Feature 0: Uninterpretable

TOP ACTIVATIONS
MAX = 3.299

,
Token,
Feature activation+0.000
then
Token then
Feature activation+0.000
they
Token they
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.448
money
Token money
Feature activation+3.299
to
Token to
Feature activation+0.000
pay
Token pay
Feature activation+0.000
these
Token these
Feature activation+0.000
victims
Token victims
Feature activation+0.000
of
Token of
Feature activation+0.000
believe
Token believe
Feature activation+0.000
she
Token she
Feature activation+0.000
does
Token does
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
stamina
Token stamina
Feature activation+3.013
,"
Token,"
Feature activation+0.000
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
target
Token target
Feature activation+0.000
then
Token then
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.370
capacity
Token capacity
Feature activation+2.383
to
Token to
Feature activation+0.000
shift
Token shift
Feature activation+0.000
things
Token things
Feature activation+0.000
globally
Token globally
Feature activation+0.000
."
Token."
Feature activation+0.000
I
Token I
Feature activation+0.000
can
Token can
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
wisdom
Token wisdom
Feature activation+1.945
to
Token to
Feature activation+0.000
know
Token know
Feature activation+0.000
the
Token the
Feature activation+0.000
difference
Token difference
Feature activation+0.000
.
Token.
Feature activation+0.000
2011
Token 2011
Feature activation+0.000
draft
Token draft
Feature activation+0.000
and
Token and
Feature activation+0.000
had
Token had
Feature activation+0.000
the
Token the
Feature activation+0.000
weight
Token weight
Feature activation+1.778
of
Token of
Feature activation+0.000
Cleveland
Token Cleveland
Feature activation+0.000
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
shoulders
Token shoulders
Feature activation+0.000
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+1.648
look
Token look
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Even
Token Even
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
He
TokenHe
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.782
capacity
Token capacity
Feature activation+1.578
to
Token to
Feature activation+0.000
influence
Token influence
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+0.000
.
Token.
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
courage
Token courage
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+1.344
energy
Token energy
Feature activation+1.343
âĢĶ
Token âĢĶ
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
you
Token you
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
courage
Token courage
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+1.344
energy
Token energy
Feature activation+1.343
âĢĶ
Token âĢĶ
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
prime
Token prime
Feature activation+0.000
minister
Token minister
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
numbers
Token numbers
Feature activation+1.171
then
Token then
Feature activation+0.000
it
Token it
Feature activation+0.000
's
Token's
Feature activation+0.000
over
Token over
Feature activation+0.000
,"
Token,"
Feature activation+0.000
the
Token the
Feature activation+0.000
scouting
Token scouting
Feature activation+0.000
department
Token department
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
analytical
Token analytical
Feature activation+1.129
department
Token department
Feature activation+0.000
says
Token says
Feature activation+0.000
he
Token he
Feature activation+0.000
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.000
I
Token I
Feature activation+0.000
can
Token can
Feature activation+0.000
get
Token get
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
out
Token out
Feature activation+1.006
of
Token of
Feature activation+0.000
them
Token them
Feature activation+0.000
.
Token.
Feature activation+0.000
So
Token So
Feature activation+0.000
I
Token I
Feature activation+0.000
and
Token and
Feature activation+0.000
future
Token future
Feature activation+0.000
generations
Token generations
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
stability
Token stability
Feature activation+0.984
needed
Token needed
Feature activation+0.000
to
Token to
Feature activation+0.000
thrive
Token thrive
Feature activation+0.000
at
Token at
Feature activation+0.000
school
Token school
Feature activation+0.000
had
Token had
Feature activation+0.000
both
Token both
Feature activation+0.000
the
Token the
Feature activation+0.000
will
Token will
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.912
means
Token means
Feature activation+0.625
to
Token to
Feature activation+0.000
fight
Token fight
Feature activation+0.000
the
Token the
Feature activation+0.000
unlimited
Token unlimited
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
Titans
Token Titans
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
edge
Token edge
Feature activation+0.805
based
Token based
Feature activation+0.000
on
Token on
Feature activation+0.000
conference
Token conference
Feature activation+0.000
record
Token record
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
He
TokenHe
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.782
capacity
Token capacity
Feature activation+1.578
to
Token to
Feature activation+0.000
influence
Token influence
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+0.000
certainly
Token certainly
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
power
Token power
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.695
room
Token room
Feature activation+0.342
to
Token to
Feature activation+0.000
cut
Token cut
Feature activation+0.000
rates
Token rates
Feature activation+0.000
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
the
Token the
Feature activation+0.000
high
Token high
Feature activation+0.000
-
Token-
Feature activation+0.000
tech
Tokentech
Feature activation+0.000
ability
Token ability
Feature activation+0.658
to
Token to
Feature activation+0.000
enrich
Token enrich
Feature activation+0.000
uranium
Token uranium
Feature activation+0.000
or
Token or
Feature activation+0.000
process
Token process
Feature activation+0.000
,
Token,
Feature activation+0.000
play
Token play
Feature activation+0.000
and
Token and
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
space
Token space
Feature activation+0.634
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
citizens
Token citizens
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
who
Token who
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
heart
Token heart
Feature activation+0.000
and
Token and
Feature activation+0.000
soul
Token soul
Feature activation+0.625
required
Token required
Feature activation+0.000
to
Token to
Feature activation+0.000
save
Token save
Feature activation+0.000
the
Token the
Feature activation+0.000
planet
Token planet
Feature activation+0.000

Top DFA by src position
MAX = 5.029

PL
Token PL
Feature activation+0.013
O
TokenO
Feature activation+0.036
have
Token have
Feature activation+0.059
the
Token the
Feature activation+0.118
funds
Token funds
Feature activation+0.264
to
Token to
Feature activation+4.699
pay
Token pay
Feature activation+0.192
the
Token the
Feature activation+0.114
families
Token families
Feature activation+0.171
of
Token of
Feature activation-0.026
the
Token the
Feature activation+0.013
the
Token the
Feature activation+0.095
"
Token "
Feature activation+0.001
st
Tokenst
Feature activation-0.006
amina
Tokenamina
Feature activation+0.035
"
Token"
Feature activation+2.287
to
Token to
Feature activation+5.029
be
Token be
Feature activation+0.112
commander
Token commander
Feature activation-0.052
in
Token in
Feature activation-0.006
chief
Token chief
Feature activation-0.027
.
Token.
Feature activation-0.020
"
Token"
Feature activation-0.028
He
TokenHe
Feature activation-0.028
has
Token has
Feature activation-0.004
the
Token the
Feature activation+0.047
capacity
Token capacity
Feature activation+0.041
to
Token to
Feature activation+3.628
influence
Token influence
Feature activation+0.096
the
Token the
Feature activation+0.031
US
Token US
Feature activation-0.119
.
Token.
Feature activation-0.056
If
Token If
Feature activation-0.158
cannot
Token cannot
Feature activation+0.099
change
Token change
Feature activation-0.056
,
Token,
Feature activation+0.057
the
Token the
Feature activation+0.057
courage
Token courage
Feature activation-0.037
to
Token to
Feature activation+4.618
change
Token change
Feature activation-0.025
the
Token the
Feature activation+0.068
things
Token things
Feature activation-0.034
I
Token I
Feature activation+0.011
can
Token can
Feature activation+0.045
draft
Token draft
Feature activation+0.008
and
Token and
Feature activation+0.015
had
Token had
Feature activation+0.003
the
Token the
Feature activation+0.035
weight
Token weight
Feature activation+0.001
of
Token of
Feature activation+3.352
Cleveland
Token Cleveland
Feature activation+0.720
on
Token on
Feature activation-0.016
his
Token his
Feature activation-0.007
shoulders
Token shoulders
Feature activation-0.035
as
Token as
Feature activation+0.012
âĢ
TokenâĢ
Feature activation+0.037
Ļ
TokenĻ
Feature activation+0.007
t
Tokent
Feature activation-0.018
have
Token have
Feature activation-0.029
the
Token the
Feature activation+0.149
muscles
Token muscles
Feature activation+4.486
and
Token and
Feature activation+0.825
he
Token he
Feature activation+0.015
didn
Token didn
Feature activation-0.118
âĢ
TokenâĢ
Feature activation+0.186
Ļ
TokenĻ
Feature activation+0.047
)
Token)
Feature activation-0.000
Rudd
Token Rudd
Feature activation-0.071
has
Token has
Feature activation-0.031
the
Token the
Feature activation+0.082
capacity
Token capacity
Feature activation+0.194
to
Token to
Feature activation+3.982
be
Token be
Feature activation+0.788
a
Token a
Feature activation+0.128
world
Token world
Feature activation-0.096
leader
Token leader
Feature activation-0.163
,"
Token,"
Feature activation-0.195
think
Token think
Feature activation-0.072
he
Token he
Feature activation-0.052
has
Token has
Feature activation-0.026
the
Token the
Feature activation+0.287
courage
Token courage
Feature activation+0.575
and
Token and
Feature activation+4.036
the
Token the
Feature activation+0.767
energy
Token energy
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
think
Token think
Feature activation-0.102
he
Token he
Feature activation-0.033
has
Token has
Feature activation-0.030
the
Token the
Feature activation+0.473
courage
Token courage
Feature activation-0.024
and
Token and
Feature activation+3.946
the
Token the
Feature activation+1.207
energy
Token energy
Feature activation+0.277
âĢĶ
Token âĢĶ
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
did
Token did
Feature activation-0.044
not
Token not
Feature activation-0.016
have
Token have
Feature activation-0.016
the
Token the
Feature activation+0.065
numbers
Token numbers
Feature activation-0.054
to
Token to
Feature activation+3.616
retain
Token retain
Feature activation-0.009
leadership
Token leadership
Feature activation-0.114
.
Token.
Feature activation-0.034
Ċ
TokenĊ
Feature activation+0.003
Ċ
TokenĊ
Feature activation+0.011
bas
Token bas
Feature activation+0.015
emen
Tokenemen
Feature activation-0.024
unless
Token unless
Feature activation-0.068
the
Token the
Feature activation+0.171
scouting
Token scouting
Feature activation+0.199
department
Token department
Feature activation+3.513
and
Token and
Feature activation+0.258
the
Token the
Feature activation+0.738
analytical
Token analytical
Feature activation+0.058
department
Token department
Feature activation+0.000
says
Token says
Feature activation+0.000
can
Token can
Feature activation+0.088
get
Token get
Feature activation+0.231
the
Token the
Feature activation+0.274
most
Token most
Feature activation+0.072
out
Token out
Feature activation+0.042
of
Token of
Feature activation+2.753
me
Token me
Feature activation-0.201
and
Token and
Feature activation-0.059
which
Token which
Feature activation-0.061
schools
Token schools
Feature activation+0.106
I
Token I
Feature activation+0.092
also
Token also
Feature activation+0.015
gives
Token gives
Feature activation-0.009
people
Token people
Feature activation-0.000
the
Token the
Feature activation+0.021
stability
Token stability
Feature activation+0.007
to
Token to
Feature activation+4.661
plan
Token plan
Feature activation-0.017
for
Token for
Feature activation+0.007
the
Token the
Feature activation+0.010
future
Token future
Feature activation-0.015
,
Token,
Feature activation-0.007
who
Token who
Feature activation-0.019
had
Token had
Feature activation+0.045
both
Token both
Feature activation+0.160
the
Token the
Feature activation+0.417
will
Token will
Feature activation+1.867
and
Token and
Feature activation+2.381
the
Token the
Feature activation+0.172
means
Token means
Feature activation+0.000
to
Token to
Feature activation+0.000
fight
Token fight
Feature activation+0.000
the
Token the
Feature activation+0.000
Bills
Token Bills
Feature activation-0.015
would
Token would
Feature activation-0.002
have
Token have
Feature activation+0.005
the
Token the
Feature activation+0.027
edge
Token edge
Feature activation+0.262
on
Token on
Feature activation+3.813
record
Token record
Feature activation-0.057
against
Token against
Feature activation-0.012
common
Token common
Feature activation-0.082
opponents
Token opponents
Feature activation-0.084
(
Token (
Feature activation-0.035
Kevin
TokenKevin
Feature activation+0.004
)
Token)
Feature activation-0.007
Rudd
Token Rudd
Feature activation-0.023
has
Token has
Feature activation+0.061
the
Token the
Feature activation+0.089
capacity
Token capacity
Feature activation+2.200
to
Token to
Feature activation+1.651
be
Token be
Feature activation+0.150
a
Token a
Feature activation+0.032
world
Token world
Feature activation-0.068
leader
Token leader
Feature activation-0.184
RBI
Token RBI
Feature activation+0.153
certainly
Token certainly
Feature activation+0.091
has
Token has
Feature activation+0.064
the
Token the
Feature activation+0.285
power
Token power
Feature activation+1.141
and
Token and
Feature activation+2.097
the
Token the
Feature activation+0.444
room
Token room
Feature activation+0.000
to
Token to
Feature activation+0.000
cut
Token cut
Feature activation+0.000
rates
Token rates
Feature activation+0.000
of
Token of
Feature activation+0.105
the
Token the
Feature activation-0.001
art
Token art
Feature activation+0.021
technological
Token technological
Feature activation+0.220
ability
Token ability
Feature activation+0.121
to
Token to
Feature activation+3.598
develop
Token develop
Feature activation+0.411
nuclear
Token nuclear
Feature activation-0.056
weapons
Token weapons
Feature activation-0.010
as
Token as
Feature activation+0.051
well
Token well
Feature activation+0.073
have
Token have
Feature activation+0.004
the
Token the
Feature activation+0.010
resources
Token resources
Feature activation+0.019
and
Token and
Feature activation+0.478
structures
Token structures
Feature activation+0.020
to
Token to
Feature activation+1.233
do
Token do
Feature activation+0.026
so
Token so
Feature activation+0.000
.
Token.
Feature activation+0.013
It
Token It
Feature activation+0.004
rest
Token rest
Feature activation-0.012
challenger
Token challenger
Feature activation-0.033
who
Token who
Feature activation-0.011
has
Token has
Feature activation+0.030
the
Token the
Feature activation+0.291
heart
Token heart
Feature activation+0.208
and
Token and
Feature activation+3.103
soul
Token soul
Feature activation+0.366
required
Token required
Feature activation+0.000
to
Token to
Feature activation+0.000
save
Token save
Feature activation+0.000
the
Token the
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.21

Head 1: 0.09

Head 2: 0.08

Head 3: 0.05

Head 4: 0.08

Head 5: 0.11

Head 6: 0.04

Head 7: 0.05

Head 8: 0.10

Head 9: 0.07

Head 10: 0.06

Head 11: 0.06

Positive logits

velt1.76

apixel1.75

pse1.67

resil1.60

1.58

ability1.56

redients1.54

antioxid1.52

Shiny1.51

nurturing1.50

knack1.47

spons1.44

requisite1.43

Ability1.42

stamps1.41

[+1.41

feat1.41

hesda1.40

htaking1.40

ascript1.39

Negative logits

IRC-1.68

UC-1.63

Example-1.57

byte-1.54

things-1.53

AU-1.51

etus-1.47

cigarettes-1.43

esta-1.41

iddles-1.41

iddle-1.40

OE-1.40

iter-1.38

Es-1.37

especially-1.37

Hat-1.36

Maker-1.35

je-1.35

oth-1.34

ibling-1.34

INTERVAL 2.969 - 3.299
CONTAINS 0.000%

believe
Token believe
Feature activation+0.000
she
Token she
Feature activation+0.000
does
Token does
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
stamina
Token stamina
Feature activation+3.013
,"
Token,"
Feature activation+0.000
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
,
Token,
Feature activation+0.000
then
Token then
Feature activation+0.000
they
Token they
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.448
money
Token money
Feature activation+3.299
to
Token to
Feature activation+0.000
pay
Token pay
Feature activation+0.000
these
Token these
Feature activation+0.000
victims
Token victims
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 2.639 - 2.969
CONTAINS 0.000%

INTERVAL 2.310 - 2.639
CONTAINS 0.000%

target
Token target
Feature activation+0.000
then
Token then
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.370
capacity
Token capacity
Feature activation+2.383
to
Token to
Feature activation+0.000
shift
Token shift
Feature activation+0.000
things
Token things
Feature activation+0.000
globally
Token globally
Feature activation+0.000
."
Token."
Feature activation+0.000

INTERVAL 1.980 - 2.310
CONTAINS 0.000%

INTERVAL 1.650 - 1.980
CONTAINS 0.000%

2011
Token 2011
Feature activation+0.000
draft
Token draft
Feature activation+0.000
and
Token and
Feature activation+0.000
had
Token had
Feature activation+0.000
the
Token the
Feature activation+0.000
weight
Token weight
Feature activation+1.778
of
Token of
Feature activation+0.000
Cleveland
Token Cleveland
Feature activation+0.000
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
shoulders
Token shoulders
Feature activation+0.000
I
Token I
Feature activation+0.000
can
Token can
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
wisdom
Token wisdom
Feature activation+1.945
to
Token to
Feature activation+0.000
know
Token know
Feature activation+0.000
the
Token the
Feature activation+0.000
difference
Token difference
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 1.320 - 1.650
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
He
TokenHe
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.782
capacity
Token capacity
Feature activation+1.578
to
Token to
Feature activation+0.000
influence
Token influence
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+0.000
.
Token.
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
courage
Token courage
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+1.344
energy
Token energy
Feature activation+1.343
âĢĶ
Token âĢĶ
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
you
Token you
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
courage
Token courage
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+1.344
energy
Token energy
Feature activation+1.343
âĢĶ
Token âĢĶ
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
you
Token you
Feature activation+0.000
have
Token have
Feature activation+0.000
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+1.648
look
Token look
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Even
Token Even
Feature activation+0.000

INTERVAL 0.990 - 1.320
CONTAINS 0.000%

the
Token the
Feature activation+0.000
prime
Token prime
Feature activation+0.000
minister
Token minister
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
numbers
Token numbers
Feature activation+1.171
then
Token then
Feature activation+0.000
it
Token it
Feature activation+0.000
's
Token's
Feature activation+0.000
over
Token over
Feature activation+0.000
,"
Token,"
Feature activation+0.000
I
Token I
Feature activation+0.000
can
Token can
Feature activation+0.000
get
Token get
Feature activation+0.000
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
out
Token out
Feature activation+1.006
of
Token of
Feature activation+0.000
them
Token them
Feature activation+0.000
.
Token.
Feature activation+0.000
So
Token So
Feature activation+0.000
I
Token I
Feature activation+0.000
the
Token the
Feature activation+0.000
scouting
Token scouting
Feature activation+0.000
department
Token department
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
analytical
Token analytical
Feature activation+1.129
department
Token department
Feature activation+0.000
says
Token says
Feature activation+0.000
he
Token he
Feature activation+0.000
can
Token can
Feature activation+0.000
be
Token be
Feature activation+0.000

INTERVAL 0.660 - 0.990
CONTAINS 0.000%

,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
Titans
Token Titans
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
edge
Token edge
Feature activation+0.805
based
Token based
Feature activation+0.000
on
Token on
Feature activation+0.000
conference
Token conference
Feature activation+0.000
record
Token record
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
He
TokenHe
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.782
capacity
Token capacity
Feature activation+1.578
to
Token to
Feature activation+0.000
influence
Token influence
Feature activation+0.000
the
Token the
Feature activation+0.000
US
Token US
Feature activation+0.000
and
Token and
Feature activation+0.000
future
Token future
Feature activation+0.000
generations
Token generations
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
stability
Token stability
Feature activation+0.984
needed
Token needed
Feature activation+0.000
to
Token to
Feature activation+0.000
thrive
Token thrive
Feature activation+0.000
at
Token at
Feature activation+0.000
school
Token school
Feature activation+0.000
certainly
Token certainly
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
power
Token power
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.695
room
Token room
Feature activation+0.342
to
Token to
Feature activation+0.000
cut
Token cut
Feature activation+0.000
rates
Token rates
Feature activation+0.000
,
Token,
Feature activation+0.000
had
Token had
Feature activation+0.000
both
Token both
Feature activation+0.000
the
Token the
Feature activation+0.000
will
Token will
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.912
means
Token means
Feature activation+0.625
to
Token to
Feature activation+0.000
fight
Token fight
Feature activation+0.000
the
Token the
Feature activation+0.000
unlimited
Token unlimited
Feature activation+0.000

INTERVAL 0.330 - 0.660
CONTAINS 0.000%

as
Token as
Feature activation+0.000
the
Token the
Feature activation+0.000
high
Token high
Feature activation+0.000
-
Token-
Feature activation+0.000
tech
Tokentech
Feature activation+0.000
ability
Token ability
Feature activation+0.658
to
Token to
Feature activation+0.000
enrich
Token enrich
Feature activation+0.000
uranium
Token uranium
Feature activation+0.000
or
Token or
Feature activation+0.000
process
Token process
Feature activation+0.000
has
Token has
Feature activation+0.000
the
Token the
Feature activation+0.000
power
Token power
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.695
room
Token room
Feature activation+0.342
to
Token to
Feature activation+0.000
cut
Token cut
Feature activation+0.000
rates
Token rates
Feature activation+0.000
,
Token,
Feature activation+0.000
yes
Token yes
Feature activation+0.000
get
Tokenget
Feature activation+0.000
arts
Token arts
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
the
Token the
Feature activation+0.000
ability
Token ability
Feature activation+0.509
to
Token to
Feature activation+0.000
judge
Token judge
Feature activation+0.000
of
Token of
Feature activation+0.000
their
Token their
Feature activation+0.000
usefulness
Token usefulness
Feature activation+0.000
month
Token month
Feature activation+0.000
,
Token,
Feature activation+0.000
then
Token then
Feature activation+0.000
they
Token they
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.448
money
Token money
Feature activation+3.299
to
Token to
Feature activation+0.000
pay
Token pay
Feature activation+0.000
these
Token these
Feature activation+0.000
victims
Token victims
Feature activation+0.000
get
Token get
Feature activation+0.000
money
Token money
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
not
Token not
Feature activation+0.000
enough
Token enough
Feature activation+0.409
âĢĵ
Token âĢĵ
Feature activation+0.000
to
Token to
Feature activation+0.000
establish
Token establish
Feature activation+0.000
peace
Token peace
Feature activation+0.000
(
Token (
Feature activation+0.000

INTERVAL 0.000 - 0.330
CONTAINS 100.000%

was
Token was
Feature activation+0.000
added
Token added
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
final
Token final
Feature activation+0.000
tax
Token tax
Feature activation+0.000
bill
Token bill
Feature activation+0.000
as
Token as
Feature activation+0.000
part
Token part
Feature activation+0.000
of
Token of
Feature activation+0.000
an
Token an
Feature activation+0.000
legal
Token legal
Feature activation+0.000
.
Token.
Feature activation+0.000
Like
Token Like
Feature activation+0.000
it
Token it
Feature activation+0.000
or
Token or
Feature activation+0.000
not
Token not
Feature activation+0.000
.
Token.
Feature activation+0.000
Now
Token Now
Feature activation+0.000
stop
Token stop
Feature activation+0.000
,"
Token,"
Feature activation+0.000
the
Token the
Feature activation+0.000
rise
Token rise
Feature activation+0.000
to
Token to
Feature activation+0.000
less
Token less
Feature activation+0.000
relig
Token relig
Feature activation+0.000
iosity
Tokeniosity
Feature activation+0.000
and
Token and
Feature activation+0.000
less
Token less
Feature activation+0.000
meaning
Token meaning
Feature activation+0.000
in
Token in
Feature activation+0.000
life
Token life
Feature activation+0.000
.
Token.
Feature activation+0.000
do
Token do
Feature activation+0.000
you
Token you
Feature activation+0.000
belong
Token belong
Feature activation+0.000
to
Token to
Feature activation+0.000
?'
Token?'
Feature activation+0.000
and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
knew
Token knew
Feature activation+0.000
what
Token what
Feature activation+0.000
was
Token was
Feature activation+0.000
happening
Token happening
Feature activation+0.000
dangerous
Token dangerous
Feature activation+0.000
is
Token is
Feature activation+0.000
he
Token he
Feature activation+0.000
?
Token?
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
asked
Token asked
Feature activation+0.000
the
Token the
Feature activation+0.000
man
Token man
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 1: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.777

Jays
Token Jays
Feature activation-0.012
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.001
baseball
Token baseball
Feature activation-0.032
cap
Token cap
Feature activation+0.052
and
Token and
Feature activation+0.214
a
Token a
Feature activation-0.074
beer
Token beer
Feature activation-0.056
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Jays
Token Jays
Feature activation-0.017
âĢ
TokenâĢ
Feature activation+0.017
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation-0.029
cap
Token cap
Feature activation-0.049
and
Token and
Feature activation+0.369
a
Token a
Feature activation-0.179
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.040
Blue
Token Blue
Feature activation+0.002
Jays
Token Jays
Feature activation-0.018
âĢ
TokenâĢ
Feature activation+0.045
Ļ
TokenĻ
Feature activation+0.004
baseball
Token baseball
Feature activation+0.121
cap
Token cap
Feature activation-0.112
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.003
less
Token less
Feature activation-0.012
lofty
Token lofty
Feature activation+0.005
goal
Token goal
Feature activation+0.008
:
Token:
Feature activation-0.055
Put
Token Put
Feature activation+0.193
a
Token a
Feature activation-0.031
Toronto
Token Toronto
Feature activation+0.018
Blue
Token Blue
Feature activation-0.002
Jays
Token Jays
Feature activation-0.018
âĢ
TokenâĢ
Feature activation+0.010
,
Token,
Feature activation-0.176
the
Token the
Feature activation-0.168
NB
Token NB
Feature activation-0.009
Space
Token Space
Feature activation-0.025
Race
Token Race
Feature activation-0.181
had
Token had
Feature activation+0.184
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.592
,
Token,
Feature activation-0.062
the
Token the
Feature activation-0.214
NB
Token NB
Feature activation-0.050
Space
Token Space
Feature activation+0.201
Race
Token Race
Feature activation-0.115
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
,
Token,
Feature activation-0.032
the
Token the
Feature activation-0.127
NB
Token NB
Feature activation-0.012
Space
Token Space
Feature activation-0.021
Race
Token Race
Feature activation-0.099
had
Token had
Feature activation+0.374
a
Token a
Feature activation+0.163
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
the
Token the
Feature activation-0.142
NB
Token NB
Feature activation-0.016
Space
Token Space
Feature activation-0.025
Race
Token Race
Feature activation-0.007
had
Token had
Feature activation+0.058
a
Token a
Feature activation+0.093
somewhat
Token somewhat
Feature activation+0.077
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
lofty
Token lofty
Feature activation-0.002
goal
Token goal
Feature activation-0.012
:
Token:
Feature activation-0.016
Put
Token Put
Feature activation+0.021
a
Token a
Feature activation-0.034
Toronto
Token Toronto
Feature activation+0.183
Blue
Token Blue
Feature activation-0.017
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.006
less
Token less
Feature activation-0.010
lofty
Token lofty
Feature activation+0.005
goal
Token goal
Feature activation-0.033
:
Token:
Feature activation-0.041
Put
Token Put
Feature activation+0.204
a
Token a
Feature activation-0.039
Toronto
Token Toronto
Feature activation-0.016
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
had
Token had
Feature activation-0.007
a
Token a
Feature activation+0.008
somewhat
Token somewhat
Feature activation+0.013
less
Token less
Feature activation-0.016
lofty
Token lofty
Feature activation+0.051
goal
Token goal
Feature activation+0.071
:
Token:
Feature activation-0.089
Put
Token Put
Feature activation+0.061
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.006
less
Token less
Feature activation-0.026
lofty
Token lofty
Feature activation-0.005
goal
Token goal
Feature activation-0.050
:
Token:
Feature activation-0.083
Put
Token Put
Feature activation+0.205
a
Token a
Feature activation+0.001
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
:
Token:
Feature activation-0.039
Put
Token Put
Feature activation+0.038
a
Token a
Feature activation-0.027
Toronto
Token Toronto
Feature activation+0.013
Blue
Token Blue
Feature activation-0.018
Jays
Token Jays
Feature activation+0.091
âĢ
TokenâĢ
Feature activation+0.013
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.005
less
Token less
Feature activation-0.018
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation-0.031
:
Token:
Feature activation-0.054
Put
Token Put
Feature activation+0.089
a
Token a
Feature activation-0.062
Toronto
Token Toronto
Feature activation+0.013
Blue
Token Blue
Feature activation-0.070
Jays
Token Jays
Feature activation-0.058
âĢ
TokenâĢ
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.007
less
Token less
Feature activation-0.005
lofty
Token lofty
Feature activation+0.005
goal
Token goal
Feature activation+0.009
:
Token:
Feature activation-0.029
Put
Token Put
Feature activation+0.110
a
Token a
Feature activation-0.030
Toronto
Token Toronto
Feature activation-0.021
Blue
Token Blue
Feature activation+0.004
Jays
Token Jays
Feature activation-0.039
âĢ
TokenâĢ
Feature activation-0.100
somewhat
Token somewhat
Feature activation+0.011
less
Token less
Feature activation-0.019
lofty
Token lofty
Feature activation+0.037
goal
Token goal
Feature activation-0.019
:
Token:
Feature activation-0.101
Put
Token Put
Feature activation+0.052
a
Token a
Feature activation-0.020
Toronto
Token Toronto
Feature activation-0.032
Blue
Token Blue
Feature activation-0.007
Jays
Token Jays
Feature activation-0.041
âĢ
TokenâĢ
Feature activation-0.092
had
Token had
Feature activation+0.005
a
Token a
Feature activation+0.036
somewhat
Token somewhat
Feature activation+0.021
less
Token less
Feature activation-0.034
lofty
Token lofty
Feature activation+0.060
goal
Token goal
Feature activation+0.777
:
Token:
Feature activation-0.033
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Race
Token Race
Feature activation-0.120
had
Token had
Feature activation+0.008
a
Token a
Feature activation+0.039
somewhat
Token somewhat
Feature activation+0.008
less
Token less
Feature activation-0.055
lofty
Token lofty
Feature activation+0.248
goal
Token goal
Feature activation+0.064
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
the
Token the
Feature activation-0.104
NB
Token NB
Feature activation-0.021
Space
Token Space
Feature activation-0.016
Race
Token Race
Feature activation-0.005
had
Token had
Feature activation+0.057
a
Token a
Feature activation+0.071
somewhat
Token somewhat
Feature activation+0.035
less
Token less
Feature activation-0.195
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation-0.111
NB
Token NB
Feature activation-0.062
Space
Token Space
Feature activation-0.018
Race
Token Race
Feature activation-0.033
had
Token had
Feature activation+0.037
a
Token a
Feature activation+0.052
somewhat
Token somewhat
Feature activation-0.025
less
Token less
Feature activation-0.148
lofty
Token lofty
Feature activation-0.046
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.05

Head 2: 0.10

Head 3: 0.09

Head 4: 0.09

Head 5: 0.07

Head 6: 0.08

Head 7: 0.08

Head 8: 0.09

Head 9: 0.08

Head 10: 0.09

Head 11: 0.09

Positive logits

FIG1.77

pint1.65

LV1.60

numbered1.57

reversible1.55

inel1.53

thirsty1.49

priced1.48

erity1.48

brewer1.46

odder1.46

LP1.45

flyer1.44

VL1.44

KD1.43

CBS1.43

freezer1.42

cumulative1.42

dose1.42

unpre1.40

Negative logits

Writing-1.76

Merit-1.75

Usage-1.75

sexism-1.71

Scholars-1.71

-1.71

Femin-1.71

-1.67

Studies-1.63

Ori-1.59

oit-1.59

ugen-1.59

Gender-1.57

proble-1.56

Problems-1.56

Writing-1.56

EVA-1.56

Cosponsors-1.54

Languages-1.54

thodox-1.53

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

is
Token is
Feature activation+0.000
foc
Token foc
Feature activation+0.000
ussed
Tokenussed
Feature activation+0.000
on
Token on
Feature activation+0.000
security
Token security
Feature activation+0.000
and
Token and
Feature activation+0.000
tends
Token tends
Feature activation+0.000
to
Token to
Feature activation+0.000
prioritize
Token prioritize
Feature activation+0.000
new
Token new
Feature activation+0.000
features
Token features
Feature activation+0.000
two
Token two
Feature activation+0.000
ng
Token ng
Feature activation+0.000
If
TokenIf
Feature activation+0.000
directives
Token directives
Feature activation+0.000
with
Token with
Feature activation+0.000
opposed
Token opposed
Feature activation+0.000
boolean
Token boolean
Feature activation+0.000
conditions
Token conditions
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
,
Token,
Feature activation+0.000
or
Token or
Feature activation+0.000
nuclear
Token nuclear
Feature activation+0.000
plant
Token plant
Feature activation+0.000
).
Token).
Feature activation+0.000
Basically
Token Basically
Feature activation+0.000
,
Token,
Feature activation+0.000
electricity
Token electricity
Feature activation+0.000
is
Token is
Feature activation+0.000
generated
Token generated
Feature activation+0.000
from
Token from
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Korean
Token Korean
Feature activation+0.000
conflict
Token conflict
Feature activation+0.000
was
Token was
Feature activation+0.000
his
Token his
Feature activation+0.000
service
Token service
Feature activation+0.000
window
Token window
Feature activation+0.000
.
Token.
Feature activation+0.000
He
Token He
Feature activation+0.000
met
Token met
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
I
TokenI
Feature activation+0.000
've
Token've
Feature activation+0.000
never
Token never
Feature activation+0.000
read
Token read
Feature activation+0.000
Vladimir
Token Vladimir
Feature activation+0.000
Nab
Token Nab
Feature activation+0.000
ok
Tokenok
Feature activation+0.000
ov
Tokenov
Feature activation+0.000
's
Token's
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 2: Uninterpretable

TOP ACTIVATIONS
MAX = 8.905

Bo
Token Bo
Feature activation+0.000
er
Tokener
Feature activation+0.000
War
Token War
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.905
copyright
Token copyright
Feature activation+0.000
Hanson
Token Hanson
Feature activation+0.000
's
Token's
Feature activation+0.000
Auction
Token Auction
Feature activation+0.000
eers
Tokeneers
Feature activation+0.000
like
Token like
Feature activation+0.000
to
Token to
Feature activation+0.000
rest
Token rest
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.510
copyright
Token copyright
Feature activation+0.000
Aden
Token Aden
Feature activation+0.000
B
Token B
Feature activation+0.000
ish
Tokenish
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
supermarket
Token supermarket
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.215
copyright
Token copyright
Feature activation+0.000
CPS
Token CPS
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Davies
Token Davies
Feature activation+0.000
during
Token during
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.043
copyright
Token copyright
Feature activation+0.000
Hanson
Token Hanson
Feature activation+0.000
's
Token's
Feature activation+0.000
Auction
Token Auction
Feature activation+0.000
eers
Tokeneers
Feature activation+0.000
to
Token to
Feature activation+0.000
neighbouring
Token neighbouring
Feature activation+0.000
countries
Token countries
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.005
copyright
Token copyright
Feature activation+0.000
AP
Token AP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
A
Token A
Feature activation+0.000
more
Token more
Feature activation+0.000
Mothers
Token Mothers
Feature activation+0.000
":
Token":
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.909
copyright
Token copyright
Feature activation+0.000
Raj
Token Raj
Feature activation+0.000
deep
Tokendeep
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ast
Tokenast
Feature activation+0.000
opol
Tokenopol
Feature activation+0.000
airport
Token airport
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.724
copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Meanwhile
Token Meanwhile
Feature activation+0.000
their
Token their
Feature activation+0.000
flights
Token flights
Feature activation+0.000
home
Token home
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.654
copyright
Token copyright
Feature activation+0.000
W
Token W
Feature activation+0.000
ael
Tokenael
Feature activation+0.000
Hussein
Token Hussein
Feature activation+0.000
/
Token/
Feature activation+0.000
's
Token's
Feature activation+0.000
air
Token air
Feature activation+0.000
strike
Token strike
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.397
copyright
Token copyright
Feature activation+0.000
AP
Token AP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
S
Token S
Feature activation+0.000
in
Token in
Feature activation+0.000
central
Token central
Feature activation+0.000
Barcelona
Token Barcelona
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.334
copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
The
Token The
Feature activation+0.000
Kh
Token Kh
Feature activation+0.000
d
Tokend
Feature activation+0.000
air
Tokenair
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.328
copyright
Token copyright
Feature activation+0.000
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Image
Token Image
Feature activation+0.454
caption
Token caption
Feature activation+0.000
police
Token police
Feature activation+0.000
in
Token in
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.218
copyright
Token copyright
Feature activation+0.000
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
had
Token had
Feature activation+0.000
lost
Token lost
Feature activation+0.000
consciousness
Token consciousness
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.205
copyright
Token copyright
Feature activation+0.000
YouTube
Token YouTube
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Davies
Token Davies
Feature activation+0.000
the
Token the
Feature activation+0.000
town
Token town
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.102
copyright
Token copyright
Feature activation+0.000
Mer
Token Mer
Feature activation+0.000
id
Tokenid
Feature activation+0.000
ith
Tokenith
Feature activation+0.000
Koh
Token Koh
Feature activation+0.000
over
Token over
Feature activation+0.000
Dund
Token Dund
Feature activation+0.000
ee
Tokenee
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.028
copyright
Token copyright
Feature activation+0.000
Alan
Token Alan
Feature activation+0.000
Mill
Token Mill
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
Image
Token Image
Feature activation+0.000
the
Token the
Feature activation+0.000
flames
Token flames
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.865
copyright
Token copyright
Feature activation+0.000
Reuters
Token Reuters
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
T
Token T
Feature activation+0.000
after
Token after
Feature activation+0.000
the
Token the
Feature activation+0.000
verdict
Token verdict
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.861
copyright
Token copyright
Feature activation+0.000
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Cecil
Token Cecil
Feature activation+0.000
the
Token the
Feature activation+0.000
lion
Token lion
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.592
copyright
Token copyright
Feature activation+0.000
Reuters
Token Reuters
Feature activation+0.000
Image
Token Image
Feature activation+0.144
caption
Token caption
Feature activation+0.000
A
Token A
Feature activation+0.000
identifying
Token identifying
Feature activation+0.000
insign
Token insign
Feature activation+0.000
ia
Tokenia
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.547
copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Men
Token Men
Feature activation+0.000
into
Token into
Feature activation+0.000
heaven
Token heaven
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.528
copyright
Token copyright
Feature activation+0.000
Press
Token Press
Feature activation+0.000
Eye
Token Eye
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000

Top DFA by src position
MAX = 8.248

dog
Token dog
Feature activation-0.001
.
Token.
Feature activation+0.011
Ċ
TokenĊ
Feature activation+0.108
Ċ
TokenĊ
Feature activation+0.098
Image
TokenImage
Feature activation+0.012
copyright
Token copyright
Feature activation+4.648
Hanson
Token Hanson
Feature activation-0.027
's
Token's
Feature activation-0.048
Auction
Token Auction
Feature activation-0.015
eers
Tokeneers
Feature activation-0.002
Image
Token Image
Feature activation-0.004
authorities
Token authorities
Feature activation-0.000
."
Token."
Feature activation-0.006
Ċ
TokenĊ
Feature activation+0.062
Ċ
TokenĊ
Feature activation+0.061
Image
TokenImage
Feature activation+0.041
copyright
Token copyright
Feature activation+8.248
Aden
Token Aden
Feature activation+0.147
B
Token B
Feature activation-0.005
ish
Tokenish
Feature activation-0.001
ar
Tokenar
Feature activation-0.003
Image
Token Image
Feature activation+0.002
lost
Token lost
Feature activation-0.000
consciousness
Token consciousness
Feature activation-0.001
Ċ
TokenĊ
Feature activation+0.023
Ċ
TokenĊ
Feature activation+0.094
Image
TokenImage
Feature activation+0.024
copyright
Token copyright
Feature activation+4.607
YouTube
Token YouTube
Feature activation+0.004
Image
Token Image
Feature activation+0.004
caption
Token caption
Feature activation+0.305
Davies
Token Davies
Feature activation-0.033
told
Token told
Feature activation+0.012
dog
Token dog
Feature activation-0.001
.
Token.
Feature activation+0.010
Ċ
TokenĊ
Feature activation+0.134
Ċ
TokenĊ
Feature activation+0.126
Image
TokenImage
Feature activation+0.024
copyright
Token copyright
Feature activation+7.840
Hanson
Token Hanson
Feature activation-0.036
's
Token's
Feature activation-0.039
Auction
Token Auction
Feature activation-0.016
eers
Tokeneers
Feature activation-0.002
Image
Token Image
Feature activation+0.001
finished
Token finished
Feature activation-0.001
.
Token.
Feature activation+0.006
Ċ
TokenĊ
Feature activation+0.204
Ċ
TokenĊ
Feature activation+0.172
Image
TokenImage
Feature activation+0.022
copyright
Token copyright
Feature activation+7.749
Reuters
Token Reuters
Feature activation-0.007
Image
Token Image
Feature activation+0.004
caption
Token caption
Feature activation+0.791
Sect
Token Sect
Feature activation+0.012
arian
Tokenarian
Feature activation+0.005
comment
Token comment
Feature activation-0.016
:
Token:
Feature activation+0.038
Ċ
TokenĊ
Feature activation+0.101
Ċ
TokenĊ
Feature activation+0.110
Image
TokenImage
Feature activation+0.012
copyright
Token copyright
Feature activation+8.116
Ar
Token Ar
Feature activation+0.067
vind
Tokenvind
Feature activation-0.007
Kejriwal
Token Kejriwal
Feature activation-0.005
Ċ
TokenĊ
Feature activation+0.032
Ċ
TokenĊ
Feature activation+0.086
several
Token several
Feature activation+0.006
trucks
Token trucks
Feature activation-0.006
Ċ
TokenĊ
Feature activation-0.323
Ċ
TokenĊ
Feature activation-0.160
Image
TokenImage
Feature activation+0.009
copyright
Token copyright
Feature activation+4.749
Getty
Token Getty
Feature activation-0.003
Images
Token Images
Feature activation-0.000
Image
Token Image
Feature activation+0.001
caption
Token caption
Feature activation+0.303
They
Token They
Feature activation+0.020
it
Token it
Feature activation-0.001
."
Token."
Feature activation-0.010
Ċ
TokenĊ
Feature activation+0.081
Ċ
TokenĊ
Feature activation+0.101
Image
TokenImage
Feature activation+0.031
copyright
Token copyright
Feature activation+7.731
W
Token W
Feature activation-0.008
ael
Tokenael
Feature activation+0.000
Hussein
Token Hussein
Feature activation+0.001
/
Token/
Feature activation-0.001
BBC
TokenBBC
Feature activation+0.001
launchers
Token launchers
Feature activation-0.001
.
Token.
Feature activation-0.010
Ċ
TokenĊ
Feature activation+0.031
Ċ
TokenĊ
Feature activation+0.082
Image
TokenImage
Feature activation+0.013
copyright
Token copyright
Feature activation+7.373
Reuters
Token Reuters
Feature activation+0.003
Image
Token Image
Feature activation+0.006
caption
Token caption
Feature activation+0.922
These
Token These
Feature activation+0.049
women
Token women
Feature activation-0.003
minded
Tokenminded
Feature activation-0.009
.
Token.
Feature activation-0.020
Ċ
TokenĊ
Feature activation+0.017
Ċ
TokenĊ
Feature activation+0.104
Image
TokenImage
Feature activation+0.023
copyright
Token copyright
Feature activation+7.417
Reuters
Token Reuters
Feature activation-0.001
Image
Token Image
Feature activation+0.001
caption
Token caption
Feature activation+0.933
Participants
Token Participants
Feature activation+0.004
formed
Token formed
Feature activation+0.001
Monday
Token Monday
Feature activation+0.004
.
Token.
Feature activation-0.005
Ċ
TokenĊ
Feature activation+0.081
Ċ
TokenĊ
Feature activation+0.131
Image
TokenImage
Feature activation+0.041
copyright
Token copyright
Feature activation+7.544
AFP
Token AFP
Feature activation-0.035
Image
Token Image
Feature activation+0.016
caption
Token caption
Feature activation+0.715
Violent
Token Violent
Feature activation+0.015
protests
Token protests
Feature activation+0.002
time
Token time
Feature activation-0.002
.
Token.
Feature activation-0.010
Ċ
TokenĊ
Feature activation+0.101
Ċ
TokenĊ
Feature activation+0.138
Image
TokenImage
Feature activation+0.031
copyright
Token copyright
Feature activation+6.837
Reuters
Token Reuters
Feature activation+0.002
Image
Token Image
Feature activation+0.010
caption
Token caption
Feature activation+0.956
Former
Token Former
Feature activation+0.029
officer
Token officer
Feature activation+0.002
death
Token death
Feature activation-0.016
.
Token.
Feature activation+0.003
Ċ
TokenĊ
Feature activation-0.018
Ċ
TokenĊ
Feature activation+0.096
Image
TokenImage
Feature activation+0.055
copyright
Token copyright
Feature activation+7.485
casc
Token casc
Feature activation+0.027
aden
Tokenaden
Feature activation-0.123
ews
Tokenews
Feature activation-0.025
.
Token.
Feature activation-0.021
co
Tokenco
Feature activation-0.040
high
Token high
Feature activation-0.001
.
Token.
Feature activation+0.009
Ċ
TokenĊ
Feature activation+0.002
Ċ
TokenĊ
Feature activation+0.031
Image
TokenImage
Feature activation+0.017
copyright
Token copyright
Feature activation+7.825
Mer
Token Mer
Feature activation+0.032
id
Tokenid
Feature activation-0.004
ith
Tokenith
Feature activation-0.002
Koh
Token Koh
Feature activation-0.003
ut
Tokenut
Feature activation-0.001
of
Token of
Feature activation-0.000
Scotland
Token Scotland
Feature activation-0.002
Ċ
TokenĊ
Feature activation+0.081
Ċ
TokenĊ
Feature activation+0.118
Image
TokenImage
Feature activation+0.018
copyright
Token copyright
Feature activation+6.499
Gordon
Token Gordon
Feature activation-0.002
Mill
Token Mill
Feature activation+0.006
igan
Tokenigan
Feature activation+0.000
Image
Token Image
Feature activation-0.009
caption
Token caption
Feature activation+0.354
Monday
Token Monday
Feature activation+0.001
.
Token.
Feature activation-0.003
Ċ
TokenĊ
Feature activation+0.014
Ċ
TokenĊ
Feature activation+0.028
Image
TokenImage
Feature activation+0.010
copyright
Token copyright
Feature activation+4.531
AFP
Token AFP
Feature activation-0.016
Image
Token Image
Feature activation+0.002
caption
Token caption
Feature activation+0.070
Violent
Token Violent
Feature activation-0.002
protests
Token protests
Feature activation-0.004
Smith
Token Smith
Feature activation-0.005
.
Token.
Feature activation-0.002
Ċ
TokenĊ
Feature activation+0.045
Ċ
TokenĊ
Feature activation+0.139
Image
TokenImage
Feature activation+0.047
copyright
Token copyright
Feature activation+6.594
Reuters
Token Reuters
Feature activation+0.006
Image
Token Image
Feature activation+0.010
caption
Token caption
Feature activation+0.980
Protesters
Token Protesters
Feature activation+0.020
took
Token took
Feature activation+0.012
<|endoftext|>
Token<|endoftext|>
Feature activation+0.828
Image
TokenImage
Feature activation+0.034
copyright
Token copyright
Feature activation+7.925
EPA
Token EPA
Feature activation-0.014
/
Token/
Feature activation-0.027
Facebook
TokenFacebook
Feature activation-0.014
Image
Token Image
Feature activation+0.015
caption
Token caption
Feature activation+0.570
several
Token several
Feature activation+0.007
trucks
Token trucks
Feature activation-0.007
Ċ
TokenĊ
Feature activation-0.360
Ċ
TokenĊ
Feature activation-0.118
Image
TokenImage
Feature activation+0.019
copyright
Token copyright
Feature activation+7.838
Getty
Token Getty
Feature activation-0.005
Images
Token Images
Feature activation+0.005
Image
Token Image
Feature activation+0.010
caption
Token caption
Feature activation+1.176
They
Token They
Feature activation+0.045
support
Token support
Feature activation-0.003
."
Token."
Feature activation-0.012
Ċ
TokenĊ
Feature activation+0.002
Ċ
TokenĊ
Feature activation+0.016
Image
TokenImage
Feature activation+0.015
copyright
Token copyright
Feature activation+6.655
Press
Token Press
Feature activation-0.026
Eye
Token Eye
Feature activation-0.003
Image
Token Image
Feature activation-0.001
caption
Token caption
Feature activation+0.198
National
Token National
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.17

Head 1: 0.24

Head 2: 0.02

Head 3: 0.06

Head 4: 0.03

Head 5: 0.13

Head 6: 0.09

Head 7: 0.02

Head 8: 0.12

Head 9: 0.05

Head 10: 0.05

Head 11: 0.03

Positive logits

closure1.63

>>>1.59

ifully1.56

tyr1.53

yright1.52

journalism1.52

VER1.51

javascript1.49

efully1.47

orney1.45

equivalents1.45

caption1.44

endif1.43

git1.42

verett1.41

WATCHED1.41

ditch1.41

refuel1.40

aback1.40

mist1.40

Negative logits

gdala-2.08

ゼウス-1.94

Init-1.74

Assass-1.68

Glac-1.67

Participant-1.67

tower-1.66

Associ-1.63

iggurat-1.60

Conquer-1.58

mite-1.56

Osama-1.55

Acceler-1.54

Orig-1.54

allion-1.53

Monstrous-1.53

Aman-1.52

atial-1.52

Found-1.51

Ogre-1.50

INTERVAL 8.014 - 8.905
CONTAINS 0.000%

during
Token during
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.043
copyright
Token copyright
Feature activation+0.000
Hanson
Token Hanson
Feature activation+0.000
's
Token's
Feature activation+0.000
Auction
Token Auction
Feature activation+0.000
eers
Tokeneers
Feature activation+0.000
Bo
Token Bo
Feature activation+0.000
er
Tokener
Feature activation+0.000
War
Token War
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.905
copyright
Token copyright
Feature activation+0.000
Hanson
Token Hanson
Feature activation+0.000
's
Token's
Feature activation+0.000
Auction
Token Auction
Feature activation+0.000
eers
Tokeneers
Feature activation+0.000
like
Token like
Feature activation+0.000
to
Token to
Feature activation+0.000
rest
Token rest
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.510
copyright
Token copyright
Feature activation+0.000
Aden
Token Aden
Feature activation+0.000
B
Token B
Feature activation+0.000
ish
Tokenish
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
supermarket
Token supermarket
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.215
copyright
Token copyright
Feature activation+0.000
CPS
Token CPS
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Davies
Token Davies
Feature activation+0.000

INTERVAL 7.124 - 8.014
CONTAINS 0.000%

in
Token in
Feature activation+0.000
central
Token central
Feature activation+0.000
Barcelona
Token Barcelona
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.334
copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
The
Token The
Feature activation+0.000
to
Token to
Feature activation+0.000
neighbouring
Token neighbouring
Feature activation+0.000
countries
Token countries
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+8.005
copyright
Token copyright
Feature activation+0.000
AP
Token AP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
A
Token A
Feature activation+0.000
their
Token their
Feature activation+0.000
flights
Token flights
Feature activation+0.000
home
Token home
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.654
copyright
Token copyright
Feature activation+0.000
W
Token W
Feature activation+0.000
ael
Tokenael
Feature activation+0.000
Hussein
Token Hussein
Feature activation+0.000
/
Token/
Feature activation+0.000
police
Token police
Feature activation+0.000
in
Token in
Feature activation+0.000
2013
Token 2013
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.218
copyright
Token copyright
Feature activation+0.000
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
more
Token more
Feature activation+0.000
Mothers
Token Mothers
Feature activation+0.000
":
Token":
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.909
copyright
Token copyright
Feature activation+0.000
Raj
Token Raj
Feature activation+0.000
deep
Tokendeep
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 6.233 - 7.124
CONTAINS 0.000%

identifying
Token identifying
Feature activation+0.000
insign
Token insign
Feature activation+0.000
ia
Tokenia
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.547
copyright
Token copyright
Feature activation+0.000
AFP
Token AFP
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
Men
Token Men
Feature activation+0.000
the
Token the
Feature activation+0.000
town
Token town
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.102
copyright
Token copyright
Feature activation+0.000
Mer
Token Mer
Feature activation+0.000
id
Tokenid
Feature activation+0.000
ith
Tokenith
Feature activation+0.000
Koh
Token Koh
Feature activation+0.000
after
Token after
Feature activation+0.000
the
Token the
Feature activation+0.000
verdict
Token verdict
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.861
copyright
Token copyright
Feature activation+0.000
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
over
Token over
Feature activation+0.000
Dund
Token Dund
Feature activation+0.000
ee
Tokenee
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+7.028
copyright
Token copyright
Feature activation+0.000
Alan
Token Alan
Feature activation+0.000
Mill
Token Mill
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
Image
Token Image
Feature activation+0.000
the
Token the
Feature activation+0.000
flames
Token flames
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.865
copyright
Token copyright
Feature activation+0.000
Reuters
Token Reuters
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
T
Token T
Feature activation+0.000

INTERVAL 5.343 - 6.233
CONTAINS 0.000%

one
Token one
Feature activation+0.000
official
Token official
Feature activation+0.000
said
Token said
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+6.077
copyright
Token copyright
Feature activation+0.000
EPA
Token EPA
Feature activation+0.000
Image
Token Image
Feature activation+0.000
caption
Token caption
Feature activation+0.000
The
Token The
Feature activation+0.000

INTERVAL 4.452 - 5.343
CONTAINS 0.000%

INTERVAL 3.562 - 4.452
CONTAINS 0.000%

ews
Tokenews
Feature activation+0.000
.
Token.
Feature activation+0.000
co
Tokenco
Feature activation+0.000
.
Token.
Feature activation+0.000
uk
Tokenuk
Feature activation+0.000
Image
Token Image
Feature activation+3.598
caption
Token caption
Feature activation+0.000
Dr
Token Dr
Feature activation+0.000
B
Token B
Feature activation+0.000
ham
Tokenham
Feature activation+0.000
bra
Tokenbra
Feature activation+0.000

INTERVAL 2.671 - 3.562
CONTAINS 0.000%

Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
Henry
Token Henry
Feature activation+0.000
Ford
Token Ford
Feature activation+0.000
Hospital
Token Hospital
Feature activation+0.000
Image
Token Image
Feature activation+2.769
caption
Token caption
Feature activation+0.000
Dr
Token Dr
Feature activation+0.000
Nag
Token Nag
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
w
Tokenw
Feature activation+0.000

INTERVAL 1.781 - 2.671
CONTAINS 0.000%

the
Token the
Feature activation+0.000
final
Token final
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Media
TokenMedia
Feature activation+2.481
playback
Token playback
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
supported
Token supported
Feature activation+0.000
on
Token on
Feature activation+0.000
ator
Tokenator
Feature activation+0.000
re
Tokenre
Feature activation+0.000
H
Token H
Feature activation+0.000
idal
Tokenidal
Feature activation+0.000
go
Tokengo
Feature activation+0.000
Image
Token Image
Feature activation+2.530
caption
Token caption
Feature activation+0.000
The
Token The
Feature activation+0.000
structure
Token structure
Feature activation+0.000
of
Token of
Feature activation+0.000
Till
Token Till
Feature activation+0.000
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
House
Token House
Feature activation+0.000
of
Token of
Feature activation+0.000
Commons
Token Commons
Feature activation+0.000
Image
Token Image
Feature activation+2.230
caption
Token caption
Feature activation+0.000
A
Token A
Feature activation+0.000
copy
Token copy
Feature activation+0.000
of
Token of
Feature activation+0.000
John
Token John
Feature activation+0.000
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
Anna
Token Anna
Feature activation+0.000
Holl
Token Holl
Feature activation+0.000
igan
Tokenigan
Feature activation+0.000
Image
Token Image
Feature activation+2.238
caption
Token caption
Feature activation+0.000
Black
Token Black
Feature activation+0.000
Pete
Token Pete
Feature activation+0.000
par
Token par
Feature activation+0.000
ades
Tokenades
Feature activation+0.000
1977
Token 1977
Feature activation+0.000
,
Token,
Feature activation+0.000
Lifetime
Token Lifetime
Feature activation+0.000
colour
Token colour
Feature activation+0.000
photograph
Token photograph
Feature activation+0.000
.
Token.
Feature activation+2.189
Copyright
Token Copyright
Feature activation+0.000
The
Token The
Feature activation+0.000
Estate
Token Estate
Feature activation+0.000
of
Token of
Feature activation+0.000
Ana
Token Ana
Feature activation+0.000

INTERVAL 0.890 - 1.781
CONTAINS 0.000%

<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
Think
Token Think
Feature activation+0.000
stock
Tokenstock
Feature activation+0.000
Image
Token Image
Feature activation+1.711
caption
Token caption
Feature activation+0.000
Domestic
Token Domestic
Feature activation+0.000
violence
Token violence
Feature activation+0.000
often
Token often
Feature activation+0.000
goes
Token goes
Feature activation+0.000
six
Token six
Feature activation+0.000
decades
Token decades
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+0.975
caption
Token caption
Feature activation+0.000
Many
Token Many
Feature activation+0.000
of
Token of
Feature activation+0.000
De
Token De
Feature activation+0.000
y
Tokeny
Feature activation+0.000
Image
TokenImage
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
Ni
Token Ni
Feature activation+0.000
all
Tokenall
Feature activation+0.000
Carson
Token Carson
Feature activation+0.000
Image
Token Image
Feature activation+1.180
caption
Token caption
Feature activation+0.000
Northern
Token Northern
Feature activation+0.000
Ireland
Token Ireland
Feature activation+0.000
people
Token people
Feature activation+0.000
's
Token's
Feature activation+0.000
passing
Token passing
Feature activation+0.000
plane
Token plane
Feature activation+0.000
overhead
Token overhead
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Image
TokenImage
Feature activation+0.957
copyright
Token copyright
Feature activation+0.000
Ja
Token Ja
Feature activation+0.000
ison
Tokenison
Feature activation+0.000
Pod
Token Pod
Feature activation+0.000
kan
Tokenkan
Feature activation+0.000
copyright
Token copyright
Feature activation+0.000
Aden
Token Aden
Feature activation+0.000
B
Token B
Feature activation+0.000
ish
Tokenish
Feature activation+0.000
ar
Tokenar
Feature activation+0.000
Image
Token Image
Feature activation+1.244
caption
Token caption
Feature activation+0.000
The
Token The
Feature activation+0.000
captured
Token captured
Feature activation+0.000
che
Token che
Feature activation+0.000
et
Tokenet
Feature activation+0.000

INTERVAL 0.000 - 0.890
CONTAINS 100.000%

general
Token general
Feature activation+0.000
manager
Token manager
Feature activation+0.000
Bro
Token Bro
Feature activation+0.000
dy
Tokendy
Feature activation+0.000
E
Token E
Feature activation+0.000
hr
Tokenhr
Feature activation+0.000
lich
Tokenlich
Feature activation+0.000
told
Token told
Feature activation+0.000
Fox
Token Fox
Feature activation+0.000
News
TokenNews
Feature activation+0.000
.
Token.
Feature activation+0.000
places
Token places
Feature activation+0.000
but
Token but
Feature activation+0.000
some
Token some
Feature activation+0.000
great
Token great
Feature activation+0.000
one
Token one
Feature activation+0.000
-
Token-
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
Doctor
Token Doctor
Feature activation+0.000
,
Token,
Feature activation+0.000
an
Token an
Feature activation+0.000
effective
Token effective
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
sp
Tokensp
Feature activation+0.000
itting
Tokenitting
Feature activation+0.000
law
Token law
Feature activation+0.000
will
Token will
Feature activation+0.000
bring
Token bring
Feature activation+0.000
down
Token down
Feature activation+0.000
inc
Token inc
Feature activation+0.000
CPS
Token CPS
Feature activation+0.000
Friday
Token Friday
Feature activation+0.000
evening
Token evening
Feature activation+0.000
.
Token.
Feature activation+0.000
d
Token d
Feature activation+0.000
p
Tokenp
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
EL
Token EL
Feature activation+0.000
P
Token P
Feature activation+0.000
AS
TokenAS
Feature activation+0.000
O
TokenO
Feature activation+0.000
shocked
Token shocked
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
person
Token person
Feature activation+0.000
drinks
Token drinks
Feature activation+0.000
heavily
Token heavily
Feature activation+0.000
and
Token and
Feature activation+0.000
enjoys
Token enjoys
Feature activation+0.000
an
Token an
Feature activation+0.000
active
Token active
Feature activation+0.000
social
Token social
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 3: Follows “Arm"

TOP ACTIVATIONS
MAX = 4.182

timed
Token timed
Feature activation+0.000
to
Token to
Feature activation+0.000
coincide
Token coincide
Feature activation+0.000
with
Token with
Feature activation+0.000
Arm
Token Arm
Feature activation+0.000
ist
Tokenist
Feature activation+4.176
ice
Tokenice
Feature activation+0.000
Day
Token Day
Feature activation+0.000
and
Token and
Feature activation+0.000
Remem
Token Remem
Feature activation+0.000
brance
Tokenbrance
Feature activation+0.000
front
Token front
Feature activation+0.000
door
Token door
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
,
Token,
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+4.057
in
Tokenin
Feature activation+1.284
-
Token-
Feature activation+0.358
arm
Tokenarm
Feature activation+0.000
with
Token with
Feature activation+0.000
her
Token her
Feature activation+0.000
been
Token been
Feature activation+0.000
an
Token an
Feature activation+0.000
example
Token example
Feature activation+0.000
of
Token of
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+4.017
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.000
Press
Token Press
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
stool
Token stool
Feature activation+0.000
,
Token,
Feature activation+0.000
arm
Token arm
Feature activation+0.000
resting
Token resting
Feature activation+3.999
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
leg
Token leg
Feature activation+0.000
,
Token,
Feature activation+0.000
answering
Token answering
Feature activation+0.000
We
Token We
Feature activation+0.000
used
Token used
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
arm
Token arm
Feature activation+0.000
in
Token in
Feature activation+3.995
arm
Token arm
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
But
Token But
Feature activation+0.000
after
Token after
Feature activation+0.000
some
Token some
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+3.947
tw
Tokentw
Feature activation+0.000
isting
Tokenisting
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
promise
Token promise
Feature activation+0.000
into
Token into
Feature activation+0.000
walking
Token walking
Feature activation+0.000
about
Token about
Feature activation+0.000
and
Token and
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+3.837
w
Tokenw
Feature activation+3.572
aving
Tokenaving
Feature activation+0.000
.
Token.
Feature activation+0.000
Poor
Token Poor
Feature activation+0.000
Mary
Token Mary
Feature activation+0.000
fire
Token fire
Feature activation+0.000
from
Token from
Feature activation+0.000
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
arm
Token arm
Feature activation+0.000
ist
Tokenist
Feature activation+3.661
ice
Tokenice
Feature activation+0.000
line
Token line
Feature activation+0.000
hit
Token hit
Feature activation+0.000
the
Token the
Feature activation+0.000
Israeli
Token Israeli
Feature activation+0.000
t
Tokent
Feature activation+0.000
cost
Token cost
Feature activation+0.000
you
Token you
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.599
a
Token a
Feature activation+2.829
leg
Token leg
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
your
Token your
Feature activation+0.000
walking
Token walking
Feature activation+0.000
about
Token about
Feature activation+0.000
and
Token and
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+3.837
w
Tokenw
Feature activation+3.572
aving
Tokenaving
Feature activation+0.000
.
Token.
Feature activation+0.000
Poor
Token Poor
Feature activation+0.000
Mary
Token Mary
Feature activation+0.000
Beard
Token Beard
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
-
Token-
Feature activation+0.000
Fix
Token Fix
Feature activation+0.000
for
Token for
Feature activation+0.000
arm
Token arm
Feature activation+0.000
v
Tokenv
Feature activation+3.471
7
Token7
Feature activation+0.000
(
Token (
Feature activation+0.000
i
Tokeni
Feature activation+0.000
4
Token4
Feature activation+0.000
s
Tokens
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Chinese
TokenChinese
Feature activation+0.000
'
Token '
Feature activation+0.000
arm
Tokenarm
Feature activation+0.000
-
Token-
Feature activation+3.458
tw
Tokentw
Feature activation+0.409
isting
Tokenisting
Feature activation+0.000
'
Token'
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Virt
Token Virt
Feature activation+0.000
ually
Tokenually
Feature activation+0.000
every
Token every
Feature activation+0.000
other
Token other
Feature activation+0.000
arm
Token arm
Feature activation+0.000
of
Token of
Feature activation+3.428
the
Token the
Feature activation+1.727
government
Token government
Feature activation+0.000
turned
Token turned
Feature activation+0.000
to
Token to
Feature activation+0.000
tactics
Token tactics
Feature activation+0.000
their
Token their
Feature activation+0.000
attempts
Token attempts
Feature activation+0.000
to
Token to
Feature activation+0.000
lawfully
Token lawfully
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.418
that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
does
Token does
Feature activation+0.000
not
Token not
Feature activation+0.000
support
Token support
Feature activation+0.000
any
Token any
Feature activation+0.000
unexpected
Token unexpected
Feature activation+0.000
movements
Token movements
Feature activation+0.000
of
Token of
Feature activation+0.000
arms
Token arms
Feature activation+0.000
or
Token or
Feature activation+3.386
legs
Token legs
Feature activation+0.000
while
Token while
Feature activation+0.000
we
Token we
Feature activation+0.000
performed
Token performed
Feature activation+0.000
the
Token the
Feature activation+0.000
because
Token because
Feature activation+0.000
I
Token I
Feature activation+0.000
had
Token had
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.372
shoulder
Token shoulder
Feature activation+0.764
injury
Token injury
Feature activation+0.000
so
Token so
Feature activation+0.000
I
Token I
Feature activation+0.000
had
Token had
Feature activation+0.000
pistol
Token pistol
Feature activation+0.000
is
Token is
Feature activation+0.000
our
Token our
Feature activation+0.000
new
Token new
Feature activation+0.000
arm
Token arm
Feature activation+0.000
brace
Token brace
Feature activation+3.299
adapter
Token adapter
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
quickly
Token quickly
Feature activation+0.000
and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
converted
Token converted
Feature activation+0.000
with
Token with
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.243
sword
Token sword
Feature activation+1.619
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
1980
Token 1980
Feature activation+0.000
s
Tokens
Feature activation+0.000
Co
Token Co
Feature activation+0.000
pp
Tokenpp
Feature activation+0.000
,
Token,
Feature activation+0.000
Joel
Token Joel
Feature activation+0.000
Arm
Token Arm
Feature activation+0.000
ia
Tokenia
Feature activation+3.221
,
Token,
Feature activation+0.000
Connor
Token Connor
Feature activation+0.000
Hel
Token Hel
Feature activation+0.000
le
Tokenle
Feature activation+0.000
buy
Tokenbuy
Feature activation+0.000
boyfriend
Token boyfriend
Feature activation+0.000
s
Tokens
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+3.210
he
Token he
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
6
Token 6
Feature activation+0.000

Top DFA by src position
MAX = 6.490

were
Token were
Feature activation-0.004
timed
Token timed
Feature activation-0.007
to
Token to
Feature activation-0.010
coincide
Token coincide
Feature activation-0.032
with
Token with
Feature activation-0.031
Arm
Token Arm
Feature activation+6.029
ist
Tokenist
Feature activation+0.206
ice
Tokenice
Feature activation+0.000
Day
Token Day
Feature activation+0.000
and
Token and
Feature activation+0.000
Remem
Token Remem
Feature activation+0.000
s
Tokens
Feature activation+0.006
front
Token front
Feature activation+0.001
door
Token door
Feature activation+0.004
Tuesday
Token Tuesday
Feature activation+0.011
,
Token,
Feature activation-0.063
arm
Token arm
Feature activation+6.299
-
Token-
Feature activation+0.097
in
Tokenin
Feature activation+0.000
-
Token-
Feature activation+0.000
arm
Tokenarm
Feature activation+0.000
with
Token with
Feature activation+0.000
has
Token has
Feature activation+0.002
been
Token been
Feature activation-0.000
an
Token an
Feature activation-0.002
example
Token example
Feature activation-0.009
of
Token of
Feature activation-0.006
arm
Token arm
Feature activation+6.166
-
Token-
Feature activation+0.114
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.000
Press
Token Press
Feature activation+0.000
foot
Token foot
Feature activation+0.503
on
Token on
Feature activation+0.030
a
Token a
Feature activation+0.037
stool
Token stool
Feature activation+0.006
,
Token,
Feature activation+0.278
arm
Token arm
Feature activation+5.310
resting
Token resting
Feature activation+0.132
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
leg
Token leg
Feature activation+0.000
,
Token,
Feature activation+0.000
.
Token.
Feature activation+0.019
We
Token We
Feature activation+0.030
used
Token used
Feature activation-0.014
to
Token to
Feature activation-0.034
be
Token be
Feature activation-0.080
arm
Token arm
Feature activation+6.490
in
Token in
Feature activation+0.192
arm
Token arm
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation-0.001
Ŀ
TokenĿ
Feature activation+0.002
But
Token But
Feature activation+0.012
after
Token after
Feature activation+0.013
some
Token some
Feature activation+0.017
arm
Token arm
Feature activation+6.024
-
Token-
Feature activation+0.096
tw
Tokentw
Feature activation+0.000
isting
Tokenisting
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
turn
Token turn
Feature activation-0.008
into
Token into
Feature activation-0.018
walking
Token walking
Feature activation+0.002
about
Token about
Feature activation+0.014
and
Token and
Feature activation+0.048
arm
Token arm
Feature activation+6.010
-
Token-
Feature activation+0.033
w
Tokenw
Feature activation+0.000
aving
Tokenaving
Feature activation+0.000
.
Token.
Feature activation+0.000
Poor
Token Poor
Feature activation+0.000
after
Token after
Feature activation+0.035
fire
Token fire
Feature activation-0.000
from
Token from
Feature activation-0.001
across
Token across
Feature activation+0.023
the
Token the
Feature activation+0.014
arm
Token arm
Feature activation+5.695
ist
Tokenist
Feature activation-0.079
ice
Tokenice
Feature activation+0.000
line
Token line
Feature activation+0.000
hit
Token hit
Feature activation+0.000
the
Token the
Feature activation+0.000
Ļ
TokenĻ
Feature activation-0.020
t
Tokent
Feature activation-0.088
cost
Token cost
Feature activation+0.087
you
Token you
Feature activation-0.000
an
Token an
Feature activation+0.061
arm
Token arm
Feature activation+5.871
and
Token and
Feature activation+0.101
a
Token a
Feature activation+0.000
leg
Token leg
Feature activation+0.000
to
Token to
Feature activation+0.000
protect
Token protect
Feature activation+0.000
turn
Token turn
Feature activation-0.006
into
Token into
Feature activation-0.009
walking
Token walking
Feature activation-0.003
about
Token about
Feature activation+0.081
and
Token and
Feature activation+0.049
arm
Token arm
Feature activation+5.636
-
Token-
Feature activation+0.080
w
Tokenw
Feature activation+0.019
aving
Tokenaving
Feature activation+0.000
.
Token.
Feature activation+0.000
Poor
Token Poor
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.001
Ċ
TokenĊ
Feature activation+0.000
-
Token-
Feature activation+0.005
Fix
Token Fix
Feature activation+0.008
for
Token for
Feature activation-0.003
arm
Token arm
Feature activation+5.496
v
Tokenv
Feature activation+0.068
7
Token7
Feature activation+0.000
(
Token (
Feature activation+0.000
i
Tokeni
Feature activation+0.000
4
Token4
Feature activation+0.000
.
Token.
Feature activation+0.005
Ċ
TokenĊ
Feature activation+0.001
Ċ
TokenĊ
Feature activation+0.003
Chinese
TokenChinese
Feature activation+0.016
'
Token '
Feature activation-0.002
arm
Tokenarm
Feature activation+5.613
-
Token-
Feature activation+0.153
tw
Tokentw
Feature activation+0.000
isting
Tokenisting
Feature activation+0.000
'
Token'
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
.
Token.
Feature activation-0.014
Virt
Token Virt
Feature activation-0.012
ually
Tokenually
Feature activation+0.010
every
Token every
Feature activation+0.005
other
Token other
Feature activation-0.007
arm
Token arm
Feature activation+6.124
of
Token of
Feature activation+0.009
the
Token the
Feature activation+0.000
government
Token government
Feature activation+0.000
turned
Token turned
Feature activation+0.000
to
Token to
Feature activation+0.000
in
Token in
Feature activation+0.021
their
Token their
Feature activation+0.025
attempts
Token attempts
Feature activation-0.003
to
Token to
Feature activation-0.064
lawfully
Token lawfully
Feature activation-0.023
arm
Token arm
Feature activation+5.901
and
Token and
Feature activation+0.201
that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
does
Token does
Feature activation+0.000
not
Token not
Feature activation+0.000
prevent
Token prevent
Feature activation+0.017
any
Token any
Feature activation+0.093
unexpected
Token unexpected
Feature activation+0.019
movements
Token movements
Feature activation+0.075
of
Token of
Feature activation+0.116
arms
Token arms
Feature activation+5.481
or
Token or
Feature activation+0.188
legs
Token legs
Feature activation+0.000
while
Token while
Feature activation+0.000
we
Token we
Feature activation+0.000
performed
Token performed
Feature activation+0.000
career
Token career
Feature activation-0.030
because
Token because
Feature activation+0.131
I
Token I
Feature activation+0.029
had
Token had
Feature activation+0.032
an
Token an
Feature activation+0.066
arm
Token arm
Feature activation+5.359
and
Token and
Feature activation+0.240
shoulder
Token shoulder
Feature activation+0.000
injury
Token injury
Feature activation+0.000
so
Token so
Feature activation+0.000
I
Token I
Feature activation+0.000
Scorpion
Token Scorpion
Feature activation+0.013
pistol
Token pistol
Feature activation+0.023
is
Token is
Feature activation+0.157
our
Token our
Feature activation+0.161
new
Token new
Feature activation-0.028
arm
Token arm
Feature activation+4.812
brace
Token brace
Feature activation+0.083
adapter
Token adapter
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
quickly
Token quickly
Feature activation+0.000
left
Token left
Feature activation+0.028
I
Token I
Feature activation+0.100
converted
Token converted
Feature activation-0.141
with
Token with
Feature activation-0.121
an
Token an
Feature activation+0.179
arm
Token arm
Feature activation+5.442
and
Token and
Feature activation+0.121
sword
Token sword
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
1980
Token 1980
Feature activation+0.000
Andrew
Token Andrew
Feature activation+0.017
Co
Token Co
Feature activation+0.001
pp
Tokenpp
Feature activation-0.000
,
Token,
Feature activation-0.057
Joel
Token Joel
Feature activation+0.037
Arm
Token Arm
Feature activation+5.230
ia
Tokenia
Feature activation+0.049
,
Token,
Feature activation+0.000
Connor
Token Connor
Feature activation+0.000
Hel
Token Hel
Feature activation+0.000
le
Tokenle
Feature activation+0.000
my
Token my
Feature activation+0.029
boyfriend
Token boyfriend
Feature activation-0.062
s
Tokens
Feature activation-0.015
âĢ
TokenâĢ
Feature activation+0.124
Ļ
TokenĻ
Feature activation+0.044
arms
Token arms
Feature activation+5.270
and
Token and
Feature activation+0.268
he
Token he
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.03

Head 2: 0.05

Head 3: 0.05

Head 4: 0.06

Head 5: 0.03

Head 6: 0.45

Head 7: 0.05

Head 8: 0.04

Head 9: 0.04

Head 10: 0.10

Head 11: 0.06

Positive logits

1.37

ochet1.32

Chao1.28

Trainer1.26

arte1.25

abet1.23

Lauder1.20

choke1.15

acer1.14

Fernandez1.11

Awakening1.11

oland1.10

Herrera1.10

hea1.10

istance1.10

NIH1.09

Ramos1.09

eye1.09

gorilla1.08

Gutierrez1.07

Negative logits

nesday-1.44

Downloadha-1.42

igmatic-1.37

phrine-1.35

soDeliveryDate-1.32

psc-1.31

fml-1.28

urnal-1.28

sites-1.27

thinkable-1.24

find-1.22

rencies-1.21

come-1.20

gem-1.18

geist-1.17

bring-1.17

risome-1.17

angible-1.16

ocre-1.14

lez-1.14

INTERVAL 3.764 - 4.182
CONTAINS 0.000%

on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
stool
Token stool
Feature activation+0.000
,
Token,
Feature activation+0.000
arm
Token arm
Feature activation+0.000
resting
Token resting
Feature activation+3.999
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
leg
Token leg
Feature activation+0.000
,
Token,
Feature activation+0.000
answering
Token answering
Feature activation+0.000
timed
Token timed
Feature activation+0.000
to
Token to
Feature activation+0.000
coincide
Token coincide
Feature activation+0.000
with
Token with
Feature activation+0.000
Arm
Token Arm
Feature activation+0.000
ist
Tokenist
Feature activation+4.176
ice
Tokenice
Feature activation+0.000
Day
Token Day
Feature activation+0.000
and
Token and
Feature activation+0.000
Remem
Token Remem
Feature activation+0.000
brance
Tokenbrance
Feature activation+0.000
been
Token been
Feature activation+0.000
an
Token an
Feature activation+0.000
example
Token example
Feature activation+0.000
of
Token of
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+4.017
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Canadian
Token Canadian
Feature activation+0.000
Press
Token Press
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
into
Token into
Feature activation+0.000
walking
Token walking
Feature activation+0.000
about
Token about
Feature activation+0.000
and
Token and
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+3.837
w
Tokenw
Feature activation+3.572
aving
Tokenaving
Feature activation+0.000
.
Token.
Feature activation+0.000
Poor
Token Poor
Feature activation+0.000
Mary
Token Mary
Feature activation+0.000
front
Token front
Feature activation+0.000
door
Token door
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
,
Token,
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+4.057
in
Tokenin
Feature activation+1.284
-
Token-
Feature activation+0.358
arm
Tokenarm
Feature activation+0.000
with
Token with
Feature activation+0.000
her
Token her
Feature activation+0.000

INTERVAL 3.345 - 3.764
CONTAINS 0.000%

their
Token their
Feature activation+0.000
attempts
Token attempts
Feature activation+0.000
to
Token to
Feature activation+0.000
lawfully
Token lawfully
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.418
that
Token that
Feature activation+0.000
he
Token he
Feature activation+0.000
does
Token does
Feature activation+0.000
not
Token not
Feature activation+0.000
support
Token support
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
-
Token-
Feature activation+0.000
Fix
Token Fix
Feature activation+0.000
for
Token for
Feature activation+0.000
arm
Token arm
Feature activation+0.000
v
Tokenv
Feature activation+3.471
7
Token7
Feature activation+0.000
(
Token (
Feature activation+0.000
i
Tokeni
Feature activation+0.000
4
Token4
Feature activation+0.000
s
Tokens
Feature activation+0.000
walking
Token walking
Feature activation+0.000
about
Token about
Feature activation+0.000
and
Token and
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+3.837
w
Tokenw
Feature activation+3.572
aving
Tokenaving
Feature activation+0.000
.
Token.
Feature activation+0.000
Poor
Token Poor
Feature activation+0.000
Mary
Token Mary
Feature activation+0.000
Beard
Token Beard
Feature activation+0.000
Virt
Token Virt
Feature activation+0.000
ually
Tokenually
Feature activation+0.000
every
Token every
Feature activation+0.000
other
Token other
Feature activation+0.000
arm
Token arm
Feature activation+0.000
of
Token of
Feature activation+3.428
the
Token the
Feature activation+1.727
government
Token government
Feature activation+0.000
turned
Token turned
Feature activation+0.000
to
Token to
Feature activation+0.000
tactics
Token tactics
Feature activation+0.000
fire
Token fire
Feature activation+0.000
from
Token from
Feature activation+0.000
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
arm
Token arm
Feature activation+0.000
ist
Tokenist
Feature activation+3.661
ice
Tokenice
Feature activation+0.000
line
Token line
Feature activation+0.000
hit
Token hit
Feature activation+0.000
the
Token the
Feature activation+0.000
Israeli
Token Israeli
Feature activation+0.000

INTERVAL 2.927 - 3.345
CONTAINS 0.000%

tried
Token tried
Feature activation+0.000
to
Token to
Feature activation+0.000
move
Token move
Feature activation+0.000
my
Token my
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+2.940
legs
Token legs
Feature activation+0.337
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
almost
Token almost
Feature activation+0.000
felt
Token felt
Feature activation+0.000
Co
Token Co
Feature activation+0.000
pp
Tokenpp
Feature activation+0.000
,
Token,
Feature activation+0.000
Joel
Token Joel
Feature activation+0.000
Arm
Token Arm
Feature activation+0.000
ia
Tokenia
Feature activation+3.221
,
Token,
Feature activation+0.000
Connor
Token Connor
Feature activation+0.000
Hel
Token Hel
Feature activation+0.000
le
Tokenle
Feature activation+0.000
buy
Tokenbuy
Feature activation+0.000
pistol
Token pistol
Feature activation+0.000
is
Token is
Feature activation+0.000
our
Token our
Feature activation+0.000
new
Token new
Feature activation+0.000
arm
Token arm
Feature activation+0.000
brace
Token brace
Feature activation+3.299
adapter
Token adapter
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
quickly
Token quickly
Feature activation+0.000
and
Token and
Feature activation+0.000
I
Token I
Feature activation+0.000
converted
Token converted
Feature activation+0.000
with
Token with
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.243
sword
Token sword
Feature activation+1.619
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
1980
Token 1980
Feature activation+0.000
s
Tokens
Feature activation+0.000
mers
Tokenmers
Feature activation+0.000
are
Token are
Feature activation+0.000
using
Token using
Feature activation+0.000
their
Token their
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+3.124
legs
Token legs
Feature activation+2.499
simultaneously
Token simultaneously
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 2.509 - 2.927
CONTAINS 0.000%

objective
Token objective
Feature activation+0.000
,
Token,
Feature activation+0.000
arm
Token arm
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.741
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+2.676
-
Token-
Feature activation+0.102
length
Tokenlength
Feature activation+0.000
assessment
Token assessment
Feature activation+0.000
of
Token of
Feature activation+0.000
these
Token these
Feature activation+0.000
a
Token a
Feature activation+0.000
human
Token human
Feature activation+0.000
being
Token being
Feature activation+0.000
with
Token with
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+2.533
lungs
Token lungs
Feature activation+0.869
and
Token and
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
when
Token when
Feature activation+0.000
his
Token his
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Love
TokenLove
Feature activation+0.000
ly
Tokenly
Feature activation+0.000
slim
Token slim
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+2.545
a
Token a
Feature activation+0.072
beautiful
Token beautiful
Feature activation+0.000
body
Token body
Feature activation+0.000
!
Token!
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ability
Token ability
Feature activation+0.000
to
Token to
Feature activation+0.000
reg
Token reg
Feature activation+0.000
row
Tokenrow
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+2.808
legs
Token legs
Feature activation+0.929
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
point
Token point
Feature activation+0.000
y
Tokeny
Feature activation+0.000
retract
Token retract
Feature activation+0.000
able
Tokenable
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+2.738
bl
Tokenbl
Feature activation+0.000
ades
Tokenades
Feature activation+0.000
Alleg
Token Alleg
Feature activation+0.000
iance
Tokeniance
Feature activation+0.000
Imperium
Token Imperium
Feature activation+0.000

INTERVAL 2.091 - 2.509
CONTAINS 0.000%

warriors
Token warriors
Feature activation+0.000
were
Token were
Feature activation+0.000
required
Token required
Feature activation+0.000
to
Token to
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+2.157
armor
Token armor
Feature activation+0.000
themselves
Token themselves
Feature activation+0.000
.
Token.
Feature activation+0.000
Hop
Token Hop
Feature activation+0.000
lite
Tokenlite
Feature activation+0.000
box
Token box
Feature activation+0.000
:
Token:
Feature activation+0.000
Size
Token Size
Feature activation+0.000
,
Token,
Feature activation+0.000
arm
Token arm
Feature activation+0.000
-
Token-
Feature activation+2.473
strength
Tokenstrength
Feature activation+0.000
,
Token,
Feature activation+0.000
athleticism
Token athleticism
Feature activation+0.000
,
Token,
Feature activation+0.000
productivity
Token productivity
Feature activation+0.000
Tehran
Token Tehran
Feature activation+0.000
had
Token had
Feature activation+0.000
helped
Token helped
Feature activation+0.000
to
Token to
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+2.176
finance
Token finance
Feature activation+0.000
his
Token his
Feature activation+0.000
kidn
Token kidn
Feature activation+0.000
appers
Tokenappers
Feature activation+0.000
.
Token.
Feature activation+0.000
in
Token in
Feature activation+0.000
1964
Token 1964
Feature activation+0.000
and
Token and
Feature activation+0.000
smuggling
Token smuggling
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+2.333
fighters
Token fighters
Feature activation+0.000
from
Token from
Feature activation+0.000
Turkey
Token Turkey
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
manufacturing
Token manufacturing
Feature activation+0.000
many
Token many
Feature activation+0.000
types
Token types
Feature activation+0.000
of
Token of
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+2.255
equipment
Token equipment
Feature activation+0.393
.
Token.
Feature activation+0.000
Iran
Token Iran
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

INTERVAL 1.673 - 2.091
CONTAINS 0.000%

officials
Token officials
Feature activation+0.000
and
Token and
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
arms
Tokenarms
Feature activation+0.000
-
Token-
Feature activation+1.981
trade
Tokentrade
Feature activation+0.000
activists
Token activists
Feature activation+0.000
say
Token say
Feature activation+0.000
they
Token they
Feature activation+0.000
had
Token had
Feature activation+0.000
staff
Token staff
Feature activation+0.000
was
Token was
Feature activation+0.000
in
Token in
Feature activation+0.000
an
Token an
Feature activation+0.000
arms
Token arms
Feature activation+0.000
-
Token-
Feature activation+1.985
race
Tokenrace
Feature activation+0.000
to
Token to
Feature activation+0.000
grow
Token grow
Feature activation+0.000
Aliens
Token Aliens
Feature activation+0.000
and
Token and
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
RA
TokenRA
Feature activation+0.000
F
TokenF
Feature activation+0.000
Arm
Token Arm
Feature activation+0.000
ou
Tokenou
Feature activation+1.968
rer
Tokenrer
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
Reaper
Token Reaper
Feature activation+0.000
U
Token U
Feature activation+0.000
the
Token the
Feature activation+0.000
Comm
Token Comm
Feature activation+0.000
iss
Tokeniss
Feature activation+0.000
ary
Tokenary
Feature activation+0.000
arm
Token arm
Feature activation+0.000
of
Token of
Feature activation+1.986
the
Token the
Feature activation+0.496
T
Token T
Feature activation+0.000
$
Token$
Feature activation+0.000
A
TokenA
Feature activation+0.000
(
Token (
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
flow
Token flow
Feature activation+0.000
of
Token of
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+1.833
munitions
Token munitions
Feature activation+0.000
from
Token from
Feature activation+0.000
Turkey
Token Turkey
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 1.255 - 1.673
CONTAINS 0.000%

ul
Tokenul
Feature activation+0.000
ok
Tokenok
Feature activation+0.000
had
Token had
Feature activation+0.000
many
Token many
Feature activation+0.000
arms
Token arms
Feature activation+0.000
,
Token,
Feature activation+1.273
and
Token and
Feature activation+0.000
they
Token they
Feature activation+0.000
all
Token all
Feature activation+0.000
grabbed
Token grabbed
Feature activation+0.000
for
Token for
Feature activation+0.000
benefiting
Token benefiting
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
American
Token American
Feature activation+0.000
arms
Token arms
Feature activation+0.000
and
Token and
Feature activation+1.540
aid
Token aid
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Multiple
TokenMultiple
Feature activation+0.000
that
Token that
Feature activation+0.000
kept
Token kept
Feature activation+0.000
the
Token the
Feature activation+0.000
arms
Token arms
Feature activation+0.000
of
Token of
Feature activation+1.457
the
Token the
Feature activation+1.372
drummer
Token drummer
Feature activation+0.000
,
Token,
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Hak
Token Hak
Feature activation+0.000
ius
Tokenius
Feature activation+0.000
converted
Token converted
Feature activation+0.000
with
Token with
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.243
sword
Token sword
Feature activation+1.619
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
1980
Token 1980
Feature activation+0.000
s
Tokens
Feature activation+0.000
Citadel
Token Citadel
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Russian
TokenRussian
Feature activation+0.000
arms
Token arms
Feature activation+0.000
,
Token,
Feature activation+1.531
financial
Token financial
Feature activation+0.000
and
Token and
Feature activation+0.000
energy
Token energy
Feature activation+0.000
companies
Token companies
Feature activation+0.000
are
Token are
Feature activation+0.000

INTERVAL 0.836 - 1.255
CONTAINS 0.000%

ERE
TokenERE
Feature activation+0.000
V
TokenV
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
(
Token (
Feature activation+0.000
Arm
TokenArm
Feature activation+0.000
Radio
TokenRadio
Feature activation+1.085
)âĢĶ
Token)âĢĶ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Armenian
Token Armenian
Feature activation+0.000
Ministry
Token Ministry
Feature activation+0.000
of
Token of
Feature activation+0.000
fear
Token fear
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
long
Token long
Feature activation+0.000
arm
Token arm
Feature activation+0.000
of
Token of
Feature activation+0.846
school
Token school
Feature activation+0.000
discipline
Token discipline
Feature activation+0.000
will
Token will
Feature activation+0.000
reach
Token reach
Feature activation+0.000
out
Token out
Feature activation+0.000
equipped
Token equipped
Feature activation+0.000
with
Token with
Feature activation+0.000
specialized
Token specialized
Feature activation+0.000
arm
Token arm
Feature activation+0.000
ament
Tokenament
Feature activation+0.000
and
Token and
Feature activation+1.198
missiles
Token missiles
Feature activation+0.000
.
Token.
Feature activation+0.000
These
Token These
Feature activation+0.000
need
Token need
Feature activation+0.000
to
Token to
Feature activation+0.000
ev
Tokenev
Feature activation+0.000
via
Token via
Feature activation+0.000
submission
Token submission
Feature activation+0.000
(
Token (
Feature activation+0.000
arm
Tokenarm
Feature activation+0.000
bar
Tokenbar
Feature activation+1.010
),
Token),
Feature activation+0.000
Round
Token Round
Feature activation+0.000
1
Token 1
Feature activation+0.000
Mal
TokenMal
Feature activation+0.000
ik
Tokenik
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
de
Token de
Feature activation+0.000
facto
Token facto
Feature activation+0.000
arm
Token arm
Feature activation+0.000
of
Token of
Feature activation+0.838
the
Token the
Feature activation+0.036
state
Token state
Feature activation+0.000
and
Token and
Feature activation+0.000
providing
Token providing
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 0.418 - 0.836
CONTAINS 0.001%

specific
Token specific
Feature activation+0.000
calls
Token calls
Feature activation+0.000
to
Token to
Feature activation+0.000
provide
Token provide
Feature activation+0.000
arms
Token arms
Feature activation+0.000
to
Token to
Feature activation+0.576
Ukraine
Token Ukraine
Feature activation+0.000
in
Token in
Feature activation+0.000
its
Token its
Feature activation+0.000
fight
Token fight
Feature activation+0.000
with
Token with
Feature activation+0.000
I
Token I
Feature activation+0.000
had
Token had
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
and
Token and
Feature activation+3.372
shoulder
Token shoulder
Feature activation+0.764
injury
Token injury
Feature activation+0.000
so
Token so
Feature activation+0.000
I
Token I
Feature activation+0.000
had
Token had
Feature activation+0.000
to
Token to
Feature activation+0.000
subsidiaries
Token subsidiaries
Feature activation+0.000
.
Token.
Feature activation+0.000
All
Token All
Feature activation+0.000
other
Token other
Feature activation+0.000
trademarks
Token trademarks
Feature activation+0.000
,
Token,
Feature activation+0.701
logos
Token logos
Feature activation+0.000
and
Token and
Feature activation+0.000
cop
Token cop
Feature activation+0.000
yrights
Tokenyrights
Feature activation+0.000
are
Token are
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
One
Token One
Feature activation+0.000
Belt
Token Belt
Feature activation+0.000
,
Token,
Feature activation+1.728
One
Token One
Feature activation+0.801
Road
Token Road
Feature activation+0.000
Initiative
Token Initiative
Feature activation+0.000
hosted
Token hosted
Feature activation+0.000
by
Token by
Feature activation+0.000
China
Token China
Feature activation+0.000
O
Token O
Feature activation+0.000
MB
TokenMB
Feature activation+0.000
is
Token is
Feature activation+0.000
an
Token an
Feature activation+0.000
arm
Token arm
Feature activation+0.000
of
Token of
Feature activation+0.606
the
Token the
Feature activation+0.586
provincial
Token provincial
Feature activation+0.000
government
Token government
Feature activation+0.000
,
Token,
Feature activation+0.000
meant
Token meant
Feature activation+0.000

INTERVAL 0.000 - 0.418
CONTAINS 99.997%

people
Token people
Feature activation+0.000
who
Token who
Feature activation+0.000
are
Token are
Feature activation+0.000
living
Token living
Feature activation+0.000
in
Token in
Feature activation+0.000
poverty
Token poverty
Feature activation+0.000
are
Token are
Feature activation+0.000
considered
Token considered
Feature activation+0.000
working
Token working
Feature activation+0.000
poor
Token poor
Feature activation+0.000
:
Token:
Feature activation+0.000
all
Token all
Feature activation+0.000
bullets
Token bullets
Feature activation+0.000
should
Token should
Feature activation+0.000
cost
Token cost
Feature activation+0.000
$
Token $
Feature activation+0.000
.
Token .
Feature activation+0.000
$
Token $
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
bullet
Token bullet
Feature activation+0.000
.
Token.
Feature activation+0.000
sublime
Token sublime
Feature activation+0.000
appearance
Token appearance
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
disguise
Token disguise
Feature activation+0.000
.
Token.
Feature activation+0.000
Somehow
Token Somehow
Feature activation+0.000
Ultra
Token Ultra
Feature activation+0.000
Magnus
Token Magnus
Feature activation+0.000
was
Token was
Feature activation+0.000
unable
Token unable
Feature activation+0.000
this
Token this
Feature activation+0.000
occasion
Token occasion
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
more
Token more
Feature activation+0.000
importantly
Token importantly
Feature activation+0.000
,
Token,
Feature activation+0.000
why
Token why
Feature activation+0.000
it
Token it
Feature activation+0.000
happens
Token happens
Feature activation+0.000
on
Token on
Feature activation+0.000
money
Token money
Feature activation+0.000
directly
Token directly
Feature activation+0.000
from
Token from
Feature activation+0.000
one
Token one
Feature activation+0.000
individual
Token individual
Feature activation+0.000
to
Token to
Feature activation+0.000
another
Token another
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 4: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.260

âĢ
TokenâĢ
Feature activation-0.066
Ļ
TokenĻ
Feature activation-0.010
baseball
Token baseball
Feature activation-0.014
cap
Token cap
Feature activation-0.191
and
Token and
Feature activation-0.219
a
Token a
Feature activation+0.076
beer
Token beer
Feature activation+0.023
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
less
Token less
Feature activation-0.037
lofty
Token lofty
Feature activation-0.007
goal
Token goal
Feature activation-0.030
:
Token:
Feature activation-0.058
Put
Token Put
Feature activation-0.069
a
Token a
Feature activation+0.041
Toronto
Token Toronto
Feature activation-0.044
Blue
Token Blue
Feature activation-0.015
Jays
Token Jays
Feature activation-0.036
âĢ
TokenâĢ
Feature activation-0.053
Ļ
TokenĻ
Feature activation-0.011
Toronto
Token Toronto
Feature activation-0.005
Blue
Token Blue
Feature activation-0.009
Jays
Token Jays
Feature activation-0.056
âĢ
TokenâĢ
Feature activation-0.088
Ļ
TokenĻ
Feature activation+0.024
baseball
Token baseball
Feature activation+0.092
cap
Token cap
Feature activation-0.056
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.347
,
Token,
Feature activation-0.019
the
Token the
Feature activation+0.029
NB
Token NB
Feature activation+0.002
Space
Token Space
Feature activation-0.015
Race
Token Race
Feature activation-0.013
had
Token had
Feature activation+0.005
a
Token a
Feature activation-0.001
,
Token,
Feature activation-0.328
the
Token the
Feature activation-0.185
NB
Token NB
Feature activation-0.018
Space
Token Space
Feature activation-0.102
Race
Token Race
Feature activation-0.241
had
Token had
Feature activation+0.004
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
,
Token,
Feature activation-0.224
the
Token the
Feature activation-0.130
NB
Token NB
Feature activation-0.086
Space
Token Space
Feature activation-0.126
Race
Token Race
Feature activation-0.145
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
the
Token the
Feature activation-0.165
NB
Token NB
Feature activation+0.008
Space
Token Space
Feature activation-0.084
Race
Token Race
Feature activation-0.160
had
Token had
Feature activation+0.048
a
Token a
Feature activation+0.077
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation-0.012
Space
Token Space
Feature activation-0.024
Race
Token Race
Feature activation-0.049
had
Token had
Feature activation-0.028
a
Token a
Feature activation+0.014
somewhat
Token somewhat
Feature activation+0.135
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
lofty
Token lofty
Feature activation-0.006
goal
Token goal
Feature activation-0.020
:
Token:
Feature activation-0.011
Put
Token Put
Feature activation-0.001
a
Token a
Feature activation+0.029
Toronto
Token Toronto
Feature activation+0.260
Blue
Token Blue
Feature activation-0.051
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
lofty
Token lofty
Feature activation-0.004
goal
Token goal
Feature activation-0.047
:
Token:
Feature activation-0.048
Put
Token Put
Feature activation-0.023
a
Token a
Feature activation-0.003
Toronto
Token Toronto
Feature activation+0.070
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.005
less
Token less
Feature activation-0.010
lofty
Token lofty
Feature activation-0.012
goal
Token goal
Feature activation-0.270
:
Token:
Feature activation-0.054
Put
Token Put
Feature activation+0.022
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation-0.030
lofty
Token lofty
Feature activation-0.021
goal
Token goal
Feature activation-0.093
:
Token:
Feature activation-0.126
Put
Token Put
Feature activation-0.125
a
Token a
Feature activation+0.067
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
less
Token less
Feature activation+0.001
lofty
Token lofty
Feature activation-0.009
goal
Token goal
Feature activation-0.036
:
Token:
Feature activation-0.030
Put
Token Put
Feature activation+0.034
a
Token a
Feature activation+0.037
Toronto
Token Toronto
Feature activation-0.015
Blue
Token Blue
Feature activation-0.036
Jays
Token Jays
Feature activation-0.037
âĢ
TokenâĢ
Feature activation-0.016
Ļ
TokenĻ
Feature activation+0.000
lofty
Token lofty
Feature activation-0.007
goal
Token goal
Feature activation-0.063
:
Token:
Feature activation-0.047
Put
Token Put
Feature activation-0.015
a
Token a
Feature activation+0.058
Toronto
Token Toronto
Feature activation+0.251
Blue
Token Blue
Feature activation-0.305
Jays
Token Jays
Feature activation-0.242
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.004
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation-0.006
goal
Token goal
Feature activation-0.020
:
Token:
Feature activation-0.034
Put
Token Put
Feature activation+0.086
a
Token a
Feature activation+0.046
Toronto
Token Toronto
Feature activation-0.026
Blue
Token Blue
Feature activation-0.016
Jays
Token Jays
Feature activation-0.066
âĢ
TokenâĢ
Feature activation-0.072
less
Token less
Feature activation+0.004
lofty
Token lofty
Feature activation-0.008
goal
Token goal
Feature activation-0.101
:
Token:
Feature activation-0.057
Put
Token Put
Feature activation+0.004
a
Token a
Feature activation+0.059
Toronto
Token Toronto
Feature activation-0.006
Blue
Token Blue
Feature activation-0.035
Jays
Token Jays
Feature activation-0.106
âĢ
TokenâĢ
Feature activation+0.027
Ļ
TokenĻ
Feature activation-0.008
,
Token,
Feature activation-0.319
the
Token the
Feature activation-0.031
NB
Token NB
Feature activation-0.009
Space
Token Space
Feature activation-0.043
Race
Token Race
Feature activation-0.063
had
Token had
Feature activation+0.059
a
Token a
Feature activation+0.011
somewhat
Token somewhat
Feature activation+0.002
less
Token less
Feature activation-0.003
lofty
Token lofty
Feature activation-0.008
goal
Token goal
Feature activation-0.576
Race
Token Race
Feature activation-0.118
had
Token had
Feature activation-0.036
a
Token a
Feature activation-0.040
somewhat
Token somewhat
Feature activation-0.023
less
Token less
Feature activation-0.024
lofty
Token lofty
Feature activation+0.083
goal
Token goal
Feature activation-0.297
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
NB
Token NB
Feature activation-0.015
Space
Token Space
Feature activation-0.016
Race
Token Race
Feature activation-0.043
had
Token had
Feature activation-0.039
a
Token a
Feature activation-0.021
somewhat
Token somewhat
Feature activation+0.065
less
Token less
Feature activation-0.052
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
Space
Token Space
Feature activation-0.038
Race
Token Race
Feature activation-0.085
had
Token had
Feature activation-0.023
a
Token a
Feature activation-0.025
somewhat
Token somewhat
Feature activation-0.053
less
Token less
Feature activation+0.065
lofty
Token lofty
Feature activation-0.203
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.06

Head 2: 0.10

Head 3: 0.08

Head 4: 0.07

Head 5: 0.10

Head 6: 0.08

Head 7: 0.08

Head 8: 0.11

Head 9: 0.07

Head 10: 0.08

Head 11: 0.09

Positive logits

Geh1.54

window1.53

Bridges1.53

September1.51

overfl1.51

Rousse1.46

October1.45

uckle1.44

Mellon1.44

Ceres1.43

Outbreak1.43

Bras1.42

bye1.41

Paraly1.41

Ney1.41

1.41

bourg1.41

Pan1.40

Chao1.39

Indra1.39

Negative logits

ariat-1.59

ardless-1.54

life-1.50

ITNESS-1.47

arian-1.42

negotiator-1.41

commandments-1.41

CAR-1.39

Software-1.39

FANTASY-1.38

pricing-1.38

ruled-1.38

urus-1.38

HEAD-1.34

negotiation-1.34

negotiating-1.34

arians-1.33

ollah-1.33

Justice-1.33

reasons-1.33

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

is
Token is
Feature activation+0.000
extend
Token extend
Feature activation+0.000
ible
Tokenible
Feature activation+0.000
by
Token by
Feature activation+0.000
expanding
Token expanding
Feature activation+0.000
the
Token the
Feature activation+0.000
size
Token size
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
Fin
Token Fin
Feature activation+0.000
ite
Tokenite
Feature activation+0.000
call
Token call
Feature activation+0.000
each
Token each
Feature activation+0.000
other
Token other
Feature activation+0.000
Mr
Token Mr
Feature activation+0.000
.
Token.
Feature activation+0.000
and
Token and
Feature activation+0.000
Mrs
Token Mrs
Feature activation+0.000
.
Token.
Feature activation+0.000
Base
Token Base
Feature activation+0.000
el
Tokenel
Feature activation+0.000
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
sold
Token sold
Feature activation+0.000
to
Token to
Feature activation+0.000
countries
Token countries
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Security
Token Security
Feature activation+0.000
Council
Token Council
Feature activation+0.000
,
Token,
Feature activation+0.000
like
Token like
Feature activation+0.000
South
Token South
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
month
Token month
Feature activation+0.000
's
Token's
Feature activation+0.000
meeting
Token meeting
Feature activation+0.000
is
Token is
Feature activation+0.000
scheduled
Token scheduled
Feature activation+0.000
for
Token for
Feature activation+0.000
May
Token May
Feature activation+0.000
30
Token 30
Feature activation+0.000
th
Tokenth
Feature activation+0.000
e
Tokene
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
others
Token others
Feature activation+0.000
.
Token.
Feature activation+0.000
Young
Token Young
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
st
Token st
Feature activation+0.000
oried
Tokenoried
Feature activation+0.000
rivalry
Token rivalry
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 5: Ultra low frequency cluster

TOP ACTIVATIONS
MAX = 0.281

by
Token by
Feature activation+0.000
pressing
Token pressing
Feature activation+0.000
and
Token and
Feature activation+0.000
holding
Token holding
Feature activation+0.000
(
Token (
Feature activation+0.000
U
TokenU
Feature activation+0.281
)
Token)
Feature activation+0.000
and
Token and
Feature activation+0.000
vice
Token vice
Feature activation+0.000
versa
Token versa
Feature activation+0.000
.
Token.
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
of
Token of
Feature activation+0.000
arrests
Token arrests
Feature activation+0.000
and
Token and
Feature activation+0.053
enforced
Token enforced
Feature activation+0.000
disappear
Token disappear
Feature activation+0.000
ances
Tokenances
Feature activation+0.000
of
Token of
Feature activation+0.000
activists
Token activists
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000

Top DFA by src position
MAX = 3.693

pressing
Token pressing
Feature activation+0.009
his
Token his
Feature activation-0.009
Special
Token Special
Feature activation+0.010
(
Token (
Feature activation+0.017
K
TokenK
Feature activation+0.008
),
Token),
Feature activation+3.693
he
Token he
Feature activation+0.004
can
Token can
Feature activation-0.008
pick
Token pick
Feature activation+0.010
up
Token up
Feature activation+0.002
things
Token things
Feature activation-0.002
midst
Token midst
Feature activation+0.113
of
Token of
Feature activation+0.113
a
Token a
Feature activation+0.096
campaign
Token campaign
Feature activation+0.220
of
Token of
Feature activation+0.181
arrests
Token arrests
Feature activation+2.104
and
Token and
Feature activation+0.082
enforced
Token enforced
Feature activation+0.000
disappear
Token disappear
Feature activation+0.000
ances
Tokenances
Feature activation+0.000
of
Token of
Feature activation+0.000
Jays
Token Jays
Feature activation-0.004
âĢ
TokenâĢ
Feature activation+0.027
Ļ
TokenĻ
Feature activation+0.016
baseball
Token baseball
Feature activation-0.042
cap
Token cap
Feature activation+0.044
and
Token and
Feature activation+0.070
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Put
Token Put
Feature activation+0.002
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.032
Blue
Token Blue
Feature activation+0.012
Jays
Token Jays
Feature activation+0.038
âĢ
TokenâĢ
Feature activation+0.078
Ļ
TokenĻ
Feature activation+0.029
baseball
Token baseball
Feature activation+0.073
cap
Token cap
Feature activation+0.067
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
,
Token,
Feature activation-0.284
the
Token the
Feature activation-0.177
NB
Token NB
Feature activation-0.084
Space
Token Space
Feature activation-0.011
Race
Token Race
Feature activation-0.008
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.482
,
Token,
Feature activation-0.432
the
Token the
Feature activation-0.181
NB
Token NB
Feature activation-0.063
Space
Token Space
Feature activation-0.019
Race
Token Race
Feature activation+0.062
had
Token had
Feature activation+0.022
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
Put
Token Put
Feature activation+0.045
a
Token a
Feature activation-0.002
Toronto
Token Toronto
Feature activation-0.025
Blue
Token Blue
Feature activation+0.022
Jays
Token Jays
Feature activation-0.011
âĢ
TokenâĢ
Feature activation+0.175
Ļ
TokenĻ
Feature activation-0.018
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Put
Token Put
Feature activation-0.017
a
Token a
Feature activation+0.021
Toronto
Token Toronto
Feature activation-0.006
Blue
Token Blue
Feature activation+0.024
Jays
Token Jays
Feature activation-0.036
âĢ
TokenâĢ
Feature activation+0.807
Ļ
TokenĻ
Feature activation+0.022
baseball
Token baseball
Feature activation-0.031
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
lofty
Token lofty
Feature activation-0.000
goal
Token goal
Feature activation-0.005
:
Token:
Feature activation+0.002
Put
Token Put
Feature activation+0.001
a
Token a
Feature activation+0.021
Toronto
Token Toronto
Feature activation+0.142
Blue
Token Blue
Feature activation-0.011
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
lofty
Token lofty
Feature activation+0.004
goal
Token goal
Feature activation-0.005
:
Token:
Feature activation+0.009
Put
Token Put
Feature activation-0.050
a
Token a
Feature activation-0.041
Toronto
Token Toronto
Feature activation+0.117
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
,
Token,
Feature activation-0.275
the
Token the
Feature activation-0.093
NB
Token NB
Feature activation-0.031
Space
Token Space
Feature activation-0.025
Race
Token Race
Feature activation+0.033
had
Token had
Feature activation+0.149
a
Token a
Feature activation+0.002
somewhat
Token somewhat
Feature activation-0.035
less
Token less
Feature activation-0.169
lofty
Token lofty
Feature activation-0.047
goal
Token goal
Feature activation-0.182
,
Token,
Feature activation-0.433
the
Token the
Feature activation-0.038
NB
Token NB
Feature activation-0.014
Space
Token Space
Feature activation+0.002
Race
Token Race
Feature activation+0.049
had
Token had
Feature activation+0.108
a
Token a
Feature activation+0.022
somewhat
Token somewhat
Feature activation+0.015
less
Token less
Feature activation-0.094
lofty
Token lofty
Feature activation-0.034
goal
Token goal
Feature activation-0.150
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation-0.017
:
Token:
Feature activation+0.010
Put
Token Put
Feature activation-0.043
a
Token a
Feature activation+0.024
Toronto
Token Toronto
Feature activation+0.391
Blue
Token Blue
Feature activation+0.131
Jays
Token Jays
Feature activation-0.037
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
:
Token:
Feature activation+0.002
Put
Token Put
Feature activation-0.012
a
Token a
Feature activation+0.001
Toronto
Token Toronto
Feature activation-0.000
Blue
Token Blue
Feature activation-0.004
Jays
Token Jays
Feature activation+0.121
âĢ
TokenâĢ
Feature activation+0.030
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation-0.194
the
Token the
Feature activation-0.132
NB
Token NB
Feature activation-0.047
Space
Token Space
Feature activation+0.004
Race
Token Race
Feature activation-0.015
had
Token had
Feature activation+0.195
a
Token a
Feature activation+0.049
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
,
Token,
Feature activation-0.125
the
Token the
Feature activation-0.125
NB
Token NB
Feature activation-0.022
Space
Token Space
Feature activation-0.009
Race
Token Race
Feature activation-0.003
had
Token had
Feature activation+0.134
a
Token a
Feature activation+0.092
somewhat
Token somewhat
Feature activation+0.084
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.021
less
Token less
Feature activation-0.040
lofty
Token lofty
Feature activation-0.026
goal
Token goal
Feature activation-0.011
:
Token:
Feature activation+0.056
Put
Token Put
Feature activation+0.144
a
Token a
Feature activation+0.141
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
a
Token a
Feature activation-0.017
somewhat
Token somewhat
Feature activation+0.019
less
Token less
Feature activation-0.010
lofty
Token lofty
Feature activation+0.011
goal
Token goal
Feature activation-0.041
:
Token:
Feature activation+0.080
Put
Token Put
Feature activation+0.029
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
,
Token,
Feature activation-0.200
the
Token the
Feature activation-0.106
NB
Token NB
Feature activation-0.069
Space
Token Space
Feature activation-0.008
Race
Token Race
Feature activation-0.053
had
Token had
Feature activation+0.131
a
Token a
Feature activation+0.026
somewhat
Token somewhat
Feature activation-0.135
less
Token less
Feature activation-0.150
lofty
Token lofty
Feature activation-0.224
goal
Token goal
Feature activation+0.000
,
Token,
Feature activation-0.162
the
Token the
Feature activation-0.091
NB
Token NB
Feature activation-0.031
Space
Token Space
Feature activation-0.007
Race
Token Race
Feature activation-0.008
had
Token had
Feature activation+0.145
a
Token a
Feature activation+0.052
somewhat
Token somewhat
Feature activation+0.035
less
Token less
Feature activation-0.137
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.14

Head 2: 0.09

Head 3: 0.08

Head 4: 0.09

Head 5: 0.07

Head 6: 0.07

Head 7: 0.08

Head 8: 0.09

Head 9: 0.07

Head 10: 0.07

Head 11: 0.07

Positive logits

socket1.89

RIP1.55

1.54

Univ1.51

socket1.51

Socket1.50

Mechdragon1.50

ushes1.50

Net1.48

ucket1.48

poll1.47

roar1.45

heck1.44

ushed1.43

sts1.43

ridor1.42

Roche1.41

net1.40

exploded1.39

dashed1.39

Negative logits

omorphic-2.06

thood-1.70

opathy-1.70

iaries-1.59

assad-1.59

omorph-1.56

mathemat-1.55

othermal-1.53

ancock-1.51

archaeological-1.49

iability-1.46

eBook-1.43

alleg-1.42

cul-1.42

languages-1.41

bapt-1.41

Languages-1.40

princip-1.40

ale-1.39

opathic-1.38

INTERVAL 0.253 - 0.281
CONTAINS 0.000%

by
Token by
Feature activation+0.000
pressing
Token pressing
Feature activation+0.000
and
Token and
Feature activation+0.000
holding
Token holding
Feature activation+0.000
(
Token (
Feature activation+0.000
U
TokenU
Feature activation+0.281
)
Token)
Feature activation+0.000
and
Token and
Feature activation+0.000
vice
Token vice
Feature activation+0.000
versa
Token versa
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 0.225 - 0.253
CONTAINS 0.000%

INTERVAL 0.197 - 0.225
CONTAINS 0.000%

INTERVAL 0.169 - 0.197
CONTAINS 0.000%

INTERVAL 0.141 - 0.169
CONTAINS 0.000%

INTERVAL 0.113 - 0.141
CONTAINS 0.000%

INTERVAL 0.084 - 0.113
CONTAINS 0.000%

INTERVAL 0.056 - 0.084
CONTAINS 0.000%

INTERVAL 0.028 - 0.056
CONTAINS 0.000%

of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
campaign
Token campaign
Feature activation+0.000
of
Token of
Feature activation+0.000
arrests
Token arrests
Feature activation+0.000
and
Token and
Feature activation+0.053
enforced
Token enforced
Feature activation+0.000
disappear
Token disappear
Feature activation+0.000
ances
Tokenances
Feature activation+0.000
of
Token of
Feature activation+0.000
activists
Token activists
Feature activation+0.000

INTERVAL 0.000 - 0.028
CONTAINS 100.000%

12
Token 12
Feature activation+0.000
GB
TokenGB
Feature activation+0.000
for
Token for
Feature activation+0.000
$
Token $
Feature activation+0.000
80
Token80
Feature activation+0.000
per
Token per
Feature activation+0.000
month
Token month
Feature activation+0.000
.
Token.
Feature activation+0.000
Adding
Token Adding
Feature activation+0.000
a
Token a
Feature activation+0.000
smartphone
Token smartphone
Feature activation+0.000
document
Token document
Feature activation+0.000
became
Token became
Feature activation+0.000
known
Token known
Feature activation+0.000
as
Token as
Feature activation+0.000
The
Token The
Feature activation+0.000
Swiss
Token Swiss
Feature activation+0.000
Statement
Token Statement
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Wh
TokenWh
Feature activation+0.000
8
Token 8
Feature activation+0.000
35
Token35
Feature activation+0.000
chipset
Token chipset
Feature activation+0.000
is
Token is
Feature activation+0.000
pretty
Token pretty
Feature activation+0.000
good
Token good
Feature activation+0.000
already
Token already
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Software
TokenSoftware
Feature activation+0.000
fans
Token fans
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
vast
Token vast
Feature activation+0.000
majority
Token majority
Feature activation+0.000
will
Token will
Feature activation+0.000
be
Token be
Feature activation+0.000
supporting
Token supporting
Feature activation+0.000
the
Token the
Feature activation+0.000
team
Token team
Feature activation+0.000
in
Token in
Feature activation+0.000
environment
Token environment
Feature activation+0.000
and
Token and
Feature activation+0.000
state
Token state
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
call
Token call
Feature activation+0.000
stack
Token stack
Feature activation+0.000
(
Token (
Feature activation+0.000
You
TokenYou
Feature activation+0.000
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 6: In local text about biases

TOP ACTIVATIONS
MAX = 5.443

omission
Token omission
Feature activation+0.000
of
Token of
Feature activation+0.000
these
Token these
Feature activation+0.000
items
Token items
Feature activation+0.000
biases
Token biases
Feature activation+2.681
the
Token the
Feature activation+5.443
estimated
Token estimated
Feature activation+0.047
savings
Token savings
Feature activation+0.000
downward
Token downward
Feature activation+0.741
.
Token.
Feature activation+0.411
Ċ
TokenĊ
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
benefited
Token benefited
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.122
in
Token in
Feature activation+4.893
both
Token both
Feature activation+2.382
scope
Token scope
Feature activation+1.691
and
Token and
Feature activation+0.000
tone
Token tone
Feature activation+0.000
of
Token of
Feature activation+0.184
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
these
Token these
Feature activation+0.000
probably
Token probably
Feature activation+0.000
bias
Token bias
Feature activation+0.956
the
Token the
Feature activation+4.834
estimated
Token estimated
Feature activation+0.247
expenditure
Token expenditure
Feature activation+0.000
reductions
Token reductions
Feature activation+0.000
and
Token and
Feature activation+0.000
tax
Token tax
Feature activation+0.000
systematic
Token systematic
Feature activation+0.000
gender
Token gender
Feature activation+0.000
bias
Token bias
Feature activation+0.971
in
Token in
Feature activation+4.468
the
Token the
Feature activation+4.275
way
Token way
Feature activation+4.662
that
Token that
Feature activation+0.426
we
Token we
Feature activation+0.770
do
Token do
Feature activation+0.727
annual
Token annual
Feature activation+0.000
merit
Token merit
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
a
Tokena
Feature activation+0.000
deliberate
Token deliberate
Feature activation+0.000
bias
Token bias
Feature activation+1.030
against
Token against
Feature activation+4.601
the
Token the
Feature activation+4.146
department
Token department
Feature activation+1.421
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
say
Token say
Feature activation+0.000
it
Token it
Feature activation+0.000
has
Token has
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+1.015
in
Token in
Feature activation+4.496
a
Token a
Feature activation+2.764
particular
Token particular
Feature activation+2.770
direction
Token direction
Feature activation+1.054
.
Token.
Feature activation+0.099
When
Token When
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
systematic
Token systematic
Feature activation+0.000
gender
Token gender
Feature activation+0.000
bias
Token bias
Feature activation+0.971
in
Token in
Feature activation+4.468
the
Token the
Feature activation+4.275
way
Token way
Feature activation+4.662
that
Token that
Feature activation+0.426
we
Token we
Feature activation+0.770
do
Token do
Feature activation+0.727
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.622
within
Token within
Feature activation+4.111
some
Token some
Feature activation+0.000
journals
Token journals
Feature activation+0.000
towards
Token towards
Feature activation+4.366
publishing
Token publishing
Feature activation+0.000
papers
Token papers
Feature activation+0.000
by
Token by
Feature activation+0.000
faculty
Token faculty
Feature activation+0.000
from
Token from
Feature activation+0.000
at
Token at
Feature activation+0.310
play
Token play
Feature activation+0.000
here
Token here
Feature activation+0.000
,
Token,
Feature activation+0.978
too
Token too
Feature activation+0.000
:
Token:
Feature activation+4.346
A
Token A
Feature activation+1.183
2015
Token 2015
Feature activation+0.000
paper
Token paper
Feature activation+0.000
in
Token in
Feature activation+0.000
Organization
Token Organization
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Media
TokenMedia
Feature activation+0.000
coverage
Token coverage
Feature activation+0.000
is
Token is
Feature activation+0.000
biased
Token biased
Feature activation+0.037
against
Token against
Feature activation+4.318
facts
Token facts
Feature activation+0.735
and
Token and
Feature activation+0.000
issues
Token issues
Feature activation+0.000
.
Token.
Feature activation+0.008
The
Token The
Feature activation+0.000
a
Token a
Feature activation+0.000
systematic
Token systematic
Feature activation+0.000
gender
Token gender
Feature activation+0.000
bias
Token bias
Feature activation+0.971
in
Token in
Feature activation+4.468
the
Token the
Feature activation+4.275
way
Token way
Feature activation+4.662
that
Token that
Feature activation+0.426
we
Token we
Feature activation+0.770
do
Token do
Feature activation+0.727
annual
Token annual
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
a
Tokena
Feature activation+0.000
deliberate
Token deliberate
Feature activation+0.000
bias
Token bias
Feature activation+1.030
against
Token against
Feature activation+4.601
the
Token the
Feature activation+4.146
department
Token department
Feature activation+1.421
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
She
Token She
Feature activation+0.000
of
Token of
Feature activation+0.000
Health
Token Health
Feature activation+0.000
testing
Token testing
Feature activation+0.000
was
Token was
Feature activation+0.000
biased
Token biased
Feature activation+0.148
in
Token in
Feature activation+4.144
favor
Token favor
Feature activation+3.832
of
Token of
Feature activation+2.487
prosecutors
Token prosecutors
Feature activation+0.000
and
Token and
Feature activation+0.360
that
Token that
Feature activation+0.000
believed
Token believed
Feature activation+0.000
the
Token the
Feature activation+0.000
media
Token media
Feature activation+0.000
was
Token was
Feature activation+0.000
biased
Token biased
Feature activation+1.080
against
Token against
Feature activation+4.144
Trump
Token Trump
Feature activation+0.421
.
Token.
Feature activation+0.000
Numbers
Token Numbers
Feature activation+0.000
like
Token like
Feature activation+0.000
that
Token that
Feature activation+0.000
show
Token show
Feature activation+0.000
evidence
Token evidence
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.622
within
Token within
Feature activation+4.111
some
Token some
Feature activation+0.000
journals
Token journals
Feature activation+0.000
towards
Token towards
Feature activation+4.366
publishing
Token publishing
Feature activation+0.000
papers
Token papers
Feature activation+0.000
training
Token training
Feature activation+0.000
in
Token in
Feature activation+0.000
dealing
Token dealing
Feature activation+0.000
with
Token with
Feature activation+0.000
biases
Token biases
Feature activation+0.000
toward
Token toward
Feature activation+4.061
other
Token other
Feature activation+1.243
communities
Token communities
Feature activation+0.000
,
Token,
Feature activation+0.000
crisis
Token crisis
Feature activation+0.000
intervention
Token intervention
Feature activation+0.000
sure
Token sure
Feature activation+0.000
they
Token they
Feature activation+0.000
're
Token're
Feature activation+0.000
not
Token not
Feature activation+0.000
biased
Token biased
Feature activation+0.011
against
Token against
Feature activation+4.028
the
Token the
Feature activation+3.858
accused
Token accused
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
I
TokenI
Feature activation+0.000
'm
Token'm
Feature activation+0.000
a
Token a
Feature activation+0.000
little
Token little
Feature activation+0.000
biased
Token biased
Feature activation+0.589
because
Token because
Feature activation+3.984
I
Token I
Feature activation+2.165
've
Token've
Feature activation+0.000
known
Token known
Feature activation+0.000
her
Token her
Feature activation+0.000
for
Token for
Feature activation+0.000
said
Token said
Feature activation+0.000
minorities
Token minorities
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.000
against
Token against
Feature activation+3.897
them
Token them
Feature activation+0.847
that
Token that
Feature activation+1.117
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
taught
Token taught
Feature activation+0.000
they
Token they
Feature activation+0.000
're
Token're
Feature activation+0.000
not
Token not
Feature activation+0.000
biased
Token biased
Feature activation+0.011
against
Token against
Feature activation+4.028
the
Token the
Feature activation+3.858
accused
Token accused
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000

Top DFA by src position
MAX = 7.218

The
Token The
Feature activation-0.023
omission
Token omission
Feature activation+0.064
of
Token of
Feature activation-0.002
these
Token these
Feature activation-0.006
items
Token items
Feature activation-0.021
biases
Token biases
Feature activation+7.218
the
Token the
Feature activation+0.070
estimated
Token estimated
Feature activation+0.000
savings
Token savings
Feature activation+0.000
downward
Token downward
Feature activation+0.000
.
Token.
Feature activation+0.000
.
Token.
Feature activation-0.001
Trump
Token Trump
Feature activation-0.007
benefited
Token benefited
Feature activation-0.014
from
Token from
Feature activation-0.186
a
Token a
Feature activation+0.007
bias
Token bias
Feature activation+6.654
in
Token in
Feature activation+0.221
both
Token both
Feature activation+0.000
scope
Token scope
Feature activation+0.000
and
Token and
Feature activation+0.000
tone
Token tone
Feature activation+0.000
assumptions
Token assumptions
Feature activation+0.090
,
Token,
Feature activation+0.036
but
Token but
Feature activation-0.018
these
Token these
Feature activation-0.039
probably
Token probably
Feature activation+0.109
bias
Token bias
Feature activation+6.520
the
Token the
Feature activation-0.035
estimated
Token estimated
Feature activation+0.000
expenditure
Token expenditure
Feature activation+0.000
reductions
Token reductions
Feature activation+0.000
and
Token and
Feature activation+0.000
evidence
Token evidence
Feature activation+0.151
of
Token of
Feature activation+0.081
a
Token a
Feature activation+0.125
systematic
Token systematic
Feature activation+0.137
gender
Token gender
Feature activation+0.201
bias
Token bias
Feature activation+4.140
in
Token in
Feature activation+1.289
the
Token the
Feature activation+0.152
way
Token way
Feature activation-0.093
that
Token that
Feature activation+0.000
we
Token we
Feature activation+0.000
reflected
Token reflected
Feature activation+0.041
âĢ
Token âĢ
Feature activation+0.064
ľ
Tokenľ
Feature activation+0.005
a
Tokena
Feature activation+0.163
deliberate
Token deliberate
Feature activation+0.139
bias
Token bias
Feature activation+4.923
against
Token against
Feature activation+0.554
the
Token the
Feature activation+0.000
department
Token department
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
to
Token to
Feature activation+0.055
say
Token say
Feature activation+0.050
it
Token it
Feature activation+0.046
has
Token has
Feature activation-0.014
a
Token a
Feature activation+0.077
bias
Token bias
Feature activation+5.813
in
Token in
Feature activation+0.244
a
Token a
Feature activation+0.000
particular
Token particular
Feature activation+0.000
direction
Token direction
Feature activation+0.000
.
Token.
Feature activation+0.000
evidence
Token evidence
Feature activation+0.168
of
Token of
Feature activation+0.032
a
Token a
Feature activation+0.101
systematic
Token systematic
Feature activation+0.123
gender
Token gender
Feature activation+0.207
bias
Token bias
Feature activation+5.120
in
Token in
Feature activation+0.216
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
that
Token that
Feature activation+0.000
we
Token we
Feature activation+0.000
findings
Token findings
Feature activation-0.004
show
Token show
Feature activation-0.002
evidence
Token evidence
Feature activation+0.048
of
Token of
Feature activation-0.000
a
Token a
Feature activation+0.046
bias
Token bias
Feature activation+5.473
within
Token within
Feature activation+0.462
some
Token some
Feature activation+0.027
journals
Token journals
Feature activation+0.009
towards
Token towards
Feature activation+0.158
publishing
Token publishing
Feature activation+0.000
There
TokenThere
Feature activation+0.052
âĢ
TokenâĢ
Feature activation-0.014
Ļ
TokenĻ
Feature activation-0.011
s
Tokens
Feature activation+0.047
another
Token another
Feature activation+0.162
bias
Token bias
Feature activation+5.496
at
Token at
Feature activation+0.053
play
Token play
Feature activation-0.004
here
Token here
Feature activation+0.051
,
Token,
Feature activation+0.032
too
Token too
Feature activation+0.043
Ċ
TokenĊ
Feature activation+0.022
Ċ
TokenĊ
Feature activation-0.016
Media
TokenMedia
Feature activation+0.016
coverage
Token coverage
Feature activation+0.063
is
Token is
Feature activation+0.056
biased
Token biased
Feature activation+5.319
against
Token against
Feature activation+0.755
facts
Token facts
Feature activation+0.000
and
Token and
Feature activation+0.000
issues
Token issues
Feature activation+0.000
.
Token.
Feature activation+0.000
evidence
Token evidence
Feature activation+0.163
of
Token of
Feature activation+0.036
a
Token a
Feature activation+0.064
systematic
Token systematic
Feature activation+0.124
gender
Token gender
Feature activation+0.158
bias
Token bias
Feature activation+4.630
in
Token in
Feature activation+0.626
the
Token the
Feature activation+0.200
way
Token way
Feature activation+0.000
that
Token that
Feature activation+0.000
we
Token we
Feature activation+0.000
reflected
Token reflected
Feature activation+0.082
âĢ
Token âĢ
Feature activation+0.042
ľ
Tokenľ
Feature activation+0.005
a
Tokena
Feature activation+0.086
deliberate
Token deliberate
Feature activation+0.153
bias
Token bias
Feature activation+3.736
against
Token against
Feature activation+1.500
the
Token the
Feature activation+0.095
department
Token department
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Department
Token Department
Feature activation+0.031
of
Token of
Feature activation-0.004
Health
Token Health
Feature activation+0.002
testing
Token testing
Feature activation+0.001
was
Token was
Feature activation+0.123
biased
Token biased
Feature activation+5.290
in
Token in
Feature activation+0.233
favor
Token favor
Feature activation+0.000
of
Token of
Feature activation+0.000
prosecutors
Token prosecutors
Feature activation+0.000
and
Token and
Feature activation+0.000
they
Token they
Feature activation+0.057
believed
Token believed
Feature activation-0.002
the
Token the
Feature activation+0.036
media
Token media
Feature activation+0.035
was
Token was
Feature activation+0.129
biased
Token biased
Feature activation+4.646
against
Token against
Feature activation+0.892
Trump
Token Trump
Feature activation+0.000
.
Token.
Feature activation+0.000
Numbers
Token Numbers
Feature activation+0.000
like
Token like
Feature activation+0.000
findings
Token findings
Feature activation-0.021
show
Token show
Feature activation-0.000
evidence
Token evidence
Feature activation+0.102
of
Token of
Feature activation+0.007
a
Token a
Feature activation+0.082
bias
Token bias
Feature activation+5.690
within
Token within
Feature activation+0.156
some
Token some
Feature activation+0.000
journals
Token journals
Feature activation+0.000
towards
Token towards
Feature activation+0.000
publishing
Token publishing
Feature activation+0.000
get
Token get
Feature activation-0.030
training
Token training
Feature activation+0.046
in
Token in
Feature activation-0.024
dealing
Token dealing
Feature activation-0.019
with
Token with
Feature activation-0.104
biases
Token biases
Feature activation+5.418
toward
Token toward
Feature activation+0.675
other
Token other
Feature activation+0.000
communities
Token communities
Feature activation+0.000
,
Token,
Feature activation+0.000
crisis
Token crisis
Feature activation+0.000
make
Token make
Feature activation-0.020
sure
Token sure
Feature activation-0.038
they
Token they
Feature activation+0.072
're
Token're
Feature activation+0.018
not
Token not
Feature activation+0.059
biased
Token biased
Feature activation+4.538
against
Token against
Feature activation+0.837
the
Token the
Feature activation+0.000
accused
Token accused
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation-0.002
I
TokenI
Feature activation+0.076
'm
Token'm
Feature activation+0.051
a
Token a
Feature activation+0.013
little
Token little
Feature activation+0.191
biased
Token biased
Feature activation+5.436
because
Token because
Feature activation+0.236
I
Token I
Feature activation+0.000
've
Token've
Feature activation+0.000
known
Token known
Feature activation+0.000
her
Token her
Feature activation+0.000
Officers
TokenOfficers
Feature activation-0.008
said
Token said
Feature activation-0.015
minorities
Token minorities
Feature activation-0.013
had
Token had
Feature activation-0.009
a
Token a
Feature activation+0.051
bias
Token bias
Feature activation+5.755
against
Token against
Feature activation+0.237
them
Token them
Feature activation+0.000
that
Token that
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
make
Token make
Feature activation-0.011
sure
Token sure
Feature activation-0.022
they
Token they
Feature activation+0.024
're
Token're
Feature activation-0.000
not
Token not
Feature activation+0.015
biased
Token biased
Feature activation+3.586
against
Token against
Feature activation+1.736
the
Token the
Feature activation+0.146
accused
Token accused
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.01

Head 2: 0.07

Head 3: 0.06

Head 4: 0.12

Head 5: 0.03

Head 6: 0.04

Head 7: 0.38

Head 8: 0.03

Head 9: 0.03

Head 10: 0.12

Head 11: 0.08

Positive logits

favor1.70

ethn1.67

biased1.65

favour1.61

favoring1.60

prejud1.57

towards1.57

partisan1.56

toward1.55

bias1.52

impartial1.50

demographics1.49

viewpoint1.48

unbiased1.48

partisans1.47

ideologically1.46

interpreting1.45

Ide1.43

channelAvailability1.42

biases1.41

Negative logits

erial-1.62

eat-1.54

raq-1.52

break-1.47

angered-1.45

prus-1.45

rip-1.45

amaz-1.44

onel-1.44

fly-1.41

miss-1.41

ydia-1.40

birds-1.40

pit-1.39

leeve-1.37

ruption-1.35

tackle-1.35

noon-1.34

iri-1.33

dr-1.33

INTERVAL 4.898 - 5.443
CONTAINS 0.000%

omission
Token omission
Feature activation+0.000
of
Token of
Feature activation+0.000
these
Token these
Feature activation+0.000
items
Token items
Feature activation+0.000
biases
Token biases
Feature activation+2.681
the
Token the
Feature activation+5.443
estimated
Token estimated
Feature activation+0.047
savings
Token savings
Feature activation+0.000
downward
Token downward
Feature activation+0.741
.
Token.
Feature activation+0.411
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 4.354 - 4.898
CONTAINS 0.000%

of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
systematic
Token systematic
Feature activation+0.000
gender
Token gender
Feature activation+0.000
bias
Token bias
Feature activation+0.971
in
Token in
Feature activation+4.468
the
Token the
Feature activation+4.275
way
Token way
Feature activation+4.662
that
Token that
Feature activation+0.426
we
Token we
Feature activation+0.770
do
Token do
Feature activation+0.727
systematic
Token systematic
Feature activation+0.000
gender
Token gender
Feature activation+0.000
bias
Token bias
Feature activation+0.971
in
Token in
Feature activation+4.468
the
Token the
Feature activation+4.275
way
Token way
Feature activation+4.662
that
Token that
Feature activation+0.426
we
Token we
Feature activation+0.770
do
Token do
Feature activation+0.727
annual
Token annual
Feature activation+0.000
merit
Token merit
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.622
within
Token within
Feature activation+4.111
some
Token some
Feature activation+0.000
journals
Token journals
Feature activation+0.000
towards
Token towards
Feature activation+4.366
publishing
Token publishing
Feature activation+0.000
papers
Token papers
Feature activation+0.000
by
Token by
Feature activation+0.000
faculty
Token faculty
Feature activation+0.000
from
Token from
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
a
Tokena
Feature activation+0.000
deliberate
Token deliberate
Feature activation+0.000
bias
Token bias
Feature activation+1.030
against
Token against
Feature activation+4.601
the
Token the
Feature activation+4.146
department
Token department
Feature activation+1.421
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
benefited
Token benefited
Feature activation+0.000
from
Token from
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.122
in
Token in
Feature activation+4.893
both
Token both
Feature activation+2.382
scope
Token scope
Feature activation+1.691
and
Token and
Feature activation+0.000
tone
Token tone
Feature activation+0.000
of
Token of
Feature activation+0.184

INTERVAL 3.810 - 4.354
CONTAINS 0.000%

show
Token show
Feature activation+0.000
evidence
Token evidence
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.622
within
Token within
Feature activation+4.111
some
Token some
Feature activation+0.000
journals
Token journals
Feature activation+0.000
towards
Token towards
Feature activation+4.366
publishing
Token publishing
Feature activation+0.000
papers
Token papers
Feature activation+0.000
believed
Token believed
Feature activation+0.000
the
Token the
Feature activation+0.000
media
Token media
Feature activation+0.000
was
Token was
Feature activation+0.000
biased
Token biased
Feature activation+1.080
against
Token against
Feature activation+4.144
Trump
Token Trump
Feature activation+0.421
.
Token.
Feature activation+0.000
Numbers
Token Numbers
Feature activation+0.000
like
Token like
Feature activation+0.000
that
Token that
Feature activation+0.000
Health
Token Health
Feature activation+0.000
testing
Token testing
Feature activation+0.000
was
Token was
Feature activation+0.000
biased
Token biased
Feature activation+0.148
in
Token in
Feature activation+4.144
favor
Token favor
Feature activation+3.832
of
Token of
Feature activation+2.487
prosecutors
Token prosecutors
Feature activation+0.000
and
Token and
Feature activation+0.360
that
Token that
Feature activation+0.000
staff
Token staff
Feature activation+0.000
land
Tokenland
Feature activation+0.000
WikiLeaks
Token WikiLeaks
Feature activation+0.000
emails
Token emails
Feature activation+0.000
show
Token show
Feature activation+0.000
bias
Token bias
Feature activation+0.000
toward
Token toward
Feature activation+3.824
Clinton
Token Clinton
Feature activation+1.268
over
Token over
Feature activation+1.522
Sanders
Token Sanders
Feature activation+0.000
--
Token --
Feature activation+0.000
and
Token and
Feature activation+0.000
said
Token said
Feature activation+0.000
minorities
Token minorities
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
bias
Token bias
Feature activation+0.000
against
Token against
Feature activation+3.897
them
Token them
Feature activation+0.847
that
Token that
Feature activation+1.117
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
taught
Token taught
Feature activation+0.000

INTERVAL 3.266 - 3.810
CONTAINS 0.000%

by
Token by
Feature activation+0.000
pointing
Token pointing
Feature activation+0.000
out
Token out
Feature activation+0.000
bias
Token bias
Feature activation+0.000
in
Token in
Feature activation+3.459
the
Token the
Feature activation+3.399
data
Token data
Feature activation+0.126
=================================
Token =================================
Feature activation+0.000
================
Token================
Feature activation+0.000
====
Token====
Feature activation+0.000
===
Token===
Feature activation+0.000
That
Token That
Feature activation+0.000
introduces
Token introduces
Feature activation+0.000
an
Token an
Feature activation+0.000
enormous
Token enormous
Feature activation+0.000
bias
Token bias
Feature activation+0.351
in
Token in
Feature activation+3.769
favor
Token favor
Feature activation+3.569
of
Token of
Feature activation+2.018
militar
Token militar
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
and
Token and
Feature activation+0.000
again
Token again
Feature activation+0.000
by
Token by
Feature activation+0.000
pointing
Token pointing
Feature activation+0.000
out
Token out
Feature activation+0.000
bias
Token bias
Feature activation+0.000
in
Token in
Feature activation+3.459
the
Token the
Feature activation+3.399
data
Token data
Feature activation+0.126
=================================
Token =================================
Feature activation+0.000
================
Token================
Feature activation+0.000
====
Token====
Feature activation+0.000
can
Token can
Feature activation+0.000
certainly
Token certainly
Feature activation+0.000
indicate
Token indicate
Feature activation+0.000
some
Token some
Feature activation+0.000
bias
Token bias
Feature activation+0.000
in
Token in
Feature activation+3.559
the
Token the
Feature activation+3.738
reporters
Token reporters
Feature activation+0.380
themselves
Token themselves
Feature activation+0.877
.
Token.
Feature activation+0.374
Fortunately
Token Fortunately
Feature activation+0.000
hit
Token hit
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
media
Token media
Feature activation+0.000
bias
Token bias
Feature activation+0.000
against
Token against
Feature activation+3.461
discussing
Token discussing
Feature activation+0.000
issues
Token issues
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 2.721 - 3.266
CONTAINS 0.000%

if
Token if
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
skewed
Token skewed
Feature activation+0.000
in
Token in
Feature activation+3.513
one
Token one
Feature activation+3.158
direction
Token direction
Feature activation+3.261
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
is
Token is
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
skewed
Token skewed
Feature activation+0.190
in
Token in
Feature activation+3.143
two
Token two
Feature activation+2.590
ways
Token ways
Feature activation+2.464
:
Token:
Feature activation+0.000
1
Token 1
Feature activation+0.000
)
Token)
Feature activation+0.000
Crawford
Token Crawford
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
ske
Token ske
Feature activation+0.000
ws
Tokenws
Feature activation+0.000
your
Token your
Feature activation+2.921
evaluation
Token evaluation
Feature activation+1.040
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
less
Token less
Feature activation+0.000
-
Token-
Feature activation+0.000
hybrid
Token hybrid
Feature activation+0.000
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
ske
Token ske
Feature activation+0.000
ws
Tokenws
Feature activation+0.766
the
Token the
Feature activation+3.230
data
Token data
Feature activation+1.894
because
Token because
Feature activation+0.601
analysts
Token analysts
Feature activation+0.000
aren
Token aren
Feature activation+0.000
't
Token't
Feature activation+0.000
an
Token an
Feature activation+0.000
institutional
Token institutional
Feature activation+0.000
ized
Tokenized
Feature activation+0.000
gender
Token gender
Feature activation+0.000
bias
Token bias
Feature activation+0.000
in
Token in
Feature activation+2.789
FIFA
Token FIFA
Feature activation+0.000
's
Token's
Feature activation+0.000
participating
Token participating
Feature activation+0.000
countries
Token countries
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 2.177 - 2.721
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
News
TokenNews
Feature activation+0.000
bias
Token bias
Feature activation+0.000
on
Token on
Feature activation+1.813
the
Token the
Feature activation+2.506
social
Token social
Feature activation+0.000
network
Token network
Feature activation+0.000
could
Token could
Feature activation+0.000
have
Token have
Feature activation+0.000
dramatic
Token dramatic
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
skewed
Token skewed
Feature activation+0.190
in
Token in
Feature activation+3.143
two
Token two
Feature activation+2.590
ways
Token ways
Feature activation+2.464
:
Token:
Feature activation+0.000
1
Token 1
Feature activation+0.000
)
Token)
Feature activation+0.000
it
Token it
Feature activation+0.000
suggesting
Token suggesting
Feature activation+0.000
a
Token a
Feature activation+0.000
strategic
Token strategic
Feature activation+0.000
bias
Token bias
Feature activation+0.077
directed
Token directed
Feature activation+1.853
against
Token against
Feature activation+2.485
Black
Token Black
Feature activation+0.000
recipients
Token recipients
Feature activation+0.000
rather
Token rather
Feature activation+0.223
than
Token than
Feature activation+0.285
in
Token in
Feature activation+1.260
bias
Token bias
Feature activation+0.000
is
Token is
Feature activation+0.274
a
Token a
Feature activation+0.852
conscious
Token conscious
Feature activation+0.499
bias
Token bias
Feature activation+0.730
about
Token about
Feature activation+2.631
certain
Token certain
Feature activation+1.586
populations
Token populations
Feature activation+0.000
based
Token based
Feature activation+1.847
upon
Token upon
Feature activation+0.506
race
Token race
Feature activation+0.000
it
Token it
Feature activation+0.000
has
Token has
Feature activation+0.000
an
Token an
Feature activation+0.000
implicit
Token implicit
Feature activation+0.000
bias
Token bias
Feature activation+0.710
for
Token for
Feature activation+2.410
neutrality
Token neutrality
Feature activation+1.044
,
Token,
Feature activation+1.248
and
Token and
Feature activation+0.024
the
Token the
Feature activation+0.011
main
Token main
Feature activation+0.000

INTERVAL 1.633 - 2.177
CONTAINS 0.001%

suburban
Token suburban
Feature activation+0.000
district
Token district
Feature activation+0.000
that
Token that
Feature activation+0.000
skewed
Token skewed
Feature activation+0.000
even
Token even
Feature activation+2.177
more
Token more
Feature activation+1.643
Republican
Token Republican
Feature activation+0.409
after
Token after
Feature activation+0.000
redist
Token redist
Feature activation+0.000
ricting
Tokenricting
Feature activation+0.000
last
Token last
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
My
TokenMy
Feature activation+0.000
obvious
Token obvious
Feature activation+0.000
bias
Token bias
Feature activation+0.000
has
Token has
Feature activation+1.710
lead
Token lead
Feature activation+0.000
me
Token me
Feature activation+0.000
to
Token to
Feature activation+0.000
highlight
Token highlight
Feature activation+0.000
one
Token one
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
biased
Token biased
Feature activation+0.000
due
Token due
Feature activation+0.847
to
Token to
Feature activation+3.249
the
Token the
Feature activation+1.988
presence
Token presence
Feature activation+0.000
of
Token of
Feature activation+0.000
notorious
Token notorious
Feature activation+0.000
human
Token human
Feature activation+0.000
rights
Token rights
Feature activation+0.000
intelligence
Token intelligence
Feature activation+0.000
to
Token to
Feature activation+0.000
fit
Token fit
Feature activation+0.000
the
Token the
Feature activation+0.000
biases
Token biases
Feature activation+0.000
of
Token of
Feature activation+2.161
Bill
Token Bill
Feature activation+0.000
Casey
Token Casey
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
the
Token the
Feature activation+0.000
agency
Token agency
Feature activation+0.000
has
Token has
Feature activation+0.000
already
Token already
Feature activation+0.000
prejud
Token prejud
Feature activation+0.000
ged
Tokenged
Feature activation+1.775
the
Token the
Feature activation+1.608
merits
Token merits
Feature activation+0.000
of
Token of
Feature activation+0.000
its
Token its
Feature activation+0.000
proposal
Token proposal
Feature activation+0.000

INTERVAL 1.089 - 1.633
CONTAINS 0.001%

ulsive
Tokenulsive
Feature activation+0.000
biases
Token biases
Feature activation+0.197
of
Token of
Feature activation+1.825
Fox
Token Fox
Feature activation+0.000
News
Token News
Feature activation+0.000
on
Token on
Feature activation+1.234
display
Token display
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
is
Token is
Feature activation+0.000
incon
Token incon
Feature activation+0.000
the
Token the
Feature activation+0.000
primary
Token primary
Feature activation+0.000
skew
Token skew
Feature activation+0.000
heavily
Token heavily
Feature activation+3.267
toward
Token toward
Feature activation+1.431
an
Token an
Feature activation+1.337
older
Token older
Feature activation+0.000
,
Token,
Feature activation+0.000
wealthier
Token wealthier
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
possess
Token possess
Feature activation+0.000
a
Token a
Feature activation+0.000
Sh
Token Sh
Feature activation+0.000
into
Tokeninto
Feature activation+0.000
bias
Token bias
Feature activation+0.000
due
Token due
Feature activation+1.364
to
Token to
Feature activation+1.233
the
Token the
Feature activation+0.625
fact
Token fact
Feature activation+1.921
none
Token none
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Tokena
Feature activation+0.000
deliberate
Token deliberate
Feature activation+0.000
bias
Token bias
Feature activation+1.030
against
Token against
Feature activation+4.601
the
Token the
Feature activation+4.146
department
Token department
Feature activation+1.421
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
She
Token She
Feature activation+0.000
added
Token added
Feature activation+0.000
has
Token has
Feature activation+0.000
an
Token an
Feature activation+0.000
implicit
Token implicit
Feature activation+0.000
bias
Token bias
Feature activation+0.666
for
Token for
Feature activation+2.388
neutrality
Token neutrality
Feature activation+1.113
,
Token,
Feature activation+1.300
and
Token and
Feature activation+0.055
the
Token the
Feature activation+0.000
main
Token main
Feature activation+0.000
neutral
Token neutral
Feature activation+0.000

INTERVAL 0.544 - 1.089
CONTAINS 0.002%

in
Token in
Feature activation+4.468
the
Token the
Feature activation+4.275
way
Token way
Feature activation+4.662
that
Token that
Feature activation+0.426
we
Token we
Feature activation+0.770
do
Token do
Feature activation+0.727
annual
Token annual
Feature activation+0.000
merit
Token merit
Feature activation+0.000
evaluations
Token evaluations
Feature activation+0.000
,"
Token,"
Feature activation+0.000
she
Token she
Feature activation+0.000
calculation
Token calculation
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
privile
Token privile
Feature activation+0.000
ging
Tokenging
Feature activation+0.656
of
Token of
Feature activation+1.489
presentation
Token presentation
Feature activation+0.000
over
Token over
Feature activation+0.591
substance
Token substance
Feature activation+0.000
,"
Token,"
Feature activation+0.000
ske
Token ske
Feature activation+0.000
wing
Tokenwing
Feature activation+0.570
her
Token her
Feature activation+1.945
results
Token results
Feature activation+0.012
in
Token in
Feature activation+1.167
the
Token the
Feature activation+0.981
game
Token game
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
They
TokenThey
Feature activation+0.000
zero
Token zero
Feature activation+0.000
systematic
Token systematic
Feature activation+0.000
or
Token or
Feature activation+0.000
general
Token general
Feature activation+0.000
judgment
Token judgment
Feature activation+0.000
against
Token against
Feature activation+0.668
infant
Token infant
Feature activation+0.000
formula
Token formula
Feature activation+0.000
or
Token or
Feature activation+0.000
bottle
Token bottle
Feature activation+0.000
feeding
Token feeding
Feature activation+0.000
they
Token they
Feature activation+0.000
believed
Token believed
Feature activation+0.000
the
Token the
Feature activation+0.000
media
Token media
Feature activation+0.000
was
Token was
Feature activation+0.000
biased
Token biased
Feature activation+1.080
against
Token against
Feature activation+4.144
Trump
Token Trump
Feature activation+0.421
.
Token.
Feature activation+0.000
Numbers
Token Numbers
Feature activation+0.000
like
Token like
Feature activation+0.000

INTERVAL 0.000 - 0.544
CONTAINS 99.996%

I
Token I
Feature activation+0.000
think
Token think
Feature activation+0.000
that
Token that
Feature activation+0.000
's
Token's
Feature activation+0.000
because
Token because
Feature activation+0.000
they
Token they
Feature activation+0.000
have
Token have
Feature activation+0.000
mum
Token mum
Feature activation+0.000
,
Token,
Feature activation+0.000
dad
Token dad
Feature activation+0.000
and
Token and
Feature activation+0.000
composed
Token composed
Feature activation+0.000
of
Token of
Feature activation+0.000
closely
Token closely
Feature activation+0.000
packed
Token packed
Feature activation+0.000
cones
Token cones
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
eye
Token eye
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
is
Token is
Feature activation+0.000
re
Token re
Feature activation+0.000
opening
Tokenopening
Feature activation+0.000
the
Token the
Feature activation+0.000
landmark
Token landmark
Feature activation+0.000
1964
Token 1964
Feature activation+0.000
law
Token law
Feature activation+0.000
for
Token for
Feature activation+0.000
revisions
Token revisions
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
pro
Tokenpro
Feature activation+0.000
pos
Tokenpos
Feature activation+0.000
als
Tokenals
Feature activation+0.000
and
Token and
Feature activation+0.000
preferences
Token preferences
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
people
Token people
Feature activation+0.000
"
Token"
Feature activation+0.000
early
Token early
Feature activation+0.000
in
Token in
Feature activation+0.000
spring
Token spring
Feature activation+0.000
training
Token training
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
NY
Token NY
Feature activation+0.000
Daily
Token Daily
Feature activation+0.000
News
Token News
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 7: In local text involving swinging / swaying

TOP ACTIVATIONS
MAX = 4.693

to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
out
Token out
Feature activation+0.000
there
Token there
Feature activation+0.000
swinging
Token swinging
Feature activation+0.000
for
Token for
Feature activation+4.693
the
Token the
Feature activation+2.062
fences
Token fences
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
the
Token the
Feature activation+0.000
with
Token with
Feature activation+0.000
a
Token a
Feature activation+0.000
single
Token single
Feature activation+0.000
clumsy
Token clumsy
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+4.683
a
Token a
Feature activation+3.504
palm
Token palm
Feature activation+0.149
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
,
Token,
Feature activation+0.000
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
board
Token board
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+3.913
a
Token a
Feature activation+3.228
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
style
Token style
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
games
Token games
Feature activation+0.000
that
Token that
Feature activation+0.000
swing
Token swing
Feature activation+0.000
it
Token it
Feature activation+3.839
one
Token one
Feature activation+2.446
way
Token way
Feature activation+0.053
or
Token or
Feature activation+0.000
the
Token the
Feature activation+0.000
other
Token other
Feature activation+0.000
a
Token a
Feature activation+0.000
single
Token single
Feature activation+0.000
clumsy
Token clumsy
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+4.683
a
Token a
Feature activation+3.504
palm
Token palm
Feature activation+0.149
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
to
Token to
Feature activation+0.000
catch
Token catch
Feature activation+0.000
a
Token a
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+3.320
the
Token the
Feature activation+2.671
lady
Token lady
Feature activation+0.000
blogger
Token blogger
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
With
Token With
Feature activation+0.000
two
Token two
Feature activation+0.000
swings
Token swings
Feature activation+0.000
of
Token of
Feature activation+3.240
the
Token the
Feature activation+2.375
bar
Token bar
Feature activation+0.000
,
Token,
Feature activation+0.000
Scar
Token Scar
Feature activation+0.000
ver
Tokenver
Feature activation+0.000
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
board
Token board
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+3.913
a
Token a
Feature activation+3.228
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
style
Token style
Feature activation+0.000
DE
Token DE
Feature activation+0.000
a
Token a
Feature activation+0.000
major
Token major
Feature activation+0.000
role
Token role
Feature activation+0.000
in
Token in
Feature activation+0.000
swinging
Token swinging
Feature activation+0.000
the
Token the
Feature activation+3.187
state
Token state
Feature activation+1.010
for
Token for
Feature activation+1.700
Obama
Token Obama
Feature activation+0.000
against
Token against
Feature activation+0.000
Romney
Token Romney
Feature activation+0.000
get
Token get
Feature activation+0.000
back
Token back
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+3.145
things
Token things
Feature activation+0.748
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
long
Token long
Feature activation+0.000
Thanksgiving
Token Thanksgiving
Feature activation+0.000
European
TokenEuropean
Feature activation+0.000
voters
Token voters
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+1.384
the
Token the
Feature activation+2.326
right
Token right
Feature activation+3.139
,
Token,
Feature activation+0.083
and
Token and
Feature activation+0.000
American
Token American
Feature activation+0.000
voters
Token voters
Feature activation+0.000
are
Token are
Feature activation+0.000
1998
Token 1998
Feature activation+0.000
-
Token -
Feature activation+0.000
a
Token a
Feature activation+0.000
sizeable
Token sizeable
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+3.000
Labor
Token Labor
Feature activation+0.842
for
Token for
Feature activation+0.737
only
Token only
Feature activation+0.000
a
Token a
Feature activation+0.000
modest
Token modest
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
perhaps
Token perhaps
Feature activation+0.000
I
Token I
Feature activation+0.000
will
Token will
Feature activation+0.000
swing
Token swing
Feature activation+0.000
my
Token my
Feature activation+2.992
bo
Token bo
Feature activation+0.000
og
Tokenog
Feature activation+0.000
er
Tokener
Feature activation+0.000
proudly
Token proudly
Feature activation+0.000
to
Token to
Feature activation+0.000
he
Token he
Feature activation+0.000
took
Token took
Feature activation+0.000
one
Token one
Feature activation+0.000
final
Token final
Feature activation+0.000
swing
Token swing
Feature activation+0.231
at
Token at
Feature activation+2.936
the
Token the
Feature activation+1.767
Americans
Token Americans
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
island
Token island
Feature activation+0.000
would
Token would
Feature activation+0.000
seem
Token seem
Feature activation+0.000
an
Token an
Feature activation+0.000
odd
Token odd
Feature activation+0.000
swing
Token swing
Feature activation+0.000
for
Token for
Feature activation+2.930
Gates
Token Gates
Feature activation+0.000
to
Token to
Feature activation+0.000
suggest
Token suggest
Feature activation+0.000
now
Token now
Feature activation+0.000
that
Token that
Feature activation+0.000
bo
Token bo
Feature activation+0.000
og
Tokenog
Feature activation+0.000
er
Tokener
Feature activation+0.000
proudly
Token proudly
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+2.858
world
Token world
Feature activation+0.000
as
Token as
Feature activation+0.000
I
Token I
Feature activation+0.000
pass
Token pass
Feature activation+0.000
by
Token by
Feature activation+0.000
to
Token to
Feature activation+0.000
catch
Token catch
Feature activation+0.000
a
Token a
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+3.320
the
Token the
Feature activation+2.671
lady
Token lady
Feature activation+0.000
blogger
Token blogger
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
a
Token a
Feature activation+0.000
hundred
Token hundred
Feature activation+0.000
times
Token times
Feature activation+0.000
,
Token,
Feature activation+0.000
sway
Token sway
Feature activation+0.000
ing
Tokening
Feature activation+2.651
to
Token to
Feature activation+1.236
its
Token its
Feature activation+2.153
gentle
Token gentle
Feature activation+0.000
melodies
Token melodies
Feature activation+0.000
with
Token with
Feature activation+0.000
didn
Token didn
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
swing
Token swing
Feature activation+0.000
at
Token at
Feature activation+2.630
it
Token it
Feature activation+0.945
because
Token because
Feature activation+0.307
I
Token I
Feature activation+0.000
thought
Token thought
Feature activation+0.000
it
Token it
Feature activation+0.000
gets
Token gets
Feature activation+0.000
bellig
Token bellig
Feature activation+0.000
erent
Tokenerent
Feature activation+0.000
,
Token,
Feature activation+0.000
swings
Token swings
Feature activation+0.000
at
Token at
Feature activation+2.623
the
Token the
Feature activation+1.997
boun
Token boun
Feature activation+0.000
cer
Tokencer
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000

Top DFA by src position
MAX = 6.737

going
Token going
Feature activation-0.003
to
Token to
Feature activation+0.007
be
Token be
Feature activation+0.038
out
Token out
Feature activation+0.035
there
Token there
Feature activation-0.084
swinging
Token swinging
Feature activation+6.462
for
Token for
Feature activation+0.270
the
Token the
Feature activation+0.000
fences
Token fences
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
rubble
Token rubble
Feature activation-0.011
with
Token with
Feature activation+0.024
a
Token a
Feature activation-0.004
single
Token single
Feature activation-0.047
clumsy
Token clumsy
Feature activation+0.081
swing
Token swing
Feature activation+6.737
of
Token of
Feature activation+0.185
a
Token a
Feature activation+0.000
palm
Token palm
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
extreme
Token extreme
Feature activation-0.054
,
Token,
Feature activation-0.139
across
Token across
Feature activation+0.087
the
Token the
Feature activation+0.020
board
Token board
Feature activation+0.004
swing
Token swing
Feature activation+5.940
to
Token to
Feature activation+0.269
a
Token a
Feature activation+0.000
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
types
Token types
Feature activation+0.007
of
Token of
Feature activation-0.014
the
Token the
Feature activation+0.010
games
Token games
Feature activation-0.013
that
Token that
Feature activation+0.057
swing
Token swing
Feature activation+5.373
it
Token it
Feature activation+0.111
one
Token one
Feature activation+0.000
way
Token way
Feature activation+0.000
or
Token or
Feature activation+0.000
the
Token the
Feature activation+0.000
rubble
Token rubble
Feature activation+0.051
with
Token with
Feature activation+0.024
a
Token a
Feature activation-0.158
single
Token single
Feature activation-0.053
clumsy
Token clumsy
Feature activation+0.067
swing
Token swing
Feature activation+5.172
of
Token of
Feature activation+0.858
a
Token a
Feature activation+0.003
palm
Token palm
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.939
to
Token to
Feature activation-0.052
catch
Token catch
Feature activation-0.123
a
Token a
Feature activation+0.068
swing
Token swing
Feature activation+5.166
of
Token of
Feature activation+0.298
the
Token the
Feature activation+0.000
lady
Token lady
Feature activation+0.000
blogger
Token blogger
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
.
Token.
Feature activation+0.009
âĢ
TokenâĢ
Feature activation+0.004
Ŀ
TokenĿ
Feature activation-0.003
With
Token With
Feature activation+0.041
two
Token two
Feature activation-0.067
swings
Token swings
Feature activation+4.975
of
Token of
Feature activation+0.187
the
Token the
Feature activation+0.000
bar
Token bar
Feature activation+0.000
,
Token,
Feature activation+0.000
Scar
Token Scar
Feature activation+0.000
extreme
Token extreme
Feature activation-0.050
,
Token,
Feature activation-0.095
across
Token across
Feature activation+0.061
the
Token the
Feature activation+0.008
board
Token board
Feature activation-0.024
swing
Token swing
Feature activation+4.622
to
Token to
Feature activation+0.746
a
Token a
Feature activation+0.204
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
played
Token played
Feature activation+0.090
a
Token a
Feature activation-0.006
major
Token major
Feature activation-0.015
role
Token role
Feature activation-0.028
in
Token in
Feature activation+0.031
swinging
Token swinging
Feature activation+5.084
the
Token the
Feature activation+0.130
state
Token state
Feature activation+0.000
for
Token for
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
against
Token against
Feature activation+0.000
to
Token to
Feature activation-0.023
get
Token get
Feature activation-0.060
back
Token back
Feature activation-0.050
in
Token in
Feature activation+0.149
the
Token the
Feature activation-0.004
swing
Token swing
Feature activation+5.287
of
Token of
Feature activation+0.304
things
Token things
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
long
Token long
Feature activation+0.000
2009
Token 2009
Feature activation-0.050
Ċ
TokenĊ
Feature activation-0.001
Ċ
TokenĊ
Feature activation-0.000
European
TokenEuropean
Feature activation-0.003
voters
Token voters
Feature activation+0.091
swing
Token swing
Feature activation+3.924
to
Token to
Feature activation+1.064
the
Token the
Feature activation+0.296
right
Token right
Feature activation+0.224
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
of
Token of
Feature activation-0.028
1998
Token 1998
Feature activation+0.014
-
Token -
Feature activation+0.020
a
Token a
Feature activation+0.141
sizeable
Token sizeable
Feature activation+0.031
swing
Token swing
Feature activation+4.508
to
Token to
Feature activation+0.316
Labor
Token Labor
Feature activation+0.000
for
Token for
Feature activation+0.000
only
Token only
Feature activation+0.000
a
Token a
Feature activation+0.000
t
Tokent
Feature activation+0.011
âĢĵ
Token âĢĵ
Feature activation-0.068
perhaps
Token perhaps
Feature activation-0.026
I
Token I
Feature activation-0.011
will
Token will
Feature activation-0.034
swing
Token swing
Feature activation+4.913
my
Token my
Feature activation+0.257
bo
Token bo
Feature activation+0.000
og
Tokenog
Feature activation+0.000
er
Tokener
Feature activation+0.000
proudly
Token proudly
Feature activation+0.000
,
Token,
Feature activation-0.014
he
Token he
Feature activation+0.021
took
Token took
Feature activation+0.015
one
Token one
Feature activation+0.036
final
Token final
Feature activation-0.107
swing
Token swing
Feature activation+4.695
at
Token at
Feature activation+0.539
the
Token the
Feature activation+0.000
Americans
Token Americans
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
it
Token it
Feature activation-0.012
would
Token would
Feature activation-0.010
seem
Token seem
Feature activation-0.043
an
Token an
Feature activation-0.023
odd
Token odd
Feature activation-0.061
swing
Token swing
Feature activation+5.060
for
Token for
Feature activation+0.118
Gates
Token Gates
Feature activation+0.000
to
Token to
Feature activation+0.000
suggest
Token suggest
Feature activation+0.000
now
Token now
Feature activation+0.000
t
Tokent
Feature activation+0.013
âĢĵ
Token âĢĵ
Feature activation-0.036
perhaps
Token perhaps
Feature activation-0.029
I
Token I
Feature activation-0.016
will
Token will
Feature activation+0.004
swing
Token swing
Feature activation+4.346
my
Token my
Feature activation+0.062
bo
Token bo
Feature activation+0.028
og
Tokenog
Feature activation-0.005
er
Tokener
Feature activation+0.056
proudly
Token proudly
Feature activation+0.154
<|endoftext|>
Token<|endoftext|>
Feature activation-0.959
to
Token to
Feature activation-0.024
catch
Token catch
Feature activation-0.114
a
Token a
Feature activation+0.036
swing
Token swing
Feature activation+3.901
of
Token of
Feature activation+0.713
the
Token the
Feature activation+0.216
lady
Token lady
Feature activation+0.000
blogger
Token blogger
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
song
Token song
Feature activation+0.065
a
Token a
Feature activation+0.038
hundred
Token hundred
Feature activation-0.033
times
Token times
Feature activation-0.030
,
Token,
Feature activation-0.176
sway
Token sway
Feature activation+4.723
ing
Tokening
Feature activation+0.387
to
Token to
Feature activation+0.000
its
Token its
Feature activation+0.000
gentle
Token gentle
Feature activation+0.000
melodies
Token melodies
Feature activation+0.000
I
Token I
Feature activation+0.015
didn
Token didn
Feature activation+0.020
âĢ
TokenâĢ
Feature activation+0.014
Ļ
TokenĻ
Feature activation-0.004
t
Tokent
Feature activation+0.043
swing
Token swing
Feature activation+3.971
at
Token at
Feature activation+0.268
it
Token it
Feature activation+0.000
because
Token because
Feature activation+0.000
I
Token I
Feature activation+0.000
thought
Token thought
Feature activation+0.000
He
Token He
Feature activation+0.035
gets
Token gets
Feature activation-0.079
bellig
Token bellig
Feature activation+0.009
erent
Tokenerent
Feature activation-0.078
,
Token,
Feature activation-0.093
swings
Token swings
Feature activation+4.677
at
Token at
Feature activation+0.266
the
Token the
Feature activation+0.000
boun
Token boun
Feature activation+0.000
cer
Tokencer
Feature activation+0.000
,
Token,
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.02

Head 2: 0.05

Head 3: 0.05

Head 4: 0.12

Head 5: 0.03

Head 6: 0.13

Head 7: 0.35

Head 8: 0.03

Head 9: 0.04

Head 10: 0.07

Head 11: 0.08

Positive logits

levers1.71

gradient1.53

compass1.51

punches1.51

sails1.49

grenades1.49

��1.48

directions1.45

torches1.43

arrow1.43

directional1.40

scissors1.37

arrows1.34

grenade1.34

envelope1.33

backwards1.33

flame1.33

Compass1.32

knots1.32

backward1.32

Negative logits

vets-1.55

vet-1.48

fol-1.47

delinquent-1.47

culosis-1.36

den-1.34

upkeep-1.32

igmat-1.30

vana-1.27

akespe-1.26

expenses-1.25

xus-1.23

Chronic-1.22

ikers-1.20

bil-1.19

adult-1.18

ukemia-1.17

ysis-1.16

expense-1.16

wives-1.15

INTERVAL 4.223 - 4.693
CONTAINS 0.000%

with
Token with
Feature activation+0.000
a
Token a
Feature activation+0.000
single
Token single
Feature activation+0.000
clumsy
Token clumsy
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+4.683
a
Token a
Feature activation+3.504
palm
Token palm
Feature activation+0.149
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
out
Token out
Feature activation+0.000
there
Token there
Feature activation+0.000
swinging
Token swinging
Feature activation+0.000
for
Token for
Feature activation+4.693
the
Token the
Feature activation+2.062
fences
Token fences
Feature activation+0.000
.
Token.
Feature activation+0.000
But
Token But
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 3.754 - 4.223
CONTAINS 0.000%

,
Token,
Feature activation+0.000
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
board
Token board
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+3.913
a
Token a
Feature activation+3.228
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
style
Token style
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
games
Token games
Feature activation+0.000
that
Token that
Feature activation+0.000
swing
Token swing
Feature activation+0.000
it
Token it
Feature activation+3.839
one
Token one
Feature activation+2.446
way
Token way
Feature activation+0.053
or
Token or
Feature activation+0.000
the
Token the
Feature activation+0.000
other
Token other
Feature activation+0.000

INTERVAL 3.285 - 3.754
CONTAINS 0.000%

<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
to
Token to
Feature activation+0.000
catch
Token catch
Feature activation+0.000
a
Token a
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+3.320
the
Token the
Feature activation+2.671
lady
Token lady
Feature activation+0.000
blogger
Token blogger
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
a
Token a
Feature activation+0.000
single
Token single
Feature activation+0.000
clumsy
Token clumsy
Feature activation+0.000
swing
Token swing
Feature activation+0.000
of
Token of
Feature activation+4.683
a
Token a
Feature activation+3.504
palm
Token palm
Feature activation+0.149
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000

INTERVAL 2.816 - 3.285
CONTAINS 0.000%

across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
board
Token board
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+3.913
a
Token a
Feature activation+3.228
3
Token 3
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
style
Token style
Feature activation+0.000
DE
Token DE
Feature activation+0.000
European
TokenEuropean
Feature activation+0.000
voters
Token voters
Feature activation+0.000
swing
Token swing
Feature activation+0.000
to
Token to
Feature activation+1.384
the
Token the
Feature activation+2.326
right
Token right
Feature activation+3.139
,
Token,
Feature activation+0.083
and
Token and
Feature activation+0.000
American
Token American
Feature activation+0.000
voters
Token voters
Feature activation+0.000
are
Token are
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
With
Token With
Feature activation+0.000
two
Token two
Feature activation+0.000
swings
Token swings
Feature activation+0.000
of
Token of
Feature activation+3.240
the
Token the
Feature activation+2.375
bar
Token bar
Feature activation+0.000
,
Token,
Feature activation+0.000
Scar
Token Scar
Feature activation+0.000
ver
Tokenver
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
perhaps
Token perhaps
Feature activation+0.000
I
Token I
Feature activation+0.000
will
Token will
Feature activation+0.000
swing
Token swing
Feature activation+0.000
my
Token my
Feature activation+2.992
bo
Token bo
Feature activation+0.000
og
Tokenog
Feature activation+0.000
er
Tokener
Feature activation+0.000
proudly
Token proudly
Feature activation+0.000
to
Token to
Feature activation+0.000
would
Token would
Feature activation+0.000
seem
Token seem
Feature activation+0.000
an
Token an
Feature activation+0.000
odd
Token odd
Feature activation+0.000
swing
Token swing
Feature activation+0.000
for
Token for
Feature activation+2.930
Gates
Token Gates
Feature activation+0.000
to
Token to
Feature activation+0.000
suggest
Token suggest
Feature activation+0.000
now
Token now
Feature activation+0.000
that
Token that
Feature activation+0.000

INTERVAL 2.346 - 2.816
CONTAINS 0.000%

demographic
Token demographic
Feature activation+0.000
group
Token group
Feature activation+0.000
critical
Token critical
Feature activation+0.000
to
Token to
Feature activation+0.000
swinging
Token swinging
Feature activation+0.000
the
Token the
Feature activation+2.521
electoral
Token electoral
Feature activation+0.918
map
Token map
Feature activation+0.343
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
support
Token support
Feature activation+0.000
six
Token six
Feature activation+0.000
years
Token years
Feature activation+0.000
later
Token later
Feature activation+0.000
you
Token you
Feature activation+0.000
swung
Token swung
Feature activation+0.000
your
Token your
Feature activation+2.571
wrong
Token wrong
Feature activation+0.254
foot
Token foot
Feature activation+0.304
to
Token to
Feature activation+1.179
drop
Token drop
Feature activation+0.000
the
Token the
Feature activation+0.000
or
Token or
Feature activation+0.000
since
Token since
Feature activation+0.000
).
Token).
Feature activation+0.000
Gonzalez
Token Gonzalez
Feature activation+0.000
swung
Token swung
Feature activation+0.000
at
Token at
Feature activation+2.364
Rivera
Token Rivera
Feature activation+0.000
's
Token's
Feature activation+0.336
0
Token 0
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
gets
Token gets
Feature activation+0.000
bellig
Token bellig
Feature activation+0.000
erent
Tokenerent
Feature activation+0.000
,
Token,
Feature activation+0.000
swings
Token swings
Feature activation+0.000
at
Token at
Feature activation+2.623
the
Token the
Feature activation+1.997
boun
Token boun
Feature activation+0.000
cer
Tokencer
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
one
Token one
Feature activation+0.000
moment
Token moment
Feature activation+0.000
of
Token of
Feature activation+0.000
brilliance
Token brilliance
Feature activation+0.000
swings
Token swings
Feature activation+0.000
the
Token the
Feature activation+2.560
result
Token result
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
READ
TokenREAD
Feature activation+0.000

INTERVAL 1.877 - 2.346
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
l
Token l
Feature activation+0.000
urch
Tokenurch
Feature activation+0.000
to
Token to
Feature activation+1.054
the
Token the
Feature activation+2.177
right
Token right
Feature activation+1.775
was
Token was
Feature activation+0.000
more
Token more
Feature activation+0.000
pronounced
Token pronounced
Feature activation+0.000
for
Token for
Feature activation+0.000
describe
Token describe
Feature activation+0.000
a
Token a
Feature activation+0.000
terrestrial
Token terrestrial
Feature activation+0.000
environment
Token environment
Feature activation+0.000
swinging
Token swinging
Feature activation+0.000
into
Token into
Feature activation+2.174
and
Token and
Feature activation+0.000
out
Token out
Feature activation+0.241
of
Token of
Feature activation+1.984
relative
Token relative
Feature activation+0.000
extremes
Token extremes
Feature activation+0.000
but
Token but
Feature activation+0.000
I
Token I
Feature activation+0.000
think
Token think
Feature activation+0.000
I
Token I
Feature activation+0.000
swayed
Token swayed
Feature activation+0.000
him
Token him
Feature activation+2.143
with
Token with
Feature activation+1.219
the
Token the
Feature activation+1.397
mention
Token mention
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
her
Token her
Feature activation+0.000
three
Token three
Feature activation+0.000
-
Token-
Feature activation+0.000
day
Tokenday
Feature activation+0.000
swing
Token swing
Feature activation+0.000
through
Token through
Feature activation+2.160
the
Token the
Feature activation+0.501
Golden
Token Golden
Feature activation+0.000
State
Token State
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
can
Token can
Feature activation+0.000
't
Token't
Feature activation+0.000
l
Token l
Feature activation+0.000
urch
Tokenurch
Feature activation+0.000
to
Token to
Feature activation+1.276
a
Token a
Feature activation+2.276
John
Token John
Feature activation+0.000
Bolton
Token Bolton
Feature activation+0.000
,"
Token,"
Feature activation+0.000
Scarborough
Token Scarborough
Feature activation+0.000
said
Token said
Feature activation+0.000

INTERVAL 1.408 - 1.877
CONTAINS 0.000%

It
Token It
Feature activation+0.000
takes
Token takes
Feature activation+0.000
three
Token three
Feature activation+0.000
separate
Token separate
Feature activation+0.000
swings
Token swings
Feature activation+0.000
at
Token at
Feature activation+1.812
the
Token the
Feature activation+1.470
ball
Token ball
Feature activation+0.000
and
Token and
Feature activation+0.000
hopes
Token hopes
Feature activation+0.000
at
Token at
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
was
Token was
Feature activation+0.000
attempting
Token attempting
Feature activation+0.000
to
Token to
Feature activation+0.000
sway
Token sway
Feature activation+0.000
the
Token the
Feature activation+1.853
outcome
Token outcome
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
election
Token election
Feature activation+0.000
in
Token in
Feature activation+0.000
oe
Tokenoe
Feature activation+0.000
that
Token that
Feature activation+0.000
failed
Token failed
Feature activation+0.000
to
Token to
Feature activation+0.000
yield
Token yield
Feature activation+0.000
to
Token to
Feature activation+1.550
officers
Token officers
Feature activation+0.000
on
Token on
Feature activation+0.000
Interstate
Token Interstate
Feature activation+0.000
85
Token 85
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
a
Token a
Feature activation+1.062
handful
Token handful
Feature activation+0.000
of
Token of
Feature activation+0.000
undecided
Token undecided
Feature activation+0.000
lawmakers
Token lawmakers
Feature activation+0.000
to
Token to
Feature activation+1.735
his
Token his
Feature activation+1.891
side
Token side
Feature activation+0.000
.
Token.
Feature activation+0.000
In
Token In
Feature activation+0.000
the
Token the
Feature activation+0.000
er
Tokener
Feature activation+0.000
ley
Tokenley
Feature activation+0.000
quickly
Token quickly
Feature activation+0.000
became
Token became
Feature activation+0.000
swayed
Token swayed
Feature activation+1.044
by
Token by
Feature activation+1.765
his
Token his
Feature activation+1.118
agent
Token agent
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000

INTERVAL 0.939 - 1.408
CONTAINS 0.001%

.
Token.
Feature activation+0.000
Travis
Token Travis
Feature activation+0.000
doesn
Token doesn
Feature activation+0.000
't
Token't
Feature activation+0.000
swing
Token swing
Feature activation+0.000
for
Token for
Feature activation+1.167
the
Token the
Feature activation+0.000
fences
Token fences
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
he
Token he
Feature activation+0.000
can
Token can
Feature activation+0.000
he
Token he
Feature activation+0.000
be
Token be
Feature activation+0.000
seen
Token seen
Feature activation+0.000
swinging
Token swinging
Feature activation+0.000
the
Token the
Feature activation+1.208
club
Token club
Feature activation+0.015
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
38
Token 38
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
the
Token the
Feature activation+0.000
biggest
Token biggest
Feature activation+0.000
swing
Token swing
Feature activation+0.000
in
Token in
Feature activation+1.226
percentage
Token percentage
Feature activation+0.000
terms
Token terms
Feature activation+0.000
in
Token in
Feature activation+0.000
Wisconsin
Token Wisconsin
Feature activation+0.000
.
Token.
Feature activation+0.000
S
Token S
Feature activation+0.000
ink
Tokenink
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
swing
Token swing
Feature activation+0.000
either
Token either
Feature activation+1.273
Matthews
Token Matthews
Feature activation+0.000
or
Token or
Feature activation+0.000
Pe
Token Pe
Feature activation+0.000
ppers
Tokenppers
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
grappling
Token grappling
Feature activation+0.000
beam
Token beam
Feature activation+0.000
to
Token to
Feature activation+0.000
swing
Token swing
Feature activation+0.000
from
Token from
Feature activation+0.985
point
Token point
Feature activation+0.000
to
Token to
Feature activation+0.000
point
Token point
Feature activation+0.000
over
Token over
Feature activation+0.034
a
Token a
Feature activation+0.000

INTERVAL 0.469 - 0.939
CONTAINS 0.001%

by
Token by
Feature activation+0.000
Russian
Token Russian
Feature activation+0.000
intelligence
Token intelligence
Feature activation+0.000
to
Token to
Feature activation+0.000
sway
Token sway
Feature activation+0.000
the
Token the
Feature activation+0.711
results
Token results
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
.
Token.
Feature activation+0.000
military
Token military
Feature activation+0.000
prepared
Token prepared
Feature activation+0.000
to
Token to
Feature activation+0.000
throw
Token throw
Feature activation+0.000
at
Token at
Feature activation+0.542
the
Token the
Feature activation+0.158
attackers
Token attackers
Feature activation+0.000
was
Token was
Feature activation+0.000
effectively
Token effectively
Feature activation+0.000
a
Token a
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
legislation
Token legislation
Feature activation+0.000
was
Token was
Feature activation+0.000
supported
Token supported
Feature activation+0.000
by
Token by
Feature activation+0.515
a
Token a
Feature activation+0.244
50
Token 50
Feature activation+0.000
-
Token-
Feature activation+0.000
19
Token19
Feature activation+0.000
vote
Token vote
Feature activation+0.000
he
Token he
Feature activation+0.000
swung
Token swung
Feature activation+0.000
the
Token the
Feature activation+1.355
golf
Token golf
Feature activation+0.000
club
Token club
Feature activation+0.358
at
Token at
Feature activation+0.472
her
Token her
Feature activation+0.000
.
Token.
Feature activation+0.000
Now
Token Now
Feature activation+0.000
here
Tokenhere
Feature activation+0.000
in
Token in
Feature activation+0.000
had
Token had
Feature activation+0.000
swung
Token swung
Feature activation+0.000
far
Token far
Feature activation+1.813
to
Token to
Feature activation+1.606
the
Token the
Feature activation+2.171
side
Token side
Feature activation+0.603
of
Token of
Feature activation+0.000
security
Token security
Feature activation+0.000
following
Token following
Feature activation+0.000
9
Token 9
Feature activation+0.000
/
Token/
Feature activation+0.000

INTERVAL 0.000 - 0.469
CONTAINS 99.998%

when
Token when
Feature activation+0.000
people
Token people
Feature activation+0.000
go
Token go
Feature activation+0.000
to
Token to
Feature activation+0.000
bed
Token bed
Feature activation+0.000
.
Token.
Feature activation+0.000
Smart
Token Smart
Feature activation+0.000
phone
Tokenphone
Feature activation+0.000
apps
Token apps
Feature activation+0.000
can
Token can
Feature activation+0.000
provide
Token provide
Feature activation+0.000
in
Token in
Feature activation+0.000
nearly
Token nearly
Feature activation+0.000
every
Token every
Feature activation+0.000
Mortal
Token Mortal
Feature activation+0.000
K
Token K
Feature activation+0.000
ombat
Tokenombat
Feature activation+0.000
fighting
Token fighting
Feature activation+0.000
game
Token game
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
one
Token one
Feature activation+0.000
graphic
Token graphic
Feature activation+0.000
representation
Token representation
Feature activation+0.000
of
Token of
Feature activation+0.000
cell
Token cell
Feature activation+0.000
division
Token division
Feature activation+0.000
over
Token over
Feature activation+0.000
time
Token time
Feature activation+0.000
called
Token called
Feature activation+0.000
a
Token a
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
watch
Tokenwatch
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
Day
Token Day
Feature activation+0.000
To
Token To
Feature activation+0.000
Remember
Token Remember
Feature activation+0.000
and
Token and
Feature activation+0.000
Vol
Token Vol
Feature activation+0.000
beat
Tokenbeat
Feature activation+0.000
sidewalk
Token sidewalk
Feature activation+0.000
in
Token in
Feature activation+0.000
front
Token front
Feature activation+0.000
of
Token of
Feature activation+0.000
Rite
Token Rite
Feature activation+0.000
Aid
Token Aid
Feature activation+0.000
was
Token was
Feature activation+0.000
stre
Token stre
Feature activation+0.000
wn
Tokenwn
Feature activation+0.000
with
Token with
Feature activation+0.000
blankets
Token blankets
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 8: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 1.042

<|endoftext|>
Token<|endoftext|>
Feature activation+0.601
,
Token,
Feature activation+0.003
the
Token the
Feature activation+0.010
NB
Token NB
Feature activation-0.006
Space
Token Space
Feature activation+0.031
Race
Token Race
Feature activation-0.027
lofty
Token lofty
Feature activation-0.011
goal
Token goal
Feature activation-0.035
:
Token:
Feature activation-0.039
Put
Token Put
Feature activation+0.122
a
Token a
Feature activation-0.068
Toronto
Token Toronto
Feature activation+0.191
Blue
Token Blue
Feature activation-0.046
Jays
Token Jays
Feature activation-0.046
âĢ
TokenâĢ
Feature activation+0.006
Ļ
TokenĻ
Feature activation-0.016
baseball
Token baseball
Feature activation+0.003
<|endoftext|>
Token<|endoftext|>
Feature activation+0.808
,
Token,
Feature activation-0.057
the
Token the
Feature activation+0.007
NB
Token NB
Feature activation-0.008
Space
Token Space
Feature activation+0.025
Race
Token Race
Feature activation-0.025
<|endoftext|>
Token<|endoftext|>
Feature activation+0.509
,
Token,
Feature activation-0.034
the
Token the
Feature activation+0.022
NB
Token NB
Feature activation+0.005
Space
Token Space
Feature activation+0.009
Race
Token Race
Feature activation-0.006
<|endoftext|>
Token<|endoftext|>
Feature activation+0.699
,
Token,
Feature activation-0.406
the
Token the
Feature activation-0.124
NB
Token NB
Feature activation-0.035
Space
Token Space
Feature activation-0.020
Race
Token Race
Feature activation-0.100
<|endoftext|>
Token<|endoftext|>
Feature activation+0.790
,
Token,
Feature activation-0.385
the
Token the
Feature activation-0.002
NB
Token NB
Feature activation-0.051
Space
Token Space
Feature activation+0.017
Race
Token Race
Feature activation-0.057
<|endoftext|>
Token<|endoftext|>
Feature activation+0.593
,
Token,
Feature activation-0.257
the
Token the
Feature activation-0.131
NB
Token NB
Feature activation-0.011
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation-0.074
<|endoftext|>
Token<|endoftext|>
Feature activation+1.042
,
Token,
Feature activation-0.206
the
Token the
Feature activation-0.130
NB
Token NB
Feature activation-0.007
Space
Token Space
Feature activation-0.015
Race
Token Race
Feature activation-0.032
<|endoftext|>
Token<|endoftext|>
Feature activation+0.804
,
Token,
Feature activation-0.057
the
Token the
Feature activation-0.008
NB
Token NB
Feature activation-0.007
Space
Token Space
Feature activation+0.023
Race
Token Race
Feature activation-0.030
<|endoftext|>
Token<|endoftext|>
Feature activation+0.800
,
Token,
Feature activation-0.054
the
Token the
Feature activation-0.022
NB
Token NB
Feature activation-0.003
Space
Token Space
Feature activation+0.037
Race
Token Race
Feature activation-0.026
<|endoftext|>
Token<|endoftext|>
Feature activation+0.772
,
Token,
Feature activation-0.219
the
Token the
Feature activation+0.014
NB
Token NB
Feature activation-0.008
Space
Token Space
Feature activation+0.040
Race
Token Race
Feature activation-0.086
<|endoftext|>
Token<|endoftext|>
Feature activation+0.858
,
Token,
Feature activation-0.088
the
Token the
Feature activation+0.011
NB
Token NB
Feature activation-0.002
Space
Token Space
Feature activation+0.052
Race
Token Race
Feature activation-0.030
<|endoftext|>
Token<|endoftext|>
Feature activation+0.832
,
Token,
Feature activation-0.170
the
Token the
Feature activation-0.006
NB
Token NB
Feature activation-0.005
Space
Token Space
Feature activation+0.021
Race
Token Race
Feature activation-0.007
<|endoftext|>
Token<|endoftext|>
Feature activation+0.694
,
Token,
Feature activation-0.095
the
Token the
Feature activation-0.031
NB
Token NB
Feature activation-0.016
Space
Token Space
Feature activation+0.007
Race
Token Race
Feature activation-0.055
<|endoftext|>
Token<|endoftext|>
Feature activation+0.884
,
Token,
Feature activation-0.085
the
Token the
Feature activation-0.007
NB
Token NB
Feature activation-0.004
Space
Token Space
Feature activation+0.006
Race
Token Race
Feature activation-0.008
<|endoftext|>
Token<|endoftext|>
Feature activation+0.688
,
Token,
Feature activation-0.036
the
Token the
Feature activation+0.008
NB
Token NB
Feature activation+0.003
Space
Token Space
Feature activation+0.031
Race
Token Race
Feature activation-0.061
<|endoftext|>
Token<|endoftext|>
Feature activation+0.654
,
Token,
Feature activation-0.622
the
Token the
Feature activation-0.040
NB
Token NB
Feature activation-0.003
Space
Token Space
Feature activation+0.003
Race
Token Race
Feature activation-0.029
<|endoftext|>
Token<|endoftext|>
Feature activation+0.556
,
Token,
Feature activation-0.427
the
Token the
Feature activation-0.089
NB
Token NB
Feature activation-0.009
Space
Token Space
Feature activation+0.016
Race
Token Race
Feature activation-0.030
<|endoftext|>
Token<|endoftext|>
Feature activation+1.023
,
Token,
Feature activation-0.260
the
Token the
Feature activation-0.096
NB
Token NB
Feature activation-0.008
Space
Token Space
Feature activation-0.009
Race
Token Race
Feature activation-0.027
<|endoftext|>
Token<|endoftext|>
Feature activation+0.771
,
Token,
Feature activation-0.329
the
Token the
Feature activation-0.096
NB
Token NB
Feature activation-0.014
Space
Token Space
Feature activation-0.008
Race
Token Race
Feature activation-0.045

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.11

Head 2: 0.10

Head 3: 0.09

Head 4: 0.09

Head 5: 0.08

Head 6: 0.07

Head 7: 0.08

Head 8: 0.10

Head 9: 0.08

Head 10: 0.07

Head 11: 0.07

Positive logits

*/(2.21

QL1.99

mercial1.97

adelphia1.94

nir1.92

ynt1.91

kid1.88

cture1.85

omorphic1.84

oke1.83

merce1.81

yrus1.79

pid1.78

mble1.77

bably1.76

anut1.76

yip1.71

ospace1.71

bris1.68

idth1.68

Negative logits

Kremlin-1.90

�士-1.88

Levin-1.83

Feldman-1.76

Doomsday-1.73

falsehood-1.66

Bernstein-1.64

Flint-1.62

Judgment-1.58

א-1.58

Nev-1.57

Toledo-1.52

ENA-1.51

Centauri-1.48

tut-1.48

recess-1.47

Stein-1.47

owitz-1.46

Freud-1.46

Sinclair-1.46

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

will
Token will
Feature activation+0.000
likely
Token likely
Feature activation+0.000
try
Token try
Feature activation+0.000
to
Token to
Feature activation+0.000
interfere
Token interfere
Feature activation+0.000
in
Token in
Feature activation+0.000
her
Token her
Feature activation+0.000
country
Token country
Feature activation+0.000
's
Token's
Feature activation+0.000
elections
Token elections
Feature activation+0.000
.
Token.
Feature activation+0.000
Every
Token Every
Feature activation+0.000
year
Token year
Feature activation+0.000
on
Token on
Feature activation+0.000
December
Token December
Feature activation+0.000
5
Token 5
Feature activation+0.000
,
Token,
Feature activation+0.000
thousands
Token thousands
Feature activation+0.000
of
Token of
Feature activation+0.000
shoppers
Token shoppers
Feature activation+0.000
visit
Token visit
Feature activation+0.000
24
Token 24
Feature activation+0.000
needed
Token needed
Feature activation+0.000
because
Token because
Feature activation+0.000
city
Token city
Feature activation+0.000
code
Token code
Feature activation+0.000
enforcement
Token enforcement
Feature activation+0.000
inspectors
Token inspectors
Feature activation+0.000
face
Token face
Feature activation+0.000
problems
Token problems
Feature activation+0.000
cracking
Token cracking
Feature activation+0.000
down
Token down
Feature activation+0.000
on
Token on
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
26
Token 26
Feature activation+0.000
-
Token-
Feature activation+0.000
year
Tokenyear
Feature activation+0.000
-
Token-
Feature activation+0.000
old
Tokenold
Feature activation+0.000
,
Token,
Feature activation+0.000
second
Token second
Feature activation+0.000
last
Token last
Feature activation+0.000
forced
Token forced
Feature activation+0.000
the
Token the
Feature activation+0.000
city
Token city
Feature activation+0.000
to
Token to
Feature activation+0.000
hire
Token hire
Feature activation+0.000
Moore
Token Moore
Feature activation+0.000
back
Token back
Feature activation+0.000
,
Token,
Feature activation+0.000
because
Token because
Feature activation+0.000
there
Token there
Feature activation+0.000
was
Token was
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 9: Fuzzy follows “ facilitate”

TOP ACTIVATIONS
MAX = 4.714

a
Token a
Feature activation+0.000
sexual
Tokensexual
Feature activation+0.000
ity
Tokenity
Feature activation+0.000
and
Token and
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
the
Token the
Feature activation+4.714
growth
Token growth
Feature activation+0.000
of
Token of
Feature activation+0.000
an
Token an
Feature activation+0.000
a
Token a
Feature activation+0.000
sexual
Tokensexual
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
employees
Token employees
Feature activation+0.000
in
Token in
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
them
Token them
Feature activation+4.632
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
public
Token public
Feature activation+0.000
information
Token information
Feature activation+0.000
before
Token before
Feature activation+0.000
can
Token can
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
fac
Tokenfac
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ate
Tokenate
Feature activation+4.264
the
Token the
Feature activation+3.141
state
Token state
Feature activation+1.537
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
for
Token for
Feature activation+0.000
its
Token its
Feature activation+0.000
part
Token part
Feature activation+0.000
in
Token in
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
the
Token the
Feature activation+4.024
fraudulent
Token fraudulent
Feature activation+0.000
representation
Token representation
Feature activation+0.000
of
Token of
Feature activation+0.000
Wells
Token Wells
Feature activation+0.000
Fargo
Token Fargo
Feature activation+0.000
section
Token section
Feature activation+0.000
al
Tokenal
Feature activation+0.000
interests
Token interests
Feature activation+0.000
,
Token,
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
the
Token the
Feature activation+4.009
demands
Token demands
Feature activation+0.000
of
Token of
Feature activation+0.000
unions
Token unions
Feature activation+0.000
and
Token and
Feature activation+0.000
old
Token old
Feature activation+0.000
against
Token against
Feature activation+0.000
them
Token them
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.327
their
Token their
Feature activation+3.400
own
Token own
Feature activation+3.863
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
the
Token the
Feature activation+0.000
necessary
Token necessary
Feature activation+0.000
laws
Token laws
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.460
the
Token the
Feature activation+3.774
release
Token release
Feature activation+0.000
of
Token of
Feature activation+0.000
detained
Token detained
Feature activation+0.000
candidates
Token candidates
Feature activation+0.000
.
Token.
Feature activation+0.000
on
Token on
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
cases
Token cases
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.951
the
Token the
Feature activation+3.556
suppression
Token suppression
Feature activation+0.139
of
Token of
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
.
Token.
Feature activation+0.000
M
Token M
Feature activation+0.000
and
Token and
Feature activation+0.000
are
Token are
Feature activation+0.000
committed
Token committed
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.908
a
Token a
Feature activation+3.410
safe
Token safe
Feature activation+0.847
and
Token and
Feature activation+0.000
secure
Token secure
Feature activation+0.000
environment
Token environment
Feature activation+0.000
for
Token for
Feature activation+0.000
turned
Token turned
Feature activation+0.000
against
Token against
Feature activation+0.000
them
Token them
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.327
their
Token their
Feature activation+3.400
own
Token own
Feature activation+3.863
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
used
Token used
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.267
the
Token the
Feature activation+3.348
production
Token production
Feature activation+0.000
of
Token of
Feature activation+0.000
platinum
Token platinum
Feature activation+0.000
coins
Token coins
Feature activation+0.000
for
Token for
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
can
Token can
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
Poland
Token Poland
Feature activation+0.432
's
Token's
Feature activation+3.316
marginal
Token marginal
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
EU
Token EU
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
essential
Token essential
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.107
the
Token the
Feature activation+3.297
internet
Token internet
Feature activation+0.499
,
Token,
Feature activation+0.000
cable
Token cable
Feature activation+0.000
TV
Token TV
Feature activation+0.000
service
Token service
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
will
Tokenwill
Feature activation+0.000
"
Token "
Feature activation+0.000
fac
Tokenfac
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ate
Tokenate
Feature activation+3.266
and
Token and
Feature activation+0.000
enable
Token enable
Feature activation+0.279
the
Token the
Feature activation+1.454
Iranian
Token Iranian
Feature activation+0.000
regime
Token regime
Feature activation+0.000
government
Token government
Feature activation+0.000
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
the
Token the
Feature activation+3.235
smooth
Token smooth
Feature activation+0.286
conducting
Token conducting
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
elections
Token elections
Feature activation+0.000
government
Token government
Feature activation+0.000
bodies
Token bodies
Feature activation+0.000
exist
Token exist
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.000
the
Token the
Feature activation+3.183
activities
Token activities
Feature activation+1.494
in
Token in
Feature activation+0.000
space
Token space
Feature activation+0.000
of
Token of
Feature activation+0.000
outside
Token outside
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
fac
Tokenfac
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ate
Tokenate
Feature activation+4.264
the
Token the
Feature activation+3.141
state
Token state
Feature activation+1.537
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
power
Token power
Feature activation+0.348
s
Tokens
Feature activation+0.000
population
Token population
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
facilitated
Token facilitated
Feature activation+0.000
the
Token the
Feature activation+3.127
rise
Token rise
Feature activation+0.000
of
Token of
Feature activation+0.000
ISIS
Token ISIS
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
raised
Token raised
Feature activation+0.000
curb
Token curb
Feature activation+0.000
to
Token to
Feature activation+0.000
help
Token help
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.616
the
Token the
Feature activation+3.117
boarding
Token boarding
Feature activation+0.000
of
Token of
Feature activation+0.000
buses
Token buses
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.000
order
Token order
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.080
the
Token the
Feature activation+3.102
early
Token early
Feature activation+0.000
recognition
Token recognition
Feature activation+0.000
and
Token and
Feature activation+0.000
management
Token management
Feature activation+0.000
of
Token of
Feature activation+0.000

Top DFA by src position
MAX = 7.702

of
Token of
Feature activation+0.105
a
Token a
Feature activation+0.064
sexual
Tokensexual
Feature activation+0.261
ity
Tokenity
Feature activation+0.167
and
Token and
Feature activation-0.047
facilitating
Token facilitating
Feature activation+6.953
the
Token the
Feature activation+0.429
growth
Token growth
Feature activation+0.000
of
Token of
Feature activation+0.000
an
Token an
Feature activation+0.000
a
Token a
Feature activation+0.000
complicity
Token complicity
Feature activation+0.323
of
Token of
Feature activation+0.101
his
Token his
Feature activation+0.046
employees
Token employees
Feature activation+0.276
in
Token in
Feature activation+0.207
facilitating
Token facilitating
Feature activation+7.702
them
Token them
Feature activation+0.235
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
public
Token public
Feature activation+0.000
information
Token information
Feature activation+0.000
.
Token.
Feature activation-0.017
It
Token It
Feature activation+0.099
can
Token can
Feature activation+0.160
âĢ
Token âĢ
Feature activation-0.101
ľ
Tokenľ
Feature activation+0.099
fac
Tokenfac
Feature activation+3.078
ilit
Tokenilit
Feature activation+3.020
ate
Tokenate
Feature activation+0.732
the
Token the
Feature activation+0.000
state
Token state
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Massachusetts
Token Massachusetts
Feature activation-0.004
for
Token for
Feature activation+0.141
its
Token its
Feature activation+0.012
part
Token part
Feature activation+0.060
in
Token in
Feature activation+0.060
facilitating
Token facilitating
Feature activation+6.544
the
Token the
Feature activation+0.267
fraudulent
Token fraudulent
Feature activation+0.000
representation
Token representation
Feature activation+0.000
of
Token of
Feature activation+0.000
Wells
Token Wells
Feature activation+0.000
its
Token its
Feature activation-0.015
section
Token section
Feature activation+0.162
al
Tokenal
Feature activation+0.001
interests
Token interests
Feature activation+0.089
,
Token,
Feature activation-0.007
facilitating
Token facilitating
Feature activation+6.782
the
Token the
Feature activation+0.215
demands
Token demands
Feature activation+0.000
of
Token of
Feature activation+0.000
unions
Token unions
Feature activation+0.000
and
Token and
Feature activation+0.000
be
Token be
Feature activation+0.051
turned
Token turned
Feature activation+0.025
against
Token against
Feature activation-0.184
them
Token them
Feature activation-0.036
to
Token to
Feature activation+0.198
facilitate
Token facilitate
Feature activation+6.721
their
Token their
Feature activation+0.320
own
Token own
Feature activation-0.030
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
amend
Token amend
Feature activation-0.093
the
Token the
Feature activation+0.026
necessary
Token necessary
Feature activation+0.097
laws
Token laws
Feature activation+0.146
to
Token to
Feature activation+0.028
facilitate
Token facilitate
Feature activation+6.667
the
Token the
Feature activation+0.122
release
Token release
Feature activation+0.000
of
Token of
Feature activation+0.000
detained
Token detained
Feature activation+0.000
candidates
Token candidates
Feature activation+0.000
rule
Token rule
Feature activation-0.027
on
Token on
Feature activation-0.031
corruption
Token corruption
Feature activation-0.001
cases
Token cases
Feature activation+0.074
to
Token to
Feature activation+0.049
facilitate
Token facilitate
Feature activation+6.257
the
Token the
Feature activation+0.218
suppression
Token suppression
Feature activation+0.000
of
Token of
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
.
Token.
Feature activation+0.000
,
Token,
Feature activation+0.026
and
Token and
Feature activation+0.027
are
Token are
Feature activation+0.245
committed
Token committed
Feature activation+0.340
to
Token to
Feature activation+0.234
facilitating
Token facilitating
Feature activation+4.762
a
Token a
Feature activation+0.375
safe
Token safe
Feature activation+0.000
and
Token and
Feature activation+0.000
secure
Token secure
Feature activation+0.000
environment
Token environment
Feature activation+0.000
be
Token be
Feature activation+0.008
turned
Token turned
Feature activation+0.144
against
Token against
Feature activation-0.311
them
Token them
Feature activation-0.007
to
Token to
Feature activation+0.111
facilitate
Token facilitate
Feature activation+7.351
their
Token their
Feature activation-0.003
own
Token own
Feature activation+0.000
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
or
Token or
Feature activation+0.021
should
Token should
Feature activation+0.056
be
Token be
Feature activation+0.084
used
Token used
Feature activation+0.045
to
Token to
Feature activation+0.281
facilitate
Token facilitate
Feature activation+5.682
the
Token the
Feature activation+0.228
production
Token production
Feature activation+0.000
of
Token of
Feature activation+0.000
platinum
Token platinum
Feature activation+0.000
coins
Token coins
Feature activation+0.000
real
Token real
Feature activation+0.025
threats
Token threats
Feature activation-0.323
.
Token.
Feature activation-0.042
It
Token It
Feature activation+0.058
can
Token can
Feature activation+0.057
facilitate
Token facilitate
Feature activation+6.743
Poland
Token Poland
Feature activation+0.569
's
Token's
Feature activation-0.060
marginal
Token marginal
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
in
Token in
Feature activation+0.000
âĢ
TokenâĢ
Feature activation-0.105
Ļ
TokenĻ
Feature activation+0.025
re
Tokenre
Feature activation+0.063
essential
Token essential
Feature activation+0.260
to
Token to
Feature activation+0.078
facilitate
Token facilitate
Feature activation+6.036
the
Token the
Feature activation+0.140
internet
Token internet
Feature activation+0.000
,
Token,
Feature activation+0.000
cable
Token cable
Feature activation+0.000
TV
Token TV
Feature activation+0.000
corners
Token corners
Feature activation-0.003
âĢĶ
TokenâĢĶ
Feature activation+0.006
will
Tokenwill
Feature activation+0.103
"
Token "
Feature activation+0.311
fac
Tokenfac
Feature activation+2.143
ilit
Tokenilit
Feature activation+2.919
ate
Tokenate
Feature activation+0.992
and
Token and
Feature activation+0.000
enable
Token enable
Feature activation+0.000
the
Token the
Feature activation+0.000
Iranian
Token Iranian
Feature activation+0.000
Erdogan
Token Erdogan
Feature activation+0.160
government
Token government
Feature activation+0.070
,
Token,
Feature activation+0.034
was
Token was
Feature activation+0.014
to
Token to
Feature activation+0.310
facilitate
Token facilitate
Feature activation+6.033
the
Token the
Feature activation+0.239
smooth
Token smooth
Feature activation+0.000
conducting
Token conducting
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
These
TokenThese
Feature activation+0.075
government
Token government
Feature activation-0.071
bodies
Token bodies
Feature activation+0.077
exist
Token exist
Feature activation+0.190
to
Token to
Feature activation+0.197
facilitate
Token facilitate
Feature activation+5.493
the
Token the
Feature activation+0.179
activities
Token activities
Feature activation+0.000
in
Token in
Feature activation+0.000
space
Token space
Feature activation+0.000
of
Token of
Feature activation+0.000
It
Token It
Feature activation+0.107
can
Token can
Feature activation+0.125
âĢ
Token âĢ
Feature activation-0.106
ľ
Tokenľ
Feature activation+0.068
fac
Tokenfac
Feature activation+1.385
ilit
Tokenilit
Feature activation+3.704
ate
Tokenate
Feature activation+0.801
the
Token the
Feature activation+0.010
state
Token state
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Ļ
TokenĻ
Feature activation-0.013
s
Tokens
Feature activation-0.026
population
Token population
Feature activation+0.003
,
Token,
Feature activation-0.030
and
Token and
Feature activation-0.086
facilitated
Token facilitated
Feature activation+6.181
the
Token the
Feature activation+0.256
rise
Token rise
Feature activation+0.000
of
Token of
Feature activation+0.000
ISIS
Token ISIS
Feature activation+0.000
.
Token.
Feature activation+0.000
a
Token a
Feature activation+0.024
raised
Token raised
Feature activation-0.367
curb
Token curb
Feature activation-0.078
to
Token to
Feature activation-0.037
help
Token help
Feature activation+0.324
facilitate
Token facilitate
Feature activation+5.931
the
Token the
Feature activation+0.244
boarding
Token boarding
Feature activation+0.000
of
Token of
Feature activation+0.000
buses
Token buses
Feature activation+0.000
.
Token.
Feature activation+0.000
dementia
Token dementia
Feature activation-0.012
,
Token,
Feature activation-0.015
in
Token in
Feature activation-0.127
order
Token order
Feature activation+0.149
to
Token to
Feature activation+0.325
facilitate
Token facilitate
Feature activation+5.659
the
Token the
Feature activation+0.254
early
Token early
Feature activation+0.000
recognition
Token recognition
Feature activation+0.000
and
Token and
Feature activation+0.000
management
Token management
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.02

Head 2: 0.06

Head 3: 0.08

Head 4: 0.18

Head 5: 0.04

Head 6: 0.06

Head 7: 0.32

Head 8: 0.04

Head 9: 0.05

Head 10: 0.08

Head 11: 0.08

Positive logits

dissemination1.74

flow1.69

workflow1.66

transfer1.62

Exit1.61

passage1.59

deliveries1.59

iHUD1.57

fulfillment1.56

creation1.54

visitation1.53

azaki1.52

access1.51

commerce1.51

delivery1.50

bidding1.50

transfers1.50

Entry1.49

transactions1.48

transfer1.47

Negative logits

quickShipAvailable-1.53

シャ-1.46

overboard-1.45

eland-1.43

-1.43

secut-1.40

Mald-1.39

rusty-1.38

stout-1.36

soaked-1.34

cloves-1.34

cknow-1.34

prone-1.33

References-1.33

minced-1.32

Adin-1.32

Parables-1.31

acher-1.31

ducks-1.28

mant-1.28

INTERVAL 4.243 - 4.714
CONTAINS 0.000%

can
Token can
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
fac
Tokenfac
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ate
Tokenate
Feature activation+4.264
the
Token the
Feature activation+3.141
state
Token state
Feature activation+1.537
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
a
Token a
Feature activation+0.000
sexual
Tokensexual
Feature activation+0.000
ity
Tokenity
Feature activation+0.000
and
Token and
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
the
Token the
Feature activation+4.714
growth
Token growth
Feature activation+0.000
of
Token of
Feature activation+0.000
an
Token an
Feature activation+0.000
a
Token a
Feature activation+0.000
sexual
Tokensexual
Feature activation+0.000
of
Token of
Feature activation+0.000
his
Token his
Feature activation+0.000
employees
Token employees
Feature activation+0.000
in
Token in
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
them
Token them
Feature activation+4.632
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
public
Token public
Feature activation+0.000
information
Token information
Feature activation+0.000
before
Token before
Feature activation+0.000

INTERVAL 3.771 - 4.243
CONTAINS 0.000%

section
Token section
Feature activation+0.000
al
Tokenal
Feature activation+0.000
interests
Token interests
Feature activation+0.000
,
Token,
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
the
Token the
Feature activation+4.009
demands
Token demands
Feature activation+0.000
of
Token of
Feature activation+0.000
unions
Token unions
Feature activation+0.000
and
Token and
Feature activation+0.000
old
Token old
Feature activation+0.000
for
Token for
Feature activation+0.000
its
Token its
Feature activation+0.000
part
Token part
Feature activation+0.000
in
Token in
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
the
Token the
Feature activation+4.024
fraudulent
Token fraudulent
Feature activation+0.000
representation
Token representation
Feature activation+0.000
of
Token of
Feature activation+0.000
Wells
Token Wells
Feature activation+0.000
Fargo
Token Fargo
Feature activation+0.000
the
Token the
Feature activation+0.000
necessary
Token necessary
Feature activation+0.000
laws
Token laws
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.460
the
Token the
Feature activation+3.774
release
Token release
Feature activation+0.000
of
Token of
Feature activation+0.000
detained
Token detained
Feature activation+0.000
candidates
Token candidates
Feature activation+0.000
.
Token.
Feature activation+0.000
against
Token against
Feature activation+0.000
them
Token them
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.327
their
Token their
Feature activation+3.400
own
Token own
Feature activation+3.863
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 3.300 - 3.771
CONTAINS 0.000%

on
Token on
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
cases
Token cases
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.951
the
Token the
Feature activation+3.556
suppression
Token suppression
Feature activation+0.139
of
Token of
Feature activation+0.000
corruption
Token corruption
Feature activation+0.000
.
Token.
Feature activation+0.000
M
Token M
Feature activation+0.000
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
used
Token used
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.267
the
Token the
Feature activation+3.348
production
Token production
Feature activation+0.000
of
Token of
Feature activation+0.000
platinum
Token platinum
Feature activation+0.000
coins
Token coins
Feature activation+0.000
for
Token for
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
can
Token can
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
Poland
Token Poland
Feature activation+0.432
's
Token's
Feature activation+3.316
marginal
Token marginal
Feature activation+0.000
ization
Tokenization
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
EU
Token EU
Feature activation+0.000
turned
Token turned
Feature activation+0.000
against
Token against
Feature activation+0.000
them
Token them
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.327
their
Token their
Feature activation+3.400
own
Token own
Feature activation+3.863
deport
Token deport
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
and
Token and
Feature activation+0.000
are
Token are
Feature activation+0.000
committed
Token committed
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.908
a
Token a
Feature activation+3.410
safe
Token safe
Feature activation+0.847
and
Token and
Feature activation+0.000
secure
Token secure
Feature activation+0.000
environment
Token environment
Feature activation+0.000
for
Token for
Feature activation+0.000

INTERVAL 2.828 - 3.300
CONTAINS 0.000%

such
Token such
Feature activation+0.000
pay
Token pay
Feature activation+0.000
offs
Tokenoffs
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.606
their
Token their
Feature activation+2.928
success
Token success
Feature activation+0.000
in
Token in
Feature activation+0.000
these
Token these
Feature activation+0.000
new
Token new
Feature activation+0.000
centers
Token centers
Feature activation+0.000
's
Token's
Feature activation+0.000
implied
Token implied
Feature activation+0.000
it
Token it
Feature activation+0.000
would
Token would
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
more
Token more
Feature activation+2.948
development
Token development
Feature activation+0.000
than
Token than
Feature activation+0.000
tearing
Token tearing
Feature activation+0.000
the
Token the
Feature activation+0.000
highway
Token highway
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
essential
Token essential
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.107
the
Token the
Feature activation+3.297
internet
Token internet
Feature activation+0.499
,
Token,
Feature activation+0.000
cable
Token cable
Feature activation+0.000
TV
Token TV
Feature activation+0.000
service
Token service
Feature activation+0.000
EC
Token EC
Feature activation+0.000
exemption
Token exemption
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
facilitates
Token facilitates
Feature activation+0.000
the
Token the
Feature activation+3.024
plants
Token plants
Feature activation+0.000
emission
Token emission
Feature activation+0.000
of
Token of
Feature activation+0.000
as
Token as
Feature activation+0.000
much
Token much
Feature activation+0.000
government
Token government
Feature activation+0.000
,
Token,
Feature activation+0.000
was
Token was
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
the
Token the
Feature activation+3.235
smooth
Token smooth
Feature activation+0.286
conducting
Token conducting
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
elections
Token elections
Feature activation+0.000

INTERVAL 2.357 - 2.828
CONTAINS 0.000%

,
Token,
Feature activation+0.000
aims
Token aims
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.458
the
Token the
Feature activation+2.175
process
Token process
Feature activation+2.398
for
Token for
Feature activation+0.483
those
Token those
Feature activation+0.000
seeking
Token seeking
Feature activation+0.000
to
Token to
Feature activation+0.000
imm
Token imm
Feature activation+0.000
crossings
Token crossings
Feature activation+0.000
were
Token were
Feature activation+0.000
facilitated
Token facilitated
Feature activation+0.000
in
Token in
Feature activation+1.016
part
Token part
Feature activation+1.274
by
Token by
Feature activation+2.391
an
Token an
Feature activation+0.331
international
Token international
Feature activation+0.000
streetcar
Token streetcar
Feature activation+0.000
system
Token system
Feature activation+0.000
that
Token that
Feature activation+0.000
must
Token must
Feature activation+0.000
experience
Token experience
Feature activation+0.000
the
Token the
Feature activation+0.000
reality
Token reality
Feature activation+0.000
facilitated
Token facilitated
Feature activation+0.000
by
Token by
Feature activation+2.426
sham
Token sham
Feature activation+0.000
ans
Tokenans
Feature activation+0.000
in
Token in
Feature activation+0.000
altered
Token altered
Feature activation+0.000
consciousness
Token consciousness
Feature activation+0.000
in
Token in
Feature activation+0.000
reducing
Token reducing
Feature activation+0.000
unemployment
Token unemployment
Feature activation+0.000
while
Token while
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.893
a
Token a
Feature activation+2.710
rewarding
Token rewarding
Feature activation+0.007
work
Token work
Feature activation+0.000
environment
Token environment
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
posts
Token posts
Feature activation+0.000
but
Token but
Feature activation+0.000
encouraged
Token encouraged
Feature activation+0.000
and
Token and
Feature activation+0.000
facilitated
Token facilitated
Feature activation+0.000
them
Token them
Feature activation+2.552
,
Token,
Feature activation+0.000
aware
Token aware
Feature activation+0.000
that
Token that
Feature activation+0.000
their
Token their
Feature activation+0.000
controversial
Token controversial
Feature activation+0.000

INTERVAL 1.886 - 2.357
CONTAINS 0.000%

in
Token in
Feature activation+0.000
no
Token no
Feature activation+0.000
rush
Token rush
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
a
Token a
Feature activation+1.917
deal
Token deal
Feature activation+0.106
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
urging
Token urging
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
keen
Token keen
Feature activation+0.000
interest
Token interest
Feature activation+0.000
in
Token in
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.828
this
Token this
Feature activation+2.078
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Des
Token Des
Feature activation+0.000
ai
Tokenai
Feature activation+0.000
study
Token study
Feature activation+0.000
population
Token population
Feature activation+0.000
that
Token that
Feature activation+0.000
can
Token can
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
our
Token our
Feature activation+1.952
understanding
Token understanding
Feature activation+0.000
of
Token of
Feature activation+0.000
past
Token past
Feature activation+0.000
and
Token and
Feature activation+0.000
future
Token future
Feature activation+0.000
national
Token national
Feature activation+0.000
space
Token space
Feature activation+0.000
strategy
Token strategy
Feature activation+0.000
and
Token and
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
better
Token better
Feature activation+2.113
coordination
Token coordination
Feature activation+0.000
of
Token of
Feature activation+0.000
space
Token space
Feature activation+0.000
activities
Token activities
Feature activation+0.000
across
Token across
Feature activation+0.000
for
Token for
Feature activation+0.000
support
Token support
Feature activation+0.000
in
Token in
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
a
Token a
Feature activation+2.697
local
Token local
Feature activation+2.031
ceasefire
Token ceasefire
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
On
TokenOn
Feature activation+0.000

INTERVAL 1.414 - 1.886
CONTAINS 0.000%

brand
Token brand
Feature activation+0.000
and
Token and
Feature activation+0.000
consumer
Token consumer
Feature activation+0.000
,
Token,
Feature activation+0.000
facilitated
Token facilitated
Feature activation+0.000
by
Token by
Feature activation+1.544
the
Token the
Feature activation+1.500
staff
Token staff
Feature activation+0.000
,
Token,
Feature activation+0.000
happens
Token happens
Feature activation+0.000
.
Token.
Feature activation+0.000
It
TokenIt
Feature activation+0.000
also
Token also
Feature activation+0.000
makes
Token makes
Feature activation+0.000
recommendations
Token recommendations
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.737
better
Token better
Feature activation+2.146
scrutiny
Token scrutiny
Feature activation+0.000
and
Token and
Feature activation+0.000
improve
Token improve
Feature activation+0.000
the
Token the
Feature activation+0.000
encryption
Token encryption
Feature activation+0.000
and
Token and
Feature activation+0.000
anonymity
Token anonymity
Feature activation+0.000
"
Token "
Feature activation+0.000
which
Tokenwhich
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.543
and
Token and
Feature activation+0.302
often
Token often
Feature activation+0.603
enable
Token enable
Feature activation+0.039
the
Token the
Feature activation+2.315
rights
Token rights
Feature activation+0.634
fac
Tokenfac
Feature activation+0.000
ilit
Tokenilit
Feature activation+0.000
ate
Tokenate
Feature activation+3.266
and
Token and
Feature activation+0.000
enable
Token enable
Feature activation+0.279
the
Token the
Feature activation+1.454
Iranian
Token Iranian
Feature activation+0.000
regime
Token regime
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
its
Token its
Feature activation+0.000
transit
Token transit
Feature activation+0.000
users
Token users
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
even
Token even
Feature activation+1.600
the
Token the
Feature activation+0.000
most
Token most
Feature activation+0.000
minor
Token minor
Feature activation+0.201
of
Token of
Feature activation+0.000
conven
Token conven
Feature activation+0.000

INTERVAL 0.943 - 1.414
CONTAINS 0.000%

using
Token using
Feature activation+0.000
a
Token a
Feature activation+0.000
computer
Token computer
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
a
Token a
Feature activation+1.191
sex
Token sex
Feature activation+0.000
crime
Token crime
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
using
Token using
Feature activation+0.000
a
Token a
Feature activation+0.000
computer
Token computer
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
a
Token a
Feature activation+1.395
sex
Token sex
Feature activation+0.000
crime
Token crime
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
(
Token (
Feature activation+0.000
C
TokenC
Feature activation+0.000
SC
TokenSC
Feature activation+0.000
)
Token)
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+1.033
connectivity
Token connectivity
Feature activation+0.000
for
Token for
Feature activation+0.000
digital
Token digital
Feature activation+0.000
services
Token services
Feature activation+0.000
in
Token in
Feature activation+0.000
by
Token by
Feature activation+0.000
Palestinians
Token Palestinians
Feature activation+0.000
and
Token and
Feature activation+0.000
formal
Token formal
Feature activation+0.000
izing
Tokenizing
Feature activation+0.000
an
Token an
Feature activation+1.023
internal
Token internal
Feature activation+0.000
system
Token system
Feature activation+0.000
of
Token of
Feature activation+0.000
movement
Token movement
Feature activation+0.000
restrictions
Token restrictions
Feature activation+0.000
ensure
Token ensure
Feature activation+0.000
,
Token,
Feature activation+0.000
even
Token even
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
,
Token,
Feature activation+0.875
people
Token people
Feature activation+1.403
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
ability
Token ability
Feature activation+0.079
to
Token to
Feature activation+0.000

INTERVAL 0.471 - 0.943
CONTAINS 0.001%

offer
Token offer
Feature activation+0.000
such
Token such
Feature activation+0.000
pay
Token pay
Feature activation+0.000
offs
Tokenoffs
Feature activation+0.000
to
Token to
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.606
their
Token their
Feature activation+2.928
success
Token success
Feature activation+0.000
in
Token in
Feature activation+0.000
these
Token these
Feature activation+0.000
new
Token new
Feature activation+0.000
successful
Token successful
Feature activation+0.000
in
Token in
Feature activation+0.000
reducing
Token reducing
Feature activation+0.000
unemployment
Token unemployment
Feature activation+0.000
while
Token while
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.893
a
Token a
Feature activation+2.710
rewarding
Token rewarding
Feature activation+0.007
work
Token work
Feature activation+0.000
environment
Token environment
Feature activation+0.000
.
Token.
Feature activation+0.000
prosecutors
Token prosecutors
Feature activation+0.000
said
Token said
Feature activation+0.000
had
Token had
Feature activation+0.000
facilitated
Token facilitated
Feature activation+0.221
at
Token at
Feature activation+0.894
least
Token least
Feature activation+0.912
$
Token $
Feature activation+0.000
180
Token180
Feature activation+0.000
million
Token million
Feature activation+0.000
in
Token in
Feature activation+0.000
sales
Token sales
Feature activation+0.000
in
Token in
Feature activation+0.000
operating
Token operating
Feature activation+0.000
expenses
Token expenses
Feature activation+0.000
may
Token may
Feature activation+0.000
facilitate
Token facilitate
Feature activation+0.000
a
Token a
Feature activation+0.892
two
Token two
Feature activation+0.000
-
Token-
Feature activation+0.000
tier
Tokentier
Feature activation+0.000
pricing
Token pricing
Feature activation+0.000
structure
Token structure
Feature activation+0.000
own
Token own
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
is
Token is
Feature activation+0.000
facilitating
Token facilitating
Feature activation+0.000
social
Token social
Feature activation+0.633
connected
Token connected
Feature activation+0.000
ness
Tokenness
Feature activation+0.000
rather
Token rather
Feature activation+0.000
than
Token than
Feature activation+0.000
divide
Token divide
Feature activation+0.000

INTERVAL 0.000 - 0.471
CONTAINS 99.998%

liver
Token liver
Feature activation+0.000
to
Token to
Feature activation+0.000
regenerate
Token regenerate
Feature activation+0.000
is
Token is
Feature activation+0.000
central
Token central
Feature activation+0.000
to
Token to
Feature activation+0.000
liver
Token liver
Feature activation+0.000
home
Token home
Feature activation+0.000
ost
Tokenost
Feature activation+0.000
asis
Tokenasis
Feature activation+0.000
.
Token.
Feature activation+0.000
faint
Token faint
Feature activation+0.000
lumin
Token lumin
Feature activation+0.000
ance
Tokenance
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
flashlight
Token flashlight
Feature activation+0.000
guiding
Token guiding
Feature activation+0.000
the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
.
Token.
Feature activation+0.000
A
Token A
Feature activation+0.000
io
Tokenio
Feature activation+0.000
P
Token P
Feature activation+0.000
iola
Tokeniola
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
all
Token all
Feature activation+0.000
-
Token-
Feature activation+0.000
time
Tokentime
Feature activation+0.000
Serie
Token Serie
Feature activation+0.000
A
Token A
Feature activation+0.000
Van
Token Van
Feature activation+0.000
Holl
Token Holl
Feature activation+0.000
en
Tokenen
Feature activation+0.000
(
Token (
Feature activation+0.000
D
TokenD
Feature activation+0.000
-
Token-
Feature activation+0.000
MD
TokenMD
Feature activation+0.000
)
Token)
Feature activation+0.000
said
Token said
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
breach
Token breach
Feature activation+0.000
by
Token by
Feature activation+0.000
The
Token The
Feature activation+0.000
Guardian
Token Guardian
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Around
TokenAround
Feature activation+0.000
90
Token 90
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 10: Follows "ir"

TOP ACTIVATIONS
MAX = 3.379

Front
Token Front
Feature activation+0.000
(
Token (
Feature activation+0.000
T
TokenT
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+3.379
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
Hay
Token Hay
Feature activation+0.000
'
Token'
Feature activation+0.000
at
Tokenat
Feature activation+0.000
report
Token report
Feature activation+0.000
found
Token found
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.939
,
Token,
Feature activation+0.000
Lebanon
Token Lebanon
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
DAM
Token DAM
Feature activation+0.000
LE
TokenLE
Feature activation+0.000
TT
TokenTT
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.855
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Bishop
Token Bishop
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
Matthews
Token Matthews
Feature activation+0.000
rover
Token rover
Feature activation+0.000
status
Token status
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.793
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Syria
Token Syria
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.770
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
The
Token The
Feature activation+0.000
com
Tokencom
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.708
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Syrian
Token Syrian
Feature activation+0.000
IES
TokenIES
Feature activation+0.000
ON
TokenON
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.696
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
By
Token By
Feature activation+0.000
2022
Token 2022
Feature activation+0.000
,
Token,
Feature activation+0.000
section
Token section
Feature activation+0.000
below
Token below
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.638
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
The
Token The
Feature activation+0.000
UN
TokenUN
Feature activation+0.000
ICH
TokenICH
Feature activation+0.000
/
Token/
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.580
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Russia
Token Russia
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
at
Tokenat
Feature activation+0.000
Tah
Token Tah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+2.437
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
(
Token (
Feature activation+0.000
H
TokenH
Feature activation+0.000
TS
TokenTS
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
IC
TokenIC
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.267
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Hundreds
Token Hundreds
Feature activation+0.000
gathered
Token gathered
Feature activation+0.000
in
Token in
Feature activation+0.000
Hay
Token Hay
Feature activation+0.000
'
Token'
Feature activation+0.000
at
Tokenat
Feature activation+0.000
Tah
Token Tah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+2.246
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
accused
Token accused
Feature activation+0.000
Assad
Token Assad
Feature activation+0.000
and
Token and
Feature activation+0.000
NI
TokenNI
Feature activation+0.000
AC
TokenAC
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.235
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Christ
Token Christ
Feature activation+0.000
Church
Token Church
Feature activation+0.000
Cathedral
Token Cathedral
Feature activation+0.000
law
Token law
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
M
TokenM
Feature activation+0.000
ETA
TokenETA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
IE
TokenIE
Feature activation+2.164
,
Token,
Feature activation+0.000
La
Token La
Feature activation+0.000
.
Token.
Feature activation+0.000
--
Token --
Feature activation+0.000
Drew
Token Drew
Feature activation+0.000
the
Token the
Feature activation+0.000
members
Token members
Feature activation+0.000
of
Token of
Feature activation+0.000
Ah
Token Ah
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+2.119
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
and
Token and
Feature activation+0.000
Je
Token Je
Feature activation+0.000
ish
Tokenish
Feature activation+0.000
AV
TokenAV
Feature activation+0.000
ES
TokenES
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.013
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Bishop
Token Bishop
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
Matthews
Token Matthews
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ah
TokenAh
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+1.970
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
's
Token's
Feature activation+0.000
advances
Token advances
Feature activation+0.000
come
Token come
Feature activation+0.000
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Be
TokenBe
Feature activation+0.000
ir
Tokenir
Feature activation+0.000
ut
Tokenut
Feature activation+1.789
(
Token (
Feature activation+0.000
AFP
TokenAFP
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
US
Token US
Feature activation+0.000
month
Token month
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
LA
Token LA
Feature activation+1.773
WN
TokenWN
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
Two
Token Two
Feature activation+0.000
restaurants
Token restaurants
Feature activation+0.000
will
Token will
Feature activation+0.000
version
Token version
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
M
TokenM
Feature activation+0.000
ETA
TokenETA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
IE
TokenIE
Feature activation+1.768
,
Token,
Feature activation+0.000
La
Token La
Feature activation+0.000
.
Token.
Feature activation+0.000
--
Token --
Feature activation+0.000
Kenny
Token Kenny
Feature activation+0.000

Top DFA by src position
MAX = 9.552

Nusra
TokenNusra
Feature activation+0.253
Front
Token Front
Feature activation+0.096
(
Token (
Feature activation-0.015
T
TokenT
Feature activation-0.036
ah
Tokenah
Feature activation+0.010
rir
Tokenrir
Feature activation+9.552
al
Token al
Feature activation+0.306
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
Hay
Token Hay
Feature activation+0.000
'
Token'
Feature activation+0.000
the
Token the
Feature activation+0.002
report
Token report
Feature activation+0.004
found
Token found
Feature activation-0.001
<|endoftext|>
Token<|endoftext|>
Feature activation-0.146
BE
TokenBE
Feature activation+0.184
IR
TokenIR
Feature activation+9.144
UT
TokenUT
Feature activation+0.131
,
Token,
Feature activation+0.000
Lebanon
Token Lebanon
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
HAL
Token HAL
Feature activation+0.025
LE
TokenLE
Feature activation-0.031
TT
TokenTT
Feature activation+0.021
/
Token/
Feature activation+0.973
FA
TokenFA
Feature activation+0.091
IR
TokenIR
Feature activation+7.997
FA
TokenFA
Feature activation-0.056
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Bishop
Token Bishop
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
and
Token and
Feature activation-0.001
rover
Token rover
Feature activation-0.005
status
Token status
Feature activation+0.004
<|endoftext|>
Token<|endoftext|>
Feature activation-0.217
BE
TokenBE
Feature activation+0.178
IR
TokenIR
Feature activation+9.061
UT
TokenUT
Feature activation+0.083
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
he
Token he
Feature activation+0.009
said
Token said
Feature activation-0.001
.
Token.
Feature activation+0.002
<|endoftext|>
Token<|endoftext|>
Feature activation-0.305
BE
TokenBE
Feature activation+0.185
IR
TokenIR
Feature activation+9.069
UT
TokenUT
Feature activation+0.168
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
.
Token.
Feature activation-0.001
com
Tokencom
Feature activation+0.005
.
Token.
Feature activation+0.004
<|endoftext|>
Token<|endoftext|>
Feature activation-0.353
BE
TokenBE
Feature activation+0.142
IR
TokenIR
Feature activation+8.979
UT
TokenUT
Feature activation+0.289
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-5.695
IES
TokenIES
Feature activation+0.194
ON
TokenON
Feature activation+0.086
/
Token/
Feature activation+0.274
FA
TokenFA
Feature activation+0.114
IR
TokenIR
Feature activation+8.410
FA
TokenFA
Feature activation+0.251
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
By
Token By
Feature activation+0.000
2022
Token 2022
Feature activation+0.000
comment
Token comment
Feature activation-0.000
section
Token section
Feature activation+0.003
below
Token below
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.207
BE
TokenBE
Feature activation+0.167
IR
TokenIR
Feature activation+8.917
UT
TokenUT
Feature activation+0.121
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
M
TokenM
Feature activation+0.043
UN
TokenUN
Feature activation+0.101
ICH
TokenICH
Feature activation+0.133
/
Token/
Feature activation+0.135
BE
TokenBE
Feature activation-0.062
IR
TokenIR
Feature activation+9.165
UT
TokenUT
Feature activation+0.081
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Hay
Token Hay
Feature activation-0.121
âĢ
TokenâĢ
Feature activation+0.170
Ļ
TokenĻ
Feature activation-0.018
at
Tokenat
Feature activation+0.402
Tah
Token Tah
Feature activation+0.116
rir
Tokenrir
Feature activation+7.585
al
Token al
Feature activation+0.205
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
(
Token (
Feature activation+0.000
H
TokenH
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-5.599
AN
TokenAN
Feature activation+0.071
IC
TokenIC
Feature activation+0.190
/
Token/
Feature activation+0.315
FA
TokenFA
Feature activation+0.166
IR
TokenIR
Feature activation+7.434
FA
TokenFA
Feature activation+0.629
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Hundreds
Token Hundreds
Feature activation+0.000
gathered
Token gathered
Feature activation+0.000
linked
Token linked
Feature activation+0.019
Hay
Token Hay
Feature activation-0.024
'
Token'
Feature activation+0.047
at
Tokenat
Feature activation+0.389
Tah
Token Tah
Feature activation+0.387
rir
Tokenrir
Feature activation+7.648
al
Token al
Feature activation+0.260
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
accused
Token accused
Feature activation+0.000
Assad
Token Assad
Feature activation+0.000
Z
TokenZ
Feature activation+0.008
NI
TokenNI
Feature activation+0.023
AC
TokenAC
Feature activation+0.043
/
Token/
Feature activation+0.651
FA
TokenFA
Feature activation+0.177
IR
TokenIR
Feature activation+7.809
FA
TokenFA
Feature activation+0.182
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Christ
Token Christ
Feature activation+0.000
Church
Token Church
Feature activation+0.000
similar
Token similar
Feature activation-0.019
law
Token law
Feature activation-0.002
<|endoftext|>
Token<|endoftext|>
Feature activation-0.023
M
TokenM
Feature activation+0.057
ETA
TokenETA
Feature activation+0.352
IR
TokenIR
Feature activation+7.114
IE
TokenIE
Feature activation+0.940
,
Token,
Feature activation+0.000
La
Token La
Feature activation+0.000
.
Token.
Feature activation+0.000
--
Token --
Feature activation+0.000
of
Token of
Feature activation+0.011
the
Token the
Feature activation+0.063
members
Token members
Feature activation+0.201
of
Token of
Feature activation+0.091
Ah
Token Ah
Feature activation-0.257
rar
Tokenrar
Feature activation+8.265
al
Token al
Feature activation+0.385
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
and
Token and
Feature activation+0.000
Je
Token Je
Feature activation+0.000
GRE
TokenGRE
Feature activation-0.025
AV
TokenAV
Feature activation+0.004
ES
TokenES
Feature activation+0.012
/
Token/
Feature activation+1.030
FA
TokenFA
Feature activation+0.134
IR
TokenIR
Feature activation+6.735
FA
TokenFA
Feature activation+0.028
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Bishop
Token Bishop
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
there
Token there
Feature activation+0.004
.
Token.
Feature activation+0.037
Ċ
TokenĊ
Feature activation-0.050
Ċ
TokenĊ
Feature activation-0.103
Ah
TokenAh
Feature activation+0.055
rar
Tokenrar
Feature activation+8.712
al
Token al
Feature activation+0.245
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
's
Token's
Feature activation+0.000
advances
Token advances
Feature activation+0.000
Mohammed
Token Mohammed
Feature activation-0.055
)
Token)
Feature activation+0.065
Ċ
TokenĊ
Feature activation+0.383
Ċ
TokenĊ
Feature activation+0.562
Be
TokenBe
Feature activation+0.118
ir
Tokenir
Feature activation+7.994
ut
Tokenut
Feature activation-0.269
(
Token (
Feature activation+0.000
AFP
TokenAFP
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
one
Token one
Feature activation+0.001
month
Token month
Feature activation+0.001
.
Token.
Feature activation+0.004
<|endoftext|>
Token<|endoftext|>
Feature activation-0.034
FA
TokenFA
Feature activation+0.920
IR
TokenIR
Feature activation+7.275
LA
Token LA
Feature activation-0.053
WN
TokenWN
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
Two
Token Two
Feature activation+0.000
restaurants
Token restaurants
Feature activation+0.000
able
Tokenable
Feature activation-0.014
version
Token version
Feature activation-0.009
<|endoftext|>
Token<|endoftext|>
Feature activation+0.038
M
TokenM
Feature activation+0.061
ETA
TokenETA
Feature activation+0.311
IR
TokenIR
Feature activation+6.797
IE
TokenIE
Feature activation+1.011
,
Token,
Feature activation+0.000
La
Token La
Feature activation+0.000
.
Token.
Feature activation+0.000
--
Token --
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.02

Head 2: 0.07

Head 3: 0.09

Head 4: 0.06

Head 5: 0.02

Head 6: 0.42

Head 7: 0.05

Head 8: 0.04

Head 9: 0.05

Head 10: 0.05

Head 11: 0.09

Positive logits

ForgeModLoader1.42

TextColor1.39

wildfire1.33

tariffs1.32

utenberg1.26

specificity1.26

istance1.25

constitu1.23

caption1.23

duty1.22

pmwiki1.22

contention1.21

1.18

nen1.17

grun1.17

fury1.17

assetsadobe1.16

ELD1.16

Rog1.15

narrowed1.14

Negative logits

ciating-2.08

thia-1.78

orative-1.76

oyer-1.75

iour-1.67

hower-1.64

olitical-1.49

adier-1.48

eus-1.46

olate-1.43

cially-1.43

cellent-1.40

admin-1.39

}.-1.38

guiActiveUn-1.37

ulative-1.36

icio-1.36

eless-1.32

enne-1.32

erella-1.32

INTERVAL 3.041 - 3.379
CONTAINS 0.000%

Front
Token Front
Feature activation+0.000
(
Token (
Feature activation+0.000
T
TokenT
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+3.379
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
Hay
Token Hay
Feature activation+0.000
'
Token'
Feature activation+0.000
at
Tokenat
Feature activation+0.000

INTERVAL 2.703 - 3.041
CONTAINS 0.000%

com
Tokencom
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.708
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Syrian
Token Syrian
Feature activation+0.000
rover
Token rover
Feature activation+0.000
status
Token status
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.793
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Syria
Token Syria
Feature activation+0.000
LE
TokenLE
Feature activation+0.000
TT
TokenTT
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.855
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Bishop
Token Bishop
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
Matthews
Token Matthews
Feature activation+0.000
report
Token report
Feature activation+0.000
found
Token found
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.939
,
Token,
Feature activation+0.000
Lebanon
Token Lebanon
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
DAM
Token DAM
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.770
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
The
Token The
Feature activation+0.000

INTERVAL 2.365 - 2.703
CONTAINS 0.000%

âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
at
Tokenat
Feature activation+0.000
Tah
Token Tah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+2.437
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
(
Token (
Feature activation+0.000
H
TokenH
Feature activation+0.000
TS
TokenTS
Feature activation+0.000
UN
TokenUN
Feature activation+0.000
ICH
TokenICH
Feature activation+0.000
/
Token/
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.580
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
Russia
Token Russia
Feature activation+0.000
IES
TokenIES
Feature activation+0.000
ON
TokenON
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.696
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
By
Token By
Feature activation+0.000
2022
Token 2022
Feature activation+0.000
,
Token,
Feature activation+0.000
section
Token section
Feature activation+0.000
below
Token below
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
BE
TokenBE
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
UT
TokenUT
Feature activation+2.638
(
Token (
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
)
Token)
Feature activation+0.000
-
Token -
Feature activation+0.000
The
Token The
Feature activation+0.000

INTERVAL 2.027 - 2.365
CONTAINS 0.000%

Hay
Token Hay
Feature activation+0.000
'
Token'
Feature activation+0.000
at
Tokenat
Feature activation+0.000
Tah
Token Tah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+2.246
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
accused
Token accused
Feature activation+0.000
Assad
Token Assad
Feature activation+0.000
and
Token and
Feature activation+0.000
NI
TokenNI
Feature activation+0.000
AC
TokenAC
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.235
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Christ
Token Christ
Feature activation+0.000
Church
Token Church
Feature activation+0.000
Cathedral
Token Cathedral
Feature activation+0.000
law
Token law
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
M
TokenM
Feature activation+0.000
ETA
TokenETA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
IE
TokenIE
Feature activation+2.164
,
Token,
Feature activation+0.000
La
Token La
Feature activation+0.000
.
Token.
Feature activation+0.000
--
Token --
Feature activation+0.000
Drew
Token Drew
Feature activation+0.000
the
Token the
Feature activation+0.000
members
Token members
Feature activation+0.000
of
Token of
Feature activation+0.000
Ah
Token Ah
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+2.119
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
and
Token and
Feature activation+0.000
Je
Token Je
Feature activation+0.000
ish
Tokenish
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
IC
TokenIC
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.267
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Hundreds
Token Hundreds
Feature activation+0.000
gathered
Token gathered
Feature activation+0.000
in
Token in
Feature activation+0.000

INTERVAL 1.689 - 2.027
CONTAINS 0.000%

AV
TokenAV
Feature activation+0.000
ES
TokenES
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+2.013
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Bishop
Token Bishop
Feature activation+0.000
Victoria
Token Victoria
Feature activation+0.000
Matthews
Token Matthews
Feature activation+0.000
Lebanon
Token Lebanon
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Be
TokenBe
Feature activation+0.000
ir
Tokenir
Feature activation+0.000
ut
Tokenut
Feature activation+1.721
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
Christmas
Token Christmas
Feature activation+0.000
celebrations
Token celebrations
Feature activation+0.000
month
Token month
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
LA
Token LA
Feature activation+1.773
WN
TokenWN
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
Two
Token Two
Feature activation+0.000
restaurants
Token restaurants
Feature activation+0.000
will
Token will
Feature activation+0.000
version
Token version
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
M
TokenM
Feature activation+0.000
ETA
TokenETA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
IE
TokenIE
Feature activation+1.768
,
Token,
Feature activation+0.000
La
Token La
Feature activation+0.000
.
Token.
Feature activation+0.000
--
Token --
Feature activation+0.000
Kenny
Token Kenny
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ah
TokenAh
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+1.970
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
's
Token's
Feature activation+0.000
advances
Token advances
Feature activation+0.000
come
Token come
Feature activation+0.000

INTERVAL 1.352 - 1.689
CONTAINS 0.000%

leader
Token leader
Feature activation+0.000
)
Token)
Feature activation+0.000
of
Token of
Feature activation+0.000
Tah
Token Tah
Feature activation+0.000
rir
Tokenrir
Feature activation+0.000
al
Token al
Feature activation+1.615
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
Hay
Token Hay
Feature activation+0.000
'
Token'
Feature activation+0.000
at
Tokenat
Feature activation+0.000
AND
TokenAND
Feature activation+0.000
ERSON
TokenERSON
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+1.401
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
The
Token The
Feature activation+0.000
Anglic
Token Anglic
Feature activation+0.000
an
Tokenan
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ALL
TokenALL
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
NA
Token NA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
N
TokenN
Feature activation+1.455
:
Token:
Feature activation+0.000
This
Token This
Feature activation+0.000
was
Token was
Feature activation+0.000
in
Token in
Feature activation+0.000
September
Token September
Feature activation+0.000
SC
Token SC
Feature activation+0.000
OTT
TokenOTT
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+1.660
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
Ac
Token Ac
Feature activation+0.000
robat
Tokenrobat
Feature activation+0.000
ics
Tokenics
Feature activation+0.000
.
Token.
Feature activation+0.000
ALL
Token ALL
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
NA
Token NA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
N
TokenN
Feature activation+1.602
:
Token:
Feature activation+0.000
[
Token [
Feature activation+0.000
trans
Tokentrans
Feature activation+0.000
lated
Tokenlated
Feature activation+0.000
]
Token]
Feature activation+0.000

INTERVAL 1.014 - 1.352
CONTAINS 0.000%

the
Token the
Feature activation+0.000
Islamist
Token Islamist
Feature activation+0.000
group
Token group
Feature activation+0.000
Ah
Token Ah
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+1.212
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
and
Token and
Feature activation+0.000
another
Token another
Feature activation+0.000
unit
Token unit
Feature activation+0.000
I
TokenI
Feature activation+0.000
AMS
TokenAMS
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+1.020
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
New
Token New
Feature activation+0.000
York
Token York
Feature activation+0.000
Yankees
Token Yankees
Feature activation+0.000
RO
TokenRO
Feature activation+0.000
LL
TokenLL
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+1.151
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
G
Token G
Feature activation+0.000
NS
TokenNS
Feature activation+0.000
Science
Token Science
Feature activation+0.000
activity
Token activity
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
â̦
Tokenâ̦
Feature activation+1.068
hindsight
Token hindsight
Feature activation+0.000
â̦
Tokenâ̦
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
If
TokenIf
Feature activation+0.000
Y
TokenY
Feature activation+0.000
TER
TokenTER
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+1.098
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
The
Token The
Feature activation+0.000
Panc
Token Panc
Feature activation+0.000
ake
Tokenake
Feature activation+0.000

INTERVAL 0.676 - 1.014
CONTAINS 0.000%

,
Token,
Feature activation+0.000
high
Token high
Feature activation+0.000
quality
Token quality
Feature activation+0.000
content
Token content
Feature activation+0.000
would
Token would
Feature activation+0.000
I
Token I
Feature activation+0.846
look
Token look
Feature activation+0.000
to
Token to
Feature activation+0.000
generate
Token generate
Feature activation+0.000
an
Token an
Feature activation+0.000
income
Token income
Feature activation+0.000
a
Token a
Feature activation+0.000
spokesman
Token spokesman
Feature activation+0.000
for
Token for
Feature activation+0.000
Ah
Token Ah
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+0.908
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
,
Token,
Feature activation+0.000
one
Token one
Feature activation+0.000
of
Token of
Feature activation+0.000
I
TokenI
Feature activation+0.000
AMS
TokenAMS
Feature activation+0.000
/
Token/
Feature activation+0.000
FA
TokenFA
Feature activation+0.000
IR
TokenIR
Feature activation+0.000
FA
TokenFA
Feature activation+0.711
X
TokenX
Feature activation+0.000
NZ
Token NZ
Feature activation+0.000
J
Token J
Feature activation+0.000
ai
Tokenai
Feature activation+0.000
P
Token P
Feature activation+0.000
Syria
Token Syria
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
TE
TokenTE
Feature activation+0.000
HR
TokenHR
Feature activation+0.000
AN
TokenAN
Feature activation+0.839
(
Token (
Feature activation+0.000
F
TokenF
Feature activation+0.000
NA
TokenNA
Feature activation+0.000
)-
Token)-
Feature activation+0.000
Leader
Token Leader
Feature activation+0.000
the
Token the
Feature activation+0.000
Sunni
Token Sunni
Feature activation+0.000
Islamist
Token Islamist
Feature activation+0.000
Ah
Token Ah
Feature activation+0.000
rar
Tokenrar
Feature activation+0.000
al
Token al
Feature activation+0.987
-
Token-
Feature activation+0.000
Sham
TokenSham
Feature activation+0.000
said
Token said
Feature activation+0.000
its
Token its
Feature activation+0.000
fighters
Token fighters
Feature activation+0.000

INTERVAL 0.338 - 0.676
CONTAINS 0.000%

(
Token(
Feature activation+0.000
CNN
TokenCNN
Feature activation+0.000
)
Token)
Feature activation+0.000
He
Token He
Feature activation+0.000
ir
Tokenir
Feature activation+0.000
to
Token to
Feature activation+0.393
a
Token a
Feature activation+0.000
construction
Token construction
Feature activation+0.000
fortune
Token fortune
Feature activation+0.000
,
Token,
Feature activation+0.000
business
Token business
Feature activation+0.000
I
Token I
Feature activation+0.000
manuel
Tokenmanuel
Feature activation+0.000
,
Token,
Feature activation+0.000
Mish
Token Mish
Feature activation+0.000
or
Tokenor
Feature activation+0.000
Ad
Token Ad
Feature activation+0.568
um
Tokenum
Feature activation+0.000
im
Tokenim
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
Bark
Token Bark
Feature activation+0.000

INTERVAL 0.000 - 0.338
CONTAINS 100.000%

against
Token against
Feature activation+0.000
this
Token this
Feature activation+0.000
unw
Token unw
Feature activation+0.000
arranted
Tokenarranted
Feature activation+0.000
act
Token act
Feature activation+0.000
by
Token by
Feature activation+0.000
DOJ
Token DOJ
Feature activation+0.000
âĢĶ
Token âĢĶ
Feature activation+0.000
Keith
Token Keith
Feature activation+0.000
Ol
Token Ol
Feature activation+0.000
ber
Tokenber
Feature activation+0.000
sense
Token sense
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
world
Token world
Feature activation+0.000
,
Token,
Feature activation+0.000
we
Token we
Feature activation+0.000
may
Token may
Feature activation+0.000
harm
Token harm
Feature activation+0.000
lessly
Tokenlessly
Feature activation+0.000
come
Token come
Feature activation+0.000
across
Token across
Feature activation+0.000
the
Token the
Feature activation+0.000
produce
Token produce
Feature activation+0.000
to
Token to
Feature activation+0.000
shine
Token shine
Feature activation+0.000
;
Token;
Feature activation+0.000
his
Token his
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
reci
Tokenreci
Feature activation+0.000
pe
Tokenpe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
two
Token two
Feature activation+0.000
people
Token people
Feature activation+0.000
were
Token were
Feature activation+0.000
detained
Token detained
Feature activation+0.000
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
police
Token police
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
On
TokenOn
Feature activation+0.000
the
Token the
Feature activation+0.000
perfect
Token perfect
Feature activation+0.000
example
Token example
Feature activation+0.000
of
Token of
Feature activation+0.000
so
Token so
Feature activation+0.000
many
Token many
Feature activation+0.000
Christians
Token Christians
Feature activation+0.000
who
Token who
Feature activation+0.000
have
Token have
Feature activation+0.000
failed
Token failed
Feature activation+0.000
to
Token to
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 11: Dead

TOP ACTIVATIONS
MAX = 0.030

Cl
Token Cl
Feature activation+0.000
ardy
Tokenardy
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Harvard
Token Harvard
Feature activation+0.000
Medical
Token Medical
Feature activation+0.030
School
Token School
Feature activation+0.000
in
Token in
Feature activation+0.000
Boston
Token Boston
Feature activation+0.000
and
Token and
Feature activation+0.000
their
Token their
Feature activation+0.000
,
Token,
Feature activation+0.000
if
Token if
Feature activation+0.000
not
Token not
Feature activation+0.000
millions
Token millions
Feature activation+0.000
,
Token,
Feature activation+0.000
of
Token of
Feature activation+0.029
young
Token young
Feature activation+0.000
women
Token women
Feature activation+0.000
whose
Token whose
Feature activation+0.000
lives
Token lives
Feature activation+0.000
have
Token have
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000

Top DFA by src position
MAX = 2.452

Jon
Token Jon
Feature activation-0.013
Cl
Token Cl
Feature activation+0.001
ardy
Tokenardy
Feature activation-0.004
at
Token at
Feature activation-0.013
the
Token the
Feature activation-0.004
Harvard
Token Harvard
Feature activation+2.452
Medical
Token Medical
Feature activation-0.068
School
Token School
Feature activation+0.000
in
Token in
Feature activation+0.000
Boston
Token Boston
Feature activation+0.000
and
Token and
Feature activation+0.000
for
Token for
Feature activation+0.348
thousands
Token thousands
Feature activation+0.329
,
Token,
Feature activation+0.386
if
Token if
Feature activation-0.047
not
Token not
Feature activation+0.001
millions
Token millions
Feature activation+1.644
,
Token,
Feature activation+0.035
of
Token of
Feature activation+0.071
young
Token young
Feature activation+0.000
women
Token women
Feature activation+0.000
whose
Token whose
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.009
less
Token less
Feature activation-0.002
lofty
Token lofty
Feature activation-0.010
goal
Token goal
Feature activation-0.014
:
Token:
Feature activation+0.024
Put
Token Put
Feature activation+0.524
a
Token a
Feature activation-0.218
Toronto
Token Toronto
Feature activation-0.079
Blue
Token Blue
Feature activation-0.017
Jays
Token Jays
Feature activation-0.021
âĢ
TokenâĢ
Feature activation+0.009
somewhat
Token somewhat
Feature activation-0.006
less
Token less
Feature activation+0.003
lofty
Token lofty
Feature activation+0.020
goal
Token goal
Feature activation-0.025
:
Token:
Feature activation+0.039
Put
Token Put
Feature activation+0.106
a
Token a
Feature activation-0.050
Toronto
Token Toronto
Feature activation-0.219
Blue
Token Blue
Feature activation-0.027
Jays
Token Jays
Feature activation-0.022
âĢ
TokenâĢ
Feature activation+0.020
,
Token,
Feature activation-0.298
the
Token the
Feature activation-0.046
NB
Token NB
Feature activation-0.027
Space
Token Space
Feature activation-0.081
Race
Token Race
Feature activation-0.244
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
,
Token,
Feature activation-0.226
the
Token the
Feature activation-0.113
NB
Token NB
Feature activation+0.025
Space
Token Space
Feature activation-0.006
Race
Token Race
Feature activation-0.453
had
Token had
Feature activation+0.061
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
Put
Token Put
Feature activation+0.254
a
Token a
Feature activation-0.023
Toronto
Token Toronto
Feature activation-0.122
Blue
Token Blue
Feature activation-0.011
Jays
Token Jays
Feature activation-0.038
âĢ
TokenâĢ
Feature activation+0.348
Ļ
TokenĻ
Feature activation+0.005
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Put
Token Put
Feature activation+0.156
a
Token a
Feature activation-0.023
Toronto
Token Toronto
Feature activation-0.099
Blue
Token Blue
Feature activation-0.027
Jays
Token Jays
Feature activation-0.036
âĢ
TokenâĢ
Feature activation+0.425
Ļ
TokenĻ
Feature activation-0.092
baseball
Token baseball
Feature activation-0.028
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
lofty
Token lofty
Feature activation-0.000
goal
Token goal
Feature activation-0.011
:
Token:
Feature activation+0.007
Put
Token Put
Feature activation+0.048
a
Token a
Feature activation-0.008
Toronto
Token Toronto
Feature activation+0.168
Blue
Token Blue
Feature activation-0.121
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.005
less
Token less
Feature activation+0.009
lofty
Token lofty
Feature activation+0.007
goal
Token goal
Feature activation-0.028
:
Token:
Feature activation+0.020
Put
Token Put
Feature activation+0.541
a
Token a
Feature activation-0.198
Toronto
Token Toronto
Feature activation+0.011
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Race
Token Race
Feature activation-0.241
had
Token had
Feature activation-0.119
a
Token a
Feature activation-0.120
somewhat
Token somewhat
Feature activation-0.077
less
Token less
Feature activation-0.093
lofty
Token lofty
Feature activation+0.026
goal
Token goal
Feature activation-0.113
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
a
Token a
Feature activation-0.036
somewhat
Token somewhat
Feature activation-0.036
less
Token less
Feature activation-0.057
lofty
Token lofty
Feature activation-0.001
goal
Token goal
Feature activation-0.004
:
Token:
Feature activation+0.105
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.007
less
Token less
Feature activation+0.015
lofty
Token lofty
Feature activation+0.007
goal
Token goal
Feature activation-0.033
:
Token:
Feature activation+0.027
Put
Token Put
Feature activation+0.337
a
Token a
Feature activation-0.032
Toronto
Token Toronto
Feature activation-0.067
Blue
Token Blue
Feature activation-0.175
Jays
Token Jays
Feature activation-0.157
âĢ
TokenâĢ
Feature activation+0.000
:
Token:
Feature activation+0.016
Put
Token Put
Feature activation+0.213
a
Token a
Feature activation-0.022
Toronto
Token Toronto
Feature activation-0.184
Blue
Token Blue
Feature activation-0.038
Jays
Token Jays
Feature activation+0.219
âĢ
TokenâĢ
Feature activation+0.032
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation-0.171
the
Token the
Feature activation-0.109
NB
Token NB
Feature activation-0.014
Space
Token Space
Feature activation+0.002
Race
Token Race
Feature activation-0.230
had
Token had
Feature activation+0.171
a
Token a
Feature activation+0.055
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
NB
Token NB
Feature activation-0.020
Space
Token Space
Feature activation-0.016
Race
Token Race
Feature activation-0.041
had
Token had
Feature activation-0.029
a
Token a
Feature activation-0.037
somewhat
Token somewhat
Feature activation+0.066
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.036
less
Token less
Feature activation-0.029
lofty
Token lofty
Feature activation-0.006
goal
Token goal
Feature activation-0.060
:
Token:
Feature activation+0.046
Put
Token Put
Feature activation+0.091
a
Token a
Feature activation+0.065
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.009
less
Token less
Feature activation+0.008
lofty
Token lofty
Feature activation+0.038
goal
Token goal
Feature activation-0.096
:
Token:
Feature activation+0.068
Put
Token Put
Feature activation+0.104
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
had
Token had
Feature activation-0.059
a
Token a
Feature activation-0.025
somewhat
Token somewhat
Feature activation-0.128
less
Token less
Feature activation-0.048
lofty
Token lofty
Feature activation-0.065
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
NB
Token NB
Feature activation-0.027
Space
Token Space
Feature activation-0.004
Race
Token Race
Feature activation-0.034
had
Token had
Feature activation-0.026
a
Token a
Feature activation-0.025
somewhat
Token somewhat
Feature activation+0.002
less
Token less
Feature activation-0.114
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.07

Head 2: 0.09

Head 3: 0.07

Head 4: 0.08

Head 5: 0.09

Head 6: 0.09

Head 7: 0.08

Head 8: 0.08

Head 9: 0.09

Head 10: 0.09

Head 11: 0.09

Positive logits

ylan1.94

quit1.92

fund1.80

end1.79

venants1.78

ories1.74

fal1.72

rf1.70

add1.69

fin1.67

conn1.64

endum1.64

ender1.62

git1.62

Donation1.62

entin1.61

aniel1.61

region1.60

own1.60

ribut1.60

Negative logits

ALSE-2.17

enthusi-2.07

unbeliev-2.00

pse-1.88

unden-1.87

cele-1.72

advoc-1.72

Electro-1.71

incendiary-1.71

satell-1.69

occas-1.67

suspic-1.66

scrut-1.66

Diesel-1.66

ocally-1.66

cloaked-1.64

celeb-1.59

shenan-1.59

gastro-1.57

nodd-1.56

INTERVAL 0.027 - 0.030
CONTAINS 0.000%

Cl
Token Cl
Feature activation+0.000
ardy
Tokenardy
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
Harvard
Token Harvard
Feature activation+0.000
Medical
Token Medical
Feature activation+0.030
School
Token School
Feature activation+0.000
in
Token in
Feature activation+0.000
Boston
Token Boston
Feature activation+0.000
and
Token and
Feature activation+0.000
their
Token their
Feature activation+0.000
,
Token,
Feature activation+0.000
if
Token if
Feature activation+0.000
not
Token not
Feature activation+0.000
millions
Token millions
Feature activation+0.000
,
Token,
Feature activation+0.000
of
Token of
Feature activation+0.029
young
Token young
Feature activation+0.000
women
Token women
Feature activation+0.000
whose
Token whose
Feature activation+0.000
lives
Token lives
Feature activation+0.000
have
Token have
Feature activation+0.000

INTERVAL 0.024 - 0.027
CONTAINS 0.000%

INTERVAL 0.021 - 0.024
CONTAINS 0.000%

INTERVAL 0.018 - 0.021
CONTAINS 0.000%

INTERVAL 0.015 - 0.018
CONTAINS 0.000%

INTERVAL 0.012 - 0.015
CONTAINS 0.000%

INTERVAL 0.009 - 0.012
CONTAINS 0.000%

INTERVAL 0.006 - 0.009
CONTAINS 0.000%

INTERVAL 0.003 - 0.006
CONTAINS 0.000%

INTERVAL 0.000 - 0.003
CONTAINS 100.000%

(
Token (
Feature activation+0.000
g
Tokeng
Feature activation+0.000
osh
Tokenosh
Feature activation+0.000
,
Token,
Feature activation+0.000
what
Token what
Feature activation+0.000
a
Token a
Feature activation+0.000
wig
Token wig
Feature activation+0.000
that
Token that
Feature activation+0.000
'd
Token'd
Feature activation+0.000
be
Token be
Feature activation+0.000
,
Token,
Feature activation+0.000
Center
Token Center
Feature activation+0.000
for
Token for
Feature activation+0.000
Immigration
Token Immigration
Feature activation+0.000
Studies
Token Studies
Feature activation+0.000
,
Token,
Feature activation+0.000
are
Token are
Feature activation+0.000
calling
Token calling
Feature activation+0.000
on
Token on
Feature activation+0.000
Trump
Token Trump
Feature activation+0.000
to
Token to
Feature activation+0.000
reject
Token reject
Feature activation+0.000
website
Token website
Feature activation+0.000
stated
Token stated
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
N
Token N
Feature activation+0.000
PS
TokenPS
Feature activation+0.000
website
Token website
Feature activation+0.000
also
Token also
Feature activation+0.000
tells
Token tells
Feature activation+0.000
to
Token to
Feature activation+0.000
watch
Token watch
Feature activation+0.000
at
Token at
Feature activation+0.000
whatever
Token whatever
Feature activation+0.000
time
Token time
Feature activation+0.000
and
Token and
Feature activation+0.000
on
Token on
Feature activation+0.000
whatever
Token whatever
Feature activation+0.000
day
Token day
Feature activation+0.000
is
Token is
Feature activation+0.000
most
Token most
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
She
TokenShe
Feature activation+0.000
also
Token also
Feature activation+0.000
applied
Token applied
Feature activation+0.000
for
Token for
Feature activation+0.000
federal
Token federal
Feature activation+0.000
disability
Token disability
Feature activation+0.000
assistance
Token assistance
Feature activation+0.000
,
Token,
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 12: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.599

Jays
Token Jays
Feature activation+0.001
âĢ
TokenâĢ
Feature activation+0.025
Ļ
TokenĻ
Feature activation+0.025
baseball
Token baseball
Feature activation-0.007
cap
Token cap
Feature activation+0.004
and
Token and
Feature activation+0.199
a
Token a
Feature activation+0.147
beer
Token beer
Feature activation+0.043
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Jays
Token Jays
Feature activation+0.013
âĢ
TokenâĢ
Feature activation+0.022
Ļ
TokenĻ
Feature activation+0.027
baseball
Token baseball
Feature activation+0.093
cap
Token cap
Feature activation-0.128
and
Token and
Feature activation+0.194
a
Token a
Feature activation+0.061
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Toronto
Token Toronto
Feature activation-0.077
Blue
Token Blue
Feature activation+0.020
Jays
Token Jays
Feature activation+0.071
âĢ
TokenâĢ
Feature activation+0.115
Ļ
TokenĻ
Feature activation+0.081
baseball
Token baseball
Feature activation+0.303
cap
Token cap
Feature activation-0.106
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
Toronto
Token Toronto
Feature activation-0.030
Blue
Token Blue
Feature activation+0.020
Jays
Token Jays
Feature activation+0.014
âĢ
TokenâĢ
Feature activation+0.034
Ļ
TokenĻ
Feature activation+0.045
baseball
Token baseball
Feature activation+0.103
cap
Token cap
Feature activation-0.225
and
Token and
Feature activation+0.046
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
,
Token,
Feature activation-0.265
the
Token the
Feature activation-0.049
NB
Token NB
Feature activation+0.006
Space
Token Space
Feature activation-0.038
Race
Token Race
Feature activation-0.172
had
Token had
Feature activation+0.271
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.808
,
Token,
Feature activation-0.306
the
Token the
Feature activation+0.022
NB
Token NB
Feature activation+0.004
Space
Token Space
Feature activation-0.044
Race
Token Race
Feature activation-0.180
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
,
Token,
Feature activation-0.203
the
Token the
Feature activation-0.079
NB
Token NB
Feature activation-0.015
Space
Token Space
Feature activation-0.049
Race
Token Race
Feature activation-0.091
had
Token had
Feature activation+0.286
a
Token a
Feature activation+0.011
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
the
Token the
Feature activation-0.104
NB
Token NB
Feature activation-0.020
Space
Token Space
Feature activation-0.020
Race
Token Race
Feature activation-0.002
had
Token had
Feature activation+0.094
a
Token a
Feature activation+0.118
somewhat
Token somewhat
Feature activation-0.009
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
lofty
Token lofty
Feature activation+0.002
goal
Token goal
Feature activation+0.002
:
Token:
Feature activation+0.017
Put
Token Put
Feature activation+0.003
a
Token a
Feature activation-0.023
Toronto
Token Toronto
Feature activation+0.300
Blue
Token Blue
Feature activation-0.043
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
lofty
Token lofty
Feature activation+0.001
goal
Token goal
Feature activation+0.011
:
Token:
Feature activation+0.060
Put
Token Put
Feature activation+0.009
a
Token a
Feature activation-0.069
Toronto
Token Toronto
Feature activation+0.071
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
a
Token a
Feature activation-0.023
somewhat
Token somewhat
Feature activation+0.006
less
Token less
Feature activation+0.002
lofty
Token lofty
Feature activation-0.031
goal
Token goal
Feature activation-0.037
:
Token:
Feature activation+0.231
Put
Token Put
Feature activation+0.054
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
a
Token a
Feature activation-0.013
somewhat
Token somewhat
Feature activation-0.031
less
Token less
Feature activation-0.021
lofty
Token lofty
Feature activation-0.007
goal
Token goal
Feature activation+0.039
:
Token:
Feature activation+0.189
Put
Token Put
Feature activation+0.028
a
Token a
Feature activation+0.045
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
:
Token:
Feature activation+0.044
Put
Token Put
Feature activation+0.043
a
Token a
Feature activation-0.002
Toronto
Token Toronto
Feature activation-0.081
Blue
Token Blue
Feature activation-0.005
Jays
Token Jays
Feature activation+0.283
âĢ
TokenâĢ
Feature activation+0.061
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
lofty
Token lofty
Feature activation+0.001
goal
Token goal
Feature activation+0.003
:
Token:
Feature activation+0.071
Put
Token Put
Feature activation+0.025
a
Token a
Feature activation-0.034
Toronto
Token Toronto
Feature activation+0.189
Blue
Token Blue
Feature activation+0.048
Jays
Token Jays
Feature activation-0.039
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
Put
Token Put
Feature activation+0.038
a
Token a
Feature activation-0.004
Toronto
Token Toronto
Feature activation-0.054
Blue
Token Blue
Feature activation+0.028
Jays
Token Jays
Feature activation+0.036
âĢ
TokenâĢ
Feature activation+0.526
Ļ
TokenĻ
Feature activation+0.007
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Put
Token Put
Feature activation+0.017
a
Token a
Feature activation-0.012
Toronto
Token Toronto
Feature activation-0.028
Blue
Token Blue
Feature activation+0.026
Jays
Token Jays
Feature activation+0.002
âĢ
TokenâĢ
Feature activation+0.599
Ļ
TokenĻ
Feature activation+0.052
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
,
Token,
Feature activation-0.416
the
Token the
Feature activation-0.001
NB
Token NB
Feature activation-0.004
Space
Token Space
Feature activation-0.032
Race
Token Race
Feature activation-0.059
had
Token had
Feature activation+0.116
a
Token a
Feature activation+0.048
somewhat
Token somewhat
Feature activation+0.018
less
Token less
Feature activation+0.035
lofty
Token lofty
Feature activation-0.063
goal
Token goal
Feature activation-0.319
the
Token the
Feature activation-0.054
NB
Token NB
Feature activation-0.017
Space
Token Space
Feature activation-0.033
Race
Token Race
Feature activation-0.121
had
Token had
Feature activation+0.030
a
Token a
Feature activation+0.073
somewhat
Token somewhat
Feature activation+0.010
less
Token less
Feature activation-0.009
lofty
Token lofty
Feature activation-0.120
goal
Token goal
Feature activation-0.089
:
Token:
Feature activation+0.000
the
Token the
Feature activation-0.075
NB
Token NB
Feature activation-0.020
Space
Token Space
Feature activation-0.010
Race
Token Race
Feature activation+0.001
had
Token had
Feature activation+0.066
a
Token a
Feature activation+0.105
somewhat
Token somewhat
Feature activation+0.019
less
Token less
Feature activation-0.150
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation-0.080
NB
Token NB
Feature activation-0.035
Space
Token Space
Feature activation-0.024
Race
Token Race
Feature activation-0.026
had
Token had
Feature activation+0.081
a
Token a
Feature activation+0.110
somewhat
Token somewhat
Feature activation+0.014
less
Token less
Feature activation-0.057
lofty
Token lofty
Feature activation-0.427
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.09

Head 1: 0.06

Head 2: 0.08

Head 3: 0.09

Head 4: 0.09

Head 5: 0.07

Head 6: 0.08

Head 7: 0.09

Head 8: 0.09

Head 9: 0.09

Head 10: 0.08

Head 11: 0.09

Positive logits

halting1.72

anx1.69

tightening1.69

ilaterally1.65

tighter1.62

cliffs1.62

shed1.61

iless1.58

environmentally1.58

insecure1.57

une1.55

mounting1.55

Koen1.54

vigilance1.51

precaution1.51

shedding1.50

mit1.50

?????-1.49

exting1.49

widening1.49

Negative logits

RAW-1.67

enery-1.63

efer-1.59

Roose-1.59

estamp-1.57

ulhu-1.57

eree-1.57

football-1.52

qus-1.50

Wrestling-1.50

Created-1.50

affle-1.49

Shots-1.49

videos-1.48

inav-1.48

Title-1.48

osaurus-1.45

mug-1.43

Encyclopedia-1.42

Soccer-1.42

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

stock
Token stock
Feature activation+0.000
market
Token market
Feature activation+0.000
indexes
Token indexes
Feature activation+0.000
,
Token,
Feature activation+0.000
while
Token while
Feature activation+0.000
they
Token they
Feature activation+0.000
both
Token both
Feature activation+0.000
watch
Token watch
Feature activation+0.000
the
Token the
Feature activation+0.000
Dow
Token Dow
Feature activation+0.000
soar
Token soar
Feature activation+0.000
What
TokenWhat
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
more
Token more
Feature activation+0.000
,
Token,
Feature activation+0.000
every
Token every
Feature activation+0.000
bite
Token bite
Feature activation+0.000
felt
Token felt
Feature activation+0.000
like
Token like
Feature activation+0.000
it
Token it
Feature activation+0.000
,
Token,
Feature activation+0.000
weapons
Token weapons
Feature activation+0.000
,
Token,
Feature activation+0.000
living
Token living
Feature activation+0.000
power
Token power
Feature activation+0.000
armor
Token armor
Feature activation+0.000
,
Token,
Feature activation+0.000
monstrous
Token monstrous
Feature activation+0.000
war
Token war
Feature activation+0.000
st
Token st
Feature activation+0.000
eeds
Tokeneeds
Feature activation+0.000
concentration
Token concentration
Feature activation+0.000
numbers
Token numbers
Feature activation+0.000
are
Token are
Feature activation+0.000
almost
Token almost
Feature activation+0.000
as
Token as
Feature activation+0.000
high
Token high
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
newer
Token newer
Feature activation+0.000
cohort
Token cohort
Feature activation+0.000
of
Token of
Feature activation+0.000
match
Token match
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Aut
TokenAut
Feature activation+0.000
omatic
Tokenomatic
Feature activation+0.000
recording
Token recording
Feature activation+0.000
and
Token and
Feature activation+0.000
storage
Token storage
Feature activation+0.000
of
Token of
Feature activation+0.000
GOT
Token GOT
Feature activation+0.000
V
TokenV
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 13: In phrases starting with “competing for”

TOP ACTIVATIONS
MAX = 3.068

between
Token between
Feature activation+0.000
males
Token males
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.549
a
Token a
Feature activation+1.874
female
Token female
Feature activation+3.068
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
E
TokenE
Feature activation+0.000
tymology
Tokentymology
Feature activation+0.000
top
Token top
Feature activation+0.000
clubs
Token clubs
Feature activation+0.000
are
Token are
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.894
entry
Token entry
Feature activation+2.902
into
Token into
Feature activation+1.517
the
Token the
Feature activation+0.864
UEFA
Token UEFA
Feature activation+0.125
Champions
Token Champions
Feature activation+0.000
League
Token League
Feature activation+0.500
also
Token also
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.104
starting
Token starting
Feature activation+1.494
ber
Token ber
Feature activation+0.525
ths
Tokenths
Feature activation+2.709
in
Token in
Feature activation+1.242
the
Token the
Feature activation+0.741
team
Token team
Feature activation+0.310
,
Token,
Feature activation+0.326
and
Token and
Feature activation+0.000
cables
Token cables
Feature activation+0.000
,
Token,
Feature activation+0.000
all
Token all
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.556
the
Token the
Feature activation+2.567
wall
Token wall
Feature activation+1.606
socket
Token socket
Feature activation+1.183
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
sometimes
Token sometimes
Feature activation+0.000
is
Token is
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.873
the
Token the
Feature activation+2.053
top
Token top
Feature activation+2.565
players
Token players
Feature activation+2.483
with
Token with
Feature activation+1.014
clubs
Token clubs
Feature activation+0.487
such
Token such
Feature activation+0.000
as
Token as
Feature activation+0.000
different
Token different
Feature activation+0.000
cities
Token cities
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.425
migrant
Token migrant
Feature activation+1.336
labour
Token labour
Feature activation+2.533
through
Token through
Feature activation+0.947
advertising
Token advertising
Feature activation+0.000
,
Token,
Feature activation+0.000
better
Token better
Feature activation+0.000
international
Token international
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.873
the
Token the
Feature activation+2.053
top
Token top
Feature activation+2.565
players
Token players
Feature activation+2.483
with
Token with
Feature activation+1.014
clubs
Token clubs
Feature activation+0.487
such
Token such
Feature activation+0.000
as
Token as
Feature activation+0.000
Chelsea
Token Chelsea
Feature activation+0.000
competed
Token competed
Feature activation+0.000
for
Token for
Feature activation+0.471
the
Token the
Feature activation+1.589
few
Token few
Feature activation+1.019
unemployed
Token unemployed
Feature activation+0.944
workers
Token workers
Feature activation+2.456
during
Token during
Feature activation+0.000
the
Token the
Feature activation+0.000
first
Token first
Feature activation+0.000
Internet
Token Internet
Feature activation+0.000
boom
Token boom
Feature activation+0.000
facing
Token facing
Feature activation+0.000
disadvantages
Token disadvantages
Feature activation+0.000
in
Token in
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.646
jobs
Token jobs
Feature activation+2.430
.
Token.
Feature activation+0.000
So
Token So
Feature activation+0.000
maybe
Token maybe
Feature activation+0.000
we
Token we
Feature activation+0.000
could
Token could
Feature activation+0.000
sides
Token sides
Feature activation+0.000
are
Token are
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.793
her
Token her
Feature activation+1.575
uterus
Token uterus
Feature activation+2.315
,
Token,
Feature activation+0.398
the
Token the
Feature activation+0.000
future
Token future
Feature activation+0.000
home
Token home
Feature activation+0.000
of
Token of
Feature activation+0.000
once
Token once
Feature activation+0.000
competed
Token competed
Feature activation+0.000
for
Token for
Feature activation+0.988
power
Token power
Feature activation+2.004
and
Token and
Feature activation+0.758
influence
Token influence
Feature activation+2.308
in
Token in
Feature activation+1.041
the
Token the
Feature activation+0.612
Arab
Token Arab
Feature activation+0.000
world
Token world
Feature activation+0.008
,
Token,
Feature activation+0.000
of
Token of
Feature activation+0.000
other
Token other
Feature activation+0.000
players
Token players
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.777
the
Token the
Feature activation+2.297
attention
Token attention
Feature activation+2.018
of
Token of
Feature activation+1.250
struggling
Token struggling
Feature activation+0.519
students
Token students
Feature activation+0.629
agree
Token agree
Feature activation+0.000
trains
Token trains
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.843
people
Token people
Feature activation+2.296
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.189
attention
Token attention
Feature activation+1.154
,
Token,
Feature activation+0.000
their
Token their
Feature activation+0.000
promises
Token promises
Feature activation+0.000
of
Token of
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.819
a
Token a
Feature activation+2.296
Stanley
Token Stanley
Feature activation+0.188
Cup
Token Cup
Feature activation+1.034
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
near
Token near
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
now
Token now
Feature activation+0.000
be
Token be
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.791
a
Token a
Feature activation+2.258
place
Token place
Feature activation+2.195
in
Token in
Feature activation+1.066
the
Token the
Feature activation+1.043
pack
Token pack
Feature activation+1.098
,
Token,
Feature activation+0.542
now
Token now
Feature activation+0.000
be
Token be
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.791
a
Token a
Feature activation+2.258
place
Token place
Feature activation+2.195
in
Token in
Feature activation+1.066
the
Token the
Feature activation+1.043
pack
Token pack
Feature activation+1.098
,
Token,
Feature activation+0.542
adding
Token adding
Feature activation+0.334
Lewis
TokenLewis
Feature activation+0.000
will
Token will
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+1.315
opportunity
Token opportunity
Feature activation+2.152
to
Token to
Feature activation+1.328
be
Token be
Feature activation+1.147
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.141
man
Token man
Feature activation+0.000
5
Token5
Feature activation+0.000
)
Token)
Feature activation+0.000
battles
Token battles
Feature activation+0.000
for
Token for
Feature activation+0.021
the
Token the
Feature activation+1.566
ball
Token ball
Feature activation+2.148
against
Token against
Feature activation+0.229
Eric
Token Eric
Feature activation+0.000
B
Token B
Feature activation+0.000
led
Tokenled
Feature activation+0.000
so
Tokenso
Feature activation+0.000
eting
Tokeneting
Feature activation+0.000
factions
Token factions
Feature activation+0.000
v
Token v
Feature activation+0.000
ied
Tokenied
Feature activation+0.000
for
Token for
Feature activation+0.238
the
Token the
Feature activation+2.140
throne
Token throne
Feature activation+1.987
and
Token and
Feature activation+1.072
counties
Token counties
Feature activation+0.354
(
Token (
Feature activation+0.474
which
Tokenwhich
Feature activation+1.486
and
Token and
Feature activation+0.000
on
Token on
Feature activation+0.000
,
Token,
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.973
that
Token that
Feature activation+2.116
title
Token title
Feature activation+1.649
with
Token with
Feature activation+1.185
the
Token the
Feature activation+0.579
technology
Token technology
Feature activation+0.251
industry
Token industry
Feature activation+0.793

Top DFA by src position
MAX = 3.404

energetic
Token energetic
Feature activation+0.052
fights
Token fights
Feature activation-0.021
between
Token between
Feature activation+0.030
males
Token males
Feature activation+0.273
competing
Token competing
Feature activation+0.502
for
Token for
Feature activation+2.553
a
Token a
Feature activation+0.931
female
Token female
Feature activation+0.315
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
's
Token's
Feature activation+0.011
top
Token top
Feature activation+0.009
clubs
Token clubs
Feature activation+0.047
are
Token are
Feature activation+0.111
competing
Token competing
Feature activation+0.393
for
Token for
Feature activation+3.372
entry
Token entry
Feature activation+0.148
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
UEFA
Token UEFA
Feature activation+0.000
Champions
Token Champions
Feature activation+0.000
e
Tokene
Feature activation+0.002
ke
Tokenke
Feature activation+0.001
are
Token are
Feature activation+0.067
also
Token also
Feature activation+0.056
competing
Token competing
Feature activation+0.311
for
Token for
Feature activation+2.726
starting
Token starting
Feature activation+0.706
ber
Token ber
Feature activation+0.122
ths
Tokenths
Feature activation+0.066
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
and
Token and
Feature activation+0.025
cables
Token cables
Feature activation+0.002
,
Token,
Feature activation+0.033
all
Token all
Feature activation-0.144
competing
Token competing
Feature activation+0.185
for
Token for
Feature activation+3.050
the
Token the
Feature activation+0.642
wall
Token wall
Feature activation+0.000
socket
Token socket
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.103
is
Token is
Feature activation-0.056
to
Token to
Feature activation+0.042
compete
Token compete
Feature activation+0.209
for
Token for
Feature activation+3.130
the
Token the
Feature activation+0.388
top
Token top
Feature activation+0.124
players
Token players
Feature activation+0.000
with
Token with
Feature activation+0.000
clubs
Token clubs
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation-0.032
with
Token with
Feature activation-0.058
different
Token different
Feature activation+0.075
cities
Token cities
Feature activation+0.017
competing
Token competing
Feature activation+0.248
for
Token for
Feature activation+3.006
migrant
Token migrant
Feature activation+0.513
labour
Token labour
Feature activation+0.185
through
Token through
Feature activation+0.000
advertising
Token advertising
Feature activation+0.000
,
Token,
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.155
is
Token is
Feature activation-0.028
to
Token to
Feature activation+0.036
compete
Token compete
Feature activation+0.305
for
Token for
Feature activation+2.551
the
Token the
Feature activation+0.404
top
Token top
Feature activation+0.463
players
Token players
Feature activation+0.078
with
Token with
Feature activation+0.000
clubs
Token clubs
Feature activation+0.000
in
Token in
Feature activation-0.029
1999
Token 1999
Feature activation+0.005
as
Token as
Feature activation-0.043
companies
Token companies
Feature activation+0.123
competed
Token competed
Feature activation+0.239
for
Token for
Feature activation+2.043
the
Token the
Feature activation+0.403
few
Token few
Feature activation+0.388
unemployed
Token unemployed
Feature activation+0.520
workers
Token workers
Feature activation+0.217
during
Token during
Feature activation+0.000
as
Token as
Feature activation+0.061
facing
Token facing
Feature activation-0.038
disadvantages
Token disadvantages
Feature activation+0.019
in
Token in
Feature activation+0.052
competing
Token competing
Feature activation+0.236
for
Token for
Feature activation+2.974
jobs
Token jobs
Feature activation+0.329
.
Token.
Feature activation+0.000
So
Token So
Feature activation+0.000
maybe
Token maybe
Feature activation+0.000
we
Token we
Feature activation+0.000
and
Token and
Feature activation+0.053
both
Token both
Feature activation+0.009
sides
Token sides
Feature activation+0.009
are
Token are
Feature activation+0.033
vying
Token vying
Feature activation+0.268
for
Token for
Feature activation+2.556
her
Token her
Feature activation+0.379
uterus
Token uterus
Feature activation+0.217
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
future
Token future
Feature activation+0.000
Three
TokenThree
Feature activation-0.010
powerful
Token powerful
Feature activation+0.022
states
Token states
Feature activation-0.014
once
Token once
Feature activation-0.032
competed
Token competed
Feature activation+0.239
for
Token for
Feature activation+2.198
power
Token power
Feature activation+0.379
and
Token and
Feature activation+1.237
influence
Token influence
Feature activation+0.077
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
number
Token number
Feature activation+0.018
of
Token of
Feature activation+0.022
other
Token other
Feature activation-0.022
players
Token players
Feature activation+0.201
vying
Token vying
Feature activation+0.309
for
Token for
Feature activation+2.619
the
Token the
Feature activation+0.530
attention
Token attention
Feature activation+0.000
of
Token of
Feature activation+0.000
struggling
Token struggling
Feature activation+0.000
students
Token students
Feature activation+0.000
ushi
Tokenushi
Feature activation-0.011
trains
Token trains
Feature activation+0.016
âĢ
TokenâĢ
Feature activation-0.029
Ŀ
TokenĿ
Feature activation+0.015
competing
Token competing
Feature activation+0.163
for
Token for
Feature activation+3.404
people
Token people
Feature activation+0.175
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
attention
Token attention
Feature activation+0.000
on
Token on
Feature activation+0.003
their
Token their
Feature activation+0.013
promises
Token promises
Feature activation+0.005
of
Token of
Feature activation+0.083
competing
Token competing
Feature activation+0.432
for
Token for
Feature activation+2.695
a
Token a
Feature activation+0.401
Stanley
Token Stanley
Feature activation+0.000
Cup
Token Cup
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.141
now
Token now
Feature activation+0.100
be
Token be
Feature activation-0.146
vying
Token vying
Feature activation+0.500
for
Token for
Feature activation+2.079
a
Token a
Feature activation+1.038
place
Token place
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
pack
Token pack
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.176
now
Token now
Feature activation+0.071
be
Token be
Feature activation-0.151
vying
Token vying
Feature activation+0.410
for
Token for
Feature activation+1.936
a
Token a
Feature activation+1.066
place
Token place
Feature activation+0.210
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
pack
Token pack
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.022
Ċ
TokenĊ
Feature activation+0.025
Lewis
TokenLewis
Feature activation+0.013
will
Token will
Feature activation+0.066
compete
Token compete
Feature activation+0.259
for
Token for
Feature activation+2.798
the
Token the
Feature activation+0.605
opportunity
Token opportunity
Feature activation+0.167
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
in
Token in
Feature activation+0.000
Morris
Token Morris
Feature activation+0.007
(
Token (
Feature activation-0.046
5
Token5
Feature activation-0.003
)
Token)
Feature activation+0.139
battles
Token battles
Feature activation+0.028
for
Token for
Feature activation+2.789
the
Token the
Feature activation+0.647
ball
Token ball
Feature activation-0.025
against
Token against
Feature activation+0.000
Eric
Token Eric
Feature activation+0.000
B
Token B
Feature activation+0.000
Comp
Token Comp
Feature activation+0.011
eting
Tokeneting
Feature activation+0.046
factions
Token factions
Feature activation+0.306
v
Token v
Feature activation+0.037
ied
Tokenied
Feature activation+0.241
for
Token for
Feature activation+2.528
the
Token the
Feature activation+0.422
throne
Token throne
Feature activation+0.000
and
Token and
Feature activation+0.000
counties
Token counties
Feature activation+0.000
(
Token (
Feature activation+0.000
off
Token off
Feature activation-0.062
and
Token and
Feature activation-0.009
on
Token on
Feature activation+0.009
,
Token,
Feature activation+0.048
vying
Token vying
Feature activation+0.737
for
Token for
Feature activation+2.652
that
Token that
Feature activation+0.256
title
Token title
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
technology
Token technology
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.01

Head 2: 0.12

Head 3: 0.09

Head 4: 0.25

Head 5: 0.02

Head 6: 0.06

Head 7: 0.17

Head 8: 0.03

Head 9: 0.04

Head 10: 0.08

Head 11: 0.12

Positive logits

immortality1.66

throne1.62

supremacy1.55

slots1.53

scraps1.46

ladder1.45

domination1.42

survival1.42

roles1.38

livelihood1.37

dominance1.35

hegemony1.34

lucrative1.33

coveted1.32

someday1.32

20191.30

titles1.30

habitat1.29

thood1.28

bearings1.28

Negative logits

actionDate-2.03

inventoryQuantity-1.73

SOURCE-1.62

clinton-1.58

SourceFile-1.51

ーク-1.41

INTON-1.37

redacted-1.34

claimer-1.33

cember-1.31

cised-1.30

Interstitial-1.30

ved-1.27

hari-1.27

lene-1.27

ighed-1.26

rican-1.26

enged-1.26

FORE-1.25

forth-1.24

INTERVAL 2.761 - 3.068
CONTAINS 0.000%

between
Token between
Feature activation+0.000
males
Token males
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.549
a
Token a
Feature activation+1.874
female
Token female
Feature activation+3.068
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
E
TokenE
Feature activation+0.000
tymology
Tokentymology
Feature activation+0.000
top
Token top
Feature activation+0.000
clubs
Token clubs
Feature activation+0.000
are
Token are
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.894
entry
Token entry
Feature activation+2.902
into
Token into
Feature activation+1.517
the
Token the
Feature activation+0.864
UEFA
Token UEFA
Feature activation+0.125
Champions
Token Champions
Feature activation+0.000
League
Token League
Feature activation+0.500

INTERVAL 2.455 - 2.761
CONTAINS 0.000%

is
Token is
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.873
the
Token the
Feature activation+2.053
top
Token top
Feature activation+2.565
players
Token players
Feature activation+2.483
with
Token with
Feature activation+1.014
clubs
Token clubs
Feature activation+0.487
such
Token such
Feature activation+0.000
as
Token as
Feature activation+0.000
also
Token also
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.104
starting
Token starting
Feature activation+1.494
ber
Token ber
Feature activation+0.525
ths
Tokenths
Feature activation+2.709
in
Token in
Feature activation+1.242
the
Token the
Feature activation+0.741
team
Token team
Feature activation+0.310
,
Token,
Feature activation+0.326
and
Token and
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.873
the
Token the
Feature activation+2.053
top
Token top
Feature activation+2.565
players
Token players
Feature activation+2.483
with
Token with
Feature activation+1.014
clubs
Token clubs
Feature activation+0.487
such
Token such
Feature activation+0.000
as
Token as
Feature activation+0.000
Chelsea
Token Chelsea
Feature activation+0.000
different
Token different
Feature activation+0.000
cities
Token cities
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.425
migrant
Token migrant
Feature activation+1.336
labour
Token labour
Feature activation+2.533
through
Token through
Feature activation+0.947
advertising
Token advertising
Feature activation+0.000
,
Token,
Feature activation+0.000
better
Token better
Feature activation+0.000
international
Token international
Feature activation+0.000
cables
Token cables
Feature activation+0.000
,
Token,
Feature activation+0.000
all
Token all
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.556
the
Token the
Feature activation+2.567
wall
Token wall
Feature activation+1.606
socket
Token socket
Feature activation+1.183
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
sometimes
Token sometimes
Feature activation+0.000

INTERVAL 2.148 - 2.455
CONTAINS 0.000%

once
Token once
Feature activation+0.000
competed
Token competed
Feature activation+0.000
for
Token for
Feature activation+0.988
power
Token power
Feature activation+2.004
and
Token and
Feature activation+0.758
influence
Token influence
Feature activation+2.308
in
Token in
Feature activation+1.041
the
Token the
Feature activation+0.612
Arab
Token Arab
Feature activation+0.000
world
Token world
Feature activation+0.008
,
Token,
Feature activation+0.000
now
Token now
Feature activation+0.000
be
Token be
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.791
a
Token a
Feature activation+2.258
place
Token place
Feature activation+2.195
in
Token in
Feature activation+1.066
the
Token the
Feature activation+1.043
pack
Token pack
Feature activation+1.098
,
Token,
Feature activation+0.542
adding
Token adding
Feature activation+0.334
trains
Token trains
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.843
people
Token people
Feature activation+2.296
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.189
attention
Token attention
Feature activation+1.154
,
Token,
Feature activation+0.000
Lewis
TokenLewis
Feature activation+0.000
will
Token will
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+1.315
opportunity
Token opportunity
Feature activation+2.152
to
Token to
Feature activation+1.328
be
Token be
Feature activation+1.147
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.141
man
Token man
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
now
Token now
Feature activation+0.000
be
Token be
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.791
a
Token a
Feature activation+2.258
place
Token place
Feature activation+2.195
in
Token in
Feature activation+1.066
the
Token the
Feature activation+1.043
pack
Token pack
Feature activation+1.098
,
Token,
Feature activation+0.542

INTERVAL 1.841 - 2.148
CONTAINS 0.000%

begun
Token begun
Feature activation+0.000
to
Token to
Feature activation+0.000
duel
Token duel
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+2.036
right
Token right
Feature activation+2.025
to
Token to
Feature activation+0.750
claim
Token claim
Feature activation+0.000
that
Token that
Feature activation+0.000
they
Token they
Feature activation+0.000
were
Token were
Feature activation+0.000
idates
Tokenidates
Feature activation+0.000
are
Token are
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.990
70
Token 70
Feature activation+1.696
seats
Token seats
Feature activation+1.874
on
Token on
Feature activation+0.714
the
Token the
Feature activation+0.518
Legislative
Token Legislative
Feature activation+0.000
Council
Token Council
Feature activation+0.379
,
Token,
Feature activation+0.945
opportunity
Token opportunity
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.980
starting
Token starting
Feature activation+1.282
jobs
Token jobs
Feature activation+1.985
,
Token,
Feature activation+0.028
along
Token along
Feature activation+0.000
with
Token with
Feature activation+0.000
incumb
Token incumb
Feature activation+0.000
ents
Tokenents
Feature activation+0.000
which
Token which
Feature activation+0.000
are
Token are
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+1.064
a
Token a
Feature activation+1.783
solution
Token solution
Feature activation+1.972
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
One
TokenOne
Feature activation+0.000
idea
Token idea
Feature activation+0.000
The
Token The
Feature activation+0.000
candidates
Token candidates
Feature activation+0.000
are
Token are
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.682
the
Token the
Feature activation+1.859
Senate
Token Senate
Feature activation+0.957
seat
Token seat
Feature activation+0.964
that
Token that
Feature activation+0.876
Jeff
Token Jeff
Feature activation+0.000
Sessions
Token Sessions
Feature activation+0.000

INTERVAL 1.534 - 1.841
CONTAINS 0.000%

armed
Token armed
Feature activation+0.000
groups
Token groups
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.651
supremacy
Token supremacy
Feature activation+1.729
in
Token in
Feature activation+1.607
Has
Token Has
Feature activation+0.000
ake
Tokenake
Feature activation+0.000
h
Tokenh
Feature activation+0.000
.
Token.
Feature activation+0.000
L
Token L
Feature activation+0.000
realistically
Token realistically
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.258
a
Token a
Feature activation+1.967
playoff
Token playoff
Feature activation+1.223
spot
Token spot
Feature activation+1.592
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
another
Token another
Feature activation+0.000
one
Token one
Feature activation+0.000
âĢĶ
TokenâĢĶ
Feature activation+0.000
the
Token the
Feature activation+1.423
K
Token K
Feature activation+0.385
A
TokenA
Feature activation+0.000
A
Token A
Feature activation+0.000
uty
Tokenuty
Feature activation+0.000
Cup
Token Cup
Feature activation+1.688
,
Token,
Feature activation+0.533
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
tournament
Token tournament
Feature activation+0.000
has
Token has
Feature activation+0.000
as
Token as
Feature activation+0.000
they
Token they
Feature activation+0.000
v
Token v
Feature activation+0.000
ied
Tokenied
Feature activation+0.000
for
Token for
Feature activation+0.018
the
Token the
Feature activation+1.836
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
The
TokenThe
Feature activation+0.790
Ultimate
Token Ultimate
Feature activation+0.000
Fighter
Token Fighter
Feature activation+0.000
continue
Token continue
Feature activation+0.000
to
Token to
Feature activation+0.000
v
Token v
Feature activation+0.000
ie
Tokenie
Feature activation+0.000
for
Token for
Feature activation+0.506
the
Token the
Feature activation+1.781
league
Token league
Feature activation+1.835
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.642
coveted
Token coveted
Feature activation+0.572

INTERVAL 1.227 - 1.534
CONTAINS 0.001%

The
Token The
Feature activation+0.000
sixteen
Token sixteen
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.338
the
Token the
Feature activation+1.525
party
Token party
Feature activation+1.870
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.376
nomination
Token nomination
Feature activation+0.866
Tea
Token Tea
Feature activation+0.000
Party
Token Party
Feature activation+0.000
started
Token started
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.289
power
Token power
Feature activation+1.400
,
Token,
Feature activation+0.000
more
Token more
Feature activation+0.000
moderate
Token moderate
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
shifted
Token shifted
Feature activation+0.000
strong
Token strong
Feature activation+0.000
[
Token [
Feature activation+0.000
to
Tokento
Feature activation+0.000
challenge
Token challenge
Feature activation+0.000
for
Token for
Feature activation+0.044
the
Token the
Feature activation+1.384
win
Token win
Feature activation+0.458
]
Token]
Feature activation+0.000
so
Token so
Feature activation+0.000
we
Token we
Feature activation+0.000
will
Token will
Feature activation+0.000
.
Token.
Feature activation+0.000
Should
Token Should
Feature activation+0.000
Oregon
Token Oregon
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.396
a
Token a
Feature activation+1.452
Pac
Token Pac
Feature activation+0.220
-
Token-
Feature activation+0.000
12
Token12
Feature activation+1.030
title
Token title
Feature activation+0.956
and
Token and
Feature activation+0.454
social
Token social
Feature activation+0.000
networking
Token networking
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.815
attention
Token attention
Feature activation+1.482
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
study
Token study
Feature activation+0.000

INTERVAL 0.920 - 1.227
CONTAINS 0.001%

seats
Token seats
Feature activation+1.874
on
Token on
Feature activation+0.714
the
Token the
Feature activation+0.518
Legislative
Token Legislative
Feature activation+0.000
Council
Token Council
Feature activation+0.379
,
Token,
Feature activation+0.945
known
Token known
Feature activation+0.379
as
Token as
Feature activation+0.000
Leg
Token Leg
Feature activation+0.000
Co
TokenCo
Feature activation+0.000
,
Token,
Feature activation+0.000
will
Token will
Feature activation+0.000
battle
Token battle
Feature activation+0.000
each
Token each
Feature activation+0.000
other
Token other
Feature activation+0.000
for
Token for
Feature activation+0.014
the
Token the
Feature activation+1.058
soul
Token soul
Feature activation+0.793
of
Token of
Feature activation+0.102
Hell
Token Hell
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
will
Token will
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.327
bragging
Token bragging
Feature activation+0.468
rights
Token rights
Feature activation+2.100
at
Token at
Feature activation+1.086
the
Token the
Feature activation+0.200
Del
Token Del
Feature activation+0.000
Mar
Token Mar
Feature activation+0.000
Th
Token Th
Feature activation+0.000
orough
Tokenorough
Feature activation+0.000
teams
Token teams
Feature activation+0.000
in
Token in
Feature activation+0.000
contention
Token contention
Feature activation+0.000
for
Token for
Feature activation+1.424
the
Token the
Feature activation+2.054
College
Token College
Feature activation+1.028
Football
Token Football
Feature activation+0.000
Playoff
Token Playoff
Feature activation+1.403
field
Token field
Feature activation+1.302
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
ec
Tokenec
Feature activation+0.000
will
Token will
Feature activation+0.000
also
Token also
Feature activation+0.000
battle
Token battle
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+1.082
Conservative
Token Conservative
Feature activation+0.335
line
Token line
Feature activation+1.129
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 0.614 - 0.920
CONTAINS 0.001%

free
Token free
Feature activation+0.000
agents
Token agents
Feature activation+0.000
vying
Token vying
Feature activation+0.000
for
Token for
Feature activation+0.790
starting
Token starting
Feature activation+1.565
jobs
Token jobs
Feature activation+0.805
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
forward
Token forward
Feature activation+0.000
to
Token to
Feature activation+0.000
competing
Token competing
Feature activation+0.000
for
Token for
Feature activation+0.135
a
Token a
Feature activation+1.086
starting
Token starting
Feature activation+0.617
job
Token job
Feature activation+0.488
elsewhere
Token elsewhere
Feature activation+0.000
by
Token by
Feature activation+0.000
this
Token this
Feature activation+0.000
time
Token time
Feature activation+0.000
operators
Token operators
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
on
Token on
Feature activation+0.000
a
Token a
Feature activation+0.304
completely
Token completely
Feature activation+0.760
level
Token level
Feature activation+0.000
playing
Token playing
Feature activation+0.000
field
Token field
Feature activation+0.000
,
Token,
Feature activation+0.000
then
Token then
Feature activation+0.000
chances
Token chances
Feature activation+0.000
better
Token better
Feature activation+0.000
to
Token to
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.122
Team
Token Team
Feature activation+0.825
USA
Token USA
Feature activation+0.943
and
Token and
Feature activation+0.232
head
Token head
Feature activation+0.000
to
Token to
Feature activation+0.000
Beijing
Token Beijing
Feature activation+0.000
see
Token see
Feature activation+0.000
the
Token the
Feature activation+0.000
teams
Token teams
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.084
a
Token a
Feature activation+0.833
share
Token share
Feature activation+0.559
of
Token of
Feature activation+1.044
that
Token that
Feature activation+0.162
pot
Token pot
Feature activation+0.315
,
Token,
Feature activation+0.000

INTERVAL 0.307 - 0.614
CONTAINS 0.002%

two
Token two
Feature activation+0.000
countries
Token countries
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.157
the
Token the
Feature activation+1.423
K
Token K
Feature activation+0.385
A
TokenA
Feature activation+0.000
A
Token A
Feature activation+0.000
uty
Tokenuty
Feature activation+0.000
Cup
Token Cup
Feature activation+1.688
,
Token,
Feature activation+0.533
down
Token down
Feature activation+0.000
and
Token and
Feature activation+0.000
fight
Token fight
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.052
title
Token title
Feature activation+0.317
again
Token again
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
am
Token am
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
fight
Token fight
Feature activation+0.000
for
Token for
Feature activation+0.000
Japan
Token Japan
Feature activation+0.393
,
Token,
Feature activation+0.000
for
Token for
Feature activation+0.000
my
Token my
Feature activation+0.000
family
Token family
Feature activation+0.000
,
Token,
Feature activation+0.000
chance
Token chance
Feature activation+0.000
to
Token to
Feature activation+0.000
not
Token not
Feature activation+0.000
only
Token only
Feature activation+0.000
compete
Token compete
Feature activation+0.000
for
Token for
Feature activation+0.329
a
Token a
Feature activation+1.583
starting
Token starting
Feature activation+0.251
job
Token job
Feature activation+0.369
but
Token but
Feature activation+0.000
play
Token play
Feature activation+0.000
feather
Token feather
Feature activation+0.017
weight
Tokenweight
Feature activation+0.000
title
Token title
Feature activation+0.582
next
Token next
Feature activation+0.114
year
Token year
Feature activation+0.050
and
Token and
Feature activation+0.452
Qu
Token Qu
Feature activation+0.000
igg
Tokenigg
Feature activation+0.000
believes
Token believes
Feature activation+0.000
he
Token he
Feature activation+0.000
can
Token can
Feature activation+0.000

INTERVAL 0.000 - 0.307
CONTAINS 99.994%

and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
haunting
Token haunting
Feature activation+0.000
sound
Token sound
Feature activation+0.000
of
Token of
Feature activation+0.000
her
Token her
Feature activation+0.000
voice
Token voice
Feature activation+0.000
softly
Token softly
Feature activation+0.000
rec
Token rec
Feature activation+0.000
iting
Tokeniting
Feature activation+0.000
nursery
Token nursery
Feature activation+0.000
being
Token being
Feature activation+0.000
made
Token made
Feature activation+0.000
for
Token for
Feature activation+0.000
Freed
Token Freed
Feature activation+0.000
the
Token the
Feature activation+0.000
Brave
Token Brave
Feature activation+0.000
Wand
Token Wand
Feature activation+0.000
erer
Tokenerer
Feature activation+0.000
and
Token and
Feature activation+0.000
tried
Token tried
Feature activation+0.000
to
Token to
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
and
Token and
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
t
Tokent
Feature activation+0.000
oug
Tokenoug
Feature activation+0.000
hen
Tokenhen
Feature activation+0.000
existing
Token existing
Feature activation+0.000
hate
Token hate
Feature activation+0.000
crime
Token crime
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
group
Token group
Feature activation+0.000
of
Token of
Feature activation+0.000
individuals
Token individuals
Feature activation+0.000
,
Token,
Feature activation+0.000
highly
Token highly
Feature activation+0.000
trained
Token trained
Feature activation+0.000
and
Token and
Feature activation+0.000
motivated
Token motivated
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
B
TokenB
Feature activation+0.000
in
Tokenin
Feature activation+0.000
Laden
Token Laden
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
precise
Token precise
Feature activation+0.000
in
Token in
Feature activation+0.000
telling
Token telling
Feature activation+0.000
America
Token America
Feature activation+0.000
the
Token the
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 14: Induction when giving image credits

TOP ACTIVATIONS
MAX = 6.042

the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+6.042
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
in
Token in
Feature activation+0.000
Peru
Token Peru
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Photo
TokenPhoto
Feature activation+5.769
by
Token by
Feature activation+0.000
Sarah
Token Sarah
Feature activation+0.000
Bud
Token Bud
Feature activation+0.000
er
Tokener
Feature activation+0.000
A
Token A
Feature activation+0.000
it
Token it
Feature activation+0.000
symbol
Token symbol
Feature activation+0.000
ises
Tokenises
Feature activation+0.000
death
Token death
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.646
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
helping
Token helping
Feature activation+0.000
yourself
Token yourself
Feature activation+0.000
to
Token to
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.641
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.078
Courtesy
Token Courtesy
Feature activation+5.637
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
as
Token as
Feature activation+0.000
fruit
Token fruit
Feature activation+0.000
or
Token or
Feature activation+0.000
cake
Token cake
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.594
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
,
Token,
Feature activation+0.000
Honduras
Token Honduras
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Photo
TokenPhoto
Feature activation+5.497
by
Token by
Feature activation+0.000
Sarah
Token Sarah
Feature activation+0.000
Bud
Token Bud
Feature activation+0.000
er
Tokener
Feature activation+0.000
A
Token A
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.467
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+5.458
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.451
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.434
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.394
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
number
Token number
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
dashboard
Token dashboard
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.385
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.353
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.286
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+5.256
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.164
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Dr
Token Dr
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+5.119
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
to
Token to
Feature activation+0.000
clean
Token clean
Feature activation+0.000
se
Tokense
Feature activation+0.000
yourself
Token yourself
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.108
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+4.989
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000

Top DFA by src position
MAX = 7.812

Civic
Token Civic
Feature activation-0.002
Center
Token Center
Feature activation-0.007
Plaza
Token Plaza
Feature activation+0.002
.
Token.
Feature activation+0.118
Photo
Token Photo
Feature activation+0.082
by
Token by
Feature activation+3.180
Chris
Token Chris
Feature activation+0.158
Stone
Token Stone
Feature activation+0.025
Thousands
Token Thousands
Feature activation+0.094
took
Token took
Feature activation+0.025
part
Token part
Feature activation+0.022
Honduras
Token Honduras
Feature activation-0.001
.
Token.
Feature activation+0.038
Ċ
TokenĊ
Feature activation+0.044
Ċ
TokenĊ
Feature activation+0.094
Photo
TokenPhoto
Feature activation+0.107
by
Token by
Feature activation+5.181
Sarah
Token Sarah
Feature activation+0.345
Bud
Token Bud
Feature activation+0.014
er
Tokener
Feature activation+0.002
A
Token A
Feature activation+0.078
normal
Token normal
Feature activation-0.029
yourself
Token yourself
Feature activation-0.000
to
Token to
Feature activation+0.001
it
Token it
Feature activation-0.001
.
Token.
Feature activation+0.024
Image
Token Image
Feature activation+0.062
via
Token via
Feature activation+3.683
SAY
Token SAY
Feature activation+0.107
S
TokenS
Feature activation+0.021
Ċ
TokenĊ
Feature activation+0.033
Ċ
TokenĊ
Feature activation+0.241
5
Token5
Feature activation+0.043
someone
Token someone
Feature activation-0.002
's
Token's
Feature activation+0.000
home
Token home
Feature activation-0.001
.
Token.
Feature activation+0.017
Image
Token Image
Feature activation+0.017
via
Token via
Feature activation+4.023
SAY
Token SAY
Feature activation+0.079
S
TokenS
Feature activation+0.013
Ċ
TokenĊ
Feature activation+0.055
Ċ
TokenĊ
Feature activation+0.195
3
Token3
Feature activation+0.036
Save
TokenSave
Feature activation+0.078
this
Token this
Feature activation+0.091
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.092
Courtesy
Token Courtesy
Feature activation+0.117
of
Token of
Feature activation+2.616
BIG
Token BIG
Feature activation+0.132
-
Token -
Feature activation+0.075
B
Token B
Feature activation+0.011
jar
Tokenjar
Feature activation-0.005
ke
Tokenke
Feature activation+0.000
someone
Token someone
Feature activation-0.003
's
Token's
Feature activation+0.001
home
Token home
Feature activation-0.001
.
Token.
Feature activation+0.032
Image
Token Image
Feature activation+0.039
via
Token via
Feature activation+7.812
SAY
Token SAY
Feature activation+0.144
S
TokenS
Feature activation+0.006
Ċ
TokenĊ
Feature activation+0.102
Ċ
TokenĊ
Feature activation+0.664
3
Token3
Feature activation+0.125
below
Token below
Feature activation-0.005
ad
Token ad
Feature activation+0.036
Ċ
TokenĊ
Feature activation+0.008
Ċ
TokenĊ
Feature activation-0.008
Photo
TokenPhoto
Feature activation+0.086
by
Token by
Feature activation+6.936
Sarah
Token Sarah
Feature activation+0.350
Bud
Token Bud
Feature activation+0.010
er
Tokener
Feature activation-0.001
Never
Token Never
Feature activation-0.003
-
Token-
Feature activation+0.009
<|endoftext|>
Token<|endoftext|>
Feature activation+2.733
took
Token took
Feature activation-0.011
part
Token part
Feature activation-0.002
in
Token in
Feature activation+0.001
the
Token the
Feature activation+0.003
San
Token San
Feature activation+0.002
Save
TokenSave
Feature activation+0.065
this
Token this
Feature activation+0.062
picture
Token picture
Feature activation-0.006
!
Token!
Feature activation+0.100
Courtesy
Token Courtesy
Feature activation+0.120
of
Token of
Feature activation+4.019
BIG
Token BIG
Feature activation+0.183
-
Token -
Feature activation+0.077
B
Token B
Feature activation+0.014
jar
Tokenjar
Feature activation-0.005
ke
Tokenke
Feature activation+0.003
<|endoftext|>
Token<|endoftext|>
Feature activation+2.756
took
Token took
Feature activation-0.013
part
Token part
Feature activation-0.002
in
Token in
Feature activation+0.001
the
Token the
Feature activation+0.004
San
Token San
Feature activation+0.003
streets
Token streets
Feature activation+0.003
of
Token of
Feature activation+0.018
downtown
Token downtown
Feature activation-0.000
.
Token.
Feature activation+0.201
Photo
Token Photo
Feature activation+0.067
by
Token by
Feature activation+4.779
Chris
Token Chris
Feature activation+0.108
Stone
Token Stone
Feature activation+0.021
Thousands
Token Thousands
Feature activation+0.069
took
Token took
Feature activation+0.014
part
Token part
Feature activation+0.009
March
Token March
Feature activation-0.005
For
Token For
Feature activation-0.003
Science
Token Science
Feature activation-0.006
.
Token.
Feature activation+0.021
Photo
Token Photo
Feature activation+0.039
by
Token by
Feature activation+4.038
Chris
Token Chris
Feature activation+0.107
Stone
Token Stone
Feature activation+0.019
Thousands
Token Thousands
Feature activation+0.034
took
Token took
Feature activation+0.004
part
Token part
Feature activation+0.018
,
Token,
Feature activation+0.001
give
Token give
Feature activation+0.004
way
Token way
Feature activation+0.006
.
Token.
Feature activation+0.004
Image
Token Image
Feature activation+0.053
via
Token via
Feature activation+5.090
SAY
Token SAY
Feature activation+0.148
S
TokenS
Feature activation+0.028
Ċ
TokenĊ
Feature activation+0.088
Ċ
TokenĊ
Feature activation+0.311
8
Token8
Feature activation+0.068
streets
Token streets
Feature activation+0.003
of
Token of
Feature activation+0.014
downtown
Token downtown
Feature activation+0.007
.
Token.
Feature activation+0.101
Photo
Token Photo
Feature activation+0.106
by
Token by
Feature activation+2.962
Chris
Token Chris
Feature activation+0.102
Stone
Token Stone
Feature activation+0.021
Thousands
Token Thousands
Feature activation+0.061
took
Token took
Feature activation+0.018
part
Token part
Feature activation+0.009
<|endoftext|>
Token<|endoftext|>
Feature activation+2.871
took
Token took
Feature activation-0.017
part
Token part
Feature activation-0.003
in
Token in
Feature activation+0.002
the
Token the
Feature activation+0.005
San
Token San
Feature activation+0.004
Save
TokenSave
Feature activation+0.040
this
Token this
Feature activation+0.044
picture
Token picture
Feature activation-0.007
!
Token!
Feature activation+0.097
Courtesy
Token Courtesy
Feature activation+0.118
of
Token of
Feature activation+4.384
BIG
Token BIG
Feature activation+0.182
-
Token -
Feature activation+0.066
B
Token B
Feature activation+0.010
jar
Tokenjar
Feature activation-0.005
ke
Tokenke
Feature activation+0.003
streets
Token streets
Feature activation+0.003
of
Token of
Feature activation+0.015
downtown
Token downtown
Feature activation+0.005
.
Token.
Feature activation+0.115
Photo
Token Photo
Feature activation+0.080
by
Token by
Feature activation+3.821
Chris
Token Chris
Feature activation+0.097
Stone
Token Stone
Feature activation+0.020
Thousands
Token Thousands
Feature activation+0.053
took
Token took
Feature activation+0.026
part
Token part
Feature activation+0.011
Save
TokenSave
Feature activation+0.018
this
Token this
Feature activation-0.033
picture
Token picture
Feature activation+0.030
!
Token!
Feature activation+0.149
Courtesy
Token Courtesy
Feature activation+0.114
of
Token of
Feature activation+4.213
BIG
Token BIG
Feature activation+0.264
-
Token -
Feature activation+0.057
B
Token B
Feature activation+0.010
jar
Tokenjar
Feature activation-0.006
ke
Tokenke
Feature activation+0.004
in
Token in
Feature activation+0.036
a
Token a
Feature activation+0.003
fist
Token fist
Feature activation+0.008
.
Token.
Feature activation+0.022
Image
Token Image
Feature activation+0.027
via
Token via
Feature activation+6.649
SAY
Token SAY
Feature activation+0.142
S
TokenS
Feature activation+0.003
Ċ
TokenĊ
Feature activation+0.040
Ċ
TokenĊ
Feature activation+0.163
When
TokenWhen
Feature activation+0.137
streets
Token streets
Feature activation+0.003
of
Token of
Feature activation+0.016
downtown
Token downtown
Feature activation+0.005
.
Token.
Feature activation+0.088
Photo
Token Photo
Feature activation+0.066
by
Token by
Feature activation+4.341
Chris
Token Chris
Feature activation+0.101
Stone
Token Stone
Feature activation+0.017
Thousands
Token Thousands
Feature activation+0.042
took
Token took
Feature activation+0.015
part
Token part
Feature activation+0.008

Decoder Weights Distribution

Head 0: 0.18

Head 1: 0.19

Head 2: 0.04

Head 3: 0.05

Head 4: 0.03

Head 5: 0.15

Head 6: 0.06

Head 7: 0.03

Head 8: 0.06

Head 9: 0.08

Head 10: 0.06

Head 11: 0.06

Positive logits

credits1.50

Nou1.48

Gors1.48

Byr1.41

breakdown1.41

photos1.37

Editors1.36

capitals1.34

unden1.34

Volks1.32

BY1.32

aftermath1.32

TED1.32

UPDATE1.30

Credits1.30

jargon1.28

Gloss1.26

Gothic1.26

BuzzFeed1.25

Goo1.25

Negative logits

emer-1.66

irtual-1.61

iage-1.59

ente-1.54

tle-1.52

iris-1.51

ses-1.51

uties-1.50

anship-1.50

rop-1.48

atson-1.47

threat-1.47

ebus-1.47

orem-1.43

cot-1.41

ilit-1.41

intent-1.41

ittee-1.40

restrial-1.40

shall-1.39

INTERVAL 5.438 - 6.042
CONTAINS 0.000%

Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+5.458
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+6.042
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
helping
Token helping
Feature activation+0.000
yourself
Token yourself
Feature activation+0.000
to
Token to
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.641
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
in
Token in
Feature activation+0.000
Peru
Token Peru
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Photo
TokenPhoto
Feature activation+5.769
by
Token by
Feature activation+0.000
Sarah
Token Sarah
Feature activation+0.000
Bud
Token Bud
Feature activation+0.000
er
Tokener
Feature activation+0.000
A
Token A
Feature activation+0.000
,
Token,
Feature activation+0.000
Honduras
Token Honduras
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Photo
TokenPhoto
Feature activation+5.497
by
Token by
Feature activation+0.000
Sarah
Token Sarah
Feature activation+0.000
Bud
Token Bud
Feature activation+0.000
er
Tokener
Feature activation+0.000
A
Token A
Feature activation+0.000

INTERVAL 4.834 - 5.438
CONTAINS 0.000%

to
Token to
Feature activation+0.000
accommodate
Token accommodate
Feature activation+0.000
waiting
Token waiting
Feature activation+0.000
patrons
Token patrons
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+4.860
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+5.119
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+5.256
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
to
Token to
Feature activation+0.000
clean
Token clean
Feature activation+0.000
se
Tokense
Feature activation+0.000
yourself
Token yourself
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+5.108
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
the
Token the
Feature activation+0.000
streets
Token streets
Feature activation+0.000
of
Token of
Feature activation+0.000
downtown
Token downtown
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+5.286
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000

INTERVAL 4.229 - 4.834
CONTAINS 0.000%

)
Token)
Feature activation+0.000
Anna
Token Anna
Feature activation+0.000
K
Token K
Feature activation+0.000
roup
Tokenroup
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+4.235
courtesy
Token courtesy
Feature activation+3.015
of
Token of
Feature activation+0.000
Water
Token Water
Feature activation+0.000
ford
Tokenford
Feature activation+0.000
Police
Token Police
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+4.349
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
Diego
Token Diego
Feature activation+0.000
Civic
Token Civic
Feature activation+0.000
Center
Token Center
Feature activation+0.000
Plaza
Token Plaza
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+4.272
by
Token by
Feature activation+0.000
Chris
Token Chris
Feature activation+0.000
Stone
Token Stone
Feature activation+0.000
Thousands
Token Thousands
Feature activation+0.000
took
Token took
Feature activation+0.000
by
Token by
Feature activation+0.000
Luc
Token Luc
Feature activation+0.000
ie
Tokenie
Feature activation+0.000
Rice
Token Rice
Feature activation+0.000
Illust
Token Illust
Feature activation+1.851
ration
Tokenration
Feature activation+4.446
and
Token and
Feature activation+0.000
Design
Token Design
Feature activation+0.000
(
Token (
Feature activation+0.000
l
Tokenl
Feature activation+0.000
uc
Tokenuc
Feature activation+0.000
that
Token that
Feature activation+0.000
you
Token you
Feature activation+0.000
receive
Token receive
Feature activation+0.000
it
Token it
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+4.449
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

INTERVAL 3.625 - 4.229
CONTAINS 0.000%

/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
)
Token)
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+3.938
by
Token by
Feature activation+0.000
Scott
Token Scott
Feature activation+0.000
Olson
Token Olson
Feature activation+0.000
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
someone
Token someone
Feature activation+0.000
down
Token down
Feature activation+0.000
in
Token in
Feature activation+0.000
public
Token public
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+3.803
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
customs
Token customs
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
culture
Token culture
Feature activation+0.000
.
Token.
Feature activation+0.000
Image
Token Image
Feature activation+3.977
via
Token via
Feature activation+0.000
SAY
Token SAY
Feature activation+0.000
S
TokenS
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Save
TokenSave
Feature activation+0.000
this
Token this
Feature activation+0.000
picture
Token picture
Feature activation+0.000
!
Token!
Feature activation+0.000
Courtesy
Token Courtesy
Feature activation+4.058
of
Token of
Feature activation+0.000
BIG
Token BIG
Feature activation+0.000
-
Token -
Feature activation+0.000
B
Token B
Feature activation+0.000
jar
Tokenjar
Feature activation+0.000
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
)
Token)
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+4.209
credit
Token credit
Feature activation+0.000
SA
Token SA
Feature activation+0.000
UL
TokenUL
Feature activation+0.000
LO
Token LO
Feature activation+0.000
EB
TokenEB
Feature activation+0.000

INTERVAL 3.021 - 3.625
CONTAINS 0.000%

21
Token 21
Feature activation+0.000
,
Token,
Feature activation+0.000
1996
Token 1996
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+3.191
by
Token by
Feature activation+0.000
Phil
Token Phil
Feature activation+0.000
Cole
Token Cole
Feature activation+0.000
/
Token/
Feature activation+0.000
All
TokenAll
Feature activation+0.000
cn
Tokencn
Feature activation+0.000
/
Token/
Feature activation+0.000
Flickr
TokenFlickr
Feature activation+0.000
)
Token)
Feature activation+0.000
(
Token (
Feature activation+0.000
photos
Tokenphotos
Feature activation+3.353
by
Token by
Feature activation+0.000
G
Token G
Feature activation+0.000
age
Tokenage
Feature activation+0.000
Sk
Token Sk
Feature activation+0.000
id
Tokenid
Feature activation+0.000
ive
Tokenive
Feature activation+0.000
Action
Token Action
Feature activation+0.000
Sports
Token Sports
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+3.268
by
Token by
Feature activation+0.000
Doug
Token Doug
Feature activation+0.000
Pens
Token Pens
Feature activation+0.000
inger
Tokeninger
Feature activation+0.000
/
Token/
Feature activation+0.000
ilateral
Tokenilateral
Feature activation+0.000
trade
Token trade
Feature activation+0.000
agreement
Token agreement
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+3.410
by
Token by
Feature activation+0.000
Ron
Token Ron
Feature activation+0.000
Sachs
Token Sachs
Feature activation+0.000
/
Token/
Feature activation+0.000
Pool
TokenPool
Feature activation+0.000
in
Token in
Feature activation+0.000
multiple
Token multiple
Feature activation+0.000
sports
Token sports
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+3.077
by
Token by
Feature activation+0.000
Rick
Token Rick
Feature activation+0.000
Wil
Token Wil
Feature activation+0.000
king
Tokenking
Feature activation+0.000
/
Token/
Feature activation+0.000

INTERVAL 2.417 - 3.021
CONTAINS 0.000%

in
Token in
Feature activation+0.000
defense
Token defense
Feature activation+0.000
spending
Token spending
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+2.675
by
Token by
Feature activation+0.000
Mark
Token Mark
Feature activation+0.000
Wilson
Token Wilson
Feature activation+0.000
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
21
Token 21
Feature activation+0.000
,
Token,
Feature activation+0.000
2017
Token 2017
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+2.550
by
Token by
Feature activation+0.000
Oliver
Token Oliver
Feature activation+0.000
Cont
Token Cont
Feature activation+0.000
re
Tokenre
Feature activation+0.000
ras
Tokenras
Feature activation+0.000
the
Token the
Feature activation+0.000
Pakistan
Token Pakistan
Feature activation+0.000
border
Token border
Feature activation+0.000
.
Token.
Feature activation+0.000
Photo
Token Photo
Feature activation+2.389
provided
Token provided
Feature activation+2.510
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
Bennett
Token Bennett
Feature activation+0.000
family
Token family
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Oval
Token Oval
Feature activation+0.000
Office
Token Office
Feature activation+0.000
couch
Token couch
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+2.683
credit
Token credit
Feature activation+0.000
BR
Token BR
Feature activation+0.000
END
TokenEND
Feature activation+0.000
AN
TokenAN
Feature activation+0.000
S
Token S
Feature activation+0.000
Anna
Token Anna
Feature activation+0.000
K
Token K
Feature activation+0.000
roup
Tokenroup
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+4.235
courtesy
Token courtesy
Feature activation+3.015
of
Token of
Feature activation+0.000
Water
Token Water
Feature activation+0.000
ford
Tokenford
Feature activation+0.000
Police
Token Police
Feature activation+0.000
Department
Token Department
Feature activation+0.000

INTERVAL 1.813 - 2.417
CONTAINS 0.000%

the
Token the
Feature activation+0.000
weekend
Token weekend
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+1.706
courtesy
Token courtesy
Feature activation+2.023
of
Token of
Feature activation+0.000
Mazda
Token Mazda
Feature activation+0.000
North
Token North
Feature activation+0.000
American
Token American
Feature activation+0.000
Operations
Token Operations
Feature activation+0.000
Photo
TokenPhoto
Feature activation+0.000
via
Token via
Feature activation+0.000
Go
Token Go
Feature activation+0.000
Fund
TokenFund
Feature activation+0.000
Me
TokenMe
Feature activation+0.000
Photo
Token Photo
Feature activation+2.328
via
Token via
Feature activation+0.000
Go
Token Go
Feature activation+0.000
Fund
TokenFund
Feature activation+0.000
Me
TokenMe
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Boston
Token Boston
Feature activation+0.000
Marathon
Token Marathon
Feature activation+0.000
bombing
Token bombing
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+2.124
by
Token by
Feature activation+0.000
James
Token James
Feature activation+0.000
Duncan
Token Duncan
Feature activation+0.000
Davidson
Token Davidson
Feature activation+0.000
/
Token/
Feature activation+0.000
address
Token address
Feature activation+0.000
to
Token to
Feature activation+0.000
Congress
Token Congress
Feature activation+0.000
.
Token.
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+2.055
by
Token by
Feature activation+0.000
Jim
Token Jim
Feature activation+0.000
Lo
Token Lo
Feature activation+0.000
Scal
Token Scal
Feature activation+0.000
zo
Tokenzo
Feature activation+0.000
ois
Tokenois
Feature activation+0.000
in
Tokenin
Feature activation+0.000
/
Token/
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Washington
Token Washington
Feature activation+0.000
Post
Token Post
Feature activation+2.378
via
Token via
Feature activation+0.000
Getty
Token Getty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
)
Token)
Feature activation+0.000
Four
Token Four
Feature activation+0.000

INTERVAL 1.208 - 1.813
CONTAINS 0.000%

his
Token his
Feature activation+0.000
granddaughter
Token granddaughter
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
(
Token(
Feature activation+1.755
via
Tokenvia
Feature activation+0.000
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
4
Token4
Feature activation+0.000
os
Tokenos
Feature activation+0.000
Bar
Token Bar
Feature activation+0.000
ria
Tokenria
Feature activation+0.000
)
Token)
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+1.485
via
Token via
Feature activation+0.000
REUTERS
Token REUTERS
Feature activation+0.000
/
Token/
Feature activation+0.000
Carl
TokenCarl
Feature activation+0.000
os
Tokenos
Feature activation+0.000
Track
TokenTrack
Feature activation+0.000
5
Token 5
Feature activation+0.000
features
Token features
Feature activation+0.000
unc
Token unc
Feature activation+0.000
redited
Tokenredited
Feature activation+0.000
production
Token production
Feature activation+1.213
from
Token from
Feature activation+0.000
Sv
Token Sv
Feature activation+0.000
idden
Tokenidden
Feature activation+0.000
and
Token and
Feature activation+0.000
unc
Token unc
Feature activation+0.000
Track
TokenTrack
Feature activation+0.000
3
Token 3
Feature activation+0.000
features
Token features
Feature activation+0.000
unc
Token unc
Feature activation+0.000
redited
Tokenredited
Feature activation+0.000
production
Token production
Feature activation+1.347
from
Token from
Feature activation+0.000
Sv
Token Sv
Feature activation+0.000
idden
Tokenidden
Feature activation+0.000
and
Token and
Feature activation+0.000
unc
Token unc
Feature activation+0.000
Track
TokenTrack
Feature activation+0.000
4
Token 4
Feature activation+0.000
features
Token features
Feature activation+0.000
unc
Token unc
Feature activation+0.000
redited
Tokenredited
Feature activation+0.000
production
Token production
Feature activation+1.501
from
Token from
Feature activation+0.000
Sv
Token Sv
Feature activation+0.000
idden
Tokenidden
Feature activation+0.000
and
Token and
Feature activation+0.000
unc
Token unc
Feature activation+0.000

INTERVAL 0.604 - 1.208
CONTAINS 0.000%

Imperial
Token Imperial
Feature activation+0.000
Base
Token Base
Feature activation+0.000
Camp
Token Camp
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Map
TokenMap
Feature activation+0.612
of
Token of
Feature activation+0.000
B
Token B
Feature activation+0.000
els
Tokenels
Feature activation+0.000
avis
Tokenavis
Feature activation+0.000
Battlefield
Token Battlefield
Feature activation+0.000
.
Token.
Feature activation+0.000
Fig
Token Fig
Feature activation+0.000
u
Tokenu
Feature activation+0.000
arts
Tokenarts
Feature activation+0.000
Action
Token Action
Feature activation+0.000
Figure
Token Figure
Feature activation+0.817
with
Token with
Feature activation+0.000
Wall
Token Wall
Feature activation+0.000
Access
Token Access
Feature activation+0.000
ory
Tokenory
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
my
Token my
Feature activation+0.000
right
Token right
Feature activation+0.000
forearm
Token forearm
Feature activation+0.000
:
Token:
Feature activation+0.000
a
Token a
Feature activation+0.000
picture
Token picture
Feature activation+0.872
of
Token of
Feature activation+0.000
my
Token my
Feature activation+0.000
Mom
Token Mom
Feature activation+0.000
surrounded
Token surrounded
Feature activation+0.000
by
Token by
Feature activation+0.000
/
Token/
Feature activation+0.000
Getty
TokenGetty
Feature activation+0.000
Images
Token Images
Feature activation+0.000
)
Token)
Feature activation+0.000
(
Token (
Feature activation+0.000
Photo
TokenPhoto
Feature activation+1.144
by
Token by
Feature activation+0.000
Joe
Token Joe
Feature activation+0.000
Ra
Token Ra
Feature activation+0.000
ed
Tokened
Feature activation+0.000
le
Tokenle
Feature activation+0.000
page
Token page
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Phot
TokenPhot
Feature activation+0.000
ographs
Tokenographs
Feature activation+0.631
by
Token by
Feature activation+0.000
Terry
Token Terry
Feature activation+0.000
Z
Token Z
Feature activation+0.000
aper
Tokenaper
Feature activation+0.000
ach
Tokenach
Feature activation+0.000

INTERVAL 0.000 - 0.604
CONTAINS 99.999%

unknown
Token unknown
Feature activation+0.000
.
Token.
Feature activation+0.000
Using
Token Using
Feature activation+0.000
clock
Token clock
Feature activation+0.000
hour
Token hour
Feature activation+0.000
to
Token to
Feature activation+0.000
document
Token document
Feature activation+0.000
eating
Token eating
Feature activation+0.000
times
Token times
Feature activation+0.000
may
Token may
Feature activation+0.000
be
Token be
Feature activation+0.000
:
Token:
Feature activation+0.000
After
Token After
Feature activation+0.000
making
Token making
Feature activation+0.000
a
Token a
Feature activation+0.000
handful
Token handful
Feature activation+0.000
of
Token of
Feature activation+0.000
starts
Token starts
Feature activation+0.000
last
Token last
Feature activation+0.000
season
Token season
Feature activation+0.000
,
Token,
Feature activation+0.000
Motor
Token Motor
Feature activation+0.000
o
Tokeno
Feature activation+0.000
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000
learned
Token learned
Feature activation+0.000
about
Token about
Feature activation+0.000
the
Token the
Feature activation+0.000
war
Token war
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
global
Token global
Feature activation+0.000
what
Token what
Feature activation+0.000
products
Token products
Feature activation+0.000
should
Token should
Feature activation+0.000
be
Token be
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
come
Token come
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
deal
Token deal
Feature activation+0.000
.
Token.
Feature activation+0.000
them
Token them
Feature activation+0.000
,
Token,
Feature activation+0.000
Mr
Token Mr
Feature activation+0.000
.
Token.
Feature activation+0.000
Bur
Token Bur
Feature activation+0.000
po
Tokenpo
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Col
Token Col
Feature activation+0.000
ton
Tokenton
Feature activation+0.000
told
Token told
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 15: In local texts involving “yelling / screaming / shouting” at/into something

TOP ACTIVATIONS
MAX = 2.471

yelling
Token yelling
Feature activation+0.000
into
Token into
Feature activation+0.000
our
Token our
Feature activation+1.105
living
Token living
Feature activation+0.386
rooms
Token rooms
Feature activation+2.170
with
Token with
Feature activation+2.471
an
Token an
Feature activation+1.696
emotional
Token emotional
Feature activation+0.813
mixture
Token mixture
Feature activation+0.572
of
Token of
Feature activation+0.576
joy
Token joy
Feature activation+0.587
all
Token all
Feature activation+0.068
day
Token day
Feature activation+0.004
screaming
Token screaming
Feature activation+0.844
at
Token at
Feature activation+0.414
everyone
Token everyone
Feature activation+1.052
,
Token,
Feature activation+2.410
even
Token even
Feature activation+1.387
the
Token the
Feature activation+0.600
5
Token 5
Feature activation+0.000
pound
Token pound
Feature activation+0.000
-
Token-
Feature activation+0.000
,
Token,
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
into
Token into
Feature activation+0.000
our
Token our
Feature activation+1.105
living
Token living
Feature activation+0.386
rooms
Token rooms
Feature activation+2.170
with
Token with
Feature activation+2.471
an
Token an
Feature activation+1.696
emotional
Token emotional
Feature activation+0.813
mixture
Token mixture
Feature activation+0.572
of
Token of
Feature activation+0.576
ists
Tokenists
Feature activation+0.000
shri
Token shri
Feature activation+0.000
ek
Tokenek
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.703
ears
Token ears
Feature activation+2.147
of
Token of
Feature activation+1.082
the
Token the
Feature activation+0.394
police
Token police
Feature activation+0.098
that
Token that
Feature activation+1.932
Gry
Token Gry
Feature activation+0.000
your
Token your
Feature activation+0.000
feet
Token feet
Feature activation+0.005
,
Token,
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
in
Token in
Feature activation+0.294
terror
Token terror
Feature activation+2.094
,
Token,
Feature activation+1.068
and
Token and
Feature activation+0.290
managing
Token managing
Feature activation+0.000
to
Token to
Feature activation+0.000
only
Token only
Feature activation+0.000
and
Token and
Feature activation+0.000
laughing
Token laughing
Feature activation+0.000
with
Token with
Feature activation+0.000
his
Token his
Feature activation+0.568
mother
Token mother
Feature activation+0.251
as
Token as
Feature activation+2.065
he
Token he
Feature activation+0.983
spends
Token spends
Feature activation+0.132
a
Token a
Feature activation+0.000
precaution
Token precaution
Feature activation+0.000
ary
Tokenary
Feature activation+0.000
time
Token time
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+1.233
abyss
Token abyss
Feature activation+1.960
like
Token like
Feature activation+2.030
this
Token this
Feature activation+0.567
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
shouting
Token shouting
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.205
ĺ
Tokenĺ
Feature activation+0.671
wow
Tokenwow
Feature activation+1.022
âĢ
TokenâĢ
Feature activation+1.362
Ļ
TokenĻ
Feature activation+1.998
.
Token.
Feature activation+0.936
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
He
TokenHe
Feature activation+0.182
pleaded
Token pleaded
Feature activation+0.000
to
Token to
Feature activation+0.000
you
Token you
Feature activation+0.000
whispers
Token whispers
Feature activation+0.000
in
Token in
Feature activation+0.007
your
Token your
Feature activation+0.944
ear
Token ear
Feature activation+1.961
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000
an
Token an
Feature activation+0.000
instance
Token instance
Feature activation+0.000
waste
Token waste
Feature activation+0.000
time
Token time
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+1.233
abyss
Token abyss
Feature activation+1.960
like
Token like
Feature activation+2.030
this
Token this
Feature activation+0.567
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
open
Token open
Feature activation+0.000
and
Token and
Feature activation+0.000
screaming
Token screaming
Feature activation+0.004
prof
Token prof
Feature activation+0.064
an
Tokenan
Feature activation+0.272
ities
Tokenities
Feature activation+1.956
at
Token at
Feature activation+1.818
the
Token the
Feature activation+1.354
couple
Token couple
Feature activation+0.728
,
Token,
Feature activation+1.254
in
Token in
Feature activation+0.612
and
Token and
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
nonsense
Token nonsense
Feature activation+0.457
into
Token into
Feature activation+1.111
the
Token the
Feature activation+1.107
mic
Token mic
Feature activation+1.942
,
Token,
Feature activation+1.090
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.318
said
Token said
Feature activation+0.041
Without
Token Without
Feature activation+0.000
the
Token the
Feature activation+0.703
ears
Token ears
Feature activation+2.147
of
Token of
Feature activation+1.082
the
Token the
Feature activation+0.394
police
Token police
Feature activation+0.098
that
Token that
Feature activation+1.932
Gry
Token Gry
Feature activation+0.000
ns
Tokenns
Feature activation+0.000
z
Tokenz
Feature activation+0.000
pan
Tokenpan
Feature activation+0.000
attended
Token attended
Feature activation+0.000
and
Token and
Feature activation+0.000
were
Token were
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
in
Token in
Feature activation+0.000
our
Token our
Feature activation+0.699
faces
Token faces
Feature activation+1.922
,"
Token,"
Feature activation+0.000
said
Token said
Feature activation+0.000
Barn
Token Barn
Feature activation+0.000
ard
Tokenard
Feature activation+0.000
,
Token,
Feature activation+0.000
,
Token,
Feature activation+0.000
cried
Token cried
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.515
dock
Token dock
Feature activation+0.033
as
Token as
Feature activation+1.917
he
Token he
Feature activation+1.115
was
Token was
Feature activation+0.747
jailed
Token jailed
Feature activation+0.000
for
Token for
Feature activation+0.000
three
Token three
Feature activation+0.000
screamed
Token screamed
Feature activation+0.000
across
Token across
Feature activation+0.254
her
Token her
Feature activation+0.548
whole
Token whole
Feature activation+0.034
body
Token body
Feature activation+1.276
as
Token as
Feature activation+1.906
she
Token she
Feature activation+0.615
forced
Token forced
Feature activation+0.000
herself
Token herself
Feature activation+0.000
into
Token into
Feature activation+0.000
a
Token a
Feature activation+0.000
and
Token and
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
into
Token into
Feature activation+0.013
the
Token the
Feature activation+0.873
microphone
Token microphone
Feature activation+1.711
as
Token as
Feature activation+1.904
she
Token she
Feature activation+0.704
belts
Token belts
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.324
song
Token song
Feature activation+0.000
bon
Token bon
Feature activation+0.000
net
Tokennet
Feature activation+0.000
and
Token and
Feature activation+0.069
yelling
Token yelling
Feature activation+0.000
as
Token as
Feature activation+0.730
if
Token if
Feature activation+1.881
they
Token they
Feature activation+1.490
had
Token had
Feature activation+1.717
been
Token been
Feature activation+0.997
run
Token run
Feature activation+0.000
into
Token into
Feature activation+0.000
"
Token"
Feature activation+0.518
and
Token and
Feature activation+1.061
shouting
Token shouting
Feature activation+0.748
prof
Token prof
Feature activation+0.000
an
Tokenan
Feature activation+0.000
ities
Tokenities
Feature activation+1.834
and
Token and
Feature activation+1.537
anti
Token anti
Feature activation+0.843
-
Token-
Feature activation+0.057
hom
Tokenhom
Feature activation+0.000
osexual
Tokenosexual
Feature activation+0.000
and
Token and
Feature activation+0.000
screaming
Token screaming
Feature activation+0.004
prof
Token prof
Feature activation+0.064
an
Tokenan
Feature activation+0.272
ities
Tokenities
Feature activation+1.956
at
Token at
Feature activation+1.818
the
Token the
Feature activation+1.354
couple
Token couple
Feature activation+0.728
,
Token,
Feature activation+1.254
in
Token in
Feature activation+0.612
which
Token which
Feature activation+0.994

Top DFA by src position
MAX = 2.393

pretty
Token pretty
Feature activation+0.019
blonde
Token blonde
Feature activation+0.018
woman
Token woman
Feature activation+0.020
,
Token,
Feature activation+0.046
yelling
Token yelling
Feature activation+0.333
into
Token into
Feature activation+2.057
our
Token our
Feature activation+0.489
living
Token living
Feature activation+0.023
rooms
Token rooms
Feature activation+0.151
with
Token with
Feature activation+0.174
an
Token an
Feature activation+0.000
stormed
Token stormed
Feature activation+0.091
around
Token around
Feature activation+0.104
all
Token all
Feature activation+0.066
day
Token day
Feature activation+0.008
screaming
Token screaming
Feature activation+0.663
at
Token at
Feature activation+1.634
everyone
Token everyone
Feature activation+0.479
,
Token,
Feature activation+0.167
even
Token even
Feature activation+0.000
the
Token the
Feature activation+0.000
5
Token 5
Feature activation+0.000
pretty
Token pretty
Feature activation+0.016
blonde
Token blonde
Feature activation+0.012
woman
Token woman
Feature activation+0.038
,
Token,
Feature activation+0.069
yelling
Token yelling
Feature activation+0.447
into
Token into
Feature activation+1.643
our
Token our
Feature activation+0.357
living
Token living
Feature activation+0.153
rooms
Token rooms
Feature activation+0.098
with
Token with
Feature activation+0.000
an
Token an
Feature activation+0.000
The
TokenThe
Feature activation-0.015
Stalin
Token Stalin
Feature activation-0.013
ists
Tokenists
Feature activation-0.015
shri
Token shri
Feature activation+0.210
ek
Tokenek
Feature activation+0.385
in
Token in
Feature activation+2.165
the
Token the
Feature activation+0.394
ears
Token ears
Feature activation+0.251
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
police
Token police
Feature activation+0.000
underneath
Token underneath
Feature activation+0.062
your
Token your
Feature activation+0.016
feet
Token feet
Feature activation+0.024
,
Token,
Feature activation-0.007
screaming
Token screaming
Feature activation+0.401
in
Token in
Feature activation+2.268
terror
Token terror
Feature activation+0.185
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
managing
Token managing
Feature activation+0.000
to
Token to
Feature activation+0.000
Ethan
Token Ethan
Feature activation+0.001
is
Token is
Feature activation+0.007
talking
Token talking
Feature activation+0.045
and
Token and
Feature activation+0.166
laughing
Token laughing
Feature activation+0.182
with
Token with
Feature activation+2.119
his
Token his
Feature activation+0.218
mother
Token mother
Feature activation+0.053
as
Token as
Feature activation+0.415
he
Token he
Feature activation+0.000
spends
Token spends
Feature activation+0.000
,
Token,
Feature activation-0.007
why
Token why
Feature activation+0.032
waste
Token waste
Feature activation+0.120
time
Token time
Feature activation+0.021
screaming
Token screaming
Feature activation+0.306
into
Token into
Feature activation+2.222
the
Token the
Feature activation+0.198
abyss
Token abyss
Feature activation+0.109
like
Token like
Feature activation+0.044
this
Token this
Feature activation+0.000
?
Token?
Feature activation+0.000
a
Token a
Feature activation+0.006
star
Token star
Feature activation+0.009
position
Token position
Feature activation+0.004
,
Token,
Feature activation+0.051
shouting
Token shouting
Feature activation+0.802
âĢ
Token âĢ
Feature activation+1.639
ĺ
Tokenĺ
Feature activation+0.061
wow
Tokenwow
Feature activation+0.207
âĢ
TokenâĢ
Feature activation+0.143
Ļ
TokenĻ
Feature activation+0.030
.
Token.
Feature activation+0.000
floor
Token floor
Feature activation+0.006
next
Token next
Feature activation+0.014
to
Token to
Feature activation-0.013
you
Token you
Feature activation+0.030
whispers
Token whispers
Feature activation+0.587
in
Token in
Feature activation+1.959
your
Token your
Feature activation+0.310
ear
Token ear
Feature activation+0.125
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000
,
Token,
Feature activation-0.123
why
Token why
Feature activation+0.005
waste
Token waste
Feature activation+0.048
time
Token time
Feature activation+0.054
screaming
Token screaming
Feature activation+0.710
into
Token into
Feature activation+1.798
the
Token the
Feature activation+0.386
abyss
Token abyss
Feature activation+0.058
like
Token like
Feature activation+0.000
this
Token this
Feature activation+0.000
?
Token?
Feature activation+0.000
's
Token's
Feature activation+0.022
doors
Token doors
Feature activation-0.013
open
Token open
Feature activation-0.087
and
Token and
Feature activation+0.072
screaming
Token screaming
Feature activation+0.799
prof
Token prof
Feature activation+1.927
an
Tokenan
Feature activation+0.002
ities
Tokenities
Feature activation+0.047
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
couple
Token couple
Feature activation+0.000
my
Token my
Feature activation+0.015
back
Token back
Feature activation+0.009
and
Token and
Feature activation+0.005
yelling
Token yelling
Feature activation+0.257
nonsense
Token nonsense
Feature activation+0.144
into
Token into
Feature activation+1.689
the
Token the
Feature activation+0.169
mic
Token mic
Feature activation+0.062
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
The
TokenThe
Feature activation-0.018
Stalin
Token Stalin
Feature activation-0.016
ists
Tokenists
Feature activation-0.029
shri
Token shri
Feature activation+0.192
ek
Tokenek
Feature activation+0.441
in
Token in
Feature activation+1.856
the
Token the
Feature activation+0.176
ears
Token ears
Feature activation+0.164
of
Token of
Feature activation+0.100
the
Token the
Feature activation+0.097
police
Token police
Feature activation+0.010
the
Token the
Feature activation-0.011
stands
Token stands
Feature activation-0.002
and
Token and
Feature activation-0.060
were
Token were
Feature activation-0.015
screaming
Token screaming
Feature activation+0.472
in
Token in
Feature activation+2.393
our
Token our
Feature activation+0.413
faces
Token faces
Feature activation+0.025
,"
Token,"
Feature activation+0.000
said
Token said
Feature activation+0.000
Barn
Token Barn
Feature activation+0.000
Plymouth
Token Plymouth
Feature activation+0.000
,
Token,
Feature activation-0.003
Devon
Token Devon
Feature activation-0.003
,
Token,
Feature activation+0.002
cried
Token cried
Feature activation+0.562
in
Token in
Feature activation+1.900
the
Token the
Feature activation+0.305
dock
Token dock
Feature activation+0.056
as
Token as
Feature activation+0.205
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
P
Token P
Feature activation+0.016
ins
Tokenins
Feature activation-0.000
and
Token and
Feature activation+0.011
needles
Token needles
Feature activation-0.011
screamed
Token screamed
Feature activation+0.434
across
Token across
Feature activation+1.853
her
Token her
Feature activation+0.107
whole
Token whole
Feature activation+0.025
body
Token body
Feature activation+0.046
as
Token as
Feature activation+0.256
she
Token she
Feature activation+0.000
-
Token-
Feature activation-0.012
b
Tokenb
Feature activation+0.001
anging
Tokenanging
Feature activation+0.033
and
Token and
Feature activation+0.079
screaming
Token screaming
Feature activation+0.211
into
Token into
Feature activation+2.048
the
Token the
Feature activation+0.231
microphone
Token microphone
Feature activation+0.088
as
Token as
Feature activation+0.215
she
Token she
Feature activation+0.000
belts
Token belts
Feature activation+0.000
the
Token the
Feature activation+0.044
bon
Token bon
Feature activation+0.008
net
Tokennet
Feature activation-0.002
and
Token and
Feature activation+0.013
yelling
Token yelling
Feature activation+0.541
as
Token as
Feature activation+1.624
if
Token if
Feature activation+0.346
they
Token they
Feature activation+0.000
had
Token had
Feature activation+0.000
been
Token been
Feature activation+0.000
run
Token run
Feature activation+0.000
,
Token,
Feature activation+0.028
Princeton
Token Princeton
Feature activation+0.001
"
Token"
Feature activation+0.066
and
Token and
Feature activation+0.095
shouting
Token shouting
Feature activation+0.598
prof
Token prof
Feature activation+1.579
an
Tokenan
Feature activation+0.004
ities
Tokenities
Feature activation+0.048
and
Token and
Feature activation+0.000
anti
Token anti
Feature activation+0.000
-
Token-
Feature activation+0.000
's
Token's
Feature activation+0.002
doors
Token doors
Feature activation-0.010
open
Token open
Feature activation-0.054
and
Token and
Feature activation+0.063
screaming
Token screaming
Feature activation+0.717
prof
Token prof
Feature activation+1.262
an
Tokenan
Feature activation-0.001
ities
Tokenities
Feature activation+0.125
at
Token at
Feature activation+0.436
the
Token the
Feature activation+0.000
couple
Token couple
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.03

Head 1: 0.01

Head 2: 0.26

Head 3: 0.10

Head 4: 0.12

Head 5: 0.03

Head 6: 0.04

Head 7: 0.10

Head 8: 0.07

Head 9: 0.04

Head 10: 0.09

Head 11: 0.11

Positive logits

plaint1.66

louder1.61

angrily1.51

incess1.51

inco1.51

loudly1.50

indign1.48

furiously1.45

uncontroll1.43

insomnia1.41

insults1.40

slurs1.38

yells1.36

behalf1.36

strokes1.35

loud1.35

gunshots1.33

redd1.31

endlessly1.31

obsc1.31

Negative logits

ngth-1.46

chnology-1.37

Pearson-1.33

accredited-1.32

ruled-1.30

Prototype-1.28

ushima-1.26

egu-1.25

══-1.23

ortium-1.23

Plant-1.23

Location-1.21

Referred-1.19

Recommended-1.19

Production-1.18

DonaldTrump-1.18

Towns-1.17

Coral-1.16

erto-1.16

OUP-1.16

INTERVAL 2.224 - 2.471
CONTAINS 0.000%

all
Token all
Feature activation+0.068
day
Token day
Feature activation+0.004
screaming
Token screaming
Feature activation+0.844
at
Token at
Feature activation+0.414
everyone
Token everyone
Feature activation+1.052
,
Token,
Feature activation+2.410
even
Token even
Feature activation+1.387
the
Token the
Feature activation+0.600
5
Token 5
Feature activation+0.000
pound
Token pound
Feature activation+0.000
-
Token-
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
into
Token into
Feature activation+0.000
our
Token our
Feature activation+1.105
living
Token living
Feature activation+0.386
rooms
Token rooms
Feature activation+2.170
with
Token with
Feature activation+2.471
an
Token an
Feature activation+1.696
emotional
Token emotional
Feature activation+0.813
mixture
Token mixture
Feature activation+0.572
of
Token of
Feature activation+0.576
joy
Token joy
Feature activation+0.587

INTERVAL 1.976 - 2.224
CONTAINS 0.000%

shouting
Token shouting
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.205
ĺ
Tokenĺ
Feature activation+0.671
wow
Tokenwow
Feature activation+1.022
âĢ
TokenâĢ
Feature activation+1.362
Ļ
TokenĻ
Feature activation+1.998
.
Token.
Feature activation+0.936
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
He
TokenHe
Feature activation+0.182
pleaded
Token pleaded
Feature activation+0.000
time
Token time
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
into
Token into
Feature activation+0.000
the
Token the
Feature activation+1.233
abyss
Token abyss
Feature activation+1.960
like
Token like
Feature activation+2.030
this
Token this
Feature activation+0.567
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
and
Token and
Feature activation+0.000
laughing
Token laughing
Feature activation+0.000
with
Token with
Feature activation+0.000
his
Token his
Feature activation+0.568
mother
Token mother
Feature activation+0.251
as
Token as
Feature activation+2.065
he
Token he
Feature activation+0.983
spends
Token spends
Feature activation+0.132
a
Token a
Feature activation+0.000
precaution
Token precaution
Feature activation+0.000
ary
Tokenary
Feature activation+0.000
,
Token,
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
into
Token into
Feature activation+0.000
our
Token our
Feature activation+1.105
living
Token living
Feature activation+0.386
rooms
Token rooms
Feature activation+2.170
with
Token with
Feature activation+2.471
an
Token an
Feature activation+1.696
emotional
Token emotional
Feature activation+0.813
mixture
Token mixture
Feature activation+0.572
of
Token of
Feature activation+0.576
ists
Tokenists
Feature activation+0.000
shri
Token shri
Feature activation+0.000
ek
Tokenek
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.703
ears
Token ears
Feature activation+2.147
of
Token of
Feature activation+1.082
the
Token the
Feature activation+0.394
police
Token police
Feature activation+0.098
that
Token that
Feature activation+1.932
Gry
Token Gry
Feature activation+0.000

INTERVAL 1.729 - 1.976
CONTAINS 0.000%

to
Token to
Feature activation+0.000
you
Token you
Feature activation+0.000
whispers
Token whispers
Feature activation+0.000
in
Token in
Feature activation+0.007
your
Token your
Feature activation+0.944
ear
Token ear
Feature activation+1.961
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
is
Token is
Feature activation+0.000
an
Token an
Feature activation+0.000
instance
Token instance
Feature activation+0.000
and
Token and
Feature activation+0.000
screaming
Token screaming
Feature activation+0.004
prof
Token prof
Feature activation+0.064
an
Tokenan
Feature activation+0.272
ities
Tokenities
Feature activation+1.956
at
Token at
Feature activation+1.818
the
Token the
Feature activation+1.354
couple
Token couple
Feature activation+0.728
,
Token,
Feature activation+1.254
in
Token in
Feature activation+0.612
which
Token which
Feature activation+0.994
the
Token the
Feature activation+0.703
ears
Token ears
Feature activation+2.147
of
Token of
Feature activation+1.082
the
Token the
Feature activation+0.394
police
Token police
Feature activation+0.098
that
Token that
Feature activation+1.932
Gry
Token Gry
Feature activation+0.000
ns
Tokenns
Feature activation+0.000
z
Tokenz
Feature activation+0.000
pan
Tokenpan
Feature activation+0.000
attended
Token attended
Feature activation+0.000
,
Token,
Feature activation+0.000
cried
Token cried
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.515
dock
Token dock
Feature activation+0.033
as
Token as
Feature activation+1.917
he
Token he
Feature activation+1.115
was
Token was
Feature activation+0.747
jailed
Token jailed
Feature activation+0.000
for
Token for
Feature activation+0.000
three
Token three
Feature activation+0.000
angry
Token angry
Feature activation+0.000
black
Token black
Feature activation+0.000
activist
Token activist
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
in
Token in
Feature activation+0.000
panic
Token panic
Feature activation+1.744
about
Token about
Feature activation+1.758
the
Token the
Feature activation+0.927
police
Token police
Feature activation+0.437
being
Token being
Feature activation+0.181
out
Token out
Feature activation+0.000

INTERVAL 1.482 - 1.729
CONTAINS 0.000%

anging
Tokenanging
Feature activation+0.000
and
Token and
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
into
Token into
Feature activation+0.013
the
Token the
Feature activation+0.873
microphone
Token microphone
Feature activation+1.711
as
Token as
Feature activation+1.904
she
Token she
Feature activation+0.704
belts
Token belts
Feature activation+0.000
out
Token out
Feature activation+0.000
the
Token the
Feature activation+0.324
screaming
Token screaming
Feature activation+0.000
and
Token and
Feature activation+0.000
crying
Token crying
Feature activation+0.137
with
Token with
Feature activation+0.828
delight
Token delight
Feature activation+1.344
as
Token as
Feature activation+1.632
they
Token they
Feature activation+1.648
open
Token open
Feature activation+0.522
the
Token the
Feature activation+0.000
presents
Token presents
Feature activation+0.000
and
Token and
Feature activation+0.000
street
Token street
Feature activation+0.000
and
Token and
Feature activation+0.000
shouting
Token shouting
Feature activation+0.000
exp
Token exp
Feature activation+0.000
let
Tokenlet
Feature activation+0.000
ives
Tokenives
Feature activation+1.532
about
Token about
Feature activation+0.673
Donald
Token Donald
Feature activation+0.000
Trump
Token Trump
Feature activation+0.005
in
Token in
Feature activation+0.197
English
Token English
Feature activation+0.616
up
Token up
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
for
Token for
Feature activation+0.421
no
Token no
Feature activation+0.405
fucking
Token fucking
Feature activation+0.659
reason
Token reason
Feature activation+1.688
.
Token.
Feature activation+0.067
But
Token But
Feature activation+0.000
you
Token you
Feature activation+0.000
know
Token know
Feature activation+0.000
what
Token what
Feature activation+0.000
someone
Token someone
Feature activation+0.000
is
Token is
Feature activation+0.000
screaming
Token screaming
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.665
ear
Token ear
Feature activation+1.615
,
Token,
Feature activation+0.000
keeping
Token keeping
Feature activation+0.000
him
Token him
Feature activation+0.018
from
Token from
Feature activation+0.040
communicating
Token communicating
Feature activation+0.000

INTERVAL 1.235 - 1.482
CONTAINS 0.001%

top
Token top
Feature activation+0.074
of
Token of
Feature activation+0.942
my
Token my
Feature activation+1.239
lungs
Token lungs
Feature activation+1.427
,
Token,
Feature activation+1.348
cheering
Token cheering
Feature activation+1.297
on
Token on
Feature activation+0.501
my
Token my
Feature activation+1.189
big
Token big
Feature activation+0.194
brother
Token brother
Feature activation+0.525
.
Token.
Feature activation+0.715
said
Token said
Feature activation+0.000
trolls
Token trolls
Feature activation+0.000
whim
Token whim
Feature activation+0.000
pering
Tokenpering
Feature activation+0.041
for
Token for
Feature activation+0.186
mercy
Token mercy
Feature activation+1.286
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
OK
TokenOK
Feature activation+0.000
,
Token,
Feature activation+0.000
.
Token.
Feature activation+0.000
She
Token She
Feature activation+0.000
screamed
Token screamed
Feature activation+0.000
in
Token in
Feature activation+0.002
pain
Token pain
Feature activation+1.333
,
Token,
Feature activation+1.269
trying
Token trying
Feature activation+1.197
to
Token to
Feature activation+0.590
push
Token push
Feature activation+0.667
me
Token me
Feature activation+0.300
off
Token off
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
sigh
Token sigh
Feature activation+0.000
with
Token with
Feature activation+0.018
relief
Token relief
Feature activation+1.386
when
Token when
Feature activation+0.623
they
Token they
Feature activation+0.388
aren
Token aren
Feature activation+0.058
't
Token't
Feature activation+0.000
.
Token.
Feature activation+0.000
he
Token he
Feature activation+0.000
whispered
Token whispered
Feature activation+0.000
threats
Token threats
Feature activation+0.068
in
Token in
Feature activation+0.365
his
Token his
Feature activation+0.981
ear
Token ear
Feature activation+1.449
.
Token.
Feature activation+0.224
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
The
TokenThe
Feature activation+0.000

INTERVAL 0.988 - 1.235
CONTAINS 0.002%

by
Token by
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
to
Token to
Feature activation+0.296
call
Token call
Feature activation+0.498
911
Token 911
Feature activation+0.375
because
Token because
Feature activation+1.051
Smith
Token Smith
Feature activation+0.297
had
Token had
Feature activation+0.383
overd
Token overd
Feature activation+0.000
osed
Tokenosed
Feature activation+0.000
in
Token in
Feature activation+0.000
roaring
Token roaring
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.048
the
Token the
Feature activation+0.000
traps
Token traps
Feature activation+0.009
to
Token to
Feature activation+1.077
reel
Token reel
Feature activation+0.104
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Kingdom
Token Kingdom
Feature activation+0.000
.
Token.
Feature activation+0.000
started
Token started
Feature activation+0.000
yelling
Token yelling
Feature activation+0.000
at
Token at
Feature activation+0.099
a
Token a
Feature activation+0.718
woman
Token woman
Feature activation+0.463
and
Token and
Feature activation+1.127
shot
Token shot
Feature activation+0.460
her
Token her
Feature activation+0.000
,
Token,
Feature activation+0.058
according
Token according
Feature activation+0.000
to
Token to
Feature activation+0.000
confronted
Token confronted
Feature activation+0.000
plaintiff
Token plaintiff
Feature activation+0.000
,
Token,
Feature activation+0.000
cursing
Token cursing
Feature activation+0.000
at
Token at
Feature activation+0.120
him
Token him
Feature activation+1.017
,
Token,
Feature activation+1.326
and
Token and
Feature activation+0.458
then
Token then
Feature activation+0.075
struck
Token struck
Feature activation+0.000
him
Token him
Feature activation+0.000
laugh
Token laugh
Feature activation+0.000
over
Token over
Feature activation+0.000
a
Token a
Feature activation+0.526
decade
Token decade
Feature activation+0.000
later
Token later
Feature activation+0.000
at
Token at
Feature activation+1.086
a
Token a
Feature activation+0.757
University
Token University
Feature activation+0.000
of
Token of
Feature activation+0.000
Florida
Token Florida
Feature activation+0.000
press
Token press
Feature activation+0.000

INTERVAL 0.741 - 0.988
CONTAINS 0.005%

ľ
Tokenľ
Feature activation+0.244
Hillary
TokenHillary
Feature activation+0.000
!
Token!
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.011
Ŀ
TokenĿ
Feature activation+0.185
and
Token and
Feature activation+0.850
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
love
Token love
Feature activation+0.000
you
Token you
Feature activation+0.000
May
Token May
Feature activation+0.000
and
Token and
Feature activation+0.000
enthusiastically
Token enthusiastically
Feature activation+0.000
chanted
Token chanted
Feature activation+0.000
her
Token her
Feature activation+0.191
name
Token name
Feature activation+0.746
.
Token.
Feature activation+0.046
Later
Token Later
Feature activation+0.000
,
Token,
Feature activation+0.000
Not
Token Not
Feature activation+0.000
ley
Tokenley
Feature activation+0.000
ace
Tokenace
Feature activation+0.000
smiled
Token smiled
Feature activation+0.000
as
Token as
Feature activation+0.000
if
Token if
Feature activation+0.397
he
Token he
Feature activation+0.440
couldn
Token couldn
Feature activation+0.775
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
quite
Token quite
Feature activation+0.111
believe
Token believe
Feature activation+0.704
away
Token away
Feature activation+0.000
.
Token.
Feature activation+0.000
He
Token He
Feature activation+0.000
chuckled
Token chuckled
Feature activation+0.000
to
Token to
Feature activation+0.000
himself
Token himself
Feature activation+0.984
,
Token,
Feature activation+1.116
rubbed
Token rubbed
Feature activation+0.000
his
Token his
Feature activation+0.000
eyes
Token eyes
Feature activation+0.000
,
Token,
Feature activation+0.000
about
Token about
Feature activation+0.000
and
Token and
Feature activation+0.181
cry
Token cry
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.681
slightest
Token slightest
Feature activation+0.959
bit
Token bit
Feature activation+1.427
of
Token of
Feature activation+1.268
pain
Token pain
Feature activation+1.403
?
Token?
Feature activation+0.000
Could
Token Could
Feature activation+0.000

INTERVAL 0.494 - 0.741
CONTAINS 0.013%

two
Token two
Feature activation+0.000
men
Token men
Feature activation+0.000
sang
Token sang
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
tune
Token tune
Feature activation+0.558
of
Token of
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Born
TokenBorn
Feature activation+0.000
to
Token to
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
cry
Token cry
Feature activation+0.000
like
Token like
Feature activation+0.000
a
Token a
Feature activation+0.680
baby
Token baby
Feature activation+0.555
.
Token.
Feature activation+0.518
Seriously
Token Seriously
Feature activation+0.000
,
Token,
Feature activation+0.000
piss
Token piss
Feature activation+0.000
yourself
Token yourself
Feature activation+0.000
if
Token if
Feature activation+0.000
was
Token was
Feature activation+0.000
pacing
Token pacing
Feature activation+0.000
,
Token,
Feature activation+0.000
concern
Token concern
Feature activation+0.000
filing
Token filing
Feature activation+0.000
her
Token her
Feature activation+0.522
stride
Token stride
Feature activation+0.041
.
Token.
Feature activation+0.000
She
Token She
Feature activation+0.000
wore
Token wore
Feature activation+0.000
a
Token a
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
po
Tokenpo
Feature activation+0.000
oping
Tokenoping
Feature activation+0.000
back
Token back
Feature activation+0.000
and
Token and
Feature activation+0.509
forth
Token forth
Feature activation+0.703
forever
Token forever
Feature activation+0.320
âĢ
TokenâĢ
Feature activation+0.179
Ŀ
TokenĿ
Feature activation+0.165
and
Token and
Feature activation+0.304
âĢ
Token âĢ
Feature activation+0.169
as
Token as
Feature activation+0.000
mobs
Token mobs
Feature activation+0.000
shouted
Token shouted
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.030
streets
Token streets
Feature activation+0.678
and
Token and
Feature activation+0.618
lit
Token lit
Feature activation+0.122
imp
Token imp
Feature activation+0.000
romptu
Tokenromptu
Feature activation+0.000
bon
Token bon
Feature activation+0.000

INTERVAL 0.247 - 0.494
CONTAINS 0.037%

disdain
Token disdain
Feature activation+0.000
ful
Tokenful
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
humans
Token humans
Feature activation+0.000
,
Token,
Feature activation+0.284
he
Token he
Feature activation+0.071
hates
Token hates
Feature activation+0.000
them
Token them
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
her
Token her
Feature activation+0.000
for
Token for
Feature activation+0.000
at
Token at
Feature activation+0.000
least
Token least
Feature activation+0.304
ten
Token ten
Feature activation+0.191
minutes
Token minutes
Feature activation+0.311
,
Token,
Feature activation+0.032
Schultz
Token Schultz
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
with
Token with
Feature activation+0.000
guilt
Token guilt
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
considerable
Token considerable
Feature activation+0.053
amount
Token amount
Feature activation+0.365
of
Token of
Feature activation+0.003
time
Token time
Feature activation+0.364
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
Sc
Token Sc
Feature activation+0.000
and
Token and
Feature activation+0.000
flies
Token flies
Feature activation+0.000
buzzing
Token buzzing
Feature activation+0.000
around
Token around
Feature activation+0.000
,
Token,
Feature activation+0.089
high
Token high
Feature activation+0.353
summer
Token summer
Feature activation+0.000
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000
were
Token were
Feature activation+0.000
in
Token in
Feature activation+0.000
began
Token began
Feature activation+0.000
to
Token to
Feature activation+0.000
soak
Token soak
Feature activation+0.000
the
Token the
Feature activation+0.000
pages
Token pages
Feature activation+0.000
as
Token as
Feature activation+0.275
I
Token I
Feature activation+0.160
tried
Token tried
Feature activation+0.000
to
Token to
Feature activation+0.000
dab
Token dab
Feature activation+0.000
them
Token them
Feature activation+0.000

INTERVAL 0.000 - 0.247
CONTAINS 99.941%

why
Token why
Feature activation+0.000
.
Token.
Feature activation+0.000
Please
Token Please
Feature activation+0.000
note
Token note
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation+0.000
following
Token following
Feature activation+0.000
specs
Token specs
Feature activation+0.000
have
Token have
Feature activation+0.000
been
Token been
Feature activation+0.000
copied
Token copied
Feature activation+0.000
depict
Token depict
Feature activation+0.000
the
Token the
Feature activation+0.000
changes
Token changes
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
population
Token population
Feature activation+0.000
between
Token between
Feature activation+0.000
the
Token the
Feature activation+0.000
old
Token old
Feature activation+0.000
(
Token (
Feature activation+0.000
65
Token65
Feature activation+0.000
nevertheless
Token nevertheless
Feature activation+0.000
entered
Token entered
Feature activation+0.000
for
Token for
Feature activation+0.000
both
Token both
Feature activation+0.000
of
Token of
Feature activation+0.000
C
Token C
Feature activation+0.000
et
Tokenet
Feature activation+0.000
in
Tokenin
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+0.000
that
Token that
Feature activation+0.000
is
Token is
Feature activation+0.000
all
Token all
Feature activation+0.000
I
Token I
Feature activation+0.000
can
Token can
Feature activation+0.000
say
Token say
Feature activation+0.000
right
Token right
Feature activation+0.000
now
Token now
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
Reports
TokenReports
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
federal
Token federal
Feature activation+0.000
government
Token government
Feature activation+0.000
is
Token is
Feature activation+0.000
pulling
Token pulling
Feature activation+0.000
the
Token the
Feature activation+0.000
plug
Token plug
Feature activation+0.000
on
Token on
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 16: Activates on the photo in “Buy |Photo”

TOP ACTIVATIONS
MAX = 6.438

Ind
TokenInd
Feature activation+0.000
y
Tokeny
Feature activation+0.000
Star
TokenStar
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.989
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Indiana
Token Indiana
Feature activation+0.000
Republican
Token Republican
Feature activation+0.000
:
Token:
Feature activation+0.000
Indianapolis
Token Indianapolis
Feature activation+0.000
Star
Token Star
Feature activation+0.000
)
Token )
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.833
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Gro
TokenGro
Feature activation+0.000
an
Tokenan
Feature activation+0.000
.
Token.
Feature activation+0.000
Community
Token Community
Feature activation+0.000
Rec
Token Rec
Feature activation+0.000
order
Tokenorder
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.791
Story
Token Story
Feature activation+3.190
Highlights
Token Highlights
Feature activation+0.000
Fox
Token Fox
Feature activation+0.000
Run
Token Run
Feature activation+0.000
Produ
Token Produ
Feature activation+0.000
Detroit
Token Detroit
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.703
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
apartment
Token apartment
Feature activation+0.000
rental
Token rental
Feature activation+0.000
Detroit
Token Detroit
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.323
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
21
Token 21
Feature activation+0.000
-
Token-
Feature activation+0.000
OUR
TokenOUR
Feature activation+0.000
N
TokenN
Feature activation+0.000
AL
TokenAL
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.306
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jo
TokenJo
Feature activation+0.000
el
Tokenel
Feature activation+0.000
Sau
Token Sau
Feature activation+0.000
Colorado
Token Colorado
Feature activation+0.000
an
Tokenan
Feature activation+0.000
library
Token library
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.226
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
There
TokenThere
Feature activation+0.000
are
Token are
Feature activation+0.000
still
Token still
Feature activation+0.000
Ad
TokenAd
Feature activation+0.000
vertis
Tokenvertis
Feature activation+0.000
er
Tokener
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.213
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Develop
TokenDevelop
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
plan
Token plan
Feature activation+0.000
UNE
TokenUNE
Feature activation+0.000
FILE
Token FILE
Feature activation+0.000
PHOTO
Token PHOTO
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.182
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
For
TokenFor
Feature activation+0.000
two
Token two
Feature activation+0.000
years
Token years
Feature activation+0.000
Journal
TokenJournal
Feature activation+0.000
&
Token &
Feature activation+0.000
Courier
Token Courier
Feature activation+0.000
)
Token )
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.108
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
C
TokenC
Feature activation+0.000
ody
Tokenody
Feature activation+0.000
Cousins
Token Cousins
Feature activation+0.000
j
Tokenj
Feature activation+0.000
.
Token.
Feature activation+0.000
com
Tokencom
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.015
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
1974
Token 1974
Feature activation+0.000
,
Token,
Feature activation+0.000
Tenn
Token Tenn
Feature activation+0.000
esse
Tokenesse
Feature activation+0.000
an
Tokenan
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.998
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jason
TokenJason
Feature activation+0.000
Is
Token Is
Feature activation+0.000
bell
Tokenbell
Feature activation+0.000
-
Token-
Feature activation+0.000
Led
TokenLed
Feature activation+0.000
ger
Tokenger
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.962
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Authorities
TokenAuthorities
Feature activation+0.000
say
Token say
Feature activation+0.000
the
Token the
Feature activation+0.000
The
Token The
Feature activation+0.000
Desert
Token Desert
Feature activation+0.000
Sun
Token Sun
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.942
Story
Token Story
Feature activation+3.266
Highlights
Token Highlights
Feature activation+0.000
Police
Token Police
Feature activation+0.000
shooting
Token shooting
Feature activation+0.000
case
Token case
Feature activation+0.000
/
Token /
Feature activation+0.000
The
Token The
Feature activation+0.000
Star
Token Star
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.767
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Z
Token Z
Feature activation+0.000
ions
Tokenions
Feature activation+0.000
Chris
TokenChris
Feature activation+0.000
May
Token May
Feature activation+0.000
hew
Tokenhew
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.762
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
CR
TokenCR
Feature activation+0.000
ES
TokenES
Feature activation+0.000
CENT
TokenCENT
Feature activation+0.000
Milwaukee
Token Milwaukee
Feature activation+0.000
Journal
Token Journal
Feature activation+0.000
Sentinel
Token Sentinel
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.709
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
W
TokenW
Feature activation+0.000
au
Tokenau
Feature activation+0.000
kes
Tokenkes
Feature activation+0.000
Detroit
Token Detroit
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.468
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
From
TokenFrom
Feature activation+0.000
a
Token a
Feature activation+0.000
tin
Token tin
Feature activation+0.000
States
Token States
Feature activation+0.000
man
Tokenman
Feature activation+0.000
Journal
Token Journal
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.429
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Oregon
TokenOregon
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
took
Token took
Feature activation+0.000
The
Token The
Feature activation+0.000
Desert
Token Desert
Feature activation+0.000
Sun
Token Sun
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+3.925
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
On
TokenOn
Feature activation+0.000
a
Token a
Feature activation+0.000
recent
Token recent
Feature activation+0.000

Top DFA by src position
MAX = 8.182

/
Token/
Feature activation-0.026
Ind
TokenInd
Feature activation-0.009
y
Tokeny
Feature activation+0.027
Star
TokenStar
Feature activation-0.011
)
Token)
Feature activation+0.151
Buy
TokenBuy
Feature activation+8.069
Photo
Token Photo
Feature activation+0.099
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Indiana
Token Indiana
Feature activation+0.000
Photo
TokenPhoto
Feature activation+0.015
:
Token:
Feature activation+0.032
Indianapolis
Token Indianapolis
Feature activation-0.050
Star
Token Star
Feature activation-0.008
)
Token )
Feature activation+0.186
Buy
TokenBuy
Feature activation+8.073
Photo
Token Photo
Feature activation-0.672
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Gro
TokenGro
Feature activation+0.000
an
Tokenan
Feature activation+0.000
The
TokenThe
Feature activation-0.003
Community
Token Community
Feature activation-0.010
Rec
Token Rec
Feature activation-0.023
order
Tokenorder
Feature activation-0.020
)
Token)
Feature activation+0.300
Buy
TokenBuy
Feature activation+8.012
Photo
Token Photo
Feature activation+0.265
Story
Token Story
Feature activation+0.000
Highlights
Token Highlights
Feature activation+0.000
Fox
Token Fox
Feature activation+0.000
Run
Token Run
Feature activation+0.000
,
Token,
Feature activation+0.020
Detroit
Token Detroit
Feature activation+0.003
Free
Token Free
Feature activation-0.011
Press
Token Press
Feature activation+0.012
)
Token)
Feature activation+0.142
Buy
TokenBuy
Feature activation+7.772
Photo
Token Photo
Feature activation+0.048
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
apartment
Token apartment
Feature activation+0.000
:
Token:
Feature activation+0.027
Detroit
Token Detroit
Feature activation-0.027
Free
Token Free
Feature activation-0.019
Press
Token Press
Feature activation+0.010
)
Token)
Feature activation+0.161
Buy
TokenBuy
Feature activation+7.407
Photo
Token Photo
Feature activation-1.325
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
21
Token 21
Feature activation+0.000
J
Token J
Feature activation-0.030
OUR
TokenOUR
Feature activation-0.004
N
TokenN
Feature activation+0.005
AL
TokenAL
Feature activation-0.002
)
Token)
Feature activation+0.164
Buy
TokenBuy
Feature activation+7.276
Photo
Token Photo
Feature activation+0.084
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jo
TokenJo
Feature activation+0.000
el
Tokenel
Feature activation+0.000
:
Token:
Feature activation-0.045
Colorado
Token Colorado
Feature activation-0.014
an
Tokenan
Feature activation-0.015
library
Token library
Feature activation-0.012
)
Token)
Feature activation+0.109
Buy
TokenBuy
Feature activation+7.977
Photo
Token Photo
Feature activation-0.204
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
There
TokenThere
Feature activation+0.000
are
Token are
Feature activation+0.000
/
Token/
Feature activation-0.041
Ad
TokenAd
Feature activation+0.012
vertis
Tokenvertis
Feature activation-0.014
er
Tokener
Feature activation-0.006
)
Token)
Feature activation+0.159
Buy
TokenBuy
Feature activation+8.182
Photo
Token Photo
Feature activation-1.035
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Develop
TokenDevelop
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
B
TokenB
Feature activation+0.013
UNE
TokenUNE
Feature activation-0.009
FILE
Token FILE
Feature activation+0.027
PHOTO
Token PHOTO
Feature activation+0.030
)
Token)
Feature activation+0.235
Buy
TokenBuy
Feature activation+7.260
Photo
Token Photo
Feature activation-0.452
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
For
TokenFor
Feature activation+0.000
two
Token two
Feature activation+0.000
/
Token/
Feature activation-0.013
Journal
TokenJournal
Feature activation+0.012
&
Token &
Feature activation-0.011
Courier
Token Courier
Feature activation+0.001
)
Token )
Feature activation+0.173
Buy
TokenBuy
Feature activation+7.866
Photo
Token Photo
Feature activation-0.795
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
C
TokenC
Feature activation+0.000
ody
Tokenody
Feature activation+0.000
pn
Tokenpn
Feature activation-0.011
j
Tokenj
Feature activation+0.001
.
Token.
Feature activation-0.001
com
Tokencom
Feature activation-0.028
)
Token)
Feature activation+0.205
Buy
TokenBuy
Feature activation+7.258
Photo
Token Photo
Feature activation-1.486
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
1974
Token 1974
Feature activation+0.000
The
Token The
Feature activation-0.007
Tenn
Token Tenn
Feature activation-0.025
esse
Tokenesse
Feature activation-0.008
an
Tokenan
Feature activation-0.003
)
Token)
Feature activation+0.220
Buy
TokenBuy
Feature activation+7.613
Photo
Token Photo
Feature activation+0.220
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jason
TokenJason
Feature activation+0.000
Is
Token Is
Feature activation+0.000
ion
Tokenion
Feature activation+0.041
-
Token-
Feature activation-0.000
Led
TokenLed
Feature activation-0.009
ger
Tokenger
Feature activation-0.001
)
Token)
Feature activation+0.309
Buy
TokenBuy
Feature activation+7.744
Photo
Token Photo
Feature activation-0.064
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Authorities
TokenAuthorities
Feature activation+0.000
say
Token say
Feature activation+0.000
:
Token:
Feature activation-0.031
The
Token The
Feature activation-0.053
Desert
Token Desert
Feature activation-0.020
Sun
Token Sun
Feature activation-0.009
)
Token)
Feature activation+0.179
Buy
TokenBuy
Feature activation+7.828
Photo
Token Photo
Feature activation-0.100
Story
Token Story
Feature activation+0.000
Highlights
Token Highlights
Feature activation+0.000
Police
Token Police
Feature activation+0.000
shooting
Token shooting
Feature activation+0.000
ye
Tokenye
Feature activation+0.001
/
Token /
Feature activation+0.038
The
Token The
Feature activation+0.002
Star
Token Star
Feature activation-0.018
)
Token)
Feature activation+0.181
Buy
TokenBuy
Feature activation+8.134
Photo
Token Photo
Feature activation+0.134
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Z
Token Z
Feature activation+0.000
/
Token/
Feature activation+0.004
Chris
TokenChris
Feature activation+0.064
May
Token May
Feature activation-0.019
hew
Tokenhew
Feature activation-0.022
)
Token)
Feature activation+0.225
Buy
TokenBuy
Feature activation+7.912
Photo
Token Photo
Feature activation+0.027
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
CR
TokenCR
Feature activation+0.000
ES
TokenES
Feature activation+0.000
,
Token,
Feature activation+0.012
Milwaukee
Token Milwaukee
Feature activation+0.011
Journal
Token Journal
Feature activation-0.012
Sentinel
Token Sentinel
Feature activation+0.001
)
Token)
Feature activation+0.188
Buy
TokenBuy
Feature activation+6.979
Photo
Token Photo
Feature activation+0.222
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
W
TokenW
Feature activation+0.000
au
Tokenau
Feature activation+0.000
,
Token,
Feature activation+0.003
Detroit
Token Detroit
Feature activation+0.016
Free
Token Free
Feature activation-0.009
Press
Token Press
Feature activation+0.007
)
Token)
Feature activation+0.195
Buy
TokenBuy
Feature activation+6.891
Photo
Token Photo
Feature activation+0.391
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
From
TokenFrom
Feature activation+0.000
a
Token a
Feature activation+0.000
,
Token,
Feature activation+0.000
States
Token States
Feature activation+0.024
man
Tokenman
Feature activation-0.013
Journal
Token Journal
Feature activation-0.009
)
Token)
Feature activation+0.067
Buy
TokenBuy
Feature activation+7.935
Photo
Token Photo
Feature activation-0.881
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Oregon
TokenOregon
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
/
Token/
Feature activation-0.004
The
Token The
Feature activation-0.010
Desert
Token Desert
Feature activation-0.029
Sun
Token Sun
Feature activation-0.015
)
Token)
Feature activation+0.258
Buy
TokenBuy
Feature activation+6.588
Photo
Token Photo
Feature activation-0.007
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
On
TokenOn
Feature activation+0.000
a
Token a
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.06

Head 2: 0.04

Head 3: 0.10

Head 4: 0.06

Head 5: 0.04

Head 6: 0.27

Head 7: 0.10

Head 8: 0.06

Head 9: 0.07

Head 10: 0.07

Head 11: 0.12

Positive logits

BuyableInstoreAndOnline1.57

emouth1.57

1.48

minded1.45

EStream1.41

guiName1.29

DragonMagazine1.26

WARD1.25

listener1.24

vertisement1.19

endra1.19

ipeg1.18

enture1.17

zeb1.17

Yourself1.16

forwarded1.16

theless1.16

DonaldTrump1.16

Ability1.14

whiff1.13

Negative logits

inals-1.41

cule-1.40

ּ-1.33

364-1.31

amput-1.30

issance-1.25

gery-1.21

hani-1.18

ographic-1.17

iom-1.16

pac-1.15

amoto-1.14

tro-1.14

-1.13

itely-1.10

geries-1.10

iterranean-1.09

psc-1.08

atl-1.08

capt-1.08

INTERVAL 5.794 - 6.438
CONTAINS 0.000%

:
Token:
Feature activation+0.000
Indianapolis
Token Indianapolis
Feature activation+0.000
Star
Token Star
Feature activation+0.000
)
Token )
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.833
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Gro
TokenGro
Feature activation+0.000
an
Tokenan
Feature activation+0.000
.
Token.
Feature activation+0.000
Ind
TokenInd
Feature activation+0.000
y
Tokeny
Feature activation+0.000
Star
TokenStar
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.989
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Indiana
Token Indiana
Feature activation+0.000
Republican
Token Republican
Feature activation+0.000

INTERVAL 5.150 - 5.794
CONTAINS 0.000%

Ad
TokenAd
Feature activation+0.000
vertis
Tokenvertis
Feature activation+0.000
er
Tokener
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.213
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Develop
TokenDevelop
Feature activation+0.000
ers
Tokeners
Feature activation+0.000
plan
Token plan
Feature activation+0.000
UNE
TokenUNE
Feature activation+0.000
FILE
Token FILE
Feature activation+0.000
PHOTO
Token PHOTO
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.182
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
For
TokenFor
Feature activation+0.000
two
Token two
Feature activation+0.000
years
Token years
Feature activation+0.000
Detroit
Token Detroit
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.703
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
With
TokenWith
Feature activation+0.000
apartment
Token apartment
Feature activation+0.000
rental
Token rental
Feature activation+0.000
Colorado
Token Colorado
Feature activation+0.000
an
Tokenan
Feature activation+0.000
library
Token library
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.226
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
There
TokenThere
Feature activation+0.000
are
Token are
Feature activation+0.000
still
Token still
Feature activation+0.000
Detroit
Token Detroit
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.323
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+0.000
21
Token 21
Feature activation+0.000
-
Token-
Feature activation+0.000

INTERVAL 4.506 - 5.150
CONTAINS 0.000%

Journal
TokenJournal
Feature activation+0.000
&
Token &
Feature activation+0.000
Courier
Token Courier
Feature activation+0.000
)
Token )
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+5.108
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
C
TokenC
Feature activation+0.000
ody
Tokenody
Feature activation+0.000
Cousins
Token Cousins
Feature activation+0.000
/
Token /
Feature activation+0.000
The
Token The
Feature activation+0.000
Star
Token Star
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.767
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Z
Token Z
Feature activation+0.000
ions
Tokenions
Feature activation+0.000
-
Token-
Feature activation+0.000
Led
TokenLed
Feature activation+0.000
ger
Tokenger
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.962
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Authorities
TokenAuthorities
Feature activation+0.000
say
Token say
Feature activation+0.000
the
Token the
Feature activation+0.000
Tenn
Token Tenn
Feature activation+0.000
esse
Tokenesse
Feature activation+0.000
an
Tokenan
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.998
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Jason
TokenJason
Feature activation+0.000
Is
Token Is
Feature activation+0.000
bell
Tokenbell
Feature activation+0.000
Chris
TokenChris
Feature activation+0.000
May
Token May
Feature activation+0.000
hew
Tokenhew
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.762
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
CR
TokenCR
Feature activation+0.000
ES
TokenES
Feature activation+0.000
CENT
TokenCENT
Feature activation+0.000

INTERVAL 3.863 - 4.506
CONTAINS 0.000%

Detroit
Token Detroit
Feature activation+0.000
Free
Token Free
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.468
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
From
TokenFrom
Feature activation+0.000
a
Token a
Feature activation+0.000
tin
Token tin
Feature activation+0.000
States
Token States
Feature activation+0.000
man
Tokenman
Feature activation+0.000
Journal
Token Journal
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.429
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Oregon
TokenOregon
Feature activation+0.000
Republicans
Token Republicans
Feature activation+0.000
took
Token took
Feature activation+0.000
OR
TokenOR
Feature activation+0.000
IDA
TokenIDA
Feature activation+0.000
TODAY
Token TODAY
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+3.906
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Update
TokenUpdate
Feature activation+0.000
,
Token,
Feature activation+0.000
4
Token 4
Feature activation+0.000
The
Token The
Feature activation+0.000
Desert
Token Desert
Feature activation+0.000
Sun
Token Sun
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+3.925
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
On
TokenOn
Feature activation+0.000
a
Token a
Feature activation+0.000
recent
Token recent
Feature activation+0.000

INTERVAL 3.219 - 3.863
CONTAINS 0.000%

Desert
Token Desert
Feature activation+0.000
Sun
Token Sun
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+4.942
Story
Token Story
Feature activation+3.266
Highlights
Token Highlights
Feature activation+0.000
Police
Token Police
Feature activation+0.000
shooting
Token shooting
Feature activation+0.000
case
Token case
Feature activation+0.000
files
Token files
Feature activation+0.000

INTERVAL 2.575 - 3.219
CONTAINS 0.000%

criminal
Token criminal
Feature activation+0.000
history
Token history
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.921
Attorney
Token Attorney
Feature activation+0.000
Mark
Token Mark
Feature activation+0.000
May
Token May
Feature activation+0.000
field
Tokenfield
Feature activation+0.000
listens
Token listens
Feature activation+0.000
och
Tokenoch
Feature activation+0.000
it
Tokenit
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.649
Sun
Token Sun
Feature activation+0.000
cast
Token cast
Feature activation+0.000
early
Token early
Feature activation+0.000
morning
Token morning
Feature activation+0.000
light
Token light
Feature activation+0.000
opinions
Token opinions
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+1.389
Photo
Token Photo
Feature activation+2.839
of
Token of
Feature activation+0.000
Yon
Token Yon
Feature activation+0.000
kers
Tokenkers
Feature activation+0.000
police
Token police
Feature activation+0.000
badge
Token badge
Feature activation+0.000
The
Token The
Feature activation+0.000
Star
Token Star
Feature activation+0.000
Press
Token Press
Feature activation+0.000
)
Token)
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.735
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
M
TokenM
Feature activation+0.000
UN
TokenUN
Feature activation+0.000
C
TokenC
Feature activation+0.000
oj
Tokenoj
Feature activation+0.000
azer
Tokenazer
Feature activation+0.000
/
Token/
Feature activation+0.000
Reuters
TokenReuters
Feature activation+0.000
Buy
Token Buy
Feature activation+0.000
Photo
Token Photo
Feature activation+3.108
Wait
Token Wait
Feature activation+0.000
1
Token 1
Feature activation+0.000
second
Token second
Feature activation+0.000
to
Token to
Feature activation+0.000
continue
Token continue
Feature activation+0.000

INTERVAL 1.931 - 2.575
CONTAINS 0.000%

/
Token/
Feature activation+0.000
GET
TokenGET
Feature activation+0.000
TY
TokenTY
Feature activation+0.000
IMAGES
Token IMAGES
Feature activation+0.000
Buy
Token Buy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.082
Wait
Token Wait
Feature activation+0.000
1
Token 1
Feature activation+0.000
second
Token second
Feature activation+0.000
to
Token to
Feature activation+0.000
continue
Token continue
Feature activation+0.000
St
Token St
Feature activation+0.000
ice
Tokenice
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.414
You
Token You
Feature activation+0.000
can
Token can
Feature activation+0.000
cut
Token cut
Feature activation+0.000
your
Token your
Feature activation+0.000
Christmas
Token Christmas
Feature activation+0.000
May
Token May
Feature activation+0.000
hew
Tokenhew
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.347
Howard
Token Howard
Feature activation+0.000
Berry
Token Berry
Feature activation+0.000
set
Token set
Feature activation+0.000
a
Token a
Feature activation+0.000
sign
Token sign
Feature activation+0.000
2015
Token 2015
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.352
E
Token E
Feature activation+0.000
ber
Tokenber
Feature activation+0.000
Road
Token Road
Feature activation+0.000
in
Token in
Feature activation+0.000
West
Token West
Feature activation+0.000
Susan
Token Susan
Feature activation+0.000
Brooks
Token Brooks
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+2.064
Eric
Token Eric
Feature activation+0.000
Hol
Token Hol
Feature activation+0.000
comb
Tokencomb
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000

INTERVAL 1.288 - 1.931
CONTAINS 0.000%

and
Token and
Feature activation+0.000
opinions
Token opinions
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+1.389
Photo
Token Photo
Feature activation+2.839
of
Token of
Feature activation+0.000
Yon
Token Yon
Feature activation+0.000
kers
Tokenkers
Feature activation+0.000
police
Token police
Feature activation+0.000
About
TokenAbout
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
DR
TokenDR
Feature activation+0.000
OP
TokenOP
Feature activation+0.000
PI
TokenPI
Feature activation+1.348
is
Token is
Feature activation+0.000
perfect
Token perfect
Feature activation+0.000
for
Token for
Feature activation+0.000
YOU
Token YOU
Feature activation+0.000
if
Token if
Feature activation+0.000
game
Token game
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Tickets
Token Tickets
Feature activation+1.705
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Once
TokenOnce
Feature activation+0.000
Tw
TokenTw
Feature activation+0.000
elve
Tokenelve
Feature activation+0.000
But
Token But
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+1.418
Cody
Token Cody
Feature activation+0.000
Cousins
Token Cousins
Feature activation+0.000
will
Token will
Feature activation+0.000
appear
Token appear
Feature activation+0.000
Thursday
Token Thursday
Feature activation+0.000
Desert
Token Desert
Feature activation+0.000
Sun
Token Sun
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+1.340
Sun
Token Sun
Feature activation+0.000
Grow
Token Grow
Feature activation+0.000
employee
Token employee
Feature activation+0.000
Diego
Token Diego
Feature activation+0.000
Camb
Token Camb
Feature activation+0.000

INTERVAL 0.644 - 1.288
CONTAINS 0.000%

the
Token the
Feature activation+0.000
region
Token region
Feature activation+0.000
."
Token."
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+1.236
Civil
Token Civil
Feature activation+0.000
rights
Token rights
Feature activation+0.000
leader
Token leader
Feature activation+0.000
Rev
Token Rev
Feature activation+0.000
.
Token.
Feature activation+0.000
opened
Token opened
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+0.876
Sun
Token Sun
Feature activation+0.000
Grow
Token Grow
Feature activation+0.000
employee
Token employee
Feature activation+0.000
Diego
Token Diego
Feature activation+0.000
Camb
Token Camb
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+0.978
Grace
Token Grace
Feature activation+0.000
Martinez
Token Martinez
Feature activation+0.000
's
Token's
Feature activation+0.000
sister
Token sister
Feature activation+0.000
,
Token,
Feature activation+0.000
standards
Token standards
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+0.669
Pl
Token Pl
Feature activation+0.000
umbing
Tokenumbing
Feature activation+0.000
and
Token and
Feature activation+0.000
heating
Token heating
Feature activation+0.000
expert
Token expert
Feature activation+0.000
Gonzalez
Token Gonzalez
Feature activation+0.000
,
Token,
Feature activation+0.000
The
Token The
Feature activation+0.000
Chronicle
Token Chronicle
Feature activation+0.000
Buy
Token Buy
Feature activation+0.000
photo
Token photo
Feature activation+0.746
Photo
Token Photo
Feature activation+0.000
:
Token:
Feature activation+0.000
Carlos
Token Carlos
Feature activation+0.000
A
Token A
Feature activation+0.000
vil
Tokenvil
Feature activation+0.000

INTERVAL 0.000 - 0.644
CONTAINS 99.999%

and
Token and
Feature activation+0.000
Random
Token Random
Feature activation+0.000
H
Token H
Feature activation+0.000
acks
Tokenacks
Feature activation+0.000
of
Token of
Feature activation+0.000
Kind
Token Kind
Feature activation+0.000
ness
Tokenness
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
is
Token is
Feature activation+0.000
being
Token being
Feature activation+0.000
the
Token the
Feature activation+0.000
ESP
Token ESP
Feature activation+0.000
8
Token8
Feature activation+0.000
266
Token266
Feature activation+0.000
development
Token development
Feature activation+0.000
scene
Token scene
Feature activation+0.000
would
Token would
Feature activation+0.000
expect
Token expect
Feature activation+0.000
:
Token:
Feature activation+0.000
WiFi
Token WiFi
Feature activation+0.000
,
Token,
Feature activation+0.000
police
Token police
Feature activation+0.000
search
Token search
Feature activation+0.000
warrant
Token warrant
Feature activation+0.000
.
Token.
Feature activation+0.000
Jesse
Token Jesse
Feature activation+0.000
Smith
Token Smith
Feature activation+0.000
,
Token,
Feature activation+0.000
24
Token 24
Feature activation+0.000
,
Token,
Feature activation+0.000
of
Token of
Feature activation+0.000
Dartmouth
Token Dartmouth
Feature activation+0.000
conserv
Token conserv
Feature activation+0.000
ed
Tokened
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
all
Token all
Feature activation+0.000
elements
Token elements
Feature activation+0.000
beautifully
Token beautifully
Feature activation+0.000
balanced
Token balanced
Feature activation+0.000
to
Token to
Feature activation+0.000
achieve
Token achieve
Feature activation+0.000
visual
Token visual
Feature activation+0.000
to
Token to
Feature activation+0.000
his
Token his
Feature activation+0.000
co
Token co
Feature activation+0.000
-
Token-
Feature activation+0.000
authors
Tokenauthors
Feature activation+0.000
describing
Token describing
Feature activation+0.000
very
Token very
Feature activation+0.000
briefly
Token briefly
Feature activation+0.000
and
Token and
Feature activation+0.000
very
Token very
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 17: Dead

TOP ACTIVATIONS
MAX = 0.306

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.271

Jays
Token Jays
Feature activation-0.015
âĢ
TokenâĢ
Feature activation+0.012
Ļ
TokenĻ
Feature activation+0.012
baseball
Token baseball
Feature activation-0.044
cap
Token cap
Feature activation-0.091
and
Token and
Feature activation+0.162
a
Token a
Feature activation-0.233
beer
Token beer
Feature activation-0.138
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Jays
Token Jays
Feature activation-0.014
âĢ
TokenâĢ
Feature activation+0.031
Ļ
TokenĻ
Feature activation+0.015
baseball
Token baseball
Feature activation-0.128
cap
Token cap
Feature activation-0.019
and
Token and
Feature activation+0.245
a
Token a
Feature activation-0.146
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
Put
Token Put
Feature activation+0.033
a
Token a
Feature activation-0.023
Toronto
Token Toronto
Feature activation+0.025
Blue
Token Blue
Feature activation-0.014
Jays
Token Jays
Feature activation-0.053
âĢ
TokenâĢ
Feature activation+0.112
Ļ
TokenĻ
Feature activation+0.044
baseball
Token baseball
Feature activation-0.252
cap
Token cap
Feature activation+0.013
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Blue
Token Blue
Feature activation-0.013
Jays
Token Jays
Feature activation-0.016
âĢ
TokenâĢ
Feature activation+0.034
Ļ
TokenĻ
Feature activation+0.025
baseball
Token baseball
Feature activation-0.115
cap
Token cap
Feature activation+0.057
and
Token and
Feature activation+0.008
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.321
,
Token,
Feature activation-0.284
the
Token the
Feature activation-0.197
NB
Token NB
Feature activation-0.035
Space
Token Space
Feature activation+0.020
Race
Token Race
Feature activation-0.319
had
Token had
Feature activation-0.089
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.321
,
Token,
Feature activation-0.170
the
Token the
Feature activation-0.155
NB
Token NB
Feature activation-0.041
Space
Token Space
Feature activation+0.005
Race
Token Race
Feature activation-0.181
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.354
,
Token,
Feature activation-0.052
the
Token the
Feature activation-0.145
NB
Token NB
Feature activation+0.005
Space
Token Space
Feature activation+0.046
Race
Token Race
Feature activation-0.188
had
Token had
Feature activation-0.149
a
Token a
Feature activation+0.044
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
NB
Token NB
Feature activation+0.007
Space
Token Space
Feature activation-0.004
Race
Token Race
Feature activation-0.011
had
Token had
Feature activation-0.028
a
Token a
Feature activation-0.048
somewhat
Token somewhat
Feature activation+0.011
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
lofty
Token lofty
Feature activation+0.007
goal
Token goal
Feature activation-0.004
:
Token:
Feature activation+0.005
Put
Token Put
Feature activation+0.015
a
Token a
Feature activation-0.036
Toronto
Token Toronto
Feature activation+0.034
Blue
Token Blue
Feature activation-0.013
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
lofty
Token lofty
Feature activation+0.009
goal
Token goal
Feature activation-0.010
:
Token:
Feature activation+0.018
Put
Token Put
Feature activation+0.071
a
Token a
Feature activation-0.097
Toronto
Token Toronto
Feature activation+0.080
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
a
Token a
Feature activation-0.040
somewhat
Token somewhat
Feature activation+0.005
less
Token less
Feature activation-0.003
lofty
Token lofty
Feature activation+0.037
goal
Token goal
Feature activation-0.076
:
Token:
Feature activation+0.072
Put
Token Put
Feature activation+0.027
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
less
Token less
Feature activation-0.025
lofty
Token lofty
Feature activation-0.006
goal
Token goal
Feature activation-0.043
:
Token:
Feature activation+0.054
Put
Token Put
Feature activation+0.064
a
Token a
Feature activation+0.099
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.001
less
Token less
Feature activation-0.002
lofty
Token lofty
Feature activation-0.003
goal
Token goal
Feature activation-0.005
:
Token:
Feature activation+0.008
Put
Token Put
Feature activation+0.032
a
Token a
Feature activation-0.013
Toronto
Token Toronto
Feature activation-0.003
Blue
Token Blue
Feature activation-0.018
Jays
Token Jays
Feature activation-0.005
âĢ
TokenâĢ
Feature activation+0.019
lofty
Token lofty
Feature activation+0.013
goal
Token goal
Feature activation-0.012
:
Token:
Feature activation+0.018
Put
Token Put
Feature activation+0.046
a
Token a
Feature activation-0.057
Toronto
Token Toronto
Feature activation+0.234
Blue
Token Blue
Feature activation-0.181
Jays
Token Jays
Feature activation-0.139
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.002
less
Token less
Feature activation-0.001
lofty
Token lofty
Feature activation+0.006
goal
Token goal
Feature activation+0.003
:
Token:
Feature activation+0.004
Put
Token Put
Feature activation+0.071
a
Token a
Feature activation-0.015
Toronto
Token Toronto
Feature activation-0.013
Blue
Token Blue
Feature activation-0.010
Jays
Token Jays
Feature activation-0.020
âĢ
TokenâĢ
Feature activation-0.032
Put
Token Put
Feature activation+0.033
a
Token a
Feature activation-0.003
Toronto
Token Toronto
Feature activation+0.006
Blue
Token Blue
Feature activation-0.023
Jays
Token Jays
Feature activation-0.058
âĢ
TokenâĢ
Feature activation+0.271
Ļ
TokenĻ
Feature activation+0.017
baseball
Token baseball
Feature activation-0.064
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.396
,
Token,
Feature activation-0.255
the
Token the
Feature activation-0.048
NB
Token NB
Feature activation+0.007
Space
Token Space
Feature activation+0.024
Race
Token Race
Feature activation-0.085
had
Token had
Feature activation-0.064
a
Token a
Feature activation-0.032
somewhat
Token somewhat
Feature activation-0.021
less
Token less
Feature activation-0.116
Race
Token Race
Feature activation-0.201
had
Token had
Feature activation-0.141
a
Token a
Feature activation-0.140
somewhat
Token somewhat
Feature activation-0.066
less
Token less
Feature activation-0.136
lofty
Token lofty
Feature activation+0.059
goal
Token goal
Feature activation-0.118
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.196
,
Token,
Feature activation-0.083
the
Token the
Feature activation-0.102
NB
Token NB
Feature activation+0.006
Space
Token Space
Feature activation-0.006
Race
Token Race
Feature activation-0.013
had
Token had
Feature activation-0.002
a
Token a
Feature activation-0.050
somewhat
Token somewhat
Feature activation-0.003
Space
Token Space
Feature activation+0.011
Race
Token Race
Feature activation-0.130
had
Token had
Feature activation-0.086
a
Token a
Feature activation-0.075
somewhat
Token somewhat
Feature activation-0.168
less
Token less
Feature activation+0.020
lofty
Token lofty
Feature activation-0.118
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.11

Head 2: 0.08

Head 3: 0.07

Head 4: 0.09

Head 5: 0.07

Head 6: 0.08

Head 7: 0.10

Head 8: 0.07

Head 9: 0.08

Head 10: 0.10

Head 11: 0.08

Positive logits

cgi2.17

osponsors2.08

WithNo2.02

ser1.97

iversary1.94

oult1.83

ongs1.81

Amen1.80

myra1.80

rament1.80

spons1.79

oons1.79

PDATE1.78

inav1.74

cest1.73

guiActiveUn1.73

pring1.73

erest1.72

ransom1.71

ospel1.70

Negative logits

flow-1.78

observable-1.69

Hew-1.66

isible-1.57

Develop-1.55

enrol-1.55

discretionary-1.54

block-1.49

overhe-1.49

research-1.47

faulty-1.47

input-1.46

brainstorm-1.46

Rowe-1.45

scan-1.44

read-1.44

reacting-1.43

stem-1.43

inherit-1.42

longitudinal-1.42

INTERVAL 0.276 - 0.306
CONTAINS 0.000%

INTERVAL 0.245 - 0.276
CONTAINS 0.000%

INTERVAL 0.214 - 0.245
CONTAINS 0.000%

INTERVAL 0.184 - 0.214
CONTAINS 0.000%

INTERVAL 0.153 - 0.184
CONTAINS 0.000%

INTERVAL 0.123 - 0.153
CONTAINS 0.000%

INTERVAL 0.092 - 0.123
CONTAINS 0.000%

INTERVAL 0.061 - 0.092
CONTAINS 0.000%

INTERVAL 0.031 - 0.061
CONTAINS 0.000%

INTERVAL 0.000 - 0.031
CONTAINS 100.000%

one
Token one
Feature activation+0.000
filling
Token filling
Feature activation+0.000
that
Token that
Feature activation+0.000
makes
Token makes
Feature activation+0.000
a
Token a
Feature activation+0.000
tour
Token tour
Feature activation+0.000
t
Tokent
Feature activation+0.000
iere
Tokeniere
Feature activation+0.000
what
Token what
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
life
Token life
Feature activation+0.000
.,
Token.,
Feature activation+0.000
'
Token'
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
said
Token said
Feature activation+0.000
Mc
Token Mc
Feature activation+0.000
ener
Tokenener
Feature activation+0.000
ney
Tokenney
Feature activation+0.000
in
Token in
Feature activation+0.000
an
Token an
Feature activation+0.000
]
Token]
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
David
TokenDavid
Feature activation+0.000
Jeff
Token Jeff
Feature activation+0.000
ries
Tokenries
Feature activation+0.000
of
Token of
Feature activation+0.000
All
Token All
Feature activation+0.000
Music
TokenMusic
Feature activation+0.000
stated
Token stated
Feature activation+0.000
,
Token,
Feature activation+0.000
state
Token state
Feature activation+0.000
feder
Token feder
Feature activation+0.000
ations
Tokenations
Feature activation+0.000
,
Token,
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
9
Token 9
Feature activation+0.000
-
Token-
Feature activation+0.000
4
Token4
Feature activation+0.000
-
Token-
Feature activation+0.000
1
Token1
Feature activation+0.000
No
Token No
Feature activation+0.000
.
Token.
Feature activation+0.000
12
Token 12
Feature activation+0.000
-
Token-
Feature activation+0.000
014
Token014
Feature activation+0.000
22
Token22
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
More
TokenMore
Feature activation+0.000
business
Token business
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 18: Dead

TOP ACTIVATIONS
MAX = 0.078

industry
Token industry
Feature activation+0.000
has
Token has
Feature activation+0.000
known
Token known
Feature activation+0.000
for
Token for
Feature activation+0.000
decades
Token decades
Feature activation+0.000
:
Token:
Feature activation+0.078
that
Token that
Feature activation+0.000
wells
Token wells
Feature activation+0.000
leak
Token leak
Feature activation+0.000
and
Token and
Feature activation+0.000
leak
Token leak
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 1.324

was
Token was
Feature activation+0.017
just
Token just
Feature activation+0.044
another
Token another
Feature activation+0.045
confirmation
Token confirmation
Feature activation+0.663
of
Token of
Feature activation+0.069
what
Token what
Feature activation+1.324
industry
Token industry
Feature activation+0.088
has
Token has
Feature activation+0.149
known
Token known
Feature activation+0.105
for
Token for
Feature activation+0.858
decades
Token decades
Feature activation+0.021
Blue
Token Blue
Feature activation+0.005
Jays
Token Jays
Feature activation+0.011
âĢ
TokenâĢ
Feature activation-0.063
Ļ
TokenĻ
Feature activation-0.022
baseball
Token baseball
Feature activation+0.015
cap
Token cap
Feature activation+0.221
and
Token and
Feature activation+0.092
a
Token a
Feature activation-0.162
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Put
Token Put
Feature activation+0.007
a
Token a
Feature activation-0.014
Toronto
Token Toronto
Feature activation+0.019
Blue
Token Blue
Feature activation+0.014
Jays
Token Jays
Feature activation-0.012
âĢ
TokenâĢ
Feature activation+0.087
Ļ
TokenĻ
Feature activation+0.015
baseball
Token baseball
Feature activation-0.036
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Blue
Token Blue
Feature activation+0.022
Jays
Token Jays
Feature activation+0.015
âĢ
TokenâĢ
Feature activation-0.043
Ļ
TokenĻ
Feature activation-0.013
baseball
Token baseball
Feature activation+0.018
cap
Token cap
Feature activation+0.275
and
Token and
Feature activation-0.088
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
,
Token,
Feature activation-0.374
the
Token the
Feature activation-0.112
NB
Token NB
Feature activation+0.041
Space
Token Space
Feature activation-0.030
Race
Token Race
Feature activation+0.214
had
Token had
Feature activation+0.216
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.484
,
Token,
Feature activation-0.283
the
Token the
Feature activation-0.035
NB
Token NB
Feature activation+0.071
Space
Token Space
Feature activation-0.133
Race
Token Race
Feature activation+0.008
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
,
Token,
Feature activation-0.188
the
Token the
Feature activation-0.098
NB
Token NB
Feature activation+0.035
Space
Token Space
Feature activation-0.049
Race
Token Race
Feature activation+0.102
had
Token had
Feature activation+0.604
a
Token a
Feature activation+0.083
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.043
Blue
Token Blue
Feature activation+0.022
Jays
Token Jays
Feature activation+0.106
âĢ
TokenâĢ
Feature activation-0.077
Ļ
TokenĻ
Feature activation-0.020
baseball
Token baseball
Feature activation+0.243
cap
Token cap
Feature activation+0.141
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
lofty
Token lofty
Feature activation+0.029
goal
Token goal
Feature activation+0.036
:
Token:
Feature activation+0.043
Put
Token Put
Feature activation-0.013
a
Token a
Feature activation-0.055
Toronto
Token Toronto
Feature activation+0.220
Blue
Token Blue
Feature activation+0.097
Jays
Token Jays
Feature activation+0.058
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.015
less
Token less
Feature activation-0.029
lofty
Token lofty
Feature activation+0.021
goal
Token goal
Feature activation+0.054
:
Token:
Feature activation+0.084
Put
Token Put
Feature activation+0.200
a
Token a
Feature activation+0.196
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
had
Token had
Feature activation+0.131
a
Token a
Feature activation+0.008
somewhat
Token somewhat
Feature activation+0.001
less
Token less
Feature activation-0.038
lofty
Token lofty
Feature activation+0.141
goal
Token goal
Feature activation+1.129
:
Token:
Feature activation+0.017
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
had
Token had
Feature activation+0.020
a
Token a
Feature activation-0.010
somewhat
Token somewhat
Feature activation+0.014
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.145
goal
Token goal
Feature activation+0.281
:
Token:
Feature activation+0.086
Put
Token Put
Feature activation+0.066
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
had
Token had
Feature activation+0.009
a
Token a
Feature activation+0.001
somewhat
Token somewhat
Feature activation+0.004
less
Token less
Feature activation-0.001
lofty
Token lofty
Feature activation+0.013
goal
Token goal
Feature activation+0.045
:
Token:
Feature activation+0.025
Put
Token Put
Feature activation+0.015
a
Token a
Feature activation-0.013
Toronto
Token Toronto
Feature activation+0.019
Blue
Token Blue
Feature activation-0.015
lofty
Token lofty
Feature activation+0.012
goal
Token goal
Feature activation+0.010
:
Token:
Feature activation+0.012
Put
Token Put
Feature activation+0.010
a
Token a
Feature activation-0.036
Toronto
Token Toronto
Feature activation+0.348
Blue
Token Blue
Feature activation-0.024
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.005
less
Token less
Feature activation-0.000
lofty
Token lofty
Feature activation+0.018
goal
Token goal
Feature activation+0.021
:
Token:
Feature activation+0.013
Put
Token Put
Feature activation+0.068
a
Token a
Feature activation-0.022
Toronto
Token Toronto
Feature activation+0.003
Blue
Token Blue
Feature activation+0.014
Jays
Token Jays
Feature activation+0.008
âĢ
TokenâĢ
Feature activation-0.043
,
Token,
Feature activation-0.173
the
Token the
Feature activation-0.075
NB
Token NB
Feature activation+0.007
Space
Token Space
Feature activation-0.007
Race
Token Race
Feature activation+0.008
had
Token had
Feature activation+0.103
a
Token a
Feature activation-0.013
somewhat
Token somewhat
Feature activation+0.029
less
Token less
Feature activation-0.170
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
lofty
Token lofty
Feature activation+0.025
goal
Token goal
Feature activation+0.024
:
Token:
Feature activation+0.033
Put
Token Put
Feature activation-0.024
a
Token a
Feature activation-0.056
Toronto
Token Toronto
Feature activation+0.074
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
Race
Token Race
Feature activation+0.147
had
Token had
Feature activation+0.030
a
Token a
Feature activation-0.042
somewhat
Token somewhat
Feature activation-0.049
less
Token less
Feature activation-0.097
lofty
Token lofty
Feature activation+0.403
goal
Token goal
Feature activation+0.250
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
,
Token,
Feature activation-0.136
the
Token the
Feature activation-0.104
NB
Token NB
Feature activation+0.012
Space
Token Space
Feature activation-0.013
Race
Token Race
Feature activation+0.010
had
Token had
Feature activation+0.145
a
Token a
Feature activation-0.013
somewhat
Token somewhat
Feature activation+0.077
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
Race
Token Race
Feature activation+0.040
had
Token had
Feature activation+0.071
a
Token a
Feature activation-0.005
somewhat
Token somewhat
Feature activation-0.158
less
Token less
Feature activation+0.024
lofty
Token lofty
Feature activation+0.249
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.07

Head 2: 0.10

Head 3: 0.08

Head 4: 0.08

Head 5: 0.09

Head 6: 0.08

Head 7: 0.09

Head 8: 0.07

Head 9: 0.08

Head 10: 0.09

Head 11: 0.09

Positive logits

BlackBerry1.74

quartz1.66

4011.66

brainstorm1.63

subordinate1.59

numer1.59

Bohem1.58

contemplated1.55

cipher1.53

reorgan1.53

BMW1.53

semic1.53

Tesla1.53

manufacturing1.52

Warrant1.52

multit1.50

Bitcoin1.48

algorithms1.46

engineering1.44

composing1.44

Negative logits

GOODMAN-2.22

ARA-2.08

ENN-2.06

iris-2.02

arium-1.94

rers-1.88

FUL-1.85

atown-1.78

agues-1.78

inant-1.76

-1.69

eers-1.67

Editors-1.67

RIPT-1.65

Collins-1.64

Ble-1.63

FOX-1.63

ESA-1.62

ester-1.62

lash-1.61

INTERVAL 0.070 - 0.078
CONTAINS 0.000%

industry
Token industry
Feature activation+0.000
has
Token has
Feature activation+0.000
known
Token known
Feature activation+0.000
for
Token for
Feature activation+0.000
decades
Token decades
Feature activation+0.000
:
Token:
Feature activation+0.078
that
Token that
Feature activation+0.000
wells
Token wells
Feature activation+0.000
leak
Token leak
Feature activation+0.000
and
Token and
Feature activation+0.000
leak
Token leak
Feature activation+0.000

INTERVAL 0.062 - 0.070
CONTAINS 0.000%

INTERVAL 0.055 - 0.062
CONTAINS 0.000%

INTERVAL 0.047 - 0.055
CONTAINS 0.000%

INTERVAL 0.039 - 0.047
CONTAINS 0.000%

INTERVAL 0.031 - 0.039
CONTAINS 0.000%

INTERVAL 0.023 - 0.031
CONTAINS 0.000%

INTERVAL 0.016 - 0.023
CONTAINS 0.000%

INTERVAL 0.008 - 0.016
CONTAINS 0.000%

INTERVAL 0.000 - 0.008
CONTAINS 100.000%

then
Token then
Feature activation+0.000
take
Token take
Feature activation+0.000
a
Token a
Feature activation+0.000
brief
Token brief
Feature activation+0.000
trip
Token trip
Feature activation+0.000
to
Token to
Feature activation+0.000
Fen
Token Fen
Feature activation+0.000
way
Tokenway
Feature activation+0.000
Park
Token Park
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
this
Token this
Feature activation+0.000
:
Token:
Feature activation+0.000
Like
Token Like
Feature activation+0.000
Loading
Token Loading
Feature activation+0.000
...
Token...
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
C
TokenC
Feature activation+0.000
ategories
Tokenategories
Feature activation+0.000
:
Token:
Feature activation+0.000
Deaths
Token Deaths
Feature activation+0.000
Day
Token Day
Feature activation+0.000
editor
Token editor
Feature activation+0.000
Stephen
Token Stephen
Feature activation+0.000
Thompson
Token Thompson
Feature activation+0.000
about
Token about
Feature activation+0.000
some
Token some
Feature activation+0.000
the
Token the
Feature activation+0.000
albums
Token albums
Feature activation+0.000
they
Token they
Feature activation+0.000
're
Token're
Feature activation+0.000
most
Token most
Feature activation+0.000
supplement
Token supplement
Feature activation+0.000
ads
Token ads
Feature activation+0.000
in
Token in
Feature activation+0.000
health
Token health
Feature activation+0.000
food
Token food
Feature activation+0.000
stores
Token stores
Feature activation+0.000
and
Token and
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
Internet
Token Internet
Feature activation+0.000
.
Token.
Feature activation+0.000
and
Token and
Feature activation+0.000
other
Token other
Feature activation+0.000
therapies
Token therapies
Feature activation+0.000
;
Token;
Feature activation+0.000
supplements
Token supplements
Feature activation+0.000
and
Token and
Feature activation+0.000
drugs
Token drugs
Feature activation+0.000
;
Token;
Feature activation+0.000
and
Token and
Feature activation+0.000
assist
Token assist
Feature activation+0.000
ive
Tokenive
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 19: Follows “ cause”

TOP ACTIVATIONS
MAX = 3.477

together
Token together
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
good
Token good
Feature activation+0.000
cause
Token cause
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+3.477
a
Token a
Feature activation+0.000
special
Token special
Feature activation+0.000
American
Token American
Feature activation+0.000
Red
Token Red
Feature activation+0.000
Cross
Token Cross
Feature activation+0.000
claims
Token claims
Feature activation+0.000
to
Token to
Feature activation+0.000
champion
Token champion
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.940
a
Token a
Feature activation+2.431
domestic
Token domestic
Feature activation+0.000
help
Token help
Feature activation+0.000
S
Token S
Feature activation+0.000
ange
Tokenange
Feature activation+0.000
oused
Tokenoused
Feature activation+0.000
and
Token and
Feature activation+0.000
funded
Token funded
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.781
a
Token a
Feature activation+2.459
separate
Token separate
Feature activation+0.000
Sikh
Token Sikh
Feature activation+0.000
state
Token state
Feature activation+0.000
.
Token.
Feature activation+0.000
order
Token order
Feature activation+0.000
to
Token to
Feature activation+0.000
advance
Token advance
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.756
Islam
Token Islam
Feature activation+0.000
.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
is
Token is
Feature activation+0.000
worrisome
Token worrisome
Feature activation+0.000
that
Token that
Feature activation+0.000
would
Token would
Feature activation+0.000
advance
Token advance
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.683
equality
Token equality
Feature activation+0.000
for
Token for
Feature activation+0.000
millions
Token millions
Feature activation+0.000
of
Token of
Feature activation+0.000
Americans
Token Americans
Feature activation+0.000
tactic
Token tactic
Feature activation+0.000
for
Token for
Feature activation+0.000
advancing
Token advancing
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.660
animal
Token animal
Feature activation+0.000
rights
Token rights
Feature activation+0.000
is
Token is
Feature activation+0.000
as
Token as
Feature activation+0.000
follows
Token follows
Feature activation+0.000
and
Token and
Feature activation+0.000
funded
Token funded
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.781
a
Token a
Feature activation+2.459
separate
Token separate
Feature activation+0.000
Sikh
Token Sikh
Feature activation+0.000
state
Token state
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
to
Token to
Feature activation+0.000
achieve
Token achieve
Feature activation+0.000
the
Token the
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.439
getting
Token getting
Feature activation+0.000
the
Token the
Feature activation+0.000
latest
Token latest
Feature activation+0.000
Android
Token Android
Feature activation+0.000
onto
Token onto
Feature activation+0.000
to
Token to
Feature activation+0.000
champion
Token champion
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.940
a
Token a
Feature activation+2.431
domestic
Token domestic
Feature activation+0.000
help
Token help
Feature activation+0.000
S
Token S
Feature activation+0.000
ange
Tokenange
Feature activation+0.000
eta
Tokeneta
Feature activation+0.000
martyr
Token martyr
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.658
the
Token the
Feature activation+2.220
Cedar
Token Cedar
Feature activation+0.000
Revolution
Token Revolution
Feature activation+0.000
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
Exit
Token Exit
Feature activation+0.000
has
Token has
Feature activation+0.000
taken
Token taken
Feature activation+0.000
up
Token up
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.137
compensation
Token compensation
Feature activation+0.000
for
Token for
Feature activation+0.000
slavery
Token slavery
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
rally
Token rally
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.009
the
Token the
Feature activation+2.118
Syrian
Token Syrian
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.000
of
Token of
Feature activation+0.000
austerity
Token austerity
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.114
a
Token a
Feature activation+1.145
budget
Token budget
Feature activation+0.000
surplus
Token surplus
Feature activation+0.000
;
Token;
Feature activation+0.000
another
Token another
Feature activation+0.000
interests
Token interests
Feature activation+0.000
and
Token and
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.091
world
Token world
Feature activation+0.196
peace
Token peace
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
I
TokenI
Feature activation+0.000
press
Token press
Feature activation+0.000
to
Token to
Feature activation+0.000
promote
Token promote
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
for
Token for
Feature activation+2.015
a
Token a
Feature activation+0.160
Mental
Token Mental
Feature activation+0.000
Patients
Token Patients
Feature activation+0.000
Union
Token Union
Feature activation+0.000
.
Token.
Feature activation+0.000
to
Token to
Feature activation+0.000
rally
Token rally
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.009
the
Token the
Feature activation+2.118
Syrian
Token Syrian
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+0.000
crus
Token crus
Feature activation+0.000
ader
Tokenader
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.923
women
Token women
Feature activation+0.717
's
Token's
Feature activation+0.000
education
Token education
Feature activation+0.000
.
Token.
Feature activation+0.000
Mc
Token Mc
Feature activation+0.000
rake
Token rake
Feature activation+0.000
up
Token up
Feature activation+0.000
the
Token the
Feature activation+0.000
lost
Token lost
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.880
Khal
Token Khal
Feature activation+0.000
istan
Tokenistan
Feature activation+0.000
and
Token and
Feature activation+0.000
used
Token used
Feature activation+0.000
that
Token that
Feature activation+0.000
wanting
Token wanting
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
with
Token with
Feature activation+1.818
the
Token the
Feature activation+1.797
hard
Token hard
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
in
Token in
Feature activation+0.000
Iran
Token Iran
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
with
Token with
Feature activation+1.818
the
Token the
Feature activation+1.797
hard
Token hard
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
in
Token in
Feature activation+0.000
Iran
Token Iran
Feature activation+0.000
,"
Token,"
Feature activation+0.000

Top DFA by src position
MAX = 7.453

coming
Token coming
Feature activation+0.073
together
Token together
Feature activation-0.042
for
Token for
Feature activation+0.151
a
Token a
Feature activation+0.166
good
Token good
Feature activation+0.062
cause
Token cause
Feature activation+7.453
âĢĵ
Token âĢĵ
Feature activation+0.585
a
Token a
Feature activation+0.000
special
Token special
Feature activation+0.000
American
Token American
Feature activation+0.000
Red
Token Red
Feature activation+0.000
which
Token which
Feature activation+0.129
claims
Token claims
Feature activation+0.181
to
Token to
Feature activation+0.069
champion
Token champion
Feature activation+0.353
the
Token the
Feature activation+0.510
cause
Token cause
Feature activation+4.985
of
Token of
Feature activation+1.022
a
Token a
Feature activation+0.000
domestic
Token domestic
Feature activation+0.000
help
Token help
Feature activation+0.000
S
Token S
Feature activation+0.000
esp
Token esp
Feature activation+0.004
oused
Tokenoused
Feature activation+0.134
and
Token and
Feature activation-0.136
funded
Token funded
Feature activation+0.080
the
Token the
Feature activation+0.092
cause
Token cause
Feature activation+6.020
of
Token of
Feature activation+1.435
a
Token a
Feature activation+0.000
separate
Token separate
Feature activation+0.000
Sikh
Token Sikh
Feature activation+0.000
state
Token state
Feature activation+0.000
in
Token in
Feature activation-0.009
order
Token order
Feature activation-0.036
to
Token to
Feature activation-0.062
advance
Token advance
Feature activation+0.513
the
Token the
Feature activation+0.238
cause
Token cause
Feature activation+5.854
of
Token of
Feature activation+1.022
Islam
Token Islam
Feature activation+0.000
.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
is
Token is
Feature activation+0.000
legislation
Token legislation
Feature activation+0.066
that
Token that
Feature activation+0.218
would
Token would
Feature activation+0.150
advance
Token advance
Feature activation+0.361
the
Token the
Feature activation+0.202
cause
Token cause
Feature activation+4.998
of
Token of
Feature activation+1.269
equality
Token equality
Feature activation+0.000
for
Token for
Feature activation+0.000
millions
Token millions
Feature activation+0.000
of
Token of
Feature activation+0.000
ed
Tokened
Feature activation+0.007
tactic
Token tactic
Feature activation+0.095
for
Token for
Feature activation+0.146
advancing
Token advancing
Feature activation+0.535
the
Token the
Feature activation+0.536
cause
Token cause
Feature activation+4.655
of
Token of
Feature activation+1.058
animal
Token animal
Feature activation+0.000
rights
Token rights
Feature activation+0.000
is
Token is
Feature activation+0.000
as
Token as
Feature activation+0.000
esp
Token esp
Feature activation-0.007
oused
Tokenoused
Feature activation+0.096
and
Token and
Feature activation-0.115
funded
Token funded
Feature activation+0.091
the
Token the
Feature activation+0.097
cause
Token cause
Feature activation+4.827
of
Token of
Feature activation+2.514
a
Token a
Feature activation-0.005
separate
Token separate
Feature activation+0.000
Sikh
Token Sikh
Feature activation+0.000
state
Token state
Feature activation+0.000
together
Token together
Feature activation+0.075
to
Token to
Feature activation+0.014
achieve
Token achieve
Feature activation+0.092
the
Token the
Feature activation+0.160
common
Token common
Feature activation+0.537
cause
Token cause
Feature activation+5.286
of
Token of
Feature activation+1.061
getting
Token getting
Feature activation+0.000
the
Token the
Feature activation+0.000
latest
Token latest
Feature activation+0.000
Android
Token Android
Feature activation+0.000
which
Token which
Feature activation+0.095
claims
Token claims
Feature activation+0.198
to
Token to
Feature activation+0.042
champion
Token champion
Feature activation+0.305
the
Token the
Feature activation+0.250
cause
Token cause
Feature activation+3.818
of
Token of
Feature activation+1.900
a
Token a
Feature activation+0.184
domestic
Token domestic
Feature activation+0.000
help
Token help
Feature activation+0.000
S
Token S
Feature activation+0.000
a
Token a
Feature activation-0.067
minor
Token minor
Feature activation-0.041
martyr
Token martyr
Feature activation+0.326
to
Token to
Feature activation+0.020
the
Token the
Feature activation+0.176
cause
Token cause
Feature activation+5.445
of
Token of
Feature activation+1.363
the
Token the
Feature activation-0.065
Cedar
Token Cedar
Feature activation+0.000
Revolution
Token Revolution
Feature activation+0.000
here
Token here
Feature activation+0.000
,
Token,
Feature activation-0.012
has
Token has
Feature activation+0.107
taken
Token taken
Feature activation+0.151
up
Token up
Feature activation+0.068
the
Token the
Feature activation+0.332
cause
Token cause
Feature activation+5.061
of
Token of
Feature activation+1.109
compensation
Token compensation
Feature activation+0.000
for
Token for
Feature activation+0.000
slavery
Token slavery
Feature activation+0.000
and
Token and
Feature activation+0.000
beginning
Token beginning
Feature activation-0.059
to
Token to
Feature activation+0.005
rally
Token rally
Feature activation+0.049
to
Token to
Feature activation+0.010
the
Token the
Feature activation+0.048
cause
Token cause
Feature activation+5.140
of
Token of
Feature activation+1.694
the
Token the
Feature activation-0.087
Syrian
Token Syrian
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
and
Token and
Feature activation+0.000
years
Token years
Feature activation+0.190
of
Token of
Feature activation+0.034
austerity
Token austerity
Feature activation+0.093
in
Token in
Feature activation+0.173
the
Token the
Feature activation+0.371
cause
Token cause
Feature activation+5.154
of
Token of
Feature activation+0.743
a
Token a
Feature activation+0.000
budget
Token budget
Feature activation+0.000
surplus
Token surplus
Feature activation+0.000
;
Token;
Feature activation+0.000
own
Token own
Feature activation+0.034
interests
Token interests
Feature activation+0.159
and
Token and
Feature activation+0.046
in
Token in
Feature activation+0.051
the
Token the
Feature activation+0.296
cause
Token cause
Feature activation+4.756
of
Token of
Feature activation+1.287
world
Token world
Feature activation+0.000
peace
Token peace
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
the
Token the
Feature activation+0.039
press
Token press
Feature activation+0.031
to
Token to
Feature activation+0.018
promote
Token promote
Feature activation-0.008
the
Token the
Feature activation+0.228
cause
Token cause
Feature activation+5.637
for
Token for
Feature activation+0.662
a
Token a
Feature activation+0.000
Mental
Token Mental
Feature activation+0.000
Patients
Token Patients
Feature activation+0.000
Union
Token Union
Feature activation+0.000
beginning
Token beginning
Feature activation-0.057
to
Token to
Feature activation-0.012
rally
Token rally
Feature activation+0.074
to
Token to
Feature activation-0.003
the
Token the
Feature activation+0.378
cause
Token cause
Feature activation+5.036
of
Token of
Feature activation+1.080
the
Token the
Feature activation+0.000
Syrian
Token Syrian
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.086
crus
Token crus
Feature activation+0.234
ader
Tokenader
Feature activation+0.103
for
Token for
Feature activation+0.307
the
Token the
Feature activation+0.559
cause
Token cause
Feature activation+4.569
of
Token of
Feature activation+0.904
women
Token women
Feature activation+0.000
's
Token's
Feature activation+0.000
education
Token education
Feature activation+0.000
.
Token.
Feature activation+0.000
to
Token to
Feature activation+0.031
rake
Token rake
Feature activation+0.009
up
Token up
Feature activation+0.334
the
Token the
Feature activation+0.197
lost
Token lost
Feature activation+0.524
cause
Token cause
Feature activation+5.353
of
Token of
Feature activation+0.806
Khal
Token Khal
Feature activation+0.000
istan
Tokenistan
Feature activation+0.000
and
Token and
Feature activation+0.000
used
Token used
Feature activation+0.000
Congress
Token Congress
Feature activation-0.181
wanting
Token wanting
Feature activation+0.053
to
Token to
Feature activation-0.050
make
Token make
Feature activation+0.460
common
Token common
Feature activation+1.334
cause
Token cause
Feature activation+4.743
with
Token with
Feature activation+0.987
the
Token the
Feature activation+0.000
hard
Token hard
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
in
Token in
Feature activation+0.000
Congress
Token Congress
Feature activation-0.114
wanting
Token wanting
Feature activation+0.061
to
Token to
Feature activation-0.002
make
Token make
Feature activation+0.283
common
Token common
Feature activation+1.455
cause
Token cause
Feature activation+3.985
with
Token with
Feature activation+1.060
the
Token the
Feature activation+0.338
hard
Token hard
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
in
Token in
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.02

Head 2: 0.13

Head 3: 0.06

Head 4: 0.19

Head 5: 0.04

Head 6: 0.05

Head 7: 0.23

Head 8: 0.04

Head 9: 0.04

Head 10: 0.10

Head 11: 0.09

Positive logits

ibur1.56

rir1.49

cellence1.39

emetery1.37

dinand1.34

kefeller1.32

Churches1.32

alist1.27

Flag1.23

negie1.21

��1.21

ilan1.20

Artists1.20

Wiki1.19

bard1.19

exile1.19

asio1.19

Scholarship1.17

iatrics1.17

cair1.17

Negative logits

pores-1.44

REPL-1.32

script-1.25

skim-1.19

redundant-1.17

vo-1.16

instructions-1.14

notice-1.13

caveat-1.13

ּ-1.12

tid-1.10

cru-1.09

Warning-1.08

availability-1.08

Warning-1.08

elcome-1.07

Provided-1.07

password-1.06

Size-1.06

ories-1.03

INTERVAL 3.130 - 3.477
CONTAINS 0.000%

together
Token together
Feature activation+0.000
for
Token for
Feature activation+0.000
a
Token a
Feature activation+0.000
good
Token good
Feature activation+0.000
cause
Token cause
Feature activation+0.000
âĢĵ
Token âĢĵ
Feature activation+3.477
a
Token a
Feature activation+0.000
special
Token special
Feature activation+0.000
American
Token American
Feature activation+0.000
Red
Token Red
Feature activation+0.000
Cross
Token Cross
Feature activation+0.000

INTERVAL 2.782 - 3.130
CONTAINS 0.000%

claims
Token claims
Feature activation+0.000
to
Token to
Feature activation+0.000
champion
Token champion
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.940
a
Token a
Feature activation+2.431
domestic
Token domestic
Feature activation+0.000
help
Token help
Feature activation+0.000
S
Token S
Feature activation+0.000
ange
Tokenange
Feature activation+0.000

INTERVAL 2.434 - 2.782
CONTAINS 0.000%

and
Token and
Feature activation+0.000
funded
Token funded
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.781
a
Token a
Feature activation+2.459
separate
Token separate
Feature activation+0.000
Sikh
Token Sikh
Feature activation+0.000
state
Token state
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
oused
Tokenoused
Feature activation+0.000
and
Token and
Feature activation+0.000
funded
Token funded
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.781
a
Token a
Feature activation+2.459
separate
Token separate
Feature activation+0.000
Sikh
Token Sikh
Feature activation+0.000
state
Token state
Feature activation+0.000
.
Token.
Feature activation+0.000
tactic
Token tactic
Feature activation+0.000
for
Token for
Feature activation+0.000
advancing
Token advancing
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.660
animal
Token animal
Feature activation+0.000
rights
Token rights
Feature activation+0.000
is
Token is
Feature activation+0.000
as
Token as
Feature activation+0.000
follows
Token follows
Feature activation+0.000
that
Token that
Feature activation+0.000
would
Token would
Feature activation+0.000
advance
Token advance
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.683
equality
Token equality
Feature activation+0.000
for
Token for
Feature activation+0.000
millions
Token millions
Feature activation+0.000
of
Token of
Feature activation+0.000
Americans
Token Americans
Feature activation+0.000
order
Token order
Feature activation+0.000
to
Token to
Feature activation+0.000
advance
Token advance
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.756
Islam
Token Islam
Feature activation+0.000
.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
is
Token is
Feature activation+0.000
worrisome
Token worrisome
Feature activation+0.000

INTERVAL 2.086 - 2.434
CONTAINS 0.000%

to
Token to
Feature activation+0.000
champion
Token champion
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.940
a
Token a
Feature activation+2.431
domestic
Token domestic
Feature activation+0.000
help
Token help
Feature activation+0.000
S
Token S
Feature activation+0.000
ange
Tokenange
Feature activation+0.000
eta
Tokeneta
Feature activation+0.000
interests
Token interests
Feature activation+0.000
and
Token and
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.091
world
Token world
Feature activation+0.196
peace
Token peace
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
I
TokenI
Feature activation+0.000
rally
Token rally
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.009
the
Token the
Feature activation+2.118
Syrian
Token Syrian
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+0.000
in
Token in
Feature activation+0.000
martyr
Token martyr
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.658
the
Token the
Feature activation+2.220
Cedar
Token Cedar
Feature activation+0.000
Revolution
Token Revolution
Feature activation+0.000
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
Exit
Token Exit
Feature activation+0.000
of
Token of
Feature activation+0.000
austerity
Token austerity
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.114
a
Token a
Feature activation+1.145
budget
Token budget
Feature activation+0.000
surplus
Token surplus
Feature activation+0.000
;
Token;
Feature activation+0.000
another
Token another
Feature activation+0.000

INTERVAL 1.739 - 2.086
CONTAINS 0.000%

to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
with
Token with
Feature activation+1.818
the
Token the
Feature activation+1.797
hard
Token hard
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
in
Token in
Feature activation+0.000
Iran
Token Iran
Feature activation+0.000
,"
Token,"
Feature activation+0.000
crus
Token crus
Feature activation+0.000
ader
Tokenader
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.923
women
Token women
Feature activation+0.717
's
Token's
Feature activation+0.000
education
Token education
Feature activation+0.000
.
Token.
Feature activation+0.000
Mc
Token Mc
Feature activation+0.000
to
Token to
Feature activation+0.000
rally
Token rally
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.009
the
Token the
Feature activation+2.118
Syrian
Token Syrian
Feature activation+0.000
opposition
Token opposition
Feature activation+0.000
and
Token and
Feature activation+0.000
,
Token,
Feature activation+0.000
wanting
Token wanting
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
with
Token with
Feature activation+1.818
the
Token the
Feature activation+1.797
hard
Token hard
Feature activation+0.000
liners
Tokenliners
Feature activation+0.000
in
Token in
Feature activation+0.000
Iran
Token Iran
Feature activation+0.000
press
Token press
Feature activation+0.000
to
Token to
Feature activation+0.000
promote
Token promote
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
for
Token for
Feature activation+2.015
a
Token a
Feature activation+0.160
Mental
Token Mental
Feature activation+0.000
Patients
Token Patients
Feature activation+0.000
Union
Token Union
Feature activation+0.000
.
Token.
Feature activation+0.000

INTERVAL 1.391 - 1.739
CONTAINS 0.000%

blame
Token blame
Feature activation+0.000
him
Token him
Feature activation+0.000
for
Token for
Feature activation+0.000
helping
Token helping
Feature activation+0.000
cause
Token cause
Feature activation+0.000
the
Token the
Feature activation+1.583
housing
Token housing
Feature activation+0.000
crisis
Token crisis
Feature activation+0.000
and
Token and
Feature activation+0.000
overhe
Token overhe
Feature activation+0.000
ated
Tokenated
Feature activation+0.000
minor
Token minor
Feature activation+0.000
martyr
Token martyr
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.658
the
Token the
Feature activation+2.220
Cedar
Token Cedar
Feature activation+0.000
Revolution
Token Revolution
Feature activation+0.000
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
t
Tokent
Feature activation+0.000
have
Token have
Feature activation+0.000
any
Token any
Feature activation+0.000
genuine
Token genuine
Feature activation+0.000
cause
Token cause
Feature activation+0.000
to
Token to
Feature activation+1.404
disrupt
Token disrupt
Feature activation+0.000
parliament
Token parliament
Feature activation+0.000
.
Token.
Feature activation+0.000
Even
Token Even
Feature activation+0.000
as
Token as
Feature activation+0.000
unite
Token unite
Feature activation+0.000
youth
Token youth
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.495
Independence
Token Independence
Feature activation+0.069
.
Token.
Feature activation+0.000
Following
Token Following
Feature activation+0.000
excerpts
Token excerpts
Feature activation+0.000
are
Token are
Feature activation+0.000
issue
Token issue
Feature activation+0.000
of
Token of
Feature activation+0.000
making
Token making
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
with
Token with
Feature activation+1.551
nationalists
Token nationalists
Feature activation+0.000
,
Token,
Feature activation+0.000
which
Token which
Feature activation+0.000
reminded
Token reminded
Feature activation+0.000
me
Token me
Feature activation+0.000

INTERVAL 1.043 - 1.391
CONTAINS 0.000%

austerity
Token austerity
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+2.114
a
Token a
Feature activation+1.145
budget
Token budget
Feature activation+0.000
surplus
Token surplus
Feature activation+0.000
;
Token;
Feature activation+0.000
another
Token another
Feature activation+0.000
fuel
Token fuel
Feature activation+0.000
government
Token government
Feature activation+0.000
to
Token to
Feature activation+0.000
support
Token support
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.301
labeling
Token labeling
Feature activation+0.000
GMO
Token GMO
Feature activation+0.000
foods
Token foods
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
formal
Token formal
Feature activation+0.000
motion
Token motion
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.172
the
Token the
Feature activation+0.291
dip
Token dip
Feature activation+0.000
,
Token,
Feature activation+0.000
He
Token He
Feature activation+0.000
aly
Tokenaly
Feature activation+0.000
even
Token even
Feature activation+0.000
ren
Token ren
Feature activation+0.000
ounced
Tokenounced
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
for
Token for
Feature activation+1.047
which
Token which
Feature activation+0.000
he
Token he
Feature activation+0.000
'd
Token'd
Feature activation+0.000
fought
Token fought
Feature activation+0.000
.
Token.
Feature activation+0.000
anged
Tokenanged
Feature activation+0.000
"
Token"
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.090
Ram
Token Ram
Feature activation+0.000
temple
Token temple
Feature activation+0.000
.
Token.
Feature activation+0.000
When
TokenWhen
Feature activation+0.000
asked
Token asked
Feature activation+0.000

INTERVAL 0.695 - 1.043
CONTAINS 0.000%

Quebec
Token Quebec
Feature activation+0.000
government
Token government
Feature activation+0.000
supports
Token supports
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
and
Token and
Feature activation+0.712
promised
Token promised
Feature activation+0.000
a
Token a
Feature activation+0.000
loan
Token loan
Feature activation+0.000
of
Token of
Feature activation+0.000
$
Token $
Feature activation+0.000
put
Token put
Feature activation+0.000
behind
Token behind
Feature activation+0.000
us
Token us
Feature activation+0.000
those
Token those
Feature activation+0.000
causes
Token causes
Feature activation+0.000
of
Token of
Feature activation+0.795
division
Token division
Feature activation+0.000
and
Token and
Feature activation+0.000
discord
Token discord
Feature activation+0.000
that
Token that
Feature activation+0.000
have
Token have
Feature activation+0.000
anyone
Token anyone
Feature activation+0.000
who
Token who
Feature activation+0.000
champions
Token champions
Feature activation+0.000
a
Token a
Feature activation+0.000
cause
Token cause
Feature activation+0.000
that
Token that
Feature activation+1.002
's
Token's
Feature activation+0.000
historical
Token historical
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
ader
Tokenader
Feature activation+0.000
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+1.923
women
Token women
Feature activation+0.717
's
Token's
Feature activation+0.000
education
Token education
Feature activation+0.000
.
Token.
Feature activation+0.000
Mc
Token Mc
Feature activation+0.000
I
TokenI
Feature activation+0.000
to
Token to
Feature activation+0.000
make
Token make
Feature activation+0.000
common
Token common
Feature activation+0.000
cause
Token cause
Feature activation+0.000
with
Token with
Feature activation+0.989
the
Token the
Feature activation+1.011
law
Token law
Feature activation+0.000
-
Token-
Feature activation+0.000
abiding
Tokenabiding
Feature activation+0.000
against
Token against
Feature activation+0.000
criminals
Token criminals
Feature activation+0.000

INTERVAL 0.348 - 0.695
CONTAINS 0.000%

the
Token the
Feature activation+0.000
way
Token way
Feature activation+0.000
that
Token that
Feature activation+0.000
a
Token a
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+0.549
action
Token action
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
formulated
Token formulated
Feature activation+0.000
."
Token."
Feature activation+0.000
uss
Tokenuss
Feature activation+0.000
ing
Tokening
Feature activation+0.000
on
Token on
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
the
Token the
Feature activation+0.379
young
Token young
Feature activation+0.000
Indian
Token Indian
Feature activation+0.000
believes
Token believes
Feature activation+0.000
in
Token in
Feature activation+0.000
.
Token.
Feature activation+0.000
importance
Token importance
Feature activation+0.000
of
Token of
Feature activation+0.000
committing
Token committing
Feature activation+0.000
to
Token to
Feature activation+0.000
causes
Token causes
Feature activation+0.000
in
Token in
Feature activation+0.435
order
Token order
Feature activation+0.000
to
Token to
Feature activation+0.000
investigate
Token investigate
Feature activation+0.000
them
Token them
Feature activation+0.000
,
Token,
Feature activation+0.000
Sanders
Token Sanders
Feature activation+0.000
has
Token has
Feature activation+0.000
championed
Token championed
Feature activation+0.000
environmental
Token environmental
Feature activation+0.000
causes
Token causes
Feature activation+0.000
,
Token,
Feature activation+0.371
stands
Token stands
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
Wall
Token Wall
Feature activation+0.000
Street
Token Street
Feature activation+0.000
s
Tokens
Feature activation+0.000
brought
Token brought
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
cause
Token cause
Feature activation+0.000
of
Token of
Feature activation+0.499
peace
Token peace
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
Middle
Token Middle
Feature activation+0.000
East
Token East
Feature activation+0.000

INTERVAL 0.000 - 0.348
CONTAINS 99.999%

s
Tokens
Feature activation+0.000
second
Token second
Feature activation+0.000
-
Token-
Feature activation+0.000
choice
Tokenchoice
Feature activation+0.000
team
Token team
Feature activation+0.000
on
Token on
Feature activation+0.000
Tuesday
Token Tuesday
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
dismal
Token dismal
Feature activation+0.000
end
Token end
Feature activation+0.000
original
Token original
Feature activation+0.000
leader
Token leader
Feature activation+0.000
,
Token,
Feature activation+0.000
Night
Token Night
Feature activation+0.000
Th
Token Th
Feature activation+0.000
ras
Tokenras
Feature activation+0.000
her
Tokenher
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
rich
Token rich
Feature activation+0.000
kid
Token kid
Feature activation+0.000
aligned
Token aligned
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
2
Token2
Feature activation+0.000
Add
Token Add
Feature activation+0.000
a
Token a
Feature activation+0.000
Layer
Token Layer
Feature activation+0.000
Mask
Token Mask
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
plant
Token plant
Feature activation+0.000
went
Token went
Feature activation+0.000
online
Token online
Feature activation+0.000
,
Token,
Feature activation+0.000
pilots
Token pilots
Feature activation+0.000
complained
Token complained
Feature activation+0.000
repeatedly
Token repeatedly
Feature activation+0.000
of
Token of
Feature activation+0.000
being
Token being
Feature activation+0.000
blinded
Token blinded
Feature activation+0.000
by
Token by
Feature activation+0.000
change
Token change
Feature activation+0.000
over
Token over
Feature activation+0.000
the
Token the
Feature activation+0.000
next
Token next
Feature activation+0.000
four
Token four
Feature activation+0.000
to
Token to
Feature activation+0.000
eight
Token eight
Feature activation+0.000
years
Token years
Feature activation+0.000
.
Token.
Feature activation+0.000
Nothing
Token Nothing
Feature activation+0.000
stays
Token stays
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 20: In text about instructions for building something

TOP ACTIVATIONS
MAX = 2.392

bolt
Token bolt
Feature activation+1.523
.
Token.
Feature activation+1.279
Screw
Token Screw
Feature activation+1.401
the
Token the
Feature activation+1.907
third
Token third
Feature activation+1.619
nut
Token nut
Feature activation+2.392
onto
Token onto
Feature activation+1.438
the
Token the
Feature activation+1.278
bolt
Token bolt
Feature activation+1.891
until
Token until
Feature activation+1.024
it
Token it
Feature activation+1.437
bolt
Token bolt
Feature activation+1.495
.
Token.
Feature activation+1.335
Screw
Token Screw
Feature activation+1.255
the
Token the
Feature activation+1.640
second
Token second
Feature activation+1.580
nut
Token nut
Feature activation+2.217
against
Token against
Feature activation+0.852
the
Token the
Feature activation+1.141
bottom
Token bottom
Feature activation+1.169
of
Token of
Feature activation+0.753
the
Token the
Feature activation+0.727
of
Token of
Feature activation+1.258
the
Token the
Feature activation+1.334
per
Token per
Feature activation+1.234
for
Tokenfor
Feature activation+0.260
ated
Tokenated
Feature activation+1.559
board
Token board
Feature activation+2.162
.
Token.
Feature activation+1.920
This
Token This
Feature activation+1.710
was
Token was
Feature activation+1.242
not
Token not
Feature activation+1.027
part
Token part
Feature activation+0.997
Put
Token Put
Feature activation+0.848
one
Token one
Feature activation+1.008
piece
Token piece
Feature activation+1.045
of
Token of
Feature activation+1.187
each
Token each
Feature activation+0.861
thread
Token thread
Feature activation+2.098
(
Token (
Feature activation+1.154
2
Token2
Feature activation+1.212
pieces
Token pieces
Feature activation+0.923
per
Token per
Feature activation+0.402
slit
Token slit
Feature activation+0.368
Cut
TokenCut
Feature activation+1.463
as
Token as
Feature activation+1.283
close
Token close
Feature activation+1.790
to
Token to
Feature activation+1.483
the
Token the
Feature activation+1.640
wire
Token wire
Feature activation+1.984
as
Token as
Feature activation+1.833
possible
Token possible
Feature activation+1.046
.
Token.
Feature activation+1.786
Ċ
TokenĊ
Feature activation+1.812
Ċ
TokenĊ
Feature activation+1.671
bolt
Token bolt
Feature activation+1.135
through
Token through
Feature activation+1.086
the
Token the
Feature activation+1.299
g
Token g
Feature activation+0.804
rom
Tokenrom
Feature activation+0.926
met
Tokenmet
Feature activation+1.978
from
Token from
Feature activation+1.160
the
Token the
Feature activation+0.979
back
Token back
Feature activation+1.168
towards
Token towards
Feature activation+1.031
the
Token the
Feature activation+1.066
outline
Token outline
Feature activation+0.697
of
Token of
Feature activation+0.627
the
Token the
Feature activation+0.739
relay
Token relay
Feature activation+0.984
.
Token.
Feature activation+1.198
Cut
Token Cut
Feature activation+1.969
two
Token two
Feature activation+1.645
prototype
Token prototype
Feature activation+1.372
circuit
Token circuit
Feature activation+1.215
boards
Token boards
Feature activation+1.587
to
Token to
Feature activation+1.660
the
Token the
Feature activation+1.334
per
Token per
Feature activation+1.234
for
Tokenfor
Feature activation+0.260
ated
Tokenated
Feature activation+1.559
board
Token board
Feature activation+2.162
.
Token.
Feature activation+1.920
This
Token This
Feature activation+1.710
was
Token was
Feature activation+1.242
not
Token not
Feature activation+1.027
part
Token part
Feature activation+0.997
of
Token of
Feature activation+0.932
.
Token.
Feature activation+1.807
Screw
Token Screw
Feature activation+1.294
the
Token the
Feature activation+1.539
first
Token first
Feature activation+1.479
nut
Token nut
Feature activation+1.553
down
Token down
Feature activation+1.910
until
Token until
Feature activation+1.504
it
Token it
Feature activation+1.843
is
Token is
Feature activation+1.347
snug
Token snug
Feature activation+1.262
against
Token against
Feature activation+0.983
two
Token two
Feature activation+1.596
device
Token device
Feature activation+1.352
boxes
Token boxes
Feature activation+1.047
.
Token.
Feature activation+1.618
The
Token The
Feature activation+1.620
larger
Token larger
Feature activation+1.910
one
Token one
Feature activation+1.653
is
Token is
Feature activation+1.753
for
Token for
Feature activation+1.164
AC
Token AC
Feature activation+0.465
and
Token and
Feature activation+0.015
the
Token the
Feature activation+0.337
carriage
Token carriage
Feature activation+1.092
bolt
Token bolt
Feature activation+1.523
.
Token.
Feature activation+1.279
Screw
Token Screw
Feature activation+1.401
the
Token the
Feature activation+1.907
third
Token third
Feature activation+1.619
nut
Token nut
Feature activation+2.392
onto
Token onto
Feature activation+1.438
the
Token the
Feature activation+1.278
bolt
Token bolt
Feature activation+1.891
relay
Token relay
Feature activation+0.891
.
Token.
Feature activation+1.178
Put
Token Put
Feature activation+1.430
together
Token together
Feature activation+1.153
the
Token the
Feature activation+1.345
circuit
Token circuit
Feature activation+1.906
using
Token using
Feature activation+1.737
the
Token the
Feature activation+1.309
28
Token 28
Feature activation+0.689
-
Token-
Feature activation+0.407
pin
Tokenpin
Feature activation+1.131
circuit
Token circuit
Feature activation+0.375
board
Token board
Feature activation+0.261
and
Token and
Feature activation+0.984
make
Token make
Feature activation+1.041
cut
Token cut
Feature activation+1.492
marks
Token marks
Feature activation+1.905
around
Token around
Feature activation+1.375
the
Token the
Feature activation+1.040
outline
Token outline
Feature activation+0.697
of
Token of
Feature activation+0.627
the
Token the
Feature activation+0.739
the
Token the
Feature activation+1.907
third
Token third
Feature activation+1.619
nut
Token nut
Feature activation+2.392
onto
Token onto
Feature activation+1.438
the
Token the
Feature activation+1.278
bolt
Token bolt
Feature activation+1.891
until
Token until
Feature activation+1.024
it
Token it
Feature activation+1.437
is
Token is
Feature activation+1.091
flush
Token flush
Feature activation+0.762
with
Token with
Feature activation+0.218
-
Token-
Feature activation+0.366
tie
Tokentie
Feature activation+0.839
cut
Token cut
Feature activation+0.745
a
Token a
Feature activation+1.655
small
Token small
Feature activation+1.480
opening
Token opening
Feature activation+1.864
and
Token and
Feature activation+1.338
fast
Token fast
Feature activation+0.989
en
Tokenen
Feature activation+1.474
a
Token a
Feature activation+1.567
g
Token g
Feature activation+0.820
Ċ
TokenĊ
Feature activation+0.678
Ċ
TokenĊ
Feature activation+0.915
I
TokenI
Feature activation+0.648
cut
Token cut
Feature activation+1.622
openings
Token openings
Feature activation+1.796
for
Token for
Feature activation+1.855
the
Token the
Feature activation+1.395
two
Token two
Feature activation+1.596
device
Token device
Feature activation+1.352
boxes
Token boxes
Feature activation+1.047
.
Token.
Feature activation+1.618
the
Token the
Feature activation+1.539
first
Token first
Feature activation+1.479
nut
Token nut
Feature activation+1.553
down
Token down
Feature activation+1.910
until
Token until
Feature activation+1.504
it
Token it
Feature activation+1.843
is
Token is
Feature activation+1.347
snug
Token snug
Feature activation+1.262
against
Token against
Feature activation+0.983
the
Token the
Feature activation+1.144
top
Token top
Feature activation+1.045
as
Token as
Feature activation+1.283
close
Token close
Feature activation+1.790
to
Token to
Feature activation+1.483
the
Token the
Feature activation+1.640
wire
Token wire
Feature activation+1.984
as
Token as
Feature activation+1.833
possible
Token possible
Feature activation+1.046
.
Token.
Feature activation+1.786
Ċ
TokenĊ
Feature activation+1.812
Ċ
TokenĊ
Feature activation+1.671
That
TokenThat
Feature activation+1.149
Ħ
TokenĦ
Feature activation+0.116
2
Token2
Feature activation+0.000
â̳
Tokenâ̳
Feature activation+0.821
down
Token down
Feature activation+0.993
the
Token the
Feature activation+0.963
bolt
Token bolt
Feature activation+1.825
.
Token.
Feature activation+1.580
Place
Token Place
Feature activation+1.342
the
Token the
Feature activation+1.198
lid
Token lid
Feature activation+1.317
on
Token on
Feature activation+0.427
the
Token the
Feature activation+1.640
wire
Token wire
Feature activation+1.984
as
Token as
Feature activation+1.833
possible
Token possible
Feature activation+1.046
.
Token.
Feature activation+1.786
Ċ
TokenĊ
Feature activation+1.812
Ċ
TokenĊ
Feature activation+1.671
That
TokenThat
Feature activation+1.149
âĢ
TokenâĢ
Feature activation+0.495
Ļ
TokenĻ
Feature activation+0.524
s
Tokens
Feature activation+0.651

Top DFA by src position
MAX = 1.359

.
Token.
Feature activation+0.004
Screw
Token Screw
Feature activation+0.127
the
Token the
Feature activation+0.029
second
Token second
Feature activation+0.020
nut
Token nut
Feature activation+0.220
onto
Token onto
Feature activation+0.483
the
Token the
Feature activation+0.027
carriage
Token carriage
Feature activation+0.015
bolt
Token bolt
Feature activation+0.091
until
Token until
Feature activation+0.041
it
Token it
Feature activation+0.004
of
Token of
Feature activation+0.004
the
Token the
Feature activation+0.006
bolt
Token bolt
Feature activation+0.068
.
Token.
Feature activation+0.100
Screw
Token Screw
Feature activation+0.182
the
Token the
Feature activation+0.262
second
Token second
Feature activation+0.096
nut
Token nut
Feature activation+0.154
against
Token against
Feature activation+0.000
the
Token the
Feature activation+0.000
bottom
Token bottom
Feature activation+0.000
the
Token the
Feature activation+0.028
per
Token per
Feature activation+0.017
for
Tokenfor
Feature activation+0.004
ated
Tokenated
Feature activation+0.020
board
Token board
Feature activation+0.266
.
Token.
Feature activation+0.741
I
Token I
Feature activation+0.017
added
Token added
Feature activation+0.043
a
Token a
Feature activation+0.062
seven
Token seven
Feature activation+0.029
inches
Token inches
Feature activation+0.042
<|endoftext|>
Token<|endoftext|>
Feature activation+0.484
the
Token the
Feature activation-0.002
embro
Token embro
Feature activation+0.032
ider
Tokenider
Feature activation-0.002
y
Tokeny
Feature activation-0.011
thread
Token thread
Feature activation+0.084
Ċ
TokenĊ
Feature activation+0.014
Ċ
TokenĊ
Feature activation+0.012
Cut
TokenCut
Feature activation+0.289
as
Token as
Feature activation+0.121
close
Token close
Feature activation+0.233
to
Token to
Feature activation+0.404
the
Token the
Feature activation+0.094
wire
Token wire
Feature activation+0.191
as
Token as
Feature activation+0.000
possible
Token possible
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.395
off
Token off
Feature activation-0.004
a
Token a
Feature activation-0.003
USB
Token USB
Feature activation-0.011
cable
Token cable
Feature activation+0.000
to
Token to
Feature activation-0.003
prototype
Token prototype
Feature activation-0.009
circuit
Token circuit
Feature activation+0.018
board
Token board
Feature activation+0.029
and
Token and
Feature activation+0.007
make
Token make
Feature activation+0.037
cut
Token cut
Feature activation+0.834
marks
Token marks
Feature activation+0.140
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.019
outline
Token outline
Feature activation+0.094
of
Token of
Feature activation+0.005
<|endoftext|>
Token<|endoftext|>
Feature activation+0.334
glue
Token glue
Feature activation+0.019
another
Token another
Feature activation+0.007
1
Token 1
Feature activation+0.007
/
Token/
Feature activation+0.000
4
Token4
Feature activation+0.000
bolt
Token bolt
Feature activation+0.086
.
Token.
Feature activation+0.074
Screw
Token Screw
Feature activation+0.236
the
Token the
Feature activation+0.151
first
Token first
Feature activation+0.144
nut
Token nut
Feature activation+0.779
down
Token down
Feature activation+0.135
until
Token until
Feature activation+0.000
it
Token it
Feature activation+0.000
is
Token is
Feature activation+0.000
snug
Token snug
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.522
glue
Token glue
Feature activation+0.007
another
Token another
Feature activation+0.004
1
Token 1
Feature activation+0.007
/
Token/
Feature activation-0.003
4
Token4
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.006
9
Token9
Feature activation+0.002
.
Token.
Feature activation+0.002
Screw
Token Screw
Feature activation+0.080
the
Token the
Feature activation+0.071
second
Token second
Feature activation+0.310
nut
Token nut
Feature activation+0.256
onto
Token onto
Feature activation+0.063
the
Token the
Feature activation+0.046
carriage
Token carriage
Feature activation+0.035
bolt
Token bolt
Feature activation+0.100
<|endoftext|>
Token<|endoftext|>
Feature activation+0.516
Cut
TokenCut
Feature activation+0.003
the
Token the
Feature activation+0.005
clasp
Token clasp
Feature activation+0.030
off
Token off
Feature activation+0.024
of
Token of
Feature activation+0.005
prototype
Token prototype
Feature activation-0.006
circuit
Token circuit
Feature activation+0.022
board
Token board
Feature activation+0.024
and
Token and
Feature activation+0.011
make
Token make
Feature activation+0.114
cut
Token cut
Feature activation+1.359
marks
Token marks
Feature activation+0.182
around
Token around
Feature activation+0.000
the
Token the
Feature activation+0.000
outline
Token outline
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.026
lid
Token lid
Feature activation+0.000
on
Token on
Feature activation+0.042
the
Token the
Feature activation+0.033
carriage
Token carriage
Feature activation+0.085
bolt
Token bolt
Feature activation+0.360
.
Token.
Feature activation+0.250
Screw
Token Screw
Feature activation+0.123
the
Token the
Feature activation+0.119
third
Token third
Feature activation+0.070
nut
Token nut
Feature activation+0.134
<|endoftext|>
Token<|endoftext|>
Feature activation+0.456
off
Token off
Feature activation-0.006
a
Token a
Feature activation-0.003
USB
Token USB
Feature activation-0.003
cable
Token cable
Feature activation+0.001
to
Token to
Feature activation-0.001
.
Token.
Feature activation+0.005
Ċ
TokenĊ
Feature activation+0.024
Ċ
TokenĊ
Feature activation+0.015
I
TokenI
Feature activation+0.038
cut
Token cut
Feature activation+0.287
openings
Token openings
Feature activation+0.727
for
Token for
Feature activation+0.452
the
Token the
Feature activation+0.000
two
Token two
Feature activation+0.000
device
Token device
Feature activation+0.000
boxes
Token boxes
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.444
cap
Token cap
Feature activation-0.000
,
Token,
Feature activation-0.000
lid
Token lid
Feature activation+0.000
and
Token and
Feature activation-0.000
can
Token can
Feature activation+0.000
the
Token the
Feature activation+0.015
wire
Token wire
Feature activation+0.149
.
Token.
Feature activation+0.062
Ċ
TokenĊ
Feature activation+0.024
Ċ
TokenĊ
Feature activation+0.018
Cut
TokenCut
Feature activation+0.518
as
Token as
Feature activation+0.067
close
Token close
Feature activation+0.288
to
Token to
Feature activation+0.462
the
Token the
Feature activation+0.113
wire
Token wire
Feature activation+0.178
second
Token second
Feature activation+0.025
nut
Token nut
Feature activation+0.101
onto
Token onto
Feature activation+0.113
the
Token the
Feature activation+0.038
carriage
Token carriage
Feature activation+0.077
bolt
Token bolt
Feature activation+0.378
until
Token until
Feature activation+0.133
it
Token it
Feature activation+0.018
is
Token is
Feature activation+0.037
about
Token about
Feature activation+0.024
1
Token 1
Feature activation+0.013
the
Token the
Feature activation+0.005
wire
Token wire
Feature activation+0.034
.
Token.
Feature activation+0.097
Ċ
TokenĊ
Feature activation+0.044
Ċ
TokenĊ
Feature activation+0.182
Cut
TokenCut
Feature activation+0.722
as
Token as
Feature activation+0.191
close
Token close
Feature activation+0.053
to
Token to
Feature activation+0.031
the
Token the
Feature activation+0.020
wire
Token wire
Feature activation+0.064

Decoder Weights Distribution

Head 0: 0.06

Head 1: 0.02

Head 2: 0.06

Head 3: 0.11

Head 4: 0.06

Head 5: 0.08

Head 6: 0.03

Head 7: 0.05

Head 8: 0.04

Head 9: 0.14

Head 10: 0.23

Head 11: 0.13

Positive logits

diagonal1.41

rotated1.33

Instruct1.33

width1.32

Setup1.29

Materials1.29

triangular1.27

resize1.27

Optional1.27

underside1.27

triangle1.26

diameter1.24

optional1.23

loop1.22

perpendicular1.22

Instructions1.20

apest1.19

Optional1.16

length1.15

fork1.15

Negative logits

BuyableInstoreAndOnline-1.49

oneliness-1.39

academ-1.25

memes-1.22

Politics-1.22

girls-1.21

fandom-1.19

executives-1.18

births-1.18

scandals-1.17

psychiatrists-1.17

societal-1.16

prostitutes-1.16

Nigeria-1.14

celebrities-1.14

friendships-1.13

arenthood-1.12

audiences-1.12

regrets-1.12

partying-1.12

INTERVAL 2.153 - 2.392
CONTAINS 0.000%

of
Token of
Feature activation+1.258
the
Token the
Feature activation+1.334
per
Token per
Feature activation+1.234
for
Tokenfor
Feature activation+0.260
ated
Tokenated
Feature activation+1.559
board
Token board
Feature activation+2.162
.
Token.
Feature activation+1.920
This
Token This
Feature activation+1.710
was
Token was
Feature activation+1.242
not
Token not
Feature activation+1.027
part
Token part
Feature activation+0.997
bolt
Token bolt
Feature activation+1.523
.
Token.
Feature activation+1.279
Screw
Token Screw
Feature activation+1.401
the
Token the
Feature activation+1.907
third
Token third
Feature activation+1.619
nut
Token nut
Feature activation+2.392
onto
Token onto
Feature activation+1.438
the
Token the
Feature activation+1.278
bolt
Token bolt
Feature activation+1.891
until
Token until
Feature activation+1.024
it
Token it
Feature activation+1.437
bolt
Token bolt
Feature activation+1.495
.
Token.
Feature activation+1.335
Screw
Token Screw
Feature activation+1.255
the
Token the
Feature activation+1.640
second
Token second
Feature activation+1.580
nut
Token nut
Feature activation+2.217
against
Token against
Feature activation+0.852
the
Token the
Feature activation+1.141
bottom
Token bottom
Feature activation+1.169
of
Token of
Feature activation+0.753
the
Token the
Feature activation+0.727

INTERVAL 1.913 - 2.153
CONTAINS 0.000%

the
Token the
Feature activation+1.334
per
Token per
Feature activation+1.234
for
Tokenfor
Feature activation+0.260
ated
Tokenated
Feature activation+1.559
board
Token board
Feature activation+2.162
.
Token.
Feature activation+1.920
This
Token This
Feature activation+1.710
was
Token was
Feature activation+1.242
not
Token not
Feature activation+1.027
part
Token part
Feature activation+0.997
of
Token of
Feature activation+0.932
bolt
Token bolt
Feature activation+1.135
through
Token through
Feature activation+1.086
the
Token the
Feature activation+1.299
g
Token g
Feature activation+0.804
rom
Tokenrom
Feature activation+0.926
met
Tokenmet
Feature activation+1.978
from
Token from
Feature activation+1.160
the
Token the
Feature activation+0.979
back
Token back
Feature activation+1.168
towards
Token towards
Feature activation+1.031
the
Token the
Feature activation+1.066
Put
Token Put
Feature activation+0.848
one
Token one
Feature activation+1.008
piece
Token piece
Feature activation+1.045
of
Token of
Feature activation+1.187
each
Token each
Feature activation+0.861
thread
Token thread
Feature activation+2.098
(
Token (
Feature activation+1.154
2
Token2
Feature activation+1.212
pieces
Token pieces
Feature activation+0.923
per
Token per
Feature activation+0.402
slit
Token slit
Feature activation+0.368
Cut
TokenCut
Feature activation+1.463
as
Token as
Feature activation+1.283
close
Token close
Feature activation+1.790
to
Token to
Feature activation+1.483
the
Token the
Feature activation+1.640
wire
Token wire
Feature activation+1.984
as
Token as
Feature activation+1.833
possible
Token possible
Feature activation+1.046
.
Token.
Feature activation+1.786
Ċ
TokenĊ
Feature activation+1.812
Ċ
TokenĊ
Feature activation+1.671
outline
Token outline
Feature activation+0.697
of
Token of
Feature activation+0.627
the
Token the
Feature activation+0.739
relay
Token relay
Feature activation+0.984
.
Token.
Feature activation+1.198
Cut
Token Cut
Feature activation+1.969
two
Token two
Feature activation+1.645
prototype
Token prototype
Feature activation+1.372
circuit
Token circuit
Feature activation+1.215
boards
Token boards
Feature activation+1.587
to
Token to
Feature activation+1.660

INTERVAL 1.674 - 1.913
CONTAINS 0.000%

bolt
Token bolt
Feature activation+0.961
up
Token up
Feature activation+1.529
through
Token through
Feature activation+1.293
the
Token the
Feature activation+1.079
bottom
Token bottom
Feature activation+1.532
.
Token.
Feature activation+1.675
Ċ
TokenĊ
Feature activation+1.249
Ċ
TokenĊ
Feature activation+0.980
Figure
TokenFigure
Feature activation+0.256
:
Token:
Feature activation+0.001
Side
Token Side
Feature activation+0.505
per
Token per
Feature activation+1.234
for
Tokenfor
Feature activation+0.260
ated
Tokenated
Feature activation+1.559
board
Token board
Feature activation+2.162
.
Token.
Feature activation+1.920
This
Token This
Feature activation+1.710
was
Token was
Feature activation+1.242
not
Token not
Feature activation+1.027
part
Token part
Feature activation+0.997
of
Token of
Feature activation+0.932
the
Token the
Feature activation+0.755
cut
Token cut
Feature activation+1.516
either
Token either
Feature activation+1.648
7
Token 7
Feature activation+1.374
or
Token or
Feature activation+0.716
14
Token 14
Feature activation+0.542
pieces
Token pieces
Feature activation+1.733
(
Token (
Feature activation+1.192
2
Token2
Feature activation+1.051
of
Token of
Feature activation+0.834
each
Token each
Feature activation+0.731
colour
Token colour
Feature activation+0.797
as
Token as
Feature activation+1.283
close
Token close
Feature activation+1.790
to
Token to
Feature activation+1.483
the
Token the
Feature activation+1.640
wire
Token wire
Feature activation+1.984
as
Token as
Feature activation+1.833
possible
Token possible
Feature activation+1.046
.
Token.
Feature activation+1.786
Ċ
TokenĊ
Feature activation+1.812
Ċ
TokenĊ
Feature activation+1.671
That
TokenThat
Feature activation+1.149
Ċ
TokenĊ
Feature activation+0.678
Ċ
TokenĊ
Feature activation+0.915
I
TokenI
Feature activation+0.648
cut
Token cut
Feature activation+1.622
openings
Token openings
Feature activation+1.796
for
Token for
Feature activation+1.855
the
Token the
Feature activation+1.395
two
Token two
Feature activation+1.596
device
Token device
Feature activation+1.352
boxes
Token boxes
Feature activation+1.047
.
Token.
Feature activation+1.618

INTERVAL 1.435 - 1.674
CONTAINS 0.001%

marking
Token marking
Feature activation+0.732
off
Token off
Feature activation+0.386
the
Token the
Feature activation+0.864
space
Token space
Feature activation+0.909
to
Token to
Feature activation+0.687
cut
Token cut
Feature activation+1.592
out
Token out
Feature activation+1.025
.
Token.
Feature activation+0.915
The
Token The
Feature activation+0.687
blue
Token blue
Feature activation+0.711
tape
Token tape
Feature activation+0.806
The
Token The
Feature activation+1.217
other
Token other
Feature activation+1.500
board
Token board
Feature activation+1.744
should
Token should
Feature activation+1.460
be
Token be
Feature activation+1.407
cut
Token cut
Feature activation+1.543
to
Token to
Feature activation+1.553
a
Token a
Feature activation+1.585
small
Token small
Feature activation+1.309
square
Token square
Feature activation+1.393
using
Token using
Feature activation+1.709
.
Token.
Feature activation+1.279
Screw
Token Screw
Feature activation+1.401
the
Token the
Feature activation+1.907
third
Token third
Feature activation+1.619
nut
Token nut
Feature activation+2.392
onto
Token onto
Feature activation+1.438
the
Token the
Feature activation+1.278
bolt
Token bolt
Feature activation+1.891
until
Token until
Feature activation+1.024
it
Token it
Feature activation+1.437
is
Token is
Feature activation+1.091
of
Token of
Feature activation+1.077
the
Token the
Feature activation+1.487
fluorescent
Token fluorescent
Feature activation+0.854
shield
Token shield
Feature activation+1.435
to
Token to
Feature activation+1.429
create
Token create
Feature activation+1.658
a
Token a
Feature activation+1.159
âĢ
Token âĢ
Feature activation+0.457
ľ
Tokenľ
Feature activation+0.132
sm
Tokensm
Feature activation+0.276
oked
Tokenoked
Feature activation+0.000
the
Token the
Feature activation+1.279
per
Token per
Feature activation+0.563
for
Tokenfor
Feature activation+0.000
ated
Tokenated
Feature activation+0.816
board
Token board
Feature activation+1.155
.
Token.
Feature activation+1.540
I
Token I
Feature activation+0.988
added
Token added
Feature activation+1.610
a
Token a
Feature activation+1.519
seven
Token seven
Feature activation+0.828
inches
Token inches
Feature activation+1.639

INTERVAL 1.196 - 1.435
CONTAINS 0.002%

with
Token with
Feature activation+0.738
glue
Token glue
Feature activation+0.930
).
Token).
Feature activation+0.817
Let
Token Let
Feature activation+0.787
the
Token the
Feature activation+0.773
glue
Token glue
Feature activation+1.416
dry
Token dry
Feature activation+0.772
completely
Token completely
Feature activation+0.310
.
Token.
Feature activation+0.504
(
Token (
Feature activation+0.507
If
TokenIf
Feature activation+0.031
shield
Token shield
Feature activation+0.960
to
Token to
Feature activation+1.523
the
Token the
Feature activation+1.418
same
Token same
Feature activation+1.164
length
Token length
Feature activation+1.551
as
Token as
Feature activation+1.322
the
Token the
Feature activation+0.996
pipe
Token pipe
Feature activation+1.105
.
Token.
Feature activation+1.648
Then
Token Then
Feature activation+1.113
use
Token use
Feature activation+1.162
Att
Token Att
Feature activation+0.278
ach
Tokenach
Feature activation+0.757
all
Token all
Feature activation+0.823
twelve
Token twelve
Feature activation+1.090
strands
Token strands
Feature activation+1.337
to
Token to
Feature activation+1.252
the
Token the
Feature activation+0.930
circle
Token circle
Feature activation+0.712
.
Token.
Feature activation+0.979
Ċ
TokenĊ
Feature activation+0.688
Ċ
TokenĊ
Feature activation+0.580
another
Token another
Feature activation+1.508
1
Token 1
Feature activation+1.498
âģ
Tokenâģ
Feature activation+0.008
Ħ
TokenĦ
Feature activation+0.314
4
Token4
Feature activation+0.548
â̳
Tokenâ̳
Feature activation+1.424
hole
Token hole
Feature activation+1.790
in
Token in
Feature activation+1.242
the
Token the
Feature activation+0.978
center
Token center
Feature activation+1.094
of
Token of
Feature activation+1.350
have
Token have
Feature activation+0.394
to
Token to
Feature activation+0.529
remove
Token remove
Feature activation+0.592
all
Token all
Feature activation+0.864
of
Token of
Feature activation+0.787
them
Token them
Feature activation+1.216
.
Token.
Feature activation+0.654
At
Token At
Feature activation+0.663
which
Token which
Feature activation+0.143
point
Token point
Feature activation+0.371
,
Token,
Feature activation+0.000

INTERVAL 0.957 - 1.196
CONTAINS 0.005%

.
Token.
Feature activation+0.769
Ċ
TokenĊ
Feature activation+0.429
Ċ
TokenĊ
Feature activation+0.453
5
Token5
Feature activation+0.298
.
Token.
Feature activation+0.302
Push
Token Push
Feature activation+1.054
the
Token the
Feature activation+0.845
carriage
Token carriage
Feature activation+0.814
bolt
Token bolt
Feature activation+0.961
up
Token up
Feature activation+1.529
through
Token through
Feature activation+1.293
Use
Token Use
Feature activation+0.600
the
Token the
Feature activation+0.473
tin
Token tin
Feature activation+0.328
sn
Token sn
Feature activation+0.286
ips
Tokenips
Feature activation+0.686
to
Token to
Feature activation+1.097
cut
Token cut
Feature activation+1.368
off
Token off
Feature activation+0.985
the
Token the
Feature activation+0.855
bottom
Token bottom
Feature activation+1.177
.
Token.
Feature activation+0.966
mount
Token mount
Feature activation+0.191
.
Token.
Feature activation+0.724
The
Token The
Feature activation+0.634
end
Token end
Feature activation+0.890
of
Token of
Feature activation+1.144
the
Token the
Feature activation+0.971
tether
Token tether
Feature activation+1.217
's
Token's
Feature activation+1.020
flat
Token flat
Feature activation+0.834
section
Token section
Feature activation+0.805
of
Token of
Feature activation+0.744
is
Token is
Feature activation+0.252
to
Token to
Feature activation+0.236
prepare
Token prepare
Feature activation+0.387
the
Token the
Feature activation+0.759
short
Token short
Feature activation+1.041
horizontal
Token horizontal
Feature activation+1.017
slider
Token slider
Feature activation+1.042
angles
Token angles
Feature activation+0.710
with
Token with
Feature activation+0.951
the
Token the
Feature activation+0.901
addition
Token addition
Feature activation+0.987
.
Token.
Feature activation+0.670
Ċ
TokenĊ
Feature activation+0.301
Ċ
TokenĊ
Feature activation+0.322
When
TokenWhen
Feature activation+0.354
cutting
Token cutting
Feature activation+0.924
across
Token across
Feature activation+1.026
the
Token the
Feature activation+1.097
grain
Token grain
Feature activation+1.140
(
Token (
Feature activation+0.782
or
Tokenor
Feature activation+0.680
on
Token on
Feature activation+0.971

INTERVAL 0.718 - 0.957
CONTAINS 0.010%

front
Token front
Feature activation+0.548
of
Token of
Feature activation+0.404
the
Token the
Feature activation+0.375
bra
Token bra
Feature activation+0.108
-
Token -
Feature activation+0.637
the
Token the
Feature activation+0.762
part
Token part
Feature activation+1.022
the
Token the
Feature activation+0.561
clamp
Token clamp
Feature activation+0.398
is
Token is
Feature activation+0.677
attached
Token attached
Feature activation+0.490
trouble
Token trouble
Feature activation+0.346
to
Token to
Feature activation+0.061
making
Token making
Feature activation+0.417
a
Token a
Feature activation+0.290
neck
Token neck
Feature activation+0.868
line
Token line
Feature activation+0.889
part
Token part
Feature activation+0.590
.
Token.
Feature activation+0.410
No
Token No
Feature activation+0.008
doubt
Token doubt
Feature activation+0.000
you
Token you
Feature activation+0.000
.
Token.
Feature activation+1.263
I
Token I
Feature activation+1.112
find
Token find
Feature activation+0.463
that
Token that
Feature activation+0.249
doubling
Token doubling
Feature activation+0.811
up
Token up
Feature activation+0.892
the
Token the
Feature activation+0.757
fl
Token fl
Feature activation+0.510
oss
Tokenoss
Feature activation+1.162
(
Token (
Feature activation+1.003
using
Tokenusing
Feature activation+0.962
cm
Tokencm
Feature activation+0.292
tart
Token tart
Feature activation+0.110
tin
Token tin
Feature activation+0.284
and
Token and
Feature activation+0.474
trim
Token trim
Feature activation+0.571
any
Token any
Feature activation+0.924
excess
Token excess
Feature activation+0.595
.
Token.
Feature activation+0.467
Repeat
Token Repeat
Feature activation+0.311
with
Token with
Feature activation+0.133
the
Token the
Feature activation+0.163
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ll
Tokenll
Feature activation+0.000
see
Token see
Feature activation+0.210
the
Token the
Feature activation+0.184
elastic
Token elastic
Feature activation+0.907
start
Token start
Feature activation+0.661
to
Token to
Feature activation+0.591
come
Token come
Feature activation+0.692
away
Token away
Feature activation+0.415
from
Token from
Feature activation+0.193

INTERVAL 0.478 - 0.718
CONTAINS 0.019%

glass
Token glass
Feature activation+0.109
,
Token,
Feature activation+0.160
scoop
Token scoop
Feature activation+0.410
the
Token the
Feature activation+0.193
firm
Token firm
Feature activation+0.000
foam
Token foam
Feature activation+0.678
from
Token from
Feature activation+0.172
your
Token your
Feature activation+0.406
bowl
Token bowl
Feature activation+0.097
and
Token and
Feature activation+0.278
doll
Token doll
Feature activation+0.196
of
Token of
Feature activation+0.658
the
Token the
Feature activation+0.488
quad
Token quad
Feature activation+0.163
cop
Tokencop
Feature activation+0.000
ter
Tokenter
Feature activation+0.000
(
Token (
Feature activation+0.521
facing
Tokenfacing
Feature activation+0.024
the
Token the
Feature activation+0.107
correct
Token correct
Feature activation+0.360
way
Token way
Feature activation+0.270
)
Token)
Feature activation+0.366
screws
Token screws
Feature activation+0.241
and
Token and
Feature activation+0.773
dis
Token dis
Feature activation+0.312
as
Tokenas
Feature activation+0.000
semble
Tokensemble
Feature activation+0.540
the
Token the
Feature activation+0.552
extension
Token extension
Feature activation+0.089
.
Token.
Feature activation+0.505
Then
Token Then
Feature activation+0.491
dis
Token dis
Feature activation+0.987
as
Tokenas
Feature activation+0.058
the
Token the
Feature activation+0.000
cookies
Token cookies
Feature activation+0.000
are
Token are
Feature activation+0.000
cool
Token cool
Feature activation+0.000
fill
Token fill
Feature activation+0.343
a
Token a
Feature activation+0.484
pastry
Token pastry
Feature activation+0.000
bag
Token bag
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
Nut
Token Nut
Feature activation+0.000
top
Token top
Feature activation+0.527
:
Token:
Feature activation+0.652
Ċ
TokenĊ
Feature activation+0.716
Ċ
TokenĊ
Feature activation+0.616
As
TokenAs
Feature activation+0.218
it
Token it
Feature activation+0.633
was
Token was
Feature activation+0.257
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.526
hook
Token hook
Feature activation+0.812
had
Token had
Feature activation+1.077

INTERVAL 0.239 - 0.478
CONTAINS 0.041%

headlights
Token headlights
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.147
tail
Token tail
Feature activation+0.477
unit
Token unit
Feature activation+0.508
is
Token is
Feature activation+0.249
un
Token un
Feature activation+0.062
apolog
Tokenapolog
Feature activation+0.000
etically
Tokenetically
Feature activation+0.000
Layer
Token Layer
Feature activation+0.000
to
Token to
Feature activation+0.000
Create
Token Create
Feature activation+0.102
Cl
Token Cl
Feature activation+0.000
ipping
Tokenipping
Feature activation+0.047
Mask
Token Mask
Feature activation+0.280
check
Token check
Feature activation+0.000
box
Tokenbox
Feature activation+0.032
and
Token and
Feature activation+0.000
click
Token click
Feature activation+0.000
OK
Token OK
Feature activation+0.000
in
Token in
Feature activation+0.000
skillet
Token skillet
Feature activation+0.000
and
Token and
Feature activation+0.000
cook
Token cook
Feature activation+0.000
on
Token on
Feature activation+0.000
each
Token each
Feature activation+0.395
side
Token side
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
P
TokenP
Feature activation+0.000
iers
Tokeniers
Feature activation+0.000
Morgan
Token Morgan
Feature activation+0.000
of
Token of
Feature activation+0.043
butter
Token butter
Feature activation+0.000
.
Token.
Feature activation+0.000
Mix
Token Mix
Feature activation+0.000
well
Token well
Feature activation+0.063
so
Token so
Feature activation+0.258
that
Token that
Feature activation+0.023
the
Token the
Feature activation+0.233
apple
Token apple
Feature activation+0.000
slices
Token slices
Feature activation+0.000
are
Token are
Feature activation+0.000
your
Token your
Feature activation+0.000
pants
Token pants
Feature activation+0.000
,
Token,
Feature activation+0.000
ensuring
Token ensuring
Feature activation+0.175
they
Token they
Feature activation+0.080
wont
Token wont
Feature activation+0.270
slip
Token slip
Feature activation+0.535
out
Token out
Feature activation+0.404
of
Token of
Feature activation+0.116
place
Token place
Feature activation+0.000
when
Token when
Feature activation+0.000

INTERVAL 0.000 - 0.239
CONTAINS 99.923%

he
Token he
Feature activation+0.000
opens
Token opens
Feature activation+0.000
doors
Token doors
Feature activation+0.000
,
Token,
Feature activation+0.000
a
Token a
Feature activation+0.000
perfect
Token perfect
Feature activation+0.000
gentleman
Token gentleman
Feature activation+0.000
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
she
Token she
Feature activation+0.000
Dragon
TokenDragon
Feature activation+0.000
Ball
Token Ball
Feature activation+0.000
Super
Token Super
Feature activation+0.000
"
Token"
Feature activation+0.000
is
Token is
Feature activation+0.000
not
Token not
Feature activation+0.000
yet
Token yet
Feature activation+0.000
out
Token out
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
we
Token we
Feature activation+0.000
answers
Token answers
Feature activation+0.000
to
Token to
Feature activation+0.000
some
Token some
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
main
Token main
Feature activation+0.000
questions
Token questions
Feature activation+0.000
surrounding
Token surrounding
Feature activation+0.000
the
Token the
Feature activation+0.000
issue
Token issue
Feature activation+0.000
.
Token.
Feature activation+0.000
B
TokenB
Feature activation+0.000
rav
Tokenrav
Feature activation+0.000
o
Tokeno
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
J
TokenJ
Feature activation+0.000
umping
Tokenumping
Feature activation+0.000
over
Token over
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
d
Tokend
Feature activation+0.000
been
Token been
Feature activation+0.000
arrested
Token arrested
Feature activation+0.000
and
Token and
Feature activation+0.000
said
Token said
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
a
Token a
Feature activation+0.000
mistake
Token mistake
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 21: Uninterpretable

TOP ACTIVATIONS
MAX = 2.857

last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Do
TokenDo
Feature activation+1.510
zens
Tokenzens
Feature activation+2.857
of
Token of
Feature activation+0.857
workers
Token workers
Feature activation+0.756
at
Token at
Feature activation+0.905
Japan
Token Japan
Feature activation+0.455
's
Token's
Feature activation+0.728
's
Token's
Feature activation+0.000
National
Token National
Feature activation+0.000
Day
Token Day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Thousands
TokenThousands
Feature activation+2.832
of
Token of
Feature activation+0.931
Catal
Token Catal
Feature activation+0.700
ans
Tokenans
Feature activation+1.404
have
Token have
Feature activation+1.055
rallied
Token rallied
Feature activation+1.574
er
Tokener
Feature activation+0.000
War
Token War
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.125
Do
TokenDo
Feature activation+1.428
zens
Tokenzens
Feature activation+2.612
of
Token of
Feature activation+1.411
previously
Token previously
Feature activation+0.932
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
published
Tokenpublished
Feature activation+0.608
Games
Token Games
Feature activation+0.185
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
More
TokenMore
Feature activation+1.156
than
Token than
Feature activation+0.862
400
Token 400
Feature activation+2.291
people
Token people
Feature activation+1.183
have
Token have
Feature activation+1.097
been
Token been
Feature activation+0.644
detained
Token detained
Feature activation+1.098
in
Token in
Feature activation+0.645
the
Token the
Feature activation+0.000
night
Token night
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
At
TokenAt
Feature activation+1.975
least
Token least
Feature activation+2.275
30
Token 30
Feature activation+1.597
protesters
Token protesters
Feature activation+0.973
were
Token were
Feature activation+0.849
arrested
Token arrested
Feature activation+0.860
in
Token in
Feature activation+0.797
executives
Token executives
Feature activation+0.000
at
Token at
Feature activation+0.000
Microsoft
Token Microsoft
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Microsoft
TokenMicrosoft
Feature activation+2.262
's
Token's
Feature activation+0.998
research
Token research
Feature activation+0.596
boss
Token boss
Feature activation+0.976
Craig
Token Craig
Feature activation+0.344
Mund
Token Mund
Feature activation+0.870
de
Token de
Feature activation+0.000
Janeiro
Token Janeiro
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Some
TokenSome
Feature activation+1.826
200
Token 200
Feature activation+2.257
,
Token,
Feature activation+0.709
000
Token000
Feature activation+1.796
people
Token people
Feature activation+1.259
have
Token have
Feature activation+1.106
protested
Token protested
Feature activation+1.468
the
Token the
Feature activation+0.000
space
Token space
Feature activation+0.000
station
Token station
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.055
"
Token"
Feature activation+2.255
It
TokenIt
Feature activation+0.665
's
Token's
Feature activation+0.046
not
Token not
Feature activation+0.000
an
Token an
Feature activation+0.000
original
Token original
Feature activation+0.000
day
Token day
Feature activation+0.000
off
Token off
Feature activation+0.000
work
Token work
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Thousands
TokenThousands
Feature activation+2.182
of
Token of
Feature activation+0.347
people
Token people
Feature activation+1.109
are
Token are
Feature activation+0.629
reported
Token reported
Feature activation+1.077
to
Token to
Feature activation+0.533
69
Token 69
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+1.711
last
Token last
Feature activation+2.171
five
Token five
Feature activation+1.179
years
Token years
Feature activation+0.999
of
Token of
Feature activation+0.242
David
Token David
Feature activation+1.580
Bowie
Token Bowie
Feature activation+0.614
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.098
About
TokenAbout
Feature activation+1.593
2
Token 2
Feature activation+1.877
,
Token,
Feature activation+1.196
000
Token000
Feature activation+2.108
Bulgar
Token Bulgar
Feature activation+1.158
ians
Tokenians
Feature activation+1.071
have
Token have
Feature activation+0.809
marched
Token marched
Feature activation+1.488
in
Token in
Feature activation+1.092
came
Token came
Feature activation+0.000
too
Token too
Feature activation+0.000
late
Token late
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+2.103
P
TokenP
Feature activation+0.350
ays
Tokenays
Feature activation+0.000
lip
Tokenlip
Feature activation+0.000
-
Token-
Feature activation+0.000
gate
Tokengate
Feature activation+0.000
Islamic
Token Islamic
Feature activation+0.000
State
Token State
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Saudi
TokenSaudi
Feature activation+2.004
Arabia
Token Arabia
Feature activation+2.098
has
Token has
Feature activation+1.001
arrested
Token arrested
Feature activation+0.814
88
Token 88
Feature activation+1.288
people
Token people
Feature activation+0.326
accused
Token accused
Feature activation+0.223
launch
Token launch
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.086
Watch
TokenWatch
Feature activation+1.416
Dogs
Token Dogs
Feature activation+1.287
'
Token'
Feature activation+2.048
publisher
Token publisher
Feature activation+1.303
has
Token has
Feature activation+0.941
apologised
Token apologised
Feature activation+0.687
after
Token after
Feature activation+0.935
a
Token a
Feature activation+0.653
devices
Token devices
Feature activation+0.000
"
Token"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.565
Car
TokenCar
Feature activation+1.328
firm
Token firm
Feature activation+2.026
Volkswagen
Token Volkswagen
Feature activation+1.469
(
Token (
Feature activation+0.731
VW
TokenVW
Feature activation+0.000
)
Token)
Feature activation+1.105
has
Token has
Feature activation+0.831
blast
Token blast
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+1.237
massive
Token massive
Feature activation+1.233
blast
Token blast
Feature activation+2.014
at
Token at
Feature activation+1.106
a
Token a
Feature activation+1.185
munitions
Token munitions
Feature activation+0.414
dump
Token dump
Feature activation+0.727
in
Token in
Feature activation+0.899
running
Token running
Feature activation+0.000
high
Token high
Feature activation+0.000
"
Token"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.004
One
TokenOne
Feature activation+2.011
passenger
Token passenger
Feature activation+1.068
,
Token,
Feature activation+0.517
Mark
Token Mark
Feature activation+0.268
Shore
Token Shore
Feature activation+0.304
y
Tokeny
Feature activation+0.637
of
Token of
Feature activation+0.000
Islamic
Token Islamic
Feature activation+0.000
State
Token State
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Saudi
TokenSaudi
Feature activation+2.004
Arabia
Token Arabia
Feature activation+2.098
has
Token has
Feature activation+1.001
arrested
Token arrested
Feature activation+0.814
88
Token 88
Feature activation+1.288
people
Token people
Feature activation+0.326
the
Token the
Feature activation+0.000
Tube
Token Tube
Feature activation+0.000
network
Token network
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
One
TokenOne
Feature activation+1.997
of
Token of
Feature activation+0.540
the
Token the
Feature activation+0.662
unions
Token unions
Feature activation+0.741
which
Token which
Feature activation+0.022
was
Token was
Feature activation+0.274
into
Token into
Feature activation+0.000
the
Token the
Feature activation+0.000
night
Token night
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
At
TokenAt
Feature activation+1.975
least
Token least
Feature activation+2.275
30
Token 30
Feature activation+1.597
protesters
Token protesters
Feature activation+0.973
were
Token were
Feature activation+0.849
arrested
Token arrested
Feature activation+0.860

Top DFA by src position
MAX = 1.994

sparked
Token sparked
Feature activation+0.053
protests
Token protests
Feature activation+0.238
last
Token last
Feature activation+0.015
week
Token week
Feature activation+0.060
Ċ
TokenĊ
Feature activation+1.199
Ċ
TokenĊ
Feature activation+1.909
Do
TokenDo
Feature activation+0.075
zens
Tokenzens
Feature activation+0.051
of
Token of
Feature activation+0.000
workers
Token workers
Feature activation+0.000
at
Token at
Feature activation+0.000
Catalonia
Token Catalonia
Feature activation+0.035
's
Token's
Feature activation+0.009
National
Token National
Feature activation+0.004
Day
Token Day
Feature activation+0.021
Ċ
TokenĊ
Feature activation+1.185
Ċ
TokenĊ
Feature activation+1.404
Thousands
TokenThousands
Feature activation+0.123
of
Token of
Feature activation+0.000
Catal
Token Catal
Feature activation+0.000
ans
Tokenans
Feature activation+0.000
have
Token have
Feature activation+0.000
Second
Token Second
Feature activation+0.007
Bo
Token Bo
Feature activation+0.016
er
Tokener
Feature activation+0.006
War
Token War
Feature activation+0.028
Ċ
TokenĊ
Feature activation+1.305
Ċ
TokenĊ
Feature activation+1.510
Do
TokenDo
Feature activation+0.119
zens
Tokenzens
Feature activation+0.016
of
Token of
Feature activation+0.000
previously
Token previously
Feature activation+0.000
-
Token-
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.781
Davis
Token Davis
Feature activation-0.001
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation-0.000
Ċ
TokenĊ
Feature activation+0.000
Big
TokenBig
Feature activation-0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.793
high
Token high
Feature activation-0.000
-
Token-
Feature activation-0.034
<|endoftext|>
Token<|endoftext|>
Feature activation+0.027
Image
TokenImage
Feature activation+0.007
copyright
Token copyright
Feature activation+0.006
longest
Token longest
Feature activation+0.009
serving
Token serving
Feature activation+0.011
executives
Token executives
Feature activation+0.013
at
Token at
Feature activation+0.064
Microsoft
Token Microsoft
Feature activation+0.228
Ċ
TokenĊ
Feature activation+1.408
Ċ
TokenĊ
Feature activation+0.954
Microsoft
TokenMicrosoft
Feature activation+0.142
's
Token's
Feature activation+0.000
research
Token research
Feature activation+0.000
boss
Token boss
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.844
Green
Token Green
Feature activation+0.000
of
Token of
Feature activation+0.002
Texas
Token Texas
Feature activation+0.003
have
Token have
Feature activation+0.001
pushed
Token pushed
Feature activation+0.001
<|endoftext|>
Token<|endoftext|>
Feature activation+1.925
Ċ
TokenĊ
Feature activation+0.028
Image
TokenImage
Feature activation+0.008
copyright
Token copyright
Feature activation+0.000
Es
Token Es
Feature activation+0.010
a
Tokena
Feature activation+0.016
<|endoftext|>
Token<|endoftext|>
Feature activation+1.735
programs
Token programs
Feature activation-0.005
,
Token,
Feature activation+0.003
he
Token he
Feature activation-0.000
watches
Token watches
Feature activation+0.006
different
Token different
Feature activation+0.002
,
Token,
Feature activation+0.078
aged
Token aged
Feature activation+0.022
69
Token 69
Feature activation+0.009
.
Token.
Feature activation+0.043
Ċ
TokenĊ
Feature activation+1.020
Ċ
TokenĊ
Feature activation+1.588
The
TokenThe
Feature activation+0.235
last
Token last
Feature activation-0.007
five
Token five
Feature activation+0.000
years
Token years
Feature activation+0.000
of
Token of
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.796
Image
TokenImage
Feature activation+0.004
caption
Token caption
Feature activation+0.009
The
Token The
Feature activation+0.060
last
Token last
Feature activation+0.004
week
Token week
Feature activation+0.069
<|endoftext|>
Token<|endoftext|>
Feature activation+1.994
description
Token description
Feature activation-0.009
with
Token with
Feature activation-0.001
an
Token an
Feature activation+0.001
in
Token in
Feature activation+0.001
-
Token-
Feature activation-0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.321
S
TokenS
Feature activation+0.001
lim
Tokenlim
Feature activation+0.001
,
Token,
Feature activation+0.003
durable
Token durable
Feature activation+0.000
,
Token,
Feature activation-0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.446
Image
TokenImage
Feature activation+0.002
copyright
Token copyright
Feature activation+0.003
Ubisoft
Token Ubisoft
Feature activation+0.078
Image
Token Image
Feature activation+0.003
caption
Token caption
Feature activation+0.036
<|endoftext|>
Token<|endoftext|>
Feature activation+1.517
Image
TokenImage
Feature activation+0.008
copyright
Token copyright
Feature activation+0.007
AFP
Token AFP
Feature activation+0.035
/
Token/
Feature activation+0.002
Getty
TokenGetty
Feature activation+0.006
<|endoftext|>
Token<|endoftext|>
Feature activation+1.596
.
Token.
Feature activation-0.001
Ċ
TokenĊ
Feature activation-0.000
Ċ
TokenĊ
Feature activation-0.000
Buy
TokenBuy
Feature activation+0.000
Photo
Token Photo
Feature activation+0.001
<|endoftext|>
Token<|endoftext|>
Feature activation+1.643
Ċ
TokenĊ
Feature activation+0.011
Media
TokenMedia
Feature activation+0.006
playback
Token playback
Feature activation-0.003
is
Token is
Feature activation+0.007
unsupported
Token unsupported
Feature activation-0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.230
S
TokenS
Feature activation+0.002
lim
Tokenlim
Feature activation+0.004
,
Token,
Feature activation+0.001
durable
Token durable
Feature activation-0.000
,
Token,
Feature activation-0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.527
Team
Token Team
Feature activation+0.001
(@
Token (@
Feature activation-0.003
Tw
TokenTw
Feature activation+0.000
itch
Tokenitch
Feature activation+0.000
y
Tokeny
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+1.886
high
Token high
Feature activation-0.001
-
Token-
Feature activation+0.001
<|endoftext|>
Token<|endoftext|>
Feature activation+0.051
Image
TokenImage
Feature activation+0.013
copyright
Token copyright
Feature activation+0.001

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.02

Head 2: 0.05

Head 3: 0.35

Head 4: 0.06

Head 5: 0.07

Head 6: 0.03

Head 7: 0.04

Head 8: 0.11

Head 9: 0.07

Head 10: 0.08

Head 11: 0.09

Positive logits

"))2.23

SHARE1.68

rote1.67

BUS1.64

TIT1.64

whisky1.63

ordable1.62

zens1.60

Sorry1.58

GREEN1.50

hello1.46

eworld1.44

taxpayer1.44

earable1.43

broadcaster1.43

owan1.43

booze1.40

cocktails1.39

CLOSE1.39

athletics1.38

Negative logits

��-1.99

Secondly-1.95

��-1.86

 -1.79

pts-1.78

thumbnails-1.77

';-1.77

Interstitial-1.75

*/-1.71

��-1.71

���-1.71

  -1.68

scrib-1.65

�醒-1.64

Democr-1.63

)*-1.58

 -1.58

however-1.57

phasis-1.57

*/-1.55

INTERVAL 2.571 - 2.857
CONTAINS 0.000%

last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Do
TokenDo
Feature activation+1.510
zens
Tokenzens
Feature activation+2.857
of
Token of
Feature activation+0.857
workers
Token workers
Feature activation+0.756
at
Token at
Feature activation+0.905
Japan
Token Japan
Feature activation+0.455
's
Token's
Feature activation+0.728
er
Tokener
Feature activation+0.000
War
Token War
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.125
Do
TokenDo
Feature activation+1.428
zens
Tokenzens
Feature activation+2.612
of
Token of
Feature activation+1.411
previously
Token previously
Feature activation+0.932
-
Token-
Feature activation+0.000
un
Tokenun
Feature activation+0.000
published
Tokenpublished
Feature activation+0.608
's
Token's
Feature activation+0.000
National
Token National
Feature activation+0.000
Day
Token Day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Thousands
TokenThousands
Feature activation+2.832
of
Token of
Feature activation+0.931
Catal
Token Catal
Feature activation+0.700
ans
Tokenans
Feature activation+1.404
have
Token have
Feature activation+1.055
rallied
Token rallied
Feature activation+1.574

INTERVAL 2.285 - 2.571
CONTAINS 0.000%

Games
Token Games
Feature activation+0.185
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
More
TokenMore
Feature activation+1.156
than
Token than
Feature activation+0.862
400
Token 400
Feature activation+2.291
people
Token people
Feature activation+1.183
have
Token have
Feature activation+1.097
been
Token been
Feature activation+0.644
detained
Token detained
Feature activation+1.098
in
Token in
Feature activation+0.645

INTERVAL 2.000 - 2.285
CONTAINS 0.000%

came
Token came
Feature activation+0.000
too
Token too
Feature activation+0.000
late
Token late
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+2.103
P
TokenP
Feature activation+0.350
ays
Tokenays
Feature activation+0.000
lip
Tokenlip
Feature activation+0.000
-
Token-
Feature activation+0.000
gate
Tokengate
Feature activation+0.000
running
Token running
Feature activation+0.000
high
Token high
Feature activation+0.000
"
Token"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.004
One
TokenOne
Feature activation+2.011
passenger
Token passenger
Feature activation+1.068
,
Token,
Feature activation+0.517
Mark
Token Mark
Feature activation+0.268
Shore
Token Shore
Feature activation+0.304
y
Tokeny
Feature activation+0.637
the
Token the
Feature activation+0.000
space
Token space
Feature activation+0.000
station
Token station
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.055
"
Token"
Feature activation+2.255
It
TokenIt
Feature activation+0.665
's
Token's
Feature activation+0.046
not
Token not
Feature activation+0.000
an
Token an
Feature activation+0.000
original
Token original
Feature activation+0.000
blast
Token blast
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+1.237
massive
Token massive
Feature activation+1.233
blast
Token blast
Feature activation+2.014
at
Token at
Feature activation+1.106
a
Token a
Feature activation+1.185
munitions
Token munitions
Feature activation+0.414
dump
Token dump
Feature activation+0.727
in
Token in
Feature activation+0.899
Islamic
Token Islamic
Feature activation+0.000
State
Token State
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Saudi
TokenSaudi
Feature activation+2.004
Arabia
Token Arabia
Feature activation+2.098
has
Token has
Feature activation+1.001
arrested
Token arrested
Feature activation+0.814
88
Token 88
Feature activation+1.288
people
Token people
Feature activation+0.326
accused
Token accused
Feature activation+0.223

INTERVAL 1.714 - 2.000
CONTAINS 0.000%

conference
Token conference
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Nick
TokenNick
Feature activation+1.221
Cle
Token Cle
Feature activation+1.220
gg
Tokengg
Feature activation+1.783
is
Token is
Feature activation+1.320
preparing
Token preparing
Feature activation+1.537
to
Token to
Feature activation+0.402
fight
Token fight
Feature activation+0.917
his
Token his
Feature activation+0.713
in
Token in
Feature activation+0.000
Hong
Token Hong
Feature activation+0.000
Kong
Token Kong
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+1.832
January
Token January
Feature activation+1.461
,
Token,
Feature activation+1.214
13
Token 13
Feature activation+1.376
groups
Token groups
Feature activation+0.833
from
Token from
Feature activation+0.763
step
Token step
Feature activation+0.000
forward
Token forward
Feature activation+0.000
"
Token"
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Twenty
TokenTwenty
Feature activation+1.883
-
Token-
Feature activation+0.687
five
Tokenfive
Feature activation+1.294
of
Token of
Feature activation+0.960
the
Token the
Feature activation+0.376
EU
Token EU
Feature activation+0.496
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Some
TokenSome
Feature activation+1.826
200
Token 200
Feature activation+2.257
,
Token,
Feature activation+0.709
000
Token000
Feature activation+1.796
people
Token people
Feature activation+1.259
have
Token have
Feature activation+1.106
protested
Token protested
Feature activation+1.468
in
Token in
Feature activation+1.011
the
Token the
Feature activation+1.113
,
Token,
Feature activation+0.000
000
Token000
Feature activation+0.000
signatures
Token signatures
Feature activation+0.113
Ċ
TokenĊ
Feature activation+0.491
Ċ
TokenĊ
Feature activation+0.991
Earlier
TokenEarlier
Feature activation+1.952
,
Token,
Feature activation+1.456
the
Token the
Feature activation+0.764
F
Token F
Feature activation+0.292
WS
TokenWS
Feature activation+0.301
said
Token said
Feature activation+0.732

INTERVAL 1.428 - 1.714
CONTAINS 0.001%

in
Token in
Feature activation+1.011
the
Token the
Feature activation+1.113
Brazilian
Token Brazilian
Feature activation+1.289
city
Token city
Feature activation+1.145
of
Token of
Feature activation+0.523
Rio
Token Rio
Feature activation+1.481
de
Token de
Feature activation+0.000
Janeiro
Token Janeiro
Feature activation+1.871
against
Token against
Feature activation+0.429
a
Token a
Feature activation+0.344
bill
Token bill
Feature activation+0.575
a
Token a
Feature activation+0.000
fresh
Token fresh
Feature activation+0.000
investigation
Token investigation
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+1.590
leak
Token leak
Feature activation+1.659
of
Token of
Feature activation+1.366
11
Token 11
Feature activation+1.228
million
Token million
Feature activation+0.855
documents
Token documents
Feature activation+1.073
Ċ
TokenĊ
Feature activation+0.000
Nick
TokenNick
Feature activation+1.221
Cle
Token Cle
Feature activation+1.220
gg
Tokengg
Feature activation+1.783
is
Token is
Feature activation+1.320
preparing
Token preparing
Feature activation+1.537
to
Token to
Feature activation+0.402
fight
Token fight
Feature activation+0.917
his
Token his
Feature activation+0.713
final
Token final
Feature activation+0.669
general
Token general
Feature activation+0.144
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
V
TokenV
Feature activation+0.962
ik
Tokenik
Feature activation+0.007
tor
Tokentor
Feature activation+0.633
Yanukovych
Token Yanukovych
Feature activation+1.609
has
Token has
Feature activation+0.841
vowed
Token vowed
Feature activation+1.085
to
Token to
Feature activation+0.817
fight
Token fight
Feature activation+0.517
for
Token for
Feature activation+0.409
es
Tokenes
Feature activation+0.882
at
Token at
Feature activation+1.454
Birmingham
Token Birmingham
Feature activation+0.884
City
Token City
Feature activation+0.046
Council
Token Council
Feature activation+0.427
say
Token say
Feature activation+1.541
£
Token £
Feature activation+0.593
600
Token600
Feature activation+0.409
m
Tokenm
Feature activation+0.296
of
Token of
Feature activation+0.255
savings
Token savings
Feature activation+0.104

INTERVAL 1.143 - 1.428
CONTAINS 0.001%

S
Token S
Feature activation+0.430
aj
Tokenaj
Feature activation+0.577
id
Tokenid
Feature activation+1.189
J
Token J
Feature activation+1.523
avid
Tokenavid
Feature activation+1.098
-
Token -
Feature activation+1.175
the
Token the
Feature activation+0.766
UK
Token UK
Feature activation+0.363
's
Token's
Feature activation+0.161
first
Token first
Feature activation+0.242
Asian
Token Asian
Feature activation+0.114
used
Token used
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Off
TokenOff
Feature activation+0.133
shore
Tokenshore
Feature activation+0.290
workers
Token workers
Feature activation+1.331
are
Token are
Feature activation+0.988
now
Token now
Feature activation+0.651
almost
Token almost
Feature activation+0.907
a
Token a
Feature activation+0.523
fifth
Token fifth
Feature activation+0.690
seen
Token seen
Feature activation+0.000
'
Token'
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
US
TokenUS
Feature activation+0.752
President
Token President
Feature activation+1.273
Donald
Token Donald
Feature activation+1.193
Trump
Token Trump
Feature activation+1.714
says
Token says
Feature activation+1.490
North
Token North
Feature activation+0.360
Korea
Token Korea
Feature activation+0.083
title
Token title
Feature activation+0.000
's
Token's
Feature activation+0.000
launch
Token launch
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.086
Watch
TokenWatch
Feature activation+1.416
Dogs
Token Dogs
Feature activation+1.287
'
Token'
Feature activation+2.048
publisher
Token publisher
Feature activation+1.303
has
Token has
Feature activation+0.941
apologised
Token apologised
Feature activation+0.687
his
Token his
Feature activation+0.000
Palestinian
Token Palestinian
Feature activation+0.000
counterpart
Token counterpart
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Israeli
TokenIsraeli
Feature activation+1.381
Prime
Token Prime
Feature activation+0.556
Minister
Token Minister
Feature activation+1.035
Benjamin
Token Benjamin
Feature activation+0.521
Netanyahu
Token Netanyahu
Feature activation+1.346
has
Token has
Feature activation+0.842

INTERVAL 0.857 - 1.143
CONTAINS 0.003%

day
Token day
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Av
TokenAv
Feature activation+0.699
iation
Tokeniation
Feature activation+1.034
fuel
Token fuel
Feature activation+0.946
has
Token has
Feature activation+0.499
been
Token been
Feature activation+1.035
sent
Token sent
Feature activation+0.849
to
Token to
Feature activation+0.484
Manchester
Token Manchester
Feature activation+0.684
,
Token,
Feature activation+1.214
13
Token 13
Feature activation+1.376
groups
Token groups
Feature activation+0.833
from
Token from
Feature activation+0.763
Hong
Token Hong
Feature activation+1.101
Kong
Token Kong
Feature activation+0.952
and
Token and
Feature activation+0.233
Taiwan
Token Taiwan
Feature activation+0.328
gathered
Token gathered
Feature activation+0.741
in
Token in
Feature activation+0.609
Tai
Token Tai
Feature activation+0.370
four
Token four
Feature activation+0.569
African
Token African
Feature activation+0.639
heads
Token heads
Feature activation+0.909
of
Token of
Feature activation+0.000
state
Token state
Feature activation+0.131
has
Token has
Feature activation+0.892
urged
Token urged
Feature activation+1.117
rebels
Token rebels
Feature activation+0.501
in
Token in
Feature activation+0.369
the
Token the
Feature activation+0.866
Democratic
Token Democratic
Feature activation+0.188
000
Token000
Feature activation+1.796
people
Token people
Feature activation+1.259
have
Token have
Feature activation+1.106
protested
Token protested
Feature activation+1.468
in
Token in
Feature activation+1.011
the
Token the
Feature activation+1.113
Brazilian
Token Brazilian
Feature activation+1.289
city
Token city
Feature activation+1.145
of
Token of
Feature activation+0.523
Rio
Token Rio
Feature activation+1.481
de
Token de
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Do
TokenDo
Feature activation+1.510
zens
Tokenzens
Feature activation+2.857
of
Token of
Feature activation+0.857
workers
Token workers
Feature activation+0.756
at
Token at
Feature activation+0.905
Japan
Token Japan
Feature activation+0.455
's
Token's
Feature activation+0.728
Ky
Token Ky
Feature activation+0.000
ush
Tokenush
Feature activation+0.000
u
Tokenu
Feature activation+0.325

INTERVAL 0.571 - 0.857
CONTAINS 0.006%

of
Token of
Feature activation+0.633
to
Token to
Feature activation+0.081
pless
Tokenpless
Feature activation+0.279
photographs
Token photographs
Feature activation+0.775
of
Token of
Feature activation+0.945
the
Token the
Feature activation+0.799
Duchess
Token Duchess
Feature activation+0.541
of
Token of
Feature activation+0.000
Cambridge
Token Cambridge
Feature activation+1.529
.
Token.
Feature activation+0.470
Ċ
TokenĊ
Feature activation+0.000
have
Token have
Feature activation+1.283
been
Token been
Feature activation+1.063
paid
Token paid
Feature activation+1.570
to
Token to
Feature activation+1.550
agony
Token agony
Feature activation+0.334
aunt
Token aunt
Feature activation+0.736
Claire
Token Claire
Feature activation+0.068
Ray
Token Ray
Feature activation+1.106
ner
Tokenner
Feature activation+1.468
,
Token,
Feature activation+0.678
who
Token who
Feature activation+0.302
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
A
TokenA
Feature activation+1.209
small
Token small
Feature activation+0.842
Italian
Token Italian
Feature activation+0.802
town
Token town
Feature activation+0.806
is
Token is
Feature activation+0.569
transforming
Token transforming
Feature activation+0.761
its
Token its
Feature activation+0.000
main
Token main
Feature activation+0.140
square
Token square
Feature activation+0.503
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Thousands
TokenThousands
Feature activation+2.182
of
Token of
Feature activation+0.347
people
Token people
Feature activation+1.109
are
Token are
Feature activation+0.629
reported
Token reported
Feature activation+1.077
to
Token to
Feature activation+0.533
be
Token be
Feature activation+1.037
staying
Token staying
Feature activation+0.826
out
Token out
Feature activation+0.163
the
Token the
Feature activation+0.340
funer
Token funer
Feature activation+0.557
als
Tokenals
Feature activation+0.610
of
Token of
Feature activation+0.842
Russian
Token Russian
Feature activation+0.300
troops
Token troops
Feature activation+0.604
who
Token who
Feature activation+0.427
fought
Token fought
Feature activation+0.287
alongside
Token alongside
Feature activation+0.262
pro
Token pro
Feature activation+0.204
-
Token-
Feature activation+0.157

INTERVAL 0.286 - 0.571
CONTAINS 0.010%

search
Token search
Feature activation+0.411
for
Token for
Feature activation+0.817
the
Token the
Feature activation+0.755
plane
Token plane
Feature activation+0.624
in
Token in
Feature activation+0.561
the
Token the
Feature activation+0.558
southern
Token southern
Feature activation+0.480
Indian
Token Indian
Feature activation+0.653
Ocean
Token Ocean
Feature activation+0.463
,
Token,
Feature activation+0.378
said
Token said
Feature activation+0.960
child
Token child
Feature activation+0.000
sexual
Token sexual
Feature activation+0.000
exploitation
Token exploitation
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
town
Token town
Feature activation+0.466
.
Token.
Feature activation+0.420
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Martin
TokenMartin
Feature activation+0.000
Kimber
Token Kimber
Feature activation+0.000
his
Token his
Feature activation+0.000
cargo
Token cargo
Feature activation+0.000
."
Token."
Feature activation+0.899
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.152
The
TokenThe
Feature activation+0.425
map
Token map
Feature activation+0.307
,
Token,
Feature activation+0.370
which
Token which
Feature activation+0.137
has
Token has
Feature activation+0.092
been
Token been
Feature activation+0.193
ihadi
Tokenihadi
Feature activation+0.000
ya
Tokenya
Feature activation+0.000
,
Token,
Feature activation+0.260
the
Token the
Feature activation+0.286
presidential
Token presidential
Feature activation+0.705
palace
Token palace
Feature activation+0.465
in
Token in
Feature activation+0.086
Cairo
Token Cairo
Feature activation+0.345
's
Token's
Feature activation+0.122
well
Token well
Feature activation+0.185
-
Token-
Feature activation+0.108
marched
Token marched
Feature activation+1.488
in
Token in
Feature activation+1.092
the
Token the
Feature activation+1.358
centre
Token centre
Feature activation+1.255
of
Token of
Feature activation+0.783
the
Token the
Feature activation+0.553
capital
Token capital
Feature activation+0.746
,
Token,
Feature activation+0.808
Sof
Token Sof
Feature activation+0.912
ia
Tokenia
Feature activation+0.588
,
Token,
Feature activation+0.951

INTERVAL 0.000 - 0.286
CONTAINS 99.978%

What
TokenWhat
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
M
Token M
Feature activation+0.000
GH
TokenGH
Feature activation+0.000
doing
Token doing
Feature activation+0.000
in
Token in
Feature activation+0.000
response
Token response
Feature activation+0.000
to
Token to
Feature activation+0.000
Ebola
Token Ebola
Feature activation+0.000
?
Token?
Feature activation+0.000
used
Token used
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
San
Token San
Feature activation+0.000
Bernardino
Token Bernardino
Feature activation+0.000
shootings
Token shootings
Feature activation+0.000
last
Token last
Feature activation+0.000
week
Token week
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
and
Token and
Feature activation+0.000
counsel
Token counsel
Feature activation+0.000
on
Token on
Feature activation+0.000
complex
Token complex
Feature activation+0.000
global
Token global
Feature activation+0.000
issues
Token issues
Feature activation+0.000
.
Token.
Feature activation+0.000
What
Token What
Feature activation+0.000
is
Token is
Feature activation+0.000
unusual
Token unusual
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
talks
Token talks
Feature activation+0.000
during
Token during
Feature activation+0.000
a
Token a
Feature activation+0.000
summit
Token summit
Feature activation+0.000
with
Token with
Feature activation+0.000
U
Token U
Feature activation+0.000
.
Token.
Feature activation+0.000
S
TokenS
Feature activation+0.000
.
Token.
Feature activation+0.000
President
Token President
Feature activation+0.000
to
Tokento
Feature activation+0.000
high
Token high
Feature activation+0.000
productivity
Token productivity
Feature activation+0.000
jobs
Token jobs
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
San
Token San
Feature activation+0.000
Francisco
Token Francisco
Feature activation+0.000
Bay
Token Bay
Feature activation+0.000
Area
Token Area
Feature activation+0.000
.
Token.
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 22: In phrase starting with “ poster”

TOP ACTIVATIONS
MAX = 7.200

âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.361
of
Token of
Feature activation+7.200
Barack
Token Barack
Feature activation+1.724
Obama
Token Obama
Feature activation+2.120
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+2.620
positively
Token positively
Feature activation+0.000
in
Token in
Feature activation+0.000
being
Token being
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.757
boy
Token boy
Feature activation+6.127
for
Token for
Feature activation+5.613
the
Token the
Feature activation+5.381
Navy
Token Navy
Feature activation+1.023
's
Token's
Feature activation+0.841
recent
Token recent
Feature activation+0.783
claim
Token claim
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+1.157
boy
Token boy
Feature activation+6.055
of
Token of
Feature activation+4.714
the
Token the
Feature activation+4.379
working
Token working
Feature activation+0.000
class
Token class
Feature activation+0.000
?
Token?
Feature activation+0.000
great
Token great
Feature activation+0.000
ful
Tokenful
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
for
Token for
Feature activation+6.034
allowing
Token allowing
Feature activation+0.000
us
Token us
Feature activation+0.000
to
Token to
Feature activation+0.000
see
Token see
Feature activation+0.000
this
Token this
Feature activation+0.000
our
Token our
Feature activation+0.000
City
Token City
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+5.900
for
Token for
Feature activation+5.783
this
Token this
Feature activation+4.410
is
Token is
Feature activation+2.433
the
Token the
Feature activation+1.305
recently
Token recently
Feature activation+0.000
City
Token City
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+5.900
for
Token for
Feature activation+5.783
this
Token this
Feature activation+4.410
is
Token is
Feature activation+2.433
the
Token the
Feature activation+1.305
recently
Token recently
Feature activation+0.000
approved
Token approved
Feature activation+0.000
or
Tokenor
Feature activation+0.000
Walker
Token Walker
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.412
boy
Token boy
Feature activation+5.702
for
Token for
Feature activation+5.131
the
Token the
Feature activation+4.434
recall
Token recall
Feature activation+0.000
and
Token and
Feature activation+0.000
repeal
Token repeal
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.786
-
Token-
Feature activation+2.137
child
Tokenchild
Feature activation+4.264
for
Token for
Feature activation+5.690
human
Token human
Feature activation+1.912
hub
Token hub
Feature activation+0.000
ris
Tokenris
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
in
Token in
Feature activation+0.000
being
Token being
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.757
boy
Token boy
Feature activation+6.127
for
Token for
Feature activation+5.613
the
Token the
Feature activation+5.381
Navy
Token Navy
Feature activation+1.023
's
Token's
Feature activation+0.841
recent
Token recent
Feature activation+0.783
policy
Token policy
Feature activation+0.000
being
Token being
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.757
boy
Token boy
Feature activation+6.127
for
Token for
Feature activation+5.613
the
Token the
Feature activation+5.381
Navy
Token Navy
Feature activation+1.023
's
Token's
Feature activation+0.841
recent
Token recent
Feature activation+0.783
policy
Token policy
Feature activation+0.000
change
Token change
Feature activation+0.000
Guns
Token Guns
Feature activation+0.000
have
Token have
Feature activation+0.000
become
Token become
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.801
child
Token child
Feature activation+5.132
of
Token of
Feature activation+4.263
American
Token American
Feature activation+0.981
individual
Token individual
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
,
Token,
Feature activation+0.000
Walker
Token Walker
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.412
boy
Token boy
Feature activation+5.702
for
Token for
Feature activation+5.131
the
Token the
Feature activation+4.434
recall
Token recall
Feature activation+0.000
and
Token and
Feature activation+0.000
repeal
Token repeal
Feature activation+0.000
movement
Token movement
Feature activation+0.000
him
Token him
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+4.810
for
Token for
Feature activation+5.079
their
Token their
Feature activation+3.005
organization
Token organization
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Let
Token Let
Feature activation+0.000
o
Tokeno
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.553
boy
Token boy
Feature activation+5.024
when
Token when
Feature activation+2.187
it
Token it
Feature activation+1.706
comes
Token comes
Feature activation+1.789
to
Token to
Feature activation+3.752
long
Token long
Feature activation+0.000
g
Token g
Feature activation+0.000
aped
Tokenaped
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
for
Token for
Feature activation+4.968
10
Token 10
Feature activation+0.346
minutes
Token minutes
Feature activation+0.000
.
Token.
Feature activation+0.000
Then
Token Then
Feature activation+0.000
he
Token he
Feature activation+0.000
ide
Tokenide
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
case
Token case
Feature activation+0.000
for
Token for
Feature activation+4.890
government
Token government
Feature activation+0.000
-
Token-
Feature activation+0.000
backed
Tokenbacked
Feature activation+0.000
terrorism
Token terrorism
Feature activation+0.000
against
Token against
Feature activation+0.000
who
Token who
Feature activation+0.000
is
Token is
Feature activation+0.000
another
Token another
Feature activation+0.000
poster
Token poster
Feature activation+0.000
child
Token child
Feature activation+4.361
for
Token for
Feature activation+4.821
progressive
Token progressive
Feature activation+0.232
Democrats
Token Democrats
Feature activation+0.000
,
Token,
Feature activation+0.000
no
Token no
Feature activation+0.000
difference
Token difference
Feature activation+0.000
"
Token"
Feature activation+0.000
him
Token him
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+4.810
for
Token for
Feature activation+5.079
their
Token their
Feature activation+3.005
organization
Token organization
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
definitely
Token definitely
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+4.799
for
Token for
Feature activation+4.328
curly
Token curly
Feature activation+0.000
hair
Token hair
Feature activation+0.000
.
Token.
Feature activation+0.000
Su
Token Su
Feature activation+0.000
s
Tokens
Feature activation+0.000
been
Token been
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+2.627
-
Token-
Feature activation+2.743
boy
Tokenboy
Feature activation+4.788
for
Token for
Feature activation+3.794
under
Token under
Feature activation+0.000
-
Token-
Feature activation+0.000
performance
Tokenperformance
Feature activation+0.000
.
Token.
Feature activation+0.000

Top DFA by src position
MAX = 9.477

it
Token it
Feature activation+0.037
âĢ
TokenâĢ
Feature activation+0.056
Ļ
TokenĻ
Feature activation+0.006
s
Tokens
Feature activation+0.143
the
Token the
Feature activation+0.050
poster
Token poster
Feature activation+9.477
of
Token of
Feature activation+0.339
Barack
Token Barack
Feature activation+0.000
Obama
Token Obama
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
so
Token so
Feature activation-0.020
positively
Token positively
Feature activation+0.029
in
Token in
Feature activation-0.007
being
Token being
Feature activation+0.071
the
Token the
Feature activation+0.342
poster
Token poster
Feature activation+8.701
boy
Token boy
Feature activation+0.191
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
Navy
Token Navy
Feature activation+0.000
's
Token's
Feature activation+0.000
apartment
Token apartment
Feature activation+0.016
claim
Token claim
Feature activation-0.107
to
Token to
Feature activation+0.023
be
Token be
Feature activation+0.372
a
Token a
Feature activation+0.419
poster
Token poster
Feature activation+7.862
boy
Token boy
Feature activation+0.430
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
working
Token working
Feature activation+0.000
class
Token class
Feature activation+0.000
really
Token really
Feature activation-0.053
great
Token great
Feature activation+0.141
ful
Tokenful
Feature activation-0.057
to
Token to
Feature activation+0.187
the
Token the
Feature activation+0.105
poster
Token poster
Feature activation+8.436
for
Token for
Feature activation+0.312
allowing
Token allowing
Feature activation+0.000
us
Token us
Feature activation+0.000
to
Token to
Feature activation+0.000
see
Token see
Feature activation+0.000
into
Token into
Feature activation-0.000
our
Token our
Feature activation+0.007
City
Token City
Feature activation+0.005
.
Token.
Feature activation+0.018
The
Token The
Feature activation+0.069
poster
Token poster
Feature activation+8.630
boy
Token boy
Feature activation+0.113
for
Token for
Feature activation+0.000
this
Token this
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
into
Token into
Feature activation-0.001
our
Token our
Feature activation-0.005
City
Token City
Feature activation+0.011
.
Token.
Feature activation+0.017
The
Token The
Feature activation+0.046
poster
Token poster
Feature activation+8.283
boy
Token boy
Feature activation+0.799
for
Token for
Feature activation-0.359
this
Token this
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
Govern
TokenGovern
Feature activation-0.013
or
Tokenor
Feature activation-0.015
Walker
Token Walker
Feature activation+0.009
is
Token is
Feature activation-0.028
the
Token the
Feature activation+0.151
poster
Token poster
Feature activation+8.647
boy
Token boy
Feature activation-0.017
for
Token for
Feature activation+0.000
the
Token the
Feature activation+0.000
recall
Token recall
Feature activation+0.000
and
Token and
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.007
s
Tokens
Feature activation+0.030
sinking
Token sinking
Feature activation-0.043
is
Token is
Feature activation+0.129
the
Token the
Feature activation+0.201
poster
Token poster
Feature activation+8.027
-
Token-
Feature activation+0.111
child
Tokenchild
Feature activation+0.493
for
Token for
Feature activation-0.368
human
Token human
Feature activation+0.000
hub
Token hub
Feature activation+0.000
so
Token so
Feature activation-0.055
positively
Token positively
Feature activation+0.086
in
Token in
Feature activation+0.082
being
Token being
Feature activation+0.083
the
Token the
Feature activation+0.198
poster
Token poster
Feature activation+7.657
boy
Token boy
Feature activation+0.889
for
Token for
Feature activation-0.376
the
Token the
Feature activation+0.000
Navy
Token Navy
Feature activation+0.000
's
Token's
Feature activation+0.000
so
Token so
Feature activation+0.001
positively
Token positively
Feature activation+0.028
in
Token in
Feature activation+0.093
being
Token being
Feature activation+0.119
the
Token the
Feature activation+0.150
poster
Token poster
Feature activation+6.799
boy
Token boy
Feature activation+0.892
for
Token for
Feature activation+0.305
the
Token the
Feature activation-0.107
Navy
Token Navy
Feature activation+0.000
's
Token's
Feature activation+0.000
.
Token.
Feature activation-0.010
Guns
Token Guns
Feature activation-0.028
have
Token have
Feature activation+0.102
become
Token become
Feature activation+0.245
the
Token the
Feature activation+0.185
poster
Token poster
Feature activation+7.535
child
Token child
Feature activation+0.045
of
Token of
Feature activation+0.000
American
Token American
Feature activation+0.000
individual
Token individual
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
Govern
TokenGovern
Feature activation-0.017
or
Tokenor
Feature activation-0.032
Walker
Token Walker
Feature activation+0.005
is
Token is
Feature activation+0.049
the
Token the
Feature activation+0.165
poster
Token poster
Feature activation+7.926
boy
Token boy
Feature activation+0.844
for
Token for
Feature activation-0.640
the
Token the
Feature activation+0.000
recall
Token recall
Feature activation+0.000
and
Token and
Feature activation+0.000
recruit
Token recruit
Feature activation-0.104
"
Token"
Feature activation+0.012
him
Token him
Feature activation+0.151
as
Token as
Feature activation+0.373
a
Token a
Feature activation+0.250
poster
Token poster
Feature activation+7.438
boy
Token boy
Feature activation+0.480
for
Token for
Feature activation+0.079
their
Token their
Feature activation+0.000
organization
Token organization
Feature activation+0.000
.
Token.
Feature activation+0.000
ared
Tokenared
Feature activation-0.089
Let
Token Let
Feature activation-0.010
o
Tokeno
Feature activation-0.023
is
Token is
Feature activation+0.091
the
Token the
Feature activation+0.088
poster
Token poster
Feature activation+8.736
boy
Token boy
Feature activation+0.081
when
Token when
Feature activation+0.000
it
Token it
Feature activation+0.000
comes
Token comes
Feature activation+0.000
to
Token to
Feature activation+0.000
and
Token and
Feature activation-0.037
g
Token g
Feature activation-0.004
aped
Tokenaped
Feature activation-0.208
at
Token at
Feature activation-0.168
the
Token the
Feature activation-0.052
poster
Token poster
Feature activation+8.149
for
Token for
Feature activation+0.376
10
Token 10
Feature activation+0.000
minutes
Token minutes
Feature activation+0.000
.
Token.
Feature activation+0.000
Then
Token Then
Feature activation+0.000
onder
Tokenonder
Feature activation+0.001
he
Tokenhe
Feature activation-0.003
ide
Tokenide
Feature activation-0.016
is
Token is
Feature activation+0.218
a
Token a
Feature activation+0.285
poster
Token poster
Feature activation+7.841
case
Token case
Feature activation+0.599
for
Token for
Feature activation-0.068
government
Token government
Feature activation+0.000
-
Token-
Feature activation+0.000
backed
Tokenbacked
Feature activation+0.000
Patrick
Token Patrick
Feature activation+0.006
,
Token,
Feature activation+0.094
who
Token who
Feature activation+0.018
is
Token is
Feature activation-0.110
another
Token another
Feature activation-0.017
poster
Token poster
Feature activation+7.430
child
Token child
Feature activation+0.416
for
Token for
Feature activation-0.097
progressive
Token progressive
Feature activation+0.000
Democrats
Token Democrats
Feature activation+0.000
,
Token,
Feature activation+0.000
recruit
Token recruit
Feature activation-0.026
"
Token"
Feature activation-0.052
him
Token him
Feature activation+0.082
as
Token as
Feature activation+0.537
a
Token a
Feature activation+0.352
poster
Token poster
Feature activation+7.166
boy
Token boy
Feature activation+0.342
for
Token for
Feature activation+0.000
their
Token their
Feature activation+0.000
organization
Token organization
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.115
Ļ
TokenĻ
Feature activation+0.012
s
Tokens
Feature activation+0.073
definitely
Token definitely
Feature activation+0.142
the
Token the
Feature activation+0.240
poster
Token poster
Feature activation+7.685
boy
Token boy
Feature activation+0.138
for
Token for
Feature activation+0.000
curly
Token curly
Feature activation+0.000
hair
Token hair
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.042
Ļ
TokenĻ
Feature activation+0.001
s
Tokens
Feature activation+0.118
been
Token been
Feature activation+0.271
the
Token the
Feature activation+0.067
poster
Token poster
Feature activation+7.185
-
Token-
Feature activation+0.142
boy
Tokenboy
Feature activation+0.056
for
Token for
Feature activation+0.000
under
Token under
Feature activation+0.000
-
Token-
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.03

Head 2: 0.08

Head 3: 0.08

Head 4: 0.10

Head 5: 0.03

Head 6: 0.07

Head 7: 0.37

Head 8: 0.04

Head 9: 0.04

Head 10: 0.07

Head 11: 0.08

Positive logits

Mothers1.50

corrid1.47

Rising1.44

posters1.42

hovah1.41

Beware1.38

dangers1.36

Hayden1.34

poster1.33

prestigious1.32

Investor1.32

gress1.30

Ultra1.28

apego1.28

Peak1.28

intolerance1.27

Spectre1.27

dissenting1.25

Vanguard1.24

Failure1.24

Negative logits

itars-1.73

-1.73

ources-1.71

hya-1.70

inations-1.69

quartered-1.58

ilage-1.48

experien-1.47

ead-1.46

displayText-1.45

-1.44

*/(-1.43

extrad-1.41

footing-1.39

JS-1.38

merce-1.38

ruct-1.37

NetMessage-1.36

ource-1.36

fing-1.36

INTERVAL 6.480 - 7.200
CONTAINS 0.000%

âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.361
of
Token of
Feature activation+7.200
Barack
Token Barack
Feature activation+1.724
Obama
Token Obama
Feature activation+2.120
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+2.620

INTERVAL 5.760 - 6.480
CONTAINS 0.000%

our
Token our
Feature activation+0.000
City
Token City
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+5.900
for
Token for
Feature activation+5.783
this
Token this
Feature activation+4.410
is
Token is
Feature activation+2.433
the
Token the
Feature activation+1.305
recently
Token recently
Feature activation+0.000
City
Token City
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+5.900
for
Token for
Feature activation+5.783
this
Token this
Feature activation+4.410
is
Token is
Feature activation+2.433
the
Token the
Feature activation+1.305
recently
Token recently
Feature activation+0.000
approved
Token approved
Feature activation+0.000
great
Token great
Feature activation+0.000
ful
Tokenful
Feature activation+0.000
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
for
Token for
Feature activation+6.034
allowing
Token allowing
Feature activation+0.000
us
Token us
Feature activation+0.000
to
Token to
Feature activation+0.000
see
Token see
Feature activation+0.000
this
Token this
Feature activation+0.000
claim
Token claim
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+1.157
boy
Token boy
Feature activation+6.055
of
Token of
Feature activation+4.714
the
Token the
Feature activation+4.379
working
Token working
Feature activation+0.000
class
Token class
Feature activation+0.000
?
Token?
Feature activation+0.000
positively
Token positively
Feature activation+0.000
in
Token in
Feature activation+0.000
being
Token being
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.757
boy
Token boy
Feature activation+6.127
for
Token for
Feature activation+5.613
the
Token the
Feature activation+5.381
Navy
Token Navy
Feature activation+1.023
's
Token's
Feature activation+0.841
recent
Token recent
Feature activation+0.783

INTERVAL 5.040 - 5.760
CONTAINS 0.000%

being
Token being
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.757
boy
Token boy
Feature activation+6.127
for
Token for
Feature activation+5.613
the
Token the
Feature activation+5.381
Navy
Token Navy
Feature activation+1.023
's
Token's
Feature activation+0.841
recent
Token recent
Feature activation+0.783
policy
Token policy
Feature activation+0.000
change
Token change
Feature activation+0.000
him
Token him
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+4.810
for
Token for
Feature activation+5.079
their
Token their
Feature activation+3.005
organization
Token organization
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Guns
Token Guns
Feature activation+0.000
have
Token have
Feature activation+0.000
become
Token become
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.801
child
Token child
Feature activation+5.132
of
Token of
Feature activation+4.263
American
Token American
Feature activation+0.981
individual
Token individual
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
,
Token,
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.786
-
Token-
Feature activation+2.137
child
Tokenchild
Feature activation+4.264
for
Token for
Feature activation+5.690
human
Token human
Feature activation+1.912
hub
Token hub
Feature activation+0.000
ris
Tokenris
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
or
Tokenor
Feature activation+0.000
Walker
Token Walker
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.412
boy
Token boy
Feature activation+5.702
for
Token for
Feature activation+5.131
the
Token the
Feature activation+4.434
recall
Token recall
Feature activation+0.000
and
Token and
Feature activation+0.000
repeal
Token repeal
Feature activation+0.000

INTERVAL 4.320 - 5.040
CONTAINS 0.000%

<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
the
Token the
Feature activation+0.000
prime
Token prime
Feature activation+0.000
poster
Token poster
Feature activation+0.644
child
Token child
Feature activation+2.690
for
Token for
Feature activation+4.718
government
Token government
Feature activation+1.746
contracts
Token contracts
Feature activation+0.000
spun
Token spun
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+1.015
-
Token-
Feature activation+3.886
game
Tokengame
Feature activation+0.746
for
Token for
Feature activation+4.071
a
Token a
Feature activation+4.325
greater
Token greater
Feature activation+0.542
underlying
Token underlying
Feature activation+0.000
trend
Token trend
Feature activation+0.000
of
Token of
Feature activation+0.000
a
Token a
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+1.157
boy
Token boy
Feature activation+6.055
of
Token of
Feature activation+4.714
the
Token the
Feature activation+4.379
working
Token working
Feature activation+0.000
class
Token class
Feature activation+0.000
?
Token?
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+5.900
for
Token for
Feature activation+5.783
this
Token this
Feature activation+4.410
is
Token is
Feature activation+2.433
the
Token the
Feature activation+1.305
recently
Token recently
Feature activation+0.000
approved
Token approved
Feature activation+0.000
Santa
Token Santa
Feature activation+0.000
g
Token g
Feature activation+0.000
aped
Tokenaped
Feature activation+0.000
at
Token at
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
for
Token for
Feature activation+4.968
10
Token 10
Feature activation+0.346
minutes
Token minutes
Feature activation+0.000
.
Token.
Feature activation+0.000
Then
Token Then
Feature activation+0.000
he
Token he
Feature activation+0.000

INTERVAL 3.600 - 4.320
CONTAINS 0.000%

even
Token even
Feature activation+0.000
hanging
Token hanging
Feature activation+0.000
up
Token up
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
of
Token of
Feature activation+3.908
the
Token the
Feature activation+2.766
Prince
Token Prince
Feature activation+0.000
on
Token on
Feature activation+0.000
his
Token his
Feature activation+0.000
bedroom
Token bedroom
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
to
Token to
Feature activation+1.645
the
Token the
Feature activation+4.240
current
Token current
Feature activation+0.468
campaign
Token campaign
Feature activation+0.000
for
Token for
Feature activation+2.228
Furious
Token Furious
Feature activation+0.000
6
Token 6
Feature activation+0.000
been
Token been
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+2.627
-
Token-
Feature activation+2.743
boy
Tokenboy
Feature activation+4.788
for
Token for
Feature activation+3.794
under
Token under
Feature activation+0.000
-
Token-
Feature activation+0.000
performance
Tokenperformance
Feature activation+0.000
.
Token.
Feature activation+0.000
Yes
Token Yes
Feature activation+0.000
ich
Tokenich
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
virtual
Token virtual
Feature activation+0.000
poster
Token poster
Feature activation+0.000
child
Token child
Feature activation+3.835
for
Token for
Feature activation+4.422
political
Token political
Feature activation+0.213
corruption
Token corruption
Feature activation+0.000
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
has
Token has
Feature activation+0.000
called
Token called
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
child
Token child
Feature activation+3.116
for
Token for
Feature activation+4.086
cr
Token cr
Feature activation+0.000
ony
Tokenony
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
and
Token and
Feature activation+0.000
corporate
Token corporate
Feature activation+0.000

INTERVAL 2.880 - 3.600
CONTAINS 0.000%

David
Token David
Feature activation+0.000
Brent
Token Brent
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+3.114
for
Token for
Feature activation+3.878
management
Token management
Feature activation+0.000
jargon
Token jargon
Feature activation+0.000
.
Token.
Feature activation+0.000
Ricky
Token Ricky
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
boy
Token boy
Feature activation+4.810
for
Token for
Feature activation+5.079
their
Token their
Feature activation+3.005
organization
Token organization
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
He
TokenHe
Feature activation+0.000
he
Token he
Feature activation+0.000
has
Token has
Feature activation+0.000
called
Token called
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+0.000
child
Token child
Feature activation+3.116
for
Token for
Feature activation+4.086
cr
Token cr
Feature activation+0.000
ony
Tokenony
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
and
Token and
Feature activation+0.000
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
first
Token first
Feature activation+0.000
poster
Token poster
Feature activation+0.000
for
Token for
Feature activation+4.450
the
Token the
Feature activation+3.405
follow
Token follow
Feature activation+0.000
-
Token-
Feature activation+0.000
up
Tokenup
Feature activation+0.000
arrived
Token arrived
Feature activation+0.000
earlier
Token earlier
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
global
Token global
Feature activation+0.000
poster
Token poster
Feature activation+0.000
children
Token children
Feature activation+3.750
for
Token for
Feature activation+3.085
how
Token how
Feature activation+1.425
outrage
Token outrage
Feature activation+0.000
ously
Tokenously
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 2.160 - 2.880
CONTAINS 0.000%

seless
Tokenseless
Feature activation+0.000
.
Token.
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
p
Tokenp
Feature activation+0.000
oster
Tokenoster
Feature activation+0.000
="
Token="
Feature activation+2.441
http
Tokenhttp
Feature activation+0.373
://
Token://
Feature activation+1.481
v
Tokenv
Feature activation+0.000
.
Token.
Feature activation+0.000
polit
Tokenpolit
Feature activation+0.000
had
Token had
Feature activation+0.000
an
Token an
Feature activation+0.000
enormous
Token enormous
Feature activation+0.000
poster
Token poster
Feature activation+0.000
of
Token of
Feature activation+3.172
a
Token a
Feature activation+2.180
still
Token still
Feature activation+0.000
shot
Token shot
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
film
Token film
Feature activation+0.000
was
Token was
Feature activation+0.000
eventually
Token eventually
Feature activation+0.000
found
Token found
Feature activation+0.000
after
Token after
Feature activation+0.000
posters
Token posters
Feature activation+0.000
showing
Token showing
Feature activation+2.602
a
Token a
Feature activation+0.000
photograph
Token photograph
Feature activation+0.000
of
Token of
Feature activation+0.000
her
Token her
Feature activation+0.000
led
Token led
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
cut
Token cut
Feature activation+0.000
away
Tokenaway
Feature activation+0.000
poster
Token poster
Feature activation+0.000
of
Token of
Feature activation+2.626
the
Token the
Feature activation+2.305
Mart
Token Mart
Feature activation+0.000
ini
Tokenini
Feature activation+0.000
9
Token 9
Feature activation+0.000
35
Token35
Feature activation+0.000
o
Tokeno
Feature activation+0.000
is
Token is
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.553
boy
Token boy
Feature activation+5.024
when
Token when
Feature activation+2.187
it
Token it
Feature activation+1.706
comes
Token comes
Feature activation+1.789
to
Token to
Feature activation+3.752
long
Token long
Feature activation+0.000
hairst
Token hairst
Feature activation+0.000

INTERVAL 1.440 - 2.160
CONTAINS 0.000%

Step
Token Step
Feature activation+0.000
an
Tokenan
Feature activation+0.000
B
Token B
Feature activation+0.000
ander
Tokenander
Feature activation+0.000
a
Tokena
Feature activation+0.000
,
Token,
Feature activation+1.580
the
Token the
Feature activation+0.000
Uk
Token Uk
Feature activation+0.000
ran
Tokenran
Feature activation+0.000
ian
Tokenian
Feature activation+0.000
WWII
Token WWII
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
theater
Token theater
Feature activation+0.000
are
Token are
Feature activation+0.000
for
Token for
Feature activation+2.361
the
Token the
Feature activation+1.446
movie
Token movie
Feature activation+0.000
"
Token "
Feature activation+0.000
Sch
TokenSch
Feature activation+0.000
lock
Tokenlock
Feature activation+0.000
,"
Token,"
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
p
Tokenp
Feature activation+0.000
oster
Tokenoster
Feature activation+0.000
="
Token="
Feature activation+2.441
http
Tokenhttp
Feature activation+0.373
://
Token://
Feature activation+1.481
v
Tokenv
Feature activation+0.000
.
Token.
Feature activation+0.000
polit
Tokenpolit
Feature activation+0.000
ico
Tokenico
Feature activation+0.000
.
Token.
Feature activation+0.000
the
Token the
Feature activation+0.000
prime
Token prime
Feature activation+0.000
poster
Token poster
Feature activation+0.644
child
Token child
Feature activation+2.690
for
Token for
Feature activation+4.718
government
Token government
Feature activation+1.746
contracts
Token contracts
Feature activation+0.000
spun
Token spun
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
control
Token control
Feature activation+0.000
waving
Token waving
Feature activation+0.000
a
Token a
Feature activation+0.000
giant
Token giant
Feature activation+0.000
poster
Token poster
Feature activation+0.000
of
Token of
Feature activation+2.602
the
Token the
Feature activation+1.886
letter
Token letter
Feature activation+0.000
L
Token L
Feature activation+0.000
.
Token.
Feature activation+0.000
Six
Token Six
Feature activation+0.000
of
Token of
Feature activation+0.000

INTERVAL 0.720 - 1.440
CONTAINS 0.000%

one
Token one
Feature activation+0.000
.
Token.
Feature activation+0.000
Children
Token Children
Feature activation+0.000
had
Token had
Feature activation+0.000
posters
Token posters
Feature activation+0.000
of
Token of
Feature activation+0.767
Spit
Token Spit
Feature activation+0.000
fires
Tokenfires
Feature activation+0.000
up
Token up
Feature activation+0.000
on
Token on
Feature activation+0.000
their
Token their
Feature activation+0.000
become
Token become
Feature activation+0.000
the
Token the
Feature activation+0.000
poster
Token poster
Feature activation+0.801
child
Token child
Feature activation+5.132
of
Token of
Feature activation+4.263
American
Token American
Feature activation+0.981
individual
Token individual
Feature activation+0.000
ism
Tokenism
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
toxic
Token toxic
Feature activation+0.000
apartment
Token apartment
Feature activation+0.000
claim
Token claim
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
a
Token a
Feature activation+0.000
poster
Token poster
Feature activation+1.157
boy
Token boy
Feature activation+6.055
of
Token of
Feature activation+4.714
the
Token the
Feature activation+4.379
working
Token working
Feature activation+0.000
class
Token class
Feature activation+0.000
military
Token military
Feature activation+0.000
,
Token,
Feature activation+0.000
who
Token who
Feature activation+0.000
held
Token held
Feature activation+0.000
posters
Token posters
Feature activation+0.000
of
Token of
Feature activation+0.778
Defense
Token Defense
Feature activation+0.000
Minister
Token Minister
Feature activation+0.000
Abdel
Token Abdel
Feature activation+0.000
F
Token F
Feature activation+0.000
att
Tokenatt
Feature activation+0.000
poster
Token poster
Feature activation+1.757
boy
Token boy
Feature activation+6.127
for
Token for
Feature activation+5.613
the
Token the
Feature activation+5.381
Navy
Token Navy
Feature activation+1.023
's
Token's
Feature activation+0.841
recent
Token recent
Feature activation+0.783
policy
Token policy
Feature activation+0.000
change
Token change
Feature activation+0.000
when
Token when
Feature activation+0.000
it
Token it
Feature activation+0.000

INTERVAL 0.000 - 0.720
CONTAINS 99.999%

with
Token with
Feature activation+0.000
113
Token 113
Feature activation+0.000
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
not
Token not
Feature activation+0.000
by
Token by
Feature activation+0.000
much
Token much
Feature activation+0.000
.
Token.
Feature activation+0.000
Stanford
Token Stanford
Feature activation+0.000
is
Token is
Feature activation+0.000
only
Token only
Feature activation+0.000
you
Token you
Feature activation+0.000
to
Token to
Feature activation+0.000
enter
Token enter
Feature activation+0.000
your
Token your
Feature activation+0.000
e
Token e
Feature activation+0.000
-
Token-
Feature activation+0.000
mail
Tokenmail
Feature activation+0.000
address
Token address
Feature activation+0.000
and
Token and
Feature activation+0.000
password
Token password
Feature activation+0.000
to
Token to
Feature activation+0.000
series
Token series
Feature activation+0.000
sponsor
Token sponsor
Feature activation+0.000
patches
Token patches
Feature activation+0.000
on
Token on
Feature activation+0.000
uniforms
Token uniforms
Feature activation+0.000
and
Token and
Feature activation+0.000
cars
Token cars
Feature activation+0.000
would
Token would
Feature activation+0.000
need
Token need
Feature activation+0.000
to
Token to
Feature activation+0.000
be
Token be
Feature activation+0.000
last
Token last
Feature activation+0.000
seven
Token seven
Feature activation+0.000
years
Token years
Feature activation+0.000
helping
Token helping
Feature activation+0.000
the
Token the
Feature activation+0.000
poor
Token poor
Feature activation+0.000
and
Token and
Feature activation+0.000
sick
Token sick
Feature activation+0.000
in
Token in
Feature activation+0.000
their
Token their
Feature activation+0.000
convent
Token convent
Feature activation+0.000
heart
Token heart
Feature activation+0.000
association
Token association
Feature activation+0.000
's
Token's
Feature activation+0.000
Circ
Token Circ
Feature activation+0.000
ulation
Tokenulation
Feature activation+0.000
journal
Token journal
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
European
Token European
Feature activation+0.000
Heart
Token Heart
Feature activation+0.000
Journal
Token Journal
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 23: Activates on “ way” in texts of the form “{verb} {pronoun} way”

TOP ACTIVATIONS
MAX = 3.464

X
Token X
Feature activation+0.000
PS
TokenPS
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
coming
Token coming
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+3.464
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
contrast
Token contrast
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
that
Token that
Feature activation+0.000
might
Token might
Feature activation+0.000
come
Token come
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+2.802
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
it
Token it
Feature activation+0.000
gives
Token gives
Feature activation+0.000
us
Token us
Feature activation+0.000
those
Token those
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
that
Token that
Feature activation+0.000
came
Token came
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+2.733
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Richardson
Token Richardson
Feature activation+0.000
said
Token said
Feature activation+0.000
whatever
Token whatever
Feature activation+0.000
criticism
Token criticism
Feature activation+0.000
is
Token is
Feature activation+0.000
thrown
Token thrown
Feature activation+0.000
its
Token its
Feature activation+0.000
way
Token way
Feature activation+2.626
.
Token.
Feature activation+0.000
Atlanta
Token Atlanta
Feature activation+0.000
may
Token may
Feature activation+0.000
have
Token have
Feature activation+0.000
lost
Token lost
Feature activation+0.000
or
Token or
Feature activation+0.000
avoiding
Token avoiding
Feature activation+0.000
damage
Token damage
Feature activation+0.000
coming
Token coming
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+2.623
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
can
Token can
Feature activation+0.000
help
Token help
Feature activation+0.000
clinch
Token clinch
Feature activation+0.000
,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
going
Token going
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+2.395
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
able
Token able
Feature activation+0.000
to
Token to
Feature activation+0.000
studio
Token studio
Feature activation+0.000
teaching
Token teaching
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
came
Token came
Feature activation+0.000
my
Token my
Feature activation+0.000
way
Token way
Feature activation+2.287
at
Token at
Feature activation+0.844
studios
Token studios
Feature activation+0.000
that
Token that
Feature activation+0.000
were
Token were
Feature activation+0.000
a
Token a
Feature activation+0.000
the
Token the
Feature activation+0.000
budget
Token budget
Feature activation+0.000
cuts
Token cuts
Feature activation+0.000
coming
Token coming
Feature activation+0.000
its
Token its
Feature activation+0.000
way
Token way
Feature activation+2.021
that
Token that
Feature activation+0.000
it
Token it
Feature activation+0.000
has
Token has
Feature activation+0.000
begun
Token begun
Feature activation+0.000
burning
Token burning
Feature activation+0.000
and
Token and
Feature activation+0.000
good
Token good
Feature activation+0.000
changes
Token changes
Feature activation+0.000
coming
Token coming
Feature activation+0.000
her
Token her
Feature activation+0.000
way
Token way
Feature activation+1.982
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
all
Token all
Feature activation+0.000
that
Token that
Feature activation+0.000
have
Token have
Feature activation+0.000
come
Token come
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.925
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Step
TokenStep
Feature activation+0.000
h
Tokenh
Feature activation+0.000
are
Token are
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
go
Token go
Feature activation+0.000
her
Token her
Feature activation+0.000
way
Token way
Feature activation+1.800
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
she
Token she
Feature activation+0.000
's
Token's
Feature activation+0.000
had
Token had
Feature activation+0.000
pl
Tokenpl
Feature activation+0.000
a
Tokena
Feature activation+0.000
are
Token are
Feature activation+0.000
coming
Token coming
Feature activation+0.000
your
Token your
Feature activation+0.000
way
Token way
Feature activation+1.680
this
Token this
Feature activation+0.000
August
Token August
Feature activation+0.000
2016
Token 2016
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
as
Token as
Feature activation+0.000
more
Token more
Feature activation+0.000
news
Token news
Feature activation+0.000
comes
Token comes
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+1.527
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
we
Token we
Feature activation+0.000
expect
Token expect
Feature activation+0.000
much
Token much
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
go
Token go
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+1.519
,
Token,
Feature activation+0.000
blood
Token blood
Feature activation+0.000
lust
Tokenlust
Feature activation+0.000
often
Token often
Feature activation+0.000
ensued
Token ensued
Feature activation+0.000
like
Token like
Feature activation+0.000
everything
Token everything
Feature activation+0.000
else
Token else
Feature activation+0.000
thrown
Token thrown
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.495
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
puzzle
Token puzzle
Feature activation+0.000
to
Token to
Feature activation+0.000
kept
Token kept
Feature activation+0.000
taxpayer
Token taxpayer
Feature activation+0.000
dollars
Token dollars
Feature activation+0.000
flowing
Token flowing
Feature activation+0.000
its
Token its
Feature activation+0.000
way
Token way
Feature activation+1.444
despite
Token despite
Feature activation+0.000
its
Token its
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
child
Token child
Feature activation+0.000
good
Token good
Feature activation+0.000
fortune
Token fortune
Feature activation+0.000
that
Token that
Feature activation+0.000
falls
Token falls
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.294
,
Token,
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
I
TokenI
Feature activation+0.000
meant
Token meant
Feature activation+0.000
day
Token day
Feature activation+0.000
sensitive
Token sensitive
Feature activation+0.000
information
Token information
Feature activation+0.000
comes
Token comes
Feature activation+0.000
my
Token my
Feature activation+0.000
way
Token way
Feature activation+1.218
.
Token.
Feature activation+0.597
I
Token I
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.000
no
Token no
Feature activation+0.000
obstacles
Token obstacles
Feature activation+0.000
stand
Token stand
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.177
.
Token.
Feature activation+0.000
No
Token No
Feature activation+0.000
door
Token door
Feature activation+0.000
is
Token is
Feature activation+0.000
left
Token left
Feature activation+0.000
have
Token have
Feature activation+0.000
essentially
Token essentially
Feature activation+0.000
gone
Token gone
Feature activation+0.000
their
Token their
Feature activation+0.000
own
Token own
Feature activation+0.000
way
Token way
Feature activation+1.143
without
Token without
Feature activation+0.000
regard
Token regard
Feature activation+0.000
to
Token to
Feature activation+0.000
Moscow
Token Moscow
Feature activation+0.000
.
Token.
Feature activation+0.000

Top DFA by src position
MAX = 6.270

and
Token and
Feature activation-0.070
X
Token X
Feature activation-0.007
PS
TokenPS
Feature activation+0.004
stuff
Token stuff
Feature activation-0.059
coming
Token coming
Feature activation+1.559
their
Token their
Feature activation+4.897
way
Token way
Feature activation+2.509
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
other
Token other
Feature activation+0.036
opportunities
Token opportunities
Feature activation+0.052
that
Token that
Feature activation+0.444
might
Token might
Feature activation+0.078
come
Token come
Feature activation+1.262
our
Token our
Feature activation+3.584
way
Token way
Feature activation+2.194
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
it
Token it
Feature activation+0.000
gives
Token gives
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-3.876
those
Token those
Feature activation-0.727
opportunities
Token opportunities
Feature activation-0.046
that
Token that
Feature activation+0.042
came
Token came
Feature activation+1.463
our
Token our
Feature activation+4.303
way
Token way
Feature activation+2.196
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Richardson
Token Richardson
Feature activation+0.000
absorb
Token absorb
Feature activation+0.008
whatever
Token whatever
Feature activation-0.038
criticism
Token criticism
Feature activation+0.362
is
Token is
Feature activation+0.030
thrown
Token thrown
Feature activation+1.503
its
Token its
Feature activation+4.537
way
Token way
Feature activation+1.550
.
Token.
Feature activation+0.000
Atlanta
Token Atlanta
Feature activation+0.000
may
Token may
Feature activation+0.000
have
Token have
Feature activation+0.000
blocking
Token blocking
Feature activation-0.140
or
Token or
Feature activation-0.154
avoiding
Token avoiding
Feature activation+0.068
damage
Token damage
Feature activation+0.077
coming
Token coming
Feature activation+2.044
their
Token their
Feature activation+4.302
way
Token way
Feature activation+1.731
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
can
Token can
Feature activation+0.000
help
Token help
Feature activation+0.000
Chad
Token Chad
Feature activation-0.000
,
Token,
Feature activation-0.002
it
Token it
Feature activation+0.034
was
Token was
Feature activation-0.009
going
Token going
Feature activation+0.252
his
Token his
Feature activation+6.270
way
Token way
Feature activation+1.304
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
able
Token able
Feature activation+0.000
new
Token new
Feature activation+0.032
studio
Token studio
Feature activation+0.003
teaching
Token teaching
Feature activation-0.011
opportunities
Token opportunities
Feature activation+0.010
came
Token came
Feature activation+1.476
my
Token my
Feature activation+3.258
way
Token way
Feature activation+1.794
at
Token at
Feature activation+0.000
studios
Token studios
Feature activation+0.000
that
Token that
Feature activation+0.000
were
Token were
Feature activation+0.000
about
Token about
Feature activation+0.446
the
Token the
Feature activation+0.045
budget
Token budget
Feature activation+0.068
cuts
Token cuts
Feature activation-0.027
coming
Token coming
Feature activation+1.505
its
Token its
Feature activation+4.064
way
Token way
Feature activation+1.925
that
Token that
Feature activation+0.000
it
Token it
Feature activation+0.000
has
Token has
Feature activation+0.000
begun
Token begun
Feature activation+0.000
prospects
Token prospects
Feature activation+0.040
and
Token and
Feature activation-0.044
good
Token good
Feature activation+0.004
changes
Token changes
Feature activation+0.021
coming
Token coming
Feature activation+2.325
her
Token her
Feature activation+3.527
way
Token way
Feature activation+1.096
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
beaten
Token beaten
Feature activation-0.294
all
Token all
Feature activation-0.221
that
Token that
Feature activation-0.085
have
Token have
Feature activation+0.230
come
Token come
Feature activation+0.955
his
Token his
Feature activation+4.129
way
Token way
Feature activation+1.820
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Step
TokenStep
Feature activation+0.000
voters
Token voters
Feature activation+0.002
are
Token are
Feature activation+0.018
expected
Token expected
Feature activation-0.144
to
Token to
Feature activation-0.155
go
Token go
Feature activation+0.278
her
Token her
Feature activation+5.176
way
Token way
Feature activation+2.071
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
she
Token she
Feature activation+0.000
's
Token's
Feature activation+0.000
Gun
Token Gun
Feature activation-0.027
pl
Tokenpl
Feature activation-0.007
a
Tokena
Feature activation-0.004
are
Token are
Feature activation-0.129
coming
Token coming
Feature activation+1.114
your
Token your
Feature activation+3.600
way
Token way
Feature activation+2.434
this
Token this
Feature activation+0.000
August
Token August
Feature activation+0.000
2016
Token 2016
Feature activation+0.000
.
Token.
Feature activation+0.000
updated
Token updated
Feature activation+0.267
as
Token as
Feature activation+0.223
more
Token more
Feature activation-0.029
news
Token news
Feature activation+0.149
comes
Token comes
Feature activation+1.315
our
Token our
Feature activation+3.521
way
Token way
Feature activation+1.322
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
we
Token we
Feature activation+0.000
expect
Token expect
Feature activation+0.000
didn
Token didn
Feature activation+0.194
âĢ
TokenâĢ
Feature activation+0.119
Ļ
TokenĻ
Feature activation+0.168
t
Tokent
Feature activation+0.220
go
Token go
Feature activation+1.067
their
Token their
Feature activation+3.133
way
Token way
Feature activation+1.117
,
Token,
Feature activation+0.000
blood
Token blood
Feature activation+0.000
lust
Tokenlust
Feature activation+0.000
often
Token often
Feature activation+0.000
,
Token,
Feature activation+0.034
like
Token like
Feature activation+0.207
everything
Token everything
Feature activation-0.119
else
Token else
Feature activation-0.070
thrown
Token thrown
Feature activation+1.582
his
Token his
Feature activation+3.302
way
Token way
Feature activation+1.560
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
puzzle
Token puzzle
Feature activation+0.000
that
Token that
Feature activation+0.053
kept
Token kept
Feature activation+0.015
taxpayer
Token taxpayer
Feature activation-0.053
dollars
Token dollars
Feature activation-0.125
flowing
Token flowing
Feature activation+0.387
its
Token its
Feature activation+5.171
way
Token way
Feature activation+1.291
despite
Token despite
Feature activation+0.000
its
Token its
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
of
Token of
Feature activation+0.038
good
Token good
Feature activation+0.001
fortune
Token fortune
Feature activation+0.012
that
Token that
Feature activation+0.189
falls
Token falls
Feature activation+1.159
his
Token his
Feature activation+3.351
way
Token way
Feature activation+1.275
,
Token,
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
I
TokenI
Feature activation+0.000
given
Token given
Feature activation+0.020
day
Token day
Feature activation+0.085
sensitive
Token sensitive
Feature activation+0.115
information
Token information
Feature activation-0.012
comes
Token comes
Feature activation+0.939
my
Token my
Feature activation+3.222
way
Token way
Feature activation+1.973
.
Token.
Feature activation+0.000
I
Token I
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
no
Token no
Feature activation+0.742
obstacles
Token obstacles
Feature activation+0.307
stand
Token stand
Feature activation+0.762
in
Token in
Feature activation+0.292
his
Token his
Feature activation+1.909
way
Token way
Feature activation+2.154
.
Token.
Feature activation+0.000
No
Token No
Feature activation+0.000
door
Token door
Feature activation+0.000
is
Token is
Feature activation+0.000
left
Token left
Feature activation+0.000
chens
Tokenchens
Feature activation-0.012
have
Token have
Feature activation-0.087
essentially
Token essentially
Feature activation-0.094
gone
Token gone
Feature activation+0.322
their
Token their
Feature activation+1.433
own
Token own
Feature activation+2.695
way
Token way
Feature activation+1.835
without
Token without
Feature activation+0.000
regard
Token regard
Feature activation+0.000
to
Token to
Feature activation+0.000
Moscow
Token Moscow
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.02

Head 2: 0.16

Head 3: 0.06

Head 4: 0.17

Head 5: 0.04

Head 6: 0.17

Head 7: 0.14

Head 8: 0.05

Head 9: 0.05

Head 10: 0.07

Head 11: 0.05

Positive logits

cano1.48

geon1.30

soon1.30

Opinion1.29

��1.28

WAYS1.28

��1.27

pour1.25

ocrat1.22

kson1.22

oxide1.19

diam1.18

autopsy1.18

krit1.18

mortem1.18

Ambro1.17

estone1.17

Nay1.17

iren1.16

illation1.16

Negative logits

distinction-1.41

Japanese-1.39

innocence-1.29

atche-1.27

div-1.26

neutrality-1.26

participation-1.25

contacts-1.24

ailability-1.24

76561-1.21

DF-1.20

partnerships-1.20

friendships-1.20

ourcing-1.19

anonymity-1.18

difference-1.18

Adinida-1.17

disembark-1.17

Temp-1.16

potential-1.16

INTERVAL 3.117 - 3.464
CONTAINS 0.000%

X
Token X
Feature activation+0.000
PS
TokenPS
Feature activation+0.000
stuff
Token stuff
Feature activation+0.000
coming
Token coming
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+3.464
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
In
TokenIn
Feature activation+0.000
contrast
Token contrast
Feature activation+0.000

INTERVAL 2.771 - 3.117
CONTAINS 0.000%

opportunities
Token opportunities
Feature activation+0.000
that
Token that
Feature activation+0.000
might
Token might
Feature activation+0.000
come
Token come
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+2.802
.
Token.
Feature activation+0.000
And
Token And
Feature activation+0.000
it
Token it
Feature activation+0.000
gives
Token gives
Feature activation+0.000
us
Token us
Feature activation+0.000

INTERVAL 2.425 - 2.771
CONTAINS 0.000%

those
Token those
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
that
Token that
Feature activation+0.000
came
Token came
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+2.733
,
Token,
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
Richardson
Token Richardson
Feature activation+0.000
said
Token said
Feature activation+0.000
or
Token or
Feature activation+0.000
avoiding
Token avoiding
Feature activation+0.000
damage
Token damage
Feature activation+0.000
coming
Token coming
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+2.623
.
Token.
Feature activation+0.000
This
Token This
Feature activation+0.000
can
Token can
Feature activation+0.000
help
Token help
Feature activation+0.000
clinch
Token clinch
Feature activation+0.000
whatever
Token whatever
Feature activation+0.000
criticism
Token criticism
Feature activation+0.000
is
Token is
Feature activation+0.000
thrown
Token thrown
Feature activation+0.000
its
Token its
Feature activation+0.000
way
Token way
Feature activation+2.626
.
Token.
Feature activation+0.000
Atlanta
Token Atlanta
Feature activation+0.000
may
Token may
Feature activation+0.000
have
Token have
Feature activation+0.000
lost
Token lost
Feature activation+0.000

INTERVAL 2.078 - 2.425
CONTAINS 0.000%

,
Token,
Feature activation+0.000
it
Token it
Feature activation+0.000
was
Token was
Feature activation+0.000
going
Token going
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+2.395
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
was
Token was
Feature activation+0.000
able
Token able
Feature activation+0.000
to
Token to
Feature activation+0.000
studio
Token studio
Feature activation+0.000
teaching
Token teaching
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
came
Token came
Feature activation+0.000
my
Token my
Feature activation+0.000
way
Token way
Feature activation+2.287
at
Token at
Feature activation+0.844
studios
Token studios
Feature activation+0.000
that
Token that
Feature activation+0.000
were
Token were
Feature activation+0.000
a
Token a
Feature activation+0.000

INTERVAL 1.732 - 2.078
CONTAINS 0.000%

and
Token and
Feature activation+0.000
good
Token good
Feature activation+0.000
changes
Token changes
Feature activation+0.000
coming
Token coming
Feature activation+0.000
her
Token her
Feature activation+0.000
way
Token way
Feature activation+1.982
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
all
Token all
Feature activation+0.000
that
Token that
Feature activation+0.000
have
Token have
Feature activation+0.000
come
Token come
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.925
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Step
TokenStep
Feature activation+0.000
h
Tokenh
Feature activation+0.000
the
Token the
Feature activation+0.000
budget
Token budget
Feature activation+0.000
cuts
Token cuts
Feature activation+0.000
coming
Token coming
Feature activation+0.000
its
Token its
Feature activation+0.000
way
Token way
Feature activation+2.021
that
Token that
Feature activation+0.000
it
Token it
Feature activation+0.000
has
Token has
Feature activation+0.000
begun
Token begun
Feature activation+0.000
burning
Token burning
Feature activation+0.000
are
Token are
Feature activation+0.000
expected
Token expected
Feature activation+0.000
to
Token to
Feature activation+0.000
go
Token go
Feature activation+0.000
her
Token her
Feature activation+0.000
way
Token way
Feature activation+1.800
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
she
Token she
Feature activation+0.000
's
Token's
Feature activation+0.000
had
Token had
Feature activation+0.000

INTERVAL 1.386 - 1.732
CONTAINS 0.000%

pl
Tokenpl
Feature activation+0.000
a
Tokena
Feature activation+0.000
are
Token are
Feature activation+0.000
coming
Token coming
Feature activation+0.000
your
Token your
Feature activation+0.000
way
Token way
Feature activation+1.680
this
Token this
Feature activation+0.000
August
Token August
Feature activation+0.000
2016
Token 2016
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
kept
Token kept
Feature activation+0.000
taxpayer
Token taxpayer
Feature activation+0.000
dollars
Token dollars
Feature activation+0.000
flowing
Token flowing
Feature activation+0.000
its
Token its
Feature activation+0.000
way
Token way
Feature activation+1.444
despite
Token despite
Feature activation+0.000
its
Token its
Feature activation+0.000
use
Token use
Feature activation+0.000
of
Token of
Feature activation+0.000
child
Token child
Feature activation+0.000
as
Token as
Feature activation+0.000
more
Token more
Feature activation+0.000
news
Token news
Feature activation+0.000
comes
Token comes
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+1.527
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
we
Token we
Feature activation+0.000
expect
Token expect
Feature activation+0.000
much
Token much
Feature activation+0.000
like
Token like
Feature activation+0.000
everything
Token everything
Feature activation+0.000
else
Token else
Feature activation+0.000
thrown
Token thrown
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.495
,
Token,
Feature activation+0.000
as
Token as
Feature activation+0.000
a
Token a
Feature activation+0.000
puzzle
Token puzzle
Feature activation+0.000
to
Token to
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
go
Token go
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+1.519
,
Token,
Feature activation+0.000
blood
Token blood
Feature activation+0.000
lust
Tokenlust
Feature activation+0.000
often
Token often
Feature activation+0.000
ensued
Token ensued
Feature activation+0.000

INTERVAL 1.039 - 1.386
CONTAINS 0.000%

no
Token no
Feature activation+0.000
obstacles
Token obstacles
Feature activation+0.000
stand
Token stand
Feature activation+0.000
in
Token in
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.177
.
Token.
Feature activation+0.000
No
Token No
Feature activation+0.000
door
Token door
Feature activation+0.000
is
Token is
Feature activation+0.000
left
Token left
Feature activation+0.000
good
Token good
Feature activation+0.000
fortune
Token fortune
Feature activation+0.000
that
Token that
Feature activation+0.000
falls
Token falls
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+1.294
,
Token,
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
I
TokenI
Feature activation+0.000
meant
Token meant
Feature activation+0.000
day
Token day
Feature activation+0.000
sensitive
Token sensitive
Feature activation+0.000
information
Token information
Feature activation+0.000
comes
Token comes
Feature activation+0.000
my
Token my
Feature activation+0.000
way
Token way
Feature activation+1.218
.
Token.
Feature activation+0.597
I
Token I
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
ve
Tokenve
Feature activation+0.000
have
Token have
Feature activation+0.000
essentially
Token essentially
Feature activation+0.000
gone
Token gone
Feature activation+0.000
their
Token their
Feature activation+0.000
own
Token own
Feature activation+0.000
way
Token way
Feature activation+1.143
without
Token without
Feature activation+0.000
regard
Token regard
Feature activation+0.000
to
Token to
Feature activation+0.000
Moscow
Token Moscow
Feature activation+0.000
.
Token.
Feature activation+0.000
another
Token another
Feature activation+0.000
ship
Token ship
Feature activation+0.000
is
Token is
Feature activation+0.000
heading
Token heading
Feature activation+0.000
her
Token her
Feature activation+0.000
way
Token way
Feature activation+1.051
,
Token,
Feature activation+0.000
but
Token but
Feature activation+0.000
it
Token it
Feature activation+0.000
's
Token's
Feature activation+0.000
too
Token too
Feature activation+0.000

INTERVAL 0.693 - 1.039
CONTAINS 0.000%

âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
object
Token object
Feature activation+0.000
so
Token so
Feature activation+0.000
much
Token much
Feature activation+0.732
to
Token to
Feature activation+0.000
the
Token the
Feature activation+0.000
name
Token name
Feature activation+0.000
as
Token as
Feature activation+0.000
much
Token much
Feature activation+0.000
and
Token and
Feature activation+0.000
did
Token did
Feature activation+0.000
things
Token things
Feature activation+0.000
their
Token their
Feature activation+0.000
own
Token own
Feature activation+0.000
way
Token way
Feature activation+0.787
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
teaching
Token teaching
Feature activation+0.000
opportunities
Token opportunities
Feature activation+0.000
came
Token came
Feature activation+0.000
my
Token my
Feature activation+0.000
way
Token way
Feature activation+2.287
at
Token at
Feature activation+0.844
studios
Token studios
Feature activation+0.000
that
Token that
Feature activation+0.000
were
Token were
Feature activation+0.000
a
Token a
Feature activation+0.000
much
Token much
Feature activation+0.000
adversity
Token adversity
Feature activation+0.000
that
Token that
Feature activation+0.000
was
Token was
Feature activation+0.000
thrown
Token thrown
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+0.990
.
Token.
Feature activation+0.000
At
Token At
Feature activation+0.000
those
Token those
Feature activation+0.000
moments
Token moments
Feature activation+0.000
,
Token,
Feature activation+0.000
were
Token were
Feature activation+0.000
gonna
Token gonna
Feature activation+0.000
do
Token do
Feature activation+0.000
it
Token it
Feature activation+0.000
their
Token their
Feature activation+0.000
way
Token way
Feature activation+0.738
.
Token.
Feature activation+0.000
Their
Token Their
Feature activation+0.000
way
Token way
Feature activation+0.000
,
Token,
Feature activation+0.000
meaning
Token meaning
Feature activation+0.000

INTERVAL 0.346 - 0.693
CONTAINS 0.000%

individual
Token individual
Feature activation+0.000
awards
Token awards
Feature activation+0.000
to
Token to
Feature activation+0.000
come
Token come
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+0.385
.
Token.
Feature activation+0.000
The
Token The
Feature activation+0.000
search
Token search
Feature activation+0.000
matches
Token matches
Feature activation+0.000
for
Token for
Feature activation+0.000
to
Token to
Feature activation+0.000
do
Token do
Feature activation+0.000
things
Token things
Feature activation+0.000
my
Token my
Feature activation+0.000
own
Token own
Feature activation+0.000
way
Token way
Feature activation+0.398
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
eventually
Token eventually
Feature activation+0.000
I
Token I
Feature activation+0.000
learned
Token learned
Feature activation+0.000
all
Token all
Feature activation+0.000
the
Token the
Feature activation+0.000
hardships
Token hardships
Feature activation+0.000
thrown
Token thrown
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+0.377
.
Token.
Feature activation+0.000
Growing
Token Growing
Feature activation+0.000
up
Token up
Feature activation+0.000
in
Token in
Feature activation+0.000
such
Token such
Feature activation+0.000
provided
Token provided
Feature activation+0.000
enough
Token enough
Feature activation+0.000
contracts
Token contracts
Feature activation+0.000
come
Token come
Feature activation+0.000
his
Token his
Feature activation+0.000
way
Token way
Feature activation+0.485
to
Token to
Feature activation+0.000
support
Token support
Feature activation+0.000
the
Token the
Feature activation+0.000
business
Token business
Feature activation+0.000
.
Token.
Feature activation+0.000
going
Token going
Feature activation+0.000
to
Token to
Feature activation+0.000
start
Token start
Feature activation+0.000
going
Token going
Feature activation+0.000
our
Token our
Feature activation+0.000
way
Token way
Feature activation+0.489
and
Token and
Feature activation+0.000
we
Token we
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000

INTERVAL 0.000 - 0.346
CONTAINS 100.000%

battery
Token battery
Feature activation+0.000
chemistry
Token chemistry
Feature activation+0.000
.
Token.
Feature activation+0.000
As
Token As
Feature activation+0.000
such
Token such
Feature activation+0.000
,
Token,
Feature activation+0.000
while
Token while
Feature activation+0.000
"
Token "
Feature activation+0.000
20
Token20
Feature activation+0.000
%
Token%
Feature activation+0.000
faster
Token faster
Feature activation+0.000
gl
Token gl
Feature activation+0.000
oved
Tokenoved
Feature activation+0.000
hand
Token hand
Feature activation+0.000
that
Token that
Feature activation+0.000
is
Token is
Feature activation+0.000
determined
Token determined
Feature activation+0.000
to
Token to
Feature activation+0.000
perform
Token perform
Feature activation+0.000
a
Token a
Feature activation+0.000
Nazi
Token Nazi
Feature activation+0.000
salute
Token salute
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
We
TokenWe
Feature activation+0.000
do
Token do
Feature activation+0.000
have
Token have
Feature activation+0.000
the
Token the
Feature activation+0.000
capability
Token capability
Feature activation+0.000
.
Token.
Feature activation+0.000
We
Token We
Feature activation+0.000
are
Token are
Feature activation+0.000
prepared
Token prepared
Feature activation+0.000
obnoxious
Token obnoxious
Feature activation+0.000
act
Token act
Feature activation+0.000
is
Token is
Feature activation+0.000
a
Token a
Feature activation+0.000
cry
Token cry
Feature activation+0.000
for
Token for
Feature activation+0.000
help
Token help
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ex
TokenEx
Feature activation+0.000
remained
Token remained
Feature activation+0.000
uncertain
Token uncertain
Feature activation+0.000
.
Token.
Feature activation+0.000
It
Token It
Feature activation+0.000
is
Token is
Feature activation+0.000
now
Token now
Feature activation+0.000
up
Token up
Feature activation+0.000
to
Token to
Feature activation+0.000
American
Token American
Feature activation+0.000
conservatives
Token conservatives
Feature activation+0.000
to
Token to
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 24: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.991

lofty
Token lofty
Feature activation-0.007
goal
Token goal
Feature activation-0.008
:
Token:
Feature activation+0.039
Put
Token Put
Feature activation-0.017
a
Token a
Feature activation-0.170
Toronto
Token Toronto
Feature activation+0.053
Blue
Token Blue
Feature activation-0.008
Jays
Token Jays
Feature activation-0.001
âĢ
TokenâĢ
Feature activation+0.009
Ļ
TokenĻ
Feature activation+0.039
baseball
Token baseball
Feature activation-0.040
Jays
Token Jays
Feature activation-0.035
âĢ
TokenâĢ
Feature activation+0.029
Ļ
TokenĻ
Feature activation+0.041
baseball
Token baseball
Feature activation-0.153
cap
Token cap
Feature activation-0.420
and
Token and
Feature activation+0.384
a
Token a
Feature activation-0.098
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
a
Token a
Feature activation-0.049
Toronto
Token Toronto
Feature activation-0.138
Blue
Token Blue
Feature activation-0.028
Jays
Token Jays
Feature activation-0.026
âĢ
TokenâĢ
Feature activation+0.085
Ļ
TokenĻ
Feature activation+0.115
baseball
Token baseball
Feature activation-0.141
cap
Token cap
Feature activation-0.092
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Jays
Token Jays
Feature activation-0.047
âĢ
TokenâĢ
Feature activation+0.021
Ļ
TokenĻ
Feature activation+0.056
baseball
Token baseball
Feature activation-0.148
cap
Token cap
Feature activation-0.494
and
Token and
Feature activation+0.080
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation-0.327
the
Token the
Feature activation-0.184
NB
Token NB
Feature activation-0.054
Space
Token Space
Feature activation-0.016
Race
Token Race
Feature activation-0.541
had
Token had
Feature activation+0.147
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.074
,
Token,
Feature activation-0.069
the
Token the
Feature activation-0.080
NB
Token NB
Feature activation-0.020
Space
Token Space
Feature activation+0.022
Race
Token Race
Feature activation-0.261
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
,
Token,
Feature activation-0.126
the
Token the
Feature activation-0.115
NB
Token NB
Feature activation-0.024
Space
Token Space
Feature activation-0.013
Race
Token Race
Feature activation-0.270
had
Token had
Feature activation+0.123
a
Token a
Feature activation+0.052
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.223
,
Token,
Feature activation-0.084
the
Token the
Feature activation-0.090
NB
Token NB
Feature activation+0.002
Space
Token Space
Feature activation-0.000
Race
Token Race
Feature activation-0.026
lofty
Token lofty
Feature activation+0.006
goal
Token goal
Feature activation-0.006
:
Token:
Feature activation+0.003
Put
Token Put
Feature activation+0.004
a
Token a
Feature activation-0.031
Toronto
Token Toronto
Feature activation+0.343
Blue
Token Blue
Feature activation+0.001
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.108
,
Token,
Feature activation-0.044
the
Token the
Feature activation-0.009
NB
Token NB
Feature activation+0.003
Space
Token Space
Feature activation+0.008
Race
Token Race
Feature activation-0.076
had
Token had
Feature activation+0.023
a
Token a
Feature activation-0.054
somewhat
Token somewhat
Feature activation+0.010
less
Token less
Feature activation+0.014
lofty
Token lofty
Feature activation+0.080
goal
Token goal
Feature activation+0.205
:
Token:
Feature activation+0.006
Put
Token Put
Feature activation-0.003
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.007
less
Token less
Feature activation-0.018
lofty
Token lofty
Feature activation-0.011
goal
Token goal
Feature activation+0.020
:
Token:
Feature activation+0.044
Put
Token Put
Feature activation+0.051
a
Token a
Feature activation-0.099
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
:
Token:
Feature activation+0.006
Put
Token Put
Feature activation-0.022
a
Token a
Feature activation-0.040
Toronto
Token Toronto
Feature activation-0.120
Blue
Token Blue
Feature activation-0.062
Jays
Token Jays
Feature activation+0.038
âĢ
TokenâĢ
Feature activation+0.005
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.043
,
Token,
Feature activation-0.070
the
Token the
Feature activation-0.027
NB
Token NB
Feature activation+0.055
Space
Token Space
Feature activation-0.008
Race
Token Race
Feature activation-0.123
had
Token had
Feature activation+0.032
a
Token a
Feature activation+0.024
somewhat
Token somewhat
Feature activation+0.015
<|endoftext|>
Token<|endoftext|>
Feature activation+0.154
,
Token,
Feature activation-0.073
the
Token the
Feature activation-0.007
NB
Token NB
Feature activation+0.003
Space
Token Space
Feature activation+0.009
Race
Token Race
Feature activation-0.013
NB
Token NB
Feature activation+0.034
Space
Token Space
Feature activation+0.009
Race
Token Race
Feature activation-0.173
had
Token had
Feature activation+0.026
a
Token a
Feature activation+0.010
somewhat
Token somewhat
Feature activation+0.040
less
Token less
Feature activation+0.036
lofty
Token lofty
Feature activation+0.030
goal
Token goal
Feature activation-0.026
:
Token:
Feature activation+0.031
Put
Token Put
Feature activation+0.009
had
Token had
Feature activation+0.065
a
Token a
Feature activation-0.044
somewhat
Token somewhat
Feature activation-0.020
less
Token less
Feature activation-0.030
lofty
Token lofty
Feature activation+0.065
goal
Token goal
Feature activation+0.991
:
Token:
Feature activation-0.082
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
had
Token had
Feature activation+0.039
a
Token a
Feature activation-0.301
somewhat
Token somewhat
Feature activation-0.109
less
Token less
Feature activation-0.050
lofty
Token lofty
Feature activation+0.094
goal
Token goal
Feature activation+0.168
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.231
,
Token,
Feature activation-0.132
the
Token the
Feature activation-0.067
NB
Token NB
Feature activation+0.004
Space
Token Space
Feature activation+0.002
Race
Token Race
Feature activation-0.023
Space
Token Space
Feature activation-0.011
Race
Token Race
Feature activation-0.161
had
Token had
Feature activation+0.019
a
Token a
Feature activation-0.218
somewhat
Token somewhat
Feature activation-0.282
less
Token less
Feature activation+0.148
lofty
Token lofty
Feature activation+0.049
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.07

Head 1: 0.06

Head 2: 0.11

Head 3: 0.09

Head 4: 0.07

Head 5: 0.08

Head 6: 0.09

Head 7: 0.10

Head 8: 0.09

Head 9: 0.07

Head 10: 0.09

Head 11: 0.08

Positive logits

gow2.08

ween1.99

driving1.97

inventoryQuantity1.93

chwitz1.80

edin1.79

hift1.79

sidel1.77

rapes1.75

raping1.69

induced1.64

suppressing1.63

somew1.61

torn1.61

raped1.60

remorse1.59

spons1.56

advertising1.55

gart1.54

whore1.52

Negative logits

��-2.01

��極-1.97

Ambrose-1.65

Byrne-1.63

Generations-1.59

Davidson-1.58

Specifications-1.58

Chiefs-1.57

Benson-1.57

Benjamin-1.56

-1.56

Quint-1.55

Bret-1.55

Holder-1.54

©-1.54

Vari-1.52

Disclosure-1.51

abulary-1.50

Ninth-1.50

Phillips-1.49

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

upon
Token upon
Feature activation+0.000
."
Token."
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
project
Token project
Feature activation+0.000
,
Token,
Feature activation+0.000
titled
Token titled
Feature activation+0.000
"
Token "
Feature activation+0.000
Af
TokenAf
Feature activation+0.000
ro
Tokenro
Feature activation+0.000
diam
Token diam
Feature activation+0.000
eters
Tokeneters
Feature activation+0.000
of
Token of
Feature activation+0.000
different
Token different
Feature activation+0.000
sections
Token sections
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
large
Token large
Feature activation+0.000
intestine
Token intestine
Feature activation+0.000
,
Token,
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
cars
Token cars
Feature activation+0.000
and
Token and
Feature activation+0.000
vans
Token vans
Feature activation+0.000
CO
Token CO
Feature activation+0.000
2
Token2
Feature activation+0.000
regulation
Token regulation
Feature activation+0.000
which
Token which
Feature activation+0.000
is
Token is
Feature activation+0.000
expected
Token expected
Feature activation+0.000
in
Token in
Feature activation+0.000
They
TokenThey
Feature activation+0.000
need
Token need
Feature activation+0.000
us
Token us
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.000
we
Token we
Feature activation+0.000
[
Token [
Feature activation+0.000
them
Tokenthem
Feature activation+0.000
].
Token].
Feature activation+0.000
There
Token There
Feature activation+0.000
are
Token are
Feature activation+0.000
said
Token said
Feature activation+0.000
the
Token the
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
dr
Tokendr
Feature activation+0.000
acon
Tokenacon
Feature activation+0.000
ian
Tokenian
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ŀ
TokenĿ
Feature activation+0.000
charges
Token charges
Feature activation+0.000
showed
Token showed
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 25: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.568

Jays
Token Jays
Feature activation-0.018
âĢ
TokenâĢ
Feature activation-0.069
Ļ
TokenĻ
Feature activation+0.023
baseball
Token baseball
Feature activation-0.029
cap
Token cap
Feature activation-0.013
and
Token and
Feature activation+0.232
a
Token a
Feature activation-0.032
beer
Token beer
Feature activation+0.074
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.078
less
Token less
Feature activation-0.017
lofty
Token lofty
Feature activation+0.007
goal
Token goal
Feature activation+0.007
:
Token:
Feature activation+0.018
Put
Token Put
Feature activation+0.235
a
Token a
Feature activation-0.006
Toronto
Token Toronto
Feature activation+0.160
Blue
Token Blue
Feature activation-0.011
Jays
Token Jays
Feature activation-0.010
âĢ
TokenâĢ
Feature activation-0.038
lofty
Token lofty
Feature activation-0.000
goal
Token goal
Feature activation-0.006
:
Token:
Feature activation+0.031
Put
Token Put
Feature activation+0.064
a
Token a
Feature activation-0.012
Toronto
Token Toronto
Feature activation+0.124
Blue
Token Blue
Feature activation-0.013
Jays
Token Jays
Feature activation-0.034
âĢ
TokenâĢ
Feature activation-0.032
Ļ
TokenĻ
Feature activation+0.089
baseball
Token baseball
Feature activation-0.017
somewhat
Token somewhat
Feature activation+0.011
less
Token less
Feature activation-0.005
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.022
Put
Token Put
Feature activation+0.258
a
Token a
Feature activation-0.061
Toronto
Token Toronto
Feature activation+0.048
Blue
Token Blue
Feature activation-0.005
Jays
Token Jays
Feature activation-0.007
âĢ
TokenâĢ
Feature activation-0.015
,
Token,
Feature activation-0.102
the
Token the
Feature activation-0.131
NB
Token NB
Feature activation+0.022
Space
Token Space
Feature activation-0.065
Race
Token Race
Feature activation-0.090
had
Token had
Feature activation+0.227
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.940
,
Token,
Feature activation-0.086
the
Token the
Feature activation-0.177
NB
Token NB
Feature activation+0.051
Space
Token Space
Feature activation+0.059
Race
Token Race
Feature activation-0.113
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
the
Token the
Feature activation-0.076
NB
Token NB
Feature activation-0.014
Space
Token Space
Feature activation-0.052
Race
Token Race
Feature activation-0.066
had
Token had
Feature activation-0.040
a
Token a
Feature activation+0.372
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation-0.011
Space
Token Space
Feature activation-0.036
Race
Token Race
Feature activation+0.009
had
Token had
Feature activation+0.064
a
Token a
Feature activation+0.084
somewhat
Token somewhat
Feature activation+0.164
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
lofty
Token lofty
Feature activation+0.002
goal
Token goal
Feature activation-0.007
:
Token:
Feature activation+0.008
Put
Token Put
Feature activation+0.032
a
Token a
Feature activation-0.013
Toronto
Token Toronto
Feature activation+0.568
Blue
Token Blue
Feature activation-0.087
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.007
less
Token less
Feature activation-0.006
lofty
Token lofty
Feature activation+0.007
goal
Token goal
Feature activation-0.012
:
Token:
Feature activation+0.014
Put
Token Put
Feature activation+0.272
a
Token a
Feature activation-0.078
Toronto
Token Toronto
Feature activation+0.143
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
a
Token a
Feature activation+0.025
somewhat
Token somewhat
Feature activation+0.015
less
Token less
Feature activation-0.015
lofty
Token lofty
Feature activation+0.012
goal
Token goal
Feature activation-0.050
:
Token:
Feature activation+0.076
Put
Token Put
Feature activation+0.067
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.001
less
Token less
Feature activation-0.015
lofty
Token lofty
Feature activation+0.006
goal
Token goal
Feature activation+0.010
:
Token:
Feature activation+0.036
Put
Token Put
Feature activation+0.177
a
Token a
Feature activation-0.023
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.009
less
Token less
Feature activation-0.002
lofty
Token lofty
Feature activation+0.002
goal
Token goal
Feature activation-0.011
:
Token:
Feature activation+0.022
Put
Token Put
Feature activation+0.113
a
Token a
Feature activation-0.009
Toronto
Token Toronto
Feature activation+0.083
Blue
Token Blue
Feature activation-0.024
Jays
Token Jays
Feature activation+0.065
âĢ
TokenâĢ
Feature activation+0.003
lofty
Token lofty
Feature activation+0.008
goal
Token goal
Feature activation-0.027
:
Token:
Feature activation+0.025
Put
Token Put
Feature activation+0.170
a
Token a
Feature activation-0.023
Toronto
Token Toronto
Feature activation+0.364
Blue
Token Blue
Feature activation-0.145
Jays
Token Jays
Feature activation-0.153
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.007
less
Token less
Feature activation-0.002
lofty
Token lofty
Feature activation+0.005
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.011
Put
Token Put
Feature activation+0.160
a
Token a
Feature activation-0.007
Toronto
Token Toronto
Feature activation+0.047
Blue
Token Blue
Feature activation-0.006
Jays
Token Jays
Feature activation-0.002
âĢ
TokenâĢ
Feature activation+0.023
Put
Token Put
Feature activation+0.085
a
Token a
Feature activation+0.003
Toronto
Token Toronto
Feature activation+0.077
Blue
Token Blue
Feature activation-0.022
Jays
Token Jays
Feature activation-0.038
âĢ
TokenâĢ
Feature activation+0.168
Ļ
TokenĻ
Feature activation+0.012
baseball
Token baseball
Feature activation-0.038
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
,
Token,
Feature activation-0.123
the
Token the
Feature activation-0.026
NB
Token NB
Feature activation-0.009
Space
Token Space
Feature activation-0.033
Race
Token Race
Feature activation-0.021
had
Token had
Feature activation+0.103
a
Token a
Feature activation+0.048
somewhat
Token somewhat
Feature activation+0.037
less
Token less
Feature activation-0.039
lofty
Token lofty
Feature activation+0.020
goal
Token goal
Feature activation-0.244
,
Token,
Feature activation-0.052
the
Token the
Feature activation-0.044
NB
Token NB
Feature activation-0.017
Space
Token Space
Feature activation-0.024
Race
Token Race
Feature activation-0.094
had
Token had
Feature activation+0.174
a
Token a
Feature activation+0.102
somewhat
Token somewhat
Feature activation+0.044
less
Token less
Feature activation-0.013
lofty
Token lofty
Feature activation+0.095
goal
Token goal
Feature activation-0.202
,
Token,
Feature activation-0.049
the
Token the
Feature activation-0.057
NB
Token NB
Feature activation-0.014
Space
Token Space
Feature activation-0.025
Race
Token Race
Feature activation+0.004
had
Token had
Feature activation+0.102
a
Token a
Feature activation+0.066
somewhat
Token somewhat
Feature activation+0.072
less
Token less
Feature activation-0.035
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
,
Token,
Feature activation-0.063
the
Token the
Feature activation-0.055
NB
Token NB
Feature activation-0.055
Space
Token Space
Feature activation-0.028
Race
Token Race
Feature activation-0.050
had
Token had
Feature activation+0.125
a
Token a
Feature activation+0.081
somewhat
Token somewhat
Feature activation+0.068
less
Token less
Feature activation-0.138
lofty
Token lofty
Feature activation-0.119
goal
Token goal
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.06

Head 2: 0.10

Head 3: 0.09

Head 4: 0.08

Head 5: 0.08

Head 6: 0.09

Head 7: 0.09

Head 8: 0.09

Head 9: 0.07

Head 10: 0.09

Head 11: 0.09

Positive logits

ibling1.67

netflix1.60

pain1.55

monitor1.53

MRI1.52

gged1.51

xiety1.47

versive1.46

infeld1.45

scan1.45

issance1.45

ospace1.44

cible1.43

trash1.41

iazep1.41

tg1.41

ritch1.39

odcast1.39

ixtape1.39

psychiat1.38

Negative logits

aye-1.58

Og-1.56

Doctrine-1.54

OUR-1.53

}:-1.43

Dates-1.42

Regulations-1.42

Nay-1.41

の�-1.40

Oy-1.38

Est-1.37

Virgin-1.37

Thou-1.37

Hilton-1.36

-1.36

Hen-1.34

-1.34

Agu-1.31

Tata-1.30

omi-1.30

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

an
Token an
Feature activation+0.000
interview
Token interview
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Once
TokenOnce
Feature activation+0.000
your
Token your
Feature activation+0.000
application
Token application
Feature activation+0.000
has
Token has
Feature activation+0.000
been
Token been
Feature activation+0.000
received
Token received
Feature activation+0.000
,
Token,
Feature activation+0.000
expense
Token expense
Feature activation+0.000
and
Token and
Feature activation+0.000
for
Token for
Feature activation+0.000
poor
Token poor
Feature activation+0.000
reasons
Token reasons
Feature activation+0.000
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
now
Token now
Feature activation+0.000
sit
Token sit
Feature activation+0.000
half
Token half
Feature activation+0.000
-
Token-
Feature activation+0.000
to
Token to
Feature activation+0.000
use
Token use
Feature activation+0.000
the
Token the
Feature activation+0.000
Internet
Token Internet
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
wanted
Token wanted
Feature activation+0.000
to
Token to
Feature activation+0.000
get
Token get
Feature activation+0.000
on
Token on
Feature activation+0.000
this
Token this
Feature activation+0.000
be
Token be
Feature activation+0.000
optimal
Token optimal
Feature activation+0.000
for
Token for
Feature activation+0.000
creating
Token creating
Feature activation+0.000
ste
Token ste
Feature activation+0.000
aks
Tokenaks
Feature activation+0.000
in
Token in
Feature activation+0.000
the
Token the
Feature activation+0.000
lab
Token lab
Feature activation+0.000
.
Token.
Feature activation+0.000
That
Token That
Feature activation+0.000
a
Token a
Feature activation+0.000
televised
Token televised
Feature activation+0.000
address
Token address
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
suggested
Token suggested
Feature activation+0.000
that
Token that
Feature activation+0.000
the
Token the
Feature activation+0.000
motive
Token motive
Feature activation+0.000
was
Token was
Feature activation+0.000
to
Token to
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 26: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.387

Jays
Token Jays
Feature activation-0.007
âĢ
TokenâĢ
Feature activation+0.026
Ļ
TokenĻ
Feature activation+0.007
baseball
Token baseball
Feature activation+0.024
cap
Token cap
Feature activation-0.063
and
Token and
Feature activation+0.121
a
Token a
Feature activation-0.209
beer
Token beer
Feature activation+0.098
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
lofty
Token lofty
Feature activation-0.001
goal
Token goal
Feature activation-0.007
:
Token:
Feature activation+0.026
Put
Token Put
Feature activation+0.135
a
Token a
Feature activation+0.019
Toronto
Token Toronto
Feature activation+0.139
Blue
Token Blue
Feature activation+0.016
Jays
Token Jays
Feature activation+0.004
âĢ
TokenâĢ
Feature activation+0.036
Ļ
TokenĻ
Feature activation+0.009
baseball
Token baseball
Feature activation+0.130
Toronto
Token Toronto
Feature activation+0.039
Blue
Token Blue
Feature activation+0.008
Jays
Token Jays
Feature activation+0.018
âĢ
TokenâĢ
Feature activation+0.122
Ļ
TokenĻ
Feature activation+0.043
baseball
Token baseball
Feature activation+0.216
cap
Token cap
Feature activation+0.152
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.002
less
Token less
Feature activation-0.016
lofty
Token lofty
Feature activation-0.001
goal
Token goal
Feature activation-0.002
:
Token:
Feature activation+0.057
Put
Token Put
Feature activation+0.162
a
Token a
Feature activation-0.029
Toronto
Token Toronto
Feature activation+0.028
Blue
Token Blue
Feature activation+0.018
Jays
Token Jays
Feature activation+0.004
âĢ
TokenâĢ
Feature activation+0.036
<|endoftext|>
Token<|endoftext|>
Feature activation-0.899
,
Token,
Feature activation-0.282
the
Token the
Feature activation-0.131
NB
Token NB
Feature activation-0.048
Space
Token Space
Feature activation-0.029
Race
Token Race
Feature activation+0.387
had
Token had
Feature activation+0.158
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.777
,
Token,
Feature activation-0.169
the
Token the
Feature activation-0.110
NB
Token NB
Feature activation-0.046
Space
Token Space
Feature activation-0.035
Race
Token Race
Feature activation+0.089
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
,
Token,
Feature activation-0.078
the
Token the
Feature activation-0.089
NB
Token NB
Feature activation-0.007
Space
Token Space
Feature activation+0.015
Race
Token Race
Feature activation+0.155
had
Token had
Feature activation+0.230
a
Token a
Feature activation+0.054
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.001
Race
Token Race
Feature activation+0.009
had
Token had
Feature activation+0.019
a
Token a
Feature activation+0.026
somewhat
Token somewhat
Feature activation+0.027
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
lofty
Token lofty
Feature activation-0.004
goal
Token goal
Feature activation-0.008
:
Token:
Feature activation+0.013
Put
Token Put
Feature activation+0.022
a
Token a
Feature activation-0.067
Toronto
Token Toronto
Feature activation+0.270
Blue
Token Blue
Feature activation+0.002
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.000
less
Token less
Feature activation-0.015
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation-0.029
:
Token:
Feature activation+0.045
Put
Token Put
Feature activation+0.121
a
Token a
Feature activation-0.113
Toronto
Token Toronto
Feature activation+0.062
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
a
Token a
Feature activation-0.020
somewhat
Token somewhat
Feature activation+0.005
less
Token less
Feature activation-0.031
lofty
Token lofty
Feature activation+0.023
goal
Token goal
Feature activation-0.072
:
Token:
Feature activation+0.140
Put
Token Put
Feature activation+0.108
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.012
less
Token less
Feature activation-0.042
lofty
Token lofty
Feature activation-0.014
goal
Token goal
Feature activation-0.044
:
Token:
Feature activation+0.151
Put
Token Put
Feature activation+0.265
a
Token a
Feature activation+0.227
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.001
less
Token less
Feature activation-0.012
lofty
Token lofty
Feature activation+0.001
goal
Token goal
Feature activation-0.006
:
Token:
Feature activation+0.034
Put
Token Put
Feature activation+0.116
a
Token a
Feature activation-0.024
Toronto
Token Toronto
Feature activation+0.021
Blue
Token Blue
Feature activation+0.006
Jays
Token Jays
Feature activation+0.062
âĢ
TokenâĢ
Feature activation+0.024
lofty
Token lofty
Feature activation+0.002
goal
Token goal
Feature activation-0.021
:
Token:
Feature activation+0.062
Put
Token Put
Feature activation+0.100
a
Token a
Feature activation-0.102
Toronto
Token Toronto
Feature activation+0.117
Blue
Token Blue
Feature activation+0.103
Jays
Token Jays
Feature activation-0.121
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.003
less
Token less
Feature activation-0.013
lofty
Token lofty
Feature activation-0.004
goal
Token goal
Feature activation-0.009
:
Token:
Feature activation+0.012
Put
Token Put
Feature activation+0.144
a
Token a
Feature activation-0.039
Toronto
Token Toronto
Feature activation+0.025
Blue
Token Blue
Feature activation+0.018
Jays
Token Jays
Feature activation+0.018
âĢ
TokenâĢ
Feature activation+0.047
Put
Token Put
Feature activation+0.057
a
Token a
Feature activation-0.038
Toronto
Token Toronto
Feature activation+0.042
Blue
Token Blue
Feature activation+0.020
Jays
Token Jays
Feature activation-0.034
âĢ
TokenâĢ
Feature activation+0.122
Ļ
TokenĻ
Feature activation-0.024
baseball
Token baseball
Feature activation+0.026
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.833
,
Token,
Feature activation-0.209
the
Token the
Feature activation-0.026
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.016
Race
Token Race
Feature activation+0.151
had
Token had
Feature activation+0.092
a
Token a
Feature activation+0.020
somewhat
Token somewhat
Feature activation+0.010
less
Token less
Feature activation-0.012
lofty
Token lofty
Feature activation+0.054
<|endoftext|>
Token<|endoftext|>
Feature activation-0.903
,
Token,
Feature activation-0.118
the
Token the
Feature activation-0.054
NB
Token NB
Feature activation-0.010
Space
Token Space
Feature activation-0.013
Race
Token Race
Feature activation+0.262
had
Token had
Feature activation+0.014
a
Token a
Feature activation+0.017
somewhat
Token somewhat
Feature activation-0.005
less
Token less
Feature activation-0.052
lofty
Token lofty
Feature activation+0.104
the
Token the
Feature activation-0.063
NB
Token NB
Feature activation-0.003
Space
Token Space
Feature activation-0.002
Race
Token Race
Feature activation+0.004
had
Token had
Feature activation+0.012
a
Token a
Feature activation+0.044
somewhat
Token somewhat
Feature activation+0.020
less
Token less
Feature activation-0.107
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
the
Token the
Feature activation-0.062
NB
Token NB
Feature activation-0.016
Space
Token Space
Feature activation-0.004
Race
Token Race
Feature activation+0.031
had
Token had
Feature activation-0.004
a
Token a
Feature activation+0.050
somewhat
Token somewhat
Feature activation-0.014
less
Token less
Feature activation-0.070
lofty
Token lofty
Feature activation-0.109
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.09

Head 1: 0.08

Head 2: 0.08

Head 3: 0.08

Head 4: 0.08

Head 5: 0.10

Head 6: 0.08

Head 7: 0.08

Head 8: 0.10

Head 9: 0.08

Head 10: 0.08

Head 11: 0.08

Positive logits

trak1.82

ertodd1.80

edes1.72

ILCS1.69

enfranch1.67

beer1.58

eka1.51

eer1.50

jriwal1.49

hypoc1.48

Oy1.44

ijuana1.41

Tile1.40

||||1.40

owntown1.40

wholesale1.38

pai1.38

ulent1.37

ABV1.37

module1.35

Negative logits

hers-1.59

etime-1.54

pressed-1.53

ties-1.43

sciences-1.42

uggest-1.37

grasped-1.36

.–-1.36

Bone-1.34

remem-1.34

raft-1.33

.",-1.33

Bret-1.33

whispered-1.32

vain-1.31

lore-1.29

persuasion-1.29

wen-1.29

ogue-1.29

instinctively-1.29

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

Ċ
TokenĊ
Feature activation+0.000
Thursday
TokenThursday
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
s
Tokens
Feature activation+0.000
action
Token action
Feature activation+0.000
was
Token was
Feature activation+0.000
the
Token the
Feature activation+0.000
15
Token 15
Feature activation+0.000
th
Tokenth
Feature activation+0.000
trade
Token trade
Feature activation+0.000
won
Token won
Feature activation+0.000
far
Token far
Feature activation+0.000
more
Token more
Feature activation+0.000
than
Token than
Feature activation+0.000
anyone
Token anyone
Feature activation+0.000
ever
Token ever
Feature activation+0.000
imagined
Token imagined
Feature activation+0.000
,
Token,
Feature activation+0.000
charging
Token charging
Feature activation+0.000
through
Token through
Feature activation+0.000
its
Token its
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
felt
Token felt
Feature activation+0.000
.
Token.
Feature activation+0.000
He
Token He
Feature activation+0.000
told
Token told
Feature activation+0.000
the
Token the
Feature activation+0.000
best
Token best
Feature activation+0.000
in
Token in
Feature activation+0.000
-
Token-
Feature activation+0.000
ring
Tokenring
Feature activation+0.000
person
Tokenperson
Feature activation+0.000
of
Token of
Feature activation+0.000
interest
Token interest
Feature activation+0.000
"
Token"
Feature activation+0.000
in
Token in
Feature activation+0.000
an
Token an
Feature activation+0.000
ongoing
Token ongoing
Feature activation+0.000
California
Token California
Feature activation+0.000
homicide
Token homicide
Feature activation+0.000
investigation
Token investigation
Feature activation+0.000
.
Token.
Feature activation+0.000
VR
Token VR
Feature activation+0.000
Chat
TokenChat
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Not
TokenNot
Feature activation+0.000
since
Token since
Feature activation+0.000
the
Token the
Feature activation+0.000
early
Token early
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 27: In local text involving redirects

TOP ACTIVATIONS
MAX = 5.571

nik
Tokennik
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.242
s
Tokens
Feature activation+4.676
here
Token here
Feature activation+3.146
.
Token.
Feature activation+5.571
For
Token For
Feature activation+3.897
the
Token the
Feature activation+2.064
slang
Token slang
Feature activation+1.092
and
Token and
Feature activation+0.000
internet
Token internet
Feature activation+0.566
ak
Tokenak
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.468
s
Tokens
Feature activation+4.646
here
Token here
Feature activation+2.883
.
Token.
Feature activation+5.476
For
Token For
Feature activation+3.657
other
Token other
Feature activation+1.579
uses
Token uses
Feature activation+2.105
,
Token,
Feature activation+1.543
see
Token see
Feature activation+0.519
ai
Tokenai
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.818
s
Tokens
Feature activation+4.150
here
Token here
Feature activation+3.006
.
Token.
Feature activation+5.405
For
Token For
Feature activation+3.765
the
Token the
Feature activation+1.489
restaurant
Token restaurant
Feature activation+0.000
chain
Token chain
Feature activation+0.000
,
Token,
Feature activation+0.644
a
Tokena
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.478
s
Tokens
Feature activation+4.526
here
Token here
Feature activation+2.410
.
Token.
Feature activation+5.309
For
Token For
Feature activation+3.727
other
Token other
Feature activation+1.118
uses
Token uses
Feature activation+1.632
,
Token,
Feature activation+1.281
see
Token see
Feature activation+0.732
Dale
Token Dale
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.249
s
Tokens
Feature activation+4.876
here
Token here
Feature activation+2.632
.
Token.
Feature activation+5.292
For
Token For
Feature activation+3.383
other
Token other
Feature activation+1.571
uses
Token uses
Feature activation+2.107
,
Token,
Feature activation+1.363
see
Token see
Feature activation+0.269
Cecil
Token Cecil
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.960
s
Tokens
Feature activation+4.089
here
Token here
Feature activation+3.116
.
Token.
Feature activation+5.266
For
Token For
Feature activation+3.417
his
Token his
Feature activation+0.000
father
Token father
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
bomb
Tokenbomb
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.250
s
Tokens
Feature activation+4.632
here
Token here
Feature activation+3.188
.
Token.
Feature activation+5.225
For
Token For
Feature activation+3.628
the
Token the
Feature activation+1.602
song
Token song
Feature activation+0.121
by
Token by
Feature activation+0.000
the
Token the
Feature activation+0.000
ani
Tokenani
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.882
s
Tokens
Feature activation+4.138
here
Token here
Feature activation+2.587
.
Token.
Feature activation+5.129
For
Token For
Feature activation+3.534
other
Token other
Feature activation+1.373
uses
Token uses
Feature activation+2.069
,
Token,
Feature activation+1.574
see
Token see
Feature activation+0.593
emon
Tokenemon
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.837
s
Tokens
Feature activation+4.065
here
Token here
Feature activation+2.754
.
Token.
Feature activation+5.114
For
Token For
Feature activation+3.590
other
Token other
Feature activation+1.093
uses
Token uses
Feature activation+1.634
,
Token,
Feature activation+1.241
see
Token see
Feature activation+0.412
let
Tokenlet
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.377
s
Tokens
Feature activation+4.567
here
Token here
Feature activation+2.376
.
Token.
Feature activation+5.106
For
Token For
Feature activation+3.285
the
Token the
Feature activation+0.000
fisher
Token fisher
Feature activation+0.000
y
Tokeny
Feature activation+0.000
patrol
Token patrol
Feature activation+0.000
"
Token"
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Dale
Token Dale
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.249
s
Tokens
Feature activation+4.876
here
Token here
Feature activation+2.632
.
Token.
Feature activation+5.292
For
Token For
Feature activation+3.383
other
Token other
Feature activation+1.571
uses
Token uses
Feature activation+2.107
uku
Tokenuku
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.085
s
Tokens
Feature activation+4.634
here
Token here
Feature activation+2.883
.
Token.
Feature activation+4.873
For
Token For
Feature activation+3.514
the
Token the
Feature activation+1.700
malware
Token malware
Feature activation+0.001
,
Token,
Feature activation+1.072
see
Token see
Feature activation+0.939
PC
Token PC
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.478
s
Tokens
Feature activation+3.743
here
Token here
Feature activation+2.547
.
Token.
Feature activation+4.866
For
Token For
Feature activation+3.398
general
Token general
Feature activation+0.308
IBM
Token IBM
Feature activation+0.000
-
Token-
Feature activation+0.000
like
Tokenlike
Feature activation+0.000
ran
Tokenran
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.047
s
Tokens
Feature activation+4.462
here
Token here
Feature activation+2.424
.
Token.
Feature activation+4.739
For
Token For
Feature activation+3.210
other
Token other
Feature activation+1.064
uses
Token uses
Feature activation+1.442
,
Token,
Feature activation+0.999
see
Token see
Feature activation+0.141
V
TokenV
Feature activation+0.000
at
Tokenat
Feature activation+0.000
nik
Tokennik
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.242
s
Tokens
Feature activation+4.676
here
Token here
Feature activation+3.146
.
Token.
Feature activation+5.571
For
Token For
Feature activation+3.897
the
Token the
Feature activation+2.064
slang
Token slang
Feature activation+1.092
An
TokenAn
Feature activation+0.000
or
Tokenor
Feature activation+0.000
ak
Tokenak
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.468
s
Tokens
Feature activation+4.646
here
Token here
Feature activation+2.883
.
Token.
Feature activation+5.476
For
Token For
Feature activation+3.657
other
Token other
Feature activation+1.579
uses
Token uses
Feature activation+2.105
"
Token"
Feature activation+0.000
D
TokenD
Feature activation+0.000
uku
Tokenuku
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.085
s
Tokens
Feature activation+4.634
here
Token here
Feature activation+2.883
.
Token.
Feature activation+4.873
For
Token For
Feature activation+3.514
the
Token the
Feature activation+1.700
malware
Token malware
Feature activation+0.001
"
Token"
Feature activation+0.000
Buzz
TokenBuzz
Feature activation+0.000
bomb
Tokenbomb
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.250
s
Tokens
Feature activation+4.632
here
Token here
Feature activation+3.188
.
Token.
Feature activation+5.225
For
Token For
Feature activation+3.628
the
Token the
Feature activation+1.602
song
Token song
Feature activation+0.121
ane
Tokenane
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.885
s
Tokens
Feature activation+3.634
here
Token here
Feature activation+2.216
.
Token.
Feature activation+4.617
For
Token For
Feature activation+2.646
the
Token the
Feature activation+1.004
rapper
Token rapper
Feature activation+0.000
,
Token,
Feature activation+0.000
see
Token see
Feature activation+0.000
A
TokenA
Feature activation+0.000
uk
Tokenuk
Feature activation+0.000
let
Tokenlet
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.377
s
Tokens
Feature activation+4.567
here
Token here
Feature activation+2.376
.
Token.
Feature activation+5.106
For
Token For
Feature activation+3.285
the
Token the
Feature activation+0.000
fisher
Token fisher
Feature activation+0.000

Top DFA by src position
MAX = 6.710

"
Token"
Feature activation+0.152
V
TokenV
Feature activation+0.141
at
Tokenat
Feature activation-0.012
nik
Tokennik
Feature activation-0.020
"
Token"
Feature activation+0.271
redirect
Token redirect
Feature activation+5.468
s
Tokens
Feature activation+0.393
here
Token here
Feature activation+1.490
.
Token.
Feature activation+0.210
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token"
Feature activation+0.130
An
TokenAn
Feature activation+0.109
or
Tokenor
Feature activation-0.010
ak
Tokenak
Feature activation-0.013
"
Token"
Feature activation+0.310
redirect
Token redirect
Feature activation+5.400
s
Tokens
Feature activation+0.350
here
Token here
Feature activation+1.720
.
Token.
Feature activation+0.184
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
"
Token"
Feature activation+0.140
T
TokenT
Feature activation+0.135
od
Tokenod
Feature activation+0.003
ai
Tokenai
Feature activation-0.004
"
Token"
Feature activation+0.200
redirect
Token redirect
Feature activation+5.591
s
Tokens
Feature activation+0.385
here
Token here
Feature activation+1.721
.
Token.
Feature activation+0.175
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
Central
TokenCentral
Feature activation-0.027
f
Token f
Feature activation-0.025
ove
Tokenove
Feature activation+0.035
a
Tokena
Feature activation+0.005
"
Token"
Feature activation+0.255
redirect
Token redirect
Feature activation+5.645
s
Tokens
Feature activation+0.336
here
Token here
Feature activation+1.432
.
Token.
Feature activation+0.170
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.537
"
Token"
Feature activation+0.131
The
TokenThe
Feature activation+0.113
Dale
Token Dale
Feature activation+0.090
"
Token"
Feature activation+0.171
redirect
Token redirect
Feature activation+5.222
s
Tokens
Feature activation+0.332
here
Token here
Feature activation+1.642
.
Token.
Feature activation+0.243
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
"
Token"
Feature activation+0.141
Lord
TokenLord
Feature activation+0.019
Robert
Token Robert
Feature activation-0.032
Cecil
Token Cecil
Feature activation-0.034
"
Token"
Feature activation+0.130
redirect
Token redirect
Feature activation+5.716
s
Tokens
Feature activation+0.366
here
Token here
Feature activation+1.472
.
Token.
Feature activation+0.232
For
Token For
Feature activation+0.000
his
Token his
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.422
"
Token"
Feature activation+0.140
Buzz
TokenBuzz
Feature activation+0.061
bomb
Tokenbomb
Feature activation-0.004
"
Token"
Feature activation+0.279
redirect
Token redirect
Feature activation+5.452
s
Tokens
Feature activation+0.348
here
Token here
Feature activation+1.400
.
Token.
Feature activation+0.210
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token"
Feature activation+0.133
Gal
TokenGal
Feature activation+0.074
v
Tokenv
Feature activation-0.003
ani
Tokenani
Feature activation+0.019
"
Token"
Feature activation+0.271
redirect
Token redirect
Feature activation+5.234
s
Tokens
Feature activation+0.319
here
Token here
Feature activation+1.628
.
Token.
Feature activation+0.205
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
"
Token"
Feature activation+0.147
H
TokenH
Feature activation+0.048
eg
Tokeneg
Feature activation-0.178
emon
Tokenemon
Feature activation-0.042
"
Token"
Feature activation+0.139
redirect
Token redirect
Feature activation+5.293
s
Tokens
Feature activation+0.331
here
Token here
Feature activation+1.552
.
Token.
Feature activation+0.166
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
"
Token"
Feature activation+0.168
A
TokenA
Feature activation-0.006
uk
Tokenuk
Feature activation-0.004
let
Tokenlet
Feature activation+0.000
"
Token"
Feature activation+0.278
redirect
Token redirect
Feature activation+5.216
s
Tokens
Feature activation+0.434
here
Token here
Feature activation+1.481
.
Token.
Feature activation+0.258
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.256
"
Token"
Feature activation+0.103
The
TokenThe
Feature activation+0.067
Dale
Token Dale
Feature activation+0.031
"
Token"
Feature activation+0.126
redirect
Token redirect
Feature activation+6.710
s
Tokens
Feature activation+0.358
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.264
"
Token"
Feature activation+0.196
D
TokenD
Feature activation+0.086
uku
Tokenuku
Feature activation-0.002
"
Token"
Feature activation+0.353
redirect
Token redirect
Feature activation+5.319
s
Tokens
Feature activation+0.323
here
Token here
Feature activation+1.617
.
Token.
Feature activation+0.184
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token"
Feature activation+0.150
IB
TokenIB
Feature activation+0.040
M
TokenM
Feature activation-0.022
PC
Token PC
Feature activation-0.092
"
Token"
Feature activation+0.351
redirect
Token redirect
Feature activation+5.228
s
Tokens
Feature activation+0.288
here
Token here
Feature activation+1.301
.
Token.
Feature activation+0.278
For
Token For
Feature activation+0.000
general
Token general
Feature activation+0.000
"
Token"
Feature activation+0.131
Air
TokenAir
Feature activation+0.026
T
TokenT
Feature activation+0.053
ran
Tokenran
Feature activation+0.001
"
Token"
Feature activation+0.152
redirect
Token redirect
Feature activation+4.605
s
Tokens
Feature activation+0.271
here
Token here
Feature activation+1.872
.
Token.
Feature activation+0.161
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
"
Token"
Feature activation+0.213
V
TokenV
Feature activation+0.088
at
Tokenat
Feature activation-0.040
nik
Tokennik
Feature activation+0.023
"
Token"
Feature activation+0.336
redirect
Token redirect
Feature activation+6.522
s
Tokens
Feature activation+0.468
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token"
Feature activation+0.173
An
TokenAn
Feature activation+0.074
or
Tokenor
Feature activation-0.091
ak
Tokenak
Feature activation-0.008
"
Token"
Feature activation+0.321
redirect
Token redirect
Feature activation+6.533
s
Tokens
Feature activation+0.446
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
other
Token other
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.066
"
Token"
Feature activation+0.278
D
TokenD
Feature activation+0.052
uku
Tokenuku
Feature activation-0.012
"
Token"
Feature activation+0.368
redirect
Token redirect
Feature activation+6.582
s
Tokens
Feature activation+0.269
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.181
"
Token"
Feature activation+0.309
Buzz
TokenBuzz
Feature activation+0.058
bomb
Tokenbomb
Feature activation-0.020
"
Token"
Feature activation+0.271
redirect
Token redirect
Feature activation+6.691
s
Tokens
Feature activation+0.337
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
S
TokenS
Feature activation+0.064
utter
Tokenutter
Feature activation-0.029
C
Token C
Feature activation+0.007
ane
Tokenane
Feature activation-0.016
"
Token"
Feature activation+0.160
redirect
Token redirect
Feature activation+5.003
s
Tokens
Feature activation+0.293
here
Token here
Feature activation+1.436
.
Token.
Feature activation+0.178
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000
"
Token"
Feature activation+0.268
A
TokenA
Feature activation-0.001
uk
Tokenuk
Feature activation-0.007
let
Tokenlet
Feature activation-0.003
"
Token"
Feature activation+0.430
redirect
Token redirect
Feature activation+5.871
s
Tokens
Feature activation+0.343
here
Token here
Feature activation+0.000
.
Token.
Feature activation+0.000
For
Token For
Feature activation+0.000
the
Token the
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.02

Head 1: 0.02

Head 2: 0.09

Head 3: 0.09

Head 4: 0.09

Head 5: 0.04

Head 6: 0.06

Head 7: 0.14

Head 8: 0.04

Head 9: 0.06

Head 10: 0.24

Head 11: 0.12

Positive logits

lisher1.54

redirect1.49

reversed1.37

deleted1.36

redistributed1.35

deletion1.33

collapsing1.30

entity1.29

edit1.27

collapsed1.26

toggle1.25

enei1.25

admin1.24

channelAvailability1.24

retrieving1.23

overd1.23

namespace1.22

control1.21

delete1.20

delete1.20

Negative logits

Pros-1.46

abal-1.45

SAM-1.43

Aid-1.42

milo-1.41

Berry-1.37

Magikarp-1.36

salsa-1.35

OPS-1.35

LIB-1.34

LAB-1.33

fruit-1.31

uds-1.28

sson-1.28

onite-1.27

confessions-1.27

sam-1.27

ollah-1.26

Chip-1.25

Sund-1.25

INTERVAL 5.014 - 5.571
CONTAINS 0.000%

Dale
Token Dale
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.249
s
Tokens
Feature activation+4.876
here
Token here
Feature activation+2.632
.
Token.
Feature activation+5.292
For
Token For
Feature activation+3.383
other
Token other
Feature activation+1.571
uses
Token uses
Feature activation+2.107
,
Token,
Feature activation+1.363
see
Token see
Feature activation+0.269
ai
Tokenai
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.818
s
Tokens
Feature activation+4.150
here
Token here
Feature activation+3.006
.
Token.
Feature activation+5.405
For
Token For
Feature activation+3.765
the
Token the
Feature activation+1.489
restaurant
Token restaurant
Feature activation+0.000
chain
Token chain
Feature activation+0.000
,
Token,
Feature activation+0.644
emon
Tokenemon
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.837
s
Tokens
Feature activation+4.065
here
Token here
Feature activation+2.754
.
Token.
Feature activation+5.114
For
Token For
Feature activation+3.590
other
Token other
Feature activation+1.093
uses
Token uses
Feature activation+1.634
,
Token,
Feature activation+1.241
see
Token see
Feature activation+0.412
Cecil
Token Cecil
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.960
s
Tokens
Feature activation+4.089
here
Token here
Feature activation+3.116
.
Token.
Feature activation+5.266
For
Token For
Feature activation+3.417
his
Token his
Feature activation+0.000
father
Token father
Feature activation+0.000
and
Token and
Feature activation+0.000
the
Token the
Feature activation+0.000
ani
Tokenani
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.882
s
Tokens
Feature activation+4.138
here
Token here
Feature activation+2.587
.
Token.
Feature activation+5.129
For
Token For
Feature activation+3.534
other
Token other
Feature activation+1.373
uses
Token uses
Feature activation+2.069
,
Token,
Feature activation+1.574
see
Token see
Feature activation+0.593

INTERVAL 4.457 - 5.014
CONTAINS 0.000%

"
Token"
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Dale
Token Dale
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.249
s
Tokens
Feature activation+4.876
here
Token here
Feature activation+2.632
.
Token.
Feature activation+5.292
For
Token For
Feature activation+3.383
other
Token other
Feature activation+1.571
uses
Token uses
Feature activation+2.107
ane
Tokenane
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.885
s
Tokens
Feature activation+3.634
here
Token here
Feature activation+2.216
.
Token.
Feature activation+4.617
For
Token For
Feature activation+2.646
the
Token the
Feature activation+1.004
rapper
Token rapper
Feature activation+0.000
,
Token,
Feature activation+0.000
see
Token see
Feature activation+0.000
f
Token f
Feature activation+0.000
ove
Tokenove
Feature activation+0.000
a
Tokena
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.478
s
Tokens
Feature activation+4.526
here
Token here
Feature activation+2.410
.
Token.
Feature activation+5.309
For
Token For
Feature activation+3.727
other
Token other
Feature activation+1.118
uses
Token uses
Feature activation+1.632
"
Token"
Feature activation+0.000
D
TokenD
Feature activation+0.000
uku
Tokenuku
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.085
s
Tokens
Feature activation+4.634
here
Token here
Feature activation+2.883
.
Token.
Feature activation+4.873
For
Token For
Feature activation+3.514
the
Token the
Feature activation+1.700
malware
Token malware
Feature activation+0.001
Air
TokenAir
Feature activation+0.000
T
TokenT
Feature activation+0.000
ran
Tokenran
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.047
s
Tokens
Feature activation+4.462
here
Token here
Feature activation+2.424
.
Token.
Feature activation+4.739
For
Token For
Feature activation+3.210
other
Token other
Feature activation+1.064
uses
Token uses
Feature activation+1.442

INTERVAL 3.900 - 4.457
CONTAINS 0.000%

elephant
Token elephant
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.318
s
Tokens
Feature activation+4.111
here
Token here
Feature activation+2.850
.
Token.
Feature activation+3.997
For
Token For
Feature activation+3.149
the
Token the
Feature activation+1.585
super
Token super
Feature activation+0.000
family
Tokenfamily
Feature activation+0.000
of
Token of
Feature activation+0.000
locks
Tokenlocks
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.589
s
Tokens
Feature activation+3.586
here
Token here
Feature activation+1.507
.
Token.
Feature activation+4.377
For
Token For
Feature activation+2.275
the
Token the
Feature activation+0.000
song
Token song
Feature activation+0.000
by
Token by
Feature activation+0.000
Alt
Token Alt
Feature activation+0.000
Gal
TokenGal
Feature activation+0.000
v
Tokenv
Feature activation+0.000
ani
Tokenani
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.882
s
Tokens
Feature activation+4.138
here
Token here
Feature activation+2.587
.
Token.
Feature activation+5.129
For
Token For
Feature activation+3.534
other
Token other
Feature activation+1.373
uses
Token uses
Feature activation+2.069
ista
Tokenista
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.366
s
Tokens
Feature activation+1.999
here
Token here
Feature activation+1.983
.
Token.
Feature activation+4.321
For
Token For
Feature activation+2.738
the
Token the
Feature activation+0.854
Brazilian
Token Brazilian
Feature activation+0.000
footballer
Token footballer
Feature activation+0.000
,
Token,
Feature activation+0.000
Lord
TokenLord
Feature activation+0.000
Robert
Token Robert
Feature activation+0.000
Cecil
Token Cecil
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.960
s
Tokens
Feature activation+4.089
here
Token here
Feature activation+3.116
.
Token.
Feature activation+5.266
For
Token For
Feature activation+3.417
his
Token his
Feature activation+0.000
father
Token father
Feature activation+0.000

INTERVAL 3.343 - 3.900
CONTAINS 0.000%

"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.242
s
Tokens
Feature activation+4.676
here
Token here
Feature activation+3.146
.
Token.
Feature activation+5.571
For
Token For
Feature activation+3.897
the
Token the
Feature activation+2.064
slang
Token slang
Feature activation+1.092
and
Token and
Feature activation+0.000
internet
Token internet
Feature activation+0.566
meme
Token meme
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.249
s
Tokens
Feature activation+4.876
here
Token here
Feature activation+2.632
.
Token.
Feature activation+5.292
For
Token For
Feature activation+3.383
other
Token other
Feature activation+1.571
uses
Token uses
Feature activation+2.107
,
Token,
Feature activation+1.363
see
Token see
Feature activation+0.269
Dale
Token Dale
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.818
s
Tokens
Feature activation+4.150
here
Token here
Feature activation+3.006
.
Token.
Feature activation+5.405
For
Token For
Feature activation+3.765
the
Token the
Feature activation+1.489
restaurant
Token restaurant
Feature activation+0.000
chain
Token chain
Feature activation+0.000
,
Token,
Feature activation+0.644
see
Token see
Feature activation+1.704
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.478
s
Tokens
Feature activation+4.526
here
Token here
Feature activation+2.410
.
Token.
Feature activation+5.309
For
Token For
Feature activation+3.727
other
Token other
Feature activation+1.118
uses
Token uses
Feature activation+1.632
,
Token,
Feature activation+1.281
see
Token see
Feature activation+0.732
F
Token F
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.085
s
Tokens
Feature activation+4.634
here
Token here
Feature activation+2.883
.
Token.
Feature activation+4.873
For
Token For
Feature activation+3.514
the
Token the
Feature activation+1.700
malware
Token malware
Feature activation+0.001
,
Token,
Feature activation+1.072
see
Token see
Feature activation+0.939
Du
Token Du
Feature activation+0.000

INTERVAL 2.786 - 3.343
CONTAINS 0.000%

"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.047
s
Tokens
Feature activation+4.462
here
Token here
Feature activation+2.424
.
Token.
Feature activation+4.739
For
Token For
Feature activation+3.210
other
Token other
Feature activation+1.064
uses
Token uses
Feature activation+1.442
,
Token,
Feature activation+0.999
see
Token see
Feature activation+0.141
Air
Token Air
Feature activation+0.000
or
Tokenor
Feature activation+0.000
ak
Tokenak
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.468
s
Tokens
Feature activation+4.646
here
Token here
Feature activation+2.883
.
Token.
Feature activation+5.476
For
Token For
Feature activation+3.657
other
Token other
Feature activation+1.579
uses
Token uses
Feature activation+2.105
,
Token,
Feature activation+1.543
D
TokenD
Feature activation+0.000
uku
Tokenuku
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.085
s
Tokens
Feature activation+4.634
here
Token here
Feature activation+2.883
.
Token.
Feature activation+4.873
For
Token For
Feature activation+3.514
the
Token the
Feature activation+1.700
malware
Token malware
Feature activation+0.001
,
Token,
Feature activation+1.072
Sea
TokenSea
Feature activation+0.000
elephant
Token elephant
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.318
s
Tokens
Feature activation+4.111
here
Token here
Feature activation+2.850
.
Token.
Feature activation+3.997
For
Token For
Feature activation+3.149
the
Token the
Feature activation+1.585
super
Token super
Feature activation+0.000
family
Tokenfamily
Feature activation+0.000
od
Tokenod
Feature activation+0.000
ai
Tokenai
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.818
s
Tokens
Feature activation+4.150
here
Token here
Feature activation+3.006
.
Token.
Feature activation+5.405
For
Token For
Feature activation+3.765
the
Token the
Feature activation+1.489
restaurant
Token restaurant
Feature activation+0.000
chain
Token chain
Feature activation+0.000

INTERVAL 2.229 - 2.786
CONTAINS 0.000%

uk
Tokenuk
Feature activation+0.000
let
Tokenlet
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.377
s
Tokens
Feature activation+4.567
here
Token here
Feature activation+2.376
.
Token.
Feature activation+5.106
For
Token For
Feature activation+3.285
the
Token the
Feature activation+0.000
fisher
Token fisher
Feature activation+0.000
y
Tokeny
Feature activation+0.000
The
TokenThe
Feature activation+0.000
Dale
Token Dale
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.249
s
Tokens
Feature activation+4.876
here
Token here
Feature activation+2.632
.
Token.
Feature activation+5.292
For
Token For
Feature activation+3.383
other
Token other
Feature activation+1.571
uses
Token uses
Feature activation+2.107
,
Token,
Feature activation+1.363
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.589
s
Tokens
Feature activation+3.586
here
Token here
Feature activation+1.507
.
Token.
Feature activation+4.377
For
Token For
Feature activation+2.275
the
Token the
Feature activation+0.000
song
Token song
Feature activation+0.000
by
Token by
Feature activation+0.000
Alt
Token Alt
Feature activation+0.000
+
Token+
Feature activation+0.000
in
Tokenin
Feature activation+0.000
ah
Tokenah
Feature activation+0.000
redirect
Token redirect
Feature activation+0.000
s
Tokens
Feature activation+1.360
here
Token here
Feature activation+1.253
.
Token.
Feature activation+2.645
For
Token For
Feature activation+1.101
other
Token other
Feature activation+0.000
uses
Token uses
Feature activation+0.526
,
Token,
Feature activation+0.000
see
Token see
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.885
s
Tokens
Feature activation+3.634
here
Token here
Feature activation+2.216
.
Token.
Feature activation+4.617
For
Token For
Feature activation+2.646
the
Token the
Feature activation+1.004
rapper
Token rapper
Feature activation+0.000
,
Token,
Feature activation+0.000
see
Token see
Feature activation+0.000
S
Token S
Feature activation+0.000

INTERVAL 1.671 - 2.229
CONTAINS 0.000%

C
Token C
Feature activation+0.000
ane
Tokenane
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.885
s
Tokens
Feature activation+3.634
here
Token here
Feature activation+2.216
.
Token.
Feature activation+4.617
For
Token For
Feature activation+2.646
the
Token the
Feature activation+1.004
rapper
Token rapper
Feature activation+0.000
,
Token,
Feature activation+0.000
redirect
Token redirect
Feature activation+1.242
s
Tokens
Feature activation+4.676
here
Token here
Feature activation+3.146
.
Token.
Feature activation+5.571
For
Token For
Feature activation+3.897
the
Token the
Feature activation+2.064
slang
Token slang
Feature activation+1.092
and
Token and
Feature activation+0.000
internet
Token internet
Feature activation+0.566
meme
Token meme
Feature activation+0.000
,
Token,
Feature activation+1.553
s
Tokens
Feature activation+4.646
here
Token here
Feature activation+2.883
.
Token.
Feature activation+5.476
For
Token For
Feature activation+3.657
other
Token other
Feature activation+1.579
uses
Token uses
Feature activation+2.105
,
Token,
Feature activation+1.543
see
Token see
Feature activation+0.519
An
Token An
Feature activation+0.000
or
Tokenor
Feature activation+0.000
ak
Tokenak
Feature activation+0.000
Bat
Token Bat
Feature activation+0.000
ista
Tokenista
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.366
s
Tokens
Feature activation+1.999
here
Token here
Feature activation+1.983
.
Token.
Feature activation+4.321
For
Token For
Feature activation+2.738
the
Token the
Feature activation+0.854
Brazilian
Token Brazilian
Feature activation+0.000
footballer
Token footballer
Feature activation+0.000
amb
Tokenamb
Feature activation+0.000
ig
Tokenig
Feature activation+0.000
uation
Tokenuation
Feature activation+0.389
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.129
Ċ
TokenĊ
Feature activation+1.729
park
Tokenpark
Feature activation+0.000
as
Tokenas
Feature activation+0.000
An
Token An
Feature activation+0.000
In
Token In
Feature activation+0.000
uit
Tokenuit
Feature activation+0.000

INTERVAL 1.114 - 1.671
CONTAINS 0.000%

amb
Tokenamb
Feature activation+0.000
ig
Tokenig
Feature activation+0.000
uation
Tokenuation
Feature activation+0.528
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.668
Ċ
TokenĊ
Feature activation+1.176
R
TokenR
Feature activation+0.310
och
Tokenoch
Feature activation+0.000
dale
Tokendale
Feature activation+0.000
Association
Token Association
Feature activation+0.000
Football
Token Football
Feature activation+0.000
redirect
Token redirect
Feature activation+0.818
s
Tokens
Feature activation+4.150
here
Token here
Feature activation+3.006
.
Token.
Feature activation+5.405
For
Token For
Feature activation+3.765
the
Token the
Feature activation+1.489
restaurant
Token restaurant
Feature activation+0.000
chain
Token chain
Feature activation+0.000
,
Token,
Feature activation+0.644
see
Token see
Feature activation+1.704
T
Token T
Feature activation+0.000
redirect
Token redirect
Feature activation+0.882
s
Tokens
Feature activation+4.138
here
Token here
Feature activation+2.587
.
Token.
Feature activation+5.129
For
Token For
Feature activation+3.534
other
Token other
Feature activation+1.373
uses
Token uses
Feature activation+2.069
,
Token,
Feature activation+1.574
see
Token see
Feature activation+0.593
Gal
Token Gal
Feature activation+0.000
v
Tokenv
Feature activation+0.000
dis
Tokendis
Feature activation+1.240
amb
Tokenamb
Feature activation+0.000
ig
Tokenig
Feature activation+0.000
uation
Tokenuation
Feature activation+0.389
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+1.129
Ċ
TokenĊ
Feature activation+1.729
park
Tokenpark
Feature activation+0.000
as
Tokenas
Feature activation+0.000
An
Token An
Feature activation+0.000
In
Token In
Feature activation+0.000
"
Token"
Feature activation+0.000
A
TokenA
Feature activation+0.000
uk
Tokenuk
Feature activation+0.000
let
Tokenlet
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+1.377
s
Tokens
Feature activation+4.567
here
Token here
Feature activation+2.376
.
Token.
Feature activation+5.106
For
Token For
Feature activation+3.285
the
Token the
Feature activation+0.000

INTERVAL 0.557 - 1.114
CONTAINS 0.000%

see
Token see
Feature activation+0.732
F
Token F
Feature activation+0.000
ove
Tokenove
Feature activation+0.000
a
Tokena
Feature activation+0.000
(
Token (
Feature activation+0.000
dis
Tokendis
Feature activation+0.643
amb
Tokenamb
Feature activation+0.000
ig
Tokenig
Feature activation+0.000
uation
Tokenuation
Feature activation+0.554
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.448
see
Token see
Feature activation+1.040
IBM
Token IBM
Feature activation+0.000
PC
Token PC
Feature activation+0.000
compatible
Token compatible
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.645
Ċ
TokenĊ
Feature activation+0.597
The
TokenThe
Feature activation+0.254
IBM
Token IBM
Feature activation+0.000
Personal
Token Personal
Feature activation+0.000
Computer
Token Computer
Feature activation+0.000
,
Token,
Feature activation+0.000
(
Token (
Feature activation+0.000
sl
Tokensl
Feature activation+0.000
ang
Tokenang
Feature activation+0.000
)
Token)
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.176
Ċ
TokenĊ
Feature activation+0.935
Tel
TokenTel
Feature activation+0.000
og
Tokenog
Feature activation+0.000
re
Tokenre
Feature activation+0.000
ika
Tokenika
Feature activation+0.000
(
Token (
Feature activation+0.403
"
Token"
Feature activation+0.000
Lord
TokenLord
Feature activation+0.000
Robert
Token Robert
Feature activation+0.000
Cecil
Token Cecil
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.960
s
Tokens
Feature activation+4.089
here
Token here
Feature activation+3.116
.
Token.
Feature activation+5.266
For
Token For
Feature activation+3.417
his
Token his
Feature activation+0.000
S
TokenS
Feature activation+0.000
utter
Tokenutter
Feature activation+0.000
C
Token C
Feature activation+0.000
ane
Tokenane
Feature activation+0.000
"
Token"
Feature activation+0.000
redirect
Token redirect
Feature activation+0.885
s
Tokens
Feature activation+3.634
here
Token here
Feature activation+2.216
.
Token.
Feature activation+4.617
For
Token For
Feature activation+2.646
the
Token the
Feature activation+1.004

INTERVAL 0.000 - 0.557
CONTAINS 99.999%

began
Token began
Feature activation+0.000
to
Token to
Feature activation+0.000
appear
Token appear
Feature activation+0.000
.
Token.
Feature activation+0.000
They
Token They
Feature activation+0.000
came
Token came
Feature activation+0.000
from
Token from
Feature activation+0.000
G
Token G
Feature activation+0.000
ull
Tokenull
Feature activation+0.000
Island
Token Island
Feature activation+0.000
,
Token,
Feature activation+0.000
that
Token that
Feature activation+0.000
of
Token of
Feature activation+0.000
other
Token other
Feature activation+0.000
Arab
Token Arab
Feature activation+0.000
-
Token-
Feature activation+0.000
Israel
TokenIsrael
Feature activation+0.000
is
Tokenis
Feature activation+0.000
,
Token,
Feature activation+0.000
he
Token he
Feature activation+0.000
said
Token said
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
t
Tokent
Feature activation+0.000
find
Token find
Feature activation+0.000
it
Token it
Feature activation+0.000
now
Token now
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Tweet
TokenTweet
Feature activation+0.000
#
Token #
Feature activation+0.000
contrast
Token contrast
Feature activation+0.000
to
Token to
Feature activation+0.000
hard
Token hard
Feature activation+0.000
-
Token-
Feature activation+0.000
line
Tokenline
Feature activation+0.000
protection
Token protection
Feature activation+0.000
ist
Tokenist
Feature activation+0.000
policies
Token policies
Feature activation+0.000
from
Token from
Feature activation+0.000
the
Token the
Feature activation+0.000
Chinese
Token Chinese
Feature activation+0.000
Nat
Token Nat
Feature activation+0.000
King
Token King
Feature activation+0.000
Cole
Token Cole
Feature activation+0.000
and
Token and
Feature activation+0.000
Sammy
Token Sammy
Feature activation+0.000
Davis
Token Davis
Feature activation+0.000
Jr
Token Jr
Feature activation+0.000
.
Token.
Feature activation+0.000
and
Token and
Feature activation+0.000
was
Token was
Feature activation+0.000
granted
Token granted
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 28: Ultra low frequency cluster

TOP ACTIVATIONS
MAX = 0.156

-
Token-
Feature activation+0.000
esque
Tokenesque
Feature activation+0.000
sens
Token sens
Feature activation+0.000
ibility
Tokenibility
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.156
more
Token more
Feature activation+0.000
than
Token than
Feature activation+0.000
one
Token one
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
that
Token that
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 1.989

whip
Token whip
Feature activation+0.129
-
Token-
Feature activation-0.001
smart
Tokensmart
Feature activation-0.012
dialogue
Token dialogue
Feature activation+0.032
,
Token,
Feature activation+0.013
a
Token a
Feature activation+1.989
Star
Token Star
Feature activation+0.048
Wars
Token Wars
Feature activation-0.005
-
Token-
Feature activation-0.005
esque
Tokenesque
Feature activation+0.031
sens
Token sens
Feature activation+0.029
Blue
Token Blue
Feature activation+0.005
Jays
Token Jays
Feature activation+0.003
âĢ
TokenâĢ
Feature activation-0.005
Ļ
TokenĻ
Feature activation+0.005
baseball
Token baseball
Feature activation+0.078
cap
Token cap
Feature activation+0.618
and
Token and
Feature activation+0.384
a
Token a
Feature activation-0.078
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Put
Token Put
Feature activation+0.057
a
Token a
Feature activation-0.106
Toronto
Token Toronto
Feature activation+0.090
Blue
Token Blue
Feature activation-0.000
Jays
Token Jays
Feature activation+0.006
âĢ
TokenâĢ
Feature activation+0.231
Ļ
TokenĻ
Feature activation-0.088
baseball
Token baseball
Feature activation+0.030
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Blue
Token Blue
Feature activation+0.005
Jays
Token Jays
Feature activation+0.002
âĢ
TokenâĢ
Feature activation+0.017
Ļ
TokenĻ
Feature activation+0.020
baseball
Token baseball
Feature activation+0.073
cap
Token cap
Feature activation+0.895
and
Token and
Feature activation-0.006
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.875
,
Token,
Feature activation-0.416
the
Token the
Feature activation-0.134
NB
Token NB
Feature activation+0.018
Space
Token Space
Feature activation+0.023
Race
Token Race
Feature activation-0.120
had
Token had
Feature activation+0.005
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.687
,
Token,
Feature activation-0.292
the
Token the
Feature activation-0.072
NB
Token NB
Feature activation+0.021
Space
Token Space
Feature activation+0.201
Race
Token Race
Feature activation-0.085
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
the
Token the
Feature activation-0.108
NB
Token NB
Feature activation+0.049
Space
Token Space
Feature activation+0.054
Race
Token Race
Feature activation-0.077
had
Token had
Feature activation-0.093
a
Token a
Feature activation+0.080
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Blue
Token Blue
Feature activation+0.028
Jays
Token Jays
Feature activation+0.116
âĢ
TokenâĢ
Feature activation+0.076
Ļ
TokenĻ
Feature activation+0.030
baseball
Token baseball
Feature activation+0.196
cap
Token cap
Feature activation+0.441
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
lofty
Token lofty
Feature activation+0.021
goal
Token goal
Feature activation-0.016
:
Token:
Feature activation-0.044
Put
Token Put
Feature activation+0.077
a
Token a
Feature activation-0.143
Toronto
Token Toronto
Feature activation+0.497
Blue
Token Blue
Feature activation-0.083
Jays
Token Jays
Feature activation+0.135
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.002
less
Token less
Feature activation-0.022
lofty
Token lofty
Feature activation+0.017
goal
Token goal
Feature activation-0.008
:
Token:
Feature activation-0.137
Put
Token Put
Feature activation+0.135
a
Token a
Feature activation+0.113
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
had
Token had
Feature activation+0.013
a
Token a
Feature activation-0.015
somewhat
Token somewhat
Feature activation+0.017
less
Token less
Feature activation-0.073
lofty
Token lofty
Feature activation+0.093
goal
Token goal
Feature activation+0.232
:
Token:
Feature activation+0.134
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.016
less
Token less
Feature activation-0.014
lofty
Token lofty
Feature activation+0.094
goal
Token goal
Feature activation-0.003
:
Token:
Feature activation-0.058
Put
Token Put
Feature activation+0.113
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
:
Token:
Feature activation-0.022
Put
Token Put
Feature activation+0.035
a
Token a
Feature activation-0.076
Toronto
Token Toronto
Feature activation+0.075
Blue
Token Blue
Feature activation-0.014
Jays
Token Jays
Feature activation+0.222
âĢ
TokenâĢ
Feature activation+0.032
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
lofty
Token lofty
Feature activation+0.009
goal
Token goal
Feature activation-0.006
:
Token:
Feature activation-0.008
Put
Token Put
Feature activation+0.020
a
Token a
Feature activation-0.064
Toronto
Token Toronto
Feature activation+0.753
Blue
Token Blue
Feature activation+0.111
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.010
less
Token less
Feature activation-0.006
lofty
Token lofty
Feature activation+0.014
goal
Token goal
Feature activation+0.001
:
Token:
Feature activation-0.023
Put
Token Put
Feature activation+0.151
a
Token a
Feature activation-0.088
Toronto
Token Toronto
Feature activation+0.052
Blue
Token Blue
Feature activation+0.003
Jays
Token Jays
Feature activation-0.012
âĢ
TokenâĢ
Feature activation+0.060
NB
Token NB
Feature activation+0.011
Space
Token Space
Feature activation+0.002
Race
Token Race
Feature activation-0.020
had
Token had
Feature activation-0.029
a
Token a
Feature activation-0.117
somewhat
Token somewhat
Feature activation+0.015
less
Token less
Feature activation-0.127
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
lofty
Token lofty
Feature activation+0.016
goal
Token goal
Feature activation-0.012
:
Token:
Feature activation-0.043
Put
Token Put
Feature activation+0.168
a
Token a
Feature activation-0.274
Toronto
Token Toronto
Feature activation+0.200
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
Race
Token Race
Feature activation-0.070
had
Token had
Feature activation-0.075
a
Token a
Feature activation-0.161
somewhat
Token somewhat
Feature activation-0.015
less
Token less
Feature activation-0.098
lofty
Token lofty
Feature activation+0.306
goal
Token goal
Feature activation+0.046
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
NB
Token NB
Feature activation+0.016
Space
Token Space
Feature activation+0.010
Race
Token Race
Feature activation-0.027
had
Token had
Feature activation-0.062
a
Token a
Feature activation-0.100
somewhat
Token somewhat
Feature activation+0.043
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
Race
Token Race
Feature activation-0.005
had
Token had
Feature activation-0.109
a
Token a
Feature activation-0.112
somewhat
Token somewhat
Feature activation-0.087
less
Token less
Feature activation-0.028
lofty
Token lofty
Feature activation+0.108
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Decoder Weights Distribution

Head 0: 0.09

Head 1: 0.07

Head 2: 0.09

Head 3: 0.10

Head 4: 0.09

Head 5: 0.08

Head 6: 0.08

Head 7: 0.07

Head 8: 0.09

Head 9: 0.07

Head 10: 0.08

Head 11: 0.08

Positive logits

Sketch1.75

heon1.63

nick1.59

isphere1.57

inged1.56

kit1.53

cookie1.53

icon1.52

��1.52

radius1.52

iery1.47

logos1.47

iven1.46

belt1.46

ETA1.46

Abbey1.46

enna1.45

din1.45

inus1.44

videos1.42

Negative logits

-2.24

'';-1.84

Own-1.71

……………………-1.64

Weak-1.60

vengeance-1.60

---------1.59

",-1.56

-------1.56

.�-1.56

SourceFile-1.55

ITNESS-1.54

-----1.53

aband-1.53

--+-1.52

ASHINGTON-1.52

«-1.51

ongevity-1.49

========-1.48

JUSTICE-1.47

INTERVAL 0.140 - 0.156
CONTAINS 0.000%

-
Token-
Feature activation+0.000
esque
Tokenesque
Feature activation+0.000
sens
Token sens
Feature activation+0.000
ibility
Tokenibility
Feature activation+0.000
,
Token,
Feature activation+0.000
and
Token and
Feature activation+0.156
more
Token more
Feature activation+0.000
than
Token than
Feature activation+0.000
one
Token one
Feature activation+0.000
relationship
Token relationship
Feature activation+0.000
that
Token that
Feature activation+0.000

INTERVAL 0.125 - 0.140
CONTAINS 0.000%

INTERVAL 0.109 - 0.125
CONTAINS 0.000%

INTERVAL 0.094 - 0.109
CONTAINS 0.000%

INTERVAL 0.078 - 0.094
CONTAINS 0.000%

INTERVAL 0.062 - 0.078
CONTAINS 0.000%

INTERVAL 0.047 - 0.062
CONTAINS 0.000%

INTERVAL 0.031 - 0.047
CONTAINS 0.000%

INTERVAL 0.016 - 0.031
CONTAINS 0.000%

INTERVAL 0.000 - 0.016
CONTAINS 100.000%

women
Token women
Feature activation+0.000
are
Token are
Feature activation+0.000
biologically
Token biologically
Feature activation+0.000
less
Token less
Feature activation+0.000
suited
Token suited
Feature activation+0.000
to
Token to
Feature activation+0.000
work
Token work
Feature activation+0.000
in
Token in
Feature activation+0.000
technology
Token technology
Feature activation+0.000
than
Token than
Feature activation+0.000
men
Token men
Feature activation+0.000
introduced
Token introduced
Feature activation+0.000
a
Token a
Feature activation+0.000
service
Token service
Feature activation+0.000
.
Token.
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
Maybe
TokenMaybe
Feature activation+0.000
this
Token this
Feature activation+0.000
FCC
Token FCC
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
We
TokenWe
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
re
Tokenre
Feature activation+0.000
very
Token very
Feature activation+0.000
proud
Token proud
Feature activation+0.000
of
Token of
Feature activation+0.000
Mar
Token Mar
Feature activation+0.000
-
Token-
Feature activation+0.000
H
Token H
Feature activation+0.000
ae
Tokenae
Feature activation+0.000
cker
Tokencker
Feature activation+0.000
singled
Token singled
Feature activation+0.000
through
Token through
Feature activation+0.000
the
Token the
Feature activation+0.000
right
Token right
Feature activation+0.000
side
Token side
Feature activation+0.000
and
Token and
Feature activation+0.000
E
Token E
Feature activation+0.000
bert
Tokenbert
Feature activation+0.000
naturally
Token naturally
Feature activation+0.000
there
Token there
Feature activation+0.000
and
Token and
Feature activation+0.000
how
Token how
Feature activation+0.000
much
Token much
Feature activation+0.000
has
Token has
Feature activation+0.000
come
Token come
Feature activation+0.000
out
Token out
Feature activation+0.000
of
Token of
Feature activation+0.000
the
Token the
Feature activation+0.000
hard
Token hard
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Feature 29: Dead

TOP ACTIVATIONS
MAX = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000

Top DFA by src position
MAX = 0.410

Ļ
TokenĻ
Feature activation-0.012
baseball
Token baseball
Feature activation-0.023
cap
Token cap
Feature activation-0.065
and
Token and
Feature activation-0.022
a
Token a
Feature activation-0.037
beer
Token beer
Feature activation+0.072
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
Jays
Token Jays
Feature activation-0.010
âĢ
TokenâĢ
Feature activation-0.019
Ļ
TokenĻ
Feature activation-0.022
baseball
Token baseball
Feature activation-0.022
cap
Token cap
Feature activation-0.107
and
Token and
Feature activation+0.185
a
Token a
Feature activation-0.079
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.006
less
Token less
Feature activation-0.002
lofty
Token lofty
Feature activation+0.019
goal
Token goal
Feature activation+0.003
:
Token:
Feature activation+0.015
Put
Token Put
Feature activation+0.030
a
Token a
Feature activation-0.032
Toronto
Token Toronto
Feature activation-0.066
Blue
Token Blue
Feature activation-0.007
Jays
Token Jays
Feature activation+0.020
âĢ
TokenâĢ
Feature activation-0.017
somewhat
Token somewhat
Feature activation-0.009
less
Token less
Feature activation-0.008
lofty
Token lofty
Feature activation-0.001
goal
Token goal
Feature activation-0.004
:
Token:
Feature activation+0.003
Put
Token Put
Feature activation+0.072
a
Token a
Feature activation-0.088
Toronto
Token Toronto
Feature activation-0.021
Blue
Token Blue
Feature activation-0.020
Jays
Token Jays
Feature activation-0.009
âĢ
TokenâĢ
Feature activation-0.013
<|endoftext|>
Token<|endoftext|>
Feature activation-1.140
,
Token,
Feature activation-0.688
the
Token the
Feature activation-0.071
NB
Token NB
Feature activation+0.029
Space
Token Space
Feature activation+0.011
Race
Token Race
Feature activation+0.410
had
Token had
Feature activation+0.168
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.073
,
Token,
Feature activation-0.566
the
Token the
Feature activation-0.076
NB
Token NB
Feature activation+0.029
Space
Token Space
Feature activation-0.086
Race
Token Race
Feature activation+0.072
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation-0.058
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.039
Race
Token Race
Feature activation+0.181
had
Token had
Feature activation-0.085
a
Token a
Feature activation+0.184
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-0.918
,
Token,
Feature activation-0.225
the
Token the
Feature activation-0.063
NB
Token NB
Feature activation-0.008
Space
Token Space
Feature activation+0.003
Race
Token Race
Feature activation+0.016
had
Token had
Feature activation-0.044
a
Token a
Feature activation-0.032
somewhat
Token somewhat
Feature activation-0.018
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.065
,
Token,
Feature activation-0.056
the
Token the
Feature activation-0.009
NB
Token NB
Feature activation-0.002
Space
Token Space
Feature activation+0.003
Race
Token Race
Feature activation+0.047
had
Token had
Feature activation+0.016
a
Token a
Feature activation+0.015
somewhat
Token somewhat
Feature activation-0.001
less
Token less
Feature activation+0.002
lofty
Token lofty
Feature activation+0.012
somewhat
Token somewhat
Feature activation+0.001
less
Token less
Feature activation+0.003
lofty
Token lofty
Feature activation+0.020
goal
Token goal
Feature activation-0.002
:
Token:
Feature activation+0.010
Put
Token Put
Feature activation+0.091
a
Token a
Feature activation-0.045
Toronto
Token Toronto
Feature activation+0.016
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Race
Token Race
Feature activation+0.078
had
Token had
Feature activation-0.039
a
Token a
Feature activation-0.007
somewhat
Token somewhat
Feature activation-0.002
less
Token less
Feature activation+0.004
lofty
Token lofty
Feature activation+0.081
goal
Token goal
Feature activation+0.066
:
Token:
Feature activation+0.069
Put
Token Put
Feature activation+0.053
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
less
Token less
Feature activation-0.021
lofty
Token lofty
Feature activation+0.006
goal
Token goal
Feature activation-0.008
:
Token:
Feature activation+0.040
Put
Token Put
Feature activation+0.033
a
Token a
Feature activation+0.156
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
:
Token:
Feature activation+0.002
Put
Token Put
Feature activation+0.043
a
Token a
Feature activation-0.027
Toronto
Token Toronto
Feature activation-0.061
Blue
Token Blue
Feature activation-0.035
Jays
Token Jays
Feature activation+0.182
âĢ
TokenâĢ
Feature activation+0.013
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
somewhat
Token somewhat
Feature activation-0.003
less
Token less
Feature activation-0.003
lofty
Token lofty
Feature activation+0.021
goal
Token goal
Feature activation+0.005
:
Token:
Feature activation+0.009
Put
Token Put
Feature activation+0.061
a
Token a
Feature activation-0.037
Toronto
Token Toronto
Feature activation+0.046
Blue
Token Blue
Feature activation-0.323
Jays
Token Jays
Feature activation-0.088
âĢ
TokenâĢ
Feature activation+0.000
Put
Token Put
Feature activation+0.067
a
Token a
Feature activation-0.040
Toronto
Token Toronto
Feature activation-0.045
Blue
Token Blue
Feature activation-0.017
Jays
Token Jays
Feature activation-0.011
âĢ
TokenâĢ
Feature activation+0.085
Ļ
TokenĻ
Feature activation-0.021
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.029
had
Token had
Feature activation+0.008
a
Token a
Feature activation+0.020
somewhat
Token somewhat
Feature activation+0.011
less
Token less
Feature activation+0.005
lofty
Token lofty
Feature activation+0.052
goal
Token goal
Feature activation+0.005
:
Token:
Feature activation+0.006
Put
Token Put
Feature activation+0.036
a
Token a
Feature activation-0.035
Toronto
Token Toronto
Feature activation-0.037
had
Token had
Feature activation-0.063
a
Token a
Feature activation-0.013
somewhat
Token somewhat
Feature activation-0.033
less
Token less
Feature activation-0.089
lofty
Token lofty
Feature activation+0.021
goal
Token goal
Feature activation+0.286
:
Token:
Feature activation-0.082
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.256
,
Token,
Feature activation-0.450
the
Token the
Feature activation-0.042
NB
Token NB
Feature activation-0.013
Space
Token Space
Feature activation-0.009
Race
Token Race
Feature activation+0.256
had
Token had
Feature activation-0.120
a
Token a
Feature activation-0.109
somewhat
Token somewhat
Feature activation-0.078
less
Token less
Feature activation-0.134
lofty
Token lofty
Feature activation+0.093
<|endoftext|>
Token<|endoftext|>
Feature activation-0.903
,
Token,
Feature activation-0.292
the
Token the
Feature activation-0.046
NB
Token NB
Feature activation-0.007
Space
Token Space
Feature activation+0.003
Race
Token Race
Feature activation+0.013
had
Token had
Feature activation-0.032
a
Token a
Feature activation-0.060
somewhat
Token somewhat
Feature activation-0.039
less
Token less
Feature activation-0.187
lofty
Token lofty
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation-1.009
,
Token,
Feature activation-0.354
the
Token the
Feature activation-0.048
NB
Token NB
Feature activation-0.037
Space
Token Space
Feature activation-0.004
Race
Token Race
Feature activation+0.068
had
Token had
Feature activation-0.043
a
Token a
Feature activation-0.069
somewhat
Token somewhat
Feature activation-0.188
less
Token less
Feature activation+0.010
lofty
Token lofty
Feature activation-0.048

Decoder Weights Distribution

Head 0: 0.08

Head 1: 0.07

Head 2: 0.07

Head 3: 0.09

Head 4: 0.08

Head 5: 0.08

Head 6: 0.09

Head 7: 0.10

Head 8: 0.08

Head 9: 0.09

Head 10: 0.09

Head 11: 0.09

Positive logits

inaction1.95

assurances1.87

persisted1.74

easing1.74

worsened1.73

heightened1.72

slack1.72

Fas1.72

funding1.71

cred1.70

vetoed1.67

deval1.67

budgets1.66

meddling1.64

scarce1.64

redist1.64

hampered1.63

rigging1.63

tightening1.63

punitive1.62

Negative logits

pill-2.19

estyles-2.06

hent-2.00

eworthy-1.96

enture-1.92

onyms-1.89

arious-1.86

arkable-1.84

rawdownloadcloneembedreportprint-1.82

pedia-1.81

tymology-1.81

amples-1.80

Orig-1.80

redients-1.80

origin-1.77

opsis-1.76

nant-1.76

igmatic-1.74

apple-1.68

xtap-1.67

INTERVAL 0.000 - 0.000
CONTAINS 100.000%

a
Token a
Feature activation+0.000
creative
Token creative
Feature activation+0.000
composition
Token composition
Feature activation+0.000
consisting
Token consisting
Feature activation+0.000
of
Token of
Feature activation+0.000
type
Token type
Feature activation+0.000
and
Token and
Feature activation+0.000
images
Token images
Feature activation+0.000
can
Token can
Feature activation+0.000
better
Token better
Feature activation+0.000
communicate
Token communicate
Feature activation+0.000
Russia
Token Russia
Feature activation+0.000
said
Token said
Feature activation+0.000
it
Token it
Feature activation+0.000
had
Token had
Feature activation+0.000
lost
Token lost
Feature activation+0.000
contact
Token contact
Feature activation+0.000
with
Token with
Feature activation+0.000
the
Token the
Feature activation+0.000
satellite
Token satellite
Feature activation+0.000
,
Token,
Feature activation+0.000
raising
Token raising
Feature activation+0.000
VR
Token VR
Feature activation+0.000
games
Token games
Feature activation+0.000
,
Token,
Feature activation+0.000
of
Token of
Feature activation+0.000
which
Token which
Feature activation+0.000
100
Token 100
Feature activation+0.000
or
Token or
Feature activation+0.000
more
Token more
Feature activation+0.000
are
Token are
Feature activation+0.000
currently
Token currently
Feature activation+0.000
in
Token in
Feature activation+0.000
Ryan
Token Ryan
Feature activation+0.000
wrote
Token wrote
Feature activation+0.000
on
Token on
Feature activation+0.000
Twitter
Token Twitter
Feature activation+0.000
.
Token.
Feature activation+0.000
âĢ
Token âĢ
Feature activation+0.000
ľ
Tokenľ
Feature activation+0.000
The
TokenThe
Feature activation+0.000
people
Token people
Feature activation+0.000
of
Token of
Feature activation+0.000
Sutherland
Token Sutherland
Feature activation+0.000
Ċ
TokenĊ
Feature activation+0.000
"
Token"
Feature activation+0.000
You
TokenYou
Feature activation+0.000
don
Token don
Feature activation+0.000
't
Token't
Feature activation+0.000
hurt
Token hurt
Feature activation+0.000
the
Token the
Feature activation+0.000
legislature
Token legislature
Feature activation+0.000
,
Token,
Feature activation+0.000
you
Token you
Feature activation+0.000
don
Token don
Feature activation+0.000

BOTTOM ACTIVATIONS
MIN = 0.000

Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
space
Tokenspace
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
-
Token-
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
into
Token into
Feature activation+0.000
near
Token near
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
<|endoftext|>
Token<|endoftext|>
Feature activation+0.000
,
Token,
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
the
Token the
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
NB
Token NB
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
âĢ
TokenâĢ
Feature activation+0.000
Ļ
TokenĻ
Feature activation+0.000
baseball
Token baseball
Feature activation+0.000
cap
Token cap
Feature activation+0.000
and
Token and
Feature activation+0.000
a
Token a
Feature activation+0.000
beer
Token beer
Feature activation+0.000
cooler
Token cooler
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Jays
Token Jays
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000
Blue
Token Blue
Feature activation+0.000
Space
Token Space
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Race
Token Race
Feature activation+0.000
had
Token had
Feature activation+0.000
a
Token a
Feature activation+0.000
somewhat
Token somewhat
Feature activation+0.000
less
Token less
Feature activation+0.000
lofty
Token lofty
Feature activation+0.000
goal
Token goal
Feature activation+0.000
:
Token:
Feature activation+0.000
Put
Token Put
Feature activation+0.000
a
Token a
Feature activation+0.000
Toronto
Token Toronto
Feature activation+0.000