TOP ACTIVATIONS
MAX = 9.475

long walks . Whenever the m ule became tired , she
Red thought about what the m op had said , and
day on , whenever the m op wasn 't available ,
all clean . After the m op had finished it would
room . Every morning the m op would wake up ready
while , sure enough the m itten floated down out of
before heading home . The m ule was friendly and smiled
! The m op would enjoy the bright
the next morning , the m ama woke the baby up
that day forward , the m ule and the little boy
itten . Now that the m itten was full , Daddy
's cleaning adventure . The m op had been released and
cream ! â The m ama said ,
. The m ittens were so thin ;
he never forgot what the m op taught him
sharing is good . The m op was not selfish anymore
were having trouble . The m ule saw this and wanted
were having trouble . The m ule saw this and wanted
purple m op . The m op lived in a big
fun !" The m op thought about it .

MIN ACTIVATIONS
MIN = -9.032

" W oof , wo of !" Spot bark
" W oof , wo of , I 'm
ed , " W oof , wo of !" Tom knew
ed , " W oof , wo of !" They both
says , " W oof , wo of , Tom !
said , " W oof , wo of !"
tail . " W oof , wo of !"
says , " W oof , wo of , my name
said , " W oof , wo of !" The leash
says , " W oof , wo of . You are
said , " W oof , wo of , wo of
" W oof , wo of !" Spot bark
said , " W oof , wo of , sorry ,
" W oof , wo of !" the dolls
says , " W oof , wo of . I understand
!" " Bang , bang , bang !" Sam
. " Bang , bang , bang !" Tom
neck l aces , grape brace lets , and grape crown
says , " W oof , wo of , hello ,
. " Bang , bang , you 're dead

INTERVAL 4.848 - 9.475
CONTAINS 0.097%

him a coat . He must be cold ."
were so excited that the m ule was helping them .
." The elderly person replied , " No
They played in the o asis for a while .
down for his nap . While he was lying down he

INTERVAL 0.221 - 4.848
CONTAINS 46.458%

truck and asks for mint ice cream . The ice cream
upon a time , there was a big elephant . He
very excited . She wanted to see the mountain up close
. Mom smiled and went back to the kitchen . She
to be saved . But no one came to help .

INTERVAL -4.405 - 0.221
CONTAINS 52.889%

Tom and Lily nod . They are OK . They are
could imagine how it would be to be as big as
" Oh no !" Lily and Ben say . They
surprise for Lily - a new doll to play with at
in . She soon realized she couldn 't swim and started

INTERVAL -9.032 - -4.405
CONTAINS 0.556%

truck !" Anna says . " It has a bow like
mom hugged them and said , " I forgive you ,
lion !" he said . " It is so cool !"
other ," Ben says . " We are twins . We
" This is ind igo , and this is violet .
Token long
Token ID890
Feature activation-1.67e+00

Pos logit contributions
ister+0.026
ths+0.025
ww+0.025
www+0.025
 avail+0.024
Neg logit contributions
 like-0.015
 everywhere-0.010
 favorite-0.009
 liked-0.009
 around-0.009
Token walks
Token ID11114
Feature activation-3.39e-01

Pos logit contributions
 like+0.040
 liked+0.030
 favorite+0.030
 everywhere+0.029
 back+0.028
Neg logit contributions
ww-0.037
ths-0.036
ister-0.036
 service-0.035
www-0.035
Token.
Token ID13
Feature activation-6.93e-02

Pos logit contributions
 like+0.004
 pad+0.002
 favorite+0.002
 dolls+0.002
 liked+0.002
Neg logit contributions
ister-0.011
ww-0.010
ths-0.010
 service-0.010
www-0.010
Token Whenever
Token ID21326
Feature activation+3.63e+00

Pos logit contributions
 like+0.001
 everywhere+0.001
 favorite+0.001
 back+0.001
 dolls+0.001
Neg logit contributions
ister-0.002
ww-0.002
www-0.002
 avail-0.002
 storm-0.002
Token the
Token ID262
Feature activation+1.98e+00

Pos logit contributions
ths+0.119
ww+0.118
ister+0.117
www+0.116
rim+0.113
Neg logit contributions
 like-0.060
 liked-0.038
 favorite-0.032
 everywhere-0.031
 all-0.031
Token m
Token ID285
Feature activation+9.48e+00

Pos logit contributions
ister+0.057
 avail+0.055
ths+0.054
ww+0.054
keys+0.053
Neg logit contributions
 like-0.039
 favorite-0.029
 liked-0.028
 back-0.026
 t-0.026
Tokenule
Token ID2261
Feature activation+5.42e+00

Pos logit contributions
 avail+0.146
ww+0.126
 storm+0.118
www+0.111
 explosion+0.111
Neg logit contributions
 like-0.298
 favorite-0.262
 liked-0.254
 pad-0.247
 and-0.245
Token became
Token ID2627
Feature activation+8.09e-01

Pos logit contributions
ww+0.171
 avail+0.170
ister+0.168
www+0.167
ths+0.163
Neg logit contributions
 like-0.096
 liked-0.069
 favorite-0.060
 everywhere-0.058
 around-0.055
Token tired
Token ID10032
Feature activation+2.13e+00

Pos logit contributions
ister+0.022
ww+0.022
ths+0.021
www+0.021
 service+0.021
Neg logit contributions
 like-0.017
 everywhere-0.012
 favorite-0.012
 liked-0.012
 around-0.011
Token,
Token ID11
Feature activation+7.33e-01

Pos logit contributions
ww+0.085
www+0.085
ths+0.084
ister+0.082
 sermon+0.081
Neg logit contributions
 like-0.027
 everywhere-0.010
 but-0.010
 and-0.010
 all-0.009
Token she
Token ID673
Feature activation+4.79e-01

Pos logit contributions
ww+0.026
ister+0.026
www+0.026
ths+0.025
vation+0.025
Neg logit contributions
 like-0.012
 liked-0.006
 all-0.006
 everywhere-0.006
 favorite-0.006
TokenRed
Token ID7738
Feature activation+1.22e+00

Pos logit contributions
ister+0.020
ww+0.020
ths+0.020
www+0.019
 service+0.019
Neg logit contributions
 like-0.021
 everywhere-0.015
 liked-0.015
 favorite-0.015
 around-0.014
Token thought
Token ID1807
Feature activation+7.13e-01

Pos logit contributions
 avail+0.036
ister+0.036
ww+0.035
ths+0.035
 service+0.035
Neg logit contributions
 like-0.023
 liked-0.017
 favorite-0.014
 everywhere-0.014
 around-0.014
Token about
Token ID546
Feature activation+6.90e-01

Pos logit contributions
ww+0.021
ister+0.021
ths+0.021
www+0.021
 avail+0.021
Neg logit contributions
 like-0.013
 everywhere-0.008
 favorite-0.008
 liked-0.008
 around-0.007
Token what
Token ID644
Feature activation+3.34e-02

Pos logit contributions
ister+0.021
ww+0.021
ths+0.021
www+0.020
 avail+0.020
Neg logit contributions
 like-0.013
 everywhere-0.008
 liked-0.008
 favorite-0.008
 all-0.007
Token the
Token ID262
Feature activation+9.03e-01

Pos logit contributions
ister+0.001
ths+0.001
ww+0.001
www+0.001
 avail+0.001
Neg logit contributions
 like-0.001
 liked-0.000
 favorite-0.000
 everywhere-0.000
 all-0.000
Token m
Token ID285
Feature activation+9.20e+00

Pos logit contributions
ister+0.024
ww+0.024
ths+0.024
keys+0.023
 avail+0.023
Neg logit contributions
 like-0.020
 favorite-0.014
 liked-0.014
 everywhere-0.013
 dolls-0.013
Tokenop
Token ID404
Feature activation+3.48e+00

Pos logit contributions
 avail+0.256
ww+0.224
 hawk+0.218
ation+0.215
 storm+0.213
Neg logit contributions
 like-0.185
 favorite-0.147
 dolls-0.141
aces-0.136
 liked-0.133
Token had
Token ID550
Feature activation-2.02e+00

Pos logit contributions
 avail+0.105
ww+0.104
ths+0.102
ister+0.100
www+0.099
Neg logit contributions
 like-0.067
 liked-0.046
 everywhere-0.043
 around-0.042
 favorite-0.040
Token said
Token ID531
Feature activation-1.30e+00

Pos logit contributions
 like+0.046
 favorite+0.034
 everywhere+0.034
 liked+0.034
 dolls+0.032
Neg logit contributions
ister-0.047
ww-0.047
www-0.046
ths-0.046
 service-0.045
Token,
Token ID11
Feature activation-4.34e+00

Pos logit contributions
 like+0.017
 favorite+0.010
 everywhere+0.010
 liked+0.009
 dolls+0.008
Neg logit contributions
ister-0.043
 service-0.041
ths-0.041
ww-0.041
www-0.040
Token and
Token ID290
Feature activation+1.34e+00

Pos logit contributions
 like+0.052
 everywhere+0.025
 favorite+0.024
 liked+0.023
 dolls+0.018
Neg logit contributions
ister-0.154
ww-0.154
ths-0.152
www-0.151
 service-0.148
Token day
Token ID1110
Feature activation+6.02e-02

Pos logit contributions
ister+0.017
ths+0.017
 avail+0.017
ww+0.017
www+0.016
Neg logit contributions
 like-0.019
 everywhere-0.015
 liked-0.014
 favorite-0.014
 around-0.014
Token on
Token ID319
Feature activation+1.48e+00

Pos logit contributions
ister+0.002
ww+0.002
ths+0.002
www+0.002
 avail+0.002
Neg logit contributions
 like-0.001
 everywhere-0.000
 liked-0.000
 favorite-0.000
 both-0.000
Token,
Token ID11
Feature activation+9.59e-01

Pos logit contributions
ister+0.051
ww+0.051
ths+0.051
 sermon+0.051
www+0.051
Neg logit contributions
 like-0.020
 everywhere-0.011
 liked-0.009
 favorite-0.009
 around-0.008
Token whenever
Token ID8797
Feature activation+2.83e+00

Pos logit contributions
ww+0.025
ths+0.025
 storm+0.024
www+0.023
body+0.023
Neg logit contributions
 like-0.016
 pad-0.014
aces-0.014
 ti-0.011
tie-0.011
Token the
Token ID262
Feature activation+1.88e+00

Pos logit contributions
ister+0.104
ww+0.103
ths+0.102
www+0.101
 sermon+0.099
Neg logit contributions
 like-0.034
 everywhere-0.015
 liked-0.015
 favorite-0.015
 dolls-0.012
Token m
Token ID285
Feature activation+9.17e+00

Pos logit contributions
ister+0.046
ww+0.046
ths+0.045
www+0.044
 avail+0.044
Neg logit contributions
 like-0.045
 favorite-0.033
 liked-0.033
 everywhere-0.032
 dolls-0.031
Tokenop
Token ID404
Feature activation+4.14e+00

Pos logit contributions
 avail+0.267
ww+0.232
www+0.219
 storm+0.215
 hawk+0.215
Neg logit contributions
 like-0.177
 favorite-0.145
 dolls-0.144
 liked-0.133
 both-0.133
Token wasn
Token ID2492
Feature activation+3.56e+00

Pos logit contributions
 avail+0.128
ww+0.124
www+0.119
ths+0.115
izing+0.114
Neg logit contributions
 like-0.082
 liked-0.062
 around-0.056
 everywhere-0.054
 all-0.053
Token't
Token ID470
Feature activation+9.80e-01

Pos logit contributions
ww+0.076
ths+0.075
 avail+0.074
www+0.074
 service+0.072
Neg logit contributions
 like-0.098
 everywhere-0.073
 both-0.073
 all-0.071
 favorite-0.070
Token available
Token ID1695
Feature activation-2.68e-01

Pos logit contributions
ister+0.027
ww+0.027
ths+0.026
www+0.026
 service+0.025
Neg logit contributions
 like-0.021
 everywhere-0.014
 liked-0.014
 favorite-0.014
 around-0.013
Token,
Token ID11
Feature activation+5.90e-01

Pos logit contributions
 like+0.004
 liked+0.003
 favorite+0.003
 everywhere+0.003
 pad+0.002
Neg logit contributions
ister-0.008
ths-0.008
ww-0.008
 service-0.008
www-0.008
Token all
Token ID477
Feature activation+1.63e-01

Pos logit contributions
 like+0.003
 everywhere+0.002
 favorite+0.002
 liked+0.002
 pad+0.002
Neg logit contributions
ister-0.005
ths-0.005
ww-0.005
 service-0.005
www-0.005
Token clean
Token ID3424
Feature activation-1.53e+00

Pos logit contributions
ww+0.005
ister+0.005
ths+0.004
www+0.004
 service+0.004
Neg logit contributions
 like-0.003
 favorite-0.002
 everywhere-0.002
 liked-0.002
 pad-0.002
Token.
Token ID13
Feature activation+8.91e-01

Pos logit contributions
 like+0.019
 everywhere+0.009
 favorite+0.009
 liked+0.009
 pad+0.007
Neg logit contributions
ister-0.054
ww-0.053
ths-0.053
www-0.052
 service-0.052
Token After
Token ID2293
Feature activation+2.38e+00

Pos logit contributions
ister+0.025
ww+0.025
ths+0.025
www+0.025
 service+0.024
Neg logit contributions
 like-0.017
 everywhere-0.011
 favorite-0.011
 liked-0.011
 dolls-0.010
Token the
Token ID262
Feature activation+1.05e+00

Pos logit contributions
ister+0.087
ww+0.086
www+0.086
 avail+0.084
asis+0.082
Neg logit contributions
 like-0.032
 all-0.018
 everywhere-0.017
 both-0.016
 liked-0.014
Token m
Token ID285
Feature activation+9.10e+00

Pos logit contributions
ister+0.027
ww+0.027
 avail+0.027
ths+0.027
www+0.026
Neg logit contributions
 like-0.024
 favorite-0.018
 love-0.017
 liked-0.017
 everywhere-0.017
Tokenop
Token ID404
Feature activation+2.96e+00

Pos logit contributions
 avail+0.256
ww+0.222
 hawk+0.214
 storm+0.214
ation+0.210
Neg logit contributions
 like-0.184
 favorite-0.140
 dolls-0.132
 both-0.130
 liked-0.129
Token had
Token ID550
Feature activation-1.11e+00

Pos logit contributions
 avail+0.094
ww+0.093
www+0.091
ister+0.090
ths+0.088
Neg logit contributions
 like-0.052
 liked-0.035
 everywhere-0.033
 around-0.033
 all-0.032
Token finished
Token ID5201
Feature activation+3.35e+00

Pos logit contributions
 like+0.028
 favorite+0.021
 everywhere+0.021
 liked+0.021
 wherever+0.020
Neg logit contributions
ww-0.024
ister-0.024
ths-0.024
www-0.023
 service-0.023
Token it
Token ID340
Feature activation+1.51e+00

Pos logit contributions
ww+0.115
ister+0.113
www+0.113
ths+0.109
 beast+0.108
Neg logit contributions
 like-0.051
 everywhere-0.029
 favorite-0.027
 liked-0.026
 all-0.025
Token would
Token ID561
Feature activation+1.75e+00

Pos logit contributions
ww+0.054
ister+0.054
www+0.053
ths+0.052
 avail+0.052
Neg logit contributions
 like-0.020
 everywhere-0.011
 favorite-0.009
 liked-0.009
 back-0.008
Token room
Token ID2119
Feature activation-5.19e-01

Pos logit contributions
 like+0.016
 everywhere+0.012
 favorite+0.012
 liked+0.012
 around+0.011
Neg logit contributions
ister-0.015
ww-0.015
ths-0.014
www-0.014
 service-0.014
Token.
Token ID13
Feature activation+4.00e-01

Pos logit contributions
 like+0.007
 pad+0.004
 everywhere+0.004
 favorite+0.004
aces+0.004
Neg logit contributions
ister-0.017
 service-0.017
ww-0.016
ths-0.016
www-0.016
Token Every
Token ID3887
Feature activation+3.49e+00

Pos logit contributions
ister+0.011
ww+0.011
www+0.011
ths+0.011
 service+0.011
Neg logit contributions
 like-0.008
 everywhere-0.005
 favorite-0.005
 liked-0.005
 back-0.005
Token morning
Token ID3329
Feature activation+3.20e+00

Pos logit contributions
keys+0.111
ths+0.108
izing+0.107
 avail+0.106
irens+0.106
Neg logit contributions
 like-0.069
 liked-0.051
 favorite-0.050
 all-0.046
 everywhere-0.046
Token the
Token ID262
Feature activation+1.87e+00

Pos logit contributions
ister+0.108
ths+0.107
ww+0.107
www+0.105
 avail+0.103
Neg logit contributions
 like-0.048
 liked-0.026
 everywhere-0.026
 favorite-0.024
 around-0.022
Token m
Token ID285
Feature activation+9.04e+00

Pos logit contributions
ister+0.048
ww+0.047
ths+0.047
 avail+0.047
www+0.046
Neg logit contributions
 like-0.042
 favorite-0.031
 liked-0.031
 everywhere-0.030
 dolls-0.029
Tokenop
Token ID404
Feature activation+3.60e+00

Pos logit contributions
 avail+0.259
ww+0.223
 storm+0.221
 service+0.214
www+0.211
Neg logit contributions
 like-0.183
 favorite-0.143
 dolls-0.138
 liked-0.134
 both-0.132
Token would
Token ID561
Feature activation+2.69e+00

Pos logit contributions
ww+0.113
 avail+0.113
www+0.111
ister+0.110
ths+0.109
Neg logit contributions
 like-0.063
 liked-0.043
 everywhere-0.040
 favorite-0.038
 around-0.038
Token wake
Token ID7765
Feature activation+2.76e+00

Pos logit contributions
ww+0.072
ths+0.071
ister+0.070
www+0.069
 avail+0.067
Neg logit contributions
 like-0.061
 liked-0.042
 everywhere-0.041
 favorite-0.040
 all-0.038
Token up
Token ID510
Feature activation+2.96e-01

Pos logit contributions
ww+0.100
www+0.097
ister+0.097
ths+0.096
 service+0.093
Neg logit contributions
 like-0.042
 everywhere-0.022
 favorite-0.022
 around-0.022
 all-0.022
Token ready
Token ID3492
Feature activation+8.95e-01

Pos logit contributions
ww+0.010
ister+0.010
www+0.010
ths+0.010
 avail+0.009
Neg logit contributions
 like-0.005
 everywhere-0.003
 favorite-0.003
 liked-0.003
 all-0.002
Token while
Token ID981
Feature activation+3.19e+00

Pos logit contributions
ister+0.046
ww+0.043
irens+0.043
ths+0.043
keys+0.042
Neg logit contributions
 like-0.024
 but-0.013
 everywhere-0.012
 around-0.012
 liked-0.012
Token,
Token ID11
Feature activation+1.24e+00

Pos logit contributions
ww+0.116
ister+0.115
www+0.115
 sermon+0.113
ths+0.112
Neg logit contributions
 like-0.043
 everywhere-0.022
 around-0.018
 wherever-0.018
 all-0.018
Token sure
Token ID1654
Feature activation+2.21e+00

Pos logit contributions
ister+0.035
ww+0.034
www+0.033
ths+0.033
 service+0.033
Neg logit contributions
 like-0.026
 everywhere-0.018
 favorite-0.017
 liked-0.017
 around-0.017
Token enough
Token ID1576
Feature activation+3.23e+00

Pos logit contributions
ths+0.062
 sermon+0.059
ister+0.059
izing+0.059
ww+0.059
Neg logit contributions
 like-0.045
 all-0.032
 found-0.032
 everywhere-0.031
 back-0.030
Token the
Token ID262
Feature activation+2.52e+00

Pos logit contributions
ww+0.112
ister+0.112
ths+0.111
www+0.110
 sermon+0.109
Neg logit contributions
 like-0.044
 everywhere-0.024
 favorite-0.020
 liked-0.020
 wherever-0.019
Token m
Token ID285
Feature activation+8.98e+00

Pos logit contributions
 avail+0.060
ww+0.060
www+0.060
ths+0.060
ister+0.060
Neg logit contributions
 like-0.060
 favorite-0.046
 back-0.045
 love-0.045
 liked-0.044
Tokenitten
Token ID2621
Feature activation+5.02e+00

Pos logit contributions
 avail+0.156
ww+0.126
www+0.120
izing+0.111
ation+0.106
Neg logit contributions
 like-0.269
 favorite-0.238
aces-0.231
 pad-0.223
 back-0.222
Token floated
Token ID29742
Feature activation+3.67e-01

Pos logit contributions
ww+0.146
 avail+0.145
ister+0.144
www+0.142
ths+0.140
Neg logit contributions
 like-0.098
 liked-0.069
 everywhere-0.066
 favorite-0.064
 around-0.063
Token down
Token ID866
Feature activation+3.29e-01

Pos logit contributions
ister+0.010
ww+0.010
ths+0.010
www+0.010
 avail+0.010
Neg logit contributions
 like-0.006
 favorite-0.004
 dolls-0.004
 liked-0.004
aces-0.004
Token out
Token ID503
Feature activation-5.10e-01

Pos logit contributions
ister+0.011
ww+0.011
ths+0.011
www+0.011
 service+0.010
Neg logit contributions
 like-0.005
 favorite-0.003
 liked-0.003
 everywhere-0.003
 dolls-0.002
Token of
Token ID286
Feature activation-3.88e-01

Pos logit contributions
 like+0.010
 everywhere+0.006
 liked+0.006
 favorite+0.006
 around+0.006
Neg logit contributions
ww-0.015
ister-0.015
www-0.015
ths-0.015
 avail-0.015
Token before
Token ID878
Feature activation+3.20e+00

Pos logit contributions
ww+0.103
ister+0.102
ths+0.101
www+0.100
 avail+0.099
Neg logit contributions
 like-0.034
 everywhere-0.015
 liked-0.015
 favorite-0.014
 around-0.012
Token heading
Token ID9087
Feature activation+1.29e+00

Pos logit contributions
ww+0.091
www+0.088
ths+0.086
ister+0.086
body+0.085
Neg logit contributions
 like-0.069
 everywhere-0.048
 all-0.047
 both-0.045
 but-0.044
Token home
Token ID1363
Feature activation+2.22e+00

Pos logit contributions
ister+0.041
ww+0.041
ths+0.040
www+0.040
 service+0.040
Neg logit contributions
 like-0.020
 liked-0.012
 favorite-0.012
 everywhere-0.012
 dolls-0.010
Token.
Token ID13
Feature activation+1.23e+00

Pos logit contributions
ww+0.083
ister+0.082
ths+0.081
www+0.081
 avail+0.079
Neg logit contributions
 like-0.024
 everywhere-0.010
 favorite-0.009
 liked-0.009
 around-0.007
Token The
Token ID383
Feature activation+1.31e+00

Pos logit contributions
ister+0.034
ww+0.034
ths+0.034
www+0.033
 service+0.032
Neg logit contributions
 like-0.025
 everywhere-0.017
 liked-0.017
 favorite-0.017
 wherever-0.015
Token m
Token ID285
Feature activation+8.93e+00

Pos logit contributions
ister+0.033
ww+0.032
ths+0.032
 avail+0.032
www+0.031
Neg logit contributions
 like-0.030
 liked-0.022
 favorite-0.022
 everywhere-0.022
 back-0.021
Tokenule
Token ID2261
Feature activation+4.57e+00

Pos logit contributions
 avail+0.142
ww+0.120
 storm+0.115
www+0.100
ths+0.094
Neg logit contributions
 like-0.276
 liked-0.236
 favorite-0.236
 and-0.232
 pad-0.232
Token was
Token ID373
Feature activation+1.48e+00

Pos logit contributions
 avail+0.142
ww+0.141
ister+0.139
www+0.135
ths+0.133
Neg logit contributions
 like-0.082
 liked-0.061
 favorite-0.051
 everywhere-0.051
 around-0.050
Token friendly
Token ID8030
Feature activation-1.85e+00

Pos logit contributions
ister+0.045
ww+0.044
ths+0.044
www+0.043
 service+0.042
Neg logit contributions
 like-0.026
 everywhere-0.017
 favorite-0.016
 liked-0.016
 around-0.014
Token and
Token ID290
Feature activation+1.49e+00

Pos logit contributions
 like+0.024
 favorite+0.017
 liked+0.014
 everywhere+0.014
aces+0.014
Neg logit contributions
 service-0.056
ister-0.056
ths-0.056
ww-0.054
 beast-0.054
Token smiled
Token ID13541
Feature activation-1.68e+00

Pos logit contributions
ister+0.046
ww+0.044
ths+0.043
irens+0.043
 avail+0.043
Neg logit contributions
 like-0.030
 liked-0.021
 favorite-0.019
 everywhere-0.019
 all-0.018
Token!
Token ID0
Feature activation+8.48e-01

Pos logit contributions
ister+0.090
ww+0.090
ths+0.089
www+0.088
 avail+0.086
Neg logit contributions
 like-0.026
 everywhere-0.010
 favorite-0.009
 liked-0.009
 around-0.006
Token
Token ID220
Feature activation+2.40e-01

Pos logit contributions
ister+0.027
ww+0.027
ths+0.026
www+0.026
 service+0.026
Neg logit contributions
 like-0.013
 everywhere-0.008
 favorite-0.008
 liked-0.008
 around-0.007
Token
Token ID198
Feature activation-2.03e-01

Pos logit contributions
ister+0.006
ama+0.006
oo+0.006
ww+0.006
iren+0.006
Neg logit contributions
 like-0.005
 liked-0.003
 favorite-0.003
 pad-0.003
 ti-0.003
Token
Token ID198
Feature activation+6.27e-01

Pos logit contributions
 like+0.003
 favorite+0.002
 everywhere+0.002
 liked+0.002
aces+0.002
Neg logit contributions
ister-0.006
ww-0.006
www-0.006
ths-0.006
 service-0.006
TokenThe
Token ID464
Feature activation+1.37e+00

Pos logit contributions
ister+0.015
ww+0.015
www+0.014
ths+0.014
 service+0.014
Neg logit contributions
 like-0.015
 everywhere-0.011
 favorite-0.011
 liked-0.011
 back-0.011
Token m
Token ID285
Feature activation+8.84e+00

Pos logit contributions
ister+0.033
ww+0.033
ths+0.032
www+0.032
 avail+0.032
Neg logit contributions
 like-0.033
 favorite-0.024
 liked-0.024
 everywhere-0.024
 dolls-0.022
Tokenop
Token ID404
Feature activation+3.27e+00

Pos logit contributions
 avail+0.234
ww+0.206
 storm+0.203
 service+0.196
www+0.193
Neg logit contributions
 like-0.195
 favorite-0.152
 dolls-0.147
 liked-0.147
 everywhere-0.141
Token would
Token ID561
Feature activation+2.23e+00

Pos logit contributions
 avail+0.101
ww+0.101
www+0.098
ister+0.097
ths+0.097
Neg logit contributions
 like-0.059
 liked-0.042
 everywhere-0.037
 favorite-0.036
 around-0.036
Token enjoy
Token ID2883
Feature activation+6.68e-01

Pos logit contributions
ww+0.060
ister+0.059
ths+0.059
www+0.057
 avail+0.056
Neg logit contributions
 like-0.050
 liked-0.035
 everywhere-0.034
 favorite-0.033
 both-0.032
Token the
Token ID262
Feature activation-1.04e+00

Pos logit contributions
ister+0.020
ww+0.020
ths+0.019
www+0.019
 service+0.019
Neg logit contributions
 like-0.012
 everywhere-0.008
 favorite-0.008
 liked-0.008
 around-0.007
Token bright
Token ID6016
Feature activation+4.69e-01

Pos logit contributions
 like+0.026
 favorite+0.019
 everywhere+0.018
 liked+0.018
 love+0.018
Neg logit contributions
ths-0.026
ister-0.026
ww-0.025
 avail-0.024
www-0.024
Token the
Token ID262
Feature activation+1.90e+00

Pos logit contributions
keys+0.072
www+0.072
ister+0.072
ww+0.072
 sermon+0.071
Neg logit contributions
 like-0.027
 all-0.020
 everywhere-0.017
 both-0.017
 where-0.015
Token next
Token ID1306
Feature activation+2.70e+00

Pos logit contributions
ister+0.061
ww+0.058
www+0.057
keys+0.056
ment+0.056
Neg logit contributions
 like-0.031
 favorite-0.025
 everywhere-0.024
 back-0.024
 t-0.024
Token morning
Token ID3329
Feature activation+3.61e+00

Pos logit contributions
 avail+0.076
eling+0.076
wen+0.076
vation+0.075
ister+0.075
Neg logit contributions
 like-0.057
 in-0.041
 favorite-0.040
 and-0.040
 but-0.039
Token,
Token ID11
Feature activation+6.72e-01

Pos logit contributions
ths+0.139
ister+0.132
 avail+0.131
ww+0.130
www+0.130
Neg logit contributions
 like-0.043
 everywhere-0.017
 liked-0.016
 and-0.013
 around-0.013
Token the
Token ID262
Feature activation+1.17e+00

Pos logit contributions
ww+0.024
ister+0.024
www+0.024
ths+0.024
 sermon+0.023
Neg logit contributions
 like-0.009
 liked-0.005
 everywhere-0.004
 all-0.004
 both-0.004
Token m
Token ID285
Feature activation+8.78e+00

Pos logit contributions
ister+0.033
ww+0.033
irens+0.032
ths+0.032
www+0.032
Neg logit contributions
 like-0.022
 favorite-0.017
 t-0.017
 everywhere-0.016
 back-0.016
Tokenama
Token ID1689
Feature activation+4.61e+00

Pos logit contributions
 avail+0.141
ww+0.114
 storm+0.110
 site+0.106
 canyon+0.105
Neg logit contributions
 like-0.262
 favorite-0.241
 liked-0.224
 t-0.217
 and-0.217
Token woke
Token ID19092
Feature activation+3.25e+00

Pos logit contributions
 avail+0.140
ister+0.137
ww+0.136
ths+0.135
www+0.134
Neg logit contributions
 like-0.086
 liked-0.060
 everywhere-0.055
 favorite-0.053
 found-0.052
Token the
Token ID262
Feature activation-1.46e+00

Pos logit contributions
ww+0.120
ister+0.116
www+0.116
 outage+0.115
 service+0.114
Neg logit contributions
 like-0.042
 everywhere-0.024
 all-0.023
 around-0.022
 back-0.021
Token baby
Token ID5156
Feature activation+3.65e+00

Pos logit contributions
 like+0.033
 favorite+0.025
 everywhere+0.023
 t+0.023
 love+0.023
Neg logit contributions
ister-0.038
ww-0.038
ths-0.038
irens-0.037
www-0.037
Token up
Token ID510
Feature activation+2.72e-01

Pos logit contributions
ww+0.117
ister+0.114
www+0.114
 avail+0.110
irens+0.108
Neg logit contributions
 like-0.068
 everywhere-0.046
 liked-0.046
 back-0.045
 around-0.043
Token that
Token ID326
Feature activation+2.08e+00

Pos logit contributions
 avail+0.013
ww+0.013
ister+0.013
www+0.013
 sermon+0.013
Neg logit contributions
 like-0.010
 everywhere-0.007
 liked-0.007
 around-0.007
 favorite-0.006
Token day
Token ID1110
Feature activation+3.07e-01

Pos logit contributions
ister+0.045
ths+0.044
ww+0.043
www+0.042
 avail+0.042
Neg logit contributions
 like-0.056
 everywhere-0.043
 liked-0.042
 favorite-0.042
 around-0.040
Token forward
Token ID2651
Feature activation+2.79e+00

Pos logit contributions
ister+0.011
ths+0.011
ww+0.011
www+0.010
 avail+0.010
Neg logit contributions
 like-0.004
 everywhere-0.002
 liked-0.002
 favorite-0.002
 around-0.002
Token,
Token ID11
Feature activation+1.78e+00

Pos logit contributions
ister+0.105
 sermon+0.102
www+0.102
ww+0.102
ths+0.101
Neg logit contributions
 like-0.035
 liked-0.016
 everywhere-0.015
 back-0.014
 around-0.013
Token the
Token ID262
Feature activation+1.47e+00

Pos logit contributions
ister+0.056
ww+0.055
ths+0.055
www+0.054
 service+0.053
Neg logit contributions
 like-0.030
 everywhere-0.018
 favorite-0.018
 liked-0.018
 wherever-0.016
Token m
Token ID285
Feature activation+8.78e+00

Pos logit contributions
ister+0.040
ww+0.040
ths+0.040
www+0.039
 avail+0.039
Neg logit contributions
 like-0.030
 favorite-0.021
 liked-0.021
 everywhere-0.021
 back-0.019
Tokenule
Token ID2261
Feature activation+4.06e+00

Pos logit contributions
 avail+0.122
ww+0.098
 storm+0.092
www+0.088
ths+0.085
Neg logit contributions
 like-0.295
 favorite-0.253
 liked-0.249
 and-0.242
 pad-0.239
Token and
Token ID290
Feature activation+7.87e-01

Pos logit contributions
ister+0.130
ww+0.129
ths+0.128
www+0.126
 service+0.124
Neg logit contributions
 like-0.064
 everywhere-0.038
 favorite-0.037
 liked-0.037
 around-0.032
Token the
Token ID262
Feature activation+4.04e-01

Pos logit contributions
ister+0.026
ww+0.026
ths+0.025
www+0.025
 service+0.025
Neg logit contributions
 like-0.012
 everywhere-0.007
 liked-0.007
 favorite-0.007
 around-0.006
Token little
Token ID1310
Feature activation+2.04e+00

Pos logit contributions
ister+0.011
ww+0.011
ths+0.011
www+0.011
 avail+0.011
Neg logit contributions
 like-0.008
 favorite-0.006
 liked-0.006
 everywhere-0.006
 back-0.005
Token boy
Token ID2933
Feature activation+1.12e+00

Pos logit contributions
ister+0.051
ww+0.051
ths+0.050
www+0.050
 service+0.048
Neg logit contributions
 like-0.047
 liked-0.033
 favorite-0.033
 everywhere-0.033
 back-0.031
Tokenitten
Token ID2621
Feature activation+2.32e+00

Pos logit contributions
 avail+0.154
 hawk+0.137
 beast+0.136
 storm+0.135
ww+0.133
Neg logit contributions
 like-0.175
 favorite-0.137
.-0.133
 both-0.132
aces-0.132
Token.
Token ID13
Feature activation+9.28e-01

Pos logit contributions
ww+0.092
ister+0.091
ths+0.090
 avail+0.090
www+0.090
Neg logit contributions
 like-0.023
 everywhere-0.007
 liked-0.005
 favorite-0.005
 around-0.005
Token Now
Token ID2735
Feature activation+1.60e+00

Pos logit contributions
ister+0.028
ww+0.028
ths+0.027
www+0.027
 service+0.027
Neg logit contributions
 like-0.017
 liked-0.011
 favorite-0.011
 everywhere-0.011
 both-0.010
Token that
Token ID326
Feature activation+2.77e+00

Pos logit contributions
ister+0.047
ww+0.047
ths+0.046
www+0.046
 service+0.045
Neg logit contributions
 like-0.030
 everywhere-0.019
 liked-0.019
 favorite-0.019
 around-0.017
Token the
Token ID262
Feature activation+1.33e+00

Pos logit contributions
ister+0.087
ths+0.085
ww+0.085
www+0.084
irens+0.083
Neg logit contributions
 like-0.051
 liked-0.031
 everywhere-0.031
 favorite-0.031
 all-0.029
Token m
Token ID285
Feature activation+8.77e+00

Pos logit contributions
ister+0.035
ww+0.034
ths+0.034
www+0.033
 avail+0.033
Neg logit contributions
 like-0.029
 favorite-0.021
 everywhere-0.021
 liked-0.021
 dolls-0.019
Tokenitten
Token ID2621
Feature activation+4.28e+00

Pos logit contributions
 avail+0.168
ww+0.139
 storm+0.138
 service+0.133
ths+0.132
Neg logit contributions
 like-0.251
 favorite-0.212
 liked-0.202
 dolls-0.198
 pad-0.197
Token was
Token ID373
Feature activation+1.83e+00

Pos logit contributions
ister+0.127
ww+0.127
ths+0.125
 avail+0.124
www+0.123
Neg logit contributions
 like-0.080
 liked-0.053
 everywhere-0.051
 favorite-0.051
 around-0.047
Token full
Token ID1336
Feature activation-1.48e+00

Pos logit contributions
ww+0.046
ister+0.046
ths+0.046
www+0.045
 storm+0.045
Neg logit contributions
 like-0.036
 liked-0.026
 favorite-0.025
 everywhere-0.025
 pad-0.025
Token,
Token ID11
Feature activation-8.09e-01

Pos logit contributions
 like+0.033
 favorite+0.027
 liked+0.026
 everywhere+0.025
 dolls+0.025
Neg logit contributions
ister-0.032
 service-0.031
ww-0.031
 storm-0.030
www-0.030
Token Daddy
Token ID26926
Feature activation+2.09e+00

Pos logit contributions
 like+0.012
 everywhere+0.007
 favorite+0.007
 liked+0.007
 around+0.006
Neg logit contributions
ister-0.027
ww-0.027
ths-0.026
www-0.026
 service-0.026
Token's
Token ID338
Feature activation+8.45e-01

Pos logit contributions
ister+0.022
ww+0.022
ths+0.022
 avail+0.022
www+0.022
Neg logit contributions
 like-0.011
 liked-0.006
 everywhere-0.006
 favorite-0.006
 around-0.005
Token cleaning
Token ID12724
Feature activation+9.01e-01

Pos logit contributions
ww+0.020
ister+0.020
www+0.020
ths+0.020
 avail+0.019
Neg logit contributions
 like-0.021
 favorite-0.015
 everywhere-0.015
 liked-0.015
 back-0.014
Token adventure
Token ID8855
Feature activation+1.85e+00

Pos logit contributions
ww+0.029
ister+0.028
www+0.028
ths+0.028
 avail+0.027
Neg logit contributions
 like-0.015
 everywhere-0.010
 favorite-0.009
 liked-0.009
 around-0.008
Token.
Token ID13
Feature activation+1.25e+00

Pos logit contributions
ister+0.066
ww+0.066
ths+0.065
www+0.064
 avail+0.063
Neg logit contributions
 like-0.023
 everywhere-0.011
 liked-0.010
 favorite-0.010
 around-0.008
Token The
Token ID383
Feature activation+1.26e+00

Pos logit contributions
ister+0.039
ww+0.039
ths+0.039
www+0.038
 avail+0.037
Neg logit contributions
 like-0.021
 everywhere-0.013
 liked-0.013
 favorite-0.012
 wherever-0.011
Token m
Token ID285
Feature activation+8.76e+00

Pos logit contributions
ister+0.031
ww+0.031
ths+0.030
www+0.030
 avail+0.029
Neg logit contributions
 like-0.030
 everywhere-0.022
 favorite-0.022
 liked-0.022
 dolls-0.020
Tokenop
Token ID404
Feature activation+2.87e+00

Pos logit contributions
 avail+0.220
ww+0.186
 storm+0.176
 hawk+0.171
www+0.171
Neg logit contributions
 like-0.204
 favorite-0.165
 liked-0.161
 dolls-0.154
 and-0.152
Token had
Token ID550
Feature activation-1.46e+00

Pos logit contributions
 avail+0.087
ww+0.087
ister+0.085
www+0.084
ths+0.084
Neg logit contributions
 like-0.053
 liked-0.037
 everywhere-0.034
 around-0.033
 favorite-0.032
Token been
Token ID587
Feature activation+1.41e+00

Pos logit contributions
 like+0.030
 everywhere+0.021
 favorite+0.021
 liked+0.020
 around+0.019
Neg logit contributions
ister-0.039
ww-0.039
ths-0.038
www-0.038
 service-0.037
Token released
Token ID2716
Feature activation-8.44e-01

Pos logit contributions
ister+0.039
ww+0.039
ths+0.038
www+0.038
 service+0.037
Neg logit contributions
 like-0.029
 everywhere-0.020
 favorite-0.019
 liked-0.019
 around-0.018
Token and
Token ID290
Feature activation+2.52e+00

Pos logit contributions
 like+0.012
 favorite+0.008
 liked+0.008
 dolls+0.008
 everywhere+0.007
Neg logit contributions
ister-0.026
ww-0.025
ths-0.025
 service-0.025
www-0.024
Token cream
Token ID8566
Feature activation-2.11e+00

Pos logit contributions
ister+0.005
ww+0.005
ths+0.005
www+0.005
 service+0.005
Neg logit contributions
 like-0.007
 everywhere-0.006
 liked-0.006
 favorite-0.006
 wherever-0.005
Token!
Token ID0
Feature activation+7.65e-01

Pos logit contributions
 like+0.038
 everywhere+0.025
 favorite+0.024
 liked+0.024
 dolls+0.022
Neg logit contributions
ister-0.063
ww-0.062
ths-0.061
www-0.061
 service-0.059
Tokenâ
Token ID22940
Feature activation+3.65e+00

Pos logit contributions
ister+0.020
ww+0.020
www+0.020
ths+0.020
 service+0.019
Neg logit contributions
 like-0.016
 everywhere-0.011
 favorite-0.011
 liked-0.011
 around-0.010
Token
Token ID26391
Feature activation+1.18e+00

Pos logit contributions
ths+0.068
 avail+0.066
ww+0.065
ister+0.065
 storm+0.064
Neg logit contributions
 like-0.108
 liked-0.084
 around-0.080
 back-0.076
 everywhere-0.076
Token The
Token ID383
Feature activation+1.42e+00

Pos logit contributions
ister+0.028
ww+0.027
www+0.026
ths+0.025
 service+0.025
Neg logit contributions
 like-0.029
 everywhere-0.023
 favorite-0.022
 liked-0.021
 back-0.021
Token m
Token ID285
Feature activation+8.75e+00

Pos logit contributions
ister+0.036
ths+0.036
ww+0.036
 avail+0.035
www+0.035
Neg logit contributions
 like-0.033
 favorite-0.024
 liked-0.023
 everywhere-0.023
 back-0.021
Tokenama
Token ID1689
Feature activation+4.04e+00

Pos logit contributions
 avail+0.126
 storm+0.106
ths+0.102
ww+0.101
izing+0.096
Neg logit contributions
 like-0.289
 favorite-0.252
 liked-0.245
aces-0.240
 pad-0.233
Token said
Token ID531
Feature activation-4.38e-01

Pos logit contributions
ister+0.104
ww+0.103
ths+0.103
 avail+0.101
www+0.100
Neg logit contributions
 like-0.089
 liked-0.063
 everywhere-0.062
 favorite-0.062
 around-0.057
Token,
Token ID11
Feature activation-3.87e+00

Pos logit contributions
 like+0.007
 everywhere+0.005
 favorite+0.005
 liked+0.005
 pad+0.004
Neg logit contributions
ister-0.013
ww-0.013
www-0.013
ths-0.013
 service-0.013
Token
Token ID6184
Feature activation+2.25e+00

Pos logit contributions
 like+0.047
 everywhere+0.022
 favorite+0.021
 liked+0.021
 around+0.016
Neg logit contributions
ister-0.137
ww-0.137
ths-0.135
www-0.134
 service-0.132
Token
Token ID95
Feature activation+2.74e+00

Pos logit contributions
ths+0.042
ww+0.041
ister+0.041
 avail+0.041
 storm+0.040
Neg logit contributions
 like-0.065
 liked-0.049
 favorite-0.048
 everywhere-0.047
 around-0.046
Token.
Token ID13
Feature activation+1.46e-01

Pos logit contributions
ister+0.047
ww+0.047
ths+0.046
www+0.046
 service+0.045
Neg logit contributions
 like-0.016
 favorite-0.008
 everywhere-0.008
 liked-0.008
 dolls-0.006
Token
Token ID220
Feature activation-4.90e-02

Pos logit contributions
ww+0.004
www+0.004
 sermon+0.004
ister+0.004
 service+0.004
Neg logit contributions
 like-0.003
 everywhere-0.002
 favorite-0.002
aces-0.002
 pad-0.002
Token
Token ID198
Feature activation-3.13e-01

Pos logit contributions
 like+0.001
 favorite+0.001
 liked+0.001
 pad+0.001
 ti+0.001
Neg logit contributions
ister-0.001
ama-0.001
oo-0.001
ww-0.001
iren-0.001
Token
Token ID198
Feature activation+5.66e-01

Pos logit contributions
 like+0.006
 favorite+0.004
 everywhere+0.004
aces+0.004
 pad+0.004
Neg logit contributions
ister-0.009
ww-0.009
www-0.009
 sermon-0.008
 service-0.008
TokenThe
Token ID464
Feature activation+1.45e+00

Pos logit contributions
ww+0.013
ister+0.013
www+0.013
 sermon+0.012
ths+0.012
Neg logit contributions
 like-0.014
 everywhere-0.011
 favorite-0.010
 liked-0.010
 wherever-0.010
Token m
Token ID285
Feature activation+8.73e+00

Pos logit contributions
ister+0.036
ww+0.035
ths+0.035
www+0.034
 avail+0.034
Neg logit contributions
 like-0.034
 favorite-0.025
 liked-0.025
 everywhere-0.025
 dolls-0.023
Tokenittens
Token ID34978
Feature activation+3.62e+00

Pos logit contributions
 avail+0.161
ww+0.141
 storm+0.133
ths+0.133
www+0.131
Neg logit contributions
 like-0.257
 favorite-0.216
 liked-0.209
 dolls-0.206
 pad-0.198
Token were
Token ID547
Feature activation+2.25e-01

Pos logit contributions
ww+0.109
ister+0.107
 avail+0.106
ths+0.105
www+0.105
Neg logit contributions
 like-0.066
 liked-0.045
 everywhere-0.043
 favorite-0.042
 both-0.039
Token so
Token ID523
Feature activation+3.08e+00

Pos logit contributions
ww+0.007
ister+0.007
ths+0.006
www+0.006
 service+0.006
Neg logit contributions
 like-0.004
 liked-0.003
 favorite-0.003
 everywhere-0.003
 pad-0.002
Token thin
Token ID7888
Feature activation-4.16e-01

Pos logit contributions
ths+0.093
ister+0.093
 avail+0.092
ww+0.092
 beast+0.091
Neg logit contributions
 like-0.056
 liked-0.037
 favorite-0.037
 everywhere-0.035
 both-0.032
Token;
Token ID26
Feature activation+1.97e+00

Pos logit contributions
 like+0.008
 favorite+0.006
 liked+0.005
 everywhere+0.005
 dolls+0.005
Neg logit contributions
ww-0.011
ister-0.011
ths-0.011
www-0.011
 service-0.011
Token he
Token ID339
Feature activation+1.02e+00

Pos logit contributions
ister+0.065
ww+0.065
www+0.063
ths+0.063
 sermon+0.061
Neg logit contributions
 like-0.051
 everywhere-0.035
 liked-0.034
 favorite-0.034
 wherever-0.032
Token never
Token ID1239
Feature activation+1.72e+00

Pos logit contributions
ister+0.029
ww+0.029
ths+0.028
www+0.028
 avail+0.027
Neg logit contributions
 like-0.021
 liked-0.014
 everywhere-0.014
 favorite-0.014
 around-0.013
Token forgot
Token ID16453
Feature activation-2.24e+00

Pos logit contributions
ister+0.048
ww+0.048
ths+0.047
www+0.047
 service+0.046
Neg logit contributions
 like-0.035
 liked-0.023
 everywhere-0.023
 favorite-0.023
 around-0.021
Token what
Token ID644
Feature activation-9.47e-01

Pos logit contributions
 like+0.038
 favorite+0.025
 liked+0.025
 pad+0.024
 everywhere+0.023
Neg logit contributions
ister-0.063
ths-0.063
 service-0.062
 storm-0.062
ww-0.061
Token the
Token ID262
Feature activation+6.12e-01

Pos logit contributions
 like+0.021
 everywhere+0.015
 favorite+0.015
 liked+0.015
 pad+0.014
Neg logit contributions
ister-0.024
ww-0.024
ths-0.023
www-0.023
 service-0.023
Token m
Token ID285
Feature activation+8.71e+00

Pos logit contributions
ister+0.014
ww+0.014
ths+0.013
www+0.013
 avail+0.013
Neg logit contributions
 like-0.016
 favorite-0.012
 everywhere-0.012
 liked-0.012
 around-0.011
Tokenop
Token ID404
Feature activation+3.11e+00

Pos logit contributions
 avail+0.140
ww+0.119
ation+0.108
ths+0.105
www+0.105
Neg logit contributions
 like-0.280
 favorite-0.237
 pad-0.226
 liked-0.225
.-0.220
Token taught
Token ID7817
Feature activation+1.36e+00

Pos logit contributions
ww+0.094
 avail+0.093
ths+0.092
ister+0.091
www+0.090
Neg logit contributions
 like-0.059
 liked-0.040
 everywhere-0.038
 around-0.036
 favorite-0.035
Token him
Token ID683
Feature activation-2.73e-01

Pos logit contributions
ister+0.041
ww+0.041
ths+0.040
www+0.040
 service+0.039
Neg logit contributions
 like-0.024
 favorite-0.015
 everywhere-0.015
 liked-0.015
 pad-0.013
Token
Token ID6184
Feature activation+2.94e+00

Pos logit contributions
 like+0.003
 everywhere+0.002
 favorite+0.002
 liked+0.002
 dolls+0.001
Neg logit contributions
ister-0.009
ths-0.009
ww-0.009
 service-0.009
www-0.009
Token
Token ID95
Feature activation+2.52e+00

Pos logit contributions
ww+0.060
 avail+0.056
ths+0.056
ister+0.056
www+0.055
Neg logit contributions
 like-0.083
 liked-0.060
 favorite-0.060
aces-0.056
 love-0.056
Token sharing
Token ID7373
Feature activation-1.49e+00

Pos logit contributions
ister+0.057
ww+0.057
www+0.056
ths+0.055
 avail+0.053
Neg logit contributions
 like-0.037
 everywhere-0.025
 favorite-0.024
 liked-0.024
 all-0.022
Token is
Token ID318
Feature activation+8.42e-01

Pos logit contributions
 like+0.032
 favorite+0.024
 pad+0.024
aces+0.022
 everywhere+0.022
Neg logit contributions
ww-0.035
ister-0.035
ths-0.034
 service-0.034
www-0.033
Token good
Token ID922
Feature activation-1.54e+00

Pos logit contributions
ths+0.019
ww+0.019
ister+0.018
www+0.018
 service+0.018
Neg logit contributions
 like-0.019
 pad-0.016
aces-0.015
 everywhere-0.015
 liked-0.015
Token.
Token ID13
Feature activation+2.97e-01

Pos logit contributions
 like+0.022
 favorite+0.012
 everywhere+0.012
 liked+0.012
 pad+0.012
Neg logit contributions
ww-0.050
ths-0.049
ister-0.049
www-0.048
 service-0.048
Token The
Token ID383
Feature activation+9.27e-01

Pos logit contributions
ww+0.009
ister+0.009
ths+0.009
www+0.009
 service+0.009
Neg logit contributions
 like-0.005
 everywhere-0.003
 favorite-0.003
 liked-0.003
 dolls-0.003
Token m
Token ID285
Feature activation+8.66e+00

Pos logit contributions
ister+0.022
ww+0.022
ths+0.022
www+0.021
 avail+0.021
Neg logit contributions
 like-0.022
 everywhere-0.016
 favorite-0.016
 liked-0.016
 dolls-0.015
Tokenop
Token ID404
Feature activation+2.57e+00

Pos logit contributions
 avail+0.256
ww+0.215
 storm+0.212
www+0.203
 service+0.202
Neg logit contributions
 like-0.163
 dolls-0.126
 favorite-0.125
 liked-0.122
 both-0.120
Token was
Token ID373
Feature activation+2.10e+00

Pos logit contributions
ww+0.082
ister+0.081
ths+0.080
 avail+0.080
www+0.080
Neg logit contributions
 like-0.043
 liked-0.027
 everywhere-0.025
 favorite-0.025
 around-0.023
Token not
Token ID407
Feature activation-6.40e-02

Pos logit contributions
ths+0.051
ister+0.051
ww+0.050
 service+0.050
 storm+0.050
Neg logit contributions
 like-0.041
 pad-0.035
aces-0.033
 everywhere-0.031
 liked-0.030
Token selfish
Token ID20363
Feature activation-2.34e+00

Pos logit contributions
 like+0.001
 everywhere+0.001
 favorite+0.001
 liked+0.001
 pad+0.001
Neg logit contributions
ww-0.002
ister-0.002
ths-0.002
www-0.002
 service-0.002
Token anymore
Token ID7471
Feature activation-1.00e+00

Pos logit contributions
 like+0.034
 ti+0.027
 favorite+0.026
 liked+0.025
 pad+0.025
Neg logit contributions
ister-0.063
ww-0.062
ths-0.062
 service-0.062
 storm-0.062
Token were
Token ID547
Feature activation+1.10e+00

Pos logit contributions
ister+0.024
ww+0.024
ths+0.024
www+0.024
 service+0.023
Neg logit contributions
 like-0.018
 everywhere-0.012
 favorite-0.012
 liked-0.012
 pad-0.011
Token having
Token ID1719
Feature activation-1.58e+00

Pos logit contributions
ister+0.031
ww+0.031
ths+0.031
www+0.030
 service+0.030
Neg logit contributions
 like-0.021
 everywhere-0.014
 favorite-0.014
 liked-0.014
 wherever-0.012
Token trouble
Token ID5876
Feature activation+7.44e-01

Pos logit contributions
 like+0.027
 everywhere+0.017
 liked+0.017
 favorite+0.017
 pad+0.015
Neg logit contributions
ww-0.047
ister-0.047
ths-0.046
www-0.046
 service-0.045
Token.
Token ID13
Feature activation+1.48e-01

Pos logit contributions
ister+0.024
 service+0.023
ths+0.023
ww+0.023
www+0.022
Neg logit contributions
 like-0.010
 favorite-0.007
 liked-0.006
 everywhere-0.006
 pad-0.006
Token The
Token ID383
Feature activation+8.53e-01

Pos logit contributions
ister+0.004
ww+0.004
ths+0.004
www+0.004
 service+0.004
Neg logit contributions
 like-0.003
 everywhere-0.002
 favorite-0.002
 liked-0.002
 wherever-0.002
Token m
Token ID285
Feature activation+8.65e+00

Pos logit contributions
ww+0.022
ister+0.022
ths+0.022
 avail+0.021
www+0.021
Neg logit contributions
 like-0.019
 favorite-0.014
 liked-0.014
 everywhere-0.013
 back-0.013
Tokenule
Token ID2261
Feature activation+4.19e+00

Pos logit contributions
 avail+0.135
ww+0.111
 storm+0.107
www+0.094
ths+0.092
Neg logit contributions
 like-0.275
 liked-0.233
 favorite-0.232
 and-0.224
 pad-0.220
Token saw
Token ID2497
Feature activation-5.38e-01

Pos logit contributions
 avail+0.130
ww+0.128
ister+0.128
 service+0.124
www+0.124
Neg logit contributions
 like-0.076
 liked-0.055
 everywhere-0.049
 favorite-0.047
 around-0.046
Token this
Token ID428
Feature activation+1.23e-01

Pos logit contributions
 like+0.009
 everywhere+0.005
 favorite+0.005
 liked+0.005
 around+0.004
Neg logit contributions
ister-0.017
ww-0.017
ths-0.017
www-0.017
 service-0.016
Token and
Token ID290
Feature activation+8.04e-01

Pos logit contributions
ister+0.005
ww+0.005
ths+0.005
www+0.004
 avail+0.004
Neg logit contributions
 like-0.001
 everywhere-0.001
 favorite-0.001
 liked-0.001
 around-0.000
Token wanted
Token ID2227
Feature activation+1.38e+00

Pos logit contributions
ister+0.024
ww+0.023
 avail+0.023
 service+0.023
 sermon+0.023
Neg logit contributions
 like-0.016
 liked-0.011
 everywhere-0.010
 favorite-0.010
 all-0.010
Token were
Token ID547
Feature activation+1.10e+00

Pos logit contributions
ister+0.024
ww+0.024
ths+0.024
www+0.024
 service+0.023
Neg logit contributions
 like-0.018
 everywhere-0.012
 favorite-0.012
 liked-0.012
 pad-0.011
Token having
Token ID1719
Feature activation-1.58e+00

Pos logit contributions
ister+0.031
ww+0.031
ths+0.031
www+0.030
 service+0.030
Neg logit contributions
 like-0.021
 everywhere-0.014
 favorite-0.014
 liked-0.014
 wherever-0.012
Token trouble
Token ID5876
Feature activation+7.44e-01

Pos logit contributions
 like+0.027
 everywhere+0.017
 liked+0.017
 favorite+0.017
 pad+0.015
Neg logit contributions
ww-0.047
ister-0.047
ths-0.046
www-0.046
 service-0.045
Token.
Token ID13
Feature activation+1.48e-01

Pos logit contributions
ister+0.024
 service+0.023
ths+0.023
ww+0.023
www+0.022
Neg logit contributions
 like-0.010
 favorite-0.007
 liked-0.006
 everywhere-0.006
 pad-0.006
Token The
Token ID383
Feature activation+8.53e-01

Pos logit contributions
ister+0.004
ww+0.004
ths+0.004
www+0.004
 service+0.004
Neg logit contributions
 like-0.003
 everywhere-0.002
 favorite-0.002
 liked-0.002
 wherever-0.002
Token m
Token ID285
Feature activation+8.65e+00

Pos logit contributions
ww+0.022
ister+0.022
ths+0.022
 avail+0.021
www+0.021
Neg logit contributions
 like-0.019
 favorite-0.014
 liked-0.014
 everywhere-0.013
 back-0.013
Tokenule
Token ID2261
Feature activation+4.19e+00

Pos logit contributions
 avail+0.135
ww+0.111
 storm+0.107
www+0.094
ths+0.092
Neg logit contributions
 like-0.275
 liked-0.233
 favorite-0.232
 and-0.224
 pad-0.220
Token saw
Token ID2497
Feature activation-5.38e-01

Pos logit contributions
 avail+0.130
ww+0.128
ister+0.128
 service+0.124
www+0.124
Neg logit contributions
 like-0.076
 liked-0.055
 everywhere-0.049
 favorite-0.047
 around-0.046
Token this
Token ID428
Feature activation+1.23e-01

Pos logit contributions
 like+0.009
 everywhere+0.005
 favorite+0.005
 liked+0.005
 around+0.004
Neg logit contributions
ister-0.017
ww-0.017
ths-0.017
www-0.017
 service-0.016
Token and
Token ID290
Feature activation+8.04e-01

Pos logit contributions
ister+0.005
ww+0.005
ths+0.005
www+0.004
 avail+0.004
Neg logit contributions
 like-0.001
 everywhere-0.001
 favorite-0.001
 liked-0.001
 around-0.000
Token wanted
Token ID2227
Feature activation+1.38e+00

Pos logit contributions
ister+0.024
ww+0.023
 avail+0.023
 service+0.023
 sermon+0.023
Neg logit contributions
 like-0.016
 liked-0.011
 everywhere-0.010
 favorite-0.010
 all-0.010
Token purple
Token ID14032
Feature activation+3.18e-01

Pos logit contributions
 like+0.013
 everywhere+0.009
 favorite+0.009
 liked+0.009
 dolls+0.008
Neg logit contributions
ister-0.017
ww-0.017
ths-0.017
www-0.016
 service-0.016
Token m
Token ID285
Feature activation+7.01e+00

Pos logit contributions
ister+0.007
ww+0.007
ths+0.007
www+0.007
 service+0.007
Neg logit contributions
 like-0.008
 everywhere-0.006
 favorite-0.006
 liked-0.006
 dolls-0.005
Tokenop
Token ID404
Feature activation+1.38e+00

Pos logit contributions
 avail+0.102
ths+0.093
ation+0.084
 situations+0.084
 storm+0.083
Neg logit contributions
 like-0.234
 favorite-0.196
 liked-0.187
 everywhere-0.187
.-0.181
Token.
Token ID13
Feature activation+9.08e-02

Pos logit contributions
ister+0.048
ww+0.048
ths+0.047
www+0.047
 service+0.046
Neg logit contributions
 like-0.017
 everywhere-0.008
 favorite-0.008
 liked-0.008
 wherever-0.006
Token The
Token ID383
Feature activation+5.91e-01

Pos logit contributions
ww+0.003
ister+0.003
www+0.003
ths+0.002
 service+0.002
Neg logit contributions
 like-0.002
 everywhere-0.001
 favorite-0.001
 liked-0.001
 back-0.001
Token m
Token ID285
Feature activation+8.65e+00

Pos logit contributions
ister+0.014
ww+0.013
ths+0.013
www+0.013
 service+0.013
Neg logit contributions
 like-0.015
 everywhere-0.011
 favorite-0.011
 liked-0.011
 around-0.010
Tokenop
Token ID404
Feature activation+2.99e+00

Pos logit contributions
 avail+0.217
 storm+0.191
ww+0.189
 service+0.187
ths+0.183
Neg logit contributions
 like-0.204
 favorite-0.161
 liked-0.156
 dolls-0.153
 everywhere-0.149
Token lived
Token ID5615
Feature activation-9.74e-01

Pos logit contributions
ister+0.088
ww+0.088
ths+0.087
www+0.086
 service+0.084
Neg logit contributions
 like-0.054
 everywhere-0.035
 favorite-0.034
 liked-0.034
 around-0.030
Token in
Token ID287
Feature activation-1.47e+00

Pos logit contributions
 like+0.015
 pad+0.011
 pads+0.010
 names+0.010
aces+0.010
Neg logit contributions
ister-0.027
oo-0.027
ths-0.026
ww-0.026
asis-0.025
Token a
Token ID257
Feature activation-1.05e+00

Pos logit contributions
 like+0.015
 everywhere+0.005
 favorite+0.004
 all+0.004
 liked+0.004
Neg logit contributions
ister-0.059
ths-0.058
www-0.056
ww-0.056
 sermon-0.055
Token big
Token ID1263
Feature activation+1.23e-01

Pos logit contributions
 like+0.017
 everywhere+0.012
 liked+0.012
 favorite+0.012
 back+0.011
Neg logit contributions
ths-0.033
ister-0.033
 avail-0.033
 service-0.033
vation-0.033
Token fun
Token ID1257
Feature activation-1.53e+00

Pos logit contributions
ww+0.006
ths+0.006
 storm+0.006
www+0.006
ister+0.006
Neg logit contributions
 like-0.005
 liked-0.004
 favorite-0.004
 pad-0.004
 everywhere-0.003
Token!"
Token ID2474
Feature activation+8.49e-02

Pos logit contributions
 like+0.027
 liked+0.022
aces+0.022
 pad+0.021
 everywhere+0.021
Neg logit contributions
ww-0.037
 service-0.036
www-0.036
ister-0.036
ths-0.036
Token
Token ID198
Feature activation-7.21e-01

Pos logit contributions
ister+0.003
ww+0.003
ths+0.003
www+0.003
 service+0.003
Neg logit contributions
 like-0.001
 everywhere-0.001
 favorite-0.001
 liked-0.001
 dolls-0.001
Token
Token ID198
Feature activation+5.29e-01

Pos logit contributions
 like+0.010
 everywhere+0.005
 favorite+0.005
 liked+0.005
 dolls+0.004
Neg logit contributions
ister-0.024
ww-0.024
ths-0.024
www-0.024
 service-0.023
TokenThe
Token ID464
Feature activation+1.30e+00

Pos logit contributions
ister+0.014
ww+0.013
 avail+0.013
 service+0.013
ths+0.013
Neg logit contributions
 like-0.012
 liked-0.009
 both-0.008
 everywhere-0.008
 favorite-0.008
Token m
Token ID285
Feature activation+8.64e+00

Pos logit contributions
ww+0.035
 avail+0.035
ister+0.035
www+0.034
vation+0.034
Neg logit contributions
 like-0.027
 favorite-0.020
 liked-0.020
 dolls-0.020
 everywhere-0.019
Tokenop
Token ID404
Feature activation+2.96e+00

Pos logit contributions
 avail+0.263
 storm+0.225
ww+0.224
 service+0.211
www+0.210
Neg logit contributions
 like-0.153
 dolls-0.120
 favorite-0.117
 liked-0.114
 both-0.110
Token thought
Token ID1807
Feature activation+7.07e-01

Pos logit contributions
ister+0.094
ww+0.094
ths+0.092
www+0.091
 avail+0.091
Neg logit contributions
 like-0.049
 liked-0.030
 everywhere-0.029
 favorite-0.029
 around-0.025
Token about
Token ID546
Feature activation+3.52e-01

Pos logit contributions
ister+0.022
ww+0.022
ths+0.022
www+0.022
 service+0.021
Neg logit contributions
 like-0.011
 everywhere-0.007
 favorite-0.007
 liked-0.007
 around-0.006
Token it
Token ID340
Feature activation+3.76e-01

Pos logit contributions
ww+0.011
ister+0.011
www+0.011
ths+0.011
 avail+0.011
Neg logit contributions
 like-0.006
 everywhere-0.004
 liked-0.004
 favorite-0.004
 all-0.003
Token.
Token ID13
Feature activation+6.95e-01

Pos logit contributions
ister+0.012
ths+0.012
ww+0.012
 service+0.012
www+0.012
Neg logit contributions
 like-0.005
 everywhere-0.002
 favorite-0.002
 liked-0.002
 dolls-0.002
Token
Token ID198
Feature activation-2.91e-01

Pos logit contributions
 like+0.013
 everywhere+0.008
 favorite+0.008
 liked+0.008
 pad+0.007
Neg logit contributions
ister-0.026
ww-0.026
ths-0.025
www-0.025
 service-0.025
Token
Token ID198
Feature activation+1.01e+00

Pos logit contributions
 like+0.003
 everywhere+0.002
 liked+0.002
 favorite+0.001
 around+0.001
Neg logit contributions
ths-0.010
ww-0.010
ister-0.010
www-0.010
 avail-0.010
Token"
Token ID1
Feature activation-4.08e+00

Pos logit contributions
ister+0.029
ths+0.029
ww+0.028
 avail+0.028
 service+0.028
Neg logit contributions
 like-0.020
 liked-0.013
 everywhere-0.013
 favorite-0.012
 back-0.012
TokenW
Token ID54
Feature activation+2.94e+00

Pos logit contributions
 like+0.078
 liked+0.055
 favorite+0.051
 everywhere+0.047
 dolls+0.045
Neg logit contributions
ths-0.119
 avail-0.119
ister-0.117
ww-0.116
www-0.116
Tokenoof
Token ID37711
Feature activation-6.94e-01

Pos logit contributions
ths+0.060
ister+0.060
ww+0.058
www+0.057
 service+0.056
Neg logit contributions
 like-0.082
 favorite-0.062
 liked-0.061
 everywhere-0.061
 pad-0.058
Token,
Token ID11
Feature activation-9.03e+00

Pos logit contributions
 like+0.013
 favorite+0.009
 liked+0.008
 everywhere+0.008
 wherever+0.007
Neg logit contributions
ww-0.021
ister-0.020
www-0.020
 service-0.019
ths-0.019
Token wo
Token ID24486
Feature activation+6.89e-01

Pos logit contributions
 like+0.174
 everywhere+0.114
 liked+0.111
 favorite+0.110
 dolls+0.099
Neg logit contributions
ths-0.262
ister-0.261
ww-0.260
www-0.255
 avail-0.253
Tokenof
Token ID1659
Feature activation-2.89e+00

Pos logit contributions
ister+0.016
ww+0.015
www+0.015
ths+0.015
 service+0.014
Neg logit contributions
 like-0.018
 favorite-0.013
 liked-0.013
 everywhere-0.013
 dolls-0.012
Token!"
Token ID2474
Feature activation-5.97e-01

Pos logit contributions
 like+0.050
 everywhere+0.031
 favorite+0.031
 liked+0.031
 around+0.027
Neg logit contributions
ister-0.088
ww-0.088
ths-0.086
www-0.086
 service-0.084
Token Spot
Token ID15899
Feature activation+1.29e+00

Pos logit contributions
 like+0.009
 liked+0.005
 favorite+0.005
 everywhere+0.004
 Lily+0.004
Neg logit contributions
ths-0.021
ister-0.021
ww-0.020
www-0.020
asis-0.020
Token bark
Token ID21405
Feature activation-1.37e-01

Pos logit contributions
ister+0.028
ww+0.028
ths+0.027
www+0.027
 service+0.026
Neg logit contributions
 like-0.033
 everywhere-0.025
 favorite-0.025
 liked-0.024
 wherever-0.023
Token
Token ID198
Feature activation-2.32e-02

Pos logit contributions
ister+0.002
ww+0.002
ths+0.002
www+0.002
iren+0.002
Neg logit contributions
 like-0.001
 favorite-0.001
 everywhere-0.001
 liked-0.001
 pad-0.001
Token
Token ID198
Feature activation+1.15e+00

Pos logit contributions
 like+0.000
 favorite+0.000
 everywhere+0.000
 liked+0.000
 pad+0.000
Neg logit contributions
ister-0.001
ww-0.001
ths-0.001
www-0.001
 service-0.001
Token"
Token ID1
Feature activation-3.76e+00

Pos logit contributions
ister+0.032
ww+0.032
ths+0.032
www+0.031
 service+0.031
Neg logit contributions
 like-0.023
 everywhere-0.015
 liked-0.015
 favorite-0.015
 both-0.014
TokenW
Token ID54
Feature activation+3.02e+00

Pos logit contributions
 like+0.092
 everywhere+0.071
 favorite+0.069
 liked+0.065
 back+0.064
Neg logit contributions
ww-0.089
www-0.086
ister-0.086
ths-0.085
 service-0.080
Tokenoof
Token ID37711
Feature activation-7.24e-01

Pos logit contributions
ths+0.062
ister+0.062
ww+0.061
 service+0.060
 avail+0.060
Neg logit contributions
 like-0.082
 favorite-0.061
 liked-0.061
 everywhere-0.060
 pad-0.058
Token,
Token ID11
Feature activation-8.89e+00

Pos logit contributions
 like+0.014
 favorite+0.010
 liked+0.010
 everywhere+0.010
 wherever+0.009
Neg logit contributions
ww-0.021
ister-0.020
www-0.020
 service-0.018
 storm-0.018
Token wo
Token ID24486
Feature activation+6.97e-01

Pos logit contributions
 like+0.165
 everywhere+0.106
 favorite+0.105
 liked+0.104
 dolls+0.092
Neg logit contributions
ister-0.261
ths-0.260
ww-0.260
www-0.254
 avail-0.252
Tokenof
Token ID1659
Feature activation-2.72e+00

Pos logit contributions
ister+0.015
ww+0.015
www+0.015
ths+0.014
 service+0.014
Neg logit contributions
 like-0.018
 favorite-0.014
 liked-0.014
 everywhere-0.014
 dolls-0.013
Token,
Token ID11
Feature activation-6.95e+00

Pos logit contributions
 like+0.056
 favorite+0.039
 everywhere+0.039
 liked+0.038
 dolls+0.035
Neg logit contributions
ww-0.074
ister-0.074
www-0.072
ths-0.070
 service-0.069
Token I
Token ID314
Feature activation-3.63e+00

Pos logit contributions
 like+0.137
 everywhere+0.090
 liked+0.088
 favorite+0.087
 back+0.078
Neg logit contributions
ths-0.200
ister-0.198
ww-0.197
 avail-0.195
www-0.194
Token'm
Token ID1101
Feature activation+1.69e+00

Pos logit contributions
 like+0.073
 everywhere+0.061
 favorite+0.057
 dolls+0.056
aces+0.055
Neg logit contributions
ister-0.088
ww-0.088
www-0.085
 service-0.082
 storm-0.081
Tokened
Token ID276
Feature activation-1.60e+00

Pos logit contributions
 like+0.018
 dolls+0.015
 wherever+0.014
 favorite+0.014
 upon+0.014
Neg logit contributions
ww-0.028
ths-0.028
 intense-0.028
 beast-0.027
 storm-0.027
Token,
Token ID11
Feature activation-4.11e+00

Pos logit contributions
 like+0.023
aces+0.021
 pad+0.020
 dolls+0.019
 skirts+0.018
Neg logit contributions
ths-0.040
 storm-0.040
 beast-0.039
ister-0.039
 intense-0.038
Token "
Token ID366
Feature activation-4.64e+00

Pos logit contributions
 like+0.065
 favorite+0.051
aces+0.047
 everywhere+0.045
 favorites+0.045
Neg logit contributions
ths-0.112
ister-0.108
ller-0.107
ww-0.106
 storm-0.106
TokenW
Token ID54
Feature activation+2.12e+00

Pos logit contributions
 like+0.118
 everywhere+0.089
 favorite+0.089
 liked+0.086
 dolls+0.082
Neg logit contributions
ww-0.105
ister-0.103
ths-0.101
www-0.101
 service-0.096
Tokenoof
Token ID37711
Feature activation-9.50e-01

Pos logit contributions
ister+0.040
ww+0.039
www+0.038
ths+0.038
 service+0.036
Neg logit contributions
 like-0.062
 everywhere-0.049
 favorite-0.048
 liked-0.048
 dolls-0.046
Token,
Token ID11
Feature activation-8.64e+00

Pos logit contributions
 like+0.019
 favorite+0.014
 liked+0.013
 everywhere+0.013
 around+0.012
Neg logit contributions
ww-0.027
ister-0.027
www-0.026
 service-0.024
 storm-0.024
Token wo
Token ID24486
Feature activation+8.42e-01

Pos logit contributions
 like+0.172
 everywhere+0.118
 favorite+0.118
 liked+0.116
 around+0.107
Neg logit contributions
ister-0.240
ww-0.238
www-0.231
ths-0.230
 service-0.225
Tokenof
Token ID1659
Feature activation-2.87e+00

Pos logit contributions
ister+0.019
ww+0.018
www+0.018
ths+0.017
 service+0.017
Neg logit contributions
 like-0.022
 favorite-0.017
 liked-0.017
 everywhere-0.016
 dolls-0.016
Token!"
Token ID2474
Feature activation-5.63e-01

Pos logit contributions
 like+0.059
 favorite+0.041
 everywhere+0.040
 liked+0.040
 dolls+0.037
Neg logit contributions
ww-0.079
ister-0.079
www-0.076
ths-0.074
 service-0.073
Token Tom
Token ID4186
Feature activation+5.05e-01

Pos logit contributions
 like+0.008
 liked+0.005
 everywhere+0.005
 favorite+0.004
 both+0.004
Neg logit contributions
ww-0.019
ister-0.019
ths-0.019
www-0.019
 avail-0.019
Token knew
Token ID2993
Feature activation+2.12e+00

Pos logit contributions
ister+0.013
ths+0.012
ww+0.012
www+0.012
 service+0.012
Neg logit contributions
 like-0.011
 everywhere-0.008
 favorite-0.008
 wherever-0.007
 dolls-0.007
Tokened
Token ID276
Feature activation-1.71e+00

Pos logit contributions
 like+0.016
 dolls+0.014
 upon+0.013
 everywhere+0.013
 wherever+0.013
Neg logit contributions
ww-0.018
www-0.017
oo-0.016
 intense-0.016
ths-0.016
Token,
Token ID11
Feature activation-3.37e+00

Pos logit contributions
 pad+0.027
 like+0.027
aces+0.027
 pads+0.024
 skirts+0.023
Neg logit contributions
ths-0.038
 avail-0.037
 storm-0.037
 service-0.037
 beast-0.037
Token "
Token ID366
Feature activation-4.85e+00

Pos logit contributions
 like+0.053
aces+0.041
 favorite+0.038
 everywhere+0.037
 favorites+0.037
Neg logit contributions
ths-0.091
ller-0.088
www-0.087
ister-0.087
ww-0.087
TokenW
Token ID54
Feature activation+2.29e+00

Pos logit contributions
 like+0.115
 everywhere+0.097
 favorite+0.089
 dolls+0.088
 back+0.087
Neg logit contributions
ww-0.115
www-0.110
ths-0.101
ister-0.101
body-0.100
Tokenoof
Token ID37711
Feature activation-1.36e+00

Pos logit contributions
ister+0.044
ww+0.043
ths+0.042
www+0.042
 service+0.040
Neg logit contributions
 like-0.066
 everywhere-0.051
 favorite-0.051
 liked-0.051
 around-0.048
Token,
Token ID11
Feature activation-8.55e+00

Pos logit contributions
 like+0.028
 favorite+0.020
 liked+0.020
 everywhere+0.019
 dolls+0.018
Neg logit contributions
ww-0.038
ister-0.037
www-0.036
 service-0.033
ths-0.033
Token wo
Token ID24486
Feature activation+5.83e-01

Pos logit contributions
 like+0.177
 everywhere+0.129
 favorite+0.127
 liked+0.123
 around+0.121
Neg logit contributions
ww-0.227
ister-0.224
www-0.218
 service-0.209
 storm-0.209
Tokenof
Token ID1659
Feature activation-2.73e+00

Pos logit contributions
ister+0.013
ww+0.013
www+0.012
ths+0.011
 service+0.011
Neg logit contributions
 like-0.016
 favorite-0.012
 liked-0.012
 everywhere-0.012
 dolls-0.011
Token!"
Token ID2474
Feature activation+6.74e-02

Pos logit contributions
 like+0.059
 favorite+0.044
 everywhere+0.042
 liked+0.041
 dolls+0.039
Neg logit contributions
ww-0.072
ister-0.071
www-0.069
ths-0.066
 service-0.065
Token They
Token ID1119
Feature activation-2.55e+00

Pos logit contributions
ister+0.002
ths+0.002
ww+0.002
 avail+0.002
www+0.002
Neg logit contributions
 like-0.001
 liked-0.001
 favorite-0.001
 everywhere-0.001
 both-0.001
Token both
Token ID1111
Feature activation-1.88e+00

Pos logit contributions
 like+0.047
joy+0.040
 hers+0.040
 Whe+0.039
aces+0.037
Neg logit contributions
ths-0.053
ww-0.051
www-0.051
 intense-0.050
 storm-0.050
Token says
Token ID1139
Feature activation-9.95e-01

Pos logit contributions
ister+0.003
ths+0.003
ww+0.003
www+0.003
 service+0.003
Neg logit contributions
 like-0.003
 everywhere-0.002
 wherever-0.002
 favorite-0.002
aces-0.002
Token,
Token ID11
Feature activation-4.06e+00

Pos logit contributions
 like+0.017
 everywhere+0.011
 favorite+0.011
 liked+0.010
 pad+0.010
Neg logit contributions
ister-0.031
ww-0.030
ths-0.029
www-0.029
 service-0.029
Token "
Token ID366
Feature activation-4.21e+00

Pos logit contributions
 like+0.057
 favorite+0.034
 everywhere+0.033
 liked+0.028
 dolls+0.028
Neg logit contributions
ister-0.131
ww-0.128
ths-0.128
www-0.126
 service-0.125
TokenW
Token ID54
Feature activation+2.55e+00

Pos logit contributions
 like+0.101
 everywhere+0.075
 favorite+0.074
 liked+0.072
 dolls+0.068
Neg logit contributions
ww-0.102
ister-0.101
ths-0.099
www-0.098
 service-0.095
Tokenoof
Token ID37711
Feature activation-1.39e+00

Pos logit contributions
ths+0.047
ister+0.046
 avail+0.045
ww+0.044
 storm+0.043
Neg logit contributions
 like-0.076
 favorite-0.057
 liked-0.057
 love-0.056
 everywhere-0.054
Token,
Token ID11
Feature activation-8.48e+00

Pos logit contributions
 like+0.026
 favorite+0.018
 liked+0.017
 everywhere+0.017
 around+0.015
Neg logit contributions
ww-0.041
ister-0.041
www-0.040
 service-0.038
ths-0.037
Token wo
Token ID24486
Feature activation+7.29e-01

Pos logit contributions
 like+0.153
 everywhere+0.096
 favorite+0.094
 liked+0.094
 back+0.083
Neg logit contributions
ths-0.256
ister-0.254
ww-0.253
 avail-0.250
www-0.248
Tokenof
Token ID1659
Feature activation-3.01e+00

Pos logit contributions
ister+0.016
ww+0.016
www+0.016
ths+0.016
 service+0.015
Neg logit contributions
 like-0.019
 favorite-0.014
 liked-0.014
 everywhere-0.014
 dolls-0.013
Token,
Token ID11
Feature activation-6.69e+00

Pos logit contributions
 like+0.054
 favorite+0.037
 everywhere+0.036
 liked+0.035
 dolls+0.032
Neg logit contributions
ister-0.090
ww-0.089
www-0.086
ths-0.084
 service-0.083
Token Tom
Token ID4186
Feature activation-2.21e+00

Pos logit contributions
 like+0.124
 everywhere+0.077
 liked+0.075
 favorite+0.074
 back+0.068
Neg logit contributions
ths-0.205
 avail-0.202
ister-0.200
ww-0.198
www-0.195
Token!
Token ID0
Feature activation-1.84e+00

Pos logit contributions
 like+0.036
 favorite+0.028
 everywhere+0.026
 liked+0.023
 dolls+0.023
Neg logit contributions
ister-0.066
ww-0.063
www-0.062
 beast-0.059
 service-0.059
Token said
Token ID531
Feature activation-9.64e-01

Pos logit contributions
 like+0.023
 everywhere+0.017
 liked+0.017
 favorite+0.017
 around+0.016
Neg logit contributions
ister-0.023
ww-0.023
ths-0.022
www-0.022
 service-0.021
Token,
Token ID11
Feature activation-4.98e+00

Pos logit contributions
 like+0.018
 pad+0.014
 everywhere+0.014
 favorite+0.013
 liked+0.012
Neg logit contributions
ister-0.027
ww-0.026
 service-0.025
www-0.025
ths-0.024
Token "
Token ID366
Feature activation-4.18e+00

Pos logit contributions
 like+0.074
 favorite+0.046
 everywhere+0.045
aces+0.042
 liked+0.040
Neg logit contributions
ister-0.154
ths-0.150
 service-0.146
ww-0.144
 sermon-0.144
TokenW
Token ID54
Feature activation+2.57e+00

Pos logit contributions
 like+0.098
 liked+0.070
 everywhere+0.069
 favorite+0.069
 wherever+0.063
Neg logit contributions
ister-0.104
ths-0.102
ww-0.102
www-0.099
 service-0.098
Tokenoof
Token ID37711
Feature activation-9.45e-01

Pos logit contributions
ister+0.050
ths+0.049
ww+0.048
www+0.047
 avail+0.046
Neg logit contributions
 like-0.074
 favorite-0.057
 liked-0.057
 everywhere-0.056
 pad-0.053
Token,
Token ID11
Feature activation-8.38e+00

Pos logit contributions
 like+0.017
 favorite+0.011
 everywhere+0.011
 liked+0.011
 around+0.010
Neg logit contributions
ww-0.028
ister-0.028
www-0.027
 service-0.026
ths-0.026
Token wo
Token ID24486
Feature activation+1.00e+00

Pos logit contributions
 like+0.149
 everywhere+0.092
 favorite+0.092
 liked+0.090
 dolls+0.081
Neg logit contributions
ths-0.257
ister-0.253
 avail-0.253
ww-0.252
www-0.249
Tokenof
Token ID1659
Feature activation-3.05e+00

Pos logit contributions
ister+0.023
ww+0.023
ths+0.022
www+0.022
 service+0.022
Neg logit contributions
 like-0.025
 everywhere-0.018
 favorite-0.018
 liked-0.018
 around-0.017
Token!"
Token ID2474
Feature activation-7.26e-01

Pos logit contributions
 like+0.054
 favorite+0.034
 everywhere+0.034
 liked+0.034
 around+0.030
Neg logit contributions
ister-0.092
ww-0.092
www-0.089
ths-0.089
 service-0.087
Token
Token ID198
Feature activation+5.52e-02

Pos logit contributions
 like+0.012
 favorite+0.008
 everywhere+0.008
 liked+0.007
 pad+0.007
Neg logit contributions
ister-0.022
ww-0.021
 service-0.021
iren-0.020
www-0.020
Token
Token ID198
Feature activation+1.17e+00

Pos logit contributions
ister+0.002
ww+0.002
ths+0.002
www+0.002
 service+0.002
Neg logit contributions
 like-0.001
 favorite-0.001
 everywhere-0.001
 liked-0.001
 pad-0.000
Token tail
Token ID7894
Feature activation-1.29e+00

Pos logit contributions
 like+0.038
 everywhere+0.027
 favorite+0.027
 liked+0.026
 around+0.024
Neg logit contributions
ister-0.046
ww-0.046
ths-0.045
www-0.045
 service-0.044
Token.
Token ID13
Feature activation-1.34e-01

Pos logit contributions
 like+0.016
 favorite+0.011
 liked+0.009
 everywhere+0.009
 wherever+0.009
Neg logit contributions
ister-0.041
ww-0.040
 service-0.039
 storm-0.039
www-0.039
Token "
Token ID366
Feature activation-4.47e+00

Pos logit contributions
 like+0.002
 everywhere+0.001
 favorite+0.001
 liked+0.001
 wherever+0.001
Neg logit contributions
ister-0.004
ww-0.004
www-0.004
ths-0.004
ama-0.004
TokenW
Token ID54
Feature activation+2.92e+00

Pos logit contributions
 like+0.103
 everywhere+0.079
 favorite+0.077
 liked+0.073
 back+0.071
Neg logit contributions
ww-0.109
ister-0.105
www-0.105
ths-0.103
 service-0.099
Tokenoof
Token ID37711
Feature activation-6.80e-01

Pos logit contributions
ister+0.056
ww+0.055
ths+0.054
www+0.053
 service+0.052
Neg logit contributions
 like-0.084
 everywhere-0.065
 favorite-0.065
 liked-0.064
 around-0.061
Token,
Token ID11
Feature activation-8.29e+00

Pos logit contributions
 like+0.013
 favorite+0.009
 everywhere+0.009
 liked+0.009
 around+0.008
Neg logit contributions
ww-0.020
ister-0.019
www-0.019
 service-0.018
ths-0.017
Token wo
Token ID24486
Feature activation+6.74e-01

Pos logit contributions
 like+0.163
 everywhere+0.110
 favorite+0.108
 liked+0.108
 around+0.097
Neg logit contributions
ister-0.234
ww-0.232
ths-0.229
www-0.227
 service-0.222
Tokenof
Token ID1659
Feature activation-2.69e+00

Pos logit contributions
ww+0.015
ister+0.015
www+0.014
ths+0.013
 service+0.013
Neg logit contributions
 like-0.018
 favorite-0.014
 liked-0.014
 everywhere-0.014
 dolls-0.013
Token!"
Token ID2474
Feature activation-5.62e-01

Pos logit contributions
 like+0.056
 favorite+0.039
 everywhere+0.039
 liked+0.038
 dolls+0.035
Neg logit contributions
ister-0.073
ww-0.073
www-0.071
ths-0.070
 service-0.068
Token
Token ID198
Feature activation-4.16e-01

Pos logit contributions
 like+0.008
 everywhere+0.005
 favorite+0.005
 liked+0.005
 dolls+0.004
Neg logit contributions
ister-0.018
ww-0.018
ths-0.018
www-0.018
 service-0.017
Token
Token ID198
Feature activation+9.31e-01

Pos logit contributions
 like+0.006
 everywhere+0.003
 favorite+0.003
 liked+0.003
 pad+0.003
Neg logit contributions
ister-0.014
ww-0.014
ths-0.013
www-0.013
 service-0.013
Token says
Token ID1139
Feature activation-1.23e+00

Pos logit contributions
ww+0.027
ister+0.026
ths+0.026
www+0.026
 service+0.025
Neg logit contributions
 like-0.027
 everywhere-0.020
 favorite-0.020
 liked-0.019
 wherever-0.019
Token,
Token ID11
Feature activation-3.87e+00

Pos logit contributions
 like+0.021
 everywhere+0.014
 favorite+0.014
 liked+0.014
 pad+0.013
Neg logit contributions
ister-0.037
ww-0.037
ths-0.035
www-0.035
 service-0.035
Token "
Token ID366
Feature activation-4.09e+00

Pos logit contributions
 like+0.057
 favorite+0.037
 everywhere+0.034
 dolls+0.032
aces+0.031
Neg logit contributions
ister-0.119
ths-0.116
ww-0.115
 storm-0.114
 service-0.114
TokenW
Token ID54
Feature activation+2.39e+00

Pos logit contributions
 like+0.090
 liked+0.064
 favorite+0.060
 everywhere+0.058
 love+0.056
Neg logit contributions
ister-0.108
ths-0.108
 avail-0.104
 service-0.103
ww-0.102
Tokenoof
Token ID37711
Feature activation-1.17e+00

Pos logit contributions
ths+0.051
ister+0.050
ww+0.049
 service+0.048
 avail+0.048
Neg logit contributions
 like-0.065
 liked-0.047
 favorite-0.047
 everywhere-0.047
 pad-0.045
Token,
Token ID11
Feature activation-8.26e+00

Pos logit contributions
 like+0.022
 favorite+0.016
 liked+0.015
 everywhere+0.015
 around+0.014
Neg logit contributions
ww-0.034
ister-0.034
www-0.033
 service-0.031
 storm-0.030
Token wo
Token ID24486
Feature activation+9.28e-01

Pos logit contributions
 like+0.154
 everywhere+0.099
 favorite+0.098
 liked+0.097
 dolls+0.086
Neg logit contributions
ister-0.243
ths-0.242
ww-0.241
www-0.237
 avail-0.234
Tokenof
Token ID1659
Feature activation-2.92e+00

Pos logit contributions
ister+0.021
ww+0.021
www+0.020
ths+0.020
 service+0.019
Neg logit contributions
 like-0.024
 favorite-0.018
 liked-0.018
 everywhere-0.018
 dolls-0.017
Token,
Token ID11
Feature activation-6.84e+00

Pos logit contributions
 like+0.054
 favorite+0.036
 everywhere+0.036
 liked+0.035
 dolls+0.032
Neg logit contributions
ister-0.086
ww-0.086
www-0.083
ths-0.081
 service-0.080
Token my
Token ID616
Feature activation-1.94e+00

Pos logit contributions
 like+0.132
 everywhere+0.086
 favorite+0.084
 liked+0.084
 back+0.076
Neg logit contributions
ths-0.200
ister-0.198
ww-0.196
www-0.193
 avail-0.193
Token name
Token ID1438
Feature activation-1.75e+00

Pos logit contributions
 like+0.038
 everywhere+0.026
 liked+0.025
 favorite+0.025
 around+0.023
Neg logit contributions
ister-0.054
ww-0.054
ths-0.053
www-0.052
 service-0.051
Token said
Token ID531
Feature activation-8.37e-01

Pos logit contributions
 like+0.007
 everywhere+0.005
 favorite+0.005
 liked+0.005
aces+0.005
Neg logit contributions
ww-0.008
www-0.008
ths-0.008
ister-0.008
 service-0.008
Token,
Token ID11
Feature activation-3.43e+00

Pos logit contributions
 like+0.017
 pad+0.013
 everywhere+0.012
 favorite+0.011
 favorites+0.011
Neg logit contributions
ww-0.022
ister-0.022
www-0.021
 service-0.020
ths-0.020
Token "
Token ID366
Feature activation-3.99e+00

Pos logit contributions
 like+0.053
 favorite+0.034
 everywhere+0.032
aces+0.031
 liked+0.030
Neg logit contributions
ister-0.102
ww-0.099
www-0.099
ths-0.098
 storm-0.096
TokenW
Token ID54
Feature activation+2.91e+00

Pos logit contributions
 like+0.102
 everywhere+0.082
 favorite+0.082
aces+0.077
 dolls+0.077
Neg logit contributions
ww-0.092
www-0.089
ths-0.082
ister-0.081
body-0.078
Tokenoof
Token ID37711
Feature activation-9.27e-01

Pos logit contributions
ister+0.056
ww+0.055
ths+0.054
www+0.053
 service+0.051
Neg logit contributions
 like-0.084
 everywhere-0.065
 favorite-0.065
 liked-0.064
 around-0.061
Token,
Token ID11
Feature activation-8.13e+00

Pos logit contributions
 like+0.018
 favorite+0.013
 liked+0.013
 everywhere+0.012
 around+0.012
Neg logit contributions
ww-0.027
ister-0.026
www-0.025
 service-0.023
 storm-0.023
Token wo
Token ID24486
Feature activation+1.21e+00

Pos logit contributions
 like+0.170
 everywhere+0.123
 favorite+0.123
 liked+0.121
 around+0.115
Neg logit contributions
ister-0.214
ww-0.212
www-0.204
 storm-0.197
 service-0.196
Tokenof
Token ID1659
Feature activation-2.54e+00

Pos logit contributions
ister+0.026
ww+0.026
www+0.025
ths+0.024
 service+0.024
Neg logit contributions
 like-0.032
 favorite-0.024
 liked-0.024
 everywhere-0.024
 dolls-0.023
Token!"
Token ID2474
Feature activation+4.60e-01

Pos logit contributions
 like+0.051
 favorite+0.036
 everywhere+0.035
 liked+0.034
 dolls+0.032
Neg logit contributions
ww-0.071
ister-0.071
www-0.068
ths-0.066
 service-0.065
Token The
Token ID383
Feature activation+1.21e+00

Pos logit contributions
ths+0.016
ister+0.016
 avail+0.016
ww+0.015
www+0.015
Neg logit contributions
 like-0.007
 liked-0.005
 and-0.004
 favorite-0.004
 Lily-0.004
Token leash
Token ID40529
Feature activation+3.35e+00

Pos logit contributions
ww+0.040
ister+0.040
www+0.039
ths+0.038
 storm+0.038
Neg logit contributions
 like-0.018
 everywhere-0.010
 favorite-0.010
 liked-0.010
 around-0.009
Token says
Token ID1139
Feature activation-7.59e-01

Pos logit contributions
ww+0.011
ister+0.010
www+0.010
 service+0.010
 sermon+0.010
Neg logit contributions
 like-0.011
 everywhere-0.009
 wherever-0.008
aces-0.008
 favorite-0.008
Token,
Token ID11
Feature activation-4.48e+00

Pos logit contributions
 like+0.015
 pad+0.012
 favorite+0.010
 liked+0.010
 everywhere+0.010
Neg logit contributions
ister-0.021
ww-0.020
 service-0.019
ths-0.019
www-0.018
Token "
Token ID366
Feature activation-4.20e+00

Pos logit contributions
 like+0.066
 favorite+0.042
 everywhere+0.041
aces+0.036
 pad+0.035
Neg logit contributions
ister-0.139
ths-0.136
 service-0.135
ww-0.134
 sermon-0.132
TokenW
Token ID54
Feature activation+2.36e+00

Pos logit contributions
 like+0.093
 everywhere+0.071
 favorite+0.067
 liked+0.063
 pad+0.063
Neg logit contributions
ww-0.109
ister-0.105
www-0.104
ths-0.101
 service-0.099
Tokenoof
Token ID37711
Feature activation-1.24e+00

Pos logit contributions
ths+0.049
ister+0.048
ww+0.048
www+0.046
 avail+0.046
Neg logit contributions
 like-0.065
 favorite-0.048
 liked-0.048
 everywhere-0.048
 love-0.045
Token,
Token ID11
Feature activation-8.08e+00

Pos logit contributions
 like+0.024
 favorite+0.018
 everywhere+0.016
 liked+0.016
 around+0.015
Neg logit contributions
ww-0.036
ister-0.035
www-0.034
 service-0.032
 storm-0.031
Token wo
Token ID24486
Feature activation+6.21e-01

Pos logit contributions
 like+0.158
 everywhere+0.108
 favorite+0.105
 liked+0.105
 around+0.096
Neg logit contributions
ister-0.227
ww-0.226
ths-0.220
www-0.220
 service-0.216
Tokenof
Token ID1659
Feature activation-3.13e+00

Pos logit contributions
ister+0.014
ww+0.014
www+0.013
ths+0.013
 service+0.012
Neg logit contributions
 like-0.016
 favorite-0.013
 liked-0.012
 everywhere-0.012
 dolls-0.012
Token.
Token ID13
Feature activation-1.28e+00

Pos logit contributions
 like+0.054
 favorite+0.039
 everywhere+0.036
 liked+0.034
 pad+0.033
Neg logit contributions
ister-0.094
ww-0.094
www-0.089
 service-0.086
oo-0.085
Token You
Token ID921
Feature activation-4.06e+00

Pos logit contributions
 like+0.033
 everywhere+0.028
 favorite+0.026
 liked+0.026
 around+0.025
Neg logit contributions
ww-0.026
ister-0.025
www-0.024
 service-0.023
 storm-0.023
Token are
Token ID389
Feature activation-6.17e-01

Pos logit contributions
 like+0.055
 favorite+0.040
 everywhere+0.039
 pad+0.037
 dolls+0.035
Neg logit contributions
ww-0.123
ister-0.121
www-0.119
 storm-0.118
 service-0.117
Token said
Token ID531
Feature activation-5.59e-01

Pos logit contributions
 like+0.004
 everywhere+0.003
 favorite+0.003
 wherever+0.003
 pad+0.003
Neg logit contributions
ww-0.004
ths-0.004
ister-0.004
www-0.004
 service-0.004
Token,
Token ID11
Feature activation-4.67e+00

Pos logit contributions
 like+0.011
 pad+0.009
 everywhere+0.008
 favorite+0.008
 liked+0.008
Neg logit contributions
ister-0.015
ww-0.015
www-0.013
 service-0.013
 storm-0.013
Token "
Token ID366
Feature activation-4.24e+00

Pos logit contributions
 like+0.079
 favorite+0.057
 favorites+0.052
aces+0.052
 everywhere+0.051
Neg logit contributions
ister-0.128
 storm-0.121
 service-0.121
ths-0.120
ller-0.119
TokenW
Token ID54
Feature activation+2.64e+00

Pos logit contributions
 like+0.103
 everywhere+0.078
 favorite+0.077
 liked+0.074
 around+0.071
Neg logit contributions
ww-0.101
ister-0.099
www-0.097
ths-0.097
 service-0.093
Tokenoof
Token ID37711
Feature activation-7.03e-01

Pos logit contributions
ister+0.050
ww+0.050
ths+0.048
www+0.048
 service+0.046
Neg logit contributions
 like-0.077
 everywhere-0.060
 favorite-0.059
 liked-0.059
 around-0.056
Token,
Token ID11
Feature activation-8.05e+00

Pos logit contributions
 like+0.013
 favorite+0.009
 liked+0.009
 everywhere+0.009
 around+0.008
Neg logit contributions
ww-0.021
ister-0.021
www-0.020
 service-0.019
 storm-0.019
Token wo
Token ID24486
Feature activation+1.15e+00

Pos logit contributions
 like+0.160
 everywhere+0.110
 favorite+0.108
 liked+0.108
 around+0.100
Neg logit contributions
ister-0.224
ww-0.223
www-0.216
ths-0.216
 service-0.212
Tokenof
Token ID1659
Feature activation-2.63e+00

Pos logit contributions
ister+0.026
ww+0.026
www+0.025
ths+0.024
 service+0.024
Neg logit contributions
 like-0.030
 favorite-0.023
 liked-0.022
 everywhere-0.022
 dolls-0.021
Token,
Token ID11
Feature activation-6.50e+00

Pos logit contributions
 like+0.047
 favorite+0.032
 everywhere+0.030
 liked+0.030
 dolls+0.027
Neg logit contributions
ister-0.079
ww-0.079
www-0.076
ths-0.074
 service-0.073
Token wo
Token ID24486
Feature activation+1.69e+00

Pos logit contributions
 like+0.137
 everywhere+0.095
 favorite+0.094
 liked+0.094
 around+0.086
Neg logit contributions
ister-0.174
ww-0.173
ths-0.168
www-0.168
 service-0.164
Tokenof
Token ID1659
Feature activation-2.40e+00

Pos logit contributions
ister+0.038
ww+0.038
ths+0.037
www+0.037
 service+0.036
Neg logit contributions
 like-0.043
 everywhere-0.032
 favorite-0.031
 liked-0.031
 dolls-0.029
Token
Token ID198
Feature activation-1.09e-01

Pos logit contributions
 like+0.015
 everywhere+0.009
 favorite+0.008
 liked+0.008
 wherever+0.007
Neg logit contributions
ister-0.031
ww-0.031
ths-0.031
www-0.030
 service-0.030
Token
Token ID198
Feature activation+8.26e-01

Pos logit contributions
 like+0.002
 everywhere+0.001
 favorite+0.001
 liked+0.001
 around+0.001
Neg logit contributions
ister-0.004
ww-0.004
ths-0.004
www-0.004
 service-0.004
Token"
Token ID1
Feature activation-3.59e+00

Pos logit contributions
ths+0.023
ister+0.023
ww+0.023
 avail+0.023
 service+0.022
Neg logit contributions
 like-0.016
 liked-0.011
 everywhere-0.010
 favorite-0.010
 both-0.010
TokenW
Token ID54
Feature activation+2.93e+00

Pos logit contributions
 like+0.073
 liked+0.051
 favorite+0.049
 everywhere+0.046
 both+0.043
Neg logit contributions
ister-0.100
ths-0.099
 avail-0.098
ww-0.097
 service-0.096
Tokenoof
Token ID37711
Feature activation-6.42e-01

Pos logit contributions
ister+0.057
ww+0.056
ths+0.056
www+0.054
 service+0.053
Neg logit contributions
 like-0.084
 everywhere-0.064
 favorite-0.064
 liked-0.064
 around-0.060
Token,
Token ID11
Feature activation-8.00e+00

Pos logit contributions
 like+0.012
 favorite+0.008
 everywhere+0.008
 liked+0.008
 wherever+0.007
Neg logit contributions
ww-0.019
ister-0.019
www-0.018
 service-0.018
ths-0.017
Token wo
Token ID24486
Feature activation+1.17e+00

Pos logit contributions
 like+0.153
 everywhere+0.101
 favorite+0.100
 liked+0.099
 around+0.089
Neg logit contributions
ister-0.230
ww-0.229
ths-0.226
www-0.223
 service-0.219
Tokenof
Token ID1659
Feature activation-2.53e+00

Pos logit contributions
ister+0.027
ww+0.027
www+0.026
ths+0.026
 service+0.025
Neg logit contributions
 like-0.030
 favorite-0.022
 liked-0.022
 everywhere-0.022
 dolls-0.020
Token!"
Token ID2474
Feature activation-8.34e-01

Pos logit contributions
 like+0.044
 everywhere+0.027
 favorite+0.027
 liked+0.027
 around+0.023
Neg logit contributions
ister-0.077
ww-0.077
ths-0.076
www-0.075
 service-0.074
Token Spot
Token ID15899
Feature activation+1.35e+00

Pos logit contributions
 like+0.010
 liked+0.004
 favorite+0.004
 everywhere+0.004
 both+0.003
Neg logit contributions
ths-0.031
ister-0.031
ww-0.030
www-0.030
 service-0.029
Token bark
Token ID21405
Feature activation-1.14e-01

Pos logit contributions
ister+0.029
ww+0.028
www+0.028
ths+0.028
 service+0.026
Neg logit contributions
 like-0.035
 everywhere-0.027
 favorite-0.026
 wherever-0.025
 liked-0.025
Token said
Token ID531
Feature activation-5.82e-01

Pos logit contributions
 like+0.003
 everywhere+0.002
 wherever+0.002
 favorite+0.002
 pad+0.002
Neg logit contributions
ths-0.003
ww-0.003
ister-0.003
www-0.003
 service-0.003
Token,
Token ID11
Feature activation-4.46e+00

Pos logit contributions
 like+0.011
 pad+0.007
 everywhere+0.007
 liked+0.007
 favorite+0.007
Neg logit contributions
ister-0.017
ww-0.017
ths-0.016
www-0.016
 service-0.016
Token "
Token ID366
Feature activation-3.93e+00

Pos logit contributions
 like+0.066
 favorite+0.041
 everywhere+0.038
 liked+0.033
 pad+0.033
Neg logit contributions
ister-0.141
ths-0.135
ww-0.135
 service-0.134
 storm-0.133
TokenW
Token ID54
Feature activation+2.38e+00

Pos logit contributions
 like+0.081
 liked+0.056
 favorite+0.053
 everywhere+0.052
 both+0.048
Neg logit contributions
ister-0.109
ths-0.108
ww-0.106
www-0.105
 service-0.104
Tokenoof
Token ID37711
Feature activation-9.88e-01

Pos logit contributions
ths+0.048
ister+0.048
ww+0.047
 service+0.046
www+0.046
Neg logit contributions
 like-0.066
 liked-0.049
 favorite-0.049
 everywhere-0.049
 love-0.046
Token,
Token ID11
Feature activation-7.99e+00

Pos logit contributions
 like+0.018
 favorite+0.013
 liked+0.012
 everywhere+0.012
 around+0.011
Neg logit contributions
ww-0.030
ister-0.030
www-0.029
 service-0.027
ths-0.027
Token wo
Token ID24486
Feature activation+1.15e+00

Pos logit contributions
 like+0.154
 everywhere+0.101
 favorite+0.100
 liked+0.099
 dolls+0.089
Neg logit contributions
ister-0.229
ths-0.229
ww-0.228
www-0.224
 avail-0.221
Tokenof
Token ID1659
Feature activation-2.78e+00

Pos logit contributions
ister+0.026
ww+0.026
ths+0.025
www+0.025
 service+0.025
Neg logit contributions
 like-0.029
 favorite-0.021
 everywhere-0.021
 liked-0.021
 dolls-0.020
Token,
Token ID11
Feature activation-6.29e+00

Pos logit contributions
 like+0.047
 favorite+0.031
 everywhere+0.030
 liked+0.029
 dolls+0.026
Neg logit contributions
ister-0.086
ww-0.086
www-0.083
ths-0.081
 service-0.080
Token sorry
Token ID7926
Feature activation-2.68e-01

Pos logit contributions
 like+0.152
 everywhere+0.111
 favorite+0.110
 liked+0.109
 dolls+0.101
Neg logit contributions
ister-0.150
ww-0.149
ths-0.148
www-0.145
 avail-0.142
Token,
Token ID11
Feature activation-4.49e+00

Pos logit contributions
 like+0.005
 everywhere+0.004
 liked+0.004
 pad+0.004
aces+0.004
Neg logit contributions
ww-0.007
ister-0.007
www-0.007
 storm-0.007
 service-0.007
Token
Token ID198
Feature activation-9.63e-01

Pos logit contributions
 like+0.027
 everywhere+0.018
 favorite+0.018
 liked+0.016
 pad+0.016
Neg logit contributions
ister-0.046
ww-0.045
www-0.044
 service-0.043
ths-0.043
Token
Token ID198
Feature activation+3.29e-01

Pos logit contributions
 like+0.016
 favorite+0.010
 everywhere+0.009
 liked+0.009
 pad+0.008
Neg logit contributions
ister-0.031
ww-0.030
www-0.029
ths-0.029
 service-0.029
Token"
Token ID1
Feature activation-3.48e+00

Pos logit contributions
ister+0.009
ww+0.009
ths+0.008
www+0.008
 service+0.008
Neg logit contributions
 like-0.007
 everywhere-0.005
 liked-0.005
 favorite-0.005
 around-0.004
TokenW
Token ID54
Feature activation+2.71e+00

Pos logit contributions
 like+0.072
 liked+0.053
 favorite+0.047
 love+0.043
 dolls+0.042
Neg logit contributions
 avail-0.098
ths-0.097
ister-0.096
asis-0.094
 service-0.093
Tokenoof
Token ID37711
Feature activation-5.14e-01

Pos logit contributions
ths+0.058
ister+0.057
ww+0.056
www+0.055
 avail+0.054
Neg logit contributions
 like-0.074
 favorite-0.054
 liked-0.054
 everywhere-0.052
 pad-0.051
Token,
Token ID11
Feature activation-7.93e+00

Pos logit contributions
 like+0.009
 favorite+0.006
 everywhere+0.006
 liked+0.006
 wherever+0.005
Neg logit contributions
ww-0.016
ister-0.016
www-0.015
ths-0.015
 service-0.015
Token wo
Token ID24486
Feature activation+1.06e+00

Pos logit contributions
 like+0.140
 everywhere+0.084
 liked+0.083
 favorite+0.082
 dolls+0.074
Neg logit contributions
ths-0.250
 avail-0.247
ister-0.241
ww-0.240
www-0.239
Tokenof
Token ID1659
Feature activation-2.74e+00

Pos logit contributions
ister+0.024
ww+0.024
www+0.023
ths+0.023
 service+0.022
Neg logit contributions
 like-0.027
 favorite-0.020
 everywhere-0.020
 liked-0.020
 dolls-0.019
Token!"
Token ID2474
Feature activation-1.22e+00

Pos logit contributions
 like+0.043
 everywhere+0.026
 liked+0.025
 favorite+0.025
 around+0.021
Neg logit contributions
ister-0.088
ww-0.087
ths-0.087
www-0.086
 service-0.084
Token the
Token ID262
Feature activation+1.89e+00

Pos logit contributions
 like+0.016
 everywhere+0.008
 liked+0.008
 favorite+0.008
 both+0.007
Neg logit contributions
ww-0.043
ths-0.043
ister-0.043
www-0.042
 service-0.041
Token dolls
Token ID36062
Feature activation-2.96e-01

Pos logit contributions
ths+0.060
 avail+0.057
ww+0.056
www+0.055
ister+0.055
Neg logit contributions
 like-0.033
 dolls-0.024
 liked-0.024
 favorite-0.023
 back-0.021
Token says
Token ID1139
Feature activation-1.22e+00

Pos logit contributions
 like+0.004
 everywhere+0.003
 liked+0.003
 favorite+0.003
 around+0.003
Neg logit contributions
ister-0.005
ww-0.005
ths-0.005
www-0.005
 service-0.005
Token,
Token ID11
Feature activation-5.18e+00

Pos logit contributions
 like+0.022
 everywhere+0.014
 favorite+0.014
 liked+0.014
 around+0.013
Neg logit contributions
ister-0.036
ww-0.036
ths-0.035
www-0.035
 service-0.034
Token "
Token ID366
Feature activation-3.58e+00

Pos logit contributions
 like+0.068
 favorite+0.039
 everywhere+0.037
 liked+0.035
aces+0.030
Neg logit contributions
ister-0.172
ww-0.167
ths-0.166
 service-0.165
www-0.163
TokenW
Token ID54
Feature activation+2.02e+00

Pos logit contributions
 like+0.081
 liked+0.057
 everywhere+0.056
 favorite+0.056
 around+0.051
Neg logit contributions
ister-0.092
ww-0.090
ths-0.090
www-0.088
 service-0.087
Tokenoof
Token ID37711
Feature activation-9.33e-01

Pos logit contributions
ths+0.043
ister+0.042
ww+0.041
 avail+0.040
 storm+0.040
Neg logit contributions
 like-0.055
 liked-0.040
 favorite-0.040
 everywhere-0.038
 love-0.038
Token,
Token ID11
Feature activation-7.87e+00

Pos logit contributions
 like+0.016
 favorite+0.011
 everywhere+0.011
 liked+0.010
 around+0.009
Neg logit contributions
ister-0.028
ww-0.028
www-0.027
ths-0.027
 service-0.027
Token wo
Token ID24486
Feature activation+9.63e-01

Pos logit contributions
 like+0.149
 everywhere+0.097
 favorite+0.095
 liked+0.094
 dolls+0.084
Neg logit contributions
ister-0.228
ths-0.228
ww-0.227
www-0.222
 avail-0.221
Tokenof
Token ID1659
Feature activation-2.74e+00

Pos logit contributions
ister+0.022
ww+0.022
ths+0.022
www+0.022
 service+0.021
Neg logit contributions
 like-0.024
 everywhere-0.018
 favorite-0.017
 liked-0.017
 around-0.016
Token.
Token ID13
Feature activation-1.40e+00

Pos logit contributions
 like+0.047
 everywhere+0.029
 favorite+0.029
 liked+0.028
 around+0.025
Neg logit contributions
ister-0.084
ww-0.084
ths-0.082
www-0.082
 service-0.080
Token I
Token ID314
Feature activation-2.95e+00

Pos logit contributions
 like+0.035
 everywhere+0.027
 favorite+0.026
 liked+0.026
 around+0.025
Neg logit contributions
ww-0.032
ister-0.031
www-0.030
 service-0.029
ths-0.029
Token understand
Token ID1833
Feature activation+1.10e-02

Pos logit contributions
 like+0.052
 everywhere+0.037
 favorite+0.036
 liked+0.033
 dolls+0.033
Neg logit contributions
ister-0.084
ww-0.084
www-0.082
 service-0.080
ths-0.080
Token!"
Token ID2474
Feature activation-6.83e-01

Pos logit contributions
 like+0.011
 favorite+0.009
 everywhere+0.009
 skirts+0.008
 liked+0.008
Neg logit contributions
ister-0.010
ww-0.010
www-0.009
 beast-0.009
 service-0.009
Token
Token ID198
Feature activation-2.07e-01

Pos logit contributions
 like+0.011
 everywhere+0.007
 favorite+0.007
 liked+0.007
 around+0.006
Neg logit contributions
ister-0.022
ww-0.021
ths-0.021
www-0.021
 service-0.021
Token
Token ID198
Feature activation+1.01e+00

Pos logit contributions
 like+0.003
 everywhere+0.002
 favorite+0.002
 liked+0.002
 pad+0.001
Neg logit contributions
ister-0.007
ww-0.007
ths-0.007
www-0.007
 service-0.007
Token"
Token ID1
Feature activation-3.82e+00

Pos logit contributions
ister+0.026
ww+0.026
ths+0.025
www+0.025
 service+0.024
Neg logit contributions
 like-0.023
 everywhere-0.016
 favorite-0.016
 liked-0.016
 around-0.015
TokenBang
Token ID43984
Feature activation-2.12e-01

Pos logit contributions
 like+0.108
 everywhere+0.089
 favorite+0.087
 where+0.083
 pad+0.083
Neg logit contributions
ww-0.075
www-0.071
ths-0.070
ister-0.069
body-0.064
Token,
Token ID11
Feature activation-7.79e+00

Pos logit contributions
 like+0.004
 favorite+0.004
 liked+0.003
 everywhere+0.003
 dolls+0.003
Neg logit contributions
ister-0.006
ww-0.006
www-0.005
ths-0.005
 beast-0.005
Token bang
Token ID20188
Feature activation-1.52e+00

Pos logit contributions
 like+0.210
 everywhere+0.159
 favorite+0.158
 liked+0.157
 around+0.148
Neg logit contributions
ister-0.164
ww-0.162
ths-0.159
www-0.157
 service-0.152
Token,
Token ID11
Feature activation-6.56e+00

Pos logit contributions
 like+0.030
 favorite+0.021
 liked+0.021
 everywhere+0.021
 dolls+0.019
Neg logit contributions
ister-0.042
ww-0.042
www-0.041
ths-0.041
 service-0.040
Token bang
Token ID20188
Feature activation-8.19e-01

Pos logit contributions
 like+0.177
 everywhere+0.135
 favorite+0.135
 liked+0.134
 around+0.126
Neg logit contributions
ister-0.136
ww-0.135
ths-0.131
www-0.130
 service-0.126
Token!"
Token ID2474
Feature activation-1.01e+00

Pos logit contributions
 like+0.017
 favorite+0.012
 liked+0.012
 everywhere+0.012
 dolls+0.011
Neg logit contributions
ww-0.022
ister-0.022
www-0.021
ths-0.021
 service-0.020
Token Sam
Token ID3409
Feature activation+2.87e-01

Pos logit contributions
 like+0.015
 everywhere+0.008
 favorite+0.008
 liked+0.008
 around+0.007
Neg logit contributions
ister-0.034
ww-0.033
ths-0.033
www-0.033
 service-0.032
Token.
Token ID13
Feature activation-1.26e+00

Pos logit contributions
 like+0.004
 Whe+0.004
 favorite+0.003
 liked+0.003
 ti+0.003
Neg logit contributions
 service-0.006
 outage-0.005
ister-0.005
asis-0.005
 site-0.005
Token
Token ID198
Feature activation-1.29e+00

Pos logit contributions
 like+0.022
 everywhere+0.015
 favorite+0.014
 liked+0.014
 pad+0.013
Neg logit contributions
ister-0.037
ww-0.037
ths-0.036
www-0.036
 service-0.036
Token
Token ID198
Feature activation-6.32e-03

Pos logit contributions
 like+0.020
 favorite+0.012
 everywhere+0.012
 liked+0.012
 pad+0.011
Neg logit contributions
ister-0.041
ww-0.040
ths-0.039
www-0.039
 service-0.039
Token"
Token ID1
Feature activation-3.54e+00

Pos logit contributions
 like+0.000
 liked+0.000
 everywhere+0.000
 favorite+0.000
 both+0.000
Neg logit contributions
ister-0.000
ww-0.000
ths-0.000
www-0.000
 avail-0.000
TokenBang
Token ID43984
Feature activation-9.45e-02

Pos logit contributions
 like+0.090
 everywhere+0.069
 favorite+0.068
 liked+0.065
 around+0.064
Neg logit contributions
ww-0.081
ister-0.078
www-0.078
ths-0.078
 service-0.075
Token,
Token ID11
Feature activation-7.71e+00

Pos logit contributions
 like+0.002
 liked+0.001
 favorite+0.001
 everywhere+0.001
 dolls+0.001
Neg logit contributions
ister-0.003
ww-0.003
www-0.003
ths-0.002
 avail-0.002
Token bang
Token ID20188
Feature activation-1.41e+00

Pos logit contributions
 like+0.199
 favorite+0.155
 liked+0.151
 everywhere+0.151
 around+0.143
Neg logit contributions
ister-0.168
ww-0.165
www-0.159
ths-0.159
 storm-0.156
Token,
Token ID11
Feature activation-6.65e+00

Pos logit contributions
 like+0.024
 everywhere+0.015
 favorite+0.015
 liked+0.015
 around+0.013
Neg logit contributions
ister-0.043
ww-0.043
ths-0.042
www-0.042
 service-0.041
Token bang
Token ID20188
Feature activation-7.43e-01

Pos logit contributions
 like+0.167
 favorite+0.126
 everywhere+0.126
 liked+0.125
 around+0.117
Neg logit contributions
ister-0.150
ww-0.149
ths-0.144
www-0.144
 service-0.140
Token!"
Token ID2474
Feature activation-1.37e+00

Pos logit contributions
 like+0.014
 everywhere+0.010
 favorite+0.009
 liked+0.009
 around+0.009
Neg logit contributions
ister-0.021
ww-0.021
ths-0.021
www-0.020
 service-0.020
Token Tom
Token ID4186
Feature activation+6.01e-01

Pos logit contributions
 like+0.019
 liked+0.009
 everywhere+0.009
 favorite+0.009
 both+0.008
Neg logit contributions
ww-0.049
ths-0.049
ister-0.049
www-0.048
 service-0.047
Token neck
Token ID7393
Feature activation-2.18e+00

Pos logit contributions
ister+0.011
ths+0.011
ww+0.011
www+0.011
 avail+0.010
Neg logit contributions
 like-0.010
 everywhere-0.007
 liked-0.007
 favorite-0.006
 both-0.006
Tokenl
Token ID75
Feature activation-1.83e+00

Pos logit contributions
 like+0.027
 favorite+0.016
 everywhere+0.015
 liked+0.015
 dolls+0.014
Neg logit contributions
ister-0.072
ww-0.070
ths-0.069
 service-0.069
www-0.068
Tokenaces
Token ID2114
Feature activation-3.08e+00

Pos logit contributions
 like+0.030
 everywhere+0.018
 favorite+0.018
 liked+0.017
 dolls+0.015
Neg logit contributions
ister-0.057
ww-0.057
ths-0.056
www-0.055
 service-0.055
Token,
Token ID11
Feature activation-3.21e+00

Pos logit contributions
 like+0.041
 dolls+0.029
 favorite+0.029
 Whe+0.026
 love+0.026
Neg logit contributions
ister-0.093
ww-0.086
ths-0.085
www-0.083
keys-0.083
Token grape
Token ID30777
Feature activation-1.81e+00

Pos logit contributions
 like+0.044
 favorite+0.025
 everywhere+0.025
 liked+0.023
 dolls+0.020
Neg logit contributions
ister-0.103
ths-0.102
ww-0.102
www-0.100
 service-0.100
Token brace
Token ID22581
Feature activation-7.69e+00

Pos logit contributions
 like+0.033
 favorite+0.022
 everywhere+0.022
 liked+0.021
 wherever+0.019
Neg logit contributions
ww-0.052
ister-0.052
ths-0.051
www-0.051
 service-0.050
Tokenlets
Token ID5289
Feature activation-5.53e+00

Pos logit contributions
 like+0.164
 Whe+0.130
 everywhere+0.130
 liked+0.127
 dolls+0.127
Neg logit contributions
ww-0.171
ller-0.157
www-0.156
ths-0.151
ink-0.148
Token,
Token ID11
Feature activation-4.21e+00

Pos logit contributions
 like+0.075
 favorite+0.055
 Whe+0.052
 liked+0.049
 dolls+0.047
Neg logit contributions
ister-0.160
 service-0.153
ww-0.151
ths-0.149
field-0.149
Token and
Token ID290
Feature activation-2.12e+00

Pos logit contributions
 like+0.075
 everywhere+0.050
 favorite+0.048
 liked+0.047
 around+0.044
Neg logit contributions
ww-0.118
ister-0.116
ths-0.115
www-0.114
 service-0.112
Token grape
Token ID30777
Feature activation-1.98e+00

Pos logit contributions
 like+0.041
 everywhere+0.027
 favorite+0.027
 liked+0.027
 dolls+0.024
Neg logit contributions
ister-0.061
ww-0.060
ths-0.060
www-0.059
 service-0.058
Token crown
Token ID12389
Feature activation-3.56e+00

Pos logit contributions
 like+0.033
 favorite+0.022
 everywhere+0.021
 liked+0.021
 wherever+0.018
Neg logit contributions
ww-0.059
ister-0.058
www-0.057
ths-0.057
 service-0.056
Token says
Token ID1139
Feature activation-1.00e+00

Pos logit contributions
 like+0.001
 everywhere+0.001
 wherever+0.001
 favorite+0.001
 pad+0.001
Neg logit contributions
ister-0.001
ths-0.001
ww-0.001
 service-0.001
www-0.001
Token,
Token ID11
Feature activation-4.14e+00

Pos logit contributions
 like+0.019
 pad+0.014
 everywhere+0.013
 favorite+0.013
 liked+0.013
Neg logit contributions
ister-0.028
ww-0.028
www-0.026
 service-0.026
ths-0.026
Token "
Token ID366
Feature activation-3.99e+00

Pos logit contributions
 like+0.061
 favorite+0.037
 everywhere+0.036
 liked+0.031
aces+0.030
Neg logit contributions
ister-0.132
ths-0.127
ww-0.127
 service-0.125
www-0.124
TokenW
Token ID54
Feature activation+2.50e+00

Pos logit contributions
 like+0.090
 liked+0.063
 everywhere+0.062
 favorite+0.062
 wherever+0.057
Neg logit contributions
ister-0.103
ths-0.101
ww-0.101
www-0.099
 service-0.098
Tokenoof
Token ID37711
Feature activation-8.42e-01

Pos logit contributions
ister+0.050
ths+0.049
ww+0.049
www+0.048
 service+0.047
Neg logit contributions
 like-0.070
 favorite-0.053
 everywhere-0.053
 liked-0.053
 pad-0.049
Token,
Token ID11
Feature activation-7.63e+00

Pos logit contributions
 like+0.015
 favorite+0.011
 everywhere+0.010
 liked+0.010
 around+0.009
Neg logit contributions
ww-0.025
ister-0.025
www-0.024
 service-0.023
 storm-0.022
Token wo
Token ID24486
Feature activation+1.26e+00

Pos logit contributions
 like+0.141
 everywhere+0.090
 favorite+0.089
 liked+0.089
 dolls+0.079
Neg logit contributions
ister-0.225
ths-0.225
ww-0.224
www-0.220
 avail-0.218
Tokenof
Token ID1659
Feature activation-2.61e+00

Pos logit contributions
ister+0.028
ww+0.028
www+0.027
ths+0.026
 service+0.026
Neg logit contributions
 like-0.033
 favorite-0.025
 liked-0.025
 everywhere-0.025
 dolls-0.023
Token,
Token ID11
Feature activation-6.17e+00

Pos logit contributions
 like+0.041
 favorite+0.026
 everywhere+0.025
 liked+0.024
 dolls+0.021
Neg logit contributions
ister-0.084
ww-0.083
www-0.081
ths-0.079
 service-0.078
Token hello
Token ID23748
Feature activation-2.11e+00

Pos logit contributions
 like+0.124
 everywhere+0.082
 favorite+0.081
 liked+0.080
 dolls+0.073
Neg logit contributions
ths-0.175
ister-0.174
ww-0.172
www-0.169
 avail-0.169
Token,
Token ID11
Feature activation-4.11e+00

Pos logit contributions
 like+0.029
 everywhere+0.020
 liked+0.019
 favorite+0.018
 dolls+0.018
Neg logit contributions
ister-0.067
ww-0.066
www-0.063
 service-0.062
ths-0.061
Token.
Token ID13
Feature activation-8.52e-01

Pos logit contributions
 like+0.010
 liked+0.007
 favorite+0.006
 everywhere+0.005
 ti+0.005
Neg logit contributions
ister-0.020
 service-0.020
ww-0.020
ths-0.020
 beast-0.019
Token
Token ID198
Feature activation-8.50e-01

Pos logit contributions
 like+0.016
 everywhere+0.011
 liked+0.011
 favorite+0.011
 back+0.010
Neg logit contributions
ister-0.022
ww-0.022
ths-0.022
www-0.022
 storm-0.021
Token
Token ID198
Feature activation+1.72e-01

Pos logit contributions
 like+0.014
 favorite+0.009
 everywhere+0.009
 liked+0.008
aces+0.008
Neg logit contributions
ister-0.026
ww-0.026
www-0.025
ths-0.025
 service-0.024
Token"
Token ID1
Feature activation-3.76e+00

Pos logit contributions
ister+0.004
ww+0.004
ths+0.004
www+0.004
 service+0.004
Neg logit contributions
 like-0.004
 everywhere-0.003
 favorite-0.003
 liked-0.003
 around-0.003
TokenBang
Token ID43984
Feature activation-2.04e-01

Pos logit contributions
 like+0.099
 everywhere+0.081
 favorite+0.079
 where+0.075
 around+0.074
Neg logit contributions
ww-0.083
www-0.078
ths-0.075
ister-0.072
 avail-0.072
Token,
Token ID11
Feature activation-7.61e+00

Pos logit contributions
 like+0.004
 liked+0.003
 favorite+0.003
 everywhere+0.003
 dolls+0.003
Neg logit contributions
ister-0.006
ww-0.006
www-0.006
ths-0.005
 beast-0.005
Token bang
Token ID20188
Feature activation-1.32e+00

Pos logit contributions
 like+0.196
 everywhere+0.147
 favorite+0.146
 liked+0.145
 around+0.135
Neg logit contributions
ister-0.169
ww-0.167
ths-0.164
www-0.162
 service-0.158
Token,
Token ID11
Feature activation-6.27e+00

Pos logit contributions
 like+0.024
 favorite+0.016
 everywhere+0.016
 liked+0.016
 dolls+0.014
Neg logit contributions
ister-0.039
ww-0.039
ths-0.038
www-0.038
 service-0.037
Token you
Token ID345
Feature activation-3.19e+00

Pos logit contributions
 like+0.157
 everywhere+0.117
 favorite+0.116
 liked+0.116
 around+0.108
Neg logit contributions
ister-0.142
ww-0.142
ths-0.139
www-0.137
 service-0.133
Token're
Token ID821
Feature activation+5.75e-01

Pos logit contributions
 like+0.061
 everywhere+0.041
 favorite+0.040
 liked+0.040
 around+0.036
Neg logit contributions
ister-0.090
ww-0.090
ths-0.088
www-0.088
 service-0.086
Token dead
Token ID2636
Feature activation-1.55e+00

Pos logit contributions
ww+0.014
ister+0.014
ths+0.014
www+0.014
 storm+0.014
Neg logit contributions
 like-0.013
 liked-0.009
 favorite-0.009
 everywhere-0.009
 pad-0.008
Token him
Token ID683
Feature activation-1.42e+00

Pos logit contributions
 like+0.018
 liked+0.012
 everywhere+0.012
 favorite+0.012
 pad+0.011
Neg logit contributions
ww-0.028
ister-0.028
ths-0.028
www-0.027
 service-0.027
Token a
Token ID257
Feature activation-6.76e-01

Pos logit contributions
 like+0.021
 liked+0.013
 everywhere+0.013
 favorite+0.012
 pad+0.011
Neg logit contributions
ister-0.043
ww-0.042
 service-0.042
ths-0.042
www-0.041
Token coat
Token ID13209
Feature activation-1.22e+00

Pos logit contributions
 like+0.015
 favorite+0.011
 everywhere+0.011
 liked+0.011
 back+0.010
Neg logit contributions
ths-0.017
ister-0.017
ww-0.017
www-0.017
 avail-0.016
Token.
Token ID13
Feature activation+2.27e-01

Pos logit contributions
 like+0.017
 everywhere+0.009
 favorite+0.009
 liked+0.009
 dolls+0.007
Neg logit contributions
ister-0.041
ww-0.041
ths-0.040
www-0.040
 service-0.040
Token He
Token ID679
Feature activation-2.63e-01

Pos logit contributions
ister+0.006
ww+0.006
ths+0.006
www+0.006
 service+0.006
Neg logit contributions
 like-0.005
 favorite-0.003
 liked-0.003
 everywhere-0.003
 both-0.003
Token must
Token ID1276
Feature activation+4.90e+00

Pos logit contributions
 like+0.005
 liked+0.003
 favorite+0.003
 everywhere+0.003
 both+0.003
Neg logit contributions
ister-0.008
ww-0.008
ths-0.008
www-0.008
 avail-0.007
Token be
Token ID307
Feature activation+5.91e-01

Pos logit contributions
ww+0.135
ths+0.134
ister+0.129
www+0.128
 sermon+0.124
Neg logit contributions
 like-0.110
 liked-0.072
 favorite-0.071
 both-0.070
 everywhere-0.069
Token cold
Token ID4692
Feature activation-2.01e+00

Pos logit contributions
ww+0.014
ths+0.014
ister+0.013
www+0.013
 service+0.013
Neg logit contributions
 like-0.013
 everywhere-0.009
 favorite-0.009
 liked-0.009
 wherever-0.009
Token."
Token ID526
Feature activation+5.57e-01

Pos logit contributions
 like+0.024
 everywhere+0.011
 favorite+0.010
 liked+0.010
 around+0.008
Neg logit contributions
ister-0.073
ww-0.072
ths-0.072
www-0.071
 service-0.070
Token
Token ID198
Feature activation+4.51e-01

Pos logit contributions
ister+0.018
ww+0.018
ths+0.017
www+0.017
 service+0.017
Neg logit contributions
 like-0.009
 everywhere-0.005
 favorite-0.005
 liked-0.005
 pad-0.004
Token
Token ID198
Feature activation+1.60e+00

Pos logit contributions
ister+0.015
ww+0.014
www+0.014
ths+0.014
 service+0.014
Neg logit contributions
 like-0.007
 everywhere-0.004
 favorite-0.004
 liked-0.004
 pad-0.004
Token were
Token ID547
Feature activation-3.00e-01

Pos logit contributions
ister+0.017
ths+0.017
ww+0.017
www+0.017
 service+0.017
Neg logit contributions
 like-0.012
 favorite-0.008
 everywhere-0.008
 liked-0.007
 wherever-0.007
Token so
Token ID523
Feature activation+2.00e+00

Pos logit contributions
 like+0.005
 everywhere+0.004
 liked+0.003
 favorite+0.003
 both+0.003
Neg logit contributions
ww-0.009
ister-0.009
www-0.009
ths-0.009
 avail-0.009
Token excited
Token ID6568
Feature activation+2.72e+00

Pos logit contributions
 avail+0.070
ww+0.066
ths+0.066
 beast+0.066
www+0.065
Neg logit contributions
 like-0.031
 liked-0.019
 both-0.017
 everywhere-0.017
 favorite-0.016
Token that
Token ID326
Feature activation+2.13e+00

Pos logit contributions
ister+0.094
ww+0.093
ths+0.092
www+0.091
 service+0.090
Neg logit contributions
 like-0.038
 everywhere-0.020
 favorite-0.019
 liked-0.019
 around-0.016
Token the
Token ID262
Feature activation+1.17e+00

Pos logit contributions
ister+0.050
ww+0.050
ths+0.049
www+0.048
 service+0.047
Neg logit contributions
 like-0.053
 everywhere-0.039
 favorite-0.038
 liked-0.038
 around-0.035
Token m
Token ID285
Feature activation+8.15e+00

Pos logit contributions
ww+0.029
ister+0.029
ths+0.029
www+0.028
 avail+0.028
Neg logit contributions
 like-0.027
 favorite-0.020
 liked-0.019
 everywhere-0.019
 back-0.019
Tokenule
Token ID2261
Feature activation+3.56e+00

Pos logit contributions
 avail+0.124
ww+0.104
 storm+0.100
www+0.089
 sermon+0.084
Neg logit contributions
 like-0.257
 favorite-0.221
 liked-0.216
aces-0.215
 both-0.213
Token was
Token ID373
Feature activation+1.73e+00

Pos logit contributions
ister+0.101
ww+0.100
ths+0.099
www+0.098
 service+0.096
Neg logit contributions
 like-0.069
 everywhere-0.046
 favorite-0.046
 liked-0.045
 dolls-0.041
Token helping
Token ID5742
Feature activation+1.92e+00

Pos logit contributions
ister+0.044
ths+0.043
ww+0.043
 service+0.042
www+0.042
Neg logit contributions
 like-0.037
 favorite-0.027
 liked-0.026
 everywhere-0.026
 pad-0.025
Token them
Token ID606
Feature activation-4.48e-01

Pos logit contributions
ister+0.060
ths+0.059
 service+0.058
ww+0.058
www+0.057
Neg logit contributions
 like-0.029
 favorite-0.019
 liked-0.018
 everywhere-0.017
 pad-0.016
Token.
Token ID13
Feature activation+4.99e-01

Pos logit contributions
 like+0.006
 favorite+0.003
 everywhere+0.003
 liked+0.003
 pad+0.003
Neg logit contributions
ister-0.015
ths-0.015
ww-0.015
www-0.015
 service-0.014
Token."
Token ID526
Feature activation+9.00e-01

Pos logit contributions
ths+0.001
ww+0.001
ister+0.001
www+0.001
 avail+0.001
Neg logit contributions
 like-0.001
 everywhere-0.000
 favorite-0.000
 liked-0.000
 around-0.000
Token
Token ID220
Feature activation+1.26e-01

Pos logit contributions
ww+0.032
ister+0.031
ths+0.031
www+0.030
 avail+0.030
Neg logit contributions
 like-0.013
 liked-0.007
 favorite-0.006
 both-0.006
 everywhere-0.006
Token
Token ID198
Feature activation+3.84e-01

Pos logit contributions
ister+0.004
ww+0.004
ths+0.004
www+0.004
 service+0.004
Neg logit contributions
 like-0.002
 favorite-0.001
 liked-0.001
 everywhere-0.001
 dolls-0.001
Token
Token ID198
Feature activation+1.26e+00

Pos logit contributions
ww+0.015
ths+0.014
 avail+0.014
www+0.014
 storm+0.014
Neg logit contributions
 like-0.004
 liked-0.001
 both-0.001
 everywhere-0.001
 back-0.001
TokenThe
Token ID464
Feature activation+1.93e+00

Pos logit contributions
 avail+0.037
ww+0.036
 storm+0.036
asis+0.035
ths+0.034
Neg logit contributions
 like-0.024
 liked-0.017
 both-0.015
 back-0.014
 but-0.014
Token elderly
Token ID13830
Feature activation+7.29e+00

Pos logit contributions
ww+0.054
ths+0.054
mmm+0.053
www+0.052
 avail+0.051
Neg logit contributions
 like-0.039
 back-0.032
 favorite-0.031
 dolls-0.030
 all-0.030
Token person
Token ID1048
Feature activation+3.82e+00

Pos logit contributions
ister+0.135
ww+0.130
ths+0.127
www+0.126
 service+0.124
Neg logit contributions
 like-0.217
 everywhere-0.172
 favorite-0.169
 liked-0.166
 wherever-0.160
Token replied
Token ID8712
Feature activation-1.67e+00

Pos logit contributions
ister+0.094
ww+0.094
ths+0.091
www+0.091
 avail+0.089
Neg logit contributions
 like-0.089
 liked-0.065
 everywhere-0.064
 favorite-0.064
 around-0.060
Token,
Token ID11
Feature activation-3.31e+00

Pos logit contributions
 like+0.032
 pad+0.022
 favorite+0.022
 everywhere+0.022
 liked+0.021
Neg logit contributions
ister-0.046
ths-0.044
ww-0.044
www-0.043
 service-0.043
Token "
Token ID366
Feature activation-3.91e+00

Pos logit contributions
 like+0.043
 favorite+0.022
 everywhere+0.022
 liked+0.019
 dolls+0.017
Neg logit contributions
ister-0.113
ths-0.111
ww-0.111
www-0.110
 service-0.108
TokenNo
Token ID2949
Feature activation+4.38e-01

Pos logit contributions
 like+0.085
 everywhere+0.059
 favorite+0.059
 liked+0.059
 around+0.054
Neg logit contributions
ister-0.102
ww-0.102
ths-0.100
www-0.099
 service-0.097
Token
Token ID198
Feature activation+3.76e-01

Pos logit contributions
 like+0.017
 favorite+0.010
 everywhere+0.010
 liked+0.010
 pad+0.008
Neg logit contributions
ister-0.037
ww-0.037
ths-0.036
www-0.036
 service-0.035
TokenThey
Token ID2990
Feature activation-2.07e+00

Pos logit contributions
ister+0.010
ww+0.010
ths+0.010
www+0.009
 service+0.009
Neg logit contributions
 like-0.008
 everywhere-0.006
 liked-0.006
 favorite-0.006
 both-0.005
Token played
Token ID2826
Feature activation-3.54e-01

Pos logit contributions
 like+0.045
aces+0.035
 hers+0.035
 favorite+0.034
 wherever+0.033
Neg logit contributions
ths-0.045
ister-0.044
 service-0.044
field-0.044
 storm-0.044
Token in
Token ID287
Feature activation-7.07e-01

Pos logit contributions
 like+0.005
 pad+0.004
 favorite+0.003
 liked+0.003
 everywhere+0.003
Neg logit contributions
 site-0.010
ister-0.010
 service-0.010
asis-0.010
ths-0.010
Token the
Token ID262
Feature activation-4.73e-01

Pos logit contributions
 like+0.014
aces+0.012
 pad+0.011
 favorite+0.011
 liked+0.011
Neg logit contributions
 service-0.017
 storm-0.016
ister-0.016
 site-0.016
ths-0.016
Token o
Token ID267
Feature activation+5.14e+00

Pos logit contributions
 like+0.011
 favorite+0.008
 everywhere+0.008
 liked+0.008
 back+0.008
Neg logit contributions
ister-0.012
ths-0.012
ww-0.011
www-0.011
 avail-0.011
Tokenasis
Token ID17765
Feature activation+6.31e-01

Pos logit contributions
 avail+0.077
www+0.060
ww+0.058
ths+0.058
 beast+0.055
Neg logit contributions
 like-0.180
 liked-0.150
 until-0.147
 favorite-0.147
 back-0.144
Token for
Token ID329
Feature activation+7.08e-01

Pos logit contributions
ister+0.023
ww+0.023
ths+0.022
www+0.022
 service+0.022
Neg logit contributions
 like-0.007
 everywhere-0.003
 favorite-0.003
 liked-0.003
 dolls-0.002
Token a
Token ID257
Feature activation-7.99e-01

Pos logit contributions
ister+0.027
ww+0.027
ths+0.027
www+0.026
 avail+0.026
Neg logit contributions
 like-0.008
 everywhere-0.003
 liked-0.003
 favorite-0.003
 both-0.002
Token while
Token ID981
Feature activation+2.38e+00

Pos logit contributions
 like+0.016
 everywhere+0.011
 favorite+0.011
 liked+0.011
aces+0.010
Neg logit contributions
ww-0.022
ister-0.022
www-0.021
ths-0.021
 service-0.021
Token.
Token ID13
Feature activation-7.17e-01

Pos logit contributions
ister+0.085
ww+0.084
ths+0.084
www+0.083
 service+0.081
Neg logit contributions
 like-0.029
 everywhere-0.014
 favorite-0.013
 liked-0.013
 around-0.010
Token down
Token ID866
Feature activation-4.50e-01

Pos logit contributions
ister+0.031
ths+0.030
 avail+0.029
 service+0.029
ww+0.029
Neg logit contributions
 like-0.017
 favorite-0.010
 pad-0.010
 liked-0.010
 everywhere-0.010
Token for
Token ID329
Feature activation+1.25e+00

Pos logit contributions
 like+0.006
 favorite+0.003
 liked+0.003
 everywhere+0.003
 dolls+0.003
Neg logit contributions
ister-0.015
ths-0.015
ww-0.015
 service-0.015
www-0.014
Token his
Token ID465
Feature activation+1.38e-01

Pos logit contributions
ister+0.045
ww+0.044
www+0.044
ths+0.043
ule+0.041
Neg logit contributions
 like-0.017
 everywhere-0.009
 all-0.008
 favorite-0.008
 liked-0.008
Token nap
Token ID25422
Feature activation+2.81e+00

Pos logit contributions
ister+0.003
ww+0.003
www+0.003
ths+0.003
 avail+0.003
Neg logit contributions
 like-0.003
 favorite-0.002
 everywhere-0.002
 liked-0.002
 dolls-0.002
Token.
Token ID13
Feature activation+1.17e+00

Pos logit contributions
ister+0.097
ww+0.096
ths+0.095
www+0.094
 service+0.093
Neg logit contributions
 like-0.037
 everywhere-0.019
 favorite-0.018
 liked-0.018
 dolls-0.014
Token While
Token ID2893
Feature activation+4.93e+00

Pos logit contributions
ister+0.037
ww+0.036
ths+0.036
www+0.035
 avail+0.034
Neg logit contributions
 like-0.021
 liked-0.013
 favorite-0.013
 everywhere-0.013
 wherever-0.012
Token he
Token ID339
Feature activation+1.32e+00

Pos logit contributions
ister+0.150
ww+0.149
www+0.147
ths+0.143
 avail+0.141
Neg logit contributions
 like-0.094
 everywhere-0.064
 all-0.061
 liked-0.060
 wherever-0.058
Token was
Token ID373
Feature activation+1.64e+00

Pos logit contributions
ister+0.043
ww+0.042
 avail+0.042
www+0.042
irens+0.042
Neg logit contributions
 like-0.022
 liked-0.016
 everywhere-0.013
 found-0.013
 favorite-0.012
Token lying
Token ID9105
Feature activation-1.46e-01

Pos logit contributions
ister+0.043
ww+0.043
ths+0.042
www+0.042
 avail+0.040
Neg logit contributions
 like-0.037
 everywhere-0.026
 liked-0.025
 favorite-0.025
 around-0.024
Token down
Token ID866
Feature activation+9.69e-01

Pos logit contributions
 like+0.002
 favorite+0.001
 everywhere+0.001
 liked+0.001
 dolls+0.001
Neg logit contributions
ister-0.005
ww-0.005
ths-0.005
www-0.005
 service-0.004
Token he
Token ID339
Feature activation+9.52e-01

Pos logit contributions
ww+0.033
ister+0.032
www+0.032
ths+0.032
 avail+0.031
Neg logit contributions
 like-0.014
 everywhere-0.008
 liked-0.008
 favorite-0.007
 back-0.007
Token truck
Token ID7779
Feature activation+1.26e+00

Pos logit contributions
 like+0.033
 everywhere+0.024
 favorite+0.024
 liked+0.024
 wherever+0.023
Neg logit contributions
ister-0.028
ww-0.028
ths-0.027
www-0.027
 service-0.027
Token and
Token ID290
Feature activation-1.23e+00

Pos logit contributions
ister+0.045
 service+0.043
ths+0.043
ww+0.043
www+0.042
Neg logit contributions
 like-0.015
 favorite-0.007
 everywhere-0.007
 liked-0.007
 dolls-0.005
Token asks
Token ID7893
Feature activation+5.90e-01

Pos logit contributions
 like+0.032
 everywhere+0.024
 favorite+0.024
 liked+0.023
 wherever+0.022
Neg logit contributions
ister-0.026
ww-0.026
ths-0.026
www-0.026
 service-0.025
Token for
Token ID329
Feature activation+1.50e+00

Pos logit contributions
ister+0.014
ama+0.013
 service+0.013
oo+0.012
 avail+0.012
Neg logit contributions
 like-0.011
 pad-0.010
aces-0.010
 favorite-0.009
 pads-0.009
Token mint
Token ID21043
Feature activation-6.87e-01

Pos logit contributions
ister+0.045
ww+0.045
ths+0.044
 service+0.044
www+0.044
Neg logit contributions
 like-0.026
 everywhere-0.017
 liked-0.017
 favorite-0.016
 pad-0.015
Token ice
Token ID4771
Feature activation+6.83e-01

Pos logit contributions
 like+0.015
 liked+0.011
 everywhere+0.011
 pad+0.011
 favorite+0.010
Neg logit contributions
ister-0.017
ww-0.017
www-0.016
ths-0.016
 service-0.016
Token cream
Token ID8566
Feature activation-2.41e+00

Pos logit contributions
ister+0.012
ww+0.012
www+0.011
 service+0.011
ths+0.011
Neg logit contributions
 like-0.020
 everywhere-0.016
 liked-0.016
 pad-0.016
 wherever-0.016
Token.
Token ID13
Feature activation-7.94e-01

Pos logit contributions
 like+0.031
 liked+0.020
 favorite+0.020
 everywhere+0.019
 dolls+0.018
Neg logit contributions
ister-0.078
 service-0.075
ww-0.075
www-0.074
ths-0.072
Token The
Token ID383
Feature activation+1.21e+00

Pos logit contributions
 like+0.016
 everywhere+0.011
 favorite+0.010
 liked+0.010
 wherever+0.009
Neg logit contributions
ister-0.022
ww-0.022
ths-0.022
www-0.021
 service-0.021
Token ice
Token ID4771
Feature activation+3.40e+00

Pos logit contributions
ister+0.028
ww+0.028
ths+0.028
www+0.028
 avail+0.027
Neg logit contributions
 like-0.030
 favorite-0.022
 liked-0.022
 everywhere-0.022
 dolls-0.021
Token cream
Token ID8566
Feature activation+7.53e-01

Pos logit contributions
ister+0.066
ww+0.065
ths+0.064
www+0.063
 service+0.061
Neg logit contributions
 like-0.097
 everywhere-0.075
 favorite-0.074
 liked-0.074
 wherever-0.070
Token upon
Token ID2402
Feature activation-2.63e+00

Pos logit contributions
 like+0.016
aces+0.011
 ti+0.010
 pad+0.009
 liked+0.009
Neg logit contributions
ister-0.029
ama-0.028
 storm-0.027
ww-0.026
ule-0.026
Token a
Token ID257
Feature activation-1.64e+00

Pos logit contributions
 like+0.042
 pad+0.032
 everywhere+0.032
aces+0.030
 favorite+0.030
Neg logit contributions
ww-0.069
ister-0.068
 sermon-0.067
body-0.066
 storm-0.066
Token time
Token ID640
Feature activation+4.37e-01

Pos logit contributions
 like+0.043
aces+0.035
 dolls+0.035
 everywhere+0.035
 wherever+0.034
Neg logit contributions
ww-0.033
www-0.031
ister-0.030
 storm-0.029
ths-0.028
Token,
Token ID11
Feature activation-1.43e+00

Pos logit contributions
ister+0.010
ww+0.010
ama+0.009
 storm+0.009
ths+0.009
Neg logit contributions
 like-0.010
aces-0.008
 pad-0.008
 favorite-0.008
 pads-0.008
Token there
Token ID612
Feature activation-5.97e-01

Pos logit contributions
 like+0.029
aces+0.025
 pad+0.023
 pads+0.023
 skirts+0.022
Neg logit contributions
ww-0.032
body-0.031
 storm-0.030
ister-0.030
ths-0.030
Token was
Token ID373
Feature activation+1.01e+00

Pos logit contributions
 like+0.015
aces+0.011
 pad+0.011
 favorite+0.011
 everywhere+0.011
Neg logit contributions
ister-0.013
ww-0.013
ths-0.012
www-0.012
 storm-0.012
Token a
Token ID257
Feature activation-7.09e-01

Pos logit contributions
ww+0.028
www+0.026
 storm+0.026
ister+0.026
body+0.026
Neg logit contributions
 like-0.016
aces-0.011
 everywhere-0.011
 pad-0.011
 liked-0.011
Token big
Token ID1263
Feature activation+3.17e-01

Pos logit contributions
 like+0.014
 liked+0.010
 favorite+0.010
 everywhere+0.010
 dolls+0.009
Neg logit contributions
ister-0.019
ww-0.019
ths-0.019
www-0.019
 service-0.019
Token elephant
Token ID20950
Feature activation+6.50e-01

Pos logit contributions
ths+0.009
ister+0.009
ww+0.009
www+0.008
 avail+0.008
Neg logit contributions
 like-0.007
 liked-0.005
 everywhere-0.004
 favorite-0.004
 back-0.004
Token.
Token ID13
Feature activation-1.14e-01

Pos logit contributions
ister+0.021
ww+0.021
ths+0.020
www+0.020
 avail+0.020
Neg logit contributions
 like-0.010
 everywhere-0.006
 liked-0.006
 favorite-0.006
 around-0.005
Token He
Token ID679
Feature activation+2.92e-01

Pos logit contributions
 like+0.002
 everywhere+0.002
 favorite+0.002
 liked+0.001
 pad+0.001
Neg logit contributions
ister-0.003
ww-0.003
www-0.003
ths-0.003
 service-0.003
Token very
Token ID845
Feature activation+2.28e+00

Pos logit contributions
ww+0.028
ister+0.028
ths+0.027
www+0.027
 avail+0.027
Neg logit contributions
 like-0.019
 everywhere-0.012
 liked-0.012
 favorite-0.012
 around-0.011
Token excited
Token ID6568
Feature activation+1.24e+00

Pos logit contributions
ths+0.070
ww+0.069
 avail+0.068
ister+0.068
irens+0.066
Neg logit contributions
 like-0.045
 liked-0.029
 favorite-0.027
 around-0.027
 both-0.025
Token.
Token ID13
Feature activation+8.71e-01

Pos logit contributions
ister+0.044
ww+0.044
ths+0.043
www+0.043
 avail+0.042
Neg logit contributions
 like-0.016
 everywhere-0.007
 favorite-0.007
 liked-0.007
 around-0.006
Token She
Token ID1375
Feature activation+7.02e-01

Pos logit contributions
ister+0.024
ww+0.024
www+0.023
ths+0.023
 service+0.023
Neg logit contributions
 like-0.017
 everywhere-0.012
 favorite-0.012
 liked-0.011
 dolls-0.011
Token wanted
Token ID2227
Feature activation+9.68e-01

Pos logit contributions
ister+0.021
ww+0.020
 avail+0.020
ths+0.020
 service+0.020
Neg logit contributions
 like-0.014
 liked-0.010
 everywhere-0.009
 favorite-0.009
 both-0.008
Token to
Token ID284
Feature activation+2.50e+00

Pos logit contributions
ister+0.036
ww+0.036
ths+0.036
www+0.035
 avail+0.034
Neg logit contributions
 like-0.011
 everywhere-0.005
 favorite-0.005
 liked-0.004
 back-0.004
Token see
Token ID766
Feature activation-1.08e+00

Pos logit contributions
ww+0.075
ths+0.074
www+0.070
ister+0.070
irens+0.070
Neg logit contributions
 like-0.054
 both-0.034
 all-0.034
 love-0.033
 liked-0.032
Token the
Token ID262
Feature activation-1.09e+00

Pos logit contributions
 like+0.019
 liked+0.013
 favorite+0.012
 pad+0.012
aces+0.012
Neg logit contributions
ister-0.031
ths-0.030
 service-0.030
 avail-0.029
ww-0.029
Token mountain
Token ID8598
Feature activation+2.27e+00

Pos logit contributions
 like+0.028
 everywhere+0.021
 favorite+0.021
 liked+0.020
 dolls+0.019
Neg logit contributions
ister-0.025
ww-0.024
ths-0.024
www-0.024
 avail-0.023
Token up
Token ID510
Feature activation+1.06e+00

Pos logit contributions
ww+0.085
www+0.082
ister+0.081
ths+0.079
 avail+0.079
Neg logit contributions
 like-0.032
 everywhere-0.015
 around-0.014
 liked-0.013
 all-0.013
Token close
Token ID1969
Feature activation-5.93e-01

Pos logit contributions
ister+0.032
ww+0.032
ths+0.032
www+0.031
 service+0.031
Neg logit contributions
 like-0.018
 everywhere-0.012
 favorite-0.012
 liked-0.011
 dolls-0.010
Token.
Token ID13
Feature activation-9.21e-01

Pos logit contributions
 like+0.004
 favorite+0.003
 pad+0.002
 ti+0.002
 liked+0.002
Neg logit contributions
ister-0.008
 service-0.008
ths-0.008
 site-0.008
 storm-0.008
Token Mom
Token ID11254
Feature activation+1.72e+00

Pos logit contributions
 like+0.018
 everywhere+0.013
 favorite+0.012
 liked+0.012
 around+0.011
Neg logit contributions
ww-0.025
ister-0.025
www-0.024
ths-0.024
 service-0.024
Token smiled
Token ID13541
Feature activation-1.99e+00

Pos logit contributions
ister+0.046
ths+0.045
ww+0.044
 service+0.043
www+0.043
Neg logit contributions
 like-0.035
 favorite-0.025
 everywhere-0.025
 wherever-0.024
 pad-0.023
Token and
Token ID290
Feature activation-2.78e-01

Pos logit contributions
aces+0.029
 like+0.029
 dolls+0.026
 Whe+0.026
 pad+0.026
Neg logit contributions
ister-0.049
 storm-0.048
ths-0.047
ww-0.047
 service-0.046
Token went
Token ID1816
Feature activation+1.27e+00

Pos logit contributions
 like+0.006
 everywhere+0.004
 liked+0.004
 favorite+0.004
 around+0.003
Neg logit contributions
ister-0.008
ww-0.008
ths-0.008
www-0.008
 service-0.007
Token back
Token ID736
Feature activation+1.89e+00

Pos logit contributions
ister+0.042
ths+0.041
ww+0.041
 service+0.041
www+0.041
Neg logit contributions
 like-0.018
 liked-0.010
 favorite-0.010
 everywhere-0.009
 pad-0.009
Token to
Token ID284
Feature activation+1.51e+00

Pos logit contributions
ister+0.057
 service+0.057
ths+0.056
ww+0.055
www+0.054
Neg logit contributions
 like-0.030
 liked-0.019
 favorite-0.018
 everywhere-0.018
aces-0.017
Token the
Token ID262
Feature activation-7.04e-01

Pos logit contributions
ister+0.045
ww+0.045
ths+0.044
www+0.044
 service+0.043
Neg logit contributions
 like-0.027
 everywhere-0.017
 favorite-0.017
 liked-0.017
 around-0.015
Token kitchen
Token ID9592
Feature activation-6.28e-01

Pos logit contributions
 like+0.017
aces+0.012
 liked+0.012
 everywhere+0.012
 favorite+0.012
Neg logit contributions
 service-0.016
ister-0.016
 storm-0.015
ths-0.015
ww-0.015
Token.
Token ID13
Feature activation-6.21e-01

Pos logit contributions
 like+0.009
 favorite+0.005
 everywhere+0.005
 liked+0.005
 wherever+0.004
Neg logit contributions
 service-0.020
ister-0.019
 site-0.019
ths-0.019
 sermon-0.018
Token She
Token ID1375
Feature activation+3.74e-01

Pos logit contributions
 like+0.014
 everywhere+0.010
 favorite+0.010
aces+0.010
 pad+0.010
Neg logit contributions
ww-0.014
www-0.014
ister-0.014
 storm-0.014
irens-0.013
Token to
Token ID284
Feature activation+1.79e+00

Pos logit contributions
ister+0.029
ww+0.029
ths+0.028
www+0.028
 service+0.028
Neg logit contributions
 like-0.011
 everywhere-0.006
 favorite-0.005
 liked-0.005
 dolls-0.004
Token be
Token ID307
Feature activation+1.42e-01

Pos logit contributions
ww+0.056
ths+0.054
www+0.053
ister+0.052
izing+0.051
Neg logit contributions
 like-0.037
 both-0.024
 all-0.023
 back-0.022
 love-0.021
Token saved
Token ID7448
Feature activation+1.23e+00

Pos logit contributions
ister+0.004
ww+0.004
ths+0.004
www+0.004
 service+0.004
Neg logit contributions
 like-0.003
 everywhere-0.002
 favorite-0.002
 liked-0.002
 dolls-0.002
Token.
Token ID13
Feature activation+1.13e+00

Pos logit contributions
ww+0.038
ister+0.038
ths+0.038
www+0.037
 service+0.037
Neg logit contributions
 like-0.021
 everywhere-0.013
 favorite-0.012
 liked-0.012
 around-0.011
Token But
Token ID887
Feature activation+1.94e+00

Pos logit contributions
ister+0.036
ww+0.036
ths+0.036
www+0.035
 service+0.035
Neg logit contributions
 like-0.017
 everywhere-0.010
 favorite-0.010
 liked-0.010
 around-0.008
Token no
Token ID645
Feature activation+1.65e+00

Pos logit contributions
ww+0.061
ister+0.061
www+0.060
iren+0.059
 sermon+0.059
Neg logit contributions
 like-0.035
 everywhere-0.021
 all-0.020
 both-0.020
 liked-0.020
Token one
Token ID530
Feature activation+2.19e+00

Pos logit contributions
ister+0.042
irens+0.040
iren+0.039
 sermon+0.038
asis+0.038
Neg logit contributions
 like-0.043
 but-0.029
 favorite-0.029
 love-0.029
 liked-0.029
Token came
Token ID1625
Feature activation-7.30e-01

Pos logit contributions
ister+0.065
ww+0.064
 avail+0.064
ths+0.063
www+0.062
Neg logit contributions
 like-0.042
 liked-0.029
 everywhere-0.027
 favorite-0.026
 around-0.025
Token to
Token ID284
Feature activation+1.44e-01

Pos logit contributions
 like+0.011
 everywhere+0.006
 favorite+0.006
 liked+0.006
 back+0.005
Neg logit contributions
ww-0.024
ister-0.024
ths-0.024
www-0.024
 service-0.023
Token help
Token ID1037
Feature activation+1.26e+00

Pos logit contributions
ww+0.005
www+0.005
ths+0.005
ister+0.005
 sermon+0.004
Neg logit contributions
 like-0.003
 liked-0.001
 everywhere-0.001
 love-0.001
 favorite-0.001
Token.
Token ID13
Feature activation+6.31e-01

Pos logit contributions
ww+0.043
ister+0.043
ths+0.043
www+0.042
 service+0.041
Neg logit contributions
 like-0.018
 everywhere-0.009
 liked-0.009
 favorite-0.008
 around-0.007
TokenTom
Token ID13787
Feature activation-2.23e-01

Pos logit contributions
ister+0.012
ww+0.012
ths+0.012
www+0.012
 service+0.012
Neg logit contributions
 like-0.010
 favorite-0.007
 everywhere-0.007
 liked-0.007
 pad-0.006
Token and
Token ID290
Feature activation-8.07e-01

Pos logit contributions
 like+0.003
 everywhere+0.002
 favorite+0.002
 wherever+0.002
 hers+0.001
Neg logit contributions
ister-0.007
ww-0.007
ths-0.007
www-0.007
 beast-0.007
Token Lily
Token ID20037
Feature activation-1.93e+00

Pos logit contributions
 like+0.009
 everywhere+0.004
 favorite+0.004
 liked+0.004
 pad+0.003
Neg logit contributions
ister-0.029
ww-0.029
ths-0.029
www-0.029
 service-0.028
Token nod
Token ID18666
Feature activation-1.50e+00

Pos logit contributions
 like+0.039
 favorite+0.030
 everywhere+0.029
 dolls+0.028
 pad+0.028
Neg logit contributions
ister-0.048
 service-0.046
ths-0.045
 storm-0.045
ww-0.045
Token.
Token ID13
Feature activation-1.22e+00

Pos logit contributions
 like+0.021
aces+0.015
 favorite+0.015
 pad+0.014
 everywhere+0.014
Neg logit contributions
ister-0.043
 storm-0.043
 service-0.042
ww-0.041
 avail-0.041
Token They
Token ID1119
Feature activation-1.82e+00

Pos logit contributions
 like+0.023
 everywhere+0.015
 favorite+0.015
 liked+0.015
 around+0.013
Neg logit contributions
ister-0.036
ww-0.036
ths-0.035
www-0.035
 service-0.034
Token are
Token ID389
Feature activation-6.78e-01

Pos logit contributions
 like+0.039
 everywhere+0.035
 dolls+0.035
aces+0.035
 pad+0.035
Neg logit contributions
ister-0.036
 storm-0.034
 service-0.033
www-0.031
ww-0.031
Token OK
Token ID7477
Feature activation+7.54e-01

Pos logit contributions
 like+0.014
 pad+0.011
aces+0.010
 wherever+0.010
 everywhere+0.010
Neg logit contributions
ister-0.017
 storm-0.016
ww-0.016
 service-0.015
ths-0.015
Token.
Token ID13
Feature activation-8.08e-01

Pos logit contributions
ister+0.026
ww+0.025
ths+0.025
 service+0.025
 storm+0.025
Neg logit contributions
 like-0.009
 favorite-0.005
 everywhere-0.005
 liked-0.005
 pad-0.004
Token They
Token ID1119
Feature activation-1.71e+00

Pos logit contributions
 like+0.015
 liked+0.009
 favorite+0.009
 everywhere+0.009
 both+0.008
Neg logit contributions
ister-0.025
ww-0.024
ths-0.024
www-0.024
 service-0.023
Token are
Token ID389
Feature activation-3.43e-01

Pos logit contributions
 like+0.035
 everywhere+0.032
 dolls+0.031
aces+0.031
 pad+0.031
Neg logit contributions
ister-0.035
 storm-0.034
 service-0.034
ww-0.031
www-0.031
Token could
Token ID714
Feature activation+2.89e+00

Pos logit contributions
ister+0.027
ww+0.027
 avail+0.026
irens+0.026
www+0.026
Neg logit contributions
 like-0.016
 liked-0.012
 everywhere-0.010
 both-0.010
 all-0.010
Token imagine
Token ID5967
Feature activation+1.02e+00

Pos logit contributions
ww+0.085
www+0.081
ths+0.080
ister+0.080
irens+0.077
Neg logit contributions
 like-0.064
 liked-0.041
 both-0.040
 favorite-0.040
 all-0.040
Token how
Token ID703
Feature activation-6.70e-01

Pos logit contributions
ister+0.030
ww+0.030
ths+0.030
www+0.029
 service+0.029
Neg logit contributions
 like-0.018
 everywhere-0.012
 favorite-0.011
 liked-0.011
 around-0.010
Token it
Token ID340
Feature activation+1.55e+00

Pos logit contributions
 like+0.012
 liked+0.008
 everywhere+0.008
 favorite+0.008
 around+0.007
Neg logit contributions
ister-0.020
ww-0.020
ths-0.020
www-0.020
 avail-0.019
Token would
Token ID561
Feature activation+1.53e+00

Pos logit contributions
ister+0.048
ww+0.048
www+0.046
ths+0.045
 avail+0.045
Neg logit contributions
 like-0.030
 liked-0.020
 everywhere-0.019
 favorite-0.018
 around-0.018
Token be
Token ID307
Feature activation-2.25e-01

Pos logit contributions
ww+0.038
ister+0.038
ths+0.037
www+0.037
 avail+0.035
Neg logit contributions
 like-0.038
 everywhere-0.027
 liked-0.026
 favorite-0.026
 back-0.025
Token to
Token ID284
Feature activation+8.78e-01

Pos logit contributions
 like+0.003
 favorite+0.002
 everywhere+0.002
 liked+0.002
 dolls+0.002
Neg logit contributions
ister-0.007
ww-0.007
ths-0.007
www-0.007
 service-0.007
Token be
Token ID307
Feature activation+1.88e-01

Pos logit contributions
ww+0.024
ths+0.023
www+0.023
ister+0.023
irens+0.022
Neg logit contributions
 like-0.022
 all-0.014
 love-0.014
 everywhere-0.014
 liked-0.014
Token as
Token ID355
Feature activation+1.78e+00

Pos logit contributions
ister+0.005
ww+0.005
ths+0.005
www+0.005
 service+0.005
Neg logit contributions
 like-0.003
 everywhere-0.002
 favorite-0.002
 liked-0.002
 dolls-0.002
Token big
Token ID1263
Feature activation-2.79e+00

Pos logit contributions
ths+0.053
ww+0.053
ister+0.053
www+0.052
 avail+0.051
Neg logit contributions
 like-0.036
 everywhere-0.023
 favorite-0.022
 liked-0.022
 all-0.021
Token as
Token ID355
Feature activation+2.88e+00

Pos logit contributions
 like+0.038
 everywhere+0.022
 liked+0.022
 favorite+0.022
 dolls+0.018
Neg logit contributions
ister-0.091
ww-0.091
ths-0.090
 service-0.089
www-0.089
Token
Token ID198
Feature activation+4.39e-01

Pos logit contributions
 like+0.013
 favorite+0.008
 everywhere+0.008
 liked+0.008
 pad+0.007
Neg logit contributions
ister-0.023
ww-0.022
www-0.022
ths-0.022
 service-0.022
Token"
Token ID1
Feature activation-3.67e+00

Pos logit contributions
ww+0.011
ister+0.010
www+0.010
ths+0.010
 service+0.009
Neg logit contributions
 like-0.011
 favorite-0.008
 everywhere-0.008
 liked-0.007
aces-0.007
TokenOh
Token ID5812
Feature activation-7.85e-01

Pos logit contributions
 like+0.086
 everywhere+0.064
 favorite+0.064
 liked+0.061
 around+0.059
Neg logit contributions
ww-0.091
ister-0.089
ths-0.087
www-0.087
 service-0.085
Token no
Token ID645
Feature activation+8.10e-01

Pos logit contributions
 like+0.013
 everywhere+0.008
 favorite+0.008
 liked+0.008
 around+0.008
Neg logit contributions
ww-0.025
ister-0.024
www-0.024
ths-0.024
 service-0.023
Token!"
Token ID2474
Feature activation-1.08e+00

Pos logit contributions
ister+0.026
ths+0.025
ww+0.025
www+0.025
 service+0.024
Neg logit contributions
 like-0.014
 everywhere-0.008
 favorite-0.008
 liked-0.008
 dolls-0.007
Token Lily
Token ID20037
Feature activation+2.05e-01

Pos logit contributions
 like+0.019
 everywhere+0.012
 favorite+0.012
 liked+0.012
 pad+0.011
Neg logit contributions
ister-0.032
ww-0.032
www-0.031
ths-0.031
 service-0.031
Token and
Token ID290
Feature activation-8.91e-01

Pos logit contributions
ister+0.006
ww+0.006
ths+0.006
www+0.006
 avail+0.006
Neg logit contributions
 like-0.004
 everywhere-0.002
 liked-0.002
 favorite-0.002
 around-0.002
Token Ben
Token ID3932
Feature activation-4.08e-01

Pos logit contributions
 like+0.012
 everywhere+0.006
 favorite+0.006
 liked+0.006
 around+0.005
Neg logit contributions
ister-0.031
ww-0.031
ths-0.031
www-0.030
 service-0.030
Token say
Token ID910
Feature activation-1.01e+00

Pos logit contributions
 like+0.011
 favorite+0.009
 everywhere+0.009
 liked+0.009
 wherever+0.008
Neg logit contributions
ister-0.008
ww-0.008
ths-0.008
www-0.008
 service-0.008
Token.
Token ID13
Feature activation-1.43e+00

Pos logit contributions
 like+0.013
 liked+0.008
 favorite+0.007
 everywhere+0.007
 pad+0.007
Neg logit contributions
ister-0.034
 service-0.033
ww-0.032
ths-0.032
www-0.032
Token They
Token ID1119
Feature activation-2.21e+00

Pos logit contributions
 like+0.028
aces+0.019
 favorite+0.019
 everywhere+0.019
 pad+0.018
Neg logit contributions
ister-0.037
ww-0.037
 service-0.036
www-0.036
ths-0.036
Token surprise
Token ID5975
Feature activation+1.18e+00

Pos logit contributions
 like+0.019
 everywhere+0.013
 liked+0.013
 favorite+0.013
 around+0.013
Neg logit contributions
ww-0.022
ister-0.022
www-0.021
ths-0.021
 service-0.021
Token for
Token ID329
Feature activation+1.69e+00

Pos logit contributions
ister+0.039
 service+0.038
ths+0.038
ww+0.038
www+0.037
Neg logit contributions
 like-0.016
 favorite-0.009
 liked-0.009
 everywhere-0.008
aces-0.008
Token Lily
Token ID20037
Feature activation-6.04e-01

Pos logit contributions
ister+0.056
ww+0.056
ths+0.055
www+0.055
 service+0.054
Neg logit contributions
 like-0.024
 everywhere-0.014
 favorite-0.013
 liked-0.013
 dolls-0.011
Token -
Token ID532
Feature activation-1.05e+00

Pos logit contributions
 like+0.009
 favorite+0.006
 liked+0.005
 everywhere+0.005
aces+0.005
Neg logit contributions
ister-0.018
ths-0.018
 service-0.018
ww-0.017
www-0.017
Token a
Token ID257
Feature activation-1.50e-01

Pos logit contributions
 like+0.017
aces+0.012
 everywhere+0.012
 favorite+0.012
 liked+0.011
Neg logit contributions
ister-0.028
ww-0.028
ths-0.028
www-0.027
 service-0.027
Token new
Token ID649
Feature activation-7.06e-01

Pos logit contributions
 like+0.003
 everywhere+0.002
 favorite+0.002
 liked+0.002
 around+0.002
Neg logit contributions
ister-0.004
ww-0.004
ths-0.004
www-0.004
 service-0.004
Token doll
Token ID3654
Feature activation-9.30e-01

Pos logit contributions
 like+0.016
 everywhere+0.012
aces+0.012
 wherever+0.011
 around+0.011
Neg logit contributions
ww-0.017
www-0.017
ister-0.017
 storm-0.016
 service-0.016
Token to
Token ID284
Feature activation+1.76e+00

Pos logit contributions
 like+0.017
 favorite+0.012
 everywhere+0.011
 liked+0.011
 dolls+0.010
Neg logit contributions
ister-0.027
ww-0.026
www-0.026
ths-0.026
 service-0.025
Token play
Token ID711
Feature activation-1.22e+00

Pos logit contributions
ister+0.052
ww+0.051
ths+0.051
www+0.050
 service+0.049
Neg logit contributions
 like-0.032
 everywhere-0.021
 favorite-0.021
 liked-0.021
 around-0.018
Token with
Token ID351
Feature activation-2.45e+00

Pos logit contributions
 like+0.017
 favorite+0.014
 pad+0.014
 everywhere+0.013
 liked+0.012
Neg logit contributions
ister-0.034
 site-0.032
ths-0.032
ww-0.032
 service-0.032
Token at
Token ID379
Feature activation+1.86e+00

Pos logit contributions
 like+0.035
 favorite+0.024
 pad+0.022
 liked+0.022
aces+0.022
Neg logit contributions
ister-0.073
ths-0.070
 service-0.070
ww-0.069
www-0.068
Token in
Token ID287
Feature activation-8.01e-01

Pos logit contributions
 like+0.010
 liked+0.009
 pad+0.008
 favorites+0.008
 favorite+0.008
Neg logit contributions
ister-0.016
 storm-0.015
ww-0.015
ths-0.015
 beast-0.014
Token.
Token ID13
Feature activation+5.00e-01

Pos logit contributions
 like+0.012
 favorite+0.009
 liked+0.009
aces+0.008
 pad+0.008
Neg logit contributions
ister-0.024
 service-0.023
 avail-0.023
 storm-0.023
ths-0.023
Token She
Token ID1375
Feature activation+5.72e-01

Pos logit contributions
ister+0.014
ww+0.014
ths+0.014
www+0.014
 service+0.014
Neg logit contributions
 like-0.010
 everywhere-0.006
 favorite-0.006
 liked-0.006
 dolls-0.005
Token soon
Token ID2582
Feature activation+1.41e+00

Pos logit contributions
ister+0.017
ww+0.017
 avail+0.016
ths+0.016
www+0.016
Neg logit contributions
 like-0.011
 liked-0.008
 favorite-0.008
 everywhere-0.007
 both-0.007
Token realized
Token ID6939
Feature activation+1.46e+00

Pos logit contributions
ww+0.037
ister+0.037
www+0.036
ths+0.036
 storm+0.035
Neg logit contributions
 like-0.029
 favorite-0.021
 everywhere-0.020
 liked-0.019
 dolls-0.019
Token she
Token ID673
Feature activation-7.88e-01

Pos logit contributions
ww+0.038
ister+0.037
www+0.037
ths+0.037
 service+0.036
Neg logit contributions
 like-0.033
 everywhere-0.023
 favorite-0.023
 liked-0.023
 around-0.021
Token couldn
Token ID3521
Feature activation+4.40e+00

Pos logit contributions
 like+0.014
 liked+0.009
 everywhere+0.009
 favorite+0.008
 around+0.008
Neg logit contributions
ww-0.025
ister-0.025
www-0.024
 avail-0.024
ths-0.024
Token't
Token ID470
Feature activation+2.38e+00

Pos logit contributions
 avail+0.125
ww+0.124
ths+0.124
 sermon+0.123
www+0.120
Neg logit contributions
 like-0.087
 all-0.067
 both-0.066
 and-0.065
.-0.065
Token swim
Token ID9422
Feature activation-6.40e-01

Pos logit contributions
ww+0.057
ister+0.056
ths+0.056
www+0.055
 avail+0.053
Neg logit contributions
 like-0.060
 liked-0.043
 everywhere-0.042
 favorite-0.042
 around-0.040
Token and
Token ID290
Feature activation+3.47e-01

Pos logit contributions
 like+0.008
 pad+0.006
 liked+0.006
 dolls+0.006
 favorite+0.006
Neg logit contributions
ister-0.020
ww-0.019
 storm-0.019
ths-0.018
www-0.018
Token started
Token ID2067
Feature activation+2.14e+00

Pos logit contributions
ister+0.010
ww+0.010
www+0.010
ths+0.010
 avail+0.010
Neg logit contributions
 like-0.007
 liked-0.005
 everywhere-0.005
 all-0.004
 favorite-0.004
Token truck
Token ID7779
Feature activation+1.44e+00

Pos logit contributions
 like+0.020
 favorite+0.015
 liked+0.015
 everywhere+0.015
 around+0.014
Neg logit contributions
ister-0.016
ww-0.015
www-0.015
 storm-0.015
 service-0.015
Token!"
Token ID2474
Feature activation-6.39e-01

Pos logit contributions
ww+0.042
ths+0.042
ister+0.042
 avail+0.041
www+0.041
Neg logit contributions
 like-0.028
 everywhere-0.018
 liked-0.017
 favorite-0.017
 around-0.016
Token Anna
Token ID11735
Feature activation+4.59e-02

Pos logit contributions
 like+0.008
 liked+0.004
 everywhere+0.004
 favorite+0.004
 both+0.004
Neg logit contributions
ths-0.023
ister-0.023
ww-0.023
www-0.022
 service-0.022
Token says
Token ID1139
Feature activation-9.07e-01

Pos logit contributions
ister+0.001
ww+0.001
www+0.001
ths+0.001
 service+0.001
Neg logit contributions
 like-0.001
 everywhere-0.001
 favorite-0.001
 wherever-0.001
 liked-0.001
Token.
Token ID13
Feature activation-1.10e+00

Pos logit contributions
 like+0.012
 favorite+0.007
 liked+0.007
 everywhere+0.007
 pad+0.007
Neg logit contributions
ister-0.029
 service-0.028
ths-0.028
ww-0.028
 sermon-0.027
Token "
Token ID366
Feature activation-4.62e+00

Pos logit contributions
 like+0.019
 favorite+0.013
 everywhere+0.012
 liked+0.011
 pad+0.011
Neg logit contributions
ister-0.032
ww-0.031
www-0.030
iren-0.030
ths-0.030
TokenIt
Token ID1026
Feature activation+2.59e-01

Pos logit contributions
 like+0.107
 everywhere+0.084
 favorite+0.081
 liked+0.076
 dolls+0.076
Neg logit contributions
ww-0.108
ister-0.106
www-0.104
ths-0.102
 service-0.100
Token has
Token ID468
Feature activation-3.14e+00

Pos logit contributions
ister+0.007
ww+0.007
ths+0.007
www+0.007
 avail+0.007
Neg logit contributions
 like-0.006
 everywhere-0.004
 liked-0.004
 favorite-0.004
 around-0.004
Token a
Token ID257
Feature activation-8.99e-01

Pos logit contributions
 like+0.047
 everywhere+0.028
 liked+0.028
 favorite+0.027
 around+0.025
Neg logit contributions
ww-0.100
www-0.096
ister-0.096
 service-0.095
ths-0.095
Token bow
Token ID9563
Feature activation-2.43e+00

Pos logit contributions
 like+0.021
 favorite+0.015
 everywhere+0.015
 liked+0.014
 back+0.014
Neg logit contributions
ths-0.024
ister-0.023
ww-0.022
 avail-0.022
 service-0.022
Token like
Token ID588
Feature activation-2.27e+00

Pos logit contributions
 like+0.038
 liked+0.027
 favorite+0.027
 everywhere+0.025
 love+0.023
Neg logit contributions
ister-0.070
ww-0.070
www-0.067
 service-0.065
 beast-0.065
Token mom
Token ID1995
Feature activation+2.31e+00

Pos logit contributions
ister+0.001
ths+0.001
 service+0.001
ww+0.001
 storm+0.001
Neg logit contributions
 like-0.001
 everywhere-0.001
 liked-0.001
 pad-0.001
 wherever-0.001
Token hugged
Token ID41130
Feature activation-8.23e-01

Pos logit contributions
ister+0.060
ths+0.059
ww+0.058
www+0.057
 service+0.057
Neg logit contributions
 like-0.049
 favorite-0.036
 everywhere-0.035
 wherever-0.034
 pad-0.033
Token them
Token ID606
Feature activation-4.71e+00

Pos logit contributions
 like+0.016
 favorite+0.013
 pad+0.013
aces+0.013
 liked+0.012
Neg logit contributions
ister-0.020
ths-0.019
irens-0.019
 service-0.019
iren-0.019
Token and
Token ID290
Feature activation-3.10e-01

Pos logit contributions
 like+0.078
 ti+0.068
 favorites+0.061
aces+0.060
 pad+0.060
Neg logit contributions
 service-0.117
ister-0.112
ths-0.110
ama-0.107
iren-0.104
Token said
Token ID531
Feature activation-6.26e-01

Pos logit contributions
 like+0.008
 everywhere+0.006
 favorite+0.006
 liked+0.006
 around+0.005
Neg logit contributions
ister-0.007
ww-0.007
ths-0.007
www-0.007
 service-0.007
Token,
Token ID11
Feature activation-4.46e+00

Pos logit contributions
 like+0.012
 pad+0.008
 favorite+0.008
 liked+0.008
 everywhere+0.008
Neg logit contributions
ister-0.018
ww-0.017
ths-0.017
 service-0.017
www-0.016
Token "
Token ID366
Feature activation-3.49e+00

Pos logit contributions
 like+0.063
 favorite+0.037
 everywhere+0.035
 liked+0.031
 pad+0.030
Neg logit contributions
ister-0.146
ths-0.144
ww-0.141
 service-0.141
www-0.138
TokenI
Token ID40
Feature activation-2.41e+00

Pos logit contributions
 like+0.071
 liked+0.050
 everywhere+0.047
 favorite+0.047
 both+0.043
Neg logit contributions
ister-0.097
ww-0.094
ths-0.094
 avail-0.093
www-0.092
Token forgive
Token ID20927
Feature activation+1.14e+00

Pos logit contributions
 like+0.053
 everywhere+0.043
 favorite+0.040
aces+0.040
 around+0.039
Neg logit contributions
ister-0.058
ww-0.057
 service-0.055
www-0.055
ths-0.053
Token you
Token ID345
Feature activation-3.15e+00

Pos logit contributions
ister+0.019
ww+0.019
ths+0.018
www+0.018
 service+0.017
Neg logit contributions
 like-0.036
 everywhere-0.028
 favorite-0.028
 liked-0.028
 pad-0.027
Token,
Token ID11
Feature activation-2.98e+00

Pos logit contributions
 like+0.052
 liked+0.036
 favorite+0.035
 everywhere+0.035
 pad+0.032
Neg logit contributions
ister-0.093
 service-0.089
ths-0.088
ww-0.088
 storm-0.086
Token lion
Token ID18744
Feature activation-1.54e+00

Pos logit contributions
 like+0.034
 everywhere+0.025
 liked+0.025
 favorite+0.025
 around+0.023
Neg logit contributions
ww-0.031
ister-0.031
www-0.030
ths-0.030
 service-0.030
Token!"
Token ID2474
Feature activation-1.78e+00

Pos logit contributions
 like+0.031
 favorite+0.022
 everywhere+0.022
 liked+0.022
 wherever+0.020
Neg logit contributions
ister-0.042
ww-0.042
www-0.040
ths-0.040
 service-0.040
Token he
Token ID339
Feature activation+1.54e-02

Pos logit contributions
 like+0.026
 everywhere+0.015
 favorite+0.015
 liked+0.014
 around+0.012
Neg logit contributions
ister-0.059
ww-0.059
ths-0.058
www-0.058
 service-0.057
Token said
Token ID531
Feature activation-3.29e-01

Pos logit contributions
ister+0.000
ww+0.000
ths+0.000
www+0.000
 service+0.000
Neg logit contributions
 like-0.000
 liked-0.000
 everywhere-0.000
 favorite-0.000
 both-0.000
Token.
Token ID13
Feature activation-1.16e+00

Pos logit contributions
 like+0.005
 pad+0.004
 liked+0.004
 everywhere+0.004
 favorite+0.003
Neg logit contributions
ister-0.009
 service-0.009
 avail-0.009
ths-0.009
oo-0.008
Token "
Token ID366
Feature activation-4.59e+00

Pos logit contributions
 like+0.020
 everywhere+0.013
 favorite+0.013
 wherever+0.012
 pad+0.012
Neg logit contributions
ister-0.034
ww-0.033
www-0.032
 service-0.032
iren-0.032
TokenIt
Token ID1026
Feature activation+1.21e-01

Pos logit contributions
 like+0.105
 everywhere+0.081
 favorite+0.079
 liked+0.074
 around+0.072
Neg logit contributions
ww-0.111
ister-0.108
www-0.106
ths-0.105
 service-0.102
Token is
Token ID318
Feature activation-2.63e-01

Pos logit contributions
ister+0.003
ths+0.003
 avail+0.003
ww+0.003
www+0.003
Neg logit contributions
 like-0.003
 liked-0.002
 everywhere-0.002
 favorite-0.002
 both-0.002
Token so
Token ID523
Feature activation+2.70e+00

Pos logit contributions
 like+0.005
 liked+0.003
 favorite+0.003
 everywhere+0.003
 wherever+0.003
Neg logit contributions
ww-0.007
ister-0.007
www-0.007
ths-0.007
 storm-0.007
Token cool
Token ID3608
Feature activation-1.47e+00

Pos logit contributions
ths+0.084
ister+0.082
 avail+0.080
 service+0.078
ww+0.078
Neg logit contributions
 like-0.052
 liked-0.033
 everywhere-0.033
 favorite-0.033
 both-0.029
Token!"
Token ID2474
Feature activation-5.55e-01

Pos logit contributions
 like+0.025
 favorite+0.016
 liked+0.016
 everywhere+0.016
 dolls+0.014
Neg logit contributions
ww-0.044
ister-0.044
www-0.043
ths-0.043
 service-0.042
Token other
Token ID584
Feature activation-2.62e+00

Pos logit contributions
 like+0.007
 liked+0.005
 favorite+0.005
 everywhere+0.005
 pad+0.005
Neg logit contributions
ww-0.008
ister-0.008
 service-0.008
www-0.008
body-0.008
Token,"
Token ID553
Feature activation-6.93e-01

Pos logit contributions
 like+0.052
 everywhere+0.035
 favorite+0.035
 liked+0.035
 wherever+0.031
Neg logit contributions
ister-0.073
ww-0.072
ths-0.071
www-0.070
 service-0.069
Token Ben
Token ID3932
Feature activation+7.80e-01

Pos logit contributions
 like+0.009
 everywhere+0.004
 favorite+0.004
 liked+0.004
 both+0.003
Neg logit contributions
ister-0.025
ww-0.025
ths-0.024
www-0.024
 service-0.024
Token says
Token ID1139
Feature activation-8.66e-01

Pos logit contributions
ister+0.010
ww+0.009
 service+0.009
www+0.009
ths+0.008
Neg logit contributions
 like-0.026
 everywhere-0.022
 favorite-0.022
 wherever-0.022
 liked-0.021
Token.
Token ID13
Feature activation-7.22e-01

Pos logit contributions
 like+0.012
 favorite+0.007
 liked+0.007
 everywhere+0.006
 pad+0.006
Neg logit contributions
ister-0.029
 service-0.028
ths-0.027
ww-0.027
 avail-0.027
Token "
Token ID366
Feature activation-4.47e+00

Pos logit contributions
 like+0.013
 everywhere+0.008
 favorite+0.008
 pad+0.008
 wherever+0.007
Neg logit contributions
ister-0.020
ww-0.020
 service-0.020
ths-0.020
www-0.020
TokenWe
Token ID1135
Feature activation-4.35e+00

Pos logit contributions
 like+0.103
 everywhere+0.079
 favorite+0.076
 liked+0.072
 around+0.070
Neg logit contributions
ww-0.106
ister-0.105
ths-0.105
www-0.102
 service-0.101
Token are
Token ID389
Feature activation-5.47e-01

Pos logit contributions
 like+0.069
joy+0.066
 hers+0.065
 where+0.064
 skirts+0.063
Neg logit contributions
 storm-0.102
ister-0.099
www-0.097
ww-0.097
 service-0.092
Token twins
Token ID20345
Feature activation-3.49e+00

Pos logit contributions
 like+0.011
 pad+0.010
 liked+0.009
 everywhere+0.009
 favorite+0.009
Neg logit contributions
ister-0.012
 service-0.012
 storm-0.012
ths-0.012
ww-0.012
Token.
Token ID13
Feature activation-5.28e-01

Pos logit contributions
 like+0.050
 favorite+0.035
 everywhere+0.034
 pad+0.034
 Whe+0.033
Neg logit contributions
 service-0.100
ister-0.098
ths-0.094
ww-0.094
keys-0.092
Token We
Token ID775
Feature activation-4.78e+00

Pos logit contributions
 like+0.014
 everywhere+0.011
 favorite+0.010
 liked+0.010
 around+0.010
Neg logit contributions
ww-0.011
ister-0.011
www-0.011
ths-0.011
 service-0.011
Token "
Token ID366
Feature activation-3.94e+00

Pos logit contributions
 like+0.073
 favorite+0.046
 everywhere+0.044
 liked+0.039
aces+0.039
Neg logit contributions
ister-0.153
ths-0.150
ww-0.147
 service-0.145
www-0.145
TokenThis
Token ID1212
Feature activation+2.71e-01

Pos logit contributions
 like+0.090
 everywhere+0.066
 favorite+0.066
 liked+0.064
 around+0.060
Neg logit contributions
ww-0.098
ister-0.096
www-0.095
ths-0.095
 service-0.090
Token is
Token ID318
Feature activation+9.35e-01

Pos logit contributions
ww+0.006
ister+0.006
www+0.006
ths+0.006
 service+0.006
Neg logit contributions
 like-0.006
 everywhere-0.005
 favorite-0.005
 liked-0.005
 wherever-0.004
Token ind
Token ID773
Feature activation-3.49e+00

Pos logit contributions
ww+0.028
ister+0.027
ths+0.027
www+0.027
 service+0.026
Neg logit contributions
 like-0.016
 liked-0.011
 everywhere-0.011
 favorite-0.010
 pad-0.010
Tokenigo
Token ID14031
Feature activation-4.29e+00

Pos logit contributions
 like+0.096
 everywhere+0.075
 liked+0.074
 favorite+0.072
 around+0.069
Neg logit contributions
ister-0.070
ths-0.069
ww-0.068
www-0.065
 avail-0.063
Token,
Token ID11
Feature activation-6.03e+00

Pos logit contributions
 like+0.073
 everywhere+0.048
 favorite+0.047
 liked+0.045
 wherever+0.040
Neg logit contributions
ww-0.129
ister-0.127
www-0.123
ths-0.121
 service-0.119
Token and
Token ID290
Feature activation-2.28e+00

Pos logit contributions
 like+0.136
 liked+0.093
 everywhere+0.093
 favorite+0.092
 dolls+0.086
Neg logit contributions
ister-0.156
ths-0.154
ww-0.153
 avail-0.152
www-0.150
Token this
Token ID428
Feature activation+5.55e-01

Pos logit contributions
 like+0.056
 everywhere+0.041
 favorite+0.041
 liked+0.041
 around+0.038
Neg logit contributions
ister-0.053
ww-0.052
ths-0.051
www-0.051
 service-0.049
Token is
Token ID318
Feature activation+7.00e-01

Pos logit contributions
ww+0.014
ister+0.014
ths+0.014
www+0.014
 service+0.013
Neg logit contributions
 like-0.012
 everywhere-0.009
 favorite-0.009
 liked-0.009
 wherever-0.008
Token violet
Token ID45725
Feature activation-6.77e-01

Pos logit contributions
ww+0.020
ister+0.020
ths+0.020
www+0.020
 service+0.019
Neg logit contributions
 like-0.013
 everywhere-0.009
 liked-0.008
 favorite-0.008
 wherever-0.008
Token.
Token ID13
Feature activation-6.21e-01

Pos logit contributions
 like+0.009
 everywhere+0.005
 liked+0.005
 favorite+0.005
 wherever+0.004
Neg logit contributions
ister-0.023
ww-0.023
ths-0.022
www-0.022
 service-0.022