TOP ACTIVATIONS
MAX = 0.440

close her eyes . V isions and dreams would appear in
about her day at day care . She had made lots
They do not fight over videos anymore . <|endoftext|> L ily
At day care , Lily met a new
Sometimes they made their own videos with their toys and a
screen . They had many videos of animals , songs ,
and Lily liked to watch videos on the big screen .
Lily didn 't like day care because she didn 't know
at Jack and said " care ful , you don 't
had to go to day care . Lily didn 't like
Da isy started making videos and put them up on
Before they left day care , the teacher gave Amelia
, they found a blue computer in their dad 's office
them to take turns watching videos . He says he will
person . She is a computer program . She can 't
TV . They had many videos of cartoons , animals and
They can 't watch any videos . They wish they had
, a phone , a computer and some papers .
he would listen to the videos . He thought they sounded
loved to play with her computer . She had many games

MIN ACTIVATIONS
MIN = -0.492

was very happy to meet Fif i . F
" Hi , I am Fif i the frog . I
and pretty . Lily and Fif i played in the fog
ch ubby little fish named Fif i . She lived in
a named Fif i . Fif i lived in a big
a unique fle a named Fif i . Fif i lived
, " Thank you , Fif i , for showing me
other . They think about Fif i 's idea . They
her family and friends . Fif i loved to explore new
store with many toys . Fif i loved to play with
and play together again !" Fif i smiled and said ,
One day , Fif i and her friends went
One day , Fif i saw a sad girl
day on , Lily and Fif i were best friends .
" Hello , I 'm Fif i . I 'm a
and had fun , but Fif i noticed something strange .
not have enough money . Fif i wanted to help Lily
cave near the reef . Fif i wondered what was inside
Later that day , Fif i and her friends were
a toy for her . Fif i found a small toy

INTERVAL 0.207 - 0.440
CONTAINS 1.523%

and said , " Let 's go home now !"
happy . That day , Sarah had become bra ver and
M ummy , I need that red stamp ! â
and said , " It 's a surprise . Would you
Sure , Tom . Let 's go . I think you

INTERVAL -0.026 - 0.207
CONTAINS 65.345%

up and kept them in his house . The
rain . But you knew that walking in it would bring
sees a tw ig , a string , and a paper
happy to help and quickly unp acked his things .
picture . Can you set it on the table for me

INTERVAL -0.259 - -0.026
CONTAINS 32.595%

fish , and he was very happy . As
and a necklace around its neck . She thought it looked
medicine . He said it would heal in a few weeks
she wanted to arrange things just right . She wanted the
laughed and said to her mum , " M ummy ,

INTERVAL -0.492 - -0.259
CONTAINS 0.537%

who could shoot the most strawberries . Sam
." L ily sn uck up on the
. M ia 's mom saw that Mia
water . L ily went to her mom and
warned her . " L ily , if you take a
Token close
Token ID1969
Feature activation+6.38e-02

Pos logit contributions
isions+0.010
 videos+0.010
care+0.010
 mechanic+0.010
 nephew+0.010
Neg logit contributions
 Fif-0.013
 bamboo-0.012
onna-0.011
iw-0.011
 pear-0.011
Token her
Token ID607
Feature activation-4.98e-02

Pos logit contributions
's+0.083
care+0.080
 videos+0.079
 Margaret+0.077
 means+0.077
Neg logit contributions
 Fif-0.123
 bamboo-0.114
iw-0.111
onna-0.108
sed-0.107
Token eyes
Token ID2951
Feature activation-1.10e-01

Pos logit contributions
 Fif+0.086
 bamboo+0.074
iw+0.071
onna+0.070
io+0.066
Neg logit contributions
's-0.082
 videos-0.080
isions-0.078
 computer-0.078
care-0.077
Token.
Token ID13
Feature activation-6.07e-03

Pos logit contributions
 Fif+0.192
 bamboo+0.171
onna+0.162
iw+0.160
 pear+0.156
Neg logit contributions
isions-0.163
 videos-0.162
care-0.161
 mechanic-0.157
's-0.156
Token V
Token ID569
Feature activation-2.97e-02

Pos logit contributions
 Fif+0.012
 bamboo+0.011
onna+0.011
iw+0.010
io+0.010
Neg logit contributions
 videos-0.008
's-0.007
isions-0.007
care-0.007
 computer-0.007
Tokenisions
Token ID3279
Feature activation+4.18e-01

Pos logit contributions
 Fif+0.058
 bamboo+0.051
onna+0.050
iw+0.049
sed+0.048
Neg logit contributions
isions-0.040
care-0.039
 videos-0.038
's-0.038
ographs-0.037
Token and
Token ID290
Feature activation+8.36e-03

Pos logit contributions
isions+0.669
 videos+0.655
care+0.653
 nephew+0.645
 mechanic+0.637
Neg logit contributions
 Fif-0.666
 bamboo-0.589
onna-0.567
iw-0.546
 pear-0.546
Token dreams
Token ID10625
Feature activation-3.72e-03

Pos logit contributions
 videos+0.012
 computer+0.011
 computers+0.011
isions+0.011
care+0.011
Neg logit contributions
 Fif-0.015
 bamboo-0.014
iw-0.013
onna-0.013
io-0.013
Token would
Token ID561
Feature activation-3.08e-02

Pos logit contributions
 Fif+0.007
 bamboo+0.006
iw+0.006
onna+0.006
 pear+0.006
Neg logit contributions
 videos-0.005
isions-0.005
care-0.005
's-0.005
 mechanic-0.005
Token appear
Token ID1656
Feature activation-8.23e-03

Pos logit contributions
 Fif+0.056
iw+0.049
 bamboo+0.049
io+0.048
sed+0.047
Neg logit contributions
 videos-0.043
's-0.043
 jet-0.042
 mechanic-0.042
isions-0.041
Token in
Token ID287
Feature activation-3.07e-02

Pos logit contributions
 Fif+0.015
 bamboo+0.014
iw+0.013
onna+0.013
sed+0.013
Neg logit contributions
 videos-0.011
care-0.011
's-0.011
isions-0.011
 means-0.011
Token about
Token ID546
Feature activation+6.57e-02

Pos logit contributions
isions+0.124
care+0.118
controlled+0.118
 mechanic+0.118
 videos+0.117
Neg logit contributions
 Fif-0.122
 bamboo-0.107
onna-0.101
 pear-0.100
iw-0.096
Token her
Token ID607
Feature activation-4.40e-02

Pos logit contributions
 videos+0.094
 computers+0.088
care+0.087
 computer+0.086
isions+0.086
Neg logit contributions
 Fif-0.121
iw-0.107
 bamboo-0.107
onna-0.106
sed-0.104
Token day
Token ID1110
Feature activation+2.95e-02

Pos logit contributions
 Fif+0.078
 bamboo+0.071
onna+0.065
 pear+0.065
iw+0.064
Neg logit contributions
isions-0.064
care-0.061
 videos-0.061
controlled-0.061
 mechanic-0.060
Token at
Token ID379
Feature activation+1.15e-01

Pos logit contributions
isions+0.057
controlled+0.054
 nephew+0.054
 mechanic+0.053
 keyboard+0.053
Neg logit contributions
 Fif-0.034
 bamboo-0.031
 blo-0.029
 pear-0.029
 ti-0.028
Token day
Token ID1110
Feature activation+2.97e-02

Pos logit contributions
 videos+0.155
care+0.151
isions+0.151
 mechanic+0.149
's+0.147
Neg logit contributions
 Fif-0.217
 bamboo-0.194
onna-0.185
iw-0.185
sed-0.181
Tokencare
Token ID6651
Feature activation+4.17e-01

Pos logit contributions
controlled+0.060
isions+0.060
 nephew+0.059
 engineer+0.059
 keyboard+0.056
Neg logit contributions
 Fif-0.030
 bamboo-0.027
onna-0.027
 ti-0.027
 blo-0.026
Token.
Token ID13
Feature activation+5.67e-03

Pos logit contributions
isions+0.657
 videos+0.643
care+0.633
 mechanic+0.631
controlled+0.627
Neg logit contributions
 Fif-0.677
 bamboo-0.597
onna-0.575
iw-0.550
 pear-0.543
Token She
Token ID1375
Feature activation-1.19e-02

Pos logit contributions
care+0.007
 videos+0.006
's+0.006
isions+0.006
 mechanic+0.006
Neg logit contributions
 Fif-0.012
 bamboo-0.011
iw-0.010
sed-0.010
onna-0.010
Token had
Token ID550
Feature activation-3.07e-02

Pos logit contributions
 Fif+0.021
 bamboo+0.019
iw+0.018
onna+0.018
 pear+0.017
Neg logit contributions
's-0.017
 videos-0.017
isions-0.017
 mechanic-0.016
care-0.016
Token made
Token ID925
Feature activation-3.54e-02

Pos logit contributions
 Fif+0.060
 bamboo+0.054
iw+0.051
onna+0.051
 pear+0.049
Neg logit contributions
isions-0.040
 videos-0.039
care-0.039
's-0.038
 mechanic-0.038
Token lots
Token ID6041
Feature activation-4.94e-02

Pos logit contributions
 Fif+0.059
 bamboo+0.054
 pear+0.049
onna+0.048
 crown+0.047
Neg logit contributions
isions-0.055
 mechanic-0.052
controlled-0.052
care-0.051
 keyboard-0.051
Token They
Token ID1119
Feature activation+1.18e-01

Pos logit contributions
 Fif+0.002
 bamboo+0.002
onna+0.002
iw+0.002
 pear+0.002
Neg logit contributions
isions-0.001
 videos-0.001
care-0.001
 mechanic-0.001
's-0.001
Token do
Token ID466
Feature activation+8.13e-02

Pos logit contributions
's+0.138
 videos+0.130
 mechanic+0.129
isions+0.128
 jet+0.127
Neg logit contributions
 Fif-0.240
 bamboo-0.220
iw-0.215
onna-0.212
sed-0.210
Token not
Token ID407
Feature activation-8.94e-02

Pos logit contributions
 videos+0.135
isions+0.133
's+0.130
care+0.130
 mechanic+0.128
Neg logit contributions
 Fif-0.130
 bamboo-0.114
iw-0.108
onna-0.107
sed-0.104
Token fight
Token ID1907
Feature activation+2.31e-01

Pos logit contributions
 Fif+0.167
onna+0.149
 bamboo+0.144
 pear+0.142
iw+0.134
Neg logit contributions
isions-0.126
cases-0.122
 Margaret-0.119
controlled-0.118
 nephew-0.117
Token over
Token ID625
Feature activation-3.60e-02

Pos logit contributions
isions+0.385
 nephew+0.367
care+0.362
controlled+0.361
 Margaret+0.360
Neg logit contributions
 Fif-0.361
 bamboo-0.316
onna-0.303
 pear-0.289
balls-0.282
Token videos
Token ID5861
Feature activation+4.16e-01

Pos logit contributions
 Fif+0.054
 bamboo+0.052
 pear+0.048
 bananas+0.044
onna+0.044
Neg logit contributions
isions-0.057
care-0.056
controlled-0.055
 mechanic-0.054
issy-0.054
Token anymore
Token ID7471
Feature activation+4.54e-02

Pos logit contributions
's+0.551
care+0.551
 videos+0.550
isions+0.534
 means+0.516
Neg logit contributions
 Fif-0.809
 bamboo-0.708
iw-0.685
onna-0.683
sed-0.677
Token.
Token ID13
Feature activation+6.08e-04

Pos logit contributions
isions+0.069
 videos+0.068
care+0.067
 mechanic+0.066
 nephew+0.066
Neg logit contributions
 Fif-0.078
 bamboo-0.069
onna-0.065
iw-0.064
 pear-0.063
Token<|endoftext|>
Token ID50256
Feature activation+1.79e-02

Pos logit contributions
isions+0.001
 videos+0.001
care+0.001
 mechanic+0.001
controlled+0.001
Neg logit contributions
 Fif-0.001
 bamboo-0.001
onna-0.001
iw-0.001
 pear-0.001
TokenL
Token ID43
Feature activation+2.04e-02

Pos logit contributions
care+0.024
's+0.023
 videos+0.023
 means+0.023
isions+0.022
Neg logit contributions
 Fif-0.034
 bamboo-0.030
sed-0.030
onna-0.029
 majesty-0.029
Tokenily
Token ID813
Feature activation-2.65e-01

Pos logit contributions
isions+0.039
 videos+0.039
 mechanic+0.038
 nephew+0.038
care+0.038
Neg logit contributions
 Fif-0.026
 bamboo-0.023
onna-0.021
iw-0.021
 pear-0.021
Token
Token ID220
Feature activation-6.51e-02

Pos logit contributions
 Fif+0.013
 bamboo+0.012
onna+0.011
iw+0.011
io+0.011
Neg logit contributions
's-0.008
care-0.008
 videos-0.008
isions-0.008
 computers-0.007
Token
Token ID198
Feature activation-3.52e-02

Pos logit contributions
 Fif+0.112
 bamboo+0.099
onna+0.094
iw+0.093
 pear+0.091
Neg logit contributions
isions-0.098
 videos-0.097
care-0.095
 mechanic-0.094
 nephew-0.093
Token
Token ID198
Feature activation-3.49e-02

Pos logit contributions
 Fif+0.060
 bamboo+0.053
onna+0.050
iw+0.050
 pear+0.048
Neg logit contributions
isions-0.054
 videos-0.053
care-0.052
 mechanic-0.051
 nephew-0.051
TokenAt
Token ID2953
Feature activation+1.22e-01

Pos logit contributions
 Fif+0.050
 bamboo+0.043
 herself+0.042
 pear+0.041
 ti+0.039
Neg logit contributions
 nephew-0.059
 videos-0.059
 films-0.058
isions-0.057
 Steven-0.057
Token day
Token ID1110
Feature activation+2.39e-02

Pos logit contributions
isions+0.202
 videos+0.191
controlled+0.188
care+0.187
's+0.185
Neg logit contributions
 Fif-0.203
 bamboo-0.178
onna-0.169
iw-0.165
 pear-0.163
Tokencare
Token ID6651
Feature activation+4.14e-01

Pos logit contributions
isions+0.048
 nephew+0.045
 engineer+0.044
 computers+0.044
urious+0.044
Neg logit contributions
 Fif-0.034
 bamboo-0.028
onna-0.027
 pear-0.026
balls-0.024
Token,
Token ID11
Feature activation-5.17e-03

Pos logit contributions
isions+0.590
 videos+0.582
care+0.573
 mechanic+0.562
 nephew+0.557
Neg logit contributions
 Fif-0.748
 bamboo-0.666
onna-0.636
iw-0.625
 pear-0.614
Token Lily
Token ID20037
Feature activation-8.33e-02

Pos logit contributions
 Fif+0.008
 bamboo+0.006
 pear+0.006
onna+0.006
iw+0.006
Neg logit contributions
isions-0.010
 videos-0.009
controlled-0.009
 mechanic-0.009
care-0.009
Token met
Token ID1138
Feature activation+3.72e-02

Pos logit contributions
 Fif+0.134
 l+0.127
 p+0.121
 pear+0.120
 bamboo+0.118
Neg logit contributions
controlled-0.123
isions-0.120
 videos-0.120
ographs-0.113
 keyboard-0.111
Token a
Token ID257
Feature activation+6.14e-02

Pos logit contributions
 videos+0.047
care+0.045
isions+0.045
's+0.045
 mechanic+0.045
Neg logit contributions
 Fif-0.073
 bamboo-0.066
onna-0.065
iw-0.064
 pear-0.061
Token new
Token ID649
Feature activation-1.78e-02

Pos logit contributions
isions+0.090
controlled+0.087
 videos+0.085
care+0.085
cases+0.083
Neg logit contributions
 Fif-0.111
 bamboo-0.098
 pear-0.095
onna-0.093
iw-0.092
Token Sometimes
Token ID8975
Feature activation+1.16e-01

Pos logit contributions
 videos+0.004
isions+0.004
care+0.004
's+0.004
 mechanic+0.004
Neg logit contributions
 Fif-0.007
 bamboo-0.006
iw-0.006
onna-0.006
 pear-0.006
Token they
Token ID484
Feature activation+1.71e-01

Pos logit contributions
isions+0.193
controlled+0.186
urious+0.181
 keyboard+0.180
arre+0.179
Neg logit contributions
 Fif-0.194
 bamboo-0.170
 pear-0.155
 strawberries-0.151
 Sara-0.145
Token made
Token ID925
Feature activation-3.96e-02

Pos logit contributions
isions+0.260
 videos+0.256
care+0.254
 mechanic+0.249
 nephew+0.247
Neg logit contributions
 Fif-0.291
 bamboo-0.258
onna-0.245
iw-0.241
 pear-0.236
Token their
Token ID511
Feature activation+1.39e-01

Pos logit contributions
 Fif+0.061
 bamboo+0.056
io+0.051
 l+0.048
 pear+0.048
Neg logit contributions
 mechanic-0.065
care-0.062
controlled-0.062
isions-0.062
 nephew-0.061
Token own
Token ID898
Feature activation+3.55e-02

Pos logit contributions
care+0.223
isions+0.222
 mechanic+0.216
controlled+0.215
 Margaret+0.213
Neg logit contributions
 Fif-0.218
 bamboo-0.195
sed-0.188
 pear-0.177
balls-0.175
Token videos
Token ID5861
Feature activation+4.14e-01

Pos logit contributions
 mechanic+0.054
care+0.053
limits+0.053
controlled+0.052
 Margaret+0.051
Neg logit contributions
 Fif-0.051
 bamboo-0.047
balls-0.046
 belly-0.046
 l-0.046
Token with
Token ID351
Feature activation+8.76e-03

Pos logit contributions
 videos+0.578
isions+0.564
care+0.561
's+0.558
 mechanic+0.548
Neg logit contributions
 Fif-0.772
 bamboo-0.681
iw-0.654
onna-0.652
sed-0.633
Token their
Token ID511
Feature activation+1.40e-01

Pos logit contributions
isions+0.012
care+0.012
 mechanic+0.012
 nephew+0.011
controlled+0.011
Neg logit contributions
 Fif-0.016
 bamboo-0.014
onna-0.013
 pear-0.013
iw-0.013
Token toys
Token ID14958
Feature activation+3.29e-02

Pos logit contributions
isions+0.198
care+0.195
 mechanic+0.192
 videos+0.189
 nephew+0.186
Neg logit contributions
 Fif-0.248
 bamboo-0.222
onna-0.209
sed-0.208
iw-0.207
Token and
Token ID290
Feature activation+1.55e-02

Pos logit contributions
isions+0.049
 videos+0.049
care+0.048
 mechanic+0.047
 nephew+0.047
Neg logit contributions
 Fif-0.056
 bamboo-0.050
onna-0.048
iw-0.047
 pear-0.046
Token a
Token ID257
Feature activation+6.65e-02

Pos logit contributions
isions+0.022
care+0.022
 mechanic+0.021
controlled+0.021
ographs+0.020
Neg logit contributions
 Fif-0.028
 bamboo-0.025
onna-0.023
 pear-0.023
sed-0.023
Token screen
Token ID3159
Feature activation+2.11e-01

Pos logit contributions
 videos+0.013
isions+0.013
care+0.013
 mechanic+0.013
's+0.013
Neg logit contributions
 Fif-0.019
 bamboo-0.017
onna-0.017
iw-0.016
sed-0.016
Token.
Token ID13
Feature activation+2.42e-03

Pos logit contributions
isions+0.324
 videos+0.318
care+0.316
 mechanic+0.310
 nephew+0.309
Neg logit contributions
 Fif-0.355
 bamboo-0.316
onna-0.299
iw-0.293
 pear-0.289
Token They
Token ID1119
Feature activation+1.19e-01

Pos logit contributions
isions+0.002
 videos+0.002
care+0.002
 mechanic+0.002
 nephew+0.002
Neg logit contributions
 Fif-0.005
 bamboo-0.005
onna-0.005
iw-0.005
 pear-0.005
Token had
Token ID550
Feature activation-3.20e-02

Pos logit contributions
isions+0.177
 videos+0.176
care+0.173
's+0.171
 mechanic+0.171
Neg logit contributions
 Fif-0.206
 bamboo-0.184
onna-0.174
iw-0.173
 pear-0.167
Token many
Token ID867
Feature activation-1.51e-02

Pos logit contributions
 Fif+0.064
 bamboo+0.058
iw+0.056
onna+0.056
 pear+0.053
Neg logit contributions
 videos-0.040
isions-0.040
's-0.039
care-0.039
 computers-0.038
Token videos
Token ID5861
Feature activation+4.13e-01

Pos logit contributions
 Fif+0.022
 bamboo+0.021
 l+0.019
io+0.019
balls+0.019
Neg logit contributions
 mechanic-0.024
care-0.024
 Margaret-0.023
isions-0.023
arse-0.023
Token of
Token ID286
Feature activation+8.39e-02

Pos logit contributions
 videos+0.391
's+0.360
care+0.353
 computer+0.341
isions+0.330
Neg logit contributions
 Fif-0.983
iw-0.877
onna-0.876
 bamboo-0.876
sed-0.863
Token animals
Token ID4695
Feature activation+1.48e-02

Pos logit contributions
isions+0.111
care+0.108
 videos+0.108
 mechanic+0.107
 nephew+0.105
Neg logit contributions
 Fif-0.159
 bamboo-0.143
onna-0.135
iw-0.134
 pear-0.132
Token,
Token ID11
Feature activation+2.66e-03

Pos logit contributions
 videos+0.020
isions+0.020
's+0.020
care+0.019
 means+0.019
Neg logit contributions
 Fif-0.028
 bamboo-0.025
onna-0.024
iw-0.024
sed-0.023
Token songs
Token ID7259
Feature activation+2.90e-02

Pos logit contributions
isions+0.004
care+0.004
 videos+0.004
 mechanic+0.004
controlled+0.004
Neg logit contributions
 Fif-0.005
 bamboo-0.004
onna-0.004
iw-0.004
 pear-0.004
Token,
Token ID11
Feature activation+2.20e-03

Pos logit contributions
's+0.033
 videos+0.033
 means+0.031
isions+0.031
care+0.031
Neg logit contributions
 Fif-0.061
iw-0.056
 bamboo-0.056
onna-0.053
 pear-0.053
Token and
Token ID290
Feature activation+1.38e-02

Pos logit contributions
 Fif+0.090
 bamboo+0.081
onna+0.077
iw+0.076
sed+0.074
Neg logit contributions
isions-0.062
's-0.061
care-0.061
 videos-0.061
 mechanic-0.060
Token Lily
Token ID20037
Feature activation-7.41e-02

Pos logit contributions
isions+0.022
 videos+0.021
care+0.021
ographs+0.020
 computers+0.020
Neg logit contributions
 Fif-0.024
 bamboo-0.021
 pear-0.020
onna-0.020
iw-0.019
Token liked
Token ID8288
Feature activation-6.85e-02

Pos logit contributions
 Fif+0.134
 bamboo+0.119
onna+0.113
iw+0.113
 pear+0.110
Neg logit contributions
isions-0.105
 videos-0.104
care-0.102
 mechanic-0.100
 nephew-0.099
Token to
Token ID284
Feature activation-2.53e-02

Pos logit contributions
 Fif+0.116
 bamboo+0.103
onna+0.098
iw+0.096
 pear+0.094
Neg logit contributions
isions-0.105
 videos-0.104
care-0.102
 mechanic-0.101
 nephew-0.100
Token watch
Token ID2342
Feature activation+6.53e-03

Pos logit contributions
 Fif+0.047
 bamboo+0.043
onna+0.041
iw+0.040
 pear+0.039
Neg logit contributions
isions-0.034
 videos-0.034
care-0.033
 mechanic-0.032
's-0.032
Token videos
Token ID5861
Feature activation+4.13e-01

Pos logit contributions
isions+0.011
controlled+0.011
cases+0.010
urious+0.010
ever+0.010
Neg logit contributions
 Fif-0.010
 bamboo-0.009
 pear-0.008
 bananas-0.008
 crown-0.008
Token on
Token ID319
Feature activation-3.76e-02

Pos logit contributions
 videos+0.545
care+0.512
's+0.508
isions+0.490
 computer+0.489
Neg logit contributions
 Fif-0.827
iw-0.725
onna-0.723
 bamboo-0.720
sed-0.716
Token the
Token ID262
Feature activation+1.33e-02

Pos logit contributions
 Fif+0.063
 bamboo+0.057
 pear+0.053
onna+0.052
io+0.052
Neg logit contributions
isions-0.058
care-0.056
 nephew-0.055
controlled-0.055
 mechanic-0.055
Token big
Token ID1263
Feature activation+1.00e-02

Pos logit contributions
isions+0.015
care+0.014
 videos+0.014
 mechanic+0.014
 nephew+0.014
Neg logit contributions
 Fif-0.028
 bamboo-0.026
onna-0.024
iw-0.024
 pear-0.024
Token screen
Token ID3159
Feature activation+2.11e-01

Pos logit contributions
 videos+0.013
isions+0.013
care+0.013
 mechanic+0.013
's+0.013
Neg logit contributions
 Fif-0.019
 bamboo-0.017
onna-0.017
iw-0.016
sed-0.016
Token.
Token ID13
Feature activation+2.42e-03

Pos logit contributions
isions+0.324
 videos+0.318
care+0.316
 mechanic+0.310
 nephew+0.309
Neg logit contributions
 Fif-0.355
 bamboo-0.316
onna-0.299
iw-0.293
 pear-0.289
Token Lily
Token ID20037
Feature activation-8.57e-02

Pos logit contributions
 Fif+0.013
 bamboo+0.011
onna+0.010
 pear+0.010
sed+0.010
Neg logit contributions
isions-0.016
 nephew-0.016
 jet-0.016
controlled-0.016
ographs-0.016
Token didn
Token ID1422
Feature activation+1.21e-01

Pos logit contributions
 Fif+0.129
 bamboo+0.110
 pear+0.108
 l+0.106
sed+0.105
Neg logit contributions
 videos-0.145
controlled-0.142
isions-0.141
ographs-0.134
 keyboard-0.134
Token't
Token ID470
Feature activation+9.94e-03

Pos logit contributions
 nephew+0.257
isions+0.247
urious+0.242
ever+0.240
ations+0.239
Neg logit contributions
 Fif-0.140
onna-0.125
 bamboo-0.109
 pear-0.102
balls-0.100
Token like
Token ID588
Feature activation-1.12e-01

Pos logit contributions
isions+0.016
ographs+0.015
 computers+0.014
 videos+0.014
 jet+0.014
Neg logit contributions
 Fif-0.017
onna-0.015
 bamboo-0.015
 pear-0.014
 ti-0.013
Token day
Token ID1110
Feature activation+2.06e-02

Pos logit contributions
 Fif+0.199
 bamboo+0.178
onna+0.169
iw+0.167
 pear+0.163
Neg logit contributions
isions-0.163
 videos-0.162
care-0.160
 mechanic-0.156
's-0.155
Tokencare
Token ID6651
Feature activation+4.12e-01

Pos logit contributions
isions+0.036
 nephew+0.034
 mechanic+0.034
urious+0.033
controlled+0.033
Neg logit contributions
 Fif-0.030
 bamboo-0.026
onna-0.026
 pear-0.024
iw-0.023
Token because
Token ID780
Feature activation+8.12e-02

Pos logit contributions
 videos+0.546
isions+0.542
care+0.537
's+0.524
 mechanic+0.520
Neg logit contributions
 Fif-0.781
 bamboo-0.706
iw-0.672
onna-0.665
 pear-0.649
Token she
Token ID673
Feature activation+1.08e-02

Pos logit contributions
isions+0.138
controlled+0.130
care+0.129
 videos+0.128
 nephew+0.128
Neg logit contributions
 Fif-0.130
 bamboo-0.115
 pear-0.107
iw-0.104
onna-0.103
Token didn
Token ID1422
Feature activation+1.22e-01

Pos logit contributions
isions+0.016
 videos+0.016
care+0.015
controlled+0.015
 mechanic+0.015
Neg logit contributions
 Fif-0.019
 bamboo-0.017
onna-0.016
iw-0.016
 pear-0.016
Token't
Token ID470
Feature activation+1.14e-02

Pos logit contributions
 nephew+0.267
isions+0.253
urious+0.246
ever+0.246
 engineer+0.244
Neg logit contributions
 Fif-0.133
onna-0.122
 bamboo-0.104
 herself-0.101
balls-0.100
Token know
Token ID760
Feature activation+1.46e-01

Pos logit contributions
isions+0.019
ographs+0.018
urious+0.017
 jet+0.017
cases+0.017
Neg logit contributions
 Fif-0.019
onna-0.016
 bamboo-0.016
 pear-0.016
 ever-0.015
Token at
Token ID379
Feature activation+1.10e-01

Pos logit contributions
isions+0.299
 videos+0.289
 mechanic+0.288
controlled+0.286
care+0.280
Neg logit contributions
 Fif-0.262
 bamboo-0.233
 pear-0.216
sed-0.215
onna-0.212
Token Jack
Token ID3619
Feature activation+9.48e-02

Pos logit contributions
 videos+0.158
isions+0.157
care+0.155
's+0.152
 mechanic+0.152
Neg logit contributions
 Fif-0.197
 bamboo-0.176
onna-0.169
iw-0.166
 pear-0.161
Token and
Token ID290
Feature activation+1.07e-02

Pos logit contributions
isions+0.162
ographs+0.157
controlled+0.157
 videos+0.155
 mechanic+0.152
Neg logit contributions
 Fif-0.142
 bamboo-0.126
onna-0.122
 pear-0.119
 belly-0.112
Token said
Token ID531
Feature activation+4.78e-02

Pos logit contributions
isions+0.014
 videos+0.014
care+0.014
 mechanic+0.013
controlled+0.013
Neg logit contributions
 Fif-0.020
 bamboo-0.018
onna-0.017
iw-0.017
 pear-0.017
Token "
Token ID366
Feature activation-8.96e-02

Pos logit contributions
 videos+0.063
's+0.063
care+0.063
isions+0.063
 mechanic+0.061
Neg logit contributions
 Fif-0.089
 bamboo-0.081
iw-0.078
onna-0.077
 pear-0.074
Tokencare
Token ID6651
Feature activation+4.11e-01

Pos logit contributions
 Fif+0.145
iw+0.126
 bamboo+0.125
onna+0.122
io+0.117
Neg logit contributions
 videos-0.144
isions-0.140
 jet-0.135
controlled-0.135
 mechanic-0.134
Tokenful
Token ID913
Feature activation-1.19e-03

Pos logit contributions
isions+0.590
 videos+0.584
care+0.577
 mechanic+0.564
 nephew+0.559
Neg logit contributions
 Fif-0.733
 bamboo-0.654
onna-0.622
iw-0.614
 pear-0.601
Token,
Token ID11
Feature activation-2.48e-03

Pos logit contributions
 Fif+0.002
 bamboo+0.002
onna+0.002
 pear+0.002
iw+0.002
Neg logit contributions
isions-0.002
care-0.002
 videos-0.002
 mechanic-0.002
 nephew-0.002
Token you
Token ID345
Feature activation+2.79e-02

Pos logit contributions
 Fif+0.004
 bamboo+0.004
onna+0.003
 pear+0.003
iw+0.003
Neg logit contributions
isions-0.004
care-0.004
 videos-0.004
controlled-0.004
 nephew-0.004
Token don
Token ID836
Feature activation+6.52e-02

Pos logit contributions
care+0.047
 videos+0.047
controlled+0.046
isions+0.045
 nephew+0.044
Neg logit contributions
 Fif-0.042
 pear-0.035
 bamboo-0.034
sed-0.034
onna-0.033
Token't
Token ID470
Feature activation+1.67e-02

Pos logit contributions
 nephew+0.132
 grandson+0.123
isions+0.123
 computers+0.122
care+0.122
Neg logit contributions
 Fif-0.077
onna-0.061
 bamboo-0.060
balls-0.060
 pear-0.053
Token had
Token ID550
Feature activation-4.26e-02

Pos logit contributions
isions+0.015
care+0.015
 videos+0.014
controlled+0.014
 jet+0.014
Neg logit contributions
 Fif-0.012
 pear-0.011
sed-0.011
 bamboo-0.011
onna-0.010
Token to
Token ID284
Feature activation-2.82e-02

Pos logit contributions
 Fif+0.080
 bamboo+0.072
iw+0.068
onna+0.068
 pear+0.065
Neg logit contributions
 videos-0.057
care-0.057
isions-0.057
's-0.057
 computers-0.056
Token go
Token ID467
Feature activation+1.03e-01

Pos logit contributions
 Fif+0.050
 bamboo+0.045
onna+0.044
 pear+0.043
iw+0.041
Neg logit contributions
isions-0.040
care-0.038
 videos-0.038
controlled-0.038
 nephew-0.038
Token to
Token ID284
Feature activation-2.84e-02

Pos logit contributions
isions+0.225
ographs+0.210
urious+0.207
 nephew+0.207
controlled+0.203
Neg logit contributions
 Fif-0.104
 bamboo-0.098
onna-0.094
 l-0.083
 crown-0.082
Token day
Token ID1110
Feature activation+1.95e-02

Pos logit contributions
 Fif+0.044
 bamboo+0.038
onna+0.037
 pear+0.036
iw+0.033
Neg logit contributions
isions-0.051
controlled-0.048
cases-0.048
ographs-0.047
 nephew-0.045
Tokencare
Token ID6651
Feature activation+4.11e-01

Pos logit contributions
controlled+0.041
 nephew+0.040
isions+0.040
ographs+0.040
 means+0.040
Neg logit contributions
onna-0.018
 Fif-0.018
 bamboo-0.015
 ti-0.014
 l-0.014
Token.
Token ID13
Feature activation-9.01e-03

Pos logit contributions
 videos+0.566
care+0.563
isions+0.559
's+0.552
 mechanic+0.542
Neg logit contributions
 Fif-0.764
 bamboo-0.689
iw-0.657
onna-0.640
 pear-0.636
Token Lily
Token ID20037
Feature activation-8.57e-02

Pos logit contributions
 Fif+0.013
 bamboo+0.011
onna+0.010
 pear+0.010
sed+0.010
Neg logit contributions
isions-0.016
 nephew-0.016
 jet-0.016
controlled-0.016
ographs-0.016
Token didn
Token ID1422
Feature activation+1.21e-01

Pos logit contributions
 Fif+0.129
 bamboo+0.110
 pear+0.108
 l+0.106
sed+0.105
Neg logit contributions
 videos-0.145
controlled-0.142
isions-0.141
ographs-0.134
 keyboard-0.134
Token't
Token ID470
Feature activation+9.94e-03

Pos logit contributions
 nephew+0.257
isions+0.247
urious+0.242
ever+0.240
ations+0.239
Neg logit contributions
 Fif-0.140
onna-0.125
 bamboo-0.109
 pear-0.102
balls-0.100
Token like
Token ID588
Feature activation-1.12e-01

Pos logit contributions
isions+0.016
ographs+0.015
 computers+0.014
 videos+0.014
 jet+0.014
Neg logit contributions
 Fif-0.017
onna-0.015
 bamboo-0.015
 pear-0.014
 ti-0.013
Token
Token ID198
Feature activation-2.87e-02

Pos logit contributions
 Fif+0.053
 bamboo+0.048
iw+0.046
onna+0.046
sed+0.045
Neg logit contributions
care-0.039
isions-0.039
 videos-0.039
's-0.038
 mechanic-0.037
TokenDa
Token ID26531
Feature activation-6.53e-02

Pos logit contributions
 Fif+0.052
 bamboo+0.046
iw+0.042
 pear+0.042
onna+0.041
Neg logit contributions
 nephew-0.041
 mechanic-0.040
 videos-0.040
isions-0.040
controlled-0.039
Tokenisy
Token ID13560
Feature activation-1.75e-01

Pos logit contributions
 Fif+0.061
 bamboo+0.048
onna+0.041
 pear+0.038
iw+0.037
Neg logit contributions
care-0.151
isions-0.151
 mechanic-0.150
 videos-0.150
 computers-0.148
Token started
Token ID2067
Feature activation+7.27e-02

Pos logit contributions
 Fif+0.314
 bamboo+0.280
onna+0.267
iw+0.263
 pear+0.258
Neg logit contributions
isions-0.248
 videos-0.245
care-0.242
 mechanic-0.236
 nephew-0.234
Token making
Token ID1642
Feature activation-3.20e-02

Pos logit contributions
isions+0.136
controlled+0.132
ographs+0.129
 mechanic+0.129
 nephew+0.126
Neg logit contributions
 Fif-0.102
onna-0.082
 bamboo-0.081
 pear-0.075
 l-0.074
Token videos
Token ID5861
Feature activation+4.11e-01

Pos logit contributions
 Fif+0.054
 bamboo+0.048
 pear+0.044
onna+0.042
 crown+0.042
Neg logit contributions
isions-0.050
controlled-0.049
 mechanic-0.049
care-0.048
 nephew-0.047
Token and
Token ID290
Feature activation+1.43e-02

Pos logit contributions
 videos+0.553
's+0.538
care+0.536
isions+0.532
 mechanic+0.509
Neg logit contributions
 Fif-0.789
 bamboo-0.694
iw-0.683
onna-0.675
sed-0.662
Token put
Token ID1234
Feature activation+3.55e-02

Pos logit contributions
 videos+0.022
isions+0.021
care+0.021
's+0.020
 mechanic+0.020
Neg logit contributions
 Fif-0.025
 bamboo-0.022
onna-0.021
iw-0.021
 pear-0.020
Token them
Token ID606
Feature activation-3.48e-02

Pos logit contributions
 videos+0.051
care+0.048
isions+0.048
's+0.047
 computers+0.047
Neg logit contributions
 Fif-0.066
onna-0.058
 bamboo-0.058
iw-0.056
sed-0.055
Token up
Token ID510
Feature activation+1.11e-01

Pos logit contributions
 Fif+0.057
 bamboo+0.050
onna+0.047
iw+0.047
 pear+0.045
Neg logit contributions
isions-0.055
 videos-0.055
care-0.054
 mechanic-0.053
 nephew-0.053
Token on
Token ID319
Feature activation-4.11e-02

Pos logit contributions
 videos+0.162
care+0.161
isions+0.159
's+0.156
 mechanic+0.156
Neg logit contributions
 Fif-0.196
 bamboo-0.173
onna-0.164
iw-0.163
sed-0.159
Token
Token ID198
Feature activation-2.83e-02

Pos logit contributions
 Fif+0.051
 bamboo+0.045
onna+0.043
iw+0.043
 pear+0.042
Neg logit contributions
isions-0.041
 videos-0.041
care-0.041
 mechanic-0.040
's-0.040
TokenBefore
Token ID8421
Feature activation+9.02e-02

Pos logit contributions
 Fif+0.051
 bamboo+0.044
onna+0.042
iw+0.041
io+0.040
Neg logit contributions
 videos-0.041
 computer-0.039
isions-0.039
 nephew-0.039
care-0.038
Token they
Token ID484
Feature activation+1.73e-01

Pos logit contributions
isions+0.130
 videos+0.124
controlled+0.124
 mechanic+0.123
's+0.122
Neg logit contributions
 Fif-0.165
 bamboo-0.145
onna-0.138
iw-0.135
 pear-0.134
Token left
Token ID1364
Feature activation-8.33e-02

Pos logit contributions
's+0.242
care+0.232
 mechanic+0.227
 videos+0.226
 nephew+0.225
Neg logit contributions
 Fif-0.321
 bamboo-0.293
iw-0.277
onna-0.275
balls-0.268
Token day
Token ID1110
Feature activation+2.33e-02

Pos logit contributions
 Fif+0.150
 bamboo+0.134
onna+0.127
iw+0.126
 pear+0.123
Neg logit contributions
 videos-0.118
isions-0.118
care-0.117
 mechanic-0.114
's-0.113
Tokencare
Token ID6651
Feature activation+4.11e-01

Pos logit contributions
isions+0.044
controlled+0.042
urious+0.041
 nephew+0.041
ographs+0.040
Neg logit contributions
 Fif-0.033
onna-0.027
 bamboo-0.027
 pear-0.025
 ti-0.024
Token,
Token ID11
Feature activation-4.48e-04

Pos logit contributions
isions+0.619
 videos+0.611
care+0.598
 mechanic+0.591
controlled+0.587
Neg logit contributions
 Fif-0.702
 bamboo-0.618
onna-0.597
iw-0.576
 pear-0.568
Token the
Token ID262
Feature activation+1.25e-02

Pos logit contributions
 Fif+0.001
 bamboo+0.001
 Flor+0.001
 Buzz+0.001
 pear+0.001
Neg logit contributions
isions-0.001
controlled-0.001
 keyboard-0.001
ever-0.001
 mechanic-0.001
Token teacher
Token ID4701
Feature activation-4.52e-02

Pos logit contributions
 videos+0.018
isions+0.018
 mechanic+0.018
care+0.018
 computer+0.017
Neg logit contributions
 Fif-0.023
 bamboo-0.020
onna-0.020
iw-0.019
io-0.018
Token gave
Token ID2921
Feature activation-2.07e-02

Pos logit contributions
 Fif+0.078
 bamboo+0.069
onna+0.067
iw+0.065
 pear+0.065
Neg logit contributions
isions-0.065
 videos-0.065
controlled-0.063
care-0.063
 keyboard-0.061
Token Amelia
Token ID49769
Feature activation+7.21e-02

Pos logit contributions
 Fif+0.038
 bamboo+0.034
onna+0.033
iw+0.032
 pear+0.031
Neg logit contributions
isions-0.029
 videos-0.028
care-0.028
's-0.028
 nephew-0.028
Token,
Token ID11
Feature activation+4.77e-04

Pos logit contributions
isions+0.052
 mechanic+0.050
 videos+0.050
 nephew+0.049
 keyboard+0.048
Neg logit contributions
 Fif-0.051
 bamboo-0.044
onna-0.041
 pear-0.041
iw-0.040
Token they
Token ID484
Feature activation+1.72e-01

Pos logit contributions
isions+0.001
 mechanic+0.001
 nephew+0.001
controlled+0.001
ographs+0.001
Neg logit contributions
 Fif-0.001
 bamboo-0.001
 pear-0.001
balls-0.001
onna-0.001
Token found
Token ID1043
Feature activation-1.15e-02

Pos logit contributions
isions+0.312
controlled+0.307
 videos+0.305
 keyboard+0.300
ographs+0.298
Neg logit contributions
 Fif-0.234
 pear-0.207
 bamboo-0.198
iw-0.188
ho-0.186
Token a
Token ID257
Feature activation+6.46e-02

Pos logit contributions
 Fif+0.021
 bamboo+0.019
 pear+0.018
sed+0.017
iw+0.017
Neg logit contributions
isions-0.017
care-0.016
 videos-0.016
 keyboard-0.015
 mechanic-0.015
Token blue
Token ID4171
Feature activation-8.98e-02

Pos logit contributions
isions+0.103
controlled+0.099
 videos+0.099
care+0.099
 means+0.097
Neg logit contributions
 Fif-0.106
 bamboo-0.097
 pear-0.093
onna-0.090
iw-0.088
Token computer
Token ID3644
Feature activation+4.10e-01

Pos logit contributions
 Fif+0.154
 bamboo+0.140
onna+0.130
iw+0.129
 pear+0.129
Neg logit contributions
isions-0.135
 videos-0.132
care-0.132
 nephew-0.128
controlled-0.128
Token in
Token ID287
Feature activation-2.59e-02

Pos logit contributions
isions+0.707
 nephew+0.698
 Margaret+0.678
 videos+0.677
care+0.670
Neg logit contributions
 Fif-0.579
 bamboo-0.533
onna-0.499
 pear-0.495
iw-0.480
Token their
Token ID511
Feature activation+1.41e-01

Pos logit contributions
 Fif+0.029
 bamboo+0.025
 Sara+0.023
 herself+0.023
 pear+0.022
Neg logit contributions
arre-0.056
isions-0.054
cases-0.054
ographs-0.053
pared-0.052
Token dad
Token ID9955
Feature activation+7.85e-02

Pos logit contributions
isions+0.201
care+0.194
ographs+0.193
controlled+0.192
 means+0.191
Neg logit contributions
 Fif-0.236
 bamboo-0.217
iw-0.212
sed-0.211
 pear-0.208
Token's
Token ID338
Feature activation+3.81e-01

Pos logit contributions
arre+0.065
 divorce+0.065
urious+0.061
 obstacle+0.060
 jet+0.060
Neg logit contributions
 Fif-0.164
 l-0.160
iw-0.157
 bamboo-0.155
 pear-0.155
Token office
Token ID2607
Feature activation+2.41e-01

Pos logit contributions
isions+0.557
ographs+0.549
lest+0.544
care+0.528
 means+0.527
Neg logit contributions
 Fif-0.572
sed-0.564
 bamboo-0.556
 belly-0.553
iw-0.541
Token them
Token ID606
Feature activation-4.58e-02

Pos logit contributions
isions+0.177
care+0.171
 mechanic+0.169
 videos+0.168
 nephew+0.168
Neg logit contributions
 Fif-0.168
 bamboo-0.147
onna-0.135
 pear-0.132
io-0.132
Token to
Token ID284
Feature activation-3.25e-02

Pos logit contributions
 Fif+0.064
 bamboo+0.055
onna+0.052
 pear+0.049
balls+0.049
Neg logit contributions
care-0.083
isions-0.083
ographs-0.081
 jet-0.081
 mechanic-0.081
Token take
Token ID1011
Feature activation+6.31e-02

Pos logit contributions
 Fif+0.056
 bamboo+0.050
onna+0.048
iw+0.047
 pear+0.046
Neg logit contributions
isions-0.048
 videos-0.047
care-0.047
 mechanic-0.046
 nephew-0.046
Token turns
Token ID4962
Feature activation+1.33e-01

Pos logit contributions
isions+0.104
urious+0.100
 keyboard+0.097
 nephew+0.096
 mechanic+0.096
Neg logit contributions
 Fif-0.097
 bamboo-0.089
io-0.079
 pear-0.078
 crown-0.077
Token watching
Token ID4964
Feature activation-4.22e-02

Pos logit contributions
urious+0.249
 nephew+0.240
isions+0.239
itionally+0.235
ographs+0.234
Neg logit contributions
 Fif-0.154
onna-0.150
balls-0.138
 bamboo-0.136
iw-0.121
Token videos
Token ID5861
Feature activation+4.10e-01

Pos logit contributions
 Fif+0.077
 bamboo+0.073
onna+0.067
 belly+0.066
 crown+0.066
Neg logit contributions
urious-0.053
isions-0.052
 nephew-0.049
ographs-0.048
itionally-0.047
Token.
Token ID13
Feature activation-9.22e-03

Pos logit contributions
isions+0.676
 nephew+0.654
 videos+0.645
 mechanic+0.639
care+0.633
Neg logit contributions
 Fif-0.618
 bamboo-0.562
onna-0.524
iw-0.502
 pear-0.500
Token He
Token ID679
Feature activation-2.89e-02

Pos logit contributions
 Fif+0.014
 bamboo+0.012
onna+0.012
iw+0.011
 pear+0.011
Neg logit contributions
isions-0.016
 videos-0.015
care-0.015
 jet-0.015
 nephew-0.015
Token says
Token ID1139
Feature activation+8.58e-02

Pos logit contributions
 Fif+0.049
 bamboo+0.044
onna+0.043
sed+0.041
 pear+0.041
Neg logit contributions
care-0.043
isions-0.042
controlled-0.040
 videos-0.039
 nephew-0.039
Token he
Token ID339
Feature activation+2.07e-02

Pos logit contributions
isions+0.116
 videos+0.114
care+0.113
 mechanic+0.110
 nephew+0.110
Neg logit contributions
 Fif-0.161
 bamboo-0.145
onna-0.138
iw-0.136
 pear-0.134
Token will
Token ID481
Feature activation-9.90e-02

Pos logit contributions
care+0.039
isions+0.038
 jet+0.037
 videos+0.037
controlled+0.036
Neg logit contributions
 Fif-0.027
 bamboo-0.023
onna-0.022
sed-0.022
iw-0.022
Token person
Token ID1048
Feature activation+1.62e-01

Pos logit contributions
 Fif+0.010
 bamboo+0.009
sed+0.009
 majesty+0.008
iw+0.008
Neg logit contributions
isions-0.007
care-0.007
controlled-0.007
cases-0.007
ographs-0.007
Token.
Token ID13
Feature activation-6.23e-04

Pos logit contributions
 videos+0.201
isions+0.200
care+0.199
's+0.199
 mechanic+0.193
Neg logit contributions
 Fif-0.320
 bamboo-0.288
onna-0.278
iw-0.273
sed-0.267
Token She
Token ID1375
Feature activation-9.98e-03

Pos logit contributions
 Fif+0.001
 bamboo+0.001
onna+0.001
iw+0.001
 pear+0.001
Neg logit contributions
isions-0.001
 videos-0.001
care-0.001
 mechanic-0.001
's-0.001
Token is
Token ID318
Feature activation-4.23e-02

Pos logit contributions
 Fif+0.017
 bamboo+0.015
iw+0.014
onna+0.014
 pear+0.013
Neg logit contributions
isions-0.016
care-0.016
 videos-0.015
 keyboard-0.015
 mechanic-0.015
Token a
Token ID257
Feature activation+6.57e-02

Pos logit contributions
 Fif+0.077
 bamboo+0.069
onna+0.065
iw+0.065
 pear+0.063
Neg logit contributions
 videos-0.060
isions-0.058
care-0.058
's-0.057
 mechanic-0.057
Token computer
Token ID3644
Feature activation+4.09e-01

Pos logit contributions
 mechanic+0.089
 videos+0.088
isions+0.086
care+0.086
 nephew+0.086
Neg logit contributions
 Fif-0.123
 bamboo-0.110
onna-0.106
iw-0.104
io-0.102
Token program
Token ID1430
Feature activation+2.51e-01

Pos logit contributions
 videos+0.569
's+0.561
isions+0.560
 mechanic+0.555
care+0.550
Neg logit contributions
 Fif-0.761
 bamboo-0.667
iw-0.637
onna-0.635
sed-0.621
Token.
Token ID13
Feature activation-3.69e-05

Pos logit contributions
 videos+0.332
isions+0.323
's+0.323
care+0.321
 mechanic+0.318
Neg logit contributions
 Fif-0.486
 bamboo-0.432
onna-0.413
iw-0.412
 pear-0.402
Token She
Token ID1375
Feature activation-9.50e-03

Pos logit contributions
 Fif+0.000
 bamboo+0.000
onna+0.000
iw+0.000
 pear+0.000
Neg logit contributions
isions-0.000
 videos-0.000
care-0.000
 mechanic-0.000
 nephew-0.000
Token can
Token ID460
Feature activation-6.71e-02

Pos logit contributions
 Fif+0.014
 bamboo+0.012
iw+0.012
onna+0.012
sed+0.011
Neg logit contributions
isions-0.016
care-0.016
 videos-0.016
 keyboard-0.015
 computer-0.015
Token't
Token ID470
Feature activation+2.03e-02

Pos logit contributions
 Fif+0.120
 bamboo+0.104
onna+0.103
 pear+0.096
iw+0.095
Neg logit contributions
isions-0.099
care-0.096
 videos-0.093
controlled-0.093
 nephew-0.091
Token TV
Token ID3195
Feature activation+3.30e-01

Pos logit contributions
 Fif+0.024
 bamboo+0.022
iw+0.022
onna+0.022
io+0.021
Neg logit contributions
's-0.011
 videos-0.011
care-0.011
 mechanic-0.010
isions-0.010
Token.
Token ID13
Feature activation+3.64e-02

Pos logit contributions
isions+0.511
 videos+0.503
care+0.497
 mechanic+0.490
 nephew+0.490
Neg logit contributions
 Fif-0.539
 bamboo-0.481
onna-0.460
iw-0.448
 pear-0.440
Token They
Token ID1119
Feature activation+1.33e-01

Pos logit contributions
isions+0.053
 videos+0.052
care+0.051
controlled+0.050
 nephew+0.050
Neg logit contributions
 Fif-0.065
 bamboo-0.057
onna-0.054
iw-0.053
 pear-0.052
Token had
Token ID550
Feature activation-2.46e-02

Pos logit contributions
isions+0.213
care+0.208
 videos+0.208
 mechanic+0.201
 nephew+0.201
Neg logit contributions
 Fif-0.214
 bamboo-0.188
onna-0.179
 pear-0.174
iw-0.173
Token many
Token ID867
Feature activation-1.54e-02

Pos logit contributions
 Fif+0.041
 bamboo+0.037
onna+0.035
iw+0.034
 pear+0.034
Neg logit contributions
isions-0.037
 videos-0.036
care-0.036
 mechanic-0.035
 nephew-0.035
Token videos
Token ID5861
Feature activation+4.09e-01

Pos logit contributions
 Fif+0.024
 bamboo+0.023
 pear+0.020
iw+0.020
onna+0.020
Neg logit contributions
care-0.024
isions-0.024
 mechanic-0.023
controlled-0.022
 Margaret-0.022
Token of
Token ID286
Feature activation+8.33e-02

Pos logit contributions
isions+0.486
 videos+0.486
care+0.475
 mechanic+0.461
's+0.460
Neg logit contributions
 Fif-0.832
 bamboo-0.750
onna-0.720
iw-0.713
 pear-0.697
Token cartoons
Token ID30070
Feature activation+2.39e-01

Pos logit contributions
isions+0.110
care+0.108
 videos+0.108
 mechanic+0.105
 nephew+0.104
Neg logit contributions
 Fif-0.158
 bamboo-0.142
onna-0.134
iw-0.133
 pear-0.131
Token,
Token ID11
Feature activation-2.67e-03

Pos logit contributions
isions+0.332
 videos+0.329
care+0.324
 mechanic+0.317
's+0.315
Neg logit contributions
 Fif-0.437
 bamboo-0.390
onna-0.372
iw-0.368
 pear-0.359
Token animals
Token ID4695
Feature activation+7.79e-03

Pos logit contributions
 Fif+0.005
 bamboo+0.004
onna+0.004
 pear+0.004
iw+0.004
Neg logit contributions
isions-0.004
care-0.004
 mechanic-0.004
controlled-0.004
 nephew-0.003
Token and
Token ID290
Feature activation+1.04e-02

Pos logit contributions
isions+0.010
 videos+0.010
care+0.010
's+0.010
 mechanic+0.010
Neg logit contributions
 Fif-0.015
 bamboo-0.013
onna-0.012
iw-0.012
 pear-0.012
Token They
Token ID1119
Feature activation+1.11e-01

Pos logit contributions
 Fif+0.010
 bamboo+0.009
onna+0.009
iw+0.008
 pear+0.008
Neg logit contributions
isions-0.007
 videos-0.006
care-0.006
 nephew-0.006
 mechanic-0.006
Token can
Token ID460
Feature activation-7.74e-02

Pos logit contributions
 videos+0.136
isions+0.135
's+0.133
care+0.132
 mechanic+0.130
Neg logit contributions
 Fif-0.220
 bamboo-0.200
iw-0.191
onna-0.191
io-0.186
Token't
Token ID470
Feature activation+1.00e-02

Pos logit contributions
 Fif+0.141
 bamboo+0.125
onna+0.120
iw+0.118
 pear+0.115
Neg logit contributions
isions-0.110
 videos-0.108
care-0.107
 mechanic-0.105
 nephew-0.104
Token watch
Token ID2342
Feature activation-1.17e-04

Pos logit contributions
isions+0.015
care+0.015
 videos+0.015
 nephew+0.014
 computers+0.014
Neg logit contributions
 Fif-0.017
 bamboo-0.015
 pear-0.014
onna-0.014
balls-0.014
Token any
Token ID597
Feature activation+8.88e-02

Pos logit contributions
 Fif+0.000
 bamboo+0.000
balls+0.000
onna+0.000
 pear+0.000
Neg logit contributions
isions-0.000
urious-0.000
 nephew-0.000
old-0.000
care-0.000
Token videos
Token ID5861
Feature activation+4.09e-01

Pos logit contributions
grand+0.097
isions+0.096
 Margaret+0.096
limits+0.095
 mechanic+0.095
Neg logit contributions
 bamboo-0.161
balls-0.160
 Fif-0.152
 l-0.149
 honey-0.146
Token.
Token ID13
Feature activation-3.66e-03

Pos logit contributions
isions+0.630
 videos+0.617
care+0.608
 nephew+0.604
 mechanic+0.602
Neg logit contributions
 Fif-0.684
 bamboo-0.610
onna-0.578
iw-0.565
 pear-0.557
Token They
Token ID1119
Feature activation+1.12e-01

Pos logit contributions
 Fif+0.007
 bamboo+0.006
onna+0.006
iw+0.006
sed+0.006
Neg logit contributions
isions-0.005
 videos-0.005
 nephew-0.005
care-0.005
 mechanic-0.005
Token wish
Token ID4601
Feature activation+1.14e-01

Pos logit contributions
 videos+0.148
's+0.147
isions+0.146
care+0.142
 mechanic+0.141
Neg logit contributions
 Fif-0.212
 bamboo-0.191
iw-0.184
onna-0.183
io-0.180
Token they
Token ID484
Feature activation+1.66e-01

Pos logit contributions
 videos+0.081
's+0.068
 computers+0.065
 jet+0.064
 computer+0.064
Neg logit contributions
 Fif-0.298
iw-0.283
 bamboo-0.278
onna-0.264
 pear-0.263
Token had
Token ID550
Feature activation-4.15e-02

Pos logit contributions
 videos+0.296
isions+0.296
care+0.293
 nephew+0.288
 mechanic+0.286
Neg logit contributions
 Fif-0.236
 bamboo-0.205
onna-0.194
 pear-0.190
iw-0.184
Token,
Token ID11
Feature activation+3.12e-03

Pos logit contributions
isions+0.329
 videos+0.323
care+0.320
 nephew+0.315
 mechanic+0.314
Neg logit contributions
 Fif-0.373
 bamboo-0.336
onna-0.320
iw-0.312
 pear-0.308
Token a
Token ID257
Feature activation+6.64e-02

Pos logit contributions
isions+0.005
urious+0.005
care+0.005
arre+0.005
 mechanic+0.005
Neg logit contributions
 Fif-0.005
 bamboo-0.005
 pear-0.004
 ti-0.004
onna-0.004
Token phone
Token ID3072
Feature activation+2.14e-01

Pos logit contributions
isions+0.093
 videos+0.091
care+0.090
controlled+0.087
 mechanic+0.087
Neg logit contributions
 Fif-0.121
 bamboo-0.109
onna-0.104
iw-0.102
 pear-0.101
Token,
Token ID11
Feature activation+3.19e-03

Pos logit contributions
isions+0.342
care+0.336
 videos+0.325
urious+0.324
controlled+0.324
Neg logit contributions
 Fif-0.335
 bamboo-0.298
onna-0.282
sed-0.281
balls-0.270
Token a
Token ID257
Feature activation+6.66e-02

Pos logit contributions
isions+0.005
care+0.005
 mechanic+0.005
urious+0.005
controlled+0.005
Neg logit contributions
 Fif-0.005
 bamboo-0.005
 pear-0.004
onna-0.004
sed-0.004
Token computer
Token ID3644
Feature activation+4.09e-01

Pos logit contributions
isions+0.093
 videos+0.089
care+0.089
controlled+0.088
ographs+0.085
Neg logit contributions
 Fif-0.120
 bamboo-0.109
onna-0.104
 pear-0.103
iw-0.102
Token and
Token ID290
Feature activation+1.83e-02

Pos logit contributions
isions+0.616
 videos+0.606
care+0.602
 mechanic+0.587
 nephew+0.586
Neg logit contributions
 Fif-0.695
 bamboo-0.620
onna-0.589
iw-0.578
 pear-0.567
Token some
Token ID617
Feature activation+1.44e-02

Pos logit contributions
arse+0.037
urious+0.037
itionally+0.036
arre+0.036
jo+0.035
Neg logit contributions
 Fif-0.020
 bamboo-0.019
 bananas-0.018
 crown-0.018
 ti-0.017
Token papers
Token ID9473
Feature activation+1.54e-01

Pos logit contributions
isions+0.020
care+0.019
itionally+0.019
urious+0.019
arse+0.019
Neg logit contributions
 Fif-0.024
 bamboo-0.023
 pear-0.021
 crown-0.021
 ti-0.021
Token.
Token ID13
Feature activation+3.47e-03

Pos logit contributions
isions+0.240
 videos+0.236
care+0.234
 mechanic+0.231
 nephew+0.230
Neg logit contributions
 Fif-0.252
 bamboo-0.224
onna-0.213
iw-0.208
 pear-0.204
Token
Token ID198
Feature activation-2.66e-02

Pos logit contributions
isions+0.005
 videos+0.005
care+0.005
's+0.005
 mechanic+0.005
Neg logit contributions
 Fif-0.006
 bamboo-0.006
onna-0.005
iw-0.005
 pear-0.005
Token he
Token ID339
Feature activation+3.17e-02

Pos logit contributions
 Fif+0.121
 bamboo+0.097
 pear+0.093
 crown+0.091
balls+0.090
Neg logit contributions
isions-0.185
controlled-0.175
arre-0.169
 divorce-0.168
care-0.168
Token would
Token ID561
Feature activation-2.84e-02

Pos logit contributions
isions+0.050
controlled+0.048
 videos+0.048
care+0.048
ographs+0.046
Neg logit contributions
 Fif-0.049
onna-0.044
 bamboo-0.043
 ti-0.042
 pear-0.042
Token listen
Token ID6004
Feature activation+3.55e-02

Pos logit contributions
 Fif+0.049
onna+0.043
 bamboo+0.043
 pear+0.041
 ti+0.041
Neg logit contributions
isions-0.041
care-0.040
controlled-0.039
 videos-0.038
 nephew-0.038
Token to
Token ID284
Feature activation-2.07e-02

Pos logit contributions
isions+0.063
 nephew+0.062
 mechanic+0.061
 jet+0.060
 videos+0.060
Neg logit contributions
 Fif-0.048
 bamboo-0.042
onna-0.042
balls-0.039
 ti-0.037
Token the
Token ID262
Feature activation+1.62e-02

Pos logit contributions
 Fif+0.029
 bamboo+0.025
onna+0.023
 pear+0.023
 crown+0.023
Neg logit contributions
isions-0.037
care-0.036
ographs-0.035
controlled-0.035
cases-0.034
Token videos
Token ID5861
Feature activation+4.08e-01

Pos logit contributions
isions+0.020
care+0.019
controlled+0.019
 Margaret+0.018
cases+0.018
Neg logit contributions
 Fif-0.030
 bamboo-0.028
balls-0.026
sed-0.026
onna-0.026
Token.
Token ID13
Feature activation+1.44e-03

Pos logit contributions
isions+0.633
 videos+0.614
care+0.605
 nephew+0.604
controlled+0.597
Neg logit contributions
 Fif-0.668
 bamboo-0.601
onna-0.571
iw-0.554
 pear-0.551
Token He
Token ID679
Feature activation-2.36e-02

Pos logit contributions
 divorce+0.003
 jet+0.003
isions+0.003
 cancelled+0.003
controlled+0.003
Neg logit contributions
 Fif-0.002
onna-0.001
 belly-0.001
 Frog-0.001
io-0.001
Token thought
Token ID1807
Feature activation-7.65e-02

Pos logit contributions
 Fif+0.036
onna+0.031
 bamboo+0.031
sed+0.030
 pear+0.030
Neg logit contributions
isions-0.038
care-0.038
 videos-0.037
controlled-0.036
ographs-0.035
Token they
Token ID484
Feature activation+1.75e-01

Pos logit contributions
 Fif+0.123
 bamboo+0.111
onna+0.109
 pear+0.103
balls+0.101
Neg logit contributions
isions-0.122
 nephew-0.114
care-0.114
urious-0.113
ographs-0.112
Token sounded
Token ID14846
Feature activation+2.33e-01

Pos logit contributions
care+0.347
isions+0.335
 mechanic+0.332
 videos+0.332
 nephew+0.331
Neg logit contributions
 Fif-0.192
 bamboo-0.163
 pear-0.159
 ti-0.154
 crown-0.150
Token loved
Token ID6151
Feature activation-2.14e-02

Pos logit contributions
 Fif+0.037
 bamboo+0.034
onna+0.033
iw+0.032
balls+0.031
Neg logit contributions
's-0.022
 means-0.021
care-0.020
 mechanic-0.020
 nephew-0.020
Token to
Token ID284
Feature activation-1.84e-02

Pos logit contributions
 Fif+0.037
 bamboo+0.032
onna+0.031
iw+0.030
 pear+0.030
Neg logit contributions
isions-0.032
 videos-0.032
care-0.032
 mechanic-0.031
 nephew-0.031
Token play
Token ID711
Feature activation+2.63e-02

Pos logit contributions
 Fif+0.034
 bamboo+0.030
onna+0.029
iw+0.028
 pear+0.028
Neg logit contributions
isions-0.026
care-0.025
 videos-0.025
 mechanic-0.025
controlled-0.025
Token with
Token ID351
Feature activation+1.25e-02

Pos logit contributions
isions+0.040
 videos+0.039
care+0.038
 nephew+0.038
 mechanic+0.038
Neg logit contributions
 Fif-0.045
 bamboo-0.041
onna-0.038
iw-0.038
 pear-0.037
Token her
Token ID607
Feature activation-3.99e-02

Pos logit contributions
isions+0.021
 videos+0.020
care+0.020
 mechanic+0.020
 nephew+0.020
Neg logit contributions
 Fif-0.020
 bamboo-0.017
onna-0.016
iw-0.016
 pear-0.016
Token computer
Token ID3644
Feature activation+4.08e-01

Pos logit contributions
 Fif+0.074
 bamboo+0.066
onna+0.064
iw+0.062
 pear+0.060
Neg logit contributions
 videos-0.055
isions-0.054
care-0.053
 mechanic-0.053
 nephew-0.053
Token.
Token ID13
Feature activation+5.44e-03

Pos logit contributions
isions+0.599
 videos+0.598
care+0.587
 mechanic+0.577
's+0.576
Neg logit contributions
 Fif-0.716
 bamboo-0.632
onna-0.602
iw-0.597
 pear-0.579
Token She
Token ID1375
Feature activation-7.39e-03

Pos logit contributions
isions+0.007
 videos+0.007
care+0.007
 mechanic+0.007
 nephew+0.007
Neg logit contributions
 Fif-0.010
 bamboo-0.009
onna-0.009
iw-0.009
 pear-0.009
Token had
Token ID550
Feature activation-2.72e-02

Pos logit contributions
 Fif+0.014
 bamboo+0.013
iw+0.012
onna+0.012
 pear+0.012
Neg logit contributions
's-0.010
 mechanic-0.010
 videos-0.009
 computers-0.009
 computer-0.009
Token many
Token ID867
Feature activation-1.08e-02

Pos logit contributions
 Fif+0.055
 bamboo+0.050
onna+0.047
iw+0.047
io+0.046
Neg logit contributions
 videos-0.034
's-0.034
isions-0.033
 computers-0.033
care-0.033
Token games
Token ID1830
Feature activation+3.24e-02

Pos logit contributions
 Fif+0.015
 bamboo+0.014
 pear+0.013
 bananas+0.013
iw+0.013
Neg logit contributions
isions-0.017
 mechanic-0.017
care-0.017
 Margaret-0.017
 divorce-0.017
Token was
Token ID373
Feature activation-2.42e-02

Pos logit contributions
 Fif+0.178
 bamboo+0.165
iw+0.162
onna+0.158
 majesty+0.148
Neg logit contributions
's-0.127
isions-0.111
 means-0.110
 videos-0.110
care-0.106
Token very
Token ID845
Feature activation-6.72e-02

Pos logit contributions
iw+0.052
 Fif+0.051
ho+0.049
io+0.049
onna+0.049
Neg logit contributions
's-0.026
 videos-0.022
-0.021
 Margaret-0.020
 means-0.020
Token happy
Token ID3772
Feature activation-5.34e-02

Pos logit contributions
 Fif+0.128
 bamboo+0.121
iw+0.115
io+0.114
onna+0.113
Neg logit contributions
's-0.082
care-0.080
 videos-0.079
isions-0.076
 means-0.076
Token to
Token ID284
Feature activation-3.78e-02

Pos logit contributions
 Fif+0.121
 bamboo+0.109
iw+0.107
io+0.106
sed+0.106
Neg logit contributions
care-0.057
's-0.055
 videos-0.053
 computer-0.047
 means-0.047
Token meet
Token ID1826
Feature activation-5.64e-02

Pos logit contributions
iw+0.096
 bamboo+0.092
 Fif+0.091
 strawberries+0.090
sed+0.089
Neg logit contributions
 pray-0.027
's-0.026
 videos-0.024
 send-0.022
 press-0.022
Token Fif
Token ID12830
Feature activation-4.92e-01

Pos logit contributions
onna+0.095
 majesty+0.090
sed+0.089
iw+0.083
 ti+0.081
Neg logit contributions
's-0.084
 videos-0.080
 means-0.079
 jet-0.075
 Margaret-0.074
Tokeni
Token ID72
Feature activation-7.34e-02

Pos logit contributions
 Fif+0.895
 ti+0.856
 bamboo+0.850
iw+0.813
 pear+0.810
Neg logit contributions
's-0.599
 means-0.555
ations-0.552
urious-0.545
care-0.530
Token.
Token ID13
Feature activation-9.26e-03

Pos logit contributions
 Fif+0.131
 bamboo+0.117
onna+0.112
iw+0.110
 pear+0.108
Neg logit contributions
isions-0.105
 videos-0.104
care-0.102
 mechanic-0.100
's-0.100
Token
Token ID198
Feature activation-4.39e-02

Pos logit contributions
 crown+0.021
 ti+0.021
 pearl+0.021
 elbow+0.021
castle+0.020
Neg logit contributions
's-0.008
 means-0.007
-0.006
 So-0.006
s-0.006
Token
Token ID198
Feature activation-4.37e-02

Pos logit contributions
 Fif+0.088
sed+0.087
 majesty+0.084
iw+0.083
 bamboo+0.082
Neg logit contributions
's-0.046
care-0.043
During-0.042
 means-0.041
 videos-0.040
TokenF
Token ID37
Feature activation-2.46e-01

Pos logit contributions
sed+0.066
 Fif+0.064
onna+0.063
 majesty+0.063
iw+0.059
Neg logit contributions
During-0.067
's-0.066
-0.063
care-0.063
isions-0.063
Token "
Token ID366
Feature activation-9.30e-02

Pos logit contributions
 Fif+0.009
 bamboo+0.008
iw+0.008
onna+0.007
io+0.007
Neg logit contributions
 videos-0.008
's-0.008
care-0.008
isions-0.008
 mechanic-0.008
TokenHi
Token ID17250
Feature activation-9.83e-02

Pos logit contributions
 Fif+0.153
 bamboo+0.135
onna+0.130
sed+0.125
iw+0.124
Neg logit contributions
isions-0.147
care-0.145
 videos-0.144
's-0.143
 mechanic-0.142
Token,
Token ID11
Feature activation-7.45e-03

Pos logit contributions
 Fif+0.183
onna+0.170
iw+0.165
 bamboo+0.164
sed+0.162
Neg logit contributions
's-0.128
care-0.121
 videos-0.121
isions-0.119
 means-0.116
Token I
Token ID314
Feature activation-9.99e-02

Pos logit contributions
 Fif+0.010
onna+0.010
iw+0.009
io+0.009
sed+0.009
Neg logit contributions
's-0.013
 videos-0.013
 mechanic-0.012
 jet-0.012
 computer-0.012
Token am
Token ID716
Feature activation-1.13e-01

Pos logit contributions
 Fif+0.161
 bamboo+0.141
onna+0.134
iw+0.131
sed+0.127
Neg logit contributions
isions-0.162
 videos-0.161
care-0.160
's-0.158
 mechanic-0.157
Token Fif
Token ID12830
Feature activation-4.92e-01

Pos logit contributions
 Fif+0.188
onna+0.171
 bamboo+0.171
iw+0.169
io+0.163
Neg logit contributions
 videos-0.168
's-0.164
care-0.162
 mechanic-0.157
 Margaret-0.157
Tokeni
Token ID72
Feature activation-7.20e-02

Pos logit contributions
 Fif+0.857
 bamboo+0.819
 ti+0.792
onna+0.772
 pear+0.757
Neg logit contributions
's-0.647
 means-0.602
care-0.584
 videos-0.577
isions-0.576
Token the
Token ID262
Feature activation+4.08e-03

Pos logit contributions
 Fif+0.134
 bamboo+0.119
onna+0.119
iw+0.113
sed+0.112
Neg logit contributions
's-0.102
 videos-0.092
isions-0.091
 nephew-0.090
care-0.089
Token frog
Token ID21264
Feature activation-3.27e-01

Pos logit contributions
 mechanic+0.006
 jet+0.006
's+0.006
 radio+0.006
 nephew+0.006
Neg logit contributions
onna-0.007
io-0.006
 Fif-0.006
iw-0.006
edly-0.006
Token.
Token ID13
Feature activation-5.67e-03

Pos logit contributions
 Fif+0.610
onna+0.531
 bamboo+0.530
sed+0.518
io+0.509
Neg logit contributions
's-0.473
 means-0.421
 videos-0.414
care-0.408
isions-0.404
Token I
Token ID314
Feature activation-1.01e-01

Pos logit contributions
io+0.010
 Fif+0.010
onna+0.010
 bamboo+0.010
iw+0.010
Neg logit contributions
's-0.007
-0.006
 videos-0.006
 Margaret-0.006
 means-0.006
Token and
Token ID290
Feature activation+4.96e-03

Pos logit contributions
 Fif+0.216
 majesty+0.207
onna+0.204
iw+0.203
sed+0.196
Neg logit contributions
's-0.085
 means-0.074
 videos-0.065
care-0.062
 computer-0.057
Token pretty
Token ID2495
Feature activation-5.55e-02

Pos logit contributions
's+0.006
 Margaret+0.005
 complicated+0.005
 jet+0.005
 videos+0.005
Neg logit contributions
io-0.010
 majesty-0.009
onna-0.009
 Fif-0.009
iw-0.009
Token.
Token ID13
Feature activation-6.96e-03

Pos logit contributions
 Fif+0.125
 majesty+0.120
onna+0.120
iw+0.120
sed+0.114
Neg logit contributions
's-0.058
 means-0.047
care-0.043
 videos-0.042
isions-0.041
Token Lily
Token ID20037
Feature activation-8.97e-02

Pos logit contributions
 crown+0.008
io+0.008
 ti+0.007
onna+0.007
castle+0.007
Neg logit contributions
's-0.013
 means-0.013
 Margaret-0.012
care-0.012
-0.012
Token and
Token ID290
Feature activation+4.05e-03

Pos logit contributions
 Fif+0.186
 bamboo+0.174
iw+0.169
onna+0.165
io+0.158
Neg logit contributions
's-0.119
 means-0.096
isions-0.094
care-0.094
 Margaret-0.093
Token Fif
Token ID12830
Feature activation-4.88e-01

Pos logit contributions
 means+0.009
's+0.009
 nephew+0.009
 Margaret+0.009
 videos+0.008
Neg logit contributions
onna-0.004
iw-0.003
 ti-0.003
io-0.003
 majesty-0.003
Tokeni
Token ID72
Feature activation-6.45e-02

Pos logit contributions
 Fif+0.892
 ti+0.889
onna+0.878
 bamboo+0.867
iw+0.851
Neg logit contributions
's-0.575
urious-0.528
 means-0.519
ations-0.503
itionally-0.489
Token played
Token ID2826
Feature activation-9.18e-02

Pos logit contributions
 Fif+0.119
onna+0.110
 bamboo+0.109
iw+0.108
 majesty+0.101
Neg logit contributions
's-0.094
 jet-0.083
isions-0.079
 means-0.079
 videos-0.079
Token in
Token ID287
Feature activation-3.35e-02

Pos logit contributions
iw+0.208
onna+0.205
 Fif+0.196
 majesty+0.192
ho+0.190
Neg logit contributions
's-0.089
 videos-0.087
 films-0.079
 radio-0.076
 computer-0.074
Token the
Token ID262
Feature activation-8.56e-04

Pos logit contributions
 Fif+0.070
onna+0.070
 majesty+0.069
io+0.068
sed+0.067
Neg logit contributions
's-0.034
 jet-0.032
 videos-0.032
 radio-0.031
 means-0.031
Token fog
Token ID19558
Feature activation-3.35e-02

Pos logit contributions
 Fif+0.002
onna+0.002
iw+0.002
io+0.002
sed+0.001
Neg logit contributions
 jet-0.001
 radio-0.001
's-0.001
 automobile-0.001
 mechanic-0.001
Token ch
Token ID442
Feature activation-9.07e-02

Pos logit contributions
isions+0.061
 videos+0.060
care+0.059
controlled+0.057
 mechanic+0.056
Neg logit contributions
 Fif-0.126
 bamboo-0.115
onna-0.110
iw-0.110
 pear-0.108
Tokenubby
Token ID38393
Feature activation+3.31e-02

Pos logit contributions
 Fif+0.186
 bamboo+0.167
onna+0.163
iw+0.159
sed+0.157
Neg logit contributions
 videos-0.105
 mechanic-0.103
isions-0.101
care-0.101
 jet-0.101
Token little
Token ID1310
Feature activation+1.41e-01

Pos logit contributions
 mechanic+0.039
 videos+0.038
 jet+0.038
 nephew+0.037
 computer+0.037
Neg logit contributions
 Fif-0.066
onna-0.060
 bamboo-0.059
iw-0.059
io-0.057
Token fish
Token ID5916
Feature activation-1.42e-01

Pos logit contributions
 videos+0.175
 mechanic+0.172
 jet+0.170
isions+0.169
 nephew+0.169
Neg logit contributions
 Fif-0.276
onna-0.246
 bamboo-0.246
iw-0.235
io-0.233
Token named
Token ID3706
Feature activation-1.24e-03

Pos logit contributions
 Fif+0.271
 bamboo+0.243
onna+0.234
iw+0.230
sed+0.225
Neg logit contributions
isions-0.183
 videos-0.183
care-0.181
's-0.178
 mechanic-0.177
Token Fif
Token ID12830
Feature activation-4.88e-01

Pos logit contributions
 Fif+0.002
 bamboo+0.001
iw+0.001
onna+0.001
 pear+0.001
Neg logit contributions
 videos-0.002
isions-0.002
care-0.002
's-0.002
 mechanic-0.002
Tokeni
Token ID72
Feature activation-6.55e-02

Pos logit contributions
 Fif+0.866
 bamboo+0.818
 ti+0.806
balls+0.794
iw+0.787
Neg logit contributions
's-0.627
urious-0.579
ations-0.559
 means-0.558
 Street-0.537
Token.
Token ID13
Feature activation-7.20e-03

Pos logit contributions
 Fif+0.127
 bamboo+0.115
onna+0.112
iw+0.109
sed+0.107
Neg logit contributions
's-0.086
 nephew-0.079
 videos-0.079
 mechanic-0.078
isions-0.078
Token She
Token ID1375
Feature activation-2.38e-02

Pos logit contributions
bled+0.008
 majesty+0.008
 elbow+0.008
kin+0.008
 ti+0.007
Neg logit contributions
 means-0.014
 On-0.014
's-0.013
 One-0.013
 They-0.013
Token lived
Token ID5615
Feature activation-3.11e-02

Pos logit contributions
 Fif+0.049
iw+0.047
onna+0.045
 bamboo+0.044
io+0.043
Neg logit contributions
's-0.028
 videos-0.024
 means-0.024
 jet-0.023
 mechanic-0.022
Token in
Token ID287
Feature activation-3.67e-02

Pos logit contributions
 majesty+0.076
onna+0.075
 seventeen+0.074
sed+0.072
drops+0.072
Neg logit contributions
 late-0.021
's-0.020
care-0.018
 who-0.015
 when-0.015
Tokena
Token ID64
Feature activation+1.47e-01

Pos logit contributions
 Fif+0.078
 bamboo+0.071
onna+0.070
iw+0.068
sed+0.068
Neg logit contributions
care-0.023
's-0.022
isions-0.022
 mechanic-0.021
 videos-0.020
Token named
Token ID3706
Feature activation+2.56e-04

Pos logit contributions
's+0.181
 videos+0.177
care+0.171
 mechanic+0.171
isions+0.170
Neg logit contributions
 Fif-0.292
onna-0.262
 bamboo-0.262
sed-0.252
iw-0.251
Token Fif
Token ID12830
Feature activation-4.86e-01

Pos logit contributions
 videos+0.000
isions+0.000
care+0.000
's+0.000
 mechanic+0.000
Neg logit contributions
 Fif-0.000
 bamboo-0.000
iw-0.000
onna-0.000
io-0.000
Tokeni
Token ID72
Feature activation-6.01e-02

Pos logit contributions
 Fif+0.850
 bamboo+0.798
balls+0.781
 ti+0.777
iw+0.774
Neg logit contributions
's-0.650
urious-0.601
 means-0.577
ations-0.565
isha-0.557
Token.
Token ID13
Feature activation-1.08e-02

Pos logit contributions
 Fif+0.115
 bamboo+0.105
onna+0.102
iw+0.100
sed+0.097
Neg logit contributions
's-0.079
 videos-0.074
 nephew-0.074
 mechanic-0.073
isions-0.072
Token Fif
Token ID12830
Feature activation-4.86e-01

Pos logit contributions
bled+0.012
castle+0.011
 majesty+0.011
 elbow+0.011
io+0.011
Neg logit contributions
 means-0.021
 On-0.021
's-0.021
 One-0.020
 During-0.019
Tokeni
Token ID72
Feature activation-6.02e-02

Pos logit contributions
 Fif+0.873
balls+0.865
 ti+0.844
 bamboo+0.835
onna+0.829
Neg logit contributions
's-0.578
urious-0.562
 means-0.552
ations-0.530
jo-0.493
Token lived
Token ID5615
Feature activation-3.20e-02

Pos logit contributions
 Fif+0.114
 bamboo+0.104
onna+0.100
iw+0.098
sed+0.095
Neg logit contributions
's-0.082
 videos-0.077
isions-0.076
 mechanic-0.075
 means-0.075
Token in
Token ID287
Feature activation-3.62e-02

Pos logit contributions
 majesty+0.080
 seventeen+0.078
onna+0.078
io+0.077
 Fif+0.075
Neg logit contributions
 late-0.022
care-0.020
's-0.020
 when-0.017
 who-0.014
Token a
Token ID257
Feature activation+5.72e-02

Pos logit contributions
 Fif+0.073
 bamboo+0.066
onna+0.064
iw+0.063
sed+0.062
Neg logit contributions
 videos-0.043
isions-0.042
care-0.042
 mechanic-0.041
's-0.041
Token big
Token ID1263
Feature activation+4.39e-03

Pos logit contributions
isions+0.081
 videos+0.079
care+0.078
 nephew+0.076
controlled+0.076
Neg logit contributions
 Fif-0.105
 bamboo-0.094
onna-0.089
iw-0.088
 pear-0.087
Token a
Token ID257
Feature activation+5.74e-02

Pos logit contributions
 Fif+0.030
 bamboo+0.027
 pear+0.025
iw+0.025
onna+0.024
Neg logit contributions
isions-0.023
 videos-0.022
care-0.022
 mechanic-0.022
 nephew-0.021
Token unique
Token ID3748
Feature activation+3.21e-02

Pos logit contributions
isions+0.062
 videos+0.061
care+0.060
 mechanic+0.058
 nephew+0.058
Neg logit contributions
 Fif-0.123
 bamboo-0.112
onna-0.107
iw-0.106
 pear-0.104
Token fle
Token ID5104
Feature activation-3.19e-02

Pos logit contributions
 mechanic+0.044
 videos+0.043
 jet+0.043
 computer+0.043
 nephew+0.041
Neg logit contributions
 Fif-0.060
onna-0.053
 bamboo-0.052
iw-0.051
io-0.051
Tokena
Token ID64
Feature activation+1.47e-01

Pos logit contributions
 Fif+0.078
 bamboo+0.071
onna+0.070
iw+0.068
sed+0.068
Neg logit contributions
care-0.023
's-0.022
isions-0.022
 mechanic-0.021
 videos-0.020
Token named
Token ID3706
Feature activation+2.56e-04

Pos logit contributions
's+0.181
 videos+0.177
care+0.171
 mechanic+0.171
isions+0.170
Neg logit contributions
 Fif-0.292
onna-0.262
 bamboo-0.262
sed-0.252
iw-0.251
Token Fif
Token ID12830
Feature activation-4.86e-01

Pos logit contributions
 videos+0.000
isions+0.000
care+0.000
's+0.000
 mechanic+0.000
Neg logit contributions
 Fif-0.000
 bamboo-0.000
iw-0.000
onna-0.000
io-0.000
Tokeni
Token ID72
Feature activation-6.01e-02

Pos logit contributions
 Fif+0.850
 bamboo+0.798
balls+0.781
 ti+0.777
iw+0.774
Neg logit contributions
's-0.650
urious-0.601
 means-0.577
ations-0.565
isha-0.557
Token.
Token ID13
Feature activation-1.08e-02

Pos logit contributions
 Fif+0.115
 bamboo+0.105
onna+0.102
iw+0.100
sed+0.097
Neg logit contributions
's-0.079
 videos-0.074
 nephew-0.074
 mechanic-0.073
isions-0.072
Token Fif
Token ID12830
Feature activation-4.86e-01

Pos logit contributions
bled+0.012
castle+0.011
 majesty+0.011
 elbow+0.011
io+0.011
Neg logit contributions
 means-0.021
 On-0.021
's-0.021
 One-0.020
 During-0.019
Tokeni
Token ID72
Feature activation-6.02e-02

Pos logit contributions
 Fif+0.873
balls+0.865
 ti+0.844
 bamboo+0.835
onna+0.829
Neg logit contributions
's-0.578
urious-0.562
 means-0.552
ations-0.530
jo-0.493
Token lived
Token ID5615
Feature activation-3.20e-02

Pos logit contributions
 Fif+0.114
 bamboo+0.104
onna+0.100
iw+0.098
sed+0.095
Neg logit contributions
's-0.082
 videos-0.077
isions-0.076
 mechanic-0.075
 means-0.075
Token,
Token ID11
Feature activation-6.16e-03

Pos logit contributions
's+0.049
 means+0.038
+0.037
care+0.036
 jet+0.035
Neg logit contributions
iw-0.097
onna-0.091
io-0.091
 bamboo-0.090
 Fif-0.089
Token "
Token ID366
Feature activation-9.36e-02

Pos logit contributions
iw+0.013
 bamboo+0.012
io+0.012
 Fif+0.012
 majesty+0.012
Neg logit contributions
's-0.007
 Margaret-0.006
care-0.006
 videos-0.006
 mechanic-0.006
TokenThank
Token ID10449
Feature activation-2.51e-01

Pos logit contributions
onna+0.159
 Fif+0.158
sed+0.150
 majesty+0.149
 bamboo+0.148
Neg logit contributions
's-0.133
care-0.127
 Margaret-0.126
 nephew-0.122
During-0.122
Token you
Token ID345
Feature activation+2.42e-02

Pos logit contributions
onna+0.612
sed+0.565
 majesty+0.562
 pear+0.539
 ti+0.536
Neg logit contributions
's-0.201
care-0.177
 means-0.165
-0.164
 videos-0.157
Token,
Token ID11
Feature activation-5.88e-03

Pos logit contributions
's+0.033
 videos+0.031
 Margaret+0.030
 computer+0.029
 nephew+0.029
Neg logit contributions
 Fif-0.045
iw-0.041
onna-0.041
sed-0.040
 bamboo-0.040
Token Fif
Token ID12830
Feature activation-4.85e-01

Pos logit contributions
io+0.005
agged+0.005
ags+0.004
aces+0.004
 ti+0.004
Neg logit contributions
 jet-0.013
's-0.012
 means-0.012
-0.012
 Margaret-0.012
Tokeni
Token ID72
Feature activation-6.12e-02

Pos logit contributions
 ti+0.922
 Fif+0.901
balls+0.889
 bamboo+0.887
onna+0.884
Neg logit contributions
's-0.461
 means-0.456
ations-0.440
urious-0.438
 Margaret-0.422
Token,
Token ID11
Feature activation-5.02e-03

Pos logit contributions
 Fif+0.116
 bamboo+0.104
onna+0.104
iw+0.102
sed+0.099
Neg logit contributions
's-0.082
 means-0.074
 videos-0.073
 nephew-0.071
 computer-0.070
Token for
Token ID329
Feature activation-1.02e-01

Pos logit contributions
io+0.010
edly+0.009
sed+0.009
iw+0.009
onna+0.009
Neg logit contributions
's-0.006
-0.006
 jet-0.005
 means-0.005
 videos-0.005
Token showing
Token ID4478
Feature activation-1.24e-01

Pos logit contributions
onna+0.206
iw+0.202
 Fif+0.198
io+0.195
 ti+0.183
Neg logit contributions
 videos-0.121
 war-0.117
's-0.112
 computer-0.105
 jet-0.105
Token me
Token ID502
Feature activation-1.07e-01

Pos logit contributions
iw+0.234
onna+0.231
 Fif+0.216
 bamboo+0.209
 pear+0.205
Neg logit contributions
's-0.176
 videos-0.149
 means-0.148
 Margaret-0.144
 jet-0.142
Token other
Token ID584
Feature activation-8.03e-02

Pos logit contributions
 Fif+0.022
 bamboo+0.018
 pear+0.017
 l+0.016
 belly+0.016
Neg logit contributions
isions-0.057
cases-0.054
ographs-0.054
care-0.053
limits-0.053
Token.
Token ID13
Feature activation-1.21e-02

Pos logit contributions
 Fif+0.102
 bamboo+0.094
 pear+0.086
onna+0.084
 belly+0.081
Neg logit contributions
isions-0.147
 mechanic-0.144
 Margaret-0.143
controlled-0.142
 nephew-0.142
Token They
Token ID1119
Feature activation+1.11e-01

Pos logit contributions
 Fif+0.028
 bamboo+0.026
iw+0.025
onna+0.025
 pear+0.024
Neg logit contributions
 videos-0.011
isions-0.011
care-0.011
 mechanic-0.010
 nephew-0.010
Token think
Token ID892
Feature activation+5.53e-02

Pos logit contributions
's+0.145
 videos+0.143
isions+0.143
 jet+0.138
 mechanic+0.138
Neg logit contributions
 Fif-0.214
 bamboo-0.195
iw-0.186
onna-0.185
io-0.178
Token about
Token ID546
Feature activation+5.84e-02

Pos logit contributions
isions+0.073
 videos+0.072
care+0.071
 mechanic+0.069
's+0.069
Neg logit contributions
 Fif-0.106
 bamboo-0.095
onna-0.091
iw-0.090
 pear-0.088
Token Fif
Token ID12830
Feature activation-4.84e-01

Pos logit contributions
 videos+0.068
's+0.064
 means+0.063
isions+0.063
 computers+0.062
Neg logit contributions
 Fif-0.120
 bamboo-0.108
iw-0.108
onna-0.107
sed-0.106
Tokeni
Token ID72
Feature activation-6.67e-02

Pos logit contributions
 Fif+0.843
 bamboo+0.770
onna+0.720
 pear+0.717
iw+0.710
Neg logit contributions
isions-0.678
's-0.675
 means-0.666
 videos-0.660
care-0.655
Token's
Token ID338
Feature activation+3.71e-01

Pos logit contributions
 Fif+0.129
 bamboo+0.118
onna+0.111
iw+0.110
 pear+0.110
Neg logit contributions
isions-0.085
care-0.084
 videos-0.082
 mechanic-0.082
 keyboard-0.079
Token idea
Token ID2126
Feature activation+7.79e-02

Pos logit contributions
 videos+0.461
isions+0.449
 mechanic+0.445
 computer+0.444
's+0.443
Neg logit contributions
 Fif-0.758
 bamboo-0.673
onna-0.642
iw-0.640
io-0.622
Token.
Token ID13
Feature activation-1.10e-02

Pos logit contributions
's+0.099
care+0.088
 videos+0.087
 means+0.087
isions+0.086
Neg logit contributions
 Fif-0.166
 bamboo-0.149
iw-0.146
 majesty-0.143
onna-0.143
Token They
Token ID1119
Feature activation+1.12e-01

Pos logit contributions
 Fif+0.024
 bamboo+0.023
onna+0.022
iw+0.022
io+0.022
Neg logit contributions
care-0.009
isions-0.009
 videos-0.009
 means-0.009
's-0.009
Token her
Token ID607
Feature activation-5.56e-02

Pos logit contributions
onna+0.009
sed+0.008
iw+0.008
stack+0.008
 Fif+0.008
Neg logit contributions
's-0.004
 videos-0.004
 jet-0.004
 films-0.004
 Peter-0.003
Token family
Token ID1641
Feature activation+2.91e-02

Pos logit contributions
onna+0.110
 Fif+0.109
 seventy+0.106
iw+0.104
ern+0.100
Neg logit contributions
 jet-0.067
's-0.063
 radio-0.060
 nephew-0.060
 computer-0.059
Token and
Token ID290
Feature activation+3.05e-03

Pos logit contributions
 videos+0.041
isions+0.040
's+0.040
care+0.040
 mechanic+0.039
Neg logit contributions
 Fif-0.053
 bamboo-0.047
onna-0.046
iw-0.045
sed-0.045
Token friends
Token ID2460
Feature activation-1.15e-01

Pos logit contributions
's+0.003
 videos+0.003
 nephew+0.003
 jet+0.003
 means+0.003
Neg logit contributions
onna-0.006
iw-0.006
 Fif-0.006
io-0.006
 bamboo-0.006
Token.
Token ID13
Feature activation-6.55e-03

Pos logit contributions
 Fif+0.239
onna+0.221
 bamboo+0.218
iw+0.216
sed+0.214
Neg logit contributions
's-0.138
 videos-0.123
care-0.121
 means-0.119
isions-0.117
Token Fif
Token ID12830
Feature activation-4.84e-01

Pos logit contributions
 majesty+0.014
 ti+0.014
io+0.014
 elbow+0.014
 crown+0.014
Neg logit contributions
's-0.005
 means-0.005
 On-0.005
During-0.004
 During-0.004
Tokeni
Token ID72
Feature activation-6.30e-02

Pos logit contributions
 Fif+0.890
balls+0.868
 ti+0.866
 bamboo+0.850
onna+0.849
Neg logit contributions
's-0.549
urious-0.533
ations-0.515
 means-0.514
itionally-0.473
Token loved
Token ID6151
Feature activation-2.89e-02

Pos logit contributions
 Fif+0.116
 bamboo+0.104
onna+0.100
iw+0.099
 pear+0.095
Neg logit contributions
's-0.087
isions-0.085
 videos-0.085
care-0.083
 mechanic-0.083
Token to
Token ID284
Feature activation-2.78e-02

Pos logit contributions
 Fif+0.057
io+0.057
iw+0.057
sed+0.056
onna+0.056
Neg logit contributions
 videos-0.033
 films-0.029
's-0.029
 computer-0.029
 computers-0.028
Token explore
Token ID7301
Feature activation+6.19e-02

Pos logit contributions
iw+0.060
 Fif+0.059
io+0.055
sed+0.055
 bamboo+0.053
Neg logit contributions
 videos-0.028
 jet-0.027
 pray-0.027
's-0.027
 press-0.025
Token new
Token ID649
Feature activation-2.06e-02

Pos logit contributions
's+0.070
 videos+0.070
care+0.064
 jet+0.063
 computers+0.063
Neg logit contributions
 Fif-0.129
sed-0.125
onna-0.120
iw-0.116
 majesty-0.114
Token store
Token ID3650
Feature activation+1.05e-01

Pos logit contributions
's+0.007
 videos+0.007
 jet+0.007
 mechanic+0.007
care+0.007
Neg logit contributions
 Fif-0.007
 bamboo-0.006
onna-0.006
io-0.006
balls-0.006
Token with
Token ID351
Feature activation-4.04e-03

Pos logit contributions
isions+0.149
 videos+0.148
care+0.146
 mechanic+0.143
's+0.142
Neg logit contributions
 Fif-0.189
 bamboo-0.169
onna-0.161
iw-0.158
 pear-0.155
Token many
Token ID867
Feature activation-2.37e-02

Pos logit contributions
onna+0.008
sed+0.008
 Fif+0.008
iw+0.008
 majesty+0.007
Neg logit contributions
 videos-0.004
's-0.004
 computers-0.004
 nephew-0.004
 mail-0.004
Token toys
Token ID14958
Feature activation+1.81e-02

Pos logit contributions
 Fif+0.042
onna+0.038
 bamboo+0.037
iw+0.036
sed+0.036
Neg logit contributions
 videos-0.034
isions-0.034
's-0.033
 computers-0.033
care-0.033
Token.
Token ID13
Feature activation-1.00e-02

Pos logit contributions
isions+0.026
care+0.025
 videos+0.025
's+0.025
 mechanic+0.025
Neg logit contributions
 Fif-0.033
 bamboo-0.029
onna-0.028
iw-0.027
 pear-0.027
Token Fif
Token ID12830
Feature activation-4.84e-01

Pos logit contributions
io+0.023
bled+0.023
 majesty+0.022
 elbow+0.022
castle+0.022
Neg logit contributions
's-0.008
 On-0.007
 means-0.007
 One-0.006
During-0.006
Tokeni
Token ID72
Feature activation-6.08e-02

Pos logit contributions
 Fif+0.868
 ti+0.856
balls+0.851
 bamboo+0.839
onna+0.830
Neg logit contributions
's-0.568
urious-0.554
 means-0.537
ations-0.529
itionally-0.483
Token loved
Token ID6151
Feature activation-3.01e-02

Pos logit contributions
 Fif+0.105
 bamboo+0.093
onna+0.088
iw+0.088
 pear+0.086
Neg logit contributions
isions-0.091
 videos-0.090
care-0.089
 mechanic-0.087
 nephew-0.086
Token to
Token ID284
Feature activation-2.86e-02

Pos logit contributions
 Fif+0.061
io+0.060
sed+0.059
 majesty+0.058
iw+0.057
Neg logit contributions
 videos-0.036
 computers-0.033
 computer-0.032
 mechanic-0.030
care-0.030
Token play
Token ID711
Feature activation+1.25e-02

Pos logit contributions
 Fif+0.060
iw+0.055
sed+0.055
 bamboo+0.054
onna+0.053
Neg logit contributions
 videos-0.032
 mechanic-0.030
 jet-0.029
 computer-0.029
 computers-0.029
Token with
Token ID351
Feature activation-3.23e-03

Pos logit contributions
 videos+0.017
care+0.016
isions+0.016
's+0.016
 mechanic+0.016
Neg logit contributions
 Fif-0.024
onna-0.021
 bamboo-0.021
iw-0.021
sed-0.021
Token and
Token ID290
Feature activation+8.60e-03

Pos logit contributions
 Fif+0.231
sed+0.211
 bamboo+0.197
io+0.197
iw+0.193
Neg logit contributions
's-0.174
care-0.148
 videos-0.140
isions-0.139
 mechanic-0.135
Token play
Token ID711
Feature activation+1.72e-02

Pos logit contributions
's+0.010
 videos+0.010
 jet+0.010
 nephew+0.010
 mechanic+0.010
Neg logit contributions
 Fif-0.016
io-0.016
iw-0.016
 bamboo-0.015
onna-0.015
Token together
Token ID1978
Feature activation+8.62e-02

Pos logit contributions
's+0.016
 videos+0.015
care+0.014
ographs+0.014
 jet+0.014
Neg logit contributions
 Fif-0.041
onna-0.038
iw-0.037
sed-0.037
ho-0.037
Token again
Token ID757
Feature activation-1.37e-01

Pos logit contributions
's+0.113
care+0.094
 means+0.090
 videos+0.085
isions+0.084
Neg logit contributions
 Fif-0.180
iw-0.161
sed-0.160
 bamboo-0.157
onna-0.152
Token!"
Token ID2474
Feature activation-9.96e-02

Pos logit contributions
 Fif+0.223
 bamboo+0.198
iw+0.188
onna+0.188
sed+0.187
Neg logit contributions
's-0.215
care-0.213
 videos-0.212
isions-0.211
 means-0.207
Token Fif
Token ID12830
Feature activation-4.83e-01

Pos logit contributions
onna+0.072
 ti+0.067
 majesty+0.066
 seventy+0.065
io+0.057
Neg logit contributions
's-0.241
 means-0.229
-0.215
 videos-0.213
care-0.212
Tokeni
Token ID72
Feature activation-5.99e-02

Pos logit contributions
 ti+0.973
onna+0.915
 crown+0.911
 bamboo+0.904
balls+0.904
Neg logit contributions
's-0.461
 means-0.458
urious-0.452
ations-0.423
ita-0.404
Token smiled
Token ID13541
Feature activation-1.60e-01

Pos logit contributions
 Fif+0.115
iw+0.109
 bamboo+0.109
onna+0.109
 majesty+0.099
Neg logit contributions
's-0.089
 means-0.071
-0.067
 nephew-0.066
 jet-0.064
Token and
Token ID290
Feature activation+9.76e-03

Pos logit contributions
 Fif+0.361
iw+0.332
 bamboo+0.326
onna+0.324
 majesty+0.312
Neg logit contributions
's-0.179
care-0.158
 videos-0.156
 mechanic-0.150
 nephew-0.149
Token said
Token ID531
Feature activation+4.81e-02

Pos logit contributions
's+0.010
 means+0.008
 jet+0.008
 Margaret+0.008
 videos+0.008
Neg logit contributions
onna-0.022
iw-0.021
 bamboo-0.021
 Fif-0.020
io-0.020
Token,
Token ID11
Feature activation-2.67e-03

Pos logit contributions
's+0.047
 means+0.034
 videos+0.033
+0.033
care+0.032
Neg logit contributions
iw-0.111
 Fif-0.107
 bamboo-0.104
onna-0.103
stack-0.101
Token
Token ID198
Feature activation-3.38e-02

Pos logit contributions
 majesty+0.012
 ti+0.012
sed+0.012
io+0.012
 crown+0.012
Neg logit contributions
's-0.006
 means-0.005
During-0.004
care-0.004
-0.004
Token
Token ID198
Feature activation-3.37e-02

Pos logit contributions
 Fif+0.060
 bamboo+0.053
onna+0.051
iw+0.050
sed+0.049
Neg logit contributions
isions-0.048
 videos-0.048
care-0.048
's-0.047
 mechanic-0.047
TokenOne
Token ID3198
Feature activation+7.03e-02

Pos logit contributions
 Fif+0.076
 majesty+0.074
sed+0.074
onna+0.073
iw+0.072
Neg logit contributions
's-0.026
During-0.026
 means-0.023
care-0.022
 jet-0.021
Token day
Token ID1110
Feature activation+2.14e-02

Pos logit contributions
isions+0.116
 videos+0.108
care+0.108
 keyboard+0.107
ographs+0.107
Neg logit contributions
 Fif-0.121
 bamboo-0.108
 pear-0.099
onna-0.098
iw-0.096
Token,
Token ID11
Feature activation-7.80e-03

Pos logit contributions
's+0.023
care+0.021
 means+0.021
During+0.018
 jet+0.018
Neg logit contributions
onna-0.043
 majesty-0.041
 Fif-0.040
io-0.040
iw-0.040
Token Fif
Token ID12830
Feature activation-4.82e-01

Pos logit contributions
onna+0.007
 majesty+0.007
io+0.006
 ti+0.006
 seventy+0.005
Neg logit contributions
's-0.017
 means-0.016
During-0.016
care-0.016
 jet-0.015
Tokeni
Token ID72
Feature activation-6.14e-02

Pos logit contributions
 Fif+0.862
 bamboo+0.812
 ti+0.806
onna+0.795
balls+0.791
Neg logit contributions
's-0.600
urious-0.575
ations-0.566
 means-0.564
care-0.530
Token and
Token ID290
Feature activation+5.88e-03

Pos logit contributions
 Fif+0.092
 bamboo+0.080
 pear+0.077
io+0.076
ho+0.074
Neg logit contributions
 videos-0.106
isions-0.106
 keyboard-0.103
controlled-0.101
ographs-0.100
Token her
Token ID607
Feature activation-5.29e-02

Pos logit contributions
's+0.007
 nephew+0.007
 videos+0.007
 means+0.007
 Margaret+0.007
Neg logit contributions
onna-0.011
iw-0.011
 Fif-0.010
io-0.010
 bamboo-0.010
Token friends
Token ID2460
Feature activation-1.12e-01

Pos logit contributions
onna+0.101
 Fif+0.099
iw+0.099
 seventy+0.088
stack+0.088
Neg logit contributions
 nephew-0.069
's-0.069
 jet-0.068
 mechanic-0.067
 videos-0.065
Token went
Token ID1816
Feature activation+3.00e-02

Pos logit contributions
 Fif+0.167
 bamboo+0.135
 pear+0.132
ho+0.128
sed+0.126
Neg logit contributions
isions-0.201
 videos-0.201
controlled-0.197
 keyboard-0.193
 complicated-0.188
Token
Token ID198
Feature activation-3.60e-02

Pos logit contributions
 ti+0.020
 majesty+0.020
sed+0.020
io+0.019
 crown+0.019
Neg logit contributions
's-0.009
 means-0.008
During-0.007
care-0.007
 On-0.007
Token
Token ID198
Feature activation-3.60e-02

Pos logit contributions
 Fif+0.064
 bamboo+0.057
onna+0.054
iw+0.054
sed+0.053
Neg logit contributions
isions-0.051
 videos-0.051
care-0.051
's-0.049
 mechanic-0.049
TokenOne
Token ID3198
Feature activation+6.98e-02

Pos logit contributions
 majesty+0.082
sed+0.081
 Fif+0.079
onna+0.079
io+0.077
Neg logit contributions
During-0.028
's-0.027
 means-0.024
care-0.022
-0.021
Token day
Token ID1110
Feature activation+2.12e-02

Pos logit contributions
isions+0.133
ographs+0.127
cases+0.123
 keyboard+0.123
 computers+0.123
Neg logit contributions
 Fif-0.108
 bamboo-0.096
 pear-0.090
iw-0.082
onna-0.081
Token,
Token ID11
Feature activation-8.55e-03

Pos logit contributions
's+0.023
 means+0.021
care+0.021
 videos+0.018
 jet+0.018
Neg logit contributions
onna-0.042
 majesty-0.041
io-0.040
 Fif-0.040
iw-0.039
Token Fif
Token ID12830
Feature activation-4.81e-01

Pos logit contributions
 majesty+0.007
onna+0.007
io+0.006
 ti+0.006
 seventy+0.005
Neg logit contributions
's-0.019
 means-0.019
During-0.018
care-0.018
 jet-0.017
Tokeni
Token ID72
Feature activation-5.87e-02

Pos logit contributions
 Fif+0.863
 ti+0.826
 bamboo+0.824
balls+0.814
onna+0.812
Neg logit contributions
's-0.582
urious-0.561
 means-0.551
ations-0.548
isha-0.516
Token saw
Token ID2497
Feature activation+6.07e-02

Pos logit contributions
 Fif+0.082
 pear+0.071
iw+0.070
io+0.069
ho+0.069
Neg logit contributions
 videos-0.103
controlled-0.101
isions-0.101
 keyboard-0.098
ographs-0.098
Token a
Token ID257
Feature activation+6.03e-02

Pos logit contributions
's+0.054
 videos+0.052
 means+0.050
 nephew+0.049
 jet+0.048
Neg logit contributions
onna-0.131
 Fif-0.130
iw-0.126
 majesty-0.124
io-0.122
Token sad
Token ID6507
Feature activation+1.69e-01

Pos logit contributions
isions+0.087
 videos+0.085
care+0.085
controlled+0.084
's+0.081
Neg logit contributions
 Fif-0.108
 bamboo-0.098
 pear-0.092
onna-0.091
iw-0.091
Token girl
Token ID2576
Feature activation+9.09e-02

Pos logit contributions
 videos+0.235
 mechanic+0.230
care+0.228
 jet+0.222
 nephew+0.222
Neg logit contributions
 Fif-0.305
 bamboo-0.272
iw-0.266
onna-0.265
io-0.255
Token day
Token ID1110
Feature activation+2.55e-02

Pos logit contributions
's+0.061
 videos+0.058
 means+0.056
 jet+0.056
care+0.056
Neg logit contributions
 Fif-0.102
 bamboo-0.092
onna-0.092
iw-0.089
sed-0.088
Token on
Token ID319
Feature activation-4.11e-02

Pos logit contributions
's+0.033
care+0.031
 Margaret+0.029
 means+0.028
 jet+0.028
Neg logit contributions
onna-0.047
 Fif-0.046
 majesty-0.046
 bamboo-0.044
iw-0.044
Token,
Token ID11
Feature activation-1.17e-03

Pos logit contributions
 Fif+0.084
iw+0.083
onna+0.082
asp+0.082
sed+0.082
Neg logit contributions
's-0.041
 means-0.037
 Margaret-0.034
 jet-0.034
care-0.034
Token Lily
Token ID20037
Feature activation-8.36e-02

Pos logit contributions
onna+0.002
sed+0.002
 ti+0.002
io+0.002
 majesty+0.002
Neg logit contributions
's-0.002
 means-0.002
-0.001
 videos-0.001
 Margaret-0.001
Token and
Token ID290
Feature activation+1.09e-02

Pos logit contributions
 Fif+0.195
 bamboo+0.186
iw+0.177
onna+0.172
sed+0.170
Neg logit contributions
's-0.099
 means-0.078
 Margaret-0.071
 videos-0.069
 jet-0.068
Token Fif
Token ID12830
Feature activation-4.81e-01

Pos logit contributions
's+0.021
 means+0.021
 Margaret+0.020
care+0.020
 nephew+0.020
Neg logit contributions
onna-0.012
 ti-0.011
sed-0.011
iw-0.011
 majesty-0.011
Tokeni
Token ID72
Feature activation-5.79e-02

Pos logit contributions
 ti+0.978
onna+0.935
balls+0.930
 Fif+0.920
 bamboo+0.908
Neg logit contributions
's-0.470
urious-0.442
ations-0.442
 means-0.435
jo-0.408
Token were
Token ID547
Feature activation-4.49e-02

Pos logit contributions
 Fif+0.114
onna+0.109
 bamboo+0.107
iw+0.106
sed+0.099
Neg logit contributions
's-0.079
 jet-0.068
 nephew-0.064
 means-0.063
 station-0.062
Token best
Token ID1266
Feature activation-1.21e-01

Pos logit contributions
onna+0.087
sed+0.086
 Fif+0.086
iw+0.085
 bamboo+0.082
Neg logit contributions
's-0.060
 Margaret-0.051
 videos-0.050
 jet-0.049
-0.048
Token friends
Token ID2460
Feature activation-1.13e-01

Pos logit contributions
 Fif+0.220
onna+0.216
stack+0.205
 seventy+0.200
 bamboo+0.200
Neg logit contributions
's-0.163
care-0.160
 videos-0.159
ever-0.153
 nephew-0.153
Token.
Token ID13
Feature activation+3.20e-04

Pos logit contributions
 Fif+0.240
 majesty+0.233
onna+0.232
sed+0.225
iw+0.221
Neg logit contributions
's-0.137
care-0.111
 videos-0.108
 Margaret-0.103
-0.102
Token"
Token ID1
Feature activation+5.10e-02

Pos logit contributions
 Fif+0.069
 bamboo+0.061
iw+0.057
onna+0.055
 pear+0.054
Neg logit contributions
 videos-0.061
isions-0.059
 nephew-0.057
 computer-0.057
 computers-0.057
TokenHello
Token ID15496
Feature activation-8.85e-02

Pos logit contributions
isions+0.077
 videos+0.076
care+0.076
 mechanic+0.074
 nephew+0.074
Neg logit contributions
 Fif-0.087
 bamboo-0.078
onna-0.075
iw-0.073
 pear-0.072
Token,
Token ID11
Feature activation-6.34e-03

Pos logit contributions
 Fif+0.164
 bamboo+0.148
onna+0.147
iw+0.143
sed+0.141
Neg logit contributions
's-0.114
care-0.114
isions-0.113
 videos-0.113
 means-0.110
Token I
Token ID314
Feature activation-9.34e-02

Pos logit contributions
onna+0.007
io+0.007
 Fif+0.007
iw+0.007
edly+0.006
Neg logit contributions
's-0.012
 jet-0.012
 means-0.012
 videos-0.011
 mechanic-0.011
Token'm
Token ID1101
Feature activation-8.82e-02

Pos logit contributions
 Fif+0.144
 bamboo+0.126
onna+0.118
iw+0.117
 pear+0.113
Neg logit contributions
isions-0.158
 videos-0.157
care-0.155
 mechanic-0.152
 nephew-0.151
Token Fif
Token ID12830
Feature activation-4.81e-01

Pos logit contributions
 Fif+0.138
 bamboo+0.126
onna+0.121
iw+0.120
io+0.115
Neg logit contributions
 videos-0.141
's-0.136
care-0.136
isions-0.136
 Margaret-0.133
Tokeni
Token ID72
Feature activation-5.84e-02

Pos logit contributions
 Fif+0.851
 bamboo+0.812
onna+0.757
 pear+0.748
 ti+0.738
Neg logit contributions
's-0.614
 means-0.587
isions-0.584
care-0.579
ever-0.571
Token.
Token ID13
Feature activation-9.00e-03

Pos logit contributions
 Fif+0.110
 bamboo+0.099
onna+0.094
iw+0.093
 pear+0.091
Neg logit contributions
isions-0.078
 videos-0.077
care-0.076
 mechanic-0.074
's-0.074
Token I
Token ID314
Feature activation-9.26e-02

Pos logit contributions
 Fif+0.016
 bamboo+0.016
io+0.016
onna+0.015
iw+0.015
Neg logit contributions
's-0.011
 videos-0.010
 means-0.010
 Margaret-0.010
care-0.010
Token'm
Token ID1101
Feature activation-8.76e-02

Pos logit contributions
 Fif+0.161
 bamboo+0.145
onna+0.137
iw+0.135
sed+0.132
Neg logit contributions
's-0.133
care-0.131
isions-0.129
 videos-0.129
 means-0.129
Token a
Token ID257
Feature activation+6.19e-02

Pos logit contributions
 Fif+0.144
iw+0.137
 bamboo+0.136
io+0.135
onna+0.134
Neg logit contributions
 videos-0.126
's-0.121
 Margaret-0.120
care-0.117
 mechanic-0.111
Token and
Token ID290
Feature activation+6.52e-03

Pos logit contributions
's+0.033
care+0.028
 videos+0.027
 means+0.026
 Peter+0.025
Neg logit contributions
 majesty-0.093
onna-0.090
bit-0.088
 Fif-0.086
igi-0.086
Token had
Token ID550
Feature activation-3.90e-02

Pos logit contributions
's+0.010
 videos+0.009
care+0.008
 jet+0.008
 Peter+0.008
Neg logit contributions
 Fif-0.011
onna-0.011
iw-0.011
 majesty-0.011
ags-0.010
Token fun
Token ID1257
Feature activation-6.71e-02

Pos logit contributions
 Fif+0.078
iw+0.075
onna+0.072
io+0.071
sed+0.070
Neg logit contributions
's-0.051
 computers-0.044
 videos-0.044
isions-0.043
 computer-0.040
Token,
Token ID11
Feature activation-5.65e-03

Pos logit contributions
 Fif+0.139
onna+0.125
iw+0.125
 bamboo+0.120
sed+0.119
Neg logit contributions
 videos-0.081
's-0.078
care-0.078
isions-0.074
 computers-0.071
Token but
Token ID475
Feature activation-6.10e-03

Pos logit contributions
 Fif+0.011
onna+0.010
iw+0.010
 bamboo+0.010
io+0.010
Neg logit contributions
's-0.007
 videos-0.007
care-0.007
isions-0.007
 computers-0.006
Token Fif
Token ID12830
Feature activation-4.80e-01

Pos logit contributions
 Fif+0.012
onna+0.011
 bamboo+0.011
iw+0.011
 majesty+0.011
Neg logit contributions
's-0.007
care-0.006
 means-0.006
 videos-0.006
 jet-0.006
Tokeni
Token ID72
Feature activation-5.98e-02

Pos logit contributions
 Fif+0.887
 bamboo+0.853
 ti+0.835
balls+0.822
onna+0.819
Neg logit contributions
's-0.574
urious-0.550
ations-0.528
 means-0.526
isions-0.500
Token noticed
Token ID6810
Feature activation+1.40e-01

Pos logit contributions
 Fif+0.109
 bamboo+0.098
onna+0.093
iw+0.092
 pear+0.090
Neg logit contributions
isions-0.083
 videos-0.083
care-0.081
 mechanic-0.080
 nephew-0.079
Token something
Token ID1223
Feature activation+1.65e-01

Pos logit contributions
's+0.133
 jet+0.125
 means+0.125
 videos+0.124
 computers+0.118
Neg logit contributions
 Fif-0.302
onna-0.295
iw-0.287
 majesty-0.284
sed-0.278
Token strange
Token ID6283
Feature activation+2.53e-01

Pos logit contributions
isions+0.181
 videos+0.178
care+0.175
 mechanic+0.171
 nephew+0.169
Neg logit contributions
 Fif-0.352
 bamboo-0.320
onna-0.306
iw-0.303
 pear-0.299
Token.
Token ID13
Feature activation-7.05e-03

Pos logit contributions
's+0.334
 videos+0.332
 means+0.326
care+0.323
isions+0.318
Neg logit contributions
 Fif-0.483
iw-0.427
 bamboo-0.424
onna-0.417
 majesty-0.412
Token not
Token ID407
Feature activation-8.81e-02

Pos logit contributions
 Fif+0.150
 bamboo+0.132
onna+0.125
iw+0.123
sed+0.119
Neg logit contributions
 videos-0.160
isions-0.160
care-0.159
's-0.158
 mechanic-0.154
Token have
Token ID423
Feature activation+2.61e-02

Pos logit contributions
 Fif+0.186
 bamboo+0.170
onna+0.163
iw+0.162
sed+0.158
Neg logit contributions
 videos-0.098
isions-0.097
care-0.096
 mechanic-0.093
's-0.093
Token enough
Token ID1576
Feature activation+7.00e-02

Pos logit contributions
 videos+0.030
care+0.030
's+0.030
isions+0.030
 mechanic+0.029
Neg logit contributions
 Fif-0.054
 bamboo-0.048
onna-0.047
iw-0.047
sed-0.045
Token money
Token ID1637
Feature activation+1.39e-01

Pos logit contributions
 videos+0.072
care+0.070
isions+0.065
's+0.065
 mechanic+0.065
Neg logit contributions
 Fif-0.157
 bamboo-0.139
iw-0.137
onna-0.137
io-0.137
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
's+0.168
care+0.167
 means+0.155
 videos+0.152
controlled+0.147
Neg logit contributions
 Fif-0.285
 bamboo-0.258
 majesty-0.254
onna-0.244
iw-0.243
Token Fif
Token ID12830
Feature activation-4.79e-01

Pos logit contributions
 ti+0.017
 majesty+0.017
onna+0.017
 crown+0.017
io+0.017
Neg logit contributions
's-0.020
 means-0.019
care-0.017
 videos-0.016
-0.016
Tokeni
Token ID72
Feature activation-5.89e-02

Pos logit contributions
 Fif+0.866
onna+0.838
 bamboo+0.836
 ti+0.823
balls+0.822
Neg logit contributions
's-0.563
 means-0.559
urious-0.552
ations-0.538
care-0.507
Token wanted
Token ID2227
Feature activation+2.21e-02

Pos logit contributions
 Fif+0.099
 bamboo+0.086
 pear+0.082
iw+0.082
sed+0.080
Neg logit contributions
isions-0.093
 videos-0.092
controlled-0.091
care-0.090
 keyboard-0.089
Token to
Token ID284
Feature activation-2.50e-02

Pos logit contributions
's+0.026
 videos+0.024
 computers+0.023
care+0.023
 mechanic+0.023
Neg logit contributions
 Fif-0.044
 majesty-0.044
sed-0.043
io-0.042
onna-0.041
Token help
Token ID1037
Feature activation-1.16e-01

Pos logit contributions
 Fif+0.050
 bamboo+0.047
sed+0.046
iw+0.045
io+0.044
Neg logit contributions
 mechanic-0.030
's-0.029
 videos-0.029
care-0.027
 jet-0.027
Token Lily
Token ID20037
Feature activation-8.37e-02

Pos logit contributions
 majesty+0.232
sed+0.224
onna+0.218
 ti+0.213
iw+0.209
Neg logit contributions
's-0.144
 videos-0.130
care-0.123
 means-0.120
 Margaret-0.112
Token cave
Token ID11527
Feature activation-7.37e-02

Pos logit contributions
isions+0.102
's+0.102
 videos+0.102
 mechanic+0.102
 jet+0.101
Neg logit contributions
 Fif-0.098
 bamboo-0.082
iw-0.080
onna-0.079
sed-0.078
Token near
Token ID1474
Feature activation-6.89e-02

Pos logit contributions
 Fif+0.130
 bamboo+0.115
onna+0.109
iw+0.108
 pear+0.105
Neg logit contributions
isions-0.108
 videos-0.107
care-0.106
 mechanic-0.104
's-0.103
Token the
Token ID262
Feature activation+8.72e-03

Pos logit contributions
 Fif+0.133
 majesty+0.130
onna+0.129
iw+0.125
sed+0.125
Neg logit contributions
's-0.078
 means-0.075
care-0.074
 videos-0.073
 jet-0.071
Token reef
Token ID25088
Feature activation-1.40e-01

Pos logit contributions
's+0.013
 jet+0.013
 mechanic+0.013
 videos+0.013
 radio+0.012
Neg logit contributions
 Fif-0.015
iw-0.014
onna-0.014
io-0.013
sed-0.013
Token.
Token ID13
Feature activation-5.93e-03

Pos logit contributions
 Fif+0.245
 bamboo+0.216
onna+0.206
iw+0.204
 pear+0.198
Neg logit contributions
isions-0.206
 videos-0.204
care-0.203
's-0.199
 mechanic-0.198
Token Fif
Token ID12830
Feature activation-4.79e-01

Pos logit contributions
 ti+0.006
io+0.006
 crown+0.006
 majesty+0.006
onna+0.006
Neg logit contributions
's-0.012
 means-0.011
care-0.010
-0.010
 Taking-0.010
Tokeni
Token ID72
Feature activation-5.88e-02

Pos logit contributions
 Fif+0.878
 bamboo+0.847
onna+0.846
 ti+0.838
balls+0.833
Neg logit contributions
's-0.566
urious-0.543
 means-0.529
ations-0.521
care-0.490
Token wondered
Token ID14028
Feature activation+2.64e-02

Pos logit contributions
 Fif+0.101
 bamboo+0.088
onna+0.083
 pear+0.083
sed+0.082
Neg logit contributions
 videos-0.090
isions-0.089
care-0.088
controlled-0.087
 mechanic-0.085
Token what
Token ID644
Feature activation+8.67e-02

Pos logit contributions
's+0.025
 videos+0.024
isions+0.024
care+0.024
 nephew+0.024
Neg logit contributions
 Fif-0.059
 bamboo-0.055
onna-0.054
iw-0.053
sed-0.052
Token was
Token ID373
Feature activation-1.19e-02

Pos logit contributions
isions+0.124
 videos+0.122
care+0.121
 mechanic+0.118
 nephew+0.117
Neg logit contributions
 Fif-0.157
 bamboo-0.140
onna-0.133
iw-0.131
 pear-0.129
Token inside
Token ID2641
Feature activation-4.02e-03

Pos logit contributions
 Fif+0.018
 bamboo+0.016
 pear+0.014
onna+0.014
iw+0.014
Neg logit contributions
isions-0.021
 videos-0.020
care-0.020
 nephew-0.020
 mechanic-0.020
Token
Token ID198
Feature activation-3.27e-02

Pos logit contributions
 Fif+0.063
 bamboo+0.059
sed+0.058
iw+0.057
 majesty+0.057
Neg logit contributions
's-0.038
care-0.038
isions-0.035
 means-0.035
 videos-0.035
TokenLater
Token ID18602
Feature activation+1.53e-01

Pos logit contributions
 Fif+0.048
onna+0.048
sed+0.047
 majesty+0.047
iw+0.046
Neg logit contributions
's-0.050
During-0.048
care-0.047
-0.046
 means-0.046
Token that
Token ID326
Feature activation+5.09e-02

Pos logit contributions
isions+0.245
 videos+0.232
controlled+0.229
 computers+0.228
 nephew+0.228
Neg logit contributions
 Fif-0.265
 bamboo-0.233
 pear-0.216
onna-0.212
iw-0.211
Token day
Token ID1110
Feature activation+2.63e-02

Pos logit contributions
isions+0.120
controlled+0.111
ographs+0.111
cases+0.110
 computers+0.110
Neg logit contributions
 Fif-0.059
 bamboo-0.049
 pear-0.047
 Buzz-0.046
 frog-0.044
Token,
Token ID11
Feature activation-2.20e-03

Pos logit contributions
's+0.027
care+0.027
 means+0.024
 jet+0.022
 videos+0.022
Neg logit contributions
onna-0.052
iw-0.051
 majesty-0.050
 Fif-0.050
 bamboo-0.049
Token Fif
Token ID12830
Feature activation-4.78e-01

Pos logit contributions
onna+0.002
 majesty+0.002
io+0.002
 ti+0.002
iw+0.002
Neg logit contributions
's-0.005
 means-0.005
care-0.005
During-0.004
 jet-0.004
Tokeni
Token ID72
Feature activation-5.77e-02

Pos logit contributions
 Fif+0.859
 bamboo+0.814
onna+0.790
 ti+0.786
iw+0.780
Neg logit contributions
's-0.606
urious-0.576
 means-0.561
ations-0.560
care-0.554
Token and
Token ID290
Feature activation+1.12e-02

Pos logit contributions
 Fif+0.084
 bamboo+0.073
io+0.073
 pear+0.072
sed+0.071
Neg logit contributions
 videos-0.101
isions-0.096
controlled-0.096
ographs-0.094
 keyboard-0.093
Token her
Token ID607
Feature activation-4.83e-02

Pos logit contributions
's+0.015
 videos+0.015
 nephew+0.015
 means+0.014
 mechanic+0.014
Neg logit contributions
 Fif-0.020
onna-0.020
iw-0.019
 bamboo-0.019
sed-0.018
Token friends
Token ID2460
Feature activation-1.09e-01

Pos logit contributions
onna+0.095
iw+0.093
 Fif+0.091
 seventy+0.082
 violet+0.081
Neg logit contributions
 jet-0.061
 nephew-0.060
 mechanic-0.059
's-0.057
 radio-0.056
Token were
Token ID547
Feature activation-4.34e-02

Pos logit contributions
 Fif+0.145
 bamboo+0.119
 pear+0.117
io+0.115
ho+0.115
Neg logit contributions
isions-0.202
 videos-0.201
controlled-0.196
 complicated-0.192
 keyboard-0.190
Token a
Token ID257
Feature activation+6.13e-02

Pos logit contributions
 videos+0.079
's+0.077
 nephew+0.076
care+0.076
 mechanic+0.075
Neg logit contributions
 Fif-0.104
onna-0.096
 bamboo-0.094
iw-0.091
sed-0.090
Token toy
Token ID13373
Feature activation+9.47e-02

Pos logit contributions
 mechanic+0.083
care+0.081
 videos+0.081
isions+0.080
 nephew+0.080
Neg logit contributions
 Fif-0.114
 bamboo-0.102
iw-0.097
onna-0.097
io-0.096
Token for
Token ID329
Feature activation-1.01e-01

Pos logit contributions
 mechanic+0.123
 videos+0.120
's+0.120
 jet+0.120
care+0.118
Neg logit contributions
 Fif-0.182
 bamboo-0.161
onna-0.161
 majesty-0.156
iw-0.154
Token her
Token ID607
Feature activation-5.08e-02

Pos logit contributions
 Fif+0.185
onna+0.181
iw+0.179
 majesty+0.176
asp+0.173
Neg logit contributions
 videos-0.134
care-0.126
 computers-0.117
's-0.113
 nephew-0.111
Token.
Token ID13
Feature activation-9.66e-03

Pos logit contributions
 Fif+0.098
 bamboo+0.087
iw+0.086
onna+0.085
io+0.084
Neg logit contributions
's-0.065
 videos-0.065
 mechanic-0.065
 nephew-0.065
 computer-0.064
Token Fif
Token ID12830
Feature activation-4.78e-01

Pos logit contributions
 majesty+0.019
io+0.019
 crown+0.019
 ti+0.019
 pearl+0.019
Neg logit contributions
's-0.010
 means-0.009
care-0.009
During-0.008
 On-0.008
Tokeni
Token ID72
Feature activation-5.74e-02

Pos logit contributions
 Fif+0.881
onna+0.834
 bamboo+0.833
balls+0.815
 ti+0.803
Neg logit contributions
's-0.579
urious-0.561
 means-0.552
ations-0.544
isions-0.522
Token found
Token ID1043
Feature activation-1.40e-02

Pos logit contributions
 Fif+0.105
 bamboo+0.093
onna+0.089
iw+0.088
 pear+0.087
Neg logit contributions
isions-0.081
 videos-0.080
care-0.078
controlled-0.076
 mechanic-0.076
Token a
Token ID257
Feature activation+6.25e-02

Pos logit contributions
onna+0.035
 majesty+0.033
castle+0.031
io+0.031
iw+0.031
Neg logit contributions
's-0.011
 means-0.009
 videos-0.007
 jet-0.007
 nephew-0.007
Token small
Token ID1402
Feature activation+8.82e-02

Pos logit contributions
 mechanic+0.084
 videos+0.083
isions+0.083
care+0.082
 computer+0.081
Neg logit contributions
 Fif-0.116
 bamboo-0.103
iw-0.098
onna-0.098
io-0.097
Token toy
Token ID13373
Feature activation+9.56e-02

Pos logit contributions
isions+0.136
 videos+0.133
care+0.131
controlled+0.128
 nephew+0.127
Neg logit contributions
 Fif-0.150
 bamboo-0.135
 pear-0.126
onna-0.126
iw-0.125
Token and
Token ID290
Feature activation+1.09e-02

Pos logit contributions
 Fif+0.302
 bamboo+0.271
onna+0.259
iw+0.256
 pear+0.250
Neg logit contributions
isions-0.219
 videos-0.218
care-0.215
 mechanic-0.211
's-0.210
Token said
Token ID531
Feature activation+4.81e-02

Pos logit contributions
's+0.011
 Margaret+0.010
 nephew+0.010
care+0.010
 means+0.009
Neg logit contributions
 Fif-0.024
 bamboo-0.023
onna-0.023
iw-0.022
io-0.021
Token,
Token ID11
Feature activation-4.02e-03

Pos logit contributions
's+0.057
care+0.053
 means+0.052
 videos+0.051
 Margaret+0.051
Neg logit contributions
 Fif-0.095
 bamboo-0.089
iw-0.088
onna-0.085
sed-0.085
Token "
Token ID366
Feature activation-8.84e-02

Pos logit contributions
 Fif+0.008
 bamboo+0.007
iw+0.007
 pear+0.007
io+0.007
Neg logit contributions
's-0.005
 Margaret-0.005
care-0.005
 videos-0.005
 mechanic-0.005
TokenLet
Token ID5756
Feature activation+2.92e-03

Pos logit contributions
 Fif+0.153
 bamboo+0.139
onna+0.134
sed+0.130
 pear+0.129
Neg logit contributions
isions-0.131
 nephew-0.129
care-0.128
 mechanic-0.127
's-0.127
Token's
Token ID338
Feature activation+3.79e-01

Pos logit contributions
 keyboard+0.003
isions+0.003
 videos+0.003
 computers+0.003
 engineer+0.002
Neg logit contributions
 Fif-0.007
 bamboo-0.006
 l-0.006
 pear-0.006
 honey-0.006
Token go
Token ID467
Feature activation+1.13e-01

Pos logit contributions
isions+0.452
 videos+0.452
care+0.442
's+0.442
 mechanic+0.439
Neg logit contributions
 Fif-0.756
 bamboo-0.686
iw-0.662
onna-0.660
sed-0.642
Token home
Token ID1363
Feature activation-1.03e-01

Pos logit contributions
's+0.155
care+0.144
 videos+0.142
 mechanic+0.142
 jet+0.140
Neg logit contributions
 Fif-0.211
sed-0.189
iw-0.184
 bamboo-0.181
io-0.179
Token now
Token ID783
Feature activation-2.25e-01

Pos logit contributions
 Fif+0.189
 bamboo+0.169
iw+0.159
sed+0.157
 pear+0.154
Neg logit contributions
's-0.146
care-0.145
isions-0.143
 videos-0.141
 mechanic-0.140
Token!"
Token ID2474
Feature activation-9.67e-02

Pos logit contributions
 Fif+0.373
 bamboo+0.329
onna+0.311
iw+0.309
 pear+0.300
Neg logit contributions
isions-0.353
 videos-0.351
care-0.347
 mechanic-0.340
's-0.338
Token
Token ID198
Feature activation-3.04e-02

Pos logit contributions
 Fif+0.160
 bamboo+0.134
iw+0.125
onna+0.123
 pear+0.122
Neg logit contributions
isions-0.160
 videos-0.159
 mechanic-0.155
controlled-0.153
 keyboard-0.152
Token happy
Token ID3772
Feature activation-4.80e-02

Pos logit contributions
 Fif+0.104
 bamboo+0.096
onna+0.092
sed+0.090
iw+0.088
Neg logit contributions
's-0.072
isions-0.070
care-0.069
 videos-0.069
 complicated-0.066
Token.
Token ID13
Feature activation-1.01e-02

Pos logit contributions
 Fif+0.093
 bamboo+0.084
onna+0.080
iw+0.079
 pear+0.078
Neg logit contributions
isions-0.062
 videos-0.062
care-0.062
's-0.060
 nephew-0.058
Token That
Token ID1320
Feature activation+7.44e-02

Pos logit contributions
 Fif+0.019
 bamboo+0.017
onna+0.016
iw+0.016
sed+0.016
Neg logit contributions
isions-0.013
 mechanic-0.013
 videos-0.013
care-0.013
controlled-0.013
Token day
Token ID1110
Feature activation+1.97e-02

Pos logit contributions
isions+0.129
care+0.125
controlled+0.125
ever+0.123
 keyboard+0.120
Neg logit contributions
 Fif-0.112
 bamboo-0.098
 pear-0.096
 l-0.094
iw-0.092
Token,
Token ID11
Feature activation-7.80e-03

Pos logit contributions
isions+0.030
 videos+0.029
 mechanic+0.029
care+0.028
controlled+0.028
Neg logit contributions
 Fif-0.034
 bamboo-0.030
onna-0.028
iw-0.028
 pear-0.028
Token Sarah
Token ID10490
Feature activation+2.10e-01

Pos logit contributions
 Fif+0.014
 bamboo+0.012
iw+0.011
 l+0.011
 Flor+0.011
Neg logit contributions
isions-0.011
 mechanic-0.011
urious-0.011
controlled-0.010
 keyboard-0.010
Token had
Token ID550
Feature activation-4.25e-02

Pos logit contributions
's+0.279
isions+0.253
 means+0.251
 videos+0.247
 Margaret+0.247
Neg logit contributions
 Fif-0.416
 bamboo-0.385
iw-0.353
onna-0.350
sed-0.346
Token become
Token ID1716
Feature activation-4.32e-02

Pos logit contributions
 Fif+0.087
 bamboo+0.079
iw+0.074
sed+0.073
onna+0.072
Neg logit contributions
's-0.058
care-0.052
isions-0.050
 computers-0.049
 Margaret-0.049
Token bra
Token ID8290
Feature activation-1.52e-01

Pos logit contributions
 Fif+0.084
 bamboo+0.076
onna+0.075
sed+0.074
iw+0.072
Neg logit contributions
 videos-0.055
's-0.053
 mechanic-0.052
care-0.052
isions-0.052
Tokenver
Token ID332
Feature activation+1.51e-01

Pos logit contributions
 Fif+0.329
 bamboo+0.297
onna+0.292
iw+0.286
sed+0.282
Neg logit contributions
 videos-0.161
isions-0.159
care-0.153
 mechanic-0.151
 nephew-0.149
Token and
Token ID290
Feature activation+6.56e-03

Pos logit contributions
isions+0.261
 videos+0.254
 keyboard+0.248
 mechanic+0.246
care+0.245
Neg logit contributions
 Fif-0.231
 bamboo-0.194
onna-0.188
iw-0.188
io-0.179
Token
Token ID241
Feature activation-2.05e-02

Pos logit contributions
isions+0.031
 videos+0.031
care+0.031
 mechanic+0.030
 nephew+0.030
Neg logit contributions
 Fif-0.038
 bamboo-0.034
onna-0.032
iw-0.032
 pear-0.031
TokenM
Token ID44
Feature activation-1.25e-01

Pos logit contributions
 Fif+0.036
 bamboo+0.033
onna+0.030
sed+0.030
iw+0.030
Neg logit contributions
care-0.030
isions-0.029
 videos-0.029
's-0.028
 mechanic-0.028
Tokenummy
Token ID13513
Feature activation-3.45e-02

Pos logit contributions
 Fif+0.231
 bamboo+0.207
onna+0.195
sed+0.194
iw+0.191
Neg logit contributions
care-0.167
isions-0.167
 mechanic-0.167
's-0.162
 videos-0.161
Token,
Token ID11
Feature activation-1.26e-03

Pos logit contributions
 Fif+0.062
 bamboo+0.056
onna+0.053
iw+0.052
 pear+0.051
Neg logit contributions
isions-0.048
 videos-0.048
care-0.047
's-0.047
 mechanic-0.047
Token I
Token ID314
Feature activation-8.98e-02

Pos logit contributions
 Fif+0.002
 bamboo+0.002
onna+0.002
iw+0.002
 pear+0.002
Neg logit contributions
isions-0.002
 videos-0.002
care-0.002
 mechanic-0.002
's-0.002
Token need
Token ID761
Feature activation+2.13e-01

Pos logit contributions
 Fif+0.170
 bamboo+0.153
onna+0.146
iw+0.144
 pear+0.141
Neg logit contributions
isions-0.120
 videos-0.118
care-0.116
 mechanic-0.114
 nephew-0.113
Token that
Token ID326
Feature activation+5.22e-02

Pos logit contributions
's+0.259
care+0.252
 videos+0.244
 mechanic+0.239
 means+0.239
Neg logit contributions
 Fif-0.433
 bamboo-0.386
io-0.378
sed-0.376
iw-0.373
Token red
Token ID2266
Feature activation-5.41e-02

Pos logit contributions
isions+0.071
 videos+0.069
care+0.067
 nephew+0.066
 mechanic+0.065
Neg logit contributions
 Fif-0.098
 bamboo-0.088
onna-0.085
iw-0.085
 pear-0.084
Token stamp
Token ID17977
Feature activation+1.31e-01

Pos logit contributions
 Fif+0.103
 bamboo+0.093
iw+0.090
onna+0.089
 pear+0.088
Neg logit contributions
isions-0.071
 videos-0.069
care-0.068
 nephew-0.067
 mechanic-0.067
Token!
Token ID0
Feature activation-1.78e-02

Pos logit contributions
's+0.184
 videos+0.182
isions+0.178
care+0.176
 means+0.175
Neg logit contributions
 Fif-0.241
 bamboo-0.212
onna-0.202
iw-0.202
sed-0.196
Tokenâ
Token ID22940
Feature activation+7.62e-02

Pos logit contributions
 Fif+0.031
 bamboo+0.027
onna+0.026
iw+0.026
 pear+0.025
Neg logit contributions
isions-0.026
 videos-0.026
care-0.025
 jet-0.025
controlled-0.025
Token and
Token ID290
Feature activation-5.51e-04

Pos logit contributions
 Fif+0.300
 bamboo+0.266
onna+0.253
iw+0.248
 pear+0.245
Neg logit contributions
isions-0.257
 videos-0.253
care-0.248
 mechanic-0.244
controlled-0.243
Token said
Token ID531
Feature activation+3.53e-02

Pos logit contributions
 Fif+0.001
 bamboo+0.001
onna+0.001
iw+0.001
 pear+0.001
Neg logit contributions
's-0.001
care-0.001
 Margaret-0.001
 means-0.001
 nephew-0.001
Token,
Token ID11
Feature activation-1.51e-02

Pos logit contributions
's+0.043
care+0.039
 means+0.039
 videos+0.038
isions+0.037
Neg logit contributions
 Fif-0.071
 bamboo-0.067
iw-0.064
onna-0.062
sed-0.061
Token "
Token ID366
Feature activation-1.00e-01

Pos logit contributions
 Fif+0.027
 bamboo+0.025
iw+0.024
 pear+0.023
onna+0.022
Neg logit contributions
's-0.022
care-0.022
 videos-0.022
isions-0.021
 mechanic-0.021
TokenIt
Token ID1026
Feature activation-2.64e-02

Pos logit contributions
 Fif+0.169
 bamboo+0.149
onna+0.142
iw+0.141
 pear+0.136
Neg logit contributions
 videos-0.153
isions-0.153
care-0.150
 mechanic-0.148
 nephew-0.147
Token's
Token ID338
Feature activation+3.65e-01

Pos logit contributions
 Fif+0.061
 bamboo+0.055
 majesty+0.053
 pear+0.053
onna+0.052
Neg logit contributions
care-0.026
isions-0.023
 videos-0.023
 intersection-0.022
 keyboard-0.022
Token a
Token ID257
Feature activation+5.33e-02

Pos logit contributions
isions+0.489
 videos+0.487
care+0.479
 mechanic+0.469
's+0.467
Neg logit contributions
 Fif-0.689
 bamboo-0.618
onna-0.591
iw-0.587
 pear-0.571
Token surprise
Token ID5975
Feature activation-1.22e-02

Pos logit contributions
 videos+0.069
isions+0.068
 mechanic+0.067
care+0.067
 computer+0.066
Neg logit contributions
 Fif-0.103
 bamboo-0.092
onna-0.089
iw-0.088
io-0.085
Token.
Token ID13
Feature activation-1.49e-02

Pos logit contributions
 Fif+0.021
 bamboo+0.018
onna+0.017
iw+0.017
 pear+0.017
Neg logit contributions
isions-0.019
 videos-0.019
care-0.018
 mechanic-0.018
's-0.018
Token Would
Token ID10928
Feature activation-5.35e-02

Pos logit contributions
 Fif+0.027
 bamboo+0.023
onna+0.022
iw+0.022
 majesty+0.021
Neg logit contributions
isions-0.022
 videos-0.022
care-0.022
 jet-0.021
 computer-0.021
Token you
Token ID345
Feature activation+1.79e-02

Pos logit contributions
 Fif+0.086
 bamboo+0.077
onna+0.070
 pear+0.069
iw+0.069
Neg logit contributions
isions-0.086
 nephew-0.085
 mechanic-0.084
controlled-0.082
care-0.082
TokenSure
Token ID19457
Feature activation-1.03e-01

Pos logit contributions
 Fif+0.167
 bamboo+0.153
onna+0.146
sed+0.143
 pear+0.139
Neg logit contributions
care-0.156
isions-0.155
 mechanic-0.154
controlled-0.153
's-0.153
Token,
Token ID11
Feature activation-1.16e-02

Pos logit contributions
 Fif+0.215
 bamboo+0.199
onna+0.196
iw+0.193
sed+0.188
Neg logit contributions
's-0.119
 mechanic-0.109
 means-0.108
 videos-0.107
 jet-0.105
Token Tom
Token ID4186
Feature activation-5.06e-02

Pos logit contributions
 Fif+0.020
io+0.019
iw+0.019
 bamboo+0.018
sed+0.018
Neg logit contributions
's-0.016
 mechanic-0.016
 videos-0.016
 nephew-0.015
care-0.015
Token.
Token ID13
Feature activation-1.68e-02

Pos logit contributions
 Fif+0.105
 bamboo+0.096
sed+0.091
iw+0.090
onna+0.090
Neg logit contributions
's-0.065
 means-0.060
care-0.055
 nephew-0.055
 mechanic-0.055
Token Let
Token ID3914
Feature activation-2.58e-02

Pos logit contributions
 Fif+0.033
 bamboo+0.031
onna+0.030
io+0.029
 pear+0.029
Neg logit contributions
 mechanic-0.021
's-0.020
 nephew-0.020
 means-0.020
 videos-0.019
Token's
Token ID338
Feature activation+3.67e-01

Pos logit contributions
 Fif+0.066
 bamboo+0.061
 pear+0.059
io+0.059
 l+0.058
Neg logit contributions
isions-0.017
 keyboard-0.016
 videos-0.015
 nephew-0.014
 computers-0.014
Token go
Token ID467
Feature activation+9.97e-02

Pos logit contributions
isions+0.406
 videos+0.402
care+0.394
 mechanic+0.389
's+0.386
Neg logit contributions
 Fif-0.777
 bamboo-0.707
onna-0.680
iw-0.675
sed-0.656
Token.
Token ID13
Feature activation-1.53e-02

Pos logit contributions
's+0.138
care+0.125
 mechanic+0.125
controlled+0.123
 means+0.121
Neg logit contributions
 Fif-0.191
sed-0.178
 bamboo-0.165
onna-0.165
 majesty-0.161
Token I
Token ID314
Feature activation-9.85e-02

Pos logit contributions
 Fif+0.031
 bamboo+0.028
onna+0.027
iw+0.027
 pear+0.026
Neg logit contributions
isions-0.018
 videos-0.018
care-0.018
 mechanic-0.017
 nephew-0.017
Token think
Token ID892
Feature activation+5.13e-02

Pos logit contributions
 Fif+0.172
 bamboo+0.153
onna+0.145
iw+0.143
sed+0.140
Neg logit contributions
isions-0.144
's-0.143
care-0.143
 mechanic-0.143
 videos-0.141
Token you
Token ID345
Feature activation+1.88e-02

Pos logit contributions
isions+0.068
 videos+0.067
care+0.066
 mechanic+0.065
's+0.064
Neg logit contributions
 Fif-0.097
 bamboo-0.088
onna-0.084
iw-0.083
 pear-0.081
Token up
Token ID510
Feature activation+1.08e-01

Pos logit contributions
 Fif+0.067
 bamboo+0.060
onna+0.057
iw+0.057
 pear+0.055
Neg logit contributions
isions-0.055
 videos-0.055
care-0.055
's-0.054
 mechanic-0.053
Token and
Token ID290
Feature activation+1.09e-02

Pos logit contributions
's+0.130
 videos+0.130
care+0.128
 means+0.124
 computers+0.123
Neg logit contributions
 Fif-0.214
 bamboo-0.189
iw-0.186
 majesty-0.184
onna-0.181
Token kept
Token ID4030
Feature activation-1.27e-01

Pos logit contributions
isions+0.016
 videos+0.016
care+0.015
controlled+0.015
 mechanic+0.015
Neg logit contributions
 Fif-0.020
 bamboo-0.017
onna-0.016
iw-0.016
 pear-0.016
Token them
Token ID606
Feature activation-3.79e-02

Pos logit contributions
 Fif+0.229
 bamboo+0.204
onna+0.195
iw+0.192
 pear+0.188
Neg logit contributions
isions-0.180
 videos-0.179
care-0.177
 mechanic-0.173
's-0.171
Token in
Token ID287
Feature activation-2.52e-02

Pos logit contributions
 Fif+0.057
 bamboo+0.053
 pear+0.048
onna+0.048
iw+0.046
Neg logit contributions
isions-0.064
 videos-0.064
isha-0.062
 nephew-0.062
 Margaret-0.061
Token his
Token ID465
Feature activation-1.34e-02

Pos logit contributions
 Fif+0.050
 bamboo+0.045
onna+0.043
iw+0.043
 pear+0.042
Neg logit contributions
isions-0.031
 videos-0.031
care-0.031
 mechanic-0.030
 nephew-0.029
Token house
Token ID2156
Feature activation+1.46e-01

Pos logit contributions
 Fif+0.022
 bamboo+0.020
onna+0.019
iw+0.018
 pear+0.018
Neg logit contributions
isions-0.021
 videos-0.021
care-0.021
 mechanic-0.020
controlled-0.020
Token.
Token ID13
Feature activation-3.73e-03

Pos logit contributions
isions+0.215
 videos+0.213
care+0.211
 mechanic+0.207
's+0.206
Neg logit contributions
 Fif-0.255
 bamboo-0.226
onna-0.214
iw-0.213
 pear-0.207
Token
Token ID198
Feature activation-3.20e-02

Pos logit contributions
 Fif+0.007
 bamboo+0.007
onna+0.006
iw+0.006
 pear+0.006
Neg logit contributions
isions-0.005
 videos-0.005
care-0.004
 mechanic-0.004
's-0.004
Token
Token ID198
Feature activation-3.17e-02

Pos logit contributions
 Fif+0.059
 bamboo+0.053
iw+0.051
onna+0.051
sed+0.050
Neg logit contributions
 videos-0.043
isions-0.043
care-0.042
's-0.042
 mechanic-0.042
TokenThe
Token ID464
Feature activation+1.06e-01

Pos logit contributions
 Fif+0.057
 bamboo+0.051
onna+0.048
iw+0.047
 pear+0.047
Neg logit contributions
 videos-0.045
 nephew-0.044
isions-0.044
care-0.044
 computers-0.043
Token rain
Token ID6290
Feature activation+1.25e-02

Pos logit contributions
 videos+0.012
isions+0.011
care+0.011
 mechanic+0.011
 jet+0.011
Neg logit contributions
 Fif-0.017
 bamboo-0.015
onna-0.015
iw-0.015
io-0.014
Token.
Token ID13
Feature activation-5.69e-03

Pos logit contributions
isions+0.019
 videos+0.019
care+0.018
 mechanic+0.018
 nephew+0.018
Neg logit contributions
 Fif-0.021
 bamboo-0.019
onna-0.018
iw-0.018
 pear-0.017
Token But
Token ID887
Feature activation+5.79e-03

Pos logit contributions
 Fif+0.011
 bamboo+0.010
onna+0.009
iw+0.009
 pear+0.009
Neg logit contributions
isions-0.008
 videos-0.007
care-0.007
 mechanic-0.007
 nephew-0.007
Token you
Token ID345
Feature activation+2.45e-02

Pos logit contributions
's+0.008
 videos+0.008
care+0.008
isions+0.008
 means+0.008
Neg logit contributions
 Fif-0.011
 bamboo-0.010
onna-0.009
iw-0.009
sed-0.009
Token knew
Token ID2993
Feature activation-1.06e-01

Pos logit contributions
isions+0.034
 videos+0.033
care+0.033
 mechanic+0.032
 nephew+0.032
Neg logit contributions
 Fif-0.046
 bamboo-0.041
onna-0.039
iw-0.038
 pear-0.038
Token that
Token ID326
Feature activation+4.92e-02

Pos logit contributions
 Fif+0.200
 bamboo+0.181
onna+0.173
iw+0.173
sed+0.168
Neg logit contributions
isions-0.138
 videos-0.137
care-0.136
's-0.134
 mechanic-0.132
Token walking
Token ID6155
Feature activation+6.59e-02

Pos logit contributions
isions+0.072
care+0.068
 videos+0.068
controlled+0.068
 mechanic+0.067
Neg logit contributions
 Fif-0.087
 bamboo-0.077
onna-0.074
 pear-0.072
iw-0.071
Token in
Token ID287
Feature activation-3.24e-02

Pos logit contributions
 videos+0.094
isions+0.094
care+0.092
's+0.090
 mechanic+0.090
Neg logit contributions
 Fif-0.118
 bamboo-0.105
onna-0.101
iw-0.099
 pear-0.097
Token it
Token ID340
Feature activation+2.02e-02

Pos logit contributions
 Fif+0.059
 bamboo+0.053
onna+0.050
iw+0.049
 pear+0.048
Neg logit contributions
isions-0.046
 videos-0.045
care-0.045
 mechanic-0.044
 nephew-0.043
Token would
Token ID561
Feature activation-3.44e-02

Pos logit contributions
isions+0.029
 videos+0.029
care+0.029
 mechanic+0.028
's+0.028
Neg logit contributions
 Fif-0.036
 bamboo-0.032
onna-0.030
iw-0.030
 pear-0.029
Token bring
Token ID2222
Feature activation+1.69e-02

Pos logit contributions
 Fif+0.061
iw+0.056
 bamboo+0.055
io+0.053
sed+0.053
Neg logit contributions
's-0.049
 videos-0.047
 mechanic-0.046
care-0.046
 jet-0.046
Token sees
Token ID7224
Feature activation+1.24e-02

Pos logit contributions
 Fif+0.052
 bamboo+0.047
onna+0.044
iw+0.044
io+0.042
Neg logit contributions
's-0.031
 means-0.028
isions-0.026
 mechanic-0.026
 videos-0.026
Token a
Token ID257
Feature activation+6.60e-02

Pos logit contributions
 videos+0.013
's+0.013
 mechanic+0.013
 computers+0.012
isions+0.012
Neg logit contributions
 Fif-0.026
onna-0.024
iw-0.023
 bamboo-0.023
 majesty-0.023
Token tw
Token ID665
Feature activation-2.18e-01

Pos logit contributions
 mechanic+0.092
 jet+0.090
 videos+0.089
 computer+0.089
isions+0.089
Neg logit contributions
 Fif-0.121
 bamboo-0.106
io-0.102
iw-0.101
onna-0.101
Tokenig
Token ID328
Feature activation-1.32e-01

Pos logit contributions
 Fif+0.456
 bamboo+0.398
iw+0.368
sed+0.366
onna+0.362
Neg logit contributions
 videos-0.249
 jet-0.248
's-0.248
care-0.246
isions-0.245
Token,
Token ID11
Feature activation+2.03e-03

Pos logit contributions
 Fif+0.231
 bamboo+0.203
iw+0.194
onna+0.191
 pear+0.185
Neg logit contributions
 videos-0.193
's-0.193
isions-0.192
care-0.189
 mechanic-0.188
Token a
Token ID257
Feature activation+6.57e-02

Pos logit contributions
 videos+0.002
isions+0.002
care+0.002
's+0.002
 mechanic+0.002
Neg logit contributions
 Fif-0.004
 bamboo-0.004
onna-0.004
iw-0.004
io-0.004
Token string
Token ID4731
Feature activation+9.31e-02

Pos logit contributions
 jet+0.100
 mechanic+0.098
 computer+0.097
 keyboard+0.095
 radio+0.095
Neg logit contributions
 Fif-0.114
io-0.101
 bamboo-0.097
iw-0.094
onna-0.093
Token,
Token ID11
Feature activation+1.84e-03

Pos logit contributions
's+0.119
 jet+0.110
 means+0.108
 mechanic+0.108
 videos+0.107
Neg logit contributions
 Fif-0.187
 bamboo-0.163
iw-0.159
 majesty-0.157
onna-0.154
Token and
Token ID290
Feature activation+1.36e-02

Pos logit contributions
's+0.002
 videos+0.002
 means+0.002
 nephew+0.002
 mechanic+0.001
Neg logit contributions
 Fif-0.004
onna-0.004
iw-0.004
 bamboo-0.004
io-0.004
Token a
Token ID257
Feature activation+6.57e-02

Pos logit contributions
 videos+0.016
isions+0.016
's+0.016
care+0.016
 mechanic+0.015
Neg logit contributions
 Fif-0.028
 bamboo-0.025
onna-0.024
iw-0.024
io-0.023
Token paper
Token ID3348
Feature activation+1.57e-01

Pos logit contributions
 jet+0.098
 mechanic+0.097
 keyboard+0.096
 computer+0.096
 radio+0.094
Neg logit contributions
 Fif-0.116
io-0.101
 bamboo-0.100
iw-0.096
onna-0.093
Token happy
Token ID3772
Feature activation-4.92e-02

Pos logit contributions
 Fif+0.029
 bamboo+0.025
onna+0.024
 pear+0.023
iw+0.023
Neg logit contributions
isions-0.031
controlled-0.029
 videos-0.029
ographs-0.029
care-0.029
Token to
Token ID284
Feature activation-3.20e-02

Pos logit contributions
 Fif+0.093
 bamboo+0.083
onna+0.079
iw+0.079
sed+0.076
Neg logit contributions
 videos-0.067
care-0.067
isions-0.067
's-0.065
 mechanic-0.065
Token help
Token ID1037
Feature activation-1.22e-01

Pos logit contributions
 Fif+0.065
 bamboo+0.059
iw+0.057
sed+0.055
onna+0.055
Neg logit contributions
 mechanic-0.039
 videos-0.039
care-0.038
's-0.038
isions-0.037
Token and
Token ID290
Feature activation+3.40e-03

Pos logit contributions
 Fif+0.147
 bamboo+0.125
 himself+0.121
 herself+0.114
onna+0.112
Neg logit contributions
ographs-0.245
urious-0.241
isions-0.241
controlled-0.230
itionally-0.223
Token quickly
Token ID2952
Feature activation-4.93e-02

Pos logit contributions
isions+0.005
care+0.005
 videos+0.005
ographs+0.005
controlled+0.005
Neg logit contributions
 Fif-0.006
 bamboo-0.005
onna-0.005
 pear-0.005
iw-0.005
Token unp
Token ID8593
Feature activation+9.62e-03

Pos logit contributions
 Fif+0.080
 bamboo+0.072
onna+0.070
 pear+0.068
 ti+0.067
Neg logit contributions
isions-0.077
 videos-0.075
ographs-0.073
controlled-0.072
 nephew-0.071
Tokenacked
Token ID6021
Feature activation+4.64e-03

Pos logit contributions
 nephew+0.018
 jet+0.017
 Margaret+0.017
 Steven+0.017
 keyboard+0.017
Neg logit contributions
 Fif-0.012
io-0.011
 bamboo-0.011
iw-0.010
onna-0.010
Token his
Token ID465
Feature activation-2.78e-02

Pos logit contributions
isions+0.008
care+0.008
 nephew+0.007
 mechanic+0.007
 keyboard+0.007
Neg logit contributions
 Fif-0.007
 bamboo-0.006
 pear-0.006
onna-0.006
io-0.006
Token things
Token ID1243
Feature activation+1.12e-01

Pos logit contributions
 Fif+0.043
 bamboo+0.039
 pear+0.036
iw+0.035
onna+0.035
Neg logit contributions
isions-0.046
 videos-0.046
care-0.045
 mechanic-0.044
 Margaret-0.044
Token.
Token ID13
Feature activation-4.02e-03

Pos logit contributions
isions+0.202
 videos+0.201
 nephew+0.198
ographs+0.194
 Margaret+0.192
Neg logit contributions
 Fif-0.146
 bamboo-0.131
onna-0.128
 pear-0.120
iw-0.119
Token
Token ID198
Feature activation-3.42e-02

Pos logit contributions
 Fif+0.006
 bamboo+0.005
onna+0.005
iw+0.005
 pear+0.005
Neg logit contributions
isions-0.007
 videos-0.007
controlled-0.006
care-0.006
 jet-0.006
Token picture
Token ID4286
Feature activation+1.95e-01

Pos logit contributions
isions+0.066
care+0.064
 videos+0.063
controlled+0.063
 nephew+0.061
Neg logit contributions
 Fif-0.091
 bamboo-0.081
iw-0.077
 pear-0.077
onna-0.077
Token.
Token ID13
Feature activation+6.74e-04

Pos logit contributions
 videos+0.283
's+0.281
isions+0.278
care+0.276
 computer+0.270
Neg logit contributions
 Fif-0.354
 bamboo-0.310
iw-0.294
onna-0.292
sed-0.286
Token Can
Token ID1680
Feature activation-2.28e-02

Pos logit contributions
 videos+0.001
isions+0.001
care+0.001
 mechanic+0.001
's+0.001
Neg logit contributions
 Fif-0.001
 bamboo-0.001
onna-0.001
iw-0.001
 pear-0.001
Token you
Token ID345
Feature activation+2.28e-02

Pos logit contributions
 Fif+0.037
 bamboo+0.033
onna+0.031
iw+0.030
 pear+0.030
Neg logit contributions
isions-0.037
 videos-0.036
care-0.036
 nephew-0.035
 mechanic-0.035
Token set
Token ID900
Feature activation-2.84e-02

Pos logit contributions
isions+0.026
 videos+0.026
care+0.026
 mechanic+0.025
's+0.025
Neg logit contributions
 Fif-0.047
 bamboo-0.043
onna-0.041
iw-0.041
 pear-0.040
Token it
Token ID340
Feature activation+1.41e-02

Pos logit contributions
 Fif+0.058
 bamboo+0.052
onna+0.050
iw+0.050
sed+0.049
Neg logit contributions
 videos-0.036
's-0.034
isions-0.034
care-0.033
 jet-0.033
Token on
Token ID319
Feature activation-4.82e-02

Pos logit contributions
 videos+0.019
care+0.018
's+0.018
isions+0.018
 mechanic+0.018
Neg logit contributions
 Fif-0.027
 bamboo-0.024
iw-0.023
onna-0.023
sed-0.023
Token the
Token ID262
Feature activation+7.04e-03

Pos logit contributions
 Fif+0.081
 bamboo+0.073
onna+0.069
iw+0.067
 pear+0.067
Neg logit contributions
isions-0.074
care-0.071
controlled-0.071
 nephew-0.070
 videos-0.070
Token table
Token ID3084
Feature activation+5.44e-02

Pos logit contributions
 videos+0.008
isions+0.008
's+0.008
 computer+0.008
 mechanic+0.008
Neg logit contributions
 Fif-0.015
 bamboo-0.013
onna-0.013
iw-0.013
 pear-0.012
Token for
Token ID329
Feature activation-1.04e-01

Pos logit contributions
isions+0.062
 videos+0.062
care+0.061
 mechanic+0.060
's+0.059
Neg logit contributions
 Fif-0.113
 bamboo-0.102
onna-0.097
iw-0.097
 pear-0.095
Token me
Token ID502
Feature activation-1.04e-01

Pos logit contributions
 Fif+0.184
 bamboo+0.164
onna+0.156
iw+0.154
 pear+0.150
Neg logit contributions
isions-0.152
 videos-0.150
care-0.148
 mechanic-0.145
 nephew-0.144
Token fish
Token ID5916
Feature activation-1.42e-01

Pos logit contributions
 jet+0.008
 mechanic+0.007
 videos+0.007
's+0.007
 computer+0.007
Neg logit contributions
 Fif-0.008
io-0.007
 bamboo-0.007
onna-0.007
iw-0.007
Token,
Token ID11
Feature activation-9.39e-03

Pos logit contributions
 Fif+0.273
 bamboo+0.242
onna+0.233
iw+0.229
sed+0.225
Neg logit contributions
isions-0.182
's-0.182
 videos-0.181
care-0.180
 mechanic-0.176
Token and
Token ID290
Feature activation+2.85e-03

Pos logit contributions
 Fif+0.018
 bamboo+0.016
io+0.015
onna+0.015
iw+0.015
Neg logit contributions
's-0.012
 videos-0.012
care-0.012
isions-0.012
 mechanic-0.012
Token he
Token ID339
Feature activation+1.78e-02

Pos logit contributions
isions+0.004
 videos+0.004
care+0.004
 mechanic+0.004
's+0.004
Neg logit contributions
 Fif-0.005
 bamboo-0.004
onna-0.004
iw-0.004
 pear-0.004
Token was
Token ID373
Feature activation-1.97e-02

Pos logit contributions
's+0.026
isions+0.023
 videos+0.023
 mechanic+0.023
 means+0.022
Neg logit contributions
 Fif-0.033
 bamboo-0.030
iw-0.029
onna-0.028
io-0.027
Token very
Token ID845
Feature activation-7.05e-02

Pos logit contributions
 Fif+0.039
iw+0.038
io+0.037
ho+0.036
 bamboo+0.036
Neg logit contributions
's-0.024
 videos-0.020
eping-0.020
-0.019
 mechanic-0.019
Token happy
Token ID3772
Feature activation-5.33e-02

Pos logit contributions
 Fif+0.139
 bamboo+0.131
io+0.129
iw+0.123
sed+0.121
Neg logit contributions
's-0.080
 complicated-0.076
care-0.074
 videos-0.072
 costly-0.070
Token.
Token ID13
Feature activation-1.41e-02

Pos logit contributions
 Fif+0.117
 bamboo+0.104
iw+0.101
onna+0.101
io+0.100
Neg logit contributions
care-0.059
's-0.057
 videos-0.055
isions-0.051
 computer-0.051
Token
Token ID198
Feature activation-3.73e-02

Pos logit contributions
 Fif+0.024
 bamboo+0.021
onna+0.019
iw+0.019
 pear+0.019
Neg logit contributions
isions-0.022
 videos-0.022
care-0.022
controlled-0.021
 computers-0.021
Token
Token ID198
Feature activation-3.75e-02

Pos logit contributions
 Fif+0.065
 bamboo+0.058
onna+0.055
iw+0.054
 pear+0.053
Neg logit contributions
isions-0.056
 videos-0.055
care-0.054
 mechanic-0.053
 nephew-0.053
TokenAs
Token ID1722
Feature activation+2.11e-01

Pos logit contributions
 Fif+0.061
 bamboo+0.055
onna+0.047
 pear+0.047
iw+0.047
Neg logit contributions
 videos-0.060
 means-0.058
care-0.057
isions-0.056
 computers-0.056
Token and
Token ID290
Feature activation+6.22e-03

Pos logit contributions
 Fif+0.271
 bamboo+0.242
 majesty+0.233
iw+0.232
sed+0.225
Neg logit contributions
's-0.160
care-0.146
 videos-0.134
 mechanic-0.133
isions-0.132
Token a
Token ID257
Feature activation+5.89e-02

Pos logit contributions
's+0.009
 videos+0.009
 means+0.008
 nephew+0.008
care+0.008
Neg logit contributions
 Fif-0.011
 bamboo-0.010
iw-0.010
onna-0.010
io-0.010
Token necklace
Token ID38987
Feature activation-1.12e-01

Pos logit contributions
 jet+0.087
 keyboard+0.087
 radio+0.087
's+0.086
 computer+0.085
Neg logit contributions
 Fif-0.105
io-0.090
iw-0.087
 bamboo-0.085
igi-0.082
Token around
Token ID1088
Feature activation+4.62e-02

Pos logit contributions
 Fif+0.257
iw+0.225
 majesty+0.222
 bamboo+0.216
onna+0.214
Neg logit contributions
's-0.124
 jet-0.097
 videos-0.096
house-0.095
care-0.094
Token its
Token ID663
Feature activation-3.38e-02

Pos logit contributions
 videos+0.053
care+0.049
's+0.048
 Margaret+0.046
 computers+0.046
Neg logit contributions
 Fif-0.097
 majesty-0.091
iw-0.089
sed-0.089
onna-0.088
Token neck
Token ID7393
Feature activation-2.07e-01

Pos logit contributions
 Fif+0.058
iw+0.055
io+0.050
Deal+0.049
onna+0.048
Neg logit contributions
's-0.054
 radio-0.049
 videos-0.048
 nephew-0.047
 computer-0.047
Token.
Token ID13
Feature activation-1.08e-02

Pos logit contributions
 Fif+0.451
iw+0.395
sed+0.389
 bamboo+0.386
 majesty+0.380
Neg logit contributions
's-0.255
care-0.218
 videos-0.210
isions-0.205
 means-0.202
Token She
Token ID1375
Feature activation-1.89e-02

Pos logit contributions
 Fif+0.022
 bamboo+0.021
io+0.020
sed+0.020
iw+0.020
Neg logit contributions
care-0.012
 videos-0.012
 Margaret-0.011
 means-0.011
 mechanic-0.011
Token thought
Token ID1807
Feature activation-8.46e-02

Pos logit contributions
 Fif+0.037
 bamboo+0.034
iw+0.033
ags+0.032
onna+0.031
Neg logit contributions
's-0.026
 means-0.023
 videos-0.022
 mechanic-0.021
 nephew-0.021
Token it
Token ID340
Feature activation+1.67e-02

Pos logit contributions
 Fif+0.174
iw+0.161
 bamboo+0.159
io+0.155
sed+0.152
Neg logit contributions
 videos-0.098
's-0.091
care-0.089
 means-0.086
 computer-0.086
Token looked
Token ID3114
Feature activation-2.77e-02

Pos logit contributions
's+0.021
 means+0.018
 videos+0.018
isions+0.017
 nephew+0.016
Neg logit contributions
 Fif-0.036
iw-0.033
onna-0.032
 bamboo-0.032
 majesty-0.031
Token medicine
Token ID9007
Feature activation+4.89e-02

Pos logit contributions
isions+0.017
 videos+0.017
care+0.016
 mechanic+0.016
's+0.016
Neg logit contributions
 Fif-0.018
 bamboo-0.016
onna-0.015
iw-0.015
 pear-0.015
Token.
Token ID13
Feature activation-5.98e-03

Pos logit contributions
's+0.059
 means+0.054
care+0.053
controlled+0.050
 videos+0.049
Neg logit contributions
 Fif-0.106
 bamboo-0.093
iw-0.093
 majesty-0.091
 pear-0.089
Token He
Token ID679
Feature activation-2.67e-02

Pos logit contributions
 Fif+0.011
 bamboo+0.010
onna+0.009
iw+0.009
 pear+0.009
Neg logit contributions
isions-0.009
 videos-0.008
care-0.008
 jet-0.008
 nephew-0.008
Token said
Token ID531
Feature activation+4.78e-02

Pos logit contributions
 Fif+0.050
 bamboo+0.045
onna+0.042
iw+0.042
 pear+0.041
Neg logit contributions
isions-0.036
's-0.036
 videos-0.035
 mechanic-0.035
 nephew-0.035
Token it
Token ID340
Feature activation+1.98e-02

Pos logit contributions
's+0.059
 videos+0.055
 means+0.054
 mechanic+0.053
care+0.053
Neg logit contributions
 Fif-0.093
 bamboo-0.086
iw-0.085
sed-0.084
io-0.081
Token would
Token ID561
Feature activation-3.30e-02

Pos logit contributions
isions+0.028
 videos+0.028
care+0.027
 mechanic+0.027
's+0.027
Neg logit contributions
 Fif-0.036
 bamboo-0.032
onna-0.031
iw-0.030
 pear-0.030
Token heal
Token ID12035
Feature activation+5.32e-02

Pos logit contributions
 Fif+0.064
io+0.059
iw+0.059
 bamboo+0.058
sed+0.055
Neg logit contributions
's-0.040
 mechanic-0.040
 pray-0.039
 videos-0.038
 jet-0.037
Token in
Token ID287
Feature activation-2.63e-02

Pos logit contributions
's+0.072
 videos+0.070
 nephew+0.068
 means+0.067
 mechanic+0.067
Neg logit contributions
 Fif-0.100
 bamboo-0.090
onna-0.088
iw-0.087
sed-0.084
Token a
Token ID257
Feature activation+6.38e-02

Pos logit contributions
 Fif+0.046
 bamboo+0.041
onna+0.039
iw+0.039
 pear+0.038
Neg logit contributions
isions-0.040
 videos-0.038
care-0.038
controlled-0.037
 mechanic-0.036
Token few
Token ID1178
Feature activation+1.08e-01

Pos logit contributions
isions+0.100
 videos+0.097
care+0.094
controlled+0.092
 means+0.092
Neg logit contributions
 Fif-0.106
 bamboo-0.099
 pear-0.092
iw-0.091
onna-0.091
Token weeks
Token ID2745
Feature activation-1.03e-01

Pos logit contributions
isions+0.194
 videos+0.190
care+0.188
 mechanic+0.185
 nephew+0.184
Neg logit contributions
 Fif-0.154
 bamboo-0.136
onna-0.124
iw-0.124
 pear-0.122
Token she
Token ID673
Feature activation+9.60e-03

Pos logit contributions
isions+0.033
 videos+0.032
care+0.032
 mechanic+0.031
 nephew+0.031
Neg logit contributions
 Fif-0.042
 bamboo-0.038
onna-0.036
iw-0.036
 pear-0.035
Token wanted
Token ID2227
Feature activation+1.54e-02

Pos logit contributions
isions+0.014
's+0.014
 videos+0.014
care+0.014
 mechanic+0.013
Neg logit contributions
 Fif-0.017
 bamboo-0.015
onna-0.014
iw-0.014
 pear-0.014
Token to
Token ID284
Feature activation-3.13e-02

Pos logit contributions
's+0.020
 videos+0.020
care+0.019
 computers+0.019
isions+0.019
Neg logit contributions
 Fif-0.030
iw-0.027
 bamboo-0.026
sed-0.026
io-0.025
Token arrange
Token ID21674
Feature activation+8.96e-02

Pos logit contributions
 Fif+0.059
 bamboo+0.054
iw+0.052
onna+0.051
sed+0.050
Neg logit contributions
isions-0.040
 videos-0.040
's-0.040
 mechanic-0.040
care-0.040
Token things
Token ID1243
Feature activation+1.09e-01

Pos logit contributions
 videos+0.120
care+0.113
isions+0.112
 computers+0.112
's+0.112
Neg logit contributions
 Fif-0.171
sed-0.152
onna-0.149
iw-0.149
 bamboo-0.148
Token just
Token ID655
Feature activation-8.00e-02

Pos logit contributions
's+0.141
care+0.139
 videos+0.135
isions+0.135
 means+0.132
Neg logit contributions
 Fif-0.213
 bamboo-0.189
sed-0.184
iw-0.182
 majesty-0.177
Token right
Token ID826
Feature activation-1.74e-01

Pos logit contributions
 Fif+0.145
 bamboo+0.129
iw+0.125
onna+0.120
sed+0.119
Neg logit contributions
's-0.115
 videos-0.111
care-0.111
 means-0.110
 computers-0.108
Token.
Token ID13
Feature activation-1.15e-02

Pos logit contributions
 Fif+0.343
 bamboo+0.305
iw+0.297
onna+0.294
sed+0.290
Neg logit contributions
's-0.223
care-0.220
 means-0.216
isions-0.213
 videos-0.212
Token She
Token ID1375
Feature activation-2.53e-02

Pos logit contributions
 Fif+0.021
 bamboo+0.018
onna+0.017
iw+0.017
 pear+0.017
Neg logit contributions
isions-0.016
 videos-0.016
controlled-0.016
 mechanic-0.016
care-0.016
Token wanted
Token ID2227
Feature activation+1.49e-02

Pos logit contributions
 Fif+0.047
 bamboo+0.042
iw+0.040
onna+0.040
sed+0.038
Neg logit contributions
isions-0.034
's-0.034
 videos-0.034
care-0.034
 mechanic-0.033
Token the
Token ID262
Feature activation-2.14e-06

Pos logit contributions
 videos+0.021
care+0.020
isions+0.020
's+0.020
 computers+0.020
Neg logit contributions
 Fif-0.028
 bamboo-0.025
iw-0.024
sed-0.023
onna-0.023
Token laughed
Token ID13818
Feature activation-1.42e-01

Pos logit contributions
 Fif+0.042
 bamboo+0.039
iw+0.037
onna+0.036
io+0.035
Neg logit contributions
's-0.033
care-0.031
 videos-0.030
isions-0.030
 jet-0.030
Token and
Token ID290
Feature activation+7.60e-03

Pos logit contributions
 Fif+0.294
iw+0.265
 bamboo+0.265
onna+0.258
 pear+0.247
Neg logit contributions
's-0.171
 videos-0.170
care-0.165
isions-0.161
 jet-0.155
Token said
Token ID531
Feature activation+3.74e-02

Pos logit contributions
's+0.011
care+0.010
 videos+0.010
 Margaret+0.010
 Kelly+0.010
Neg logit contributions
 Fif-0.013
 bamboo-0.013
iw-0.013
onna-0.013
io-0.012
Token to
Token ID284
Feature activation-3.28e-02

Pos logit contributions
's+0.038
care+0.034
 videos+0.034
 means+0.031
+0.031
Neg logit contributions
iw-0.082
 bamboo-0.079
 Fif-0.078
igi-0.075
io-0.075
Token her
Token ID607
Feature activation-5.69e-02

Pos logit contributions
iw+0.060
balls+0.057
 bamboo+0.056
sed+0.055
onna+0.055
Neg logit contributions
's-0.046
 pray-0.044
 videos-0.043
 Margaret-0.043
 Kelly-0.043
Token mum
Token ID25682
Feature activation-1.80e-01

Pos logit contributions
 Fif+0.103
onna+0.095
 bamboo+0.095
iw+0.094
io+0.087
Neg logit contributions
 videos-0.083
 computer-0.081
's-0.080
 nephew-0.079
 jet-0.078
Token,
Token ID11
Feature activation-7.56e-03

Pos logit contributions
 Fif+0.344
 bamboo+0.312
iw+0.303
onna+0.300
sed+0.295
Neg logit contributions
's-0.237
 videos-0.232
isions-0.225
care-0.223
 nephew-0.220
Token "
Token ID366
Feature activation-9.58e-02

Pos logit contributions
 bamboo+0.015
 Fif+0.015
iw+0.014
io+0.014
onna+0.014
Neg logit contributions
's-0.009
care-0.009
 Margaret-0.008
-0.008
 Steven-0.008
TokenM
Token ID44
Feature activation-1.36e-01

Pos logit contributions
 Fif+0.187
 bamboo+0.174
sed+0.171
onna+0.167
 majesty+0.161
Neg logit contributions
care-0.130
's-0.129
isions-0.128
-0.124
ographs-0.121
Tokenummy
Token ID13513
Feature activation-4.33e-02

Pos logit contributions
 Fif+0.253
 bamboo+0.238
sed+0.228
balls+0.224
onna+0.217
Neg logit contributions
care-0.171
isions-0.168
urious-0.164
isha-0.162
 means-0.160
Token,
Token ID11
Feature activation-6.84e-03

Pos logit contributions
 Fif+0.080
 bamboo+0.072
iw+0.068
onna+0.068
sed+0.065
Neg logit contributions
's-0.059
 videos-0.059
isions-0.058
care-0.058
 mechanic-0.057
Token who
Token ID508
Feature activation+2.37e-01

Pos logit contributions
isions+0.029
 videos+0.029
's+0.028
care+0.028
 nephew+0.027
Neg logit contributions
 Fif-0.062
 bamboo-0.056
onna-0.056
sed-0.054
iw-0.054
Token could
Token ID714
Feature activation-7.23e-02

Pos logit contributions
isions+0.365
 videos+0.362
care+0.357
 mechanic+0.349
 nephew+0.346
Neg logit contributions
 Fif-0.399
 bamboo-0.353
onna-0.334
iw-0.331
 pear-0.323
Token shoot
Token ID2686
Feature activation+1.93e-01

Pos logit contributions
 Fif+0.143
 bamboo+0.130
iw+0.128
onna+0.122
 majesty+0.121
Neg logit contributions
's-0.086
 mechanic-0.085
 videos-0.085
isions-0.084
care-0.082
Token the
Token ID262
Feature activation+1.52e-02

Pos logit contributions
isions+0.285
 videos+0.280
care+0.278
 mechanic+0.274
 nephew+0.271
Neg logit contributions
 Fif-0.337
 bamboo-0.300
onna-0.282
iw-0.280
 pear-0.275
Token most
Token ID749
Feature activation-6.86e-02

Pos logit contributions
's+0.031
+0.028
isions+0.027
care+0.027
 computer+0.027
Neg logit contributions
 Fif-0.019
onna-0.017
sed-0.016
 bamboo-0.015
io-0.015
Token strawberries
Token ID36973
Feature activation-3.37e-01

Pos logit contributions
 Fif+0.119
sed+0.103
 bamboo+0.103
onna+0.100
iw+0.098
Neg logit contributions
's-0.102
isions-0.101
 videos-0.100
care-0.099
 means-0.097
Token.
Token ID13
Feature activation+2.11e-03

Pos logit contributions
 Fif+0.663
 bamboo+0.585
 majesty+0.568
iw+0.565
sed+0.564
Neg logit contributions
's-0.436
care-0.415
 videos-0.411
isions-0.410
 means-0.405
Token
Token ID220
Feature activation-5.75e-02

Pos logit contributions
isions+0.003
 videos+0.003
 mechanic+0.003
controlled+0.003
care+0.003
Neg logit contributions
 Fif-0.003
 bamboo-0.003
onna-0.003
iw-0.003
 pear-0.003
Token
Token ID198
Feature activation-2.45e-02

Pos logit contributions
 Fif+0.095
 bamboo+0.083
onna+0.080
iw+0.078
 pear+0.076
Neg logit contributions
isions-0.090
 videos-0.090
care-0.087
 mechanic-0.087
 nephew-0.086
Token
Token ID198
Feature activation-2.50e-02

Pos logit contributions
 Fif+0.045
 bamboo+0.041
onna+0.038
iw+0.038
sed+0.038
Neg logit contributions
isions-0.034
care-0.034
 videos-0.033
's-0.033
 mechanic-0.032
TokenSam
Token ID16305
Feature activation-4.16e-02

Pos logit contributions
 Fif+0.039
 bamboo+0.034
iw+0.033
onna+0.032
 pear+0.031
Neg logit contributions
 videos-0.041
 nephew-0.040
 mechanic-0.039
isions-0.039
controlled-0.039
Token."
Token ID526
Feature activation-1.15e-01

Pos logit contributions
 Fif+0.401
 bamboo+0.352
sed+0.331
onna+0.329
iw+0.328
Neg logit contributions
's-0.331
care-0.324
isions-0.318
 videos-0.310
 means-0.308
Token
Token ID220
Feature activation-6.09e-02

Pos logit contributions
 Fif+0.224
 bamboo+0.207
onna+0.195
io+0.194
iw+0.192
Neg logit contributions
's-0.150
care-0.150
 videos-0.142
isions-0.140
 Margaret-0.136
Token
Token ID198
Feature activation-2.98e-02

Pos logit contributions
 Fif+0.101
 bamboo+0.088
onna+0.085
iw+0.082
 pear+0.080
Neg logit contributions
isions-0.097
 videos-0.096
care-0.093
 mechanic-0.093
 nephew-0.092
Token
Token ID198
Feature activation-3.01e-02

Pos logit contributions
 Fif+0.061
sed+0.056
 bamboo+0.056
iw+0.054
io+0.053
Neg logit contributions
's-0.034
care-0.034
 videos-0.032
isions-0.032
 Margaret-0.031
TokenL
Token ID43
Feature activation+1.41e-02

Pos logit contributions
 Fif+0.059
 bamboo+0.053
onna+0.051
iw+0.050
 pear+0.049
Neg logit contributions
isions-0.038
 videos-0.038
care-0.037
 mechanic-0.037
 nephew-0.036
Tokenily
Token ID813
Feature activation-2.75e-01

Pos logit contributions
isions+0.032
 videos+0.031
care+0.031
's+0.031
 mechanic+0.031
Neg logit contributions
 Fif-0.014
 bamboo-0.011
onna-0.011
iw-0.010
sed-0.010
Token sn
Token ID3013
Feature activation-1.77e-01

Pos logit contributions
 Fif+0.536
 bamboo+0.489
onna+0.462
iw+0.461
sed+0.432
Neg logit contributions
's-0.394
 mechanic-0.345
isions-0.343
 nephew-0.338
 means-0.335
Tokenuck
Token ID1347
Feature activation+5.27e-02

Pos logit contributions
 Fif+0.283
 bamboo+0.250
onna+0.242
sed+0.235
 majesty+0.230
Neg logit contributions
care-0.285
old-0.278
controlled-0.272
's-0.268
 mechanic-0.265
Token up
Token ID510
Feature activation+1.04e-01

Pos logit contributions
's+0.062
 videos+0.061
care+0.060
 Margaret+0.056
 means+0.056
Neg logit contributions
 Fif-0.109
onna-0.099
iw-0.097
 bamboo-0.097
sed-0.095
Token on
Token ID319
Feature activation-4.35e-02

Pos logit contributions
care+0.123
's+0.113
 videos+0.112
 mechanic+0.107
 computer+0.102
Neg logit contributions
 Fif-0.214
 majesty-0.203
 bamboo-0.202
sed-0.189
io-0.186
Token the
Token ID262
Feature activation+1.08e-02

Pos logit contributions
 Fif+0.087
 bamboo+0.080
 majesty+0.077
onna+0.077
iw+0.077
Neg logit contributions
 videos-0.054
care-0.053
 computer-0.051
's-0.051
 mechanic-0.050
Token.
Token ID13
Feature activation+2.78e-03

Pos logit contributions
care+0.130
 videos+0.130
's+0.129
isions+0.123
 computer+0.121
Neg logit contributions
 Fif-0.204
iw-0.183
 bamboo-0.182
onna-0.179
sed-0.173
Token
Token ID220
Feature activation-6.02e-02

Pos logit contributions
's+0.003
care+0.003
+0.003
 Margaret+0.003
 means+0.003
Neg logit contributions
 bamboo-0.006
 Fif-0.005
 crown-0.005
 ti-0.005
iw-0.005
Token
Token ID198
Feature activation-2.98e-02

Pos logit contributions
 Fif+0.105
 bamboo+0.094
onna+0.089
iw+0.088
 pear+0.086
Neg logit contributions
isions-0.089
 videos-0.088
care-0.087
 mechanic-0.085
's-0.084
Token
Token ID198
Feature activation-2.96e-02

Pos logit contributions
 Fif+0.059
 bamboo+0.057
sed+0.054
iw+0.054
 crown+0.053
Neg logit contributions
care-0.035
's-0.034
 Margaret-0.031
isions-0.031
-0.031
TokenM
Token ID44
Feature activation-1.25e-01

Pos logit contributions
 Fif+0.045
 bamboo+0.041
onna+0.038
iw+0.038
sed+0.037
Neg logit contributions
care-0.049
isions-0.049
's-0.049
 videos-0.048
 mechanic-0.047
Tokenia
Token ID544
Feature activation-2.74e-01

Pos logit contributions
 Fif+0.138
 bamboo+0.116
iw+0.098
 ti+0.097
onna+0.096
Neg logit contributions
isions-0.259
's-0.258
care-0.257
urious-0.256
 videos-0.253
Token's
Token ID338
Feature activation+3.80e-01

Pos logit contributions
 Fif+0.559
 bamboo+0.509
onna+0.487
iw+0.484
 pear+0.469
Neg logit contributions
isions-0.322
 videos-0.318
's-0.317
care-0.315
 mechanic-0.309
Token mom
Token ID1995
Feature activation-4.48e-02

Pos logit contributions
 mechanic+0.518
 videos+0.511
 computer+0.510
's+0.505
 jet+0.505
Neg logit contributions
 Fif-0.730
iw-0.646
 bamboo-0.646
onna-0.629
io-0.602
Token saw
Token ID2497
Feature activation+5.76e-02

Pos logit contributions
 Fif+0.081
 bamboo+0.072
onna+0.069
iw+0.068
 pear+0.067
Neg logit contributions
isions-0.064
 videos-0.063
care-0.062
 mechanic-0.061
controlled-0.060
Token that
Token ID326
Feature activation+5.37e-02

Pos logit contributions
 videos+0.066
care+0.065
isions+0.065
's+0.065
 means+0.062
Neg logit contributions
 Fif-0.120
 bamboo-0.108
onna-0.106
sed-0.104
iw-0.104
Token Mia
Token ID32189
Feature activation-2.01e-01

Pos logit contributions
 videos+0.083
care+0.083
isions+0.082
's+0.081
 mechanic+0.079
Neg logit contributions
 Fif-0.089
 bamboo-0.080
onna-0.075
iw-0.075
sed-0.073
Token water
Token ID1660
Feature activation-1.12e-01

Pos logit contributions
's+0.007
 jet+0.007
 videos+0.006
care+0.006
isions+0.006
Neg logit contributions
 Fif-0.007
onna-0.006
iw-0.006
sed-0.006
io-0.006
Token.
Token ID13
Feature activation-1.56e-02

Pos logit contributions
 Fif+0.213
 bamboo+0.187
iw+0.179
onna+0.178
sed+0.177
Neg logit contributions
's-0.151
care-0.149
isions-0.146
 videos-0.145
 mechanic-0.141
Token
Token ID198
Feature activation-3.85e-02

Pos logit contributions
 Fif+0.029
 bamboo+0.027
onna+0.025
iw+0.025
sed+0.025
Neg logit contributions
care-0.021
's-0.021
 videos-0.020
isions-0.020
 mechanic-0.020
Token
Token ID198
Feature activation-3.85e-02

Pos logit contributions
 Fif+0.062
 bamboo+0.054
onna+0.051
iw+0.049
 pear+0.049
Neg logit contributions
isions-0.064
 videos-0.063
 nephew-0.061
 mechanic-0.061
controlled-0.060
TokenL
Token ID43
Feature activation+1.28e-02

Pos logit contributions
 Fif+0.060
 bamboo+0.053
 pear+0.048
onna+0.047
iw+0.046
Neg logit contributions
 nephew-0.065
 videos-0.065
isions-0.065
 means-0.063
 computers-0.062
Tokenily
Token ID813
Feature activation-2.78e-01

Pos logit contributions
isions+0.031
 videos+0.031
care+0.031
 mechanic+0.030
's+0.030
Neg logit contributions
 Fif-0.011
 bamboo-0.008
onna-0.007
iw-0.007
 pear-0.006
Token went
Token ID1816
Feature activation+2.70e-02

Pos logit contributions
 Fif+0.469
 bamboo+0.403
 pear+0.375
sed+0.374
io+0.369
Neg logit contributions
 videos-0.439
isions-0.436
care-0.432
controlled-0.427
 keyboard-0.421
Token to
Token ID284
Feature activation-2.72e-02

Pos logit contributions
's+0.030
care+0.030
 videos+0.029
 mechanic+0.028
 computer+0.027
Neg logit contributions
 Fif-0.053
 majesty-0.051
sed-0.050
onna-0.050
iw-0.050
Token her
Token ID607
Feature activation-5.55e-02

Pos logit contributions
 Fif+0.049
 bamboo+0.043
onna+0.041
iw+0.041
 pear+0.040
Neg logit contributions
isions-0.040
 videos-0.039
care-0.039
 mechanic-0.038
 nephew-0.037
Token mom
Token ID1995
Feature activation-5.05e-02

Pos logit contributions
 Fif+0.082
 bamboo+0.073
 pear+0.067
sed+0.065
onna+0.064
Neg logit contributions
isions-0.098
care-0.097
 videos-0.097
controlled-0.096
ographs-0.093
Token and
Token ID290
Feature activation+1.72e-03

Pos logit contributions
 Fif+0.065
 bamboo+0.059
iw+0.056
 pear+0.055
onna+0.054
Neg logit contributions
isions-0.092
controlled-0.091
 complicated-0.091
ographs-0.089
 keyboard-0.088
Token warned
Token ID7728
Feature activation+1.02e-01

Pos logit contributions
 Fif+0.115
 bamboo+0.102
iw+0.098
onna+0.097
 majesty+0.091
Neg logit contributions
's-0.092
isions-0.084
care-0.082
 mechanic-0.082
 nephew-0.081
Token her
Token ID607
Feature activation-5.33e-02

Pos logit contributions
 videos+0.160
isions+0.158
care+0.157
 mechanic+0.155
's+0.154
Neg logit contributions
 Fif-0.172
 bamboo-0.151
iw-0.144
onna-0.144
sed-0.139
Token.
Token ID13
Feature activation-8.30e-03

Pos logit contributions
 Fif+0.081
 bamboo+0.071
onna+0.066
 pear+0.066
balls+0.062
Neg logit contributions
isions-0.097
ographs-0.093
care-0.092
 videos-0.091
controlled-0.091
Token "
Token ID366
Feature activation-9.58e-02

Pos logit contributions
 Fif+0.015
iw+0.014
 bamboo+0.014
io+0.013
sed+0.013
Neg logit contributions
 mechanic-0.011
 videos-0.011
care-0.011
 Margaret-0.010
 nephew-0.010
TokenL
Token ID43
Feature activation+9.82e-03

Pos logit contributions
 Fif+0.129
iw+0.103
 bamboo+0.101
onna+0.095
 pear+0.090
Neg logit contributions
 videos-0.189
isions-0.185
cases-0.183
 computers-0.181
 computer-0.180
Tokenily
Token ID813
Feature activation-2.77e-01

Pos logit contributions
care+0.022
isions+0.022
 videos+0.021
's+0.021
 mechanic+0.021
Neg logit contributions
 Fif-0.010
 bamboo-0.008
onna-0.007
sed-0.007
iw-0.007
Token,
Token ID11
Feature activation-7.08e-03

Pos logit contributions
 Fif+0.450
 bamboo+0.394
onna+0.373
iw+0.369
 pear+0.366
Neg logit contributions
isions-0.450
 videos-0.447
care-0.438
controlled-0.425
 keyboard-0.424
Token if
Token ID611
Feature activation+1.06e-01

Pos logit contributions
 Fif+0.013
 bamboo+0.012
iw+0.012
io+0.012
onna+0.012
Neg logit contributions
's-0.009
 videos-0.009
 mechanic-0.009
 computer-0.009
 jet-0.009
Token you
Token ID345
Feature activation+2.23e-02

Pos logit contributions
isions+0.140
 videos+0.139
care+0.137
 mechanic+0.134
's+0.133
Neg logit contributions
 Fif-0.200
 bamboo-0.180
onna-0.172
iw-0.170
 pear-0.166
Token take
Token ID1011
Feature activation+6.24e-02

Pos logit contributions
isions+0.030
 videos+0.030
care+0.029
 mechanic+0.028
 nephew+0.028
Neg logit contributions
 Fif-0.042
 bamboo-0.038
onna-0.036
iw-0.036
 pear-0.035
Token a
Token ID257
Feature activation+5.65e-02

Pos logit contributions
 videos+0.077
isions+0.075
care+0.075
's+0.074
 mechanic+0.073
Neg logit contributions
 Fif-0.124
 bamboo-0.111
onna-0.109
iw-0.107
sed-0.104