TOP ACTIVATIONS
MAX = 0.233

be cow boys and cow girls . They put on hats
my cow boys and cow girls . I love you ."
play cow boys and cow girls , and pretend to ride
the blocks , compare the sizes , and sort the colors
shoot ac orns with their tails . The ac orns flew
them how to move their tails in the water and soon
" Let 's hold our tails together . That way ,
jumped , w agging their tails . Then ,
cars of different colors and sizes . They make a line
cubes of different colors and sizes . They build towers and
cars of different colors and sizes . They like to drive
cars of different colors and sizes . They make zoom sounds
cubes of different colors and sizes . They can make towers
they 're different colors and sizes ." Max started
cars of different colors and sizes . They make roads with
it with holes of different sizes . I will make it
cars of different colors and sizes . They liked to zoom
took turns to measure their tails . The mouse 's tail
boats of different colors and sizes . They liked to make
Jack and w agged their tails . Jack was so happy

MIN ACTIVATIONS
MIN = -0.171

His birthday had ended in disaster . <|endoftext|> Jimmy was new
Mom came and saw the disaster . She was not angry
again . But unb ek nown to him , the king
But then disaster struck ! Suddenly , the
done ? This is a disaster . You did not listen
before . But then , disaster struck ! On the day
came running and saw the disaster . She turned off the
laughing . " What a disaster ! You two and Rex
to start his journey when disaster struck . All the things
was delighted . But then disaster struck , he put the
. <|endoftext|> Jimmy was 3 - years - old and he
But then , disaster ! The top suddenly stopped
One day , a 3 - year - old girl named
One day , a three - year - old boy was
day , a little 3 - year - old boy named
day , a little 3 - year - old girl found
A 3 - year - old child noticed
day , a little 3 - year - old girl was
One day , a three - year - old boy was
classroom . Jake was 3 - years - old and the

INTERVAL 0.132 - 0.233
CONTAINS 0.008%

dogs . Do you like dogs ?" Tom asks .
together . Lily played her songs as best as she could
of colors , shapes and sizes . She couldn 't believe
She tells her friends the names of the flowers and what
dolls of different colors and sizes . She was very organized

INTERVAL 0.031 - 0.132
CONTAINS 15.821%

. " You have the blue boat !" "
and the cars and the trees . The hot red balloon
holes and rust . It can hurt you ," Tom says
to make up stories with their dolls and cars . Lily
. They wait for the music to start . They are

INTERVAL -0.070 - 0.031
CONTAINS 81.994%

touch them , feeling the soft pet als . It was
toy grill that he played with every day .
Mom . He learned his lesson . The moral
them . They learned that it 's okay to ask for
. He stirred and stirred until everything was mixed together .

INTERVAL -0.171 - -0.070
CONTAINS 2.177%

try too . He looked for another vine . He saw
they would see Comet Hale - B opp again someday ,
She felt better and went back outside to play with her
Lily and Anna . She took them to the bedroom .
the boy thanked his mom for helping him sort all his
Token be
Token ID307
Feature activation+1.94e-02

Pos logit contributions
 sizes+0.014
bows+0.012
 planets+0.012
 cultures+0.012
 ABC+0.012
Neg logit contributions
 late-0.014
Start-0.012
 aback-0.012
O-0.012
sm-0.012
Token cow
Token ID9875
Feature activation+3.58e-02

Pos logit contributions
 sizes+0.039
 planets+0.036
bows+0.034
 clubs+0.034
 toes+0.033
Neg logit contributions
 late-0.028
 aback-0.025
O-0.025
Start-0.025
sm-0.023
Tokenboys
Token ID13202
Feature activation+9.04e-02

Pos logit contributions
 sizes+0.041
 planets+0.033
 ABC+0.033
 cultures+0.033
hips+0.032
Neg logit contributions
 late-0.079
sm-0.072
 aback-0.069
 severe-0.067
 poisonous-0.067
Token and
Token ID290
Feature activation+5.23e-03

Pos logit contributions
 sizes+0.150
 planets+0.128
 ABC+0.127
 cultures+0.125
 toes+0.123
Neg logit contributions
 late-0.167
 aback-0.142
sm-0.139
O-0.134
Start-0.134
Token cow
Token ID9875
Feature activation+3.68e-02

Pos logit contributions
 sizes+0.008
 planets+0.007
 clubs+0.007
 ABC+0.007
bows+0.007
Neg logit contributions
 late-0.010
 aback-0.009
Start-0.009
sm-0.009
O-0.009
Tokengirls
Token ID36960
Feature activation+2.33e-01

Pos logit contributions
 sizes+0.041
 ABC+0.035
 cultures+0.032
bows+0.032
 planets+0.032
Neg logit contributions
 late-0.082
sm-0.075
 aback-0.071
 heaven-0.069
 another-0.069
Token.
Token ID13
Feature activation+1.22e-02

Pos logit contributions
 sizes+0.455
 cultures+0.389
 ABC+0.386
 planets+0.381
bows+0.379
Neg logit contributions
 late-0.328
 another-0.272
 here-0.270
 aback-0.268
sm-0.265
Token They
Token ID1119
Feature activation+6.15e-02

Pos logit contributions
 sizes+0.019
 toes+0.016
 necks+0.016
 planets+0.015
 clubs+0.015
Neg logit contributions
 late-0.024
 severe-0.021
sm-0.021
 aback-0.021
 poisonous-0.020
Token put
Token ID1234
Feature activation-5.47e-02

Pos logit contributions
 sizes+0.112
 planets+0.098
bows+0.096
 ABC+0.095
 clubs+0.094
Neg logit contributions
 late-0.104
 aback-0.089
Start-0.087
sm-0.086
O-0.086
Token on
Token ID319
Feature activation-7.42e-03

Pos logit contributions
 late+0.094
Start+0.083
 aback+0.082
O+0.079
sm+0.078
Neg logit contributions
 sizes-0.098
 planets-0.087
 clubs-0.085
bows-0.084
 toes-0.082
Token hats
Token ID23910
Feature activation+8.32e-02

Pos logit contributions
 late+0.011
 aback+0.010
sm+0.009
Start+0.009
O+0.009
Neg logit contributions
 sizes-0.014
bows-0.013
 planets-0.013
 ABC-0.012
 cultures-0.012
Token my
Token ID616
Feature activation-6.20e-03

Pos logit contributions
 sizes+0.131
 planets+0.117
bows+0.116
 ABC+0.115
 clubs+0.113
Neg logit contributions
 late-0.079
 aback-0.065
Start-0.062
sm-0.062
O-0.062
Token cow
Token ID9875
Feature activation+3.49e-02

Pos logit contributions
 late+0.010
 aback+0.009
Start+0.009
sm+0.009
O+0.009
Neg logit contributions
 sizes-0.011
 planets-0.010
bows-0.010
 ABC-0.010
 clubs-0.010
Tokenboys
Token ID13202
Feature activation+9.28e-02

Pos logit contributions
 sizes+0.056
bows+0.049
 planets+0.048
 ABC+0.047
 toes+0.045
Neg logit contributions
 late-0.067
 aback-0.059
Start-0.059
O-0.058
izz-0.056
Token and
Token ID290
Feature activation+3.29e-03

Pos logit contributions
 sizes+0.136
bows+0.119
 planets+0.115
 ABC+0.112
 clubs+0.109
Neg logit contributions
 late-0.188
Start-0.165
 aback-0.164
O-0.162
sm-0.158
Token cow
Token ID9875
Feature activation+3.26e-02

Pos logit contributions
 sizes+0.006
 planets+0.005
 clubs+0.005
 toes+0.005
 ABC+0.005
Neg logit contributions
 late-0.006
 aback-0.005
Start-0.005
izz-0.005
sm-0.005
Tokengirls
Token ID36960
Feature activation+2.30e-01

Pos logit contributions
 sizes+0.053
bows+0.045
 planets+0.045
 ABC+0.044
 clubs+0.043
Neg logit contributions
 late-0.062
 aback-0.054
Start-0.054
O-0.053
sm-0.051
Token.
Token ID13
Feature activation+2.66e-02

Pos logit contributions
 sizes+0.433
 planets+0.381
bows+0.374
 ABC+0.372
 clubs+0.367
Neg logit contributions
 late-0.369
 aback-0.313
Start-0.304
sm-0.304
O-0.302
Token I
Token ID314
Feature activation-6.05e-02

Pos logit contributions
 sizes+0.045
 planets+0.039
bows+0.038
 ABC+0.037
 cultures+0.037
Neg logit contributions
 late-0.047
 aback-0.042
Start-0.041
sm-0.041
O-0.040
Token love
Token ID1842
Feature activation+3.20e-02

Pos logit contributions
 late+0.124
 aback+0.111
O+0.109
Start+0.109
izz+0.107
Neg logit contributions
 sizes-0.078
bows-0.071
 ABC-0.063
 neck-0.059
 pretending-0.059
Token you
Token ID345
Feature activation+1.70e-03

Pos logit contributions
 sizes+0.066
 planets+0.060
 ABC+0.058
 cultures+0.057
 toes+0.057
Neg logit contributions
 late-0.044
 aback-0.041
Start-0.040
O-0.037
izz-0.037
Token."
Token ID526
Feature activation-2.21e-02

Pos logit contributions
 sizes+0.003
 planets+0.003
 ABC+0.002
bows+0.002
 toes+0.002
Neg logit contributions
 late-0.003
 aback-0.003
Start-0.003
O-0.003
sm-0.003
Token play
Token ID711
Feature activation+5.29e-02

Pos logit contributions
 sizes+0.014
 planets+0.012
bows+0.012
 ABC+0.011
 cultures+0.011
Neg logit contributions
 late-0.018
 aback-0.015
O-0.015
Start-0.015
sm-0.015
Token cow
Token ID9875
Feature activation+3.80e-02

Pos logit contributions
 sizes+0.066
 planets+0.063
bows+0.060
 ABC+0.060
 clubs+0.060
Neg logit contributions
 late-0.108
izz-0.103
 severe-0.101
 aback-0.100
O-0.098
Tokenboys
Token ID13202
Feature activation+9.31e-02

Pos logit contributions
 sizes+0.052
 planets+0.043
 ABC+0.042
 cultures+0.041
bows+0.041
Neg logit contributions
 late-0.079
 aback-0.070
sm-0.069
Start-0.067
O-0.066
Token and
Token ID290
Feature activation+2.83e-03

Pos logit contributions
 sizes+0.147
 planets+0.125
 ABC+0.122
 cultures+0.121
bows+0.120
Neg logit contributions
 late-0.177
 aback-0.156
sm-0.150
Start-0.148
O-0.148
Token cow
Token ID9875
Feature activation+3.87e-02

Pos logit contributions
 sizes+0.004
 planets+0.004
bows+0.004
 ABC+0.004
 clubs+0.004
Neg logit contributions
 late-0.006
 aback-0.005
Start-0.005
sm-0.005
O-0.005
Tokengirls
Token ID36960
Feature activation+2.30e-01

Pos logit contributions
 sizes+0.048
 ABC+0.040
 planets+0.040
 cultures+0.039
bows+0.039
Neg logit contributions
 late-0.081
sm-0.072
 aback-0.071
 poisoned-0.068
 severe-0.067
Token,
Token ID11
Feature activation+4.39e-02

Pos logit contributions
 sizes+0.406
 cultures+0.348
 planets+0.343
 ABC+0.339
azes+0.336
Neg logit contributions
 late-0.355
 aback-0.306
 another-0.303
 here-0.301
sm-0.287
Token and
Token ID290
Feature activation-2.90e-03

Pos logit contributions
 sizes+0.072
 planets+0.062
bows+0.062
 ABC+0.061
 cultures+0.060
Neg logit contributions
 late-0.081
 aback-0.069
sm-0.068
Start-0.067
O-0.067
Token pretend
Token ID16614
Feature activation+6.99e-02

Pos logit contributions
 late+0.005
 aback+0.005
Start+0.005
O+0.004
sm+0.004
Neg logit contributions
 sizes-0.005
 planets-0.004
bows-0.004
 ABC-0.004
 clubs-0.004
Token to
Token ID284
Feature activation+1.48e-02

Pos logit contributions
 sizes+0.118
 planets+0.102
bows+0.100
 ABC+0.099
 clubs+0.098
Neg logit contributions
 late-0.125
 aback-0.116
Start-0.111
O-0.111
izz-0.107
Token ride
Token ID6594
Feature activation+1.65e-02

Pos logit contributions
 sizes+0.021
 planets+0.018
bows+0.017
 ABC+0.017
 clubs+0.017
Neg logit contributions
 late-0.031
 aback-0.028
sm-0.027
Start-0.027
izz-0.027
Token the
Token ID262
Feature activation+3.77e-02

Pos logit contributions
 sizes+0.151
 planets+0.136
 ABC+0.131
bows+0.131
 clubs+0.128
Neg logit contributions
 late-0.119
 aback-0.106
Start-0.102
O-0.102
sm-0.099
Token blocks
Token ID7021
Feature activation+3.93e-02

Pos logit contributions
 sizes+0.047
bows+0.041
 cultures+0.040
 clubs+0.038
 necks+0.038
Neg logit contributions
 late-0.080
 aback-0.067
sm-0.067
Start-0.066
O-0.064
Token,
Token ID11
Feature activation+5.48e-02

Pos logit contributions
 sizes+0.062
bows+0.053
 planets+0.052
 ABC+0.050
 toes+0.050
Neg logit contributions
 late-0.074
O-0.070
 aback-0.070
Start-0.070
 severe-0.067
Token compare
Token ID8996
Feature activation+4.10e-02

Pos logit contributions
 sizes+0.095
 planets+0.083
bows+0.082
 ABC+0.081
 clubs+0.079
Neg logit contributions
 late-0.096
 aback-0.083
Start-0.081
O-0.081
sm-0.080
Token the
Token ID262
Feature activation+4.30e-02

Pos logit contributions
 sizes+0.077
 planets+0.068
bows+0.066
 ABC+0.066
 clubs+0.065
Neg logit contributions
 late-0.067
 aback-0.058
Start-0.056
O-0.056
sm-0.056
Token sizes
Token ID10620
Feature activation+1.68e-01

Pos logit contributions
 sizes+0.054
bows+0.048
 cultures+0.047
 planets+0.046
 clubs+0.046
Neg logit contributions
 late-0.089
 aback-0.078
Start-0.075
sm-0.074
O-0.072
Token,
Token ID11
Feature activation+6.17e-02

Pos logit contributions
 sizes+0.320
 planets+0.278
bows+0.277
 ABC+0.269
 clubs+0.269
Neg logit contributions
 late-0.271
 aback-0.236
Start-0.230
O-0.229
sm-0.222
Token and
Token ID290
Feature activation-4.82e-04

Pos logit contributions
 sizes+0.093
bows+0.082
 ABC+0.081
 planets+0.080
 cultures+0.080
Neg logit contributions
 late-0.116
 aback-0.097
Start-0.094
O-0.094
sm-0.093
Token sort
Token ID3297
Feature activation+1.73e-02

Pos logit contributions
 late+0.001
 aback+0.001
Start+0.001
sm+0.001
O+0.001
Neg logit contributions
 sizes-0.001
bows-0.001
 ABC-0.001
 planets-0.001
 clubs-0.001
Token the
Token ID262
Feature activation+3.60e-02

Pos logit contributions
 sizes+0.032
 planets+0.028
bows+0.028
 ABC+0.027
 clubs+0.027
Neg logit contributions
 late-0.029
O-0.025
 aback-0.025
Start-0.025
 severe-0.025
Token colors
Token ID7577
Feature activation+1.17e-01

Pos logit contributions
 sizes+0.043
bows+0.038
 cultures+0.037
inas+0.036
 clubs+0.036
Neg logit contributions
 late-0.076
 aback-0.066
Start-0.064
sm-0.064
 my-0.062
Token shoot
Token ID2686
Feature activation+4.02e-02

Pos logit contributions
 sizes+0.116
 planets+0.099
bows+0.099
 ABC+0.098
 cultures+0.096
Neg logit contributions
 late-0.136
 aback-0.118
Start-0.116
O-0.115
sm-0.114
Token ac
Token ID936
Feature activation-5.64e-03

Pos logit contributions
 sizes+0.075
 planets+0.067
 clubs+0.065
 languages+0.063
 nations+0.063
Neg logit contributions
 late-0.066
izz-0.058
 severe-0.056
 aback-0.056
O-0.055
Tokenorns
Token ID19942
Feature activation+2.39e-02

Pos logit contributions
 late+0.010
 aback+0.008
Start+0.008
sm+0.008
 severe+0.008
Neg logit contributions
 sizes-0.010
bows-0.009
 clubs-0.009
 planets-0.009
 ABC-0.009
Token with
Token ID351
Feature activation-6.95e-03

Pos logit contributions
 sizes+0.041
 planets+0.036
bows+0.035
 ABC+0.035
 cultures+0.035
Neg logit contributions
 late-0.042
 aback-0.037
sm-0.036
Start-0.035
O-0.035
Token their
Token ID511
Feature activation+5.23e-02

Pos logit contributions
 late+0.009
 aback+0.008
Start+0.007
O+0.007
sm+0.007
Neg logit contributions
 sizes-0.015
 planets-0.014
 clubs-0.013
bows-0.013
 ABC-0.013
Token tails
Token ID30514
Feature activation+1.67e-01

Pos logit contributions
 sizes+0.080
bows+0.072
 planets+0.072
 ABC+0.070
 cultures+0.070
Neg logit contributions
 late-0.092
sm-0.083
 aback-0.083
O-0.082
 poisonous-0.082
Token.
Token ID13
Feature activation+2.03e-02

Pos logit contributions
 sizes+0.312
 ABC+0.302
 cultures+0.301
bows+0.290
 planets+0.285
Neg logit contributions
 late-0.241
 aback-0.200
sm-0.190
 another-0.187
 poisoned-0.183
Token The
Token ID383
Feature activation+2.98e-02

Pos logit contributions
 sizes+0.039
 planets+0.034
bows+0.034
 ABC+0.034
 clubs+0.033
Neg logit contributions
 late-0.032
 aback-0.028
Start-0.027
sm-0.027
O-0.027
Token ac
Token ID936
Feature activation-1.90e-02

Pos logit contributions
 sizes+0.057
 planets+0.049
bows+0.049
 ABC+0.048
 cultures+0.048
Neg logit contributions
 late-0.049
O-0.043
Start-0.041
 aback-0.041
sm-0.040
Tokenorns
Token ID19942
Feature activation+1.32e-03

Pos logit contributions
 late+0.032
 aback+0.027
Start+0.026
sm+0.025
 severe+0.025
Neg logit contributions
 sizes-0.035
bows-0.031
 planets-0.030
 clubs-0.030
 ABC-0.029
Token flew
Token ID13112
Feature activation-5.52e-03

Pos logit contributions
 sizes+0.002
 planets+0.002
 ABC+0.002
bows+0.002
 cultures+0.002
Neg logit contributions
 late-0.002
 aback-0.002
sm-0.002
Start-0.002
O-0.002
Token them
Token ID606
Feature activation+9.22e-03

Pos logit contributions
 sizes+0.016
 clubs+0.015
 planets+0.014
 patterns+0.014
 languages+0.014
Neg logit contributions
 late-0.017
 severe-0.015
Start-0.015
sm-0.014
O-0.014
Token how
Token ID703
Feature activation-5.55e-03

Pos logit contributions
 sizes+0.021
 planets+0.019
bows+0.018
 ABC+0.018
 clubs+0.018
Neg logit contributions
 late-0.012
 aback-0.010
Start-0.009
O-0.009
sm-0.009
Token to
Token ID284
Feature activation+7.79e-03

Pos logit contributions
 late+0.009
 aback+0.008
O+0.008
Start+0.008
izz+0.008
Neg logit contributions
 sizes-0.010
 planets-0.009
 clubs-0.009
 cultures-0.009
 toes-0.008
Token move
Token ID1445
Feature activation+1.54e-02

Pos logit contributions
 sizes+0.012
 planets+0.011
bows+0.010
 ABC+0.010
 clubs+0.010
Neg logit contributions
 late-0.015
 aback-0.013
sm-0.013
Start-0.013
O-0.013
Token their
Token ID511
Feature activation+5.19e-02

Pos logit contributions
 sizes+0.032
 planets+0.028
bows+0.028
 clubs+0.028
 ABC+0.027
Neg logit contributions
 late-0.022
 aback-0.018
Start-0.018
O-0.018
sm-0.018
Token tails
Token ID30514
Feature activation+1.67e-01

Pos logit contributions
 sizes+0.075
bows+0.070
 planets+0.061
 ABC+0.061
inas+0.061
Neg logit contributions
 late-0.098
 aback-0.087
 poisoned-0.085
Start-0.085
sm-0.084
Token in
Token ID287
Feature activation+2.55e-02

Pos logit contributions
 sizes+0.304
 ABC+0.271
bows+0.270
 cultures+0.265
 planets+0.265
Neg logit contributions
 late-0.278
 aback-0.238
Start-0.223
sm-0.221
O-0.215
Token the
Token ID262
Feature activation+2.71e-02

Pos logit contributions
 sizes+0.060
 planets+0.054
 clubs+0.053
 ABC+0.053
bows+0.053
Neg logit contributions
 late-0.029
 aback-0.023
sm-0.023
Start-0.022
O-0.022
Token water
Token ID1660
Feature activation+5.37e-02

Pos logit contributions
 sizes+0.049
 planets+0.043
bows+0.042
 ABC+0.041
 clubs+0.041
Neg logit contributions
 late-0.046
 aback-0.040
sm-0.039
Start-0.039
O-0.039
Token and
Token ID290
Feature activation+1.17e-03

Pos logit contributions
 sizes+0.083
 planets+0.073
 ABC+0.073
bows+0.071
 cultures+0.071
Neg logit contributions
 late-0.104
 aback-0.091
sm-0.086
Start-0.086
O-0.084
Token soon
Token ID2582
Feature activation-1.68e-02

Pos logit contributions
 sizes+0.002
 ABC+0.002
bows+0.002
 planets+0.002
 cultures+0.002
Neg logit contributions
 late-0.002
 aback-0.002
Start-0.002
sm-0.002
O-0.002
Token "
Token ID366
Feature activation-3.40e-03

Pos logit contributions
 sizes+0.041
 planets+0.034
bows+0.033
 ABC+0.033
 clubs+0.032
Neg logit contributions
 late-0.083
 aback-0.074
Start-0.073
O-0.073
sm-0.073
TokenLet
Token ID5756
Feature activation-3.00e-03

Pos logit contributions
 late+0.005
 poisonous+0.004
 severe+0.004
 aback+0.004
 poisoned+0.004
Neg logit contributions
 sizes-0.007
bows-0.006
 planets-0.006
 cultures-0.006
 ABC-0.006
Token's
Token ID338
Feature activation+6.47e-02

Pos logit contributions
 late+0.007
sm+0.006
izz+0.006
 poisonous+0.006
plug+0.006
Neg logit contributions
 sizes-0.003
 ABC-0.003
 clubs-0.003
 languages-0.003
 patterns-0.003
Token hold
Token ID1745
Feature activation+3.78e-03

Pos logit contributions
 sizes+0.109
 planets+0.094
bows+0.093
 ABC+0.092
 clubs+0.090
Neg logit contributions
 late-0.118
 aback-0.102
Start-0.100
sm-0.099
O-0.099
Token our
Token ID674
Feature activation+5.24e-02

Pos logit contributions
 sizes+0.007
 clubs+0.006
 toes+0.006
 planets+0.006
 patterns+0.006
Neg logit contributions
 late-0.006
 severe-0.005
Start-0.005
O-0.005
 poisonous-0.005
Token tails
Token ID30514
Feature activation+1.66e-01

Pos logit contributions
 sizes+0.081
bows+0.073
 planets+0.068
 ABC+0.067
 cultures+0.066
Neg logit contributions
 late-0.097
 aback-0.086
sm-0.084
Start-0.082
 severe-0.082
Token together
Token ID1978
Feature activation+5.44e-02

Pos logit contributions
 sizes+0.296
 planets+0.259
bows+0.255
 ABC+0.252
 clubs+0.249
Neg logit contributions
 late-0.284
 aback-0.244
Start-0.240
O-0.238
sm-0.237
Token.
Token ID13
Feature activation+3.98e-02

Pos logit contributions
 sizes+0.095
 planets+0.082
bows+0.081
 ABC+0.080
 clubs+0.079
Neg logit contributions
 late-0.095
 aback-0.082
O-0.081
Start-0.081
sm-0.080
Token That
Token ID1320
Feature activation+5.22e-03

Pos logit contributions
 sizes+0.074
 planets+0.065
bows+0.064
 ABC+0.064
 clubs+0.062
Neg logit contributions
 late-0.065
 aback-0.055
sm-0.054
Start-0.054
O-0.054
Token way
Token ID835
Feature activation+2.88e-02

Pos logit contributions
 sizes+0.007
 planets+0.006
bows+0.006
 ABC+0.006
 clubs+0.006
Neg logit contributions
 late-0.011
 aback-0.010
Start-0.010
O-0.010
sm-0.010
Token,
Token ID11
Feature activation+4.91e-02

Pos logit contributions
 sizes+0.050
 planets+0.043
bows+0.042
 ABC+0.042
 clubs+0.042
Neg logit contributions
 late-0.051
 aback-0.044
O-0.044
Start-0.044
sm-0.044
Token jumped
Token ID11687
Feature activation+1.72e-02

Pos logit contributions
 late+0.006
 aback+0.006
O+0.005
Start+0.005
sm+0.005
Neg logit contributions
 sizes-0.007
 ABC-0.006
bows-0.006
 cultures-0.006
 planets-0.006
Token,
Token ID11
Feature activation+5.89e-02

Pos logit contributions
 sizes+0.031
 cultures+0.028
 ABC+0.028
 planets+0.027
 hobbies+0.027
Neg logit contributions
 late-0.031
 aback-0.027
Start-0.025
O-0.025
sm-0.024
Token w
Token ID266
Feature activation+3.27e-02

Pos logit contributions
 sizes+0.099
 cultures+0.093
 planets+0.091
 ABC+0.091
 languages+0.089
Neg logit contributions
 late-0.105
 aback-0.091
Start-0.087
sm-0.085
O-0.083
Tokenagging
Token ID16406
Feature activation+1.33e-01

Pos logit contributions
 sizes+0.040
 planets+0.033
bows+0.033
 ABC+0.032
 cultures+0.031
Neg logit contributions
 late-0.074
 aback-0.067
Start-0.066
O-0.065
sm-0.065
Token their
Token ID511
Feature activation+5.71e-02

Pos logit contributions
 sizes+0.242
 ABC+0.233
 cultures+0.230
bows+0.223
 planets+0.216
Neg logit contributions
 late-0.198
sm-0.165
 aback-0.164
Start-0.155
O-0.154
Token tails
Token ID30514
Feature activation+1.64e-01

Pos logit contributions
 sizes+0.060
bows+0.057
hips+0.056
 ABC+0.056
 cultures+0.056
Neg logit contributions
 late-0.120
sm-0.112
 poisonous-0.108
O-0.105
 severe-0.104
Token.
Token ID13
Feature activation+2.29e-02

Pos logit contributions
 cultures+0.314
 ABC+0.311
 sizes+0.298
 hobbies+0.291
azes+0.286
Neg logit contributions
 late-0.231
 aback-0.194
 another-0.191
sm-0.182
 away-0.181
Token
Token ID198
Feature activation-1.42e-02

Pos logit contributions
 sizes+0.042
 planets+0.037
bows+0.037
 ABC+0.036
 clubs+0.036
Neg logit contributions
 late-0.038
 aback-0.033
sm-0.032
Start-0.032
O-0.032
Token
Token ID198
Feature activation-1.87e-02

Pos logit contributions
 late+0.024
 aback+0.022
sm+0.020
 poisoned+0.020
Start+0.020
Neg logit contributions
 sizes-0.024
 ABC-0.021
 planets-0.021
 clubs-0.020
bows-0.020
TokenThen
Token ID6423
Feature activation+4.45e-02

Pos logit contributions
 aback+0.033
 late+0.033
sm+0.029
 poisonous+0.029
 poisoned+0.029
Neg logit contributions
 sizes-0.029
 planets-0.025
 clubs-0.025
 necks-0.025
 ABC-0.024
Token,
Token ID11
Feature activation+1.69e-02

Pos logit contributions
 sizes+0.077
bows+0.067
 cultures+0.065
 ABC+0.065
 planets+0.062
Neg logit contributions
 late-0.081
O-0.069
 aback-0.068
Start-0.067
sm-0.065
Token cars
Token ID5006
Feature activation+4.73e-02

Pos logit contributions
 late+0.078
sm+0.069
 aback+0.068
 my+0.067
O+0.066
Neg logit contributions
bows-0.047
 sizes-0.044
hips-0.042
 cultures-0.042
azes-0.040
Token of
Token ID286
Feature activation-1.47e-02

Pos logit contributions
 sizes+0.071
 planets+0.061
 cultures+0.060
 ABC+0.059
 toes+0.059
Neg logit contributions
 late-0.095
 aback-0.083
sm-0.081
Start-0.079
O-0.079
Token different
Token ID1180
Feature activation+1.79e-02

Pos logit contributions
 late+0.023
 aback+0.019
O+0.019
Start+0.019
sm+0.018
Neg logit contributions
 sizes-0.028
bows-0.025
 planets-0.025
 ABC-0.024
 cultures-0.024
Token colors
Token ID7577
Feature activation+1.15e-01

Pos logit contributions
hips+0.018
azes+0.017
 mast+0.017
sing+0.017
igo+0.017
Neg logit contributions
 late-0.036
 aback-0.033
 my-0.031
sm-0.031
 another-0.030
Token and
Token ID290
Feature activation+5.96e-03

Pos logit contributions
 sizes+0.193
 cultures+0.184
 ABC+0.179
 planets+0.176
 toes+0.174
Neg logit contributions
 late-0.191
sm-0.151
 aback-0.151
 my-0.141
Start-0.140
Token sizes
Token ID10620
Feature activation+1.63e-01

Pos logit contributions
pit+0.004
utes+0.004
azes+0.004
castle+0.003
bles+0.003
Neg logit contributions
 late-0.014
 my-0.013
 another-0.013
sm-0.012
 aback-0.012
Token.
Token ID13
Feature activation+3.16e-02

Pos logit contributions
 sizes+0.279
 ABC+0.269
 cultures+0.265
 planets+0.263
 cheeks+0.254
Neg logit contributions
 late-0.258
sm-0.218
 aback-0.209
Start-0.196
O-0.190
Token They
Token ID1119
Feature activation+6.34e-02

Pos logit contributions
 sizes+0.062
bows+0.055
 planets+0.055
 ABC+0.054
 cultures+0.054
Neg logit contributions
 late-0.050
O-0.040
 aback-0.040
Start-0.039
sm-0.039
Token make
Token ID787
Feature activation+2.87e-02

Pos logit contributions
 sizes+0.116
 ABC+0.108
bows+0.105
 cultures+0.104
azes+0.104
Neg logit contributions
 late-0.090
 aback-0.075
Start-0.073
O-0.072
sm-0.071
Token a
Token ID257
Feature activation+3.15e-02

Pos logit contributions
 sizes+0.048
 cultures+0.042
bows+0.042
 toes+0.040
 cheeks+0.040
Neg logit contributions
 late-0.051
Start-0.042
 another-0.041
 aback-0.040
sm-0.040
Token line
Token ID1627
Feature activation-4.86e-02

Pos logit contributions
 sizes+0.065
bows+0.063
 cultures+0.059
aces+0.058
azes+0.057
Neg logit contributions
 late-0.043
 my-0.033
 poisonous-0.031
Start-0.031
O-0.031
Token cubes
Token ID34896
Feature activation+1.38e-02

Pos logit contributions
 late+0.087
sm+0.075
 aback+0.075
Start+0.071
O+0.070
Neg logit contributions
bows-0.055
 sizes-0.054
 cultures-0.048
hips-0.048
 ABC-0.048
Token of
Token ID286
Feature activation-3.21e-02

Pos logit contributions
 sizes+0.026
 planets+0.023
bows+0.023
 ABC+0.022
 clubs+0.022
Neg logit contributions
 late-0.022
 aback-0.019
Start-0.019
O-0.019
sm-0.018
Token different
Token ID1180
Feature activation+6.89e-03

Pos logit contributions
 late+0.044
 aback+0.036
O+0.035
sm+0.035
Start+0.035
Neg logit contributions
 sizes-0.067
bows-0.061
 planets-0.061
 ABC-0.060
 cultures-0.059
Token colors
Token ID7577
Feature activation+1.13e-01

Pos logit contributions
bows+0.008
hips+0.007
 sizes+0.007
 ABC+0.007
tails+0.007
Neg logit contributions
 late-0.015
 aback-0.013
sm-0.012
 another-0.012
 my-0.012
Token and
Token ID290
Feature activation+5.39e-03

Pos logit contributions
 sizes+0.190
 cultures+0.178
 ABC+0.175
 planets+0.174
 necks+0.169
Neg logit contributions
 late-0.195
sm-0.158
 aback-0.156
O-0.147
izz-0.146
Token sizes
Token ID10620
Feature activation+1.63e-01

Pos logit contributions
utes+0.004
azes+0.004
 ABC+0.004
hips+0.004
bows+0.004
Neg logit contributions
 late-0.013
 my-0.011
sm-0.011
 another-0.011
 aback-0.011
Token.
Token ID13
Feature activation+3.08e-02

Pos logit contributions
 sizes+0.274
 ABC+0.261
 planets+0.257
 cultures+0.255
 necks+0.245
Neg logit contributions
 late-0.267
sm-0.226
 aback-0.218
Start-0.206
O-0.202
Token They
Token ID1119
Feature activation+6.33e-02

Pos logit contributions
 sizes+0.058
 planets+0.052
bows+0.051
 ABC+0.050
 cultures+0.050
Neg logit contributions
 late-0.050
 aback-0.042
Start-0.041
O-0.041
sm-0.040
Token build
Token ID1382
Feature activation+3.49e-02

Pos logit contributions
 sizes+0.112
 ABC+0.102
bows+0.101
 planets+0.099
 cultures+0.098
Neg logit contributions
 late-0.099
 aback-0.083
Start-0.081
sm-0.081
O-0.080
Token towers
Token ID18028
Feature activation+8.35e-02

Pos logit contributions
 sizes+0.068
inas+0.065
bows+0.065
 cultures+0.064
 cheeks+0.063
Neg logit contributions
 late-0.051
 another-0.043
 aback-0.036
 my-0.036
Start-0.035
Token and
Token ID290
Feature activation-5.53e-03

Pos logit contributions
 sizes+0.167
 cultures+0.160
 hobbies+0.159
 cheeks+0.157
 habitats+0.154
Neg logit contributions
 late-0.123
 another-0.097
sm-0.083
 here-0.080
 my-0.080
Token cars
Token ID5006
Feature activation+4.58e-02

Pos logit contributions
 late+0.080
sm+0.071
 aback+0.070
Start+0.068
 my+0.067
Neg logit contributions
bows-0.048
 sizes-0.048
hips-0.042
 cultures-0.042
 ABC-0.041
Token of
Token ID286
Feature activation-1.58e-02

Pos logit contributions
 sizes+0.072
 planets+0.063
 ABC+0.061
 cultures+0.061
 toes+0.060
Neg logit contributions
 late-0.088
 aback-0.076
sm-0.075
Start-0.073
O-0.072
Token different
Token ID1180
Feature activation+1.81e-02

Pos logit contributions
 late+0.025
 aback+0.020
Start+0.020
O+0.020
sm+0.020
Neg logit contributions
 sizes-0.030
bows-0.027
 planets-0.026
 ABC-0.026
 cultures-0.026
Token colors
Token ID7577
Feature activation+1.13e-01

Pos logit contributions
hips+0.018
 mast+0.017
sing+0.017
igo+0.017
azes+0.017
Neg logit contributions
 late-0.037
 aback-0.034
sm-0.032
 my-0.032
 another-0.030
Token and
Token ID290
Feature activation+6.48e-03

Pos logit contributions
 sizes+0.187
 cultures+0.176
 ABC+0.173
 planets+0.170
 necks+0.169
Neg logit contributions
 late-0.190
sm-0.152
 aback-0.151
 my-0.143
Start-0.140
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
pit+0.004
utes+0.004
azes+0.004
ahs+0.004
bles+0.004
Neg logit contributions
 late-0.016
 my-0.015
 another-0.014
sm-0.014
 aback-0.014
Token.
Token ID13
Feature activation+2.89e-02

Pos logit contributions
 sizes+0.276
 ABC+0.263
 planets+0.258
 cultures+0.257
 necks+0.249
Neg logit contributions
 late-0.260
sm-0.221
 aback-0.210
Start-0.198
O-0.191
Token They
Token ID1119
Feature activation+6.24e-02

Pos logit contributions
 sizes+0.055
bows+0.049
 planets+0.049
 ABC+0.048
 clubs+0.048
Neg logit contributions
 late-0.047
O-0.038
 aback-0.038
Start-0.037
sm-0.037
Token like
Token ID588
Feature activation+5.61e-02

Pos logit contributions
 sizes+0.112
 ABC+0.102
 toes+0.099
bows+0.099
 cultures+0.098
Neg logit contributions
 late-0.095
 aback-0.079
Start-0.077
sm-0.076
O-0.076
Token to
Token ID284
Feature activation+1.27e-02

Pos logit contributions
 sizes+0.125
bows+0.117
 cheeks+0.114
 necks+0.114
 planets+0.114
Neg logit contributions
 late-0.067
sm-0.047
 my-0.044
 another-0.043
Start-0.042
Token drive
Token ID3708
Feature activation+2.31e-04

Pos logit contributions
 sizes+0.021
 planets+0.019
bows+0.018
 ABC+0.018
 cultures+0.018
Neg logit contributions
 late-0.022
O-0.019
Start-0.019
 aback-0.019
sm-0.018
Token cars
Token ID5006
Feature activation+4.88e-02

Pos logit contributions
 late+0.074
sm+0.067
 aback+0.063
Start+0.062
 my+0.062
Neg logit contributions
bows-0.042
 sizes-0.040
hips-0.037
 cultures-0.037
inas-0.036
Token of
Token ID286
Feature activation-1.69e-02

Pos logit contributions
 sizes+0.082
 cultures+0.075
 planets+0.075
 toes+0.073
 ABC+0.073
Neg logit contributions
 late-0.086
sm-0.072
 aback-0.070
Start-0.065
O-0.064
Token different
Token ID1180
Feature activation+1.92e-02

Pos logit contributions
 late+0.023
 aback+0.018
Start+0.018
O+0.017
sm+0.017
Neg logit contributions
 sizes-0.035
bows-0.032
 planets-0.032
 cheeks-0.031
 cultures-0.031
Token colors
Token ID7577
Feature activation+1.20e-01

Pos logit contributions
 mast+0.019
hips+0.019
sing+0.018
 cheeks+0.018
 disco+0.018
Neg logit contributions
 late-0.039
 aback-0.035
sm-0.034
 my-0.033
 another-0.032
Token and
Token ID290
Feature activation+1.04e-02

Pos logit contributions
 cultures+0.213
 sizes+0.211
 habitats+0.205
 cheeks+0.203
 necks+0.203
Neg logit contributions
 late-0.173
sm-0.135
 another-0.126
 aback-0.126
 my-0.123
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
pit+0.009
castle+0.008
utes+0.007
azes+0.007
iked+0.007
Neg logit contributions
 late-0.023
 my-0.022
 the-0.021
 another-0.021
Start-0.020
Token.
Token ID13
Feature activation+2.17e-02

Pos logit contributions
 sizes+0.289
 ABC+0.281
 planets+0.278
 cultures+0.277
 habitats+0.269
Neg logit contributions
 late-0.234
sm-0.201
 aback-0.184
Start-0.175
 another-0.172
Token They
Token ID1119
Feature activation+6.74e-02

Pos logit contributions
 sizes+0.044
bows+0.041
 planets+0.040
 ABC+0.039
 cultures+0.039
Neg logit contributions
 late-0.032
O-0.026
 aback-0.025
Start-0.024
sm-0.024
Token make
Token ID787
Feature activation+2.88e-02

Pos logit contributions
 sizes+0.132
 ABC+0.124
bows+0.122
azes+0.121
 cultures+0.121
Neg logit contributions
 late-0.086
 aback-0.070
O-0.068
Start-0.067
sm-0.066
Token zoom
Token ID19792
Feature activation+2.52e-02

Pos logit contributions
 sizes+0.046
 cultures+0.042
bows+0.042
 cheeks+0.040
 toes+0.040
Neg logit contributions
 late-0.051
Start-0.042
 another-0.041
 aback-0.041
O-0.040
Token sounds
Token ID5238
Feature activation+8.80e-02

Pos logit contributions
 sizes+0.040
 cultures+0.036
 toes+0.035
 planets+0.035
 necks+0.035
Neg logit contributions
 late-0.045
 aback-0.039
Start-0.038
sm-0.037
O-0.037
Token cubes
Token ID34896
Feature activation+1.45e-02

Pos logit contributions
 late+0.078
sm+0.067
 aback+0.067
O+0.065
Start+0.065
Neg logit contributions
bows-0.047
 sizes-0.046
 cultures-0.041
hips-0.041
inas-0.041
Token of
Token ID286
Feature activation-3.04e-02

Pos logit contributions
 sizes+0.028
 planets+0.026
 ABC+0.025
 cultures+0.025
bows+0.025
Neg logit contributions
 late-0.023
 aback-0.018
sm-0.018
Start-0.017
O-0.017
Token different
Token ID1180
Feature activation+1.06e-02

Pos logit contributions
 late+0.044
 aback+0.036
Start+0.035
O+0.035
sm+0.035
Neg logit contributions
 sizes-0.061
 planets-0.055
bows-0.055
 ABC-0.054
 cultures-0.054
Token colors
Token ID7577
Feature activation+1.18e-01

Pos logit contributions
bows+0.011
hips+0.011
tails+0.011
inas+0.011
 cheeks+0.011
Neg logit contributions
 late-0.023
 aback-0.020
sm-0.019
 another-0.018
Start-0.018
Token and
Token ID290
Feature activation+6.92e-03

Pos logit contributions
 cultures+0.214
 sizes+0.211
 habitats+0.209
 ABC+0.204
 planets+0.203
Neg logit contributions
 late-0.174
sm-0.133
 another-0.126
izz-0.123
 aback-0.121
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
utes+0.006
pit+0.006
azes+0.006
ahs+0.005
 habitats+0.005
Neg logit contributions
 late-0.015
 my-0.014
 another-0.014
 the-0.013
 some-0.013
Token.
Token ID13
Feature activation+2.25e-02

Pos logit contributions
 sizes+0.291
 ABC+0.287
 planets+0.286
 cultures+0.282
 habitats+0.279
Neg logit contributions
 late-0.233
sm-0.197
 aback-0.177
 another-0.170
Start-0.170
Token They
Token ID1119
Feature activation+6.65e-02

Pos logit contributions
 sizes+0.046
 planets+0.042
bows+0.041
 clubs+0.041
 cultures+0.040
Neg logit contributions
 late-0.032
O-0.026
Start-0.026
 aback-0.026
sm-0.025
Token can
Token ID460
Feature activation+7.81e-02

Pos logit contributions
 sizes+0.125
 ABC+0.116
bows+0.116
 planets+0.113
 cultures+0.112
Neg logit contributions
 late-0.093
 aback-0.076
O-0.075
Start-0.074
sm-0.073
Token make
Token ID787
Feature activation+3.21e-02

Pos logit contributions
 sizes+0.140
bows+0.126
 cultures+0.124
 planets+0.124
 languages+0.120
Neg logit contributions
 late-0.126
O-0.106
Start-0.103
 aback-0.101
izz-0.099
Token towers
Token ID18028
Feature activation+9.38e-02

Pos logit contributions
 sizes+0.056
bows+0.053
 cultures+0.052
 cheeks+0.050
 habitats+0.050
Neg logit contributions
 late-0.053
 another-0.043
 aback-0.041
Start-0.041
 my-0.041
Token they
Token ID484
Feature activation+3.27e-02

Pos logit contributions
 sizes+0.002
 planets+0.002
 ABC+0.002
 cultures+0.002
 nations+0.002
Neg logit contributions
 late-0.002
 aback-0.002
Start-0.002
O-0.002
sm-0.002
Token're
Token ID821
Feature activation+6.83e-02

Pos logit contributions
 sizes+0.051
 planets+0.043
bows+0.042
 ABC+0.040
 cultures+0.040
Neg logit contributions
 late-0.067
O-0.058
 aback-0.058
Start-0.057
sm-0.057
Token different
Token ID1180
Feature activation+2.25e-02

Pos logit contributions
 sizes+0.126
 planets+0.109
bows+0.105
 ABC+0.103
 toes+0.102
Neg logit contributions
 late-0.114
 aback-0.103
O-0.102
Start-0.102
sm-0.099
Token colors
Token ID7577
Feature activation+1.24e-01

Pos logit contributions
 sizes+0.039
 clubs+0.036
 necks+0.036
 ABC+0.035
 planets+0.035
Neg logit contributions
 late-0.037
 aback-0.030
sm-0.029
 my-0.028
Start-0.027
Token and
Token ID290
Feature activation+2.91e-03

Pos logit contributions
 sizes+0.241
 cultures+0.221
 ABC+0.220
 planets+0.219
 clubs+0.219
Neg logit contributions
 late-0.182
 aback-0.142
sm-0.141
Start-0.132
O-0.130
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
 sizes+0.004
 ABC+0.003
bows+0.003
 clubs+0.003
 planets+0.003
Neg logit contributions
 late-0.006
 aback-0.005
sm-0.005
Start-0.005
 my-0.005
Token."
Token ID526
Feature activation-1.91e-02

Pos logit contributions
 sizes+0.297
 planets+0.269
 ABC+0.267
 cultures+0.262
 clubs+0.261
Neg logit contributions
 late-0.256
 aback-0.216
sm-0.212
Start-0.206
O-0.201
Token
Token ID198
Feature activation-1.24e-02

Pos logit contributions
 late+0.022
 aback+0.016
Start+0.016
O+0.016
sm+0.014
Neg logit contributions
 sizes-0.048
 planets-0.043
 cultures-0.043
bows-0.042
 nations-0.041
Token
Token ID198
Feature activation-1.66e-02

Pos logit contributions
 late+0.018
 aback+0.015
Start+0.015
O+0.015
sm+0.015
Neg logit contributions
 sizes-0.025
 planets-0.022
bows-0.022
 ABC-0.022
 clubs-0.022
TokenMax
Token ID11518
Feature activation+4.56e-02

Pos logit contributions
 late+0.025
O+0.021
Start+0.021
 aback+0.020
sm+0.020
Neg logit contributions
 sizes-0.034
bows-0.030
 planets-0.030
 ABC-0.029
 toes-0.029
Token started
Token ID2067
Feature activation+9.45e-03

Pos logit contributions
 sizes+0.079
 planets+0.069
bows+0.068
 ABC+0.067
 cultures+0.066
Neg logit contributions
 late-0.080
 aback-0.070
Start-0.068
O-0.068
sm-0.067
Token cars
Token ID5006
Feature activation+4.84e-02

Pos logit contributions
 late+0.079
sm+0.072
 aback+0.067
 my+0.067
O+0.067
Neg logit contributions
bows-0.047
 sizes-0.044
hips-0.042
 cultures-0.041
 cheeks-0.040
Token of
Token ID286
Feature activation-1.45e-02

Pos logit contributions
 sizes+0.084
 planets+0.077
 cultures+0.077
 toes+0.077
 cheeks+0.074
Neg logit contributions
 late-0.083
sm-0.070
 aback-0.067
Start-0.062
O-0.062
Token different
Token ID1180
Feature activation+1.78e-02

Pos logit contributions
 late+0.021
Start+0.017
 aback+0.017
O+0.017
sm+0.017
Neg logit contributions
 sizes-0.028
bows-0.026
 planets-0.025
 cultures-0.025
 ABC-0.025
Token colors
Token ID7577
Feature activation+1.16e-01

Pos logit contributions
 mast+0.018
hips+0.018
sing+0.017
 cheeks+0.016
 disco+0.016
Neg logit contributions
 late-0.036
 aback-0.032
sm-0.032
 my-0.030
 another-0.029
Token and
Token ID290
Feature activation+7.41e-03

Pos logit contributions
 cultures+0.205
 sizes+0.205
 antics+0.197
 habitats+0.197
 ABC+0.195
Neg logit contributions
 late-0.170
sm-0.132
 aback-0.123
 another-0.121
 my-0.119
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
pit+0.005
utes+0.005
castle+0.004
azes+0.004
bles+0.004
Neg logit contributions
 late-0.018
 my-0.016
 another-0.016
 the-0.016
Start-0.015
Token.
Token ID13
Feature activation+2.25e-02

Pos logit contributions
 sizes+0.290
 ABC+0.289
 planets+0.284
 cultures+0.283
 habitats+0.278
Neg logit contributions
 late-0.227
sm-0.194
 aback-0.175
 another-0.163
Start-0.161
Token They
Token ID1119
Feature activation+6.40e-02

Pos logit contributions
 sizes+0.040
 planets+0.035
 ABC+0.034
bows+0.034
 toes+0.033
Neg logit contributions
 late-0.039
 aback-0.035
sm-0.034
Start-0.033
O-0.033
Token make
Token ID787
Feature activation+2.90e-02

Pos logit contributions
 sizes+0.120
 ABC+0.112
bows+0.110
azes+0.109
 cultures+0.108
Neg logit contributions
 late-0.088
 aback-0.071
O-0.070
sm-0.069
Start-0.069
Token roads
Token ID9725
Feature activation+6.36e-02

Pos logit contributions
 sizes+0.045
 cultures+0.040
bows+0.040
 toes+0.039
 cheeks+0.038
Neg logit contributions
 late-0.054
Start-0.044
 aback-0.043
 another-0.043
sm-0.043
Token with
Token ID351
Feature activation-1.79e-02

Pos logit contributions
 sizes+0.117
 cheeks+0.112
 toes+0.112
 habitats+0.111
 cultures+0.111
Neg logit contributions
 late-0.093
 another-0.073
 aback-0.067
 away-0.065
 my-0.064
Token it
Token ID340
Feature activation-4.57e-02

Pos logit contributions
 sizes+0.070
 planets+0.065
 patterns+0.063
 ABC+0.063
 toes+0.061
Neg logit contributions
 late-0.043
 aback-0.037
O-0.035
sm-0.035
 severe-0.035
Token with
Token ID351
Feature activation-8.08e-03

Pos logit contributions
 late+0.077
 aback+0.066
Start+0.065
O+0.065
sm+0.064
Neg logit contributions
 sizes-0.083
 planets-0.073
bows-0.072
 ABC-0.071
 clubs-0.070
Token holes
Token ID10421
Feature activation+4.16e-02

Pos logit contributions
 late+0.011
 aback+0.010
izz+0.009
Start+0.009
sm+0.009
Neg logit contributions
 sizes-0.017
 planets-0.016
 patterns-0.015
 toes-0.015
 clubs-0.015
Token of
Token ID286
Feature activation+9.36e-03

Pos logit contributions
 sizes+0.064
 planets+0.055
bows+0.054
 ABC+0.053
 clubs+0.052
Neg logit contributions
 late-0.081
 aback-0.071
Start-0.070
O-0.070
sm-0.069
Token different
Token ID1180
Feature activation+3.26e-02

Pos logit contributions
 sizes+0.021
 planets+0.020
 patterns+0.019
 clubs+0.019
 toes+0.018
Neg logit contributions
 late-0.012
 aback-0.010
izz-0.010
Start-0.010
sm-0.010
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
 sizes+0.036
bows+0.032
 ABC+0.032
hips+0.030
 planets+0.030
Neg logit contributions
 late-0.072
 aback-0.065
sm-0.063
Start-0.063
O-0.061
Token.
Token ID13
Feature activation+2.42e-02

Pos logit contributions
 sizes+0.269
 planets+0.236
 ABC+0.234
bows+0.233
 cultures+0.229
Neg logit contributions
 late-0.289
 aback-0.248
sm-0.243
Start-0.240
O-0.238
Token I
Token ID314
Feature activation-5.25e-02

Pos logit contributions
 sizes+0.048
 planets+0.043
bows+0.042
 ABC+0.042
 clubs+0.041
Neg logit contributions
 late-0.036
 aback-0.031
Start-0.030
sm-0.030
O-0.030
Token will
Token ID481
Feature activation+7.58e-02

Pos logit contributions
 late+0.105
 aback+0.091
O+0.091
 severe+0.090
Start+0.089
Neg logit contributions
 sizes-0.078
bows-0.067
 planets-0.062
 ABC-0.061
 patterns-0.060
Token make
Token ID787
Feature activation+3.42e-02

Pos logit contributions
 sizes+0.138
 planets+0.121
bows+0.121
 ABC+0.119
 cultures+0.117
Neg logit contributions
 late-0.126
 aback-0.106
Start-0.106
O-0.104
sm-0.101
Token it
Token ID340
Feature activation-4.53e-02

Pos logit contributions
 sizes+0.074
 planets+0.069
 patterns+0.067
 ABC+0.066
 clubs+0.065
Neg logit contributions
 late-0.045
 aback-0.038
O-0.037
sm-0.036
 severe-0.036
Token cars
Token ID5006
Feature activation+5.01e-02

Pos logit contributions
 late+0.083
sm+0.075
 aback+0.071
 severe+0.070
O+0.070
Neg logit contributions
 sizes-0.045
bows-0.044
hips-0.039
 cultures-0.039
 planets-0.038
Token of
Token ID286
Feature activation-1.27e-02

Pos logit contributions
 sizes+0.106
 cultures+0.099
 planets+0.099
 toes+0.097
 cheeks+0.095
Neg logit contributions
 late-0.067
sm-0.053
 aback-0.051
Start-0.044
O-0.044
Token different
Token ID1180
Feature activation+1.84e-02

Pos logit contributions
 late+0.017
 aback+0.014
Start+0.014
O+0.014
sm+0.014
Neg logit contributions
 sizes-0.026
bows-0.024
 planets-0.024
 ABC-0.023
 cultures-0.023
Token colors
Token ID7577
Feature activation+1.18e-01

Pos logit contributions
 mast+0.019
hips+0.019
 disco+0.018
sing+0.017
igo+0.017
Neg logit contributions
 late-0.037
 aback-0.033
sm-0.033
 another-0.030
 my-0.030
Token and
Token ID290
Feature activation+6.62e-03

Pos logit contributions
 cultures+0.233
 sizes+0.225
 habitats+0.223
 antics+0.223
 necks+0.220
Neg logit contributions
 late-0.152
sm-0.116
 another-0.107
 aback-0.103
out-0.097
Token sizes
Token ID10620
Feature activation+1.62e-01

Pos logit contributions
pit+0.005
utes+0.005
castle+0.005
azes+0.005
 spine+0.004
Neg logit contributions
 late-0.015
sm-0.013
 the-0.013
 another-0.013
 my-0.013
Token.
Token ID13
Feature activation+2.32e-02

Pos logit contributions
 ABC+0.306
 planets+0.304
 cultures+0.303
 habitats+0.300
 sizes+0.299
Neg logit contributions
 late-0.200
sm-0.178
 aback-0.150
 another-0.148
out-0.140
Token They
Token ID1119
Feature activation+6.77e-02

Pos logit contributions
 sizes+0.044
bows+0.039
 planets+0.039
 ABC+0.038
 cultures+0.038
Neg logit contributions
 late-0.037
O-0.030
 aback-0.030
Start-0.030
sm-0.029
Token liked
Token ID8288
Feature activation+5.72e-02

Pos logit contributions
 sizes+0.109
 ABC+0.099
bows+0.099
 cultures+0.098
 planets+0.097
Neg logit contributions
 late-0.115
O-0.099
sm-0.099
 aback-0.095
Start-0.094
Token to
Token ID284
Feature activation+1.91e-02

Pos logit contributions
 sizes+0.134
bows+0.123
 planets+0.121
hips+0.119
 toes+0.119
Neg logit contributions
 late-0.065
sm-0.047
O-0.044
Start-0.044
 aback-0.043
Token zoom
Token ID19792
Feature activation+3.43e-02

Pos logit contributions
 sizes+0.032
 planets+0.029
bows+0.028
 cultures+0.027
 ABC+0.027
Neg logit contributions
 late-0.032
 aback-0.028
Start-0.028
O-0.028
sm-0.027
Token took
Token ID1718
Feature activation-7.12e-02

Pos logit contributions
 sizes+0.037
 planets+0.033
bows+0.032
 ABC+0.032
 clubs+0.031
Neg logit contributions
 late-0.037
 aback-0.032
Start-0.031
O-0.031
sm-0.031
Token turns
Token ID4962
Feature activation+4.22e-02

Pos logit contributions
 late+0.110
O+0.095
Start+0.094
 severe+0.093
sm+0.093
Neg logit contributions
 sizes-0.140
 planets-0.128
 clubs-0.123
 toes-0.123
 patterns-0.121
Token to
Token ID284
Feature activation-5.70e-04

Pos logit contributions
 sizes+0.072
 cultures+0.065
 planets+0.064
bows+0.063
 hobbies+0.063
Neg logit contributions
 late-0.078
sm-0.064
 aback-0.063
Start-0.062
 poisoned-0.059
Token measure
Token ID3953
Feature activation+1.58e-03

Pos logit contributions
 late+0.001
 aback+0.001
sm+0.001
Start+0.001
O+0.001
Neg logit contributions
 sizes-0.001
 planets-0.001
bows-0.001
 ABC-0.001
 clubs-0.001
Token their
Token ID511
Feature activation+4.84e-02

Pos logit contributions
 sizes+0.004
 planets+0.003
bows+0.003
 ABC+0.003
 clubs+0.003
Neg logit contributions
 late-0.002
 aback-0.002
Start-0.001
sm-0.001
O-0.001
Token tails
Token ID30514
Feature activation+1.61e-01

Pos logit contributions
 sizes+0.068
bows+0.058
 planets+0.057
 cultures+0.057
hips+0.056
Neg logit contributions
 late-0.096
 aback-0.083
sm-0.081
Start-0.080
 poisonous-0.080
Token.
Token ID13
Feature activation+1.71e-02

Pos logit contributions
 sizes+0.295
 cultures+0.277
 ABC+0.273
 hobbies+0.265
bows+0.265
Neg logit contributions
 late-0.247
 aback-0.195
 another-0.190
sm-0.185
Start-0.181
Token The
Token ID383
Feature activation+3.07e-02

Pos logit contributions
 sizes+0.030
 planets+0.027
bows+0.026
 ABC+0.026
 clubs+0.025
Neg logit contributions
 late-0.030
 aback-0.026
sm-0.025
Start-0.025
O-0.025
Token mouse
Token ID10211
Feature activation-2.40e-03

Pos logit contributions
 sizes+0.063
bows+0.055
 clubs+0.054
 toes+0.054
 planets+0.054
Neg logit contributions
 late-0.046
O-0.037
Start-0.037
sm-0.037
 aback-0.036
Token's
Token ID338
Feature activation+5.12e-02

Pos logit contributions
 late+0.004
 aback+0.004
Start+0.004
sm+0.004
O+0.004
Neg logit contributions
 sizes-0.004
 planets-0.004
bows-0.004
 ABC-0.003
 cultures-0.003
Token tail
Token ID7894
Feature activation+6.55e-02

Pos logit contributions
 sizes+0.077
 cultures+0.066
 planets+0.065
bows+0.064
hips+0.063
Neg logit contributions
 late-0.096
 aback-0.080
 another-0.079
 poisonous-0.079
 poisoned-0.078
Token boats
Token ID16659
Feature activation+3.19e-02

Pos logit contributions
 late+0.077
sm+0.069
 aback+0.066
O+0.064
Start+0.064
Neg logit contributions
bows-0.053
 sizes-0.051
 ABC-0.047
hips-0.047
azes-0.045
Token of
Token ID286
Feature activation-1.90e-02

Pos logit contributions
 sizes+0.063
 ABC+0.060
 planets+0.058
 cheeks+0.057
igo+0.057
Neg logit contributions
 late-0.045
sm-0.037
 aback-0.036
 poisoned-0.031
Start-0.030
Token different
Token ID1180
Feature activation+1.54e-02

Pos logit contributions
 late+0.026
 aback+0.021
O+0.021
sm+0.021
Start+0.021
Neg logit contributions
 sizes-0.039
bows-0.036
 ABC-0.035
 planets-0.035
 cultures-0.034
Token colors
Token ID7577
Feature activation+1.18e-01

Pos logit contributions
bows+0.016
azes+0.016
hips+0.016
sing+0.015
igo+0.015
Neg logit contributions
 late-0.031
sm-0.028
 aback-0.028
 my-0.026
Start-0.026
Token and
Token ID290
Feature activation+7.96e-03

Pos logit contributions
 cultures+0.215
 sizes+0.214
 ABC+0.214
 antics+0.205
 cheeks+0.205
Neg logit contributions
 late-0.157
sm-0.132
 aback-0.117
 another-0.115
 here-0.110
Token sizes
Token ID10620
Feature activation+1.61e-01

Pos logit contributions
pit+0.005
azes+0.004
hips+0.004
 ABC+0.004
iece+0.004
Neg logit contributions
 late-0.020
sm-0.018
 aback-0.017
 my-0.017
 another-0.017
Token.
Token ID13
Feature activation+2.35e-02

Pos logit contributions
 ABC+0.293
 sizes+0.287
 planets+0.280
 cultures+0.278
azes+0.271
Neg logit contributions
 late-0.217
sm-0.201
 aback-0.176
 another-0.163
 here-0.160
Token They
Token ID1119
Feature activation+6.65e-02

Pos logit contributions
 sizes+0.043
 planets+0.037
bows+0.037
 ABC+0.037
 clubs+0.036
Neg logit contributions
 late-0.040
 aback-0.034
Start-0.033
O-0.033
sm-0.033
Token liked
Token ID8288
Feature activation+5.79e-02

Pos logit contributions
 sizes+0.104
 ABC+0.094
bows+0.094
 cultures+0.090
 planets+0.090
Neg logit contributions
 late-0.117
sm-0.103
O-0.102
 aback-0.101
Start-0.098
Token to
Token ID284
Feature activation+1.73e-02

Pos logit contributions
 sizes+0.134
bows+0.123
 planets+0.120
hips+0.119
 ABC+0.118
Neg logit contributions
 late-0.068
sm-0.050
 aback-0.048
O-0.048
Start-0.048
Token make
Token ID787
Feature activation+4.07e-02

Pos logit contributions
 sizes+0.029
bows+0.025
 planets+0.025
 ABC+0.025
 cultures+0.024
Neg logit contributions
 late-0.029
 aback-0.025
O-0.025
Start-0.025
sm-0.025
Token Jack
Token ID3619
Feature activation+3.89e-02

Pos logit contributions
 late+0.020
 aback+0.017
sm+0.017
Start+0.017
O+0.017
Neg logit contributions
 sizes-0.023
 planets-0.020
 clubs-0.020
 toes-0.020
 ABC-0.020
Token and
Token ID290
Feature activation-1.60e-02

Pos logit contributions
 sizes+0.061
 planets+0.053
bows+0.053
 ABC+0.052
 cultures+0.051
Neg logit contributions
 late-0.075
 aback-0.066
Start-0.064
sm-0.063
O-0.062
Token w
Token ID266
Feature activation+3.55e-02

Pos logit contributions
 late+0.031
sm+0.029
O+0.029
izz+0.028
 aback+0.027
Neg logit contributions
 sizes-0.022
 neck-0.019
 toes-0.019
 planets-0.018
 clubs-0.018
Tokenagged
Token ID14655
Feature activation+5.77e-02

Pos logit contributions
 sizes+0.070
 planets+0.062
bows+0.062
 ABC+0.061
 cultures+0.060
Neg logit contributions
 late-0.054
 aback-0.047
Start-0.045
sm-0.045
O-0.045
Token their
Token ID511
Feature activation+5.81e-02

Pos logit contributions
 sizes+0.134
 ABC+0.124
bows+0.124
 planets+0.123
 cultures+0.120
Neg logit contributions
 late-0.067
 aback-0.053
Start-0.049
sm-0.049
O-0.047
Token tails
Token ID30514
Feature activation+1.61e-01

Pos logit contributions
 sizes+0.065
bows+0.060
 ABC+0.057
 planets+0.055
 cultures+0.054
Neg logit contributions
 late-0.130
sm-0.115
 aback-0.113
 poisonous-0.113
O-0.113
Token.
Token ID13
Feature activation+1.94e-02

Pos logit contributions
 sizes+0.314
 cultures+0.298
 ABC+0.297
bows+0.291
 planets+0.290
Neg logit contributions
 late-0.249
 aback-0.200
sm-0.194
 another-0.185
Start-0.185
Token Jack
Token ID3619
Feature activation+2.39e-02

Pos logit contributions
 sizes+0.034
 planets+0.030
bows+0.030
 ABC+0.029
 clubs+0.029
Neg logit contributions
 late-0.034
 aback-0.029
Start-0.029
O-0.028
sm-0.028
Token was
Token ID373
Feature activation+2.47e-02

Pos logit contributions
 sizes+0.039
 planets+0.034
bows+0.034
 ABC+0.033
 clubs+0.033
Neg logit contributions
 late-0.044
 aback-0.038
Start-0.037
sm-0.037
O-0.037
Token so
Token ID523
Feature activation-2.06e-02

Pos logit contributions
 sizes+0.037
 neck+0.031
 racing+0.031
 pretending+0.031
 toes+0.030
Neg logit contributions
 late-0.043
O-0.042
sm-0.041
Start-0.041
 aback-0.040
Token happy
Token ID3772
Feature activation-1.83e-02

Pos logit contributions
 late+0.038
O+0.035
 aback+0.035
sm+0.034
Start+0.034
Neg logit contributions
 sizes-0.030
 toes-0.024
bows-0.024
 ABC-0.024
 planets-0.023
Token His
Token ID2399
Feature activation+3.64e-02

Pos logit contributions
sm+0.007
 late+0.007
 poisonous+0.007
 poisoned+0.007
izz+0.007
Neg logit contributions
 sizes-0.005
 ABC-0.005
 planets-0.005
 toes-0.004
 cheeks-0.004
Token birthday
Token ID10955
Feature activation+9.46e-03

Pos logit contributions
 sizes+0.069
bows+0.060
hips+0.058
 cultures+0.057
 planets+0.057
Neg logit contributions
 late-0.060
 severe-0.050
 poisoned-0.049
 poisonous-0.049
O-0.048
Token had
Token ID550
Feature activation-1.03e-02

Pos logit contributions
 sizes+0.016
 planets+0.014
bows+0.013
 ABC+0.013
 cultures+0.013
Neg logit contributions
 late-0.017
 aback-0.015
sm-0.015
Start-0.014
O-0.014
Token ended
Token ID4444
Feature activation+4.96e-04

Pos logit contributions
 late+0.016
 aback+0.015
sm+0.014
izz+0.014
O+0.014
Neg logit contributions
 sizes-0.018
 planets-0.016
bows-0.016
 clubs-0.016
 ABC-0.016
Token in
Token ID287
Feature activation-1.72e-02

Pos logit contributions
 sizes+0.001
 planets+0.001
 cultures+0.001
bows+0.001
 ABC+0.001
Neg logit contributions
 late-0.001
 aback-0.001
sm-0.001
Start-0.001
 severe-0.001
Token disaster
Token ID9336
Feature activation-1.71e-01

Pos logit contributions
 late+0.022
 aback+0.020
O+0.020
Start+0.020
sm+0.019
Neg logit contributions
 sizes-0.035
 planets-0.032
 ABC-0.031
 clubs-0.030
bows-0.030
Token.
Token ID13
Feature activation-1.00e-02

Pos logit contributions
 late+0.320
 aback+0.286
O+0.282
Start+0.279
sm+0.275
Neg logit contributions
 sizes-0.269
bows-0.229
 planets-0.228
 ABC-0.219
 clubs-0.215
Token<|endoftext|>
Token ID50256
Feature activation-4.56e-02

Pos logit contributions
 late+0.017
 poisonous+0.017
sm+0.016
 poisoned+0.016
O+0.016
Neg logit contributions
 sizes-0.015
 ABC-0.014
 planets-0.013
 languages-0.012
bows-0.012
TokenJimmy
Token ID40335
Feature activation-1.40e-02

Pos logit contributions
 late+0.053
 aback+0.043
Start+0.041
sm+0.041
O+0.040
Neg logit contributions
 sizes-0.105
 planets-0.095
bows-0.094
 ABC-0.094
 clubs-0.092
Token was
Token ID373
Feature activation-1.87e-02

Pos logit contributions
 late+0.027
 aback+0.024
Start+0.024
O+0.024
sm+0.023
Neg logit contributions
 sizes-0.021
 planets-0.018
bows-0.018
 ABC-0.018
 toes-0.017
Token new
Token ID649
Feature activation-1.77e-02

Pos logit contributions
 late+0.038
O+0.037
 aback+0.037
Start+0.036
sm+0.036
Neg logit contributions
 sizes-0.022
 planets-0.020
 ABC-0.020
 toes-0.020
 pretending-0.019
TokenMom
Token ID29252
Feature activation-4.38e-02

Pos logit contributions
 late+0.044
 aback+0.040
sm+0.037
 poisoned+0.035
 poisonous+0.035
Neg logit contributions
 sizes-0.059
 ABC-0.055
 planets-0.054
 clubs-0.053
 toes-0.053
Token came
Token ID1625
Feature activation-4.47e-02

Pos logit contributions
 late+0.069
 aback+0.060
Start+0.059
O+0.058
sm+0.058
Neg logit contributions
 sizes-0.085
 planets-0.075
bows-0.074
 cultures-0.073
 ABC-0.073
Token and
Token ID290
Feature activation-2.94e-02

Pos logit contributions
 late+0.061
 aback+0.050
Start+0.048
sm+0.048
O+0.047
Neg logit contributions
 sizes-0.098
bows-0.088
 planets-0.087
 ABC-0.085
 cultures-0.085
Token saw
Token ID2497
Feature activation-6.18e-02

Pos logit contributions
 late+0.056
vious+0.051
 aback+0.050
ores+0.050
 severe+0.049
Neg logit contributions
 pretending-0.030
 waving-0.028
 sizes-0.028
 laughing-0.027
 neck-0.027
Token the
Token ID262
Feature activation+5.04e-03

Pos logit contributions
 late+0.083
 aback+0.070
 severe+0.069
Start+0.068
O+0.068
Neg logit contributions
 sizes-0.122
 planets-0.114
 ABC-0.110
 patterns-0.109
 clubs-0.109
Token disaster
Token ID9336
Feature activation-1.67e-01

Pos logit contributions
 sizes+0.009
 ABC+0.008
 planets+0.008
bows+0.008
 toes+0.007
Neg logit contributions
 late-0.009
 aback-0.008
Start-0.008
O-0.008
sm-0.007
Token.
Token ID13
Feature activation-1.41e-02

Pos logit contributions
 late+0.338
 aback+0.302
Start+0.294
O+0.293
sm+0.291
Neg logit contributions
 sizes-0.243
 planets-0.205
bows-0.205
 ABC-0.199
 clubs-0.193
Token She
Token ID1375
Feature activation-2.75e-02

Pos logit contributions
 late+0.023
sm+0.021
 aback+0.021
O+0.021
izz+0.020
Neg logit contributions
 sizes-0.023
 ABC-0.020
bows-0.020
 planets-0.020
 cheeks-0.019
Token was
Token ID373
Feature activation+5.61e-03

Pos logit contributions
 late+0.058
 aback+0.051
 heaven+0.048
Start+0.047
Aid+0.046
Neg logit contributions
 sizes-0.034
 planets-0.029
 ABC-0.029
 neck-0.029
 clubs-0.028
Token not
Token ID407
Feature activation-3.49e-02

Pos logit contributions
 sizes+0.007
 pretending+0.006
 racing+0.006
 neck+0.006
 toes+0.006
Neg logit contributions
 late-0.011
 aback-0.010
O-0.010
sm-0.010
Start-0.010
Token angry
Token ID7954
Feature activation-2.08e-02

Pos logit contributions
 late+0.062
O+0.060
 aback+0.059
Start+0.057
sm+0.056
Neg logit contributions
 sizes-0.051
 toes-0.044
 pretending-0.044
 neck-0.044
 ABC-0.044
Token again
Token ID757
Feature activation-7.16e-02

Pos logit contributions
 sizes+0.002
 planets+0.002
bows+0.002
 ABC+0.002
 toes+0.002
Neg logit contributions
 late-0.002
O-0.002
Start-0.002
 aback-0.002
izz-0.002
Token.
Token ID13
Feature activation+7.85e-03

Pos logit contributions
 late+0.144
 aback+0.129
Start+0.127
O+0.126
sm+0.125
Neg logit contributions
 sizes-0.105
 planets-0.089
bows-0.087
 ABC-0.085
 cultures-0.084
Token But
Token ID887
Feature activation+2.73e-02

Pos logit contributions
 sizes+0.013
 planets+0.012
 ABC+0.012
bows+0.011
 toes+0.011
Neg logit contributions
 late-0.014
sm-0.012
 aback-0.012
O-0.012
Start-0.012
Token unb
Token ID22619
Feature activation-4.91e-02

Pos logit contributions
 sizes+0.049
 planets+0.043
bows+0.042
 ABC+0.042
 cultures+0.041
Neg logit contributions
 late-0.046
 aback-0.040
Start-0.039
sm-0.039
O-0.039
Tokenek
Token ID988
Feature activation-1.38e-02

Pos logit contributions
 late+0.072
 aback+0.060
Start+0.059
sm+0.058
O+0.058
Neg logit contributions
 sizes-0.100
 planets-0.089
bows-0.088
 ABC-0.087
 clubs-0.086
Tokennown
Token ID3408
Feature activation-1.67e-01

Pos logit contributions
 late+0.025
 aback+0.021
Start+0.020
sm+0.020
O+0.020
Neg logit contributions
 sizes-0.023
bows-0.020
 planets-0.020
 ABC-0.020
 necks-0.019
Token to
Token ID284
Feature activation-7.48e-02

Pos logit contributions
 late+0.266
 severe+0.223
 aback+0.220
Start+0.219
O+0.216
Neg logit contributions
 sizes-0.293
 planets-0.270
 clubs-0.266
bows-0.259
 ABC-0.259
Token him
Token ID683
Feature activation-1.78e-02

Pos logit contributions
 late+0.109
 aback+0.092
Start+0.091
sm+0.089
O+0.087
Neg logit contributions
 sizes-0.149
 planets-0.136
 ABC-0.133
 toes-0.130
 clubs-0.130
Token,
Token ID11
Feature activation+1.12e-02

Pos logit contributions
 late+0.035
 aback+0.030
sm+0.029
Start+0.029
O+0.029
Neg logit contributions
 sizes-0.028
bows-0.024
 planets-0.024
 ABC-0.023
 clubs-0.023
Token the
Token ID262
Feature activation+5.12e-03

Pos logit contributions
 sizes+0.017
 planets+0.015
 neck+0.015
 cultures+0.015
 toes+0.015
Neg logit contributions
izz-0.018
 late-0.018
O-0.018
 poisonous-0.017
sm-0.017
Token king
Token ID5822
Feature activation-3.14e-02

Pos logit contributions
 sizes+0.008
bows+0.007
 clubs+0.006
 ABC+0.006
 cultures+0.006
Neg logit contributions
 late-0.010
Start-0.009
 aback-0.009
sm-0.009
O-0.008
Token
Token ID220
Feature activation+1.15e-02

Pos logit contributions
 sizes+0.007
bows+0.006
 planets+0.006
 ABC+0.006
 clubs+0.006
Neg logit contributions
 late-0.006
 aback-0.005
sm-0.005
O-0.005
Start-0.005
Token
Token ID198
Feature activation-1.89e-02

Pos logit contributions
 sizes+0.019
bows+0.018
 ABC+0.017
hips+0.017
 planets+0.017
Neg logit contributions
 late-0.019
 aback-0.017
 poisonous-0.017
sm-0.017
 poisoned-0.017
Token
Token ID198
Feature activation-2.25e-02

Pos logit contributions
 late+0.032
 aback+0.029
 poisonous+0.029
 poisoned+0.029
sm+0.028
Neg logit contributions
 sizes-0.029
 ABC-0.026
 clubs-0.026
bows-0.025
 planets-0.025
TokenBut
Token ID1537
Feature activation+2.40e-02

Pos logit contributions
 late+0.035
 aback+0.034
 poisonous+0.033
 poisoned+0.033
 heaven+0.030
Neg logit contributions
 sizes-0.036
 clubs-0.033
 racing-0.032
utes-0.032
 ABC-0.032
Token then
Token ID788
Feature activation-4.63e-02

Pos logit contributions
 sizes+0.040
 planets+0.035
bows+0.034
 toes+0.034
 ABC+0.034
Neg logit contributions
 late-0.042
 aback-0.037
sm-0.037
Start-0.037
O-0.036
Token disaster
Token ID9336
Feature activation-1.66e-01

Pos logit contributions
 late+0.109
 aback+0.099
sm+0.098
O+0.097
Start+0.097
Neg logit contributions
 sizes-0.049
 planets-0.040
bows-0.040
 ABC-0.038
 clubs-0.038
Token struck
Token ID7425
Feature activation-5.64e-02

Pos logit contributions
 late+0.227
 aback+0.193
O+0.185
Start+0.182
 severe+0.181
Neg logit contributions
 sizes-0.342
bows-0.308
 planets-0.300
 ABC-0.294
 clubs-0.289
Token!
Token ID0
Feature activation-2.16e-02

Pos logit contributions
 late+0.094
 aback+0.080
Start+0.078
sm+0.078
O+0.078
Neg logit contributions
 sizes-0.103
 planets-0.090
bows-0.090
 ABC-0.088
 clubs-0.087
Token Suddenly
Token ID24975
Feature activation-2.17e-02

Pos logit contributions
 late+0.034
sm+0.034
 poisonous+0.033
 poisoned+0.032
O+0.032
Neg logit contributions
 sizes-0.033
 ABC-0.032
bows-0.031
 cheeks-0.030
 planets-0.029
Token,
Token ID11
Feature activation+5.71e-03

Pos logit contributions
 late+0.056
sm+0.051
 aback+0.050
O+0.050
Start+0.050
Neg logit contributions
 sizes-0.016
bows-0.013
 planets-0.013
 clubs-0.013
 ABC-0.012
Token the
Token ID262
Feature activation-2.88e-03

Pos logit contributions
 sizes+0.009
 planets+0.009
bows+0.009
 ABC+0.008
 clubs+0.008
Neg logit contributions
sm-0.008
 aback-0.008
 late-0.008
 poisonous-0.008
Aid-0.007
Token done
Token ID1760
Feature activation+1.36e-03

Pos logit contributions
 late+0.067
 aback+0.060
Start+0.058
O+0.057
sm+0.057
Neg logit contributions
 sizes-0.057
bows-0.049
 planets-0.048
 ABC-0.048
 clubs-0.048
Token?
Token ID30
Feature activation-1.16e-02

Pos logit contributions
 sizes+0.003
 planets+0.003
bows+0.002
 ABC+0.002
 clubs+0.002
Neg logit contributions
 late-0.002
 aback-0.002
Start-0.002
sm-0.002
O-0.002
Token This
Token ID770
Feature activation-2.88e-02

Pos logit contributions
 late+0.020
 aback+0.019
 severe+0.018
 poisonous+0.018
sm+0.018
Neg logit contributions
 sizes-0.018
 ABC-0.016
bows-0.016
 planets-0.016
 clubs-0.015
Token is
Token ID318
Feature activation+2.63e-02

Pos logit contributions
 late+0.054
 aback+0.047
Start+0.046
sm+0.046
O+0.046
Neg logit contributions
 sizes-0.047
 planets-0.041
bows-0.040
 ABC-0.039
 cultures-0.039
Token a
Token ID257
Feature activation+1.21e-02

Pos logit contributions
 sizes+0.046
 ABC+0.040
 planets+0.039
 winning+0.039
 pretending+0.039
Neg logit contributions
 aback-0.043
Start-0.040
 late-0.038
handed-0.038
izz-0.037
Token disaster
Token ID9336
Feature activation-1.65e-01

Pos logit contributions
 sizes+0.024
 planets+0.021
bows+0.021
 ABC+0.020
 clubs+0.020
Neg logit contributions
 late-0.019
 aback-0.016
Start-0.015
sm-0.015
O-0.015
Token.
Token ID13
Feature activation-6.95e-03

Pos logit contributions
 late+0.305
 aback+0.289
O+0.284
 severe+0.278
Start+0.276
Neg logit contributions
 sizes-0.250
bows-0.216
 planets-0.211
 ABC-0.201
 pretending-0.195
Token You
Token ID921
Feature activation-2.14e-02

Pos logit contributions
 late+0.013
 aback+0.012
 poisonous+0.011
 severe+0.011
sm+0.011
Neg logit contributions
 sizes-0.011
 ABC-0.010
 planets-0.009
bows-0.009
 clubs-0.009
Token did
Token ID750
Feature activation+8.49e-02

Pos logit contributions
 late+0.046
 aback+0.045
Start+0.042
O+0.041
Aid+0.039
Neg logit contributions
 sizes-0.026
 planets-0.023
bows-0.022
 clubs-0.022
 nations-0.021
Token not
Token ID407
Feature activation-2.39e-02

Pos logit contributions
 sizes+0.095
 ABC+0.083
 planets+0.080
bows+0.079
 clubs+0.078
Neg logit contributions
 late-0.186
 aback-0.182
sm-0.172
O-0.170
izz-0.169
Token listen
Token ID6004
Feature activation+3.72e-02

Pos logit contributions
 late+0.041
 aback+0.041
O+0.038
Start+0.037
 severe+0.037
Neg logit contributions
 sizes-0.038
bows-0.033
 nations-0.032
 planets-0.031
 cultures-0.031
Token before
Token ID878
Feature activation+5.24e-03

Pos logit contributions
 sizes+0.004
 planets+0.003
bows+0.003
 ABC+0.003
 cultures+0.003
Neg logit contributions
 late-0.004
 aback-0.003
sm-0.003
Start-0.003
O-0.003
Token.
Token ID13
Feature activation+6.65e-03

Pos logit contributions
 sizes+0.009
bows+0.007
 planets+0.007
 cultures+0.007
 ABC+0.007
Neg logit contributions
 late-0.010
 aback-0.008
Start-0.008
sm-0.008
O-0.008
Token But
Token ID887
Feature activation+2.90e-02

Pos logit contributions
 sizes+0.010
 toes+0.009
 ABC+0.009
 planets+0.009
 cheeks+0.008
Neg logit contributions
 late-0.012
sm-0.012
 poisonous-0.011
 severe-0.011
 aback-0.011
Token then
Token ID788
Feature activation-4.08e-02

Pos logit contributions
 sizes+0.044
 toes+0.041
 planets+0.039
 cheeks+0.037
 ABC+0.037
Neg logit contributions
 late-0.052
 aback-0.051
sm-0.051
Start-0.050
 severe-0.048
Token,
Token ID11
Feature activation+2.40e-02

Pos logit contributions
 late+0.071
 aback+0.062
Start+0.060
sm+0.060
O+0.060
Neg logit contributions
 sizes-0.071
 planets-0.063
bows-0.062
 ABC-0.061
 clubs-0.060
Token disaster
Token ID9336
Feature activation-1.63e-01

Pos logit contributions
 sizes+0.041
 toes+0.039
 planets+0.038
 ABC+0.036
 racing+0.034
Neg logit contributions
 aback-0.038
 late-0.037
sm-0.037
Start-0.036
O-0.034
Token struck
Token ID7425
Feature activation-5.70e-02

Pos logit contributions
 late+0.297
 aback+0.280
 severe+0.266
O+0.266
Start+0.263
Neg logit contributions
 sizes-0.246
bows-0.218
 planets-0.208
 ABC-0.201
 toes-0.192
Token!
Token ID0
Feature activation-1.97e-02

Pos logit contributions
 late+0.103
 aback+0.090
Start+0.088
sm+0.087
O+0.087
Neg logit contributions
 sizes-0.096
 planets-0.083
bows-0.082
 ABC-0.081
 clubs-0.080
Token On
Token ID1550
Feature activation-3.64e-02

Pos logit contributions
sm+0.036
 poisonous+0.035
 aback+0.035
 severe+0.034
 poisoned+0.034
Neg logit contributions
 ABC-0.027
 sizes-0.026
 toes-0.025
 cheeks-0.024
 planets-0.022
Token the
Token ID262
Feature activation+6.43e-03

Pos logit contributions
 late+0.046
 aback+0.042
sm+0.041
Start+0.040
O+0.038
Neg logit contributions
 sizes-0.078
 planets-0.074
 ABC-0.071
 toes-0.070
 clubs-0.070
Token day
Token ID1110
Feature activation-5.54e-03

Pos logit contributions
 sizes+0.009
bows+0.008
 planets+0.008
 ABC+0.007
 cultures+0.007
Neg logit contributions
 late-0.013
Start-0.011
sm-0.011
 aback-0.011
O-0.011
Token came
Token ID1625
Feature activation-4.05e-02

Pos logit contributions
 late+0.049
 aback+0.041
Start+0.040
O+0.040
sm+0.040
Neg logit contributions
 sizes-0.056
 planets-0.049
bows-0.049
 ABC-0.048
 clubs-0.047
Token running
Token ID2491
Feature activation-4.93e-03

Pos logit contributions
 late+0.069
 aback+0.059
Start+0.058
O+0.057
sm+0.057
Neg logit contributions
 sizes-0.073
 planets-0.064
bows-0.063
 ABC-0.062
 clubs-0.061
Token and
Token ID290
Feature activation-3.19e-02

Pos logit contributions
 late+0.010
O+0.009
Start+0.009
izz+0.009
 aback+0.009
Neg logit contributions
 sizes-0.006
 planets-0.006
bows-0.005
 ABC-0.005
 toes-0.005
Token saw
Token ID2497
Feature activation-5.95e-02

Pos logit contributions
 late+0.057
O+0.053
 aback+0.050
 severe+0.050
Aid+0.049
Neg logit contributions
 sizes-0.047
 pretending-0.045
 necks-0.041
 waving-0.041
 neck-0.040
Token the
Token ID262
Feature activation+1.10e-02

Pos logit contributions
 late+0.091
 aback+0.086
Start+0.082
 severe+0.080
Aid+0.079
Neg logit contributions
 sizes-0.100
 planets-0.091
 patterns-0.091
 pretending-0.089
 ABC-0.089
Token disaster
Token ID9336
Feature activation-1.61e-01

Pos logit contributions
 sizes+0.020
bows+0.018
 ABC+0.018
 planets+0.017
 patterns+0.017
Neg logit contributions
 late-0.018
 aback-0.017
Start-0.016
O-0.016
izz-0.015
Token.
Token ID13
Feature activation-9.51e-03

Pos logit contributions
 late+0.325
 aback+0.293
Start+0.284
O+0.283
sm+0.280
Neg logit contributions
 sizes-0.235
 planets-0.198
bows-0.198
 ABC-0.191
 clubs-0.185
Token She
Token ID1375
Feature activation-2.78e-02

Pos logit contributions
 aback+0.017
sm+0.017
Aid+0.017
ited+0.017
 late+0.016
Neg logit contributions
 sizes-0.012
 ABC-0.011
bows-0.010
 cheeks-0.010
azes-0.010
Token turned
Token ID2900
Feature activation-1.70e-02

Pos logit contributions
 late+0.058
 aback+0.054
 heaven+0.049
Aid+0.049
handed+0.049
Neg logit contributions
 sizes-0.033
 planets-0.029
 neck-0.028
 necks-0.028
 ABC-0.027
Token off
Token ID572
Feature activation-3.56e-02

Pos logit contributions
handed+0.038
 heaven+0.034
Gu+0.034
Start+0.034
Aid+0.034
Neg logit contributions
 sizes-0.014
 lines-0.013
 toes-0.013
,-0.012
 cheeks-0.012
Token the
Token ID262
Feature activation+1.48e-02

Pos logit contributions
handed+0.069
Aid+0.067
 aback+0.067
vous+0.063
izz+0.063
Neg logit contributions
,-0.039
 ABC-0.037
 stories-0.035
 sizes-0.034
bows-0.033
Token laughing
Token ID14376
Feature activation+3.40e-02

Pos logit contributions
 sizes+0.060
 planets+0.050
bows+0.050
 ABC+0.049
 toes+0.048
Neg logit contributions
 late-0.088
 aback-0.078
sm-0.076
O-0.076
Start-0.076
Token.
Token ID13
Feature activation+1.17e-02

Pos logit contributions
 sizes+0.041
 planets+0.034
bows+0.034
 toes+0.033
 ABC+0.031
Neg logit contributions
 late-0.072
 aback-0.068
izz-0.067
Start-0.067
O-0.067
Token "
Token ID366
Feature activation-1.93e-02

Pos logit contributions
 sizes+0.017
 ABC+0.014
bows+0.014
 planets+0.014
 toes+0.013
Neg logit contributions
 late-0.023
 aback-0.021
sm-0.020
Start-0.020
O-0.020
TokenWhat
Token ID2061
Feature activation-2.99e-02

Pos logit contributions
 aback+0.028
 late+0.027
 poisonous+0.025
 poisoned+0.025
 severe+0.024
Neg logit contributions
 sizes-0.037
 planets-0.032
bows-0.032
 necks-0.031
 ABC-0.031
Token a
Token ID257
Feature activation-3.83e-03

Pos logit contributions
 late+0.056
 aback+0.048
sm+0.048
Start+0.048
O+0.047
Neg logit contributions
 sizes-0.049
 planets-0.042
bows-0.042
 ABC-0.041
 cultures-0.041
Token disaster
Token ID9336
Feature activation-1.60e-01

Pos logit contributions
 late+0.006
 aback+0.005
Start+0.005
sm+0.005
O+0.005
Neg logit contributions
 sizes-0.008
 planets-0.007
bows-0.007
 ABC-0.007
 clubs-0.007
Token!
Token ID0
Feature activation-1.77e-02

Pos logit contributions
 late+0.294
 severe+0.274
 aback+0.274
O+0.264
Aid+0.264
Neg logit contributions
 sizes-0.242
bows-0.212
 planets-0.203
 ABC-0.193
 pretending-0.192
Token You
Token ID921
Feature activation-1.61e-02

Pos logit contributions
 late+0.030
 aback+0.028
 poisonous+0.027
 severe+0.027
sm+0.027
Neg logit contributions
 sizes-0.029
 ABC-0.026
 toes-0.024
 planets-0.024
bows-0.024
Token two
Token ID734
Feature activation+3.28e-02

Pos logit contributions
 late+0.036
 aback+0.033
O+0.032
Start+0.032
sm+0.031
Neg logit contributions
 sizes-0.019
 planets-0.017
 clubs-0.015
bows-0.015
 nations-0.015
Token and
Token ID290
Feature activation-2.78e-02

Pos logit contributions
 sizes+0.057
 planets+0.048
 toes+0.048
bows+0.048
 clubs+0.047
Neg logit contributions
 late-0.058
 aback-0.053
O-0.051
Start-0.051
sm-0.048
Token Rex
Token ID17853
Feature activation+5.93e-02

Pos logit contributions
 late+0.055
 aback+0.052
Start+0.050
sm+0.049
izz+0.048
Neg logit contributions
 sizes-0.042
 toes-0.039
 planets-0.036
 neck-0.035
 cheeks-0.035
Token to
Token ID284
Feature activation-1.93e-02

Pos logit contributions
 late+0.029
 aback+0.025
Start+0.024
O+0.024
sm+0.024
Neg logit contributions
 sizes-0.039
 planets-0.034
bows-0.034
 ABC-0.034
 toes-0.033
Token start
Token ID923
Feature activation+4.94e-03

Pos logit contributions
 late+0.038
 aback+0.034
sm+0.033
Start+0.032
O+0.032
Neg logit contributions
 sizes-0.029
 planets-0.026
 ABC-0.025
 clubs-0.025
bows-0.025
Token his
Token ID465
Feature activation+5.19e-03

Pos logit contributions
 sizes+0.010
bows+0.009
 planets+0.009
 cultures+0.009
 ABC+0.009
Neg logit contributions
 late-0.007
 aback-0.006
Start-0.006
sm-0.006
O-0.006
Token journey
Token ID7002
Feature activation+1.05e-02

Pos logit contributions
 sizes+0.009
bows+0.008
 planets+0.008
 cultures+0.007
 ABC+0.007
Neg logit contributions
 late-0.009
 aback-0.008
O-0.008
Start-0.008
sm-0.008
Token when
Token ID618
Feature activation+3.63e-02

Pos logit contributions
 sizes+0.018
 planets+0.016
 ABC+0.015
bows+0.015
 cultures+0.015
Neg logit contributions
 late-0.019
 aback-0.017
Start-0.016
sm-0.016
O-0.015
Token disaster
Token ID9336
Feature activation-1.59e-01

Pos logit contributions
 sizes+0.078
bows+0.070
 planets+0.067
 cultures+0.066
 necks+0.066
Neg logit contributions
 late-0.054
O-0.041
Start-0.041
 aback-0.040
sm-0.040
Token struck
Token ID7425
Feature activation-5.08e-02

Pos logit contributions
 late+0.281
 aback+0.251
O+0.243
Start+0.241
sm+0.239
Neg logit contributions
 sizes-0.269
bows-0.233
 planets-0.231
 ABC-0.224
 clubs-0.217
Token.
Token ID13
Feature activation-6.49e-03

Pos logit contributions
 late+0.096
 aback+0.082
Start+0.079
sm+0.078
O+0.078
Neg logit contributions
 sizes-0.085
bows-0.073
 planets-0.073
 ABC-0.073
 cultures-0.072
Token All
Token ID1439
Feature activation-1.80e-02

Pos logit contributions
 late+0.012
sm+0.011
 severe+0.011
 poisonous+0.011
izz+0.011
Neg logit contributions
 sizes-0.010
 ABC-0.009
 planets-0.008
 toes-0.008
bows-0.008
Token the
Token ID262
Feature activation-1.63e-02

Pos logit contributions
 late+0.025
O+0.025
Start+0.025
 aback+0.023
 severe+0.023
Neg logit contributions
 sizes-0.035
 planets-0.032
 toes-0.029
 nations-0.029
 languages-0.029
Token things
Token ID1243
Feature activation+3.70e-02

Pos logit contributions
 late+0.031
 aback+0.026
Start+0.025
 severe+0.025
O+0.025
Neg logit contributions
 sizes-0.025
bows-0.023
hips-0.022
 planets-0.022
 ABC-0.021
Token was
Token ID373
Feature activation+2.54e-02

Pos logit contributions
 sizes+0.052
 planets+0.045
 neck+0.043
 ABC+0.043
 necks+0.043
Neg logit contributions
 late-0.056
 aback-0.047
sm-0.046
Start-0.046
O-0.045
Token delighted
Token ID20707
Feature activation-1.26e-02

Pos logit contributions
 sizes+0.038
 ABC+0.032
 pretending+0.032
 toes+0.032
 neck+0.032
Neg logit contributions
 late-0.047
O-0.044
sm-0.044
 aback-0.043
Start-0.043
Token.
Token ID13
Feature activation+1.31e-02

Pos logit contributions
 late+0.020
 aback+0.017
sm+0.016
Start+0.016
O+0.015
Neg logit contributions
 sizes-0.026
 planets-0.023
 ABC-0.023
hips-0.022
bows-0.022
Token But
Token ID887
Feature activation+3.41e-02

Pos logit contributions
 sizes+0.023
 planets+0.020
bows+0.020
 ABC+0.019
 clubs+0.019
Neg logit contributions
 late-0.023
 aback-0.020
sm-0.019
Start-0.019
O-0.019
Token then
Token ID788
Feature activation-4.08e-02

Pos logit contributions
 sizes+0.057
 planets+0.049
 toes+0.049
 clubs+0.048
 necks+0.047
Neg logit contributions
 late-0.060
sm-0.054
 aback-0.054
Start-0.053
O-0.052
Token disaster
Token ID9336
Feature activation-1.58e-01

Pos logit contributions
 late+0.069
 aback+0.059
Start+0.058
O+0.057
sm+0.057
Neg logit contributions
 sizes-0.075
 planets-0.065
bows-0.065
 ABC-0.064
 cultures-0.063
Token struck
Token ID7425
Feature activation-5.36e-02

Pos logit contributions
 late+0.215
 aback+0.191
 severe+0.183
O+0.180
Start+0.178
Neg logit contributions
 sizes-0.314
bows-0.285
 planets-0.269
 ABC-0.263
 pretending-0.257
Token,
Token ID11
Feature activation+3.22e-03

Pos logit contributions
 late+0.082
 aback+0.068
Start+0.066
O+0.065
sm+0.065
Neg logit contributions
 sizes-0.108
 planets-0.096
bows-0.095
 ABC-0.094
 cultures-0.093
Token he
Token ID339
Feature activation-4.30e-02

Pos logit contributions
 sizes+0.005
 planets+0.004
 clubs+0.004
bows+0.004
 ABC+0.004
Neg logit contributions
 late-0.006
sm-0.005
 aback-0.005
O-0.005
Start-0.005
Token put
Token ID1234
Feature activation-7.47e-02

Pos logit contributions
 late+0.074
 aback+0.064
Start+0.062
sm+0.062
O+0.061
Neg logit contributions
 sizes-0.077
 planets-0.067
bows-0.066
 ABC-0.066
 clubs-0.065
Token the
Token ID262
Feature activation+1.50e-02

Pos logit contributions
 late+0.117
Start+0.107
 severe+0.104
sm+0.102
O+0.102
Neg logit contributions
 sizes-0.139
 planets-0.124
 clubs-0.123
 toes-0.122
 neck-0.119
Token.
Token ID13
Feature activation+5.41e-03

Pos logit contributions
 late+0.011
 aback+0.010
Start+0.009
O+0.009
sm+0.009
Neg logit contributions
 sizes-0.009
 planets-0.008
bows-0.007
 ABC-0.007
 clubs-0.007
Token<|endoftext|>
Token ID50256
Feature activation-4.18e-02

Pos logit contributions
 sizes+0.009
 ABC+0.007
 planets+0.007
 cultures+0.007
bows+0.007
Neg logit contributions
 late-0.009
 poisonous-0.009
 poisoned-0.008
 severe-0.008
sm-0.008
TokenJimmy
Token ID40335
Feature activation-1.39e-02

Pos logit contributions
 late+0.050
 aback+0.041
Start+0.039
sm+0.039
O+0.038
Neg logit contributions
 sizes-0.095
 planets-0.086
bows-0.085
 ABC-0.085
 clubs-0.083
Token was
Token ID373
Feature activation-1.09e-02

Pos logit contributions
 late+0.027
 aback+0.023
Start+0.023
O+0.023
sm+0.023
Neg logit contributions
 sizes-0.021
 planets-0.019
bows-0.018
 ABC-0.018
 cultures-0.017
Token 3
Token ID513
Feature activation-8.02e-02

Pos logit contributions
 late+0.021
 aback+0.021
Start+0.020
O+0.020
sm+0.020
Neg logit contributions
 sizes-0.014
 planets-0.012
 toes-0.012
 pretending-0.012
 ABC-0.012
Token-
Token ID12
Feature activation-1.58e-01

Pos logit contributions
 late+0.167
 aback+0.148
sm+0.146
Start+0.145
O+0.144
Neg logit contributions
 sizes-0.113
 planets-0.095
bows-0.095
 ABC-0.092
 clubs-0.090
Tokenyears
Token ID19002
Feature activation-8.81e-02

Pos logit contributions
 late+0.268
 aback+0.239
Start+0.236
O+0.235
 severe+0.221
Neg logit contributions
 sizes-0.241
 clubs-0.219
 ABC-0.216
 planets-0.216
 toes-0.215
Token-
Token ID12
Feature activation-1.34e-01

Pos logit contributions
 late+0.112
 aback+0.092
Start+0.089
O+0.088
sm+0.084
Neg logit contributions
 sizes-0.196
 planets-0.176
bows-0.174
 ABC-0.174
 clubs-0.171
Tokenold
Token ID727
Feature activation-4.46e-02

Pos logit contributions
 late+0.179
 aback+0.150
O+0.147
Start+0.146
 severe+0.135
Neg logit contributions
 sizes-0.275
 ABC-0.249
 planets-0.248
 clubs-0.246
 cultures-0.246
Token and
Token ID290
Feature activation-3.64e-02

Pos logit contributions
 late+0.081
 aback+0.069
sm+0.068
Start+0.064
O+0.062
Neg logit contributions
 sizes-0.077
bows-0.069
 planets-0.067
hips-0.067
 ABC-0.066
Token he
Token ID339
Feature activation-2.42e-02

Pos logit contributions
 late+0.062
 aback+0.052
Start+0.051
sm+0.051
O+0.050
Neg logit contributions
 sizes-0.066
bows-0.058
 planets-0.058
 ABC-0.057
 clubs-0.056
Token
Token ID198
Feature activation-5.17e-03

Pos logit contributions
 sizes+0.038
 planets+0.034
bows+0.034
 ABC+0.033
 cultures+0.033
Neg logit contributions
 late-0.033
 aback-0.028
Start-0.027
O-0.027
sm-0.026
Token
Token ID198
Feature activation-9.57e-03

Pos logit contributions
 late+0.008
 aback+0.008
sm+0.007
 poisoned+0.007
 poisonous+0.007
Neg logit contributions
 sizes-0.009
 planets-0.008
 ABC-0.008
 clubs-0.008
hips-0.007
TokenBut
Token ID1537
Feature activation+2.69e-02

Pos logit contributions
 late+0.016
 aback+0.015
sm+0.014
 poisonous+0.014
 poisoned+0.014
Neg logit contributions
 sizes-0.016
 clubs-0.014
 planets-0.014
 ABC-0.014
 grid-0.014
Token then
Token ID788
Feature activation-4.06e-02

Pos logit contributions
 sizes+0.052
bows+0.046
 planets+0.046
 ABC+0.046
 cultures+0.045
Neg logit contributions
 late-0.043
 aback-0.036
O-0.035
Start-0.035
sm-0.034
Token,
Token ID11
Feature activation+2.44e-02

Pos logit contributions
 late+0.076
 aback+0.066
Start+0.065
O+0.064
sm+0.063
Neg logit contributions
 sizes-0.069
bows-0.059
 ABC-0.058
 planets-0.058
 cultures-0.057
Token disaster
Token ID9336
Feature activation-1.57e-01

Pos logit contributions
 sizes+0.047
 planets+0.042
bows+0.041
 ABC+0.041
 clubs+0.040
Neg logit contributions
 late-0.038
 aback-0.032
Start-0.032
sm-0.031
O-0.031
Token!
Token ID0
Feature activation-1.56e-02

Pos logit contributions
 late+0.220
 aback+0.187
 severe+0.183
O+0.177
Aid+0.175
Neg logit contributions
 sizes-0.313
bows-0.284
 planets-0.278
 ABC-0.261
 toes-0.260
Token The
Token ID383
Feature activation+2.46e-02

Pos logit contributions
 late+0.027
sm+0.024
 aback+0.024
Start+0.023
O+0.023
Neg logit contributions
 sizes-0.026
 clubs-0.023
 planets-0.023
 ABC-0.022
bows-0.022
Token top
Token ID1353
Feature activation-5.11e-02

Pos logit contributions
 sizes+0.041
bows+0.035
 planets+0.035
 cultures+0.034
 clubs+0.033
Neg logit contributions
 late-0.045
 aback-0.039
Start-0.038
O-0.038
sm-0.038
Token suddenly
Token ID6451
Feature activation-2.83e-02

Pos logit contributions
 late+0.079
 aback+0.066
Start+0.064
O+0.064
sm+0.064
Neg logit contributions
 sizes-0.100
 planets-0.089
bows-0.088
 ABC-0.087
 clubs-0.086
Token stopped
Token ID5025
Feature activation-3.76e-02

Pos logit contributions
 late+0.049
 aback+0.043
Start+0.042
sm+0.041
O+0.041
Neg logit contributions
 sizes-0.049
 planets-0.043
bows-0.043
 ABC-0.042
 clubs-0.041
TokenOne
Token ID3198
Feature activation-3.84e-02

Pos logit contributions
 late+0.061
 aback+0.052
 severe+0.050
Start+0.049
 poisonous+0.049
Neg logit contributions
 sizes-0.103
 planets-0.094
bows-0.094
 ABC-0.093
 clubs-0.090
Token day
Token ID1110
Feature activation-3.60e-02

Pos logit contributions
 aback+0.089
 late+0.089
Start+0.085
plug+0.083
O+0.083
Neg logit contributions
 sizes-0.038
 ABC-0.031
 toes-0.031
 planets-0.030
bows-0.028
Token,
Token ID11
Feature activation+1.27e-03

Pos logit contributions
 late+0.093
 aback+0.086
Start+0.084
sm+0.083
 severe+0.083
Neg logit contributions
 sizes-0.028
 toes-0.023
 planets-0.022
 ABC-0.021
bows-0.021
Token a
Token ID257
Feature activation-2.93e-02

Pos logit contributions
 sizes+0.003
 planets+0.002
 toes+0.002
bows+0.002
 clubs+0.002
Neg logit contributions
 late-0.002
 aback-0.001
 severe-0.001
Start-0.001
sm-0.001
Token 3
Token ID513
Feature activation-7.79e-02

Pos logit contributions
 late+0.039
O+0.031
sm+0.030
Start+0.030
 aback+0.030
Neg logit contributions
 sizes-0.065
 planets-0.058
bows-0.058
 cultures-0.057
 clubs-0.056
Token-
Token ID12
Feature activation-1.54e-01

Pos logit contributions
 late+0.129
 aback+0.115
Start+0.106
O+0.103
 severe+0.101
Neg logit contributions
 sizes-0.141
 ABC-0.126
 planets-0.125
 toes-0.123
bows-0.121
Tokenyear
Token ID1941
Feature activation-1.14e-01

Pos logit contributions
 late+0.172
 aback+0.147
Start+0.136
O+0.133
 severe+0.127
Neg logit contributions
 sizes-0.341
 planets-0.313
 ABC-0.308
hips-0.307
 nations-0.307
Token-
Token ID12
Feature activation-1.44e-01

Pos logit contributions
 late+0.154
 aback+0.133
Start+0.128
O+0.127
sm+0.117
Neg logit contributions
 sizes-0.239
 planets-0.215
 ABC-0.214
bows-0.212
 toes-0.211
Tokenold
Token ID727
Feature activation-5.29e-02

Pos logit contributions
 late+0.186
 aback+0.159
Start+0.150
O+0.150
 severe+0.140
Neg logit contributions
 sizes-0.303
 planets-0.275
 ABC-0.272
bows-0.271
 clubs-0.270
Token girl
Token ID2576
Feature activation-4.38e-02

Pos logit contributions
 late+0.094
 aback+0.085
Start+0.083
O+0.082
 severe+0.082
Neg logit contributions
 sizes-0.082
 planets-0.075
 ABC-0.071
 clubs-0.070
 neck-0.070
Token named
Token ID3706
Feature activation-1.00e-02

Pos logit contributions
 late+0.080
 aback+0.070
Start+0.068
O+0.067
sm+0.066
Neg logit contributions
 sizes-0.072
 planets-0.063
bows-0.062
 ABC-0.061
 clubs-0.061
TokenOne
Token ID3198
Feature activation-4.03e-02

Pos logit contributions
 late+0.064
 aback+0.052
Start+0.050
sm+0.050
O+0.049
Neg logit contributions
 sizes-0.113
 planets-0.103
bows-0.102
 ABC-0.101
 clubs-0.099
Token day
Token ID1110
Feature activation-3.63e-02

Pos logit contributions
 late+0.093
 aback+0.091
Start+0.088
O+0.086
plug+0.085
Neg logit contributions
 sizes-0.042
 planets-0.034
 ABC-0.034
 toes-0.033
bows-0.031
Token,
Token ID11
Feature activation-2.64e-05

Pos logit contributions
 late+0.087
 aback+0.079
Start+0.077
sm+0.077
O+0.077
Neg logit contributions
 sizes-0.040
 planets-0.032
bows-0.031
 ABC-0.030
 clubs-0.029
Token a
Token ID257
Feature activation-3.28e-02

Pos logit contributions
 late+0.000
 aback+0.000
Start+0.000
sm+0.000
O+0.000
Neg logit contributions
 sizes-0.000
 planets-0.000
bows-0.000
 ABC-0.000
 clubs-0.000
Token three
Token ID1115
Feature activation-6.93e-02

Pos logit contributions
 late+0.043
O+0.034
sm+0.033
Start+0.033
 aback+0.032
Neg logit contributions
 sizes-0.074
 planets-0.066
bows-0.066
 cultures-0.065
 clubs-0.064
Token-
Token ID12
Feature activation-1.53e-01

Pos logit contributions
 late+0.101
 aback+0.087
Start+0.081
O+0.079
sm+0.077
Neg logit contributions
 sizes-0.142
 planets-0.127
 ABC-0.125
bows-0.124
 clubs-0.123
Tokenyear
Token ID1941
Feature activation-1.11e-01

Pos logit contributions
 late+0.158
 aback+0.130
Start+0.122
O+0.122
 severe+0.117
Neg logit contributions
 sizes-0.354
 planets-0.327
 clubs-0.320
bows-0.319
 cultures-0.319
Token-
Token ID12
Feature activation-1.41e-01

Pos logit contributions
 late+0.140
 aback+0.116
Start+0.111
O+0.111
sm+0.107
Neg logit contributions
 sizes-0.247
 planets-0.223
bows-0.220
 ABC-0.219
 clubs-0.217
Tokenold
Token ID727
Feature activation-5.36e-02

Pos logit contributions
 late+0.171
 aback+0.142
O+0.136
Start+0.135
 severe+0.126
Neg logit contributions
 sizes-0.313
 planets-0.285
bows-0.282
 ABC-0.279
 clubs-0.279
Token boy
Token ID2933
Feature activation-3.12e-02

Pos logit contributions
 late+0.090
 aback+0.078
Start+0.077
O+0.075
 severe+0.074
Neg logit contributions
 sizes-0.094
 planets-0.085
bows-0.082
 clubs-0.082
 ABC-0.081
Token was
Token ID373
Feature activation-1.01e-02

Pos logit contributions
 late+0.053
 aback+0.045
sm+0.045
O+0.044
Start+0.044
Neg logit contributions
 sizes-0.057
 planets-0.050
bows-0.050
 ABC-0.049
 cultures-0.048
Token day
Token ID1110
Feature activation-2.88e-02

Pos logit contributions
 aback+0.062
Start+0.058
 late+0.056
plug+0.056
O+0.055
Neg logit contributions
 sizes-0.017
 ABC-0.015
 toes-0.013
 planets-0.013
 hobbies-0.011
Token,
Token ID11
Feature activation+8.83e-03

Pos logit contributions
 late+0.071
 aback+0.065
Start+0.064
sm+0.063
O+0.063
Neg logit contributions
 sizes-0.028
 planets-0.023
bows-0.022
 ABC-0.022
 toes-0.021
Token a
Token ID257
Feature activation-1.78e-02

Pos logit contributions
 sizes+0.018
 planets+0.017
 toes+0.016
 ABC+0.016
bows+0.016
Neg logit contributions
 late-0.012
 aback-0.011
Start-0.010
izz-0.010
 severe-0.010
Token little
Token ID1310
Feature activation-3.69e-02

Pos logit contributions
 late+0.028
 aback+0.026
sm+0.023
Start+0.023
izz+0.023
Neg logit contributions
 sizes-0.032
 ABC-0.030
 planets-0.028
bows-0.028
 toes-0.027
Token 3
Token ID513
Feature activation-7.50e-02

Pos logit contributions
 late+0.064
O+0.058
izz+0.058
sm+0.057
 aback+0.057
Neg logit contributions
 sizes-0.061
 planets-0.053
bows-0.052
 ABC-0.051
 clubs-0.050
Token-
Token ID12
Feature activation-1.53e-01

Pos logit contributions
 late+0.133
 aback+0.117
O+0.109
Start+0.108
 severe+0.105
Neg logit contributions
 sizes-0.130
 ABC-0.115
 planets-0.112
 toes-0.111
 languages-0.110
Tokenyear
Token ID1941
Feature activation-1.10e-01

Pos logit contributions
 late+0.192
 aback+0.161
O+0.153
Start+0.151
 severe+0.151
Neg logit contributions
 sizes-0.312
 nations-0.288
 ABC-0.286
 planets-0.285
hips-0.285
Token-
Token ID12
Feature activation-1.39e-01

Pos logit contributions
 late+0.150
 aback+0.129
O+0.126
Start+0.125
sm+0.116
Neg logit contributions
 sizes-0.228
 ABC-0.204
 planets-0.204
bows-0.200
 clubs-0.200
Tokenold
Token ID727
Feature activation-5.44e-02

Pos logit contributions
 late+0.197
 aback+0.169
O+0.165
Start+0.159
 severe+0.155
Neg logit contributions
 sizes-0.273
 ABC-0.247
 planets-0.246
bows-0.246
 cultures-0.246
Token boy
Token ID2933
Feature activation-2.80e-02

Pos logit contributions
 late+0.091
 aback+0.077
O+0.077
Start+0.076
sm+0.074
Neg logit contributions
 sizes-0.098
 planets-0.087
bows-0.085
 ABC-0.085
 cultures-0.083
Token named
Token ID3706
Feature activation-8.57e-03

Pos logit contributions
 late+0.047
 aback+0.040
Start+0.039
O+0.039
sm+0.038
Neg logit contributions
 sizes-0.050
 planets-0.044
bows-0.044
 ABC-0.043
 clubs-0.043
Token day
Token ID1110
Feature activation-2.85e-02

Pos logit contributions
 aback+0.069
 late+0.065
Start+0.064
plug+0.062
O+0.061
Neg logit contributions
 sizes-0.029
 planets-0.024
 ABC-0.024
 toes-0.024
bows-0.021
Token,
Token ID11
Feature activation+6.66e-03

Pos logit contributions
 late+0.063
 aback+0.056
O+0.056
Start+0.055
sm+0.055
Neg logit contributions
 sizes-0.038
 planets-0.031
bows-0.031
 cultures-0.030
 ABC-0.030
Token a
Token ID257
Feature activation-2.08e-02

Pos logit contributions
 sizes+0.013
 planets+0.012
 toes+0.012
 ABC+0.012
 clubs+0.012
Neg logit contributions
 late-0.009
 aback-0.009
 severe-0.008
izz-0.008
sm-0.008
Token little
Token ID1310
Feature activation-3.85e-02

Pos logit contributions
 late+0.030
 aback+0.025
Start+0.024
sm+0.024
O+0.024
Neg logit contributions
 sizes-0.042
 planets-0.038
bows-0.037
 ABC-0.037
 clubs-0.037
Token 3
Token ID513
Feature activation-7.67e-02

Pos logit contributions
 late+0.064
 aback+0.057
Start+0.056
O+0.055
sm+0.055
Neg logit contributions
 sizes-0.067
 planets-0.060
bows-0.059
 clubs-0.057
 ABC-0.057
Token-
Token ID12
Feature activation-1.53e-01

Pos logit contributions
 late+0.125
 aback+0.110
Start+0.102
O+0.100
 severe+0.098
Neg logit contributions
 sizes-0.143
 ABC-0.125
 planets-0.125
 toes-0.122
bows-0.122
Tokenyear
Token ID1941
Feature activation-1.12e-01

Pos logit contributions
 late+0.184
 aback+0.152
 severe+0.142
Start+0.141
O+0.140
Neg logit contributions
 sizes-0.319
 planets-0.300
 nations-0.297
hips-0.295
 languages-0.293
Token-
Token ID12
Feature activation-1.40e-01

Pos logit contributions
 late+0.149
 aback+0.128
Start+0.123
O+0.123
sm+0.114
Neg logit contributions
 sizes-0.237
 planets-0.214
 ABC-0.211
bows-0.210
 clubs-0.208
Tokenold
Token ID727
Feature activation-5.31e-02

Pos logit contributions
 late+0.190
 aback+0.162
O+0.154
Start+0.152
 severe+0.148
Neg logit contributions
 sizes-0.285
 planets-0.263
bows-0.258
 nations-0.257
 ABC-0.257
Token girl
Token ID2576
Feature activation-4.45e-02

Pos logit contributions
 late+0.093
 aback+0.080
Start+0.080
O+0.079
 severe+0.078
Neg logit contributions
 sizes-0.089
 planets-0.081
bows-0.078
 ABC-0.077
 clubs-0.076
Token found
Token ID1043
Feature activation-9.22e-02

Pos logit contributions
 late+0.078
 aback+0.067
Start+0.066
O+0.065
sm+0.064
Neg logit contributions
 sizes-0.077
 planets-0.068
bows-0.067
 ABC-0.066
 clubs-0.066
Token
Token ID220
Feature activation+7.58e-03

Pos logit contributions
 late+0.005
sm+0.005
 severe+0.005
 aback+0.004
 poisonous+0.004
Neg logit contributions
 sizes-0.004
 ABC-0.003
 planets-0.003
 toes-0.003
bows-0.003
Token
Token ID198
Feature activation-2.76e-02

Pos logit contributions
 sizes+0.012
 ABC+0.011
 planets+0.011
hips+0.011
bows+0.011
Neg logit contributions
 late-0.013
 aback-0.012
sm-0.011
O-0.011
 poisoned-0.011
Token
Token ID198
Feature activation-3.06e-02

Pos logit contributions
 late+0.048
 aback+0.044
 poisoned+0.042
 poisonous+0.042
sm+0.042
Neg logit contributions
 sizes-0.043
 ABC-0.039
 planets-0.037
 languages-0.035
 clubs-0.035
TokenA
Token ID32
Feature activation-1.48e-02

Pos logit contributions
 aback+0.049
 late+0.046
 poisoned+0.042
Aid+0.042
sm+0.041
Neg logit contributions
 sizes-0.050
 ABC-0.045
 planets-0.044
 necks-0.044
 cultures-0.043
Token 3
Token ID513
Feature activation-7.01e-02

Pos logit contributions
 late+0.023
 aback+0.021
sm+0.019
Start+0.019
O+0.019
Neg logit contributions
 sizes-0.027
 ABC-0.025
 planets-0.024
bows-0.023
 clubs-0.023
Token-
Token ID12
Feature activation-1.52e-01

Pos logit contributions
 late+0.121
 aback+0.106
O+0.100
Start+0.099
 severe+0.096
Neg logit contributions
 sizes-0.123
 planets-0.108
 ABC-0.107
 languages-0.105
 nations-0.105
Tokenyear
Token ID1941
Feature activation-1.06e-01

Pos logit contributions
 late+0.187
 aback+0.159
O+0.152
Start+0.147
 severe+0.145
Neg logit contributions
 sizes-0.309
 nations-0.291
 languages-0.289
hips-0.287
 planets-0.285
Token-
Token ID12
Feature activation-1.40e-01

Pos logit contributions
 late+0.145
 aback+0.125
Start+0.120
O+0.120
sm+0.113
Neg logit contributions
 sizes-0.219
 planets-0.198
 ABC-0.197
 clubs-0.193
bows-0.193
Tokenold
Token ID727
Feature activation-4.78e-02

Pos logit contributions
 late+0.198
 aback+0.172
O+0.168
Start+0.161
 severe+0.156
Neg logit contributions
 sizes-0.271
 nations-0.249
 planets-0.248
 languages-0.247
 cultures-0.245
Token child
Token ID1200
Feature activation-5.89e-02

Pos logit contributions
 late+0.082
O+0.071
 aback+0.070
Start+0.069
 severe+0.068
Neg logit contributions
 sizes-0.082
 planets-0.074
 ABC-0.071
 nations-0.070
bows-0.070
Token noticed
Token ID6810
Feature activation-6.73e-02

Pos logit contributions
 late+0.095
 aback+0.081
Start+0.079
sm+0.079
O+0.078
Neg logit contributions
 sizes-0.112
 planets-0.098
bows-0.098
 ABC-0.096
 cultures-0.095
Token day
Token ID1110
Feature activation-2.64e-02

Pos logit contributions
 aback+0.060
 late+0.059
Start+0.057
plug+0.056
 severe+0.054
Neg logit contributions
 sizes-0.026
 toes-0.021
 ABC-0.021
 planets-0.021
bows-0.019
Token,
Token ID11
Feature activation+9.51e-03

Pos logit contributions
 late+0.058
 aback+0.051
O+0.051
sm+0.050
Start+0.050
Neg logit contributions
 sizes-0.037
bows-0.030
 planets-0.030
 cultures-0.029
 ABC-0.028
Token a
Token ID257
Feature activation-1.97e-02

Pos logit contributions
 sizes+0.020
 planets+0.019
 ABC+0.018
 toes+0.018
bows+0.018
Neg logit contributions
 late-0.012
 aback-0.011
Start-0.010
sm-0.010
izz-0.010
Token little
Token ID1310
Feature activation-3.74e-02

Pos logit contributions
 late+0.029
 aback+0.024
Start+0.023
sm+0.023
O+0.023
Neg logit contributions
 sizes-0.039
 planets-0.035
 ABC-0.035
bows-0.035
 clubs-0.034
Token 3
Token ID513
Feature activation-7.64e-02

Pos logit contributions
 late+0.063
 aback+0.056
Start+0.055
sm+0.054
 severe+0.054
Neg logit contributions
 sizes-0.065
 planets-0.057
bows-0.056
 clubs-0.056
 toes-0.054
Token-
Token ID12
Feature activation-1.52e-01

Pos logit contributions
 late+0.114
 aback+0.098
Start+0.092
O+0.090
 severe+0.088
Neg logit contributions
 sizes-0.152
 planets-0.135
 ABC-0.133
bows-0.132
 clubs-0.131
Tokenyear
Token ID1941
Feature activation-1.07e-01

Pos logit contributions
 late+0.170
 aback+0.135
Start+0.127
 severe+0.126
O+0.125
Neg logit contributions
 sizes-0.341
 planets-0.313
bows-0.309
 nations-0.308
 clubs-0.307
Token-
Token ID12
Feature activation-1.38e-01

Pos logit contributions
 late+0.133
 aback+0.110
Start+0.106
O+0.105
sm+0.102
Neg logit contributions
 sizes-0.238
 planets-0.215
bows-0.212
 ABC-0.211
 clubs-0.209
Tokenold
Token ID727
Feature activation-5.05e-02

Pos logit contributions
 late+0.172
 aback+0.142
O+0.134
Start+0.134
 severe+0.128
Neg logit contributions
 sizes-0.306
 planets-0.278
bows-0.275
 ABC-0.272
 cultures-0.271
Token girl
Token ID2576
Feature activation-4.47e-02

Pos logit contributions
 late+0.083
 aback+0.070
Start+0.069
O+0.069
sm+0.067
Neg logit contributions
 sizes-0.092
 planets-0.083
bows-0.081
 ABC-0.080
 clubs-0.079
Token was
Token ID373
Feature activation-3.22e-03

Pos logit contributions
 late+0.072
 aback+0.061
Start+0.060
O+0.059
sm+0.059
Neg logit contributions
 sizes-0.084
 planets-0.075
bows-0.074
 ABC-0.073
 clubs-0.072
TokenOne
Token ID3198
Feature activation-3.92e-02

Pos logit contributions
 late+0.066
 aback+0.055
Start+0.053
sm+0.052
 severe+0.052
Neg logit contributions
 sizes-0.109
 planets-0.099
bows-0.099
 ABC-0.097
 clubs-0.095
Token day
Token ID1110
Feature activation-3.54e-02

Pos logit contributions
 late+0.091
 aback+0.090
Start+0.086
plug+0.084
O+0.084
Neg logit contributions
 sizes-0.039
 planets-0.032
 ABC-0.032
 toes-0.031
bows-0.029
Token,
Token ID11
Feature activation+1.79e-03

Pos logit contributions
 late+0.084
 aback+0.076
Start+0.074
O+0.074
sm+0.074
Neg logit contributions
 sizes-0.040
 planets-0.032
bows-0.032
 ABC-0.031
 cultures-0.030
Token a
Token ID257
Feature activation-3.04e-02

Pos logit contributions
 sizes+0.004
 planets+0.004
bows+0.004
 toes+0.004
 ABC+0.004
Neg logit contributions
 late-0.002
 aback-0.002
Start-0.002
sm-0.002
 severe-0.002
Token three
Token ID1115
Feature activation-6.48e-02

Pos logit contributions
 late+0.039
O+0.031
sm+0.030
Start+0.029
 aback+0.028
Neg logit contributions
 sizes-0.070
bows-0.063
 planets-0.062
 cultures-0.061
 clubs-0.061
Token-
Token ID12
Feature activation-1.51e-01

Pos logit contributions
 late+0.103
 aback+0.091
Start+0.085
O+0.082
 severe+0.080
Neg logit contributions
 sizes-0.125
 planets-0.111
 ABC-0.110
 toes-0.108
 clubs-0.107
Tokenyear
Token ID1941
Feature activation-1.11e-01

Pos logit contributions
 late+0.154
 aback+0.127
Start+0.117
O+0.117
 severe+0.111
Neg logit contributions
 sizes-0.357
 planets-0.329
bows-0.321
 ABC-0.321
 cultures-0.321
Token-
Token ID12
Feature activation-1.41e-01

Pos logit contributions
 late+0.146
 aback+0.123
Start+0.118
O+0.118
sm+0.111
Neg logit contributions
 sizes-0.240
 planets-0.217
 ABC-0.214
bows-0.213
 clubs-0.211
Tokenold
Token ID727
Feature activation-5.37e-02

Pos logit contributions
 late+0.171
 aback+0.141
O+0.134
Start+0.134
 severe+0.125
Neg logit contributions
 sizes-0.314
 planets-0.286
bows-0.282
 ABC-0.281
 cultures-0.279
Token boy
Token ID2933
Feature activation-2.90e-02

Pos logit contributions
 late+0.091
 aback+0.080
Start+0.079
O+0.078
 severe+0.076
Neg logit contributions
 sizes-0.093
 planets-0.084
bows-0.081
 ABC-0.080
 clubs-0.080
Token was
Token ID373
Feature activation-9.93e-03

Pos logit contributions
 late+0.052
 aback+0.045
Start+0.044
O+0.044
sm+0.043
Neg logit contributions
 sizes-0.050
 planets-0.043
bows-0.043
 ABC-0.042
 clubs-0.042
Token classroom
Token ID15806
Feature activation-1.41e-02

Pos logit contributions
 late+0.039
O+0.031
sm+0.030
Start+0.029
 aback+0.027
Neg logit contributions
 sizes-0.070
bows-0.062
 planets-0.062
 cultures-0.061
 clubs-0.061
Token.
Token ID13
Feature activation+6.88e-03

Pos logit contributions
 late+0.022
sm+0.019
Start+0.018
O+0.016
 aback+0.016
Neg logit contributions
 sizes-0.027
 planets-0.025
 cultures-0.025
hips-0.024
 necks-0.023
Token Jake
Token ID14757
Feature activation+2.29e-02

Pos logit contributions
 sizes+0.013
bows+0.012
 ABC+0.012
 planets+0.012
 toes+0.011
Neg logit contributions
 late-0.010
 severe-0.009
 aback-0.009
 poisonous-0.009
 poisoned-0.009
Token was
Token ID373
Feature activation+1.64e-02

Pos logit contributions
 sizes+0.037
 planets+0.032
bows+0.031
 ABC+0.031
 clubs+0.030
Neg logit contributions
 late-0.043
 aback-0.038
Start-0.037
O-0.037
sm-0.037
Token 3
Token ID513
Feature activation-6.77e-02

Pos logit contributions
 sizes+0.031
 planets+0.027
bows+0.027
 cultures+0.026
 clubs+0.026
Neg logit contributions
 late-0.028
 aback-0.023
Start-0.022
sm-0.022
O-0.022
Token-
Token ID12
Feature activation-1.51e-01

Pos logit contributions
 late+0.117
sm+0.103
Start+0.098
 aback+0.098
O+0.095
Neg logit contributions
 sizes-0.118
bows-0.106
 planets-0.104
hips-0.103
 cultures-0.101
Tokenyears
Token ID19002
Feature activation-7.57e-02

Pos logit contributions
 late+0.211
 aback+0.178
Start+0.171
O+0.169
sm+0.163
Neg logit contributions
 sizes-0.312
 planets-0.280
bows-0.276
 ABC-0.275
 clubs-0.273
Token-
Token ID12
Feature activation-1.26e-01

Pos logit contributions
 late+0.087
 aback+0.068
sm+0.066
Start+0.065
O+0.064
Neg logit contributions
 sizes-0.177
 planets-0.161
bows-0.160
 ABC-0.157
 cultures-0.156
Tokenold
Token ID727
Feature activation-3.79e-02

Pos logit contributions
 late+0.140
 aback+0.109
sm+0.107
Start+0.105
O+0.103
Neg logit contributions
 sizes-0.303
 planets-0.275
bows-0.272
 ABC-0.269
 clubs-0.266
Token and
Token ID290
Feature activation-2.83e-02

Pos logit contributions
 late+0.060
sm+0.052
 aback+0.051
Start+0.046
 my+0.045
Neg logit contributions
 sizes-0.071
hips-0.068
bows-0.065
 planets-0.064
 ABC-0.064
Token the
Token ID262
Feature activation+3.45e-03

Pos logit contributions
 late+0.049
O+0.040
Start+0.040
sm+0.040
 aback+0.039
Neg logit contributions
 sizes-0.050
bows-0.046
 planets-0.045
 clubs-0.044
 necks-0.044
Token dogs
Token ID6844
Feature activation+1.34e-01

Pos logit contributions
 sizes+0.121
 planets+0.110
 ABC+0.107
 clubs+0.105
bows+0.104
Neg logit contributions
 late-0.075
 aback-0.066
Start-0.062
O-0.061
izz-0.060
Token.
Token ID13
Feature activation+1.77e-02

Pos logit contributions
 sizes+0.212
 planets+0.183
bows+0.180
 ABC+0.178
 clubs+0.174
Neg logit contributions
 late-0.255
 aback-0.221
Start-0.217
sm-0.216
O-0.215
Token Do
Token ID2141
Feature activation+4.19e-02

Pos logit contributions
 sizes+0.034
 planets+0.030
bows+0.030
 ABC+0.029
 clubs+0.029
Neg logit contributions
 late-0.028
 aback-0.024
Start-0.023
sm-0.023
O-0.023
Token you
Token ID345
Feature activation-2.00e-02

Pos logit contributions
 sizes+0.118
bows+0.110
pit+0.106
 mast+0.106
 planets+0.105
Neg logit contributions
 late-0.022
O-0.021
 my-0.021
Start-0.016
 another-0.015
Token like
Token ID588
Feature activation+6.01e-02

Pos logit contributions
 late+0.029
 aback+0.024
Start+0.023
sm+0.023
O+0.023
Neg logit contributions
 sizes-0.041
 planets-0.037
bows-0.037
 clubs-0.037
 ABC-0.036
Token dogs
Token ID6844
Feature activation+1.38e-01

Pos logit contributions
 sizes+0.129
 planets+0.117
 ABC+0.114
bows+0.113
 clubs+0.113
Neg logit contributions
 late-0.082
 aback-0.070
Start-0.066
sm-0.065
O-0.065
Token?"
Token ID1701
Feature activation-8.22e-02

Pos logit contributions
 sizes+0.245
bows+0.220
 planets+0.217
 ABC+0.215
 clubs+0.212
Neg logit contributions
 late-0.225
 aback-0.209
Start-0.200
O-0.197
sm-0.196
Token Tom
Token ID4186
Feature activation+1.20e-02

Pos logit contributions
 late+0.149
 aback+0.142
sm+0.139
Start+0.133
 severe+0.133
Neg logit contributions
 sizes-0.130
 planets-0.114
 ABC-0.110
 clubs-0.108
 nations-0.106
Token asks
Token ID7893
Feature activation-3.19e-02

Pos logit contributions
 sizes+0.026
bows+0.023
 clubs+0.023
 planets+0.023
hips+0.023
Neg logit contributions
 late-0.016
 aback-0.014
Start-0.013
O-0.012
sm-0.012
Token.
Token ID13
Feature activation+8.38e-03

Pos logit contributions
 late+0.057
 aback+0.048
sm+0.047
O+0.046
Start+0.046
Neg logit contributions
 sizes-0.057
bows-0.050
 planets-0.049
 cultures-0.048
 ABC-0.047
Token
Token ID198
Feature activation-1.70e-02

Pos logit contributions
 sizes+0.011
 ABC+0.010
 planets+0.010
 clubs+0.010
 neck+0.009
Neg logit contributions
 late-0.015
 poisonous-0.015
 aback-0.015
sm-0.014
 severe-0.014
Token together
Token ID1978
Feature activation+5.33e-02

Pos logit contributions
 sizes+0.064
 cultures+0.055
 ABC+0.055
 planets+0.053
 clubs+0.053
Neg logit contributions
 late-0.058
sm-0.047
 aback-0.047
Start-0.044
 another-0.043
Token.
Token ID13
Feature activation+1.68e-02

Pos logit contributions
 sizes+0.090
 planets+0.076
bows+0.076
 clubs+0.075
 cultures+0.075
Neg logit contributions
 late-0.097
 aback-0.082
sm-0.081
Start-0.080
O-0.079
Token Lily
Token ID20037
Feature activation-1.17e-02

Pos logit contributions
 sizes+0.032
 planets+0.028
bows+0.027
 ABC+0.027
 cultures+0.027
Neg logit contributions
 late-0.028
 aback-0.023
Start-0.023
O-0.023
sm-0.022
Token played
Token ID2826
Feature activation+7.07e-02

Pos logit contributions
 late+0.018
 aback+0.015
Start+0.015
O+0.015
sm+0.015
Neg logit contributions
 sizes-0.022
hips-0.020
 cultures-0.019
bows-0.019
 clubs-0.019
Token her
Token ID607
Feature activation+2.05e-02

Pos logit contributions
 sizes+0.135
 planets+0.118
bows+0.117
 ABC+0.115
 cultures+0.114
Neg logit contributions
 late-0.114
 aback-0.096
Start-0.094
O-0.093
sm-0.092
Token songs
Token ID7259
Feature activation+1.39e-01

Pos logit contributions
 sizes+0.036
hips+0.029
 cultures+0.028
 clubs+0.028
 nations+0.028
Neg logit contributions
 late-0.033
 my-0.029
 another-0.029
sm-0.028
Start-0.027
Token as
Token ID355
Feature activation+1.84e-02

Pos logit contributions
 sizes+0.241
 cultures+0.208
 planets+0.205
 ABC+0.203
 clubs+0.202
Neg logit contributions
 late-0.249
 aback-0.210
sm-0.207
Start-0.201
O-0.198
Token best
Token ID1266
Feature activation+4.22e-02

Pos logit contributions
 sizes+0.040
 clubs+0.036
 necks+0.035
 cultures+0.035
 cheeks+0.035
Neg logit contributions
 late-0.026
sm-0.020
O-0.019
Start-0.018
 severe-0.017
Token as
Token ID355
Feature activation+1.22e-02

Pos logit contributions
 sizes+0.078
bows+0.068
 planets+0.068
 ABC+0.067
 cultures+0.066
Neg logit contributions
 late-0.070
 aback-0.059
sm-0.058
Start-0.058
O-0.058
Token she
Token ID673
Feature activation-3.89e-02

Pos logit contributions
 sizes+0.027
 necks+0.025
 clubs+0.025
 cultures+0.024
 languages+0.024
Neg logit contributions
 late-0.016
sm-0.012
O-0.011
 another-0.011
 my-0.010
Token could
Token ID714
Feature activation+8.55e-02

Pos logit contributions
 late+0.071
 aback+0.062
Start+0.061
sm+0.061
O+0.061
Neg logit contributions
 sizes-0.065
 planets-0.055
bows-0.055
 ABC-0.054
 cultures-0.053
Token of
Token ID286
Feature activation-2.65e-02

Pos logit contributions
 sizes+0.005
 ABC+0.005
 planets+0.005
bows+0.005
 cultures+0.005
Neg logit contributions
 late-0.001
sm-0.001
 aback-0.001
Start-0.001
 poisonous-0.001
Token colors
Token ID7577
Feature activation+1.06e-01

Pos logit contributions
 late+0.046
 aback+0.040
sm+0.040
Start+0.040
O+0.039
Neg logit contributions
 sizes-0.045
bows-0.040
 planets-0.039
 ABC-0.039
 clubs-0.038
Token,
Token ID11
Feature activation+5.26e-02

Pos logit contributions
 sizes+0.181
 ABC+0.180
 cultures+0.180
 hobbies+0.179
 habitats+0.179
Neg logit contributions
 late-0.156
sm-0.132
 aback-0.129
 another-0.119
Start-0.118
Token shapes
Token ID15268
Feature activation+1.03e-01

Pos logit contributions
 sizes+0.072
hips+0.070
 ABC+0.068
utes+0.067
bows+0.066
Neg logit contributions
 late-0.096
 aback-0.084
sm-0.081
Start-0.080
 poisonous-0.077
Token and
Token ID290
Feature activation-5.53e-04

Pos logit contributions
 ABC+0.182
 sizes+0.182
 cultures+0.180
 habitats+0.179
 hobbies+0.177
Neg logit contributions
 late-0.142
sm-0.129
 aback-0.120
 another-0.115
 here-0.112
Token sizes
Token ID10620
Feature activation+1.48e-01

Pos logit contributions
 late+0.001
 aback+0.001
sm+0.001
Start+0.001
 another+0.001
Neg logit contributions
hips-0.001
utes-0.001
bles-0.001
azes-0.001
 ABC-0.001
Token.
Token ID13
Feature activation+1.36e-02

Pos logit contributions
 ABC+0.264
 sizes+0.258
 cultures+0.255
 planets+0.253
 hobbies+0.252
Neg logit contributions
 late-0.203
sm-0.183
 aback-0.176
Start-0.160
 another-0.156
Token She
Token ID1375
Feature activation+2.62e-03

Pos logit contributions
 sizes+0.026
 planets+0.023
bows+0.023
 ABC+0.023
 cultures+0.022
Neg logit contributions
 late-0.021
 aback-0.018
sm-0.018
O-0.018
Start-0.018
Token couldn
Token ID3521
Feature activation+1.08e-02

Pos logit contributions
 sizes+0.004
bows+0.004
 planets+0.004
 ABC+0.004
 cultures+0.004
Neg logit contributions
 late-0.005
 aback-0.004
O-0.004
Start-0.004
sm-0.004
Token't
Token ID470
Feature activation-2.69e-02

Pos logit contributions
 ABC+0.012
bows+0.011
 sizes+0.011
 racing+0.010
hips+0.010
Neg logit contributions
 late-0.023
 aback-0.022
sm-0.021
 poisonous-0.021
 poisoned-0.020
Token believe
Token ID1975
Feature activation+3.41e-02

Pos logit contributions
 late+0.049
 aback+0.043
Start+0.041
sm+0.041
O+0.041
Neg logit contributions
 sizes-0.045
 planets-0.039
bows-0.039
 ABC-0.038
 toes-0.037
Token She
Token ID1375
Feature activation-4.82e-03

Pos logit contributions
 sizes+0.023
 planets+0.021
bows+0.020
 ABC+0.020
 clubs+0.020
Neg logit contributions
 late-0.016
 aback-0.013
Start-0.013
O-0.013
sm-0.013
Token tells
Token ID4952
Feature activation+1.26e-02

Pos logit contributions
 late+0.008
 aback+0.007
Start+0.007
sm+0.007
O+0.007
Neg logit contributions
 sizes-0.008
bows-0.007
 planets-0.007
 ABC-0.007
 clubs-0.007
Token her
Token ID607
Feature activation-8.95e-03

Pos logit contributions
 sizes+0.027
 planets+0.024
bows+0.024
 ABC+0.024
 clubs+0.023
Neg logit contributions
 late-0.017
 aback-0.014
Start-0.014
sm-0.014
O-0.014
Token friends
Token ID2460
Feature activation+8.94e-02

Pos logit contributions
 late+0.017
 aback+0.015
Start+0.015
sm+0.015
O+0.015
Neg logit contributions
 sizes-0.014
 planets-0.012
bows-0.012
 ABC-0.012
 clubs-0.011
Token the
Token ID262
Feature activation+2.91e-02

Pos logit contributions
 sizes+0.181
 planets+0.160
bows+0.159
 ABC+0.157
 clubs+0.155
Neg logit contributions
 late-0.129
 aback-0.108
sm-0.106
Start-0.105
O-0.102
Token names
Token ID3891
Feature activation+1.33e-01

Pos logit contributions
 sizes+0.046
 planets+0.039
bows+0.037
 toes+0.037
 ABC+0.037
Neg logit contributions
 late-0.054
Start-0.046
 aback-0.046
sm-0.046
O-0.045
Token of
Token ID286
Feature activation-1.85e-03

Pos logit contributions
 sizes+0.256
 planets+0.227
bows+0.225
 ABC+0.222
 clubs+0.219
Neg logit contributions
 late-0.209
 aback-0.179
Start-0.174
O-0.172
sm-0.169
Token the
Token ID262
Feature activation+3.49e-02

Pos logit contributions
 late+0.002
 aback+0.002
Start+0.002
O+0.002
sm+0.002
Neg logit contributions
 sizes-0.004
 planets-0.004
bows-0.004
 ABC-0.004
 clubs-0.004
Token flowers
Token ID12734
Feature activation+6.72e-02

Pos logit contributions
 sizes+0.053
 planets+0.044
 ABC+0.044
hips+0.044
 toes+0.043
Neg logit contributions
 late-0.064
sm-0.056
Start-0.054
 aback-0.053
O-0.053
Token and
Token ID290
Feature activation-1.02e-02

Pos logit contributions
 sizes+0.108
 planets+0.093
 ABC+0.092
bows+0.091
 clubs+0.090
Neg logit contributions
 late-0.123
sm-0.106
 aback-0.105
Start-0.103
O-0.101
Token what
Token ID644
Feature activation+2.48e-03

Pos logit contributions
 late+0.015
 aback+0.013
Start+0.012
sm+0.012
O+0.012
Neg logit contributions
 sizes-0.021
 planets-0.018
bows-0.018
 ABC-0.018
 clubs-0.018
Token dolls
Token ID36062
Feature activation+4.37e-02

Pos logit contributions
 late+0.095
sm+0.086
 aback+0.085
Start+0.084
O+0.084
Neg logit contributions
 sizes-0.046
bows-0.040
 planets-0.038
 ABC-0.037
hips-0.037
Token of
Token ID286
Feature activation-2.42e-02

Pos logit contributions
 sizes+0.076
 planets+0.069
 ABC+0.068
 cultures+0.067
 mast+0.066
Neg logit contributions
 late-0.074
sm-0.063
 aback-0.061
Start-0.056
izz-0.056
Token different
Token ID1180
Feature activation+1.47e-02

Pos logit contributions
 late+0.038
 aback+0.032
Start+0.031
O+0.031
sm+0.031
Neg logit contributions
 sizes-0.046
 planets-0.041
bows-0.041
 ABC-0.040
 clubs-0.040
Token colors
Token ID7577
Feature activation+1.16e-01

Pos logit contributions
 sizes+0.012
 mast+0.011
hips+0.011
 planets+0.011
 clubs+0.010
Neg logit contributions
 late-0.034
 aback-0.032
sm-0.032
Start-0.031
O-0.030
Token and
Token ID290
Feature activation+4.48e-03

Pos logit contributions
 sizes+0.206
 cultures+0.203
 mast+0.197
 ABC+0.195
 planets+0.194
Neg logit contributions
 late-0.173
sm-0.142
 aback-0.136
 another-0.130
O-0.128
Token sizes
Token ID10620
Feature activation+1.53e-01

Pos logit contributions
hips+0.003
utes+0.003
 mast+0.003
castle+0.003
pit+0.003
Neg logit contributions
 late-0.010
 my-0.009
sm-0.009
 medicine-0.009
O-0.009
Token.
Token ID13
Feature activation+1.55e-02

Pos logit contributions
 sizes+0.284
 ABC+0.276
 planets+0.275
 cultures+0.274
 mast+0.268
Neg logit contributions
 late-0.211
sm-0.185
 aback-0.170
Start-0.158
 another-0.157
Token She
Token ID1375
Feature activation+1.45e-03

Pos logit contributions
 sizes+0.028
 ABC+0.025
 planets+0.025
bows+0.024
 toes+0.024
Neg logit contributions
 late-0.026
 aback-0.023
 severe-0.023
sm-0.023
Start-0.022
Token was
Token ID373
Feature activation+3.33e-02

Pos logit contributions
 sizes+0.002
bows+0.002
 planets+0.002
 cultures+0.002
 ABC+0.002
Neg logit contributions
 late-0.003
O-0.002
 aback-0.002
sm-0.002
Start-0.002
Token very
Token ID845
Feature activation-4.61e-02

Pos logit contributions
 sizes+0.061
 planets+0.053
bows+0.052
 ABC+0.052
 toes+0.050
Neg logit contributions
 late-0.055
 aback-0.049
Start-0.048
O-0.047
sm-0.047
Token organized
Token ID8389
Feature activation-1.14e-02

Pos logit contributions
 late+0.077
 aback+0.075
Start+0.070
O+0.069
 severe+0.067
Neg logit contributions
 sizes-0.080
 toes-0.069
 ABC-0.069
bows-0.068
 planets-0.067
Token.
Token ID13
Feature activation+9.36e-03

Pos logit contributions
 sizes+0.031
 planets+0.029
 cultures+0.029
 nations+0.028
 ABC+0.028
Neg logit contributions
 late-0.028
 aback-0.023
sm-0.021
Start-0.020
 severe-0.020
Token "
Token ID366
Feature activation-1.51e-02

Pos logit contributions
 sizes+0.017
bows+0.015
 planets+0.015
 ABC+0.014
 clubs+0.014
Neg logit contributions
 late-0.015
 aback-0.013
sm-0.013
Start-0.013
 severe-0.013
TokenYou
Token ID1639
Feature activation+2.23e-02

Pos logit contributions
 aback+0.022
 poisonous+0.020
 late+0.019
 poisoned+0.019
 severe+0.018
Neg logit contributions
 sizes-0.027
 necks-0.024
 ABC-0.023
 clubs-0.023
 neck-0.023
Token have
Token ID423
Feature activation-1.30e-02

Pos logit contributions
 sizes+0.032
 clubs+0.027
 planets+0.027
hips+0.026
bows+0.026
Neg logit contributions
 late-0.047
sm-0.041
 aback-0.040
O-0.040
Start-0.039
Token the
Token ID262
Feature activation+2.46e-02

Pos logit contributions
 late+0.020
izz+0.018
sm+0.018
 aback+0.018
O+0.017
Neg logit contributions
 sizes-0.025
 clubs-0.022
 planets-0.022
 necks-0.021
 ABC-0.021
Token blue
Token ID4171
Feature activation+6.38e-02

Pos logit contributions
 sizes+0.037
bows+0.034
 cultures+0.032
 ABC+0.032
 planets+0.032
Neg logit contributions
 late-0.048
 aback-0.041
Start-0.039
O-0.039
 poisoned-0.038
Token boat
Token ID8848
Feature activation-2.13e-02

Pos logit contributions
 sizes+0.105
bows+0.094
 cultures+0.089
 ABC+0.088
 planets+0.086
Neg logit contributions
 late-0.113
 aback-0.098
O-0.096
sm-0.095
Start-0.094
Token!"
Token ID2474
Feature activation-4.03e-02

Pos logit contributions
 late+0.039
O+0.036
Start+0.035
izz+0.035
 aback+0.035
Neg logit contributions
 sizes-0.035
bows-0.030
 clubs-0.028
 planets-0.028
hips-0.028
Token
Token ID198
Feature activation-1.54e-02

Pos logit contributions
 late+0.063
 aback+0.055
sm+0.054
Start+0.053
 severe+0.052
Neg logit contributions
 sizes-0.075
 planets-0.066
 ABC-0.065
bows-0.065
 clubs-0.064
Token
Token ID198
Feature activation-1.85e-02

Pos logit contributions
 late+0.026
 aback+0.024
 poisoned+0.023
 severe+0.023
sm+0.023
Neg logit contributions
 sizes-0.024
 ABC-0.022
 planets-0.021
 clubs-0.021
hips-0.021
Token"
Token ID1
Feature activation-4.76e-02

Pos logit contributions
 late+0.029
 aback+0.026
sm+0.025
Start+0.024
 poisoned+0.024
Neg logit contributions
 sizes-0.034
 clubs-0.030
 planets-0.029
 necks-0.029
bows-0.028
Token and
Token ID290
Feature activation-1.54e-02

Pos logit contributions
 sizes+0.084
 planets+0.073
 ABC+0.073
bows+0.072
 necks+0.071
Neg logit contributions
 late-0.075
sm-0.065
 aback-0.062
Start-0.060
izz-0.057
Token the
Token ID262
Feature activation+1.99e-02

Pos logit contributions
 late+0.024
 aback+0.020
sm+0.020
Start+0.019
O+0.019
Neg logit contributions
 sizes-0.029
bows-0.026
 ABC-0.026
 planets-0.026
hips-0.025
Token cars
Token ID5006
Feature activation+4.22e-02

Pos logit contributions
 sizes+0.036
bows+0.032
 planets+0.031
 cultures+0.031
hips+0.030
Neg logit contributions
 late-0.034
 aback-0.027
Start-0.027
O-0.027
sm-0.027
Token and
Token ID290
Feature activation-1.20e-02

Pos logit contributions
 sizes+0.072
 ABC+0.064
 planets+0.063
 cultures+0.063
 languages+0.061
Neg logit contributions
 late-0.072
sm-0.062
 aback-0.060
Start-0.059
 poisoned-0.058
Token the
Token ID262
Feature activation+2.31e-02

Pos logit contributions
 late+0.014
 aback+0.011
Start+0.011
sm+0.011
O+0.011
Neg logit contributions
 sizes-0.028
 planets-0.025
bows-0.025
 ABC-0.025
 clubs-0.024
Token trees
Token ID7150
Feature activation+8.13e-02

Pos logit contributions
 sizes+0.040
bows+0.036
 cultures+0.035
 planets+0.034
hips+0.034
Neg logit contributions
 late-0.040
 aback-0.032
Start-0.032
O-0.032
sm-0.032
Token.
Token ID13
Feature activation+1.21e-02

Pos logit contributions
 sizes+0.147
 ABC+0.137
 planets+0.134
 cultures+0.133
 hobbies+0.131
Neg logit contributions
 late-0.131
sm-0.111
 aback-0.103
 poisonous-0.100
Start-0.100
Token The
Token ID383
Feature activation+2.71e-02

Pos logit contributions
 sizes+0.024
bows+0.021
 planets+0.021
 clubs+0.020
 ABC+0.020
Neg logit contributions
 late-0.018
sm-0.016
 aback-0.016
 severe-0.016
O-0.015
Token hot
Token ID3024
Feature activation-2.99e-02

Pos logit contributions
 sizes+0.048
bows+0.042
 planets+0.041
 cultures+0.040
 clubs+0.040
Neg logit contributions
 late-0.047
 aback-0.039
Start-0.039
sm-0.039
O-0.039
Token red
Token ID2266
Feature activation+1.49e-02

Pos logit contributions
 late+0.057
 aback+0.049
O+0.048
sm+0.048
Start+0.048
Neg logit contributions
 sizes-0.047
 planets-0.040
bows-0.040
 ABC-0.039
 cultures-0.039
Token balloon
Token ID21190
Feature activation-2.04e-02

Pos logit contributions
 sizes+0.024
 planets+0.021
bows+0.021
 ABC+0.021
 clubs+0.020
Neg logit contributions
 late-0.028
 aback-0.025
Start-0.024
sm-0.024
O-0.024
Token holes
Token ID10421
Feature activation+2.03e-02

Pos logit contributions
 late+0.022
 aback+0.020
izz+0.019
sm+0.019
Start+0.018
Neg logit contributions
 sizes-0.025
 planets-0.022
 clubs-0.022
 neck-0.022
 toes-0.021
Token and
Token ID290
Feature activation-1.52e-02

Pos logit contributions
 sizes+0.033
 planets+0.028
bows+0.028
 toes+0.027
 clubs+0.027
Neg logit contributions
 late-0.038
 aback-0.034
Start-0.034
O-0.034
sm-0.033
Token rust
Token ID17000
Feature activation+2.69e-02

Pos logit contributions
 late+0.031
 aback+0.028
sm+0.027
Start+0.027
izz+0.027
Neg logit contributions
 sizes-0.023
 planets-0.020
 toes-0.019
 necks-0.019
 clubs-0.019
Token.
Token ID13
Feature activation+4.81e-03

Pos logit contributions
 sizes+0.047
 planets+0.039
bows+0.038
 toes+0.037
 clubs+0.036
Neg logit contributions
 late-0.050
 aback-0.046
Start-0.045
O-0.043
 severe-0.042
Token It
Token ID632
Feature activation-1.11e-02

Pos logit contributions
 sizes+0.009
 planets+0.008
 ABC+0.008
bows+0.008
 clubs+0.007
Neg logit contributions
 late-0.008
 aback-0.007
sm-0.007
Start-0.007
 severe-0.006
Token can
Token ID460
Feature activation+6.71e-02

Pos logit contributions
 late+0.025
 aback+0.023
O+0.023
 severe+0.023
sm+0.022
Neg logit contributions
 sizes-0.014
 planets-0.012
bows-0.011
 necks-0.011
 ABC-0.010
Token hurt
Token ID5938
Feature activation+1.91e-02

Pos logit contributions
 sizes+0.081
hips+0.078
 neck+0.073
 planets+0.073
 toes+0.072
Neg logit contributions
 aback-0.145
 late-0.141
sm-0.137
Start-0.130
handed-0.121
Token you
Token ID345
Feature activation-2.53e-02

Pos logit contributions
 sizes+0.039
 planets+0.035
 necks+0.034
 clubs+0.033
 toes+0.033
Neg logit contributions
 late-0.029
 aback-0.025
Start-0.024
 severe-0.023
sm-0.023
Token,"
Token ID553
Feature activation-6.21e-02

Pos logit contributions
 late+0.044
 aback+0.040
Start+0.040
O+0.039
sm+0.039
Neg logit contributions
 sizes-0.043
bows-0.037
 planets-0.036
 ABC-0.036
 toes-0.036
Token Tom
Token ID4186
Feature activation+1.52e-02

Pos logit contributions
 late+0.107
 aback+0.095
sm+0.091
Start+0.091
O+0.088
Neg logit contributions
 sizes-0.109
 planets-0.096
 ABC-0.095
bows-0.094
 toes-0.093
Token says
Token ID1139
Feature activation+1.67e-02

Pos logit contributions
 sizes+0.027
 planets+0.023
bows+0.023
 ABC+0.023
 clubs+0.023
Neg logit contributions
 late-0.026
 aback-0.023
Start-0.022
sm-0.022
O-0.022
Token to
Token ID284
Feature activation+7.96e-03

Pos logit contributions
 sizes+0.124
bows+0.115
 planets+0.112
 ABC+0.111
 cultures+0.110
Neg logit contributions
 late-0.065
sm-0.048
 aback-0.048
O-0.047
Start-0.046
Token make
Token ID787
Feature activation+3.27e-02

Pos logit contributions
 sizes+0.012
 planets+0.010
bows+0.010
 ABC+0.009
 clubs+0.009
Neg logit contributions
 late-0.016
 aback-0.014
Start-0.014
O-0.014
sm-0.014
Token up
Token ID510
Feature activation-3.36e-02

Pos logit contributions
 sizes+0.061
bows+0.054
 planets+0.053
 cultures+0.053
 ABC+0.053
Neg logit contributions
 late-0.052
 aback-0.044
Start-0.044
sm-0.042
O-0.042
Token stories
Token ID3923
Feature activation+1.17e-01

Pos logit contributions
 late+0.072
sm+0.062
Start+0.061
 aback+0.060
 severe+0.059
Neg logit contributions
 sizes-0.044
 planets-0.036
bows-0.036
 ABC-0.036
 necks-0.035
Token with
Token ID351
Feature activation+3.40e-03

Pos logit contributions
 sizes+0.212
 cultures+0.193
 necks+0.191
 planets+0.189
 toes+0.187
Neg logit contributions
 late-0.196
 aback-0.156
sm-0.149
Start-0.147
 another-0.143
Token their
Token ID511
Feature activation+5.29e-02

Pos logit contributions
 sizes+0.007
bows+0.007
 ABC+0.007
 cultures+0.007
 planets+0.007
Neg logit contributions
 late-0.004
sm-0.003
O-0.003
 aback-0.003
Start-0.003
Token dolls
Token ID36062
Feature activation+7.10e-02

Pos logit contributions
 sizes+0.080
bows+0.076
 cultures+0.072
hips+0.070
iece+0.070
Neg logit contributions
 late-0.087
sm-0.076
 my-0.075
 poisonous-0.075
 medicine-0.073
Token and
Token ID290
Feature activation+4.61e-03

Pos logit contributions
 sizes+0.117
 planets+0.103
 ABC+0.102
 cultures+0.101
bows+0.099
Neg logit contributions
 late-0.129
 aback-0.107
sm-0.107
 severe-0.103
Start-0.103
Token cars
Token ID5006
Feature activation+6.32e-02

Pos logit contributions
bows+0.007
 sizes+0.007
hips+0.007
 ABC+0.007
igo+0.006
Neg logit contributions
 late-0.007
 my-0.007
 heaven-0.006
 another-0.006
 severe-0.006
Token.
Token ID13
Feature activation+1.51e-02

Pos logit contributions
 sizes+0.105
 planets+0.092
 ABC+0.091
 cultures+0.090
bows+0.088
Neg logit contributions
 late-0.115
 aback-0.097
sm-0.096
Start-0.094
O-0.093
Token Lily
Token ID20037
Feature activation-1.22e-02

Pos logit contributions
 sizes+0.030
 planets+0.027
bows+0.026
 ABC+0.026
 clubs+0.026
Neg logit contributions
 late-0.023
 aback-0.020
Start-0.019
sm-0.019
O-0.019
Token.
Token ID13
Feature activation+1.98e-02

Pos logit contributions
 sizes+0.100
 planets+0.091
 ABC+0.091
 cultures+0.090
bows+0.089
Neg logit contributions
 late-0.087
sm-0.068
 aback-0.068
 another-0.064
 severe-0.063
Token They
Token ID1119
Feature activation+6.03e-02

Pos logit contributions
 sizes+0.032
 planets+0.028
bows+0.027
 ABC+0.027
 cultures+0.027
Neg logit contributions
 late-0.038
 aback-0.032
Start-0.031
O-0.031
sm-0.031
Token wait
Token ID4043
Feature activation-3.08e-03

Pos logit contributions
 sizes+0.100
 planets+0.086
bows+0.085
 ABC+0.084
 cultures+0.083
Neg logit contributions
 late-0.109
 aback-0.095
Start-0.093
sm-0.092
O-0.092
Token for
Token ID329
Feature activation-8.83e-02

Pos logit contributions
 late+0.005
sm+0.004
 aback+0.004
Start+0.003
O+0.003
Neg logit contributions
 sizes-0.006
 planets-0.006
bows-0.006
 cultures-0.006
 clubs-0.006
Token the
Token ID262
Feature activation+2.21e-02

Pos logit contributions
 late+0.113
 aback+0.090
Start+0.088
O+0.088
sm+0.087
Neg logit contributions
 sizes-0.197
bows-0.177
 planets-0.177
 ABC-0.174
 cultures-0.172
Token music
Token ID2647
Feature activation+7.75e-02

Pos logit contributions
 sizes+0.034
bows+0.030
 planets+0.030
 clubs+0.029
 cultures+0.029
Neg logit contributions
 late-0.042
O-0.036
 aback-0.035
sm-0.035
Start-0.035
Token to
Token ID284
Feature activation-8.47e-03

Pos logit contributions
 sizes+0.132
 planets+0.116
 ABC+0.113
bows+0.113
 cultures+0.111
Neg logit contributions
 late-0.138
 aback-0.118
sm-0.116
Start-0.115
O-0.113
Token start
Token ID923
Feature activation+1.27e-02

Pos logit contributions
 late+0.016
 aback+0.014
sm+0.014
Start+0.014
O+0.013
Neg logit contributions
 sizes-0.014
 planets-0.012
 ABC-0.012
bows-0.012
 clubs-0.012
Token.
Token ID13
Feature activation+1.42e-02

Pos logit contributions
 sizes+0.024
 planets+0.021
bows+0.021
 cultures+0.020
hips+0.020
Neg logit contributions
 late-0.021
sm-0.017
 aback-0.016
 another-0.015
Start-0.015
Token They
Token ID1119
Feature activation+5.75e-02

Pos logit contributions
 sizes+0.023
 planets+0.020
bows+0.020
 ABC+0.019
 cultures+0.019
Neg logit contributions
 late-0.027
 aback-0.023
Start-0.023
O-0.023
sm-0.022
Token are
Token ID389
Feature activation+3.49e-02

Pos logit contributions
 sizes+0.094
 planets+0.082
bows+0.081
 ABC+0.080
 clubs+0.078
Neg logit contributions
 late-0.107
 aback-0.093
Start-0.091
sm-0.090
O-0.090
Token touch
Token ID3638
Feature activation-1.01e-02

Pos logit contributions
 late+0.047
 severe+0.041
sm+0.037
 poisoned+0.037
 infected+0.036
Neg logit contributions
 sizes-0.029
 patterns-0.027
 clubs-0.027
 toes-0.026
 planets-0.026
Token them
Token ID606
Feature activation+2.51e-03

Pos logit contributions
 late+0.013
 aback+0.011
Start+0.010
O+0.010
sm+0.010
Neg logit contributions
 sizes-0.022
 planets-0.020
bows-0.020
 ABC-0.020
 clubs-0.020
Token,
Token ID11
Feature activation+5.74e-02

Pos logit contributions
 sizes+0.005
 cultures+0.004
 planets+0.004
 ABC+0.004
 habitats+0.004
Neg logit contributions
 late-0.004
 aback-0.003
 another-0.003
Start-0.003
sm-0.003
Token feeling
Token ID4203
Feature activation-6.41e-02

Pos logit contributions
 sizes+0.083
 planets+0.071
bows+0.070
 ABC+0.069
 cultures+0.067
Neg logit contributions
 late-0.118
 aback-0.104
Start-0.102
sm-0.101
O-0.100
Token the
Token ID262
Feature activation+1.29e-02

Pos logit contributions
 late+0.099
O+0.086
plug+0.084
Start+0.084
 aback+0.083
Neg logit contributions
 sizes-0.122
 planets-0.104
 toes-0.102
 patterns-0.102
 neck-0.099
Token soft
Token ID2705
Feature activation+2.96e-02

Pos logit contributions
 sizes+0.025
 ABC+0.024
 cultures+0.024
 planets+0.024
azes+0.023
Neg logit contributions
 late-0.018
 aback-0.015
 poisonous-0.015
 poisoned-0.014
O-0.014
Token pet
Token ID4273
Feature activation+1.75e-02

Pos logit contributions
hips+0.040
 ABC+0.039
 sizes+0.039
azes+0.039
 nations+0.039
Neg logit contributions
 late-0.052
 aback-0.048
 poisonous-0.047
sm-0.046
O-0.045
Tokenals
Token ID874
Feature activation-4.66e-03

Pos logit contributions
 sizes+0.031
 planets+0.028
 ABC+0.027
 cultures+0.027
bows+0.027
Neg logit contributions
 late-0.029
 aback-0.025
sm-0.025
O-0.025
Start-0.024
Token.
Token ID13
Feature activation+1.54e-02

Pos logit contributions
 late+0.008
 aback+0.007
sm+0.007
Start+0.007
O+0.007
Neg logit contributions
 sizes-0.008
 ABC-0.007
 planets-0.007
 cultures-0.007
bows-0.007
Token It
Token ID632
Feature activation-1.39e-02

Pos logit contributions
 sizes+0.025
 planets+0.021
 ABC+0.021
bows+0.021
 clubs+0.021
Neg logit contributions
 late-0.028
sm-0.025
 severe-0.025
 aback-0.025
O-0.024
Token was
Token ID373
Feature activation+2.64e-02

Pos logit contributions
 late+0.032
 aback+0.027
 severe+0.026
sm+0.026
Aid+0.026
Neg logit contributions
 sizes-0.019
 planets-0.015
 patterns-0.015
 toes-0.014
bows-0.014
Token toy
Token ID13373
Feature activation-6.45e-03

Pos logit contributions
 late+0.015
 aback+0.012
sm+0.012
Start+0.011
O+0.011
Neg logit contributions
 sizes-0.017
 planets-0.015
bows-0.015
 cultures-0.015
hips-0.015
Token grill
Token ID29901
Feature activation+1.33e-02

Pos logit contributions
 late+0.010
sm+0.008
 poisoned+0.008
O+0.008
 poisonous+0.008
Neg logit contributions
 sizes-0.012
bows-0.011
 cultures-0.011
 ABC-0.010
 languages-0.010
Token that
Token ID326
Feature activation-3.29e-02

Pos logit contributions
 sizes+0.029
 cultures+0.027
bows+0.027
 planets+0.027
 ABC+0.026
Neg logit contributions
 late-0.017
sm-0.013
 aback-0.012
 poisoned-0.012
O-0.012
Token he
Token ID339
Feature activation-1.13e-02

Pos logit contributions
 late+0.064
 aback+0.060
Start+0.059
 severe+0.056
sm+0.056
Neg logit contributions
 sizes-0.052
 planets-0.045
 ABC-0.043
bows-0.042
 toes-0.041
Token played
Token ID2826
Feature activation+7.90e-02

Pos logit contributions
 late+0.025
 aback+0.024
Start+0.023
 severe+0.022
O+0.022
Neg logit contributions
 sizes-0.015
 planets-0.012
 ABC-0.011
 necks-0.011
 neck-0.011
Token with
Token ID351
Feature activation+7.85e-03

Pos logit contributions
 sizes+0.140
 planets+0.124
bows+0.122
 ABC+0.120
 clubs+0.118
Neg logit contributions
 late-0.134
 aback-0.118
Start-0.114
O-0.113
sm-0.113
Token every
Token ID790
Feature activation+2.90e-02

Pos logit contributions
 sizes+0.013
 planets+0.011
bows+0.011
 ABC+0.011
 clubs+0.011
Neg logit contributions
 late-0.014
 aback-0.012
Start-0.012
sm-0.012
O-0.012
Token day
Token ID1110
Feature activation+2.50e-02

Pos logit contributions
 sizes+0.028
 toes+0.023
 ABC+0.022
 planets+0.022
 pretending+0.020
Neg logit contributions
 aback-0.069
 late-0.068
Start-0.067
izz-0.065
O-0.065
Token.
Token ID13
Feature activation+1.59e-02

Pos logit contributions
 sizes+0.047
 cultures+0.043
bows+0.043
 ABC+0.042
 hobbies+0.041
Neg logit contributions
 late-0.043
sm-0.032
 aback-0.031
Start-0.030
 another-0.030
Token
Token ID198
Feature activation-1.45e-02

Pos logit contributions
 sizes+0.034
 planets+0.031
bows+0.031
 cultures+0.031
 clubs+0.030
Neg logit contributions
 late-0.021
 aback-0.017
O-0.017
Start-0.017
sm-0.016
Token
Token ID198
Feature activation-1.96e-02

Pos logit contributions
 late+0.023
 aback+0.021
Start+0.020
sm+0.019
 severe+0.019
Neg logit contributions
 sizes-0.026
 ABC-0.023
 planets-0.023
bows-0.023
 clubs-0.022
Token Mom
Token ID11254
Feature activation-2.71e-02

Pos logit contributions
 late+0.131
 aback+0.122
Start+0.120
sm+0.120
O+0.114
Neg logit contributions
 sizes-0.184
 planets-0.168
 patterns-0.166
 toes-0.166
 ABC-0.166
Token.
Token ID13
Feature activation+2.47e-03

Pos logit contributions
 late+0.049
 aback+0.043
Start+0.042
sm+0.041
O+0.041
Neg logit contributions
 sizes-0.046
 planets-0.040
bows-0.039
 ABC-0.039
 clubs-0.038
Token He
Token ID679
Feature activation+1.32e-02

Pos logit contributions
 sizes+0.004
 planets+0.003
 ABC+0.003
 toes+0.003
bows+0.003
Neg logit contributions
 late-0.004
sm-0.004
 aback-0.004
izz-0.004
Start-0.004
Token learned
Token ID4499
Feature activation+5.49e-02

Pos logit contributions
 sizes+0.019
 planets+0.017
 neck+0.017
 necks+0.016
 ABC+0.016
Neg logit contributions
 late-0.026
 aback-0.024
sm-0.022
Start-0.022
izz-0.022
Token his
Token ID465
Feature activation+1.35e-02

Pos logit contributions
 sizes+0.088
 ABC+0.085
 planets+0.081
 clubs+0.080
 hobbies+0.080
Neg logit contributions
izz-0.078
 aback-0.076
 late-0.076
ited-0.075
sm-0.073
Token lesson
Token ID11483
Feature activation-5.44e-03

Pos logit contributions
 sizes+0.028
bows+0.025
 planets+0.025
 clubs+0.024
 necks+0.024
Neg logit contributions
 late-0.021
 severe-0.016
O-0.016
Start-0.016
sm-0.016
Token.
Token ID13
Feature activation+4.32e-03

Pos logit contributions
 late+0.010
 aback+0.008
sm+0.008
Start+0.008
 severe+0.008
Neg logit contributions
 sizes-0.010
 planets-0.009
bows-0.009
 toes-0.009
 ABC-0.008
Token
Token ID198
Feature activation-2.82e-02

Pos logit contributions
 sizes+0.007
 planets+0.006
 ABC+0.006
bows+0.006
 toes+0.006
Neg logit contributions
 late-0.007
sm-0.007
 aback-0.007
izz-0.007
O-0.007
Token
Token ID198
Feature activation-3.18e-02

Pos logit contributions
 late+0.045
 aback+0.042
Start+0.040
sm+0.039
 poisoned+0.039
Neg logit contributions
 sizes-0.047
 planets-0.042
 ABC-0.042
 clubs-0.040
hips-0.040
TokenThe
Token ID464
Feature activation+1.94e-02

Pos logit contributions
 late+0.049
 aback+0.048
sm+0.043
 poisoned+0.042
Start+0.042
Neg logit contributions
 sizes-0.055
 planets-0.049
 ABC-0.048
 clubs-0.047
hips-0.047
Token moral
Token ID6573
Feature activation-3.28e-02

Pos logit contributions
 sizes+0.033
bows+0.029
 planets+0.028
 cultures+0.027
 clubs+0.027
Neg logit contributions
 late-0.036
O-0.030
sm-0.030
 aback-0.030
Start-0.030
Token them
Token ID606
Feature activation+3.65e-03

Pos logit contributions
 sizes+0.148
 planets+0.132
 ABC+0.130
bows+0.129
 clubs+0.128
Neg logit contributions
 late-0.115
 aback-0.097
Start-0.095
O-0.094
sm-0.093
Token.
Token ID13
Feature activation+1.27e-02

Pos logit contributions
 sizes+0.006
 planets+0.005
bows+0.005
 ABC+0.005
 clubs+0.005
Neg logit contributions
 late-0.007
 aback-0.006
O-0.006
Start-0.006
sm-0.006
Token They
Token ID1119
Feature activation+5.79e-02

Pos logit contributions
 sizes+0.021
 planets+0.018
bows+0.017
 ABC+0.017
 clubs+0.017
Neg logit contributions
 late-0.023
 aback-0.021
Start-0.020
sm-0.020
 severe-0.020
Token learned
Token ID4499
Feature activation+6.17e-02

Pos logit contributions
 sizes+0.090
 planets+0.081
 clubs+0.077
bows+0.074
 nations+0.074
Neg logit contributions
 late-0.112
 aback-0.099
Start-0.097
O-0.093
sm-0.092
Token that
Token ID326
Feature activation-2.77e-02

Pos logit contributions
 sizes+0.119
 planets+0.109
 clubs+0.106
 ABC+0.105
 cultures+0.105
Neg logit contributions
 late-0.085
izz-0.077
O-0.076
 aback-0.076
sm-0.074
Token it
Token ID340
Feature activation-6.83e-02

Pos logit contributions
 late+0.043
 aback+0.033
sm+0.032
Start+0.032
O+0.032
Neg logit contributions
 sizes-0.056
bows-0.051
 planets-0.049
 clubs-0.048
 ABC-0.048
Token's
Token ID338
Feature activation+5.15e-02

Pos logit contributions
 late+0.129
 aback+0.111
O+0.111
Start+0.110
sm+0.109
Neg logit contributions
 sizes-0.110
 planets-0.096
bows-0.092
 ABC-0.092
 cultures-0.089
Token okay
Token ID8788
Feature activation+4.01e-02

Pos logit contributions
 sizes+0.084
 planets+0.072
 ABC+0.068
 cultures+0.068
bows+0.066
Neg logit contributions
 late-0.086
 aback-0.083
Start-0.082
sm-0.081
izz-0.081
Token to
Token ID284
Feature activation-3.54e-03

Pos logit contributions
 sizes+0.081
 planets+0.072
bows+0.071
 ABC+0.071
 cultures+0.069
Neg logit contributions
 late-0.059
 aback-0.050
Start-0.049
O-0.049
sm-0.048
Token ask
Token ID1265
Feature activation+1.41e-02

Pos logit contributions
 late+0.006
 aback+0.005
Start+0.005
O+0.005
sm+0.005
Neg logit contributions
 sizes-0.006
 planets-0.005
bows-0.005
 ABC-0.005
 clubs-0.005
Token for
Token ID329
Feature activation-7.78e-02

Pos logit contributions
 sizes+0.024
 planets+0.021
 ABC+0.021
 cultures+0.021
 toes+0.020
Neg logit contributions
 late-0.024
O-0.022
 aback-0.021
Start-0.021
izz-0.021
Token.
Token ID13
Feature activation+2.84e-02

Pos logit contributions
 late+0.072
Start+0.063
 aback+0.063
O+0.063
sm+0.061
Neg logit contributions
 sizes-0.062
bows-0.052
 planets-0.052
 toes-0.051
 clubs-0.051
Token He
Token ID679
Feature activation+2.41e-02

Pos logit contributions
 sizes+0.049
 planets+0.043
 ABC+0.042
bows+0.042
 clubs+0.041
Neg logit contributions
 late-0.049
 aback-0.043
sm-0.042
 severe-0.042
Start-0.041
Token stirred
Token ID33091
Feature activation-1.19e-02

Pos logit contributions
 sizes+0.037
 planets+0.032
 neck+0.031
 toes+0.030
 necks+0.030
Neg logit contributions
 late-0.047
 aback-0.042
Start-0.041
O-0.040
sm-0.039
Token and
Token ID290
Feature activation-5.31e-03

Pos logit contributions
 late+0.018
 aback+0.015
Start+0.015
sm+0.015
O+0.015
Neg logit contributions
 sizes-0.024
 planets-0.021
bows-0.020
 ABC-0.020
 toes-0.020
Token stirred
Token ID33091
Feature activation-1.26e-02

Pos logit contributions
 late+0.008
 aback+0.008
Start+0.007
O+0.007
 severe+0.007
Neg logit contributions
 sizes-0.010
 planets-0.009
 patterns-0.008
bows-0.008
 toes-0.008
Token until
Token ID1566
Feature activation+7.84e-03

Pos logit contributions
 late+0.023
 aback+0.020
O+0.020
Start+0.019
sm+0.019
Neg logit contributions
 sizes-0.021
 planets-0.018
bows-0.018
 ABC-0.018
 cultures-0.018
Token everything
Token ID2279
Feature activation-3.26e-02

Pos logit contributions
 sizes+0.012
 toes+0.011
 planets+0.011
 patterns+0.010
 cheeks+0.010
Neg logit contributions
 aback-0.013
 late-0.013
sm-0.013
Start-0.013
 severe-0.012
Token was
Token ID373
Feature activation+1.50e-02

Pos logit contributions
 late+0.070
 aback+0.063
Start+0.063
O+0.063
 severe+0.062
Neg logit contributions
 sizes-0.045
 planets-0.037
bows-0.036
 patterns-0.035
 toes-0.033
Token mixed
Token ID7668
Feature activation-4.12e-02

Pos logit contributions
 sizes+0.028
 planets+0.024
bows+0.023
 toes+0.023
 ABC+0.023
Neg logit contributions
 late-0.024
 aback-0.022
Start-0.021
sm-0.021
O-0.021
Token together
Token ID1978
Feature activation+3.83e-02

Pos logit contributions
 late+0.072
 aback+0.062
Start+0.061
O+0.060
sm+0.060
Neg logit contributions
 sizes-0.072
 planets-0.063
bows-0.062
 ABC-0.061
 clubs-0.060
Token.
Token ID13
Feature activation+2.57e-02

Pos logit contributions
 sizes+0.063
 planets+0.057
bows+0.057
 ABC+0.057
 cultures+0.056
Neg logit contributions
 late-0.069
 aback-0.058
sm-0.058
Start-0.056
O-0.054
Token try
Token ID1949
Feature activation+3.85e-02

Pos logit contributions
 late+0.008
 aback+0.007
sm+0.007
Start+0.007
 severe+0.007
Neg logit contributions
 sizes-0.007
bows-0.006
 planets-0.006
 ABC-0.006
 clubs-0.006
Token too
Token ID1165
Feature activation-2.51e-03

Pos logit contributions
 sizes+0.085
 planets+0.077
 ABC+0.075
bows+0.075
 clubs+0.074
Neg logit contributions
 late-0.050
 aback-0.041
Start-0.039
sm-0.039
O-0.039
Token.
Token ID13
Feature activation+1.01e-02

Pos logit contributions
 late+0.005
 aback+0.005
Start+0.004
O+0.004
sm+0.004
Neg logit contributions
 sizes-0.004
 planets-0.003
bows-0.003
 ABC-0.003
 clubs-0.003
Token He
Token ID679
Feature activation+2.77e-02

Pos logit contributions
 sizes+0.016
 toes+0.013
 ABC+0.013
bows+0.013
 patterns+0.013
Neg logit contributions
 late-0.019
sm-0.017
 severe-0.017
 aback-0.017
 poisonous-0.017
Token looked
Token ID3114
Feature activation-4.29e-03

Pos logit contributions
 sizes+0.046
 planets+0.040
 neck+0.039
 necks+0.039
 toes+0.039
Neg logit contributions
 late-0.052
 aback-0.044
 severe-0.043
sm-0.042
Start-0.042
Token for
Token ID329
Feature activation-8.94e-02

Pos logit contributions
 late+0.008
O+0.007
 severe+0.007
izz+0.007
Start+0.007
Neg logit contributions
 sizes-0.007
 toes-0.006
 planets-0.006
 patterns-0.006
 clubs-0.006
Token another
Token ID1194
Feature activation-2.29e-02

Pos logit contributions
 late+0.136
 severe+0.108
 aback+0.105
Start+0.105
plug+0.104
Neg logit contributions
 sizes-0.179
 patterns-0.166
 toes-0.165
 planets-0.165
 neck-0.161
Token vine
Token ID17793
Feature activation+4.12e-03

Pos logit contributions
 late+0.037
 aback+0.032
Start+0.031
sm+0.031
O+0.030
Neg logit contributions
 sizes-0.042
 planets-0.038
 ABC-0.037
bows-0.037
 cultures-0.037
Token.
Token ID13
Feature activation+3.28e-03

Pos logit contributions
 sizes+0.007
 ABC+0.007
 cultures+0.007
 planets+0.007
bows+0.007
Neg logit contributions
 late-0.007
 aback-0.006
sm-0.006
Start-0.005
 poisoned-0.005
Token He
Token ID679
Feature activation+1.97e-02

Pos logit contributions
 sizes+0.005
 toes+0.004
 ABC+0.004
bows+0.004
 planets+0.004
Neg logit contributions
 late-0.006
sm-0.005
 severe-0.005
 aback-0.005
 poisonous-0.005
Token saw
Token ID2497
Feature activation-4.59e-02

Pos logit contributions
 sizes+0.039
 planets+0.034
bows+0.034
 ABC+0.034
 clubs+0.033
Neg logit contributions
 late-0.030
 aback-0.026
Start-0.025
O-0.025
sm-0.025
Token they
Token ID484
Feature activation+1.71e-02

Pos logit contributions
 late+0.052
 aback+0.047
Start+0.046
O+0.045
sm+0.044
Neg logit contributions
 sizes-0.065
 planets-0.059
 ABC-0.056
bows-0.055
 cultures-0.055
Token would
Token ID561
Feature activation+7.50e-02

Pos logit contributions
 sizes+0.019
 planets+0.016
bows+0.015
 ABC+0.015
 cultures+0.015
Neg logit contributions
 late-0.041
 aback-0.037
Start-0.036
O-0.036
sm-0.036
Token see
Token ID766
Feature activation-3.06e-02

Pos logit contributions
 sizes+0.145
bows+0.125
 planets+0.125
 ABC+0.124
 clubs+0.124
Neg logit contributions
 late-0.120
 aback-0.098
Start-0.096
O-0.093
sm-0.092
Token Comet
Token ID36238
Feature activation+6.83e-02

Pos logit contributions
 late+0.042
 aback+0.037
Start+0.037
O+0.037
izz+0.034
Neg logit contributions
 sizes-0.065
 planets-0.061
 ABC-0.056
 cultures-0.056
 patterns-0.056
Token Hale
Token ID35056
Feature activation+4.32e-02

Pos logit contributions
 sizes+0.114
 clubs+0.098
 planets+0.096
 cultures+0.095
 ABC+0.095
Neg logit contributions
 late-0.122
 aback-0.102
sm-0.102
Start-0.100
 severe-0.096
Token-
Token ID12
Feature activation-7.68e-02

Pos logit contributions
 sizes+0.105
 ABC+0.096
bows+0.094
 clubs+0.094
 planets+0.093
Neg logit contributions
 late-0.041
 aback-0.033
sm-0.032
Start-0.028
 severe-0.027
TokenB
Token ID33
Feature activation+7.00e-02

Pos logit contributions
 late+0.131
 aback+0.113
Start+0.109
O+0.109
sm+0.107
Neg logit contributions
 sizes-0.138
 planets-0.122
bows-0.121
 ABC-0.119
 cultures-0.117
Tokenopp
Token ID10365
Feature activation+1.08e-02

Pos logit contributions
 sizes+0.117
 planets+0.106
 ABC+0.097
 clubs+0.097
 cultures+0.096
Neg logit contributions
 late-0.123
 aback-0.111
sm-0.111
Start-0.106
O-0.105
Token again
Token ID757
Feature activation-5.75e-02

Pos logit contributions
 sizes+0.013
 planets+0.011
 clubs+0.011
 ABC+0.010
 cultures+0.010
Neg logit contributions
 late-0.024
sm-0.021
 aback-0.021
Start-0.020
 severe-0.020
Token someday
Token ID25580
Feature activation+3.29e-02

Pos logit contributions
 late+0.113
 aback+0.100
Start+0.098
O+0.098
sm+0.097
Neg logit contributions
 sizes-0.087
 planets-0.075
bows-0.073
 ABC-0.072
 cultures-0.071
Token,
Token ID11
Feature activation+3.61e-02

Pos logit contributions
 sizes+0.057
bows+0.049
 planets+0.049
 ABC+0.049
 clubs+0.048
Neg logit contributions
 late-0.059
 aback-0.048
sm-0.047
Start-0.046
 severe-0.045
Token She
Token ID1375
Feature activation-1.02e-02

Pos logit contributions
 sizes+0.008
 planets+0.007
bows+0.007
 ABC+0.007
 clubs+0.007
Neg logit contributions
 late-0.006
 aback-0.005
Start-0.005
O-0.005
sm-0.005
Token felt
Token ID2936
Feature activation-2.81e-02

Pos logit contributions
 late+0.017
 aback+0.015
Start+0.014
O+0.014
sm+0.014
Neg logit contributions
 sizes-0.019
bows-0.016
 planets-0.016
 ABC-0.016
 cultures-0.016
Token better
Token ID1365
Feature activation-6.83e-03

Pos logit contributions
 late+0.047
izz+0.046
plug+0.045
O+0.045
Start+0.043
Neg logit contributions
 sizes-0.043
 planets-0.037
 patterns-0.036
 clubs-0.035
 cultures-0.035
Token and
Token ID290
Feature activation-2.71e-02

Pos logit contributions
 late+0.013
 aback+0.011
Start+0.011
sm+0.011
O+0.011
Neg logit contributions
 sizes-0.011
 planets-0.009
bows-0.009
 ABC-0.009
 clubs-0.009
Token went
Token ID1816
Feature activation-5.85e-02

Pos logit contributions
 late+0.044
 aback+0.038
Start+0.037
O+0.036
sm+0.036
Neg logit contributions
 sizes-0.051
 planets-0.045
bows-0.044
 ABC-0.044
 clubs-0.043
Token back
Token ID736
Feature activation-8.72e-02

Pos logit contributions
 late+0.073
 aback+0.059
Start+0.056
O+0.056
sm+0.056
Neg logit contributions
 sizes-0.133
 planets-0.120
bows-0.119
 ABC-0.117
 clubs-0.116
Token outside
Token ID2354
Feature activation-1.50e-02

Pos logit contributions
O+0.150
 late+0.142
Start+0.140
izz+0.132
 severe+0.132
Neg logit contributions
 sizes-0.128
 planets-0.126
 pretending-0.125
 clubs-0.122
bows-0.120
Token to
Token ID284
Feature activation-1.68e-02

Pos logit contributions
 late+0.023
 aback+0.020
Start+0.020
O+0.019
sm+0.019
Neg logit contributions
 sizes-0.029
 planets-0.025
bows-0.025
 ABC-0.025
 clubs-0.024
Token play
Token ID711
Feature activation+5.07e-02

Pos logit contributions
 late+0.034
 aback+0.031
O+0.030
Start+0.030
sm+0.030
Neg logit contributions
 sizes-0.024
 planets-0.021
 ABC-0.020
bows-0.020
 clubs-0.020
Token with
Token ID351
Feature activation-5.23e-03

Pos logit contributions
 sizes+0.069
 planets+0.062
 clubs+0.061
bows+0.060
 patterns+0.059
Neg logit contributions
 late-0.101
 aback-0.092
 severe-0.090
Start-0.089
izz-0.088
Token her
Token ID607
Feature activation+1.08e-02

Pos logit contributions
 late+0.008
 aback+0.007
Start+0.007
plug+0.006
O+0.006
Neg logit contributions
 sizes-0.010
 planets-0.010
 patterns-0.009
 clubs-0.009
 toes-0.009
Token Lily
Token ID20037
Feature activation+1.90e-03

Pos logit contributions
 late+0.104
 severe+0.101
sm+0.096
eh+0.093
vous+0.092
Neg logit contributions
 shirts-0.078
 sizes-0.078
 toes-0.077
 planets-0.076
 patterns-0.074
Token and
Token ID290
Feature activation-2.47e-02

Pos logit contributions
 sizes+0.003
 planets+0.003
bows+0.003
 ABC+0.003
 cultures+0.003
Neg logit contributions
 late-0.003
 aback-0.003
sm-0.003
Start-0.003
O-0.003
Token Anna
Token ID11735
Feature activation+1.05e-02

Pos logit contributions
 late+0.046
 aback+0.041
sm+0.040
handed+0.040
O+0.040
Neg logit contributions
 sizes-0.036
 cheeks-0.034
 pretending-0.034
 shirts-0.033
 toes-0.033
Token.
Token ID13
Feature activation+5.99e-03

Pos logit contributions
 sizes+0.020
 planets+0.018
bows+0.018
 ABC+0.018
 cultures+0.017
Neg logit contributions
 late-0.017
 aback-0.014
Start-0.014
sm-0.014
O-0.014
Token She
Token ID1375
Feature activation-8.32e-03

Pos logit contributions
 sizes+0.010
 ABC+0.009
 planets+0.009
bows+0.008
 toes+0.008
Neg logit contributions
 late-0.010
 aback-0.009
sm-0.009
 severe-0.009
 poisoned-0.009
Token took
Token ID1718
Feature activation-7.20e-02

Pos logit contributions
 late+0.016
 aback+0.014
Start+0.013
O+0.013
Aid+0.013
Neg logit contributions
 sizes-0.012
 neck-0.010
 necks-0.010
 clubs-0.010
 ABC-0.010
Token them
Token ID606
Feature activation+3.81e-03

Pos logit contributions
 late+0.131
 severe+0.125
 poisonous+0.121
sm+0.120
eh+0.119
Neg logit contributions
 toes-0.094
 shirts-0.092
 patterns-0.091
 sizes-0.091
 planets-0.090
Token to
Token ID284
Feature activation-1.48e-02

Pos logit contributions
 sizes+0.008
 clubs+0.007
bows+0.007
 planets+0.007
 toes+0.007
Neg logit contributions
 late-0.005
O-0.005
Start-0.005
 aback-0.004
sm-0.004
Token the
Token ID262
Feature activation+3.59e-02

Pos logit contributions
vious+0.024
 aback+0.023
Aid+0.023
 late+0.023
sm+0.023
Neg logit contributions
 ABC-0.021
 pretending-0.020
 neck-0.020
 patterns-0.020
 sizes-0.020
Token bedroom
Token ID14043
Feature activation+3.36e-02

Pos logit contributions
 sizes+0.052
 ABC+0.046
 planets+0.046
 clubs+0.046
bows+0.045
Neg logit contributions
 late-0.069
 aback-0.064
Start-0.061
sm-0.061
izz-0.060
Token.
Token ID13
Feature activation+8.73e-03

Pos logit contributions
 sizes+0.049
 planets+0.041
bows+0.041
 ABC+0.040
 clubs+0.039
Neg logit contributions
 late-0.068
 aback-0.061
O-0.060
Start-0.060
sm-0.059
Token the
Token ID262
Feature activation+2.05e-02

Pos logit contributions
 late+0.007
 aback+0.006
Start+0.006
O+0.006
sm+0.006
Neg logit contributions
 sizes-0.006
 planets-0.005
 toes-0.005
 clubs-0.005
bows-0.005
Token boy
Token ID2933
Feature activation+1.16e-02

Pos logit contributions
 sizes+0.039
 planets+0.034
bows+0.034
 ABC+0.033
 cultures+0.033
Neg logit contributions
 late-0.033
 aback-0.028
Start-0.027
O-0.027
sm-0.027
Token thanked
Token ID26280
Feature activation+5.48e-02

Pos logit contributions
 sizes+0.019
 planets+0.017
bows+0.016
 ABC+0.016
 toes+0.016
Neg logit contributions
 late-0.021
 aback-0.019
Start-0.018
O-0.018
sm-0.018
Token his
Token ID465
Feature activation+1.15e-02

Pos logit contributions
 sizes+0.109
 planets+0.099
 ABC+0.097
 cultures+0.096
bows+0.095
Neg logit contributions
 late-0.080
 aback-0.069
Start-0.066
sm-0.065
O-0.063
Token mom
Token ID1995
Feature activation-1.80e-02

Pos logit contributions
 sizes+0.022
bows+0.019
 planets+0.018
 clubs+0.018
hips+0.018
Neg logit contributions
 late-0.019
sm-0.016
 aback-0.016
Start-0.016
 severe-0.015
Token for
Token ID329
Feature activation-7.72e-02

Pos logit contributions
 late+0.027
sm+0.022
 aback+0.021
 severe+0.019
 another+0.018
Neg logit contributions
 sizes-0.036
 planets-0.033
bows-0.033
 clubs-0.032
 necks-0.032
Token helping
Token ID5742
Feature activation+7.63e-02

Pos logit contributions
 late+0.153
 aback+0.139
Start+0.134
O+0.133
sm+0.132
Neg logit contributions
 sizes-0.111
 planets-0.097
 ABC-0.093
 clubs-0.093
 cultures-0.091
Token him
Token ID683
Feature activation+3.36e-02

Pos logit contributions
 sizes+0.148
 planets+0.131
bows+0.129
 ABC+0.129
 clubs+0.127
Neg logit contributions
 late-0.119
 aback-0.101
Start-0.099
sm-0.097
O-0.097
Token sort
Token ID3297
Feature activation+1.61e-02

Pos logit contributions
 sizes+0.052
 planets+0.044
bows+0.044
 ABC+0.043
 clubs+0.042
Neg logit contributions
 late-0.065
 aback-0.057
Start-0.056
O-0.056
sm-0.056
Token all
Token ID477
Feature activation-3.56e-02

Pos logit contributions
 sizes+0.032
 planets+0.028
 clubs+0.028
bows+0.027
 toes+0.027
Neg logit contributions
 late-0.024
O-0.021
Start-0.021
sm-0.020
 severe-0.019
Token his
Token ID465
Feature activation+3.77e-03

Pos logit contributions
 late+0.045
O+0.042
Start+0.040
 severe+0.037
 poisoned+0.037
Neg logit contributions
 sizes-0.079
 planets-0.069
 patterns-0.067
 languages-0.066
 cultures-0.066