TOP ACTIVATIONS
MAX = 9.668

his head out of his bur row and saw something new
who lived in his cozy bur row . He loved to
mist and into their cozy bur rows . The end .
able to escape from his bur row and play in the
and showed him around their bur row . They took turns
and soon they found their bur row . The animals were
enough it disappeared into a bur row . The
peek ed out of his bur row . He
ies lived in a cozy bur row near a small lake
until it reached its cozy bur row . The end .
he finally reached his cozy bur row . From then on
Benny lived in a cozy bur row with his family .
was too strong and the bur row started to fill with
from the river to their bur row .
he could back to his bur row . When
got m - mad and r - ran away and I
fluffy and lived in a bur row in the forest .
who lived in the same bur row . They were very
it to his friends ' bur rows . They all thought
ate their popcorn while it r ained outside . Lily was

MIN ACTIVATIONS
MIN = -9.759

to share and wait . How about we clean up this
play with other toys . How about this puzzle ?" She
downstairs for a snack . How does that sound ?"
that you both like . How about a cartoon about animals
in the living room . How does that sound ?"
are plenty of them . How about that ?"
is safe and clean . How does that sound ?"
just one of you . How would you feel if someone
you for saying sorry . How about we go home and
will be sad too . How can we fix it ?"
share and take turns . How about this : you can
. Sharing is caring . How about you take turns playing
back and relax together . How does that sound ?"
on the big slide . How was it ?" Sara 's
you can use safely . How about that ?" he said
play with something else . How about the blocks ?" Tim
magic book ? Really ? How much do you want for
play and watch cartoons . How about that ?"
can play pretend together . How about we pretend to be
go the other way . How does that sound ?"

INTERVAL 4.812 - 9.668
CONTAINS 0.622%

ained and r ained and r ained . The rain was
. But it was too wind y outside . So the
home . It 's too wind y here ."
be careful , because the tunnel was fragile .
as he walked home . <|endoftext|> Once upon a time there

INTERVAL -0.045 - 4.812
CONTAINS 55.764%

prince ?" The little prince showed his mum the
a small girl named Lily . She lived in a warm
She said , " Tom , why don 't you take
tried to dig a hole for more light , but the
<|endoftext|> Once upon a time , there was a little girl

INTERVAL -4.902 - -0.045
CONTAINS 42.318%

. He runs to the musician and says ,
hon k . They see a big truck . The truck
! He said he was here to help her make friends
she tried to climb up on a chair to put a
. The frog was so shocked he jumped straight out of

INTERVAL -9.759 - -4.902
CONTAINS 1.296%

boy was gone . " Where are you ?" she shouted
is the best summer ever !" Anna says .
ze makes the blocks fall down . The building is gone
, little spirit , â said F awn .
the old man , " What is it ?"
Token his
Token ID465
Feature activation+2.32e+00

Pos logit contributions
 blonde+0.030
 Eva+0.030
chard+0.030
 wrists+0.030
 Robert+0.030
Neg logit contributions
 of-0.019
loo-0.018
ick-0.017
el-0.016
iren-0.015
Token head
Token ID1182
Feature activation+9.87e-01

Pos logit contributions
loo+0.083
 of+0.075
ick+0.069
 ig+0.069
auc+0.066
Neg logit contributions
 Eva-0.103
chard-0.102
 Robert-0.102
 wrists-0.100
 blonde-0.097
Token out
Token ID503
Feature activation-1.08e+00

Pos logit contributions
loo+0.027
 of+0.024
ick+0.022
iren+0.021
auc+0.021
Neg logit contributions
 Eva-0.052
 Robert-0.052
chard-0.052
 wrists-0.052
 blonde-0.050
Token of
Token ID286
Feature activation+4.11e+00

Pos logit contributions
 Eva+0.061
 Robert+0.060
chard+0.059
 wrists+0.059
 blonde+0.058
Neg logit contributions
loo-0.028
iren-0.021
auc-0.021
ick-0.021
arl-0.020
Token his
Token ID465
Feature activation+1.09e+00

Pos logit contributions
loo+0.157
auc+0.135
ick+0.134
iren+0.132
 of+0.129
Neg logit contributions
 Eva-0.179
 Robert-0.179
chard-0.176
 wrists-0.175
 blonde-0.170
Token bur
Token ID4356
Feature activation+8.15e+00

Pos logit contributions
loo+0.038
 of+0.033
ick+0.032
iren+0.031
auc+0.031
Neg logit contributions
 Eva-0.050
chard-0.050
 Robert-0.050
 wrists-0.049
 blonde-0.047
Tokenrow
Token ID808
Feature activation+5.51e+00

Pos logit contributions
loo+0.164
auc+0.124
pour+0.121
iren+0.120
 of+0.103
Neg logit contributions
chard-0.466
 wrists-0.461
 blonde-0.450
 Eva-0.448
 Robert-0.438
Token and
Token ID290
Feature activation-2.25e-01

Pos logit contributions
loo+0.175
iren+0.145
ick+0.142
auc+0.141
 of+0.139
Neg logit contributions
chard-0.267
 Robert-0.265
 Eva-0.263
 wrists-0.260
 blonde-0.254
Token saw
Token ID2497
Feature activation-2.42e-01

Pos logit contributions
 Eva+0.009
 Robert+0.009
chard+0.009
 wrists+0.009
 blonde+0.009
Neg logit contributions
loo-0.009
 of-0.008
ick-0.008
iren-0.008
auc-0.008
Token something
Token ID1223
Feature activation-2.98e+00

Pos logit contributions
 wrists+0.011
chard+0.011
 Eva+0.011
 Robert+0.011
 blonde+0.010
Neg logit contributions
loo-0.008
 of-0.008
ick-0.007
auc-0.007
 ig-0.007
Token new
Token ID649
Feature activation-1.87e+00

Pos logit contributions
 Eva+0.125
 wrists+0.124
 Robert+0.123
chard+0.123
 blonde+0.118
Neg logit contributions
loo-0.117
 of-0.106
ick-0.101
auc-0.099
iren-0.098
Token who
Token ID508
Feature activation-4.55e+00

Pos logit contributions
chard+0.103
 Eva+0.103
 Robert+0.103
 wrists+0.102
 blonde+0.098
Neg logit contributions
loo-0.101
 of-0.087
ick-0.087
iren-0.086
auc-0.086
Token lived
Token ID5615
Feature activation+1.32e+00

Pos logit contributions
 wrists+0.191
chard+0.186
 Eva+0.184
 Robert+0.183
 blonde+0.169
Neg logit contributions
loo-0.176
 of-0.172
ick-0.159
auc-0.147
iren-0.147
Token in
Token ID287
Feature activation-2.19e+00

Pos logit contributions
loo+0.041
 of+0.035
ick+0.035
iren+0.033
auc+0.033
Neg logit contributions
 Eva-0.066
 Robert-0.065
 wrists-0.065
chard-0.065
 blonde-0.063
Token his
Token ID465
Feature activation+4.54e-01

Pos logit contributions
 Eva+0.083
 Robert+0.083
chard+0.081
 wrists+0.079
 blonde+0.078
Neg logit contributions
loo-0.097
auc-0.085
iren-0.084
ick-0.083
 of-0.083
Token cozy
Token ID37438
Feature activation+2.02e+00

Pos logit contributions
loo+0.015
iren+0.013
auc+0.013
ick+0.013
 of+0.013
Neg logit contributions
 Eva-0.022
 Robert-0.021
chard-0.021
 wrists-0.021
 blonde-0.021
Token bur
Token ID4356
Feature activation+7.65e+00

Pos logit contributions
loo+0.069
auc+0.064
iren+0.063
ick+0.058
arl+0.057
Neg logit contributions
 wrists-0.094
chard-0.093
 Robert-0.093
 blonde-0.092
 Eva-0.092
Tokenrow
Token ID808
Feature activation+5.10e+00

Pos logit contributions
loo+0.162
auc+0.141
iren+0.133
pour+0.130
ick+0.114
Neg logit contributions
chard-0.435
 wrists-0.418
 blonde-0.410
 Robert-0.394
 Eva-0.394
Token.
Token ID13
Feature activation+6.30e-01

Pos logit contributions
loo+0.159
iren+0.131
ick+0.129
auc+0.128
 of+0.125
Neg logit contributions
chard-0.254
 Robert-0.250
 Eva-0.249
 wrists-0.249
 blonde-0.242
Token He
Token ID679
Feature activation-7.73e-01

Pos logit contributions
 of+0.019
loo+0.018
 late+0.015
 ig+0.015
ick+0.015
Neg logit contributions
 wrists-0.031
chard-0.030
cious-0.029
 yes-0.028
rice-0.028
Token loved
Token ID6151
Feature activation-2.60e-01

Pos logit contributions
chard+0.038
 Robert+0.038
 Eva+0.038
 wrists+0.037
 Edward+0.035
Neg logit contributions
 of-0.023
loo-0.021
 ig-0.020
ick-0.019
row-0.018
Token to
Token ID284
Feature activation-3.38e-02

Pos logit contributions
 Eva+0.012
 Robert+0.012
chard+0.012
 wrists+0.012
 blonde+0.012
Neg logit contributions
loo-0.009
 of-0.007
ick-0.007
iren-0.007
auc-0.007
Token mist
Token ID4020
Feature activation+2.06e+00

Pos logit contributions
loo+0.038
ick+0.033
 of+0.032
iren+0.032
auc+0.032
Neg logit contributions
 Eva-0.047
chard-0.047
 Robert-0.047
 wrists-0.047
 blonde-0.046
Token and
Token ID290
Feature activation-3.41e-01

Pos logit contributions
loo+0.066
iren+0.055
ick+0.055
auc+0.054
 of+0.053
Neg logit contributions
 Eva-0.099
 Robert-0.098
chard-0.098
 wrists-0.097
 blonde-0.095
Token into
Token ID656
Feature activation-4.33e-01

Pos logit contributions
 Robert+0.014
 wrists+0.014
 Eva+0.014
 blonde+0.014
chard+0.014
Neg logit contributions
loo-0.013
 of-0.012
pour-0.011
auc-0.010
ick-0.010
Token their
Token ID511
Feature activation-1.03e-01

Pos logit contributions
chard+0.019
 Eva+0.019
 Robert+0.019
 wrists+0.019
 blonde+0.018
Neg logit contributions
loo-0.016
 of-0.014
ick-0.014
iren-0.013
auc-0.013
Token cozy
Token ID37438
Feature activation+1.98e+00

Pos logit contributions
 Eva+0.005
chard+0.005
 Robert+0.005
 wrists+0.005
 blonde+0.005
Neg logit contributions
loo-0.003
 of-0.003
ick-0.003
iren-0.003
auc-0.003
Token bur
Token ID4356
Feature activation+7.47e+00

Pos logit contributions
loo+0.071
auc+0.066
iren+0.065
ick+0.059
arl+0.058
Neg logit contributions
 wrists-0.090
 blonde-0.089
 Eva-0.085
chard-0.085
 Claire-0.084
Tokenrows
Token ID8516
Feature activation+9.52e-01

Pos logit contributions
loo+0.162
auc+0.133
iren+0.130
pour+0.121
ick+0.108
Neg logit contributions
 wrists-0.398
chard-0.396
 blonde-0.389
cious-0.386
 Eva-0.378
Token.
Token ID13
Feature activation+3.70e-01

Pos logit contributions
loo+0.030
ick+0.024
iren+0.024
 of+0.024
auc+0.024
Neg logit contributions
chard-0.047
 Eva-0.047
 Robert-0.047
 wrists-0.047
 blonde-0.045
Token The
Token ID383
Feature activation+2.44e-02

Pos logit contributions
loo+0.014
 of+0.013
ick+0.011
iren+0.011
auc+0.011
Neg logit contributions
 wrists-0.017
chard-0.016
 blonde-0.015
cious-0.015
 Eva-0.014
Token end
Token ID886
Feature activation+4.00e+00

Pos logit contributions
loo+0.001
 of+0.001
ick+0.001
auc+0.001
iren+0.001
Neg logit contributions
 Eva-0.001
 Robert-0.001
chard-0.001
 wrists-0.001
 blonde-0.001
Token.
Token ID13
Feature activation+2.55e-01

Pos logit contributions
loo+0.150
arl+0.134
iren+0.132
auc+0.131
ick+0.124
Neg logit contributions
 Eva-0.166
 Robert-0.157
 blonde-0.154
 Joy-0.153
 aloud-0.151
Token able
Token ID1498
Feature activation-2.19e+00

Pos logit contributions
loo+0.093
 of+0.082
ick+0.080
iren+0.078
auc+0.077
Neg logit contributions
 Eva-0.108
chard-0.107
 wrists-0.107
 Robert-0.107
 blonde-0.103
Token to
Token ID284
Feature activation-6.13e-03

Pos logit contributions
 Robert+0.103
 Eva+0.103
chard+0.103
 wrists+0.102
 blonde+0.098
Neg logit contributions
loo-0.075
ick-0.063
iren-0.062
auc-0.062
 of-0.059
Token escape
Token ID6654
Feature activation+1.02e+00

Pos logit contributions
 Eva+0.000
 Robert+0.000
chard+0.000
 wrists+0.000
 blonde+0.000
Neg logit contributions
loo-0.000
ick-0.000
 of-0.000
auc-0.000
iren-0.000
Token from
Token ID422
Feature activation-5.19e-01

Pos logit contributions
loo+0.033
ick+0.027
iren+0.027
auc+0.026
 of+0.026
Neg logit contributions
 Robert-0.050
chard-0.050
 Eva-0.050
 wrists-0.049
 blonde-0.047
Token his
Token ID465
Feature activation+9.35e-02

Pos logit contributions
 wrists+0.024
 Eva+0.024
 Robert+0.024
chard+0.024
 blonde+0.023
Neg logit contributions
loo-0.018
 of-0.016
ick-0.015
iren-0.015
auc-0.014
Token bur
Token ID4356
Feature activation+7.46e+00

Pos logit contributions
loo+0.003
 of+0.003
ick+0.003
iren+0.003
auc+0.003
Neg logit contributions
 Eva-0.004
chard-0.004
 Robert-0.004
 wrists-0.004
 blonde-0.004
Tokenrow
Token ID808
Feature activation+4.86e+00

Pos logit contributions
loo+0.155
auc+0.120
iren+0.112
pour+0.110
ick+0.105
Neg logit contributions
chard-0.414
 wrists-0.410
 blonde-0.393
rice-0.385
 Eva-0.384
Token and
Token ID290
Feature activation-1.66e-01

Pos logit contributions
loo+0.158
iren+0.131
ick+0.129
auc+0.127
arl+0.122
Neg logit contributions
chard-0.232
 Robert-0.230
 Eva-0.227
 wrists-0.225
 blonde-0.219
Token play
Token ID711
Feature activation+7.90e-02

Pos logit contributions
 Eva+0.007
 Robert+0.007
chard+0.007
 wrists+0.007
 blonde+0.006
Neg logit contributions
loo-0.007
ick-0.006
 of-0.006
iren-0.006
auc-0.006
Token in
Token ID287
Feature activation-2.48e+00

Pos logit contributions
loo+0.002
 of+0.002
ick+0.002
auc+0.002
iren+0.002
Neg logit contributions
 Eva-0.004
chard-0.004
 wrists-0.004
 Robert-0.004
 blonde-0.004
Token the
Token ID262
Feature activation+5.17e-01

Pos logit contributions
 Robert+0.102
 Eva+0.101
chard+0.098
 wrists+0.096
 Claire+0.095
Neg logit contributions
loo-0.101
auc-0.089
iren-0.088
arl-0.086
ick-0.086
Token and
Token ID290
Feature activation-3.64e-02

Pos logit contributions
 Eva+0.164
chard+0.163
 Robert+0.163
 wrists+0.162
 blonde+0.157
Neg logit contributions
loo-0.095
 of-0.079
ick-0.078
iren-0.076
auc-0.075
Token showed
Token ID3751
Feature activation-1.24e+00

Pos logit contributions
chard+0.002
 Eva+0.002
 Robert+0.002
 wrists+0.002
 blonde+0.001
Neg logit contributions
loo-0.001
 of-0.001
ick-0.001
iren-0.001
auc-0.001
Token him
Token ID683
Feature activation-3.26e-01

Pos logit contributions
chard+0.061
 wrists+0.058
 blonde+0.058
 Eva+0.057
 Robert+0.056
Neg logit contributions
loo-0.039
 of-0.039
ick-0.035
auc-0.034
iren-0.032
Token around
Token ID1088
Feature activation-1.91e+00

Pos logit contributions
chard+0.015
 Robert+0.015
 Eva+0.015
 wrists+0.015
 blonde+0.015
Neg logit contributions
loo-0.011
 of-0.010
ick-0.009
auc-0.009
 ig-0.009
Token their
Token ID511
Feature activation-2.61e-01

Pos logit contributions
 Eva+0.095
chard+0.095
 Robert+0.094
 wrists+0.094
 blonde+0.091
Neg logit contributions
loo-0.059
 of-0.051
ick-0.049
auc-0.047
iren-0.047
Token bur
Token ID4356
Feature activation+7.39e+00

Pos logit contributions
chard+0.012
 Eva+0.012
 Robert+0.012
 blonde+0.011
 wrists+0.011
Neg logit contributions
loo-0.009
 of-0.008
 ig-0.008
ick-0.008
 tunnel-0.008
Tokenrow
Token ID808
Feature activation+4.66e+00

Pos logit contributions
loo+0.128
auc+0.091
pour+0.090
iren+0.090
ick+0.088
Neg logit contributions
 wrists-0.452
chard-0.441
 blonde-0.434
 Eva-0.426
 Robert-0.418
Token.
Token ID13
Feature activation+5.21e-01

Pos logit contributions
loo+0.140
ick+0.114
 of+0.114
iren+0.113
auc+0.111
Neg logit contributions
chard-0.235
 Eva-0.235
 Robert-0.234
 wrists-0.234
 blonde-0.225
Token They
Token ID1119
Feature activation-1.44e-01

Pos logit contributions
loo+0.019
 of+0.019
auc+0.017
ick+0.017
iren+0.017
Neg logit contributions
 wrists-0.022
chard-0.022
cious-0.020
 Eva-0.020
 blonde-0.020
Token took
Token ID1718
Feature activation+4.48e-02

Pos logit contributions
 Eva+0.006
chard+0.006
 Robert+0.006
 wrists+0.006
 blonde+0.006
Neg logit contributions
loo-0.005
 of-0.005
ick-0.005
 ig-0.005
auc-0.005
Token turns
Token ID4962
Feature activation-1.23e+00

Pos logit contributions
loo+0.002
 of+0.001
ick+0.001
auc+0.001
iren+0.001
Neg logit contributions
chard-0.002
 wrists-0.002
 Eva-0.002
 Robert-0.002
 blonde-0.002
Token and
Token ID290
Feature activation-3.22e-01

Pos logit contributions
 of+0.011
loo+0.010
ick+0.009
pour+0.009
 late+0.009
Neg logit contributions
 wrists-0.016
 Eva-0.015
 Robert-0.015
chard-0.015
 blonde-0.015
Token soon
Token ID2582
Feature activation+8.67e-01

Pos logit contributions
 wrists+0.014
 Robert+0.014
 Eva+0.014
 blonde+0.014
chard+0.014
Neg logit contributions
loo-0.011
 of-0.011
ick-0.009
pour-0.009
auc-0.009
Token they
Token ID484
Feature activation+9.57e-01

Pos logit contributions
loo+0.028
 of+0.027
ick+0.023
pour+0.022
auc+0.021
Neg logit contributions
 wrists-0.042
chard-0.041
 blonde-0.039
 Robert-0.039
 Eva-0.039
Token found
Token ID1043
Feature activation+1.18e+00

Pos logit contributions
loo+0.034
 of+0.032
ick+0.029
pour+0.028
auc+0.028
Neg logit contributions
 wrists-0.043
 Eva-0.042
chard-0.042
 Robert-0.041
 blonde-0.041
Token their
Token ID511
Feature activation-2.37e-01

Pos logit contributions
loo+0.040
 of+0.039
ick+0.033
auc+0.033
 ig+0.032
Neg logit contributions
 wrists-0.055
chard-0.055
 blonde-0.054
 Eva-0.054
 Robert-0.054
Token bur
Token ID4356
Feature activation+7.31e+00

Pos logit contributions
 Eva+0.011
 Robert+0.011
chard+0.011
 wrists+0.010
 blonde+0.010
Neg logit contributions
loo-0.008
 of-0.007
 ig-0.007
ick-0.007
 tunnel-0.007
Tokenrow
Token ID808
Feature activation+4.70e+00

Pos logit contributions
loo+0.159
iren+0.133
auc+0.128
pour+0.119
ick+0.113
Neg logit contributions
chard-0.404
 wrists-0.393
 blonde-0.379
 Robert-0.375
 Eva-0.375
Token.
Token ID13
Feature activation+3.62e-01

Pos logit contributions
loo+0.150
iren+0.125
ick+0.124
 of+0.121
auc+0.120
Neg logit contributions
chard-0.228
 Robert-0.226
 Eva-0.226
 wrists-0.224
 blonde-0.216
Token The
Token ID383
Feature activation+6.06e-02

Pos logit contributions
loo+0.012
 of+0.012
iren+0.011
auc+0.010
ick+0.010
Neg logit contributions
 wrists-0.016
chard-0.016
 blonde-0.015
cious-0.015
 Eva-0.014
Token animals
Token ID4695
Feature activation-1.98e+00

Pos logit contributions
loo+0.002
 of+0.002
ick+0.002
 ig+0.002
auc+0.002
Neg logit contributions
 Eva-0.003
 Robert-0.003
 wrists-0.003
chard-0.003
 blonde-0.003
Token were
Token ID547
Feature activation+8.74e-01

Pos logit contributions
 Eva+0.087
 wrists+0.086
 Robert+0.086
chard+0.085
 blonde+0.083
Neg logit contributions
loo-0.073
 of-0.068
ick-0.063
auc-0.061
iren-0.061
Token enough
Token ID1576
Feature activation+4.38e-01

Pos logit contributions
loo+0.020
 of+0.020
ick+0.018
pour+0.016
 ig+0.016
Neg logit contributions
 wrists-0.033
chard-0.032
 blonde-0.030
 Eva-0.030
 Robert-0.029
Token it
Token ID340
Feature activation+1.78e-01

Pos logit contributions
loo+0.015
 of+0.013
ick+0.012
iren+0.012
auc+0.012
Neg logit contributions
chard-0.021
 Eva-0.021
 wrists-0.021
 Robert-0.021
 blonde-0.020
Token disappeared
Token ID12120
Feature activation+1.93e+00

Pos logit contributions
loo+0.007
 of+0.006
ick+0.006
iren+0.006
auc+0.006
Neg logit contributions
 Eva-0.008
 Robert-0.008
chard-0.008
 wrists-0.008
 blonde-0.007
Token into
Token ID656
Feature activation-4.59e-01

Pos logit contributions
loo+0.066
iren+0.056
auc+0.053
ick+0.053
 of+0.051
Neg logit contributions
chard-0.091
 Robert-0.091
 Eva-0.090
 wrists-0.088
 blonde-0.087
Token a
Token ID257
Feature activation-1.07e+00

Pos logit contributions
 Eva+0.019
 Robert+0.019
chard+0.019
 wrists+0.019
 blonde+0.018
Neg logit contributions
loo-0.018
auc-0.016
iren-0.016
ick-0.016
 of-0.015
Token bur
Token ID4356
Feature activation+7.26e+00

Pos logit contributions
chard+0.048
 Eva+0.048
 Robert+0.048
 wrists+0.048
 blonde+0.046
Neg logit contributions
loo-0.038
ick-0.033
 of-0.032
iren-0.032
auc-0.032
Tokenrow
Token ID808
Feature activation+4.57e+00

Pos logit contributions
loo+0.174
auc+0.143
 shower+0.136
iren+0.135
pour+0.134
Neg logit contributions
atively-0.324
chard-0.318
 blonde-0.314
 wrists-0.314
ya-0.313
Token.
Token ID13
Feature activation+3.26e-01

Pos logit contributions
loo+0.154
iren+0.133
arl+0.126
ick+0.125
auc+0.120
Neg logit contributions
chard-0.206
 Robert-0.200
 Eva-0.196
 Claire-0.190
 wrists-0.190
Token
Token ID198
Feature activation+1.15e-01

Pos logit contributions
loo+0.011
 of+0.011
 late+0.010
iren+0.010
ick+0.010
Neg logit contributions
 wrists-0.015
chard-0.014
cious-0.013
 blonde-0.013
 Eva-0.013
Token
Token ID198
Feature activation+4.52e-02

Pos logit contributions
loo+0.004
 of+0.003
iren+0.003
ick+0.003
auc+0.003
Neg logit contributions
 wrists-0.006
chard-0.006
 Robert-0.005
 Eva-0.005
 blonde-0.005
TokenThe
Token ID464
Feature activation-1.53e-01

Pos logit contributions
loo+0.002
 of+0.002
iren+0.002
ick+0.002
auc+0.001
Neg logit contributions
 wrists-0.002
chard-0.002
 Robert-0.002
 Eva-0.002
 blonde-0.002
Token peek
Token ID27185
Feature activation-9.04e-01

Pos logit contributions
chard+0.059
 Robert+0.058
 Eva+0.058
 wrists+0.057
 blonde+0.055
Neg logit contributions
loo-0.052
 of-0.048
ick-0.045
auc-0.044
iren-0.044
Tokened
Token ID276
Feature activation-1.74e+00

Pos logit contributions
chard+0.037
 Robert+0.037
 Eva+0.037
 wrists+0.037
 blonde+0.035
Neg logit contributions
loo-0.036
 of-0.033
ick-0.032
auc-0.032
iren-0.031
Token out
Token ID503
Feature activation-2.19e+00

Pos logit contributions
 wrists+0.087
chard+0.087
 Eva+0.086
 Robert+0.085
 blonde+0.083
Neg logit contributions
loo-0.053
 of-0.048
ick-0.045
auc-0.043
iren-0.042
Token of
Token ID286
Feature activation+3.12e+00

Pos logit contributions
 Eva+0.125
 Robert+0.122
 Claire+0.121
 yes+0.121
 blonde+0.118
Neg logit contributions
loo-0.057
iren-0.043
auc-0.041
ick-0.039
arl-0.037
Token his
Token ID465
Feature activation-1.32e-01

Pos logit contributions
loo+0.119
ick+0.102
auc+0.102
 of+0.101
iren+0.100
Neg logit contributions
 Eva-0.136
 Robert-0.135
chard-0.135
 wrists-0.133
 blonde-0.128
Token bur
Token ID4356
Feature activation+7.25e+00

Pos logit contributions
 wrists+0.005
chard+0.005
 Robert+0.005
 Eva+0.005
 blonde+0.005
Neg logit contributions
loo-0.005
iren-0.005
auc-0.005
ick-0.005
 of-0.005
Tokenrow
Token ID808
Feature activation+4.60e+00

Pos logit contributions
loo+0.166
auc+0.131
pour+0.130
iren+0.124
 of+0.118
Neg logit contributions
 wrists-0.394
chard-0.376
 blonde-0.367
cious-0.364
ya-0.361
Token.
Token ID13
Feature activation+5.01e-01

Pos logit contributions
loo+0.158
iren+0.132
auc+0.128
ick+0.126
arl+0.124
Neg logit contributions
chard-0.213
 Robert-0.210
 Eva-0.210
 wrists-0.207
 blonde-0.201
Token
Token ID198
Feature activation+2.01e-01

Pos logit contributions
loo+0.018
 of+0.017
iren+0.015
pour+0.015
ick+0.015
Neg logit contributions
 wrists-0.023
chard-0.022
cious-0.021
 Eva-0.020
 blonde-0.020
Token
Token ID198
Feature activation+1.64e-01

Pos logit contributions
loo+0.007
 of+0.007
ick+0.006
iren+0.006
auc+0.006
Neg logit contributions
 wrists-0.009
chard-0.009
 Robert-0.009
 Eva-0.009
 blonde-0.009
TokenHe
Token ID1544
Feature activation+1.44e-01

Pos logit contributions
loo+0.006
ick+0.005
 of+0.005
auc+0.005
iren+0.005
Neg logit contributions
 Eva-0.008
 Robert-0.008
chard-0.008
 wrists-0.007
 blonde-0.007
Tokenies
Token ID444
Feature activation-1.45e-01

Pos logit contributions
loo+0.051
iren+0.045
pour+0.041
auc+0.039
arl+0.037
Neg logit contributions
 wrists-0.097
 Eva-0.097
 Claire-0.096
chard-0.096
cious-0.095
Token lived
Token ID5615
Feature activation+8.36e-01

Pos logit contributions
chard+0.004
 Robert+0.004
 Eva+0.004
 wrists+0.004
 blonde+0.004
Neg logit contributions
loo-0.008
iren-0.007
ick-0.007
auc-0.007
arl-0.007
Token in
Token ID287
Feature activation-2.33e+00

Pos logit contributions
loo+0.026
 of+0.022
ick+0.022
iren+0.021
auc+0.021
Neg logit contributions
 Eva-0.041
 Robert-0.041
 wrists-0.041
chard-0.041
 blonde-0.040
Token a
Token ID257
Feature activation-7.79e-01

Pos logit contributions
 Eva+0.083
 Robert+0.082
chard+0.080
 wrists+0.078
 blonde+0.078
Neg logit contributions
loo-0.111
auc-0.097
iren-0.097
ick-0.094
 of-0.093
Token cozy
Token ID37438
Feature activation+1.91e+00

Pos logit contributions
chard+0.033
 Eva+0.033
 Robert+0.032
 wrists+0.032
 blonde+0.031
Neg logit contributions
loo-0.030
iren-0.026
 of-0.026
auc-0.026
ick-0.026
Token bur
Token ID4356
Feature activation+7.24e+00

Pos logit contributions
loo+0.069
auc+0.065
iren+0.063
ick+0.056
arl+0.055
Neg logit contributions
chard-0.085
 wrists-0.084
 Claire-0.084
 Eva-0.083
 Robert-0.083
Tokenrow
Token ID808
Feature activation+4.76e+00

Pos logit contributions
loo+0.154
pour+0.138
auc+0.134
iren+0.125
 accidents+0.121
Neg logit contributions
chard-0.379
 wrists-0.368
cious-0.367
 blonde-0.358
ya-0.352
Token near
Token ID1474
Feature activation-1.36e+00

Pos logit contributions
loo+0.157
iren+0.134
auc+0.128
ick+0.127
arl+0.125
Neg logit contributions
chard-0.229
 Robert-0.222
 Eva-0.221
 wrists-0.219
 blonde-0.212
Token a
Token ID257
Feature activation-7.72e-01

Pos logit contributions
 wrists+0.064
chard+0.064
 Robert+0.064
 Eva+0.064
 blonde+0.062
Neg logit contributions
loo-0.044
 of-0.041
ick-0.039
 ig-0.037
auc-0.036
Token small
Token ID1402
Feature activation+1.87e+00

Pos logit contributions
chard+0.036
 Eva+0.035
 Robert+0.035
 wrists+0.035
 blonde+0.034
Neg logit contributions
loo-0.027
 of-0.023
ick-0.023
iren-0.023
auc-0.022
Token lake
Token ID13546
Feature activation-5.57e-01

Pos logit contributions
loo+0.067
iren+0.059
auc+0.058
ick+0.056
arl+0.055
Neg logit contributions
chard-0.085
 Robert-0.081
 Eva-0.081
 wrists-0.081
 Claire-0.079
Token until
Token ID1566
Feature activation+4.06e-01

Pos logit contributions
 Eva+0.056
chard+0.056
 Robert+0.056
 wrists+0.056
 blonde+0.054
Neg logit contributions
loo-0.044
 of-0.038
ick-0.037
iren-0.037
auc-0.036
Token it
Token ID340
Feature activation+2.02e-01

Pos logit contributions
loo+0.015
 of+0.013
ick+0.013
iren+0.013
auc+0.013
Neg logit contributions
 Eva-0.017
chard-0.017
 Robert-0.017
 wrists-0.017
 blonde-0.017
Token reached
Token ID4251
Feature activation-4.93e-01

Pos logit contributions
loo+0.008
 of+0.007
ick+0.007
iren+0.007
auc+0.007
Neg logit contributions
 Eva-0.008
chard-0.008
 Robert-0.008
 wrists-0.008
 blonde-0.008
Token its
Token ID663
Feature activation-7.77e-01

Pos logit contributions
chard+0.023
 wrists+0.023
 Eva+0.022
 Robert+0.022
 blonde+0.022
Neg logit contributions
loo-0.017
 of-0.015
ick-0.015
iren-0.014
auc-0.014
Token cozy
Token ID37438
Feature activation+1.71e+00

Pos logit contributions
 Robert+0.037
 Eva+0.037
chard+0.036
 wrists+0.036
 blonde+0.035
Neg logit contributions
loo-0.026
 of-0.023
ick-0.022
iren-0.021
auc-0.021
Token bur
Token ID4356
Feature activation+7.24e+00

Pos logit contributions
loo+0.057
iren+0.053
auc+0.053
ick+0.047
arl+0.047
Neg logit contributions
 wrists-0.079
 blonde-0.078
chard-0.078
 Eva-0.077
 Robert-0.077
Tokenrow
Token ID808
Feature activation+4.63e+00

Pos logit contributions
loo+0.158
iren+0.126
auc+0.125
pour+0.114
ick+0.104
Neg logit contributions
chard-0.383
 wrists-0.368
atively-0.362
 blonde-0.362
rice-0.356
Token.
Token ID13
Feature activation+1.71e-01

Pos logit contributions
loo+0.154
iren+0.129
auc+0.124
ick+0.123
arl+0.118
Neg logit contributions
chard-0.220
 Robert-0.216
 Eva-0.215
 wrists-0.212
 blonde-0.208
Token The
Token ID383
Feature activation-3.15e-02

Pos logit contributions
loo+0.006
 of+0.006
ick+0.006
pour+0.006
 late+0.006
Neg logit contributions
 wrists-0.007
chard-0.007
 blonde-0.006
 Eva-0.006
cious-0.006
Token end
Token ID886
Feature activation+3.84e+00

Pos logit contributions
 Eva+0.001
 wrists+0.001
 Robert+0.001
chard+0.001
 blonde+0.001
Neg logit contributions
loo-0.001
 of-0.001
ick-0.001
iren-0.001
auc-0.001
Token.
Token ID13
Feature activation+2.20e-01

Pos logit contributions
loo+0.136
arl+0.119
auc+0.119
iren+0.117
ick+0.114
Neg logit contributions
 Eva-0.166
 Robert-0.163
 blonde-0.160
 aloud-0.154
rice-0.153
Token he
Token ID339
Feature activation+5.64e-01

Pos logit contributions
loo+0.007
 of+0.006
ick+0.006
iren+0.006
auc+0.006
Neg logit contributions
 wrists-0.009
 Eva-0.009
chard-0.009
 Robert-0.009
 blonde-0.009
Token finally
Token ID3443
Feature activation+3.65e-01

Pos logit contributions
loo+0.021
 of+0.020
ick+0.018
pour+0.017
auc+0.017
Neg logit contributions
 wrists-0.025
chard-0.024
 Robert-0.024
 blonde-0.024
 Eva-0.023
Token reached
Token ID4251
Feature activation-6.28e-01

Pos logit contributions
loo+0.015
 of+0.014
ick+0.013
iren+0.013
auc+0.013
Neg logit contributions
 wrists-0.015
chard-0.015
 Robert-0.014
 Eva-0.014
 blonde-0.014
Token his
Token ID465
Feature activation-3.65e-01

Pos logit contributions
 wrists+0.030
chard+0.030
 Eva+0.030
 Robert+0.030
 blonde+0.029
Neg logit contributions
loo-0.021
 of-0.018
ick-0.017
iren-0.016
auc-0.016
Token cozy
Token ID37438
Feature activation+1.60e+00

Pos logit contributions
chard+0.017
 Robert+0.017
 Eva+0.017
 wrists+0.016
 blonde+0.016
Neg logit contributions
loo-0.012
 of-0.011
 ig-0.010
ick-0.010
 tunnel-0.010
Token bur
Token ID4356
Feature activation+7.23e+00

Pos logit contributions
loo+0.051
auc+0.045
iren+0.045
ick+0.042
 of+0.041
Neg logit contributions
 wrists-0.078
 Eva-0.077
 Robert-0.077
chard-0.076
 blonde-0.075
Tokenrow
Token ID808
Feature activation+4.58e+00

Pos logit contributions
loo+0.154
auc+0.120
iren+0.118
pour+0.111
ick+0.103
Neg logit contributions
chard-0.400
 wrists-0.392
 Eva-0.382
 blonde-0.376
 Robert-0.376
Token.
Token ID13
Feature activation+1.31e-01

Pos logit contributions
loo+0.150
iren+0.126
ick+0.122
auc+0.121
arl+0.117
Neg logit contributions
chard-0.219
 Robert-0.218
 Eva-0.216
 wrists-0.211
 blonde-0.205
Token From
Token ID3574
Feature activation-2.68e-01

Pos logit contributions
loo+0.004
 of+0.004
iren+0.004
ick+0.004
 late+0.004
Neg logit contributions
 wrists-0.006
chard-0.006
cious-0.005
 blonde-0.005
nam-0.005
Token then
Token ID788
Feature activation-1.30e+00

Pos logit contributions
 Eva+0.012
 wrists+0.012
chard+0.012
 Robert+0.012
 blonde+0.011
Neg logit contributions
loo-0.010
 of-0.009
ick-0.008
iren-0.008
auc-0.008
Token on
Token ID319
Feature activation-2.14e+00

Pos logit contributions
 Robert+0.060
 Eva+0.059
 blonde+0.057
 yes+0.056
 wrists+0.056
Neg logit contributions
loo-0.045
auc-0.041
rim-0.038
 outage-0.037
iren-0.036
Token Benny
Token ID44275
Feature activation-9.46e-01

Pos logit contributions
loo+0.022
 of+0.021
ick+0.020
auc+0.019
iren+0.019
Neg logit contributions
 wrists-0.025
 Eva-0.025
chard-0.024
 Robert-0.023
cious-0.023
Token lived
Token ID5615
Feature activation+8.93e-01

Pos logit contributions
chard+0.038
 Eva+0.038
 wrists+0.038
 Robert+0.037
 blonde+0.035
Neg logit contributions
loo-0.036
 of-0.035
ick-0.033
 ig-0.032
 tunnel-0.031
Token in
Token ID287
Feature activation-2.45e+00

Pos logit contributions
loo+0.030
 of+0.025
ick+0.025
iren+0.024
auc+0.024
Neg logit contributions
chard-0.043
 Eva-0.043
 Robert-0.042
 wrists-0.042
 blonde-0.041
Token a
Token ID257
Feature activation-8.74e-01

Pos logit contributions
 Robert+0.083
 Eva+0.081
chard+0.078
 blonde+0.076
 Claire+0.074
Neg logit contributions
loo-0.122
auc-0.113
iren-0.109
arl-0.107
row-0.104
Token cozy
Token ID37438
Feature activation+1.73e+00

Pos logit contributions
chard+0.038
 Eva+0.037
 Robert+0.037
 wrists+0.036
 blonde+0.036
Neg logit contributions
loo-0.034
iren-0.029
auc-0.029
 of-0.029
ick-0.029
Token bur
Token ID4356
Feature activation+7.23e+00

Pos logit contributions
loo+0.060
auc+0.058
iren+0.056
ick+0.049
arl+0.049
Neg logit contributions
chard-0.079
 Robert-0.078
 wrists-0.077
 Eva-0.077
 Claire-0.076
Tokenrow
Token ID808
Feature activation+4.65e+00

Pos logit contributions
loo+0.148
auc+0.139
iren+0.128
pour+0.126
ick+0.111
Neg logit contributions
chard-0.405
 wrists-0.375
 blonde-0.370
cious-0.359
ya-0.355
Token with
Token ID351
Feature activation-2.08e+00

Pos logit contributions
loo+0.161
iren+0.139
auc+0.134
ick+0.131
arl+0.129
Neg logit contributions
chard-0.219
 Robert-0.210
 Eva-0.208
 wrists-0.205
 blonde-0.203
Token his
Token ID465
Feature activation+5.86e-02

Pos logit contributions
 wrists+0.096
 Eva+0.095
 Robert+0.093
chard+0.093
rice+0.089
Neg logit contributions
loo-0.075
 of-0.068
ick-0.064
auc-0.060
iren-0.060
Token family
Token ID1641
Feature activation+2.31e+00

Pos logit contributions
loo+0.002
iren+0.002
auc+0.002
 of+0.002
ick+0.002
Neg logit contributions
chard-0.003
 wrists-0.003
 Robert-0.003
 Eva-0.003
 blonde-0.003
Token.
Token ID13
Feature activation+6.97e-01

Pos logit contributions
loo+0.076
iren+0.064
auc+0.063
ick+0.061
arl+0.060
Neg logit contributions
chard-0.113
 Robert-0.112
 wrists-0.111
 Eva-0.110
 blonde-0.107
Token was
Token ID373
Feature activation+6.88e-01

Pos logit contributions
loo+0.214
ick+0.183
iren+0.183
auc+0.181
 of+0.179
Neg logit contributions
 Eva-0.200
 wrists-0.199
 Robert-0.197
chard-0.197
 blonde-0.191
Token too
Token ID1165
Feature activation+1.47e+00

Pos logit contributions
loo+0.029
auc+0.024
iren+0.023
ick+0.023
arl+0.022
Neg logit contributions
chard-0.029
 Eva-0.028
 Robert-0.028
 blonde-0.027
 wrists-0.027
Token strong
Token ID1913
Feature activation+1.05e+00

Pos logit contributions
loo+0.060
auc+0.048
iren+0.048
 ig+0.045
rim+0.044
Neg logit contributions
chard-0.061
 Eva-0.058
 blonde-0.058
 Claire-0.056
 Robert-0.056
Token and
Token ID290
Feature activation-2.81e-01

Pos logit contributions
loo+0.045
auc+0.036
rim+0.036
pour+0.035
iren+0.035
Neg logit contributions
 Robert-0.044
 Eva-0.041
 Stella-0.039
 Claire-0.038
 blonde-0.038
Token the
Token ID262
Feature activation+3.28e-01

Pos logit contributions
chard+0.012
 Eva+0.012
 Robert+0.012
 wrists+0.012
 blonde+0.011
Neg logit contributions
loo-0.011
ick-0.010
 of-0.009
iren-0.009
auc-0.009
Token bur
Token ID4356
Feature activation+7.22e+00

Pos logit contributions
loo+0.009
 of+0.008
ick+0.007
iren+0.007
auc+0.007
Neg logit contributions
 Eva-0.017
 Robert-0.017
chard-0.017
 wrists-0.017
 blonde-0.017
Tokenrow
Token ID808
Feature activation+4.66e+00

Pos logit contributions
loo+0.202
iren+0.160
 ig+0.158
auc+0.157
pour+0.150
Neg logit contributions
 wrists-0.333
atively-0.332
chard-0.327
rice-0.321
cious-0.319
Token started
Token ID2067
Feature activation+1.52e+00

Pos logit contributions
loo+0.188
iren+0.160
ick+0.157
auc+0.156
 of+0.155
Neg logit contributions
chard-0.185
 wrists-0.185
 Eva-0.184
 Robert-0.183
 blonde-0.179
Token to
Token ID284
Feature activation-1.73e-01

Pos logit contributions
loo+0.060
iren+0.048
auc+0.048
ick+0.047
arl+0.046
Neg logit contributions
 wrists-0.066
chard-0.065
 Eva-0.064
 blonde-0.064
 Robert-0.063
Token fill
Token ID6070
Feature activation+3.58e-01

Pos logit contributions
chard+0.008
 Eva+0.008
 wrists+0.008
 Robert+0.008
 blonde+0.008
Neg logit contributions
loo-0.006
iren-0.005
auc-0.005
 of-0.005
ick-0.005
Token with
Token ID351
Feature activation-2.19e+00

Pos logit contributions
loo+0.012
 of+0.011
ick+0.011
auc+0.010
pour+0.010
Neg logit contributions
chard-0.016
 Robert-0.016
 Eva-0.016
 wrists-0.016
 blonde-0.016
Token from
Token ID422
Feature activation-5.14e-01

Pos logit contributions
loo+0.071
arl+0.056
pour+0.055
iren+0.055
ick+0.053
Neg logit contributions
 yes-0.079
 wrists-0.074
chard-0.074
 Robert-0.073
 Eva-0.071
Token the
Token ID262
Feature activation+4.23e-01

Pos logit contributions
chard+0.023
 Robert+0.023
 Eva+0.023
 wrists+0.022
 blonde+0.021
Neg logit contributions
loo-0.019
 of-0.016
ick-0.016
auc-0.016
iren-0.016
Token river
Token ID7850
Feature activation+1.19e+00

Pos logit contributions
loo+0.014
 of+0.012
ick+0.012
iren+0.011
auc+0.011
Neg logit contributions
chard-0.020
 Robert-0.020
 Eva-0.020
 wrists-0.020
 blonde-0.019
Token to
Token ID284
Feature activation-1.57e-01

Pos logit contributions
loo+0.040
ick+0.033
iren+0.032
auc+0.031
arl+0.031
Neg logit contributions
chard-0.058
 Robert-0.056
 Eva-0.056
 wrists-0.056
 Claire-0.053
Token their
Token ID511
Feature activation-3.28e-02

Pos logit contributions
chard+0.006
 Eva+0.006
 Robert+0.006
 Claire+0.005
 wrists+0.005
Neg logit contributions
loo-0.007
iren-0.006
auc-0.006
ick-0.006
arl-0.006
Token bur
Token ID4356
Feature activation+7.22e+00

Pos logit contributions
chard+0.002
 wrists+0.002
 Robert+0.002
 Eva+0.002
 blonde+0.002
Neg logit contributions
loo-0.001
 of-0.001
ick-0.001
iren-0.001
auc-0.001
Tokenrow
Token ID808
Feature activation+4.82e+00

Pos logit contributions
loo+0.174
pour+0.149
 accidents+0.148
auc+0.141
iren+0.135
Neg logit contributions
chard-0.341
 wrists-0.320
rice-0.300
atively-0.299
ya-0.299
Token.
Token ID13
Feature activation+1.20e-01

Pos logit contributions
loo+0.163
iren+0.137
ick+0.132
arl+0.129
auc+0.129
Neg logit contributions
chard-0.229
 Robert-0.219
 Eva-0.217
 wrists-0.214
 Claire-0.208
Token
Token ID220
Feature activation-2.47e+00

Pos logit contributions
loo+0.004
 of+0.004
iren+0.004
ick+0.004
pour+0.004
Neg logit contributions
 wrists-0.005
chard-0.005
cious-0.005
 Eva-0.005
 blonde-0.005
Token
Token ID198
Feature activation+1.96e-01

Pos logit contributions
chard+0.106
 wrists+0.106
 Eva+0.106
 Robert+0.106
 blonde+0.102
Neg logit contributions
loo-0.094
 of-0.083
ick-0.081
iren-0.079
auc-0.078
Token
Token ID198
Feature activation+2.03e-01

Pos logit contributions
loo+0.007
 of+0.006
iren+0.006
ick+0.006
auc+0.005
Neg logit contributions
 wrists-0.009
chard-0.009
 Eva-0.009
 Robert-0.009
 blonde-0.009
Token he
Token ID339
Feature activation+1.02e+00

Pos logit contributions
 Robert+0.072
 Eva+0.072
chard+0.072
 wrists+0.071
 blonde+0.068
Neg logit contributions
loo-0.063
 of-0.054
ick-0.054
iren-0.053
auc-0.053
Token could
Token ID714
Feature activation-3.15e-01

Pos logit contributions
 of+0.032
loo+0.031
ick+0.029
row+0.026
pour+0.026
Neg logit contributions
 Robert-0.050
 Eva-0.050
 wrists-0.049
chard-0.049
 Claire-0.047
Token back
Token ID736
Feature activation-1.26e+00

Pos logit contributions
 wrists+0.017
chard+0.017
 Robert+0.017
 Eva+0.017
 blonde+0.016
Neg logit contributions
loo-0.009
 of-0.008
ick-0.007
iren-0.007
auc-0.007
Token to
Token ID284
Feature activation-1.76e-01

Pos logit contributions
 Eva+0.048
chard+0.048
 Robert+0.048
 wrists+0.046
 Joy+0.045
Neg logit contributions
loo-0.049
auc-0.046
rim-0.045
iren-0.045
ick-0.044
Token his
Token ID465
Feature activation-1.22e-02

Pos logit contributions
chard+0.008
 wrists+0.008
 Robert+0.008
 Eva+0.008
 blonde+0.008
Neg logit contributions
loo-0.006
 of-0.005
ick-0.005
iren-0.005
 ig-0.005
Token bur
Token ID4356
Feature activation+7.18e+00

Pos logit contributions
 Eva+0.001
 Robert+0.001
chard+0.001
 wrists+0.001
 blonde+0.001
Neg logit contributions
loo-0.000
 of-0.000
ick-0.000
iren-0.000
auc-0.000
Tokenrow
Token ID808
Feature activation+4.56e+00

Pos logit contributions
loo+0.167
auc+0.135
iren+0.125
pour+0.122
ick+0.119
Neg logit contributions
chard-0.367
 wrists-0.359
 blonde-0.347
 Eva-0.342
rice-0.341
Token.
Token ID13
Feature activation+5.30e-01

Pos logit contributions
loo+0.160
iren+0.135
auc+0.131
arl+0.129
ick+0.129
Neg logit contributions
chard-0.208
 Robert-0.204
 Eva-0.202
 wrists-0.198
 blonde-0.195
Token
Token ID198
Feature activation+4.02e-01

Pos logit contributions
loo+0.018
 of+0.017
ick+0.016
iren+0.016
 late+0.015
Neg logit contributions
 wrists-0.026
chard-0.024
 Eva-0.023
 blonde-0.023
cious-0.022
Token
Token ID198
Feature activation+3.66e-01

Pos logit contributions
loo+0.014
 of+0.013
iren+0.012
ick+0.012
auc+0.011
Neg logit contributions
 wrists-0.019
chard-0.019
 Eva-0.018
 Robert-0.018
 blonde-0.018
TokenWhen
Token ID2215
Feature activation-3.20e-01

Pos logit contributions
loo+0.016
 of+0.015
ick+0.014
iren+0.014
auc+0.014
Neg logit contributions
 Eva-0.013
chard-0.013
 Robert-0.013
 wrists-0.013
 blonde-0.013
Token got
Token ID1392
Feature activation+2.12e+00

Pos logit contributions
loo+0.028
 of+0.026
ick+0.025
auc+0.024
iren+0.024
Neg logit contributions
chard-0.041
 Robert-0.040
 Eva-0.040
 wrists-0.039
 blonde-0.038
Token m
Token ID285
Feature activation+6.71e-01

Pos logit contributions
loo+0.076
 of+0.067
ick+0.065
iren+0.064
auc+0.064
Neg logit contributions
 Eva-0.096
 Robert-0.095
chard-0.095
 wrists-0.095
 blonde-0.091
Token-
Token ID12
Feature activation+1.38e+00

Pos logit contributions
loo+0.021
 tunnel+0.017
pour+0.016
auc+0.016
 of+0.016
Neg logit contributions
 wrists-0.032
 blonde-0.032
 Eva-0.032
cious-0.031
 Robert-0.031
Tokenmad
Token ID9937
Feature activation-4.04e-01

Pos logit contributions
loo+0.046
 of+0.041
iren+0.040
ick+0.039
auc+0.039
Neg logit contributions
 Robert-0.065
 wrists-0.065
 Eva-0.065
chard-0.064
 blonde-0.062
Token and
Token ID290
Feature activation-6.76e-02

Pos logit contributions
 Eva+0.019
 wrists+0.019
 Robert+0.019
 blonde+0.019
chard+0.018
Neg logit contributions
loo-0.014
ick-0.011
auc-0.011
iren-0.011
 of-0.011
Token r
Token ID374
Feature activation+7.18e+00

Pos logit contributions
chard+0.003
 Robert+0.003
 Eva+0.003
 wrists+0.003
 blonde+0.003
Neg logit contributions
loo-0.002
 of-0.002
iren-0.002
ick-0.002
auc-0.002
Token-
Token ID12
Feature activation+1.26e+00

Pos logit contributions
loo+0.220
 of+0.201
 tunnel+0.197
 storm+0.188
 outage+0.185
Neg logit contributions
 wrists-0.334
 blonde-0.322
 Eva-0.314
chard-0.307
 blond-0.302
Tokenran
Token ID2596
Feature activation-1.14e+00

Pos logit contributions
loo+0.042
 of+0.036
iren+0.036
ick+0.035
auc+0.035
Neg logit contributions
 Robert-0.060
 Eva-0.060
 wrists-0.059
chard-0.059
 blonde-0.057
Token away
Token ID1497
Feature activation+4.62e-01

Pos logit contributions
chard+0.056
 Eva+0.054
 Robert+0.054
 wrists+0.053
 Claire+0.051
Neg logit contributions
loo-0.036
 of-0.033
ick-0.033
iren-0.032
auc-0.030
Token and
Token ID290
Feature activation-2.05e-01

Pos logit contributions
loo+0.015
iren+0.012
auc+0.012
ick+0.012
 of+0.011
Neg logit contributions
 Eva-0.022
 wrists-0.022
 Robert-0.022
chard-0.022
 blonde-0.021
Token I
Token ID314
Feature activation-1.04e+00

Pos logit contributions
chard+0.008
 Eva+0.008
 Robert+0.008
 wrists+0.008
 blonde+0.008
Neg logit contributions
loo-0.008
 of-0.007
ick-0.007
iren-0.007
auc-0.007
Token fluffy
Token ID39145
Feature activation+2.52e-01

Pos logit contributions
loo+0.001
 of+0.001
ick+0.001
iren+0.001
auc+0.001
Neg logit contributions
 Eva-0.001
 Robert-0.001
 wrists-0.001
chard-0.001
 blonde-0.001
Token and
Token ID290
Feature activation-1.86e-02

Pos logit contributions
loo+0.006
 of+0.005
ick+0.005
 ig+0.004
iren+0.004
Neg logit contributions
 Eva-0.014
chard-0.014
 wrists-0.014
 Robert-0.014
 blonde-0.013
Token lived
Token ID5615
Feature activation+8.71e-01

Pos logit contributions
 Eva+0.001
 wrists+0.001
 Robert+0.001
chard+0.001
 blonde+0.001
Neg logit contributions
loo-0.001
 of-0.001
ick-0.001
 ig-0.001
iren-0.001
Token in
Token ID287
Feature activation-2.52e+00

Pos logit contributions
loo+0.025
 of+0.022
ick+0.021
iren+0.020
auc+0.019
Neg logit contributions
 wrists-0.045
 Eva-0.045
 Robert-0.045
chard-0.044
 blonde-0.043
Token a
Token ID257
Feature activation-9.00e-01

Pos logit contributions
 Robert+0.088
 Eva+0.088
chard+0.086
 blonde+0.083
 wrists+0.081
Neg logit contributions
loo-0.119
auc-0.107
iren-0.104
ick-0.103
arl-0.102
Token bur
Token ID4356
Feature activation+7.18e+00

Pos logit contributions
chard+0.039
 Eva+0.039
 Robert+0.038
 wrists+0.038
 blonde+0.037
Neg logit contributions
loo-0.034
auc-0.029
ick-0.029
iren-0.029
 of-0.029
Tokenrow
Token ID808
Feature activation+4.74e+00

Pos logit contributions
loo+0.163
auc+0.154
pour+0.141
 accidents+0.139
iren+0.134
Neg logit contributions
chard-0.367
 wrists-0.341
 blonde-0.337
cious-0.337
ya-0.331
Token in
Token ID287
Feature activation-2.46e+00

Pos logit contributions
loo+0.159
iren+0.136
auc+0.130
ick+0.128
arl+0.128
Neg logit contributions
chard-0.229
 Robert-0.216
 Eva-0.214
 wrists-0.213
 blonde-0.209
Token the
Token ID262
Feature activation+6.70e-01

Pos logit contributions
 Eva+0.089
 Robert+0.089
chard+0.088
 Claire+0.083
 blonde+0.082
Neg logit contributions
loo-0.111
auc-0.104
iren-0.101
arl-0.100
ick-0.098
Token forest
Token ID8222
Feature activation+9.07e-01

Pos logit contributions
loo+0.025
ick+0.022
iren+0.021
 of+0.021
auc+0.021
Neg logit contributions
chard-0.030
 Eva-0.028
 blonde-0.028
 Robert-0.028
 wrists-0.027
Token.
Token ID13
Feature activation+8.07e-01

Pos logit contributions
loo+0.027
ick+0.022
iren+0.022
 of+0.022
auc+0.022
Neg logit contributions
chard-0.046
 Eva-0.046
 Robert-0.046
 wrists-0.045
 blonde-0.044
Token who
Token ID508
Feature activation-4.60e+00

Pos logit contributions
 Eva+0.108
chard+0.108
 Robert+0.108
 wrists+0.107
 blonde+0.103
Neg logit contributions
loo-0.101
 of-0.089
ick-0.087
iren-0.086
auc-0.085
Token lived
Token ID5615
Feature activation+7.54e-01

Pos logit contributions
 wrists+0.181
chard+0.181
 Eva+0.178
 Robert+0.178
 blonde+0.163
Neg logit contributions
loo-0.188
 of-0.180
ick-0.174
auc-0.162
iren-0.158
Token in
Token ID287
Feature activation-2.39e+00

Pos logit contributions
loo+0.023
 of+0.020
ick+0.019
iren+0.018
auc+0.018
Neg logit contributions
 Eva-0.038
 Robert-0.038
 wrists-0.038
chard-0.038
 blonde-0.037
Token the
Token ID262
Feature activation+7.18e-01

Pos logit contributions
 wrists+0.103
chard+0.102
 Robert+0.100
 Eva+0.100
 blonde+0.096
Neg logit contributions
loo-0.088
 of-0.081
ick-0.079
iren-0.074
auc-0.073
Token same
Token ID976
Feature activation+6.19e-01

Pos logit contributions
loo+0.023
 of+0.020
ick+0.019
iren+0.018
auc+0.018
Neg logit contributions
 wrists-0.035
 Eva-0.035
 Robert-0.035
chard-0.035
 blonde-0.033
Token bur
Token ID4356
Feature activation+7.17e+00

Pos logit contributions
loo+0.023
ick+0.019
 of+0.019
iren+0.019
auc+0.019
Neg logit contributions
 Eva-0.027
chard-0.027
 Robert-0.027
 wrists-0.027
 blonde-0.026
Tokenrow
Token ID808
Feature activation+4.68e+00

Pos logit contributions
loo+0.159
pour+0.136
auc+0.131
ick+0.127
iren+0.122
Neg logit contributions
chard-0.382
 wrists-0.379
cious-0.364
 blonde-0.352
ya-0.351
Token.
Token ID13
Feature activation+9.82e-01

Pos logit contributions
loo+0.142
iren+0.116
ick+0.115
auc+0.112
 of+0.111
Neg logit contributions
chard-0.237
 Robert-0.233
 Eva-0.233
 wrists-0.233
 blonde-0.224
Token They
Token ID1119
Feature activation+2.53e-01

Pos logit contributions
 of+0.032
loo+0.032
 ig+0.028
ick+0.028
 late+0.027
Neg logit contributions
 wrists-0.044
cious-0.041
chard-0.041
rice-0.041
 yes-0.040
Token were
Token ID547
Feature activation+1.08e+00

Pos logit contributions
 of+0.009
loo+0.008
ick+0.008
 ig+0.007
ful+0.007
Neg logit contributions
 Eva-0.011
 wrists-0.011
 Robert-0.011
chard-0.011
 blonde-0.010
Token very
Token ID845
Feature activation-2.34e+00

Pos logit contributions
loo+0.037
 of+0.036
ick+0.034
 ig+0.031
iren+0.030
Neg logit contributions
 wrists-0.048
chard-0.047
 Eva-0.047
 Robert-0.047
 blonde-0.045
Token it
Token ID340
Feature activation+8.64e-02

Pos logit contributions
loo+0.148
 of+0.130
ick+0.128
iren+0.126
auc+0.125
Neg logit contributions
chard-0.147
 Eva-0.146
 Robert-0.145
 wrists-0.145
 blonde-0.139
Token to
Token ID284
Feature activation-4.00e-01

Pos logit contributions
loo+0.003
iren+0.003
ick+0.003
auc+0.003
arl+0.003
Neg logit contributions
 Robert-0.004
 Eva-0.004
chard-0.004
 wrists-0.004
 blonde-0.004
Token his
Token ID465
Feature activation-4.50e-01

Pos logit contributions
 wrists+0.019
chard+0.019
 Eva+0.018
 Robert+0.018
 blonde+0.018
Neg logit contributions
loo-0.013
 of-0.012
ick-0.011
iren-0.011
auc-0.011
Token friends
Token ID2460
Feature activation-8.54e-01

Pos logit contributions
chard+0.021
 Eva+0.020
 Robert+0.020
 wrists+0.020
 blonde+0.020
Neg logit contributions
loo-0.015
 of-0.014
 ig-0.013
ick-0.013
 tunnel-0.012
Token'
Token ID6
Feature activation-1.87e-01

Pos logit contributions
 Eva+0.047
 wrists+0.046
chard+0.046
 Robert+0.046
 blonde+0.045
Neg logit contributions
loo-0.021
 of-0.020
ick-0.018
auc-0.017
iren-0.016
Token bur
Token ID4356
Feature activation+7.16e+00

Pos logit contributions
 Eva+0.009
chard+0.009
 Robert+0.009
 wrists+0.009
 blonde+0.009
Neg logit contributions
loo-0.006
 of-0.005
ick-0.005
iren-0.005
auc-0.005
Tokenrows
Token ID8516
Feature activation+7.16e-01

Pos logit contributions
loo+0.161
auc+0.125
iren+0.125
pour+0.123
ick+0.111
Neg logit contributions
chard-0.389
 wrists-0.369
 Robert-0.364
 Eva-0.363
rice-0.359
Token.
Token ID13
Feature activation+1.74e-01

Pos logit contributions
loo+0.022
ick+0.018
 of+0.018
iren+0.018
auc+0.018
Neg logit contributions
 Robert-0.036
 Eva-0.036
chard-0.036
 wrists-0.035
 blonde-0.034
Token They
Token ID1119
Feature activation-1.26e-01

Pos logit contributions
loo+0.006
 of+0.006
ick+0.005
iren+0.005
auc+0.005
Neg logit contributions
 wrists-0.008
chard-0.008
 blonde-0.007
cious-0.007
 Eva-0.007
Token all
Token ID477
Feature activation+4.56e+00

Pos logit contributions
 Eva+0.006
 wrists+0.006
 blonde+0.006
chard+0.006
 Robert+0.006
Neg logit contributions
loo-0.004
 of-0.004
ick-0.004
pour-0.003
 ig-0.003
Token thought
Token ID1807
Feature activation+3.84e-01

Pos logit contributions
loo+0.209
 of+0.193
ick+0.185
auc+0.181
iren+0.180
Neg logit contributions
chard-0.159
 Eva-0.158
 wrists-0.158
 Robert-0.157
 blonde-0.149
Token ate
Token ID15063
Feature activation+1.19e+00

Pos logit contributions
 Robert+0.083
 Eva+0.082
chard+0.079
 wrists+0.078
 Claire+0.076
Neg logit contributions
loo-0.067
auc-0.058
iren-0.057
ick-0.057
arl-0.056
Token their
Token ID511
Feature activation+1.62e-01

Pos logit contributions
loo+0.038
 of+0.035
ick+0.031
auc+0.031
iren+0.030
Neg logit contributions
 wrists-0.057
chard-0.057
 blonde-0.057
 Robert-0.057
 Eva-0.057
Token popcorn
Token ID38914
Feature activation-4.61e+00

Pos logit contributions
loo+0.006
 of+0.005
 ig+0.005
ick+0.005
auc+0.005
Neg logit contributions
 Robert-0.007
 Eva-0.007
chard-0.007
 blonde-0.007
 wrists-0.007
Token while
Token ID981
Feature activation-1.97e+00

Pos logit contributions
 Eva+0.227
chard+0.227
 Robert+0.226
 wrists+0.225
 blonde+0.217
Neg logit contributions
loo-0.146
 of-0.122
ick-0.121
iren-0.118
auc-0.117
Token it
Token ID340
Feature activation+6.01e-01

Pos logit contributions
 Eva+0.087
 Robert+0.087
chard+0.087
 wrists+0.086
 blonde+0.082
Neg logit contributions
loo-0.073
 of-0.062
ick-0.062
iren-0.062
auc-0.062
Token r
Token ID374
Feature activation+7.16e+00

Pos logit contributions
loo+0.022
 of+0.019
ick+0.018
auc+0.018
iren+0.018
Neg logit contributions
chard-0.027
 Robert-0.027
 Eva-0.027
 wrists-0.027
 blonde-0.026
Tokenained
Token ID1328
Feature activation+4.13e+00

Pos logit contributions
loo+0.151
 of+0.146
 tunnel+0.143
 outage+0.138
 ig+0.120
Neg logit contributions
chard-0.378
 wrists-0.366
 Eva-0.351
 blonde-0.345
 Robert-0.343
Token outside
Token ID2354
Feature activation-1.51e+00

Pos logit contributions
loo+0.150
auc+0.123
iren+0.123
ick+0.121
rim+0.121
Neg logit contributions
 Robert-0.176
chard-0.176
 Eva-0.175
 Claire-0.170
 wrists-0.166
Token.
Token ID13
Feature activation+5.15e-01

Pos logit contributions
 Eva+0.066
 Robert+0.065
 wrists+0.064
 Claire+0.064
chard+0.064
Neg logit contributions
loo-0.055
iren-0.047
arl-0.046
ick-0.045
auc-0.045
Token Lily
Token ID20037
Feature activation+1.66e-01

Pos logit contributions
loo+0.018
 of+0.018
auc+0.016
iren+0.016
ick+0.016
Neg logit contributions
chard-0.023
 wrists-0.023
cious-0.021
 Eva-0.021
 Robert-0.021
Token was
Token ID373
Feature activation+5.47e-01

Pos logit contributions
loo+0.006
 of+0.006
ick+0.005
auc+0.005
iren+0.005
Neg logit contributions
chard-0.008
 Robert-0.007
 wrists-0.007
 Eva-0.007
 blonde-0.007
Token to
Token ID284
Feature activation-3.23e-01

Pos logit contributions
loo+0.023
ick+0.019
 of+0.019
iren+0.018
auc+0.018
Neg logit contributions
chard-0.034
 Eva-0.034
 Robert-0.034
 wrists-0.034
 blonde-0.033
Token share
Token ID2648
Feature activation-4.62e+00

Pos logit contributions
 Eva+0.013
 wrists+0.013
 Robert+0.013
chard+0.013
 blonde+0.013
Neg logit contributions
loo-0.013
auc-0.011
ick-0.011
iren-0.011
 of-0.011
Token and
Token ID290
Feature activation-2.04e-01

Pos logit contributions
 Eva+0.245
 Robert+0.244
chard+0.244
 wrists+0.240
 blonde+0.238
Neg logit contributions
loo-0.124
 of-0.113
ick-0.100
iren-0.098
auc-0.098
Token wait
Token ID4043
Feature activation-1.75e-01

Pos logit contributions
 wrists+0.009
 Eva+0.009
chard+0.009
 Robert+0.009
 blonde+0.008
Neg logit contributions
loo-0.008
ick-0.007
iren-0.007
auc-0.007
 of-0.006
Token.
Token ID13
Feature activation+7.94e-01

Pos logit contributions
chard+0.010
 wrists+0.010
 Robert+0.010
 Eva+0.010
 blonde+0.010
Neg logit contributions
loo-0.004
 of-0.004
ick-0.003
auc-0.003
arl-0.003
Token How
Token ID1374
Feature activation-9.76e+00

Pos logit contributions
loo+0.034
 of+0.030
ick+0.030
iren+0.030
auc+0.029
Neg logit contributions
chard-0.030
 Eva-0.030
 Robert-0.030
 wrists-0.030
 blonde-0.028
Token about
Token ID546
Feature activation-2.54e+00

Pos logit contributions
 Eva+0.376
 Robert+0.374
chard+0.374
 wrists+0.369
 blonde+0.355
Neg logit contributions
loo-0.414
ick-0.360
 of-0.359
iren-0.359
auc-0.352
Token we
Token ID356
Feature activation+1.06e+00

Pos logit contributions
 Eva+0.121
chard+0.121
 Robert+0.121
 wrists+0.120
 blonde+0.116
Neg logit contributions
loo-0.085
 of-0.073
ick-0.071
iren-0.070
auc-0.070
Token clean
Token ID3424
Feature activation+3.45e+00

Pos logit contributions
loo+0.043
iren+0.037
auc+0.037
ick+0.036
 of+0.035
Neg logit contributions
chard-0.045
 Robert-0.044
 Eva-0.044
 wrists-0.043
 Claire-0.041
Token up
Token ID510
Feature activation-5.84e+00

Pos logit contributions
loo+0.139
 of+0.121
ick+0.120
iren+0.118
auc+0.117
Neg logit contributions
chard-0.141
 Eva-0.141
 Robert-0.141
 wrists-0.139
 blonde-0.135
Token this
Token ID428
Feature activation-7.08e-02

Pos logit contributions
 Eva+0.275
chard+0.273
 Robert+0.272
 wrists+0.272
 Claire+0.258
Neg logit contributions
loo-0.200
ick-0.166
iren-0.165
auc-0.160
 of-0.158
Token play
Token ID711
Feature activation-3.78e-01

Pos logit contributions
 Robert+0.018
 Eva+0.018
 wrists+0.017
chard+0.017
 Claire+0.017
Neg logit contributions
loo-0.018
auc-0.015
iren-0.015
ick-0.015
pour-0.014
Token with
Token ID351
Feature activation-2.59e+00

Pos logit contributions
chard+0.020
 Eva+0.020
 Robert+0.020
 wrists+0.020
 blonde+0.020
Neg logit contributions
loo-0.010
 of-0.009
ick-0.008
auc-0.008
iren-0.007
Token other
Token ID584
Feature activation-1.40e-01

Pos logit contributions
chard+0.139
 Eva+0.138
 Robert+0.138
 wrists+0.137
 blonde+0.135
Neg logit contributions
loo-0.068
 of-0.060
ick-0.055
auc-0.053
iren-0.052
Token toys
Token ID14958
Feature activation-6.45e-01

Pos logit contributions
chard+0.007
 Eva+0.007
 Robert+0.007
 wrists+0.007
 blonde+0.007
Neg logit contributions
loo-0.004
 of-0.004
ick-0.003
auc-0.003
iren-0.003
Token.
Token ID13
Feature activation+8.37e-01

Pos logit contributions
chard+0.031
 Robert+0.031
 Eva+0.031
 wrists+0.031
 Claire+0.029
Neg logit contributions
loo-0.021
iren-0.017
ick-0.017
auc-0.017
 of-0.017
Token How
Token ID1374
Feature activation-9.75e+00

Pos logit contributions
loo+0.038
 of+0.035
ick+0.034
auc+0.033
iren+0.033
Neg logit contributions
chard-0.029
 wrists-0.029
 Eva-0.028
 Robert-0.028
 blonde-0.027
Token about
Token ID546
Feature activation-2.64e+00

Pos logit contributions
chard+0.450
 wrists+0.447
 Robert+0.439
 Eva+0.435
 blonde+0.420
Neg logit contributions
loo-0.354
 of-0.340
ick-0.303
auc-0.298
arl-0.275
Token this
Token ID428
Feature activation+7.94e-02

Pos logit contributions
 Eva+0.132
chard+0.132
 wrists+0.131
 Robert+0.130
 blonde+0.129
Neg logit contributions
loo-0.078
 of-0.070
ick-0.064
iren-0.064
auc-0.063
Token puzzle
Token ID15027
Feature activation+9.48e-01

Pos logit contributions
loo+0.002
 of+0.002
ick+0.002
iren+0.002
auc+0.002
Neg logit contributions
 Eva-0.004
 Robert-0.004
chard-0.004
 wrists-0.004
 blonde-0.004
Token?"
Token ID1701
Feature activation-3.95e+00

Pos logit contributions
loo+0.027
ick+0.022
iren+0.022
auc+0.021
arl+0.020
Neg logit contributions
chard-0.049
 Robert-0.049
 Eva-0.049
 wrists-0.048
 Claire-0.047
Token She
Token ID1375
Feature activation-6.58e-01

Pos logit contributions
chard+0.200
 wrists+0.199
cious+0.193
 blonde+0.189
 Robert+0.186
Neg logit contributions
 of-0.117
loo-0.114
iren-0.097
ick-0.093
auc-0.091
Token downstairs
Token ID34624
Feature activation-1.77e+00

Pos logit contributions
loo+0.024
auc+0.021
ick+0.021
rim+0.021
iren+0.020
Neg logit contributions
 Robert-0.025
chard-0.025
 wrists-0.025
 Eva-0.024
 Claire-0.023
Token for
Token ID329
Feature activation+1.14e+00

Pos logit contributions
 Eva+0.087
 Robert+0.087
chard+0.087
 wrists+0.086
 blonde+0.082
Neg logit contributions
loo-0.057
iren-0.046
auc-0.045
ick-0.045
 of-0.045
Token a
Token ID257
Feature activation-1.22e+00

Pos logit contributions
loo+0.041
 of+0.036
ick+0.035
iren+0.035
auc+0.034
Neg logit contributions
 Eva-0.051
 Robert-0.051
chard-0.051
 wrists-0.050
 blonde-0.048
Token snack
Token ID26906
Feature activation-8.42e-01

Pos logit contributions
 wrists+0.055
 Eva+0.054
chard+0.054
 Robert+0.054
 blonde+0.052
Neg logit contributions
loo-0.044
 of-0.039
ick-0.038
iren-0.036
auc-0.036
Token.
Token ID13
Feature activation+9.01e-01

Pos logit contributions
 Eva+0.035
 Robert+0.035
chard+0.035
 wrists+0.035
 Claire+0.034
Neg logit contributions
loo-0.031
iren-0.028
ick-0.026
rim-0.026
auc-0.026
Token How
Token ID1374
Feature activation-9.74e+00

Pos logit contributions
loo+0.038
 of+0.034
ick+0.033
iren+0.033
auc+0.033
Neg logit contributions
chard-0.035
 wrists-0.034
 Robert-0.034
 Eva-0.034
 blonde-0.032
Token does
Token ID857
Feature activation-1.11e+00

Pos logit contributions
 wrists+0.427
chard+0.426
 Robert+0.417
 Eva+0.415
 blonde+0.401
Neg logit contributions
loo-0.374
 of-0.350
ick-0.323
auc-0.315
iren-0.295
Token that
Token ID326
Feature activation+3.03e+00

Pos logit contributions
 Robert+0.051
 Eva+0.050
chard+0.048
 wrists+0.047
 Claire+0.046
Neg logit contributions
loo-0.041
auc-0.035
ick-0.035
iren-0.034
arl-0.034
Token sound
Token ID2128
Feature activation+8.88e-01

Pos logit contributions
loo+0.107
ick+0.092
iren+0.090
auc+0.089
arl+0.087
Neg logit contributions
 Eva-0.137
 Robert-0.137
chard-0.134
 yes-0.133
 Claire-0.129
Token?"
Token ID1701
Feature activation-3.74e+00

Pos logit contributions
loo+0.028
ick+0.024
iren+0.022
auc+0.022
 of+0.022
Neg logit contributions
 Eva-0.043
 Robert-0.043
chard-0.043
 wrists-0.042
 blonde-0.041
Token
Token ID198
Feature activation-3.12e-01

Pos logit contributions
 wrists+0.190
chard+0.189
 Robert+0.183
 Eva+0.179
 blonde+0.179
Neg logit contributions
 of-0.110
loo-0.109
ick-0.093
iren-0.092
pour-0.088
Token that
Token ID326
Feature activation+3.06e+00

Pos logit contributions
loo+0.005
 of+0.004
ick+0.004
iren+0.004
auc+0.004
Neg logit contributions
 Eva-0.009
 Robert-0.009
chard-0.009
 wrists-0.009
 blonde-0.008
Token you
Token ID345
Feature activation+1.23e-01

Pos logit contributions
loo+0.104
ick+0.090
iren+0.088
auc+0.085
arl+0.084
Neg logit contributions
 Robert-0.143
 Eva-0.143
chard-0.141
 wrists-0.140
 Claire-0.134
Token both
Token ID1111
Feature activation+1.92e+00

Pos logit contributions
loo+0.004
 of+0.003
ick+0.003
auc+0.003
iren+0.003
Neg logit contributions
chard-0.006
 Robert-0.006
 Eva-0.006
 wrists-0.006
 blonde-0.006
Token like
Token ID588
Feature activation-2.55e+00

Pos logit contributions
loo+0.071
ick+0.060
iren+0.059
auc+0.059
 of+0.057
Neg logit contributions
 Eva-0.086
 Robert-0.085
chard-0.085
 wrists-0.085
 blonde-0.081
Token.
Token ID13
Feature activation+9.34e-01

Pos logit contributions
 Eva+0.129
chard+0.129
 Robert+0.128
 wrists+0.127
 blonde+0.124
Neg logit contributions
loo-0.077
 of-0.067
ick-0.063
iren-0.062
auc-0.061
Token How
Token ID1374
Feature activation-9.74e+00

Pos logit contributions
loo+0.040
 of+0.037
ick+0.036
iren+0.035
auc+0.035
Neg logit contributions
chard-0.035
 wrists-0.034
 Eva-0.033
 Robert-0.033
 blonde-0.032
Token about
Token ID546
Feature activation-2.44e+00

Pos logit contributions
chard+0.447
 wrists+0.441
 Robert+0.437
 Eva+0.433
 blonde+0.419
Neg logit contributions
loo-0.357
 of-0.334
ick-0.307
auc-0.298
arl-0.277
Token a
Token ID257
Feature activation-1.26e+00

Pos logit contributions
 Eva+0.114
 Robert+0.114
chard+0.114
 wrists+0.113
 blonde+0.109
Neg logit contributions
loo-0.084
 of-0.071
ick-0.071
iren-0.069
auc-0.069
Token cartoon
Token ID16251
Feature activation-5.36e+00

Pos logit contributions
 wrists+0.058
chard+0.058
 Eva+0.058
 Robert+0.056
 blonde+0.055
Neg logit contributions
loo-0.044
 of-0.040
ick-0.037
iren-0.036
auc-0.035
Token about
Token ID546
Feature activation-2.44e+00

Pos logit contributions
 Eva+0.271
 Robert+0.270
chard+0.267
 wrists+0.263
 Claire+0.254
Neg logit contributions
loo-0.159
ick-0.134
auc-0.127
iren-0.127
arl-0.123
Token animals
Token ID4695
Feature activation-1.88e+00

Pos logit contributions
 Eva+0.111
chard+0.111
 Robert+0.111
 wrists+0.110
 blonde+0.109
Neg logit contributions
loo-0.085
 of-0.076
ick-0.072
iren-0.069
auc-0.069
Token in
Token ID287
Feature activation-2.64e+00

Pos logit contributions
loo+0.136
iren+0.120
ick+0.119
auc+0.117
rim+0.115
Neg logit contributions
 Robert-0.157
 Eva-0.156
 wrists-0.152
chard-0.151
 Claire-0.148
Token the
Token ID262
Feature activation+2.29e-01

Pos logit contributions
 Robert+0.093
chard+0.093
 Eva+0.089
 wrists+0.088
 Claire+0.086
Neg logit contributions
loo-0.123
auc-0.113
iren-0.111
arl-0.109
rim-0.108
Token living
Token ID2877
Feature activation+1.53e+00

Pos logit contributions
loo+0.008
ick+0.007
 of+0.007
iren+0.007
auc+0.007
Neg logit contributions
chard-0.010
 Robert-0.010
 Eva-0.010
 wrists-0.010
 blonde-0.010
Token room
Token ID2119
Feature activation+2.43e+00

Pos logit contributions
loo+0.047
iren+0.044
ick+0.042
 of+0.042
arl+0.042
Neg logit contributions
 wrists-0.072
 Eva-0.069
chard-0.069
 Robert-0.068
 Claire-0.065
Token.
Token ID13
Feature activation+6.86e-01

Pos logit contributions
loo+0.083
iren+0.070
ick+0.070
auc+0.068
 of+0.067
Neg logit contributions
 Eva-0.113
 Robert-0.112
 wrists-0.112
chard-0.112
 blonde-0.107
Token How
Token ID1374
Feature activation-9.74e+00

Pos logit contributions
loo+0.030
 of+0.026
ick+0.026
iren+0.025
auc+0.025
Neg logit contributions
chard-0.026
 Eva-0.026
 wrists-0.026
 Robert-0.026
 blonde-0.024
Token does
Token ID857
Feature activation-1.21e+00

Pos logit contributions
 wrists+0.442
chard+0.441
 Robert+0.437
 Eva+0.435
 blonde+0.420
Neg logit contributions
loo-0.358
 of-0.329
ick-0.304
auc-0.297
iren-0.280
Token that
Token ID326
Feature activation+3.01e+00

Pos logit contributions
 Robert+0.053
 Eva+0.053
chard+0.053
 wrists+0.051
 blonde+0.049
Neg logit contributions
loo-0.045
auc-0.039
ick-0.039
iren-0.038
arl-0.037
Token sound
Token ID2128
Feature activation+8.54e-01

Pos logit contributions
loo+0.105
ick+0.091
arl+0.090
iren+0.089
auc+0.088
Neg logit contributions
 Robert-0.135
chard-0.135
 Eva-0.134
 yes-0.131
 Claire-0.126
Token?"
Token ID1701
Feature activation-4.36e+00

Pos logit contributions
loo+0.025
ick+0.021
iren+0.020
 of+0.020
auc+0.020
Neg logit contributions
chard-0.044
 Eva-0.044
 Robert-0.044
 wrists-0.043
 blonde-0.041
Token
Token ID198
Feature activation-2.60e-01

Pos logit contributions
 wrists+0.223
chard+0.214
 blonde+0.211
 Eva+0.210
 Robert+0.209
Neg logit contributions
loo-0.134
 of-0.129
iren-0.112
ick-0.109
auc-0.106
Token are
Token ID389
Feature activation+6.33e-01

Pos logit contributions
 Robert+0.001
 Eva+0.001
 wrists+0.001
chard+0.001
 Claire+0.001
Neg logit contributions
loo-0.001
iren-0.001
auc-0.001
ick-0.001
arl-0.001
Token plenty
Token ID6088
Feature activation-1.07e-02

Pos logit contributions
loo+0.024
 of+0.021
ick+0.020
iren+0.020
auc+0.020
Neg logit contributions
 Eva-0.027
 Robert-0.027
chard-0.027
 wrists-0.027
 blonde-0.026
Token of
Token ID286
Feature activation+3.08e+00

Pos logit contributions
 yes+0.001
 Claire+0.001
 Eva+0.001
 Robert+0.001
chard+0.001
Neg logit contributions
loo-0.000
rim-0.000
el-0.000
 ig-0.000
ick-0.000
Token them
Token ID606
Feature activation-3.13e+00

Pos logit contributions
loo+0.116
ick+0.103
auc+0.099
pour+0.099
iren+0.097
Neg logit contributions
 wrists-0.131
 Robert-0.129
 blonde-0.127
 Eva-0.126
chard-0.123
Token.
Token ID13
Feature activation+8.98e-01

Pos logit contributions
chard+0.149
 Robert+0.148
 Eva+0.148
 wrists+0.147
 blonde+0.140
Neg logit contributions
loo-0.104
iren-0.086
ick-0.086
auc-0.085
 of-0.082
Token How
Token ID1374
Feature activation-9.73e+00

Pos logit contributions
loo+0.039
 of+0.035
ick+0.034
iren+0.033
auc+0.033
Neg logit contributions
chard-0.034
 wrists-0.034
 Eva-0.034
 Robert-0.033
 blonde-0.032
Token about
Token ID546
Feature activation-2.48e+00

Pos logit contributions
chard+0.439
 wrists+0.437
 Robert+0.434
 Eva+0.433
 blonde+0.416
Neg logit contributions
loo-0.358
 of-0.329
ick-0.305
auc-0.300
iren-0.285
Token that
Token ID326
Feature activation+2.98e+00

Pos logit contributions
 Eva+0.123
chard+0.122
 Robert+0.122
 wrists+0.121
 blonde+0.119
Neg logit contributions
loo-0.077
 of-0.067
ick-0.064
auc-0.062
iren-0.062
Token?"
Token ID1701
Feature activation-3.87e+00

Pos logit contributions
loo+0.100
ick+0.085
iren+0.083
auc+0.083
arl+0.079
Neg logit contributions
 Robert-0.140
chard-0.139
 Eva-0.139
 wrists-0.133
 yes-0.133
Token
Token ID198
Feature activation-2.85e-01

Pos logit contributions
 wrists+0.186
chard+0.186
 Robert+0.181
 Eva+0.180
 blonde+0.175
Neg logit contributions
loo-0.126
 of-0.120
iren-0.107
ick-0.107
auc-0.103
Token
Token ID198
Feature activation-2.51e-01

Pos logit contributions
 Eva+0.011
 Robert+0.011
 wrists+0.011
chard+0.011
 blonde+0.010
Neg logit contributions
loo-0.012
ick-0.010
auc-0.010
iren-0.010
 of-0.010
Token is
Token ID318
Feature activation+1.09e+00

Pos logit contributions
 Eva+0.015
 Robert+0.015
chard+0.015
 wrists+0.015
 blonde+0.014
Neg logit contributions
loo-0.010
ick-0.009
iren-0.009
 of-0.009
auc-0.009
Token safe
Token ID3338
Feature activation+1.62e+00

Pos logit contributions
loo+0.036
 of+0.031
ick+0.031
iren+0.029
auc+0.029
Neg logit contributions
 wrists-0.051
 Eva-0.051
 Robert-0.051
chard-0.051
 blonde-0.049
Token and
Token ID290
Feature activation+8.38e-02

Pos logit contributions
loo+0.060
iren+0.053
auc+0.051
ick+0.050
rim+0.048
Neg logit contributions
 Robert-0.071
chard-0.071
 Eva-0.070
 wrists-0.068
 Claire-0.067
Token clean
Token ID3424
Feature activation+3.64e+00

Pos logit contributions
loo+0.004
ick+0.003
iren+0.003
auc+0.003
 of+0.003
Neg logit contributions
chard-0.003
 Robert-0.003
 Eva-0.003
 wrists-0.003
 blonde-0.003
Token.
Token ID13
Feature activation+7.31e-01

Pos logit contributions
loo+0.117
 of+0.098
ick+0.097
iren+0.096
auc+0.095
Neg logit contributions
 Eva-0.177
 Robert-0.177
chard-0.177
 wrists-0.176
 blonde-0.169
Token How
Token ID1374
Feature activation-9.73e+00

Pos logit contributions
loo+0.030
 of+0.027
ick+0.026
iren+0.025
auc+0.025
Neg logit contributions
chard-0.029
 wrists-0.029
 Eva-0.029
 Robert-0.028
 blonde-0.028
Token does
Token ID857
Feature activation-1.04e+00

Pos logit contributions
chard+0.451
 wrists+0.451
 Robert+0.436
 Eva+0.434
 blonde+0.427
Neg logit contributions
loo-0.350
 of-0.329
ick-0.300
auc-0.289
 ig-0.268
Token that
Token ID326
Feature activation+3.05e+00

Pos logit contributions
 Robert+0.049
 Eva+0.049
chard+0.048
 wrists+0.047
 Claire+0.045
Neg logit contributions
loo-0.037
ick-0.031
auc-0.031
iren-0.030
arl-0.029
Token sound
Token ID2128
Feature activation+1.01e+00

Pos logit contributions
loo+0.103
ick+0.088
auc+0.087
iren+0.085
arl+0.083
Neg logit contributions
 Robert-0.145
 Eva-0.144
chard-0.140
 yes-0.138
 Claire-0.135
Token?"
Token ID1701
Feature activation-4.06e+00

Pos logit contributions
loo+0.030
ick+0.025
iren+0.024
auc+0.024
 of+0.023
Neg logit contributions
 Eva-0.051
 Robert-0.051
chard-0.051
 wrists-0.050
 Claire-0.048
Token
Token ID198
Feature activation-2.50e-01

Pos logit contributions
 wrists+0.202
chard+0.201
 Robert+0.191
 blonde+0.191
 Eva+0.191
Neg logit contributions
loo-0.123
 of-0.117
ick-0.106
iren-0.104
pour-0.098
Token just
Token ID655
Feature activation+1.64e+00

Pos logit contributions
loo+0.104
 of+0.088
ick+0.088
iren+0.087
auc+0.086
Neg logit contributions
chard-0.129
 Eva-0.129
 Robert-0.129
 wrists-0.128
 blonde-0.123
Token one
Token ID530
Feature activation+4.36e-01

Pos logit contributions
loo+0.061
 of+0.055
ick+0.052
iren+0.051
auc+0.051
Neg logit contributions
 Eva-0.070
chard-0.070
 wrists-0.070
 Robert-0.070
 blonde-0.067
Token of
Token ID286
Feature activation+3.12e+00

Pos logit contributions
loo+0.013
 of+0.011
ick+0.011
iren+0.010
auc+0.010
Neg logit contributions
 Eva-0.022
chard-0.022
 Robert-0.022
 wrists-0.022
 blonde-0.021
Token you
Token ID345
Feature activation+1.84e-02

Pos logit contributions
loo+0.119
ick+0.101
auc+0.100
iren+0.099
 of+0.096
Neg logit contributions
 Eva-0.137
 Robert-0.136
 wrists-0.135
chard-0.134
 Claire-0.128
Token.
Token ID13
Feature activation+6.84e-01

Pos logit contributions
loo+0.001
 of+0.000
ick+0.000
iren+0.000
auc+0.000
Neg logit contributions
chard-0.001
 Eva-0.001
 Robert-0.001
 wrists-0.001
 blonde-0.001
Token How
Token ID1374
Feature activation-9.72e+00

Pos logit contributions
loo+0.031
ick+0.028
 of+0.027
iren+0.027
auc+0.027
Neg logit contributions
 Robert-0.024
 Eva-0.024
chard-0.024
 wrists-0.023
 blonde-0.023
Token would
Token ID561
Feature activation-2.84e+00

Pos logit contributions
 Eva+0.388
chard+0.387
 Robert+0.387
 wrists+0.385
 blonde+0.369
Neg logit contributions
loo-0.403
 of-0.355
ick-0.349
auc-0.342
iren-0.342
Token you
Token ID345
Feature activation+1.46e-03

Pos logit contributions
 Robert+0.151
 Eva+0.151
chard+0.150
 wrists+0.150
 blonde+0.144
Neg logit contributions
loo-0.081
ick-0.065
iren-0.064
 of-0.064
auc-0.063
Token feel
Token ID1254
Feature activation-1.21e+00

Pos logit contributions
loo+0.000
iren+0.000
auc+0.000
ick+0.000
 of+0.000
Neg logit contributions
chard-0.000
 Robert-0.000
 Eva-0.000
 wrists-0.000
 blonde-0.000
Token if
Token ID611
Feature activation-3.36e+00

Pos logit contributions
chard+0.051
 Robert+0.051
 Eva+0.051
 wrists+0.050
 blonde+0.048
Neg logit contributions
loo-0.047
ick-0.040
iren-0.040
 of-0.040
auc-0.039
Token someone
Token ID2130
Feature activation-1.12e+00

Pos logit contributions
 Eva+0.167
chard+0.167
 Robert+0.166
 wrists+0.164
 blonde+0.158
Neg logit contributions
loo-0.106
ick-0.088
iren-0.087
 of-0.087
auc-0.086
Token you
Token ID345
Feature activation+1.74e-01

Pos logit contributions
loo+0.075
 of+0.061
ick+0.059
iren+0.057
auc+0.057
Neg logit contributions
chard-0.162
 Eva-0.162
 Robert-0.161
 wrists-0.161
 blonde-0.156
Token for
Token ID329
Feature activation+1.19e+00

Pos logit contributions
loo+0.005
 of+0.004
ick+0.004
iren+0.004
auc+0.004
Neg logit contributions
chard-0.009
 Eva-0.009
 wrists-0.009
 Robert-0.009
 blonde-0.009
Token saying
Token ID2282
Feature activation-7.52e-01

Pos logit contributions
loo+0.041
 of+0.036
ick+0.035
iren+0.034
auc+0.033
Neg logit contributions
 Eva-0.055
chard-0.055
 Robert-0.055
 wrists-0.054
 blonde-0.053
Token sorry
Token ID7926
Feature activation-1.12e+00

Pos logit contributions
chard+0.033
 Eva+0.032
 wrists+0.031
 Robert+0.031
 blonde+0.031
Neg logit contributions
loo-0.028
 of-0.026
iren-0.025
ick-0.024
arl-0.024
Token.
Token ID13
Feature activation+1.03e+00

Pos logit contributions
 Eva+0.053
 Robert+0.052
 wrists+0.052
chard+0.050
 Claire+0.049
Neg logit contributions
loo-0.038
ick-0.032
iren-0.032
auc-0.031
 of-0.030
Token How
Token ID1374
Feature activation-9.72e+00

Pos logit contributions
loo+0.043
 of+0.038
ick+0.037
iren+0.037
auc+0.037
Neg logit contributions
chard-0.041
 Eva-0.040
 wrists-0.040
 Robert-0.040
 blonde-0.038
Token about
Token ID546
Feature activation-2.53e+00

Pos logit contributions
 wrists+0.431
chard+0.429
 Eva+0.419
 Robert+0.418
 blonde+0.403
Neg logit contributions
loo-0.373
 of-0.347
ick-0.320
auc-0.315
iren-0.295
Token we
Token ID356
Feature activation+1.18e+00

Pos logit contributions
 Eva+0.123
chard+0.123
 Robert+0.122
 wrists+0.122
 blonde+0.119
Neg logit contributions
loo-0.081
 of-0.070
ick-0.067
iren-0.066
auc-0.066
Token go
Token ID467
Feature activation+7.46e-01

Pos logit contributions
loo+0.052
iren+0.044
auc+0.044
ick+0.042
 of+0.040
Neg logit contributions
chard-0.047
 Robert-0.046
 wrists-0.045
 Eva-0.044
 yes-0.044
Token home
Token ID1363
Feature activation+3.11e+00

Pos logit contributions
loo+0.029
ick+0.025
auc+0.025
iren+0.025
arl+0.024
Neg logit contributions
chard-0.030
 Robert-0.030
 wrists-0.029
 Eva-0.029
 yes-0.028
Token and
Token ID290
Feature activation-1.04e-01

Pos logit contributions
loo+0.115
iren+0.099
auc+0.098
ick+0.097
arl+0.090
Neg logit contributions
chard-0.135
 Eva-0.134
 Robert-0.133
 wrists-0.130
 Claire-0.127
Token will
Token ID481
Feature activation-1.22e+00

Pos logit contributions
 Robert+0.020
 Eva+0.020
chard+0.019
 wrists+0.019
 blonde+0.019
Neg logit contributions
loo-0.017
 of-0.015
ick-0.015
iren-0.014
auc-0.014
Token be
Token ID307
Feature activation+1.55e+00

Pos logit contributions
 Robert+0.049
 wrists+0.049
chard+0.049
 Eva+0.049
 blonde+0.046
Neg logit contributions
loo-0.053
auc-0.045
ick-0.044
iren-0.043
 of-0.042
Token sad
Token ID6507
Feature activation+7.82e-01

Pos logit contributions
loo+0.052
 of+0.049
iren+0.045
ick+0.045
auc+0.044
Neg logit contributions
 Eva-0.071
 Robert-0.071
chard-0.070
 wrists-0.070
 blonde-0.068
Token too
Token ID1165
Feature activation+1.15e+00

Pos logit contributions
loo+0.035
iren+0.031
auc+0.030
rim+0.029
ick+0.029
Neg logit contributions
 wrists-0.027
chard-0.027
 Robert-0.027
 Eva-0.026
 Claire-0.026
Token.
Token ID13
Feature activation+8.73e-01

Pos logit contributions
loo+0.042
auc+0.035
ick+0.035
iren+0.035
 of+0.035
Neg logit contributions
 Eva-0.052
 Robert-0.052
chard-0.052
 wrists-0.051
 blonde-0.049
Token How
Token ID1374
Feature activation-9.72e+00

Pos logit contributions
loo+0.035
 of+0.031
iren+0.030
ick+0.030
auc+0.030
Neg logit contributions
chard-0.036
 wrists-0.035
 Robert-0.035
 Eva-0.035
 blonde-0.033
Token can
Token ID460
Feature activation-4.23e-01

Pos logit contributions
chard+0.399
 Eva+0.399
 Robert+0.398
 wrists+0.398
 blonde+0.381
Neg logit contributions
loo-0.392
 of-0.348
ick-0.338
auc-0.332
iren-0.329
Token we
Token ID356
Feature activation+1.14e+00

Pos logit contributions
 Eva+0.021
 Robert+0.021
chard+0.021
 wrists+0.021
 blonde+0.020
Neg logit contributions
loo-0.013
ick-0.011
iren-0.010
auc-0.010
 of-0.010
Token fix
Token ID4259
Feature activation+1.29e+00

Pos logit contributions
loo+0.048
auc+0.041
iren+0.040
ick+0.040
arl+0.037
Neg logit contributions
chard-0.046
 Robert-0.045
 wrists-0.045
 Eva-0.043
 yes-0.042
Token it
Token ID340
Feature activation-4.18e-01

Pos logit contributions
loo+0.045
 of+0.041
ick+0.039
iren+0.038
auc+0.037
Neg logit contributions
chard-0.059
 blonde-0.057
 Eva-0.057
 Robert-0.057
 wrists-0.056
Token?"
Token ID1701
Feature activation-5.13e+00

Pos logit contributions
chard+0.020
 Robert+0.019
 Eva+0.019
 wrists+0.019
 yes+0.018
Neg logit contributions
loo-0.014
ick-0.012
iren-0.012
auc-0.011
arl-0.011
Token share
Token ID2648
Feature activation-4.62e+00

Pos logit contributions
 wrists+0.012
 Eva+0.011
 Robert+0.011
chard+0.011
 blonde+0.011
Neg logit contributions
loo-0.014
auc-0.012
ick-0.011
iren-0.011
 of-0.011
Token and
Token ID290
Feature activation-1.23e-01

Pos logit contributions
chard+0.246
 Robert+0.245
 Eva+0.245
 wrists+0.241
 blonde+0.239
Neg logit contributions
loo-0.123
 of-0.110
ick-0.099
iren-0.098
auc-0.096
Token take
Token ID1011
Feature activation+8.52e-02

Pos logit contributions
 wrists+0.005
 Eva+0.005
 Robert+0.005
chard+0.005
 blonde+0.005
Neg logit contributions
loo-0.005
ick-0.005
auc-0.004
iren-0.004
pour-0.004
Token turns
Token ID4962
Feature activation-1.37e+00

Pos logit contributions
loo+0.003
 of+0.003
ick+0.003
iren+0.003
auc+0.002
Neg logit contributions
chard-0.004
 Eva-0.004
 Robert-0.004
 wrists-0.004
 blonde-0.004
Token.
Token ID13
Feature activation+9.45e-01

Pos logit contributions
 wrists+0.061
 Robert+0.060
 Eva+0.060
chard+0.059
 blonde+0.057
Neg logit contributions
loo-0.052
ick-0.042
auc-0.042
 ig-0.040
pour-0.040
Token How
Token ID1374
Feature activation-9.72e+00

Pos logit contributions
loo+0.039
 of+0.034
ick+0.034
iren+0.033
auc+0.033
Neg logit contributions
 Eva-0.038
 Robert-0.038
chard-0.037
 wrists-0.037
 blonde-0.036
Token about
Token ID546
Feature activation-2.54e+00

Pos logit contributions
 Eva+0.395
 Robert+0.392
chard+0.390
 wrists+0.385
 blonde+0.373
Neg logit contributions
loo-0.392
iren-0.338
ick-0.338
 of-0.331
auc-0.328
Token this
Token ID428
Feature activation-1.49e-01

Pos logit contributions
chard+0.124
 Eva+0.124
 Robert+0.124
 wrists+0.123
 blonde+0.120
Neg logit contributions
loo-0.081
 of-0.070
ick-0.067
iren-0.066
auc-0.066
Token:
Token ID25
Feature activation-1.91e+00

Pos logit contributions
 Eva+0.007
 Robert+0.007
chard+0.007
 wrists+0.007
 blonde+0.007
Neg logit contributions
loo-0.005
ick-0.004
auc-0.004
iren-0.004
 of-0.004
Token you
Token ID345
Feature activation+4.04e-03

Pos logit contributions
 Eva+0.084
 Robert+0.084
 wrists+0.083
chard+0.083
 blonde+0.081
Neg logit contributions
loo-0.069
 of-0.065
iren-0.059
auc-0.059
ick-0.058
Token can
Token ID460
Feature activation-4.21e-01

Pos logit contributions
loo+0.000
iren+0.000
auc+0.000
ick+0.000
 of+0.000
Neg logit contributions
chard-0.000
 Robert-0.000
 wrists-0.000
 Eva-0.000
 blonde-0.000
Token.
Token ID13
Feature activation+9.50e-01

Pos logit contributions
 Eva+0.239
chard+0.239
 Robert+0.238
 wrists+0.236
 blonde+0.232
Neg logit contributions
loo-0.127
 of-0.110
ick-0.102
iren-0.100
auc-0.100
Token Sharing
Token ID36644
Feature activation-4.05e+00

Pos logit contributions
loo+0.040
 of+0.035
ick+0.034
iren+0.034
auc+0.034
Neg logit contributions
chard-0.037
 Eva-0.037
 Robert-0.037
 wrists-0.037
 blonde-0.035
Token is
Token ID318
Feature activation+9.36e-01

Pos logit contributions
 wrists+0.184
 Robert+0.182
 Eva+0.178
chard+0.170
 Claire+0.166
Neg logit contributions
loo-0.150
ick-0.130
auc-0.124
 ig-0.123
pour-0.117
Token caring
Token ID18088
Feature activation-3.08e+00

Pos logit contributions
loo+0.031
 of+0.030
iren+0.027
ick+0.027
auc+0.026
Neg logit contributions
 Eva-0.043
chard-0.043
 Robert-0.042
 wrists-0.042
 blonde-0.042
Token.
Token ID13
Feature activation+1.01e+00

Pos logit contributions
 Eva+0.132
 Robert+0.131
 wrists+0.131
 Claire+0.123
chard+0.121
Neg logit contributions
loo-0.113
iren-0.098
pour-0.096
ick-0.095
auc-0.095
Token How
Token ID1374
Feature activation-9.72e+00

Pos logit contributions
loo+0.042
 of+0.037
ick+0.036
iren+0.036
auc+0.036
Neg logit contributions
chard-0.040
 Eva-0.039
 Robert-0.039
 wrists-0.039
 blonde-0.038
Token about
Token ID546
Feature activation-2.49e+00

Pos logit contributions
 Eva+0.394
 Robert+0.392
chard+0.391
 wrists+0.386
 blonde+0.372
Neg logit contributions
loo-0.393
ick-0.339
iren-0.338
 of-0.336
auc-0.331
Token you
Token ID345
Feature activation+5.60e-02

Pos logit contributions
 Eva+0.128
 blonde+0.128
chard+0.126
 Robert+0.125
 wrists+0.124
Neg logit contributions
loo-0.067
 of-0.064
iren-0.055
ick-0.055
auc-0.053
Token take
Token ID1011
Feature activation+4.29e-02

Pos logit contributions
loo+0.002
ick+0.002
iren+0.002
auc+0.002
 of+0.002
Neg logit contributions
chard-0.002
 Robert-0.002
 Eva-0.002
 wrists-0.002
 blonde-0.002
Token turns
Token ID4962
Feature activation-1.31e+00

Pos logit contributions
loo+0.001
 of+0.001
iren+0.001
ick+0.001
auc+0.001
Neg logit contributions
chard-0.002
 Eva-0.002
 wrists-0.002
 Robert-0.002
 blonde-0.002
Token playing
Token ID2712
Feature activation-2.75e+00

Pos logit contributions
 wrists+0.059
 Robert+0.058
chard+0.057
 Eva+0.057
 Claire+0.054
Neg logit contributions
loo-0.049
 ig-0.040
ick-0.039
auc-0.039
pour-0.037
Token back
Token ID736
Feature activation-1.34e+00

Pos logit contributions
chard+0.055
 Robert+0.055
 Eva+0.055
 wrists+0.054
 blonde+0.051
Neg logit contributions
loo-0.052
iren-0.045
ick-0.045
auc-0.044
 of-0.044
Token and
Token ID290
Feature activation-7.70e-02

Pos logit contributions
 Eva+0.065
chard+0.065
 Robert+0.065
 wrists+0.065
 blonde+0.063
Neg logit contributions
loo-0.043
 of-0.037
ick-0.036
iren-0.035
auc-0.035
Token relax
Token ID8960
Feature activation-1.55e+00

Pos logit contributions
chard+0.003
 Eva+0.003
 Robert+0.003
 wrists+0.003
 blonde+0.003
Neg logit contributions
loo-0.003
ick-0.003
iren-0.003
 of-0.003
auc-0.003
Token together
Token ID1978
Feature activation-2.01e+00

Pos logit contributions
 Eva+0.079
 Robert+0.079
chard+0.078
 wrists+0.078
 blonde+0.076
Neg logit contributions
loo-0.047
 of-0.039
ick-0.038
iren-0.037
auc-0.037
Token.
Token ID13
Feature activation+7.03e-01

Pos logit contributions
chard+0.101
 Eva+0.101
 Robert+0.101
 wrists+0.100
 blonde+0.099
Neg logit contributions
loo-0.061
 of-0.054
ick-0.051
iren-0.047
auc-0.047
Token How
Token ID1374
Feature activation-9.72e+00

Pos logit contributions
loo+0.029
 of+0.026
ick+0.025
iren+0.025
auc+0.025
Neg logit contributions
chard-0.028
 Eva-0.027
 wrists-0.027
 Robert-0.027
 blonde-0.026
Token does
Token ID857
Feature activation-1.19e+00

Pos logit contributions
chard+0.431
 wrists+0.429
 Robert+0.425
 Eva+0.424
 blonde+0.407
Neg logit contributions
loo-0.366
 of-0.332
ick-0.313
auc-0.305
iren-0.292
Token that
Token ID326
Feature activation+3.03e+00

Pos logit contributions
 Robert+0.052
 Eva+0.052
chard+0.051
 wrists+0.050
 blonde+0.048
Neg logit contributions
loo-0.045
auc-0.039
ick-0.039
iren-0.038
 of-0.037
Token sound
Token ID2128
Feature activation+9.74e-01

Pos logit contributions
loo+0.101
ick+0.086
auc+0.086
iren+0.082
arl+0.080
Neg logit contributions
 Robert-0.143
 Eva-0.143
chard-0.140
 wrists-0.136
 yes-0.136
Token?"
Token ID1701
Feature activation-4.25e+00

Pos logit contributions
loo+0.030
ick+0.025
auc+0.024
iren+0.024
 of+0.023
Neg logit contributions
chard-0.048
 Robert-0.048
 Eva-0.048
 wrists-0.046
 Claire-0.045
Token
Token ID198
Feature activation-3.30e-01

Pos logit contributions
chard+0.204
 wrists+0.203
 Robert+0.199
 Eva+0.197
 blonde+0.191
Neg logit contributions
loo-0.144
 of-0.132
iren-0.123
ick-0.120
auc-0.114
Token on
Token ID319
Feature activation-2.14e+00

Pos logit contributions
loo+0.056
ick+0.049
auc+0.047
iren+0.044
rim+0.043
Neg logit contributions
chard-0.076
 wrists-0.076
 Robert-0.075
 Eva-0.075
 blonde-0.071
Token the
Token ID262
Feature activation+2.19e-01

Pos logit contributions
chard+0.101
 Eva+0.099
 wrists+0.099
 Robert+0.099
 blonde+0.097
Neg logit contributions
loo-0.071
 of-0.063
ick-0.060
iren-0.058
auc-0.057
Token big
Token ID1263
Feature activation+2.31e+00

Pos logit contributions
loo+0.008
 of+0.007
iren+0.006
auc+0.006
ick+0.006
Neg logit contributions
 Eva-0.010
chard-0.010
 Robert-0.010
 wrists-0.010
 blonde-0.010
Token slide
Token ID10649
Feature activation-1.89e+00

Pos logit contributions
loo+0.092
 of+0.080
ick+0.080
iren+0.078
auc+0.078
Neg logit contributions
 Eva-0.095
 Robert-0.094
chard-0.094
 wrists-0.094
 blonde-0.090
Token.
Token ID13
Feature activation+1.06e+00

Pos logit contributions
chard+0.094
 Eva+0.094
 Robert+0.094
 wrists+0.093
 blonde+0.090
Neg logit contributions
loo-0.059
ick-0.048
 of-0.048
iren-0.048
auc-0.047
Token How
Token ID1374
Feature activation-9.71e+00

Pos logit contributions
loo+0.044
 of+0.039
ick+0.038
iren+0.037
auc+0.037
Neg logit contributions
chard-0.042
 Eva-0.042
 wrists-0.042
 Robert-0.042
 blonde-0.040
Token was
Token ID373
Feature activation+3.70e-01

Pos logit contributions
chard+0.453
 wrists+0.452
 Robert+0.442
 Eva+0.439
 blonde+0.427
Neg logit contributions
loo-0.352
 of-0.325
ick-0.297
auc-0.292
iren-0.273
Token it
Token ID340
Feature activation-3.98e-01

Pos logit contributions
loo+0.009
 of+0.008
iren+0.006
ick+0.006
auc+0.006
Neg logit contributions
 wrists-0.021
 blonde-0.021
 Eva-0.020
chard-0.020
 Robert-0.020
Token?"
Token ID1701
Feature activation-5.27e+00

Pos logit contributions
chard+0.019
 Eva+0.019
 Robert+0.019
 yes+0.018
 Claire+0.018
Neg logit contributions
loo-0.012
iren-0.011
ick-0.010
auc-0.010
arl-0.010
Token Sara
Token ID24799
Feature activation-1.80e+00

Pos logit contributions
 wrists+0.251
chard+0.247
 Robert+0.247
 Eva+0.243
 blonde+0.238
Neg logit contributions
loo-0.173
 of-0.168
iren-0.149
ick-0.144
auc-0.143
Token's
Token ID338
Feature activation+9.81e-01

Pos logit contributions
chard+0.086
 Robert+0.083
 wrists+0.083
 Eva+0.083
 blonde+0.082
Neg logit contributions
loo-0.059
 of-0.054
iren-0.051
ick-0.051
auc-0.050
Token you
Token ID345
Feature activation+1.41e-01

Pos logit contributions
loo+0.103
iren+0.088
ick+0.088
arl+0.084
 of+0.084
Neg logit contributions
 Eva-0.142
 Robert-0.141
chard-0.139
 wrists-0.139
 Claire-0.132
Token can
Token ID460
Feature activation-3.29e-01

Pos logit contributions
loo+0.004
 of+0.004
ick+0.003
auc+0.003
iren+0.003
Neg logit contributions
 Eva-0.007
 Robert-0.007
 wrists-0.007
chard-0.007
 blonde-0.007
Token use
Token ID779
Feature activation+2.90e+00

Pos logit contributions
chard+0.013
 Robert+0.013
 wrists+0.013
 yes+0.012
 blonde+0.012
Neg logit contributions
loo-0.014
auc-0.012
iren-0.011
pour-0.011
rim-0.011
Token safely
Token ID11512
Feature activation+1.12e+00

Pos logit contributions
loo+0.094
ick+0.077
 of+0.076
iren+0.076
auc+0.076
Neg logit contributions
 Robert-0.142
chard-0.141
 Eva-0.141
 wrists-0.141
 blonde-0.135
Token.
Token ID13
Feature activation+7.29e-01

Pos logit contributions
loo+0.040
auc+0.033
iren+0.033
ick+0.033
pour+0.031
Neg logit contributions
chard-0.052
 Robert-0.050
 wrists-0.050
 Eva-0.049
 blonde-0.047
Token How
Token ID1374
Feature activation-9.71e+00

Pos logit contributions
loo+0.031
 of+0.029
ick+0.028
iren+0.027
auc+0.027
Neg logit contributions
chard-0.027
 wrists-0.026
 Eva-0.026
 Robert-0.026
 blonde-0.025
Token about
Token ID546
Feature activation-2.51e+00

Pos logit contributions
chard+0.460
 wrists+0.458
 Robert+0.449
 Eva+0.449
 blonde+0.432
Neg logit contributions
loo-0.339
 of-0.323
ick-0.289
auc-0.281
 ig-0.258
Token that
Token ID326
Feature activation+3.03e+00

Pos logit contributions
 Robert+0.116
 Eva+0.116
chard+0.116
 wrists+0.115
 blonde+0.111
Neg logit contributions
loo-0.088
 of-0.074
ick-0.074
iren-0.073
auc-0.072
Token?"
Token ID1701
Feature activation-5.35e+00

Pos logit contributions
loo+0.099
ick+0.083
auc+0.081
iren+0.080
 of+0.079
Neg logit contributions
chard-0.148
 Robert-0.148
 Eva-0.147
 wrists-0.144
 blonde-0.139
Token he
Token ID339
Feature activation+6.54e-01

Pos logit contributions
 wrists+0.264
chard+0.252
 Eva+0.250
 Robert+0.248
 blonde+0.245
Neg logit contributions
loo-0.163
 of-0.157
ick-0.142
iren-0.139
auc-0.134
Token said
Token ID531
Feature activation+7.09e-01

Pos logit contributions
loo+0.024
 of+0.023
iren+0.022
ick+0.022
auc+0.021
Neg logit contributions
chard-0.027
 Eva-0.026
 wrists-0.026
 Robert-0.026
 blonde-0.025
Token play
Token ID711
Feature activation-3.89e-01

Pos logit contributions
 Eva+0.017
 Robert+0.017
 wrists+0.017
chard+0.017
 blonde+0.016
Neg logit contributions
loo-0.016
auc-0.014
ick-0.014
iren-0.014
 of-0.013
Token with
Token ID351
Feature activation-2.50e+00

Pos logit contributions
 blonde+0.022
 Eva+0.022
chard+0.022
 wrists+0.022
 Robert+0.022
Neg logit contributions
loo-0.009
 of-0.009
ick-0.007
iren-0.007
auc-0.007
Token something
Token ID1223
Feature activation-3.64e+00

Pos logit contributions
 blonde+0.129
 Robert+0.129
chard+0.129
 Eva+0.129
 wrists+0.127
Neg logit contributions
loo-0.062
 of-0.060
ick-0.054
iren-0.049
auc-0.047
Token else
Token ID2073
Feature activation-4.49e+00

Pos logit contributions
chard+0.133
 Eva+0.133
 Robert+0.132
 wrists+0.132
 blonde+0.127
Neg logit contributions
loo-0.162
 of-0.148
ick-0.142
auc-0.140
iren-0.139
Token.
Token ID13
Feature activation+8.91e-01

Pos logit contributions
 Robert+0.249
 Eva+0.248
chard+0.248
 blonde+0.247
 wrists+0.247
Neg logit contributions
loo-0.109
 of-0.100
ick-0.087
auc-0.081
iren-0.076
Token How
Token ID1374
Feature activation-9.71e+00

Pos logit contributions
loo+0.039
 of+0.035
ick+0.034
iren+0.034
auc+0.034
Neg logit contributions
chard-0.033
 wrists-0.033
 Eva-0.032
 Robert-0.032
 blonde-0.031
Token about
Token ID546
Feature activation-2.50e+00

Pos logit contributions
 wrists+0.434
chard+0.432
 Robert+0.425
 Eva+0.423
 blonde+0.410
Neg logit contributions
loo-0.366
 of-0.348
ick-0.313
auc-0.308
iren-0.285
Token the
Token ID262
Feature activation+9.90e-02

Pos logit contributions
 Eva+0.122
 wrists+0.121
chard+0.121
 blonde+0.120
 Robert+0.120
Neg logit contributions
loo-0.077
 of-0.070
ick-0.063
iren-0.063
auc-0.062
Token blocks
Token ID7021
Feature activation-4.38e-01

Pos logit contributions
loo+0.004
 of+0.003
ick+0.003
iren+0.003
auc+0.003
Neg logit contributions
 Eva-0.004
 Robert-0.004
 wrists-0.004
chard-0.004
 blonde-0.004
Token?"
Token ID1701
Feature activation-3.88e+00

Pos logit contributions
chard+0.021
 Robert+0.020
 Eva+0.020
 wrists+0.020
 yes+0.019
Neg logit contributions
loo-0.014
ick-0.012
iren-0.012
auc-0.011
 ig-0.011
Token Tim
Token ID5045
Feature activation+9.35e-01

Pos logit contributions
chard+0.192
 wrists+0.191
cious+0.185
 blonde+0.181
 Robert+0.179
Neg logit contributions
 of-0.120
loo-0.118
iren-0.099
ick-0.097
 late-0.095
Token magic
Token ID5536
Feature activation-4.04e-01

Pos logit contributions
chard+0.007
 wrists+0.007
 Robert+0.007
 Eva+0.007
 blonde+0.007
Neg logit contributions
loo-0.005
 of-0.004
iren-0.004
ick-0.004
auc-0.004
Token book
Token ID1492
Feature activation-1.40e+00

Pos logit contributions
chard+0.019
 Robert+0.019
 Eva+0.019
 wrists+0.019
 blonde+0.018
Neg logit contributions
loo-0.014
 of-0.012
ick-0.012
iren-0.011
auc-0.011
Token?
Token ID30
Feature activation-6.85e-01

Pos logit contributions
 Eva+0.069
 Robert+0.069
chard+0.067
 wrists+0.066
 Claire+0.064
Neg logit contributions
loo-0.043
ick-0.038
iren-0.035
auc-0.035
rim-0.034
Token Really
Token ID16123
Feature activation-4.97e+00

Pos logit contributions
chard+0.024
 wrists+0.024
 Robert+0.024
 Eva+0.023
 blonde+0.022
Neg logit contributions
loo-0.031
 of-0.028
auc-0.028
ick-0.027
iren-0.027
Token?
Token ID30
Feature activation-7.07e-01

Pos logit contributions
chard+0.285
 Eva+0.285
 Robert+0.283
 wrists+0.280
 blonde+0.272
Neg logit contributions
loo-0.117
auc-0.088
iren-0.088
ick-0.088
 of-0.086
Token How
Token ID1374
Feature activation-9.71e+00

Pos logit contributions
chard+0.026
 wrists+0.026
 Robert+0.026
 Eva+0.025
 blonde+0.024
Neg logit contributions
loo-0.030
 of-0.028
auc-0.027
ick-0.027
iren-0.026
Token much
Token ID881
Feature activation+1.63e+00

Pos logit contributions
chard+0.496
 wrists+0.493
 Robert+0.483
 Eva+0.476
 blonde+0.459
Neg logit contributions
loo-0.304
 of-0.292
ick-0.256
auc-0.251
arl-0.227
Token do
Token ID466
Feature activation-2.79e-01

Pos logit contributions
loo+0.054
ick+0.047
auc+0.046
iren+0.045
arl+0.043
Neg logit contributions
 Eva-0.079
 Robert-0.077
chard-0.075
 wrists-0.074
 blonde-0.073
Token you
Token ID345
Feature activation+2.17e-02

Pos logit contributions
chard+0.015
 Eva+0.015
 Robert+0.014
 wrists+0.014
 blonde+0.014
Neg logit contributions
loo-0.008
 of-0.007
ick-0.006
iren-0.006
auc-0.006
Token want
Token ID765
Feature activation-2.04e-01

Pos logit contributions
loo+0.001
 of+0.001
ick+0.001
iren+0.001
auc+0.001
Neg logit contributions
chard-0.001
 Eva-0.001
 Robert-0.001
 wrists-0.001
 blonde-0.001
Token for
Token ID329
Feature activation+1.10e+00

Pos logit contributions
 Eva+0.010
 Robert+0.010
chard+0.010
 wrists+0.010
 blonde+0.009
Neg logit contributions
loo-0.007
ick-0.006
 of-0.005
iren-0.005
auc-0.005
Token play
Token ID711
Feature activation-2.77e-01

Pos logit contributions
 Eva+0.011
 Robert+0.011
 wrists+0.011
chard+0.011
 blonde+0.010
Neg logit contributions
loo-0.011
auc-0.010
iren-0.010
 of-0.009
ick-0.009
Token and
Token ID290
Feature activation-2.52e-01

Pos logit contributions
 Eva+0.015
 Robert+0.015
chard+0.015
 wrists+0.015
 blonde+0.014
Neg logit contributions
loo-0.008
 of-0.007
ick-0.006
auc-0.006
iren-0.006
Token watch
Token ID2342
Feature activation-5.17e-01

Pos logit contributions
 Eva+0.011
 Robert+0.011
chard+0.011
 wrists+0.011
 blonde+0.010
Neg logit contributions
loo-0.010
 of-0.008
ick-0.008
iren-0.008
auc-0.008
Token cartoons
Token ID30070
Feature activation-5.58e+00

Pos logit contributions
chard+0.026
 Eva+0.025
 wrists+0.025
 Robert+0.025
 blonde+0.025
Neg logit contributions
loo-0.015
 of-0.014
ick-0.013
auc-0.012
iren-0.012
Token.
Token ID13
Feature activation+7.15e-01

Pos logit contributions
 Eva+0.286
chard+0.286
 Robert+0.286
 wrists+0.285
 blonde+0.279
Neg logit contributions
loo-0.162
 of-0.138
ick-0.133
iren-0.129
auc-0.128
Token How
Token ID1374
Feature activation-9.71e+00

Pos logit contributions
loo+0.030
 of+0.028
ick+0.027
iren+0.027
auc+0.026
Neg logit contributions
chard-0.027
 wrists-0.027
 Eva-0.026
 Robert-0.026
 blonde-0.025
Token about
Token ID546
Feature activation-2.57e+00

Pos logit contributions
chard+0.484
 wrists+0.482
 Robert+0.471
 Eva+0.461
 blonde+0.451
Neg logit contributions
 of-0.318
loo-0.313
ick-0.267
auc-0.260
 ig-0.231
Token that
Token ID326
Feature activation+2.96e+00

Pos logit contributions
chard+0.127
 Eva+0.126
 Robert+0.126
 wrists+0.125
 blonde+0.123
Neg logit contributions
loo-0.079
 of-0.070
ick-0.066
iren-0.064
auc-0.064
Token?"
Token ID1701
Feature activation-4.09e+00

Pos logit contributions
loo+0.098
ick+0.081
auc+0.080
iren+0.080
arl+0.076
Neg logit contributions
 Eva-0.143
 Robert-0.143
chard-0.141
 wrists-0.137
 Claire-0.135
Token
Token ID198
Feature activation-1.23e-01

Pos logit contributions
 wrists+0.212
chard+0.206
 blonde+0.206
cious+0.205
 Eva+0.195
Neg logit contributions
 of-0.110
loo-0.108
iren-0.095
 late-0.091
ick-0.085
Token
Token ID198
Feature activation-1.18e-01

Pos logit contributions
chard+0.006
 wrists+0.006
 blonde+0.006
 Robert+0.006
 Eva+0.006
Neg logit contributions
loo-0.004
 of-0.004
iren-0.003
ick-0.003
auc-0.003
Token can
Token ID460
Feature activation-2.79e-01

Pos logit contributions
loo+0.044
iren+0.038
auc+0.037
ick+0.036
rim+0.035
Neg logit contributions
chard-0.052
 Robert-0.052
 wrists-0.051
 Eva-0.051
 Claire-0.048
Token play
Token ID711
Feature activation-4.10e-01

Pos logit contributions
 wrists+0.011
 Robert+0.011
 Eva+0.010
chard+0.010
 blonde+0.010
Neg logit contributions
loo-0.012
auc-0.010
iren-0.010
pour-0.010
ick-0.010
Token pretend
Token ID16614
Feature activation-2.05e+00

Pos logit contributions
 Eva+0.020
chard+0.020
 Robert+0.020
 wrists+0.020
 blonde+0.020
Neg logit contributions
loo-0.013
 of-0.011
ick-0.011
auc-0.010
iren-0.010
Token together
Token ID1978
Feature activation-1.88e+00

Pos logit contributions
 Robert+0.096
 Eva+0.096
chard+0.095
 wrists+0.095
 Claire+0.090
Neg logit contributions
loo-0.069
ick-0.059
iren-0.058
auc-0.056
arl-0.056
Token.
Token ID13
Feature activation+8.88e-01

Pos logit contributions
 Eva+0.095
chard+0.095
 Robert+0.095
 wrists+0.094
 blonde+0.091
Neg logit contributions
loo-0.057
 of-0.047
ick-0.046
iren-0.045
auc-0.045
Token How
Token ID1374
Feature activation-9.71e+00

Pos logit contributions
loo+0.037
 of+0.034
iren+0.033
ick+0.033
auc+0.032
Neg logit contributions
chard-0.034
 wrists-0.034
 Eva-0.033
 Robert-0.033
 blonde-0.032
Token about
Token ID546
Feature activation-2.49e+00

Pos logit contributions
chard+0.430
 wrists+0.428
 Robert+0.424
 Eva+0.424
 blonde+0.406
Neg logit contributions
loo-0.366
 of-0.334
ick-0.313
auc-0.307
iren-0.295
Token we
Token ID356
Feature activation+1.19e+00

Pos logit contributions
 Eva+0.117
chard+0.117
 Robert+0.117
 wrists+0.116
 blonde+0.113
Neg logit contributions
loo-0.085
 of-0.073
ick-0.071
iren-0.070
auc-0.069
Token pretend
Token ID16614
Feature activation-2.05e+00

Pos logit contributions
loo+0.056
iren+0.048
auc+0.047
ick+0.045
rim+0.044
Neg logit contributions
chard-0.043
 Robert-0.043
 wrists-0.042
 yes-0.042
 Eva-0.041
Token to
Token ID284
Feature activation-2.99e-01

Pos logit contributions
 Robert+0.100
 Eva+0.100
chard+0.099
 wrists+0.098
 blonde+0.093
Neg logit contributions
loo-0.066
ick-0.056
iren-0.055
auc-0.054
 of-0.054
Token be
Token ID307
Feature activation+1.63e+00

Pos logit contributions
 wrists+0.012
 Robert+0.012
 Eva+0.012
chard+0.012
 blonde+0.011
Neg logit contributions
loo-0.012
ick-0.011
iren-0.010
auc-0.010
arl-0.010
Token go
Token ID467
Feature activation+4.62e-01

Pos logit contributions
 Eva+0.049
 Robert+0.049
chard+0.049
 wrists+0.048
 blonde+0.047
Neg logit contributions
loo-0.043
 of-0.037
ick-0.036
auc-0.036
iren-0.036
Token the
Token ID262
Feature activation-2.52e-02

Pos logit contributions
loo+0.016
auc+0.013
ick+0.013
iren+0.013
 of+0.013
Neg logit contributions
chard-0.022
 Robert-0.022
 Eva-0.021
 wrists-0.021
 blonde-0.020
Token other
Token ID584
Feature activation+1.42e-02

Pos logit contributions
 Eva+0.001
 wrists+0.001
 Robert+0.001
 blonde+0.001
chard+0.001
Neg logit contributions
loo-0.001
 of-0.001
ick-0.001
 ig-0.001
iren-0.001
Token way
Token ID835
Feature activation+9.12e-01

Pos logit contributions
loo+0.000
 of+0.000
ick+0.000
iren+0.000
auc+0.000
Neg logit contributions
chard-0.001
 Eva-0.001
 Robert-0.001
 wrists-0.001
 blonde-0.001
Token.
Token ID13
Feature activation+2.78e-01

Pos logit contributions
loo+0.033
auc+0.028
iren+0.028
ick+0.027
rim+0.026
Neg logit contributions
 Robert-0.041
chard-0.040
 Eva-0.040
 wrists-0.039
 Claire-0.038
Token How
Token ID1374
Feature activation-9.70e+00

Pos logit contributions
loo+0.010
 of+0.010
ick+0.009
iren+0.009
arl+0.009
Neg logit contributions
chard-0.012
 wrists-0.012
 Eva-0.011
 Robert-0.011
 blonde-0.011
Token does
Token ID857
Feature activation-1.35e+00

Pos logit contributions
chard+0.458
 wrists+0.458
 Eva+0.450
 Robert+0.449
 blonde+0.433
Neg logit contributions
loo-0.333
 of-0.316
ick-0.287
auc-0.277
pour-0.258
Token that
Token ID326
Feature activation+2.96e+00

Pos logit contributions
 Robert+0.059
 Eva+0.059
chard+0.057
 wrists+0.057
 blonde+0.055
Neg logit contributions
loo-0.052
auc-0.045
ick-0.044
iren-0.043
 of-0.042
Token sound
Token ID2128
Feature activation+1.05e+00

Pos logit contributions
loo+0.108
auc+0.090
iren+0.089
ick+0.088
arl+0.086
Neg logit contributions
 Robert-0.135
 Eva-0.132
 yes-0.126
chard-0.126
 wrists-0.125
Token?"
Token ID1701
Feature activation-4.40e+00

Pos logit contributions
loo+0.035
ick+0.028
auc+0.028
iren+0.028
 of+0.027
Neg logit contributions
 Robert-0.050
 Eva-0.050
chard-0.050
 wrists-0.049
 blonde-0.047
Token
Token ID198
Feature activation-5.85e-02

Pos logit contributions
chard+0.204
 wrists+0.199
cious+0.195
 blonde+0.192
 Eva+0.184
Neg logit contributions
 of-0.144
loo-0.136
iren-0.124
ick-0.119
pour-0.117
Tokenained
Token ID1328
Feature activation+3.62e+00

Pos logit contributions
loo+0.150
 of+0.129
 ig+0.113
 outage+0.106
 tunnel+0.104
Neg logit contributions
 wrists-0.342
 blonde-0.340
chard-0.336
 Eva-0.328
 Robert-0.321
Token and
Token ID290
Feature activation-3.82e-01

Pos logit contributions
loo+0.142
auc+0.119
iren+0.119
ick+0.116
arl+0.116
Neg logit contributions
 Robert-0.150
 Eva-0.147
chard-0.146
 Claire-0.140
 wrists-0.140
Token r
Token ID374
Feature activation+6.54e+00

Pos logit contributions
 Eva+0.016
chard+0.016
 wrists+0.016
 Robert+0.016
 blonde+0.015
Neg logit contributions
loo-0.015
ick-0.013
iren-0.013
auc-0.012
arl-0.012
Tokenained
Token ID1328
Feature activation+3.55e+00

Pos logit contributions
loo+0.146
 of+0.125
 ig+0.103
iren+0.096
 tunnel+0.096
Neg logit contributions
 wrists-0.356
 blonde-0.349
chard-0.342
 Eva-0.336
 Robert-0.328
Token and
Token ID290
Feature activation-3.67e-01

Pos logit contributions
loo+0.137
auc+0.115
iren+0.114
ick+0.112
arl+0.110
Neg logit contributions
 Robert-0.149
 Eva-0.148
chard-0.147
 wrists-0.142
 blonde-0.140
Token r
Token ID374
Feature activation+6.51e+00

Pos logit contributions
chard+0.016
 wrists+0.016
 Eva+0.016
 Robert+0.016
 blonde+0.015
Neg logit contributions
loo-0.014
ick-0.012
iren-0.012
auc-0.012
arl-0.012
Tokenained
Token ID1328
Feature activation+3.51e+00

Pos logit contributions
loo+0.151
 of+0.127
 ig+0.104
iren+0.102
auc+0.099
Neg logit contributions
 wrists-0.352
 blonde-0.343
chard-0.335
 Eva-0.326
rice-0.323
Token.
Token ID13
Feature activation+3.34e-01

Pos logit contributions
loo+0.131
auc+0.109
iren+0.107
ick+0.106
rim+0.104
Neg logit contributions
 Robert-0.151
 Eva-0.150
chard-0.150
 wrists-0.145
 blonde-0.143
Token The
Token ID383
Feature activation-2.01e-01

Pos logit contributions
 of+0.013
loo+0.012
ick+0.012
pour+0.011
 late+0.011
Neg logit contributions
chard-0.014
 wrists-0.013
cious-0.012
 Eva-0.012
 Robert-0.012
Token rain
Token ID6290
Feature activation+4.86e+00

Pos logit contributions
 Eva+0.010
chard+0.010
 Robert+0.010
 wrists+0.010
 blonde+0.010
Neg logit contributions
loo-0.006
 of-0.005
ick-0.005
iren-0.005
auc-0.005
Token was
Token ID373
Feature activation+3.52e-01

Pos logit contributions
loo+0.209
ick+0.177
iren+0.176
auc+0.176
arl+0.176
Neg logit contributions
 wrists-0.183
 Eva-0.179
 blonde-0.176
 Robert-0.175
chard-0.175
Token.
Token ID13
Feature activation+7.08e-01

Pos logit contributions
loo+0.033
iren+0.027
ick+0.027
auc+0.026
 of+0.026
Neg logit contributions
chard-0.052
 Eva-0.052
 Robert-0.052
 wrists-0.051
 blonde-0.049
Token But
Token ID887
Feature activation+8.98e-03

Pos logit contributions
 of+0.024
loo+0.022
ick+0.020
iren+0.020
pour+0.020
Neg logit contributions
 wrists-0.032
chard-0.031
cious-0.030
 Eva-0.029
rice-0.028
Token it
Token ID340
Feature activation+2.51e-01

Pos logit contributions
loo+0.000
 of+0.000
ick+0.000
iren+0.000
auc+0.000
Neg logit contributions
 Eva-0.000
 wrists-0.000
 Robert-0.000
chard-0.000
 blonde-0.000
Token was
Token ID373
Feature activation+8.12e-01

Pos logit contributions
loo+0.010
iren+0.008
ick+0.008
 of+0.008
auc+0.008
Neg logit contributions
chard-0.011
 Eva-0.011
 Robert-0.011
 wrists-0.010
 blonde-0.010
Token too
Token ID1165
Feature activation+1.49e+00

Pos logit contributions
loo+0.032
auc+0.028
iren+0.027
ick+0.026
arl+0.026
Neg logit contributions
chard-0.036
 Eva-0.034
 Robert-0.034
 blonde-0.034
 wrists-0.033
Token wind
Token ID2344
Feature activation+6.04e+00

Pos logit contributions
loo+0.065
auc+0.058
iren+0.057
rim+0.056
ick+0.052
Neg logit contributions
chard-0.058
 Eva-0.052
 blonde-0.051
 Robert-0.050
 Claire-0.048
Tokeny
Token ID88
Feature activation+5.54e-01

Pos logit contributions
loo+0.198
auc+0.159
iren+0.156
pour+0.155
 ig+0.148
Neg logit contributions
chard-0.292
 Robert-0.286
 wrists-0.281
 Eva-0.279
rice-0.275
Token outside
Token ID2354
Feature activation-1.51e+00

Pos logit contributions
loo+0.020
iren+0.017
auc+0.017
ick+0.016
pour+0.016
Neg logit contributions
 Robert-0.025
chard-0.025
 Eva-0.025
 wrists-0.024
 blonde-0.023
Token.
Token ID13
Feature activation+7.33e-01

Pos logit contributions
 Robert+0.063
chard+0.063
 Eva+0.062
 wrists+0.061
 blonde+0.061
Neg logit contributions
loo-0.058
iren-0.051
arl-0.049
rim-0.048
auc-0.048
Token So
Token ID1406
Feature activation-2.26e+00

Pos logit contributions
 of+0.023
loo+0.021
ick+0.020
iren+0.019
pour+0.019
Neg logit contributions
chard-0.035
 wrists-0.035
cious-0.033
ya-0.032
 Eva-0.032
Token the
Token ID262
Feature activation+7.01e-01

Pos logit contributions
 wrists+0.134
chard+0.133
 Eva+0.131
 Robert+0.129
 blonde+0.127
Neg logit contributions
loo-0.047
 of-0.043
ick-0.038
 ig-0.032
pour-0.032
Token home
Token ID1363
Feature activation+2.61e+00

Pos logit contributions
loo+0.024
auc+0.022
ick+0.021
iren+0.021
arl+0.020
Neg logit contributions
chard-0.028
 Robert-0.026
 Eva-0.025
 Claire-0.024
 wrists-0.024
Token.
Token ID13
Feature activation+8.74e-01

Pos logit contributions
loo+0.090
iren+0.075
auc+0.075
ick+0.074
 of+0.072
Neg logit contributions
 Eva-0.122
chard-0.122
 Robert-0.122
 wrists-0.119
 Claire-0.115
Token It
Token ID632
Feature activation-1.27e+00

Pos logit contributions
loo+0.034
 of+0.032
auc+0.030
iren+0.030
ick+0.030
Neg logit contributions
chard-0.036
 Eva-0.035
 wrists-0.035
 Robert-0.035
 blonde-0.033
Token's
Token ID338
Feature activation+1.14e+00

Pos logit contributions
 Eva+0.051
 wrists+0.051
 Robert+0.051
chard+0.050
 Claire+0.047
Neg logit contributions
loo-0.051
iren-0.045
ick-0.044
arl-0.043
 of-0.043
Token too
Token ID1165
Feature activation+1.16e+00

Pos logit contributions
loo+0.042
 of+0.036
ick+0.035
iren+0.035
auc+0.035
Neg logit contributions
chard-0.051
 Eva-0.051
 Robert-0.051
 wrists-0.050
 blonde-0.048
Token wind
Token ID2344
Feature activation+5.60e+00

Pos logit contributions
loo+0.044
auc+0.038
iren+0.038
ick+0.037
 of+0.037
Neg logit contributions
chard-0.050
 Eva-0.049
 Robert-0.049
 wrists-0.047
 blonde-0.046
Tokeny
Token ID88
Feature activation+2.69e-01

Pos logit contributions
loo+0.179
ick+0.147
auc+0.146
 of+0.144
iren+0.144
Neg logit contributions
 wrists-0.268
chard-0.265
 Eva-0.265
 Robert-0.265
rice-0.251
Token here
Token ID994
Feature activation-3.32e+00

Pos logit contributions
loo+0.009
iren+0.008
ick+0.008
auc+0.008
 of+0.007
Neg logit contributions
chard-0.012
 Robert-0.012
 Eva-0.012
 wrists-0.012
 blonde-0.011
Token."
Token ID526
Feature activation-2.99e+00

Pos logit contributions
 Eva+0.161
chard+0.161
 Robert+0.161
 wrists+0.160
 blonde+0.155
Neg logit contributions
loo-0.106
 of-0.092
ick-0.089
iren-0.086
auc-0.085
Token
Token ID198
Feature activation+7.28e-03

Pos logit contributions
chard+0.130
 wrists+0.130
 Eva+0.129
 Robert+0.129
 blonde+0.124
Neg logit contributions
loo-0.114
 of-0.101
ick-0.097
iren-0.097
auc-0.095
Token
Token ID198
Feature activation+2.00e-02

Pos logit contributions
loo+0.000
ick+0.000
auc+0.000
iren+0.000
pour+0.000
Neg logit contributions
 Eva-0.000
 Robert-0.000
chard-0.000
 wrists-0.000
 Claire-0.000
Token be
Token ID307
Feature activation+1.97e+00

Pos logit contributions
 Eva+0.007
 Robert+0.007
chard+0.007
 wrists+0.007
 blonde+0.007
Neg logit contributions
loo-0.007
auc-0.006
ick-0.006
iren-0.006
 of-0.006
Token careful
Token ID8161
Feature activation+1.67e+00

Pos logit contributions
loo+0.072
 of+0.068
ick+0.064
iren+0.063
auc+0.061
Neg logit contributions
chard-0.085
 wrists-0.085
 Robert-0.085
 Eva-0.085
 blonde-0.079
Token,
Token ID11
Feature activation+2.98e-01

Pos logit contributions
loo+0.060
auc+0.048
iren+0.048
ick+0.047
pour+0.047
Neg logit contributions
 Robert-0.078
 Eva-0.077
 wrists-0.076
chard-0.075
 blonde-0.073
Token because
Token ID780
Feature activation+2.33e+00

Pos logit contributions
loo+0.014
 of+0.012
ick+0.012
iren+0.012
auc+0.012
Neg logit contributions
 Eva-0.010
 Robert-0.010
chard-0.010
 wrists-0.010
 blonde-0.009
Token the
Token ID262
Feature activation+3.28e-01

Pos logit contributions
loo+0.110
arl+0.107
rim+0.104
iren+0.102
ick+0.102
Neg logit contributions
chard-0.075
 Robert-0.074
 Eva-0.069
 yes-0.067
 blonde-0.066
Token tunnel
Token ID13275
Feature activation+5.36e+00

Pos logit contributions
loo+0.007
rim+0.006
iren+0.006
ick+0.006
arl+0.006
Neg logit contributions
 blonde-0.018
 wrists-0.018
 Robert-0.017
 Eva-0.017
chard-0.017
Token was
Token ID373
Feature activation+6.99e-01

Pos logit contributions
loo+0.216
ick+0.187
iren+0.187
arl+0.186
auc+0.181
Neg logit contributions
 Robert-0.212
 wrists-0.210
 Eva-0.210
chard-0.207
 blonde-0.199
Token fragile
Token ID21049
Feature activation+2.48e+00

Pos logit contributions
loo+0.031
auc+0.027
iren+0.026
ick+0.026
arl+0.025
Neg logit contributions
chard-0.027
 Robert-0.026
 Eva-0.026
 blonde-0.025
 wrists-0.025
Token.
Token ID13
Feature activation+3.88e-01

Pos logit contributions
loo+0.087
iren+0.073
ick+0.070
auc+0.069
arl+0.069
Neg logit contributions
 Robert-0.115
 wrists-0.113
 Eva-0.113
chard-0.112
 blonde-0.110
Token
Token ID220
Feature activation-2.32e+00

Pos logit contributions
loo+0.016
 of+0.015
ick+0.014
iren+0.014
auc+0.014
Neg logit contributions
chard-0.015
 wrists-0.015
 Eva-0.015
 Robert-0.015
 blonde-0.014
Token
Token ID198
Feature activation+3.07e-01

Pos logit contributions
chard+0.102
 wrists+0.102
 Eva+0.101
 Robert+0.101
 blonde+0.097
Neg logit contributions
loo-0.087
 of-0.076
ick-0.075
iren-0.073
auc-0.073
Token as
Token ID355
Feature activation-1.95e+00

Pos logit contributions
loo+0.012
 of+0.009
ick+0.009
iren+0.009
auc+0.009
Neg logit contributions
 Eva-0.022
 Robert-0.022
chard-0.022
 wrists-0.022
 blonde-0.021
Token he
Token ID339
Feature activation+5.81e-01

Pos logit contributions
 wrists+0.093
chard+0.091
 Eva+0.090
 Robert+0.089
 blonde+0.087
Neg logit contributions
loo-0.067
 of-0.060
ick-0.057
auc-0.053
iren-0.053
Token walked
Token ID6807
Feature activation-6.59e-01

Pos logit contributions
loo+0.024
 of+0.022
ick+0.021
iren+0.020
auc+0.020
Neg logit contributions
chard-0.024
 Eva-0.023
 Robert-0.023
 wrists-0.023
 blonde-0.022
Token home
Token ID1363
Feature activation+3.06e+00

Pos logit contributions
 Eva+0.033
 wrists+0.032
chard+0.032
 Robert+0.032
 blonde+0.031
Neg logit contributions
loo-0.021
 of-0.018
ick-0.017
iren-0.017
auc-0.017
Token.
Token ID13
Feature activation+2.04e-01

Pos logit contributions
loo+0.102
 of+0.085
ick+0.085
auc+0.083
iren+0.083
Neg logit contributions
 Eva-0.146
chard-0.146
 Robert-0.146
 wrists-0.145
 blonde-0.140
Token<|endoftext|>
Token ID50256
Feature activation+4.84e+00

Pos logit contributions
 of+0.006
loo+0.006
 late+0.005
iren+0.005
ick+0.004
Neg logit contributions
 wrists-0.011
chard-0.011
 blonde-0.010
cious-0.010
 Eva-0.010
TokenOnce
Token ID7454
Feature activation+7.43e-01

Pos logit contributions
loo+0.160
 of+0.143
ick+0.134
iren+0.128
pour+0.123
Neg logit contributions
 wrists-0.244
chard-0.239
 Eva-0.231
 Robert-0.228
 blonde-0.228
Token upon
Token ID2402
Feature activation+2.03e+00

Pos logit contributions
loo+0.025
iren+0.021
auc+0.021
ick+0.021
 of+0.020
Neg logit contributions
 Robert-0.035
 Eva-0.035
chard-0.035
 wrists-0.033
 blonde-0.033
Token a
Token ID257
Feature activation-1.02e+00

Pos logit contributions
 of+0.062
loo+0.060
ick+0.050
 late+0.049
 ig+0.048
Neg logit contributions
 wrists-0.106
chard-0.100
 Eva-0.098
 Robert-0.097
rice-0.094
Token time
Token ID640
Feature activation+2.64e+00

Pos logit contributions
chard+0.058
 wrists+0.057
 Robert+0.055
 Eva+0.054
 Claire+0.052
Neg logit contributions
loo-0.025
 of-0.023
ick-0.023
arl-0.019
 ig-0.018
Token there
Token ID612
Feature activation-1.31e+00

Pos logit contributions
loo+0.070
 of+0.058
ick+0.056
auc+0.053
iren+0.053
Neg logit contributions
 wrists-0.145
chard-0.144
 Eva-0.143
 Robert-0.143
 blonde-0.139
Token prince
Token ID19716
Feature activation-2.46e+00

Pos logit contributions
loo+0.040
 of+0.034
ick+0.033
auc+0.032
 ig+0.030
Neg logit contributions
 Eva-0.065
chard-0.065
 Robert-0.065
 wrists-0.064
 blonde-0.062
Token?"
Token ID1701
Feature activation-3.90e+00

Pos logit contributions
 Robert+0.142
 Eva+0.141
chard+0.139
 wrists+0.136
 blonde+0.133
Neg logit contributions
loo-0.058
iren-0.045
ick-0.044
auc-0.044
 of-0.043
Token
Token ID198
Feature activation-7.79e-02

Pos logit contributions
 wrists+0.197
chard+0.191
 Eva+0.183
 blonde+0.180
cious+0.179
Neg logit contributions
loo-0.115
 of-0.112
iren-0.106
ick-0.097
auc-0.094
Token
Token ID198
Feature activation-6.62e-02

Pos logit contributions
 wrists+0.004
chard+0.004
 Eva+0.004
 blonde+0.004
 Robert+0.004
Neg logit contributions
loo-0.003
 of-0.002
iren-0.002
ick-0.002
auc-0.002
TokenThe
Token ID464
Feature activation-1.64e-01

Pos logit contributions
 Eva+0.003
chard+0.003
 Robert+0.003
 wrists+0.003
 blonde+0.003
Neg logit contributions
loo-0.003
 of-0.002
ick-0.002
iren-0.002
auc-0.002
Token little
Token ID1310
Feature activation+1.26e+00

Pos logit contributions
 wrists+0.006
 Eva+0.005
chard+0.005
 Robert+0.005
 blonde+0.005
Neg logit contributions
loo-0.008
 of-0.007
ick-0.007
 ig-0.007
auc-0.007
Token prince
Token ID19716
Feature activation-2.49e+00

Pos logit contributions
loo+0.061
 of+0.055
ick+0.054
auc+0.053
iren+0.053
Neg logit contributions
 Eva-0.041
 wrists-0.041
chard-0.041
 Robert-0.041
 blonde-0.039
Token showed
Token ID3751
Feature activation-1.35e+00

Pos logit contributions
chard+0.108
 wrists+0.104
 blonde+0.102
 Eva+0.101
 Robert+0.100
Neg logit contributions
loo-0.088
 of-0.087
ick-0.080
 ig-0.075
iren-0.074
Token his
Token ID465
Feature activation-1.76e-01

Pos logit contributions
chard+0.063
 wrists+0.062
 Eva+0.062
 Robert+0.061
 blonde+0.060
Neg logit contributions
loo-0.047
 of-0.042
ick-0.040
iren-0.039
auc-0.039
Token mum
Token ID25682
Feature activation+7.69e-01

Pos logit contributions
chard+0.009
 Eva+0.009
 Robert+0.008
 wrists+0.008
 blonde+0.008
Neg logit contributions
loo-0.005
 of-0.005
 ig-0.005
 tunnel-0.004
ick-0.004
Token the
Token ID262
Feature activation+3.46e-01

Pos logit contributions
 of+0.022
loo+0.021
ick+0.018
iren+0.018
 ig+0.018
Neg logit contributions
chard-0.039
 wrists-0.039
 blonde-0.038
 Eva-0.038
 Robert-0.037
Token a
Token ID257
Feature activation-7.67e-01

Pos logit contributions
loo+0.028
 of+0.028
ick+0.025
 ig+0.023
 late+0.023
Neg logit contributions
 wrists-0.042
chard-0.040
 Eva-0.040
 Robert-0.039
rice-0.039
Token small
Token ID1402
Feature activation+1.74e+00

Pos logit contributions
 wrists+0.035
 Eva+0.035
 Robert+0.035
chard+0.033
 blonde+0.033
Neg logit contributions
loo-0.027
 of-0.024
ick-0.023
 ig-0.022
auc-0.021
Token girl
Token ID2576
Feature activation-1.24e+00

Pos logit contributions
loo+0.067
iren+0.058
auc+0.058
 of+0.058
ick+0.058
Neg logit contributions
chard-0.073
 Robert-0.072
 Eva-0.072
 wrists-0.071
 blonde-0.069
Token named
Token ID3706
Feature activation+1.40e-01

Pos logit contributions
chard+0.040
 Eva+0.039
 wrists+0.039
 Robert+0.038
 blonde+0.038
Neg logit contributions
loo-0.061
iren-0.054
ick-0.053
arl-0.053
auc-0.053
Token Lily
Token ID20037
Feature activation+1.81e-01

Pos logit contributions
 of+0.004
loo+0.004
ick+0.004
arl+0.003
auc+0.003
Neg logit contributions
 wrists-0.007
 Robert-0.007
chard-0.007
 Edward-0.007
 Percy-0.006
Token.
Token ID13
Feature activation+1.22e+00

Pos logit contributions
loo+0.006
iren+0.005
auc+0.005
ick+0.005
arl+0.005
Neg logit contributions
chard-0.009
 Robert-0.009
 Eva-0.009
 wrists-0.008
 blonde-0.008
Token She
Token ID1375
Feature activation+8.87e-02

Pos logit contributions
loo+0.047
 of+0.042
ick+0.040
auc+0.039
iren+0.039
Neg logit contributions
chard-0.052
 Robert-0.052
 wrists-0.052
 Eva-0.052
 blonde-0.048
Token lived
Token ID5615
Feature activation+6.94e-01

Pos logit contributions
loo+0.005
 of+0.004
ick+0.004
iren+0.004
auc+0.004
Neg logit contributions
 Eva-0.003
chard-0.003
 wrists-0.003
 Robert-0.003
 blonde-0.002
Token in
Token ID287
Feature activation-2.32e+00

Pos logit contributions
loo+0.027
iren+0.024
auc+0.024
ick+0.023
 outage+0.022
Neg logit contributions
chard-0.030
 Eva-0.028
 Robert-0.028
 wrists-0.027
 Claire-0.027
Token a
Token ID257
Feature activation-7.92e-01

Pos logit contributions
 Eva+0.087
 Robert+0.086
chard+0.086
 wrists+0.084
 blonde+0.082
Neg logit contributions
loo-0.103
 of-0.090
ick-0.089
auc-0.089
iren-0.089
Token warm
Token ID5814
Feature activation+3.31e+00

Pos logit contributions
chard+0.032
 Eva+0.031
 Robert+0.031
 wrists+0.031
 blonde+0.030
Neg logit contributions
loo-0.033
iren-0.029
auc-0.028
 of-0.028
ick-0.028
TokenShe
Token ID3347
Feature activation+7.19e-01

Pos logit contributions
loo+0.010
 of+0.010
iren+0.009
ick+0.008
auc+0.008
Neg logit contributions
 wrists-0.018
chard-0.018
 blonde-0.017
 Robert-0.017
 Eva-0.017
Token said
Token ID531
Feature activation+7.13e-01

Pos logit contributions
 of+0.025
loo+0.024
auc+0.022
 dri+0.021
ick+0.021
Neg logit contributions
chard-0.032
 Robert-0.031
 Eva-0.030
 blonde-0.029
 wrists-0.029
Token,
Token ID11
Feature activation+2.15e-01

Pos logit contributions
 of+0.017
loo+0.014
ful+0.013
auc+0.012
ick+0.012
Neg logit contributions
chard-0.041
 wrists-0.040
 Eva-0.038
 blonde-0.038
 Robert-0.038
Token "
Token ID366
Feature activation-3.37e+00

Pos logit contributions
 of+0.008
loo+0.007
iren+0.007
ful+0.007
arl+0.007
Neg logit contributions
 wrists-0.009
chard-0.008
 blonde-0.008
 Eva-0.008
 Robert-0.008
TokenTom
Token ID13787
Feature activation+6.56e-01

Pos logit contributions
 wrists+0.145
chard+0.144
 Eva+0.138
 Robert+0.138
 blonde+0.134
Neg logit contributions
loo-0.115
 of-0.108
iren-0.105
pour-0.102
auc-0.099
Token,
Token ID11
Feature activation+1.75e-01

Pos logit contributions
loo+0.018
 of+0.015
ick+0.014
iren+0.014
auc+0.014
Neg logit contributions
 Eva-0.035
 Robert-0.035
chard-0.035
 wrists-0.035
 blonde-0.034
Token why
Token ID1521
Feature activation-6.35e+00

Pos logit contributions
loo+0.006
 of+0.006
ick+0.006
iren+0.005
auc+0.005
Neg logit contributions
chard-0.008
 Eva-0.008
 Robert-0.008
 wrists-0.008
 blonde-0.007
Token don
Token ID836
Feature activation+6.64e-01

Pos logit contributions
 Eva+0.271
 Robert+0.269
chard+0.262
 wrists+0.258
 blonde+0.251
Neg logit contributions
loo-0.246
iren-0.215
ick-0.211
auc-0.207
 of-0.203
Token't
Token ID470
Feature activation+3.53e+00

Pos logit contributions
loo+0.031
auc+0.027
 ig+0.027
row+0.026
ick+0.026
Neg logit contributions
 Robert-0.025
 Eva-0.023
 Stella-0.022
chard-0.022
 Edward-0.021
Token you
Token ID345
Feature activation+3.91e-01

Pos logit contributions
loo+0.147
iren+0.123
auc+0.123
ick+0.121
arl+0.120
Neg logit contributions
 Robert-0.157
 Eva-0.152
chard-0.148
 wrists-0.146
 Claire-0.145
Token take
Token ID1011
Feature activation+2.29e-01

Pos logit contributions
loo+0.015
 of+0.012
ick+0.012
iren+0.012
auc+0.012
Neg logit contributions
 Eva-0.017
chard-0.017
 Robert-0.017
 wrists-0.017
 blonde-0.016
Token tried
Token ID3088
Feature activation+2.20e+00

Pos logit contributions
 Eva+0.013
 Robert+0.013
chard+0.013
 wrists+0.013
 blonde+0.013
Neg logit contributions
loo-0.013
 of-0.011
ick-0.011
iren-0.011
auc-0.011
Token to
Token ID284
Feature activation+2.54e-01

Pos logit contributions
loo+0.067
 of+0.058
ick+0.055
iren+0.054
auc+0.053
Neg logit contributions
 Eva-0.110
chard-0.110
 Robert-0.109
 wrists-0.109
 blonde-0.106
Token dig
Token ID3100
Feature activation+1.41e+00

Pos logit contributions
loo+0.010
 of+0.008
ick+0.008
iren+0.008
auc+0.008
Neg logit contributions
 Eva-0.011
 Robert-0.011
chard-0.011
 wrists-0.011
 blonde-0.010
Token a
Token ID257
Feature activation-6.21e-01

Pos logit contributions
loo+0.050
 of+0.045
ick+0.043
auc+0.042
iren+0.041
Neg logit contributions
chard-0.064
 wrists-0.064
 Eva-0.064
 Robert-0.064
 blonde-0.062
Token hole
Token ID7604
Feature activation+6.96e+00

Pos logit contributions
chard+0.032
 wrists+0.031
 Eva+0.031
 Robert+0.031
 blonde+0.030
Neg logit contributions
loo-0.018
iren-0.015
ick-0.015
auc-0.014
arl-0.014
Token for
Token ID329
Feature activation+1.49e+00

Pos logit contributions
loo+0.237
iren+0.202
ick+0.195
auc+0.191
arl+0.183
Neg logit contributions
chard-0.325
 Eva-0.324
 Robert-0.322
 wrists-0.313
 Claire-0.306
Token more
Token ID517
Feature activation-3.46e-01

Pos logit contributions
loo+0.061
auc+0.053
iren+0.052
ick+0.052
 of+0.052
Neg logit contributions
 Eva-0.061
 Robert-0.060
chard-0.060
 wrists-0.059
 blonde-0.057
Token light
Token ID1657
Feature activation+3.76e+00

Pos logit contributions
 wrists+0.014
 Robert+0.014
 blonde+0.014
chard+0.014
 Eva+0.014
Neg logit contributions
loo-0.013
iren-0.012
ick-0.012
auc-0.011
arl-0.011
Token,
Token ID11
Feature activation+1.47e-01

Pos logit contributions
loo+0.126
iren+0.107
ick+0.106
auc+0.102
 of+0.099
Neg logit contributions
 Eva-0.175
 wrists-0.175
 Robert-0.173
chard-0.172
 blonde-0.166
Token but
Token ID475
Feature activation-2.16e+00

Pos logit contributions
loo+0.007
 of+0.006
ick+0.006
auc+0.006
iren+0.006
Neg logit contributions
 Eva-0.005
 Robert-0.005
chard-0.005
 wrists-0.005
 blonde-0.005
Token the
Token ID262
Feature activation+2.86e-01

Pos logit contributions
 wrists+0.089
 Eva+0.088
 Robert+0.088
chard+0.087
 blonde+0.086
Neg logit contributions
loo-0.085
 of-0.076
ick-0.073
iren-0.070
auc-0.070
Token<|endoftext|>
Token ID50256
Feature activation+4.89e+00

Pos logit contributions
loo+0.022
 of+0.021
ick+0.018
iren+0.018
pour+0.017
Neg logit contributions
chard-0.032
 wrists-0.032
 blonde-0.029
 Eva-0.029
 Robert-0.029
TokenOnce
Token ID7454
Feature activation+1.10e+00

Pos logit contributions
loo+0.175
 of+0.153
ick+0.148
iren+0.143
auc+0.141
Neg logit contributions
 wrists-0.225
chard-0.224
 Eva-0.221
 Robert-0.220
 blonde-0.214
Token upon
Token ID2402
Feature activation+2.43e+00

Pos logit contributions
loo+0.038
iren+0.033
auc+0.033
ick+0.031
pour+0.031
Neg logit contributions
 Eva-0.050
 Robert-0.050
chard-0.049
 wrists-0.047
 Claire-0.047
Token a
Token ID257
Feature activation-6.76e-01

Pos logit contributions
 of+0.074
loo+0.069
ick+0.060
 late+0.059
 ig+0.055
Neg logit contributions
 wrists-0.129
chard-0.121
 Robert-0.117
 Eva-0.116
rice-0.113
Token time
Token ID640
Feature activation+2.77e+00

Pos logit contributions
chard+0.040
 wrists+0.039
 Robert+0.037
 Eva+0.037
nam+0.036
Neg logit contributions
loo-0.015
 of-0.014
ick-0.014
arl-0.011
 ig-0.011
Token,
Token ID11
Feature activation+1.12e+00

Pos logit contributions
loo+0.080
 of+0.065
ick+0.064
iren+0.064
auc+0.063
Neg logit contributions
 Eva-0.145
 Robert-0.144
chard-0.144
 wrists-0.142
 blonde-0.139
Token there
Token ID612
Feature activation-1.18e+00

Pos logit contributions
loo+0.030
 of+0.026
ick+0.024
auc+0.022
iren+0.022
Neg logit contributions
 wrists-0.062
 Robert-0.061
 Eva-0.061
chard-0.060
 blonde-0.058
Token was
Token ID373
Feature activation+9.65e-01

Pos logit contributions
 Eva+0.052
 Robert+0.052
chard+0.052
 wrists+0.051
 blonde+0.050
Neg logit contributions
loo-0.043
 of-0.037
ick-0.037
iren-0.037
auc-0.036
Token a
Token ID257
Feature activation-6.75e-01

Pos logit contributions
 of+0.029
loo+0.028
ick+0.025
 late+0.024
 ig+0.024
Neg logit contributions
 wrists-0.050
chard-0.047
 Eva-0.046
rice-0.046
 Robert-0.045
Token little
Token ID1310
Feature activation+1.71e+00

Pos logit contributions
 wrists+0.031
 Eva+0.031
 Robert+0.030
chard+0.029
 blonde+0.029
Neg logit contributions
loo-0.024
 of-0.021
ick-0.021
 ig-0.020
auc-0.019
Token girl
Token ID2576
Feature activation-1.16e+00

Pos logit contributions
loo+0.071
iren+0.062
auc+0.061
ick+0.061
 of+0.061
Neg logit contributions
chard-0.067
 Robert-0.066
 Eva-0.066
 wrists-0.065
 blonde-0.063
Token.
Token ID13
Feature activation+9.99e-01

Pos logit contributions
chard+0.030
 Eva+0.029
 Robert+0.029
 wrists+0.029
 blonde+0.029
Neg logit contributions
loo-0.012
 of-0.012
ick-0.010
iren-0.009
auc-0.009
Token
Token ID198
Feature activation+4.01e-01

Pos logit contributions
loo+0.038
 of+0.034
iren+0.033
ick+0.033
auc+0.032
Neg logit contributions
chard-0.043
 Eva-0.043
 wrists-0.043
 Robert-0.042
 blonde-0.041
Token
Token ID198
Feature activation+4.20e-01

Pos logit contributions
loo+0.016
 of+0.014
ick+0.014
iren+0.013
auc+0.013
Neg logit contributions
 Eva-0.017
chard-0.017
 Robert-0.017
 wrists-0.017
 blonde-0.016
TokenHe
Token ID1544
Feature activation+3.24e-01

Pos logit contributions
loo+0.019
ick+0.017
auc+0.017
 of+0.016
iren+0.016
Neg logit contributions
 Robert-0.015
chard-0.015
 Eva-0.015
 wrists-0.014
 blonde-0.014
Token runs
Token ID4539
Feature activation-6.61e-01

Pos logit contributions
loo+0.010
 of+0.010
auc+0.009
iren+0.009
ick+0.009
Neg logit contributions
chard-0.016
 Robert-0.015
 Eva-0.015
 wrists-0.015
 blonde-0.015
Token to
Token ID284
Feature activation-1.98e-01

Pos logit contributions
 wrists+0.034
chard+0.034
 Eva+0.034
 Robert+0.034
 blonde+0.033
Neg logit contributions
loo-0.019
 of-0.017
iren-0.016
ick-0.016
auc-0.015
Token the
Token ID262
Feature activation+4.94e-01

Pos logit contributions
 wrists+0.010
 blonde+0.010
cious+0.010
chard+0.010
 Eva+0.010
Neg logit contributions
loo-0.005
 of-0.005
iren-0.005
ick-0.004
pour-0.004
Token musician
Token ID21623
Feature activation-4.10e+00

Pos logit contributions
loo+0.020
 of+0.017
ick+0.017
iren+0.017
auc+0.017
Neg logit contributions
 Eva-0.020
chard-0.020
 Robert-0.020
 wrists-0.020
 blonde-0.019
Token and
Token ID290
Feature activation+5.79e-01

Pos logit contributions
chard+0.212
 Eva+0.211
 Robert+0.210
 wrists+0.210
 blonde+0.209
Neg logit contributions
loo-0.115
 of-0.105
iren-0.094
ick-0.093
auc-0.090
Token says
Token ID1139
Feature activation+3.79e-01

Pos logit contributions
loo+0.018
 of+0.017
iren+0.017
auc+0.015
 dri+0.014
Neg logit contributions
chard-0.027
 blonde-0.027
 Robert-0.026
 Eva-0.026
 wrists-0.026
Token,
Token ID11
Feature activation+7.53e-01

Pos logit contributions
 of+0.011
loo+0.011
iren+0.009
auc+0.009
ick+0.009
Neg logit contributions
 wrists-0.020
chard-0.020
 Eva-0.019
 blonde-0.019
rice-0.018
Token hon
Token ID3032
Feature activation-1.84e+00

Pos logit contributions
loo+0.049
ick+0.042
ained+0.041
auc+0.041
pour+0.039
Neg logit contributions
 Eva-0.050
 Robert-0.049
 wrists-0.047
 Claire-0.045
 blonde-0.045
Tokenk
Token ID74
Feature activation-2.75e+00

Pos logit contributions
 wrists+0.051
 Eva+0.050
rice+0.049
cious+0.049
 Robert+0.048
Neg logit contributions
loo-0.097
pour-0.082
auc-0.081
ick-0.079
 of-0.079
Token.
Token ID13
Feature activation+9.69e-01

Pos logit contributions
 Eva+0.100
 Robert+0.097
 yes+0.095
 wrists+0.093
 Edward+0.091
Neg logit contributions
loo-0.111
ick-0.092
auc-0.091
rim-0.089
iren-0.087
Token They
Token ID1119
Feature activation+5.26e-02

Pos logit contributions
loo+0.041
 of+0.036
ick+0.035
auc+0.035
iren+0.035
Neg logit contributions
chard-0.038
 Robert-0.038
 Eva-0.038
 wrists-0.037
 blonde-0.036
Token see
Token ID766
Feature activation-1.16e-02

Pos logit contributions
loo+0.002
 of+0.002
ick+0.002
iren+0.002
auc+0.002
Neg logit contributions
 Eva-0.002
 Robert-0.002
chard-0.002
 wrists-0.002
 blonde-0.002
Token a
Token ID257
Feature activation-1.13e+00

Pos logit contributions
 Eva+0.000
 Robert+0.000
 wrists+0.000
chard+0.000
 blonde+0.000
Neg logit contributions
loo-0.001
ick-0.000
iren-0.000
auc-0.000
arl-0.000
Token big
Token ID1263
Feature activation+2.37e+00

Pos logit contributions
chard+0.046
 Robert+0.045
 Eva+0.044
 wrists+0.044
 blonde+0.042
Neg logit contributions
loo-0.043
auc-0.039
ick-0.039
 of-0.038
iren-0.036
Token truck
Token ID7779
Feature activation-1.17e-01

Pos logit contributions
loo+0.093
auc+0.088
ick+0.086
ained+0.085
rim+0.081
Neg logit contributions
 Robert-0.092
 wrists-0.090
 Eva-0.090
chard-0.089
 blonde-0.085
Token.
Token ID13
Feature activation+9.70e-01

Pos logit contributions
 Robert+0.004
chard+0.004
 Eva+0.004
 wrists+0.004
 yes+0.004
Neg logit contributions
loo-0.004
ick-0.004
auc-0.004
rim-0.004
arl-0.004
Token The
Token ID383
Feature activation-1.52e-01

Pos logit contributions
loo+0.049
 of+0.044
ick+0.044
iren+0.043
auc+0.043
Neg logit contributions
 Eva-0.029
 Robert-0.029
chard-0.029
 wrists-0.029
 blonde-0.028
Token truck
Token ID7779
Feature activation-1.34e-01

Pos logit contributions
 Robert+0.005
chard+0.005
 wrists+0.005
 yes+0.005
 blonde+0.005
Neg logit contributions
loo-0.006
ful-0.006
auc-0.006
angle-0.006
ick-0.006
Token!
Token ID0
Feature activation-2.13e-03

Pos logit contributions
 Robert+0.119
 Eva+0.119
chard+0.119
 wrists+0.117
 blonde+0.114
Neg logit contributions
loo-0.075
ick-0.061
iren-0.061
auc-0.060
 of-0.060
Token He
Token ID679
Feature activation-7.88e-01

Pos logit contributions
 wrists+0.000
chard+0.000
 Eva+0.000
 Robert+0.000
 blonde+0.000
Neg logit contributions
loo-0.000
 of-0.000
iren-0.000
ick-0.000
auc-0.000
Token said
Token ID531
Feature activation+8.37e-01

Pos logit contributions
 wrists+0.037
 Eva+0.037
chard+0.036
 Robert+0.036
 blonde+0.035
Neg logit contributions
 of-0.026
loo-0.026
ick-0.022
 ig-0.021
auc-0.021
Token he
Token ID339
Feature activation+9.04e-01

Pos logit contributions
 of+0.018
loo+0.018
ick+0.013
auc+0.013
iren+0.012
Neg logit contributions
 wrists-0.050
chard-0.049
 Eva-0.048
 blonde-0.048
 Robert-0.047
Token was
Token ID373
Feature activation+5.69e-01

Pos logit contributions
 of+0.028
loo+0.026
ick+0.024
auc+0.021
 ig+0.021
Neg logit contributions
 wrists-0.046
chard-0.044
 Eva-0.044
 Robert-0.043
 Stella-0.041
Token here
Token ID994
Feature activation-3.04e+00

Pos logit contributions
 of+0.018
loo+0.017
ick+0.016
 ig+0.014
 late+0.014
Neg logit contributions
 wrists-0.029
 Eva-0.026
 blonde-0.026
chard-0.026
cious-0.026
Token to
Token ID284
Feature activation-1.77e-01

Pos logit contributions
 wrists+0.159
 Eva+0.159
chard+0.158
 Robert+0.157
 blonde+0.153
Neg logit contributions
loo-0.086
 of-0.077
ick-0.070
auc-0.066
iren-0.066
Token help
Token ID1037
Feature activation+2.56e+00

Pos logit contributions
 Eva+0.008
chard+0.008
 Robert+0.008
 wrists+0.008
 blonde+0.007
Neg logit contributions
loo-0.007
 of-0.006
ick-0.006
iren-0.005
auc-0.005
Token her
Token ID607
Feature activation-2.57e-01

Pos logit contributions
loo+0.083
 of+0.069
ick+0.068
iren+0.067
auc+0.066
Neg logit contributions
 Eva-0.126
 Robert-0.126
chard-0.125
 wrists-0.125
 blonde-0.120
Token make
Token ID787
Feature activation+6.90e-01

Pos logit contributions
 Eva+0.012
chard+0.012
 Robert+0.012
 wrists+0.012
 blonde+0.012
Neg logit contributions
loo-0.009
 of-0.007
ick-0.007
iren-0.007
auc-0.007
Token friends
Token ID2460
Feature activation-1.09e+00

Pos logit contributions
loo+0.027
 of+0.024
ick+0.024
iren+0.023
auc+0.023
Neg logit contributions
chard-0.029
 Eva-0.028
 Robert-0.028
 wrists-0.028
 blonde-0.027
Token she
Token ID673
Feature activation+1.49e+00

Pos logit contributions
 of+0.022
loo+0.022
ick+0.019
iren+0.017
pour+0.017
Neg logit contributions
 wrists-0.036
 Eva-0.036
 Robert-0.036
cious-0.035
chard-0.034
Token tried
Token ID3088
Feature activation+1.02e+00

Pos logit contributions
loo+0.073
arl+0.066
iren+0.066
ick+0.064
pour+0.062
Neg logit contributions
 Eva-0.048
 Robert-0.048
chard-0.047
 wrists-0.046
 blonde-0.045
Token to
Token ID284
Feature activation-1.08e-01

Pos logit contributions
loo+0.035
 of+0.029
ick+0.029
iren+0.028
auc+0.028
Neg logit contributions
 Eva-0.049
 Robert-0.048
chard-0.048
 wrists-0.048
 blonde-0.046
Token climb
Token ID12080
Feature activation+1.06e+00

Pos logit contributions
 wrists+0.004
chard+0.004
 Eva+0.004
 Robert+0.004
 blonde+0.004
Neg logit contributions
loo-0.005
iren-0.004
auc-0.004
arl-0.004
ick-0.004
Token up
Token ID510
Feature activation-5.68e+00

Pos logit contributions
loo+0.035
 of+0.031
ick+0.030
iren+0.029
auc+0.028
Neg logit contributions
 wrists-0.050
 Eva-0.050
 Robert-0.049
chard-0.049
 blonde-0.048
Token on
Token ID319
Feature activation-1.89e+00

Pos logit contributions
 Eva+0.239
 Robert+0.238
chard+0.234
 wrists+0.231
 blonde+0.229
Neg logit contributions
loo-0.227
auc-0.194
iren-0.189
ick-0.188
arl-0.184
Token a
Token ID257
Feature activation-9.13e-01

Pos logit contributions
chard+0.090
 Eva+0.088
 Robert+0.088
 wrists+0.088
 blonde+0.084
Neg logit contributions
loo-0.061
 of-0.055
ick-0.053
iren-0.051
pour-0.049
Token chair
Token ID5118
Feature activation-3.68e-01

Pos logit contributions
 Eva+0.040
chard+0.039
 Robert+0.039
 wrists+0.039
 blonde+0.038
Neg logit contributions
loo-0.034
 of-0.029
ick-0.029
iren-0.029
auc-0.029
Token to
Token ID284
Feature activation-9.63e-02

Pos logit contributions
 Robert+0.015
chard+0.015
 Eva+0.015
 blonde+0.015
 wrists+0.014
Neg logit contributions
loo-0.014
arl-0.012
rim-0.012
auc-0.012
iren-0.012
Token put
Token ID1234
Feature activation+1.41e-01

Pos logit contributions
 Robert+0.005
 wrists+0.005
 Eva+0.005
chard+0.004
 blonde+0.004
Neg logit contributions
loo-0.003
 of-0.003
ick-0.003
iren-0.003
auc-0.003
Token a
Token ID257
Feature activation-9.37e-01

Pos logit contributions
loo+0.005
 of+0.004
ick+0.004
iren+0.004
auc+0.004
Neg logit contributions
 Eva-0.007
chard-0.007
 Robert-0.007
 wrists-0.007
 blonde-0.006
Token.
Token ID13
Feature activation+7.87e-01

Pos logit contributions
loo+0.058
iren+0.047
auc+0.047
ick+0.047
arl+0.045
Neg logit contributions
 wrists-0.068
chard-0.067
 Robert-0.066
 Eva-0.066
 blonde-0.065
Token The
Token ID383
Feature activation+1.63e-01

Pos logit contributions
loo+0.027
 of+0.027
iren+0.025
ick+0.024
 late+0.024
Neg logit contributions
chard-0.035
 wrists-0.035
cious-0.033
 Eva-0.033
 blonde-0.032
Token frog
Token ID21264
Feature activation-3.44e+00

Pos logit contributions
loo+0.006
 of+0.005
ick+0.005
iren+0.005
auc+0.005
Neg logit contributions
 Robert-0.007
 Eva-0.007
 wrists-0.007
chard-0.007
 blonde-0.007
Token was
Token ID373
Feature activation+7.52e-01

Pos logit contributions
 Eva+0.142
 Robert+0.142
chard+0.142
 wrists+0.141
 blonde+0.136
Neg logit contributions
loo-0.136
 of-0.121
ick-0.117
iren-0.115
auc-0.114
Token so
Token ID523
Feature activation-2.80e+00

Pos logit contributions
loo+0.033
auc+0.029
ick+0.028
iren+0.028
 of+0.027
Neg logit contributions
chard-0.029
 Eva-0.029
 Robert-0.028
 wrists-0.028
 blonde-0.027
Token shocked
Token ID11472
Feature activation-6.97e-02

Pos logit contributions
chard+0.094
 Eva+0.089
 Robert+0.088
 blonde+0.085
 wrists+0.085
Neg logit contributions
loo-0.138
auc-0.122
ick-0.120
 ig-0.113
iren-0.112
Token he
Token ID339
Feature activation+1.05e+00

Pos logit contributions
 Robert+0.003
 Eva+0.003
chard+0.003
 wrists+0.003
 Stella+0.003
Neg logit contributions
loo-0.003
auc-0.002
ick-0.002
iren-0.002
arl-0.002
Token jumped
Token ID11687
Feature activation-3.60e-01

Pos logit contributions
loo+0.040
 of+0.037
ick+0.035
iren+0.034
auc+0.033
Neg logit contributions
 Eva-0.045
 Robert-0.044
chard-0.044
 wrists-0.043
 blonde-0.043
Token straight
Token ID3892
Feature activation-7.04e-01

Pos logit contributions
 Robert+0.014
chard+0.014
 Eva+0.014
 wrists+0.013
 Claire+0.013
Neg logit contributions
loo-0.015
ick-0.012
iren-0.012
auc-0.012
 ig-0.012
Token out
Token ID503
Feature activation-2.15e+00

Pos logit contributions
 Robert+0.032
 Eva+0.031
 wrists+0.031
chard+0.031
 blonde+0.030
Neg logit contributions
loo-0.026
ick-0.022
auc-0.021
iren-0.021
 of-0.021
Token of
Token ID286
Feature activation+3.37e+00

Pos logit contributions
 Eva+0.105
 Joy+0.103
 yes+0.103
 aloud+0.102
 Robert+0.099
Neg logit contributions
loo-0.064
ained-0.053
auc-0.047
ick-0.046
w-0.044
Token boy
Token ID2933
Feature activation-9.14e-01

Pos logit contributions
loo+0.022
 of+0.019
iren+0.019
ick+0.019
auc+0.019
Neg logit contributions
 Robert-0.026
chard-0.026
 Eva-0.026
 wrists-0.025
 blonde-0.024
Token was
Token ID373
Feature activation+7.82e-01

Pos logit contributions
 Eva+0.042
 wrists+0.041
chard+0.041
 Robert+0.040
 blonde+0.040
Neg logit contributions
loo-0.032
 of-0.031
ick-0.027
 ig-0.027
auc-0.026
Token gone
Token ID3750
Feature activation+4.51e+00

Pos logit contributions
loo+0.028
 of+0.026
ick+0.024
iren+0.024
auc+0.023
Neg logit contributions
 wrists-0.036
 Eva-0.035
 Robert-0.035
chard-0.034
 blonde-0.034
Token.
Token ID13
Feature activation+8.72e-01

Pos logit contributions
loo+0.165
auc+0.144
iren+0.143
ick+0.134
rim+0.131
Neg logit contributions
 Robert-0.203
chard-0.202
 Eva-0.196
 wrists-0.193
 blonde-0.189
Token "
Token ID366
Feature activation-3.14e+00

Pos logit contributions
loo+0.031
 of+0.027
iren+0.026
auc+0.025
ick+0.025
Neg logit contributions
 wrists-0.040
chard-0.039
 Eva-0.038
 Robert-0.038
 yes-0.037
TokenWhere
Token ID8496
Feature activation-5.05e+00

Pos logit contributions
 Robert+0.115
chard+0.115
 Eva+0.114
 wrists+0.111
 blonde+0.108
Neg logit contributions
loo-0.142
ick-0.125
 of-0.124
auc-0.122
iren-0.122
Token are
Token ID389
Feature activation+8.58e-01

Pos logit contributions
 Robert+0.199
 Eva+0.198
chard+0.191
 blonde+0.182
 yes+0.181
Neg logit contributions
loo-0.208
iren-0.186
ick-0.182
arl-0.178
auc-0.178
Token you
Token ID345
Feature activation+3.38e-01

Pos logit contributions
loo+0.024
 of+0.019
ick+0.019
iren+0.019
auc+0.019
Neg logit contributions
 Robert-0.046
 Eva-0.046
chard-0.046
 wrists-0.045
 blonde-0.044
Token?"
Token ID1701
Feature activation-3.21e+00

Pos logit contributions
loo+0.013
iren+0.012
auc+0.012
ick+0.011
 of+0.011
Neg logit contributions
 Robert-0.014
chard-0.014
 Eva-0.014
 Claire-0.013
 wrists-0.013
Token she
Token ID673
Feature activation+1.38e+00

Pos logit contributions
 wrists+0.151
chard+0.147
 Eva+0.145
 Robert+0.144
 blonde+0.141
Neg logit contributions
loo-0.102
 of-0.099
iren-0.090
ick-0.085
auc-0.082
Token shouted
Token ID17293
Feature activation+4.44e-01

Pos logit contributions
 of+0.042
loo+0.040
ick+0.037
iren+0.037
 ig+0.036
Neg logit contributions
chard-0.061
 wrists-0.061
 Eva-0.060
 Robert-0.060
 blonde-0.058
Token is
Token ID318
Feature activation+8.71e-01

Pos logit contributions
loo+0.006
ick+0.005
iren+0.005
 of+0.005
auc+0.005
Neg logit contributions
 Robert-0.011
 Eva-0.011
 wrists-0.011
chard-0.011
 blonde-0.010
Token the
Token ID262
Feature activation+2.28e-01

Pos logit contributions
loo+0.029
 of+0.027
ick+0.024
 ig+0.024
 tunnel+0.023
Neg logit contributions
 wrists-0.042
chard-0.041
 blonde-0.040
 Eva-0.039
cious-0.038
Token best
Token ID1266
Feature activation-3.63e+00

Pos logit contributions
loo+0.008
 of+0.007
ick+0.007
iren+0.007
auc+0.007
Neg logit contributions
 Eva-0.010
chard-0.010
 Robert-0.010
 wrists-0.010
 blonde-0.010
Token summer
Token ID3931
Feature activation+2.85e+00

Pos logit contributions
 Robert+0.162
 Eva+0.153
 wrists+0.150
 Claire+0.147
 Stella+0.143
Neg logit contributions
loo-0.122
iren-0.120
ick-0.116
auc-0.115
arl-0.114
Token ever
Token ID1683
Feature activation+2.00e+00

Pos logit contributions
loo+0.086
arl+0.080
iren+0.077
ick+0.076
auc+0.076
Neg logit contributions
 Robert-0.141
 Eva-0.139
 wrists-0.131
 Claire-0.131
chard-0.129
Token!"
Token ID2474
Feature activation-5.08e+00

Pos logit contributions
loo+0.072
iren+0.062
ick+0.062
auc+0.061
 of+0.059
Neg logit contributions
 Eva-0.090
 Robert-0.090
chard-0.087
 wrists-0.087
 Claire-0.083
Token Anna
Token ID11735
Feature activation-3.10e+00

Pos logit contributions
chard+0.215
 wrists+0.212
 Robert+0.212
 Eva+0.212
 blonde+0.205
Neg logit contributions
loo-0.196
 of-0.176
ick-0.168
iren-0.167
auc-0.166
Token says
Token ID1139
Feature activation+2.69e-01

Pos logit contributions
chard+0.164
 blonde+0.154
 Robert+0.154
 wrists+0.154
 Eva+0.152
Neg logit contributions
loo-0.088
 of-0.086
ick-0.074
iren-0.074
auc-0.074
Token.
Token ID13
Feature activation+8.19e-01

Pos logit contributions
loo+0.006
 of+0.006
ick+0.005
iren+0.005
auc+0.005
Neg logit contributions
chard-0.015
 Robert-0.015
 Eva-0.015
 wrists-0.015
 blonde-0.015
Token
Token ID198
Feature activation-1.93e-02

Pos logit contributions
 of+0.027
loo+0.025
iren+0.024
arl+0.023
ick+0.022
Neg logit contributions
chard-0.039
 wrists-0.039
 Robert-0.038
 Eva-0.037
cious-0.037
Token
Token ID198
Feature activation-1.77e-03

Pos logit contributions
 Eva+0.001
 Robert+0.001
chard+0.001
 wrists+0.001
 blonde+0.001
Neg logit contributions
loo-0.001
 of-0.001
ick-0.001
auc-0.001
iren-0.001
Tokenze
Token ID2736
Feature activation-2.89e+00

Pos logit contributions
chard+0.090
 Robert+0.089
 Eva+0.087
 wrists+0.085
 Claire+0.081
Neg logit contributions
loo-0.100
 of-0.091
ick-0.089
iren-0.087
auc-0.087
Token makes
Token ID1838
Feature activation+1.43e+00

Pos logit contributions
 Eva+0.122
 wrists+0.121
 Robert+0.120
 blonde+0.116
chard+0.113
Neg logit contributions
loo-0.111
ick-0.093
arl-0.090
auc-0.090
 ig-0.088
Token the
Token ID262
Feature activation+4.84e-01

Pos logit contributions
loo+0.064
auc+0.054
iren+0.054
ick+0.054
arl+0.053
Neg logit contributions
 Eva-0.053
 Robert-0.052
 wrists-0.051
 blonde-0.050
chard-0.048
Token blocks
Token ID7021
Feature activation-4.77e-01

Pos logit contributions
loo+0.018
ick+0.016
 of+0.016
iren+0.016
auc+0.016
Neg logit contributions
 wrists-0.020
 Eva-0.020
chard-0.020
 Robert-0.020
 blonde-0.020
Token fall
Token ID2121
Feature activation+4.91e+00

Pos logit contributions
 yes+0.018
 wrists+0.016
 blonde+0.016
 Eva+0.015
 aloud+0.015
Neg logit contributions
loo-0.021
enn-0.018
iren-0.018
rim-0.017
arl-0.017
Token down
Token ID866
Feature activation-4.92e+00

Pos logit contributions
loo+0.232
rim+0.203
auc+0.189
iren+0.180
enn+0.177
Neg logit contributions
 wrists-0.157
 aloud-0.152
 blonde-0.152
 Robert-0.150
 yes-0.145
Token.
Token ID13
Feature activation+1.18e+00

Pos logit contributions
 Robert+0.183
 Eva+0.176
chard+0.174
 aloud+0.168
 Claire+0.163
Neg logit contributions
loo-0.202
rim-0.186
auc-0.176
arl-0.175
row-0.168
Token The
Token ID383
Feature activation+5.30e-02

Pos logit contributions
loo+0.039
 of+0.038
ick+0.034
auc+0.033
iren+0.033
Neg logit contributions
chard-0.056
 wrists-0.055
cious-0.052
 Robert-0.052
 blonde-0.052
Token building
Token ID2615
Feature activation-3.75e-02

Pos logit contributions
loo+0.002
ick+0.002
arl+0.002
auc+0.002
iren+0.002
Neg logit contributions
 wrists-0.002
 Eva-0.002
chard-0.002
 blonde-0.002
 Robert-0.002
Token is
Token ID318
Feature activation+1.13e+00

Pos logit contributions
 wrists+0.002
 Eva+0.002
 blonde+0.001
 Robert+0.001
 yes+0.001
Neg logit contributions
loo-0.001
arl-0.001
rim-0.001
ick-0.001
pour-0.001
Token gone
Token ID3750
Feature activation+4.36e+00

Pos logit contributions
loo+0.050
auc+0.042
ick+0.041
iren+0.040
arp+0.039
Neg logit contributions
chard-0.044
 blonde-0.043
 wrists-0.042
 Eva-0.042
 Robert-0.041
Token,
Token ID11
Feature activation+2.55e-01

Pos logit contributions
 wrists+0.313
chard+0.311
 Eva+0.309
 Robert+0.307
 blonde+0.294
Neg logit contributions
loo-0.121
 of-0.107
ick-0.099
 ig-0.085
pour-0.084
Token little
Token ID1310
Feature activation+1.50e+00

Pos logit contributions
loo+0.011
 of+0.010
ick+0.009
iren+0.009
auc+0.009
Neg logit contributions
 Eva-0.010
chard-0.010
 Robert-0.010
 wrists-0.010
 blonde-0.009
Token spirit
Token ID4437
Feature activation+4.95e-01

Pos logit contributions
loo+0.055
 of+0.048
ick+0.047
auc+0.045
iren+0.045
Neg logit contributions
 Eva-0.067
chard-0.067
 Robert-0.067
 wrists-0.067
 blonde-0.063
Token,
Token ID11
Feature activation+2.75e-01

Pos logit contributions
loo+0.015
ick+0.012
iren+0.012
auc+0.012
 of+0.012
Neg logit contributions
 Robert-0.025
 Eva-0.025
chard-0.025
 wrists-0.024
 blonde-0.024
Tokenâ
Token ID22940
Feature activation+3.74e+00

Pos logit contributions
loo+0.010
 of+0.009
ick+0.009
auc+0.008
iren+0.008
Neg logit contributions
 wrists-0.012
chard-0.012
 Eva-0.012
 Robert-0.012
 blonde-0.012
Token
Token ID26391
Feature activation-6.38e+00

Pos logit contributions
loo+0.201
 of+0.181
ick+0.180
iren+0.178
auc+0.177
Neg logit contributions
 Eva-0.102
chard-0.102
 Robert-0.102
 wrists-0.102
 blonde-0.095
Token said
Token ID531
Feature activation+8.94e-01

Pos logit contributions
chard+0.260
 wrists+0.258
 Eva+0.257
 Robert+0.255
 Claire+0.242
Neg logit contributions
loo-0.254
 of-0.223
ick-0.223
iren-0.214
auc-0.209
Token F
Token ID376
Feature activation+6.13e-02

Pos logit contributions
loo+0.021
 of+0.020
ick+0.017
auc+0.015
iren+0.015
Neg logit contributions
 wrists-0.052
chard-0.051
 Eva-0.050
 Robert-0.049
 blonde-0.049
Tokenawn
Token ID3832
Feature activation-2.93e+00

Pos logit contributions
loo+0.002
ick+0.002
auc+0.002
 of+0.002
arl+0.002
Neg logit contributions
 wrists-0.003
 Eva-0.003
 Robert-0.003
chard-0.003
 blonde-0.003
Token.
Token ID13
Feature activation+4.99e-01

Pos logit contributions
chard+0.166
 Eva+0.166
 wrists+0.165
 Robert+0.165
 blonde+0.161
Neg logit contributions
loo-0.073
 of-0.061
ick-0.058
auc-0.055
iren-0.053
Token
Token ID6184
Feature activation-1.37e+00

Pos logit contributions
loo+0.020
 of+0.019
ick+0.018
iren+0.018
auc+0.018
Neg logit contributions
chard-0.020
 Eva-0.020
 wrists-0.020
 Robert-0.019
 blonde-0.018
Token the
Token ID262
Feature activation+3.83e-01

Pos logit contributions
 of+0.001
loo+0.001
ick+0.001
auc+0.001
iren+0.001
Neg logit contributions
 wrists-0.003
chard-0.003
 blonde-0.003
 Eva-0.003
cious-0.003
Token old
Token ID1468
Feature activation+8.96e-01

Pos logit contributions
loo+0.014
 of+0.012
ick+0.012
auc+0.012
iren+0.012
Neg logit contributions
 wrists-0.017
 Eva-0.017
 Robert-0.017
chard-0.017
 blonde-0.016
Token man
Token ID582
Feature activation-1.32e+00

Pos logit contributions
loo+0.033
 of+0.029
iren+0.028
ick+0.027
auc+0.027
Neg logit contributions
 Eva-0.040
chard-0.040
 wrists-0.039
 Robert-0.039
 blonde-0.038
Token,
Token ID11
Feature activation+2.96e-01

Pos logit contributions
 wrists+0.074
 blonde+0.073
chard+0.069
 Eva+0.069
 Robert+0.069
Neg logit contributions
 of-0.036
loo-0.032
ick-0.026
auc-0.026
ful-0.026
Token "
Token ID366
Feature activation-3.56e+00

Pos logit contributions
 of+0.013
loo+0.012
ick+0.011
ful+0.011
iren+0.011
Neg logit contributions
 wrists-0.012
 blonde-0.011
 Eva-0.011
cious-0.010
 Robert-0.010
TokenWhat
Token ID2061
Feature activation-7.31e+00

Pos logit contributions
 wrists+0.146
 Eva+0.142
 blonde+0.140
 Robert+0.139
chard+0.136
Neg logit contributions
loo-0.138
 of-0.127
iren-0.123
auc-0.120
pour-0.120
Token is
Token ID318
Feature activation+1.03e+00

Pos logit contributions
 wrists+0.385
chard+0.383
 Eva+0.371
 Robert+0.368
 blonde+0.366
Neg logit contributions
loo-0.211
 of-0.198
ick-0.171
auc-0.164
iren-0.152
Token it
Token ID340
Feature activation+2.33e-01

Pos logit contributions
loo+0.034
 of+0.030
ick+0.029
auc+0.028
iren+0.028
Neg logit contributions
 wrists-0.049
 Eva-0.049
chard-0.048
 Robert-0.047
 blonde-0.047
Token?"
Token ID1701
Feature activation-3.86e+00

Pos logit contributions
loo+0.006
 of+0.005
ick+0.005
iren+0.005
auc+0.005
Neg logit contributions
 Eva-0.012
chard-0.012
 Robert-0.012
 wrists-0.012
 blonde-0.012
Token
Token ID198
Feature activation-4.34e-02

Pos logit contributions
 wrists+0.194
chard+0.185
 blonde+0.184
cious+0.182
 Eva+0.176
Neg logit contributions
 of-0.121
loo-0.112
iren-0.109
pour-0.098
 late-0.096
Token
Token ID198
Feature activation-5.39e-02

Pos logit contributions
 wrists+0.002
chard+0.002
 blonde+0.002
 Eva+0.002
 Robert+0.002
Neg logit contributions
loo-0.001
 of-0.001
iren-0.001
ick-0.001
auc-0.001