TOP ACTIVATIONS
MAX = 0.310

She cried and threw a tant rum . Her mom tried
listen . She threw a tant rum in the store .
and so she threw a tant rum when her parents said
her feet and threw a tant rum . She grabbed the
to cry and threw a tant rum . She didn 't
. He kept throwing a tant rum . His mom was
sad . She threw a tant rum and cried loudly .
and started to throw a tant rum . Suddenly , she
love and honesty , no matter what . <|endoftext|> Jack was
smart . It does not matter if you have long hair
kept crying and throwing a tant rum . Finally
Billy was sad . No matter how hard he tried ,
. She knew that no matter how bitter life got ,
ami figure . But no matter how hard she tried ,
remember , it doesn 't matter if your painting is not
She was reminded that no matter what , she had a
always love you , no matter what ." <|endoftext|> L ily
and felt content . No matter what had happened before he
again and again . No matter how many times they came
forget to dream , no matter how far away they seemed

MIN ACTIVATIONS
MIN = -0.262

down the tree and scam per away . The
and fierce winds started to blow . Tim my was scared
sleeping , okay ? Whis per means to talk very softly
excited to feel the wind blow through their hair and spread
, the wind started to blow and it made Lily sh
, the wind started to blow and the trees began to
his trunk and started to blow . A bunch of delicious
the other balls . Cas per started to display them in
. Har per 's mom put the bowls
The fan started to blow gently and soon Molly felt
the rare rac coon scam per away . Luckily
kids would join together to blow bubbles and have a good
Soon the wind started to blow and the waves started to
fast it made the wind blow . The car
, the wind started to blow harder and harder . Suddenly
, the wind started to blow , and it started to
. The wind started to blow hard and it began to
, " You have to blow on the sail and steer
The wind began to blow harder and the child laughed
sounds . He likes to blow the trumpet in the morning

INTERVAL 0.167 - 0.310
CONTAINS 0.943%

one could open . No matter how hard they tried they
tree . " Wow , this is a nice
him out . <|endoftext|> Once upon a time there was a
friends forever . <|endoftext|> Once upon a time , there was
cake . " Oh no !" Lily and Ben

INTERVAL 0.024 - 0.167
CONTAINS 32.854%

." <|endoftext|> Once upon a time , there was a small
old . One day , Sarah went to the
, Leo ." But Leo was angry . He
who liked to play with fire work . They found some
other . They love their globe and their planet . They

INTERVAL -0.119 - 0.024
CONTAINS 61.421%

They must be happy ." They walk closer to
The girl then gave him a hug and said
a field of grass . The bright green grass was very
anyway . After a little while , Mam ma
good and make someone happy . The end . <|endoftext|> Once

INTERVAL -0.262 - -0.119
CONTAINS 4.783%

and says hello . Anna says , " Hi , dad
Mom my , can I have this toy please ?" Mia
and not bother them when they are working . They learned
. They g iggle and have fun . The
go on your ship ?" Tom agreed and they went on
Token She
Token ID1375
Feature activation+8.16e-02

Pos logit contributions
 unlike+0.054
 matter+0.050
 superpower+0.049
 perspective+0.048
eway+0.048
Neg logit contributions
 Ben-0.041
Ben-0.037
 bat-0.036
 says-0.035
 next-0.035
Token cried
Token ID16896
Feature activation-3.64e-02

Pos logit contributions
 unlike+0.157
 matter+0.149
 superpower+0.142
 blonde+0.139
 tactic+0.134
Neg logit contributions
ila-0.099
line-0.095
Ben-0.093
 Ben-0.092
 bat-0.091
Token and
Token ID290
Feature activation-7.73e-02

Pos logit contributions
 Ben+0.013
 bat+0.012
Ben+0.009
 se+0.009
 next+0.008
Neg logit contributions
 unlike-0.108
 superpower-0.101
 matter-0.101
 perspective-0.100
eway-0.099
Token threw
Token ID9617
Feature activation-1.38e-01

Pos logit contributions
ila+0.062
 Ben+0.062
Ben+0.061
 bat+0.061
 se+0.060
Neg logit contributions
 unlike-0.187
 matter-0.172
 superpower-0.171
 blonde-0.168
 perspective-0.166
Token a
Token ID257
Feature activation+1.99e-03

Pos logit contributions
 Ben+0.130
 bat+0.114
 each+0.114
 put+0.107
 se+0.105
Neg logit contributions
 unlike-0.346
 superpower-0.318
 perspective-0.316
eway-0.314
 former-0.309
Token tant
Token ID24246
Feature activation+3.10e-01

Pos logit contributions
 unlike+0.003
 former+0.003
eway+0.003
 matter+0.003
 superpower+0.003
Neg logit contributions
 bat-0.004
 se-0.004
Ben-0.003
 slide-0.003
 Ben-0.003
Tokenrum
Token ID6582
Feature activation+2.09e-01

Pos logit contributions
 unlike+0.277
 former+0.236
 blonde+0.226
 matter+0.224
 superpower+0.223
Neg logit contributions
ila-0.731
 se-0.701
Ben-0.689
 Ben-0.689
 bat-0.689
Token.
Token ID13
Feature activation+2.87e-02

Pos logit contributions
 unlike+0.565
 matter+0.546
 superpower+0.534
 tactic+0.532
 perspective+0.519
Neg logit contributions
 Ben-0.128
 next-0.112
 each-0.107
 bat-0.105
ila-0.101
Token Her
Token ID2332
Feature activation+1.39e-01

Pos logit contributions
 unlike+0.053
 matter+0.052
 perspective+0.049
eway+0.048
 superpower+0.048
Neg logit contributions
 Ben-0.046
Ben-0.040
 bat-0.039
 next-0.039
 says-0.038
Token mom
Token ID1995
Feature activation-3.33e-02

Pos logit contributions
 unlike+0.385
eway+0.356
 perspective+0.352
 tactic+0.351
 matter+0.350
Neg logit contributions
 bat-0.088
 Ben-0.086
 se-0.083
 next-0.077
 slide-0.071
Token tried
Token ID3088
Feature activation+2.49e-02

Pos logit contributions
ila+0.040
 Ben+0.039
Ben+0.039
 bat+0.038
 se+0.036
Neg logit contributions
 unlike-0.066
 matter-0.061
 superpower-0.058
 blonde-0.057
 tactic-0.057
Token listen
Token ID6004
Feature activation-1.19e-01

Pos logit contributions
ila+0.040
 Ben+0.040
 se+0.040
Ben+0.039
 bat+0.038
Neg logit contributions
 matter-0.075
 unlike-0.075
 superpower-0.069
 perspective-0.068
 tactic-0.067
Token.
Token ID13
Feature activation+1.75e-02

Pos logit contributions
 Ben+0.108
 bat+0.104
Ben+0.103
ila+0.103
 se+0.101
Neg logit contributions
 unlike-0.281
 superpower-0.257
 matter-0.257
 perspective-0.252
 blonde-0.251
Token She
Token ID1375
Feature activation+7.20e-02

Pos logit contributions
 unlike+0.031
 matter+0.028
 superpower+0.028
 perspective+0.027
 tactic+0.027
Neg logit contributions
 Ben-0.026
Ben-0.025
 bat-0.025
ila-0.025
 se-0.024
Token threw
Token ID9617
Feature activation-1.43e-01

Pos logit contributions
 unlike+0.136
 matter+0.127
 superpower+0.122
 blonde+0.121
 former+0.116
Neg logit contributions
ila-0.089
 Ben-0.087
line-0.086
 bat-0.085
Ben-0.085
Token a
Token ID257
Feature activation-4.38e-03

Pos logit contributions
 Ben+0.110
 bat+0.104
Ben+0.101
ila+0.101
 se+0.100
Neg logit contributions
 unlike-0.359
 superpower-0.330
 matter-0.327
 perspective-0.323
 blonde-0.321
Token tant
Token ID24246
Feature activation+3.10e-01

Pos logit contributions
 bat+0.009
 se+0.009
 Ben+0.009
Ben+0.009
ila+0.008
Neg logit contributions
 unlike-0.006
 matter-0.005
 superpower-0.005
 blonde-0.005
 perspective-0.005
Tokenrum
Token ID6582
Feature activation+2.08e-01

Pos logit contributions
 unlike+0.282
 former+0.236
 blonde+0.228
 matter+0.227
 superpower+0.222
Neg logit contributions
ila-0.738
 se-0.700
Ben-0.696
 Ben-0.691
 bat-0.687
Token in
Token ID287
Feature activation+3.15e-02

Pos logit contributions
 unlike+0.545
 superpower+0.500
 matter+0.499
 perspective+0.491
 blonde+0.491
Neg logit contributions
 bat-0.125
 Ben-0.124
Ben-0.123
ila-0.122
 se-0.120
Token the
Token ID262
Feature activation-6.25e-02

Pos logit contributions
 unlike+0.079
 matter+0.074
 superpower+0.073
 blonde+0.072
 perspective+0.072
Neg logit contributions
 Ben-0.022
Ben-0.022
 bat-0.022
ila-0.022
 se-0.021
Token store
Token ID3650
Feature activation-3.85e-02

Pos logit contributions
ila+0.059
 Ben+0.055
Ben+0.055
 se+0.051
 bat+0.048
Neg logit contributions
 unlike-0.137
 matter-0.132
 superpower-0.131
 tactic-0.128
 perspective-0.125
Token.
Token ID13
Feature activation+1.82e-02

Pos logit contributions
Ben+0.033
ila+0.032
 bat+0.031
 se+0.030
 Ben+0.029
Neg logit contributions
 unlike-0.092
 matter-0.083
 perspective-0.082
 blonde-0.082
 superpower-0.082
Token and
Token ID290
Feature activation-7.79e-02

Pos logit contributions
 unlike+0.113
 matter+0.105
 superpower+0.105
 perspective+0.103
 blonde+0.102
Neg logit contributions
 Ben-0.020
 bat-0.019
ila-0.018
Ben-0.018
 se-0.018
Token so
Token ID523
Feature activation+1.09e-01

Pos logit contributions
 Ben+0.115
 bat+0.113
ila+0.112
Ben+0.111
 se+0.110
Neg logit contributions
 unlike-0.140
 superpower-0.125
 perspective-0.123
 matter-0.122
 tactic-0.121
Token she
Token ID673
Feature activation+1.28e-02

Pos logit contributions
 unlike+0.248
 superpower+0.226
 matter+0.226
 perspective+0.221
 tactic+0.220
Neg logit contributions
 Ben-0.108
 bat-0.105
ila-0.104
Ben-0.104
 se-0.102
Token threw
Token ID9617
Feature activation-1.38e-01

Pos logit contributions
 unlike+0.027
 matter+0.025
 superpower+0.024
 blonde+0.024
 tactic+0.023
Neg logit contributions
 Ben-0.014
ila-0.014
Ben-0.014
 bat-0.014
 se-0.014
Token a
Token ID257
Feature activation+6.10e-03

Pos logit contributions
ila+0.157
Ben+0.156
 bat+0.156
 Ben+0.155
 se+0.153
Neg logit contributions
 unlike-0.286
 matter-0.262
 superpower-0.256
 blonde-0.254
 tactic-0.250
Token tant
Token ID24246
Feature activation+3.10e-01

Pos logit contributions
 unlike+0.014
 matter+0.012
 superpower+0.012
 blonde+0.012
eway+0.012
Neg logit contributions
 bat-0.007
 se-0.007
Ben-0.006
 Ben-0.006
ila-0.006
Tokenrum
Token ID6582
Feature activation+2.08e-01

Pos logit contributions
 unlike+0.271
 former+0.236
 matter+0.230
 blonde+0.227
 superpower+0.216
Neg logit contributions
ila-0.728
 se-0.697
 bat-0.682
 Ben-0.679
Ben-0.673
Token when
Token ID618
Feature activation+1.19e-01

Pos logit contributions
 unlike+0.537
 matter+0.519
 superpower+0.507
 tactic+0.499
 blonde+0.489
Neg logit contributions
 Ben-0.144
 bat-0.128
 next-0.127
ila-0.126
 each-0.120
Token her
Token ID607
Feature activation+7.26e-02

Pos logit contributions
 unlike+0.274
 superpower+0.250
 matter+0.250
 tactic+0.245
 perspective+0.244
Neg logit contributions
 Ben-0.115
 bat-0.111
ila-0.110
Ben-0.110
 se-0.108
Token parents
Token ID3397
Feature activation-1.34e-02

Pos logit contributions
 unlike+0.179
 matter+0.164
 superpower+0.161
 perspective+0.161
 tactic+0.160
Neg logit contributions
 Ben-0.060
 bat-0.059
 se-0.058
Ben-0.056
ila-0.056
Token said
Token ID531
Feature activation-5.44e-02

Pos logit contributions
ila+0.015
Ben+0.015
 Ben+0.014
 bat+0.014
 se+0.014
Neg logit contributions
 unlike-0.026
 matter-0.025
 superpower-0.023
 blonde-0.023
 darkened-0.023
Token her
Token ID607
Feature activation+6.28e-02

Pos logit contributions
 unlike+0.038
 superpower+0.036
 perspective+0.035
 matter+0.035
eway+0.034
Neg logit contributions
 Ben-0.013
 each-0.011
 bat-0.011
 put-0.010
 and-0.010
Token feet
Token ID3625
Feature activation-3.40e-02

Pos logit contributions
 unlike+0.140
 matter+0.126
eway+0.125
 superpower+0.124
 tactic+0.122
Neg logit contributions
 Ben-0.071
 bat-0.068
 se-0.064
Ben-0.063
 next-0.063
Token and
Token ID290
Feature activation-8.06e-02

Pos logit contributions
 Ben+0.017
ila+0.016
Ben+0.016
 bat+0.016
 se+0.015
Neg logit contributions
 unlike-0.093
 matter-0.086
 superpower-0.085
 blonde-0.084
 perspective-0.084
Token threw
Token ID9617
Feature activation-1.43e-01

Pos logit contributions
 Ben+0.075
 bat+0.070
 se+0.069
Ben+0.068
ila+0.067
Neg logit contributions
 unlike-0.192
 superpower-0.178
 perspective-0.176
 matter-0.175
eway-0.172
Token a
Token ID257
Feature activation-1.77e-03

Pos logit contributions
 Ben+0.125
 bat+0.112
ila+0.109
 se+0.108
Ben+0.107
Neg logit contributions
 unlike-0.354
 superpower-0.329
 perspective-0.322
 matter-0.319
 tactic-0.317
Token tant
Token ID24246
Feature activation+3.08e-01

Pos logit contributions
 bat+0.004
 se+0.004
Ben+0.004
 Ben+0.004
ila+0.004
Neg logit contributions
 unlike-0.002
 superpower-0.001
 former-0.001
 matter-0.001
eway-0.001
Tokenrum
Token ID6582
Feature activation+2.04e-01

Pos logit contributions
 unlike+0.260
 former+0.215
 matter+0.211
 blonde+0.208
 superpower+0.205
Neg logit contributions
ila-0.743
 se-0.705
Ben-0.697
 Ben-0.695
 bat-0.694
Token.
Token ID13
Feature activation+2.07e-02

Pos logit contributions
 unlike+0.522
 matter+0.495
 superpower+0.490
 tactic+0.481
 perspective+0.474
Neg logit contributions
 Ben-0.147
ila-0.133
 bat-0.132
Ben-0.129
 next-0.125
Token She
Token ID1375
Feature activation+7.36e-02

Pos logit contributions
 unlike+0.038
 matter+0.037
 superpower+0.035
 tactic+0.035
 perspective+0.034
Neg logit contributions
 Ben-0.033
Ben-0.029
ila-0.029
 bat-0.028
 se-0.027
Token grabbed
Token ID13176
Feature activation-8.05e-02

Pos logit contributions
 unlike+0.166
 superpower+0.152
 matter+0.151
 perspective+0.149
 tactic+0.147
Neg logit contributions
 Ben-0.074
 bat-0.073
Ben-0.072
 se-0.072
ila-0.071
Token the
Token ID262
Feature activation-6.11e-02

Pos logit contributions
 Ben+0.066
 bat+0.065
Ben+0.064
ila+0.064
 se+0.063
Neg logit contributions
 unlike-0.195
 matter-0.179
 superpower-0.178
 blonde-0.174
 perspective-0.174
Token to
Token ID284
Feature activation+7.30e-02

Pos logit contributions
 unlike+0.148
 matter+0.134
 superpower+0.134
 perspective+0.131
 blonde+0.130
Neg logit contributions
 Ben-0.069
 bat-0.068
 se-0.066
Ben-0.065
ila-0.065
Token cry
Token ID3960
Feature activation-3.02e-03

Pos logit contributions
 unlike+0.146
 superpower+0.131
 matter+0.130
 perspective+0.127
 blonde+0.127
Neg logit contributions
 Ben-0.092
 bat-0.091
Ben-0.090
ila-0.089
 se-0.089
Token and
Token ID290
Feature activation-8.00e-02

Pos logit contributions
 Ben+0.002
 bat+0.002
Ben+0.002
 se+0.002
ila+0.002
Neg logit contributions
 unlike-0.008
 superpower-0.008
 matter-0.008
 blonde-0.007
 perspective-0.007
Token threw
Token ID9617
Feature activation-1.41e-01

Pos logit contributions
 Ben+0.084
 bat+0.078
 se+0.075
 put+0.075
Ben+0.071
Neg logit contributions
 unlike-0.190
 superpower-0.173
 perspective-0.171
eway-0.170
 blonde-0.166
Token a
Token ID257
Feature activation+4.52e-04

Pos logit contributions
 Ben+0.127
 bat+0.114
Ben+0.106
ila+0.106
 se+0.105
Neg logit contributions
 unlike-0.356
 superpower-0.326
 perspective-0.322
eway-0.319
 former-0.318
Token tant
Token ID24246
Feature activation+3.07e-01

Pos logit contributions
 unlike+0.001
 former+0.001
eway+0.001
 meanings+0.001
 avail+0.001
Neg logit contributions
 bat-0.001
 se-0.001
 slide-0.001
Ben-0.001
 dove-0.001
Tokenrum
Token ID6582
Feature activation+2.06e-01

Pos logit contributions
 unlike+0.273
 former+0.232
 blonde+0.224
 matter+0.221
 superpower+0.217
Neg logit contributions
ila-0.725
 se-0.690
 Ben-0.683
Ben-0.682
 bat-0.682
Token.
Token ID13
Feature activation+2.41e-02

Pos logit contributions
 unlike+0.535
 matter+0.505
 superpower+0.501
 tactic+0.491
 perspective+0.486
Neg logit contributions
 Ben-0.142
 bat-0.128
ila-0.126
 next-0.123
Ben-0.122
Token She
Token ID1375
Feature activation+7.96e-02

Pos logit contributions
 unlike+0.046
 matter+0.045
eway+0.044
 perspective+0.044
 tactic+0.043
Neg logit contributions
 Ben-0.037
 bat-0.030
 Tom-0.030
 next-0.030
Ben-0.029
Token didn
Token ID1422
Feature activation+1.02e-01

Pos logit contributions
 unlike+0.163
 superpower+0.146
 matter+0.145
 perspective+0.143
 blonde+0.142
Neg logit contributions
 Ben-0.098
 bat-0.097
Ben-0.095
 se-0.095
ila-0.094
Token't
Token ID470
Feature activation+1.12e-02

Pos logit contributions
 unlike+0.116
 superpower+0.088
 matter+0.086
eway+0.085
 Tears+0.084
Neg logit contributions
 dads-0.212
 se-0.208
 pige-0.207
Ben-0.207
ila-0.206
Token.
Token ID13
Feature activation+1.38e-02

Pos logit contributions
 Ben+0.088
 bat+0.084
Ben+0.080
 se+0.079
ila+0.079
Neg logit contributions
 unlike-0.319
 superpower-0.294
 matter-0.293
 blonde-0.288
 perspective-0.287
Token He
Token ID679
Feature activation+1.10e-01

Pos logit contributions
 unlike+0.024
 matter+0.022
 perspective+0.021
 superpower+0.021
 tactic+0.020
Neg logit contributions
 Ben-0.024
Ben-0.022
ila-0.022
 bat-0.021
 says-0.021
Token kept
Token ID4030
Feature activation-2.73e-02

Pos logit contributions
 unlike+0.224
 matter+0.203
 superpower+0.201
 blonde+0.196
 tactic+0.192
Neg logit contributions
 Ben-0.131
 bat-0.130
ila-0.129
Ben-0.128
 se-0.125
Token throwing
Token ID9644
Feature activation-1.12e-01

Pos logit contributions
Ben+0.041
ila+0.040
 bat+0.038
 se+0.038
 Ben+0.037
Neg logit contributions
 unlike-0.047
 matter-0.042
 superpower-0.042
 blonde-0.041
 Tears-0.040
Token a
Token ID257
Feature activation-6.53e-03

Pos logit contributions
 Ben+0.108
 bat+0.107
ila+0.106
Ben+0.106
 se+0.104
Neg logit contributions
 unlike-0.254
 matter-0.232
 superpower-0.231
 blonde-0.226
 perspective-0.226
Token tant
Token ID24246
Feature activation+3.07e-01

Pos logit contributions
 bat+0.008
 se+0.008
Ben+0.008
 Ben+0.008
 slide+0.007
Neg logit contributions
 unlike-0.014
 matter-0.012
 superpower-0.012
 blonde-0.012
eway-0.012
Tokenrum
Token ID6582
Feature activation+2.04e-01

Pos logit contributions
 unlike+0.228
 blonde+0.170
 matter+0.169
 superpower+0.169
 former+0.160
Neg logit contributions
ila-0.768
Ben-0.747
 se-0.747
 Ben-0.747
 bat-0.743
Token.
Token ID13
Feature activation+1.50e-02

Pos logit contributions
 unlike+0.498
 matter+0.455
 superpower+0.455
 perspective+0.446
 blonde+0.445
Neg logit contributions
 Ben-0.161
 bat-0.159
Ben-0.157
ila-0.157
 se-0.154
Token His
Token ID2399
Feature activation+1.69e-01

Pos logit contributions
 unlike+0.027
 matter+0.025
 perspective+0.024
 superpower+0.024
 tactic+0.024
Neg logit contributions
 Ben-0.026
Ben-0.023
ila-0.023
 bat-0.022
 Tom-0.022
Token mom
Token ID1995
Feature activation-4.13e-02

Pos logit contributions
 unlike+0.457
 matter+0.416
eway+0.411
 tactic+0.410
 perspective+0.410
Neg logit contributions
 bat-0.116
 Ben-0.116
 se-0.103
Ben-0.098
 next-0.098
Token was
Token ID373
Feature activation+8.49e-02

Pos logit contributions
ila+0.050
 bat+0.050
 Ben+0.050
Ben+0.049
 dads+0.047
Neg logit contributions
 unlike-0.078
 matter-0.072
 superpower-0.068
 blonde-0.068
 darkened-0.066
Token sad
Token ID6507
Feature activation+1.28e-01

Pos logit contributions
 Ben+0.113
Ben+0.112
 bat+0.112
ila+0.112
 se+0.110
Neg logit contributions
 unlike-0.155
 matter-0.139
 superpower-0.137
 blonde-0.134
 perspective-0.133
Token.
Token ID13
Feature activation+1.94e-02

Pos logit contributions
 unlike+0.367
 superpower+0.345
eway+0.344
 blonde+0.344
 matter+0.344
Neg logit contributions
 Ben-0.070
 and-0.058
 next-0.058
 each-0.055
 they-0.054
Token She
Token ID1375
Feature activation+7.39e-02

Pos logit contributions
 unlike+0.041
 superpower+0.038
 matter+0.038
 perspective+0.037
eway+0.036
Neg logit contributions
 Ben-0.024
Ben-0.022
ila-0.021
 bat-0.021
 says-0.021
Token threw
Token ID9617
Feature activation-1.45e-01

Pos logit contributions
 unlike+0.164
eway+0.153
 perspective+0.153
 superpower+0.151
 tactic+0.147
Neg logit contributions
 Ben-0.083
 put-0.082
 bat-0.078
Ben-0.078
 se-0.078
Token a
Token ID257
Feature activation-2.61e-03

Pos logit contributions
 Ben+0.086
 bat+0.078
ila+0.076
Ben+0.075
 se+0.073
Neg logit contributions
 unlike-0.394
 superpower-0.365
 matter-0.360
 perspective-0.359
 blonde-0.355
Token tant
Token ID24246
Feature activation+3.04e-01

Pos logit contributions
 bat+0.005
 se+0.005
Ben+0.005
 slide+0.004
 Ben+0.004
Neg logit contributions
 unlike-0.004
 former-0.004
eway-0.003
 superpower-0.003
 perspective-0.003
Tokenrum
Token ID6582
Feature activation+2.02e-01

Pos logit contributions
 unlike+0.268
 former+0.237
 matter+0.224
 blonde+0.219
 superpower+0.215
Neg logit contributions
ila-0.722
 se-0.677
Ben-0.670
 Ben-0.667
 bat-0.662
Token and
Token ID290
Feature activation-8.23e-02

Pos logit contributions
 unlike+0.572
 matter+0.546
 superpower+0.540
 tactic+0.532
 blonde+0.524
Neg logit contributions
 Ben-0.090
 bat-0.076
ila-0.075
Ben-0.073
 next-0.071
Token cried
Token ID16896
Feature activation-4.43e-02

Pos logit contributions
 Ben+0.098
 bat+0.097
Ben+0.096
ila+0.096
 se+0.095
Neg logit contributions
 unlike-0.169
 superpower-0.153
 matter-0.153
 perspective-0.149
 blonde-0.148
Token loudly
Token ID23112
Feature activation+7.31e-02

Pos logit contributions
 Ben+0.034
 bat+0.031
 each+0.030
 next+0.028
 and+0.028
Neg logit contributions
 unlike-0.114
 matter-0.107
 superpower-0.107
 tactic-0.106
eway-0.106
Token.
Token ID13
Feature activation+2.03e-02

Pos logit contributions
 unlike+0.191
 matter+0.182
 perspective+0.179
 tactic+0.179
 superpower+0.177
Neg logit contributions
 Ben-0.053
 next-0.044
 bat-0.043
 se-0.039
 each-0.039
Token and
Token ID290
Feature activation-8.39e-02

Pos logit contributions
 unlike+0.130
 matter+0.118
 superpower+0.117
 tactic+0.115
 perspective+0.115
Neg logit contributions
Ben-0.031
ila-0.031
 se-0.029
 bat-0.029
 Ben-0.026
Token started
Token ID2067
Feature activation+6.20e-02

Pos logit contributions
Ben+0.103
ila+0.102
 bat+0.100
 Ben+0.099
 se+0.097
Neg logit contributions
 unlike-0.167
 matter-0.153
 superpower-0.148
 blonde-0.146
 tactic-0.144
Token to
Token ID284
Feature activation+6.86e-02

Pos logit contributions
 unlike+0.143
 matter+0.132
 superpower+0.131
 perspective+0.129
 blonde+0.127
Neg logit contributions
 Ben-0.061
 bat-0.061
 se-0.058
Ben-0.056
ila-0.055
Token throw
Token ID3714
Feature activation-2.05e-01

Pos logit contributions
 unlike+0.139
 superpower+0.122
 perspective+0.119
eway+0.119
 matter+0.118
Neg logit contributions
 bat-0.092
 Ben-0.092
 se-0.087
Ben-0.086
 put-0.085
Token a
Token ID257
Feature activation-3.82e-03

Pos logit contributions
 Ben+0.173
 bat+0.161
Ben+0.156
ila+0.153
 se+0.150
Neg logit contributions
 unlike-0.513
 superpower-0.471
 matter-0.465
 perspective-0.464
eway-0.457
Token tant
Token ID24246
Feature activation+3.04e-01

Pos logit contributions
 bat+0.005
 se+0.005
 Ben+0.005
Ben+0.005
ila+0.005
Neg logit contributions
 unlike-0.008
 former-0.007
 superpower-0.007
 matter-0.007
 blonde-0.007
Tokenrum
Token ID6582
Feature activation+2.02e-01

Pos logit contributions
 unlike+0.253
 blonde+0.198
 superpower+0.195
 matter+0.195
 former+0.191
Neg logit contributions
ila-0.736
Ben-0.711
 Ben-0.710
 se-0.710
 bat-0.707
Token.
Token ID13
Feature activation+2.02e-02

Pos logit contributions
 unlike+0.511
 matter+0.476
 superpower+0.474
 tactic+0.462
 perspective+0.461
Neg logit contributions
 Ben-0.151
 bat-0.142
ila-0.140
Ben-0.138
 se-0.135
Token Suddenly
Token ID24975
Feature activation+1.90e-01

Pos logit contributions
 unlike+0.039
 matter+0.037
eway+0.036
 superpower+0.035
 perspective+0.035
Neg logit contributions
 Ben-0.032
Ben-0.027
 bat-0.027
 next-0.026
 says-0.026
Token,
Token ID11
Feature activation+1.02e-02

Pos logit contributions
 unlike+0.471
 matter+0.432
 superpower+0.431
 tactic+0.426
 perspective+0.425
Neg logit contributions
 Ben-0.159
 bat-0.147
Ben-0.146
ila-0.145
 se-0.141
Token she
Token ID673
Feature activation+6.13e-03

Pos logit contributions
 unlike+0.027
 perspective+0.026
 tactic+0.026
 superpower+0.026
eway+0.026
Neg logit contributions
 Ben-0.009
 they-0.007
 bat-0.007
ila-0.006
Ben-0.006
Token love
Token ID1842
Feature activation-1.49e-02

Pos logit contributions
 Ben+0.006
 each+0.005
 bat+0.005
 se+0.005
ila+0.005
Neg logit contributions
 unlike-0.013
 superpower-0.012
eway-0.012
 darkened-0.012
 tactic-0.012
Token and
Token ID290
Feature activation-7.41e-02

Pos logit contributions
 Ben+0.010
 bat+0.010
ila+0.010
Ben+0.010
 se+0.009
Neg logit contributions
 unlike-0.038
 superpower-0.035
 matter-0.035
 tactic-0.035
 blonde-0.035
Token honesty
Token ID24345
Feature activation+1.73e-02

Pos logit contributions
 Ben+0.078
 bat+0.078
 se+0.076
Ben+0.075
ila+0.074
Neg logit contributions
 unlike-0.162
 superpower-0.147
 matter-0.146
 perspective-0.143
 blonde-0.143
Token,
Token ID11
Feature activation+1.88e-02

Pos logit contributions
 unlike+0.043
 superpower+0.039
 matter+0.039
 perspective+0.039
 blonde+0.039
Neg logit contributions
 Ben-0.013
 bat-0.013
Ben-0.012
ila-0.012
 se-0.012
Token no
Token ID645
Feature activation+3.29e-02

Pos logit contributions
 unlike+0.039
 matter+0.035
 superpower+0.035
 blonde+0.034
 perspective+0.034
Neg logit contributions
Ben-0.022
ila-0.022
 Ben-0.022
 bat-0.021
 se-0.021
Token matter
Token ID2300
Feature activation+3.03e-01

Pos logit contributions
 unlike+0.028
eway+0.024
 Terr+0.021
 superpower+0.018
inct+0.018
Neg logit contributions
 Ben-0.083
 each-0.081
 put-0.081
 se-0.081
 bat-0.081
Token what
Token ID644
Feature activation+8.20e-02

Pos logit contributions
 unlike+0.638
 superpower+0.576
 blonde+0.567
 Terr+0.557
 darkened+0.556
Neg logit contributions
 Ben-0.369
ila-0.342
 each-0.337
 bat-0.330
 we-0.326
Token.
Token ID13
Feature activation+2.83e-02

Pos logit contributions
 unlike+0.194
eway+0.179
 tactic+0.178
 superpower+0.176
 perspective+0.176
Neg logit contributions
 Ben-0.076
 bat-0.071
 se-0.068
ila-0.067
Ben-0.067
Token<|endoftext|>
Token ID50256
Feature activation+9.09e-03

Pos logit contributions
 unlike+0.064
 superpower+0.058
 matter+0.057
 perspective+0.056
 blonde+0.056
Neg logit contributions
 se-0.027
 bat-0.027
ila-0.027
Ben-0.026
 Ben-0.026
TokenJack
Token ID14295
Feature activation+2.23e-02

Pos logit contributions
 unlike+0.020
 superpower+0.018
 matter+0.018
 blonde+0.018
 perspective+0.018
Neg logit contributions
 bat-0.009
 se-0.009
ila-0.009
Ben-0.009
 Ben-0.009
Token was
Token ID373
Feature activation+9.31e-02

Pos logit contributions
 unlike+0.047
 superpower+0.043
 matter+0.042
 blonde+0.042
 perspective+0.040
Neg logit contributions
ila-0.023
Ben-0.022
 se-0.021
 dads-0.021
 bat-0.021
Token smart
Token ID4451
Feature activation-1.23e-02

Pos logit contributions
Ben+0.120
ila+0.119
 bat+0.115
 se+0.114
 Ben+0.110
Neg logit contributions
 unlike-0.147
 matter-0.140
 superpower-0.130
 blonde-0.129
 meanings-0.121
Token.
Token ID13
Feature activation+2.57e-02

Pos logit contributions
ila+0.011
Ben+0.011
 bat+0.011
 se+0.011
 Ben+0.011
Neg logit contributions
 unlike-0.029
 matter-0.026
 superpower-0.026
 perspective-0.026
 blonde-0.025
Token It
Token ID632
Feature activation+6.45e-02

Pos logit contributions
 unlike+0.042
 superpower+0.036
 matter+0.036
 blonde+0.036
 Tears+0.034
Neg logit contributions
 se-0.042
 bat-0.040
ila-0.039
Ben-0.039
 dove-0.038
Token does
Token ID857
Feature activation+5.19e-02

Pos logit contributions
 unlike+0.144
 matter+0.132
 superpower+0.129
 tactic+0.124
 Terr+0.122
Neg logit contributions
Ben-0.065
ila-0.063
 Ben-0.062
 bat-0.062
line-0.059
Token not
Token ID407
Feature activation-3.39e-02

Pos logit contributions
 unlike+0.106
 matter+0.099
 superpower+0.096
 blonde+0.092
 tactic+0.092
Neg logit contributions
 Ben-0.061
Ben-0.061
 se-0.060
ila-0.060
 bat-0.059
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
ila+0.039
 Ben+0.038
Ben+0.038
 se+0.037
 bat+0.037
Neg logit contributions
 unlike-0.069
 matter-0.068
 superpower-0.063
 blonde-0.061
 perspective-0.060
Token if
Token ID611
Feature activation+8.38e-02

Pos logit contributions
 unlike+0.565
 matter+0.522
 superpower+0.510
 perspective+0.502
 tactic+0.492
Neg logit contributions
 bat-0.405
Ben-0.403
 se-0.402
 Ben-0.395
ila-0.394
Token you
Token ID345
Feature activation-1.19e-01

Pos logit contributions
 unlike+0.207
 matter+0.200
 superpower+0.191
 blonde+0.187
 perspective+0.187
Neg logit contributions
 se-0.056
Ben-0.055
 bat-0.054
ila-0.052
line-0.045
Token have
Token ID423
Feature activation-1.34e-01

Pos logit contributions
ila+0.129
Ben+0.118
 se+0.115
 Ben+0.112
 pige+0.112
Neg logit contributions
 matter-0.237
 unlike-0.224
 superpower-0.219
 blonde-0.214
 tactic-0.210
Token long
Token ID890
Feature activation-7.10e-03

Pos logit contributions
 Ben+0.153
 bat+0.149
Ben+0.146
 se+0.145
ila+0.145
Neg logit contributions
 unlike-0.289
 superpower-0.259
 matter-0.254
eway-0.253
 perspective-0.252
Token hair
Token ID4190
Feature activation+7.59e-02

Pos logit contributions
Ben+0.007
 Ben+0.007
 bat+0.007
ila+0.007
 se+0.007
Neg logit contributions
 unlike-0.015
 matter-0.014
 superpower-0.014
 blonde-0.014
 perspective-0.014
Token kept
Token ID4030
Feature activation-3.01e-02

Pos logit contributions
 unlike+0.146
 matter+0.133
 superpower+0.131
 blonde+0.128
 perspective+0.127
Neg logit contributions
 Ben-0.088
ila-0.087
 bat-0.087
Ben-0.086
 se-0.085
Token crying
Token ID13774
Feature activation+5.60e-02

Pos logit contributions
ila+0.044
 Ben+0.044
Ben+0.044
 bat+0.044
 se+0.043
Neg logit contributions
 unlike-0.053
 matter-0.047
 superpower-0.047
 blonde-0.046
 perspective-0.045
Token and
Token ID290
Feature activation-8.55e-02

Pos logit contributions
 unlike+0.159
 superpower+0.149
 matter+0.148
eway+0.146
 perspective+0.146
Neg logit contributions
 Ben-0.025
 bat-0.022
Ben-0.021
ila-0.021
 se-0.020
Token throwing
Token ID9644
Feature activation-1.15e-01

Pos logit contributions
 Ben+0.113
 bat+0.108
 se+0.105
Ben+0.105
ila+0.104
Neg logit contributions
 unlike-0.172
 superpower-0.156
 matter-0.152
 perspective-0.151
eway-0.150
Token a
Token ID257
Feature activation-4.28e-03

Pos logit contributions
 Ben+0.099
 bat+0.091
Ben+0.085
 each+0.084
 se+0.081
Neg logit contributions
 unlike-0.291
 superpower-0.268
 perspective-0.267
eway-0.265
 matter-0.260
Token tant
Token ID24246
Feature activation+3.02e-01

Pos logit contributions
 bat+0.008
 se+0.007
Ben+0.007
 Ben+0.007
 slide+0.007
Neg logit contributions
 unlike-0.007
eway-0.006
 former-0.006
 superpower-0.006
 matter-0.006
Tokenrum
Token ID6582
Feature activation+2.00e-01

Pos logit contributions
 unlike+0.266
 former+0.221
 blonde+0.213
 matter+0.211
 superpower+0.210
Neg logit contributions
ila-0.722
Ben-0.684
 se-0.682
 Ben-0.679
 bat-0.677
Token.
Token ID13
Feature activation+1.79e-02

Pos logit contributions
 unlike+0.529
 matter+0.509
 superpower+0.502
 tactic+0.496
 perspective+0.485
Neg logit contributions
 Ben-0.132
 next-0.116
 each-0.110
ila-0.110
 bat-0.108
Token
Token ID198
Feature activation-2.40e-02

Pos logit contributions
 unlike+0.044
 matter+0.043
 superpower+0.042
 perspective+0.041
eway+0.041
Neg logit contributions
 Ben-0.019
Ben-0.015
 next-0.014
 Tom-0.014
 bat-0.013
Token
Token ID198
Feature activation-2.34e-02

Pos logit contributions
 Ben+0.017
Ben+0.017
ila+0.017
 bat+0.016
 se+0.015
Neg logit contributions
 unlike-0.061
 matter-0.057
 superpower-0.057
 blonde-0.055
 perspective-0.055
TokenFinally
Token ID11158
Feature activation+1.86e-01

Pos logit contributions
Ben+0.035
 Ben+0.034
 bat+0.034
ila+0.034
 se+0.032
Neg logit contributions
 unlike-0.042
 superpower-0.038
 matter-0.037
 perspective-0.036
 tactic-0.036
TokenBilly
Token ID49640
Feature activation+5.67e-02

Pos logit contributions
 unlike+0.017
 superpower+0.015
 matter+0.015
 perspective+0.015
 blonde+0.015
Neg logit contributions
 Ben-0.008
 bat-0.008
ila-0.008
Ben-0.008
 se-0.008
Token was
Token ID373
Feature activation+8.96e-02

Pos logit contributions
 unlike+0.116
 blonde+0.105
 superpower+0.105
 matter+0.104
 perspective+0.100
Neg logit contributions
ila-0.065
Ben-0.061
 se-0.061
 bat-0.061
 Ben-0.059
Token sad
Token ID6507
Feature activation+1.36e-01

Pos logit contributions
 unlike+0.190
 matter+0.174
 superpower+0.173
 perspective+0.170
 tactic+0.169
Neg logit contributions
 Ben-0.101
 bat-0.099
ila-0.096
Ben-0.096
 se-0.096
Token.
Token ID13
Feature activation+2.52e-02

Pos logit contributions
 unlike+0.355
 matter+0.333
 superpower+0.331
 tactic+0.330
eway+0.327
Neg logit contributions
 bat-0.088
 Ben-0.087
 se-0.077
 each-0.075
 next-0.074
Token No
Token ID1400
Feature activation+8.54e-02

Pos logit contributions
 unlike+0.044
 superpower+0.043
 perspective+0.042
 matter+0.042
 former+0.039
Neg logit contributions
 Ben-0.038
 bat-0.036
 Tom-0.034
 next-0.032
 each-0.032
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.169
 superpower+0.151
eway+0.148
 Terr+0.145
 blonde+0.144
Neg logit contributions
 Ben-0.111
 bat-0.110
 se-0.109
 next-0.105
Ben-0.104
Token how
Token ID703
Feature activation+4.58e-02

Pos logit contributions
 unlike+0.652
 superpower+0.603
 blonde+0.600
 Terr+0.593
 former+0.581
Neg logit contributions
 Ben-0.349
 next-0.316
 we-0.315
 each-0.315
ila-0.313
Token hard
Token ID1327
Feature activation+4.53e-02

Pos logit contributions
 unlike+0.090
 matter+0.081
 superpower+0.081
 perspective+0.078
 blonde+0.078
Neg logit contributions
 Ben-0.059
 bat-0.058
ila-0.057
Ben-0.057
 se-0.057
Token he
Token ID339
Feature activation+1.70e-02

Pos logit contributions
 unlike+0.097
 matter+0.088
 superpower+0.087
 blonde+0.085
 perspective+0.084
Neg logit contributions
ila-0.050
Ben-0.049
 Ben-0.049
 bat-0.049
 se-0.048
Token tried
Token ID3088
Feature activation+2.45e-02

Pos logit contributions
 unlike+0.031
 matter+0.028
 superpower+0.028
 blonde+0.027
 tactic+0.026
Neg logit contributions
ila-0.024
 Ben-0.024
Ben-0.024
 bat-0.023
 se-0.023
Token,
Token ID11
Feature activation+1.67e-02

Pos logit contributions
 unlike+0.062
 superpower+0.058
eway+0.058
 tactic+0.058
 matter+0.058
Neg logit contributions
 Ben-0.018
 bat-0.017
 next-0.015
 and-0.015
 each-0.015
Token.
Token ID13
Feature activation+2.54e-02

Pos logit contributions
 Ben+0.041
 bat+0.040
ila+0.040
Ben+0.040
 se+0.039
Neg logit contributions
 unlike-0.071
 matter-0.064
 superpower-0.064
 perspective-0.062
 blonde-0.062
Token She
Token ID1375
Feature activation+8.21e-02

Pos logit contributions
 unlike+0.048
 matter+0.044
 superpower+0.043
 perspective+0.042
 tactic+0.042
Neg logit contributions
 Ben-0.036
ila-0.034
Ben-0.033
 bat-0.033
 says-0.032
Token knew
Token ID2993
Feature activation+4.00e-02

Pos logit contributions
 unlike+0.148
 superpower+0.132
 matter+0.131
 former+0.123
 rarely+0.121
Neg logit contributions
ila-0.116
Ben-0.113
 Ben-0.113
line-0.110
 bat-0.108
Token that
Token ID326
Feature activation-1.02e-01

Pos logit contributions
 unlike+0.106
 matter+0.098
 superpower+0.098
 perspective+0.096
 blonde+0.096
Neg logit contributions
 Ben-0.024
 bat-0.024
Ben-0.023
ila-0.023
 se-0.022
Token no
Token ID645
Feature activation+3.18e-02

Pos logit contributions
ila+0.111
 Ben+0.111
Ben+0.111
 bat+0.110
 se+0.108
Neg logit contributions
 unlike-0.220
 matter-0.199
 superpower-0.197
 perspective-0.193
 blonde-0.192
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.051
 Terr+0.045
eway+0.045
 superpower+0.043
 blonde+0.043
Neg logit contributions
 next-0.053
 Ben-0.052
 put-0.051
 each-0.051
 bat-0.050
Token how
Token ID703
Feature activation+4.57e-02

Pos logit contributions
 unlike+0.607
 superpower+0.548
 blonde+0.541
 matter+0.529
 tactic+0.527
Neg logit contributions
 Ben-0.383
ila-0.365
 bat-0.357
 se-0.348
 next-0.348
Token bitter
Token ID12922
Feature activation+7.26e-02

Pos logit contributions
 unlike+0.090
 matter+0.081
 superpower+0.081
 tactic+0.079
 blonde+0.079
Neg logit contributions
 Ben-0.059
 bat-0.057
 se-0.057
ila-0.057
Ben-0.056
Token life
Token ID1204
Feature activation+5.37e-03

Pos logit contributions
 unlike+0.167
 superpower+0.155
 matter+0.152
 perspective+0.152
 blonde+0.151
Neg logit contributions
 Ben-0.068
 bat-0.065
ila-0.063
 se-0.063
Ben-0.063
Token got
Token ID1392
Feature activation-5.82e-02

Pos logit contributions
 unlike+0.011
 matter+0.010
 superpower+0.010
 blonde+0.009
 tactic+0.009
Neg logit contributions
Ben-0.006
 Ben-0.006
ila-0.006
 bat-0.006
 se-0.006
Token,
Token ID11
Feature activation+1.77e-02

Pos logit contributions
 Ben+0.051
 bat+0.049
 se+0.049
ila+0.046
 next+0.046
Neg logit contributions
 unlike-0.137
 superpower-0.127
 perspective-0.125
 tactic-0.125
eway-0.125
Tokenami
Token ID6277
Feature activation-1.86e-02

Pos logit contributions
 Ben+0.067
 bat+0.066
Ben+0.064
ila+0.064
 se+0.062
Neg logit contributions
 unlike-0.075
 matter-0.068
 superpower-0.067
 perspective-0.067
 blonde-0.065
Token figure
Token ID3785
Feature activation-2.06e-02

Pos logit contributions
 Ben+0.016
 se+0.016
 bat+0.015
ila+0.015
Ben+0.015
Neg logit contributions
 unlike-0.044
 matter-0.041
 superpower-0.040
 blonde-0.040
 tactic-0.040
Token.
Token ID13
Feature activation+2.77e-02

Pos logit contributions
Ben+0.018
 Ben+0.018
 bat+0.018
ila+0.017
 se+0.017
Neg logit contributions
 unlike-0.049
 superpower-0.044
 matter-0.044
 blonde-0.043
 perspective-0.043
Token But
Token ID887
Feature activation+1.12e-01

Pos logit contributions
 unlike+0.056
 matter+0.054
 superpower+0.054
 perspective+0.052
 avail+0.051
Neg logit contributions
 Ben-0.034
 put-0.030
 says-0.030
 goes-0.030
 next-0.030
Token no
Token ID645
Feature activation+3.35e-02

Pos logit contributions
 unlike+0.283
 superpower+0.275
 perspective+0.272
eway+0.272
 blonde+0.270
Neg logit contributions
 next-0.079
 Ben-0.078
 put-0.074
 they-0.069
 we-0.066
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.053
 Terr+0.049
eway+0.048
 superpower+0.045
 Tears+0.044
Neg logit contributions
 Ben-0.058
 put-0.058
 se-0.056
 next-0.056
ila-0.055
Token how
Token ID703
Feature activation+4.49e-02

Pos logit contributions
 unlike+0.624
 superpower+0.571
 blonde+0.556
 tactic+0.552
 Terr+0.546
Neg logit contributions
 Ben-0.376
ila-0.353
 bat-0.339
 next-0.338
 each-0.337
Token hard
Token ID1327
Feature activation+4.52e-02

Pos logit contributions
 unlike+0.086
 matter+0.077
 superpower+0.077
 perspective+0.075
 blonde+0.074
Neg logit contributions
 Ben-0.060
 bat-0.059
Ben-0.059
ila-0.058
 se-0.058
Token she
Token ID673
Feature activation+1.34e-02

Pos logit contributions
 unlike+0.107
 superpower+0.099
 perspective+0.098
 tactic+0.097
 matter+0.097
Neg logit contributions
 Ben-0.040
 bat-0.037
ila-0.036
 se-0.036
Ben-0.036
Token tried
Token ID3088
Feature activation+2.49e-02

Pos logit contributions
 unlike+0.024
 matter+0.023
 superpower+0.022
 tactic+0.021
 blonde+0.020
Neg logit contributions
ila-0.018
 Ben-0.018
Ben-0.018
 dads-0.018
 pige-0.018
Token,
Token ID11
Feature activation+1.83e-02

Pos logit contributions
 unlike+0.064
 superpower+0.059
 matter+0.059
eway+0.059
 tactic+0.059
Neg logit contributions
 Ben-0.018
 bat-0.016
 next-0.015
ila-0.015
 se-0.014
Token remember
Token ID3505
Feature activation-2.05e-02

Pos logit contributions
Ben+0.000
ila+0.000
 bat+0.000
 se+0.000
 dads+0.000
Neg logit contributions
 unlike-0.001
 matter-0.001
 blonde-0.001
 superpower-0.001
 alas-0.001
Token,
Token ID11
Feature activation+1.66e-02

Pos logit contributions
 Ben+0.019
 bat+0.019
ila+0.019
Ben+0.019
 se+0.018
Neg logit contributions
 unlike-0.047
 matter-0.043
 superpower-0.043
 perspective-0.042
 blonde-0.042
Token it
Token ID340
Feature activation-1.63e-02

Pos logit contributions
 unlike+0.034
 matter+0.032
 superpower+0.030
 blonde+0.030
 former+0.029
Neg logit contributions
ila-0.020
Ben-0.020
 se-0.019
 bat-0.018
 Ben-0.017
Token doesn
Token ID1595
Feature activation+4.49e-02

Pos logit contributions
Ben+0.017
ila+0.016
 bat+0.016
 Ben+0.016
 se+0.016
Neg logit contributions
 unlike-0.036
 matter-0.033
 superpower-0.033
 tactic-0.032
 perspective-0.031
Token't
Token ID470
Feature activation+1.37e-02

Pos logit contributions
 unlike+0.061
 matter+0.051
 superpower+0.050
 tactic+0.046
 blonde+0.045
Neg logit contributions
Ben-0.085
ila-0.085
 Ben-0.083
 dads-0.083
 se-0.083
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.019
 superpower+0.016
 matter+0.016
 perspective+0.015
 blonde+0.015
Neg logit contributions
 Ben-0.026
 bat-0.026
ila-0.025
Ben-0.025
 se-0.025
Token if
Token ID611
Feature activation+8.45e-02

Pos logit contributions
 unlike+0.551
 matter+0.495
 superpower+0.490
 perspective+0.479
 blonde+0.473
Neg logit contributions
 Ben-0.422
 bat-0.422
Ben-0.420
ila-0.417
 se-0.415
Token your
Token ID534
Feature activation+8.34e-03

Pos logit contributions
 unlike+0.210
 matter+0.197
 superpower+0.193
 perspective+0.189
 blonde+0.189
Neg logit contributions
Ben-0.061
 bat-0.060
ila-0.060
 se-0.059
 Ben-0.058
Token painting
Token ID12036
Feature activation+3.23e-02

Pos logit contributions
 unlike+0.016
 matter+0.014
 superpower+0.014
 perspective+0.013
 blonde+0.013
Neg logit contributions
 Ben-0.011
 bat-0.011
ila-0.011
Ben-0.011
 se-0.011
Token is
Token ID318
Feature activation-4.39e-03

Pos logit contributions
 unlike+0.070
 matter+0.065
 superpower+0.064
 perspective+0.062
 tactic+0.062
Neg logit contributions
Ben-0.034
 Ben-0.033
 bat-0.033
ila-0.033
 se-0.032
Token not
Token ID407
Feature activation-3.05e-02

Pos logit contributions
 Ben+0.005
 bat+0.005
 se+0.005
ila+0.005
Ben+0.005
Neg logit contributions
 unlike-0.009
 superpower-0.008
 matter-0.008
 tactic-0.008
 perspective-0.008
Token She
Token ID1375
Feature activation+8.04e-02

Pos logit contributions
 unlike+0.059
 matter+0.055
 superpower+0.054
 perspective+0.053
 tactic+0.053
Neg logit contributions
 Ben-0.025
Ben-0.022
 bat-0.022
ila-0.021
 se-0.019
Token was
Token ID373
Feature activation+9.09e-02

Pos logit contributions
 unlike+0.148
 superpower+0.134
 matter+0.134
 former+0.126
 blonde+0.125
Neg logit contributions
ila-0.109
Ben-0.107
 Ben-0.106
 bat-0.102
line-0.102
Token reminded
Token ID14516
Feature activation+9.00e-02

Pos logit contributions
 unlike+0.179
 superpower+0.157
 matter+0.156
 blonde+0.153
 perspective+0.149
Neg logit contributions
Ben-0.120
ila-0.117
 Ben-0.116
 bat-0.114
 se-0.113
Token that
Token ID326
Feature activation-1.04e-01

Pos logit contributions
 unlike+0.215
 matter+0.196
 superpower+0.195
 tactic+0.189
 blonde+0.188
Neg logit contributions
 bat-0.076
Ben-0.076
 Ben-0.074
 se-0.073
ila-0.073
Token no
Token ID645
Feature activation+3.20e-02

Pos logit contributions
ila+0.119
Ben+0.117
 bat+0.110
 se+0.108
 Ben+0.107
Neg logit contributions
 unlike-0.221
 matter-0.200
 superpower-0.195
 perspective-0.191
 blonde-0.188
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.053
 Terr+0.047
eway+0.047
 superpower+0.045
 blonde+0.045
Neg logit contributions
 Ben-0.053
 next-0.052
 bat-0.051
 se-0.051
 put-0.050
Token what
Token ID644
Feature activation+8.14e-02

Pos logit contributions
 unlike+0.634
 superpower+0.575
 blonde+0.572
 Terr+0.561
 darkened+0.556
Neg logit contributions
 Ben-0.365
ila-0.339
 each-0.336
 next-0.329
 we-0.328
Token,
Token ID11
Feature activation+1.68e-02

Pos logit contributions
 unlike+0.172
eway+0.157
 tactic+0.156
 perspective+0.156
 former+0.154
Neg logit contributions
 Ben-0.097
 bat-0.091
 we-0.087
 se-0.087
 next-0.086
Token she
Token ID673
Feature activation+1.10e-02

Pos logit contributions
 unlike+0.039
 superpower+0.036
 matter+0.036
 tactic+0.036
eway+0.035
Neg logit contributions
 Ben-0.015
 bat-0.014
ila-0.014
Ben-0.014
 se-0.014
Token had
Token ID550
Feature activation-3.26e-02

Pos logit contributions
 unlike+0.021
 matter+0.019
 superpower+0.018
 blonde+0.017
 perspective+0.017
Neg logit contributions
Ben-0.015
 Ben-0.015
ila-0.015
 bat-0.015
 se-0.014
Token a
Token ID257
Feature activation+3.95e-03

Pos logit contributions
 Ben+0.042
 bat+0.041
Ben+0.040
ila+0.040
 se+0.040
Neg logit contributions
 unlike-0.065
 superpower-0.058
 matter-0.058
 perspective-0.057
 tactic-0.057
Token always
Token ID1464
Feature activation+1.18e-02

Pos logit contributions
 unlike+0.113
 matter+0.110
 superpower+0.108
 blonde+0.105
 perspective+0.102
Neg logit contributions
ila-0.053
Ben-0.050
 Ben-0.048
 se-0.048
 pige-0.048
Token love
Token ID1842
Feature activation-1.55e-02

Pos logit contributions
 matter+0.023
 unlike+0.023
 superpower+0.022
 blonde+0.021
 former+0.020
Neg logit contributions
ila-0.014
Ben-0.013
 pige-0.013
 Ben-0.013
 se-0.012
Token you
Token ID345
Feature activation-1.20e-01

Pos logit contributions
Ben+0.015
ila+0.014
 bat+0.014
 se+0.013
line+0.012
Neg logit contributions
 unlike-0.035
 matter-0.032
 superpower-0.032
 blonde-0.031
 perspective-0.031
Token,
Token ID11
Feature activation+1.08e-02

Pos logit contributions
ila+0.117
 se+0.116
Ben+0.112
 bat+0.108
 Ben+0.107
Neg logit contributions
 unlike-0.269
 matter-0.249
 superpower-0.248
 blonde-0.241
 perspective-0.239
Token no
Token ID645
Feature activation+3.19e-02

Pos logit contributions
 unlike+0.020
 matter+0.019
 blonde+0.017
 superpower+0.017
 alas+0.016
Neg logit contributions
Ben-0.014
ila-0.014
 se-0.014
 bat-0.013
ars-0.013
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.049
eway+0.041
 Terr+0.041
 superpower+0.040
 blonde+0.039
Neg logit contributions
 Ben-0.059
 bat-0.058
 se-0.056
 next-0.055
 put-0.055
Token what
Token ID644
Feature activation+8.00e-02

Pos logit contributions
 unlike+0.646
 superpower+0.578
 blonde+0.568
 matter+0.567
 tactic+0.559
Neg logit contributions
 Ben-0.355
ila-0.336
 bat-0.333
Ben-0.326
 se-0.318
Token."
Token ID526
Feature activation-5.10e-02

Pos logit contributions
 unlike+0.188
 matter+0.173
 superpower+0.173
 blonde+0.169
 perspective+0.168
Neg logit contributions
Ben-0.068
 bat-0.068
 se-0.068
ila-0.067
 Ben-0.067
Token<|endoftext|>
Token ID50256
Feature activation+8.01e-03

Pos logit contributions
 se+0.064
ila+0.061
 bat+0.060
 wave+0.059
 dove+0.058
Neg logit contributions
 unlike-0.096
 Tears-0.089
 blonde-0.088
 superpower-0.086
 Terr-0.086
TokenL
Token ID43
Feature activation-4.55e-02

Pos logit contributions
 unlike+0.013
 superpower+0.012
 blonde+0.012
 Tears+0.012
 perspective+0.011
Neg logit contributions
 se-0.011
 dads-0.011
 bat-0.011
ila-0.011
Ben-0.011
Tokenily
Token ID813
Feature activation-1.02e-01

Pos logit contributions
 bat+0.054
 se+0.054
 wave+0.052
 dads+0.049
 pige+0.049
Neg logit contributions
 unlike-0.082
 matter-0.081
 tactic-0.080
 blonde-0.076
eway-0.076
Token and
Token ID290
Feature activation-7.52e-02

Pos logit contributions
 unlike+0.056
 matter+0.052
 superpower+0.052
 perspective+0.051
 tactic+0.051
Neg logit contributions
 Ben-0.012
 bat-0.011
 se-0.011
Ben-0.011
ila-0.011
Token felt
Token ID2936
Feature activation+1.26e-02

Pos logit contributions
ila+0.081
Ben+0.079
 bat+0.074
 Ben+0.072
 se+0.066
Neg logit contributions
 unlike-0.168
 blonde-0.147
 matter-0.147
 superpower-0.144
 Tears-0.142
Token content
Token ID2695
Feature activation+6.68e-02

Pos logit contributions
 unlike+0.027
 matter+0.024
 superpower+0.023
 blonde+0.023
 tactic+0.023
Neg logit contributions
Ben-0.014
ila-0.014
 Ben-0.014
 bat-0.014
 se-0.014
Token.
Token ID13
Feature activation+2.99e-02

Pos logit contributions
 unlike+0.156
 superpower+0.140
 matter+0.139
 perspective+0.138
 blonde+0.137
Neg logit contributions
ila-0.062
Ben-0.062
 Ben-0.062
 bat-0.060
 se-0.058
Token No
Token ID1400
Feature activation+8.70e-02

Pos logit contributions
 unlike+0.047
 matter+0.043
 superpower+0.041
 perspective+0.040
 tactic+0.040
Neg logit contributions
 Ben-0.053
 bat-0.049
Ben-0.049
ila-0.048
 se-0.047
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.090
eway+0.071
 superpower+0.070
 blonde+0.066
 former+0.064
Neg logit contributions
 Ben-0.198
 se-0.195
 bat-0.193
Ben-0.192
ila-0.192
Token what
Token ID644
Feature activation+8.39e-02

Pos logit contributions
 unlike+0.657
 blonde+0.616
 superpower+0.610
 Terr+0.597
 darkened+0.588
Neg logit contributions
 Ben-0.360
 each-0.318
 we-0.317
 next-0.314
ila-0.310
Token had
Token ID550
Feature activation-3.13e-02

Pos logit contributions
 unlike+0.167
 superpower+0.149
 matter+0.148
 perspective+0.147
 tactic+0.147
Neg logit contributions
 Ben-0.111
 bat-0.107
 se-0.106
Ben-0.105
ila-0.104
Token happened
Token ID3022
Feature activation+7.92e-02

Pos logit contributions
 bat+0.034
Ben+0.033
ila+0.033
 Ben+0.033
 se+0.032
Neg logit contributions
 unlike-0.068
 matter-0.062
 superpower-0.061
 blonde-0.059
 tactic-0.059
Token before
Token ID878
Feature activation+7.24e-02

Pos logit contributions
 unlike+0.183
 superpower+0.166
 matter+0.166
 perspective+0.163
 blonde+0.162
Neg logit contributions
 Ben-0.073
 bat-0.072
ila-0.072
Ben-0.072
 se-0.070
Token he
Token ID339
Feature activation+1.76e-02

Pos logit contributions
 unlike+0.162
 matter+0.146
 superpower+0.146
 blonde+0.143
 perspective+0.142
Neg logit contributions
ila-0.073
Ben-0.073
 Ben-0.071
 bat-0.070
 se-0.068
Token again
Token ID757
Feature activation+7.57e-02

Pos logit contributions
Ben+0.015
ila+0.015
 bat+0.015
 Ben+0.015
 se+0.014
Neg logit contributions
 unlike-0.039
 superpower-0.035
 perspective-0.035
 matter-0.035
 tactic-0.034
Token and
Token ID290
Feature activation-7.53e-02

Pos logit contributions
 unlike+0.160
 superpower+0.147
 perspective+0.146
 matter+0.141
 former+0.140
Neg logit contributions
Ben-0.083
ila-0.082
 se-0.076
 Ben-0.075
 bat-0.075
Token again
Token ID757
Feature activation+7.61e-02

Pos logit contributions
ila+0.092
Ben+0.088
 pige+0.083
 bat+0.081
irts+0.080
Neg logit contributions
 unlike-0.138
 matter-0.132
 Tears-0.130
 alas-0.126
 terribly-0.123
Token.
Token ID13
Feature activation+2.90e-02

Pos logit contributions
 unlike+0.155
 superpower+0.142
 perspective+0.140
 matter+0.138
 blonde+0.133
Neg logit contributions
ila-0.088
Ben-0.088
 bat-0.081
 se-0.080
 pige-0.079
Token No
Token ID1400
Feature activation+8.61e-02

Pos logit contributions
 unlike+0.062
 matter+0.057
 superpower+0.056
 perspective+0.055
 blonde+0.055
Neg logit contributions
 Ben-0.033
 bat-0.032
Ben-0.031
ila-0.031
 se-0.030
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.127
 superpower+0.107
 blonde+0.103
 perspective+0.103
eway+0.103
Neg logit contributions
 Ben-0.157
 bat-0.153
 se-0.153
ila-0.153
Ben-0.152
Token how
Token ID703
Feature activation+4.58e-02

Pos logit contributions
 unlike+0.614
 blonde+0.552
 superpower+0.551
 matter+0.534
 tactic+0.533
Neg logit contributions
 Ben-0.384
 bat-0.363
ila-0.358
 each-0.348
 se-0.346
Token many
Token ID867
Feature activation-1.66e-02

Pos logit contributions
 unlike+0.081
 matter+0.073
 superpower+0.073
 perspective+0.071
 blonde+0.069
Neg logit contributions
Ben-0.066
 bat-0.065
ila-0.064
 Ben-0.064
 se-0.063
Token times
Token ID1661
Feature activation+2.89e-02

Pos logit contributions
Ben+0.023
ila+0.022
 Ben+0.021
 bat+0.021
 se+0.021
Neg logit contributions
 unlike-0.032
 matter-0.030
 superpower-0.028
 blonde-0.028
 tactic-0.027
Token they
Token ID484
Feature activation-1.01e-01

Pos logit contributions
 unlike+0.067
 matter+0.063
 superpower+0.062
eway+0.060
 tactic+0.060
Neg logit contributions
Ben-0.026
ila-0.023
 bat-0.022
 se-0.022
 Ben-0.021
Token came
Token ID1625
Feature activation+1.64e-02

Pos logit contributions
ila+0.132
Ben+0.130
 Ben+0.129
 bat+0.125
 se+0.124
Neg logit contributions
 unlike-0.189
 matter-0.179
 superpower-0.173
 perspective-0.167
 blonde-0.166
Token forget
Token ID6044
Feature activation-7.78e-02

Pos logit contributions
 Ben+0.034
ila+0.033
Ben+0.033
 bat+0.032
 se+0.031
Neg logit contributions
 unlike-0.099
 matter-0.092
 superpower-0.091
 perspective-0.089
 tactic-0.089
Token to
Token ID284
Feature activation+7.78e-02

Pos logit contributions
 Ben+0.055
 bat+0.050
ila+0.047
 se+0.047
Ben+0.045
Neg logit contributions
 unlike-0.202
 superpower-0.186
 blonde-0.186
 matter-0.186
 tactic-0.182
Token dream
Token ID4320
Feature activation+9.41e-03

Pos logit contributions
 unlike+0.175
 matter+0.164
 superpower+0.162
 perspective+0.158
 blonde+0.157
Neg logit contributions
Ben-0.075
ila-0.075
 Ben-0.074
 bat-0.071
 se-0.070
Token,
Token ID11
Feature activation+1.71e-02

Pos logit contributions
 unlike+0.018
 matter+0.016
 superpower+0.016
 perspective+0.015
eway+0.015
Neg logit contributions
Ben-0.013
 bat-0.012
ila-0.012
 Ben-0.012
 se-0.012
Token no
Token ID645
Feature activation+3.26e-02

Pos logit contributions
 unlike+0.036
 matter+0.032
 superpower+0.032
 perspective+0.031
 blonde+0.031
Neg logit contributions
Ben-0.020
ila-0.020
 Ben-0.019
 bat-0.019
 se-0.019
Token matter
Token ID2300
Feature activation+3.02e-01

Pos logit contributions
 unlike+0.036
 Terr+0.029
 blonde+0.028
eway+0.027
 superpower+0.027
Neg logit contributions
 Ben-0.073
 se-0.073
 next-0.072
 bat-0.071
 wave-0.070
Token how
Token ID703
Feature activation+4.62e-02

Pos logit contributions
 unlike+0.631
 blonde+0.579
 superpower+0.570
 tactic+0.555
 former+0.554
Neg logit contributions
 Ben-0.369
ila-0.345
 bat-0.336
 next-0.332
 each-0.328
Token far
Token ID1290
Feature activation-6.02e-02

Pos logit contributions
 unlike+0.094
 matter+0.085
 superpower+0.085
 blonde+0.083
 tactic+0.083
Neg logit contributions
 Ben-0.056
 bat-0.055
ila-0.055
 se-0.054
Ben-0.054
Token away
Token ID1497
Feature activation+1.89e-01

Pos logit contributions
Ben+0.063
 bat+0.061
ila+0.061
 se+0.060
 Ben+0.059
Neg logit contributions
 unlike-0.131
 matter-0.123
 superpower-0.118
 tactic-0.114
 blonde-0.113
Token they
Token ID484
Feature activation-1.01e-01

Pos logit contributions
 unlike+0.432
 superpower+0.394
 matter+0.394
 blonde+0.386
 perspective+0.386
Neg logit contributions
 Ben-0.183
 bat-0.178
ila-0.175
Ben-0.173
 se-0.172
Token seemed
Token ID3947
Feature activation+1.44e-01

Pos logit contributions
Ben+0.137
ila+0.137
 Ben+0.137
 bat+0.136
 se+0.133
Neg logit contributions
 unlike-0.186
 matter-0.170
 superpower-0.166
 perspective-0.161
 tactic-0.159
Token down
Token ID866
Feature activation+1.12e-01

Pos logit contributions
 Ben+0.109
 bat+0.104
ila+0.099
 se+0.098
Ben+0.097
Neg logit contributions
 unlike-0.387
 matter-0.360
 superpower-0.359
 blonde-0.353
 tactic-0.352
Token the
Token ID262
Feature activation-6.88e-02

Pos logit contributions
 unlike+0.308
 matter+0.286
 superpower+0.286
 blonde+0.281
 perspective+0.279
Neg logit contributions
 Ben-0.056
 bat-0.053
ila-0.052
Ben-0.050
 se-0.050
Token tree
Token ID5509
Feature activation-4.01e-02

Pos logit contributions
ila+0.091
 Ben+0.089
Ben+0.088
 bat+0.085
 se+0.085
Neg logit contributions
 unlike-0.133
 matter-0.120
 superpower-0.120
 perspective-0.117
 blonde-0.116
Token and
Token ID290
Feature activation-9.08e-02

Pos logit contributions
Ben+0.036
ila+0.035
 Ben+0.035
 bat+0.034
 se+0.034
Neg logit contributions
 unlike-0.094
 superpower-0.086
 matter-0.086
 perspective-0.085
 blonde-0.084
Token scam
Token ID19126
Feature activation-8.89e-02

Pos logit contributions
ila+0.084
Ben+0.083
 Ben+0.081
 bat+0.080
 se+0.079
Neg logit contributions
 unlike-0.209
 matter-0.192
 superpower-0.190
 perspective-0.185
 blonde-0.185
Tokenper
Token ID525
Feature activation-2.57e-01

Pos logit contributions
Ben+0.146
 se+0.146
 bat+0.143
 Ben+0.141
 says+0.140
Neg logit contributions
 unlike-0.137
eway-0.120
 superpower-0.120
 blonde-0.119
 perspective-0.116
Token away
Token ID1497
Feature activation+1.59e-01

Pos logit contributions
 Ben+0.332
 bat+0.328
ila+0.326
Ben+0.325
 se+0.322
Neg logit contributions
 unlike-0.499
 matter-0.447
 superpower-0.446
 perspective-0.434
 blonde-0.433
Token.
Token ID13
Feature activation+1.28e-02

Pos logit contributions
 unlike+0.389
 matter+0.369
 superpower+0.362
 blonde+0.356
 tactic+0.354
Neg logit contributions
 Ben-0.124
 bat-0.123
 se-0.117
 next-0.113
ila-0.110
Token
Token ID198
Feature activation-3.22e-02

Pos logit contributions
 unlike+0.028
 matter+0.026
 superpower+0.025
 perspective+0.025
 tactic+0.025
Neg logit contributions
 Ben-0.015
Ben-0.014
 bat-0.014
ila-0.013
 se-0.013
Token
Token ID198
Feature activation-3.15e-02

Pos logit contributions
 Ben+0.030
 bat+0.030
 se+0.030
ila+0.029
Ben+0.028
Neg logit contributions
 unlike-0.073
 superpower-0.067
 matter-0.065
 perspective-0.065
 blonde-0.065
TokenThe
Token ID464
Feature activation-4.24e-02

Pos logit contributions
 Ben+0.038
 se+0.037
 bat+0.037
ila+0.036
Ben+0.035
Neg logit contributions
 unlike-0.064
 superpower-0.058
 matter-0.057
 perspective-0.056
 blonde-0.056
Token and
Token ID290
Feature activation-9.44e-02

Pos logit contributions
 unlike+0.376
 superpower+0.354
 matter+0.351
 tactic+0.348
 perspective+0.344
Neg logit contributions
 se-0.087
 Ben-0.085
 bat-0.083
 next-0.077
ila-0.071
Token fierce
Token ID14800
Feature activation+2.93e-02

Pos logit contributions
 Ben+0.122
 se+0.121
 bat+0.118
ila+0.113
 put+0.113
Neg logit contributions
 unlike-0.182
 superpower-0.170
 perspective-0.167
 tactic-0.166
 blonde-0.161
Token winds
Token ID13520
Feature activation-7.40e-02

Pos logit contributions
 unlike+0.062
 superpower+0.056
 matter+0.056
 tactic+0.056
 perspective+0.055
Neg logit contributions
 Ben-0.034
 se-0.032
 bat-0.032
ila-0.031
Ben-0.030
Token started
Token ID2067
Feature activation+4.12e-02

Pos logit contributions
 Ben+0.077
Ben+0.076
ila+0.075
 bat+0.074
 se+0.072
Neg logit contributions
 unlike-0.163
 matter-0.147
 superpower-0.145
 perspective-0.144
 blonde-0.141
Token to
Token ID284
Feature activation+4.30e-02

Pos logit contributions
 unlike+0.095
 tactic+0.090
 superpower+0.090
 matter+0.089
 blonde+0.088
Neg logit contributions
 se-0.037
 bat-0.036
 Ben-0.035
 next-0.034
 put-0.032
Token blow
Token ID6611
Feature activation-2.56e-01

Pos logit contributions
 unlike+0.092
 superpower+0.083
 matter+0.082
 blonde+0.081
 perspective+0.081
Neg logit contributions
 bat-0.047
 Ben-0.047
 se-0.046
Ben-0.046
ila-0.046
Token.
Token ID13
Feature activation+1.49e-03

Pos logit contributions
 Ben+0.261
 bat+0.260
 se+0.252
Ben+0.241
ila+0.239
Neg logit contributions
 unlike-0.571
 superpower-0.525
 matter-0.515
 perspective-0.513
 tactic-0.511
Token Tim
Token ID5045
Feature activation-3.60e-02

Pos logit contributions
 unlike+0.003
 matter+0.003
 superpower+0.003
 perspective+0.003
 tactic+0.003
Neg logit contributions
 Ben-0.002
Ben-0.002
 bat-0.002
ila-0.002
 se-0.002
Tokenmy
Token ID1820
Feature activation+6.26e-02

Pos logit contributions
 Ben+0.040
ila+0.040
 bat+0.039
Ben+0.039
 se+0.039
Neg logit contributions
 unlike-0.076
 matter-0.069
 superpower-0.069
 blonde-0.067
 perspective-0.067
Token was
Token ID373
Feature activation+6.89e-02

Pos logit contributions
 unlike+0.124
 matter+0.111
 superpower+0.110
 blonde+0.108
 perspective+0.107
Neg logit contributions
ila-0.078
 Ben-0.078
Ben-0.076
 bat-0.075
 se-0.072
Token scared
Token ID12008
Feature activation+9.00e-02

Pos logit contributions
 unlike+0.134
 matter+0.128
 superpower+0.126
 perspective+0.124
 tactic+0.123
Neg logit contributions
 bat-0.086
 Ben-0.086
 se-0.083
 put-0.083
 next-0.081
Token sleeping
Token ID11029
Feature activation-4.76e-02

Pos logit contributions
 unlike+0.000
 matter+0.000
 superpower+0.000
 perspective+0.000
 blonde+0.000
Neg logit contributions
 Ben-0.000
 bat-0.000
ila-0.000
Ben-0.000
 se-0.000
Token,
Token ID11
Feature activation-2.59e-03

Pos logit contributions
 Ben+0.040
ila+0.039
 bat+0.039
Ben+0.039
 se+0.039
Neg logit contributions
 unlike-0.113
 superpower-0.104
 matter-0.103
 perspective-0.101
 blonde-0.101
Token okay
Token ID8788
Feature activation+1.31e-01

Pos logit contributions
 Ben+0.003
ila+0.003
 bat+0.003
Ben+0.003
 se+0.003
Neg logit contributions
 unlike-0.006
 matter-0.005
 superpower-0.005
 blonde-0.005
 perspective-0.005
Token?
Token ID30
Feature activation+8.90e-03

Pos logit contributions
 unlike+0.340
 blonde+0.315
 superpower+0.313
 perspective+0.312
 tactic+0.309
Neg logit contributions
 Ben-0.097
 bat-0.089
 next-0.084
 we-0.082
 put-0.077
Token Whis
Token ID28424
Feature activation-6.22e-02

Pos logit contributions
 unlike+0.019
 matter+0.017
 perspective+0.017
 superpower+0.017
 tactic+0.017
Neg logit contributions
 Ben-0.011
 bat-0.010
Ben-0.010
ila-0.010
 says-0.010
Tokenper
Token ID525
Feature activation-2.56e-01

Pos logit contributions
 bat+0.084
 se+0.080
Ben+0.073
 Ben+0.073
 goes+0.073
Neg logit contributions
 unlike-0.115
 superpower-0.107
 perspective-0.106
 blonde-0.103
 matter-0.102
Token means
Token ID1724
Feature activation+3.35e-02

Pos logit contributions
 Ben+0.259
ila+0.254
 bat+0.249
Ben+0.247
 se+0.244
Neg logit contributions
 unlike-0.572
 matter-0.518
 superpower-0.518
 perspective-0.509
 blonde-0.507
Token to
Token ID284
Feature activation+4.73e-02

Pos logit contributions
 unlike+0.075
 matter+0.068
 superpower+0.067
 tactic+0.067
 perspective+0.067
Neg logit contributions
 Ben-0.035
 bat-0.034
ila-0.032
 se-0.032
 next-0.031
Token talk
Token ID1561
Feature activation-4.75e-02

Pos logit contributions
 unlike+0.104
 matter+0.101
 superpower+0.097
 blonde+0.094
 perspective+0.093
Neg logit contributions
ila-0.044
Ben-0.043
 se-0.042
 Ben-0.041
 bat-0.041
Token very
Token ID845
Feature activation+1.03e-01

Pos logit contributions
 Ben+0.037
 next+0.034
 bat+0.033
 and+0.033
 goes+0.032
Neg logit contributions
 unlike-0.118
 matter-0.110
 blonde-0.110
 perspective-0.110
 tactic-0.110
Token softly
Token ID26625
Feature activation+1.02e-02

Pos logit contributions
 unlike+0.164
 matter+0.144
 superpower+0.143
 blonde+0.138
 perspective+0.137
Neg logit contributions
 Ben-0.169
 bat-0.168
Ben-0.167
ila-0.167
 se-0.166
Token excited
Token ID6568
Feature activation-4.96e-03

Pos logit contributions
 unlike+0.027
 matter+0.024
 superpower+0.024
 perspective+0.023
 tactic+0.023
Neg logit contributions
 Ben-0.016
 bat-0.016
ila-0.016
Ben-0.016
 se-0.015
Token to
Token ID284
Feature activation+4.75e-02

Pos logit contributions
 Ben+0.005
ila+0.005
Ben+0.005
 bat+0.005
 se+0.005
Neg logit contributions
 unlike-0.011
 matter-0.010
 superpower-0.010
 perspective-0.010
 blonde-0.010
Token feel
Token ID1254
Feature activation-1.70e-02

Pos logit contributions
 unlike+0.106
 superpower+0.101
 matter+0.101
 blonde+0.097
 tactic+0.097
Neg logit contributions
ila-0.044
Ben-0.044
 Ben-0.039
 pige-0.038
 se-0.037
Token the
Token ID262
Feature activation-7.71e-02

Pos logit contributions
 Ben+0.016
Ben+0.016
ila+0.016
 bat+0.016
 se+0.015
Neg logit contributions
 unlike-0.039
 matter-0.035
 superpower-0.035
 blonde-0.034
 perspective-0.034
Token wind
Token ID2344
Feature activation-8.55e-02

Pos logit contributions
 bat+0.089
 se+0.086
 Ben+0.085
Ben+0.082
ila+0.079
Neg logit contributions
 unlike-0.167
 superpower-0.148
eway-0.146
 tactic-0.146
 matter-0.145
Token blow
Token ID6611
Feature activation-2.55e-01

Pos logit contributions
Ben+0.102
ila+0.100
 Ben+0.097
 bat+0.089
 dads+0.086
Neg logit contributions
 unlike-0.182
 matter-0.160
 superpower-0.157
 perspective-0.155
 blonde-0.153
Token through
Token ID832
Feature activation+7.59e-02

Pos logit contributions
 Ben+0.269
 bat+0.265
ila+0.264
Ben+0.264
 se+0.259
Neg logit contributions
 unlike-0.554
 matter-0.503
 superpower-0.501
 perspective-0.489
 blonde-0.489
Token their
Token ID511
Feature activation-3.49e-02

Pos logit contributions
 unlike+0.212
 superpower+0.196
 matter+0.196
 tactic+0.192
 blonde+0.192
Neg logit contributions
 Ben-0.036
 bat-0.034
Ben-0.032
ila-0.032
 se-0.032
Token hair
Token ID4190
Feature activation+4.73e-02

Pos logit contributions
ila+0.050
 Ben+0.048
Ben+0.048
 bat+0.046
 se+0.046
Neg logit contributions
 unlike-0.063
 superpower-0.058
 matter-0.057
 blonde-0.056
 perspective-0.055
Token and
Token ID290
Feature activation-1.07e-01

Pos logit contributions
 unlike+0.112
 matter+0.102
 superpower+0.101
 blonde+0.100
 perspective+0.100
Neg logit contributions
ila-0.041
Ben-0.041
 Ben-0.039
 bat-0.038
 se-0.037
Token spread
Token ID4104
Feature activation-1.54e-01

Pos logit contributions
ila+0.096
Ben+0.094
 Ben+0.093
 bat+0.091
 se+0.089
Neg logit contributions
 unlike-0.250
 matter-0.229
 superpower-0.227
 blonde-0.224
 perspective-0.222
Token,
Token ID11
Feature activation+1.44e-02

Pos logit contributions
 unlike+0.276
 superpower+0.251
 matter+0.251
 perspective+0.245
 tactic+0.245
Neg logit contributions
 Ben-0.129
 bat-0.125
Ben-0.124
ila-0.123
 se-0.123
Token the
Token ID262
Feature activation-5.33e-02

Pos logit contributions
 unlike+0.033
 superpower+0.030
 matter+0.030
 perspective+0.030
 tactic+0.030
Neg logit contributions
 Ben-0.014
 bat-0.013
ila-0.013
Ben-0.013
 se-0.013
Token wind
Token ID2344
Feature activation-8.18e-02

Pos logit contributions
 Ben+0.065
 bat+0.064
Ben+0.064
 se+0.064
ila+0.063
Neg logit contributions
 unlike-0.107
 matter-0.096
 superpower-0.096
 perspective-0.094
 blonde-0.093
Token started
Token ID2067
Feature activation+4.34e-02

Pos logit contributions
 Ben+0.100
Ben+0.099
ila+0.096
 bat+0.095
 se+0.088
Neg logit contributions
 unlike-0.167
 matter-0.149
 superpower-0.146
 perspective-0.144
 darkened-0.144
Token to
Token ID284
Feature activation+4.66e-02

Pos logit contributions
 unlike+0.099
 superpower+0.094
 tactic+0.094
 matter+0.093
 blonde+0.092
Neg logit contributions
 se-0.041
 bat-0.040
 next-0.038
 Ben-0.037
 put-0.036
Token blow
Token ID6611
Feature activation-2.52e-01

Pos logit contributions
 unlike+0.104
 matter+0.100
 superpower+0.094
 perspective+0.094
 darkened+0.092
Neg logit contributions
ila-0.047
Ben-0.046
 Ben-0.045
 bat-0.041
 dads-0.039
Token and
Token ID290
Feature activation-8.99e-02

Pos logit contributions
 bat+0.208
 Ben+0.207
 se+0.200
Ben+0.187
 next+0.183
Neg logit contributions
 unlike-0.614
 superpower-0.571
 matter-0.555
 perspective-0.554
 blonde-0.554
Token it
Token ID340
Feature activation-3.68e-02

Pos logit contributions
 Ben+0.062
 bat+0.060
 se+0.058
Ben+0.055
ila+0.055
Neg logit contributions
 unlike-0.231
 superpower-0.214
 matter-0.212
 perspective-0.211
 blonde-0.211
Token made
Token ID925
Feature activation-9.30e-02

Pos logit contributions
 Ben+0.046
 bat+0.045
Ben+0.045
ila+0.045
 se+0.044
Neg logit contributions
 unlike-0.073
 matter-0.066
 superpower-0.066
 perspective-0.064
 blonde-0.064
Token Lily
Token ID20037
Feature activation-5.45e-02

Pos logit contributions
 Ben+0.061
 bat+0.051
 se+0.048
Ben+0.048
 each+0.047
Neg logit contributions
 unlike-0.252
 superpower-0.236
 tactic-0.230
 perspective-0.230
 blonde-0.227
Token sh
Token ID427
Feature activation-6.68e-02

Pos logit contributions
 bat+0.059
 Ben+0.057
 se+0.055
ila+0.053
 put+0.052
Neg logit contributions
 unlike-0.120
 matter-0.109
 superpower-0.109
 perspective-0.107
 blonde-0.105
Token,
Token ID11
Feature activation+6.51e-04

Pos logit contributions
 next+0.100
 Ben+0.098
 bat+0.092
ila+0.088
 se+0.087
Neg logit contributions
 unlike-0.317
 matter-0.302
 blonde-0.289
 superpower-0.287
 meanings-0.286
Token the
Token ID262
Feature activation-7.32e-02

Pos logit contributions
 unlike+0.002
 tactic+0.001
 superpower+0.001
 matter+0.001
 perspective+0.001
Neg logit contributions
 Ben-0.001
 they-0.001
 bat-0.001
Ben-0.001
ila-0.001
Token wind
Token ID2344
Feature activation-8.00e-02

Pos logit contributions
 Ben+0.088
 bat+0.087
Ben+0.086
ila+0.085
 se+0.085
Neg logit contributions
 unlike-0.150
 matter-0.135
 superpower-0.134
 tactic-0.131
 perspective-0.131
Token started
Token ID2067
Feature activation+4.26e-02

Pos logit contributions
 Ben+0.101
Ben+0.098
ila+0.098
 bat+0.096
 se+0.090
Neg logit contributions
 unlike-0.161
 matter-0.144
 superpower-0.142
 perspective-0.138
 darkened-0.137
Token to
Token ID284
Feature activation+4.75e-02

Pos logit contributions
 unlike+0.094
 matter+0.087
 superpower+0.087
 tactic+0.085
 blonde+0.085
Neg logit contributions
 bat-0.043
 se-0.043
 Ben-0.043
Ben-0.040
ila-0.040
Token blow
Token ID6611
Feature activation-2.51e-01

Pos logit contributions
 unlike+0.101
 matter+0.099
 darkened+0.091
 superpower+0.091
 perspective+0.090
Neg logit contributions
ila-0.053
Ben-0.050
 Ben-0.049
 pige-0.046
 dads-0.045
Token and
Token ID290
Feature activation-9.70e-02

Pos logit contributions
 bat+0.190
 Ben+0.189
 se+0.183
Ben+0.167
 next+0.164
Neg logit contributions
 unlike-0.629
 superpower-0.589
 perspective-0.577
 matter-0.573
 tactic-0.571
Token the
Token ID262
Feature activation-7.55e-02

Pos logit contributions
 Ben+0.082
 bat+0.081
 se+0.078
Ben+0.078
ila+0.077
Neg logit contributions
 unlike-0.233
 superpower-0.214
 matter-0.213
 perspective-0.210
 blonde-0.209
Token trees
Token ID7150
Feature activation+8.10e-03

Pos logit contributions
 Ben+0.078
ila+0.077
Ben+0.076
 bat+0.075
 se+0.073
Neg logit contributions
 unlike-0.166
 matter-0.152
 superpower-0.151
 perspective-0.148
 blonde-0.147
Token began
Token ID2540
Feature activation+7.69e-02

Pos logit contributions
 unlike+0.017
 matter+0.015
 superpower+0.015
 blonde+0.014
 perspective+0.014
Neg logit contributions
 Ben-0.010
Ben-0.010
ila-0.010
 bat-0.009
 se-0.009
Token to
Token ID284
Feature activation+4.97e-02

Pos logit contributions
 unlike+0.182
 matter+0.172
 superpower+0.170
 tactic+0.169
eway+0.166
Neg logit contributions
 se-0.069
 bat-0.067
 Ben-0.063
 next-0.061
 put-0.058
Token his
Token ID465
Feature activation+7.78e-02

Pos logit contributions
 unlike+0.108
 matter+0.103
 superpower+0.099
 former+0.096
 blonde+0.096
Neg logit contributions
Ben-0.043
ila-0.043
 se-0.039
 bat-0.039
 goes-0.039
Token trunk
Token ID21427
Feature activation+5.41e-02

Pos logit contributions
 unlike+0.168
 matter+0.150
 superpower+0.147
 perspective+0.147
 blonde+0.146
Neg logit contributions
 Ben-0.090
 bat-0.089
 se-0.084
Ben-0.083
ila-0.083
Token and
Token ID290
Feature activation-9.75e-02

Pos logit contributions
 unlike+0.143
 matter+0.131
 superpower+0.131
 perspective+0.129
 tactic+0.129
Neg logit contributions
Ben-0.032
ila-0.031
 Ben-0.031
 bat-0.029
 se-0.028
Token started
Token ID2067
Feature activation+4.73e-02

Pos logit contributions
ila+0.087
Ben+0.086
 bat+0.082
 Ben+0.081
 se+0.079
Neg logit contributions
 unlike-0.227
 matter-0.209
 superpower-0.207
 blonde-0.202
 tactic-0.200
Token to
Token ID284
Feature activation+5.01e-02

Pos logit contributions
 unlike+0.104
 matter+0.094
 superpower+0.094
 perspective+0.092
 blonde+0.091
Neg logit contributions
 Ben-0.050
 bat-0.050
 se-0.048
ila-0.048
Ben-0.048
Token blow
Token ID6611
Feature activation-2.50e-01

Pos logit contributions
 unlike+0.116
 superpower+0.105
 matter+0.105
 perspective+0.103
 tactic+0.102
Neg logit contributions
 bat-0.048
 Ben-0.047
 se-0.046
Ben-0.045
ila-0.044
Token.
Token ID13
Feature activation+4.63e-03

Pos logit contributions
ila+0.233
Ben+0.232
 Ben+0.231
 bat+0.230
 se+0.225
Neg logit contributions
 unlike-0.569
 matter-0.521
 superpower-0.518
 perspective-0.505
 blonde-0.505
Token A
Token ID317
Feature activation+7.63e-02

Pos logit contributions
 unlike+0.008
 matter+0.007
 superpower+0.007
 perspective+0.007
 tactic+0.007
Neg logit contributions
 Ben-0.007
 bat-0.007
Ben-0.007
ila-0.007
 se-0.007
Token bunch
Token ID7684
Feature activation+3.01e-02

Pos logit contributions
 unlike+0.170
eway+0.152
 tactic+0.151
 superpower+0.150
 matter+0.150
Neg logit contributions
 se-0.085
 bat-0.085
Ben-0.082
 Ben-0.082
 slide-0.077
Token of
Token ID286
Feature activation-7.97e-04

Pos logit contributions
 unlike+0.064
 superpower+0.055
 matter+0.054
 darkened+0.053
eway+0.053
Neg logit contributions
Ben-0.032
 Ben-0.032
 bat-0.031
ila-0.031
 dads-0.031
Token delicious
Token ID12625
Feature activation+6.47e-02

Pos logit contributions
 Ben+0.001
 bat+0.001
 se+0.001
ila+0.001
Ben+0.001
Neg logit contributions
 unlike-0.002
 superpower-0.002
 matter-0.002
 perspective-0.002
 tactic-0.002
Token the
Token ID262
Feature activation-7.25e-02

Pos logit contributions
 unlike+0.098
 superpower+0.089
 blonde+0.088
 matter+0.088
 perspective+0.087
Neg logit contributions
 Ben-0.047
 bat-0.044
 se-0.044
Ben-0.041
ila-0.041
Token other
Token ID584
Feature activation+3.01e-02

Pos logit contributions
 bat+0.090
 Ben+0.090
 se+0.088
Ben+0.087
ila+0.085
Neg logit contributions
 unlike-0.147
 superpower-0.131
 matter-0.128
 blonde-0.127
 perspective-0.127
Token balls
Token ID11333
Feature activation-4.08e-02

Pos logit contributions
 unlike+0.064
 matter+0.061
 blonde+0.058
 superpower+0.057
 perspective+0.057
Neg logit contributions
ila-0.032
 se-0.031
Ben-0.031
 bat-0.030
 Ben-0.029
Token.
Token ID13
Feature activation+1.03e-03

Pos logit contributions
Ben+0.042
ila+0.041
 Ben+0.038
 bat+0.037
 se+0.037
Neg logit contributions
 unlike-0.092
 matter-0.081
 superpower-0.080
 perspective-0.080
 tactic-0.079
Token Cas
Token ID11294
Feature activation-5.04e-02

Pos logit contributions
 unlike+0.002
 matter+0.002
 superpower+0.002
 perspective+0.002
 tactic+0.002
Neg logit contributions
 Ben-0.001
Ben-0.001
ila-0.001
 bat-0.001
 says-0.001
Tokenper
Token ID525
Feature activation-2.50e-01

Pos logit contributions
 bat+0.048
 se+0.046
Ben+0.045
 Ben+0.045
ila+0.045
Neg logit contributions
 unlike-0.116
 superpower-0.105
 matter-0.105
 perspective-0.103
 blonde-0.101
Token started
Token ID2067
Feature activation+4.82e-02

Pos logit contributions
 Ben+0.297
 bat+0.295
Ben+0.292
ila+0.290
 se+0.281
Neg logit contributions
 unlike-0.505
 matter-0.454
 superpower-0.451
 blonde-0.440
 tactic-0.438
Token to
Token ID284
Feature activation+5.20e-02

Pos logit contributions
 unlike+0.117
 matter+0.109
 perspective+0.106
eway+0.106
 superpower+0.106
Neg logit contributions
 bat-0.043
 se-0.042
 Ben-0.041
 put-0.039
 next-0.036
Token display
Token ID3359
Feature activation-5.48e-02

Pos logit contributions
 unlike+0.120
 superpower+0.105
eway+0.104
 perspective+0.103
 blonde+0.102
Neg logit contributions
 bat-0.058
 se-0.054
 Ben-0.054
 put-0.054
 say-0.052
Token them
Token ID606
Feature activation-5.67e-03

Pos logit contributions
 Ben+0.038
 bat+0.031
 se+0.029
 each+0.029
ila+0.028
Neg logit contributions
 unlike-0.148
 perspective-0.137
 matter-0.136
 superpower-0.136
 tactic-0.134
Token in
Token ID287
Feature activation+1.36e-02

Pos logit contributions
 Ben+0.005
 bat+0.005
 next+0.005
 each+0.004
 se+0.004
Neg logit contributions
 unlike-0.014
 superpower-0.013
 matter-0.013
 blonde-0.013
 perspective-0.013
Token.
Token ID13
Feature activation+7.71e-03

Pos logit contributions
 Ben+0.004
 bat+0.004
 se+0.004
ila+0.004
Ben+0.004
Neg logit contributions
 unlike-0.016
 matter-0.015
 superpower-0.015
 blonde-0.014
 perspective-0.014
Token
Token ID220
Feature activation+1.06e-01

Pos logit contributions
 unlike+0.015
 matter+0.014
 superpower+0.014
 perspective+0.014
 tactic+0.013
Neg logit contributions
 Ben-0.011
Ben-0.010
 bat-0.010
ila-0.010
 se-0.009
Token
Token ID198
Feature activation-3.39e-02

Pos logit contributions
 unlike+0.245
 superpower+0.223
 matter+0.220
 perspective+0.218
 tactic+0.218
Neg logit contributions
 bat-0.095
 se-0.095
 Ben-0.092
Ben-0.092
ila-0.091
Token
Token ID198
Feature activation-3.18e-02

Pos logit contributions
 Ben+0.028
Ben+0.028
 bat+0.027
ila+0.027
 se+0.027
Neg logit contributions
 unlike-0.082
 matter-0.075
 superpower-0.075
 blonde-0.073
 perspective-0.073
TokenHar
Token ID13587
Feature activation-8.90e-02

Pos logit contributions
Ben+0.041
 Ben+0.041
 bat+0.040
ila+0.040
 se+0.039
Neg logit contributions
 unlike-0.063
 matter-0.056
 superpower-0.056
 blonde-0.054
 perspective-0.054
Tokenper
Token ID525
Feature activation-2.50e-01

Pos logit contributions
 bat+0.131
 Ben+0.128
Ben+0.126
 se+0.126
ila+0.123
Neg logit contributions
 unlike-0.159
 matter-0.142
 superpower-0.141
 perspective-0.139
 tactic-0.137
Token's
Token ID338
Feature activation-4.11e-02

Pos logit contributions
 Ben+0.318
 bat+0.315
Ben+0.313
ila+0.311
 se+0.303
Neg logit contributions
 unlike-0.486
 matter-0.435
 superpower-0.434
 blonde-0.423
 perspective-0.421
Token mom
Token ID1995
Feature activation-4.97e-02

Pos logit contributions
 Ben+0.043
 bat+0.042
 se+0.042
Ben+0.039
 next+0.039
Neg logit contributions
 unlike-0.094
eway-0.085
 matter-0.085
 tactic-0.083
 superpower-0.082
Token put
Token ID1234
Feature activation-2.21e-01

Pos logit contributions
 bat+0.058
Ben+0.058
ila+0.057
 Ben+0.057
 se+0.054
Neg logit contributions
 unlike-0.101
 matter-0.092
 superpower-0.090
 blonde-0.089
 perspective-0.087
Token the
Token ID262
Feature activation-6.88e-02

Pos logit contributions
Ben+0.173
 Ben+0.173
ila+0.173
 bat+0.172
 se+0.166
Neg logit contributions
 unlike-0.538
 matter-0.493
 superpower-0.493
 blonde-0.482
 tactic-0.478
Token bowls
Token ID34796
Feature activation-3.14e-02

Pos logit contributions
 se+0.083
 bat+0.083
 Ben+0.081
Ben+0.078
ila+0.077
Neg logit contributions
 unlike-0.142
 perspective-0.128
 superpower-0.127
eway-0.127
 matter-0.127
Token
Token ID198
Feature activation-3.87e-02

Pos logit contributions
 bat+0.040
 se+0.040
 Ben+0.039
 goes+0.039
ila+0.039
Neg logit contributions
 unlike-0.082
 superpower-0.075
eway-0.074
 perspective-0.074
 tactic-0.074
TokenThe
Token ID464
Feature activation-4.36e-02

Pos logit contributions
 Ben+0.046
ila+0.045
 bat+0.045
 se+0.044
 goes+0.043
Neg logit contributions
 unlike-0.077
 superpower-0.070
 matter-0.069
 perspective-0.069
 blonde-0.068
Token fan
Token ID4336
Feature activation+5.40e-02

Pos logit contributions
 Ben+0.048
 bat+0.047
Ben+0.046
ila+0.046
 se+0.046
Neg logit contributions
 unlike-0.094
 matter-0.085
 superpower-0.085
 perspective-0.083
 blonde-0.083
Token started
Token ID2067
Feature activation+4.96e-02

Pos logit contributions
 unlike+0.113
 matter+0.103
 superpower+0.102
 perspective+0.100
 blonde+0.100
Neg logit contributions
 Ben-0.061
 bat-0.060
ila-0.060
Ben-0.060
 se-0.059
Token to
Token ID284
Feature activation+5.34e-02

Pos logit contributions
 unlike+0.117
 matter+0.110
 superpower+0.108
 perspective+0.107
 tactic+0.107
Neg logit contributions
 se-0.045
 bat-0.045
 Ben-0.043
 next-0.040
 put-0.039
Token blow
Token ID6611
Feature activation-2.49e-01

Pos logit contributions
 unlike+0.116
 matter+0.107
 superpower+0.105
 perspective+0.103
 blonde+0.103
Neg logit contributions
 Ben-0.056
ila-0.056
Ben-0.055
 bat-0.054
 se-0.053
Token gently
Token ID15165
Feature activation-1.95e-02

Pos logit contributions
 Ben+0.198
 bat+0.193
Ben+0.186
 se+0.186
ila+0.183
Neg logit contributions
 unlike-0.617
 superpower-0.568
 matter-0.564
 perspective-0.556
 blonde-0.554
Token and
Token ID290
Feature activation-9.68e-02

Pos logit contributions
ila+0.016
Ben+0.016
 Ben+0.016
 bat+0.015
 se+0.015
Neg logit contributions
 unlike-0.047
 matter-0.043
 superpower-0.043
 blonde-0.042
 tactic-0.042
Token soon
Token ID2582
Feature activation+2.47e-04

Pos logit contributions
 Ben+0.088
 bat+0.087
ila+0.086
Ben+0.086
 se+0.084
Neg logit contributions
 unlike-0.224
 matter-0.205
 superpower-0.205
 blonde-0.200
 perspective-0.200
Token Molly
Token ID30236
Feature activation-1.83e-02

Pos logit contributions
 unlike+0.001
 matter+0.001
 superpower+0.001
 blonde+0.001
 perspective+0.001
Neg logit contributions
 bat-0.000
Ben-0.000
ila-0.000
 se-0.000
 Ben-0.000
Token felt
Token ID2936
Feature activation-6.28e-03

Pos logit contributions
ila+0.021
Ben+0.020
 Ben+0.020
 bat+0.020
 se+0.019
Neg logit contributions
 unlike-0.038
 superpower-0.034
 blonde-0.033
 matter-0.033
 darkened-0.032
Token the
Token ID262
Feature activation-7.18e-02

Pos logit contributions
 Ben+0.131
ila+0.115
 bat+0.112
Ben+0.110
 each+0.109
Neg logit contributions
 unlike-0.299
 superpower-0.275
 matter-0.271
 tactic-0.269
 perspective-0.265
Token rare
Token ID4071
Feature activation+4.63e-02

Pos logit contributions
 next+0.089
 Ben+0.084
 se+0.083
 bat+0.082
 slide+0.082
Neg logit contributions
 unlike-0.152
 tactic-0.141
eway-0.135
 superpower-0.135
rum-0.132
Token rac
Token ID3444
Feature activation-7.66e-02

Pos logit contributions
 unlike+0.095
 superpower+0.085
 tactic+0.084
 alas+0.083
 matter+0.082
Neg logit contributions
 next-0.055
 slide-0.055
 Ben-0.055
 se-0.054
 bat-0.054
Tokencoon
Token ID20912
Feature activation+1.23e-02

Pos logit contributions
 Ben+0.135
 bat+0.135
 se+0.132
Ben+0.129
 says+0.129
Neg logit contributions
 unlike-0.110
eway-0.100
 superpower-0.096
 perspective-0.094
 matter-0.093
Token scam
Token ID19126
Feature activation-8.36e-02

Pos logit contributions
 unlike+0.028
 matter+0.025
 superpower+0.025
 perspective+0.025
 blonde+0.024
Neg logit contributions
Ben-0.012
ila-0.012
 Ben-0.012
 bat-0.012
 se-0.012
Tokenper
Token ID525
Feature activation-2.49e-01

Pos logit contributions
 se+0.131
Ben+0.131
 says+0.126
 bat+0.126
 dads+0.125
Neg logit contributions
 unlike-0.126
eway-0.124
 perspective-0.115
 terribly-0.114
 Tears-0.113
Token away
Token ID1497
Feature activation+1.64e-01

Pos logit contributions
 Ben+0.256
Ben+0.253
 bat+0.252
ila+0.252
 se+0.247
Neg logit contributions
 unlike-0.547
 matter-0.496
 superpower-0.495
 perspective-0.487
 blonde-0.482
Token.
Token ID13
Feature activation+8.64e-03

Pos logit contributions
 unlike+0.425
 superpower+0.399
 matter+0.399
 blonde+0.390
 tactic+0.388
Neg logit contributions
 Ben-0.108
 bat-0.102
 next-0.098
 se-0.097
ila-0.093
Token
Token ID198
Feature activation-3.29e-02

Pos logit contributions
 unlike+0.019
 matter+0.019
 superpower+0.018
 perspective+0.018
 tactic+0.018
Neg logit contributions
 Ben-0.010
Ben-0.009
ila-0.008
 bat-0.008
 says-0.008
Token
Token ID198
Feature activation-3.25e-02

Pos logit contributions
 se+0.035
 bat+0.034
 Ben+0.033
ila+0.033
Ben+0.032
Neg logit contributions
 unlike-0.069
 superpower-0.063
eway-0.063
 Tears-0.061
 blonde-0.061
TokenLuckily
Token ID42332
Feature activation+1.74e-02

Pos logit contributions
 Ben+0.037
 se+0.036
 bat+0.036
ila+0.034
 says+0.034
Neg logit contributions
 unlike-0.067
 superpower-0.061
eway-0.060
 blonde-0.060
 perspective-0.059
Token kids
Token ID3988
Feature activation-7.52e-02

Pos logit contributions
 Ben+0.078
 bat+0.077
ila+0.077
Ben+0.077
 se+0.075
Neg logit contributions
 unlike-0.147
 matter-0.134
 superpower-0.133
 perspective-0.130
 blonde-0.130
Token would
Token ID561
Feature activation+8.12e-02

Pos logit contributions
 Ben+0.087
Ben+0.087
ila+0.086
 bat+0.085
 se+0.082
Neg logit contributions
 unlike-0.156
 superpower-0.139
 matter-0.139
 perspective-0.137
 blonde-0.136
Token join
Token ID4654
Feature activation-1.06e-01

Pos logit contributions
 unlike+0.182
 matter+0.168
 superpower+0.166
 perspective+0.163
 blonde+0.161
Neg logit contributions
ila-0.079
 Ben-0.079
Ben-0.078
 bat-0.074
 se-0.073
Token together
Token ID1978
Feature activation+6.42e-02

Pos logit contributions
 bat+0.086
 Ben+0.085
Ben+0.085
ila+0.084
 se+0.084
Neg logit contributions
 unlike-0.255
 matter-0.234
 superpower-0.234
 perspective-0.229
 blonde-0.229
Token to
Token ID284
Feature activation+5.24e-02

Pos logit contributions
 unlike+0.157
 superpower+0.144
 matter+0.144
 perspective+0.141
 blonde+0.140
Neg logit contributions
ila-0.050
Ben-0.050
 Ben-0.049
 bat-0.049
 se-0.048
Token blow
Token ID6611
Feature activation-2.49e-01

Pos logit contributions
 unlike+0.118
 matter+0.112
 superpower+0.110
 perspective+0.108
 blonde+0.107
Neg logit contributions
ila-0.048
Ben-0.048
 Ben-0.046
 se-0.045
 bat-0.044
Token bubbles
Token ID25037
Feature activation-2.04e-02

Pos logit contributions
ila+0.264
Ben+0.264
 Ben+0.262
 bat+0.258
 se+0.255
Neg logit contributions
 unlike-0.531
 matter-0.483
 superpower-0.482
 tactic-0.469
 blonde-0.469
Token and
Token ID290
Feature activation-9.38e-02

Pos logit contributions
ila+0.022
Ben+0.022
 Ben+0.021
 dads+0.019
 se+0.019
Neg logit contributions
 unlike-0.044
 superpower-0.039
 matter-0.038
eway-0.038
 tactic-0.038
Token have
Token ID423
Feature activation-1.54e-01

Pos logit contributions
ila+0.090
Ben+0.088
 Ben+0.084
 bat+0.082
 se+0.081
Neg logit contributions
 unlike-0.213
 matter-0.197
 superpower-0.195
 blonde-0.190
 perspective-0.188
Token a
Token ID257
Feature activation-1.19e-02

Pos logit contributions
 Ben+0.183
 bat+0.181
 se+0.177
Ben+0.177
ila+0.175
Neg logit contributions
 unlike-0.317
 superpower-0.285
 matter-0.284
 perspective-0.278
 tactic-0.277
Token good
Token ID922
Feature activation-2.44e-02

Pos logit contributions
 bat+0.017
 se+0.016
 Ben+0.016
Ben+0.016
 slide+0.015
Neg logit contributions
 unlike-0.023
eway-0.020
 matter-0.019
 tactic-0.019
 perspective-0.019
Token Soon
Token ID15894
Feature activation+5.89e-02

Pos logit contributions
 unlike+0.026
 matter+0.024
 superpower+0.024
 perspective+0.023
 tactic+0.023
Neg logit contributions
 Ben-0.019
Ben-0.018
 bat-0.017
ila-0.017
 se-0.017
Token the
Token ID262
Feature activation-5.89e-02

Pos logit contributions
 unlike+0.135
 matter+0.123
 superpower+0.122
 blonde+0.119
 perspective+0.119
Neg logit contributions
 bat-0.055
Ben-0.054
ila-0.054
 se-0.053
 Ben-0.053
Token wind
Token ID2344
Feature activation-7.92e-02

Pos logit contributions
ila+0.064
Ben+0.062
 Ben+0.062
 bat+0.059
 says+0.055
Neg logit contributions
 unlike-0.126
 matter-0.116
 superpower-0.115
 perspective-0.113
 blonde-0.110
Token started
Token ID2067
Feature activation+4.98e-02

Pos logit contributions
 Ben+0.094
Ben+0.093
ila+0.092
 bat+0.090
 se+0.086
Neg logit contributions
 unlike-0.164
 matter-0.148
 superpower-0.145
 perspective-0.141
 darkened-0.140
Token to
Token ID284
Feature activation+5.10e-02

Pos logit contributions
 unlike+0.111
 superpower+0.104
 matter+0.103
 tactic+0.102
 blonde+0.101
Neg logit contributions
 se-0.049
 bat-0.049
 Ben-0.048
 next-0.044
ila-0.043
Token blow
Token ID6611
Feature activation-2.49e-01

Pos logit contributions
 unlike+0.111
 matter+0.109
 darkened+0.101
 superpower+0.099
 perspective+0.098
Neg logit contributions
ila-0.057
Ben-0.054
 Ben-0.052
 pige-0.048
 dads-0.048
Token and
Token ID290
Feature activation-9.08e-02

Pos logit contributions
 Ben+0.218
 bat+0.215
 se+0.209
Ben+0.207
ila+0.204
Neg logit contributions
 unlike-0.589
 superpower-0.541
 matter-0.537
 perspective-0.529
 blonde-0.528
Token the
Token ID262
Feature activation-6.67e-02

Pos logit contributions
ila+0.075
Ben+0.074
 Ben+0.070
 bat+0.069
 se+0.067
Neg logit contributions
 unlike-0.220
 matter-0.203
 superpower-0.200
 perspective-0.194
 blonde-0.194
Token waves
Token ID9813
Feature activation-1.70e-01

Pos logit contributions
ila+0.070
 Ben+0.068
Ben+0.068
 bat+0.064
 se+0.061
Neg logit contributions
 unlike-0.146
 matter-0.135
 superpower-0.133
 perspective-0.131
 blonde-0.129
Token started
Token ID2067
Feature activation+4.78e-02

Pos logit contributions
 Ben+0.204
Ben+0.203
ila+0.202
 bat+0.197
 se+0.191
Neg logit contributions
 unlike-0.346
 matter-0.312
 superpower-0.306
 blonde-0.299
 perspective-0.299
Token to
Token ID284
Feature activation+5.05e-02

Pos logit contributions
 unlike+0.104
 matter+0.095
 superpower+0.095
 perspective+0.092
 tactic+0.092
Neg logit contributions
 Ben-0.050
 bat-0.050
 se-0.050
ila-0.048
Ben-0.048
Token fast
Token ID3049
Feature activation-9.27e-02

Pos logit contributions
 unlike+0.191
 matter+0.176
 superpower+0.173
 perspective+0.171
 blonde+0.169
Neg logit contributions
ila-0.083
 Ben-0.083
 bat-0.082
 se-0.081
Ben-0.081
Token it
Token ID340
Feature activation-3.19e-02

Pos logit contributions
Ben+0.069
ila+0.068
 bat+0.065
 se+0.064
 Ben+0.064
Neg logit contributions
 unlike-0.233
 superpower-0.210
 matter-0.209
 perspective-0.208
 blonde-0.203
Token made
Token ID925
Feature activation-8.85e-02

Pos logit contributions
 Ben+0.040
Ben+0.039
ila+0.039
 bat+0.039
 se+0.038
Neg logit contributions
 unlike-0.064
 matter-0.057
 superpower-0.057
 perspective-0.055
 blonde-0.055
Token the
Token ID262
Feature activation-7.13e-02

Pos logit contributions
Ben+0.093
 bat+0.089
ila+0.088
 se+0.087
 dads+0.082
Neg logit contributions
 unlike-0.194
 matter-0.178
 superpower-0.171
eway-0.169
 perspective-0.166
Token wind
Token ID2344
Feature activation-8.09e-02

Pos logit contributions
ila+0.081
 Ben+0.081
Ben+0.081
 bat+0.079
 se+0.077
Neg logit contributions
 unlike-0.148
 matter-0.136
 superpower-0.134
 perspective-0.132
 blonde-0.130
Token blow
Token ID6611
Feature activation-2.48e-01

Pos logit contributions
Ben+0.093
 Ben+0.091
ila+0.090
 bat+0.081
 se+0.080
Neg logit contributions
 unlike-0.175
 matter-0.156
 superpower-0.153
 perspective-0.151
 tactic-0.147
Token.
Token ID13
Feature activation+1.06e-02

Pos logit contributions
 Ben+0.271
ila+0.270
Ben+0.270
 bat+0.263
 se+0.259
Neg logit contributions
 unlike-0.531
 matter-0.484
 superpower-0.476
 perspective-0.466
 tactic-0.462
Token
Token ID198
Feature activation-3.51e-02

Pos logit contributions
 unlike+0.020
 matter+0.018
 superpower+0.018
 perspective+0.017
 tactic+0.017
Neg logit contributions
 Ben-0.015
 bat-0.014
Ben-0.014
ila-0.014
 se-0.014
Token
Token ID198
Feature activation-3.43e-02

Pos logit contributions
 bat+0.031
 Ben+0.031
 se+0.031
ila+0.030
Ben+0.030
Neg logit contributions
 unlike-0.081
 superpower-0.074
 matter-0.073
 perspective-0.073
 tactic-0.072
TokenThe
Token ID464
Feature activation-4.27e-02

Pos logit contributions
 Ben+0.051
 bat+0.050
ila+0.050
 se+0.050
Ben+0.049
Neg logit contributions
 unlike-0.060
 matter-0.054
 superpower-0.053
 perspective-0.052
 blonde-0.052
Token car
Token ID1097
Feature activation+1.44e-02

Pos logit contributions
ila+0.054
 Ben+0.053
Ben+0.053
 says+0.048
 se+0.048
Neg logit contributions
 unlike-0.080
 matter-0.078
 superpower-0.076
 perspective-0.072
 blonde-0.072
Token,
Token ID11
Feature activation-1.44e-03

Pos logit contributions
 Ben+0.113
Ben+0.085
 they+0.084
ila+0.081
 bat+0.081
Neg logit contributions
 unlike-0.340
 perspective-0.318
 tactic-0.317
 superpower-0.313
eway-0.310
Token the
Token ID262
Feature activation-7.39e-02

Pos logit contributions
 Ben+0.001
 bat+0.001
Ben+0.001
ila+0.001
 se+0.001
Neg logit contributions
 unlike-0.004
 superpower-0.003
 perspective-0.003
 tactic-0.003
 matter-0.003
Token wind
Token ID2344
Feature activation-8.36e-02

Pos logit contributions
 Ben+0.076
ila+0.074
 bat+0.074
Ben+0.074
 se+0.072
Neg logit contributions
 unlike-0.163
 matter-0.148
 superpower-0.148
 blonde-0.145
 perspective-0.145
Token started
Token ID2067
Feature activation+4.56e-02

Pos logit contributions
 Ben+0.095
Ben+0.093
ila+0.093
 bat+0.092
 se+0.088
Neg logit contributions
 unlike-0.176
 matter-0.159
 superpower-0.157
 perspective-0.153
 blonde-0.152
Token to
Token ID284
Feature activation+4.90e-02

Pos logit contributions
 unlike+0.106
 superpower+0.101
 tactic+0.100
 matter+0.100
eway+0.100
Neg logit contributions
 se-0.042
 bat-0.042
 Ben-0.038
 next-0.037
 put-0.035
Token blow
Token ID6611
Feature activation-2.48e-01

Pos logit contributions
 unlike+0.107
 matter+0.106
 darkened+0.097
 superpower+0.096
 tactic+0.094
Neg logit contributions
ila-0.050
Ben-0.049
 Ben-0.048
 pige-0.046
 dads-0.045
Token harder
Token ID7069
Feature activation+3.74e-02

Pos logit contributions
 Ben+0.219
 bat+0.217
 se+0.208
Ben+0.206
ila+0.200
Neg logit contributions
 unlike-0.588
 superpower-0.547
 perspective-0.534
 matter-0.534
 blonde-0.528
Token and
Token ID290
Feature activation-9.79e-02

Pos logit contributions
 unlike+0.093
 matter+0.086
 perspective+0.084
 superpower+0.084
 tactic+0.083
Neg logit contributions
ila-0.029
Ben-0.029
 Ben-0.027
 bat-0.026
 se-0.024
Token harder
Token ID7069
Feature activation+3.81e-02

Pos logit contributions
 Ben+0.114
 bat+0.111
 se+0.109
Ben+0.109
ila+0.109
Neg logit contributions
 unlike-0.206
 superpower-0.187
 matter-0.184
 perspective-0.183
 blonde-0.183
Token.
Token ID13
Feature activation+3.97e-03

Pos logit contributions
 unlike+0.088
 matter+0.080
 superpower+0.080
 perspective+0.079
 blonde+0.078
Neg logit contributions
Ben-0.035
ila-0.035
 Ben-0.035
 bat-0.034
 se-0.033
Token Suddenly
Token ID24975
Feature activation+1.73e-01

Pos logit contributions
 unlike+0.009
 matter+0.009
 perspective+0.008
 superpower+0.008
 tactic+0.008
Neg logit contributions
 Ben-0.005
Ben-0.004
 bat-0.004
ila-0.004
 Tom-0.004
Token,
Token ID11
Feature activation+1.05e-03

Pos logit contributions
 unlike+0.522
 superpower+0.485
 matter+0.485
 blonde+0.475
 perspective+0.475
Neg logit contributions
 Ben-0.076
 bat-0.073
Ben-0.071
ila-0.071
 se-0.069
Token the
Token ID262
Feature activation-6.36e-02

Pos logit contributions
 unlike+0.002
 matter+0.002
 blonde+0.002
 superpower+0.002
 perspective+0.002
Neg logit contributions
ila-0.001
Ben-0.001
 bat-0.001
 Ben-0.001
 se-0.001
Token wind
Token ID2344
Feature activation-8.07e-02

Pos logit contributions
 bat+0.071
 Ben+0.071
 se+0.071
Ben+0.068
 next+0.068
Neg logit contributions
 unlike-0.136
 tactic-0.123
 superpower-0.122
eway-0.120
 matter-0.120
Token started
Token ID2067
Feature activation+4.93e-02

Pos logit contributions
 Ben+0.101
Ben+0.099
ila+0.099
 bat+0.097
 se+0.093
Neg logit contributions
 unlike-0.163
 matter-0.144
 superpower-0.143
 perspective-0.141
 blonde-0.139
Token to
Token ID284
Feature activation+5.10e-02

Pos logit contributions
 unlike+0.113
 superpower+0.108
 matter+0.107
 tactic+0.107
 blonde+0.105
Neg logit contributions
 se-0.045
 bat-0.043
 Ben-0.042
 next-0.041
 each-0.039
Token blow
Token ID6611
Feature activation-2.48e-01

Pos logit contributions
 unlike+0.109
 matter+0.102
 superpower+0.098
 perspective+0.096
 tactic+0.095
Neg logit contributions
 Ben-0.057
ila-0.057
Ben-0.056
 bat-0.053
 se-0.051
Token,
Token ID11
Feature activation+6.82e-04

Pos logit contributions
 bat+0.219
 Ben+0.218
 se+0.215
 next+0.199
Ben+0.197
Neg logit contributions
 unlike-0.580
 superpower-0.545
 perspective-0.527
 tactic-0.527
 matter-0.527
Token and
Token ID290
Feature activation-9.31e-02

Pos logit contributions
 unlike+0.002
 matter+0.002
 superpower+0.002
 blonde+0.001
 perspective+0.001
Neg logit contributions
ila-0.001
Ben-0.001
 Ben-0.001
 bat-0.001
 se-0.000
Token it
Token ID340
Feature activation-3.45e-02

Pos logit contributions
 Ben+0.057
 bat+0.054
 se+0.052
ila+0.052
Ben+0.052
Neg logit contributions
 unlike-0.246
 superpower-0.228
 matter-0.227
 perspective-0.224
 tactic-0.223
Token started
Token ID2067
Feature activation+4.96e-02

Pos logit contributions
 Ben+0.043
ila+0.043
Ben+0.043
 bat+0.042
 se+0.041
Neg logit contributions
 unlike-0.069
 matter-0.062
 superpower-0.061
 blonde-0.059
 perspective-0.059
Token to
Token ID284
Feature activation+5.20e-02

Pos logit contributions
 unlike+0.120
 tactic+0.116
 matter+0.115
 superpower+0.114
eway+0.112
Neg logit contributions
 se-0.041
 bat-0.040
 next-0.037
 Ben-0.036
 put-0.035
Token.
Token ID13
Feature activation+1.42e-03

Pos logit contributions
 unlike+0.019
 superpower+0.018
 matter+0.017
 tactic+0.017
 blonde+0.017
Neg logit contributions
 bat-0.010
 se-0.010
 Ben-0.010
Ben-0.009
 slide-0.009
Token The
Token ID383
Feature activation-2.35e-02

Pos logit contributions
 unlike+0.003
 matter+0.003
 superpower+0.003
 perspective+0.003
 tactic+0.003
Neg logit contributions
 Ben-0.002
 bat-0.001
Ben-0.001
ila-0.001
 se-0.001
Token wind
Token ID2344
Feature activation-7.82e-02

Pos logit contributions
 Ben+0.025
 se+0.025
 bat+0.025
Ben+0.024
 next+0.024
Neg logit contributions
 unlike-0.052
 superpower-0.047
 tactic-0.046
 matter-0.046
eway-0.046
Token started
Token ID2067
Feature activation+4.82e-02

Pos logit contributions
 Ben+0.096
Ben+0.094
 bat+0.092
ila+0.092
 se+0.089
Neg logit contributions
 unlike-0.159
 matter-0.143
 superpower-0.141
 perspective-0.136
 blonde-0.136
Token to
Token ID284
Feature activation+5.14e-02

Pos logit contributions
 unlike+0.114
 superpower+0.110
 matter+0.109
 tactic+0.109
 blonde+0.108
Neg logit contributions
 se-0.041
 bat-0.039
 next-0.037
 Ben-0.036
 put-0.035
Token blow
Token ID6611
Feature activation-2.47e-01

Pos logit contributions
 unlike+0.104
 matter+0.098
 superpower+0.092
 perspective+0.091
 tactic+0.091
Neg logit contributions
ila-0.063
Ben-0.062
 Ben-0.062
 bat-0.057
 dads-0.056
Token hard
Token ID1327
Feature activation+1.61e-02

Pos logit contributions
 bat+0.178
 Ben+0.175
 se+0.169
 next+0.156
Ben+0.155
Neg logit contributions
 unlike-0.627
 superpower-0.590
 perspective-0.577
 matter-0.571
 blonde-0.570
Token and
Token ID290
Feature activation-9.90e-02

Pos logit contributions
 unlike+0.043
 superpower+0.040
 matter+0.039
 blonde+0.039
 perspective+0.039
Neg logit contributions
 Ben-0.010
 bat-0.009
 se-0.009
Ben-0.008
ila-0.008
Token it
Token ID340
Feature activation-3.70e-02

Pos logit contributions
 Ben+0.077
 bat+0.074
 se+0.071
Ben+0.069
ila+0.068
Neg logit contributions
 unlike-0.247
 superpower-0.229
 perspective-0.226
 matter-0.226
 tactic-0.224
Token began
Token ID2540
Feature activation+8.10e-02

Pos logit contributions
 bat+0.042
 Ben+0.042
 se+0.042
Ben+0.041
ila+0.041
Neg logit contributions
 unlike-0.077
 superpower-0.070
 matter-0.069
 perspective-0.069
 blonde-0.068
Token to
Token ID284
Feature activation+5.15e-02

Pos logit contributions
 unlike+0.200
 matter+0.191
 tactic+0.189
 superpower+0.189
 blonde+0.187
Neg logit contributions
 se-0.065
 bat-0.063
 next-0.058
 Ben-0.056
 put-0.053
Token,
Token ID11
Feature activation-5.53e-04

Pos logit contributions
 Ben+0.076
 bat+0.073
 se+0.072
ila+0.072
Ben+0.071
Neg logit contributions
 unlike-0.188
 matter-0.175
 superpower-0.172
 perspective-0.170
 tactic-0.169
Token "
Token ID366
Feature activation+6.06e-02

Pos logit contributions
 Ben+0.001
ila+0.001
 bat+0.001
Ben+0.001
 se+0.001
Neg logit contributions
 unlike-0.001
 superpower-0.001
 matter-0.001
 blonde-0.001
 perspective-0.001
TokenYou
Token ID1639
Feature activation-3.22e-02

Pos logit contributions
 unlike+0.120
 matter+0.110
 superpower+0.108
 perspective+0.106
 blonde+0.105
Neg logit contributions
 Ben-0.076
 bat-0.075
ila-0.074
 se-0.074
Ben-0.071
Token have
Token ID423
Feature activation-1.50e-01

Pos logit contributions
ila+0.036
 Ben+0.036
Ben+0.036
 bat+0.035
 se+0.035
Neg logit contributions
 unlike-0.067
 matter-0.062
 superpower-0.061
 blonde-0.059
 perspective-0.059
Token to
Token ID284
Feature activation+5.39e-02

Pos logit contributions
ila+0.174
Ben+0.170
 Ben+0.170
 bat+0.166
 se+0.163
Neg logit contributions
 unlike-0.313
 matter-0.284
 superpower-0.284
 blonde-0.275
 perspective-0.274
Token blow
Token ID6611
Feature activation-2.47e-01

Pos logit contributions
 matter+0.109
 superpower+0.102
 unlike+0.098
 former+0.097
 perspective+0.095
Neg logit contributions
ila-0.066
 pige-0.063
 says-0.057
Ben-0.057
 Ben-0.056
Token on
Token ID319
Feature activation+8.99e-03

Pos logit contributions
ila+0.252
Ben+0.244
 Ben+0.237
 se+0.234
 bat+0.229
Neg logit contributions
 unlike-0.539
 matter-0.498
 superpower-0.494
 tactic-0.475
 perspective-0.474
Token the
Token ID262
Feature activation-7.78e-02

Pos logit contributions
 unlike+0.022
 matter+0.021
 superpower+0.021
 perspective+0.020
 blonde+0.020
Neg logit contributions
 Ben-0.007
ila-0.007
 bat-0.007
Ben-0.006
 se-0.006
Token sail
Token ID14936
Feature activation-1.53e-01

Pos logit contributions
ila+0.109
Ben+0.097
 Ben+0.093
 says+0.087
 dads+0.083
Neg logit contributions
 unlike-0.150
 matter-0.141
 superpower-0.139
 perspective-0.136
 former-0.132
Token and
Token ID290
Feature activation-9.95e-02

Pos logit contributions
ila+0.163
Ben+0.162
 Ben+0.158
 se+0.156
 bat+0.151
Neg logit contributions
 unlike-0.327
 matter-0.300
 superpower-0.295
 perspective-0.294
 tactic-0.289
Token steer
Token ID27401
Feature activation-1.64e-01

Pos logit contributions
ila+0.103
Ben+0.094
 Ben+0.085
 pige+0.080
 se+0.080
Neg logit contributions
 unlike-0.223
 matter-0.211
 superpower-0.205
 perspective-0.196
 former-0.196
Token
Token ID198
Feature activation-3.92e-02

Pos logit contributions
 se+0.043
 bat+0.042
 Ben+0.041
ila+0.040
 goes+0.040
Neg logit contributions
 unlike-0.082
 superpower-0.076
 perspective-0.075
eway-0.075
 tactic-0.073
TokenThe
Token ID464
Feature activation-4.37e-02

Pos logit contributions
 Ben+0.050
 se+0.048
 bat+0.048
ila+0.047
 says+0.047
Neg logit contributions
 unlike-0.076
 perspective-0.069
 superpower-0.068
 matter-0.068
eway-0.067
Token wind
Token ID2344
Feature activation-8.01e-02

Pos logit contributions
 Ben+0.064
 bat+0.063
ila+0.063
Ben+0.063
 se+0.062
Neg logit contributions
 unlike-0.078
 matter-0.069
 superpower-0.069
 perspective-0.067
 blonde-0.067
Token began
Token ID2540
Feature activation+8.39e-02

Pos logit contributions
 Ben+0.100
Ben+0.098
ila+0.096
 bat+0.095
 se+0.089
Neg logit contributions
 unlike-0.161
 matter-0.145
 superpower-0.143
 perspective-0.138
 darkened-0.137
Token to
Token ID284
Feature activation+5.52e-02

Pos logit contributions
 unlike+0.188
 matter+0.172
 superpower+0.172
 perspective+0.168
 tactic+0.168
Neg logit contributions
 Ben-0.083
 bat-0.082
 se-0.080
ila-0.079
Ben-0.078
Token blow
Token ID6611
Feature activation-2.47e-01

Pos logit contributions
 matter+0.118
 unlike+0.118
 superpower+0.109
 darkened+0.107
 perspective+0.107
Neg logit contributions
ila-0.061
 Ben-0.058
Ben-0.057
 pige-0.053
 dads-0.051
Token harder
Token ID7069
Feature activation+4.08e-02

Pos logit contributions
 Ben+0.192
 bat+0.186
 se+0.180
Ben+0.175
ila+0.167
Neg logit contributions
 unlike-0.612
 superpower-0.567
 matter-0.556
 perspective-0.555
 tactic-0.549
Token and
Token ID290
Feature activation-9.75e-02

Pos logit contributions
 unlike+0.106
 matter+0.098
 superpower+0.097
 perspective+0.096
 blonde+0.095
Neg logit contributions
 Ben-0.026
ila-0.026
Ben-0.026
 bat-0.026
 se-0.024
Token the
Token ID262
Feature activation-7.17e-02

Pos logit contributions
 Ben+0.126
 bat+0.124
Ben+0.123
ila+0.123
 se+0.122
Neg logit contributions
 unlike-0.191
 superpower-0.171
 matter-0.170
 perspective-0.166
 blonde-0.166
Token child
Token ID1200
Feature activation+5.19e-02

Pos logit contributions
 Ben+0.096
ila+0.096
Ben+0.095
 bat+0.095
 se+0.092
Neg logit contributions
 unlike-0.135
 matter-0.122
 superpower-0.121
 perspective-0.118
 blonde-0.118
Token laughed
Token ID13818
Feature activation-5.32e-02

Pos logit contributions
 unlike+0.104
 matter+0.094
 superpower+0.093
 blonde+0.091
 perspective+0.090
Neg logit contributions
ila-0.062
 Ben-0.062
Ben-0.062
 bat-0.061
 se-0.058
Token sounds
Token ID5238
Feature activation+2.62e-02

Pos logit contributions
 unlike+0.062
 superpower+0.055
 matter+0.055
 perspective+0.054
 blonde+0.053
Neg logit contributions
 Ben-0.045
 bat-0.045
Ben-0.044
ila-0.044
 se-0.044
Token.
Token ID13
Feature activation+8.04e-03

Pos logit contributions
 unlike+0.063
 matter+0.056
 superpower+0.056
 blonde+0.055
 perspective+0.055
Neg logit contributions
ila-0.022
Ben-0.022
 Ben-0.022
 bat-0.022
 se-0.021
Token He
Token ID679
Feature activation+9.58e-02

Pos logit contributions
 unlike+0.014
 blonde+0.012
 superpower+0.012
 matter+0.012
 Tears+0.012
Neg logit contributions
 se-0.012
 bat-0.011
ila-0.011
Ben-0.011
 next-0.011
Token likes
Token ID7832
Feature activation-1.64e-02

Pos logit contributions
 unlike+0.192
 matter+0.177
 superpower+0.177
 blonde+0.174
 tactic+0.160
Neg logit contributions
ila-0.104
Ben-0.104
line-0.103
 bat-0.100
 next-0.097
Token to
Token ID284
Feature activation+4.90e-02

Pos logit contributions
ila+0.017
Ben+0.017
 bat+0.017
 Ben+0.017
 se+0.016
Neg logit contributions
 unlike-0.035
 matter-0.032
 superpower-0.032
 perspective-0.031
 blonde-0.031
Token blow
Token ID6611
Feature activation-2.46e-01

Pos logit contributions
 unlike+0.105
 matter+0.102
 superpower+0.099
 blonde+0.097
 tactic+0.095
Neg logit contributions
Ben-0.048
ila-0.048
 Ben-0.047
 pige-0.044
 next-0.043
Token the
Token ID262
Feature activation-8.29e-02

Pos logit contributions
ila+0.238
Ben+0.236
 bat+0.226
 Ben+0.225
 se+0.225
Neg logit contributions
 unlike-0.546
 matter-0.502
 superpower-0.495
 tactic-0.486
 blonde-0.482
Token trumpet
Token ID40334
Feature activation+8.28e-02

Pos logit contributions
 Ben+0.135
 bat+0.135
 se+0.133
Ben+0.132
ila+0.132
Neg logit contributions
 unlike-0.135
 matter-0.117
 superpower-0.117
 perspective-0.114
 blonde-0.113
Token in
Token ID287
Feature activation+8.76e-03

Pos logit contributions
 unlike+0.194
 matter+0.177
 superpower+0.176
 tactic+0.172
 blonde+0.170
Neg logit contributions
ila-0.074
Ben-0.073
 Ben-0.073
 bat-0.070
 se-0.069
Token the
Token ID262
Feature activation-8.25e-02

Pos logit contributions
 unlike+0.020
 matter+0.018
 superpower+0.018
 blonde+0.018
 perspective+0.018
Neg logit contributions
ila-0.008
 Ben-0.008
 bat-0.008
Ben-0.008
 se-0.008
Token morning
Token ID3329
Feature activation+6.03e-02

Pos logit contributions
ila+0.097
Ben+0.093
 Ben+0.092
 se+0.088
 bat+0.087
Neg logit contributions
 unlike-0.163
 superpower-0.153
 matter-0.152
 blonde-0.151
 tactic-0.145
Token one
Token ID530
Feature activation+2.42e-02

Pos logit contributions
 unlike+0.024
 Terr+0.022
eway+0.022
 tactic+0.021
 superpower+0.021
Neg logit contributions
 Ben-0.011
 bat-0.010
 next-0.010
 se-0.010
 put-0.010
Token could
Token ID714
Feature activation+3.70e-02

Pos logit contributions
 unlike+0.050
 matter+0.046
 superpower+0.045
 perspective+0.045
 blonde+0.045
Neg logit contributions
 Ben-0.028
 bat-0.028
ila-0.028
 se-0.028
Ben-0.027
Token open
Token ID1280
Feature activation-6.53e-02

Pos logit contributions
 unlike+0.090
 matter+0.083
 superpower+0.083
 perspective+0.081
 blonde+0.081
Neg logit contributions
 bat-0.030
 Ben-0.029
ila-0.029
 se-0.029
Ben-0.028
Token.
Token ID13
Feature activation+2.13e-04

Pos logit contributions
 Ben+0.046
 bat+0.041
 next+0.040
 se+0.039
 put+0.039
Neg logit contributions
 unlike-0.167
 superpower-0.156
 tactic-0.153
 matter-0.152
 blonde-0.152
Token No
Token ID1400
Feature activation+5.97e-02

Pos logit contributions
 unlike+0.000
 matter+0.000
 perspective+0.000
 superpower+0.000
 tactic+0.000
Neg logit contributions
 Ben-0.000
 bat-0.000
Ben-0.000
ila-0.000
 put-0.000
Token matter
Token ID2300
Feature activation+2.83e-01

Pos logit contributions
 unlike+0.134
 Terr+0.125
eway+0.124
 former+0.121
 Tears+0.118
Neg logit contributions
 put-0.063
 Ben-0.063
 next-0.060
 bat-0.060
 se-0.059
Token how
Token ID703
Feature activation+2.77e-02

Pos logit contributions
 unlike+0.595
 superpower+0.537
 blonde+0.535
 matter+0.523
 tactic+0.520
Neg logit contributions
 Ben-0.339
ila-0.322
 bat-0.318
 next-0.304
 each-0.303
Token hard
Token ID1327
Feature activation+1.99e-02

Pos logit contributions
 unlike+0.058
 matter+0.053
 superpower+0.053
 perspective+0.052
 blonde+0.051
Neg logit contributions
Ben-0.031
 Ben-0.031
 bat-0.031
ila-0.031
 se-0.030
Token they
Token ID484
Feature activation-1.22e-01

Pos logit contributions
 unlike+0.042
 matter+0.039
 superpower+0.038
 blonde+0.037
 perspective+0.036
Neg logit contributions
ila-0.023
Ben-0.023
 se-0.022
 bat-0.021
 Ben-0.021
Token tried
Token ID3088
Feature activation+3.77e-03

Pos logit contributions
 Ben+0.154
 bat+0.152
ila+0.151
Ben+0.151
 se+0.150
Neg logit contributions
 unlike-0.241
 superpower-0.216
 matter-0.215
 perspective-0.210
 blonde-0.210
Token they
Token ID484
Feature activation-1.22e-01

Pos logit contributions
 unlike+0.009
 superpower+0.009
 matter+0.009
 blonde+0.009
 tactic+0.009
Neg logit contributions
 Ben-0.003
 bat-0.003
 each-0.002
 next-0.002
 they-0.002
Token tree
Token ID5509
Feature activation-2.64e-02

Pos logit contributions
ila+0.059
Ben+0.054
 Ben+0.052
 dads+0.051
 next+0.049
Neg logit contributions
 unlike-0.080
 matter-0.074
 perspective-0.073
 superpower-0.072
 blonde-0.071
Token.
Token ID13
Feature activation+1.34e-02

Pos logit contributions
Ben+0.031
ila+0.031
 Ben+0.030
 dads+0.029
 se+0.027
Neg logit contributions
 unlike-0.055
 perspective-0.050
 matter-0.049
 tactic-0.048
 Tears-0.047
Token
Token ID198
Feature activation-2.91e-02

Pos logit contributions
 unlike+0.026
 matter+0.024
 superpower+0.024
 perspective+0.023
 blonde+0.023
Neg logit contributions
 Ben-0.018
 bat-0.017
Ben-0.017
ila-0.017
 se-0.016
Token
Token ID198
Feature activation-2.87e-02

Pos logit contributions
 se+0.033
 bat+0.032
 Ben+0.031
ila+0.031
Ben+0.030
Neg logit contributions
 unlike-0.058
 tactic-0.053
eway-0.053
 perspective-0.052
 superpower-0.052
Token"
Token ID1
Feature activation-7.34e-02

Pos logit contributions
 Ben+0.047
 se+0.047
ila+0.046
 bat+0.046
Ben+0.045
Neg logit contributions
 unlike-0.046
 matter-0.040
 superpower-0.039
eway-0.039
 blonde-0.038
TokenWow
Token ID22017
Feature activation+1.80e-01

Pos logit contributions
Ben+0.105
 Ben+0.103
 bat+0.102
 se+0.099
ila+0.099
Neg logit contributions
 unlike-0.137
 superpower-0.124
 matter-0.121
 tactic-0.120
 perspective-0.119
Token,
Token ID11
Feature activation+7.27e-03

Pos logit contributions
 unlike+0.390
 matter+0.353
 superpower+0.351
 perspective+0.345
 blonde+0.343
Neg logit contributions
 bat-0.187
 se-0.187
Ben-0.185
ila-0.184
 Ben-0.177
Token this
Token ID428
Feature activation-6.79e-02

Pos logit contributions
 unlike+0.016
 superpower+0.015
 perspective+0.015
 blonde+0.014
 matter+0.014
Neg logit contributions
 Ben-0.008
 bat-0.008
ila-0.007
Ben-0.007
 se-0.007
Token is
Token ID318
Feature activation-1.56e-02

Pos logit contributions
 Ben+0.065
Ben+0.065
ila+0.064
 bat+0.064
 se+0.063
Neg logit contributions
 unlike-0.154
 matter-0.141
 superpower-0.141
 perspective-0.137
 tactic-0.136
Token a
Token ID257
Feature activation-3.53e-03

Pos logit contributions
 Ben+0.017
 bat+0.015
 next+0.014
 put+0.014
 we+0.014
Neg logit contributions
 unlike-0.034
 perspective-0.034
 superpower-0.033
 tactic-0.033
 blonde-0.033
Token nice
Token ID3621
Feature activation+1.98e-02

Pos logit contributions
 bat+0.004
 se+0.004
 Ben+0.004
Ben+0.004
 slide+0.004
Neg logit contributions
 unlike-0.008
 tactic-0.007
 blonde-0.007
eway-0.007
 superpower-0.007
Token him
Token ID683
Feature activation+4.24e-02

Pos logit contributions
 Ben+0.023
 bat+0.022
Ben+0.022
ila+0.022
 se+0.022
Neg logit contributions
 unlike-0.033
 superpower-0.030
 matter-0.029
 blonde-0.029
 perspective-0.029
Token out
Token ID503
Feature activation+1.03e-01

Pos logit contributions
 unlike+0.100
 matter+0.092
 superpower+0.092
 perspective+0.089
 blonde+0.089
Neg logit contributions
 Ben-0.037
Ben-0.036
ila-0.036
 bat-0.036
 se-0.035
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
 unlike+0.253
 matter+0.232
 superpower+0.232
 blonde+0.227
 perspective+0.227
Neg logit contributions
 Ben-0.081
 bat-0.079
ila-0.077
 se-0.077
Ben-0.076
Token<|endoftext|>
Token ID50256
Feature activation-1.14e-02

Pos logit contributions
 unlike+0.023
 superpower+0.021
 perspective+0.020
 matter+0.020
 Tears+0.020
Neg logit contributions
 se-0.012
 bat-0.012
ila-0.012
Ben-0.011
 Ben-0.011
TokenOnce
Token ID7454
Feature activation+9.64e-02

Pos logit contributions
 se+0.012
 bat+0.012
ila+0.012
Ben+0.012
 Ben+0.012
Neg logit contributions
 unlike-0.024
 superpower-0.022
eway-0.021
 perspective-0.021
 matter-0.021
Token upon
Token ID2402
Feature activation+1.92e-01

Pos logit contributions
 unlike+0.111
 matter+0.091
 superpower+0.091
 perspective+0.086
 blonde+0.086
Neg logit contributions
 Ben-0.201
 bat-0.199
ila-0.199
Ben-0.199
 se-0.197
Token a
Token ID257
Feature activation-7.63e-03

Pos logit contributions
 unlike+0.441
 matter+0.403
 superpower+0.401
 blonde+0.393
 perspective+0.392
Neg logit contributions
 Ben-0.176
 bat-0.173
ila-0.173
Ben-0.172
 se-0.169
Token time
Token ID640
Feature activation+7.29e-02

Pos logit contributions
 se+0.014
ila+0.014
 Ben+0.014
 bat+0.014
line+0.013
Neg logit contributions
 unlike-0.010
 matter-0.009
 blonde-0.009
 superpower-0.009
 perspective-0.009
Token there
Token ID612
Feature activation+6.02e-02

Pos logit contributions
 unlike+0.177
 matter+0.164
 superpower+0.163
 perspective+0.160
 blonde+0.159
Neg logit contributions
 Ben-0.061
ila-0.058
 bat-0.058
Ben-0.057
 se-0.055
Token was
Token ID373
Feature activation+7.45e-02

Pos logit contributions
 unlike+0.135
 superpower+0.126
 former+0.125
 matter+0.123
 perspective+0.122
Neg logit contributions
 Ben-0.066
ila-0.062
Ben-0.061
 bat-0.061
 se-0.058
Token a
Token ID257
Feature activation-6.68e-03

Pos logit contributions
 unlike+0.164
 matter+0.147
 blonde+0.146
 superpower+0.146
 perspective+0.141
Neg logit contributions
ila-0.076
 bat-0.074
Ben-0.074
 se-0.074
 Ben-0.073
Token friends
Token ID2460
Feature activation+2.77e-02

Pos logit contributions
 unlike+0.170
 superpower+0.153
 matter+0.152
 perspective+0.149
 blonde+0.148
Neg logit contributions
 Ben-0.093
Ben-0.092
ila-0.092
 bat-0.092
 se-0.090
Token forever
Token ID8097
Feature activation+1.67e-01

Pos logit contributions
 unlike+0.064
 superpower+0.059
 matter+0.059
 perspective+0.057
 blonde+0.057
Neg logit contributions
 Ben-0.025
 bat-0.025
ila-0.025
Ben-0.025
 se-0.024
Token.
Token ID13
Feature activation+2.32e-02

Pos logit contributions
 unlike+0.357
 perspective+0.325
 superpower+0.318
 matter+0.318
 tactic+0.303
Neg logit contributions
ila-0.194
Ben-0.190
 Ben-0.181
 bat-0.171
 se-0.170
Token<|endoftext|>
Token ID50256
Feature activation+5.10e-03

Pos logit contributions
 unlike+0.046
 Tears+0.041
 superpower+0.041
 perspective+0.040
 matter+0.039
Neg logit contributions
 se-0.029
ila-0.028
 bat-0.027
Ben-0.027
 dove-0.026
TokenOnce
Token ID7454
Feature activation+1.12e-01

Pos logit contributions
 unlike+0.010
eway+0.009
 Tears+0.009
 superpower+0.009
 tactic+0.009
Neg logit contributions
 se-0.006
 dads-0.006
ila-0.006
 bat-0.006
Ben-0.006
Token upon
Token ID2402
Feature activation+2.08e-01

Pos logit contributions
 unlike+0.127
 matter+0.105
 superpower+0.104
 perspective+0.101
 tactic+0.099
Neg logit contributions
 bat-0.233
 Ben-0.231
Ben-0.231
 se-0.230
ila-0.229
Token a
Token ID257
Feature activation+2.33e-03

Pos logit contributions
 unlike+0.453
 matter+0.413
 superpower+0.408
 blonde+0.404
 tactic+0.398
Neg logit contributions
ila-0.211
 Ben-0.206
Ben-0.203
 se-0.203
 bat-0.202
Token time
Token ID640
Feature activation+8.68e-02

Pos logit contributions
 unlike+0.003
 matter+0.003
 blonde+0.003
 superpower+0.003
 perspective+0.002
Neg logit contributions
 se-0.004
ila-0.004
 Ben-0.004
line-0.004
 bat-0.004
Token,
Token ID11
Feature activation+1.36e-02

Pos logit contributions
 unlike+0.197
 superpower+0.179
 matter+0.178
 blonde+0.175
 perspective+0.174
Neg logit contributions
 bat-0.082
 Ben-0.081
Ben-0.081
 se-0.081
ila-0.081
Token there
Token ID612
Feature activation+7.23e-02

Pos logit contributions
 unlike+0.033
 superpower+0.031
 perspective+0.030
 former+0.030
 matter+0.030
Neg logit contributions
 Ben-0.012
Ben-0.011
 bat-0.011
ila-0.011
 next-0.011
Token was
Token ID373
Feature activation+8.89e-02

Pos logit contributions
 unlike+0.157
 superpower+0.144
 matter+0.141
 former+0.141
 blonde+0.140
Neg logit contributions
 Ben-0.084
 bat-0.079
Ben-0.078
ila-0.077
 se-0.075
Token cake
Token ID12187
Feature activation+1.32e-02

Pos logit contributions
 se+0.069
 Ben+0.067
 bat+0.065
 slide+0.064
 next+0.063
Neg logit contributions
 unlike-0.150
eway-0.139
 superpower-0.134
 perspective-0.133
 matter-0.133
Token.
Token ID13
Feature activation+1.07e-02

Pos logit contributions
 unlike+0.029
 matter+0.026
 superpower+0.026
 blonde+0.025
 perspective+0.025
Neg logit contributions
Ben-0.013
ila-0.013
 bat-0.013
 Ben-0.012
 se-0.012
Token
Token ID198
Feature activation-3.09e-02

Pos logit contributions
 unlike+0.025
 matter+0.023
 superpower+0.023
 perspective+0.022
 tactic+0.022
Neg logit contributions
 Ben-0.010
Ben-0.010
 bat-0.010
ila-0.010
 se-0.009
Token
Token ID198
Feature activation-3.07e-02

Pos logit contributions
 se+0.046
 wave+0.044
 dads+0.043
 pige+0.043
 bat+0.043
Neg logit contributions
 unlike-0.045
eway-0.044
 Tears-0.041
 Terr-0.040
 Where-0.040
Token"
Token ID1
Feature activation-7.87e-02

Pos logit contributions
 se+0.040
 wave+0.039
 waves+0.039
 dove+0.038
 pige+0.038
Neg logit contributions
 unlike-0.045
Andy-0.044
Fear-0.043
 total-0.043
 slightly-0.043
TokenOh
Token ID5812
Feature activation+2.03e-01

Pos logit contributions
 Ben+0.144
 bat+0.143
ila+0.140
 goes+0.135
 says+0.134
Neg logit contributions
 unlike-0.102
 matter-0.095
 blonde-0.089
 superpower-0.087
 perspective-0.083
Token no
Token ID645
Feature activation+2.57e-02

Pos logit contributions
 unlike+0.483
 superpower+0.440
 perspective+0.436
 tactic+0.434
 matter+0.433
Neg logit contributions
 Ben-0.208
ila-0.188
 bat-0.182
Ben-0.180
 se-0.178
Token!"
Token ID2474
Feature activation-6.65e-02

Pos logit contributions
 unlike+0.061
 superpower+0.056
 matter+0.055
 blonde+0.054
 tactic+0.054
Neg logit contributions
 Ben-0.023
 bat-0.022
ila-0.022
Ben-0.022
 se-0.022
Token Lily
Token ID20037
Feature activation-4.54e-02

Pos logit contributions
 bat+0.046
 se+0.045
Ben+0.045
ila+0.044
 slide+0.044
Neg logit contributions
 unlike-0.147
 blonde-0.139
 matter-0.136
 Tears-0.135
 Nobody-0.131
Token and
Token ID290
Feature activation-9.21e-02

Pos logit contributions
ila+0.035
Ben+0.035
 se+0.034
 bat+0.033
line+0.032
Neg logit contributions
 unlike-0.101
 tactic-0.097
 Tears-0.097
 Terr-0.094
 blonde-0.092
Token Ben
Token ID3932
Feature activation-2.07e-01

Pos logit contributions
ars+0.100
ites+0.096
is+0.089
aks+0.086
 pige+0.085
Neg logit contributions
 blonde-0.172
 Tears-0.163
 avail-0.162
 matter-0.154
 disgust-0.150
Token."
Token ID526
Feature activation-5.19e-02

Pos logit contributions
Ben+0.024
 bat+0.023
 Ben+0.023
ila+0.023
 se+0.023
Neg logit contributions
 unlike-0.053
 matter-0.049
 superpower-0.048
 perspective-0.047
 blonde-0.046
Token<|endoftext|>
Token ID50256
Feature activation+6.24e-03

Pos logit contributions
 Ben+0.044
Ben+0.036
 bat+0.035
 says+0.032
 they+0.031
Neg logit contributions
 unlike-0.137
 matter-0.130
 perspective-0.126
 superpower-0.125
eway-0.122
TokenOnce
Token ID7454
Feature activation+1.13e-01

Pos logit contributions
 unlike+0.012
 superpower+0.011
 matter+0.011
 blonde+0.011
 perspective+0.011
Neg logit contributions
 bat-0.008
 se-0.008
 Ben-0.008
ila-0.008
Ben-0.008
Token upon
Token ID2402
Feature activation+2.07e-01

Pos logit contributions
 unlike+0.148
 superpower+0.125
 matter+0.125
 blonde+0.119
 perspective+0.119
Neg logit contributions
 Ben-0.220
ila-0.217
 bat-0.215
Ben-0.215
 se-0.212
Token a
Token ID257
Feature activation+2.04e-03

Pos logit contributions
 unlike+0.489
 superpower+0.447
 matter+0.447
 perspective+0.438
 blonde+0.435
Neg logit contributions
 Ben-0.184
 bat-0.180
Ben-0.178
ila-0.177
 se-0.174
Token time
Token ID640
Feature activation+8.58e-02

Pos logit contributions
 unlike+0.003
 matter+0.003
 superpower+0.003
 blonde+0.003
 perspective+0.003
Neg logit contributions
 Ben-0.003
ila-0.003
 se-0.003
 bat-0.003
Ben-0.003
Token,
Token ID11
Feature activation+1.40e-02

Pos logit contributions
 unlike+0.208
 matter+0.192
 superpower+0.191
 perspective+0.187
 blonde+0.186
Neg logit contributions
 Ben-0.073
ila-0.069
 bat-0.068
Ben-0.068
 se-0.066
Token there
Token ID612
Feature activation+7.29e-02

Pos logit contributions
 unlike+0.035
 former+0.034
 superpower+0.034
 perspective+0.033
 matter+0.033
Neg logit contributions
 Ben-0.012
ila-0.010
Ben-0.010
 next-0.010
 they-0.009
Token was
Token ID373
Feature activation+8.77e-02

Pos logit contributions
 unlike+0.171
 former+0.163
 superpower+0.161
 perspective+0.157
 matter+0.155
Neg logit contributions
 Ben-0.075
Ben-0.067
 bat-0.066
ila-0.065
 put-0.064
Token a
Token ID257
Feature activation+2.75e-03

Pos logit contributions
 unlike+0.199
 matter+0.180
 superpower+0.179
 blonde+0.177
 perspective+0.175
Neg logit contributions
ila-0.083
 Ben-0.083
 bat-0.083
Ben-0.082
 se-0.081
Token small
Token ID1402
Feature activation+3.80e-02

Pos logit contributions
 unlike+0.004
 matter+0.004
 superpower+0.004
 blonde+0.004
 perspective+0.004
Neg logit contributions
ila-0.004
 Ben-0.004
 goes-0.004
 says-0.004
irts-0.004
Token old
Token ID1468
Feature activation-5.37e-02

Pos logit contributions
 unlike+0.116
 blonde+0.104
 superpower+0.103
 perspective+0.100
 former+0.097
Neg logit contributions
Ben-0.103
 bat-0.094
ila-0.092
 Ben-0.089
 throw-0.087
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
 Ben+0.019
 and+0.019
 each+0.017
 next+0.013
 bat+0.012
Neg logit contributions
 unlike-0.164
eway-0.159
 superpower-0.153
 matter-0.149
 former-0.149
Token
Token ID220
Feature activation+1.19e-01

Pos logit contributions
 unlike+0.024
 superpower+0.022
 matter+0.021
 perspective+0.021
 tactic+0.021
Neg logit contributions
 Ben-0.015
Ben-0.014
 bat-0.014
ila-0.014
 next-0.013
Token
Token ID198
Feature activation-2.84e-02

Pos logit contributions
 unlike+0.291
 matter+0.268
 superpower+0.268
 perspective+0.262
 blonde+0.261
Neg logit contributions
 Ben-0.094
ila-0.090
 bat-0.090
Ben-0.090
 se-0.086
TokenOne
Token ID3198
Feature activation+4.87e-02

Pos logit contributions
 Ben+0.023
 bat+0.022
Ben+0.022
ila+0.022
 se+0.022
Neg logit contributions
 unlike-0.069
 matter-0.063
 superpower-0.063
 perspective-0.062
 blonde-0.062
Token day
Token ID1110
Feature activation+2.79e-02

Pos logit contributions
 unlike+0.087
 matter+0.081
 superpower+0.076
 blonde+0.073
 tactic+0.073
Neg logit contributions
Ben-0.067
ila-0.066
 se-0.066
 Ben-0.066
 bat-0.063
Token,
Token ID11
Feature activation+7.17e-03

Pos logit contributions
 unlike+0.079
 former+0.077
 superpower+0.074
 perspective+0.074
 tactic+0.073
Neg logit contributions
 Ben-0.016
 they-0.013
Ben-0.011
 bat-0.011
 next-0.011
Token Sarah
Token ID10490
Feature activation+1.03e-01

Pos logit contributions
 unlike+0.014
 former+0.014
 superpower+0.014
 perspective+0.013
eway+0.013
Neg logit contributions
 Ben-0.010
 they-0.010
 next-0.009
 put-0.009
 Tom-0.009
Token went
Token ID1816
Feature activation-1.22e-01

Pos logit contributions
 unlike+0.184
 blonde+0.167
 matter+0.164
 shouldn+0.157
 darkened+0.156
Neg logit contributions
 dads-0.134
ila-0.130
line-0.126
 bat-0.117
Ben-0.116
Token to
Token ID284
Feature activation+6.84e-02

Pos logit contributions
Ben+0.188
ila+0.184
 Ben+0.170
 se+0.166
 pige+0.165
Neg logit contributions
 unlike-0.210
 superpower-0.192
 tactic-0.175
eway-0.173
 matter-0.172
Token the
Token ID262
Feature activation-6.38e-02

Pos logit contributions
 unlike+0.159
 matter+0.155
 superpower+0.150
 blonde+0.148
 tactic+0.147
Neg logit contributions
ila-0.055
Ben-0.052
 se-0.049
 bat-0.047
line-0.045
Token,
Token ID11
Feature activation+8.04e-05

Pos logit contributions
 unlike+0.164
 matter+0.151
 superpower+0.150
 perspective+0.146
 blonde+0.146
Neg logit contributions
 Ben-0.062
 bat-0.062
ila-0.062
Ben-0.062
 se-0.061
Token Leo
Token ID19632
Feature activation-1.41e-01

Pos logit contributions
 unlike+0.000
 matter+0.000
 superpower+0.000
 tactic+0.000
 blonde+0.000
Neg logit contributions
Ben-0.000
ila-0.000
 bat-0.000
 se-0.000
 Ben-0.000
Token."
Token ID526
Feature activation-6.39e-02

Pos logit contributions
Ben+0.132
ila+0.130
 bat+0.129
 Ben+0.129
 se+0.128
Neg logit contributions
 unlike-0.316
 matter-0.291
 superpower-0.289
 perspective-0.278
 blonde-0.277
Token
Token ID198
Feature activation-3.32e-02

Pos logit contributions
 bat+0.060
 se+0.060
ila+0.059
Ben+0.058
 Ben+0.056
Neg logit contributions
 unlike-0.144
 superpower-0.130
 tactic-0.128
 blonde-0.127
 perspective-0.127
Token
Token ID198
Feature activation-3.29e-02

Pos logit contributions
 se+0.043
 bat+0.042
 wave+0.039
 goes+0.039
 put+0.039
Neg logit contributions
 unlike-0.060
eway-0.056
 Tears-0.055
 superpower-0.053
 Terr-0.053
TokenBut
Token ID1537
Feature activation+7.72e-02

Pos logit contributions
 se+0.035
 bat+0.034
ila+0.032
 Ben+0.032
 put+0.031
Neg logit contributions
 unlike-0.069
 superpower-0.063
eway-0.063
 perspective-0.062
 tactic-0.061
Token Leo
Token ID19632
Feature activation-1.40e-01

Pos logit contributions
 unlike+0.196
 superpower+0.182
 matter+0.182
 perspective+0.180
 blonde+0.179
Neg logit contributions
 Ben-0.060
ila-0.054
 bat-0.053
Ben-0.053
 se-0.050
Token was
Token ID373
Feature activation+7.77e-02

Pos logit contributions
 bat+0.175
 Ben+0.175
ila+0.175
Ben+0.173
 se+0.172
Neg logit contributions
 unlike-0.272
 matter-0.244
 superpower-0.243
 blonde-0.236
 tactic-0.232
Token angry
Token ID7954
Feature activation+4.07e-02

Pos logit contributions
 unlike+0.169
 matter+0.153
 superpower+0.152
 blonde+0.149
 perspective+0.148
Neg logit contributions
 Ben-0.084
 bat-0.083
Ben-0.082
ila-0.082
 se-0.081
Token.
Token ID13
Feature activation+9.96e-03

Pos logit contributions
 unlike+0.092
 matter+0.083
 superpower+0.081
 tactic+0.080
 perspective+0.080
Neg logit contributions
 bat-0.039
Ben-0.039
 se-0.038
ila-0.038
 Ben-0.036
Token He
Token ID679
Feature activation+1.06e-01

Pos logit contributions
 matter+0.018
 unlike+0.018
 perspective+0.017
 superpower+0.016
 former+0.015
Neg logit contributions
 Ben-0.017
ila-0.015
Ben-0.015
 says-0.014
 bat-0.014
Token who
Token ID508
Feature activation-7.95e-03

Pos logit contributions
ila+0.010
Ben+0.009
 Ben+0.009
 bat+0.009
 se+0.009
Neg logit contributions
 unlike-0.021
 superpower-0.019
 perspective-0.018
 matter-0.018
 blonde-0.018
Token liked
Token ID8288
Feature activation+5.06e-02

Pos logit contributions
 Ben+0.011
ila+0.011
 bat+0.011
Ben+0.011
 se+0.011
Neg logit contributions
 unlike-0.015
 matter-0.013
 superpower-0.013
 blonde-0.013
 tactic-0.012
Token to
Token ID284
Feature activation+7.51e-02

Pos logit contributions
 unlike+0.110
 matter+0.099
 superpower+0.099
 perspective+0.097
 blonde+0.097
Neg logit contributions
 Ben-0.054
 bat-0.053
ila-0.053
Ben-0.053
 se-0.052
Token play
Token ID711
Feature activation-9.03e-02

Pos logit contributions
 unlike+0.184
 matter+0.171
 superpower+0.171
 blonde+0.168
 tactic+0.165
Neg logit contributions
ila-0.057
 Ben-0.056
Ben-0.056
 bat-0.053
 se-0.052
Token with
Token ID351
Feature activation-9.81e-03

Pos logit contributions
 Ben+0.087
 bat+0.085
Ben+0.084
ila+0.084
 se+0.083
Neg logit contributions
 unlike-0.207
 matter-0.189
 superpower-0.189
 perspective-0.184
 blonde-0.184
Token fire
Token ID2046
Feature activation+9.37e-02

Pos logit contributions
 Ben+0.007
 bat+0.006
 se+0.006
 each+0.006
Ben+0.006
Neg logit contributions
 unlike-0.027
 matter-0.025
 superpower-0.025
 perspective-0.024
eway-0.024
Tokenwork
Token ID1818
Feature activation+6.80e-02

Pos logit contributions
 unlike+0.237
 matter+0.220
 blonde+0.219
 superpower+0.217
 tactic+0.213
Neg logit contributions
 Ben-0.070
 bat-0.068
 se-0.065
 next-0.063
 each-0.061
Token.
Token ID13
Feature activation+2.20e-02

Pos logit contributions
 unlike+0.164
 matter+0.150
 superpower+0.150
 perspective+0.146
 blonde+0.146
Neg logit contributions
 Ben-0.056
ila-0.055
Ben-0.055
 bat-0.055
 se-0.054
Token They
Token ID1119
Feature activation-2.90e-02

Pos logit contributions
 unlike+0.031
 matter+0.027
 superpower+0.027
 blonde+0.026
 tactic+0.026
Neg logit contributions
ila-0.039
 se-0.039
 bat-0.039
Ben-0.039
 Ben-0.038
Token found
Token ID1043
Feature activation-1.12e-02

Pos logit contributions
 Ben+0.032
ila+0.032
Ben+0.032
 bat+0.031
 se+0.030
Neg logit contributions
 unlike-0.061
 matter-0.056
 superpower-0.055
 blonde-0.054
 perspective-0.053
Token some
Token ID617
Feature activation-1.64e-02

Pos logit contributions
 Ben+0.009
 se+0.009
 bat+0.009
ila+0.008
 each+0.008
Neg logit contributions
 unlike-0.027
 matter-0.026
 superpower-0.026
 perspective-0.025
 blonde-0.025
Token other
Token ID584
Feature activation+4.67e-02

Pos logit contributions
 Ben+0.163
 bat+0.161
ila+0.159
Ben+0.159
 se+0.157
Neg logit contributions
 unlike-0.452
 matter-0.417
 superpower-0.414
 perspective-0.406
 blonde-0.404
Token.
Token ID13
Feature activation+2.62e-02

Pos logit contributions
 unlike+0.102
 superpower+0.094
 matter+0.093
 tactic+0.092
 perspective+0.092
Neg logit contributions
ila-0.050
Ben-0.047
 se-0.046
 Ben-0.044
 bat-0.044
Token They
Token ID1119
Feature activation-3.21e-02

Pos logit contributions
 unlike+0.061
 matter+0.057
 superpower+0.057
 tactic+0.055
 perspective+0.055
Neg logit contributions
 Ben-0.025
ila-0.024
Ben-0.023
 bat-0.023
 se-0.022
Token love
Token ID1842
Feature activation-1.60e-02

Pos logit contributions
ila+0.033
Ben+0.031
 se+0.030
 Ben+0.030
 pige+0.030
Neg logit contributions
 matter-0.065
 unlike-0.063
 superpower-0.061
 blonde-0.060
 perspective-0.059
Token their
Token ID511
Feature activation-4.93e-03

Pos logit contributions
ila+0.013
Ben+0.013
 bat+0.012
 se+0.012
 Ben+0.011
Neg logit contributions
 unlike-0.039
 matter-0.036
 superpower-0.035
 perspective-0.035
 blonde-0.034
Token globe
Token ID13342
Feature activation+4.77e-02

Pos logit contributions
 Ben+0.006
 bat+0.006
ila+0.006
Ben+0.006
 se+0.006
Neg logit contributions
 unlike-0.010
 matter-0.009
 superpower-0.009
 blonde-0.009
 perspective-0.009
Token and
Token ID290
Feature activation-8.20e-02

Pos logit contributions
 unlike+0.114
 matter+0.102
 superpower+0.102
 perspective+0.100
 tactic+0.099
Neg logit contributions
Ben-0.043
ila-0.042
 Ben-0.040
 bat-0.039
 se-0.038
Token their
Token ID511
Feature activation-5.21e-03

Pos logit contributions
ila+0.087
Ben+0.086
 se+0.080
 Ben+0.079
 bat+0.078
Neg logit contributions
 unlike-0.179
 matter-0.165
 superpower-0.164
 blonde-0.159
 perspective-0.158
Token planet
Token ID5440
Feature activation+7.95e-05

Pos logit contributions
 Ben+0.006
 bat+0.006
Ben+0.006
ila+0.006
 se+0.006
Neg logit contributions
 unlike-0.010
 matter-0.009
 superpower-0.009
 blonde-0.009
 perspective-0.009
Token.
Token ID13
Feature activation+2.64e-02

Pos logit contributions
 unlike+0.000
 superpower+0.000
 matter+0.000
 perspective+0.000
 tactic+0.000
Neg logit contributions
Ben-0.000
ila-0.000
 Ben-0.000
 bat-0.000
 se-0.000
Token They
Token ID1119
Feature activation-3.17e-02

Pos logit contributions
 unlike+0.062
 matter+0.057
 superpower+0.056
 perspective+0.055
 tactic+0.055
Neg logit contributions
 Ben-0.025
 bat-0.024
Ben-0.024
ila-0.024
 se-0.023
Token They
Token ID1119
Feature activation-5.51e-02

Pos logit contributions
 unlike+0.016
 superpower+0.014
 matter+0.014
 blonde+0.014
 perspective+0.013
Neg logit contributions
 se-0.013
ila-0.012
 bat-0.012
Ben-0.012
 Ben-0.012
Token must
Token ID1276
Feature activation+1.56e-01

Pos logit contributions
ila+0.060
Ben+0.060
 Ben+0.056
 se+0.052
 bat+0.051
Neg logit contributions
 unlike-0.112
 matter-0.109
 superpower-0.104
 blonde-0.100
 former-0.099
Token be
Token ID307
Feature activation-1.25e-01

Pos logit contributions
 matter+0.312
 unlike+0.301
 former+0.298
 superpower+0.288
eway+0.277
Neg logit contributions
Ben-0.161
 se-0.159
ila-0.158
 Ben-0.154
 pige-0.153
Token happy
Token ID3772
Feature activation+1.70e-02

Pos logit contributions
 Ben+0.152
 bat+0.150
ila+0.148
Ben+0.148
 se+0.147
Neg logit contributions
 unlike-0.255
 superpower-0.230
 matter-0.230
 perspective-0.225
 blonde-0.223
Token."
Token ID526
Feature activation-6.44e-02

Pos logit contributions
 unlike+0.040
 superpower+0.035
 matter+0.035
 perspective+0.035
 tactic+0.034
Neg logit contributions
Ben-0.016
ila-0.016
 bat-0.015
 se-0.015
 Ben-0.015
Token
Token ID198
Feature activation-3.84e-02

Pos logit contributions
 bat+0.062
 se+0.062
ila+0.060
Ben+0.059
 Ben+0.058
Neg logit contributions
 unlike-0.143
 superpower-0.129
 matter-0.129
 blonde-0.128
 tactic-0.127
Token
Token ID198
Feature activation-3.77e-02

Pos logit contributions
 se+0.050
 bat+0.049
 goes+0.048
 blow+0.048
 wave+0.047
Neg logit contributions
 unlike-0.064
eway-0.060
 Tears-0.059
 tactic-0.058
 perspective-0.058
TokenThey
Token ID2990
Feature activation-1.00e-01

Pos logit contributions
 se+0.039
 Ben+0.038
 bat+0.038
 goes+0.036
 waves+0.035
Neg logit contributions
 unlike-0.076
 matter-0.072
 perspective-0.071
eway-0.070
 superpower-0.070
Token walk
Token ID2513
Feature activation-1.29e-01

Pos logit contributions
ila+0.103
 pige+0.099
Ben+0.096
 we+0.094
 next+0.094
Neg logit contributions
 matter-0.190
 blonde-0.184
 unlike-0.181
 superpower-0.181
 darkened-0.174
Token closer
Token ID5699
Feature activation+1.32e-01

Pos logit contributions
Ben+0.158
ila+0.157
 se+0.157
 bat+0.156
 Ben+0.155
Neg logit contributions
 unlike-0.251
 superpower-0.226
 matter-0.225
 tactic-0.224
 perspective-0.222
Token to
Token ID284
Feature activation+5.61e-02

Pos logit contributions
 unlike+0.279
 superpower+0.251
 perspective+0.248
 tactic+0.246
 matter+0.246
Neg logit contributions
 bat-0.145
Ben-0.145
ila-0.141
 se-0.140
 Ben-0.137
Token
Token ID198
Feature activation-1.79e-02

Pos logit contributions
 unlike+0.297
 superpower+0.271
 matter+0.269
 blonde+0.265
 perspective+0.264
Neg logit contributions
 bat-0.106
 Ben-0.105
 se-0.104
ila-0.104
Ben-0.104
Token
Token ID198
Feature activation-1.75e-02

Pos logit contributions
 se+0.023
 bat+0.023
 goes+0.022
 wave+0.021
 dove+0.021
Neg logit contributions
 unlike-0.032
 Terr-0.030
eway-0.030
 Tears-0.029
 tactic-0.028
TokenThe
Token ID464
Feature activation-2.56e-02

Pos logit contributions
 Ben+0.022
 says+0.020
 bat+0.020
 se+0.020
ila+0.020
Neg logit contributions
 unlike-0.035
eway-0.030
 matter-0.030
 superpower-0.030
 perspective-0.030
Token girl
Token ID2576
Feature activation+1.33e-01

Pos logit contributions
 Ben+0.031
 bat+0.031
ila+0.031
Ben+0.031
 se+0.031
Neg logit contributions
 unlike-0.052
 superpower-0.047
 matter-0.047
 perspective-0.046
 blonde-0.046
Token then
Token ID788
Feature activation-1.08e-01

Pos logit contributions
 unlike+0.249
 superpower+0.218
 matter+0.212
 blonde+0.211
 Tears+0.208
Neg logit contributions
ila-0.179
Ben-0.177
 Ben-0.172
 bat-0.169
 dads-0.165
Token gave
Token ID2921
Feature activation-8.25e-02

Pos logit contributions
ila+0.119
Ben+0.119
 bat+0.113
 Ben+0.111
line+0.109
Neg logit contributions
 unlike-0.222
 matter-0.200
 blonde-0.196
 superpower-0.195
 darkened-0.195
Token him
Token ID683
Feature activation+5.80e-02

Pos logit contributions
 bat+0.089
Ben+0.087
ila+0.086
 se+0.085
 Ben+0.084
Neg logit contributions
 unlike-0.176
 matter-0.162
 superpower-0.159
 perspective-0.156
 blonde-0.156
Token a
Token ID257
Feature activation+1.10e-03

Pos logit contributions
 unlike+0.118
 superpower+0.110
 matter+0.108
 blonde+0.107
 Tears+0.103
Neg logit contributions
Ben-0.064
ila-0.063
 se-0.061
 bat-0.059
line-0.058
Token hug
Token ID16225
Feature activation-3.14e-02

Pos logit contributions
 unlike+0.003
 matter+0.003
 perspective+0.003
eway+0.002
 superpower+0.002
Neg logit contributions
 bat-0.001
 se-0.001
 Ben-0.001
Ben-0.001
ila-0.001
Token and
Token ID290
Feature activation-7.77e-02

Pos logit contributions
 Ben+0.015
 bat+0.015
ila+0.014
Ben+0.014
 se+0.014
Neg logit contributions
 unlike-0.086
 matter-0.080
 superpower-0.080
 perspective-0.078
 blonde-0.078
Token said
Token ID531
Feature activation-5.77e-02

Pos logit contributions
Ben+0.068
ila+0.067
 bat+0.063
 Ben+0.059
line+0.057
Neg logit contributions
 unlike-0.183
 matter-0.167
 superpower-0.165
 blonde-0.162
 tactic-0.159
Token a
Token ID257
Feature activation-4.30e-03

Pos logit contributions
 unlike+0.258
 matter+0.238
 superpower+0.238
 perspective+0.233
 tactic+0.233
Neg logit contributions
 Ben-0.049
 bat-0.048
Ben-0.045
 se-0.045
ila-0.044
Token field
Token ID2214
Feature activation-4.16e-02

Pos logit contributions
ila+0.005
 Ben+0.005
Ben+0.005
 bat+0.005
 se+0.005
Neg logit contributions
 unlike-0.009
 superpower-0.008
 matter-0.008
 perspective-0.008
 blonde-0.008
Token of
Token ID286
Feature activation+9.81e-03

Pos logit contributions
Ben+0.044
ila+0.044
 Ben+0.043
 se+0.041
 bat+0.041
Neg logit contributions
 unlike-0.091
 perspective-0.082
eway-0.080
 tactic-0.080
 superpower-0.080
Token grass
Token ID8701
Feature activation-3.42e-02

Pos logit contributions
 unlike+0.020
 matter+0.018
 blonde+0.018
 superpower+0.018
 perspective+0.018
Neg logit contributions
ila-0.012
 Ben-0.011
Ben-0.011
 bat-0.011
 se-0.011
Token.
Token ID13
Feature activation+1.89e-02

Pos logit contributions
 Ben+0.027
 bat+0.027
ila+0.027
Ben+0.027
 se+0.026
Neg logit contributions
 unlike-0.083
 superpower-0.076
 matter-0.076
 perspective-0.075
 blonde-0.074
Token The
Token ID383
Feature activation-7.81e-03

Pos logit contributions
 unlike+0.038
 superpower+0.036
 matter+0.035
 perspective+0.035
 tactic+0.035
Neg logit contributions
 Ben-0.025
 bat-0.023
Ben-0.022
 put-0.021
 Tom-0.021
Token bright
Token ID6016
Feature activation+3.89e-02

Pos logit contributions
ila+0.013
 Ben+0.011
 gives+0.010
Ben+0.010
 says+0.010
Neg logit contributions
 unlike-0.013
 matter-0.012
 perspective-0.011
 blonde-0.011
 darkened-0.011
Token green
Token ID4077
Feature activation-4.19e-03

Pos logit contributions
 unlike+0.081
 matter+0.074
 perspective+0.071
 blonde+0.070
heart+0.069
Neg logit contributions
ila-0.048
Ben-0.046
 Ben-0.044
 bat-0.040
 se-0.039
Token grass
Token ID8701
Feature activation-3.37e-02

Pos logit contributions
ila+0.005
 Ben+0.005
Ben+0.004
 bat+0.004
 se+0.004
Neg logit contributions
 unlike-0.009
 matter-0.008
 perspective-0.008
 superpower-0.008
 blonde-0.008
Token was
Token ID373
Feature activation+7.97e-02

Pos logit contributions
 Ben+0.042
 bat+0.041
ila+0.041
Ben+0.041
 se+0.040
Neg logit contributions
 unlike-0.067
 matter-0.062
 superpower-0.061
 perspective-0.060
 blonde-0.059
Token very
Token ID845
Feature activation+1.17e-01

Pos logit contributions
 unlike+0.170
 superpower+0.164
 perspective+0.163
 tactic+0.163
 matter+0.161
Neg logit contributions
 bat-0.086
 Ben-0.085
 se-0.082
 next-0.080
 put-0.079
Token anyway
Token ID6949
Feature activation+1.59e-01

Pos logit contributions
ila+0.011
Ben+0.011
 Ben+0.011
 se+0.010
 bat+0.010
Neg logit contributions
 unlike-0.023
 superpower-0.021
 matter-0.020
 tactic-0.020
 blonde-0.020
Token.
Token ID13
Feature activation+2.83e-02

Pos logit contributions
 unlike+0.340
 matter+0.306
eway+0.297
 superpower+0.297
 tactic+0.294
Neg logit contributions
Ben-0.185
 bat-0.173
 Ben-0.169
 se-0.168
ila-0.166
Token
Token ID198
Feature activation-1.72e-02

Pos logit contributions
 unlike+0.053
 matter+0.048
 superpower+0.047
 perspective+0.046
 blonde+0.045
Neg logit contributions
 Ben-0.039
 bat-0.039
ila-0.038
Ben-0.038
 se-0.038
Token
Token ID198
Feature activation-1.70e-02

Pos logit contributions
 Ben+0.020
 se+0.020
 bat+0.019
 goes+0.019
 wave+0.019
Neg logit contributions
 unlike-0.033
eway-0.031
 superpower-0.030
 Tears-0.030
 tactic-0.029
TokenAfter
Token ID3260
Feature activation+1.28e-01

Pos logit contributions
 Ben+0.025
 bat+0.023
 se+0.023
ila+0.022
 says+0.022
Neg logit contributions
 unlike-0.030
eway-0.027
 superpower-0.026
 perspective-0.026
 tactic-0.026
Token a
Token ID257
Feature activation+1.51e-03

Pos logit contributions
 unlike+0.330
 superpower+0.304
 matter+0.302
 blonde+0.297
 perspective+0.295
Neg logit contributions
Ben-0.082
ila-0.081
 Ben-0.080
 bat-0.080
 se-0.077
Token little
Token ID1310
Feature activation+6.75e-02

Pos logit contributions
 unlike+0.002
adoes+0.002
eway+0.002
 blonde+0.002
 perspective+0.002
Neg logit contributions
 bat-0.003
 se-0.003
Ben-0.003
 Ben-0.002
 next-0.002
Token while
Token ID981
Feature activation+1.77e-01

Pos logit contributions
 unlike+0.127
 matter+0.114
 superpower+0.114
 perspective+0.110
 blonde+0.110
Neg logit contributions
 Ben-0.091
ila-0.090
 bat-0.090
Ben-0.090
 se-0.088
Token,
Token ID11
Feature activation+1.77e-02

Pos logit contributions
 unlike+0.412
 matter+0.375
 superpower+0.374
 blonde+0.365
 perspective+0.365
Neg logit contributions
Ben-0.161
 bat-0.159
 Ben-0.159
ila-0.155
 se-0.153
Token Mam
Token ID29926
Feature activation-3.65e-02

Pos logit contributions
 unlike+0.047
 superpower+0.043
 matter+0.043
 perspective+0.042
 blonde+0.042
Neg logit contributions
 Ben-0.011
 bat-0.010
ila-0.010
 se-0.010
Ben-0.010
Tokenma
Token ID2611
Feature activation+2.35e-02

Pos logit contributions
 bat+0.054
Ben+0.053
 pige+0.051
 Ben+0.051
ashes+0.048
Neg logit contributions
 matter-0.058
 unlike-0.055
heart-0.054
 exactly-0.052
 tactic-0.052
Token good
Token ID922
Feature activation-1.45e-02

Pos logit contributions
 unlike+0.315
 superpower+0.287
 perspective+0.286
 matter+0.284
 blonde+0.282
Neg logit contributions
 Ben-0.181
 bat-0.175
 se-0.173
ila-0.173
Ben-0.170
Token and
Token ID290
Feature activation-8.58e-02

Pos logit contributions
 Ben+0.011
 bat+0.010
ila+0.010
Ben+0.010
 se+0.010
Neg logit contributions
 unlike-0.037
 superpower-0.034
 matter-0.034
 blonde-0.033
 perspective-0.033
Token make
Token ID787
Feature activation-9.67e-02

Pos logit contributions
ila+0.104
Ben+0.104
 Ben+0.103
 bat+0.102
 se+0.099
Neg logit contributions
 unlike-0.174
 matter-0.160
 superpower-0.156
 blonde-0.152
 tactic-0.151
Token someone
Token ID2130
Feature activation+6.51e-02

Pos logit contributions
 bat+0.114
Ben+0.113
 Ben+0.113
ila+0.113
 se+0.111
Neg logit contributions
 unlike-0.198
 matter-0.180
 superpower-0.178
 perspective-0.173
 blonde-0.173
Token happy
Token ID3772
Feature activation+2.40e-02

Pos logit contributions
 unlike+0.138
 matter+0.125
 superpower+0.124
 blonde+0.121
 perspective+0.121
Neg logit contributions
 Ben-0.073
Ben-0.071
ila-0.071
 bat-0.071
 se-0.069
Token.
Token ID13
Feature activation+1.70e-02

Pos logit contributions
 unlike+0.065
 blonde+0.061
 superpower+0.060
 perspective+0.060
 matter+0.060
Neg logit contributions
 Ben-0.015
 next-0.013
 bat-0.012
 says-0.012
 se-0.011
Token The
Token ID383
Feature activation-1.00e-02

Pos logit contributions
 unlike+0.031
 Tears+0.028
 superpower+0.027
 Terr+0.027
 blonde+0.026
Neg logit contributions
 se-0.024
ila-0.023
 bat-0.023
 wave-0.022
Ben-0.022
Token end
Token ID886
Feature activation-2.04e-02

Pos logit contributions
ila+0.011
 Ben+0.010
Ben+0.010
 bat+0.010
 se+0.010
Neg logit contributions
 unlike-0.021
 matter-0.021
 superpower-0.020
 blonde-0.019
 perspective-0.019
Token.
Token ID13
Feature activation+1.80e-02

Pos logit contributions
 Ben+0.012
 bat+0.011
Ben+0.011
ila+0.011
 next+0.011
Neg logit contributions
 unlike-0.056
 blonde-0.052
 perspective-0.051
 matter-0.051
 superpower-0.051
Token<|endoftext|>
Token ID50256
Feature activation-3.60e-03

Pos logit contributions
 unlike+0.036
 Tears+0.031
 matter+0.031
 superpower+0.031
 Terr+0.031
Neg logit contributions
 se-0.022
ila-0.022
 dads-0.021
 bat-0.021
 wave-0.020
TokenOnce
Token ID7454
Feature activation+1.04e-01

Pos logit contributions
 se+0.004
 dads+0.004
ila+0.004
 bat+0.004
Ben+0.004
Neg logit contributions
 unlike-0.007
 superpower-0.007
 Tears-0.006
eway-0.006
 tactic-0.006
Token and
Token ID290
Feature activation-9.41e-02

Pos logit contributions
 unlike+0.009
 superpower+0.008
 matter+0.008
 blonde+0.008
 tactic+0.008
Neg logit contributions
 bat-0.004
Ben-0.004
 se-0.004
ila-0.004
 Ben-0.003
Token says
Token ID1139
Feature activation-1.29e-01

Pos logit contributions
Ben+0.087
ila+0.086
 bat+0.083
 se+0.075
 pige+0.070
Neg logit contributions
 unlike-0.209
 matter-0.197
 superpower-0.193
 blonde-0.190
 tactic-0.188
Token hello
Token ID23748
Feature activation+1.01e-01

Pos logit contributions
 bat+0.122
Ben+0.121
 Ben+0.120
ila+0.119
 se+0.117
Neg logit contributions
 unlike-0.293
 superpower-0.267
 matter-0.265
 blonde-0.262
 perspective-0.258
Token.
Token ID13
Feature activation+1.52e-02

Pos logit contributions
 unlike+0.280
 matter+0.264
 superpower+0.260
 perspective+0.256
 tactic+0.256
Neg logit contributions
 Ben-0.064
 next-0.055
 each-0.050
 bat-0.049
 and-0.049
Token Anna
Token ID11735
Feature activation-1.08e-01

Pos logit contributions
 unlike+0.030
 matter+0.027
 superpower+0.027
 perspective+0.027
 tactic+0.026
Neg logit contributions
 Ben-0.021
Ben-0.019
 bat-0.019
ila-0.019
 se-0.019
Token says
Token ID1139
Feature activation-1.28e-01

Pos logit contributions
ila+0.097
line+0.091
 se+0.090
Ben+0.089
 bat+0.085
Neg logit contributions
 unlike-0.228
 superpower-0.223
 Terr-0.211
 blonde-0.210
 Tears-0.206
Token,
Token ID11
Feature activation-1.04e-04

Pos logit contributions
 bat+0.158
Ben+0.155
ila+0.148
 se+0.144
 Ben+0.139
Neg logit contributions
 unlike-0.250
 superpower-0.226
 blonde-0.226
 matter-0.219
 tactic-0.212
Token "
Token ID366
Feature activation+6.71e-02

Pos logit contributions
ila+0.000
 bat+0.000
Ben+0.000
 se+0.000
 Ben+0.000
Neg logit contributions
 unlike-0.000
 superpower-0.000
 blonde-0.000
 matter-0.000
 Tears-0.000
TokenHi
Token ID17250
Feature activation+1.32e-01

Pos logit contributions
 unlike+0.100
 matter+0.093
 blonde+0.089
 perspective+0.088
 superpower+0.086
Neg logit contributions
 bat-0.110
 Ben-0.108
ila-0.106
 se-0.104
 put-0.102
Token,
Token ID11
Feature activation+7.45e-04

Pos logit contributions
 unlike+0.286
 matter+0.264
 superpower+0.261
 blonde+0.257
 perspective+0.253
Neg logit contributions
 bat-0.134
 se-0.131
ila-0.131
Ben-0.130
 Ben-0.124
Token dad
Token ID9955
Feature activation-1.45e-01

Pos logit contributions
 unlike+0.002
 matter+0.001
 superpower+0.001
 blonde+0.001
 tactic+0.001
Neg logit contributions
 bat-0.001
Ben-0.001
ila-0.001
 se-0.001
irts-0.001
TokenMom
Token ID29252
Feature activation-3.88e-02

Pos logit contributions
 unlike+0.165
 superpower+0.157
 tactic+0.152
 matter+0.151
 perspective+0.150
Neg logit contributions
Ben-0.065
 Ben-0.059
ila-0.058
 bat-0.058
 se-0.057
Tokenmy
Token ID1820
Feature activation+7.72e-02

Pos logit contributions
Ben+0.048
ila+0.048
 bat+0.048
 Ben+0.047
 se+0.047
Neg logit contributions
 unlike-0.076
 matter-0.070
 superpower-0.068
 blonde-0.067
 perspective-0.067
Token,
Token ID11
Feature activation+5.37e-03

Pos logit contributions
 unlike+0.161
 matter+0.147
 superpower+0.143
 blonde+0.143
 perspective+0.138
Neg logit contributions
ila-0.088
Ben-0.086
 bat-0.084
 se-0.083
 Ben-0.081
Token can
Token ID460
Feature activation+4.91e-02

Pos logit contributions
 unlike+0.011
 matter+0.010
 superpower+0.010
 blonde+0.010
 perspective+0.010
Neg logit contributions
ila-0.006
Ben-0.006
 bat-0.006
 Ben-0.006
 se-0.006
Token I
Token ID314
Feature activation-1.52e-01

Pos logit contributions
 unlike+0.112
 matter+0.105
 blonde+0.101
 superpower+0.099
 alas+0.098
Neg logit contributions
Ben-0.045
ila-0.044
 bat-0.043
 se-0.040
 dads-0.038
Token have
Token ID423
Feature activation-1.42e-01

Pos logit contributions
ila+0.139
 Ben+0.138
Ben+0.134
 se+0.125
 bat+0.122
Neg logit contributions
 unlike-0.352
 matter-0.329
 superpower-0.325
 blonde-0.315
 tactic-0.310
Token this
Token ID428
Feature activation-7.01e-02

Pos logit contributions
 Ben+0.107
 bat+0.104
 se+0.101
Ben+0.100
ila+0.100
Neg logit contributions
 unlike-0.355
 superpower-0.326
 matter-0.326
 perspective-0.321
 tactic-0.320
Token toy
Token ID13373
Feature activation+1.10e-01

Pos logit contributions
 bat+0.083
 Ben+0.081
 slide+0.077
 se+0.075
 next+0.075
Neg logit contributions
 unlike-0.148
 perspective-0.137
 blonde-0.136
 matter-0.133
eway-0.133
Token please
Token ID3387
Feature activation+6.21e-02

Pos logit contributions
 unlike+0.262
 matter+0.243
 blonde+0.242
 perspective+0.241
 superpower+0.241
Neg logit contributions
 Ben-0.098
 bat-0.097
 se-0.092
Ben-0.088
ila-0.087
Token?"
Token ID1701
Feature activation-3.32e-02

Pos logit contributions
 unlike+0.109
 matter+0.100
 superpower+0.097
 tactic+0.090
 blonde+0.090
Neg logit contributions
ila-0.091
Ben-0.089
 Ben-0.083
 se-0.082
 dads-0.080
Token Mia
Token ID32189
Feature activation-7.99e-02

Pos logit contributions
 Ben+0.031
ila+0.029
 bat+0.029
Ben+0.029
 se+0.028
Neg logit contributions
 unlike-0.079
 superpower-0.072
 matter-0.072
 perspective-0.071
 blonde-0.070
Token and
Token ID290
Feature activation-1.05e-01

Pos logit contributions
ila+0.007
Ben+0.007
 se+0.007
 bat+0.007
 Ben+0.007
Neg logit contributions
 unlike-0.018
 matter-0.016
 superpower-0.016
 perspective-0.015
 blonde-0.015
Token not
Token ID407
Feature activation-5.98e-02

Pos logit contributions
ila+0.090
Ben+0.089
 bat+0.084
 Ben+0.083
 se+0.081
Neg logit contributions
 unlike-0.250
 matter-0.230
 superpower-0.228
 blonde-0.220
 perspective-0.219
Token bother
Token ID11393
Feature activation-8.79e-02

Pos logit contributions
 Ben+0.064
 bat+0.063
 se+0.061
Ben+0.061
ila+0.060
Neg logit contributions
 unlike-0.131
 superpower-0.119
 matter-0.117
 blonde-0.116
 tactic-0.116
Token them
Token ID606
Feature activation-8.60e-03

Pos logit contributions
 Ben+0.075
 bat+0.072
ila+0.071
Ben+0.071
 se+0.069
Neg logit contributions
 unlike-0.212
 superpower-0.194
 matter-0.193
 blonde-0.190
 perspective-0.190
Token when
Token ID618
Feature activation+9.83e-02

Pos logit contributions
 Ben+0.008
ila+0.008
 bat+0.008
Ben+0.008
 se+0.008
Neg logit contributions
 unlike-0.020
 matter-0.018
 superpower-0.018
 perspective-0.018
 blonde-0.017
Token they
Token ID484
Feature activation-1.29e-01

Pos logit contributions
 unlike+0.272
 matter+0.253
 superpower+0.250
 perspective+0.245
 blonde+0.245
Neg logit contributions
Ben-0.042
 bat-0.041
ila-0.041
 se-0.040
 Ben-0.038
Token are
Token ID389
Feature activation-8.67e-02

Pos logit contributions
ila+0.128
Ben+0.127
 Ben+0.127
 bat+0.122
 se+0.120
Neg logit contributions
 unlike-0.275
 matter-0.262
 superpower-0.254
 perspective-0.249
 tactic-0.245
Token working
Token ID1762
Feature activation-3.93e-02

Pos logit contributions
 Ben+0.110
Ben+0.109
 bat+0.108
ila+0.108
 se+0.107
Neg logit contributions
 unlike-0.171
 matter-0.153
 superpower-0.153
 blonde-0.149
 perspective-0.148
Token.
Token ID13
Feature activation+1.97e-03

Pos logit contributions
 Ben+0.038
ila+0.037
Ben+0.037
 bat+0.037
 se+0.036
Neg logit contributions
 unlike-0.089
 matter-0.081
 superpower-0.081
 perspective-0.079
 blonde-0.079
Token They
Token ID1119
Feature activation-5.70e-02

Pos logit contributions
 unlike+0.005
 matter+0.004
 superpower+0.004
 perspective+0.004
 tactic+0.004
Neg logit contributions
 Ben-0.002
Ben-0.002
ila-0.002
 bat-0.002
 se-0.002
Token learned
Token ID4499
Feature activation-7.02e-02

Pos logit contributions
ila+0.059
Ben+0.058
 Ben+0.058
 bat+0.055
 se+0.054
Neg logit contributions
 unlike-0.117
 matter-0.113
 superpower-0.110
 perspective-0.106
 blonde-0.106
Token.
Token ID13
Feature activation+2.98e-02

Pos logit contributions
 bat+0.055
Ben+0.053
 se+0.052
ila+0.052
 dads+0.051
Neg logit contributions
 unlike-0.113
 matter-0.098
 superpower-0.097
 blonde-0.097
 slightly-0.096
Token They
Token ID1119
Feature activation-2.84e-02

Pos logit contributions
 unlike+0.053
 blonde+0.048
 Tears+0.047
 superpower+0.046
 Nobody+0.046
Neg logit contributions
 se-0.043
ila-0.040
 bat-0.040
 dads-0.039
 next-0.038
Token g
Token ID308
Feature activation+5.22e-02

Pos logit contributions
ila+0.034
Ben+0.032
 pige+0.031
 dads+0.030
 next+0.030
Neg logit contributions
 unlike-0.051
 matter-0.050
 superpower-0.048
 blonde-0.048
 perspective-0.046
Tokeniggle
Token ID24082
Feature activation-6.93e-02

Pos logit contributions
 unlike+0.091
 superpower+0.078
 blonde+0.077
 perspective+0.076
 Tears+0.075
Neg logit contributions
Ben-0.079
 se-0.078
ila-0.077
 bat-0.076
 next-0.074
Token and
Token ID290
Feature activation-8.03e-02

Pos logit contributions
ila+0.049
Ben+0.046
 se+0.044
 bat+0.042
 Ben+0.040
Neg logit contributions
 unlike-0.175
 superpower-0.160
 matter-0.158
 perspective-0.157
 blonde-0.156
Token have
Token ID423
Feature activation-1.30e-01

Pos logit contributions
ila+0.110
Ben+0.102
iza+0.098
ily+0.097
 pige+0.096
Neg logit contributions
 unlike-0.133
 matter-0.131
 blonde-0.128
 superpower-0.125
 fascination-0.119
Token fun
Token ID1257
Feature activation+1.92e-02

Pos logit contributions
ila+0.178
Ben+0.173
 bat+0.169
 Ben+0.168
 se+0.168
Neg logit contributions
 unlike-0.240
 matter-0.218
 blonde-0.215
 superpower-0.214
 perspective-0.208
Token.
Token ID13
Feature activation+3.03e-02

Pos logit contributions
 unlike+0.044
 superpower+0.040
 perspective+0.040
 matter+0.040
 blonde+0.039
Neg logit contributions
Ben-0.018
 Ben-0.018
ila-0.018
 bat-0.017
 se-0.017
Token
Token ID198
Feature activation-1.62e-02

Pos logit contributions
 unlike+0.062
 blonde+0.056
 Tears+0.055
 superpower+0.055
 matter+0.054
Neg logit contributions
 se-0.036
ila-0.034
 bat-0.033
Ben-0.031
 slide-0.031
Token
Token ID198
Feature activation-1.58e-02

Pos logit contributions
 se+0.024
 bat+0.022
 wave+0.022
 dads+0.022
line+0.022
Neg logit contributions
 unlike-0.025
eway-0.023
 Tears-0.023
 Finally-0.022
 Before-0.022
TokenThe
Token ID464
Feature activation-2.10e-02

Pos logit contributions
 se+0.024
 bat+0.023
 wave+0.022
 goes+0.022
 dads+0.022
Neg logit contributions
 unlike-0.023
 slightly-0.021
 darkened-0.021
 total-0.021
 alas-0.021
Token go
Token ID467
Feature activation-1.94e-01

Pos logit contributions
 matter+0.105
 unlike+0.104
 superpower+0.098
 former+0.095
 Frankie+0.094
Neg logit contributions
ila-0.057
Ben-0.053
line-0.048
 Ben-0.047
 goes-0.046
Token on
Token ID319
Feature activation+7.81e-03

Pos logit contributions
Ben+0.296
ila+0.279
 Ben+0.266
Tom+0.247
 pige+0.245
Neg logit contributions
 unlike-0.318
 superpower-0.293
 matter-0.287
eway-0.274
 tactic-0.271
Token your
Token ID534
Feature activation-1.35e-02

Pos logit contributions
 unlike+0.016
 superpower+0.015
 matter+0.014
 perspective+0.014
 former+0.014
Neg logit contributions
ila-0.009
Ben-0.009
 se-0.008
line-0.008
 Ben-0.008
Token ship
Token ID4074
Feature activation-5.64e-02

Pos logit contributions
ila+0.017
Ben+0.016
 Ben+0.015
line+0.013
 se+0.013
Neg logit contributions
 unlike-0.027
 superpower-0.026
 matter-0.026
 tactic-0.025
 former-0.025
Token?"
Token ID1701
Feature activation-4.23e-02

Pos logit contributions
ila+0.064
Ben+0.063
 Ben+0.061
 se+0.060
 bat+0.059
Neg logit contributions
 unlike-0.118
 matter-0.108
 superpower-0.107
 tactic-0.103
 perspective-0.103
Token Tom
Token ID4186
Feature activation-1.75e-01

Pos logit contributions
 Ben+0.049
 bat+0.049
ila+0.049
Ben+0.049
 se+0.048
Neg logit contributions
 unlike-0.088
 matter-0.079
 superpower-0.079
 blonde-0.077
 perspective-0.077
Token agreed
Token ID4987
Feature activation-2.45e-02

Pos logit contributions
 Ben+0.149
 bat+0.147
ila+0.145
Ben+0.145
 se+0.142
Neg logit contributions
 unlike-0.419
 matter-0.384
 superpower-0.383
 blonde-0.374
 perspective-0.374
Token and
Token ID290
Feature activation-9.75e-02

Pos logit contributions
ila+0.026
Ben+0.024
 pige+0.022
 se+0.021
 goes+0.021
Neg logit contributions
 unlike-0.054
 tactic-0.049
 matter-0.049
 superpower-0.048
 Tears-0.047
Token they
Token ID484
Feature activation-1.22e-01

Pos logit contributions
Ben+0.094
ila+0.093
 pige+0.080
irts+0.078
 bat+0.076
Neg logit contributions
 unlike-0.211
 Frankie-0.193
 matter-0.190
 Mill-0.187
 superpower-0.186
Token went
Token ID1816
Feature activation-1.30e-01

Pos logit contributions
ila+0.153
Ben+0.144
 Ben+0.137
 pige+0.136
 bat+0.127
Neg logit contributions
 unlike-0.229
 matter-0.222
 superpower-0.216
 darkened-0.212
 blonde-0.210
Token on
Token ID319
Feature activation+8.06e-03

Pos logit contributions
Ben+0.204
ila+0.192
 Ben+0.184
 pige+0.183
 se+0.170
Neg logit contributions
 unlike-0.203
eway-0.196
 superpower-0.193
 tactic-0.184
 terribly-0.181