TOP ACTIVATIONS
MAX = 0.517

She f iddled and pro dd ed for a few minutes
. She poked and pro dd ed the thick skeleton with
. E dd ie was about to give
. It poked and pro dd ed at the trap until
sister !" E dd ie went immediately to his
fire work . <|endoftext|> E dd ie was a three year
um ma kissed him good night and urged him to go
bunny said , " Good night sheep ! I 'm gonna
the crib and said good night to her mom . She
He said good night to the moon before he
toys and hugged Mom good night . They went to their
and said , " Good night , Joe ."
her new doll a good night hug and put it away
bed and kissed her good night . The little
them . She said good night and went off to bed
are the sun 's good night kiss to the earth ."
. Then she said good night to the nos y little
, Jack y said good night to the skull and went
anymore . She said good night to the cat and went
it away . " Good night , Brown ie ," she

MIN ACTIVATIONS
MIN = -0.523

the pedals or the handle bars . She felt sad .
slowly pulled on the handle bars . He slowly rolled down
as sels on the handle bars and a bell that sounded
how to turn the handle bars and keep her balance .
blue wheels and pink handle bars . She liked to go
. He grabbed the handle bars with his little hands .
. He grabbed the handle bars and he quickly pushed on
twisted , and the handle bars were bent .
how to turn the handle bars . Finally
, held onto the handle bars carefully and slowly rolled over
the red flags . Both groups were very happy . They
She grabbed onto its handle bars and put her feet onto
let go of the handle bars . Suddenly , Joey heard
were sorting themselves into two groups . He was curious and
toys and put them in groups . He counted to zero
bird sat on the handle bars and held on tight .
they liked to live in groups and talk to each other
note attached to the handle bars . The note said ,
land , there were two groups of animals . They were
They were all organizing into groups and chatting .

INTERVAL 0.257 - 0.517
CONTAINS 1.734%

castle and started to eat it . He ate the sand
friend Sam . She put it in her backpack and went
, " Let 's hold it together ."
, they remembered how nice it was to share different things
and covered the city . It was so big that it

INTERVAL -0.003 - 0.257
CONTAINS 49.913%

had st ung her . Sam ran to her
They were very happy and danced together . But
farm , so he got a camera . He put the
said . Ben went near the chicken and reached
<|endoftext|> S ue was very excited ! Today she was going

INTERVAL -0.263 - -0.003
CONTAINS 47.790%

loved to slide down it every day . But one day
together . One day , the clown had a big game
. " Grab my hand , Ben . I
throw their balls to each other and catch them . They
's getting closer , let 's go ." Rob tremb led

INTERVAL -0.523 - -0.263
CONTAINS 0.564%

help you !" <|endoftext|> Once upon a time there was a
had not touched the or nam ents . They hoped Mom
She was scared at first , but then she saw
exciting adventure . <|endoftext|> Once upon a time , there was
by himself . <|endoftext|> Once upon a time , there was
Token She
Token ID1375
Feature activation+6.12e-02

Pos logit contributions
dd+0.064
night+0.052
ez+0.046
 pity+0.044
amb+0.044
Neg logit contributions
 groups-0.071
bars-0.070
 towns-0.063
 younger-0.062
 fighters-0.062
Token f
Token ID277
Feature activation-1.84e-01

Pos logit contributions
dd+0.106
night+0.087
ez+0.079
 pity+0.075
amb+0.074
Neg logit contributions
 groups-0.115
bars-0.112
 towns-0.100
 fighters-0.100
 younger-0.100
Tokeniddled
Token ID34897
Feature activation+7.35e-02

Pos logit contributions
 groups+0.358
bars+0.355
 fighters+0.313
 towns+0.312
 eight+0.310
Neg logit contributions
dd-0.307
night-0.255
ez-0.224
 pity-0.218
uggle-0.210
Token and
Token ID290
Feature activation+8.20e-02

Pos logit contributions
dd+0.124
night+0.103
ez+0.090
amb+0.088
 pity+0.087
Neg logit contributions
 groups-0.142
bars-0.137
 towns-0.124
 eight-0.124
 younger-0.122
Token pro
Token ID386
Feature activation+6.10e-02

Pos logit contributions
dd+0.147
night+0.125
ez+0.112
 pity+0.108
faced+0.104
Neg logit contributions
 groups-0.146
bars-0.144
 fighters-0.131
 department-0.128
ents-0.128
Tokendd
Token ID1860
Feature activation+5.17e-01

Pos logit contributions
dd+0.086
night+0.075
 pity+0.063
ez+0.062
uggle+0.061
Neg logit contributions
bars-0.127
 groups-0.127
 eight-0.114
ents-0.113
 towns-0.112
Tokened
Token ID276
Feature activation+2.24e-01

Pos logit contributions
dd+0.816
night+0.665
ez+0.577
-+0.546
amb+0.500
Neg logit contributions
 groups-1.107
bars-1.051
 younger-0.970
ents-0.954
 older-0.952
Token for
Token ID329
Feature activation+2.78e-02

Pos logit contributions
dd+0.379
night+0.316
ez+0.284
 pity+0.274
amb+0.259
Neg logit contributions
 groups-0.425
bars-0.420
 fighters-0.374
 towns-0.369
 younger-0.366
Token a
Token ID257
Feature activation+7.52e-02

Pos logit contributions
dd+0.044
night+0.034
ez+0.031
 pity+0.031
 it+0.029
Neg logit contributions
bars-0.057
 groups-0.055
 fighters-0.051
 department-0.050
lights-0.050
Token few
Token ID1178
Feature activation-9.30e-02

Pos logit contributions
dd+0.142
night+0.122
ez+0.108
 pity+0.106
faced+0.103
Neg logit contributions
 groups-0.128
bars-0.127
 towns-0.112
 fighters-0.111
ents-0.109
Token minutes
Token ID2431
Feature activation-9.45e-02

Pos logit contributions
bars+0.182
 groups+0.167
 fighters+0.161
ains+0.160
ents+0.159
Neg logit contributions
dd-0.155
night-0.131
ez-0.121
--0.120
 pity-0.118
Token.
Token ID13
Feature activation-3.36e-02

Pos logit contributions
dd+0.198
night+0.165
ez+0.150
amb+0.138
-+0.138
Neg logit contributions
 groups-0.221
bars-0.217
 towns-0.193
lights-0.193
 younger-0.191
Token She
Token ID1375
Feature activation-4.77e-03

Pos logit contributions
 groups+0.067
bars+0.063
lights+0.060
 fighters+0.060
 bridges+0.058
Neg logit contributions
dd-0.052
night-0.044
amb-0.039
ez-0.038
 Now-0.038
Token poked
Token ID47320
Feature activation+1.08e-01

Pos logit contributions
 groups+0.010
bars+0.009
 towns+0.008
 fighters+0.008
lights+0.008
Neg logit contributions
dd-0.008
night-0.006
ez-0.006
 pity-0.005
amb-0.005
Token and
Token ID290
Feature activation+5.78e-02

Pos logit contributions
dd+0.207
night+0.175
ez+0.156
uggle+0.152
 pity+0.151
Neg logit contributions
 groups-0.180
bars-0.179
 eight-0.157
 towns-0.157
 younger-0.154
Token pro
Token ID386
Feature activation+2.59e-02

Pos logit contributions
dd+0.097
night+0.078
ez+0.072
 pity+0.068
faced+0.065
Neg logit contributions
 groups-0.114
bars-0.110
lights-0.100
 fighters-0.099
 towns-0.099
Tokendd
Token ID1860
Feature activation+4.84e-01

Pos logit contributions
dd+0.015
night+0.014
 pity+0.009
faced+0.009
uggle+0.009
Neg logit contributions
bars-0.072
 groups-0.070
ents-0.066
 eight-0.065
 towns-0.065
Tokened
Token ID276
Feature activation+1.91e-01

Pos logit contributions
dd+0.497
night+0.346
ez+0.274
-+0.234
oo+0.207
Neg logit contributions
 groups-1.302
bars-1.261
 younger-1.179
 older-1.156
lights-1.156
Token the
Token ID262
Feature activation-1.40e-01

Pos logit contributions
dd+0.330
night+0.275
ez+0.249
 pity+0.239
amb+0.229
Neg logit contributions
 groups-0.360
bars-0.353
 fighters-0.315
 towns-0.311
 younger-0.309
Token thick
Token ID6546
Feature activation-7.76e-02

Pos logit contributions
 groups+0.273
bars+0.269
 towns+0.244
 eight+0.242
 younger+0.241
Neg logit contributions
dd-0.233
night-0.188
ez-0.169
amb-0.160
 pity-0.159
Token skeleton
Token ID18328
Feature activation+1.02e-01

Pos logit contributions
 groups+0.149
bars+0.146
 towns+0.132
 eight+0.131
 younger+0.129
Neg logit contributions
dd-0.133
night-0.107
ez-0.096
amb-0.093
 pity-0.092
Token with
Token ID351
Feature activation+8.54e-02

Pos logit contributions
dd+0.152
night+0.125
-+0.117
ez+0.117
ggy+0.111
Neg logit contributions
 groups-0.211
bars-0.198
 towns-0.184
lights-0.183
 younger-0.182
Token.
Token ID13
Feature activation-3.98e-02

Pos logit contributions
dd+0.044
night+0.034
-+0.032
 it+0.029
 inside+0.028
Neg logit contributions
 groups-0.069
bars-0.064
 towns-0.063
 eight-0.061
 younger-0.060
Token
Token ID220
Feature activation+6.50e-03

Pos logit contributions
 groups+0.080
bars+0.079
 fighters+0.071
 towns+0.071
 eight+0.070
Neg logit contributions
dd-0.063
night-0.051
ez-0.045
 pity-0.043
amb-0.043
Token
Token ID198
Feature activation+4.16e-02

Pos logit contributions
dd+0.007
night+0.005
ez+0.004
uggle+0.004
oo+0.004
Neg logit contributions
 groups-0.016
bars-0.015
 towns-0.014
 eight-0.014
 fighters-0.014
Token
Token ID198
Feature activation+4.06e-02

Pos logit contributions
dd+0.059
night+0.043
ez+0.038
amb+0.037
oo+0.036
Neg logit contributions
 groups-0.093
bars-0.092
 fighters-0.084
 younger-0.083
 department-0.083
TokenE
Token ID36
Feature activation+9.83e-02

Pos logit contributions
dd+0.080
night+0.062
ez+0.058
amb+0.055
oo+0.055
Neg logit contributions
bars-0.071
 groups-0.071
 younger-0.062
 fighters-0.062
 department-0.062
Tokendd
Token ID1860
Feature activation+4.83e-01

Pos logit contributions
dd+0.037
night+0.015
ez+0.003
 pity+0.000
faced-0.005
Neg logit contributions
 groups-0.304
bars-0.303
 fighters-0.285
 towns-0.284
 eight-0.280
Tokenie
Token ID494
Feature activation-3.57e-03

Pos logit contributions
dd+0.693
night+0.549
ez+0.452
-+0.430
oo+0.416
Neg logit contributions
 groups-1.074
bars-1.070
 fastest-0.985
 younger-0.969
 thirteen-0.954
Token was
Token ID373
Feature activation+2.60e-02

Pos logit contributions
 groups+0.007
bars+0.007
lights+0.006
 fighters+0.006
 towns+0.006
Neg logit contributions
dd-0.006
night-0.005
ez-0.005
 pity-0.004
faced-0.004
Token about
Token ID546
Feature activation+8.77e-02

Pos logit contributions
dd+0.044
night+0.036
ez+0.034
 pity+0.032
faced+0.030
Neg logit contributions
bars-0.050
 groups-0.049
lights-0.044
 fighters-0.044
 towns-0.043
Token to
Token ID284
Feature activation-2.74e-02

Pos logit contributions
dd+0.160
night+0.134
ez+0.121
 pity+0.116
amb+0.114
Neg logit contributions
 groups-0.157
bars-0.155
 towns-0.136
 fighters-0.136
 younger-0.135
Token give
Token ID1577
Feature activation-5.43e-03

Pos logit contributions
 groups+0.049
bars+0.049
 fighters+0.043
 towns+0.043
 younger+0.042
Neg logit contributions
dd-0.050
night-0.042
ez-0.038
 pity-0.036
amb-0.035
Token.
Token ID13
Feature activation-3.21e-02

Pos logit contributions
dd+0.011
night+0.009
ez+0.008
 pity+0.008
faced+0.008
Neg logit contributions
bars-0.011
 groups-0.011
 fighters-0.009
 towns-0.009
lights-0.009
Token It
Token ID632
Feature activation+2.76e-01

Pos logit contributions
bars+0.065
 groups+0.065
lights+0.058
 fighters+0.057
 fastest+0.056
Neg logit contributions
dd-0.050
night-0.042
ez-0.037
 pity-0.035
 Now-0.035
Token poked
Token ID47320
Feature activation+1.03e-01

Pos logit contributions
dd+0.430
night+0.334
ez+0.306
 pity+0.295
faced+0.288
Neg logit contributions
bars-0.593
 groups-0.577
lights-0.516
 towns-0.509
 bridges-0.504
Token and
Token ID290
Feature activation+5.70e-02

Pos logit contributions
dd+0.182
night+0.151
ez+0.135
amb+0.129
 pity+0.129
Neg logit contributions
 groups-0.191
bars-0.183
 eight-0.167
 towns-0.167
 fighters-0.164
Token pro
Token ID386
Feature activation+2.14e-02

Pos logit contributions
dd+0.098
night+0.079
ez+0.072
 pity+0.069
faced+0.067
Neg logit contributions
bars-0.112
 groups-0.111
lights-0.097
 towns-0.096
 younger-0.095
Tokendd
Token ID1860
Feature activation+4.81e-01

Pos logit contributions
night+0.008
dd+0.006
uggle+0.004
 coincidence+0.004
 pity+0.004
Neg logit contributions
 groups-0.062
bars-0.060
 eight-0.059
ents-0.059
 towns-0.057
Tokened
Token ID276
Feature activation+1.90e-01

Pos logit contributions
dd+0.514
night+0.352
ez+0.282
-+0.246
oo+0.217
Neg logit contributions
 groups-1.307
bars-1.277
 younger-1.181
 older-1.162
lights-1.149
Token at
Token ID379
Feature activation+1.09e-01

Pos logit contributions
dd+0.307
night+0.250
ez+0.224
 pity+0.216
amb+0.204
Neg logit contributions
 groups-0.382
bars-0.381
 fighters-0.337
 younger-0.334
 towns-0.334
Token the
Token ID262
Feature activation-1.45e-01

Pos logit contributions
dd+0.196
night+0.160
ez+0.149
 pity+0.146
oo+0.140
Neg logit contributions
bars-0.200
 groups-0.193
lights-0.170
 fighters-0.169
 department-0.167
Token trap
Token ID12840
Feature activation+1.57e-01

Pos logit contributions
 groups+0.295
bars+0.284
 towns+0.259
 fighters+0.259
 eight+0.258
Neg logit contributions
dd-0.234
night-0.190
ez-0.170
amb-0.160
 pity-0.159
Token until
Token ID1566
Feature activation-8.10e-02

Pos logit contributions
dd+0.230
night+0.192
ez+0.169
 it+0.167
-+0.166
Neg logit contributions
bars-0.326
 groups-0.319
 towns-0.290
 fighters-0.285
lights-0.284
Token sister
Token ID6621
Feature activation-1.63e-01

Pos logit contributions
dd+0.037
night+0.031
ez+0.028
 pity+0.027
amb+0.027
Neg logit contributions
 groups-0.032
bars-0.031
 towns-0.027
 fighters-0.027
 younger-0.027
Token!"
Token ID2474
Feature activation+5.06e-02

Pos logit contributions
bars+0.324
 groups+0.310
lights+0.286
 narrower+0.283
 towns+0.280
Neg logit contributions
dd-0.248
night-0.224
ez-0.197
--0.187
amb-0.181
Token
Token ID198
Feature activation+3.26e-02

Pos logit contributions
dd+0.093
night+0.077
ez+0.072
 pity+0.069
faced+0.067
Neg logit contributions
bars-0.088
 groups-0.088
 eight-0.076
 towns-0.076
 younger-0.076
Token
Token ID198
Feature activation+3.20e-02

Pos logit contributions
dd+0.061
night+0.055
ez+0.049
 pity+0.048
faced+0.047
Neg logit contributions
 groups-0.053
bars-0.051
 eight-0.046
 towns-0.045
 younger-0.045
TokenE
Token ID36
Feature activation+9.00e-02

Pos logit contributions
dd+0.060
night+0.050
ez+0.045
 pity+0.043
amb+0.043
Neg logit contributions
 groups-0.056
bars-0.055
 towns-0.049
 fighters-0.049
 younger-0.048
Tokendd
Token ID1860
Feature activation+4.74e-01

Pos logit contributions
dd+0.075
night+0.055
ez+0.046
 pity+0.044
faced+0.039
Neg logit contributions
 groups-0.232
bars-0.231
 fighters-0.217
 towns-0.214
 eight-0.213
Tokenie
Token ID494
Feature activation-1.57e-02

Pos logit contributions
dd+0.717
night+0.604
ez+0.501
oo+0.479
-+0.478
Neg logit contributions
 groups-1.015
bars-1.003
 fastest-0.935
 towns-0.912
 bridges-0.909
Token went
Token ID1816
Feature activation+3.60e-02

Pos logit contributions
 groups+0.030
bars+0.029
 towns+0.026
 fighters+0.026
lights+0.026
Neg logit contributions
dd-0.027
night-0.022
ez-0.020
 pity-0.019
amb-0.019
Token immediately
Token ID3393
Feature activation+3.44e-02

Pos logit contributions
dd+0.068
night+0.055
ez+0.052
amb+0.048
 pity+0.048
Neg logit contributions
 groups-0.064
bars-0.062
 older-0.057
 younger-0.056
 towns-0.056
Token to
Token ID284
Feature activation-3.54e-02

Pos logit contributions
dd+0.049
night+0.041
ez+0.035
 pity+0.034
-+0.033
Neg logit contributions
bars-0.073
 groups-0.072
 fighters-0.065
lights-0.065
ains-0.064
Token his
Token ID465
Feature activation-1.11e-02

Pos logit contributions
 groups+0.059
bars+0.055
 younger+0.051
 older+0.051
 towns+0.050
Neg logit contributions
dd-0.073
night-0.060
ez-0.056
amb-0.054
faced-0.052
Token fire
Token ID2046
Feature activation-1.94e-02

Pos logit contributions
 groups+0.254
bars+0.243
 towns+0.219
 fighters+0.218
 younger+0.216
Neg logit contributions
dd-0.296
night-0.248
ez-0.231
amb-0.221
 pity-0.219
Tokenwork
Token ID1818
Feature activation-1.58e-01

Pos logit contributions
 groups+0.035
bars+0.034
ents+0.030
 eight+0.030
 narrower+0.029
Neg logit contributions
dd-0.033
night-0.031
--0.028
ez-0.026
ake-0.026
Token.
Token ID13
Feature activation-5.08e-02

Pos logit contributions
bars+0.308
 groups+0.303
 towns+0.270
 eight+0.269
 fighters+0.269
Neg logit contributions
dd-0.246
night-0.224
ez-0.187
--0.182
uggle-0.180
Token<|endoftext|>
Token ID50256
Feature activation+1.80e-01

Pos logit contributions
 groups+0.087
bars+0.087
 eight+0.078
 towns+0.078
 younger+0.076
Neg logit contributions
dd-0.096
night-0.076
ez-0.072
 pity-0.070
amb-0.067
TokenE
Token ID36
Feature activation+8.55e-02

Pos logit contributions
dd+0.356
night+0.300
 pity+0.266
ez+0.266
amb+0.263
Neg logit contributions
 groups-0.299
bars-0.278
 towns-0.261
 eight-0.252
ill-0.248
Tokendd
Token ID1860
Feature activation+4.68e-01

Pos logit contributions
dd+0.107
night+0.088
 pity+0.083
ez+0.082
faced+0.074
Neg logit contributions
bars-0.178
 groups-0.178
 fighters-0.166
 eight-0.166
 towns-0.161
Tokenie
Token ID494
Feature activation-1.63e-02

Pos logit contributions
dd+0.655
night+0.559
ez+0.444
oo+0.435
-+0.431
Neg logit contributions
bars-1.026
 groups-1.013
 fastest-0.948
 younger-0.927
 thirteen-0.910
Token was
Token ID373
Feature activation+1.56e-02

Pos logit contributions
 groups+0.028
bars+0.027
 eight+0.025
 towns+0.025
 older+0.024
Neg logit contributions
dd-0.030
night-0.024
 pity-0.022
ez-0.022
faced-0.022
Token a
Token ID257
Feature activation+3.49e-02

Pos logit contributions
dd+0.029
night+0.025
ez+0.023
amb+0.022
uggle+0.022
Neg logit contributions
 groups-0.027
 older-0.025
 younger-0.025
bars-0.024
 eight-0.024
Token three
Token ID1115
Feature activation-2.03e-01

Pos logit contributions
dd+0.073
night+0.062
ez+0.060
amb+0.056
 pity+0.055
Neg logit contributions
 groups-0.053
bars-0.050
 younger-0.046
 older-0.046
 eight-0.046
Token year
Token ID614
Feature activation-1.10e-02

Pos logit contributions
 groups+0.427
bars+0.422
 towns+0.378
 fighters+0.378
 younger+0.376
Neg logit contributions
dd-0.306
night-0.246
ez-0.217
 pity-0.204
amb-0.200
Tokenum
Token ID388
Feature activation+1.43e-01

Pos logit contributions
 groups+0.312
bars+0.283
 towns+0.267
 younger+0.261
 older+0.260
Neg logit contributions
dd-0.343
night-0.287
amb-0.268
ez-0.259
oo-0.251
Tokenma
Token ID2611
Feature activation-7.52e-02

Pos logit contributions
dd+0.202
night+0.167
ez+0.142
-+0.132
amb+0.129
Neg logit contributions
 groups-0.313
bars-0.305
 towns-0.280
 fastest-0.278
 fighters-0.276
Token kissed
Token ID27946
Feature activation+8.85e-02

Pos logit contributions
 groups+0.153
bars+0.148
 towns+0.135
lights+0.134
 fastest+0.133
Neg logit contributions
dd-0.117
night-0.099
ez-0.089
--0.081
 pity-0.080
Token him
Token ID683
Feature activation+2.39e-01

Pos logit contributions
dd+0.133
night+0.109
ez+0.094
 pity+0.089
amb+0.086
Neg logit contributions
 groups-0.187
bars-0.184
 towns-0.166
 fighters-0.165
 younger-0.164
Token good
Token ID922
Feature activation+1.38e-01

Pos logit contributions
dd+0.399
night+0.334
ez+0.298
 pity+0.282
amb+0.275
Neg logit contributions
 groups-0.461
bars-0.453
 towns-0.405
 fighters-0.404
 eight-0.399
Tokennight
Token ID3847
Feature activation+4.28e-01

Pos logit contributions
dd+0.240
night+0.187
ez+0.177
amb+0.168
 pity+0.167
Neg logit contributions
 groups-0.256
bars-0.253
 fighters-0.227
 younger-0.225
 towns-0.225
Token and
Token ID290
Feature activation+9.22e-02

Pos logit contributions
dd+0.722
night+0.613
ez+0.538
 pity+0.505
amb+0.502
Neg logit contributions
 groups-0.824
bars-0.812
 towns-0.723
 fighters-0.717
 younger-0.709
Token urged
Token ID11643
Feature activation-2.50e-03

Pos logit contributions
dd+0.162
night+0.133
ez+0.121
 pity+0.115
amb+0.115
Neg logit contributions
 groups-0.170
bars-0.168
 towns-0.149
 fighters-0.149
 younger-0.148
Token him
Token ID683
Feature activation+2.26e-01

Pos logit contributions
 groups+0.005
bars+0.005
 fighters+0.004
 towns+0.004
 younger+0.004
Neg logit contributions
dd-0.004
night-0.003
ez-0.003
 pity-0.003
amb-0.003
Token to
Token ID284
Feature activation+5.30e-03

Pos logit contributions
dd+0.359
night+0.317
ez+0.259
ggy+0.252
 pity+0.251
Neg logit contributions
 groups-0.461
bars-0.443
 towns-0.398
lights-0.396
ains-0.392
Token go
Token ID467
Feature activation+3.22e-02

Pos logit contributions
dd+0.010
night+0.008
ez+0.007
 pity+0.007
amb+0.007
Neg logit contributions
 groups-0.010
bars-0.009
 towns-0.008
 fighters-0.008
 younger-0.008
Token bunny
Token ID44915
Feature activation+2.19e-01

Pos logit contributions
dd+0.035
night+0.030
 pity+0.025
-+0.024
oo+0.024
Neg logit contributions
bars-0.063
 groups-0.061
lights-0.055
ents-0.054
 towns-0.054
Token said
Token ID531
Feature activation+1.31e-01

Pos logit contributions
dd+0.378
night+0.311
ez+0.280
 pity+0.269
amb+0.263
Neg logit contributions
 groups-0.410
bars-0.401
 towns-0.358
 fighters-0.357
 eight-0.355
Token,
Token ID11
Feature activation+5.43e-02

Pos logit contributions
dd+0.251
night+0.208
ez+0.192
 pity+0.187
uggle+0.186
Neg logit contributions
 groups-0.216
bars-0.209
 eight-0.191
 fighters-0.188
 younger-0.188
Token "
Token ID366
Feature activation-5.81e-02

Pos logit contributions
dd+0.109
night+0.094
ez+0.086
 pity+0.083
amb+0.082
Neg logit contributions
 groups-0.086
bars-0.085
 towns-0.073
 fighters-0.072
lights-0.072
TokenGood
Token ID10248
Feature activation+2.77e-02

Pos logit contributions
 groups+0.102
bars+0.102
 towns+0.090
 eight+0.090
 fighters+0.090
Neg logit contributions
dd-0.104
night-0.089
ez-0.079
 pity-0.077
uggle-0.075
Tokennight
Token ID3847
Feature activation+4.27e-01

Pos logit contributions
dd+0.040
ez+0.031
to+0.027
faced+0.026
uggle+0.025
Neg logit contributions
 fastest-0.052
 groups-0.052
bars-0.051
 fighters-0.051
 older-0.051
Token sheep
Token ID15900
Feature activation+4.43e-02

Pos logit contributions
dd+0.761
night+0.622
ez+0.573
 pity+0.550
amb+0.537
Neg logit contributions
 groups-0.779
bars-0.754
 fighters-0.680
 towns-0.678
 eight-0.678
Token!
Token ID0
Feature activation+5.77e-02

Pos logit contributions
dd+0.073
night+0.062
ez+0.054
 pity+0.051
amb+0.051
Neg logit contributions
 groups-0.088
bars-0.087
lights-0.077
 towns-0.077
ents-0.076
Token I
Token ID314
Feature activation-1.07e-01

Pos logit contributions
dd+0.107
night+0.089
ez+0.081
 pity+0.078
amb+0.076
Neg logit contributions
 groups-0.101
bars-0.100
 towns-0.088
 fighters-0.088
 eight-0.087
Token'm
Token ID1101
Feature activation+6.94e-02

Pos logit contributions
 groups+0.169
bars+0.158
 department+0.148
 eight+0.147
 fighters+0.146
Neg logit contributions
dd-0.204
night-0.165
faced-0.151
ez-0.151
amb-0.150
Token gonna
Token ID8066
Feature activation+2.34e-02

Pos logit contributions
dd+0.122
night+0.104
ez+0.093
amb+0.091
 pity+0.089
Neg logit contributions
 groups-0.125
bars-0.119
 older-0.113
 younger-0.113
 eight-0.110
Token the
Token ID262
Feature activation-1.21e-01

Pos logit contributions
bars+0.331
 groups+0.316
 fighters+0.287
 department+0.282
 fastest+0.279
Neg logit contributions
dd-0.365
night-0.305
ez-0.287
 pity-0.283
 it-0.261
Token crib
Token ID48083
Feature activation+6.72e-02

Pos logit contributions
 groups+0.202
bars+0.198
 towns+0.173
 fighters+0.172
 younger+0.171
Neg logit contributions
dd-0.236
night-0.200
ez-0.182
 pity-0.175
amb-0.173
Token and
Token ID290
Feature activation+7.52e-02

Pos logit contributions
dd+0.067
night+0.066
-+0.061
 it+0.051
amb+0.049
Neg logit contributions
bars-0.152
 groups-0.148
ains-0.143
 narrower-0.141
 thirteen-0.138
Token said
Token ID531
Feature activation+9.52e-02

Pos logit contributions
dd+0.141
night+0.117
ez+0.108
 pity+0.102
amb+0.102
Neg logit contributions
 groups-0.131
bars-0.129
 towns-0.114
 fighters-0.113
 younger-0.113
Token good
Token ID922
Feature activation+1.14e-01

Pos logit contributions
dd+0.169
night+0.139
ez+0.129
 pity+0.122
amb+0.120
Neg logit contributions
 groups-0.174
bars-0.170
 eight-0.154
 towns-0.152
 younger-0.152
Tokennight
Token ID3847
Feature activation+4.21e-01

Pos logit contributions
dd+0.146
ez+0.099
faced+0.088
uggle+0.080
ggy+0.076
Neg logit contributions
bars-0.250
 older-0.243
 department-0.243
 fighters-0.242
 younger-0.238
Token to
Token ID284
Feature activation-7.36e-03

Pos logit contributions
dd+0.768
night+0.652
ez+0.580
 pity+0.555
amb+0.549
Neg logit contributions
 groups-0.758
bars-0.744
 towns-0.656
 fighters-0.650
 younger-0.644
Token her
Token ID607
Feature activation+1.22e-01

Pos logit contributions
 groups+0.013
bars+0.013
 towns+0.011
 fighters+0.011
 younger+0.011
Neg logit contributions
dd-0.014
night-0.012
ez-0.010
 pity-0.010
amb-0.010
Token mom
Token ID1995
Feature activation-6.90e-03

Pos logit contributions
dd+0.214
night+0.186
ez+0.163
 pity+0.159
amb+0.155
Neg logit contributions
 groups-0.221
bars-0.220
 towns-0.193
 fighters-0.192
 eight-0.190
Token.
Token ID13
Feature activation-1.81e-02

Pos logit contributions
 groups+0.015
bars+0.014
 towns+0.013
 narrower+0.013
lights+0.013
Neg logit contributions
dd-0.010
night-0.009
ez-0.007
--0.007
amb-0.007
Token She
Token ID1375
Feature activation+1.10e-02

Pos logit contributions
 groups+0.035
bars+0.035
 towns+0.031
 fighters+0.031
 younger+0.030
Neg logit contributions
dd-0.030
night-0.025
ez-0.022
amb-0.021
 pity-0.021
Token
Token ID198
Feature activation+7.00e-02

Pos logit contributions
dd+0.010
night+0.008
ez+0.008
 pity+0.007
amb+0.007
Neg logit contributions
 groups-0.012
bars-0.012
 towns-0.011
 fighters-0.011
 younger-0.011
Token
Token ID198
Feature activation+6.46e-02

Pos logit contributions
dd+0.108
night+0.095
 pity+0.077
amb+0.076
ez+0.075
Neg logit contributions
 groups-0.143
bars-0.129
 eight-0.128
 towns-0.126
ents-0.124
TokenHe
Token ID1544
Feature activation-9.38e-02

Pos logit contributions
dd+0.149
night+0.133
ez+0.122
 pity+0.120
amb+0.116
Neg logit contributions
 groups-0.084
bars-0.079
 towns-0.069
 eight-0.068
 fighters-0.066
Token said
Token ID531
Feature activation+9.33e-02

Pos logit contributions
 groups+0.176
bars+0.173
 towns+0.153
 fighters+0.153
 younger+0.152
Neg logit contributions
dd-0.163
night-0.136
ez-0.122
 pity-0.117
amb-0.114
Token good
Token ID922
Feature activation+1.16e-01

Pos logit contributions
dd+0.161
night+0.131
ez+0.122
 pity+0.115
amb+0.113
Neg logit contributions
 groups-0.175
bars-0.171
 eight-0.156
 towns-0.154
 fighters-0.154
Tokennight
Token ID3847
Feature activation+4.21e-01

Pos logit contributions
dd+0.191
ez+0.138
faced+0.126
ggy+0.122
uggle+0.120
Neg logit contributions
bars-0.218
 fighters-0.216
 department-0.214
 older-0.214
 towns-0.212
Token to
Token ID284
Feature activation-5.65e-03

Pos logit contributions
dd+0.750
night+0.613
ez+0.564
 pity+0.545
amb+0.527
Neg logit contributions
 groups-0.765
bars-0.750
 fighters-0.669
 towns-0.667
 eight-0.666
Token the
Token ID262
Feature activation-1.22e-01

Pos logit contributions
 groups+0.010
bars+0.010
 fighters+0.009
 towns+0.009
 department+0.009
Neg logit contributions
dd-0.010
night-0.009
ez-0.008
 pity-0.008
amb-0.007
Token moon
Token ID8824
Feature activation-9.95e-02

Pos logit contributions
 groups+0.221
bars+0.215
 towns+0.193
 younger+0.192
 fighters+0.191
Neg logit contributions
dd-0.222
night-0.184
ez-0.168
amb-0.159
 pity-0.159
Token before
Token ID878
Feature activation-2.77e-01

Pos logit contributions
 groups+0.209
bars+0.201
 towns+0.183
 eight+0.182
 younger+0.181
Neg logit contributions
dd-0.143
night-0.130
ez-0.109
--0.104
amb-0.101
Token he
Token ID339
Feature activation-2.32e-02

Pos logit contributions
 groups+0.519
bars+0.515
 towns+0.450
 fighters+0.448
 younger+0.448
Neg logit contributions
dd-0.491
night-0.407
ez-0.361
 pity-0.348
oo-0.337
Token toys
Token ID14958
Feature activation-6.63e-03

Pos logit contributions
 groups+0.046
bars+0.045
 towns+0.040
 younger+0.040
 eight+0.039
Neg logit contributions
dd-0.048
night-0.040
ez-0.036
amb-0.035
 pity-0.034
Token and
Token ID290
Feature activation+8.10e-02

Pos logit contributions
 groups+0.015
bars+0.015
 eight+0.014
 towns+0.014
 younger+0.014
Neg logit contributions
dd-0.009
night-0.007
--0.006
ez-0.005
uggle-0.005
Token hugged
Token ID41130
Feature activation+1.10e-01

Pos logit contributions
dd+0.145
night+0.121
ez+0.109
 pity+0.104
amb+0.103
Neg logit contributions
 groups-0.148
bars-0.145
 towns-0.129
 fighters-0.128
 younger-0.127
Token Mom
Token ID11254
Feature activation-1.27e-02

Pos logit contributions
dd+0.240
night+0.199
ez+0.189
faced+0.183
amb+0.181
Neg logit contributions
 groups-0.158
bars-0.148
 towns-0.135
 fighters-0.129
lights-0.129
Token good
Token ID922
Feature activation+1.22e-01

Pos logit contributions
 groups+0.027
bars+0.026
 eight+0.024
 towns+0.024
 fighters+0.024
Neg logit contributions
dd-0.017
night-0.016
ez-0.014
--0.013
amb-0.012
Tokennight
Token ID3847
Feature activation+4.20e-01

Pos logit contributions
dd+0.119
ggy+0.073
faced+0.067
ez+0.067
amb+0.063
Neg logit contributions
bars-0.294
 groups-0.289
 directions-0.288
 department-0.286
 fighters-0.284
Token.
Token ID13
Feature activation-1.31e-02

Pos logit contributions
dd+0.802
night+0.673
ez+0.617
 pity+0.594
amb+0.583
Neg logit contributions
 groups-0.716
bars-0.700
 towns-0.616
 fighters-0.615
 younger-0.610
Token They
Token ID1119
Feature activation-3.66e-02

Pos logit contributions
 groups+0.023
bars+0.022
 towns+0.020
 fighters+0.020
 younger+0.019
Neg logit contributions
dd-0.025
night-0.021
ez-0.019
 pity-0.018
amb-0.018
Token went
Token ID1816
Feature activation+6.31e-02

Pos logit contributions
 groups+0.069
bars+0.068
 towns+0.060
 fighters+0.060
 younger+0.060
Neg logit contributions
dd-0.063
night-0.053
ez-0.047
 pity-0.045
amb-0.044
Token to
Token ID284
Feature activation-1.25e-02

Pos logit contributions
dd+0.124
night+0.098
ez+0.096
 pity+0.090
oof+0.088
Neg logit contributions
 groups-0.107
bars-0.103
 older-0.095
 towns-0.093
 younger-0.092
Token their
Token ID511
Feature activation-4.29e-02

Pos logit contributions
 groups+0.022
bars+0.021
 younger+0.020
 older+0.019
 eight+0.019
Neg logit contributions
dd-0.024
night-0.019
ez-0.018
amb-0.017
faced-0.017
Token and
Token ID290
Feature activation+7.80e-02

Pos logit contributions
dd+0.174
night+0.142
ez+0.125
 pity+0.118
amb+0.116
Neg logit contributions
 groups-0.230
bars-0.225
 towns-0.203
 fighters-0.202
 younger-0.201
Token said
Token ID531
Feature activation+9.24e-02

Pos logit contributions
dd+0.133
night+0.110
ez+0.099
 pity+0.094
amb+0.094
Neg logit contributions
 groups-0.149
bars-0.146
 towns-0.130
 fighters-0.130
 younger-0.129
Token,
Token ID11
Feature activation+2.36e-02

Pos logit contributions
dd+0.166
night+0.139
ez+0.128
 pity+0.122
amb+0.122
Neg logit contributions
 groups-0.164
bars-0.162
 eight-0.144
 fighters-0.143
 towns-0.143
Token "
Token ID366
Feature activation-8.27e-02

Pos logit contributions
dd+0.042
night+0.035
ez+0.031
amb+0.030
 it+0.030
Neg logit contributions
 groups-0.043
bars-0.042
 towns-0.037
 fastest-0.037
lights-0.037
TokenGood
Token ID10248
Feature activation+1.50e-02

Pos logit contributions
bars+0.157
 groups+0.156
 towns+0.138
 fighters+0.138
 eight+0.136
Neg logit contributions
dd-0.139
night-0.117
ez-0.103
 pity-0.100
uggle-0.097
Tokennight
Token ID3847
Feature activation+4.20e-01

Pos logit contributions
dd+0.020
ez+0.014
ggy+0.012
faced+0.012
 pity+0.011
Neg logit contributions
bars-0.033
 groups-0.032
 older-0.032
 younger-0.031
 fighters-0.031
Token,
Token ID11
Feature activation+1.99e-02

Pos logit contributions
dd+0.702
night+0.598
ez+0.516
 pity+0.493
amb+0.487
Neg logit contributions
 groups-0.808
bars-0.805
 towns-0.705
 fighters-0.696
ents-0.693
Token Joe
Token ID5689
Feature activation-5.36e-03

Pos logit contributions
dd+0.032
night+0.027
ez+0.023
 pity+0.023
amb+0.022
Neg logit contributions
bars-0.039
 groups-0.039
 towns-0.034
 fighters-0.034
lights-0.034
Token."
Token ID526
Feature activation-6.77e-02

Pos logit contributions
 groups+0.011
bars+0.011
lights+0.010
 towns+0.009
ents+0.009
Neg logit contributions
dd-0.009
night-0.008
ez-0.007
--0.006
oo-0.006
Token
Token ID198
Feature activation+5.77e-02

Pos logit contributions
 groups+0.134
bars+0.132
 towns+0.118
 fighters+0.117
 younger+0.117
Neg logit contributions
dd-0.111
night-0.092
ez-0.082
 pity-0.077
amb-0.077
Token
Token ID198
Feature activation+5.45e-02

Pos logit contributions
dd+0.098
night+0.089
 pity+0.076
ez+0.075
faced+0.072
Neg logit contributions
 groups-0.105
bars-0.099
 eight-0.093
 towns-0.092
ents-0.090
Token her
Token ID607
Feature activation+1.35e-01

Pos logit contributions
 groups+0.030
bars+0.029
 older+0.028
 younger+0.027
 eight+0.027
Neg logit contributions
dd-0.031
night-0.025
ez-0.023
amb-0.023
ager-0.022
Token new
Token ID649
Feature activation-2.13e-02

Pos logit contributions
dd+0.212
night+0.192
ez+0.165
 pity+0.164
faced+0.152
Neg logit contributions
 groups-0.256
bars-0.255
 fighters-0.230
 towns-0.227
 fastest-0.226
Token doll
Token ID3654
Feature activation+1.74e-01

Pos logit contributions
 groups+0.041
bars+0.040
 younger+0.036
 older+0.036
 towns+0.036
Neg logit contributions
dd-0.038
night-0.030
ez-0.027
amb-0.026
 pity-0.026
Token a
Token ID257
Feature activation+7.46e-02

Pos logit contributions
dd+0.272
night+0.233
ez+0.201
 pity+0.192
faced+0.188
Neg logit contributions
 groups-0.349
bars-0.346
 towns-0.307
 fighters-0.306
 younger-0.303
Token good
Token ID922
Feature activation+1.22e-01

Pos logit contributions
dd+0.152
night+0.114
ez+0.114
amb+0.110
oo+0.103
Neg logit contributions
 groups-0.126
bars-0.123
 younger-0.117
 older-0.115
 eight-0.115
Tokennight
Token ID3847
Feature activation+4.18e-01

Pos logit contributions
dd+0.262
ez+0.181
amb+0.181
cha+0.170
to+0.164
Neg logit contributions
 older-0.178
 younger-0.175
 groups-0.174
 department-0.169
 size-0.169
Token hug
Token ID16225
Feature activation+1.13e-01

Pos logit contributions
dd+0.776
night+0.600
ez+0.575
 pity+0.561
faced+0.551
Neg logit contributions
 groups-0.727
bars-0.709
 eight-0.642
 fighters-0.638
 younger-0.636
Token and
Token ID290
Feature activation+8.14e-02

Pos logit contributions
dd+0.196
night+0.160
ez+0.145
 pity+0.139
amb+0.136
Neg logit contributions
 groups-0.212
bars-0.209
 towns-0.186
 fighters-0.186
 younger-0.184
Token put
Token ID1234
Feature activation+2.34e-02

Pos logit contributions
dd+0.151
night+0.118
ez+0.117
amb+0.112
faced+0.110
Neg logit contributions
 groups-0.138
bars-0.137
 younger-0.123
 fighters-0.122
 towns-0.121
Token it
Token ID340
Feature activation+3.18e-01

Pos logit contributions
dd+0.044
amb+0.034
faced+0.034
night+0.033
ggy+0.033
Neg logit contributions
bars-0.038
 groups-0.037
 older-0.034
 forks-0.034
 directions-0.034
Token away
Token ID1497
Feature activation+9.54e-02

Pos logit contributions
dd+0.474
night+0.398
ez+0.336
 pity+0.315
uggle+0.311
Neg logit contributions
 groups-0.674
bars-0.660
 fighters-0.599
 towns-0.598
 younger-0.589
Token bed
Token ID3996
Feature activation+1.55e-02

Pos logit contributions
dd+0.217
night+0.188
ez+0.163
 pity+0.158
uggle+0.154
Neg logit contributions
 groups-0.237
bars-0.234
 fighters-0.210
 towns-0.208
 eight-0.205
Token and
Token ID290
Feature activation+7.66e-02

Pos logit contributions
dd+0.021
night+0.018
-+0.015
ez+0.013
uggle+0.013
Neg logit contributions
 groups-0.034
bars-0.033
 towns-0.031
 bridges-0.031
 fighters-0.031
Token kissed
Token ID27946
Feature activation+5.27e-02

Pos logit contributions
dd+0.139
night+0.114
ez+0.106
amb+0.102
 pity+0.100
Neg logit contributions
 groups-0.137
bars-0.135
 younger-0.119
 fighters-0.119
 towns-0.118
Token her
Token ID607
Feature activation+1.22e-01

Pos logit contributions
dd+0.081
night+0.063
ez+0.058
amb+0.054
 pity+0.054
Neg logit contributions
 groups-0.109
bars-0.107
 towns-0.096
 younger-0.096
 fighters-0.096
Token good
Token ID922
Feature activation+1.16e-01

Pos logit contributions
dd+0.234
night+0.177
ez+0.173
amb+0.165
 pity+0.162
Neg logit contributions
 groups-0.212
bars-0.209
 younger-0.189
 older-0.187
 eight-0.185
Tokennight
Token ID3847
Feature activation+4.18e-01

Pos logit contributions
dd+0.126
ez+0.071
amb+0.061
faced+0.061
oof+0.057
Neg logit contributions
bars-0.287
 groups-0.282
 older-0.271
 younger-0.270
 department-0.265
Token.
Token ID13
Feature activation-1.33e-02

Pos logit contributions
dd+0.786
night+0.658
ez+0.602
 pity+0.578
amb+0.569
Neg logit contributions
 groups-0.721
bars-0.706
 towns-0.622
 fighters-0.620
 younger-0.616
Token
Token ID198
Feature activation+6.19e-02

Pos logit contributions
 groups+0.025
bars+0.025
 towns+0.022
 fighters+0.022
 younger+0.022
Neg logit contributions
dd-0.023
night-0.019
ez-0.017
 pity-0.016
amb-0.016
Token
Token ID198
Feature activation+5.84e-02

Pos logit contributions
dd+0.100
night+0.088
 pity+0.076
amb+0.074
ez+0.073
Neg logit contributions
 groups-0.116
bars-0.108
 eight-0.105
ents-0.101
 directions-0.100
TokenThe
Token ID464
Feature activation-2.31e-01

Pos logit contributions
dd+0.124
night+0.109
ez+0.100
 pity+0.097
amb+0.095
Neg logit contributions
 groups-0.086
bars-0.083
 towns-0.072
 eight-0.071
 fighters-0.071
Token little
Token ID1310
Feature activation-6.49e-02

Pos logit contributions
 groups+0.311
bars+0.303
 older+0.262
 younger+0.262
 towns+0.260
Neg logit contributions
dd-0.523
night-0.443
ez-0.425
amb-0.410
 pity-0.401
Token them
Token ID606
Feature activation+4.06e-02

Pos logit contributions
dd+0.045
night+0.038
ez+0.033
 pity+0.032
amb+0.030
Neg logit contributions
 groups-0.056
bars-0.055
 fighters-0.050
 towns-0.049
 eight-0.049
Token.
Token ID13
Feature activation-1.22e-02

Pos logit contributions
dd+0.045
night+0.042
-+0.039
 it+0.037
ggy+0.035
Neg logit contributions
bars-0.091
 groups-0.087
 eight-0.083
lights-0.082
 thirteen-0.082
Token She
Token ID1375
Feature activation+1.59e-02

Pos logit contributions
 groups+0.024
bars+0.024
 fighters+0.022
 towns+0.022
 fastest+0.021
Neg logit contributions
dd-0.019
night-0.016
ez-0.014
amb-0.013
 pity-0.013
Token said
Token ID531
Feature activation+9.17e-02

Pos logit contributions
dd+0.027
night+0.023
ez+0.020
 pity+0.019
amb+0.019
Neg logit contributions
 groups-0.030
bars-0.030
 towns-0.026
 fighters-0.026
 younger-0.026
Token good
Token ID922
Feature activation+1.22e-01

Pos logit contributions
dd+0.155
night+0.128
ez+0.116
 pity+0.110
amb+0.109
Neg logit contributions
 groups-0.176
bars-0.172
 towns-0.154
 fighters-0.154
 eight-0.153
Tokennight
Token ID3847
Feature activation+4.17e-01

Pos logit contributions
dd+0.186
ez+0.136
faced+0.130
ggy+0.126
uggle+0.123
Neg logit contributions
bars-0.243
 groups-0.234
 older-0.232
 fighters-0.230
 department-0.230
Token and
Token ID290
Feature activation+7.22e-02

Pos logit contributions
dd+0.719
night+0.588
ez+0.537
 pity+0.516
amb+0.502
Neg logit contributions
 groups-0.780
bars-0.765
 fighters-0.685
 towns-0.683
 eight-0.681
Token went
Token ID1816
Feature activation+6.46e-02

Pos logit contributions
dd+0.131
night+0.109
ez+0.100
 pity+0.095
amb+0.094
Neg logit contributions
 groups-0.129
bars-0.127
 towns-0.112
 fighters-0.112
 younger-0.111
Token off
Token ID572
Feature activation-3.49e-03

Pos logit contributions
dd+0.129
night+0.104
ez+0.100
oof+0.095
 pity+0.095
Neg logit contributions
 groups-0.104
bars-0.101
 older-0.093
 younger-0.091
 towns-0.091
Token to
Token ID284
Feature activation-3.74e-03

Pos logit contributions
 groups+0.006
bars+0.006
 fighters+0.006
 towns+0.006
 younger+0.005
Neg logit contributions
dd-0.006
night-0.005
ez-0.005
 pity-0.004
amb-0.004
Token bed
Token ID3996
Feature activation+9.07e-03

Pos logit contributions
 groups+0.007
bars+0.007
 younger+0.007
 older+0.007
 towns+0.007
Neg logit contributions
dd-0.006
night-0.005
ez-0.004
amb-0.004
faced-0.004
Token are
Token ID389
Feature activation-1.34e-01

Pos logit contributions
 groups+0.332
bars+0.310
 department+0.297
 younger+0.295
 towns+0.293
Neg logit contributions
dd-0.317
night-0.263
 it-0.248
ez-0.236
 wish-0.234
Token the
Token ID262
Feature activation-1.18e-01

Pos logit contributions
bars+0.286
 groups+0.254
 towns+0.236
ains+0.233
oes+0.230
Neg logit contributions
dd-0.197
 it-0.157
 pity-0.153
night-0.145
 happened-0.140
Token sun
Token ID4252
Feature activation+6.16e-02

Pos logit contributions
bars+0.202
 groups+0.200
ents+0.177
 towns+0.171
lights+0.170
Neg logit contributions
dd-0.209
night-0.186
 pity-0.164
ez-0.159
 wish-0.154
Token's
Token ID338
Feature activation-8.40e-02

Pos logit contributions
dd+0.082
night+0.073
-+0.065
 it+0.061
ake+0.059
Neg logit contributions
 groups-0.135
bars-0.126
 towns-0.119
 eight-0.117
 narrower-0.117
Token good
Token ID922
Feature activation+1.23e-01

Pos logit contributions
bars+0.153
 groups+0.152
ents+0.132
 towns+0.130
 fighters+0.130
Neg logit contributions
dd-0.147
night-0.130
 pity-0.111
ez-0.109
--0.106
Tokennight
Token ID3847
Feature activation+4.15e-01

Pos logit contributions
dd+0.190
night+0.165
 happened+0.136
-+0.136
 it+0.131
Neg logit contributions
 groups-0.247
bars-0.246
 towns-0.223
ents-0.220
 eight-0.219
Token kiss
Token ID9245
Feature activation+8.08e-02

Pos logit contributions
dd+0.654
night+0.576
ez+0.469
-+0.459
 pity+0.440
Neg logit contributions
 groups-0.839
bars-0.829
 towns-0.738
 younger-0.727
ents-0.723
Token to
Token ID284
Feature activation-7.84e-03

Pos logit contributions
dd+0.133
night+0.113
ez+0.098
 pity+0.093
oo+0.090
Neg logit contributions
 groups-0.160
bars-0.156
 towns-0.140
 younger-0.138
 fighters-0.137
Token the
Token ID262
Feature activation-1.21e-01

Pos logit contributions
 groups+0.014
bars+0.014
 fighters+0.012
 towns+0.012
 department+0.012
Neg logit contributions
dd-0.014
night-0.012
ez-0.011
 pity-0.011
oo-0.010
Token earth
Token ID4534
Feature activation-1.15e-02

Pos logit contributions
 groups+0.199
bars+0.196
 fighters+0.169
 towns+0.168
 eight+0.166
Neg logit contributions
dd-0.237
night-0.204
ez-0.183
 pity-0.178
amb-0.173
Token."
Token ID526
Feature activation-6.28e-02

Pos logit contributions
 groups+0.023
bars+0.023
 narrower+0.021
 eight+0.021
 towns+0.021
Neg logit contributions
dd-0.016
night-0.014
--0.012
ez-0.011
 it-0.011
Token.
Token ID13
Feature activation-1.09e-02

Pos logit contributions
dd+0.224
night+0.188
amb+0.172
ez+0.171
oof+0.169
Neg logit contributions
 groups-0.179
bars-0.154
 towns-0.149
 bridges-0.147
lights-0.144
Token Then
Token ID3244
Feature activation+3.54e-02

Pos logit contributions
 groups+0.019
bars+0.019
 towns+0.017
 eight+0.017
 younger+0.017
Neg logit contributions
dd-0.020
night-0.017
ez-0.015
 pity-0.015
faced-0.014
Token she
Token ID673
Feature activation+4.09e-02

Pos logit contributions
dd+0.063
night+0.052
ez+0.048
 pity+0.045
amb+0.045
Neg logit contributions
 groups-0.065
bars-0.063
 towns-0.057
 eight-0.056
 fighters-0.056
Token said
Token ID531
Feature activation+8.98e-02

Pos logit contributions
dd+0.069
night+0.057
ez+0.050
 pity+0.048
amb+0.046
Neg logit contributions
 groups-0.079
bars-0.079
 towns-0.070
 fighters-0.069
lights-0.069
Token good
Token ID922
Feature activation+1.15e-01

Pos logit contributions
dd+0.155
night+0.124
ez+0.120
oof+0.112
 pity+0.111
Neg logit contributions
 groups-0.165
bars-0.163
 eight-0.151
 older-0.149
 younger-0.149
Tokennight
Token ID3847
Feature activation+4.15e-01

Pos logit contributions
dd+0.150
faced+0.099
ez+0.096
ggy+0.091
Here+0.086
Neg logit contributions
bars-0.248
 older-0.244
oes-0.242
 department-0.241
 fighters-0.238
Token to
Token ID284
Feature activation-9.57e-03

Pos logit contributions
dd+0.778
night+0.622
ez+0.598
 pity+0.585
uggle+0.567
Neg logit contributions
 groups-0.698
bars-0.687
 eight-0.618
 fighters-0.616
 younger-0.615
Token the
Token ID262
Feature activation-1.27e-01

Pos logit contributions
 groups+0.016
bars+0.015
 towns+0.014
 younger+0.014
 eight+0.014
Neg logit contributions
dd-0.019
night-0.016
ez-0.015
amb-0.014
faced-0.014
Token nos
Token ID43630
Feature activation-2.52e-03

Pos logit contributions
 groups+0.226
bars+0.223
 towns+0.196
 fighters+0.195
 younger+0.193
Neg logit contributions
dd-0.232
night-0.195
ez-0.175
 pity-0.169
amb-0.165
Tokeny
Token ID88
Feature activation-8.34e-02

Pos logit contributions
 groups+0.004
bars+0.004
 towns+0.003
 fighters+0.003
 younger+0.003
Neg logit contributions
dd-0.005
night-0.004
ez-0.004
amb-0.004
 pity-0.004
Token little
Token ID1310
Feature activation-6.65e-02

Pos logit contributions
 groups+0.164
bars+0.161
 towns+0.144
 fighters+0.144
 younger+0.143
Neg logit contributions
dd-0.138
night-0.113
ez-0.101
 pity-0.096
amb-0.094
Token,
Token ID11
Feature activation+1.89e-02

Pos logit contributions
dd+0.369
night+0.306
ez+0.272
 pity+0.261
amb+0.255
Neg logit contributions
 groups-0.413
bars-0.407
 towns-0.361
 fighters-0.361
 younger-0.358
Token Jack
Token ID3619
Feature activation+8.71e-02

Pos logit contributions
dd+0.028
night+0.023
ez+0.019
 pity+0.019
amb+0.018
Neg logit contributions
 groups-0.040
bars-0.040
 fighters-0.036
 towns-0.036
 eight-0.035
Tokeny
Token ID88
Feature activation-8.38e-02

Pos logit contributions
dd+0.137
night+0.115
ez+0.101
-+0.093
amb+0.093
Neg logit contributions
 groups-0.179
bars-0.176
lights-0.160
 towns-0.158
 bridges-0.158
Token said
Token ID531
Feature activation+8.39e-02

Pos logit contributions
 groups+0.162
bars+0.160
lights+0.142
 fighters+0.141
 towns+0.141
Neg logit contributions
dd-0.138
night-0.119
ez-0.102
 pity-0.097
amb-0.095
Token good
Token ID922
Feature activation+1.08e-01

Pos logit contributions
dd+0.127
night+0.097
ez+0.093
oof+0.086
amb+0.085
Neg logit contributions
 groups-0.173
bars-0.171
 eight-0.159
 younger-0.156
 older-0.156
Tokennight
Token ID3847
Feature activation+4.14e-01

Pos logit contributions
dd+0.134
ez+0.093
faced+0.083
ggy+0.078
Here+0.073
Neg logit contributions
bars-0.241
 older-0.237
 department-0.236
 fighters-0.233
 groups-0.232
Token to
Token ID284
Feature activation-1.32e-02

Pos logit contributions
dd+0.757
night+0.604
ez+0.576
 pity+0.558
uggle+0.543
Neg logit contributions
 groups-0.725
bars-0.711
 fighters-0.641
 eight-0.638
 younger-0.637
Token the
Token ID262
Feature activation-1.33e-01

Pos logit contributions
 groups+0.024
bars+0.024
 towns+0.021
 younger+0.021
 fighters+0.021
Neg logit contributions
dd-0.024
night-0.020
ez-0.018
 pity-0.017
amb-0.017
Token skull
Token ID14511
Feature activation+1.50e-01

Pos logit contributions
 groups+0.228
bars+0.212
 older+0.210
 eight+0.209
 younger+0.206
Neg logit contributions
dd-0.252
ez-0.204
night-0.198
amb-0.196
oo-0.190
Token and
Token ID290
Feature activation+6.81e-02

Pos logit contributions
dd+0.213
night+0.200
ez+0.159
-+0.157
faced+0.145
Neg logit contributions
 groups-0.311
bars-0.301
 towns-0.280
 bridges-0.272
ains-0.272
Token went
Token ID1816
Feature activation+5.78e-02

Pos logit contributions
dd+0.131
ez+0.104
night+0.104
amb+0.096
faced+0.095
Neg logit contributions
 groups-0.111
bars-0.110
 younger-0.098
 fighters-0.098
 towns-0.098
Token anymore
Token ID7471
Feature activation-5.17e-02

Pos logit contributions
dd+0.054
night+0.046
-+0.041
 it+0.039
ez+0.039
Neg logit contributions
 groups-0.075
bars-0.073
 fighters-0.070
ains-0.069
 bowls-0.068
Token.
Token ID13
Feature activation-2.32e-02

Pos logit contributions
bars+0.097
 groups+0.097
 towns+0.085
 fighters+0.084
 younger+0.084
Neg logit contributions
dd-0.090
night-0.075
ez-0.068
 pity-0.063
ggy-0.062
Token She
Token ID1375
Feature activation+1.11e-02

Pos logit contributions
 groups+0.050
bars+0.048
 fighters+0.044
 towns+0.043
lights+0.043
Neg logit contributions
dd-0.033
night-0.027
ez-0.024
amb-0.024
 pity-0.023
Token said
Token ID531
Feature activation+8.63e-02

Pos logit contributions
dd+0.021
night+0.017
ez+0.016
 pity+0.015
uggle+0.015
Neg logit contributions
 groups-0.019
bars-0.019
 younger-0.017
 eight-0.017
 towns-0.017
Token good
Token ID922
Feature activation+1.10e-01

Pos logit contributions
dd+0.135
night+0.107
ez+0.099
 pity+0.093
amb+0.092
Neg logit contributions
 groups-0.173
bars-0.172
 eight-0.158
 towns-0.156
 younger-0.156
Tokennight
Token ID3847
Feature activation+4.14e-01

Pos logit contributions
dd+0.140
ez+0.086
faced+0.083
ggy+0.077
Here+0.073
Neg logit contributions
bars-0.257
 older-0.241
 department-0.241
oes-0.238
 fighters-0.236
Token to
Token ID284
Feature activation-1.26e-02

Pos logit contributions
dd+0.740
night+0.598
ez+0.557
 pity+0.539
uggle+0.520
Neg logit contributions
 groups-0.744
bars-0.733
 fighters-0.656
 eight-0.655
 towns-0.652
Token the
Token ID262
Feature activation-1.29e-01

Pos logit contributions
 groups+0.022
bars+0.022
 towns+0.020
 younger+0.020
 eight+0.020
Neg logit contributions
dd-0.023
night-0.019
ez-0.017
 pity-0.016
faced-0.016
Token cat
Token ID3797
Feature activation+1.43e-01

Pos logit contributions
 groups+0.226
bars+0.221
 towns+0.195
 fighters+0.194
 younger+0.192
Neg logit contributions
dd-0.241
night-0.203
ez-0.184
 pity-0.177
amb-0.174
Token and
Token ID290
Feature activation+6.87e-02

Pos logit contributions
dd+0.225
night+0.191
ez+0.168
 pity+0.156
amb+0.154
Neg logit contributions
 groups-0.287
bars-0.283
 towns-0.253
 fighters-0.251
 eight-0.249
Token went
Token ID1816
Feature activation+5.78e-02

Pos logit contributions
dd+0.126
night+0.104
ez+0.096
 pity+0.091
amb+0.090
Neg logit contributions
 groups-0.121
bars-0.120
 towns-0.106
 fighters-0.106
 younger-0.105
Token it
Token ID340
Feature activation+3.18e-01

Pos logit contributions
dd+0.044
amb+0.034
faced+0.034
night+0.033
ggy+0.033
Neg logit contributions
bars-0.038
 groups-0.037
 older-0.034
 forks-0.034
 directions-0.034
Token away
Token ID1497
Feature activation+9.54e-02

Pos logit contributions
dd+0.474
night+0.398
ez+0.336
 pity+0.315
uggle+0.311
Neg logit contributions
 groups-0.674
bars-0.660
 fighters-0.599
 towns-0.598
 younger-0.589
Token.
Token ID13
Feature activation-1.09e-02

Pos logit contributions
dd+0.140
night+0.127
-+0.103
ez+0.099
 pity+0.097
Neg logit contributions
bars-0.203
 groups-0.201
ents-0.178
 eight-0.178
 fighters-0.178
Token "
Token ID366
Feature activation-8.37e-02

Pos logit contributions
 groups+0.019
bars+0.019
 eight+0.017
 younger+0.017
ents+0.016
Neg logit contributions
dd-0.021
night-0.017
ez-0.016
 pity-0.015
faced-0.015
TokenGood
Token ID10248
Feature activation+1.50e-02

Pos logit contributions
bars+0.115
 groups+0.104
 eight+0.102
 highlight+0.098
 fastest+0.097
Neg logit contributions
dd-0.167
night-0.148
--0.136
 pity-0.135
oo-0.135
Tokennight
Token ID3847
Feature activation+4.14e-01

Pos logit contributions
dd+0.022
ez+0.017
to+0.015
faced+0.015
Aid+0.014
Neg logit contributions
 first-0.027
 size-0.027
 fighters-0.027
 department-0.027
 eight-0.026
Token,
Token ID11
Feature activation+2.29e-02

Pos logit contributions
dd+0.752
night+0.608
ez+0.577
 pity+0.547
faced+0.536
Neg logit contributions
 groups-0.739
bars-0.707
 eight-0.651
 fighters-0.650
 older-0.649
Token Brown
Token ID4373
Feature activation+9.66e-02

Pos logit contributions
dd+0.040
night+0.032
ez+0.030
 pity+0.028
amb+0.028
Neg logit contributions
 groups-0.043
bars-0.042
 younger-0.038
 older-0.038
 eight-0.038
Tokenie
Token ID494
Feature activation+5.89e-03

Pos logit contributions
dd+0.144
night+0.122
-+0.103
ez+0.102
oo+0.096
Neg logit contributions
 groups-0.208
bars-0.207
 towns-0.184
lights-0.183
 bridges-0.181
Token,"
Token ID553
Feature activation+9.91e-02

Pos logit contributions
dd+0.010
night+0.008
ez+0.007
 pity+0.007
amb+0.007
Neg logit contributions
 groups-0.011
bars-0.011
 fighters-0.010
 towns-0.010
 younger-0.010
Token she
Token ID673
Feature activation+4.10e-02

Pos logit contributions
dd+0.200
night+0.165
ez+0.152
uggle+0.149
faced+0.148
Neg logit contributions
bars-0.159
 groups-0.152
 younger-0.143
 older-0.140
ents-0.137
Token the
Token ID262
Feature activation-1.55e-01

Pos logit contributions
 groups+0.172
bars+0.169
 fighters+0.155
 towns+0.154
 department+0.153
Neg logit contributions
dd-0.125
night-0.101
ez-0.089
 pity-0.087
oo-0.085
Token pedals
Token ID42563
Feature activation-8.27e-02

Pos logit contributions
 groups+0.276
bars+0.248
lights+0.243
 fighters+0.239
 department+0.236
Neg logit contributions
dd-0.265
night-0.231
 pity-0.219
 puppy-0.201
 wish-0.200
Token or
Token ID393
Feature activation-1.26e-01

Pos logit contributions
 groups+0.182
 towns+0.164
 fighters+0.163
 department+0.161
 bowls+0.160
Neg logit contributions
dd-0.108
--0.088
 it-0.086
night-0.086
oo-0.077
Token the
Token ID262
Feature activation-1.52e-01

Pos logit contributions
 groups+0.239
bars+0.225
 fighters+0.210
lights+0.207
 towns+0.207
Neg logit contributions
dd-0.211
night-0.178
 pity-0.161
ez-0.159
faced-0.150
Token handle
Token ID5412
Feature activation-1.23e-01

Pos logit contributions
 groups+0.289
bars+0.263
lights+0.254
 fighters+0.251
 department+0.250
Neg logit contributions
dd-0.240
night-0.204
 pity-0.196
 puppy-0.177
ez-0.176
Tokenbars
Token ID34046
Feature activation-5.23e-01

Pos logit contributions
 groups+0.282
 towns+0.253
 fighters+0.245
 department+0.243
 highlight+0.240
Neg logit contributions
dd-0.148
--0.131
 it-0.130
night-0.122
 happened-0.110
Token.
Token ID13
Feature activation-4.85e-02

Pos logit contributions
 groups+1.052
bars+0.944
 towns+0.936
 fighters+0.912
 eight+0.911
Neg logit contributions
dd-0.852
night-0.700
ez-0.627
--0.621
 pity-0.607
Token She
Token ID1375
Feature activation-1.52e-02

Pos logit contributions
 groups+0.091
bars+0.089
 towns+0.079
 fighters+0.079
 younger+0.079
Neg logit contributions
dd-0.085
night-0.071
ez-0.063
 pity-0.060
amb-0.059
Token felt
Token ID2936
Feature activation-1.19e-01

Pos logit contributions
 groups+0.027
bars+0.027
 towns+0.023
 fighters+0.023
 eight+0.023
Neg logit contributions
dd-0.028
night-0.024
ez-0.021
 pity-0.020
amb-0.020
Token sad
Token ID6507
Feature activation-8.98e-03

Pos logit contributions
 groups+0.204
bars+0.199
 older+0.177
 younger+0.176
 eight+0.174
Neg logit contributions
dd-0.226
night-0.194
amb-0.172
ez-0.171
oof-0.165
Token.
Token ID13
Feature activation-4.83e-02

Pos logit contributions
 groups+0.021
bars+0.020
 towns+0.019
 fighters+0.018
 eight+0.018
Neg logit contributions
dd-0.011
night-0.009
--0.008
ez-0.007
 it-0.007
Token slowly
Token ID6364
Feature activation-1.02e-01

Pos logit contributions
dd+0.081
night+0.068
ez+0.063
 pity+0.060
uggle+0.059
Neg logit contributions
bars-0.075
 groups-0.074
 towns-0.065
 younger-0.064
 fighters-0.064
Token pulled
Token ID5954
Feature activation-2.48e-02

Pos logit contributions
 groups+0.202
bars+0.184
 fighters+0.177
 department+0.176
lights+0.173
Neg logit contributions
dd-0.169
night-0.143
ez-0.123
--0.118
 pity-0.117
Token on
Token ID319
Feature activation+9.04e-02

Pos logit contributions
bars+0.046
 groups+0.046
 younger+0.040
 older+0.040
 towns+0.040
Neg logit contributions
dd-0.044
night-0.037
ez-0.033
amb-0.031
uggle-0.031
Token the
Token ID262
Feature activation-1.57e-01

Pos logit contributions
dd+0.170
night+0.143
ez+0.130
 pity+0.125
amb+0.122
Neg logit contributions
 groups-0.157
bars-0.154
 towns-0.136
 fighters-0.135
 younger-0.134
Token handle
Token ID5412
Feature activation-1.23e-01

Pos logit contributions
 groups+0.361
bars+0.335
 fighters+0.324
ents+0.318
 department+0.318
Neg logit contributions
dd-0.194
night-0.160
 pity-0.133
ez-0.125
faced-0.123
Tokenbars
Token ID34046
Feature activation-5.22e-01

Pos logit contributions
 groups+0.189
ains+0.156
 towns+0.153
 eight+0.152
illas+0.149
Neg logit contributions
--0.226
dd-0.219
night-0.215
 it-0.214
amb-0.193
Token.
Token ID13
Feature activation-4.42e-02

Pos logit contributions
 groups+1.115
 towns+0.989
 eight+0.980
 department+0.970
 fighters+0.969
Neg logit contributions
dd-0.772
night-0.643
--0.597
amb-0.567
ez-0.552
Token He
Token ID679
Feature activation-7.48e-02

Pos logit contributions
 groups+0.077
bars+0.076
 towns+0.066
 younger+0.066
 fighters+0.066
Neg logit contributions
dd-0.084
night-0.070
ez-0.064
 pity-0.061
amb-0.060
Token slowly
Token ID6364
Feature activation-1.03e-01

Pos logit contributions
 groups+0.136
bars+0.133
 towns+0.118
 fighters+0.118
 younger+0.117
Neg logit contributions
dd-0.135
night-0.113
ez-0.102
 pity-0.097
amb-0.096
Token rolled
Token ID11686
Feature activation-4.16e-02

Pos logit contributions
 groups+0.203
bars+0.185
 fighters+0.177
lights+0.176
 department+0.175
Neg logit contributions
dd-0.171
night-0.145
ez-0.127
--0.119
 pity-0.119
Token down
Token ID866
Feature activation+2.42e-02

Pos logit contributions
bars+0.079
 groups+0.078
 towns+0.069
 eight+0.069
 bridges+0.069
Neg logit contributions
dd-0.073
night-0.061
ez-0.056
oof-0.051
 pity-0.051
Tokenas
Token ID292
Feature activation-3.18e-01

Pos logit contributions
dd+0.027
night+0.021
ez+0.017
 pity+0.016
amb+0.016
Neg logit contributions
 groups-0.050
bars-0.049
 towns-0.045
 fighters-0.045
 younger-0.044
Tokensels
Token ID14002
Feature activation-1.40e-01

Pos logit contributions
 groups+0.655
bars+0.622
 towns+0.573
 fighters+0.573
 younger+0.572
Neg logit contributions
dd-0.497
night-0.410
ez-0.361
amb-0.335
--0.334
Token on
Token ID319
Feature activation+9.05e-02

Pos logit contributions
 groups+0.297
bars+0.276
 towns+0.261
 fighters+0.259
 younger+0.259
Neg logit contributions
dd-0.206
night-0.180
--0.146
 happened-0.139
 it-0.138
Token the
Token ID262
Feature activation-1.55e-01

Pos logit contributions
dd+0.121
night+0.094
ez+0.080
 pity+0.076
amb+0.073
Neg logit contributions
 groups-0.207
bars-0.205
 fighters-0.186
 towns-0.185
 younger-0.184
Token handle
Token ID5412
Feature activation-1.21e-01

Pos logit contributions
 groups+0.269
bars+0.262
 fighters+0.239
 department+0.233
ents+0.229
Neg logit contributions
dd-0.280
night-0.250
 pity-0.213
ez-0.213
--0.202
Tokenbars
Token ID34046
Feature activation-5.19e-01

Pos logit contributions
 groups+0.252
 towns+0.218
 fighters+0.215
 eight+0.215
 younger+0.213
Neg logit contributions
dd-0.180
night-0.158
--0.141
 it-0.137
amb-0.127
Token and
Token ID290
Feature activation+4.31e-02

Pos logit contributions
 groups+1.055
bars+0.963
 towns+0.926
 fighters+0.919
 eight+0.913
Neg logit contributions
dd-0.836
night-0.702
ez-0.617
amb-0.586
--0.584
Token a
Token ID257
Feature activation+3.46e-02

Pos logit contributions
dd+0.075
night+0.065
ez+0.058
 pity+0.057
 it+0.054
Neg logit contributions
 groups-0.077
bars-0.074
 fighters-0.067
 department-0.067
ains-0.066
Token bell
Token ID8966
Feature activation+1.10e-01

Pos logit contributions
dd+0.058
night+0.049
ez+0.043
 pity+0.042
amb+0.040
Neg logit contributions
 groups-0.066
bars-0.065
 fighters-0.058
 towns-0.058
 younger-0.057
Token that
Token ID326
Feature activation+1.12e-01

Pos logit contributions
dd+0.181
night+0.155
ez+0.136
-+0.131
amb+0.129
Neg logit contributions
 groups-0.216
bars-0.196
 towns-0.187
 fighters-0.183
 eight-0.183
Token sounded
Token ID14846
Feature activation-8.95e-02

Pos logit contributions
dd+0.191
night+0.161
ez+0.142
 pity+0.140
uggle+0.129
Neg logit contributions
 groups-0.214
bars-0.204
ents-0.185
 fighters-0.185
 eight-0.184
Token how
Token ID703
Feature activation-1.34e-01

Pos logit contributions
dd+0.031
night+0.027
ez+0.023
 pity+0.023
amb+0.022
Neg logit contributions
 groups-0.038
bars-0.037
 fighters-0.034
 towns-0.033
 eight-0.033
Token to
Token ID284
Feature activation-3.51e-02

Pos logit contributions
 groups+0.287
bars+0.273
lights+0.258
 fighters+0.256
ains+0.255
Neg logit contributions
dd-0.194
night-0.161
--0.141
 it-0.133
 pity-0.130
Token turn
Token ID1210
Feature activation-2.04e-01

Pos logit contributions
 groups+0.068
bars+0.065
 department+0.060
lights+0.060
 fighters+0.060
Neg logit contributions
dd-0.058
night-0.047
ez-0.042
 pity-0.042
amb-0.042
Token the
Token ID262
Feature activation-1.53e-01

Pos logit contributions
 groups+0.418
 fighters+0.391
 department+0.391
bars+0.376
 eight+0.370
Neg logit contributions
dd-0.296
night-0.242
 it-0.228
--0.225
 pity-0.215
Token handle
Token ID5412
Feature activation-1.20e-01

Pos logit contributions
 groups+0.290
 department+0.259
ill+0.257
lights+0.255
 fighters+0.254
Neg logit contributions
dd-0.232
night-0.209
 pity-0.197
--0.185
 wish-0.183
Tokenbars
Token ID34046
Feature activation-5.18e-01

Pos logit contributions
 groups+0.108
 bowls+0.074
 fighters+0.071
 department+0.070
 towns+0.066
Neg logit contributions
dd-0.300
--0.298
night-0.293
 it-0.287
uggle-0.267
Token and
Token ID290
Feature activation+4.68e-02

Pos logit contributions
 groups+1.137
 department+1.006
 eight+1.005
 bowls+1.002
 fighters+0.999
Neg logit contributions
dd-0.698
night-0.614
--0.561
 it-0.527
ez-0.511
Token keep
Token ID1394
Feature activation-7.51e-03

Pos logit contributions
dd+0.071
night+0.062
 it+0.057
 pity+0.057
ez+0.053
Neg logit contributions
 groups-0.091
 department-0.080
ains-0.079
bars-0.079
 fighters-0.079
Token her
Token ID607
Feature activation+9.62e-02

Pos logit contributions
bars+0.015
 groups+0.014
 younger+0.013
 older+0.013
 towns+0.013
Neg logit contributions
dd-0.013
night-0.011
ez-0.009
faced-0.009
amb-0.009
Token balance
Token ID5236
Feature activation-6.93e-02

Pos logit contributions
dd+0.153
night+0.136
 pity+0.122
-+0.117
ez+0.115
Neg logit contributions
 groups-0.183
 department-0.167
bars-0.167
 fighters-0.163
ill-0.159
Token.
Token ID13
Feature activation-4.66e-02

Pos logit contributions
 groups+0.141
 department+0.133
 fighters+0.131
bars+0.131
 eight+0.130
Neg logit contributions
dd-0.106
night-0.085
--0.080
 pity-0.075
ez-0.075
Token blue
Token ID4171
Feature activation-1.71e-01

Pos logit contributions
dd+0.115
night+0.093
ez+0.082
 pity+0.079
amb+0.076
Neg logit contributions
 groups-0.157
bars-0.153
 fighters-0.140
 towns-0.138
 department-0.138
Token wheels
Token ID13666
Feature activation-1.28e-01

Pos logit contributions
bars+0.293
 groups+0.290
 towns+0.249
 eight+0.247
ents+0.245
Neg logit contributions
dd-0.334
night-0.279
ez-0.256
 pity-0.245
faced-0.241
Token and
Token ID290
Feature activation+4.74e-02

Pos logit contributions
 groups+0.313
 department+0.283
 towns+0.281
 fighters+0.279
 narrower+0.275
Neg logit contributions
dd-0.139
--0.111
night-0.108
amb-0.101
 it-0.097
Token pink
Token ID11398
Feature activation-4.99e-02

Pos logit contributions
dd+0.081
night+0.066
ez+0.059
 pity+0.058
amb+0.056
Neg logit contributions
 groups-0.090
bars-0.087
 fighters-0.079
 department-0.078
 towns-0.078
Token handle
Token ID5412
Feature activation-1.17e-01

Pos logit contributions
 groups+0.082
bars+0.081
 towns+0.070
 fighters+0.070
 younger+0.069
Neg logit contributions
dd-0.099
night-0.084
ez-0.077
 pity-0.074
amb-0.072
Tokenbars
Token ID34046
Feature activation-5.18e-01

Pos logit contributions
 groups+0.193
 eight+0.164
 department+0.163
 fighters+0.163
 towns+0.160
Neg logit contributions
dd-0.215
--0.191
night-0.189
 it-0.182
amb-0.175
Token.
Token ID13
Feature activation-4.47e-02

Pos logit contributions
 groups+0.934
bars+0.887
 towns+0.809
 fighters+0.805
 eight+0.800
Neg logit contributions
dd-0.940
night-0.795
ez-0.717
 pity-0.692
amb-0.680
Token She
Token ID1375
Feature activation-1.23e-02

Pos logit contributions
 groups+0.078
bars+0.076
 fighters+0.070
lights+0.068
 department+0.068
Neg logit contributions
dd-0.079
night-0.070
ez-0.061
amb-0.060
uggle-0.059
Token liked
Token ID8288
Feature activation+1.29e-02

Pos logit contributions
 groups+0.024
bars+0.024
 towns+0.022
 fighters+0.021
 younger+0.021
Neg logit contributions
dd-0.020
night-0.017
ez-0.015
 pity-0.014
amb-0.014
Token to
Token ID284
Feature activation-3.57e-02

Pos logit contributions
dd+0.022
night+0.018
ez+0.016
 pity+0.015
amb+0.015
Neg logit contributions
 groups-0.025
bars-0.024
 fighters-0.022
 department-0.021
lights-0.021
Token go
Token ID467
Feature activation-6.78e-03

Pos logit contributions
bars+0.069
 groups+0.066
 towns+0.060
 younger+0.059
 older+0.059
Neg logit contributions
dd-0.065
night-0.052
ez-0.048
amb-0.043
faced-0.043
Token.
Token ID13
Feature activation-4.78e-02

Pos logit contributions
dd+0.062
night+0.055
-+0.052
amb+0.051
uggle+0.048
Neg logit contributions
 groups-0.092
 eight-0.083
 towns-0.082
lights-0.081
 fighters-0.081
Token He
Token ID679
Feature activation-7.02e-02

Pos logit contributions
 groups+0.099
bars+0.093
lights+0.086
 fastest+0.085
 fighters+0.084
Neg logit contributions
dd-0.068
night-0.062
 Now-0.055
amb-0.051
 It-0.051
Token grabbed
Token ID13176
Feature activation-4.06e-03

Pos logit contributions
 groups+0.132
bars+0.131
 towns+0.115
 fighters+0.115
 younger+0.114
Neg logit contributions
dd-0.122
night-0.101
ez-0.091
 pity-0.087
amb-0.085
Token the
Token ID262
Feature activation-1.54e-01

Pos logit contributions
bars+0.006
 groups+0.006
 eight+0.006
 older+0.005
 younger+0.005
Neg logit contributions
dd-0.008
night-0.007
ez-0.007
uggle-0.006
amb-0.006
Token handle
Token ID5412
Feature activation-1.19e-01

Pos logit contributions
 groups+0.333
bars+0.307
 fighters+0.299
 department+0.296
lights+0.296
Neg logit contributions
dd-0.214
night-0.174
 pity-0.154
ez-0.146
amb-0.141
Tokenbars
Token ID34046
Feature activation-5.17e-01

Pos logit contributions
 groups+0.207
ains+0.173
 towns+0.171
 department+0.169
 fighters+0.168
Neg logit contributions
dd-0.209
--0.200
night-0.197
amb-0.190
 it-0.180
Token with
Token ID351
Feature activation+7.37e-02

Pos logit contributions
 groups+1.209
 towns+1.081
bars+1.070
 fighters+1.069
 department+1.063
Neg logit contributions
dd-0.662
night-0.530
--0.464
amb-0.462
ez-0.439
Token his
Token ID465
Feature activation-9.26e-03

Pos logit contributions
dd+0.125
night+0.103
ez+0.093
 pity+0.090
amb+0.087
Neg logit contributions
 groups-0.141
bars-0.138
 fighters-0.125
 towns-0.123
 department-0.123
Token little
Token ID1310
Feature activation-9.03e-02

Pos logit contributions
 groups+0.018
bars+0.017
 fighters+0.016
 department+0.016
 towns+0.016
Neg logit contributions
dd-0.014
night-0.012
 pity-0.011
ez-0.011
--0.010
Token hands
Token ID2832
Feature activation-1.06e-01

Pos logit contributions
 groups+0.149
bars+0.142
 fighters+0.130
lights+0.130
ains+0.128
Neg logit contributions
dd-0.167
night-0.149
 pity-0.133
ez-0.133
amb-0.128
Token.
Token ID13
Feature activation-4.95e-02

Pos logit contributions
 groups+0.247
 towns+0.226
bars+0.219
lights+0.218
ains+0.218
Neg logit contributions
dd-0.132
night-0.106
--0.094
amb-0.089
 it-0.088
Token.
Token ID13
Feature activation-4.90e-02

Pos logit contributions
 groups+0.142
ains+0.125
 towns+0.124
bars+0.124
 eight+0.124
Neg logit contributions
dd-0.083
night-0.080
--0.077
amb-0.065
faced-0.065
Token He
Token ID679
Feature activation-7.32e-02

Pos logit contributions
 groups+0.104
bars+0.102
 fighters+0.094
lights+0.093
 fastest+0.092
Neg logit contributions
dd-0.069
night-0.060
ez-0.050
amb-0.048
 Now-0.048
Token grabbed
Token ID13176
Feature activation-5.44e-03

Pos logit contributions
 groups+0.132
bars+0.130
 towns+0.115
 fighters+0.114
 younger+0.114
Neg logit contributions
dd-0.132
night-0.111
ez-0.100
 pity-0.096
amb-0.095
Token the
Token ID262
Feature activation-1.55e-01

Pos logit contributions
bars+0.008
 groups+0.008
 eight+0.008
 older+0.007
 younger+0.007
Neg logit contributions
dd-0.011
night-0.010
amb-0.009
uggle-0.009
ake-0.009
Token handle
Token ID5412
Feature activation-1.19e-01

Pos logit contributions
 groups+0.307
bars+0.304
 towns+0.271
 fighters+0.270
 younger+0.268
Neg logit contributions
dd-0.254
night-0.208
ez-0.186
 pity-0.176
amb-0.173
Tokenbars
Token ID34046
Feature activation-5.17e-01

Pos logit contributions
 groups+0.249
ains+0.219
 fighters+0.214
 towns+0.214
 department+0.212
Neg logit contributions
dd-0.165
--0.155
night-0.151
amb-0.140
 it-0.139
Token and
Token ID290
Feature activation+4.24e-02

Pos logit contributions
 groups+1.223
 towns+1.100
 fighters+1.092
 department+1.088
ains+1.078
Neg logit contributions
dd-0.645
night-0.509
--0.455
amb-0.445
ez-0.423
Token he
Token ID339
Feature activation-5.17e-02

Pos logit contributions
dd+0.080
night+0.067
ez+0.062
 pity+0.059
amb+0.058
Neg logit contributions
bars-0.073
 groups-0.073
 towns-0.063
 younger-0.063
 eight-0.063
Token quickly
Token ID2952
Feature activation-3.34e-02

Pos logit contributions
 groups+0.095
bars+0.092
 towns+0.082
 fighters+0.082
 younger+0.081
Neg logit contributions
dd-0.092
night-0.077
ez-0.070
 pity-0.067
amb-0.065
Token pushed
Token ID7121
Feature activation-6.56e-02

Pos logit contributions
 groups+0.064
bars+0.061
 fighters+0.056
 towns+0.056
lights+0.055
Neg logit contributions
dd-0.058
night-0.048
ez-0.043
 pity-0.041
amb-0.039
Token on
Token ID319
Feature activation+9.42e-02

Pos logit contributions
 groups+0.136
bars+0.136
 eight+0.122
 towns+0.122
 younger+0.121
Neg logit contributions
dd-0.102
night-0.084
ez-0.072
uggle-0.069
amb-0.068
Token twisted
Token ID19074
Feature activation-2.90e-02

Pos logit contributions
 groups+0.332
bars+0.326
 towns+0.306
ains+0.296
ill+0.288
Neg logit contributions
dd-0.307
 pity-0.233
night-0.226
ez-0.224
 it-0.222
Token,
Token ID11
Feature activation-7.41e-03

Pos logit contributions
 groups+0.067
bars+0.064
 fighters+0.060
 towns+0.060
 eight+0.060
Neg logit contributions
dd-0.037
night-0.029
--0.025
ez-0.025
 pity-0.023
Token and
Token ID290
Feature activation+4.66e-02

Pos logit contributions
 groups+0.014
bars+0.014
 towns+0.012
 fighters+0.012
 younger+0.012
Neg logit contributions
dd-0.012
night-0.010
ez-0.009
 pity-0.009
 it-0.009
Token the
Token ID262
Feature activation-1.52e-01

Pos logit contributions
dd+0.062
 it+0.054
night+0.044
 happened+0.042
 pity+0.042
Neg logit contributions
bars-0.105
 groups-0.104
 fastest-0.092
ains-0.091
 highlight-0.091
Token handle
Token ID5412
Feature activation-1.19e-01

Pos logit contributions
 groups+0.303
bars+0.295
 fighters+0.265
 towns+0.264
 eight+0.262
Neg logit contributions
dd-0.249
night-0.205
ez-0.182
 pity-0.176
amb-0.168
Tokenbars
Token ID34046
Feature activation-5.16e-01

Pos logit contributions
 groups+0.240
 towns+0.211
bars+0.210
 eight+0.206
 younger+0.205
Neg logit contributions
dd-0.192
night-0.154
ez-0.142
--0.139
 pity-0.132
Token were
Token ID547
Feature activation-1.80e-01

Pos logit contributions
 groups+1.032
bars+0.954
 towns+0.913
 eight+0.893
 fighters+0.888
Neg logit contributions
dd-0.848
night-0.695
ez-0.636
 pity-0.608
--0.594
Token bent
Token ID17157
Feature activation+1.20e-02

Pos logit contributions
 groups+0.343
 towns+0.320
bars+0.319
ains+0.311
oes+0.295
Neg logit contributions
dd-0.296
 it-0.227
 pity-0.225
ez-0.218
--0.214
Token.
Token ID13
Feature activation-4.76e-02

Pos logit contributions
dd+0.015
night+0.013
-+0.012
 it+0.011
amb+0.010
Neg logit contributions
 groups-0.027
bars-0.025
 eight-0.024
 towns-0.024
 bowls-0.024
Token
Token ID198
Feature activation+3.37e-02

Pos logit contributions
 groups+0.096
bars+0.096
 towns+0.084
 fastest+0.083
 younger+0.083
Neg logit contributions
dd-0.075
night-0.062
ez-0.055
 pity-0.052
amb-0.051
Token
Token ID198
Feature activation+3.28e-02

Pos logit contributions
dd+0.058
night+0.050
ez+0.044
 pity+0.042
amb+0.041
Neg logit contributions
 groups-0.063
bars-0.060
 eight-0.056
 towns-0.055
 fighters-0.055
Token how
Token ID703
Feature activation-1.32e-01

Pos logit contributions
dd+0.181
night+0.158
ez+0.142
 pity+0.139
uggle+0.134
Neg logit contributions
 groups-0.169
bars-0.164
 fighters-0.147
 towns-0.144
 eight-0.144
Token to
Token ID284
Feature activation-3.33e-02

Pos logit contributions
 groups+0.274
bars+0.254
 fighters+0.244
 department+0.243
lights+0.241
Neg logit contributions
dd-0.207
night-0.168
--0.144
 it-0.140
 pity-0.139
Token turn
Token ID1210
Feature activation-2.02e-01

Pos logit contributions
 groups+0.063
bars+0.061
 fighters+0.055
 department+0.055
 towns+0.054
Neg logit contributions
dd-0.057
night-0.047
ez-0.042
 pity-0.042
amb-0.040
Token the
Token ID262
Feature activation-1.52e-01

Pos logit contributions
 groups+0.416
 fighters+0.390
 department+0.389
bars+0.385
 eight+0.372
Neg logit contributions
dd-0.295
night-0.237
 it-0.216
--0.212
ez-0.210
Token handle
Token ID5412
Feature activation-1.17e-01

Pos logit contributions
 groups+0.298
 department+0.267
ill+0.266
 fighters+0.266
lights+0.263
Neg logit contributions
dd-0.223
night-0.194
 pity-0.182
--0.165
 news-0.165
Tokenbars
Token ID34046
Feature activation-5.15e-01

Pos logit contributions
 groups+0.146
 bowls+0.111
 fighters+0.110
 department+0.107
 towns+0.107
Neg logit contributions
dd-0.255
--0.249
night-0.243
 it-0.241
uggle-0.218
Token.
Token ID13
Feature activation-4.30e-02

Pos logit contributions
 groups+1.137
 department+1.011
 towns+1.007
 eight+1.006
 fighters+1.002
Neg logit contributions
dd-0.705
night-0.601
--0.560
 it-0.533
ez-0.501
Token
Token ID220
Feature activation+9.74e-04

Pos logit contributions
 groups+0.081
bars+0.080
 towns+0.071
 younger+0.071
 fighters+0.070
Neg logit contributions
dd-0.077
night-0.062
ez-0.057
 pity-0.054
amb-0.053
Token
Token ID198
Feature activation+3.75e-02

Pos logit contributions
dd+0.001
night+0.001
ez+0.001
uggle+0.001
 pity+0.001
Neg logit contributions
 groups-0.002
bars-0.002
 fighters-0.002
 towns-0.002
 eight-0.002
Token
Token ID198
Feature activation+3.66e-02

Pos logit contributions
dd+0.070
night+0.061
ez+0.054
 pity+0.052
amb+0.050
Neg logit contributions
 groups-0.066
bars-0.063
 towns-0.057
 eight-0.057
 younger-0.056
TokenFinally
Token ID11158
Feature activation-7.86e-02

Pos logit contributions
dd+0.076
night+0.065
ez+0.060
 pity+0.058
amb+0.057
Neg logit contributions
 groups-0.056
bars-0.055
 fighters-0.048
 younger-0.047
 towns-0.047
Token,
Token ID11
Feature activation-7.97e-03

Pos logit contributions
 groups+0.184
 fighters+0.166
 towns+0.165
 department+0.162
ains+0.162
Neg logit contributions
dd-0.104
night-0.086
--0.077
amb-0.073
 it-0.071
Token held
Token ID2714
Feature activation+1.71e-03

Pos logit contributions
 groups+0.017
bars+0.015
 fighters+0.015
 department+0.015
lights+0.015
Neg logit contributions
dd-0.012
night-0.009
ez-0.009
 it-0.008
 pity-0.008
Token onto
Token ID4291
Feature activation-3.32e-02

Pos logit contributions
dd+0.003
night+0.003
ez+0.002
faced+0.002
 pity+0.002
Neg logit contributions
bars-0.003
 groups-0.003
 eight-0.003
 older-0.003
 younger-0.003
Token the
Token ID262
Feature activation-1.50e-01

Pos logit contributions
 groups+0.057
bars+0.057
 eight+0.050
 towns+0.050
 younger+0.049
Neg logit contributions
dd-0.063
night-0.053
ez-0.048
 pity-0.046
amb-0.045
Token handle
Token ID5412
Feature activation-1.18e-01

Pos logit contributions
 groups+0.276
bars+0.255
lights+0.242
 fighters+0.237
 department+0.233
Neg logit contributions
dd-0.261
night-0.222
 pity-0.201
ez-0.200
faced-0.188
Tokenbars
Token ID34046
Feature activation-5.15e-01

Pos logit contributions
 groups+0.182
 fighters+0.159
 department+0.156
 towns+0.156
ains+0.155
Neg logit contributions
--0.222
dd-0.208
 it-0.205
night-0.201
amb-0.199
Token carefully
Token ID7773
Feature activation-3.20e-02

Pos logit contributions
 groups+1.139
 fighters+1.033
 towns+1.029
 department+1.027
ains+0.996
Neg logit contributions
dd-0.726
night-0.588
--0.561
amb-0.547
 it-0.501
Token and
Token ID290
Feature activation+4.68e-02

Pos logit contributions
 groups+0.069
 department+0.067
 fighters+0.067
ains+0.065
 towns+0.063
Neg logit contributions
dd-0.038
--0.032
night-0.032
 it-0.030
ez-0.027
Token slowly
Token ID6364
Feature activation-9.66e-02

Pos logit contributions
dd+0.084
night+0.071
ez+0.064
 pity+0.061
amb+0.060
Neg logit contributions
 groups-0.085
bars-0.083
 fighters-0.074
 towns-0.074
 younger-0.073
Token rolled
Token ID11686
Feature activation-3.90e-02

Pos logit contributions
 groups+0.192
ains+0.171
 fighters+0.171
 department+0.170
bars+0.168
Neg logit contributions
dd-0.158
night-0.133
--0.117
ez-0.115
 pity-0.109
Token over
Token ID625
Feature activation-6.13e-02

Pos logit contributions
 groups+0.076
bars+0.075
 towns+0.067
 fighters+0.066
 younger+0.066
Neg logit contributions
dd-0.066
night-0.054
ez-0.048
 pity-0.046
amb-0.045
Token the
Token ID262
Feature activation-1.57e-01

Pos logit contributions
dd+0.255
night+0.210
 pity+0.191
ez+0.189
faced+0.179
Neg logit contributions
bars-0.297
 groups-0.295
 fighters-0.261
lights-0.261
 towns-0.260
Token red
Token ID2266
Feature activation-5.48e-02

Pos logit contributions
bars+0.284
ents+0.251
lights+0.247
 department+0.231
ups+0.228
Neg logit contributions
dd-0.235
night-0.217
 pity-0.211
 name-0.209
 news-0.203
Token flags
Token ID9701
Feature activation-1.86e-01

Pos logit contributions
bars+0.094
 groups+0.091
ents+0.084
 younger+0.084
 department+0.082
Neg logit contributions
dd-0.094
night-0.084
--0.073
 pity-0.071
ez-0.070
Token.
Token ID13
Feature activation-4.90e-02

Pos logit contributions
 groups+0.377
bars+0.374
 towns+0.354
 eight+0.353
 younger+0.342
Neg logit contributions
dd-0.247
night-0.203
 it-0.185
--0.177
 happened-0.175
Token Both
Token ID5747
Feature activation-2.22e-01

Pos logit contributions
bars+0.103
 groups+0.098
 fastest+0.095
 highlight+0.091
lights+0.090
Neg logit contributions
night-0.060
dd-0.060
 Now-0.057
 It-0.054
amb-0.048
Token groups
Token ID2628
Feature activation-5.15e-01

Pos logit contributions
bars+0.472
ric+0.439
lands+0.416
ifying+0.410
oes+0.403
Neg logit contributions
dd-0.266
 it-0.259
--0.242
night-0.236
 happened-0.227
Token were
Token ID547
Feature activation-1.81e-01

Pos logit contributions
bars+0.932
 groups+0.900
lights+0.832
 eight+0.811
 narrower+0.809
Neg logit contributions
dd-0.861
night-0.713
 happened-0.637
 it-0.636
 pity-0.633
Token very
Token ID845
Feature activation+1.29e-02

Pos logit contributions
bars+0.409
 groups+0.361
lights+0.356
oes+0.347
ains+0.343
Neg logit contributions
dd-0.253
 pity-0.181
night-0.179
ggy-0.174
--0.168
Token happy
Token ID3772
Feature activation+8.34e-02

Pos logit contributions
dd+0.016
night+0.011
 pity+0.010
-+0.010
ggy+0.010
Neg logit contributions
bars-0.031
 groups-0.028
ains-0.028
lights-0.027
 department-0.027
Token.
Token ID13
Feature activation-5.07e-02

Pos logit contributions
dd+0.139
night+0.116
ez+0.102
 pity+0.097
amb+0.095
Neg logit contributions
bars-0.162
 groups-0.160
 eight-0.141
 younger-0.141
 towns-0.141
Token They
Token ID1119
Feature activation-7.08e-02

Pos logit contributions
bars+0.106
 groups+0.102
 fastest+0.096
 narrower+0.094
lights+0.094
Neg logit contributions
dd-0.068
night-0.062
 Now-0.061
 It-0.055
amb-0.054
Token She
Token ID1375
Feature activation-1.20e-02

Pos logit contributions
 groups+0.079
bars+0.078
 younger+0.068
 eight+0.068
 towns+0.068
Neg logit contributions
dd-0.095
night-0.077
ez-0.075
 pity-0.070
ggy-0.067
Token grabbed
Token ID13176
Feature activation+2.74e-03

Pos logit contributions
bars+0.022
 groups+0.022
 towns+0.019
 eight+0.019
 younger+0.019
Neg logit contributions
dd-0.022
night-0.018
ez-0.017
 pity-0.016
uggle-0.016
Token onto
Token ID4291
Feature activation-2.84e-02

Pos logit contributions
dd+0.006
night+0.005
ager+0.005
ake+0.005
 coincidence+0.005
Neg logit contributions
bars-0.004
 groups-0.004
 eight-0.004
 older-0.003
 first-0.003
Token its
Token ID663
Feature activation+1.18e-01

Pos logit contributions
bars+0.041
 groups+0.040
 eight+0.037
 older+0.036
 younger+0.036
Neg logit contributions
dd-0.063
night-0.054
ez-0.050
amb-0.049
faced-0.047
Token handle
Token ID5412
Feature activation-1.16e-01

Pos logit contributions
dd+0.203
night+0.177
amb+0.152
 pity+0.151
ez+0.151
Neg logit contributions
 groups-0.219
bars-0.209
 fighters-0.190
 towns-0.190
 department-0.187
Tokenbars
Token ID34046
Feature activation-5.15e-01

Pos logit contributions
 groups+0.310
 fighters+0.282
 towns+0.281
 department+0.280
ains+0.278
Neg logit contributions
dd-0.085
--0.083
night-0.076
amb-0.074
 it-0.062
Token and
Token ID290
Feature activation+4.85e-02

Pos logit contributions
 groups+1.163
bars+1.072
 towns+1.042
 fighters+1.041
 department+1.027
Neg logit contributions
dd-0.700
night-0.564
amb-0.483
ez-0.478
--0.460
Token put
Token ID1234
Feature activation+1.71e-04

Pos logit contributions
dd+0.094
night+0.079
ez+0.074
 pity+0.070
uggle+0.069
Neg logit contributions
bars-0.082
 groups-0.081
 towns-0.070
 younger-0.069
 eight-0.069
Token her
Token ID607
Feature activation+1.04e-01

Pos logit contributions
dd+0.000
night+0.000
ez+0.000
faced+0.000
ake+0.000
Neg logit contributions
bars-0.000
 groups-0.000
 bridges-0.000
 eight-0.000
 older-0.000
Token feet
Token ID3625
Feature activation-2.26e-01

Pos logit contributions
dd+0.169
night+0.149
amb+0.129
 pity+0.129
ez+0.128
Neg logit contributions
 groups-0.197
bars-0.175
 fighters-0.172
 department-0.171
 towns-0.169
Token onto
Token ID4291
Feature activation-2.94e-02

Pos logit contributions
 groups+0.547
 fighters+0.500
 towns+0.496
bars+0.495
 department+0.492
Neg logit contributions
dd-0.263
night-0.196
--0.169
amb-0.161
ez-0.153
Token let
Token ID1309
Feature activation-1.32e-01

Pos logit contributions
dd+0.087
night+0.072
ez+0.067
amb+0.064
uggle+0.064
Neg logit contributions
 groups-0.079
bars-0.079
 towns-0.071
 eight-0.069
 younger-0.069
Token go
Token ID467
Feature activation-6.01e-03

Pos logit contributions
 groups+0.194
bars+0.183
 younger+0.182
 older+0.180
 towns+0.178
Neg logit contributions
dd-0.275
ez-0.226
night-0.221
amb-0.217
oof-0.211
Token of
Token ID286
Feature activation-1.64e-01

Pos logit contributions
 groups+0.010
bars+0.010
 towns+0.009
 older+0.009
 younger+0.009
Neg logit contributions
dd-0.012
night-0.010
ez-0.009
 pity-0.009
faced-0.008
Token the
Token ID262
Feature activation-1.50e-01

Pos logit contributions
bars+0.298
 groups+0.292
 fighters+0.260
 department+0.258
 eight+0.256
Neg logit contributions
dd-0.289
night-0.240
ez-0.216
 pity-0.216
uggle-0.201
Token handle
Token ID5412
Feature activation-1.16e-01

Pos logit contributions
 groups+0.250
bars+0.245
 fighters+0.214
 towns+0.212
lights+0.211
Neg logit contributions
dd-0.288
night-0.248
ez-0.220
 pity-0.217
uggle-0.211
Tokenbars
Token ID34046
Feature activation-5.14e-01

Pos logit contributions
 groups+0.230
 eight+0.210
 towns+0.204
ains+0.201
 fighters+0.198
Neg logit contributions
dd-0.148
--0.147
night-0.139
 it-0.138
amb-0.125
Token.
Token ID13
Feature activation-4.48e-02

Pos logit contributions
 groups+1.053
 eight+0.930
 towns+0.926
 department+0.921
bars+0.920
Neg logit contributions
dd-0.809
night-0.679
--0.611
amb-0.611
ez-0.587
Token Suddenly
Token ID24975
Feature activation+8.76e-02

Pos logit contributions
 groups+0.089
bars+0.088
 fighters+0.078
 towns+0.077
lights+0.077
Neg logit contributions
dd-0.072
night-0.060
ez-0.052
amb-0.051
 pity-0.050
Token,
Token ID11
Feature activation-8.21e-03

Pos logit contributions
dd+0.131
night+0.104
 pity+0.084
ez+0.083
-+0.083
Neg logit contributions
bars-0.189
 groups-0.188
 younger-0.166
 fighters-0.166
 fastest-0.164
Token Joey
Token ID26154
Feature activation+8.24e-02

Pos logit contributions
 groups+0.016
bars+0.016
 younger+0.014
 fighters+0.014
 towns+0.014
Neg logit contributions
dd-0.014
night-0.012
ez-0.010
 pity-0.010
amb-0.010
Token heard
Token ID2982
Feature activation-4.49e-02

Pos logit contributions
dd+0.134
night+0.103
ez+0.097
-+0.092
 happened+0.087
Neg logit contributions
 groups-0.168
bars-0.161
lights-0.153
 fighters-0.146
 sizes-0.145
Token were
Token ID547
Feature activation-1.79e-01

Pos logit contributions
bars+0.135
lights+0.127
 groups+0.123
 fastest+0.122
 younger+0.119
Neg logit contributions
dd-0.102
 it-0.073
night-0.069
 happened-0.068
 wrote-0.067
Token sorting
Token ID29407
Feature activation-1.01e-01

Pos logit contributions
bars+0.353
 groups+0.325
lights+0.301
 department+0.294
 fighters+0.290
Neg logit contributions
dd-0.300
night-0.230
 pity-0.217
ez-0.213
oo-0.209
Token themselves
Token ID2405
Feature activation-1.74e-01

Pos logit contributions
bars+0.182
 groups+0.181
 fighters+0.160
 younger+0.158
 towns+0.158
Neg logit contributions
dd-0.181
night-0.150
ez-0.136
 pity-0.132
amb-0.126
Token into
Token ID656
Feature activation-2.24e-01

Pos logit contributions
bars+0.400
 groups+0.357
 fighters+0.348
 younger+0.345
 department+0.345
Neg logit contributions
dd-0.209
night-0.155
--0.153
 it-0.143
 inside-0.138
Token two
Token ID734
Feature activation-2.11e-01

Pos logit contributions
bars+0.415
 groups+0.351
 fighters+0.348
 department+0.347
 fastest+0.342
Neg logit contributions
dd-0.383
 pity-0.307
night-0.302
ez-0.300
 it-0.295
Token groups
Token ID2628
Feature activation-5.13e-01

Pos logit contributions
bars+0.370
oes+0.322
 department+0.304
 fighters+0.303
lights+0.292
Neg logit contributions
dd-0.373
--0.314
 pity-0.292
ez-0.291
 puppy-0.287
Token.
Token ID13
Feature activation-4.57e-02

Pos logit contributions
bars+1.042
 groups+1.014
lights+0.963
 younger+0.957
 eight+0.954
Neg logit contributions
dd-0.701
night-0.543
--0.522
 it-0.477
ez-0.468
Token He
Token ID679
Feature activation-7.22e-02

Pos logit contributions
 groups+0.097
bars+0.094
 department+0.089
 fighters+0.088
 eight+0.087
Neg logit contributions
dd-0.063
night-0.049
ez-0.042
amb-0.041
 Now-0.041
Token was
Token ID373
Feature activation+1.78e-02

Pos logit contributions
bars+0.152
 groups+0.147
lights+0.135
 younger+0.132
 directions+0.131
Neg logit contributions
dd-0.109
night-0.084
ez-0.075
 pity-0.071
--0.070
Token curious
Token ID11040
Feature activation+5.01e-02

Pos logit contributions
dd+0.027
night+0.020
 pity+0.020
ez+0.019
oo+0.019
Neg logit contributions
bars-0.038
lights-0.034
 groups-0.033
 department-0.031
ains-0.031
Token and
Token ID290
Feature activation+4.21e-02

Pos logit contributions
dd+0.068
night+0.059
ez+0.048
-+0.045
 pity+0.045
Neg logit contributions
 groups-0.107
bars-0.107
 eight-0.097
 fighters-0.096
 towns-0.096
Token toys
Token ID14958
Feature activation-4.12e-02

Pos logit contributions
bars+0.016
 groups+0.015
 thirteen+0.014
ents+0.014
ill+0.014
Neg logit contributions
dd-0.012
night-0.010
--0.009
 pity-0.009
ez-0.009
Token and
Token ID290
Feature activation+4.70e-02

Pos logit contributions
 groups+0.093
bars+0.090
lights+0.086
 younger+0.086
 eight+0.085
Neg logit contributions
dd-0.045
 it-0.038
--0.035
night-0.035
 inside-0.032
Token put
Token ID1234
Feature activation-8.44e-03

Pos logit contributions
dd+0.084
night+0.072
 pity+0.065
ez+0.063
amb+0.060
Neg logit contributions
 groups-0.081
bars-0.080
 fighters-0.070
lights-0.070
ains-0.069
Token them
Token ID606
Feature activation+7.14e-03

Pos logit contributions
 groups+0.015
bars+0.015
 towns+0.013
 fighters+0.013
 younger+0.013
Neg logit contributions
dd-0.015
night-0.013
ez-0.011
amb-0.011
 pity-0.011
Token in
Token ID287
Feature activation+1.65e-02

Pos logit contributions
dd+0.009
night+0.008
ez+0.006
-+0.006
 pity+0.006
Neg logit contributions
bars-0.016
 groups-0.016
 fighters-0.014
 younger-0.014
 towns-0.014
Token groups
Token ID2628
Feature activation-5.12e-01

Pos logit contributions
dd+0.020
 it+0.017
 pity+0.016
ez+0.015
night+0.015
Neg logit contributions
bars-0.037
 department-0.033
 fighters-0.033
ents-0.033
ups-0.033
Token.
Token ID13
Feature activation-4.75e-02

Pos logit contributions
 groups+1.037
bars+1.001
lights+0.974
 fighters+0.973
 narrower+0.968
Neg logit contributions
dd-0.697
night-0.578
--0.517
ez-0.492
 it-0.483
Token He
Token ID679
Feature activation-6.93e-02

Pos logit contributions
 groups+0.090
bars+0.088
 towns+0.079
 fighters+0.078
lights+0.077
Neg logit contributions
dd-0.080
night-0.069
ez-0.060
amb-0.058
 pity-0.058
Token counted
Token ID14789
Feature activation-2.29e-01

Pos logit contributions
 groups+0.127
bars+0.126
lights+0.111
 towns+0.110
 fighters+0.108
Neg logit contributions
dd-0.123
night-0.102
ez-0.093
 pity-0.087
amb-0.086
Token to
Token ID284
Feature activation-3.58e-02

Pos logit contributions
 groups+0.488
bars+0.477
 highlight+0.464
 forget+0.457
 fastest+0.449
Neg logit contributions
 it-0.262
dd-0.229
--0.225
 him-0.182
 inside-0.181
Token zero
Token ID6632
Feature activation-4.03e-02

Pos logit contributions
bars+0.051
lights+0.046
 groups+0.041
ains+0.040
 department+0.039
Neg logit contributions
dd-0.065
 it-0.064
 pity-0.063
--0.060
night-0.057
Token bird
Token ID6512
Feature activation+7.92e-02

Pos logit contributions
 groups+0.152
bars+0.148
 towns+0.133
 older+0.132
 younger+0.132
Neg logit contributions
dd-0.143
night-0.117
ez-0.106
amb-0.100
uggle-0.099
Token sat
Token ID3332
Feature activation+7.15e-02

Pos logit contributions
dd+0.128
night+0.103
ez+0.096
-+0.089
amb+0.087
Neg logit contributions
 groups-0.161
bars-0.155
 fighters-0.140
 department-0.139
lights-0.139
Token on
Token ID319
Feature activation+1.03e-01

Pos logit contributions
dd+0.115
night+0.094
ez+0.084
 pity+0.080
amb+0.078
Neg logit contributions
 groups-0.144
bars-0.142
 towns-0.127
 fighters-0.127
 younger-0.126
Token the
Token ID262
Feature activation-1.45e-01

Pos logit contributions
dd+0.207
night+0.176
ez+0.161
 pity+0.154
amb+0.154
Neg logit contributions
 groups-0.168
bars-0.163
 towns-0.145
 eight-0.144
 fighters-0.143
Token handle
Token ID5412
Feature activation-1.11e-01

Pos logit contributions
 groups+0.293
 department+0.254
 fighters+0.254
ill+0.252
lights+0.246
Neg logit contributions
dd-0.221
night-0.187
--0.169
amb-0.165
 puppy-0.164
Tokenbars
Token ID34046
Feature activation-5.09e-01

Pos logit contributions
 groups+0.242
 bowls+0.207
 department+0.207
 fighters+0.205
 forks+0.200
Neg logit contributions
dd-0.144
--0.138
night-0.137
amb-0.123
 it-0.123
Token and
Token ID290
Feature activation+5.74e-02

Pos logit contributions
 groups+1.171
bars+1.062
 towns+1.041
 fighters+1.039
 department+1.035
Neg logit contributions
dd-0.676
night-0.548
ez-0.471
amb-0.466
--0.459
Token held
Token ID2714
Feature activation+1.89e-03

Pos logit contributions
dd+0.108
night+0.090
ez+0.083
 pity+0.079
amb+0.078
Neg logit contributions
 groups-0.100
bars-0.099
 towns-0.087
 fighters-0.087
 younger-0.086
Token on
Token ID319
Feature activation+1.02e-01

Pos logit contributions
dd+0.004
night+0.003
uggle+0.003
ake+0.003
faced+0.003
Neg logit contributions
bars-0.003
 groups-0.003
 eight-0.003
 younger-0.003
 older-0.003
Token tight
Token ID5381
Feature activation-2.22e-03

Pos logit contributions
dd+0.164
night+0.137
ez+0.120
amb+0.116
 pity+0.116
Neg logit contributions
 groups-0.204
bars-0.199
 fighters-0.179
lights-0.178
 towns-0.177
Token.
Token ID13
Feature activation-3.48e-02

Pos logit contributions
 groups+0.004
bars+0.004
 fighters+0.004
 department+0.004
ains+0.004
Neg logit contributions
dd-0.004
night-0.003
ez-0.003
--0.003
amb-0.003
Token they
Token ID484
Feature activation-1.19e-01

Pos logit contributions
dd+0.111
night+0.092
ez+0.083
 pity+0.080
amb+0.078
Neg logit contributions
 groups-0.119
bars-0.117
 towns-0.103
 fighters-0.103
 younger-0.102
Token liked
Token ID8288
Feature activation+1.57e-02

Pos logit contributions
 groups+0.178
bars+0.174
 fighters+0.151
 towns+0.148
 eight+0.146
Neg logit contributions
dd-0.253
night-0.220
ez-0.200
amb-0.194
 pity-0.191
Token to
Token ID284
Feature activation-3.62e-02

Pos logit contributions
dd+0.028
night+0.023
ez+0.021
 pity+0.020
oof+0.019
Neg logit contributions
bars-0.028
 groups-0.028
 towns-0.025
 fighters-0.025
 department-0.024
Token live
Token ID2107
Feature activation+9.71e-02

Pos logit contributions
 groups+0.066
bars+0.064
 eight+0.057
 towns+0.057
 younger+0.057
Neg logit contributions
dd-0.066
night-0.055
ez-0.049
 pity-0.046
amb-0.046
Token in
Token ID287
Feature activation+1.47e-02

Pos logit contributions
dd+0.172
night+0.143
ez+0.129
 pity+0.123
amb+0.121
Neg logit contributions
 groups-0.181
bars-0.175
 towns-0.158
 fighters-0.157
 younger-0.155
Token groups
Token ID2628
Feature activation-5.08e-01

Pos logit contributions
dd+0.026
night+0.022
ez+0.021
 pity+0.020
 it+0.019
Neg logit contributions
bars-0.027
 groups-0.024
 fighters-0.023
 department-0.023
ents-0.022
Token and
Token ID290
Feature activation+4.59e-02

Pos logit contributions
bars+1.068
 groups+1.008
 eight+0.993
 narrower+0.993
 department+0.974
Neg logit contributions
dd-0.650
night-0.548
--0.502
 it-0.479
ez-0.470
Token talk
Token ID1561
Feature activation+2.92e-02

Pos logit contributions
dd+0.078
night+0.066
 pity+0.061
ez+0.060
 it+0.058
Neg logit contributions
bars-0.084
 groups-0.079
lights-0.071
ains-0.070
 department-0.070
Token to
Token ID284
Feature activation-3.65e-02

Pos logit contributions
dd+0.043
night+0.039
-+0.032
ez+0.032
 pity+0.030
Neg logit contributions
bars-0.059
 groups-0.059
 eight-0.052
 younger-0.052
 towns-0.052
Token each
Token ID1123
Feature activation-1.14e-01

Pos logit contributions
bars+0.064
 groups+0.063
 fighters+0.055
 towns+0.054
 department+0.054
Neg logit contributions
dd-0.067
night-0.057
ez-0.052
 pity-0.050
amb-0.048
Token other
Token ID584
Feature activation-2.11e-01

Pos logit contributions
bars+0.181
lights+0.161
 groups+0.158
 department+0.150
 fighters+0.149
Neg logit contributions
dd-0.228
night-0.199
ez-0.189
--0.184
 pity-0.184
Token note
Token ID3465
Feature activation+1.27e-01

Pos logit contributions
dd+0.108
night+0.090
ez+0.087
amb+0.082
oo+0.078
Neg logit contributions
 groups-0.082
bars-0.079
 younger-0.073
 older-0.072
 eight-0.072
Token attached
Token ID7223
Feature activation+4.71e-02

Pos logit contributions
dd+0.208
night+0.170
ez+0.154
 pity+0.146
amb+0.143
Neg logit contributions
bars-0.252
 groups-0.251
 fighters-0.221
 older-0.221
 younger-0.220
Token to
Token ID284
Feature activation-2.33e-02

Pos logit contributions
dd+0.071
night+0.061
ez+0.052
-+0.051
 pity+0.049
Neg logit contributions
 groups-0.098
bars-0.092
 towns-0.087
lights-0.085
 fighters-0.084
Token the
Token ID262
Feature activation-1.44e-01

Pos logit contributions
 groups+0.042
bars+0.042
 fighters+0.037
 towns+0.037
lights+0.037
Neg logit contributions
dd-0.041
night-0.035
ez-0.031
 pity-0.031
faced-0.029
Token handle
Token ID5412
Feature activation-1.07e-01

Pos logit contributions
 groups+0.260
bars+0.250
 fighters+0.225
lights+0.223
 towns+0.223
Neg logit contributions
dd-0.254
night-0.218
ez-0.192
 pity-0.191
faced-0.183
Tokenbars
Token ID34046
Feature activation-5.06e-01

Pos logit contributions
 groups+0.226
 towns+0.203
 narrower+0.192
 eight+0.191
 bridges+0.191
Neg logit contributions
dd-0.145
night-0.134
--0.126
 it-0.119
amb-0.110
Token.
Token ID13
Feature activation-3.56e-02

Pos logit contributions
 groups+0.974
bars+0.891
 towns+0.857
 eight+0.839
 fighters+0.832
Neg logit contributions
dd-0.861
night-0.733
ez-0.650
amb-0.626
 pity-0.619
Token The
Token ID383
Feature activation-6.49e-02

Pos logit contributions
 groups+0.073
bars+0.072
 fighters+0.065
 towns+0.065
 younger+0.064
Neg logit contributions
dd-0.055
night-0.045
ez-0.039
 pity-0.038
amb-0.037
Token note
Token ID3465
Feature activation+1.26e-01

Pos logit contributions
bars+0.128
 groups+0.124
 eight+0.116
 older+0.114
 towns+0.112
Neg logit contributions
dd-0.118
night-0.090
ez-0.089
amb-0.083
oo-0.079
Token said
Token ID531
Feature activation+7.47e-02

Pos logit contributions
dd+0.215
night+0.177
ez+0.160
 pity+0.151
amb+0.150
Neg logit contributions
 groups-0.242
bars-0.239
 towns-0.212
 fighters-0.211
 younger-0.210
Token,
Token ID11
Feature activation-1.37e-03

Pos logit contributions
dd+0.135
night+0.112
ez+0.102
 pity+0.097
amb+0.096
Neg logit contributions
 groups-0.136
bars-0.133
 towns-0.119
 eight-0.119
 fighters-0.118
Token land
Token ID1956
Feature activation-2.29e-03

Pos logit contributions
dd+0.062
night+0.052
ez+0.047
 pity+0.045
amb+0.044
Neg logit contributions
 groups-0.065
bars-0.064
 towns-0.056
 fighters-0.056
 younger-0.056
Token,
Token ID11
Feature activation+1.06e-02

Pos logit contributions
bars+0.005
 groups+0.005
 fighters+0.005
 department+0.005
 younger+0.005
Neg logit contributions
dd-0.003
night-0.002
--0.002
oo-0.002
 it-0.002
Token there
Token ID612
Feature activation-1.18e-02

Pos logit contributions
dd+0.016
night+0.012
 it+0.012
 pity+0.010
uggle+0.010
Neg logit contributions
bars-0.022
 groups-0.021
 department-0.020
 fighters-0.020
lights-0.019
Token were
Token ID547
Feature activation-1.64e-01

Pos logit contributions
 groups+0.022
bars+0.022
 towns+0.019
 fighters+0.019
 younger+0.019
Neg logit contributions
dd-0.021
night-0.017
ez-0.016
 pity-0.015
amb-0.015
Token two
Token ID734
Feature activation-1.96e-01

Pos logit contributions
bars+0.358
 groups+0.314
ge+0.302
lights+0.299
iful+0.298
Neg logit contributions
dd-0.215
 it-0.174
 news-0.169
night-0.162
ggy-0.146
Token groups
Token ID2628
Feature activation-5.02e-01

Pos logit contributions
bars+0.407
oes+0.344
 groups+0.329
 fighters+0.326
 department+0.325
Neg logit contributions
dd-0.294
--0.251
night-0.216
 puppy-0.212
 news-0.211
Token of
Token ID286
Feature activation-1.55e-01

Pos logit contributions
bars+0.886
 groups+0.842
 younger+0.769
 eight+0.766
 fighters+0.765
Neg logit contributions
dd-0.873
night-0.751
--0.662
 happened-0.652
ez-0.646
Token animals
Token ID4695
Feature activation-1.09e-01

Pos logit contributions
bars+0.233
 groups+0.221
 fighters+0.192
lights+0.191
 department+0.189
Neg logit contributions
dd-0.326
night-0.277
ez-0.257
 pity-0.252
amb-0.242
Token.
Token ID13
Feature activation-3.57e-02

Pos logit contributions
bars+0.204
 groups+0.202
 younger+0.178
 towns+0.177
 fighters+0.176
Neg logit contributions
dd-0.191
night-0.157
ez-0.143
 pity-0.135
amb-0.134
Token They
Token ID1119
Feature activation-5.56e-02

Pos logit contributions
bars+0.077
 groups+0.072
 fighters+0.067
 department+0.067
 fastest+0.067
Neg logit contributions
dd-0.048
night-0.040
 Now-0.034
 It-0.034
ez-0.034
Token were
Token ID547
Feature activation-1.68e-01

Pos logit contributions
bars+0.129
 groups+0.115
lights+0.114
ents+0.109
 younger+0.109
Neg logit contributions
dd-0.076
night-0.056
 pity-0.049
 it-0.049
ez-0.048
Token They
Token ID1119
Feature activation-5.61e-02

Pos logit contributions
bars+0.063
 groups+0.061
 fastest+0.059
 fighters+0.058
lights+0.058
Neg logit contributions
dd-0.035
night-0.034
 It-0.029
 Now-0.029
amb-0.027
Token were
Token ID547
Feature activation-1.67e-01

Pos logit contributions
bars+0.110
lights+0.103
 groups+0.102
 fastest+0.097
 bridges+0.094
Neg logit contributions
dd-0.091
night-0.067
 it-0.065
 pity-0.064
 happened-0.063
Token all
Token ID477
Feature activation-6.77e-02

Pos logit contributions
bars+0.316
 groups+0.299
lights+0.274
 department+0.266
 fighters+0.264
Neg logit contributions
dd-0.287
night-0.228
 pity-0.210
ez-0.208
oo-0.204
Token organizing
Token ID16924
Feature activation-7.24e-02

Pos logit contributions
bars+0.136
 groups+0.127
lights+0.122
ains+0.120
 department+0.118
Neg logit contributions
dd-0.100
night-0.082
ez-0.077
 pity-0.074
faced-0.072
Token into
Token ID656
Feature activation-2.14e-01

Pos logit contributions
bars+0.131
 groups+0.129
 fighters+0.116
 eight+0.115
ents+0.114
Neg logit contributions
dd-0.127
night-0.106
ez-0.095
 pity-0.094
uggle-0.089
Token groups
Token ID2628
Feature activation-5.02e-01

Pos logit contributions
bars+0.375
 department+0.339
 fighters+0.334
oes+0.330
lights+0.316
Neg logit contributions
dd-0.361
 pity-0.299
 it-0.292
--0.285
night-0.283
Token and
Token ID290
Feature activation+5.72e-02

Pos logit contributions
bars+1.050
 eight+1.011
 narrower+1.010
 groups+1.009
 department+0.987
Neg logit contributions
dd-0.599
--0.483
night-0.477
 it-0.425
uggle-0.394
Token chatting
Token ID32046
Feature activation-1.04e-01

Pos logit contributions
dd+0.095
night+0.079
 it+0.071
 pity+0.071
ez+0.068
Neg logit contributions
bars-0.104
 groups-0.099
ains-0.090
lights-0.090
 eight-0.089
Token.
Token ID13
Feature activation-3.20e-02

Pos logit contributions
 groups+0.204
bars+0.199
 towns+0.179
 fighters+0.177
 younger+0.177
Neg logit contributions
dd-0.173
night-0.141
ez-0.127
 pity-0.120
amb-0.119
Token
Token ID198
Feature activation+4.40e-02

Pos logit contributions
bars+0.065
 groups+0.063
 fighters+0.059
lights+0.058
 fastest+0.058
Neg logit contributions
dd-0.046
night-0.040
 Now-0.034
amb-0.034
ez-0.033
Token
Token ID198
Feature activation+4.32e-02

Pos logit contributions
dd+0.074
night+0.062
ez+0.055
 pity+0.052
amb+0.051
Neg logit contributions
 groups-0.086
bars-0.083
 towns-0.075
 eight-0.075
 younger-0.074
Tokencastle
Token ID18676
Feature activation-9.08e-02

Pos logit contributions
dd+0.082
night+0.071
-+0.066
amb+0.061
 it+0.059
Neg logit contributions
 groups-0.119
bars-0.119
 eight-0.112
 younger-0.111
 towns-0.110
Token and
Token ID290
Feature activation+8.85e-02

Pos logit contributions
 groups+0.197
bars+0.183
 eight+0.181
 towns+0.179
 department+0.177
Neg logit contributions
dd-0.121
night-0.102
--0.091
 it-0.084
amb-0.084
Token started
Token ID2067
Feature activation-5.90e-02

Pos logit contributions
dd+0.157
night+0.130
ez+0.119
 pity+0.112
amb+0.112
Neg logit contributions
 groups-0.164
bars-0.161
 towns-0.143
 fighters-0.143
 younger-0.142
Token to
Token ID284
Feature activation+1.20e-02

Pos logit contributions
bars+0.114
 groups+0.113
 department+0.100
 fighters+0.100
lights+0.100
Neg logit contributions
dd-0.095
night-0.078
ez-0.070
 pity-0.068
--0.066
Token eat
Token ID4483
Feature activation+1.58e-01

Pos logit contributions
dd+0.022
night+0.019
ez+0.017
 pity+0.017
faced+0.016
Neg logit contributions
bars-0.021
 groups-0.020
 fighters-0.018
lights-0.018
ents-0.018
Token it
Token ID340
Feature activation+3.26e-01

Pos logit contributions
dd+0.284
night+0.237
ez+0.213
 pity+0.206
amb+0.200
Neg logit contributions
 groups-0.287
bars-0.284
 towns-0.251
 fighters-0.250
 younger-0.248
Token.
Token ID13
Feature activation-9.72e-04

Pos logit contributions
dd+0.438
night+0.377
-+0.327
 it+0.304
ggy+0.296
Neg logit contributions
bars-0.709
 groups-0.694
 towns-0.663
 eight-0.642
 fighters-0.638
Token He
Token ID679
Feature activation-3.00e-02

Pos logit contributions
 groups+0.002
bars+0.002
 eight+0.002
 towns+0.002
 younger+0.002
Neg logit contributions
dd-0.002
night-0.001
ez-0.001
 pity-0.001
amb-0.001
Token ate
Token ID15063
Feature activation+1.48e-01

Pos logit contributions
 groups+0.058
bars+0.057
 towns+0.051
 fighters+0.050
 younger+0.050
Neg logit contributions
dd-0.051
night-0.042
ez-0.038
 pity-0.036
faced-0.035
Token the
Token ID262
Feature activation-1.19e-01

Pos logit contributions
dd+0.278
night+0.233
ez+0.216
amb+0.208
 pity+0.201
Neg logit contributions
 groups-0.258
bars-0.247
 bridges-0.222
 eight-0.222
 towns-0.220
Token sand
Token ID6450
Feature activation+5.33e-02

Pos logit contributions
 groups+0.225
bars+0.221
 towns+0.197
 fighters+0.196
 younger+0.195
Neg logit contributions
dd-0.207
night-0.172
ez-0.154
 pity-0.147
amb-0.144
Token friend
Token ID1545
Feature activation+2.77e-02

Pos logit contributions
dd+0.241
night+0.200
ez+0.190
amb+0.182
uggle+0.181
Neg logit contributions
 groups-0.154
bars-0.135
 older-0.135
 younger-0.134
 fighters-0.127
Token Sam
Token ID3409
Feature activation-2.65e-02

Pos logit contributions
dd+0.039
night+0.033
ez+0.029
 pity+0.027
ggy+0.027
Neg logit contributions
bars-0.059
 groups-0.057
lights-0.053
 towns-0.052
 fighters-0.051
Token.
Token ID13
Feature activation-4.28e-02

Pos logit contributions
bars+0.051
 groups+0.050
lights+0.046
 fighters+0.045
 towns+0.045
Neg logit contributions
dd-0.043
night-0.037
ez-0.033
 pity-0.030
--0.030
Token She
Token ID1375
Feature activation-6.96e-03

Pos logit contributions
bars+0.083
 groups+0.082
lights+0.074
 fighters+0.073
 towns+0.073
Neg logit contributions
dd-0.066
night-0.058
amb-0.049
faced-0.049
 pity-0.049
Token put
Token ID1234
Feature activation-1.93e-03

Pos logit contributions
 groups+0.013
bars+0.012
 fighters+0.011
 younger+0.011
 eight+0.011
Neg logit contributions
dd-0.012
night-0.011
ez-0.010
uggle-0.009
 pity-0.009
Token it
Token ID340
Feature activation+2.90e-01

Pos logit contributions
 groups+0.003
bars+0.003
 eight+0.003
 bridges+0.003
 forks+0.003
Neg logit contributions
dd-0.004
night-0.003
amb-0.003
ager-0.003
des-0.003
Token in
Token ID287
Feature activation+2.31e-02

Pos logit contributions
dd+0.387
night+0.309
ez+0.259
 pity+0.241
amb+0.238
Neg logit contributions
 groups-0.655
bars-0.646
 fighters-0.590
 towns-0.589
 younger-0.580
Token her
Token ID607
Feature activation+1.03e-01

Pos logit contributions
dd+0.032
night+0.026
ez+0.023
 pity+0.022
oof+0.020
Neg logit contributions
bars-0.051
 groups-0.049
 fighters-0.046
 towns-0.045
 department-0.045
Token backpack
Token ID22526
Feature activation-8.77e-03

Pos logit contributions
dd+0.207
night+0.178
ez+0.162
 pity+0.157
amb+0.154
Neg logit contributions
 groups-0.166
bars-0.163
 towns-0.142
 fighters-0.142
 younger-0.139
Token and
Token ID290
Feature activation+4.98e-02

Pos logit contributions
 groups+0.020
bars+0.019
 fighters+0.019
 towns+0.019
 eight+0.018
Neg logit contributions
dd-0.010
night-0.009
--0.007
ez-0.007
amb-0.006
Token went
Token ID1816
Feature activation+4.48e-02

Pos logit contributions
dd+0.092
night+0.077
ez+0.071
 pity+0.067
amb+0.066
Neg logit contributions
 groups-0.089
bars-0.087
 towns-0.077
 fighters-0.077
 younger-0.076
Token,
Token ID11
Feature activation-7.56e-03

Pos logit contributions
dd+0.116
night+0.096
ez+0.087
 pity+0.083
amb+0.081
Neg logit contributions
 groups-0.125
bars-0.123
 towns-0.109
 fighters-0.109
 younger-0.108
Token "
Token ID366
Feature activation-1.12e-01

Pos logit contributions
 groups+0.014
bars+0.014
 towns+0.012
 fastest+0.012
lights+0.012
Neg logit contributions
dd-0.013
night-0.011
oo-0.010
 it-0.010
faced-0.010
TokenLet
Token ID5756
Feature activation+1.12e-02

Pos logit contributions
 groups+0.211
bars+0.209
 towns+0.185
 fighters+0.184
 eight+0.183
Neg logit contributions
dd-0.194
night-0.162
ez-0.146
 pity-0.140
amb-0.136
Token's
Token ID338
Feature activation-1.08e-01

Pos logit contributions
dd+0.028
night+0.024
ez+0.023
amb+0.022
 pity+0.022
Neg logit contributions
 groups-0.012
bars-0.012
 older-0.010
 younger-0.010
 towns-0.010
Token hold
Token ID1745
Feature activation-2.31e-02

Pos logit contributions
 groups+0.190
bars+0.184
 towns+0.164
 fighters+0.163
 younger+0.162
Neg logit contributions
dd-0.202
night-0.169
ez-0.155
 pity-0.146
amb-0.144
Token it
Token ID340
Feature activation+2.87e-01

Pos logit contributions
 groups+0.040
bars+0.039
 eight+0.035
 towns+0.034
 older+0.034
Neg logit contributions
dd-0.043
night-0.037
ez-0.033
amb-0.032
 pity-0.032
Token together
Token ID1978
Feature activation+2.90e-02

Pos logit contributions
dd+0.412
night+0.372
-+0.315
amb+0.300
uggle+0.290
Neg logit contributions
 groups-0.588
bars-0.571
 towns-0.547
 fighters-0.525
 younger-0.517
Token."
Token ID526
Feature activation-9.44e-02

Pos logit contributions
dd+0.035
night+0.030
 it+0.028
-+0.028
oo+0.023
Neg logit contributions
bars-0.065
 groups-0.063
 narrower-0.060
 towns-0.059
lights-0.059
Token
Token ID220
Feature activation-4.35e-03

Pos logit contributions
 groups+0.187
bars+0.184
 fighters+0.166
 towns+0.165
 younger+0.163
Neg logit contributions
dd-0.152
night-0.127
ez-0.110
amb-0.106
 pity-0.106
Token
Token ID198
Feature activation+3.19e-02

Pos logit contributions
 groups+0.009
bars+0.009
 towns+0.008
 fighters+0.008
 eight+0.008
Neg logit contributions
dd-0.007
night-0.005
ez-0.005
 pity-0.004
amb-0.004
Token
Token ID198
Feature activation+3.11e-02

Pos logit contributions
dd+0.053
night+0.044
ez+0.040
 pity+0.038
amb+0.036
Neg logit contributions
 groups-0.063
bars-0.061
 towns-0.055
 fighters-0.054
 eight-0.054
Token,
Token ID11
Feature activation+6.75e-03

Pos logit contributions
bars+0.081
 groups+0.081
 eight+0.072
ents+0.072
ains+0.070
Neg logit contributions
dd-0.056
night-0.050
--0.043
 pity-0.040
 happened-0.040
Token they
Token ID484
Feature activation-1.11e-01

Pos logit contributions
dd+0.012
night+0.010
 it+0.009
ez+0.009
 pity+0.009
Neg logit contributions
bars-0.013
 groups-0.012
 department-0.011
 fighters-0.011
 younger-0.011
Token remembered
Token ID12086
Feature activation-7.97e-02

Pos logit contributions
bars+0.208
 groups+0.205
lights+0.184
 younger+0.182
 bridges+0.180
Neg logit contributions
dd-0.192
night-0.159
ez-0.139
 pity-0.136
amb-0.134
Token how
Token ID703
Feature activation-1.19e-01

Pos logit contributions
bars+0.163
 groups+0.157
 fastest+0.141
lights+0.140
 fighters+0.139
Neg logit contributions
dd-0.128
night-0.107
ez-0.091
 pity-0.090
 it-0.087
Token nice
Token ID3621
Feature activation+7.43e-02

Pos logit contributions
bars+0.244
 groups+0.242
 fighters+0.216
 towns+0.214
 department+0.213
Neg logit contributions
dd-0.190
night-0.154
ez-0.132
 pity-0.129
faced-0.125
Token it
Token ID340
Feature activation+3.00e-01

Pos logit contributions
dd+0.133
night+0.111
ez+0.099
 pity+0.096
amb+0.094
Neg logit contributions
 groups-0.135
bars-0.133
 towns-0.117
 fighters-0.117
 younger-0.117
Token was
Token ID373
Feature activation+3.36e-02

Pos logit contributions
dd+0.436
night+0.359
 happened+0.316
ez+0.302
 pity+0.299
Neg logit contributions
bars-0.634
 groups-0.630
 eight-0.583
ents-0.567
 towns-0.558
Token to
Token ID284
Feature activation-2.24e-02

Pos logit contributions
dd+0.053
night+0.044
ez+0.037
-+0.037
 happened+0.037
Neg logit contributions
bars-0.070
 groups-0.067
lights-0.061
 eight-0.061
 towns-0.060
Token share
Token ID2648
Feature activation+4.67e-02

Pos logit contributions
 groups+0.040
bars+0.038
 towns+0.035
 eight+0.034
 younger+0.034
Neg logit contributions
dd-0.042
night-0.035
ez-0.032
amb-0.030
 pity-0.030
Token different
Token ID1180
Feature activation-1.03e-01

Pos logit contributions
dd+0.088
night+0.073
ez+0.066
amb+0.065
 pity+0.062
Neg logit contributions
 groups-0.087
bars-0.078
 towns-0.073
 bridges-0.072
 directions-0.071
Token things
Token ID1243
Feature activation+6.36e-02

Pos logit contributions
bars+0.162
 groups+0.158
 fighters+0.139
 department+0.137
 eight+0.136
Neg logit contributions
dd-0.205
night-0.178
ez-0.162
 pity-0.156
amb-0.153
Token and
Token ID290
Feature activation+4.83e-02

Pos logit contributions
dd+0.088
-+0.076
night+0.073
 it+0.068
oo+0.062
Neg logit contributions
bars-0.153
 groups-0.150
 forks-0.143
 bowls-0.142
lights-0.133
Token covered
Token ID5017
Feature activation-8.84e-02

Pos logit contributions
dd+0.077
 it+0.066
night+0.065
 pity+0.059
 happened+0.058
Neg logit contributions
 groups-0.093
bars-0.090
ill-0.080
ains-0.079
 towns-0.078
Token the
Token ID262
Feature activation-1.45e-01

Pos logit contributions
bars+0.172
 groups+0.167
ents+0.150
 younger+0.149
 eight+0.146
Neg logit contributions
dd-0.150
night-0.124
ez-0.109
--0.105
 pity-0.103
Token city
Token ID1748
Feature activation-1.13e-01

Pos logit contributions
bars+0.248
 groups+0.226
lights+0.203
ents+0.203
ill+0.199
Neg logit contributions
dd-0.255
night-0.244
 news-0.214
 pity-0.210
 wish-0.207
Token.
Token ID13
Feature activation-4.55e-02

Pos logit contributions
 groups+0.236
bars+0.225
 eight+0.219
ains+0.213
 fighters+0.212
Neg logit contributions
dd-0.135
--0.128
night-0.122
 it-0.117
faced-0.107
Token It
Token ID632
Feature activation+2.70e-01

Pos logit contributions
bars+0.100
 groups+0.099
 fastest+0.090
lights+0.090
 fighters+0.090
Neg logit contributions
dd-0.057
night-0.048
 Now-0.043
 It-0.041
ez-0.041
Token was
Token ID373
Feature activation+2.25e-02

Pos logit contributions
dd+0.364
night+0.291
ez+0.258
 happened+0.243
 it+0.239
Neg logit contributions
bars-0.610
 groups-0.606
 forks-0.540
ents-0.529
 towns-0.528
Token so
Token ID523
Feature activation-1.07e-01

Pos logit contributions
dd+0.035
night+0.029
 pity+0.027
ez+0.026
 it+0.026
Neg logit contributions
bars-0.047
 groups-0.044
 towns-0.040
ains-0.039
oes-0.039
Token big
Token ID1263
Feature activation-6.53e-02

Pos logit contributions
bars+0.215
 groups+0.212
 fighters+0.188
 towns+0.187
ents+0.184
Neg logit contributions
dd-0.174
night-0.144
 pity-0.123
ez-0.123
faced-0.116
Token that
Token ID326
Feature activation+1.17e-01

Pos logit contributions
 groups+0.134
bars+0.132
 towns+0.119
 fighters+0.119
 younger+0.117
Neg logit contributions
dd-0.101
night-0.082
ez-0.073
 pity-0.069
amb-0.068
Token it
Token ID340
Feature activation+2.89e-01

Pos logit contributions
dd+0.143
night+0.108
 it+0.105
 pity+0.087
 happened+0.085
Neg logit contributions
 groups-0.274
bars-0.270
ents-0.251
 department-0.241
 eight-0.241
Token had
Token ID550
Feature activation-5.81e-02

Pos logit contributions
dd+0.011
night+0.009
amb+0.008
ez+0.008
 pity+0.008
Neg logit contributions
 groups-0.011
bars-0.010
 eight-0.009
 fighters-0.009
 towns-0.009
Token st
Token ID336
Feature activation-4.06e-02

Pos logit contributions
bars+0.123
 groups+0.119
lights+0.111
 towns+0.108
 fighters+0.108
Neg logit contributions
dd-0.091
night-0.072
ez-0.070
 pity-0.066
 happened-0.061
Tokenung
Token ID2150
Feature activation+1.18e-01

Pos logit contributions
 groups+0.098
bars+0.097
 towns+0.090
 fighters+0.089
 younger+0.088
Neg logit contributions
dd-0.051
night-0.037
ez-0.034
oo-0.033
amb-0.032
Token her
Token ID607
Feature activation+1.20e-01

Pos logit contributions
dd+0.163
night+0.130
ez+0.111
 pity+0.106
amb+0.098
Neg logit contributions
 groups-0.267
bars-0.262
 fighters-0.240
 towns-0.238
 younger-0.234
Token.
Token ID13
Feature activation-2.08e-02

Pos logit contributions
dd+0.211
night+0.183
ez+0.165
 pity+0.158
amb+0.155
Neg logit contributions
 groups-0.219
bars-0.213
 towns-0.195
 fighters-0.191
 department-0.185
Token
Token ID198
Feature activation+5.81e-02

Pos logit contributions
 groups+0.044
bars+0.043
 towns+0.039
 fighters+0.039
 younger+0.039
Neg logit contributions
dd-0.031
night-0.025
ez-0.022
 pity-0.021
amb-0.021
Token
Token ID198
Feature activation+5.55e-02

Pos logit contributions
dd+0.115
night+0.106
ez+0.090
amb+0.088
 pity+0.087
Neg logit contributions
 groups-0.094
bars-0.087
 eight-0.083
ents-0.080
 towns-0.078
TokenSam
Token ID16305
Feature activation+2.79e-02

Pos logit contributions
dd+0.107
night+0.090
ez+0.082
 pity+0.079
amb+0.078
Neg logit contributions
 groups-0.095
bars-0.094
 fighters-0.082
 towns-0.082
 younger-0.081
Token ran
Token ID4966
Feature activation-1.47e-02

Pos logit contributions
dd+0.044
night+0.035
ez+0.034
 pity+0.030
ggy+0.030
Neg logit contributions
 groups-0.058
bars-0.058
lights-0.052
 fighters-0.051
 towns-0.051
Token to
Token ID284
Feature activation-1.75e-02

Pos logit contributions
 groups+0.020
bars+0.019
 eight+0.018
 older+0.018
 department+0.018
Neg logit contributions
dd-0.033
ez-0.027
night-0.027
amb-0.024
oof-0.024
Token her
Token ID607
Feature activation+1.17e-01

Pos logit contributions
 groups+0.023
 older+0.020
bars+0.020
 younger+0.020
 eight+0.019
Neg logit contributions
dd-0.042
night-0.036
ez-0.034
amb-0.033
faced-0.031
Token They
Token ID1119
Feature activation-6.83e-02

Pos logit contributions
 groups+0.088
bars+0.085
 towns+0.076
 fighters+0.076
 eight+0.076
Neg logit contributions
dd-0.099
night-0.082
ez-0.076
 pity-0.073
amb-0.071
Token were
Token ID547
Feature activation-1.81e-01

Pos logit contributions
bars+0.132
 groups+0.132
 fighters+0.118
lights+0.118
 department+0.116
Neg logit contributions
dd-0.114
night-0.094
ez-0.082
 pity-0.079
amb-0.077
Token very
Token ID845
Feature activation+1.31e-02

Pos logit contributions
bars+0.376
 groups+0.364
lights+0.323
 department+0.322
 towns+0.322
Neg logit contributions
dd-0.283
night-0.229
ez-0.203
 pity-0.197
oo-0.191
Token happy
Token ID3772
Feature activation+8.44e-02

Pos logit contributions
dd+0.018
night+0.014
ez+0.012
 pity+0.012
amb+0.011
Neg logit contributions
 groups-0.029
bars-0.029
 towns-0.026
 fighters-0.026
 eight-0.026
Token and
Token ID290
Feature activation+4.24e-02

Pos logit contributions
dd+0.159
night+0.131
ez+0.124
 pity+0.120
faced+0.117
Neg logit contributions
 groups-0.148
bars-0.139
 fighters-0.131
 towns-0.129
 department-0.126
Token danced
Token ID39480
Feature activation+2.49e-02

Pos logit contributions
dd+0.077
night+0.062
ez+0.061
 pity+0.056
amb+0.055
Neg logit contributions
 groups-0.075
bars-0.073
 towns-0.067
 younger-0.067
 fighters-0.066
Token together
Token ID1978
Feature activation+2.78e-02

Pos logit contributions
dd+0.046
night+0.037
ez+0.036
 pity+0.034
oof+0.033
Neg logit contributions
 groups-0.044
bars-0.040
 towns-0.039
 fighters-0.038
lights-0.037
Token.
Token ID13
Feature activation-5.23e-02

Pos logit contributions
dd+0.045
night+0.038
ez+0.033
 pity+0.031
-+0.031
Neg logit contributions
bars-0.056
 groups-0.054
 fighters-0.048
 department-0.048
 towns-0.048
Token
Token ID198
Feature activation+2.45e-02

Pos logit contributions
 groups+0.086
bars+0.082
 eight+0.076
 towns+0.075
 fighters+0.074
Neg logit contributions
dd-0.105
night-0.086
ez-0.080
 pity-0.078
amb-0.074
Token
Token ID198
Feature activation+2.39e-02

Pos logit contributions
dd+0.051
night+0.046
 pity+0.041
ez+0.041
amb+0.040
Neg logit contributions
 groups-0.035
 eight-0.031
bars-0.030
ents-0.030
 bridges-0.029
TokenBut
Token ID1537
Feature activation-3.02e-02

Pos logit contributions
dd+0.054
night+0.049
ez+0.045
 pity+0.044
amb+0.044
Neg logit contributions
 groups-0.031
bars-0.028
 eight-0.026
 towns-0.025
 bridges-0.025
Token farm
Token ID5318
Feature activation-2.09e-03

Pos logit contributions
 groups+0.005
bars+0.005
 fighters+0.005
 towns+0.004
 younger+0.004
Neg logit contributions
dd-0.005
night-0.004
ez-0.004
 pity-0.004
faced-0.004
Token,
Token ID11
Feature activation-3.89e-03

Pos logit contributions
bars+0.004
 groups+0.004
lights+0.004
 eight+0.004
 thirteen+0.004
Neg logit contributions
dd-0.003
night-0.003
--0.002
ggy-0.002
ez-0.002
Token so
Token ID523
Feature activation-1.01e-01

Pos logit contributions
bars+0.006
 groups+0.006
lights+0.006
 towns+0.005
 fighters+0.005
Neg logit contributions
dd-0.007
night-0.006
 it-0.006
 pity-0.006
ez-0.006
Token he
Token ID339
Feature activation-4.57e-02

Pos logit contributions
 groups+0.162
bars+0.161
 eight+0.140
 towns+0.140
 fighters+0.139
Neg logit contributions
dd-0.200
night-0.170
ez-0.159
 pity-0.151
amb-0.149
Token got
Token ID1392
Feature activation-7.25e-03

Pos logit contributions
 groups+0.084
bars+0.082
 towns+0.073
 fighters+0.073
 younger+0.072
Neg logit contributions
dd-0.081
night-0.068
ez-0.062
 pity-0.059
amb-0.058
Token a
Token ID257
Feature activation+4.70e-02

Pos logit contributions
 groups+0.013
bars+0.012
 older+0.011
 younger+0.011
 eight+0.011
Neg logit contributions
dd-0.014
night-0.011
ez-0.010
amb-0.010
oof-0.010
Token camera
Token ID4676
Feature activation+4.33e-02

Pos logit contributions
dd+0.098
ez+0.077
night+0.075
amb+0.074
 Now+0.067
Neg logit contributions
 groups-0.080
 younger-0.074
 older-0.074
 eight-0.072
bars-0.071
Token.
Token ID13
Feature activation-4.34e-02

Pos logit contributions
dd+0.075
night+0.063
ez+0.056
amb+0.053
 pity+0.053
Neg logit contributions
 groups-0.081
bars-0.079
 towns-0.071
 fighters-0.070
 eight-0.070
Token He
Token ID679
Feature activation-6.70e-02

Pos logit contributions
 groups+0.080
bars+0.078
 fighters+0.070
 towns+0.070
 younger+0.069
Neg logit contributions
dd-0.077
night-0.065
ez-0.058
 pity-0.055
amb-0.055
Token put
Token ID1234
Feature activation-1.22e-03

Pos logit contributions
 groups+0.123
bars+0.120
 fighters+0.107
 towns+0.106
 eight+0.106
Neg logit contributions
dd-0.119
night-0.100
ez-0.090
 pity-0.087
uggle-0.085
Token the
Token ID262
Feature activation-1.50e-01

Pos logit contributions
 groups+0.002
bars+0.002
 eight+0.002
 bridges+0.001
 older+0.001
Neg logit contributions
dd-0.003
night-0.002
ez-0.002
amb-0.002
ggy-0.002
Token said
Token ID531
Feature activation+6.79e-02

Pos logit contributions
dd+0.112
night+0.091
ez+0.080
 pity+0.076
amb+0.074
Neg logit contributions
 groups-0.152
bars-0.149
 fighters-0.134
 towns-0.134
 younger-0.134
Token.
Token ID13
Feature activation-3.96e-02

Pos logit contributions
dd+0.120
night+0.103
ez+0.091
 pity+0.087
amb+0.085
Neg logit contributions
 groups-0.126
bars-0.122
 towns-0.110
 fighters-0.109
 eight-0.107
Token
Token ID198
Feature activation+3.57e-02

Pos logit contributions
 groups+0.079
bars+0.077
 towns+0.069
 fighters+0.069
 younger+0.069
Neg logit contributions
dd-0.065
night-0.053
ez-0.047
 pity-0.045
amb-0.044
Token
Token ID198
Feature activation+3.46e-02

Pos logit contributions
dd+0.062
night+0.052
ez+0.047
 pity+0.045
faced+0.043
Neg logit contributions
 groups-0.067
bars-0.065
 towns-0.059
 eight-0.058
 younger-0.058
TokenBen
Token ID11696
Feature activation+4.42e-02

Pos logit contributions
dd+0.049
night+0.031
ggy+0.029
amb+0.028
faced+0.027
Neg logit contributions
bars-0.081
lights-0.077
 department-0.074
 groups-0.074
 fighters-0.073
Token went
Token ID1816
Feature activation+4.19e-02

Pos logit contributions
dd+0.075
night+0.061
ez+0.055
 pity+0.053
amb+0.052
Neg logit contributions
 groups-0.085
bars-0.084
 towns-0.074
lights-0.074
 fighters-0.074
Token near
Token ID1474
Feature activation-8.79e-02

Pos logit contributions
dd+0.080
night+0.067
ez+0.062
 pity+0.058
amb+0.058
Neg logit contributions
 groups-0.073
bars-0.071
 towns-0.063
 fighters-0.062
 younger-0.062
Token the
Token ID262
Feature activation-1.46e-01

Pos logit contributions
bars+0.158
 groups+0.142
 fighters+0.134
lights+0.134
 department+0.133
Neg logit contributions
dd-0.157
night-0.135
 it-0.127
 pity-0.126
ez-0.126
Token chicken
Token ID9015
Feature activation+9.35e-02

Pos logit contributions
bars+0.319
 groups+0.313
lights+0.284
 towns+0.280
 fighters+0.278
Neg logit contributions
dd-0.211
night-0.169
ez-0.145
 pity-0.143
faced-0.135
Token and
Token ID290
Feature activation+5.15e-02

Pos logit contributions
dd+0.141
night+0.121
ez+0.110
uggle+0.103
 pity+0.102
Neg logit contributions
bars-0.187
 groups-0.182
lights-0.166
 fighters-0.165
 fastest-0.163
Token reached
Token ID4251
Feature activation-6.87e-02

Pos logit contributions
dd+0.088
night+0.073
ez+0.065
 pity+0.063
faced+0.061
Neg logit contributions
 groups-0.098
bars-0.097
 towns-0.086
 fighters-0.085
lights-0.085
Token<|endoftext|>
Token ID50256
Feature activation+1.98e-01

Pos logit contributions
 groups+0.083
bars+0.080
 fighters+0.073
 fastest+0.073
 older+0.073
Neg logit contributions
dd-0.052
night-0.044
amb-0.038
ez-0.037
 Now-0.037
TokenS
Token ID50
Feature activation+8.55e-02

Pos logit contributions
dd+0.384
night+0.326
ez+0.293
 pity+0.287
faced+0.281
Neg logit contributions
 groups-0.332
bars-0.320
 towns-0.288
 eight-0.280
 fighters-0.280
Tokenue
Token ID518
Feature activation-1.33e-01

Pos logit contributions
dd+0.145
night+0.122
ez+0.107
uggle+0.104
oo+0.103
Neg logit contributions
 groups-0.162
bars-0.158
 fighters-0.144
 sizes-0.143
lights-0.143
Token was
Token ID373
Feature activation+2.57e-02

Pos logit contributions
 groups+0.239
bars+0.236
 towns+0.212
 eight+0.210
 fighters+0.206
Neg logit contributions
dd-0.233
night-0.192
 pity-0.171
ez-0.171
faced-0.167
Token very
Token ID845
Feature activation+2.37e-02

Pos logit contributions
dd+0.052
night+0.045
ez+0.040
amb+0.040
uable+0.039
Neg logit contributions
 groups-0.040
 older-0.040
 younger-0.038
 eight-0.037
bars-0.035
Token excited
Token ID6568
Feature activation+1.39e-01

Pos logit contributions
dd+0.040
night+0.033
ez+0.030
 pity+0.028
amb+0.028
Neg logit contributions
 groups-0.046
bars-0.045
 younger-0.041
 older-0.040
 eight-0.040
Token!
Token ID0
Feature activation+2.75e-02

Pos logit contributions
dd+0.257
night+0.214
ez+0.197
 pity+0.189
amb+0.185
Neg logit contributions
 groups-0.246
bars-0.241
 towns-0.213
 fighters-0.212
 younger-0.210
Token Today
Token ID6288
Feature activation+4.86e-02

Pos logit contributions
dd+0.048
night+0.040
ez+0.036
 pity+0.034
amb+0.034
Neg logit contributions
 groups-0.051
bars-0.051
 towns-0.045
 fighters-0.045
 younger-0.044
Token she
Token ID673
Feature activation+2.01e-02

Pos logit contributions
dd+0.081
night+0.067
ez+0.060
 pity+0.057
amb+0.056
Neg logit contributions
 groups-0.094
bars-0.093
 towns-0.083
 fighters-0.082
 younger-0.082
Token was
Token ID373
Feature activation+2.52e-02

Pos logit contributions
dd+0.040
night+0.033
ez+0.032
 pity+0.031
uggle+0.030
Neg logit contributions
 groups-0.032
bars-0.031
 eight-0.028
 older-0.027
 towns-0.027
Token going
Token ID1016
Feature activation+1.52e-02

Pos logit contributions
dd+0.044
night+0.036
ez+0.033
 pity+0.031
amb+0.031
Neg logit contributions
 groups-0.048
bars-0.047
 towns-0.042
 fighters-0.041
 younger-0.041
Token loved
Token ID6151
Feature activation+1.63e-01

Pos logit contributions
dd+0.129
night+0.109
ez+0.099
 pity+0.097
uggle+0.091
Neg logit contributions
 groups-0.132
bars-0.130
 towns-0.117
 fighters-0.114
 eight-0.114
Token to
Token ID284
Feature activation-3.23e-02

Pos logit contributions
dd+0.321
night+0.271
ez+0.250
 pity+0.238
amb+0.235
Neg logit contributions
 groups-0.275
bars-0.272
 towns-0.238
 fighters-0.236
 younger-0.236
Token slide
Token ID10649
Feature activation-7.74e-03

Pos logit contributions
 groups+0.062
bars+0.061
 towns+0.055
 younger+0.054
 fighters+0.054
Neg logit contributions
dd-0.055
night-0.045
ez-0.041
 pity-0.039
amb-0.038
Token down
Token ID866
Feature activation+3.05e-02

Pos logit contributions
 groups+0.018
 department+0.017
 fighters+0.017
 eight+0.016
bars+0.016
Neg logit contributions
dd-0.009
--0.007
 it-0.007
night-0.007
oo-0.006
Token it
Token ID340
Feature activation+2.88e-01

Pos logit contributions
dd+0.043
 it+0.042
-+0.036
night+0.034
 Now+0.033
Neg logit contributions
 groups-0.059
bars-0.058
 department-0.056
 fighters-0.056
 younger-0.055
Token every
Token ID790
Feature activation-1.23e-01

Pos logit contributions
dd+0.363
night+0.312
-+0.284
 it+0.270
ez+0.251
Neg logit contributions
 groups-0.620
 fighters-0.596
bars-0.595
 department-0.595
 towns-0.574
Token day
Token ID1110
Feature activation+1.35e-01

Pos logit contributions
bars+0.270
 groups+0.269
 eight+0.246
 older+0.243
 younger+0.243
Neg logit contributions
dd-0.175
night-0.130
ez-0.120
 pity-0.113
amb-0.111
Token.
Token ID13
Feature activation-4.41e-02

Pos logit contributions
dd+0.222
night+0.182
ez+0.157
-+0.154
amb+0.151
Neg logit contributions
 groups-0.265
bars-0.256
 fighters-0.242
 department-0.242
 towns-0.234
Token But
Token ID887
Feature activation+1.37e-01

Pos logit contributions
 groups+0.083
bars+0.082
 towns+0.072
 younger+0.072
 eight+0.071
Neg logit contributions
dd-0.078
night-0.064
ez-0.058
 pity-0.055
amb-0.054
Token one
Token ID530
Feature activation-4.18e-02

Pos logit contributions
dd+0.285
night+0.239
ez+0.226
 pity+0.217
amb+0.215
Neg logit contributions
 groups-0.212
bars-0.208
 eight-0.180
 older-0.180
 younger-0.178
Token day
Token ID1110
Feature activation+1.34e-01

Pos logit contributions
 groups+0.066
 eight+0.065
bars+0.065
 older+0.062
 first+0.061
Neg logit contributions
dd-0.078
icol-0.059
 pity-0.058
amb-0.057
night-0.057
Token together
Token ID1978
Feature activation+4.21e-02

Pos logit contributions
dd+0.229
night+0.191
ez+0.173
 pity+0.165
amb+0.163
Neg logit contributions
 groups-0.228
bars-0.225
 towns-0.198
 fighters-0.197
 younger-0.196
Token.
Token ID13
Feature activation-3.64e-02

Pos logit contributions
dd+0.065
night+0.052
ez+0.046
-+0.045
 it+0.044
Neg logit contributions
bars-0.089
 groups-0.086
 younger-0.078
 department-0.078
 older-0.078
Token One
Token ID1881
Feature activation-1.22e-01

Pos logit contributions
 groups+0.070
bars+0.068
 towns+0.061
 fighters+0.061
 younger+0.060
Neg logit contributions
dd-0.063
night-0.052
ez-0.047
 pity-0.044
amb-0.044
Token day
Token ID1110
Feature activation+1.40e-01

Pos logit contributions
 groups+0.283
bars+0.279
 towns+0.255
 younger+0.254
 fighters+0.253
Neg logit contributions
dd-0.159
night-0.120
ez-0.105
 pity-0.098
amb-0.096
Token,
Token ID11
Feature activation+2.36e-03

Pos logit contributions
dd+0.248
night+0.207
ez+0.187
 pity+0.179
amb+0.176
Neg logit contributions
 groups-0.260
bars-0.254
 towns-0.227
 eight-0.226
 fighters-0.226
Token the
Token ID262
Feature activation-1.46e-01

Pos logit contributions
dd+0.005
night+0.004
ez+0.003
 pity+0.003
faced+0.003
Neg logit contributions
bars-0.004
 groups-0.004
ents-0.003
 towns-0.003
 fighters-0.003
Token clown
Token ID25573
Feature activation+1.65e-01

Pos logit contributions
 groups+0.290
bars+0.283
 older+0.273
 fastest+0.267
 towns+0.266
Neg logit contributions
dd-0.240
ez-0.188
night-0.187
amb-0.167
oof-0.159
Token had
Token ID550
Feature activation-7.46e-02

Pos logit contributions
dd+0.272
night+0.224
ez+0.199
 pity+0.186
amb+0.186
Neg logit contributions
 groups-0.325
bars-0.323
 towns-0.285
lights-0.285
 younger-0.283
Token a
Token ID257
Feature activation+4.76e-02

Pos logit contributions
bars+0.167
 groups+0.165
lights+0.155
 department+0.149
 fighters+0.148
Neg logit contributions
dd-0.101
night-0.081
ez-0.067
 pity-0.065
 it-0.064
Token big
Token ID1263
Feature activation-6.24e-02

Pos logit contributions
dd+0.107
ez+0.087
amb+0.086
night+0.084
uggle+0.077
Neg logit contributions
 groups-0.075
 younger-0.070
 older-0.068
 eight-0.065
 fastest-0.063
Token game
Token ID983
Feature activation+4.68e-03

Pos logit contributions
 groups+0.107
 older+0.097
 younger+0.094
bars+0.092
 eight+0.087
Neg logit contributions
dd-0.134
night-0.101
ez-0.100
amb-0.099
to-0.095
Token.
Token ID13
Feature activation-1.76e-02

Pos logit contributions
 groups+0.223
bars+0.207
 towns+0.198
 fighters+0.189
 eight+0.188
Neg logit contributions
dd-0.175
night-0.152
ez-0.134
--0.129
amb-0.126
Token
Token ID198
Feature activation+5.79e-02

Pos logit contributions
 groups+0.035
bars+0.034
 towns+0.031
 fighters+0.030
 younger+0.030
Neg logit contributions
dd-0.029
night-0.024
ez-0.021
 pity-0.020
amb-0.020
Token
Token ID198
Feature activation+5.46e-02

Pos logit contributions
dd+0.102
night+0.089
ez+0.077
 pity+0.074
amb+0.073
Neg logit contributions
 groups-0.108
bars-0.105
 eight-0.094
 towns-0.093
ents-0.092
Token"
Token ID1
Feature activation-3.56e-02

Pos logit contributions
dd+0.101
night+0.081
ez+0.075
 pity+0.072
amb+0.071
Neg logit contributions
 groups-0.098
bars-0.098
 fighters-0.087
 younger-0.086
 department-0.086
TokenGrab
Token ID48400
Feature activation+7.24e-02

Pos logit contributions
 groups+0.064
bars+0.063
 eight+0.057
 towns+0.056
 fighters+0.055
Neg logit contributions
dd-0.063
night-0.054
ez-0.047
amb-0.047
oo-0.046
Token my
Token ID616
Feature activation-1.11e-01

Pos logit contributions
dd+0.149
night+0.132
ez+0.118
faced+0.114
amb+0.114
Neg logit contributions
 groups-0.108
bars-0.105
 eight-0.096
 towns-0.091
 older-0.091
Token hand
Token ID1021
Feature activation-1.12e-01

Pos logit contributions
 groups+0.203
bars+0.183
 towns+0.179
 fighters+0.177
lights+0.174
Neg logit contributions
dd-0.185
night-0.175
--0.158
 pity-0.154
ez-0.147
Token,
Token ID11
Feature activation+1.38e-02

Pos logit contributions
 groups+0.266
bars+0.249
 towns+0.245
lights+0.234
 younger+0.233
Neg logit contributions
dd-0.142
night-0.113
--0.098
amb-0.095
ez-0.093
Token Ben
Token ID3932
Feature activation-1.38e-02

Pos logit contributions
dd+0.026
night+0.022
ez+0.020
 pity+0.019
amb+0.019
Neg logit contributions
 groups-0.024
bars-0.024
 towns-0.021
 fighters-0.021
 younger-0.021
Token.
Token ID13
Feature activation-2.26e-02

Pos logit contributions
 groups+0.026
bars+0.025
 fighters+0.022
 eight+0.022
 towns+0.022
Neg logit contributions
dd-0.024
night-0.020
ez-0.018
amb-0.017
 pity-0.017
Token I
Token ID314
Feature activation-1.31e-01

Pos logit contributions
bars+0.041
 groups+0.041
 eight+0.037
 towns+0.036
 fighters+0.036
Neg logit contributions
dd-0.041
night-0.033
ez-0.029
 pity-0.028
faced-0.028
Token throw
Token ID3714
Feature activation+6.97e-02

Pos logit contributions
bars+0.109
 groups+0.108
lights+0.096
 fighters+0.096
 towns+0.094
Neg logit contributions
dd-0.097
night-0.081
 pity-0.073
ez-0.072
uggle-0.072
Token their
Token ID511
Feature activation-6.56e-02

Pos logit contributions
dd+0.135
night+0.114
ez+0.102
amb+0.100
 pity+0.096
Neg logit contributions
 groups-0.121
bars-0.116
 towns-0.103
 eight-0.102
 bridges-0.102
Token balls
Token ID11333
Feature activation-1.15e-01

Pos logit contributions
bars+0.123
 groups+0.108
 fighters+0.108
 department+0.104
 younger+0.103
Neg logit contributions
dd-0.102
night-0.087
--0.083
ez-0.082
 pity-0.081
Token to
Token ID284
Feature activation-3.36e-02

Pos logit contributions
 groups+0.245
bars+0.243
 fighters+0.222
 towns+0.221
 younger+0.220
Neg logit contributions
dd-0.164
night-0.133
ez-0.114
uggle-0.109
amb-0.107
Token each
Token ID1123
Feature activation-1.12e-01

Pos logit contributions
bars+0.056
 groups+0.054
 fighters+0.048
 department+0.047
lights+0.047
Neg logit contributions
dd-0.064
night-0.055
 pity-0.051
ez-0.050
oo-0.048
Token other
Token ID584
Feature activation-2.06e-01

Pos logit contributions
bars+0.144
 groups+0.137
 fighters+0.120
lights+0.120
 department+0.116
Neg logit contributions
dd-0.251
night-0.223
ez-0.209
 pity-0.205
--0.201
Token and
Token ID290
Feature activation+5.08e-02

Pos logit contributions
bars+0.458
 groups+0.431
 fighters+0.412
 eight+0.411
 department+0.405
Neg logit contributions
dd-0.251
night-0.223
--0.192
ez-0.188
uggle-0.174
Token catch
Token ID4929
Feature activation+2.02e-01

Pos logit contributions
dd+0.088
night+0.074
ez+0.066
 pity+0.064
amb+0.061
Neg logit contributions
 groups-0.095
bars-0.094
 fighters-0.083
 towns-0.082
 younger-0.081
Token them
Token ID606
Feature activation+1.37e-02

Pos logit contributions
dd+0.388
night+0.328
ez+0.295
amb+0.287
 pity+0.279
Neg logit contributions
 groups-0.355
bars-0.336
 towns-0.301
 younger-0.298
 older-0.296
Token.
Token ID13
Feature activation-3.91e-02

Pos logit contributions
dd+0.019
night+0.016
ez+0.015
-+0.014
 it+0.013
Neg logit contributions
bars-0.029
 groups-0.028
 towns-0.026
 eight-0.026
lights-0.025
Token They
Token ID1119
Feature activation-5.93e-02

Pos logit contributions
 groups+0.064
bars+0.063
 towns+0.055
 fighters+0.055
 younger+0.054
Neg logit contributions
dd-0.078
night-0.066
ez-0.061
 pity-0.058
amb-0.057
Token's
Token ID338
Feature activation-9.70e-02

Pos logit contributions
dd+0.275
night+0.240
ez+0.211
-+0.205
 pity+0.204
Neg logit contributions
 groups-0.283
bars-0.282
lights-0.249
ents-0.241
 towns-0.241
Token getting
Token ID1972
Feature activation-1.75e-01

Pos logit contributions
bars+0.209
 groups+0.201
lights+0.179
 towns+0.179
 fighters+0.178
Neg logit contributions
dd-0.150
night-0.121
 pity-0.105
ez-0.105
oo-0.101
Token closer
Token ID5699
Feature activation+1.26e-02

Pos logit contributions
bars+0.341
 groups+0.338
 fighters+0.300
 towns+0.298
ains+0.296
Neg logit contributions
dd-0.287
night-0.239
 pity-0.209
 it-0.206
--0.193
Token,
Token ID11
Feature activation+9.59e-03

Pos logit contributions
dd+0.016
night+0.014
-+0.012
 it+0.011
ez+0.010
Neg logit contributions
bars-0.028
 groups-0.028
 towns-0.025
 younger-0.025
lights-0.025
Token let
Token ID1309
Feature activation-1.16e-01

Pos logit contributions
dd+0.016
night+0.013
ez+0.011
amb+0.011
 pity+0.011
Neg logit contributions
 groups-0.019
bars-0.018
 towns-0.016
 fighters-0.016
 younger-0.016
Token's
Token ID338
Feature activation-9.59e-02

Pos logit contributions
 groups+0.155
bars+0.151
 younger+0.132
 eight+0.132
 towns+0.131
Neg logit contributions
dd-0.256
night-0.222
ez-0.208
 pity-0.202
amb-0.200
Token go
Token ID467
Feature activation+8.54e-03

Pos logit contributions
 groups+0.167
bars+0.161
 towns+0.145
 younger+0.144
 eight+0.143
Neg logit contributions
dd-0.181
night-0.152
ez-0.140
 pity-0.131
amb-0.130
Token."
Token ID526
Feature activation-7.62e-02

Pos logit contributions
dd+0.015
night+0.012
ez+0.011
 pity+0.011
faced+0.010
Neg logit contributions
 groups-0.016
bars-0.016
 towns-0.014
 fighters-0.014
 younger-0.014
Token Rob
Token ID3851
Feature activation+2.80e-02

Pos logit contributions
 groups+0.145
bars+0.143
 towns+0.127
 fighters+0.126
 eight+0.125
Neg logit contributions
dd-0.131
night-0.109
ez-0.097
 pity-0.093
amb-0.091
Token tremb
Token ID29868
Feature activation-5.46e-02

Pos logit contributions
dd+0.047
night+0.039
ez+0.034
amb+0.033
 pity+0.032
Neg logit contributions
 groups-0.055
bars-0.054
 towns-0.048
 bridges-0.048
lights-0.048
Tokenled
Token ID992
Feature activation+9.02e-02

Pos logit contributions
 groups+0.112
bars+0.110
 towns+0.100
lights+0.099
 older+0.098
Neg logit contributions
dd-0.083
night-0.068
ez-0.059
uggle-0.058
 pity-0.058
Token help
Token ID1037
Feature activation+2.67e-03

Pos logit contributions
 groups+0.039
bars+0.038
 towns+0.034
 fighters+0.034
 younger+0.033
Neg logit contributions
dd-0.037
night-0.031
ez-0.028
 pity-0.027
amb-0.026
Token you
Token ID345
Feature activation+1.09e-02

Pos logit contributions
dd+0.004
night+0.003
 it+0.003
-+0.003
ez+0.003
Neg logit contributions
bars-0.006
 groups-0.006
lights-0.005
 department-0.005
 fighters-0.005
Token!"
Token ID2474
Feature activation+6.51e-02

Pos logit contributions
dd+0.016
night+0.014
uggle+0.012
-+0.012
ez+0.012
Neg logit contributions
bars-0.022
 groups-0.021
lights-0.020
 fighters-0.019
 towns-0.019
Token<|endoftext|>
Token ID50256
Feature activation+2.04e-01

Pos logit contributions
dd+0.085
night+0.072
ez+0.062
amb+0.060
 Now+0.059
Neg logit contributions
 groups-0.145
bars-0.142
 fastest-0.134
 department-0.132
 fighters-0.131
TokenOnce
Token ID7454
Feature activation-1.01e-01

Pos logit contributions
dd+0.393
night+0.333
ez+0.301
 pity+0.293
amb+0.287
Neg logit contributions
 groups-0.345
bars-0.335
 towns-0.298
 fighters-0.293
 eight-0.292
Token upon
Token ID2402
Feature activation-3.10e-01

Pos logit contributions
 groups+0.109
bars+0.103
 fighters+0.094
 towns+0.092
 sizes+0.090
Neg logit contributions
dd-0.247
night-0.219
 it-0.208
 happened-0.201
--0.197
Token a
Token ID257
Feature activation+5.31e-02

Pos logit contributions
 groups+0.443
bars+0.384
 eight+0.377
 older+0.356
 younger+0.351
Neg logit contributions
dd-0.676
night-0.573
amb-0.558
ez-0.544
 pity-0.525
Token time
Token ID640
Feature activation+9.11e-02

Pos logit contributions
dd+0.088
night+0.071
ez+0.066
amb+0.065
 pity+0.064
Neg logit contributions
 groups-0.102
bars-0.093
 eight-0.090
 younger-0.087
 older-0.087
Token there
Token ID612
Feature activation-1.44e-02

Pos logit contributions
dd+0.148
night+0.120
ez+0.106
 pity+0.100
amb+0.099
Neg logit contributions
 groups-0.181
bars-0.180
 fighters-0.161
 towns-0.160
 younger-0.160
Token was
Token ID373
Feature activation+3.49e-02

Pos logit contributions
 groups+0.022
bars+0.020
ill+0.018
 towns+0.018
 fighters+0.018
Neg logit contributions
dd-0.030
night-0.026
amb-0.024
ez-0.024
 pity-0.023
Token a
Token ID257
Feature activation+5.20e-02

Pos logit contributions
dd+0.077
 pity+0.074
night+0.072
uable+0.072
person+0.070
Neg logit contributions
 groups-0.040
 eight-0.036
 older-0.030
 first-0.030
 younger-0.028
Token had
Token ID550
Feature activation-7.28e-02

Pos logit contributions
 groups+0.207
bars+0.202
 younger+0.182
 older+0.180
 towns+0.178
Neg logit contributions
dd-0.200
night-0.165
ez-0.147
 pity-0.143
ggy-0.139
Token not
Token ID407
Feature activation-1.00e-01

Pos logit contributions
bars+0.151
 groups+0.147
lights+0.137
 towns+0.128
 department+0.126
Neg logit contributions
dd-0.115
night-0.097
 pity-0.085
ez-0.084
ggy-0.083
Token touched
Token ID12615
Feature activation-2.41e-02

Pos logit contributions
bars+0.221
lights+0.214
 groups+0.212
ains+0.201
 department+0.191
Neg logit contributions
dd-0.134
night-0.113
 happened-0.108
ez-0.105
 pity-0.097
Token the
Token ID262
Feature activation-1.45e-01

Pos logit contributions
bars+0.042
 groups+0.041
 department+0.038
 fighters+0.037
ents+0.037
Neg logit contributions
dd-0.041
night-0.035
 it-0.033
ez-0.032
 pity-0.032
Token or
Token ID393
Feature activation-1.13e-01

Pos logit contributions
bars+0.274
 groups+0.266
ill+0.254
 fighters+0.248
lights+0.247
Neg logit contributions
dd-0.186
 puppy-0.171
 pity-0.168
night-0.162
 jewel-0.161
Tokennam
Token ID7402
Feature activation-2.98e-01

Pos logit contributions
 groups+0.196
bars+0.189
lands+0.172
 fighters+0.169
lights+0.168
Neg logit contributions
dd-0.185
night-0.167
 it-0.162
 happened-0.156
 pity-0.150
Tokenents
Token ID658
Feature activation-4.20e-01

Pos logit contributions
 groups+0.198
 department+0.194
 fighters+0.184
bars+0.175
 narrower+0.172
Neg logit contributions
dd-0.788
--0.697
amb-0.692
ear-0.684
ake-0.679
Token.
Token ID13
Feature activation-3.49e-02

Pos logit contributions
 groups+0.865
bars+0.830
 eight+0.770
 towns+0.770
 fighters+0.759
Neg logit contributions
dd-0.620
night-0.532
--0.485
 it-0.473
ez-0.470
Token They
Token ID1119
Feature activation-5.20e-02

Pos logit contributions
 groups+0.058
bars+0.057
 towns+0.050
 fighters+0.050
 younger+0.049
Neg logit contributions
dd-0.068
night-0.058
ez-0.053
 pity-0.051
amb-0.050
Token hoped
Token ID10719
Feature activation-5.49e-02

Pos logit contributions
 groups+0.108
bars+0.107
 fighters+0.096
 towns+0.096
 younger+0.095
Neg logit contributions
dd-0.079
night-0.064
ez-0.057
 pity-0.054
amb-0.052
Token Mom
Token ID11254
Feature activation-3.61e-02

Pos logit contributions
 groups+0.102
bars+0.099
 fastest+0.088
 eight+0.086
 towns+0.085
Neg logit contributions
dd-0.096
night-0.084
 it-0.072
 pity-0.069
ggy-0.069
Token
Token ID198
Feature activation+3.43e-02

Pos logit contributions
dd+0.062
night+0.052
ez+0.046
 pity+0.044
amb+0.043
Neg logit contributions
 groups-0.064
bars-0.062
 eight-0.057
 towns-0.057
 younger-0.055
TokenShe
Token ID3347
Feature activation-8.20e-02

Pos logit contributions
dd+0.076
night+0.064
ez+0.060
 pity+0.058
ggy+0.058
Neg logit contributions
bars-0.050
 groups-0.049
 fighters-0.044
lights-0.042
 department-0.042
Token was
Token ID373
Feature activation+1.90e-02

Pos logit contributions
 groups+0.133
bars+0.126
 eight+0.115
 younger+0.112
ents+0.112
Neg logit contributions
dd-0.162
night-0.139
uggle-0.130
ez-0.126
amb-0.125
Token scared
Token ID12008
Feature activation+1.05e-02

Pos logit contributions
dd+0.035
night+0.030
ez+0.027
 pity+0.026
amb+0.025
Neg logit contributions
 groups-0.034
bars-0.033
 towns-0.029
 fighters-0.029
 younger-0.029
Token at
Token ID379
Feature activation+9.80e-02

Pos logit contributions
dd+0.016
night+0.013
ez+0.012
-+0.011
 it+0.011
Neg logit contributions
 groups-0.022
bars-0.022
 fighters-0.020
 towns-0.020
ains-0.019
Token first
Token ID717
Feature activation-3.82e-01

Pos logit contributions
dd+0.255
night+0.234
oo+0.218
 pity+0.217
uggle+0.216
Neg logit contributions
 groups-0.083
bars-0.083
lights-0.066
 fighters-0.065
 towns-0.063
Token,
Token ID11
Feature activation-7.47e-03

Pos logit contributions
 groups+0.687
bars+0.673
 towns+0.596
 fighters+0.592
 younger+0.588
Neg logit contributions
dd-0.698
night-0.583
ez-0.528
 pity-0.506
amb-0.498
Token but
Token ID475
Feature activation-1.72e-01

Pos logit contributions
 groups+0.010
bars+0.010
 fighters+0.009
 towns+0.009
 department+0.008
Neg logit contributions
dd-0.017
night-0.014
ez-0.013
 pity-0.013
amb-0.013
Token then
Token ID788
Feature activation-4.54e-02

Pos logit contributions
 groups+0.271
bars+0.270
 eight+0.240
 older+0.240
 younger+0.240
Neg logit contributions
dd-0.345
night-0.292
ez-0.274
amb-0.267
uggle-0.258
Token she
Token ID673
Feature activation+1.22e-02

Pos logit contributions
 groups+0.083
bars+0.081
 towns+0.073
 younger+0.072
 eight+0.072
Neg logit contributions
dd-0.081
night-0.067
ez-0.062
amb-0.060
uggle-0.058
Token saw
Token ID2497
Feature activation+2.34e-02

Pos logit contributions
dd+0.021
night+0.017
ez+0.015
 pity+0.014
-+0.014
Neg logit contributions
bars-0.024
 groups-0.024
 towns-0.021
lights-0.021
 fighters-0.021
Token exciting
Token ID7895
Feature activation+1.66e-02

Pos logit contributions
 groups+0.007
bars+0.007
 older+0.007
 younger+0.006
 towns+0.006
Neg logit contributions
dd-0.006
night-0.005
ez-0.004
amb-0.004
faced-0.004
Token adventure
Token ID8855
Feature activation+1.03e-01

Pos logit contributions
dd+0.033
ez+0.026
night+0.024
amb+0.024
faced+0.024
Neg logit contributions
 groups-0.029
 towns-0.027
 directions-0.026
 older-0.026
 bridges-0.026
Token.
Token ID13
Feature activation-3.60e-02

Pos logit contributions
dd+0.179
night+0.149
ez+0.131
-+0.126
ggy+0.124
Neg logit contributions
 groups-0.196
bars-0.189
 eight-0.173
 towns-0.170
 fighters-0.169
Token<|endoftext|>
Token ID50256
Feature activation+1.98e-01

Pos logit contributions
 groups+0.057
bars+0.057
 eight+0.052
ents+0.051
 fighters+0.050
Neg logit contributions
dd-0.071
night-0.059
ez-0.054
 pity-0.053
faced-0.050
TokenOnce
Token ID7454
Feature activation-1.11e-01

Pos logit contributions
dd+0.374
night+0.316
ez+0.281
 pity+0.277
amb+0.269
Neg logit contributions
 groups-0.347
bars-0.330
 towns-0.301
 eight-0.299
 fighters-0.291
Token upon
Token ID2402
Feature activation-3.19e-01

Pos logit contributions
 groups+0.128
bars+0.123
 fighters+0.108
 towns+0.108
 department+0.102
Neg logit contributions
dd-0.267
night-0.234
oo-0.212
ez-0.211
--0.210
Token a
Token ID257
Feature activation+4.45e-02

Pos logit contributions
 groups+0.409
 eight+0.360
bars+0.338
 first+0.322
 older+0.322
Neg logit contributions
dd-0.737
night-0.619
amb-0.617
ez-0.581
 pity-0.569
Token time
Token ID640
Feature activation+8.22e-02

Pos logit contributions
dd+0.078
night+0.065
ez+0.061
 pity+0.060
amb+0.059
Neg logit contributions
 groups-0.079
bars-0.069
 eight-0.069
 younger-0.065
 older-0.065
Token,
Token ID11
Feature activation-3.28e-03

Pos logit contributions
dd+0.129
night+0.104
ez+0.091
oo+0.086
amb+0.086
Neg logit contributions
bars-0.167
 groups-0.167
 fighters-0.149
 towns-0.149
 younger-0.148
Token there
Token ID612
Feature activation-2.33e-02

Pos logit contributions
 groups+0.005
bars+0.005
 eight+0.005
 older+0.004
 towns+0.004
Neg logit contributions
dd-0.007
night-0.006
ez-0.005
amb-0.005
 pity-0.005
Token was
Token ID373
Feature activation+2.36e-02

Pos logit contributions
 groups+0.035
bars+0.033
 eight+0.030
 towns+0.029
ill+0.029
Neg logit contributions
dd-0.049
night-0.042
amb-0.039
ez-0.038
 pity-0.037
Token by
Token ID416
Feature activation-2.47e-02

Pos logit contributions
 groups+0.136
bars+0.130
ents+0.122
 eight+0.121
 department+0.121
Neg logit contributions
dd-0.087
night-0.077
ez-0.063
--0.061
 it-0.060
Token himself
Token ID2241
Feature activation-5.76e-02

Pos logit contributions
bars+0.050
 groups+0.048
 department+0.046
ents+0.044
 fighters+0.044
Neg logit contributions
dd-0.039
 it-0.029
night-0.029
ez-0.029
oo-0.028
Token.
Token ID13
Feature activation-2.86e-02

Pos logit contributions
bars+0.131
 groups+0.126
 bridges+0.123
lights+0.121
 department+0.121
Neg logit contributions
dd-0.066
night-0.054
--0.054
 it-0.049
ggy-0.042
Token<|endoftext|>
Token ID50256
Feature activation+1.99e-01

Pos logit contributions
 groups+0.054
bars+0.053
 towns+0.047
 fighters+0.047
 eight+0.047
Neg logit contributions
dd-0.050
night-0.041
ez-0.037
 pity-0.036
amb-0.034
TokenOnce
Token ID7454
Feature activation-1.08e-01

Pos logit contributions
dd+0.395
night+0.339
ez+0.301
 pity+0.298
faced+0.292
Neg logit contributions
 groups-0.325
bars-0.308
 towns-0.282
 eight-0.274
 fighters-0.271
Token upon
Token ID2402
Feature activation-3.11e-01

Pos logit contributions
 groups+0.105
bars+0.101
 fighters+0.087
 towns+0.086
 department+0.082
Neg logit contributions
dd-0.279
night-0.246
 it-0.225
ez-0.224
--0.223
Token a
Token ID257
Feature activation+5.67e-02

Pos logit contributions
 groups+0.388
 eight+0.341
bars+0.309
 first+0.301
 older+0.301
Neg logit contributions
dd-0.721
amb-0.623
night-0.622
ez-0.574
icol-0.572
Token time
Token ID640
Feature activation+9.22e-02

Pos logit contributions
dd+0.101
night+0.084
amb+0.080
ez+0.078
 pity+0.077
Neg logit contributions
 groups-0.099
 eight-0.086
bars-0.084
 younger-0.082
 older-0.081
Token,
Token ID11
Feature activation+2.28e-03

Pos logit contributions
dd+0.153
night+0.125
ez+0.112
 pity+0.106
amb+0.105
Neg logit contributions
 groups-0.180
bars-0.178
 towns-0.159
 fighters-0.158
 younger-0.157
Token there
Token ID612
Feature activation-1.40e-02

Pos logit contributions
dd+0.005
night+0.004
ez+0.004
 pity+0.004
amb+0.003
Neg logit contributions
 groups-0.004
bars-0.003
 eight-0.003
 towns-0.003
 older-0.003
Token was
Token ID373
Feature activation+3.58e-02

Pos logit contributions
 groups+0.022
bars+0.021
 eight+0.019
 towns+0.019
ill+0.018
Neg logit contributions
dd-0.028
night-0.025
amb-0.022
ez-0.022
 pity-0.021