TOP ACTIVATIONS
MAX = 0.289

he went exploring . <|endoftext|> Once upon a time , there
he went exploring . <|endoftext|> Once upon a time , there
fun playing together . <|endoftext|> Once upon a time , there
got lost again . <|endoftext|> Once upon a time , there
. The end . <|endoftext|> Once upon a time , there
he had seen . <|endoftext|> Once upon a time , there
place to explore . <|endoftext|> Once upon a time there was
the cozy rug . <|endoftext|> Once upon a time , there
a great day ! <|endoftext|> Once upon a time , there
to play together . <|endoftext|> Once upon a time , there
discover new things . <|endoftext|> Once upon a time , there
explore the world ! <|endoftext|> Once upon a time , there
in the field . <|endoftext|> Once upon a time , there
get to them . <|endoftext|> Once upon a time , there
came to cartoons ! <|endoftext|> Once upon a time there was
she loved them ! <|endoftext|> Once there was a goose who
on the ride . <|endoftext|> Once upon a time , there
time playing together . <|endoftext|> Once upon a time , there
was truly mighty ! <|endoftext|> Once upon a time , there
best day ever ! <|endoftext|> Once upon a time , there

MIN ACTIVATIONS
MIN = -0.240

not touch anything without her permission . Lily promised to follow
never go outside without her permission . <|endoftext|> Once upon a
me help you clean your wounds and give you some cookies
never sign things without my permission !" The mother
not do things without their permission . Even though it may
. Let me clean your wounds and put some band -
't explore it without my permission ." The man
without asking his dad 's permission . He knew it was
away from someone without their permission . St ealing something without
can come in without our permission ." " Yes
always better to ask for permission than to take something without
foolish things and always ask permission before entering someone 's tree
King and asked for his permission . The King said ,
someone without first asking their permission . We should ask the
belongings , and ask for permission before using them . <|endoftext|>
things without your mom 's permission . Next time , Josh
his mom 's records without permission . He would repeat to
you do not know without permission . They can be dangerous
promised to always ask for permission before drawing on anything again
jokes on others without their permission ." The little

INTERVAL 0.157 - 0.289
CONTAINS 1.837%

read her books . One day , her mom came in
!" <|endoftext|> Once upon a time , there was a house
big pile of hay . There were the white cows !
. <|endoftext|> Once upon a time , there was a graceful
free insect companion . <|endoftext|> Once upon a time there was

INTERVAL 0.025 - 0.157
CONTAINS 33.750%

game . Bob agreed and together , they played
. Max tried to pick up the stick , but it
happy and kind . They said , " We
in the park . They liked to dig in the sand
, but she needed something else . She needed some coal

INTERVAL -0.108 - 0.025
CONTAINS 62.237%

cooked for you and dad and Sam . I wanted to
ostr ich . <|endoftext|> L ily and Ben were playing in
the kids on board . The bus liked to take the
the best original recipe . The end . <|endoftext|> Once upon
ate every bite and said , " This is the best

INTERVAL -0.240 - -0.108
CONTAINS 2.174%

your colors so I can draw one too ?" Mia thought
. I will warn your mom . I will call her
showed them how the workers put the parts together , how
they didn 't find the shoes , they had fun looking
I was playing with my cars . I like them a
Token he
Token ID339
Feature activation+9.34e-02

Pos logit contributions
und+0.076
+0.066
nexpected+0.064
aded+0.063
ued+0.062
Neg logit contributions
 wounds-0.071
reating-0.064
pins-0.063
 gloves-0.059
 knee-0.059
Token went
Token ID1816
Feature activation-9.85e-03

Pos logit contributions
und+0.159
+0.142
nexpected+0.132
aded+0.127
ued+0.124
Neg logit contributions
 wounds-0.169
pins-0.146
reating-0.145
 decide-0.144
 gloves-0.143
Token exploring
Token ID13504
Feature activation+4.81e-02

Pos logit contributions
 wounds+0.017
reating+0.015
pins+0.014
 knee+0.013
 gloves+0.013
Neg logit contributions
und-0.018
-0.016
ued-0.015
aded-0.015
nexpected-0.014
Token.
Token ID13
Feature activation-1.44e-02

Pos logit contributions
und+0.096
+0.082
nexpected+0.082
aded+0.079
ued+0.079
Neg logit contributions
 wounds-0.074
pins-0.066
reating-0.066
 knee-0.060
 gloves-0.059
Token<|endoftext|>
Token ID50256
Feature activation+1.47e-01

Pos logit contributions
 wounds+0.024
pins+0.022
reating+0.022
line+0.021
 knee+0.021
Neg logit contributions
und-0.025
nexpected-0.022
aded-0.021
-0.021
ued-0.020
TokenOnce
Token ID7454
Feature activation+2.86e-01

Pos logit contributions
und+0.281
nexpected+0.267
aded+0.265
ued+0.256
+0.245
Neg logit contributions
 wounds-0.187
pins-0.170
reating-0.168
line-0.159
 jacket-0.159
Token upon
Token ID2402
Feature activation+1.91e-01

Pos logit contributions
und+0.450
+0.390
nexpected+0.375
aded+0.369
ued+0.347
Neg logit contributions
 wounds-0.530
pins-0.503
reating-0.497
line-0.469
 knee-0.468
Token a
Token ID257
Feature activation+6.75e-02

Pos logit contributions
und+0.284
+0.239
nexpected+0.228
aded+0.224
time+0.217
Neg logit contributions
 wounds-0.381
pins-0.347
reating-0.337
line-0.321
bars-0.316
Token time
Token ID640
Feature activation+2.40e-01

Pos logit contributions
und+0.091
+0.073
nexpected+0.072
ued+0.072
uttered+0.069
Neg logit contributions
 wounds-0.140
 gloves-0.129
 bike-0.124
 knee-0.124
 jacket-0.122
Token,
Token ID11
Feature activation+1.06e-01

Pos logit contributions
und+0.508
aded+0.487
nexpected+0.469
ued+0.447
ited+0.442
Neg logit contributions
pins-0.258
 bike-0.246
 back-0.236
line-0.236
 knee-0.232
Token there
Token ID612
Feature activation+2.34e-01

Pos logit contributions
und+0.188
+0.187
nexpected+0.179
aded+0.173
 contemplation+0.169
Neg logit contributions
 back-0.143
 wounds-0.141
pins-0.136
 knee-0.136
 bike-0.132
Token he
Token ID339
Feature activation+9.34e-02

Pos logit contributions
und+0.076
+0.066
nexpected+0.064
aded+0.063
ued+0.062
Neg logit contributions
 wounds-0.071
reating-0.064
pins-0.063
 gloves-0.059
 knee-0.059
Token went
Token ID1816
Feature activation-9.85e-03

Pos logit contributions
und+0.159
+0.142
nexpected+0.132
aded+0.127
ued+0.124
Neg logit contributions
 wounds-0.169
pins-0.146
reating-0.145
 decide-0.144
 gloves-0.143
Token exploring
Token ID13504
Feature activation+4.81e-02

Pos logit contributions
 wounds+0.017
reating+0.015
pins+0.014
 knee+0.013
 gloves+0.013
Neg logit contributions
und-0.018
-0.016
ued-0.015
aded-0.015
nexpected-0.014
Token.
Token ID13
Feature activation-1.44e-02

Pos logit contributions
und+0.096
+0.082
nexpected+0.082
aded+0.079
ued+0.079
Neg logit contributions
 wounds-0.074
pins-0.066
reating-0.066
 knee-0.060
 gloves-0.059
Token<|endoftext|>
Token ID50256
Feature activation+1.47e-01

Pos logit contributions
 wounds+0.024
pins+0.022
reating+0.022
line+0.021
 knee+0.021
Neg logit contributions
und-0.025
nexpected-0.022
aded-0.021
-0.021
ued-0.020
TokenOnce
Token ID7454
Feature activation+2.86e-01

Pos logit contributions
und+0.281
nexpected+0.267
aded+0.265
ued+0.256
+0.245
Neg logit contributions
 wounds-0.187
pins-0.170
reating-0.168
line-0.159
 jacket-0.159
Token upon
Token ID2402
Feature activation+1.91e-01

Pos logit contributions
und+0.450
+0.390
nexpected+0.375
aded+0.369
ued+0.347
Neg logit contributions
 wounds-0.530
pins-0.503
reating-0.497
line-0.469
 knee-0.468
Token a
Token ID257
Feature activation+6.75e-02

Pos logit contributions
und+0.284
+0.239
nexpected+0.228
aded+0.224
time+0.217
Neg logit contributions
 wounds-0.381
pins-0.347
reating-0.337
line-0.321
bars-0.316
Token time
Token ID640
Feature activation+2.40e-01

Pos logit contributions
und+0.091
+0.073
nexpected+0.072
ued+0.072
uttered+0.069
Neg logit contributions
 wounds-0.140
 gloves-0.129
 bike-0.124
 knee-0.124
 jacket-0.122
Token,
Token ID11
Feature activation+1.06e-01

Pos logit contributions
und+0.508
aded+0.487
nexpected+0.469
ued+0.447
ited+0.442
Neg logit contributions
pins-0.258
 bike-0.246
 back-0.236
line-0.236
 knee-0.232
Token there
Token ID612
Feature activation+2.34e-01

Pos logit contributions
und+0.188
+0.187
nexpected+0.179
aded+0.173
 contemplation+0.169
Neg logit contributions
 back-0.143
 wounds-0.141
pins-0.136
 knee-0.136
 bike-0.132
Token fun
Token ID1257
Feature activation+8.29e-02

Pos logit contributions
und+0.150
+0.131
nexpected+0.130
aded+0.126
ued+0.120
Neg logit contributions
 wounds-0.142
pins-0.129
reating-0.129
line-0.122
 knee-0.119
Token playing
Token ID2712
Feature activation-2.79e-02

Pos logit contributions
und+0.144
+0.123
ued+0.120
nexpected+0.120
aded+0.117
Neg logit contributions
 wounds-0.147
reating-0.129
pins-0.128
 gloves-0.122
 knee-0.122
Token together
Token ID1978
Feature activation+1.26e-01

Pos logit contributions
 wounds+0.053
reating+0.047
pins+0.047
 gloves+0.044
 knee+0.044
Neg logit contributions
und-0.046
-0.039
nexpected-0.038
ued-0.037
aded-0.037
Token.
Token ID13
Feature activation-4.13e-02

Pos logit contributions
und+0.307
nexpected+0.277
aded+0.275
+0.275
ued+0.273
Neg logit contributions
 wounds-0.129
reating-0.105
pins-0.101
 even-0.097
 gloves-0.096
Token<|endoftext|>
Token ID50256
Feature activation+1.52e-01

Pos logit contributions
 wounds+0.073
pins+0.066
reating+0.065
 knee+0.062
 gloves+0.061
Neg logit contributions
und-0.071
-0.060
nexpected-0.060
aded-0.059
ued-0.058
TokenOnce
Token ID7454
Feature activation+2.83e-01

Pos logit contributions
und+0.270
+0.233
nexpected+0.229
aded+0.223
ued+0.219
Neg logit contributions
 wounds-0.263
reating-0.236
pins-0.233
 gloves-0.218
 knee-0.217
Token upon
Token ID2402
Feature activation+1.82e-01

Pos logit contributions
und+0.402
+0.332
nexpected+0.320
aded+0.310
ued+0.303
Neg logit contributions
 wounds-0.592
reating-0.547
pins-0.544
 gloves-0.513
 knee-0.512
Token a
Token ID257
Feature activation+6.15e-02

Pos logit contributions
und+0.296
+0.259
nexpected+0.244
aded+0.236
time+0.232
Neg logit contributions
 wounds-0.345
pins-0.309
reating-0.305
line-0.283
bars-0.282
Token time
Token ID640
Feature activation+2.22e-01

Pos logit contributions
und+0.085
ued+0.067
used+0.066
nexpected+0.066
+0.066
Neg logit contributions
 wounds-0.123
 gloves-0.113
pins-0.110
 knee-0.109
 bike-0.109
Token,
Token ID11
Feature activation+1.01e-01

Pos logit contributions
und+0.472
aded+0.435
nexpected+0.417
ued+0.417
+0.402
Neg logit contributions
 bike-0.233
pins-0.233
 knee-0.232
 wounds-0.232
 back-0.225
Token there
Token ID612
Feature activation+2.26e-01

Pos logit contributions
und+0.218
+0.209
aded+0.191
ued+0.188
nexpected+0.188
Neg logit contributions
 back-0.116
 even-0.111
 mom-0.110
 wounds-0.101
 knee-0.101
Token got
Token ID1392
Feature activation+1.48e-02

Pos logit contributions
 wounds+0.028
reating+0.026
pins+0.026
 gloves+0.024
line+0.024
Neg logit contributions
und-0.029
-0.025
nexpected-0.025
aded-0.024
ued-0.023
Token lost
Token ID2626
Feature activation+3.42e-02

Pos logit contributions
und+0.031
+0.027
nexpected+0.026
aded+0.025
ued+0.025
Neg logit contributions
 wounds-0.022
reating-0.019
pins-0.019
 knee-0.017
 gloves-0.017
Token again
Token ID757
Feature activation-2.08e-02

Pos logit contributions
und+0.074
+0.062
nexpected+0.061
aded+0.059
ued+0.059
Neg logit contributions
 wounds-0.048
pins-0.042
reating-0.041
 decide-0.037
 gloves-0.037
Token.
Token ID13
Feature activation-2.74e-02

Pos logit contributions
 wounds+0.021
reating+0.016
pins+0.016
 knee+0.014
 gloves+0.014
Neg logit contributions
und-0.052
-0.048
nexpected-0.048
aded-0.047
ued-0.046
Token<|endoftext|>
Token ID50256
Feature activation+1.43e-01

Pos logit contributions
 wounds+0.044
reating+0.040
pins+0.040
 knee+0.038
 gloves+0.037
Neg logit contributions
und-0.049
nexpected-0.043
aded-0.043
-0.042
ued-0.041
TokenOnce
Token ID7454
Feature activation+2.82e-01

Pos logit contributions
und+0.279
nexpected+0.263
aded+0.260
ued+0.251
+0.243
Neg logit contributions
 wounds-0.188
pins-0.169
reating-0.169
line-0.156
 jacket-0.155
Token upon
Token ID2402
Feature activation+1.82e-01

Pos logit contributions
und+0.448
+0.389
nexpected+0.374
aded+0.369
ued+0.347
Neg logit contributions
 wounds-0.518
pins-0.492
reating-0.486
 knee-0.458
line-0.455
Token a
Token ID257
Feature activation+6.20e-02

Pos logit contributions
und+0.265
+0.222
nexpected+0.212
aded+0.207
time+0.205
Neg logit contributions
 wounds-0.370
pins-0.339
reating-0.326
line-0.314
bars-0.310
Token time
Token ID640
Feature activation+2.32e-01

Pos logit contributions
und+0.088
ued+0.071
+0.071
nexpected+0.070
uttered+0.070
Neg logit contributions
 wounds-0.122
 gloves-0.113
 bike-0.109
 knee-0.109
 jacket-0.106
Token,
Token ID11
Feature activation+9.73e-02

Pos logit contributions
und+0.486
aded+0.461
nexpected+0.448
ued+0.426
+0.422
Neg logit contributions
pins-0.259
 bike-0.246
 back-0.238
 wounds-0.235
line-0.233
Token there
Token ID612
Feature activation+2.27e-01

Pos logit contributions
und+0.176
+0.175
nexpected+0.164
aded+0.160
 contemplation+0.155
Neg logit contributions
 back-0.135
 wounds-0.134
pins-0.130
 knee-0.128
 bike-0.126
Token.
Token ID13
Feature activation-5.52e-02

Pos logit contributions
 wounds+0.066
reating+0.054
pins+0.053
 gloves+0.046
 knee+0.046
Neg logit contributions
und-0.161
-0.145
nexpected-0.141
aded-0.138
ued-0.137
Token The
Token ID383
Feature activation-2.41e-02

Pos logit contributions
 wounds+0.112
pins+0.098
reating+0.098
 gloves+0.094
 knee+0.093
Neg logit contributions
und-0.087
-0.075
nexpected-0.068
mon-0.065
 Luckily-0.064
Token end
Token ID886
Feature activation-2.68e-02

Pos logit contributions
 wounds+0.034
reating+0.033
pins+0.031
line+0.031
rings+0.030
Neg logit contributions
-0.044
day-0.042
und-0.042
time-0.041
nexpected-0.040
Token.
Token ID13
Feature activation-4.74e-02

Pos logit contributions
 wounds+0.034
pins+0.029
reating+0.028
line+0.026
 knee+0.026
Neg logit contributions
und-0.061
-0.054
nexpected-0.054
time-0.051
aded-0.050
Token<|endoftext|>
Token ID50256
Feature activation+1.49e-01

Pos logit contributions
 wounds+0.104
pins+0.091
reating+0.090
 gloves+0.088
 knee+0.087
Neg logit contributions
und-0.067
-0.059
nexpected-0.050
mon-0.049
day-0.049
TokenOnce
Token ID7454
Feature activation+2.82e-01

Pos logit contributions
und+0.268
+0.231
nexpected+0.228
aded+0.223
ued+0.219
Neg logit contributions
 wounds-0.252
reating-0.226
pins-0.223
 gloves-0.208
 knee-0.207
Token upon
Token ID2402
Feature activation+1.82e-01

Pos logit contributions
und+0.399
+0.330
nexpected+0.318
aded+0.309
ued+0.300
Neg logit contributions
 wounds-0.589
reating-0.545
pins-0.541
 gloves-0.509
 knee-0.508
Token a
Token ID257
Feature activation+5.92e-02

Pos logit contributions
und+0.290
+0.255
nexpected+0.238
time+0.231
aded+0.230
Neg logit contributions
 wounds-0.349
pins-0.311
reating-0.308
line-0.285
bars-0.285
Token time
Token ID640
Feature activation+2.19e-01

Pos logit contributions
und+0.080
+0.062
nexpected+0.062
ued+0.061
aded+0.061
Neg logit contributions
 wounds-0.121
 gloves-0.111
pins-0.109
 bike-0.108
 knee-0.108
Token,
Token ID11
Feature activation+9.74e-02

Pos logit contributions
und+0.469
aded+0.434
nexpected+0.413
ued+0.410
+0.396
Neg logit contributions
 bike-0.229
pins-0.229
 knee-0.225
 wounds-0.222
 back-0.215
Token there
Token ID612
Feature activation+2.25e-01

Pos logit contributions
und+0.216
+0.203
aded+0.192
nexpected+0.185
ued+0.183
Neg logit contributions
 back-0.105
 even-0.105
 mom-0.103
."-0.100
 bike-0.095
Token he
Token ID339
Feature activation+9.34e-02

Pos logit contributions
und+0.167
aded+0.145
nexpected+0.143
+0.143
ued+0.143
Neg logit contributions
 wounds-0.072
pins-0.063
reating-0.063
 gloves-0.058
 knee-0.058
Token had
Token ID550
Feature activation+1.03e-01

Pos logit contributions
und+0.159
+0.142
nexpected+0.132
aded+0.127
ued+0.124
Neg logit contributions
 wounds-0.169
pins-0.146
reating-0.145
 decide-0.144
 gloves-0.143
Token seen
Token ID1775
Feature activation+7.30e-02

Pos logit contributions
und+0.184
+0.164
nexpected+0.154
ued+0.150
aded+0.149
Neg logit contributions
 wounds-0.178
reating-0.157
pins-0.156
 gloves-0.151
 knee-0.147
Token.
Token ID13
Feature activation-2.91e-02

Pos logit contributions
und+0.141
nexpected+0.121
+0.121
aded+0.118
ued+0.114
Neg logit contributions
 wounds-0.115
pins-0.103
reating-0.100
 knee-0.092
 sax-0.092
Token<|endoftext|>
Token ID50256
Feature activation+1.40e-01

Pos logit contributions
 wounds+0.045
pins+0.041
reating+0.041
line+0.039
 knee+0.039
Neg logit contributions
und-0.053
nexpected-0.047
aded-0.046
-0.045
ued-0.044
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.250
nexpected+0.236
aded+0.233
ued+0.225
+0.218
Neg logit contributions
 wounds-0.198
pins-0.182
reating-0.180
line-0.172
 jacket-0.170
Token upon
Token ID2402
Feature activation+1.85e-01

Pos logit contributions
und+0.457
+0.402
nexpected+0.384
aded+0.380
ued+0.354
Neg logit contributions
 wounds-0.503
pins-0.482
reating-0.474
line-0.452
 bike-0.445
Token a
Token ID257
Feature activation+6.35e-02

Pos logit contributions
und+0.273
+0.228
nexpected+0.219
aded+0.214
time+0.210
Neg logit contributions
 wounds-0.371
pins-0.340
reating-0.331
line-0.315
bars-0.310
Token time
Token ID640
Feature activation+2.35e-01

Pos logit contributions
und+0.089
+0.072
ued+0.071
nexpected+0.071
uttered+0.070
Neg logit contributions
 wounds-0.128
 gloves-0.117
 bike-0.113
 knee-0.113
 jacket-0.110
Token,
Token ID11
Feature activation+9.99e-02

Pos logit contributions
und+0.504
aded+0.482
nexpected+0.466
ued+0.446
ited+0.443
Neg logit contributions
pins-0.239
 bike-0.228
 back-0.226
line-0.220
 knee-0.216
Token there
Token ID612
Feature activation+2.30e-01

Pos logit contributions
+0.190
und+0.184
nexpected+0.173
aded+0.168
 contemplation+0.165
Neg logit contributions
 back-0.134
 wounds-0.129
 knee-0.124
pins-0.123
 even-0.121
Token place
Token ID1295
Feature activation+8.06e-02

Pos logit contributions
und+0.244
+0.219
nexpected+0.210
aded+0.207
ued+0.204
Neg logit contributions
 wounds-0.152
reating-0.131
pins-0.126
 gloves-0.115
 decide-0.111
Token to
Token ID284
Feature activation+9.07e-04

Pos logit contributions
und+0.191
+0.173
nexpected+0.169
aded+0.165
ued+0.163
Neg logit contributions
 wounds-0.092
reating-0.078
pins-0.076
 gloves-0.068
 knee-0.068
Token explore
Token ID7301
Feature activation-1.64e-02

Pos logit contributions
und+0.002
+0.002
nexpected+0.002
aded+0.002
ued+0.002
Neg logit contributions
 wounds-0.001
reating-0.001
pins-0.001
 gloves-0.001
 knee-0.001
Token.
Token ID13
Feature activation-1.36e-02

Pos logit contributions
 wounds+0.023
pins+0.020
reating+0.020
 knee+0.018
 sax+0.017
Neg logit contributions
und-0.035
nexpected-0.030
-0.030
ued-0.029
aded-0.029
Token<|endoftext|>
Token ID50256
Feature activation+1.40e-01

Pos logit contributions
 wounds+0.022
reating+0.021
pins+0.020
 gloves+0.019
 knee+0.019
Neg logit contributions
und-0.024
nexpected-0.021
aded-0.020
-0.020
ued-0.019
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.260
nexpected+0.248
aded+0.244
ued+0.237
+0.225
Neg logit contributions
 wounds-0.187
reating-0.171
pins-0.170
 gloves-0.160
 jacket-0.159
Token upon
Token ID2402
Feature activation+1.86e-01

Pos logit contributions
und+0.434
+0.373
nexpected+0.360
aded+0.355
ued+0.335
Neg logit contributions
 wounds-0.536
pins-0.504
reating-0.500
 knee-0.472
line-0.470
Token a
Token ID257
Feature activation+6.37e-02

Pos logit contributions
und+0.279
+0.235
nexpected+0.225
aded+0.219
time+0.214
Neg logit contributions
 wounds-0.371
pins-0.339
reating-0.329
line-0.314
bars-0.308
Token time
Token ID640
Feature activation+2.35e-01

Pos logit contributions
und+0.088
+0.070
nexpected+0.069
ued+0.069
uttered+0.068
Neg logit contributions
 wounds-0.130
 gloves-0.120
 knee-0.115
 bike-0.115
 jacket-0.113
Token there
Token ID612
Feature activation+2.35e-01

Pos logit contributions
und+0.501
aded+0.481
nexpected+0.467
ued+0.442
ited+0.438
Neg logit contributions
pins-0.246
 bike-0.234
 back-0.229
line-0.225
 even-0.221
Token was
Token ID373
Feature activation+9.76e-02

Pos logit contributions
und+0.461
ued+0.440
aded+0.431
nexpected+0.431
+0.428
Neg logit contributions
 even-0.271
 mom-0.263
 wounds-0.261
 jacket-0.257
 knee-0.257
Token the
Token ID262
Feature activation-8.67e-02

Pos logit contributions
und+0.162
+0.143
nexpected+0.140
aded+0.137
ued+0.135
Neg logit contributions
 wounds-0.109
reating-0.095
pins-0.094
 gloves-0.086
 knee-0.086
Token cozy
Token ID37438
Feature activation-6.84e-03

Pos logit contributions
reating+0.151
 pilots+0.143
pins+0.141
 wounds+0.140
rings+0.138
Neg logit contributions
und-0.117
time-0.112
nexpected-0.106
ued-0.106
-0.104
Token rug
Token ID14477
Feature activation-2.16e-02

Pos logit contributions
 wounds+0.011
reating+0.009
pins+0.009
 decide+0.009
 pilots+0.008
Neg logit contributions
und-0.013
-0.012
nexpected-0.011
aded-0.011
ued-0.011
Token.
Token ID13
Feature activation-4.66e-02

Pos logit contributions
 wounds+0.019
pins+0.016
reating+0.016
 knee+0.014
line+0.013
Neg logit contributions
und-0.055
nexpected-0.051
-0.050
aded-0.050
ued-0.048
Token<|endoftext|>
Token ID50256
Feature activation+1.47e-01

Pos logit contributions
 wounds+0.073
pins+0.066
reating+0.065
 gloves+0.061
 sax+0.060
Neg logit contributions
und-0.088
aded-0.077
nexpected-0.077
-0.075
ued-0.074
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.265
+0.228
nexpected+0.226
aded+0.221
ued+0.216
Neg logit contributions
 wounds-0.248
reating-0.223
pins-0.221
 gloves-0.206
 knee-0.205
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.416
+0.348
nexpected+0.336
aded+0.330
ued+0.317
Neg logit contributions
 wounds-0.564
reating-0.525
pins-0.524
 gloves-0.488
 knee-0.487
Token a
Token ID257
Feature activation+5.91e-02

Pos logit contributions
und+0.298
+0.259
nexpected+0.245
aded+0.236
time+0.231
Neg logit contributions
 wounds-0.347
pins-0.312
reating-0.308
line-0.287
 sax-0.283
Token time
Token ID640
Feature activation+2.20e-01

Pos logit contributions
und+0.083
used+0.067
aded+0.067
nexpected+0.066
ued+0.066
Neg logit contributions
 wounds-0.115
 gloves-0.105
 bike-0.103
 knee-0.102
pins-0.101
Token,
Token ID11
Feature activation+9.95e-02

Pos logit contributions
und+0.475
aded+0.452
nexpected+0.428
ued+0.415
+0.411
Neg logit contributions
 bike-0.225
pins-0.219
 wounds-0.212
 back-0.211
 knee-0.211
Token there
Token ID612
Feature activation+2.27e-01

Pos logit contributions
und+0.224
+0.219
aded+0.209
nexpected+0.202
 contemplation+0.195
Neg logit contributions
 back-0.103
 mom-0.097
 even-0.096
.-0.085
 bike-0.084
Token a
Token ID257
Feature activation+5.54e-02

Pos logit contributions
und+0.137
+0.121
nexpected+0.120
aded+0.117
ued+0.114
Neg logit contributions
 wounds-0.116
reating-0.105
pins-0.104
line-0.095
 gloves-0.094
Token great
Token ID1049
Feature activation+1.30e-01

Pos logit contributions
und+0.111
nexpected+0.095
+0.095
aded+0.092
ued+0.091
Neg logit contributions
 wounds-0.085
reating-0.074
pins-0.072
 jacket-0.072
 gloves-0.071
Token day
Token ID1110
Feature activation+1.78e-01

Pos logit contributions
und+0.197
+0.166
nexpected+0.160
aded+0.155
ued+0.152
Neg logit contributions
 wounds-0.263
reating-0.239
pins-0.237
 gloves-0.223
 knee-0.222
Token!
Token ID0
Feature activation+2.20e-04

Pos logit contributions
und+0.414
+0.375
nexpected+0.374
aded+0.359
ued+0.356
Neg logit contributions
 wounds-0.202
reating-0.173
pins-0.172
 gloves-0.155
 knee-0.154
Token<|endoftext|>
Token ID50256
Feature activation+1.35e-01

Pos logit contributions
und+0.000
nexpected+0.000
+0.000
aded+0.000
ued+0.000
Neg logit contributions
 wounds-0.000
reating-0.000
pins-0.000
 knee-0.000
 gloves-0.000
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.233
nexpected+0.217
aded+0.210
ued+0.206
+0.201
Neg logit contributions
 wounds-0.209
reating-0.192
pins-0.192
line-0.180
 jacket-0.179
Token upon
Token ID2402
Feature activation+1.85e-01

Pos logit contributions
und+0.453
+0.395
nexpected+0.380
aded+0.373
ued+0.350
Neg logit contributions
 wounds-0.512
pins-0.488
reating-0.478
 knee-0.455
line-0.455
Token a
Token ID257
Feature activation+6.03e-02

Pos logit contributions
und+0.290
+0.243
nexpected+0.236
aded+0.232
 Larry+0.223
Neg logit contributions
 wounds-0.359
pins-0.326
reating-0.318
line-0.300
bars-0.295
Token time
Token ID640
Feature activation+2.34e-01

Pos logit contributions
und+0.076
+0.060
nexpected+0.059
ued+0.057
aded+0.055
Neg logit contributions
 wounds-0.133
 gloves-0.121
 knee-0.118
pins-0.118
reating-0.118
Token,
Token ID11
Feature activation+9.70e-02

Pos logit contributions
und+0.495
aded+0.469
nexpected+0.466
ued+0.438
ited+0.433
Neg logit contributions
pins-0.244
 back-0.229
 bike-0.226
line-0.226
 knee-0.222
Token there
Token ID612
Feature activation+2.28e-01

Pos logit contributions
und+0.168
+0.164
nexpected+0.157
aded+0.149
 contemplation+0.145
Neg logit contributions
 wounds-0.144
 back-0.140
pins-0.140
 knee-0.136
 bike-0.133
Token to
Token ID284
Feature activation+2.66e-02

Pos logit contributions
 wounds+0.015
reating+0.013
pins+0.013
 gloves+0.012
 knee+0.012
Neg logit contributions
und-0.019
-0.017
nexpected-0.016
aded-0.016
ued-0.016
Token play
Token ID711
Feature activation-6.06e-02

Pos logit contributions
und+0.057
+0.051
nexpected+0.048
aded+0.047
ued+0.047
Neg logit contributions
 wounds-0.037
reating-0.034
pins-0.032
line-0.029
 gloves-0.028
Token together
Token ID1978
Feature activation+1.18e-01

Pos logit contributions
 wounds+0.118
reating+0.112
pins+0.110
line+0.101
bars+0.101
Neg logit contributions
und-0.096
-0.081
nexpected-0.076
aded-0.074
ued-0.072
Token.
Token ID13
Feature activation-4.11e-02

Pos logit contributions
und+0.288
+0.259
nexpected+0.254
aded+0.249
ued+0.247
Neg logit contributions
 wounds-0.130
reating-0.109
pins-0.107
 gloves-0.095
 knee-0.094
Token<|endoftext|>
Token ID50256
Feature activation+1.48e-01

Pos logit contributions
 wounds+0.070
pins+0.063
reating+0.063
 gloves+0.058
 knee+0.057
Neg logit contributions
und-0.073
nexpected-0.063
aded-0.063
-0.063
ued-0.061
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.256
+0.220
nexpected+0.213
aded+0.207
ued+0.204
Neg logit contributions
 wounds-0.265
reating-0.238
pins-0.235
 gloves-0.221
 knee-0.220
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.397
+0.328
nexpected+0.317
aded+0.308
ued+0.299
Neg logit contributions
 wounds-0.587
reating-0.543
pins-0.540
 gloves-0.508
 knee-0.507
Token a
Token ID257
Feature activation+5.97e-02

Pos logit contributions
und+0.300
+0.262
nexpected+0.245
aded+0.239
time+0.233
Neg logit contributions
 wounds-0.347
pins-0.311
reating-0.308
line-0.284
 sax-0.283
Token time
Token ID640
Feature activation+2.22e-01

Pos logit contributions
und+0.080
ued+0.064
nexpected+0.063
aded+0.063
+0.063
Neg logit contributions
 wounds-0.121
 gloves-0.110
pins-0.108
 bike-0.108
 knee-0.106
Token,
Token ID11
Feature activation+1.00e-01

Pos logit contributions
und+0.475
aded+0.448
nexpected+0.427
ued+0.420
+0.405
Neg logit contributions
 bike-0.223
pins-0.222
 wounds-0.218
 knee-0.217
 back-0.213
Token there
Token ID612
Feature activation+2.28e-01

Pos logit contributions
und+0.219
+0.206
aded+0.197
nexpected+0.195
ued+0.184
Neg logit contributions
 even-0.114
 back-0.114
 mom-0.108
 bike-0.101
 dad-0.099
Token discover
Token ID7073
Feature activation-6.99e-03

Pos logit contributions
 wounds+0.072
reating+0.063
pins+0.062
 gloves+0.057
 knee+0.057
Neg logit contributions
und-0.109
-0.096
nexpected-0.095
aded-0.092
ued-0.092
Token new
Token ID649
Feature activation-3.44e-02

Pos logit contributions
 wounds+0.011
pins+0.011
 knee+0.010
line+0.009
reating+0.009
Neg logit contributions
und-0.013
nexpected-0.011
aded-0.011
 Larry-0.010
ued-0.010
Token things
Token ID1243
Feature activation-2.35e-02

Pos logit contributions
 wounds+0.057
pins+0.056
reating+0.050
line+0.050
 knee+0.050
Neg logit contributions
und-0.062
nexpected-0.051
aded-0.048
 Larry-0.047
ued-0.047
Token.
Token ID13
Feature activation-5.29e-02

Pos logit contributions
 wounds+0.033
reating+0.027
pins+0.026
 knee+0.026
 gloves+0.025
Neg logit contributions
und-0.051
-0.044
ued-0.043
nexpected-0.043
aded-0.041
Token<|endoftext|>
Token ID50256
Feature activation+1.38e-01

Pos logit contributions
 wounds+0.127
pins+0.112
reating+0.112
 knee+0.106
 gloves+0.105
Neg logit contributions
und-0.066
-0.056
 Luckily-0.046
day-0.045
nexpected-0.045
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.256
nexpected+0.222
+0.221
aded+0.218
ued+0.214
Neg logit contributions
 wounds-0.223
reating-0.200
pins-0.197
 gloves-0.185
 knee-0.183
Token upon
Token ID2402
Feature activation+1.79e-01

Pos logit contributions
und+0.390
+0.322
nexpected+0.309
aded+0.299
ued+0.290
Neg logit contributions
 wounds-0.594
reating-0.551
pins-0.548
 knee-0.518
 gloves-0.515
Token a
Token ID257
Feature activation+5.70e-02

Pos logit contributions
und+0.256
+0.216
nexpected+0.205
time+0.201
aded+0.197
Neg logit contributions
 wounds-0.367
pins-0.336
reating-0.324
line-0.311
bars-0.308
Token time
Token ID640
Feature activation+2.26e-01

Pos logit contributions
und+0.075
+0.059
ued+0.058
nexpected+0.058
aded+0.056
Neg logit contributions
 wounds-0.120
 gloves-0.110
 knee-0.107
pins-0.106
 bike-0.106
Token,
Token ID11
Feature activation+9.53e-02

Pos logit contributions
und+0.470
aded+0.443
nexpected+0.428
ued+0.414
+0.401
Neg logit contributions
pins-0.242
 bike-0.238
 knee-0.230
 wounds-0.229
 back-0.228
Token there
Token ID612
Feature activation+2.23e-01

Pos logit contributions
und+0.178
+0.174
nexpected+0.161
aded+0.158
 contemplation+0.152
Neg logit contributions
 wounds-0.130
 back-0.128
 bike-0.122
pins-0.122
 knee-0.121
Token explore
Token ID7301
Feature activation-1.48e-02

Pos logit contributions
 wounds+0.020
reating+0.017
pins+0.017
 gloves+0.015
 knee+0.015
Neg logit contributions
und-0.036
-0.032
nexpected-0.031
aded-0.031
ued-0.030
Token the
Token ID262
Feature activation-6.30e-02

Pos logit contributions
 wounds+0.022
pins+0.020
reating+0.019
 knee+0.018
 sax+0.017
Neg logit contributions
und-0.031
nexpected-0.026
-0.026
ued-0.025
aded-0.025
Token world
Token ID995
Feature activation+6.35e-02

Pos logit contributions
 wounds+0.110
pins+0.102
reating+0.098
 knee+0.091
line+0.091
Neg logit contributions
und-0.110
-0.092
nexpected-0.091
ued-0.088
aded-0.088
Token!
Token ID0
Feature activation-9.34e-03

Pos logit contributions
und+0.147
+0.133
nexpected+0.131
aded+0.128
ued+0.125
Neg logit contributions
 wounds-0.075
reating-0.065
pins-0.064
 gloves-0.057
 knee-0.056
Token<|endoftext|>
Token ID50256
Feature activation+1.35e-01

Pos logit contributions
 wounds+0.016
reating+0.015
pins+0.014
 knee+0.013
 gloves+0.013
Neg logit contributions
und-0.016
nexpected-0.014
-0.014
aded-0.014
ued-0.013
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.231
nexpected+0.218
aded+0.213
ued+0.206
+0.197
Neg logit contributions
 wounds-0.203
pins-0.189
reating-0.188
line-0.178
 jacket-0.177
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.433
+0.372
nexpected+0.359
aded+0.354
ued+0.332
Neg logit contributions
 wounds-0.535
pins-0.506
reating-0.499
line-0.472
 knee-0.471
Token a
Token ID257
Feature activation+6.32e-02

Pos logit contributions
und+0.271
+0.226
nexpected+0.217
aded+0.213
 Larry+0.208
Neg logit contributions
 wounds-0.370
pins-0.338
reating-0.328
line-0.314
bars-0.309
Token time
Token ID640
Feature activation+2.36e-01

Pos logit contributions
und+0.086
+0.068
nexpected+0.067
ued+0.067
aded+0.065
Neg logit contributions
 wounds-0.131
 gloves-0.120
 knee-0.117
 bike-0.116
pins-0.115
Token,
Token ID11
Feature activation+9.88e-02

Pos logit contributions
und+0.493
aded+0.471
nexpected+0.460
ued+0.433
ited+0.426
Neg logit contributions
pins-0.261
line-0.244
 bike-0.243
 back-0.241
 wounds-0.234
Token there
Token ID612
Feature activation+2.30e-01

Pos logit contributions
und+0.171
+0.168
nexpected+0.162
aded+0.155
 contemplation+0.150
Neg logit contributions
 wounds-0.146
 back-0.141
pins-0.139
 knee-0.137
 bike-0.133
Token in
Token ID287
Feature activation+2.62e-02

Pos logit contributions
und+0.230
nexpected+0.208
+0.207
aded+0.203
ued+0.203
Neg logit contributions
 wounds-0.120
reating-0.109
pins-0.105
 gloves-0.100
 bike-0.095
Token the
Token ID262
Feature activation-6.30e-02

Pos logit contributions
und+0.054
+0.048
nexpected+0.046
aded+0.045
ued+0.044
Neg logit contributions
 wounds-0.038
reating-0.034
pins-0.034
 knee-0.031
 gloves-0.030
Token field
Token ID2214
Feature activation+5.67e-02

Pos logit contributions
 wounds+0.107
pins+0.093
reating+0.091
 gloves+0.091
 jacket+0.089
Neg logit contributions
und-0.118
-0.103
nexpected-0.100
aded-0.098
ued-0.096
Token.
Token ID13
Feature activation-5.44e-02

Pos logit contributions
und+0.135
+0.122
nexpected+0.121
aded+0.118
ued+0.117
Neg logit contributions
 wounds-0.063
reating-0.053
pins-0.052
 knee-0.047
 gloves-0.047
Token<|endoftext|>
Token ID50256
Feature activation+1.46e-01

Pos logit contributions
 wounds+0.111
reating+0.098
pins+0.097
 gloves+0.093
 knee+0.093
Neg logit contributions
und-0.085
-0.073
nexpected-0.065
mon-0.064
 Luckily-0.063
TokenOnce
Token ID7454
Feature activation+2.81e-01

Pos logit contributions
und+0.257
+0.221
nexpected+0.216
aded+0.210
ued+0.206
Neg logit contributions
 wounds-0.257
reating-0.231
pins-0.228
 gloves-0.214
 knee-0.213
Token upon
Token ID2402
Feature activation+1.82e-01

Pos logit contributions
und+0.395
+0.326
nexpected+0.315
aded+0.304
ued+0.297
Neg logit contributions
 wounds-0.591
reating-0.544
pins-0.540
 gloves-0.510
 knee-0.509
Token a
Token ID257
Feature activation+5.80e-02

Pos logit contributions
und+0.293
+0.255
nexpected+0.240
aded+0.234
time+0.231
Neg logit contributions
 wounds-0.345
pins-0.310
reating-0.308
bars-0.284
line-0.283
Token time
Token ID640
Feature activation+2.19e-01

Pos logit contributions
und+0.078
+0.061
nexpected+0.061
ued+0.060
used+0.060
Neg logit contributions
 wounds-0.119
 gloves-0.108
pins-0.107
 bike-0.106
 knee-0.105
Token,
Token ID11
Feature activation+9.62e-02

Pos logit contributions
und+0.463
aded+0.424
nexpected+0.410
ued+0.400
+0.390
Neg logit contributions
 bike-0.234
pins-0.233
 wounds-0.228
 knee-0.226
 back-0.224
Token there
Token ID612
Feature activation+2.24e-01

Pos logit contributions
und+0.202
+0.188
nexpected+0.174
aded+0.174
live+0.164
Neg logit contributions
 back-0.121
 even-0.117
 wounds-0.111
 mom-0.111
 bike-0.110
Token get
Token ID651
Feature activation-1.08e-01

Pos logit contributions
 wounds+0.037
reating+0.032
pins+0.030
 gloves+0.029
bars+0.028
Neg logit contributions
und-0.045
-0.041
ued-0.038
date-0.036
aded-0.036
Token to
Token ID284
Feature activation-1.66e-02

Pos logit contributions
 wounds+0.155
reating+0.136
pins+0.134
 gloves+0.123
 knee+0.122
Neg logit contributions
und-0.227
-0.201
nexpected-0.196
aded-0.191
ued-0.189
Token them
Token ID606
Feature activation-7.68e-02

Pos logit contributions
 wounds+0.024
pins+0.022
reating+0.021
 gloves+0.019
bars+0.019
Neg logit contributions
und-0.034
-0.029
nexpected-0.028
aded-0.028
ued-0.027
Token.
Token ID13
Feature activation-5.27e-02

Pos logit contributions
 wounds+0.092
reating+0.076
pins+0.075
 gloves+0.067
 knee+0.066
Neg logit contributions
und-0.180
-0.160
nexpected-0.153
ued-0.151
aded-0.150
Token<|endoftext|>
Token ID50256
Feature activation+1.47e-01

Pos logit contributions
 wounds+0.096
reating+0.087
pins+0.086
 gloves+0.081
 knee+0.080
Neg logit contributions
und-0.090
-0.078
nexpected-0.075
aded-0.072
ued-0.071
TokenOnce
Token ID7454
Feature activation+2.80e-01

Pos logit contributions
und+0.278
nexpected+0.242
+0.239
aded+0.237
ued+0.232
Neg logit contributions
 wounds-0.231
reating-0.207
pins-0.203
 gloves-0.189
 knee-0.189
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.404
+0.335
nexpected+0.324
aded+0.315
ued+0.306
Neg logit contributions
 wounds-0.577
reating-0.536
pins-0.532
 knee-0.500
 gloves-0.500
Token a
Token ID257
Feature activation+5.92e-02

Pos logit contributions
und+0.306
+0.269
nexpected+0.253
aded+0.245
time+0.239
Neg logit contributions
 wounds-0.340
pins-0.305
reating-0.302
 sax-0.279
line-0.277
Token time
Token ID640
Feature activation+2.21e-01

Pos logit contributions
und+0.080
ued+0.063
+0.062
nexpected+0.062
used+0.062
Neg logit contributions
 wounds-0.120
 gloves-0.110
pins-0.108
 bike-0.107
 knee-0.106
Token,
Token ID11
Feature activation+9.85e-02

Pos logit contributions
und+0.487
aded+0.465
nexpected+0.437
ued+0.431
acked+0.418
Neg logit contributions
 bike-0.207
pins-0.199
 back-0.197
 knee-0.195
 even-0.191
Token there
Token ID612
Feature activation+2.26e-01

Pos logit contributions
und+0.217
+0.199
aded+0.193
nexpected+0.190
 contemplation+0.183
Neg logit contributions
 back-0.113
 even-0.110
 mom-0.103
."-0.098
 bike-0.096
Token came
Token ID1625
Feature activation-8.13e-03

Pos logit contributions
und+0.280
+0.249
nexpected+0.246
aded+0.241
ued+0.239
Neg logit contributions
 wounds-0.166
reating-0.144
pins-0.141
 knee-0.130
 gloves-0.130
Token to
Token ID284
Feature activation+2.08e-02

Pos logit contributions
 wounds+0.011
reating+0.009
pins+0.009
 gloves+0.009
 knee+0.008
Neg logit contributions
und-0.018
-0.016
nexpected-0.015
aded-0.015
ued-0.015
Token cartoons
Token ID30070
Feature activation-2.19e-02

Pos logit contributions
und+0.043
+0.038
nexpected+0.037
aded+0.036
ued+0.036
Neg logit contributions
 wounds-0.030
reating-0.027
pins-0.027
 gloves-0.024
 knee-0.024
Token!
Token ID0
Feature activation-5.86e-03

Pos logit contributions
 wounds+0.023
reating+0.021
pins+0.020
 knee+0.018
line+0.018
Neg logit contributions
und-0.053
-0.048
nexpected-0.048
aded-0.047
ued-0.045
Token<|endoftext|>
Token ID50256
Feature activation+1.40e-01

Pos logit contributions
 wounds+0.010
reating+0.009
pins+0.009
 knee+0.008
 gloves+0.008
Neg logit contributions
und-0.010
nexpected-0.009
-0.009
aded-0.009
ued-0.008
TokenOnce
Token ID7454
Feature activation+2.80e-01

Pos logit contributions
und+0.262
nexpected+0.252
aded+0.250
ued+0.240
+0.226
Neg logit contributions
 wounds-0.183
reating-0.167
pins-0.165
 bike-0.156
 jacket-0.155
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.441
+0.376
nexpected+0.367
aded+0.364
ued+0.338
Neg logit contributions
 wounds-0.525
pins-0.498
reating-0.492
line-0.465
 knee-0.463
Token a
Token ID257
Feature activation+6.27e-02

Pos logit contributions
und+0.270
+0.227
nexpected+0.215
aded+0.209
time+0.207
Neg logit contributions
 wounds-0.374
pins-0.340
reating-0.329
line-0.314
bars-0.309
Token time
Token ID640
Feature activation+2.35e-01

Pos logit contributions
und+0.085
+0.067
nexpected+0.067
ued+0.066
uttered+0.064
Neg logit contributions
 wounds-0.131
 gloves-0.120
 knee-0.117
 bike-0.116
 jacket-0.113
Token there
Token ID612
Feature activation+2.35e-01

Pos logit contributions
und+0.506
aded+0.492
nexpected+0.481
ited+0.454
ued+0.454
Neg logit contributions
pins-0.226
 bike-0.215
 back-0.214
line-0.208
 even-0.206
Token was
Token ID373
Feature activation+9.66e-02

Pos logit contributions
und+0.457
ued+0.445
nexpected+0.434
+0.431
aded+0.430
Neg logit contributions
 even-0.271
 mom-0.264
 wounds-0.258
 knee-0.258
 jacket-0.255
Token she
Token ID673
Feature activation+5.32e-02

Pos logit contributions
 wounds+0.109
reating+0.093
pins+0.091
 gloves+0.083
 knee+0.082
Neg logit contributions
und-0.200
-0.178
nexpected-0.175
aded-0.171
ued-0.169
Token loved
Token ID6151
Feature activation+1.56e-02

Pos logit contributions
und+0.091
+0.080
nexpected+0.076
aded+0.073
ued+0.071
Neg logit contributions
 wounds-0.096
pins-0.084
reating-0.084
 decide-0.081
 gloves-0.081
Token them
Token ID606
Feature activation-4.25e-02

Pos logit contributions
und+0.030
+0.026
nexpected+0.026
aded+0.025
ued+0.025
Neg logit contributions
 wounds-0.025
reating-0.022
pins-0.022
 gloves-0.020
 knee-0.020
Token!
Token ID0
Feature activation-1.77e-02

Pos logit contributions
 wounds+0.063
pins+0.053
reating+0.052
 gloves+0.047
line+0.047
Neg logit contributions
und-0.087
-0.078
ued-0.074
nexpected-0.073
aded-0.072
Token<|endoftext|>
Token ID50256
Feature activation+1.37e-01

Pos logit contributions
 wounds+0.034
reating+0.030
pins+0.030
 gloves+0.028
 knee+0.028
Neg logit contributions
und-0.029
-0.025
nexpected-0.023
aded-0.023
ued-0.023
TokenOnce
Token ID7454
Feature activation+2.80e-01

Pos logit contributions
und+0.251
nexpected+0.239
aded+0.236
ued+0.229
+0.219
Neg logit contributions
 wounds-0.188
reating-0.174
pins-0.172
 gloves-0.162
 jacket-0.161
Token there
Token ID612
Feature activation+2.43e-01

Pos logit contributions
und+0.435
+0.375
nexpected+0.361
aded+0.356
ued+0.335
Neg logit contributions
 wounds-0.528
pins-0.503
reating-0.498
line-0.467
 knee-0.465
Token was
Token ID373
Feature activation+1.07e-01

Pos logit contributions
und+0.432
aded+0.385
nexpected+0.381
+0.378
ued+0.372
Neg logit contributions
 wounds-0.363
pins-0.346
reating-0.344
 knee-0.343
line-0.340
Token a
Token ID257
Feature activation+5.97e-02

Pos logit contributions
und+0.233
+0.206
nexpected+0.206
ued+0.200
aded+0.196
Neg logit contributions
 wounds-0.136
reating-0.119
pins-0.119
 gloves-0.116
 knee-0.113
Token goose
Token ID37246
Feature activation+7.26e-02

Pos logit contributions
und+0.079
+0.072
mon+0.061
nexpected+0.061
aded+0.060
Neg logit contributions
 wounds-0.129
reating-0.123
pins-0.114
bars-0.109
 stations-0.107
Token who
Token ID508
Feature activation+8.68e-02

Pos logit contributions
und+0.152
+0.133
nexpected+0.129
aded+0.125
ued+0.125
Neg logit contributions
 wounds-0.108
reating-0.094
pins-0.091
 gloves-0.085
 knee-0.081
Token on
Token ID319
Feature activation+4.39e-02

Pos logit contributions
und+0.112
aded+0.090
+0.089
nexpected+0.089
ued+0.087
Neg logit contributions
 wounds-0.098
pins-0.093
reating-0.087
line-0.081
 decide-0.080
Token the
Token ID262
Feature activation-1.01e-01

Pos logit contributions
und+0.078
+0.069
nexpected+0.064
time+0.060
day+0.059
Neg logit contributions
 wounds-0.078
pins-0.070
reating-0.067
line-0.065
 knee-0.064
Token ride
Token ID6594
Feature activation-7.46e-02

Pos logit contributions
 wounds+0.142
reating+0.132
pins+0.131
rings+0.112
line+0.110
Neg logit contributions
und-0.201
nexpected-0.183
-0.183
ued-0.182
time-0.173
Token.
Token ID13
Feature activation-5.21e-02

Pos logit contributions
 wounds+0.097
pins+0.079
reating+0.079
 gloves+0.071
 knee+0.070
Neg logit contributions
und-0.171
-0.150
nexpected-0.144
ued-0.143
aded-0.142
Token<|endoftext|>
Token ID50256
Feature activation+1.49e-01

Pos logit contributions
 wounds+0.106
reating+0.097
pins+0.094
 gloves+0.090
 knee+0.089
Neg logit contributions
und-0.081
-0.069
nexpected-0.064
ued-0.061
aded-0.060
TokenOnce
Token ID7454
Feature activation+2.80e-01

Pos logit contributions
und+0.271
+0.234
nexpected+0.232
aded+0.227
ued+0.223
Neg logit contributions
 wounds-0.247
reating-0.221
pins-0.218
 gloves-0.203
 knee-0.203
Token upon
Token ID2402
Feature activation+1.85e-01

Pos logit contributions
und+0.394
+0.327
nexpected+0.315
aded+0.307
ued+0.298
Neg logit contributions
 wounds-0.584
reating-0.543
pins-0.540
 knee-0.507
 gloves-0.507
Token a
Token ID257
Feature activation+5.72e-02

Pos logit contributions
und+0.293
+0.257
nexpected+0.240
time+0.233
aded+0.232
Neg logit contributions
 wounds-0.356
pins-0.318
reating-0.315
 sax-0.292
bars-0.292
Token time
Token ID640
Feature activation+2.19e-01

Pos logit contributions
und+0.078
used+0.062
+0.062
nexpected+0.061
ued+0.061
Neg logit contributions
 wounds-0.114
 gloves-0.104
pins-0.103
 bike-0.102
 knee-0.101
Token,
Token ID11
Feature activation+9.78e-02

Pos logit contributions
und+0.460
aded+0.431
nexpected+0.411
ued+0.404
+0.393
Neg logit contributions
 bike-0.234
pins-0.234
 wounds-0.229
 knee-0.225
 back-0.220
Token there
Token ID612
Feature activation+2.25e-01

Pos logit contributions
und+0.205
+0.189
aded+0.184
nexpected+0.179
ued+0.173
Neg logit contributions
 back-0.122
 even-0.114
 bike-0.111
 mom-0.110
."-0.108
Token time
Token ID640
Feature activation+1.55e-01

Pos logit contributions
und+0.148
nexpected+0.116
+0.114
used+0.112
ued+0.112
Neg logit contributions
 wounds-0.206
 bike-0.191
 knee-0.188
 jacket-0.188
 gloves-0.188
Token playing
Token ID2712
Feature activation-2.21e-02

Pos logit contributions
und+0.310
+0.274
nexpected+0.272
aded+0.260
ued+0.260
Neg logit contributions
 wounds-0.233
reating-0.202
pins-0.200
 gloves-0.188
 knee-0.187
Token together
Token ID1978
Feature activation+1.30e-01

Pos logit contributions
 wounds+0.039
reating+0.033
 gloves+0.033
pins+0.033
 knee+0.032
Neg logit contributions
und-0.038
nexpected-0.034
-0.034
ued-0.033
aded-0.032
Token.
Token ID13
Feature activation-4.01e-02

Pos logit contributions
und+0.357
nexpected+0.331
+0.328
aded+0.324
ued+0.319
Neg logit contributions
 wounds-0.092
reating-0.071
pins-0.067
 gloves-0.060
 knee-0.057
Token<|endoftext|>
Token ID50256
Feature activation+1.49e-01

Pos logit contributions
 wounds+0.057
pins+0.054
reating+0.051
line+0.049
 knee+0.048
Neg logit contributions
und-0.076
nexpected-0.068
aded-0.067
ued-0.067
-0.065
TokenOnce
Token ID7454
Feature activation+2.79e-01

Pos logit contributions
und+0.276
nexpected+0.239
+0.238
aded+0.234
ued+0.229
Neg logit contributions
 wounds-0.241
reating-0.216
pins-0.213
 gloves-0.199
 knee-0.197
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.425
+0.357
nexpected+0.344
aded+0.337
ued+0.325
Neg logit contributions
 wounds-0.550
pins-0.513
reating-0.513
 knee-0.476
 gloves-0.475
Token a
Token ID257
Feature activation+6.04e-02

Pos logit contributions
und+0.310
+0.270
nexpected+0.257
aded+0.250
ued+0.242
Neg logit contributions
 wounds-0.337
pins-0.301
reating-0.299
line-0.274
 sax-0.274
Token time
Token ID640
Feature activation+2.20e-01

Pos logit contributions
und+0.087
used+0.071
ued+0.070
nexpected+0.070
+0.069
Neg logit contributions
 wounds-0.115
 gloves-0.105
 bike-0.102
 knee-0.101
 jacket-0.101
Token,
Token ID11
Feature activation+1.01e-01

Pos logit contributions
und+0.502
aded+0.479
nexpected+0.452
ued+0.445
+0.436
Neg logit contributions
 bike-0.189
 back-0.184
 mom-0.179
 knee-0.178
pins-0.175
Token there
Token ID612
Feature activation+2.27e-01

Pos logit contributions
und+0.231
+0.224
aded+0.210
nexpected+0.206
 contemplation+0.202
Neg logit contributions
 back-0.100
 even-0.097
 mom-0.095
.-0.089
."-0.084
Token was
Token ID373
Feature activation+5.16e-02

Pos logit contributions
und+0.026
+0.024
nexpected+0.023
aded+0.023
ued+0.022
Neg logit contributions
 wounds-0.015
reating-0.012
pins-0.012
 knee-0.012
 gloves-0.012
Token truly
Token ID4988
Feature activation+1.15e-01

Pos logit contributions
und+0.110
+0.100
nexpected+0.098
mon+0.089
uttered+0.089
Neg logit contributions
 wounds-0.073
 knee-0.060
 gloves-0.059
reating-0.059
pins-0.058
Token mighty
Token ID18680
Feature activation+7.40e-02

Pos logit contributions
und+0.259
+0.229
nexpected+0.224
aded+0.208
live+0.207
Neg logit contributions
 wounds-0.154
reating-0.125
 knee-0.125
 gloves-0.124
pins-0.124
Token!
Token ID0
Feature activation-8.94e-03

Pos logit contributions
und+0.157
+0.139
nexpected+0.134
aded+0.132
ued+0.130
Neg logit contributions
 wounds-0.102
pins-0.090
reating-0.089
line-0.078
 knee-0.078
Token<|endoftext|>
Token ID50256
Feature activation+1.37e-01

Pos logit contributions
 wounds+0.016
reating+0.014
pins+0.014
 knee+0.013
 gloves+0.013
Neg logit contributions
und-0.015
nexpected-0.013
-0.013
aded-0.013
ued-0.012
TokenOnce
Token ID7454
Feature activation+2.79e-01

Pos logit contributions
und+0.227
nexpected+0.219
aded+0.211
ued+0.204
+0.197
Neg logit contributions
 wounds-0.212
reating-0.197
pins-0.195
line-0.189
 jacket-0.184
Token upon
Token ID2402
Feature activation+1.84e-01

Pos logit contributions
und+0.453
+0.400
nexpected+0.384
aded+0.376
ued+0.353
Neg logit contributions
 wounds-0.502
pins-0.478
reating-0.472
 knee-0.448
line-0.447
Token a
Token ID257
Feature activation+6.09e-02

Pos logit contributions
und+0.269
+0.220
nexpected+0.215
aded+0.212
 Larry+0.207
Neg logit contributions
 wounds-0.372
pins-0.343
reating-0.329
line-0.316
bars-0.311
Token time
Token ID640
Feature activation+2.33e-01

Pos logit contributions
und+0.080
+0.064
nexpected+0.062
ued+0.061
aded+0.058
Neg logit contributions
 wounds-0.130
 gloves-0.119
 knee-0.116
 bike-0.114
reating-0.114
Token,
Token ID11
Feature activation+9.90e-02

Pos logit contributions
und+0.500
aded+0.480
nexpected+0.467
ited+0.448
ued+0.447
Neg logit contributions
pins-0.223
 back-0.216
 bike-0.215
 knee-0.208
 even-0.208
Token there
Token ID612
Feature activation+2.27e-01

Pos logit contributions
und+0.176
+0.172
nexpected+0.166
aded+0.157
 contemplation+0.155
Neg logit contributions
 wounds-0.139
 back-0.138
pins-0.133
 knee-0.133
 even-0.128
Token best
Token ID1266
Feature activation+9.14e-02

Pos logit contributions
 wounds+0.049
reating+0.044
pins+0.044
line+0.041
 knee+0.039
Neg logit contributions
und-0.041
-0.039
nexpected-0.034
aded-0.033
time-0.033
Token day
Token ID1110
Feature activation+1.67e-01

Pos logit contributions
und+0.190
nexpected+0.159
ued+0.159
+0.157
aded+0.156
Neg logit contributions
 wounds-0.135
reating-0.124
pins-0.115
 gloves-0.115
 knee-0.115
Token ever
Token ID1683
Feature activation+8.13e-02

Pos logit contributions
und+0.300
+0.252
nexpected+0.251
aded+0.243
ued+0.242
Neg logit contributions
 wounds-0.287
pins-0.262
reating-0.260
 knee-0.242
 gloves-0.242
Token!
Token ID0
Feature activation-1.83e-02

Pos logit contributions
und+0.153
+0.132
nexpected+0.129
aded+0.124
ued+0.124
Neg logit contributions
 wounds-0.138
reating-0.120
pins-0.119
 gloves-0.113
 knee-0.112
Token<|endoftext|>
Token ID50256
Feature activation+1.47e-01

Pos logit contributions
 wounds+0.040
reating+0.035
pins+0.035
 gloves+0.034
 knee+0.033
Neg logit contributions
und-0.027
-0.023
 Luckily-0.021
mon-0.020
nexpected-0.020
TokenOnce
Token ID7454
Feature activation+2.79e-01

Pos logit contributions
und+0.263
+0.226
nexpected+0.223
aded+0.218
ued+0.214
Neg logit contributions
 wounds-0.251
reating-0.225
pins-0.222
 gloves-0.208
 knee-0.207
Token upon
Token ID2402
Feature activation+1.83e-01

Pos logit contributions
und+0.399
+0.330
nexpected+0.320
aded+0.312
ued+0.300
Neg logit contributions
 wounds-0.576
reating-0.539
pins-0.537
 knee-0.501
 gloves-0.500
Token a
Token ID257
Feature activation+5.81e-02

Pos logit contributions
und+0.297
+0.262
nexpected+0.246
aded+0.239
time+0.234
Neg logit contributions
 wounds-0.346
pins-0.309
reating-0.303
line-0.282
 sax-0.281
Token time
Token ID640
Feature activation+2.19e-01

Pos logit contributions
und+0.079
nexpected+0.061
+0.061
ued+0.061
aded+0.061
Neg logit contributions
 wounds-0.118
 gloves-0.108
pins-0.106
 bike-0.106
 jacket-0.105
Token,
Token ID11
Feature activation+9.98e-02

Pos logit contributions
und+0.480
aded+0.453
nexpected+0.432
ued+0.422
+0.413
Neg logit contributions
pins-0.208
 bike-0.207
 knee-0.200
 back-0.196
."-0.191
Token there
Token ID612
Feature activation+2.27e-01

Pos logit contributions
und+0.220
+0.211
aded+0.197
nexpected+0.194
 contemplation+0.189
Neg logit contributions
 back-0.107
 even-0.103
 mom-0.100
."-0.099
.-0.090
Token not
Token ID407
Feature activation-6.37e-02

Pos logit contributions
 wounds+0.164
reating+0.143
pins+0.141
 gloves+0.130
 knee+0.129
Neg logit contributions
und-0.235
-0.207
nexpected-0.202
aded-0.198
ued-0.195
Token touch
Token ID3638
Feature activation-1.10e-01

Pos logit contributions
 wounds+0.085
reating+0.073
pins+0.072
 gloves+0.066
 knee+0.066
Neg logit contributions
und-0.141
-0.126
nexpected-0.123
aded-0.121
ued-0.119
Token anything
Token ID1997
Feature activation-8.46e-02

Pos logit contributions
 wounds+0.160
pins+0.144
reating+0.141
 decide+0.130
line+0.129
Neg logit contributions
und-0.226
nexpected-0.189
-0.189
aded-0.179
ued-0.178
Token without
Token ID1231
Feature activation-5.64e-02

Pos logit contributions
 wounds+0.106
reating+0.091
pins+0.089
 gloves+0.081
 knee+0.081
Neg logit contributions
und-0.192
-0.171
nexpected-0.167
ued-0.163
aded-0.163
Token her
Token ID607
Feature activation-1.30e-01

Pos logit contributions
 wounds+0.087
reating+0.077
pins+0.076
 knee+0.070
 gloves+0.070
Neg logit contributions
und-0.111
-0.097
nexpected-0.096
aded-0.093
ued-0.091
Token permission
Token ID7170
Feature activation-2.40e-01

Pos logit contributions
 wounds+0.158
reating+0.152
pins+0.146
line+0.129
 sax+0.121
Neg logit contributions
und-0.280
-0.254
nexpected-0.249
ued-0.240
aded-0.239
Token.
Token ID13
Feature activation-7.08e-02

Pos logit contributions
 wounds+0.339
reating+0.274
 sax+0.258
 gloves+0.256
pins+0.250
Neg logit contributions
und-0.503
-0.449
ued-0.436
time-0.418
mon-0.416
Token Lily
Token ID20037
Feature activation-1.95e-02

Pos logit contributions
 wounds+0.119
reating+0.107
pins+0.106
 knee+0.099
 gloves+0.098
Neg logit contributions
und-0.128
nexpected-0.110
-0.110
aded-0.109
ued-0.105
Token promised
Token ID8072
Feature activation-1.36e-01

Pos logit contributions
 wounds+0.033
 knee+0.028
 gloves+0.027
pins+0.027
 jacket+0.027
Neg logit contributions
und-0.037
-0.033
ued-0.031
nexpected-0.030
aded-0.030
Token to
Token ID284
Feature activation-7.05e-02

Pos logit contributions
 wounds+0.241
reating+0.223
 sax+0.213
pins+0.212
 stations+0.206
Neg logit contributions
und-0.216
-0.190
nexpected-0.182
 Luckily-0.173
time-0.169
Token follow
Token ID1061
Feature activation-7.20e-02

Pos logit contributions
 wounds+0.112
reating+0.099
pins+0.096
 sax+0.089
line+0.088
Neg logit contributions
und-0.136
-0.121
nexpected-0.116
aded-0.112
ued-0.108
Token never
Token ID1239
Feature activation-1.71e-02

Pos logit contributions
 wounds+0.139
reating+0.122
pins+0.120
 gloves+0.111
 knee+0.111
Neg logit contributions
und-0.190
-0.167
nexpected-0.163
aded-0.159
ued-0.157
Token go
Token ID467
Feature activation-1.11e-01

Pos logit contributions
 wounds+0.022
reating+0.019
pins+0.018
 gloves+0.017
 knee+0.017
Neg logit contributions
und-0.039
-0.035
nexpected-0.034
aded-0.033
ued-0.033
Token outside
Token ID2354
Feature activation-1.76e-02

Pos logit contributions
 wounds+0.174
reating+0.154
pins+0.153
 gloves+0.141
 knee+0.141
Neg logit contributions
und-0.218
-0.191
nexpected-0.185
aded-0.181
ued-0.179
Token without
Token ID1231
Feature activation-4.87e-02

Pos logit contributions
 wounds+0.025
pins+0.021
reating+0.021
 gloves+0.019
 jacket+0.019
Neg logit contributions
und-0.038
-0.033
nexpected-0.032
aded-0.031
ued-0.031
Token her
Token ID607
Feature activation-1.17e-01

Pos logit contributions
 wounds+0.076
reating+0.067
pins+0.066
 gloves+0.061
 knee+0.061
Neg logit contributions
und-0.096
-0.084
nexpected-0.083
aded-0.080
ued-0.079
Token permission
Token ID7170
Feature activation-2.38e-01

Pos logit contributions
 wounds+0.151
reating+0.148
pins+0.140
line+0.131
 sax+0.126
Neg logit contributions
und-0.232
-0.216
nexpected-0.205
aded-0.201
ued-0.200
Token.
Token ID13
Feature activation-6.65e-02

Pos logit contributions
 wounds+0.379
 sax+0.301
pins+0.291
reating+0.290
 gloves+0.289
Neg logit contributions
und-0.445
-0.409
time-0.396
ued-0.385
day-0.384
Token<|endoftext|>
Token ID50256
Feature activation+1.27e-01

Pos logit contributions
 wounds+0.134
reating+0.120
pins+0.119
 gloves+0.113
 knee+0.112
Neg logit contributions
und-0.105
-0.091
nexpected-0.082
mon-0.077
 Luckily-0.077
TokenOnce
Token ID7454
Feature activation+2.63e-01

Pos logit contributions
und+0.258
nexpected+0.239
aded+0.236
ued+0.227
+0.222
Neg logit contributions
 wounds-0.164
reating-0.145
pins-0.144
 gloves-0.130
line-0.130
Token upon
Token ID2402
Feature activation+1.64e-01

Pos logit contributions
und+0.397
+0.336
nexpected+0.325
aded+0.320
ued+0.303
Neg logit contributions
 wounds-0.515
pins-0.484
reating-0.482
 knee-0.449
 bike-0.447
Token a
Token ID257
Feature activation+4.14e-02

Pos logit contributions
und+0.263
+0.225
nexpected+0.217
aded+0.209
time+0.205
Neg logit contributions
 wounds-0.309
pins-0.282
reating-0.277
line-0.259
bars-0.255
Token me
Token ID502
Feature activation-8.67e-02

Pos logit contributions
 wounds+0.118
reating+0.108
pins+0.107
 knee+0.099
line+0.098
Neg logit contributions
und-0.109
nexpected-0.092
-0.089
aded-0.089
ued-0.087
Token help
Token ID1037
Feature activation-1.47e-01

Pos logit contributions
 wounds+0.121
reating+0.109
pins+0.103
 sax+0.093
 gloves+0.092
Neg logit contributions
und-0.183
-0.166
nexpected-0.158
aded-0.158
ued-0.154
Token you
Token ID345
Feature activation-6.82e-02

Pos logit contributions
 wounds+0.280
pins+0.271
reating+0.255
 dashboard+0.247
bars+0.247
Neg logit contributions
und-0.213
nexpected-0.176
aded-0.163
time-0.162
 Larry-0.162
Token clean
Token ID3424
Feature activation-9.01e-02

Pos logit contributions
 wounds+0.090
reating+0.079
pins+0.075
 sax+0.071
 gloves+0.068
Neg logit contributions
und-0.148
-0.131
aded-0.126
ued-0.125
nexpected-0.123
Token your
Token ID534
Feature activation-1.26e-01

Pos logit contributions
 wounds+0.160
pins+0.151
reating+0.147
 gloves+0.139
bars+0.138
Neg logit contributions
und-0.149
nexpected-0.120
-0.113
ued-0.106
 Larry-0.106
Token wounds
Token ID14129
Feature activation-2.37e-01

Pos logit contributions
reating+0.174
pins+0.163
line+0.144
bars+0.138
 wounds+0.137
Neg logit contributions
und-0.251
nexpected-0.229
aded-0.226
-0.221
 Luckily-0.210
Token and
Token ID290
Feature activation-1.19e-01

Pos logit contributions
 wounds+0.336
reating+0.288
pins+0.272
 sax+0.262
 gloves+0.261
Neg logit contributions
und-0.503
-0.445
ued-0.411
nexpected-0.403
time-0.394
Token give
Token ID1577
Feature activation-1.22e-01

Pos logit contributions
 wounds+0.155
reating+0.150
pins+0.142
 sax+0.128
line+0.126
Neg logit contributions
und-0.242
-0.216
nexpected-0.212
aded-0.206
ued-0.202
Token you
Token ID345
Feature activation-6.22e-02

Pos logit contributions
 wounds+0.192
pins+0.175
reating+0.173
 knee+0.161
line+0.160
Neg logit contributions
und-0.231
nexpected-0.201
-0.194
aded-0.192
ued-0.189
Token some
Token ID617
Feature activation-6.92e-02

Pos logit contributions
 wounds+0.103
reating+0.095
line+0.093
pins+0.093
 sax+0.089
Neg logit contributions
und-0.107
nexpected-0.095
-0.091
aded-0.090
ued-0.089
Token cookies
Token ID14746
Feature activation-7.29e-02

Pos logit contributions
 wounds+0.101
line+0.093
pins+0.093
reating+0.091
 sax+0.085
Neg logit contributions
und-0.130
-0.117
nexpected-0.116
aded-0.114
ued-0.111
Token never
Token ID1239
Feature activation-1.43e-02

Pos logit contributions
 wounds+0.007
reating+0.006
pins+0.006
 gloves+0.006
 knee+0.006
Neg logit contributions
und-0.010
-0.009
nexpected-0.009
aded-0.009
ued-0.009
Token sign
Token ID1051
Feature activation-8.26e-02

Pos logit contributions
 wounds+0.016
reating+0.014
pins+0.013
 decide+0.013
 knee+0.013
Neg logit contributions
und-0.034
-0.030
nexpected-0.030
aded-0.029
ued-0.029
Token things
Token ID1243
Feature activation-3.69e-02

Pos logit contributions
 wounds+0.126
pins+0.109
reating+0.109
 gloves+0.101
 knee+0.100
Neg logit contributions
und-0.166
-0.144
nexpected-0.142
aded-0.137
ued-0.137
Token without
Token ID1231
Feature activation-4.98e-02

Pos logit contributions
 wounds+0.053
reating+0.047
pins+0.046
 gloves+0.043
 knee+0.042
Neg logit contributions
und-0.077
-0.068
nexpected-0.066
aded-0.065
ued-0.064
Token my
Token ID616
Feature activation-9.36e-02

Pos logit contributions
 wounds+0.080
reating+0.070
pins+0.067
 gloves+0.065
 knee+0.064
Neg logit contributions
und-0.098
-0.089
nexpected-0.082
ued-0.081
aded-0.080
Token permission
Token ID7170
Feature activation-2.33e-01

Pos logit contributions
 wounds+0.130
 knee+0.103
 gloves+0.101
reating+0.100
pins+0.100
Neg logit contributions
und-0.213
-0.183
nexpected-0.179
aded-0.178
ued-0.176
Token!"
Token ID2474
Feature activation-6.09e-02

Pos logit contributions
 wounds+0.340
reating+0.279
pins+0.261
 sax+0.260
 gloves+0.258
Neg logit contributions
und-0.487
-0.438
ued-0.423
mon-0.409
nexpected-0.404
Token
Token ID198
Feature activation+6.26e-02

Pos logit contributions
 wounds+0.102
pins+0.090
reating+0.090
 knee+0.088
line+0.085
Neg logit contributions
und-0.107
nexpected-0.096
aded-0.095
ued-0.092
-0.091
Token
Token ID198
Feature activation+6.48e-02

Pos logit contributions
und+0.112
+0.096
nexpected+0.094
aded+0.091
ued+0.090
Neg logit contributions
 wounds-0.110
reating-0.099
pins-0.098
 gloves-0.092
 knee-0.091
TokenThe
Token ID464
Feature activation-1.29e-02

Pos logit contributions
und+0.139
+0.126
nexpected+0.122
aded+0.120
ued+0.117
Neg logit contributions
 wounds-0.090
reating-0.078
pins-0.075
 gloves-0.070
 knee-0.069
Token mother
Token ID2802
Feature activation-5.59e-02

Pos logit contributions
 wounds+0.024
reating+0.022
pins+0.021
 gloves+0.020
line+0.020
Neg logit contributions
und-0.022
-0.019
nexpected-0.018
aded-0.018
ued-0.018
Token not
Token ID407
Feature activation-3.57e-02

Pos logit contributions
 wounds+0.123
reating+0.105
pins+0.103
 gloves+0.095
 knee+0.095
Neg logit contributions
und-0.204
-0.182
nexpected-0.178
aded-0.174
ued-0.172
Token do
Token ID466
Feature activation+8.44e-03

Pos logit contributions
 wounds+0.040
reating+0.034
pins+0.032
 decide+0.031
 knee+0.031
Neg logit contributions
und-0.084
-0.078
nexpected-0.077
aded-0.075
ued-0.074
Token things
Token ID1243
Feature activation-4.99e-02

Pos logit contributions
und+0.017
+0.015
nexpected+0.015
aded+0.014
ued+0.014
Neg logit contributions
 wounds-0.012
reating-0.011
pins-0.011
 gloves-0.010
 knee-0.010
Token without
Token ID1231
Feature activation-4.84e-02

Pos logit contributions
 wounds+0.081
reating+0.072
pins+0.071
 gloves+0.066
 knee+0.066
Neg logit contributions
und-0.095
-0.083
nexpected-0.081
aded-0.079
ued-0.078
Token their
Token ID511
Feature activation-6.08e-02

Pos logit contributions
 wounds+0.071
reating+0.062
pins+0.060
 gloves+0.056
 knee+0.056
Neg logit contributions
und-0.101
-0.090
nexpected-0.087
aded-0.085
ued-0.084
Token permission
Token ID7170
Feature activation-2.30e-01

Pos logit contributions
 wounds+0.079
reating+0.067
pins+0.066
 gloves+0.060
 knee+0.060
Neg logit contributions
und-0.137
-0.122
nexpected-0.120
aded-0.117
ued-0.116
Token.
Token ID13
Feature activation-6.94e-02

Pos logit contributions
 wounds+0.286
reating+0.230
pins+0.226
 gloves+0.214
 sax+0.208
Neg logit contributions
und-0.528
-0.474
ued-0.452
nexpected-0.449
aded-0.440
Token Even
Token ID3412
Feature activation-1.38e-02

Pos logit contributions
 wounds+0.117
reating+0.106
pins+0.104
 knee+0.099
line+0.095
Neg logit contributions
und-0.121
nexpected-0.107
aded-0.106
ued-0.103
-0.103
Token though
Token ID996
Feature activation+8.24e-02

Pos logit contributions
 wounds+0.018
reating+0.015
pins+0.015
 gloves+0.014
 knee+0.013
Neg logit contributions
und-0.031
-0.028
nexpected-0.027
aded-0.027
ued-0.027
Token it
Token ID340
Feature activation+1.19e-02

Pos logit contributions
und+0.182
+0.163
nexpected+0.158
ued+0.157
aded+0.156
Neg logit contributions
 wounds-0.108
reating-0.092
pins-0.090
 gloves-0.086
 knee-0.083
Token may
Token ID743
Feature activation-2.51e-02

Pos logit contributions
und+0.030
date+0.027
+0.027
aded+0.027
imm+0.026
Neg logit contributions
 wounds-0.010
 back-0.008
 even-0.007
 knee-0.007
."-0.006
Token.
Token ID13
Feature activation-1.01e-01

Pos logit contributions
 wounds+0.058
pins+0.050
reating+0.048
 knee+0.044
 gloves+0.044
Neg logit contributions
und-0.097
-0.082
nexpected-0.082
ued-0.079
aded-0.079
Token Let
Token ID3914
Feature activation-5.88e-02

Pos logit contributions
 wounds+0.183
reating+0.164
pins+0.161
 gloves+0.151
 knee+0.150
Neg logit contributions
und-0.177
-0.153
nexpected-0.146
aded-0.140
ued-0.140
Token me
Token ID502
Feature activation-9.17e-02

Pos logit contributions
 wounds+0.106
reating+0.095
pins+0.095
 knee+0.087
 gloves+0.086
Neg logit contributions
und-0.104
-0.087
nexpected-0.087
aded-0.083
ued-0.083
Token clean
Token ID3424
Feature activation-7.81e-02

Pos logit contributions
 wounds+0.123
reating+0.107
pins+0.104
 gloves+0.094
 knee+0.093
Neg logit contributions
und-0.200
-0.180
nexpected-0.174
aded-0.171
ued-0.169
Token your
Token ID534
Feature activation-1.16e-01

Pos logit contributions
 wounds+0.134
pins+0.125
reating+0.120
 decide+0.113
 gloves+0.113
Neg logit contributions
und-0.137
nexpected-0.110
-0.107
 Larry-0.101
ued-0.101
Token wounds
Token ID14129
Feature activation-2.28e-01

Pos logit contributions
reating+0.145
pins+0.145
line+0.130
 wounds+0.120
bars+0.115
Neg logit contributions
und-0.243
-0.216
nexpected-0.216
aded-0.210
ued-0.203
Token and
Token ID290
Feature activation-1.10e-01

Pos logit contributions
 wounds+0.318
reating+0.269
pins+0.257
 gloves+0.246
 sax+0.241
Neg logit contributions
und-0.490
-0.436
ued-0.409
nexpected-0.402
aded-0.389
Token put
Token ID1234
Feature activation-1.22e-01

Pos logit contributions
 wounds+0.146
reating+0.138
pins+0.132
 sax+0.119
line+0.118
Neg logit contributions
und-0.225
-0.202
nexpected-0.195
ued-0.189
aded-0.189
Token some
Token ID617
Feature activation-5.38e-02

Pos logit contributions
 wounds+0.181
reating+0.159
pins+0.157
 knee+0.144
 gloves+0.143
Neg logit contributions
und-0.251
-0.220
nexpected-0.215
aded-0.209
ued-0.207
Token band
Token ID4097
Feature activation-1.36e-01

Pos logit contributions
 wounds+0.080
pins+0.075
line+0.073
reating+0.072
 sax+0.068
Neg logit contributions
und-0.100
-0.087
nexpected-0.087
aded-0.086
ued-0.084
Token-
Token ID12
Feature activation-1.06e-01

Pos logit contributions
 wounds+0.268
reating+0.248
pins+0.237
 decide+0.230
 sax+0.225
Neg logit contributions
und-0.209
-0.168
ued-0.164
aded-0.164
nexpected-0.161
Token't
Token ID470
Feature activation-2.88e-02

Pos logit contributions
 wounds+0.049
reating+0.043
pins+0.043
 gloves+0.040
 knee+0.039
Neg logit contributions
und-0.067
-0.058
nexpected-0.057
aded-0.056
ued-0.055
Token explore
Token ID7301
Feature activation-5.77e-02

Pos logit contributions
 wounds+0.038
reating+0.032
pins+0.032
 gloves+0.029
 knee+0.029
Neg logit contributions
und-0.064
-0.057
nexpected-0.056
aded-0.055
ued-0.055
Token it
Token ID340
Feature activation-1.09e-02

Pos logit contributions
 wounds+0.089
pins+0.085
reating+0.078
 knee+0.077
 hem+0.074
Neg logit contributions
und-0.113
nexpected-0.093
 Larry-0.090
-0.088
ued-0.088
Token without
Token ID1231
Feature activation-4.09e-02

Pos logit contributions
 wounds+0.010
reating+0.009
pins+0.008
 gloves+0.007
 decide+0.007
Neg logit contributions
und-0.028
-0.026
nexpected-0.025
aded-0.025
ued-0.024
Token my
Token ID616
Feature activation-8.81e-02

Pos logit contributions
 wounds+0.067
reating+0.058
pins+0.056
 gloves+0.054
 knee+0.053
Neg logit contributions
und-0.079
-0.072
nexpected-0.067
aded-0.066
ued-0.066
Token permission
Token ID7170
Feature activation-2.28e-01

Pos logit contributions
 wounds+0.127
reating+0.120
pins+0.117
line+0.105
 gloves+0.100
Neg logit contributions
und-0.174
-0.155
nexpected-0.150
ued-0.147
aded-0.147
Token."
Token ID526
Feature activation-7.21e-02

Pos logit contributions
 wounds+0.299
reating+0.225
pins+0.219
 sax+0.218
 gloves+0.216
Neg logit contributions
und-0.502
ued-0.439
-0.436
mon-0.419
used-0.419
Token
Token ID198
Feature activation+6.03e-02

Pos logit contributions
 wounds+0.119
pins+0.104
reating+0.104
 knee+0.099
 gloves+0.097
Neg logit contributions
und-0.134
nexpected-0.116
-0.114
aded-0.114
ued-0.110
Token
Token ID198
Feature activation+6.28e-02

Pos logit contributions
und+0.103
+0.088
nexpected+0.085
aded+0.082
ued+0.081
Neg logit contributions
 wounds-0.111
reating-0.101
pins-0.100
 gloves-0.094
 knee-0.093
TokenThe
Token ID464
Feature activation-1.78e-02

Pos logit contributions
und+0.128
+0.112
nexpected+0.110
aded+0.107
ued+0.106
Neg logit contributions
 wounds-0.095
reating-0.084
pins-0.082
 gloves-0.076
 knee-0.076
Token man
Token ID582
Feature activation+5.31e-03

Pos logit contributions
 wounds+0.026
pins+0.023
reating+0.023
 gloves+0.021
 knee+0.020
Neg logit contributions
und-0.037
-0.032
nexpected-0.032
aded-0.031
ued-0.031
Token without
Token ID1231
Feature activation-4.70e-02

Pos logit contributions
 wounds+0.113
reating+0.098
pins+0.096
 knee+0.085
 gloves+0.085
Neg logit contributions
und-0.244
-0.218
nexpected-0.216
aded-0.212
ued-0.208
Token asking
Token ID4737
Feature activation-6.82e-02

Pos logit contributions
 wounds+0.071
reating+0.062
pins+0.060
 gloves+0.057
 knee+0.055
Neg logit contributions
und-0.097
-0.087
nexpected-0.082
ued-0.081
aded-0.080
Token his
Token ID465
Feature activation-9.48e-02

Pos logit contributions
 wounds+0.090
reating+0.076
pins+0.076
 knee+0.070
 gloves+0.069
Neg logit contributions
und-0.148
nexpected-0.130
-0.129
aded-0.125
ued-0.125
Token dad
Token ID9955
Feature activation-1.44e-01

Pos logit contributions
 wounds+0.093
reating+0.076
pins+0.074
 gloves+0.065
 knee+0.065
Neg logit contributions
und-0.243
-0.219
nexpected-0.215
aded-0.211
ued-0.209
Token's
Token ID338
Feature activation-1.01e-01

Pos logit contributions
 wounds+0.180
reating+0.158
pins+0.153
 gloves+0.138
 knee+0.137
Neg logit contributions
und-0.327
-0.294
nexpected-0.288
aded-0.284
ued-0.277
Token permission
Token ID7170
Feature activation-2.27e-01

Pos logit contributions
 wounds+0.126
pins+0.112
reating+0.112
 gloves+0.099
line+0.098
Neg logit contributions
und-0.218
nexpected-0.194
-0.194
aded-0.190
ued-0.188
Token.
Token ID13
Feature activation-6.58e-02

Pos logit contributions
 wounds+0.276
reating+0.218
pins+0.216
 gloves+0.203
 knee+0.196
Neg logit contributions
und-0.528
-0.472
nexpected-0.456
ued-0.452
aded-0.445
Token He
Token ID679
Feature activation+4.59e-02

Pos logit contributions
 wounds+0.094
reating+0.087
pins+0.083
 decide+0.077
 knee+0.077
Neg logit contributions
und-0.128
aded-0.116
nexpected-0.115
-0.112
ued-0.110
Token knew
Token ID2993
Feature activation-4.00e-02

Pos logit contributions
und+0.088
+0.080
aded+0.074
nexpected+0.073
ued+0.072
Neg logit contributions
 wounds-0.074
 decide-0.063
pins-0.060
 gloves-0.060
 even-0.060
Token it
Token ID340
Feature activation-2.10e-02

Pos logit contributions
 wounds+0.064
reating+0.057
pins+0.055
 gloves+0.052
 knee+0.051
Neg logit contributions
und-0.078
-0.069
nexpected-0.066
aded-0.065
ued-0.064
Token was
Token ID373
Feature activation-3.65e-03

Pos logit contributions
 wounds+0.030
 even+0.029
 back+0.026
 gloves+0.026
pins+0.026
Neg logit contributions
und-0.042
-0.039
aded-0.037
nexpected-0.034
ued-0.034
Token away
Token ID1497
Feature activation+6.12e-02

Pos logit contributions
 wounds+0.071
reating+0.063
pins+0.062
 gloves+0.058
 knee+0.057
Neg logit contributions
und-0.089
-0.078
nexpected-0.076
aded-0.074
ued-0.073
Token from
Token ID422
Feature activation-1.21e-02

Pos logit contributions
und+0.134
+0.119
nexpected+0.117
aded+0.114
ued+0.113
Neg logit contributions
 wounds-0.081
reating-0.070
pins-0.069
 gloves-0.063
 knee-0.063
Token someone
Token ID2130
Feature activation-1.16e-01

Pos logit contributions
 wounds+0.016
reating+0.014
pins+0.013
 decide+0.012
 knee+0.012
Neg logit contributions
und-0.027
-0.025
nexpected-0.023
ued-0.023
aded-0.023
Token without
Token ID1231
Feature activation-4.79e-02

Pos logit contributions
 wounds+0.166
reating+0.138
pins+0.136
 gloves+0.131
 knee+0.129
Neg logit contributions
und-0.242
-0.223
nexpected-0.215
aded-0.213
ued-0.203
Token their
Token ID511
Feature activation-5.40e-02

Pos logit contributions
 wounds+0.071
reating+0.062
pins+0.058
 gloves+0.056
 knee+0.054
Neg logit contributions
und-0.100
-0.093
nexpected-0.086
ued-0.084
aded-0.084
Token permission
Token ID7170
Feature activation-2.26e-01

Pos logit contributions
 wounds+0.073
reating+0.061
pins+0.060
 gloves+0.056
 knee+0.056
Neg logit contributions
und-0.121
-0.108
nexpected-0.104
aded-0.103
ued-0.101
Token.
Token ID13
Feature activation-6.94e-02

Pos logit contributions
 wounds+0.274
reating+0.227
pins+0.221
 gloves+0.205
 knee+0.198
Neg logit contributions
und-0.527
-0.471
nexpected-0.453
ued-0.448
aded-0.443
Token St
Token ID520
Feature activation-4.88e-02

Pos logit contributions
 wounds+0.108
reating+0.096
pins+0.095
line+0.088
 knee+0.088
Neg logit contributions
und-0.128
nexpected-0.116
aded-0.115
ued-0.111
-0.108
Tokenealing
Token ID26919
Feature activation+3.89e-02

Pos logit contributions
 wounds+0.094
pins+0.085
reating+0.085
 knee+0.080
 gloves+0.079
Neg logit contributions
und-0.078
-0.066
nexpected-0.063
aded-0.061
ued-0.061
Token something
Token ID1223
Feature activation+7.87e-04

Pos logit contributions
und+0.078
+0.073
aded+0.067
nexpected+0.067
ued+0.063
Neg logit contributions
 wounds-0.059
reating-0.048
pins-0.047
 knee-0.046
 sax-0.046
Token without
Token ID1231
Feature activation-2.28e-02

Pos logit contributions
und+0.002
+0.001
nexpected+0.001
aded+0.001
live+0.001
Neg logit contributions
 wounds-0.001
pins-0.001
reating-0.001
 jacket-0.001
 gloves-0.001
Token can
Token ID460
Feature activation-5.18e-02

Pos logit contributions
 wounds+0.088
 bike+0.073
 gloves+0.072
reating+0.072
 decide+0.072
Neg logit contributions
und-0.145
-0.136
nexpected-0.128
aded-0.128
ued-0.124
Token come
Token ID1282
Feature activation-1.07e-01

Pos logit contributions
 wounds+0.077
reating+0.068
pins+0.065
 gloves+0.060
 knee+0.060
Neg logit contributions
und-0.106
-0.095
nexpected-0.092
ued-0.090
aded-0.089
Token in
Token ID287
Feature activation+4.73e-03

Pos logit contributions
 wounds+0.183
reating+0.163
pins+0.149
 sax+0.149
 gloves+0.149
Neg logit contributions
und-0.195
-0.167
aded-0.150
used-0.150
ued-0.149
Token without
Token ID1231
Feature activation-4.46e-02

Pos logit contributions
und+0.012
+0.010
nexpected+0.010
aded+0.010
ued+0.010
Neg logit contributions
 wounds-0.005
reating-0.004
pins-0.004
 knee-0.004
 gloves-0.004
Token our
Token ID674
Feature activation-5.43e-02

Pos logit contributions
 wounds+0.068
reating+0.060
pins+0.059
 gloves+0.055
 knee+0.055
Neg logit contributions
und-0.089
-0.078
nexpected-0.077
aded-0.075
ued-0.073
Token permission
Token ID7170
Feature activation-2.26e-01

Pos logit contributions
 wounds+0.080
pins+0.068
reating+0.068
 gloves+0.064
 knee+0.063
Neg logit contributions
und-0.113
-0.099
nexpected-0.097
aded-0.095
ued-0.094
Token."
Token ID526
Feature activation-6.43e-02

Pos logit contributions
 wounds+0.302
reating+0.234
 sax+0.231
 gloves+0.216
pins+0.209
Neg logit contributions
und-0.498
-0.440
ued-0.429
mon-0.414
time-0.411
Token
Token ID198
Feature activation+6.54e-02

Pos logit contributions
 wounds+0.109
reating+0.098
pins+0.096
 gloves+0.090
 knee+0.089
Neg logit contributions
und-0.119
-0.103
nexpected-0.100
aded-0.097
ued-0.096
Token
Token ID198
Feature activation+6.74e-02

Pos logit contributions
und+0.095
+0.078
nexpected+0.074
aded+0.072
mon+0.071
Neg logit contributions
 wounds-0.135
reating-0.129
pins-0.123
 gloves-0.118
 knee-0.118
Token"
Token ID1
Feature activation-8.38e-02

Pos logit contributions
und+0.131
+0.114
nexpected+0.111
aded+0.108
ued+0.107
Neg logit contributions
 wounds-0.109
reating-0.097
pins-0.095
 gloves-0.089
 knee-0.088
TokenYes
Token ID5297
Feature activation+1.87e-03

Pos logit contributions
 wounds+0.125
pins+0.110
reating+0.110
 gloves+0.102
 sax+0.100
Neg logit contributions
und-0.172
-0.148
nexpected-0.146
aded-0.143
ued-0.142
Token always
Token ID1464
Feature activation+1.39e-02

Pos logit contributions
 wounds+0.092
reating+0.078
pins+0.075
 gloves+0.071
 decide+0.071
Neg logit contributions
und-0.145
-0.130
nexpected-0.121
aded-0.118
ued-0.116
Token better
Token ID1365
Feature activation-3.42e-02

Pos logit contributions
und+0.035
+0.033
nexpected+0.030
aded+0.030
ued+0.029
Neg logit contributions
 wounds-0.015
reating-0.012
 gloves-0.012
pins-0.011
 decide-0.011
Token to
Token ID284
Feature activation-3.93e-02

Pos logit contributions
 wounds+0.047
pins+0.039
reating+0.038
 gloves+0.036
 knee+0.036
Neg logit contributions
und-0.076
-0.069
nexpected-0.066
aded-0.065
ued-0.064
Token ask
Token ID1265
Feature activation-1.42e-01

Pos logit contributions
 wounds+0.046
reating+0.039
pins+0.038
 knee+0.035
 gloves+0.035
Neg logit contributions
und-0.092
-0.083
nexpected-0.081
aded-0.079
ued-0.078
Token for
Token ID329
Feature activation-4.30e-02

Pos logit contributions
 wounds+0.184
reating+0.159
pins+0.157
 gloves+0.143
 knee+0.142
Neg logit contributions
und-0.314
-0.279
nexpected-0.275
aded-0.268
ued-0.265
Token permission
Token ID7170
Feature activation-2.26e-01

Pos logit contributions
 wounds+0.060
pins+0.056
reating+0.055
line+0.050
 sax+0.049
Neg logit contributions
und-0.086
nexpected-0.077
-0.076
aded-0.075
ued-0.073
Token than
Token ID621
Feature activation-3.90e-02

Pos logit contributions
 wounds+0.350
pins+0.295
reating+0.294
 sax+0.279
 gloves+0.275
Neg logit contributions
und-0.440
-0.384
nexpected-0.371
ued-0.365
aded-0.355
Token to
Token ID284
Feature activation-6.39e-02

Pos logit contributions
 wounds+0.057
reating+0.050
pins+0.049
 gloves+0.045
 knee+0.045
Neg logit contributions
und-0.080
-0.071
nexpected-0.069
aded-0.068
ued-0.067
Token take
Token ID1011
Feature activation-1.39e-01

Pos logit contributions
 wounds+0.092
pins+0.081
reating+0.080
 knee+0.072
 gloves+0.072
Neg logit contributions
und-0.131
-0.114
nexpected-0.111
ued-0.108
aded-0.107
Token something
Token ID1223
Feature activation-5.96e-02

Pos logit contributions
 wounds+0.192
reating+0.158
pins+0.154
 gloves+0.150
 knee+0.143
Neg logit contributions
und-0.307
-0.281
nexpected-0.266
aded-0.260
ued-0.256
Token without
Token ID1231
Feature activation-6.10e-02

Pos logit contributions
 wounds+0.093
reating+0.085
pins+0.083
 knee+0.077
 gloves+0.076
Neg logit contributions
und-0.111
-0.095
nexpected-0.094
ued-0.093
aded-0.092
Token foolish
Token ID19538
Feature activation+4.31e-02

Pos logit contributions
und+0.176
+0.161
nexpected+0.153
aded+0.149
ued+0.149
Neg logit contributions
 wounds-0.131
pins-0.111
reating-0.111
 knee-0.105
 aer-0.104
Token things
Token ID1243
Feature activation-2.88e-02

Pos logit contributions
und+0.092
+0.079
nexpected+0.078
aded+0.077
ued+0.076
Neg logit contributions
 wounds-0.060
pins-0.053
reating-0.052
 gloves-0.047
 sax-0.046
Token and
Token ID290
Feature activation-7.02e-02

Pos logit contributions
 wounds+0.040
reating+0.035
pins+0.034
 gloves+0.032
 knee+0.031
Neg logit contributions
und-0.061
-0.054
nexpected-0.053
aded-0.052
ued-0.051
Token always
Token ID1464
Feature activation+1.35e-02

Pos logit contributions
 wounds+0.109
reating+0.094
pins+0.093
 gloves+0.087
 knee+0.087
Neg logit contributions
und-0.140
-0.125
nexpected-0.121
aded-0.118
ued-0.115
Token ask
Token ID1265
Feature activation-1.38e-01

Pos logit contributions
und+0.027
+0.026
nexpected+0.024
aded+0.023
ued+0.023
Neg logit contributions
 wounds-0.020
 decide-0.017
reating-0.017
 gloves-0.016
 jacket-0.016
Token permission
Token ID7170
Feature activation-2.25e-01

Pos logit contributions
 wounds+0.191
pins+0.169
reating+0.165
 knee+0.153
 jacket+0.152
Neg logit contributions
und-0.289
-0.251
nexpected-0.250
aded-0.242
ued-0.239
Token before
Token ID878
Feature activation-6.31e-02

Pos logit contributions
 wounds+0.341
pins+0.286
reating+0.275
 sax+0.271
 gloves+0.267
Neg logit contributions
und-0.446
-0.383
nexpected-0.375
ued-0.372
mon-0.360
Token entering
Token ID8218
Feature activation-2.56e-02

Pos logit contributions
 wounds+0.096
pins+0.084
reating+0.083
 gloves+0.078
 knee+0.078
Neg logit contributions
und-0.126
nexpected-0.109
-0.107
aded-0.106
ued-0.106
Token someone
Token ID2130
Feature activation-8.36e-02

Pos logit contributions
 wounds+0.036
reating+0.032
pins+0.032
 knee+0.029
 gloves+0.029
Neg logit contributions
und-0.054
-0.047
nexpected-0.046
aded-0.045
ued-0.045
Token's
Token ID338
Feature activation-5.64e-02

Pos logit contributions
 wounds+0.117
reating+0.098
pins+0.097
 decide+0.095
 gloves+0.093
Neg logit contributions
und-0.176
-0.159
nexpected-0.157
aded-0.151
ued-0.144
Token tree
Token ID5509
Feature activation+8.22e-03

Pos logit contributions
 wounds+0.085
reating+0.074
pins+0.073
 gloves+0.068
 knee+0.068
Neg logit contributions
und-0.115
-0.101
nexpected-0.099
aded-0.096
ued-0.095
Token King
Token ID2677
Feature activation-5.65e-02

Pos logit contributions
 wounds+0.136
pins+0.124
reating+0.121
 gloves+0.109
line+0.109
Neg logit contributions
und-0.171
-0.151
nexpected-0.145
aded-0.142
ued-0.140
Token and
Token ID290
Feature activation-4.51e-02

Pos logit contributions
 wounds+0.064
reating+0.057
pins+0.055
 knee+0.050
line+0.050
Neg logit contributions
und-0.133
nexpected-0.121
-0.120
aded-0.117
ued-0.114
Token asked
Token ID1965
Feature activation-6.09e-02

Pos logit contributions
 wounds+0.070
reating+0.063
pins+0.061
 decide+0.058
 gloves+0.057
Neg logit contributions
und-0.088
-0.079
nexpected-0.076
aded-0.075
ued-0.072
Token for
Token ID329
Feature activation-1.24e-02

Pos logit contributions
 wounds+0.086
reating+0.075
pins+0.073
 gloves+0.068
 knee+0.067
Neg logit contributions
und-0.129
-0.115
nexpected-0.112
aded-0.109
ued-0.108
Token his
Token ID465
Feature activation-7.39e-02

Pos logit contributions
 wounds+0.018
pins+0.016
reating+0.016
 knee+0.015
 sax+0.014
Neg logit contributions
und-0.025
nexpected-0.022
-0.022
aded-0.021
ued-0.021
Token permission
Token ID7170
Feature activation-2.25e-01

Pos logit contributions
 wounds+0.105
reating+0.091
pins+0.090
 gloves+0.083
 knee+0.082
Neg logit contributions
und-0.156
-0.138
nexpected-0.135
aded-0.131
ued-0.130
Token.
Token ID13
Feature activation-6.22e-02

Pos logit contributions
 wounds+0.317
reating+0.258
pins+0.255
 gloves+0.241
 sax+0.240
Neg logit contributions
und-0.477
-0.422
ued-0.404
nexpected-0.404
aded-0.392
Token The
Token ID383
Feature activation-3.82e-02

Pos logit contributions
 wounds+0.092
pins+0.086
reating+0.086
line+0.083
 knee+0.079
Neg logit contributions
und-0.116
nexpected-0.106
aded-0.104
ued-0.098
-0.097
Token King
Token ID2677
Feature activation-2.17e-02

Pos logit contributions
 wounds+0.061
pins+0.054
reating+0.053
 gloves+0.049
 knee+0.049
Neg logit contributions
und-0.073
-0.065
nexpected-0.063
ued-0.061
aded-0.061
Token said
Token ID531
Feature activation-5.85e-03

Pos logit contributions
 even+0.027
 wounds+0.027
 aer+0.025
 knee+0.024
 bike+0.024
Neg logit contributions
und-0.045
-0.043
nexpected-0.039
aded-0.038
ued-0.037
Token,
Token ID11
Feature activation+9.02e-03

Pos logit contributions
 wounds+0.009
pins+0.008
reating+0.008
 knee+0.008
 gloves+0.008
Neg logit contributions
und-0.012
-0.010
nexpected-0.010
aded-0.010
ued-0.009
Token someone
Token ID2130
Feature activation-1.03e-01

Pos logit contributions
und+0.003
+0.003
nexpected+0.003
aded+0.003
ued+0.003
Neg logit contributions
 wounds-0.002
reating-0.002
pins-0.002
 gloves-0.002
 knee-0.002
Token without
Token ID1231
Feature activation-4.30e-02

Pos logit contributions
 wounds+0.134
reating+0.113
pins+0.112
 gloves+0.105
 knee+0.102
Neg logit contributions
und-0.229
-0.208
nexpected-0.204
aded-0.199
ued-0.192
Token first
Token ID717
Feature activation-3.58e-02

Pos logit contributions
 wounds+0.070
reating+0.062
pins+0.060
 gloves+0.057
 knee+0.056
Neg logit contributions
und-0.083
-0.074
nexpected-0.069
ued-0.068
aded-0.068
Token asking
Token ID4737
Feature activation-6.46e-02

Pos logit contributions
 wounds+0.026
pins+0.020
reating+0.020
 gloves+0.017
 knee+0.017
Neg logit contributions
und-0.099
-0.091
nexpected-0.090
aded-0.088
ued-0.086
Token their
Token ID511
Feature activation-4.70e-02

Pos logit contributions
 wounds+0.071
reating+0.059
pins+0.058
 gloves+0.052
 knee+0.052
Neg logit contributions
und-0.156
-0.140
nexpected-0.138
aded-0.135
ued-0.134
Token permission
Token ID7170
Feature activation-2.25e-01

Pos logit contributions
 wounds+0.058
reating+0.052
pins+0.051
line+0.046
 gloves+0.045
Neg logit contributions
und-0.105
-0.094
nexpected-0.092
aded-0.090
ued-0.089
Token.
Token ID13
Feature activation-8.07e-02

Pos logit contributions
 wounds+0.273
reating+0.216
pins+0.212
 gloves+0.201
 sax+0.198
Neg logit contributions
und-0.523
-0.466
ued-0.451
nexpected-0.445
mon-0.440
Token We
Token ID775
Feature activation-7.70e-04

Pos logit contributions
 wounds+0.124
pins+0.116
reating+0.114
line+0.105
 knee+0.105
Neg logit contributions
und-0.148
nexpected-0.132
aded-0.130
ued-0.125
-0.125
Token should
Token ID815
Feature activation+2.54e-03

Pos logit contributions
 wounds+0.001
 decide+0.001
pins+0.001
 even+0.001
 jacket+0.001
Neg logit contributions
und-0.001
nexpected-0.001
-0.001
aded-0.001
live-0.001
Token ask
Token ID1265
Feature activation-1.30e-01

Pos logit contributions
und+0.006
+0.005
nexpected+0.005
aded+0.005
ued+0.005
Neg logit contributions
 wounds-0.003
pins-0.003
reating-0.003
 decide-0.002
 knee-0.002
Token the
Token ID262
Feature activation-1.00e-01

Pos logit contributions
 wounds+0.188
reating+0.166
pins+0.165
 knee+0.153
 gloves+0.153
Neg logit contributions
und-0.264
-0.228
nexpected-0.228
aded-0.220
ued-0.220
Token belongings
Token ID30985
Feature activation-5.34e-02

Pos logit contributions
 wounds+0.136
reating+0.115
pins+0.112
 gloves+0.104
 knee+0.102
Neg logit contributions
und-0.232
-0.210
nexpected-0.203
aded-0.200
ued-0.197
Token,
Token ID11
Feature activation+1.94e-03

Pos logit contributions
 wounds+0.063
reating+0.051
pins+0.050
 knee+0.047
 gloves+0.046
Neg logit contributions
und-0.125
-0.111
ued-0.108
nexpected-0.108
aded-0.105
Token and
Token ID290
Feature activation-7.31e-02

Pos logit contributions
und+0.004
+0.004
nexpected+0.004
aded+0.003
ued+0.003
Neg logit contributions
 wounds-0.003
reating-0.002
pins-0.002
 knee-0.002
 gloves-0.002
Token ask
Token ID1265
Feature activation-1.38e-01

Pos logit contributions
 wounds+0.103
reating+0.089
pins+0.087
 gloves+0.081
 knee+0.080
Neg logit contributions
und-0.156
-0.140
nexpected-0.136
aded-0.133
ued-0.131
Token for
Token ID329
Feature activation-4.26e-02

Pos logit contributions
 wounds+0.180
pins+0.155
reating+0.153
 knee+0.142
 gloves+0.141
Neg logit contributions
und-0.302
-0.264
nexpected-0.263
aded-0.255
ued-0.253
Token permission
Token ID7170
Feature activation-2.25e-01

Pos logit contributions
 wounds+0.058
pins+0.054
reating+0.053
 knee+0.049
line+0.049
Neg logit contributions
und-0.085
nexpected-0.077
-0.075
aded-0.074
ued-0.072
Token before
Token ID878
Feature activation-7.50e-02

Pos logit contributions
 wounds+0.324
pins+0.267
reating+0.266
 gloves+0.254
 sax+0.252
Neg logit contributions
und-0.464
-0.406
nexpected-0.394
ued-0.392
mon-0.377
Token using
Token ID1262
Feature activation-5.99e-02

Pos logit contributions
 wounds+0.114
pins+0.099
reating+0.099
 gloves+0.091
line+0.090
Neg logit contributions
und-0.152
nexpected-0.131
-0.129
aded-0.127
ued-0.126
Token them
Token ID606
Feature activation-8.11e-02

Pos logit contributions
 wounds+0.086
reating+0.076
pins+0.075
 knee+0.069
 gloves+0.068
Neg logit contributions
und-0.124
-0.109
nexpected-0.107
aded-0.104
ued-0.103
Token.
Token ID13
Feature activation-7.30e-02

Pos logit contributions
 wounds+0.093
pins+0.078
reating+0.077
 gloves+0.071
 knee+0.068
Neg logit contributions
und-0.193
-0.175
nexpected-0.171
aded-0.169
ued-0.163
Token<|endoftext|>
Token ID50256
Feature activation+1.17e-01

Pos logit contributions
 wounds+0.127
reating+0.114
pins+0.113
 knee+0.105
 gloves+0.105
Neg logit contributions
und-0.129
-0.110
nexpected-0.110
aded-0.107
ued-0.105
Token things
Token ID1243
Feature activation-4.56e-02

Pos logit contributions
und+0.026
+0.023
nexpected+0.022
aded+0.022
ued+0.021
Neg logit contributions
 wounds-0.017
reating-0.014
pins-0.014
 knee-0.013
 gloves-0.013
Token without
Token ID1231
Feature activation-4.87e-02

Pos logit contributions
 wounds+0.069
reating+0.060
pins+0.060
 gloves+0.055
 knee+0.055
Neg logit contributions
und-0.092
-0.081
nexpected-0.080
aded-0.078
ued-0.076
Token your
Token ID534
Feature activation-1.15e-01

Pos logit contributions
 wounds+0.070
reating+0.061
pins+0.059
 gloves+0.055
 knee+0.055
Neg logit contributions
und-0.103
-0.092
nexpected-0.088
aded-0.086
ued-0.086
Token mom
Token ID1995
Feature activation-1.56e-01

Pos logit contributions
 wounds+0.147
reating+0.146
pins+0.135
line+0.120
 gloves+0.119
Neg logit contributions
und-0.243
-0.219
nexpected-0.215
aded-0.210
ued-0.204
Token's
Token ID338
Feature activation-9.89e-02

Pos logit contributions
 wounds+0.208
reating+0.179
pins+0.179
 knee+0.165
 gloves+0.163
Neg logit contributions
und-0.342
-0.306
nexpected-0.302
aded-0.298
ued-0.290
Token permission
Token ID7170
Feature activation-2.24e-01

Pos logit contributions
 wounds+0.113
reating+0.110
pins+0.103
line+0.090
 sax+0.090
Neg logit contributions
und-0.221
nexpected-0.197
-0.195
aded-0.192
ued-0.188
Token.
Token ID13
Feature activation-6.89e-02

Pos logit contributions
 wounds+0.301
reating+0.242
pins+0.233
 gloves+0.226
 sax+0.222
Neg logit contributions
und-0.494
-0.441
ued-0.421
nexpected-0.412
mon-0.407
Token Next
Token ID7406
Feature activation+6.46e-02

Pos logit contributions
 wounds+0.117
reating+0.104
pins+0.101
 knee+0.096
 gloves+0.094
Neg logit contributions
und-0.123
-0.109
aded-0.107
nexpected-0.107
ued-0.104
Token time
Token ID640
Feature activation+1.49e-01

Pos logit contributions
und+0.123
uttered+0.115
+0.111
imm+0.110
inged+0.110
Neg logit contributions
 wounds-0.098
 even-0.097
reating-0.087
 bra-0.087
 aer-0.086
Token,
Token ID11
Feature activation+2.05e-02

Pos logit contributions
und+0.248
+0.220
nexpected+0.212
aded+0.209
ued+0.198
Neg logit contributions
 wounds-0.268
pins-0.249
reating-0.248
 gloves-0.230
line-0.230
Token Josh
Token ID8518
Feature activation+1.57e-03

Pos logit contributions
und+0.043
+0.039
aded+0.037
nexpected+0.037
ued+0.036
Neg logit contributions
 wounds-0.030
reating-0.026
pins-0.026
 decide-0.024
 gloves-0.024
Token his
Token ID465
Feature activation-9.85e-02

Pos logit contributions
 wounds+0.196
reating+0.173
pins+0.170
 gloves+0.160
 knee+0.160
Neg logit contributions
und-0.228
-0.199
nexpected-0.195
aded-0.189
ued-0.187
Token mom
Token ID1995
Feature activation-1.66e-01

Pos logit contributions
pins+0.159
 wounds+0.154
reating+0.151
line+0.150
bars+0.134
Neg logit contributions
und-0.163
-0.149
nexpected-0.145
time-0.134
ued-0.134
Token's
Token ID338
Feature activation-9.45e-02

Pos logit contributions
 wounds+0.250
pins+0.216
reating+0.199
 gloves+0.192
 sax+0.188
Neg logit contributions
und-0.335
-0.287
nexpected-0.269
date-0.267
day-0.266
Token records
Token ID4406
Feature activation-1.05e-01

Pos logit contributions
pins+0.135
reating+0.128
line+0.125
 wounds+0.121
bars+0.117
Neg logit contributions
und-0.170
nexpected-0.155
-0.154
aded-0.146
time-0.145
Token without
Token ID1231
Feature activation-5.82e-02

Pos logit contributions
 wounds+0.158
pins+0.134
reating+0.126
 gloves+0.119
 decide+0.116
Neg logit contributions
und-0.212
-0.184
nexpected-0.178
time-0.176
aded-0.174
Token permission
Token ID7170
Feature activation-2.24e-01

Pos logit contributions
 wounds+0.104
reating+0.093
pins+0.093
 knee+0.086
 gloves+0.086
Neg logit contributions
und-0.101
-0.086
nexpected-0.085
aded-0.082
ued-0.080
Token.
Token ID13
Feature activation-6.88e-02

Pos logit contributions
 wounds+0.329
reating+0.248
 sax+0.247
pins+0.245
 gloves+0.244
Neg logit contributions
und-0.468
-0.417
time-0.399
ued-0.396
day-0.383
Token He
Token ID679
Feature activation+4.71e-02

Pos logit contributions
 wounds+0.123
reating+0.111
pins+0.110
 gloves+0.103
 knee+0.103
Neg logit contributions
und-0.119
-0.102
nexpected-0.100
aded-0.097
ued-0.095
Token would
Token ID561
Feature activation-3.97e-02

Pos logit contributions
und+0.093
+0.083
nexpected+0.079
aded+0.078
ued+0.077
Neg logit contributions
 wounds-0.074
pins-0.062
reating-0.062
 gloves-0.060
 decide-0.059
Token repeat
Token ID9585
Feature activation-1.67e-02

Pos logit contributions
 wounds+0.067
reating+0.060
pins+0.059
 gloves+0.055
 knee+0.055
Neg logit contributions
und-0.073
-0.063
nexpected-0.061
aded-0.059
ued-0.059
Token to
Token ID284
Feature activation-1.36e-02

Pos logit contributions
 wounds+0.025
pins+0.023
reating+0.022
 knee+0.020
 gloves+0.020
Neg logit contributions
und-0.033
nexpected-0.028
-0.028
aded-0.027
ued-0.027
Token you
Token ID345
Feature activation-5.00e-02

Pos logit contributions
 wounds+0.081
pins+0.068
reating+0.067
 sax+0.065
 gloves+0.063
Neg logit contributions
und-0.113
-0.097
ued-0.094
time-0.093
nexpected-0.090
Token do
Token ID466
Feature activation+1.64e-02

Pos logit contributions
 wounds+0.060
reating+0.052
pins+0.050
 gloves+0.045
 knee+0.045
Neg logit contributions
und-0.116
-0.103
nexpected-0.101
ued-0.099
aded-0.099
Token not
Token ID407
Feature activation-4.48e-02

Pos logit contributions
und+0.038
+0.034
nexpected+0.034
aded+0.033
ued+0.033
Neg logit contributions
 wounds-0.019
reating-0.016
pins-0.016
 gloves-0.015
 knee-0.014
Token know
Token ID760
Feature activation-4.56e-02

Pos logit contributions
 wounds+0.064
pins+0.055
reating+0.052
 sax+0.049
bars+0.049
Neg logit contributions
und-0.093
ued-0.078
aded-0.078
-0.077
nexpected-0.076
Token without
Token ID1231
Feature activation-5.78e-02

Pos logit contributions
 wounds+0.064
pins+0.057
reating+0.053
 knee+0.050
line+0.050
Neg logit contributions
und-0.095
nexpected-0.080
-0.079
ued-0.077
aded-0.076
Token permission
Token ID7170
Feature activation-2.23e-01

Pos logit contributions
 wounds+0.092
pins+0.082
reating+0.082
 knee+0.075
 gloves+0.075
Neg logit contributions
und-0.111
-0.096
nexpected-0.095
aded-0.092
ued-0.090
Token.
Token ID13
Feature activation-7.63e-02

Pos logit contributions
 wounds+0.303
pins+0.238
reating+0.237
 sax+0.232
 gloves+0.229
Neg logit contributions
und-0.488
-0.419
ued-0.418
nexpected-0.399
time-0.398
Token They
Token ID1119
Feature activation+4.43e-02

Pos logit contributions
 wounds+0.127
reating+0.114
pins+0.113
 knee+0.107
 gloves+0.104
Neg logit contributions
und-0.137
nexpected-0.119
aded-0.118
-0.118
ued-0.114
Token can
Token ID460
Feature activation-2.95e-02

Pos logit contributions
und+0.090
+0.081
nexpected+0.079
aded+0.078
mon+0.073
Neg logit contributions
 wounds-0.067
pins-0.058
 decide-0.057
 gloves-0.054
reating-0.054
Token be
Token ID307
Feature activation-9.50e-02

Pos logit contributions
 wounds+0.049
reating+0.043
pins+0.042
 gloves+0.040
 knee+0.039
Neg logit contributions
und-0.056
-0.049
nexpected-0.047
aded-0.046
ued-0.045
Token dangerous
Token ID4923
Feature activation+3.84e-03

Pos logit contributions
 wounds+0.142
reating+0.122
pins+0.121
 gloves+0.113
 knee+0.112
Neg logit contributions
und-0.196
-0.174
nexpected-0.169
aded-0.164
ued-0.163
Token promised
Token ID8072
Feature activation-1.60e-01

Pos logit contributions
 wounds+0.141
reating+0.124
pins+0.124
 decide+0.116
 knee+0.115
Neg logit contributions
und-0.179
-0.160
nexpected-0.150
aded-0.149
ued-0.149
Token to
Token ID284
Feature activation-7.76e-02

Pos logit contributions
 wounds+0.296
pins+0.265
 gloves+0.262
reating+0.260
 sax+0.257
Neg logit contributions
und-0.251
-0.212
nexpected-0.208
 Luckily-0.190
aded-0.185
Token always
Token ID1464
Feature activation-4.38e-02

Pos logit contributions
 wounds+0.126
reating+0.108
pins+0.108
 gloves+0.102
 knee+0.099
Neg logit contributions
und-0.149
-0.130
nexpected-0.129
aded-0.121
mon-0.118
Token ask
Token ID1265
Feature activation-1.54e-01

Pos logit contributions
 wounds+0.053
reating+0.045
pins+0.043
 knee+0.040
 gloves+0.040
Neg logit contributions
und-0.102
-0.092
nexpected-0.089
aded-0.088
ued-0.087
Token for
Token ID329
Feature activation-5.14e-02

Pos logit contributions
 wounds+0.230
pins+0.211
 jacket+0.209
 sax+0.202
 knee+0.201
Neg logit contributions
und-0.262
nexpected-0.243
-0.227
aded-0.223
 Larry-0.223
Token permission
Token ID7170
Feature activation-2.23e-01

Pos logit contributions
 wounds+0.075
pins+0.072
reating+0.068
 jacket+0.066
 sax+0.066
Neg logit contributions
und-0.093
nexpected-0.089
aded-0.082
-0.081
mon-0.078
Token before
Token ID878
Feature activation-8.37e-02

Pos logit contributions
 wounds+0.347
 sax+0.289
pins+0.287
 gloves+0.286
 jacket+0.284
Neg logit contributions
und-0.402
-0.364
nexpected-0.354
time-0.350
day-0.342
Token drawing
Token ID8263
Feature activation-5.24e-02

Pos logit contributions
 wounds+0.138
reating+0.124
pins+0.124
line+0.114
 gloves+0.114
Neg logit contributions
und-0.158
ued-0.132
nexpected-0.131
-0.131
aded-0.130
Token on
Token ID319
Feature activation+2.64e-02

Pos logit contributions
 wounds+0.080
reating+0.068
pins+0.068
 knee+0.064
 jacket+0.064
Neg logit contributions
und-0.104
-0.090
nexpected-0.088
aded-0.084
mon-0.083
Token anything
Token ID1997
Feature activation-8.74e-02

Pos logit contributions
und+0.053
nexpected+0.046
+0.045
mon+0.042
aded+0.041
Neg logit contributions
 wounds-0.040
reating-0.035
pins-0.035
 knee-0.032
 gloves-0.032
Token again
Token ID757
Feature activation-8.58e-02

Pos logit contributions
 wounds+0.115
reating+0.100
pins+0.098
 gloves+0.090
 knee+0.088
Neg logit contributions
und-0.193
-0.172
nexpected-0.168
aded-0.166
ued-0.163
Token jokes
Token ID14532
Feature activation-4.83e-02

Pos logit contributions
 wounds+0.126
reating+0.108
pins+0.104
 knee+0.101
 gloves+0.099
Neg logit contributions
und-0.176
-0.159
nexpected-0.156
aded-0.153
ued-0.149
Token on
Token ID319
Feature activation+3.72e-02

Pos logit contributions
 wounds+0.056
reating+0.049
pins+0.048
 gloves+0.043
 knee+0.043
Neg logit contributions
und-0.113
-0.102
nexpected-0.101
aded-0.099
ued-0.096
Token others
Token ID1854
Feature activation-1.14e-01

Pos logit contributions
und+0.080
+0.071
nexpected+0.069
aded+0.068
ued+0.067
Neg logit contributions
 wounds-0.052
reating-0.045
pins-0.044
 gloves-0.041
 knee-0.040
Token without
Token ID1231
Feature activation-4.22e-02

Pos logit contributions
 wounds+0.106
reating+0.085
pins+0.083
 gloves+0.072
 knee+0.071
Neg logit contributions
und-0.296
-0.269
nexpected-0.264
aded-0.259
ued-0.257
Token their
Token ID511
Feature activation-5.54e-02

Pos logit contributions
 wounds+0.069
reating+0.061
pins+0.060
 gloves+0.057
 knee+0.056
Neg logit contributions
und-0.080
-0.070
nexpected-0.068
aded-0.066
ued-0.066
Token permission
Token ID7170
Feature activation-2.22e-01

Pos logit contributions
 wounds+0.076
reating+0.068
pins+0.066
 gloves+0.060
 knee+0.059
Neg logit contributions
und-0.118
-0.105
nexpected-0.103
aded-0.101
ued-0.099
Token."
Token ID526
Feature activation-6.44e-02

Pos logit contributions
 wounds+0.279
reating+0.216
pins+0.210
 gloves+0.208
 sax+0.207
Neg logit contributions
und-0.499
ued-0.443
-0.436
used-0.419
mon-0.418
Token
Token ID198
Feature activation+7.11e-02

Pos logit contributions
 wounds+0.076
 knee+0.062
pins+0.062
 jacket+0.059
 decide+0.059
Neg logit contributions
und-0.140
nexpected-0.123
aded-0.123
ued-0.121
-0.115
Token
Token ID198
Feature activation+7.35e-02

Pos logit contributions
und+0.143
+0.126
nexpected+0.124
aded+0.122
ued+0.120
Neg logit contributions
 wounds-0.101
pins-0.086
reating-0.084
 knee-0.078
 gloves-0.078
TokenThe
Token ID464
Feature activation-6.27e-04

Pos logit contributions
und+0.152
+0.137
aded+0.125
nexpected+0.124
uttered+0.123
Neg logit contributions
 wounds-0.101
pins-0.082
 decide-0.081
 jacket-0.076
 gloves-0.076
Token little
Token ID1310
Feature activation+1.15e-01

Pos logit contributions
 wounds+0.001
reating+0.001
pins+0.001
 gloves+0.001
 knee+0.001
Neg logit contributions
und-0.001
-0.001
nexpected-0.001
aded-0.001
ued-0.001
Token read
Token ID1100
Feature activation-7.97e-02

Pos logit contributions
 wounds+0.117
reating+0.116
pins+0.106
 stations+0.102
 sax+0.100
Neg logit contributions
und-0.115
-0.110
nexpected-0.107
 Luckily-0.103
ued-0.100
Token her
Token ID607
Feature activation-7.42e-02

Pos logit contributions
 wounds+0.135
 knee+0.116
line+0.115
pins+0.115
 decide+0.112
Neg logit contributions
und-0.141
time-0.123
nexpected-0.120
-0.117
ued-0.111
Token books
Token ID3835
Feature activation-7.11e-02

Pos logit contributions
reating+0.109
line+0.105
 wounds+0.104
pins+0.103
 decide+0.093
Neg logit contributions
und-0.137
nexpected-0.129
-0.128
 Luckily-0.120
ued-0.120
Token.
Token ID13
Feature activation-4.46e-02

Pos logit contributions
 wounds+0.113
reating+0.085
 knee+0.084
 gloves+0.083
 sax+0.082
Neg logit contributions
und-0.145
-0.131
time-0.127
ued-0.124
day-0.119
Token One
Token ID1881
Feature activation+3.57e-02

Pos logit contributions
 wounds+0.104
reating+0.092
pins+0.090
 gloves+0.085
 sax+0.085
Neg logit contributions
und-0.058
-0.051
nexpected-0.044
 Luckily-0.043
mon-0.041
Token day
Token ID1110
Feature activation+1.93e-01

Pos logit contributions
und+0.053
+0.045
nexpected+0.043
aded+0.042
ued+0.041
Neg logit contributions
 wounds-0.073
reating-0.067
pins-0.066
 gloves-0.062
 knee-0.062
Token,
Token ID11
Feature activation+5.93e-02

Pos logit contributions
und+0.403
aded+0.351
+0.344
ued+0.339
nexpected+0.335
Neg logit contributions
pins-0.236
 wounds-0.234
reating-0.226
 bike-0.226
 knee-0.225
Token her
Token ID607
Feature activation-2.99e-02

Pos logit contributions
und+0.147
aded+0.127
+0.126
nexpected+0.125
ued+0.125
Neg logit contributions
 wounds-0.064
pins-0.057
reating-0.056
 gloves-0.053
 knee-0.052
Token mom
Token ID1995
Feature activation-8.77e-02

Pos logit contributions
 wounds+0.031
reating+0.029
pins+0.026
line+0.025
 stations+0.022
Neg logit contributions
und-0.071
-0.067
nexpected-0.065
aded-0.063
 Luckily-0.061
Token came
Token ID1625
Feature activation-2.72e-02

Pos logit contributions
 wounds+0.080
reating+0.080
pins+0.079
 gloves+0.071
 jacket+0.070
Neg logit contributions
und-0.215
-0.204
ued-0.195
nexpected-0.190
aded-0.189
Token in
Token ID287
Feature activation+5.61e-02

Pos logit contributions
 wounds+0.043
reating+0.038
pins+0.037
 gloves+0.034
 knee+0.034
Neg logit contributions
und-0.054
-0.047
nexpected-0.046
aded-0.045
ued-0.044
Token!"
Token ID2474
Feature activation-6.84e-02

Pos logit contributions
 wounds+0.101
reating+0.089
pins+0.088
 knee+0.081
 gloves+0.080
Neg logit contributions
und-0.139
-0.120
nexpected-0.117
aded-0.115
ued-0.113
Token<|endoftext|>
Token ID50256
Feature activation+1.28e-01

Pos logit contributions
 wounds+0.118
reating+0.106
pins+0.105
 gloves+0.098
 knee+0.098
Neg logit contributions
und-0.124
-0.107
nexpected-0.105
aded-0.102
ued-0.100
TokenOnce
Token ID7454
Feature activation+2.61e-01

Pos logit contributions
und+0.188
nexpected+0.164
aded+0.157
ued+0.156
+0.154
Neg logit contributions
 wounds-0.244
reating-0.226
pins-0.223
 gloves-0.215
 knee-0.211
Token upon
Token ID2402
Feature activation+1.73e-01

Pos logit contributions
und+0.393
+0.332
nexpected+0.318
aded+0.312
ued+0.301
Neg logit contributions
 wounds-0.517
pins-0.483
reating-0.483
 knee-0.448
 gloves-0.447
Token a
Token ID257
Feature activation+5.19e-02

Pos logit contributions
und+0.262
+0.226
nexpected+0.214
aded+0.207
time+0.204
Neg logit contributions
 wounds-0.340
pins-0.310
reating-0.304
line-0.285
bars-0.281
Token time
Token ID640
Feature activation+2.12e-01

Pos logit contributions
und+0.077
used+0.063
ued+0.062
uttered+0.062
+0.061
Neg logit contributions
 wounds-0.097
 gloves-0.089
 bike-0.085
 knee-0.084
 decide-0.083
Token,
Token ID11
Feature activation+8.48e-02

Pos logit contributions
und+0.425
aded+0.388
nexpected+0.371
+0.362
ued+0.362
Neg logit contributions
 wounds-0.255
pins-0.245
 bike-0.243
 knee-0.238
reating-0.231
Token there
Token ID612
Feature activation+2.18e-01

Pos logit contributions
und+0.164
+0.155
nexpected+0.145
aded+0.143
live+0.137
Neg logit contributions
 even-0.115
 wounds-0.111
 back-0.105
 knee-0.104
 bike-0.104
Token was
Token ID373
Feature activation+8.19e-02

Pos logit contributions
und+0.469
ued+0.457
igsaw+0.439
+0.434
used+0.429
Neg logit contributions
 even-0.218
 mom-0.198
 wounds-0.192
 decide-0.187
 jacket-0.184
Token a
Token ID257
Feature activation+4.10e-02

Pos logit contributions
und+0.169
+0.148
nexpected+0.146
aded+0.142
ued+0.141
Neg logit contributions
 wounds-0.119
reating-0.106
pins-0.104
 gloves-0.097
 knee-0.096
Token house
Token ID2156
Feature activation+8.91e-02

Pos logit contributions
und+0.053
+0.050
mon+0.043
nexpected+0.043
aded+0.042
Neg logit contributions
 wounds-0.087
reating-0.083
pins-0.077
bars-0.075
 stations-0.072
Token big
Token ID1263
Feature activation+4.23e-02

Pos logit contributions
 wounds+0.047
reating+0.042
pins+0.042
 decide+0.038
 knee+0.038
Neg logit contributions
und-0.047
-0.041
nexpected-0.038
mon-0.038
ued-0.037
Token pile
Token ID14540
Feature activation+3.97e-02

Pos logit contributions
und+0.082
+0.072
nexpected+0.070
aded+0.067
ued+0.067
Neg logit contributions
 wounds-0.067
reating-0.059
pins-0.058
 gloves-0.053
 knee-0.053
Token of
Token ID286
Feature activation+6.76e-02

Pos logit contributions
und+0.078
+0.069
nexpected+0.067
aded+0.065
ued+0.065
Neg logit contributions
 wounds-0.061
reating-0.054
pins-0.054
 gloves-0.050
 knee-0.050
Token hay
Token ID27678
Feature activation+7.04e-05

Pos logit contributions
und+0.147
nexpected+0.126
aded+0.124
+0.123
ued+0.120
Neg logit contributions
 wounds-0.087
pins-0.081
reating-0.076
line-0.071
 knee-0.069
Token.
Token ID13
Feature activation-5.07e-02

Pos logit contributions
und+0.000
+0.000
nexpected+0.000
aded+0.000
ued+0.000
Neg logit contributions
 wounds-0.000
pins-0.000
reating-0.000
 gloves-0.000
 knee-0.000
Token There
Token ID1318
Feature activation+1.62e-01

Pos logit contributions
 wounds+0.078
pins+0.068
 gloves+0.066
 knee+0.065
reating+0.065
Neg logit contributions
und-0.095
nexpected-0.086
aded-0.083
-0.081
ued-0.080
Token were
Token ID547
Feature activation+7.63e-02

Pos logit contributions
und+0.331
+0.312
nexpected+0.288
igsaw+0.286
aded+0.285
Neg logit contributions
 wounds-0.203
 even-0.196
pins-0.191
line-0.188
 gloves-0.185
Token the
Token ID262
Feature activation-3.57e-02

Pos logit contributions
und+0.164
+0.153
nexpected+0.145
ued+0.141
aded+0.138
Neg logit contributions
 wounds-0.107
reating-0.091
 gloves-0.089
pins-0.088
 knee-0.081
Token white
Token ID2330
Feature activation+2.51e-02

Pos logit contributions
 wounds+0.050
reating+0.044
pins+0.043
 knee+0.039
 gloves+0.039
Neg logit contributions
und-0.076
-0.067
nexpected-0.065
aded-0.064
ued-0.063
Token cows
Token ID22575
Feature activation-3.22e-02

Pos logit contributions
und+0.058
+0.052
nexpected+0.051
ued+0.050
aded+0.050
Neg logit contributions
 wounds-0.031
reating-0.026
 gloves-0.026
pins-0.025
 knee-0.024
Token!
Token ID0
Feature activation-3.40e-02

Pos logit contributions
 wounds+0.040
pins+0.036
line+0.034
 gloves+0.033
 jacket+0.033
Neg logit contributions
und-0.070
nexpected-0.066
aded-0.062
-0.062
ued-0.061
Token.
Token ID13
Feature activation-5.45e-02

Pos logit contributions
und+0.009
nexpected+0.009
+0.009
aded+0.008
live+0.008
Neg logit contributions
 wounds-0.005
pins-0.004
reating-0.004
line-0.004
 decide-0.004
Token<|endoftext|>
Token ID50256
Feature activation+1.25e-01

Pos logit contributions
 wounds+0.095
pins+0.087
reating+0.087
line+0.081
 knee+0.080
Neg logit contributions
und-0.093
nexpected-0.082
aded-0.079
-0.078
ued-0.075
TokenOnce
Token ID7454
Feature activation+2.62e-01

Pos logit contributions
und+0.189
nexpected+0.183
aded+0.175
ued+0.166
+0.160
Neg logit contributions
 wounds-0.203
reating-0.195
pins-0.193
line-0.188
 gloves-0.182
Token upon
Token ID2402
Feature activation+1.69e-01

Pos logit contributions
und+0.421
+0.362
aded+0.351
nexpected+0.351
ued+0.321
Neg logit contributions
 wounds-0.478
pins-0.460
reating-0.450
line-0.426
 bike-0.425
Token a
Token ID257
Feature activation+4.67e-02

Pos logit contributions
und+0.276
+0.235
nexpected+0.229
aded+0.223
ued+0.217
Neg logit contributions
 wounds-0.314
pins-0.286
reating-0.280
line-0.262
 sax-0.257
Token time
Token ID640
Feature activation+2.10e-01

Pos logit contributions
und+0.071
uttered+0.057
used+0.056
+0.056
nexpected+0.056
Neg logit contributions
 wounds-0.085
 gloves-0.079
 bike-0.076
 even-0.075
 knee-0.074
Token,
Token ID11
Feature activation+8.90e-02

Pos logit contributions
und+0.449
aded+0.426
nexpected+0.405
+0.392
acked+0.392
Neg logit contributions
pins-0.208
 bike-0.207
 even-0.200
 back-0.197
 wounds-0.193
Token there
Token ID612
Feature activation+2.15e-01

Pos logit contributions
und+0.170
+0.159
nexpected+0.154
aded+0.151
 contemplation+0.142
Neg logit contributions
 even-0.117
 back-0.116
 wounds-0.114
 knee-0.111
 bike-0.109
Token was
Token ID373
Feature activation+7.82e-02

Pos logit contributions
und+0.446
ued+0.431
+0.418
aded+0.411
used+0.403
Neg logit contributions
 even-0.234
 wounds-0.215
 mom-0.212
 aer-0.207
 knee-0.207
Token a
Token ID257
Feature activation+3.31e-02

Pos logit contributions
und+0.170
+0.149
nexpected+0.147
ued+0.143
aded+0.142
Neg logit contributions
 wounds-0.105
reating-0.093
pins-0.091
 gloves-0.087
 knee-0.084
Token graceful
Token ID44363
Feature activation+8.43e-02

Pos logit contributions
und+0.046
+0.043
nexpected+0.038
mon+0.037
aded+0.037
Neg logit contributions
 wounds-0.068
reating-0.064
pins-0.060
bars-0.058
 stations-0.055
Tokenfree
Token ID5787
Feature activation+2.36e-03

Pos logit contributions
 wounds+0.128
reating+0.108
pins+0.106
 gloves+0.097
 knee+0.096
Neg logit contributions
und-0.239
-0.213
nexpected-0.208
aded-0.204
ued-0.202
Token insect
Token ID12833
Feature activation-6.24e-02

Pos logit contributions
und+0.005
+0.005
nexpected+0.005
ued+0.004
aded+0.004
Neg logit contributions
 wounds-0.003
 knee-0.003
pins-0.002
reating-0.002
 gloves-0.002
Token companion
Token ID15185
Feature activation-4.38e-02

Pos logit contributions
 wounds+0.071
reating+0.060
pins+0.059
 gloves+0.053
 knee+0.052
Neg logit contributions
und-0.149
-0.133
nexpected-0.131
aded-0.128
ued-0.127
Token.
Token ID13
Feature activation-5.63e-02

Pos logit contributions
 wounds+0.042
reating+0.035
pins+0.034
 gloves+0.029
 knee+0.029
Neg logit contributions
und-0.112
-0.102
nexpected-0.100
aded-0.099
ued-0.096
Token<|endoftext|>
Token ID50256
Feature activation+1.31e-01

Pos logit contributions
 wounds+0.086
pins+0.077
reating+0.076
 knee+0.071
line+0.071
Neg logit contributions
und-0.106
nexpected-0.095
aded-0.092
-0.090
ued-0.089
TokenOnce
Token ID7454
Feature activation+2.65e-01

Pos logit contributions
und+0.195
nexpected+0.185
aded+0.179
ued+0.169
+0.160
Neg logit contributions
 wounds-0.217
reating-0.207
pins-0.206
line-0.201
 gloves-0.197
Token upon
Token ID2402
Feature activation+1.73e-01

Pos logit contributions
und+0.424
+0.364
aded+0.355
nexpected+0.355
ued+0.327
Neg logit contributions
 wounds-0.483
pins-0.464
reating-0.455
 bike-0.431
line-0.430
Token a
Token ID257
Feature activation+5.12e-02

Pos logit contributions
und+0.299
+0.258
nexpected+0.250
aded+0.244
ued+0.239
Neg logit contributions
 wounds-0.307
pins-0.277
reating-0.274
line-0.253
 knee-0.249
Token time
Token ID640
Feature activation+2.12e-01

Pos logit contributions
und+0.076
+0.061
uttered+0.061
used+0.060
aded+0.060
Neg logit contributions
 wounds-0.096
 gloves-0.088
 bike-0.085
 jacket-0.083
 even-0.083
Token there
Token ID612
Feature activation+2.25e-01

Pos logit contributions
und+0.480
aded+0.462
nexpected+0.444
ited+0.434
acked+0.429
Neg logit contributions
 bike-0.170
 mom-0.167
 even-0.165
pins-0.162
 back-0.161
Token was
Token ID373
Feature activation+8.56e-02

Pos logit contributions
und+0.462
ued+0.450
+0.434
aded+0.424
igsaw+0.415
Neg logit contributions
 even-0.240
 mom-0.216
 knee-0.216
 wounds-0.215
 jacket-0.213
Token game
Token ID983
Feature activation+2.55e-02

Pos logit contributions
 wounds+0.116
reating+0.104
pins+0.104
 gloves+0.097
 knee+0.097
Neg logit contributions
und-0.098
-0.084
nexpected-0.081
aded-0.079
ued-0.077
Token.
Token ID13
Feature activation-5.39e-02

Pos logit contributions
und+0.059
+0.053
nexpected+0.052
aded+0.051
ued+0.050
Neg logit contributions
 wounds-0.031
reating-0.026
pins-0.026
 gloves-0.023
 knee-0.023
Token
Token ID198
Feature activation+6.31e-02

Pos logit contributions
 wounds+0.068
pins+0.067
line+0.062
 bike+0.061
 Miss+0.060
Neg logit contributions
und-0.098
aded-0.097
nexpected-0.094
ued-0.087
ited-0.083
Token
Token ID198
Feature activation+6.72e-02

Pos logit contributions
und+0.130
nexpected+0.118
aded+0.116
+0.114
ued+0.110
Neg logit contributions
 wounds-0.079
pins-0.070
reating-0.068
line-0.061
 bike-0.061
TokenBob
Token ID18861
Feature activation+7.03e-02

Pos logit contributions
und+0.116
nexpected+0.104
+0.104
aded+0.103
uttered+0.102
Neg logit contributions
 wounds-0.105
reating-0.099
pins-0.096
 bike-0.090
 gloves-0.090
Token agreed
Token ID4987
Feature activation+3.74e-02

Pos logit contributions
und+0.149
+0.140
nexpected+0.125
aded+0.123
ued+0.121
Neg logit contributions
 wounds-0.092
 even-0.082
 gloves-0.077
 decide-0.077
pins-0.076
Token and
Token ID290
Feature activation-5.98e-02

Pos logit contributions
und+0.084
+0.074
nexpected+0.072
aded+0.072
ued+0.069
Neg logit contributions
 wounds-0.049
pins-0.043
reating-0.041
 knee-0.040
 jacket-0.039
Token together
Token ID1978
Feature activation+1.28e-01

Pos logit contributions
 wounds+0.075
pins+0.070
 even+0.066
reating+0.066
 gloves+0.063
Neg logit contributions
und-0.129
-0.118
nexpected-0.114
aded-0.111
ued-0.109
Token,
Token ID11
Feature activation+2.69e-02

Pos logit contributions
und+0.333
+0.298
nexpected+0.292
live+0.279
aded+0.277
Neg logit contributions
 even-0.111
pins-0.106
 back-0.096
 wounds-0.088
 mom-0.084
Token they
Token ID484
Feature activation+5.60e-02

Pos logit contributions
und+0.051
nexpected+0.044
+0.044
aded+0.044
ued+0.043
Neg logit contributions
 wounds-0.042
reating-0.039
pins-0.038
 gloves-0.036
 bike-0.035
Token played
Token ID2826
Feature activation-6.62e-02

Pos logit contributions
und+0.103
+0.092
nexpected+0.089
ued+0.085
aded+0.085
Neg logit contributions
 wounds-0.090
pins-0.086
reating-0.083
 gloves-0.081
 decide-0.080
Token.
Token ID13
Feature activation-4.03e-02

Pos logit contributions
 wounds+0.007
reating+0.006
pins+0.005
 gloves+0.005
 knee+0.005
Neg logit contributions
und-0.024
-0.022
nexpected-0.022
aded-0.021
ued-0.021
Token Max
Token ID5436
Feature activation+2.40e-02

Pos logit contributions
 wounds+0.066
pins+0.058
 gloves+0.056
reating+0.056
 knee+0.054
Neg logit contributions
und-0.077
-0.066
nexpected-0.066
aded-0.066
ued-0.064
Token tried
Token ID3088
Feature activation-2.16e-02

Pos logit contributions
und+0.055
+0.052
igsaw+0.048
date+0.046
aded+0.046
Neg logit contributions
 wounds-0.029
 gloves-0.027
 even-0.026
 c-0.026
 jacket-0.026
Token to
Token ID284
Feature activation+5.89e-03

Pos logit contributions
 wounds+0.034
reating+0.030
pins+0.030
 knee+0.028
 gloves+0.028
Neg logit contributions
und-0.042
-0.037
nexpected-0.036
aded-0.035
ued-0.034
Token pick
Token ID2298
Feature activation-7.23e-02

Pos logit contributions
und+0.010
+0.009
nexpected+0.009
aded+0.009
ued+0.008
Neg logit contributions
 wounds-0.010
reating-0.009
pins-0.009
bars-0.008
 pilots-0.008
Token up
Token ID510
Feature activation+3.67e-02

Pos logit contributions
 wounds+0.112
reating+0.099
pins+0.098
 knee+0.091
 gloves+0.089
Neg logit contributions
und-0.143
-0.124
nexpected-0.123
aded-0.118
ued-0.118
Token the
Token ID262
Feature activation-7.58e-02

Pos logit contributions
und+0.075
+0.065
nexpected+0.065
aded+0.062
ued+0.062
Neg logit contributions
 wounds-0.054
reating-0.048
pins-0.047
 knee-0.043
line-0.043
Token stick
Token ID4859
Feature activation-8.80e-02

Pos logit contributions
 wounds+0.083
 gloves+0.067
pins+0.067
 bike+0.064
reating+0.064
Neg logit contributions
und-0.188
-0.168
nexpected-0.166
aded-0.162
ued-0.161
Token,
Token ID11
Feature activation+1.03e-02

Pos logit contributions
 wounds+0.134
pins+0.127
reating+0.122
 knee+0.120
 gloves+0.117
Neg logit contributions
und-0.171
nexpected-0.153
aded-0.151
-0.150
ued-0.141
Token but
Token ID475
Feature activation+4.78e-02

Pos logit contributions
und+0.019
+0.017
nexpected+0.016
ued+0.016
mon+0.016
Neg logit contributions
 wounds-0.016
reating-0.015
pins-0.014
 pilots-0.013
 stations-0.013
Token it
Token ID340
Feature activation+1.09e-02

Pos logit contributions
und+0.096
+0.084
nexpected+0.082
aded+0.080
ued+0.079
Neg logit contributions
 wounds-0.073
reating-0.065
pins-0.064
 gloves-0.059
 knee-0.059
Token happy
Token ID3772
Feature activation+1.17e-01

Pos logit contributions
und+0.116
+0.100
aded+0.099
nexpected+0.098
ued+0.096
Neg logit contributions
 wounds-0.085
reating-0.078
pins-0.074
 gloves-0.068
line-0.067
Token and
Token ID290
Feature activation-5.93e-02

Pos logit contributions
und+0.265
+0.239
nexpected+0.228
aded+0.225
 Larry+0.221
Neg logit contributions
 wounds-0.144
reating-0.125
pins-0.123
 gloves-0.109
line-0.107
Token kind
Token ID1611
Feature activation+4.99e-03

Pos logit contributions
 wounds+0.087
reating+0.077
pins+0.073
 gloves+0.069
 knee+0.068
Neg logit contributions
und-0.122
-0.108
nexpected-0.105
aded-0.103
ued-0.102
Token.
Token ID13
Feature activation-7.97e-02

Pos logit contributions
und+0.010
+0.009
time+0.009
ued+0.008
mon+0.008
Neg logit contributions
 wounds-0.007
reating-0.006
pins-0.006
 gloves-0.005
 hem-0.005
Token
Token ID198
Feature activation+3.67e-02

Pos logit contributions
 wounds+0.173
reating+0.159
pins+0.150
 knee+0.149
 gloves+0.144
Neg logit contributions
und-0.113
-0.097
nexpected-0.085
mon-0.083
 Luckily-0.083
Token
Token ID198
Feature activation+4.19e-02

Pos logit contributions
und+0.050
mon+0.040
aded+0.039
nexpected+0.038
 Larry+0.037
Neg logit contributions
 wounds-0.073
reating-0.071
pins-0.070
bars-0.070
 knee-0.068
TokenThey
Token ID2990
Feature activation+4.64e-02

Pos logit contributions
und+0.067
mon+0.055
nexpected+0.054
aded+0.053
ued+0.052
Neg logit contributions
 wounds-0.079
 knee-0.074
reating-0.074
pins-0.071
bars-0.069
Token said
Token ID531
Feature activation+6.30e-03

Pos logit contributions
und+0.064
+0.054
nexpected+0.048
aded+0.047
live+0.047
Neg logit contributions
 wounds-0.103
reating-0.095
pins-0.085
 knee-0.084
bars-0.082
Token,
Token ID11
Feature activation+1.28e-02

Pos logit contributions
und+0.013
+0.011
nexpected+0.011
aded+0.011
ued+0.011
Neg logit contributions
 wounds-0.009
reating-0.008
pins-0.008
 gloves-0.008
 knee-0.007
Token "
Token ID366
Feature activation-5.28e-02

Pos logit contributions
und+0.020
+0.018
nexpected+0.017
mon+0.016
aded+0.016
Neg logit contributions
 wounds-0.025
reating-0.023
pins-0.021
 gloves-0.020
 knee-0.020
TokenWe
Token ID1135
Feature activation-1.01e-02

Pos logit contributions
 wounds+0.099
reating+0.087
pins+0.086
 knee+0.083
 gloves+0.079
Neg logit contributions
und-0.092
-0.082
nexpected-0.076
aded-0.072
day-0.071
Token in
Token ID287
Feature activation+7.36e-02

Pos logit contributions
und+0.014
+0.012
nexpected+0.012
ued+0.012
aded+0.011
Neg logit contributions
 wounds-0.011
 gloves-0.010
 knee-0.010
reating-0.010
 jacket-0.009
Token the
Token ID262
Feature activation-2.72e-02

Pos logit contributions
und+0.164
+0.147
nexpected+0.144
aded+0.142
ued+0.141
Neg logit contributions
 wounds-0.096
reating-0.083
pins-0.081
 gloves-0.075
 knee-0.073
Token park
Token ID3952
Feature activation+4.04e-02

Pos logit contributions
 wounds+0.042
reating+0.037
pins+0.036
 gloves+0.035
 knee+0.034
Neg logit contributions
und-0.054
-0.047
nexpected-0.046
aded-0.045
ued-0.045
Token.
Token ID13
Feature activation-2.18e-02

Pos logit contributions
und+0.105
+0.094
nexpected+0.094
aded+0.091
ued+0.090
Neg logit contributions
 wounds-0.038
reating-0.031
pins-0.030
 knee-0.026
 gloves-0.026
Token They
Token ID1119
Feature activation+7.47e-02

Pos logit contributions
 wounds+0.033
pins+0.031
reating+0.030
 gloves+0.029
line+0.029
Neg logit contributions
und-0.042
nexpected-0.038
-0.037
ued-0.036
aded-0.035
Token liked
Token ID8288
Feature activation+4.46e-02

Pos logit contributions
und+0.100
+0.080
nexpected+0.076
aded+0.074
ued+0.072
Neg logit contributions
 wounds-0.168
reating-0.155
pins-0.146
 knee-0.143
 gloves-0.139
Token to
Token ID284
Feature activation+9.55e-03

Pos logit contributions
und+0.080
+0.067
nexpected+0.067
aded+0.066
mon+0.062
Neg logit contributions
 wounds-0.076
reating-0.067
pins-0.067
 knee-0.063
 decide-0.061
Token dig
Token ID3100
Feature activation-1.29e-02

Pos logit contributions
und+0.018
+0.017
aded+0.016
nexpected+0.015
ued+0.015
Neg logit contributions
 wounds-0.015
reating-0.014
pins-0.012
bars-0.012
 pilots-0.012
Token in
Token ID287
Feature activation+5.69e-02

Pos logit contributions
 wounds+0.018
reating+0.016
pins+0.015
 gloves+0.014
 knee+0.014
Neg logit contributions
und-0.027
-0.024
nexpected-0.024
aded-0.023
ued-0.023
Token the
Token ID262
Feature activation-4.18e-02

Pos logit contributions
und+0.100
+0.079
nexpected+0.079
aded+0.076
mon+0.075
Neg logit contributions
 wounds-0.097
reating-0.090
pins-0.089
 knee-0.084
 sax-0.083
Token sand
Token ID6450
Feature activation-7.21e-03

Pos logit contributions
 wounds+0.059
reating+0.055
pins+0.054
line+0.046
 knee+0.046
Neg logit contributions
und-0.088
-0.074
aded-0.074
nexpected-0.073
mon-0.072
Token,
Token ID11
Feature activation+1.97e-02

Pos logit contributions
 wounds+0.082
pins+0.079
reating+0.073
 gloves+0.070
line+0.069
Neg logit contributions
und-0.148
nexpected-0.137
-0.134
aded-0.132
ued-0.126
Token but
Token ID475
Feature activation+5.62e-02

Pos logit contributions
und+0.042
+0.037
nexpected+0.036
aded+0.035
ued+0.035
Neg logit contributions
 wounds-0.027
reating-0.024
pins-0.023
 knee-0.021
 sax-0.021
Token she
Token ID673
Feature activation+3.40e-02

Pos logit contributions
und+0.142
+0.123
aded+0.119
ued+0.116
nexpected+0.115
Neg logit contributions
 wounds-0.062
pins-0.056
 even-0.051
reating-0.051
 gloves-0.050
Token needed
Token ID2622
Feature activation-2.96e-02

Pos logit contributions
und+0.066
+0.055
ued+0.054
aded+0.051
nexpected+0.051
Neg logit contributions
 wounds-0.055
 knee-0.050
 even-0.049
 gloves-0.048
pins-0.047
Token something
Token ID1223
Feature activation-5.83e-03

Pos logit contributions
 wounds+0.045
reating+0.040
pins+0.039
 knee+0.036
 sax+0.036
Neg logit contributions
und-0.059
-0.052
nexpected-0.051
aded-0.049
ued-0.048
Token else
Token ID2073
Feature activation+5.78e-02

Pos logit contributions
 wounds+0.008
reating+0.007
pins+0.007
 gloves+0.007
 jacket+0.006
Neg logit contributions
und-0.013
-0.011
nexpected-0.011
aded-0.011
ued-0.010
Token.
Token ID13
Feature activation-3.64e-02

Pos logit contributions
und+0.145
+0.131
nexpected+0.128
aded+0.127
ued+0.124
Neg logit contributions
 wounds-0.062
pins-0.049
reating-0.048
 gloves-0.047
 jacket-0.045
Token She
Token ID1375
Feature activation+5.10e-02

Pos logit contributions
 wounds+0.049
pins+0.043
reating+0.042
 decide+0.042
 knee+0.041
Neg logit contributions
und-0.076
aded-0.066
nexpected-0.065
-0.065
ued-0.063
Token needed
Token ID2622
Feature activation-1.21e-02

Pos logit contributions
und+0.096
+0.087
ued+0.076
nexpected+0.075
aded+0.074
Neg logit contributions
 wounds-0.087
 gloves-0.077
 decide-0.074
 knee-0.073
 jacket-0.073
Token some
Token ID617
Feature activation-8.41e-03

Pos logit contributions
 wounds+0.017
reating+0.015
pins+0.015
 gloves+0.014
 knee+0.013
Neg logit contributions
und-0.026
-0.023
nexpected-0.022
aded-0.022
ued-0.021
Token coal
Token ID5655
Feature activation+2.17e-02

Pos logit contributions
 wounds+0.011
reating+0.010
pins+0.009
 gloves+0.009
 knee+0.009
Neg logit contributions
und-0.019
-0.017
nexpected-0.016
aded-0.016
ued-0.016
Token cooked
Token ID15847
Feature activation-9.41e-03

Pos logit contributions
 wounds+0.082
pins+0.073
 decide+0.071
 gloves+0.068
 knee+0.065
Neg logit contributions
und-0.154
-0.135
nexpected-0.134
ued-0.133
aded-0.133
Token for
Token ID329
Feature activation-1.02e-02

Pos logit contributions
 wounds+0.015
pins+0.013
reating+0.013
 sax+0.012
 knee+0.012
Neg logit contributions
und-0.018
nexpected-0.015
-0.015
aded-0.015
ued-0.014
Token you
Token ID345
Feature activation-4.23e-02

Pos logit contributions
 wounds+0.016
reating+0.014
pins+0.014
 knee+0.013
 sax+0.013
Neg logit contributions
und-0.020
-0.018
nexpected-0.017
aded-0.017
ued-0.016
Token and
Token ID290
Feature activation-9.43e-02

Pos logit contributions
 wounds+0.047
reating+0.039
pins+0.039
 gloves+0.035
 knee+0.034
Neg logit contributions
und-0.103
-0.092
nexpected-0.091
aded-0.089
ued-0.088
Token dad
Token ID9955
Feature activation-1.05e-01

Pos logit contributions
 wounds+0.130
reating+0.109
pins+0.109
 gloves+0.103
 knee+0.103
Neg logit contributions
und-0.207
-0.184
nexpected-0.178
ued-0.175
aded-0.175
Token and
Token ID290
Feature activation-9.94e-02

Pos logit contributions
 wounds+0.087
reating+0.070
pins+0.068
 gloves+0.060
 knee+0.059
Neg logit contributions
und-0.285
-0.260
nexpected-0.256
aded-0.255
ued-0.250
Token Sam
Token ID3409
Feature activation-7.42e-02

Pos logit contributions
 wounds+0.129
reating+0.111
pins+0.110
 gloves+0.099
 knee+0.099
Neg logit contributions
und-0.222
-0.197
nexpected-0.193
aded-0.189
ued-0.186
Token.
Token ID13
Feature activation-1.07e-01

Pos logit contributions
 wounds+0.052
pins+0.036
reating+0.035
 gloves+0.035
 knee+0.035
Neg logit contributions
und-0.206
-0.191
nexpected-0.188
aded-0.187
ued-0.181
Token I
Token ID314
Feature activation-5.51e-02

Pos logit contributions
 wounds+0.198
reating+0.178
pins+0.172
 gloves+0.162
 jacket+0.160
Neg logit contributions
und-0.182
-0.157
nexpected-0.147
aded-0.142
mon-0.140
Token wanted
Token ID2227
Feature activation-3.22e-02

Pos logit contributions
 wounds+0.071
 decide+0.064
pins+0.061
 even+0.058
 gloves+0.058
Neg logit contributions
und-0.121
-0.107
nexpected-0.104
ued-0.103
aded-0.102
Token to
Token ID284
Feature activation-2.81e-02

Pos logit contributions
 wounds+0.051
pins+0.045
reating+0.044
 knee+0.041
 gloves+0.041
Neg logit contributions
und-0.062
-0.054
nexpected-0.053
aded-0.052
ued-0.050
Token ostr
Token ID42387
Feature activation-5.18e-02

Pos logit contributions
und+0.025
nexpected+0.021
+0.021
aded+0.021
ued+0.020
Neg logit contributions
 wounds-0.020
pins-0.019
reating-0.019
line-0.017
 sax-0.016
Tokenich
Token ID488
Feature activation-4.89e-02

Pos logit contributions
 wounds+0.075
reating+0.066
pins+0.065
 gloves+0.060
 knee+0.059
Neg logit contributions
und-0.108
-0.095
nexpected-0.093
aded-0.091
ued-0.090
Token.
Token ID13
Feature activation-7.67e-02

Pos logit contributions
 wounds+0.058
reating+0.049
pins+0.048
 gloves+0.043
 knee+0.042
Neg logit contributions
und-0.115
-0.102
nexpected-0.100
aded-0.098
ued-0.098
Token<|endoftext|>
Token ID50256
Feature activation+1.22e-01

Pos logit contributions
 wounds+0.137
reating+0.123
pins+0.121
 gloves+0.114
 knee+0.113
Neg logit contributions
und-0.135
-0.116
nexpected-0.113
aded-0.110
ued-0.108
TokenL
Token ID43
Feature activation+3.70e-02

Pos logit contributions
und+0.168
nexpected+0.158
aded+0.148
ued+0.145
+0.137
Neg logit contributions
 wounds-0.228
reating-0.216
pins-0.211
 gloves-0.206
line-0.204
Tokenily
Token ID813
Feature activation+1.18e-02

Pos logit contributions
und+0.074
+0.065
nexpected+0.063
aded+0.062
ued+0.061
Neg logit contributions
 wounds-0.057
reating-0.050
pins-0.050
 gloves-0.046
 knee-0.046
Token and
Token ID290
Feature activation+6.30e-04

Pos logit contributions
und+0.025
+0.023
nexpected+0.022
aded+0.022
ued+0.022
Neg logit contributions
 wounds-0.016
pins-0.014
reating-0.014
 gloves-0.013
 jacket-0.013
Token Ben
Token ID3932
Feature activation+2.26e-02

Pos logit contributions
und+0.001
+0.001
nexpected+0.001
ued+0.001
aded+0.001
Neg logit contributions
 wounds-0.001
reating-0.001
pins-0.001
line-0.001
 gloves-0.001
Token were
Token ID547
Feature activation+9.16e-02

Pos logit contributions
und+0.039
+0.034
nexpected+0.033
aded+0.032
ued+0.031
Neg logit contributions
 wounds-0.040
reating-0.036
pins-0.036
 knee-0.033
 gloves-0.033
Token playing
Token ID2712
Feature activation+7.73e-03

Pos logit contributions
und+0.199
nexpected+0.182
+0.177
ued+0.169
aded+0.162
Neg logit contributions
 wounds-0.122
 knee-0.107
pins-0.101
reating-0.100
 gloves-0.099
Token in
Token ID287
Feature activation+7.78e-02

Pos logit contributions
und+0.015
+0.013
nexpected+0.013
ued+0.013
aded+0.012
Neg logit contributions
 wounds-0.012
 knee-0.010
 gloves-0.010
reating-0.010
pins-0.010
Token the
Token ID262
Feature activation-5.05e-02

Pos logit contributions
und+0.011
+0.009
nexpected+0.009
aded+0.009
ued+0.009
Neg logit contributions
 wounds-0.007
reating-0.006
pins-0.006
 gloves-0.005
 knee-0.005
Token kids
Token ID3988
Feature activation-2.70e-02

Pos logit contributions
 wounds+0.082
pins+0.074
reating+0.070
 knee+0.069
line+0.067
Neg logit contributions
und-0.092
aded-0.077
nexpected-0.076
day-0.076
-0.075
Token on
Token ID319
Feature activation+5.82e-02

Pos logit contributions
 wounds+0.039
reating+0.033
pins+0.031
 gloves+0.029
 knee+0.029
Neg logit contributions
und-0.057
-0.050
ued-0.049
aded-0.047
nexpected-0.047
Token board
Token ID3096
Feature activation-2.79e-02

Pos logit contributions
und+0.112
+0.097
nexpected+0.094
ued+0.091
aded+0.087
Neg logit contributions
 wounds-0.095
reating-0.082
pins-0.081
 knee-0.078
line-0.075
Token.
Token ID13
Feature activation-3.52e-02

Pos logit contributions
 wounds+0.028
reating+0.023
pins+0.022
 gloves+0.020
 knee+0.019
Neg logit contributions
und-0.071
-0.063
nexpected-0.062
ued-0.061
aded-0.061
Token The
Token ID383
Feature activation-2.98e-02

Pos logit contributions
 wounds+0.056
pins+0.051
reating+0.050
 gloves+0.047
line+0.047
Neg logit contributions
und-0.067
nexpected-0.058
-0.057
aded-0.056
ued-0.055
Token bus
Token ID1323
Feature activation-4.15e-02

Pos logit contributions
 wounds+0.045
reating+0.039
pins+0.035
 knee+0.034
rings+0.033
Neg logit contributions
und-0.053
time-0.049
 year-0.049
-0.048
ued-0.048
Token liked
Token ID8288
Feature activation+3.79e-02

Pos logit contributions
 wounds+0.077
pins+0.070
reating+0.070
 knee+0.066
 gloves+0.066
Neg logit contributions
und-0.068
-0.058
nexpected-0.056
aded-0.055
ued-0.053
Token to
Token ID284
Feature activation+9.96e-03

Pos logit contributions
und+0.072
+0.061
nexpected+0.060
aded+0.058
ued+0.058
Neg logit contributions
 wounds-0.062
reating-0.054
pins-0.054
 knee-0.051
 jacket-0.049
Token take
Token ID1011
Feature activation-8.68e-02

Pos logit contributions
und+0.016
ued+0.014
+0.013
date+0.013
time+0.013
Neg logit contributions
 wounds-0.018
reating-0.015
pins-0.015
rings-0.014
line-0.014
Token the
Token ID262
Feature activation-8.20e-02

Pos logit contributions
 wounds+0.137
reating+0.121
pins+0.120
 knee+0.111
 gloves+0.111
Neg logit contributions
und-0.170
-0.147
nexpected-0.145
aded-0.141
ued-0.140
Token the
Token ID262
Feature activation-1.18e-01

Pos logit contributions
 wounds+0.160
pins+0.154
reating+0.152
 sax+0.141
line+0.140
Neg logit contributions
und-0.176
-0.171
day-0.162
aded-0.158
 craving-0.158
Token best
Token ID1266
Feature activation+2.61e-02

Pos logit contributions
 wounds+0.198
pins+0.182
reating+0.180
line+0.167
 gloves+0.160
Neg logit contributions
und-0.211
-0.190
nexpected-0.181
aded-0.176
ued-0.172
Token original
Token ID2656
Feature activation+6.80e-03

Pos logit contributions
und+0.052
+0.046
nexpected+0.045
aded+0.044
ued+0.043
Neg logit contributions
 wounds-0.040
pins-0.035
reating-0.034
 gloves-0.031
line-0.031
Token recipe
Token ID8364
Feature activation-2.90e-02

Pos logit contributions
und+0.014
+0.012
nexpected+0.012
ued+0.012
aded+0.012
Neg logit contributions
 wounds-0.010
reating-0.009
pins-0.008
 gloves-0.008
 knee-0.008
Token.
Token ID13
Feature activation-7.26e-02

Pos logit contributions
 wounds+0.032
reating+0.027
pins+0.026
 gloves+0.023
 knee+0.023
Neg logit contributions
und-0.070
-0.063
nexpected-0.062
aded-0.061
ued-0.060
Token The
Token ID383
Feature activation-4.72e-02

Pos logit contributions
 wounds+0.161
reating+0.140
pins+0.139
 gloves+0.135
 knee+0.134
Neg logit contributions
und-0.105
-0.092
nexpected-0.077
 Luckily-0.077
mon-0.076
Token end
Token ID886
Feature activation-3.91e-02

Pos logit contributions
 wounds+0.055
pins+0.049
reating+0.048
line+0.045
 sax+0.041
Neg logit contributions
und-0.103
-0.101
nexpected-0.098
aded-0.095
ued-0.093
Token.
Token ID13
Feature activation-5.66e-02

Pos logit contributions
 wounds+0.047
pins+0.041
reating+0.039
line+0.036
 knee+0.035
Neg logit contributions
und-0.092
-0.082
nexpected-0.081
aded-0.077
time-0.076
Token<|endoftext|>
Token ID50256
Feature activation+1.32e-01

Pos logit contributions
 wounds+0.127
pins+0.111
reating+0.110
 gloves+0.108
 knee+0.107
Neg logit contributions
und-0.078
-0.069
nexpected-0.058
day-0.057
mon-0.056
TokenOnce
Token ID7454
Feature activation+2.67e-01

Pos logit contributions
und+0.253
nexpected+0.226
aded+0.221
+0.218
ued+0.216
Neg logit contributions
 wounds-0.196
reating-0.175
pins-0.173
 gloves-0.160
 knee-0.159
Token upon
Token ID2402
Feature activation+1.68e-01

Pos logit contributions
und+0.390
+0.327
nexpected+0.313
aded+0.306
ued+0.293
Neg logit contributions
 wounds-0.543
reating-0.509
pins-0.507
 knee-0.474
 gloves-0.470
Token ate
Token ID15063
Feature activation+1.97e-02

Pos logit contributions
und+0.071
+0.060
nexpected+0.058
aded+0.056
ued+0.054
Neg logit contributions
 wounds-0.093
reating-0.086
pins-0.083
 knee-0.079
 gloves-0.078
Token every
Token ID790
Feature activation-1.54e-02

Pos logit contributions
und+0.041
+0.036
nexpected+0.036
aded+0.035
ued+0.034
Neg logit contributions
 wounds-0.028
reating-0.025
pins-0.024
 gloves-0.022
 knee-0.022
Token bite
Token ID13197
Feature activation-7.37e-03

Pos logit contributions
 wounds+0.023
 knee+0.021
reating+0.020
 gloves+0.020
 jacket+0.020
Neg logit contributions
und-0.033
nexpected-0.026
aded-0.026
-0.025
ued-0.025
Token and
Token ID290
Feature activation-6.49e-02

Pos logit contributions
 wounds+0.009
reating+0.008
pins+0.008
 gloves+0.007
 knee+0.007
Neg logit contributions
und-0.017
-0.015
nexpected-0.015
aded-0.014
ued-0.014
Token said
Token ID531
Feature activation-2.38e-02

Pos logit contributions
 wounds+0.117
reating+0.104
pins+0.104
 gloves+0.098
 knee+0.097
Neg logit contributions
und-0.113
-0.097
nexpected-0.094
aded-0.092
ued-0.091
Token,
Token ID11
Feature activation-7.18e-03

Pos logit contributions
 wounds+0.044
pins+0.040
reating+0.039
 knee+0.038
 gloves+0.037
Neg logit contributions
und-0.041
-0.035
nexpected-0.033
aded-0.032
ued-0.031
Token "
Token ID366
Feature activation-6.64e-02

Pos logit contributions
 wounds+0.014
reating+0.012
pins+0.011
 knee+0.011
 gloves+0.011
Neg logit contributions
und-0.012
-0.011
nexpected-0.010
ued-0.009
mon-0.009
TokenThis
Token ID1212
Feature activation-4.46e-03

Pos logit contributions
 wounds+0.130
reating+0.115
pins+0.113
 knee+0.109
 gloves+0.106
Neg logit contributions
und-0.109
-0.096
nexpected-0.090
aded-0.086
ued-0.083
Token is
Token ID318
Feature activation-4.89e-02

Pos logit contributions
 wounds+0.006
 gloves+0.006
 decide+0.005
reating+0.005
 bike+0.005
Neg logit contributions
und-0.009
ued-0.008
uttered-0.008
aded-0.008
-0.008
Token the
Token ID262
Feature activation-1.05e-01

Pos logit contributions
 wounds+0.074
reating+0.065
pins+0.064
 gloves+0.060
 knee+0.058
Neg logit contributions
und-0.101
-0.088
nexpected-0.085
aded-0.084
ued-0.083
Token best
Token ID1266
Feature activation+4.67e-02

Pos logit contributions
 wounds+0.196
pins+0.178
reating+0.177
line+0.162
 sax+0.161
Neg logit contributions
und-0.169
-0.150
nexpected-0.142
aded-0.138
ued-0.134
Token your
Token ID534
Feature activation-1.19e-01

Pos logit contributions
 wounds+0.209
pins+0.181
line+0.177
reating+0.174
 sax+0.169
Neg logit contributions
und-0.192
nexpected-0.175
aded-0.162
-0.159
ued-0.153
Token colors
Token ID7577
Feature activation-8.63e-02

Pos logit contributions
 wounds+0.178
reating+0.162
pins+0.157
line+0.142
 gloves+0.140
Neg logit contributions
und-0.236
-0.215
nexpected-0.211
aded-0.202
ued-0.196
Token so
Token ID523
Feature activation+2.46e-02

Pos logit contributions
 wounds+0.158
reating+0.126
pins+0.125
 stations+0.123
 gloves+0.120
Neg logit contributions
und-0.151
-0.130
time-0.121
nexpected-0.120
mon-0.115
Token I
Token ID314
Feature activation-8.00e-02

Pos logit contributions
und+0.037
nexpected+0.034
+0.032
time+0.032
mon+0.031
Neg logit contributions
 wounds-0.049
reating-0.042
 gloves-0.041
pins-0.041
 stations-0.040
Token can
Token ID460
Feature activation-4.98e-02

Pos logit contributions
 wounds+0.136
reating+0.121
pins+0.119
 gloves+0.111
 knee+0.111
Neg logit contributions
und-0.147
-0.128
nexpected-0.124
aded-0.120
ued-0.119
Token draw
Token ID3197
Feature activation-1.43e-01

Pos logit contributions
 wounds+0.085
reating+0.069
pins+0.066
 stations+0.064
 gloves+0.063
Neg logit contributions
und-0.093
-0.090
nexpected-0.079
ued-0.076
aded-0.075
Token one
Token ID530
Feature activation-8.58e-02

Pos logit contributions
 wounds+0.274
pins+0.236
reating+0.222
 knee+0.220
 decide+0.215
Neg logit contributions
und-0.235
-0.192
nexpected-0.192
time-0.188
aded-0.178
Token too
Token ID1165
Feature activation-3.69e-02

Pos logit contributions
 wounds+0.127
pins+0.108
reating+0.108
 gloves+0.099
 knee+0.098
Neg logit contributions
und-0.177
-0.157
nexpected-0.152
aded-0.148
ued-0.147
Token?"
Token ID1701
Feature activation-6.19e-02

Pos logit contributions
 wounds+0.045
reating+0.041
pins+0.040
 gloves+0.036
 knee+0.035
Neg logit contributions
und-0.083
-0.074
nexpected-0.073
aded-0.072
ued-0.071
Token Mia
Token ID32189
Feature activation+2.07e-02

Pos logit contributions
 wounds+0.064
 knee+0.053
 say+0.050
 pants+0.050
 decide+0.049
Neg logit contributions
und-0.141
ued-0.132
aded-0.128
-0.126
itting-0.122
Token thought
Token ID1807
Feature activation+3.44e-02

Pos logit contributions
und+0.045
+0.040
aded+0.038
ued+0.037
nexpected+0.036
Neg logit contributions
 wounds-0.027
 decide-0.024
 knee-0.024
 even-0.023
 jacket-0.023
Token.
Token ID13
Feature activation-1.07e-01

Pos logit contributions
und+0.008
+0.007
nexpected+0.006
ued+0.006
mon+0.006
Neg logit contributions
 wounds-0.004
reating-0.003
pins-0.003
 sax-0.003
 gloves-0.003
Token I
Token ID314
Feature activation-5.55e-02

Pos logit contributions
 wounds+0.195
reating+0.177
pins+0.171
 gloves+0.162
 jacket+0.159
Neg logit contributions
und-0.186
-0.162
nexpected-0.154
ued-0.147
aded-0.147
Token will
Token ID481
Feature activation-5.37e-02

Pos logit contributions
 wounds+0.067
 decide+0.060
 say+0.054
pins+0.053
 fix+0.051
Neg logit contributions
und-0.123
-0.118
nexpected-0.114
aded-0.111
 Larry-0.111
Token warn
Token ID9828
Feature activation-8.83e-02

Pos logit contributions
 wounds+0.073
reating+0.064
pins+0.063
 gloves+0.057
 knee+0.056
Neg logit contributions
und-0.117
-0.104
nexpected-0.101
aded-0.098
ued-0.098
Token your
Token ID534
Feature activation-1.15e-01

Pos logit contributions
 wounds+0.165
pins+0.155
 gloves+0.149
 jacket+0.146
reating+0.145
Neg logit contributions
und-0.135
nexpected-0.116
-0.110
mon-0.108
ued-0.108
Token mom
Token ID1995
Feature activation-1.86e-01

Pos logit contributions
 wounds+0.152
reating+0.144
pins+0.138
 sax+0.129
line+0.126
Neg logit contributions
und-0.241
-0.216
nexpected-0.215
aded-0.210
mon-0.203
Token.
Token ID13
Feature activation-1.19e-01

Pos logit contributions
 wounds+0.223
reating+0.177
pins+0.176
 sax+0.165
 gloves+0.164
Neg logit contributions
und-0.436
-0.378
nexpected-0.369
mon-0.361
ued-0.359
Token I
Token ID314
Feature activation-6.48e-02

Pos logit contributions
 wounds+0.226
reating+0.210
pins+0.199
 gloves+0.190
 jacket+0.185
Neg logit contributions
und-0.198
-0.172
nexpected-0.159
 Luckily-0.153
ued-0.153
Token will
Token ID481
Feature activation-5.65e-02

Pos logit contributions
 wounds+0.085
 decide+0.073
pins+0.069
reating+0.064
 gloves+0.064
Neg logit contributions
und-0.141
-0.130
nexpected-0.128
aded-0.127
ued-0.122
Token call
Token ID869
Feature activation-9.52e-02

Pos logit contributions
 wounds+0.078
reating+0.069
pins+0.068
 gloves+0.062
line+0.061
Neg logit contributions
und-0.121
-0.108
nexpected-0.105
aded-0.101
ued-0.101
Token her
Token ID607
Feature activation-9.75e-02

Pos logit contributions
 wounds+0.176
pins+0.162
reating+0.156
 gloves+0.153
 stations+0.153
Neg logit contributions
und-0.133
nexpected-0.117
mon-0.108
time-0.106
ued-0.106
Token showed
Token ID3751
Feature activation-3.97e-02

Pos logit contributions
und+0.078
+0.068
nexpected+0.065
ued+0.063
aded+0.063
Neg logit contributions
 wounds-0.073
pins-0.066
 gloves-0.064
reating-0.064
 jacket-0.061
Token them
Token ID606
Feature activation-5.73e-02

Pos logit contributions
 wounds+0.057
pins+0.048
reating+0.048
 knee+0.045
 gloves+0.044
Neg logit contributions
und-0.084
nexpected-0.072
-0.072
aded-0.070
ued-0.070
Token how
Token ID703
Feature activation-6.51e-02

Pos logit contributions
 wounds+0.088
reating+0.077
pins+0.076
 knee+0.071
 gloves+0.070
Neg logit contributions
und-0.114
-0.098
nexpected-0.098
aded-0.095
ued-0.094
Token the
Token ID262
Feature activation-1.06e-01

Pos logit contributions
 wounds+0.113
reating+0.099
 knee+0.096
pins+0.095
 gloves+0.091
Neg logit contributions
und-0.114
nexpected-0.096
aded-0.095
-0.094
ued-0.092
Token workers
Token ID3259
Feature activation-8.54e-02

Pos logit contributions
 wounds+0.152
reating+0.136
pins+0.135
 knee+0.128
bars+0.120
Neg logit contributions
und-0.207
-0.175
aded-0.173
nexpected-0.172
ued-0.171
Token put
Token ID1234
Feature activation-1.17e-01

Pos logit contributions
 wounds+0.152
pins+0.143
reating+0.138
 gloves+0.132
line+0.131
Neg logit contributions
und-0.147
-0.126
nexpected-0.125
aded-0.119
ued-0.116
Token the
Token ID262
Feature activation-9.76e-02

Pos logit contributions
 wounds+0.171
reating+0.151
pins+0.145
 knee+0.139
 gloves+0.132
Neg logit contributions
und-0.239
-0.207
nexpected-0.202
aded-0.199
ued-0.196
Token parts
Token ID3354
Feature activation-8.92e-02

Pos logit contributions
 wounds+0.135
reating+0.123
pins+0.119
 knee+0.111
line+0.104
Neg logit contributions
und-0.201
-0.173
aded-0.171
nexpected-0.171
ued-0.169
Token together
Token ID1978
Feature activation+9.53e-02

Pos logit contributions
 wounds+0.171
reating+0.151
 knee+0.139
 sax+0.139
pins+0.139
Neg logit contributions
und-0.140
-0.115
ued-0.114
aded-0.111
nexpected-0.108
Token,
Token ID11
Feature activation-5.28e-03

Pos logit contributions
und+0.218
+0.194
nexpected+0.188
ued+0.185
aded+0.185
Neg logit contributions
 wounds-0.117
reating-0.099
pins-0.093
 knee-0.087
 gloves-0.084
Token how
Token ID703
Feature activation-6.12e-02

Pos logit contributions
 wounds+0.007
reating+0.006
pins+0.006
 knee+0.006
 jacket+0.005
Neg logit contributions
und-0.011
-0.010
nexpected-0.010
aded-0.009
ued-0.009
Token they
Token ID484
Feature activation+3.16e-02

Pos logit contributions
und+0.171
+0.151
nexpected+0.148
ued+0.148
aded+0.145
Neg logit contributions
 wounds-0.103
pins-0.090
reating-0.089
 gloves-0.087
 bike-0.083
Token didn
Token ID1422
Feature activation-1.53e-02

Pos logit contributions
und+0.059
nexpected+0.052
aded+0.051
+0.049
ued+0.048
Neg logit contributions
pins-0.049
 wounds-0.047
 decide-0.046
line-0.044
 jacket-0.044
Token't
Token ID470
Feature activation-1.91e-02

Pos logit contributions
 wounds+0.039
reating+0.036
pins+0.036
 gloves+0.034
 knee+0.034
Neg logit contributions
und-0.015
-0.012
nexpected-0.011
aded-0.010
ued-0.010
Token find
Token ID1064
Feature activation-1.11e-01

Pos logit contributions
 wounds+0.030
reating+0.025
 knee+0.022
 gloves+0.022
pins+0.022
Neg logit contributions
und-0.036
-0.034
ued-0.032
aded-0.030
day-0.030
Token the
Token ID262
Feature activation-9.92e-02

Pos logit contributions
 wounds+0.161
reating+0.142
pins+0.140
 knee+0.128
line+0.127
Neg logit contributions
und-0.229
-0.200
nexpected-0.197
aded-0.191
ued-0.191
Token shoes
Token ID10012
Feature activation-1.49e-01

Pos logit contributions
 wounds+0.154
reating+0.139
pins+0.130
 pilots+0.123
 knee+0.119
Neg logit contributions
und-0.165
-0.159
ued-0.147
 craving-0.146
nexpected-0.144
Token,
Token ID11
Feature activation-6.44e-03

Pos logit contributions
 wounds+0.242
pins+0.216
reating+0.216
 gloves+0.200
 knee+0.199
Neg logit contributions
und-0.281
-0.243
nexpected-0.241
aded-0.234
ued-0.228
Token they
Token ID484
Feature activation+3.55e-02

Pos logit contributions
 wounds+0.010
pins+0.009
reating+0.009
 gloves+0.008
 knee+0.008
Neg logit contributions
und-0.013
-0.011
nexpected-0.011
ued-0.011
aded-0.011
Token had
Token ID550
Feature activation+3.56e-02

Pos logit contributions
und+0.065
+0.056
nexpected+0.056
aded+0.053
ued+0.051
Neg logit contributions
 wounds-0.060
pins-0.057
 decide-0.053
 gloves-0.053
 jacket-0.052
Token fun
Token ID1257
Feature activation+6.93e-02

Pos logit contributions
und+0.072
+0.064
nexpected+0.062
aded+0.060
ued+0.060
Neg logit contributions
 wounds-0.054
pins-0.047
reating-0.047
 gloves-0.045
 jacket-0.043
Token looking
Token ID2045
Feature activation+5.20e-02

Pos logit contributions
und+0.145
+0.130
nexpected+0.127
aded+0.123
ued+0.122
Neg logit contributions
 wounds-0.099
reating-0.087
pins-0.085
 gloves-0.080
 knee-0.078
Token I
Token ID314
Feature activation-6.05e-02

Pos logit contributions
 wounds+0.199
reating+0.189
pins+0.176
line+0.162
 gloves+0.161
Neg logit contributions
und-0.190
-0.168
nexpected-0.162
 Luckily-0.155
mon-0.150
Token was
Token ID373
Feature activation-8.40e-03

Pos logit contributions
 wounds+0.104
 decide+0.092
pins+0.089
 knee+0.087
 jacket+0.087
Neg logit contributions
und-0.104
-0.093
aded-0.087
ued-0.087
nexpected-0.086
Token playing
Token ID2712
Feature activation-4.03e-02

Pos logit contributions
 wounds+0.014
reating+0.013
pins+0.013
 gloves+0.012
 knee+0.012
Neg logit contributions
und-0.015
-0.013
nexpected-0.013
aded-0.013
ued-0.012
Token with
Token ID351
Feature activation+1.56e-02

Pos logit contributions
 wounds+0.070
reating+0.067
pins+0.067
bars+0.062
rings+0.060
Neg logit contributions
und-0.069
-0.058
time-0.056
nexpected-0.054
aded-0.053
Token my
Token ID616
Feature activation-7.00e-02

Pos logit contributions
und+0.029
+0.025
nexpected+0.025
aded+0.024
ued+0.023
Neg logit contributions
 wounds-0.025
pins-0.022
reating-0.022
line-0.021
 decide-0.020
Token cars
Token ID5006
Feature activation-1.21e-01

Pos logit contributions
reating+0.097
 wounds+0.096
pins+0.091
line+0.088
bars+0.083
Neg logit contributions
und-0.129
-0.124
nexpected-0.114
aded-0.113
time-0.108
Token.
Token ID13
Feature activation-1.07e-01

Pos logit contributions
 wounds+0.166
reating+0.134
pins+0.132
 sax+0.121
 gloves+0.119
Neg logit contributions
und-0.263
-0.230
time-0.215
ued-0.208
day-0.206
Token I
Token ID314
Feature activation-5.83e-02

Pos logit contributions
 wounds+0.212
reating+0.198
pins+0.183
line+0.172
 decide+0.170
Neg logit contributions
und-0.169
-0.146
 Luckily-0.135
nexpected-0.133
 Larry-0.131
Token like
Token ID588
Feature activation+2.33e-02

Pos logit contributions
 wounds+0.092
 decide+0.082
pins+0.079
 gloves+0.077
 jacket+0.076
Neg logit contributions
und-0.110
-0.098
aded-0.093
nexpected-0.093
ued-0.092
Token them
Token ID606
Feature activation-6.66e-02

Pos logit contributions
und+0.037
nexpected+0.031
+0.031
aded+0.029
mon+0.029
Neg logit contributions
 wounds-0.042
pins-0.038
line-0.036
 decide-0.036
reating-0.036
Token a
Token ID257
Feature activation-5.45e-02

Pos logit contributions
 wounds+0.076
reating+0.063
pins+0.063
 gloves+0.054
 jacket+0.053
Neg logit contributions
und-0.158
-0.142
nexpected-0.137
ued-0.135
aded-0.134