TOP ACTIVATIONS
MAX = 0.259

She hid behind an ash tr ay . "
cigarette and using an ash tr ay . L
gave him a small ash tr ay as a gift .
the box and turned a dial . He looked through the
window sill with a tight lid . The
and accidentally dropped her ash tr ay on the ground .
in and found an ash tr ay . Lucy thought it
che ws the food and swall ows it .
bowl , it snapped its lid shut and opened again .
play with lamps and ash tr ays anymore .
hit her mom 's ash tr ay . The ash tr
cr un ches them and swall ows them . He makes
: Sharing is caring . Pride goes before a fall .
a copper pot with a lid . It was very old
They grabbed an old ash tr ay and put it in
his head and a spl int on his leg . They
her mom what an ash tr ay was , and her
he sp ied an ash tr ay . He picked up
noise in the bushes . Care fully , he moved closer
your fingers in the ash tr ay , so Jon had

MIN ACTIVATIONS
MIN = -0.218

" You are the best kids in the world . I
" You are the best kids . You are very smart
" They are such good kids . They deserve a nice
. You are the best kids and the best learners ."
" They are such good kids . They deserve a nice
Bob o loved to make kids laugh and play . He
M um continued :" Let 's start baking !"
L ily and Ben are kids who live next door .
L ily and Ben are kids who like to play outside
saw what happened . Those kids were mean . They took
She said , " Hi kids , I have a magic
Tim and Mia are adventurous kids . They like to explore
loved to make all the kids laugh . One day ,
to make fun of other kids . They saw a ch
" Hello , kids ," the truck driver said
<|endoftext|> Sam and Ben are kids who like to play with
ila and Tom were adventurous kids . They liked to explore
The giant loved children and often gave them everything
He wanted to make the kids happy . One
kids . You are good kids . You are smart .

INTERVAL 0.140 - 0.259
CONTAINS 0.877%

how to do it . Then , a
so it wouldn 't fall off again . From that day
a button bracelet , a button flower , and a button
Just try your best ." L ily took
a big hole in the top and puts a cone on

INTERVAL 0.021 - 0.140
CONTAINS 30.291%

to preschool . At preschool , she loved learning
, a truck and a book . She gives them hugs
animals too ." The rabbit thanked the fairy and went on
had a wonderful time . When it was time to go
the canvas was his dad 's . He did not know

INTERVAL -0.099 - 0.021
CONTAINS 64.990%

little girl named Jane . She loved to swim in the
the phone ." L ila and Sam follow mom
met a friendly monkey . The monkey said , " Hello
the zipper over and over and couldn 't get it to
their paws and claws to the flower . But they could

INTERVAL -0.218 - -0.099
CONTAINS 3.842%

Lily . She loved to play in her backyard every day
They smile . " Thank you , sheep cloud
are tools ." Tom nods . He says sorry
, there was a little boy who was very thirsty .
there lived a 3 year old called Lucy . One day
Token She
Token ID1375
Feature activation-5.82e-03

Pos logit contributions
ipop+0.019
ometer+0.018
 lid+0.017
cases+0.016
 off+0.015
Neg logit contributions
Playing-0.016
iling-0.015
Fighting-0.015
 sports-0.014
Leave-0.014
Token hid
Token ID24519
Feature activation-2.95e-03

Pos logit contributions
Playing+0.009
iling+0.009
Fighting+0.009
 years+0.008
Leave+0.008
Neg logit contributions
ometer-0.011
ipop-0.011
 lid-0.010
cases-0.010
ouse-0.009
Token behind
Token ID2157
Feature activation+2.75e-02

Pos logit contributions
Playing+0.005
iling+0.005
Fighting+0.005
 sports+0.005
hing+0.005
Neg logit contributions
ipop-0.005
ometer-0.005
 lid-0.005
cases-0.004
 off-0.004
Token an
Token ID281
Feature activation+1.24e-01

Pos logit contributions
 lid+0.040
ipop+0.039
ometer+0.038
 off+0.036
 eyes+0.034
Neg logit contributions
Playing-0.059
iling-0.058
Fighting-0.054
hing-0.051
 disl-0.051
Token ash
Token ID12530
Feature activation+1.06e-01

Pos logit contributions
ometer+0.208
ipop+0.205
 lid+0.201
 off+0.177
cases+0.169
Neg logit contributions
Playing-0.236
Fighting-0.218
iling-0.215
Leave-0.207
Why-0.198
Tokentr
Token ID2213
Feature activation+2.29e-01

Pos logit contributions
ipop+0.135
ometer+0.128
cases+0.111
 lid+0.106
ided+0.093
Neg logit contributions
iling-0.219
Playing-0.218
 town-0.211
Why-0.210
 years-0.208
Tokenay
Token ID323
Feature activation+1.47e-02

Pos logit contributions
ipop+0.506
ometer+0.497
 lid+0.463
cases+0.452
 prayer+0.440
Neg logit contributions
hing-0.224
 course-0.216
e-0.210
iling-0.208
room-0.201
Token.
Token ID13
Feature activation+2.15e-02

Pos logit contributions
ometer+0.030
ipop+0.029
 lid+0.027
cases+0.025
atively+0.023
Neg logit contributions
Playing-0.022
iling-0.022
Fighting-0.021
 owning-0.020
hing-0.020
Token
Token ID198
Feature activation-1.12e-02

Pos logit contributions
ipop+0.046
ometer+0.046
 lid+0.042
cases+0.040
ouse+0.038
Neg logit contributions
Playing-0.031
iling-0.028
Fighting-0.028
Why-0.026
 town-0.026
Token
Token ID198
Feature activation-1.77e-02

Pos logit contributions
Playing+0.020
iling+0.019
Fighting+0.018
Leave+0.017
 sports+0.017
Neg logit contributions
ometer-0.020
ipop-0.020
 lid-0.018
cases-0.017
atively-0.016
Token"
Token ID1
Feature activation-1.58e-01

Pos logit contributions
 owning+0.029
iling+0.028
Playing+0.026
 sports+0.026
 playing+0.025
Neg logit contributions
ometer-0.030
 lid-0.029
ipop-0.027
 off-0.026
atively-0.025
Token cigarette
Token ID17779
Feature activation+9.51e-02

Pos logit contributions
ometer+0.142
 lid+0.133
ipop+0.132
cases+0.107
 off+0.106
Neg logit contributions
iling-0.167
Playing-0.165
Fighting-0.151
Leave-0.147
 owning-0.146
Token and
Token ID290
Feature activation+9.43e-03

Pos logit contributions
ometer+0.160
ipop+0.153
 lid+0.144
cases+0.129
 off+0.125
Neg logit contributions
Playing-0.186
iling-0.179
Fighting-0.174
 sports-0.166
Leave-0.165
Token using
Token ID1262
Feature activation+1.86e-02

Pos logit contributions
ometer+0.016
ipop+0.016
 lid+0.015
 off+0.014
cases+0.013
Neg logit contributions
iling-0.017
Playing-0.016
Fighting-0.015
hing-0.015
 sports-0.015
Token an
Token ID281
Feature activation+1.24e-01

Pos logit contributions
ometer+0.032
ipop+0.032
 lid+0.029
cases+0.027
 off+0.025
Neg logit contributions
Playing-0.034
iling-0.033
Fighting-0.031
 sports-0.030
Leave-0.030
Token ash
Token ID12530
Feature activation+9.92e-02

Pos logit contributions
ipop+0.245
ometer+0.242
 lid+0.222
cases+0.210
atively+0.196
Neg logit contributions
Playing-0.193
iling-0.191
Fighting-0.179
 sports-0.170
 owning-0.170
Tokentr
Token ID2213
Feature activation+2.24e-01

Pos logit contributions
ipop+0.120
ometer+0.098
cases+0.096
 lid+0.093
atively+0.085
Neg logit contributions
Playing-0.205
Why-0.200
iling-0.197
hing-0.196
e-0.194
Tokenay
Token ID323
Feature activation+1.66e-02

Pos logit contributions
ipop+0.497
 lid+0.463
ometer+0.462
atively+0.449
 goodbye+0.437
Neg logit contributions
hing-0.229
Playing-0.213
iling-0.209
e-0.207
organized-0.204
Token.
Token ID13
Feature activation+2.10e-02

Pos logit contributions
ometer+0.031
ipop+0.031
 lid+0.029
cases+0.027
atively+0.026
Neg logit contributions
Playing-0.027
iling-0.026
Fighting-0.025
hing-0.025
e-0.024
Token
Token ID198
Feature activation-1.44e-02

Pos logit contributions
ipop+0.042
ometer+0.040
 lid+0.036
ided+0.035
cases+0.034
Neg logit contributions
Playing-0.033
Why-0.031
Fighting-0.030
 town-0.030
 sports-0.029
Token
Token ID198
Feature activation-2.08e-02

Pos logit contributions
Playing+0.019
 town+0.017
iling+0.017
Why+0.016
Fighting+0.016
Neg logit contributions
ipop-0.033
ometer-0.033
 lid-0.030
cases-0.029
atively-0.028
TokenL
Token ID43
Feature activation-7.24e-02

Pos logit contributions
iling+0.037
Playing+0.037
Fighting+0.035
hing+0.035
 owning+0.035
Neg logit contributions
ometer-0.035
ipop-0.035
 lid-0.033
cases-0.030
 off-0.028
Token gave
Token ID2921
Feature activation+7.63e-02

Pos logit contributions
ipop+0.111
ometer+0.107
 lid+0.100
cases+0.090
 off+0.087
Neg logit contributions
iling-0.150
Playing-0.148
Fighting-0.142
Leave-0.137
hing-0.135
Token him
Token ID683
Feature activation+2.08e-02

Pos logit contributions
ipop+0.172
ometer+0.171
 lid+0.161
 off+0.153
cases+0.147
Neg logit contributions
Playing-0.100
iling-0.097
Fighting-0.090
Leave-0.083
hing-0.081
Token a
Token ID257
Feature activation+1.11e-01

Pos logit contributions
ipop+0.035
ometer+0.035
 lid+0.032
cases+0.030
 off+0.028
Neg logit contributions
Playing-0.039
iling-0.038
Fighting-0.036
Leave-0.034
 owning-0.034
Token small
Token ID1402
Feature activation+1.50e-02

Pos logit contributions
ometer+0.208
ipop+0.203
cases+0.189
 lid+0.185
rily+0.177
Neg logit contributions
 sports-0.162
Playing-0.162
 town-0.154
Fighting-0.149
 owning-0.148
Token ash
Token ID12530
Feature activation+1.03e-01

Pos logit contributions
ometer+0.024
ipop+0.023
 lid+0.021
cases+0.020
atively+0.018
Neg logit contributions
Playing-0.028
iling-0.027
 sports-0.027
Fighting-0.027
 owning-0.026
Tokentr
Token ID2213
Feature activation+2.22e-01

Pos logit contributions
ipop+0.108
ometer+0.101
 lid+0.092
cases+0.085
atively+0.070
Neg logit contributions
Playing-0.230
iling-0.221
hing-0.220
 sports-0.219
e-0.217
Tokenay
Token ID323
Feature activation+1.56e-02

Pos logit contributions
ipop+0.470
 lid+0.455
ometer+0.453
atively+0.412
 prayer+0.411
Neg logit contributions
hing-0.235
Playing-0.226
 repairing-0.223
e-0.217
iling-0.212
Token as
Token ID355
Feature activation+8.32e-02

Pos logit contributions
ometer+0.028
ipop+0.028
 lid+0.025
cases+0.023
 off+0.022
Neg logit contributions
Playing-0.028
iling-0.028
Fighting-0.026
Leave-0.025
 sports-0.025
Token a
Token ID257
Feature activation+1.21e-01

Pos logit contributions
ometer+0.140
ipop+0.138
 lid+0.125
cases+0.115
 off+0.107
Neg logit contributions
Playing-0.156
iling-0.153
Fighting-0.146
Leave-0.139
hing-0.138
Token gift
Token ID6979
Feature activation+1.10e-01

Pos logit contributions
ometer+0.189
ipop+0.188
cases+0.166
 lid+0.166
atively+0.148
Neg logit contributions
Playing-0.226
 sports-0.212
Fighting-0.210
 owning-0.208
iling-0.207
Token.
Token ID13
Feature activation+2.53e-02

Pos logit contributions
ometer+0.188
ipop+0.186
 lid+0.169
cases+0.155
atively+0.141
Neg logit contributions
Playing-0.203
iling-0.198
Fighting-0.189
 sports-0.182
 owning-0.181
Token the
Token ID262
Feature activation+3.69e-03

Pos logit contributions
Playing+0.058
iling+0.056
Fighting+0.054
Leave+0.052
hing+0.051
Neg logit contributions
ometer-0.051
ipop-0.050
 lid-0.045
cases-0.042
 off-0.042
Token box
Token ID3091
Feature activation+9.44e-02

Pos logit contributions
ometer+0.007
ipop+0.007
 lid+0.006
cases+0.005
 off+0.005
Neg logit contributions
Playing-0.007
iling-0.007
Fighting-0.006
hing-0.006
Leave-0.006
Token and
Token ID290
Feature activation+3.39e-02

Pos logit contributions
ometer+0.168
ipop+0.166
 lid+0.152
cases+0.140
 off+0.129
Neg logit contributions
Playing-0.167
iling-0.163
Fighting-0.155
 sports-0.148
Leave-0.148
Token turned
Token ID2900
Feature activation+1.53e-01

Pos logit contributions
ometer+0.060
ipop+0.059
 lid+0.054
cases+0.050
 off+0.047
Neg logit contributions
Playing-0.061
iling-0.059
Fighting-0.057
 sports-0.054
Leave-0.054
Token a
Token ID257
Feature activation+1.25e-01

Pos logit contributions
ipop+0.270
ometer+0.270
 lid+0.235
cases+0.221
atively+0.205
Neg logit contributions
Playing-0.270
iling-0.257
Fighting-0.253
 sports-0.243
hing-0.241
Token dial
Token ID5980
Feature activation+2.20e-01

Pos logit contributions
ipop+0.211
ometer+0.209
cases+0.180
 lid+0.178
atively+0.162
Neg logit contributions
Playing-0.225
iling-0.216
Fighting-0.213
 sports-0.205
 owning-0.204
Token.
Token ID13
Feature activation+3.24e-02

Pos logit contributions
ometer+0.363
ipop+0.362
 lid+0.329
cases+0.289
atively+0.271
Neg logit contributions
Playing-0.410
iling-0.396
Fighting-0.382
hing-0.371
 owning-0.365
Token He
Token ID679
Feature activation+4.53e-03

Pos logit contributions
ometer+0.057
ipop+0.056
 lid+0.052
cases+0.048
atively+0.044
Neg logit contributions
Playing-0.059
Fighting-0.054
iling-0.054
 sports-0.052
Why-0.051
Token looked
Token ID3114
Feature activation+1.41e-01

Pos logit contributions
ometer+0.008
ipop+0.008
 lid+0.007
cases+0.006
 off+0.006
Neg logit contributions
Playing-0.008
iling-0.008
Fighting-0.008
 sports-0.007
Leave-0.007
Token through
Token ID832
Feature activation+4.55e-03

Pos logit contributions
ometer+0.270
ipop+0.264
 lid+0.240
cases+0.223
atively+0.205
Neg logit contributions
Playing-0.234
iling-0.224
Fighting-0.214
 sports-0.209
 owning-0.206
Token the
Token ID262
Feature activation-1.40e-02

Pos logit contributions
 lid+0.008
ometer+0.007
ipop+0.007
 off+0.007
 eyes+0.007
Neg logit contributions
Playing-0.009
iling-0.008
Fighting-0.008
hing-0.008
Leave-0.007
Token window
Token ID4324
Feature activation+1.04e-01

Pos logit contributions
Playing+0.006
iling+0.006
Fighting+0.005
hing+0.005
Leave+0.005
Neg logit contributions
ometer-0.004
ipop-0.004
 lid-0.004
cases-0.003
 off-0.003
Token sill
Token ID49276
Feature activation+1.37e-01

Pos logit contributions
ometer+0.187
ipop+0.185
 lid+0.169
cases+0.154
atively+0.141
Neg logit contributions
Playing-0.174
iling-0.168
Fighting-0.161
 years-0.157
 sports-0.156
Token with
Token ID351
Feature activation+7.11e-02

Pos logit contributions
ometer+0.263
ipop+0.241
 lid+0.230
cases+0.210
ouse+0.190
Neg logit contributions
iling-0.220
Playing-0.214
e-0.203
hing-0.202
 years-0.202
Token a
Token ID257
Feature activation+1.15e-01

Pos logit contributions
ipop+0.116
ometer+0.116
 lid+0.112
 off+0.099
cases+0.096
Neg logit contributions
Playing-0.136
iling-0.132
Fighting-0.125
 sports-0.118
 owning-0.117
Token tight
Token ID5381
Feature activation+1.51e-01

Pos logit contributions
ometer+0.215
ipop+0.212
cases+0.190
 lid+0.177
atively+0.170
Neg logit contributions
Playing-0.179
Fighting-0.171
iling-0.168
Leave-0.167
 town-0.165
Token lid
Token ID19789
Feature activation+2.18e-01

Pos logit contributions
ometer+0.252
ipop+0.245
cases+0.221
 lid+0.205
atively+0.184
Neg logit contributions
iling-0.260
Playing-0.257
e-0.246
 town-0.246
Leave-0.245
Token.
Token ID13
Feature activation+2.01e-02

Pos logit contributions
ometer+0.421
ipop+0.412
 lid+0.375
cases+0.344
ouse+0.321
Neg logit contributions
Playing-0.340
iling-0.326
hing-0.309
Fighting-0.307
 owning-0.300
Token
Token ID220
Feature activation+1.58e-01

Pos logit contributions
ipop+0.033
ometer+0.033
 lid+0.029
cases+0.027
ouse+0.025
Neg logit contributions
Playing-0.038
Fighting-0.035
iling-0.035
 sports-0.034
 town-0.034
Token
Token ID198
Feature activation-8.58e-03

Pos logit contributions
ipop+0.277
ometer+0.276
 lid+0.252
cases+0.231
atively+0.213
Neg logit contributions
Playing-0.284
iling-0.272
Fighting-0.263
 owning-0.249
 sports-0.249
Token
Token ID198
Feature activation-1.54e-02

Pos logit contributions
iling+0.018
Playing+0.018
Fighting+0.017
 owning+0.016
 sports+0.016
Neg logit contributions
ometer-0.012
ipop-0.012
 lid-0.011
cases-0.010
 off-0.009
TokenThe
Token ID464
Feature activation-2.82e-02

Pos logit contributions
iling+0.024
 owning+0.023
Playing+0.022
 playing+0.022
 years+0.021
Neg logit contributions
 lid-0.027
ometer-0.026
ipop-0.026
cases-0.024
 off-0.024
Token and
Token ID290
Feature activation+1.76e-02

Pos logit contributions
ometer+0.172
ipop+0.169
 lid+0.155
cases+0.143
 off+0.133
Neg logit contributions
Playing-0.177
iling-0.172
Fighting-0.165
 sports-0.157
Leave-0.157
Token accidentally
Token ID14716
Feature activation+4.67e-02

Pos logit contributions
ometer+0.033
ipop+0.033
 lid+0.031
cases+0.028
 off+0.027
Neg logit contributions
Playing-0.029
iling-0.029
Fighting-0.027
hing-0.025
 sports-0.025
Token dropped
Token ID5710
Feature activation+4.95e-02

Pos logit contributions
ometer+0.081
ipop+0.080
 lid+0.072
cases+0.068
atively+0.062
Neg logit contributions
Playing-0.083
iling-0.080
Fighting-0.076
e-0.075
Leave-0.075
Token her
Token ID607
Feature activation+6.64e-02

Pos logit contributions
ometer+0.092
ipop+0.091
 lid+0.084
cases+0.078
 off+0.072
Neg logit contributions
Playing-0.084
iling-0.081
Fighting-0.078
 sports-0.074
Leave-0.074
Token ash
Token ID12530
Feature activation+8.50e-02

Pos logit contributions
ipop+0.111
ometer+0.107
 lid+0.098
cases+0.096
atively+0.086
Neg logit contributions
Playing-0.111
iling-0.110
e-0.106
 years-0.103
 playing-0.103
Tokentr
Token ID2213
Feature activation+2.17e-01

Pos logit contributions
ipop+0.112
ometer+0.093
cases+0.087
ae+0.082
 lid+0.080
Neg logit contributions
Why-0.160
ting-0.160
iling-0.159
e-0.158
hing-0.156
Tokenay
Token ID323
Feature activation-5.04e-03

Pos logit contributions
ipop+0.491
ometer+0.464
 lid+0.453
 prayer+0.452
atively+0.451
Neg logit contributions
hing-0.179
e-0.172
 repairing-0.168
organized-0.159
opl-0.157
Token on
Token ID319
Feature activation+8.27e-02

Pos logit contributions
Playing+0.009
iling+0.009
Fighting+0.008
hing+0.008
 sports+0.008
Neg logit contributions
ometer-0.009
ipop-0.009
 lid-0.008
cases-0.007
atively-0.007
Token the
Token ID262
Feature activation-9.69e-03

Pos logit contributions
ometer+0.157
 lid+0.153
ipop+0.151
 off+0.141
cases+0.128
Neg logit contributions
Playing-0.143
iling-0.133
Fighting-0.127
Leave-0.118
hing-0.116
Token ground
Token ID2323
Feature activation+5.07e-02

Pos logit contributions
Playing+0.020
iling+0.019
Fighting+0.018
 owning+0.016
Leave+0.016
Neg logit contributions
ometer-0.017
 lid-0.017
ipop-0.015
 off-0.014
 door-0.014
Token.
Token ID13
Feature activation+1.70e-02

Pos logit contributions
ometer+0.088
ipop+0.087
 lid+0.079
cases+0.073
atively+0.067
Neg logit contributions
Playing-0.092
iling-0.090
Fighting-0.086
 sports-0.082
Leave-0.082
Token in
Token ID287
Feature activation+1.01e-02

Pos logit contributions
ometer+0.255
ipop+0.252
 lid+0.229
cases+0.215
atively+0.198
Neg logit contributions
Playing-0.221
iling-0.214
Fighting-0.203
 sports-0.195
 owning-0.195
Token and
Token ID290
Feature activation-2.37e-03

Pos logit contributions
 lid+0.019
ipop+0.018
ometer+0.017
 off+0.016
 eyes+0.016
Neg logit contributions
Playing-0.017
iling-0.016
Fighting-0.016
hing-0.015
e-0.014
Token found
Token ID1043
Feature activation-1.37e-03

Pos logit contributions
Playing+0.003
iling+0.003
Fighting+0.003
 sports+0.003
 owning+0.003
Neg logit contributions
ometer-0.005
ipop-0.005
 lid-0.005
cases-0.004
atively-0.004
Token an
Token ID281
Feature activation+1.07e-01

Pos logit contributions
Playing+0.002
Fighting+0.002
 owning+0.002
iling+0.002
Why+0.002
Neg logit contributions
ipop-0.003
ometer-0.003
cases-0.003
 lid-0.002
ouse-0.002
Token ash
Token ID12530
Feature activation+8.42e-02

Pos logit contributions
ometer+0.207
ipop+0.203
 lid+0.191
cases+0.173
 off+0.165
Neg logit contributions
Playing-0.176
iling-0.168
Fighting-0.162
Leave-0.153
 sports-0.151
Tokentr
Token ID2213
Feature activation+2.16e-01

Pos logit contributions
ipop+0.111
ometer+0.096
cases+0.090
 reluctantly+0.080
ided+0.078
Neg logit contributions
iling-0.167
Playing-0.167
 town-0.166
Why-0.164
 playing-0.162
Tokenay
Token ID323
Feature activation+6.88e-04

Pos logit contributions
ipop+0.473
ometer+0.464
 lid+0.430
 prayer+0.422
 goodbye+0.411
Neg logit contributions
hing-0.216
iling-0.214
Playing-0.204
e-0.197
 course-0.195
Token.
Token ID13
Feature activation+1.61e-02

Pos logit contributions
ometer+0.001
ipop+0.001
 lid+0.001
cases+0.001
atively+0.001
Neg logit contributions
Playing-0.001
iling-0.001
Fighting-0.001
 owning-0.001
hing-0.001
Token Lucy
Token ID22162
Feature activation-1.54e-03

Pos logit contributions
ipop+0.029
ometer+0.029
 lid+0.026
cases+0.024
atively+0.022
Neg logit contributions
Playing-0.028
iling-0.027
Fighting-0.026
 owning-0.025
 sports-0.025
Token thought
Token ID1807
Feature activation+1.37e-01

Pos logit contributions
iling+0.003
Fighting+0.003
Playing+0.003
 sports+0.003
Leave+0.003
Neg logit contributions
ometer-0.003
ipop-0.003
 lid-0.002
cases-0.002
 mouth-0.002
Token it
Token ID340
Feature activation+9.87e-02

Pos logit contributions
ometer+0.228
ipop+0.224
 lid+0.202
cases+0.187
atively+0.170
Neg logit contributions
Playing-0.262
iling-0.255
Fighting-0.244
 owning-0.236
 sports-0.235
Token che
Token ID1125
Feature activation+1.44e-02

Pos logit contributions
ometer+0.033
ipop+0.032
 lid+0.028
cases+0.027
ouse+0.026
Neg logit contributions
Playing-0.030
iling-0.030
Leave-0.028
Fighting-0.028
Why-0.028
Tokenws
Token ID18504
Feature activation+1.04e-01

Pos logit contributions
ometer+0.023
ipop+0.022
 lid+0.020
cases+0.019
atively+0.017
Neg logit contributions
iling-0.029
Playing-0.028
e-0.026
Fighting-0.026
hing-0.026
Token the
Token ID262
Feature activation-1.25e-02

Pos logit contributions
ometer+0.179
ipop+0.176
 lid+0.166
 off+0.147
cases+0.145
Neg logit contributions
Playing-0.192
iling-0.188
Fighting-0.180
Leave-0.172
 sports-0.171
Token food
Token ID2057
Feature activation+1.89e-03

Pos logit contributions
Playing+0.025
iling+0.024
Fighting+0.023
hing+0.022
Leave+0.021
Neg logit contributions
ipop-0.021
ometer-0.021
 lid-0.020
cases-0.017
 off-0.016
Token and
Token ID290
Feature activation+2.42e-02

Pos logit contributions
ometer+0.003
ipop+0.003
 lid+0.003
cases+0.003
 off+0.003
Neg logit contributions
Playing-0.004
iling-0.003
Fighting-0.003
Leave-0.003
 sports-0.003
Token swall
Token ID20301
Feature activation+2.16e-01

Pos logit contributions
ipop+0.042
ometer+0.042
 lid+0.039
cases+0.035
 off+0.034
Neg logit contributions
Playing-0.044
iling-0.042
Fighting-0.041
hing-0.039
 sports-0.039
Tokenows
Token ID1666
Feature activation+1.14e-01

Pos logit contributions
ipop+0.265
ometer+0.188
 lid+0.170
cases+0.146
ouse+0.144
Neg logit contributions
iling-0.460
Playing-0.453
hing-0.449
 Playing-0.438
 sports-0.437
Token it
Token ID340
Feature activation+1.51e-01

Pos logit contributions
ometer+0.208
ipop+0.208
 lid+0.187
cases+0.176
atively+0.161
Neg logit contributions
Playing-0.190
iling-0.185
Fighting-0.179
 sports-0.171
hing-0.170
Token.
Token ID13
Feature activation+2.30e-02

Pos logit contributions
ipop+0.310
ometer+0.286
 lid+0.275
cases+0.265
ouse+0.252
Neg logit contributions
iling-0.194
Playing-0.192
e-0.190
Why-0.187
 town-0.179
Token
Token ID198
Feature activation-7.19e-03

Pos logit contributions
ipop+0.044
ometer+0.043
 lid+0.039
cases+0.037
atively+0.034
Neg logit contributions
Playing-0.039
Fighting-0.036
iling-0.036
Leave-0.034
 owning-0.034
Token
Token ID198
Feature activation-1.45e-02

Pos logit contributions
iling+0.014
Playing+0.014
Fighting+0.014
 sports+0.013
Leave+0.013
Neg logit contributions
ometer-0.011
ipop-0.011
 lid-0.010
cases-0.009
 off-0.008
Token bowl
Token ID9396
Feature activation+1.09e-01

Pos logit contributions
ometer+0.165
ipop+0.162
 lid+0.144
cases+0.138
ouse+0.131
Neg logit contributions
Playing-0.078
iling-0.073
 years-0.072
 playing-0.070
 town-0.070
Token,
Token ID11
Feature activation-1.07e-02

Pos logit contributions
ometer+0.226
ipop+0.217
 lid+0.202
cases+0.195
atively+0.177
Neg logit contributions
Playing-0.150
iling-0.149
e-0.135
Fighting-0.134
 sports-0.133
Token it
Token ID340
Feature activation+1.12e-01

Pos logit contributions
Playing+0.020
iling+0.020
Fighting+0.019
hing+0.018
 owning+0.018
Neg logit contributions
ometer-0.018
ipop-0.018
 lid-0.016
cases-0.015
 off-0.014
Token snapped
Token ID20821
Feature activation+1.39e-01

Pos logit contributions
ometer+0.192
ipop+0.190
 lid+0.173
cases+0.159
 off+0.146
Neg logit contributions
Playing-0.203
iling-0.198
Fighting-0.189
 sports-0.181
Leave-0.180
Token its
Token ID663
Feature activation+1.41e-01

Pos logit contributions
ometer+0.260
ipop+0.257
 lid+0.230
cases+0.219
atively+0.198
Neg logit contributions
Playing-0.230
iling-0.227
Fighting-0.214
 sports-0.208
 years-0.206
Token lid
Token ID19789
Feature activation+2.15e-01

Pos logit contributions
ometer+0.194
ipop+0.193
cases+0.164
 lid+0.161
atively+0.139
Neg logit contributions
iling-0.286
Playing-0.282
 sports-0.269
e-0.268
 years-0.268
Token shut
Token ID4423
Feature activation+1.54e-01

Pos logit contributions
ometer+0.432
ipop+0.419
 lid+0.377
cases+0.350
ouse+0.330
Neg logit contributions
iling-0.322
Playing-0.320
hing-0.294
Fighting-0.294
e-0.288
Token and
Token ID290
Feature activation+4.20e-02

Pos logit contributions
ometer+0.269
ipop+0.268
 lid+0.240
cases+0.225
atively+0.203
Neg logit contributions
Playing-0.273
iling-0.270
Fighting-0.257
 sports-0.247
 years-0.245
Token opened
Token ID4721
Feature activation+1.53e-01

Pos logit contributions
ometer+0.074
ipop+0.073
 lid+0.066
cases+0.061
atively+0.056
Neg logit contributions
Playing-0.076
iling-0.073
Fighting-0.070
Leave-0.067
 owning-0.067
Token again
Token ID757
Feature activation+6.61e-02

Pos logit contributions
ipop+0.349
ometer+0.344
cases+0.306
 lid+0.301
ouse+0.288
Neg logit contributions
Playing-0.183
iling-0.182
Fighting-0.175
 town-0.173
Why-0.172
Token.
Token ID13
Feature activation+1.00e-02

Pos logit contributions
ometer+0.101
ipop+0.100
 lid+0.092
cases+0.081
 off+0.081
Neg logit contributions
Playing-0.135
iling-0.129
Fighting-0.127
Leave-0.122
hing-0.121
Token play
Token ID711
Feature activation-1.03e-01

Pos logit contributions
ometer+0.121
ipop+0.115
 lid+0.107
 off+0.100
cases+0.091
Neg logit contributions
Playing-0.143
Fighting-0.135
iling-0.134
 sports-0.129
 owning-0.129
Token with
Token ID351
Feature activation+3.69e-02

Pos logit contributions
iling+0.227
Playing+0.225
years+0.214
Fighting+0.213
Leave+0.203
Neg logit contributions
ometer-0.136
 lid-0.129
 off-0.119
ipop-0.108
 mouth-0.101
Token lamps
Token ID32209
Feature activation+5.30e-03

Pos logit contributions
ometer+0.057
ipop+0.053
 lid+0.052
 mouth+0.046
 off+0.045
Neg logit contributions
iling-0.076
Playing-0.076
Fighting-0.074
hing-0.072
Leave-0.072
Token and
Token ID290
Feature activation+2.14e-02

Pos logit contributions
ometer+0.008
ipop+0.008
cases+0.007
 lid+0.007
 off+0.006
Neg logit contributions
Playing-0.011
iling-0.010
Fighting-0.010
Leave-0.010
Now-0.010
Token ash
Token ID12530
Feature activation+7.51e-02

Pos logit contributions
ometer+0.036
ipop+0.035
 lid+0.034
cases+0.031
 off+0.031
Neg logit contributions
iling-0.040
Playing-0.038
Fighting-0.035
hing-0.035
Now-0.034
Tokentr
Token ID2213
Feature activation+2.14e-01

Pos logit contributions
ometer+0.076
ipop+0.075
 lid+0.063
cases+0.054
 off+0.046
Neg logit contributions
Playing-0.191
iling-0.187
Fighting-0.182
 sports-0.176
Leave-0.176
Tokenays
Token ID592
Feature activation+1.66e-01

Pos logit contributions
ipop+0.408
ometer+0.385
 lid+0.377
atively+0.331
 prayer+0.318
Neg logit contributions
Playing-0.327
iling-0.323
hing-0.313
Fighting-0.295
organized-0.288
Token anymore
Token ID7471
Feature activation-3.58e-02

Pos logit contributions
ipop+0.240
ometer+0.236
 lid+0.219
 off+0.203
cases+0.187
Neg logit contributions
Playing-0.349
iling-0.340
Fighting-0.332
Leave-0.318
 disl-0.309
Token.
Token ID13
Feature activation+2.83e-02

Pos logit contributions
iling+0.079
Playing+0.077
Fighting+0.076
Leave+0.075
hing+0.075
Neg logit contributions
ipop-0.048
ometer-0.048
 lid-0.040
 off-0.035
cases-0.035
Token
Token ID198
Feature activation+3.75e-03

Pos logit contributions
ipop+0.061
ometer+0.057
 lid+0.056
cases+0.053
atively+0.051
Neg logit contributions
Playing-0.045
Why-0.039
Fighting-0.039
 town-0.038
 Playing-0.037
Token
Token ID198
Feature activation-2.87e-03

Pos logit contributions
ometer+0.007
ipop+0.007
 lid+0.006
cases+0.006
atively+0.005
Neg logit contributions
Playing-0.007
iling-0.006
Fighting-0.006
 sports-0.006
 owning-0.006
Token hit
Token ID2277
Feature activation+2.92e-02

Pos logit contributions
ometer+0.069
ipop+0.068
 lid+0.062
cases+0.057
atively+0.052
Neg logit contributions
Playing-0.074
iling-0.072
Fighting-0.068
Leave-0.066
 sports-0.066
Token her
Token ID607
Feature activation+6.79e-02

Pos logit contributions
ometer+0.053
ipop+0.053
 lid+0.049
 off+0.044
cases+0.044
Neg logit contributions
Playing-0.051
iling-0.049
Fighting-0.047
hing-0.044
Leave-0.044
Token mom
Token ID1995
Feature activation+1.00e-02

Pos logit contributions
ometer+0.109
ipop+0.108
 lid+0.097
cases+0.090
atively+0.081
Neg logit contributions
Playing-0.131
iling-0.128
Fighting-0.122
 sports-0.118
Leave-0.117
Token's
Token ID338
Feature activation+8.90e-02

Pos logit contributions
ipop+0.015
ometer+0.015
 off+0.014
 lid+0.013
cases+0.012
Neg logit contributions
Playing-0.021
iling-0.020
Fighting-0.020
 sports-0.019
 disl-0.018
Token ash
Token ID12530
Feature activation+8.58e-02

Pos logit contributions
ometer+0.122
ipop+0.118
 lid+0.112
 off+0.089
cases+0.089
Neg logit contributions
Playing-0.204
iling-0.197
Fighting-0.189
hing-0.181
 owning-0.179
Tokentr
Token ID2213
Feature activation+2.13e-01

Pos logit contributions
ipop+0.106
ometer+0.106
 lid+0.089
cases+0.085
atively+0.072
Neg logit contributions
Playing-0.184
iling-0.184
Fighting-0.173
Why-0.173
hing-0.173
Tokenay
Token ID323
Feature activation-3.00e-03

Pos logit contributions
ipop+0.465
ometer+0.450
 lid+0.430
 prayer+0.419
atively+0.408
Neg logit contributions
hing-0.211
e-0.205
iling-0.203
Playing-0.195
 course-0.194
Token.
Token ID13
Feature activation+1.28e-02

Pos logit contributions
Playing+0.006
iling+0.005
Fighting+0.005
Leave+0.005
 sports+0.005
Neg logit contributions
ometer-0.005
ipop-0.005
 lid-0.005
cases-0.004
 off-0.004
Token The
Token ID383
Feature activation-2.65e-02

Pos logit contributions
ipop+0.023
ometer+0.023
 lid+0.021
cases+0.019
atively+0.018
Neg logit contributions
Playing-0.023
Fighting-0.021
iling-0.021
Why-0.020
Leave-0.020
Token ash
Token ID12530
Feature activation+4.64e-02

Pos logit contributions
Playing+0.049
iling+0.049
Fighting+0.046
hing+0.044
Leave+0.043
Neg logit contributions
ipop-0.046
ometer-0.046
 lid-0.044
cases-0.037
 off-0.035
Tokentr
Token ID2213
Feature activation+1.85e-01

Pos logit contributions
ipop+0.056
ometer+0.053
 lid+0.045
cases+0.043
atively+0.036
Neg logit contributions
Playing-0.102
iling-0.097
Leave-0.096
Fighting-0.095
Why-0.095
Token cr
Token ID1067
Feature activation+1.05e-01

Pos logit contributions
ometer+0.012
ipop+0.012
 lid+0.011
cases+0.010
atively+0.009
Neg logit contributions
Playing-0.017
iling-0.016
Fighting-0.016
Leave-0.015
 sports-0.015
Tokenun
Token ID403
Feature activation+2.36e-02

Pos logit contributions
ipop+0.197
ometer+0.195
 lid+0.176
cases+0.161
atively+0.151
Neg logit contributions
Playing-0.174
iling-0.171
hing-0.162
Fighting-0.161
Leave-0.154
Tokenches
Token ID2052
Feature activation+1.14e-01

Pos logit contributions
ipop+0.037
ometer+0.036
 lid+0.032
cases+0.031
 off+0.030
Neg logit contributions
Playing-0.047
iling-0.043
Fighting-0.043
 sports-0.042
Why-0.042
Token them
Token ID606
Feature activation+7.07e-02

Pos logit contributions
ometer+0.176
ipop+0.168
 lid+0.160
 off+0.159
cases+0.143
Neg logit contributions
Playing-0.232
iling-0.224
Fighting-0.210
Leave-0.203
 sports-0.203
Token and
Token ID290
Feature activation+4.15e-02

Pos logit contributions
ometer+0.119
ipop+0.117
 lid+0.107
cases+0.100
 off+0.096
Neg logit contributions
Playing-0.133
iling-0.129
Fighting-0.125
 sports-0.118
Leave-0.118
Token swall
Token ID20301
Feature activation+2.13e-01

Pos logit contributions
ometer+0.075
ipop+0.071
 lid+0.064
cases+0.059
ouse+0.059
Neg logit contributions
Playing-0.072
iling-0.068
Leave-0.068
Fighting-0.067
Why-0.066
Tokenows
Token ID1666
Feature activation+1.12e-01

Pos logit contributions
ipop+0.232
ometer+0.181
 lid+0.166
cases+0.119
ouse+0.114
Neg logit contributions
Playing-0.492
iling-0.485
hing-0.479
 sports-0.467
 Playing-0.466
Token them
Token ID606
Feature activation+9.13e-02

Pos logit contributions
ometer+0.200
ipop+0.197
 lid+0.181
cases+0.166
 off+0.157
Neg logit contributions
Playing-0.202
iling-0.197
Fighting-0.187
 sports-0.177
 owning-0.177
Token.
Token ID13
Feature activation+2.11e-02

Pos logit contributions
ometer+0.158
ipop+0.156
 lid+0.142
cases+0.130
 off+0.120
Neg logit contributions
Playing-0.167
iling-0.162
Fighting-0.155
 sports-0.148
Leave-0.148
Token He
Token ID679
Feature activation+7.23e-03

Pos logit contributions
ipop+0.038
ometer+0.038
 lid+0.035
cases+0.032
atively+0.030
Neg logit contributions
Playing-0.038
Fighting-0.035
iling-0.035
Why-0.033
 owning-0.033
Token makes
Token ID1838
Feature activation+2.14e-02

Pos logit contributions
ometer+0.011
ipop+0.011
 lid+0.010
cases+0.009
atively+0.008
Neg logit contributions
Playing-0.015
iling-0.014
Fighting-0.014
Leave-0.013
 owning-0.013
Token:
Token ID25
Feature activation-6.53e-02

Pos logit contributions
Playing+0.051
iling+0.050
Fighting+0.048
Leave+0.047
hing+0.047
Neg logit contributions
ometer-0.037
ipop-0.036
 lid-0.032
cases-0.028
 off-0.027
Token Sharing
Token ID36644
Feature activation-4.75e-02

Pos logit contributions
iling+0.101
hing+0.093
Playing+0.093
 years+0.088
 sports+0.088
Neg logit contributions
ometer-0.130
ipop-0.119
 lid-0.113
 off-0.103
 mouth-0.102
Token is
Token ID318
Feature activation-4.17e-02

Pos logit contributions
iling+0.106
Playing+0.104
years+0.103
Leave+0.101
Fighting+0.099
Neg logit contributions
ometer-0.060
ipop-0.056
 lid-0.053
 off-0.048
 mouth-0.042
Token caring
Token ID18088
Feature activation-9.56e-03

Pos logit contributions
iling+0.069
hing+0.065
Playing+0.064
Leave+0.061
e+0.060
Neg logit contributions
ometer-0.076
ipop-0.071
 lid-0.069
 off-0.068
 prayer-0.059
Token.
Token ID13
Feature activation+1.91e-03

Pos logit contributions
Playing+0.020
iling+0.020
Fighting+0.019
Leave+0.019
years+0.018
Neg logit contributions
ometer-0.014
ipop-0.013
 lid-0.013
 off-0.011
cases-0.011
Token Pride
Token ID21735
Feature activation+2.13e-01

Pos logit contributions
ometer+0.004
ipop+0.004
 lid+0.003
cases+0.003
 off+0.003
Neg logit contributions
Playing-0.003
iling-0.003
Fighting-0.003
 sports-0.003
Leave-0.003
Token goes
Token ID2925
Feature activation+8.54e-03

Pos logit contributions
ometer+0.338
ipop+0.326
 lid+0.300
cases+0.273
 off+0.260
Neg logit contributions
Playing-0.424
iling-0.411
Fighting-0.395
Leave-0.382
 sports-0.379
Token before
Token ID878
Feature activation+9.49e-02

Pos logit contributions
ometer+0.013
ipop+0.011
 lid+0.011
 off+0.011
cases+0.010
Neg logit contributions
Playing-0.018
iling-0.018
Fighting-0.017
hing-0.016
Leave-0.016
Token a
Token ID257
Feature activation+1.22e-01

Pos logit contributions
ometer+0.168
 lid+0.161
ipop+0.156
 off+0.153
 eyes+0.143
Neg logit contributions
Playing-0.153
iling-0.148
hing-0.147
years-0.145
Fighting-0.140
Token fall
Token ID2121
Feature activation+7.28e-02

Pos logit contributions
ometer+0.200
ipop+0.195
 lid+0.190
 off+0.156
cases+0.151
Neg logit contributions
Playing-0.240
iling-0.234
Fighting-0.221
hing-0.214
Leave-0.210
Token.
Token ID13
Feature activation+1.31e-02

Pos logit contributions
ometer+0.114
ipop+0.103
 lid+0.097
 off+0.094
cases+0.089
Neg logit contributions
Playing-0.152
iling-0.147
Fighting-0.141
 sports-0.135
 years-0.135
Token a
Token ID257
Feature activation+7.48e-02

Pos logit contributions
iling+0.076
Playing+0.075
Fighting+0.071
hing+0.069
Leave+0.068
Neg logit contributions
ometer-0.057
ipop-0.056
 lid-0.053
cases-0.047
 off-0.044
Token copper
Token ID15317
Feature activation+2.64e-02

Pos logit contributions
ometer+0.120
ipop+0.116
 lid+0.116
 mouth+0.089
 door+0.088
Neg logit contributions
iling-0.154
Playing-0.150
Fighting-0.142
hing-0.140
Leave-0.135
Token pot
Token ID1787
Feature activation+9.12e-02

Pos logit contributions
ometer+0.037
ipop+0.037
 lid+0.033
cases+0.030
 off+0.027
Neg logit contributions
Playing-0.056
iling-0.055
Fighting-0.053
 sports-0.051
 owning-0.051
Token with
Token ID351
Feature activation+6.17e-02

Pos logit contributions
ipop+0.150
ometer+0.146
 lid+0.141
cases+0.126
 off+0.123
Neg logit contributions
Playing-0.176
Fighting-0.169
iling-0.169
Leave-0.159
 sports-0.155
Token a
Token ID257
Feature activation+1.13e-01

Pos logit contributions
ometer+0.094
ipop+0.093
 lid+0.092
 off+0.077
cases+0.075
Neg logit contributions
Playing-0.125
iling-0.124
Fighting-0.117
Leave-0.111
hing-0.110
Token lid
Token ID19789
Feature activation+2.13e-01

Pos logit contributions
ipop+0.191
ometer+0.190
cases+0.167
 lid+0.159
atively+0.150
Neg logit contributions
Playing-0.201
iling-0.190
Fighting-0.187
 owning-0.182
 sports-0.182
Token.
Token ID13
Feature activation+1.59e-02

Pos logit contributions
ometer+0.388
ipop+0.381
 lid+0.347
cases+0.320
atively+0.297
Neg logit contributions
Playing-0.368
iling-0.358
Fighting-0.339
 owning-0.326
 sports-0.324
Token It
Token ID632
Feature activation+2.31e-02

Pos logit contributions
ipop+0.029
ometer+0.029
 lid+0.026
cases+0.025
 off+0.023
Neg logit contributions
Playing-0.027
Fighting-0.025
iling-0.025
Why-0.024
 town-0.024
Token was
Token ID373
Feature activation-2.75e-02

Pos logit contributions
ometer+0.047
ipop+0.046
 lid+0.042
cases+0.040
 off+0.037
Neg logit contributions
Playing-0.036
iling-0.033
Fighting-0.032
Why-0.031
 years-0.030
Token very
Token ID845
Feature activation-2.95e-02

Pos logit contributions
Playing+0.047
iling+0.045
Fighting+0.043
 owning+0.042
 sports+0.041
Neg logit contributions
ometer-0.051
ipop-0.050
 lid-0.045
cases-0.043
atively-0.039
Token old
Token ID1468
Feature activation-1.30e-01

Pos logit contributions
Playing+0.043
iling+0.040
 years+0.038
Fighting+0.038
 owning+0.038
Neg logit contributions
ipop-0.061
ometer-0.060
cases-0.053
 lid-0.053
atively-0.049
Token They
Token ID1119
Feature activation-3.56e-02

Pos logit contributions
ometer+0.019
ipop+0.018
 lid+0.017
cases+0.016
 off+0.015
Neg logit contributions
Playing-0.027
iling-0.027
Fighting-0.025
hing-0.025
Leave-0.025
Token grabbed
Token ID13176
Feature activation+1.99e-02

Pos logit contributions
iling+0.072
Fighting+0.067
Playing+0.066
 disl+0.064
Leave+0.064
Neg logit contributions
ipop-0.055
ometer-0.052
 lid-0.050
cases-0.049
 off-0.049
Token an
Token ID281
Feature activation+1.08e-01

Pos logit contributions
ometer+0.035
ipop+0.035
 lid+0.032
cases+0.029
 off+0.028
Neg logit contributions
Playing-0.036
iling-0.035
Fighting-0.034
Leave-0.032
 owning-0.032
Token old
Token ID1468
Feature activation-7.08e-02

Pos logit contributions
ometer+0.196
ipop+0.195
 lid+0.184
cases+0.165
 off+0.161
Neg logit contributions
Playing-0.192
iling-0.181
Fighting-0.177
Leave-0.166
Why-0.166
Token ash
Token ID12530
Feature activation+9.22e-02

Pos logit contributions
Playing+0.128
iling+0.125
Fighting+0.119
 sports+0.114
Leave+0.113
Neg logit contributions
ometer-0.124
ipop-0.122
 lid-0.112
cases-0.103
 off-0.095
Tokentr
Token ID2213
Feature activation+2.12e-01

Pos logit contributions
ipop+0.113
ometer+0.104
cases+0.092
 lid+0.087
ouse+0.078
Neg logit contributions
Playing-0.188
iling-0.187
 sports-0.183
hing-0.183
 town-0.182
Tokenay
Token ID323
Feature activation+3.19e-03

Pos logit contributions
ipop+0.488
ometer+0.452
 lid+0.447
 prayer+0.420
ae+0.417
Neg logit contributions
hing-0.204
Playing-0.179
e-0.175
 course-0.175
iling-0.174
Token and
Token ID290
Feature activation+3.50e-02

Pos logit contributions
ometer+0.006
ipop+0.006
 lid+0.005
cases+0.005
 off+0.005
Neg logit contributions
Playing-0.006
iling-0.006
Fighting-0.005
Leave-0.005
 sports-0.005
Token put
Token ID1234
Feature activation-5.92e-03

Pos logit contributions
ometer+0.061
ipop+0.061
 lid+0.055
cases+0.051
 off+0.048
Neg logit contributions
Playing-0.063
iling-0.062
Fighting-0.059
Leave-0.056
 sports-0.056
Token it
Token ID340
Feature activation+1.08e-01

Pos logit contributions
Playing+0.011
iling+0.011
Fighting+0.011
 sports+0.010
 owning+0.010
Neg logit contributions
ometer-0.010
ipop-0.010
 lid-0.008
cases-0.008
atively-0.007
Token in
Token ID287
Feature activation+3.09e-02

Pos logit contributions
ometer+0.215
ipop+0.212
 lid+0.193
cases+0.181
ouse+0.167
Neg logit contributions
Playing-0.166
iling-0.159
Fighting-0.150
 sports-0.147
 owning-0.144
Token his
Token ID465
Feature activation+8.88e-02

Pos logit contributions
ometer+0.176
ipop+0.174
 lid+0.160
cases+0.149
 off+0.139
Neg logit contributions
Playing-0.146
iling-0.142
Fighting-0.135
 sports-0.128
Leave-0.128
Token head
Token ID1182
Feature activation+8.85e-02

Pos logit contributions
ometer+0.121
ipop+0.121
 lid+0.104
cases+0.095
atively+0.085
Neg logit contributions
Playing-0.189
iling-0.184
Fighting-0.179
 years-0.175
Leave-0.175
Token and
Token ID290
Feature activation+3.01e-02

Pos logit contributions
ipop+0.171
ometer+0.170
 lid+0.158
cases+0.144
atively+0.136
Neg logit contributions
Playing-0.138
iling-0.130
Fighting-0.126
e-0.122
hing-0.121
Token a
Token ID257
Feature activation+1.17e-01

Pos logit contributions
ometer+0.055
ipop+0.054
 lid+0.049
cases+0.044
atively+0.042
Neg logit contributions
Playing-0.052
iling-0.048
Fighting-0.047
Leave-0.046
 sports-0.046
Token spl
Token ID4328
Feature activation+4.97e-02

Pos logit contributions
ometer+0.183
ipop+0.183
 lid+0.156
cases+0.155
atively+0.139
Neg logit contributions
Playing-0.223
Fighting-0.210
iling-0.208
Leave-0.206
 sports-0.205
Tokenint
Token ID600
Feature activation+2.12e-01

Pos logit contributions
ipop+0.076
ometer+0.075
 lid+0.072
cases+0.062
 off+0.061
Neg logit contributions
Playing-0.096
iling-0.095
hing-0.089
Fighting-0.089
e-0.087
Token on
Token ID319
Feature activation+9.83e-02

Pos logit contributions
 lid+0.313
ipop+0.311
ometer+0.287
cases+0.272
atively+0.246
Neg logit contributions
e-0.363
hing-0.363
Playing-0.327
iling-0.324
ting-0.305
Token his
Token ID465
Feature activation+8.71e-02

Pos logit contributions
ometer+0.199
ipop+0.196
 lid+0.184
cases+0.169
 off+0.162
Neg logit contributions
Playing-0.153
iling-0.148
Fighting-0.139
 owning-0.131
Leave-0.131
Token leg
Token ID1232
Feature activation+4.38e-02

Pos logit contributions
ipop+0.119
ometer+0.115
 lid+0.097
cases+0.093
atively+0.083
Neg logit contributions
Playing-0.179
iling-0.173
 years-0.173
Fighting-0.170
 town-0.170
Token.
Token ID13
Feature activation+2.85e-02

Pos logit contributions
ometer+0.077
ipop+0.076
 lid+0.069
cases+0.064
atively+0.059
Neg logit contributions
Playing-0.078
iling-0.075
Fighting-0.072
 sports-0.069
Leave-0.069
Token They
Token ID1119
Feature activation-1.47e-02

Pos logit contributions
ipop+0.061
ometer+0.059
 lid+0.054
cases+0.053
ouse+0.049
Neg logit contributions
Playing-0.043
Fighting-0.038
Why-0.038
 Playing-0.037
 town-0.036
Token her
Token ID607
Feature activation+6.05e-02

Pos logit contributions
ipop+0.113
ometer+0.107
 off+0.096
 lid+0.096
 goodbye+0.085
Neg logit contributions
iling-0.149
Playing-0.143
Fighting-0.140
Leave-0.135
hing-0.135
Token mom
Token ID1995
Feature activation+3.91e-03

Pos logit contributions
ometer+0.065
ipop+0.062
 lid+0.055
cases+0.049
atively+0.044
Neg logit contributions
Playing-0.151
 owning-0.141
iling-0.141
Fighting-0.140
 sports-0.139
Token what
Token ID644
Feature activation+1.28e-02

Pos logit contributions
ometer+0.007
ipop+0.007
 lid+0.006
cases+0.006
atively+0.005
Neg logit contributions
Playing-0.007
iling-0.007
Fighting-0.007
 owning-0.006
 sports-0.006
Token an
Token ID281
Feature activation+9.92e-02

Pos logit contributions
ometer+0.018
ipop+0.018
 lid+0.015
 off+0.013
cases+0.012
Neg logit contributions
iling-0.029
hing-0.027
Leave-0.026
Playing-0.026
 sidelines-0.026
Token ash
Token ID12530
Feature activation+8.17e-02

Pos logit contributions
ometer+0.211
ipop+0.208
 lid+0.194
cases+0.179
 off+0.170
Neg logit contributions
Playing-0.145
iling-0.137
Fighting-0.132
Leave-0.125
 sports-0.123
Tokentr
Token ID2213
Feature activation+2.12e-01

Pos logit contributions
ipop+0.088
ometer+0.078
 lid+0.073
cases+0.069
 off+0.058
Neg logit contributions
Playing-0.188
iling-0.179
Why-0.178
Fighting-0.176
 owning-0.175
Tokenay
Token ID323
Feature activation-5.26e-03

Pos logit contributions
ipop+0.448
 lid+0.427
ometer+0.418
atively+0.396
 prayer+0.380
Neg logit contributions
Playing-0.227
hing-0.226
iling-0.218
 repairing-0.218
e-0.215
Token was
Token ID373
Feature activation-2.26e-02

Pos logit contributions
Playing+0.010
iling+0.010
Fighting+0.009
 sports+0.009
Leave+0.009
Neg logit contributions
ometer-0.009
ipop-0.009
 lid-0.008
cases-0.007
 off-0.007
Token,
Token ID11
Feature activation-1.95e-02

Pos logit contributions
Playing+0.045
iling+0.043
Fighting+0.042
Leave+0.042
hing+0.041
Neg logit contributions
ometer-0.035
ipop-0.035
 lid-0.032
cases-0.029
 off-0.028
Token and
Token ID290
Feature activation+1.01e-02

Pos logit contributions
iling+0.040
Playing+0.038
hing+0.036
Fighting+0.036
e+0.035
Neg logit contributions
ipop-0.029
ometer-0.028
 lid-0.026
 off-0.023
cases-0.023
Token her
Token ID607
Feature activation+6.13e-02

Pos logit contributions
ometer+0.021
ipop+0.020
 lid+0.019
cases+0.018
ouse+0.017
Neg logit contributions
Playing-0.016
Fighting-0.014
iling-0.014
Why-0.014
 owning-0.014
Token he
Token ID339
Feature activation+2.29e-02

Pos logit contributions
ometer+0.119
ipop+0.118
 lid+0.105
cases+0.091
 off+0.086
Neg logit contributions
iling-0.220
Playing-0.218
Fighting-0.208
hing-0.202
Leave-0.197
Token sp
Token ID599
Feature activation+8.37e-02

Pos logit contributions
ipop+0.040
ometer+0.038
 lid+0.037
cases+0.033
 off+0.032
Neg logit contributions
iling-0.041
Playing-0.040
Fighting-0.040
hing-0.037
 sports-0.037
Tokenied
Token ID798
Feature activation+5.70e-02

Pos logit contributions
ometer+0.151
ipop+0.141
 lid+0.132
 prayer+0.119
 mouth+0.113
Neg logit contributions
iling-0.155
Playing-0.148
hing-0.138
Fighting-0.131
 sports-0.130
Token an
Token ID281
Feature activation+1.07e-01

Pos logit contributions
ometer+0.096
ipop+0.095
 lid+0.088
cases+0.079
 off+0.076
Neg logit contributions
Playing-0.108
iling-0.105
Fighting-0.101
Leave-0.096
 owning-0.094
Token ash
Token ID12530
Feature activation+8.56e-02

Pos logit contributions
ipop+0.193
ometer+0.192
 lid+0.183
 off+0.166
 door+0.156
Neg logit contributions
Playing-0.188
Fighting-0.174
iling-0.168
Leave-0.166
Why-0.157
Tokentr
Token ID2213
Feature activation+2.10e-01

Pos logit contributions
ipop+0.093
ometer+0.088
cases+0.071
 lid+0.070
atively+0.056
Neg logit contributions
Playing-0.196
iling-0.196
Why-0.188
 town-0.187
e-0.186
Tokenay
Token ID323
Feature activation+1.20e-03

Pos logit contributions
ipop+0.470
ometer+0.452
 lid+0.437
 prayer+0.407
atively+0.403
Neg logit contributions
hing-0.217
iling-0.202
Playing-0.193
 course-0.193
e-0.192
Token.
Token ID13
Feature activation+1.44e-02

Pos logit contributions
ometer+0.002
ipop+0.002
 lid+0.002
cases+0.002
 off+0.002
Neg logit contributions
Playing-0.002
iling-0.002
Fighting-0.002
Leave-0.002
 sports-0.002
Token He
Token ID679
Feature activation-1.21e-02

Pos logit contributions
ipop+0.027
ometer+0.026
 lid+0.024
cases+0.023
atively+0.021
Neg logit contributions
Playing-0.025
Fighting-0.023
iling-0.022
 town-0.022
Why-0.022
Token picked
Token ID6497
Feature activation+6.14e-02

Pos logit contributions
iling+0.023
Playing+0.022
Fighting+0.022
Leave+0.021
hing+0.021
Neg logit contributions
ometer-0.020
ipop-0.020
 lid-0.018
cases-0.017
 off-0.016
Token up
Token ID510
Feature activation+7.68e-02

Pos logit contributions
ipop+0.117
ometer+0.117
cases+0.104
 lid+0.099
atively+0.095
Neg logit contributions
Playing-0.096
iling-0.092
e-0.091
Fighting-0.090
 owning-0.090
Token noise
Token ID7838
Feature activation+8.63e-02

Pos logit contributions
ometer+0.054
ipop+0.053
 lid+0.051
 off+0.045
cases+0.044
Neg logit contributions
Playing-0.057
iling-0.053
Fighting-0.051
years-0.049
 owning-0.048
Token in
Token ID287
Feature activation+3.84e-02

Pos logit contributions
ometer+0.113
ipop+0.109
 off+0.101
 lid+0.099
cases+0.096
Neg logit contributions
Playing-0.187
iling-0.179
Fighting-0.177
years-0.175
 sports-0.168
Token the
Token ID262
Feature activation+3.59e-03

Pos logit contributions
 lid+0.069
ometer+0.068
ipop+0.066
 off+0.062
cases+0.057
Neg logit contributions
Playing-0.069
iling-0.062
Fighting-0.061
 sports-0.057
 owning-0.057
Token bushes
Token ID37413
Feature activation+1.80e-03

Pos logit contributions
ometer+0.006
 lid+0.006
ipop+0.006
 off+0.005
cases+0.005
Neg logit contributions
Playing-0.007
iling-0.007
Fighting-0.006
 owning-0.006
 sports-0.006
Token.
Token ID13
Feature activation+1.18e-02

Pos logit contributions
ometer+0.002
ipop+0.002
 lid+0.002
cases+0.002
 off+0.002
Neg logit contributions
Playing-0.004
iling-0.004
Fighting-0.004
Leave-0.004
 disl-0.004
Token Care
Token ID7276
Feature activation+2.10e-01

Pos logit contributions
ometer+0.019
ipop+0.019
 lid+0.017
cases+0.016
 off+0.014
Neg logit contributions
Playing-0.023
iling-0.022
Fighting-0.021
 sports-0.020
Leave-0.020
Tokenfully
Token ID2759
Feature activation+1.34e-01

Pos logit contributions
ometer+0.317
ipop+0.314
 lid+0.282
cases+0.254
 off+0.228
Neg logit contributions
Playing-0.428
iling-0.419
Fighting-0.402
 sports-0.386
Leave-0.386
Token,
Token ID11
Feature activation+8.62e-03

Pos logit contributions
ometer+0.232
ipop+0.228
 lid+0.208
cases+0.191
atively+0.174
Neg logit contributions
Playing-0.245
iling-0.238
Fighting-0.227
Leave-0.218
 owning-0.218
Token he
Token ID339
Feature activation+2.55e-02

Pos logit contributions
ometer+0.016
ipop+0.016
 lid+0.015
cases+0.013
 off+0.013
Neg logit contributions
Playing-0.015
iling-0.015
Fighting-0.014
hing-0.013
Leave-0.013
Token moved
Token ID3888
Feature activation+3.67e-02

Pos logit contributions
ometer+0.043
ipop+0.042
 lid+0.039
cases+0.036
 off+0.033
Neg logit contributions
Playing-0.048
iling-0.047
Fighting-0.045
 sports-0.043
Leave-0.043
Token closer
Token ID5699
Feature activation+1.49e-01

Pos logit contributions
ometer+0.061
ipop+0.060
 lid+0.055
cases+0.050
 off+0.048
Neg logit contributions
Playing-0.071
iling-0.069
Fighting-0.066
 owning-0.063
Leave-0.062
Token your
Token ID534
Feature activation+3.42e-02

Pos logit contributions
Playing+0.045
iling+0.043
Fighting+0.042
Leave+0.040
 sports+0.039
Neg logit contributions
ometer-0.035
ipop-0.033
 lid-0.032
 off-0.028
cases-0.027
Token fingers
Token ID9353
Feature activation+4.92e-02

Pos logit contributions
ometer+0.060
ipop+0.059
 lid+0.054
cases+0.049
 off+0.046
Neg logit contributions
Playing-0.062
iling-0.061
Fighting-0.058
 sports-0.055
Leave-0.055
Token in
Token ID287
Feature activation+2.66e-02

Pos logit contributions
ometer+0.081
ipop+0.080
 lid+0.073
cases+0.067
 off+0.062
Neg logit contributions
Playing-0.094
iling-0.091
Fighting-0.087
Leave-0.083
 sports-0.083
Token the
Token ID262
Feature activation-2.73e-03

Pos logit contributions
 lid+0.041
ometer+0.038
 off+0.037
ipop+0.036
 mouth+0.034
Neg logit contributions
Playing-0.055
iling-0.051
Fighting-0.048
hing-0.048
Leave-0.046
Token ash
Token ID12530
Feature activation+8.07e-02

Pos logit contributions
Playing+0.005
iling+0.005
Fighting+0.004
 sports+0.004
hing+0.004
Neg logit contributions
ometer-0.005
 lid-0.005
ipop-0.005
 mouth-0.004
 door-0.004
Tokentr
Token ID2213
Feature activation+2.10e-01

Pos logit contributions
ipop+0.084
ometer+0.075
 lid+0.068
cases+0.058
atively+0.051
Neg logit contributions
Playing-0.192
iling-0.190
hing-0.182
Fighting-0.182
 years-0.181
Tokenay
Token ID323
Feature activation-7.46e-03

Pos logit contributions
ipop+0.453
 lid+0.426
ometer+0.421
atively+0.395
 goodbye+0.376
Neg logit contributions
hing-0.233
iling-0.226
Playing-0.225
e-0.214
 repairing-0.211
Token,
Token ID11
Feature activation-2.47e-03

Pos logit contributions
Playing+0.014
iling+0.014
Fighting+0.013
Leave+0.013
 owning+0.012
Neg logit contributions
ometer-0.013
ipop-0.013
 lid-0.011
cases-0.011
 off-0.010
Token so
Token ID523
Feature activation+5.98e-02

Pos logit contributions
Playing+0.005
iling+0.005
Fighting+0.005
hing+0.004
 sports+0.004
Neg logit contributions
ometer-0.004
ipop-0.004
 lid-0.003
 off-0.003
cases-0.003
Token Jon
Token ID5966
Feature activation-5.37e-02

Pos logit contributions
ometer+0.094
ipop+0.093
 lid+0.084
cases+0.076
 off+0.069
Neg logit contributions
Playing-0.119
iling-0.116
Fighting-0.112
 sports-0.107
Leave-0.107
Token had
Token ID550
Feature activation-1.99e-02

Pos logit contributions
Playing+0.096
iling+0.094
Fighting+0.089
 sports+0.085
Leave+0.085
Neg logit contributions
ometer-0.094
ipop-0.093
 lid-0.086
cases-0.079
atively-0.072
Token "
Token ID366
Feature activation-1.63e-01

Pos logit contributions
Playing+0.022
iling+0.022
Fighting+0.021
 sports+0.020
Leave+0.020
Neg logit contributions
ometer-0.016
ipop-0.016
 lid-0.014
cases-0.013
 off-0.011
TokenYou
Token ID1639
Feature activation-8.64e-03

Pos logit contributions
iling+0.194
 sports+0.167
 owning+0.167
 years+0.164
 disl+0.160
Neg logit contributions
ometer-0.356
ipop-0.352
 lid-0.319
antly-0.300
cases-0.295
Token are
Token ID389
Feature activation-7.46e-02

Pos logit contributions
Playing+0.015
iling+0.015
Fighting+0.013
Leave+0.013
 sports+0.013
Neg logit contributions
ipop-0.016
ometer-0.016
 lid-0.015
cases-0.014
 off-0.013
Token the
Token ID262
Feature activation-7.21e-02

Pos logit contributions
iling+0.151
Playing+0.146
Leave+0.141
Fighting+0.138
 arrest+0.135
Neg logit contributions
ometer-0.103
ipop-0.100
 lid-0.097
 off-0.094
 name-0.079
Token best
Token ID1266
Feature activation-5.56e-02

Pos logit contributions
Playing+0.185
iling+0.185
Fighting+0.174
Leave+0.171
hing+0.170
Neg logit contributions
ometer-0.072
ipop-0.069
 lid-0.062
cases-0.044
 off-0.043
Token kids
Token ID3988
Feature activation-2.18e-01

Pos logit contributions
Playing+0.093
iling+0.091
 sports+0.087
Fighting+0.085
 town+0.085
Neg logit contributions
ometer-0.100
ipop-0.099
 lid-0.093
cases-0.090
atively-0.084
Token in
Token ID287
Feature activation-2.68e-03

Pos logit contributions
Playing+0.392
iling+0.377
Fighting+0.376
Leave+0.364
hing+0.353
Neg logit contributions
ipop-0.392
ometer-0.380
 lid-0.322
cases-0.313
 off-0.304
Token the
Token ID262
Feature activation-2.66e-02

Pos logit contributions
Playing+0.006
hing+0.006
Fighting+0.006
Leave+0.005
ican+0.005
Neg logit contributions
 off-0.003
 lid-0.003
 mom-0.003
ipop-0.003
 door-0.003
Token world
Token ID995
Feature activation-1.24e-01

Pos logit contributions
iling+0.040
hing+0.040
Playing+0.040
years+0.039
Leave+0.037
Neg logit contributions
 lid-0.055
ometer-0.051
 door-0.049
ipop-0.048
 bag-0.046
Token.
Token ID13
Feature activation-3.27e-03

Pos logit contributions
Playing+0.293
Leave+0.284
Fighting+0.281
iling+0.269
years+0.262
Neg logit contributions
ometer-0.151
ipop-0.151
cases-0.124
 off-0.114
 lid-0.113
Token I
Token ID314
Feature activation-8.20e-02

Pos logit contributions
Playing+0.006
iling+0.006
Fighting+0.005
hing+0.005
 years+0.005
Neg logit contributions
ometer-0.006
ipop-0.005
 lid-0.005
cases-0.004
 off-0.004
Token "
Token ID366
Feature activation-1.66e-01

Pos logit contributions
Playing+0.011
iling+0.011
Fighting+0.010
 sports+0.010
Leave+0.010
Neg logit contributions
ometer-0.008
ipop-0.008
 lid-0.007
cases-0.006
 off-0.006
TokenYou
Token ID1639
Feature activation-5.21e-03

Pos logit contributions
iling+0.215
 repairing+0.192
 sports+0.187
 owning+0.185
 destroying+0.185
Neg logit contributions
ipop-0.338
ometer-0.328
 lid-0.302
 reluctantly-0.275
antly-0.275
Token are
Token ID389
Feature activation-7.17e-02

Pos logit contributions
Playing+0.009
iling+0.009
Fighting+0.008
Leave+0.008
 sports+0.008
Neg logit contributions
ipop-0.010
ometer-0.010
 lid-0.009
cases-0.008
 off-0.008
Token the
Token ID262
Feature activation-6.84e-02

Pos logit contributions
iling+0.136
Playing+0.134
Fighting+0.126
Leave+0.126
hing+0.122
Neg logit contributions
ometer-0.107
ipop-0.104
 lid-0.102
 off-0.097
 name-0.084
Token best
Token ID1266
Feature activation-5.05e-02

Pos logit contributions
iling+0.197
Playing+0.189
 disl+0.176
years+0.175
hing+0.175
Neg logit contributions
ometer-0.057
 lid-0.048
ipop-0.044
 key-0.032
 name-0.029
Token kids
Token ID3988
Feature activation-2.16e-01

Pos logit contributions
Playing+0.077
iling+0.075
 sports+0.072
Fighting+0.070
 town+0.070
Neg logit contributions
ipop-0.099
ometer-0.098
 lid-0.091
cases-0.087
atively-0.082
Token.
Token ID13
Feature activation-7.16e-03

Pos logit contributions
Playing+0.381
Fighting+0.365
iling+0.362
Leave+0.352
hing+0.348
Neg logit contributions
ipop-0.395
ometer-0.381
 lid-0.327
cases-0.311
 off-0.309
Token You
Token ID921
Feature activation-1.25e-02

Pos logit contributions
Playing+0.013
iling+0.013
Fighting+0.012
hing+0.012
 sports+0.011
Neg logit contributions
ometer-0.013
ipop-0.012
 lid-0.011
 off-0.009
cases-0.009
Token are
Token ID389
Feature activation-7.01e-02

Pos logit contributions
Playing+0.023
iling+0.022
Leave+0.021
Fighting+0.021
 sports+0.020
Neg logit contributions
ipop-0.021
ometer-0.021
 lid-0.020
 off-0.017
cases-0.017
Token very
Token ID845
Feature activation-4.88e-02

Pos logit contributions
iling+0.149
Playing+0.148
Leave+0.138
Fighting+0.137
hing+0.135
Neg logit contributions
ometer-0.087
 off-0.084
ipop-0.083
 lid-0.081
 name-0.066
Token smart
Token ID4451
Feature activation-1.11e-01

Pos logit contributions
Playing+0.086
iling+0.083
Fighting+0.081
Leave+0.080
hing+0.076
Neg logit contributions
ometer-0.087
ipop-0.086
 lid-0.078
cases-0.070
ouse-0.068
Token "
Token ID366
Feature activation-1.78e-01

Pos logit contributions
iling+0.044
Playing+0.040
 sports+0.040
Fighting+0.040
e+0.039
Neg logit contributions
ometer-0.021
ipop-0.021
 lid-0.019
cases-0.016
 off-0.015
TokenThey
Token ID2990
Feature activation-2.26e-02

Pos logit contributions
iling+0.215
Playing+0.198
 sports+0.193
 owning+0.191
 years+0.185
Neg logit contributions
ometer-0.398
ipop-0.391
 lid-0.365
cases-0.350
 off-0.326
Token are
Token ID389
Feature activation-7.44e-02

Pos logit contributions
Playing+0.048
Leave+0.046
years+0.045
Fighting+0.045
Why+0.043
Neg logit contributions
ometer-0.027
 lid-0.027
ipop-0.027
ows-0.025
 name-0.024
Token such
Token ID884
Feature activation-6.21e-02

Pos logit contributions
iling+0.153
Playing+0.147
hing+0.141
Leave+0.139
Fighting+0.139
Neg logit contributions
ometer-0.100
 off-0.098
ipop-0.096
 lid-0.093
 mouth-0.081
Token good
Token ID922
Feature activation-1.12e-01

Pos logit contributions
iling+0.142
Playing+0.140
hing+0.136
Leave+0.130
Fighting+0.129
Neg logit contributions
ometer-0.075
 lid-0.068
ipop-0.065
 off-0.057
 name-0.056
Token kids
Token ID3988
Feature activation-2.12e-01

Pos logit contributions
Playing+0.178
iling+0.169
Fighting+0.163
Leave+0.155
hing+0.154
Neg logit contributions
ometer-0.224
ipop-0.221
 lid-0.203
cases-0.190
 off-0.183
Token.
Token ID13
Feature activation-1.13e-02

Pos logit contributions
Playing+0.483
Fighting+0.466
iling+0.465
Leave+0.464
hing+0.456
Neg logit contributions
ipop-0.271
ometer-0.271
cases-0.209
 lid-0.204
 off-0.197
Token They
Token ID1119
Feature activation-3.94e-02

Pos logit contributions
iling+0.021
Playing+0.021
Fighting+0.020
hing+0.020
 owning+0.019
Neg logit contributions
ometer-0.019
ipop-0.018
 lid-0.016
cases-0.015
 off-0.014
Token deserve
Token ID10925
Feature activation-3.91e-02

Pos logit contributions
Playing+0.081
years+0.079
iling+0.078
Leave+0.077
 sports+0.075
Neg logit contributions
ometer-0.051
 lid-0.048
ipop-0.048
 name-0.043
ows-0.043
Token a
Token ID257
Feature activation+8.00e-02

Pos logit contributions
iling+0.076
Playing+0.076
Fighting+0.071
hing+0.070
Leave+0.068
Neg logit contributions
ometer-0.060
ipop-0.060
 lid-0.055
 off-0.052
cases-0.048
Token nice
Token ID3621
Feature activation-4.27e-02

Pos logit contributions
ipop+0.149
ometer+0.147
 lid+0.135
cases+0.133
atively+0.126
Neg logit contributions
Playing-0.128
 sports-0.122
 town-0.118
Fighting-0.117
iling-0.115
Token.
Token ID13
Feature activation+1.42e-02

Pos logit contributions
Playing+0.059
iling+0.057
Fighting+0.055
Leave+0.054
hing+0.053
Neg logit contributions
ometer-0.039
ipop-0.039
 lid-0.035
cases-0.030
 off-0.030
Token You
Token ID921
Feature activation-2.05e-03

Pos logit contributions
ometer+0.027
ipop+0.026
 lid+0.024
cases+0.022
 off+0.021
Neg logit contributions
Playing-0.023
iling-0.023
Fighting-0.022
 sports-0.021
hing-0.021
Token are
Token ID389
Feature activation-6.43e-02

Pos logit contributions
Playing+0.004
iling+0.004
Fighting+0.003
Leave+0.003
 sports+0.003
Neg logit contributions
ipop-0.004
ometer-0.003
 lid-0.003
cases-0.003
 off-0.003
Token the
Token ID262
Feature activation-6.57e-02

Pos logit contributions
iling+0.132
Playing+0.131
Fighting+0.123
Leave+0.123
hing+0.121
Neg logit contributions
ometer-0.085
ipop-0.081
 lid-0.080
 off-0.079
 prayer-0.065
Token best
Token ID1266
Feature activation-4.77e-02

Pos logit contributions
iling+0.176
Playing+0.174
Fighting+0.162
hing+0.159
 disl+0.158
Neg logit contributions
ometer-0.060
ipop-0.053
 lid-0.053
 off-0.036
cases-0.032
Token kids
Token ID3988
Feature activation-2.12e-01

Pos logit contributions
Playing+0.077
iling+0.075
Fighting+0.072
 sports+0.071
 town+0.070
Neg logit contributions
ipop-0.090
ometer-0.090
 lid-0.083
cases-0.079
atively-0.073
Token and
Token ID290
Feature activation-1.41e-02

Pos logit contributions
Playing+0.373
Fighting+0.356
iling+0.356
Leave+0.343
hing+0.337
Neg logit contributions
ipop-0.385
ometer-0.372
 lid-0.320
cases-0.307
 off-0.306
Token the
Token ID262
Feature activation-4.08e-02

Pos logit contributions
iling+0.027
Playing+0.026
hing+0.025
Fighting+0.024
Leave+0.024
Neg logit contributions
ometer-0.024
ipop-0.023
 lid-0.021
cases-0.019
 off-0.019
Token best
Token ID1266
Feature activation-3.12e-02

Pos logit contributions
iling+0.089
Playing+0.084
years+0.080
hing+0.080
e+0.076
Neg logit contributions
 lid-0.058
ometer-0.053
ipop-0.051
 key-0.047
 door-0.046
Token learners
Token ID46184
Feature activation-1.36e-01

Pos logit contributions
Playing+0.052
iling+0.051
Fighting+0.048
 sports+0.046
Leave+0.046
Neg logit contributions
ometer-0.059
ipop-0.058
 lid-0.054
cases-0.050
 off-0.046
Token."
Token ID526
Feature activation+3.54e-02

Pos logit contributions
Playing+0.314
Fighting+0.295
 disl+0.289
iling+0.289
Leave+0.287
Neg logit contributions
ipop-0.163
ometer-0.158
 lid-0.147
 off-0.126
cases-0.118
Token "
Token ID366
Feature activation-1.76e-01

Pos logit contributions
iling+0.041
Playing+0.038
 years+0.036
 sports+0.036
e+0.036
Neg logit contributions
ometer-0.019
ipop-0.018
 lid-0.017
cases-0.014
 off-0.013
TokenThey
Token ID2990
Feature activation-2.69e-02

Pos logit contributions
iling+0.235
Playing+0.233
Fighting+0.213
 owning+0.210
 sports+0.208
Neg logit contributions
ometer-0.385
ipop-0.379
 lid-0.353
cases-0.332
atively-0.314
Token are
Token ID389
Feature activation-7.79e-02

Pos logit contributions
Playing+0.053
Fighting+0.049
Leave+0.048
iling+0.047
years+0.047
Neg logit contributions
ometer-0.042
ipop-0.041
 lid-0.041
cases-0.036
 off-0.035
Token such
Token ID884
Feature activation-6.34e-02

Pos logit contributions
Playing+0.156
iling+0.153
Fighting+0.145
hing+0.145
Leave+0.141
Neg logit contributions
ipop-0.115
ometer-0.114
 lid-0.109
 off-0.102
cases-0.094
Token good
Token ID922
Feature activation-1.11e-01

Pos logit contributions
Playing+0.140
iling+0.138
hing+0.131
Fighting+0.127
Leave+0.125
Neg logit contributions
ometer-0.085
 lid-0.082
ipop-0.077
 off-0.068
 name-0.064
Token kids
Token ID3988
Feature activation-2.11e-01

Pos logit contributions
Playing+0.178
iling+0.166
Fighting+0.163
hing+0.153
Leave+0.153
Neg logit contributions
ometer-0.220
ipop-0.218
 lid-0.203
cases-0.187
 off-0.183
Token.
Token ID13
Feature activation-1.57e-02

Pos logit contributions
Playing+0.470
iling+0.446
Fighting+0.445
Leave+0.440
hing+0.430
Neg logit contributions
ometer-0.288
ipop-0.287
 lid-0.231
cases-0.228
 off-0.219
Token They
Token ID1119
Feature activation-4.01e-02

Pos logit contributions
Playing+0.028
iling+0.028
Fighting+0.026
 sports+0.025
 owning+0.025
Neg logit contributions
ometer-0.027
ipop-0.027
 lid-0.025
cases-0.023
 off-0.021
Token deserve
Token ID10925
Feature activation-4.15e-02

Pos logit contributions
Playing+0.082
iling+0.077
years+0.077
Fighting+0.076
Leave+0.075
Neg logit contributions
ometer-0.055
 lid-0.055
ipop-0.054
cases-0.047
ows-0.047
Token a
Token ID257
Feature activation+7.54e-02

Pos logit contributions
Playing+0.083
iling+0.083
Fighting+0.078
hing+0.076
Leave+0.075
Neg logit contributions
ipop-0.062
ometer-0.061
 lid-0.057
 off-0.053
cases-0.049
Token nice
Token ID3621
Feature activation-4.42e-02

Pos logit contributions
ometer+0.131
ipop+0.128
 lid+0.116
cases+0.115
atively+0.106
Neg logit contributions
Playing-0.131
iling-0.123
 sports-0.123
Fighting-0.121
 town-0.120
Token Bob
Token ID5811
Feature activation-3.79e-02

Pos logit contributions
ometer+0.004
ipop+0.004
 lid+0.003
cases+0.003
 off+0.003
Neg logit contributions
Playing-0.004
iling-0.004
Fighting-0.004
Leave-0.004
 owning-0.004
Tokeno
Token ID78
Feature activation+1.05e-04

Pos logit contributions
 sidelines+0.100
Leave+0.097
Fighting+0.097
Playing+0.093
iling+0.090
Neg logit contributions
ometer-0.042
ipop-0.037
ouse-0.028
bie-0.024
 lid-0.023
Token loved
Token ID6151
Feature activation-1.27e-01

Pos logit contributions
ometer+0.000
ipop+0.000
ouse+0.000
 replied+0.000
 lid+0.000
Neg logit contributions
 sidelines-0.000
Leave-0.000
Fighting-0.000
iling-0.000
ting-0.000
Token to
Token ID284
Feature activation-4.03e-02

Pos logit contributions
iling+0.344
Playing+0.335
Fighting+0.326
hing+0.323
Leave+0.319
Neg logit contributions
ipop-0.110
ometer-0.107
 off-0.081
 lid-0.075
 eyes-0.065
Token make
Token ID787
Feature activation-2.93e-02

Pos logit contributions
Playing+0.069
iling+0.068
Fighting+0.066
years+0.059
 years+0.058
Neg logit contributions
ipop-0.076
ometer-0.072
 lid-0.069
 off-0.064
cases-0.062
Token kids
Token ID3988
Feature activation-2.11e-01

Pos logit contributions
Playing+0.051
iling+0.049
Fighting+0.047
Leave+0.045
hing+0.044
Neg logit contributions
ometer-0.053
ipop-0.052
 lid-0.049
cases-0.046
 off-0.045
Token laugh
Token ID6487
Feature activation-3.71e-02

Pos logit contributions
iling+0.357
Playing+0.356
Fighting+0.337
Leave+0.331
Now+0.324
Neg logit contributions
ipop-0.374
ometer-0.373
 lid-0.344
cases-0.338
 off-0.332
Token and
Token ID290
Feature activation-2.27e-02

Pos logit contributions
Playing+0.072
iling+0.071
Fighting+0.069
Leave+0.068
 disl+0.065
Neg logit contributions
ometer-0.058
ipop-0.056
 lid-0.052
 off-0.050
cases-0.050
Token play
Token ID711
Feature activation-1.31e-01

Pos logit contributions
iling+0.031
Playing+0.030
Fighting+0.028
Leave+0.026
 years+0.025
Neg logit contributions
ometer-0.049
ipop-0.049
 lid-0.046
 off-0.044
 mouth-0.041
Token.
Token ID13
Feature activation+1.84e-03

Pos logit contributions
iling+0.239
Playing+0.238
Fighting+0.228
years+0.222
Leave+0.218
Neg logit contributions
 lid-0.197
 off-0.191
ometer-0.181
ipop-0.181
atively-0.169
Token He
Token ID679
Feature activation-2.45e-02

Pos logit contributions
ipop+0.003
ometer+0.003
 lid+0.003
cases+0.002
 off+0.002
Neg logit contributions
Playing-0.004
iling-0.003
 sports-0.003
Fighting-0.003
 town-0.003
Token
Token ID198
Feature activation-1.24e-02

Pos logit contributions
ometer+0.049
ipop+0.048
 lid+0.043
cases+0.042
ouse+0.037
Neg logit contributions
Playing-0.047
 sports-0.043
Fighting-0.043
iling-0.043
Why-0.042
Token
Token ID198
Feature activation-1.95e-02

Pos logit contributions
Playing+0.020
iling+0.019
Fighting+0.018
 sports+0.018
Why+0.017
Neg logit contributions
ometer-0.024
ipop-0.024
 lid-0.022
cases-0.021
atively-0.019
TokenM
Token ID44
Feature activation+7.51e-02

Pos logit contributions
Playing+0.034
iling+0.033
Fighting+0.032
 owning+0.030
 sports+0.030
Neg logit contributions
ometer-0.035
ipop-0.034
 lid-0.032
cases-0.029
 off-0.027
Tokenum
Token ID388
Feature activation-7.20e-03

Pos logit contributions
ometer+0.094
ipop+0.083
 lid+0.081
cases+0.072
 off+0.064
Neg logit contributions
iling-0.174
Playing-0.173
 years-0.162
 sports-0.159
Fighting-0.156
Token continued
Token ID3767
Feature activation-1.49e-02

Pos logit contributions
iling+0.016
Playing+0.016
Fighting+0.016
Leave+0.016
 owning+0.015
Neg logit contributions
ipop-0.009
ometer-0.008
 lid-0.007
cases-0.007
 off-0.006
Token:"
Token ID11097
Feature activation-2.10e-01

Pos logit contributions
Playing+0.029
iling+0.028
Fighting+0.027
 sports+0.025
Leave+0.025
Neg logit contributions
ometer-0.025
ipop-0.024
 lid-0.022
cases-0.020
 off-0.019
TokenLet
Token ID5756
Feature activation-2.27e-02

Pos logit contributions
iling+0.282
Playing+0.278
Fighting+0.258
hing+0.249
 owning+0.247
Neg logit contributions
ipop-0.456
ometer-0.451
 lid-0.415
cases-0.384
 off-0.368
Token's
Token ID338
Feature activation+2.38e-02

Pos logit contributions
Playing+0.041
iling+0.040
Fighting+0.038
 sports+0.037
Leave+0.036
Neg logit contributions
ometer-0.040
ipop-0.040
 lid-0.036
cases-0.033
atively-0.031
Token start
Token ID923
Feature activation-9.53e-03

Pos logit contributions
ometer+0.041
ipop+0.041
 lid+0.038
cases+0.034
 off+0.033
Neg logit contributions
Playing-0.043
iling-0.042
Fighting-0.041
Leave-0.038
 owning-0.038
Token baking
Token ID16871
Feature activation-5.02e-02

Pos logit contributions
Playing+0.015
iling+0.015
Fighting+0.014
Leave+0.014
hing+0.013
Neg logit contributions
ometer-0.019
ipop-0.018
 lid-0.017
cases-0.016
 off-0.016
Token!"
Token ID2474
Feature activation-6.75e-03

Pos logit contributions
Playing+0.088
iling+0.086
Fighting+0.082
 sports+0.078
 owning+0.078
Neg logit contributions
ometer-0.090
ipop-0.089
 lid-0.081
cases-0.075
 off-0.069
TokenL
Token ID43
Feature activation-7.99e-02

Pos logit contributions
Playing+0.151
iling+0.151
Fighting+0.146
 disl+0.142
Leave+0.140
Neg logit contributions
ometer-0.110
ipop-0.110
 lid-0.104
cases-0.084
 off-0.084
Tokenily
Token ID813
Feature activation-6.26e-02

Pos logit contributions
iling+0.105
Playing+0.103
 years+0.097
e+0.096
 town+0.095
Neg logit contributions
ometer-0.186
ipop-0.170
 lid-0.155
cases-0.152
atively-0.146
Token and
Token ID290
Feature activation-3.58e-03

Pos logit contributions
Fighting+0.126
Playing+0.122
years+0.122
iling+0.121
 sidelines+0.121
Neg logit contributions
ipop-0.090
ometer-0.087
 lid-0.076
 closes-0.073
cases-0.072
Token Ben
Token ID3932
Feature activation-3.62e-02

Pos logit contributions
Playing+0.007
iling+0.006
Fighting+0.006
hing+0.006
 sports+0.006
Neg logit contributions
ometer-0.006
ipop-0.006
 lid-0.006
cases-0.005
 off-0.005
Token are
Token ID389
Feature activation-6.85e-02

Pos logit contributions
Fighting+0.075
iling+0.072
 sidelines+0.071
years+0.071
Playing+0.070
Neg logit contributions
ipop-0.047
 lid-0.042
ometer-0.042
 closes-0.039
cases-0.038
Token kids
Token ID3988
Feature activation-2.10e-01

Pos logit contributions
Playing+0.082
iling+0.080
Fighting+0.075
hing+0.071
Leave+0.071
Neg logit contributions
ometer-0.160
ipop-0.160
 lid-0.148
cases-0.140
 off-0.136
Token who
Token ID508
Feature activation-3.04e-02

Pos logit contributions
iling+0.340
Playing+0.339
Fighting+0.326
hing+0.316
Leave+0.315
Neg logit contributions
ometer-0.388
ipop-0.383
 lid-0.336
cases-0.323
 off-0.319
Token live
Token ID2107
Feature activation-1.12e-01

Pos logit contributions
iling+0.049
Fighting+0.047
Leave+0.046
Playing+0.045
years+0.044
Neg logit contributions
ipop-0.055
ometer-0.053
 name-0.048
cases-0.046
ows-0.045
Token next
Token ID1306
Feature activation-2.53e-03

Pos logit contributions
Playing+0.177
iling+0.174
Fighting+0.168
Leave+0.158
hing+0.156
Neg logit contributions
ometer-0.219
ipop-0.218
 lid-0.201
cases-0.184
 off-0.178
Token door
Token ID3420
Feature activation+7.81e-02

Pos logit contributions
 town+0.005
iling+0.005
 years+0.005
Playing+0.005
 sports+0.005
Neg logit contributions
ipop-0.004
ometer-0.003
 lid-0.003
cases-0.003
atively-0.003
Token.
Token ID13
Feature activation-8.35e-04

Pos logit contributions
ometer+0.128
ipop+0.126
 lid+0.115
cases+0.105
 off+0.096
Neg logit contributions
Playing-0.149
iling-0.145
Fighting-0.139
 sports-0.133
Leave-0.133
TokenL
Token ID43
Feature activation-8.25e-02

Pos logit contributions
Playing+0.171
 disl+0.170
iling+0.170
Fighting+0.169
years+0.167
Neg logit contributions
ipop-0.094
 lid-0.093
ometer-0.092
 off-0.068
antly-0.065
Tokenily
Token ID813
Feature activation-6.25e-02

Pos logit contributions
iling+0.132
Playing+0.128
e+0.121
 years+0.117
 town+0.117
Neg logit contributions
ometer-0.172
ipop-0.154
 lid-0.144
cases-0.139
atively-0.127
Token and
Token ID290
Feature activation-1.85e-03

Pos logit contributions
Fighting+0.128
Playing+0.128
years+0.127
iling+0.126
 sidelines+0.122
Neg logit contributions
ipop-0.085
ometer-0.080
 lid-0.073
 closes-0.068
 reluctantly-0.068
Token Ben
Token ID3932
Feature activation-3.74e-02

Pos logit contributions
Playing+0.004
iling+0.004
Fighting+0.004
Leave+0.003
 owning+0.003
Neg logit contributions
ipop-0.003
ometer-0.003
 lid-0.002
cases-0.002
 off-0.002
Token are
Token ID389
Feature activation-6.98e-02

Pos logit contributions
Fighting+0.078
iling+0.078
years+0.077
Playing+0.075
 sidelines+0.075
Neg logit contributions
ipop-0.044
 lid-0.040
 reluctantly-0.037
 closes-0.037
ometer-0.037
Token kids
Token ID3988
Feature activation-2.10e-01

Pos logit contributions
Playing+0.075
iling+0.073
Fighting+0.068
hing+0.064
Leave+0.064
Neg logit contributions
ometer-0.172
ipop-0.172
 lid-0.161
cases-0.152
 off-0.147
Token who
Token ID508
Feature activation-3.28e-02

Pos logit contributions
iling+0.396
Playing+0.395
Fighting+0.382
hing+0.380
Now+0.380
Neg logit contributions
ometer-0.323
ipop-0.319
 lid-0.271
 off-0.265
cases-0.261
Token like
Token ID588
Feature activation-9.94e-02

Pos logit contributions
iling+0.052
Fighting+0.050
Playing+0.049
Leave+0.048
years+0.047
Neg logit contributions
ipop-0.060
ometer-0.057
 name-0.050
ows-0.050
cases-0.050
Token to
Token ID284
Feature activation-2.90e-02

Pos logit contributions
iling+0.306
hing+0.280
Playing+0.275
e+0.274
 years+0.274
Neg logit contributions
ometer-0.067
ipop-0.063
 off-0.036
 goodbye-0.025
 lid-0.025
Token play
Token ID711
Feature activation-1.37e-01

Pos logit contributions
iling+0.037
Playing+0.037
Fighting+0.035
Leave+0.030
 owning+0.030
Neg logit contributions
ipop-0.067
ometer-0.064
 lid-0.060
 off-0.057
cases-0.056
Token outside
Token ID2354
Feature activation-3.67e-02

Pos logit contributions
iling+0.251
Playing+0.245
Fighting+0.243
Leave+0.228
years+0.228
Neg logit contributions
ipop-0.217
ometer-0.216
 lid-0.208
 off-0.197
cases-0.174
Token saw
Token ID2497
Feature activation-4.81e-02

Pos logit contributions
Playing+0.145
iling+0.140
Fighting+0.136
 sports+0.128
 owning+0.127
Neg logit contributions
ometer-0.152
ipop-0.151
 lid-0.138
cases-0.125
 off-0.119
Token what
Token ID644
Feature activation-1.57e-02

Pos logit contributions
Playing+0.078
iling+0.075
Fighting+0.073
hing+0.070
Leave+0.069
Neg logit contributions
ipop-0.091
ometer-0.090
 lid-0.084
 off-0.081
cases-0.079
Token happened
Token ID3022
Feature activation-2.81e-02

Pos logit contributions
Playing+0.030
iling+0.029
Fighting+0.028
hing+0.027
Leave+0.027
Neg logit contributions
ometer-0.026
ipop-0.026
 lid-0.023
cases-0.021
 off-0.020
Token.
Token ID13
Feature activation-7.31e-03

Pos logit contributions
Playing+0.058
iling+0.057
Fighting+0.055
hing+0.053
Leave+0.053
Neg logit contributions
ometer-0.042
ipop-0.041
 lid-0.036
cases-0.034
 off-0.031
Token Those
Token ID5845
Feature activation-2.61e-02

Pos logit contributions
Playing+0.012
iling+0.012
Fighting+0.011
Leave+0.011
Why+0.011
Neg logit contributions
ipop-0.014
ometer-0.013
 lid-0.012
cases-0.012
atively-0.011
Token kids
Token ID3988
Feature activation-2.09e-01

Pos logit contributions
Playing+0.046
iling+0.045
Fighting+0.043
 supporting+0.041
years+0.040
Neg logit contributions
ipop-0.045
ometer-0.044
 lid-0.043
cases-0.043
 off-0.039
Token were
Token ID547
Feature activation-8.62e-02

Pos logit contributions
Playing+0.424
Leave+0.414
Fighting+0.413
years+0.399
Now+0.396
Neg logit contributions
ipop-0.306
ometer-0.289
 lid-0.261
cases-0.259
ows-0.240
Token mean
Token ID1612
Feature activation-9.24e-02

Pos logit contributions
Playing+0.152
iling+0.149
Fighting+0.140
hing+0.138
Leave+0.135
Neg logit contributions
ometer-0.153
ipop-0.152
 lid-0.138
cases-0.133
 off-0.133
Token.
Token ID13
Feature activation-4.61e-03

Pos logit contributions
Playing+0.184
iling+0.176
Fighting+0.172
Leave+0.167
hing+0.165
Neg logit contributions
ometer-0.142
ipop-0.139
 lid-0.127
cases-0.123
 off-0.117
Token They
Token ID1119
Feature activation-3.47e-02

Pos logit contributions
Playing+0.008
iling+0.008
Fighting+0.008
Leave+0.007
 sports+0.007
Neg logit contributions
ometer-0.008
ipop-0.008
 lid-0.008
cases-0.007
 off-0.006
Token took
Token ID1718
Feature activation+4.63e-02

Pos logit contributions
Playing+0.068
iling+0.064
 sports+0.062
Fighting+0.062
Now+0.060
Neg logit contributions
ipop-0.053
ometer-0.051
 lid-0.050
cases-0.047
 off-0.044
Token She
Token ID1375
Feature activation+4.67e-03

Pos logit contributions
ometer+0.034
ipop+0.033
 lid+0.031
cases+0.028
 off+0.026
Neg logit contributions
Playing-0.031
iling-0.029
Fighting-0.028
 sports-0.027
 owning-0.027
Token said
Token ID531
Feature activation+1.21e-01

Pos logit contributions
ipop+0.006
ometer+0.006
 lid+0.005
cases+0.005
 off+0.004
Neg logit contributions
iling-0.010
Playing-0.010
Fighting-0.010
hing-0.009
Leave-0.009
Token,
Token ID11
Feature activation-1.21e-02

Pos logit contributions
ipop+0.191
ometer+0.191
 lid+0.169
cases+0.155
 off+0.145
Neg logit contributions
Playing-0.240
iling-0.238
Fighting-0.226
hing-0.216
Leave-0.216
Token "
Token ID366
Feature activation-1.71e-01

Pos logit contributions
Playing+0.025
iling+0.024
Fighting+0.023
 sports+0.022
 owning+0.022
Neg logit contributions
ometer-0.018
ipop-0.018
 lid-0.016
cases-0.015
 off-0.013
TokenHi
Token ID17250
Feature activation-3.22e-02

Pos logit contributions
iling+0.243
 years+0.224
 owning+0.217
 repairing+0.217
e+0.203
Neg logit contributions
ipop-0.339
ometer-0.327
cases-0.304
 lid-0.296
 off-0.278
Token kids
Token ID3988
Feature activation-2.09e-01

Pos logit contributions
Playing+0.066
iling+0.065
Fighting+0.064
hing+0.062
Leave+0.062
Neg logit contributions
ipop-0.048
ometer-0.047
 lid-0.043
cases-0.039
 off-0.037
Token,
Token ID11
Feature activation-4.91e-02

Pos logit contributions
Playing+0.554
Leave+0.549
Fighting+0.542
Now+0.530
hing+0.520
Neg logit contributions
ipop-0.187
ometer-0.165
 lid-0.113
 off-0.104
 mouth-0.102
Token I
Token ID314
Feature activation-9.62e-02

Pos logit contributions
iling+0.104
Playing+0.099
 years+0.095
Fighting+0.094
Now+0.093
Neg logit contributions
ipop-0.074
ometer-0.067
 lid-0.063
 off-0.058
 goodbye-0.056
Token have
Token ID423
Feature activation-5.82e-02

Pos logit contributions
iling+0.180
Playing+0.177
Fighting+0.176
Leave+0.164
 sidelines+0.164
Neg logit contributions
ipop-0.156
ometer-0.149
 lid-0.129
ouse-0.128
 name-0.124
Token a
Token ID257
Feature activation+7.85e-02

Pos logit contributions
iling+0.121
Playing+0.121
hing+0.116
Fighting+0.115
Leave+0.108
Neg logit contributions
ipop-0.083
ometer-0.080
 lid-0.074
cases-0.068
 off-0.067
Token magic
Token ID5536
Feature activation+6.72e-02

Pos logit contributions
ometer+0.146
ipop+0.144
 lid+0.132
cases+0.126
atively+0.116
Neg logit contributions
Playing-0.131
iling-0.125
Fighting-0.121
 sports-0.119
 owning-0.117
TokenTim
Token ID14967
Feature activation-9.73e-02

Pos logit contributions
iling+0.155
Playing+0.154
Fighting+0.151
 disl+0.147
Leave+0.144
Neg logit contributions
ometer-0.104
ipop-0.103
 lid-0.099
 off-0.079
cases-0.075
Token and
Token ID290
Feature activation-1.12e-02

Pos logit contributions
Playing+0.187
Fighting+0.180
Leave+0.174
iling+0.174
 sports+0.165
Neg logit contributions
ipop-0.164
ometer-0.157
 lid-0.139
 off-0.127
cases-0.118
Token Mia
Token ID32189
Feature activation+6.49e-03

Pos logit contributions
Playing+0.017
Leave+0.015
 town+0.015
e+0.015
 sports+0.015
Neg logit contributions
ometer-0.023
ipop-0.022
 lid-0.021
cases-0.021
atively-0.019
Token are
Token ID389
Feature activation-6.74e-02

Pos logit contributions
ipop+0.008
 lid+0.007
 closes+0.007
 off+0.007
ometer+0.006
Neg logit contributions
Fighting-0.014
Playing-0.014
years-0.014
 sidelines-0.013
iling-0.013
Token adventurous
Token ID39770
Feature activation-1.43e-01

Pos logit contributions
Playing+0.083
iling+0.078
Fighting+0.072
 sports+0.071
 owning+0.070
Neg logit contributions
ometer-0.161
ipop-0.158
 lid-0.149
cases-0.140
atively-0.135
Token kids
Token ID3988
Feature activation-2.07e-01

Pos logit contributions
Playing+0.268
iling+0.260
Fighting+0.253
hing+0.240
Leave+0.240
Neg logit contributions
ipop-0.245
ometer-0.241
 lid-0.220
cases-0.198
 off-0.192
Token.
Token ID13
Feature activation+1.64e-03

Pos logit contributions
Playing+0.436
iling+0.435
Fighting+0.424
Leave+0.415
hing+0.412
Neg logit contributions
ipop-0.285
ometer-0.279
 lid-0.231
cases-0.217
 off-0.216
Token They
Token ID1119
Feature activation-4.11e-02

Pos logit contributions
ometer+0.002
ipop+0.002
 lid+0.002
cases+0.002
 off+0.002
Neg logit contributions
Playing-0.003
iling-0.003
Fighting-0.003
Leave-0.003
 owning-0.003
Token like
Token ID588
Feature activation-1.04e-01

Pos logit contributions
iling+0.071
Fighting+0.070
Playing+0.069
Leave+0.064
ting+0.064
Neg logit contributions
ipop-0.066
 name-0.063
ometer-0.061
 lid-0.058
 off-0.057
Token to
Token ID284
Feature activation-3.08e-02

Pos logit contributions
iling+0.308
Playing+0.286
hing+0.281
 years+0.279
Fighting+0.279
Neg logit contributions
ometer-0.076
ipop-0.065
 off-0.048
 lid-0.037
cases-0.033
Token explore
Token ID7301
Feature activation-6.50e-02

Pos logit contributions
Playing+0.050
iling+0.049
Fighting+0.048
years+0.044
ting+0.042
Neg logit contributions
ipop-0.058
ometer-0.056
 lid-0.054
 off-0.052
 name-0.050
Token loved
Token ID6151
Feature activation-1.30e-01

Pos logit contributions
Fighting+0.045
iling+0.045
Playing+0.043
Leave+0.041
hing+0.041
Neg logit contributions
ipop-0.046
ometer-0.043
 lid-0.039
ouse-0.037
cases-0.037
Token to
Token ID284
Feature activation-4.05e-02

Pos logit contributions
iling+0.340
Playing+0.333
hing+0.324
Fighting+0.323
Leave+0.316
Neg logit contributions
ipop-0.125
ometer-0.113
 off-0.096
 lid-0.088
 eyes-0.083
Token make
Token ID787
Feature activation-2.91e-02

Pos logit contributions
Playing+0.068
iling+0.065
Fighting+0.064
years+0.057
 years+0.056
Neg logit contributions
ipop-0.078
ometer-0.070
 lid-0.070
 off-0.068
 goodbye-0.064
Token all
Token ID477
Feature activation-4.06e-02

Pos logit contributions
Playing+0.050
iling+0.048
Fighting+0.046
Leave+0.044
Why+0.043
Neg logit contributions
ometer-0.052
ipop-0.052
 lid-0.049
cases-0.046
 off-0.045
Token the
Token ID262
Feature activation-6.37e-02

Pos logit contributions
Playing+0.070
iling+0.065
Fighting+0.064
Leave+0.063
assing+0.059
Neg logit contributions
ipop-0.072
ometer-0.067
 off-0.066
 lid-0.064
 mouth-0.056
Token kids
Token ID3988
Feature activation-2.07e-01

Pos logit contributions
Playing+0.126
iling+0.118
Fighting+0.114
Leave+0.112
Why+0.109
Neg logit contributions
 lid-0.102
ipop-0.091
ometer-0.089
 off-0.084
 eyes-0.083
Token laugh
Token ID6487
Feature activation-3.23e-02

Pos logit contributions
iling+0.370
Playing+0.367
Fighting+0.355
Leave+0.350
Now+0.348
Neg logit contributions
ipop-0.356
ometer-0.336
cases-0.312
 off-0.308
 lid-0.303
Token.
Token ID13
Feature activation+4.67e-03

Pos logit contributions
Playing+0.063
iling+0.062
Fighting+0.060
Leave+0.059
 disl+0.056
Neg logit contributions
ometer-0.051
ipop-0.049
 lid-0.046
cases-0.044
 off-0.044
Token One
Token ID1881
Feature activation+1.00e-01

Pos logit contributions
ometer+0.006
ipop+0.006
 lid+0.005
cases+0.005
 off+0.004
Neg logit contributions
Playing-0.011
iling-0.010
Fighting-0.010
 sports-0.010
 owning-0.010
Token day
Token ID1110
Feature activation-2.08e-02

Pos logit contributions
ipop+0.191
ometer+0.187
 lid+0.167
 off+0.148
cases+0.147
Neg logit contributions
Playing-0.170
iling-0.166
Fighting-0.166
Leave-0.156
 owning-0.149
Token,
Token ID11
Feature activation-2.79e-02

Pos logit contributions
Playing+0.036
iling+0.034
Fighting+0.032
 sports+0.032
 owning+0.032
Neg logit contributions
ometer-0.038
ipop-0.037
 lid-0.034
cases-0.032
atively-0.029
Token to
Token ID284
Feature activation-3.33e-02

Pos logit contributions
iling+0.279
hing+0.264
Playing+0.258
e+0.257
Fighting+0.250
Neg logit contributions
ometer-0.068
ipop-0.064
 off-0.058
 lid-0.047
 eyes-0.040
Token make
Token ID787
Feature activation-2.67e-02

Pos logit contributions
Playing+0.059
iling+0.058
Fighting+0.054
years+0.051
 owning+0.051
Neg logit contributions
ipop-0.059
ometer-0.058
 lid-0.054
 off-0.050
cases-0.049
Token fun
Token ID1257
Feature activation-1.46e-01

Pos logit contributions
Playing+0.047
iling+0.046
Fighting+0.043
Leave+0.042
hing+0.041
Neg logit contributions
ometer-0.048
ipop-0.047
 lid-0.044
cases-0.040
 off-0.039
Token of
Token ID286
Feature activation-1.18e-01

Pos logit contributions
Playing+0.283
iling+0.276
Fighting+0.268
Leave+0.259
years+0.254
Neg logit contributions
ometer-0.228
ipop-0.222
 lid-0.212
cases-0.197
 off-0.189
Token other
Token ID584
Feature activation-1.09e-01

Pos logit contributions
iling+0.232
years+0.216
Now+0.209
vous+0.207
hing+0.207
Neg logit contributions
 eyes-0.176
 off-0.171
 mouth-0.167
 lid-0.164
 arms-0.154
Token kids
Token ID3988
Feature activation-2.07e-01

Pos logit contributions
Playing+0.187
Now+0.185
iling+0.183
Fighting+0.181
years+0.179
Neg logit contributions
 eyes-0.169
 lid-0.168
ipop-0.157
 off-0.155
cases-0.155
Token.
Token ID13
Feature activation+2.76e-03

Pos logit contributions
iling+0.492
Now+0.489
Playing+0.478
Fighting+0.478
Leave+0.477
Neg logit contributions
ipop-0.231
ometer-0.211
cases-0.199
 off-0.175
ows-0.170
Token They
Token ID1119
Feature activation-3.74e-02

Pos logit contributions
ipop+0.005
ometer+0.005
 lid+0.004
cases+0.004
ouse+0.004
Neg logit contributions
Playing-0.005
iling-0.005
Fighting-0.005
Why-0.005
 town-0.004
Token saw
Token ID2497
Feature activation-5.40e-02

Pos logit contributions
iling+0.077
Fighting+0.069
Playing+0.068
Now+0.067
 sports+0.067
Neg logit contributions
ipop-0.058
ometer-0.054
 lid-0.049
cases-0.047
atively-0.043
Token a
Token ID257
Feature activation+6.96e-02

Pos logit contributions
iling+0.124
Playing+0.114
Fighting+0.109
hing+0.108
e+0.106
Neg logit contributions
ipop-0.068
ometer-0.067
 lid-0.066
 off-0.059
 eyes-0.056
Token ch
Token ID442
Feature activation+1.70e-02

Pos logit contributions
ometer+0.110
ipop+0.108
 lid+0.103
cases+0.085
ouse+0.081
Neg logit contributions
iling-0.142
Playing-0.139
Fighting-0.132
hing-0.127
Leave-0.126
Token
Token ID198
Feature activation-1.86e-02

Pos logit contributions
ipop+0.024
ometer+0.024
 lid+0.022
cases+0.020
ouse+0.019
Neg logit contributions
Playing-0.022
Fighting-0.020
iling-0.020
 sports-0.019
Why-0.019
Token
Token ID198
Feature activation-2.55e-02

Pos logit contributions
Playing+0.031
iling+0.028
Fighting+0.028
 town+0.027
Why+0.027
Neg logit contributions
ipop-0.037
ometer-0.036
 lid-0.033
cases-0.030
atively-0.029
Token"
Token ID1
Feature activation-1.64e-01

Pos logit contributions
iling+0.050
Playing+0.047
 owning+0.047
Fighting+0.047
hing+0.046
Neg logit contributions
ometer-0.039
ipop-0.038
 lid-0.035
cases-0.034
 off-0.030
TokenHello
Token ID15496
Feature activation-2.44e-02

Pos logit contributions
iling+0.270
 years+0.243
Playing+0.239
 owning+0.236
Fighting+0.229
Neg logit contributions
ometer-0.311
ipop-0.307
 lid-0.298
cases-0.271
atively-0.254
Token,
Token ID11
Feature activation-5.60e-02

Pos logit contributions
Playing+0.041
iling+0.041
 sports+0.038
 owning+0.037
Fighting+0.037
Neg logit contributions
ometer-0.044
ipop-0.044
 lid-0.040
cases-0.037
atively-0.035
Token kids
Token ID3988
Feature activation-2.07e-01

Pos logit contributions
Playing+0.099
iling+0.098
Fighting+0.092
Leave+0.087
 years+0.087
Neg logit contributions
ometer-0.102
ipop-0.100
 lid-0.095
cases-0.087
 off-0.081
Token,"
Token ID553
Feature activation-2.94e-02

Pos logit contributions
Playing+0.510
Fighting+0.497
Leave+0.493
iling+0.473
Now+0.465
Neg logit contributions
ipop-0.228
ometer-0.227
 lid-0.186
 off-0.153
cases-0.152
Token the
Token ID262
Feature activation-5.00e-02

Pos logit contributions
Playing+0.060
iling+0.058
Fighting+0.056
 sports+0.054
Leave+0.054
Neg logit contributions
ometer-0.045
ipop-0.045
 lid-0.040
cases-0.036
 off-0.033
Token truck
Token ID7779
Feature activation+7.07e-04

Pos logit contributions
iling+0.115
Playing+0.110
Fighting+0.103
 owning+0.101
 years+0.101
Neg logit contributions
ometer-0.069
 lid-0.064
ipop-0.063
cases-0.056
 door-0.049
Token driver
Token ID4639
Feature activation-5.26e-02

Pos logit contributions
ipop+0.001
ometer+0.001
 lid+0.001
rily+0.001
cases+0.001
Neg logit contributions
Playing-0.001
 town-0.001
Why-0.001
Fighting-0.001
 years-0.001
Token said
Token ID531
Feature activation+1.11e-01

Pos logit contributions
Playing+0.098
iling+0.092
e+0.090
 years+0.088
Fighting+0.087
Neg logit contributions
ometer-0.088
ipop-0.083
 lid-0.081
cases-0.069
atively-0.065
Token<|endoftext|>
Token ID50256
Feature activation-7.51e-02

Pos logit contributions
ipop+0.027
ometer+0.027
 lid+0.024
cases+0.022
atively+0.021
Neg logit contributions
Playing-0.028
iling-0.027
Fighting-0.026
 sports-0.025
Why-0.025
TokenSam
Token ID16305
Feature activation-6.16e-02

Pos logit contributions
Playing+0.153
iling+0.152
Fighting+0.150
 disl+0.145
Leave+0.144
Neg logit contributions
ometer-0.105
ipop-0.105
 lid-0.099
 off-0.079
cases-0.078
Token and
Token ID290
Feature activation-1.57e-03

Pos logit contributions
Fighting+0.136
Playing+0.133
Leave+0.130
 sidelines+0.129
years+0.125
Neg logit contributions
ipop-0.081
ometer-0.069
 lid-0.062
 off-0.061
 reluctantly-0.061
Token Ben
Token ID3932
Feature activation-3.31e-02

Pos logit contributions
Playing+0.002
iling+0.002
Fighting+0.002
 sports+0.002
Leave+0.002
Neg logit contributions
ometer-0.003
ipop-0.003
 lid-0.003
cases-0.003
atively-0.003
Token are
Token ID389
Feature activation-6.48e-02

Pos logit contributions
Fighting+0.070
iling+0.067
years+0.067
 sidelines+0.066
Playing+0.064
Neg logit contributions
ipop-0.040
 lid-0.035
 name-0.034
 reluctantly-0.033
ometer-0.033
Token kids
Token ID3988
Feature activation-2.06e-01

Pos logit contributions
Playing+0.099
iling+0.095
Fighting+0.090
 sports+0.085
 owning+0.085
Neg logit contributions
ometer-0.134
ipop-0.133
 lid-0.123
cases-0.115
atively-0.107
Token who
Token ID508
Feature activation-3.08e-02

Pos logit contributions
Playing+0.407
iling+0.404
Fighting+0.392
Leave+0.381
hing+0.378
Neg logit contributions
ometer-0.314
ipop-0.311
 lid-0.265
cases-0.250
 off-0.243
Token like
Token ID588
Feature activation-9.68e-02

Pos logit contributions
iling+0.051
Fighting+0.049
Leave+0.048
Playing+0.047
years+0.046
Neg logit contributions
ipop-0.053
ometer-0.051
 name-0.048
ows-0.044
cases-0.043
Token to
Token ID284
Feature activation-2.85e-02

Pos logit contributions
iling+0.276
Playing+0.255
hing+0.253
Fighting+0.252
 years+0.249
Neg logit contributions
ipop-0.081
ometer-0.080
 off-0.048
 lid-0.045
cases-0.039
Token play
Token ID711
Feature activation-1.37e-01

Pos logit contributions
iling+0.037
Playing+0.037
Fighting+0.035
 years+0.031
Leave+0.031
Neg logit contributions
ipop-0.065
ometer-0.062
 lid-0.058
 off-0.054
cases-0.054
Token with
Token ID351
Feature activation+3.10e-03

Pos logit contributions
iling+0.248
Playing+0.242
Fighting+0.240
Leave+0.223
 disl+0.221
Neg logit contributions
ipop-0.224
ometer-0.219
 lid-0.210
 off-0.197
atively-0.179
Tokenila
Token ID10102
Feature activation-5.27e-02

Pos logit contributions
iling+0.112
Playing+0.108
e+0.102
 years+0.101
 town+0.098
Neg logit contributions
ometer-0.177
ipop-0.161
 lid-0.151
cases-0.144
ouse-0.135
Token and
Token ID290
Feature activation-5.52e-03

Pos logit contributions
Fighting+0.122
Playing+0.122
 sidelines+0.119
years+0.119
Leave+0.115
Neg logit contributions
ipop-0.058
ometer-0.055
 lid-0.049
 closes-0.048
 off-0.042
Token Tom
Token ID4186
Feature activation-6.15e-02

Pos logit contributions
Playing+0.009
iling+0.008
Fighting+0.008
 sports+0.008
Leave+0.008
Neg logit contributions
ometer-0.011
ipop-0.011
 lid-0.010
cases-0.009
 off-0.009
Token were
Token ID547
Feature activation-6.65e-02

Pos logit contributions
Fighting+0.129
Playing+0.124
 sidelines+0.124
years+0.122
iling+0.120
Neg logit contributions
ipop-0.079
 lid-0.069
ometer-0.068
 closes-0.064
 off-0.063
Token adventurous
Token ID39770
Feature activation-1.41e-01

Pos logit contributions
Playing+0.107
iling+0.103
Fighting+0.097
 sports+0.095
 owning+0.094
Neg logit contributions
ometer-0.132
ipop-0.130
 lid-0.120
cases-0.112
atively-0.106
Token kids
Token ID3988
Feature activation-2.06e-01

Pos logit contributions
Playing+0.276
iling+0.263
Fighting+0.260
hing+0.248
Leave+0.246
Neg logit contributions
ipop-0.231
ometer-0.227
 lid-0.205
cases-0.184
 off-0.179
Token.
Token ID13
Feature activation+8.89e-04

Pos logit contributions
Playing+0.396
iling+0.392
Fighting+0.382
Leave+0.376
hing+0.372
Neg logit contributions
ipop-0.321
ometer-0.320
 lid-0.267
cases-0.254
 off-0.248
Token They
Token ID1119
Feature activation-4.07e-02

Pos logit contributions
ometer+0.002
ipop+0.002
 lid+0.001
cases+0.001
 off+0.001
Neg logit contributions
Playing-0.002
iling-0.002
Why-0.002
Fighting-0.001
 sports-0.001
Token liked
Token ID8288
Feature activation-1.01e-01

Pos logit contributions
iling+0.077
Fighting+0.073
ting+0.069
Leave+0.069
Playing+0.068
Neg logit contributions
ipop-0.064
ometer-0.060
 name-0.057
 lid-0.054
atively-0.053
Token to
Token ID284
Feature activation-3.78e-02

Pos logit contributions
iling+0.301
Playing+0.282
hing+0.281
Leave+0.277
Fighting+0.276
Neg logit contributions
ometer-0.066
ipop-0.054
 off-0.049
 lid-0.035
 eyes-0.031
Token explore
Token ID7301
Feature activation-6.84e-02

Pos logit contributions
iling+0.061
Playing+0.061
Fighting+0.059
years+0.054
ting+0.052
Neg logit contributions
ipop-0.071
ometer-0.070
 lid-0.067
 off-0.064
 name-0.060
Token
Token ID198
Feature activation-2.13e-02

Pos logit contributions
ipop+0.013
ometer+0.013
 lid+0.012
cases+0.011
 off+0.010
Neg logit contributions
Playing-0.014
iling-0.013
Fighting-0.012
 sports-0.012
 town-0.012
Token
Token ID198
Feature activation-2.76e-02

Pos logit contributions
Playing+0.039
iling+0.037
Fighting+0.036
 sports+0.034
 owning+0.034
Neg logit contributions
ometer-0.037
ipop-0.037
 lid-0.034
cases-0.031
 off-0.029
TokenThe
Token ID464
Feature activation-3.56e-02

Pos logit contributions
iling+0.063
ting+0.059
 owning+0.059
 disl+0.057
 arrest+0.057
Neg logit contributions
ometer-0.027
 lid-0.024
ipop-0.023
cases-0.021
atively-0.020
Token giant
Token ID6175
Feature activation-4.35e-02

Pos logit contributions
Playing+0.068
iling+0.067
Fighting+0.064
hing+0.061
Leave+0.061
Neg logit contributions
ipop-0.060
 lid-0.058
ometer-0.058
 door-0.047
cases-0.046
Token loved
Token ID6151
Feature activation-1.22e-01

Pos logit contributions
Fighting+0.081
Playing+0.080
iling+0.079
Leave+0.076
hing+0.075
Neg logit contributions
ipop-0.072
ometer-0.069
 lid-0.062
cases-0.057
 off-0.054
Token children
Token ID1751
Feature activation-2.06e-01

Pos logit contributions
iling+0.315
Playing+0.303
hing+0.299
Leave+0.290
Fighting+0.290
Neg logit contributions
ipop-0.123
ometer-0.117
 off-0.093
 lid-0.081
 eyes-0.081
Token and
Token ID290
Feature activation-1.95e-02

Pos logit contributions
iling+0.437
Now+0.427
Playing+0.422
Fighting+0.419
Stop+0.418
Neg logit contributions
ipop-0.297
ometer-0.284
 off-0.241
 mouth-0.227
 goodbye-0.219
Token often
Token ID1690
Feature activation+7.71e-03

Pos logit contributions
iling+0.034
Playing+0.033
Fighting+0.031
Leave+0.031
hing+0.030
Neg logit contributions
ipop-0.035
ometer-0.034
 lid-0.030
cases-0.029
 off-0.028
Token gave
Token ID2921
Feature activation+4.50e-02

Pos logit contributions
ipop+0.012
ometer+0.012
 lid+0.010
cases+0.010
 off+0.010
Neg logit contributions
Playing-0.015
Fighting-0.014
iling-0.014
Leave-0.014
Now-0.013
Token them
Token ID606
Feature activation+3.02e-02

Pos logit contributions
ometer+0.094
ipop+0.093
 lid+0.086
cases+0.081
 off+0.079
Neg logit contributions
Playing-0.066
iling-0.064
Fighting-0.061
Leave-0.058
 owning-0.056
Token everything
Token ID2279
Feature activation+7.18e-03

Pos logit contributions
ometer+0.051
ipop+0.051
 lid+0.046
cases+0.042
 off+0.040
Neg logit contributions
Playing-0.056
iling-0.055
Fighting-0.052
Leave-0.050
 owning-0.050
Token He
Token ID679
Feature activation-8.54e-03

Pos logit contributions
ometer+0.029
ipop+0.028
 lid+0.025
cases+0.023
 off+0.021
Neg logit contributions
Playing-0.039
iling-0.038
Fighting-0.036
 sports-0.035
 owning-0.035
Token wanted
Token ID2227
Feature activation-5.42e-02

Pos logit contributions
Playing+0.016
iling+0.016
Fighting+0.015
Leave+0.014
hing+0.014
Neg logit contributions
ipop-0.014
ometer-0.014
 lid-0.013
cases-0.012
 off-0.011
Token to
Token ID284
Feature activation-3.22e-02

Pos logit contributions
Playing+0.132
iling+0.131
Fighting+0.124
hing+0.123
Leave+0.121
Neg logit contributions
ipop-0.061
ometer-0.061
 lid-0.052
 off-0.048
cases-0.045
Token make
Token ID787
Feature activation-2.59e-02

Pos logit contributions
Playing+0.063
iling+0.059
Fighting+0.057
years+0.055
 owning+0.052
Neg logit contributions
ipop-0.052
ometer-0.051
 lid-0.049
 off-0.047
 reluctantly-0.041
Token the
Token ID262
Feature activation-6.85e-02

Pos logit contributions
Playing+0.045
iling+0.044
Fighting+0.042
Leave+0.040
 owning+0.040
Neg logit contributions
ometer-0.047
ipop-0.046
 lid-0.043
cases-0.039
 off-0.037
Token kids
Token ID3988
Feature activation-2.06e-01

Pos logit contributions
Playing+0.127
iling+0.120
Fighting+0.114
Leave+0.112
hing+0.110
Neg logit contributions
ipop-0.118
ometer-0.115
 lid-0.114
cases-0.095
 mouth-0.093
Token happy
Token ID3772
Feature activation-1.07e-01

Pos logit contributions
Playing+0.389
iling+0.383
Fighting+0.356
Now+0.355
Leave+0.342
Neg logit contributions
ipop-0.342
ometer-0.327
 lid-0.298
 off-0.290
cases-0.290
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
Playing+0.227
iling+0.216
Fighting+0.213
Leave+0.207
hing+0.203
Neg logit contributions
ipop-0.155
ometer-0.151
 lid-0.138
 off-0.133
cases-0.123
Token
Token ID198
Feature activation-1.42e-02

Pos logit contributions
ipop+0.019
ometer+0.019
 lid+0.017
cases+0.016
 off+0.014
Neg logit contributions
Playing-0.023
iling-0.022
Fighting-0.022
 sports-0.021
 town-0.021
Token
Token ID198
Feature activation-2.04e-02

Pos logit contributions
Playing+0.025
iling+0.023
Fighting+0.022
 sports+0.021
 town+0.021
Neg logit contributions
ipop-0.027
ometer-0.027
 lid-0.024
cases-0.023
atively-0.021
TokenOne
Token ID3198
Feature activation+8.00e-02

Pos logit contributions
iling+0.042
 owning+0.039
Playing+0.038
ting+0.038
hing+0.037
Neg logit contributions
ometer-0.027
 lid-0.025
ipop-0.024
cases-0.023
atively-0.021
Token kids
Token ID3988
Feature activation-1.94e-01

Pos logit contributions
Playing+0.112
iling+0.109
Fighting+0.104
 sports+0.100
Leave+0.100
Neg logit contributions
ometer-0.121
ipop-0.116
 lid-0.107
cases-0.099
atively-0.092
Token.
Token ID13
Feature activation-1.28e-02

Pos logit contributions
Playing+0.463
Fighting+0.441
Leave+0.439
iling+0.438
hing+0.425
Neg logit contributions
ometer-0.227
ipop-0.227
 lid-0.180
cases-0.162
 off-0.158
Token You
Token ID921
Feature activation-1.70e-02

Pos logit contributions
Playing+0.023
iling+0.023
Fighting+0.021
hing+0.020
 sports+0.020
Neg logit contributions
ometer-0.023
ipop-0.022
 lid-0.020
cases-0.018
 off-0.017
Token are
Token ID389
Feature activation-7.84e-02

Pos logit contributions
Playing+0.032
iling+0.032
Fighting+0.030
 sports+0.030
Leave+0.029
Neg logit contributions
 lid-0.026
ipop-0.026
ometer-0.025
 off-0.023
icked-0.022
Token good
Token ID922
Feature activation-1.09e-01

Pos logit contributions
iling+0.167
Playing+0.162
hing+0.152
Fighting+0.151
Leave+0.150
Neg logit contributions
ometer-0.098
 off-0.096
ipop-0.096
 lid-0.094
 mouth-0.080
Token kids
Token ID3988
Feature activation-2.06e-01

Pos logit contributions
Playing+0.170
iling+0.163
Fighting+0.156
hing+0.149
Leave+0.148
Neg logit contributions
ometer-0.219
ipop-0.218
 lid-0.199
cases-0.186
 off-0.180
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
Playing+0.433
iling+0.417
Fighting+0.408
Leave+0.397
hing+0.391
Neg logit contributions
ometer-0.302
ipop-0.298
 lid-0.258
cases-0.243
 off-0.224
Token You
Token ID921
Feature activation-1.60e-02

Pos logit contributions
iling+0.021
Playing+0.021
Fighting+0.020
hing+0.020
 sports+0.019
Neg logit contributions
ometer-0.021
ipop-0.020
 lid-0.018
cases-0.016
 off-0.015
Token are
Token ID389
Feature activation-7.67e-02

Pos logit contributions
Playing+0.031
iling+0.030
Fighting+0.029
 sports+0.029
Leave+0.028
Neg logit contributions
ipop-0.023
 lid-0.023
ometer-0.023
 off-0.021
icked-0.020
Token smart
Token ID4451
Feature activation-1.21e-01

Pos logit contributions
iling+0.167
Playing+0.162
hing+0.152
Fighting+0.151
Leave+0.150
Neg logit contributions
 off-0.093
ometer-0.092
ipop-0.090
 lid-0.087
 mouth-0.075
Token.
Token ID13
Feature activation-1.30e-02

Pos logit contributions
Playing+0.245
iling+0.235
Fighting+0.229
hing+0.221
Leave+0.219
Neg logit contributions
ometer-0.183
ipop-0.181
 lid-0.166
cases-0.150
 off-0.143
Token how
Token ID703
Feature activation+1.29e-02

Pos logit contributions
ometer+0.073
ipop+0.073
 lid+0.067
cases+0.062
 off+0.059
Neg logit contributions
Playing-0.068
iling-0.067
Fighting-0.064
Leave-0.061
 sports-0.060
Token to
Token ID284
Feature activation-9.28e-03

Pos logit contributions
ometer+0.019
ipop+0.018
 lid+0.016
cases+0.014
 off+0.014
Neg logit contributions
Playing-0.028
iling-0.027
Fighting-0.026
hing-0.026
Leave-0.025
Token do
Token ID466
Feature activation-1.16e-01

Pos logit contributions
Playing+0.018
Fighting+0.018
iling+0.017
Leave+0.017
years+0.017
Neg logit contributions
ometer-0.014
ipop-0.014
 lid-0.014
 off-0.012
cases-0.012
Token it
Token ID340
Feature activation+7.04e-02

Pos logit contributions
Playing+0.252
iling+0.248
Fighting+0.238
Leave+0.230
hing+0.229
Neg logit contributions
ometer-0.159
ipop-0.156
 lid-0.143
cases-0.130
 off-0.124
Token.
Token ID13
Feature activation+8.69e-03

Pos logit contributions
ometer+0.112
ipop+0.109
 lid+0.100
cases+0.092
 off+0.088
Neg logit contributions
Playing-0.140
iling-0.136
Fighting-0.131
Leave-0.127
 years-0.123
Token
Token ID220
Feature activation+1.48e-01

Pos logit contributions
ipop+0.018
ometer+0.016
 lid+0.016
cases+0.015
atively+0.014
Neg logit contributions
Playing-0.014
 town-0.012
Fighting-0.012
Why-0.012
 Playing-0.012
Token
Token ID198
Feature activation-1.63e-02

Pos logit contributions
ipop+0.322
ometer+0.306
 lid+0.290
cases+0.266
atively+0.257
Neg logit contributions
Playing-0.220
Fighting-0.191
iling-0.187
 town-0.181
Why-0.180
Token
Token ID198
Feature activation-2.26e-02

Pos logit contributions
Playing+0.026
iling+0.024
Fighting+0.024
 years+0.023
 town+0.023
Neg logit contributions
ipop-0.033
ometer-0.033
 lid-0.030
cases-0.028
atively-0.026
TokenThen
Token ID6423
Feature activation+1.54e-02

Pos logit contributions
iling+0.042
Playing+0.042
 owning+0.041
Fighting+0.040
hing+0.039
Neg logit contributions
ometer-0.036
ipop-0.034
 lid-0.033
cases-0.031
atively-0.027
Token,
Token ID11
Feature activation-2.43e-02

Pos logit contributions
ometer+0.036
ipop+0.034
 lid+0.032
cases+0.031
ows+0.030
Neg logit contributions
Playing-0.019
 playing-0.019
 town-0.019
 sports-0.018
 years-0.018
Token a
Token ID257
Feature activation+7.81e-02

Pos logit contributions
Playing+0.039
Why+0.036
 sports+0.036
iling+0.036
 town+0.036
Neg logit contributions
ometer-0.048
ipop-0.046
 lid-0.043
cases-0.040
ouse-0.038
Token so
Token ID523
Feature activation+7.13e-02

Pos logit contributions
ometer+0.234
ipop+0.226
 lid+0.210
cases+0.190
 off+0.189
Neg logit contributions
Playing-0.279
iling-0.269
Fighting-0.258
Leave-0.245
 sports-0.242
Token it
Token ID340
Feature activation+1.14e-01

Pos logit contributions
ometer+0.099
ipop+0.094
 lid+0.081
 off+0.077
cases+0.074
Neg logit contributions
iling-0.158
Playing-0.155
Fighting-0.149
hing-0.144
 owning-0.141
Token wouldn
Token ID3636
Feature activation+7.46e-02

Pos logit contributions
ipop+0.175
ometer+0.175
 lid+0.156
cases+0.139
atively+0.128
Neg logit contributions
Playing-0.227
iling-0.219
Fighting-0.211
Leave-0.204
hing-0.203
Token't
Token ID470
Feature activation+7.48e-02

Pos logit contributions
ometer+0.115
ipop+0.109
 lid+0.101
 off+0.092
cases+0.091
Neg logit contributions
Playing-0.154
iling-0.151
Fighting-0.143
 sports-0.141
 owning-0.136
Token fall
Token ID2121
Feature activation+4.31e-02

Pos logit contributions
ometer+0.124
ipop+0.121
 lid+0.111
 off+0.109
cases+0.100
Neg logit contributions
Playing-0.141
iling-0.136
Fighting-0.130
years-0.125
 sports-0.124
Token off
Token ID572
Feature activation+1.45e-01

Pos logit contributions
ometer+0.060
ipop+0.057
 lid+0.052
 off+0.048
cases+0.047
Neg logit contributions
Playing-0.095
iling-0.093
Fighting-0.089
 sports-0.085
 years-0.084
Token again
Token ID757
Feature activation+4.74e-02

Pos logit contributions
ometer+0.190
 off+0.185
ipop+0.184
 lid+0.175
cases+0.155
Neg logit contributions
Playing-0.328
iling-0.309
Fighting-0.299
hing-0.280
Leave-0.276
Token.
Token ID13
Feature activation+1.48e-02

Pos logit contributions
ometer+0.062
ipop+0.059
 lid+0.054
 off+0.053
cases+0.048
Neg logit contributions
Playing-0.106
iling-0.100
Fighting-0.099
hing-0.095
Leave-0.094
Token From
Token ID3574
Feature activation+1.32e-02

Pos logit contributions
ipop+0.025
ometer+0.025
 lid+0.023
cases+0.021
atively+0.019
Neg logit contributions
Playing-0.028
iling-0.026
Fighting-0.026
 sports-0.025
Leave-0.024
Token that
Token ID326
Feature activation-7.52e-03

Pos logit contributions
ipop+0.024
ometer+0.024
 lid+0.022
 off+0.021
cases+0.020
Neg logit contributions
Playing-0.022
iling-0.022
Fighting-0.021
hing-0.020
Leave-0.020
Token day
Token ID1110
Feature activation-2.35e-02

Pos logit contributions
Playing+0.011
iling+0.010
Fighting+0.010
 sports+0.009
Leave+0.009
Neg logit contributions
ometer-0.016
ipop-0.016
 lid-0.015
cases-0.014
 off-0.013
Token a
Token ID257
Feature activation+9.54e-02

Pos logit contributions
ometer+0.004
ipop+0.004
 lid+0.004
cases+0.003
 off+0.003
Neg logit contributions
Playing-0.004
iling-0.004
Fighting-0.004
 sports-0.003
hing-0.003
Token button
Token ID4936
Feature activation+1.59e-01

Pos logit contributions
ometer+0.155
ipop+0.154
 lid+0.137
cases+0.129
atively+0.117
Neg logit contributions
Playing-0.181
iling-0.175
Fighting-0.169
 sports-0.163
 owning-0.162
Token bracelet
Token ID42893
Feature activation+8.52e-02

Pos logit contributions
ometer+0.280
ipop+0.276
 lid+0.252
cases+0.231
atively+0.214
Neg logit contributions
Playing-0.282
iling-0.275
Fighting-0.262
 sports-0.251
 owning-0.250
Token,
Token ID11
Feature activation+8.93e-04

Pos logit contributions
ometer+0.152
ipop+0.150
 lid+0.137
cases+0.124
atively+0.118
Neg logit contributions
Playing-0.145
iling-0.142
Fighting-0.133
 owning-0.131
 sports-0.130
Token a
Token ID257
Feature activation+9.90e-02

Pos logit contributions
ometer+0.002
ipop+0.002
 lid+0.001
cases+0.001
 off+0.001
Neg logit contributions
Playing-0.002
iling-0.002
Fighting-0.001
 sports-0.001
hing-0.001
Token button
Token ID4936
Feature activation+1.62e-01

Pos logit contributions
ometer+0.163
ipop+0.161
 lid+0.143
cases+0.136
atively+0.124
Neg logit contributions
Playing-0.185
iling-0.178
Fighting-0.173
 sports-0.167
 owning-0.166
Token flower
Token ID15061
Feature activation+8.13e-02

Pos logit contributions
ometer+0.285
ipop+0.281
 lid+0.257
cases+0.235
atively+0.218
Neg logit contributions
Playing-0.286
iling-0.279
Fighting-0.265
 owning-0.255
 sports-0.254
Token,
Token ID11
Feature activation+3.25e-03

Pos logit contributions
ometer+0.142
ipop+0.141
 lid+0.128
cases+0.118
atively+0.109
Neg logit contributions
Playing-0.145
iling-0.141
Fighting-0.134
 sports-0.129
 owning-0.129
Token and
Token ID290
Feature activation+2.68e-02

Pos logit contributions
ometer+0.006
ipop+0.006
 lid+0.005
cases+0.005
 off+0.005
Neg logit contributions
Playing-0.006
iling-0.006
Fighting-0.005
 sports-0.005
hing-0.005
Token a
Token ID257
Feature activation+1.08e-01

Pos logit contributions
ometer+0.046
ipop+0.045
 lid+0.041
cases+0.038
 off+0.035
Neg logit contributions
Playing-0.049
iling-0.048
Fighting-0.046
 sports-0.044
Leave-0.044
Token button
Token ID4936
Feature activation+1.62e-01

Pos logit contributions
ometer+0.182
ipop+0.180
 lid+0.162
cases+0.151
 off+0.138
Neg logit contributions
Playing-0.201
iling-0.195
Fighting-0.187
 sports-0.180
 owning-0.179
Token Just
Token ID2329
Feature activation+1.35e-01

Pos logit contributions
Playing+0.006
iling+0.006
Fighting+0.006
 sports+0.006
Leave+0.006
Neg logit contributions
ometer-0.006
ipop-0.006
 lid-0.005
cases-0.005
 off-0.004
Token try
Token ID1949
Feature activation+8.19e-02

Pos logit contributions
ometer+0.226
ipop+0.225
 lid+0.204
 off+0.184
cases+0.175
Neg logit contributions
Playing-0.247
iling-0.241
Fighting-0.233
Leave-0.218
hing-0.218
Token your
Token ID534
Feature activation+3.71e-02

Pos logit contributions
ometer+0.126
ipop+0.123
 lid+0.110
cases+0.100
 off+0.097
Neg logit contributions
Playing-0.165
iling-0.163
Fighting-0.155
Leave-0.149
hing-0.149
Token best
Token ID1266
Feature activation-1.80e-02

Pos logit contributions
ipop+0.062
ometer+0.059
cases+0.053
atively+0.052
 lid+0.050
Neg logit contributions
Playing-0.068
 sports-0.067
Fighting-0.065
 playing-0.065
iling-0.064
Token."
Token ID526
Feature activation+2.89e-02

Pos logit contributions
Playing+0.028
 sports+0.028
iling+0.027
 playing+0.027
 owning+0.026
Neg logit contributions
ipop-0.034
ometer-0.034
 lid-0.032
cases-0.031
atively-0.029
Token
Token ID220
Feature activation+1.56e-01

Pos logit contributions
ometer+0.043
ipop+0.042
 lid+0.037
cases+0.035
atively+0.031
Neg logit contributions
Playing-0.061
Fighting-0.057
iling-0.056
Why-0.054
 sports-0.054
Token
Token ID198
Feature activation-5.32e-03

Pos logit contributions
ipop+0.325
ometer+0.307
 lid+0.296
cases+0.272
atively+0.259
Neg logit contributions
Playing-0.240
Fighting-0.215
iling-0.212
 town-0.198
 Playing-0.198
Token
Token ID198
Feature activation-1.15e-02

Pos logit contributions
iling+0.011
Playing+0.011
Fighting+0.010
 owning+0.010
hing+0.010
Neg logit contributions
ometer-0.008
ipop-0.008
 lid-0.007
cases-0.006
 off-0.006
TokenL
Token ID43
Feature activation-6.98e-02

Pos logit contributions
iling+0.020
 owning+0.020
hing+0.019
ting+0.019
 arrest+0.018
Neg logit contributions
ometer-0.018
 lid-0.017
ipop-0.017
cases-0.015
 reluctantly-0.015
Tokenily
Token ID813
Feature activation-4.66e-02

Pos logit contributions
Playing+0.126
Fighting+0.119
iling+0.118
Leave+0.115
hing+0.113
Neg logit contributions
ipop-0.123
ometer-0.115
 lid-0.108
cases-0.097
atively-0.092
Token took
Token ID1718
Feature activation+4.03e-02

Pos logit contributions
Playing+0.085
iling+0.084
Fighting+0.080
 owning+0.077
Leave+0.076
Neg logit contributions
ometer-0.080
ipop-0.079
 lid-0.072
cases-0.066
 off-0.062
Token a
Token ID257
Feature activation+9.79e-02

Pos logit contributions
ipop+0.096
ometer+0.095
 lid+0.083
cases+0.083
ouse+0.077
Neg logit contributions
Playing-0.060
iling-0.056
Fighting-0.055
e-0.054
hing-0.053
Token big
Token ID1263
Feature activation-2.12e-02

Pos logit contributions
ipop+0.177
ometer+0.169
cases+0.149
 lid+0.142
atively+0.135
Neg logit contributions
Playing-0.166
iling-0.159
Fighting-0.156
 town-0.154
 sports-0.153
Token hole
Token ID7604
Feature activation+9.25e-02

Pos logit contributions
Playing+0.040
iling+0.039
Fighting+0.037
 sports+0.036
hing+0.036
Neg logit contributions
ometer-0.035
ipop-0.035
 lid-0.030
cases-0.029
atively-0.026
Token in
Token ID287
Feature activation+4.50e-02

Pos logit contributions
ometer+0.200
ipop+0.197
 lid+0.180
cases+0.173
ouse+0.160
Neg logit contributions
Playing-0.126
iling-0.124
Fighting-0.112
 years-0.109
 owning-0.109
Token the
Token ID262
Feature activation+1.07e-02

Pos logit contributions
ometer+0.090
 lid+0.088
ipop+0.088
 off+0.078
cases+0.076
Neg logit contributions
Playing-0.071
iling-0.068
Fighting-0.063
hing-0.060
Leave-0.060
Token top
Token ID1353
Feature activation+1.55e-01

Pos logit contributions
ometer+0.015
ipop+0.014
 lid+0.014
cases+0.011
 off+0.010
Neg logit contributions
Playing-0.024
iling-0.024
Fighting-0.023
 owning-0.022
 sports-0.022
Token and
Token ID290
Feature activation+2.70e-02

Pos logit contributions
ometer+0.300
ipop+0.296
 lid+0.274
cases+0.254
 off+0.238
Neg logit contributions
Playing-0.252
iling-0.244
Fighting-0.233
 sports-0.220
Leave-0.220
Token puts
Token ID7584
Feature activation+5.97e-03

Pos logit contributions
ipop+0.047
ometer+0.047
 lid+0.040
cases+0.037
ouse+0.036
Neg logit contributions
Playing-0.049
iling-0.047
Fighting-0.045
Why-0.045
Leave-0.044
Token a
Token ID257
Feature activation+9.93e-02

Pos logit contributions
ometer+0.012
ipop+0.011
cases+0.010
 lid+0.010
ouse+0.009
Neg logit contributions
Playing-0.009
iling-0.009
Fighting-0.009
 owning-0.009
e-0.009
Token cone
Token ID27763
Feature activation+4.41e-02

Pos logit contributions
ipop+0.184
ometer+0.173
cases+0.170
atively+0.146
 reluctantly+0.142
Neg logit contributions
 town-0.154
Playing-0.146
e-0.143
 sports-0.141
 years-0.138
Token on
Token ID319
Feature activation+9.11e-02

Pos logit contributions
ipop+0.097
ometer+0.096
cases+0.090
 lid+0.086
 prayer+0.085
Neg logit contributions
e-0.052
Playing-0.046
 playing-0.046
hing-0.044
iling-0.044
Token to
Token ID284
Feature activation-1.85e-02

Pos logit contributions
 playing+0.048
 sports+0.044
 town+0.043
Playing+0.043
 outdoors+0.041
Neg logit contributions
ometer-0.068
ipop-0.065
 lid-0.061
cases-0.057
atively-0.057
Token preschool
Token ID38725
Feature activation-1.01e-01

Pos logit contributions
Playing+0.033
iling+0.032
Fighting+0.030
Leave+0.029
hing+0.029
Neg logit contributions
ipop-0.033
ometer-0.032
 lid-0.030
 off-0.027
cases-0.027
Token.
Token ID13
Feature activation+4.91e-03

Pos logit contributions
Playing+0.198
iling+0.196
Fighting+0.189
Leave+0.181
hing+0.181
Neg logit contributions
ipop-0.158
ometer-0.156
 lid-0.141
cases-0.133
 off-0.122
Token
Token ID198
Feature activation-2.34e-02

Pos logit contributions
ipop+0.009
ometer+0.009
 lid+0.008
cases+0.008
atively+0.007
Neg logit contributions
Playing-0.008
iling-0.008
Fighting-0.007
 town-0.007
Why-0.007
Token
Token ID198
Feature activation-2.92e-02

Pos logit contributions
Playing+0.036
iling+0.032
 town+0.032
Fighting+0.032
Why+0.031
Neg logit contributions
ipop-0.049
ometer-0.048
 lid-0.044
cases-0.042
atively-0.040
TokenAt
Token ID2953
Feature activation+3.28e-02

Pos logit contributions
iling+0.050
Playing+0.049
Fighting+0.048
hing+0.046
 owning+0.046
Neg logit contributions
ometer-0.050
ipop-0.049
 lid-0.046
cases-0.043
 off-0.041
Token preschool
Token ID38725
Feature activation-1.03e-01

Pos logit contributions
ometer+0.083
ipop+0.082
 lid+0.077
cases+0.073
atively+0.069
Neg logit contributions
Playing-0.035
iling-0.033
Fighting-0.031
 sports-0.029
 owning-0.029
Token,
Token ID11
Feature activation-2.66e-02

Pos logit contributions
Fighting+0.242
iling+0.240
Playing+0.239
hing+0.236
 disl+0.231
Neg logit contributions
ipop-0.118
ometer-0.107
cases-0.100
 lid-0.097
 off-0.090
Token she
Token ID673
Feature activation+1.40e-02

Pos logit contributions
Playing+0.045
iling+0.043
Fighting+0.041
 sports+0.040
 owning+0.039
Neg logit contributions
ometer-0.050
ipop-0.049
 lid-0.045
cases-0.042
ouse-0.039
Token loved
Token ID6151
Feature activation-1.28e-01

Pos logit contributions
ipop+0.022
ometer+0.021
cases+0.019
 lid+0.019
 reluctantly+0.018
Neg logit contributions
iling-0.027
Fighting-0.026
Playing-0.025
hing-0.025
Leave-0.025
Token learning
Token ID4673
Feature activation-5.79e-02

Pos logit contributions
iling+0.288
hing+0.280
Leave+0.267
Playing+0.261
Fighting+0.256
Neg logit contributions
ipop-0.167
ometer-0.161
 off-0.154
 eyes-0.139
 door-0.129
Token,
Token ID11
Feature activation+9.10e-03

Pos logit contributions
Playing+0.012
Fighting+0.012
iling+0.012
Leave+0.011
hing+0.011
Neg logit contributions
ipop-0.009
ometer-0.008
 lid-0.007
cases-0.006
ouse-0.006
Token a
Token ID257
Feature activation+1.00e-01

Pos logit contributions
ometer+0.015
ipop+0.015
 lid+0.014
cases+0.013
 off+0.012
Neg logit contributions
Playing-0.017
iling-0.017
Fighting-0.016
hing-0.015
Leave-0.015
Token truck
Token ID7779
Feature activation+5.53e-02

Pos logit contributions
ometer+0.170
ipop+0.167
 lid+0.152
cases+0.142
 off+0.130
Neg logit contributions
Playing-0.184
iling-0.177
Fighting-0.171
 sports-0.166
 owning-0.164
Token and
Token ID290
Feature activation+3.11e-02

Pos logit contributions
ometer+0.087
ipop+0.086
 lid+0.073
cases+0.066
 off+0.062
Neg logit contributions
Playing-0.115
iling-0.114
Fighting-0.113
Leave-0.108
hing-0.103
Token a
Token ID257
Feature activation+1.10e-01

Pos logit contributions
ometer+0.064
ipop+0.062
 lid+0.058
cases+0.053
ouse+0.051
Neg logit contributions
Playing-0.048
 sports-0.043
iling-0.043
Fighting-0.043
Leave-0.042
Token book
Token ID1492
Feature activation+8.96e-02

Pos logit contributions
ometer+0.186
ipop+0.182
 lid+0.164
cases+0.159
atively+0.143
Neg logit contributions
Playing-0.198
iling-0.186
Fighting-0.183
 sports-0.182
 owning-0.178
Token.
Token ID13
Feature activation+1.74e-02

Pos logit contributions
ometer+0.177
ipop+0.174
 lid+0.163
cases+0.146
ouse+0.137
Neg logit contributions
Playing-0.136
iling-0.129
 sports-0.121
 years-0.121
 owning-0.121
Token She
Token ID1375
Feature activation+7.01e-03

Pos logit contributions
ometer+0.028
ipop+0.027
 lid+0.025
cases+0.022
 off+0.021
Neg logit contributions
iling-0.034
Playing-0.034
Fighting-0.032
hing-0.031
 years-0.031
Token gives
Token ID3607
Feature activation+4.92e-02

Pos logit contributions
ometer+0.012
ipop+0.012
 lid+0.011
cases+0.010
 off+0.009
Neg logit contributions
Playing-0.013
iling-0.013
Fighting-0.012
Leave-0.012
hing-0.012
Token them
Token ID606
Feature activation+2.93e-02

Pos logit contributions
ometer+0.098
ipop+0.098
 lid+0.091
cases+0.083
 off+0.081
Neg logit contributions
Playing-0.077
iling-0.075
Fighting-0.071
Leave-0.067
hing-0.066
Token hugs
Token ID40657
Feature activation+6.58e-02

Pos logit contributions
ometer+0.053
ipop+0.052
 lid+0.048
cases+0.045
atively+0.041
Neg logit contributions
Playing-0.051
iling-0.049
Fighting-0.047
 sports-0.045
 owning-0.044
Token animals
Token ID4695
Feature activation-1.11e-01

Pos logit contributions
Playing+0.097
Fighting+0.089
iling+0.085
Now+0.079
Leave+0.079
Neg logit contributions
 lid-0.130
ometer-0.128
ipop-0.124
 off-0.123
 eyes-0.117
Token too
Token ID1165
Feature activation+6.07e-02

Pos logit contributions
Playing+0.232
Fighting+0.217
iling+0.216
Leave+0.212
years+0.209
Neg logit contributions
ometer-0.166
ipop-0.152
 lid-0.150
 off-0.148
cases-0.125
Token."
Token ID526
Feature activation+2.85e-02

Pos logit contributions
ometer+0.093
ipop+0.089
 lid+0.083
cases+0.074
 off+0.070
Neg logit contributions
Playing-0.125
iling-0.121
Fighting-0.117
Leave-0.113
 sports-0.111
Token The
Token ID383
Feature activation-2.25e-02

Pos logit contributions
ometer+0.045
ipop+0.044
 lid+0.040
 off+0.036
 mouth+0.035
Neg logit contributions
Playing-0.054
iling-0.054
Leave-0.051
Fighting-0.050
 owning-0.050
Token rabbit
Token ID22746
Feature activation-5.87e-02

Pos logit contributions
iling+0.045
Playing+0.043
Fighting+0.042
hing+0.041
e+0.039
Neg logit contributions
 lid-0.037
ometer-0.035
ipop-0.034
 mother-0.028
 key-0.027
Token thanked
Token ID26280
Feature activation+5.17e-02

Pos logit contributions
Playing+0.109
iling+0.107
Fighting+0.103
Leave+0.098
 owning+0.098
Neg logit contributions
ipop-0.099
ometer-0.099
 lid-0.090
cases-0.081
 off-0.077
Token the
Token ID262
Feature activation-2.18e-02

Pos logit contributions
ipop+0.076
 off+0.074
 lid+0.073
 eyes+0.069
ometer+0.068
Neg logit contributions
Playing-0.100
iling-0.099
Fighting-0.091
organized-0.088
hing-0.086
Token fairy
Token ID25607
Feature activation-8.66e-03

Pos logit contributions
Playing+0.042
iling+0.041
Fighting+0.039
hing+0.037
Leave+0.037
Neg logit contributions
ometer-0.036
ipop-0.036
 lid-0.035
 off-0.028
cases-0.027
Token and
Token ID290
Feature activation+3.23e-02

Pos logit contributions
iling+0.012
Playing+0.012
e+0.011
 sports+0.011
 years+0.011
Neg logit contributions
ometer-0.018
ipop-0.017
 lid-0.016
cases-0.015
atively-0.014
Token went
Token ID1816
Feature activation-1.25e-02

Pos logit contributions
ometer+0.063
ipop+0.060
 lid+0.056
cases+0.052
ouse+0.048
Neg logit contributions
Playing-0.053
iling-0.050
Fighting-0.049
 sports-0.047
 owning-0.047
Token on
Token ID319
Feature activation+5.69e-02

Pos logit contributions
 town+0.022
 playing+0.022
 sports+0.022
Playing+0.022
e+0.021
Neg logit contributions
ometer-0.021
ipop-0.020
 lid-0.019
cases-0.017
atively-0.016
Token had
Token ID550
Feature activation-1.58e-02

Pos logit contributions
ipop+0.023
ometer+0.022
 lid+0.020
 off+0.019
cases+0.019
Neg logit contributions
iling-0.024
Playing-0.023
Fighting-0.022
Leave-0.021
hing-0.021
Token a
Token ID257
Feature activation+9.95e-02

Pos logit contributions
iling+0.035
Playing+0.035
Fighting+0.033
hing+0.032
Leave+0.032
Neg logit contributions
ipop-0.020
ometer-0.020
 lid-0.019
 off-0.017
cases-0.017
Token wonderful
Token ID7932
Feature activation-6.80e-02

Pos logit contributions
ometer+0.193
ipop+0.192
 lid+0.178
cases+0.163
 off+0.153
Neg logit contributions
Playing-0.160
iling-0.157
Fighting-0.149
Leave-0.141
hing-0.140
Token time
Token ID640
Feature activation+5.42e-02

Pos logit contributions
Playing+0.113
iling+0.110
Fighting+0.104
 sports+0.099
 owning+0.099
Neg logit contributions
ometer-0.129
ipop-0.128
 lid-0.117
cases-0.109
 off-0.101
Token.
Token ID13
Feature activation+1.05e-02

Pos logit contributions
ipop+0.065
ometer+0.063
 off+0.054
cases+0.054
 lid+0.054
Neg logit contributions
Fighting-0.120
Playing-0.120
Leave-0.118
iling-0.117
Why-0.114
Token When
Token ID1649
Feature activation+1.05e-01

Pos logit contributions
ometer+0.017
ipop+0.017
 lid+0.015
cases+0.014
 off+0.013
Neg logit contributions
Playing-0.020
iling-0.020
Fighting-0.019
 sports-0.018
Leave-0.018
Token it
Token ID340
Feature activation+9.44e-02

Pos logit contributions
ometer+0.182
ipop+0.176
 lid+0.159
cases+0.150
atively+0.140
Neg logit contributions
Playing-0.196
iling-0.186
Fighting-0.181
 sports-0.179
 owning-0.176
Token was
Token ID373
Feature activation-2.87e-02

Pos logit contributions
ometer+0.185
ipop+0.182
 lid+0.167
cases+0.155
 off+0.144
Neg logit contributions
Playing-0.152
iling-0.148
Fighting-0.140
 sports-0.133
Leave-0.133
Token time
Token ID640
Feature activation+1.74e-02

Pos logit contributions
Playing+0.046
iling+0.045
Fighting+0.042
 sports+0.041
 owning+0.041
Neg logit contributions
ometer-0.058
ipop-0.056
 lid-0.051
cases-0.048
atively-0.044
Token to
Token ID284
Feature activation-1.54e-03

Pos logit contributions
ometer+0.036
ipop+0.033
 lid+0.033
cases+0.031
ouse+0.030
Neg logit contributions
 playing-0.025
 being-0.022
 owning-0.021
 doing-0.021
Playing-0.020
Token go
Token ID467
Feature activation-4.34e-02

Pos logit contributions
Playing+0.003
iling+0.003
Fighting+0.003
 sports+0.003
hing+0.003
Neg logit contributions
ometer-0.003
ipop-0.003
 lid-0.002
cases-0.002
atively-0.002
Token the
Token ID262
Feature activation-2.61e-02

Pos logit contributions
ipop+0.073
ometer+0.072
 lid+0.065
cases+0.062
 off+0.059
Neg logit contributions
Playing-0.082
iling-0.081
Fighting-0.077
Leave-0.074
hing-0.073
Token canvas
Token ID21978
Feature activation+2.78e-02

Pos logit contributions
Playing+0.042
iling+0.041
Fighting+0.038
Leave+0.037
 sports+0.037
Neg logit contributions
ometer-0.051
ipop-0.051
 lid-0.047
cases-0.043
 off-0.041
Token was
Token ID373
Feature activation-1.79e-02

Pos logit contributions
ometer+0.046
ipop+0.045
cases+0.039
 lid+0.039
 off+0.035
Neg logit contributions
Playing-0.055
iling-0.054
Fighting-0.051
Leave-0.050
Now-0.049
Token his
Token ID465
Feature activation+4.11e-02

Pos logit contributions
Playing+0.029
iling+0.028
Fighting+0.027
 owning+0.027
 sports+0.026
Neg logit contributions
ometer-0.034
ipop-0.034
 lid-0.031
cases-0.029
atively-0.027
Token dad
Token ID9955
Feature activation-6.80e-02

Pos logit contributions
ometer+0.062
ipop+0.061
 lid+0.055
cases+0.050
 off+0.045
Neg logit contributions
Playing-0.084
iling-0.082
Fighting-0.079
 sports-0.076
 owning-0.076
Token's
Token ID338
Feature activation+6.55e-02

Pos logit contributions
Playing+0.155
iling+0.148
Fighting+0.148
Now+0.147
 sports+0.145
Neg logit contributions
ipop-0.088
ometer-0.086
 lid-0.071
cases-0.069
 off-0.066
Token.
Token ID13
Feature activation+2.16e-02

Pos logit contributions
ometer+0.105
ipop+0.100
 lid+0.094
cases+0.080
ouse+0.075
Neg logit contributions
Playing-0.132
iling-0.131
hing-0.123
Fighting-0.123
Leave-0.121
Token He
Token ID679
Feature activation+7.34e-03

Pos logit contributions
ipop+0.038
ometer+0.035
 lid+0.032
cases+0.030
atively+0.028
Neg logit contributions
Playing-0.040
Fighting-0.038
Why-0.037
iling-0.037
 town-0.037
Token did
Token ID750
Feature activation+3.93e-02

Pos logit contributions
ometer+0.015
ipop+0.015
 lid+0.014
cases+0.013
ouse+0.012
Neg logit contributions
Playing-0.011
iling-0.010
 years-0.010
Fighting-0.010
Leave-0.009
Token not
Token ID407
Feature activation+5.61e-02

Pos logit contributions
ometer+0.059
ipop+0.058
 lid+0.052
cases+0.046
atively+0.042
Neg logit contributions
Playing-0.083
iling-0.080
Fighting-0.079
 owning-0.076
Why-0.076
Token know
Token ID760
Feature activation+4.85e-02

Pos logit contributions
ometer+0.103
ipop+0.102
 lid+0.095
 off+0.088
cases+0.085
Neg logit contributions
Playing-0.095
iling-0.090
Fighting-0.087
 sports-0.082
Leave-0.082
Token little
Token ID1310
Feature activation-4.61e-02

Pos logit contributions
ometer+0.079
ipop+0.077
 lid+0.071
ouse+0.057
 mother+0.057
Neg logit contributions
iling-0.123
Fighting-0.115
Playing-0.114
hing-0.112
Leave-0.108
Token girl
Token ID2576
Feature activation-5.81e-02

Pos logit contributions
iling+0.100
Playing+0.099
Fighting+0.097
Leave+0.091
Now+0.090
Neg logit contributions
ometer-0.069
ipop-0.068
 lid-0.060
 mouth-0.046
ouse-0.046
Token named
Token ID3706
Feature activation-2.07e-02

Pos logit contributions
Fighting+0.092
iling+0.089
Playing+0.088
hing+0.088
Leave+0.084
Neg logit contributions
ometer-0.110
ipop-0.108
 lid-0.097
 off-0.090
 mouth-0.089
Token Jane
Token ID12091
Feature activation+5.56e-03

Pos logit contributions
Playing+0.035
iling+0.033
Fighting+0.031
 sports+0.030
 owning+0.030
Neg logit contributions
ometer-0.040
ipop-0.039
 lid-0.036
cases-0.034
atively-0.031
Token.
Token ID13
Feature activation-7.96e-04

Pos logit contributions
ometer+0.008
ipop+0.008
 lid+0.007
 off+0.006
cases+0.006
Neg logit contributions
Playing-0.012
Fighting-0.011
iling-0.011
Leave-0.011
 sports-0.011
Token She
Token ID1375
Feature activation-1.24e-02

Pos logit contributions
Playing+0.001
iling+0.001
Fighting+0.001
 sports+0.001
Leave+0.001
Neg logit contributions
ometer-0.001
ipop-0.001
 lid-0.001
cases-0.001
 off-0.001
Token loved
Token ID6151
Feature activation-1.33e-01

Pos logit contributions
iling+0.020
Fighting+0.019
 sidelines+0.018
hing+0.018
Leave+0.018
Neg logit contributions
ometer-0.023
ipop-0.023
 lid-0.021
ouse-0.021
 off-0.018
Token to
Token ID284
Feature activation-4.59e-02

Pos logit contributions
iling+0.360
hing+0.342
Playing+0.332
Leave+0.331
e+0.330
Neg logit contributions
ipop-0.109
ometer-0.106
 off-0.095
 eyes-0.079
 lid-0.070
Token swim
Token ID9422
Feature activation-8.99e-03

Pos logit contributions
iling+0.074
Playing+0.073
Fighting+0.071
Leave+0.063
years+0.062
Neg logit contributions
ipop-0.092
ometer-0.087
 lid-0.083
 off-0.077
cases-0.073
Token in
Token ID287
Feature activation-9.29e-03

Pos logit contributions
Playing+0.012
iling+0.011
 sports+0.010
Fighting+0.010
 years+0.010
Neg logit contributions
ometer-0.020
ipop-0.020
 lid-0.019
cases-0.018
atively-0.017
Token the
Token ID262
Feature activation-3.55e-02

Pos logit contributions
iling+0.021
Playing+0.021
Fighting+0.020
hing+0.020
coon+0.019
Neg logit contributions
 off-0.012
 lid-0.011
 eyes-0.010
 door-0.010
 blanket-0.009
Token the
Token ID262
Feature activation-8.56e-03

Pos logit contributions
ometer+0.089
ipop+0.088
 lid+0.080
cases+0.077
atively+0.076
Neg logit contributions
Playing-0.054
iling-0.052
 years-0.049
Fighting-0.049
e-0.049
Token phone
Token ID3072
Feature activation+1.09e-01

Pos logit contributions
Playing+0.018
iling+0.018
Fighting+0.017
 owning+0.016
 sports+0.016
Neg logit contributions
ometer-0.012
ipop-0.012
 lid-0.010
cases-0.010
atively-0.009
Token."
Token ID526
Feature activation+3.58e-02

Pos logit contributions
ipop+0.192
ometer+0.185
 lid+0.177
atively+0.154
cases+0.147
Neg logit contributions
Playing-0.181
iling-0.175
Fighting-0.166
 years-0.164
 town-0.164
Token
Token ID198
Feature activation-1.28e-03

Pos logit contributions
ipop+0.080
ometer+0.079
 lid+0.072
cases+0.070
atively+0.069
Neg logit contributions
Playing-0.051
Fighting-0.043
 town-0.042
 Playing-0.042
Why-0.041
Token
Token ID198
Feature activation-7.64e-03

Pos logit contributions
Playing+0.002
iling+0.002
Fighting+0.002
 sports+0.002
Leave+0.002
Neg logit contributions
ometer-0.002
ipop-0.002
 lid-0.002
cases-0.002
 off-0.002
TokenL
Token ID43
Feature activation-6.21e-02

Pos logit contributions
iling+0.012
Playing+0.011
hing+0.011
 owning+0.011
 sports+0.010
Neg logit contributions
ometer-0.015
ipop-0.015
 lid-0.014
cases-0.013
 off-0.012
Tokenila
Token ID10102
Feature activation-3.01e-02

Pos logit contributions
Playing+0.112
iling+0.111
Fighting+0.104
 years+0.100
e+0.099
Neg logit contributions
ometer-0.112
ipop-0.107
 lid-0.098
cases-0.092
 off-0.084
Token and
Token ID290
Feature activation+2.29e-02

Pos logit contributions
Playing+0.049
iling+0.047
Fighting+0.045
Leave+0.043
 sports+0.043
Neg logit contributions
ometer-0.058
ipop-0.058
 lid-0.053
cases-0.049
 off-0.046
Token Sam
Token ID3409
Feature activation-7.34e-03

Pos logit contributions
ometer+0.037
ipop+0.037
 lid+0.033
cases+0.030
 off+0.028
Neg logit contributions
Playing-0.044
iling-0.044
Fighting-0.042
Leave-0.040
hing-0.040
Token follow
Token ID1061
Feature activation+6.25e-02

Pos logit contributions
Playing+0.015
iling+0.014
Fighting+0.014
 owning+0.013
Leave+0.013
Neg logit contributions
ipop-0.011
ometer-0.011
 lid-0.011
cases-0.010
 off-0.009
Token mom
Token ID1995
Feature activation+3.21e-03

Pos logit contributions
ipop+0.100
ometer+0.098
 lid+0.087
cases+0.080
atively+0.075
Neg logit contributions
Playing-0.121
iling-0.117
Fighting-0.115
 town-0.113
 years-0.112
Token met
Token ID1138
Feature activation-4.06e-02

Pos logit contributions
ipop+0.018
ometer+0.018
 lid+0.017
cases+0.016
 off+0.016
Neg logit contributions
iling-0.020
Fighting-0.018
Playing-0.018
Leave-0.017
 disl-0.017
Token a
Token ID257
Feature activation+7.72e-02

Pos logit contributions
Playing+0.077
iling+0.076
Fighting+0.072
Leave+0.069
hing+0.069
Neg logit contributions
ometer-0.067
ipop-0.066
 lid-0.060
cases-0.055
 off-0.052
Token friendly
Token ID8030
Feature activation-5.73e-02

Pos logit contributions
ometer+0.119
ipop+0.110
 lid+0.108
 mother+0.088
 prayer+0.086
Neg logit contributions
iling-0.166
Playing-0.149
hing-0.148
Fighting-0.145
years-0.142
Token monkey
Token ID21657
Feature activation-5.89e-02

Pos logit contributions
iling+0.103
Playing+0.101
Fighting+0.096
Leave+0.092
Now+0.090
Neg logit contributions
ometer-0.105
ipop-0.104
 lid-0.095
 door-0.081
 mouth-0.080
Token.
Token ID13
Feature activation+1.68e-02

Pos logit contributions
Playing+0.074
iling+0.073
Fighting+0.069
Leave+0.064
 sports+0.063
Neg logit contributions
ometer-0.135
ipop-0.134
 lid-0.123
cases-0.116
 off-0.111
Token The
Token ID383
Feature activation-3.12e-02

Pos logit contributions
ometer+0.031
ipop+0.031
 lid+0.027
cases+0.026
 mouth+0.024
Neg logit contributions
iling-0.030
Playing-0.029
Fighting-0.027
Leave-0.027
hing-0.026
Token monkey
Token ID21657
Feature activation-9.27e-02

Pos logit contributions
Playing+0.048
iling+0.048
Fighting+0.046
hing+0.043
Leave+0.042
Neg logit contributions
ipop-0.063
ometer-0.062
 lid-0.059
cases-0.052
 mouth-0.050
Token said
Token ID531
Feature activation+1.22e-01

Pos logit contributions
Playing+0.178
iling+0.177
Fighting+0.171
Leave+0.164
 sports+0.162
Neg logit contributions
ometer-0.149
ipop-0.148
 lid-0.131
cases-0.122
 off-0.111
Token,
Token ID11
Feature activation-8.69e-03

Pos logit contributions
ometer+0.215
ipop+0.210
 lid+0.195
cases+0.179
atively+0.164
Neg logit contributions
Playing-0.217
iling-0.209
Fighting-0.201
 sports-0.196
 owning-0.195
Token "
Token ID366
Feature activation-1.71e-01

Pos logit contributions
Playing+0.018
iling+0.018
Fighting+0.017
 sports+0.016
Leave+0.016
Neg logit contributions
ometer-0.013
ipop-0.013
 lid-0.011
cases-0.010
 off-0.009
TokenHello
Token ID15496
Feature activation-2.16e-02

Pos logit contributions
iling+0.224
 owning+0.205
 years+0.204
Playing+0.194
 repairing+0.194
Neg logit contributions
ipop-0.354
ometer-0.350
 lid-0.319
cases-0.302
 off-0.290
Token the
Token ID262
Feature activation-1.54e-02

Pos logit contributions
ometer+0.105
ipop+0.102
 lid+0.101
 off+0.096
cases+0.091
Neg logit contributions
Playing-0.104
iling-0.103
Fighting-0.094
hing-0.092
Leave-0.089
Token zipper
Token ID48992
Feature activation+6.00e-02

Pos logit contributions
Playing+0.033
iling+0.032
Fighting+0.031
 sports+0.030
Leave+0.030
Neg logit contributions
ometer-0.022
ipop-0.022
 lid-0.018
cases-0.017
atively-0.015
Token over
Token ID625
Feature activation+5.92e-02

Pos logit contributions
ometer+0.109
ipop+0.107
 lid+0.098
cases+0.092
 off+0.086
Neg logit contributions
Playing-0.105
iling-0.103
Fighting-0.098
Leave-0.093
 sports-0.093
Token and
Token ID290
Feature activation+1.85e-02

Pos logit contributions
 lid+0.083
ipop+0.082
ometer+0.076
 off+0.073
 eyes+0.071
Neg logit contributions
iling-0.129
hing-0.123
Playing-0.121
Fighting-0.119
 disl-0.117
Token over
Token ID625
Feature activation+5.56e-02

Pos logit contributions
ometer+0.031
ipop+0.030
 lid+0.028
cases+0.026
 off+0.024
Neg logit contributions
Playing-0.035
iling-0.035
Fighting-0.033
 sports-0.031
Leave-0.031
Token and
Token ID290
Feature activation+1.43e-02

Pos logit contributions
ipop+0.070
ometer+0.069
 lid+0.067
 off+0.058
atively+0.054
Neg logit contributions
iling-0.126
Playing-0.125
Fighting-0.121
hing-0.121
 disl-0.115
Token couldn
Token ID3521
Feature activation-4.88e-02

Pos logit contributions
ometer+0.024
ipop+0.024
 lid+0.022
cases+0.020
 off+0.019
Neg logit contributions
Playing-0.027
iling-0.026
Fighting-0.025
 sports-0.024
Leave-0.024
Token't
Token ID470
Feature activation+5.33e-02

Pos logit contributions
Playing+0.086
iling+0.084
Fighting+0.080
 sports+0.076
Leave+0.076
Neg logit contributions
ometer-0.088
ipop-0.087
 lid-0.079
cases-0.073
 off-0.068
Token get
Token ID651
Feature activation+3.93e-04

Pos logit contributions
ometer+0.087
ipop+0.086
 lid+0.081
 off+0.074
atively+0.071
Neg logit contributions
Playing-0.103
iling-0.102
Fighting-0.098
 owning-0.090
 Playing-0.089
Token it
Token ID340
Feature activation+8.63e-02

Pos logit contributions
ometer+0.001
ipop+0.001
 lid+0.001
cases+0.001
atively+0.001
Neg logit contributions
Playing-0.001
iling-0.001
Fighting-0.001
 owning-0.001
 sports-0.001
Token to
Token ID284
Feature activation+3.99e-03

Pos logit contributions
ometer+0.146
ipop+0.144
 lid+0.131
cases+0.120
atively+0.110
Neg logit contributions
Playing-0.161
iling-0.157
Fighting-0.150
 sports-0.144
 owning-0.144
Token their
Token ID511
Feature activation+5.17e-02

Pos logit contributions
ipop+0.177
ometer+0.175
 lid+0.161
cases+0.158
ouse+0.145
Neg logit contributions
Playing-0.118
Fighting-0.108
iling-0.107
 sports-0.107
 town-0.106
Token paws
Token ID49193
Feature activation+2.41e-02

Pos logit contributions
ipop+0.064
ometer+0.063
cases+0.054
 lid+0.048
atively+0.045
Neg logit contributions
Playing-0.115
iling-0.114
 years-0.111
Fighting-0.108
 town-0.108
Token and
Token ID290
Feature activation+3.11e-02

Pos logit contributions
ipop+0.044
ometer+0.044
 lid+0.040
cases+0.037
atively+0.034
Neg logit contributions
Playing-0.041
iling-0.039
Fighting-0.038
 sports-0.037
hing-0.037
Token claws
Token ID28421
Feature activation+1.21e-01

Pos logit contributions
ometer+0.060
ipop+0.059
 lid+0.053
cases+0.052
ouse+0.047
Neg logit contributions
Playing-0.050
e-0.046
Leave-0.046
iling-0.045
Fighting-0.045
Token to
Token ID284
Feature activation+3.28e-02

Pos logit contributions
ometer+0.199
ipop+0.196
 lid+0.178
cases+0.164
 off+0.153
Neg logit contributions
Playing-0.231
iling-0.227
Fighting-0.216
 sports-0.207
 owning-0.206
Token the
Token ID262
Feature activation-1.22e-02

Pos logit contributions
ipop+0.052
ometer+0.052
 lid+0.051
 off+0.047
atively+0.044
Neg logit contributions
Playing-0.064
iling-0.062
Fighting-0.060
 sports-0.055
 Playing-0.055
Token flower
Token ID15061
Feature activation+3.91e-02

Pos logit contributions
Playing+0.022
iling+0.022
Fighting+0.021
hing+0.020
Leave+0.020
Neg logit contributions
ometer-0.022
 lid-0.021
ipop-0.021
 off-0.017
cases-0.017
Token.
Token ID13
Feature activation+2.73e-02

Pos logit contributions
ometer+0.065
ipop+0.063
 lid+0.058
cases+0.052
 off+0.050
Neg logit contributions
Playing-0.076
iling-0.074
Fighting-0.071
 sports-0.067
Leave-0.067
Token But
Token ID887
Feature activation+6.50e-02

Pos logit contributions
ometer+0.042
ipop+0.040
 lid+0.037
cases+0.033
 off+0.030
Neg logit contributions
Playing-0.056
iling-0.055
Fighting-0.053
 sports-0.050
hing-0.050
Token they
Token ID484
Feature activation+2.07e-02

Pos logit contributions
ometer+0.111
ipop+0.110
 lid+0.101
cases+0.092
 off+0.089
Neg logit contributions
iling-0.121
Playing-0.120
Fighting-0.113
hing-0.108
 sports-0.107
Token could
Token ID714
Feature activation+4.62e-04

Pos logit contributions
ipop+0.036
ometer+0.036
 lid+0.033
cases+0.030
atively+0.028
Neg logit contributions
iling-0.037
Playing-0.037
Fighting-0.035
 sports-0.035
Leave-0.033
Token Lily
Token ID20037
Feature activation-3.41e-02

Pos logit contributions
Playing+0.031
iling+0.030
Fighting+0.027
 sports+0.027
Why+0.027
Neg logit contributions
ometer-0.043
ipop-0.042
 lid-0.038
cases-0.037
atively-0.035
Token.
Token ID13
Feature activation+3.06e-03

Pos logit contributions
iling+0.075
Playing+0.074
Fighting+0.074
Leave+0.072
 sports+0.071
Neg logit contributions
ometer-0.043
ipop-0.040
 lid-0.035
 off-0.031
ouse-0.029
Token She
Token ID1375
Feature activation-8.67e-03

Pos logit contributions
ometer+0.005
ipop+0.005
 lid+0.004
cases+0.004
 off+0.004
Neg logit contributions
Playing-0.006
iling-0.006
Fighting-0.006
Leave-0.005
hing-0.005
Token loved
Token ID6151
Feature activation-1.32e-01

Pos logit contributions
iling+0.013
Fighting+0.012
hing+0.011
Playing+0.011
Leave+0.011
Neg logit contributions
ometer-0.018
ipop-0.017
 lid-0.016
ouse-0.015
 off-0.014
Token to
Token ID284
Feature activation-4.39e-02

Pos logit contributions
iling+0.361
hing+0.344
Playing+0.338
Fighting+0.333
Leave+0.332
Neg logit contributions
ipop-0.109
ometer-0.107
 off-0.086
 eyes-0.069
 lid-0.069
Token play
Token ID711
Feature activation-1.46e-01

Pos logit contributions
iling+0.065
Playing+0.065
Fighting+0.062
Leave+0.055
 years+0.054
Neg logit contributions
ipop-0.093
ometer-0.089
 lid-0.084
 off-0.078
cases-0.076
Token in
Token ID287
Feature activation-2.71e-02

Pos logit contributions
iling+0.240
Playing+0.235
Fighting+0.234
years+0.219
Leave+0.210
Neg logit contributions
ipop-0.258
ometer-0.254
 lid-0.251
 off-0.238
cases-0.215
Token her
Token ID607
Feature activation+2.84e-02

Pos logit contributions
iling+0.061
Playing+0.058
hing+0.056
Fighting+0.056
years+0.053
Neg logit contributions
 off-0.033
 lid-0.032
 door-0.028
ipop-0.027
 mom-0.027
Token backyard
Token ID24296
Feature activation-4.85e-02

Pos logit contributions
ometer+0.056
ipop+0.054
 lid+0.051
cases+0.047
 off+0.044
Neg logit contributions
Playing-0.046
iling-0.045
Fighting-0.043
hing-0.041
Leave-0.040
Token every
Token ID790
Feature activation+3.41e-02

Pos logit contributions
Playing+0.099
iling+0.098
Fighting+0.097
Leave+0.092
hing+0.091
Neg logit contributions
ometer-0.069
ipop-0.069
 lid-0.065
cases-0.057
 off-0.056
Token day
Token ID1110
Feature activation-2.29e-02

Pos logit contributions
ometer+0.077
ipop+0.076
 lid+0.070
cases+0.068
atively+0.062
Neg logit contributions
Playing-0.045
iling-0.043
Fighting-0.040
 sports-0.039
Leave-0.038
Token They
Token ID1119
Feature activation-1.73e-02

Pos logit contributions
ometer+0.040
ipop+0.040
 lid+0.036
cases+0.033
atively+0.030
Neg logit contributions
Playing-0.045
iling-0.043
Fighting-0.042
 sports-0.040
Leave-0.040
Token smile
Token ID8212
Feature activation-1.07e-02

Pos logit contributions
Playing+0.033
iling+0.033
Fighting+0.031
 sports+0.030
 owning+0.029
Neg logit contributions
ipop-0.028
ometer-0.027
 lid-0.026
 off-0.023
cases-0.023
Token.
Token ID13
Feature activation+2.36e-02

Pos logit contributions
Playing+0.018
iling+0.017
Fighting+0.017
 sports+0.016
 owning+0.016
Neg logit contributions
ipop-0.020
ometer-0.020
 lid-0.018
cases-0.017
atively-0.016
Token
Token ID198
Feature activation-1.90e-03

Pos logit contributions
ipop+0.042
ometer+0.042
 lid+0.038
cases+0.036
atively+0.033
Neg logit contributions
Playing-0.043
iling-0.040
Fighting-0.039
 sports-0.037
Leave-0.037
Token
Token ID198
Feature activation-7.86e-03

Pos logit contributions
iling+0.004
Playing+0.004
Fighting+0.004
 sports+0.004
Leave+0.004
Neg logit contributions
ometer-0.003
ipop-0.003
 lid-0.003
cases-0.002
 off-0.002
Token"
Token ID1
Feature activation-1.48e-01

Pos logit contributions
iling+0.015
 owning+0.015
 sports+0.014
hing+0.014
 disl+0.013
Neg logit contributions
ometer-0.012
 lid-0.011
ipop-0.011
 off-0.009
cases-0.009
TokenThank
Token ID10449
Feature activation-8.83e-02

Pos logit contributions
iling+0.198
Playing+0.172
 years+0.172
 owning+0.170
 sports+0.169
Neg logit contributions
ipop-0.321
ometer-0.316
 lid-0.307
atively-0.265
cases-0.264
Token you
Token ID345
Feature activation-5.13e-02

Pos logit contributions
Playing+0.114
iling+0.109
Fighting+0.105
 owning+0.100
 sports+0.099
Neg logit contributions
ipop-0.194
ometer-0.193
 lid-0.183
 off-0.163
cases-0.163
Token,
Token ID11
Feature activation-4.50e-02

Pos logit contributions
 years+0.069
iling+0.064
Playing+0.062
e+0.061
 playing+0.060
Neg logit contributions
ometer-0.110
ipop-0.109
cases-0.099
 lid-0.097
ouse-0.092
Token sheep
Token ID15900
Feature activation-7.60e-02

Pos logit contributions
Playing+0.079
iling+0.078
Fighting+0.074
Leave+0.070
 sports+0.070
Neg logit contributions
ometer-0.081
ipop-0.081
 lid-0.075
cases-0.067
 off-0.064
Token cloud
Token ID6279
Feature activation-6.63e-03

Pos logit contributions
Playing+0.179
iling+0.170
Fighting+0.168
Leave+0.163
 owning+0.160
Neg logit contributions
ometer-0.098
ipop-0.098
 lid-0.082
 off-0.068
cases-0.067
Token are
Token ID389
Feature activation-7.18e-02

Pos logit contributions
ometer+0.100
ipop+0.095
 lid+0.087
cases+0.072
 off+0.070
Neg logit contributions
Playing-0.165
iling-0.159
Fighting-0.157
Leave-0.151
hing-0.151
Token tools
Token ID4899
Feature activation+2.33e-02

Pos logit contributions
Playing+0.130
iling+0.129
hing+0.121
Fighting+0.120
Leave+0.118
Neg logit contributions
ipop-0.118
ometer-0.117
 lid-0.111
 off-0.103
cases-0.094
Token."
Token ID526
Feature activation+1.44e-02

Pos logit contributions
ometer+0.035
ipop+0.033
 lid+0.030
cases+0.029
 off+0.028
Neg logit contributions
Playing-0.049
iling-0.047
Fighting-0.045
Leave-0.044
years-0.043
Token
Token ID198
Feature activation-2.37e-02

Pos logit contributions
ometer+0.025
ipop+0.024
 lid+0.022
cases+0.020
atively+0.019
Neg logit contributions
Playing-0.027
iling-0.026
Fighting-0.025
 sports-0.024
Leave-0.024
Token
Token ID198
Feature activation-2.81e-02

Pos logit contributions
Playing+0.045
iling+0.045
Fighting+0.042
 sports+0.041
 owning+0.041
Neg logit contributions
ometer-0.038
ipop-0.038
 lid-0.034
cases-0.031
 off-0.029
TokenTom
Token ID13787
Feature activation-1.00e-01

Pos logit contributions
iling+0.046
 owning+0.045
Playing+0.043
hing+0.043
ting+0.043
Neg logit contributions
ometer-0.049
ipop-0.047
 lid-0.044
cases-0.039
eth-0.039
Token nods
Token ID34833
Feature activation+4.11e-02

Pos logit contributions
Playing+0.198
iling+0.193
Fighting+0.187
 owning+0.182
Leave+0.180
Neg logit contributions
ipop-0.156
ometer-0.156
 lid-0.140
cases-0.129
 off-0.119
Token.
Token ID13
Feature activation+1.01e-02

Pos logit contributions
ometer+0.096
ipop+0.091
 lid+0.086
cases+0.085
roller+0.073
Neg logit contributions
Playing-0.054
Why-0.053
iling-0.051
 years-0.050
 sports-0.048
Token He
Token ID679
Feature activation+2.37e-03

Pos logit contributions
ipop+0.018
ometer+0.017
 lid+0.016
cases+0.016
ouse+0.014
Neg logit contributions
Playing-0.018
Fighting-0.017
Why-0.017
 town-0.016
iling-0.016
Token says
Token ID1139
Feature activation+1.28e-01

Pos logit contributions
ometer+0.004
ipop+0.004
 lid+0.003
cases+0.003
atively+0.003
Neg logit contributions
Playing-0.005
iling-0.005
Fighting-0.004
Leave-0.004
 owning-0.004
Token sorry
Token ID7926
Feature activation+7.57e-03

Pos logit contributions
ometer+0.238
ipop+0.230
 lid+0.217
cases+0.205
atively+0.187
Neg logit contributions
Playing-0.215
iling-0.199
Fighting-0.196
 owning-0.192
 sports-0.192
Token,
Token ID11
Feature activation-3.63e-02

Pos logit contributions
ometer+0.041
ipop+0.040
 lid+0.037
cases+0.034
 off+0.032
Neg logit contributions
Playing-0.040
iling-0.039
Fighting-0.037
 sports-0.036
 owning-0.036
Token there
Token ID612
Feature activation+5.21e-02

Pos logit contributions
Playing+0.063
iling+0.060
Fighting+0.056
Why+0.056
 sports+0.055
Neg logit contributions
ometer-0.067
ipop-0.066
 lid-0.060
cases-0.057
 off-0.051
Token was
Token ID373
Feature activation-5.90e-02

Pos logit contributions
ometer+0.065
ipop+0.065
 lid+0.062
 off+0.049
atively+0.047
Neg logit contributions
iling-0.116
Fighting-0.115
Playing-0.113
Leave-0.110
years-0.108
Token a
Token ID257
Feature activation+5.41e-02

Pos logit contributions
iling+0.137
Playing+0.133
Fighting+0.132
hing+0.126
Leave+0.124
Neg logit contributions
ipop-0.067
ometer-0.064
 lid-0.062
 off-0.052
cases-0.046
Token little
Token ID1310
Feature activation-4.99e-02

Pos logit contributions
ometer+0.076
ipop+0.074
 lid+0.069
 mother+0.055
ouse+0.055
Neg logit contributions
iling-0.122
Fighting-0.114
Playing-0.114
hing-0.111
Leave-0.107
Token boy
Token ID2933
Feature activation-1.00e-01

Pos logit contributions
Playing+0.110
iling+0.109
Fighting+0.106
Leave+0.100
Now+0.098
Neg logit contributions
ometer-0.073
ipop-0.072
 lid-0.064
 mouth-0.049
ouse-0.048
Token who
Token ID508
Feature activation-1.95e-02

Pos logit contributions
Fighting+0.170
Playing+0.162
iling+0.160
hing+0.158
Leave+0.157
Neg logit contributions
ometer-0.178
ipop-0.174
 lid-0.148
 mouth-0.143
 off-0.139
Token was
Token ID373
Feature activation-4.80e-02

Pos logit contributions
iling+0.031
Fighting+0.029
Playing+0.029
ting+0.028
 sidelines+0.027
Neg logit contributions
ometer-0.039
ipop-0.036
 lid-0.034
ouse-0.032
 name-0.030
Token very
Token ID845
Feature activation-4.19e-02

Pos logit contributions
iling+0.101
Playing+0.096
Fighting+0.092
hing+0.092
Now+0.090
Neg logit contributions
ometer-0.075
ipop-0.067
 lid-0.061
 mouth-0.055
 off-0.054
Token thirsty
Token ID47124
Feature activation-9.21e-02

Pos logit contributions
Playing+0.074
iling+0.073
Fighting+0.071
hing+0.068
Leave+0.067
Neg logit contributions
ometer-0.076
ipop-0.075
 lid-0.067
cases-0.061
ouse-0.058
Token.
Token ID13
Feature activation-8.05e-03

Pos logit contributions
Playing+0.203
iling+0.195
Fighting+0.194
Leave+0.186
 disl+0.184
Neg logit contributions
ometer-0.123
ipop-0.121
 lid-0.115
 off-0.102
cases-0.095
Token there
Token ID612
Feature activation+5.24e-02

Pos logit contributions
Playing+0.062
iling+0.060
Fighting+0.057
 sports+0.055
 owning+0.055
Neg logit contributions
ometer-0.057
ipop-0.056
 lid-0.051
cases-0.047
 off-0.043
Token lived
Token ID5615
Feature activation-1.45e-01

Pos logit contributions
ometer+0.062
ipop+0.062
 lid+0.061
 off+0.047
 mouth+0.046
Neg logit contributions
iling-0.119
Fighting-0.118
Playing-0.115
Leave-0.113
years-0.110
Token a
Token ID257
Feature activation+4.63e-02

Pos logit contributions
iling+0.343
Fighting+0.326
Playing+0.326
 disl+0.320
hing+0.316
Neg logit contributions
 lid-0.154
ipop-0.152
ometer-0.150
 off-0.133
icked-0.112
Token 3
Token ID513
Feature activation-1.00e-01

Pos logit contributions
ometer+0.066
ipop+0.064
 lid+0.060
 mother+0.049
ouse+0.048
Neg logit contributions
iling-0.104
Fighting-0.096
Playing-0.095
hing-0.094
Leave-0.090
Token year
Token ID614
Feature activation+2.35e-02

Pos logit contributions
Playing+0.181
Fighting+0.172
iling+0.168
Leave+0.163
 owning+0.161
Neg logit contributions
ipop-0.179
ometer-0.176
 lid-0.166
cases-0.147
ouse-0.138
Token old
Token ID1468
Feature activation-1.21e-01

Pos logit contributions
ometer+0.046
ipop+0.046
 lid+0.042
cases+0.039
 off+0.037
Neg logit contributions
Playing-0.037
iling-0.036
Fighting-0.034
Leave-0.033
 sports-0.032
Token called
Token ID1444
Feature activation+2.08e-02

Pos logit contributions
iling+0.234
Playing+0.230
Fighting+0.222
hing+0.212
Leave+0.212
Neg logit contributions
ometer-0.207
ipop-0.206
 lid-0.182
cases-0.156
 eyes-0.150
Token Lucy
Token ID22162
Feature activation+4.62e-03

Pos logit contributions
ometer+0.039
ipop+0.038
 lid+0.035
cases+0.032
 off+0.030
Neg logit contributions
Playing-0.036
iling-0.035
Fighting-0.034
Leave-0.032
hing-0.031
Token.
Token ID13
Feature activation+3.14e-03

Pos logit contributions
ometer+0.007
ipop+0.006
 lid+0.005
cases+0.005
 off+0.005
Neg logit contributions
Playing-0.010
iling-0.010
Fighting-0.010
Leave-0.009
 sports-0.009
Token One
Token ID1881
Feature activation+9.63e-02

Pos logit contributions
ometer+0.004
ipop+0.004
 lid+0.004
cases+0.003
atively+0.003
Neg logit contributions
Playing-0.007
iling-0.007
Fighting-0.007
Leave-0.006
hing-0.006
Token day
Token ID1110
Feature activation-2.56e-02

Pos logit contributions
ometer+0.183
ipop+0.182
 lid+0.162
cases+0.142
ouse+0.141
Neg logit contributions
iling-0.163
Playing-0.162
Fighting-0.158
Leave-0.151
hing-0.143