TOP ACTIVATIONS
MAX = 32.548

by one 1 , 2 ,
said I kept my promise !
said don â t
shouted Run Tony , Run !
himself Sometimes it â
â Come on now , Katie
â Hello ! Where are you
perfect bag now I 'm very
, Thanks , Jack . You
, We use the gun .
, It means I 'm gonna
, Hope you 'll be more
, Will you come listen to
, This box is probably too
, Let me take a look
, I was dreaming about this
, That â s
, I can fly , I
, That â s
, That 's the most important

MIN ACTIVATIONS
MIN = -21.954

saw some more che wed threads in the house . Fl
she continued picking her p ears . With a smile on
, do not touch the pins . They are not toys
" Let 's replace the rocks at the bottom of the
and found lots of gold coins and shiny jewels .
to spray some nice smelling liquid around the garden . Sarah
continued to make fancy mud designs . <|endoftext|> One day ,
that allowed people to buy stamps . Her mom my gave
saw a big bottle of shampoo on the shelf . It
good to have so much clutter . You can trip and
the birds could make their nests and feel safe . The
Let 's gather up some rocks and see who can make
you will never touch my axe again ," his dad said
tree with many ripe p ears . They wanted to pick
chest and found lots of gold coins and shiny jewels .
the area clean by making stacks of sticks .
a grand hall filled with pots of the magic liquid ,
load your wagon with p ears near a wild bear .
to be careful when collecting rocks and always make sure she
. They had to get stitches and shots . They had

INTERVAL 18.922 - 32.548
CONTAINS 0.194%

that were all mixed up . Jane didn 't know what
â What â s
â Why don â
dead squirrel in the grass . She stared at it for
could be felt all around . <|endoftext|> Once upon a time

INTERVAL 5.297 - 18.922
CONTAINS 21.992%

for dinner . She felt a wave of relief as their
her . She ran after the dog and the man and
" Please , Lily , you have to take the pill
kept asking until finally , his parents agreed to help him
, " Yes , m ama . I can be quiet

INTERVAL -8.329 - 5.297
CONTAINS 66.789%

He understood that it was much too cold outside , and
was scared of the dark , loud noises , and even
Jill laughed and said ,
It went far and wide on the big blue sea .
but he did not know how . He asked , "

INTERVAL -21.954 - -8.329
CONTAINS 11.025%

drove over it , hon king and be eping .
. They jumped into the pool and threw a ball for
dog . Tim liked to play outside in the dirt with
looked closely and saw that it was a sm elly sausage
He wants to hug the dog . He runs to the
Token by
Token ID416
Feature activation+1.48e+01

Pos logit contributions
Un+0.023
David+0.023
 Fro+0.023
 concession+0.022
Jack+0.022
Neg logit contributions
hole-0.040
 pieces-0.040
 on-0.039
 atop-0.037
 edges-0.037
Token one
Token ID530
Feature activation+1.14e+01

Pos logit contributions
Un+0.166
Jack+0.164
 Fro+0.164
David+0.161
Jim+0.158
Neg logit contributions
hole-0.172
 pieces-0.167
 atop-0.158
 lid-0.153
 edges-0.153
Token
Token ID6184
Feature activation+1.22e+01

Pos logit contributions
Jack+0.117
Un+0.116
 concession+0.116
 Fro+0.116
Jim+0.114
Neg logit contributions
ps-0.124
hole-0.124
 atop-0.117
 pieces-0.117
 tangled-0.116
Token
Token ID95
Feature activation+1.31e+01

Pos logit contributions
Un+0.123
Jack+0.122
David+0.122
 Fro+0.121
 concession+0.116
Neg logit contributions
 pieces-0.161
hole-0.158
 on-0.147
 atop-0.147
 edges-0.146
Token
Token ID26391
Feature activation+1.85e+01

Pos logit contributions
Jack+0.119
Un+0.119
 Fro+0.118
David+0.118
Jim+0.112
Neg logit contributions
hole-0.182
 pieces-0.179
 atop-0.170
 on-0.169
 edges-0.168
Token
Token ID447
Feature activation+2.22e+01

Pos logit contributions
 concession+0.169
Un+0.167
Da+0.167
 Fro+0.165
Jack+0.165
Neg logit contributions
hole-0.251
 on-0.237
 pieces-0.232
 atop-0.226
 falling-0.221
Token
Token ID250
Feature activation+1.71e+01

Pos logit contributions
Un+0.208
David+0.206
 Fro+0.205
Jack+0.205
Kids+0.197
Neg logit contributions
hole-0.294
 pieces-0.289
 atop-0.278
 on-0.276
 edges-0.268
Token 1
Token ID352
Feature activation+5.25e+00

Pos logit contributions
 Fro+0.114
Un+0.113
Jack+0.112
David+0.112
 concession+0.108
Neg logit contributions
hole-0.274
 pieces-0.273
 on-0.259
 atop-0.258
 edges-0.257
Token,
Token ID11
Feature activation+1.15e+01

Pos logit contributions
Un+0.063
 Fro+0.061
Jim+0.061
David+0.061
Jack+0.061
Neg logit contributions
hole-0.054
 edges-0.053
 pieces-0.053
 falling-0.051
 tangled-0.051
Token 2
Token ID362
Feature activation+5.85e+00

Pos logit contributions
Un+0.113
David+0.110
Jack+0.110
 Fro+0.107
 concession+0.104
Neg logit contributions
hole-0.153
 pieces-0.150
 atop-0.146
ps-0.143
 edges-0.141
Token,
Token ID11
Feature activation+1.10e+01

Pos logit contributions
Un+0.067
David+0.066
Jack+0.066
Jim+0.066
 Fro+0.064
Neg logit contributions
hole-0.066
 atop-0.062
 tangled-0.062
ps-0.062
 edges-0.061
Token said
Token ID531
Feature activation+1.33e+01

Pos logit contributions
hole+0.098
 pieces+0.095
 edges+0.091
ps+0.091
 lid+0.091
Neg logit contributions
 Fro-0.113
Jack-0.104
Un-0.104
enny-0.101
 His-0.101
Token
Token ID6184
Feature activation+1.49e+01

Pos logit contributions
Un+0.193
David+0.189
Jack+0.189
Kids+0.185
 concession+0.185
Neg logit contributions
 on-0.112
 pieces-0.111
hole-0.110
 atop-0.102
ps-0.101
Token
Token ID95
Feature activation+1.32e+01

Pos logit contributions
Un+0.153
Jack+0.152
David+0.151
 Fro+0.149
Jim+0.144
Neg logit contributions
hole-0.190
 pieces-0.184
 atop-0.175
 edges-0.174
 on-0.174
Token
Token ID26391
Feature activation+1.88e+01

Pos logit contributions
Un+0.122
Jack+0.122
David+0.121
 Fro+0.120
Jim+0.114
Neg logit contributions
hole-0.182
 pieces-0.181
 atop-0.170
 on-0.169
 edges-0.169
Token
Token ID129
Feature activation+1.44e+01

Pos logit contributions
 concession+0.187
Da+0.182
Un+0.181
Jack+0.179
Kids+0.175
Neg logit contributions
hole-0.241
 on-0.234
 pieces-0.219
 atop-0.214
 trunk-0.211
Token
Token ID241
Feature activation+2.13e+01

Pos logit contributions
Jack+0.136
Un+0.136
David+0.134
 Fro+0.133
 concession+0.128
Neg logit contributions
hole-0.193
 pieces-0.190
 atop-0.180
 edges-0.179
 on-0.179
TokenI
Token ID40
Feature activation+1.38e+01

Pos logit contributions
 Fro+0.179
 concession+0.175
oph+0.173
Un+0.172
Da+0.168
Neg logit contributions
 on-0.301
hole-0.300
 pieces-0.294
 edges-0.279
 atop-0.277
Token kept
Token ID4030
Feature activation+6.17e+00

Pos logit contributions
Un+0.142
David+0.140
Jack+0.139
 Fro+0.136
 concession+0.134
Neg logit contributions
hole-0.174
 pieces-0.171
 on-0.166
 atop-0.162
 edges-0.157
Token my
Token ID616
Feature activation+9.87e+00

Pos logit contributions
David+0.075
Un+0.074
Jack+0.074
 Fro+0.074
Johnny+0.073
Neg logit contributions
hole-0.069
 pieces-0.065
 lid-0.061
 edges-0.061
 atop-0.061
Token promise
Token ID6991
Feature activation+2.12e+00

Pos logit contributions
 Fro+0.119
 concession+0.114
Un+0.113
Jack+0.111
David+0.111
Neg logit contributions
hole-0.111
 atop-0.103
 pieces-0.102
ps-0.096
 edges-0.095
Token!
Token ID0
Feature activation+1.61e+01

Pos logit contributions
Un+0.027
Jack+0.027
David+0.027
 Fro+0.027
 concession+0.026
Neg logit contributions
 pieces-0.021
hole-0.021
 atop-0.019
 edges-0.019
ps-0.019
Token said
Token ID531
Feature activation+1.33e+01

Pos logit contributions
hole+0.193
 pieces+0.191
 edges+0.187
els+0.187
 atop+0.184
Neg logit contributions
Un-0.114
Jack-0.113
David-0.111
 concession-0.107
enny-0.107
Token
Token ID6184
Feature activation+1.49e+01

Pos logit contributions
Un+0.193
David+0.189
Jack+0.189
Kids+0.185
 concession+0.185
Neg logit contributions
 on-0.112
 pieces-0.111
hole-0.110
 atop-0.102
ps-0.101
Token
Token ID95
Feature activation+1.32e+01

Pos logit contributions
Un+0.153
Jack+0.152
David+0.151
 Fro+0.149
Jim+0.144
Neg logit contributions
hole-0.190
 pieces-0.184
 atop-0.175
 edges-0.174
 on-0.174
Token
Token ID26391
Feature activation+1.88e+01

Pos logit contributions
Un+0.122
Jack+0.122
David+0.121
 Fro+0.120
Jim+0.114
Neg logit contributions
hole-0.182
 pieces-0.181
 atop-0.170
 on-0.169
 edges-0.169
Token
Token ID129
Feature activation+1.44e+01

Pos logit contributions
 concession+0.187
Da+0.182
Un+0.181
Jack+0.179
Kids+0.175
Neg logit contributions
hole-0.241
 on-0.234
 pieces-0.219
 atop-0.214
 trunk-0.211
Token
Token ID241
Feature activation+2.13e+01

Pos logit contributions
Jack+0.136
Un+0.136
David+0.134
 Fro+0.133
 concession+0.128
Neg logit contributions
hole-0.193
 pieces-0.190
 atop-0.180
 edges-0.179
 on-0.179
Tokendon
Token ID9099
Feature activation+1.34e+01

Pos logit contributions
 Fro+0.179
 concession+0.175
oph+0.173
Un+0.172
Da+0.168
Neg logit contributions
 on-0.301
hole-0.300
 pieces-0.294
 edges-0.279
 atop-0.277
Tokenâ
Token ID22940
Feature activation+1.58e+01

Pos logit contributions
 Fro+0.180
 concession+0.178
Jack+0.176
Da+0.175
Un+0.175
Neg logit contributions
 on-0.124
hole-0.118
 pieces-0.114
 button-0.108
 hardest-0.107
Token
Token ID26391
Feature activation+1.81e+01

Pos logit contributions
Jack+0.094
Un+0.093
David+0.093
 Fro+0.091
Kids+0.088
Neg logit contributions
 pieces-0.273
hole-0.268
 atop-0.255
 edges-0.255
 on-0.254
Token
Token ID8151
Feature activation+1.36e+01

Pos logit contributions
 concession+0.170
Jack+0.170
Un+0.170
Da+0.169
David+0.164
Neg logit contributions
hole-0.234
 on-0.227
 pieces-0.219
 atop-0.211
 ladder-0.208
Tokent
Token ID83
Feature activation+1.05e+01

Pos logit contributions
Un+0.151
Jack+0.150
David+0.150
 Fro+0.148
Jim+0.143
Neg logit contributions
hole-0.161
 pieces-0.160
 atop-0.149
 on-0.148
 edges-0.147
Token shouted
Token ID17293
Feature activation+1.07e+01

Pos logit contributions
bones+0.180
cks+0.179
 splits+0.175
hole+0.175
ps+0.171
Neg logit contributions
Jack-0.140
 Pete-0.139
 Benny-0.139
 Fro-0.139
 him-0.138
Token
Token ID6184
Feature activation+1.38e+01

Pos logit contributions
Un+0.164
Jack+0.159
David+0.159
Kids+0.157
 concession+0.156
Neg logit contributions
 on-0.085
 pieces-0.081
hole-0.081
ps-0.079
 atop-0.076
Token
Token ID95
Feature activation+1.33e+01

Pos logit contributions
Un+0.141
Da+0.139
Jack+0.138
David+0.135
Jim+0.131
Neg logit contributions
hole-0.178
ps-0.167
 pieces-0.163
 on-0.163
 edges-0.161
Token
Token ID26391
Feature activation+1.87e+01

Pos logit contributions
 Fro+0.129
Un+0.128
Jack+0.127
Da+0.124
David+0.123
Neg logit contributions
hole-0.177
 pieces-0.166
ps-0.164
 edges-0.162
 on-0.161
Token
Token ID129
Feature activation+1.45e+01

Pos logit contributions
 concession+0.182
Da+0.180
Un+0.177
Jack+0.175
Kids+0.173
Neg logit contributions
hole-0.245
 on-0.235
 pieces-0.221
 atop-0.219
 trunk-0.214
Token
Token ID241
Feature activation+2.11e+01

Pos logit contributions
Jack+0.140
Un+0.135
 Fro+0.135
Da+0.134
 concession+0.134
Neg logit contributions
hole-0.189
ps-0.181
 pieces-0.177
 edges-0.176
 atop-0.174
TokenRun
Token ID10987
Feature activation+5.56e+00

Pos logit contributions
 Fro+0.182
oph+0.180
 concession+0.179
Un+0.175
Da+0.173
Neg logit contributions
 on-0.293
hole-0.289
 pieces-0.282
 edges-0.269
 atop-0.266
Token Tony
Token ID8832
Feature activation+8.00e+00

Pos logit contributions
Un+0.078
Jack+0.078
David+0.078
 Fro+0.077
Jim+0.075
Neg logit contributions
hole-0.049
 pieces-0.049
 on-0.045
 atop-0.044
 edges-0.044
Token,
Token ID11
Feature activation+1.05e+01

Pos logit contributions
Un+0.105
 Fro+0.104
Jack+0.104
David+0.104
Kids+0.100
Neg logit contributions
 pieces-0.078
hole-0.078
 atop-0.072
 edges-0.071
 falling-0.069
Token Run
Token ID5660
Feature activation+2.03e+00

Pos logit contributions
Un+0.101
oph+0.099
Jack+0.099
David+0.097
Da+0.096
Neg logit contributions
 on-0.137
hole-0.134
 pieces-0.133
 falling-0.123
 lid-0.122
Token!
Token ID0
Feature activation+1.64e+01

Pos logit contributions
Jack+0.029
David+0.029
Un+0.029
 Fro+0.028
Jim+0.028
Neg logit contributions
hole-0.018
 pieces-0.017
 on-0.017
 edges-0.016
ps-0.016
Token himself
Token ID2241
Feature activation+1.51e+01

Pos logit contributions
hole+0.026
 pieces+0.026
 atop+0.024
 on+0.024
 edges+0.024
Neg logit contributions
Un-0.020
Jack-0.019
David-0.019
 Fro-0.019
 concession-0.018
Token
Token ID6184
Feature activation+1.27e+01

Pos logit contributions
 Fro+0.179
Jack+0.179
David+0.174
Un+0.169
Jim+0.169
Neg logit contributions
hole-0.164
 pieces-0.160
 atop-0.147
 edges-0.147
ps-0.146
Token
Token ID95
Feature activation+1.32e+01

Pos logit contributions
Un+0.130
Jack+0.129
David+0.129
 Fro+0.127
Da+0.123
Neg logit contributions
hole-0.161
 pieces-0.153
 atop-0.150
 on-0.147
 edges-0.145
Token
Token ID26391
Feature activation+1.85e+01

Pos logit contributions
 Fro+0.121
Un+0.120
Jack+0.120
David+0.119
Da+0.114
Neg logit contributions
hole-0.181
 pieces-0.175
 atop-0.168
 on-0.167
 edges-0.167
Token
Token ID129
Feature activation+1.39e+01

Pos logit contributions
 concession+0.180
Da+0.176
Un+0.176
Kids+0.174
Jack+0.171
Neg logit contributions
hole-0.242
 on-0.229
 pieces-0.217
 atop-0.216
 trunk-0.210
Token
Token ID241
Feature activation+2.09e+01

Pos logit contributions
Jack+0.127
Un+0.127
David+0.125
 Fro+0.124
Jim+0.118
Neg logit contributions
hole-0.191
 pieces-0.190
 atop-0.179
 on-0.178
 edges-0.177
TokenSometimes
Token ID15468
Feature activation+8.89e+00

Pos logit contributions
oph+0.178
 Fro+0.175
 concession+0.172
Da+0.169
David+0.167
Neg logit contributions
 on-0.293
hole-0.289
 pieces-0.286
 edges-0.275
 atop-0.268
Token it
Token ID340
Feature activation+5.50e+00

Pos logit contributions
Un+0.110
Jack+0.109
 Fro+0.109
David+0.108
Kids+0.104
Neg logit contributions
hole-0.095
 pieces-0.092
 atop-0.087
ps-0.086
 edges-0.084
Tokenâ
Token ID22940
Feature activation+1.34e+01

Pos logit contributions
Un+0.062
 Fro+0.061
Jack+0.061
 concession+0.059
Kids+0.059
Neg logit contributions
hole-0.064
 pieces-0.061
 atop-0.060
 edges-0.057
ps-0.057
Token
Token ID26391
Feature activation+1.55e+01

Pos logit contributions
Jack+0.073
Kids+0.072
Un+0.072
David+0.070
 Fro+0.066
Neg logit contributions
 pieces-0.243
hole-0.237
 lid-0.227
 atop-0.224
 edges-0.224
Token
Token ID8151
Feature activation+1.11e+01

Pos logit contributions
Un+0.142
Jack+0.142
 Fro+0.139
Da+0.139
 concession+0.139
Neg logit contributions
hole-0.207
 pieces-0.200
 on-0.197
 atop-0.189
 falling-0.188
Token
Token ID198
Feature activation+2.18e+01

Pos logit contributions
hole+0.048
 pieces+0.048
 atop+0.045
 edges+0.044
ps+0.043
Neg logit contributions
Un-0.045
Jack-0.045
David-0.045
 Fro-0.045
 concession-0.043
Token
Token ID198
Feature activation+1.92e+01

Pos logit contributions
 concession+0.222
oph+0.220
latable+0.210
 Fro+0.207
ory+0.195
Neg logit contributions
 on-0.254
 pieces-0.252
hole-0.248
 atop-0.236
 slot-0.227
Tokenâ
Token ID22940
Feature activation+1.85e+01

Pos logit contributions
 Fro+0.182
 concession+0.177
oph+0.171
latable+0.170
elt+0.168
Neg logit contributions
 pieces-0.226
hole-0.225
 on-0.219
 atop-0.210
 slot-0.208
Token
Token ID26391
Feature activation+2.02e+01

Pos logit contributions
Jack+0.130
David+0.127
Un+0.123
Jim+0.118
Kids+0.118
Neg logit contributions
 pieces-0.312
hole-0.295
 on-0.287
 lid-0.284
 edges-0.283
Token
Token ID129
Feature activation+1.51e+01

Pos logit contributions
 concession+0.195
Un+0.193
Kids+0.189
 Fro+0.187
Jack+0.186
Neg logit contributions
hole-0.270
 pieces-0.251
 on-0.248
 atop-0.243
 falling-0.241
Token
Token ID241
Feature activation+2.08e+01

Pos logit contributions
Jack+0.155
Un+0.155
David+0.153
 Fro+0.153
 concession+0.147
Neg logit contributions
hole-0.189
 pieces-0.185
 atop-0.176
 edges-0.175
ps-0.175
TokenCome
Token ID16773
Feature activation+6.79e+00

Pos logit contributions
 Fro+0.168
oph+0.161
 concession+0.161
Un+0.157
Da+0.151
Neg logit contributions
hole-0.308
 on-0.301
 pieces-0.298
 atop-0.287
 edges-0.286
Token on
Token ID319
Feature activation+7.47e+00

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
oph+0.128
Da+0.128
Neg logit contributions
 on-0.028
hole-0.025
 pieces-0.022
 atop-0.021
 falling-0.019
Token now
Token ID783
Feature activation+5.85e+00

Pos logit contributions
Un+0.092
Jack+0.091
David+0.091
 Fro+0.089
Jim+0.088
Neg logit contributions
 pieces-0.079
hole-0.079
 on-0.075
 atop-0.073
 falling-0.072
Token,
Token ID11
Feature activation+9.88e+00

Pos logit contributions
Un+0.081
Jack+0.081
David+0.081
 Fro+0.080
Jim+0.078
Neg logit contributions
hole-0.053
 pieces-0.052
 atop-0.048
 on-0.047
 edges-0.047
Token Katie
Token ID24721
Feature activation+8.25e+00

Pos logit contributions
Un+0.100
Jack+0.099
David+0.098
Da+0.096
Jim+0.095
Neg logit contributions
 on-0.124
 pieces-0.122
hole-0.122
 edges-0.112
 falling-0.112
Token
Token ID198
Feature activation+2.18e+01

Pos logit contributions
hole+0.056
 pieces+0.055
 atop+0.051
ps+0.050
 edges+0.050
Neg logit contributions
Un-0.060
Jack-0.059
David-0.059
 Fro-0.057
 concession-0.057
Token
Token ID198
Feature activation+1.92e+01

Pos logit contributions
 concession+0.222
oph+0.220
latable+0.210
 Fro+0.207
ory+0.195
Neg logit contributions
 on-0.254
 pieces-0.252
hole-0.248
 atop-0.236
 slot-0.227
Tokenâ
Token ID22940
Feature activation+1.85e+01

Pos logit contributions
 Fro+0.182
 concession+0.177
oph+0.171
latable+0.170
elt+0.168
Neg logit contributions
 pieces-0.226
hole-0.225
 on-0.219
 atop-0.210
 slot-0.208
Token
Token ID26391
Feature activation+2.02e+01

Pos logit contributions
Jack+0.130
David+0.127
Un+0.123
Jim+0.118
Kids+0.118
Neg logit contributions
 pieces-0.312
hole-0.295
 on-0.287
 lid-0.284
 edges-0.283
Token
Token ID129
Feature activation+1.51e+01

Pos logit contributions
 concession+0.195
Un+0.193
Kids+0.189
 Fro+0.187
Jack+0.186
Neg logit contributions
hole-0.270
 pieces-0.251
 on-0.248
 atop-0.243
 falling-0.241
Token
Token ID241
Feature activation+2.08e+01

Pos logit contributions
Jack+0.155
Un+0.155
David+0.153
 Fro+0.153
 concession+0.147
Neg logit contributions
hole-0.189
 pieces-0.185
 atop-0.176
 edges-0.175
ps-0.175
TokenHello
Token ID15496
Feature activation+5.71e+00

Pos logit contributions
 Fro+0.168
oph+0.161
 concession+0.161
Un+0.157
Da+0.151
Neg logit contributions
hole-0.308
 on-0.301
 pieces-0.298
 atop-0.287
 edges-0.286
Token!
Token ID0
Feature activation+1.69e+01

Pos logit contributions
 Fro+0.068
Jack+0.068
Un+0.067
David+0.066
Johnny+0.065
Neg logit contributions
 pieces-0.064
hole-0.064
ps-0.061
 edges-0.060
 atop-0.059
Token Where
Token ID6350
Feature activation+1.05e+01

Pos logit contributions
Un+0.147
David+0.144
 concession+0.144
 Fro+0.144
Jack+0.144
Neg logit contributions
hole-0.233
 pieces-0.222
 on-0.221
 atop-0.215
 edges-0.207
Token are
Token ID389
Feature activation+7.22e+00

Pos logit contributions
Un+0.128
Jack+0.127
David+0.126
 Fro+0.126
 concession+0.121
Neg logit contributions
hole-0.113
 pieces-0.112
 on-0.104
 atop-0.103
 edges-0.102
Token you
Token ID345
Feature activation+4.65e+00

Pos logit contributions
Un+0.086
David+0.085
 concession+0.084
 Fro+0.083
enny+0.082
Neg logit contributions
 pieces-0.082
 on-0.076
 changing-0.070
 falling-0.069
 hardest-0.069
Token perfect
Token ID2818
Feature activation+1.00e+01

Pos logit contributions
Un+0.220
Jack+0.220
David+0.218
 Fro+0.216
Jim+0.209
Neg logit contributions
hole-0.185
 pieces-0.185
 atop-0.170
 on-0.169
 edges-0.168
Token bag
Token ID6131
Feature activation+1.57e+00

Pos logit contributions
 Fro+0.128
Jack+0.127
Un+0.125
David+0.124
 concession+0.124
Neg logit contributions
hole-0.097
 pieces-0.090
 atop-0.089
ps-0.085
els-0.084
Token
Token ID6184
Feature activation+9.11e+00

Pos logit contributions
Jack+0.020
 concession+0.020
enny+0.020
Un+0.020
David+0.019
Neg logit contributions
hole-0.014
 pieces-0.014
ps-0.014
els-0.014
 atop-0.014
Token
Token ID95
Feature activation+1.14e+01

Pos logit contributions
Un+0.089
Jack+0.089
David+0.088
 Fro+0.086
 concession+0.085
Neg logit contributions
 pieces-0.120
hole-0.119
ps-0.112
 atop-0.112
 falling-0.111
Token
Token ID26391
Feature activation+1.67e+01

Pos logit contributions
Un+0.110
Jack+0.110
David+0.109
 Fro+0.109
Jim+0.104
Neg logit contributions
hole-0.152
 pieces-0.150
 atop-0.141
 on-0.140
 edges-0.140
Token
Token ID447
Feature activation+2.05e+01

Pos logit contributions
Un+0.167
 Fro+0.165
Jack+0.164
 concession+0.164
Da+0.164
Neg logit contributions
hole-0.214
 on-0.205
 pieces-0.199
 atop-0.191
 falling-0.187
Token
Token ID250
Feature activation+1.62e+01

Pos logit contributions
Un+0.209
David+0.207
 Fro+0.206
Jack+0.204
Kids+0.198
Neg logit contributions
hole-0.255
 pieces-0.252
 on-0.242
 atop-0.241
 edges-0.231
Token now
Token ID783
Feature activation+9.01e+00

Pos logit contributions
Un+0.104
Jack+0.103
David+0.102
 Fro+0.095
Jim+0.093
Neg logit contributions
hole-0.271
 pieces-0.270
 atop-0.259
ps-0.257
 falling-0.256
Token I
Token ID314
Feature activation+9.55e+00

Pos logit contributions
Un+0.095
David+0.094
Jack+0.094
 Fro+0.093
 concession+0.090
Neg logit contributions
 pieces-0.111
hole-0.110
 on-0.107
 atop-0.103
 edges-0.102
Token'm
Token ID1101
Feature activation+9.80e+00

Pos logit contributions
Jack+0.101
Un+0.099
 Fro+0.099
David+0.099
Kids+0.095
Neg logit contributions
hole-0.118
 pieces-0.118
 atop-0.111
 edges-0.110
ps-0.109
Token very
Token ID845
Feature activation+1.08e+01

Pos logit contributions
 Fro+0.103
Un+0.101
David+0.101
Jack+0.100
 concession+0.097
Neg logit contributions
hole-0.125
 pieces-0.125
 edges-0.116
 atop-0.114
 lid-0.114
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.145
ps+0.141
cks+0.139
 hardest+0.130
 increasing+0.129
Neg logit contributions
 wife-0.111
Her-0.109
His-0.108
Jack-0.108
 Her-0.107
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenThanks
Token ID9690
Feature activation+7.41e+00

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token,
Token ID11
Feature activation+1.06e+01

Pos logit contributions
Un+0.103
Jack+0.103
David+0.102
 Fro+0.102
Jim+0.098
Neg logit contributions
hole-0.067
 pieces-0.067
 atop-0.061
 on-0.060
 edges-0.060
Token Jack
Token ID3619
Feature activation+8.42e+00

Pos logit contributions
Un+0.102
David+0.100
Jack+0.100
 Fro+0.096
Da+0.095
Neg logit contributions
 pieces-0.138
 on-0.137
hole-0.136
 falling-0.127
 edges-0.127
Token.
Token ID13
Feature activation+1.64e+01

Pos logit contributions
 Fro+0.108
Jack+0.107
David+0.105
Un+0.104
Jim+0.101
Neg logit contributions
hole-0.085
 pieces-0.082
 atop-0.077
 edges-0.077
 lid-0.075
Token You
Token ID921
Feature activation+8.72e+00

Pos logit contributions
Un+0.162
David+0.159
 concession+0.158
Da+0.157
Jack+0.157
Neg logit contributions
hole-0.208
 pieces-0.195
 on-0.192
 edges-0.183
 trunk-0.182
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
 pieces+0.013
 on+0.013
hole+0.012
 atop+0.012
 falling+0.012
Neg logit contributions
David-0.013
Jack-0.013
Un-0.013
Jim-0.012
 Fro-0.012
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenWe
Token ID1135
Feature activation+1.11e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token use
Token ID779
Feature activation+5.17e+00

Pos logit contributions
Jack+0.129
David+0.127
 Fro+0.126
Un+0.125
Jim+0.121
Neg logit contributions
hole-0.128
 pieces-0.126
 atop-0.118
 edges-0.117
ps-0.115
Token the
Token ID262
Feature activation+8.08e+00

Pos logit contributions
Jack+0.051
Un+0.051
 Fro+0.050
David+0.050
 concession+0.048
Neg logit contributions
hole-0.066
 atop-0.063
 lid-0.058
ps-0.057
 globe-0.057
Token gun
Token ID2485
Feature activation-1.30e+00

Pos logit contributions
 concession+0.095
 Fro+0.094
 shepherd+0.093
 wife+0.092
 please+0.089
Neg logit contributions
hole-0.086
cks-0.079
 atop-0.076
bones-0.074
ps-0.070
Token.
Token ID13
Feature activation+1.53e+01

Pos logit contributions
hole+0.015
 atop+0.014
 pieces+0.014
 lid+0.014
 edges+0.014
Neg logit contributions
Un-0.013
 concession-0.013
Jack-0.013
 Fro-0.012
David-0.012
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.153
 pieces+0.149
 atop+0.147
 edges+0.144
ps+0.144
Neg logit contributions
 Fro-0.101
Un-0.099
Jack-0.098
David-0.095
Jim-0.093
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenIt
Token ID1026
Feature activation+8.28e+00

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token means
Token ID1724
Feature activation+7.62e+00

Pos logit contributions
Jack+0.080
Un+0.078
 Fro+0.078
 concession+0.077
Kids+0.076
Neg logit contributions
hole-0.106
 pieces-0.103
 atop-0.102
 falling-0.097
 edges-0.097
Token I
Token ID314
Feature activation+8.43e+00

Pos logit contributions
 Fro+0.071
Jack+0.071
Un+0.071
David+0.071
Johnny+0.068
Neg logit contributions
hole-0.105
 pieces-0.101
 atop-0.098
ps-0.096
 edges-0.094
Token'm
Token ID1101
Feature activation+9.12e+00

Pos logit contributions
Jack+0.082
 Fro+0.082
Un+0.081
David+0.080
Kids+0.077
Neg logit contributions
 pieces-0.111
hole-0.111
 atop-0.104
 edges-0.103
ps-0.102
Token gonna
Token ID8066
Feature activation+9.08e+00

Pos logit contributions
 Fro+0.096
Un+0.096
David+0.093
Jack+0.091
 concession+0.089
Neg logit contributions
hole-0.120
 pieces-0.119
 lid-0.109
 edges-0.109
 atop-0.108
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.093
 pieces+0.091
 atop+0.083
 edges+0.081
ps+0.080
Neg logit contributions
Un-0.165
Jack-0.164
David-0.163
 Fro-0.162
Kids-0.157
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenHope
Token ID34456
Feature activation+8.26e+00

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token you
Token ID345
Feature activation+8.97e+00

Pos logit contributions
 Fro+0.096
Jack+0.093
Un+0.091
 concession+0.090
David+0.089
Neg logit contributions
hole-0.093
 pieces-0.086
 atop-0.082
 edges-0.082
ps-0.081
Token'll
Token ID1183
Feature activation+8.86e+00

Pos logit contributions
 Fro+0.078
Jack+0.078
David+0.073
 please+0.071
 concession+0.070
Neg logit contributions
hole-0.123
 atop-0.122
 pieces-0.121
els-0.121
bones-0.121
Token be
Token ID307
Feature activation+6.28e+00

Pos logit contributions
Un+0.090
Jack+0.090
David+0.089
 Fro+0.088
Jim+0.085
Neg logit contributions
hole-0.113
 pieces-0.112
 atop-0.105
 on-0.105
 edges-0.104
Token more
Token ID517
Feature activation+5.23e+00

Pos logit contributions
Jack+0.073
Un+0.073
David+0.073
 Fro+0.072
Jim+0.070
Neg logit contributions
hole-0.071
 pieces-0.070
 atop-0.065
 edges-0.065
ps-0.063
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
 concession+0.032
oph+0.030
David+0.030
 Fro+0.029
Un+0.029
Neg logit contributions
 on-0.039
 pieces-0.038
hole-0.037
 atop-0.035
 trunk-0.035
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenWill
Token ID8743
Feature activation+9.59e+00

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token you
Token ID345
Feature activation+7.41e+00

Pos logit contributions
Un+0.105
Jack+0.105
 Fro+0.104
David+0.104
Jim+0.100
Neg logit contributions
hole-0.114
 pieces-0.114
 atop-0.107
 edges-0.105
 on-0.104
Token come
Token ID1282
Feature activation+4.89e+00

Pos logit contributions
 Fro+0.070
Un+0.068
Jack+0.068
David+0.067
 concession+0.065
Neg logit contributions
 pieces-0.100
hole-0.098
 edges-0.093
 atop-0.093
els-0.091
Token listen
Token ID6004
Feature activation+3.04e+00

Pos logit contributions
Un+0.070
Jack+0.070
David+0.069
 Fro+0.069
 concession+0.067
Neg logit contributions
hole-0.042
 pieces-0.042
 atop-0.038
 edges-0.038
 on-0.037
Token to
Token ID284
Feature activation+7.86e+00

Pos logit contributions
Un+0.030
Jack+0.030
David+0.029
 concession+0.028
 Fro+0.028
Neg logit contributions
 pieces-0.039
hole-0.039
 atop-0.037
ps-0.037
 lid-0.037
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.134
 lid+0.132
ps+0.132
 button+0.127
 globe+0.126
Neg logit contributions
Jack-0.117
Un-0.116
David-0.112
Da-0.107
He-0.106
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenThis
Token ID1212
Feature activation+1.01e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token box
Token ID3091
Feature activation+2.04e+00

Pos logit contributions
 Fro+0.125
 concession+0.124
Jack+0.123
David+0.122
Kids+0.121
Neg logit contributions
hole-0.103
 atop-0.098
ps-0.094
 pieces-0.089
 edges-0.089
Token is
Token ID318
Feature activation+7.09e+00

Pos logit contributions
 concession+0.024
Jack+0.023
enny+0.023
addy+0.022
 Fro+0.022
Neg logit contributions
 atop-0.020
hole-0.020
 pieces-0.019
 falling-0.019
 edges-0.018
Token probably
Token ID2192
Feature activation+7.39e+00

Pos logit contributions
Un+0.065
 concession+0.060
 please+0.059
Johnny+0.059
 Fro+0.058
Neg logit contributions
hole-0.094
els-0.092
ps-0.091
 atop-0.090
 pieces-0.088
Token too
Token ID1165
Feature activation+6.48e+00

Pos logit contributions
 Fro+0.078
 concession+0.077
Un+0.076
Jack+0.073
David+0.071
Neg logit contributions
hole-0.088
 pieces-0.084
els-0.082
 atop-0.080
 lid-0.080
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.056
 pieces+0.056
 atop+0.052
 edges+0.052
ps+0.052
Neg logit contributions
Jack-0.040
Un-0.039
 Fro-0.039
David-0.039
Jim-0.037
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenLet
Token ID5756
Feature activation+9.26e+00

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token me
Token ID502
Feature activation+8.32e+00

Pos logit contributions
 Fro+0.139
Un+0.138
David+0.136
 concession+0.135
Da+0.134
Neg logit contributions
 on-0.078
 pieces-0.076
 way-0.069
hole-0.068
 button-0.066
Token take
Token ID1011
Feature activation+2.29e+00

Pos logit contributions
Un+0.090
Jack+0.088
David+0.088
 Fro+0.086
Kids+0.084
Neg logit contributions
hole-0.102
 pieces-0.101
 on-0.099
 atop-0.095
ps-0.093
Token a
Token ID257
Feature activation+8.18e+00

Pos logit contributions
Jack+0.025
Un+0.024
 Fro+0.024
David+0.024
Jim+0.023
Neg logit contributions
hole-0.029
 pieces-0.028
 atop-0.027
 edges-0.026
 falling-0.026
Token look
Token ID804
Feature activation+3.02e+00

Pos logit contributions
Jack+0.091
 Fro+0.089
Un+0.087
enny+0.086
David+0.086
Neg logit contributions
ps-0.100
hole-0.100
 pieces-0.096
els-0.095
 atop-0.094
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
 pieces+0.192
hole+0.186
ps+0.184
 edges+0.179
els+0.178
Neg logit contributions
Un-0.183
His-0.175
Jack-0.174
 Fro-0.173
David-0.173
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenI
Token ID40
Feature activation+1.33e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token was
Token ID373
Feature activation+9.44e+00

Pos logit contributions
Un+0.124
David+0.123
Jack+0.122
 Fro+0.120
 concession+0.116
Neg logit contributions
hole-0.181
 pieces-0.179
 on-0.172
 atop-0.169
 edges-0.166
Token dreaming
Token ID28368
Feature activation+2.79e+00

Pos logit contributions
 Fro+0.103
Un+0.102
David+0.102
Jack+0.101
 concession+0.097
Neg logit contributions
hole-0.117
 pieces-0.115
 atop-0.106
 lid-0.106
 edges-0.106
Token about
Token ID546
Feature activation+7.09e+00

Pos logit contributions
Jack+0.030
Un+0.029
 Fro+0.029
 concession+0.028
David+0.028
Neg logit contributions
hole-0.034
 pieces-0.033
 atop-0.032
els-0.031
 edges-0.031
Token this
Token ID428
Feature activation+4.83e+00

Pos logit contributions
Un+0.078
 Fro+0.078
Jack+0.078
David+0.077
 concession+0.074
Neg logit contributions
hole-0.087
 pieces-0.082
 atop-0.079
 lid-0.078
ps-0.077
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.060
 pieces+0.059
 atop+0.055
ps+0.054
 edges+0.054
Neg logit contributions
Jack-0.070
David-0.070
Un-0.070
 Fro-0.069
Jim-0.067
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenThat
Token ID2504
Feature activation+1.20e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Tokenâ
Token ID22940
Feature activation+1.43e+01

Pos logit contributions
Jack+0.142
Un+0.142
 Fro+0.140
David+0.140
 concession+0.136
Neg logit contributions
hole-0.133
 pieces-0.129
 atop-0.125
ps-0.123
 edges-0.120
Token
Token ID26391
Feature activation+1.64e+01

Pos logit contributions
Jack+0.090
Kids+0.090
David+0.087
Un+0.084
Johnny+0.082
Neg logit contributions
 pieces-0.250
hole-0.240
 edges-0.231
 lid-0.230
ps-0.229
Token
Token ID8151
Feature activation+1.27e+01

Pos logit contributions
Un+0.159
Jack+0.157
 Fro+0.156
 concession+0.156
David+0.154
Neg logit contributions
hole-0.212
 pieces-0.205
 on-0.203
 atop-0.194
 falling-0.191
Tokens
Token ID82
Feature activation+1.13e+01

Pos logit contributions
Jack+0.134
 Fro+0.132
David+0.131
Un+0.131
Jim+0.126
Neg logit contributions
 pieces-0.157
hole-0.156
 atop-0.147
 edges-0.147
ps-0.146
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.058
 pieces+0.057
els+0.054
ool+0.054
 edges+0.053
Neg logit contributions
Un-0.023
Jack-0.022
 Fro-0.021
His-0.021
David-0.021
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenI
Token ID40
Feature activation+1.33e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token can
Token ID460
Feature activation+1.05e+01

Pos logit contributions
Un+0.124
David+0.123
Jack+0.122
 Fro+0.120
 concession+0.116
Neg logit contributions
hole-0.181
 pieces-0.179
 on-0.172
 atop-0.169
 edges-0.166
Token fly
Token ID6129
Feature activation-4.31e-01

Pos logit contributions
Un+0.131
 Fro+0.128
David+0.127
Da+0.127
Jack+0.125
Neg logit contributions
hole-0.114
 pieces-0.112
 on-0.109
 falling-0.103
 edges-0.103
Token,
Token ID11
Feature activation+9.50e+00

Pos logit contributions
 pieces+0.004
hole+0.004
 on+0.004
 atop+0.004
 edges+0.004
Neg logit contributions
David-0.006
Un-0.006
Jack-0.006
 Fro-0.006
Kids-0.006
Token I
Token ID314
Feature activation+8.92e+00

Pos logit contributions
Un+0.104
Jack+0.103
David+0.102
Da+0.100
Jim+0.099
Neg logit contributions
 on-0.113
 pieces-0.112
hole-0.111
 falling-0.102
 edges-0.101
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
hole+0.128
 pieces+0.125
 atop+0.123
els+0.122
ps+0.120
Neg logit contributions
David-0.088
Un-0.087
Jack-0.085
 Fro-0.084
His-0.083
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenThat
Token ID2504
Feature activation+1.20e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Tokenâ
Token ID22940
Feature activation+1.43e+01

Pos logit contributions
Jack+0.142
Un+0.142
 Fro+0.140
David+0.140
 concession+0.136
Neg logit contributions
hole-0.133
 pieces-0.129
 atop-0.125
ps-0.123
 edges-0.120
Token
Token ID26391
Feature activation+1.64e+01

Pos logit contributions
Jack+0.090
Kids+0.090
David+0.087
Un+0.084
Johnny+0.082
Neg logit contributions
 pieces-0.250
hole-0.240
 edges-0.231
 lid-0.230
ps-0.229
Token
Token ID8151
Feature activation+1.27e+01

Pos logit contributions
Un+0.159
Jack+0.157
 Fro+0.156
 concession+0.156
David+0.154
Neg logit contributions
hole-0.212
 pieces-0.205
 on-0.203
 atop-0.194
 falling-0.191
Tokens
Token ID82
Feature activation+1.13e+01

Pos logit contributions
Jack+0.134
 Fro+0.132
David+0.131
Un+0.131
Jim+0.126
Neg logit contributions
 pieces-0.157
hole-0.156
 atop-0.147
 edges-0.147
ps-0.146
Token,
Token ID11
Feature activation+1.69e+01

Pos logit contributions
Un+0.030
 Fro+0.029
Jack+0.029
David+0.029
Da+0.029
Neg logit contributions
 pieces-0.058
hole-0.057
 on-0.055
 edges-0.053
 atop-0.053
Token
Token ID6184
Feature activation+1.28e+01

Pos logit contributions
Un+0.176
Jack+0.174
David+0.174
 Fro+0.171
Kids+0.166
Neg logit contributions
 pieces-0.211
hole-0.210
 on-0.200
 atop-0.196
 edges-0.194
Token
Token ID95
Feature activation+1.24e+01

Pos logit contributions
Un+0.133
Jack+0.132
David+0.132
 Fro+0.131
 concession+0.126
Neg logit contributions
hole-0.161
 pieces-0.158
 atop-0.151
 on-0.149
 edges-0.147
Token
Token ID26391
Feature activation+1.83e+01

Pos logit contributions
 Fro+0.124
Un+0.122
Jack+0.122
David+0.120
Jim+0.116
Neg logit contributions
hole-0.161
 pieces-0.157
 atop-0.149
 edges-0.148
 on-0.148
Token
Token ID129
Feature activation+1.33e+01

Pos logit contributions
 concession+0.172
Un+0.170
Da+0.170
 Fro+0.168
Jack+0.168
Neg logit contributions
hole-0.246
 on-0.231
 pieces-0.226
 atop-0.221
 falling-0.217
Token
Token ID241
Feature activation+2.04e+01

Pos logit contributions
Jack+0.130
Un+0.130
David+0.129
 Fro+0.128
 concession+0.123
Neg logit contributions
hole-0.174
 pieces-0.172
 atop-0.162
 on-0.161
 edges-0.161
TokenThat
Token ID2504
Feature activation+1.20e+01

Pos logit contributions
 Fro+0.173
oph+0.172
 concession+0.167
Un+0.162
Da+0.162
Neg logit contributions
hole-0.286
 on-0.286
 pieces-0.280
 edges-0.270
 atop-0.263
Token's
Token ID338
Feature activation+1.15e+01

Pos logit contributions
Jack+0.142
Un+0.142
 Fro+0.140
David+0.140
 concession+0.136
Neg logit contributions
hole-0.133
 pieces-0.129
 atop-0.125
ps-0.123
 edges-0.120
Token the
Token ID262
Feature activation+9.42e+00

Pos logit contributions
Un+0.118
Jack+0.114
 Fro+0.114
David+0.112
Johnny+0.110
Neg logit contributions
 pieces-0.152
hole-0.151
 atop-0.142
ps-0.137
 edges-0.136
Token most
Token ID749
Feature activation+7.37e+00

Pos logit contributions
Un+0.119
David+0.116
 Fro+0.115
 concession+0.115
Jack+0.114
Neg logit contributions
hole-0.103
 atop-0.091
 pieces-0.088
ps-0.088
els-0.088
Token important
Token ID1593
Feature activation+7.11e+00

Pos logit contributions
Jack+0.084
Johnny+0.080
 concession+0.079
Un+0.079
Kids+0.079
Neg logit contributions
hole-0.090
ps-0.086
 pieces-0.083
 atop-0.082
 lid-0.080
Token saw
Token ID2497
Feature activation-6.85e+00

Pos logit contributions
 pieces+0.075
hole+0.074
 on+0.070
 atop+0.068
 falling+0.067
Neg logit contributions
Un-0.075
David-0.074
Jack-0.074
 Fro-0.073
Jim-0.071
Token some
Token ID617
Feature activation-1.06e+01

Pos logit contributions
 pieces+0.088
hole+0.086
 on+0.084
 atop+0.082
ps+0.082
Neg logit contributions
David-0.071
Jack-0.070
Un-0.069
Kids-0.066
Jim-0.066
Token more
Token ID517
Feature activation-1.05e+01

Pos logit contributions
hole+0.126
 atop+0.111
ps+0.106
 globe+0.105
 edges+0.104
Neg logit contributions
 Fro-0.124
Jack-0.124
 concession-0.123
Un-0.119
David-0.117
Token che
Token ID1125
Feature activation-1.47e+01

Pos logit contributions
hole+0.105
 atop+0.100
ps+0.097
 hardest+0.093
 lid+0.088
Neg logit contributions
Jack-0.136
 Fro-0.132
enny-0.130
 concession-0.127
Johnny-0.127
Tokenwed
Token ID19103
Feature activation-1.66e+01

Pos logit contributions
hole+0.146
 button+0.139
 ladder+0.139
 changing+0.138
 lid+0.137
Neg logit contributions
 Fro-0.177
Jack-0.176
ackers-0.175
enny-0.174
Jim-0.170
Token threads
Token ID14390
Feature activation-2.12e+01

Pos logit contributions
hole+0.184
 code+0.177
 mant+0.161
 slot+0.161
 atop+0.160
Neg logit contributions
 Fro-0.201
 him-0.187
Jack-0.183
He-0.178
 He-0.178
Token in
Token ID287
Feature activation-7.98e+00

Pos logit contributions
 edges+0.239
ps+0.236
hole+0.235
 lid+0.235
els+0.232
Neg logit contributions
Jack-0.206
 Fro-0.199
enny-0.195
He-0.190
They-0.189
Token the
Token ID262
Feature activation-5.57e+00

Pos logit contributions
hole+0.117
 atop+0.111
ool+0.105
 hardest+0.105
 edges+0.103
Neg logit contributions
 Fro-0.071
Un-0.070
Jack-0.065
His-0.064
Da-0.062
Token house
Token ID2156
Feature activation-1.61e+01

Pos logit contributions
ps+0.056
hole+0.056
cks+0.055
els+0.054
 atop+0.054
Neg logit contributions
Un-0.070
Jack-0.070
Da-0.068
 Fro-0.067
Brother-0.067
Token.
Token ID13
Feature activation+3.52e+00

Pos logit contributions
 atop+0.172
hole+0.169
ps+0.167
els+0.167
 edges+0.166
Neg logit contributions
Un-0.172
Jack-0.172
 concession-0.168
 Fro-0.167
David-0.164
Token Fl
Token ID1610
Feature activation-1.97e+00

Pos logit contributions
David+0.026
Jack+0.024
Un+0.024
Jim+0.023
Johnny+0.022
Neg logit contributions
 pieces-0.057
hole-0.056
 on-0.055
 edges-0.055
 lid-0.054
Token she
Token ID673
Feature activation-4.87e+00

Pos logit contributions
hole+0.035
 pieces+0.034
 edges+0.032
ps+0.032
 lid+0.032
Neg logit contributions
Jack-0.022
Un-0.022
 Fro-0.022
David-0.022
Johnny-0.021
Token continued
Token ID3767
Feature activation-7.59e+00

Pos logit contributions
hole+0.070
 pieces+0.067
 edges+0.066
ps+0.065
 atop+0.065
Neg logit contributions
Jack-0.043
Un-0.042
 Fro-0.042
Kids-0.040
David-0.040
Token picking
Token ID10868
Feature activation-1.54e+01

Pos logit contributions
hole+0.105
 pieces+0.094
 edges+0.093
 lid+0.093
 wires+0.088
Neg logit contributions
 Fro-0.077
Un-0.075
His-0.071
David-0.070
 His-0.069
Token her
Token ID607
Feature activation-8.34e+00

Pos logit contributions
hole+0.182
 code+0.162
 falling+0.147
 edges+0.145
ool+0.145
Neg logit contributions
 Fro-0.200
Da-0.175
 Pete-0.166
Un-0.166
Jack-0.165
Token p
Token ID279
Feature activation-1.15e+01

Pos logit contributions
hole+0.123
 slot+0.108
 code+0.103
bones+0.101
 atop+0.100
Neg logit contributions
 Fro-0.086
 Her-0.070
 His-0.070
omo-0.069
 concession-0.069
Tokenears
Token ID4127
Feature activation-2.12e+01

Pos logit contributions
hole+0.125
 pieces+0.124
 atop+0.115
 on+0.114
 edges+0.113
Neg logit contributions
Un-0.139
 Fro-0.138
Jack-0.138
David-0.138
 concession-0.133
Token.
Token ID13
Feature activation+4.48e+00

Pos logit contributions
hole+0.202
ps+0.201
 edges+0.197
 pieces+0.197
 lid+0.190
Neg logit contributions
 Fro-0.261
Un-0.258
Jack-0.252
His-0.244
enny-0.240
Token With
Token ID2080
Feature activation-4.73e+00

Pos logit contributions
 concession+0.032
Jim+0.032
Un+0.032
David+0.032
Jack+0.031
Neg logit contributions
hole-0.068
 on-0.067
 pieces-0.067
 edges-0.064
 atop-0.064
Token a
Token ID257
Feature activation-3.79e+00

Pos logit contributions
 pieces+0.056
hole+0.054
 on+0.053
 atop+0.051
 falling+0.051
Neg logit contributions
Jack-0.053
David-0.053
Un-0.052
Kids-0.051
 concession-0.051
Token smile
Token ID8212
Feature activation-1.27e+01

Pos logit contributions
hole+0.043
 pieces+0.042
ps+0.040
 edges+0.040
 atop+0.039
Neg logit contributions
Un-0.045
Jack-0.045
David-0.044
 Fro-0.044
Jim-0.042
Token on
Token ID319
Feature activation-7.73e+00

Pos logit contributions
 pieces+0.098
hole+0.095
 edges+0.093
ps+0.084
 wires+0.084
Neg logit contributions
Un-0.197
 Fro-0.193
His-0.190
David-0.189
Jack-0.186
Token,
Token ID11
Feature activation-8.42e-01

Pos logit contributions
hole+0.106
 edges+0.102
 pieces+0.101
 atop+0.101
ool+0.100
Neg logit contributions
 please-0.058
 Fro-0.051
 Max-0.051
 Ben-0.050
Un-0.049
Token do
Token ID466
Feature activation-5.66e+00

Pos logit contributions
 pieces+0.011
hole+0.011
 on+0.010
 edges+0.010
 atop+0.010
Neg logit contributions
David-0.009
Jack-0.009
Un-0.009
 Fro-0.008
 concession-0.008
Token not
Token ID407
Feature activation-5.09e+00

Pos logit contributions
hole+0.082
 pieces+0.077
 atop+0.075
 edges+0.072
ps+0.072
Neg logit contributions
Un-0.053
 Fro-0.053
Jack-0.050
David-0.049
Jim-0.048
Token touch
Token ID3638
Feature activation-1.36e+01

Pos logit contributions
hole+0.069
 pieces+0.067
 atop+0.065
ps+0.061
 edges+0.061
Neg logit contributions
 Fro-0.052
Un-0.051
Jack-0.049
David-0.046
His-0.046
Token the
Token ID262
Feature activation-7.55e+00

Pos logit contributions
hole+0.183
 atop+0.176
 code+0.172
 globe+0.166
 lid+0.163
Neg logit contributions
 Fro-0.141
 Mitt-0.129
Un-0.116
Johnny-0.116
 him-0.115
Token pins
Token ID20567
Feature activation-2.11e+01

Pos logit contributions
hole+0.085
 atop+0.084
ool+0.073
 existed+0.072
 underneath+0.072
Neg logit contributions
 Fro-0.088
Ben-0.087
Kids-0.082
Sam-0.079
L-0.079
Token.
Token ID13
Feature activation+1.50e+00

Pos logit contributions
hole+0.227
bones+0.224
 lid+0.221
 atop+0.219
 edges+0.219
Neg logit contributions
 please-0.228
They-0.214
 Fro-0.205
He-0.204
Jack-0.204
Token They
Token ID1119
Feature activation-5.12e+00

Pos logit contributions
Jack+0.013
David+0.013
 concession+0.013
Un+0.013
oph+0.013
Neg logit contributions
hole-0.021
 pieces-0.020
 on-0.020
 edges-0.019
ps-0.019
Token are
Token ID389
Feature activation-6.34e+00

Pos logit contributions
hole+0.074
 atop+0.074
bones+0.068
 pieces+0.067
ool+0.067
Neg logit contributions
 Fro-0.039
 please-0.037
They-0.034
 him-0.034
He-0.033
Token not
Token ID407
Feature activation-7.60e+00

Pos logit contributions
hole+0.093
 atop+0.084
 globe+0.083
 angle+0.082
 lid+0.081
Neg logit contributions
 Fro-0.058
 Mitt-0.053
Un-0.053
 wife-0.050
 uncle-0.048
Token toys
Token ID14958
Feature activation-1.70e+01

Pos logit contributions
hole+0.096
 lid+0.093
 atop+0.092
 globe+0.090
els+0.085
Neg logit contributions
 Fro-0.082
Un-0.076
 wife-0.073
 Mitt-0.071
Da-0.068
Token"
Token ID1
Feature activation+1.95e+00

Pos logit contributions
 Fro+0.029
enny+0.024
ascal+0.022
 shepherd+0.022
ory+0.022
Neg logit contributions
hole-0.046
 on-0.044
 pieces-0.044
 atop-0.042
 edges-0.042
TokenLet
Token ID5756
Feature activation-6.24e+00

Pos logit contributions
Un+0.016
Jack+0.016
David+0.016
 Fro+0.016
Jim+0.015
Neg logit contributions
hole-0.029
 pieces-0.028
 atop-0.027
 on-0.027
 edges-0.027
Token's
Token ID338
Feature activation-4.40e+00

Pos logit contributions
 on+0.050
 pieces+0.049
 way+0.044
 button+0.042
hole+0.041
Neg logit contributions
 Fro-0.095
enny-0.094
 concession-0.094
Un-0.092
 coach-0.092
Token replace
Token ID6330
Feature activation-1.47e+01

Pos logit contributions
 pieces+0.060
hole+0.060
 atop+0.056
 edges+0.056
ps+0.055
Neg logit contributions
Un-0.041
Jack-0.040
David-0.040
 Fro-0.040
Jim-0.038
Token the
Token ID262
Feature activation-7.76e+00

Pos logit contributions
hole+0.196
 atop+0.182
 falling+0.179
 lid+0.178
ool+0.177
Neg logit contributions
 Fro-0.146
Jack-0.132
Un-0.131
David-0.128
 concession-0.126
Token rocks
Token ID12586
Feature activation-2.11e+01

Pos logit contributions
hole+0.101
bones+0.100
ool+0.100
els+0.100
ps+0.097
Neg logit contributions
 fisherman-0.074
 wife-0.068
 coach-0.062
 dolphin-0.059
I-0.058
Token at
Token ID379
Feature activation-8.53e+00

Pos logit contributions
ps+0.218
 falling+0.213
 edges+0.207
 lid+0.207
hole+0.205
Neg logit contributions
Jack-0.222
They-0.218
He-0.212
 concession-0.210
 Fro-0.209
Token the
Token ID262
Feature activation-6.80e+00

Pos logit contributions
 lid+0.121
ps+0.117
hole+0.117
 splits+0.114
 tangled+0.111
Neg logit contributions
 Fro-0.071
 Pete-0.064
 Sammy-0.061
 Dad-0.061
Un-0.060
Token bottom
Token ID4220
Feature activation-1.58e+01

Pos logit contributions
hole+0.086
els+0.086
ps+0.084
ool+0.081
cks+0.076
Neg logit contributions
 fisherman-0.070
Jack-0.066
 coach-0.066
I-0.065
 concession-0.065
Token of
Token ID286
Feature activation-7.77e+00

Pos logit contributions
ps+0.172
bones+0.152
 falling+0.152
 atop+0.147
 tangled+0.147
Neg logit contributions
Jack-0.195
Jim-0.184
Un-0.181
David-0.178
 Fro-0.177
Token the
Token ID262
Feature activation-6.76e+00

Pos logit contributions
hole+0.116
 atop+0.104
ool+0.103
 splits+0.101
 lid+0.101
Neg logit contributions
 Fro-0.073
Un-0.066
David-0.062
Jack-0.062
 Pete-0.062
Token and
Token ID290
Feature activation-3.34e+00

Pos logit contributions
ps+0.235
els+0.231
 edges+0.227
 problems+0.227
bones+0.224
Neg logit contributions
 His-0.097
Jack-0.097
 Fro-0.097
He-0.096
His-0.095
Token found
Token ID1043
Feature activation-8.51e+00

Pos logit contributions
hole+0.040
 pieces+0.039
ps+0.037
 edges+0.037
 falling+0.037
Neg logit contributions
Jack-0.038
 Fro-0.036
Un-0.036
David-0.036
Jim-0.035
Token lots
Token ID6041
Feature activation-1.23e+01

Pos logit contributions
hole+0.122
 falling+0.119
 changing+0.112
 edges+0.112
ool+0.110
Neg logit contributions
 Fro-0.074
Jack-0.071
 His-0.068
David-0.066
 him-0.064
Token of
Token ID286
Feature activation-8.86e+00

Pos logit contributions
hole+0.184
ps+0.182
 hardest+0.182
 atop+0.177
 falling+0.174
Neg logit contributions
Jack-0.092
 concession-0.090
 Fro-0.086
enny-0.082
He-0.081
Token gold
Token ID3869
Feature activation-2.06e+01

Pos logit contributions
hole+0.090
ps+0.084
 atop+0.083
 falling+0.081
 hardest+0.080
Neg logit contributions
Jack-0.113
 concession-0.109
 Fro-0.107
David-0.106
He-0.104
Token coins
Token ID10796
Feature activation-2.09e+01

Pos logit contributions
 hardest+0.170
 angle+0.165
 falling+0.162
 changing+0.162
hole+0.162
Neg logit contributions
He-0.257
ory-0.249
 him-0.247
 Fro-0.247
 He-0.243
Token and
Token ID290
Feature activation-5.05e+00

Pos logit contributions
 splits+0.210
 edges+0.206
 hardest+0.206
 connects+0.204
ps+0.202
Neg logit contributions
 Fro-0.221
ory-0.219
David-0.216
ete-0.209
He-0.209
Token shiny
Token ID22441
Feature activation-1.54e+01

Pos logit contributions
hole+0.039
 atop+0.035
ps+0.033
 falling+0.033
ool+0.031
Neg logit contributions
Jack-0.076
 Fro-0.076
Un-0.075
David-0.074
 concession-0.073
Token jewels
Token ID42701
Feature activation-1.84e+01

Pos logit contributions
hole+0.134
 falling+0.132
 hardest+0.131
 socket+0.121
 changing+0.120
Neg logit contributions
 him-0.188
ory-0.187
 He-0.186
He-0.184
 His-0.181
Token.
Token ID13
Feature activation+4.32e+00

Pos logit contributions
 hardest+0.196
 edges+0.191
els+0.188
ps+0.185
 wires+0.185
Neg logit contributions
 Fro-0.182
He-0.179
enny-0.179
Jack-0.178
ory-0.175
Token
Token ID198
Feature activation+2.51e+00

Pos logit contributions
Un+0.034
David+0.031
 Fro+0.031
 concession+0.031
Jim+0.030
Neg logit contributions
 pieces-0.067
hole-0.066
 on-0.064
 edges-0.062
 atop-0.061
Token to
Token ID284
Feature activation-5.22e+00

Pos logit contributions
hole+0.060
 pieces+0.058
 atop+0.055
ps+0.054
 edges+0.054
Neg logit contributions
Jack-0.044
Un-0.043
 Fro-0.043
David-0.043
Kids-0.041
Token spray
Token ID11662
Feature activation-1.88e+01

Pos logit contributions
hole+0.062
 pieces+0.059
 atop+0.058
 hardest+0.053
 globe+0.052
Neg logit contributions
 Fro-0.060
Jack-0.057
Kids-0.057
Un-0.056
 Mitt-0.054
Token some
Token ID617
Feature activation-1.19e+01

Pos logit contributions
hole+0.291
 code+0.267
 slot+0.254
 plank+0.248
 atop+0.246
Neg logit contributions
 Fro-0.172
 Mitt-0.170
 Sammy-0.153
They-0.150
 Sandy-0.141
Token nice
Token ID3621
Feature activation-1.27e+01

Pos logit contributions
hole+0.167
 globe+0.152
 atop+0.147
bones+0.144
 code+0.144
Neg logit contributions
 Mitt-0.113
 Fro-0.112
They-0.109
Kids-0.106
Ben-0.101
Token smelling
Token ID48090
Feature activation-1.60e+01

Pos logit contributions
hole+0.150
 atop+0.139
 code+0.134
 slot+0.130
 hardest+0.128
Neg logit contributions
 Fro-0.149
Kids-0.142
Da-0.137
 Mitt-0.136
Jack-0.133
Token liquid
Token ID8122
Feature activation-2.09e+01

Pos logit contributions
hole+0.234
 atop+0.200
 code+0.200
 hardest+0.192
eking+0.187
Neg logit contributions
 Fro-0.160
 Mitt-0.157
Kids-0.145
They-0.140
Da-0.140
Token around
Token ID1088
Feature activation-1.18e+01

Pos logit contributions
hole+0.274
 slot+0.246
 plank+0.235
 code+0.235
 angle+0.234
Neg logit contributions
They-0.211
 Fro-0.210
 Pete-0.190
 Mitt-0.188
Un-0.184
Token the
Token ID262
Feature activation-6.64e+00

Pos logit contributions
hole+0.182
 code+0.161
 hardest+0.159
 atop+0.159
ps+0.157
Neg logit contributions
Jack-0.101
 Fro-0.097
Da-0.096
 Mitt-0.091
Un-0.090
Token garden
Token ID11376
Feature activation-1.58e+01

Pos logit contributions
hole+0.080
ps+0.073
 atop+0.073
bones+0.073
 code+0.071
Neg logit contributions
Ben-0.071
 Mitt-0.071
 Fro-0.070
Kids-0.070
They-0.067
Token.
Token ID13
Feature activation+3.91e+00

Pos logit contributions
hole+0.188
 atop+0.181
ps+0.176
 hardest+0.171
bones+0.171
Neg logit contributions
 Fro-0.158
They-0.148
Un-0.143
Da-0.142
 Mitt-0.138
Token Sarah
Token ID10490
Feature activation-4.16e+00

Pos logit contributions
 Fro+0.032
Un+0.032
 concession+0.031
David+0.030
Jack+0.030
Neg logit contributions
hole-0.057
 pieces-0.057
 on-0.055
 edges-0.053
 atop-0.053
Token continued
Token ID3767
Feature activation-8.16e+00

Pos logit contributions
hole+0.121
 pieces+0.117
 atop+0.111
 edges+0.111
 lid+0.110
Neg logit contributions
Jack-0.091
Un-0.090
 Fro-0.090
David-0.088
Kids-0.085
Token to
Token ID284
Feature activation-5.24e+00

Pos logit contributions
hole+0.128
 pieces+0.116
 lid+0.113
 edges+0.112
ps+0.112
Neg logit contributions
 Fro-0.068
Un-0.064
Jack-0.063
David-0.060
Johnny-0.059
Token make
Token ID787
Feature activation-1.16e+01

Pos logit contributions
hole+0.072
 pieces+0.072
 atop+0.067
 edges+0.067
 on+0.067
Neg logit contributions
Jack-0.048
Un-0.048
David-0.048
 Fro-0.048
Kids-0.045
Token fancy
Token ID14996
Feature activation-1.46e+01

Pos logit contributions
 pieces+0.109
hole+0.097
 on+0.096
 edges+0.093
 falling+0.090
Neg logit contributions
Un-0.168
David-0.164
Jack-0.162
Kids-0.160
 concession-0.160
Token mud
Token ID17492
Feature activation-1.92e+01

Pos logit contributions
hole+0.127
 pieces+0.118
 atop+0.113
ps+0.107
 on+0.106
Neg logit contributions
 Fro-0.209
Jack-0.209
Un-0.206
David-0.206
 concession-0.201
Token designs
Token ID9824
Feature activation-2.09e+01

Pos logit contributions
hole+0.179
 pieces+0.170
 atop+0.160
 edges+0.155
ps+0.150
Neg logit contributions
 Fro-0.263
Jack-0.260
Un-0.260
David-0.255
 concession-0.251
Token.
Token ID13
Feature activation+4.59e+00

Pos logit contributions
hole+0.197
 pieces+0.174
 edges+0.173
 atop+0.173
ps+0.171
Neg logit contributions
Jack-0.276
Un-0.266
 Fro-0.265
David-0.265
Johnny-0.259
Token<|endoftext|>
Token ID50256
Feature activation-2.38e+00

Pos logit contributions
David+0.049
Jim+0.046
Un+0.046
 concession+0.045
Jack+0.045
Neg logit contributions
 pieces-0.061
 on-0.060
hole-0.058
 edges-0.055
 atop-0.055
TokenOne
Token ID3198
Feature activation-2.59e+00

Pos logit contributions
hole+0.040
 pieces+0.040
 atop+0.038
 edges+0.038
ps+0.038
Neg logit contributions
Jack-0.015
Un-0.014
David-0.014
 Fro-0.014
Jim-0.013
Token day
Token ID1110
Feature activation-5.35e+00

Pos logit contributions
 pieces+0.016
hole+0.015
 on+0.015
 atop+0.014
 falling+0.013
Neg logit contributions
David-0.044
Un-0.043
Jack-0.043
 Fro-0.043
Kids-0.042
Token,
Token ID11
Feature activation+1.07e+00

Pos logit contributions
hole+0.051
 pieces+0.050
 atop+0.045
 edges+0.045
ps+0.045
Neg logit contributions
Jack-0.074
Un-0.073
David-0.072
 Fro-0.072
Jim-0.070
Token that
Token ID326
Feature activation-4.42e+00

Pos logit contributions
 hardest+0.184
 edges+0.178
 pieces+0.177
ps+0.176
hole+0.170
Neg logit contributions
Un-0.196
David-0.188
Jack-0.186
 concession-0.185
Johnny-0.183
Token allowed
Token ID3142
Feature activation-1.07e+01

Pos logit contributions
hole+0.068
ps+0.064
 atop+0.063
els+0.061
 edges+0.061
Neg logit contributions
Jack-0.037
Un-0.037
Johnny-0.035
Sarah-0.034
Kids-0.033
Token people
Token ID661
Feature activation-1.34e+01

Pos logit contributions
hole+0.174
 pieces+0.169
 atop+0.165
 edges+0.164
 hardest+0.159
Neg logit contributions
 Fro-0.073
Jack-0.072
Un-0.071
David-0.070
Kids-0.069
Token to
Token ID284
Feature activation-6.44e+00

Pos logit contributions
 lid+0.206
 pieces+0.205
 edges+0.203
hole+0.198
ps+0.193
Neg logit contributions
Un-0.101
David-0.097
Jack-0.093
 Fro-0.092
 concession-0.091
Token buy
Token ID2822
Feature activation-1.50e+01

Pos logit contributions
hole+0.094
 pieces+0.092
 edges+0.091
 hardest+0.089
 atop+0.089
Neg logit contributions
Un-0.060
David-0.056
 Fro-0.053
Jack-0.052
His-0.051
Token stamps
Token ID25560
Feature activation-2.08e+01

Pos logit contributions
hole+0.196
 atop+0.181
 falling+0.179
ps+0.177
 lid+0.175
Neg logit contributions
 him-0.158
Un-0.148
Jack-0.144
 uncle-0.143
David-0.142
Token.
Token ID13
Feature activation+4.35e+00

Pos logit contributions
hole+0.213
 edges+0.210
 lid+0.206
ps+0.205
 pieces+0.202
Neg logit contributions
David-0.238
 Fro-0.236
Jack-0.235
Un-0.234
They-0.228
Token Her
Token ID2332
Feature activation-2.26e+00

Pos logit contributions
David+0.032
Jim+0.031
Un+0.031
Jack+0.030
Tony+0.030
Neg logit contributions
 pieces-0.072
 edges-0.070
hole-0.069
 doors-0.069
 lid-0.068
Token mom
Token ID1995
Feature activation-4.67e+00

Pos logit contributions
 pieces+0.033
hole+0.032
 on+0.031
 edges+0.031
 atop+0.030
Neg logit contributions
Jack-0.020
Un-0.020
David-0.020
 Fro-0.019
Jim-0.019
Tokenmy
Token ID1820
Feature activation-5.11e+00

Pos logit contributions
hole+0.061
 pieces+0.061
 atop+0.057
 on+0.057
 edges+0.057
Neg logit contributions
Un-0.046
Jack-0.046
David-0.046
 Fro-0.045
Jim-0.043
Token gave
Token ID2921
Feature activation-8.71e+00

Pos logit contributions
hole+0.069
 pieces+0.067
 edges+0.066
 atop+0.065
ps+0.064
Neg logit contributions
Jack-0.050
Un-0.049
David-0.049
Johnny-0.047
 Fro-0.047
Token saw
Token ID2497
Feature activation-7.90e+00

Pos logit contributions
hole+0.043
 pieces+0.043
 atop+0.040
 on+0.039
 edges+0.039
Neg logit contributions
Un-0.041
Jack-0.041
David-0.040
 Fro-0.040
 concession-0.039
Token a
Token ID257
Feature activation-7.66e+00

Pos logit contributions
 pieces+0.081
 on+0.074
 edges+0.067
 wires+0.065
ps+0.065
Neg logit contributions
Un-0.115
David-0.114
oph-0.110
Jack-0.109
Jim-0.108
Token big
Token ID1263
Feature activation-9.15e+00

Pos logit contributions
hole+0.075
ps+0.070
 pieces+0.067
 atop+0.066
els+0.064
Neg logit contributions
Jack-0.106
Un-0.104
 Fro-0.103
 concession-0.102
David-0.102
Token bottle
Token ID9294
Feature activation-1.88e+01

Pos logit contributions
hole+0.091
ps+0.084
 atop+0.081
 hardest+0.078
 pieces+0.076
Neg logit contributions
 Fro-0.123
 concession-0.121
Jack-0.120
Un-0.114
Johnny-0.112
Token of
Token ID286
Feature activation-9.87e+00

Pos logit contributions
hole+0.239
ps+0.236
 hardest+0.235
 edges+0.230
 problems+0.230
Neg logit contributions
Jack-0.184
Un-0.174
 Fro-0.173
His-0.169
They-0.166
Token shampoo
Token ID47259
Feature activation-2.08e+01

Pos logit contributions
hole+0.126
 atop+0.114
 hardest+0.106
 existed+0.105
ps+0.104
Neg logit contributions
 Fro-0.114
Jack-0.105
 Mitt-0.104
 concession-0.103
Kids-0.102
Token on
Token ID319
Feature activation-1.06e+01

Pos logit contributions
hole+0.170
 atop+0.153
 hardest+0.149
 pieces+0.144
 angles+0.139
Neg logit contributions
 Fro-0.286
addy-0.277
Kids-0.275
Jack-0.269
 concession-0.267
Token the
Token ID262
Feature activation-5.65e+00

Pos logit contributions
hole+0.135
 atop+0.126
 pieces+0.121
 hardest+0.120
ps+0.120
Neg logit contributions
Jack-0.110
 Fro-0.110
Un-0.109
His-0.105
David-0.102
Token shelf
Token ID18316
Feature activation-1.60e+01

Pos logit contributions
hole+0.047
 atop+0.046
els+0.041
cks+0.041
 pieces+0.040
Neg logit contributions
 concession-0.080
 Fro-0.079
Un-0.078
addy-0.078
Kids-0.077
Token.
Token ID13
Feature activation+3.66e+00

Pos logit contributions
ps+0.182
 hardest+0.177
 atop+0.174
 matters+0.172
els+0.170
Neg logit contributions
Jack-0.158
Un-0.155
They-0.155
His-0.154
He-0.153
Token It
Token ID632
Feature activation-7.32e+00

Pos logit contributions
Un+0.040
David+0.039
 concession+0.039
Johnny+0.036
Billy+0.035
Neg logit contributions
 pieces-0.049
 on-0.048
 lid-0.046
 edges-0.045
hole-0.041
Token good
Token ID922
Feature activation-8.70e+00

Pos logit contributions
hole+0.098
 atop+0.097
els+0.092
 lid+0.089
bones+0.088
Neg logit contributions
Un-0.075
Jack-0.071
Johnny-0.070
 please-0.069
They-0.068
Token to
Token ID284
Feature activation-5.18e+00

Pos logit contributions
hole+0.111
 atop+0.109
 slot+0.098
 lid+0.098
 globe+0.097
Neg logit contributions
 Fro-0.084
 coach-0.082
Jack-0.081
Un-0.081
 please-0.080
Token have
Token ID423
Feature activation-8.44e+00

Pos logit contributions
hole+0.067
 pieces+0.067
 atop+0.065
 on+0.061
ps+0.061
Neg logit contributions
Jack-0.053
Un-0.053
David-0.051
 Fro-0.050
Kids-0.050
Token so
Token ID523
Feature activation-4.34e+00

Pos logit contributions
hole+0.121
 atop+0.119
 globe+0.111
 lid+0.106
 angle+0.101
Neg logit contributions
Jack-0.076
Un-0.072
 His-0.070
 Mitt-0.070
They-0.069
Token much
Token ID881
Feature activation-8.99e+00

Pos logit contributions
hole+0.059
 atop+0.053
 globe+0.050
 lid+0.049
ps+0.048
Neg logit contributions
Un-0.049
Jack-0.046
They-0.045
Ben-0.043
enny-0.042
Token clutter
Token ID45343
Feature activation-2.08e+01

Pos logit contributions
hole+0.103
 atop+0.103
 globe+0.092
 hardest+0.090
cks+0.086
Neg logit contributions
Jack-0.096
 concession-0.096
 His-0.096
Un-0.095
Kids-0.095
Token.
Token ID13
Feature activation+3.68e+00

Pos logit contributions
hole+0.225
 lid+0.218
 atop+0.213
 trunk+0.210
ool+0.210
Neg logit contributions
Un-0.222
Jack-0.211
 concession-0.211
They-0.208
Johnny-0.202
Token You
Token ID921
Feature activation-5.11e+00

Pos logit contributions
David+0.035
 Fro+0.035
Jack+0.035
Un+0.034
 concession+0.033
Neg logit contributions
hole-0.049
 pieces-0.048
 on-0.046
 edges-0.046
ps-0.045
Token can
Token ID460
Feature activation-4.97e+00

Pos logit contributions
 atop+0.065
hole+0.064
bones+0.060
 pieces+0.060
 edges+0.059
Neg logit contributions
Jack-0.049
Un-0.046
enny-0.046
 Fro-0.046
 please-0.045
Token trip
Token ID5296
Feature activation-1.58e+01

Pos logit contributions
 pieces+0.050
hole+0.050
 on+0.050
 falling+0.048
 edges+0.047
Neg logit contributions
David-0.065
Un-0.064
 Fro-0.064
Da-0.063
Jim-0.063
Token and
Token ID290
Feature activation-3.75e+00

Pos logit contributions
hole+0.184
 pieces+0.177
 atop+0.173
 lid+0.167
els+0.166
Neg logit contributions
Jack-0.181
Un-0.177
David-0.172
 concession-0.169
Johnny-0.167
Token the
Token ID262
Feature activation-4.63e+00

Pos logit contributions
hole+0.067
 pieces+0.066
 atop+0.062
 on+0.061
 edges+0.061
Neg logit contributions
Un-0.052
Jack-0.052
David-0.051
 Fro-0.051
Jim-0.048
Token birds
Token ID10087
Feature activation-1.31e+01

Pos logit contributions
hole+0.066
ool+0.066
cks+0.063
bones+0.061
 atop+0.060
Neg logit contributions
 wife-0.039
Ben-0.036
David-0.036
Jack-0.035
Un-0.035
Token could
Token ID714
Feature activation-4.71e+00

Pos logit contributions
 pieces+0.170
hole+0.168
 edges+0.167
 lid+0.164
els+0.162
Neg logit contributions
 please-0.102
He-0.102
Jack-0.101
They-0.100
 Fro-0.100
Token make
Token ID787
Feature activation-1.10e+01

Pos logit contributions
hole+0.056
 pieces+0.056
 atop+0.052
 on+0.052
 edges+0.051
Neg logit contributions
Un-0.052
Jack-0.052
David-0.052
 Fro-0.051
Jim-0.049
Token their
Token ID511
Feature activation-6.86e+00

Pos logit contributions
hole+0.148
 atop+0.134
 pieces+0.132
 lid+0.130
 edges+0.129
Neg logit contributions
 Fro-0.109
Jack-0.106
Un-0.105
David-0.104
 concession-0.100
Token nests
Token ID44382
Feature activation-2.07e+01

Pos logit contributions
hole+0.079
 atop+0.075
ps+0.071
ool+0.065
cks+0.065
Neg logit contributions
 Fro-0.077
 concession-0.069
 wife-0.068
 Mitt-0.068
 uncle-0.068
Token and
Token ID290
Feature activation-4.30e+00

Pos logit contributions
hole+0.221
 pieces+0.212
 edges+0.206
els+0.204
 atop+0.203
Neg logit contributions
 Fro-0.228
Jack-0.223
Un-0.223
David-0.219
 concession-0.217
Token feel
Token ID1254
Feature activation-9.45e+00

Pos logit contributions
hole+0.049
 pieces+0.045
 atop+0.044
ps+0.042
 edges+0.042
Neg logit contributions
Un-0.050
 Fro-0.049
Jack-0.047
 concession-0.047
David-0.047
Token safe
Token ID3338
Feature activation-1.58e+01

Pos logit contributions
hole+0.113
 pieces+0.106
els+0.099
 edges+0.098
 atop+0.098
Neg logit contributions
 Fro-0.107
Un-0.105
Jack-0.104
David-0.101
 concession-0.101
Token.
Token ID13
Feature activation+4.12e+00

Pos logit contributions
hole+0.169
els+0.164
 lid+0.162
ool+0.162
 pieces+0.159
Neg logit contributions
Un-0.168
 concession-0.161
 Fro-0.160
They-0.159
He-0.156
Token The
Token ID383
Feature activation-1.59e+00

Pos logit contributions
Un+0.038
David+0.037
 concession+0.036
Kids+0.036
Jack+0.035
Neg logit contributions
hole-0.058
 pieces-0.057
 on-0.054
 edges-0.053
 trunk-0.053
TokenLet
Token ID5756
Feature activation-5.94e+00

Pos logit contributions
 Fro+0.012
 concession+0.011
enny+0.010
Un+0.010
David+0.010
Neg logit contributions
 pieces-0.022
hole-0.021
 on-0.020
 edges-0.020
 tangled-0.020
Token's
Token ID338
Feature activation-4.16e+00

Pos logit contributions
 way+0.037
 on+0.036
 pieces+0.035
 broken+0.030
 button+0.028
Neg logit contributions
 Fro-0.105
enny-0.104
 concession-0.102
David-0.101
Tony-0.101
Token gather
Token ID6431
Feature activation-1.44e+01

Pos logit contributions
hole+0.051
 pieces+0.050
 on+0.049
 atop+0.047
 falling+0.047
Neg logit contributions
Un-0.045
Jack-0.045
David-0.044
 Fro-0.043
Da-0.043
Token up
Token ID510
Feature activation-1.10e+01

Pos logit contributions
hole+0.189
 falling+0.177
 code+0.171
 atop+0.169
 changing+0.164
Neg logit contributions
 Fro-0.145
Jack-0.140
 Pete-0.129
 concession-0.125
 Mitt-0.121
Token some
Token ID617
Feature activation-1.20e+01

Pos logit contributions
hole+0.146
 atop+0.131
 falling+0.129
 lid+0.128
 pieces+0.127
Neg logit contributions
Jack-0.114
 Fro-0.110
Jim-0.106
 concession-0.106
Un-0.106
Token rocks
Token ID12586
Feature activation-2.07e+01

Pos logit contributions
hole+0.137
ps+0.134
bones+0.125
 falling+0.124
 globe+0.124
Neg logit contributions
 concession-0.121
 please-0.119
 Fro-0.119
 wife-0.114
Jack-0.114
Token and
Token ID290
Feature activation-5.61e+00

Pos logit contributions
 lid+0.219
ps+0.217
 edges+0.217
 falling+0.215
hole+0.213
Neg logit contributions
 Fro-0.219
Jack-0.218
 concession-0.210
enny-0.209
Un-0.207
Token see
Token ID766
Feature activation-7.32e+00

Pos logit contributions
hole+0.059
 pieces+0.059
 atop+0.054
 on+0.053
 edges+0.053
Neg logit contributions
Un-0.069
Jack-0.069
David-0.068
 Fro-0.068
 concession-0.066
Token who
Token ID508
Feature activation-5.91e+00

Pos logit contributions
 lid+0.109
bones+0.108
 splits+0.107
 falling+0.106
hole+0.106
Neg logit contributions
 Fro-0.048
Un-0.047
Jack-0.047
They-0.047
Johnny-0.043
Token can
Token ID460
Feature activation-6.44e+00

Pos logit contributions
ps+0.073
 lid+0.071
hole+0.071
 falling+0.069
 edges+0.068
Neg logit contributions
Jack-0.056
 concession-0.052
Un-0.050
 Fro-0.050
They-0.050
Token make
Token ID787
Feature activation-1.17e+01

Pos logit contributions
hole+0.077
 pieces+0.076
 atop+0.071
 edges+0.070
ps+0.069
Neg logit contributions
Jack-0.070
 Fro-0.069
Un-0.068
David-0.067
Jim-0.065
Token you
Token ID345
Feature activation-7.48e+00

Pos logit contributions
hole+0.086
 pieces+0.084
 atop+0.078
 edges+0.077
 falling+0.077
Neg logit contributions
Jack-0.073
Un-0.072
David-0.072
 Fro-0.071
 concession-0.069
Token will
Token ID481
Feature activation-7.50e+00

Pos logit contributions
hole+0.105
 pieces+0.103
 atop+0.100
 falling+0.099
ps+0.099
Neg logit contributions
Jack-0.065
Un-0.064
David-0.062
 Fro-0.061
 concession-0.061
Token never
Token ID1239
Feature activation-7.43e+00

Pos logit contributions
hole+0.105
 pieces+0.105
 atop+0.102
 edges+0.096
els+0.095
Neg logit contributions
Jack-0.068
David-0.067
Un-0.067
 Fro-0.066
His-0.065
Token touch
Token ID3638
Feature activation-1.39e+01

Pos logit contributions
hole+0.099
 pieces+0.098
 atop+0.096
 edges+0.092
ps+0.092
Neg logit contributions
Un-0.071
Jack-0.071
 Fro-0.070
David-0.068
His-0.065
Token my
Token ID616
Feature activation-7.23e+00

Pos logit contributions
hole+0.214
ool+0.201
 code+0.198
 lid+0.197
 atop+0.195
Neg logit contributions
 him-0.113
Jack-0.088
 Max-0.088
 Mitt-0.086
 Dad-0.086
Token axe
Token ID23272
Feature activation-2.07e+01

Pos logit contributions
bones+0.075
ps+0.073
 atop+0.073
hole+0.069
ool+0.068
Neg logit contributions
Un-0.084
 uncle-0.084
 Fro-0.079
 Mitt-0.075
 him-0.075
Token again
Token ID757
Feature activation-1.17e+01

Pos logit contributions
 atop+0.248
hole+0.247
ps+0.246
els+0.241
 lid+0.240
Neg logit contributions
Un-0.206
 Fro-0.200
Jack-0.192
 concession-0.189
His-0.184
Token,"
Token ID553
Feature activation+4.22e+00

Pos logit contributions
hole+0.109
ps+0.106
 atop+0.105
 lid+0.104
 pieces+0.102
Neg logit contributions
Un-0.162
Jack-0.154
 Fro-0.150
David-0.147
Jim-0.145
Token his
Token ID465
Feature activation-2.19e+00

Pos logit contributions
 concession+0.050
latable+0.050
enny+0.047
oph+0.047
Da+0.046
Neg logit contributions
 on-0.052
 pieces-0.048
 hard-0.047
 broken-0.046
 tangled-0.045
Token dad
Token ID9955
Feature activation-5.68e+00

Pos logit contributions
 pieces+0.040
hole+0.040
 on+0.038
 edges+0.038
 atop+0.038
Neg logit contributions
Jack-0.011
Un-0.010
 Fro-0.010
David-0.010
 concession-0.009
Token said
Token ID531
Feature activation-5.25e+00

Pos logit contributions
hole+0.075
 pieces+0.074
 edges+0.072
 atop+0.072
 falling+0.071
Neg logit contributions
Jack-0.055
Un-0.055
David-0.054
 Fro-0.053
His-0.052
Token tree
Token ID5509
Feature activation-1.61e+01

Pos logit contributions
ps+0.147
hole+0.137
 hardest+0.123
 angles+0.119
 code+0.119
Neg logit contributions
Un-0.257
He-0.252
Jack-0.251
She-0.251
 concession-0.247
Token with
Token ID351
Feature activation-8.46e+00

Pos logit contributions
ps+0.172
 pieces+0.162
 hardest+0.160
 edges+0.158
hole+0.157
Neg logit contributions
Un-0.189
 Fro-0.181
Jack-0.181
They-0.176
He-0.176
Token many
Token ID867
Feature activation-1.19e+01

Pos logit contributions
hole+0.085
 pieces+0.076
 atop+0.076
ps+0.073
 falling+0.072
Neg logit contributions
Jack-0.111
 Fro-0.111
Un-0.110
David-0.109
Jim-0.106
Token ripe
Token ID29036
Feature activation-1.72e+01

Pos logit contributions
hole+0.125
ps+0.122
 slot+0.116
 globe+0.115
 code+0.114
Neg logit contributions
He-0.146
Jack-0.143
She-0.137
His-0.136
Da-0.136
Token p
Token ID279
Feature activation-1.20e+01

Pos logit contributions
hole+0.185
 code+0.175
 remote+0.173
ps+0.171
 slot+0.163
Neg logit contributions
 Fro-0.200
oph-0.193
He-0.190
Jack-0.189
 He-0.184
Tokenears
Token ID4127
Feature activation-2.06e+01

Pos logit contributions
hole+0.146
 pieces+0.145
 atop+0.137
ps+0.137
 on+0.136
Neg logit contributions
 Fro-0.126
Un-0.124
Jack-0.122
David-0.122
oph-0.121
Token.
Token ID13
Feature activation+4.59e+00

Pos logit contributions
hole+0.200
 code+0.197
ps+0.196
 hardest+0.194
 remote+0.189
Neg logit contributions
Un-0.230
They-0.228
 Fro-0.226
Jack-0.221
He-0.220
Token They
Token ID1119
Feature activation-2.87e+00

Pos logit contributions
Johnny+0.058
David+0.056
Un+0.052
Billy+0.051
Harry+0.051
Neg logit contributions
 pieces-0.054
 on-0.053
 edges-0.050
 lid-0.050
 ladder-0.046
Token wanted
Token ID2227
Feature activation-8.75e+00

Pos logit contributions
 pieces+0.034
hole+0.032
 on+0.031
 edges+0.030
 trunk+0.030
Neg logit contributions
Un-0.032
 concession-0.032
 Fro-0.032
David-0.032
Kids-0.031
Token to
Token ID284
Feature activation-5.10e+00

Pos logit contributions
hole+0.155
 atop+0.142
ps+0.140
 lid+0.135
ool+0.135
Neg logit contributions
 Fro-0.060
Jack-0.056
Da-0.050
Jim-0.050
Un-0.049
Token pick
Token ID2298
Feature activation-1.45e+01

Pos logit contributions
hole+0.058
 pieces+0.057
 atop+0.053
 edges+0.052
ps+0.052
Neg logit contributions
Jack-0.060
Un-0.060
 Fro-0.059
David-0.059
Jim-0.057
Token chest
Token ID7721
Feature activation-1.55e+01

Pos logit contributions
 falling+0.053
ps+0.052
 atop+0.051
els+0.051
hole+0.051
Neg logit contributions
 concession-0.070
His-0.069
He-0.069
 Fro-0.069
Jack-0.069
Token and
Token ID290
Feature activation-3.34e+00

Pos logit contributions
ps+0.235
els+0.231
 edges+0.227
 problems+0.227
bones+0.224
Neg logit contributions
 His-0.097
Jack-0.097
 Fro-0.097
He-0.096
His-0.095
Token found
Token ID1043
Feature activation-8.51e+00

Pos logit contributions
hole+0.040
 pieces+0.039
ps+0.037
 edges+0.037
 falling+0.037
Neg logit contributions
Jack-0.038
 Fro-0.036
Un-0.036
David-0.036
Jim-0.035
Token lots
Token ID6041
Feature activation-1.23e+01

Pos logit contributions
hole+0.122
 falling+0.119
 changing+0.112
 edges+0.112
ool+0.110
Neg logit contributions
 Fro-0.074
Jack-0.071
 His-0.068
David-0.066
 him-0.064
Token of
Token ID286
Feature activation-8.86e+00

Pos logit contributions
hole+0.184
ps+0.182
 hardest+0.182
 atop+0.177
 falling+0.174
Neg logit contributions
Jack-0.092
 concession-0.090
 Fro-0.086
enny-0.082
He-0.081
Token gold
Token ID3869
Feature activation-2.06e+01

Pos logit contributions
hole+0.090
ps+0.084
 atop+0.083
 falling+0.081
 hardest+0.080
Neg logit contributions
Jack-0.113
 concession-0.109
 Fro-0.107
David-0.106
He-0.104
Token coins
Token ID10796
Feature activation-2.09e+01

Pos logit contributions
 hardest+0.170
 angle+0.165
 falling+0.162
 changing+0.162
hole+0.162
Neg logit contributions
He-0.257
ory-0.249
 him-0.247
 Fro-0.247
 He-0.243
Token and
Token ID290
Feature activation-5.05e+00

Pos logit contributions
 splits+0.210
 edges+0.206
 hardest+0.206
 connects+0.204
ps+0.202
Neg logit contributions
 Fro-0.221
ory-0.219
David-0.216
ete-0.209
He-0.209
Token shiny
Token ID22441
Feature activation-1.54e+01

Pos logit contributions
hole+0.039
 atop+0.035
ps+0.033
 falling+0.033
ool+0.031
Neg logit contributions
Jack-0.076
 Fro-0.076
Un-0.075
David-0.074
 concession-0.073
Token jewels
Token ID42701
Feature activation-1.84e+01

Pos logit contributions
hole+0.134
 falling+0.132
 hardest+0.131
 socket+0.121
 changing+0.120
Neg logit contributions
 him-0.188
ory-0.187
 He-0.186
He-0.184
 His-0.181
Token.
Token ID13
Feature activation+4.32e+00

Pos logit contributions
 hardest+0.196
 edges+0.191
els+0.188
ps+0.185
 wires+0.185
Neg logit contributions
 Fro-0.182
He-0.179
enny-0.179
Jack-0.178
ory-0.175
Token the
Token ID262
Feature activation-6.04e+00

Pos logit contributions
hole+0.187
 lid+0.170
ool+0.169
 code+0.162
 pieces+0.162
Neg logit contributions
 Fro-0.083
Un-0.069
 him-0.068
 Mitt-0.066
 Dod-0.064
Token area
Token ID1989
Feature activation-1.65e+01

Pos logit contributions
hole+0.085
els+0.084
bones+0.079
cks+0.077
ps+0.077
Neg logit contributions
 Fro-0.062
He-0.054
 He-0.052
 shepherd-0.052
Da-0.052
Token clean
Token ID3424
Feature activation-1.45e+01

Pos logit contributions
hole+0.192
ps+0.180
els+0.171
 atop+0.170
 edges+0.168
Neg logit contributions
Jack-0.170
Da-0.166
 Fro-0.166
His-0.161
They-0.159
Token by
Token ID416
Feature activation-6.41e+00

Pos logit contributions
hole+0.168
 atop+0.159
ps+0.156
els+0.152
 lid+0.151
Neg logit contributions
 Fro-0.151
 concession-0.139
Un-0.139
Jack-0.138
 He-0.135
Token making
Token ID1642
Feature activation-1.20e+01

Pos logit contributions
hole+0.090
 slot+0.081
 atop+0.080
 code+0.080
 lid+0.080
Neg logit contributions
 Fro-0.059
They-0.054
Un-0.053
 Mitt-0.052
 him-0.049
Token stacks
Token ID24285
Feature activation-2.06e+01

Pos logit contributions
hole+0.133
 atop+0.119
 pieces+0.111
 falling+0.109
 globe+0.108
Neg logit contributions
 Fro-0.152
Jack-0.148
Un-0.144
David-0.143
 concession-0.139
Token of
Token ID286
Feature activation-8.73e+00

Pos logit contributions
hole+0.249
 atop+0.235
 edges+0.225
ps+0.222
 lid+0.217
Neg logit contributions
They-0.218
Un-0.210
Jack-0.209
 Fro-0.202
He-0.201
Token sticks
Token ID16461
Feature activation-2.01e+01

Pos logit contributions
hole+0.087
 atop+0.083
 hardest+0.075
ps+0.074
 existed+0.069
Neg logit contributions
 Fro-0.120
Jack-0.111
 concession-0.108
Un-0.108
Johnny-0.106
Token.
Token ID13
Feature activation+4.64e+00

Pos logit contributions
hole+0.251
 atop+0.236
 edges+0.234
 slot+0.233
els+0.231
Neg logit contributions
They-0.191
 Fro-0.187
He-0.177
 He-0.166
Un-0.166
Token
Token ID198
Feature activation+2.30e+00

Pos logit contributions
Un+0.038
David+0.038
Jack+0.037
 Fro+0.036
 concession+0.036
Neg logit contributions
hole-0.069
 pieces-0.069
 on-0.065
 atop-0.065
 edges-0.064
Token
Token ID198
Feature activation+3.34e+00

Pos logit contributions
Un+0.018
 concession+0.018
 Fro+0.018
David+0.018
Jack+0.017
Neg logit contributions
hole-0.034
 pieces-0.034
 on-0.033
 atop-0.032
 edges-0.031
Token a
Token ID257
Feature activation-5.30e+00

Pos logit contributions
hole+0.081
 plank+0.078
ps+0.078
ool+0.078
 splits+0.077
Neg logit contributions
Un-0.047
Jack-0.047
David-0.046
 uncle-0.044
 Her-0.044
Token grand
Token ID4490
Feature activation-5.66e+00

Pos logit contributions
ps+0.068
els+0.066
hole+0.062
cks+0.062
 atop+0.059
Neg logit contributions
Un-0.060
Jack-0.060
 concession-0.059
David-0.059
Johnny-0.056
Token hall
Token ID6899
Feature activation-1.67e+01

Pos logit contributions
 atop+0.067
hole+0.066
 pieces+0.065
 tangled+0.065
 on+0.064
Neg logit contributions
Jack-0.061
David-0.061
oph-0.060
 concession-0.059
Jim-0.059
Token filled
Token ID5901
Feature activation-1.29e+01

Pos logit contributions
ps+0.183
 pieces+0.181
els+0.178
 edges+0.175
 lid+0.172
Neg logit contributions
 concession-0.187
Un-0.183
David-0.181
Jack-0.181
He-0.170
Token with
Token ID351
Feature activation-9.88e+00

Pos logit contributions
hole+0.144
ps+0.134
 pieces+0.131
 hardest+0.130
 edges+0.129
Neg logit contributions
Jack-0.157
Un-0.150
David-0.148
 concession-0.146
Da-0.144
Token pots
Token ID32195
Feature activation-2.06e+01

Pos logit contributions
hole+0.111
 atop+0.102
ool+0.101
ps+0.097
cks+0.096
Neg logit contributions
 concession-0.115
David-0.108
Jack-0.108
Un-0.106
 Fro-0.106
Token of
Token ID286
Feature activation-9.24e+00

Pos logit contributions
 hardest+0.266
hole+0.262
 edges+0.260
 splits+0.260
 atop+0.258
Neg logit contributions
David-0.163
Jack-0.162
Un-0.160
 concession-0.157
He-0.156
Token the
Token ID262
Feature activation-6.50e+00

Pos logit contributions
hole+0.104
 hardest+0.096
 atop+0.093
ps+0.089
 existed+0.084
Neg logit contributions
 concession-0.112
 Fro-0.102
 uncle-0.102
Un-0.101
 Mitt-0.099
Token magic
Token ID5536
Feature activation-1.51e+01

Pos logit contributions
cks+0.075
els+0.070
ool+0.068
bones+0.067
 atop+0.066
Neg logit contributions
 concession-0.076
Jack-0.069
addy-0.063
 triple-0.060
 wife-0.060
Token liquid
Token ID8122
Feature activation-1.85e+01

Pos logit contributions
cks+0.135
 atop+0.121
ps+0.121
 hardest+0.119
 tangled+0.117
Neg logit contributions
 concession-0.185
Jack-0.176
He-0.172
 shepherd-0.170
addy-0.169
Token,
Token ID11
Feature activation-5.33e+00

Pos logit contributions
hole+0.190
 plank+0.178
 hardest+0.176
 pieces+0.175
 edges+0.174
Neg logit contributions
They-0.191
Jack-0.181
Un-0.179
 concession-0.179
 Fro-0.175
Token load
Token ID3440
Feature activation-1.70e+01

Pos logit contributions
hole+0.096
 atop+0.089
 lid+0.087
 pieces+0.087
els+0.086
Neg logit contributions
 Fro-0.066
Un-0.062
Kids-0.057
Jack-0.057
 His-0.055
Token your
Token ID534
Feature activation-8.09e+00

Pos logit contributions
hole+0.192
 atop+0.157
 code+0.150
 slot+0.150
 falling+0.143
Neg logit contributions
 Fro-0.216
Jack-0.207
 concession-0.201
Un-0.200
Da-0.194
Token wagon
Token ID31860
Feature activation-1.77e+01

Pos logit contributions
hole+0.106
 atop+0.093
 slot+0.089
 code+0.086
 connects+0.083
Neg logit contributions
 Fro-0.092
 His-0.080
omo-0.077
 concession-0.076
His-0.075
Token with
Token ID351
Feature activation-8.28e+00

Pos logit contributions
hole+0.184
 pieces+0.172
 atop+0.171
ps+0.168
 edges+0.167
Neg logit contributions
Jack-0.207
 Fro-0.206
 concession-0.205
Un-0.205
His-0.197
Token p
Token ID279
Feature activation-1.13e+01

Pos logit contributions
hole+0.108
 atop+0.096
 lid+0.087
 hardest+0.087
 pieces+0.087
Neg logit contributions
 Fro-0.090
Un-0.085
Jack-0.083
 Mitt-0.081
 concession-0.081
Tokenears
Token ID4127
Feature activation-2.06e+01

Pos logit contributions
hole+0.130
 pieces+0.130
 atop+0.122
 on+0.119
 edges+0.119
Neg logit contributions
 Fro-0.128
Un-0.126
Jack-0.125
David-0.125
 concession-0.121
Token near
Token ID1474
Feature activation-1.00e+01

Pos logit contributions
hole+0.222
 atop+0.208
ps+0.208
 pieces+0.207
 edges+0.205
Neg logit contributions
Un-0.236
 Fro-0.236
Jack-0.233
 concession-0.231
enny-0.228
Token a
Token ID257
Feature activation-4.39e+00

Pos logit contributions
hole+0.140
 atop+0.131
 pieces+0.130
 edges+0.125
ps+0.124
Neg logit contributions
 Fro-0.094
Jack-0.091
Un-0.091
David-0.085
Jim-0.083
Token wild
Token ID4295
Feature activation-1.03e+01

Pos logit contributions
ps+0.047
els+0.043
hole+0.041
 atop+0.040
cks+0.039
Neg logit contributions
Un-0.062
 Fro-0.061
 concession-0.059
Jack-0.058
Da-0.056
Token bear
Token ID6842
Feature activation-1.51e+01

Pos logit contributions
hole+0.104
ps+0.098
 atop+0.098
 pieces+0.097
 hardest+0.092
Neg logit contributions
 Fro-0.136
Un-0.128
Jack-0.127
 concession-0.126
Da-0.122
Token.
Token ID13
Feature activation+4.07e+00

Pos logit contributions
 pieces+0.135
hole+0.133
 atop+0.126
 edges+0.125
ps+0.124
Neg logit contributions
 Fro-0.211
Un-0.209
Jack-0.204
 concession-0.199
David-0.198
Token to
Token ID284
Feature activation-5.84e+00

Pos logit contributions
hole+0.100
 atop+0.093
 pieces+0.093
ps+0.090
 lid+0.089
Neg logit contributions
Jack-0.069
 Fro-0.069
Un-0.067
David-0.067
Da-0.064
Token be
Token ID307
Feature activation-8.62e+00

Pos logit contributions
hole+0.073
 pieces+0.073
 atop+0.068
 on+0.068
 edges+0.067
Neg logit contributions
Un-0.061
Jack-0.061
David-0.060
 Fro-0.060
Jim-0.057
Token careful
Token ID8161
Feature activation-1.23e+01

Pos logit contributions
hole+0.099
 pieces+0.095
 atop+0.089
 edges+0.088
 lid+0.087
Neg logit contributions
Un-0.102
Jack-0.102
David-0.100
 Fro-0.100
Da-0.096
Token when
Token ID618
Feature activation-3.91e+00

Pos logit contributions
hole+0.140
 pieces+0.136
 atop+0.129
ps+0.126
 lid+0.126
Neg logit contributions
Jack-0.142
Un-0.140
David-0.137
 Fro-0.135
His-0.133
Token collecting
Token ID13157
Feature activation-1.66e+01

Pos logit contributions
hole+0.063
 pieces+0.057
ps+0.056
 lid+0.056
 atop+0.056
Neg logit contributions
 Fro-0.031
Jack-0.030
Un-0.030
David-0.028
Kids-0.028
Token rocks
Token ID12586
Feature activation-2.06e+01

Pos logit contributions
hole+0.156
 atop+0.131
 slot+0.123
 globe+0.121
 changing+0.116
Neg logit contributions
 Fro-0.228
Jack-0.209
His-0.209
He-0.206
 His-0.202
Token and
Token ID290
Feature activation-5.44e+00

Pos logit contributions
hole+0.241
 lid+0.232
 edges+0.232
ps+0.226
 atop+0.225
Neg logit contributions
 Fro-0.215
Jack-0.201
Un-0.201
enny-0.199
He-0.194
Token always
Token ID1464
Feature activation-6.52e+00

Pos logit contributions
 pieces+0.060
hole+0.059
 on+0.055
 edges+0.054
 atop+0.054
Neg logit contributions
Un-0.066
Jack-0.065
David-0.065
 Fro-0.064
Jim-0.063
Token make
Token ID787
Feature activation-1.15e+01

Pos logit contributions
hole+0.094
 pieces+0.093
 atop+0.088
 lid+0.086
 edges+0.086
Neg logit contributions
Un-0.055
 Fro-0.055
Jack-0.054
His-0.053
David-0.052
Token sure
Token ID1654
Feature activation-1.11e+01

Pos logit contributions
 pieces+0.143
hole+0.138
 on+0.132
 edges+0.130
 falling+0.129
Neg logit contributions
Un-0.125
David-0.124
Jack-0.123
 Fro-0.120
 concession-0.120
Token she
Token ID673
Feature activation-8.21e+00

Pos logit contributions
hole+0.155
 pieces+0.150
 atop+0.145
 edges+0.142
ps+0.141
Neg logit contributions
Jack-0.101
Un-0.101
 Fro-0.100
David-0.099
Kids-0.095
Token.
Token ID13
Feature activation+3.38e+00

Pos logit contributions
 atop+0.171
els+0.169
ool+0.164
 pieces+0.163
hole+0.161
Neg logit contributions
Un-0.159
 Fro-0.150
 His-0.148
 uncle-0.147
His-0.143
Token They
Token ID1119
Feature activation-3.57e+00

Pos logit contributions
 concession+0.032
David+0.032
Jack+0.030
Un+0.030
oph+0.030
Neg logit contributions
 pieces-0.047
 on-0.045
 lid-0.045
hole-0.044
 edges-0.044
Token had
Token ID550
Feature activation-5.96e+00

Pos logit contributions
hole+0.047
 pieces+0.046
 atop+0.044
 edges+0.043
 on+0.043
Neg logit contributions
Un-0.035
Jack-0.035
 Fro-0.035
David-0.035
Kids-0.033
Token to
Token ID284
Feature activation-5.44e+00

Pos logit contributions
hole+0.077
 pieces+0.073
 atop+0.072
 edges+0.069
ps+0.068
Neg logit contributions
Un-0.062
 Fro-0.062
Jack-0.061
David-0.061
Jim-0.058
Token get
Token ID651
Feature activation-9.46e+00

Pos logit contributions
 pieces+0.058
 on+0.058
ps+0.056
hole+0.056
 falling+0.056
Neg logit contributions
Jack-0.068
enny-0.068
David-0.067
 concession-0.065
oph-0.065
Token stitches
Token ID28096
Feature activation-2.06e+01

Pos logit contributions
hole+0.099
 pieces+0.095
 atop+0.089
 edges+0.088
ps+0.085
Neg logit contributions
Un-0.121
 Fro-0.120
Jack-0.120
David-0.118
 concession-0.115
Token and
Token ID290
Feature activation-5.41e+00

Pos logit contributions
 edges+0.228
hole+0.228
 atop+0.226
 lid+0.225
 pieces+0.224
Neg logit contributions
Un-0.213
Jack-0.212
enny-0.208
 Fro-0.208
David-0.204
Token shots
Token ID6934
Feature activation-1.77e+01

Pos logit contributions
 pieces+0.053
 on+0.049
hole+0.049
 falling+0.047
 edges+0.046
Neg logit contributions
Jack-0.076
David-0.076
oph-0.074
Jim-0.074
 Fro-0.073
Token.
Token ID13
Feature activation+2.84e+00

Pos logit contributions
hole+0.203
 hardest+0.198
 atop+0.193
els+0.192
 edges+0.192
Neg logit contributions
Un-0.188
He-0.186
Jack-0.185
 Fro-0.184
His-0.183
Token They
Token ID1119
Feature activation-3.82e+00

Pos logit contributions
 concession+0.027
David+0.026
oph+0.025
Un+0.025
Jim+0.024
Neg logit contributions
 pieces-0.040
 on-0.039
 edges-0.038
hole-0.038
 lid-0.037
Token had
Token ID550
Feature activation-6.21e+00

Pos logit contributions
hole+0.050
 pieces+0.049
 atop+0.047
 edges+0.046
 on+0.045
Neg logit contributions
Un-0.038
Jack-0.038
David-0.038
 Fro-0.037
Kids-0.036
Token that
Token ID326
Feature activation+1.64e+01

Pos logit contributions
hole+0.069
 pieces+0.068
 atop+0.065
 edges+0.064
ps+0.064
Neg logit contributions
Jack-0.046
 Fro-0.046
Un-0.045
David-0.045
Jim-0.043
Token were
Token ID547
Feature activation+1.07e+01

Pos logit contributions
 Fro+0.163
Jack+0.162
Un+0.155
David+0.149
 concession+0.149
Neg logit contributions
hole-0.225
 atop-0.202
ps-0.200
 pieces-0.196
 edges-0.195
Token all
Token ID477
Feature activation+1.06e+01

Pos logit contributions
 Fro+0.105
Un+0.103
Jack+0.102
 please+0.096
David+0.096
Neg logit contributions
hole-0.148
 lid-0.124
 globe-0.121
ool-0.121
 atop-0.120
Token mixed
Token ID7668
Feature activation+1.21e+00

Pos logit contributions
Jack+0.098
 Jack+0.092
 He+0.092
 His+0.091
 concession+0.090
Neg logit contributions
hole-0.137
 lid-0.130
els-0.128
 plank-0.122
 atop-0.121
Token up
Token ID510
Feature activation+4.60e+00

Pos logit contributions
Jack+0.015
Johnny+0.014
They+0.014
oph+0.013
addy+0.013
Neg logit contributions
hole-0.013
 code-0.012
ps-0.012
 slot-0.012
 atop-0.011
Token.
Token ID13
Feature activation+1.89e+01

Pos logit contributions
Jack+0.053
 Fro+0.051
 concession+0.050
Un+0.049
David+0.049
Neg logit contributions
hole-0.055
 atop-0.048
ps-0.046
 lid-0.045
 code-0.045
Token Jane
Token ID12091
Feature activation+1.12e+01

Pos logit contributions
Un+0.149
David+0.146
 concession+0.144
Jack+0.141
 Fro+0.139
Neg logit contributions
hole-0.274
 pieces-0.273
 on-0.264
 atop-0.256
 edges-0.255
Token didn
Token ID1422
Feature activation+9.56e+00

Pos logit contributions
Jack+0.105
Un+0.097
David+0.096
 Fro+0.095
Jim+0.095
Neg logit contributions
hole-0.155
 pieces-0.150
 atop-0.146
 lid-0.145
 edges-0.143
Token't
Token ID470
Feature activation+8.49e+00

Pos logit contributions
Un+0.110
David+0.109
Jack+0.109
 Fro+0.109
Kids+0.105
Neg logit contributions
 pieces-0.108
hole-0.108
 on-0.102
 atop-0.100
 falling-0.099
Token know
Token ID760
Feature activation+4.67e+00

Pos logit contributions
Jack+0.068
David+0.066
 Fro+0.065
Un+0.064
Kids+0.062
Neg logit contributions
hole-0.131
 pieces-0.129
 atop-0.125
 edges-0.121
els-0.119
Token what
Token ID644
Feature activation+8.24e+00

Pos logit contributions
Jack+0.033
 Fro+0.032
David+0.032
Un+0.031
Johnny+0.031
Neg logit contributions
hole-0.073
 atop-0.070
 pieces-0.070
ps-0.070
 edges-0.069
Token
Token ID198
Feature activation+1.89e+01

Pos logit contributions
Un+0.177
David+0.176
oph+0.172
Jack+0.170
 concession+0.170
Neg logit contributions
hole-0.331
 on-0.327
 pieces-0.323
 atop-0.307
 trunk-0.307
Token
Token ID198
Feature activation+1.84e+01

Pos logit contributions
oph+0.195
 concession+0.194
 Fro+0.182
Un+0.181
David+0.179
Neg logit contributions
 pieces-0.223
hole-0.217
 on-0.216
 atop-0.200
 hard-0.198
Tokenâ
Token ID22940
Feature activation+1.73e+01

Pos logit contributions
 Fro+0.173
oph+0.169
 concession+0.164
latable+0.152
ory+0.150
Neg logit contributions
 pieces-0.224
hole-0.222
 on-0.215
 atop-0.209
 tangled-0.202
Token
Token ID26391
Feature activation+1.91e+01

Pos logit contributions
Jack+0.116
David+0.112
Un+0.105
Jim+0.105
Kids+0.105
Neg logit contributions
 pieces-0.306
hole-0.281
 lid-0.280
 on-0.278
 edges-0.273
Token
Token ID129
Feature activation+1.39e+01

Pos logit contributions
 concession+0.181
Un+0.180
 Fro+0.177
Kids+0.176
Jack+0.176
Neg logit contributions
hole-0.258
 pieces-0.243
 on-0.237
 atop-0.233
 falling-0.233
Token
Token ID241
Feature activation+2.03e+01

Pos logit contributions
Un+0.138
Jack+0.137
David+0.137
 Fro+0.135
Jim+0.130
Neg logit contributions
 pieces-0.183
hole-0.182
 atop-0.170
 on-0.170
 edges-0.168
TokenWhat
Token ID2061
Feature activation+1.22e+01

Pos logit contributions
 Fro+0.164
oph+0.157
 concession+0.156
Un+0.151
David+0.149
Neg logit contributions
hole-0.300
 on-0.293
 pieces-0.292
 edges-0.281
 atop-0.280
Tokenâ
Token ID22940
Feature activation+1.29e+01

Pos logit contributions
Un+0.147
David+0.146
 Fro+0.145
Jack+0.145
 concession+0.140
Neg logit contributions
 pieces-0.133
hole-0.132
 on-0.124
 atop-0.121
 edges-0.120
Token
Token ID26391
Feature activation+1.54e+01

Pos logit contributions
Jack+0.080
David+0.078
Kids+0.074
Un+0.073
Johnny+0.072
Neg logit contributions
 pieces-0.236
hole-0.223
 edges-0.216
 lid-0.215
 matches-0.212
Token
Token ID8151
Feature activation+1.08e+01

Pos logit contributions
 concession+0.134
 Fro+0.132
Un+0.132
Jack+0.128
Kids+0.127
Neg logit contributions
hole-0.212
 pieces-0.204
 on-0.201
 atop-0.196
 falling-0.194
Tokens
Token ID82
Feature activation+1.09e+01

Pos logit contributions
Jack+0.124
David+0.122
 Fro+0.121
Un+0.121
Jim+0.117
Neg logit contributions
 pieces-0.125
hole-0.123
 edges-0.117
ps-0.116
 atop-0.116
Token
Token ID198
Feature activation+1.71e+01

Pos logit contributions
David+0.160
oph+0.154
Un+0.153
Jack+0.151
 concession+0.150
Neg logit contributions
 on-0.275
hole-0.270
 pieces-0.268
 trunk-0.261
 atop-0.257
Token
Token ID198
Feature activation+1.74e+01

Pos logit contributions
oph+0.182
 concession+0.179
David+0.169
 Fro+0.167
 coach+0.167
Neg logit contributions
 pieces-0.198
 on-0.192
hole-0.189
 hard-0.177
 trunk-0.176
Tokenâ
Token ID22940
Feature activation+1.61e+01

Pos logit contributions
 Fro+0.164
oph+0.161
 concession+0.154
addy+0.143
ory+0.143
Neg logit contributions
 pieces-0.208
hole-0.206
 on-0.201
 atop-0.197
 trunk-0.190
Token
Token ID26391
Feature activation+1.79e+01

Pos logit contributions
Jack+0.101
David+0.097
Un+0.094
Jim+0.093
Kids+0.093
Neg logit contributions
 pieces-0.293
hole-0.268
 lid-0.265
 on-0.265
 matches-0.260
Token
Token ID129
Feature activation+1.31e+01

Pos logit contributions
 concession+0.165
Un+0.165
Jack+0.163
Kids+0.162
 Fro+0.161
Neg logit contributions
hole-0.244
 pieces-0.231
 on-0.224
 falling-0.221
 atop-0.220
Token
Token ID241
Feature activation+1.97e+01

Pos logit contributions
Un+0.123
David+0.122
Jack+0.121
 Fro+0.120
Jim+0.115
Neg logit contributions
 pieces-0.180
hole-0.179
 atop-0.167
 on-0.167
 edges-0.165
TokenWhy
Token ID5195
Feature activation+1.34e+01

Pos logit contributions
 Fro+0.160
oph+0.158
 concession+0.155
Un+0.149
David+0.148
Neg logit contributions
hole-0.286
 on-0.281
 pieces-0.279
 edges-0.269
 atop-0.267
Token don
Token ID836
Feature activation+1.03e+01

Pos logit contributions
Un+0.150
Jack+0.149
David+0.148
 Fro+0.147
 concession+0.142
Neg logit contributions
hole-0.157
 pieces-0.157
 on-0.147
 atop-0.145
 edges-0.144
Tokenâ
Token ID22940
Feature activation+1.29e+01

Pos logit contributions
 Fro+0.136
David+0.135
Un+0.133
Jack+0.133
oph+0.132
Neg logit contributions
 on-0.098
 pieces-0.097
hole-0.094
 falling-0.090
 atop-0.089
Token
Token ID26391
Feature activation+1.50e+01

Pos logit contributions
Jack+0.071
David+0.070
Un+0.069
 Fro+0.066
Jim+0.065
Neg logit contributions
 pieces-0.235
hole-0.228
 edges-0.218
 lid-0.216
 atop-0.216
Token
Token ID8151
Feature activation+9.37e+00

Pos logit contributions
 concession+0.131
Jack+0.128
Kids+0.128
 Fro+0.128
Un+0.128
Neg logit contributions
hole-0.205
 pieces-0.197
 on-0.195
 falling-0.189
 atop-0.187
Token dead
Token ID2636
Feature activation+8.73e+00

Pos logit contributions
 pieces+0.118
hole+0.117
ps+0.112
 edges+0.110
els+0.107
Neg logit contributions
 Fro-0.167
Jack-0.164
Un-0.162
David-0.160
Jim-0.157
Token squirrel
Token ID33039
Feature activation+5.96e+00

Pos logit contributions
Jack+0.118
 Fro+0.118
Un+0.117
David+0.115
 concession+0.113
Neg logit contributions
hole-0.081
 pieces-0.075
 atop-0.072
 edges-0.069
ps-0.069
Token in
Token ID287
Feature activation+1.11e+01

Pos logit contributions
 Fro+0.072
Jack+0.071
Un+0.069
David+0.068
enny+0.067
Neg logit contributions
hole-0.063
 pieces-0.062
 edges-0.059
 lid-0.058
 atop-0.058
Token the
Token ID262
Feature activation+1.30e+01

Pos logit contributions
Jack+0.125
 Fro+0.124
Un+0.124
David+0.121
Jim+0.119
Neg logit contributions
hole-0.129
 pieces-0.122
 atop-0.119
 edges-0.115
 hardest-0.113
Token grass
Token ID8701
Feature activation+4.95e+00

Pos logit contributions
 Fro+0.168
Un+0.166
Jack+0.165
 concession+0.164
David+0.161
Neg logit contributions
hole-0.133
 pieces-0.118
 atop-0.117
ps-0.117
els-0.116
Token.
Token ID13
Feature activation+1.90e+01

Pos logit contributions
 Fro+0.065
Jack+0.063
Un+0.063
David+0.062
Jim+0.061
Neg logit contributions
 pieces-0.045
hole-0.044
 atop-0.042
 edges-0.041
ps-0.041
Token She
Token ID1375
Feature activation+1.22e+01

Pos logit contributions
David+0.151
Un+0.151
 concession+0.149
Jack+0.144
oph+0.141
Neg logit contributions
hole-0.268
 pieces-0.264
 on-0.262
 trunk-0.252
 atop-0.250
Token stared
Token ID18484
Feature activation+3.37e+00

Pos logit contributions
Un+0.132
Jack+0.132
David+0.131
 Fro+0.130
Jim+0.125
Neg logit contributions
hole-0.148
 pieces-0.148
 atop-0.138
 on-0.137
 edges-0.136
Token at
Token ID379
Feature activation+8.26e+00

Pos logit contributions
Un+0.047
Jack+0.047
David+0.047
 Fro+0.046
Kids+0.045
Neg logit contributions
hole-0.030
 pieces-0.030
 atop-0.028
 on-0.027
 edges-0.027
Token it
Token ID340
Feature activation+2.43e+00

Pos logit contributions
Un+0.097
Jack+0.097
 Fro+0.096
David+0.096
Jim+0.092
Neg logit contributions
hole-0.093
 pieces-0.092
 atop-0.086
 edges-0.084
 on-0.084
Token for
Token ID329
Feature activation+7.07e+00

Pos logit contributions
 Fro+0.019
 Her+0.019
 please+0.019
 His+0.019
 She+0.019
Neg logit contributions
bones-0.033
 plank-0.032
 slot-0.032
hole-0.032
 pits-0.031
Token could
Token ID714
Feature activation+9.19e+00

Pos logit contributions
Jack+0.021
 Fro+0.021
 concession+0.021
David+0.020
Un+0.020
Neg logit contributions
hole-0.018
 edges-0.017
 pieces-0.017
ps-0.017
els-0.017
Token be
Token ID307
Feature activation+5.32e+00

Pos logit contributions
Un+0.090
Jack+0.089
 Fro+0.088
David+0.088
 concession+0.085
Neg logit contributions
hole-0.119
 pieces-0.117
 atop-0.110
 edges-0.109
 lid-0.107
Token felt
Token ID2936
Feature activation+7.09e+00

Pos logit contributions
David+0.060
Jack+0.059
Un+0.058
 Fro+0.058
Da+0.057
Neg logit contributions
hole-0.063
 lid-0.061
 button-0.058
 pieces-0.056
 edges-0.055
Token all
Token ID477
Feature activation+8.82e+00

Pos logit contributions
Jack+0.071
 Fro+0.070
Un+0.068
David+0.067
 concession+0.066
Neg logit contributions
hole-0.094
 pieces-0.087
els-0.085
 atop-0.083
 lid-0.083
Token around
Token ID1088
Feature activation+5.71e+00

Pos logit contributions
Jack+0.103
Da+0.097
 Fro+0.096
 concession+0.092
 His+0.092
Neg logit contributions
hole-0.101
 lid-0.096
els-0.093
 code-0.088
 edges-0.087
Token.
Token ID13
Feature activation+1.90e+01

Pos logit contributions
Jack+0.072
 Fro+0.069
Un+0.068
David+0.068
Jim+0.067
Neg logit contributions
hole-0.063
 pieces-0.057
 lid-0.054
ps-0.054
 edges-0.053
Token<|endoftext|>
Token ID50256
Feature activation+1.33e+01

Pos logit contributions
Un+0.166
 concession+0.161
David+0.159
 Fro+0.153
Jack+0.153
Neg logit contributions
hole-0.258
 pieces-0.255
 on-0.244
 edges-0.237
 atop-0.236
TokenOnce
Token ID7454
Feature activation+1.32e+01

Pos logit contributions
Jack+0.080
 Fro+0.075
David+0.071
Jim+0.070
Un+0.070
Neg logit contributions
hole-0.232
 pieces-0.228
ps-0.223
 edges-0.222
 falling-0.221
Token upon
Token ID2402
Feature activation+1.61e+01

Pos logit contributions
Un+0.126
David+0.124
Jack+0.123
 Fro+0.121
 concession+0.118
Neg logit contributions
hole-0.178
 pieces-0.177
 on-0.167
 atop-0.166
 edges-0.164
Token a
Token ID257
Feature activation+1.40e+01

Pos logit contributions
Kids+0.310
ete+0.302
oph+0.302
enny+0.296
 concession+0.296
Neg logit contributions
 on-0.104
 pieces-0.097
 hard-0.079
 broken-0.068
 glowing-0.065
Token time
Token ID640
Feature activation+1.03e+01

Pos logit contributions
 concession+0.281
David+0.279
Kids+0.277
Johnny+0.274
 Fro+0.272
Neg logit contributions
 hard-0.040
 pieces-0.037
 on-0.031
 falling-0.029
 edges-0.027
Token for
Token ID329
Feature activation+8.56e+00

Pos logit contributions
David+0.033
 Fro+0.031
Jack+0.030
enny+0.030
Un+0.030
Neg logit contributions
els-0.102
 pieces-0.099
 matches-0.097
 globe-0.094
 bulb-0.094
Token dinner
Token ID8073
Feature activation-3.08e-01

Pos logit contributions
 Fro+0.082
Jack+0.079
Un+0.079
David+0.078
Jim+0.073
Neg logit contributions
hole-0.125
 atop-0.114
 pieces-0.113
 edges-0.109
els-0.109
Token.
Token ID13
Feature activation+1.81e+01

Pos logit contributions
hole+0.003
 pieces+0.003
 atop+0.003
ps+0.002
els+0.002
Neg logit contributions
Jack-0.004
Un-0.004
David-0.004
 Fro-0.004
Jim-0.004
Token She
Token ID1375
Feature activation+1.14e+01

Pos logit contributions
Un+0.151
 concession+0.148
David+0.148
 Fro+0.138
Jack+0.138
Neg logit contributions
hole-0.247
 pieces-0.246
 on-0.238
 edges-0.227
 trunk-0.227
Token felt
Token ID2936
Feature activation+7.01e+00

Pos logit contributions
Un+0.124
Jack+0.124
David+0.124
 Fro+0.122
Jim+0.117
Neg logit contributions
 pieces-0.137
hole-0.136
 on-0.127
 atop-0.126
 edges-0.125
Token a
Token ID257
Feature activation+8.67e+00

Pos logit contributions
Un+0.085
Jack+0.085
David+0.084
 Fro+0.084
 concession+0.081
Neg logit contributions
hole-0.076
 pieces-0.075
 atop-0.070
 edges-0.069
 on-0.069
Token wave
Token ID6769
Feature activation+1.02e+00

Pos logit contributions
Un+0.100
Jack+0.100
David+0.096
 Fro+0.096
 concession+0.094
Neg logit contributions
hole-0.100
 pieces-0.097
 atop-0.092
ps-0.091
els-0.090
Token of
Token ID286
Feature activation+6.81e+00

Pos logit contributions
Un+0.012
Jack+0.012
 Fro+0.011
David+0.011
 concession+0.011
Neg logit contributions
 pieces-0.012
hole-0.011
 edges-0.011
 atop-0.011
els-0.011
Token relief
Token ID8259
Feature activation+2.76e-01

Pos logit contributions
Jack+0.068
 Fro+0.068
David+0.066
Jim+0.065
 concession+0.065
Neg logit contributions
hole-0.092
 pieces-0.082
 atop-0.080
 edges-0.079
els-0.078
Token as
Token ID355
Feature activation+8.36e+00

Pos logit contributions
Un+0.003
 Fro+0.003
Jack+0.003
David+0.003
 concession+0.003
Neg logit contributions
hole-0.003
 pieces-0.003
 hardest-0.003
 edges-0.003
els-0.003
Token their
Token ID511
Feature activation+7.65e+00

Pos logit contributions
Jack+0.060
Un+0.060
David+0.059
 Fro+0.058
Jim+0.055
Neg logit contributions
hole-0.133
 pieces-0.131
 atop-0.124
 edges-0.124
ps-0.123
Token her
Token ID607
Feature activation+3.46e+00

Pos logit contributions
 Fro+0.021
Jack+0.019
Da+0.018
Un+0.018
 Her+0.018
Neg logit contributions
hole-0.031
 atop-0.029
 pieces-0.027
 lid-0.027
 edges-0.027
Token.
Token ID13
Feature activation+1.31e+01

Pos logit contributions
 Fro+0.035
 uncle+0.030
 wife+0.029
 coach+0.029
 Pete+0.026
Neg logit contributions
hole-0.047
bones-0.046
 atop-0.045
ps-0.045
ool-0.043
Token She
Token ID1375
Feature activation+7.85e+00

Pos logit contributions
 concession+0.126
David+0.113
Jack+0.109
oph+0.109
Jim+0.105
Neg logit contributions
hole-0.183
 on-0.180
 pieces-0.173
 trunk-0.170
 edges-0.168
Token ran
Token ID4966
Feature activation+1.11e+00

Pos logit contributions
David+0.099
Jack+0.096
Un+0.096
 concession+0.093
 Fro+0.093
Neg logit contributions
 pieces-0.085
hole-0.081
 on-0.079
ps-0.076
 tangled-0.075
Token after
Token ID706
Feature activation+4.52e+00

Pos logit contributions
David+0.015
Un+0.015
Jack+0.015
 Fro+0.014
Jim+0.014
Neg logit contributions
 pieces-0.011
hole-0.011
 on-0.010
 edges-0.010
 atop-0.010
Token the
Token ID262
Feature activation+5.72e+00

Pos logit contributions
Un+0.053
David+0.053
Jack+0.052
 Fro+0.051
 concession+0.050
Neg logit contributions
 pieces-0.051
hole-0.050
 on-0.048
 atop-0.047
 falling-0.046
Token dog
Token ID3290
Feature activation-1.39e+00

Pos logit contributions
 Fro+0.037
Un+0.035
Jack+0.035
 concession+0.034
Kids+0.034
Neg logit contributions
hole-0.098
 atop-0.094
els-0.092
ps-0.091
 pieces-0.089
Token and
Token ID290
Feature activation+6.07e+00

Pos logit contributions
hole+0.018
 atop+0.018
 pieces+0.018
 edges+0.018
ool+0.017
Neg logit contributions
 Fro-0.015
Un-0.012
 His-0.012
 Her-0.012
Jack-0.012
Token the
Token ID262
Feature activation+5.57e+00

Pos logit contributions
 Fro+0.068
Un+0.067
Jack+0.067
David+0.066
Kids+0.063
Neg logit contributions
hole-0.073
 pieces-0.071
 atop-0.068
 edges-0.067
ps-0.065
Token man
Token ID582
Feature activation-1.30e+00

Pos logit contributions
 Fro+0.041
Un+0.040
Jack+0.039
David+0.038
 concession+0.038
Neg logit contributions
hole-0.090
 atop-0.085
els-0.083
ps-0.082
 pieces-0.081
Token and
Token ID290
Feature activation+6.18e+00

Pos logit contributions
 edges+0.017
 hardest+0.017
 atop+0.016
hole+0.016
ool+0.016
Neg logit contributions
 Fro-0.013
 His-0.012
His-0.011
 Her-0.011
 him-0.011
Token"
Token ID1
Feature activation+1.44e+01

Pos logit contributions
 Fro+0.161
 concession+0.153
oph+0.150
enny+0.136
David+0.135
Neg logit contributions
 pieces-0.192
hole-0.192
 on-0.185
 atop-0.178
 tangled-0.178
TokenPlease
Token ID5492
Feature activation+9.45e+00

Pos logit contributions
 Fro+0.107
David+0.105
 concession+0.105
enny+0.102
ete+0.101
Neg logit contributions
 pieces-0.230
 tangled-0.215
 on-0.213
 edges-0.210
ps-0.209
Token,
Token ID11
Feature activation+1.00e+01

Pos logit contributions
 Fro+0.119
Jack+0.116
Un+0.113
David+0.111
Kids+0.109
Neg logit contributions
hole-0.105
 pieces-0.101
 atop-0.096
 edges-0.092
ps-0.089
Token Lily
Token ID20037
Feature activation+6.18e+00

Pos logit contributions
Un+0.084
David+0.084
Jack+0.082
 Fro+0.077
 concession+0.077
Neg logit contributions
 pieces-0.146
hole-0.145
 on-0.141
 lid-0.137
 falling-0.137
Token,
Token ID11
Feature activation+9.26e+00

Pos logit contributions
Un+0.087
Jack+0.087
 Fro+0.086
David+0.086
Jim+0.083
Neg logit contributions
hole-0.055
 pieces-0.055
 atop-0.050
 edges-0.049
 on-0.048
Token you
Token ID345
Feature activation+5.75e+00

Pos logit contributions
David+0.102
Un+0.101
Jack+0.099
 Fro+0.095
Da+0.094
Neg logit contributions
 pieces-0.112
hole-0.110
 on-0.110
 falling-0.106
ps-0.104
Token have
Token ID423
Feature activation+4.88e+00

Pos logit contributions
 Fro+0.065
Jack+0.063
Jim+0.058
enny+0.058
Un+0.058
Neg logit contributions
hole-0.072
 atop-0.068
 pieces-0.066
bones-0.065
 edges-0.064
Token to
Token ID284
Feature activation+5.44e+00

Pos logit contributions
Jack+0.047
Un+0.046
 Fro+0.046
David+0.046
Jim+0.044
Neg logit contributions
hole-0.067
 pieces-0.063
 atop-0.062
ps-0.060
 edges-0.059
Token take
Token ID1011
Feature activation-6.21e-01

Pos logit contributions
Un+0.063
David+0.062
Jack+0.061
 concession+0.060
 Fro+0.059
Neg logit contributions
 pieces-0.063
hole-0.061
 on-0.059
ps-0.058
 falling-0.058
Token the
Token ID262
Feature activation+4.30e+00

Pos logit contributions
hole+0.009
 atop+0.009
 globe+0.008
 pieces+0.008
 changing+0.008
Neg logit contributions
Jack-0.006
 Fro-0.006
Un-0.005
Da-0.005
Jim-0.005
Token pill
Token ID9582
Feature activation-3.68e+00

Pos logit contributions
 Fro+0.051
 coach+0.048
 concession+0.047
 wife+0.046
 shepherd+0.045
Neg logit contributions
hole-0.053
 atop-0.047
cks-0.044
els-0.044
bones-0.043
Token kept
Token ID4030
Feature activation+1.08e+01

Pos logit contributions
hole+0.159
 pieces+0.158
 atop+0.146
 edges+0.145
ps+0.144
Neg logit contributions
 Fro-0.206
Un-0.205
Jack-0.205
David-0.204
 concession-0.197
Token asking
Token ID4737
Feature activation+6.54e+00

Pos logit contributions
Jack+0.130
 Fro+0.130
David+0.129
Un+0.127
Johnny+0.123
Neg logit contributions
hole-0.120
 pieces-0.113
 edges-0.106
 lid-0.105
 atop-0.105
Token until
Token ID1566
Feature activation+1.20e+01

Pos logit contributions
Jack+0.076
 Fro+0.074
David+0.074
Un+0.074
His+0.071
Neg logit contributions
hole-0.075
 pieces-0.073
 atop-0.069
 edges-0.068
 falling-0.067
Token finally
Token ID3443
Feature activation+1.07e+01

Pos logit contributions
Jack+0.096
David+0.094
Un+0.094
 Fro+0.094
Jim+0.089
Neg logit contributions
hole-0.184
 pieces-0.179
 edges-0.170
ps-0.169
 atop-0.168
Token,
Token ID11
Feature activation+1.24e+01

Pos logit contributions
Jack+0.112
David+0.111
 Fro+0.109
Un+0.109
Jim+0.106
Neg logit contributions
hole-0.135
 pieces-0.134
 edges-0.126
 atop-0.123
ps-0.122
Token his
Token ID465
Feature activation+1.24e+01

Pos logit contributions
Jack+0.127
David+0.127
Un+0.126
 Fro+0.124
Jim+0.120
Neg logit contributions
hole-0.158
 pieces-0.156
 edges-0.147
ps-0.146
 atop-0.145
Token parents
Token ID3397
Feature activation+7.72e+00

Pos logit contributions
Un+0.102
 Fro+0.101
Jack+0.101
David+0.101
 concession+0.096
Neg logit contributions
hole-0.182
 pieces-0.177
 atop-0.172
 edges-0.166
ps-0.166
Token agreed
Token ID4987
Feature activation+4.72e+00

Pos logit contributions
David+0.077
Jack+0.077
Un+0.073
Johnny+0.071
Jim+0.070
Neg logit contributions
 pieces-0.102
 edges-0.101
hole-0.100
 atop-0.095
ps-0.094
Token to
Token ID284
Feature activation+8.13e+00

Pos logit contributions
Jack+0.055
David+0.054
Un+0.053
 Fro+0.052
Jim+0.051
Neg logit contributions
hole-0.055
 pieces-0.053
 atop-0.050
 edges-0.050
ps-0.049
Token help
Token ID1037
Feature activation+2.94e+00

Pos logit contributions
Un+0.088
Jack+0.087
David+0.087
 Fro+0.086
 concession+0.083
Neg logit contributions
hole-0.098
 pieces-0.098
 atop-0.091
 on-0.091
 edges-0.090
Token him
Token ID683
Feature activation+3.83e+00

Pos logit contributions
Jack+0.027
Un+0.027
David+0.027
 Fro+0.026
Jim+0.025
Neg logit contributions
hole-0.041
 pieces-0.040
 atop-0.038
 edges-0.038
ps-0.037
Token,
Token ID11
Feature activation+1.45e+01

Pos logit contributions
Un+0.193
David+0.189
Jack+0.189
Kids+0.185
 concession+0.185
Neg logit contributions
 on-0.112
 pieces-0.111
hole-0.110
 atop-0.102
ps-0.101
Token "
Token ID366
Feature activation+1.82e+01

Pos logit contributions
Un+0.136
Jack+0.135
David+0.134
 Fro+0.133
 concession+0.127
Neg logit contributions
hole-0.195
 pieces-0.195
 on-0.183
 atop-0.183
 edges-0.181
TokenYes
Token ID5297
Feature activation+1.13e+01

Pos logit contributions
 Fro+0.167
 concession+0.156
oph+0.145
addy+0.143
ete+0.141
Neg logit contributions
hole-0.233
 pieces-0.231
 atop-0.221
 on-0.220
 edges-0.217
Token,
Token ID11
Feature activation+1.29e+01

Pos logit contributions
Jack+0.148
 Fro+0.146
David+0.146
Un+0.146
Jim+0.142
Neg logit contributions
hole-0.111
 pieces-0.110
 atop-0.103
 edges-0.099
ps-0.098
Token m
Token ID285
Feature activation+1.06e+01

Pos logit contributions
Un+0.120
David+0.119
Jack+0.117
 Fro+0.114
 concession+0.113
Neg logit contributions
 pieces-0.174
hole-0.172
 on-0.170
 edges-0.160
 falling-0.159
Tokenama
Token ID1689
Feature activation+7.27e+00

Pos logit contributions
Un+0.091
Jack+0.090
David+0.090
 Fro+0.089
 concession+0.085
Neg logit contributions
 pieces-0.154
hole-0.154
 atop-0.145
 edges-0.144
 on-0.144
Token.
Token ID13
Feature activation+1.59e+01

Pos logit contributions
 Fro+0.095
Un+0.094
Jack+0.089
David+0.089
Kids+0.087
Neg logit contributions
hole-0.073
 pieces-0.071
 edges-0.068
 atop-0.067
 tangled-0.067
Token I
Token ID314
Feature activation+1.07e+01

Pos logit contributions
David+0.154
Jack+0.152
Un+0.147
 Fro+0.142
Jim+0.142
Neg logit contributions
hole-0.213
 pieces-0.205
 on-0.200
 edges-0.193
 atop-0.193
Token can
Token ID460
Feature activation+8.61e+00

Pos logit contributions
Jack+0.097
 Fro+0.096
Un+0.095
David+0.093
Kids+0.090
Neg logit contributions
hole-0.151
 pieces-0.150
 atop-0.142
 edges-0.141
 falling-0.137
Token be
Token ID307
Feature activation+4.04e+00

Pos logit contributions
Un+0.105
David+0.104
Jack+0.104
 Fro+0.103
Kids+0.100
Neg logit contributions
hole-0.093
 pieces-0.093
 on-0.087
 edges-0.085
 atop-0.085
Token quiet
Token ID5897
Feature activation-8.84e-01

Pos logit contributions
 Fro+0.046
Un+0.046
Jack+0.046
David+0.046
Jim+0.044
Neg logit contributions
hole-0.048
 pieces-0.046
 atop-0.043
 edges-0.043
 lid-0.043
Token He
Token ID679
Feature activation+6.18e+00

Pos logit contributions
David+0.075
 Fro+0.071
Un+0.071
Kids+0.070
Jack+0.070
Neg logit contributions
 pieces-0.209
hole-0.206
 on-0.197
 edges-0.196
 atop-0.193
Token understood
Token ID7247
Feature activation+1.71e+00

Pos logit contributions
Jack+0.064
Un+0.061
 Fro+0.060
David+0.059
Jim+0.058
Neg logit contributions
hole-0.082
 atop-0.075
 pieces-0.074
 falling-0.074
 edges-0.073
Token that
Token ID326
Feature activation+3.76e+00

Pos logit contributions
Jack+0.017
Un+0.017
David+0.017
 Fro+0.017
Jim+0.016
Neg logit contributions
hole-0.023
 pieces-0.022
 edges-0.021
 atop-0.021
ps-0.020
Token it
Token ID340
Feature activation-2.22e-01

Pos logit contributions
Jack+0.031
Un+0.030
 Fro+0.030
David+0.029
Jim+0.028
Neg logit contributions
hole-0.058
ps-0.054
 atop-0.053
 lid-0.052
els-0.052
Token was
Token ID373
Feature activation+5.85e-01

Pos logit contributions
hole+0.003
 atop+0.003
 pieces+0.003
 edges+0.003
 lid+0.003
Neg logit contributions
Jack-0.002
Un-0.002
 Fro-0.002
 concession-0.002
David-0.002
Token much
Token ID881
Feature activation-6.76e-01

Pos logit contributions
Un+0.005
Jack+0.005
His+0.005
 Fro+0.005
David+0.005
Neg logit contributions
hole-0.009
 lid-0.008
 globe-0.008
 pieces-0.007
els-0.007
Token too
Token ID1165
Feature activation+8.07e-01

Pos logit contributions
hole+0.007
ps+0.006
 atop+0.006
 globe+0.006
 edges+0.006
Neg logit contributions
Jack-0.009
Un-0.008
 Fro-0.008
 concession-0.008
David-0.008
Token cold
Token ID4692
Feature activation-5.80e+00

Pos logit contributions
Un+0.010
Jack+0.010
David+0.009
He+0.009
Johnny+0.009
Neg logit contributions
hole-0.010
bones-0.008
els-0.008
ps-0.008
 atop-0.008
Token outside
Token ID2354
Feature activation-1.93e+00

Pos logit contributions
hole+0.067
 pieces+0.065
 atop+0.062
 lid+0.061
 edges+0.060
Neg logit contributions
Jack-0.065
Un-0.065
 Fro-0.063
David-0.063
 concession-0.062
Token,
Token ID11
Feature activation+3.75e+00

Pos logit contributions
hole+0.022
 pieces+0.020
 atop+0.020
ps+0.020
 edges+0.019
Neg logit contributions
Jack-0.022
Un-0.021
 Fro-0.021
David-0.021
 concession-0.021
Token and
Token ID290
Feature activation+5.35e+00

Pos logit contributions
Un+0.041
Jack+0.041
David+0.041
 Fro+0.040
Kids+0.039
Neg logit contributions
 pieces-0.045
hole-0.045
 on-0.042
 atop-0.041
 edges-0.041
Token was
Token ID373
Feature activation-8.40e-01

Pos logit contributions
David+0.021
Un+0.020
Jack+0.020
oph+0.020
Jim+0.020
Neg logit contributions
 on-0.021
 pieces-0.021
hole-0.020
 falling-0.020
ps-0.020
Token scared
Token ID12008
Feature activation-7.50e+00

Pos logit contributions
 pieces+0.008
hole+0.008
 on+0.008
 falling+0.008
 atop+0.008
Neg logit contributions
Jack-0.011
David-0.011
Un-0.011
 Fro-0.010
Kids-0.010
Token of
Token ID286
Feature activation-4.01e+00

Pos logit contributions
 pieces+0.078
hole+0.077
 on+0.073
 edges+0.071
 atop+0.070
Neg logit contributions
David-0.096
Un-0.095
Jack-0.095
 Fro-0.093
 concession-0.092
Token the
Token ID262
Feature activation-3.25e+00

Pos logit contributions
hole+0.055
 atop+0.052
ool+0.049
 lid+0.049
 pieces+0.049
Neg logit contributions
 Fro-0.043
Un-0.041
Jack-0.039
David-0.038
Johnny-0.037
Token dark
Token ID3223
Feature activation-8.90e+00

Pos logit contributions
hole+0.042
 atop+0.038
els+0.037
cks+0.036
ps+0.035
Neg logit contributions
 Fro-0.038
 concession-0.036
Un-0.036
 coach-0.035
Kids-0.035
Token,
Token ID11
Feature activation-1.12e+00

Pos logit contributions
hole+0.086
 pieces+0.082
 atop+0.082
ps+0.078
 hardest+0.075
Neg logit contributions
Un-0.116
Jack-0.115
 Fro-0.113
David-0.112
Kids-0.111
Token loud
Token ID7812
Feature activation-7.36e+00

Pos logit contributions
hole+0.014
 pieces+0.013
els+0.013
ps+0.013
 atop+0.013
Neg logit contributions
 Fro-0.012
Un-0.012
Jack-0.012
David-0.011
His-0.011
Token noises
Token ID26782
Feature activation-1.04e+01

Pos logit contributions
hole+0.089
 atop+0.083
 pieces+0.081
 hardest+0.077
els+0.077
Neg logit contributions
Un-0.083
 Fro-0.082
Jack-0.080
 concession-0.078
Kids-0.078
Token,
Token ID11
Feature activation-9.04e-01

Pos logit contributions
hole+0.136
 lid+0.134
 hardest+0.130
 pieces+0.124
ool+0.123
Neg logit contributions
Un-0.108
Jack-0.107
 Fro-0.107
 him-0.106
His-0.100
Token and
Token ID290
Feature activation+7.32e-01

Pos logit contributions
hole+0.012
 pieces+0.011
 atop+0.011
 edges+0.011
 lid+0.011
Neg logit contributions
 Fro-0.009
Un-0.009
Jack-0.009
David-0.009
 concession-0.008
Token even
Token ID772
Feature activation-3.86e+00

Pos logit contributions
David+0.006
Un+0.006
Jack+0.006
 Fro+0.006
oph+0.006
Neg logit contributions
 pieces-0.011
hole-0.010
 on-0.010
 falling-0.010
 edges-0.010
Token
Token ID26391
Feature activation+6.76e+00

Pos logit contributions
Jack+0.039
David+0.037
Kids+0.035
Un+0.035
Johnny+0.034
Neg logit contributions
 pieces-0.099
hole-0.092
 lid-0.092
 atop-0.090
 on-0.090
Token Jill
Token ID22772
Feature activation+1.68e+00

Pos logit contributions
Un+0.065
Kids+0.065
Da+0.063
 concession+0.063
 Fro+0.063
Neg logit contributions
hole-0.091
 pieces-0.090
 on-0.085
 edges-0.084
 falling-0.082
Token laughed
Token ID13818
Feature activation-3.48e+00

Pos logit contributions
Jack+0.016
 Fro+0.015
David+0.015
Un+0.014
Jim+0.014
Neg logit contributions
hole-0.023
 pieces-0.023
 atop-0.022
 edges-0.022
ps-0.022
Token and
Token ID290
Feature activation+4.72e+00

Pos logit contributions
 on+0.031
 pieces+0.029
hole+0.027
 atop+0.026
 edges+0.025
Neg logit contributions
Kids-0.052
David-0.052
Un-0.052
 concession-0.051
Jack-0.050
Token said
Token ID531
Feature activation+1.50e+00

Pos logit contributions
Un+0.051
Jack+0.051
David+0.051
 Fro+0.050
 concession+0.049
Neg logit contributions
hole-0.057
 pieces-0.057
 on-0.053
 atop-0.053
 edges-0.052
Token,
Token ID11
Feature activation+3.43e+00

Pos logit contributions
Jack+0.021
Un+0.021
 Fro+0.020
David+0.020
Jim+0.020
Neg logit contributions
hole-0.014
 pieces-0.014
 atop-0.012
 edges-0.012
 on-0.012
Token
Token ID6184
Feature activation+2.76e+00

Pos logit contributions
Un+0.029
Jack+0.029
David+0.029
 Fro+0.029
Jim+0.027
Neg logit contributions
hole-0.051
 pieces-0.050
 atop-0.047
 edges-0.047
ps-0.047
Token
Token ID95
Feature activation+4.08e+00

Pos logit contributions
Un+0.029
David+0.029
Jack+0.029
 Fro+0.028
Da+0.028
Neg logit contributions
hole-0.034
 pieces-0.033
 atop-0.032
 on-0.032
 edges-0.031
Token
Token ID26391
Feature activation+5.54e+00

Pos logit contributions
Un+0.039
 Fro+0.038
Jack+0.038
David+0.038
 concession+0.036
Neg logit contributions
hole-0.055
 pieces-0.054
 edges-0.051
 atop-0.051
 on-0.050
Token
Token ID129
Feature activation+1.63e+00

Pos logit contributions
Kids+0.064
Da+0.063
 Fro+0.063
 concession+0.062
Un+0.062
Neg logit contributions
hole-0.064
 pieces-0.063
 on-0.060
 edges-0.057
 falling-0.056
Token
Token ID241
Feature activation+9.03e+00

Pos logit contributions
Un+0.013
Jack+0.013
David+0.013
 Fro+0.013
Kids+0.012
Neg logit contributions
hole-0.024
 pieces-0.024
 atop-0.023
 on-0.022
 edges-0.022
Token It
Token ID632
Feature activation+5.81e+00

Pos logit contributions
 concession+0.135
Un+0.133
David+0.131
Jack+0.127
oph+0.124
Neg logit contributions
 pieces-0.201
hole-0.201
 on-0.195
 edges-0.187
 lid-0.186
Token went
Token ID1816
Feature activation+2.24e+00

Pos logit contributions
 Fro+0.047
Jack+0.043
His+0.043
Un+0.041
 concession+0.040
Neg logit contributions
hole-0.090
els-0.089
ps-0.082
 pieces-0.081
ool-0.081
Token far
Token ID1290
Feature activation-3.77e-01

Pos logit contributions
 Fro+0.027
His+0.024
 concession+0.024
Un+0.023
Jack+0.023
Neg logit contributions
hole-0.027
els-0.026
 pieces-0.026
 edges-0.024
ps-0.024
Token and
Token ID290
Feature activation+7.62e+00

Pos logit contributions
els+0.004
 pieces+0.004
ps+0.004
 lid+0.004
hole+0.004
Neg logit contributions
 Fro-0.004
He-0.004
Jack-0.004
Un-0.004
David-0.004
Token wide
Token ID3094
Feature activation+1.42e-01

Pos logit contributions
 Fro+0.078
Jack+0.073
Un+0.068
 concession+0.067
Johnny+0.066
Neg logit contributions
els-0.115
hole-0.107
ool-0.106
ps-0.099
 pieces-0.094
Token on
Token ID319
Feature activation+2.85e+00

Pos logit contributions
 Fro+0.001
 him+0.001
 Pete+0.001
 He+0.001
 His+0.001
Neg logit contributions
els-0.002
ool-0.002
 pieces-0.002
 splits-0.002
 slot-0.002
Token the
Token ID262
Feature activation+6.40e+00

Pos logit contributions
 Fro+0.025
Un+0.024
David+0.023
Jack+0.023
Jim+0.022
Neg logit contributions
hole-0.044
ool-0.038
 hardest-0.038
 atop-0.038
els-0.037
Token big
Token ID1263
Feature activation+5.70e+00

Pos logit contributions
Un+0.068
David+0.067
 concession+0.067
 Fro+0.065
Jack+0.065
Neg logit contributions
hole-0.078
els-0.075
 atop-0.072
 pieces-0.070
 tangled-0.069
Token blue
Token ID4171
Feature activation+2.98e+00

Pos logit contributions
 Fro+0.052
David+0.052
Un+0.051
Jack+0.050
 concession+0.050
Neg logit contributions
hole-0.078
ps-0.077
 atop-0.073
 tangled-0.071
els-0.071
Token sea
Token ID5417
Feature activation-1.37e+00

Pos logit contributions
Jack+0.035
Un+0.035
David+0.035
 Fro+0.034
Kids+0.033
Neg logit contributions
hole-0.033
 pieces-0.032
 atop-0.031
ps-0.030
 edges-0.030
Token.
Token ID13
Feature activation+1.40e+01

Pos logit contributions
els+0.017
ps+0.016
hole+0.015
ool+0.015
 tangled+0.015
Neg logit contributions
 Fro-0.013
Jack-0.013
He-0.012
They-0.012
 Pete-0.012
Token but
Token ID475
Feature activation+7.11e+00

Pos logit contributions
Un+0.058
Jack+0.057
David+0.057
 Fro+0.055
Kids+0.055
Neg logit contributions
 pieces-0.053
hole-0.052
 on-0.050
 falling-0.048
 edges-0.048
Token he
Token ID339
Feature activation+4.82e+00

Pos logit contributions
David+0.057
Un+0.057
Jack+0.055
oph+0.053
Kids+0.053
Neg logit contributions
 pieces-0.109
 on-0.106
hole-0.105
 falling-0.101
 atop-0.100
Token did
Token ID750
Feature activation+3.11e+00

Pos logit contributions
David+0.057
Un+0.056
Jack+0.055
 concession+0.054
 Fro+0.054
Neg logit contributions
 pieces-0.054
 on-0.051
hole-0.051
 atop-0.048
 tangled-0.048
Token not
Token ID407
Feature activation+4.11e+00

Pos logit contributions
His+0.023
Un+0.021
Jack+0.020
 Fro+0.020
David+0.019
Neg logit contributions
hole-0.052
 pieces-0.051
 lid-0.047
 edges-0.047
 atop-0.047
Token know
Token ID760
Feature activation-3.91e-01

Pos logit contributions
Un+0.025
Jack+0.025
 Fro+0.025
David+0.024
Jim+0.023
Neg logit contributions
hole-0.070
 pieces-0.070
 atop-0.069
 edges-0.066
 lid-0.066
Token how
Token ID703
Feature activation+4.86e+00

Pos logit contributions
hole+0.006
 atop+0.006
 lid+0.006
 edges+0.006
 pieces+0.006
Neg logit contributions
 Fro-0.002
His-0.002
Un-0.002
Jack-0.002
Johnny-0.002
Token.
Token ID13
Feature activation+1.12e+01

Pos logit contributions
Jack+0.054
Un+0.054
 Fro+0.052
His+0.052
David+0.051
Neg logit contributions
hole-0.063
 pieces-0.056
 atop-0.056
ps-0.054
 lid-0.054
Token He
Token ID679
Feature activation+5.09e+00

Pos logit contributions
David+0.070
 concession+0.068
oph+0.066
Jack+0.065
Un+0.065
Neg logit contributions
 pieces-0.193
hole-0.186
 edges-0.184
 on-0.183
 lid-0.181
Token asked
Token ID1965
Feature activation+3.21e+00

Pos logit contributions
David+0.059
Jack+0.058
Un+0.058
 Fro+0.057
 concession+0.057
Neg logit contributions
 pieces-0.058
hole-0.057
 on-0.054
 edges-0.053
ps-0.052
Token,
Token ID11
Feature activation+5.70e+00

Pos logit contributions
 Fro+0.023
Jack+0.021
Un+0.021
His+0.020
Jim+0.019
Neg logit contributions
hole-0.056
 pieces-0.052
 atop-0.051
 tangled-0.050
 edges-0.050
Token "
Token ID366
Feature activation+9.07e+00

Pos logit contributions
Un+0.035
His+0.033
 Fro+0.032
 His+0.031
Jack+0.030
Neg logit contributions
hole-0.105
 pieces-0.102
ool-0.098
els-0.098
 edges-0.097
Token drove
Token ID10357
Feature activation-8.59e+00

Pos logit contributions
 lid+0.178
 mant+0.175
 hardest+0.174
ool+0.173
hole+0.171
Neg logit contributions
 played-0.054
They-0.042
He-0.037
Jack-0.036
 each-0.035
Token over
Token ID625
Feature activation-4.00e+00

Pos logit contributions
ool+0.111
els+0.108
hole+0.108
 teaspoon+0.106
 lid+0.105
Neg logit contributions
 him-0.082
They-0.078
 He-0.077
His-0.074
 each-0.074
Token it
Token ID340
Feature activation-7.62e+00

Pos logit contributions
ps+0.056
hole+0.055
ool+0.055
 atop+0.053
eking+0.052
Neg logit contributions
Jack-0.034
Un-0.034
 him-0.032
They-0.032
Jim-0.031
Token,
Token ID11
Feature activation-1.34e+00

Pos logit contributions
els+0.132
ps+0.125
bones+0.122
cks+0.118
ocol+0.118
Neg logit contributions
 He-0.051
 She-0.046
 As-0.044
 They-0.042
As-0.041
Token hon
Token ID3032
Feature activation-9.97e+00

Pos logit contributions
ool+0.020
els+0.019
hole+0.018
 lid+0.017
 code+0.017
Neg logit contributions
 Fro-0.011
 concession-0.010
 He-0.010
Un-0.010
Jack-0.010
Tokenking
Token ID3364
Feature activation-1.06e+01

Pos logit contributions
 pieces+0.116
hole+0.113
 atop+0.107
 lid+0.106
 edges+0.104
Neg logit contributions
Un-0.112
 Fro-0.111
David-0.111
Jack-0.109
His-0.104
Token and
Token ID290
Feature activation+2.97e-01

Pos logit contributions
hole+0.157
ool+0.144
 tangled+0.137
 atop+0.135
 hardest+0.133
Neg logit contributions
 Fro-0.089
His-0.088
They-0.087
 His-0.086
He-0.081
Token be
Token ID307
Feature activation-4.76e+00

Pos logit contributions
 Fro+0.003
Un+0.003
 Max+0.003
 him+0.003
Jack+0.003
Neg logit contributions
ool-0.004
hole-0.004
 lid-0.004
els-0.004
ently-0.003
Tokeneping
Token ID7213
Feature activation-1.21e+01

Pos logit contributions
 lid+0.061
ool+0.060
hole+0.060
 button+0.056
 globe+0.054
Neg logit contributions
Un-0.051
They-0.048
David-0.048
Billy-0.048
Johnny-0.047
Token.
Token ID13
Feature activation+8.22e+00

Pos logit contributions
hole+0.161
 atop+0.146
ool+0.143
 tangled+0.142
 bulb+0.141
Neg logit contributions
 Fro-0.110
 His-0.110
His-0.109
They-0.105
Un-0.105
Token
Token ID198
Feature activation+6.03e+00

Pos logit contributions
Jack+0.045
 Fro+0.041
 His+0.041
Un+0.041
David+0.039
Neg logit contributions
hole-0.148
 pieces-0.147
ps-0.145
 falling-0.143
 atop-0.142
Token.
Token ID13
Feature activation+9.93e+00

Pos logit contributions
Un+0.018
Jack+0.018
 Fro+0.018
David+0.018
Jim+0.018
Neg logit contributions
hole-0.009
 pieces-0.009
 atop-0.008
 edges-0.008
ps-0.008
Token They
Token ID1119
Feature activation+2.34e+00

Pos logit contributions
David+0.101
oph+0.099
 concession+0.099
Jack+0.098
Johnny+0.095
Neg logit contributions
 on-0.133
 pieces-0.127
hole-0.120
 falling-0.119
 ladder-0.119
Token jumped
Token ID11687
Feature activation-6.48e+00

Pos logit contributions
Un+0.027
David+0.027
Jack+0.026
 concession+0.026
 Fro+0.026
Neg logit contributions
 pieces-0.027
hole-0.026
 on-0.026
ps-0.025
 atop-0.025
Token into
Token ID656
Feature activation-2.28e+00

Pos logit contributions
hole+0.067
 pieces+0.063
 atop+0.058
 edges+0.058
els+0.057
Neg logit contributions
 Fro-0.085
Jack-0.083
Un-0.081
 concession-0.080
Kids-0.078
Token the
Token ID262
Feature activation-7.45e-01

Pos logit contributions
 pieces+0.022
hole+0.021
 on+0.020
 edges+0.019
ps+0.019
Neg logit contributions
David-0.031
Un-0.031
Jack-0.031
 Fro-0.030
enny-0.030
Token pool
Token ID5933
Feature activation-1.07e+01

Pos logit contributions
hole+0.008
 pieces+0.008
 atop+0.007
ps+0.007
 falling+0.007
Neg logit contributions
Jack-0.009
 Fro-0.009
 concession-0.009
Un-0.009
Kids-0.009
Token and
Token ID290
Feature activation+9.78e-01

Pos logit contributions
hole+0.159
 atop+0.157
els+0.155
 tangled+0.151
 hardest+0.148
Neg logit contributions
 Fro-0.080
Jack-0.078
 concession-0.076
Un-0.076
They-0.074
Token threw
Token ID9617
Feature activation-7.40e+00

Pos logit contributions
 Fro+0.010
Jack+0.010
Un+0.010
Kids+0.010
 concession+0.010
Neg logit contributions
hole-0.013
 pieces-0.012
 atop-0.011
 edges-0.011
els-0.011
Token a
Token ID257
Feature activation-9.14e-01

Pos logit contributions
hole+0.098
 pieces+0.093
 atop+0.090
 edges+0.087
 falling+0.087
Neg logit contributions
 Fro-0.075
Un-0.074
Jack-0.074
David-0.071
Jim-0.070
Token ball
Token ID2613
Feature activation-1.12e+01

Pos logit contributions
hole+0.010
 atop+0.009
ps+0.009
els+0.009
 pieces+0.009
Neg logit contributions
Jack-0.011
 Fro-0.011
Un-0.011
His-0.011
 concession-0.010
Token for
Token ID329
Feature activation-2.63e+00

Pos logit contributions
hole+0.108
 pieces+0.102
 atop+0.098
 edges+0.096
 hardest+0.093
Neg logit contributions
 Fro-0.150
Un-0.144
Jack-0.143
His-0.141
 concession-0.138
Token dog
Token ID3290
Feature activation-6.80e+00

Pos logit contributions
 on+0.009
 hard+0.008
 pieces+0.008
 broken+0.008
 matches+0.007
Neg logit contributions
enny-0.010
oph-0.010
ete-0.010
Tony-0.009
ree-0.009
Token.
Token ID13
Feature activation+6.40e+00

Pos logit contributions
hole+0.080
 pieces+0.074
 atop+0.070
ps+0.070
 edges+0.070
Neg logit contributions
 Fro-0.085
Un-0.082
Jack-0.081
Kids-0.080
 His-0.078
Token Tim
Token ID5045
Feature activation-1.47e+00

Pos logit contributions
David+0.047
Un+0.047
Jack+0.045
 concession+0.045
oph+0.044
Neg logit contributions
 pieces-0.098
hole-0.097
 on-0.095
 edges-0.092
 falling-0.091
Token liked
Token ID8288
Feature activation-4.36e+00

Pos logit contributions
 pieces+0.019
hole+0.019
 on+0.018
 edges+0.017
 atop+0.017
Neg logit contributions
Un-0.015
David-0.015
Jack-0.015
 Fro-0.015
Jim-0.014
Token to
Token ID284
Feature activation-2.66e+00

Pos logit contributions
 pieces+0.053
hole+0.050
 on+0.050
 falling+0.048
 edges+0.048
Neg logit contributions
Jack-0.048
David-0.048
Un-0.047
 Fro-0.046
Jim-0.046
Token play
Token ID711
Feature activation-8.69e+00

Pos logit contributions
 pieces+0.033
 on+0.032
 trunk+0.032
 falling+0.032
ps+0.031
Neg logit contributions
David-0.028
Jack-0.028
Un-0.028
enny-0.027
oph-0.027
Token outside
Token ID2354
Feature activation-9.42e+00

Pos logit contributions
 pieces+0.074
hole+0.073
 on+0.067
 atop+0.067
 edges+0.065
Neg logit contributions
Un-0.126
David-0.126
Jack-0.125
 Fro-0.124
Jim-0.121
Token in
Token ID287
Feature activation-5.24e+00

Pos logit contributions
hole+0.093
 pieces+0.087
ps+0.082
 atop+0.081
 edges+0.080
Neg logit contributions
Jack-0.125
Un-0.124
 Fro-0.123
 concession-0.119
David-0.119
Token the
Token ID262
Feature activation-4.62e+00

Pos logit contributions
 pieces+0.045
 on+0.041
hole+0.040
 falling+0.037
 edges+0.037
Neg logit contributions
David-0.079
oph-0.079
Un-0.078
 Fro-0.078
enny-0.078
Token dirt
Token ID13647
Feature activation-1.50e+01

Pos logit contributions
 pieces+0.037
 on+0.036
 edges+0.033
hole+0.033
 atop+0.032
Neg logit contributions
David-0.071
 Fro-0.069
enny-0.069
Jack-0.068
Un-0.067
Token with
Token ID351
Feature activation-6.66e+00

Pos logit contributions
hole+0.189
 atop+0.180
els+0.175
ps+0.174
bones+0.174
Neg logit contributions
 Fro-0.135
They-0.132
Jack-0.130
 He-0.129
He-0.128
Token looked
Token ID3114
Feature activation-8.03e+00

Pos logit contributions
 pieces+0.020
 on+0.019
hole+0.018
 edges+0.018
 broken+0.018
Neg logit contributions
David-0.024
 concession-0.024
 Fro-0.023
Un-0.023
Kids-0.023
Token closely
Token ID7173
Feature activation-1.08e+01

Pos logit contributions
hole+0.060
 pieces+0.060
 on+0.054
 atop+0.054
 edges+0.052
Neg logit contributions
Un-0.125
David-0.124
Jack-0.124
 Fro-0.122
Kids-0.120
Token and
Token ID290
Feature activation-2.06e+00

Pos logit contributions
hole+0.139
 pieces+0.136
ps+0.132
 code+0.131
 falling+0.128
Neg logit contributions
Jack-0.107
Un-0.103
David-0.102
 Fro-0.101
Jim-0.099
Token saw
Token ID2497
Feature activation-6.06e+00

Pos logit contributions
hole+0.024
 pieces+0.024
 on+0.023
 atop+0.023
 falling+0.023
Neg logit contributions
Un-0.023
 Fro-0.022
David-0.022
 concession-0.022
Kids-0.022
Token that
Token ID326
Feature activation-4.41e+00

Pos logit contributions
 pieces+0.071
hole+0.071
 on+0.066
 atop+0.065
 edges+0.065
Neg logit contributions
Un-0.068
Jack-0.068
David-0.068
 Fro-0.067
Kids-0.065
Token it
Token ID340
Feature activation-9.61e+00

Pos logit contributions
hole+0.049
 pieces+0.046
ps+0.045
 atop+0.044
 edges+0.043
Neg logit contributions
Jack-0.054
 Fro-0.053
Un-0.053
David-0.053
Jim-0.051
Token was
Token ID373
Feature activation-7.60e+00

Pos logit contributions
hole+0.133
 pieces+0.128
 atop+0.123
 edges+0.121
ps+0.121
Neg logit contributions
Jack-0.091
Un-0.088
 Fro-0.088
David-0.084
Johnny-0.083
Token a
Token ID257
Feature activation-6.81e+00

Pos logit contributions
hole+0.092
 pieces+0.091
 atop+0.085
 edges+0.084
 on+0.083
Neg logit contributions
Un-0.083
Jack-0.083
David-0.082
 Fro-0.082
Jim-0.078
Token sm
Token ID895
Feature activation-1.01e+01

Pos logit contributions
hole+0.078
ps+0.073
 pieces+0.070
els+0.069
 atop+0.068
Neg logit contributions
Jack-0.086
Un-0.085
 Fro-0.085
 concession-0.082
David-0.081
Tokenelly
Token ID6148
Feature activation-1.24e+01

Pos logit contributions
hole+0.106
 pieces+0.102
 on+0.097
 edges+0.094
ool+0.094
Neg logit contributions
David-0.128
Un-0.124
 Fro-0.123
Harry-0.122
 concession-0.121
Token sausage
Token ID31889
Feature activation-1.72e+01

Pos logit contributions
hole+0.152
ps+0.139
cks+0.136
 atop+0.136
 pieces+0.132
Neg logit contributions
 Fro-0.149
 concession-0.132
Jack-0.132
 Mitt-0.129
Kids-0.129
Token He
Token ID679
Feature activation+1.36e+00

Pos logit contributions
David+0.051
Jack+0.049
Un+0.048
Da+0.045
oph+0.045
Neg logit contributions
 pieces-0.138
hole-0.134
 on-0.132
 edges-0.131
 trunk-0.129
Token wants
Token ID3382
Feature activation-4.23e+00

Pos logit contributions
David+0.019
Jack+0.019
Un+0.019
 Fro+0.018
Jim+0.018
Neg logit contributions
 pieces-0.012
hole-0.012
 on-0.011
 atop-0.011
 edges-0.011
Token to
Token ID284
Feature activation-1.27e+00

Pos logit contributions
hole+0.067
 atop+0.062
 pieces+0.059
 lid+0.058
ps+0.058
Neg logit contributions
 Fro-0.038
Un-0.036
Jack-0.035
David-0.033
Jim-0.033
Token hug
Token ID16225
Feature activation-7.36e+00

Pos logit contributions
 on+0.013
 trunk+0.013
 pieces+0.013
 falling+0.012
 ring+0.012
Neg logit contributions
David-0.017
enny-0.016
Jack-0.016
oph-0.016
 concession-0.016
Token the
Token ID262
Feature activation-1.25e+00

Pos logit contributions
hole+0.093
 pieces+0.089
 atop+0.085
 edges+0.083
ps+0.082
Neg logit contributions
Un-0.081
 Fro-0.080
Jack-0.079
David-0.077
Kids-0.075
Token dog
Token ID3290
Feature activation-8.74e+00

Pos logit contributions
hole+0.020
 atop+0.019
 pieces+0.019
ps+0.018
els+0.018
Neg logit contributions
Un-0.009
 Fro-0.009
Jack-0.009
David-0.009
 concession-0.008
Token.
Token ID13
Feature activation+7.03e+00

Pos logit contributions
hole+0.104
 pieces+0.096
 atop+0.096
 edges+0.095
 lid+0.092
Neg logit contributions
 Fro-0.111
 His-0.100
His-0.097
 Her-0.094
Un-0.093
Token He
Token ID679
Feature activation+7.19e-01

Pos logit contributions
David+0.047
Da+0.045
Un+0.045
 concession+0.044
Harry+0.044
Neg logit contributions
 pieces-0.118
hole-0.113
 on-0.112
 edges-0.111
 trunk-0.109
Token runs
Token ID4539
Feature activation-5.91e+00

Pos logit contributions
David+0.010
Jack+0.010
Un+0.010
Johnny+0.010
Da+0.010
Neg logit contributions
 pieces-0.006
hole-0.006
 on-0.006
ps-0.006
 edges-0.006
Token to
Token ID284
Feature activation-1.70e+00

Pos logit contributions
 pieces+0.062
hole+0.061
 on+0.059
 atop+0.056
 edges+0.056
Neg logit contributions
David-0.074
Un-0.074
Jack-0.073
 Fro-0.072
Jim-0.070
Token the
Token ID262
Feature activation-7.36e-01

Pos logit contributions
 pieces+0.017
hole+0.017
 on+0.016
 atop+0.016
 falling+0.016
Neg logit contributions
Un-0.022
Jack-0.022
David-0.022
 Fro-0.021
Kids-0.021