TOP ACTIVATIONS
MAX = 27.443

s so interesting here !" Her mother said
that are much swe eter !" <|endoftext|> Once there was a
longer until they cool down ." But Lily didn
My necklace is gone forever ." <|endoftext|> Once upon a time
t quite finished yet . So the bear reached out
in the red je ep !" His mom heard him and
he got even mad der ! He turned to Bob and
gonna be an exciting journey !" he exclaimed .
by one 1 , 2 ,
want a boss y cow !" Her mom said
himself Sometimes it â
said I kept my promise !
said don â t
this drain by cleaning it ." He ran inside and asked
most important people in life ." The bear was
stop making so much noise !" Mia says . " You
cream could taste so good !" Lily exclaimed .
a tum my ac he ." L ily
much fun playing with it ." L ily
to record more songs too ". Everyone laughed . Mrs Jones

MIN ACTIVATIONS
MIN = -23.495

real treasure chest filled with gold coins and jewels .
you will never touch my axe again ," his dad said
found a chest filled with gold and jewels . From that
treasure chest filled with gold coins and jewels .
treasure chest full of shiny gold coins ! Tim , Sally
prevent accident by not putting rocks in our mouths ."
should not have taken your sack . I was curious .
not have knocked down your tower . That was not nice
opened a treasure chest of gold coins .
could not keep filling the bucket . The rain was heavy
chest and found lots of gold coins and shiny jewels .
, do not touch the pins . They are not toys
he didn 't lose his wallet again . And Lily felt
chest full of shiny gold coins ! Tim , Sally ,
Maybe the treasure is not gold or jewels ," L ila
load your wagon with p ears near a wild bear .
she continued picking her p ears . With a smile on
Did you kids find my gold watch ?"
a treasure chest of gold coins . Every
they fought over the shiny gold . The puppy was getting

INTERVAL 14.708 - 27.443
CONTAINS 1.612%

their blanket back ! " Now that bunny won
You 'll find out soon ." Conf used ,
hurt himself . Tom was upset . He
'm bored of my toys ," Mia said . She felt
. The cat said , " What is that , Skip

INTERVAL 1.974 - 14.708
CONTAINS 35.304%

at the beach . <|endoftext|> L ily and Ben were twins
. <|endoftext|> Once upon a time , there was a little
ice cream . One day , Tim my went to the
you to the stars !" The little girl laughed and said
his red flag . One day , it started to rain

INTERVAL -10.761 - 1.974
CONTAINS 56.164%

trying to find out what other people were doing . One
upon a time , there was a brave knight named Max
the cloth as a scarf and wrapped it around their necks
pe b bles !" she says . " No
played a game with his friend . He wanted to win

INTERVAL -23.495 - -10.761
CONTAINS 6.919%

It 's how I became rich ." Lily didn 't understand
. She looked in the mirror and smiled .
. He landed on the ground and did not move .
dirt . It was a coin ! He put the coin
the beach and find sand dollars . On
Token
Token ID8151
Feature activation+1.28e+01

Pos logit contributions
His+0.185
Jack+0.183
They+0.173
 His+0.168
He+0.166
Neg logit contributions
hole-0.253
 globe-0.247
ps-0.243
 bulb-0.243
 pieces-0.241
Tokens
Token ID82
Feature activation+1.19e+01

Pos logit contributions
His+0.156
Jack+0.156
They+0.148
 His+0.145
He+0.143
Neg logit contributions
hole-0.178
 globe-0.173
 bulb-0.173
ps-0.172
 pieces-0.170
Token so
Token ID523
Feature activation+1.15e+01

Pos logit contributions
His+0.146
Jack+0.144
They+0.140
 His+0.138
He+0.136
Neg logit contributions
hole-0.166
 globe-0.164
ps-0.161
 bulb-0.158
 pieces-0.157
Token interesting
Token ID3499
Feature activation+3.07e+00

Pos logit contributions
His+0.162
Jack+0.162
They+0.156
 His+0.152
He+0.152
Neg logit contributions
hole-0.144
 globe-0.139
ps-0.137
 bulb-0.135
 pieces-0.133
Token here
Token ID994
Feature activation+4.41e+00

Pos logit contributions
Jack+0.041
His+0.041
They+0.041
 His+0.039
He+0.038
Neg logit contributions
hole-0.039
ps-0.038
 globe-0.038
 bulb-0.037
 pieces-0.036
Token!"
Token ID2474
Feature activation+1.99e+01

Pos logit contributions
Jack+0.057
His+0.057
They+0.056
 His+0.054
He+0.054
Neg logit contributions
hole-0.058
ps-0.057
 globe-0.057
 bulb-0.056
ool-0.055
Token
Token ID198
Feature activation+1.69e+01

Pos logit contributions
His+0.165
Jack+0.163
They+0.155
As+0.149
Un+0.148
Neg logit contributions
hole-0.311
 pieces-0.308
 trunk-0.307
 bulb-0.303
 ladder-0.297
Token
Token ID198
Feature activation+1.72e+01

Pos logit contributions
 concession+0.147
His+0.138
 His+0.136
 please+0.134
 As+0.133
Neg logit contributions
 trunk-0.228
 pieces-0.222
ool-0.221
hole-0.219
 tree-0.219
TokenHer
Token ID9360
Feature activation+1.12e+01

Pos logit contributions
 His+0.139
 As+0.129
 concession+0.120
 please+0.115
idents+0.115
Neg logit contributions
 trunk-0.236
 pieces-0.234
 globe-0.232
ool-0.228
hole-0.227
Token mother
Token ID2802
Feature activation+5.18e+00

Pos logit contributions
His+0.103
Jack+0.102
They+0.096
 His+0.093
He+0.091
Neg logit contributions
hole-0.193
 globe-0.188
ps-0.188
 bulb-0.188
 pieces-0.186
Token said
Token ID531
Feature activation+6.90e+00

Pos logit contributions
Jack+0.055
His+0.052
They+0.050
 His+0.049
He+0.048
Neg logit contributions
 globe-0.084
hole-0.083
 bulb-0.081
 edges-0.081
ool-0.080
Token that
Token ID326
Feature activation+8.96e+00

Pos logit contributions
They+0.020
His+0.020
Jack+0.020
He+0.019
 His+0.019
Neg logit contributions
hole-0.017
ps-0.017
 globe-0.016
ool-0.016
 bulb-0.016
Token are
Token ID389
Feature activation+6.29e+00

Pos logit contributions
Jack+0.082
His+0.078
They+0.077
 His+0.076
He+0.074
Neg logit contributions
hole-0.162
ps-0.157
ool-0.151
els-0.148
 globe-0.148
Token much
Token ID881
Feature activation+2.24e+00

Pos logit contributions
They+0.081
His+0.080
Jack+0.080
He+0.078
 His+0.074
Neg logit contributions
hole-0.091
 globe-0.085
ps-0.081
ool-0.079
ening-0.077
Token swe
Token ID3490
Feature activation+4.99e+00

Pos logit contributions
They+0.028
Jack+0.027
He+0.027
His+0.026
 please+0.025
Neg logit contributions
hole-0.032
 globe-0.031
ps-0.031
ool-0.029
els-0.028
Tokeneter
Token ID2357
Feature activation-2.28e+00

Pos logit contributions
His+0.073
Jack+0.070
He+0.066
Jim+0.064
They+0.064
Neg logit contributions
hole-0.064
 bulb-0.061
ool-0.061
els-0.060
 trunk-0.060
Token!"
Token ID2474
Feature activation+1.98e+01

Pos logit contributions
hole+0.027
 bulb+0.026
 globe+0.025
ool+0.025
 plank+0.025
Neg logit contributions
They-0.032
Jack-0.031
His-0.031
 His-0.031
He-0.030
Token<|endoftext|>
Token ID50256
Feature activation+1.14e+01

Pos logit contributions
His+0.162
Jack+0.159
Un+0.159
As+0.155
 concession+0.155
Neg logit contributions
 pieces-0.298
 trunk-0.289
 bulb-0.287
hole-0.283
 ladder-0.281
TokenOnce
Token ID7454
Feature activation+1.16e+01

Pos logit contributions
Jack+0.103
His+0.103
They+0.096
 His+0.093
He+0.092
Neg logit contributions
hole-0.198
ps-0.194
 bulb-0.194
 globe-0.193
 pieces-0.190
Token there
Token ID612
Feature activation+9.09e+00

Pos logit contributions
His+0.121
Jack+0.121
They+0.115
 His+0.110
He+0.108
Neg logit contributions
hole-0.182
 bulb-0.177
 globe-0.176
 pieces-0.175
ps-0.173
Token was
Token ID373
Feature activation+1.21e+01

Pos logit contributions
His+0.119
Jack+0.118
They+0.116
He+0.110
David+0.109
Neg logit contributions
hole-0.117
 bulb-0.115
 globe-0.115
 pieces-0.112
 trunk-0.110
Token a
Token ID257
Feature activation+8.19e+00

Pos logit contributions
His+0.183
David+0.168
Johnny+0.166
Jack+0.166
Un+0.163
Neg logit contributions
 pieces-0.136
 falling-0.134
 bulb-0.132
 hanging-0.131
 trunk-0.128
Token longer
Token ID2392
Feature activation+5.19e+00

Pos logit contributions
Jack+0.115
His+0.110
They+0.108
 His+0.106
He+0.105
Neg logit contributions
hole-0.152
ps-0.151
 globe-0.147
ool-0.146
els-0.142
Token until
Token ID1566
Feature activation+1.14e+01

Pos logit contributions
His+0.063
 His+0.062
They+0.062
Jack+0.061
He+0.060
Neg logit contributions
 globe-0.071
hole-0.069
ool-0.068
els-0.066
 bulb-0.066
Token they
Token ID484
Feature activation+8.25e+00

Pos logit contributions
His+0.117
Jack+0.117
 His+0.108
They+0.107
He+0.103
Neg logit contributions
hole-0.188
ps-0.182
ool-0.182
 globe-0.181
 bulb-0.177
Token cool
Token ID3608
Feature activation+1.72e+00

Pos logit contributions
Jack+0.089
His+0.088
They+0.084
 His+0.081
He+0.080
Neg logit contributions
hole-0.129
 globe-0.126
ps-0.124
 bulb-0.124
 pieces-0.124
Token down
Token ID866
Feature activation+2.98e+00

Pos logit contributions
Jack+0.028
His+0.028
They+0.028
 His+0.027
He+0.026
Neg logit contributions
hole-0.016
 globe-0.016
ool-0.016
 bulb-0.016
 trunk-0.015
Token."
Token ID526
Feature activation+1.97e+01

Pos logit contributions
They+0.031
Jack+0.031
His+0.030
 His+0.028
He+0.028
Neg logit contributions
ool-0.049
hole-0.047
 bulb-0.046
 globe-0.046
els-0.045
Token
Token ID198
Feature activation+1.69e+01

Pos logit contributions
His+0.174
Un+0.172
Jack+0.169
 concession+0.165
David+0.160
Neg logit contributions
 trunk-0.281
 bulb-0.277
hole-0.276
 globe-0.275
 pieces-0.273
Token
Token ID198
Feature activation+1.73e+01

Pos logit contributions
 concession+0.149
 please+0.147
His+0.142
Thanks+0.139
 As+0.138
Neg logit contributions
 globe-0.224
 trunk-0.222
 pieces-0.218
 bulb-0.218
ool-0.214
TokenBut
Token ID1537
Feature activation+1.20e+01

Pos logit contributions
 His+0.128
 As+0.123
elt+0.120
 please+0.112
 concession+0.112
Neg logit contributions
 globe-0.254
 bulb-0.248
 trunk-0.242
 pieces-0.241
cut-0.240
Token Lily
Token ID20037
Feature activation+6.21e+00

Pos logit contributions
His+0.124
Jack+0.123
They+0.117
 His+0.112
He+0.112
Neg logit contributions
hole-0.190
 bulb-0.187
 globe-0.186
ps-0.185
 pieces-0.184
Token didn
Token ID1422
Feature activation+7.89e+00

Pos logit contributions
His+0.068
Jack+0.067
They+0.064
 His+0.062
He+0.061
Neg logit contributions
hole-0.095
 bulb-0.093
ps-0.093
 globe-0.093
 pieces-0.092
TokenMy
Token ID3666
Feature activation+1.31e+01

Pos logit contributions
 globe+0.263
ps+0.257
 bulb+0.256
hole+0.256
ool+0.254
Neg logit contributions
His-0.155
They-0.152
Jack-0.150
He-0.146
 His-0.144
Token necklace
Token ID38987
Feature activation-1.62e+00

Pos logit contributions
His+0.167
Jack+0.165
They+0.159
 His+0.158
He+0.154
Neg logit contributions
hole-0.173
 globe-0.169
ps-0.168
 bulb-0.166
ool-0.164
Token is
Token ID318
Feature activation+7.21e+00

Pos logit contributions
ps+0.021
ening+0.020
hole+0.020
els+0.020
 bulb+0.020
Neg logit contributions
His-0.021
They-0.020
He-0.020
Jack-0.020
 His-0.019
Token gone
Token ID3750
Feature activation+3.34e-01

Pos logit contributions
They+0.063
 please+0.062
His+0.062
He+0.060
 His+0.057
Neg logit contributions
 globe-0.122
hole-0.118
els-0.117
ool-0.116
ps-0.115
Token forever
Token ID8097
Feature activation+2.68e+00

Pos logit contributions
They+0.004
Jack+0.004
He+0.004
 please+0.004
His+0.004
Neg logit contributions
els-0.005
ps-0.005
 globe-0.004
ool-0.004
hole-0.004
Token."
Token ID526
Feature activation+1.97e+01

Pos logit contributions
They+0.033
His+0.032
He+0.032
 His+0.031
Jack+0.031
Neg logit contributions
 globe-0.035
ps-0.035
els-0.035
ool-0.034
hole-0.034
Token<|endoftext|>
Token ID50256
Feature activation+1.18e+01

Pos logit contributions
Un+0.181
 concession+0.180
Jack+0.179
His+0.177
David+0.174
Neg logit contributions
 trunk-0.273
 pieces-0.266
hole-0.264
 bulb-0.264
 ladder-0.260
TokenOnce
Token ID7454
Feature activation+1.15e+01

Pos logit contributions
Jack+0.089
His+0.087
They+0.081
 His+0.078
He+0.077
Neg logit contributions
hole-0.223
ps-0.219
 bulb-0.219
 globe-0.217
ool-0.214
Token upon
Token ID2402
Feature activation+1.55e+01

Pos logit contributions
His+0.116
Jack+0.115
They+0.109
 His+0.105
He+0.103
Neg logit contributions
hole-0.186
 globe-0.181
 bulb-0.181
ps-0.180
 pieces-0.179
Token a
Token ID257
Feature activation+1.28e+01

Pos logit contributions
ete+0.262
Kids+0.262
enny+0.249
David+0.247
Brother+0.245
Neg logit contributions
 pieces-0.146
 glowing-0.126
 trunk-0.119
ps-0.119
 edges-0.117
Token time
Token ID640
Feature activation+9.93e+00

Pos logit contributions
His+0.254
 His+0.245
They+0.242
Jack+0.241
Johnny+0.241
Neg logit contributions
ool-0.072
 trunk-0.070
 pieces-0.069
 falling-0.066
 globe-0.064
Token
Token ID8151
Feature activation+2.18e+01

Pos logit contributions
His+0.072
They+0.069
Jack+0.068
Johnny+0.067
David+0.066
Neg logit contributions
 trunk-0.147
hole-0.144
 bulb-0.141
 pieces-0.141
 edges-0.140
Tokent
Token ID83
Feature activation+1.17e+01

Pos logit contributions
His+0.321
Jack+0.318
They+0.302
Un+0.295
He+0.292
Neg logit contributions
hole-0.251
 globe-0.232
ool-0.228
 bulb-0.225
 trunk-0.222
Token quite
Token ID2407
Feature activation+1.08e+01

Pos logit contributions
His+0.142
Jack+0.135
 His+0.135
He+0.134
They+0.133
Neg logit contributions
 globe-0.171
hole-0.164
 pieces-0.160
 lid-0.160
 bulb-0.160
Token finished
Token ID5201
Feature activation+3.86e+00

Pos logit contributions
His+0.120
 His+0.115
Jack+0.112
He+0.110
They+0.110
Neg logit contributions
 globe-0.169
hole-0.168
 bulb-0.163
ool-0.163
 pieces-0.160
Token yet
Token ID1865
Feature activation+5.06e+00

Pos logit contributions
His+0.048
Jack+0.047
They+0.047
 His+0.046
He+0.044
Neg logit contributions
hole-0.054
 globe-0.052
ool-0.052
 bulb-0.050
ps-0.050
Token.
Token ID13
Feature activation+1.97e+01

Pos logit contributions
His+0.068
Jack+0.067
They+0.067
 His+0.066
He+0.064
Neg logit contributions
hole-0.065
ps-0.063
 globe-0.063
ool-0.061
 bulb-0.060
Token So
Token ID1406
Feature activation+1.27e+01

Pos logit contributions
Jack+0.142
His+0.141
They+0.134
David+0.118
He+0.118
Neg logit contributions
hole-0.357
 trunk-0.350
 globe-0.347
 bulb-0.346
 pieces-0.339
Token the
Token ID262
Feature activation+1.10e+01

Pos logit contributions
His+0.143
Jack+0.142
They+0.135
 His+0.130
He+0.129
Neg logit contributions
hole-0.190
 globe-0.185
 bulb-0.185
ps-0.184
 pieces-0.183
Token bear
Token ID6842
Feature activation+1.97e+00

Pos logit contributions
Jack+0.153
His+0.146
Johnny+0.138
They+0.138
 His+0.137
Neg logit contributions
hole-0.153
ps-0.138
ool-0.135
 globe-0.133
ening-0.133
Token reached
Token ID4251
Feature activation+4.45e+00

Pos logit contributions
His+0.023
Jack+0.023
They+0.022
 His+0.022
He+0.021
Neg logit contributions
hole-0.029
 globe-0.029
 bulb-0.028
ps-0.028
ool-0.028
Token out
Token ID503
Feature activation+2.41e+00

Pos logit contributions
His+0.059
 His+0.058
Jack+0.056
They+0.055
He+0.054
Neg logit contributions
 globe-0.058
hole-0.058
ool-0.057
 bulb-0.056
els-0.052
Token in
Token ID287
Feature activation+1.28e+01

Pos logit contributions
Jack+0.048
His+0.047
They+0.044
He+0.039
David+0.039
Neg logit contributions
hole-0.157
 bulb-0.155
 globe-0.155
 pieces-0.154
 trunk-0.154
Token the
Token ID262
Feature activation+1.15e+01

Pos logit contributions
His+0.143
Jack+0.142
They+0.135
 His+0.132
He+0.130
Neg logit contributions
hole-0.194
 globe-0.189
 bulb-0.189
ps-0.188
 pieces-0.185
Token red
Token ID2266
Feature activation+4.12e+00

Pos logit contributions
Jack+0.173
His+0.169
He+0.166
They+0.166
 His+0.163
Neg logit contributions
hole-0.127
ps-0.122
ool-0.118
els-0.116
 globe-0.113
Token je
Token ID11223
Feature activation+2.93e+00

Pos logit contributions
His+0.067
 His+0.065
Jack+0.065
They+0.064
He+0.062
Neg logit contributions
ps-0.040
hole-0.039
ool-0.035
 falling-0.035
 globe-0.035
Tokenep
Token ID538
Feature activation-3.06e-01

Pos logit contributions
His+0.053
Jack+0.051
He+0.050
They+0.049
Johnny+0.048
Neg logit contributions
 bulb-0.025
ps-0.025
ening-0.024
ool-0.024
 globe-0.024
Token!"
Token ID2474
Feature activation+1.96e+01

Pos logit contributions
 bulb+0.004
 globe+0.004
ool+0.004
 bulbs+0.004
 plank+0.004
Neg logit contributions
His-0.004
They-0.004
 His-0.004
He-0.004
As-0.004
Token His
Token ID2399
Feature activation+1.18e+01

Pos logit contributions
Jack+0.159
His+0.155
Un+0.152
 concession+0.151
Johnny+0.151
Neg logit contributions
 pieces-0.298
 trunk-0.293
 ladder-0.286
hole-0.282
 bulb-0.280
Token mom
Token ID1995
Feature activation+6.44e+00

Pos logit contributions
Jack+0.125
His+0.120
They+0.117
As+0.111
He+0.108
Neg logit contributions
 trunk-0.184
ps-0.183
 pieces-0.183
 bulb-0.182
hole-0.182
Token heard
Token ID2982
Feature activation+6.10e+00

Pos logit contributions
His+0.071
Jack+0.070
They+0.068
 His+0.067
He+0.065
Neg logit contributions
 globe-0.100
hole-0.098
 bulb-0.096
ool-0.096
 plank-0.095
Token him
Token ID683
Feature activation+3.26e+00

Pos logit contributions
Jack+0.052
His+0.051
They+0.048
As+0.048
Thanks+0.046
Neg logit contributions
ps-0.104
 pieces-0.103
 bulb-0.098
ening-0.098
hole-0.097
Token and
Token ID290
Feature activation+9.50e+00

Pos logit contributions
His+0.044
Jack+0.042
They+0.042
 His+0.041
He+0.040
Neg logit contributions
hole-0.042
 globe-0.042
 bulb-0.041
 pieces-0.041
ool-0.041
Token he
Token ID339
Feature activation+1.35e+01

Pos logit contributions
His+0.001
Jack+0.001
They+0.001
 His+0.001
He+0.001
Neg logit contributions
hole-0.003
 globe-0.003
 bulb-0.003
ps-0.003
 pieces-0.003
Token got
Token ID1392
Feature activation+7.21e+00

Pos logit contributions
His+0.150
Jack+0.148
They+0.143
 His+0.141
He+0.140
Neg logit contributions
hole-0.205
 globe-0.202
 bulb-0.199
ool-0.196
ps-0.195
Token even
Token ID772
Feature activation+7.27e+00

Pos logit contributions
His+0.082
Jack+0.078
He+0.077
They+0.077
 His+0.077
Neg logit contributions
 globe-0.120
hole-0.115
ool-0.114
 bulb-0.108
ening-0.102
Token mad
Token ID8805
Feature activation+2.51e+00

Pos logit contributions
His+0.093
Jack+0.092
They+0.089
He+0.087
 His+0.085
Neg logit contributions
 globe-0.100
hole-0.099
ool-0.095
 bulb-0.095
ps-0.094
Tokender
Token ID1082
Feature activation+1.14e+01

Pos logit contributions
His+0.039
Jack+0.039
They+0.038
 His+0.037
He+0.037
Neg logit contributions
hole-0.027
 globe-0.026
 bulb-0.026
ps-0.025
 pieces-0.025
Token!
Token ID0
Feature activation+1.96e+01

Pos logit contributions
His+0.170
Jack+0.169
They+0.166
He+0.162
 His+0.158
Neg logit contributions
 globe-0.132
hole-0.132
 bulb-0.130
ool-0.125
ps-0.124
Token He
Token ID679
Feature activation+1.06e+01

Pos logit contributions
Jack+0.127
They+0.126
His+0.125
As+0.112
Un+0.110
Neg logit contributions
hole-0.360
 trunk-0.357
 globe-0.353
 bulb-0.350
 pieces-0.348
Token turned
Token ID2900
Feature activation+1.51e+00

Pos logit contributions
His+0.128
Jack+0.128
They+0.121
 His+0.119
He+0.117
Neg logit contributions
hole-0.151
 globe-0.147
 bulb-0.146
ps-0.145
 pieces-0.144
Token to
Token ID284
Feature activation+8.74e+00

Pos logit contributions
His+0.023
Jack+0.023
He+0.022
 His+0.021
They+0.021
Neg logit contributions
 globe-0.019
hole-0.019
ool-0.018
 bulb-0.018
 plank-0.017
Token Bob
Token ID5811
Feature activation+2.30e+00

Pos logit contributions
His+0.106
Jack+0.106
They+0.104
As+0.100
He+0.098
Neg logit contributions
ps-0.120
 pieces-0.116
 bulb-0.115
 trunk-0.114
 globe-0.113
Token and
Token ID290
Feature activation+9.80e+00

Pos logit contributions
His+0.029
Jack+0.028
 His+0.028
They+0.027
He+0.027
Neg logit contributions
ool-0.031
hole-0.031
 globe-0.031
 bulb-0.031
 pieces-0.031
Token gonna
Token ID8066
Feature activation+1.21e+01

Pos logit contributions
 His+0.121
They+0.120
His+0.119
Jack+0.114
He+0.114
Neg logit contributions
hole-0.150
 globe-0.149
 atop-0.141
ps-0.141
ool-0.140
Token be
Token ID307
Feature activation+6.05e+00

Pos logit contributions
His+0.142
Jack+0.141
They+0.134
 His+0.130
He+0.129
Neg logit contributions
hole-0.174
 bulb-0.170
 globe-0.170
ps-0.169
 pieces-0.167
Token an
Token ID281
Feature activation+5.91e+00

Pos logit contributions
Jack+0.067
They+0.066
His+0.065
He+0.061
As+0.061
Neg logit contributions
 globe-0.101
hole-0.093
els-0.087
ool-0.087
 lid-0.087
Token exciting
Token ID7895
Feature activation+1.19e+00

Pos logit contributions
 please+0.079
Jack+0.072
His+0.069
He+0.069
 His+0.068
Neg logit contributions
 globe-0.079
ps-0.077
cks-0.074
ls-0.074
 atop-0.071
Token journey
Token ID7002
Feature activation-3.52e+00

Pos logit contributions
Jack+0.016
 please+0.015
His+0.015
They+0.015
 His+0.014
Neg logit contributions
ps-0.015
hole-0.015
els-0.014
 atop-0.013
 wil-0.013
Token!"
Token ID2474
Feature activation+1.96e+01

Pos logit contributions
ps+0.047
 bulb+0.046
els+0.046
 globe+0.045
ool+0.043
Neg logit contributions
They-0.045
Jack-0.044
His-0.043
He-0.042
 please-0.041
Token he
Token ID339
Feature activation+1.02e+01

Pos logit contributions
His+0.165
Jack+0.162
Un+0.159
They+0.157
As+0.155
Neg logit contributions
 pieces-0.299
hole-0.293
 trunk-0.293
 bulb-0.284
 ladder-0.284
Token exclaimed
Token ID33552
Feature activation+7.20e+00

Pos logit contributions
Jack+0.129
His+0.128
They+0.123
He+0.118
 His+0.116
Neg logit contributions
hole-0.134
ps-0.133
 bulb-0.132
 pieces-0.130
 trunk-0.129
Token.
Token ID13
Feature activation+1.79e+01

Pos logit contributions
His+0.120
Jack+0.119
They+0.115
 His+0.113
He+0.112
Neg logit contributions
hole-0.069
 globe-0.067
 bulb-0.066
ps-0.066
 pieces-0.065
Token
Token ID198
Feature activation+1.60e+01

Pos logit contributions
Jack+0.148
They+0.144
His+0.141
As+0.134
Un+0.133
Neg logit contributions
 trunk-0.280
 globe-0.277
hole-0.272
 pieces-0.270
 ladder-0.268
Token
Token ID198
Feature activation+1.67e+01

Pos logit contributions
 concession+0.133
 please+0.125
 His+0.118
Un+0.118
Jack+0.118
Neg logit contributions
 trunk-0.221
 pieces-0.219
 globe-0.216
ool-0.214
hole-0.213
Token by
Token ID416
Feature activation+1.23e+01

Pos logit contributions
His+0.049
Jack+0.047
David+0.044
 His+0.043
Un+0.042
Neg logit contributions
hole-0.105
 globe-0.103
 pieces-0.101
 bulb-0.101
ool-0.101
Token one
Token ID530
Feature activation+8.14e+00

Pos logit contributions
His+0.142
Jack+0.140
They+0.139
 His+0.132
He+0.130
Neg logit contributions
hole-0.181
ool-0.177
 globe-0.174
 bulb-0.171
 pieces-0.169
Token
Token ID6184
Feature activation+8.65e+00

Pos logit contributions
His+0.095
He+0.095
They+0.094
 please+0.092
Jack+0.092
Neg logit contributions
ps-0.113
 globe-0.110
hole-0.107
ool-0.105
els-0.102
Token
Token ID95
Feature activation+9.59e+00

Pos logit contributions
His+0.105
Jack+0.104
They+0.099
 His+0.098
He+0.095
Neg logit contributions
hole-0.123
 pieces-0.121
ps-0.120
 bulb-0.120
 globe-0.120
Token
Token ID26391
Feature activation+1.55e+01

Pos logit contributions
Jack+0.102
His+0.102
They+0.096
 His+0.093
He+0.092
Neg logit contributions
hole-0.150
 globe-0.145
ps-0.145
 bulb-0.144
ool-0.142
Token
Token ID447
Feature activation+1.96e+01

Pos logit contributions
His+0.130
Jack+0.124
They+0.116
 His+0.110
He+0.108
Neg logit contributions
hole-0.275
 globe-0.268
 trunk-0.263
 ladder-0.254
 pieces-0.253
Token
Token ID250
Feature activation+1.66e+01

Pos logit contributions
His+0.217
Jack+0.213
They+0.204
 His+0.201
He+0.192
Neg logit contributions
hole-0.289
 globe-0.288
 bulb-0.279
 trunk-0.276
 pieces-0.274
Token 1
Token ID352
Feature activation+2.01e+00

Pos logit contributions
His+0.121
Jack+0.117
 His+0.114
They+0.107
As+0.097
Neg logit contributions
hole-0.304
 pieces-0.295
 globe-0.294
 bulb-0.293
ps-0.289
Token,
Token ID11
Feature activation+1.01e+01

Pos logit contributions
They+0.027
His+0.027
He+0.026
Jack+0.026
 His+0.026
Neg logit contributions
 bulb-0.026
 globe-0.025
ps-0.025
hole-0.024
 edges-0.024
Token 2
Token ID362
Feature activation+2.18e+00

Pos logit contributions
Jack+0.104
His+0.103
They+0.101
He+0.097
 His+0.094
Neg logit contributions
hole-0.160
 globe-0.159
ps-0.158
 bulb-0.157
ool-0.154
Token,
Token ID11
Feature activation+9.53e+00

Pos logit contributions
They+0.029
Jack+0.028
His+0.028
He+0.027
As+0.026
Neg logit contributions
 bulb-0.030
 globe-0.030
hole-0.029
ps-0.029
ool-0.027
Token want
Token ID765
Feature activation+6.61e+00

Pos logit contributions
His+0.125
Jack+0.124
They+0.117
 His+0.114
He+0.112
Neg logit contributions
hole-0.180
 globe-0.175
ps-0.174
 bulb-0.174
 pieces-0.172
Token a
Token ID257
Feature activation+7.01e+00

Pos logit contributions
His+0.062
Jack+0.062
They+0.058
 His+0.058
He+0.055
Neg logit contributions
hole-0.113
 globe-0.110
 bulb-0.109
ps-0.108
ool-0.108
Token boss
Token ID6478
Feature activation-1.10e+00

Pos logit contributions
Jack+0.088
His+0.088
 please+0.085
 His+0.083
Sure+0.080
Neg logit contributions
hole-0.102
ps-0.101
els-0.099
ening-0.093
 atop-0.091
Tokeny
Token ID88
Feature activation+3.56e+00

Pos logit contributions
hole+0.011
 globe+0.011
ps+0.011
 bulb+0.011
 pieces+0.011
Neg logit contributions
Jack-0.018
His-0.018
They-0.017
 His-0.017
As-0.016
Token cow
Token ID9875
Feature activation-3.63e+00

Pos logit contributions
His+0.048
Jack+0.048
They+0.045
 His+0.045
He+0.044
Neg logit contributions
hole-0.046
ps-0.044
 globe-0.044
 bulb-0.043
 pieces-0.043
Token!"
Token ID2474
Feature activation+1.96e+01

Pos logit contributions
hole+0.040
 globe+0.039
ps+0.038
 bulb+0.038
ool+0.038
Neg logit contributions
His-0.055
Jack-0.055
They-0.053
 His-0.052
He-0.051
Token
Token ID198
Feature activation+1.60e+01

Pos logit contributions
Jack+0.161
His+0.154
As+0.152
 concession+0.151
They+0.148
Neg logit contributions
 trunk-0.299
 pieces-0.296
 bulb-0.293
 ladder-0.290
hole-0.289
Token
Token ID198
Feature activation+1.66e+01

Pos logit contributions
 concession+0.154
 please+0.140
elt+0.135
latable+0.135
Thanks+0.134
Neg logit contributions
 trunk-0.205
ool-0.200
 globe-0.198
 pieces-0.197
hole-0.196
TokenHer
Token ID9360
Feature activation+1.14e+01

Pos logit contributions
 His+0.130
 As+0.124
elt+0.123
 concession+0.120
 Fro+0.116
Neg logit contributions
 globe-0.232
 trunk-0.231
 bulb-0.225
 pieces-0.225
cut-0.222
Token mom
Token ID1995
Feature activation+5.33e+00

Pos logit contributions
His+0.110
Jack+0.110
They+0.103
 His+0.099
He+0.098
Neg logit contributions
hole-0.189
 bulb-0.186
 globe-0.185
ps-0.184
 pieces-0.183
Token said
Token ID531
Feature activation+7.30e+00

Pos logit contributions
Jack+0.062
His+0.061
They+0.059
 His+0.057
He+0.056
Neg logit contributions
hole-0.079
 globe-0.078
 bulb-0.077
ool-0.076
 pieces-0.076
Token himself
Token ID2241
Feature activation+1.13e+01

Pos logit contributions
His+0.010
Jack+0.010
They+0.010
 His+0.009
He+0.009
Neg logit contributions
hole-0.014
 globe-0.014
ps-0.014
 bulb-0.014
 pieces-0.014
Token
Token ID6184
Feature activation+9.64e+00

Pos logit contributions
His+0.145
Jack+0.142
 His+0.138
They+0.137
He+0.136
Neg logit contributions
hole-0.150
 globe-0.148
ool-0.146
 bulb-0.146
ps-0.144
Token
Token ID95
Feature activation+1.00e+01

Pos logit contributions
Jack+0.118
His+0.116
They+0.113
He+0.107
As+0.106
Neg logit contributions
hole-0.133
 globe-0.128
 bulb-0.127
ool-0.126
ps-0.126
Token
Token ID26391
Feature activation+1.60e+01

Pos logit contributions
Jack+0.110
His+0.107
They+0.102
 His+0.097
He+0.096
Neg logit contributions
hole-0.154
 globe-0.148
ps-0.147
ool-0.145
 bulb-0.145
Token
Token ID129
Feature activation+1.07e+01

Pos logit contributions
His+0.137
Jack+0.134
They+0.123
 His+0.118
He+0.115
Neg logit contributions
hole-0.276
 globe-0.268
 trunk-0.264
 ladder-0.254
ool-0.251
Token
Token ID241
Feature activation+1.96e+01

Pos logit contributions
His+0.123
Jack+0.122
They+0.116
 His+0.113
He+0.112
Neg logit contributions
hole-0.158
 globe-0.154
 bulb-0.154
ps-0.153
 pieces-0.152
TokenSometimes
Token ID15468
Feature activation+7.52e+00

Pos logit contributions
Jack+0.153
His+0.150
 His+0.141
David+0.130
ete+0.130
Neg logit contributions
hole-0.329
 globe-0.322
 edges-0.316
 pieces-0.315
 trunk-0.310
Token it
Token ID340
Feature activation+2.73e+00

Pos logit contributions
His+0.101
Jack+0.101
They+0.098
 His+0.095
He+0.094
Neg logit contributions
hole-0.096
 globe-0.094
ps-0.093
 bulb-0.092
ool-0.090
Tokenâ
Token ID22940
Feature activation+9.93e+00

Pos logit contributions
They+0.034
His+0.034
Jack+0.034
 His+0.033
He+0.033
Neg logit contributions
hole-0.037
 globe-0.037
ps-0.036
 bulb-0.035
ool-0.035
Token
Token ID26391
Feature activation+1.30e+01

Pos logit contributions
Jack+0.076
His+0.075
They+0.069
 His+0.069
He+0.065
Neg logit contributions
 bulb-0.188
hole-0.186
 globe-0.186
 pieces-0.186
ps-0.181
Token
Token ID8151
Feature activation+8.55e+00

Pos logit contributions
His+0.144
Jack+0.138
They+0.129
 His+0.128
He+0.126
Neg logit contributions
hole-0.199
 globe-0.194
 trunk-0.188
 pieces-0.187
ool-0.186
Token said
Token ID531
Feature activation+1.07e+01

Pos logit contributions
 globe+0.120
hole+0.119
ool+0.119
ps+0.118
 bulb+0.118
Neg logit contributions
His-0.115
 His-0.114
They-0.112
He-0.112
Jack-0.112
Token
Token ID6184
Feature activation+1.02e+01

Pos logit contributions
Jack+0.161
His+0.159
They+0.154
He+0.150
As+0.149
Neg logit contributions
hole-0.116
ps-0.115
 bulb-0.113
 globe-0.112
 pieces-0.110
Token
Token ID95
Feature activation+9.36e+00

Pos logit contributions
His+0.127
Jack+0.127
They+0.121
 His+0.117
He+0.117
Neg logit contributions
hole-0.140
 globe-0.137
 bulb-0.136
ps-0.136
ool-0.133
Token
Token ID26391
Feature activation+1.58e+01

Pos logit contributions
His+0.104
Jack+0.104
They+0.099
 His+0.096
He+0.094
Neg logit contributions
hole-0.142
 globe-0.138
ps-0.137
 bulb-0.137
 pieces-0.135
Token
Token ID129
Feature activation+1.06e+01

Pos logit contributions
His+0.140
Jack+0.136
They+0.122
He+0.119
 His+0.118
Neg logit contributions
hole-0.268
 globe-0.264
 trunk-0.258
 ladder-0.250
ool-0.246
Token
Token ID241
Feature activation+1.96e+01

Pos logit contributions
His+0.126
Jack+0.126
They+0.119
 His+0.116
He+0.115
Neg logit contributions
hole-0.153
 globe-0.149
 bulb-0.149
ps-0.149
 pieces-0.147
TokenI
Token ID40
Feature activation+1.33e+01

Pos logit contributions
His+0.147
Jack+0.144
 His+0.135
They+0.121
Johnny+0.118
Neg logit contributions
hole-0.342
 globe-0.334
 pieces-0.325
 edges-0.320
 trunk-0.318
Token kept
Token ID4030
Feature activation+5.31e+00

Pos logit contributions
Jack+0.148
His+0.147
They+0.139
 His+0.135
He+0.131
Neg logit contributions
hole-0.197
 bulb-0.192
 globe-0.191
ool-0.190
ps-0.190
Token my
Token ID616
Feature activation+7.13e+00

Pos logit contributions
His+0.072
Jack+0.071
They+0.069
 His+0.067
Johnny+0.066
Neg logit contributions
hole-0.069
 globe-0.068
ool-0.067
 bulb-0.065
 pieces-0.063
Token promise
Token ID6991
Feature activation+2.45e-01

Pos logit contributions
His+0.096
 His+0.095
Jack+0.094
They+0.089
He+0.089
Neg logit contributions
hole-0.090
 globe-0.086
ool-0.084
ps-0.083
 bulb-0.082
Token!
Token ID0
Feature activation+1.62e+01

Pos logit contributions
His+0.004
Jack+0.003
They+0.003
 His+0.003
He+0.003
Neg logit contributions
hole-0.003
 globe-0.003
 bulb-0.003
ps-0.003
 pieces-0.003
Token said
Token ID531
Feature activation+1.07e+01

Pos logit contributions
els+0.218
ool+0.217
ps+0.216
 globe+0.216
hole+0.216
Neg logit contributions
His-0.124
They-0.123
Jack-0.121
He-0.119
 His-0.116
Token
Token ID6184
Feature activation+1.02e+01

Pos logit contributions
Jack+0.161
His+0.159
They+0.154
He+0.150
As+0.149
Neg logit contributions
hole-0.116
ps-0.115
 bulb-0.113
 globe-0.112
 pieces-0.110
Token
Token ID95
Feature activation+9.36e+00

Pos logit contributions
His+0.127
Jack+0.127
They+0.121
 His+0.117
He+0.117
Neg logit contributions
hole-0.140
 globe-0.137
 bulb-0.136
ps-0.136
ool-0.133
Token
Token ID26391
Feature activation+1.58e+01

Pos logit contributions
His+0.104
Jack+0.104
They+0.099
 His+0.096
He+0.094
Neg logit contributions
hole-0.142
 globe-0.138
ps-0.137
 bulb-0.137
 pieces-0.135
Token
Token ID129
Feature activation+1.06e+01

Pos logit contributions
His+0.140
Jack+0.136
They+0.122
He+0.119
 His+0.118
Neg logit contributions
hole-0.268
 globe-0.264
 trunk-0.258
 ladder-0.250
ool-0.246
Token
Token ID241
Feature activation+1.96e+01

Pos logit contributions
His+0.126
Jack+0.126
They+0.119
 His+0.116
He+0.115
Neg logit contributions
hole-0.153
 globe-0.149
 bulb-0.149
ps-0.149
 pieces-0.147
Tokendon
Token ID9099
Feature activation+1.18e+01

Pos logit contributions
His+0.147
Jack+0.144
 His+0.135
They+0.121
Johnny+0.118
Neg logit contributions
hole-0.342
 globe-0.334
 pieces-0.325
 edges-0.320
 trunk-0.318
Tokenâ
Token ID22940
Feature activation+1.14e+01

Pos logit contributions
Jack+0.159
His+0.154
Jim+0.145
Harry+0.143
Luke+0.142
Neg logit contributions
ool-0.138
hole-0.136
 ladder-0.133
 button-0.132
 trunk-0.130
Token
Token ID26391
Feature activation+1.52e+01

Pos logit contributions
His+0.095
Jack+0.093
They+0.086
 His+0.086
He+0.082
Neg logit contributions
hole-0.206
 bulb-0.204
 pieces-0.204
 globe-0.203
ps-0.201
Token
Token ID8151
Feature activation+1.04e+01

Pos logit contributions
His+0.155
Jack+0.149
They+0.137
 His+0.135
He+0.134
Neg logit contributions
hole-0.240
 globe-0.236
 trunk-0.231
 ladder-0.226
ool-0.222
Tokent
Token ID83
Feature activation+9.30e+00

Pos logit contributions
His+0.127
Jack+0.127
They+0.121
 His+0.118
He+0.116
Neg logit contributions
hole-0.146
 globe-0.142
 bulb-0.142
ps-0.141
 pieces-0.140
Token this
Token ID428
Feature activation+4.65e+00

Pos logit contributions
Jack+0.043
His+0.043
They+0.043
 His+0.040
He+0.040
Neg logit contributions
hole-0.043
 globe-0.042
ool-0.041
 trunk-0.041
ps-0.041
Token drain
Token ID14782
Feature activation-4.82e+00

Pos logit contributions
Jack+0.059
They+0.059
 please+0.058
His+0.057
He+0.056
Neg logit contributions
ps-0.060
hole-0.059
 globe-0.056
els-0.054
cks-0.052
Token by
Token ID416
Feature activation+7.76e+00

Pos logit contributions
hole+0.062
 globe+0.062
ool+0.060
ps+0.060
 trunk+0.059
Neg logit contributions
They-0.062
Jack-0.060
His-0.058
He-0.057
 please-0.056
Token cleaning
Token ID12724
Feature activation-2.70e+00

Pos logit contributions
They+0.090
His+0.086
Jack+0.084
As+0.080
Her+0.079
Neg logit contributions
 globe-0.119
hole-0.117
ool-0.110
 atop-0.106
 lid-0.106
Token it
Token ID340
Feature activation-1.23e+00

Pos logit contributions
 globe+0.036
hole+0.035
 trunk+0.035
ool+0.035
ps+0.034
Neg logit contributions
They-0.034
Jack-0.034
His-0.032
Johnny-0.031
He-0.030
Token."
Token ID526
Feature activation+1.96e+01

Pos logit contributions
 globe+0.018
hole+0.017
els+0.016
bones+0.016
ps+0.016
Neg logit contributions
They-0.015
 please-0.014
His-0.014
As-0.014
He-0.014
Token He
Token ID679
Feature activation+1.07e+01

Pos logit contributions
 concession+0.174
Un+0.173
His+0.172
Jack+0.172
David+0.170
Neg logit contributions
 trunk-0.272
 pieces-0.266
 bulb-0.263
hole-0.259
 ladder-0.257
Token ran
Token ID4966
Feature activation+3.38e+00

Pos logit contributions
His+0.140
Jack+0.140
They+0.134
 His+0.130
He+0.129
Neg logit contributions
hole-0.141
 globe-0.138
 bulb-0.136
ps-0.135
 pieces-0.133
Token inside
Token ID2641
Feature activation+2.92e+00

Pos logit contributions
Jack+0.048
His+0.048
They+0.046
 His+0.045
He+0.045
Neg logit contributions
hole-0.040
ps-0.039
 pieces-0.039
 bulb-0.038
 globe-0.038
Token and
Token ID290
Feature activation+9.68e+00

Pos logit contributions
His+0.035
Jack+0.035
They+0.033
 His+0.032
He+0.032
Neg logit contributions
hole-0.042
 bulb-0.040
 globe-0.040
ps-0.040
 pieces-0.040
Token asked
Token ID1965
Feature activation+8.15e+00

Pos logit contributions
Jack+0.126
His+0.125
They+0.118
 His+0.116
He+0.114
Neg logit contributions
hole-0.131
 globe-0.129
ps-0.128
 bulb-0.127
ool-0.126
Token most
Token ID749
Feature activation+8.14e+00

Pos logit contributions
Jack+0.213
His+0.206
They+0.205
He+0.202
 His+0.197
Neg logit contributions
hole-0.159
ps-0.146
ool-0.144
els-0.140
ening-0.140
Token important
Token ID1593
Feature activation+6.32e+00

Pos logit contributions
Jack+0.109
His+0.103
They+0.103
He+0.101
As+0.098
Neg logit contributions
hole-0.109
ps-0.108
 globe-0.106
 bulb-0.101
els-0.101
Token people
Token ID661
Feature activation+1.64e+00

Pos logit contributions
Jack+0.098
His+0.098
They+0.095
 His+0.095
He+0.093
Neg logit contributions
hole-0.067
ps-0.065
els-0.061
 globe-0.061
 bulb-0.060
Token in
Token ID287
Feature activation+8.82e+00

Pos logit contributions
They+0.019
Jack+0.018
His+0.017
He+0.017
 His+0.016
Neg logit contributions
ool-0.025
hole-0.025
 lid-0.024
 globe-0.024
 edges-0.024
Token life
Token ID1204
Feature activation-1.10e+00

Pos logit contributions
Jack+0.096
They+0.093
His+0.091
 His+0.089
He+0.085
Neg logit contributions
hole-0.136
ool-0.132
 globe-0.131
ps-0.128
 bulb-0.127
Token."
Token ID526
Feature activation+1.96e+01

Pos logit contributions
ool+0.014
ps+0.014
hole+0.014
 globe+0.013
 bulb+0.013
Neg logit contributions
They-0.015
Jack-0.015
His-0.014
He-0.014
 His-0.014
Token
Token ID198
Feature activation+1.65e+01

Pos logit contributions
His+0.164
Jack+0.161
Un+0.156
They+0.151
 concession+0.151
Neg logit contributions
hole-0.295
 trunk-0.295
 pieces-0.291
 globe-0.285
 bulb-0.285
Token
Token ID198
Feature activation+1.68e+01

Pos logit contributions
 please+0.140
His+0.140
 concession+0.134
 His+0.133
Jack+0.132
Neg logit contributions
 pieces-0.230
 trunk-0.229
 globe-0.227
hole-0.224
ool-0.223
TokenThe
Token ID464
Feature activation+1.03e+01

Pos logit contributions
 His+0.124
 please+0.112
 As+0.112
elt+0.109
 concession+0.102
Neg logit contributions
 globe-0.252
 pieces-0.250
 trunk-0.247
 bulb-0.245
ool-0.242
Token bear
Token ID6842
Feature activation+1.44e+00

Pos logit contributions
Jack+0.125
His+0.119
He+0.116
They+0.115
 His+0.110
Neg logit contributions
hole-0.160
ps-0.143
ool-0.143
ening-0.139
els-0.138
Token was
Token ID373
Feature activation+8.03e+00

Pos logit contributions
His+0.016
Jack+0.015
 His+0.015
They+0.015
He+0.014
Neg logit contributions
hole-0.023
 globe-0.022
 bulb-0.022
ps-0.022
ool-0.022
Token stop
Token ID2245
Feature activation+7.98e+00

Pos logit contributions
hole+0.086
ening+0.082
 atop+0.080
ool+0.079
 plank+0.076
Neg logit contributions
 His-0.072
His-0.067
 Her-0.067
Her-0.064
 please-0.064
Token making
Token ID1642
Feature activation+3.27e+00

Pos logit contributions
Jack+0.115
His+0.115
They+0.112
 His+0.108
He+0.107
Neg logit contributions
hole-0.095
 globe-0.092
 bulb-0.091
ool-0.091
ps-0.089
Token so
Token ID523
Feature activation+9.17e+00

Pos logit contributions
Jack+0.038
His+0.037
They+0.036
 His+0.034
He+0.034
Neg logit contributions
hole-0.049
 globe-0.048
ool-0.047
 bulb-0.046
 trunk-0.046
Token much
Token ID881
Feature activation+3.99e+00

Pos logit contributions
They+0.122
Jack+0.121
His+0.121
He+0.115
 His+0.111
Neg logit contributions
hole-0.128
 globe-0.122
ool-0.118
 bulb-0.118
ps-0.115
Token noise
Token ID7838
Feature activation-2.64e+00

Pos logit contributions
They+0.053
Jack+0.051
He+0.050
His+0.048
 His+0.047
Neg logit contributions
hole-0.055
 globe-0.053
ool-0.051
ps-0.051
 trunk-0.050
Token!"
Token ID2474
Feature activation+1.95e+01

Pos logit contributions
ool+0.037
hole+0.036
 globe+0.035
 trunk+0.035
 stem+0.034
Neg logit contributions
They-0.036
He-0.032
Jack-0.032
His-0.032
As-0.030
Token Mia
Token ID32189
Feature activation+1.02e+01

Pos logit contributions
Jack+0.155
His+0.155
As+0.146
They+0.146
Un+0.142
Neg logit contributions
 pieces-0.308
 trunk-0.306
hole-0.303
 bulb-0.302
ps-0.299
Token says
Token ID1139
Feature activation+8.06e+00

Pos logit contributions
His+0.125
Jack+0.125
They+0.120
 His+0.117
He+0.115
Neg logit contributions
hole-0.143
 globe-0.140
 bulb-0.138
ps-0.137
 pieces-0.136
Token.
Token ID13
Feature activation+1.81e+01

Pos logit contributions
His+0.113
 His+0.109
They+0.107
Jack+0.106
Her+0.103
Neg logit contributions
hole-0.105
 globe-0.103
ool-0.102
 plank-0.102
 pieces-0.099
Token "
Token ID366
Feature activation+1.52e+01

Pos logit contributions
Jack+0.146
His+0.139
Johnny+0.128
David+0.126
They+0.124
Neg logit contributions
 trunk-0.310
 bulb-0.307
 globe-0.304
hole-0.303
ps-0.301
TokenYou
Token ID1639
Feature activation+8.37e+00

Pos logit contributions
ete+0.152
 His+0.146
enny+0.134
 Fro+0.133
Johnny+0.128
Neg logit contributions
 globe-0.231
 edges-0.222
 bulb-0.215
 pieces-0.212
 object-0.211
Tokencream
Token ID36277
Feature activation+8.99e+00

Pos logit contributions
hole+0.026
 globe+0.025
 bulb+0.025
ps+0.024
ool+0.022
Neg logit contributions
His-0.077
Jack-0.074
 His-0.074
They-0.073
He-0.072
Token could
Token ID714
Feature activation+1.32e+01

Pos logit contributions
His+0.100
Jack+0.099
They+0.095
 His+0.093
He+0.091
Neg logit contributions
hole-0.135
ps-0.132
 globe-0.131
 bulb-0.131
ool-0.129
Token taste
Token ID6938
Feature activation+7.91e-01

Pos logit contributions
Jack+0.171
His+0.165
They+0.157
Jim+0.155
David+0.154
Neg logit contributions
 bulb-0.178
hole-0.175
ps-0.173
 pieces-0.172
 globe-0.170
Token so
Token ID523
Feature activation+1.01e+01

Pos logit contributions
They+0.009
Jack+0.009
His+0.009
He+0.009
 His+0.009
Neg logit contributions
hole-0.012
 globe-0.012
 trunk-0.011
ps-0.011
ool-0.011
Token good
Token ID922
Feature activation+2.25e+00

Pos logit contributions
His+0.130
Jack+0.129
They+0.125
 His+0.121
He+0.120
Neg logit contributions
hole-0.143
 globe-0.137
ps-0.134
 bulb-0.133
 pieces-0.130
Token!"
Token ID2474
Feature activation+1.95e+01

Pos logit contributions
Jack+0.030
His+0.029
They+0.029
 His+0.028
He+0.028
Neg logit contributions
hole-0.030
 globe-0.029
 bulb-0.029
ool-0.028
 trunk-0.028
Token Lily
Token ID20037
Feature activation+9.75e+00

Pos logit contributions
His+0.169
Jack+0.168
Un+0.167
David+0.164
 concession+0.164
Neg logit contributions
 pieces-0.284
 trunk-0.277
 bulb-0.274
 ladder-0.267
hole-0.264
Token exclaimed
Token ID33552
Feature activation+7.59e+00

Pos logit contributions
His+0.136
Jack+0.134
Un+0.126
David+0.126
They+0.125
Neg logit contributions
hole-0.114
 pieces-0.114
ps-0.112
 bulb-0.112
 trunk-0.109
Token.
Token ID13
Feature activation+1.80e+01

Pos logit contributions
Jack+0.129
His+0.129
They+0.125
 His+0.122
He+0.122
Neg logit contributions
hole-0.069
 bulb-0.067
ps-0.066
 globe-0.066
 pieces-0.065
Token
Token ID198
Feature activation+1.58e+01

Pos logit contributions
His+0.147
Jack+0.146
They+0.134
He+0.131
David+0.129
Neg logit contributions
 trunk-0.291
hole-0.289
 globe-0.288
 bulb-0.286
 pieces-0.281
Token
Token ID198
Feature activation+1.66e+01

Pos logit contributions
 please+0.136
 concession+0.136
His+0.131
 His+0.127
Jack+0.126
Neg logit contributions
 pieces-0.217
hole-0.217
 trunk-0.215
ool-0.215
 globe-0.214
Token a
Token ID257
Feature activation+1.07e+01

Pos logit contributions
 globe+0.131
hole+0.130
 bulb+0.127
ool+0.124
ening+0.121
Neg logit contributions
His-0.092
Jack-0.091
They-0.089
 His-0.085
 please-0.085
Token tum
Token ID11814
Feature activation+3.96e+00

Pos logit contributions
Jack+0.150
His+0.142
He+0.139
 please+0.137
They+0.135
Neg logit contributions
ps-0.143
els-0.139
hole-0.139
ening-0.124
ool-0.123
Tokenmy
Token ID1820
Feature activation+4.34e+00

Pos logit contributions
His+0.054
Jack+0.053
They+0.051
 His+0.050
He+0.050
Neg logit contributions
hole-0.051
 globe-0.050
 bulb-0.049
ps-0.049
 pieces-0.049
Token ac
Token ID936
Feature activation+3.29e+00

Pos logit contributions
His+0.055
He+0.054
 His+0.053
As+0.051
They+0.051
Neg logit contributions
 globe-0.061
ps-0.057
els-0.056
hole-0.056
ool-0.056
Tokenhe
Token ID258
Feature activation+4.59e+00

Pos logit contributions
His+0.053
Jack+0.052
He+0.051
They+0.050
 His+0.049
Neg logit contributions
hole-0.034
ps-0.034
 bulb-0.033
 globe-0.033
 pieces-0.032
Token."
Token ID526
Feature activation+1.95e+01

Pos logit contributions
His+0.064
Jack+0.062
 His+0.060
They+0.059
He+0.059
Neg logit contributions
hole-0.059
ps-0.056
 globe-0.055
 bulb-0.055
ool-0.054
Token
Token ID220
Feature activation+1.47e+01

Pos logit contributions
Jack+0.165
They+0.164
 concession+0.164
His+0.161
David+0.161
Neg logit contributions
 trunk-0.289
 bulb-0.278
hole-0.276
 pieces-0.276
 ladder-0.273
Token
Token ID198
Feature activation+1.67e+01

Pos logit contributions
Jack+0.120
His+0.118
They+0.107
David+0.103
Un+0.102
Neg logit contributions
hole-0.253
ool-0.242
 trunk-0.242
 pieces-0.241
 globe-0.239
Token
Token ID198
Feature activation+1.72e+01

Pos logit contributions
 concession+0.132
 please+0.126
His+0.122
David+0.120
They+0.119
Neg logit contributions
 trunk-0.245
hole-0.241
 pieces-0.241
ool-0.239
 globe-0.239
TokenL
Token ID43
Feature activation+1.33e+01

Pos logit contributions
 His+0.119
 concession+0.117
 Fro+0.113
 As+0.113
 please+0.109
Neg logit contributions
 trunk-0.254
 globe-0.250
 bulb-0.246
 pieces-0.245
hole-0.242
Tokenily
Token ID813
Feature activation+7.46e+00

Pos logit contributions
His+0.170
Jack+0.168
They+0.163
 His+0.159
He+0.156
Neg logit contributions
hole-0.181
ps-0.176
 bulb-0.176
 globe-0.174
 pieces-0.173
Token much
Token ID881
Feature activation+8.85e+00

Pos logit contributions
Jack+0.209
His+0.209
They+0.198
 His+0.191
He+0.188
Neg logit contributions
hole-0.248
 bulb-0.245
 globe-0.242
ps-0.241
 pieces-0.240
Token fun
Token ID1257
Feature activation+4.50e+00

Pos logit contributions
Jack+0.092
They+0.087
His+0.085
He+0.084
 His+0.079
Neg logit contributions
hole-0.150
ps-0.149
 globe-0.144
els-0.142
ool-0.138
Token playing
Token ID2712
Feature activation+2.88e+00

Pos logit contributions
Jack+0.064
They+0.062
His+0.062
He+0.058
 His+0.058
Neg logit contributions
hole-0.056
ps-0.053
 bulb-0.052
 globe-0.052
ool-0.052
Token with
Token ID351
Feature activation+6.99e+00

Pos logit contributions
Jack+0.041
His+0.041
They+0.039
 His+0.038
He+0.038
Neg logit contributions
hole-0.035
 globe-0.033
 bulb-0.033
ps-0.033
ool-0.033
Token it
Token ID340
Feature activation-9.47e-01

Pos logit contributions
His+0.074
Jack+0.073
They+0.069
 His+0.068
He+0.067
Neg logit contributions
hole-0.109
 bulb-0.107
ps-0.107
 globe-0.107
 pieces-0.106
Token."
Token ID526
Feature activation+1.95e+01

Pos logit contributions
hole+0.014
 globe+0.013
ps+0.013
els+0.013
ool+0.012
Neg logit contributions
Jack-0.012
They-0.011
His-0.011
 His-0.011
As-0.011
Token
Token ID220
Feature activation+1.44e+01

Pos logit contributions
Un+0.178
His+0.177
Jack+0.175
David+0.171
 concession+0.168
Neg logit contributions
 trunk-0.273
 pieces-0.262
 bulb-0.260
 globe-0.257
hole-0.257
Token
Token ID198
Feature activation+1.69e+01

Pos logit contributions
His+0.117
Jack+0.114
They+0.098
Un+0.098
David+0.097
Neg logit contributions
hole-0.251
ps-0.240
 globe-0.240
 pieces-0.239
 trunk-0.238
Token
Token ID198
Feature activation+1.73e+01

Pos logit contributions
 concession+0.137
 please+0.136
His+0.132
Un+0.129
David+0.128
Neg logit contributions
 globe-0.233
 pieces-0.232
 trunk-0.232
 bulb-0.226
hole-0.224
TokenL
Token ID43
Feature activation+1.31e+01

Pos logit contributions
 His+0.124
 As+0.112
 please+0.110
 concession+0.108
elt+0.107
Neg logit contributions
 globe-0.253
 bulb-0.249
 pieces-0.248
 trunk-0.247
cut-0.244
Tokenily
Token ID813
Feature activation+6.95e+00

Pos logit contributions
His+0.170
Jack+0.168
They+0.161
 His+0.159
He+0.156
Neg logit contributions
hole-0.176
ps-0.172
 bulb-0.171
 globe-0.170
 pieces-0.168
Token to
Token ID284
Feature activation+1.46e+01

Pos logit contributions
ps+0.058
els+0.054
hole+0.051
 atop+0.049
 shines+0.048
Neg logit contributions
Jack-0.069
His-0.068
He-0.064
They-0.063
As-0.062
Token record
Token ID1700
Feature activation+1.19e+00

Pos logit contributions
Jack+0.182
His+0.181
They+0.171
He+0.169
As+0.166
Neg logit contributions
 bulb-0.200
ps-0.195
 trunk-0.192
 pieces-0.191
hole-0.189
Token more
Token ID517
Feature activation+4.14e+00

Pos logit contributions
His+0.013
Jack+0.013
They+0.013
 His+0.012
He+0.012
Neg logit contributions
hole-0.018
 globe-0.018
ool-0.018
 bulb-0.017
ps-0.017
Token songs
Token ID7259
Feature activation-1.25e-01

Pos logit contributions
They+0.055
His+0.052
Jack+0.052
He+0.051
 His+0.050
Neg logit contributions
hole-0.055
 globe-0.055
ool-0.052
ps-0.051
 trunk-0.050
Token too
Token ID1165
Feature activation+6.74e+00

Pos logit contributions
ool+0.002
hole+0.002
 globe+0.002
 plank+0.002
 bulb+0.002
Neg logit contributions
They-0.002
His-0.001
Jack-0.001
He-0.001
 His-0.001
Token".
Token ID1911
Feature activation+1.94e+01

Pos logit contributions
They+0.085
Jack+0.076
He+0.075
As+0.071
 He+0.070
Neg logit contributions
 globe-0.102
els-0.101
ps-0.101
hole-0.100
ool-0.098
Token Everyone
Token ID11075
Feature activation+7.11e+00

Pos logit contributions
His+0.153
Jack+0.151
They+0.147
Un+0.142
As+0.140
Neg logit contributions
hole-0.303
 trunk-0.302
 pieces-0.298
 globe-0.293
 bulb-0.289
Token laughed
Token ID13818
Feature activation+2.14e+00

Pos logit contributions
His+0.082
Jack+0.082
They+0.079
 His+0.076
He+0.075
Neg logit contributions
hole-0.107
ool-0.104
 globe-0.104
 bulb-0.102
ps-0.101
Token.
Token ID13
Feature activation+1.68e+01

Pos logit contributions
His+0.029
Jack+0.029
They+0.028
 His+0.027
He+0.027
Neg logit contributions
hole-0.027
ps-0.026
 globe-0.026
 bulb-0.026
 pieces-0.025
Token Mrs
Token ID9074
Feature activation+9.68e+00

Pos logit contributions
His+0.120
Jack+0.117
They+0.105
He+0.100
As+0.097
Neg logit contributions
hole-0.312
 globe-0.307
 bulb-0.304
 pieces-0.303
ps-0.302
Token Jones
Token ID5437
Feature activation+1.70e+00

Pos logit contributions
Jack+0.148
They+0.146
His+0.145
 His+0.141
He+0.139
Neg logit contributions
hole-0.105
ool-0.101
 bulb-0.100
 pieces-0.099
ps-0.098
Token real
Token ID1103
Feature activation-1.23e+01

Pos logit contributions
ps+0.091
els+0.091
hole+0.082
ening+0.082
cks+0.077
Neg logit contributions
Jack-0.092
His-0.092
He-0.091
 please-0.088
As-0.084
Token treasure
Token ID14068
Feature activation-2.04e+01

Pos logit contributions
els+0.168
ps+0.168
hole+0.165
cks+0.157
ool+0.156
Neg logit contributions
His-0.154
He-0.148
 His-0.146
Jack-0.139
 He-0.139
Token chest
Token ID7721
Feature activation-1.96e+01

Pos logit contributions
ening+0.258
 blades+0.242
ps+0.241
els+0.237
ool+0.229
Neg logit contributions
They-0.253
He-0.241
 He-0.230
His-0.228
 please-0.224
Token filled
Token ID5901
Feature activation-1.30e+01

Pos logit contributions
 blades+0.307
els+0.301
ps+0.292
ening+0.288
 prope+0.280
Neg logit contributions
They-0.192
He-0.169
 please-0.167
 he-0.162
 He-0.161
Token with
Token ID351
Feature activation-1.01e+01

Pos logit contributions
hole+0.184
ps+0.177
 bulb+0.175
ool+0.172
 globe+0.172
Neg logit contributions
Jack-0.163
His-0.159
They-0.158
As-0.146
He-0.146
Token gold
Token ID3869
Feature activation-2.35e+01

Pos logit contributions
ool+0.123
hole+0.120
 globe+0.112
 bulb+0.108
 ladder+0.108
Neg logit contributions
They-0.139
He-0.138
 him-0.136
His-0.136
 His-0.135
Token coins
Token ID10796
Feature activation-2.28e+01

Pos logit contributions
 angle+0.279
 falling+0.266
 bulb+0.263
ening+0.259
ool+0.257
Neg logit contributions
He-0.301
They-0.298
 He-0.284
His-0.262
 His-0.253
Token and
Token ID290
Feature activation-2.65e+00

Pos logit contributions
ool+0.337
 bulb+0.337
 blades+0.331
 bolt+0.326
ps+0.324
Neg logit contributions
They-0.233
He-0.212
 he-0.187
 He-0.186
His-0.183
Token jewels
Token ID42701
Feature activation-1.88e+01

Pos logit contributions
hole+0.021
 globe+0.020
ool+0.020
ps+0.020
 bulb+0.019
Neg logit contributions
His-0.049
Jack-0.048
They-0.047
 His-0.047
He-0.047
Token.
Token ID13
Feature activation+8.99e+00

Pos logit contributions
ool+0.275
els+0.268
 angle+0.268
 blades+0.267
ening+0.263
Neg logit contributions
They-0.202
He-0.187
 He-0.166
 he-0.165
His-0.162
Token
Token ID220
Feature activation+2.95e+00

Pos logit contributions
David+0.132
His+0.128
Un+0.127
Jack+0.126
Jim+0.123
Neg logit contributions
 trunk-0.108
 pieces-0.105
 box-0.103
hole-0.101
 lid-0.101
Token you
Token ID345
Feature activation-5.49e+00

Pos logit contributions
hole+0.088
 globe+0.086
 bulb+0.085
ps+0.085
 pieces+0.084
Neg logit contributions
His-0.063
Jack-0.063
They-0.060
 His-0.058
He-0.057
Token will
Token ID481
Feature activation-4.67e+00

Pos logit contributions
hole+0.091
 globe+0.090
 bulb+0.090
ps+0.090
ool+0.088
Neg logit contributions
His-0.053
Jack-0.052
They-0.051
 His-0.048
He-0.048
Token never
Token ID1239
Feature activation-4.72e+00

Pos logit contributions
hole+0.076
 globe+0.075
 pieces+0.074
ps+0.073
 bulb+0.073
Neg logit contributions
His-0.049
They-0.047
Jack-0.047
He-0.044
 His-0.044
Token touch
Token ID3638
Feature activation-1.33e+01

Pos logit contributions
 globe+0.074
hole+0.073
ps+0.071
 bulb+0.070
 pieces+0.070
Neg logit contributions
His-0.052
Jack-0.051
They-0.049
 His-0.048
He-0.048
Token my
Token ID616
Feature activation-7.80e+00

Pos logit contributions
ool+0.258
hole+0.250
 globe+0.247
ps+0.235
 lid+0.232
Neg logit contributions
 him-0.103
His-0.083
 His-0.083
 He-0.082
Jack-0.082
Token axe
Token ID23272
Feature activation-2.33e+01

Pos logit contributions
ool+0.090
ps+0.089
aro+0.089
bones+0.082
iest+0.081
Neg logit contributions
He-0.108
 He-0.107
 His-0.106
 him-0.104
His-0.103
Token again
Token ID757
Feature activation-1.06e+01

Pos logit contributions
ps+0.370
 globe+0.368
ool+0.360
els+0.359
hole+0.353
Neg logit contributions
His-0.230
He-0.225
 please-0.221
They-0.218
Jack-0.216
Token,"
Token ID553
Feature activation+7.72e+00

Pos logit contributions
 globe+0.124
ps+0.120
 bulb+0.117
hole+0.117
ool+0.115
Neg logit contributions
They-0.162
His-0.159
Jack-0.158
He-0.154
 His-0.149
Token his
Token ID465
Feature activation-2.19e+00

Pos logit contributions
As+0.087
Jack+0.087
latable+0.085
enny+0.084
 concession+0.083
Neg logit contributions
 pieces-0.100
 ladder-0.098
 ring-0.097
 broken-0.096
 tangled-0.095
Token dad
Token ID9955
Feature activation-6.55e+00

Pos logit contributions
hole+0.045
 globe+0.045
 bulb+0.045
ps+0.044
ool+0.044
Neg logit contributions
His-0.013
Jack-0.013
They-0.011
 His-0.011
He-0.011
Token said
Token ID531
Feature activation-2.35e+00

Pos logit contributions
hole+0.097
 globe+0.097
 bulb+0.096
ps+0.094
ool+0.092
Neg logit contributions
His-0.077
Jack-0.075
They-0.072
 His-0.072
He-0.069
Token found
Token ID1043
Feature activation-5.68e+00

Pos logit contributions
Jack+0.001
His+0.001
They+0.001
 His+0.001
He+0.001
Neg logit contributions
ool-0.002
hole-0.002
ps-0.002
 bulb-0.002
 globe-0.002
Token a
Token ID257
Feature activation-4.63e+00

Pos logit contributions
ool+0.095
 bulb+0.093
 globe+0.092
 falling+0.091
hole+0.091
Neg logit contributions
His-0.055
 His-0.054
He-0.053
Jack-0.051
They-0.051
Token chest
Token ID7721
Feature activation-1.77e+01

Pos logit contributions
ps+0.061
els+0.060
ening+0.053
ool+0.053
hole+0.052
Neg logit contributions
Jack-0.069
He-0.068
His-0.067
They-0.066
As-0.065
Token filled
Token ID5901
Feature activation-1.23e+01

Pos logit contributions
ps+0.254
els+0.251
 blades+0.241
ool+0.230
 bulb+0.229
Neg logit contributions
He-0.216
His-0.214
They-0.205
Jack-0.198
 He-0.190
Token with
Token ID351
Feature activation-9.45e+00

Pos logit contributions
hole+0.185
ps+0.185
 bulb+0.178
ool+0.176
 falling+0.172
Neg logit contributions
Jack-0.146
His-0.143
They-0.139
He-0.130
As-0.128
Token gold
Token ID3869
Feature activation-2.29e+01

Pos logit contributions
ool+0.123
hole+0.115
 globe+0.113
 bulb+0.112
ps+0.111
Neg logit contributions
He-0.128
They-0.126
His-0.126
 His-0.124
Jack-0.124
Token and
Token ID290
Feature activation-2.78e+00

Pos logit contributions
 wil+0.277
 angle+0.276
ps+0.273
 falling+0.273
 bulb+0.272
Neg logit contributions
He-0.294
They-0.280
 He-0.273
His-0.245
 he-0.240
Token jewels
Token ID42701
Feature activation-1.93e+01

Pos logit contributions
ps+0.026
hole+0.026
ool+0.026
 globe+0.024
els+0.024
Neg logit contributions
His-0.047
Jack-0.046
They-0.046
He-0.046
 His-0.045
Token.
Token ID13
Feature activation+8.06e+00

Pos logit contributions
ool+0.274
els+0.269
ps+0.267
 bulb+0.261
 blades+0.257
Neg logit contributions
He-0.229
They-0.222
His-0.206
 He-0.204
 he-0.194
Token From
Token ID3574
Feature activation-1.82e+00

Pos logit contributions
Jack+0.056
His+0.056
They+0.052
Her+0.046
David+0.046
Neg logit contributions
hole-0.155
 globe-0.151
 pieces-0.150
 trunk-0.150
 bulb-0.147
Token that
Token ID326
Feature activation+2.01e+00

Pos logit contributions
hole+0.032
ps+0.032
ool+0.032
 bulb+0.032
 globe+0.031
Neg logit contributions
His-0.016
Jack-0.016
They-0.015
He-0.014
 His-0.014
Token treasure
Token ID14068
Feature activation-2.04e+01

Pos logit contributions
els+0.168
ps+0.168
hole+0.165
cks+0.157
ool+0.156
Neg logit contributions
His-0.154
He-0.148
 His-0.146
Jack-0.139
 He-0.139
Token chest
Token ID7721
Feature activation-1.96e+01

Pos logit contributions
ening+0.258
 blades+0.242
ps+0.241
els+0.237
ool+0.229
Neg logit contributions
They-0.253
He-0.241
 He-0.230
His-0.228
 please-0.224
Token filled
Token ID5901
Feature activation-1.30e+01

Pos logit contributions
 blades+0.307
els+0.301
ps+0.292
ening+0.288
 prope+0.280
Neg logit contributions
They-0.192
He-0.169
 please-0.167
 he-0.162
 He-0.161
Token with
Token ID351
Feature activation-1.01e+01

Pos logit contributions
hole+0.184
ps+0.177
 bulb+0.175
ool+0.172
 globe+0.172
Neg logit contributions
Jack-0.163
His-0.159
They-0.158
As-0.146
He-0.146
Token gold
Token ID3869
Feature activation-2.35e+01

Pos logit contributions
ool+0.123
hole+0.120
 globe+0.112
 bulb+0.108
 ladder+0.108
Neg logit contributions
They-0.139
He-0.138
 him-0.136
His-0.136
 His-0.135
Token coins
Token ID10796
Feature activation-2.28e+01

Pos logit contributions
 angle+0.279
 falling+0.266
 bulb+0.263
ening+0.259
ool+0.257
Neg logit contributions
He-0.301
They-0.298
 He-0.284
His-0.262
 His-0.253
Token and
Token ID290
Feature activation-2.65e+00

Pos logit contributions
ool+0.337
 bulb+0.337
 blades+0.331
 bolt+0.326
ps+0.324
Neg logit contributions
They-0.233
He-0.212
 he-0.187
 He-0.186
His-0.183
Token jewels
Token ID42701
Feature activation-1.88e+01

Pos logit contributions
hole+0.021
 globe+0.020
ool+0.020
ps+0.020
 bulb+0.019
Neg logit contributions
His-0.049
Jack-0.048
They-0.047
 His-0.047
He-0.047
Token.
Token ID13
Feature activation+8.99e+00

Pos logit contributions
ool+0.275
els+0.268
 angle+0.268
 blades+0.267
ening+0.263
Neg logit contributions
They-0.202
He-0.187
 He-0.166
 he-0.165
His-0.162
Token
Token ID220
Feature activation+2.95e+00

Pos logit contributions
David+0.132
His+0.128
Un+0.127
Jack+0.126
Jim+0.123
Neg logit contributions
 trunk-0.108
 pieces-0.105
 box-0.103
hole-0.101
 lid-0.101
Token
Token ID198
Feature activation+7.09e+00

Pos logit contributions
Jack+0.025
His+0.025
David+0.023
Un+0.022
Johnny+0.022
Neg logit contributions
hole-0.050
 globe-0.048
 pieces-0.048
 trunk-0.048
ps-0.047
Token treasure
Token ID14068
Feature activation-1.76e+01

Pos logit contributions
ps+0.055
hole+0.054
els+0.053
ening+0.050
ool+0.048
Neg logit contributions
Jack-0.061
His-0.060
They-0.057
He-0.057
 His-0.056
Token chest
Token ID7721
Feature activation-1.68e+01

Pos logit contributions
ps+0.199
els+0.188
ening+0.187
 angles+0.183
 blades+0.182
Neg logit contributions
Jack-0.256
His-0.253
He-0.252
They-0.250
 His-0.245
Token full
Token ID1336
Feature activation-9.71e+00

Pos logit contributions
ps+0.263
els+0.259
 blades+0.248
 bulb+0.240
 bulbs+0.238
Neg logit contributions
His-0.183
Jack-0.182
They-0.178
He-0.176
 His-0.169
Token of
Token ID286
Feature activation-6.24e+00

Pos logit contributions
ps+0.197
 globe+0.185
els+0.184
 blades+0.183
 plank+0.182
Neg logit contributions
 His-0.066
Jack-0.065
His-0.063
They-0.059
He-0.059
Token shiny
Token ID22441
Feature activation-1.70e+01

Pos logit contributions
hole+0.072
ps+0.071
ool+0.071
et+0.064
ening+0.064
Neg logit contributions
Jack-0.091
 His-0.090
He-0.089
His-0.088
 He-0.087
Token gold
Token ID3869
Feature activation-2.27e+01

Pos logit contributions
ps+0.181
 wil+0.172
ool+0.167
 falling+0.166
hole+0.165
Neg logit contributions
 He-0.244
He-0.234
 His-0.231
His-0.225
 him-0.223
Token coins
Token ID10796
Feature activation-2.21e+01

Pos logit contributions
 wil+0.315
 angle+0.312
 angles+0.309
ps+0.306
 bulb+0.297
Neg logit contributions
 He-0.242
He-0.237
They-0.232
 he-0.210
 His-0.207
Token!
Token ID0
Feature activation+8.26e+00

Pos logit contributions
 bulb+0.340
ps+0.336
 blades+0.328
 prope+0.322
ool+0.321
Neg logit contributions
They-0.221
He-0.205
 he-0.194
 He-0.193
Jack-0.191
Token Tim
Token ID5045
Feature activation-2.36e+00

Pos logit contributions
His+0.077
Jack+0.070
Her+0.070
Luke+0.069
They+0.068
Neg logit contributions
 ladder-0.132
 trunk-0.131
 pieces-0.130
 chim-0.127
ool-0.126
Token,
Token ID11
Feature activation+2.20e+00

Pos logit contributions
hole+0.035
 bulb+0.035
ool+0.035
ps+0.035
 globe+0.035
Neg logit contributions
Jack-0.027
His-0.027
 His-0.025
They-0.025
He-0.025
Token Sally
Token ID25737
Feature activation-2.74e+00

Pos logit contributions
Jack+0.016
His+0.015
They+0.014
 His+0.014
He+0.013
Neg logit contributions
hole-0.043
ps-0.042
 globe-0.042
 bulb-0.042
ool-0.042
Token prevent
Token ID2948
Feature activation-9.15e+00

Pos logit contributions
hole+0.012
 bulb+0.011
 globe+0.011
ps+0.011
 pieces+0.011
Neg logit contributions
His-0.010
Jack-0.010
They-0.009
 His-0.009
He-0.009
Token accident
Token ID5778
Feature activation-1.17e+01

Pos logit contributions
 globe+0.165
hole+0.163
ool+0.162
 bulb+0.151
 trunk+0.148
Neg logit contributions
They-0.079
Jack-0.078
 him-0.074
His-0.070
 His-0.070
Token by
Token ID416
Feature activation-3.95e+00

Pos logit contributions
 globe+0.147
hole+0.134
 bulb+0.133
 trunk+0.132
els+0.128
Neg logit contributions
They-0.169
Jack-0.155
He-0.155
As-0.155
His-0.154
Token not
Token ID407
Feature activation-3.51e+00

Pos logit contributions
hole+0.058
 globe+0.057
ool+0.056
 code+0.052
 atop+0.051
Neg logit contributions
They-0.047
 His-0.044
His-0.043
 Sammy-0.042
Jack-0.042
Token putting
Token ID5137
Feature activation-1.43e+01

Pos logit contributions
 globe+0.063
hole+0.059
ool+0.058
 lid+0.056
els+0.056
Neg logit contributions
They-0.035
His-0.034
 please-0.034
He-0.033
As-0.032
Token rocks
Token ID12586
Feature activation-2.26e+01

Pos logit contributions
hole+0.249
ool+0.246
 globe+0.241
 angle+0.223
 code+0.222
Neg logit contributions
 him-0.140
They-0.116
Jack-0.115
 his-0.108
His-0.108
Token in
Token ID287
Feature activation-8.01e+00

Pos logit contributions
 globe+0.308
ool+0.285
 chim+0.279
 lid+0.275
 angles+0.275
Neg logit contributions
They-0.289
He-0.288
 please-0.280
His-0.270
Jack-0.249
Token our
Token ID674
Feature activation-7.90e+00

Pos logit contributions
 globe+0.139
hole+0.134
ool+0.133
ps+0.133
anger+0.130
Neg logit contributions
They-0.069
Jack-0.064
His-0.061
He-0.059
 him-0.056
Token mouths
Token ID28552
Feature activation-1.63e+01

Pos logit contributions
hole+0.096
ool+0.095
ps+0.092
 globe+0.088
 atop+0.086
Neg logit contributions
 His-0.110
His-0.109
Jack-0.104
He-0.103
 He-0.099
Token."
Token ID526
Feature activation+9.64e+00

Pos logit contributions
 globe+0.239
ool+0.215
 bulb+0.213
ps+0.212
els+0.210
Neg logit contributions
They-0.201
 please-0.191
He-0.187
His-0.181
As-0.180
Token
Token ID198
Feature activation+7.83e+00

Pos logit contributions
His+0.079
Jack+0.078
Johnny+0.075
David+0.072
Un+0.069
Neg logit contributions
hole-0.167
 bulb-0.166
 pieces-0.165
 globe-0.164
ps-0.163
Token should
Token ID815
Feature activation-2.30e+00

Pos logit contributions
 bulb+0.013
ps+0.012
 globe+0.012
 pieces+0.012
hole+0.012
Neg logit contributions
Jack-0.014
His-0.014
David-0.013
They-0.013
Jim-0.013
Token not
Token ID407
Feature activation-2.69e+00

Pos logit contributions
 ring+0.027
ps+0.027
 bulb+0.026
 falling+0.025
 trunk+0.025
Neg logit contributions
Jack-0.033
Un-0.033
As-0.032
His-0.032
Da-0.032
Token have
Token ID423
Feature activation-5.34e+00

Pos logit contributions
hole+0.038
 bulb+0.038
 trunk+0.037
ps+0.037
 pieces+0.037
Neg logit contributions
Jack-0.032
His-0.032
They-0.029
 His-0.029
David-0.029
Token taken
Token ID2077
Feature activation-1.23e+01

Pos logit contributions
hole+0.077
 globe+0.076
 bulb+0.074
ool+0.073
ps+0.072
Neg logit contributions
Jack-0.067
His-0.067
They-0.064
 His-0.061
He-0.060
Token your
Token ID534
Feature activation-8.70e+00

Pos logit contributions
hole+0.201
 globe+0.198
 bulb+0.194
ool+0.192
 atop+0.187
Neg logit contributions
Jack-0.128
His-0.126
They-0.122
 His-0.117
He-0.114
Token sack
Token ID23704
Feature activation-2.26e+01

Pos logit contributions
hole+0.128
ps+0.122
et+0.120
ening+0.120
ool+0.116
Neg logit contributions
 His-0.093
Jack-0.092
They-0.091
 Something-0.090
His-0.090
Token.
Token ID13
Feature activation+4.53e+00

Pos logit contributions
hole+0.291
els+0.289
 bulb+0.289
 globe+0.288
ening+0.287
Neg logit contributions
They-0.293
His-0.291
He-0.283
Jack-0.278
 His-0.265
Token I
Token ID314
Feature activation-1.45e+00

Pos logit contributions
Jack+0.061
David+0.060
As+0.058
Jim+0.057
Harry+0.057
Neg logit contributions
 globe-0.050
 bulb-0.050
 trunk-0.050
ps-0.047
 edges-0.047
Token was
Token ID373
Feature activation-4.39e+00

Pos logit contributions
 bulb+0.017
ps+0.017
hole+0.017
 globe+0.017
 pieces+0.016
Neg logit contributions
Jack-0.020
His-0.020
David-0.019
Jim-0.019
They-0.019
Token curious
Token ID11040
Feature activation-1.14e+01

Pos logit contributions
hole+0.048
 globe+0.047
 bulb+0.046
ps+0.046
 pieces+0.045
Neg logit contributions
His-0.068
Jack-0.068
They-0.065
 His-0.064
He-0.063
Token.
Token ID13
Feature activation+4.78e+00

Pos logit contributions
hole+0.134
 globe+0.134
 bulb+0.131
ool+0.129
ps+0.127
Neg logit contributions
His-0.166
They-0.164
Jack-0.164
He-0.160
 His-0.159
Token not
Token ID407
Feature activation-2.20e+00

Pos logit contributions
 bulb+0.025
ps+0.025
 trunk+0.024
 ring+0.024
 globe+0.024
Neg logit contributions
Jack-0.023
His-0.022
Un-0.021
David-0.021
Jim-0.021
Token have
Token ID423
Feature activation-4.97e+00

Pos logit contributions
hole+0.036
 globe+0.035
ps+0.035
 bulb+0.034
 pieces+0.034
Neg logit contributions
His-0.023
Jack-0.023
They-0.021
 His-0.021
He-0.020
Token knocked
Token ID13642
Feature activation-1.10e+01

Pos logit contributions
hole+0.077
 globe+0.076
ool+0.074
 bulb+0.073
ps+0.072
Neg logit contributions
Jack-0.058
His-0.057
They-0.056
 His-0.052
As-0.051
Token down
Token ID866
Feature activation-1.07e+01

Pos logit contributions
ool+0.163
hole+0.159
 globe+0.156
 trunk+0.152
 bulb+0.148
Neg logit contributions
They-0.139
Jack-0.135
His-0.125
 him-0.124
He-0.119
Token your
Token ID534
Feature activation-8.33e+00

Pos logit contributions
hole+0.182
ool+0.174
 bulb+0.173
 globe+0.172
ening+0.168
Neg logit contributions
His-0.109
They-0.109
Jack-0.104
 His-0.101
He-0.100
Token tower
Token ID10580
Feature activation-2.24e+01

Pos logit contributions
ool+0.105
hole+0.103
ening+0.102
 atop+0.101
ps+0.091
Neg logit contributions
 His-0.104
 Dad-0.103
 He-0.100
Jack-0.100
 coach-0.099
Token.
Token ID13
Feature activation+5.30e+00

Pos logit contributions
ening+0.313
 globe+0.309
 atop+0.304
ps+0.302
hole+0.296
Neg logit contributions
They-0.280
His-0.264
He-0.254
 His-0.243
Jack-0.241
Token That
Token ID1320
Feature activation-3.19e+00

Pos logit contributions
Jack+0.057
His+0.051
David+0.050
Jim+0.050
As+0.047
Neg logit contributions
 globe-0.080
 bulb-0.078
hole-0.078
ps-0.077
 chim-0.076
Token was
Token ID373
Feature activation-3.65e+00

Pos logit contributions
hole+0.051
ps+0.050
ool+0.048
 atop+0.047
 globe+0.047
Neg logit contributions
Jack-0.034
They-0.033
His-0.032
 His-0.032
He-0.031
Token not
Token ID407
Feature activation-4.17e+00

Pos logit contributions
hole+0.058
 globe+0.055
ool+0.054
 bulb+0.054
ps+0.054
Neg logit contributions
His-0.040
Jack-0.040
They-0.039
 His-0.036
He-0.036
Token nice
Token ID3621
Feature activation-1.02e+01

Pos logit contributions
hole+0.067
 globe+0.066
ool+0.063
 lid+0.062
els+0.062
Neg logit contributions
They-0.046
His-0.045
Jack-0.043
 His-0.041
He-0.041
Token opened
Token ID4721
Feature activation-1.14e+01

Pos logit contributions
ps+0.063
hole+0.062
ool+0.060
ening+0.059
els+0.059
Neg logit contributions
 His-0.031
Jack-0.031
His-0.028
 Her-0.028
They-0.028
Token a
Token ID257
Feature activation-5.79e+00

Pos logit contributions
 globe+0.166
ps+0.166
 plank+0.163
ool+0.163
 bulb+0.159
Neg logit contributions
They-0.111
His-0.111
 As-0.107
As-0.107
 him-0.106
Token treasure
Token ID14068
Feature activation-1.97e+01

Pos logit contributions
ps+0.086
els+0.074
cks+0.069
ening+0.067
hole+0.065
Neg logit contributions
 please-0.083
He-0.081
Jack-0.079
His-0.078
Un-0.076
Token chest
Token ID7721
Feature activation-1.89e+01

Pos logit contributions
ps+0.229
ening+0.213
ool+0.201
els+0.196
 blades+0.195
Neg logit contributions
 please-0.272
He-0.263
They-0.258
His-0.258
 He-0.252
Token of
Token ID286
Feature activation-5.87e+00

Pos logit contributions
ps+0.292
els+0.285
 blades+0.282
 pieces+0.269
 edges+0.269
Neg logit contributions
 please-0.194
They-0.189
He-0.186
His-0.184
 He-0.169
Token gold
Token ID3869
Feature activation-2.23e+01

Pos logit contributions
hole+0.088
ool+0.084
ps+0.083
 globe+0.082
ening+0.079
Neg logit contributions
 His-0.067
His-0.066
Jack-0.063
 him-0.063
They-0.062
Token coins
Token ID10796
Feature activation-2.20e+01

Pos logit contributions
ps+0.279
 angle+0.271
 bulb+0.267
ening+0.265
 angles+0.264
Neg logit contributions
He-0.243
 He-0.241
 he-0.233
They-0.231
 please-0.228
Token.
Token ID13
Feature activation+8.84e+00

Pos logit contributions
ool+0.322
 blades+0.317
ps+0.316
 bulb+0.315
 globe+0.309
Neg logit contributions
They-0.223
He-0.210
 please-0.209
 he-0.204
His-0.203
Token
Token ID220
Feature activation+2.80e+00

Pos logit contributions
His+0.066
Jack+0.066
They+0.060
 His+0.057
He+0.057
Neg logit contributions
hole-0.166
 globe-0.162
ps-0.161
 bulb-0.161
 pieces-0.160
Token
Token ID198
Feature activation+7.05e+00

Pos logit contributions
His+0.021
Jack+0.021
They+0.018
 His+0.017
Johnny+0.017
Neg logit contributions
hole-0.052
 globe-0.050
 trunk-0.049
 pieces-0.049
ps-0.049
Token
Token ID198
Feature activation+8.06e+00

Pos logit contributions
His+0.043
Jack+0.043
They+0.038
 His+0.037
He+0.035
Neg logit contributions
hole-0.141
 globe-0.138
ps-0.136
 pieces-0.136
 bulb-0.136
Token could
Token ID714
Feature activation-6.05e-01

Pos logit contributions
 globe+0.071
hole+0.069
ool+0.068
ps+0.068
 bulb+0.067
Neg logit contributions
Jack-0.044
His-0.043
They-0.042
He-0.040
 His-0.040
Token not
Token ID407
Feature activation-2.20e+00

Pos logit contributions
 falling+0.008
 bulb+0.007
 scattered+0.007
ps+0.007
 ladder+0.007
Neg logit contributions
Jack-0.009
His-0.009
David-0.009
enny-0.008
Johnny-0.008
Token keep
Token ID1394
Feature activation-8.91e+00

Pos logit contributions
hole+0.033
 globe+0.032
ps+0.031
 pieces+0.031
ool+0.031
Neg logit contributions
His-0.026
Jack-0.025
They-0.025
He-0.023
 His-0.023
Token filling
Token ID12591
Feature activation-1.75e+01

Pos logit contributions
ool+0.144
hole+0.143
 globe+0.132
 code+0.128
 lid+0.125
Neg logit contributions
 him-0.104
 His-0.090
They-0.089
Jack-0.088
 Fro-0.088
Token the
Token ID262
Feature activation-8.09e+00

Pos logit contributions
hole+0.308
ool+0.292
 globe+0.276
 bulb+0.275
ps+0.271
Neg logit contributions
They-0.163
His-0.163
Jack-0.162
 him-0.157
He-0.151
Token bucket
Token ID19236
Feature activation-2.22e+01

Pos logit contributions
ool+0.106
hole+0.103
ps+0.101
els+0.094
cut+0.093
Neg logit contributions
They-0.107
Jack-0.101
 They-0.099
He-0.092
As-0.091
Token.
Token ID13
Feature activation+7.84e+00

Pos logit contributions
ps+0.308
 globe+0.288
hole+0.288
els+0.285
ening+0.274
Neg logit contributions
They-0.278
His-0.272
He-0.272
Jack-0.261
 He-0.243
Token The
Token ID383
Feature activation-1.11e+00

Pos logit contributions
Jack+0.036
His+0.035
They+0.032
As+0.026
He+0.026
Neg logit contributions
hole-0.168
 bulb-0.166
 pieces-0.166
 trunk-0.165
 globe-0.164
Token rain
Token ID6290
Feature activation-1.39e+01

Pos logit contributions
hole+0.014
ool+0.014
 globe+0.012
els+0.011
ps+0.011
Neg logit contributions
They-0.017
Jack-0.016
He-0.016
His-0.015
As-0.015
Token was
Token ID373
Feature activation-4.38e+00

Pos logit contributions
hole+0.201
 globe+0.196
ool+0.189
 bulb+0.189
ps+0.186
Neg logit contributions
Jack-0.164
They-0.163
His-0.160
He-0.151
As-0.150
Token heavy
Token ID4334
Feature activation-1.30e+01

Pos logit contributions
hole+0.066
 globe+0.065
ool+0.062
els+0.060
 bulb+0.060
Neg logit contributions
His-0.052
They-0.050
Jack-0.049
He-0.048
 His-0.047
Token chest
Token ID7721
Feature activation-1.62e+01

Pos logit contributions
ps+0.052
 bulb+0.051
ool+0.050
els+0.049
 falling+0.049
Neg logit contributions
His-0.089
He-0.088
 His-0.087
Jack-0.084
 He-0.083
Token and
Token ID290
Feature activation-2.10e-01

Pos logit contributions
ps+0.280
ool+0.271
els+0.270
 bulb+0.270
 blades+0.269
Neg logit contributions
They-0.151
His-0.146
 His-0.142
He-0.140
As-0.136
Token found
Token ID1043
Feature activation-5.97e+00

Pos logit contributions
hole+0.003
ps+0.003
ool+0.003
 bulb+0.003
 globe+0.003
Neg logit contributions
Jack-0.003
His-0.003
They-0.003
 His-0.003
He-0.002
Token lots
Token ID6041
Feature activation-1.11e+01

Pos logit contributions
ool+0.107
hole+0.107
 bulb+0.105
 falling+0.105
ening+0.104
Neg logit contributions
 His-0.049
Jack-0.048
His-0.047
They-0.047
He-0.043
Token of
Token ID286
Feature activation-7.08e+00

Pos logit contributions
ps+0.207
 globe+0.202
hole+0.201
 bulb+0.199
ool+0.193
Neg logit contributions
Jack-0.085
He-0.083
His-0.081
They-0.079
 His-0.076
Token gold
Token ID3869
Feature activation-2.22e+01

Pos logit contributions
ps+0.078
hole+0.078
ool+0.076
 bulb+0.074
 globe+0.073
Neg logit contributions
Jack-0.108
He-0.105
His-0.104
 His-0.103
They-0.100
Token coins
Token ID10796
Feature activation-2.17e+01

Pos logit contributions
 angle+0.226
 bulb+0.221
 stem+0.205
 wil+0.204
ening+0.203
Neg logit contributions
He-0.340
They-0.336
 He-0.323
 him-0.314
His-0.308
Token and
Token ID290
Feature activation-2.21e+00

Pos logit contributions
 bulb+0.289
 blades+0.287
ool+0.286
 chim+0.280
ps+0.276
Neg logit contributions
They-0.269
He-0.251
His-0.239
Jack-0.234
 He-0.228
Token shiny
Token ID22441
Feature activation-1.61e+01

Pos logit contributions
hole+0.017
ool+0.017
ps+0.016
 globe+0.016
 bulb+0.016
Neg logit contributions
His-0.041
Jack-0.041
 His-0.040
They-0.040
He-0.040
Token jewels
Token ID42701
Feature activation-1.97e+01

Pos logit contributions
hole+0.147
 bulb+0.146
ool+0.146
 falling+0.146
 globe+0.141
Neg logit contributions
He-0.248
 He-0.248
His-0.246
 His-0.245
 him-0.240
Token.
Token ID13
Feature activation+8.29e+00

Pos logit contributions
ool+0.269
 blades+0.268
 bulb+0.265
ps+0.264
els+0.264
Neg logit contributions
They-0.228
He-0.217
His-0.203
Jack-0.199
 he-0.198
Token,
Token ID11
Feature activation+8.50e-01

Pos logit contributions
ool+0.097
 globe+0.095
hole+0.095
 bulb+0.092
 blades+0.092
Neg logit contributions
 please-0.056
They-0.048
 Max-0.046
 him-0.045
Jack-0.045
Token do
Token ID466
Feature activation-4.20e+00

Pos logit contributions
His+0.009
Jack+0.009
They+0.009
 His+0.008
He+0.008
Neg logit contributions
 bulb-0.013
hole-0.013
ps-0.013
 globe-0.013
 pieces-0.012
Token not
Token ID407
Feature activation-2.49e+00

Pos logit contributions
hole+0.072
 globe+0.069
 bulb+0.069
ps+0.068
 pieces+0.067
Neg logit contributions
His-0.041
Jack-0.040
They-0.039
 His-0.037
He-0.036
Token touch
Token ID3638
Feature activation-1.24e+01

Pos logit contributions
hole+0.039
 globe+0.038
ps+0.037
 pieces+0.036
ool+0.036
Neg logit contributions
His-0.028
They-0.027
Jack-0.027
 His-0.026
He-0.026
Token the
Token ID262
Feature activation-7.13e+00

Pos logit contributions
 globe+0.220
hole+0.211
ool+0.210
 atop+0.200
 code+0.197
Neg logit contributions
They-0.102
 him-0.101
His-0.100
Jack-0.098
 His-0.098
Token pins
Token ID20567
Feature activation-2.22e+01

Pos logit contributions
hole+0.087
ool+0.087
ening+0.085
 atop+0.085
 globe+0.081
Neg logit contributions
They-0.094
Ben-0.088
Jack-0.087
Sam-0.085
L-0.084
Token.
Token ID13
Feature activation+4.65e+00

Pos logit contributions
 globe+0.310
 trunk+0.297
ool+0.296
ening+0.293
 chim+0.292
Neg logit contributions
They-0.283
 please-0.275
He-0.252
His-0.245
Jack-0.234
Token They
Token ID1119
Feature activation-2.87e+00

Pos logit contributions
Jack+0.047
His+0.042
Jim+0.040
David+0.040
As+0.038
Neg logit contributions
 bulb-0.076
hole-0.075
ps-0.074
 trunk-0.073
 globe-0.073
Token are
Token ID389
Feature activation-3.98e+00

Pos logit contributions
hole+0.047
ool+0.047
 globe+0.047
ening+0.046
 atop+0.046
Neg logit contributions
They-0.027
 please-0.026
He-0.024
As-0.022
 His-0.022
Token not
Token ID407
Feature activation-4.41e+00

Pos logit contributions
 globe+0.070
hole+0.069
ool+0.065
 angle+0.065
anger+0.062
Neg logit contributions
They-0.039
 please-0.035
He-0.034
His-0.033
 His-0.033
Token toys
Token ID14958
Feature activation-1.66e+01

Pos logit contributions
 globe+0.070
hole+0.065
ool+0.064
 lid+0.063
 atop+0.061
Neg logit contributions
They-0.051
His-0.045
 His-0.045
He-0.045
 please-0.045
Token he
Token ID339
Feature activation-7.31e+00

Pos logit contributions
hole+0.152
 globe+0.148
 bulb+0.148
ps+0.147
 pieces+0.145
Neg logit contributions
His-0.085
Jack-0.084
They-0.079
 His-0.077
He-0.075
Token didn
Token ID1422
Feature activation-6.61e+00

Pos logit contributions
hole+0.109
 globe+0.109
 bulb+0.107
ps+0.105
ool+0.105
Neg logit contributions
His-0.082
Jack-0.082
 His-0.078
They-0.077
He-0.077
Token't
Token ID470
Feature activation-3.59e+00

Pos logit contributions
 bulb+0.090
hole+0.090
ps+0.090
 pieces+0.089
ool+0.089
Neg logit contributions
Jack-0.080
His-0.077
They-0.074
Jim-0.072
David-0.072
Token lose
Token ID4425
Feature activation-1.16e+01

Pos logit contributions
 globe+0.054
hole+0.054
ps+0.053
 pieces+0.052
 bulb+0.052
Neg logit contributions
His-0.042
Jack-0.040
They-0.040
He-0.039
 His-0.039
Token his
Token ID465
Feature activation-8.06e+00

Pos logit contributions
ool+0.216
hole+0.210
 globe+0.208
 chim+0.202
 bulb+0.201
Neg logit contributions
His-0.099
 His-0.092
He-0.089
Jack-0.088
 him-0.087
Token wallet
Token ID13008
Feature activation-2.22e+01

Pos logit contributions
ool+0.127
hole+0.125
ps+0.125
 bulb+0.123
ening+0.123
Neg logit contributions
 His-0.093
His-0.090
 He-0.078
He-0.077
 Dad-0.073
Token again
Token ID757
Feature activation-1.06e+01

Pos logit contributions
ps+0.312
 globe+0.309
 blades+0.306
 bulb+0.304
 chim+0.302
Neg logit contributions
His-0.274
 His-0.262
He-0.260
 He-0.248
Jack-0.241
Token.
Token ID13
Feature activation+7.91e+00

Pos logit contributions
 globe+0.148
ps+0.145
ool+0.142
 bulb+0.140
hole+0.137
Neg logit contributions
His-0.142
 His-0.135
They-0.135
Jack-0.133
He-0.133
Token And
Token ID843
Feature activation+1.67e-01

Pos logit contributions
 unison+0.079
David+0.079
Danny+0.077
Tony+0.076
Her+0.076
Neg logit contributions
 way-0.107
 pieces-0.102
 trunk-0.100
 wall-0.097
 tree-0.096
Token Lily
Token ID20037
Feature activation-3.05e+00

Pos logit contributions
His+0.002
Un+0.002
They+0.002
Jack+0.002
Her+0.002
Neg logit contributions
 pieces-0.002
 way-0.002
 on-0.002
 box-0.002
 globe-0.002
Token felt
Token ID2936
Feature activation-4.42e+00

Pos logit contributions
hole+0.049
 globe+0.048
 bulb+0.047
ps+0.047
ool+0.047
Neg logit contributions
His-0.032
Jack-0.032
They-0.030
 His-0.029
He-0.029
Token chest
Token ID7721
Feature activation-1.68e+01

Pos logit contributions
ps+0.199
els+0.188
ening+0.187
 angles+0.183
 blades+0.182
Neg logit contributions
Jack-0.256
His-0.253
He-0.252
They-0.250
 His-0.245
Token full
Token ID1336
Feature activation-9.71e+00

Pos logit contributions
ps+0.263
els+0.259
 blades+0.248
 bulb+0.240
 bulbs+0.238
Neg logit contributions
His-0.183
Jack-0.182
They-0.178
He-0.176
 His-0.169
Token of
Token ID286
Feature activation-6.24e+00

Pos logit contributions
ps+0.197
 globe+0.185
els+0.184
 blades+0.183
 plank+0.182
Neg logit contributions
 His-0.066
Jack-0.065
His-0.063
They-0.059
He-0.059
Token shiny
Token ID22441
Feature activation-1.70e+01

Pos logit contributions
hole+0.072
ps+0.071
ool+0.071
et+0.064
ening+0.064
Neg logit contributions
Jack-0.091
 His-0.090
He-0.089
His-0.088
 He-0.087
Token gold
Token ID3869
Feature activation-2.27e+01

Pos logit contributions
ps+0.181
 wil+0.172
ool+0.167
 falling+0.166
hole+0.165
Neg logit contributions
 He-0.244
He-0.234
 His-0.231
His-0.225
 him-0.223
Token coins
Token ID10796
Feature activation-2.21e+01

Pos logit contributions
 wil+0.315
 angle+0.312
 angles+0.309
ps+0.306
 bulb+0.297
Neg logit contributions
 He-0.242
He-0.237
They-0.232
 he-0.210
 His-0.207
Token!
Token ID0
Feature activation+8.26e+00

Pos logit contributions
 bulb+0.340
ps+0.336
 blades+0.328
 prope+0.322
ool+0.321
Neg logit contributions
They-0.221
He-0.205
 he-0.194
 He-0.193
Jack-0.191
Token Tim
Token ID5045
Feature activation-2.36e+00

Pos logit contributions
His+0.077
Jack+0.070
Her+0.070
Luke+0.069
They+0.068
Neg logit contributions
 ladder-0.132
 trunk-0.131
 pieces-0.130
 chim-0.127
ool-0.126
Token,
Token ID11
Feature activation+2.20e+00

Pos logit contributions
hole+0.035
 bulb+0.035
ool+0.035
ps+0.035
 globe+0.035
Neg logit contributions
Jack-0.027
His-0.027
 His-0.025
They-0.025
He-0.025
Token Sally
Token ID25737
Feature activation-2.74e+00

Pos logit contributions
Jack+0.016
His+0.015
They+0.014
 His+0.014
He+0.013
Neg logit contributions
hole-0.043
ps-0.042
 globe-0.042
 bulb-0.042
ool-0.042
Token,
Token ID11
Feature activation+1.89e+00

Pos logit contributions
hole+0.044
ool+0.044
ps+0.043
 bulb+0.042
ening+0.042
Neg logit contributions
Jack-0.030
His-0.029
 His-0.028
They-0.027
He-0.027
TokenMaybe
Token ID13300
Feature activation-2.80e+00

Pos logit contributions
His+0.045
Jack+0.045
They+0.041
 His+0.041
He+0.039
Neg logit contributions
hole-0.099
 globe-0.097
 bulb-0.097
ps-0.097
 pieces-0.096
Token the
Token ID262
Feature activation-4.06e+00

Pos logit contributions
hole+0.051
ps+0.051
 globe+0.050
ool+0.050
 bulb+0.049
Neg logit contributions
Jack-0.023
They-0.023
His-0.023
 His-0.020
As-0.020
Token treasure
Token ID14068
Feature activation-1.81e+01

Pos logit contributions
hole+0.061
ool+0.059
els+0.059
ening+0.057
ps+0.055
Neg logit contributions
He-0.052
They-0.050
Dad-0.049
His-0.047
Jack-0.047
Token is
Token ID318
Feature activation-5.48e+00

Pos logit contributions
hole+0.217
ening+0.217
ool+0.216
 edges+0.208
 bulb+0.206
Neg logit contributions
They-0.248
He-0.233
 please-0.226
His-0.223
 He-0.222
Token not
Token ID407
Feature activation-4.83e+00

Pos logit contributions
hole+0.095
ool+0.092
 globe+0.089
els+0.088
 blades+0.086
Neg logit contributions
They-0.052
 please-0.051
His-0.045
 Dad-0.045
 His-0.044
Token gold
Token ID3869
Feature activation-2.21e+01

Pos logit contributions
els+0.080
ool+0.079
 globe+0.079
hole+0.079
 lid+0.075
Neg logit contributions
They-0.045
 please-0.043
He-0.041
 He-0.040
 His-0.040
Token or
Token ID393
Feature activation-3.59e+00

Pos logit contributions
 bulb+0.303
 wil+0.299
 globe+0.298
hole+0.297
 mant+0.295
Neg logit contributions
 please-0.246
They-0.243
He-0.239
 He-0.234
His-0.224
Token jewels
Token ID42701
Feature activation-2.14e+01

Pos logit contributions
ool+0.050
hole+0.049
ps+0.046
 globe+0.046
 atop+0.045
Neg logit contributions
His-0.043
They-0.042
 His-0.041
He-0.040
 please-0.040
Token,"
Token ID553
Feature activation+7.54e+00

Pos logit contributions
ps+0.293
els+0.290
 bulb+0.289
 globe+0.285
ool+0.284
Neg logit contributions
They-0.255
 please-0.254
He-0.239
Jack-0.228
His-0.224
Token L
Token ID406
Feature activation-2.67e-02

Pos logit contributions
Jack+0.084
His+0.079
As+0.072
Jim+0.071
David+0.071
Neg logit contributions
hole-0.112
 pieces-0.112
ps-0.109
 globe-0.108
 bulb-0.108
Tokenila
Token ID10102
Feature activation-2.71e+00

Pos logit contributions
ps+0.000
hole+0.000
els+0.000
 globe+0.000
ool+0.000
Neg logit contributions
Jack-0.000
His-0.000
 His-0.000
They-0.000
As-0.000
Token load
Token ID3440
Feature activation-1.70e+01

Pos logit contributions
 globe+0.065
hole+0.065
ool+0.062
ps+0.061
els+0.060
Neg logit contributions
 His-0.042
They-0.042
His-0.041
Jack-0.041
As-0.040
Token your
Token ID534
Feature activation-8.14e+00

Pos logit contributions
hole+0.229
 globe+0.208
ool+0.197
 angle+0.193
 bulb+0.192
Neg logit contributions
They-0.240
His-0.228
Jack-0.225
He-0.221
 His-0.216
Token wagon
Token ID31860
Feature activation-2.03e+01

Pos logit contributions
hole+0.116
ool+0.104
 globe+0.099
 atop+0.098
ening+0.097
Neg logit contributions
 His-0.106
His-0.102
 He-0.096
He-0.095
 please-0.092
Token with
Token ID351
Feature activation-7.62e+00

Pos logit contributions
hole+0.265
ool+0.264
ps+0.262
 globe+0.256
els+0.254
Neg logit contributions
His-0.255
He-0.248
They-0.247
Jack-0.245
 His-0.233
Token p
Token ID279
Feature activation-1.37e+01

Pos logit contributions
hole+0.118
 globe+0.110
ool+0.106
 atop+0.103
 plank+0.099
Neg logit contributions
They-0.088
His-0.085
 His-0.084
Jack-0.084
 him-0.082
Tokenears
Token ID4127
Feature activation-2.20e+01

Pos logit contributions
 globe+0.163
hole+0.159
 bulb+0.154
ps+0.153
 pieces+0.152
Neg logit contributions
His-0.193
Jack-0.192
They-0.191
 His-0.185
He-0.183
Token near
Token ID1474
Feature activation-9.63e+00

Pos logit contributions
ps+0.298
hole+0.295
 globe+0.294
 bulb+0.288
ool+0.287
Neg logit contributions
They-0.265
Jack-0.261
His-0.258
He-0.256
 please-0.239
Token a
Token ID257
Feature activation-4.73e+00

Pos logit contributions
hole+0.164
 globe+0.160
ool+0.156
ps+0.155
 bulb+0.151
Neg logit contributions
Jack-0.091
They-0.090
His-0.088
 His-0.083
He-0.079
Token wild
Token ID4295
Feature activation-1.25e+01

Pos logit contributions
ps+0.057
els+0.051
hole+0.045
cks+0.043
 atop+0.042
Neg logit contributions
He-0.080
 please-0.078
Jack-0.078
His-0.077
Un-0.074
Token bear
Token ID6842
Feature activation-1.70e+01

Pos logit contributions
ps+0.144
hole+0.139
els+0.130
 atop+0.127
cks+0.125
Neg logit contributions
 His-0.179
His-0.178
Jack-0.177
 He-0.176
 please-0.173
Token.
Token ID13
Feature activation+7.15e+00

Pos logit contributions
ps+0.200
hole+0.195
 globe+0.193
 bulb+0.191
ool+0.190
Neg logit contributions
Jack-0.237
His-0.236
They-0.232
He-0.225
 His-0.224
Token she
Token ID673
Feature activation-4.03e+00

Pos logit contributions
hole+0.002
ps+0.002
 globe+0.002
ool+0.002
 pieces+0.002
Neg logit contributions
Jack-0.001
His-0.001
They-0.001
He-0.001
 His-0.001
Token continued
Token ID3767
Feature activation-5.65e+00

Pos logit contributions
 globe+0.067
ool+0.066
hole+0.066
 bulb+0.065
ps+0.065
Neg logit contributions
Jack-0.041
His-0.040
They-0.040
As-0.038
He-0.037
Token picking
Token ID10868
Feature activation-1.44e+01

Pos logit contributions
hole+0.099
ool+0.095
 globe+0.093
 lid+0.089
 edges+0.089
Neg logit contributions
His-0.056
 His-0.054
They-0.053
Her-0.051
Jack-0.048
Token her
Token ID607
Feature activation-7.25e+00

Pos logit contributions
hole+0.209
ool+0.194
 globe+0.193
 code+0.187
 plank+0.177
Neg logit contributions
They-0.179
 His-0.173
 him-0.170
 He-0.170
His-0.168
Token p
Token ID279
Feature activation-1.40e+01

Pos logit contributions
hole+0.114
ool+0.101
ening+0.100
 globe+0.095
cut+0.094
Neg logit contributions
 His-0.084
His-0.077
 Her-0.077
 So-0.074
 He-0.074
Tokenears
Token ID4127
Feature activation-2.20e+01

Pos logit contributions
hole+0.156
 globe+0.154
 bulb+0.149
ps+0.149
 pieces+0.145
Neg logit contributions
His-0.208
Jack-0.205
They-0.202
 His-0.200
He-0.195
Token.
Token ID13
Feature activation+7.77e+00

Pos logit contributions
ps+0.283
 globe+0.275
hole+0.268
 bulb+0.267
ool+0.267
Neg logit contributions
They-0.294
His-0.289
He-0.279
Jack-0.276
 His-0.262
Token With
Token ID2080
Feature activation-3.08e+00

Pos logit contributions
His+0.062
Jack+0.056
Jim+0.052
They+0.051
He+0.049
Neg logit contributions
hole-0.141
 bulb-0.138
ool-0.135
 globe-0.135
 trunk-0.135
Token a
Token ID257
Feature activation-2.64e+00

Pos logit contributions
hole+0.047
ps+0.046
 bulb+0.046
 pieces+0.046
 globe+0.046
Neg logit contributions
Jack-0.033
His-0.033
They-0.032
 His-0.031
He-0.030
Token smile
Token ID8212
Feature activation-1.30e+01

Pos logit contributions
hole+0.034
ps+0.033
 globe+0.032
 bulb+0.032
els+0.032
Neg logit contributions
His-0.037
Jack-0.036
They-0.035
He-0.034
 His-0.034
Token on
Token ID319
Feature activation-5.29e+00

Pos logit contributions
 globe+0.139
ool+0.138
hole+0.136
 edges+0.135
ps+0.135
Neg logit contributions
His-0.211
They-0.198
 His-0.196
He-0.194
Her-0.191
TokenDid
Token ID11633
Feature activation-3.91e+00

Pos logit contributions
 His+0.033
His+0.029
 As+0.029
Jack+0.028
ete+0.027
Neg logit contributions
 globe-0.073
hole-0.073
 pieces-0.073
 bulb-0.072
 plank-0.069
Token you
Token ID345
Feature activation-5.22e+00

Pos logit contributions
 pieces+0.056
hole+0.055
 bulb+0.055
ps+0.055
 trunk+0.054
Neg logit contributions
His-0.045
Jack-0.044
They-0.041
He-0.040
David-0.040
Token kids
Token ID3988
Feature activation-1.22e+01

Pos logit contributions
hole+0.080
 globe+0.079
ps+0.078
 pieces+0.077
 bulb+0.077
Neg logit contributions
His-0.059
Jack-0.058
They-0.055
 His-0.054
He-0.054
Token find
Token ID1064
Feature activation-8.94e+00

Pos logit contributions
hole+0.192
 globe+0.189
ps+0.187
 bulb+0.186
 pieces+0.186
Neg logit contributions
His-0.130
Jack-0.130
They-0.123
 His-0.120
He-0.118
Token my
Token ID616
Feature activation-7.05e+00

Pos logit contributions
 globe+0.165
hole+0.164
ool+0.163
 ladder+0.154
ening+0.154
Neg logit contributions
Jack-0.070
 him-0.069
His-0.068
 His-0.067
They-0.064
Token gold
Token ID3869
Feature activation-2.20e+01

Pos logit contributions
ool+0.084
ening+0.080
ps+0.080
hole+0.078
pan+0.074
Neg logit contributions
 His-0.104
 He-0.096
 Dad-0.094
His-0.093
 him-0.089
Token watch
Token ID2342
Feature activation-1.68e+01

Pos logit contributions
ening+0.248
ps+0.241
hole+0.236
 falling+0.232
 blades+0.226
Neg logit contributions
His-0.323
 His-0.314
 please-0.307
He-0.307
 He-0.300
Token?"
Token ID1701
Feature activation+8.18e+00

Pos logit contributions
ps+0.248
hole+0.240
 angles+0.238
 globe+0.236
 bulb+0.233
Neg logit contributions
 please-0.189
His-0.189
They-0.187
 His-0.184
Jack-0.182
Token
Token ID220
Feature activation+2.87e+00

Pos logit contributions
Jack+0.089
His+0.088
They+0.087
David+0.085
Un+0.083
Neg logit contributions
 pieces-0.118
hole-0.117
 trunk-0.117
 bulb-0.116
 globe-0.115
Token
Token ID198
Feature activation+7.29e+00

Pos logit contributions
Jack+0.023
His+0.022
David+0.020
Jim+0.020
Un+0.020
Neg logit contributions
hole-0.051
 pieces-0.050
ps-0.049
 bulb-0.048
 trunk-0.048
Token
Token ID198
Feature activation+8.26e+00

Pos logit contributions
His+0.058
David+0.057
Jack+0.056
Jim+0.055
Un+0.053
Neg logit contributions
 pieces-0.118
 bulb-0.118
 globe-0.116
hole-0.114
 trunk-0.114
Token a
Token ID257
Feature activation-5.79e+00

Pos logit contributions
 globe+0.166
ps+0.166
 plank+0.163
ool+0.163
 bulb+0.159
Neg logit contributions
They-0.111
His-0.111
 As-0.107
As-0.107
 him-0.106
Token treasure
Token ID14068
Feature activation-1.97e+01

Pos logit contributions
ps+0.086
els+0.074
cks+0.069
ening+0.067
hole+0.065
Neg logit contributions
 please-0.083
He-0.081
Jack-0.079
His-0.078
Un-0.076
Token chest
Token ID7721
Feature activation-1.89e+01

Pos logit contributions
ps+0.229
ening+0.213
ool+0.201
els+0.196
 blades+0.195
Neg logit contributions
 please-0.272
He-0.263
They-0.258
His-0.258
 He-0.252
Token of
Token ID286
Feature activation-5.87e+00

Pos logit contributions
ps+0.292
els+0.285
 blades+0.282
 pieces+0.269
 edges+0.269
Neg logit contributions
 please-0.194
They-0.189
He-0.186
His-0.184
 He-0.169
Token gold
Token ID3869
Feature activation-2.23e+01

Pos logit contributions
hole+0.088
ool+0.084
ps+0.083
 globe+0.082
ening+0.079
Neg logit contributions
 His-0.067
His-0.066
Jack-0.063
 him-0.063
They-0.062
Token coins
Token ID10796
Feature activation-2.20e+01

Pos logit contributions
ps+0.279
 angle+0.271
 bulb+0.267
ening+0.265
 angles+0.264
Neg logit contributions
He-0.243
 He-0.241
 he-0.233
They-0.231
 please-0.228
Token.
Token ID13
Feature activation+8.84e+00

Pos logit contributions
ool+0.322
 blades+0.317
ps+0.316
 bulb+0.315
 globe+0.309
Neg logit contributions
They-0.223
He-0.210
 please-0.209
 he-0.204
His-0.203
Token
Token ID220
Feature activation+2.80e+00

Pos logit contributions
His+0.066
Jack+0.066
They+0.060
 His+0.057
He+0.057
Neg logit contributions
hole-0.166
 globe-0.162
ps-0.161
 bulb-0.161
 pieces-0.160
Token
Token ID198
Feature activation+7.05e+00

Pos logit contributions
His+0.021
Jack+0.021
They+0.018
 His+0.017
Johnny+0.017
Neg logit contributions
hole-0.052
 globe-0.050
 trunk-0.049
 pieces-0.049
ps-0.049
Token
Token ID198
Feature activation+8.06e+00

Pos logit contributions
His+0.043
Jack+0.043
They+0.038
 His+0.037
He+0.035
Neg logit contributions
hole-0.141
 globe-0.138
ps-0.136
 pieces-0.136
 bulb-0.136
TokenEvery
Token ID6109
Feature activation+4.06e-01

Pos logit contributions
Jack+0.077
His+0.076
 His+0.074
Johnny+0.071
They+0.067
Neg logit contributions
hole-0.129
 globe-0.125
 pieces-0.124
ps-0.122
 trunk-0.122
Token they
Token ID484
Feature activation-4.15e+00

Pos logit contributions
hole+0.028
ool+0.027
ps+0.027
 bulb+0.027
 globe+0.027
Neg logit contributions
Jack-0.009
They-0.009
His-0.008
He-0.007
 His-0.007
Token fought
Token ID8350
Feature activation-1.40e+01

Pos logit contributions
ool+0.072
hole+0.071
 globe+0.071
 bulb+0.069
ps+0.069
Neg logit contributions
Jack-0.042
His-0.040
They-0.039
He-0.037
 His-0.035
Token over
Token ID625
Feature activation-5.37e+00

Pos logit contributions
ool+0.209
 bulb+0.206
 globe+0.203
hole+0.201
 egg+0.199
Neg logit contributions
They-0.180
His-0.167
Jack-0.163
He-0.160
 His-0.156
Token the
Token ID262
Feature activation-5.98e+00

Pos logit contributions
ool+0.095
hole+0.088
ps+0.086
 globe+0.085
 atop+0.085
Neg logit contributions
They-0.054
He-0.050
His-0.050
Jack-0.049
 His-0.047
Token shiny
Token ID22441
Feature activation-1.57e+01

Pos logit contributions
ool+0.091
els+0.088
hole+0.084
ying+0.083
ips+0.079
Neg logit contributions
He-0.068
Jack-0.064
 He-0.063
 him-0.062
His-0.057
Token gold
Token ID3869
Feature activation-2.20e+01

Pos logit contributions
ps+0.195
 falling+0.180
aro+0.180
eking+0.180
 fastest+0.176
Neg logit contributions
 His-0.197
His-0.194
 He-0.194
 him-0.190
 his-0.183
Token.
Token ID13
Feature activation+8.58e+00

Pos logit contributions
ps+0.323
eking+0.323
 wil+0.323
 tangled+0.314
hole+0.312
Neg logit contributions
 He-0.224
He-0.212
 His-0.201
They-0.198
His-0.194
Token The
Token ID383
Feature activation-7.79e-01

Pos logit contributions
His+0.073
Jack+0.071
David+0.066
He+0.065
Un+0.064
Neg logit contributions
 globe-0.146
hole-0.145
 trunk-0.145
 pieces-0.142
 bulb-0.142
Token puppy
Token ID26188
Feature activation-1.06e+01

Pos logit contributions
hole+0.013
ool+0.012
ps+0.012
els+0.012
 angles+0.012
Neg logit contributions
Jack-0.008
His-0.008
They-0.008
He-0.008
As-0.008
Token was
Token ID373
Feature activation-3.39e+00

Pos logit contributions
hole+0.182
ool+0.177
 globe+0.174
 bulb+0.173
ps+0.173
Neg logit contributions
Jack-0.105
His-0.104
They-0.099
As-0.098
 His-0.098
Token getting
Token ID1972
Feature activation-8.58e+00

Pos logit contributions
hole+0.050
 globe+0.050
ool+0.049
 bulb+0.048
ps+0.047
Neg logit contributions
His-0.040
Jack-0.039
They-0.038
 His-0.037
He-0.037
Token their
Token ID511
Feature activation+8.66e+00

Pos logit contributions
His+0.111
Jack+0.111
They+0.108
 His+0.107
He+0.105
Neg logit contributions
 globe-0.118
hole-0.117
ool-0.110
 bulb-0.105
 trunk-0.101
Token blanket
Token ID18447
Feature activation-4.70e+00

Pos logit contributions
Jack+0.124
His+0.123
 His+0.121
They+0.120
He+0.115
Neg logit contributions
hole-0.099
ool-0.092
 globe-0.090
ening-0.088
ps-0.087
Token back
Token ID736
Feature activation+3.24e+00

Pos logit contributions
els+0.065
 angles+0.064
 globe+0.063
hole+0.062
ool+0.062
Neg logit contributions
They-0.056
His-0.053
He-0.052
Jack-0.051
 His-0.050
Token!
Token ID0
Feature activation+1.70e+01

Pos logit contributions
They+0.041
His+0.039
 His+0.038
He+0.038
Jack+0.038
Neg logit contributions
ool-0.041
 globe-0.041
hole-0.040
ps-0.040
els-0.039
Token
Token ID198
Feature activation+1.50e+01

Pos logit contributions
Jack+0.130
His+0.126
They+0.113
David+0.111
Jim+0.108
Neg logit contributions
hole-0.299
 trunk-0.294
 globe-0.294
 bulb-0.290
ool-0.284
Token
Token ID198
Feature activation+1.60e+01

Pos logit contributions
His+0.150
Jack+0.148
 His+0.145
 please+0.145
Un+0.137
Neg logit contributions
 trunk-0.197
 pieces-0.195
hole-0.194
 globe-0.191
ool-0.190
Token"
Token ID1
Feature activation+1.48e+01

Pos logit contributions
 His+0.128
 As+0.110
 Fro+0.106
 He+0.104
 please+0.100
Neg logit contributions
 globe-0.238
 pieces-0.235
hole-0.233
 trunk-0.232
 bulb-0.229
TokenNow
Token ID3844
Feature activation+5.40e+00

Pos logit contributions
His+0.132
 His+0.129
Jack+0.128
They+0.120
 As+0.115
Neg logit contributions
 globe-0.247
hole-0.243
 bulb-0.242
 pieces-0.242
ps-0.241
Token that
Token ID326
Feature activation+7.74e+00

Pos logit contributions
Jack+0.060
His+0.060
They+0.058
 His+0.056
He+0.055
Neg logit contributions
hole-0.081
 globe-0.079
ps-0.078
 bulb-0.078
ool-0.078
Token bunny
Token ID44915
Feature activation-3.24e+00

Pos logit contributions
Jack+0.101
They+0.100
His+0.096
As+0.094
 His+0.094
Neg logit contributions
ps-0.101
els-0.099
hole-0.099
 globe-0.092
ool-0.091
Token won
Token ID1839
Feature activation+2.32e+00

Pos logit contributions
hole+0.039
els+0.036
ps+0.036
 globe+0.036
 bulb+0.035
Neg logit contributions
Jack-0.045
They-0.044
His-0.043
 His-0.042
He-0.042
TokenYou
Token ID1639
Feature activation+6.91e+00

Pos logit contributions
 His+0.094
His+0.093
Jack+0.087
 please+0.085
 concession+0.084
Neg logit contributions
hole-0.221
 globe-0.217
 bulb-0.215
 pieces-0.209
 edges-0.205
Token'll
Token ID1183
Feature activation+5.26e+00

Pos logit contributions
Jack+0.080
They+0.076
His+0.076
 His+0.073
He+0.073
Neg logit contributions
hole-0.101
 globe-0.100
ool-0.099
els-0.097
ening-0.097
Token find
Token ID1064
Feature activation+7.14e-01

Pos logit contributions
His+0.047
They+0.045
Jack+0.045
He+0.042
 His+0.042
Neg logit contributions
hole-0.092
 globe-0.090
 pieces-0.090
ool-0.089
els-0.089
Token out
Token ID503
Feature activation-2.46e-01

Pos logit contributions
Jack+0.006
His+0.005
They+0.005
 him+0.005
 His+0.005
Neg logit contributions
 globe-0.013
hole-0.013
ool-0.013
 ladder-0.012
 bulb-0.012
Token soon
Token ID2582
Feature activation-6.64e-01

Pos logit contributions
hole+0.004
 globe+0.004
ool+0.004
 bulb+0.004
ps+0.003
Neg logit contributions
Jack-0.003
His-0.003
They-0.003
As-0.003
He-0.003
Token."
Token ID526
Feature activation+1.54e+01

Pos logit contributions
 bulb+0.009
 globe+0.009
ool+0.009
els+0.009
 bulbs+0.009
Neg logit contributions
They-0.008
 please-0.008
Jack-0.007
As-0.007
 As-0.007
Token
Token ID198
Feature activation+1.37e+01

Pos logit contributions
Jack+0.121
His+0.117
They+0.116
Un+0.114
 concession+0.110
Neg logit contributions
hole-0.258
 trunk-0.258
 globe-0.254
 bulb-0.249
 pieces-0.248
Token
Token ID198
Feature activation+1.46e+01

Pos logit contributions
Jack+0.101
His+0.100
They+0.098
Un+0.097
 As+0.094
Neg logit contributions
 globe-0.219
 trunk-0.217
hole-0.216
 bulb-0.216
 pieces-0.215
TokenConf
Token ID18546
Feature activation+9.16e+00

Pos logit contributions
 As+0.095
 His+0.093
 While+0.093
 concession+0.087
 Fro+0.085
Neg logit contributions
 globe-0.235
 bulb-0.233
 pieces-0.232
cut-0.231
 trunk-0.228
Tokenused
Token ID1484
Feature activation+1.85e+00

Pos logit contributions
His+0.131
Jack+0.130
They+0.124
 His+0.123
He+0.119
Neg logit contributions
 pieces-0.114
ool-0.110
hole-0.107
ps-0.107
 bulb-0.107
Token,
Token ID11
Feature activation+8.25e+00

Pos logit contributions
Jack+0.026
His+0.026
 His+0.025
They+0.024
He+0.024
Neg logit contributions
hole-0.024
 globe-0.023
ps-0.023
 bulb-0.022
ool-0.022
Token hurt
Token ID5938
Feature activation-1.87e+00

Pos logit contributions
His+0.061
Jack+0.061
They+0.057
 His+0.055
He+0.055
Neg logit contributions
hole-0.094
 bulb-0.092
ps-0.092
 globe-0.092
 pieces-0.091
Token himself
Token ID2241
Feature activation-1.66e+00

Pos logit contributions
hole+0.032
ps+0.032
 bulb+0.032
 globe+0.032
 pieces+0.031
Neg logit contributions
His-0.017
Jack-0.017
They-0.015
 His-0.015
He-0.015
Token.
Token ID13
Feature activation+1.62e+01

Pos logit contributions
 globe+0.022
ps+0.022
els+0.022
 bulb+0.021
ool+0.021
Neg logit contributions
His-0.021
They-0.020
He-0.020
As-0.019
 His-0.019
Token
Token ID220
Feature activation+9.95e+00

Pos logit contributions
His+0.142
Jack+0.141
They+0.139
David+0.125
 concession+0.124
Neg logit contributions
 trunk-0.256
hole-0.248
 chim-0.245
 ladder-0.239
 bulb-0.234
Token
Token ID198
Feature activation+1.39e+01

Pos logit contributions
Jack+0.071
His+0.070
They+0.061
David+0.057
As+0.056
Neg logit contributions
hole-0.188
 globe-0.181
ps-0.180
 pieces-0.180
 bulb-0.179
Token
Token ID198
Feature activation+1.49e+01

Pos logit contributions
His+0.101
 please+0.099
Jack+0.098
 His+0.097
They+0.097
Neg logit contributions
hole-0.214
 trunk-0.214
 globe-0.212
 pieces-0.209
ool-0.205
TokenTom
Token ID13787
Feature activation+5.60e+00

Pos logit contributions
 His+0.104
 As+0.096
 please+0.094
 concession+0.090
elt+0.089
Neg logit contributions
 trunk-0.224
cut-0.218
 tree-0.214
 globe-0.213
 pieces-0.213
Token was
Token ID373
Feature activation+6.64e+00

Pos logit contributions
His+0.072
Jack+0.071
They+0.068
 His+0.066
He+0.065
Neg logit contributions
hole-0.075
ps-0.073
 bulb-0.072
 pieces-0.072
 globe-0.072
Token upset
Token ID9247
Feature activation-3.06e-01

Pos logit contributions
Jack+0.089
His+0.089
They+0.085
 His+0.082
He+0.082
Neg logit contributions
hole-0.086
ps-0.083
 bulb-0.083
 globe-0.083
 pieces-0.082
Token.
Token ID13
Feature activation+1.50e+01

Pos logit contributions
hole+0.004
 bulb+0.003
ps+0.003
 pieces+0.003
 globe+0.003
Neg logit contributions
Jack-0.004
His-0.004
They-0.004
 His-0.004
He-0.004
Token He
Token ID679
Feature activation+6.67e+00

Pos logit contributions
Jack+0.073
They+0.068
His+0.065
As+0.054
David+0.051
Neg logit contributions
 trunk-0.315
hole-0.315
 pieces-0.304
 globe-0.301
 edges-0.301
Token'm
Token ID1101
Feature activation+1.03e+01

Pos logit contributions
Jack+0.117
His+0.115
They+0.110
 His+0.107
He+0.106
Neg logit contributions
hole-0.176
 globe-0.174
ps-0.171
 bulb-0.169
 pieces-0.168
Token bored
Token ID17533
Feature activation-4.33e-01

Pos logit contributions
Jack+0.099
His+0.095
They+0.093
 please+0.093
 His+0.092
Neg logit contributions
 globe-0.181
hole-0.176
ool-0.167
 pieces-0.165
 bulb-0.164
Token of
Token ID286
Feature activation+6.83e+00

Pos logit contributions
ps+0.005
 globe+0.005
hole+0.005
 bulb+0.005
 atop+0.005
Neg logit contributions
Jack-0.006
They-0.006
His-0.006
He-0.006
 His-0.006
Token my
Token ID616
Feature activation+5.74e+00

Pos logit contributions
Jack+0.065
His+0.061
They+0.059
 His+0.059
Jim+0.055
Neg logit contributions
hole-0.120
ool-0.116
 globe-0.115
 bulb-0.110
ps-0.110
Token toys
Token ID14958
Feature activation-4.91e+00

Pos logit contributions
Jack+0.066
 His+0.065
His+0.064
They+0.060
 please+0.059
Neg logit contributions
hole-0.084
ps-0.079
ool-0.079
els-0.076
 atop-0.075
Token,"
Token ID553
Feature activation+1.67e+01

Pos logit contributions
ps+0.062
 globe+0.061
els+0.060
ool+0.059
hole+0.059
Neg logit contributions
They-0.064
Jack-0.063
His-0.062
He-0.061
 please-0.061
Token Mia
Token ID32189
Feature activation+8.89e+00

Pos logit contributions
Jack+0.161
His+0.159
They+0.155
As+0.152
David+0.146
Neg logit contributions
 bulb-0.259
 pieces-0.259
hole-0.253
 trunk-0.251
 globe-0.250
Token said
Token ID531
Feature activation+8.84e+00

Pos logit contributions
His+0.113
Jack+0.112
They+0.107
 His+0.104
He+0.103
Neg logit contributions
hole-0.120
ps-0.116
 bulb-0.116
 globe-0.116
 pieces-0.115
Token.
Token ID13
Feature activation+1.62e+01

Pos logit contributions
His+0.134
Jack+0.131
 His+0.129
They+0.125
He+0.121
Neg logit contributions
hole-0.103
 globe-0.101
ool-0.097
 bulb-0.096
 pieces-0.096
Token She
Token ID1375
Feature activation+8.95e+00

Pos logit contributions
Jack+0.165
His+0.155
David+0.152
Johnny+0.149
As+0.143
Neg logit contributions
 bulb-0.248
 trunk-0.243
 globe-0.240
 pieces-0.239
hole-0.238
Token felt
Token ID2936
Feature activation+5.20e+00

Pos logit contributions
His+0.116
Jack+0.115
They+0.110
 His+0.107
He+0.106
Neg logit contributions
hole-0.118
 bulb-0.116
ps-0.115
 globe-0.114
 pieces-0.114
Token.
Token ID13
Feature activation+1.84e+01

Pos logit contributions
His+0.059
Jack+0.058
They+0.058
 His+0.057
He+0.055
Neg logit contributions
hole-0.057
 globe-0.055
 bulb-0.054
ool-0.054
 pieces-0.053
Token The
Token ID383
Feature activation+1.10e+01

Pos logit contributions
Jack+0.143
His+0.137
David+0.125
They+0.120
Jim+0.118
Neg logit contributions
 trunk-0.323
hole-0.315
 bulb-0.314
 globe-0.310
 chim-0.310
Token cat
Token ID3797
Feature activation+2.79e+00

Pos logit contributions
Jack+0.132
His+0.128
They+0.127
He+0.124
 His+0.122
Neg logit contributions
hole-0.166
ps-0.148
ening-0.145
ool-0.145
 atop-0.143
Token said
Token ID531
Feature activation+7.43e+00

Pos logit contributions
His+0.032
Jack+0.032
They+0.031
 His+0.031
He+0.030
Neg logit contributions
hole-0.043
 globe-0.041
ool-0.040
 bulb-0.039
ps-0.039
Token,
Token ID11
Feature activation+8.85e+00

Pos logit contributions
His+0.098
Jack+0.097
They+0.093
 His+0.092
He+0.090
Neg logit contributions
hole-0.102
 globe-0.097
 bulb-0.095
ps-0.095
ool-0.095
Token "
Token ID366
Feature activation+1.49e+01

Pos logit contributions
His+0.079
Jack+0.076
They+0.074
 His+0.074
He+0.071
Neg logit contributions
hole-0.157
ool-0.152
ps-0.152
 globe-0.152
 pieces-0.151
TokenWhat
Token ID2061
Feature activation+9.74e+00

Pos logit contributions
 His+0.114
Jack+0.111
His+0.111
David+0.105
They+0.099
Neg logit contributions
 globe-0.255
 pieces-0.254
 bulb-0.253
hole-0.252
ps-0.247
Token is
Token ID318
Feature activation+7.06e+00

Pos logit contributions
His+0.142
Jack+0.141
They+0.135
 His+0.133
He+0.131
Neg logit contributions
hole-0.114
 globe-0.111
 bulb-0.111
ps-0.110
 pieces-0.109
Token that
Token ID326
Feature activation+5.53e+00

Pos logit contributions
Jack+0.082
His+0.080
They+0.079
 His+0.077
Her+0.073
Neg logit contributions
hole-0.109
 globe-0.099
ool-0.097
 pieces-0.094
 trunk-0.094
Token,
Token ID11
Feature activation+9.21e+00

Pos logit contributions
 His+0.067
His+0.067
They+0.065
Jack+0.065
 please+0.065
Neg logit contributions
hole-0.080
 globe-0.076
ps-0.075
els-0.075
 pieces-0.072
Token Skip
Token ID32214
Feature activation+2.06e+00

Pos logit contributions
His+0.135
Jack+0.134
They+0.127
 His+0.125
He+0.123
Neg logit contributions
 bulb-0.108
hole-0.106
 globe-0.105
 pieces-0.103
ps-0.103
Token at
Token ID379
Feature activation+3.41e-01

Pos logit contributions
hole+0.079
 globe+0.077
 bulb+0.076
ool+0.076
ps+0.075
Neg logit contributions
His-0.043
Jack-0.041
They-0.041
 His-0.039
He-0.037
Token the
Token ID262
Feature activation-5.31e-01

Pos logit contributions
Jack+0.003
His+0.003
They+0.003
 His+0.003
He+0.003
Neg logit contributions
hole-0.006
ool-0.006
 globe-0.006
ps-0.006
 bulb-0.006
Token beach
Token ID10481
Feature activation-9.90e+00

Pos logit contributions
hole+0.008
ool+0.007
els+0.007
ps+0.007
ening+0.007
Neg logit contributions
Jack-0.007
They-0.007
His-0.007
He-0.006
David-0.006
Token.
Token ID13
Feature activation+1.14e+01

Pos logit contributions
hole+0.137
ool+0.133
 bulb+0.128
ps+0.127
 globe+0.126
Neg logit contributions
They-0.129
His-0.120
 His-0.119
He-0.114
 He-0.112
Token<|endoftext|>
Token ID50256
Feature activation+4.80e+00

Pos logit contributions
Jack+0.133
His+0.124
Johnny+0.122
Sarah+0.120
David+0.119
Neg logit contributions
 globe-0.147
 hole-0.146
 chim-0.145
 trunk-0.143
 edges-0.140
TokenL
Token ID43
Feature activation+7.87e+00

Pos logit contributions
Jack+0.054
His+0.053
They+0.051
He+0.049
 His+0.049
Neg logit contributions
hole-0.074
 bulb-0.072
ps-0.072
 globe-0.071
ool-0.071
Tokenily
Token ID813
Feature activation+4.58e+00

Pos logit contributions
His+0.109
Jack+0.105
 His+0.101
He+0.100
Un+0.099
Neg logit contributions
 pieces-0.100
hole-0.100
ps-0.099
els-0.096
ool-0.095
Token and
Token ID290
Feature activation+7.37e+00

Pos logit contributions
Jack+0.056
His+0.056
They+0.054
 His+0.052
He+0.051
Neg logit contributions
hole-0.065
ps-0.062
 globe-0.062
 bulb-0.062
ool-0.061
Token Ben
Token ID3932
Feature activation+3.89e+00

Pos logit contributions
enny+0.111
As+0.108
 concession+0.107
Johnny+0.104
His+0.102
Neg logit contributions
 on-0.080
 ring-0.078
 bulb-0.078
 trunk-0.077
 tree-0.076
Token were
Token ID547
Feature activation+5.01e+00

Pos logit contributions
His+0.048
Jack+0.048
They+0.045
 His+0.044
He+0.043
Neg logit contributions
hole-0.052
 globe-0.052
 bulb-0.052
ps-0.051
 pieces-0.051
Token twins
Token ID20345
Feature activation-1.97e+00

Pos logit contributions
His+0.081
Jack+0.079
They+0.075
He+0.075
 His+0.074
Neg logit contributions
ps-0.051
hole-0.051
 pieces-0.050
 bulb-0.050
 falling-0.049
Token.
Token ID13
Feature activation+1.00e+01

Pos logit contributions
 bulb+0.029
 globe+0.028
 pieces+0.026
 atop+0.026
 glowing+0.026
Neg logit contributions
Jack-0.116
Jim-0.111
Kids-0.111
David-0.108
Luke-0.107
Token<|endoftext|>
Token ID50256
Feature activation+3.38e+00

Pos logit contributions
Jack+0.068
His+0.065
They+0.061
David+0.056
Jim+0.055
Neg logit contributions
hole-0.191
 trunk-0.190
 globe-0.190
 bulb-0.188
 pieces-0.186
TokenOnce
Token ID7454
Feature activation+4.55e+00

Pos logit contributions
His+0.021
Jack+0.020
They+0.018
 His+0.018
He+0.017
Neg logit contributions
hole-0.068
 globe-0.067
 bulb-0.066
 pieces-0.066
ps-0.066
Token upon
Token ID2402
Feature activation+5.90e+00

Pos logit contributions
His+0.052
Jack+0.051
They+0.050
He+0.047
David+0.046
Neg logit contributions
hole-0.067
 bulb-0.065
 globe-0.065
 pieces-0.064
ool-0.064
Token a
Token ID257
Feature activation+4.40e+00

Pos logit contributions
ete+0.119
Kids+0.116
David+0.112
enny+0.112
 coach+0.110
Neg logit contributions
 on-0.040
 pieces-0.040
 glowing-0.035
 broken-0.033
 trunk-0.031
Token time
Token ID640
Feature activation+2.57e+00

Pos logit contributions
His+0.092
Johnny+0.087
 His+0.087
Jack+0.086
David+0.086
Neg logit contributions
ool-0.024
 trunk-0.020
hole-0.020
 globe-0.019
 falling-0.019
Token,
Token ID11
Feature activation+7.14e+00

Pos logit contributions
His+0.043
Jack+0.042
They+0.041
He+0.040
 His+0.040
Neg logit contributions
hole-0.024
 globe-0.024
 bulb-0.023
 pieces-0.023
ps-0.022
Token there
Token ID612
Feature activation+8.26e-01

Pos logit contributions
Jack+0.120
His+0.120
David+0.115
They+0.115
 His+0.113
Neg logit contributions
 bulb-0.066
hole-0.065
 globe-0.064
 pieces-0.062
ps-0.061
Token was
Token ID373
Feature activation+2.64e+00

Pos logit contributions
His+0.011
Jack+0.011
They+0.011
David+0.011
He+0.010
Neg logit contributions
 globe-0.010
 bulb-0.010
hole-0.010
 pieces-0.010
 trunk-0.010
Token a
Token ID257
Feature activation-2.17e-01

Pos logit contributions
His+0.043
enny+0.042
ete+0.041
Kids+0.041
David+0.041
Neg logit contributions
 pieces-0.026
 bulb-0.022
 glowing-0.022
 on-0.022
 globe-0.022
Token little
Token ID1310
Feature activation-2.36e+00

Pos logit contributions
hole+0.003
ps+0.003
 globe+0.003
 bulb+0.003
 pieces+0.003
Neg logit contributions
His-0.003
Jack-0.003
They-0.003
 His-0.003
He-0.003
Token ice
Token ID4771
Feature activation-1.04e+01

Pos logit contributions
Jack+0.059
His+0.059
David+0.055
They+0.055
Jim+0.055
Neg logit contributions
 pieces-0.041
 bulb-0.040
 falling-0.038
 trunk-0.038
 globe-0.038
Token cream
Token ID8566
Feature activation-8.81e+00

Pos logit contributions
hole+0.129
 bulb+0.117
ening+0.115
ps+0.111
 globe+0.108
Neg logit contributions
They-0.156
His-0.151
Jack-0.147
He-0.143
 His-0.141
Token.
Token ID13
Feature activation+1.38e+01

Pos logit contributions
hole+0.119
 bulb+0.112
ps+0.110
ool+0.110
ening+0.110
Neg logit contributions
They-0.122
His-0.116
 His-0.113
He-0.111
 He-0.110
Token One
Token ID1881
Feature activation+4.50e+00

Pos logit contributions
Jack+0.130
They+0.121
His+0.121
David+0.116
Jim+0.113
Neg logit contributions
 trunk-0.230
 globe-0.228
hole-0.227
 bulb-0.225
ool-0.221
Token day
Token ID1110
Feature activation+1.84e+00

Pos logit contributions
Jack+0.090
His+0.089
They+0.088
David+0.086
He+0.085
Neg logit contributions
 bulb-0.027
 globe-0.027
 pieces-0.025
ool-0.025
hole-0.025
Token,
Token ID11
Feature activation+8.26e+00

Pos logit contributions
Jack+0.031
His+0.031
They+0.030
He+0.029
David+0.029
Neg logit contributions
hole-0.017
 globe-0.016
 bulb-0.016
 pieces-0.016
 falling-0.016
Token Tim
Token ID5045
Feature activation+3.87e+00

Pos logit contributions
Kids+0.090
enny+0.090
David+0.090
ete+0.089
Rachel+0.087
Neg logit contributions
 falling-0.123
 bulb-0.122
 globe-0.121
 ladder-0.119
 hanging-0.116
Tokenmy
Token ID1820
Feature activation+3.32e+00

Pos logit contributions
His+0.057
Jack+0.055
They+0.054
 His+0.053
Her+0.052
Neg logit contributions
hole-0.045
 pieces-0.045
 bulb-0.044
 trunk-0.043
ps-0.042
Token went
Token ID1816
Feature activation-9.29e-02

Pos logit contributions
His+0.052
They+0.049
Jack+0.049
David+0.049
Of+0.049
Neg logit contributions
 pieces-0.036
 bulb-0.033
 box-0.033
 way-0.032
 scattered-0.032
Token to
Token ID284
Feature activation+4.09e+00

Pos logit contributions
hole+0.001
ps+0.001
 bulb+0.001
 globe+0.001
 pieces+0.001
Neg logit contributions
Jack-0.001
His-0.001
They-0.001
 His-0.001
He-0.001
Token the
Token ID262
Feature activation+1.21e+00

Pos logit contributions
They+0.072
His+0.070
Jack+0.069
He+0.068
And+0.068
Neg logit contributions
 way-0.036
 trunk-0.035
 pieces-0.035
 bulb-0.035
 box-0.034
Token you
Token ID345
Feature activation+1.34e+01

Pos logit contributions
ps+0.244
hole+0.243
 plank+0.232
ool+0.225
 code+0.220
Neg logit contributions
They-0.184
Jack-0.166
As-0.161
 She-0.155
Da-0.154
Token to
Token ID284
Feature activation+1.10e+01

Pos logit contributions
Jack+0.160
His+0.158
They+0.156
 His+0.151
He+0.148
Neg logit contributions
hole-0.192
 globe-0.188
 bulb-0.184
ool-0.182
ps-0.181
Token the
Token ID262
Feature activation+9.18e+00

Pos logit contributions
His+0.125
Jack+0.124
They+0.118
 His+0.114
He+0.113
Neg logit contributions
hole-0.162
 bulb-0.160
ps-0.159
 globe-0.159
 pieces-0.157
Token stars
Token ID5788
Feature activation-3.68e+00

Pos logit contributions
Jack+0.124
They+0.121
His+0.119
Un+0.117
 please+0.117
Neg logit contributions
hole-0.114
ening-0.106
ps-0.104
els-0.104
ool-0.103
Token!"
Token ID2474
Feature activation+1.85e+01

Pos logit contributions
ps+0.044
hole+0.044
 globe+0.043
ool+0.043
els+0.043
Neg logit contributions
They-0.050
Jack-0.048
His-0.047
He-0.047
 please-0.046
Token The
Token ID383
Feature activation+1.08e+01

Pos logit contributions
Jack+0.149
His+0.149
They+0.137
Un+0.136
As+0.135
Neg logit contributions
 pieces-0.298
hole-0.297
 trunk-0.294
 bulb-0.292
 globe-0.288
Token little
Token ID1310
Feature activation+9.56e+00

Pos logit contributions
Jack+0.142
His+0.137
They+0.134
He+0.133
 His+0.126
Neg logit contributions
hole-0.151
ool-0.138
ps-0.135
ening-0.133
els-0.131
Token girl
Token ID2576
Feature activation+6.39e+00

Pos logit contributions
Jack+0.109
His+0.103
They+0.099
He+0.097
 His+0.097
Neg logit contributions
hole-0.146
ps-0.142
ool-0.138
 falling-0.135
 globe-0.132
Token laughed
Token ID13818
Feature activation+2.73e+00

Pos logit contributions
His+0.075
Jack+0.074
They+0.071
 His+0.070
He+0.068
Neg logit contributions
hole-0.094
 globe-0.091
 bulb-0.090
ps-0.090
 pieces-0.089
Token and
Token ID290
Feature activation+1.02e+01

Pos logit contributions
Jack+0.042
His+0.042
They+0.040
David+0.039
Jim+0.039
Neg logit contributions
hole-0.028
ps-0.028
 bulb-0.028
 globe-0.028
 pieces-0.028
Token said
Token ID531
Feature activation+7.69e+00

Pos logit contributions
Jack+0.114
His+0.113
They+0.109
 His+0.105
He+0.102
Neg logit contributions
hole-0.160
 globe-0.154
ool-0.154
 bulb-0.153
ps-0.152
Token his
Token ID465
Feature activation+1.69e+00

Pos logit contributions
Jack+0.009
His+0.009
They+0.009
As+0.008
 His+0.008
Neg logit contributions
 pieces-0.018
 bulb-0.018
hole-0.018
ps-0.018
 globe-0.018
Token red
Token ID2266
Feature activation-4.40e+00

Pos logit contributions
His+0.025
Jack+0.024
 His+0.024
They+0.023
He+0.023
Neg logit contributions
hole-0.021
ool-0.019
ps-0.019
ening-0.019
 globe-0.018
Token flag
Token ID6056
Feature activation-1.10e+01

Pos logit contributions
hole+0.048
 globe+0.046
ps+0.046
 bulb+0.046
 pieces+0.045
Neg logit contributions
His-0.068
Jack-0.067
They-0.065
 His-0.064
He-0.063
Token.
Token ID13
Feature activation+1.43e+01

Pos logit contributions
ps+0.141
hole+0.135
ening+0.134
ool+0.128
 globe+0.127
Neg logit contributions
His-0.153
 His-0.146
They-0.146
He-0.142
 He-0.134
Token One
Token ID1881
Feature activation+5.60e+00

Pos logit contributions
Jack+0.142
David+0.128
Jim+0.126
Luke+0.125
They+0.121
Neg logit contributions
 trunk-0.240
 globe-0.233
 bulb-0.232
 chim-0.231
ool-0.227
Token day
Token ID1110
Feature activation+3.19e+00

Pos logit contributions
Jack+0.106
His+0.106
They+0.103
 His+0.101
He+0.101
Neg logit contributions
hole-0.040
 bulb-0.039
 globe-0.039
 pieces-0.037
ool-0.037
Token,
Token ID11
Feature activation+9.55e+00

Pos logit contributions
Jack+0.054
His+0.053
They+0.052
He+0.051
David+0.050
Neg logit contributions
hole-0.028
 globe-0.028
 bulb-0.028
 pieces-0.028
 falling-0.027
Token it
Token ID340
Feature activation-4.16e-02

Pos logit contributions
enny+0.118
David+0.115
Kids+0.114
Luke+0.113
ete+0.113
Neg logit contributions
 bulb-0.135
 falling-0.133
 globe-0.132
 ladder-0.129
 hanging-0.126
Token started
Token ID2067
Feature activation+9.14e-01

Pos logit contributions
 bulb+0.000
 pieces+0.000
 falling+0.000
hole+0.000
ps+0.000
Neg logit contributions
Jack-0.001
David-0.001
His-0.001
Jim-0.001
They-0.001
Token to
Token ID284
Feature activation+3.86e+00

Pos logit contributions
Jack+0.011
His+0.010
They+0.010
He+0.010
Jim+0.009
Neg logit contributions
hole-0.013
 pieces-0.013
 bulb-0.013
 falling-0.013
ps-0.013
Token rain
Token ID6290
Feature activation-9.61e+00

Pos logit contributions
Jack+0.058
His+0.057
They+0.055
 His+0.054
He+0.053
Neg logit contributions
 bulb-0.044
hole-0.043
 globe-0.042
ps-0.042
 pieces-0.042
Token trying
Token ID2111
Feature activation-7.03e+00

Pos logit contributions
ool+0.004
hole+0.004
ps+0.004
els+0.004
 pieces+0.004
Neg logit contributions
His-0.003
Jack-0.003
 His-0.003
He-0.003
 He-0.003
Token to
Token ID284
Feature activation-2.26e+00

Pos logit contributions
hole+0.139
 globe+0.137
 bulb+0.134
ool+0.134
ps+0.133
Neg logit contributions
His-0.053
 His-0.050
They-0.048
Jack-0.045
He-0.045
Token find
Token ID1064
Feature activation-8.01e+00

Pos logit contributions
hole+0.032
 globe+0.031
 bulb+0.031
ps+0.031
 pieces+0.031
Neg logit contributions
His-0.028
Jack-0.027
They-0.026
 His-0.025
He-0.025
Token out
Token ID503
Feature activation-9.46e+00

Pos logit contributions
hole+0.133
 globe+0.133
ool+0.129
ps+0.128
 bulb+0.126
Neg logit contributions
His-0.079
Jack-0.077
 His-0.076
He-0.071
They-0.071
Token what
Token ID644
Feature activation-4.25e+00

Pos logit contributions
 globe+0.175
hole+0.175
 bulb+0.171
ps+0.171
ool+0.168
Neg logit contributions
His-0.078
 His-0.072
He-0.071
They-0.071
Jack-0.069
Token other
Token ID584
Feature activation-8.17e+00

Pos logit contributions
hole+0.072
ps+0.071
 globe+0.068
 bulb+0.067
ool+0.067
Neg logit contributions
Jack-0.043
His-0.043
 His-0.040
They-0.040
He-0.039
Token people
Token ID661
Feature activation-1.31e+01

Pos logit contributions
ool+0.134
ps+0.131
hole+0.126
 plank+0.124
 globe+0.123
Neg logit contributions
They-0.081
 Her-0.078
As-0.077
 She-0.077
He-0.076
Token were
Token ID547
Feature activation-7.62e+00

Pos logit contributions
ool+0.250
 lid+0.243
 edges+0.242
 plank+0.240
hole+0.240
Neg logit contributions
Jack-0.105
They-0.102
He-0.093
 him-0.092
 His-0.090
Token doing
Token ID1804
Feature activation-9.28e+00

Pos logit contributions
ool+0.131
 globe+0.128
 button+0.118
 prope+0.118
 bulb+0.118
Neg logit contributions
They-0.076
Jack-0.068
David-0.067
 him-0.066
 please-0.066
Token.
Token ID13
Feature activation+8.22e+00

Pos logit contributions
ool+0.122
 bulb+0.121
hole+0.120
 globe+0.117
 trunk+0.116
Neg logit contributions
They-0.122
Jack-0.121
His-0.116
 His-0.113
He-0.112
Token One
Token ID1881
Feature activation-1.02e+00

Pos logit contributions
Jack+0.076
His+0.074
They+0.071
David+0.071
Johnny+0.069
Neg logit contributions
 trunk-0.138
hole-0.137
 globe-0.135
 pieces-0.132
 bulb-0.131
Token upon
Token ID2402
Feature activation+3.17e+00

Pos logit contributions
His+0.020
Jack+0.020
They+0.019
 His+0.018
He+0.018
Neg logit contributions
hole-0.029
 bulb-0.028
 globe-0.028
 pieces-0.028
ps-0.027
Token a
Token ID257
Feature activation+1.83e+00

Pos logit contributions
ete+0.063
enny+0.061
Kids+0.060
David+0.060
 Matthew+0.060
Neg logit contributions
 pieces-0.022
 on-0.021
 glowing-0.018
 broken-0.018
 branch-0.017
Token time
Token ID640
Feature activation-4.86e-02

Pos logit contributions
His+0.038
Johnny+0.036
Jack+0.036
 His+0.036
David+0.036
Neg logit contributions
ool-0.010
 globe-0.009
 trunk-0.009
hole-0.009
 pieces-0.008
Token,
Token ID11
Feature activation+4.96e+00

Pos logit contributions
hole+0.001
 globe+0.000
 bulb+0.000
ps+0.000
 pieces+0.000
Neg logit contributions
His-0.001
Jack-0.001
They-0.001
 His-0.001
He-0.001
Token there
Token ID612
Feature activation-1.72e+00

Pos logit contributions
Jack+0.076
His+0.076
They+0.074
 His+0.072
He+0.071
Neg logit contributions
hole-0.055
ps-0.054
 globe-0.053
 pieces-0.052
 bulb-0.051
Token was
Token ID373
Feature activation+4.06e-01

Pos logit contributions
 globe+0.021
 bulb+0.021
hole+0.021
 pieces+0.020
 trunk+0.020
Neg logit contributions
His-0.023
Jack-0.023
They-0.022
David-0.022
He-0.022
Token a
Token ID257
Feature activation-2.45e+00

Pos logit contributions
His+0.006
David+0.006
Johnny+0.006
enny+0.006
He+0.006
Neg logit contributions
 pieces-0.004
 bulb-0.004
 globe-0.004
 trunk-0.004
 egg-0.004
Token brave
Token ID14802
Feature activation-1.03e+01

Pos logit contributions
hole+0.030
 globe+0.029
 bulb+0.029
ps+0.029
 pieces+0.029
Neg logit contributions
His-0.034
Jack-0.034
They-0.033
 His-0.032
He-0.032
Token knight
Token ID22062
Feature activation-1.39e+01

Pos logit contributions
hole+0.162
ps+0.156
els+0.149
 atop+0.146
 pieces+0.144
Neg logit contributions
Jack-0.116
His-0.109
He-0.106
 His-0.105
They-0.104
Token named
Token ID3706
Feature activation-2.12e+00

Pos logit contributions
hole+0.226
ps+0.220
 bulb+0.217
 pieces+0.215
 globe+0.214
Neg logit contributions
His-0.143
They-0.137
Jack-0.136
 His-0.130
He-0.127
Token Max
Token ID5436
Feature activation-3.95e+00

Pos logit contributions
hole+0.040
 globe+0.040
ps+0.039
 pieces+0.039
 bulb+0.039
Neg logit contributions
His-0.015
Jack-0.015
They-0.014
He-0.013
 His-0.013
Token the
Token ID262
Feature activation-3.28e+00

Pos logit contributions
hole+0.118
ool+0.116
 globe+0.114
 atop+0.111
 trunk+0.109
Neg logit contributions
They-0.029
 His-0.028
Jack-0.027
His-0.026
 him-0.025
Token cloth
Token ID16270
Feature activation-1.47e+01

Pos logit contributions
cks+0.052
bones+0.050
hole+0.045
et+0.045
els+0.045
Neg logit contributions
 please-0.035
He-0.035
His-0.033
 His-0.033
 Something-0.032
Token as
Token ID355
Feature activation+1.04e+00

Pos logit contributions
ool+0.237
hole+0.235
ps+0.230
 globe+0.229
 edges+0.225
Neg logit contributions
They-0.142
His-0.133
Jack-0.127
 His-0.126
He-0.124
Token a
Token ID257
Feature activation-6.19e-01

Pos logit contributions
They+0.009
His+0.009
Jack+0.009
 His+0.008
He+0.008
Neg logit contributions
hole-0.019
 globe-0.018
 trunk-0.018
ool-0.017
ps-0.017
Token scarf
Token ID36953
Feature activation-1.45e+01

Pos logit contributions
ps+0.009
els+0.009
cks+0.009
hole+0.008
et+0.008
Neg logit contributions
 please-0.008
His-0.008
Sure-0.007
 His-0.007
Jack-0.007
Token and
Token ID290
Feature activation+1.26e+00

Pos logit contributions
 edges+0.207
ool+0.207
hole+0.205
els+0.204
 angles+0.204
Neg logit contributions
They-0.165
His-0.159
 His-0.151
He-0.149
 please-0.149
Token wrapped
Token ID12908
Feature activation-1.11e+01

Pos logit contributions
Jack+0.016
His+0.016
They+0.015
 His+0.015
He+0.015
Neg logit contributions
hole-0.017
 bulb-0.017
 pieces-0.016
ps-0.016
 globe-0.016
Token it
Token ID340
Feature activation-1.05e+01

Pos logit contributions
hole+0.174
ool+0.171
 chim+0.166
 bulb+0.165
 trunk+0.164
Neg logit contributions
His-0.118
They-0.117
Jack-0.117
 His-0.108
He-0.105
Token around
Token ID1088
Feature activation-6.96e+00

Pos logit contributions
 chim+0.172
hole+0.163
els+0.153
 bolt+0.152
ool+0.151
Neg logit contributions
His-0.125
They-0.124
He-0.110
 His-0.108
As-0.106
Token their
Token ID511
Feature activation-4.15e+00

Pos logit contributions
hole+0.131
ool+0.125
 trunk+0.120
 bulb+0.120
ps+0.120
Neg logit contributions
Jack-0.061
His-0.060
They-0.058
 His-0.054
He-0.050
Token necks
Token ID44184
Feature activation-1.67e+01

Pos logit contributions
hole+0.051
ool+0.050
 atop+0.047
iest+0.046
ening+0.045
Neg logit contributions
His-0.058
They-0.056
 His-0.056
He-0.053
Jack-0.053
Token pe
Token ID613
Feature activation-1.36e+01

Pos logit contributions
hole+0.087
ening+0.079
ool+0.077
 compartment+0.074
 globe+0.074
Neg logit contributions
 Fro-0.060
They-0.054
 His-0.052
His-0.051
He-0.051
Tokenb
Token ID65
Feature activation-1.33e+01

Pos logit contributions
 globe+0.120
 bulb+0.112
hole+0.111
 chim+0.106
 box+0.103
Neg logit contributions
His-0.238
They-0.235
He-0.224
 His-0.212
As-0.212
Tokenbles
Token ID7689
Feature activation-1.67e+01

Pos logit contributions
hole+0.125
 bulb+0.120
ening+0.119
 globe+0.116
 pieces+0.114
Neg logit contributions
His-0.227
They-0.224
He-0.214
 His-0.212
Jack-0.211
Token!"
Token ID2474
Feature activation+1.07e+01

Pos logit contributions
 globe+0.209
hole+0.209
ening+0.206
 bulb+0.205
 angles+0.198
Neg logit contributions
They-0.231
He-0.211
 please-0.209
Jack-0.202
His-0.201
Token she
Token ID673
Feature activation+1.71e+00

Pos logit contributions
Johnny+0.173
David+0.168
Billy+0.166
Harry+0.156
Jack+0.150
Neg logit contributions
 pieces-0.121
 lid-0.111
 edges-0.109
 ladder-0.108
 hole-0.105
Token says
Token ID1139
Feature activation+8.98e-01

Pos logit contributions
Jack+0.031
His+0.030
Johnny+0.029
David+0.029
They+0.029
Neg logit contributions
ps-0.014
 pieces-0.014
hole-0.014
 bulb-0.014
 trunk-0.013
Token.
Token ID13
Feature activation+1.07e+01

Pos logit contributions
His+0.014
 His+0.014
Jack+0.014
They+0.013
He+0.013
Neg logit contributions
hole-0.010
 globe-0.010
 bulb-0.009
ool-0.009
 pieces-0.009
Token
Token ID198
Feature activation+9.07e+00

Pos logit contributions
Johnny+0.107
Jack+0.100
David+0.095
His+0.091
Billy+0.090
Neg logit contributions
ps-0.168
 ladder-0.168
 pieces-0.168
 lid-0.168
 trunk-0.167
Token
Token ID198
Feature activation+1.00e+01

Pos logit contributions
David+0.091
Johnny+0.089
Jack+0.088
 concession+0.087
Thanks+0.087
Neg logit contributions
hole-0.119
 pieces-0.119
 ladder-0.118
 globe-0.117
 bulb-0.117
Token"
Token ID1
Feature activation+7.54e+00

Pos logit contributions
Johnny+0.135
 His+0.135
David+0.128
 Johnny+0.127
 As+0.123
Neg logit contributions
 pieces-0.110
 edges-0.108
hole-0.104
 box-0.102
 ladder-0.101
TokenNo
Token ID2949
Feature activation-5.21e-01

Pos logit contributions
 His+0.070
David+0.067
ete+0.067
Jack+0.066
enny+0.065
Neg logit contributions
 globe-0.121
 bulb-0.115
 pieces-0.115
 edges-0.109
 object-0.108
Token played
Token ID2826
Feature activation-5.15e+00

Pos logit contributions
His+0.030
Jack+0.029
They+0.028
 His+0.028
He+0.027
Neg logit contributions
hole-0.031
 pieces-0.031
 bulb-0.031
ps-0.030
 globe-0.030
Token a
Token ID257
Feature activation-3.75e-01

Pos logit contributions
 pieces+0.064
ps+0.063
hole+0.062
 globe+0.062
 bulb+0.062
Neg logit contributions
Jack-0.069
His-0.069
They-0.066
David-0.065
 His-0.065
Token game
Token ID983
Feature activation-9.29e+00

Pos logit contributions
hole+0.004
ps+0.004
 globe+0.004
 bulb+0.004
 pieces+0.004
Neg logit contributions
His-0.006
Jack-0.006
They-0.006
 His-0.006
He-0.006
Token with
Token ID351
Feature activation-2.26e+00

Pos logit contributions
hole+0.130
 globe+0.127
ps+0.127
 bulb+0.127
 pieces+0.123
Neg logit contributions
His-0.115
Jack-0.114
They-0.109
 His-0.106
He-0.105
Token his
Token ID465
Feature activation-1.43e+00

Pos logit contributions
hole+0.048
 globe+0.047
 bulb+0.046
ool+0.046
ps+0.045
Neg logit contributions
His-0.014
Jack-0.014
 His-0.013
They-0.012
He-0.011
Token friend
Token ID1545
Feature activation-5.80e+00

Pos logit contributions
hole+0.022
ps+0.021
 bulb+0.021
 globe+0.021
ool+0.021
Neg logit contributions
His-0.016
Jack-0.016
 His-0.015
They-0.015
He-0.015
Token.
Token ID13
Feature activation+1.04e+01

Pos logit contributions
hole+0.097
 globe+0.096
ps+0.095
 bulb+0.095
ool+0.093
Neg logit contributions
Jack-0.056
His-0.056
 His-0.052
They-0.052
He-0.049
Token He
Token ID679
Feature activation+1.59e+00

Pos logit contributions
Jack+0.066
They+0.057
David+0.056
His+0.056
Jim+0.052
Neg logit contributions
 trunk-0.205
 bulb-0.203
 globe-0.200
 chim-0.199
ps-0.198
Token wanted
Token ID2227
Feature activation-1.90e+00

Pos logit contributions
Jack+0.018
His+0.017
They+0.016
 His+0.016
As+0.016
Neg logit contributions
hole-0.026
ps-0.025
 globe-0.025
 bulb-0.024
ool-0.024
Token to
Token ID284
Feature activation+4.08e-01

Pos logit contributions
hole+0.034
 globe+0.033
 bulb+0.033
ps+0.033
ool+0.033
Neg logit contributions
Jack-0.017
His-0.017
 His-0.016
They-0.016
He-0.015
Token win
Token ID1592
Feature activation-9.94e+00

Pos logit contributions
Jack+0.006
His+0.006
They+0.005
Un+0.005
David+0.005
Neg logit contributions
 pieces-0.005
ps-0.005
 bulb-0.005
 trunk-0.005
 ring-0.005
Token It
Token ID632
Feature activation-5.11e+00

Pos logit contributions
Jack+0.074
His+0.070
David+0.067
Jim+0.067
Her+0.064
Neg logit contributions
 trunk-0.094
hole-0.093
 bulb-0.091
 globe-0.090
 chim-0.090
Token's
Token ID338
Feature activation-2.75e+00

Pos logit contributions
hole+0.078
ool+0.075
 globe+0.074
 atop+0.071
 falling+0.071
Neg logit contributions
They-0.059
 His-0.053
As-0.052
 They-0.050
 As-0.050
Token how
Token ID703
Feature activation-1.02e+00

Pos logit contributions
hole+0.053
 globe+0.052
 pieces+0.049
ps+0.049
ening+0.048
Neg logit contributions
They-0.024
As-0.021
 please-0.020
His-0.019
Un-0.018
Token I
Token ID314
Feature activation-1.49e+00

Pos logit contributions
hole+0.013
 globe+0.013
 bulb+0.013
ps+0.013
 pieces+0.013
Neg logit contributions
His-0.013
Jack-0.013
They-0.013
 His-0.012
He-0.012
Token became
Token ID2627
Feature activation-6.56e+00

Pos logit contributions
hole+0.024
 globe+0.023
 pieces+0.023
ool+0.023
ps+0.023
Neg logit contributions
His-0.015
Jack-0.015
They-0.015
He-0.014
 His-0.014
Token rich
Token ID5527
Feature activation-1.64e+01

Pos logit contributions
 globe+0.115
ool+0.112
hole+0.111
 lid+0.104
 trunk+0.102
Neg logit contributions
They-0.062
 please-0.056
 His-0.056
He-0.055
As-0.055
Token."
Token ID526
Feature activation+9.36e+00

Pos logit contributions
ool+0.242
hole+0.239
 bulb+0.231
 globe+0.227
ps+0.225
Neg logit contributions
They-0.186
His-0.175
He-0.173
Jack-0.172
As-0.168
Token Lily
Token ID20037
Feature activation-2.44e+00

Pos logit contributions
David+0.113
Jack+0.110
Un+0.110
Jim+0.109
Harry+0.107
Neg logit contributions
 pieces-0.129
 trunk-0.127
 box-0.124
 lid-0.122
hole-0.120
Token didn
Token ID1422
Feature activation-1.64e+00

Pos logit contributions
 pieces+0.030
 bulb+0.029
hole+0.029
 globe+0.029
ps+0.029
Neg logit contributions
His-0.032
Jack-0.032
They-0.031
He-0.030
David-0.030
Token't
Token ID470
Feature activation-7.28e-01

Pos logit contributions
ps+0.021
 bulb+0.021
 globe+0.021
hole+0.021
ool+0.021
Neg logit contributions
Jack-0.021
His-0.020
They-0.019
Jim-0.019
David-0.019
Token understand
Token ID1833
Feature activation-6.74e+00

Pos logit contributions
hole+0.013
 globe+0.013
 bulb+0.013
ps+0.013
 pieces+0.013
Neg logit contributions
His-0.006
Jack-0.006
They-0.006
 His-0.006
He-0.005
Token.
Token ID13
Feature activation+1.07e+01

Pos logit contributions
 globe+0.055
 plank+0.052
 trunk+0.052
ool+0.051
 pieces+0.051
Neg logit contributions
His-0.050
They-0.048
He-0.047
Jack-0.047
 His-0.046
Token She
Token ID1375
Feature activation+1.83e+00

Pos logit contributions
His+0.091
Jack+0.087
David+0.087
Jim+0.082
Un+0.081
Neg logit contributions
 pieces-0.184
hole-0.182
 trunk-0.181
 bulb-0.180
 lid-0.178
Token looked
Token ID3114
Feature activation-2.90e+00

Pos logit contributions
His+0.022
Jack+0.022
They+0.021
 His+0.020
He+0.020
Neg logit contributions
hole-0.026
 globe-0.026
ps-0.026
 bulb-0.025
 pieces-0.025
Token in
Token ID287
Feature activation-2.98e+00

Pos logit contributions
 globe+0.039
ool+0.039
hole+0.038
ps+0.038
 bulb+0.037
Neg logit contributions
Jack-0.037
His-0.036
They-0.035
He-0.034
 His-0.034
Token the
Token ID262
Feature activation-3.39e+00

Pos logit contributions
hole+0.047
ool+0.047
 globe+0.047
ps+0.047
 bulb+0.046
Neg logit contributions
Jack-0.031
His-0.030
They-0.029
He-0.028
 His-0.028
Token mirror
Token ID10162
Feature activation-1.40e+01

Pos logit contributions
hole+0.036
ps+0.035
 globe+0.035
 bulb+0.034
ool+0.034
Neg logit contributions
His-0.053
Jack-0.053
They-0.051
 His-0.051
He-0.050
Token and
Token ID290
Feature activation+1.93e+00

Pos logit contributions
ps+0.235
ool+0.234
 globe+0.233
 bulb+0.227
 plank+0.221
Neg logit contributions
They-0.125
Jack-0.119
His-0.114
As-0.109
 As-0.109
Token smiled
Token ID13541
Feature activation-6.11e+00

Pos logit contributions
Jack+0.022
His+0.021
They+0.020
 His+0.020
He+0.019
Neg logit contributions
ool-0.030
ps-0.029
hole-0.029
 globe-0.029
 bulb-0.028
Token.
Token ID13
Feature activation+1.02e+01

Pos logit contributions
 bulb+0.035
 pieces+0.034
hole+0.032
 glowing+0.031
ps+0.031
Neg logit contributions
Jack-0.126
Jim-0.121
David-0.120
Kids-0.120
His-0.120
Token
Token ID198
Feature activation+8.37e+00

Pos logit contributions
His+0.082
Jack+0.080
David+0.074
Jim+0.072
They+0.071
Neg logit contributions
hole-0.179
 bulb-0.178
 pieces-0.178
 trunk-0.176
 globe-0.173
Token
Token ID198
Feature activation+9.48e+00

Pos logit contributions
His+0.067
Jack+0.065
David+0.061
 His+0.060
Un+0.058
Neg logit contributions
hole-0.142
 globe-0.139
 pieces-0.138
 bulb-0.137
 trunk-0.136
Token.
Token ID13
Feature activation+9.08e+00

Pos logit contributions
els+0.187
ool+0.187
 bulb+0.184
 bulbs+0.182
 globe+0.182
Neg logit contributions
They-0.133
He-0.124
His-0.123
 He-0.116
 His-0.111
Token He
Token ID679
Feature activation+1.67e-01

Pos logit contributions
Jack+0.077
Sarah+0.074
They+0.074
As+0.074
Johnny+0.072
Neg logit contributions
 lid-0.147
 pieces-0.145
 edges-0.143
 hole-0.143
 trunk-0.140
Token landed
Token ID11406
Feature activation-7.72e+00

Pos logit contributions
His+0.002
They+0.002
Jack+0.002
He+0.002
 His+0.002
Neg logit contributions
ool-0.002
hole-0.002
 globe-0.002
 bulb-0.002
ening-0.002
Token on
Token ID319
Feature activation-4.44e+00

Pos logit contributions
ool+0.097
 bulb+0.088
hole+0.087
 globe+0.085
 angles+0.081
Neg logit contributions
He-0.128
His-0.127
They-0.125
 He-0.114
Jack-0.114
Token the
Token ID262
Feature activation-3.86e+00

Pos logit contributions
 pieces+0.058
ps+0.057
hole+0.056
 globe+0.056
 bulb+0.055
Neg logit contributions
Jack-0.057
His-0.057
They-0.056
As-0.053
He-0.053
Token ground
Token ID2323
Feature activation-1.19e+01

Pos logit contributions
els+0.040
ool+0.039
hole+0.036
ening+0.036
 atop+0.034
Neg logit contributions
His-0.065
He-0.063
They-0.062
 His-0.060
Jack-0.058
Token and
Token ID290
Feature activation+9.07e-01

Pos logit contributions
ool+0.240
els+0.237
 bulb+0.233
 bulbs+0.230
 globe+0.229
Neg logit contributions
He-0.086
They-0.083
His-0.080
 He-0.078
As-0.074
Token did
Token ID750
Feature activation-1.76e+00

Pos logit contributions
His+0.012
Jack+0.011
They+0.011
 His+0.011
He+0.011
Neg logit contributions
ool-0.013
hole-0.012
 bulb-0.012
 globe-0.012
els-0.011
Token not
Token ID407
Feature activation+7.68e-01

Pos logit contributions
ps+0.028
 falling+0.027
hole+0.027
 pieces+0.027
 globe+0.026
Neg logit contributions
Jack-0.019
His-0.018
They-0.017
David-0.017
Johnny-0.017
Token move
Token ID1445
Feature activation-1.19e+01

Pos logit contributions
They+0.006
His+0.006
Jack+0.006
He+0.006
As+0.005
Neg logit contributions
hole-0.015
ool-0.015
 globe-0.014
ening-0.014
 atop-0.014
Token.
Token ID13
Feature activation+9.01e+00

Pos logit contributions
 globe+0.159
ool+0.158
hole+0.157
 bulb+0.157
 stem+0.146
Neg logit contributions
His-0.164
He-0.158
They-0.156
Jack-0.152
 His-0.152
Token dirt
Token ID13647
Feature activation-1.62e+01

Pos logit contributions
ps+0.046
ening+0.041
ool+0.040
els+0.039
hole+0.037
Neg logit contributions
He-0.079
His-0.078
Jack-0.075
 His-0.075
As-0.074
Token.
Token ID13
Feature activation+8.76e+00

Pos logit contributions
 globe+0.244
 bulb+0.243
ps+0.243
 prope+0.242
 mant+0.236
Neg logit contributions
He-0.168
They-0.165
 He-0.161
His-0.146
Jack-0.135
Token It
Token ID632
Feature activation-3.76e+00

Pos logit contributions
Jack+0.052
His+0.050
They+0.049
As+0.042
Johnny+0.041
Neg logit contributions
hole-0.174
 trunk-0.174
 pieces-0.173
 globe-0.172
ps-0.170
Token was
Token ID373
Feature activation-2.42e+00

Pos logit contributions
hole+0.068
ps+0.066
 globe+0.064
ool+0.064
ening+0.064
Neg logit contributions
His-0.034
Jack-0.032
He-0.032
 His-0.031
They-0.031
Token a
Token ID257
Feature activation-3.90e+00

Pos logit contributions
hole+0.046
 globe+0.044
ool+0.043
ening+0.043
ps+0.042
Neg logit contributions
His-0.020
 His-0.018
He-0.017
Jack-0.016
They-0.016
Token coin
Token ID10752
Feature activation-1.79e+01

Pos logit contributions
ps+0.056
els+0.052
hole+0.050
ening+0.048
cks+0.046
Neg logit contributions
His-0.055
Jack-0.055
He-0.055
 please-0.053
 His-0.050
Token!
Token ID0
Feature activation+9.08e+00

Pos logit contributions
ps+0.258
 angles+0.244
 blades+0.242
ening+0.242
els+0.238
Neg logit contributions
He-0.204
His-0.200
They-0.196
Jack-0.186
 He-0.182
Token He
Token ID679
Feature activation-1.31e-01

Pos logit contributions
Jack+0.062
His+0.060
They+0.059
As+0.052
Her+0.051
Neg logit contributions
hole-0.172
 trunk-0.171
 pieces-0.171
 globe-0.168
ool-0.167
Token put
Token ID1234
Feature activation-8.06e+00

Pos logit contributions
hole+0.002
 globe+0.002
ps+0.002
 bulb+0.002
ool+0.002
Neg logit contributions
His-0.002
Jack-0.002
They-0.001
He-0.001
 His-0.001
Token the
Token ID262
Feature activation-4.53e+00

Pos logit contributions
ool+0.172
 chim+0.168
 channel+0.164
 code+0.162
 bulb+0.161
Neg logit contributions
 him-0.056
 his-0.047
 He-0.045
His-0.044
He-0.041
Token coin
Token ID10752
Feature activation-1.80e+01

Pos logit contributions
ening+0.052
ool+0.051
els+0.046
ps+0.045
hole+0.044
Neg logit contributions
His-0.073
He-0.071
 His-0.066
 He-0.062
Jack-0.061
Token the
Token ID262
Feature activation-1.41e-01

Pos logit contributions
His+0.019
Jack+0.019
They+0.018
He+0.018
Jim+0.018
Neg logit contributions
 pieces-0.013
hole-0.013
 bulb-0.013
 globe-0.013
 trunk-0.013
Token beach
Token ID10481
Feature activation-1.04e+01

Pos logit contributions
hole+0.002
 globe+0.001
 bulb+0.001
 pieces+0.001
ps+0.001
Neg logit contributions
His-0.002
Jack-0.002
They-0.002
 His-0.002
He-0.002
Token and
Token ID290
Feature activation+3.17e+00

Pos logit contributions
ps+0.144
hole+0.142
 bulb+0.141
 globe+0.139
ool+0.139
Neg logit contributions
Jack-0.127
His-0.126
They-0.125
 His-0.118
He-0.117
Token find
Token ID1064
Feature activation-1.81e+00

Pos logit contributions
Jack+0.041
His+0.040
They+0.039
Jim+0.038
David+0.037
Neg logit contributions
 trunk-0.042
 bulb-0.041
 ladder-0.040
 ring-0.040
hole-0.040
Token sand
Token ID6450
Feature activation-1.27e+01

Pos logit contributions
hole+0.028
 globe+0.027
ps+0.027
 bulb+0.027
 pieces+0.026
Neg logit contributions
His-0.020
Jack-0.020
They-0.019
 His-0.019
He-0.018
Token dollars
Token ID5054
Feature activation-1.30e+01

Pos logit contributions
hole+0.143
ool+0.139
 bulb+0.138
ening+0.137
ps+0.137
Neg logit contributions
They-0.187
His-0.185
Jack-0.182
He-0.177
 His-0.175
Token.
Token ID13
Feature activation+1.23e+01

Pos logit contributions
 bulb+0.181
 globe+0.168
ool+0.168
ening+0.168
 chim+0.166
Neg logit contributions
They-0.171
His-0.170
 His-0.163
Jack-0.162
He-0.160
Token
Token ID220
Feature activation+6.15e+00

Pos logit contributions
Da+0.124
Billy+0.123
Harry+0.119
Sarah+0.116
Jack+0.113
Neg logit contributions
hole-0.189
 edges-0.189
 lid-0.185
 way-0.184
 trunk-0.182
Token
Token ID198
Feature activation+9.66e+00

Pos logit contributions
Jack+0.053
His+0.051
Un+0.047
Thanks+0.046
Da+0.045
Neg logit contributions
hole-0.105
 pieces-0.098
 globe-0.097
 trunk-0.096
ool-0.095
Token
Token ID198
Feature activation+1.07e+01

Pos logit contributions
Jack+0.078
Da+0.075
Harry+0.074
Un+0.074
Thanks+0.074
Neg logit contributions
hole-0.151
 pieces-0.143
 lid-0.136
ool-0.135
 trunk-0.134
TokenOn
Token ID2202
Feature activation+2.17e+00

Pos logit contributions
 Benny+0.081
 Sammy+0.081
 Billy+0.078
 Buddy+0.077
Da+0.076
Neg logit contributions
hole-0.171
 pieces-0.165
 lid-0.160
 sides-0.158
 edges-0.158