TOP ACTIVATIONS
MAX = 0.191

had ripped right off its hinges ! Lily felt bad for
door had fallen off its hinges and Jill was lying on
calling out " Hello ?". A little girl came running into
was a little girl named A ire l le . She
was a small fair . A little boy went to the
a dead wheat field . A young boy went out to
said " Free Candy ". A man with a beard and
day something magical happened . A gentle rain started pouring from
there in the garden . A three year old girl was
there was a theater . A little girl named Lucy was
in a big park . A little boy named Tim was
day something special happened . A big bird flew by and
was a little girl named A yl a . She was
, something magical happened . A big rainbow appeared in the
something very special happened . A tiny fairy appeared out of
big surprise was there . A big , fluffy bunny hopped
, something magical happened . A little girl has seen the
, something special happened . A nice man came to the
a big food contest . A kind judge was there to
key could unlock anything ! A little girl named Quick was

MIN ACTIVATIONS
MIN = -0.216

you for supplying us with panc ake mix , mom and
mom came back with the panc ake mix and they made
to come back with the panc ake mix , but she
he became a blue berry panc ake . It was not
into many pieces and the panc ake batter spilled all over
don 't you enter a panc ake cooking contest ?" Tim
not to stop stirring the panc ake batter .
and supply us with more panc ake mix . But it
where they could get more panc ake mix .
pointed to the blue berry panc ake . Mom
that they ran out of panc ake mix . Lily was
and continued to stir the panc ake batter .
. I was tired and crank y . I appreciate your
. They are tired and crank y . They say ,
. You are tired and crank y . You need to
the bag will not be mad at them . They hope
please . Do not be mad . Do not take our
. You 're tired and crank y ." S
the car and not be mad at them .
You are making my dog mad . He does not like

INTERVAL 0.089 - 0.191
CONTAINS 5.014%

They told them what they saw . " Mom
day , the hipp o saw a little bird . The
boy who was playing with a toy that she wanted to
his garden . He picked up the mel on
you can spin too !" So the little girl and the

INTERVAL -0.013 - 0.089
CONTAINS 51.001%

underneath the carpet . The puppet had a friendly face and
and put it on a leaf . She wanted to weigh
From that day on , Lily learned to be serious when
crash . One day , he saw a new
, a man named John , came to take him for

INTERVAL -0.114 - -0.013
CONTAINS 41.775%

. He says , " No , you go away .
poke her teeth . It hurt and tasted bad .
again . He kept jumping up and down until he was
J enny 's friends were very excited when
. She thought for a moment and smiled .

INTERVAL -0.216 - -0.114
CONTAINS 2.209%

, but she made her unhappy . She said she was
a carrot . He gives it to L ila . He
hurt your lip . Or it might stick to something else
bitter . It 's worth it ," the doctor said .
and she wanted to try it . She reached
Token had
Token ID550
Feature activation+8.13e-02

Pos logit contributions
 cardboard+0.068
lf+0.064
 camps+0.062
ts+0.062
 fossils+0.055
Neg logit contributions
eal-0.135
arre-0.129
umpy-0.124
irk-0.121
 back-0.121
Token ripped
Token ID19551
Feature activation+2.39e-02

Pos logit contributions
lf+0.110
 cardboard+0.107
ts+0.100
 camps+0.096
 trek+0.089
Neg logit contributions
eal-0.162
arre-0.156
umpy-0.147
irk-0.145
vious-0.142
Token right
Token ID826
Feature activation-1.22e-01

Pos logit contributions
lf+0.040
 cardboard+0.038
ts+0.037
 camps+0.036
 trek+0.036
Neg logit contributions
eal-0.038
arre-0.037
 back-0.036
umpy-0.035
irk-0.034
Token off
Token ID572
Feature activation-4.58e-02

Pos logit contributions
eal+0.229
arre+0.218
irk+0.206
umpy+0.206
ags+0.200
Neg logit contributions
 cardboard-0.183
lf-0.181
ts-0.165
 camps-0.160
 trek-0.156
Token its
Token ID663
Feature activation-6.32e-02

Pos logit contributions
eal+0.076
arre+0.074
 back+0.072
umpy+0.072
irk+0.069
Neg logit contributions
 cardboard-0.069
lf-0.068
ts-0.065
 camps-0.065
 trek-0.062
Token hinges
Token ID41442
Feature activation+1.91e-01

Pos logit contributions
eal+0.105
arre+0.102
irk+0.094
umpy+0.093
vious+0.088
Neg logit contributions
 cardboard-0.111
lf-0.107
ts-0.101
 camps-0.097
 blasted-0.095
Token!
Token ID0
Feature activation-2.84e-02

Pos logit contributions
lf+0.304
ts+0.302
 cardboard+0.295
 camps+0.275
 trek+0.270
Neg logit contributions
arre-0.314
eal-0.311
umpy-0.299
 back-0.291
irk-0.279
Token Lily
Token ID20037
Feature activation+6.18e-03

Pos logit contributions
eal+0.055
arre+0.052
irk+0.049
umpy+0.049
awk+0.049
Neg logit contributions
 cardboard-0.042
lf-0.040
ts-0.035
 camps-0.034
 pretend-0.033
Token felt
Token ID2936
Feature activation+2.21e-02

Pos logit contributions
 cardboard+0.009
 camps+0.008
lf+0.008
ts+0.008
 fossils+0.008
Neg logit contributions
eal-0.012
arre-0.012
umpy-0.011
 back-0.011
 disappointed-0.010
Token bad
Token ID2089
Feature activation-8.83e-02

Pos logit contributions
 cardboard+0.036
ts+0.033
lf+0.033
 pretend+0.032
 shovel+0.030
Neg logit contributions
eal-0.039
arre-0.039
irk-0.035
umpy-0.035
awk-0.034
Token for
Token ID329
Feature activation-6.82e-02

Pos logit contributions
eal+0.169
arre+0.162
umpy+0.151
irk+0.149
ags+0.147
Neg logit contributions
 cardboard-0.132
lf-0.125
ts-0.116
 camps-0.112
 pretend-0.110
Token door
Token ID3420
Feature activation+2.43e-02

Pos logit contributions
 cardboard+0.097
 camps+0.093
ts+0.089
 trek+0.086
lf+0.085
Neg logit contributions
eal-0.142
arre-0.137
 back-0.135
umpy-0.135
irk-0.129
Token had
Token ID550
Feature activation+8.03e-02

Pos logit contributions
 camps+0.036
 cardboard+0.034
 seas+0.034
ts+0.033
 fossils+0.033
Neg logit contributions
 back-0.041
eal-0.039
 disappointed-0.039
umpy-0.037
 mad-0.037
Token fallen
Token ID9292
Feature activation+4.21e-02

Pos logit contributions
lf+0.116
 cardboard+0.104
 camps+0.104
ts+0.104
 trek+0.095
Neg logit contributions
eal-0.142
 back-0.140
arre-0.134
umpy-0.130
ags-0.128
Token off
Token ID572
Feature activation-4.54e-02

Pos logit contributions
lf+0.084
ts+0.078
 cardboard+0.076
 trek+0.072
 fossils+0.072
Neg logit contributions
 back-0.056
eal-0.053
arre-0.053
umpy-0.051
ags-0.049
Token its
Token ID663
Feature activation-6.81e-02

Pos logit contributions
 back+0.068
eal+0.066
arre+0.065
umpy+0.065
irk+0.063
Neg logit contributions
lf-0.072
ts-0.070
 cardboard-0.070
 trek-0.067
 blasted-0.066
Token hinges
Token ID41442
Feature activation+1.87e-01

Pos logit contributions
eal+0.126
arre+0.121
umpy+0.116
irk+0.114
ags+0.111
Neg logit contributions
lf-0.099
 cardboard-0.097
ts-0.093
 camps-0.086
 blasted-0.084
Token and
Token ID290
Feature activation-2.68e-02

Pos logit contributions
lf+0.301
ts+0.292
 cardboard+0.288
 camps+0.268
 trek+0.268
Neg logit contributions
arre-0.305
eal-0.302
umpy-0.294
 back-0.284
irk-0.277
Token Jill
Token ID22772
Feature activation+1.19e-02

Pos logit contributions
eal+0.046
 back+0.045
arre+0.044
 disappointed+0.042
 mad+0.042
Neg logit contributions
lf-0.038
 cardboard-0.037
 camps-0.036
ts-0.034
 trek-0.034
Token was
Token ID373
Feature activation+2.74e-02

Pos logit contributions
 cardboard+0.019
lf+0.018
 camps+0.017
ts+0.017
 trek+0.016
Neg logit contributions
eal-0.021
arre-0.020
umpy-0.019
irk-0.019
 back-0.018
Token lying
Token ID9105
Feature activation-3.05e-03

Pos logit contributions
 cardboard+0.044
lf+0.043
ts+0.040
 camps+0.039
 trek+0.038
Neg logit contributions
eal-0.048
arre-0.047
umpy-0.044
irk-0.043
ags-0.041
Token on
Token ID319
Feature activation-1.85e-02

Pos logit contributions
eal+0.006
arre+0.006
umpy+0.005
irk+0.005
vious+0.005
Neg logit contributions
 cardboard-0.004
lf-0.004
ts-0.004
 camps-0.004
 trek-0.004
Token calling
Token ID4585
Feature activation-1.27e-02

Pos logit contributions
 cardboard+0.053
lf+0.052
ts+0.049
 camps+0.046
 blasted+0.045
Neg logit contributions
eal-0.075
arre-0.073
umpy-0.069
irk-0.068
ags-0.065
Token out
Token ID503
Feature activation-1.55e-02

Pos logit contributions
eal+0.027
arre+0.027
umpy+0.025
irk+0.025
vious+0.024
Neg logit contributions
 cardboard-0.016
lf-0.016
 called-0.013
 pretend-0.013
ts-0.013
Token "
Token ID366
Feature activation+1.41e-03

Pos logit contributions
eal+0.033
arre+0.032
irk+0.030
umpy+0.030
agree+0.029
Neg logit contributions
 cardboard-0.020
lf-0.019
ts-0.017
 camps-0.017
 pretend-0.016
TokenHello
Token ID15496
Feature activation+4.18e-02

Pos logit contributions
lf+0.002
 cardboard+0.002
ts+0.002
 blasted+0.002
 trek+0.002
Neg logit contributions
eal-0.002
arre-0.002
irk-0.002
 fault-0.002
ags-0.002
Token?".
Token ID43634
Feature activation-2.98e-02

Pos logit contributions
 cardboard+0.056
lf+0.055
ts+0.050
 camps+0.048
 blasted+0.046
Neg logit contributions
eal-0.085
arre-0.082
umpy-0.077
irk-0.077
ags-0.074
Token A
Token ID317
Feature activation+1.85e-01

Pos logit contributions
eal+0.057
arre+0.056
umpy+0.051
irk+0.050
ags+0.050
Neg logit contributions
 cardboard-0.045
lf-0.044
 called-0.038
ts-0.038
 pretend-0.038
Token little
Token ID1310
Feature activation+1.16e-01

Pos logit contributions
 cardboard+0.281
lf+0.276
ts+0.258
 camps+0.246
 blasted+0.238
Neg logit contributions
eal-0.341
arre-0.330
umpy-0.308
irk-0.308
ags-0.290
Token girl
Token ID2576
Feature activation+3.85e-02

Pos logit contributions
 cardboard+0.207
lf+0.198
ts+0.183
 camps+0.179
 shovel+0.178
Neg logit contributions
eal-0.187
arre-0.185
umpy-0.167
ags-0.165
irk-0.163
Token came
Token ID1625
Feature activation+2.58e-02

Pos logit contributions
 cardboard+0.048
lf+0.045
ts+0.044
 camps+0.042
 seas+0.038
Neg logit contributions
eal-0.082
arre-0.077
irk-0.074
umpy-0.073
 disappointed-0.071
Token running
Token ID2491
Feature activation-8.16e-03

Pos logit contributions
 cardboard+0.045
lf+0.043
 trek+0.041
 camps+0.041
 called+0.040
Neg logit contributions
eal-0.042
arre-0.040
irk-0.037
umpy-0.036
vious-0.035
Token into
Token ID656
Feature activation+4.63e-02

Pos logit contributions
arre+0.016
eal+0.016
ags+0.015
umpy+0.015
ealing+0.015
Neg logit contributions
 cardboard-0.011
lf-0.010
 pretend-0.010
 called-0.010
 pretended-0.010
Token was
Token ID373
Feature activation+3.68e-02

Pos logit contributions
 cardboard+0.101
ts+0.095
lf+0.094
 swords+0.088
 camps+0.088
Neg logit contributions
eal-0.109
arre-0.108
umpy-0.105
 back-0.103
irk-0.101
Token a
Token ID257
Feature activation+1.63e-01

Pos logit contributions
lf+0.057
ts+0.053
 blasted+0.049
 cardboard+0.047
 fossils+0.046
Neg logit contributions
eal-0.064
arre-0.064
ags-0.063
umpy-0.060
 back-0.060
Token little
Token ID1310
Feature activation+1.09e-01

Pos logit contributions
 cardboard+0.232
lf+0.230
ts+0.229
 trek+0.211
 blasted+0.210
Neg logit contributions
eal-0.296
umpy-0.280
arre-0.280
irk-0.270
 back-0.260
Token girl
Token ID2576
Feature activation+3.40e-02

Pos logit contributions
 cardboard+0.208
lf+0.208
ts+0.197
 camps+0.188
 trek+0.187
Neg logit contributions
eal-0.155
arre-0.147
umpy-0.137
irk-0.137
awk-0.125
Token named
Token ID3706
Feature activation+1.03e-01

Pos logit contributions
ts+0.040
 loot+0.040
 camps+0.039
 fossils+0.039
 cardboard+0.038
Neg logit contributions
 back-0.064
umpy-0.064
eal-0.063
irk-0.062
 disappointed-0.060
Token A
Token ID317
Feature activation+1.82e-01

Pos logit contributions
lf+0.174
 cardboard+0.173
ts+0.166
 camps+0.157
 blasted+0.157
Neg logit contributions
eal-0.171
arre-0.163
umpy-0.153
irk-0.149
 back-0.145
Tokenire
Token ID557
Feature activation+2.67e-02

Pos logit contributions
 cardboard+0.232
lf+0.227
ts+0.210
 camps+0.196
 blasted+0.191
Neg logit contributions
eal-0.380
arre-0.368
umpy-0.348
irk-0.347
vious-0.328
Tokenl
Token ID75
Feature activation-3.69e-02

Pos logit contributions
 cardboard+0.041
 camps+0.037
lf+0.036
 trek+0.036
ts+0.036
Neg logit contributions
umpy-0.046
eal-0.046
arre-0.045
irk-0.043
itter-0.041
Tokenle
Token ID293
Feature activation-4.58e-02

Pos logit contributions
eal+0.060
arre+0.059
umpy+0.059
irk+0.056
ags+0.055
Neg logit contributions
 cardboard-0.061
lf-0.057
 camps-0.054
ts-0.054
 shovel-0.052
Token.
Token ID13
Feature activation-9.39e-03

Pos logit contributions
umpy+0.072
eal+0.069
arre+0.065
itter+0.065
 back+0.065
Neg logit contributions
 cardboard-0.074
 dice-0.069
 fossils-0.066
 swords-0.066
ts-0.065
Token She
Token ID1375
Feature activation+6.63e-02

Pos logit contributions
eal+0.010
arre+0.010
irk+0.009
umpy+0.009
ags+0.008
Neg logit contributions
 cardboard-0.021
lf-0.021
ts-0.020
 camps-0.020
 blasted-0.019
Token was
Token ID373
Feature activation+4.02e-02

Pos logit contributions
 cardboard+0.095
lf+0.094
ts+0.092
 blasted+0.081
 camps+0.080
Neg logit contributions
eal-0.142
arre-0.139
 back-0.131
umpy-0.131
irk-0.130
Token a
Token ID257
Feature activation+1.67e-01

Pos logit contributions
lf+0.061
ts+0.056
 blasted+0.051
 cardboard+0.050
 fossils+0.047
Neg logit contributions
eal-0.075
arre-0.073
ags-0.070
awk-0.068
vious-0.068
Token small
Token ID1402
Feature activation+8.08e-02

Pos logit contributions
ts+0.255
 cardboard+0.251
lf+0.242
 blasted+0.238
 trek+0.230
Neg logit contributions
eal-0.281
umpy-0.280
arre-0.268
irk-0.263
 back-0.249
Token fair
Token ID3148
Feature activation-4.43e-02

Pos logit contributions
lf+0.146
 cardboard+0.146
ts+0.139
 blasted+0.131
 camps+0.131
Neg logit contributions
eal-0.123
arre-0.117
umpy-0.110
irk-0.110
awk-0.101
Token.
Token ID13
Feature activation-1.45e-02

Pos logit contributions
eal+0.069
arre+0.064
umpy+0.063
irk+0.062
awk+0.057
Neg logit contributions
 cardboard-0.078
lf-0.076
ts-0.076
 blasted-0.070
 trek-0.070
Token A
Token ID317
Feature activation+1.82e-01

Pos logit contributions
eal+0.017
umpy+0.017
irk+0.017
arre+0.016
agree+0.015
Neg logit contributions
ts-0.030
lf-0.028
 cardboard-0.027
 blasted-0.026
 camps-0.025
Token little
Token ID1310
Feature activation+1.13e-01

Pos logit contributions
 cardboard+0.239
lf+0.234
ts+0.226
 camps+0.206
 trek+0.204
Neg logit contributions
eal-0.364
arre-0.348
umpy-0.336
irk-0.333
 back-0.311
Token boy
Token ID2933
Feature activation+5.18e-02

Pos logit contributions
 cardboard+0.202
lf+0.191
ts+0.176
 camps+0.174
 shovel+0.174
Neg logit contributions
eal-0.185
arre-0.184
umpy-0.163
irk-0.162
ags-0.159
Token went
Token ID1816
Feature activation+4.44e-02

Pos logit contributions
 camps+0.064
ts+0.063
 cardboard+0.061
 miles+0.060
lf+0.058
Neg logit contributions
umpy-0.098
eal-0.097
irk-0.092
 back-0.090
 disappointed-0.089
Token to
Token ID284
Feature activation+3.33e-04

Pos logit contributions
 cardboard+0.075
lf+0.070
 camps+0.066
ts+0.065
 trek+0.064
Neg logit contributions
eal-0.078
arre-0.076
umpy-0.070
irk-0.070
ealing-0.067
Token the
Token ID262
Feature activation-3.16e-02

Pos logit contributions
lf+0.001
 cardboard+0.000
ts+0.000
 camps+0.000
 blasted+0.000
Neg logit contributions
eal-0.001
arre-0.001
 back-0.001
irk-0.000
ags-0.000
Token a
Token ID257
Feature activation+1.65e-01

Pos logit contributions
lf+0.060
ts+0.058
 blasted+0.053
 cardboard+0.050
 hunt+0.050
Neg logit contributions
arre-0.062
eal-0.061
ags-0.059
 back-0.058
umpy-0.057
Token dead
Token ID2636
Feature activation-5.70e-03

Pos logit contributions
ts+0.235
 cardboard+0.233
lf+0.232
 trek+0.213
 blasted+0.212
Neg logit contributions
eal-0.300
arre-0.284
umpy-0.281
irk-0.274
 back-0.265
Token wheat
Token ID17135
Feature activation+1.25e-02

Pos logit contributions
eal+0.011
arre+0.010
umpy+0.010
irk+0.010
ags+0.009
Neg logit contributions
 cardboard-0.009
lf-0.008
ts-0.008
 camps-0.008
 blasted-0.007
Token field
Token ID2214
Feature activation+5.01e-02

Pos logit contributions
 cardboard+0.017
lf+0.016
ts+0.015
 camps+0.015
 blasted+0.014
Neg logit contributions
eal-0.025
arre-0.024
umpy-0.023
irk-0.022
ags-0.021
Token.
Token ID13
Feature activation-1.27e-02

Pos logit contributions
 cardboard+0.067
lf+0.066
ts+0.061
 camps+0.058
 blasted+0.056
Neg logit contributions
eal-0.101
arre-0.098
umpy-0.092
irk-0.092
ags-0.088
Token A
Token ID317
Feature activation+1.81e-01

Pos logit contributions
eal+0.017
arre+0.016
umpy+0.015
irk+0.015
ags+0.014
Neg logit contributions
 cardboard-0.026
lf-0.026
ts-0.024
 camps-0.024
 blasted-0.023
Token young
Token ID1862
Feature activation+6.03e-02

Pos logit contributions
 cardboard+0.213
lf+0.208
ts+0.189
 camps+0.177
 blasted+0.170
Neg logit contributions
eal-0.396
arre-0.385
umpy-0.364
irk-0.364
ags-0.347
Token boy
Token ID2933
Feature activation+4.84e-02

Pos logit contributions
 cardboard+0.105
lf+0.100
ts+0.091
 shovel+0.089
 camps+0.089
Neg logit contributions
eal-0.103
arre-0.102
ags-0.093
irk-0.091
umpy-0.089
Token went
Token ID1816
Feature activation+4.37e-02

Pos logit contributions
 camps+0.064
ts+0.062
lf+0.062
 cardboard+0.061
 dice+0.058
Neg logit contributions
eal-0.084
umpy-0.084
crow-0.080
 back-0.080
vious-0.080
Token out
Token ID503
Feature activation-4.72e-03

Pos logit contributions
 cardboard+0.067
 called+0.058
lf+0.058
 hunt+0.057
 camps+0.056
Neg logit contributions
eal-0.087
arre-0.083
irk-0.080
umpy-0.078
ealing-0.077
Token to
Token ID284
Feature activation+2.64e-03

Pos logit contributions
eal+0.008
arre+0.007
umpy+0.007
irk+0.007
ags+0.006
Neg logit contributions
 cardboard-0.008
lf-0.008
ts-0.008
 camps-0.007
 trek-0.007
Token said
Token ID531
Feature activation-3.29e-02

Pos logit contributions
 cardboard+0.005
lf+0.005
ts+0.004
 pretend+0.004
 shovel+0.004
Neg logit contributions
eal-0.009
arre-0.009
irk-0.008
umpy-0.008
ags-0.008
Token "
Token ID366
Feature activation-3.03e-03

Pos logit contributions
eal+0.080
arre+0.078
umpy+0.076
irk+0.075
ags+0.072
Neg logit contributions
lf-0.034
 cardboard-0.034
ts-0.029
 pretend-0.027
 called-0.026
TokenFree
Token ID11146
Feature activation-9.51e-03

Pos logit contributions
eal+0.006
ags+0.005
arre+0.005
ealing+0.005
 fault+0.005
Neg logit contributions
lf-0.004
 cardboard-0.004
 pretended-0.004
 pretend-0.003
 blasted-0.003
Token Candy
Token ID24680
Feature activation+5.97e-03

Pos logit contributions
eal+0.015
irk+0.014
arre+0.014
umpy+0.013
vious+0.013
Neg logit contributions
lf-0.017
 cardboard-0.017
ts-0.016
 pretend-0.015
 camps-0.015
Token".
Token ID1911
Feature activation-3.15e-02

Pos logit contributions
 cardboard+0.008
lf+0.008
ts+0.007
 camps+0.007
 pretend+0.007
Neg logit contributions
eal-0.012
arre-0.012
irk-0.011
umpy-0.011
ags-0.011
Token A
Token ID317
Feature activation+1.80e-01

Pos logit contributions
eal+0.062
arre+0.060
umpy+0.056
vious+0.055
ags+0.055
Neg logit contributions
 cardboard-0.040
lf-0.040
 called-0.038
 pretend-0.035
 named-0.035
Token man
Token ID582
Feature activation-8.77e-04

Pos logit contributions
 cardboard+0.265
lf+0.260
ts+0.240
 camps+0.229
 blasted+0.222
Neg logit contributions
eal-0.342
arre-0.331
umpy-0.309
irk-0.309
ags-0.295
Token with
Token ID351
Feature activation-1.71e-02

Pos logit contributions
eal+0.002
arre+0.002
irk+0.002
 disappointed+0.001
umpy+0.001
Neg logit contributions
 cardboard-0.001
 camps-0.001
lf-0.001
ts-0.001
 seas-0.001
Token a
Token ID257
Feature activation+1.62e-01

Pos logit contributions
eal+0.037
arre+0.036
irk+0.035
umpy+0.034
vious+0.034
Neg logit contributions
lf-0.020
 cardboard-0.018
ts-0.018
 camps-0.017
 trek-0.016
Token beard
Token ID21213
Feature activation+6.61e-02

Pos logit contributions
 cardboard+0.219
lf+0.200
 shovel+0.179
 pretend+0.173
 camps+0.170
Neg logit contributions
eal-0.341
arre-0.330
ags-0.317
umpy-0.315
vious-0.309
Token and
Token ID290
Feature activation-1.14e-02

Pos logit contributions
 cardboard+0.091
lf+0.086
ts+0.082
 camps+0.080
 trek+0.075
Neg logit contributions
eal-0.130
arre-0.125
umpy-0.118
irk-0.118
vious-0.113
Token day
Token ID1110
Feature activation-3.55e-03

Pos logit contributions
eal+0.069
arre+0.068
irk+0.063
umpy+0.063
ags+0.060
Neg logit contributions
 cardboard-0.046
lf-0.045
ts-0.041
 camps-0.039
 trek-0.038
Token something
Token ID1223
Feature activation+3.34e-02

Pos logit contributions
eal+0.008
arre+0.008
irk+0.007
umpy+0.007
ags+0.007
Neg logit contributions
 cardboard-0.004
lf-0.004
ts-0.003
 camps-0.003
 trek-0.003
Token magical
Token ID10883
Feature activation+1.99e-02

Pos logit contributions
 cardboard+0.046
lf+0.044
ts+0.041
 camps+0.038
 pretend+0.038
Neg logit contributions
eal-0.069
arre-0.067
irk-0.063
umpy-0.062
ags-0.061
Token happened
Token ID3022
Feature activation+4.06e-03

Pos logit contributions
 cardboard+0.027
lf+0.026
ts+0.024
 camps+0.023
 blasted+0.022
Neg logit contributions
eal-0.040
arre-0.039
umpy-0.036
irk-0.036
vious-0.035
Token.
Token ID13
Feature activation-1.49e-02

Pos logit contributions
 cardboard+0.005
lf+0.005
ts+0.004
 camps+0.004
 pretend+0.004
Neg logit contributions
eal-0.009
arre-0.009
irk-0.008
umpy-0.008
ags-0.008
Token A
Token ID317
Feature activation+1.80e-01

Pos logit contributions
eal+0.027
arre+0.025
irk+0.024
umpy+0.023
ags+0.023
Neg logit contributions
 cardboard-0.024
lf-0.023
 camps-0.021
 trek-0.021
ts-0.020
Token gentle
Token ID10296
Feature activation-5.80e-02

Pos logit contributions
 cardboard+0.264
lf+0.260
ts+0.252
 camps+0.235
 blasted+0.227
Neg logit contributions
eal-0.328
arre-0.319
umpy-0.301
irk-0.300
vious-0.279
Token rain
Token ID6290
Feature activation-1.32e-02

Pos logit contributions
arre+0.106
eal+0.104
ags+0.098
umpy+0.095
Fighting+0.092
Neg logit contributions
 cardboard-0.094
lf-0.093
 trek-0.079
 pretend-0.078
ts-0.077
Token started
Token ID2067
Feature activation+2.95e-02

Pos logit contributions
eal+0.027
arre+0.026
irk+0.025
ags+0.025
umpy+0.025
Neg logit contributions
lf-0.017
 cardboard-0.017
ts-0.015
 blasted-0.015
 pretended-0.014
Token pouring
Token ID23147
Feature activation-2.83e-02

Pos logit contributions
 cardboard+0.047
lf+0.044
ts+0.041
 blasted+0.040
 trek+0.040
Neg logit contributions
eal-0.056
arre-0.052
irk-0.050
umpy-0.049
ags-0.048
Token from
Token ID422
Feature activation-4.46e-03

Pos logit contributions
eal+0.055
arre+0.052
irk+0.051
umpy+0.049
agree+0.046
Neg logit contributions
 cardboard-0.043
lf-0.038
ts-0.038
 camps-0.036
 shovel-0.035
Token there
Token ID612
Feature activation+6.40e-02

Pos logit contributions
eal+0.033
arre+0.032
irk+0.030
umpy+0.030
vious+0.029
Neg logit contributions
 cardboard-0.019
lf-0.018
ts-0.017
 camps-0.016
 blasted-0.015
Token in
Token ID287
Feature activation+3.74e-02

Pos logit contributions
 cardboard+0.087
lf+0.083
ts+0.077
 camps+0.073
 blasted+0.072
Neg logit contributions
eal-0.125
arre-0.123
umpy-0.114
irk-0.113
 back-0.109
Token the
Token ID262
Feature activation-2.96e-02

Pos logit contributions
lf+0.055
ts+0.052
 blasted+0.052
 cardboard+0.051
 camps+0.050
Neg logit contributions
eal-0.058
 back-0.057
arre-0.056
irk-0.056
ags-0.055
Token garden
Token ID11376
Feature activation+6.46e-02

Pos logit contributions
eal+0.070
arre+0.066
irk+0.064
umpy+0.061
ealing+0.060
Neg logit contributions
 cardboard-0.036
lf-0.031
 camps-0.030
 pretend-0.027
 seas-0.026
Token.
Token ID13
Feature activation-1.43e-02

Pos logit contributions
 cardboard+0.101
lf+0.099
ts+0.094
 camps+0.090
 blasted+0.088
Neg logit contributions
eal-0.108
arre-0.106
umpy-0.102
irk-0.096
 back-0.096
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.017
arre+0.016
umpy+0.015
irk+0.014
agree+0.014
Neg logit contributions
lf-0.030
 cardboard-0.030
ts-0.029
 camps-0.028
 blasted-0.027
Token three
Token ID1115
Feature activation+1.12e-01

Pos logit contributions
 cardboard+0.226
lf+0.221
ts+0.219
 camps+0.197
 trek+0.197
Neg logit contributions
eal-0.365
arre-0.351
umpy-0.340
irk-0.334
 back-0.318
Token year
Token ID614
Feature activation+1.33e-02

Pos logit contributions
 cardboard+0.182
 camps+0.169
 blasted+0.168
ts+0.165
 trek+0.164
Neg logit contributions
eal-0.198
umpy-0.184
arre-0.180
irk-0.173
vious-0.164
Token old
Token ID1468
Feature activation+1.60e-02

Pos logit contributions
 cardboard+0.024
lf+0.024
ts+0.021
 camps+0.020
 called+0.020
Neg logit contributions
eal-0.023
arre-0.022
irk-0.020
ags-0.019
Fighting-0.019
Token girl
Token ID2576
Feature activation+3.21e-02

Pos logit contributions
 cardboard+0.029
ts+0.029
lf+0.029
 camps+0.028
 blasted+0.027
Neg logit contributions
eal-0.023
arre-0.021
irk-0.021
umpy-0.020
 back-0.019
Token was
Token ID373
Feature activation+3.86e-02

Pos logit contributions
ts+0.038
 camps+0.037
lf+0.035
 cardboard+0.035
 seas+0.032
Neg logit contributions
eal-0.064
 back-0.060
irk-0.059
umpy-0.059
 disappointed-0.058
Token there
Token ID612
Feature activation+6.83e-02

Pos logit contributions
 cardboard+0.086
lf+0.086
ts+0.079
 trek+0.078
 camps+0.076
Neg logit contributions
arre-0.128
eal-0.127
umpy-0.124
 back-0.122
ags-0.117
Token was
Token ID373
Feature activation+3.71e-02

Pos logit contributions
 cardboard+0.105
ts+0.099
lf+0.098
 swords+0.092
 fossils+0.092
Neg logit contributions
arre-0.116
eal-0.114
umpy-0.109
 back-0.107
irk-0.104
Token a
Token ID257
Feature activation+1.64e-01

Pos logit contributions
lf+0.058
ts+0.057
 blasted+0.052
 hunt+0.050
 cardboard+0.050
Neg logit contributions
arre-0.063
eal-0.061
ags-0.059
 back-0.059
umpy-0.057
Token theater
Token ID13766
Feature activation+5.71e-02

Pos logit contributions
ts+0.246
 cardboard+0.241
lf+0.240
 trek+0.231
 blasted+0.225
Neg logit contributions
eal-0.279
umpy-0.266
arre-0.264
irk-0.255
 back-0.250
Token.
Token ID13
Feature activation-1.37e-02

Pos logit contributions
 cardboard+0.083
ts+0.077
 camps+0.073
lf+0.073
 trek+0.072
Neg logit contributions
eal-0.106
arre-0.100
umpy-0.100
irk-0.095
 back-0.093
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.018
arre+0.017
umpy+0.016
irk+0.016
agree+0.015
Neg logit contributions
ts-0.027
 cardboard-0.027
lf-0.026
 camps-0.025
 trek-0.024
Token little
Token ID1310
Feature activation+1.11e-01

Pos logit contributions
 cardboard+0.291
ts+0.290
lf+0.272
 trek+0.267
 camps+0.260
Neg logit contributions
eal-0.292
umpy-0.274
arre-0.267
irk-0.256
 back-0.254
Token girl
Token ID2576
Feature activation+3.45e-02

Pos logit contributions
 cardboard+0.210
lf+0.206
ts+0.194
 camps+0.188
 blasted+0.184
Neg logit contributions
eal-0.166
arre-0.161
umpy-0.146
irk-0.146
ags-0.137
Token named
Token ID3706
Feature activation+1.04e-01

Pos logit contributions
ts+0.045
 camps+0.042
 swords+0.040
 cardboard+0.040
 loot+0.038
Neg logit contributions
eal-0.067
umpy-0.065
 back-0.063
irk-0.062
itter-0.061
Token Lucy
Token ID22162
Feature activation+1.75e-02

Pos logit contributions
 cardboard+0.182
lf+0.179
ts+0.174
 camps+0.167
 trek+0.161
Neg logit contributions
eal-0.169
arre-0.158
umpy-0.152
irk-0.147
 back-0.142
Token was
Token ID373
Feature activation+4.33e-02

Pos logit contributions
 cardboard+0.028
lf+0.027
 camps+0.026
 seas+0.026
 dice+0.025
Neg logit contributions
eal-0.030
umpy-0.028
 back-0.027
arre-0.026
irk-0.026
Token in
Token ID287
Feature activation+3.78e-02

Pos logit contributions
 cardboard+0.072
ts+0.069
 camps+0.066
 loot+0.065
 trek+0.065
Neg logit contributions
umpy-0.066
eal-0.063
arre-0.062
vious-0.059
 back-0.059
Token a
Token ID257
Feature activation+1.62e-01

Pos logit contributions
lf+0.060
 camps+0.060
ts+0.059
 cardboard+0.059
 loot+0.059
Neg logit contributions
eal-0.048
 back-0.048
agree-0.048
ags-0.047
arre-0.047
Token big
Token ID1263
Feature activation+7.40e-02

Pos logit contributions
 cardboard+0.200
lf+0.198
ts+0.185
 camps+0.171
 trek+0.166
Neg logit contributions
eal-0.341
arre-0.333
umpy-0.315
irk-0.311
vious-0.296
Token park
Token ID3952
Feature activation+4.62e-02

Pos logit contributions
 cardboard+0.094
lf+0.091
ts+0.081
 camps+0.078
 blasted+0.075
Neg logit contributions
eal-0.159
arre-0.154
irk-0.145
umpy-0.145
ags-0.139
Token.
Token ID13
Feature activation-1.46e-02

Pos logit contributions
 cardboard+0.077
ts+0.073
lf+0.072
 camps+0.071
 trek+0.067
Neg logit contributions
eal-0.074
umpy-0.070
arre-0.067
irk-0.065
 back-0.065
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.017
arre+0.015
umpy+0.015
irk+0.014
vious+0.014
Neg logit contributions
lf-0.031
ts-0.031
 cardboard-0.031
 camps-0.030
 blasted-0.028
Token little
Token ID1310
Feature activation+1.06e-01

Pos logit contributions
 cardboard+0.237
lf+0.232
ts+0.217
 camps+0.204
 trek+0.197
Neg logit contributions
eal-0.364
arre-0.352
umpy-0.334
irk-0.331
vious-0.313
Token boy
Token ID2933
Feature activation+4.88e-02

Pos logit contributions
 cardboard+0.179
lf+0.168
ts+0.155
 shovel+0.153
 camps+0.151
Neg logit contributions
eal-0.185
arre-0.184
irk-0.164
umpy-0.163
ags-0.161
Token named
Token ID3706
Feature activation+1.01e-01

Pos logit contributions
 camps+0.067
ts+0.063
 cardboard+0.060
 miles+0.058
 trek+0.058
Neg logit contributions
umpy-0.086
eal-0.085
 back-0.080
crow-0.079
irk-0.079
Token Tim
Token ID5045
Feature activation+3.88e-02

Pos logit contributions
lf+0.194
ts+0.183
 cardboard+0.178
 hours+0.171
 blasted+0.168
Neg logit contributions
 back-0.131
eal-0.125
umpy-0.107
ags-0.106
agree-0.104
Token was
Token ID373
Feature activation+4.11e-02

Pos logit contributions
 cardboard+0.071
lf+0.071
ts+0.066
 camps+0.064
 blasted+0.063
Neg logit contributions
eal-0.057
umpy-0.055
arre-0.054
irk-0.051
 back-0.050
Token day
Token ID1110
Feature activation-7.20e-03

Pos logit contributions
 cardboard+0.149
lf+0.146
ts+0.132
 camps+0.129
 trek+0.127
Neg logit contributions
eal-0.169
arre-0.165
irk-0.151
umpy-0.148
ags-0.145
Token something
Token ID1223
Feature activation+2.93e-02

Pos logit contributions
eal+0.012
arre+0.011
 back+0.011
ealing+0.011
 mad+0.010
Neg logit contributions
lf-0.011
 cardboard-0.010
ts-0.010
 camps-0.010
 blasted-0.009
Token special
Token ID2041
Feature activation-5.93e-02

Pos logit contributions
 cardboard+0.044
lf+0.043
ts+0.039
 camps+0.037
 blasted+0.036
Neg logit contributions
eal-0.056
arre-0.055
irk-0.051
umpy-0.051
ags-0.049
Token happened
Token ID3022
Feature activation+1.60e-03

Pos logit contributions
eal+0.100
arre+0.095
irk+0.089
umpy+0.089
vious+0.087
Neg logit contributions
 cardboard-0.095
lf-0.092
ts-0.090
 camps-0.088
 trek-0.083
Token.
Token ID13
Feature activation-1.22e-02

Pos logit contributions
 cardboard+0.002
lf+0.002
ts+0.002
 camps+0.002
 blasted+0.002
Neg logit contributions
eal-0.003
arre-0.003
irk-0.003
umpy-0.003
ags-0.003
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.018
arre+0.017
umpy+0.016
irk+0.016
vious+0.015
Neg logit contributions
lf-0.023
 cardboard-0.023
ts-0.022
 camps-0.021
 blasted-0.020
Token big
Token ID1263
Feature activation+7.49e-02

Pos logit contributions
 cardboard+0.236
lf+0.232
ts+0.220
 camps+0.205
 trek+0.197
Neg logit contributions
eal-0.360
arre-0.348
umpy-0.330
irk-0.329
vious-0.310
Token bird
Token ID6512
Feature activation+1.95e-02

Pos logit contributions
 cardboard+0.100
lf+0.098
ts+0.090
 camps+0.085
 blasted+0.082
Neg logit contributions
eal-0.153
arre-0.148
umpy-0.139
irk-0.139
ags-0.132
Token flew
Token ID13112
Feature activation+6.40e-02

Pos logit contributions
 cardboard+0.025
lf+0.024
ts+0.024
 camps+0.024
 fossils+0.023
Neg logit contributions
eal-0.037
arre-0.036
vious-0.036
umpy-0.035
awk-0.035
Token by
Token ID416
Feature activation-6.00e-02

Pos logit contributions
 cardboard+0.107
 called+0.096
 blasted+0.093
 trek+0.092
lf+0.092
Neg logit contributions
eal-0.111
arre-0.111
irk-0.101
umpy-0.098
 fault-0.096
Token and
Token ID290
Feature activation-8.44e-03

Pos logit contributions
eal+0.098
arre+0.095
umpy+0.092
irk+0.090
 back+0.090
Neg logit contributions
lf-0.096
 cardboard-0.095
ts-0.088
 camps-0.088
 blasted-0.084
Token was
Token ID373
Feature activation+3.68e-02

Pos logit contributions
 cardboard+0.099
ts+0.094
lf+0.090
 swords+0.088
 camps+0.088
Neg logit contributions
eal-0.106
arre-0.103
umpy-0.098
 back-0.098
irk-0.097
Token a
Token ID257
Feature activation+1.62e-01

Pos logit contributions
lf+0.064
ts+0.060
 blasted+0.057
 hunt+0.053
 cardboard+0.051
Neg logit contributions
arre-0.055
eal-0.053
ags-0.053
 back-0.053
vious-0.052
Token little
Token ID1310
Feature activation+1.04e-01

Pos logit contributions
ts+0.239
 cardboard+0.237
lf+0.236
 trek+0.217
 blasted+0.216
Neg logit contributions
eal-0.282
arre-0.267
umpy-0.266
irk-0.260
 back-0.249
Token girl
Token ID2576
Feature activation+3.18e-02

Pos logit contributions
lf+0.201
 cardboard+0.201
ts+0.191
 camps+0.182
 trek+0.181
Neg logit contributions
eal-0.146
arre-0.138
irk-0.128
umpy-0.128
awk-0.116
Token named
Token ID3706
Feature activation+9.81e-02

Pos logit contributions
 loot+0.044
ts+0.043
 fossils+0.041
 cardboard+0.041
 dice+0.041
Neg logit contributions
 back-0.056
umpy-0.054
irk-0.053
eal-0.053
olding-0.052
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
 cardboard+0.160
lf+0.158
ts+0.150
 camps+0.143
 blasted+0.140
Neg logit contributions
eal-0.172
arre-0.165
umpy-0.155
irk-0.153
ags-0.145
Tokenyl
Token ID2645
Feature activation-2.37e-02

Pos logit contributions
 cardboard+0.241
lf+0.236
ts+0.217
 camps+0.206
 blasted+0.199
Neg logit contributions
eal-0.361
arre-0.351
umpy-0.329
irk-0.329
ags-0.313
Tokena
Token ID64
Feature activation+2.08e-02

Pos logit contributions
eal+0.033
arre+0.032
umpy+0.031
irk+0.029
vious+0.028
Neg logit contributions
 cardboard-0.046
lf-0.042
 camps-0.042
ts-0.042
 shovel-0.041
Token.
Token ID13
Feature activation-9.71e-03

Pos logit contributions
 cardboard+0.029
ts+0.026
 camps+0.026
 trek+0.025
 blasted+0.025
Neg logit contributions
eal-0.038
umpy-0.038
irk-0.035
arre-0.035
 back-0.035
Token She
Token ID1375
Feature activation+6.59e-02

Pos logit contributions
eal+0.014
arre+0.014
umpy+0.012
irk+0.012
ealing+0.012
Neg logit contributions
 cardboard-0.019
lf-0.018
 camps-0.017
ts-0.017
 pretend-0.017
Token was
Token ID373
Feature activation+3.75e-02

Pos logit contributions
 cardboard+0.101
 camps+0.093
lf+0.091
 shovel+0.089
ts+0.087
Neg logit contributions
eal-0.117
umpy-0.113
arre-0.109
irk-0.103
 back-0.101
Token,
Token ID11
Feature activation+7.54e-02

Pos logit contributions
 cardboard+0.002
lf+0.002
ts+0.002
 camps+0.002
 blasted+0.002
Neg logit contributions
eal-0.004
arre-0.004
umpy-0.004
irk-0.004
 back-0.004
Token something
Token ID1223
Feature activation+3.48e-02

Pos logit contributions
 cardboard+0.120
lf+0.117
ts+0.114
 camps+0.106
 blasted+0.105
Neg logit contributions
eal-0.124
arre-0.119
 back-0.110
irk-0.108
umpy-0.108
Token magical
Token ID10883
Feature activation+1.73e-02

Pos logit contributions
 cardboard+0.048
lf+0.046
ts+0.043
 pretend+0.041
 called+0.040
Neg logit contributions
arre-0.071
eal-0.071
irk-0.066
umpy-0.065
ags-0.065
Token happened
Token ID3022
Feature activation+6.50e-03

Pos logit contributions
 cardboard+0.027
lf+0.026
ts+0.025
 camps+0.025
 trek+0.023
Neg logit contributions
eal-0.029
arre-0.027
irk-0.026
umpy-0.026
vious-0.026
Token.
Token ID13
Feature activation-1.16e-02

Pos logit contributions
 cardboard+0.008
lf+0.008
ts+0.007
 camps+0.007
 trek+0.007
Neg logit contributions
eal-0.014
arre-0.014
irk-0.013
umpy-0.013
ags-0.012
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.020
arre+0.020
umpy+0.018
irk+0.018
ags+0.017
Neg logit contributions
 cardboard-0.019
lf-0.018
ts-0.017
 camps-0.016
 blasted-0.016
Token big
Token ID1263
Feature activation+7.86e-02

Pos logit contributions
 cardboard+0.251
lf+0.250
ts+0.247
 camps+0.227
 trek+0.220
Neg logit contributions
eal-0.328
arre-0.319
umpy-0.304
irk-0.303
 back-0.283
Token rainbow
Token ID27223
Feature activation+4.02e-02

Pos logit contributions
 cardboard+0.111
lf+0.110
ts+0.104
 camps+0.099
 trek+0.095
Neg logit contributions
eal-0.150
arre-0.145
irk-0.137
umpy-0.136
vious-0.129
Token appeared
Token ID4120
Feature activation+6.40e-02

Pos logit contributions
 cardboard+0.050
ts+0.048
lf+0.047
 camps+0.047
 trek+0.044
Neg logit contributions
eal-0.079
arre-0.076
umpy-0.072
vious-0.071
irk-0.071
Token in
Token ID287
Feature activation+3.91e-02

Pos logit contributions
lf+0.098
 cardboard+0.096
ts+0.092
 camps+0.082
 blasted+0.080
Neg logit contributions
eal-0.114
arre-0.112
umpy-0.106
 back-0.105
irk-0.104
Token the
Token ID262
Feature activation-3.13e-02

Pos logit contributions
lf+0.067
ts+0.067
 cardboard+0.062
 blasted+0.061
 trek+0.061
Neg logit contributions
 back-0.053
arre-0.050
eal-0.050
 aback-0.049
ags-0.048
Token something
Token ID1223
Feature activation+3.20e-02

Pos logit contributions
 camps+0.123
ts+0.122
 trek+0.122
 hunt+0.118
lf+0.114
Neg logit contributions
 back-0.074
 mad-0.065
 disappointed-0.062
 Fighting-0.061
eal-0.061
Token very
Token ID845
Feature activation-3.62e-02

Pos logit contributions
 cardboard+0.049
lf+0.048
ts+0.044
 pretend+0.041
 blasted+0.041
Neg logit contributions
eal-0.060
arre-0.060
irk-0.055
umpy-0.055
ags-0.053
Token special
Token ID2041
Feature activation-5.74e-02

Pos logit contributions
arre+0.067
eal+0.067
irk+0.061
ags+0.059
umpy+0.059
Neg logit contributions
 cardboard-0.056
lf-0.055
ts-0.050
 pretend-0.047
 camps-0.047
Token happened
Token ID3022
Feature activation+1.47e-03

Pos logit contributions
eal+0.092
arre+0.086
irk+0.083
umpy+0.082
vious+0.080
Neg logit contributions
 cardboard-0.093
ts-0.091
 camps-0.090
lf-0.088
 trek-0.084
Token.
Token ID13
Feature activation-1.46e-02

Pos logit contributions
 cardboard+0.002
lf+0.002
ts+0.002
 camps+0.002
 blasted+0.002
Neg logit contributions
eal-0.003
arre-0.003
irk-0.003
umpy-0.003
ags-0.003
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.021
arre+0.020
umpy+0.019
irk+0.019
agree+0.018
Neg logit contributions
lf-0.027
 cardboard-0.027
ts-0.026
 camps-0.025
 blasted-0.024
Token tiny
Token ID7009
Feature activation+9.51e-02

Pos logit contributions
 cardboard+0.239
ts+0.239
lf+0.235
 camps+0.219
 trek+0.214
Neg logit contributions
eal-0.343
arre-0.340
umpy-0.320
irk-0.315
 back-0.296
Token fairy
Token ID25607
Feature activation+4.24e-02

Pos logit contributions
lf+0.152
 cardboard+0.152
ts+0.149
 camps+0.141
 trek+0.138
Neg logit contributions
eal-0.162
arre-0.156
irk-0.147
umpy-0.147
vious-0.137
Token appeared
Token ID4120
Feature activation+6.28e-02

Pos logit contributions
 camps+0.058
ts+0.056
 cardboard+0.054
 fossils+0.053
 dice+0.053
Neg logit contributions
eal-0.073
 back-0.070
vious-0.068
arre-0.067
umpy-0.066
Token out
Token ID503
Feature activation-5.16e-03

Pos logit contributions
 cardboard+0.088
lf+0.088
ts+0.081
 camps+0.076
 blasted+0.074
Neg logit contributions
eal-0.121
arre-0.118
umpy-0.112
irk-0.111
ags-0.106
Token of
Token ID286
Feature activation+2.65e-02

Pos logit contributions
eal+0.007
umpy+0.007
arre+0.007
irk+0.007
vious+0.007
Neg logit contributions
lf-0.009
 cardboard-0.009
ts-0.009
 camps-0.008
 shovel-0.008
Token big
Token ID1263
Feature activation+7.76e-02

Pos logit contributions
 cardboard+0.222
lf+0.217
ts+0.199
 camps+0.190
 blasted+0.183
Neg logit contributions
eal-0.320
arre-0.310
umpy-0.291
irk-0.291
ags-0.277
Token surprise
Token ID5975
Feature activation+1.11e-02

Pos logit contributions
 cardboard+0.119
lf+0.115
ts+0.104
 camps+0.101
 blasted+0.097
Neg logit contributions
eal-0.146
arre-0.142
umpy-0.133
irk-0.131
ags-0.125
Token was
Token ID373
Feature activation+3.76e-02

Pos logit contributions
 cardboard+0.017
lf+0.016
ts+0.016
 camps+0.015
 blasted+0.014
Neg logit contributions
eal-0.020
arre-0.020
irk-0.018
umpy-0.018
ags-0.017
Token there
Token ID612
Feature activation+7.03e-02

Pos logit contributions
 cardboard+0.048
lf+0.047
ts+0.047
 camps+0.043
 blasted+0.042
Neg logit contributions
eal-0.071
arre-0.069
 back-0.066
irk-0.065
umpy-0.065
Token.
Token ID13
Feature activation-1.84e-02

Pos logit contributions
 cardboard+0.095
lf+0.093
ts+0.086
 camps+0.081
 blasted+0.079
Neg logit contributions
eal-0.142
arre-0.137
umpy-0.129
irk-0.129
ags-0.123
Token A
Token ID317
Feature activation+1.79e-01

Pos logit contributions
eal+0.031
arre+0.030
umpy+0.028
irk+0.028
ags+0.027
Neg logit contributions
 cardboard-0.030
lf-0.030
ts-0.028
 camps-0.027
 trek-0.026
Token big
Token ID1263
Feature activation+7.68e-02

Pos logit contributions
ts+0.273
lf+0.258
 cardboard+0.257
 camps+0.239
 trek+0.238
Neg logit contributions
eal-0.305
arre-0.293
irk-0.280
umpy-0.278
 back-0.273
Token,
Token ID11
Feature activation+6.19e-02

Pos logit contributions
 cardboard+0.109
lf+0.108
ts+0.104
 camps+0.098
 trek+0.095
Neg logit contributions
eal-0.146
arre-0.140
irk-0.132
umpy-0.131
vious-0.124
Token fluffy
Token ID39145
Feature activation-5.89e-04

Pos logit contributions
 cardboard+0.086
lf+0.085
ts+0.079
 camps+0.075
 blasted+0.073
Neg logit contributions
eal-0.121
arre-0.117
umpy-0.110
irk-0.110
ags-0.104
Token bunny
Token ID44915
Feature activation+4.60e-02

Pos logit contributions
eal+0.001
arre+0.001
irk+0.001
umpy+0.001
ags+0.001
Neg logit contributions
 cardboard-0.001
lf-0.001
ts-0.001
 camps-0.001
 blasted-0.001
Token hopped
Token ID45230
Feature activation+6.28e-02

Pos logit contributions
 cardboard+0.058
lf+0.058
ts+0.058
 camps+0.053
 fossils+0.048
Neg logit contributions
eal-0.090
arre-0.084
umpy-0.083
irk-0.082
 back-0.081
Token,
Token ID11
Feature activation+7.47e-02

Pos logit contributions
eal+0.005
arre+0.005
umpy+0.004
irk+0.004
awk+0.004
Neg logit contributions
 cardboard-0.002
lf-0.002
ts-0.002
 camps-0.002
 blasted-0.002
Token something
Token ID1223
Feature activation+3.41e-02

Pos logit contributions
 cardboard+0.111
lf+0.109
ts+0.108
 camps+0.103
 blasted+0.102
Neg logit contributions
eal-0.118
arre-0.115
 back-0.112
 mad-0.104
Fighting-0.103
Token magical
Token ID10883
Feature activation+1.75e-02

Pos logit contributions
 cardboard+0.047
lf+0.044
ts+0.042
 pretend+0.040
 called+0.038
Neg logit contributions
eal-0.072
arre-0.072
irk-0.066
ags-0.065
umpy-0.065
Token happened
Token ID3022
Feature activation+7.13e-03

Pos logit contributions
 cardboard+0.025
lf+0.025
ts+0.024
 camps+0.023
 trek+0.022
Neg logit contributions
eal-0.032
arre-0.030
umpy-0.029
irk-0.029
vious-0.028
Token.
Token ID13
Feature activation-1.24e-02

Pos logit contributions
 cardboard+0.009
lf+0.009
ts+0.008
 camps+0.007
 trek+0.007
Neg logit contributions
eal-0.016
arre-0.015
irk-0.014
umpy-0.014
ags-0.014
Token A
Token ID317
Feature activation+1.78e-01

Pos logit contributions
eal+0.021
arre+0.020
irk+0.019
umpy+0.019
ags+0.018
Neg logit contributions
 cardboard-0.021
lf-0.020
ts-0.019
 camps-0.018
 trek-0.018
Token little
Token ID1310
Feature activation+1.13e-01

Pos logit contributions
 cardboard+0.252
lf+0.251
ts+0.250
 camps+0.229
 trek+0.222
Neg logit contributions
eal-0.323
arre-0.314
umpy-0.300
irk-0.298
 back-0.280
Token girl
Token ID2576
Feature activation+3.54e-02

Pos logit contributions
 cardboard+0.192
lf+0.191
ts+0.179
 camps+0.171
 blasted+0.167
Neg logit contributions
eal-0.187
arre-0.180
irk-0.167
umpy-0.167
vious-0.156
Token has
Token ID468
Feature activation+5.52e-02

Pos logit contributions
ts+0.048
 cardboard+0.047
lf+0.046
 camps+0.045
 fossils+0.041
Neg logit contributions
eal-0.068
irk-0.061
arre-0.061
 back-0.061
 disappointed-0.061
Token seen
Token ID1775
Feature activation+7.99e-02

Pos logit contributions
lf+0.067
 cardboard+0.067
ts+0.061
 camps+0.059
 blasted+0.054
Neg logit contributions
eal-0.116
arre-0.113
umpy-0.108
irk-0.106
vious-0.104
Token the
Token ID262
Feature activation-3.22e-02

Pos logit contributions
 cardboard+0.107
lf+0.106
ts+0.099
 camps+0.096
 blasted+0.093
Neg logit contributions
eal-0.152
arre-0.149
umpy-0.140
irk-0.139
 back-0.135
Token,
Token ID11
Feature activation+7.02e-02

Pos logit contributions
arre+0.180
eal+0.176
umpy+0.172
ags+0.171
irk+0.171
Neg logit contributions
lf-0.073
 cardboard-0.073
ts-0.068
 pretended-0.068
 called-0.062
Token something
Token ID1223
Feature activation+3.07e-02

Pos logit contributions
 cardboard+0.093
lf+0.091
ts+0.082
 camps+0.079
 trek+0.075
Neg logit contributions
eal-0.146
arre-0.140
irk-0.133
umpy-0.132
ags-0.127
Token special
Token ID2041
Feature activation-4.40e-02

Pos logit contributions
 cardboard+0.048
lf+0.046
ts+0.043
 camps+0.040
 pretend+0.039
Neg logit contributions
eal-0.058
arre-0.056
irk-0.052
umpy-0.052
ags-0.050
Token happened
Token ID3022
Feature activation+1.05e-02

Pos logit contributions
eal+0.084
arre+0.081
umpy+0.076
irk+0.076
vious+0.073
Neg logit contributions
 cardboard-0.062
lf-0.062
ts-0.058
 camps-0.055
 trek-0.053
Token.
Token ID13
Feature activation-1.60e-02

Pos logit contributions
 cardboard+0.012
lf+0.012
ts+0.010
 pretend+0.010
 camps+0.009
Neg logit contributions
eal-0.024
arre-0.023
irk-0.022
umpy-0.022
ags-0.021
Token A
Token ID317
Feature activation+1.78e-01

Pos logit contributions
eal+0.028
arre+0.027
irk+0.025
umpy+0.025
ags+0.024
Neg logit contributions
 cardboard-0.026
lf-0.026
 camps-0.023
ts-0.023
 trek-0.023
Token nice
Token ID3621
Feature activation-5.03e-02

Pos logit contributions
 cardboard+0.258
lf+0.254
ts+0.240
 camps+0.226
 blasted+0.219
Neg logit contributions
eal-0.337
arre-0.326
umpy-0.306
irk-0.306
vious-0.287
Token man
Token ID582
Feature activation+1.51e-03

Pos logit contributions
eal+0.090
arre+0.089
umpy+0.082
ags+0.081
irk+0.080
Neg logit contributions
 cardboard-0.081
lf-0.079
ts-0.069
 camps-0.068
 shovel-0.068
Token came
Token ID1625
Feature activation+2.37e-02

Pos logit contributions
 cardboard+0.002
lf+0.002
ts+0.002
 camps+0.002
 seas+0.002
Neg logit contributions
eal-0.003
arre-0.003
irk-0.003
 back-0.003
umpy-0.003
Token to
Token ID284
Feature activation+2.36e-03

Pos logit contributions
 cardboard+0.035
lf+0.032
 trek+0.030
 camps+0.030
ts+0.030
Neg logit contributions
eal-0.046
arre-0.044
irk-0.041
umpy-0.041
ags-0.039
Token the
Token ID262
Feature activation-3.37e-02

Pos logit contributions
 cardboard+0.003
lf+0.003
ts+0.003
 camps+0.003
 blasted+0.003
Neg logit contributions
eal-0.004
arre-0.004
irk-0.004
umpy-0.004
ags-0.004
Token a
Token ID257
Feature activation+1.62e-01

Pos logit contributions
ts+0.073
lf+0.068
 blasted+0.066
 loot+0.065
 hunt+0.064
Neg logit contributions
 back-0.051
arre-0.046
vious-0.046
eal-0.046
umpy-0.045
Token big
Token ID1263
Feature activation+7.29e-02

Pos logit contributions
ts+0.240
 cardboard+0.236
lf+0.235
 trek+0.218
 blasted+0.215
Neg logit contributions
eal-0.283
arre-0.274
umpy-0.267
irk-0.258
 back-0.250
Token food
Token ID2057
Feature activation+1.26e-02

Pos logit contributions
 cardboard+0.100
lf+0.099
ts+0.091
 camps+0.086
 blasted+0.084
Neg logit contributions
eal-0.145
arre-0.140
umpy-0.131
irk-0.131
ags-0.125
Token contest
Token ID8414
Feature activation+2.75e-02

Pos logit contributions
ts+0.019
 cardboard+0.018
lf+0.018
 trek+0.017
 blasted+0.017
Neg logit contributions
eal-0.021
arre-0.021
umpy-0.021
irk-0.021
 back-0.020
Token.
Token ID13
Feature activation-1.54e-02

Pos logit contributions
 cardboard+0.045
ts+0.043
lf+0.042
 camps+0.041
 trek+0.039
Neg logit contributions
eal-0.045
arre-0.043
umpy-0.043
irk-0.040
 back-0.039
Token A
Token ID317
Feature activation+1.78e-01

Pos logit contributions
eal+0.023
arre+0.023
umpy+0.021
irk+0.021
vious+0.020
Neg logit contributions
lf-0.028
 cardboard-0.027
ts-0.027
 camps-0.025
 trek-0.025
Token kind
Token ID1611
Feature activation-5.00e-02

Pos logit contributions
 cardboard+0.264
lf+0.258
ts+0.248
 camps+0.233
 trek+0.229
Neg logit contributions
eal-0.328
arre-0.319
umpy-0.300
irk-0.295
vious-0.276
Token judge
Token ID5052
Feature activation+2.30e-02

Pos logit contributions
arre+0.081
eal+0.081
ags+0.071
Fighting+0.071
agree+0.070
Neg logit contributions
 cardboard-0.089
lf-0.088
ts-0.075
 camps-0.075
 shovel-0.075
Token was
Token ID373
Feature activation+3.78e-02

Pos logit contributions
 camps+0.028
 cardboard+0.027
ts+0.026
lf+0.025
 trek+0.025
Neg logit contributions
eal-0.046
umpy-0.044
arre-0.044
vious-0.043
irk-0.042
Token there
Token ID612
Feature activation+6.48e-02

Pos logit contributions
 cardboard+0.051
lf+0.050
ts+0.046
 camps+0.045
 trek+0.043
Neg logit contributions
eal-0.075
arre-0.073
umpy-0.069
irk-0.068
ags-0.065
Token to
Token ID284
Feature activation+2.78e-03

Pos logit contributions
 cardboard+0.101
lf+0.100
ts+0.093
 camps+0.090
 blasted+0.087
Neg logit contributions
eal-0.112
arre-0.110
umpy-0.104
irk-0.100
 back-0.097
Token key
Token ID1994
Feature activation+3.97e-02

Pos logit contributions
ts+0.048
lf+0.046
 camps+0.045
 cardboard+0.045
 pretended+0.043
Neg logit contributions
 back-0.064
eal-0.062
 cheek-0.058
umpy-0.057
 disappointed-0.056
Token could
Token ID714
Feature activation+1.01e-01

Pos logit contributions
 camps+0.057
ts+0.054
 cardboard+0.054
 trek+0.050
lf+0.050
Neg logit contributions
eal-0.068
 back-0.067
umpy-0.066
 disappointed-0.064
olding-0.061
Token unlock
Token ID12116
Feature activation+7.63e-02

Pos logit contributions
 camps+0.126
 cardboard+0.126
lf+0.121
ts+0.119
 hours+0.111
Neg logit contributions
eal-0.189
 back-0.179
umpy-0.176
agree-0.176
 disappointed-0.171
Token anything
Token ID1997
Feature activation-4.79e-03

Pos logit contributions
lf+0.120
 camps+0.118
 cardboard+0.117
 trek+0.112
 blasted+0.111
Neg logit contributions
 back-0.115
eal-0.108
ags-0.102
 disappointed-0.100
arre-0.100
Token!
Token ID0
Feature activation-1.39e-02

Pos logit contributions
 back+0.008
 disappointed+0.007
umpy+0.007
eal+0.007
 forward+0.007
Neg logit contributions
lf-0.007
 camps-0.007
 cardboard-0.007
 trek-0.007
ts-0.007
Token A
Token ID317
Feature activation+1.78e-01

Pos logit contributions
eal+0.020
arre+0.019
umpy+0.018
irk+0.018
ags+0.017
Neg logit contributions
lf-0.026
 cardboard-0.026
ts-0.025
 camps-0.024
 blasted-0.023
Token little
Token ID1310
Feature activation+1.03e-01

Pos logit contributions
ts+0.315
 camps+0.276
lf+0.275
 trek+0.275
 cardboard+0.266
Neg logit contributions
 back-0.256
eal-0.250
umpy-0.245
 cheek-0.242
irk-0.235
Token girl
Token ID2576
Feature activation+3.22e-02

Pos logit contributions
lf+0.222
ts+0.213
 cardboard+0.208
 camps+0.205
 trek+0.201
Neg logit contributions
eal-0.121
arre-0.107
irk-0.107
umpy-0.104
 back-0.101
Token named
Token ID3706
Feature activation+9.72e-02

Pos logit contributions
 camps+0.049
ts+0.048
 swords+0.043
 seas+0.043
 cardboard+0.043
Neg logit contributions
eal-0.054
 back-0.053
umpy-0.050
 disappointed-0.049
irk-0.049
Token Quick
Token ID12029
Feature activation+1.02e-01

Pos logit contributions
lf+0.180
 cardboard+0.180
 camps+0.177
ts+0.174
 trek+0.166
Neg logit contributions
eal-0.138
arre-0.122
umpy-0.122
 back-0.122
itter-0.118
Token was
Token ID373
Feature activation+3.78e-02

Pos logit contributions
 cardboard+0.150
lf+0.148
ts+0.135
 camps+0.133
 blasted+0.127
Neg logit contributions
eal-0.195
arre-0.185
irk-0.175
umpy-0.175
vious-0.168
Token you
Token ID345
Feature activation-1.05e-01

Pos logit contributions
eal+0.051
arre+0.050
umpy+0.047
irk+0.046
vious+0.044
Neg logit contributions
lf-0.037
 cardboard-0.036
ts-0.034
 camps-0.030
 pretend-0.030
Token for
Token ID329
Feature activation-6.94e-02

Pos logit contributions
eal+0.231
arre+0.225
irk+0.213
umpy+0.212
vious+0.204
Neg logit contributions
 cardboard-0.126
lf-0.122
 camps-0.105
ts-0.104
 blasted-0.099
Token supplying
Token ID28099
Feature activation-8.35e-03

Pos logit contributions
eal+0.125
arre+0.123
irk+0.111
ags+0.111
umpy+0.111
Neg logit contributions
 cardboard-0.103
lf-0.100
 camps-0.096
ts-0.093
 trek-0.091
Token us
Token ID514
Feature activation-1.28e-01

Pos logit contributions
eal+0.014
arre+0.014
umpy+0.013
irk+0.013
vious+0.012
Neg logit contributions
 cardboard-0.014
lf-0.014
ts-0.013
 camps-0.013
 pretend-0.012
Token with
Token ID351
Feature activation-2.82e-02

Pos logit contributions
eal+0.282
arre+0.270
irk+0.269
umpy+0.257
vious+0.248
Neg logit contributions
 cardboard-0.158
lf-0.153
ts-0.145
 camps-0.133
 pretend-0.129
Token panc
Token ID22154
Feature activation-2.16e-01

Pos logit contributions
eal+0.051
arre+0.050
umpy+0.046
irk+0.046
ags+0.045
Neg logit contributions
lf-0.042
 cardboard-0.042
ts-0.039
 camps-0.037
 blasted-0.037
Tokenake
Token ID539
Feature activation-6.51e-02

Pos logit contributions
eal+0.272
arre+0.249
irk+0.237
umpy+0.235
vious+0.218
Neg logit contributions
lf-0.455
 cardboard-0.454
ts-0.423
 camps-0.409
 trek-0.401
Token mix
Token ID5022
Feature activation-3.84e-02

Pos logit contributions
eal+0.137
irk+0.132
arre+0.132
umpy+0.127
vious+0.122
Neg logit contributions
lf-0.087
 cardboard-0.084
ts-0.077
 camps-0.072
 pretend-0.070
Token,
Token ID11
Feature activation+5.50e-02

Pos logit contributions
eal+0.080
arre+0.077
irk+0.074
umpy+0.073
agree+0.069
Neg logit contributions
 cardboard-0.051
lf-0.049
ts-0.046
 camps-0.043
 pretend-0.040
Token mom
Token ID1995
Feature activation-3.98e-02

Pos logit contributions
 cardboard+0.070
lf+0.068
 camps+0.059
 pretend+0.059
 pretended+0.057
Neg logit contributions
eal-0.117
arre-0.114
umpy-0.111
irk-0.110
ags-0.107
Token and
Token ID290
Feature activation-1.66e-02

Pos logit contributions
eal+0.074
arre+0.070
irk+0.067
ags+0.065
agree+0.065
Neg logit contributions
 cardboard-0.060
lf-0.059
 camps-0.053
 pretended-0.052
ts-0.052
Token mom
Token ID1995
Feature activation-3.88e-02

Pos logit contributions
eal+0.019
arre+0.018
irk+0.017
 back+0.017
 mad+0.016
Neg logit contributions
 cardboard-0.019
lf-0.018
ts-0.018
 camps-0.017
 trek-0.016
Token came
Token ID1625
Feature activation+1.59e-02

Pos logit contributions
eal+0.068
arre+0.066
irk+0.059
 back+0.058
umpy+0.058
Neg logit contributions
 cardboard-0.065
lf-0.064
ts-0.061
 camps-0.057
 fossils-0.056
Token back
Token ID736
Feature activation-1.65e-01

Pos logit contributions
 cardboard+0.028
lf+0.026
ts+0.025
 camps+0.025
 pretend+0.024
Neg logit contributions
eal-0.026
arre-0.025
umpy-0.024
irk-0.023
vious-0.023
Token with
Token ID351
Feature activation-2.73e-02

Pos logit contributions
eal+0.303
arre+0.293
umpy+0.273
irk+0.272
vious+0.258
Neg logit contributions
 cardboard-0.258
lf-0.252
ts-0.234
 camps-0.226
 trek-0.218
Token the
Token ID262
Feature activation-3.97e-02

Pos logit contributions
eal+0.054
arre+0.052
irk+0.050
 back+0.049
ags+0.049
Neg logit contributions
lf-0.037
 cardboard-0.034
 blasted-0.032
 camps-0.032
ts-0.031
Token panc
Token ID22154
Feature activation-2.15e-01

Pos logit contributions
eal+0.074
arre+0.071
irk+0.067
umpy+0.067
ags+0.063
Neg logit contributions
lf-0.059
 cardboard-0.059
ts-0.055
 camps-0.052
 blasted-0.051
Tokenake
Token ID539
Feature activation-6.50e-02

Pos logit contributions
eal+0.291
arre+0.264
irk+0.250
umpy+0.247
vious+0.239
Neg logit contributions
lf-0.448
 cardboard-0.439
ts-0.400
 camps-0.390
 trek-0.384
Token mix
Token ID5022
Feature activation-3.93e-02

Pos logit contributions
eal+0.144
arre+0.141
irk+0.134
umpy+0.133
ags+0.128
Neg logit contributions
 cardboard-0.080
lf-0.079
ts-0.067
 camps-0.061
 pretend-0.060
Token and
Token ID290
Feature activation-1.28e-02

Pos logit contributions
eal+0.088
arre+0.086
umpy+0.082
irk+0.082
agree+0.079
Neg logit contributions
 cardboard-0.045
ts-0.041
lf-0.041
 shovel-0.035
 called-0.035
Token they
Token ID484
Feature activation-1.86e-02

Pos logit contributions
eal+0.026
arre+0.025
irk+0.023
umpy+0.023
 back+0.022
Neg logit contributions
 cardboard-0.016
lf-0.016
 camps-0.015
ts-0.015
 trek-0.014
Token made
Token ID925
Feature activation-2.15e-02

Pos logit contributions
eal+0.036
arre+0.035
 back+0.032
 disappointed+0.031
irk+0.031
Neg logit contributions
 cardboard-0.026
lf-0.026
 camps-0.023
ts-0.023
 fossils-0.022
Token to
Token ID284
Feature activation-3.42e-04

Pos logit contributions
eal+0.050
arre+0.048
umpy+0.043
irk+0.043
ags+0.041
Neg logit contributions
 cardboard-0.063
lf-0.062
ts-0.059
 camps-0.057
 blasted-0.056
Token come
Token ID1282
Feature activation-7.68e-02

Pos logit contributions
eal+0.001
arre+0.001
umpy+0.001
irk+0.001
ags+0.001
Neg logit contributions
 cardboard-0.001
lf-0.001
ts-0.000
 camps-0.000
 blasted-0.000
Token back
Token ID736
Feature activation-1.67e-01

Pos logit contributions
eal+0.108
arre+0.103
irk+0.094
umpy+0.094
vious+0.087
Neg logit contributions
 cardboard-0.154
lf-0.150
ts-0.142
 camps-0.139
 blasted-0.135
Token with
Token ID351
Feature activation-2.52e-02

Pos logit contributions
eal+0.301
arre+0.292
umpy+0.273
irk+0.271
 back+0.262
Neg logit contributions
 cardboard-0.257
lf-0.256
ts-0.243
 camps-0.222
 blasted-0.222
Token the
Token ID262
Feature activation-3.96e-02

Pos logit contributions
eal+0.045
arre+0.043
irk+0.042
ags+0.041
 back+0.040
Neg logit contributions
lf-0.038
 cardboard-0.037
 blasted-0.035
ts-0.034
 camps-0.033
Token panc
Token ID22154
Feature activation-2.13e-01

Pos logit contributions
eal+0.077
arre+0.075
irk+0.072
umpy+0.070
agree+0.068
Neg logit contributions
 cardboard-0.058
lf-0.055
 camps-0.052
ts-0.051
 pretend-0.049
Tokenake
Token ID539
Feature activation-6.24e-02

Pos logit contributions
eal+0.295
arre+0.267
irk+0.259
umpy+0.253
vious+0.245
Neg logit contributions
lf-0.433
 cardboard-0.418
 camps-0.383
ts-0.376
 trek-0.368
Token mix
Token ID5022
Feature activation-3.60e-02

Pos logit contributions
eal+0.132
arre+0.128
irk+0.122
umpy+0.121
ags+0.116
Neg logit contributions
 cardboard-0.081
lf-0.080
ts-0.071
 camps-0.067
 blasted-0.064
Token,
Token ID11
Feature activation+5.72e-02

Pos logit contributions
eal+0.079
arre+0.078
irk+0.075
umpy+0.074
agree+0.072
Neg logit contributions
 cardboard-0.041
lf-0.041
ts-0.038
 camps-0.033
 pretend-0.033
Token but
Token ID475
Feature activation-1.35e-03

Pos logit contributions
 cardboard+0.064
lf+0.061
ts+0.055
 pretended+0.054
 pretend+0.053
Neg logit contributions
eal-0.130
arre-0.127
umpy-0.123
irk-0.123
awk-0.117
Token she
Token ID673
Feature activation+1.29e-04

Pos logit contributions
eal+0.003
arre+0.003
umpy+0.003
irk+0.003
ags+0.003
Neg logit contributions
lf-0.002
 cardboard-0.002
ts-0.001
 pretended-0.001
 pretend-0.001
Token he
Token ID339
Feature activation+1.92e-02

Pos logit contributions
eal+0.027
arre+0.026
 back+0.025
irk+0.024
 disappointed+0.024
Neg logit contributions
 cardboard-0.020
lf-0.020
 camps-0.019
ts-0.019
 trek-0.018
Token became
Token ID2627
Feature activation+2.39e-02

Pos logit contributions
 cardboard+0.027
lf+0.026
 camps+0.024
ts+0.023
 fossils+0.023
Neg logit contributions
eal-0.037
arre-0.037
umpy-0.033
 back-0.033
 disappointed-0.032
Token a
Token ID257
Feature activation+1.54e-01

Pos logit contributions
 cardboard+0.038
lf+0.035
ts+0.035
 camps+0.032
 pretend+0.032
Neg logit contributions
eal-0.044
arre-0.042
irk-0.040
umpy-0.040
awk-0.038
Token blue
Token ID4171
Feature activation+1.92e-02

Pos logit contributions
lf+0.237
 cardboard+0.229
ts+0.222
 trek+0.200
 blasted+0.200
Neg logit contributions
eal-0.276
arre-0.272
umpy-0.253
irk-0.248
 back-0.238
Tokenberry
Token ID8396
Feature activation+3.72e-02

Pos logit contributions
lf+0.034
ts+0.033
 cardboard+0.033
 blasted+0.030
 trek+0.030
Neg logit contributions
eal-0.030
arre-0.028
irk-0.027
umpy-0.027
vious-0.025
Token panc
Token ID22154
Feature activation-2.13e-01

Pos logit contributions
 cardboard+0.052
lf+0.051
ts+0.047
 camps+0.045
 blasted+0.044
Neg logit contributions
eal-0.073
arre-0.071
umpy-0.066
irk-0.066
ags-0.063
Tokenake
Token ID539
Feature activation-6.10e-02

Pos logit contributions
eal+0.241
arre+0.232
umpy+0.203
irk+0.201
ags+0.183
Neg logit contributions
 cardboard-0.481
lf-0.471
ts-0.456
 camps-0.439
 blasted-0.435
Token.
Token ID13
Feature activation-2.52e-02

Pos logit contributions
eal+0.125
arre+0.121
irk+0.115
umpy+0.113
ags+0.109
Neg logit contributions
lf-0.083
 cardboard-0.083
ts-0.075
 camps-0.072
 pretend-0.067
Token It
Token ID632
Feature activation-1.18e-02

Pos logit contributions
eal+0.056
arre+0.054
irk+0.052
umpy+0.051
awk+0.049
Neg logit contributions
 cardboard-0.030
lf-0.029
ts-0.025
 camps-0.025
 blasted-0.023
Token was
Token ID373
Feature activation+2.97e-02

Pos logit contributions
eal+0.028
arre+0.027
irk+0.026
umpy+0.026
ags+0.025
Neg logit contributions
 cardboard-0.012
lf-0.012
ts-0.010
 camps-0.010
 blasted-0.009
Token not
Token ID407
Feature activation-3.43e-02

Pos logit contributions
 cardboard+0.040
lf+0.039
ts+0.035
 camps+0.034
 blasted+0.032
Neg logit contributions
eal-0.061
arre-0.059
irk-0.056
umpy-0.056
ags-0.053
Token into
Token ID656
Feature activation+4.14e-02

Pos logit contributions
 cardboard+0.029
lf+0.028
ts+0.026
 camps+0.025
 trek+0.025
Neg logit contributions
eal-0.035
arre-0.035
umpy-0.032
irk-0.032
ags-0.030
Token many
Token ID867
Feature activation+8.83e-02

Pos logit contributions
 trek+0.056
lf+0.055
 camps+0.055
 blasted+0.053
 cardboard+0.053
Neg logit contributions
arre-0.077
eal-0.070
umpy-0.069
irk-0.067
 mad-0.066
Token pieces
Token ID5207
Feature activation+2.34e-02

Pos logit contributions
 cardboard+0.125
lf+0.120
 trek+0.116
 camps+0.116
 blasted+0.115
Neg logit contributions
arre-0.162
eal-0.160
umpy-0.151
irk-0.144
vious-0.141
Token and
Token ID290
Feature activation-1.57e-02

Pos logit contributions
 cardboard+0.033
lf+0.033
ts+0.031
 camps+0.030
 trek+0.029
Neg logit contributions
eal-0.044
arre-0.044
umpy-0.041
irk-0.040
vious-0.039
Token the
Token ID262
Feature activation-3.90e-02

Pos logit contributions
arre+0.025
eal+0.025
 back+0.023
 disappointed+0.023
 mad+0.023
Neg logit contributions
 camps-0.024
 trek-0.023
lf-0.023
 cardboard-0.023
ts-0.021
Token panc
Token ID22154
Feature activation-2.12e-01

Pos logit contributions
eal+0.081
arre+0.079
irk+0.075
umpy+0.075
agree+0.071
Neg logit contributions
 cardboard-0.052
lf-0.050
ts-0.047
 camps-0.043
 blasted-0.041
Tokenake
Token ID539
Feature activation-6.10e-02

Pos logit contributions
eal+0.257
arre+0.237
irk+0.218
umpy+0.216
vious+0.199
Neg logit contributions
 cardboard-0.461
lf-0.459
ts-0.431
 camps-0.416
 blasted-0.410
Token batter
Token ID4984
Feature activation-4.15e-02

Pos logit contributions
eal+0.138
arre+0.134
irk+0.128
umpy+0.127
ags+0.123
Neg logit contributions
 cardboard-0.068
lf-0.068
ts-0.061
 camps-0.055
 blasted-0.054
Token spilled
Token ID31896
Feature activation+3.69e-02

Pos logit contributions
eal+0.083
arre+0.081
umpy+0.074
 back+0.073
irk+0.072
Neg logit contributions
 cardboard-0.053
lf-0.053
 camps-0.050
ts-0.048
 trek-0.047
Token all
Token ID477
Feature activation-7.75e-02

Pos logit contributions
 cardboard+0.051
lf+0.050
ts+0.046
 camps+0.044
 blasted+0.042
Neg logit contributions
eal-0.074
arre-0.071
umpy-0.067
irk-0.067
ags-0.064
Token over
Token ID625
Feature activation-9.91e-02

Pos logit contributions
eal+0.123
arre+0.116
irk+0.115
ealing+0.110
umpy+0.106
Neg logit contributions
 cardboard-0.147
lf-0.132
ts-0.126
 blasted-0.122
 called-0.122
Token don
Token ID836
Feature activation-8.51e-02

Pos logit contributions
eal+0.018
arre+0.018
umpy+0.016
irk+0.016
ealing+0.015
Neg logit contributions
 cardboard-0.014
lf-0.013
ts-0.012
 camps-0.012
 trek-0.012
Token't
Token ID470
Feature activation-1.23e-02

Pos logit contributions
arre+0.131
eal+0.127
umpy+0.119
agree+0.113
Fighting+0.111
Neg logit contributions
 cardboard-0.150
lf-0.141
ts-0.141
 camps-0.139
 trek-0.136
Token you
Token ID345
Feature activation-9.90e-02

Pos logit contributions
eal+0.020
arre+0.019
umpy+0.017
irk+0.017
vious+0.017
Neg logit contributions
 cardboard-0.021
lf-0.020
ts-0.020
 camps-0.019
 trek-0.019
Token enter
Token ID3802
Feature activation+5.87e-02

Pos logit contributions
eal+0.182
arre+0.177
umpy+0.160
awk+0.156
vious+0.155
Neg logit contributions
lf-0.152
 cardboard-0.148
 camps-0.137
ts-0.137
 trek-0.132
Token a
Token ID257
Feature activation+1.56e-01

Pos logit contributions
lf+0.095
 cardboard+0.093
ts+0.088
 blasted+0.087
 trek+0.086
Neg logit contributions
eal-0.094
arre-0.092
ags-0.086
 back-0.085
umpy-0.083
Token panc
Token ID22154
Feature activation-2.12e-01

Pos logit contributions
 cardboard+0.220
lf+0.217
ts+0.205
 camps+0.189
 blasted+0.187
Neg logit contributions
eal-0.300
arre-0.292
umpy-0.272
irk-0.270
vious-0.256
Tokenake
Token ID539
Feature activation-6.03e-02

Pos logit contributions
eal+0.252
arre+0.236
irk+0.214
umpy+0.214
ags+0.194
Neg logit contributions
 cardboard-0.464
lf-0.461
ts-0.433
 camps-0.423
 blasted-0.414
Token cooking
Token ID10801
Feature activation-1.48e-02

Pos logit contributions
eal+0.127
arre+0.123
irk+0.117
umpy+0.116
ags+0.111
Neg logit contributions
 cardboard-0.077
lf-0.077
ts-0.068
 camps-0.066
 blasted-0.063
Token contest
Token ID8414
Feature activation+2.68e-02

Pos logit contributions
eal+0.031
arre+0.030
irk+0.029
umpy+0.029
Fighting+0.027
Neg logit contributions
 cardboard-0.019
lf-0.019
ts-0.017
 camps-0.016
 pretend-0.016
Token?"
Token ID1701
Feature activation-2.54e-02

Pos logit contributions
 cardboard+0.038
lf+0.037
ts+0.034
 camps+0.032
 trek+0.032
Neg logit contributions
eal-0.053
arre-0.051
umpy-0.048
irk-0.047
vious-0.045
Token Tim
Token ID5045
Feature activation+3.88e-02

Pos logit contributions
eal+0.052
arre+0.050
umpy+0.047
irk+0.047
ags+0.045
Neg logit contributions
 cardboard-0.034
lf-0.034
ts-0.031
 camps-0.029
 blasted-0.028
Token not
Token ID407
Feature activation-3.35e-02

Pos logit contributions
eal+0.030
arre+0.029
umpy+0.027
irk+0.027
ags+0.026
Neg logit contributions
 cardboard-0.021
lf-0.020
ts-0.019
 camps-0.017
 blasted-0.017
Token to
Token ID284
Feature activation-1.82e-04

Pos logit contributions
eal+0.068
arre+0.065
irk+0.063
umpy+0.061
ags+0.061
Neg logit contributions
 cardboard-0.046
lf-0.044
ts-0.041
 pretend-0.040
 shovel-0.039
Token stop
Token ID2245
Feature activation-9.15e-02

Pos logit contributions
eal+0.000
arre+0.000
irk+0.000
umpy+0.000
ags+0.000
Neg logit contributions
 cardboard-0.000
lf-0.000
ts-0.000
 camps-0.000
 pretend-0.000
Token stirring
Token ID26547
Feature activation-3.80e-02

Pos logit contributions
eal+0.206
arre+0.195
umpy+0.190
irk+0.189
agree+0.179
Neg logit contributions
 cardboard-0.112
lf-0.108
ts-0.102
 pretend-0.093
 pretended-0.089
Token the
Token ID262
Feature activation-3.84e-02

Pos logit contributions
eal+0.089
arre+0.086
irk+0.083
umpy+0.083
vious+0.078
Neg logit contributions
 cardboard-0.045
lf-0.043
ts-0.041
 pretend-0.037
 camps-0.035
Token panc
Token ID22154
Feature activation-2.11e-01

Pos logit contributions
eal+0.079
arre+0.078
irk+0.075
umpy+0.074
agree+0.070
Neg logit contributions
 cardboard-0.052
lf-0.051
ts-0.047
 pretend-0.043
 camps-0.043
Tokenake
Token ID539
Feature activation-5.95e-02

Pos logit contributions
eal+0.294
irk+0.267
arre+0.258
umpy+0.253
vious+0.239
Neg logit contributions
lf-0.425
 cardboard-0.412
ts-0.384
 camps-0.372
 blasted-0.365
Token batter
Token ID4984
Feature activation-4.04e-02

Pos logit contributions
eal+0.131
arre+0.127
irk+0.124
umpy+0.120
ags+0.115
Neg logit contributions
lf-0.074
 cardboard-0.072
ts-0.065
 camps-0.059
 pretend-0.059
Token.
Token ID13
Feature activation-2.43e-02

Pos logit contributions
eal+0.083
arre+0.080
irk+0.076
umpy+0.075
ags+0.071
Neg logit contributions
 cardboard-0.055
lf-0.054
ts-0.050
 camps-0.047
 pretend-0.045
Token
Token ID198
Feature activation-8.44e-03

Pos logit contributions
eal+0.054
umpy+0.052
irk+0.051
arre+0.051
vious+0.051
Neg logit contributions
 cardboard-0.027
lf-0.026
 called-0.023
 pretend-0.022
 pretended-0.021
Token
Token ID198
Feature activation-8.00e-03

Pos logit contributions
eal+0.015
arre+0.014
umpy+0.013
irk+0.013
ags+0.013
Neg logit contributions
 cardboard-0.014
lf-0.013
ts-0.012
 camps-0.012
 blasted-0.012
Token and
Token ID290
Feature activation-1.23e-02

Pos logit contributions
 cardboard+0.105
lf+0.103
ts+0.096
 camps+0.091
 trek+0.090
Neg logit contributions
eal-0.118
arre-0.114
umpy-0.107
irk-0.105
vious-0.100
Token supply
Token ID5127
Feature activation-8.67e-03

Pos logit contributions
eal+0.025
arre+0.024
umpy+0.022
irk+0.022
ags+0.021
Neg logit contributions
 cardboard-0.017
lf-0.016
ts-0.015
 camps-0.014
 blasted-0.014
Token us
Token ID514
Feature activation-1.19e-01

Pos logit contributions
eal+0.017
arre+0.017
umpy+0.016
irk+0.016
agree+0.015
Neg logit contributions
 cardboard-0.012
lf-0.011
ts-0.011
 camps-0.010
 pretend-0.010
Token with
Token ID351
Feature activation-2.22e-02

Pos logit contributions
eal+0.264
arre+0.258
irk+0.255
umpy+0.246
agree+0.235
Neg logit contributions
 cardboard-0.140
lf-0.133
ts-0.130
 pretend-0.116
 camps-0.115
Token more
Token ID517
Feature activation-6.22e-02

Pos logit contributions
eal+0.045
arre+0.043
irk+0.041
umpy+0.041
ags+0.038
Neg logit contributions
 cardboard-0.031
lf-0.030
ts-0.028
 camps-0.026
 blasted-0.025
Token panc
Token ID22154
Feature activation-2.11e-01

Pos logit contributions
eal+0.118
arre+0.115
irk+0.108
umpy+0.108
agree+0.102
Neg logit contributions
 cardboard-0.092
lf-0.090
ts-0.084
 camps-0.080
 pretend-0.076
Tokenake
Token ID539
Feature activation-6.00e-02

Pos logit contributions
eal+0.256
arre+0.238
irk+0.221
umpy+0.218
vious+0.200
Neg logit contributions
 cardboard-0.452
lf-0.451
ts-0.424
 camps-0.411
 trek-0.402
Token mix
Token ID5022
Feature activation-3.31e-02

Pos logit contributions
eal+0.122
arre+0.119
irk+0.118
umpy+0.112
Fighting+0.109
Neg logit contributions
lf-0.081
 cardboard-0.080
ts-0.074
 camps-0.069
 pretend-0.067
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
eal+0.067
arre+0.066
irk+0.063
umpy+0.062
agree+0.059
Neg logit contributions
 cardboard-0.044
lf-0.043
ts-0.041
 camps-0.038
 pretend-0.036
Token But
Token ID887
Feature activation+1.34e-03

Pos logit contributions
eal+0.036
arre+0.034
irk+0.033
umpy+0.032
ags+0.031
Neg logit contributions
 cardboard-0.036
lf-0.034
 camps-0.032
ts-0.032
 pretend-0.031
Token it
Token ID340
Feature activation-1.21e-01

Pos logit contributions
 cardboard+0.002
lf+0.002
ts+0.002
 camps+0.002
 trek+0.002
Neg logit contributions
eal-0.002
arre-0.002
umpy-0.002
irk-0.002
 back-0.002
Token where
Token ID810
Feature activation+4.58e-02

Pos logit contributions
eal+0.063
arre+0.060
umpy+0.057
irk+0.057
ags+0.054
Neg logit contributions
 cardboard-0.044
lf-0.043
ts-0.039
 camps-0.038
 trek-0.036
Token they
Token ID484
Feature activation-1.30e-02

Pos logit contributions
 cardboard+0.043
lf+0.040
 pretend+0.035
ts+0.034
 called+0.032
Neg logit contributions
eal-0.118
arre-0.115
ags-0.111
umpy-0.110
irk-0.110
Token could
Token ID714
Feature activation+9.77e-02

Pos logit contributions
eal+0.028
arre+0.028
umpy+0.026
 back+0.026
vious+0.026
Neg logit contributions
lf-0.016
 cardboard-0.016
 camps-0.014
ts-0.013
 trek-0.013
Token get
Token ID651
Feature activation-5.06e-02

Pos logit contributions
 cardboard+0.143
lf+0.141
ts+0.134
 camps+0.127
 trek+0.124
Neg logit contributions
eal-0.182
arre-0.178
umpy-0.166
irk-0.160
ags-0.158
Token more
Token ID517
Feature activation-6.18e-02

Pos logit contributions
eal+0.109
arre+0.107
irk+0.105
umpy+0.104
agree+0.097
Neg logit contributions
 cardboard-0.065
lf-0.058
ts-0.057
 pretend-0.054
 camps-0.053
Token panc
Token ID22154
Feature activation-2.11e-01

Pos logit contributions
arre+0.110
eal+0.108
irk+0.105
umpy+0.103
agree+0.101
Neg logit contributions
 cardboard-0.097
lf-0.093
ts-0.088
 pretend-0.086
 camps-0.085
Tokenake
Token ID539
Feature activation-5.94e-02

Pos logit contributions
eal+0.264
arre+0.249
irk+0.230
umpy+0.229
vious+0.208
Neg logit contributions
 cardboard-0.443
lf-0.441
ts-0.416
 camps-0.402
 blasted-0.393
Token mix
Token ID5022
Feature activation-3.32e-02

Pos logit contributions
eal+0.137
irk+0.137
arre+0.133
umpy+0.131
Fighting+0.128
Neg logit contributions
lf-0.064
 cardboard-0.061
ts-0.053
 pretend-0.051
 camps-0.048
Token.
Token ID13
Feature activation-1.93e-02

Pos logit contributions
eal+0.076
arre+0.075
irk+0.073
umpy+0.072
Fighting+0.069
Neg logit contributions
 cardboard-0.036
lf-0.033
ts-0.032
 pretend-0.029
 camps-0.028
Token
Token ID220
Feature activation-9.38e-02

Pos logit contributions
umpy+0.040
eal+0.040
vious+0.039
arre+0.038
irk+0.038
Neg logit contributions
 cardboard-0.024
lf-0.023
 pretend-0.021
 called-0.020
 pretended-0.020
Token
Token ID198
Feature activation-3.57e-03

Pos logit contributions
eal+0.192
arre+0.182
irk+0.171
vious+0.170
umpy+0.169
Neg logit contributions
 cardboard-0.123
lf-0.121
ts-0.108
 camps-0.104
 trek-0.099
Token pointed
Token ID6235
Feature activation+1.75e-02

Pos logit contributions
 cardboard+0.116
lf+0.109
ts+0.105
 camps+0.097
 blasted+0.091
Neg logit contributions
eal-0.186
arre-0.175
irk-0.161
umpy-0.159
 back-0.154
Token to
Token ID284
Feature activation-3.02e-03

Pos logit contributions
 cardboard+0.022
 pretend+0.020
 shovel+0.020
 swords+0.019
 pretended+0.019
Neg logit contributions
arre-0.039
umpy-0.036
eal-0.034
agree-0.034
ivering-0.034
Token the
Token ID262
Feature activation-3.51e-02

Pos logit contributions
eal+0.006
arre+0.005
irk+0.005
 back+0.005
umpy+0.005
Neg logit contributions
lf-0.004
 cardboard-0.004
ts-0.004
 camps-0.004
 blasted-0.004
Token blue
Token ID4171
Feature activation+2.49e-02

Pos logit contributions
eal+0.071
arre+0.068
umpy+0.065
irk+0.064
vious+0.061
Neg logit contributions
 cardboard-0.048
lf-0.047
ts-0.043
 camps-0.041
 blasted-0.040
Tokenberry
Token ID8396
Feature activation+4.51e-02

Pos logit contributions
 cardboard+0.039
lf+0.038
ts+0.036
 camps+0.034
 blasted+0.034
Neg logit contributions
eal-0.045
arre-0.043
irk-0.040
umpy-0.040
vious-0.038
Token panc
Token ID22154
Feature activation-2.10e-01

Pos logit contributions
 cardboard+0.058
lf+0.057
ts+0.052
 camps+0.049
 blasted+0.048
Neg logit contributions
eal-0.094
arre-0.091
umpy-0.086
irk-0.086
ags-0.082
Tokenake
Token ID539
Feature activation-6.03e-02

Pos logit contributions
eal+0.264
arre+0.251
umpy+0.230
irk+0.229
vious+0.211
Neg logit contributions
lf-0.441
 cardboard-0.440
ts-0.409
 camps-0.397
 blasted-0.390
Token.
Token ID13
Feature activation-1.90e-02

Pos logit contributions
eal+0.129
arre+0.127
irk+0.122
umpy+0.122
vious+0.115
Neg logit contributions
lf-0.079
 cardboard-0.074
ts-0.068
 camps-0.061
 pretend-0.061
Token
Token ID198
Feature activation-7.72e-03

Pos logit contributions
eal+0.033
arre+0.032
irk+0.030
umpy+0.030
vious+0.029
Neg logit contributions
lf-0.032
 cardboard-0.032
 camps-0.028
ts-0.027
 blasted-0.027
Token
Token ID198
Feature activation-7.39e-03

Pos logit contributions
eal+0.016
arre+0.015
irk+0.014
vious+0.014
awk+0.014
Neg logit contributions
lf-0.010
 cardboard-0.010
 camps-0.009
 trek-0.008
 blasted-0.008
TokenMom
Token ID29252
Feature activation+2.33e-02

Pos logit contributions
eal+0.014
arre+0.013
irk+0.013
umpy+0.012
ags+0.012
Neg logit contributions
lf-0.012
 cardboard-0.011
ts-0.010
 camps-0.010
 blasted-0.010
Token that
Token ID326
Feature activation-3.52e-03

Pos logit contributions
eal+0.124
arre+0.120
irk+0.113
umpy+0.113
ags+0.109
Neg logit contributions
 cardboard-0.084
lf-0.083
ts-0.075
 camps-0.070
 blasted-0.068
Token they
Token ID484
Feature activation-1.21e-02

Pos logit contributions
eal+0.008
arre+0.008
irk+0.007
umpy+0.007
ags+0.007
Neg logit contributions
 cardboard-0.004
lf-0.004
ts-0.004
 camps-0.004
 pretend-0.003
Token ran
Token ID4966
Feature activation+4.42e-02

Pos logit contributions
eal+0.030
arre+0.029
umpy+0.028
irk+0.027
vious+0.026
Neg logit contributions
 cardboard-0.011
lf-0.011
ts-0.010
 camps-0.009
 trek-0.008
Token out
Token ID503
Feature activation-1.13e-02

Pos logit contributions
 cardboard+0.069
lf+0.062
 camps+0.061
 pretend+0.061
ts+0.059
Neg logit contributions
eal-0.082
arre-0.077
irk-0.077
umpy-0.076
ealing-0.073
Token of
Token ID286
Feature activation+2.46e-02

Pos logit contributions
eal+0.022
arre+0.021
irk+0.020
umpy+0.020
ealing+0.019
Neg logit contributions
 cardboard-0.017
lf-0.016
ts-0.015
 camps-0.014
 trek-0.014
Token panc
Token ID22154
Feature activation-2.10e-01

Pos logit contributions
 cardboard+0.036
lf+0.035
ts+0.033
 camps+0.031
 blasted+0.031
Neg logit contributions
eal-0.046
arre-0.045
irk-0.042
umpy-0.041
ags-0.040
Tokenake
Token ID539
Feature activation-5.86e-02

Pos logit contributions
eal+0.250
arre+0.234
umpy+0.216
irk+0.215
vious+0.194
Neg logit contributions
 cardboard-0.457
lf-0.452
ts-0.425
 camps-0.416
 trek-0.406
Token mix
Token ID5022
Feature activation-3.27e-02

Pos logit contributions
eal+0.109
arre+0.106
umpy+0.098
irk+0.097
ags+0.093
Neg logit contributions
 cardboard-0.088
lf-0.086
ts-0.080
 camps-0.077
 blasted-0.075
Token.
Token ID13
Feature activation-1.71e-02

Pos logit contributions
eal+0.070
arre+0.068
irk+0.065
umpy+0.065
agree+0.061
Neg logit contributions
 cardboard-0.041
lf-0.039
ts-0.038
 camps-0.034
 pretend-0.034
Token Lily
Token ID20037
Feature activation+1.18e-02

Pos logit contributions
eal+0.035
arre+0.034
irk+0.033
umpy+0.033
awk+0.032
Neg logit contributions
 cardboard-0.023
lf-0.021
 pretend-0.019
 pretended-0.018
 camps-0.018
Token was
Token ID373
Feature activation+3.65e-02

Pos logit contributions
lf+0.012
 cardboard+0.011
ts+0.010
 pretended+0.010
 pretend+0.009
Neg logit contributions
eal-0.028
arre-0.027
irk-0.027
ags-0.026
umpy-0.026
Token and
Token ID290
Feature activation-1.40e-02

Pos logit contributions
eal+0.044
arre+0.043
irk+0.042
umpy+0.041
ags+0.040
Neg logit contributions
 cardboard-0.023
lf-0.022
 pretended-0.020
 pretend-0.019
ts-0.019
Token continued
Token ID3767
Feature activation+5.41e-02

Pos logit contributions
eal+0.028
arre+0.027
irk+0.025
umpy+0.024
 back+0.024
Neg logit contributions
 cardboard-0.019
lf-0.018
ts-0.017
 camps-0.017
 trek-0.015
Token to
Token ID284
Feature activation+2.39e-03

Pos logit contributions
 cardboard+0.086
lf+0.084
ts+0.076
 pretend+0.075
 camps+0.074
Neg logit contributions
eal-0.103
arre-0.098
umpy-0.098
irk-0.096
vious-0.094
Token stir
Token ID11240
Feature activation-3.00e-02

Pos logit contributions
lf+0.003
 cardboard+0.003
ts+0.003
 camps+0.003
 trek+0.003
Neg logit contributions
eal-0.005
arre-0.005
irk-0.004
umpy-0.004
ags-0.004
Token the
Token ID262
Feature activation-3.66e-02

Pos logit contributions
eal+0.066
arre+0.064
umpy+0.062
irk+0.062
agree+0.058
Neg logit contributions
 cardboard-0.037
lf-0.036
ts-0.035
 pretend-0.031
 camps-0.030
Token panc
Token ID22154
Feature activation-2.08e-01

Pos logit contributions
eal+0.073
arre+0.070
irk+0.068
umpy+0.067
vious+0.063
Neg logit contributions
 cardboard-0.054
lf-0.052
ts-0.046
 pretend-0.044
 camps-0.043
Tokenake
Token ID539
Feature activation-5.79e-02

Pos logit contributions
eal+0.262
arre+0.240
irk+0.229
umpy+0.224
vious+0.209
Neg logit contributions
lf-0.443
 cardboard-0.439
ts-0.409
 camps-0.397
 blasted-0.389
Token batter
Token ID4984
Feature activation-3.81e-02

Pos logit contributions
eal+0.121
arre+0.117
irk+0.111
umpy+0.110
ags+0.105
Neg logit contributions
 cardboard-0.077
lf-0.076
ts-0.068
 camps-0.064
 blasted-0.062
Token.
Token ID13
Feature activation-1.88e-02

Pos logit contributions
eal+0.073
arre+0.071
umpy+0.066
irk+0.066
ags+0.064
Neg logit contributions
 cardboard-0.054
lf-0.053
ts-0.049
 camps-0.047
 trek-0.046
Token
Token ID220
Feature activation-9.33e-02

Pos logit contributions
eal+0.039
arre+0.037
irk+0.036
umpy+0.036
vious+0.035
Neg logit contributions
 cardboard-0.024
lf-0.023
 pretend-0.021
 camps-0.020
 pretended-0.020
Token
Token ID198
Feature activation-3.91e-03

Pos logit contributions
eal+0.187
arre+0.179
irk+0.168
vious+0.165
umpy+0.165
Neg logit contributions
 cardboard-0.126
lf-0.124
ts-0.111
 camps-0.107
 trek-0.103
Token.
Token ID13
Feature activation-3.97e-02

Pos logit contributions
eal+0.274
irk+0.255
arre+0.250
ags+0.239
umpy+0.233
Neg logit contributions
 cardboard-0.120
lf-0.112
 pretend-0.105
ts-0.102
 called-0.100
Token I
Token ID314
Feature activation-4.29e-03

Pos logit contributions
eal+0.063
arre+0.061
irk+0.055
umpy+0.054
awk+0.051
Neg logit contributions
 cardboard-0.070
lf-0.066
 pretend-0.063
ts-0.061
 named-0.061
Token was
Token ID373
Feature activation+2.25e-02

Pos logit contributions
eal+0.009
arre+0.008
ags+0.008
irk+0.008
umpy+0.008
Neg logit contributions
 cardboard-0.005
 pretend-0.005
lf-0.005
ts-0.005
 pretended-0.005
Token tired
Token ID10032
Feature activation-4.33e-02

Pos logit contributions
 cardboard+0.033
ts+0.031
lf+0.030
 pretend+0.029
 called+0.029
Neg logit contributions
eal-0.044
irk-0.042
arre-0.041
umpy-0.039
agree-0.038
Token and
Token ID290
Feature activation-2.71e-02

Pos logit contributions
eal+0.096
arre+0.092
irk+0.085
ags+0.083
ealing+0.083
Neg logit contributions
 cardboard-0.052
lf-0.050
ts-0.044
 camps-0.044
 pretend-0.041
Token crank
Token ID30425
Feature activation-1.81e-01

Pos logit contributions
eal+0.046
arre+0.045
irk+0.042
ags+0.041
umpy+0.040
Neg logit contributions
 cardboard-0.045
lf-0.044
ts-0.043
 pretend-0.040
 pretended-0.040
Tokeny
Token ID88
Feature activation+3.62e-02

Pos logit contributions
eal+0.360
arre+0.335
ags+0.329
irk+0.320
Aud+0.317
Neg logit contributions
 called-0.208
ts-0.198
 pretended-0.196
lf-0.196
 pretend-0.174
Token.
Token ID13
Feature activation-4.04e-02

Pos logit contributions
 cardboard+0.040
lf+0.037
ts+0.035
 pretend+0.034
 camps+0.032
Neg logit contributions
eal-0.083
arre-0.079
irk-0.077
ags-0.073
agree-0.072
Token I
Token ID314
Feature activation-8.68e-04

Pos logit contributions
eal+0.076
arre+0.074
irk+0.067
umpy+0.065
ags+0.064
Neg logit contributions
 cardboard-0.061
lf-0.058
 pretend-0.052
 camps-0.052
ts-0.051
Token appreciate
Token ID9144
Feature activation-9.55e-03

Pos logit contributions
eal+0.002
arre+0.002
ags+0.002
irk+0.002
umpy+0.002
Neg logit contributions
 cardboard-0.001
 pretended-0.001
 pretend-0.001
ts-0.001
lf-0.001
Token your
Token ID534
Feature activation-1.05e-01

Pos logit contributions
eal+0.019
arre+0.018
umpy+0.017
irk+0.017
vious+0.015
Neg logit contributions
 cardboard-0.014
lf-0.014
ts-0.013
 camps-0.012
 pretend-0.012
Token.
Token ID13
Feature activation-3.60e-02

Pos logit contributions
eal+0.259
arre+0.251
umpy+0.235
irk+0.235
ags+0.224
Neg logit contributions
 cardboard-0.176
lf-0.173
ts-0.159
 camps-0.151
 blasted-0.146
Token They
Token ID1119
Feature activation+5.56e-02

Pos logit contributions
eal+0.081
arre+0.073
awk+0.070
vious+0.069
irk+0.068
Neg logit contributions
 cardboard-0.043
 camps-0.036
 pretend-0.034
 shovel-0.034
 called-0.033
Token are
Token ID389
Feature activation-1.67e-02

Pos logit contributions
 cardboard+0.083
lf+0.080
ts+0.077
 camps+0.071
 pretend+0.071
Neg logit contributions
eal-0.104
arre-0.101
umpy-0.095
irk-0.095
ags-0.090
Token tired
Token ID10032
Feature activation-4.16e-02

Pos logit contributions
eal+0.035
arre+0.033
irk+0.032
umpy+0.030
 aback+0.029
Neg logit contributions
 cardboard-0.024
ts-0.022
lf-0.022
 called-0.022
 pretend-0.021
Token and
Token ID290
Feature activation-3.13e-02

Pos logit contributions
eal+0.080
arre+0.078
umpy+0.074
irk+0.074
ags+0.070
Neg logit contributions
 cardboard-0.059
lf-0.058
ts-0.054
 camps-0.051
 blasted-0.050
Token crank
Token ID30425
Feature activation-1.80e-01

Pos logit contributions
arre+0.053
eal+0.053
irk+0.048
ags+0.045
itter+0.045
Neg logit contributions
ts-0.053
 cardboard-0.051
 pretend-0.050
 pretended-0.049
 called-0.048
Tokeny
Token ID88
Feature activation+3.19e-02

Pos logit contributions
eal+0.385
joy+0.334
arre+0.332
 disapproval+0.332
 aback+0.332
Neg logit contributions
 called-0.194
ts-0.192
lf-0.182
 pretended-0.177
 pretend-0.172
Token.
Token ID13
Feature activation-3.58e-02

Pos logit contributions
 cardboard+0.037
lf+0.034
ts+0.033
 pretend+0.031
 pretended+0.029
Neg logit contributions
eal-0.073
arre-0.069
irk-0.066
umpy-0.064
ags-0.063
Token They
Token ID1119
Feature activation+5.56e-02

Pos logit contributions
eal+0.087
arre+0.079
awk+0.076
vious+0.076
irk+0.075
Neg logit contributions
 cardboard-0.036
 pretend-0.030
 camps-0.029
 shovel-0.028
lf-0.026
Token say
Token ID910
Feature activation-5.60e-02

Pos logit contributions
 cardboard+0.079
lf+0.074
ts+0.073
 pretend+0.070
 trek+0.067
Neg logit contributions
eal-0.108
arre-0.107
umpy-0.100
irk-0.099
ags-0.094
Token,
Token ID11
Feature activation+4.26e-02

Pos logit contributions
eal+0.161
arre+0.157
irk+0.154
umpy+0.154
vious+0.146
Neg logit contributions
 cardboard-0.038
lf-0.032
ts-0.026
 pretend-0.026
 shovel-0.023
Token.
Token ID13
Feature activation-3.60e-02

Pos logit contributions
eal+0.153
arre+0.146
irk+0.142
agree+0.140
umpy+0.135
Neg logit contributions
 cardboard-0.060
lf-0.058
ts-0.054
 pretend-0.051
 camps-0.050
Token You
Token ID921
Feature activation-5.48e-02

Pos logit contributions
eal+0.061
arre+0.061
irk+0.052
ocked+0.051
 mad+0.050
Neg logit contributions
 cardboard-0.059
 pretend-0.052
 named-0.050
lf-0.049
 shovel-0.049
Token are
Token ID389
Feature activation-1.74e-02

Pos logit contributions
eal+0.106
arre+0.104
umpy+0.097
irk+0.095
vious+0.093
Neg logit contributions
lf-0.079
 cardboard-0.079
ts-0.070
 camps-0.070
 trek-0.066
Token tired
Token ID10032
Feature activation-4.14e-02

Pos logit contributions
eal+0.035
arre+0.034
irk+0.032
umpy+0.031
awk+0.030
Neg logit contributions
 cardboard-0.025
lf-0.024
ts-0.024
 pretend-0.021
 pretended-0.021
Token and
Token ID290
Feature activation-3.00e-02

Pos logit contributions
eal+0.088
arre+0.084
irk+0.079
umpy+0.077
ags+0.076
Neg logit contributions
 cardboard-0.053
lf-0.051
ts-0.047
 camps-0.045
 pretend-0.042
Token crank
Token ID30425
Feature activation-1.80e-01

Pos logit contributions
arre+0.054
eal+0.054
irk+0.050
ags+0.047
umpy+0.045
Neg logit contributions
ts-0.047
 cardboard-0.046
 pretend-0.044
lf-0.042
 called-0.042
Tokeny
Token ID88
Feature activation+3.32e-02

Pos logit contributions
eal+0.379
arre+0.330
irk+0.330
joy+0.328
ags+0.323
Neg logit contributions
 called-0.199
ts-0.188
lf-0.184
 pretend-0.177
 pretended-0.176
Token.
Token ID13
Feature activation-3.66e-02

Pos logit contributions
 cardboard+0.036
lf+0.033
ts+0.032
 pretend+0.031
 camps+0.028
Neg logit contributions
eal-0.078
arre-0.073
irk-0.071
ags-0.067
agree-0.067
Token You
Token ID921
Feature activation-5.46e-02

Pos logit contributions
eal+0.058
arre+0.056
irk+0.049
umpy+0.046
ocked+0.046
Neg logit contributions
 cardboard-0.065
lf-0.058
 pretend-0.057
ts-0.056
 named-0.056
Token need
Token ID761
Feature activation-1.71e-03

Pos logit contributions
eal+0.109
arre+0.106
umpy+0.099
irk+0.099
ags+0.094
Neg logit contributions
 cardboard-0.076
lf-0.074
ts-0.068
 camps-0.065
 blasted-0.063
Token to
Token ID284
Feature activation-8.64e-03

Pos logit contributions
eal+0.003
arre+0.003
umpy+0.003
irk+0.003
ealing+0.002
Neg logit contributions
 cardboard-0.003
lf-0.003
ts-0.003
 camps-0.003
 trek-0.002
Token the
Token ID262
Feature activation-4.69e-02

Pos logit contributions
eal+0.038
arre+0.037
umpy+0.035
irk+0.034
awk+0.033
Neg logit contributions
 cardboard-0.018
lf-0.016
ts-0.015
 pretend-0.013
 camps-0.013
Token bag
Token ID6131
Feature activation+1.07e-02

Pos logit contributions
eal+0.062
arre+0.061
umpy+0.058
irk+0.057
 back+0.052
Neg logit contributions
lf-0.090
ts-0.089
 cardboard-0.088
 blasted-0.084
 trek-0.083
Token will
Token ID481
Feature activation-4.59e-02

Pos logit contributions
 cardboard+0.014
ts+0.013
lf+0.013
 pretend+0.012
 camps+0.012
Neg logit contributions
eal-0.022
arre-0.022
irk-0.021
umpy-0.020
Fighting-0.019
Token not
Token ID407
Feature activation-4.40e-02

Pos logit contributions
eal+0.088
arre+0.086
irk+0.079
umpy+0.079
vious+0.075
Neg logit contributions
 cardboard-0.069
lf-0.066
ts-0.062
 camps-0.059
 pretend-0.057
Token be
Token ID307
Feature activation-7.21e-02

Pos logit contributions
eal+0.077
arre+0.074
irk+0.069
umpy+0.069
ags+0.065
Neg logit contributions
 cardboard-0.072
lf-0.070
ts-0.066
 camps-0.063
 blasted-0.061
Token mad
Token ID8805
Feature activation-1.78e-01

Pos logit contributions
eal+0.119
arre+0.115
irk+0.107
umpy+0.105
agree+0.099
Neg logit contributions
 cardboard-0.127
lf-0.121
ts-0.120
 camps-0.110
 pretended-0.109
Token at
Token ID379
Feature activation-4.76e-02

Pos logit contributions
eal+0.347
arre+0.320
irk+0.310
agree+0.297
ags+0.294
Neg logit contributions
 cardboard-0.251
lf-0.243
ts-0.231
 camps-0.229
 pretend-0.220
Token them
Token ID606
Feature activation-1.15e-01

Pos logit contributions
eal+0.081
arre+0.079
irk+0.073
umpy+0.073
ags+0.071
Neg logit contributions
 cardboard-0.077
lf-0.076
ts-0.070
 camps-0.069
 blasted-0.067
Token.
Token ID13
Feature activation-3.85e-02

Pos logit contributions
eal+0.264
arre+0.245
irk+0.241
umpy+0.225
ags+0.224
Neg logit contributions
 cardboard-0.131
lf-0.126
ts-0.115
 camps-0.112
 pretend-0.111
Token They
Token ID1119
Feature activation+5.73e-02

Pos logit contributions
eal+0.084
arre+0.080
irk+0.075
awk+0.074
umpy+0.073
Neg logit contributions
 cardboard-0.047
lf-0.040
 pretend-0.039
ts-0.038
 camps-0.038
Token hope
Token ID2911
Feature activation-1.62e-02

Pos logit contributions
 cardboard+0.085
lf+0.079
ts+0.078
 pretend+0.074
 camps+0.073
Neg logit contributions
eal-0.108
arre-0.107
umpy-0.100
irk-0.099
ags-0.093
Token please
Token ID3387
Feature activation-6.86e-02

Pos logit contributions
 cardboard+0.061
lf+0.059
 pretend+0.055
ts+0.055
 camps+0.054
Neg logit contributions
eal-0.068
arre-0.067
irk-0.063
umpy-0.063
ags-0.059
Token.
Token ID13
Feature activation-4.43e-02

Pos logit contributions
eal+0.125
arre+0.120
umpy+0.112
irk+0.111
ags+0.108
Neg logit contributions
 cardboard-0.105
lf-0.103
ts-0.096
 camps-0.093
 trek-0.092
Token Do
Token ID2141
Feature activation+2.63e-02

Pos logit contributions
eal+0.060
arre+0.058
umpy+0.052
irk+0.052
ags+0.048
Neg logit contributions
 cardboard-0.090
lf-0.088
ts-0.083
 camps-0.081
 blasted-0.079
Token not
Token ID407
Feature activation-4.66e-02

Pos logit contributions
 cardboard+0.047
lf+0.045
 camps+0.044
 blasted+0.043
 trek+0.042
Neg logit contributions
eal-0.039
arre-0.038
agree-0.034
umpy-0.034
irk-0.034
Token be
Token ID307
Feature activation-7.39e-02

Pos logit contributions
eal+0.087
arre+0.085
umpy+0.079
irk+0.078
vious+0.075
Neg logit contributions
 cardboard-0.069
lf-0.068
ts-0.064
 camps-0.061
 trek-0.059
Token mad
Token ID8805
Feature activation-1.78e-01

Pos logit contributions
eal+0.114
irk+0.107
arre+0.106
agree+0.100
umpy+0.099
Neg logit contributions
 cardboard-0.132
 called-0.128
ts-0.126
 pretend-0.120
 named-0.119
Token.
Token ID13
Feature activation-4.45e-02

Pos logit contributions
eal+0.394
arre+0.363
irk+0.358
agree+0.345
ags+0.340
Neg logit contributions
lf-0.194
 cardboard-0.190
ts-0.173
 camps-0.172
 pretend-0.171
Token Do
Token ID2141
Feature activation+2.62e-02

Pos logit contributions
eal+0.073
arre+0.071
irk+0.065
umpy+0.064
ags+0.060
Neg logit contributions
 cardboard-0.078
lf-0.075
ts-0.070
 pretend-0.068
 camps-0.067
Token not
Token ID407
Feature activation-4.69e-02

Pos logit contributions
 cardboard+0.045
lf+0.044
 camps+0.042
ts+0.040
 blasted+0.040
Neg logit contributions
eal-0.041
arre-0.041
umpy-0.036
irk-0.036
agree-0.036
Token take
Token ID1011
Feature activation-4.52e-02

Pos logit contributions
eal+0.089
arre+0.086
umpy+0.081
irk+0.081
ags+0.076
Neg logit contributions
 cardboard-0.069
lf-0.068
ts-0.063
 camps-0.060
 blasted-0.058
Token our
Token ID674
Feature activation-8.87e-02

Pos logit contributions
eal+0.096
arre+0.094
umpy+0.090
irk+0.089
agree+0.084
Neg logit contributions
 cardboard-0.059
ts-0.055
lf-0.053
 pretend-0.048
 camps-0.046
Token.
Token ID13
Feature activation-4.01e-02

Pos logit contributions
 cardboard+0.009
lf+0.009
ts+0.008
 camps+0.008
 pretend+0.007
Neg logit contributions
eal-0.016
arre-0.015
irk-0.014
umpy-0.014
agree-0.013
Token You
Token ID921
Feature activation-5.52e-02

Pos logit contributions
eal+0.057
arre+0.054
irk+0.049
umpy+0.049
awk+0.045
Neg logit contributions
 cardboard-0.079
lf-0.076
ts-0.072
 camps-0.070
 blasted-0.068
Token're
Token ID821
Feature activation-2.89e-02

Pos logit contributions
eal+0.103
arre+0.103
umpy+0.095
irk+0.092
vious+0.091
Neg logit contributions
lf-0.082
 cardboard-0.081
ts-0.073
 camps-0.073
 trek-0.069
Token tired
Token ID10032
Feature activation-3.88e-02

Pos logit contributions
eal+0.056
irk+0.051
arre+0.050
umpy+0.048
awk+0.047
Neg logit contributions
 cardboard-0.047
lf-0.043
ts-0.043
 called-0.039
 pretend-0.038
Token and
Token ID290
Feature activation-2.86e-02

Pos logit contributions
eal+0.084
arre+0.080
irk+0.076
umpy+0.073
ags+0.073
Neg logit contributions
 cardboard-0.048
lf-0.046
ts-0.041
 camps-0.040
 pretend-0.038
Token crank
Token ID30425
Feature activation-1.78e-01

Pos logit contributions
eal+0.046
arre+0.044
irk+0.041
ags+0.039
umpy+0.038
Neg logit contributions
 cardboard-0.051
ts-0.049
lf-0.048
 pretend-0.046
 pretended-0.045
Tokeny
Token ID88
Feature activation+3.49e-02

Pos logit contributions
eal+0.339
ags+0.299
Aud+0.296
irk+0.296
arre+0.286
Neg logit contributions
 called-0.221
lf-0.212
ts-0.208
 pretended-0.206
 cardboard-0.197
Token."
Token ID526
Feature activation-4.40e-02

Pos logit contributions
 cardboard+0.041
lf+0.037
ts+0.035
 pretend+0.034
 camps+0.033
Neg logit contributions
eal-0.079
irk-0.073
arre-0.072
ags-0.068
agree-0.066
Token
Token ID198
Feature activation-1.80e-02

Pos logit contributions
eal+0.090
arre+0.085
umpy+0.078
vious+0.077
irk+0.077
Neg logit contributions
 cardboard-0.062
lf-0.060
ts-0.053
 shovel-0.051
 pretend-0.050
Token
Token ID198
Feature activation-1.76e-02

Pos logit contributions
eal+0.028
arre+0.027
irk+0.026
umpy+0.026
Fighting+0.024
Neg logit contributions
 cardboard-0.032
lf-0.031
ts-0.030
 camps-0.029
 blasted-0.029
TokenS
Token ID50
Feature activation-5.76e-02

Pos logit contributions
eal+0.032
arre+0.031
umpy+0.029
irk+0.029
ags+0.028
Neg logit contributions
 cardboard-0.027
lf-0.027
ts-0.025
 camps-0.024
 trek-0.023
Token the
Token ID262
Feature activation-4.74e-02

Pos logit contributions
lf+0.026
 cardboard+0.026
ts+0.024
 camps+0.024
 trek+0.023
Neg logit contributions
eal-0.032
arre-0.031
irk-0.029
umpy-0.029
 back-0.028
Token car
Token ID1097
Feature activation+4.05e-02

Pos logit contributions
eal+0.089
arre+0.086
umpy+0.080
irk+0.080
ags+0.076
Neg logit contributions
 cardboard-0.072
lf-0.070
ts-0.065
 camps-0.062
 blasted-0.060
Token and
Token ID290
Feature activation-3.07e-02

Pos logit contributions
 cardboard+0.062
lf+0.060
ts+0.056
 camps+0.053
 blasted+0.052
Neg logit contributions
eal-0.076
arre-0.073
irk-0.069
umpy-0.068
ags-0.065
Token not
Token ID407
Feature activation-4.30e-02

Pos logit contributions
eal+0.067
arre+0.066
umpy+0.063
irk+0.061
ealing+0.059
Neg logit contributions
 cardboard-0.040
lf-0.038
ts-0.035
 pretend-0.033
 pretended-0.032
Token be
Token ID307
Feature activation-7.31e-02

Pos logit contributions
eal+0.076
arre+0.074
umpy+0.069
irk+0.068
vious+0.065
Neg logit contributions
 cardboard-0.068
lf-0.067
ts-0.062
 camps-0.061
 trek-0.059
Token mad
Token ID8805
Feature activation-1.78e-01

Pos logit contributions
eal+0.121
irk+0.112
arre+0.111
umpy+0.102
agree+0.102
Neg logit contributions
 cardboard-0.132
ts-0.127
lf-0.126
 pretended-0.116
 pretend-0.112
Token at
Token ID379
Feature activation-4.71e-02

Pos logit contributions
eal+0.345
arre+0.321
irk+0.308
agree+0.301
ags+0.296
Neg logit contributions
 cardboard-0.252
lf-0.243
 camps-0.229
ts-0.228
 pretend-0.217
Token them
Token ID606
Feature activation-1.14e-01

Pos logit contributions
arre+0.073
eal+0.072
irk+0.067
ags+0.067
 back+0.067
Neg logit contributions
 cardboard-0.078
lf-0.078
 camps-0.073
 blasted-0.072
 trek-0.071
Token.
Token ID13
Feature activation-3.69e-02

Pos logit contributions
eal+0.258
arre+0.243
irk+0.236
umpy+0.226
ags+0.224
Neg logit contributions
 cardboard-0.133
lf-0.130
ts-0.116
 camps-0.113
 pretend-0.113
Token
Token ID198
Feature activation-1.58e-02

Pos logit contributions
eal+0.075
arre+0.072
umpy+0.067
irk+0.066
vious+0.065
Neg logit contributions
 cardboard-0.050
lf-0.047
ts-0.043
 camps-0.043
 pretend-0.042
Token
Token ID198
Feature activation-1.53e-02

Pos logit contributions
eal+0.024
arre+0.023
umpy+0.021
irk+0.020
vious+0.019
Neg logit contributions
 cardboard-0.030
lf-0.029
ts-0.028
 camps-0.027
 trek-0.026
Token You
Token ID921
Feature activation-5.05e-02

Pos logit contributions
arre+0.064
eal+0.062
irk+0.055
umpy+0.053
ags+0.053
Neg logit contributions
 cardboard-0.055
lf-0.050
 pretend-0.050
 blasted-0.048
ts-0.047
Token are
Token ID389
Feature activation-1.59e-02

Pos logit contributions
eal+0.090
arre+0.089
umpy+0.084
vious+0.080
awk+0.080
Neg logit contributions
lf-0.082
 cardboard-0.078
 camps-0.071
ts-0.069
 trek-0.066
Token making
Token ID1642
Feature activation-4.64e-02

Pos logit contributions
eal+0.031
arre+0.030
irk+0.028
umpy+0.027
vious+0.026
Neg logit contributions
 cardboard-0.024
lf-0.023
ts-0.022
 camps-0.020
 pretend-0.020
Token my
Token ID616
Feature activation-7.68e-02

Pos logit contributions
eal+0.109
irk+0.103
arre+0.102
umpy+0.100
olding+0.095
Neg logit contributions
 cardboard-0.052
ts-0.047
lf-0.046
 pretend-0.043
 camps-0.038
Token dog
Token ID3290
Feature activation+3.80e-03

Pos logit contributions
arre+0.147
eal+0.146
umpy+0.138
irk+0.136
 aback+0.134
Neg logit contributions
 cardboard-0.109
 pretend-0.107
lf-0.099
ts-0.097
 hunt-0.096
Token mad
Token ID8805
Feature activation-1.78e-01

Pos logit contributions
 cardboard+0.006
 pretend+0.006
ts+0.006
lf+0.006
 hunt+0.005
Neg logit contributions
eal-0.007
irk-0.007
arre-0.006
ags-0.006
umpy-0.006
Token.
Token ID13
Feature activation-3.80e-02

Pos logit contributions
eal+0.386
arre+0.365
irk+0.351
agree+0.333
Fighting+0.330
Neg logit contributions
 cardboard-0.209
lf-0.203
ts-0.188
 camps-0.178
 pretend-0.175
Token He
Token ID679
Feature activation+7.90e-02

Pos logit contributions
eal+0.063
arre+0.062
irk+0.054
 fault+0.051
awk+0.050
Neg logit contributions
 cardboard-0.063
lf-0.061
 called-0.057
 pretend-0.057
 named-0.056
Token does
Token ID857
Feature activation-1.97e-02

Pos logit contributions
 cardboard+0.128
lf+0.127
ts+0.116
 camps+0.113
 trek+0.109
Neg logit contributions
eal-0.139
arre-0.135
umpy-0.125
irk-0.123
vious-0.119
Token not
Token ID407
Feature activation-4.37e-02

Pos logit contributions
eal+0.046
arre+0.045
irk+0.043
umpy+0.042
vious+0.040
Neg logit contributions
 cardboard-0.022
lf-0.021
ts-0.021
 pretend-0.020
 pretended-0.017
Token like
Token ID588
Feature activation+1.62e-02

Pos logit contributions
eal+0.086
arre+0.085
irk+0.079
umpy+0.077
ags+0.074
Neg logit contributions
 cardboard-0.060
ts-0.057
lf-0.056
 pretend-0.054
 pretended-0.052
Token They
Token ID1119
Feature activation+5.67e-02

Pos logit contributions
eal+0.067
arre+0.065
umpy+0.060
irk+0.059
vious+0.058
Neg logit contributions
 cardboard-0.052
lf-0.049
ts-0.046
 camps-0.044
 pretend-0.044
Token told
Token ID1297
Feature activation-3.96e-02

Pos logit contributions
 cardboard+0.096
lf+0.095
 camps+0.091
 fossils+0.085
 dice+0.084
Neg logit contributions
eal-0.087
 back-0.086
 disappointed-0.085
 mad-0.083
 ashamed-0.080
Token them
Token ID606
Feature activation-1.12e-01

Pos logit contributions
eal+0.075
arre+0.074
umpy+0.070
irk+0.068
vious+0.064
Neg logit contributions
 cardboard-0.061
lf-0.058
ts-0.056
 pretend-0.051
 camps-0.051
Token what
Token ID644
Feature activation+1.45e-03

Pos logit contributions
eal+0.244
arre+0.233
irk+0.224
umpy+0.222
ags+0.215
Neg logit contributions
 cardboard-0.142
lf-0.135
ts-0.129
 camps-0.115
 pretend-0.115
Token they
Token ID484
Feature activation-1.98e-02

Pos logit contributions
 cardboard+0.002
lf+0.002
ts+0.002
 camps+0.002
 blasted+0.002
Neg logit contributions
eal-0.003
arre-0.003
umpy-0.003
irk-0.003
ags-0.003
Token saw
Token ID2497
Feature activation+9.11e-02

Pos logit contributions
eal+0.042
arre+0.041
umpy+0.039
irk+0.039
vious+0.038
Neg logit contributions
 cardboard-0.025
lf-0.025
 camps-0.022
ts-0.021
 trek-0.020
Token.
Token ID13
Feature activation-3.71e-02

Pos logit contributions
 cardboard+0.142
lf+0.139
 camps+0.126
ts+0.123
 blasted+0.122
Neg logit contributions
eal-0.163
arre-0.156
umpy-0.150
irk-0.148
vious-0.144
Token
Token ID198
Feature activation-1.68e-02

Pos logit contributions
eal+0.082
arre+0.078
vious+0.072
umpy+0.072
awk+0.071
Neg logit contributions
 cardboard-0.047
lf-0.040
 pretend-0.038
 called-0.037
 shovel-0.037
Token
Token ID198
Feature activation-1.64e-02

Pos logit contributions
eal+0.025
arre+0.024
umpy+0.023
irk+0.023
ags+0.021
Neg logit contributions
 cardboard-0.031
lf-0.030
ts-0.029
 camps-0.028
 blasted-0.027
Token"
Token ID1
Feature activation-2.89e-02

Pos logit contributions
eal+0.026
arre+0.026
umpy+0.024
irk+0.023
Fighting+0.022
Neg logit contributions
 cardboard-0.029
lf-0.028
ts-0.026
 camps-0.026
 trek-0.025
TokenMom
Token ID29252
Feature activation+1.76e-02

Pos logit contributions
eal+0.053
arre+0.050
irk+0.047
umpy+0.047
ags+0.045
Neg logit contributions
 cardboard-0.046
lf-0.046
ts-0.043
 camps-0.040
 blasted-0.039
Token day
Token ID1110
Feature activation-8.31e-03

Pos logit contributions
 cardboard+0.142
lf+0.139
ts+0.129
 camps+0.125
 blasted+0.121
Neg logit contributions
eal-0.155
arre-0.151
irk-0.139
umpy-0.136
ags-0.132
Token,
Token ID11
Feature activation+6.14e-02

Pos logit contributions
eal+0.012
 back+0.012
arre+0.012
ealing+0.011
awk+0.011
Neg logit contributions
lf-0.015
 cardboard-0.014
ts-0.013
 camps-0.013
 trek-0.012
Token the
Token ID262
Feature activation-3.45e-02

Pos logit contributions
lf+0.117
 cardboard+0.108
 camps+0.105
 trek+0.100
ts+0.098
Neg logit contributions
eal-0.076
 back-0.074
 disappointed-0.072
 mad-0.070
 ashamed-0.065
Token hipp
Token ID18568
Feature activation+1.71e-02

Pos logit contributions
eal+0.061
arre+0.058
umpy+0.057
 back+0.053
ags+0.053
Neg logit contributions
lf-0.054
 cardboard-0.053
ts-0.049
 trek-0.047
 blasted-0.046
Tokeno
Token ID78
Feature activation+3.19e-02

Pos logit contributions
 cardboard+0.035
lf+0.034
ts+0.032
 camps+0.031
 trek+0.030
Neg logit contributions
eal-0.023
arre-0.023
umpy-0.021
irk-0.020
vious-0.019
Token saw
Token ID2497
Feature activation+9.26e-02

Pos logit contributions
 cardboard+0.043
lf+0.042
ts+0.038
 camps+0.035
 trek+0.033
Neg logit contributions
arre-0.065
eal-0.065
umpy-0.062
vious-0.058
irk-0.057
Token a
Token ID257
Feature activation+1.57e-01

Pos logit contributions
lf+0.116
 cardboard+0.111
ts+0.099
 camps+0.098
 trek+0.095
Neg logit contributions
eal-0.191
arre-0.186
umpy-0.175
vious-0.175
 back-0.174
Token little
Token ID1310
Feature activation+9.85e-02

Pos logit contributions
lf+0.220
 cardboard+0.215
ts+0.211
 camps+0.191
 trek+0.190
Neg logit contributions
eal-0.293
arre-0.287
umpy-0.272
irk-0.270
 back-0.255
Token bird
Token ID6512
Feature activation+1.55e-02

Pos logit contributions
 cardboard+0.153
lf+0.144
ts+0.135
 camps+0.132
 shovel+0.127
Neg logit contributions
eal-0.183
arre-0.178
umpy-0.164
irk-0.163
ags-0.160
Token.
Token ID13
Feature activation-1.51e-02

Pos logit contributions
lf+0.023
 cardboard+0.022
ts+0.021
 camps+0.020
 fossils+0.019
Neg logit contributions
eal-0.028
arre-0.028
umpy-0.027
irk-0.026
vious-0.026
Token The
Token ID383
Feature activation+3.09e-02

Pos logit contributions
arre+0.024
eal+0.024
irk+0.021
umpy+0.021
ags+0.020
Neg logit contributions
 cardboard-0.027
lf-0.026
ts-0.025
 blasted-0.024
 camps-0.023
Token boy
Token ID2933
Feature activation+4.31e-02

Pos logit contributions
 cardboard+0.213
lf+0.211
ts+0.199
 camps+0.186
 trek+0.181
Neg logit contributions
eal-0.307
arre-0.300
irk-0.280
umpy-0.280
vious-0.262
Token who
Token ID508
Feature activation+6.67e-02

Pos logit contributions
 cardboard+0.058
lf+0.057
 camps+0.054
ts+0.054
 blasted+0.051
Neg logit contributions
eal-0.084
arre-0.083
irk-0.079
umpy-0.077
 mad-0.077
Token was
Token ID373
Feature activation+3.45e-02

Pos logit contributions
 cardboard+0.076
lf+0.072
ts+0.066
 camps+0.065
 blasted+0.060
Neg logit contributions
eal-0.147
arre-0.144
umpy-0.137
irk-0.137
vious-0.131
Token playing
Token ID2712
Feature activation+1.55e-02

Pos logit contributions
 cardboard+0.062
lf+0.061
ts+0.057
 camps+0.055
 blasted+0.053
Neg logit contributions
eal-0.056
arre-0.053
irk-0.049
umpy-0.049
ags-0.046
Token with
Token ID351
Feature activation-2.53e-02

Pos logit contributions
 cardboard+0.019
lf+0.018
ts+0.016
 camps+0.015
 blasted+0.015
Neg logit contributions
eal-0.034
arre-0.033
irk-0.031
umpy-0.031
ags-0.030
Token a
Token ID257
Feature activation+1.57e-01

Pos logit contributions
eal+0.040
arre+0.040
 back+0.039
ags+0.038
irk+0.037
Neg logit contributions
lf-0.040
 blasted-0.038
 camps-0.036
 cardboard-0.036
 trek-0.035
Token toy
Token ID13373
Feature activation+2.06e-02

Pos logit contributions
 cardboard+0.216
lf+0.214
ts+0.200
 camps+0.188
 blasted+0.183
Neg logit contributions
eal-0.308
arre-0.300
umpy-0.280
irk-0.280
ags-0.265
Token that
Token ID326
Feature activation-7.46e-03

Pos logit contributions
lf+0.035
ts+0.035
 trek+0.034
 cardboard+0.031
 blasted+0.031
Neg logit contributions
 back-0.031
arre-0.028
eal-0.027
umpy-0.027
vious-0.027
Token she
Token ID673
Feature activation-3.22e-03

Pos logit contributions
eal+0.016
arre+0.016
irk+0.015
umpy+0.015
vious+0.014
Neg logit contributions
 cardboard-0.008
lf-0.008
ts-0.008
 camps-0.007
 trek-0.007
Token wanted
Token ID2227
Feature activation+5.00e-02

Pos logit contributions
eal+0.006
arre+0.006
umpy+0.006
 disappointed+0.005
vious+0.005
Neg logit contributions
 cardboard-0.005
 camps-0.004
lf-0.004
 shovel-0.004
 seas-0.004
Token to
Token ID284
Feature activation-3.60e-03

Pos logit contributions
 cardboard+0.095
lf+0.094
ts+0.089
 camps+0.086
 blasted+0.084
Neg logit contributions
eal-0.072
arre-0.069
umpy-0.064
irk-0.063
ags-0.059
Token his
Token ID465
Feature activation-4.03e-02

Pos logit contributions
lf+0.057
ts+0.056
 blasted+0.056
 cardboard+0.056
 pretended+0.054
Neg logit contributions
eal-0.036
 back-0.035
irk-0.033
 mad-0.033
ags-0.033
Token garden
Token ID11376
Feature activation+5.96e-02

Pos logit contributions
eal+0.080
arre+0.078
umpy+0.073
irk+0.073
ags+0.069
Neg logit contributions
 cardboard-0.056
lf-0.055
ts-0.051
 camps-0.048
 blasted-0.047
Token.
Token ID13
Feature activation-1.89e-02

Pos logit contributions
 cardboard+0.093
lf+0.092
ts+0.088
 camps+0.083
 blasted+0.083
Neg logit contributions
eal-0.101
arre-0.100
umpy-0.094
irk-0.092
ags-0.088
Token
Token ID198
Feature activation-5.79e-03

Pos logit contributions
eal+0.034
arre+0.033
umpy+0.030
irk+0.030
ags+0.029
Neg logit contributions
 cardboard-0.030
lf-0.029
ts-0.027
 camps-0.026
 trek-0.025
Token
Token ID198
Feature activation-5.12e-03

Pos logit contributions
eal+0.009
arre+0.009
irk+0.008
umpy+0.008
ags+0.007
Neg logit contributions
 cardboard-0.010
lf-0.010
ts-0.010
 camps-0.009
 blasted-0.009
TokenHe
Token ID1544
Feature activation+1.00e-01

Pos logit contributions
eal+0.009
arre+0.009
umpy+0.008
irk+0.008
vious+0.008
Neg logit contributions
 cardboard-0.008
lf-0.008
ts-0.008
 camps-0.007
 blasted-0.007
Token picked
Token ID6497
Feature activation+1.80e-02

Pos logit contributions
 cardboard+0.160
 camps+0.155
 fossils+0.140
 loot+0.140
 miles+0.139
Neg logit contributions
eal-0.163
arre-0.143
 back-0.138
 disappointed-0.138
 mad-0.134
Token up
Token ID510
Feature activation-7.23e-02

Pos logit contributions
lf+0.031
 cardboard+0.031
ts+0.029
 camps+0.028
 blasted+0.027
Neg logit contributions
eal-0.029
arre-0.028
irk-0.026
umpy-0.026
vious-0.024
Token the
Token ID262
Feature activation-3.79e-02

Pos logit contributions
eal+0.121
arre+0.113
 back+0.112
vious+0.106
irk+0.104
Neg logit contributions
lf-0.118
 cardboard-0.110
 camps-0.106
 blasted-0.104
ts-0.103
Token mel
Token ID7758
Feature activation+7.20e-03

Pos logit contributions
eal+0.074
arre+0.072
umpy+0.068
irk+0.068
vious+0.065
Neg logit contributions
lf-0.052
 cardboard-0.051
ts-0.050
 camps-0.046
 blasted-0.045
Tokenon
Token ID261
Feature activation+1.69e-02

Pos logit contributions
 cardboard+0.010
ts+0.009
lf+0.009
 camps+0.009
 trek+0.008
Neg logit contributions
eal-0.014
arre-0.014
umpy-0.013
irk-0.013
vious-0.012
Token you
Token ID345
Feature activation-1.06e-01

Pos logit contributions
 cardboard+0.075
lf+0.073
ts+0.065
 camps+0.063
 trek+0.061
Neg logit contributions
eal-0.117
arre-0.117
irk-0.108
umpy-0.107
ags-0.106
Token can
Token ID460
Feature activation+3.58e-02

Pos logit contributions
eal+0.222
arre+0.214
umpy+0.202
irk+0.197
vious+0.195
Neg logit contributions
lf-0.142
 cardboard-0.137
ts-0.127
 camps-0.125
 trek-0.114
Token spin
Token ID7906
Feature activation-1.56e-02

Pos logit contributions
 cardboard+0.046
lf+0.044
ts+0.040
 pretend+0.039
 blasted+0.038
Neg logit contributions
eal-0.074
arre-0.073
irk-0.070
umpy-0.068
ags-0.065
Token too
Token ID1165
Feature activation-1.29e-01

Pos logit contributions
eal+0.032
arre+0.031
irk+0.029
umpy+0.029
olding+0.027
Neg logit contributions
 cardboard-0.022
lf-0.021
ts-0.020
 pretend-0.019
 camps-0.018
Token!"
Token ID2474
Feature activation-3.03e-02

Pos logit contributions
eal+0.237
arre+0.234
umpy+0.230
irk+0.217
vious+0.210
Neg logit contributions
 cardboard-0.188
lf-0.187
ts-0.178
 camps-0.163
 blasted-0.162
Token So
Token ID1406
Feature activation+9.51e-02

Pos logit contributions
arre+0.058
eal+0.056
umpy+0.055
 fault+0.052
irk+0.052
Neg logit contributions
 cardboard-0.043
lf-0.040
 called-0.038
ts-0.037
 trek-0.037
Token the
Token ID262
Feature activation-3.92e-02

Pos logit contributions
 cardboard+0.113
lf+0.110
ts+0.099
 camps+0.093
 blasted+0.091
Neg logit contributions
eal-0.210
arre-0.206
umpy-0.195
irk-0.194
ags-0.186
Token little
Token ID1310
Feature activation+9.97e-02

Pos logit contributions
arre+0.066
eal+0.061
irk+0.057
umpy+0.056
ags+0.055
Neg logit contributions
 cardboard-0.067
lf-0.066
ts-0.063
 camps-0.061
 pretend-0.060
Token girl
Token ID2576
Feature activation+2.78e-02

Pos logit contributions
 cardboard+0.156
lf+0.144
 camps+0.143
 shovel+0.143
 pretend+0.141
Neg logit contributions
arre-0.184
eal-0.159
ags-0.154
umpy-0.154
agree-0.151
Token and
Token ID290
Feature activation-1.12e-02

Pos logit contributions
 cardboard+0.034
lf+0.033
ts+0.030
 camps+0.029
 blasted+0.028
Neg logit contributions
eal-0.059
arre-0.058
umpy-0.054
irk-0.054
ags-0.052
Token the
Token ID262
Feature activation-3.88e-02

Pos logit contributions
arre+0.026
umpy+0.026
eal+0.026
irk+0.025
ags+0.024
Neg logit contributions
lf-0.012
 cardboard-0.012
ts-0.011
 pretended-0.010
 trek-0.010
Token underneath
Token ID14638
Feature activation+3.15e-02

Pos logit contributions
eal+0.017
 back+0.016
ags+0.016
irk+0.016
 mad+0.015
Neg logit contributions
lf-0.015
 cardboard-0.015
ts-0.014
 blasted-0.014
 trek-0.013
Token the
Token ID262
Feature activation-3.22e-02

Pos logit contributions
lf+0.046
 cardboard+0.045
ts+0.043
 camps+0.041
 blasted+0.040
Neg logit contributions
eal-0.058
arre-0.056
irk-0.053
umpy-0.052
ags-0.050
Token carpet
Token ID20710
Feature activation+8.88e-02

Pos logit contributions
eal+0.066
arre+0.064
umpy+0.062
irk+0.061
 back+0.059
Neg logit contributions
lf-0.039
 cardboard-0.038
ts-0.038
 trek-0.034
 camps-0.034
Token.
Token ID13
Feature activation-1.87e-02

Pos logit contributions
 cardboard+0.126
lf+0.123
ts+0.116
 camps+0.110
 blasted+0.105
Neg logit contributions
eal-0.171
arre-0.166
umpy-0.156
irk-0.155
vious-0.147
Token The
Token ID383
Feature activation+3.15e-02

Pos logit contributions
eal+0.034
arre+0.034
umpy+0.031
ags+0.031
irk+0.031
Neg logit contributions
 cardboard-0.029
lf-0.027
 called-0.025
 shovel-0.024
 pretend-0.024
Token puppet
Token ID30095
Feature activation+4.49e-03

Pos logit contributions
lf+0.052
 cardboard+0.052
ts+0.052
 trek+0.049
 camps+0.047
Neg logit contributions
eal-0.052
arre-0.047
irk-0.047
umpy-0.046
 back-0.044
Token had
Token ID550
Feature activation+8.56e-02

Pos logit contributions
 cardboard+0.005
lf+0.005
ts+0.005
 camps+0.004
 blasted+0.004
Neg logit contributions
eal-0.010
arre-0.010
umpy-0.009
irk-0.009
ags-0.009
Token a
Token ID257
Feature activation+1.58e-01

Pos logit contributions
lf+0.098
 cardboard+0.098
ts+0.088
 camps+0.084
 blasted+0.079
Neg logit contributions
eal-0.188
arre-0.182
umpy-0.173
irk-0.172
vious-0.165
Token friendly
Token ID8030
Feature activation-3.11e-02

Pos logit contributions
 cardboard+0.209
lf+0.207
ts+0.193
 camps+0.181
 blasted+0.174
Neg logit contributions
eal-0.321
arre-0.312
umpy-0.292
irk-0.292
ags-0.276
Token face
Token ID1986
Feature activation-5.90e-02

Pos logit contributions
arre+0.053
eal+0.052
umpy+0.049
Fighting+0.047
vious+0.047
Neg logit contributions
 cardboard-0.054
lf-0.051
 pretend-0.046
 shovel-0.045
ts-0.045
Token and
Token ID290
Feature activation-8.96e-03

Pos logit contributions
eal+0.114
arre+0.111
umpy+0.105
irk+0.104
vious+0.099
Neg logit contributions
 cardboard-0.082
lf-0.080
ts-0.076
 camps-0.072
 trek-0.069
Token and
Token ID290
Feature activation-1.50e-02

Pos logit contributions
 cardboard+0.062
lf+0.060
ts+0.055
 camps+0.052
 blasted+0.050
Neg logit contributions
eal-0.101
arre-0.098
irk-0.093
umpy-0.092
ags-0.089
Token put
Token ID1234
Feature activation+2.83e-03

Pos logit contributions
eal+0.030
arre+0.028
irk+0.027
 back+0.026
umpy+0.026
Neg logit contributions
lf-0.020
 cardboard-0.020
ts-0.018
 camps-0.018
 trek-0.017
Token it
Token ID340
Feature activation-1.28e-01

Pos logit contributions
 cardboard+0.004
lf+0.003
ts+0.003
 camps+0.003
 shovel+0.003
Neg logit contributions
eal-0.006
arre-0.006
umpy-0.006
irk-0.006
vious-0.005
Token on
Token ID319
Feature activation-1.54e-02

Pos logit contributions
eal+0.258
arre+0.251
irk+0.237
umpy+0.235
vious+0.222
Neg logit contributions
 cardboard-0.180
lf-0.172
ts-0.162
 camps-0.153
 trek-0.147
Token a
Token ID257
Feature activation+1.56e-01

Pos logit contributions
eal+0.030
arre+0.029
umpy+0.027
irk+0.027
ags+0.026
Neg logit contributions
 cardboard-0.022
lf-0.021
ts-0.020
 camps-0.019
 blasted-0.018
Token leaf
Token ID12835
Feature activation+1.58e-02

Pos logit contributions
 cardboard+0.209
lf+0.207
ts+0.194
 camps+0.181
 blasted+0.178
Neg logit contributions
eal-0.310
arre-0.299
umpy-0.284
irk-0.281
ags-0.267
Token.
Token ID13
Feature activation-1.97e-02

Pos logit contributions
 cardboard+0.021
lf+0.021
ts+0.019
 camps+0.018
 blasted+0.017
Neg logit contributions
eal-0.032
arre-0.031
irk-0.030
umpy-0.029
ags-0.028
Token She
Token ID1375
Feature activation+5.23e-02

Pos logit contributions
eal+0.033
arre+0.032
umpy+0.030
irk+0.030
ags+0.028
Neg logit contributions
 cardboard-0.033
lf-0.033
ts-0.030
 camps-0.029
 trek-0.028
Token wanted
Token ID2227
Feature activation+4.85e-02

Pos logit contributions
lf+0.074
 cardboard+0.073
ts+0.065
 camps+0.063
 blasted+0.060
Neg logit contributions
eal-0.099
arre-0.094
vious-0.085
irk-0.085
 disappointed-0.084
Token to
Token ID284
Feature activation-4.26e-03

Pos logit contributions
 cardboard+0.086
lf+0.084
ts+0.079
 camps+0.076
 trek+0.074
Neg logit contributions
eal-0.080
arre-0.077
umpy-0.071
irk-0.071
ags-0.066
Token weigh
Token ID10164
Feature activation+1.07e-02

Pos logit contributions
eal+0.008
arre+0.007
umpy+0.007
awk+0.006
 back+0.006
Neg logit contributions
lf-0.007
 cardboard-0.006
 camps-0.006
ts-0.006
 trek-0.006
Token From
Token ID3574
Feature activation+1.30e-02

Pos logit contributions
eal+0.067
arre+0.063
umpy+0.059
irk+0.059
vious+0.058
Neg logit contributions
 cardboard-0.045
lf-0.043
 camps-0.038
 pretend-0.037
 called-0.037
Token that
Token ID326
Feature activation-8.01e-03

Pos logit contributions
 cardboard+0.017
lf+0.016
ts+0.015
 blasted+0.015
 camps+0.015
Neg logit contributions
eal-0.026
arre-0.025
irk-0.024
umpy-0.024
ags-0.023
Token day
Token ID1110
Feature activation-1.43e-02

Pos logit contributions
eal+0.016
arre+0.015
irk+0.014
umpy+0.014
ags+0.014
Neg logit contributions
 cardboard-0.012
lf-0.012
 camps-0.011
ts-0.010
 trek-0.010
Token on
Token ID319
Feature activation-1.68e-02

Pos logit contributions
eal+0.026
arre+0.026
umpy+0.024
irk+0.024
ags+0.023
Neg logit contributions
 cardboard-0.022
lf-0.021
ts-0.020
 camps-0.019
 trek-0.018
Token,
Token ID11
Feature activation+5.28e-02

Pos logit contributions
eal+0.037
arre+0.036
umpy+0.034
irk+0.034
ags+0.033
Neg logit contributions
 cardboard-0.018
lf-0.018
ts-0.017
 camps-0.016
 blasted-0.015
Token Lily
Token ID20037
Feature activation+3.87e-03

Pos logit contributions
 cardboard+0.077
lf+0.076
ts+0.070
 camps+0.067
 trek+0.064
Neg logit contributions
eal-0.102
arre-0.098
irk-0.092
umpy-0.092
ags-0.088
Token learned
Token ID4499
Feature activation-1.67e-02

Pos logit contributions
lf+0.004
 cardboard+0.004
 pretended+0.004
ts+0.004
 blasted+0.004
Neg logit contributions
eal-0.009
irk-0.008
arre-0.008
ags-0.008
awk-0.008
Token to
Token ID284
Feature activation-6.21e-03

Pos logit contributions
eal+0.044
arre+0.040
irk+0.039
umpy+0.039
ealing+0.039
Neg logit contributions
 cardboard-0.017
 called-0.014
lf-0.013
 treasure-0.012
 pretend-0.012
Token be
Token ID307
Feature activation-7.14e-02

Pos logit contributions
eal+0.012
arre+0.012
umpy+0.011
irk+0.011
vious+0.011
Neg logit contributions
 cardboard-0.009
lf-0.008
ts-0.008
 pretend-0.008
 trek-0.007
Token serious
Token ID2726
Feature activation-7.82e-02

Pos logit contributions
arre+0.138
eal+0.138
irk+0.131
umpy+0.124
awk+0.123
Neg logit contributions
 cardboard-0.100
ts-0.094
 called-0.092
 hours-0.090
 pretend-0.089
Token when
Token ID618
Feature activation+5.29e-02

Pos logit contributions
eal+0.154
arre+0.150
umpy+0.141
irk+0.140
ags+0.133
Neg logit contributions
 cardboard-0.109
lf-0.107
ts-0.099
 camps-0.094
 blasted-0.091
Token crash
Token ID7014
Feature activation+3.37e-02

Pos logit contributions
eal+0.033
arre+0.033
irk+0.030
umpy+0.030
 back+0.029
Neg logit contributions
lf-0.024
 cardboard-0.024
 camps-0.022
ts-0.022
 trek-0.020
Token.
Token ID13
Feature activation-2.21e-02

Pos logit contributions
 cardboard+0.050
lf+0.050
ts+0.046
 camps+0.044
 trek+0.043
Neg logit contributions
eal-0.062
arre-0.060
umpy-0.056
irk-0.055
ags-0.053
Token
Token ID198
Feature activation-6.87e-03

Pos logit contributions
eal+0.047
arre+0.045
umpy+0.043
irk+0.042
awk+0.041
Neg logit contributions
 cardboard-0.027
 camps-0.025
lf-0.024
 pretend-0.023
 trek-0.023
Token
Token ID198
Feature activation-6.34e-03

Pos logit contributions
eal+0.012
arre+0.011
umpy+0.011
irk+0.010
ags+0.010
Neg logit contributions
 cardboard-0.011
lf-0.011
ts-0.011
 camps-0.010
 blasted-0.010
TokenOne
Token ID3198
Feature activation+8.57e-02

Pos logit contributions
eal+0.015
arre+0.014
irk+0.014
umpy+0.014
ags+0.013
Neg logit contributions
 cardboard-0.007
lf-0.006
ts-0.006
 camps-0.005
 trek-0.005
Token day
Token ID1110
Feature activation-1.03e-02

Pos logit contributions
 cardboard+0.131
lf+0.130
 camps+0.115
ts+0.114
 pretend+0.111
Neg logit contributions
eal-0.162
arre-0.159
irk-0.147
ags-0.142
umpy-0.141
Token,
Token ID11
Feature activation+5.89e-02

Pos logit contributions
eal+0.017
arre+0.016
 back+0.015
ealing+0.015
 forward+0.014
Neg logit contributions
lf-0.016
 cardboard-0.016
ts-0.015
 blasted-0.014
 camps-0.014
Token he
Token ID339
Feature activation+1.88e-02

Pos logit contributions
 cardboard+0.107
lf+0.106
ts+0.099
 blasted+0.094
 trek+0.093
Neg logit contributions
eal-0.087
arre-0.081
 back-0.075
 disappointed-0.074
irk-0.074
Token saw
Token ID2497
Feature activation+9.20e-02

Pos logit contributions
 cardboard+0.029
ts+0.028
lf+0.028
 fossils+0.025
 loot+0.024
Neg logit contributions
arre-0.034
eal-0.033
umpy-0.032
 back-0.030
vious-0.029
Token a
Token ID257
Feature activation+1.55e-01

Pos logit contributions
lf+0.113
 cardboard+0.109
ts+0.101
 camps+0.098
 blasted+0.097
Neg logit contributions
eal-0.186
arre-0.182
irk-0.171
 back-0.170
vious-0.169
Token new
Token ID649
Feature activation-5.80e-02

Pos logit contributions
ts+0.223
lf+0.223
 cardboard+0.220
 trek+0.201
 blasted+0.198
Neg logit contributions
eal-0.273
arre-0.268
irk-0.254
umpy-0.252
 back-0.245
Token,
Token ID11
Feature activation+6.25e-02

Pos logit contributions
ts+0.015
 cardboard+0.014
lf+0.014
 camps+0.013
 seas+0.013
Neg logit contributions
arre-0.014
eal-0.014
umpy-0.013
 back-0.013
vious-0.013
Token a
Token ID257
Feature activation+1.58e-01

Pos logit contributions
lf+0.088
 cardboard+0.086
ts+0.081
 camps+0.076
 blasted+0.072
Neg logit contributions
eal-0.119
arre-0.116
irk-0.111
 back-0.107
umpy-0.106
Token man
Token ID582
Feature activation-4.78e-05

Pos logit contributions
 cardboard+0.202
lf+0.196
ts+0.176
 camps+0.170
 blasted+0.162
Neg logit contributions
eal-0.334
arre-0.324
irk-0.304
umpy-0.304
ags-0.295
Token named
Token ID3706
Feature activation+9.97e-02

Pos logit contributions
eal+0.000
arre+0.000
 back+0.000
irk+0.000
aring+0.000
Neg logit contributions
 cardboard-0.000
lf-0.000
ts-0.000
 camps-0.000
 seas-0.000
Token John
Token ID1757
Feature activation+2.76e-02

Pos logit contributions
lf+0.179
 cardboard+0.173
ts+0.163
 hours+0.157
 blasted+0.157
Neg logit contributions
eal-0.140
arre-0.138
 back-0.128
umpy-0.127
 aback-0.124
Token,
Token ID11
Feature activation+6.21e-02

Pos logit contributions
lf+0.042
 cardboard+0.041
ts+0.038
 seas+0.037
 dice+0.036
Neg logit contributions
 back-0.048
eal-0.047
arre-0.046
umpy-0.043
irk-0.043
Token came
Token ID1625
Feature activation+2.19e-02

Pos logit contributions
lf+0.093
 cardboard+0.091
ts+0.086
 camps+0.078
 seas+0.077
Neg logit contributions
arre-0.113
eal-0.113
 back-0.105
irk-0.104
umpy-0.104
Token to
Token ID284
Feature activation+1.24e-03

Pos logit contributions
 cardboard+0.034
 camps+0.031
lf+0.030
 trek+0.030
 called+0.030
Neg logit contributions
eal-0.040
arre-0.038
irk-0.037
ealing-0.035
agree-0.035
Token take
Token ID1011
Feature activation-3.72e-02

Pos logit contributions
lf+0.002
 cardboard+0.002
ts+0.002
 camps+0.002
 blasted+0.002
Neg logit contributions
eal-0.002
arre-0.002
irk-0.002
 back-0.002
ags-0.002
Token him
Token ID683
Feature activation-5.48e-02

Pos logit contributions
eal+0.074
arre+0.072
umpy+0.066
irk+0.066
ags+0.065
Neg logit contributions
lf-0.052
 cardboard-0.050
ts-0.046
 camps-0.043
 blasted-0.043
Token for
Token ID329
Feature activation-6.65e-02

Pos logit contributions
eal+0.090
arre+0.084
irk+0.081
umpy+0.079
Fighting+0.075
Neg logit contributions
 cardboard-0.100
lf-0.091
ts-0.088
 camps-0.088
 trek-0.085
Token.
Token ID13
Feature activation-3.56e-02

Pos logit contributions
 cardboard+0.094
lf+0.092
ts+0.085
 camps+0.080
 blasted+0.077
Neg logit contributions
eal-0.139
arre-0.135
irk-0.126
umpy-0.126
ags-0.120
Token He
Token ID679
Feature activation+8.01e-02

Pos logit contributions
eal+0.058
arre+0.056
umpy+0.051
irk+0.051
ags+0.048
Neg logit contributions
 cardboard-0.063
lf-0.062
ts-0.058
 camps-0.056
 blasted-0.055
Token says
Token ID1139
Feature activation-3.43e-02

Pos logit contributions
lf+0.140
 cardboard+0.140
ts+0.128
 camps+0.124
 blasted+0.122
Neg logit contributions
eal-0.129
arre-0.125
umpy-0.114
irk-0.113
vious-0.109
Token,
Token ID11
Feature activation+4.39e-02

Pos logit contributions
eal+0.101
arre+0.097
irk+0.094
umpy+0.093
ags+0.092
Neg logit contributions
 cardboard-0.021
lf-0.017
ts-0.014
 shovel-0.012
 swords-0.012
Token "
Token ID366
Feature activation-8.13e-03

Pos logit contributions
 cardboard+0.056
lf+0.054
ts+0.048
 shovel+0.045
 pretend+0.045
Neg logit contributions
eal-0.097
arre-0.092
irk-0.088
umpy-0.088
awk-0.087
TokenNo
Token ID2949
Feature activation-6.46e-02

Pos logit contributions
eal+0.014
arre+0.013
irk+0.012
ags+0.012
 fault+0.012
Neg logit contributions
lf-0.013
 cardboard-0.013
ts-0.013
 pretend-0.012
 pretended-0.012
Token,
Token ID11
Feature activation+4.68e-02

Pos logit contributions
eal+0.148
arre+0.145
irk+0.137
umpy+0.136
ags+0.131
Neg logit contributions
 cardboard-0.069
lf-0.067
ts-0.062
 camps-0.057
 blasted-0.055
Token you
Token ID345
Feature activation-1.05e-01

Pos logit contributions
 cardboard+0.075
lf+0.074
ts+0.070
 camps+0.068
 trek+0.066
Neg logit contributions
eal-0.083
arre-0.079
irk-0.073
umpy-0.073
vious-0.070
Token go
Token ID467
Feature activation-4.01e-02

Pos logit contributions
arre+0.186
eal+0.173
umpy+0.172
vious+0.161
ealing+0.159
Neg logit contributions
lf-0.179
 cardboard-0.175
 camps-0.162
ts-0.161
 trek-0.159
Token away
Token ID1497
Feature activation-9.95e-02

Pos logit contributions
eal+0.055
arre+0.053
umpy+0.050
 back+0.047
irk+0.047
Neg logit contributions
 cardboard-0.080
lf-0.080
ts-0.075
 camps-0.072
 trek-0.072
Token.
Token ID13
Feature activation-3.86e-02

Pos logit contributions
umpy+0.160
arre+0.158
eal+0.156
 back+0.146
irk+0.146
Neg logit contributions
 cardboard-0.166
lf-0.161
ts-0.152
 trek-0.151
 blasted-0.146
Token poke
Token ID22620
Feature activation-1.68e-03

Pos logit contributions
eal+0.039
arre+0.037
irk+0.034
umpy+0.034
 back+0.034
Neg logit contributions
 cardboard-0.025
lf-0.025
 camps-0.023
ts-0.022
 trek-0.022
Token her
Token ID607
Feature activation-7.32e-02

Pos logit contributions
eal+0.003
arre+0.003
umpy+0.003
irk+0.003
Fighting+0.003
Neg logit contributions
 cardboard-0.003
lf-0.002
ts-0.002
 pretend-0.002
 camps-0.002
Token teeth
Token ID9941
Feature activation+2.42e-02

Pos logit contributions
eal+0.136
arre+0.134
irk+0.125
umpy+0.124
Fighting+0.117
Neg logit contributions
 cardboard-0.114
lf-0.110
ts-0.103
 camps-0.098
 pretend-0.097
Token.
Token ID13
Feature activation-2.64e-02

Pos logit contributions
 cardboard+0.032
lf+0.031
ts+0.028
 camps+0.026
 blasted+0.026
Neg logit contributions
eal-0.051
arre-0.049
irk-0.046
umpy-0.046
ags-0.044
Token It
Token ID632
Feature activation-1.19e-02

Pos logit contributions
eal+0.054
arre+0.051
irk+0.049
awk+0.048
umpy+0.048
Neg logit contributions
 cardboard-0.037
lf-0.035
ts-0.031
 camps-0.030
 pretend-0.029
Token hurt
Token ID5938
Feature activation-6.65e-02

Pos logit contributions
eal+0.025
arre+0.024
umpy+0.023
irk+0.022
vious+0.022
Neg logit contributions
 cardboard-0.015
lf-0.014
ts-0.013
 camps-0.013
 trek-0.012
Token and
Token ID290
Feature activation-2.08e-02

Pos logit contributions
eal+0.134
arre+0.130
umpy+0.122
irk+0.122
ags+0.116
Neg logit contributions
 cardboard-0.091
lf-0.089
ts-0.082
 camps-0.078
 blasted-0.075
Token tasted
Token ID29187
Feature activation-6.41e-03

Pos logit contributions
eal+0.040
arre+0.039
irk+0.036
umpy+0.036
ags+0.035
Neg logit contributions
 cardboard-0.031
lf-0.030
ts-0.028
 camps-0.026
 blasted-0.026
Token bad
Token ID2089
Feature activation-9.07e-02

Pos logit contributions
eal+0.013
arre+0.012
irk+0.012
umpy+0.011
agree+0.011
Neg logit contributions
 cardboard-0.009
ts-0.009
lf-0.009
 pretend-0.008
 camps-0.008
Token.
Token ID13
Feature activation-2.75e-02

Pos logit contributions
eal+0.184
arre+0.179
umpy+0.168
irk+0.168
ags+0.160
Neg logit contributions
 cardboard-0.122
lf-0.119
ts-0.110
 camps-0.104
 blasted-0.100
Token
Token ID198
Feature activation-1.01e-02

Pos logit contributions
eal+0.057
arre+0.053
irk+0.050
awk+0.049
umpy+0.049
Neg logit contributions
 cardboard-0.038
lf-0.037
ts-0.032
 camps-0.030
 pretend-0.030
Token again
Token ID757
Feature activation-1.14e-01

Pos logit contributions
eal+0.023
arre+0.023
umpy+0.022
irk+0.021
ags+0.021
Neg logit contributions
 cardboard-0.012
lf-0.012
ts-0.011
 camps-0.010
 blasted-0.010
Token.
Token ID13
Feature activation-2.01e-02

Pos logit contributions
eal+0.230
arre+0.221
irk+0.208
umpy+0.208
 back+0.201
Neg logit contributions
 cardboard-0.150
lf-0.147
ts-0.133
 camps-0.129
 trek-0.122
Token He
Token ID679
Feature activation+8.37e-02

Pos logit contributions
eal+0.037
arre+0.036
irk+0.033
umpy+0.033
vious+0.032
Neg logit contributions
 cardboard-0.030
lf-0.030
ts-0.027
 camps-0.027
 trek-0.027
Token kept
Token ID4030
Feature activation-2.07e-02

Pos logit contributions
 cardboard+0.102
lf+0.099
ts+0.090
 camps+0.085
 blasted+0.080
Neg logit contributions
eal-0.181
arre-0.175
umpy-0.164
irk-0.164
vious-0.157
Token jumping
Token ID14284
Feature activation-3.82e-02

Pos logit contributions
eal+0.049
umpy+0.049
arre+0.048
irk+0.046
vious+0.045
Neg logit contributions
 cardboard-0.022
lf-0.020
ts-0.019
 pretend-0.018
 camps-0.017
Token up
Token ID510
Feature activation-7.29e-02

Pos logit contributions
eal+0.079
arre+0.077
irk+0.073
umpy+0.073
ags+0.069
Neg logit contributions
 cardboard-0.050
lf-0.048
ts-0.045
 camps-0.043
 trek-0.041
Token and
Token ID290
Feature activation-1.22e-02

Pos logit contributions
eal+0.139
arre+0.135
umpy+0.126
irk+0.126
ags+0.120
Neg logit contributions
lf-0.104
 cardboard-0.103
ts-0.095
 camps-0.089
 blasted-0.088
Token down
Token ID866
Feature activation-6.36e-02

Pos logit contributions
eal+0.020
arre+0.019
umpy+0.019
irk+0.019
ags+0.018
Neg logit contributions
 cardboard-0.021
lf-0.021
 pretended-0.021
ts-0.020
 pretend-0.020
Token until
Token ID1566
Feature activation+6.91e-02

Pos logit contributions
eal+0.141
arre+0.137
irk+0.129
umpy+0.129
ags+0.123
Neg logit contributions
 cardboard-0.076
lf-0.073
ts-0.067
 camps-0.063
 trek-0.061
Token he
Token ID339
Feature activation+1.63e-02

Pos logit contributions
 cardboard+0.078
lf+0.076
ts+0.069
 camps+0.065
 blasted+0.062
Neg logit contributions
eal-0.153
arre-0.148
irk-0.139
umpy-0.139
vious-0.134
Token was
Token ID373
Feature activation+2.96e-02

Pos logit contributions
 cardboard+0.021
lf+0.020
ts+0.018
 camps+0.017
 shovel+0.016
Neg logit contributions
eal-0.035
arre-0.034
umpy-0.032
irk-0.031
vious-0.030
Token
Token ID220
Feature activation-9.28e-02

Pos logit contributions
eal+0.056
arre+0.053
umpy+0.049
irk+0.049
ags+0.046
Neg logit contributions
 cardboard-0.064
lf-0.063
ts-0.059
 camps-0.057
 blasted-0.055
Token
Token ID198
Feature activation-1.37e-02

Pos logit contributions
eal+0.217
arre+0.196
irk+0.192
vious+0.192
umpy+0.183
Neg logit contributions
 cardboard-0.095
lf-0.091
ts-0.079
 camps-0.076
 called-0.074
Token
Token ID198
Feature activation-1.35e-02

Pos logit contributions
eal+0.028
arre+0.027
irk+0.025
vious+0.025
umpy+0.025
Neg logit contributions
lf-0.018
 cardboard-0.018
ts-0.016
 camps-0.016
 trek-0.015
TokenJ
Token ID41
Feature activation+1.91e-03

Pos logit contributions
eal+0.028
arre+0.026
irk+0.025
umpy+0.024
ags+0.024
Neg logit contributions
lf-0.019
 cardboard-0.018
ts-0.016
 camps-0.015
 trek-0.015
Tokenenny
Token ID11870
Feature activation+4.00e-02

Pos logit contributions
lf+0.003
 cardboard+0.003
ts+0.003
 blasted+0.003
 trek+0.003
Neg logit contributions
eal-0.003
arre-0.003
agree-0.003
ags-0.003
Fighting-0.003
Token's
Token ID338
Feature activation-7.02e-02

Pos logit contributions
lf+0.046
 cardboard+0.044
ts+0.040
 pretended+0.038
 trek+0.037
Neg logit contributions
eal-0.087
arre-0.086
irk-0.083
umpy-0.082
ags-0.082
Token friends
Token ID2460
Feature activation+2.48e-02

Pos logit contributions
eal+0.117
arre+0.113
umpy+0.105
irk+0.104
ags+0.098
Neg logit contributions
 cardboard-0.120
lf-0.118
ts-0.110
 camps-0.106
 blasted-0.103
Token were
Token ID547
Feature activation+5.22e-02

Pos logit contributions
 cardboard+0.024
lf+0.023
ts+0.021
 pretended+0.019
 blasted+0.019
Neg logit contributions
eal-0.058
arre-0.056
umpy-0.055
irk-0.055
ags-0.054
Token very
Token ID845
Feature activation-3.29e-02

Pos logit contributions
 called+0.070
 cardboard+0.070
lf+0.068
 named+0.064
ts+0.064
Neg logit contributions
eal-0.106
irk-0.101
awk-0.096
arre-0.095
agree-0.093
Token excited
Token ID6568
Feature activation+6.00e-04

Pos logit contributions
eal+0.052
arre+0.050
ags+0.050
irk+0.047
agree+0.047
Neg logit contributions
 cardboard-0.054
lf-0.054
ts-0.052
 called-0.052
 pretend-0.050
Token when
Token ID618
Feature activation+6.47e-02

Pos logit contributions
lf+0.001
 cardboard+0.001
ts+0.001
 pretend+0.001
 camps+0.001
Neg logit contributions
eal-0.001
arre-0.001
umpy-0.001
irk-0.001
ags-0.001
Token.
Token ID13
Feature activation-2.44e-02

Pos logit contributions
eal+0.001
arre+0.001
umpy+0.001
irk+0.001
ags+0.001
Neg logit contributions
 cardboard-0.001
lf-0.001
ts-0.001
 camps-0.001
 blasted-0.001
Token She
Token ID1375
Feature activation+5.79e-02

Pos logit contributions
eal+0.044
arre+0.044
awk+0.041
ags+0.040
umpy+0.040
Neg logit contributions
 cardboard-0.036
lf-0.034
 shovel-0.032
 camps-0.031
 called-0.031
Token thought
Token ID1807
Feature activation-8.34e-03

Pos logit contributions
 cardboard+0.078
lf+0.075
ts+0.074
 camps+0.069
 fossils+0.064
Neg logit contributions
eal-0.120
arre-0.113
irk-0.105
umpy-0.104
 back-0.102
Token for
Token ID329
Feature activation-6.37e-02

Pos logit contributions
eal+0.013
arre+0.012
irk+0.011
umpy+0.011
 back+0.011
Neg logit contributions
lf-0.014
 cardboard-0.014
 camps-0.013
ts-0.013
 blasted-0.013
Token a
Token ID257
Feature activation+1.57e-01

Pos logit contributions
 back+0.102
 aback+0.097
 firm+0.096
eal+0.095
agree+0.093
Neg logit contributions
 camps-0.097
lf-0.092
 blasted-0.088
 cardboard-0.084
 trek-0.082
Token moment
Token ID2589
Feature activation-6.65e-02

Pos logit contributions
lf+0.329
 cardboard+0.327
ts+0.318
 camps+0.303
 blasted+0.296
Neg logit contributions
eal-0.193
arre-0.184
irk-0.166
umpy-0.164
agree-0.149
Token and
Token ID290
Feature activation-1.40e-02

Pos logit contributions
eal+0.142
arre+0.138
umpy+0.129
irk+0.129
ags+0.124
Neg logit contributions
 cardboard-0.084
lf-0.083
ts-0.074
 camps-0.070
 trek-0.068
Token smiled
Token ID13541
Feature activation-2.26e-02

Pos logit contributions
eal+0.029
arre+0.028
 back+0.026
irk+0.026
 disappointed+0.026
Neg logit contributions
 cardboard-0.018
lf-0.017
ts-0.017
 camps-0.017
 trek-0.015
Token.
Token ID13
Feature activation-2.35e-02

Pos logit contributions
arre+0.056
eal+0.055
umpy+0.052
awk+0.051
Fighting+0.051
Neg logit contributions
 cardboard-0.020
 pretend-0.017
 pretended-0.017
lf-0.016
 camps-0.016
Token
Token ID220
Feature activation-9.69e-02

Pos logit contributions
eal+0.043
arre+0.042
umpy+0.039
irk+0.039
awk+0.038
Neg logit contributions
 cardboard-0.036
lf-0.034
 camps-0.031
 shovel-0.031
 trek-0.030
Token
Token ID198
Feature activation-1.23e-02

Pos logit contributions
eal+0.207
arre+0.190
irk+0.186
vious+0.182
umpy+0.181
Neg logit contributions
 cardboard-0.121
lf-0.113
ts-0.105
 camps-0.100
 trek-0.099
Token,
Token ID11
Feature activation+5.05e-02

Pos logit contributions
eal+0.219
arre+0.209
umpy+0.197
irk+0.194
ealing+0.192
Neg logit contributions
 cardboard-0.100
lf-0.100
 pretend-0.085
ts-0.080
 pretended-0.079
Token but
Token ID475
Feature activation-6.60e-03

Pos logit contributions
 cardboard+0.043
lf+0.037
 pretend+0.037
 pretended+0.036
 called+0.036
Neg logit contributions
eal-0.131
arre-0.129
umpy-0.122
ealing-0.118
arling-0.117
Token she
Token ID673
Feature activation-5.49e-03

Pos logit contributions
eal+0.014
arre+0.014
umpy+0.013
irk+0.013
awk+0.013
Neg logit contributions
lf-0.009
 cardboard-0.008
ts-0.008
 pretend-0.007
 pretended-0.007
Token made
Token ID925
Feature activation-2.19e-02

Pos logit contributions
eal+0.012
arre+0.011
umpy+0.011
irk+0.011
vious+0.010
Neg logit contributions
 cardboard-0.007
lf-0.007
ts-0.006
 camps-0.006
 blasted-0.006
Token her
Token ID607
Feature activation-7.25e-02

Pos logit contributions
eal+0.053
arre+0.052
umpy+0.051
irk+0.050
awk+0.047
Neg logit contributions
 cardboard-0.022
lf-0.020
ts-0.019
 pretend-0.018
 camps-0.016
Token unhappy
Token ID19283
Feature activation-1.30e-01

Pos logit contributions
eal+0.154
arre+0.151
umpy+0.142
arling+0.138
irk+0.137
Neg logit contributions
 cardboard-0.098
lf-0.091
 pretend-0.089
 pretended-0.079
 camps-0.078
Token.
Token ID13
Feature activation-3.28e-02

Pos logit contributions
eal+0.337
arre+0.322
umpy+0.301
ealing+0.299
awk+0.299
Neg logit contributions
lf-0.122
 cardboard-0.107
 pretend-0.097
 called-0.087
ts-0.086
Token She
Token ID1375
Feature activation+5.11e-02

Pos logit contributions
eal+0.054
arre+0.052
umpy+0.046
irk+0.046
awk+0.045
Neg logit contributions
 cardboard-0.057
lf-0.055
 pretend-0.051
 camps-0.050
ts-0.049
Token said
Token ID531
Feature activation-2.82e-02

Pos logit contributions
 cardboard+0.083
lf+0.079
 camps+0.074
ts+0.072
 fossils+0.069
Neg logit contributions
eal-0.093
arre-0.085
irk-0.080
 disappointed-0.079
vious-0.079
Token she
Token ID673
Feature activation-5.53e-03

Pos logit contributions
eal+0.068
arre+0.066
umpy+0.064
irk+0.062
ags+0.061
Neg logit contributions
 cardboard-0.032
lf-0.032
ts-0.027
 pretend-0.025
 jewels-0.023
Token was
Token ID373
Feature activation+2.56e-02

Pos logit contributions
eal+0.012
arre+0.012
umpy+0.011
irk+0.011
vious+0.011
Neg logit contributions
 cardboard-0.007
lf-0.007
ts-0.006
 camps-0.006
 trek-0.005
Token a
Token ID257
Feature activation+1.53e-01

Pos logit contributions
eal+0.024
 back+0.023
arre+0.022
irk+0.022
umpy+0.022
Neg logit contributions
lf-0.022
 cardboard-0.020
ts-0.019
 camps-0.019
 blasted-0.018
Token carrot
Token ID40562
Feature activation+4.95e-02

Pos logit contributions
lf+0.227
ts+0.218
 camps+0.201
 cardboard+0.200
 trek+0.194
Neg logit contributions
eal-0.265
 back-0.255
irk-0.254
umpy-0.247
arre-0.246
Token.
Token ID13
Feature activation-2.25e-02

Pos logit contributions
lf+0.078
 cardboard+0.076
ts+0.074
 trek+0.069
 camps+0.068
Neg logit contributions
eal-0.084
arre-0.081
 back-0.077
umpy-0.077
irk-0.076
Token He
Token ID679
Feature activation+8.05e-02

Pos logit contributions
eal+0.033
arre+0.031
irk+0.029
umpy+0.029
ags+0.027
Neg logit contributions
lf-0.042
 cardboard-0.042
ts-0.040
 camps-0.038
 blasted-0.038
Token gives
Token ID3607
Feature activation-4.19e-02

Pos logit contributions
lf+0.159
 cardboard+0.147
ts+0.138
 hours+0.136
 trek+0.133
Neg logit contributions
eal-0.110
arre-0.102
vious-0.094
 disappointed-0.093
irk-0.090
Token it
Token ID340
Feature activation-1.27e-01

Pos logit contributions
eal+0.072
arre+0.066
irk+0.063
 back+0.062
ags+0.061
Neg logit contributions
lf-0.069
 cardboard-0.067
ts-0.061
 camps-0.061
 blasted-0.060
Token to
Token ID284
Feature activation-1.15e-02

Pos logit contributions
eal+0.222
arre+0.216
umpy+0.200
irk+0.200
ags+0.188
Neg logit contributions
 cardboard-0.208
lf-0.201
ts-0.189
 camps-0.182
 blasted-0.177
Token L
Token ID406
Feature activation-1.84e-02

Pos logit contributions
eal+0.022
arre+0.021
irk+0.020
umpy+0.019
ags+0.019
Neg logit contributions
 cardboard-0.017
lf-0.017
ts-0.016
 camps-0.015
 blasted-0.015
Tokenila
Token ID10102
Feature activation-1.19e-02

Pos logit contributions
eal+0.029
arre+0.028
irk+0.026
umpy+0.026
agree+0.025
Neg logit contributions
lf-0.032
 cardboard-0.032
ts-0.030
 camps-0.028
 trek-0.028
Token.
Token ID13
Feature activation-2.62e-02

Pos logit contributions
eal+0.026
arre+0.026
irk+0.024
umpy+0.024
agree+0.023
Neg logit contributions
 cardboard-0.014
lf-0.013
ts-0.012
 camps-0.011
 pretend-0.011
Token He
Token ID679
Feature activation+8.00e-02

Pos logit contributions
eal+0.044
arre+0.043
umpy+0.039
irk+0.039
ags+0.037
Neg logit contributions
 cardboard-0.045
lf-0.044
ts-0.041
 camps-0.039
 blasted-0.038
Token hurt
Token ID5938
Feature activation-6.87e-02

Pos logit contributions
eal+0.055
arre+0.053
irk+0.050
umpy+0.050
ags+0.047
Neg logit contributions
 cardboard-0.041
lf-0.039
ts-0.037
 pretend-0.034
 camps-0.034
Token your
Token ID534
Feature activation-9.84e-02

Pos logit contributions
eal+0.130
arre+0.125
irk+0.115
umpy+0.113
ags+0.108
Neg logit contributions
 cardboard-0.106
lf-0.103
ts-0.098
 pretend-0.088
 camps-0.088
Token lip
Token ID10645
Feature activation-5.69e-02

Pos logit contributions
eal+0.149
arre+0.146
umpy+0.132
irk+0.131
vious+0.123
Neg logit contributions
 cardboard-0.187
lf-0.182
ts-0.172
 camps-0.164
 pretend-0.163
Token.
Token ID13
Feature activation-3.68e-02

Pos logit contributions
eal+0.124
arre+0.117
irk+0.114
umpy+0.108
ags+0.107
Neg logit contributions
 cardboard-0.072
lf-0.071
ts-0.065
 pretend-0.059
 camps-0.058
Token Or
Token ID1471
Feature activation+1.77e-02

Pos logit contributions
eal+0.053
arre+0.051
irk+0.046
umpy+0.046
ags+0.042
Neg logit contributions
 cardboard-0.071
lf-0.069
ts-0.065
 camps-0.063
 blasted-0.062
Token it
Token ID340
Feature activation-1.24e-01

Pos logit contributions
lf+0.025
 cardboard+0.025
 camps+0.023
ts+0.022
 trek+0.022
Neg logit contributions
eal-0.033
arre-0.032
irk-0.030
umpy-0.030
vious-0.029
Token might
Token ID1244
Feature activation-3.61e-03

Pos logit contributions
eal+0.252
irk+0.239
arre+0.233
umpy+0.219
ags+0.219
Neg logit contributions
 cardboard-0.178
ts-0.160
lf-0.159
 pretend-0.154
 blasted-0.148
Token stick
Token ID4859
Feature activation+3.04e-02

Pos logit contributions
eal+0.007
arre+0.007
irk+0.006
umpy+0.006
ags+0.006
Neg logit contributions
 cardboard-0.006
lf-0.005
ts-0.005
 camps-0.005
 blasted-0.005
Token to
Token ID284
Feature activation-7.17e-03

Pos logit contributions
 cardboard+0.046
lf+0.041
ts+0.040
 pretend+0.038
 camps+0.036
Neg logit contributions
eal-0.061
arre-0.057
irk-0.056
umpy-0.053
Fighting-0.050
Token something
Token ID1223
Feature activation+1.64e-02

Pos logit contributions
eal+0.014
arre+0.013
umpy+0.013
irk+0.013
vious+0.012
Neg logit contributions
 cardboard-0.011
lf-0.010
ts-0.010
 camps-0.009
 pretend-0.009
Token else
Token ID2073
Feature activation-3.85e-03

Pos logit contributions
 cardboard+0.020
lf+0.019
ts+0.018
 camps+0.017
 blasted+0.016
Neg logit contributions
eal-0.036
arre-0.035
irk-0.033
umpy-0.033
ags-0.031
Token bitter
Token ID12922
Feature activation-9.13e-02

Pos logit contributions
eal+0.150
arre+0.140
irk+0.138
umpy+0.132
vious+0.124
Neg logit contributions
 cardboard-0.124
ts-0.115
lf-0.115
 camps-0.107
 pretend-0.106
Token.
Token ID13
Feature activation-4.10e-02

Pos logit contributions
eal+0.202
arre+0.193
irk+0.185
umpy+0.179
Fighting+0.177
Neg logit contributions
 cardboard-0.106
lf-0.099
ts-0.093
 camps-0.088
 pretend-0.087
Token It
Token ID632
Feature activation-1.35e-02

Pos logit contributions
eal+0.064
arre+0.061
irk+0.056
umpy+0.056
ags+0.052
Neg logit contributions
 cardboard-0.074
lf-0.072
ts-0.067
 camps-0.066
 blasted-0.064
Token's
Token ID338
Feature activation-7.82e-02

Pos logit contributions
eal+0.030
irk+0.029
ags+0.027
arre+0.027
umpy+0.026
Neg logit contributions
 cardboard-0.017
 pretend-0.016
 matches-0.015
ts-0.015
 pretended-0.015
Token worth
Token ID2861
Feature activation-5.44e-02

Pos logit contributions
eal+0.150
arre+0.140
irk+0.137
umpy+0.132
vious+0.124
Neg logit contributions
 cardboard-0.124
lf-0.114
ts-0.113
 camps-0.106
 pretend-0.105
Token it
Token ID340
Feature activation-1.24e-01

Pos logit contributions
eal+0.105
arre+0.102
umpy+0.097
irk+0.096
ags+0.093
Neg logit contributions
 cardboard-0.075
lf-0.075
 camps-0.067
ts-0.065
 blasted-0.064
Token,"
Token ID553
Feature activation+2.84e-02

Pos logit contributions
eal+0.267
arre+0.257
irk+0.248
umpy+0.234
Fighting+0.229
Neg logit contributions
 cardboard-0.159
lf-0.152
ts-0.143
 camps-0.131
 pretend-0.130
Token the
Token ID262
Feature activation-4.74e-02

Pos logit contributions
 cardboard+0.031
lf+0.026
ts+0.025
 pretend+0.023
 called+0.022
Neg logit contributions
eal-0.071
arre-0.070
umpy-0.068
irk-0.062
ags-0.062
Token doctor
Token ID6253
Feature activation-1.90e-02

Pos logit contributions
eal+0.073
arre+0.072
irk+0.064
umpy+0.063
agree+0.061
Neg logit contributions
 cardboard-0.089
lf-0.088
ts-0.082
 camps-0.079
 pretend-0.078
Token said
Token ID531
Feature activation-3.04e-02

Pos logit contributions
eal+0.040
arre+0.040
umpy+0.038
irk+0.037
ags+0.037
Neg logit contributions
lf-0.023
 cardboard-0.023
ts-0.021
 pretended-0.020
 blasted-0.019
Token.
Token ID13
Feature activation-3.72e-02

Pos logit contributions
arre+0.076
eal+0.074
umpy+0.069
Fighting+0.068
agree+0.067
Neg logit contributions
 cardboard-0.027
lf-0.026
ts-0.025
 pretend-0.020
 called-0.020
Token and
Token ID290
Feature activation-1.75e-02

Pos logit contributions
ts+0.077
lf+0.077
 cardboard+0.077
 trek+0.074
 blasted+0.072
Neg logit contributions
eal-0.092
arre-0.090
 back-0.089
irk-0.087
umpy-0.086
Token she
Token ID673
Feature activation+6.07e-04

Pos logit contributions
eal+0.027
 disappointed+0.025
 mad+0.024
arre+0.024
vious+0.023
Neg logit contributions
 camps-0.029
 cardboard-0.029
lf-0.028
ts-0.027
 blasted-0.027
Token wanted
Token ID2227
Feature activation+4.37e-02

Pos logit contributions
 cardboard+0.001
lf+0.001
 camps+0.001
ts+0.001
 blasted+0.001
Neg logit contributions
eal-0.001
arre-0.001
umpy-0.001
irk-0.001
 disappointed-0.001
Token to
Token ID284
Feature activation-8.11e-04

Pos logit contributions
 cardboard+0.071
lf+0.069
ts+0.065
 camps+0.062
 blasted+0.060
Neg logit contributions
eal-0.078
arre-0.075
umpy-0.070
irk-0.070
ags-0.066
Token try
Token ID1949
Feature activation-2.90e-02

Pos logit contributions
eal+0.002
arre+0.001
umpy+0.001
irk+0.001
vious+0.001
Neg logit contributions
 cardboard-0.001
lf-0.001
ts-0.001
 camps-0.001
 blasted-0.001
Token it
Token ID340
Feature activation-1.22e-01

Pos logit contributions
eal+0.070
arre+0.067
umpy+0.065
irk+0.065
vious+0.062
Neg logit contributions
 cardboard-0.030
lf-0.028
ts-0.026
 pretend-0.023
 camps-0.023
Token.
Token ID13
Feature activation-2.18e-02

Pos logit contributions
eal+0.255
arre+0.248
irk+0.234
umpy+0.232
vious+0.221
Neg logit contributions
 cardboard-0.159
lf-0.152
ts-0.140
 camps-0.132
 pretend-0.128
Token
Token ID198
Feature activation-7.91e-03

Pos logit contributions
eal+0.035
arre+0.034
irk+0.032
umpy+0.032
ags+0.031
Neg logit contributions
 cardboard-0.038
lf-0.036
 pretend-0.034
 trek-0.033
ts-0.033
Token
Token ID198
Feature activation-7.95e-03

Pos logit contributions
eal+0.013
arre+0.012
umpy+0.012
irk+0.011
ags+0.011
Neg logit contributions
 cardboard-0.014
lf-0.014
ts-0.013
 camps-0.012
 blasted-0.012
TokenShe
Token ID3347
Feature activation+7.99e-02

Pos logit contributions
eal+0.013
arre+0.012
umpy+0.011
irk+0.011
vious+0.010
Neg logit contributions
 cardboard-0.014
lf-0.014
ts-0.013
 camps-0.013
 blasted-0.012
Token reached
Token ID4251
Feature activation+5.91e-02

Pos logit contributions
 cardboard+0.114
lf+0.104
ts+0.101
 camps+0.100
 fossils+0.092
Neg logit contributions
eal-0.160
arre-0.152
umpy-0.135
irk-0.135
 disappointed-0.134