TOP ACTIVATIONS
MAX = 4.791

no ! We destroyed the apple !" Lily cried .
go back to the pool side , Mia . Maybe we
going to eat you for dinner !" Tim my was so
want to go in the cave . It 's too ugly
not pull them off the stems , and we could not
want to play in the hut anymore ," Lily said .
not to fall into the pond . It 's deep and
We want to eat the banana , but we can 't
won 't swim in the pond again . I will be
want to play in the barrel anymore . It is too
't want to eat the pit . I only want to
you to clean up your room ! Now , I have
's not play in the forest again ," L ila said
Let 's go down the pit ," Ben said . "
sit in the old arm chair . It is wet and
children ! You ruined the cake ! You have to say
should not fight over the banana . You should share and
No , we cannot go outside ," Mia says .
We should not bother the frogs . They live here and
's never go into a cave again !" L ila said

MIN ACTIVATIONS
MIN = -5.067

it would give her a special power . One
parents gave him a pe b ble . He had never
He gave him a beautiful golden sword . The
and always wore a spark ly dress . One day ,
Pixie gave her a spark ly wand , which she could
bedroom and put on the pretty buckle .
green dress with a spark ly belt . She couldn 't
she liked to wear her special dress . It was purple
unique girl with a spark ly smile and bright eyes that
yellow dress with a spark ly cape . One
feeling very proud wearing her special suit . She will never
to her and put a comforting hand on her shoulder .
had just given him a shiny new pistol . Joey had
she put on her spark ly shoes . She
mom my gave her a cool mask to wear . It
she found a shiny pe b ble and picked it up
. He put on a blue shirt , blue pants ,
She picked up a shiny golden sword and went out into
fairy gave Lily a spark ly wand and said to make
her mum put on her special clothes and said to her

INTERVAL 2.326 - 4.791
CONTAINS 0.311%

At first , none of the other children wanted anything to
want to build a wide tower , like a castle !"
. Now we can enjoy cocoa and stay healthy too !"
very grateful . Mrs . Johnson asked Lily
" Yes , it is safe to eat ," she said

INTERVAL -0.138 - 2.326
CONTAINS 56.956%

was no longer mad at her ! <|endoftext|> Once upon a
with their work . The palace was neat and tidy once
Tom and Mia are shocked . They drop the bow and
Suddenly , the paper boat started to fill with
, sweet ie . Painting is fun , but it can

INTERVAL -2.603 - -0.138
CONTAINS 42.394%

saw that it was a whale . The whale
saw that it was the perfect size . Sm iling ,
gathered around a funny - looking machine . John
path . Suddenly , the path said hello back ! They
. She was having so much fun that she started to

INTERVAL -5.067 - -2.603
CONTAINS 0.338%

showed her a beautiful dress with colorful flowers on it .
neck . T illy spent every day playing
light , like a spark ly star . John
, Tim my found a shiny rock on the ground .
s new skill . <|endoftext|> One day , a 3
Token no
Token ID645
Feature activation+1.59e+00

Pos logit contributions
 mess+0.041
 wasteful+0.038
 overflow+0.038
 roughly+0.036
 neither+0.036
Neg logit contributions
 name-0.046
ka-0.044
ance-0.043
dot-0.042
 ti-0.041
Token!
Token ID0
Feature activation+9.65e-01

Pos logit contributions
 neither+0.078
 overflow+0.078
 wasteful+0.078
 unfortunately+0.070
 ruining+0.070
Neg logit contributions
ka-0.042
 name-0.041
 powers-0.037
dot-0.035
 aboard-0.034
Token We
Token ID775
Feature activation+1.32e+00

Pos logit contributions
 mess+0.046
 neither+0.043
 wasteful+0.042
 overflow+0.041
 ruining+0.040
Neg logit contributions
 name-0.032
ka-0.029
ance-0.029
dot-0.028
anced-0.027
Token destroyed
Token ID6572
Feature activation+2.90e+00

Pos logit contributions
 neither+0.058
 mess+0.057
 structures+0.056
 wasteful+0.055
 overflow+0.054
Neg logit contributions
 name-0.043
tie-0.040
ka-0.039
ocket-0.036
aved-0.036
Token the
Token ID262
Feature activation+2.13e+00

Pos logit contributions
 overflow+0.137
 forgot+0.129
 neither+0.128
 wasteful+0.126
ivery+0.122
Neg logit contributions
 aboard-0.088
ka-0.081
 name-0.081
ance-0.070
 Magic-0.070
Token apple
Token ID17180
Feature activation+4.79e+00

Pos logit contributions
 neither+0.100
 overflow+0.091
 wasteful+0.087
 ruining+0.082
 roughly+0.082
Neg logit contributions
 name-0.075
orn-0.069
tie-0.066
 amulet-0.062
 Crown-0.061
Token!"
Token ID2474
Feature activation+8.88e-01

Pos logit contributions
 wasteful+0.259
 overflow+0.239
 sudden+0.237
 neither+0.237
Sadly+0.234
Neg logit contributions
orn-0.091
 aboard-0.081
tie-0.081
ed-0.067
--0.066
Token Lily
Token ID20037
Feature activation+2.89e+00

Pos logit contributions
 mess+0.030
 neither+0.030
 wasteful+0.029
 overflow+0.026
 ruining+0.026
Neg logit contributions
 name-0.041
ka-0.038
dot-0.037
leaf-0.037
tie-0.036
Token cried
Token ID16896
Feature activation+2.57e-01

Pos logit contributions
 overflow+0.163
 structures+0.161
 flood+0.159
 wasteful+0.158
 earthquake+0.154
Neg logit contributions
ed-0.056
 name-0.055
 hugs-0.049
u-0.047
tie-0.046
Token.
Token ID13
Feature activation+5.92e-01

Pos logit contributions
 neither+0.011
 mess+0.011
 wasteful+0.011
 ruining+0.010
 overflow+0.010
Neg logit contributions
ka-0.009
 name-0.009
tie-0.009
ance-0.009
dot-0.008
Token
Token ID198
Feature activation+2.95e-01

Pos logit contributions
 mess+0.028
 neither+0.027
 wasteful+0.027
 overflow+0.025
 ruining+0.025
Neg logit contributions
 name-0.020
tie-0.019
ka-0.018
dot-0.018
ance-0.018
Token go
Token ID467
Feature activation-1.03e+00

Pos logit contributions
 neither+0.062
 wasteful+0.061
 mess+0.058
 roughly+0.054
 mosquitoes+0.054
Neg logit contributions
 name-0.040
tie-0.031
ance-0.030
 aboard-0.030
ka-0.030
Token back
Token ID736
Feature activation+1.04e+00

Pos logit contributions
ka+0.041
tie+0.040
ance+0.040
anced+0.039
dot+0.039
Neg logit contributions
 mess-0.041
 neither-0.040
 overflow-0.035
 fighting-0.034
 wasteful-0.034
Token to
Token ID284
Feature activation+1.30e+00

Pos logit contributions
 neither+0.052
 wasteful+0.051
 mess+0.049
 overflow+0.046
 ruining+0.045
Neg logit contributions
 aboard-0.034
 name-0.032
ka-0.031
ance-0.030
tie-0.029
Token the
Token ID262
Feature activation+1.08e+00

Pos logit contributions
 neither+0.060
 wasteful+0.057
 roughly+0.053
 mess+0.053
 overflow+0.050
Neg logit contributions
 name-0.045
 aboard-0.044
ance-0.040
 Magic-0.039
 outer-0.039
Token pool
Token ID5933
Feature activation+3.26e+00

Pos logit contributions
 neither+0.046
 wasteful+0.040
 mess+0.038
 roughly+0.037
 ruining+0.036
Neg logit contributions
 name-0.044
 aboard-0.040
ka-0.040
 Crown-0.039
tie-0.039
Tokenside
Token ID1589
Feature activation+4.32e+00

Pos logit contributions
 wasteful+0.155
 neither+0.149
 ruining+0.149
 mosquitoes+0.145
 spitting+0.139
Neg logit contributions
 aboard-0.095
 name-0.089
tie-0.085
ka-0.082
 suit-0.079
Token,
Token ID11
Feature activation+2.21e+00

Pos logit contributions
 wasteful+0.237
 neither+0.219
 ruining+0.210
 mosquitoes+0.204
 roughly+0.195
Neg logit contributions
 name-0.130
 aboard-0.116
 powers-0.099
 outfit-0.098
 costume-0.096
Token Mia
Token ID32189
Feature activation+1.25e+00

Pos logit contributions
 wasteful+0.078
 neither+0.073
 roughly+0.071
oving+0.070
 mess+0.068
Neg logit contributions
 name-0.104
 aboard-0.086
 hello-0.081
ance-0.078
 hi-0.078
Token.
Token ID13
Feature activation+3.78e-01

Pos logit contributions
 neither+0.053
 wasteful+0.050
 mess+0.048
 overflow+0.047
 ruining+0.047
Neg logit contributions
 name-0.044
 aboard-0.043
ka-0.043
ance-0.042
tie-0.042
Token Maybe
Token ID6674
Feature activation+1.25e+00

Pos logit contributions
 mess+0.015
 neither+0.015
 wasteful+0.014
 overflow+0.013
 ruining+0.013
Neg logit contributions
 name-0.015
tie-0.015
ka-0.015
ance-0.015
dot-0.014
Token we
Token ID356
Feature activation+2.58e-01

Pos logit contributions
 mess+0.040
 wasteful+0.039
 neither+0.037
 overflow+0.035
 roughly+0.035
Neg logit contributions
 name-0.058
 aboard-0.058
ance-0.055
tie-0.052
dot-0.052
Token going
Token ID1016
Feature activation-6.96e-01

Pos logit contributions
 neither+0.036
 mess+0.036
 wasteful+0.033
 structures+0.032
 overflow+0.032
Neg logit contributions
 name-0.038
dot-0.036
leaf-0.036
 aboard-0.035
tie-0.034
Token to
Token ID284
Feature activation+1.34e+00

Pos logit contributions
ka+0.026
ance+0.025
tie+0.025
leaf+0.025
anced+0.025
Neg logit contributions
 mess-0.030
 neither-0.030
 wasteful-0.027
 fighting-0.026
 ruining-0.026
Token eat
Token ID4483
Feature activation+1.47e+00

Pos logit contributions
 neither+0.063
 wasteful+0.059
 mess+0.055
 roughly+0.054
 structures+0.051
Neg logit contributions
 name-0.049
 aboard-0.042
 Magic-0.041
leaf-0.040
 inspect-0.040
Token you
Token ID345
Feature activation+2.02e+00

Pos logit contributions
 neither+0.063
 mess+0.058
 wasteful+0.057
 overflow+0.055
 forgot+0.052
Neg logit contributions
 name-0.053
 aboard-0.052
tie-0.051
shirt-0.051
ance-0.051
Token for
Token ID329
Feature activation+1.15e+00

Pos logit contributions
 neither+0.091
 wasteful+0.087
 mess+0.081
 overflow+0.079
Uh+0.077
Neg logit contributions
 name-0.069
 aboard-0.068
tie-0.067
ka-0.066
ance-0.066
Token dinner
Token ID8073
Feature activation+4.25e+00

Pos logit contributions
 neither+0.055
 wasteful+0.051
 mess+0.051
 overflow+0.049
 roughly+0.046
Neg logit contributions
 powers-0.035
dot-0.035
ance-0.035
 name-0.035
 aboard-0.034
Token!"
Token ID2474
Feature activation+1.15e+00

Pos logit contributions
 wasteful+0.211
Uh+0.199
 overflow+0.198
 neither+0.198
 structures+0.190
Neg logit contributions
tie-0.109
 ti-0.104
 aboard-0.101
ka-0.099
 name-0.098
Token Tim
Token ID5045
Feature activation+6.80e-01

Pos logit contributions
 mess+0.041
 neither+0.040
 wasteful+0.040
 structures+0.036
 ruining+0.035
Neg logit contributions
 name-0.051
tie-0.048
shirt-0.047
Aid-0.046
dot-0.046
Tokenmy
Token ID1820
Feature activation+1.40e+00

Pos logit contributions
 neither+0.034
 wasteful+0.032
 mess+0.032
 overflow+0.030
 ruining+0.029
Neg logit contributions
ka-0.021
 name-0.021
tie-0.020
shirt-0.019
ance-0.019
Token was
Token ID373
Feature activation+1.32e+00

Pos logit contributions
 neither+0.067
 wasteful+0.067
 mess+0.064
 overflow+0.061
 structures+0.061
Neg logit contributions
 name-0.044
tie-0.041
ka-0.041
shirt-0.040
dot-0.039
Token so
Token ID523
Feature activation+1.30e+00

Pos logit contributions
 mess+0.049
 neither+0.048
 wasteful+0.047
 overflow+0.046
 structures+0.046
Neg logit contributions
 name-0.055
dot-0.047
tie-0.047
 aboard-0.045
leaf-0.045
Token want
Token ID765
Feature activation+2.12e-01

Pos logit contributions
 neither+0.003
 mess+0.002
 wasteful+0.002
 overflow+0.002
 fighting+0.002
Neg logit contributions
ka-0.002
tie-0.002
 name-0.002
ance-0.002
dot-0.002
Token to
Token ID284
Feature activation+8.96e-01

Pos logit contributions
 neither+0.010
 mess+0.010
 wasteful+0.010
 overflow+0.009
 roughly+0.009
Neg logit contributions
ka-0.006
 name-0.006
tie-0.006
ance-0.006
dot-0.006
Token go
Token ID467
Feature activation-6.94e-01

Pos logit contributions
 wasteful+0.044
 neither+0.043
 mess+0.042
 mosquitoes+0.038
 roughly+0.037
Neg logit contributions
 name-0.029
ance-0.025
tie-0.025
 aboard-0.025
dot-0.025
Token in
Token ID287
Feature activation+9.92e-01

Pos logit contributions
ka+0.025
leaf+0.024
tie+0.024
ance+0.024
shirt+0.024
Neg logit contributions
 mess-0.031
 neither-0.031
 wasteful-0.028
 fighting-0.027
 ruining-0.027
Token the
Token ID262
Feature activation+3.80e-02

Pos logit contributions
 neither+0.043
 wasteful+0.043
 mess+0.040
 overflow+0.037
 roughly+0.036
Neg logit contributions
 aboard-0.035
 name-0.034
ka-0.033
ance-0.032
tie-0.032
Token cave
Token ID11527
Feature activation+4.18e+00

Pos logit contributions
 neither+0.002
 mess+0.001
 wasteful+0.001
 overflow+0.001
 fighting+0.001
Neg logit contributions
ka-0.001
tie-0.001
 name-0.001
ance-0.001
dot-0.001
Token.
Token ID13
Feature activation+3.07e-01

Pos logit contributions
 wasteful+0.194
 mosquitoes+0.188
 spitting+0.185
 neither+0.185
 ruining+0.182
Neg logit contributions
 aboard-0.116
tie-0.115
ka-0.109
 name-0.108
 ti-0.103
Token It
Token ID632
Feature activation+9.23e-01

Pos logit contributions
 mess+0.014
 neither+0.014
 wasteful+0.013
 overflow+0.012
 ruining+0.012
Neg logit contributions
 name-0.010
ka-0.010
tie-0.010
dot-0.010
ance-0.010
Token's
Token ID338
Feature activation+1.17e+00

Pos logit contributions
 neither+0.038
 wasteful+0.037
 mess+0.036
 roughly+0.034
 overflow+0.033
Neg logit contributions
 name-0.033
ka-0.033
 aboard-0.032
ance-0.032
tie-0.031
Token too
Token ID1165
Feature activation+1.10e+00

Pos logit contributions
 neither+0.043
 wasteful+0.038
 mess+0.037
 structures+0.036
 overflow+0.036
Neg logit contributions
 name-0.048
 aboard-0.045
tie-0.044
 powers-0.043
ka-0.043
Token ugly
Token ID13400
Feature activation+8.72e-01

Pos logit contributions
 neither+0.046
 wasteful+0.041
 mess+0.040
 ruining+0.040
 fighting+0.039
Neg logit contributions
ka-0.041
 name-0.041
 aboard-0.040
tie-0.039
dot-0.037
Token not
Token ID407
Feature activation+6.16e-01

Pos logit contributions
 wasteful+0.064
 structures+0.062
 neither+0.061
 rats+0.060
 mess+0.058
Neg logit contributions
 name-0.075
tie-0.059
 polish-0.056
ad-0.056
 Magic-0.055
Token pull
Token ID2834
Feature activation+1.03e+00

Pos logit contributions
 neither+0.028
 mess+0.026
 wasteful+0.026
 ruining+0.024
 rats+0.024
Neg logit contributions
 name-0.022
tie-0.021
ka-0.021
shirt-0.020
dot-0.020
Token them
Token ID606
Feature activation+2.56e+00

Pos logit contributions
 mess+0.042
 neither+0.042
 wasteful+0.040
 overflow+0.040
 roughly+0.037
Neg logit contributions
ka-0.038
tie-0.035
 name-0.035
 aboard-0.035
ance-0.034
Token off
Token ID572
Feature activation+2.72e+00

Pos logit contributions
 wasteful+0.134
 neither+0.128
 mosquitoes+0.126
Uh+0.125
 spitting+0.123
Neg logit contributions
 aboard-0.064
tie-0.058
 name-0.057
ance-0.054
ka-0.051
Token the
Token ID262
Feature activation+7.93e-01

Pos logit contributions
 neither+0.123
 wasteful+0.115
 mess+0.111
 overflow+0.108
 ruining+0.106
Neg logit contributions
 aboard-0.084
ka-0.082
tie-0.078
 name-0.078
 outer-0.076
Token stems
Token ID21552
Feature activation+4.14e+00

Pos logit contributions
 neither+0.032
 wasteful+0.029
 mess+0.027
 overflow+0.026
 fighting+0.026
Neg logit contributions
 name-0.031
tie-0.031
ka-0.031
shirt-0.031
dot-0.030
Token,
Token ID11
Feature activation+2.34e+00

Pos logit contributions
 wasteful+0.203
 neither+0.196
 ruining+0.190
 mosquitoes+0.186
 rats+0.184
Neg logit contributions
tie-0.109
 aboard-0.099
 name-0.097
ance-0.094
ka-0.089
Token and
Token ID290
Feature activation+2.26e+00

Pos logit contributions
 mess+0.077
 neither+0.075
 anth+0.074
 wasteful+0.073
 roughly+0.073
Neg logit contributions
 name-0.098
tie-0.083
 aboard-0.082
shirt-0.081
 rib-0.080
Token we
Token ID356
Feature activation+9.12e-01

Pos logit contributions
 neither+0.083
 mess+0.077
 overflow+0.075
 wasteful+0.073
 roughly+0.073
Neg logit contributions
 name-0.088
 aboard-0.079
 powers-0.079
tie-0.079
leaf-0.079
Token could
Token ID714
Feature activation+1.77e+00

Pos logit contributions
 neither+0.041
 mess+0.040
 wasteful+0.039
 structures+0.037
 mosquitoes+0.037
Neg logit contributions
 name-0.030
tie-0.029
ka-0.029
ance-0.027
leaf-0.026
Token not
Token ID407
Feature activation+7.28e-01

Pos logit contributions
 wasteful+0.073
 structures+0.069
 neither+0.068
 rats+0.066
 erupted+0.064
Neg logit contributions
 name-0.070
tie-0.055
 polish-0.051
 aboard-0.051
 inspect-0.050
Token want
Token ID765
Feature activation+4.66e-01

Pos logit contributions
 neither+0.019
 mess+0.018
 wasteful+0.017
 structures+0.016
 overflow+0.016
Neg logit contributions
 name-0.015
tie-0.014
ka-0.014
dot-0.014
leaf-0.014
Token to
Token ID284
Feature activation+1.09e+00

Pos logit contributions
 neither+0.022
 mess+0.021
 wasteful+0.021
 overflow+0.019
 roughly+0.019
Neg logit contributions
ka-0.015
 name-0.015
tie-0.015
ance-0.014
dot-0.014
Token play
Token ID711
Feature activation+1.10e+00

Pos logit contributions
 neither+0.048
 wasteful+0.048
 mess+0.046
 structures+0.043
 roughly+0.042
Neg logit contributions
 name-0.040
tie-0.035
ka-0.035
dot-0.034
ance-0.033
Token in
Token ID287
Feature activation+2.01e+00

Pos logit contributions
 neither+0.049
 wasteful+0.046
 mess+0.045
 overflow+0.042
 erupted+0.040
Neg logit contributions
 name-0.038
tie-0.037
ka-0.035
 aboard-0.035
 powers-0.035
Token the
Token ID262
Feature activation+1.23e+00

Pos logit contributions
 neither+0.099
 wasteful+0.098
 forgot+0.089
 overflow+0.088
 erupted+0.087
Neg logit contributions
 outer-0.055
 aboard-0.053
 powers-0.051
ance-0.048
 Magic-0.047
Token hut
Token ID40812
Feature activation+4.13e+00

Pos logit contributions
 neither+0.057
 wasteful+0.048
 fighting+0.046
 overflow+0.045
 mess+0.044
Neg logit contributions
 name-0.044
orn-0.041
 amulet-0.040
 cloak-0.040
tie-0.040
Token anymore
Token ID7471
Feature activation+2.90e+00

Pos logit contributions
 wasteful+0.196
 neither+0.195
 overflow+0.181
 mosquitoes+0.181
 ruining+0.181
Neg logit contributions
 aboard-0.106
tie-0.101
 name-0.099
 ti-0.092
 amulet-0.089
Token,"
Token ID553
Feature activation+1.17e+00

Pos logit contributions
 wasteful+0.150
 overflow+0.145
 neither+0.141
 ruining+0.132
 mess+0.130
Neg logit contributions
 name-0.076
 aboard-0.070
 powers-0.068
ance-0.067
 ti-0.065
Token Lily
Token ID20037
Feature activation+1.75e+00

Pos logit contributions
 mess+0.049
 neither+0.047
 overflow+0.047
 wasteful+0.046
 ruining+0.045
Neg logit contributions
 name-0.042
 Magic-0.036
dot-0.036
shirt-0.036
ka-0.036
Token said
Token ID531
Feature activation-2.80e-02

Pos logit contributions
 overflow+0.084
 wasteful+0.084
 structures+0.082
 ruining+0.081
 flood+0.080
Neg logit contributions
 name-0.051
 gown-0.045
shirt-0.044
ka-0.044
ed-0.044
Token.
Token ID13
Feature activation+4.06e-01

Pos logit contributions
ka+0.001
tie+0.001
 name+0.001
ance+0.001
dot+0.001
Neg logit contributions
 neither-0.001
 mess-0.001
 wasteful-0.001
 overflow-0.001
 fighting-0.001
Token not
Token ID407
Feature activation+1.22e+00

Pos logit contributions
 wasteful+0.065
 mess+0.064
 neither+0.063
 ruining+0.058
 overflow+0.057
Neg logit contributions
 name-0.054
tie-0.052
ka-0.050
ance-0.049
shirt-0.049
Token to
Token ID284
Feature activation+1.90e+00

Pos logit contributions
 neither+0.063
 mess+0.061
 wasteful+0.060
 roughly+0.056
 overflow+0.056
Neg logit contributions
 name-0.034
tie-0.031
 aboard-0.028
ka-0.028
ance-0.028
Token fall
Token ID2121
Feature activation+1.51e+00

Pos logit contributions
 neither+0.092
 wasteful+0.091
 mess+0.090
 structures+0.086
 unfortunately+0.085
Neg logit contributions
 name-0.059
tie-0.046
 Crown-0.043
dot-0.043
ka-0.042
Token into
Token ID656
Feature activation+1.26e+00

Pos logit contributions
 wasteful+0.062
 neither+0.061
 mess+0.061
 roughly+0.054
 ruining+0.054
Neg logit contributions
 aboard-0.055
ka-0.055
tie-0.054
 name-0.053
ance-0.052
Token the
Token ID262
Feature activation+1.41e-01

Pos logit contributions
 neither+0.051
 mess+0.049
 wasteful+0.048
 overflow+0.045
 roughly+0.045
Neg logit contributions
 name-0.044
ka-0.043
 aboard-0.043
ance-0.042
 powers-0.042
Token pond
Token ID16723
Feature activation+4.12e+00

Pos logit contributions
 neither+0.005
 mess+0.005
 wasteful+0.005
 roughly+0.004
 overflow+0.004
Neg logit contributions
ka-0.006
 name-0.006
tie-0.006
dot-0.006
ance-0.006
Token.
Token ID13
Feature activation+1.13e+00

Pos logit contributions
Uh+0.201
 wasteful+0.200
 neither+0.200
 ruining+0.193
 mess+0.189
Neg logit contributions
 name-0.092
tie-0.090
 aboard-0.088
ance-0.078
 gown-0.077
Token It
Token ID632
Feature activation+1.81e+00

Pos logit contributions
 mess+0.056
 neither+0.053
 wasteful+0.050
 overflow+0.048
 ruining+0.047
Neg logit contributions
 name-0.037
dot-0.033
ance-0.033
ka-0.033
tie-0.032
Token's
Token ID338
Feature activation+1.03e+00

Pos logit contributions
 mess+0.091
 wasteful+0.090
 neither+0.090
 roughly+0.088
 structures+0.086
Neg logit contributions
 name-0.046
 aboard-0.039
ka-0.038
ance-0.038
 shone-0.037
Token deep
Token ID2769
Feature activation+1.32e+00

Pos logit contributions
 neither+0.030
 mess+0.028
 structures+0.026
 wasteful+0.025
 overflow+0.025
Neg logit contributions
 name-0.052
tie-0.048
ka-0.047
 powers-0.046
dot-0.046
Token and
Token ID290
Feature activation+1.48e+00

Pos logit contributions
 neither+0.059
 wasteful+0.055
 mess+0.054
 overflow+0.050
 ruining+0.050
Neg logit contributions
ka-0.047
 name-0.044
 aboard-0.044
tie-0.044
 powers-0.042
TokenWe
Token ID1135
Feature activation-2.46e-01

Pos logit contributions
 mess+0.006
 neither+0.006
 wasteful+0.006
 overflow+0.005
 ruining+0.005
Neg logit contributions
ka-0.007
tie-0.007
 name-0.007
leaf-0.007
dot-0.007
Token want
Token ID765
Feature activation+1.34e+00

Pos logit contributions
ka+0.010
tie+0.010
ance+0.010
 name+0.010
dot+0.010
Neg logit contributions
 neither-0.010
 mess-0.010
 wasteful-0.009
 fighting-0.008
 overflow-0.008
Token to
Token ID284
Feature activation+8.70e-01

Pos logit contributions
 neither+0.072
 wasteful+0.070
 mess+0.069
 overflow+0.067
 roughly+0.066
Neg logit contributions
 name-0.033
leaf-0.031
 powers-0.030
 aboard-0.030
tie-0.029
Token eat
Token ID4483
Feature activation+2.07e+00

Pos logit contributions
 neither+0.042
 wasteful+0.041
 mess+0.040
 fighting+0.036
 ruining+0.036
Neg logit contributions
 name-0.030
tie-0.027
ka-0.026
dot-0.026
ance-0.025
Token the
Token ID262
Feature activation+6.29e-01

Pos logit contributions
 neither+0.094
 overflow+0.091
 wasteful+0.089
 forgot+0.088
 mess+0.087
Neg logit contributions
 name-0.066
 aboard-0.065
tie-0.059
ka-0.059
shirt-0.057
Token banana
Token ID25996
Feature activation+4.09e+00

Pos logit contributions
 neither+0.025
 wasteful+0.023
 mess+0.023
 overflow+0.021
 roughly+0.021
Neg logit contributions
 name-0.026
ka-0.025
tie-0.025
dot-0.024
shirt-0.024
Token,
Token ID11
Feature activation+1.84e+00

Pos logit contributions
 wasteful+0.190
 ruining+0.184
 overflow+0.182
 sudden+0.179
 neither+0.179
Neg logit contributions
tie-0.110
 name-0.110
 aboard-0.106
orn-0.096
 powers-0.095
Token but
Token ID475
Feature activation+1.47e+00

Pos logit contributions
 wasteful+0.066
 neither+0.066
 mess+0.066
 roughly+0.062
 overflow+0.060
Neg logit contributions
 name-0.080
tie-0.067
 aboard-0.063
uit-0.063
shirt-0.063
Token we
Token ID356
Feature activation+1.29e+00

Pos logit contributions
 mess+0.052
 overflow+0.049
 roughly+0.049
 neither+0.048
 wasteful+0.047
Neg logit contributions
 name-0.062
 aboard-0.055
 Crown-0.054
tie-0.053
 Magic-0.053
Token can
Token ID460
Feature activation+2.04e+00

Pos logit contributions
 neither+0.059
 mess+0.056
 wasteful+0.055
 structures+0.054
 mosquitoes+0.052
Neg logit contributions
 name-0.043
tie-0.039
ka-0.037
leaf-0.036
ocket-0.036
Token't
Token ID470
Feature activation+1.90e+00

Pos logit contributions
 wasteful+0.095
 mess+0.088
 neither+0.087
 rats+0.085
 mosquitoes+0.083
Neg logit contributions
 name-0.080
tie-0.065
ad-0.061
 Crown-0.058
dot-0.057
Token won
Token ID1839
Feature activation+7.83e-01

Pos logit contributions
ka+0.006
tie+0.006
ance+0.006
 name+0.006
dot+0.006
Neg logit contributions
 neither-0.006
 mess-0.006
 wasteful-0.006
 overflow-0.005
 fighting-0.005
Token't
Token ID470
Feature activation+1.33e+00

Pos logit contributions
 neither+0.028
 mess+0.028
 wasteful+0.028
 overflow+0.025
 rats+0.024
Neg logit contributions
 name-0.032
ka-0.032
tie-0.031
ance-0.031
 aboard-0.031
Token swim
Token ID9422
Feature activation+1.84e+00

Pos logit contributions
 neither+0.058
 wasteful+0.055
 rats+0.052
 mess+0.051
 structures+0.051
Neg logit contributions
 name-0.049
tie-0.046
ka-0.043
leaf-0.041
yo-0.041
Token in
Token ID287
Feature activation+1.94e+00

Pos logit contributions
 mess+0.070
 neither+0.070
 wasteful+0.070
 ruining+0.066
 mosquitoes+0.064
Neg logit contributions
tie-0.070
ka-0.070
 aboard-0.067
 name-0.067
ance-0.062
Token the
Token ID262
Feature activation-3.12e-03

Pos logit contributions
 neither+0.090
 wasteful+0.086
 overflow+0.079
keys+0.078
 forgot+0.078
Neg logit contributions
 outer-0.056
 aboard-0.056
 powers-0.055
ka-0.054
 name-0.053
Token pond
Token ID16723
Feature activation+4.06e+00

Pos logit contributions
ka+0.000
tie+0.000
 name+0.000
ance+0.000
dot+0.000
Neg logit contributions
 neither-0.000
 mess-0.000
 wasteful-0.000
 overflow-0.000
 fighting-0.000
Token again
Token ID757
Feature activation+2.79e+00

Pos logit contributions
 wasteful+0.200
 neither+0.195
 ruining+0.193
Uh+0.189
 overflow+0.186
Neg logit contributions
tie-0.090
 aboard-0.086
 name-0.081
 ti-0.079
ka-0.079
Token.
Token ID13
Feature activation-2.33e-01

Pos logit contributions
 neither+0.139
 wasteful+0.138
 overflow+0.135
 ruining+0.129
 mess+0.127
Neg logit contributions
 aboard-0.066
 name-0.064
tie-0.062
ka-0.061
 powers-0.058
Token I
Token ID314
Feature activation-6.11e-02

Pos logit contributions
ka+0.010
tie+0.010
ance+0.009
 name+0.009
dot+0.009
Neg logit contributions
 neither-0.009
 mess-0.008
 wasteful-0.008
 fighting-0.007
 overflow-0.007
Token will
Token ID481
Feature activation+1.52e-01

Pos logit contributions
ka+0.002
tie+0.002
 name+0.002
ance+0.002
dot+0.002
Neg logit contributions
 neither-0.003
 mess-0.003
 wasteful-0.003
 overflow-0.002
 fighting-0.002
Token be
Token ID307
Feature activation+6.86e-01

Pos logit contributions
 neither+0.006
 mess+0.006
 wasteful+0.006
 overflow+0.005
 roughly+0.005
Neg logit contributions
 name-0.006
ka-0.006
tie-0.006
ance-0.006
dot-0.006
Token want
Token ID765
Feature activation+9.01e-01

Pos logit contributions
 neither+0.029
 mess+0.028
 wasteful+0.027
 overflow+0.025
 ruining+0.025
Neg logit contributions
 name-0.021
tie-0.020
ka-0.019
dot-0.019
leaf-0.019
Token to
Token ID284
Feature activation+1.33e+00

Pos logit contributions
 mess+0.047
 neither+0.047
 wasteful+0.046
 overflow+0.044
 roughly+0.043
Neg logit contributions
tie-0.022
 name-0.022
ka-0.022
 aboard-0.021
leaf-0.021
Token play
Token ID711
Feature activation+1.58e+00

Pos logit contributions
 wasteful+0.065
 neither+0.064
 mess+0.061
 rats+0.058
 mosquitoes+0.058
Neg logit contributions
 name-0.044
ka-0.036
tie-0.036
 aboard-0.034
dot-0.034
Token in
Token ID287
Feature activation+2.11e+00

Pos logit contributions
 neither+0.074
 wasteful+0.070
 mess+0.067
 overflow+0.066
 forgot+0.065
Neg logit contributions
 name-0.051
tie-0.047
 aboard-0.047
ka-0.045
 ti-0.044
Token the
Token ID262
Feature activation+2.50e-01

Pos logit contributions
 neither+0.106
 wasteful+0.106
 forgot+0.099
 overflow+0.096
 mess+0.094
Neg logit contributions
 outer-0.055
 aboard-0.055
 powers-0.047
 Magic-0.047
 name-0.046
Token barrel
Token ID9036
Feature activation+4.02e+00

Pos logit contributions
 neither+0.011
 mess+0.010
 wasteful+0.010
 overflow+0.009
 ruining+0.009
Neg logit contributions
 name-0.009
ka-0.009
tie-0.009
dot-0.009
 aboard-0.009
Token anymore
Token ID7471
Feature activation+2.86e+00

Pos logit contributions
 wasteful+0.227
 sudden+0.215
Uh+0.210
 neither+0.209
 mosquitoes+0.209
Neg logit contributions
tie-0.074
 aboard-0.072
ane-0.060
 name-0.059
ance-0.058
Token.
Token ID13
Feature activation+3.12e-01

Pos logit contributions
 wasteful+0.158
 overflow+0.155
 neither+0.147
 mess+0.146
 ruining+0.141
Neg logit contributions
 name-0.064
 aboard-0.060
 powers-0.055
ance-0.055
tie-0.053
Token It
Token ID632
Feature activation+1.61e+00

Pos logit contributions
 mess+0.014
 neither+0.013
 wasteful+0.012
 overflow+0.012
 ruining+0.011
Neg logit contributions
 name-0.011
ka-0.011
tie-0.011
dot-0.011
ance-0.011
Token is
Token ID318
Feature activation+1.30e+00

Pos logit contributions
 neither+0.091
 wasteful+0.088
 roughly+0.085
 overflow+0.084
 mess+0.084
Neg logit contributions
 name-0.033
 aboard-0.029
ka-0.028
ance-0.026
 shone-0.025
Token too
Token ID1165
Feature activation+1.24e+00

Pos logit contributions
 neither+0.046
 mess+0.040
 wasteful+0.039
 overflow+0.038
 structures+0.037
Neg logit contributions
 name-0.056
tie-0.051
ka-0.051
 aboard-0.050
 powers-0.049
Token't
Token ID470
Feature activation-2.38e-01

Pos logit contributions
 mess+0.027
 neither+0.027
 wasteful+0.026
 overflow+0.023
 roughly+0.022
Neg logit contributions
 name-0.032
tie-0.032
ance-0.031
ka-0.031
dot-0.031
Token want
Token ID765
Feature activation+3.98e-01

Pos logit contributions
ka+0.009
tie+0.009
ance+0.009
 name+0.009
dot+0.009
Neg logit contributions
 neither-0.010
 mess-0.010
 wasteful-0.009
 overflow-0.008
 fighting-0.008
Token to
Token ID284
Feature activation+1.29e+00

Pos logit contributions
 neither+0.019
 mess+0.019
 wasteful+0.018
 overflow+0.017
 roughly+0.017
Neg logit contributions
tie-0.012
ka-0.012
 name-0.012
dot-0.011
ance-0.011
Token eat
Token ID4483
Feature activation+1.81e+00

Pos logit contributions
 wasteful+0.059
 neither+0.059
 mess+0.055
 rats+0.052
 structures+0.052
Neg logit contributions
 name-0.046
tie-0.040
ka-0.038
dot-0.037
 Crown-0.036
Token the
Token ID262
Feature activation+8.33e-01

Pos logit contributions
 neither+0.079
 wasteful+0.075
 overflow+0.074
 forgot+0.072
 mess+0.071
Neg logit contributions
 name-0.059
 aboard-0.057
ka-0.056
tie-0.055
ance-0.054
Token pit
Token ID6028
Feature activation+4.00e+00

Pos logit contributions
 neither+0.032
 wasteful+0.029
 mess+0.027
 roughly+0.026
 overflow+0.026
Neg logit contributions
 name-0.036
ka-0.034
tie-0.034
 aboard-0.033
 Crown-0.032
Token.
Token ID13
Feature activation-1.52e-01

Pos logit contributions
 wasteful+0.186
Sadly+0.183
 ruining+0.180
 sudden+0.175
 overflow+0.175
Neg logit contributions
 aboard-0.109
tie-0.108
ka-0.106
iest-0.096
ed-0.091
Token I
Token ID314
Feature activation-1.14e-01

Pos logit contributions
ka+0.005
tie+0.004
 name+0.004
ance+0.004
dot+0.004
Neg logit contributions
 neither-0.008
 mess-0.007
 wasteful-0.007
 overflow-0.007
 fighting-0.007
Token only
Token ID691
Feature activation+2.08e-02

Pos logit contributions
ka+0.005
tie+0.005
 name+0.005
ance+0.005
dot+0.005
Neg logit contributions
 neither-0.004
 mess-0.004
 wasteful-0.004
 overflow-0.004
 fighting-0.004
Token want
Token ID765
Feature activation+8.16e-01

Pos logit contributions
 neither+0.001
 mess+0.001
 wasteful+0.001
 overflow+0.001
 fighting+0.001
Neg logit contributions
ka-0.001
tie-0.001
 name-0.001
ance-0.001
dot-0.001
Token to
Token ID284
Feature activation+1.38e+00

Pos logit contributions
 neither+0.041
 mess+0.040
 wasteful+0.039
 overflow+0.037
 roughly+0.036
Neg logit contributions
 name-0.023
tie-0.022
ka-0.022
 aboard-0.022
 powers-0.021
Token you
Token ID345
Feature activation+1.15e+00

Pos logit contributions
 mess+0.067
 wasteful+0.066
 neither+0.066
 overflow+0.064
 roughly+0.059
Neg logit contributions
ance-0.079
 name-0.076
ka-0.075
 Magic-0.072
 aboard-0.072
Token to
Token ID284
Feature activation+1.25e+00

Pos logit contributions
 mess+0.042
 neither+0.041
 wasteful+0.039
 overflow+0.036
 ruining+0.035
Neg logit contributions
 name-0.047
tie-0.046
ance-0.046
ka-0.044
dot-0.044
Token clean
Token ID3424
Feature activation+1.23e+00

Pos logit contributions
 neither+0.055
 wasteful+0.052
 mess+0.051
 structures+0.047
 rats+0.047
Neg logit contributions
 name-0.046
tie-0.041
ance-0.041
ka-0.040
shirt-0.039
Token up
Token ID510
Feature activation+1.95e+00

Pos logit contributions
 neither+0.059
 wasteful+0.055
 mess+0.052
 overflow+0.052
 erupted+0.050
Neg logit contributions
ka-0.039
 name-0.038
 aboard-0.038
tie-0.036
shirt-0.036
Token your
Token ID534
Feature activation+1.27e+00

Pos logit contributions
 neither+0.082
 wasteful+0.076
 mess+0.075
 overflow+0.073
 quarrel+0.070
Neg logit contributions
 aboard-0.066
ance-0.066
ka-0.064
 name-0.064
shirt-0.062
Token room
Token ID2119
Feature activation+3.99e+00

Pos logit contributions
 neither+0.051
 wasteful+0.044
 unfortunately+0.038
 overflow+0.038
 mosquitoes+0.037
Neg logit contributions
 name-0.057
 powers-0.051
ka-0.051
 gown-0.048
anger-0.047
Token!
Token ID0
Feature activation+2.98e-01

Pos logit contributions
 mosquitoes+0.173
 neither+0.172
Uh+0.168
 wasteful+0.168
 snakes+0.163
Neg logit contributions
ka-0.111
tie-0.110
 powers-0.109
 name-0.109
ance-0.108
Token Now
Token ID2735
Feature activation+1.17e+00

Pos logit contributions
 mess+0.013
 neither+0.013
 wasteful+0.012
 overflow+0.011
 ruining+0.011
Neg logit contributions
 name-0.011
ka-0.010
tie-0.010
ance-0.010
dot-0.010
Token,
Token ID11
Feature activation+8.32e-01

Pos logit contributions
 wasteful+0.042
 neither+0.042
 mess+0.041
 overflow+0.038
 roughly+0.037
Neg logit contributions
 name-0.049
ance-0.047
 aboard-0.046
tie-0.045
ka-0.044
Token I
Token ID314
Feature activation-8.33e-02

Pos logit contributions
 neither+0.030
 mess+0.030
 wasteful+0.029
 overflow+0.026
 roughly+0.026
Neg logit contributions
 name-0.036
tie-0.033
ance-0.033
 aboard-0.033
ka-0.032
Token have
Token ID423
Feature activation+6.10e-01

Pos logit contributions
ka+0.003
tie+0.003
 name+0.003
ance+0.003
dot+0.003
Neg logit contributions
 neither-0.004
 mess-0.004
 wasteful-0.003
 overflow-0.003
 fighting-0.003
Token's
Token ID338
Feature activation+5.65e-01

Pos logit contributions
 wasteful+0.035
 neither+0.035
 mess+0.033
 roughly+0.032
 overflow+0.031
Neg logit contributions
 name-0.036
ance-0.033
ka-0.033
 aboard-0.032
tie-0.032
Token not
Token ID407
Feature activation+1.24e+00

Pos logit contributions
 neither+0.027
 mess+0.026
 wasteful+0.026
 ruining+0.023
 roughly+0.023
Neg logit contributions
 name-0.019
tie-0.017
ka-0.017
ance-0.017
 aboard-0.016
Token play
Token ID711
Feature activation+1.47e+00

Pos logit contributions
 wasteful+0.061
 neither+0.058
 mess+0.056
 floor+0.053
 structures+0.052
Neg logit contributions
 name-0.041
 aboard-0.034
tie-0.033
ance-0.032
ka-0.031
Token in
Token ID287
Feature activation+2.09e+00

Pos logit contributions
 neither+0.070
 wasteful+0.067
 mess+0.063
 overflow+0.060
 anth+0.060
Neg logit contributions
 name-0.046
 aboard-0.043
 powers-0.041
tie-0.040
ka-0.039
Token the
Token ID262
Feature activation+1.11e+00

Pos logit contributions
 wasteful+0.104
 neither+0.103
 forgot+0.095
 overflow+0.093
 mess+0.091
Neg logit contributions
 aboard-0.057
 outer-0.053
 powers-0.048
 name-0.046
ance-0.045
Token forest
Token ID8222
Feature activation+3.97e+00

Pos logit contributions
 neither+0.046
 wasteful+0.039
 mess+0.034
 fighting+0.034
 roughly+0.034
Neg logit contributions
 name-0.048
orn-0.043
 cloak-0.042
tie-0.041
 amulet-0.040
Token again
Token ID757
Feature activation+2.99e+00

Pos logit contributions
 wasteful+0.189
 neither+0.185
 ruining+0.175
 overflow+0.175
 spitting+0.175
Neg logit contributions
 aboard-0.103
tie-0.086
ka-0.085
 powers-0.085
 ti-0.083
Token,"
Token ID553
Feature activation-5.81e-01

Pos logit contributions
 wasteful+0.146
 neither+0.143
 overflow+0.141
 ruining+0.135
 mess+0.132
Neg logit contributions
 aboard-0.087
 name-0.079
ka-0.075
 powers-0.069
 outer-0.067
Token L
Token ID406
Feature activation-9.14e-01

Pos logit contributions
tie+0.027
ka+0.027
ance+0.026
dot+0.026
shirt+0.026
Neg logit contributions
 neither-0.021
 mess-0.019
 wasteful-0.018
 fighting-0.016
 overflow-0.016
Tokenila
Token ID10102
Feature activation+1.13e+00

Pos logit contributions
tie+0.037
 name+0.036
ka+0.035
shirt+0.035
 amulet+0.035
Neg logit contributions
 neither-0.037
 mess-0.035
 wasteful-0.033
 fighting-0.031
 mosquitoes-0.030
Token said
Token ID531
Feature activation-7.00e-01

Pos logit contributions
 mess+0.047
 overflow+0.046
 wasteful+0.046
 neither+0.045
 ruining+0.044
Neg logit contributions
tie-0.041
 name-0.040
ka-0.039
leaf-0.038
shirt-0.038
TokenLet
Token ID5756
Feature activation+9.11e-01

Pos logit contributions
 neither+0.010
 mess+0.010
 wasteful+0.009
 ruining+0.008
 overflow+0.008
Neg logit contributions
tie-0.008
ka-0.008
 name-0.008
dot-0.008
shirt-0.008
Token's
Token ID338
Feature activation+8.76e-01

Pos logit contributions
 neither+0.037
 wasteful+0.037
 mess+0.035
 roughly+0.034
 ruining+0.033
Neg logit contributions
 name-0.035
ka-0.032
ance-0.032
 aboard-0.032
tie-0.032
Token go
Token ID467
Feature activation-1.63e+00

Pos logit contributions
 wasteful+0.042
 neither+0.042
 mess+0.042
 mosquitoes+0.038
 rats+0.037
Neg logit contributions
 name-0.030
tie-0.025
 aboard-0.024
ka-0.024
ance-0.024
Token down
Token ID866
Feature activation+1.37e+00

Pos logit contributions
ka+0.065
anced+0.064
ance+0.064
tie+0.063
ocket+0.063
Neg logit contributions
 mess-0.064
 neither-0.062
 fighting-0.056
 overflow-0.055
 wasteful-0.051
Token the
Token ID262
Feature activation+1.79e+00

Pos logit contributions
 wasteful+0.062
 neither+0.061
 mess+0.060
 overflow+0.054
 ruining+0.053
Neg logit contributions
 aboard-0.049
ka-0.043
ance-0.043
 name-0.043
tie-0.041
Token pit
Token ID6028
Feature activation+3.96e+00

Pos logit contributions
 neither+0.077
 wasteful+0.074
 roughly+0.063
 unfortunately+0.062
 mosquitoes+0.061
Neg logit contributions
 name-0.070
 aboard-0.064
orn-0.059
anger-0.058
 amulet-0.057
Token,"
Token ID553
Feature activation+1.49e+00

Pos logit contributions
 wasteful+0.202
 mosquitoes+0.192
 ruining+0.184
 neither+0.183
 spitting+0.182
Neg logit contributions
 aboard-0.120
tie-0.098
 name-0.095
ka-0.094
anger-0.088
Token Ben
Token ID3932
Feature activation+9.20e-01

Pos logit contributions
 mess+0.055
 wasteful+0.052
 neither+0.050
 ruining+0.050
 overflow+0.050
Neg logit contributions
 name-0.062
ka-0.060
shirt-0.056
dot-0.054
 Magic-0.054
Token said
Token ID531
Feature activation-1.19e-01

Pos logit contributions
 neither+0.036
 mess+0.036
 wasteful+0.036
 ruining+0.034
 overflow+0.033
Neg logit contributions
 name-0.036
ka-0.035
tie-0.035
shirt-0.034
dot-0.033
Token.
Token ID13
Feature activation+2.94e-01

Pos logit contributions
ka+0.005
tie+0.005
 name+0.004
ance+0.004
dot+0.004
Neg logit contributions
 neither-0.005
 mess-0.005
 wasteful-0.005
 overflow-0.004
 fighting-0.004
Token "
Token ID366
Feature activation+2.04e-01

Pos logit contributions
 mess+0.013
 neither+0.013
 wasteful+0.012
 ruining+0.011
 overflow+0.011
Neg logit contributions
 name-0.011
ka-0.010
tie-0.010
dot-0.010
shirt-0.010
Token sit
Token ID1650
Feature activation+9.46e-01

Pos logit contributions
 wasteful+0.055
 neither+0.054
 mess+0.051
 roughly+0.049
 structures+0.048
Neg logit contributions
 name-0.034
dot-0.029
ance-0.028
tie-0.028
ka-0.028
Token in
Token ID287
Feature activation+1.83e+00

Pos logit contributions
 neither+0.046
 mess+0.045
 wasteful+0.044
 overflow+0.042
 structures+0.042
Neg logit contributions
ka-0.028
tie-0.027
 name-0.027
 aboard-0.026
ance-0.025
Token the
Token ID262
Feature activation-2.04e-01

Pos logit contributions
 wasteful+0.089
 neither+0.088
 forgot+0.080
 overflow+0.080
keys+0.079
Neg logit contributions
 aboard-0.052
 outer-0.048
ka-0.047
 name-0.045
ance-0.043
Token old
Token ID1468
Feature activation+1.42e+00

Pos logit contributions
ka+0.008
tie+0.008
 name+0.008
ance+0.008
dot+0.008
Neg logit contributions
 mess-0.008
 neither-0.008
 wasteful-0.007
 overflow-0.007
 fighting-0.007
Token arm
Token ID3211
Feature activation+2.46e+00

Pos logit contributions
 neither+0.062
 wasteful+0.058
 overflow+0.054
 ruining+0.052
 mess+0.050
Neg logit contributions
 name-0.051
 gown-0.050
ka-0.049
tie-0.048
shirt-0.048
Tokenchair
Token ID16337
Feature activation+3.96e+00

Pos logit contributions
 neither+0.126
 wasteful+0.122
 overflow+0.115
 ruining+0.115
 mess+0.112
Neg logit contributions
tie-0.066
shirt-0.065
ance-0.064
 gown-0.062
anger-0.061
Token.
Token ID13
Feature activation+5.31e-01

Pos logit contributions
 wasteful+0.198
 anth+0.188
 structures+0.188
 neither+0.188
 overflow+0.187
Neg logit contributions
tie-0.094
 aboard-0.086
ance-0.083
 gown-0.079
ka-0.078
Token It
Token ID632
Feature activation+9.89e-01

Pos logit contributions
 mess+0.021
 neither+0.020
 wasteful+0.018
 ruining+0.017
 overflow+0.017
Neg logit contributions
 name-0.023
tie-0.021
ka-0.021
dot-0.021
ance-0.021
Token is
Token ID318
Feature activation+1.41e+00

Pos logit contributions
 neither+0.048
 wasteful+0.046
 mess+0.045
 roughly+0.043
 overflow+0.043
Neg logit contributions
 name-0.029
ka-0.028
tie-0.026
 aboard-0.026
ance-0.026
Token wet
Token ID9583
Feature activation+2.14e+00

Pos logit contributions
 neither+0.049
 mess+0.042
 structures+0.042
 overflow+0.042
 wasteful+0.042
Neg logit contributions
 name-0.061
ka-0.056
tie-0.055
 aboard-0.055
dot-0.053
Token and
Token ID290
Feature activation+1.18e+00

Pos logit contributions
 neither+0.110
 wasteful+0.107
 anth+0.098
 overflow+0.097
 ruining+0.097
Neg logit contributions
tie-0.056
ka-0.055
 aboard-0.052
 name-0.048
ance-0.047
Token children
Token ID1751
Feature activation+1.94e+00

Pos logit contributions
 neither+0.038
 wasteful+0.037
 overflow+0.037
 mess+0.036
 roughly+0.034
Neg logit contributions
ka-0.053
 name-0.053
ance-0.050
 aboard-0.050
 powers-0.048
Token!
Token ID0
Feature activation+3.09e-01

Pos logit contributions
 overflow+0.092
 neither+0.088
 wasteful+0.086
 structures+0.083
 anth+0.083
Neg logit contributions
ka-0.053
 name-0.053
tie-0.051
 aboard-0.049
leaf-0.048
Token You
Token ID921
Feature activation+1.24e+00

Pos logit contributions
 neither+0.014
 mess+0.014
 wasteful+0.013
 overflow+0.012
 ruining+0.012
Neg logit contributions
ka-0.011
 name-0.011
ance-0.010
dot-0.010
tie-0.010
Token ruined
Token ID20484
Feature activation+2.80e+00

Pos logit contributions
 neither+0.052
 mess+0.051
 wasteful+0.047
 overflow+0.047
 structures+0.046
Neg logit contributions
 name-0.043
leaf-0.042
ka-0.042
ance-0.041
tie-0.041
Token the
Token ID262
Feature activation+1.26e+00

Pos logit contributions
 overflow+0.128
 neither+0.127
 forgot+0.123
 erupted+0.122
Uh+0.122
Neg logit contributions
 name-0.075
ka-0.072
 aboard-0.071
ance-0.066
 Magic-0.064
Token cake
Token ID12187
Feature activation+3.94e+00

Pos logit contributions
 neither+0.046
 wasteful+0.039
 roughly+0.037
 overflow+0.036
 ruining+0.035
Neg logit contributions
 name-0.059
ka-0.054
tie-0.052
 aboard-0.052
 Crown-0.052
Token!
Token ID0
Feature activation+2.18e-01

Pos logit contributions
 wasteful+0.213
 neither+0.210
Uh+0.199
 overflow+0.197
 mosquitoes+0.196
Neg logit contributions
 name-0.069
tie-0.063
 aboard-0.063
ance-0.057
anger-0.056
Token You
Token ID921
Feature activation+1.36e+00

Pos logit contributions
 neither+0.010
 mess+0.010
 wasteful+0.009
 overflow+0.009
 ruining+0.008
Neg logit contributions
ka-0.008
 name-0.007
tie-0.007
ance-0.007
dot-0.007
Token have
Token ID423
Feature activation+1.09e+00

Pos logit contributions
 neither+0.057
 mess+0.055
 wasteful+0.052
 structures+0.051
 overflow+0.051
Neg logit contributions
 name-0.047
leaf-0.046
ka-0.046
tie-0.046
ance-0.045
Token to
Token ID284
Feature activation+1.09e+00

Pos logit contributions
 neither+0.052
 mess+0.051
 overflow+0.050
 wasteful+0.048
 roughly+0.048
Neg logit contributions
 name-0.029
ka-0.029
tie-0.028
ance-0.028
leaf-0.028
Token say
Token ID910
Feature activation+9.86e-01

Pos logit contributions
 neither+0.048
 mess+0.042
 wasteful+0.042
 rats+0.041
 snakes+0.040
Neg logit contributions
 name-0.041
ance-0.037
ka-0.036
ad-0.035
leaf-0.035
Token should
Token ID815
Feature activation+1.08e+00

Pos logit contributions
 mess+0.042
 neither+0.042
 wasteful+0.040
 overflow+0.038
 structures+0.038
Neg logit contributions
 name-0.027
tie-0.025
ance-0.024
ka-0.024
leaf-0.023
Token not
Token ID407
Feature activation+1.99e+00

Pos logit contributions
 neither+0.044
 mess+0.041
 wasteful+0.041
 rats+0.038
 structures+0.038
Neg logit contributions
 name-0.044
tie-0.038
ance-0.037
ka-0.037
dot-0.036
Token fight
Token ID1907
Feature activation+1.88e+00

Pos logit contributions
 neither+0.094
 wasteful+0.090
 mess+0.085
 mosquitoes+0.083
 structures+0.082
Neg logit contributions
 name-0.071
tie-0.058
ad-0.055
orn-0.054
shirt-0.053
Token over
Token ID625
Feature activation+2.27e+00

Pos logit contributions
 overflow+0.084
 neither+0.084
 wasteful+0.082
 mess+0.082
 flood+0.080
Neg logit contributions
 name-0.058
 aboard-0.053
ka-0.052
tie-0.051
ance-0.050
Token the
Token ID262
Feature activation+1.79e+00

Pos logit contributions
 overflow+0.100
 neither+0.098
 mess+0.097
 wasteful+0.095
 timer+0.091
Neg logit contributions
 name-0.064
 aboard-0.063
tie-0.059
 powers-0.059
 Crown-0.059
Token banana
Token ID25996
Feature activation+3.94e+00

Pos logit contributions
 neither+0.081
 wasteful+0.072
 overflow+0.071
 roughly+0.068
oving+0.065
Neg logit contributions
 name-0.067
 Crown-0.060
orn-0.057
tie-0.057
ka-0.057
Token.
Token ID13
Feature activation+1.66e-01

Pos logit contributions
 wasteful+0.191
 sudden+0.189
 overflow+0.189
 neither+0.184
Uh+0.178
Neg logit contributions
anger-0.087
orn-0.085
ance-0.085
 name-0.085
tie-0.084
Token You
Token ID921
Feature activation+1.18e+00

Pos logit contributions
 mess+0.008
 neither+0.008
 wasteful+0.008
 overflow+0.007
 ruining+0.007
Neg logit contributions
 name-0.005
ka-0.005
tie-0.005
ance-0.005
dot-0.005
Token should
Token ID815
Feature activation+1.14e+00

Pos logit contributions
 mess+0.058
 neither+0.056
 wasteful+0.054
 overflow+0.052
 structures+0.052
Neg logit contributions
 name-0.036
tie-0.034
ance-0.032
ka-0.032
leaf-0.032
Token share
Token ID2648
Feature activation+2.00e+00

Pos logit contributions
 neither+0.048
 wasteful+0.044
 mess+0.044
 rats+0.042
 overflow+0.041
Neg logit contributions
 name-0.045
tie-0.039
ance-0.038
ka-0.038
dot-0.037
Token and
Token ID290
Feature activation+2.01e+00

Pos logit contributions
 neither+0.101
 overflow+0.099
 flood+0.094
 wasteful+0.094
 erupted+0.090
Neg logit contributions
 name-0.058
tie-0.051
 aboard-0.048
 Crown-0.047
ka-0.046
TokenNo
Token ID2949
Feature activation+1.06e+00

Pos logit contributions
 mess+0.040
 neither+0.036
 wasteful+0.036
 structures+0.034
 ruining+0.034
Neg logit contributions
leaf-0.033
tie-0.032
 name-0.032
shirt-0.032
ka-0.031
Token,
Token ID11
Feature activation+1.60e+00

Pos logit contributions
 wasteful+0.051
 neither+0.049
 overflow+0.047
 mess+0.047
 roughly+0.047
Neg logit contributions
 name-0.032
 powers-0.029
ka-0.029
ance-0.029
 aboard-0.028
Token we
Token ID356
Feature activation+7.64e-01

Pos logit contributions
 mess+0.053
 roughly+0.051
 wasteful+0.051
oving+0.049
 structures+0.048
Neg logit contributions
 name-0.068
 aboard-0.061
ocket-0.061
ance-0.060
 hello-0.060
Token cannot
Token ID2314
Feature activation-9.90e-02

Pos logit contributions
 neither+0.040
 mess+0.040
 wasteful+0.039
 structures+0.037
 overflow+0.037
Neg logit contributions
 name-0.021
tie-0.018
ka-0.017
dot-0.016
ance-0.016
Token go
Token ID467
Feature activation+8.38e-01

Pos logit contributions
ka+0.004
tie+0.004
 name+0.004
ance+0.004
dot+0.004
Neg logit contributions
 neither-0.004
 mess-0.004
 wasteful-0.004
 overflow-0.003
 fighting-0.003
Token outside
Token ID2354
Feature activation+3.93e+00

Pos logit contributions
 wasteful+0.043
 neither+0.043
 mess+0.043
 mosquitoes+0.039
 roughly+0.039
Neg logit contributions
 aboard-0.023
 name-0.022
ka-0.021
tie-0.021
dot-0.020
Token,"
Token ID553
Feature activation+1.14e+00

Pos logit contributions
 wasteful+0.206
 neither+0.188
Uh+0.178
 ruining+0.176
 mess+0.176
Neg logit contributions
 aboard-0.103
 name-0.088
 ti-0.079
ka-0.078
 gown-0.077
Token Mia
Token ID32189
Feature activation+8.79e-01

Pos logit contributions
 mess+0.037
 wasteful+0.035
 neither+0.035
 ruining+0.033
oving+0.031
Neg logit contributions
 name-0.055
ka-0.048
 aboard-0.048
dot-0.047
 Magic-0.047
Token says
Token ID1139
Feature activation-3.67e-01

Pos logit contributions
 neither+0.037
 wasteful+0.035
 mess+0.035
 overflow+0.034
 ruining+0.033
Neg logit contributions
 name-0.034
ka-0.031
ance-0.031
tie-0.030
leaf-0.029
Token.
Token ID13
Feature activation+5.73e-01

Pos logit contributions
ka+0.015
tie+0.015
 name+0.014
ance+0.014
dot+0.014
Neg logit contributions
 neither-0.015
 mess-0.014
 wasteful-0.013
 fighting-0.012
 overflow-0.012
Token
Token ID198
Feature activation+3.77e-01

Pos logit contributions
 mess+0.027
 neither+0.027
 wasteful+0.026
 overflow+0.024
 ruining+0.024
Neg logit contributions
 name-0.019
tie-0.018
dot-0.017
 aboard-0.017
ance-0.017
TokenWe
Token ID1135
Feature activation+5.74e-02

Pos logit contributions
 neither+0.015
 mess+0.015
 wasteful+0.014
 ruining+0.013
 overflow+0.013
Neg logit contributions
tie-0.013
ka-0.013
 name-0.013
dot-0.013
leaf-0.012
Token should
Token ID815
Feature activation+5.39e-01

Pos logit contributions
 neither+0.003
 mess+0.002
 wasteful+0.002
 overflow+0.002
 ruining+0.002
Neg logit contributions
ka-0.002
tie-0.002
 name-0.002
ance-0.002
dot-0.002
Token not
Token ID407
Feature activation+1.06e+00

Pos logit contributions
 neither+0.021
 mess+0.021
 wasteful+0.020
 overflow+0.018
 roughly+0.018
Neg logit contributions
 name-0.022
tie-0.021
ka-0.020
ance-0.020
dot-0.020
Token bother
Token ID11393
Feature activation+1.85e+00

Pos logit contributions
 neither+0.048
 wasteful+0.047
 mess+0.046
 rats+0.042
 structures+0.042
Neg logit contributions
 name-0.039
tie-0.035
dot-0.032
ance-0.032
ka-0.031
Token the
Token ID262
Feature activation+1.05e+00

Pos logit contributions
 mess+0.072
 wasteful+0.066
 neither+0.064
 overflow+0.063
 roughly+0.059
Neg logit contributions
 aboard-0.075
 name-0.073
ance-0.070
tie-0.068
ka-0.067
Token frogs
Token ID37475
Feature activation+3.91e+00

Pos logit contributions
 neither+0.048
 wasteful+0.047
 mess+0.044
 roughly+0.044
 ruining+0.042
Neg logit contributions
 name-0.038
ka-0.033
 aboard-0.032
tie-0.032
 Crown-0.031
Token.
Token ID13
Feature activation+8.98e-01

Pos logit contributions
 wasteful+0.176
 neither+0.170
 structures+0.169
Uh+0.166
 mouths+0.165
Neg logit contributions
 aboard-0.099
tie-0.097
 name-0.095
ance-0.088
orn-0.087
Token They
Token ID1119
Feature activation+1.31e+00

Pos logit contributions
 mess+0.036
 neither+0.033
 wasteful+0.031
 ruining+0.029
 overflow+0.029
Neg logit contributions
 name-0.039
tie-0.036
ka-0.035
dot-0.035
ance-0.035
Token live
Token ID2107
Feature activation+6.87e-01

Pos logit contributions
 neither+0.058
 mess+0.058
 wasteful+0.057
 roughly+0.053
 structures+0.052
Neg logit contributions
 name-0.044
tie-0.041
ance-0.040
ka-0.039
leaf-0.039
Token here
Token ID994
Feature activation+2.73e+00

Pos logit contributions
 neither+0.031
 mess+0.030
 wasteful+0.029
 overflow+0.027
 ruining+0.026
Neg logit contributions
 name-0.023
ka-0.023
tie-0.023
 aboard-0.023
ance-0.022
Token and
Token ID290
Feature activation+2.24e+00

Pos logit contributions
 mess+0.114
 wasteful+0.113
 neither+0.112
 overflow+0.107
 ruining+0.102
Neg logit contributions
 name-0.097
tie-0.096
 aboard-0.094
ka-0.087
ance-0.086
Token's
Token ID338
Feature activation+5.26e-01

Pos logit contributions
 wasteful+0.044
 neither+0.043
 mess+0.041
 overflow+0.041
 ruining+0.041
Neg logit contributions
 name-0.040
ance-0.036
ka-0.036
 aboard-0.035
tie-0.035
Token never
Token ID1239
Feature activation+1.08e+00

Pos logit contributions
 neither+0.025
 mess+0.024
 wasteful+0.024
 rats+0.022
 roughly+0.022
Neg logit contributions
 name-0.018
tie-0.016
ka-0.016
ance-0.016
dot-0.015
Token go
Token ID467
Feature activation+2.52e-01

Pos logit contributions
 neither+0.052
 wasteful+0.050
 mess+0.047
 roughly+0.045
 mosquitoes+0.044
Neg logit contributions
 name-0.036
tie-0.031
ka-0.031
ance-0.030
 Crown-0.030
Token into
Token ID656
Feature activation+2.04e+00

Pos logit contributions
 neither+0.012
 mess+0.012
 wasteful+0.012
 overflow+0.011
 roughly+0.011
Neg logit contributions
tie-0.008
 name-0.008
ka-0.008
 aboard-0.007
dot-0.007
Token a
Token ID257
Feature activation+1.10e+00

Pos logit contributions
 neither+0.099
 wasteful+0.093
 overflow+0.088
 mess+0.087
 forgot+0.084
Neg logit contributions
 aboard-0.061
 outer-0.060
 powers-0.054
ance-0.052
 name-0.051
Token cave
Token ID11527
Feature activation+3.91e+00

Pos logit contributions
 neither+0.049
 wasteful+0.042
 mess+0.041
 ruining+0.039
 overflow+0.038
Neg logit contributions
 name-0.040
 aboard-0.037
ka-0.037
dot-0.037
tie-0.037
Token again
Token ID757
Feature activation+3.07e+00

Pos logit contributions
 wasteful+0.175
 neither+0.174
 mouths+0.168
 mosquitoes+0.167
 structures+0.164
Neg logit contributions
 aboard-0.117
tie-0.108
 ti-0.105
ka-0.102
 name-0.101
Token!"
Token ID2474
Feature activation+8.54e-02

Pos logit contributions
 wasteful+0.154
 neither+0.147
 overflow+0.143
 structures+0.139
 ruining+0.138
Neg logit contributions
 aboard-0.090
 name-0.080
ka-0.071
 powers-0.067
 ti-0.067
Token L
Token ID406
Feature activation-3.88e-01

Pos logit contributions
 neither+0.003
 mess+0.002
 wasteful+0.002
 overflow+0.002
 ruining+0.002
Neg logit contributions
ka-0.004
 name-0.004
tie-0.004
dot-0.004
ance-0.004
Tokenila
Token ID10102
Feature activation+1.49e+00

Pos logit contributions
tie+0.014
ka+0.013
 name+0.013
dot+0.013
shirt+0.013
Neg logit contributions
 neither-0.018
 mess-0.017
 wasteful-0.016
 fighting-0.015
 overflow-0.015
Token said
Token ID531
Feature activation-9.58e-01

Pos logit contributions
 wasteful+0.071
 overflow+0.070
 mess+0.069
 neither+0.069
 ruining+0.069
Neg logit contributions
tie-0.048
 name-0.045
ka-0.044
shirt-0.044
leaf-0.043
Token it
Token ID340
Feature activation-1.63e+00

Pos logit contributions
ka+0.070
leaf+0.065
tie+0.065
ance+0.065
anger+0.064
Neg logit contributions
 mess-0.071
 neither-0.071
 wasteful-0.060
 not-0.059
 floor-0.059
Token would
Token ID561
Feature activation-1.45e+00

Pos logit contributions
leaf+0.074
ka+0.071
tie+0.068
dot+0.068
 aboard+0.067
Neg logit contributions
 neither-0.062
 mess-0.061
 floor-0.052
 wasteful-0.050
 overflow-0.049
Token give
Token ID1577
Feature activation-2.20e-01

Pos logit contributions
ka+0.054
ance+0.053
leaf+0.052
dot+0.051
 aboard+0.051
Neg logit contributions
 mess-0.064
 neither-0.061
 overflow-0.057
 wasteful-0.054
 not-0.054
Token her
Token ID607
Feature activation-3.31e+00

Pos logit contributions
ka+0.008
tie+0.008
 name+0.008
dot+0.008
ance+0.008
Neg logit contributions
 neither-0.010
 mess-0.009
 wasteful-0.009
 fighting-0.008
 overflow-0.008
Token a
Token ID257
Feature activation-4.03e+00

Pos logit contributions
leaf+0.159
orn+0.159
anced+0.155
reetings+0.153
lynn+0.150
Neg logit contributions
 mess-0.109
 neither-0.098
 not-0.092
 both-0.088
 messy-0.077
Token special
Token ID2041
Feature activation-5.07e+00

Pos logit contributions
ance+0.241
anced+0.239
ully+0.207
lynn+0.204
umper+0.204
Neg logit contributions
 mess-0.135
 messy-0.088
 not-0.088
 bit-0.086
 fuss-0.082
Token power
Token ID1176
Feature activation-2.43e+00

Pos logit contributions
ance+0.248
anced+0.244
umper+0.241
aved+0.225
igh+0.224
Neg logit contributions
 mess-0.203
 floor-0.151
 mistake-0.147
 game-0.144
 noise-0.143
Token.
Token ID13
Feature activation-1.73e-01

Pos logit contributions
ka+0.119
anced+0.113
leaf+0.113
dot+0.112
ance+0.109
Neg logit contributions
 mess-0.081
 neither-0.075
 floor-0.065
 wasteful-0.061
 overflow-0.060
Token
Token ID198
Feature activation-3.76e-01

Pos logit contributions
ka+0.007
tie+0.007
ance+0.007
 name+0.006
dot+0.006
Neg logit contributions
 neither-0.007
 mess-0.007
 wasteful-0.006
 fighting-0.006
 overflow-0.006
Token
Token ID198
Feature activation+6.08e-01

Pos logit contributions
ka+0.016
tie+0.015
ance+0.015
 name+0.015
dot+0.015
Neg logit contributions
 neither-0.014
 mess-0.014
 wasteful-0.013
 fighting-0.012
 overflow-0.012
TokenOne
Token ID3198
Feature activation-7.52e-01

Pos logit contributions
 neither+0.025
 mess+0.024
 wasteful+0.023
 ruining+0.021
 overflow+0.021
Neg logit contributions
tie-0.025
 name-0.024
ka-0.024
dot-0.024
shirt-0.023
Token parents
Token ID3397
Feature activation-1.43e+00

Pos logit contributions
tie+0.081
ance+0.081
orn+0.080
ka+0.080
anced+0.079
Neg logit contributions
 neither-0.034
 mess-0.033
 fighting-0.027
 quarrel-0.025
 wasteful-0.024
Token gave
Token ID2921
Feature activation-1.33e+00

Pos logit contributions
ka+0.060
dot+0.059
tie+0.058
 powers+0.056
ance+0.056
Neg logit contributions
 mess-0.054
 neither-0.053
 fighting-0.049
 wasteful-0.046
 ruining-0.044
Token him
Token ID683
Feature activation-2.37e+00

Pos logit contributions
tie+0.054
dot+0.053
ka+0.053
anger+0.053
 amulet+0.052
Neg logit contributions
 neither-0.051
 mess-0.050
 fighting-0.045
 wasteful-0.043
 mice-0.041
Token a
Token ID257
Feature activation-2.79e+00

Pos logit contributions
 amulet+0.104
ka+0.104
ocket+0.103
anced+0.102
tie+0.102
Neg logit contributions
 neither-0.081
 mess-0.071
 wasteful-0.068
 both-0.062
 raw-0.061
Token pe
Token ID613
Feature activation-3.90e+00

Pos logit contributions
ance+0.146
anced+0.140
ka+0.129
dot+0.123
tie+0.123
Neg logit contributions
 mess-0.099
 neither-0.078
 quarrel-0.077
 flood-0.072
 fight-0.070
Tokenb
Token ID65
Feature activation-5.04e+00

Pos logit contributions
anced+0.155
ance+0.139
 Magic+0.139
icol+0.139
leaf+0.137
Neg logit contributions
 mess-0.142
 neither-0.136
 not-0.125
 wasteful-0.124
 pies-0.113
Tokenble
Token ID903
Feature activation-1.76e+00

Pos logit contributions
leaf+0.282
 outer+0.272
Ta+0.266
shirt+0.258
chery+0.256
Neg logit contributions
 neither-0.107
 not-0.093
 both-0.061
 mess-0.058
 pies-0.056
Token.
Token ID13
Feature activation-1.50e+00

Pos logit contributions
ka+0.079
dot+0.075
anced+0.075
leaf+0.075
ance+0.074
Neg logit contributions
 neither-0.062
 mess-0.060
 wasteful-0.053
 fighting-0.050
 roughly-0.050
Token He
Token ID679
Feature activation-1.28e+00

Pos logit contributions
ka+0.069
ocket+0.066
 powers+0.066
orn+0.065
ance+0.064
Neg logit contributions
 neither-0.053
 mess-0.047
 wasteful-0.044
 fighting-0.043
 roughly-0.042
Token had
Token ID550
Feature activation-1.22e+00

Pos logit contributions
ka+0.054
ance+0.054
tie+0.053
dot+0.053
leaf+0.051
Neg logit contributions
 mess-0.047
 neither-0.046
 wasteful-0.041
 fighting-0.039
 roughly-0.038
Token never
Token ID1239
Feature activation-7.10e-01

Pos logit contributions
ka+0.050
tie+0.048
ance+0.048
anger+0.047
 name+0.047
Neg logit contributions
 mess-0.045
 neither-0.045
 wasteful-0.039
 fighting-0.038
 floor-0.038
Token He
Token ID679
Feature activation-1.70e-01

Pos logit contributions
 mess+0.025
 neither+0.025
 wasteful+0.024
 ruining+0.022
 overflow+0.022
Neg logit contributions
 name-0.020
ka-0.019
tie-0.019
ance-0.019
shirt-0.019
Token gave
Token ID2921
Feature activation-4.75e-01

Pos logit contributions
ka+0.007
tie+0.007
 name+0.007
ance+0.007
dot+0.007
Neg logit contributions
 neither-0.007
 mess-0.007
 wasteful-0.006
 overflow-0.006
 fighting-0.006
Token him
Token ID683
Feature activation-2.08e+00

Pos logit contributions
ka+0.021
tie+0.020
dot+0.020
 name+0.020
ance+0.020
Neg logit contributions
 neither-0.017
 mess-0.017
 wasteful-0.015
 fighting-0.014
 ruining-0.014
Token a
Token ID257
Feature activation-3.07e+00

Pos logit contributions
ka+0.096
anger+0.095
dot+0.092
anced+0.092
tie+0.091
Neg logit contributions
 neither-0.070
 mess-0.062
 wasteful-0.060
 mice-0.057
 rats-0.057
Token beautiful
Token ID4950
Feature activation-3.95e+00

Pos logit contributions
ance+0.162
anced+0.154
ka+0.143
dot+0.140
ancing+0.131
Neg logit contributions
 mess-0.111
 wasteful-0.082
 fuss-0.079
 neither-0.078
 fight-0.076
Token golden
Token ID10861
Feature activation-4.98e+00

Pos logit contributions
ance+0.177
anced+0.164
aved+0.153
leaf+0.149
dot+0.147
Neg logit contributions
 mess-0.175
 floor-0.136
 mistake-0.131
 table-0.130
 pie-0.127
Token sword
Token ID8429
Feature activation-1.99e+00

Pos logit contributions
ance+0.220
anced+0.210
leaf+0.196
dot+0.190
aved+0.189
Neg logit contributions
 mess-0.217
 floor-0.176
 mistake-0.162
 table-0.159
 pie-0.158
Token.
Token ID13
Feature activation-1.23e-01

Pos logit contributions
ka+0.096
leaf+0.092
dot+0.091
ance+0.089
anced+0.089
Neg logit contributions
 neither-0.063
 mess-0.061
 fighting-0.051
 roughly-0.050
 not-0.049
Token
Token ID198
Feature activation-1.95e-01

Pos logit contributions
ka+0.005
tie+0.005
 name+0.005
ance+0.004
dot+0.004
Neg logit contributions
 neither-0.005
 mess-0.005
 wasteful-0.005
 overflow-0.004
 fighting-0.004
Token
Token ID198
Feature activation+8.53e-01

Pos logit contributions
ka+0.008
tie+0.008
 name+0.007
ance+0.007
dot+0.007
Neg logit contributions
 neither-0.008
 mess-0.008
 wasteful-0.007
 overflow-0.007
 fighting-0.007
TokenThe
Token ID464
Feature activation-1.11e+00

Pos logit contributions
 neither+0.034
 mess+0.033
 wasteful+0.030
 ruining+0.028
 fighting+0.028
Neg logit contributions
shirt-0.035
tie-0.035
leaf-0.035
ka-0.034
yo-0.034
Token and
Token ID290
Feature activation-1.54e+00

Pos logit contributions
anced+0.070
leaf+0.069
tie+0.069
ka+0.068
ance+0.068
Neg logit contributions
 mess-0.051
 neither-0.043
 structures-0.041
 floor-0.040
 snakes-0.040
Token always
Token ID1464
Feature activation-1.88e+00

Pos logit contributions
ka+0.075
 aboard+0.075
anger+0.074
tie+0.072
ance+0.072
Neg logit contributions
 neither-0.048
 mess-0.047
 not-0.043
 wasteful-0.041
 floor-0.039
Token wore
Token ID12408
Feature activation-2.76e+00

Pos logit contributions
anger+0.091
leaf+0.087
ance+0.085
tie+0.084
 aboard+0.083
Neg logit contributions
 neither-0.064
 mess-0.061
 fighting-0.056
 ruining-0.055
 forgot-0.055
Token a
Token ID257
Feature activation-2.88e+00

Pos logit contributions
anced+0.151
ane+0.146
ance+0.144
oured+0.142
anger+0.142
Neg logit contributions
 mess-0.074
 not-0.064
 both-0.062
 mice-0.056
 structures-0.053
Token spark
Token ID9009
Feature activation-2.28e+00

Pos logit contributions
ance+0.179
anced+0.170
leaf+0.148
 aboard+0.145
ancing+0.141
Neg logit contributions
 mess-0.103
 floor-0.069
 messy-0.068
 not-0.065
 roughly-0.061
Tokenly
Token ID306
Feature activation-4.98e+00

Pos logit contributions
anced+0.102
ance+0.096
leaf+0.094
dot+0.091
 aboard+0.090
Neg logit contributions
 mess-0.088
 neither-0.080
 floor-0.071
 anymore-0.071
 wasteful-0.069
Token dress
Token ID6576
Feature activation-2.31e+00

Pos logit contributions
ance+0.240
leaf+0.228
anced+0.226
iddled+0.204
year+0.194
Neg logit contributions
 mess-0.206
 floor-0.191
 bit-0.158
 table-0.154
 while-0.153
Token.
Token ID13
Feature activation-6.66e-01

Pos logit contributions
leaf+0.121
anced+0.120
ka+0.116
dot+0.115
 amulet+0.114
Neg logit contributions
 neither-0.064
 mess-0.061
 not-0.052
 floor-0.052
 anymore-0.051
Token One
Token ID1881
Feature activation-1.48e+00

Pos logit contributions
ka+0.033
leaf+0.032
tie+0.032
ance+0.031
 aboard+0.031
Neg logit contributions
 neither-0.021
 mess-0.020
 wasteful-0.018
 fighting-0.017
 roughly-0.017
Token day
Token ID1110
Feature activation-6.14e-01

Pos logit contributions
anced+0.083
ance+0.081
ka+0.080
leaf+0.079
tie+0.079
Neg logit contributions
 mess-0.037
 neither-0.035
 wasteful-0.028
 floor-0.028
 roughly-0.027
Token,
Token ID11
Feature activation+2.39e-01

Pos logit contributions
tie+0.030
leaf+0.030
ka+0.030
ance+0.029
shirt+0.029
Neg logit contributions
 neither-0.020
 mess-0.018
 wasteful-0.016
 fighting-0.016
 roughly-0.015
Token Pixie
Token ID46687
Feature activation-1.39e+00

Pos logit contributions
 neither+0.005
 mess+0.005
 wasteful+0.005
 overflow+0.004
 ruining+0.004
Neg logit contributions
 name-0.005
ka-0.005
tie-0.005
ance-0.005
dot-0.005
Token gave
Token ID2921
Feature activation-1.57e+00

Pos logit contributions
ka+0.059
tie+0.058
dot+0.057
leaf+0.056
shirt+0.056
Neg logit contributions
 mess-0.051
 neither-0.051
 wasteful-0.044
 fighting-0.041
 roughly-0.041
Token her
Token ID607
Feature activation-2.54e+00

Pos logit contributions
dot+0.073
ka+0.072
tie+0.071
shirt+0.070
anger+0.069
Neg logit contributions
 neither-0.053
 mess-0.046
 Amy-0.043
 fighting-0.041
 mice-0.040
Token a
Token ID257
Feature activation-3.32e+00

Pos logit contributions
anced+0.119
tie+0.117
dot+0.116
orn+0.114
ka+0.114
Neg logit contributions
 mess-0.085
 neither-0.081
 not-0.071
 both-0.066
 floor-0.064
Token spark
Token ID9009
Feature activation-2.92e+00

Pos logit contributions
ance+0.191
anced+0.189
dot+0.166
ka+0.165
ancing+0.162
Neg logit contributions
 mess-0.106
 not-0.071
 bit-0.070
 neither-0.069
 million-0.066
Tokenly
Token ID306
Feature activation-4.96e+00

Pos logit contributions
dot+0.130
anced+0.128
lynn+0.123
leaf+0.123
 Crown+0.123
Neg logit contributions
 mess-0.099
 neither-0.091
 anymore-0.073
 floor-0.072
 worse-0.072
Token wand
Token ID11569
Feature activation-2.42e+00

Pos logit contributions
ance+0.206
anced+0.200
leaf+0.188
lynn+0.185
ully+0.181
Neg logit contributions
 mess-0.208
 floor-0.171
 bit-0.164
 longer-0.153
 noise-0.152
Token,
Token ID11
Feature activation-1.28e+00

Pos logit contributions
ka+0.121
dot+0.118
ance+0.115
leaf+0.115
yo+0.115
Neg logit contributions
 neither-0.074
 mess-0.068
 fighting-0.057
 roughly-0.056
 wasteful-0.054
Token which
Token ID543
Feature activation-9.37e-01

Pos logit contributions
ka+0.062
ance+0.060
dot+0.058
anger+0.057
tie+0.057
Neg logit contributions
 neither-0.042
 mess-0.042
 wasteful-0.036
 fighting-0.034
 ruining-0.033
Token she
Token ID673
Feature activation-1.85e+00

Pos logit contributions
ka+0.041
dot+0.039
tie+0.039
shirt+0.038
ance+0.038
Neg logit contributions
 neither-0.036
 mess-0.035
 wasteful-0.031
 fighting-0.029
 overflow-0.029
Token could
Token ID714
Feature activation-1.39e+00

Pos logit contributions
dot+0.080
ka+0.080
ance+0.079
anger+0.078
 aboard+0.078
Neg logit contributions
 mess-0.063
 neither-0.062
 forgot-0.055
 wasteful-0.053
 roughly-0.051
Token bedroom
Token ID14043
Feature activation+1.12e+00

Pos logit contributions
leaf+0.174
ance+0.158
reetings+0.155
tie+0.154
 aboard+0.150
Neg logit contributions
 mess-0.110
 floor-0.101
 messy-0.085
 beds-0.083
 room-0.080
Token and
Token ID290
Feature activation-1.04e+00

Pos logit contributions
 neither+0.050
 wasteful+0.049
 mess+0.047
 overflow+0.044
 mosquitoes+0.044
Neg logit contributions
 name-0.038
tie-0.037
ance-0.036
ka-0.036
 gown-0.035
Token put
Token ID1234
Feature activation-3.33e+00

Pos logit contributions
ka+0.040
ance+0.038
tie+0.038
anced+0.037
dot+0.037
Neg logit contributions
 neither-0.044
 mess-0.044
 wasteful-0.039
 fighting-0.037
 ruining-0.037
Token on
Token ID319
Feature activation-2.56e+00

Pos logit contributions
reetings+0.207
leaf+0.202
anced+0.197
found+0.197
chery+0.197
Neg logit contributions
 both-0.065
 mess-0.056
 neither-0.053
 Lily-0.052
 they-0.046
Token the
Token ID262
Feature activation-4.36e+00

Pos logit contributions
reetings+0.151
leaf+0.145
orn+0.144
ane+0.143
anced+0.141
Neg logit contributions
 mess-0.063
 both-0.059
 not-0.052
 floor-0.049
 neither-0.046
Token pretty
Token ID2495
Feature activation-4.93e+00

Pos logit contributions
anced+0.217
ully+0.216
enna+0.214
igh+0.212
year+0.208
Neg logit contributions
 mess-0.157
 floor-0.157
 noise-0.120
 ground-0.119
 floors-0.115
Token buckle
Token ID44983
Feature activation-1.04e+00

Pos logit contributions
leaf+0.249
year+0.246
ance+0.232
anced+0.230
uttered+0.226
Neg logit contributions
 mess-0.186
 floor-0.175
 table-0.132
 floors-0.130
 windows-0.121
Token.
Token ID13
Feature activation-1.01e+00

Pos logit contributions
ka+0.043
leaf+0.041
dot+0.041
 name+0.040
ance+0.040
Neg logit contributions
 neither-0.041
 mess-0.039
 wasteful-0.035
 roughly-0.034
 fighting-0.034
Token
Token ID220
Feature activation-7.86e-01

Pos logit contributions
ka+0.048
leaf+0.046
tie+0.046
ance+0.046
 aboard+0.045
Neg logit contributions
 neither-0.033
 mess-0.032
 wasteful-0.028
 fighting-0.027
 roughly-0.027
Token
Token ID198
Feature activation-2.98e-01

Pos logit contributions
ka+0.035
tie+0.034
dot+0.033
leaf+0.033
ance+0.033
Neg logit contributions
 neither-0.028
 mess-0.027
 wasteful-0.025
 fighting-0.023
 roughly-0.023
Token
Token ID198
Feature activation+3.38e-01

Pos logit contributions
ka+0.012
tie+0.012
ance+0.012
 name+0.012
dot+0.011
Neg logit contributions
 neither-0.012
 mess-0.011
 wasteful-0.011
 fighting-0.010
 overflow-0.010
Token green
Token ID4077
Feature activation-1.98e+00

Pos logit contributions
ance+0.068
anced+0.065
ka+0.063
tie+0.060
 aboard+0.060
Neg logit contributions
 mess-0.061
 neither-0.055
 wasteful-0.052
 floor-0.050
 ruining-0.049
Token dress
Token ID6576
Feature activation-1.10e+00

Pos logit contributions
leaf+0.082
anced+0.081
ance+0.080
dot+0.078
tie+0.076
Neg logit contributions
 mess-0.080
 floor-0.072
 neither-0.068
 table-0.061
 overflow-0.061
Token with
Token ID351
Feature activation-2.35e+00

Pos logit contributions
ka+0.052
leaf+0.052
dot+0.051
anced+0.051
ance+0.050
Neg logit contributions
 neither-0.035
 mess-0.035
 wasteful-0.029
 floor-0.028
 overflow-0.028
Token a
Token ID257
Feature activation-1.98e+00

Pos logit contributions
int+0.098
orn+0.096
tie+0.096
anger+0.096
 aboard+0.095
Neg logit contributions
 mess-0.083
 neither-0.079
 floor-0.073
 mice-0.072
 rats-0.071
Token spark
Token ID9009
Feature activation-2.47e+00

Pos logit contributions
ance+0.084
anced+0.079
 aboard+0.074
leaf+0.074
ka+0.072
Neg logit contributions
 mess-0.089
 floor-0.074
 neither-0.073
 wasteful-0.067
 fighting-0.067
Tokenly
Token ID306
Feature activation-4.91e+00

Pos logit contributions
anced+0.111
ance+0.108
leaf+0.107
lynn+0.103
reetings+0.102
Neg logit contributions
 mess-0.093
 neither-0.077
 floor-0.076
 anymore-0.068
 fighting-0.065
Token belt
Token ID10999
Feature activation-2.60e+00

Pos logit contributions
ance+0.218
leaf+0.205
anced+0.198
year+0.189
found+0.186
Neg logit contributions
 mess-0.212
 floor-0.199
 table-0.161
 noise-0.152
 bit-0.143
Token.
Token ID13
Feature activation-1.46e+00

Pos logit contributions
ka+0.124
leaf+0.120
anced+0.120
Aid+0.119
ance+0.116
Neg logit contributions
 mess-0.081
 neither-0.081
 floor-0.074
 fighting-0.070
 not-0.068
Token She
Token ID1375
Feature activation-1.42e+00

Pos logit contributions
ka+0.072
leaf+0.072
tie+0.071
 powers+0.070
 amulet+0.069
Neg logit contributions
 mess-0.045
 neither-0.044
 floor-0.037
 wasteful-0.035
 roughly-0.035
Token couldn
Token ID3521
Feature activation-1.14e+00

Pos logit contributions
ance+0.061
 aboard+0.061
leaf+0.061
tie+0.060
anger+0.060
Neg logit contributions
 neither-0.050
 mess-0.050
 forgot-0.042
 floor-0.041
 wasteful-0.041
Token't
Token ID470
Feature activation-1.26e+00

Pos logit contributions
ka+0.066
tie+0.065
shirt+0.063
leaf+0.063
 name+0.062
Neg logit contributions
 neither-0.026
 mess-0.024
 wasteful-0.019
 fighting-0.018
 overflow-0.018
Token she
Token ID673
Feature activation-1.13e+00

Pos logit contributions
 neither+0.005
 mess+0.005
 wasteful+0.005
 overflow+0.004
 ruining+0.004
Neg logit contributions
ka-0.005
 name-0.005
tie-0.005
ance-0.005
dot-0.005
Token liked
Token ID8288
Feature activation+2.57e-02

Pos logit contributions
ka+0.042
tie+0.041
ance+0.040
dot+0.040
 aboard+0.040
Neg logit contributions
 mess-0.047
 neither-0.047
 wasteful-0.043
 roughly-0.040
 fighting-0.040
Token to
Token ID284
Feature activation-1.33e+00

Pos logit contributions
 neither+0.001
 mess+0.001
 wasteful+0.001
 overflow+0.001
 fighting+0.001
Neg logit contributions
ka-0.001
tie-0.001
 name-0.001
ance-0.001
dot-0.001
Token wear
Token ID5806
Feature activation-2.98e+00

Pos logit contributions
ka+0.056
tie+0.053
ance+0.052
 aboard+0.052
leaf+0.051
Neg logit contributions
 mess-0.053
 neither-0.050
 overflow-0.045
 wasteful-0.045
 quarrel-0.045
Token her
Token ID607
Feature activation-3.50e+00

Pos logit contributions
anced+0.167
leaf+0.163
ance+0.160
reetings+0.158
int+0.157
Neg logit contributions
 mess-0.078
 both-0.063
 not-0.058
 they-0.058
 floor-0.056
Token special
Token ID2041
Feature activation-4.90e+00

Pos logit contributions
ance+0.187
year+0.183
leaf+0.180
anced+0.178
reetings+0.176
Neg logit contributions
 mess-0.105
 not-0.089
 floor-0.085
 neither-0.083
 messy-0.078
Token dress
Token ID6576
Feature activation-1.93e+00

Pos logit contributions
ance+0.223
year+0.214
anced+0.213
leaf+0.212
aved+0.207
Neg logit contributions
 mess-0.196
 floor-0.167
 quarrel-0.147
 table-0.145
 fight-0.137
Token.
Token ID13
Feature activation-1.01e+00

Pos logit contributions
leaf+0.097
dot+0.093
ka+0.092
anced+0.089
 aboard+0.088
Neg logit contributions
 neither-0.061
 mess-0.058
 floor-0.049
 roughly-0.046
 not-0.046
Token It
Token ID632
Feature activation-7.05e-01

Pos logit contributions
ka+0.048
tie+0.046
leaf+0.046
ance+0.045
 aboard+0.045
Neg logit contributions
 neither-0.033
 mess-0.030
 wasteful-0.028
 fighting-0.027
 roughly-0.026
Token was
Token ID373
Feature activation-1.75e+00

Pos logit contributions
ka+0.029
tie+0.029
leaf+0.028
dot+0.028
ance+0.028
Neg logit contributions
 neither-0.028
 mess-0.027
 wasteful-0.024
 overflow-0.023
 fighting-0.022
Token purple
Token ID14032
Feature activation-1.31e+00

Pos logit contributions
ance+0.080
leaf+0.080
ka+0.078
anger+0.078
tie+0.077
Neg logit contributions
 mess-0.061
 neither-0.060
 wasteful-0.053
 not-0.053
 ruining-0.050
Token unique
Token ID3748
Feature activation-1.30e+00

Pos logit contributions
ance+0.034
tie+0.034
ka+0.033
anced+0.033
 aboard+0.032
Neg logit contributions
 mess-0.025
 neither-0.024
 wasteful-0.023
 fighting-0.020
 overflow-0.020
Token girl
Token ID2576
Feature activation-9.31e-01

Pos logit contributions
anced+0.064
ance+0.063
ka+0.061
tie+0.061
dot+0.060
Neg logit contributions
 mess-0.045
 neither-0.039
 wasteful-0.036
 floor-0.034
 fighting-0.034
Token with
Token ID351
Feature activation-1.60e+00

Pos logit contributions
ka+0.045
anced+0.045
ance+0.044
tie+0.043
leaf+0.043
Neg logit contributions
 neither-0.030
 mess-0.028
 wasteful-0.026
 fighting-0.025
 roughly-0.024
Token a
Token ID257
Feature activation-9.62e-01

Pos logit contributions
 aboard+0.077
ka+0.073
tie+0.072
anger+0.071
int+0.071
Neg logit contributions
 neither-0.052
 mess-0.051
 wasteful-0.047
 fighting-0.045
 mice-0.045
Token spark
Token ID9009
Feature activation-2.12e+00

Pos logit contributions
ance+0.043
ka+0.041
anced+0.041
 aboard+0.040
tie+0.040
Neg logit contributions
 mess-0.038
 neither-0.035
 wasteful-0.032
 fighting-0.030
 overflow-0.029
Tokenly
Token ID306
Feature activation-4.90e+00

Pos logit contributions
anced+0.089
ance+0.087
dot+0.087
leaf+0.086
lynn+0.084
Neg logit contributions
 mess-0.083
 neither-0.077
 floor-0.066
 fighting-0.064
 wasteful-0.064
Token smile
Token ID8212
Feature activation-2.80e+00

Pos logit contributions
leaf+0.217
ance+0.217
anced+0.198
iddled+0.193
ancing+0.190
Neg logit contributions
 mess-0.209
 floor-0.179
 noise-0.148
 hole-0.146
 table-0.143
Token and
Token ID290
Feature activation-1.67e+00

Pos logit contributions
leaf+0.138
ance+0.121
lynn+0.121
dot+0.120
 amulet+0.120
Neg logit contributions
 neither-0.089
 mess-0.087
 fighting-0.078
 floor-0.074
 anymore-0.071
Token bright
Token ID6016
Feature activation-3.10e+00

Pos logit contributions
ka+0.071
anger+0.071
tie+0.069
 aboard+0.069
orn+0.069
Neg logit contributions
 neither-0.061
 mess-0.061
 not-0.052
 wasteful-0.051
 fighting-0.051
Token eyes
Token ID2951
Feature activation-3.05e+00

Pos logit contributions
ance+0.161
leaf+0.156
chery+0.155
cher+0.155
lynn+0.152
Neg logit contributions
 mess-0.103
 floor-0.081
 noise-0.069
 fighting-0.067
 faces-0.066
Token that
Token ID326
Feature activation-1.71e+00

Pos logit contributions
lynn+0.182
leaf+0.180
 amulet+0.178
yo+0.178
ance+0.177
Neg logit contributions
 mess-0.061
 neither-0.054
 fighting-0.053
 anymore-0.047
 floor-0.041
Token yellow
Token ID7872
Feature activation-2.89e+00

Pos logit contributions
ance+0.068
anced+0.066
ka+0.061
tie+0.060
leaf+0.060
Neg logit contributions
 mess-0.054
 neither-0.047
 wasteful-0.045
 floor-0.042
 fighting-0.041
Token dress
Token ID6576
Feature activation-8.72e-01

Pos logit contributions
anced+0.113
leaf+0.112
ance+0.110
iddled+0.102
aved+0.101
Neg logit contributions
 mess-0.130
 floor-0.116
 neither-0.111
 table-0.105
 longer-0.100
Token with
Token ID351
Feature activation-2.52e+00

Pos logit contributions
ka+0.041
leaf+0.040
anced+0.040
dot+0.040
ance+0.040
Neg logit contributions
 neither-0.028
 mess-0.027
 wasteful-0.023
 overflow-0.022
 floor-0.022
Token a
Token ID257
Feature activation-2.21e+00

Pos logit contributions
int+0.115
anger+0.110
Aid+0.110
lynn+0.109
 aboard+0.109
Neg logit contributions
 mess-0.081
 neither-0.077
 floor-0.072
 mice-0.072
 both-0.070
Token spark
Token ID9009
Feature activation-1.96e+00

Pos logit contributions
ance+0.096
anced+0.091
leaf+0.086
 aboard+0.084
ka+0.082
Neg logit contributions
 mess-0.097
 floor-0.081
 neither-0.079
 wasteful-0.074
 fighting-0.073
Tokenly
Token ID306
Feature activation-4.89e+00

Pos logit contributions
anced+0.085
ance+0.082
leaf+0.082
dot+0.080
lynn+0.077
Neg logit contributions
 mess-0.076
 neither-0.068
 floor-0.063
 fighting-0.058
 wasteful-0.057
Token cape
Token ID36023
Feature activation-2.03e+00

Pos logit contributions
ance+0.214
leaf+0.209
anced+0.202
year+0.189
aved+0.185
Neg logit contributions
 mess-0.212
 floor-0.203
 table-0.164
 noise-0.156
 ground-0.144
Token.
Token ID13
Feature activation+2.98e-01

Pos logit contributions
ka+0.099
leaf+0.096
Aid+0.096
dot+0.093
shirt+0.092
Neg logit contributions
 neither-0.065
 mess-0.062
 fighting-0.056
 floor-0.053
 ruining-0.053
Token
Token ID198
Feature activation-1.94e-01

Pos logit contributions
 neither+0.012
 mess+0.012
 wasteful+0.012
 overflow+0.011
 ruining+0.011
Neg logit contributions
 name-0.012
ka-0.011
tie-0.011
dot-0.011
shirt-0.011
Token
Token ID198
Feature activation+2.53e-01

Pos logit contributions
ka+0.008
tie+0.008
 name+0.007
ance+0.007
dot+0.007
Neg logit contributions
 neither-0.008
 mess-0.008
 wasteful-0.007
 fighting-0.006
 overflow-0.006
TokenOne
Token ID3198
Feature activation-9.23e-01

Pos logit contributions
 neither+0.009
 mess+0.009
 wasteful+0.008
 overflow+0.008
 ruining+0.008
Neg logit contributions
ka-0.011
tie-0.011
 name-0.011
dot-0.011
shirt-0.011
Token feeling
Token ID4203
Feature activation-1.29e+00

Pos logit contributions
 neither+0.013
 mess+0.013
 wasteful+0.013
 overflow+0.011
 mosquitoes+0.011
Neg logit contributions
ka-0.013
 name-0.013
tie-0.012
ance-0.012
dot-0.012
Token very
Token ID845
Feature activation-1.75e-01

Pos logit contributions
ka+0.070
anger+0.069
tie+0.069
anced+0.069
ance+0.068
Neg logit contributions
 neither-0.034
 mess-0.033
 wasteful-0.031
 roughly-0.026
 overflow-0.025
Token proud
Token ID6613
Feature activation-6.75e-01

Pos logit contributions
ka+0.008
tie+0.008
 name+0.008
ance+0.008
dot+0.007
Neg logit contributions
 neither-0.006
 mess-0.006
 wasteful-0.006
 overflow-0.005
 fighting-0.005
Token wearing
Token ID5762
Feature activation-2.41e+00

Pos logit contributions
ka+0.029
tie+0.028
dot+0.028
 name+0.028
ance+0.028
Neg logit contributions
 neither-0.025
 mess-0.025
 wasteful-0.022
 fighting-0.021
 roughly-0.021
Token her
Token ID607
Feature activation-3.41e+00

Pos logit contributions
anced+0.124
ka+0.122
leaf+0.121
dot+0.121
ance+0.121
Neg logit contributions
 mess-0.072
 not-0.059
 roughly-0.057
 neither-0.057
 floor-0.055
Token special
Token ID2041
Feature activation-4.89e+00

Pos logit contributions
ance+0.173
anced+0.164
leaf+0.155
dot+0.151
aved+0.149
Neg logit contributions
 mess-0.125
 not-0.100
 floor-0.092
 quarrel-0.090
 neither-0.089
Token suit
Token ID6050
Feature activation-1.67e+00

Pos logit contributions
ance+0.211
anced+0.191
leaf+0.182
aved+0.173
tie+0.170
Neg logit contributions
 mess-0.234
 floor-0.185
 mistake-0.185
 fuss-0.176
 quarrel-0.174
Token.
Token ID13
Feature activation-1.32e+00

Pos logit contributions
ka+0.077
leaf+0.076
dot+0.075
ance+0.072
 name+0.072
Neg logit contributions
 neither-0.056
 mess-0.055
 wasteful-0.046
 roughly-0.045
 fighting-0.045
Token She
Token ID1375
Feature activation-1.27e+00

Pos logit contributions
ka+0.068
tie+0.065
leaf+0.063
 powers+0.062
dot+0.062
Neg logit contributions
 neither-0.040
 mess-0.035
 wasteful-0.033
 roughly-0.032
 fighting-0.031
Token will
Token ID481
Feature activation-1.58e+00

Pos logit contributions
ka+0.059
tie+0.057
dot+0.057
leaf+0.056
anger+0.056
Neg logit contributions
 mess-0.044
 neither-0.042
 wasteful-0.037
 forgot-0.036
 roughly-0.034
Token never
Token ID1239
Feature activation+5.53e-01

Pos logit contributions
ka+0.092
shirt+0.089
dot+0.088
leaf+0.087
tie+0.087
Neg logit contributions
 neither-0.037
 mess-0.035
 not-0.030
 overflow-0.027
 wasteful-0.026
Token to
Token ID284
Feature activation-1.49e+00

Pos logit contributions
 neither+0.004
 mess+0.004
 wasteful+0.003
 overflow+0.003
 ruining+0.003
Neg logit contributions
ka-0.003
tie-0.003
 name-0.003
dot-0.003
ance-0.003
Token her
Token ID607
Feature activation-3.54e-01

Pos logit contributions
tie+0.073
ka+0.072
ance+0.071
shirt+0.069
anger+0.069
Neg logit contributions
 mess-0.049
 neither-0.048
 fighting-0.046
 wasteful-0.043
 ruining-0.041
Token and
Token ID290
Feature activation-8.15e-03

Pos logit contributions
ka+0.016
tie+0.015
ance+0.015
dot+0.015
leaf+0.015
Neg logit contributions
 mess-0.013
 neither-0.013
 wasteful-0.012
 fighting-0.011
 ruining-0.010
Token put
Token ID1234
Feature activation-2.19e+00

Pos logit contributions
ka+0.000
tie+0.000
 name+0.000
ance+0.000
dot+0.000
Neg logit contributions
 neither-0.000
 mess-0.000
 wasteful-0.000
 overflow-0.000
 fighting-0.000
Token a
Token ID257
Feature activation-2.28e+00

Pos logit contributions
dot+0.116
leaf+0.113
 amulet+0.113
tie+0.112
ka+0.109
Neg logit contributions
 neither-0.063
 mess-0.055
 wasteful-0.050
 fighting-0.048
 both-0.046
Token comforting
Token ID34228
Feature activation-4.89e+00

Pos logit contributions
ance+0.112
anced+0.107
tie+0.096
dot+0.096
ka+0.092
Neg logit contributions
 mess-0.099
 wasteful-0.077
 neither-0.074
 floor-0.073
 overflow-0.073
Token hand
Token ID1021
Feature activation-2.45e+00

Pos logit contributions
ance+0.225
anced+0.212
ane+0.200
amb+0.190
tie+0.189
Neg logit contributions
 mess-0.204
 floor-0.169
 noise-0.159
 table-0.151
 fuss-0.147
Token on
Token ID319
Feature activation-6.16e-01

Pos logit contributions
ka+0.114
 Crown+0.112
leaf+0.110
ance+0.109
 amulet+0.109
Neg logit contributions
 neither-0.082
 mess-0.072
 fighting-0.066
 roughly-0.065
 wasteful-0.063
Token her
Token ID607
Feature activation-2.06e+00

Pos logit contributions
ka+0.027
tie+0.026
ance+0.026
dot+0.026
shirt+0.025
Neg logit contributions
 neither-0.023
 mess-0.022
 wasteful-0.020
 fighting-0.019
 ruining-0.019
Token shoulder
Token ID8163
Feature activation-3.21e-01

Pos logit contributions
orn+0.106
leaf+0.105
anced+0.105
ka+0.104
ance+0.104
Neg logit contributions
 mess-0.068
 floor-0.060
 neither-0.050
 mistake-0.048
 not-0.047
Token.
Token ID13
Feature activation+1.78e-01

Pos logit contributions
ka+0.013
tie+0.012
 name+0.012
ance+0.012
dot+0.012
Neg logit contributions
 neither-0.013
 mess-0.013
 wasteful-0.012
 overflow-0.011
 fighting-0.011
Token had
Token ID550
Feature activation-1.30e+00

Pos logit contributions
dot+0.077
ka+0.076
ance+0.076
tie+0.075
leaf+0.074
Neg logit contributions
 mess-0.057
 neither-0.056
 fighting-0.050
 wasteful-0.048
 ruining-0.044
Token just
Token ID655
Feature activation-2.26e+00

Pos logit contributions
ka+0.052
tie+0.051
ance+0.051
dot+0.050
anger+0.050
Neg logit contributions
 mess-0.052
 neither-0.051
 wasteful-0.046
 fighting-0.045
 ruining-0.043
Token given
Token ID1813
Feature activation-1.50e+00

Pos logit contributions
anger+0.100
ance+0.092
lor+0.092
orn+0.091
 name+0.090
Neg logit contributions
 mess-0.084
 neither-0.084
 fighting-0.080
 wasteful-0.075
 ruining-0.072
Token him
Token ID683
Feature activation-2.40e+00

Pos logit contributions
tie+0.062
ka+0.061
dot+0.061
anger+0.061
anced+0.061
Neg logit contributions
 neither-0.056
 mess-0.056
 wasteful-0.049
 fighting-0.048
 pies-0.044
Token a
Token ID257
Feature activation-3.27e+00

Pos logit contributions
ka+0.108
anced+0.106
orn+0.103
dot+0.103
anger+0.103
Neg logit contributions
 neither-0.085
 mess-0.075
 wasteful-0.073
 not-0.067
 roughly-0.066
Token shiny
Token ID22441
Feature activation-4.84e+00

Pos logit contributions
ance+0.178
anced+0.174
ka+0.153
ed+0.149
dot+0.148
Neg logit contributions
 mess-0.119
 quarrel-0.091
 fight-0.087
 neither-0.086
 mistake-0.085
Token new
Token ID649
Feature activation-4.08e+00

Pos logit contributions
ance+0.231
anced+0.227
leaf+0.218
dot+0.204
lor+0.202
Neg logit contributions
 mess-0.203
 floor-0.153
 mistake-0.152
 longer-0.137
 table-0.137
Token pistol
Token ID16790
Feature activation-1.98e+00

Pos logit contributions
ance+0.211
anced+0.199
leaf+0.181
year+0.181
aved+0.180
Neg logit contributions
 mess-0.155
 floor-0.120
 mistake-0.120
 quarrel-0.115
 fight-0.111
Token.
Token ID13
Feature activation-4.64e-01

Pos logit contributions
ka+0.094
dot+0.090
leaf+0.089
anced+0.089
ance+0.087
Neg logit contributions
 neither-0.067
 mess-0.061
 wasteful-0.054
 fighting-0.054
 roughly-0.053
Token Joey
Token ID26154
Feature activation-1.04e+00

Pos logit contributions
ka+0.018
tie+0.017
ance+0.017
dot+0.017
leaf+0.017
Neg logit contributions
 neither-0.019
 mess-0.019
 wasteful-0.018
 fighting-0.017
 overflow-0.016
Token had
Token ID550
Feature activation-2.18e+00

Pos logit contributions
ance+0.045
ka+0.044
tie+0.044
leaf+0.044
dot+0.044
Neg logit contributions
 mess-0.037
 neither-0.036
 wasteful-0.032
 fighting-0.031
 roughly-0.030
Token she
Token ID673
Feature activation-8.72e-01

Pos logit contributions
leaf+0.056
tie+0.054
ka+0.054
dot+0.053
orn+0.052
Neg logit contributions
 neither-0.048
 mess-0.046
 wasteful-0.039
 fighting-0.039
 roughly-0.037
Token put
Token ID1234
Feature activation-3.33e+00

Pos logit contributions
leaf+0.036
tie+0.035
ka+0.035
ance+0.035
 name+0.034
Neg logit contributions
 mess-0.034
 neither-0.034
 wasteful-0.029
 fighting-0.028
 floor-0.028
Token on
Token ID319
Feature activation-1.88e+00

Pos logit contributions
reetings+0.215
leaf+0.214
found+0.206
chery+0.201
int+0.198
Neg logit contributions
 both-0.065
 neither-0.055
 Lily-0.048
 they-0.047
 mess-0.045
Token her
Token ID607
Feature activation-3.61e+00

Pos logit contributions
reetings+0.099
orn+0.098
leaf+0.098
anced+0.095
tie+0.094
Neg logit contributions
 mess-0.058
 neither-0.049
 floor-0.047
 both-0.047
 not-0.046
Token spark
Token ID9009
Feature activation-2.81e+00

Pos logit contributions
leaf+0.177
ance+0.171
orn+0.170
year+0.167
aved+0.165
Neg logit contributions
 mess-0.130
 floor-0.115
 not-0.107
 messy-0.098
 neither-0.096
Tokenly
Token ID306
Feature activation-4.82e+00

Pos logit contributions
leaf+0.130
anced+0.118
 Magic+0.117
dot+0.117
 name+0.115
Neg logit contributions
 mess-0.102
 floor-0.087
 neither-0.083
 anymore-0.073
 roughly-0.068
Token shoes
Token ID10012
Feature activation-1.71e+00

Pos logit contributions
leaf+0.258
ance+0.234
year+0.230
anced+0.223
 aboard+0.221
Neg logit contributions
 mess-0.170
 floor-0.159
 table-0.121
 messy-0.112
 floors-0.107
Token.
Token ID13
Feature activation-1.91e+00

Pos logit contributions
leaf+0.084
ka+0.081
 name+0.081
anced+0.080
dot+0.080
Neg logit contributions
 neither-0.055
 mess-0.051
 roughly-0.043
 fighting-0.042
 floor-0.042
Token
Token ID198
Feature activation-6.11e-01

Pos logit contributions
ka+0.100
leaf+0.099
 powers+0.098
tie+0.098
 amulet+0.096
Neg logit contributions
 mess-0.051
 neither-0.050
 floor-0.044
 roughly-0.042
 fighting-0.041
Token
Token ID198
Feature activation-2.06e-01

Pos logit contributions
ka+0.026
tie+0.026
ance+0.025
dot+0.025
anced+0.025
Neg logit contributions
 neither-0.022
 mess-0.022
 wasteful-0.020
 roughly-0.018
 fighting-0.018
TokenShe
Token ID3347
Feature activation-1.27e+00

Pos logit contributions
ka+0.010
tie+0.010
 name+0.010
ance+0.010
dot+0.010
Neg logit contributions
 neither-0.006
 mess-0.006
 wasteful-0.006
 overflow-0.005
 fighting-0.005
Token mom
Token ID1995
Feature activation+5.22e-01

Pos logit contributions
leaf+0.103
tie+0.098
 aboard+0.098
ka+0.097
 amulet+0.097
Neg logit contributions
 neither-0.033
 mess-0.032
 fighting-0.021
 not-0.021
 floor-0.020
Tokenmy
Token ID1820
Feature activation+1.44e-01

Pos logit contributions
 neither+0.023
 mess+0.022
 wasteful+0.022
 overflow+0.020
 ruining+0.020
Neg logit contributions
ka-0.018
 name-0.018
tie-0.018
 aboard-0.017
ance-0.017
Token gave
Token ID2921
Feature activation-1.58e-01

Pos logit contributions
 neither+0.006
 mess+0.006
 wasteful+0.005
 overflow+0.005
 ruining+0.005
Neg logit contributions
ka-0.006
tie-0.006
 name-0.006
ance-0.005
dot-0.005
Token her
Token ID607
Feature activation-2.48e+00

Pos logit contributions
ka+0.005
tie+0.005
 name+0.005
ance+0.005
dot+0.005
Neg logit contributions
 neither-0.007
 mess-0.007
 wasteful-0.007
 fighting-0.006
 overflow-0.006
Token a
Token ID257
Feature activation-3.96e+00

Pos logit contributions
anced+0.126
reetings+0.126
orn+0.122
 amulet+0.121
ance+0.120
Neg logit contributions
 neither-0.070
 mess-0.067
 wasteful-0.056
 not-0.054
 both-0.054
Token cool
Token ID3608
Feature activation-4.82e+00

Pos logit contributions
ance+0.247
anced+0.242
ancing+0.206
ed+0.203
aved+0.202
Neg logit contributions
 mess-0.129
 messy-0.088
 table-0.083
 quarrel-0.081
 fight-0.078
Token mask
Token ID9335
Feature activation-2.15e+00

Pos logit contributions
ance+0.234
reetings+0.232
anced+0.230
year+0.226
lor+0.224
Neg logit contributions
 mess-0.195
 table-0.142
 floor-0.142
 bit-0.127
 mistake-0.122
Token to
Token ID284
Feature activation-2.14e+00

Pos logit contributions
ka+0.100
dot+0.094
ance+0.092
anced+0.092
 amulet+0.092
Neg logit contributions
 neither-0.073
 mess-0.069
 roughly-0.060
 fighting-0.058
 wasteful-0.057
Token wear
Token ID5806
Feature activation-2.60e+00

Pos logit contributions
ka+0.084
 aboard+0.083
anced+0.083
dot+0.081
ance+0.081
Neg logit contributions
 mess-0.087
 neither-0.079
 wasteful-0.075
 not-0.075
 overflow-0.074
Token.
Token ID13
Feature activation+3.25e-01

Pos logit contributions
leaf+0.138
anced+0.138
dot+0.135
 amulet+0.133
int+0.132
Neg logit contributions
 mess-0.068
 not-0.059
 neither-0.058
 both-0.053
 floor-0.053
Token It
Token ID632
Feature activation-6.10e-01

Pos logit contributions
 neither+0.012
 mess+0.011
 wasteful+0.011
 ruining+0.010
 overflow+0.010
Neg logit contributions
 name-0.014
ka-0.014
tie-0.014
dot-0.013
shirt-0.013
Token she
Token ID673
Feature activation-2.83e-01

Pos logit contributions
 mess+0.030
 neither+0.029
 wasteful+0.028
 overflow+0.025
 ruining+0.025
Neg logit contributions
 name-0.033
ka-0.033
tie-0.032
dot-0.031
shirt-0.031
Token found
Token ID1043
Feature activation-1.13e+00

Pos logit contributions
ka+0.012
tie+0.011
 name+0.011
ance+0.011
leaf+0.011
Neg logit contributions
 neither-0.011
 mess-0.011
 wasteful-0.010
 fighting-0.009
 overflow-0.009
Token a
Token ID257
Feature activation-3.03e-01

Pos logit contributions
ka+0.046
tie+0.044
anced+0.044
 aboard+0.044
dot+0.043
Neg logit contributions
 neither-0.045
 mess-0.041
 wasteful-0.040
 rats-0.039
 mice-0.037
Token shiny
Token ID22441
Feature activation-3.16e+00

Pos logit contributions
ka+0.012
ance+0.012
tie+0.012
dot+0.012
 name+0.011
Neg logit contributions
 mess-0.012
 neither-0.012
 wasteful-0.011
 overflow-0.010
 fighting-0.010
Token pe
Token ID613
Feature activation-3.85e+00

Pos logit contributions
ance+0.139
anced+0.136
leaf+0.132
dot+0.124
found+0.124
Neg logit contributions
 mess-0.145
 floor-0.116
 neither-0.109
 wasteful-0.104
 fighting-0.103
Tokenb
Token ID65
Feature activation-4.76e+00

Pos logit contributions
anced+0.164
leaf+0.149
ance+0.143
 Music+0.135
 farewell+0.134
Neg logit contributions
 mess-0.151
 neither-0.127
 not-0.126
 wasteful-0.121
 floor-0.117
Tokenble
Token ID903
Feature activation-1.12e+00

Pos logit contributions
leaf+0.286
 outer+0.263
shirt+0.262
chery+0.260
 farewell+0.259
Neg logit contributions
 neither-0.101
 not-0.088
 mess-0.053
 fighting-0.048
 both-0.045
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
ka+0.046
anced+0.044
ance+0.044
 name+0.043
leaf+0.043
Neg logit contributions
 neither-0.045
 mess-0.044
 wasteful-0.039
 fighting-0.039
 roughly-0.038
Token picked
Token ID6497
Feature activation-3.14e+00

Pos logit contributions
ance+0.043
ka+0.041
tie+0.041
 aboard+0.040
anced+0.040
Neg logit contributions
 mess-0.039
 neither-0.038
 wasteful-0.036
 roughly-0.032
 fighting-0.031
Token it
Token ID340
Feature activation-1.10e+00

Pos logit contributions
anced+0.197
chery+0.193
ka+0.191
uster+0.190
leaf+0.190
Neg logit contributions
 neither-0.048
 mess-0.045
 not-0.035
 fighting-0.029
 both-0.027
Token up
Token ID510
Feature activation-4.16e-01

Pos logit contributions
leaf+0.047
ka+0.046
ance+0.046
anced+0.045
dot+0.045
Neg logit contributions
 neither-0.043
 mess-0.042
 fighting-0.035
 roughly-0.035
 wasteful-0.035
Token.
Token ID13
Feature activation-5.17e-01

Pos logit contributions
ka+0.168
anced+0.156
leaf+0.155
dot+0.155
 amulet+0.153
Neg logit contributions
 neither-0.085
 mess-0.071
 not-0.068
 they-0.061
 floor-0.060
Token He
Token ID679
Feature activation-2.20e+00

Pos logit contributions
ka+0.021
tie+0.020
ance+0.020
leaf+0.020
dot+0.020
Neg logit contributions
 neither-0.020
 mess-0.020
 wasteful-0.019
 overflow-0.017
 fighting-0.017
Token put
Token ID1234
Feature activation-3.35e+00

Pos logit contributions
ance+0.105
leaf+0.103
tie+0.102
dot+0.100
anger+0.100
Neg logit contributions
 mess-0.071
 neither-0.063
 forgot-0.060
 floor-0.056
 stomp-0.052
Token on
Token ID319
Feature activation-2.50e+00

Pos logit contributions
reetings+0.176
found+0.175
leaf+0.175
lor+0.169
dot+0.166
Neg logit contributions
 neither-0.086
 both-0.086
 mess-0.074
 they-0.073
 fighting-0.064
Token a
Token ID257
Feature activation-3.81e+00

Pos logit contributions
reetings+0.141
leaf+0.139
orn+0.138
ocket+0.137
anced+0.133
Neg logit contributions
 mess-0.066
 floor-0.055
 both-0.054
 neither-0.052
 not-0.052
Token blue
Token ID4171
Feature activation-4.76e+00

Pos logit contributions
ance+0.205
anced+0.198
 aboard+0.164
reetings+0.161
ancing+0.158
Neg logit contributions
 mess-0.155
 floor-0.129
 messy-0.115
 table-0.113
 timer-0.109
Token shirt
Token ID10147
Feature activation-2.46e+00

Pos logit contributions
reetings+0.246
leaf+0.241
ance+0.238
anced+0.235
year+0.234
Neg logit contributions
 floor-0.138
 mess-0.136
 table-0.127
 longer-0.106
 both-0.102
Token,
Token ID11
Feature activation-2.76e+00

Pos logit contributions
leaf+0.142
ka+0.138
anced+0.134
dot+0.133
ance+0.131
Neg logit contributions
 neither-0.053
 mess-0.048
 floor-0.046
 not-0.041
 roughly-0.037
Token blue
Token ID4171
Feature activation-3.81e+00

Pos logit contributions
orn+0.130
ance+0.129
anger+0.128
ane+0.125
reetings+0.123
Neg logit contributions
 mess-0.082
 neither-0.082
 floor-0.078
 not-0.076
 they-0.074
Token pants
Token ID12581
Feature activation-2.43e+00

Pos logit contributions
leaf+0.194
year+0.192
reetings+0.188
ocket+0.180
 amulet+0.176
Neg logit contributions
 floor-0.102
 mess-0.097
 table-0.087
 mouths-0.086
 cups-0.085
Token,
Token ID11
Feature activation-2.52e+00

Pos logit contributions
leaf+0.137
ka+0.132
 amulet+0.132
 name+0.131
orn+0.129
Neg logit contributions
 neither-0.057
 mess-0.052
 floor-0.045
 fighting-0.044
 they-0.039
Token She
Token ID1375
Feature activation-1.64e+00

Pos logit contributions
ka+0.057
tie+0.054
orn+0.052
leaf+0.052
ocket+0.052
Neg logit contributions
 neither-0.035
 mess-0.032
 wasteful-0.030
 roughly-0.029
 floor-0.029
Token picked
Token ID6497
Feature activation-2.55e+00

Pos logit contributions
leaf+0.073
tie+0.073
ka+0.072
dot+0.071
ocket+0.070
Neg logit contributions
 mess-0.055
 neither-0.054
 forgot-0.047
 quarrel-0.045
 fighting-0.045
Token up
Token ID510
Feature activation-1.46e+00

Pos logit contributions
och+0.121
ka+0.121
aved+0.120
ocket+0.120
uster+0.118
Neg logit contributions
 neither-0.069
 mess-0.065
 mosquitoes-0.064
 snakes-0.058
 mice-0.058
Token a
Token ID257
Feature activation-1.76e+00

Pos logit contributions
tie+0.062
ka+0.060
leaf+0.060
dot+0.059
anced+0.059
Neg logit contributions
 neither-0.052
 mess-0.051
 rats-0.047
 mosquitoes-0.045
 mice-0.045
Token shiny
Token ID22441
Feature activation-4.10e+00

Pos logit contributions
ance+0.084
anced+0.083
dot+0.075
ka+0.073
leaf+0.072
Neg logit contributions
 mess-0.071
 neither-0.055
 floor-0.054
 wasteful-0.053
 quarrel-0.051
Token golden
Token ID10861
Feature activation-4.75e+00

Pos logit contributions
anced+0.183
ance+0.181
leaf+0.179
dot+0.164
aved+0.164
Neg logit contributions
 mess-0.181
 floor-0.146
 mistake-0.126
 table-0.124
 tower-0.120
Token sword
Token ID8429
Feature activation-1.41e+00

Pos logit contributions
ance+0.220
anced+0.214
leaf+0.203
aved+0.194
iddled+0.188
Neg logit contributions
 mess-0.204
 floor-0.173
 table-0.150
 tower-0.145
 pie-0.144
Token and
Token ID290
Feature activation-1.17e+00

Pos logit contributions
ka+0.072
leaf+0.070
dot+0.069
shirt+0.067
Aid+0.067
Neg logit contributions
 neither-0.042
 mess-0.040
 fighting-0.034
 floor-0.033
 roughly-0.032
Token went
Token ID1816
Feature activation-3.36e+00

Pos logit contributions
ka+0.049
tie+0.047
dot+0.047
leaf+0.046
ance+0.046
Neg logit contributions
 neither-0.043
 mess-0.042
 wasteful-0.037
 roughly-0.036
 fighting-0.035
Token out
Token ID503
Feature activation-2.09e+00

Pos logit contributions
orn+0.189
ocket+0.186
anced+0.185
igh+0.184
ka+0.181
Neg logit contributions
 not-0.079
 mess-0.072
 after-0.068
 they-0.065
 ruining-0.060
Token into
Token ID656
Feature activation-1.08e+00

Pos logit contributions
tie+0.107
ka+0.106
yo+0.102
orn+0.102
Aid+0.101
Neg logit contributions
 mess-0.058
 neither-0.058
 fighting-0.055
 snakes-0.050
 ruining-0.049
Token fairy
Token ID25607
Feature activation-1.90e-01

Pos logit contributions
tie+0.055
shirt+0.052
anced+0.051
dot+0.049
ance+0.049
Neg logit contributions
 mess-0.067
 neither-0.059
 wasteful-0.057
 fighting-0.056
 floor-0.054
Token gave
Token ID2921
Feature activation-3.85e-01

Pos logit contributions
ka+0.007
tie+0.007
 name+0.007
ance+0.007
dot+0.007
Neg logit contributions
 neither-0.008
 mess-0.008
 wasteful-0.007
 fighting-0.007
 overflow-0.007
Token Lily
Token ID20037
Feature activation-2.12e+00

Pos logit contributions
tie+0.019
ka+0.019
dot+0.019
 name+0.019
leaf+0.019
Neg logit contributions
 neither-0.012
 mess-0.011
 wasteful-0.010
 fighting-0.009
 overflow-0.009
Token a
Token ID257
Feature activation-3.36e+00

Pos logit contributions
anger+0.096
ka+0.095
anced+0.091
leaf+0.091
dot+0.089
Neg logit contributions
 neither-0.070
 mess-0.065
 both-0.057
 not-0.057
 mice-0.056
Token spark
Token ID9009
Feature activation-2.88e+00

Pos logit contributions
ance+0.206
anced+0.199
dot+0.177
ka+0.176
lynn+0.173
Neg logit contributions
 mess-0.097
 not-0.060
 neither-0.058
 overflow-0.057
 wasteful-0.057
Tokenly
Token ID306
Feature activation-4.74e+00

Pos logit contributions
anced+0.143
dot+0.136
ance+0.134
aved+0.133
lynn+0.133
Neg logit contributions
 mess-0.089
 neither-0.075
 floor-0.062
 longer-0.060
 worse-0.060
Token wand
Token ID11569
Feature activation-2.55e+00

Pos logit contributions
ance+0.195
anced+0.182
leaf+0.176
lynn+0.174
aden+0.167
Neg logit contributions
 mess-0.205
 floor-0.172
 bit-0.160
 noise-0.150
 longer-0.149
Token and
Token ID290
Feature activation-4.24e-01

Pos logit contributions
ka+0.144
dot+0.136
ance+0.135
leaf+0.135
yo+0.135
Neg logit contributions
 neither-0.064
 mess-0.057
 not-0.048
 fighting-0.047
 roughly-0.046
Token said
Token ID531
Feature activation-1.40e+00

Pos logit contributions
ka+0.019
tie+0.019
 name+0.018
dot+0.018
ance+0.018
Neg logit contributions
 neither-0.015
 mess-0.015
 wasteful-0.013
 fighting-0.012
 overflow-0.012
Token to
Token ID284
Feature activation-1.46e-01

Pos logit contributions
leaf+0.077
tie+0.076
ka+0.076
shirt+0.076
Aid+0.074
Neg logit contributions
 neither-0.038
 mess-0.031
 wasteful-0.027
 fighting-0.027
 not-0.026
Token make
Token ID787
Feature activation-2.96e-01

Pos logit contributions
ka+0.007
tie+0.007
 name+0.007
dot+0.007
ance+0.007
Neg logit contributions
 neither-0.005
 mess-0.005
 wasteful-0.004
 overflow-0.004
 fighting-0.004
Token her
Token ID607
Feature activation-1.23e+00

Pos logit contributions
ka+0.018
tie+0.018
ance+0.018
leaf+0.018
 name+0.018
Neg logit contributions
 neither-0.013
 mess-0.013
 wasteful-0.012
 fighting-0.011
 roughly-0.010
Token mum
Token ID25682
Feature activation-1.06e+00

Pos logit contributions
tie+0.066
ka+0.066
ance+0.065
dot+0.064
 aboard+0.064
Neg logit contributions
 mess-0.033
 neither-0.031
 fighting-0.025
 wasteful-0.025
 mosquitoes-0.024
Token put
Token ID1234
Feature activation-1.53e+00

Pos logit contributions
ka+0.041
tie+0.040
dot+0.040
ance+0.040
anced+0.039
Neg logit contributions
 neither-0.043
 mess-0.042
 wasteful-0.039
 fighting-0.036
 overflow-0.035
Token on
Token ID319
Feature activation-2.73e-01

Pos logit contributions
dot+0.059
leaf+0.059
ance+0.059
anced+0.058
 amulet+0.058
Neg logit contributions
 neither-0.061
 mess-0.057
 wasteful-0.053
 raw-0.051
 fighting-0.050
Token her
Token ID607
Feature activation-3.07e+00

Pos logit contributions
tie+0.010
ka+0.010
 name+0.010
ance+0.010
dot+0.010
Neg logit contributions
 neither-0.011
 mess-0.011
 wasteful-0.011
 fighting-0.010
 overflow-0.010
Token special
Token ID2041
Feature activation-4.73e+00

Pos logit contributions
ance+0.146
anced+0.139
ka+0.135
tie+0.134
leaf+0.131
Neg logit contributions
 mess-0.105
 neither-0.093
 floor-0.091
 quarrel-0.086
 raw-0.086
Token clothes
Token ID8242
Feature activation-1.17e+00

Pos logit contributions
ance+0.218
anced+0.208
aved+0.199
tie+0.197
leaf+0.195
Neg logit contributions
 mess-0.183
 floor-0.160
 quarrel-0.150
 table-0.146
 fight-0.144
Token and
Token ID290
Feature activation-7.61e-01

Pos logit contributions
ka+0.056
leaf+0.055
dot+0.054
 name+0.053
ance+0.052
Neg logit contributions
 neither-0.040
 mess-0.037
 wasteful-0.032
 fighting-0.032
 roughly-0.031
Token said
Token ID531
Feature activation-8.22e-01

Pos logit contributions
ka+0.032
tie+0.032
ance+0.031
dot+0.031
leaf+0.031
Neg logit contributions
 neither-0.029
 mess-0.028
 wasteful-0.026
 overflow-0.024
 fighting-0.023
Token to
Token ID284
Feature activation+1.56e-01

Pos logit contributions
tie+0.040
ka+0.040
leaf+0.039
shirt+0.038
 name+0.038
Neg logit contributions
 neither-0.026
 mess-0.024
 wasteful-0.022
 fighting-0.021
 overflow-0.020
Token her
Token ID607
Feature activation-1.86e+00

Pos logit contributions
 neither+0.007
 mess+0.007
 wasteful+0.007
 overflow+0.006
 fighting+0.006
Neg logit contributions
ka-0.005
 name-0.005
tie-0.005
ance-0.005
dot-0.005
TokenAt
Token ID2953
Feature activation+5.29e-01

Pos logit contributions
 neither+0.027
 mess+0.026
 wasteful+0.026
 ruining+0.024
 overflow+0.023
Neg logit contributions
 name-0.019
tie-0.018
ka-0.018
dot-0.018
shirt-0.017
Token first
Token ID717
Feature activation+1.57e+00

Pos logit contributions
 neither+0.020
 mess+0.019
 wasteful+0.019
 overflow+0.017
 roughly+0.017
Neg logit contributions
 aboard-0.021
ka-0.021
tie-0.021
ance-0.021
 name-0.021
Token,
Token ID11
Feature activation+1.44e+00

Pos logit contributions
 wasteful+0.078
 mess+0.073
 neither+0.073
 overflow+0.073
 ruining+0.072
Neg logit contributions
 name-0.041
 aboard-0.041
ance-0.039
 powers-0.038
orn-0.038
Token none
Token ID4844
Feature activation+4.56e-01

Pos logit contributions
 mess+0.074
 wasteful+0.071
 overflow+0.066
 neither+0.065
 ruining+0.064
Neg logit contributions
 name-0.040
tie-0.038
 aboard-0.038
 shone-0.034
 Magic-0.034
Token of
Token ID286
Feature activation+5.01e-02

Pos logit contributions
 neither+0.022
 mess+0.022
 wasteful+0.021
 overflow+0.020
 ruining+0.020
Neg logit contributions
tie-0.014
ka-0.013
 name-0.013
ance-0.013
 aboard-0.013
Token the
Token ID262
Feature activation+2.85e+00

Pos logit contributions
 neither+0.002
 mess+0.002
 wasteful+0.002
 overflow+0.002
 ruining+0.002
Neg logit contributions
ka-0.002
tie-0.002
 name-0.002
ance-0.002
dot-0.002
Token other
Token ID584
Feature activation+2.34e+00

Pos logit contributions
 neither+0.134
 wasteful+0.127
iders+0.119
 roughly+0.119
oving+0.118
Neg logit contributions
 name-0.099
 compass-0.078
 suitcase-0.074
 armour-0.073
 amulet-0.071
Token children
Token ID1751
Feature activation+3.51e-01

Pos logit contributions
 wasteful+0.102
 neither+0.094
 ruining+0.094
 roughly+0.090
 overflow+0.090
Neg logit contributions
 name-0.079
 aboard-0.074
 shone-0.072
ed-0.070
anced-0.068
Token wanted
Token ID2227
Feature activation+2.20e-01

Pos logit contributions
 neither+0.014
 mess+0.014
 wasteful+0.013
 overflow+0.012
 ruining+0.012
Neg logit contributions
tie-0.014
ka-0.014
 name-0.013
ance-0.013
dot-0.013
Token anything
Token ID1997
Feature activation+6.45e-01

Pos logit contributions
 neither+0.011
 mess+0.011
 wasteful+0.010
 overflow+0.009
 ruining+0.009
Neg logit contributions
ka-0.007
 name-0.007
tie-0.007
ance-0.006
dot-0.006
Token to
Token ID284
Feature activation+4.23e-01

Pos logit contributions
 neither+0.026
 mess+0.025
 wasteful+0.025
 overflow+0.023
 rats+0.022
Neg logit contributions
tie-0.024
 name-0.024
 aboard-0.023
ka-0.023
dot-0.023
Token want
Token ID765
Feature activation+1.38e+00

Pos logit contributions
ka+0.014
tie+0.013
ance+0.013
dot+0.013
leaf+0.013
Neg logit contributions
 neither-0.013
 mess-0.012
 wasteful-0.011
 fighting-0.010
 overflow-0.010
Token to
Token ID284
Feature activation+7.13e-01

Pos logit contributions
 neither+0.065
 mess+0.063
 wasteful+0.063
 overflow+0.061
 roughly+0.060
Neg logit contributions
 name-0.041
ka-0.040
 aboard-0.040
tie-0.039
 powers-0.038
Token build
Token ID1382
Feature activation+1.02e+00

Pos logit contributions
 neither+0.032
 wasteful+0.030
 mess+0.030
 mosquitoes+0.028
 rats+0.028
Neg logit contributions
 name-0.026
ka-0.024
tie-0.024
dot-0.023
ance-0.023
Token a
Token ID257
Feature activation+2.42e+00

Pos logit contributions
 neither+0.048
 wasteful+0.045
 mess+0.045
 overflow+0.043
 roughly+0.041
Neg logit contributions
 name-0.032
ka-0.032
 aboard-0.031
ance-0.030
tie-0.030
Token wide
Token ID3094
Feature activation+2.06e+00

Pos logit contributions
 neither+0.107
 wasteful+0.097
 ruining+0.089
 mosquitoes+0.087
 wasting+0.086
Neg logit contributions
 name-0.097
 Crown-0.080
yo-0.078
 aboard-0.077
ka-0.077
Token tower
Token ID10580
Feature activation+3.08e+00

Pos logit contributions
 neither+0.079
 wasteful+0.076
 mosquitoes+0.065
 ruining+0.065
 wasting+0.063
Neg logit contributions
 name-0.094
ka-0.089
 aboard-0.084
tie-0.083
 Crown-0.083
Token,
Token ID11
Feature activation+1.71e+00

Pos logit contributions
 wasteful+0.146
 mosquitoes+0.135
 neither+0.135
 sudden+0.134
 overflow+0.130
Neg logit contributions
 aboard-0.097
tie-0.085
 name-0.085
ka-0.079
ance-0.076
Token like
Token ID588
Feature activation+1.09e+00

Pos logit contributions
 mess+0.059
 neither+0.056
 wasteful+0.056
 roughly+0.051
 overflow+0.051
Neg logit contributions
 name-0.077
tie-0.068
shirt-0.066
 aboard-0.064
 Magic-0.064
Token a
Token ID257
Feature activation+2.39e+00

Pos logit contributions
 neither+0.042
 mess+0.041
 wasteful+0.040
 overflow+0.037
 roughly+0.036
Neg logit contributions
 name-0.042
ka-0.042
dot-0.041
tie-0.041
 aboard-0.041
Token castle
Token ID16669
Feature activation+2.98e+00

Pos logit contributions
 neither+0.096
 wasteful+0.088
 ruining+0.077
 overflow+0.077
 wasting+0.074
Neg logit contributions
 name-0.104
 Crown-0.090
tie-0.087
 ti-0.087
ka-0.086
Token!"
Token ID2474
Feature activation+1.01e+00

Pos logit contributions
 wasteful+0.139
 neither+0.137
 overflow+0.131
 sudden+0.128
 mosquitoes+0.127
Neg logit contributions
 name-0.090
 aboard-0.085
tie-0.083
 ti-0.079
ka-0.078
Token.
Token ID13
Feature activation-5.91e-01

Pos logit contributions
 neither+0.034
 mess+0.032
 wasteful+0.032
 overflow+0.030
 ruining+0.029
Neg logit contributions
 name-0.033
ka-0.032
tie-0.032
ance-0.031
 aboard-0.031
Token Now
Token ID2735
Feature activation+8.76e-01

Pos logit contributions
ka+0.023
tie+0.022
dot+0.021
ance+0.021
leaf+0.021
Neg logit contributions
 neither-0.025
 mess-0.023
 wasteful-0.023
 fighting-0.021
 overflow-0.020
Token we
Token ID356
Feature activation+2.47e-01

Pos logit contributions
 neither+0.033
 wasteful+0.033
 mess+0.032
 roughly+0.029
 overflow+0.029
Neg logit contributions
 name-0.036
ance-0.034
 aboard-0.033
ka-0.033
tie-0.033
Token can
Token ID460
Feature activation+8.17e-01

Pos logit contributions
 neither+0.012
 mess+0.012
 wasteful+0.012
 overflow+0.011
 ruining+0.011
Neg logit contributions
 name-0.007
ka-0.007
tie-0.007
ance-0.007
dot-0.007
Token enjoy
Token ID2883
Feature activation+9.99e-01

Pos logit contributions
 neither+0.034
 wasteful+0.034
 mess+0.033
 rats+0.030
 roughly+0.029
Neg logit contributions
 name-0.033
tie-0.029
ance-0.029
ka-0.029
dot-0.028
Token cocoa
Token ID35845
Feature activation+2.94e+00

Pos logit contributions
 neither+0.043
 mess+0.040
 wasteful+0.040
 overflow+0.037
 roughly+0.036
Neg logit contributions
 name-0.036
 aboard-0.035
dot-0.034
tie-0.034
ka-0.034
Token and
Token ID290
Feature activation+1.73e+00

Pos logit contributions
 wasteful+0.122
 neither+0.111
 mess+0.110
Uh+0.106
 overflow+0.103
Neg logit contributions
 aboard-0.111
ance-0.107
tie-0.104
 name-0.103
ka-0.101
Token stay
Token ID2652
Feature activation+3.70e-01

Pos logit contributions
 neither+0.075
 mess+0.070
 wasteful+0.069
 roughly+0.066
 overflow+0.063
Neg logit contributions
 name-0.064
leaf-0.055
tie-0.055
dot-0.054
 powers-0.054
Token healthy
Token ID5448
Feature activation+2.42e+00

Pos logit contributions
 neither+0.014
 mess+0.014
 wasteful+0.013
 overflow+0.012
 rats+0.012
Neg logit contributions
 name-0.015
ka-0.015
tie-0.015
dot-0.014
 aboard-0.014
Token too
Token ID1165
Feature activation+3.26e+00

Pos logit contributions
 wasteful+0.102
 neither+0.101
 mess+0.097
 overflow+0.095
 erupted+0.088
Neg logit contributions
 aboard-0.088
tie-0.087
 name-0.085
ka-0.084
ance-0.082
Token!"
Token ID2474
Feature activation+4.79e-01

Pos logit contributions
 wasteful+0.167
 mess+0.156
Uh+0.153
 neither+0.153
 mosquitoes+0.151
Neg logit contributions
 name-0.086
ka-0.079
 aboard-0.078
tie-0.075
 ti-0.073
Token very
Token ID845
Feature activation+1.17e+00

Pos logit contributions
 neither+0.034
 mess+0.033
 wasteful+0.031
 structures+0.030
 overflow+0.030
Neg logit contributions
 name-0.031
ka-0.030
tie-0.030
dot-0.030
leaf-0.029
Token grateful
Token ID14066
Feature activation-5.79e-01

Pos logit contributions
 neither+0.054
 mess+0.050
 structures+0.050
 snakes+0.048
 mosquitoes+0.047
Neg logit contributions
 name-0.041
ka-0.036
leaf-0.036
tie-0.034
dot-0.034
Token.
Token ID13
Feature activation+2.17e+00

Pos logit contributions
ka+0.024
dot+0.023
tie+0.023
ance+0.023
leaf+0.023
Neg logit contributions
 neither-0.023
 mess-0.022
 wasteful-0.020
 fighting-0.019
 roughly-0.019
Token
Token ID220
Feature activation+4.77e-01

Pos logit contributions
 wasteful+0.127
 neither+0.126
 ruining+0.122
 mosquitoes+0.121
 overflow+0.120
Neg logit contributions
 name-0.052
tie-0.039
 Magic-0.036
 aboard-0.035
shirt-0.034
Token
Token ID198
Feature activation+1.84e+00

Pos logit contributions
 neither+0.021
 mess+0.021
 wasteful+0.020
 overflow+0.018
 ruining+0.018
Neg logit contributions
 name-0.017
tie-0.016
ka-0.016
ance-0.016
dot-0.016
Token
Token ID198
Feature activation+2.38e+00

Pos logit contributions
 neither+0.106
 wasteful+0.102
 mess+0.098
 ruining+0.096
 rats+0.092
Neg logit contributions
 name-0.047
tie-0.041
shirt-0.040
 aboard-0.040
dot-0.039
TokenMrs
Token ID27034
Feature activation+1.46e-01

Pos logit contributions
 neither+0.115
 snakes+0.094
 drown+0.091
 wasteful+0.090
 mouths+0.089
Neg logit contributions
shirt-0.087
 name-0.081
tie-0.081
yo-0.080
new-0.076
Token.
Token ID13
Feature activation+9.53e-01

Pos logit contributions
 neither+0.006
 mess+0.006
 wasteful+0.006
 overflow+0.005
 ruining+0.005
Neg logit contributions
ka-0.005
tie-0.005
 name-0.005
ance-0.005
dot-0.005
Token Johnson
Token ID5030
Feature activation+8.35e-01

Pos logit contributions
 neither+0.032
 mess+0.031
 wasteful+0.030
 overflow+0.027
 mosquitoes+0.026
Neg logit contributions
 name-0.045
ka-0.042
tie-0.042
 aboard-0.041
anced-0.041
Token asked
Token ID1965
Feature activation+1.18e+00

Pos logit contributions
 neither+0.035
 mess+0.034
 wasteful+0.033
 overflow+0.031
 ruining+0.031
Neg logit contributions
ka-0.030
tie-0.030
 name-0.030
 aboard-0.028
dot-0.028
Token Lily
Token ID20037
Feature activation-5.34e-01

Pos logit contributions
 neither+0.050
 mess+0.049
 overflow+0.047
 wasteful+0.046
 ruining+0.044
Neg logit contributions
 name-0.042
ance-0.039
ka-0.038
tie-0.038
 aboard-0.038
Token "
Token ID366
Feature activation-3.68e-01

Pos logit contributions
 neither+0.043
 mess+0.043
 wasteful+0.041
 overflow+0.038
 ruining+0.037
Neg logit contributions
 name-0.045
anced-0.040
dot-0.040
tie-0.039
ka-0.039
TokenYes
Token ID5297
Feature activation+1.86e+00

Pos logit contributions
ka+0.016
ance+0.015
 name+0.015
tie+0.015
 aboard+0.015
Neg logit contributions
 neither-0.013
 mess-0.013
 wasteful-0.012
 roughly-0.011
 overflow-0.011
Token,
Token ID11
Feature activation+1.39e+00

Pos logit contributions
 wasteful+0.098
 mess+0.095
 ruining+0.091
 overflow+0.091
 structures+0.091
Neg logit contributions
 name-0.042
ka-0.042
anger-0.036
 aboard-0.035
tie-0.034
Token it
Token ID340
Feature activation-5.03e-02

Pos logit contributions
 mess+0.047
 wasteful+0.045
 neither+0.044
 roughly+0.041
 overflow+0.041
Neg logit contributions
 name-0.062
tie-0.058
 aboard-0.057
yo-0.056
ance-0.056
Token is
Token ID318
Feature activation+5.50e-01

Pos logit contributions
ka+0.002
tie+0.002
 name+0.002
ance+0.002
dot+0.002
Neg logit contributions
 neither-0.002
 mess-0.002
 wasteful-0.002
 overflow-0.002
 fighting-0.002
Token safe
Token ID3338
Feature activation+2.48e+00

Pos logit contributions
 neither+0.020
 mess+0.019
 wasteful+0.019
 overflow+0.018
 fighting+0.017
Neg logit contributions
 name-0.024
ka-0.023
tie-0.023
 aboard-0.022
dot-0.022
Token to
Token ID284
Feature activation+8.98e-01

Pos logit contributions
 neither+0.108
 wasteful+0.106
 overflow+0.101
 mess+0.100
 ruining+0.100
Neg logit contributions
 aboard-0.082
tie-0.081
 name-0.080
ka-0.076
 powers-0.076
Token eat
Token ID4483
Feature activation+2.88e+00

Pos logit contributions
 neither+0.036
 wasteful+0.032
 mess+0.032
 structures+0.029
 fighting+0.029
Neg logit contributions
 name-0.038
tie-0.035
dot-0.033
ka-0.033
 Crown-0.033
Token,"
Token ID553
Feature activation+1.36e+00

Pos logit contributions
 neither+0.135
 overflow+0.123
 wasteful+0.118
Uh+0.114
 ruining+0.114
Neg logit contributions
 name-0.083
 powers-0.083
tie-0.082
 aboard-0.081
ka-0.081
Token she
Token ID673
Feature activation+4.30e-01

Pos logit contributions
 mess+0.063
 wasteful+0.062
 overflow+0.060
 neither+0.059
 ruining+0.059
Neg logit contributions
 name-0.045
ka-0.038
shirt-0.037
 shone-0.037
tie-0.037
Token said
Token ID531
Feature activation+1.23e-01

Pos logit contributions
 neither+0.016
 mess+0.015
 wasteful+0.014
 overflow+0.014
 ruining+0.014
Neg logit contributions
 name-0.019
ka-0.018
tie-0.018
leaf-0.018
dot-0.017
Token was
Token ID373
Feature activation-1.35e-01

Pos logit contributions
ka+0.039
dot+0.038
tie+0.037
ance+0.037
 name+0.037
Neg logit contributions
 neither-0.038
 mess-0.037
 wasteful-0.033
 fighting-0.032
 roughly-0.031
Token no
Token ID645
Feature activation+4.73e-01

Pos logit contributions
ka+0.006
tie+0.006
 name+0.006
ance+0.006
dot+0.006
Neg logit contributions
 neither-0.005
 mess-0.005
 wasteful-0.004
 fighting-0.004
 overflow-0.004
Token longer
Token ID2392
Feature activation+5.52e-01

Pos logit contributions
 neither+0.010
 mess+0.009
 wasteful+0.009
 overflow+0.008
 structures+0.007
Neg logit contributions
ka-0.028
tie-0.027
 name-0.026
dot-0.026
leaf-0.026
Token mad
Token ID8805
Feature activation+7.44e-01

Pos logit contributions
 neither+0.017
 mess+0.016
 wasteful+0.015
 overflow+0.014
 roughly+0.013
Neg logit contributions
ka-0.027
 name-0.026
tie-0.026
dot-0.025
ance-0.025
Token at
Token ID379
Feature activation+6.66e-01

Pos logit contributions
 mess+0.031
 neither+0.031
 wasteful+0.031
 overflow+0.028
 ruining+0.027
Neg logit contributions
 name-0.027
tie-0.026
ka-0.026
ance-0.026
 aboard-0.025
Token her
Token ID607
Feature activation+5.59e-01

Pos logit contributions
 neither+0.028
 mess+0.027
 wasteful+0.027
 overflow+0.025
 roughly+0.024
Neg logit contributions
 name-0.024
ka-0.024
ance-0.024
tie-0.023
 aboard-0.023
Token!
Token ID0
Feature activation+3.20e-01

Pos logit contributions
 neither+0.024
 mess+0.023
 wasteful+0.022
 overflow+0.021
 ruining+0.020
Neg logit contributions
 name-0.020
tie-0.020
ka-0.019
ance-0.019
dot-0.019
Token<|endoftext|>
Token ID50256
Feature activation-1.88e+00

Pos logit contributions
 mess+0.014
 neither+0.014
 wasteful+0.013
 overflow+0.013
 ruining+0.012
Neg logit contributions
 name-0.011
tie-0.011
ka-0.011
ance-0.011
dot-0.011
TokenOnce
Token ID7454
Feature activation+1.01e-01

Pos logit contributions
ka+0.078
tie+0.077
leaf+0.077
dot+0.076
shirt+0.075
Neg logit contributions
 neither-0.067
 mess-0.066
 fighting-0.061
 roughly-0.056
 Amy-0.056
Token upon
Token ID2402
Feature activation-7.77e-01

Pos logit contributions
 neither+0.005
 mess+0.005
 wasteful+0.005
 overflow+0.005
 ruining+0.005
Neg logit contributions
ka-0.003
tie-0.003
 name-0.003
ance-0.003
dot-0.003
Token a
Token ID257
Feature activation+1.43e+00

Pos logit contributions
ka+0.029
ance+0.028
tie+0.028
anced+0.027
 aboard+0.026
Neg logit contributions
 neither-0.033
 mess-0.033
 wasteful-0.030
 fighting-0.029
 roughly-0.028
Token with
Token ID351
Feature activation+2.80e-01

Pos logit contributions
 neither+0.030
 wasteful+0.030
 mess+0.029
 overflow+0.027
 ruining+0.026
Neg logit contributions
 name-0.022
ka-0.022
tie-0.022
 aboard-0.022
ance-0.022
Token their
Token ID511
Feature activation-2.22e-01

Pos logit contributions
 neither+0.008
 mess+0.008
 wasteful+0.008
 overflow+0.007
 roughly+0.007
Neg logit contributions
 name-0.014
ka-0.013
tie-0.013
ance-0.013
dot-0.013
Token work
Token ID670
Feature activation+7.68e-03

Pos logit contributions
ka+0.009
tie+0.009
ance+0.008
 name+0.008
dot+0.008
Neg logit contributions
 mess-0.009
 neither-0.009
 wasteful-0.008
 fighting-0.008
 overflow-0.008
Token.
Token ID13
Feature activation+1.32e-03

Pos logit contributions
 neither+0.000
 mess+0.000
 wasteful+0.000
 overflow+0.000
 fighting+0.000
Neg logit contributions
ka-0.000
tie-0.000
 name-0.000
ance-0.000
dot-0.000
Token The
Token ID383
Feature activation+1.29e-02

Pos logit contributions
 neither+0.000
 mess+0.000
 wasteful+0.000
 overflow+0.000
 ruining+0.000
Neg logit contributions
ka-0.000
tie-0.000
 name-0.000
ance-0.000
dot-0.000
Token palace
Token ID20562
Feature activation+1.28e+00

Pos logit contributions
 neither+0.000
 mess+0.000
 wasteful+0.000
 overflow+0.000
 fighting+0.000
Neg logit contributions
ka-0.001
tie-0.001
 name-0.001
ance-0.001
dot-0.001
Token was
Token ID373
Feature activation+1.91e-01

Pos logit contributions
 neither+0.061
 wasteful+0.061
 mess+0.059
 ruining+0.057
 overflow+0.056
Neg logit contributions
 name-0.037
ka-0.036
tie-0.036
 shone-0.035
ance-0.035
Token neat
Token ID15049
Feature activation-2.17e-01

Pos logit contributions
 neither+0.007
 mess+0.006
 wasteful+0.006
 overflow+0.005
 fighting+0.005
Neg logit contributions
ka-0.008
tie-0.008
 name-0.008
dot-0.008
ance-0.008
Token and
Token ID290
Feature activation+4.21e-01

Pos logit contributions
ka+0.009
tie+0.009
 name+0.009
ance+0.009
dot+0.009
Neg logit contributions
 neither-0.008
 mess-0.008
 wasteful-0.007
 overflow-0.007
 fighting-0.007
Token tidy
Token ID43044
Feature activation+7.01e-01

Pos logit contributions
 neither+0.014
 mess+0.014
 wasteful+0.013
 overflow+0.012
 fighting+0.012
Neg logit contributions
ka-0.019
tie-0.018
 name-0.018
ance-0.018
dot-0.018
Token once
Token ID1752
Feature activation+7.34e-01

Pos logit contributions
 neither+0.030
 wasteful+0.028
 mess+0.028
 overflow+0.025
 mosquitoes+0.025
Neg logit contributions
ka-0.026
tie-0.025
 name-0.025
ance-0.024
 aboard-0.024
TokenTom
Token ID13787
Feature activation+1.28e+00

Pos logit contributions
 mess+0.087
 neither+0.085
 quarrel+0.076
 wasteful+0.076
 anymore+0.075
Neg logit contributions
 name-0.056
da-0.056
dot-0.055
ka-0.055
orn-0.054
Token and
Token ID290
Feature activation-2.39e-01

Pos logit contributions
 neither+0.062
 wasteful+0.058
 overflow+0.054
 mess+0.053
 unfortunately+0.052
Neg logit contributions
 name-0.041
ka-0.040
ance-0.038
tie-0.037
 aboard-0.037
Token Mia
Token ID32189
Feature activation+1.22e+00

Pos logit contributions
ka+0.015
tie+0.015
ance+0.014
 name+0.014
dot+0.014
Neg logit contributions
 neither-0.004
 mess-0.004
 wasteful-0.004
 overflow-0.003
 fighting-0.003
Token are
Token ID389
Feature activation+1.96e+00

Pos logit contributions
 neither+0.052
 wasteful+0.051
 mess+0.049
 structures+0.045
 rats+0.044
Neg logit contributions
 name-0.046
tie-0.041
ance-0.041
ka-0.041
dot-0.039
Token shocked
Token ID11472
Feature activation+7.52e-01

Pos logit contributions
 wasteful+0.069
 neither+0.068
 timer+0.064
 mess+0.064
 overflow+0.064
Neg logit contributions
 name-0.079
tie-0.071
 aboard-0.066
ocket-0.066
ka-0.066
Token.
Token ID13
Feature activation+1.28e+00

Pos logit contributions
 neither+0.037
 wasteful+0.036
 mess+0.035
 overflow+0.033
 ruining+0.032
Neg logit contributions
 name-0.022
ka-0.021
tie-0.021
 aboard-0.020
dot-0.020
Token They
Token ID1119
Feature activation+5.39e-01

Pos logit contributions
 mess+0.050
 wasteful+0.048
 neither+0.048
 ruining+0.044
 overflow+0.042
Neg logit contributions
 name-0.058
dot-0.049
ka-0.048
anced-0.048
ance-0.047
Token drop
Token ID4268
Feature activation+5.90e-01

Pos logit contributions
 neither+0.021
 mess+0.019
 wasteful+0.019
 overflow+0.017
 rats+0.017
Neg logit contributions
 name-0.022
ka-0.021
tie-0.021
leaf-0.020
dot-0.020
Token the
Token ID262
Feature activation-1.16e+00

Pos logit contributions
 mess+0.022
 neither+0.022
 wasteful+0.021
 ruining+0.019
 overflow+0.019
Neg logit contributions
 name-0.023
ka-0.023
tie-0.023
 aboard-0.023
ance-0.022
Token bow
Token ID9563
Feature activation+1.58e-01

Pos logit contributions
ance+0.038
dot+0.037
shirt+0.036
anced+0.036
ka+0.035
Neg logit contributions
 mess-0.059
 neither-0.053
 fighting-0.052
 wasteful-0.049
 floor-0.049
Token and
Token ID290
Feature activation+9.34e-01

Pos logit contributions
 neither+0.007
 mess+0.007
 wasteful+0.006
 overflow+0.006
 ruining+0.006
Neg logit contributions
tie-0.006
ka-0.006
 name-0.006
ance-0.006
dot-0.005
Token
Token ID198
Feature activation+5.70e-01

Pos logit contributions
ka+0.002
tie+0.002
 name+0.002
ance+0.002
dot+0.002
Neg logit contributions
 neither-0.002
 mess-0.002
 wasteful-0.002
 overflow-0.002
 fighting-0.002
Token
Token ID198
Feature activation+1.66e+00

Pos logit contributions
 neither+0.027
 mess+0.026
 wasteful+0.026
 ruining+0.024
 overflow+0.024
Neg logit contributions
 name-0.019
tie-0.018
ka-0.018
dot-0.017
shirt-0.017
TokenSuddenly
Token ID38582
Feature activation+8.96e-01

Pos logit contributions
 neither+0.070
 wasteful+0.063
 mess+0.062
 mouths+0.057
 overflow+0.056
Neg logit contributions
shirt-0.065
tie-0.064
 name-0.062
dot-0.062
leaf-0.062
Token,
Token ID11
Feature activation+1.58e+00

Pos logit contributions
 mess+0.040
 wasteful+0.039
 neither+0.039
 overflow+0.036
 ruining+0.035
Neg logit contributions
ka-0.030
 name-0.030
ance-0.029
 aboard-0.029
tie-0.028
Token the
Token ID262
Feature activation-6.67e-01

Pos logit contributions
 mess+0.074
 wasteful+0.073
 neither+0.069
 overflow+0.067
 anth+0.066
Neg logit contributions
 name-0.048
ka-0.046
 shone-0.043
ance-0.043
 aboard-0.043
Token paper
Token ID3348
Feature activation+2.97e-01

Pos logit contributions
ka+0.027
ance+0.027
tie+0.027
dot+0.027
anced+0.026
Neg logit contributions
 mess-0.027
 neither-0.026
 wasteful-0.023
 fighting-0.023
 floor-0.022
Token boat
Token ID8848
Feature activation+3.80e-01

Pos logit contributions
 neither+0.013
 mess+0.012
 wasteful+0.012
 overflow+0.011
 ruining+0.011
Neg logit contributions
ka-0.011
tie-0.011
 name-0.011
ance-0.010
dot-0.010
Token started
Token ID2067
Feature activation-1.88e-01

Pos logit contributions
 neither+0.016
 mess+0.016
 wasteful+0.015
 overflow+0.014
 ruining+0.014
Neg logit contributions
ka-0.014
tie-0.014
 name-0.014
 aboard-0.013
ance-0.013
Token to
Token ID284
Feature activation+1.16e+00

Pos logit contributions
ka+0.006
tie+0.006
 name+0.006
ance+0.006
dot+0.006
Neg logit contributions
 neither-0.009
 mess-0.009
 wasteful-0.008
 fighting-0.008
 overflow-0.008
Token fill
Token ID6070
Feature activation+1.07e+00

Pos logit contributions
 neither+0.052
 wasteful+0.050
 mess+0.046
 roughly+0.045
 rats+0.045
Neg logit contributions
 name-0.039
 aboard-0.038
ka-0.037
 shone-0.036
 outer-0.035
Token with
Token ID351
Feature activation-1.55e-01

Pos logit contributions
 neither+0.054
 wasteful+0.052
 mess+0.049
 ruining+0.047
 overflow+0.047
Neg logit contributions
 name-0.032
ka-0.032
 aboard-0.030
tie-0.029
ance-0.028
Token,
Token ID11
Feature activation+2.14e+00

Pos logit contributions
 neither+0.042
 overflow+0.040
 wasteful+0.040
 mess+0.039
 flood+0.036
Neg logit contributions
 name-0.036
ka-0.036
 powers-0.035
ance-0.034
tie-0.033
Token sweet
Token ID6029
Feature activation+7.51e-01

Pos logit contributions
 neither+0.080
 overflow+0.077
 erupted+0.076
 wasteful+0.073
 roughly+0.073
Neg logit contributions
 name-0.087
 Magic-0.073
 rib-0.071
ka-0.071
 ti-0.069
Tokenie
Token ID494
Feature activation+7.78e-01

Pos logit contributions
 neither+0.033
 wasteful+0.030
 mess+0.029
 ruining+0.029
 overflow+0.028
Neg logit contributions
ka-0.028
 name-0.027
ance-0.026
tie-0.025
shirt-0.025
Token.
Token ID13
Feature activation-1.53e+00

Pos logit contributions
 neither+0.031
 mess+0.030
 wasteful+0.029
 overflow+0.028
 ruining+0.027
Neg logit contributions
ka-0.030
 name-0.030
tie-0.029
ance-0.028
 aboard-0.028
Token Painting
Token ID48892
Feature activation-3.00e-01

Pos logit contributions
ka+0.072
tie+0.071
 aboard+0.068
leaf+0.067
ocket+0.067
Neg logit contributions
 neither-0.051
 wasteful-0.046
 mess-0.041
 fighting-0.040
 roughly-0.039
Token is
Token ID318
Feature activation+5.56e-01

Pos logit contributions
ka+0.011
tie+0.011
 name+0.010
dot+0.010
ance+0.010
Neg logit contributions
 neither-0.013
 mess-0.013
 wasteful-0.012
 overflow-0.011
 fighting-0.011
Token fun
Token ID1257
Feature activation+1.29e+00

Pos logit contributions
 neither+0.019
 mess+0.019
 wasteful+0.017
 overflow+0.017
 fighting+0.016
Neg logit contributions
 name-0.024
tie-0.023
ka-0.023
ance-0.023
 aboard-0.023
Token,
Token ID11
Feature activation+1.29e+00

Pos logit contributions
 neither+0.061
 wasteful+0.057
 overflow+0.054
 mess+0.053
 ruining+0.051
Neg logit contributions
 name-0.041
 powers-0.041
ka-0.039
tie-0.039
 aboard-0.038
Token but
Token ID475
Feature activation+4.90e-01

Pos logit contributions
 neither+0.040
 wasteful+0.038
 mess+0.037
 overflow+0.034
 roughly+0.033
Neg logit contributions
 name-0.061
tie-0.057
ka-0.055
shirt-0.055
 aboard-0.053
Token it
Token ID340
Feature activation+9.98e-01

Pos logit contributions
 neither+0.015
 mess+0.015
 wasteful+0.013
 overflow+0.012
 roughly+0.012
Neg logit contributions
 name-0.023
tie-0.023
ka-0.023
ance-0.023
 aboard-0.022
Token can
Token ID460
Feature activation+1.11e+00

Pos logit contributions
 neither+0.042
 mess+0.040
 wasteful+0.040
 overflow+0.037
 roughly+0.037
Neg logit contributions
 name-0.035
ka-0.035
ance-0.034
tie-0.033
 aboard-0.032
Token saw
Token ID2497
Feature activation-7.97e-01

Pos logit contributions
 mess+0.026
 neither+0.026
 wasteful+0.025
 overflow+0.023
 ruining+0.022
Neg logit contributions
tie-0.022
ka-0.021
 name-0.021
 aboard-0.020
dot-0.020
Token that
Token ID326
Feature activation+2.87e-01

Pos logit contributions
ka+0.031
tie+0.030
shirt+0.030
dot+0.030
leaf+0.029
Neg logit contributions
 neither-0.034
 mess-0.031
 wasteful-0.030
 fighting-0.029
 rats-0.027
Token it
Token ID340
Feature activation+3.63e-01

Pos logit contributions
 mess+0.011
 neither+0.011
 wasteful+0.011
 overflow+0.010
 ruining+0.010
Neg logit contributions
 name-0.011
tie-0.011
ka-0.011
 aboard-0.011
ance-0.011
Token was
Token ID373
Feature activation-3.40e-01

Pos logit contributions
 neither+0.017
 mess+0.016
 wasteful+0.016
 overflow+0.014
 roughly+0.014
Neg logit contributions
ka-0.012
tie-0.012
 name-0.012
 aboard-0.011
ance-0.011
Token a
Token ID257
Feature activation-2.81e-01

Pos logit contributions
ka+0.012
tie+0.012
ance+0.012
dot+0.012
shirt+0.012
Neg logit contributions
 neither-0.015
 mess-0.015
 wasteful-0.014
 fighting-0.013
 ruining-0.013
Token whale
Token ID22206
Feature activation-4.43e-01

Pos logit contributions
ka+0.011
tie+0.011
ance+0.011
 name+0.011
dot+0.011
Neg logit contributions
 neither-0.011
 mess-0.011
 wasteful-0.010
 overflow-0.009
 fighting-0.009
Token.
Token ID13
Feature activation-8.46e-01

Pos logit contributions
ka+0.017
tie+0.017
ance+0.017
 name+0.016
dot+0.016
Neg logit contributions
 neither-0.018
 mess-0.018
 wasteful-0.016
 fighting-0.016
 ruining-0.015
Token
Token ID198
Feature activation-9.21e-01

Pos logit contributions
ka+0.037
tie+0.036
ance+0.036
leaf+0.036
 powers+0.035
Neg logit contributions
 neither-0.031
 mess-0.029
 wasteful-0.027
 fighting-0.026
 overflow-0.025
Token
Token ID198
Feature activation-8.16e-01

Pos logit contributions
ka+0.043
ance+0.042
tie+0.042
anced+0.041
dot+0.041
Neg logit contributions
 neither-0.030
 mess-0.030
 wasteful-0.026
 roughly-0.025
 fighting-0.025
TokenThe
Token ID464
Feature activation+4.14e-01

Pos logit contributions
ka+0.036
ance+0.036
anced+0.035
tie+0.035
leaf+0.035
Neg logit contributions
 mess-0.028
 neither-0.028
 wasteful-0.026
 roughly-0.024
 overflow-0.023
Token whale
Token ID22206
Feature activation+9.57e-01

Pos logit contributions
 neither+0.017
 mess+0.016
 wasteful+0.016
 roughly+0.015
 ruining+0.014
Neg logit contributions
 name-0.016
ka-0.016
 aboard-0.016
tie-0.015
leaf-0.015
Token saw
Token ID2497
Feature activation-7.11e-01

Pos logit contributions
 neither+0.000
 mess+0.000
 wasteful+0.000
 overflow+0.000
 fighting+0.000
Neg logit contributions
ka-0.000
tie-0.000
 name-0.000
ance-0.000
dot-0.000
Token that
Token ID326
Feature activation+9.25e-02

Pos logit contributions
ka+0.032
tie+0.031
dot+0.030
 name+0.030
shirt+0.030
Neg logit contributions
 neither-0.026
 mess-0.025
 wasteful-0.023
 fighting-0.021
 ruining-0.021
Token it
Token ID340
Feature activation-8.10e-02

Pos logit contributions
 neither+0.004
 mess+0.004
 wasteful+0.003
 overflow+0.003
 ruining+0.003
Neg logit contributions
ka-0.004
tie-0.004
 name-0.004
ance-0.004
dot-0.004
Token was
Token ID373
Feature activation-4.11e-03

Pos logit contributions
ka+0.003
tie+0.003
 name+0.003
ance+0.003
dot+0.003
Neg logit contributions
 neither-0.003
 mess-0.003
 wasteful-0.003
 overflow-0.003
 fighting-0.003
Token the
Token ID262
Feature activation+8.28e-02

Pos logit contributions
ka+0.000
tie+0.000
 name+0.000
ance+0.000
dot+0.000
Neg logit contributions
 neither-0.000
 mess-0.000
 wasteful-0.000
 overflow-0.000
 fighting-0.000
Token perfect
Token ID2818
Feature activation-6.03e-01

Pos logit contributions
 neither+0.004
 mess+0.003
 wasteful+0.003
 overflow+0.003
 fighting+0.003
Neg logit contributions
ka-0.003
 name-0.003
tie-0.003
ance-0.003
dot-0.003
Token size
Token ID2546
Feature activation-2.02e+00

Pos logit contributions
ka+0.025
tie+0.025
ance+0.025
dot+0.024
leaf+0.024
Neg logit contributions
 mess-0.024
 neither-0.023
 wasteful-0.020
 fighting-0.019
 overflow-0.019
Token.
Token ID13
Feature activation+6.43e-02

Pos logit contributions
ka+0.090
 name+0.087
tie+0.087
dot+0.086
Aid+0.085
Neg logit contributions
 mess-0.072
 neither-0.071
 wasteful-0.061
 floor-0.059
 fighting-0.058
Token Sm
Token ID2439
Feature activation+2.69e-01

Pos logit contributions
 neither+0.003
 mess+0.003
 wasteful+0.003
 overflow+0.003
 ruining+0.002
Neg logit contributions
ka-0.002
tie-0.002
 name-0.002
ance-0.002
dot-0.002
Tokeniling
Token ID4386
Feature activation-1.63e-01

Pos logit contributions
 neither+0.008
 mess+0.008
 wasteful+0.007
 overflow+0.006
 ruining+0.006
Neg logit contributions
ka-0.014
tie-0.014
ance-0.013
dot-0.013
 name-0.013
Token,
Token ID11
Feature activation+8.67e-01

Pos logit contributions
ka+0.007
tie+0.007
 name+0.007
dot+0.007
ance+0.007
Neg logit contributions
 neither-0.006
 mess-0.006
 wasteful-0.006
 fighting-0.005
 overflow-0.005
Token gathered
Token ID9272
Feature activation-9.57e-01

Pos logit contributions
ka+0.029
ance+0.028
tie+0.028
 name+0.028
dot+0.028
Neg logit contributions
 mess-0.027
 neither-0.027
 fighting-0.025
 wasteful-0.025
 ruining-0.023
Token around
Token ID1088
Feature activation+7.97e-02

Pos logit contributions
ka+0.037
ance+0.037
tie+0.037
dot+0.037
 name+0.036
Neg logit contributions
 neither-0.039
 mess-0.038
 wasteful-0.035
 fighting-0.033
 roughly-0.032
Token a
Token ID257
Feature activation-1.26e-01

Pos logit contributions
 neither+0.003
 mess+0.003
 wasteful+0.003
 overflow+0.003
 ruining+0.003
Neg logit contributions
ka-0.003
tie-0.003
 name-0.003
dot-0.003
ance-0.003
Token funny
Token ID8258
Feature activation-1.64e+00

Pos logit contributions
ka+0.005
tie+0.005
 name+0.005
ance+0.005
dot+0.005
Neg logit contributions
 neither-0.005
 mess-0.005
 wasteful-0.005
 overflow-0.004
 fighting-0.004
Token-
Token ID12
Feature activation-2.12e+00

Pos logit contributions
ance+0.066
anced+0.063
tie+0.061
ka+0.060
leaf+0.059
Neg logit contributions
 mess-0.075
 neither-0.063
 floor-0.060
 fighting-0.060
 quarrel-0.059
Tokenlooking
Token ID11534
Feature activation-8.39e-01

Pos logit contributions
anced+0.089
ance+0.083
 powers+0.077
 name+0.077
 amulet+0.077
Neg logit contributions
 neither-0.075
 roughly-0.073
 mess-0.071
 wasteful-0.070
 fighting-0.065
Token machine
Token ID4572
Feature activation-7.57e-01

Pos logit contributions
ka+0.034
ance+0.034
tie+0.034
dot+0.033
anced+0.033
Neg logit contributions
 mess-0.033
 neither-0.031
 wasteful-0.029
 fighting-0.027
 overflow-0.027
Token.
Token ID13
Feature activation+9.20e-01

Pos logit contributions
ka+0.031
tie+0.030
ance+0.030
 name+0.030
dot+0.030
Neg logit contributions
 neither-0.029
 mess-0.029
 wasteful-0.026
 fighting-0.025
 ruining-0.024
Token
Token ID198
Feature activation+6.51e-01

Pos logit contributions
 mess+0.047
 neither+0.047
 wasteful+0.046
 ruining+0.043
 overflow+0.043
Neg logit contributions
 name-0.028
tie-0.025
 aboard-0.024
dot-0.024
shirt-0.024
Token
Token ID198
Feature activation+1.48e+00

Pos logit contributions
 neither+0.031
 mess+0.030
 wasteful+0.030
 ruining+0.027
 overflow+0.027
Neg logit contributions
 name-0.022
tie-0.021
ka-0.021
 aboard-0.020
dot-0.020
TokenJohn
Token ID7554
Feature activation+4.16e-01

Pos logit contributions
 neither+0.067
 mess+0.066
 wasteful+0.063
 ruining+0.059
 snakes+0.058
Neg logit contributions
 name-0.052
shirt-0.051
tie-0.051
dot-0.050
leaf-0.050
Token path
Token ID3108
Feature activation-5.83e-01

Pos logit contributions
ka+0.001
tie+0.001
 name+0.001
ance+0.001
dot+0.001
Neg logit contributions
 neither-0.001
 mess-0.001
 wasteful-0.001
 overflow-0.001
 fighting-0.001
Token.
Token ID13
Feature activation+7.99e-01

Pos logit contributions
ka+0.024
tie+0.024
 name+0.023
dot+0.023
ance+0.023
Neg logit contributions
 neither-0.023
 mess-0.022
 wasteful-0.020
 fighting-0.019
 overflow-0.019
Token Suddenly
Token ID24975
Feature activation+6.09e-01

Pos logit contributions
 mess+0.031
 neither+0.028
 wasteful+0.028
 ruining+0.025
 overflow+0.025
Neg logit contributions
 name-0.036
tie-0.034
ka-0.033
dot-0.033
anced-0.033
Token,
Token ID11
Feature activation+1.13e+00

Pos logit contributions
 mess+0.026
 neither+0.024
 wasteful+0.024
 overflow+0.022
 ruining+0.022
Neg logit contributions
ka-0.023
 name-0.022
ance-0.022
tie-0.022
 aboard-0.022
Token the
Token ID262
Feature activation-1.51e-01

Pos logit contributions
 mess+0.047
 wasteful+0.042
 neither+0.039
 overflow+0.038
 ruining+0.038
Neg logit contributions
 aboard-0.044
 name-0.044
ka-0.044
tie-0.043
ance-0.042
Token path
Token ID3108
Feature activation-2.06e-01

Pos logit contributions
ka+0.006
tie+0.006
 name+0.006
dot+0.006
ance+0.006
Neg logit contributions
 neither-0.006
 mess-0.006
 wasteful-0.005
 overflow-0.005
 fighting-0.005
Token said
Token ID531
Feature activation-9.36e-01

Pos logit contributions
ka+0.008
tie+0.007
 name+0.007
dot+0.007
ance+0.007
Neg logit contributions
 neither-0.009
 mess-0.009
 wasteful-0.008
 fighting-0.007
 overflow-0.007
Token hello
Token ID23748
Feature activation-9.78e-01

Pos logit contributions
ka+0.045
leaf+0.044
tie+0.043
shirt+0.043
dot+0.043
Neg logit contributions
 neither-0.033
 mess-0.028
 wasteful-0.027
 fighting-0.026
 overflow-0.024
Token back
Token ID736
Feature activation-8.88e-01

Pos logit contributions
ka+0.043
leaf+0.043
dot+0.041
shirt+0.041
orn+0.041
Neg logit contributions
 neither-0.036
 mess-0.034
 fighting-0.030
 wasteful-0.029
 overflow-0.028
Token!
Token ID0
Feature activation+5.39e-01

Pos logit contributions
ka+0.040
leaf+0.039
dot+0.039
shirt+0.039
tie+0.038
Neg logit contributions
 neither-0.032
 mess-0.030
 wasteful-0.027
 fighting-0.027
 roughly-0.026
Token They
Token ID1119
Feature activation-6.52e-02

Pos logit contributions
 mess+0.023
 neither+0.022
 wasteful+0.021
 ruining+0.019
 overflow+0.019
Neg logit contributions
 name-0.021
ka-0.021
tie-0.020
dot-0.020
ance-0.020
Token.
Token ID13
Feature activation-1.94e+00

Pos logit contributions
ka+0.032
leaf+0.032
dot+0.031
ance+0.030
 aboard+0.030
Neg logit contributions
 neither-0.027
 mess-0.026
 wasteful-0.023
 fighting-0.023
 roughly-0.022
Token She
Token ID1375
Feature activation-3.10e-02

Pos logit contributions
leaf+0.102
ka+0.101
tie+0.097
ocket+0.097
shirt+0.093
Neg logit contributions
 neither-0.056
 mess-0.050
 structures-0.047
 fighting-0.046
 floor-0.045
Token was
Token ID373
Feature activation+3.83e-01

Pos logit contributions
ka+0.001
tie+0.001
 name+0.001
ance+0.001
dot+0.001
Neg logit contributions
 neither-0.001
 mess-0.001
 wasteful-0.001
 overflow-0.001
 fighting-0.001
Token having
Token ID1719
Feature activation-2.09e-01

Pos logit contributions
 neither+0.014
 mess+0.014
 wasteful+0.013
 overflow+0.012
 roughly+0.012
Neg logit contributions
 name-0.016
ka-0.016
tie-0.015
dot-0.015
leaf-0.015
Token so
Token ID523
Feature activation+1.47e-01

Pos logit contributions
ka+0.009
tie+0.009
 name+0.009
ance+0.008
dot+0.008
Neg logit contributions
 neither-0.008
 mess-0.008
 wasteful-0.007
 fighting-0.007
 ruining-0.007
Token much
Token ID881
Feature activation-9.52e-01

Pos logit contributions
 neither+0.005
 mess+0.005
 wasteful+0.005
 overflow+0.004
 ruining+0.004
Neg logit contributions
ka-0.007
tie-0.007
 name-0.006
dot-0.006
ance-0.006
Token fun
Token ID1257
Feature activation+1.01e-02

Pos logit contributions
ka+0.059
ance+0.058
tie+0.057
anced+0.057
ocket+0.056
Neg logit contributions
 mess-0.021
 neither-0.017
 wasteful-0.014
 fighting-0.014
 roughly-0.013
Token that
Token ID326
Feature activation-3.57e-01

Pos logit contributions
 neither+0.000
 mess+0.000
 wasteful+0.000
 overflow+0.000
 fighting+0.000
Neg logit contributions
ka-0.000
tie-0.000
 name-0.000
ance-0.000
dot-0.000
Token she
Token ID673
Feature activation+1.28e+00

Pos logit contributions
ka+0.017
tie+0.016
leaf+0.016
 name+0.016
ance+0.016
Neg logit contributions
 neither-0.012
 mess-0.012
 wasteful-0.011
 fighting-0.010
 roughly-0.010
Token started
Token ID2067
Feature activation+2.17e-01

Pos logit contributions
 neither+0.048
 wasteful+0.046
 mess+0.046
 structures+0.044
 overflow+0.042
Neg logit contributions
ka-0.052
tie-0.049
 name-0.049
shirt-0.048
u-0.047
Token to
Token ID284
Feature activation+5.51e-01

Pos logit contributions
 neither+0.010
 mess+0.010
 wasteful+0.010
 overflow+0.009
 roughly+0.009
Neg logit contributions
ka-0.007
 name-0.007
tie-0.007
dot-0.007
ance-0.006
Token showed
Token ID3751
Feature activation-1.66e+00

Pos logit contributions
ka+0.021
ance+0.021
tie+0.021
 name+0.021
dot+0.021
Neg logit contributions
 neither-0.026
 mess-0.026
 wasteful-0.024
 fighting-0.022
 roughly-0.022
Token her
Token ID607
Feature activation-1.99e+00

Pos logit contributions
dot+0.073
tie+0.073
ocket+0.071
leaf+0.070
 name+0.070
Neg logit contributions
 neither-0.055
 mess-0.051
 mice-0.047
 rats-0.045
 roughly-0.044
Token a
Token ID257
Feature activation-2.47e+00

Pos logit contributions
ka+0.096
leaf+0.093
dot+0.092
tie+0.091
orn+0.090
Neg logit contributions
 neither-0.067
 mess-0.063
 fighting-0.054
 not-0.052
 wasteful-0.050
Token beautiful
Token ID4950
Feature activation-3.79e+00

Pos logit contributions
ance+0.119
anced+0.110
ka+0.102
dot+0.101
tie+0.101
Neg logit contributions
 mess-0.102
 floor-0.082
 neither-0.077
 quarrel-0.075
 wasteful-0.072
Token dress
Token ID6576
Feature activation-2.34e+00

Pos logit contributions
ance+0.188
anced+0.174
aved+0.165
leaf+0.160
int+0.153
Neg logit contributions
 mess-0.171
 floor-0.137
 mistake-0.121
 table-0.119
 noise-0.117
Token with
Token ID351
Feature activation-2.65e+00

Pos logit contributions
ka+0.128
leaf+0.126
dot+0.122
Aid+0.121
ance+0.119
Neg logit contributions
 neither-0.062
 mess-0.061
 floor-0.052
 not-0.047
 fighting-0.045
Token colorful
Token ID20239
Feature activation-3.30e+00

Pos logit contributions
int+0.115
anger+0.111
orn+0.109
Aid+0.109
ance+0.109
Neg logit contributions
 mess-0.093
 neither-0.092
 floor-0.081
 fighting-0.080
 mice-0.079
Token flowers
Token ID12734
Feature activation-1.78e+00

Pos logit contributions
ance+0.142
leaf+0.132
anced+0.131
int+0.131
Aid+0.130
Neg logit contributions
 mess-0.132
 floor-0.118
 structures-0.097
 fighting-0.095
 table-0.094
Token on
Token ID319
Feature activation-8.56e-01

Pos logit contributions
ka+0.087
leaf+0.082
shirt+0.082
ance+0.082
Aid+0.081
Neg logit contributions
 neither-0.055
 mess-0.054
 fighting-0.048
 ruining-0.044
 floor-0.043
Token it
Token ID340
Feature activation-8.02e-01

Pos logit contributions
ka+0.038
ance+0.038
leaf+0.038
dot+0.037
tie+0.037
Neg logit contributions
 neither-0.029
 mess-0.029
 wasteful-0.025
 roughly-0.024
 fighting-0.024
Token.
Token ID13
Feature activation-1.26e+00

Pos logit contributions
ka+0.034
dot+0.032
tie+0.032
leaf+0.032
ance+0.032
Neg logit contributions
 neither-0.031
 mess-0.030
 wasteful-0.027
 fighting-0.026
 roughly-0.025
Token neck
Token ID7393
Feature activation-9.17e-01

Pos logit contributions
 neither+0.018
 mess+0.017
 wasteful+0.017
 overflow+0.015
 ruining+0.015
Neg logit contributions
 name-0.014
tie-0.014
ka-0.014
dot-0.013
ance-0.013
Token.
Token ID13
Feature activation-1.70e+00

Pos logit contributions
ka+0.040
leaf+0.038
dot+0.038
ance+0.037
 name+0.037
Neg logit contributions
 neither-0.034
 mess-0.033
 wasteful-0.029
 roughly-0.028
 fighting-0.027
Token
Token ID220
Feature activation-1.09e+00

Pos logit contributions
ka+0.082
leaf+0.080
tie+0.079
orn+0.077
ocket+0.077
Neg logit contributions
 neither-0.056
 mess-0.050
 floor-0.044
 fighting-0.044
 wasteful-0.043
Token
Token ID198
Feature activation-7.57e-01

Pos logit contributions
ka+0.052
tie+0.049
leaf+0.049
dot+0.049
 powers+0.047
Neg logit contributions
 neither-0.037
 mess-0.034
 wasteful-0.031
 fighting-0.030
 roughly-0.029
Token
Token ID198
Feature activation-7.20e-01

Pos logit contributions
ka+0.034
tie+0.033
ance+0.032
anced+0.032
dot+0.032
Neg logit contributions
 neither-0.026
 mess-0.025
 wasteful-0.022
 fighting-0.021
 roughly-0.021
TokenT
Token ID51
Feature activation-3.15e+00

Pos logit contributions
ka+0.032
tie+0.031
anced+0.030
leaf+0.030
ance+0.030
Neg logit contributions
 mess-0.026
 neither-0.025
 wasteful-0.023
 roughly-0.022
 overflow-0.021
Tokenilly
Token ID6548
Feature activation-4.21e-01

Pos logit contributions
leaf+0.136
tie+0.130
hy+0.127
chery+0.126
dot+0.124
Neg logit contributions
 mess-0.119
 neither-0.089
 floor-0.086
 anth-0.085
 wasteful-0.085
Token spent
Token ID3377
Feature activation+5.95e-01

Pos logit contributions
ka+0.016
tie+0.016
 name+0.016
dot+0.016
ance+0.016
Neg logit contributions
 neither-0.017
 mess-0.017
 wasteful-0.016
 fighting-0.015
 overflow-0.014
Token every
Token ID790
Feature activation-1.19e+00

Pos logit contributions
 neither+0.025
 mess+0.024
 wasteful+0.024
 overflow+0.022
 rats+0.020
Neg logit contributions
ka-0.022
tie-0.022
 name-0.022
dot-0.021
ance-0.021
Token day
Token ID1110
Feature activation+4.83e-01

Pos logit contributions
ance+0.056
anced+0.054
ka+0.054
 name+0.053
leaf+0.052
Neg logit contributions
 neither-0.041
 mess-0.040
 wasteful-0.034
 roughly-0.034
 fighting-0.033
Token playing
Token ID2712
Feature activation+1.56e-01

Pos logit contributions
 neither+0.021
 mess+0.020
 wasteful+0.020
 overflow+0.018
 rats+0.018
Neg logit contributions
ka-0.017
tie-0.017
dot-0.017
 name-0.016
ance-0.016
Token light
Token ID1657
Feature activation-2.01e+00

Pos logit contributions
ka+0.023
ance+0.023
tie+0.022
dot+0.022
anced+0.022
Neg logit contributions
 mess-0.029
 neither-0.029
 wasteful-0.027
 fighting-0.025
 ruining-0.025
Token,
Token ID11
Feature activation-1.04e+00

Pos logit contributions
ance+0.081
ka+0.081
dot+0.081
leaf+0.081
anced+0.079
Neg logit contributions
 mess-0.079
 neither-0.075
 fighting-0.065
 wasteful-0.064
 floor-0.063
Token like
Token ID588
Feature activation-1.77e+00

Pos logit contributions
ka+0.043
ance+0.043
tie+0.041
dot+0.041
anced+0.041
Neg logit contributions
 neither-0.040
 mess-0.040
 wasteful-0.036
 fighting-0.034
 ruining-0.034
Token a
Token ID257
Feature activation-8.98e-02

Pos logit contributions
ka+0.072
tie+0.070
shirt+0.070
dot+0.069
yo+0.067
Neg logit contributions
 neither-0.069
 mess-0.066
 fighting-0.060
 wasteful-0.060
 snakes-0.059
Token spark
Token ID9009
Feature activation-2.14e+00

Pos logit contributions
ka+0.003
tie+0.003
 name+0.003
ance+0.003
dot+0.003
Neg logit contributions
 neither-0.004
 mess-0.004
 wasteful-0.004
 overflow-0.004
 fighting-0.004
Tokenly
Token ID306
Feature activation-2.71e+00

Pos logit contributions
anced+0.088
ance+0.087
leaf+0.087
dot+0.086
lynn+0.082
Neg logit contributions
 mess-0.087
 neither-0.075
 anymore-0.068
 fighting-0.066
 floor-0.066
Token star
Token ID3491
Feature activation-1.68e+00

Pos logit contributions
ance+0.107
leaf+0.100
anced+0.099
dot+0.095
iddled+0.093
Neg logit contributions
 mess-0.121
 floor-0.105
 neither-0.098
 snakes-0.094
 rats-0.091
Token.
Token ID13
Feature activation+8.72e-01

Pos logit contributions
leaf+0.072
ka+0.070
ance+0.070
dot+0.070
shirt+0.069
Neg logit contributions
 mess-0.062
 neither-0.060
 fighting-0.052
 wasteful-0.050
 ruining-0.050
Token
Token ID198
Feature activation+7.78e-02

Pos logit contributions
 neither+0.043
 mess+0.042
 wasteful+0.042
 overflow+0.039
 ruining+0.038
Neg logit contributions
 name-0.028
 aboard-0.026
leaf-0.026
tie-0.026
dot-0.025
Token
Token ID198
Feature activation+1.24e+00

Pos logit contributions
 neither+0.003
 mess+0.003
 wasteful+0.003
 overflow+0.003
 ruining+0.003
Neg logit contributions
ka-0.003
tie-0.003
 name-0.003
dot-0.003
ance-0.003
TokenJohn
Token ID7554
Feature activation-7.98e-01

Pos logit contributions
 neither+0.058
 mess+0.054
 wasteful+0.052
 overflow+0.049
 snakes+0.047
Neg logit contributions
leaf-0.044
tie-0.044
shirt-0.043
yo-0.042
dot-0.042
Token,
Token ID11
Feature activation+8.27e-01

Pos logit contributions
ka+0.010
tie+0.010
 name+0.010
ance+0.010
leaf+0.010
Neg logit contributions
 neither-0.008
 mess-0.008
 wasteful-0.007
 fighting-0.007
 overflow-0.007
Token Tim
Token ID5045
Feature activation+6.09e-01

Pos logit contributions
 mess+0.026
 wasteful+0.025
 neither+0.023
 ruining+0.020
 overflow+0.020
Neg logit contributions
 name-0.040
tie-0.040
ka-0.039
 aboard-0.038
dot-0.038
Tokenmy
Token ID1820
Feature activation+1.19e+00

Pos logit contributions
 neither+0.030
 wasteful+0.028
 mess+0.028
 overflow+0.026
 ruining+0.026
Neg logit contributions
ka-0.019
 name-0.018
tie-0.018
ance-0.018
orn-0.017
Token found
Token ID1043
Feature activation-8.34e-01

Pos logit contributions
 neither+0.055
 wasteful+0.053
 mess+0.049
 ruining+0.049
 overflow+0.047
Neg logit contributions
 name-0.040
ka-0.038
tie-0.038
 aboard-0.037
dot-0.036
Token a
Token ID257
Feature activation-1.37e+00

Pos logit contributions
ka+0.032
tie+0.031
 name+0.031
dot+0.031
anced+0.031
Neg logit contributions
 neither-0.035
 mess-0.033
 wasteful-0.032
 fighting-0.029
 rats-0.029
Token shiny
Token ID22441
Feature activation-3.08e+00

Pos logit contributions
ance+0.064
anced+0.062
ka+0.059
dot+0.058
leaf+0.058
Neg logit contributions
 mess-0.055
 neither-0.048
 wasteful-0.045
 fighting-0.043
 overflow-0.042
Token rock
Token ID3881
Feature activation-9.34e-01

Pos logit contributions
anced+0.137
leaf+0.137
ance+0.136
dot+0.127
found+0.125
Neg logit contributions
 mess-0.137
 floor-0.107
 neither-0.105
 wasteful-0.098
 fighting-0.097
Token on
Token ID319
Feature activation+3.74e-01

Pos logit contributions
ka+0.040
 name+0.038
ance+0.038
leaf+0.038
anced+0.038
Neg logit contributions
 neither-0.037
 mess-0.036
 wasteful-0.032
 fighting-0.032
 ruining-0.031
Token the
Token ID262
Feature activation-2.30e-01

Pos logit contributions
 neither+0.015
 mess+0.015
 wasteful+0.015
 overflow+0.013
 ruining+0.013
Neg logit contributions
 name-0.014
ka-0.014
tie-0.014
ance-0.013
 aboard-0.013
Token ground
Token ID2323
Feature activation+3.28e-01

Pos logit contributions
ka+0.013
tie+0.013
 name+0.012
ance+0.012
dot+0.012
Neg logit contributions
 mess-0.006
 neither-0.006
 wasteful-0.005
 overflow-0.004
 ruining-0.004
Token.
Token ID13
Feature activation-2.40e-01

Pos logit contributions
 neither+0.015
 mess+0.014
 wasteful+0.014
 overflow+0.013
 ruining+0.013
Neg logit contributions
tie-0.011
ka-0.011
 name-0.011
ance-0.011
dot-0.011
Token
Token ID8151
Feature activation+2.98e-01

Pos logit contributions
ka+0.025
tie+0.025
 name+0.024
leaf+0.024
ance+0.024
Neg logit contributions
 neither-0.028
 mess-0.027
 wasteful-0.025
 fighting-0.024
 roughly-0.024
Tokens
Token ID82
Feature activation-1.78e+00

Pos logit contributions
 mess+0.013
 neither+0.013
 wasteful+0.012
 overflow+0.011
 ruining+0.011
Neg logit contributions
ka-0.011
tie-0.010
 name-0.010
ance-0.010
dot-0.010
Token new
Token ID649
Feature activation-1.98e+00

Pos logit contributions
ka+0.082
dot+0.078
orn+0.077
tie+0.076
shirt+0.075
Neg logit contributions
 neither-0.065
 mess-0.064
 fighting-0.058
 not-0.054
 ruining-0.052
Token skill
Token ID5032
Feature activation-1.61e+00

Pos logit contributions
ance+0.083
ka+0.083
ocket+0.083
tie+0.081
anced+0.081
Neg logit contributions
 mess-0.081
 neither-0.070
 fighting-0.064
 quarrel-0.064
 mistake-0.062
Token.
Token ID13
Feature activation-1.80e+00

Pos logit contributions
ka+0.072
 name+0.070
dot+0.069
Aid+0.067
tie+0.067
Neg logit contributions
 neither-0.058
 mess-0.053
 fighting-0.047
 wasteful-0.047
 roughly-0.045
Token<|endoftext|>
Token ID50256
Feature activation-2.78e+00

Pos logit contributions
ka+0.091
tie+0.090
ocket+0.086
 powers+0.085
leaf+0.085
Neg logit contributions
 neither-0.054
 mess-0.045
 fighting-0.042
 wasteful-0.042
 Amy-0.039
TokenOne
Token ID3198
Feature activation-1.00e+00

Pos logit contributions
tie+0.129
ocket+0.128
leaf+0.128
ka+0.127
dot+0.126
Neg logit contributions
 mess-0.080
 neither-0.078
 fighting-0.076
 Amy-0.069
 mice-0.067
Token day
Token ID1110
Feature activation-8.17e-01

Pos logit contributions
tie+0.048
ka+0.047
ance+0.046
anced+0.046
leaf+0.046
Neg logit contributions
 neither-0.033
 mess-0.032
 wasteful-0.028
 fighting-0.026
 roughly-0.026
Token,
Token ID11
Feature activation-1.11e+00

Pos logit contributions
tie+0.040
leaf+0.039
ka+0.039
dot+0.038
ance+0.038
Neg logit contributions
 neither-0.026
 mess-0.024
 wasteful-0.021
 fighting-0.021
 roughly-0.020
Token a
Token ID257
Feature activation-1.27e+00

Pos logit contributions
ance+0.046
leaf+0.046
ka+0.045
ocket+0.045
tie+0.044
Neg logit contributions
 neither-0.044
 mess-0.038
 fighting-0.036
 wasteful-0.036
 roughly-0.035
Token 3
Token ID513
Feature activation+4.43e-01

Pos logit contributions
ance+0.073
anced+0.072
tie+0.069
ka+0.068
 aboard+0.067
Neg logit contributions
 mess-0.032
 neither-0.031
 wasteful-0.028
 fighting-0.024
 roughly-0.023