TOP ACTIVATIONS
MAX = 0.090

he was still wearing his muddy blue socks and said ,
. It 's not your fault . Maybe we can play
bird . You admit your fault and ask for mercy .
" It 's not your fault , my d arl ings
. It 's not your fault . Fl uffy was old
. It 's not your fault . The rope was old
loved to play in his sandbox with his sp ade .
. It 's not my fault . I was too little
" It 's not my fault . I can 't help
. He was wearing his brand new icy blue jacket as
it was Tim my 's fault but Tim my felt guilty
so you can build your sandbox ." So
" It 's your fault !" Lily said .
" It 's your fault , Sam !" Tom shouted
make something special in his sandbox . He grabbed a p
caul iflower growing near her sandbox . She was very excited
lunch , playing in their sandbox and running around in the
and it was his own fault . Johnny smiled
" It 's your fault , Ben !" Lily said
was still wearing her p aj amas .

MIN ACTIVATIONS
MIN = -0.113

and bumped against the books hel ves . He smiled as
was ready , the b unn ies hopped over to the
on the row of books hel ves . <|endoftext|> Tom was
spotted a row of books hel ves in her grandparents '
walked over to the books hel ves and began to read
park , the g az eb o would be there waiting
sometimes , even the l itt lest things can lead to
one of the happiest re ind eers in the North Pole
He was the happiest re ind eer in the North Pole
. It was called " cham om ile " and it
quiet compared to the noisy se ag ull s outside .
the shore , and the se ag ull s flying in
on top . The b unn ies wanted to have a
did not feel the water se eping in . They did
apped together until all the se ag ull s flew away
. She used a m itt to take out the pan
very special toy for L ott ie . It was a
the surf and saw the se ag ull up in the
So the b unn ies and Bob enjoyed some
watched the waves and the se ag ull s . They

INTERVAL 0.039 - 0.090
CONTAINS 0.474%

He saw Jill with her math book . He did not
like to play with their toy cars . They have many
loud that it hurt his friend 's ears . His friends
the sink and wash your doll 's arm ."
knew it was going to rain soon ! S

INTERVAL -0.012 - 0.039
CONTAINS 82.191%

isn 't yours . Always be honest and never lie .
The dad tried to fix the truck but it
boy to have a pistol . John thought
� � Do you want to try doing some math ?
that he started spinning and tw irling around the yard .

INTERVAL -0.062 - -0.012
CONTAINS 17.281%

Sam was sad . She wanted to help him . She
friendly squirrel . " I have a plan . Follow me
his mission to help any animal who was lost . He
through the park , L ola saw something special - a
small ladder inside so that birds could climb in and out

INTERVAL -0.113 - -0.062
CONTAINS 0.055%

The moral of the story is that sometimes we need
want to see the b unn ies . They are the
measure again to see how tall it is . They hope
baby frog ! T ad smiled and carefully picked
. They learned that being kind and loving was better .
Token he
Token ID339
Feature activation+2.05e-02

Pos logit contributions
 brand+0.002
uit+0.002
ieved+0.002
but+0.002
 exactly+0.002
Neg logit contributions
 nation-0.001
 e-0.001
 person-0.001
 knock-0.001
 guys-0.001
Token was
Token ID373
Feature activation+1.85e-03

Pos logit contributions
but+0.045
uit+0.045
 exactly+0.041
 brand+0.040
ags+0.040
Neg logit contributions
 knock-0.014
ce-0.011
 ever-0.011
 nation-0.010
unts-0.010
Token still
Token ID991
Feature activation+1.78e-02

Pos logit contributions
uit+0.003
but+0.003
 brand+0.003
 exactly+0.003
 perched+0.003
Neg logit contributions
 knock-0.002
 nation-0.002
unts-0.002
 e-0.002
 person-0.002
Token wearing
Token ID5762
Feature activation+1.59e-02

Pos logit contributions
uit+0.031
but+0.031
ishly+0.028
ieved+0.028
Here+0.028
Neg logit contributions
 nation-0.018
 person-0.017
 e-0.017
 knock-0.017
ot-0.017
Token his
Token ID465
Feature activation+3.33e-02

Pos logit contributions
but+0.029
ieved+0.027
 perched+0.027
shadow+0.026
sometimes+0.025
Neg logit contributions
 turns-0.016
 guys-0.016
ot-0.015
 wind-0.015
 breaks-0.014
Token muddy
Token ID39085
Feature activation+8.96e-02

Pos logit contributions
 perched+0.063
but+0.059
sometimes+0.059
 exactly+0.058
 altogether+0.056
Neg logit contributions
 guys-0.033
ot-0.033
ia-0.031
ubby-0.030
ce-0.030
Token blue
Token ID4171
Feature activation+4.95e-02

Pos logit contributions
 perched+0.161
 exactly+0.153
sometimes+0.146
but+0.145
ieved+0.140
Neg logit contributions
ot-0.108
 guys-0.102
ia-0.093
 wind-0.089
 team-0.086
Token socks
Token ID24359
Feature activation+5.33e-02

Pos logit contributions
 perched+0.090
 exactly+0.086
ieved+0.083
 unconscious+0.082
but+0.081
Neg logit contributions
 guys-0.054
ot-0.053
 team-0.051
ia-0.049
ce-0.049
Token and
Token ID290
Feature activation-5.40e-04

Pos logit contributions
 ang+0.105
 perched+0.103
shadow+0.100
Here+0.100
 elbows+0.098
Neg logit contributions
 ever-0.036
 person-0.027
 nation-0.027
 wind-0.026
 of-0.026
Token said
Token ID531
Feature activation+2.18e-03

Pos logit contributions
 knock+0.001
ot+0.001
 grown+0.000
 nation+0.000
 e+0.000
Neg logit contributions
uit-0.001
but-0.001
ieved-0.001
 atop-0.001
 altogether-0.001
Token,
Token ID11
Feature activation-1.52e-03

Pos logit contributions
uit+0.003
 brand+0.003
 exactly+0.003
but+0.003
 unconscious+0.003
Neg logit contributions
 nation-0.003
 knock-0.002
ce-0.002
 Find-0.002
 guys-0.002
Token.
Token ID13
Feature activation-1.08e-03

Pos logit contributions
 brand+0.019
 exactly+0.018
uit+0.018
but+0.017
 perched+0.016
Neg logit contributions
 nation-0.023
ce-0.023
unts-0.022
 knock-0.022
ia-0.022
Token It
Token ID632
Feature activation-1.60e-02

Pos logit contributions
 nation+0.001
ce+0.001
unts+0.001
ye+0.001
 knock+0.001
Neg logit contributions
 brand-0.002
but-0.002
 exactly-0.002
uit-0.002
 perched-0.002
Token's
Token ID338
Feature activation-1.58e-02

Pos logit contributions
 nation+0.019
ass+0.018
unts+0.018
ce+0.017
ye+0.017
Neg logit contributions
 exactly-0.025
 brand-0.024
 perched-0.023
but-0.023
 struggling-0.022
Token not
Token ID407
Feature activation+7.19e-03

Pos logit contributions
 knock+0.019
 blast+0.019
 nation+0.018
 wind+0.018
 spe+0.018
Neg logit contributions
but-0.025
uit-0.023
ishly-0.022
 brand-0.022
iddled-0.021
Token your
Token ID534
Feature activation+2.95e-02

Pos logit contributions
but+0.013
uit+0.012
 brand+0.011
ishly+0.011
 exactly+0.011
Neg logit contributions
 guys-0.008
 knock-0.008
 nation-0.008
 blast-0.007
 wind-0.007
Token fault
Token ID8046
Feature activation+8.84e-02

Pos logit contributions
but+0.044
 perched+0.044
uit+0.043
 exactly+0.043
ieved+0.042
Neg logit contributions
 nation-0.042
 knock-0.039
 blast-0.037
 guys-0.036
 person-0.035
Token.
Token ID13
Feature activation-7.92e-04

Pos logit contributions
 brand+0.136
uit+0.126
 exactly+0.126
 perched+0.120
but+0.117
Neg logit contributions
 guys-0.110
 knock-0.108
unts-0.108
 nation-0.108
ia-0.105
Token Maybe
Token ID6674
Feature activation-1.03e-02

Pos logit contributions
 nation+0.001
ce+0.001
ye+0.001
 knock+0.001
 guys+0.001
Neg logit contributions
but-0.001
 brand-0.001
 exactly-0.001
uit-0.001
sometimes-0.001
Token we
Token ID356
Feature activation-6.48e-03

Pos logit contributions
 nation+0.011
 knock+0.010
ce+0.009
 blast+0.009
unts+0.009
Neg logit contributions
 brand-0.018
 exactly-0.017
uit-0.017
but-0.017
 perched-0.016
Token can
Token ID460
Feature activation-1.54e-02

Pos logit contributions
 nation+0.005
 knock+0.004
 guys+0.004
 blast+0.004
unts+0.004
Neg logit contributions
 brand-0.013
uit-0.013
 exactly-0.013
but-0.013
 perched-0.012
Token play
Token ID711
Feature activation+1.17e-02

Pos logit contributions
 nation+0.019
 knock+0.018
 blast+0.017
unts+0.017
ce+0.017
Neg logit contributions
 brand-0.024
 exactly-0.023
but-0.022
uit-0.022
 perched-0.022
Token bird
Token ID6512
Feature activation-2.58e-03

Pos logit contributions
 brand+0.004
 exactly+0.004
uit+0.004
but+0.004
 perched+0.004
Neg logit contributions
 nation-0.002
 knock-0.002
ce-0.002
unts-0.002
 spe-0.002
Token.
Token ID13
Feature activation+5.93e-04

Pos logit contributions
 nation+0.004
ce+0.003
ia+0.003
ot+0.003
ass+0.003
Neg logit contributions
 brand-0.004
but-0.003
uit-0.003
 exactly-0.003
 unconscious-0.003
Token You
Token ID921
Feature activation+1.24e-02

Pos logit contributions
 exactly+0.001
 brand+0.001
but+0.001
uit+0.001
 perched+0.001
Neg logit contributions
 nation-0.001
 knock-0.001
unts-0.001
ce-0.001
 blast-0.001
Token admit
Token ID9159
Feature activation+1.03e-02

Pos logit contributions
 brand+0.019
 exactly+0.019
but+0.018
uit+0.018
 perched+0.018
Neg logit contributions
 nation-0.014
 knock-0.013
ce-0.013
unts-0.013
 blast-0.013
Token your
Token ID534
Feature activation+2.78e-02

Pos logit contributions
 brand+0.017
uit+0.017
but+0.016
 exactly+0.016
ieved+0.016
Neg logit contributions
 nation-0.011
 knock-0.011
 guys-0.011
 Find-0.010
ce-0.010
Token fault
Token ID8046
Feature activation+8.70e-02

Pos logit contributions
uit+0.056
but+0.055
 exactly+0.054
ieved+0.054
 perched+0.054
Neg logit contributions
 nation-0.025
unts-0.021
 person-0.020
ot-0.020
 knock-0.020
Token and
Token ID290
Feature activation+1.01e-02

Pos logit contributions
 brand+0.167
Here+0.154
 ang+0.153
 inspected+0.150
 perched+0.149
Neg logit contributions
 ever-0.085
 guys-0.079
ot-0.075
 team-0.073
unts-0.073
Token ask
Token ID1265
Feature activation-2.18e-02

Pos logit contributions
uit+0.019
but+0.019
 brand+0.018
 exactly+0.018
 altogether+0.017
Neg logit contributions
 nation-0.010
 knock-0.009
 Find-0.009
 goes-0.009
 comes-0.008
Token for
Token ID329
Feature activation+1.10e-02

Pos logit contributions
 guys+0.021
 nation+0.021
ot+0.020
ce+0.019
 knock+0.019
Neg logit contributions
 brand-0.041
 perched-0.038
but-0.037
iddled-0.036
uit-0.036
Token mercy
Token ID17703
Feature activation+3.34e-03

Pos logit contributions
but+0.021
 brand+0.020
uit+0.020
 exactly+0.020
 perched+0.019
Neg logit contributions
 knock-0.010
 blast-0.010
 nation-0.009
 person-0.008
 Find-0.008
Token.
Token ID13
Feature activation-2.38e-03

Pos logit contributions
 brand+0.006
uit+0.005
 exactly+0.005
but+0.005
 perched+0.005
Neg logit contributions
 nation-0.004
unts-0.004
 ever-0.004
 team-0.004
ass-0.004
Token"
Token ID1
Feature activation+1.10e-02

Pos logit contributions
 brand+0.003
 exactly+0.003
uit+0.003
but+0.003
 perched+0.003
Neg logit contributions
 nation-0.002
 knock-0.002
ce-0.002
unts-0.002
 blast-0.002
TokenIt
Token ID1026
Feature activation+5.49e-03

Pos logit contributions
 exactly+0.017
 brand+0.016
uit+0.016
 perched+0.015
but+0.015
Neg logit contributions
 nation-0.013
 knock-0.013
ce-0.013
 spe-0.013
unts-0.013
Token's
Token ID338
Feature activation-8.00e-03

Pos logit contributions
 brand+0.011
uit+0.011
but+0.011
 exactly+0.011
 perched+0.010
Neg logit contributions
 knock-0.004
 nation-0.003
ce-0.003
 happens-0.003
 guys-0.003
Token not
Token ID407
Feature activation+1.16e-02

Pos logit contributions
 guys+0.011
 nation+0.011
 knock+0.011
 wind+0.011
 storm+0.011
Neg logit contributions
but-0.011
uit-0.010
 brand-0.010
iddled-0.009
 perched-0.009
Token your
Token ID534
Feature activation+3.27e-02

Pos logit contributions
but+0.023
uit+0.021
 brand+0.021
 exactly+0.019
ishly+0.019
Neg logit contributions
 guys-0.012
 nation-0.011
 wind-0.010
unts-0.010
 knock-0.010
Token fault
Token ID8046
Feature activation+8.64e-02

Pos logit contributions
ieved+0.057
but+0.057
iddled+0.056
uit+0.053
 brand+0.051
Neg logit contributions
 nation-0.038
 person-0.035
 spe-0.035
 guys-0.035
 knock-0.035
Token,
Token ID11
Feature activation+6.63e-03

Pos logit contributions
 brand+0.147
 exactly+0.136
uit+0.135
 perched+0.129
but+0.128
Neg logit contributions
 guys-0.097
unts-0.093
unn-0.093
 nation-0.092
 knock-0.092
Token my
Token ID616
Feature activation+3.39e-02

Pos logit contributions
 exactly+0.010
 brand+0.010
uit+0.009
but+0.009
 perched+0.009
Neg logit contributions
 nation-0.008
unts-0.008
ce-0.008
ass-0.008
 knock-0.007
Token d
Token ID288
Feature activation-3.99e-03

Pos logit contributions
but+0.069
uit+0.065
ieved+0.064
 unconscious+0.062
iddled+0.062
Neg logit contributions
 spe-0.029
 nation-0.028
 guys-0.026
 person-0.025
 knock-0.025
Tokenarl
Token ID7063
Feature activation-3.59e-02

Pos logit contributions
 nation+0.005
 knock+0.005
 blast+0.004
ce+0.004
 spe+0.004
Neg logit contributions
 exactly-0.006
uit-0.006
 brand-0.006
but-0.006
 perched-0.005
Tokenings
Token ID654
Feature activation+3.51e-02

Pos logit contributions
 nation+0.060
 knock+0.057
ce+0.055
 blast+0.055
ass+0.055
Neg logit contributions
 exactly-0.038
 brand-0.038
uit-0.036
but-0.036
 perched-0.034
Token.
Token ID13
Feature activation-1.70e-03

Pos logit contributions
 brand+0.012
but+0.011
 exactly+0.011
uit+0.011
 perched+0.010
Neg logit contributions
unts-0.007
 nation-0.007
 team-0.007
 comes-0.007
ot-0.007
Token It
Token ID632
Feature activation-1.44e-02

Pos logit contributions
 nation+0.002
 knock+0.002
ce+0.002
unts+0.002
ass+0.002
Neg logit contributions
 brand-0.003
 exactly-0.003
but-0.003
uit-0.003
 perched-0.003
Token's
Token ID338
Feature activation-1.54e-02

Pos logit contributions
 nation+0.014
 knock+0.013
ce+0.012
unts+0.012
ass+0.012
Neg logit contributions
 brand-0.026
 exactly-0.025
but-0.024
uit-0.024
 perched-0.024
Token not
Token ID407
Feature activation+3.84e-03

Pos logit contributions
 wind+0.018
 nation+0.018
ia+0.017
 blast+0.017
 knock+0.017
Neg logit contributions
but-0.026
uit-0.023
iddled-0.023
 brand-0.023
 inspected-0.021
Token your
Token ID534
Feature activation+2.71e-02

Pos logit contributions
but+0.008
uit+0.007
 brand+0.007
 exactly+0.007
ishly+0.007
Neg logit contributions
 guys-0.004
 nation-0.003
 wind-0.003
unts-0.003
 knock-0.003
Token fault
Token ID8046
Feature activation+8.60e-02

Pos logit contributions
but+0.048
ieved+0.048
iddled+0.047
 brand+0.047
 exactly+0.046
Neg logit contributions
 nation-0.033
 team-0.028
 guys-0.027
 person-0.027
 knock-0.027
Token.
Token ID13
Feature activation-2.32e-03

Pos logit contributions
 brand+0.146
 exactly+0.130
 perched+0.129
uit+0.128
 inspected+0.124
Neg logit contributions
 guys-0.104
 nation-0.098
 ever-0.098
 team-0.098
unts-0.096
Token Fl
Token ID1610
Feature activation-1.05e-02

Pos logit contributions
 nation+0.003
 knock+0.003
ce+0.003
 guys+0.003
ye+0.003
Neg logit contributions
 brand-0.003
but-0.003
 exactly-0.003
uit-0.003
 unconscious-0.003
Tokenuffy
Token ID15352
Feature activation+2.51e-02

Pos logit contributions
 knock+0.016
 spe+0.016
 Find+0.016
ye+0.016
ce+0.015
Neg logit contributions
 exactly-0.011
uit-0.011
 brand-0.011
but-0.010
 struggling-0.010
Token was
Token ID373
Feature activation-5.73e-03

Pos logit contributions
 brand+0.055
but+0.055
uit+0.055
 exactly+0.054
 perched+0.049
Neg logit contributions
 nation-0.015
 knock-0.014
 comes-0.014
 breaks-0.013
 wakes-0.013
Token old
Token ID1468
Feature activation-1.32e-02

Pos logit contributions
 nation+0.007
 knock+0.006
 spe+0.006
unts+0.006
 blast+0.006
Neg logit contributions
but-0.009
uit-0.009
 brand-0.008
 exactly-0.008
sometimes-0.008
Token.
Token ID13
Feature activation+2.40e-04

Pos logit contributions
 nation+0.021
 knock+0.020
ce+0.020
unts+0.020
ass+0.020
Neg logit contributions
 brand-0.013
 exactly-0.013
but-0.012
uit-0.012
 perched-0.011
Token It
Token ID632
Feature activation-1.63e-02

Pos logit contributions
 exactly+0.000
 brand+0.000
but+0.000
uit+0.000
 perched+0.000
Neg logit contributions
 nation-0.000
unts-0.000
 knock-0.000
ce-0.000
 blast-0.000
Token's
Token ID338
Feature activation-1.94e-02

Pos logit contributions
 nation+0.017
unts+0.015
ce+0.015
 knock+0.015
ass+0.015
Neg logit contributions
 exactly-0.028
 brand-0.027
but-0.026
uit-0.026
 perched-0.026
Token not
Token ID407
Feature activation+2.41e-03

Pos logit contributions
 knock+0.023
 nation+0.023
 wind+0.022
 spe+0.022
 blast+0.022
Neg logit contributions
but-0.031
uit-0.028
 brand-0.028
iddled-0.027
ishly-0.026
Token your
Token ID534
Feature activation+2.48e-02

Pos logit contributions
but+0.004
uit+0.004
 brand+0.004
ishly+0.004
iddled+0.004
Neg logit contributions
 guys-0.003
 nation-0.003
 knock-0.002
 wind-0.002
 blast-0.002
Token fault
Token ID8046
Feature activation+8.51e-02

Pos logit contributions
but+0.046
iddled+0.045
ieved+0.045
uit+0.043
 brand+0.043
Neg logit contributions
 nation-0.029
 person-0.025
 team-0.024
 knock-0.023
 guys-0.023
Token.
Token ID13
Feature activation+4.05e-04

Pos logit contributions
 brand+0.152
 exactly+0.139
uit+0.139
 perched+0.137
Here+0.135
Neg logit contributions
 guys-0.093
 ever-0.088
 nation-0.085
unts-0.084
ot-0.083
Token The
Token ID383
Feature activation+1.52e-03

Pos logit contributions
 exactly+0.001
 brand+0.001
but+0.001
uit+0.001
 perched+0.001
Neg logit contributions
 nation-0.001
 knock-0.000
ce-0.000
unts-0.000
 guys-0.000
Token rope
Token ID17182
Feature activation+6.32e-03

Pos logit contributions
but+0.003
 exactly+0.003
ieved+0.003
iddled+0.003
 perched+0.003
Neg logit contributions
 nation-0.002
 storm-0.001
 wind-0.001
 person-0.001
 knock-0.001
Token was
Token ID373
Feature activation-1.47e-02

Pos logit contributions
 brand+0.015
but+0.014
uit+0.014
 exactly+0.014
nam+0.014
Neg logit contributions
 knock-0.003
 comes-0.003
 goes-0.003
 nation-0.003
 breaks-0.003
Token old
Token ID1468
Feature activation-2.11e-02

Pos logit contributions
 spe+0.017
 knock+0.017
 blast+0.016
 wind+0.016
 nation+0.016
Neg logit contributions
but-0.025
uit-0.023
ishly-0.021
sometimes-0.021
 brand-0.021
Token loved
Token ID6151
Feature activation+1.44e-02

Pos logit contributions
 brand+0.003
 exactly+0.003
but+0.003
 perched+0.003
 struggling+0.003
Neg logit contributions
 nation-0.006
ass-0.006
 pm-0.006
 Danger-0.006
ye-0.006
Token to
Token ID284
Feature activation-6.20e-03

Pos logit contributions
 brand+0.015
 exactly+0.014
but+0.013
uit+0.013
 perched+0.013
Neg logit contributions
 nation-0.025
ass-0.024
ce-0.024
unts-0.023
 knock-0.023
Token play
Token ID711
Feature activation+2.71e-02

Pos logit contributions
 nation+0.008
ce+0.007
unts+0.007
ass+0.007
 knock+0.007
Neg logit contributions
 brand-0.009
 exactly-0.009
uit-0.009
but-0.009
 perched-0.008
Token in
Token ID287
Feature activation+3.44e-03

Pos logit contributions
 exactly+0.030
 brand+0.030
uit+0.028
but+0.028
 perched+0.027
Neg logit contributions
 nation-0.044
 knock-0.042
 blast-0.041
ce-0.041
 guys-0.041
Token his
Token ID465
Feature activation+7.91e-03

Pos logit contributions
 brand+0.003
 exactly+0.003
uit+0.003
 perched+0.003
but+0.003
Neg logit contributions
 nation-0.006
oke-0.006
ass-0.006
unts-0.006
unn-0.006
Token sandbox
Token ID35204
Feature activation+8.45e-02

Pos logit contributions
 brand+0.009
 exactly+0.008
uit+0.008
but+0.007
 unconscious+0.007
Neg logit contributions
ass-0.013
unts-0.013
 spe-0.013
ce-0.013
 nation-0.013
Token with
Token ID351
Feature activation+1.94e-04

Pos logit contributions
 brand+0.145
 perched+0.136
 unconscious+0.136
ishly+0.134
uit+0.134
Neg logit contributions
 nation-0.091
ia-0.080
 team-0.079
 knock-0.077
 person-0.077
Token his
Token ID465
Feature activation+3.93e-03

Pos logit contributions
 exactly+0.000
 brand+0.000
but+0.000
 perched+0.000
uit+0.000
Neg logit contributions
 nation-0.000
 knock-0.000
 guys-0.000
unts-0.000
 blast-0.000
Token sp
Token ID599
Feature activation+2.55e-02

Pos logit contributions
but+0.007
ieved+0.007
 exactly+0.007
 perched+0.007
uit+0.007
Neg logit contributions
 nation-0.004
 team-0.004
 blast-0.003
unts-0.003
 guys-0.003
Tokenade
Token ID671
Feature activation+4.48e-02

Pos logit contributions
 exactly+0.033
 brand+0.032
but+0.030
uit+0.028
 upon+0.025
Neg logit contributions
 nation-0.038
 spe-0.038
 knock-0.037
 blast-0.037
 Find-0.036
Token.
Token ID13
Feature activation+2.49e-03

Pos logit contributions
 brand+0.077
 perched+0.075
ieved+0.074
shadow+0.073
 unconscious+0.072
Neg logit contributions
 nation-0.050
 team-0.047
ia-0.046
 guys-0.045
ot-0.044
Token.
Token ID13
Feature activation+1.55e-03

Pos logit contributions
 nation+0.005
 knock+0.005
ce+0.005
unts+0.005
 blast+0.005
Neg logit contributions
 brand-0.009
 exactly-0.009
but-0.009
uit-0.009
 perched-0.009
Token It
Token ID632
Feature activation-1.96e-02

Pos logit contributions
 brand+0.003
 exactly+0.002
but+0.002
uit+0.002
 unconscious+0.002
Neg logit contributions
 nation-0.002
ce-0.002
unts-0.002
ye-0.002
ass-0.002
Token's
Token ID338
Feature activation-1.89e-02

Pos logit contributions
 nation+0.020
ce+0.018
 knock+0.018
unts+0.018
ass+0.018
Neg logit contributions
 exactly-0.034
 brand-0.034
but-0.032
uit-0.032
 perched-0.031
Token not
Token ID407
Feature activation+4.14e-03

Pos logit contributions
 nation+0.021
 knock+0.020
 spe+0.020
 guys+0.020
 blast+0.019
Neg logit contributions
but-0.032
uit-0.030
 brand-0.029
iddled-0.027
sometimes-0.026
Token my
Token ID616
Feature activation+4.51e-02

Pos logit contributions
but+0.008
uit+0.008
 brand+0.007
ishly+0.007
iddled+0.007
Neg logit contributions
 guys-0.004
 nation-0.004
 knock-0.004
 wind-0.003
 blast-0.003
Token fault
Token ID8046
Feature activation+8.40e-02

Pos logit contributions
but+0.093
ieved+0.091
iddled+0.090
uit+0.086
 perched+0.086
Neg logit contributions
 nation-0.040
 guys-0.039
 person-0.036
 team-0.033
 knock-0.033
Token.
Token ID13
Feature activation-1.13e-03

Pos logit contributions
 brand+0.152
 exactly+0.141
 perched+0.139
uit+0.138
Here+0.135
Neg logit contributions
 guys-0.089
 ever-0.082
unts-0.081
 nation-0.080
 team-0.079
Token I
Token ID314
Feature activation+1.30e-02

Pos logit contributions
 nation+0.001
ce+0.001
 knock+0.001
ye+0.001
 guys+0.001
Neg logit contributions
 brand-0.002
but-0.002
 exactly-0.002
uit-0.001
 unconscious-0.001
Token was
Token ID373
Feature activation-5.00e-03

Pos logit contributions
 brand+0.023
uit+0.022
 exactly+0.022
but+0.022
 unconscious+0.021
Neg logit contributions
 nation-0.013
 knock-0.013
unts-0.012
 blast-0.012
 Find-0.012
Token too
Token ID1165
Feature activation+3.63e-03

Pos logit contributions
 nation+0.007
 knock+0.007
 blast+0.007
unts+0.007
 spe+0.006
Neg logit contributions
uit-0.007
 brand-0.007
but-0.006
 exactly-0.006
 perched-0.006
Token little
Token ID1310
Feature activation-1.68e-02

Pos logit contributions
 brand+0.006
 exactly+0.006
but+0.005
uit+0.005
 perched+0.005
Neg logit contributions
 nation-0.004
 knock-0.004
ce-0.004
 blast-0.004
ass-0.004
Token "
Token ID366
Feature activation+1.32e-02

Pos logit contributions
uit+0.000
 brand+0.000
 exactly+0.000
 ease+0.000
 intact+0.000
Neg logit contributions
 knock-0.000
 e-0.000
 comes-0.000
 wind-0.000
oke-0.000
TokenIt
Token ID1026
Feature activation-9.41e-04

Pos logit contributions
 brand+0.019
 exactly+0.018
uit+0.017
 perched+0.016
 unconscious+0.016
Neg logit contributions
 nation-0.018
ce-0.017
ass-0.017
 knock-0.017
 Find-0.017
Token's
Token ID338
Feature activation-9.09e-03

Pos logit contributions
 nation+0.001
 knock+0.001
ce+0.001
unts+0.001
ass+0.001
Neg logit contributions
 brand-0.002
 exactly-0.002
but-0.002
uit-0.002
 perched-0.002
Token not
Token ID407
Feature activation+1.12e-02

Pos logit contributions
 nation+0.011
 knock+0.011
 e+0.011
 guys+0.011
 storm+0.011
Neg logit contributions
but-0.014
uit-0.013
 brand-0.013
 exactly-0.012
sometimes-0.011
Token my
Token ID616
Feature activation+4.70e-02

Pos logit contributions
but+0.020
uit+0.019
ishly+0.018
 brand+0.018
ieved+0.017
Neg logit contributions
 guys-0.012
 nation-0.011
 wind-0.011
 knock-0.011
 blast-0.010
Token fault
Token ID8046
Feature activation+8.32e-02

Pos logit contributions
ieved+0.084
uit+0.080
but+0.080
 exactly+0.079
 perched+0.078
Neg logit contributions
 nation-0.057
 guys-0.051
 person-0.051
 team-0.048
 storm-0.047
Token.
Token ID13
Feature activation+5.84e-03

Pos logit contributions
 brand+0.151
 exactly+0.138
 perched+0.135
uit+0.135
Here+0.133
Neg logit contributions
 guys-0.087
 nation-0.083
unts-0.077
 team-0.077
oke-0.077
Token I
Token ID314
Feature activation+1.81e-02

Pos logit contributions
but+0.007
 exactly+0.007
 brand+0.007
uit+0.007
 perched+0.007
Neg logit contributions
 nation-0.008
 guys-0.008
 knock-0.008
 spe-0.008
 blast-0.007
Token can
Token ID460
Feature activation-7.81e-03

Pos logit contributions
 brand+0.034
uit+0.033
but+0.033
 exactly+0.032
 unconscious+0.031
Neg logit contributions
 nation-0.017
 knock-0.016
ass-0.015
 blast-0.015
unts-0.015
Token't
Token ID470
Feature activation-2.88e-03

Pos logit contributions
 nation+0.007
 knock+0.007
 blast+0.006
ce+0.006
unts+0.006
Neg logit contributions
 brand-0.014
 exactly-0.014
but-0.013
uit-0.013
 perched-0.013
Token help
Token ID1037
Feature activation-1.27e-02

Pos logit contributions
 nation+0.003
 knock+0.003
 blast+0.003
ce+0.003
 guys+0.002
Neg logit contributions
 brand-0.005
 exactly-0.005
uit-0.005
but-0.005
 perched-0.005
Token.
Token ID13
Feature activation+1.69e-03

Pos logit contributions
 brand+0.028
 perched+0.028
 ang+0.028
 unconscious+0.028
 exactly+0.027
Neg logit contributions
 nation-0.027
 knock-0.025
 team-0.024
ce-0.024
 guys-0.023
Token He
Token ID679
Feature activation-3.18e-03

Pos logit contributions
uit+0.003
 exactly+0.003
 brand+0.003
 perched+0.003
but+0.003
Neg logit contributions
 knock-0.002
 nation-0.002
 Find-0.002
air-0.002
 Danger-0.002
Token was
Token ID373
Feature activation-3.56e-03

Pos logit contributions
 nation+0.003
 knock+0.003
ce+0.003
 blast+0.003
ot+0.003
Neg logit contributions
 brand-0.006
 exactly-0.006
uit-0.006
but-0.005
 unconscious-0.005
Token wearing
Token ID5762
Feature activation+1.65e-02

Pos logit contributions
 knock+0.005
 nation+0.005
 blast+0.005
ce+0.005
 e+0.005
Neg logit contributions
uit-0.005
but-0.005
 brand-0.005
 exactly-0.004
 altogether-0.004
Token his
Token ID465
Feature activation+2.69e-02

Pos logit contributions
but+0.027
 perched+0.025
sometimes+0.024
ishly+0.024
 unconscious+0.024
Neg logit contributions
 guys-0.021
 blast-0.021
 nation-0.019
 turns-0.019
 wind-0.018
Token brand
Token ID4508
Feature activation+8.30e-02

Pos logit contributions
 perched+0.055
sometimes+0.051
but+0.050
 exactly+0.050
uit+0.048
Neg logit contributions
 nation-0.026
 guys-0.025
 blast-0.023
 team-0.023
ot-0.022
Token new
Token ID649
Feature activation+6.04e-02

Pos logit contributions
 perched+0.125
 unconscious+0.122
uit+0.119
 struggling+0.118
ishly+0.117
Neg logit contributions
ia-0.091
 nation-0.091
ce-0.090
ot-0.090
 knock-0.090
Token icy
Token ID30284
Feature activation+5.05e-02

Pos logit contributions
 perched+0.109
sometimes+0.100
but+0.099
 exactly+0.097
 unconscious+0.096
Neg logit contributions
 nation-0.074
 guys-0.069
ot-0.066
 team-0.065
 blast-0.063
Token blue
Token ID4171
Feature activation+4.17e-02

Pos logit contributions
 exactly+0.076
 perched+0.076
but+0.070
sometimes+0.068
 brand+0.067
Neg logit contributions
 nation-0.068
 guys-0.068
ot-0.066
 knock-0.063
ia-0.062
Token jacket
Token ID15224
Feature activation+1.66e-02

Pos logit contributions
 perched+0.074
 exactly+0.070
ieved+0.068
sometimes+0.068
 ang+0.067
Neg logit contributions
 guys-0.048
ot-0.046
 nation-0.046
 team-0.044
 wake-0.042
Token as
Token ID355
Feature activation-2.28e-02

Pos logit contributions
 ang+0.030
Here+0.027
Acc+0.027
 elbows+0.027
 exactly+0.027
Neg logit contributions
 ever-0.015
 guys-0.015
ce-0.014
--0.014
 person-0.013
Token it
Token ID340
Feature activation+9.40e-04

Pos logit contributions
 brand+0.007
 exactly+0.007
uit+0.007
but+0.006
 perched+0.006
Neg logit contributions
 nation-0.007
 knock-0.007
ce-0.006
unts-0.006
 guys-0.006
Token was
Token ID373
Feature activation+2.00e-03

Pos logit contributions
 brand+0.001
 exactly+0.001
uit+0.001
but+0.001
 perched+0.001
Neg logit contributions
 nation-0.001
 knock-0.001
ce-0.001
unts-0.001
 guys-0.001
Token Tim
Token ID5045
Feature activation-4.86e-04

Pos logit contributions
but+0.003
uit+0.003
 brand+0.003
 perched+0.003
 exactly+0.002
Neg logit contributions
 guys-0.003
 knock-0.002
 storm-0.002
 nation-0.002
unts-0.002
Tokenmy
Token ID1820
Feature activation-1.29e-02

Pos logit contributions
ot+0.001
ny+0.001
oke+0.001
unts+0.001
ia+0.001
Neg logit contributions
 brand-0.001
 perched-0.001
but-0.001
 exactly-0.001
 atop-0.001
Token's
Token ID338
Feature activation-1.52e-02

Pos logit contributions
 nation+0.009
 knock+0.009
unts+0.008
ce+0.008
 Find+0.007
Neg logit contributions
 brand-0.026
uit-0.026
 exactly-0.025
but-0.025
 perched-0.024
Token fault
Token ID8046
Feature activation+8.29e-02

Pos logit contributions
 knock+0.014
 nation+0.014
ot+0.013
ia+0.013
 guys+0.013
Neg logit contributions
 exactly-0.027
 perched-0.027
uit-0.026
 brand-0.026
but-0.025
Token but
Token ID475
Feature activation+2.21e-02

Pos logit contributions
 brand+0.120
 exactly+0.111
 perched+0.111
uit+0.110
but+0.105
Neg logit contributions
unts-0.099
 nation-0.098
 knock-0.097
 guys-0.096
oke-0.093
Token Tim
Token ID5045
Feature activation+2.83e-02

Pos logit contributions
uit+0.035
but+0.034
 brand+0.033
 perched+0.032
 atop+0.030
Neg logit contributions
 knock-0.024
 nation-0.024
 guys-0.023
 Find-0.022
 person-0.022
Tokenmy
Token ID1820
Feature activation+2.16e-02

Pos logit contributions
 brand+0.049
but+0.047
 perched+0.042
nam+0.042
uit+0.042
Neg logit contributions
 comes-0.031
ny-0.030
oke-0.030
ia-0.029
unts-0.029
Token felt
Token ID2936
Feature activation+4.56e-03

Pos logit contributions
uit+0.038
 brand+0.038
but+0.036
 perched+0.035
 exactly+0.035
Neg logit contributions
 knock-0.020
unts-0.020
 nation-0.020
ith-0.019
oke-0.018
Token guilty
Token ID6717
Feature activation+3.47e-02

Pos logit contributions
uit+0.011
but+0.010
bat+0.009
 perched+0.009
 exactly+0.009
Neg logit contributions
 knock-0.003
 wind-0.003
unts-0.003
 Find-0.002
 happens-0.002
Token so
Token ID523
Feature activation+4.20e-03

Pos logit contributions
iddled+0.013
 struggling+0.013
 ang+0.012
 brand+0.012
 inspected+0.012
Neg logit contributions
 ever-0.008
 guys-0.007
 nation-0.007
 team-0.006
 go-0.006
Token you
Token ID345
Feature activation-5.78e-03

Pos logit contributions
 brand+0.007
 exactly+0.007
but+0.006
uit+0.006
 perched+0.006
Neg logit contributions
 nation-0.005
 knock-0.005
ce-0.004
unts-0.004
ass-0.004
Token can
Token ID460
Feature activation-1.13e-02

Pos logit contributions
 nation+0.005
 guys+0.005
 knock+0.005
 blast+0.005
ass+0.004
Neg logit contributions
but-0.011
 brand-0.011
 exactly-0.011
uit-0.010
 unconscious-0.010
Token build
Token ID1382
Feature activation+2.50e-02

Pos logit contributions
 nation+0.011
 knock+0.010
 blast+0.010
ce+0.010
 guys+0.010
Neg logit contributions
 brand-0.020
 exactly-0.020
but-0.019
uit-0.019
 perched-0.019
Token your
Token ID534
Feature activation+1.95e-02

Pos logit contributions
but+0.037
uit+0.035
 unconscious+0.034
 brand+0.033
 exactly+0.033
Neg logit contributions
 nation-0.034
 guys-0.033
 blast-0.031
 knock-0.030
 Find-0.029
Token sandbox
Token ID35204
Feature activation+8.27e-02

Pos logit contributions
but+0.035
uit+0.035
 exactly+0.034
 brand+0.033
sometimes+0.032
Neg logit contributions
 nation-0.022
 knock-0.019
 guys-0.018
 blast-0.018
 spe-0.017
Token."
Token ID526
Feature activation+3.33e-03

Pos logit contributions
 unconscious+0.135
 brand+0.127
uit+0.127
 struggling+0.124
iddled+0.122
Neg logit contributions
 nation-0.104
 guys-0.093
 knock-0.086
 team-0.085
ia-0.083
Token
Token ID220
Feature activation+6.97e-03

Pos logit contributions
 brand+0.006
 exactly+0.006
 struggling+0.005
 intact+0.005
uit+0.005
Neg logit contributions
 Find-0.003
 nation-0.003
ce-0.003
air-0.003
ell-0.003
Token
Token ID198
Feature activation+2.71e-03

Pos logit contributions
 brand+0.012
 exactly+0.011
uit+0.010
 perched+0.010
 struggling+0.010
Neg logit contributions
unts-0.008
ia-0.008
oke-0.008
ass-0.007
 Find-0.007
Token
Token ID198
Feature activation+5.11e-03

Pos logit contributions
 brand+0.004
 exactly+0.004
 struggling+0.004
 perched+0.004
uit+0.004
Neg logit contributions
ce-0.003
unts-0.003
ia-0.003
 knock-0.003
ot-0.003
TokenSo
Token ID2396
Feature activation+2.42e-02

Pos logit contributions
 brand+0.009
 exactly+0.008
 perched+0.007
uit+0.007
 unconscious+0.007
Neg logit contributions
 nation-0.006
ce-0.006
unts-0.006
 knock-0.006
 blast-0.006
Token
Token ID198
Feature activation-2.32e-03

Pos logit contributions
 nation+0.007
 knock+0.006
ce+0.006
unts+0.006
 blast+0.006
Neg logit contributions
 brand-0.004
 exactly-0.004
uit-0.004
but-0.004
 perched-0.004
Token"
Token ID1
Feature activation+5.17e-03

Pos logit contributions
 nation+0.003
 guys+0.003
 spe+0.003
ass+0.003
 knock+0.003
Neg logit contributions
 exactly-0.003
but-0.003
 brand-0.003
uit-0.003
 unconscious-0.003
TokenIt
Token ID1026
Feature activation+4.34e-03

Pos logit contributions
 exactly+0.007
 brand+0.007
uit+0.007
but+0.007
 perched+0.007
Neg logit contributions
 nation-0.007
 knock-0.007
unts-0.006
ce-0.006
 blast-0.006
Token's
Token ID338
Feature activation-1.02e-02

Pos logit contributions
 exactly+0.007
 brand+0.007
uit+0.007
but+0.007
 perched+0.007
Neg logit contributions
 nation-0.005
ce-0.004
unts-0.004
 knock-0.004
ass-0.004
Token your
Token ID534
Feature activation+2.31e-02

Pos logit contributions
 nation+0.016
 knock+0.015
ce+0.015
unts+0.015
 blast+0.014
Neg logit contributions
 brand-0.012
 exactly-0.012
but-0.012
uit-0.012
 perched-0.011
Token fault
Token ID8046
Feature activation+8.27e-02

Pos logit contributions
but+0.039
 perched+0.037
uit+0.036
 exactly+0.036
 brand+0.036
Neg logit contributions
 nation-0.028
 guys-0.026
 blast-0.025
 knock-0.024
ye-0.024
Token!"
Token ID2474
Feature activation+7.69e-03

Pos logit contributions
 brand+0.151
but+0.139
Here+0.139
uit+0.139
 exactly+0.138
Neg logit contributions
 guys-0.080
ia-0.078
 nation-0.076
 knock-0.076
unts-0.075
Token Lily
Token ID20037
Feature activation+1.78e-02

Pos logit contributions
 brand+0.016
 exactly+0.016
uit+0.015
but+0.015
 perched+0.015
Neg logit contributions
 nation-0.005
 knock-0.005
 blast-0.005
unts-0.005
 Find-0.005
Token said
Token ID531
Feature activation+1.86e-02

Pos logit contributions
 brand+0.025
 exactly+0.024
uit+0.023
but+0.022
 perched+0.021
Neg logit contributions
 knock-0.024
unts-0.024
ce-0.023
 nation-0.023
 e-0.022
Token.
Token ID13
Feature activation+2.08e-03

Pos logit contributions
 brand+0.022
 exactly+0.021
uit+0.020
 perched+0.019
but+0.019
Neg logit contributions
 nation-0.028
 knock-0.028
unts-0.028
ce-0.028
ot-0.027
Token
Token ID198
Feature activation-3.01e-03

Pos logit contributions
 exactly+0.003
 brand+0.003
uit+0.003
 perched+0.003
ieved+0.003
Neg logit contributions
unts-0.003
 knock-0.002
 Find-0.002
 nation-0.002
ce-0.002
Token
Token ID198
Feature activation-7.70e-03

Pos logit contributions
 nation+0.018
 knock+0.017
unts+0.017
ce+0.017
ass+0.016
Neg logit contributions
 brand-0.013
 exactly-0.013
uit-0.012
but-0.012
 perched-0.011
Token"
Token ID1
Feature activation+7.08e-03

Pos logit contributions
 nation+0.010
 guys+0.009
 spe+0.009
ass+0.009
 knock+0.009
Neg logit contributions
 exactly-0.011
but-0.011
 brand-0.011
uit-0.011
 unconscious-0.010
TokenIt
Token ID1026
Feature activation+4.60e-03

Pos logit contributions
 brand+0.012
 exactly+0.012
uit+0.011
but+0.011
 perched+0.011
Neg logit contributions
 nation-0.007
 knock-0.007
ce-0.007
unts-0.007
 blast-0.007
Token's
Token ID338
Feature activation-8.89e-03

Pos logit contributions
 brand+0.009
 exactly+0.009
uit+0.008
but+0.008
 perched+0.008
Neg logit contributions
 nation-0.004
 knock-0.004
ce-0.004
unts-0.003
 guys-0.003
Token your
Token ID534
Feature activation+2.54e-02

Pos logit contributions
 nation+0.014
 knock+0.014
 guys+0.014
 blast+0.014
unts+0.014
Neg logit contributions
 brand-0.010
but-0.010
uit-0.009
 exactly-0.009
 perched-0.008
Token fault
Token ID8046
Feature activation+8.25e-02

Pos logit contributions
ieved+0.037
but+0.037
 brand+0.036
iddled+0.036
uit+0.035
Neg logit contributions
 nation-0.032
ot-0.031
 spe-0.031
 person-0.031
 knock-0.030
Token,
Token ID11
Feature activation+2.81e-03

Pos logit contributions
 brand+0.152
uit+0.138
Here+0.138
 exactly+0.137
but+0.136
Neg logit contributions
 guys-0.078
ia-0.077
unts-0.075
 knock-0.074
 ever-0.073
Token Sam
Token ID3409
Feature activation+2.15e-03

Pos logit contributions
 exactly+0.005
 brand+0.005
uit+0.004
but+0.004
 perched+0.004
Neg logit contributions
 nation-0.003
ce-0.003
unts-0.003
 knock-0.003
 spe-0.003
Token!"
Token ID2474
Feature activation+1.01e-02

Pos logit contributions
 brand+0.003
but+0.003
 exactly+0.003
uit+0.003
 unconscious+0.003
Neg logit contributions
 knock-0.003
ce-0.003
 nation-0.003
unts-0.003
ia-0.002
Token Tom
Token ID4186
Feature activation+1.12e-02

Pos logit contributions
 brand+0.024
 perched+0.023
uit+0.023
 exactly+0.023
 atop+0.023
Neg logit contributions
 knock-0.005
 blast-0.005
ia-0.004
 wind-0.004
unts-0.004
Token shouted
Token ID17293
Feature activation+2.38e-02

Pos logit contributions
 brand+0.016
 exactly+0.015
uit+0.014
but+0.014
 perched+0.013
Neg logit contributions
 knock-0.016
ce-0.015
unts-0.015
ia-0.014
 nation-0.014
Token make
Token ID787
Feature activation+1.00e-02

Pos logit contributions
 brand+0.009
 exactly+0.009
uit+0.009
but+0.009
 perched+0.008
Neg logit contributions
 nation-0.006
ce-0.006
unts-0.006
ass-0.006
 knock-0.006
Token something
Token ID1223
Feature activation-5.64e-03

Pos logit contributions
but+0.011
uit+0.011
 brand+0.011
 exactly+0.010
 unconscious+0.010
Neg logit contributions
 nation-0.015
 guys-0.015
 knock-0.015
 blast-0.015
 e-0.015
Token special
Token ID2041
Feature activation-3.44e-03

Pos logit contributions
 knock+0.005
 nation+0.005
 dive+0.004
ye+0.004
 wake+0.004
Neg logit contributions
uit-0.012
 unconscious-0.011
ishly-0.011
ieved-0.010
but-0.010
Token in
Token ID287
Feature activation+2.79e-03

Pos logit contributions
 ever+0.003
 nation+0.003
 guys+0.002
 person+0.002
 dive+0.002
Neg logit contributions
 unconscious-0.007
 elbows-0.007
 ang-0.007
iddled-0.006
 brand-0.006
Token his
Token ID465
Feature activation+1.06e-02

Pos logit contributions
 brand+0.003
 exactly+0.003
but+0.003
uit+0.003
 perched+0.003
Neg logit contributions
 nation-0.004
 knock-0.004
 blast-0.004
ce-0.004
unts-0.004
Token sandbox
Token ID35204
Feature activation+8.21e-02

Pos logit contributions
 exactly+0.014
but+0.014
 brand+0.013
uit+0.013
 perched+0.013
Neg logit contributions
 nation-0.016
 knock-0.014
 blast-0.014
 Find-0.014
 guys-0.014
Token.
Token ID13
Feature activation-8.84e-04

Pos logit contributions
 brand+0.154
iddled+0.151
 unconscious+0.148
 aback+0.141
 perched+0.141
Neg logit contributions
 nation-0.084
 person-0.074
 team-0.072
ia-0.066
 guys-0.065
Token He
Token ID679
Feature activation-2.80e-04

Pos logit contributions
 nation+0.001
 knock+0.001
 Find+0.001
ce+0.001
 guys+0.001
Neg logit contributions
 exactly-0.001
 brand-0.001
uit-0.001
but-0.001
 perched-0.001
Token grabbed
Token ID13176
Feature activation+3.29e-02

Pos logit contributions
 nation+0.000
 knock+0.000
unts+0.000
ce+0.000
ass+0.000
Neg logit contributions
 exactly-0.000
 brand-0.000
uit-0.000
but-0.000
 perched-0.000
Token a
Token ID257
Feature activation-3.95e-03

Pos logit contributions
but+0.036
 brand+0.036
 exactly+0.035
uit+0.033
 perched+0.032
Neg logit contributions
 knock-0.051
 nation-0.051
 blast-0.050
 guys-0.050
 e-0.049
Token p
Token ID279
Feature activation+3.22e-02

Pos logit contributions
 nation+0.004
 knock+0.004
 blast+0.004
 spe+0.004
ye+0.004
Neg logit contributions
but-0.007
 exactly-0.007
sometimes-0.006
uit-0.006
 brand-0.006
Token caul
Token ID46087
Feature activation-4.82e-02

Pos logit contributions
 exactly+0.015
but+0.015
 brand+0.015
uit+0.015
 perched+0.014
Neg logit contributions
 nation-0.009
 knock-0.008
ce-0.007
 spe-0.007
 blast-0.007
Tokeniflower
Token ID42642
Feature activation-2.48e-02

Pos logit contributions
 nation+0.078
 blast+0.075
 knock+0.075
 guys+0.075
 Danger+0.075
Neg logit contributions
 brand-0.054
 exactly-0.054
but-0.051
 perched-0.051
 struggling-0.047
Token growing
Token ID3957
Feature activation-4.42e-03

Pos logit contributions
 nation+0.031
 person+0.029
 knock+0.029
ce+0.028
 e+0.028
Neg logit contributions
 brand-0.037
 ang-0.035
ieved-0.034
but-0.034
 aback-0.034
Token near
Token ID1474
Feature activation+1.15e-02

Pos logit contributions
 spe+0.006
ce+0.006
 nation+0.006
ith+0.005
air+0.005
Neg logit contributions
 brand-0.006
uit-0.006
but-0.006
 exactly-0.006
 unconscious-0.005
Token her
Token ID607
Feature activation+1.07e-02

Pos logit contributions
but+0.016
 brand+0.015
 exactly+0.015
ieved+0.015
sometimes+0.015
Neg logit contributions
 nation-0.015
 knock-0.014
 e-0.014
unts-0.014
ot-0.014
Token sandbox
Token ID35204
Feature activation+8.20e-02

Pos logit contributions
 exactly+0.019
uit+0.018
 brand+0.018
ieved+0.018
but+0.018
Neg logit contributions
 nation-0.012
 team-0.010
 knock-0.010
 person-0.010
unts-0.009
Token.
Token ID13
Feature activation+3.17e-03

Pos logit contributions
 brand+0.153
 aback+0.137
 unconscious+0.136
ieved+0.135
iddled+0.133
Neg logit contributions
 nation-0.084
ia-0.083
 person-0.077
 pm-0.070
 knock-0.068
Token She
Token ID1375
Feature activation+3.69e-03

Pos logit contributions
 exactly+0.005
 brand+0.005
 perched+0.005
uit+0.005
but+0.005
Neg logit contributions
 Find-0.004
 nation-0.004
ce-0.003
unts-0.003
 knock-0.003
Token was
Token ID373
Feature activation-2.55e-03

Pos logit contributions
 brand+0.006
 exactly+0.005
uit+0.005
but+0.005
 unconscious+0.005
Neg logit contributions
 nation-0.004
 knock-0.004
ce-0.004
unts-0.004
 blast-0.004
Token very
Token ID845
Feature activation+2.40e-02

Pos logit contributions
 knock+0.004
 nation+0.004
 e+0.004
unts+0.004
 spe+0.004
Neg logit contributions
uit-0.003
but-0.003
 brand-0.003
 exactly-0.003
 altogether-0.003
Token excited
Token ID6568
Feature activation-2.92e-03

Pos logit contributions
 exactly+0.035
 brand+0.035
but+0.033
uit+0.032
 perched+0.031
Neg logit contributions
 nation-0.031
ce-0.029
 knock-0.029
 blast-0.029
unts-0.028
Token lunch
Token ID9965
Feature activation+2.64e-02

Pos logit contributions
 nation+0.001
 guys+0.001
 knock+0.001
 team+0.001
 spe+0.001
Neg logit contributions
uit-0.001
but-0.001
 exactly-0.001
 perched-0.001
 brand-0.001
Token,
Token ID11
Feature activation-3.11e-04

Pos logit contributions
 perched+0.039
 brand+0.038
 exactly+0.037
 unconscious+0.037
 ang+0.037
Neg logit contributions
 nation-0.035
 guys-0.034
 team-0.032
ce-0.031
ia-0.031
Token playing
Token ID2712
Feature activation+2.26e-02

Pos logit contributions
 nation+0.000
 knock+0.000
 blast+0.000
 spe+0.000
 e+0.000
Neg logit contributions
 brand-0.000
uit-0.000
iddled-0.000
but-0.000
 exactly-0.000
Token in
Token ID287
Feature activation+1.20e-03

Pos logit contributions
 exactly+0.038
 brand+0.037
uit+0.036
but+0.036
 perched+0.035
Neg logit contributions
 nation-0.029
 guys-0.027
 team-0.026
 blast-0.025
ce-0.023
Token their
Token ID511
Feature activation-3.08e-03

Pos logit contributions
but+0.002
 exactly+0.002
uit+0.002
 brand+0.002
 perched+0.002
Neg logit contributions
 nation-0.001
 blast-0.001
 knock-0.001
unts-0.001
 spe-0.001
Token sandbox
Token ID35204
Feature activation+8.20e-02

Pos logit contributions
 nation+0.004
 knock+0.004
ce+0.004
unts+0.004
 blast+0.004
Neg logit contributions
 brand-0.004
 exactly-0.004
but-0.004
uit-0.004
 perched-0.004
Token and
Token ID290
Feature activation-4.15e-03

Pos logit contributions
 brand+0.150
 unconscious+0.140
 perched+0.139
iddled+0.139
 aback+0.137
Neg logit contributions
 nation-0.099
ia-0.095
 team-0.089
 person-0.079
ot-0.078
Token running
Token ID2491
Feature activation+1.23e-02

Pos logit contributions
 nation+0.006
 knock+0.006
 blast+0.005
 guys+0.005
 spe+0.005
Neg logit contributions
uit-0.006
but-0.006
 exactly-0.006
 brand-0.006
ieved-0.005
Token around
Token ID1088
Feature activation+7.73e-03

Pos logit contributions
 exactly+0.017
uit+0.017
 brand+0.017
but+0.016
sometimes+0.015
Neg logit contributions
 nation-0.018
 guys-0.017
 blast-0.016
 knock-0.016
ce-0.015
Token in
Token ID287
Feature activation+2.95e-03

Pos logit contributions
uit+0.013
 brand+0.012
 exactly+0.012
ieved+0.011
 reassured+0.011
Neg logit contributions
 nation-0.011
 team-0.009
 blast-0.009
 knock-0.009
 spe-0.009
Token the
Token ID262
Feature activation+7.80e-04

Pos logit contributions
but+0.005
 exactly+0.005
uit+0.005
 brand+0.005
 perched+0.004
Neg logit contributions
 nation-0.004
 blast-0.003
 knock-0.003
unts-0.003
 team-0.003
Token and
Token ID290
Feature activation-4.71e-03

Pos logit contributions
 guys+0.018
ce+0.018
 comes+0.018
 ever+0.017
air+0.017
Neg logit contributions
 brand-0.036
Here-0.034
 perched-0.034
 exactly-0.033
nam-0.032
Token it
Token ID340
Feature activation-7.88e-03

Pos logit contributions
 knock+0.006
 nation+0.005
 Find+0.005
 ever+0.005
 blast+0.005
Neg logit contributions
uit-0.007
but-0.007
 brand-0.007
 exactly-0.006
 perched-0.006
Token was
Token ID373
Feature activation-3.48e-03

Pos logit contributions
 knock+0.008
ce+0.007
 nation+0.007
 goes+0.007
 happens+0.007
Neg logit contributions
 brand-0.014
uit-0.013
 exactly-0.013
but-0.013
 thanking-0.012
Token his
Token ID465
Feature activation+5.02e-03

Pos logit contributions
 time+0.003
 wind+0.003
 knock+0.003
 ever+0.003
 storm+0.003
Neg logit contributions
but-0.006
ishly-0.005
uit-0.005
oding-0.005
uddled-0.005
Token own
Token ID898
Feature activation+1.52e-02

Pos logit contributions
 perched+0.010
but+0.010
uit+0.009
ieved+0.009
 exactly+0.009
Neg logit contributions
 nation-0.005
 team-0.004
 knock-0.004
 blast-0.004
ot-0.004
Token fault
Token ID8046
Feature activation+8.13e-02

Pos logit contributions
 perched+0.029
but+0.027
 exactly+0.027
 brand+0.026
uit+0.026
Neg logit contributions
 nation-0.015
 team-0.014
 person-0.014
ot-0.013
 guys-0.013
Token.
Token ID13
Feature activation+4.00e-03

Pos logit contributions
 brand+0.121
Here+0.116
 perched+0.109
 ang+0.108
iddled+0.108
Neg logit contributions
 ever-0.101
 team-0.095
 person-0.093
 nation-0.092
unts-0.090
Token
Token ID198
Feature activation+3.26e-04

Pos logit contributions
 perched+0.006
 exactly+0.006
uit+0.005
 brand+0.005
 struggling+0.005
Neg logit contributions
 Find-0.005
unts-0.005
 knock-0.005
oke-0.005
ass-0.005
Token
Token ID198
Feature activation+3.63e-03

Pos logit contributions
 exactly+0.001
ishly+0.000
 perched+0.000
 brand+0.000
uit+0.000
Neg logit contributions
unts-0.000
 knock-0.000
ce-0.000
ot-0.000
 Find-0.000
TokenJohnny
Token ID44960
Feature activation-1.12e-02

Pos logit contributions
 perched+0.005
 exactly+0.004
 brand+0.004
uit+0.004
ishly+0.004
Neg logit contributions
 knock-0.005
unts-0.005
ce-0.005
ot-0.005
 Find-0.005
Token smiled
Token ID13541
Feature activation+1.23e-02

Pos logit contributions
ce+0.012
 nation+0.012
 knock+0.011
air+0.011
unts+0.011
Neg logit contributions
 brand-0.019
 exactly-0.018
but-0.018
uit-0.018
 perched-0.016
Token
Token ID198
Feature activation-9.74e-03

Pos logit contributions
 nation+0.020
unts+0.020
 knock+0.020
ce+0.019
 Find+0.019
Neg logit contributions
 brand-0.017
 exactly-0.017
uit-0.016
but-0.016
 perched-0.015
Token"
Token ID1
Feature activation+3.74e-03

Pos logit contributions
 nation+0.012
 guys+0.011
 knock+0.011
 spe+0.011
ass+0.011
Neg logit contributions
 exactly-0.015
 brand-0.015
but-0.014
uit-0.014
 perched-0.014
TokenIt
Token ID1026
Feature activation+7.14e-04

Pos logit contributions
 brand+0.005
 exactly+0.005
but+0.005
uit+0.005
 perched+0.005
Neg logit contributions
 nation-0.005
ass-0.005
 knock-0.004
ce-0.004
unts-0.004
Token's
Token ID338
Feature activation-1.68e-02

Pos logit contributions
 brand+0.001
 exactly+0.001
uit+0.001
but+0.001
 perched+0.001
Neg logit contributions
 nation-0.001
 knock-0.001
ce-0.001
unts-0.001
 blast-0.001
Token your
Token ID534
Feature activation+1.94e-02

Pos logit contributions
 nation+0.023
 knock+0.022
unts+0.021
ce+0.021
 blast+0.021
Neg logit contributions
 brand-0.023
 exactly-0.023
but-0.022
uit-0.022
 perched-0.021
Token fault
Token ID8046
Feature activation+8.13e-02

Pos logit contributions
but+0.033
 perched+0.032
iddled+0.032
 brand+0.032
ieved+0.031
Neg logit contributions
 nation-0.023
ia-0.021
 knock-0.021
 team-0.020
 blast-0.020
Token,
Token ID11
Feature activation-6.65e-04

Pos logit contributions
 brand+0.159
Here+0.152
uit+0.145
but+0.142
 exactly+0.141
Neg logit contributions
 guys-0.071
ia-0.069
 nation-0.067
 ever-0.067
unts-0.066
Token Ben
Token ID3932
Feature activation-2.24e-02

Pos logit contributions
 nation+0.001
 knock+0.001
ce+0.001
unts+0.001
 blast+0.001
Neg logit contributions
 brand-0.001
 exactly-0.001
but-0.001
uit-0.001
 perched-0.001
Token!"
Token ID2474
Feature activation+6.85e-03

Pos logit contributions
 nation+0.031
unts+0.029
 knock+0.029
ce+0.029
 guys+0.028
Neg logit contributions
 brand-0.032
 exactly-0.030
but-0.030
uit-0.029
 unconscious-0.027
Token Lily
Token ID20037
Feature activation+1.90e-02

Pos logit contributions
uit+0.016
 brand+0.016
 exactly+0.016
 perched+0.016
but+0.016
Neg logit contributions
 knock-0.003
ot-0.003
 nation-0.002
 blast-0.002
unts-0.002
Token said
Token ID531
Feature activation+1.90e-02

Pos logit contributions
 brand+0.029
uit+0.027
 exactly+0.027
but+0.025
 atop+0.024
Neg logit contributions
 knock-0.024
unts-0.024
ce-0.023
 nation-0.022
oke-0.022
Token was
Token ID373
Feature activation+1.74e-03

Pos logit contributions
 exactly+0.043
 brand+0.041
 perched+0.038
 struggling+0.038
uit+0.037
Neg logit contributions
 nation-0.048
ye-0.046
 blast-0.045
unts-0.044
ce-0.044
Token still
Token ID991
Feature activation+1.22e-02

Pos logit contributions
 brand+0.002
 exactly+0.002
uit+0.002
but+0.002
 perched+0.002
Neg logit contributions
 nation-0.003
 knock-0.003
ce-0.002
ass-0.002
unts-0.002
Token wearing
Token ID5762
Feature activation+1.49e-02

Pos logit contributions
uit+0.021
but+0.020
ishly+0.018
 brand+0.018
Here+0.017
Neg logit contributions
 nation-0.014
 e-0.012
 ever-0.012
 knock-0.012
 person-0.012
Token her
Token ID607
Feature activation+2.96e-02

Pos logit contributions
but+0.029
sometimes+0.026
 unconscious+0.026
uit+0.026
ishly+0.026
Neg logit contributions
 wind-0.013
 blast-0.012
 turns-0.012
 guys-0.012
 spe-0.012
Token p
Token ID279
Feature activation+6.69e-02

Pos logit contributions
 unconscious+0.058
but+0.057
uit+0.056
sometimes+0.056
iddled+0.056
Neg logit contributions
 nation-0.024
 spe-0.022
 blast-0.022
ia-0.021
ot-0.021
Tokenaj
Token ID1228
Feature activation+8.07e-02

Pos logit contributions
 brand+0.049
 exactly+0.049
but+0.044
uit+0.044
 perched+0.041
Neg logit contributions
 nation-0.134
 knock-0.128
ce-0.127
unts-0.126
ass-0.126
Tokenamas
Token ID17485
Feature activation+2.58e-02

Pos logit contributions
 brand+0.063
 exactly+0.057
uit+0.056
but+0.052
 perched+0.044
Neg logit contributions
 nation-0.166
 knock-0.159
ass-0.155
unts-0.155
 Find-0.153
Token.
Token ID13
Feature activation-2.07e-03

Pos logit contributions
 brand+0.039
 perched+0.038
 ang+0.038
 exactly+0.038
 elbows+0.035
Neg logit contributions
 ever-0.030
ce-0.027
te-0.026
ot-0.025
 wake-0.025
Token
Token ID220
Feature activation+2.01e-03

Pos logit contributions
 nation+0.003
 Find+0.003
 knock+0.002
ce+0.002
unts+0.002
Neg logit contributions
 exactly-0.003
 brand-0.003
uit-0.003
but-0.003
 perched-0.003
Token
Token ID198
Feature activation-3.59e-03

Pos logit contributions
 brand+0.004
 ang+0.003
 exactly+0.003
ishly+0.003
 perched+0.003
Neg logit contributions
ia-0.002
unts-0.002
 Find-0.002
ass-0.002
ot-0.002
Token
Token ID198
Feature activation-8.13e-05

Pos logit contributions
 nation+0.005
ce+0.005
 knock+0.005
unts+0.005
 Find+0.005
Neg logit contributions
 brand-0.004
 exactly-0.004
uit-0.004
 perched-0.004
but-0.004
Token and
Token ID290
Feature activation+8.35e-03

Pos logit contributions
 brand+0.029
 exactly+0.028
sometimes+0.026
 thanking+0.026
but+0.026
Neg logit contributions
air-0.025
 guys-0.024
 knock-0.024
 blast-0.024
 wind-0.024
Token bumped
Token ID38034
Feature activation+2.14e-02

Pos logit contributions
 exactly+0.014
 brand+0.013
uit+0.013
but+0.013
sometimes+0.012
Neg logit contributions
 knock-0.009
 blast-0.009
 nation-0.009
ce-0.008
 spe-0.008
Token against
Token ID1028
Feature activation+9.54e-03

Pos logit contributions
 brand+0.027
 exactly+0.025
uit+0.025
but+0.023
ieved+0.022
Neg logit contributions
ce-0.031
 nation-0.030
 guys-0.030
air-0.030
unts-0.030
Token the
Token ID262
Feature activation-4.29e-03

Pos logit contributions
 brand+0.014
 exactly+0.014
but+0.013
uit+0.013
 unconscious+0.013
Neg logit contributions
 nation-0.012
ce-0.011
 knock-0.011
 guys-0.011
ass-0.011
Token books
Token ID3835
Feature activation-7.16e-03

Pos logit contributions
 nation+0.005
 knock+0.005
ce+0.004
 blast+0.004
unts+0.004
Neg logit contributions
 exactly-0.007
 brand-0.007
but-0.007
uit-0.007
 perched-0.006
Tokenhel
Token ID2978
Feature activation-1.13e-01

Pos logit contributions
 knock+0.009
 nation+0.009
air+0.009
ce+0.009
unts+0.009
Neg logit contributions
 brand-0.010
 exactly-0.010
 perched-0.009
uit-0.009
 thanking-0.009
Tokenves
Token ID1158
Feature activation-2.03e-02

Pos logit contributions
 nation+0.118
unts+0.112
 blast+0.110
 knock+0.106
ye+0.105
Neg logit contributions
 exactly-0.193
but-0.186
 brand-0.182
uit-0.181
 perched-0.179
Token.
Token ID13
Feature activation-2.09e-03

Pos logit contributions
 ever+0.021
 knock+0.021
 nation+0.020
air+0.020
 go+0.020
Neg logit contributions
 brand-0.034
 exactly-0.031
 ang-0.030
aj-0.030
Here-0.030
Token He
Token ID679
Feature activation-5.80e-03

Pos logit contributions
 nation+0.003
 knock+0.003
ce+0.003
unts+0.003
ass+0.003
Neg logit contributions
 brand-0.003
 exactly-0.003
uit-0.003
but-0.003
 perched-0.002
Token smiled
Token ID13541
Feature activation+5.04e-03

Pos logit contributions
 nation+0.007
 knock+0.007
ce+0.006
 blast+0.006
 spe+0.006
Neg logit contributions
 brand-0.009
 exactly-0.009
uit-0.009
but-0.009
sometimes-0.008
Token as
Token ID355
Feature activation-6.49e-03

Pos logit contributions
uit+0.010
ieved+0.010
Here+0.009
 elbows+0.009
 unconscious+0.009
Neg logit contributions
 guys-0.003
 nation-0.003
--0.003
 wind-0.003
 bye-0.003
Token was
Token ID373
Feature activation-2.40e-02

Pos logit contributions
 nation+0.011
 knock+0.011
ce+0.010
unts+0.010
ass+0.010
Neg logit contributions
 brand-0.012
 exactly-0.012
but-0.011
uit-0.011
 perched-0.011
Token ready
Token ID3492
Feature activation-1.59e-02

Pos logit contributions
 knock+0.016
 blast+0.016
 nation+0.015
 wind+0.015
 storm+0.014
Neg logit contributions
but-0.050
 brand-0.046
sometimes-0.046
uit-0.046
ishly-0.046
Token,
Token ID11
Feature activation-5.40e-03

Pos logit contributions
 nation+0.014
 guys+0.013
ce+0.012
 comes+0.012
ia+0.012
Neg logit contributions
 brand-0.030
 unconscious-0.028
but-0.028
 perched-0.028
 exactly-0.028
Token the
Token ID262
Feature activation-1.04e-02

Pos logit contributions
 nation+0.006
 knock+0.006
 guys+0.006
 blast+0.006
ce+0.005
Neg logit contributions
 brand-0.009
 exactly-0.008
but-0.008
uit-0.008
 unconscious-0.008
Token b
Token ID275
Feature activation-5.45e-02

Pos logit contributions
 nation+0.009
 knock+0.009
 guys+0.008
 blast+0.008
ce+0.008
Neg logit contributions
 exactly-0.020
but-0.020
 brand-0.019
uit-0.019
 perched-0.018
Tokenunn
Token ID20935
Feature activation-1.10e-01

Pos logit contributions
 nation+0.043
 blast+0.040
 knock+0.039
 guys+0.035
 Find+0.034
Neg logit contributions
 exactly-0.109
 brand-0.105
uit-0.104
but-0.104
 perched-0.100
Tokenies
Token ID444
Feature activation-1.09e-02

Pos logit contributions
 nation+0.122
ce+0.112
 knock+0.110
unts+0.110
ass+0.109
Neg logit contributions
 brand-0.183
 exactly-0.179
uit-0.172
but-0.170
 perched-0.164
Token hopped
Token ID45230
Feature activation-1.48e-02

Pos logit contributions
 nation+0.012
 knock+0.010
ce+0.010
 blast+0.010
 team+0.009
Neg logit contributions
 brand-0.021
uit-0.019
 exactly-0.019
but-0.018
 unconscious-0.018
Token over
Token ID625
Feature activation+5.70e-03

Pos logit contributions
ce+0.017
ny+0.016
ot+0.015
 ever+0.015
ye+0.015
Neg logit contributions
 exactly-0.022
 brand-0.022
sometimes-0.021
oding-0.021
nam-0.021
Token to
Token ID284
Feature activation-7.21e-03

Pos logit contributions
 exactly+0.010
ieved+0.009
oding+0.009
 brand+0.009
but+0.009
Neg logit contributions
 spe-0.006
 nation-0.006
 blast-0.006
ce-0.006
 ever-0.005
Token the
Token ID262
Feature activation+1.61e-03

Pos logit contributions
 nation+0.009
 knock+0.009
 blast+0.008
 guys+0.008
 spe+0.008
Neg logit contributions
 exactly-0.011
 brand-0.011
but-0.011
uit-0.011
 unconscious-0.010
Token on
Token ID319
Feature activation-4.44e-03

Pos logit contributions
 nation+0.024
 knock+0.022
 guys+0.022
 blast+0.020
 wake+0.020
Neg logit contributions
 perched-0.026
 unconscious-0.026
uit-0.026
but-0.026
ishly-0.025
Token the
Token ID262
Feature activation-4.74e-05

Pos logit contributions
 nation+0.005
 ever+0.005
 spe+0.005
ce+0.005
 month+0.005
Neg logit contributions
but-0.007
 brand-0.007
oding-0.007
uit-0.007
ieved-0.006
Token row
Token ID5752
Feature activation-1.27e-02

Pos logit contributions
 nation+0.000
unts+0.000
ass+0.000
ce+0.000
 knock+0.000
Neg logit contributions
 brand-0.000
 exactly-0.000
but-0.000
uit-0.000
 perched-0.000
Token of
Token ID286
Feature activation+2.29e-04

Pos logit contributions
 nation+0.013
 ever+0.010
ce+0.010
ot+0.010
ass+0.010
Neg logit contributions
 brand-0.025
 exactly-0.023
 unconscious-0.023
 perched-0.022
nam-0.022
Token books
Token ID3835
Feature activation+5.06e-03

Pos logit contributions
 exactly+0.000
 brand+0.000
but+0.000
uit+0.000
 perched+0.000
Neg logit contributions
 nation-0.000
 knock-0.000
 blast-0.000
ce-0.000
 spe-0.000
Tokenhel
Token ID2978
Feature activation-1.10e-01

Pos logit contributions
 ang+0.008
 brand+0.008
 unconscious+0.008
 elbows+0.008
 exactly+0.008
Neg logit contributions
 nation-0.006
 ever-0.006
ce-0.005
air-0.005
 month-0.005
Tokenves
Token ID1158
Feature activation-1.05e-02

Pos logit contributions
 nation+0.106
 knock+0.097
unts+0.096
 blast+0.094
 Find+0.093
Neg logit contributions
 exactly-0.194
 brand-0.191
but-0.186
uit-0.183
 perched-0.180
Token.
Token ID13
Feature activation+3.17e-03

Pos logit contributions
 ever+0.013
 nation+0.012
 comes+0.010
 guys+0.010
 goes+0.010
Neg logit contributions
 ang-0.018
 brand-0.017
 unconscious-0.016
 elbows-0.016
 exactly-0.015
Token<|endoftext|>
Token ID50256
Feature activation+1.89e-02

Pos logit contributions
 brand+0.004
 exactly+0.004
uit+0.004
but+0.004
 perched+0.004
Neg logit contributions
 nation-0.004
 Find-0.004
ass-0.004
ce-0.004
unts-0.004
TokenTom
Token ID13787
Feature activation-2.37e-02

Pos logit contributions
 brand+0.028
 exactly+0.028
but+0.024
 ang+0.023
uit+0.023
Neg logit contributions
ce-0.026
 nation-0.025
ass-0.024
unts-0.023
 Find-0.023
Token was
Token ID373
Feature activation-2.53e-03

Pos logit contributions
ce+0.032
 comes+0.031
 nation+0.031
ith+0.031
 goes+0.031
Neg logit contributions
but-0.034
 brand-0.034
 exactly-0.032
nam-0.029
uit-0.029
Token spotted
Token ID13489
Feature activation+8.71e-03

Pos logit contributions
 brand+0.007
 exactly+0.007
but+0.007
uit+0.007
 perched+0.006
Neg logit contributions
 nation-0.005
 knock-0.005
ce-0.005
unts-0.004
 guys-0.004
Token a
Token ID257
Feature activation+1.50e-03

Pos logit contributions
but+0.015
ieved+0.015
 afloat+0.014
shadow+0.014
arse+0.014
Neg logit contributions
 person-0.007
 nation-0.007
 e-0.006
 Halloween-0.006
head-0.006
Token row
Token ID5752
Feature activation-3.99e-03

Pos logit contributions
 exactly+0.003
but+0.003
uit+0.003
 brand+0.003
 perched+0.002
Neg logit contributions
 nation-0.002
 knock-0.001
 spe-0.001
 blast-0.001
ce-0.001
Token of
Token ID286
Feature activation-3.78e-03

Pos logit contributions
 nation+0.002
ot+0.002
 person+0.002
 knock+0.002
 team+0.002
Neg logit contributions
 brand-0.010
 exactly-0.009
but-0.009
 perched-0.008
 unconscious-0.008
Token books
Token ID3835
Feature activation+9.44e-03

Pos logit contributions
 nation+0.004
 e+0.003
 blast+0.003
 spe+0.003
 guys+0.003
Neg logit contributions
but-0.007
 exactly-0.007
uit-0.007
 brand-0.006
ieved-0.006
Tokenhel
Token ID2978
Feature activation-1.08e-01

Pos logit contributions
 brand+0.016
 exactly+0.015
 ang+0.015
 unconscious+0.014
 elbows+0.014
Neg logit contributions
 knock-0.010
 nation-0.009
ce-0.009
 ever-0.009
air-0.008
Tokenves
Token ID1158
Feature activation-8.03e-03

Pos logit contributions
 nation+0.086
ce+0.079
 knock+0.077
ass+0.072
ia+0.072
Neg logit contributions
 brand-0.212
 exactly-0.206
uit-0.202
but-0.197
 unconscious-0.193
Token in
Token ID287
Feature activation+2.16e-03

Pos logit contributions
 ever+0.007
 nation+0.007
 comes+0.006
 goes+0.006
 go+0.006
Neg logit contributions
 ang-0.015
 brand-0.015
 exactly-0.014
Acc-0.013
 Okay-0.013
Token her
Token ID607
Feature activation+1.14e-02

Pos logit contributions
but+0.004
 brand+0.004
 exactly+0.003
ishly+0.003
 unconscious+0.003
Neg logit contributions
 nation-0.003
 blast-0.002
 spe-0.002
 knock-0.002
 month-0.002
Token grandparents
Token ID28571
Feature activation+1.33e-02

Pos logit contributions
but+0.018
 exactly+0.018
uit+0.017
 brand+0.017
sometimes+0.017
Neg logit contributions
 nation-0.015
 blast-0.013
ce-0.012
 Find-0.012
 knock-0.012
Token'
Token ID6
Feature activation-1.45e-02

Pos logit contributions
but+0.024
 exactly+0.024
 brand+0.024
 unconscious+0.023
 ang+0.023
Neg logit contributions
 nation-0.013
ce-0.012
 knock-0.011
 guys-0.010
 Find-0.010
Token walked
Token ID6807
Feature activation+2.56e-04

Pos logit contributions
 brand+0.004
 exactly+0.004
uit+0.004
but+0.004
 unconscious+0.004
Neg logit contributions
 nation-0.004
 knock-0.003
ce-0.003
 blast-0.003
unts-0.003
Token over
Token ID625
Feature activation+7.54e-03

Pos logit contributions
 brand+0.000
 exactly+0.000
sometimes+0.000
 unconscious+0.000
 struggling+0.000
Neg logit contributions
ce-0.000
 nation-0.000
 knock-0.000
 guys-0.000
ot-0.000
Token to
Token ID284
Feature activation-2.94e-03

Pos logit contributions
 brand+0.011
 exactly+0.011
uit+0.010
but+0.010
 perched+0.010
Neg logit contributions
 nation-0.010
ce-0.010
 blast-0.009
 knock-0.009
 spe-0.009
Token the
Token ID262
Feature activation+4.12e-04

Pos logit contributions
 nation+0.004
 knock+0.004
 blast+0.004
 guys+0.004
ass+0.004
Neg logit contributions
 brand-0.004
 exactly-0.004
but-0.004
uit-0.004
 perched-0.004
Token books
Token ID3835
Feature activation+3.18e-03

Pos logit contributions
 brand+0.001
 exactly+0.001
but+0.001
uit+0.001
 perched+0.001
Neg logit contributions
 nation-0.001
 knock-0.000
ce-0.000
unts-0.000
ass-0.000
Tokenhel
Token ID2978
Feature activation-1.06e-01

Pos logit contributions
 brand+0.005
 exactly+0.005
 ang+0.005
 perched+0.005
 unconscious+0.004
Neg logit contributions
 nation-0.004
 knock-0.003
ce-0.003
air-0.003
 blast-0.003
Tokenves
Token ID1158
Feature activation-7.38e-03

Pos logit contributions
 nation+0.107
unts+0.099
 blast+0.097
 knock+0.097
ye+0.095
Neg logit contributions
 exactly-0.186
 brand-0.181
but-0.178
uit-0.175
 perched-0.172
Token and
Token ID290
Feature activation-7.08e-03

Pos logit contributions
 nation+0.004
 ever+0.004
 go+0.004
 wake+0.004
 comes+0.004
Neg logit contributions
 ang-0.015
 brand-0.015
Acc-0.014
 unconscious-0.014
 exactly-0.014
Token began
Token ID2540
Feature activation-1.31e-02

Pos logit contributions
 nation+0.008
 knock+0.008
 blast+0.007
 guys+0.007
ce+0.007
Neg logit contributions
 brand-0.012
 exactly-0.012
uit-0.011
but-0.011
 unconscious-0.011
Token to
Token ID284
Feature activation-1.21e-02

Pos logit contributions
 knock+0.016
 nation+0.016
ce+0.015
 blast+0.015
 spe+0.015
Neg logit contributions
 brand-0.020
 exactly-0.019
uit-0.019
ieved-0.019
but-0.019
Token read
Token ID1100
Feature activation-3.69e-03

Pos logit contributions
 knock+0.012
 nation+0.011
 blast+0.010
 wake+0.009
 guys+0.009
Neg logit contributions
 exactly-0.023
but-0.023
 brand-0.023
uit-0.022
 unconscious-0.022
Token park
Token ID3952
Feature activation+3.51e-03

Pos logit contributions
 exactly+0.012
but+0.012
uit+0.011
 brand+0.011
 perched+0.011
Neg logit contributions
 nation-0.007
 knock-0.007
 blast-0.006
 storm-0.006
 person-0.006
Token,
Token ID11
Feature activation+2.71e-03

Pos logit contributions
 brand+0.006
uit+0.006
 exactly+0.005
 unconscious+0.005
iddled+0.005
Neg logit contributions
 nation-0.004
 knock-0.004
 person-0.003
 comes-0.003
 ever-0.003
Token the
Token ID262
Feature activation-1.35e-03

Pos logit contributions
uit+0.005
 brand+0.004
but+0.004
 atop+0.004
iddled+0.004
Neg logit contributions
 nation-0.003
 knock-0.003
ot-0.003
 wake-0.003
 blast-0.003
Token g
Token ID308
Feature activation+3.46e-03

Pos logit contributions
 nation+0.001
 knock+0.001
 blast+0.001
 wind+0.001
 storm+0.001
Neg logit contributions
 exactly-0.003
but-0.003
uit-0.003
 brand-0.002
 perched-0.002
Tokenaz
Token ID1031
Feature activation-7.96e-02

Pos logit contributions
 brand+0.006
 exactly+0.006
uit+0.005
but+0.005
 unconscious+0.005
Neg logit contributions
 spe-0.004
 nation-0.004
 Find-0.003
 knock-0.003
 blast-0.003
Tokeneb
Token ID1765
Feature activation-1.06e-01

Pos logit contributions
 nation+0.085
ce+0.083
 knock+0.077
ass+0.077
air+0.077
Neg logit contributions
 brand-0.135
 exactly-0.133
uit-0.125
but-0.122
 perched-0.119
Tokeno
Token ID78
Feature activation-2.30e-02

Pos logit contributions
ce+0.093
ass+0.091
ia+0.090
 nation+0.090
ot+0.086
Neg logit contributions
 brand-0.198
but-0.185
 exactly-0.185
 unconscious-0.180
uit-0.178
Token would
Token ID561
Feature activation-6.05e-03

Pos logit contributions
 nation+0.023
ce+0.022
 knock+0.021
 ever+0.021
 spe+0.021
Neg logit contributions
 brand-0.040
but-0.039
 exactly-0.039
uit-0.038
nam-0.037
Token be
Token ID307
Feature activation+3.30e-03

Pos logit contributions
 knock+0.005
 wake+0.004
 nation+0.004
 blast+0.004
 ever+0.003
Neg logit contributions
 brand-0.013
but-0.013
 unconscious-0.013
iddled-0.013
 exactly-0.013
Token there
Token ID612
Feature activation+3.48e-03

Pos logit contributions
but+0.006
ishly+0.005
uit+0.005
stack+0.005
nam+0.005
Neg logit contributions
ot-0.003
 wind-0.003
 blast-0.003
 ever-0.003
 time-0.003
Token waiting
Token ID4953
Feature activation+4.27e-03

Pos logit contributions
iddled+0.007
 elbows+0.007
 ang+0.007
 unconscious+0.007
 brand+0.006
Neg logit contributions
 ever-0.004
 nation-0.003
 month-0.003
 team-0.002
 person-0.002
Token sometimes
Token ID3360
Feature activation-5.61e-04

Pos logit contributions
 brand+0.009
uit+0.009
 exactly+0.009
but+0.009
 perched+0.008
Neg logit contributions
 nation-0.008
 knock-0.008
 guys-0.007
 spe-0.007
 blast-0.007
Token,
Token ID11
Feature activation+5.00e-03

Pos logit contributions
 nation+0.001
 knock+0.001
 guys+0.001
 blast+0.001
unts+0.001
Neg logit contributions
 brand-0.001
 exactly-0.001
uit-0.001
but-0.001
 perched-0.001
Token even
Token ID772
Feature activation+8.93e-03

Pos logit contributions
 brand+0.007
 exactly+0.007
uit+0.007
but+0.007
 perched+0.006
Neg logit contributions
 nation-0.007
 knock-0.006
 guys-0.006
 blast-0.006
 spe-0.006
Token the
Token ID262
Feature activation-1.01e-02

Pos logit contributions
 brand+0.009
uit+0.009
 exactly+0.009
but+0.009
 perched+0.008
Neg logit contributions
 nation-0.015
 knock-0.014
 guys-0.014
 spe-0.014
 blast-0.014
Token l
Token ID300
Feature activation-1.66e-03

Pos logit contributions
 nation+0.010
 knock+0.009
 blast+0.009
 guys+0.009
 spe+0.008
Neg logit contributions
 exactly-0.018
uit-0.018
 brand-0.018
but-0.018
 perched-0.017
Tokenitt
Token ID715
Feature activation-1.06e-01

Pos logit contributions
 nation+0.002
 knock+0.001
ce+0.001
unts+0.001
 blast+0.001
Neg logit contributions
 exactly-0.003
 brand-0.003
but-0.003
uit-0.003
 perched-0.003
Tokenlest
Token ID32712
Feature activation-8.23e-04

Pos logit contributions
 nation+0.100
ce+0.091
unts+0.090
 knock+0.088
ass+0.088
Neg logit contributions
 brand-0.194
 exactly-0.186
uit-0.185
but-0.180
 perched-0.171
Token things
Token ID1243
Feature activation-1.40e-02

Pos logit contributions
 nation+0.001
ass+0.001
 guys+0.001
 person+0.001
 e+0.001
Neg logit contributions
 brand-0.002
uit-0.001
 exactly-0.001
but-0.001
 perched-0.001
Token can
Token ID460
Feature activation-2.24e-02

Pos logit contributions
 nation+0.012
 knock+0.011
 comes+0.011
 happens+0.011
 guys+0.011
Neg logit contributions
uit-0.028
 brand-0.027
but-0.026
 exactly-0.026
 perched-0.025
Token lead
Token ID1085
Feature activation-1.87e-02

Pos logit contributions
 knock+0.017
 nation+0.017
 guys+0.014
 blast+0.014
unts+0.014
Neg logit contributions
 brand-0.045
 exactly-0.044
but-0.043
uit-0.043
 unconscious-0.042
Token to
Token ID284
Feature activation-1.73e-02

Pos logit contributions
 nation+0.023
 guys+0.022
 ever+0.021
ot+0.020
 Find+0.020
Neg logit contributions
 brand-0.030
uit-0.029
but-0.028
 exactly-0.028
 perched-0.026
Token one
Token ID530
Feature activation-4.42e-03

Pos logit contributions
 nation+0.019
 knock+0.018
 blast+0.018
unts+0.017
 e+0.017
Neg logit contributions
uit-0.019
 brand-0.018
but-0.018
 altogether-0.017
 exactly-0.017
Token of
Token ID286
Feature activation-2.87e-02

Pos logit contributions
 nation+0.005
 knock+0.005
ce+0.005
 blast+0.005
ass+0.005
Neg logit contributions
uit-0.007
 brand-0.007
 exactly-0.007
but-0.006
 perched-0.006
Token the
Token ID262
Feature activation-1.48e-02

Pos logit contributions
 eight+0.030
 nation+0.030
ot+0.029
 July+0.029
 blast+0.028
Neg logit contributions
 brand-0.050
but-0.049
uit-0.044
uddled-0.044
 exactly-0.044
Token happiest
Token ID49414
Feature activation+8.89e-03

Pos logit contributions
 nation+0.016
 team+0.014
 person+0.013
 knock+0.013
 guys+0.012
Neg logit contributions
but-0.028
 brand-0.027
uit-0.027
 exactly-0.027
 perched-0.027
Token re
Token ID302
Feature activation-8.01e-02

Pos logit contributions
uit+0.016
 brand+0.016
 exactly+0.015
but+0.015
 unconscious+0.014
Neg logit contributions
 nation-0.010
 person-0.008
 team-0.008
 guys-0.008
 knock-0.008
Tokenind
Token ID521
Feature activation-1.05e-01

Pos logit contributions
 nation+0.063
 knock+0.059
 blast+0.059
 Find+0.056
unts+0.056
Neg logit contributions
 exactly-0.156
 brand-0.153
but-0.150
uit-0.147
 perched-0.146
Tokeneers
Token ID47619
Feature activation-3.47e-02

Pos logit contributions
 nation+0.063
ce+0.053
 knock+0.052
ass+0.050
unts+0.050
Neg logit contributions
 brand-0.229
 exactly-0.227
uit-0.219
but-0.217
 perched-0.212
Token in
Token ID287
Feature activation-1.50e-02

Pos logit contributions
 ever+0.040
 nation+0.039
 person+0.031
 spe+0.031
 team+0.031
Neg logit contributions
 brand-0.062
uit-0.061
 ang-0.059
 exactly-0.056
aj-0.055
Token the
Token ID262
Feature activation+5.44e-03

Pos logit contributions
 nation+0.018
 blast+0.015
 spe+0.014
 knock+0.014
unts+0.014
Neg logit contributions
but-0.026
 brand-0.026
 exactly-0.025
uit-0.025
 unconscious-0.024
Token North
Token ID2258
Feature activation-2.79e-02

Pos logit contributions
 exactly+0.011
but+0.011
uit+0.010
 brand+0.010
 perched+0.009
Neg logit contributions
 nation-0.006
ce-0.004
 guys-0.004
 blast-0.004
ye-0.004
Token Pole
Token ID29385
Feature activation-2.67e-02

Pos logit contributions
 nation+0.029
ce+0.027
 knock+0.027
unts+0.026
ass+0.026
Neg logit contributions
 exactly-0.047
 brand-0.047
but-0.044
uit-0.044
 perched-0.043
Token He
Token ID679
Feature activation-4.88e-03

Pos logit contributions
 nation+0.008
 knock+0.008
unts+0.008
ass+0.008
ce+0.008
Neg logit contributions
 brand-0.007
 exactly-0.006
but-0.006
uit-0.006
 perched-0.006
Token was
Token ID373
Feature activation-7.07e-03

Pos logit contributions
 nation+0.006
 knock+0.006
ce+0.006
unts+0.006
 blast+0.006
Neg logit contributions
 brand-0.007
 exactly-0.007
uit-0.007
but-0.007
 perched-0.006
Token the
Token ID262
Feature activation-6.66e-03

Pos logit contributions
 nation+0.009
 knock+0.009
ce+0.009
 blast+0.009
unts+0.009
Neg logit contributions
 brand-0.010
 exactly-0.010
uit-0.010
but-0.010
 unconscious-0.009
Token happiest
Token ID49414
Feature activation+7.02e-03

Pos logit contributions
 nation+0.009
 knock+0.008
ce+0.008
ass+0.008
unts+0.008
Neg logit contributions
 brand-0.009
 exactly-0.009
but-0.009
uit-0.009
 perched-0.008
Token re
Token ID302
Feature activation-7.86e-02

Pos logit contributions
 exactly+0.013
 brand+0.013
but+0.013
uit+0.013
 perched+0.013
Neg logit contributions
 nation-0.006
 knock-0.005
unts-0.005
 Find-0.005
ce-0.005
Tokenind
Token ID521
Feature activation-1.05e-01

Pos logit contributions
 blast+0.095
 Find+0.088
 happens+0.086
 knock+0.086
 guys+0.083
Neg logit contributions
but-0.126
 exactly-0.126
 perched-0.122
 brand-0.115
uit-0.112
Tokeneer
Token ID28153
Feature activation-4.59e-02

Pos logit contributions
 blast+0.099
 knock+0.096
 Find+0.096
unts+0.094
 nation+0.093
Neg logit contributions
but-0.182
 exactly-0.179
 perched-0.175
uit-0.168
 brand-0.167
Token in
Token ID287
Feature activation-9.11e-03

Pos logit contributions
 nation+0.057
 ever+0.053
ce+0.048
 guys+0.048
ass+0.048
Neg logit contributions
 brand-0.076
uit-0.072
 exactly-0.070
 ang-0.067
but-0.066
Token the
Token ID262
Feature activation+6.95e-03

Pos logit contributions
 nation+0.013
 knock+0.012
 blast+0.011
ce+0.011
ass+0.011
Neg logit contributions
 brand-0.013
 exactly-0.013
but-0.012
uit-0.012
 perched-0.011
Token North
Token ID2258
Feature activation-2.32e-02

Pos logit contributions
 brand+0.010
 exactly+0.010
 perched+0.010
uit+0.010
but+0.009
Neg logit contributions
 nation-0.008
 knock-0.008
unts-0.008
ass-0.008
 Find-0.008
Token Pole
Token ID29385
Feature activation-2.48e-02

Pos logit contributions
 nation+0.024
 knock+0.022
 blast+0.022
unts+0.022
ass+0.022
Neg logit contributions
 brand-0.039
 exactly-0.038
but-0.037
uit-0.037
 perched-0.036
Token.
Token ID13
Feature activation-1.39e-03

Pos logit contributions
 brand+0.025
 unconscious+0.025
 exactly+0.025
 perched+0.024
 ang+0.024
Neg logit contributions
 nation-0.020
ce-0.018
 guys-0.018
ot-0.018
ia-0.018
Token It
Token ID632
Feature activation-1.85e-02

Pos logit contributions
 Find+0.001
 nation+0.001
unts+0.001
ass+0.001
ce+0.001
Neg logit contributions
 brand-0.003
 exactly-0.003
uit-0.002
but-0.002
 ang-0.002
Token was
Token ID373
Feature activation-1.18e-02

Pos logit contributions
 nation+0.018
 knock+0.017
ce+0.016
 spe+0.016
 blast+0.016
Neg logit contributions
uit-0.034
 brand-0.034
but-0.033
 exactly-0.033
 unconscious-0.031
Token called
Token ID1444
Feature activation-2.04e-02

Pos logit contributions
 spe+0.013
 nation+0.013
 knock+0.013
 wind+0.013
 e+0.012
Neg logit contributions
but-0.020
uit-0.019
oding-0.017
sometimes-0.017
ishly-0.017
Token "
Token ID366
Feature activation-2.38e-03

Pos logit contributions
 nation+0.025
 knock+0.023
 blast+0.023
 guys+0.023
ce+0.023
Neg logit contributions
 exactly-0.031
but-0.031
 brand-0.031
uit-0.029
 perched-0.029
Tokencham
Token ID49869
Feature activation-1.04e-01

Pos logit contributions
ce+0.003
 nation+0.003
ass+0.002
ot+0.002
air+0.002
Neg logit contributions
 brand-0.004
 exactly-0.004
uit-0.004
 struggling-0.004
 unconscious-0.004
Tokenom
Token ID296
Feature activation-3.39e-02

Pos logit contributions
 nation+0.098
 knock+0.091
ce+0.087
unts+0.086
 blast+0.086
Neg logit contributions
 exactly-0.185
 brand-0.184
uit-0.177
but-0.177
 perched-0.171
Tokenile
Token ID576
Feature activation+1.01e-03

Pos logit contributions
 nation+0.033
ce+0.031
 knock+0.030
unts+0.029
 guys+0.029
Neg logit contributions
 brand-0.061
 exactly-0.060
but-0.057
uit-0.057
 perched-0.055
Token"
Token ID1
Feature activation-1.12e-02

Pos logit contributions
 brand+0.002
 exactly+0.002
but+0.002
uit+0.002
 unconscious+0.001
Neg logit contributions
 nation-0.001
ce-0.001
 guys-0.001
 knock-0.001
ass-0.001
Token and
Token ID290
Feature activation-2.54e-03

Pos logit contributions
ce+0.014
ot+0.013
ass+0.013
 guys+0.013
 nation+0.013
Neg logit contributions
 brand-0.018
 exactly-0.016
 struggling-0.016
 unconscious-0.015
ishly-0.015
Token it
Token ID340
Feature activation-6.25e-03

Pos logit contributions
 knock+0.003
 nation+0.003
 spe+0.003
 e+0.002
 blast+0.002
Neg logit contributions
 brand-0.004
 exactly-0.004
uit-0.004
but-0.004
 unconscious-0.004
Token quiet
Token ID5897
Feature activation+5.92e-03

Pos logit contributions
 nation+0.014
ass+0.014
ce+0.013
unts+0.013
 Find+0.013
Neg logit contributions
 brand-0.017
 exactly-0.017
 perched-0.016
uit-0.016
but-0.015
Token compared
Token ID3688
Feature activation-3.21e-02

Pos logit contributions
 brand+0.010
 exactly+0.010
uit+0.010
but+0.009
ieved+0.009
Neg logit contributions
 nation-0.007
 guys-0.007
ot-0.007
 wake-0.007
 wind-0.006
Token to
Token ID284
Feature activation-2.18e-02

Pos logit contributions
ot+0.035
 nation+0.033
 wind+0.030
 ever+0.030
 person+0.030
Neg logit contributions
 brand-0.057
uit-0.054
ieved-0.051
 perched-0.050
ishly-0.049
Token the
Token ID262
Feature activation-1.17e-02

Pos logit contributions
 nation+0.025
 knock+0.024
 blast+0.023
ot+0.022
unts+0.021
Neg logit contributions
but-0.037
 brand-0.037
uit-0.036
 exactly-0.035
 perched-0.033
Token noisy
Token ID31210
Feature activation-6.07e-03

Pos logit contributions
 nation+0.013
 person+0.011
 wind+0.011
 storm+0.011
 knock+0.011
Neg logit contributions
but-0.022
uit-0.021
 exactly-0.021
ieved-0.021
iddled-0.020
Token se
Token ID384
Feature activation-1.04e-01

Pos logit contributions
 nation+0.007
 guys+0.006
 knock+0.005
unts+0.005
 person+0.005
Neg logit contributions
 exactly-0.012
but-0.011
 brand-0.011
uit-0.011
 perched-0.011
Tokenag
Token ID363
Feature activation-2.20e-02

Pos logit contributions
 nation+0.130
ce+0.123
unts+0.122
ass+0.121
 knock+0.119
Neg logit contributions
 brand-0.154
 exactly-0.152
uit-0.145
but-0.144
 perched-0.143
Tokenull
Token ID724
Feature activation-4.45e-02

Pos logit contributions
ot+0.016
 nation+0.015
ia+0.014
ass+0.014
unts+0.013
Neg logit contributions
 brand-0.051
 exactly-0.046
uit-0.045
 ang-0.044
nam-0.044
Tokens
Token ID82
Feature activation-3.12e-02

Pos logit contributions
ot+0.045
 nation+0.044
ce+0.043
air+0.042
 knock+0.040
Neg logit contributions
 brand-0.076
uit-0.074
 exactly-0.073
but-0.069
Here-0.069
Token outside
Token ID2354
Feature activation+1.75e-02

Pos logit contributions
 nation+0.034
 ever+0.033
ot+0.033
 team+0.029
unts+0.028
Neg logit contributions
uit-0.053
but-0.051
 brand-0.051
rob-0.048
ieved-0.048
Token.
Token ID13
Feature activation+3.34e-03

Pos logit contributions
 brand+0.030
uit+0.028
ieved+0.028
 exactly+0.027
but+0.027
Neg logit contributions
 nation-0.021
ot-0.019
 person-0.018
 ever-0.018
 team-0.017
Token the
Token ID262
Feature activation-7.78e-03

Pos logit contributions
 nation+0.023
 knock+0.022
ce+0.021
unts+0.021
 blast+0.021
Neg logit contributions
 brand-0.021
 exactly-0.021
but-0.019
uit-0.019
 perched-0.018
Token shore
Token ID15191
Feature activation-1.76e-02

Pos logit contributions
 nation+0.009
 knock+0.008
 blast+0.008
 guys+0.008
ye+0.008
Neg logit contributions
 exactly-0.013
 brand-0.013
but-0.013
uit-0.013
 perched-0.012
Token,
Token ID11
Feature activation-1.27e-02

Pos logit contributions
 nation+0.025
 knock+0.023
ce+0.022
air+0.022
unts+0.022
Neg logit contributions
 brand-0.025
 exactly-0.024
 unconscious-0.023
 perched-0.022
uit-0.022
Token and
Token ID290
Feature activation-1.28e-02

Pos logit contributions
 knock+0.015
 e+0.013
 nation+0.013
 wake+0.013
 blast+0.013
Neg logit contributions
uit-0.021
 brand-0.020
?'-0.020
?,-0.019
 exactly-0.019
Token the
Token ID262
Feature activation-1.07e-02

Pos logit contributions
 knock+0.015
 nation+0.014
 spe+0.013
 blast+0.013
 e+0.013
Neg logit contributions
uit-0.022
 brand-0.021
 exactly-0.021
but-0.020
?'-0.019
Token se
Token ID384
Feature activation-1.04e-01

Pos logit contributions
 nation+0.010
 wind+0.009
 knock+0.008
 blast+0.008
 storm+0.008
Neg logit contributions
but-0.022
 exactly-0.022
uit-0.021
ieved-0.020
 brand-0.020
Tokenag
Token ID363
Feature activation-2.55e-02

Pos logit contributions
 nation+0.167
 knock+0.159
 spe+0.157
 blast+0.154
 Find+0.153
Neg logit contributions
 exactly-0.118
 brand-0.113
uit-0.109
but-0.108
 perched-0.099
Tokenull
Token ID724
Feature activation-4.41e-02

Pos logit contributions
 nation+0.017
unts+0.014
ce+0.014
ass+0.014
 knock+0.014
Neg logit contributions
 brand-0.054
 exactly-0.053
uit-0.051
but-0.051
 perched-0.049
Tokens
Token ID82
Feature activation-3.78e-02

Pos logit contributions
 nation+0.052
 knock+0.051
ce+0.049
unts+0.048
 spe+0.048
Neg logit contributions
 brand-0.068
 exactly-0.068
uit-0.065
but-0.064
 unconscious-0.060
Token flying
Token ID7348
Feature activation+4.70e-03

Pos logit contributions
 spe+0.041
unts+0.039
 blast+0.038
 nation+0.038
 wind+0.037
Neg logit contributions
 brand-0.063
but-0.063
ieved-0.062
 exactly-0.062
uit-0.061
Token in
Token ID287
Feature activation-5.93e-03

Pos logit contributions
 brand+0.007
uit+0.007
 thanking+0.007
 exactly+0.007
but+0.006
Neg logit contributions
ot-0.006
 nation-0.006
 ever-0.006
 spe-0.005
 wind-0.005
Token on
Token ID319
Feature activation-7.71e-03

Pos logit contributions
 brand+0.000
but+0.000
 exactly+0.000
uit+0.000
 perched+0.000
Neg logit contributions
 nation-0.000
 guys-0.000
 knock-0.000
unts-0.000
ce-0.000
Token top
Token ID1353
Feature activation+1.41e-02

Pos logit contributions
 nation+0.008
 blast+0.008
 knock+0.008
ia+0.007
 breaks+0.007
Neg logit contributions
but-0.015
 brand-0.014
uit-0.014
uddled-0.013
 exactly-0.013
Token.
Token ID13
Feature activation+4.51e-03

Pos logit contributions
 brand+0.026
uit+0.023
 exactly+0.023
nam+0.022
ishly+0.022
Neg logit contributions
ia-0.016
 nation-0.015
 knock-0.015
ce-0.015
air-0.015
Token The
Token ID383
Feature activation-2.49e-04

Pos logit contributions
 brand+0.008
 exactly+0.008
uit+0.008
 ang+0.008
 struggling+0.007
Neg logit contributions
 Find-0.004
 nation-0.004
ot-0.004
 Danger-0.004
 Go-0.004
Token b
Token ID275
Feature activation-4.44e-02

Pos logit contributions
 nation+0.000
 knock+0.000
 blast+0.000
 team+0.000
 storm+0.000
Neg logit contributions
but-0.000
 exactly-0.000
uit-0.000
 brand-0.000
 perched-0.000
Tokenunn
Token ID20935
Feature activation-1.03e-01

Pos logit contributions
 nation+0.043
 knock+0.040
 blast+0.039
 guys+0.037
unts+0.037
Neg logit contributions
 exactly-0.080
 brand-0.077
but-0.076
uit-0.075
 perched-0.074
Tokenies
Token ID444
Feature activation+4.66e-03

Pos logit contributions
 nation+0.096
unn+0.088
ass+0.085
unts+0.084
ot+0.083
Neg logit contributions
 brand-0.203
uit-0.187
 exactly-0.185
but-0.178
 unconscious-0.172
Token wanted
Token ID2227
Feature activation-1.47e-02

Pos logit contributions
 brand+0.008
uit+0.008
but+0.008
 exactly+0.008
 unconscious+0.007
Neg logit contributions
 nation-0.006
 knock-0.005
 goes-0.005
 comes-0.005
 team-0.004
Token to
Token ID284
Feature activation-4.54e-03

Pos logit contributions
 nation+0.017
 guys+0.016
 knock+0.016
 spe+0.016
 ever+0.016
Neg logit contributions
 brand-0.023
but-0.022
uit-0.022
 exactly-0.021
 unconscious-0.021
Token have
Token ID423
Feature activation+1.15e-03

Pos logit contributions
 knock+0.004
 nation+0.004
 blast+0.003
 guys+0.003
 wake+0.003
Neg logit contributions
 brand-0.010
but-0.009
 exactly-0.009
 unconscious-0.009
uit-0.009
Token a
Token ID257
Feature activation+1.12e-03

Pos logit contributions
but+0.002
uit+0.002
oding+0.002
 unconscious+0.002
 brand+0.002
Neg logit contributions
 blast-0.001
 nation-0.001
 spe-0.001
 turns-0.001
 guys-0.001
Token did
Token ID750
Feature activation-2.25e-02

Pos logit contributions
 knock+0.020
 nation+0.020
 blast+0.018
 guys+0.016
 sneak+0.016
Neg logit contributions
but-0.041
 brand-0.041
uit-0.039
 unconscious-0.038
sometimes-0.038
Token not
Token ID407
Feature activation-2.48e-04

Pos logit contributions
 knock+0.024
 blast+0.021
 nation+0.020
ia+0.018
 turns+0.017
Neg logit contributions
uit-0.044
but-0.042
 unconscious-0.042
ishly-0.041
sometimes-0.039
Token feel
Token ID1254
Feature activation-8.14e-03

Pos logit contributions
 nation+0.000
 knock+0.000
unts+0.000
ce+0.000
 blast+0.000
Neg logit contributions
 brand-0.000
 exactly-0.000
but-0.000
uit-0.000
 perched-0.000
Token the
Token ID262
Feature activation-2.39e-03

Pos logit contributions
 nation+0.008
 wind+0.008
 knock+0.007
 guys+0.007
unts+0.007
Neg logit contributions
uit-0.016
but-0.016
 brand-0.014
ieved-0.014
ishly-0.014
Token water
Token ID1660
Feature activation-1.41e-03

Pos logit contributions
 nation+0.003
 guys+0.002
 wind+0.002
 storm+0.002
 knock+0.002
Neg logit contributions
but-0.005
 perched-0.004
 exactly-0.004
uit-0.004
ieved-0.004
Token se
Token ID384
Feature activation-1.02e-01

Pos logit contributions
 nation+0.002
 knock+0.002
 guys+0.002
 spe+0.002
 e+0.002
Neg logit contributions
 brand-0.002
 exactly-0.002
but-0.002
uit-0.002
 perched-0.002
Tokeneping
Token ID7213
Feature activation-1.42e-02

Pos logit contributions
 nation+0.146
 knock+0.136
 spe+0.133
 blast+0.131
ce+0.130
Neg logit contributions
 exactly-0.137
 brand-0.134
uit-0.129
but-0.127
 perched-0.119
Token in
Token ID287
Feature activation-3.27e-03

Pos logit contributions
 guys+0.020
 nation+0.019
 knock+0.019
 e+0.018
 wind+0.018
Neg logit contributions
 brand-0.020
 exactly-0.019
but-0.019
uit-0.018
 perched-0.017
Token.
Token ID13
Feature activation+1.21e-03

Pos logit contributions
 nation+0.004
 spe+0.003
 knock+0.003
 blast+0.003
unts+0.003
Neg logit contributions
but-0.006
 brand-0.005
 exactly-0.005
uit-0.005
ieved-0.005
Token They
Token ID1119
Feature activation-1.87e-02

Pos logit contributions
 brand+0.003
 exactly+0.003
uit+0.002
 perched+0.002
but+0.002
Neg logit contributions
 nation-0.001
 Find-0.001
unts-0.001
 knock-0.001
 Danger-0.001
Token did
Token ID750
Feature activation-1.65e-02

Pos logit contributions
 knock+0.020
 nation+0.019
 blast+0.018
 spe+0.017
 guys+0.017
Neg logit contributions
 brand-0.035
but-0.034
uit-0.033
 exactly-0.032
sometimes-0.031
Tokenapped
Token ID6320
Feature activation-1.65e-02

Pos logit contributions
 nation+0.008
 knock+0.008
ce+0.007
unts+0.007
ass+0.007
Neg logit contributions
 brand-0.009
 exactly-0.008
but-0.008
uit-0.008
 perched-0.008
Token together
Token ID1978
Feature activation+2.11e-03

Pos logit contributions
ot+0.020
 nation+0.018
unts+0.017
ia+0.016
 blast+0.016
Neg logit contributions
uit-0.027
but-0.027
 brand-0.027
 perched-0.026
sometimes-0.026
Token until
Token ID1566
Feature activation+3.28e-03

Pos logit contributions
uit+0.003
 brand+0.003
 exactly+0.003
but+0.003
sometimes+0.003
Neg logit contributions
 nation-0.002
 knock-0.002
 blast-0.002
 team-0.002
ot-0.002
Token all
Token ID477
Feature activation-2.75e-02

Pos logit contributions
 brand+0.005
 exactly+0.005
but+0.005
uit+0.005
 perched+0.005
Neg logit contributions
 nation-0.004
 blast-0.004
 knock-0.004
ce-0.004
 spe-0.003
Token the
Token ID262
Feature activation-2.93e-02

Pos logit contributions
 nation+0.035
 knock+0.032
ce+0.032
unts+0.032
ass+0.031
Neg logit contributions
 brand-0.041
 exactly-0.040
but-0.038
uit-0.038
 perched-0.037
Token se
Token ID384
Feature activation-1.02e-01

Pos logit contributions
 nation+0.029
 knock+0.026
ce+0.025
 blast+0.025
unts+0.025
Neg logit contributions
 exactly-0.052
 brand-0.052
but-0.051
uit-0.050
 perched-0.049
Tokenag
Token ID363
Feature activation-3.39e-02

Pos logit contributions
 nation+0.127
ce+0.125
unts+0.123
ass+0.121
ia+0.117
Neg logit contributions
 brand-0.151
 exactly-0.146
but-0.141
uit-0.139
 perched-0.138
Tokenull
Token ID724
Feature activation-4.89e-02

Pos logit contributions
ot+0.028
 nation+0.027
ia+0.027
ce+0.025
unts+0.024
Neg logit contributions
 brand-0.071
 exactly-0.065
but-0.064
uit-0.064
 ang-0.062
Tokens
Token ID82
Feature activation-3.55e-02

Pos logit contributions
ce+0.052
 spe+0.051
 nation+0.051
 knock+0.050
 e+0.048
Neg logit contributions
 brand-0.081
 exactly-0.080
uit-0.079
but-0.076
 unconscious-0.072
Token flew
Token ID13112
Feature activation-1.35e-02

Pos logit contributions
ot+0.042
 ever+0.041
 spe+0.039
 nation+0.039
 wind+0.038
Neg logit contributions
but-0.057
 brand-0.056
uit-0.055
 exactly-0.053
sometimes-0.052
Token away
Token ID1497
Feature activation+8.02e-03

Pos logit contributions
ot+0.016
 wind+0.016
ny+0.015
 ever+0.014
 nation+0.014
Neg logit contributions
 brand-0.020
sometimes-0.019
 elbow-0.019
 struggling-0.019
but-0.019
Token.
Token ID13
Feature activation-1.74e-03

Pos logit contributions
 knock+0.003
air+0.003
ce+0.003
 nation+0.003
ia+0.003
Neg logit contributions
 brand-0.003
uit-0.003
 exactly-0.003
 perched-0.003
 unconscious-0.003
Token She
Token ID1375
Feature activation-1.40e-03

Pos logit contributions
 Find+0.002
 nation+0.002
 knock+0.002
unts+0.002
 Danger+0.002
Neg logit contributions
 exactly-0.003
 brand-0.003
uit-0.003
but-0.003
 perched-0.002
Token used
Token ID973
Feature activation-1.02e-02

Pos logit contributions
 nation+0.002
 knock+0.002
ce+0.002
 blast+0.002
 spe+0.002
Neg logit contributions
 brand-0.002
 exactly-0.002
uit-0.002
but-0.002
 unconscious-0.002
Token a
Token ID257
Feature activation-1.21e-02

Pos logit contributions
 spe+0.011
 blast+0.010
 turns+0.010
 eight+0.010
 knock+0.010
Neg logit contributions
but-0.017
sometimes-0.016
 brand-0.016
uit-0.016
ieved-0.016
Token m
Token ID285
Feature activation-1.43e-02

Pos logit contributions
 knock+0.010
 blast+0.010
 nation+0.009
 spe+0.009
unts+0.008
Neg logit contributions
but-0.025
sometimes-0.025
uit-0.024
 exactly-0.024
ishly-0.024
Tokenitt
Token ID715
Feature activation-1.02e-01

Pos logit contributions
 nation+0.016
 guys+0.014
 blast+0.014
ye+0.014
 knock+0.013
Neg logit contributions
 exactly-0.025
 brand-0.024
but-0.023
uit-0.023
 perched-0.021
Token to
Token ID284
Feature activation-2.34e-02

Pos logit contributions
 nation+0.110
ce+0.102
 knock+0.102
unts+0.101
ass+0.100
Neg logit contributions
 brand-0.168
 exactly-0.165
uit-0.158
but-0.158
 perched-0.152
Token take
Token ID1011
Feature activation-1.82e-02

Pos logit contributions
 knock+0.022
 blast+0.020
 nation+0.020
ot+0.020
ce+0.019
Neg logit contributions
 brand-0.044
but-0.043
 exactly-0.043
 unconscious-0.042
uit-0.041
Token out
Token ID503
Feature activation-6.56e-03

Pos logit contributions
 nation+0.034
unts+0.032
 knock+0.032
ye+0.031
oke+0.031
Neg logit contributions
 exactly-0.017
 brand-0.016
uit-0.015
but-0.014
 perched-0.013
Token the
Token ID262
Feature activation-8.51e-03

Pos logit contributions
 nation+0.008
 knock+0.008
ce+0.008
ot+0.008
 spe+0.007
Neg logit contributions
but-0.010
 brand-0.010
ieved-0.009
uit-0.009
 perched-0.009
Token pan
Token ID3425
Feature activation-3.55e-02

Pos logit contributions
 knock+0.008
 nation+0.008
 spe+0.008
 blast+0.008
 e+0.007
Neg logit contributions
uit-0.016
but-0.016
 exactly-0.015
 perched-0.015
 brand-0.015
Token very
Token ID845
Feature activation+2.49e-02

Pos logit contributions
but+0.006
uit+0.005
sometimes+0.005
 exactly+0.005
 perched+0.005
Neg logit contributions
 knock-0.003
 nation-0.003
 blast-0.003
 spe-0.003
 person-0.003
Token special
Token ID2041
Feature activation+4.04e-03

Pos logit contributions
 brand+0.035
 exactly+0.035
but+0.031
 perched+0.029
uit+0.029
Neg logit contributions
 nation-0.033
ass-0.032
ce-0.032
 blast-0.032
unts-0.032
Token toy
Token ID13373
Feature activation+2.69e-02

Pos logit contributions
but+0.008
 exactly+0.007
ieved+0.007
sometimes+0.007
 unconscious+0.007
Neg logit contributions
 nation-0.004
 knock-0.004
 person-0.004
 blast-0.003
 wake-0.003
Token for
Token ID329
Feature activation-3.55e-03

Pos logit contributions
ieved+0.044
 exactly+0.039
 perched+0.039
 brand+0.039
 ang+0.039
Neg logit contributions
 person-0.032
 guys-0.031
 nation-0.031
 knock-0.030
 e-0.028
Token L
Token ID406
Feature activation-2.25e-02

Pos logit contributions
 nation+0.006
 knock+0.006
ce+0.006
unts+0.006
ass+0.006
Neg logit contributions
 brand-0.004
 exactly-0.004
but-0.003
uit-0.003
 perched-0.003
Tokenott
Token ID1252
Feature activation-1.01e-01

Pos logit contributions
 nation+0.028
 knock+0.026
 blast+0.025
unts+0.025
 spe+0.025
Neg logit contributions
 exactly-0.035
 brand-0.034
but-0.033
uit-0.033
 perched-0.031
Tokenie
Token ID494
Feature activation-1.04e-02

Pos logit contributions
ass+0.087
ce+0.086
 nation+0.086
ia+0.084
ot+0.079
Neg logit contributions
 brand-0.196
 exactly-0.181
but-0.176
 perched-0.175
uit-0.174
Token.
Token ID13
Feature activation-2.26e-03

Pos logit contributions
 nation+0.012
 person+0.011
 team+0.011
ot+0.011
ce+0.011
Neg logit contributions
 brand-0.016
 exactly-0.016
iddled-0.016
but-0.016
ieved-0.016
Token It
Token ID632
Feature activation-1.70e-02

Pos logit contributions
 nation+0.003
ce+0.003
unts+0.003
 knock+0.003
ass+0.003
Neg logit contributions
 brand-0.003
 exactly-0.003
but-0.003
uit-0.003
 perched-0.003
Token was
Token ID373
Feature activation-4.25e-03

Pos logit contributions
 knock+0.016
ce+0.015
 nation+0.015
 spe+0.015
 e+0.014
Neg logit contributions
 exactly-0.029
but-0.029
uit-0.028
 brand-0.028
 unconscious-0.027
Token a
Token ID257
Feature activation-3.50e-03

Pos logit contributions
 spe+0.005
 knock+0.005
 e+0.005
ot+0.005
 wind+0.005
Neg logit contributions
but-0.007
ishly-0.006
uit-0.006
sometimes-0.006
oding-0.006
Token the
Token ID262
Feature activation-2.79e-04

Pos logit contributions
 knock+0.013
 nation+0.013
 blast+0.012
 Danger+0.012
ce+0.012
Neg logit contributions
 brand-0.017
 exactly-0.016
but-0.016
uit-0.015
 perched-0.015
Token surf
Token ID9053
Feature activation-6.11e-03

Pos logit contributions
 nation+0.000
 knock+0.000
 person+0.000
 e+0.000
ye+0.000
Neg logit contributions
 exactly-0.001
uit-0.001
but-0.000
 perched-0.000
 brand-0.000
Token and
Token ID290
Feature activation-6.86e-03

Pos logit contributions
 nation+0.006
ce+0.006
 team+0.005
air+0.005
ye+0.005
Neg logit contributions
 exactly-0.011
 brand-0.011
 perched-0.010
 unconscious-0.010
 struggling-0.010
Token saw
Token ID2497
Feature activation-1.32e-02

Pos logit contributions
 knock+0.007
 nation+0.007
 blast+0.007
 e+0.006
ye+0.006
Neg logit contributions
 brand-0.011
uit-0.011
 exactly-0.011
but-0.011
sometimes-0.010
Token the
Token ID262
Feature activation-1.11e-02

Pos logit contributions
 e+0.013
 nation+0.013
 knock+0.012
 Mir+0.012
 person+0.012
Neg logit contributions
ieved-0.022
but-0.021
uit-0.021
aj-0.021
 brand-0.021
Token se
Token ID384
Feature activation-1.01e-01

Pos logit contributions
 nation+0.011
 knock+0.009
unts+0.009
 blast+0.009
 guys+0.009
Neg logit contributions
 exactly-0.021
uit-0.021
 brand-0.020
but-0.020
 perched-0.020
Tokenag
Token ID363
Feature activation-2.40e-02

Pos logit contributions
 nation+0.185
 spe+0.177
 blast+0.176
 knock+0.175
 guys+0.171
Neg logit contributions
 exactly-0.095
 brand-0.091
uit-0.088
but-0.088
ieved-0.073
Tokenull
Token ID724
Feature activation-5.16e-02

Pos logit contributions
 nation+0.016
ass+0.013
ce+0.013
 knock+0.013
unts+0.013
Neg logit contributions
 brand-0.051
 exactly-0.050
uit-0.049
but-0.048
 perched-0.047
Token up
Token ID510
Feature activation-2.87e-02

Pos logit contributions
 knock+0.065
 nation+0.063
 e+0.058
 spe+0.057
ce+0.056
Neg logit contributions
 brand-0.075
uit-0.075
 exactly-0.074
but-0.073
 unconscious-0.066
Token in
Token ID287
Feature activation-6.35e-03

Pos logit contributions
 ever+0.026
ot+0.026
 nation+0.026
 spe+0.026
unts+0.025
Neg logit contributions
 brand-0.051
uit-0.051
ishly-0.046
 exactly-0.045
 unconscious-0.044
Token the
Token ID262
Feature activation-3.29e-03

Pos logit contributions
 nation+0.008
 blast+0.008
 spe+0.008
 knock+0.007
unts+0.007
Neg logit contributions
 brand-0.010
but-0.010
 exactly-0.010
uit-0.009
 perched-0.009
Token
Token ID198
Feature activation-6.90e-04

Pos logit contributions
 brand+0.013
 exactly+0.013
uit+0.012
 intact+0.012
 perched+0.012
Neg logit contributions
ot-0.007
 Find-0.007
unts-0.007
 nation-0.007
ce-0.006
Token
Token ID198
Feature activation+1.90e-03

Pos logit contributions
unts+0.001
ot+0.001
ia+0.001
ce+0.001
 knock+0.001
Neg logit contributions
 brand-0.001
 exactly-0.001
uit-0.001
 perched-0.001
but-0.001
TokenSo
Token ID2396
Feature activation+2.29e-02

Pos logit contributions
 brand+0.003
 exactly+0.003
uit+0.003
 perched+0.003
but+0.003
Neg logit contributions
 nation-0.002
unts-0.002
ce-0.002
ot-0.002
 knock-0.002
Token the
Token ID262
Feature activation-9.85e-04

Pos logit contributions
 brand+0.040
but+0.037
 exactly+0.037
uit+0.037
 perched+0.035
Neg logit contributions
 nation-0.025
 blast-0.023
ce-0.022
 knock-0.022
ot-0.022
Token b
Token ID275
Feature activation-4.33e-02

Pos logit contributions
 nation+0.001
 knock+0.001
ce+0.001
unts+0.001
 blast+0.001
Neg logit contributions
 brand-0.002
 exactly-0.002
but-0.002
uit-0.002
 perched-0.002
Tokenunn
Token ID20935
Feature activation-1.01e-01

Pos logit contributions
 nation+0.030
 knock+0.029
 blast+0.029
 guys+0.025
 Find+0.024
Neg logit contributions
 exactly-0.090
uit-0.087
but-0.086
 brand-0.085
 perched-0.082
Tokenies
Token ID444
Feature activation+6.24e-03

Pos logit contributions
 nation+0.107
ce+0.098
ass+0.096
unts+0.096
 knock+0.095
Neg logit contributions
 brand-0.175
 exactly-0.168
uit-0.163
but-0.161
 perched-0.155
Token and
Token ID290
Feature activation-2.05e-03

Pos logit contributions
 brand+0.012
 exactly+0.010
but+0.010
uit+0.010
 unconscious+0.010
Neg logit contributions
 nation-0.007
 knock-0.007
 goes-0.006
 team-0.006
ot-0.006
Token Bob
Token ID5811
Feature activation+3.60e-03

Pos logit contributions
 nation+0.002
 knock+0.002
ce+0.002
 spe+0.002
 guys+0.002
Neg logit contributions
 brand-0.003
 exactly-0.003
but-0.003
uit-0.003
 unconscious-0.003
Token enjoyed
Token ID8359
Feature activation-5.02e-04

Pos logit contributions
 brand+0.006
 exactly+0.006
but+0.006
 unconscious+0.005
uit+0.005
Neg logit contributions
 nation-0.004
ce-0.004
unts-0.004
 knock-0.004
ith-0.004
Token some
Token ID617
Feature activation-1.74e-02

Pos logit contributions
 nation+0.000
 knock+0.000
 spe+0.000
 blast+0.000
 team+0.000
Neg logit contributions
uit-0.001
but-0.001
 unconscious-0.001
oding-0.001
ieved-0.001
Token watched
Token ID7342
Feature activation-7.75e-04

Pos logit contributions
 nation+0.020
 knock+0.019
 spe+0.019
 blast+0.018
unts+0.018
Neg logit contributions
 brand-0.030
 exactly-0.029
uit-0.028
but-0.028
 unconscious-0.027
Token the
Token ID262
Feature activation-1.74e-03

Pos logit contributions
 nation+0.001
 knock+0.001
ce+0.001
 guys+0.001
unts+0.001
Neg logit contributions
 brand-0.001
 exactly-0.001
uit-0.001
but-0.001
 perched-0.001
Token waves
Token ID9813
Feature activation-3.52e-02

Pos logit contributions
 nation+0.002
 knock+0.002
 blast+0.001
 guys+0.001
 storm+0.001
Neg logit contributions
 exactly-0.003
uit-0.003
but-0.003
 brand-0.003
 perched-0.003
Token and
Token ID290
Feature activation-1.47e-02

Pos logit contributions
 nation+0.044
 knock+0.042
unts+0.041
ce+0.041
 blast+0.040
Neg logit contributions
 brand-0.053
 exactly-0.053
but-0.050
uit-0.049
 perched-0.047
Token the
Token ID262
Feature activation-1.06e-02

Pos logit contributions
 nation+0.020
 knock+0.019
 spe+0.019
 blast+0.018
ce+0.018
Neg logit contributions
uit-0.021
 exactly-0.021
 brand-0.021
but-0.021
sometimes-0.019
Token se
Token ID384
Feature activation-1.01e-01

Pos logit contributions
 nation+0.011
 wind+0.010
 knock+0.010
 storm+0.009
 blast+0.009
Neg logit contributions
 exactly-0.021
but-0.020
uit-0.020
 brand-0.019
ieved-0.018
Tokenag
Token ID363
Feature activation-2.52e-02

Pos logit contributions
 nation+0.156
 knock+0.147
 blast+0.144
 spe+0.144
 guys+0.141
Neg logit contributions
 exactly-0.122
 brand-0.119
uit-0.114
but-0.113
 perched-0.105
Tokenull
Token ID724
Feature activation-5.16e-02

Pos logit contributions
 nation+0.013
ot+0.012
ia+0.011
ce+0.011
ass+0.011
Neg logit contributions
 brand-0.059
 exactly-0.057
but-0.054
uit-0.054
 perched-0.053
Tokens
Token ID82
Feature activation-4.24e-02

Pos logit contributions
 nation+0.064
 knock+0.062
ce+0.061
unts+0.059
 blast+0.059
Neg logit contributions
 brand-0.076
 exactly-0.076
uit-0.073
but-0.072
 unconscious-0.068
Token.
Token ID13
Feature activation-8.67e-03

Pos logit contributions
 spe+0.045
 nation+0.044
 wind+0.043
 wake+0.042
 guys+0.042
Neg logit contributions
uit-0.071
but-0.070
 exactly-0.069
 brand-0.066
ieved-0.065
Token They
Token ID1119
Feature activation-1.89e-02

Pos logit contributions
 nation+0.008
 knock+0.008
ce+0.008
unts+0.008
 blast+0.007
Neg logit contributions
 exactly-0.015
 brand-0.015
uit-0.015
but-0.015
 perched-0.014
Token He
Token ID679
Feature activation-5.24e-03

Pos logit contributions
 exactly+0.003
 perched+0.002
 brand+0.002
uit+0.002
but+0.002
Neg logit contributions
 nation-0.001
 Find-0.001
ia-0.001
ass-0.001
unts-0.001
Token saw
Token ID2497
Feature activation-2.30e-03

Pos logit contributions
 nation+0.006
 knock+0.006
ce+0.006
unts+0.006
 blast+0.006
Neg logit contributions
 brand-0.008
 exactly-0.008
uit-0.008
but-0.008
 perched-0.007
Token Jill
Token ID22772
Feature activation+4.86e-03

Pos logit contributions
 knock+0.002
 nation+0.002
 wind+0.002
 guys+0.002
 person+0.002
Neg logit contributions
but-0.004
ieved-0.004
uit-0.004
 perched-0.004
shadow-0.003
Token with
Token ID351
Feature activation+1.98e-03

Pos logit contributions
 brand+0.009
 exactly+0.009
uit+0.008
but+0.008
?,+0.008
Neg logit contributions
 knock-0.005
 nation-0.005
ce-0.005
 wake-0.004
air-0.004
Token her
Token ID607
Feature activation+8.15e-03

Pos logit contributions
but+0.003
 exactly+0.003
uit+0.003
sometimes+0.003
 perched+0.003
Neg logit contributions
 spe-0.002
 nation-0.002
 knock-0.002
 blast-0.002
 wind-0.002
Token math
Token ID10688
Feature activation+5.14e-02

Pos logit contributions
 perched+0.016
but+0.016
sometimes+0.016
uit+0.016
iddled+0.016
Neg logit contributions
 nation-0.008
 team-0.007
 knock-0.007
 page-0.007
 person-0.006
Token book
Token ID1492
Feature activation+1.94e-02

Pos logit contributions
 exactly+0.094
uit+0.094
but+0.094
 perched+0.093
 brand+0.091
Neg logit contributions
 nation-0.053
 guys-0.048
 knock-0.046
 blast-0.044
 breaks-0.043
Token.
Token ID13
Feature activation+6.87e-04

Pos logit contributions
 perched+0.030
 ang+0.030
but+0.030
 brand+0.030
 exactly+0.029
Neg logit contributions
 nation-0.020
 knock-0.019
 ever-0.019
ia-0.018
ce-0.018
Token He
Token ID679
Feature activation-3.84e-03

Pos logit contributions
 exactly+0.001
 brand+0.001
 perched+0.001
uit+0.001
 struggling+0.001
Neg logit contributions
 nation-0.001
 Find-0.001
ia-0.001
unts-0.001
ass-0.001
Token did
Token ID750
Feature activation-7.32e-03

Pos logit contributions
 nation+0.005
 knock+0.005
ce+0.004
unts+0.004
 blast+0.004
Neg logit contributions
 brand-0.006
 exactly-0.006
uit-0.006
but-0.006
 perched-0.005
Token not
Token ID407
Feature activation+1.59e-02

Pos logit contributions
 knock+0.007
ia+0.006
 blast+0.005
 nation+0.005
 breaks+0.005
Neg logit contributions
uit-0.014
 perched-0.014
ishly-0.014
 unconscious-0.014
 brand-0.013
Token like
Token ID588
Feature activation+2.94e-02

Pos logit contributions
 nation+0.012
 knock+0.011
 wake+0.011
unts+0.011
 blast+0.011
Neg logit contributions
 brand-0.019
 exactly-0.018
 unconscious-0.018
but-0.018
 perched-0.018
Token to
Token ID284
Feature activation+3.02e-03

Pos logit contributions
 exactly+0.043
 brand+0.042
uit+0.041
but+0.038
 perched+0.037
Neg logit contributions
 nation-0.041
 knock-0.038
 guys-0.038
 blast-0.037
unts-0.036
Token play
Token ID711
Feature activation+2.66e-02

Pos logit contributions
 exactly+0.005
 brand+0.005
but+0.005
uit+0.005
 perched+0.005
Neg logit contributions
 nation-0.003
 knock-0.003
 blast-0.003
 guys-0.003
ot-0.003
Token with
Token ID351
Feature activation-2.66e-03

Pos logit contributions
 brand+0.044
 exactly+0.043
 unconscious+0.043
 thanking+0.043
but+0.042
Neg logit contributions
 nation-0.034
 guys-0.033
 team-0.033
 blast-0.030
 dive-0.025
Token their
Token ID511
Feature activation-5.46e-03

Pos logit contributions
 nation+0.003
 guys+0.003
 knock+0.003
 blast+0.003
 spe+0.003
Neg logit contributions
but-0.004
 exactly-0.004
uit-0.004
 brand-0.004
 perched-0.004
Token toy
Token ID13373
Feature activation+3.96e-02

Pos logit contributions
 nation+0.007
 knock+0.006
 team+0.006
 guys+0.006
 blast+0.006
Neg logit contributions
 exactly-0.009
but-0.009
uit-0.009
 brand-0.008
 perched-0.008
Token cars
Token ID5006
Feature activation+3.15e-02

Pos logit contributions
 brand+0.075
 exactly+0.074
but+0.073
ieved+0.072
 perched+0.071
Neg logit contributions
 nation-0.041
 guys-0.038
 team-0.035
 knock-0.030
 person-0.029
Token.
Token ID13
Feature activation-6.46e-03

Pos logit contributions
 brand+0.057
 exactly+0.052
 perched+0.049
 unconscious+0.048
ieved+0.048
Neg logit contributions
 nation-0.038
 team-0.036
 guys-0.035
ot-0.032
unts-0.031
Token They
Token ID1119
Feature activation-1.02e-02

Pos logit contributions
 nation+0.007
 knock+0.006
 blast+0.006
unts+0.006
ce+0.006
Neg logit contributions
 brand-0.011
 exactly-0.011
but-0.010
uit-0.010
 perched-0.010
Token have
Token ID423
Feature activation-4.38e-04

Pos logit contributions
 nation+0.009
ce+0.008
ass+0.008
unts+0.008
 knock+0.008
Neg logit contributions
 brand-0.019
 exactly-0.019
uit-0.018
but-0.018
 perched-0.017
Token many
Token ID867
Feature activation-2.20e-02

Pos logit contributions
 nation+0.001
 blast+0.001
 knock+0.001
 guys+0.001
 spe+0.001
Neg logit contributions
but-0.001
 exactly-0.001
uit-0.001
 brand-0.001
 unconscious-0.001
Token loud
Token ID7812
Feature activation-2.56e-02

Pos logit contributions
 brand+0.035
 exactly+0.035
 something+0.029
but+0.028
 himself+0.026
Neg logit contributions
ass-0.046
oke-0.044
 nation-0.044
uders-0.044
 Danger-0.043
Token that
Token ID326
Feature activation-1.12e-02

Pos logit contributions
 nation+0.035
 guys+0.033
ce+0.033
 knock+0.033
 blast+0.032
Neg logit contributions
 exactly-0.035
 brand-0.035
uit-0.033
but-0.032
 unconscious-0.031
Token it
Token ID340
Feature activation-1.87e-02

Pos logit contributions
 nation+0.018
ce+0.017
unts+0.017
 knock+0.017
ass+0.017
Neg logit contributions
 exactly-0.012
 brand-0.012
but-0.011
uit-0.011
 perched-0.011
Token hurt
Token ID5938
Feature activation+1.10e-02

Pos logit contributions
 nation+0.023
 knock+0.021
ce+0.021
unts+0.021
 blast+0.021
Neg logit contributions
 brand-0.028
 exactly-0.028
uit-0.027
but-0.027
 perched-0.026
Token his
Token ID465
Feature activation+1.03e-02

Pos logit contributions
 brand+0.016
but+0.014
uit+0.014
 exactly+0.014
 perched+0.013
Neg logit contributions
 nation-0.015
 guys-0.015
ia-0.014
ass-0.014
 ever-0.014
Token friend
Token ID1545
Feature activation+4.00e-02

Pos logit contributions
 exactly+0.018
 brand+0.017
but+0.017
uit+0.017
 perched+0.017
Neg logit contributions
 nation-0.011
 guys-0.010
 knock-0.010
ot-0.009
 blast-0.009
Token's
Token ID338
Feature activation-1.53e-02

Pos logit contributions
 brand+0.070
but+0.066
 exactly+0.065
sometimes+0.060
uit+0.059
Neg logit contributions
 guys-0.044
 nation-0.043
ot-0.043
ce-0.042
 e-0.041
Token ears
Token ID11368
Feature activation+3.04e-03

Pos logit contributions
 nation+0.014
 knock+0.012
 guys+0.012
 wake+0.012
unts+0.012
Neg logit contributions
 exactly-0.030
but-0.030
uit-0.029
 brand-0.029
 perched-0.028
Token.
Token ID13
Feature activation-1.77e-03

Pos logit contributions
 brand+0.005
 exactly+0.005
but+0.005
ieved+0.005
uit+0.005
Neg logit contributions
 wake-0.003
 comes-0.003
 guys-0.003
 nation-0.003
 ever-0.003
Token His
Token ID2399
Feature activation+6.45e-03

Pos logit contributions
 nation+0.003
 knock+0.002
unts+0.002
ce+0.002
 Find+0.002
Neg logit contributions
 brand-0.002
 exactly-0.002
uit-0.002
but-0.002
 unconscious-0.002
Token friends
Token ID2460
Feature activation+3.16e-02

Pos logit contributions
 brand+0.010
 exactly+0.010
but+0.009
uit+0.009
 perched+0.009
Neg logit contributions
 nation-0.008
 knock-0.008
ce-0.007
unts-0.007
 blast-0.007
Token the
Token ID262
Feature activation+8.43e-03

Pos logit contributions
 nation+0.012
 knock+0.012
 blast+0.012
ce+0.011
 Find+0.011
Neg logit contributions
 exactly-0.013
 brand-0.013
but-0.013
uit-0.013
 perched-0.012
Token sink
Token ID14595
Feature activation-8.26e-03

Pos logit contributions
 exactly+0.013
but+0.013
uit+0.012
 perched+0.012
 brand+0.012
Neg logit contributions
 nation-0.011
 knock-0.010
 guys-0.010
 blast-0.010
ye-0.009
Token and
Token ID290
Feature activation-5.82e-04

Pos logit contributions
 nation+0.010
ce+0.008
 knock+0.008
head+0.008
 team+0.008
Neg logit contributions
 brand-0.014
 exactly-0.013
 perched-0.013
uit-0.013
 unconscious-0.013
Token wash
Token ID13502
Feature activation+1.48e-02

Pos logit contributions
 knock+0.001
 nation+0.001
 blast+0.000
 Find+0.000
 spe+0.000
Neg logit contributions
 exactly-0.001
uit-0.001
but-0.001
 brand-0.001
 unconscious-0.001
Token your
Token ID534
Feature activation+1.53e-02

Pos logit contributions
 brand+0.026
but+0.025
uit+0.025
 exactly+0.025
 perched+0.025
Neg logit contributions
ot-0.014
 nation-0.013
ce-0.013
 guys-0.013
air-0.013
Token doll
Token ID3654
Feature activation+4.71e-02

Pos logit contributions
but+0.027
 exactly+0.027
uit+0.026
 brand+0.026
 perched+0.025
Neg logit contributions
 nation-0.016
 knock-0.015
 guys-0.014
ce-0.014
 blast-0.014
Token's
Token ID338
Feature activation-1.96e-02

Pos logit contributions
 brand+0.069
 exactly+0.067
uit+0.066
but+0.064
 perched+0.062
Neg logit contributions
 nation-0.060
 knock-0.057
ce-0.057
ot-0.056
 guys-0.056
Token arm
Token ID3211
Feature activation+1.78e-02

Pos logit contributions
ot+0.018
 nation+0.018
 spe+0.017
 knock+0.017
 e+0.016
Neg logit contributions
but-0.037
uit-0.035
 exactly-0.035
sometimes-0.035
 brand-0.034
Token."
Token ID526
Feature activation+1.23e-03

Pos logit contributions
 exactly+0.032
uit+0.031
 brand+0.030
 perched+0.029
 ang+0.028
Neg logit contributions
ce-0.017
ot-0.016
 knock-0.016
air-0.016
 guys-0.016
Token
Token ID198
Feature activation-7.40e-04

Pos logit contributions
 brand+0.002
 exactly+0.002
 perched+0.002
uit+0.002
 atop+0.002
Neg logit contributions
 Find-0.001
unts-0.001
ia-0.001
ot-0.001
 knock-0.001
Token
Token ID198
Feature activation+1.15e-03

Pos logit contributions
unts+0.001
 knock+0.001
ia+0.001
ce+0.001
ot+0.001
Neg logit contributions
 brand-0.001
 exactly-0.001
uit-0.001
ishly-0.001
 perched-0.001
Token knew
Token ID2993
Feature activation-4.85e-03

Pos logit contributions
 nation+0.004
 knock+0.003
unts+0.003
ass+0.003
ce+0.003
Neg logit contributions
 exactly-0.004
 brand-0.004
but-0.003
 perched-0.003
uit-0.003
Token it
Token ID340
Feature activation-1.60e-03

Pos logit contributions
 knock+0.005
 nation+0.005
 ever+0.005
ia+0.005
 spe+0.005
Neg logit contributions
 brand-0.008
uit-0.008
but-0.007
iddled-0.007
ieved-0.007
Token was
Token ID373
Feature activation+1.05e-03

Pos logit contributions
 nation+0.002
 knock+0.002
ce+0.002
unts+0.002
ass+0.002
Neg logit contributions
 brand-0.002
 exactly-0.002
but-0.002
uit-0.002
 perched-0.002
Token going
Token ID1016
Feature activation-1.56e-02

Pos logit contributions
but+0.002
uit+0.002
 brand+0.002
ishly+0.002
iddled+0.002
Neg logit contributions
 storm-0.001
 wind-0.001
 nation-0.001
 spe-0.001
 guys-0.001
Token to
Token ID284
Feature activation-1.04e-02

Pos logit contributions
 guys+0.019
 nation+0.019
 knock+0.018
 storm+0.018
ce+0.018
Neg logit contributions
 exactly-0.023
uit-0.022
 brand-0.022
but-0.021
 unconscious-0.020
Token rain
Token ID6290
Feature activation+3.99e-02

Pos logit contributions
 nation+0.010
 knock+0.009
 blast+0.009
ce+0.009
unts+0.009
Neg logit contributions
 brand-0.019
 exactly-0.019
but-0.018
uit-0.018
 perched-0.017
Token soon
Token ID2582
Feature activation-3.53e-02

Pos logit contributions
 brand+0.055
 exactly+0.054
 perched+0.049
 unconscious+0.049
uit+0.048
Neg logit contributions
 nation-0.054
 knock-0.053
 guys-0.053
ce-0.051
 blast-0.050
Token!
Token ID0
Feature activation-1.89e-03

Pos logit contributions
 nation+0.039
 knock+0.038
 ever+0.038
 comes+0.037
 wake+0.037
Neg logit contributions
 brand-0.057
uit-0.053
iddled-0.053
but-0.053
 exactly-0.053
Token
Token ID198
Feature activation-5.55e-03

Pos logit contributions
 Find+0.003
 nation+0.003
 knock+0.003
ia+0.003
unts+0.003
Neg logit contributions
 brand-0.003
 exactly-0.002
uit-0.002
but-0.002
 perched-0.002
Token
Token ID198
Feature activation-1.48e-03

Pos logit contributions
unts+0.007
 knock+0.007
ia+0.007
 nation+0.007
ce+0.007
Neg logit contributions
 brand-0.008
 exactly-0.007
uit-0.007
but-0.007
 unconscious-0.006
TokenS
Token ID50
Feature activation-1.40e-02

Pos logit contributions
 nation+0.002
 knock+0.002
ce+0.002
unts+0.002
ass+0.002
Neg logit contributions
 exactly-0.002
 brand-0.002
but-0.002
uit-0.002
 perched-0.002
Token isn
Token ID2125
Feature activation+8.11e-03

Pos logit contributions
 nation+0.010
 guys+0.009
 Find+0.009
 knock+0.009
 happens+0.009
Neg logit contributions
 brand-0.023
uit-0.023
 exactly-0.022
but-0.022
 perched-0.021
Token't
Token ID470
Feature activation-2.21e-02

Pos logit contributions
 exactly+0.012
 brand+0.012
but+0.012
uit+0.012
 perched+0.011
Neg logit contributions
 nation-0.010
 spe-0.009
ce-0.009
unts-0.009
ye-0.009
Token yours
Token ID12431
Feature activation+2.99e-02

Pos logit contributions
 guys+0.029
 nation+0.028
 Find+0.025
ass+0.025
unts+0.025
Neg logit contributions
 brand-0.034
uit-0.032
but-0.032
 exactly-0.030
ishly-0.030
Token.
Token ID13
Feature activation-5.58e-03

Pos logit contributions
 brand+0.056
 perched+0.052
 ang+0.052
ieved+0.051
 unconscious+0.051
Neg logit contributions
 ever-0.030
 nation-0.027
 guys-0.026
ass-0.026
 person-0.025
Token Always
Token ID16622
Feature activation+4.92e-03

Pos logit contributions
 nation+0.007
 knock+0.007
ce+0.007
 spe+0.007
 guys+0.007
Neg logit contributions
 exactly-0.008
 brand-0.008
but-0.008
uit-0.007
 perched-0.007
Token be
Token ID307
Feature activation+5.73e-03

Pos logit contributions
 brand+0.009
but+0.009
 exactly+0.009
uit+0.009
 unconscious+0.008
Neg logit contributions
 nation-0.005
 knock-0.004
 Find-0.004
 blast-0.004
unts-0.004
Token honest
Token ID5508
Feature activation-3.36e-02

Pos logit contributions
uit+0.011
 brand+0.010
but+0.010
 altogether+0.010
bat+0.009
Neg logit contributions
 nation-0.005
unts-0.005
ith-0.005
 knock-0.005
 Find-0.005
Token and
Token ID290
Feature activation-1.31e-02

Pos logit contributions
 nation+0.037
 person+0.034
ot+0.033
ith+0.033
 guys+0.033
Neg logit contributions
 brand-0.057
uit-0.053
 perched-0.052
 ang-0.051
 exactly-0.051
Token never
Token ID1239
Feature activation-1.87e-02

Pos logit contributions
 nation+0.015
 knock+0.014
unts+0.013
 Find+0.013
 blast+0.013
Neg logit contributions
 brand-0.022
uit-0.021
 exactly-0.021
but-0.021
 perched-0.020
Token lie
Token ID6486
Feature activation-1.39e-02

Pos logit contributions
 nation+0.017
 knock+0.017
 Find+0.015
 guys+0.014
unts+0.014
Neg logit contributions
 brand-0.036
uit-0.035
but-0.035
 exactly-0.034
 unconscious-0.031
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
 guys+0.016
 nation+0.016
 ever+0.015
ot+0.015
ce+0.015
Neg logit contributions
 brand-0.020
uit-0.020
ieved-0.019
 exactly-0.018
Here-0.018
Token
Token ID198
Feature activation-3.96e-03

Pos logit contributions
 nation+0.001
 Find+0.001
ass+0.001
unts+0.001
ce+0.001
Neg logit contributions
 brand-0.002
 exactly-0.002
uit-0.002
but-0.002
 perched-0.002
Token
Token ID198
Feature activation-1.67e-04

Pos logit contributions
ce+0.006
unts+0.006
 nation+0.006
 knock+0.006
 blast+0.005
Neg logit contributions
 brand-0.005
 exactly-0.005
uit-0.005
but-0.005
 perched-0.004
TokenThe
Token ID464
Feature activation+7.22e-03

Pos logit contributions
 nation+0.000
 knock+0.000
ce+0.000
unts+0.000
 blast+0.000
Neg logit contributions
 brand-0.000
 exactly-0.000
uit-0.000
but-0.000
 perched-0.000
Token dad
Token ID9955
Feature activation-2.04e-02

Pos logit contributions
 brand+0.013
 exactly+0.013
uit+0.012
but+0.012
 unconscious+0.011
Neg logit contributions
 nation-0.007
ass-0.007
ce-0.007
 Find-0.006
unts-0.006
Token tried
Token ID3088
Feature activation+1.39e-03

Pos logit contributions
 nation+0.027
 knock+0.025
unts+0.025
ce+0.025
ass+0.025
Neg logit contributions
 exactly-0.028
 brand-0.028
uit-0.027
but-0.027
 perched-0.026
Token to
Token ID284
Feature activation+1.32e-02

Pos logit contributions
 brand+0.002
ieved+0.002
but+0.002
 exactly+0.002
uit+0.002
Neg logit contributions
 knock-0.001
 blast-0.001
 turns-0.001
 wake-0.001
 e-0.001
Token fix
Token ID4259
Feature activation+2.83e-02

Pos logit contributions
 brand+0.025
 exactly+0.025
but+0.024
 perched+0.023
uit+0.023
Neg logit contributions
 knock-0.013
 blast-0.012
 nation-0.011
 guys-0.010
 wake-0.010
Token the
Token ID262
Feature activation+1.76e-02

Pos logit contributions
 brand+0.044
but+0.044
 exactly+0.043
uit+0.042
 perched+0.041
Neg logit contributions
 nation-0.031
 knock-0.030
ce-0.030
 guys-0.029
 blast-0.029
Token truck
Token ID7779
Feature activation+2.04e-02

Pos logit contributions
 exactly+0.027
but+0.027
 brand+0.026
uit+0.026
 perched+0.026
Neg logit contributions
 nation-0.022
 knock-0.021
 guys-0.020
 blast-0.020
unts-0.019
Token but
Token ID475
Feature activation+1.81e-02

Pos logit contributions
 brand+0.032
 perched+0.032
 thanking+0.031
 exactly+0.031
uit+0.030
Neg logit contributions
 knock-0.023
 nation-0.021
air-0.020
 guys-0.019
ot-0.019
Token it
Token ID340
Feature activation-1.66e-03

Pos logit contributions
 brand+0.033
uit+0.033
but+0.032
 exactly+0.031
 perched+0.030
Neg logit contributions
 knock-0.015
 nation-0.014
 ever-0.014
 blast-0.014
ot-0.014
Token boy
Token ID2933
Feature activation-2.16e-02

Pos logit contributions
 nation+0.009
 knock+0.008
 guys+0.008
 blast+0.008
ye+0.007
Neg logit contributions
 exactly-0.017
but-0.016
uit-0.016
 brand-0.016
 perched-0.016
Token to
Token ID284
Feature activation-1.87e-03

Pos logit contributions
 nation+0.025
 ever+0.024
 knock+0.023
ce+0.022
air+0.022
Neg logit contributions
 brand-0.036
uit-0.034
 exactly-0.033
but-0.032
iddled-0.032
Token have
Token ID423
Feature activation+1.02e-02

Pos logit contributions
 knock+0.001
 nation+0.001
 blast+0.001
 wake+0.001
 guys+0.001
Neg logit contributions
 brand-0.004
but-0.004
 exactly-0.004
 unconscious-0.004
 perched-0.004
Token a
Token ID257
Feature activation+1.11e-02

Pos logit contributions
but+0.019
oding+0.017
ishly+0.017
uit+0.017
 brand+0.016
Neg logit contributions
 ever-0.011
 turns-0.011
 blast-0.011
 breaks-0.010
 happens-0.010
Token pistol
Token ID16790
Feature activation+9.38e-03

Pos logit contributions
but+0.018
 exactly+0.018
uit+0.017
 perched+0.017
sometimes+0.016
Neg logit contributions
 nation-0.013
 knock-0.013
 blast-0.012
ye-0.012
 guys-0.011
Token.
Token ID13
Feature activation+9.79e-03

Pos logit contributions
 brand+0.016
iddled+0.016
ieved+0.016
 exactly+0.015
Here+0.015
Neg logit contributions
 ever-0.010
 guys-0.010
 nation-0.010
 team-0.009
 person-0.009
Token
Token ID220
Feature activation+7.50e-03

Pos logit contributions
 brand+0.015
 exactly+0.015
uit+0.014
 perched+0.014
but+0.013
Neg logit contributions
 Find-0.012
 nation-0.012
unts-0.011
 Danger-0.011
ass-0.011
Token
Token ID198
Feature activation+5.33e-03

Pos logit contributions
 brand+0.013
 exactly+0.012
 perched+0.012
ishly+0.011
uit+0.011
Neg logit contributions
oke-0.008
unts-0.008
ia-0.007
 Find-0.007
ass-0.007
Token
Token ID198
Feature activation+8.93e-03

Pos logit contributions
 brand+0.008
 exactly+0.008
uit+0.007
 perched+0.007
 struggling+0.007
Neg logit contributions
unts-0.007
ce-0.007
 knock-0.007
 Find-0.006
ye-0.006
TokenJohn
Token ID7554
Feature activation+1.03e-04

Pos logit contributions
 brand+0.013
 exactly+0.013
uit+0.012
 perched+0.012
but+0.011
Neg logit contributions
 nation-0.012
unts-0.011
 knock-0.011
ce-0.011
 Find-0.011
Token thought
Token ID1807
Feature activation+1.63e-02

Pos logit contributions
 brand+0.000
 exactly+0.000
uit+0.000
but+0.000
 perched+0.000
Neg logit contributions
ce-0.000
 nation-0.000
 knock-0.000
ith-0.000
air-0.000
Token�
Token ID129
Feature activation-3.80e-02

Pos logit contributions
 nation+0.043
 knock+0.041
ass+0.040
 Danger+0.039
unts+0.038
Neg logit contributions
 brand-0.037
 exactly-0.035
but-0.035
 perched-0.034
uit-0.034
Token�
Token ID241
Feature activation+1.24e-02

Pos logit contributions
 nation+0.055
 spe+0.050
 blast+0.050
ye+0.049
 guys+0.049
Neg logit contributions
 brand-0.050
 exactly-0.050
 perched-0.048
but-0.047
uit-0.045
TokenDo
Token ID5211
Feature activation-2.01e-02

Pos logit contributions
 exactly+0.020
 brand+0.020
uit+0.019
but+0.017
ishly+0.017
Neg logit contributions
 nation-0.014
unts-0.014
ce-0.014
 spe-0.013
 blast-0.013
Token you
Token ID345
Feature activation+1.85e-02

Pos logit contributions
 nation+0.032
unts+0.030
 spe+0.030
 knock+0.029
ce+0.029
Neg logit contributions
 exactly-0.025
 brand-0.023
but-0.022
uit-0.022
 perched-0.021
Token want
Token ID765
Feature activation+2.67e-03

Pos logit contributions
 brand+0.035
 exactly+0.035
but+0.034
uit+0.033
 perched+0.032
Neg logit contributions
 nation-0.016
 knock-0.015
ass-0.014
unts-0.014
 blast-0.014
Token to
Token ID284
Feature activation+5.21e-03

Pos logit contributions
 exactly+0.005
 unconscious+0.005
 brand+0.005
 intact+0.005
but+0.005
Neg logit contributions
 Find-0.003
 guys-0.003
 happens-0.002
ye-0.002
 ever-0.002
Token try
Token ID1949
Feature activation+7.32e-04

Pos logit contributions
 brand+0.009
 exactly+0.009
uit+0.008
but+0.008
 perched+0.008
Neg logit contributions
 nation-0.005
ce-0.005
 knock-0.005
unts-0.005
ass-0.005
Token doing
Token ID1804
Feature activation+1.91e-02

Pos logit contributions
 unconscious+0.001
but+0.001
ishly+0.001
 exactly+0.001
 brand+0.001
Neg logit contributions
ce-0.001
 guys-0.001
 Find-0.001
 blast-0.001
 dive-0.001
Token some
Token ID617
Feature activation-2.71e-03

Pos logit contributions
but+0.025
 unconscious+0.024
uit+0.023
ishly+0.022
sometimes+0.021
Neg logit contributions
 guys-0.028
 blast-0.028
 nation-0.026
 Find-0.026
 knock-0.026
Token math
Token ID10688
Feature activation+2.89e-02

Pos logit contributions
 nation+0.003
 knock+0.003
 guys+0.003
 blast+0.003
ce+0.003
Neg logit contributions
but-0.004
 brand-0.004
uit-0.004
 exactly-0.004
 unconscious-0.004
Token?
Token ID30
Feature activation+1.55e-02

Pos logit contributions
 unconscious+0.041
 brand+0.040
 exactly+0.039
uit+0.039
 thanking+0.038
Neg logit contributions
 nation-0.041
 guys-0.040
 team-0.037
ce-0.037
 blast-0.036
Token that
Token ID326
Feature activation-1.10e-02

Pos logit contributions
 knock+0.000
 e+0.000
 spe+0.000
 nation+0.000
 blast+0.000
Neg logit contributions
 brand-0.000
uit-0.000
iddled-0.000
 exactly-0.000
ieved-0.000
Token he
Token ID339
Feature activation+7.65e-03

Pos logit contributions
 nation+0.019
 knock+0.018
ce+0.018
unts+0.018
ass+0.018
Neg logit contributions
 brand-0.011
 exactly-0.011
uit-0.010
but-0.010
 perched-0.010
Token started
Token ID2067
Feature activation-5.80e-03

Pos logit contributions
 brand+0.013
 exactly+0.013
uit+0.013
but+0.012
 unconscious+0.012
Neg logit contributions
 nation-0.008
 knock-0.008
ce-0.008
 blast-0.007
unts-0.007
Token spinning
Token ID19493
Feature activation+3.59e-02

Pos logit contributions
 nation+0.007
 knock+0.007
 blast+0.007
unts+0.007
ce+0.007
Neg logit contributions
uit-0.009
 brand-0.009
 exactly-0.009
but-0.008
 perched-0.007
Token and
Token ID290
Feature activation-1.62e-03

Pos logit contributions
uit+0.063
 exactly+0.062
 thanking+0.059
nam+0.058
 unconscious+0.058
Neg logit contributions
 guys-0.034
ot-0.033
 wind-0.033
ce-0.032
 turns-0.032
Token tw
Token ID665
Feature activation-4.17e-03

Pos logit contributions
 knock+0.002
 nation+0.002
 blast+0.002
ce+0.002
unts+0.002
Neg logit contributions
 exactly-0.003
uit-0.002
 brand-0.002
but-0.002
 unconscious-0.002
Tokenirling
Token ID24297
Feature activation+1.29e-02

Pos logit contributions
 nation+0.006
ce+0.005
 knock+0.005
unts+0.005
 blast+0.005
Neg logit contributions
 brand-0.006
 exactly-0.006
but-0.006
uit-0.006
 perched-0.005
Token around
Token ID1088
Feature activation+4.14e-03

Pos logit contributions
uit+0.021
 brand+0.021
 exactly+0.021
 struggling+0.020
 unconscious+0.020
Neg logit contributions
 guys-0.014
 wind-0.013
ot-0.013
 nation-0.013
 ever-0.013
Token the
Token ID262
Feature activation-1.36e-02

Pos logit contributions
 brand+0.006
 exactly+0.006
uit+0.006
 perched+0.005
ieved+0.005
Neg logit contributions
 nation-0.006
ce-0.005
 knock-0.005
 blast-0.005
 guys-0.005
Token yard
Token ID12699
Feature activation-1.21e-02

Pos logit contributions
 nation+0.017
 knock+0.015
ce+0.015
unts+0.015
 blast+0.015
Neg logit contributions
 exactly-0.022
 brand-0.021
uit-0.021
but-0.020
 perched-0.020
Token.
Token ID13
Feature activation-3.86e-03

Pos logit contributions
 nation+0.017
 knock+0.015
ce+0.015
 team+0.015
 ever+0.015
Neg logit contributions
 brand-0.017
 exactly-0.016
uit-0.016
 unconscious-0.015
?,-0.015
Token Sam
Token ID3409
Feature activation+1.02e-02

Pos logit contributions
but+0.011
ieved+0.010
 exactly+0.010
 brand+0.010
uit+0.010
Neg logit contributions
 knock-0.007
ia-0.006
 nation-0.006
ot-0.006
 e-0.006
Token was
Token ID373
Feature activation+1.03e-03

Pos logit contributions
but+0.021
 brand+0.020
 exactly+0.020
uit+0.019
sometimes+0.018
Neg logit contributions
 knock-0.009
 wake-0.008
 nation-0.008
 e-0.007
ce-0.007
Token sad
Token ID6507
Feature activation+2.49e-02

Pos logit contributions
but+0.002
uit+0.002
 brand+0.002
 exactly+0.002
sometimes+0.001
Neg logit contributions
 knock-0.001
 nation-0.001
unts-0.001
 e-0.001
 wind-0.001
Token.
Token ID13
Feature activation-1.56e-03

Pos logit contributions
 brand+0.040
uit+0.039
 perched+0.039
ieved+0.038
 exactly+0.038
Neg logit contributions
 knock-0.030
 nation-0.027
unts-0.026
ot-0.026
ia-0.026
Token She
Token ID1375
Feature activation-1.60e-04

Pos logit contributions
 nation+0.002
 knock+0.002
unts+0.002
 Find+0.002
ce+0.002
Neg logit contributions
 exactly-0.002
 brand-0.002
uit-0.002
but-0.002
 perched-0.002
Token wanted
Token ID2227
Feature activation-1.33e-02

Pos logit contributions
 nation+0.000
 knock+0.000
unts+0.000
ce+0.000
 Find+0.000
Neg logit contributions
 brand-0.000
 exactly-0.000
uit-0.000
but-0.000
 perched-0.000
Token to
Token ID284
Feature activation+2.43e-03

Pos logit contributions
 nation+0.017
 knock+0.017
 guys+0.016
 Find+0.016
 blast+0.016
Neg logit contributions
 brand-0.019
 exactly-0.019
 perched-0.018
but-0.018
uit-0.018
Token help
Token ID1037
Feature activation-5.08e-03

Pos logit contributions
 exactly+0.004
 brand+0.004
but+0.004
uit+0.004
 perched+0.004
Neg logit contributions
 nation-0.003
 knock-0.003
 blast-0.002
ce-0.002
unts-0.002
Token him
Token ID683
Feature activation+1.53e-02

Pos logit contributions
 nation+0.008
 knock+0.008
ce+0.008
unts+0.008
ass+0.008
Neg logit contributions
 brand-0.006
 exactly-0.006
but-0.005
uit-0.005
 perched-0.005
Token.
Token ID13
Feature activation+8.90e-04

Pos logit contributions
 brand+0.028
Here+0.027
 unconscious+0.027
uit+0.026
 exactly+0.026
Neg logit contributions
 knock-0.017
 go-0.015
 wake-0.014
 dive-0.014
 team-0.014
Token She
Token ID1375
Feature activation+1.49e-03

Pos logit contributions
 brand+0.001
 exactly+0.001
uit+0.001
 perched+0.001
 unconscious+0.001
Neg logit contributions
 nation-0.001
 Find-0.001
unts-0.001
 knock-0.001
ot-0.001
Token friendly
Token ID8030
Feature activation-6.05e-03

Pos logit contributions
 exactly+0.031
but+0.030
 brand+0.030
uit+0.030
 perched+0.029
Neg logit contributions
 nation-0.012
 knock-0.012
ce-0.011
unts-0.011
 spe-0.010
Token squirrel
Token ID33039
Feature activation+4.48e-03

Pos logit contributions
 nation+0.006
 knock+0.005
unts+0.005
ce+0.005
 blast+0.005
Neg logit contributions
 brand-0.011
 exactly-0.011
uit-0.010
but-0.010
 perched-0.010
Token.
Token ID13
Feature activation-3.13e-03

Pos logit contributions
 brand+0.006
 exactly+0.005
uit+0.005
but+0.005
 unconscious+0.005
Neg logit contributions
ce-0.007
 nation-0.006
 knock-0.006
ass-0.006
ot-0.006
Token "
Token ID366
Feature activation+8.75e-03

Pos logit contributions
 Find+0.003
oke+0.003
ce+0.003
unts+0.003
 knock+0.003
Neg logit contributions
 brand-0.005
 exactly-0.005
uit-0.005
 intact-0.005
ieved-0.005
TokenI
Token ID40
Feature activation+1.22e-02

Pos logit contributions
 brand+0.014
 exactly+0.014
uit+0.013
 struggling+0.012
 perched+0.012
Neg logit contributions
 nation-0.010
ce-0.010
air-0.010
 Find-0.010
ass-0.010
Token have
Token ID423
Feature activation-1.18e-02

Pos logit contributions
 brand+0.023
but+0.022
uit+0.022
 exactly+0.022
 unconscious+0.021
Neg logit contributions
 nation-0.011
 knock-0.011
 blast-0.010
 spe-0.010
ass-0.010
Token a
Token ID257
Feature activation-5.08e-03

Pos logit contributions
 nation+0.016
 Find+0.016
 eight+0.015
 blast+0.015
 spe+0.015
Neg logit contributions
but-0.017
uit-0.017
 brand-0.016
 exactly-0.016
 unconscious-0.015
Token plan
Token ID1410
Feature activation-3.18e-03

Pos logit contributions
 nation+0.005
 knock+0.005
 blast+0.005
 spe+0.005
ye+0.005
Neg logit contributions
but-0.010
 exactly-0.009
sometimes-0.009
uit-0.009
ieved-0.008
Token.
Token ID13
Feature activation+1.17e-03

Pos logit contributions
 ever+0.004
 nation+0.003
 Find+0.003
 person+0.003
ce+0.003
Neg logit contributions
 brand-0.006
 ang-0.006
 unconscious-0.005
 exactly-0.005
ieved-0.005
Token Follow
Token ID7281
Feature activation-6.20e-05

Pos logit contributions
 exactly+0.002
 brand+0.002
but+0.002
uit+0.002
 perched+0.002
Neg logit contributions
 nation-0.001
 knock-0.001
ce-0.001
unts-0.001
 blast-0.001
Token me
Token ID502
Feature activation+8.98e-03

Pos logit contributions
 nation+0.000
ce+0.000
air+0.000
ot+0.000
 guys+0.000
Neg logit contributions
 brand-0.000
 exactly-0.000
but-0.000
 unconscious-0.000
uit-0.000
Token his
Token ID465
Feature activation+9.93e-03

Pos logit contributions
 nation+0.019
 team+0.017
 ever+0.017
 wake+0.017
 blast+0.017
Neg logit contributions
uit-0.022
but-0.021
ieved-0.021
 brand-0.021
 unconscious-0.021
Token mission
Token ID4365
Feature activation-5.59e-03

Pos logit contributions
 brand+0.015
 exactly+0.015
but+0.014
uit+0.014
 perched+0.014
Neg logit contributions
 nation-0.012
 knock-0.012
unts-0.011
ce-0.011
 blast-0.011
Token to
Token ID284
Feature activation+9.40e-03

Pos logit contributions
 ever+0.005
 nation+0.005
 person+0.005
ot+0.004
 grown+0.004
Neg logit contributions
 perched-0.010
ishly-0.010
 unconscious-0.010
nam-0.010
Here-0.010
Token help
Token ID1037
Feature activation-1.08e-02

Pos logit contributions
 brand+0.017
but+0.016
 exactly+0.016
uit+0.016
 unconscious+0.016
Neg logit contributions
 nation-0.009
 knock-0.009
 blast-0.009
ass-0.008
 Find-0.008
Token any
Token ID597
Feature activation+3.94e-03

Pos logit contributions
 nation+0.018
ce+0.017
 spe+0.017
 knock+0.017
 blast+0.017
Neg logit contributions
 exactly-0.012
 brand-0.011
uit-0.011
but-0.010
 perched-0.010
Token animal
Token ID5044
Feature activation-2.64e-02

Pos logit contributions
 exactly+0.006
 brand+0.006
uit+0.005
 unconscious+0.005
 struggling+0.005
Neg logit contributions
 nation-0.005
ce-0.005
ass-0.005
 knock-0.005
unts-0.005
Token who
Token ID508
Feature activation-4.88e-03

Pos logit contributions
 nation+0.031
 knock+0.030
 person+0.028
 guys+0.028
 team+0.027
Neg logit contributions
 brand-0.043
uit-0.042
but-0.042
ieved-0.040
 exactly-0.040
Token was
Token ID373
Feature activation-1.64e-02

Pos logit contributions
 nation+0.006
 knock+0.005
ass+0.005
 Find+0.005
 guys+0.005
Neg logit contributions
 brand-0.008
but-0.008
uit-0.008
 exactly-0.008
 unconscious-0.007
Token lost
Token ID2626
Feature activation-4.26e-03

Pos logit contributions
 nation+0.019
 ever+0.018
 spe+0.018
 knock+0.018
unts+0.017
Neg logit contributions
uit-0.028
but-0.027
 brand-0.024
iddled-0.024
ishly-0.023
Token.
Token ID13
Feature activation-3.76e-03

Pos logit contributions
 ever+0.005
 nation+0.005
 guys+0.005
 person+0.004
 spe+0.004
Neg logit contributions
ieved-0.007
 brand-0.007
but-0.007
uit-0.007
iddled-0.007
Token He
Token ID679
Feature activation-1.08e-02

Pos logit contributions
 nation+0.006
 knock+0.006
 spe+0.006
ce+0.006
 guys+0.005
Neg logit contributions
 exactly-0.004
 brand-0.004
but-0.004
uit-0.004
 unconscious-0.004
Token through
Token ID832
Feature activation-1.31e-02

Pos logit contributions
 nation+0.016
 knock+0.014
ot+0.014
 guys+0.014
ce+0.014
Neg logit contributions
 brand-0.018
 exactly-0.017
uit-0.016
but-0.016
 perched-0.015
Token the
Token ID262
Feature activation-6.25e-03

Pos logit contributions
 nation+0.020
unts+0.020
 knock+0.019
ye+0.019
ass+0.019
Neg logit contributions
 brand-0.015
 exactly-0.015
but-0.014
uit-0.014
 perched-0.014
Token park
Token ID3952
Feature activation-2.17e-03

Pos logit contributions
 spe+0.012
oke+0.011
ass+0.011
 pm+0.011
unts+0.011
Neg logit contributions
 brand-0.005
but-0.005
 perched-0.004
 exactly-0.004
 upon-0.004
Token,
Token ID11
Feature activation-8.62e-03

Pos logit contributions
 nation+0.003
unts+0.003
ce+0.003
 blast+0.003
ye+0.003
Neg logit contributions
 exactly-0.003
 brand-0.003
but-0.003
uit-0.002
 perched-0.002
Token L
Token ID406
Feature activation-2.36e-02

Pos logit contributions
 nation+0.011
 knock+0.011
 blast+0.011
 guys+0.010
ce+0.010
Neg logit contributions
 brand-0.012
 exactly-0.012
uit-0.011
but-0.011
 perched-0.011
Tokenola
Token ID5708
Feature activation-1.44e-02

Pos logit contributions
 nation+0.023
 spe+0.023
 knock+0.022
 blast+0.022
unts+0.021
Neg logit contributions
 exactly-0.041
but-0.040
 brand-0.040
uit-0.039
 upon-0.037
Token saw
Token ID2497
Feature activation+5.77e-03

Pos logit contributions
 nation+0.020
 knock+0.019
ce+0.018
unts+0.018
ass+0.018
Neg logit contributions
 brand-0.019
 exactly-0.019
but-0.018
uit-0.018
 perched-0.018
Token something
Token ID1223
Feature activation-1.87e-02

Pos logit contributions
but+0.009
uit+0.009
ieved+0.009
 exactly+0.009
 brand+0.008
Neg logit contributions
 nation-0.006
 person-0.006
 wind-0.006
ia-0.006
ot-0.006
Token special
Token ID2041
Feature activation-1.03e-02

Pos logit contributions
 knock+0.021
 nation+0.019
 spe+0.018
 e+0.017
unts+0.016
Neg logit contributions
uit-0.032
but-0.031
 exactly-0.030
ieved-0.029
 unconscious-0.029
Token -
Token ID532
Feature activation+1.38e-02

Pos logit contributions
 ever+0.011
 nation+0.010
 happens+0.010
 comes+0.009
 person+0.009
Neg logit contributions
 brand-0.018
 ang-0.018
ieved-0.017
 unconscious-0.017
uit-0.016
Token a
Token ID257
Feature activation+1.03e-03

Pos logit contributions
uit+0.016
but+0.016
 exactly+0.016
 brand+0.016
ieved+0.015
Neg logit contributions
 nation-0.022
 spe-0.020
 knock-0.020
 blast-0.019
 guys-0.019
Token small
Token ID1402
Feature activation-2.51e-03

Pos logit contributions
 nation+0.002
 knock+0.002
 blast+0.002
 spe+0.002
unts+0.002
Neg logit contributions
 exactly-0.003
but-0.003
uit-0.003
sometimes-0.003
 brand-0.003
Token ladder
Token ID18002
Feature activation+3.51e-02

Pos logit contributions
 nation+0.003
 knock+0.003
 spe+0.003
 blast+0.003
unts+0.003
Neg logit contributions
 exactly-0.004
 brand-0.004
but-0.004
uit-0.004
 perched-0.004
Token inside
Token ID2641
Feature activation+1.16e-02

Pos logit contributions
 brand+0.049
 exactly+0.049
uit+0.046
 thanking+0.044
but+0.044
Neg logit contributions
 spe-0.044
 knock-0.044
ot-0.042
air-0.042
 nation-0.042
Token so
Token ID523
Feature activation+1.10e-02

Pos logit contributions
ishly+0.024
 brand+0.024
ieved+0.023
uit+0.023
aj+0.022
Neg logit contributions
 ever-0.008
 of-0.008
 person-0.007
 nation-0.006
 month-0.006
Token that
Token ID326
Feature activation-2.22e-02

Pos logit contributions
 brand+0.016
 exactly+0.016
uit+0.015
but+0.015
 perched+0.015
Neg logit contributions
 nation-0.014
 knock-0.013
ce-0.013
unts-0.013
 spe-0.013
Token birds
Token ID10087
Feature activation-1.25e-02

Pos logit contributions
 nation+0.030
 knock+0.029
ce+0.028
 guys+0.027
 blast+0.027
Neg logit contributions
 brand-0.031
 exactly-0.030
uit-0.030
but-0.029
 perched-0.027
Token could
Token ID714
Feature activation-1.41e-02

Pos logit contributions
 nation+0.011
 comes+0.011
 ever+0.010
 spe+0.010
 knock+0.010
Neg logit contributions
uit-0.025
 brand-0.025
 exactly-0.023
but-0.023
 unconscious-0.022
Token climb
Token ID12080
Feature activation+1.77e-02

Pos logit contributions
 knock+0.012
 spe+0.010
 blast+0.010
 wake+0.009
 nation+0.009
Neg logit contributions
 brand-0.029
 exactly-0.029
 unconscious-0.029
but-0.029
uit-0.029
Token in
Token ID287
Feature activation+8.51e-03

Pos logit contributions
 exactly+0.025
 brand+0.025
uit+0.024
but+0.024
 reassured+0.023
Neg logit contributions
 spe-0.023
 nation-0.022
ce-0.022
 knock-0.021
 guys-0.021
Token and
Token ID290
Feature activation-2.59e-03

Pos logit contributions
 brand+0.013
but+0.013
 exactly+0.012
ieved+0.012
uit+0.012
Neg logit contributions
 spe-0.011
 nation-0.010
 blast-0.010
 ever-0.009
 knock-0.009
Token out
Token ID503
Feature activation+1.24e-02

Pos logit contributions
 knock+0.003
 nation+0.002
 blast+0.002
 spe+0.002
 Find+0.002
Neg logit contributions
uit-0.005
 exactly-0.005
but-0.005
 brand-0.005
 unconscious-0.004
Token
Token ID198
Feature activation-1.67e-03

Pos logit contributions
unts+0.007
 knock+0.007
 nation+0.006
ce+0.006
 Find+0.006
Neg logit contributions
 exactly-0.006
 brand-0.005
uit-0.005
but-0.005
 perched-0.005
TokenThe
Token ID464
Feature activation+2.18e-03

Pos logit contributions
 nation+0.002
 knock+0.002
ce+0.002
unts+0.002
 blast+0.002
Neg logit contributions
 exactly-0.002
 brand-0.002
uit-0.002
but-0.002
 perched-0.002
Token moral
Token ID6573
Feature activation-2.14e-02

Pos logit contributions
 brand+0.003
 exactly+0.003
uit+0.003
but+0.003
 struggling+0.003
Neg logit contributions
ass-0.003
 nation-0.002
 spe-0.002
air-0.002
ce-0.002
Token of
Token ID286
Feature activation-1.99e-02

Pos logit contributions
 knock+0.029
 guys+0.028
 nation+0.027
 spe+0.027
 blast+0.026
Neg logit contributions
 exactly-0.029
 brand-0.027
but-0.026
 perched-0.025
sometimes-0.025
Token the
Token ID262
Feature activation-2.09e-02

Pos logit contributions
 nation+0.030
 knock+0.028
unts+0.027
ce+0.027
ass+0.027
Neg logit contributions
 brand-0.025
 exactly-0.025
but-0.023
uit-0.023
 perched-0.022
Token story
Token ID1621
Feature activation-6.80e-02

Pos logit contributions
air+0.035
 spe+0.034
ot+0.033
ass+0.033
venth+0.032
Neg logit contributions
 brand-0.022
 exactly-0.021
 thanking-0.019
 shining-0.019
 struggling-0.018
Token is
Token ID318
Feature activation-2.37e-02

Pos logit contributions
 nation+0.074
 spe+0.073
ce+0.072
 blast+0.071
 knock+0.071
Neg logit contributions
 exactly-0.111
 brand-0.105
but-0.103
uit-0.099
 perched-0.096
Token that
Token ID326
Feature activation-3.77e-02

Pos logit contributions
 nation+0.041
 knock+0.040
ce+0.040
 spe+0.039
 guys+0.039
Neg logit contributions
 exactly-0.025
 brand-0.023
but-0.022
uit-0.020
 perched-0.020
Token sometimes
Token ID3360
Feature activation-2.91e-02

Pos logit contributions
 nation+0.052
 knock+0.048
ce+0.047
unts+0.047
 blast+0.047
Neg logit contributions
 brand-0.052
 exactly-0.051
uit-0.049
but-0.049
 perched-0.047
Token we
Token ID356
Feature activation-1.01e-02

Pos logit contributions
ce+0.037
ass+0.036
 nation+0.036
unts+0.036
ia+0.035
Neg logit contributions
 exactly-0.042
 brand-0.041
 perched-0.038
but-0.037
 struggling-0.037
Token need
Token ID761
Feature activation-3.95e-03

Pos logit contributions
 nation+0.010
 knock+0.009
ce+0.009
unts+0.009
ass+0.009
Neg logit contributions
 brand-0.018
 exactly-0.018
uit-0.017
but-0.017
 perched-0.016
Token want
Token ID765
Feature activation-1.49e-02

Pos logit contributions
 brand+0.004
 exactly+0.004
uit+0.004
but+0.004
 perched+0.004
Neg logit contributions
 nation-0.002
ce-0.002
 knock-0.002
unts-0.002
ass-0.002
Token to
Token ID284
Feature activation-4.33e-03

Pos logit contributions
 nation+0.020
 guys+0.019
 spe+0.018
 knock+0.018
 blast+0.018
Neg logit contributions
uit-0.021
 brand-0.021
but-0.021
 exactly-0.021
 perched-0.020
Token see
Token ID766
Feature activation-5.86e-03

Pos logit contributions
 nation+0.004
 knock+0.003
 blast+0.003
unts+0.003
 guys+0.003
Neg logit contributions
 brand-0.009
 exactly-0.008
but-0.008
uit-0.008
 unconscious-0.008
Token the
Token ID262
Feature activation-3.29e-03

Pos logit contributions
 nation+0.007
 guys+0.006
 spe+0.006
 knock+0.006
 e+0.005
Neg logit contributions
but-0.010
uit-0.010
 unconscious-0.009
 brand-0.009
ieved-0.009
Token b
Token ID275
Feature activation-4.76e-02

Pos logit contributions
 nation+0.003
 guys+0.003
 knock+0.003
 spe+0.003
 team+0.003
Neg logit contributions
but-0.006
 exactly-0.006
uit-0.006
 brand-0.006
 perched-0.006
Tokenunn
Token ID20935
Feature activation-9.64e-02

Pos logit contributions
 blast+0.057
 knock+0.054
 nation+0.053
 Find+0.049
 happens+0.047
Neg logit contributions
but-0.075
 exactly-0.074
uit-0.072
 brand-0.071
 perched-0.069
Tokenies
Token ID444
Feature activation-4.27e-03

Pos logit contributions
 nation+0.112
 knock+0.103
ce+0.103
unts+0.102
ass+0.101
Neg logit contributions
 brand-0.152
 exactly-0.151
uit-0.144
but-0.143
 perched-0.138
Token.
Token ID13
Feature activation-2.18e-03

Pos logit contributions
 nation+0.006
 knock+0.005
 wake+0.005
 ever+0.005
 dive+0.004
Neg logit contributions
uit-0.007
 brand-0.007
but-0.006
 unconscious-0.006
 exactly-0.006
Token They
Token ID1119
Feature activation-1.33e-02

Pos logit contributions
 nation+0.003
 knock+0.002
unts+0.002
 Find+0.002
ass+0.002
Neg logit contributions
 brand-0.004
 exactly-0.003
uit-0.003
but-0.003
 perched-0.003
Token are
Token ID389
Feature activation-1.20e-02

Pos logit contributions
 nation+0.013
 knock+0.013
 blast+0.011
 guys+0.011
unts+0.011
Neg logit contributions
 brand-0.026
but-0.025
 exactly-0.024
uit-0.024
 unconscious-0.024
Token the
Token ID262
Feature activation-1.09e-02

Pos logit contributions
 nation+0.014
 knock+0.013
unts+0.013
 spe+0.012
 blast+0.012
Neg logit contributions
uit-0.020
but-0.020
 brand-0.019
sometimes-0.018
 exactly-0.018
Token measure
Token ID3953
Feature activation-1.19e-03

Pos logit contributions
 nation+0.004
 knock+0.004
 blast+0.004
 wind+0.004
 eight+0.004
Neg logit contributions
 perched-0.008
but-0.008
uit-0.008
 exactly-0.008
ishly-0.008
Token again
Token ID757
Feature activation-2.82e-02

Pos logit contributions
 nation+0.002
 knock+0.001
ot+0.001
ia+0.001
unts+0.001
Neg logit contributions
 brand-0.002
 unconscious-0.002
 perched-0.002
uit-0.002
 exactly-0.002
Token to
Token ID284
Feature activation-2.20e-02

Pos logit contributions
 nation+0.039
ce+0.036
ot+0.035
 knock+0.035
unts+0.034
Neg logit contributions
 brand-0.041
 exactly-0.039
uit-0.038
but-0.037
 unconscious-0.037
Token see
Token ID766
Feature activation-1.62e-02

Pos logit contributions
 nation+0.024
ce+0.023
unts+0.022
ass+0.022
ye+0.022
Neg logit contributions
 brand-0.035
 exactly-0.035
uit-0.033
but-0.033
 perched-0.032
Token how
Token ID703
Feature activation-2.46e-02

Pos logit contributions
 nation+0.020
 knock+0.019
ot+0.018
unts+0.017
 comes+0.017
Neg logit contributions
uit-0.026
but-0.025
 unconscious-0.024
ishly-0.024
 brand-0.023
Token tall
Token ID7331
Feature activation-7.46e-02

Pos logit contributions
ass+0.033
 nation+0.033
ce+0.033
unn+0.032
 guys+0.031
Neg logit contributions
 brand-0.034
 exactly-0.033
but-0.031
 perched-0.028
 upon-0.028
Token it
Token ID340
Feature activation-3.79e-02

Pos logit contributions
 nation+0.098
 knock+0.091
ce+0.089
 blast+0.088
unts+0.088
Neg logit contributions
 brand-0.108
 exactly-0.107
uit-0.102
but-0.101
 perched-0.097
Token is
Token ID318
Feature activation-2.02e-02

Pos logit contributions
 nation+0.031
 knock+0.029
ce+0.028
 blast+0.027
unts+0.027
Neg logit contributions
 brand-0.072
 exactly-0.072
uit-0.070
but-0.070
 perched-0.067
Token.
Token ID13
Feature activation-1.09e-02

Pos logit contributions
 knock+0.016
 nation+0.016
 goes+0.015
 ever+0.015
 go+0.015
Neg logit contributions
uit-0.039
ishly-0.039
Here-0.037
iddled-0.036
 unconscious-0.036
Token They
Token ID1119
Feature activation-1.92e-02

Pos logit contributions
 nation+0.012
 knock+0.011
unts+0.011
ce+0.011
 blast+0.011
Neg logit contributions
 exactly-0.017
 brand-0.017
uit-0.017
but-0.017
 perched-0.016
Token hope
Token ID2911
Feature activation-1.94e-02

Pos logit contributions
 nation+0.018
 knock+0.018
 blast+0.016
unts+0.015
 guys+0.015
Neg logit contributions
 brand-0.036
 exactly-0.036
but-0.035
uit-0.035
 unconscious-0.034
Token baby
Token ID5156
Feature activation+1.62e-02

Pos logit contributions
 nation+0.013
 knock+0.012
 blast+0.011
 spe+0.011
 storm+0.010
Neg logit contributions
but-0.029
uit-0.027
 exactly-0.027
sometimes-0.026
 unconscious-0.025
Token frog
Token ID21264
Feature activation+3.38e-02

Pos logit contributions
 brand+0.030
but+0.029
 exactly+0.029
uit+0.028
ieved+0.027
Neg logit contributions
 nation-0.017
 spe-0.016
 person-0.015
 knock-0.015
 comes-0.015
Token!
Token ID0
Feature activation+1.55e-02

Pos logit contributions
 brand+0.063
nam+0.058
but+0.057
 exactly+0.056
 ang+0.054
Neg logit contributions
 nation-0.037
 ever-0.036
 comes-0.036
 team-0.033
ce-0.032
Token
Token ID198
Feature activation-6.18e-03

Pos logit contributions
 brand+0.029
 exactly+0.029
uit+0.027
but+0.026
 struggling+0.025
Neg logit contributions
 Find-0.015
 nation-0.014
 knock-0.013
 Danger-0.013
ce-0.013
Token
Token ID198
Feature activation-3.98e-03

Pos logit contributions
unts+0.008
ce+0.008
ia+0.008
ot+0.007
ass+0.007
Neg logit contributions
 brand-0.010
 exactly-0.009
uit-0.009
ishly-0.009
but-0.009
TokenT
Token ID51
Feature activation-6.92e-02

Pos logit contributions
 nation+0.005
ce+0.004
unts+0.004
 knock+0.004
ot+0.004
Neg logit contributions
 brand-0.007
 exactly-0.006
uit-0.006
 perched-0.006
but-0.006
Tokenad
Token ID324
Feature activation-2.62e-02

Pos logit contributions
 nation+0.068
ce+0.062
 knock+0.062
unts+0.061
ass+0.061
Neg logit contributions
 brand-0.121
 exactly-0.120
but-0.115
uit-0.114
 perched-0.112
Token smiled
Token ID13541
Feature activation+3.82e-03

Pos logit contributions
ce+0.029
ia+0.028
 nation+0.028
ith+0.026
oke+0.026
Neg logit contributions
 brand-0.047
uit-0.046
but-0.046
 exactly-0.044
nam-0.041
Token and
Token ID290
Feature activation+3.49e-03

Pos logit contributions
uit+0.007
 unconscious+0.006
Here+0.006
sometimes+0.006
ieved+0.006
Neg logit contributions
ot-0.004
 bye-0.003
 nation-0.003
 wind-0.003
 team-0.003
Token carefully
Token ID7773
Feature activation-1.34e-02

Pos logit contributions
uit+0.005
 brand+0.005
but+0.005
 exactly+0.005
sometimes+0.005
Neg logit contributions
 spe-0.004
ot-0.004
 knock-0.004
 nation-0.004
ce-0.004
Token picked
Token ID6497
Feature activation+5.94e-03

Pos logit contributions
ce+0.015
 spe+0.014
 knock+0.014
ot+0.014
 e+0.013
Neg logit contributions
 exactly-0.022
 brand-0.022
but-0.021
uit-0.021
 unconscious-0.020
Token.
Token ID13
Feature activation-1.77e-02

Pos logit contributions
 guys+0.039
 nation+0.038
unts+0.036
 team+0.035
 knock+0.034
Neg logit contributions
 brand-0.041
 perched-0.037
 exactly-0.036
uit-0.036
but-0.036
Token They
Token ID1119
Feature activation-2.39e-02

Pos logit contributions
 nation+0.023
 knock+0.023
ce+0.023
 guys+0.022
ye+0.022
Neg logit contributions
 brand-0.024
 exactly-0.023
but-0.023
uit-0.022
 perched-0.021
Token learned
Token ID4499
Feature activation-3.46e-02

Pos logit contributions
 nation+0.026
 knock+0.025
unts+0.024
 blast+0.023
 guys+0.022
Neg logit contributions
 brand-0.040
but-0.040
uit-0.039
 exactly-0.039
 perched-0.037
Token that
Token ID326
Feature activation-1.23e-02

Pos logit contributions
 knock+0.028
 nation+0.027
unts+0.027
 happens+0.027
ot+0.026
Neg logit contributions
uit-0.068
ieved-0.066
 perched-0.061
iddled-0.061
 atop-0.061
Token being
Token ID852
Feature activation-1.06e-02

Pos logit contributions
 nation+0.016
 knock+0.015
unts+0.015
 guys+0.014
 spe+0.014
Neg logit contributions
 brand-0.018
 exactly-0.018
uit-0.017
but-0.017
 perched-0.016
Token kind
Token ID1611
Feature activation-6.36e-02

Pos logit contributions
 nation+0.010
unts+0.009
 knock+0.009
 Find+0.008
 blast+0.008
Neg logit contributions
uit-0.019
 brand-0.019
 exactly-0.019
but-0.019
 perched-0.018
Token and
Token ID290
Feature activation-1.64e-02

Pos logit contributions
 nation+0.075
unts+0.070
 Find+0.067
 guys+0.067
 knock+0.066
Neg logit contributions
 brand-0.106
 exactly-0.102
but-0.098
uit-0.096
 perched-0.095
Token loving
Token ID14442
Feature activation-4.55e-02

Pos logit contributions
 nation+0.021
ce+0.020
ass+0.020
 knock+0.020
unts+0.019
Neg logit contributions
 brand-0.024
 exactly-0.023
but-0.022
uit-0.022
 perched-0.021
Token was
Token ID373
Feature activation-1.33e-02

Pos logit contributions
 nation+0.047
 comes+0.042
unts+0.042
 person+0.041
 guys+0.041
Neg logit contributions
 brand-0.087
 exactly-0.083
but-0.081
 perched-0.081
uit-0.080
Token better
Token ID1365
Feature activation-3.60e-02

Pos logit contributions
 nation+0.012
 knock+0.012
unts+0.012
 person+0.011
 blast+0.010
Neg logit contributions
uit-0.024
but-0.024
 brand-0.023
ishly-0.022
iddled-0.022
Token.
Token ID13
Feature activation-1.63e-02

Pos logit contributions
 nation+0.052
unts+0.049
 knock+0.048
ce+0.047
 Find+0.047
Neg logit contributions
 brand-0.048
 exactly-0.047
uit-0.046
but-0.044
 perched-0.043