TOP ACTIVATIONS
MAX = 4.381

, he wore the biggest grin on his face , even
and daddy decided to take Annie to the park to play
friends . They were called Maggie and Lily . They liked
" Where are we going Jessie ?" Mom my asked .
girl still had a proud grin on her face . She
the man always had a smile on his face .
The puppy had a big grin on his face and he
went home and enjoyed his freshly caught fish . <|endoftext|> Once
, still wearing her biggest smile for the day . <|endoftext|>
, organized and having a smile on his face was the
to side on the cur vy branch and was soon high
him with a big happy grin . They hugged each other
both always had a big smile on their faces . <|endoftext|>
The special menu prevented Laura and Alice from having an
One day , Sarah took Mary to the pool for the
Amy always had a big smile on her face . She
. He had a big grin on his face and said
! Everyone had a big grin on their face . It
people and always had a smile on her face .
. He had a big grin on his face , adm

MIN ACTIVATIONS
MIN = -4.817

ended up in such bad situations . The
him how to think of situations from different perspectives , and
parents could lead to dangerous situations . From that day on
also help you in tough situations . From that day on
could help him in scary situations . He continued to balance
out of difficult and sticky situations . <|endoftext|> The family had
to stay calm in scary situations . So
lately . With work and err ands , they didn 't
cr ay ons . It er ases their animals and flowers
angry ". The 3 - year - old thought hard
ases their drawings . It er ases their cr ay ons
liked to tie yarn around things , like his toys and
on their papers . It er ases their drawings . It
was important to value all living creatures . <|endoftext|> Once upon
be happy . Even though situations can be serious , it
tried twisting it around other things that he had , like
The 3 - year - old thought hard and smiled
a loud noise . It er ases part of their mountain
brown and it had many holes in it . Rose was
things can come from bad situations . <|endoftext|> Once upon a

INTERVAL 2.082 - 4.381
CONTAINS 0.353%

the cage . They wished they never had to leave him
called Lucy was adm iring Maggie . Lucy was only three
were two best friends , M ina and Nicole . They
a little fish named Nem o . Nem o lived in
butterflies . Eventually , the sun started to set and the

INTERVAL -0.218 - 2.082
CONTAINS 62.445%

a girl named Anna . Anna was three years old and
Anna starts to cry . She is hurt and
stay in the perfect sky . <|endoftext|> Once there was a
cold wind and heard a strange noise . "
, look at my flower ! It bloom ed !" she

INTERVAL -2.517 - -0.218
CONTAINS 37.053%

treasure map !" Ben said . " Let 's follow it
with excitement and cl apped his hands .
friend , Sue . They played on the swings and in
to put on top of her tower . She reached for
go for a walk in the park . He slowly walked

INTERVAL -4.817 - -2.517
CONTAINS 0.149%

no because it was too expensive . L ily
was singing ! The little people were so happy to see
of bugs and small swimming things , but he kept looking
it out . They all sw am over to the flame
and the 3 - year - old followed him , laughing
Token,
Token ID11
Feature activation+1.62e-02

Pos logit contributions
 band+0.021
aid+0.020
P+0.019
 gol+0.018
 basketball+0.018
Neg logit contributions
rant-0.017
 enclosure-0.016
iously-0.016
 horizon-0.016
 peak-0.016
Token he
Token ID339
Feature activation+1.64e-01

Pos logit contributions
rant+0.000
 enclosure+0.000
 horizon+0.000
 peak+0.000
sing+0.000
Neg logit contributions
P-0.000
 band-0.000
 jacket-0.000
aid-0.000
 gum-0.000
Token wore
Token ID12408
Feature activation-6.21e-01

Pos logit contributions
rant+0.005
 enclosure+0.004
 horizon+0.004
sing+0.004
asser+0.004
Neg logit contributions
 band-0.005
 gum-0.005
P-0.005
 basketball-0.005
 matters-0.004
Token the
Token ID262
Feature activation+5.98e-01

Pos logit contributions
 band+0.011
P+0.010
 gol+0.009
 gum+0.009
Ol+0.009
Neg logit contributions
rant-0.024
 horizon-0.022
 peak-0.021
 enclosure-0.021
 definitely-0.020
Token biggest
Token ID4094
Feature activation+9.46e-01

Pos logit contributions
rant+0.018
sing+0.017
 horizon+0.017
 enclosure+0.017
 doorstep+0.015
Neg logit contributions
 band-0.017
P-0.017
 gum-0.016
 basketball-0.016
 matters-0.016
Token grin
Token ID16770
Feature activation+4.36e+00

Pos logit contributions
rant+0.030
 enclosure+0.028
 horizon+0.027
sing+0.027
 doorstep+0.025
Neg logit contributions
 band-0.025
P-0.025
 gum-0.024
 matters-0.023
 basketball-0.023
Token on
Token ID319
Feature activation+4.76e-01

Pos logit contributions
rant+0.173
 horizon+0.154
 enclosure+0.152
sing+0.151
iously+0.146
Neg logit contributions
 band-0.084
P-0.080
 gum-0.071
 basketball-0.071
 matters-0.068
Token his
Token ID465
Feature activation-5.55e-01

Pos logit contributions
rant+0.020
 horizon+0.018
 peak+0.017
iously+0.017
 enclosure+0.017
Neg logit contributions
 band-0.008
 jacket-0.006
 gum-0.006
aid-0.006
P-0.006
Token face
Token ID1986
Feature activation+1.02e+00

Pos logit contributions
 band+0.015
P+0.014
 gum+0.014
 basketball+0.014
 matters+0.013
Neg logit contributions
rant-0.019
sing-0.017
 enclosure-0.017
 horizon-0.016
 doorstep-0.015
Token,
Token ID11
Feature activation+6.23e-01

Pos logit contributions
rant+0.031
 horizon+0.027
 enclosure+0.027
sing+0.027
iously+0.025
Neg logit contributions
 band-0.031
P-0.030
 matters-0.028
 gum-0.028
 basketball-0.027
Token even
Token ID772
Feature activation-4.88e-01

Pos logit contributions
rant+0.017
sing+0.016
 horizon+0.015
 enclosure+0.015
iously+0.013
Neg logit contributions
P-0.020
 band-0.020
 gum-0.018
 matters-0.018
 basketball-0.018
Token and
Token ID290
Feature activation-7.49e-01

Pos logit contributions
rant+0.029
 enclosure+0.026
 horizon+0.025
sing+0.025
 doorstep+0.023
Neg logit contributions
 band-0.026
P-0.024
 gum-0.024
 basketball-0.023
 matters-0.022
Token daddy
Token ID35822
Feature activation+5.48e-01

Pos logit contributions
P+0.021
 band+0.021
aid+0.021
 gum+0.021
 basketball+0.018
Neg logit contributions
rant-0.020
sing-0.019
iously-0.017
 enclosure-0.017
 horizon-0.017
Token decided
Token ID3066
Feature activation+7.17e-01

Pos logit contributions
rant+0.019
 enclosure+0.017
 horizon+0.016
sing+0.016
 doorstep+0.015
Neg logit contributions
 band-0.015
 gum-0.014
P-0.014
 basketball-0.013
 matters-0.012
Token to
Token ID284
Feature activation-6.21e-02

Pos logit contributions
rant+0.017
iously+0.015
 horizon+0.015
 enclosure+0.015
 doorstep+0.015
Neg logit contributions
P-0.024
 band-0.023
aid-0.022
Ol-0.022
 gum-0.022
Token take
Token ID1011
Feature activation+1.44e-01

Pos logit contributions
P+0.002
 band+0.002
 gum+0.002
aid+0.002
Ol+0.002
Neg logit contributions
rant-0.001
 enclosure-0.001
 doorstep-0.001
 peak-0.001
 horizon-0.001
Token Annie
Token ID25860
Feature activation+4.19e+00

Pos logit contributions
rant+0.005
 enclosure+0.004
 horizon+0.004
ples+0.004
iously+0.004
Neg logit contributions
 band-0.003
aid-0.003
Ol-0.003
P-0.003
 gol-0.003
Token to
Token ID284
Feature activation+2.74e-01

Pos logit contributions
rant+0.139
 enclosure+0.127
sing+0.124
 horizon+0.120
 doorstep+0.116
Neg logit contributions
 band-0.115
P-0.107
 gum-0.106
 basketball-0.102
 matters-0.098
Token the
Token ID262
Feature activation+5.36e-01

Pos logit contributions
rant+0.009
 enclosure+0.009
sing+0.009
 doorstep+0.009
 horizon+0.009
Neg logit contributions
P-0.006
aid-0.006
 band-0.006
 gum-0.005
Ol-0.005
Token park
Token ID3952
Feature activation+1.71e+00

Pos logit contributions
rant+0.017
 enclosure+0.016
sing+0.015
 horizon+0.015
 doorstep+0.014
Neg logit contributions
 band-0.015
P-0.014
 gum-0.014
 matters-0.013
 basketball-0.013
Token to
Token ID284
Feature activation-3.57e-01

Pos logit contributions
rant+0.049
 enclosure+0.048
sing+0.043
 horizon+0.042
 doorstep+0.041
Neg logit contributions
 band-0.053
P-0.050
 gum-0.050
 matters-0.048
Ol-0.046
Token play
Token ID711
Feature activation-3.48e-01

Pos logit contributions
 band+0.011
 gum+0.011
P+0.011
aid+0.010
 matters+0.010
Neg logit contributions
rant-0.010
 enclosure-0.009
 horizon-0.008
 doorstep-0.008
 peak-0.008
Token friends
Token ID2460
Feature activation-1.42e-01

Pos logit contributions
 band+0.002
 matters+0.002
 gum+0.002
P+0.002
 basketball+0.002
Neg logit contributions
 enclosure-0.008
sing-0.008
rant-0.008
asser-0.007
 horizon-0.007
Token.
Token ID13
Feature activation+1.44e+00

Pos logit contributions
 band+0.003
P+0.003
 gum+0.003
 basketball+0.003
 matters+0.003
Neg logit contributions
rant-0.005
 horizon-0.005
sing-0.005
 enclosure-0.005
 doorstep-0.005
Token They
Token ID1119
Feature activation+3.97e-01

Pos logit contributions
rant+0.048
 horizon+0.042
sing+0.042
 enclosure+0.042
asser+0.039
Neg logit contributions
 band-0.040
P-0.038
 gum-0.038
 basketball-0.037
 matters-0.035
Token were
Token ID547
Feature activation+1.02e+00

Pos logit contributions
rant+0.013
 enclosure+0.011
sing+0.011
 horizon+0.010
asser+0.010
Neg logit contributions
 band-0.012
 gum-0.011
P-0.011
 basketball-0.010
 matters-0.010
Token called
Token ID1444
Feature activation+1.03e+00

Pos logit contributions
rant+0.038
sing+0.035
 enclosure+0.035
 horizon+0.033
asser+0.032
Neg logit contributions
 band-0.024
P-0.021
 gum-0.021
 basketball-0.020
 matters-0.019
Token Maggie
Token ID33014
Feature activation+4.17e+00

Pos logit contributions
sing+0.034
 enclosure+0.033
rant+0.033
 horizon+0.029
 doorstep+0.028
Neg logit contributions
P-0.029
 band-0.026
aid-0.024
 gum-0.023
 matters-0.023
Token and
Token ID290
Feature activation-6.95e-01

Pos logit contributions
rant+0.157
 horizon+0.142
sing+0.138
 enclosure+0.137
asser+0.132
Neg logit contributions
 band-0.104
 gum-0.088
P-0.088
 basketball-0.084
 mar-0.079
Token Lily
Token ID20037
Feature activation+3.43e-01

Pos logit contributions
 band+0.016
 gum+0.014
P+0.014
 matters+0.013
 basketball+0.013
Neg logit contributions
sing-0.025
rant-0.025
 enclosure-0.022
 definitely-0.021
elope-0.020
Token.
Token ID13
Feature activation+1.11e+00

Pos logit contributions
rant+0.013
 horizon+0.012
 enclosure+0.012
sing+0.012
asser+0.011
Neg logit contributions
 band-0.008
 gum-0.007
P-0.007
 basketball-0.007
 matters-0.007
Token They
Token ID1119
Feature activation+3.64e-01

Pos logit contributions
rant+0.045
 doorstep+0.043
sing+0.043
iously+0.042
 horizon+0.042
Neg logit contributions
 his-0.019
 P-0.017
 matters-0.017
P-0.017
 there-0.015
Token liked
Token ID8288
Feature activation+8.25e-01

Pos logit contributions
rant+0.012
 enclosure+0.010
sing+0.010
 horizon+0.010
asser+0.009
Neg logit contributions
 band-0.011
 gum-0.010
P-0.010
 basketball-0.009
 matters-0.009
Token "
Token ID366
Feature activation+1.37e+00

Pos logit contributions
rant+0.052
asser+0.047
sing+0.047
ns+0.044
ines+0.043
Neg logit contributions
 his-0.026
 matters-0.025
 there-0.023
 the-0.022
 their-0.022
TokenWhere
Token ID8496
Feature activation+1.19e-01

Pos logit contributions
rant+0.052
sing+0.048
 enclosure+0.048
 horizon+0.047
asser+0.045
Neg logit contributions
 band-0.032
P-0.030
 gum-0.027
 basketball-0.027
 matters-0.025
Token are
Token ID389
Feature activation-1.38e+00

Pos logit contributions
rant+0.003
sing+0.003
 enclosure+0.003
 definitely+0.003
asser+0.003
Neg logit contributions
P-0.004
 band-0.004
 basketball-0.003
 gum-0.003
 gol-0.003
Token we
Token ID356
Feature activation+1.60e-01

Pos logit contributions
 band+0.038
 gum+0.035
P+0.035
 basketball+0.034
aid+0.034
Neg logit contributions
rant-0.041
 enclosure-0.040
 horizon-0.038
 doorstep-0.036
sing-0.035
Token going
Token ID1016
Feature activation-3.05e-01

Pos logit contributions
rant+0.005
sing+0.005
 enclosure+0.005
 horizon+0.005
asser+0.004
Neg logit contributions
 band-0.004
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
Token Jessie
Token ID43425
Feature activation+4.16e+00

Pos logit contributions
 band+0.008
P+0.007
 gum+0.007
 basketball+0.007
 matters+0.006
Neg logit contributions
rant-0.011
sing-0.010
 enclosure-0.010
 horizon-0.010
asser-0.009
Token?"
Token ID1701
Feature activation+8.97e-01

Pos logit contributions
rant+0.145
sing+0.131
 enclosure+0.131
 horizon+0.128
asser+0.120
Neg logit contributions
 band-0.110
P-0.099
 gum-0.098
 basketball-0.095
 matters-0.092
Token Mom
Token ID11254
Feature activation-9.39e-01

Pos logit contributions
rant+0.034
 enclosure+0.031
sing+0.030
 horizon+0.030
asser+0.028
Neg logit contributions
 band-0.021
P-0.019
 gum-0.019
 basketball-0.018
 matters-0.017
Tokenmy
Token ID1820
Feature activation-1.72e+00

Pos logit contributions
 band+0.028
P+0.026
 gum+0.026
 basketball+0.025
 matters+0.024
Neg logit contributions
rant-0.029
 enclosure-0.026
sing-0.026
 horizon-0.025
 doorstep-0.023
Token asked
Token ID1965
Feature activation+5.36e-01

Pos logit contributions
P+0.052
 gum+0.050
 basketball+0.050
 band+0.048
 matters+0.047
Neg logit contributions
rant-0.048
 enclosure-0.042
bell-0.042
sing-0.041
iously-0.041
Token.
Token ID13
Feature activation-2.14e-01

Pos logit contributions
rant+0.017
sing+0.015
 enclosure+0.015
iously+0.014
 horizon+0.014
Neg logit contributions
 band-0.016
P-0.015
 gum-0.015
 basketball-0.015
 mar-0.014
Token girl
Token ID2576
Feature activation-2.97e-01

Pos logit contributions
 horizon+0.005
rant+0.005
 enclosure+0.004
sing+0.004
 doorstep+0.004
Neg logit contributions
P-0.004
Ol-0.004
 gum-0.004
 matters-0.004
 band-0.004
Token still
Token ID991
Feature activation-7.13e-01

Pos logit contributions
 band+0.008
P+0.008
 gum+0.008
 basketball+0.007
 matters+0.007
Neg logit contributions
rant-0.010
 enclosure-0.009
sing-0.009
 horizon-0.009
 doorstep-0.008
Token had
Token ID550
Feature activation+4.15e-01

Pos logit contributions
 band+0.020
P+0.019
 gum+0.019
 basketball+0.018
 matters+0.017
Neg logit contributions
rant-0.023
sing-0.021
 enclosure-0.021
 horizon-0.020
iously-0.019
Token a
Token ID257
Feature activation+4.72e-01

Pos logit contributions
rant+0.013
 enclosure+0.012
 horizon+0.012
sing+0.012
 peak+0.011
Neg logit contributions
 band-0.011
P-0.011
 basketball-0.010
 gum-0.010
Ol-0.010
Token proud
Token ID6613
Feature activation+1.52e+00

Pos logit contributions
rant+0.016
sing+0.015
 enclosure+0.015
 horizon+0.015
 doorstep+0.014
Neg logit contributions
 band-0.012
P-0.011
 gum-0.011
 basketball-0.011
 matters-0.011
Token grin
Token ID16770
Feature activation+4.15e+00

Pos logit contributions
sing+0.043
rant+0.043
 horizon+0.042
 enclosure+0.041
 doorstep+0.038
Neg logit contributions
P-0.046
 band-0.043
 gum-0.042
Ol-0.041
 matters-0.041
Token on
Token ID319
Feature activation+5.15e-01

Pos logit contributions
rant+0.155
 enclosure+0.140
 horizon+0.137
sing+0.135
iously+0.130
Neg logit contributions
 band-0.094
P-0.089
 gum-0.080
 basketball-0.079
 matters-0.078
Token her
Token ID607
Feature activation-4.89e-01

Pos logit contributions
rant+0.022
 enclosure+0.020
 horizon+0.020
sing+0.019
iously+0.019
Neg logit contributions
 band-0.009
P-0.008
 gum-0.008
 basketball-0.007
 mar-0.007
Token face
Token ID1986
Feature activation+6.24e-01

Pos logit contributions
 band+0.013
 gum+0.012
P+0.011
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.018
 enclosure-0.016
sing-0.015
 horizon-0.015
asser-0.014
Token.
Token ID13
Feature activation-2.04e-01

Pos logit contributions
rant+0.019
 enclosure+0.017
 horizon+0.017
sing+0.017
iously+0.016
Neg logit contributions
 band-0.019
P-0.018
 gum-0.017
 basketball-0.017
 matters-0.017
Token She
Token ID1375
Feature activation-1.18e-01

Pos logit contributions
 band+0.007
P+0.007
 gum+0.006
 basketball+0.006
 matters+0.006
Neg logit contributions
rant-0.005
 enclosure-0.005
 horizon-0.005
sing-0.005
 doorstep-0.004
Token the
Token ID262
Feature activation+4.72e-02

Pos logit contributions
rant+0.019
 enclosure+0.017
sing+0.017
 horizon+0.017
iously+0.016
Neg logit contributions
 band-0.014
P-0.013
 gum-0.012
 basketball-0.012
 matters-0.011
Token man
Token ID582
Feature activation+8.30e-01

Pos logit contributions
rant+0.002
 horizon+0.002
 enclosure+0.002
sing+0.002
 doorstep+0.002
Neg logit contributions
 band-0.001
P-0.001
 basketball-0.001
 gum-0.001
 matters-0.001
Token always
Token ID1464
Feature activation-4.62e-01

Pos logit contributions
rant+0.025
 enclosure+0.023
sing+0.022
 horizon+0.022
 doorstep+0.020
Neg logit contributions
 band-0.025
P-0.023
 gum-0.023
 basketball-0.023
 matters-0.021
Token had
Token ID550
Feature activation+1.07e+00

Pos logit contributions
 band+0.013
 gum+0.013
 basketball+0.013
P+0.013
 baseball+0.012
Neg logit contributions
rant-0.014
sing-0.012
 enclosure-0.012
 horizon-0.011
ples-0.011
Token a
Token ID257
Feature activation+1.13e+00

Pos logit contributions
rant+0.031
 definitely+0.028
ples+0.028
 enclosure+0.028
anne+0.026
Neg logit contributions
 basketball-0.026
 band-0.025
 gol-0.024
P-0.024
 jacket-0.023
Token smile
Token ID8212
Feature activation+4.06e+00

Pos logit contributions
rant+0.034
sing+0.034
 enclosure+0.034
 horizon+0.033
 doorstep+0.032
Neg logit contributions
P-0.028
 band-0.027
 matters-0.027
 gum-0.026
 basketball-0.025
Token on
Token ID319
Feature activation+7.33e-01

Pos logit contributions
rant+0.132
sing+0.117
 enclosure+0.116
 horizon+0.114
ples+0.107
Neg logit contributions
P-0.108
 band-0.107
 gum-0.098
 matters-0.097
 basketball-0.096
Token his
Token ID465
Feature activation-6.78e-01

Pos logit contributions
rant+0.032
 enclosure+0.029
sing+0.028
iously+0.027
 horizon+0.027
Neg logit contributions
 band-0.012
P-0.010
 gum-0.010
 basketball-0.009
 jacket-0.009
Token face
Token ID1986
Feature activation+1.01e+00

Pos logit contributions
 band+0.020
P+0.019
 gum+0.018
 basketball+0.018
 matters+0.017
Neg logit contributions
rant-0.021
sing-0.020
 enclosure-0.019
 horizon-0.019
 doorstep-0.018
Token.
Token ID13
Feature activation+4.86e-01

Pos logit contributions
rant+0.031
 enclosure+0.027
 horizon+0.027
sing+0.027
iously+0.024
Neg logit contributions
 band-0.030
P-0.030
 gum-0.027
 matters-0.027
 basketball-0.027
Token
Token ID220
Feature activation+1.13e+00

Pos logit contributions
rant+0.020
sing+0.018
 enclosure+0.018
 horizon+0.017
 doorstep+0.017
Neg logit contributions
 band-0.010
P-0.009
 gum-0.009
 basketball-0.009
 matters-0.008
Token The
Token ID383
Feature activation+2.03e-01

Pos logit contributions
rant+0.009
 enclosure+0.009
sing+0.008
 horizon+0.008
 doorstep+0.007
Neg logit contributions
 band-0.011
P-0.010
 gum-0.010
 basketball-0.009
 matters-0.009
Token puppy
Token ID26188
Feature activation+3.21e-01

Pos logit contributions
 horizon+0.006
rant+0.006
 doorstep+0.006
sing+0.006
 enclosure+0.005
Neg logit contributions
P-0.005
 band-0.005
 matters-0.005
Ol-0.005
aid-0.005
Token had
Token ID550
Feature activation+7.65e-01

Pos logit contributions
rant+0.010
 enclosure+0.009
sing+0.009
 horizon+0.009
asser+0.008
Neg logit contributions
 band-0.009
P-0.009
 gum-0.009
 basketball-0.008
 matters-0.008
Token a
Token ID257
Feature activation+1.83e-01

Pos logit contributions
rant+0.027
 enclosure+0.026
sing+0.026
 horizon+0.025
 doorstep+0.024
Neg logit contributions
 band-0.016
P-0.016
 basketball-0.015
 gum-0.015
 matters-0.014
Token big
Token ID1263
Feature activation+3.36e-01

Pos logit contributions
sing+0.006
rant+0.006
 enclosure+0.006
 horizon+0.005
 doorstep+0.005
Neg logit contributions
 band-0.004
P-0.004
 matters-0.004
 gum-0.004
aid-0.004
Token grin
Token ID16770
Feature activation+4.05e+00

Pos logit contributions
rant+0.012
sing+0.011
 enclosure+0.011
 horizon+0.010
 doorstep+0.010
Neg logit contributions
 band-0.008
P-0.008
 gum-0.007
 matters-0.007
 basketball-0.007
Token on
Token ID319
Feature activation+7.62e-01

Pos logit contributions
rant+0.166
 enclosure+0.154
sing+0.149
 horizon+0.147
asser+0.142
Neg logit contributions
 band-0.074
P-0.070
 matters-0.061
 gum-0.061
Ol-0.060
Token his
Token ID465
Feature activation-5.75e-01

Pos logit contributions
rant+0.043
 enclosure+0.040
sing+0.040
 horizon+0.039
 doorstep+0.038
Neg logit contributions
 band-0.004
P-0.003
 gum-0.002
 basketball-0.002
 matters-0.001
Token face
Token ID1986
Feature activation+4.94e-01

Pos logit contributions
 band+0.014
 gum+0.012
 basketball+0.012
P+0.012
 his+0.011
Neg logit contributions
rant-0.022
 enclosure-0.020
 horizon-0.019
sing-0.019
asser-0.018
Token and
Token ID290
Feature activation-3.58e-02

Pos logit contributions
rant+0.016
 enclosure+0.015
sing+0.014
 horizon+0.014
 doorstep+0.013
Neg logit contributions
 band-0.014
P-0.013
 gum-0.012
 basketball-0.012
 matters-0.012
Token he
Token ID339
Feature activation-3.14e-01

Pos logit contributions
P+0.001
 band+0.001
aid+0.001
 gum+0.001
Ol+0.001
Neg logit contributions
sing-0.001
 enclosure-0.001
rant-0.001
 horizon-0.001
iously-0.001
Token went
Token ID1816
Feature activation-5.79e-01

Pos logit contributions
rant+0.019
 enclosure+0.017
sing+0.017
 horizon+0.016
 doorstep+0.015
Neg logit contributions
P-0.015
 gum-0.015
 band-0.015
 basketball-0.014
 matters-0.014
Token home
Token ID1363
Feature activation+5.96e-01

Pos logit contributions
 band+0.016
 matters+0.016
P+0.016
 gum+0.016
Ol+0.015
Neg logit contributions
rant-0.018
 enclosure-0.016
 horizon-0.015
sing-0.015
 entrance-0.014
Token and
Token ID290
Feature activation-3.61e-01

Pos logit contributions
rant+0.017
sing+0.017
 enclosure+0.016
 horizon+0.015
 doorstep+0.015
Neg logit contributions
P-0.018
 band-0.017
 gum-0.017
 matters-0.017
 basketball-0.016
Token enjoyed
Token ID8359
Feature activation+8.74e-01

Pos logit contributions
P+0.010
 band+0.010
 gum+0.010
aid+0.009
 basketball+0.009
Neg logit contributions
rant-0.011
 enclosure-0.010
sing-0.009
 doorstep-0.009
 horizon-0.009
Token his
Token ID465
Feature activation+4.24e-02

Pos logit contributions
rant+0.029
 enclosure+0.028
sing+0.027
 horizon+0.026
 peak+0.025
Neg logit contributions
P-0.019
 band-0.018
Ol-0.017
aid-0.017
 matters-0.016
Token freshly
Token ID29026
Feature activation+3.95e+00

Pos logit contributions
rant+0.001
sing+0.001
 enclosure+0.001
 horizon+0.001
 doorstep+0.001
Neg logit contributions
 band-0.001
P-0.001
 gum-0.001
 matters-0.001
 basketball-0.001
Token caught
Token ID4978
Feature activation+6.37e-01

Pos logit contributions
rant+0.134
 enclosure+0.121
sing+0.118
 horizon+0.115
asser+0.110
Neg logit contributions
 band-0.103
 gum-0.094
 basketball-0.094
P-0.093
 matters-0.089
Token fish
Token ID5916
Feature activation+3.02e-01

Pos logit contributions
rant+0.020
 enclosure+0.018
 horizon+0.018
sing+0.017
 peak+0.017
Neg logit contributions
P-0.017
Ol-0.016
 band-0.016
 matters-0.015
aid-0.015
Token.
Token ID13
Feature activation+5.17e-01

Pos logit contributions
rant+0.009
 enclosure+0.009
sing+0.009
 horizon+0.008
 doorstep+0.008
Neg logit contributions
 band-0.008
P-0.008
 matters-0.008
 gum-0.008
 basketball-0.007
Token<|endoftext|>
Token ID50256
Feature activation+2.01e-01

Pos logit contributions
rant+0.019
 enclosure+0.017
sing+0.017
 horizon+0.017
asser+0.016
Neg logit contributions
 band-0.013
P-0.012
 gum-0.012
 basketball-0.011
 matters-0.011
TokenOnce
Token ID7454
Feature activation-1.42e-01

Pos logit contributions
rant+0.006
 enclosure+0.006
 horizon+0.005
asser+0.005
sing+0.005
Neg logit contributions
 band-0.006
 gum-0.005
P-0.005
 basketball-0.005
aid-0.005
Token,
Token ID11
Feature activation+1.32e-01

Pos logit contributions
rant+0.009
sing+0.008
 horizon+0.008
 enclosure+0.008
 doorstep+0.008
Neg logit contributions
P-0.008
 band-0.008
 gum-0.008
 matters-0.007
 basketball-0.007
Token still
Token ID991
Feature activation-1.09e+00

Pos logit contributions
rant+0.004
sing+0.003
 enclosure+0.003
 horizon+0.003
 doorstep+0.003
Neg logit contributions
 band-0.004
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
Token wearing
Token ID5762
Feature activation-8.74e-01

Pos logit contributions
P+0.035
 band+0.034
 gum+0.032
 basketball+0.031
Ol+0.030
Neg logit contributions
rant-0.030
sing-0.029
 enclosure-0.027
 horizon-0.027
 doorstep-0.025
Token her
Token ID607
Feature activation-4.24e-01

Pos logit contributions
P+0.021
 gol+0.020
 band+0.018
Ol+0.018
aid+0.017
Neg logit contributions
rant-0.029
 horizon-0.025
ples-0.024
 definitely-0.023
bell-0.023
Token biggest
Token ID4094
Feature activation+9.56e-01

Pos logit contributions
 band+0.010
P+0.010
 matters+0.009
 gum+0.008
 gol+0.008
Neg logit contributions
rant-0.014
sing-0.014
 horizon-0.014
 enclosure-0.013
asser-0.013
Token smile
Token ID8212
Feature activation+3.95e+00

Pos logit contributions
rant+0.033
 enclosure+0.033
 horizon+0.033
sing+0.032
 doorstep+0.032
Neg logit contributions
P-0.019
 matters-0.018
 band-0.017
Ol-0.016
aid-0.015
Token for
Token ID329
Feature activation+1.18e+00

Pos logit contributions
rant+0.162
 horizon+0.144
sing+0.142
 enclosure+0.140
 doorstep+0.135
Neg logit contributions
 band-0.075
P-0.073
 basketball-0.064
 gum-0.063
Ol-0.060
Token the
Token ID262
Feature activation+4.59e-01

Pos logit contributions
rant+0.046
sing+0.043
 enclosure+0.040
 horizon+0.040
iously+0.040
Neg logit contributions
P-0.022
 band-0.022
 gum-0.019
 basketball-0.019
 matters-0.019
Token day
Token ID1110
Feature activation+2.24e+00

Pos logit contributions
 horizon+0.017
rant+0.016
sing+0.015
 doorstep+0.015
 previous+0.015
Neg logit contributions
P-0.010
 matters-0.009
 band-0.009
aid-0.007
 gum-0.007
Token.
Token ID13
Feature activation+9.62e-02

Pos logit contributions
rant+0.072
 enclosure+0.064
sing+0.064
 horizon+0.064
 doorstep+0.059
Neg logit contributions
 band-0.063
P-0.061
 gum-0.058
 basketball-0.057
 matters-0.055
Token<|endoftext|>
Token ID50256
Feature activation-2.90e-01

Pos logit contributions
rant+0.004
 enclosure+0.004
sing+0.004
 horizon+0.003
asser+0.003
Neg logit contributions
 band-0.002
P-0.002
 gum-0.002
 basketball-0.002
 matters-0.002
Token,
Token ID11
Feature activation+8.25e-01

Pos logit contributions
rant+0.002
 horizon+0.001
 enclosure+0.001
sing+0.001
iously+0.001
Neg logit contributions
P-0.002
 band-0.002
Ol-0.002
 basketball-0.002
 gum-0.002
Token organized
Token ID8389
Feature activation-6.92e-01

Pos logit contributions
rant+0.021
 horizon+0.020
sing+0.019
 enclosure+0.019
iously+0.018
Neg logit contributions
P-0.028
 band-0.025
 basketball-0.023
Ol-0.022
 gum-0.022
Token and
Token ID290
Feature activation-3.39e-01

Pos logit contributions
P+0.025
 band+0.022
Ol+0.021
aid+0.021
 basketball+0.020
Neg logit contributions
rant-0.018
 horizon-0.015
 enclosure-0.014
iously-0.013
sing-0.013
Token having
Token ID1719
Feature activation+6.74e-01

Pos logit contributions
P+0.010
 band+0.009
Ol+0.008
 basketball+0.008
 gum+0.008
Neg logit contributions
rant-0.010
sing-0.009
 horizon-0.009
 enclosure-0.009
 definitely-0.009
Token a
Token ID257
Feature activation+8.27e-01

Pos logit contributions
rant+0.018
 definitely+0.017
 horizon+0.016
 enclosure+0.016
 previous+0.016
Neg logit contributions
P-0.019
 band-0.018
aid-0.016
 jacket-0.016
 basketball-0.016
Token smile
Token ID8212
Feature activation+3.93e+00

Pos logit contributions
rant+0.028
sing+0.026
 horizon+0.026
 enclosure+0.026
asser+0.025
Neg logit contributions
 band-0.017
P-0.017
 gum-0.016
 matters-0.016
 basketball-0.015
Token on
Token ID319
Feature activation+5.94e-01

Pos logit contributions
rant+0.142
sing+0.124
 enclosure+0.123
 horizon+0.121
ples+0.116
Neg logit contributions
P-0.092
 band-0.090
 gum-0.079
 basketball-0.077
Ol-0.076
Token his
Token ID465
Feature activation-6.97e-01

Pos logit contributions
rant+0.027
 enclosure+0.023
iously+0.023
 horizon+0.023
asser+0.023
Neg logit contributions
 band-0.009
P-0.008
aid-0.006
 gol-0.006
Ol-0.006
Token face
Token ID1986
Feature activation+7.89e-01

Pos logit contributions
 band+0.020
P+0.020
 gum+0.018
 basketball+0.018
 matters+0.018
Neg logit contributions
rant-0.021
sing-0.020
 horizon-0.019
 enclosure-0.019
 doorstep-0.018
Token was
Token ID373
Feature activation+1.07e+00

Pos logit contributions
rant+0.024
 enclosure+0.020
 horizon+0.020
sing+0.020
iously+0.019
Neg logit contributions
 band-0.024
P-0.024
 gum-0.022
 basketball-0.021
 matters-0.021
Token the
Token ID262
Feature activation+9.47e-03

Pos logit contributions
rant+0.033
sing+0.027
 horizon+0.027
 definitely+0.027
 previous+0.026
Neg logit contributions
P-0.031
 band-0.030
 basketball-0.028
Ol-0.027
 gum-0.027
Token to
Token ID284
Feature activation+3.31e-01

Pos logit contributions
rant+0.008
 enclosure+0.008
 horizon+0.008
sing+0.007
iously+0.007
Neg logit contributions
P-0.009
 band-0.009
 matters-0.009
 gum-0.008
 basketball-0.008
Token side
Token ID1735
Feature activation-1.93e-01

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.012
 peak+0.011
Neg logit contributions
 band-0.007
P-0.006
 gum-0.006
 basketball-0.006
 matters-0.005
Token on
Token ID319
Feature activation-3.68e-01

Pos logit contributions
P+0.006
 band+0.005
 matters+0.005
 gum+0.005
 basketball+0.005
Neg logit contributions
rant-0.006
 enclosure-0.006
 horizon-0.005
sing-0.005
 peak-0.005
Token the
Token ID262
Feature activation+5.15e-01

Pos logit contributions
 band+0.008
P+0.008
aid+0.007
 matters+0.007
 basketball+0.007
Neg logit contributions
rant-0.014
 enclosure-0.013
 horizon-0.013
 peak-0.013
sing-0.012
Token cur
Token ID1090
Feature activation+1.67e-01

Pos logit contributions
rant+0.020
sing+0.016
ples+0.016
asser+0.016
 enclosure+0.016
Neg logit contributions
 band-0.014
 gum-0.012
 basketball-0.011
 his-0.011
P-0.011
Tokenvy
Token ID7670
Feature activation+3.92e+00

Pos logit contributions
rant+0.005
sing+0.005
 enclosure+0.005
 horizon+0.005
 doorstep+0.004
Neg logit contributions
 band-0.005
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
Token branch
Token ID8478
Feature activation+6.83e-01

Pos logit contributions
rant+0.126
asser+0.102
sing+0.099
 horizon+0.097
 enclosure+0.094
Neg logit contributions
 band-0.123
 gum-0.105
 basketball-0.097
P-0.096
 his-0.095
Token and
Token ID290
Feature activation-2.02e-01

Pos logit contributions
rant+0.021
 enclosure+0.020
sing+0.019
 horizon+0.019
iously+0.017
Neg logit contributions
P-0.020
 band-0.020
 matters-0.019
 gum-0.018
 basketball-0.018
Token was
Token ID373
Feature activation-1.43e-01

Pos logit contributions
P+0.006
 band+0.006
 gum+0.006
 basketball+0.005
 matters+0.005
Neg logit contributions
rant-0.006
 enclosure-0.005
sing-0.005
 peak-0.005
 horizon-0.005
Token soon
Token ID2582
Feature activation-7.77e-01

Pos logit contributions
 band+0.004
P+0.004
 gum+0.004
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.004
sing-0.004
 enclosure-0.004
 horizon-0.004
iously-0.003
Token high
Token ID1029
Feature activation+6.02e-01

Pos logit contributions
P+0.024
 band+0.024
 gum+0.021
 matters+0.021
 basketball+0.021
Neg logit contributions
rant-0.022
sing-0.021
 horizon-0.021
 enclosure-0.021
 peak-0.020
Token him
Token ID683
Feature activation-3.10e-01

Pos logit contributions
rant+0.017
 enclosure+0.016
 horizon+0.015
sing+0.014
 peak+0.014
Neg logit contributions
P-0.010
 band-0.010
 gum-0.009
Ol-0.008
 basketball-0.008
Token with
Token ID351
Feature activation-9.55e-01

Pos logit contributions
 band+0.009
P+0.008
 gum+0.008
 basketball+0.007
 matters+0.007
Neg logit contributions
rant-0.010
 enclosure-0.010
sing-0.009
 horizon-0.009
 doorstep-0.009
Token a
Token ID257
Feature activation-2.22e-01

Pos logit contributions
P+0.023
 band+0.021
 matters+0.019
 gol+0.019
Ol+0.019
Neg logit contributions
rant-0.030
 enclosure-0.029
 peak-0.028
 horizon-0.028
 entrance-0.027
Token big
Token ID1263
Feature activation+9.98e-02

Pos logit contributions
P+0.006
 matters+0.006
 band+0.006
 gum+0.005
aid+0.005
Neg logit contributions
sing-0.006
rant-0.006
 enclosure-0.006
 horizon-0.006
 peak-0.006
Token happy
Token ID3772
Feature activation+8.47e-01

Pos logit contributions
rant+0.003
sing+0.003
 enclosure+0.003
 horizon+0.003
asser+0.003
Neg logit contributions
 band-0.003
P-0.002
 gum-0.002
 matters-0.002
 basketball-0.002
Token grin
Token ID16770
Feature activation+3.92e+00

Pos logit contributions
sing+0.024
rant+0.023
 enclosure+0.023
 horizon+0.023
 doorstep+0.021
Neg logit contributions
P-0.026
 matters-0.024
 band-0.024
 gum-0.023
Ol-0.023
Token.
Token ID13
Feature activation-4.04e-01

Pos logit contributions
rant+0.136
 enclosure+0.119
sing+0.119
 horizon+0.116
 doorstep+0.110
Neg logit contributions
 band-0.099
P-0.096
 gum-0.087
 basketball-0.086
Ol-0.085
Token They
Token ID1119
Feature activation-1.79e-01

Pos logit contributions
 band+0.011
P+0.010
 gum+0.009
 basketball+0.009
 matters+0.009
Neg logit contributions
rant-0.014
 enclosure-0.013
 horizon-0.013
sing-0.013
 doorstep-0.012
Token hugged
Token ID41130
Feature activation+1.43e+00

Pos logit contributions
 band+0.004
 gum+0.004
P+0.004
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.006
 enclosure-0.006
sing-0.006
 horizon-0.006
 doorstep-0.005
Token each
Token ID1123
Feature activation+1.03e+00

Pos logit contributions
rant+0.038
sing+0.036
 enclosure+0.033
 horizon+0.033
iously+0.032
Neg logit contributions
P-0.047
 band-0.045
 gum-0.042
 basketball-0.041
 matters-0.041
Token other
Token ID584
Feature activation-3.44e-01

Pos logit contributions
rant+0.049
sing+0.046
 enclosure+0.045
 horizon+0.045
asser+0.042
Neg logit contributions
 band-0.016
P-0.013
 gum-0.012
 basketball-0.011
 matters-0.011
Token both
Token ID1111
Feature activation+3.09e-01

Pos logit contributions
rant+0.006
 enclosure+0.006
sing+0.005
 horizon+0.005
asser+0.005
Neg logit contributions
 band-0.006
 gum-0.006
P-0.005
 basketball-0.005
 matters-0.005
Token always
Token ID1464
Feature activation-5.32e-01

Pos logit contributions
rant+0.010
 enclosure+0.009
sing+0.009
 horizon+0.009
asser+0.008
Neg logit contributions
 band-0.009
P-0.008
 gum-0.008
 basketball-0.008
 matters-0.008
Token had
Token ID550
Feature activation+9.13e-01

Pos logit contributions
 band+0.016
 gum+0.015
P+0.015
 basketball+0.015
 matters+0.013
Neg logit contributions
rant-0.016
sing-0.015
 enclosure-0.014
 horizon-0.013
asser-0.013
Token a
Token ID257
Feature activation+8.63e-01

Pos logit contributions
rant+0.028
 enclosure+0.027
sing+0.025
asser+0.024
ples+0.024
Neg logit contributions
 band-0.022
 basketball-0.021
P-0.020
 gum-0.020
 gol-0.019
Token big
Token ID1263
Feature activation+7.77e-01

Pos logit contributions
rant+0.032
 enclosure+0.030
sing+0.030
 horizon+0.029
asser+0.027
Neg logit contributions
 band-0.020
P-0.018
 gum-0.018
 basketball-0.017
 matters-0.017
Token smile
Token ID8212
Feature activation+3.91e+00

Pos logit contributions
rant+0.028
 enclosure+0.024
 horizon+0.024
sing+0.024
iously+0.023
Neg logit contributions
 band-0.021
 gum-0.018
P-0.018
 basketball-0.018
 matters-0.017
Token on
Token ID319
Feature activation+7.10e-01

Pos logit contributions
rant+0.145
 enclosure+0.131
sing+0.131
 horizon+0.124
 doorstep+0.118
Neg logit contributions
 band-0.088
P-0.086
 gum-0.080
 basketball-0.079
 matters-0.075
Token their
Token ID511
Feature activation-1.67e-01

Pos logit contributions
rant+0.030
 enclosure+0.028
sing+0.028
 horizon+0.027
asser+0.026
Neg logit contributions
 band-0.013
P-0.011
 gum-0.011
 basketball-0.011
 matters-0.010
Token faces
Token ID6698
Feature activation+6.20e-01

Pos logit contributions
 band+0.004
 gum+0.004
P+0.004
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.006
 enclosure-0.005
sing-0.005
 horizon-0.005
asser-0.005
Token.
Token ID13
Feature activation+3.94e-01

Pos logit contributions
rant+0.019
 enclosure+0.017
sing+0.016
 horizon+0.016
 doorstep+0.015
Neg logit contributions
 band-0.019
P-0.018
 gum-0.017
 matters-0.017
 basketball-0.017
Token<|endoftext|>
Token ID50256
Feature activation-1.74e-01

Pos logit contributions
rant+0.016
sing+0.014
 enclosure+0.014
 horizon+0.014
 doorstep+0.013
Neg logit contributions
 band-0.008
 gum-0.008
P-0.008
 basketball-0.007
 matters-0.007
Token
Token ID198
Feature activation+6.35e-01

Pos logit contributions
rant+0.005
sing+0.005
 enclosure+0.005
 horizon+0.005
iously+0.005
Neg logit contributions
 band-0.002
P-0.001
 gum-0.001
 matters-0.001
 his-0.001
TokenThe
Token ID464
Feature activation-6.10e-01

Pos logit contributions
rant+0.026
sing+0.026
 enclosure+0.024
 horizon+0.023
iously+0.023
Neg logit contributions
P-0.014
 band-0.013
 gum-0.011
 his-0.010
 basketball-0.010
Token special
Token ID2041
Feature activation+6.24e-01

Pos logit contributions
 band+0.016
P+0.015
 gum+0.015
 basketball+0.014
 his+0.014
Neg logit contributions
rant-0.022
sing-0.020
 enclosure-0.019
 horizon-0.019
iously-0.018
Token menu
Token ID6859
Feature activation-1.56e-01

Pos logit contributions
rant+0.019
sing+0.017
 enclosure+0.017
 horizon+0.016
iously+0.015
Neg logit contributions
 band-0.020
 gum-0.018
P-0.018
 basketball-0.017
 matters-0.017
Token prevented
Token ID13351
Feature activation+9.18e-01

Pos logit contributions
 band+0.004
P+0.004
 gum+0.004
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.005
 enclosure-0.005
 horizon-0.005
sing-0.005
 doorstep-0.004
Token Laura
Token ID16753
Feature activation+3.91e+00

Pos logit contributions
rant+0.030
 enclosure+0.027
sing+0.027
 horizon+0.027
asser+0.025
Neg logit contributions
 band-0.026
P-0.023
 gum-0.023
 basketball-0.023
 matters-0.022
Token and
Token ID290
Feature activation-7.35e-01

Pos logit contributions
rant+0.146
 enclosure+0.134
sing+0.133
 horizon+0.130
 doorstep+0.123
Neg logit contributions
 band-0.092
P-0.084
 gum-0.082
 basketball-0.079
 matters-0.076
Token Alice
Token ID14862
Feature activation+1.57e+00

Pos logit contributions
P+0.020
 band+0.019
 gum+0.018
 basketball+0.018
Ol+0.016
Neg logit contributions
rant-0.025
 enclosure-0.023
sing-0.023
 horizon-0.022
iously-0.021
Token from
Token ID422
Feature activation-7.56e-01

Pos logit contributions
rant+0.055
 enclosure+0.050
sing+0.050
 horizon+0.048
 doorstep+0.045
Neg logit contributions
 band-0.041
P-0.038
 gum-0.037
 basketball-0.036
 matters-0.034
Token having
Token ID1719
Feature activation+7.52e-01

Pos logit contributions
 band+0.019
 gum+0.017
P+0.017
 basketball+0.017
 matters+0.016
Neg logit contributions
rant-0.027
 enclosure-0.025
sing-0.024
 horizon-0.024
asser-0.023
Token an
Token ID281
Feature activation-1.91e-01

Pos logit contributions
rant+0.026
 enclosure+0.024
 horizon+0.024
sing+0.023
asser+0.022
Neg logit contributions
 band-0.018
P-0.017
 basketball-0.016
 gum-0.015
 matters-0.015
TokenOne
Token ID3198
Feature activation+1.04e+00

Pos logit contributions
rant+0.008
sing+0.007
 enclosure+0.007
 horizon+0.007
iously+0.007
Neg logit contributions
P-0.006
 band-0.006
 basketball-0.005
 gum-0.005
 matters-0.005
Token day
Token ID1110
Feature activation-6.37e-01

Pos logit contributions
rant+0.032
sing+0.029
 enclosure+0.028
 horizon+0.028
 peak+0.025
Neg logit contributions
 band-0.031
 gum-0.029
P-0.028
 basketball-0.027
 matters-0.026
Token,
Token ID11
Feature activation+1.95e-01

Pos logit contributions
 band+0.018
 gum+0.016
P+0.016
 basketball+0.015
aid+0.015
Neg logit contributions
rant-0.020
 enclosure-0.018
 horizon-0.018
sing-0.018
iously-0.017
Token Sarah
Token ID10490
Feature activation+7.91e-01

Pos logit contributions
rant+0.005
 enclosure+0.004
sing+0.004
 horizon+0.004
 doorstep+0.004
Neg logit contributions
 band-0.007
P-0.007
 gum-0.007
 basketball-0.006
 matters-0.006
Token took
Token ID1718
Feature activation+4.85e-01

Pos logit contributions
rant+0.025
 enclosure+0.023
 horizon+0.022
sing+0.021
 doorstep+0.020
Neg logit contributions
 band-0.023
P-0.021
 gum-0.021
 basketball-0.020
 matters-0.020
Token Mary
Token ID5335
Feature activation+3.90e+00

Pos logit contributions
rant+0.015
 enclosure+0.013
anne+0.012
ples+0.012
sing+0.012
Neg logit contributions
 band-0.012
P-0.011
Ol-0.011
 jacket-0.011
 gum-0.011
Token to
Token ID284
Feature activation+6.15e-01

Pos logit contributions
rant+0.118
 enclosure+0.114
sing+0.107
 horizon+0.101
 doorstep+0.101
Neg logit contributions
 band-0.114
P-0.112
 gum-0.105
 matters-0.102
aid-0.099
Token the
Token ID262
Feature activation+4.09e-01

Pos logit contributions
rant+0.021
 enclosure+0.020
sing+0.020
 doorstep+0.019
 horizon+0.019
Neg logit contributions
P-0.015
aid-0.013
 band-0.013
 gum-0.013
 matters-0.012
Token pool
Token ID5933
Feature activation+1.44e+00

Pos logit contributions
rant+0.012
 enclosure+0.011
 horizon+0.011
sing+0.011
 doorstep+0.010
Neg logit contributions
 band-0.012
P-0.012
 matters-0.011
 gum-0.011
aid-0.011
Token for
Token ID329
Feature activation+1.05e+00

Pos logit contributions
rant+0.049
 enclosure+0.044
sing+0.043
 horizon+0.043
 doorstep+0.040
Neg logit contributions
 band-0.040
P-0.037
 gum-0.036
 matters-0.035
 basketball-0.034
Token the
Token ID262
Feature activation+1.62e-01

Pos logit contributions
rant+0.043
sing+0.039
 enclosure+0.039
 horizon+0.038
 doorstep+0.037
Neg logit contributions
 band-0.020
P-0.019
 gum-0.018
 basketball-0.017
 matters-0.016
Token Amy
Token ID14235
Feature activation+1.29e+00

Pos logit contributions
rant+0.044
 horizon+0.041
iously+0.040
 enclosure+0.040
sing+0.039
Neg logit contributions
 basketball-0.015
 his-0.014
 gum-0.014
 band-0.013
 matters-0.012
Token always
Token ID1464
Feature activation-5.29e-01

Pos logit contributions
rant+0.046
sing+0.042
 enclosure+0.042
 horizon+0.041
iously+0.039
Neg logit contributions
 band-0.034
P-0.031
 gum-0.030
 basketball-0.029
 matters-0.029
Token had
Token ID550
Feature activation+6.14e-01

Pos logit contributions
 band+0.016
 gum+0.016
P+0.015
 basketball+0.015
 pins+0.014
Neg logit contributions
rant-0.015
sing-0.014
 enclosure-0.013
asser-0.012
 horizon-0.012
Token a
Token ID257
Feature activation+5.64e-01

Pos logit contributions
rant+0.017
sing+0.017
 enclosure+0.017
 horizon+0.015
 previous+0.015
Neg logit contributions
 band-0.015
P-0.014
 gol-0.014
aid-0.014
Ol-0.014
Token big
Token ID1263
Feature activation+6.16e-01

Pos logit contributions
sing+0.018
rant+0.018
 enclosure+0.017
 horizon+0.016
 doorstep+0.016
Neg logit contributions
 band-0.014
P-0.013
 matters-0.013
 gum-0.013
aid-0.012
Token smile
Token ID8212
Feature activation+3.88e+00

Pos logit contributions
rant+0.020
sing+0.019
 enclosure+0.019
 horizon+0.018
asser+0.017
Neg logit contributions
 band-0.017
P-0.016
 gum-0.015
 basketball-0.015
 matters-0.015
Token on
Token ID319
Feature activation+6.11e-01

Pos logit contributions
rant+0.152
 enclosure+0.138
sing+0.138
 horizon+0.134
 doorstep+0.128
Neg logit contributions
 band-0.083
P-0.077
 gum-0.073
 basketball-0.071
 matters-0.069
Token her
Token ID607
Feature activation-4.05e-01

Pos logit contributions
rant+0.029
 enclosure+0.027
sing+0.027
 horizon+0.027
asser+0.026
Neg logit contributions
 band-0.008
P-0.007
 gum-0.006
 basketball-0.006
 matters-0.005
Token face
Token ID1986
Feature activation+7.37e-01

Pos logit contributions
 band+0.012
P+0.012
 gum+0.011
 matters+0.011
 basketball+0.011
Neg logit contributions
rant-0.012
sing-0.012
 enclosure-0.011
 horizon-0.011
 doorstep-0.010
Token.
Token ID13
Feature activation+4.03e-01

Pos logit contributions
rant+0.024
 enclosure+0.022
sing+0.022
 horizon+0.021
 doorstep+0.020
Neg logit contributions
 band-0.021
P-0.019
 gum-0.019
 basketball-0.018
 matters-0.018
Token She
Token ID1375
Feature activation+3.30e-01

Pos logit contributions
rant+0.017
iously+0.015
sing+0.015
 horizon+0.014
 enclosure+0.014
Neg logit contributions
 basketball-0.008
 gum-0.008
 band-0.008
 his-0.008
 matters-0.007
Token.
Token ID13
Feature activation-2.49e-01

Pos logit contributions
rant+0.074
 enclosure+0.068
sing+0.065
 horizon+0.063
 doorstep+0.060
Neg logit contributions
 band-0.071
P-0.069
 gum-0.065
 basketball-0.064
 matters-0.062
Token He
Token ID679
Feature activation-2.94e-02

Pos logit contributions
 band+0.007
 gum+0.006
P+0.006
 basketball+0.006
 matters+0.006
Neg logit contributions
rant-0.009
sing-0.008
 enclosure-0.008
 horizon-0.007
asser-0.007
Token had
Token ID550
Feature activation+5.89e-01

Pos logit contributions
 band+0.001
P+0.001
 gum+0.001
 basketball+0.001
 matters+0.001
Neg logit contributions
rant-0.001
 enclosure-0.001
sing-0.001
 horizon-0.001
asser-0.001
Token a
Token ID257
Feature activation+4.57e-01

Pos logit contributions
rant+0.015
 enclosure+0.014
sing+0.014
 horizon+0.013
 doorstep+0.013
Neg logit contributions
P-0.019
 band-0.019
 basketball-0.018
 gum-0.018
 matters-0.017
Token big
Token ID1263
Feature activation+5.58e-01

Pos logit contributions
rant+0.016
sing+0.015
 enclosure+0.014
 horizon+0.014
 doorstep+0.013
Neg logit contributions
 band-0.012
P-0.011
 gum-0.011
 basketball-0.010
 matters-0.010
Token grin
Token ID16770
Feature activation+3.88e+00

Pos logit contributions
rant+0.019
sing+0.017
 enclosure+0.017
 horizon+0.017
asser+0.015
Neg logit contributions
 band-0.015
P-0.014
 gum-0.014
 basketball-0.013
 matters-0.013
Token on
Token ID319
Feature activation+8.34e-01

Pos logit contributions
rant+0.136
 enclosure+0.126
sing+0.124
 horizon+0.123
 doorstep+0.116
Neg logit contributions
P-0.094
 band-0.088
 matters-0.083
 gum-0.081
Ol-0.080
Token his
Token ID465
Feature activation-6.69e-01

Pos logit contributions
rant+0.043
 enclosure+0.041
sing+0.041
 horizon+0.040
asser+0.038
Neg logit contributions
 band-0.008
P-0.006
 gum-0.006
 basketball-0.005
 matters-0.004
Token face
Token ID1986
Feature activation+5.08e-01

Pos logit contributions
 band+0.018
 gum+0.016
P+0.016
 basketball+0.015
 matters+0.015
Neg logit contributions
rant-0.024
 enclosure-0.021
sing-0.021
 horizon-0.020
asser-0.019
Token and
Token ID290
Feature activation+8.49e-02

Pos logit contributions
rant+0.015
 enclosure+0.014
 horizon+0.014
sing+0.014
iously+0.013
Neg logit contributions
 band-0.015
P-0.015
 matters-0.014
 gum-0.014
 basketball-0.014
Token said
Token ID531
Feature activation-9.22e-02

Pos logit contributions
rant+0.003
 enclosure+0.003
sing+0.003
 horizon+0.002
 doorstep+0.002
Neg logit contributions
P-0.002
 band-0.002
 gum-0.002
 basketball-0.002
Ol-0.002
Token!
Token ID0
Feature activation+3.68e-01

Pos logit contributions
rant+0.052
 enclosure+0.048
sing+0.047
 horizon+0.046
 doorstep+0.043
Neg logit contributions
 band-0.036
 gum-0.034
 basketball-0.032
P-0.032
 matters-0.029
Token Everyone
Token ID11075
Feature activation-1.77e+00

Pos logit contributions
rant+0.011
 enclosure+0.010
sing+0.010
 horizon+0.009
 doorstep+0.009
Neg logit contributions
 band-0.012
P-0.011
 gum-0.011
 basketball-0.010
 matters-0.010
Token had
Token ID550
Feature activation+5.75e-01

Pos logit contributions
 band+0.046
P+0.044
 gum+0.043
 basketball+0.041
Ol+0.038
Neg logit contributions
rant-0.060
 enclosure-0.053
sing-0.052
 horizon-0.051
 doorstep-0.048
Token a
Token ID257
Feature activation+2.07e-01

Pos logit contributions
rant+0.023
 enclosure+0.021
sing+0.021
 horizon+0.021
asser+0.019
Neg logit contributions
 band-0.013
 gum-0.011
P-0.011
 basketball-0.011
 matters-0.010
Token big
Token ID1263
Feature activation+3.10e-01

Pos logit contributions
rant+0.008
 enclosure+0.007
 horizon+0.007
sing+0.007
iously+0.007
Neg logit contributions
 band-0.005
 gum-0.004
 basketball-0.004
P-0.004
 mar-0.004
Token grin
Token ID16770
Feature activation+3.86e+00

Pos logit contributions
rant+0.012
 enclosure+0.011
 horizon+0.010
sing+0.010
iously+0.010
Neg logit contributions
 band-0.008
 gum-0.007
 basketball-0.007
P-0.006
 mar-0.006
Token on
Token ID319
Feature activation+7.62e-01

Pos logit contributions
rant+0.146
 enclosure+0.132
sing+0.131
 horizon+0.129
 doorstep+0.122
Neg logit contributions
 band-0.083
P-0.080
 gum-0.074
 matters-0.073
 basketball-0.072
Token their
Token ID511
Feature activation+5.65e-03

Pos logit contributions
rant+0.038
 enclosure+0.036
sing+0.035
 horizon+0.035
 doorstep+0.033
Neg logit contributions
 band-0.010
P-0.008
 gum-0.008
 basketball-0.007
 matters-0.007
Token face
Token ID1986
Feature activation+5.28e-01

Pos logit contributions
rant+0.000
 enclosure+0.000
 horizon+0.000
sing+0.000
asser+0.000
Neg logit contributions
 band-0.000
 gum-0.000
 basketball-0.000
P-0.000
 matters-0.000
Token.
Token ID13
Feature activation-9.06e-02

Pos logit contributions
rant+0.017
 enclosure+0.015
sing+0.015
 horizon+0.015
 doorstep+0.014
Neg logit contributions
 band-0.015
P-0.014
 gum-0.013
 basketball-0.013
 matters-0.013
Token It
Token ID632
Feature activation-7.04e-01

Pos logit contributions
 band+0.003
 gum+0.003
 basketball+0.002
P+0.002
 matters+0.002
Neg logit contributions
rant-0.003
sing-0.003
 enclosure-0.003
 horizon-0.002
asser-0.002
Token people
Token ID661
Feature activation-2.61e+00

Pos logit contributions
rant+0.015
 enclosure+0.014
 horizon+0.013
sing+0.013
 doorstep+0.013
Neg logit contributions
P-0.007
 band-0.007
 gum-0.006
 basketball-0.006
Ol-0.006
Token and
Token ID290
Feature activation-2.43e-01

Pos logit contributions
P+0.079
 band+0.074
 gum+0.073
 basketball+0.067
Ol+0.066
Neg logit contributions
rant-0.078
 enclosure-0.069
sing-0.066
 horizon-0.063
ples-0.060
Token always
Token ID1464
Feature activation-3.79e-01

Pos logit contributions
P+0.007
 band+0.007
 gum+0.006
 basketball+0.006
Ol+0.006
Neg logit contributions
rant-0.007
sing-0.007
 enclosure-0.007
 horizon-0.007
 doorstep-0.006
Token had
Token ID550
Feature activation+6.71e-01

Pos logit contributions
 band+0.010
P+0.010
 gum+0.010
 basketball+0.009
 matters+0.009
Neg logit contributions
rant-0.013
sing-0.011
 enclosure-0.011
 horizon-0.011
asser-0.010
Token a
Token ID257
Feature activation+8.64e-01

Pos logit contributions
rant+0.021
 enclosure+0.019
 horizon+0.018
sing+0.018
asser+0.017
Neg logit contributions
P-0.017
 band-0.017
Ol-0.015
 basketball-0.015
 gol-0.015
Token smile
Token ID8212
Feature activation+3.86e+00

Pos logit contributions
rant+0.027
sing+0.026
 enclosure+0.026
 horizon+0.024
 doorstep+0.024
Neg logit contributions
P-0.022
 band-0.022
 gum-0.020
 matters-0.020
 basketball-0.019
Token on
Token ID319
Feature activation+3.55e-01

Pos logit contributions
rant+0.154
 enclosure+0.138
sing+0.138
 horizon+0.134
 doorstep+0.127
Neg logit contributions
 band-0.080
P-0.075
 gum-0.070
 basketball-0.068
 matters-0.065
Token her
Token ID607
Feature activation-3.84e-01

Pos logit contributions
rant+0.014
 enclosure+0.013
 horizon+0.013
asser+0.012
 doorstep+0.012
Neg logit contributions
 band-0.007
P-0.006
 gum-0.005
 basketball-0.005
 gol-0.005
Token face
Token ID1986
Feature activation+8.03e-01

Pos logit contributions
P+0.012
 band+0.012
 gum+0.011
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.011
sing-0.011
 horizon-0.010
 enclosure-0.010
 doorstep-0.009
Token.
Token ID13
Feature activation+1.44e-01

Pos logit contributions
rant+0.026
 enclosure+0.023
sing+0.022
 horizon+0.022
 doorstep+0.021
Neg logit contributions
 band-0.023
P-0.022
 gum-0.021
 basketball-0.021
 matters-0.020
Token
Token ID198
Feature activation-2.26e-01

Pos logit contributions
rant+0.005
sing+0.005
 enclosure+0.005
 horizon+0.005
iously+0.004
Neg logit contributions
 band-0.003
P-0.003
 gum-0.003
 basketball-0.003
 matters-0.003
Token.
Token ID13
Feature activation-4.02e-01

Pos logit contributions
rant+0.039
 horizon+0.035
 enclosure+0.035
sing+0.035
 peak+0.032
Neg logit contributions
P-0.039
 band-0.039
 gum-0.036
Ol-0.034
 matters-0.034
Token He
Token ID679
Feature activation-4.09e-01

Pos logit contributions
 band+0.010
P+0.009
 gum+0.009
 basketball+0.009
 matters+0.008
Neg logit contributions
rant-0.014
 enclosure-0.013
sing-0.013
 horizon-0.013
 doorstep-0.012
Token had
Token ID550
Feature activation+6.04e-01

Pos logit contributions
P+0.012
 band+0.012
 gum+0.011
 basketball+0.011
 matters+0.010
Neg logit contributions
rant-0.013
 enclosure-0.011
sing-0.011
 horizon-0.011
 peak-0.010
Token a
Token ID257
Feature activation+4.80e-01

Pos logit contributions
rant+0.016
 enclosure+0.015
sing+0.014
 horizon+0.014
 definitely+0.013
Neg logit contributions
P-0.019
 band-0.018
 basketball-0.017
 matters-0.017
 gum-0.016
Token big
Token ID1263
Feature activation+6.46e-01

Pos logit contributions
sing+0.014
rant+0.013
 enclosure+0.013
 horizon+0.013
 peak+0.012
Neg logit contributions
P-0.013
 matters-0.013
 band-0.012
 gum-0.012
aid-0.012
Token grin
Token ID16770
Feature activation+3.86e+00

Pos logit contributions
rant+0.020
sing+0.020
 enclosure+0.019
 horizon+0.019
 peak+0.017
Neg logit contributions
P-0.017
 band-0.017
 matters-0.016
 gum-0.016
Ol-0.015
Token on
Token ID319
Feature activation+7.62e-01

Pos logit contributions
rant+0.140
 enclosure+0.129
 horizon+0.127
sing+0.126
iously+0.119
Neg logit contributions
P-0.084
 band-0.083
 matters-0.078
 gum-0.073
Ol-0.072
Token his
Token ID465
Feature activation-7.13e-01

Pos logit contributions
rant+0.036
 enclosure+0.033
 horizon+0.032
sing+0.032
iously+0.032
Neg logit contributions
 band-0.010
P-0.008
 gum-0.007
Ol-0.006
 basketball-0.006
Token face
Token ID1986
Feature activation+4.97e-01

Pos logit contributions
 band+0.021
P+0.020
 gum+0.019
 matters+0.019
 basketball+0.019
Neg logit contributions
rant-0.022
sing-0.021
 horizon-0.020
 enclosure-0.020
 doorstep-0.018
Token,
Token ID11
Feature activation+2.13e-01

Pos logit contributions
rant+0.015
 enclosure+0.013
 horizon+0.013
sing+0.013
iously+0.012
Neg logit contributions
 band-0.015
P-0.015
 matters-0.014
 gum-0.014
 basketball-0.013
Token adm
Token ID6178
Feature activation-1.96e-01

Pos logit contributions
rant+0.006
sing+0.006
 enclosure+0.005
 horizon+0.005
iously+0.005
Neg logit contributions
P-0.007
 band-0.007
 matters-0.006
 gum-0.006
 basketball-0.006
Token ended
Token ID4444
Feature activation-1.15e+00

Pos logit contributions
P+0.030
 band+0.028
Ol+0.027
aid+0.026
 gum+0.025
Neg logit contributions
rant-0.027
 enclosure-0.024
iously-0.023
sing-0.023
 horizon-0.023
Token up
Token ID510
Feature activation-1.47e-01

Pos logit contributions
P+0.035
 band+0.034
Ol+0.033
aid+0.033
 gum+0.030
Neg logit contributions
rant-0.032
sing-0.031
 enclosure-0.030
 horizon-0.029
iously-0.029
Token in
Token ID287
Feature activation-3.11e-01

Pos logit contributions
P+0.004
 band+0.004
Ol+0.004
 gum+0.004
aid+0.004
Neg logit contributions
rant-0.004
 enclosure-0.004
 horizon-0.004
sing-0.004
iously-0.004
Token such
Token ID884
Feature activation-4.46e-01

Pos logit contributions
P+0.007
 band+0.007
 gum+0.006
aid+0.006
Ol+0.006
Neg logit contributions
rant-0.011
 enclosure-0.011
sing-0.010
 horizon-0.010
 doorstep-0.010
Token bad
Token ID2089
Feature activation-2.15e-01

Pos logit contributions
 band+0.011
P+0.011
Ol+0.010
 gum+0.010
 basketball+0.010
Neg logit contributions
rant-0.014
sing-0.014
 enclosure-0.013
 horizon-0.013
iously-0.013
Token situations
Token ID7445
Feature activation-4.82e+00

Pos logit contributions
P+0.006
 band+0.006
Ol+0.005
 gum+0.005
 basketball+0.005
Neg logit contributions
rant-0.007
sing-0.006
 enclosure-0.006
 horizon-0.006
 doorstep-0.006
Token.
Token ID13
Feature activation+1.96e-01

Pos logit contributions
 band+0.144
P+0.141
 gum+0.134
 basketball+0.127
Ol+0.125
Neg logit contributions
rant-0.145
 enclosure-0.132
sing-0.128
 horizon-0.126
asser-0.117
Token
Token ID220
Feature activation+1.03e+00

Pos logit contributions
rant+0.006
 enclosure+0.006
sing+0.005
 horizon+0.005
 doorstep+0.005
Neg logit contributions
 band-0.006
P-0.005
 gum-0.005
 basketball-0.005
 matters-0.005
Token
Token ID198
Feature activation+2.68e-01

Pos logit contributions
rant+0.028
 enclosure+0.027
 horizon+0.025
sing+0.025
 doorstep+0.023
Neg logit contributions
 band-0.033
P-0.030
 gum-0.029
 basketball-0.028
 jacket-0.028
Token
Token ID198
Feature activation+4.74e-01

Pos logit contributions
rant+0.008
 enclosure+0.008
 horizon+0.007
sing+0.007
 doorstep+0.007
Neg logit contributions
 band-0.007
 gum-0.007
 basketball-0.006
P-0.006
 matters-0.006
TokenThe
Token ID464
Feature activation-2.19e-01

Pos logit contributions
rant+0.014
 enclosure+0.014
 horizon+0.013
sing+0.012
 doorstep+0.012
Neg logit contributions
 band-0.013
 gum-0.012
 matters-0.012
 basketball-0.012
 mar-0.012
Token him
Token ID683
Feature activation-3.60e-01

Pos logit contributions
P+0.015
 band+0.013
 gum+0.012
 basketball+0.012
Ol+0.012
Neg logit contributions
rant-0.020
 enclosure-0.018
 horizon-0.017
iously-0.017
 peak-0.016
Token how
Token ID703
Feature activation-6.66e-01

Pos logit contributions
P+0.012
 band+0.011
 gum+0.010
Ol+0.009
 gol+0.009
Neg logit contributions
 enclosure-0.010
rant-0.010
iously-0.008
sing-0.008
 horizon-0.008
Token to
Token ID284
Feature activation-5.06e-01

Pos logit contributions
P+0.023
 band+0.023
 gum+0.021
Ol+0.021
 jacket+0.021
Neg logit contributions
rant-0.012
 horizon-0.011
 previous-0.011
 enclosure-0.010
iously-0.010
Token think
Token ID892
Feature activation+9.46e-01

Pos logit contributions
P+0.016
 band+0.015
 gum+0.015
 basketball+0.014
 matters+0.014
Neg logit contributions
rant-0.015
 enclosure-0.013
sing-0.013
 horizon-0.012
 peak-0.012
Token of
Token ID286
Feature activation-3.98e-01

Pos logit contributions
rant+0.026
 horizon+0.023
 enclosure+0.021
sing+0.020
 peak+0.020
Neg logit contributions
P-0.033
 band-0.030
Ol-0.028
 basketball-0.027
 gum-0.027
Token situations
Token ID7445
Feature activation-4.19e+00

Pos logit contributions
P+0.010
 band+0.009
aid+0.008
 basketball+0.008
Ol+0.008
Neg logit contributions
rant-0.014
 enclosure-0.013
 horizon-0.013
sing-0.013
 peak-0.012
Token from
Token ID422
Feature activation-1.52e+00

Pos logit contributions
 band+0.127
P+0.126
 gum+0.115
 basketball+0.110
 matters+0.109
Neg logit contributions
rant-0.126
 enclosure-0.115
 horizon-0.109
sing-0.103
iously-0.100
Token different
Token ID1180
Feature activation-1.23e+00

Pos logit contributions
 band+0.034
P+0.033
 gum+0.031
 basketball+0.029
Ol+0.027
Neg logit contributions
rant-0.059
 enclosure-0.055
 horizon-0.054
sing-0.052
iously-0.049
Token perspectives
Token ID22582
Feature activation-1.06e+00

Pos logit contributions
 band+0.038
P+0.037
 gum+0.035
 basketball+0.033
aid+0.033
Neg logit contributions
rant-0.036
 enclosure-0.035
 horizon-0.034
iously-0.031
sing-0.030
Token,
Token ID11
Feature activation+4.81e-01

Pos logit contributions
 band+0.034
P+0.033
 gum+0.030
 basketball+0.030
Ol+0.029
Neg logit contributions
rant-0.032
 enclosure-0.028
 horizon-0.027
sing-0.025
iously-0.024
Token and
Token ID290
Feature activation-4.82e-01

Pos logit contributions
rant+0.013
sing+0.012
 enclosure+0.012
 horizon+0.012
iously+0.011
Neg logit contributions
P-0.016
 band-0.015
 gum-0.014
 basketball-0.013
 matters-0.013
Token parents
Token ID3397
Feature activation-8.68e-01

Pos logit contributions
P+0.013
Ol+0.011
 band+0.011
B+0.010
 basketball+0.010
Neg logit contributions
sing-0.013
 horizon-0.012
 doorstep-0.012
 enclosure-0.012
iously-0.011
Token could
Token ID714
Feature activation-5.31e-01

Pos logit contributions
P+0.026
 band+0.026
 gum+0.023
 basketball+0.023
Ol+0.022
Neg logit contributions
rant-0.027
 enclosure-0.024
sing-0.023
 horizon-0.023
 doorstep-0.022
Token lead
Token ID1085
Feature activation-6.70e-01

Pos logit contributions
P+0.016
 band+0.016
 gum+0.015
 basketball+0.014
 matters+0.014
Neg logit contributions
rant-0.015
sing-0.014
 horizon-0.014
 enclosure-0.014
 doorstep-0.013
Token to
Token ID284
Feature activation-1.11e+00

Pos logit contributions
 band+0.017
P+0.017
 gum+0.016
 basketball+0.016
aid+0.015
Neg logit contributions
rant-0.022
 enclosure-0.020
 horizon-0.019
sing-0.019
iously-0.018
Token dangerous
Token ID4923
Feature activation-2.92e-01

Pos logit contributions
P+0.034
 band+0.030
aid+0.029
 gum+0.028
Ol+0.027
Neg logit contributions
sing-0.030
rant-0.029
 enclosure-0.029
 horizon-0.028
 previous-0.028
Token situations
Token ID7445
Feature activation-4.14e+00

Pos logit contributions
 band+0.007
P+0.006
 gum+0.006
Ol+0.006
 basketball+0.005
Neg logit contributions
rant-0.011
 enclosure-0.010
 horizon-0.010
sing-0.010
iously-0.009
Token.
Token ID13
Feature activation-1.64e-01

Pos logit contributions
 band+0.123
P+0.122
 gum+0.114
 matters+0.110
 basketball+0.109
Neg logit contributions
rant-0.124
 enclosure-0.114
sing-0.109
 horizon-0.108
asser-0.102
Token From
Token ID3574
Feature activation+1.29e+00

Pos logit contributions
 band+0.005
 gum+0.004
P+0.004
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.005
sing-0.005
 enclosure-0.005
 horizon-0.005
 peak-0.004
Token that
Token ID326
Feature activation+3.38e-01

Pos logit contributions
rant+0.033
 horizon+0.029
 enclosure+0.028
sing+0.027
iously+0.025
Neg logit contributions
 band-0.044
 gum-0.042
P-0.041
 basketball-0.041
 matters-0.040
Token day
Token ID1110
Feature activation+6.17e-01

Pos logit contributions
rant+0.010
sing+0.009
 enclosure+0.009
 horizon+0.009
asser+0.008
Neg logit contributions
 band-0.010
 gum-0.009
P-0.009
 basketball-0.009
 mar-0.008
Token on
Token ID319
Feature activation-1.24e-01

Pos logit contributions
rant+0.027
sing+0.025
 enclosure+0.025
 horizon+0.024
asser+0.024
Neg logit contributions
 band-0.009
 gum-0.008
 basketball-0.008
P-0.007
 mar-0.007
Token also
Token ID635
Feature activation-7.27e-01

Pos logit contributions
 band+0.043
P+0.042
 gum+0.041
 basketball+0.041
 matters+0.038
Neg logit contributions
rant-0.042
 enclosure-0.037
sing-0.037
 horizon-0.037
 doorstep-0.033
Token help
Token ID1037
Feature activation-3.05e-01

Pos logit contributions
P+0.023
 band+0.023
 gum+0.022
 basketball+0.022
 mar+0.021
Neg logit contributions
rant-0.019
sing-0.018
 enclosure-0.017
 horizon-0.016
 doorstep-0.016
Token you
Token ID345
Feature activation-1.54e+00

Pos logit contributions
P+0.009
 band+0.008
aid+0.008
 basketball+0.008
 gol+0.008
Neg logit contributions
rant-0.009
 enclosure-0.008
 horizon-0.008
sing-0.008
 doorstep-0.007
Token in
Token ID287
Feature activation-4.47e-01

Pos logit contributions
 band+0.047
P+0.046
 gum+0.044
 basketball+0.042
 mar+0.040
Neg logit contributions
rant-0.045
 enclosure-0.042
sing-0.041
 horizon-0.040
 doorstep-0.038
Token tough
Token ID5802
Feature activation-1.06e+00

Pos logit contributions
 band+0.015
P+0.015
 gum+0.014
 basketball+0.014
aid+0.013
Neg logit contributions
rant-0.011
 enclosure-0.010
sing-0.010
 horizon-0.009
asser-0.009
Token situations
Token ID7445
Feature activation-4.02e+00

Pos logit contributions
 band+0.030
P+0.029
 gum+0.027
 basketball+0.027
aid+0.027
Neg logit contributions
rant-0.031
 enclosure-0.031
 horizon-0.029
sing-0.028
iously-0.028
Token.
Token ID13
Feature activation+1.92e-01

Pos logit contributions
 band+0.121
P+0.118
 gum+0.112
 basketball+0.108
 matters+0.104
Neg logit contributions
rant-0.121
 enclosure-0.112
 horizon-0.104
sing-0.104
asser-0.098
Token From
Token ID3574
Feature activation+1.19e+00

Pos logit contributions
rant+0.005
 enclosure+0.005
 horizon+0.005
sing+0.005
 doorstep+0.004
Neg logit contributions
 band-0.006
P-0.006
 gum-0.006
 basketball-0.005
 matters-0.005
Token that
Token ID326
Feature activation+9.67e-02

Pos logit contributions
rant+0.028
 horizon+0.024
 enclosure+0.023
sing+0.022
iously+0.020
Neg logit contributions
 band-0.044
P-0.042
 gum-0.041
 basketball-0.041
Ol-0.040
Token day
Token ID1110
Feature activation+4.34e-01

Pos logit contributions
rant+0.003
 enclosure+0.002
sing+0.002
 horizon+0.002
asser+0.002
Neg logit contributions
 band-0.003
P-0.003
 gum-0.003
 basketball-0.003
aid-0.003
Token on
Token ID319
Feature activation-2.47e-01

Pos logit contributions
rant+0.016
asser+0.014
 enclosure+0.014
sing+0.014
 horizon+0.013
Neg logit contributions
 band-0.009
 gum-0.008
 basketball-0.008
 mar-0.008
 baseball-0.007
Token could
Token ID714
Feature activation-1.14e-02

Pos logit contributions
rant+0.000
sing+0.000
 horizon+0.000
iously+0.000
 previous+0.000
Neg logit contributions
P-0.000
 band-0.000
aid-0.000
 gol-0.000
Ol-0.000
Token help
Token ID1037
Feature activation+2.70e-01

Pos logit contributions
P+0.000
 gum+0.000
 band+0.000
 basketball+0.000
 matters+0.000
Neg logit contributions
rant-0.000
sing-0.000
 enclosure-0.000
 peak-0.000
 horizon-0.000
Token him
Token ID683
Feature activation-4.57e-01

Pos logit contributions
rant+0.007
sing+0.007
 enclosure+0.007
 doorstep+0.006
 horizon+0.006
Neg logit contributions
P-0.008
aid-0.007
 band-0.007
 gum-0.006
 gol-0.006
Token in
Token ID287
Feature activation-2.25e-01

Pos logit contributions
P+0.014
 band+0.013
 gum+0.012
Ol+0.012
 matters+0.012
Neg logit contributions
rant-0.012
sing-0.012
 enclosure-0.011
 peak-0.011
 horizon-0.011
Token scary
Token ID14343
Feature activation-1.92e-01

Pos logit contributions
P+0.006
 band+0.006
aid+0.005
 gum+0.005
 basketball+0.005
Neg logit contributions
rant-0.007
 enclosure-0.007
sing-0.006
 horizon-0.006
 peak-0.006
Token situations
Token ID7445
Feature activation-4.01e+00

Pos logit contributions
 band+0.005
P+0.005
aid+0.005
 gum+0.005
Ol+0.005
Neg logit contributions
rant-0.005
 enclosure-0.005
 horizon-0.005
sing-0.005
 peak-0.005
Token.
Token ID13
Feature activation+2.72e-01

Pos logit contributions
 band+0.122
P+0.119
 gum+0.113
 basketball+0.108
 matters+0.107
Neg logit contributions
rant-0.119
 enclosure-0.109
 horizon-0.103
sing-0.103
asser-0.095
Token He
Token ID679
Feature activation+2.19e-02

Pos logit contributions
rant+0.010
 enclosure+0.009
sing+0.009
 horizon+0.009
asser+0.008
Neg logit contributions
 band-0.006
P-0.006
 gum-0.006
 basketball-0.006
 matters-0.005
Token continued
Token ID3767
Feature activation+1.39e-01

Pos logit contributions
rant+0.001
 enclosure+0.001
sing+0.001
 horizon+0.001
 doorstep+0.001
Neg logit contributions
 band-0.001
 gum-0.001
P-0.001
 basketball-0.001
 matters-0.001
Token to
Token ID284
Feature activation-2.75e-01

Pos logit contributions
rant+0.004
 enclosure+0.004
sing+0.004
 horizon+0.003
asser+0.003
Neg logit contributions
 band-0.004
P-0.004
 gum-0.004
aid-0.004
 matters-0.004
Token balance
Token ID5236
Feature activation-9.56e-01

Pos logit contributions
P+0.009
 gum+0.009
 band+0.009
aid+0.008
 basketball+0.008
Neg logit contributions
 enclosure-0.006
rant-0.006
sing-0.006
 peak-0.006
 horizon-0.006
Token out
Token ID503
Feature activation-7.22e-01

Pos logit contributions
 band+0.047
P+0.046
 gum+0.043
 basketball+0.040
 matters+0.040
Neg logit contributions
rant-0.049
sing-0.045
 horizon-0.044
 enclosure-0.044
 doorstep-0.041
Token of
Token ID286
Feature activation-1.37e+00

Pos logit contributions
P+0.020
 band+0.020
 gum+0.019
 basketball+0.018
Ol+0.017
Neg logit contributions
rant-0.023
 enclosure-0.021
sing-0.020
 horizon-0.020
iously-0.019
Token difficult
Token ID2408
Feature activation-5.86e-01

Pos logit contributions
P+0.034
 band+0.032
aid+0.030
Ol+0.029
 gum+0.027
Neg logit contributions
rant-0.046
 horizon-0.042
 enclosure-0.042
sing-0.041
 previous-0.039
Token and
Token ID290
Feature activation-3.25e-01

Pos logit contributions
P+0.017
 band+0.017
aid+0.015
Ol+0.015
 gum+0.014
Neg logit contributions
rant-0.017
 horizon-0.017
 enclosure-0.016
iously-0.016
sing-0.016
Token sticky
Token ID23408
Feature activation+2.11e-01

Pos logit contributions
 band+0.009
P+0.009
 gum+0.009
 basketball+0.008
Ol+0.008
Neg logit contributions
rant-0.010
sing-0.009
 horizon-0.009
 enclosure-0.009
iously-0.008
Token situations
Token ID7445
Feature activation-4.01e+00

Pos logit contributions
rant+0.007
iously+0.006
 horizon+0.006
sing+0.006
 enclosure+0.006
Neg logit contributions
P-0.006
 band-0.005
Ol-0.005
J-0.005
aid-0.005
Token.
Token ID13
Feature activation-1.01e-01

Pos logit contributions
P+0.123
 band+0.122
 gum+0.112
 basketball+0.109
 matters+0.106
Neg logit contributions
rant-0.119
 enclosure-0.105
 horizon-0.101
sing-0.100
asser-0.094
Token<|endoftext|>
Token ID50256
Feature activation-3.04e-01

Pos logit contributions
 band+0.002
P+0.002
 gum+0.002
 basketball+0.002
 matters+0.002
Neg logit contributions
rant-0.004
 enclosure-0.003
sing-0.003
 horizon-0.003
asser-0.003
TokenThe
Token ID464
Feature activation-4.65e-01

Pos logit contributions
 band+0.009
 gum+0.008
P+0.008
aid+0.008
 basketball+0.008
Neg logit contributions
rant-0.009
 enclosure-0.008
 horizon-0.008
asser-0.007
sing-0.007
Token family
Token ID1641
Feature activation+5.17e-02

Pos logit contributions
P+0.013
 band+0.013
 matters+0.013
aid+0.012
Ol+0.012
Neg logit contributions
 horizon-0.013
rant-0.012
 enclosure-0.012
sing-0.012
 doorstep-0.011
Token had
Token ID550
Feature activation+3.79e-01

Pos logit contributions
rant+0.002
 enclosure+0.002
 horizon+0.001
sing+0.001
 doorstep+0.001
Neg logit contributions
 band-0.001
 gum-0.001
P-0.001
 basketball-0.001
 matters-0.001
Token to
Token ID284
Feature activation-8.19e-01

Pos logit contributions
rant+0.001
 enclosure+0.001
sing+0.001
 doorstep+0.001
 horizon+0.001
Neg logit contributions
P-0.001
 band-0.001
 gum-0.001
 basketball-0.001
 mar-0.001
Token stay
Token ID2652
Feature activation-4.52e-01

Pos logit contributions
P+0.029
 gum+0.028
 band+0.027
 basketball+0.027
 matters+0.026
Neg logit contributions
rant-0.019
 enclosure-0.018
sing-0.018
 doorstep-0.018
 horizon-0.017
Token calm
Token ID9480
Feature activation+5.38e-02

Pos logit contributions
 gum+0.012
 band+0.012
P+0.011
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.015
 enclosure-0.013
sing-0.013
 horizon-0.012
asser-0.012
Token in
Token ID287
Feature activation+4.92e-01

Pos logit contributions
rant+0.002
 horizon+0.001
 enclosure+0.001
sing+0.001
iously+0.001
Neg logit contributions
 band-0.002
P-0.002
 gum-0.002
 basketball-0.002
 matters-0.002
Token scary
Token ID14343
Feature activation+3.79e-03

Pos logit contributions
rant+0.021
 enclosure+0.019
sing+0.019
 horizon+0.019
 doorstep+0.018
Neg logit contributions
 band-0.008
P-0.008
 gum-0.008
 basketball-0.007
 matters-0.007
Token situations
Token ID7445
Feature activation-3.88e+00

Pos logit contributions
rant+0.000
 enclosure+0.000
 horizon+0.000
sing+0.000
 doorstep+0.000
Neg logit contributions
 band-0.000
P-0.000
 gum-0.000
 basketball-0.000
Ol-0.000
Token.
Token ID13
Feature activation-4.66e-01

Pos logit contributions
 band+0.107
P+0.101
 gum+0.098
 basketball+0.095
 matters+0.093
Neg logit contributions
rant-0.129
 enclosure-0.118
sing-0.116
 horizon-0.113
asser-0.106
Token
Token ID220
Feature activation+5.38e-01

Pos logit contributions
 band+0.013
P+0.012
 gum+0.012
 basketball+0.012
 matters+0.011
Neg logit contributions
rant-0.016
 enclosure-0.014
sing-0.014
 horizon-0.013
asser-0.012
Token
Token ID198
Feature activation+1.65e-01

Pos logit contributions
rant+0.017
 enclosure+0.016
sing+0.015
 horizon+0.015
asser+0.014
Neg logit contributions
 band-0.015
P-0.014
 gum-0.014
 basketball-0.013
 mar-0.013
Token
Token ID198
Feature activation+5.97e-01

Pos logit contributions
rant+0.005
 enclosure+0.005
sing+0.005
 horizon+0.005
asser+0.005
Neg logit contributions
 band-0.004
 gum-0.004
P-0.004
 basketball-0.004
 matters-0.004
TokenSo
Token ID2396
Feature activation+4.15e-01

Pos logit contributions
rant+0.018
 enclosure+0.017
 doorstep+0.016
sing+0.016
 horizon+0.016
Neg logit contributions
 band-0.015
 matters-0.015
 gum-0.014
 mar-0.013
 penn-0.013
Token lately
Token ID16537
Feature activation-1.21e+00

Pos logit contributions
P+0.006
 band+0.006
Ol+0.005
 gum+0.005
 matters+0.005
Neg logit contributions
rant-0.005
sing-0.004
 enclosure-0.004
 horizon-0.004
iously-0.004
Token.
Token ID13
Feature activation-2.26e-01

Pos logit contributions
 band+0.035
P+0.035
 matters+0.032
 gum+0.032
 basketball+0.032
Neg logit contributions
rant-0.037
 enclosure-0.033
sing-0.033
 horizon-0.033
iously-0.030
Token With
Token ID2080
Feature activation-6.80e-01

Pos logit contributions
 band+0.005
 his+0.005
 matters+0.005
 basketball+0.004
 gum+0.004
Neg logit contributions
rant-0.008
 enclosure-0.007
sing-0.007
asser-0.007
 horizon-0.007
Token work
Token ID670
Feature activation-1.58e+00

Pos logit contributions
 band+0.015
P+0.014
 basketball+0.014
 gum+0.013
 matters+0.012
Neg logit contributions
rant-0.026
sing-0.023
 enclosure-0.022
 horizon-0.022
iously-0.021
Token and
Token ID290
Feature activation+1.90e-01

Pos logit contributions
P+0.048
 basketball+0.044
 gum+0.043
Ol+0.041
 band+0.041
Neg logit contributions
rant-0.044
 horizon-0.038
iously-0.037
bell-0.037
sing-0.036
Token err
Token ID11454
Feature activation-3.87e+00

Pos logit contributions
rant+0.006
 horizon+0.005
sing+0.005
iously+0.005
 enclosure+0.005
Neg logit contributions
P-0.006
 band-0.005
 basketball-0.005
 gum-0.005
Ol-0.005
Tokenands
Token ID1746
Feature activation-1.36e+00

Pos logit contributions
 band+0.122
 gum+0.111
P+0.110
 basketball+0.107
 matters+0.106
Neg logit contributions
rant-0.115
sing-0.107
 enclosure-0.105
 horizon-0.104
 doorstep-0.094
Token,
Token ID11
Feature activation-3.92e-01

Pos logit contributions
P+0.041
 gum+0.039
 basketball+0.038
 band+0.035
 baseball+0.035
Neg logit contributions
rant-0.039
iously-0.034
 horizon-0.034
sing-0.033
 enclosure-0.033
Token they
Token ID484
Feature activation-2.35e-01

Pos logit contributions
 band+0.012
P+0.012
 gum+0.012
 basketball+0.011
 hockey+0.011
Neg logit contributions
rant-0.011
 horizon-0.010
sing-0.010
 enclosure-0.009
iously-0.009
Token didn
Token ID1422
Feature activation+5.36e-01

Pos logit contributions
 gum+0.007
P+0.007
 band+0.007
 basketball+0.007
 matters+0.006
Neg logit contributions
rant-0.007
sing-0.006
 enclosure-0.006
 horizon-0.006
 doorstep-0.006
Token't
Token ID470
Feature activation+6.83e-01

Pos logit contributions
rant+0.013
 enclosure+0.011
sing+0.011
 horizon+0.011
 doorstep+0.010
Neg logit contributions
 band-0.020
P-0.019
 gum-0.018
 basketball-0.018
 matters-0.017
Token cr
Token ID1067
Feature activation-1.81e+00

Pos logit contributions
 band+0.024
 gum+0.022
P+0.022
 basketball+0.021
 matters+0.020
Neg logit contributions
rant-0.031
 enclosure-0.028
sing-0.027
 horizon-0.027
asser-0.025
Tokenay
Token ID323
Feature activation-7.30e-02

Pos logit contributions
 band+0.051
P+0.048
 gum+0.047
 basketball+0.044
 matters+0.041
Neg logit contributions
rant-0.059
 enclosure-0.056
sing-0.055
 horizon-0.053
iously-0.049
Tokenons
Token ID684
Feature activation-1.17e+00

Pos logit contributions
 band+0.001
P+0.001
 basketball+0.001
 gum+0.001
 matters+0.001
Neg logit contributions
rant-0.003
 enclosure-0.002
sing-0.002
asser-0.002
 horizon-0.002
Token.
Token ID13
Feature activation-1.65e-01

Pos logit contributions
P+0.039
 band+0.037
 gum+0.035
Ol+0.035
 matters+0.034
Neg logit contributions
rant-0.031
 enclosure-0.029
sing-0.028
 horizon-0.026
asser-0.024
Token It
Token ID632
Feature activation-6.46e-01

Pos logit contributions
 band+0.006
P+0.006
 gum+0.006
 jacket+0.005
 basketball+0.005
Neg logit contributions
rant-0.004
sing-0.003
 enclosure-0.003
 horizon-0.003
asser-0.003
Token er
Token ID1931
Feature activation-3.82e+00

Pos logit contributions
 band+0.019
 gum+0.016
P+0.016
 basketball+0.016
 matters+0.016
Neg logit contributions
rant-0.021
 horizon-0.019
sing-0.019
 enclosure-0.019
 doorstep-0.018
Tokenases
Token ID1386
Feature activation-6.16e-01

Pos logit contributions
 band+0.096
P+0.087
 gum+0.086
 basketball+0.084
 matters+0.081
Neg logit contributions
rant-0.136
 enclosure-0.121
sing-0.120
 horizon-0.116
asser-0.113
Token their
Token ID511
Feature activation-9.30e-01

Pos logit contributions
P+0.017
 band+0.015
 gum+0.015
 matters+0.014
 basketball+0.014
Neg logit contributions
rant-0.020
 enclosure-0.018
 horizon-0.017
sing-0.016
 entrance-0.016
Token animals
Token ID4695
Feature activation-1.35e+00

Pos logit contributions
 band+0.024
 gum+0.021
P+0.021
 basketball+0.021
 matters+0.020
Neg logit contributions
rant-0.034
 enclosure-0.030
sing-0.030
 horizon-0.029
asser-0.028
Token and
Token ID290
Feature activation-2.99e-02

Pos logit contributions
P+0.041
 band+0.040
 gum+0.038
 matters+0.036
Ol+0.036
Neg logit contributions
rant-0.038
 enclosure-0.038
sing-0.036
 horizon-0.033
 doorstep-0.032
Token flowers
Token ID12734
Feature activation-9.35e-01

Pos logit contributions
P+0.001
 band+0.001
 gum+0.001
 basketball+0.001
Ol+0.001
Neg logit contributions
rant-0.001
 enclosure-0.001
sing-0.001
 horizon-0.001
 peak-0.001
Token angry
Token ID7954
Feature activation-2.62e-01

Pos logit contributions
rant+0.009
 horizon+0.008
 enclosure+0.008
sing+0.007
asser+0.007
Neg logit contributions
 band-0.011
P-0.011
Ol-0.010
 basketball-0.010
 gum-0.010
Token".
Token ID1911
Feature activation-5.13e-01

Pos logit contributions
 band+0.007
P+0.007
 gum+0.007
 matters+0.007
 basketball+0.007
Neg logit contributions
rant-0.008
sing-0.007
 enclosure-0.007
 horizon-0.007
iously-0.007
Token
Token ID198
Feature activation-5.91e-01

Pos logit contributions
 band+0.015
P+0.014
 gum+0.014
 matters+0.014
aid+0.013
Neg logit contributions
 enclosure-0.014
rant-0.013
sing-0.013
 horizon-0.012
 doorstep-0.011
TokenThe
Token ID464
Feature activation-8.01e-01

Pos logit contributions
 band+0.018
 gum+0.017
 mar+0.016
 basketball+0.016
 jacket+0.016
Neg logit contributions
rant-0.016
 enclosure-0.015
 horizon-0.014
sing-0.013
anne-0.013
Token 3
Token ID513
Feature activation-3.54e-01

Pos logit contributions
 matters+0.021
P+0.021
 band+0.020
 gum+0.018
Ol+0.018
Neg logit contributions
rant-0.024
sing-0.024
 horizon-0.024
 enclosure-0.024
 doorstep-0.021
Token-
Token ID12
Feature activation-3.77e+00

Pos logit contributions
 band+0.013
P+0.012
 gum+0.011
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.009
sing-0.008
 enclosure-0.008
 horizon-0.008
 doorstep-0.007
Tokenyear
Token ID1941
Feature activation-1.13e+00

Pos logit contributions
 band+0.141
 gum+0.138
 basketball+0.137
 matters+0.131
P+0.130
Neg logit contributions
rant-0.084
 enclosure-0.070
sing-0.068
iously-0.062
 horizon-0.060
Token-
Token ID12
Feature activation-3.64e+00

Pos logit contributions
 band+0.032
P+0.030
 gum+0.030
 basketball+0.029
 matters+0.028
Neg logit contributions
rant-0.036
 enclosure-0.033
sing-0.032
 horizon-0.032
iously-0.030
Tokenold
Token ID727
Feature activation-2.48e+00

Pos logit contributions
 band+0.106
 gum+0.102
 basketball+0.101
 matters+0.097
P+0.096
Neg logit contributions
rant-0.110
 enclosure-0.099
sing-0.095
iously-0.091
 horizon-0.090
Token thought
Token ID1807
Feature activation+2.19e-02

Pos logit contributions
P+0.066
 band+0.065
 gum+0.065
 basketball+0.062
 matters+0.061
Neg logit contributions
rant-0.078
 enclosure-0.074
sing-0.069
iously-0.069
 horizon-0.067
Token hard
Token ID1327
Feature activation-3.89e-01

Pos logit contributions
rant+0.001
iously+0.001
sing+0.001
 horizon+0.001
 enclosure+0.000
Neg logit contributions
P-0.001
 band-0.001
 basketball-0.001
 gum-0.001
 mar-0.001
Tokenases
Token ID1386
Feature activation-5.57e-01

Pos logit contributions
 band+0.094
P+0.084
 gum+0.084
 basketball+0.083
 matters+0.079
Neg logit contributions
rant-0.129
 enclosure-0.114
sing-0.113
 horizon-0.109
asser-0.108
Token their
Token ID511
Feature activation-8.84e-01

Pos logit contributions
P+0.015
 band+0.013
 gum+0.012
 matters+0.012
Ol+0.012
Neg logit contributions
rant-0.019
 enclosure-0.017
 horizon-0.016
sing-0.015
 entrance-0.015
Token drawings
Token ID23388
Feature activation-1.37e+00

Pos logit contributions
 band+0.026
 gum+0.024
P+0.023
 basketball+0.023
 matters+0.022
Neg logit contributions
rant-0.029
 enclosure-0.025
sing-0.025
 horizon-0.025
asser-0.023
Token.
Token ID13
Feature activation-1.50e-01

Pos logit contributions
P+0.044
 band+0.043
 gum+0.041
Ol+0.040
 matters+0.039
Neg logit contributions
rant-0.038
 enclosure-0.034
sing-0.033
 horizon-0.033
 entrance-0.030
Token It
Token ID632
Feature activation-5.14e-01

Pos logit contributions
 band+0.006
P+0.005
 gum+0.005
 jacket+0.005
 basketball+0.005
Neg logit contributions
rant-0.003
sing-0.003
 enclosure-0.003
 horizon-0.003
asser-0.002
Token er
Token ID1931
Feature activation-3.71e+00

Pos logit contributions
 band+0.015
 gum+0.013
P+0.013
 basketball+0.013
 matters+0.013
Neg logit contributions
rant-0.017
sing-0.015
 horizon-0.015
 enclosure-0.015
 doorstep-0.014
Tokenases
Token ID1386
Feature activation-5.74e-01

Pos logit contributions
 band+0.093
P+0.084
 gum+0.083
 basketball+0.082
 matters+0.078
Neg logit contributions
rant-0.132
 enclosure-0.117
sing-0.116
 horizon-0.112
asser-0.110
Token their
Token ID511
Feature activation-8.92e-01

Pos logit contributions
P+0.016
 band+0.014
 gum+0.014
 matters+0.014
Ol+0.013
Neg logit contributions
rant-0.019
 enclosure-0.016
 horizon-0.016
sing-0.015
 entrance-0.015
Token cr
Token ID1067
Feature activation-1.81e+00

Pos logit contributions
 band+0.024
 gum+0.022
P+0.022
 basketball+0.021
 matters+0.020
Neg logit contributions
rant-0.031
 enclosure-0.028
sing-0.027
 horizon-0.027
asser-0.025
Tokenay
Token ID323
Feature activation-7.30e-02

Pos logit contributions
 band+0.051
P+0.048
 gum+0.047
 basketball+0.044
 matters+0.041
Neg logit contributions
rant-0.059
 enclosure-0.056
sing-0.055
 horizon-0.053
iously-0.049
Tokenons
Token ID684
Feature activation-1.17e+00

Pos logit contributions
 band+0.001
P+0.001
 basketball+0.001
 gum+0.001
 matters+0.001
Neg logit contributions
rant-0.003
 enclosure-0.002
sing-0.002
asser-0.002
 horizon-0.002
Token liked
Token ID8288
Feature activation+2.57e-01

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.011
 horizon+0.011
 doorstep+0.011
Neg logit contributions
 band-0.010
P-0.009
 gum-0.009
 basketball-0.009
 matters-0.009
Token to
Token ID284
Feature activation-1.10e+00

Pos logit contributions
rant+0.007
 enclosure+0.007
sing+0.006
 horizon+0.006
asser+0.006
Neg logit contributions
P-0.008
 band-0.008
 matters-0.007
 gum-0.007
aid-0.007
Token tie
Token ID9839
Feature activation-3.74e-01

Pos logit contributions
 band+0.030
P+0.028
 gum+0.027
 basketball+0.026
 matters+0.025
Neg logit contributions
rant-0.037
sing-0.033
 enclosure-0.033
 horizon-0.032
asser-0.030
Token yarn
Token ID21181
Feature activation-6.04e-01

Pos logit contributions
P+0.009
 band+0.008
 gol+0.008
aid+0.007
 jacket+0.007
Neg logit contributions
rant-0.013
 enclosure-0.011
 horizon-0.011
ples-0.011
sing-0.010
Token around
Token ID1088
Feature activation+8.70e-02

Pos logit contributions
P+0.018
 band+0.018
 matters+0.016
aid+0.016
Ol+0.015
Neg logit contributions
rant-0.017
 enclosure-0.016
 horizon-0.014
sing-0.014
asser-0.014
Token things
Token ID1243
Feature activation-3.69e+00

Pos logit contributions
rant+0.004
 enclosure+0.003
 horizon+0.003
sing+0.003
ples+0.003
Neg logit contributions
P-0.002
 band-0.002
aid-0.001
 matters-0.001
 gol-0.001
Token,
Token ID11
Feature activation+7.38e-01

Pos logit contributions
 band+0.104
P+0.099
 gum+0.093
 basketball+0.090
 matters+0.088
Neg logit contributions
rant-0.121
 enclosure-0.107
sing-0.103
 horizon-0.103
asser-0.097
Token like
Token ID588
Feature activation-5.91e-01

Pos logit contributions
rant+0.019
sing+0.019
 enclosure+0.017
 horizon+0.016
 peak+0.014
Neg logit contributions
P-0.025
 band-0.024
 matters-0.022
 gum-0.021
Ol-0.021
Token his
Token ID465
Feature activation-5.97e-01

Pos logit contributions
P+0.017
 matters+0.015
 band+0.014
aid+0.014
Ol+0.013
Neg logit contributions
rant-0.016
 enclosure-0.016
sing-0.015
ples-0.015
 horizon-0.014
Token toys
Token ID14958
Feature activation-1.48e+00

Pos logit contributions
P+0.015
 band+0.013
aid+0.012
 matters+0.012
Ol+0.012
Neg logit contributions
sing-0.021
rant-0.019
 enclosure-0.019
 horizon-0.019
 doorstep-0.018
Token and
Token ID290
Feature activation+5.09e-01

Pos logit contributions
P+0.045
 band+0.043
 matters+0.040
Ol+0.040
 gum+0.039
Neg logit contributions
rant-0.042
 enclosure-0.037
 horizon-0.036
sing-0.035
asser-0.034
Token on
Token ID319
Feature activation+1.17e-02

Pos logit contributions
 band+0.027
P+0.026
 gum+0.024
 matters+0.024
 basketball+0.023
Neg logit contributions
rant-0.055
sing-0.049
 enclosure-0.048
 horizon-0.047
asser-0.047
Token their
Token ID511
Feature activation-5.97e-01

Pos logit contributions
rant+0.000
 horizon+0.000
 enclosure+0.000
sing+0.000
 peak+0.000
Neg logit contributions
 band-0.000
P-0.000
 matters-0.000
 gum-0.000
aid-0.000
Token papers
Token ID9473
Feature activation-1.86e+00

Pos logit contributions
 band+0.018
 gum+0.016
 basketball+0.015
 mar+0.015
P+0.015
Neg logit contributions
rant-0.021
 enclosure-0.017
 horizon-0.017
sing-0.016
asser-0.016
Token.
Token ID13
Feature activation-1.17e-01

Pos logit contributions
 band+0.053
P+0.050
 gum+0.048
 basketball+0.047
 matters+0.045
Neg logit contributions
rant-0.060
 enclosure-0.054
sing-0.053
 horizon-0.053
 doorstep-0.049
Token It
Token ID632
Feature activation-4.69e-01

Pos logit contributions
 band+0.004
P+0.004
 gum+0.004
 jacket+0.004
 matters+0.004
Neg logit contributions
rant-0.003
sing-0.002
 enclosure-0.002
 horizon-0.002
asser-0.002
Token er
Token ID1931
Feature activation-3.68e+00

Pos logit contributions
 band+0.014
 gum+0.012
P+0.012
 matters+0.012
 basketball+0.012
Neg logit contributions
rant-0.015
sing-0.014
 horizon-0.014
 enclosure-0.013
 doorstep-0.013
Tokenases
Token ID1386
Feature activation-5.57e-01

Pos logit contributions
 band+0.094
P+0.084
 gum+0.084
 basketball+0.083
 matters+0.079
Neg logit contributions
rant-0.129
 enclosure-0.114
sing-0.113
 horizon-0.109
asser-0.108
Token their
Token ID511
Feature activation-8.84e-01

Pos logit contributions
P+0.015
 band+0.013
 gum+0.012
 matters+0.012
Ol+0.012
Neg logit contributions
rant-0.019
 enclosure-0.017
 horizon-0.016
sing-0.015
 entrance-0.015
Token drawings
Token ID23388
Feature activation-1.37e+00

Pos logit contributions
 band+0.026
 gum+0.024
P+0.023
 basketball+0.023
 matters+0.022
Neg logit contributions
rant-0.029
 enclosure-0.025
sing-0.025
 horizon-0.025
asser-0.023
Token.
Token ID13
Feature activation-1.50e-01

Pos logit contributions
P+0.044
 band+0.043
 gum+0.041
Ol+0.040
 matters+0.039
Neg logit contributions
rant-0.038
 enclosure-0.034
sing-0.033
 horizon-0.033
 entrance-0.030
Token It
Token ID632
Feature activation-5.14e-01

Pos logit contributions
 band+0.006
P+0.005
 gum+0.005
 jacket+0.005
 basketball+0.005
Neg logit contributions
rant-0.003
sing-0.003
 enclosure-0.003
 horizon-0.003
asser-0.002
Token was
Token ID373
Feature activation+8.49e-01

Pos logit contributions
rant+0.001
 enclosure+0.001
sing+0.001
 horizon+0.001
 doorstep+0.001
Neg logit contributions
 band-0.001
P-0.001
 gum-0.001
 basketball-0.001
 matters-0.001
Token important
Token ID1593
Feature activation-1.27e+00

Pos logit contributions
rant+0.028
sing+0.024
 enclosure+0.023
 horizon+0.023
asser+0.022
Neg logit contributions
 band-0.024
P-0.023
 gum-0.022
 basketball-0.021
Ol-0.021
Token to
Token ID284
Feature activation-7.06e-01

Pos logit contributions
 band+0.036
P+0.034
 gum+0.032
 basketball+0.032
 matters+0.031
Neg logit contributions
rant-0.041
 enclosure-0.038
sing-0.037
 horizon-0.036
asser-0.034
Token value
Token ID1988
Feature activation+6.51e-02

Pos logit contributions
 band+0.023
P+0.022
 gum+0.022
 basketball+0.022
 matters+0.020
Neg logit contributions
rant-0.020
 enclosure-0.017
sing-0.017
 horizon-0.016
 doorstep-0.015
Token all
Token ID477
Feature activation-3.35e-01

Pos logit contributions
rant+0.002
 horizon+0.002
 enclosure+0.002
sing+0.002
iously+0.002
Neg logit contributions
P-0.002
 band-0.002
Ol-0.001
 basketball-0.001
 gol-0.001
Token living
Token ID2877
Feature activation-3.65e+00

Pos logit contributions
 band+0.006
P+0.006
 gum+0.006
 basketball+0.005
Ol+0.005
Neg logit contributions
rant-0.014
 enclosure-0.013
 horizon-0.012
sing-0.012
 doorstep-0.012
Token creatures
Token ID8109
Feature activation-2.58e+00

Pos logit contributions
 band+0.079
P+0.071
 gum+0.069
 basketball+0.066
Ol+0.062
Neg logit contributions
rant-0.145
 enclosure-0.129
sing-0.125
 horizon-0.125
asser-0.122
Token.
Token ID13
Feature activation+2.40e-02

Pos logit contributions
 band+0.077
P+0.074
 gum+0.070
Ol+0.067
 basketball+0.066
Neg logit contributions
rant-0.078
 enclosure-0.070
 horizon-0.067
sing-0.065
asser-0.062
Token<|endoftext|>
Token ID50256
Feature activation-2.59e-01

Pos logit contributions
rant+0.001
sing+0.001
 enclosure+0.001
 horizon+0.001
iously+0.001
Neg logit contributions
 band-0.000
 gum-0.000
P-0.000
 basketball-0.000
 matters-0.000
TokenOnce
Token ID7454
Feature activation-1.70e-01

Pos logit contributions
 band+0.006
 gum+0.006
P+0.006
 basketball+0.005
 matters+0.005
Neg logit contributions
rant-0.009
 enclosure-0.009
 horizon-0.008
sing-0.008
asser-0.008
Token upon
Token ID2402
Feature activation-3.13e-01

Pos logit contributions
 band+0.005
P+0.005
 gum+0.005
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.005
 enclosure-0.005
sing-0.005
 horizon-0.005
asser-0.004
Token be
Token ID307
Feature activation+9.51e-02

Pos logit contributions
rant+0.007
sing+0.006
 horizon+0.006
 enclosure+0.006
 doorstep+0.006
Neg logit contributions
P-0.007
 band-0.007
 gum-0.007
 basketball-0.006
Ol-0.006
Token happy
Token ID3772
Feature activation+8.97e-02

Pos logit contributions
rant+0.003
sing+0.003
 enclosure+0.003
ples+0.003
 horizon+0.003
Neg logit contributions
 band-0.002
 gum-0.002
 basketball-0.002
P-0.002
 mar-0.002
Token.
Token ID13
Feature activation-2.18e-01

Pos logit contributions
rant+0.003
sing+0.002
 horizon+0.002
 enclosure+0.002
iously+0.002
Neg logit contributions
 band-0.003
P-0.003
 gum-0.002
 basketball-0.002
Ol-0.002
Token Even
Token ID3412
Feature activation-1.33e+00

Pos logit contributions
 band+0.006
 gum+0.006
P+0.006
 basketball+0.006
 matters+0.006
Neg logit contributions
rant-0.007
sing-0.006
 enclosure-0.006
 horizon-0.006
asser-0.006
Token though
Token ID996
Feature activation-1.01e+00

Pos logit contributions
 band+0.040
P+0.040
Ol+0.038
 gum+0.037
 basketball+0.036
Neg logit contributions
rant-0.038
 horizon-0.031
iously-0.028
sing-0.028
ples-0.028
Token situations
Token ID7445
Feature activation-3.65e+00

Pos logit contributions
P+0.026
 band+0.024
aid+0.023
Ol+0.022
 basketball+0.021
Neg logit contributions
 horizon-0.033
rant-0.032
 enclosure-0.032
sing-0.031
 peak-0.031
Token can
Token ID460
Feature activation-1.33e+00

Pos logit contributions
P+0.122
 band+0.112
 gum+0.106
 basketball+0.105
Ol+0.102
Neg logit contributions
rant-0.103
 horizon-0.090
 enclosure-0.090
sing-0.086
iously-0.080
Token be
Token ID307
Feature activation-5.99e-02

Pos logit contributions
 band+0.040
P+0.038
 gum+0.038
 basketball+0.037
 matters+0.035
Neg logit contributions
rant-0.040
 enclosure-0.036
sing-0.036
 horizon-0.036
 doorstep-0.032
Token serious
Token ID2726
Feature activation-2.93e-01

Pos logit contributions
P+0.002
 band+0.002
 gum+0.002
 basketball+0.002
Ol+0.002
Neg logit contributions
rant-0.002
sing-0.002
 enclosure-0.002
 horizon-0.002
ples-0.001
Token,
Token ID11
Feature activation+1.92e-01

Pos logit contributions
P+0.009
 band+0.008
 gum+0.007
Ol+0.007
aid+0.007
Neg logit contributions
rant-0.009
 horizon-0.008
 enclosure-0.008
sing-0.008
iously-0.008
Token it
Token ID340
Feature activation-5.49e-01

Pos logit contributions
 horizon+0.006
rant+0.006
 previous+0.005
 peak+0.005
sing+0.005
Neg logit contributions
P-0.005
 band-0.005
Ol-0.004
aid-0.004
 basketball-0.004
Token tried
Token ID3088
Feature activation+1.57e-01

Pos logit contributions
rant+0.028
 enclosure+0.025
sing+0.025
 horizon+0.024
asser+0.023
Neg logit contributions
 band-0.019
P-0.017
 gum-0.017
 basketball-0.017
 matters-0.016
Token twisting
Token ID36373
Feature activation-2.17e-01

Pos logit contributions
rant+0.004
 horizon+0.004
sing+0.004
 enclosure+0.004
 peak+0.004
Neg logit contributions
P-0.005
 band-0.005
aid-0.004
 matters-0.004
 gum-0.004
Token it
Token ID340
Feature activation-5.70e-01

Pos logit contributions
P+0.006
 band+0.006
aid+0.005
Ol+0.005
 matters+0.005
Neg logit contributions
rant-0.007
 horizon-0.006
 peak-0.006
 enclosure-0.006
sing-0.006
Token around
Token ID1088
Feature activation-1.58e-01

Pos logit contributions
 band+0.017
P+0.016
 gum+0.015
 basketball+0.014
 matters+0.014
Neg logit contributions
rant-0.018
sing-0.016
 horizon-0.015
 enclosure-0.015
iously-0.015
Token other
Token ID584
Feature activation-4.59e-01

Pos logit contributions
P+0.004
 band+0.004
aid+0.003
 matters+0.003
 gum+0.003
Neg logit contributions
rant-0.006
sing-0.005
 horizon-0.005
 enclosure-0.005
 peak-0.005
Token things
Token ID1243
Feature activation-3.65e+00

Pos logit contributions
P+0.013
 band+0.013
 matters+0.012
 gum+0.011
aid+0.011
Neg logit contributions
rant-0.014
 enclosure-0.014
sing-0.013
 horizon-0.013
 doorstep-0.012
Token that
Token ID326
Feature activation-8.39e-01

Pos logit contributions
P+0.114
 band+0.106
Ol+0.096
 gum+0.096
 matters+0.094
Neg logit contributions
rant-0.112
sing-0.097
 horizon-0.096
 enclosure-0.093
 peak-0.090
Token he
Token ID339
Feature activation-7.08e-01

Pos logit contributions
P+0.025
 band+0.025
aid+0.022
Ol+0.021
 gum+0.020
Neg logit contributions
rant-0.023
sing-0.020
 horizon-0.020
 peak-0.018
 previous-0.018
Token had
Token ID550
Feature activation+4.35e-01

Pos logit contributions
 band+0.021
P+0.020
 gum+0.019
 basketball+0.019
 matters+0.018
Neg logit contributions
rant-0.022
sing-0.019
 enclosure-0.019
 horizon-0.018
asser-0.017
Token,
Token ID11
Feature activation+2.28e-01

Pos logit contributions
rant+0.013
 enclosure+0.011
sing+0.011
 horizon+0.011
asser+0.010
Neg logit contributions
 band-0.013
P-0.013
 gum-0.012
 basketball-0.012
 matters-0.012
Token like
Token ID588
Feature activation-6.57e-01

Pos logit contributions
rant+0.006
sing+0.006
 horizon+0.005
 enclosure+0.005
iously+0.005
Neg logit contributions
P-0.008
 band-0.007
 matters-0.007
 gum-0.007
 basketball-0.006
Token
Token ID198
Feature activation-5.91e-01

Pos logit contributions
 band+0.015
P+0.014
 gum+0.014
 matters+0.014
aid+0.013
Neg logit contributions
 enclosure-0.014
rant-0.013
sing-0.013
 horizon-0.012
 doorstep-0.011
TokenThe
Token ID464
Feature activation-8.01e-01

Pos logit contributions
 band+0.018
 gum+0.017
 mar+0.016
 basketball+0.016
 jacket+0.016
Neg logit contributions
rant-0.016
 enclosure-0.015
 horizon-0.014
sing-0.013
anne-0.013
Token 3
Token ID513
Feature activation-3.54e-01

Pos logit contributions
 matters+0.021
P+0.021
 band+0.020
 gum+0.018
Ol+0.018
Neg logit contributions
rant-0.024
sing-0.024
 horizon-0.024
 enclosure-0.024
 doorstep-0.021
Token-
Token ID12
Feature activation-3.77e+00

Pos logit contributions
 band+0.013
P+0.012
 gum+0.011
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.009
sing-0.008
 enclosure-0.008
 horizon-0.008
 doorstep-0.007
Tokenyear
Token ID1941
Feature activation-1.13e+00

Pos logit contributions
 band+0.141
 gum+0.138
 basketball+0.137
 matters+0.131
P+0.130
Neg logit contributions
rant-0.084
 enclosure-0.070
sing-0.068
iously-0.062
 horizon-0.060
Token-
Token ID12
Feature activation-3.64e+00

Pos logit contributions
 band+0.032
P+0.030
 gum+0.030
 basketball+0.029
 matters+0.028
Neg logit contributions
rant-0.036
 enclosure-0.033
sing-0.032
 horizon-0.032
iously-0.030
Tokenold
Token ID727
Feature activation-2.48e+00

Pos logit contributions
 band+0.106
 gum+0.102
 basketball+0.101
 matters+0.097
P+0.096
Neg logit contributions
rant-0.110
 enclosure-0.099
sing-0.095
iously-0.091
 horizon-0.090
Token thought
Token ID1807
Feature activation+2.19e-02

Pos logit contributions
P+0.066
 band+0.065
 gum+0.065
 basketball+0.062
 matters+0.061
Neg logit contributions
rant-0.078
 enclosure-0.074
sing-0.069
iously-0.069
 horizon-0.067
Token hard
Token ID1327
Feature activation-3.89e-01

Pos logit contributions
rant+0.001
iously+0.001
sing+0.001
 horizon+0.001
 enclosure+0.000
Neg logit contributions
P-0.001
 band-0.001
 basketball-0.001
 gum-0.001
 mar-0.001
Token and
Token ID290
Feature activation-6.92e-01

Pos logit contributions
 band+0.011
P+0.011
 gum+0.010
 basketball+0.010
 matters+0.010
Neg logit contributions
rant-0.012
sing-0.011
 enclosure-0.011
 horizon-0.011
iously-0.010
Token smiled
Token ID13541
Feature activation+7.67e-01

Pos logit contributions
P+0.019
 band+0.018
 gum+0.018
 matters+0.017
 basketball+0.017
Neg logit contributions
rant-0.021
iously-0.020
 enclosure-0.019
sing-0.019
 horizon-0.019
Token a
Token ID257
Feature activation+3.82e-01

Pos logit contributions
rant+0.018
sing+0.017
 enclosure+0.017
 horizon+0.017
 doorstep+0.016
Neg logit contributions
P-0.009
 band-0.009
 gum-0.008
 basketball-0.008
 matters-0.008
Token loud
Token ID7812
Feature activation-2.34e-01

Pos logit contributions
rant+0.018
 enclosure+0.015
iously+0.015
ples+0.014
 horizon+0.014
Neg logit contributions
 band-0.008
 gum-0.006
 basketball-0.005
 his-0.005
P-0.005
Token noise
Token ID7838
Feature activation-5.15e-01

Pos logit contributions
 band+0.005
 gum+0.004
 his+0.004
 basketball+0.003
P+0.003
Neg logit contributions
rant-0.010
 enclosure-0.009
 horizon-0.009
sing-0.009
 doorstep-0.009
Token.
Token ID13
Feature activation+3.22e-01

Pos logit contributions
 band+0.012
 gum+0.010
P+0.010
 basketball+0.010
 his+0.010
Neg logit contributions
rant-0.020
 enclosure-0.018
sing-0.018
 horizon-0.018
 doorstep-0.017
Token It
Token ID632
Feature activation-1.96e-01

Pos logit contributions
rant+0.010
 enclosure+0.009
sing+0.009
 horizon+0.008
asser+0.008
Neg logit contributions
 band-0.010
P-0.009
 gum-0.009
 basketball-0.009
 matters-0.009
Token er
Token ID1931
Feature activation-3.63e+00

Pos logit contributions
 band+0.005
 his+0.004
 matters+0.004
 gum+0.004
 mar+0.004
Neg logit contributions
rant-0.007
sing-0.006
 horizon-0.006
 doorstep-0.006
iously-0.006
Tokenases
Token ID1386
Feature activation-5.82e-01

Pos logit contributions
 band+0.105
P+0.096
 gum+0.095
 basketball+0.093
 matters+0.090
Neg logit contributions
rant-0.118
 enclosure-0.106
sing-0.104
 horizon-0.101
asser-0.096
Token part
Token ID636
Feature activation-8.32e-02

Pos logit contributions
 band+0.012
P+0.012
 gum+0.011
 basketball+0.011
 matters+0.010
Neg logit contributions
rant-0.023
 enclosure-0.021
sing-0.020
 horizon-0.020
asser-0.019
Token of
Token ID286
Feature activation+7.34e-02

Pos logit contributions
P+0.002
 band+0.002
 gum+0.002
Ol+0.002
 basketball+0.002
Neg logit contributions
rant-0.003
 enclosure-0.002
sing-0.002
 horizon-0.002
 entrance-0.002
Token their
Token ID511
Feature activation-7.66e-01

Pos logit contributions
rant+0.003
 enclosure+0.003
sing+0.003
 horizon+0.003
 peak+0.003
Neg logit contributions
 band-0.001
P-0.001
 gum-0.001
 basketball-0.001
 matters-0.001
Token mountain
Token ID8598
Feature activation+5.51e-01

Pos logit contributions
 band+0.019
P+0.018
 gum+0.017
 basketball+0.017
 matters+0.016
Neg logit contributions
rant-0.028
 enclosure-0.025
sing-0.025
 horizon-0.024
asser-0.023
Token brown
Token ID7586
Feature activation-6.85e-01

Pos logit contributions
rant+0.017
sing+0.015
 enclosure+0.015
 horizon+0.015
 doorstep+0.013
Neg logit contributions
 band-0.016
P-0.015
 gum-0.014
 basketball-0.014
 matters-0.014
Token and
Token ID290
Feature activation-5.26e-01

Pos logit contributions
 band+0.018
 gum+0.017
P+0.017
 basketball+0.016
 matters+0.015
Neg logit contributions
rant-0.024
 enclosure-0.021
sing-0.021
 horizon-0.021
asser-0.019
Token it
Token ID340
Feature activation-1.04e+00

Pos logit contributions
 band+0.015
P+0.014
 basketball+0.013
 gum+0.013
 matters+0.013
Neg logit contributions
rant-0.017
sing-0.015
 horizon-0.015
 enclosure-0.015
asser-0.013
Token had
Token ID550
Feature activation+3.59e-01

Pos logit contributions
 band+0.030
P+0.029
 gum+0.028
 basketball+0.027
 matters+0.026
Neg logit contributions
rant-0.033
sing-0.029
 enclosure-0.028
 horizon-0.028
 doorstep-0.026
Token many
Token ID867
Feature activation-1.25e+00

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.012
 doorstep+0.011
Neg logit contributions
 band-0.009
P-0.008
 gum-0.008
 basketball-0.008
 matters-0.007
Token holes
Token ID10421
Feature activation-3.63e+00

Pos logit contributions
 band+0.034
P+0.032
 gum+0.031
 basketball+0.030
 matters+0.029
Neg logit contributions
rant-0.042
 enclosure-0.038
sing-0.038
 horizon-0.037
asser-0.034
Token in
Token ID287
Feature activation+2.04e-01

Pos logit contributions
 band+0.093
P+0.085
 gum+0.083
 basketball+0.080
 matters+0.078
Neg logit contributions
rant-0.130
 enclosure-0.116
sing-0.116
 horizon-0.114
asser-0.106
Token it
Token ID340
Feature activation-1.13e+00

Pos logit contributions
rant+0.009
 enclosure+0.008
 horizon+0.008
sing+0.008
 doorstep+0.007
Neg logit contributions
 band-0.004
P-0.004
 gum-0.003
 basketball-0.003
 matters-0.003
Token.
Token ID13
Feature activation-3.85e-01

Pos logit contributions
 band+0.032
P+0.030
 gum+0.029
 basketball+0.028
 matters+0.027
Neg logit contributions
rant-0.037
 enclosure-0.033
sing-0.033
 horizon-0.033
asser-0.030
Token Rose
Token ID8049
Feature activation-7.26e-01

Pos logit contributions
 band+0.010
P+0.009
 gum+0.009
 basketball+0.008
 matters+0.008
Neg logit contributions
rant-0.014
 enclosure-0.012
sing-0.012
 horizon-0.012
asser-0.011
Token was
Token ID373
Feature activation+2.07e-01

Pos logit contributions
 band+0.018
P+0.016
 gum+0.016
 basketball+0.016
 matters+0.015
Neg logit contributions
rant-0.026
 enclosure-0.024
sing-0.024
 horizon-0.023
asser-0.022
Token things
Token ID1243
Feature activation-2.17e+00

Pos logit contributions
rant+0.036
 enclosure+0.033
sing+0.033
 horizon+0.033
iously+0.031
Neg logit contributions
 band-0.016
P-0.014
 basketball-0.013
 gum-0.013
Ol-0.012
Token can
Token ID460
Feature activation-1.11e+00

Pos logit contributions
P+0.070
 band+0.067
 gum+0.062
Ol+0.062
 basketball+0.061
Neg logit contributions
rant-0.062
 horizon-0.053
sing-0.053
 enclosure-0.051
asser-0.048
Token come
Token ID1282
Feature activation-8.67e-03

Pos logit contributions
 band+0.031
 gum+0.027
P+0.027
 basketball+0.026
 matters+0.026
Neg logit contributions
rant-0.038
 enclosure-0.035
sing-0.034
 horizon-0.033
asser-0.031
Token from
Token ID422
Feature activation-1.35e+00

Pos logit contributions
 band+0.000
P+0.000
 gum+0.000
 basketball+0.000
 matters+0.000
Neg logit contributions
rant-0.000
 enclosure-0.000
sing-0.000
 horizon-0.000
asser-0.000
Token bad
Token ID2089
Feature activation+4.21e-01

Pos logit contributions
 band+0.036
P+0.034
 gum+0.032
aid+0.032
Ol+0.031
Neg logit contributions
rant-0.045
sing-0.041
 horizon-0.041
 enclosure-0.040
 doorstep-0.036
Token situations
Token ID7445
Feature activation-3.62e+00

Pos logit contributions
rant+0.015
 horizon+0.013
 enclosure+0.013
sing+0.013
 doorstep+0.012
Neg logit contributions
 band-0.010
P-0.009
 basketball-0.009
 gum-0.009
Ol-0.009
Token.
Token ID13
Feature activation+3.29e-01

Pos logit contributions
 band+0.101
P+0.099
 gum+0.092
 basketball+0.089
 matters+0.087
Neg logit contributions
rant-0.118
 enclosure-0.107
 horizon-0.102
sing-0.102
asser-0.096
Token<|endoftext|>
Token ID50256
Feature activation-2.79e-01

Pos logit contributions
rant+0.015
sing+0.013
 enclosure+0.013
 horizon+0.013
iously+0.013
Neg logit contributions
 band-0.006
 gum-0.005
P-0.005
 basketball-0.005
 matters-0.005
TokenOnce
Token ID7454
Feature activation-1.98e-01

Pos logit contributions
 band+0.007
P+0.006
 gum+0.006
 basketball+0.006
 matters+0.006
Neg logit contributions
rant-0.010
 enclosure-0.009
 horizon-0.009
sing-0.009
asser-0.008
Token upon
Token ID2402
Feature activation-3.19e-01

Pos logit contributions
 band+0.006
P+0.006
 gum+0.005
 basketball+0.005
 matters+0.005
Neg logit contributions
rant-0.006
 enclosure-0.005
sing-0.005
 horizon-0.005
asser-0.005
Token a
Token ID257
Feature activation-9.21e-01

Pos logit contributions
 band+0.005
P+0.005
 gum+0.005
 basketball+0.004
 matters+0.004
Neg logit contributions
rant-0.014
 enclosure-0.013
sing-0.013
 horizon-0.013
asser-0.012
Token the
Token ID262
Feature activation+4.57e-01

Pos logit contributions
rant+0.024
 enclosure+0.024
sing+0.023
 horizon+0.022
 doorstep+0.021
Neg logit contributions
P-0.014
 band-0.012
 gum-0.011
 basketball-0.010
Ol-0.010
Token cage
Token ID17186
Feature activation+1.01e+00

Pos logit contributions
sing+0.014
 horizon+0.014
rant+0.014
 enclosure+0.014
 entrance+0.012
Neg logit contributions
P-0.012
 matters-0.012
 band-0.011
 gum-0.011
 basketball-0.010
Token.
Token ID13
Feature activation+1.54e-01

Pos logit contributions
rant+0.032
 enclosure+0.030
sing+0.029
 horizon+0.028
 doorstep+0.026
Neg logit contributions
P-0.029
 band-0.029
 gum-0.028
 basketball-0.027
 matters-0.026
Token They
Token ID1119
Feature activation+4.12e-01

Pos logit contributions
rant+0.005
 enclosure+0.004
sing+0.004
 horizon+0.004
asser+0.004
Neg logit contributions
 band-0.005
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
Token wished
Token ID16555
Feature activation+6.94e-02

Pos logit contributions
rant+0.014
 enclosure+0.012
sing+0.012
 horizon+0.012
 doorstep+0.011
Neg logit contributions
 band-0.011
P-0.011
 gum-0.011
 basketball-0.010
 matters-0.010
Token they
Token ID484
Feature activation+2.71e+00

Pos logit contributions
sing+0.002
rant+0.002
 horizon+0.002
 enclosure+0.002
iously+0.002
Neg logit contributions
P-0.002
 band-0.002
 basketball-0.001
aid-0.001
 gum-0.001
Token never
Token ID1239
Feature activation+2.08e+00

Pos logit contributions
rant+0.078
 enclosure+0.068
sing+0.068
 horizon+0.066
 doorstep+0.062
Neg logit contributions
 band-0.088
P-0.083
 gum-0.082
 basketball-0.080
 matters-0.076
Token had
Token ID550
Feature activation+1.87e+00

Pos logit contributions
rant+0.063
 enclosure+0.059
sing+0.057
 horizon+0.054
 doorstep+0.051
Neg logit contributions
P-0.059
 band-0.058
 gum-0.057
 basketball-0.056
Ol-0.052
Token to
Token ID284
Feature activation+3.17e-02

Pos logit contributions
rant+0.056
 enclosure+0.052
sing+0.051
 doorstep+0.046
 horizon+0.046
Neg logit contributions
P-0.051
 band-0.048
 basketball-0.047
Ol-0.047
 gum-0.045
Token leave
Token ID2666
Feature activation-7.65e-01

Pos logit contributions
 enclosure+0.001
rant+0.001
sing+0.001
 doorstep+0.001
 horizon+0.001
Neg logit contributions
P-0.001
 gum-0.001
 band-0.001
 basketball-0.001
 matters-0.001
Token him
Token ID683
Feature activation-1.80e-01

Pos logit contributions
 band+0.018
P+0.017
 gum+0.016
 basketball+0.015
 matters+0.015
Neg logit contributions
rant-0.029
 enclosure-0.027
sing-0.026
 horizon-0.026
 doorstep-0.024
Token called
Token ID1444
Feature activation-2.18e-01

Pos logit contributions
 band+0.012
P+0.011
 gum+0.011
 basketball+0.010
 matters+0.010
Neg logit contributions
rant-0.015
 enclosure-0.014
sing-0.014
 horizon-0.013
asser-0.012
Token Lucy
Token ID22162
Feature activation+1.49e+00

Pos logit contributions
 band+0.007
P+0.006
 gum+0.006
 basketball+0.006
aid+0.006
Neg logit contributions
rant-0.007
 enclosure-0.006
sing-0.006
 horizon-0.005
 doorstep-0.005
Token was
Token ID373
Feature activation+9.86e-02

Pos logit contributions
rant+0.057
sing+0.054
 horizon+0.052
 enclosure+0.051
asser+0.048
Neg logit contributions
 band-0.035
P-0.031
 gum-0.029
 basketball-0.029
 matters-0.028
Token adm
Token ID6178
Feature activation+2.96e-02

Pos logit contributions
rant+0.003
sing+0.003
 enclosure+0.003
 horizon+0.002
asser+0.002
Neg logit contributions
 band-0.003
P-0.003
 gum-0.003
 basketball-0.003
 matters-0.003
Tokeniring
Token ID3428
Feature activation+1.76e-01

Pos logit contributions
rant+0.001
 horizon+0.001
 enclosure+0.001
sing+0.000
iously+0.000
Neg logit contributions
 band-0.001
 gum-0.001
 matters-0.001
 basketball-0.001
P-0.001
Token Maggie
Token ID33014
Feature activation+3.03e+00

Pos logit contributions
rant+0.007
 enclosure+0.006
 horizon+0.006
sing+0.006
iously+0.006
Neg logit contributions
P-0.004
 band-0.004
aid-0.003
Ol-0.003
 gum-0.003
Token.
Token ID13
Feature activation-7.72e-03

Pos logit contributions
rant+0.110
sing+0.100
 enclosure+0.099
 horizon+0.097
asser+0.092
Neg logit contributions
 band-0.076
 gum-0.068
P-0.067
 basketball-0.065
 matters-0.063
Token Lucy
Token ID22162
Feature activation+4.03e-01

Pos logit contributions
 band+0.000
 basketball+0.000
 his+0.000
 gum+0.000
P+0.000
Neg logit contributions
rant-0.000
sing-0.000
 enclosure-0.000
 horizon-0.000
asser-0.000
Token was
Token ID373
Feature activation+5.59e-01

Pos logit contributions
rant+0.015
sing+0.014
 enclosure+0.013
 horizon+0.013
asser+0.013
Neg logit contributions
 band-0.010
P-0.009
 gum-0.009
 basketball-0.008
 matters-0.008
Token only
Token ID691
Feature activation-3.32e-01

Pos logit contributions
rant+0.017
sing+0.016
 enclosure+0.015
 horizon+0.015
 doorstep+0.014
Neg logit contributions
 band-0.017
P-0.016
 gum-0.015
 basketball-0.015
 matters-0.015
Token three
Token ID1115
Feature activation-4.52e-01

Pos logit contributions
 band+0.012
P+0.012
 gum+0.011
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.008
sing-0.007
 enclosure-0.007
 horizon-0.006
iously-0.006
Token were
Token ID547
Feature activation-1.23e+00

Pos logit contributions
 band+0.009
P+0.009
 gum+0.008
aid+0.008
 basketball+0.008
Neg logit contributions
rant-0.009
 enclosure-0.009
sing-0.008
 horizon-0.008
 doorstep-0.008
Token two
Token ID734
Feature activation-1.80e+00

Pos logit contributions
 band+0.024
 gum+0.020
 basketball+0.019
 matters+0.019
aid+0.019
Neg logit contributions
rant-0.049
 enclosure-0.045
 horizon-0.044
sing-0.044
 doorstep-0.041
Token best
Token ID1266
Feature activation-2.34e-01

Pos logit contributions
 band+0.023
P+0.019
 gum+0.019
 basketball+0.017
 matters+0.016
Neg logit contributions
rant-0.086
 enclosure-0.081
sing-0.080
 horizon-0.079
asser-0.075
Token friends
Token ID2460
Feature activation-6.00e-02

Pos logit contributions
 band+0.003
 matters+0.003
 gum+0.003
P+0.003
 basketball+0.003
Neg logit contributions
 enclosure-0.010
sing-0.010
rant-0.010
asser-0.009
 horizon-0.009
Token,
Token ID11
Feature activation+2.31e+00

Pos logit contributions
 band+0.001
P+0.001
 gum+0.001
 basketball+0.001
 matters+0.001
Neg logit contributions
rant-0.002
 horizon-0.002
sing-0.002
 enclosure-0.002
 doorstep-0.002
Token M
Token ID337
Feature activation+2.13e+00

Pos logit contributions
rant+0.082
sing+0.080
 enclosure+0.078
 horizon+0.069
astically+0.066
Neg logit contributions
 band-0.053
P-0.049
 matters-0.047
 gum-0.047
aid-0.044
Tokenina
Token ID1437
Feature activation+9.63e-01

Pos logit contributions
rant+0.070
 enclosure+0.063
 horizon+0.061
sing+0.061
 peak+0.058
Neg logit contributions
 band-0.061
P-0.055
 gum-0.054
 basketball-0.052
 mar-0.049
Token and
Token ID290
Feature activation-6.22e-01

Pos logit contributions
rant+0.035
sing+0.031
 enclosure+0.031
 horizon+0.031
asser+0.029
Neg logit contributions
 band-0.025
P-0.022
 gum-0.022
 basketball-0.021
 matters-0.020
Token Nicole
Token ID25176
Feature activation+8.01e-02

Pos logit contributions
 band+0.016
 gum+0.015
P+0.014
aid+0.014
 pins+0.013
Neg logit contributions
rant-0.019
sing-0.018
 enclosure-0.017
ples-0.016
anne-0.016
Token.
Token ID13
Feature activation+9.92e-01

Pos logit contributions
rant+0.003
 enclosure+0.003
sing+0.003
 horizon+0.003
asser+0.002
Neg logit contributions
 band-0.002
P-0.002
 gum-0.002
 basketball-0.002
 matters-0.002
Token They
Token ID1119
Feature activation+2.57e-01

Pos logit contributions
rant+0.034
 enclosure+0.031
sing+0.031
 horizon+0.030
 doorstep+0.029
Neg logit contributions
 band-0.026
 basketball-0.024
 gum-0.024
P-0.024
 matters-0.022
Token a
Token ID257
Feature activation-6.70e-01

Pos logit contributions
 band+0.007
P+0.006
 gum+0.006
 basketball+0.006
 matters+0.006
Neg logit contributions
rant-0.017
sing-0.016
 enclosure-0.016
 horizon-0.015
asser-0.014
Token little
Token ID1310
Feature activation-1.26e+00

Pos logit contributions
 band+0.015
P+0.013
 gum+0.013
 basketball+0.013
 matters+0.012
Neg logit contributions
rant-0.026
 enclosure-0.024
sing-0.024
 horizon-0.023
asser-0.022
Token fish
Token ID5916
Feature activation+6.37e-01

Pos logit contributions
 band+0.028
P+0.026
 matters+0.026
 gum+0.026
 basketball+0.024
Neg logit contributions
rant-0.045
sing-0.043
 enclosure-0.042
 horizon-0.041
asser-0.038
Token named
Token ID3706
Feature activation+7.50e-01

Pos logit contributions
rant+0.023
 horizon+0.021
sing+0.021
 enclosure+0.020
iously+0.020
Neg logit contributions
 band-0.016
P-0.015
 gum-0.014
 basketball-0.014
 matters-0.013
Token Nem
Token ID22547
Feature activation+1.78e+00

Pos logit contributions
rant+0.033
 enclosure+0.030
sing+0.030
 horizon+0.030
 doorstep+0.028
Neg logit contributions
 band-0.013
P-0.012
 gum-0.011
 basketball-0.011
 matters-0.010
Tokeno
Token ID78
Feature activation+2.11e+00

Pos logit contributions
rant+0.062
 horizon+0.059
sing+0.056
 enclosure+0.056
 doorstep+0.054
Neg logit contributions
 band-0.047
P-0.046
 gum-0.042
 basketball-0.041
 his-0.040
Token.
Token ID13
Feature activation+1.18e+00

Pos logit contributions
rant+0.082
 horizon+0.075
sing+0.074
 enclosure+0.072
iously+0.069
Neg logit contributions
 band-0.050
P-0.045
 gum-0.042
 basketball-0.042
 his-0.040
Token Nem
Token ID22547
Feature activation+6.26e-01

Pos logit contributions
rant+0.052
 enclosure+0.048
sing+0.048
 horizon+0.048
asser+0.045
Neg logit contributions
 band-0.020
P-0.017
 gum-0.017
 basketball-0.016
 matters-0.015
Tokeno
Token ID78
Feature activation+1.45e+00

Pos logit contributions
rant+0.022
 horizon+0.021
sing+0.020
 enclosure+0.020
 doorstep+0.020
Neg logit contributions
 band-0.016
P-0.015
 gum-0.014
 basketball-0.014
 matters-0.014
Token lived
Token ID5615
Feature activation-1.53e+00

Pos logit contributions
rant+0.055
sing+0.050
 horizon+0.050
 enclosure+0.049
iously+0.047
Neg logit contributions
 band-0.035
P-0.032
 gum-0.030
 basketball-0.029
 matters-0.029
Token in
Token ID287
Feature activation+1.09e-01

Pos logit contributions
 band+0.036
 gum+0.035
 matters+0.033
 basketball+0.032
P+0.031
Neg logit contributions
rant-0.053
sing-0.048
 enclosure-0.048
asser-0.046
 horizon-0.044
Token butterflies
Token ID44571
Feature activation-9.59e-01

Pos logit contributions
rant+0.001
 enclosure+0.001
 horizon+0.001
sing+0.001
iously+0.001
Neg logit contributions
 band-0.001
P-0.001
 gum-0.001
 basketball-0.001
Ol-0.001
Token.
Token ID13
Feature activation+4.29e-01

Pos logit contributions
 band+0.027
P+0.026
 gum+0.025
 matters+0.024
 basketball+0.024
Neg logit contributions
rant-0.031
 enclosure-0.028
sing-0.027
 horizon-0.027
asser-0.025
Token Eventually
Token ID16178
Feature activation-4.38e-01

Pos logit contributions
rant+0.012
sing+0.011
 enclosure+0.011
 horizon+0.010
asser+0.010
Neg logit contributions
 band-0.014
 gum-0.013
P-0.013
 basketball-0.013
 matters-0.012
Token,
Token ID11
Feature activation+5.62e-01

Pos logit contributions
P+0.013
 band+0.013
aid+0.012
 basketball+0.012
 gum+0.012
Neg logit contributions
rant-0.011
 horizon-0.010
 enclosure-0.010
 peak-0.010
iously-0.010
Token the
Token ID262
Feature activation+2.44e-01

Pos logit contributions
rant+0.019
 horizon+0.019
sing+0.018
 peak+0.018
 enclosure+0.018
Neg logit contributions
P-0.013
 band-0.012
aid-0.011
 gum-0.011
Ol-0.011
Token sun
Token ID4252
Feature activation+2.41e+00

Pos logit contributions
rant+0.008
sing+0.008
 horizon+0.007
 enclosure+0.007
 doorstep+0.007
Neg logit contributions
 band-0.006
P-0.006
 gum-0.006
 basketball-0.006
 matters-0.006
Token started
Token ID2067
Feature activation+9.49e-01

Pos logit contributions
rant+0.082
 horizon+0.075
sing+0.072
 enclosure+0.072
 doorstep+0.068
Neg logit contributions
 band-0.062
P-0.061
 basketball-0.060
 gum-0.059
 matters-0.056
Token to
Token ID284
Feature activation+6.36e-01

Pos logit contributions
rant+0.032
sing+0.029
 enclosure+0.028
 horizon+0.028
 doorstep+0.026
Neg logit contributions
 band-0.026
P-0.025
 gum-0.024
 basketball-0.023
 matters-0.023
Token set
Token ID900
Feature activation+4.11e-01

Pos logit contributions
rant+0.021
 enclosure+0.019
sing+0.019
 horizon+0.018
asser+0.017
Neg logit contributions
 band-0.018
P-0.017
 gum-0.017
 basketball-0.016
 matters-0.016
Token and
Token ID290
Feature activation+3.61e-04

Pos logit contributions
rant+0.011
 horizon+0.011
sing+0.010
 enclosure+0.010
 doorstep+0.010
Neg logit contributions
P-0.013
 band-0.013
 matters-0.012
 basketball-0.012
 gum-0.012
Token the
Token ID262
Feature activation+4.67e-01

Pos logit contributions
rant+0.000
 horizon+0.000
sing+0.000
 enclosure+0.000
 peak+0.000
Neg logit contributions
P-0.000
 band-0.000
 gum-0.000
 basketball-0.000
Ol-0.000
Token a
Token ID257
Feature activation-1.11e+00

Pos logit contributions
 band+0.019
P+0.016
 gum+0.016
 matters+0.015
 basketball+0.015
Neg logit contributions
rant-0.034
sing-0.032
 enclosure-0.031
 horizon-0.030
asser-0.028
Token girl
Token ID2576
Feature activation+3.13e-01

Pos logit contributions
 band+0.026
P+0.023
 gum+0.023
 matters+0.022
 basketball+0.022
Neg logit contributions
rant-0.041
sing-0.038
 enclosure-0.038
 horizon-0.037
asser-0.035
Token named
Token ID3706
Feature activation+2.63e-01

Pos logit contributions
rant+0.012
 horizon+0.011
sing+0.011
 enclosure+0.010
iously+0.010
Neg logit contributions
 band-0.008
P-0.007
 gum-0.007
 basketball-0.007
 matters-0.006
Token Anna
Token ID11735
Feature activation+1.66e+00

Pos logit contributions
rant+0.012
 enclosure+0.011
sing+0.011
 horizon+0.011
asser+0.011
Neg logit contributions
 band-0.004
P-0.003
 gum-0.003
 basketball-0.003
 matters-0.003
Token.
Token ID13
Feature activation+9.18e-01

Pos logit contributions
rant+0.068
 horizon+0.063
sing+0.062
 enclosure+0.061
iously+0.059
Neg logit contributions
 band-0.036
P-0.032
 gum-0.030
 basketball-0.029
 his-0.029
Token Anna
Token ID11735
Feature activation+7.29e-01

Pos logit contributions
rant+0.035
 enclosure+0.032
sing+0.032
 horizon+0.031
iously+0.031
Neg logit contributions
 band-0.021
 gum-0.020
 basketball-0.020
P-0.018
 his-0.018
Token was
Token ID373
Feature activation+5.90e-01

Pos logit contributions
rant+0.026
sing+0.023
 enclosure+0.023
 horizon+0.023
iously+0.021
Neg logit contributions
 band-0.020
P-0.018
 gum-0.018
 basketball-0.017
 matters-0.017
Token three
Token ID1115
Feature activation-5.78e-01

Pos logit contributions
rant+0.020
sing+0.018
 enclosure+0.018
 horizon+0.017
asser+0.016
Neg logit contributions
 band-0.016
P-0.015
 gum-0.015
 basketball-0.014
 matters-0.014
Token years
Token ID812
Feature activation-4.62e-01

Pos logit contributions
 band+0.022
P+0.020
 gum+0.020
 basketball+0.019
 mar+0.019
Neg logit contributions
rant-0.015
sing-0.013
 horizon-0.013
 enclosure-0.012
 peak-0.011
Token old
Token ID1468
Feature activation-1.65e+00

Pos logit contributions
 band+0.013
P+0.013
 gum+0.012
 basketball+0.012
 matters+0.012
Neg logit contributions
rant-0.015
 enclosure-0.013
sing-0.013
 horizon-0.013
 doorstep-0.012
Token and
Token ID290
Feature activation-7.83e-01

Pos logit contributions
 band+0.039
 gum+0.033
 basketball+0.031
P+0.029
 his+0.029
Neg logit contributions
rant-0.070
sing-0.062
 enclosure-0.062
 horizon-0.061
 peak-0.058
Token
Token ID198
Feature activation-7.30e-01

Pos logit contributions
 band+0.017
P+0.016
 gum+0.015
 basketball+0.015
 matters+0.014
Neg logit contributions
rant-0.019
 enclosure-0.017
sing-0.017
 horizon-0.016
asser-0.015
Token
Token ID198
Feature activation-3.78e-01

Pos logit contributions
 band+0.015
P+0.014
 gum+0.013
 basketball+0.013
 matters+0.012
Neg logit contributions
rant-0.030
 enclosure-0.027
sing-0.027
 horizon-0.026
asser-0.025
TokenAnna
Token ID31160
Feature activation+1.91e-01

Pos logit contributions
 band+0.010
 gum+0.009
 basketball+0.009
P+0.008
 matters+0.008
Neg logit contributions
rant-0.013
 enclosure-0.012
sing-0.011
 horizon-0.011
asser-0.011
Token starts
Token ID4940
Feature activation-1.07e-02

Pos logit contributions
rant+0.007
sing+0.006
 horizon+0.006
 enclosure+0.006
 doorstep+0.005
Neg logit contributions
 band-0.005
 gum-0.005
P-0.005
 matters-0.004
 basketball-0.004
Token to
Token ID284
Feature activation-7.92e-01

Pos logit contributions
P+0.000
 matters+0.000
 band+0.000
 arrows+0.000
 basketball+0.000
Neg logit contributions
rant-0.000
sing-0.000
 enclosure-0.000
 horizon-0.000
iously-0.000
Token cry
Token ID3960
Feature activation+3.50e-01

Pos logit contributions
P+0.023
 band+0.022
 gum+0.021
 basketball+0.020
aid+0.020
Neg logit contributions
rant-0.024
sing-0.022
 enclosure-0.022
 horizon-0.020
asser-0.020
Token.
Token ID13
Feature activation-4.89e-01

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.011
 doorstep+0.011
Neg logit contributions
 band-0.009
 gum-0.008
P-0.008
 basketball-0.008
 matters-0.007
Token She
Token ID1375
Feature activation+5.07e-01

Pos logit contributions
 band+0.014
 gum+0.013
 basketball+0.012
P+0.012
 matters+0.012
Neg logit contributions
rant-0.017
 enclosure-0.015
sing-0.014
 horizon-0.014
asser-0.013
Token is
Token ID318
Feature activation+1.27e-01

Pos logit contributions
rant+0.018
 horizon+0.017
sing+0.017
 enclosure+0.016
 doorstep+0.016
Neg logit contributions
 band-0.013
 gum-0.012
P-0.012
 matters-0.012
 his-0.012
Token hurt
Token ID5938
Feature activation-6.65e-01

Pos logit contributions
rant+0.005
sing+0.004
 enclosure+0.004
 horizon+0.004
asser+0.004
Neg logit contributions
 band-0.003
P-0.003
 gum-0.003
 basketball-0.003
 matters-0.003
Token and
Token ID290
Feature activation-7.59e-01

Pos logit contributions
 band+0.018
P+0.016
 gum+0.016
 basketball+0.016
 matters+0.015
Neg logit contributions
rant-0.023
 enclosure-0.021
sing-0.021
 horizon-0.020
 doorstep-0.019
Token stay
Token ID2652
Feature activation-3.82e-01

Pos logit contributions
rant+0.010
 enclosure+0.009
sing+0.009
 horizon+0.009
 doorstep+0.008
Neg logit contributions
 band-0.011
P-0.011
 gum-0.011
 basketball-0.010
 matters-0.010
Token in
Token ID287
Feature activation+2.96e-01

Pos logit contributions
 band+0.010
 gum+0.009
P+0.009
 basketball+0.009
 matters+0.008
Neg logit contributions
rant-0.013
 enclosure-0.012
sing-0.012
 horizon-0.012
 doorstep-0.011
Token the
Token ID262
Feature activation+2.13e-01

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.012
 doorstep+0.011
Neg logit contributions
 band-0.005
P-0.005
 gum-0.004
 basketball-0.004
 matters-0.004
Token perfect
Token ID2818
Feature activation+1.23e+00

Pos logit contributions
rant+0.006
 horizon+0.005
 enclosure+0.005
sing+0.005
 doorstep+0.004
Neg logit contributions
 band-0.007
P-0.007
 gum-0.007
 basketball-0.007
 matters-0.007
Token sky
Token ID6766
Feature activation+5.99e-01

Pos logit contributions
rant+0.034
 enclosure+0.031
 horizon+0.030
sing+0.030
 doorstep+0.027
Neg logit contributions
 band-0.040
P-0.039
 gum-0.038
 basketball-0.037
 matters-0.036
Token.
Token ID13
Feature activation+3.16e-01

Pos logit contributions
rant+0.019
 enclosure+0.017
sing+0.017
 horizon+0.016
 doorstep+0.015
Neg logit contributions
 band-0.018
P-0.017
 gum-0.016
 basketball-0.016
 matters-0.015
Token<|endoftext|>
Token ID50256
Feature activation+2.75e-01

Pos logit contributions
rant+0.012
 enclosure+0.011
sing+0.011
 horizon+0.010
asser+0.010
Neg logit contributions
 band-0.008
P-0.007
 gum-0.007
 basketball-0.007
 matters-0.006
TokenOnce
Token ID7454
Feature activation-9.93e-02

Pos logit contributions
rant+0.008
 enclosure+0.008
 horizon+0.007
asser+0.007
sing+0.007
Neg logit contributions
 band-0.008
 gum-0.007
 basketball-0.007
P-0.007
aid-0.007
Token there
Token ID612
Feature activation-4.24e-01

Pos logit contributions
 band+0.003
P+0.003
 gum+0.003
 basketball+0.003
 matters+0.003
Neg logit contributions
rant-0.003
 enclosure-0.003
sing-0.003
 horizon-0.002
asser-0.002
Token was
Token ID373
Feature activation-6.17e-01

Pos logit contributions
 band+0.012
P+0.012
 gum+0.012
 basketball+0.011
aid+0.011
Neg logit contributions
rant-0.013
 enclosure-0.012
sing-0.011
 horizon-0.011
 peak-0.011
Token a
Token ID257
Feature activation-7.69e-01

Pos logit contributions
 band+0.013
P+0.011
 gum+0.011
 basketball+0.010
 matters+0.010
Neg logit contributions
rant-0.024
sing-0.023
 enclosure-0.022
 horizon-0.021
asser-0.020
Token cold
Token ID4692
Feature activation+3.78e-01

Pos logit contributions
 band+0.002
 gum+0.001
P+0.001
 basketball+0.001
 his+0.001
Neg logit contributions
rant-0.002
 enclosure-0.002
sing-0.002
 horizon-0.002
iously-0.002
Token wind
Token ID2344
Feature activation+1.02e+00

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.011
asser+0.011
Neg logit contributions
 band-0.010
P-0.009
 gum-0.009
 basketball-0.009
 matters-0.009
Token and
Token ID290
Feature activation+2.41e-01

Pos logit contributions
rant+0.038
sing+0.034
 enclosure+0.034
 horizon+0.034
asser+0.032
Neg logit contributions
 band-0.024
P-0.022
 gum-0.022
 basketball-0.021
 matters-0.020
Token heard
Token ID2982
Feature activation+1.23e+00

Pos logit contributions
rant+0.008
sing+0.008
 enclosure+0.008
 horizon+0.008
 peak+0.007
Neg logit contributions
P-0.006
 band-0.006
 basketball-0.006
 gum-0.005
 matters-0.005
Token a
Token ID257
Feature activation+2.89e-01

Pos logit contributions
rant+0.045
 enclosure+0.042
sing+0.041
 horizon+0.041
 entrance+0.037
Neg logit contributions
P-0.031
 band-0.027
 gum-0.025
 basketball-0.025
 matters-0.025
Token strange
Token ID6283
Feature activation-7.45e-02

Pos logit contributions
rant+0.010
 enclosure+0.009
sing+0.009
 horizon+0.009
asser+0.008
Neg logit contributions
 band-0.008
 gum-0.007
P-0.007
 basketball-0.006
 matters-0.006
Token noise
Token ID7838
Feature activation-4.21e-01

Pos logit contributions
 band+0.002
 gum+0.001
P+0.001
 basketball+0.001
 matters+0.001
Neg logit contributions
rant-0.003
 enclosure-0.003
sing-0.003
 horizon-0.003
asser-0.003
Token.
Token ID13
Feature activation-4.40e-01

Pos logit contributions
 band+0.011
P+0.010
 gum+0.010
 basketball+0.010
 matters+0.009
Neg logit contributions
rant-0.015
 enclosure-0.013
sing-0.013
 horizon-0.013
asser-0.012
Token
Token ID198
Feature activation-5.39e-01

Pos logit contributions
 band+0.013
P+0.012
 gum+0.012
 basketball+0.011
 matters+0.011
Neg logit contributions
rant-0.013
 enclosure-0.012
sing-0.012
 horizon-0.012
asser-0.011
Token
Token ID198
Feature activation-1.66e-01

Pos logit contributions
 band+0.012
P+0.011
 gum+0.011
 basketball+0.010
 matters+0.010
Neg logit contributions
rant-0.021
 enclosure-0.019
sing-0.019
 horizon-0.019
asser-0.018
Token"
Token ID1
Feature activation+5.84e-01

Pos logit contributions
 band+0.004
 gum+0.004
 basketball+0.004
 matters+0.004
 mar+0.004
Neg logit contributions
rant-0.006
 enclosure-0.005
 horizon-0.005
asser-0.005
sing-0.005
Token,
Token ID11
Feature activation+6.81e-01

Pos logit contributions
rant+0.025
sing+0.022
 enclosure+0.022
 horizon+0.022
asser+0.021
Neg logit contributions
 band-0.013
 gum-0.012
P-0.011
 basketball-0.011
 matters-0.010
Token look
Token ID804
Feature activation-1.42e+00

Pos logit contributions
rant+0.021
 enclosure+0.019
sing+0.019
 horizon+0.018
 doorstep+0.017
Neg logit contributions
 band-0.020
P-0.019
 gum-0.018
 basketball-0.017
 matters-0.017
Token at
Token ID379
Feature activation+7.48e-01

Pos logit contributions
 band+0.033
P+0.029
 gum+0.029
 basketball+0.028
 matters+0.027
Neg logit contributions
rant-0.054
 enclosure-0.050
sing-0.050
 horizon-0.049
 doorstep-0.046
Token my
Token ID616
Feature activation+1.54e-01

Pos logit contributions
rant+0.031
 enclosure+0.028
 horizon+0.028
sing+0.028
asser+0.026
Neg logit contributions
 band-0.015
P-0.013
 gum-0.013
 basketball-0.012
 matters-0.011
Token flower
Token ID15061
Feature activation+2.27e-01

Pos logit contributions
rant+0.006
 enclosure+0.005
 horizon+0.005
sing+0.005
asser+0.005
Neg logit contributions
 band-0.004
 gum-0.003
 basketball-0.003
 friends-0.003
 matters-0.003
Token!
Token ID0
Feature activation+2.62e-01

Pos logit contributions
rant+0.008
sing+0.007
 enclosure+0.007
 horizon+0.007
asser+0.006
Neg logit contributions
 band-0.006
 gum-0.006
P-0.006
 basketball-0.006
 matters-0.005
Token It
Token ID632
Feature activation-4.57e-01

Pos logit contributions
rant+0.009
sing+0.008
 enclosure+0.008
 horizon+0.007
asser+0.007
Neg logit contributions
 band-0.007
 gum-0.007
 basketball-0.007
P-0.007
 matters-0.007
Token bloom
Token ID29955
Feature activation+2.27e-01

Pos logit contributions
 band+0.010
 matters+0.009
P+0.009
 gum+0.008
 mar+0.008
Neg logit contributions
rant-0.017
sing-0.017
 enclosure-0.017
 horizon-0.016
iously-0.015
Tokened
Token ID276
Feature activation+8.61e-01

Pos logit contributions
 enclosure+0.008
rant+0.008
sing+0.008
 horizon+0.007
 doorstep+0.007
Neg logit contributions
 band-0.005
P-0.005
 gum-0.005
 basketball-0.005
 matters-0.005
Token!"
Token ID2474
Feature activation+3.41e-01

Pos logit contributions
rant+0.029
 enclosure+0.026
sing+0.026
 horizon+0.025
asser+0.023
Neg logit contributions
 band-0.024
P-0.022
 gum-0.022
 basketball-0.021
 matters-0.020
Token she
Token ID673
Feature activation-1.87e+00

Pos logit contributions
rant+0.010
sing+0.009
 enclosure+0.009
bell+0.009
 doorstep+0.009
Neg logit contributions
P-0.008
 band-0.007
 gum-0.007
Ol-0.007
 medals-0.007
Token treasure
Token ID14068
Feature activation+3.91e-01

Pos logit contributions
rant+0.004
 enclosure+0.004
sing+0.004
 horizon+0.004
iously+0.004
Neg logit contributions
 band-0.003
P-0.003
 gum-0.003
 basketball-0.003
 matters-0.003
Token map
Token ID3975
Feature activation+5.12e-01

Pos logit contributions
rant+0.011
sing+0.009
 horizon+0.009
 enclosure+0.009
asser+0.008
Neg logit contributions
 band-0.013
 gum-0.012
P-0.012
 basketball-0.012
 matters-0.012
Token!"
Token ID2474
Feature activation+6.55e-01

Pos logit contributions
rant+0.014
sing+0.013
 horizon+0.012
 enclosure+0.012
asser+0.011
Neg logit contributions
 band-0.017
 gum-0.016
P-0.015
 basketball-0.015
 matters-0.015
Token Ben
Token ID3932
Feature activation-9.31e-01

Pos logit contributions
rant+0.022
 enclosure+0.019
 horizon+0.018
sing+0.018
iously+0.017
Neg logit contributions
 band-0.017
P-0.016
 gum-0.016
 basketball-0.015
aid-0.015
Token said
Token ID531
Feature activation+4.49e-01

Pos logit contributions
 band+0.021
P+0.020
 gum+0.019
 basketball+0.019
 matters+0.019
Neg logit contributions
rant-0.034
 enclosure-0.031
sing-0.031
iously-0.029
 horizon-0.029
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
sing+0.010
rant+0.010
iously+0.009
 enclosure+0.008
bell+0.008
Neg logit contributions
P-0.017
 band-0.015
 gum-0.015
Ol-0.014
 matters-0.014
Token "
Token ID366
Feature activation-2.86e-02

Pos logit contributions
 band+0.032
P+0.030
 matters+0.029
 gum+0.029
 basketball+0.028
Neg logit contributions
rant-0.027
 enclosure-0.024
sing-0.023
 horizon-0.022
iously-0.021
TokenLet
Token ID5756
Feature activation-3.73e-01

Pos logit contributions
 band+0.001
P+0.001
 gum+0.001
 basketball+0.001
 matters+0.000
Neg logit contributions
rant-0.001
 enclosure-0.001
sing-0.001
 horizon-0.001
iously-0.001
Token's
Token ID338
Feature activation-6.76e-01

Pos logit contributions
 band+0.012
 gum+0.011
 basketball+0.011
P+0.011
aid+0.011
Neg logit contributions
rant-0.010
 enclosure-0.009
 horizon-0.009
sing-0.008
 doorstep-0.008
Token follow
Token ID1061
Feature activation+2.05e-01

Pos logit contributions
 band+0.019
P+0.017
 gum+0.017
 basketball+0.017
 matters+0.016
Neg logit contributions
rant-0.023
sing-0.020
 enclosure-0.020
 horizon-0.020
asser-0.018
Token it
Token ID340
Feature activation-4.82e-01

Pos logit contributions
rant+0.008
 enclosure+0.007
sing+0.007
 horizon+0.007
 peak+0.007
Neg logit contributions
 band-0.004
P-0.004
 gum-0.004
 basketball-0.004
aid-0.004
Token with
Token ID351
Feature activation-1.11e-01

Pos logit contributions
rant+0.023
sing+0.021
 horizon+0.020
 enclosure+0.019
iously+0.019
Neg logit contributions
P-0.032
 band-0.030
 gum-0.029
 matters-0.029
aid-0.027
Token excitement
Token ID14067
Feature activation+1.41e-01

Pos logit contributions
 band+0.003
P+0.003
 gum+0.003
 basketball+0.003
 matters+0.003
Neg logit contributions
rant-0.003
 horizon-0.003
 enclosure-0.003
sing-0.003
 peak-0.003
Token and
Token ID290
Feature activation-2.43e-01

Pos logit contributions
rant+0.004
sing+0.004
 enclosure+0.004
 horizon+0.004
iously+0.004
Neg logit contributions
 band-0.004
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
Token cl
Token ID537
Feature activation-6.69e-01

Pos logit contributions
P+0.008
 gum+0.007
Ol+0.007
 band+0.007
 basketball+0.006
Neg logit contributions
sing-0.006
rant-0.006
iously-0.006
 enclosure-0.006
 doorstep-0.006
Tokenapped
Token ID6320
Feature activation-2.56e-01

Pos logit contributions
 band+0.017
 gum+0.016
P+0.016
 basketball+0.015
 matters+0.015
Neg logit contributions
rant-0.024
 enclosure-0.022
sing-0.022
 horizon-0.021
 doorstep-0.020
Token his
Token ID465
Feature activation-5.22e-01

Pos logit contributions
P+0.006
 gum+0.006
 band+0.006
 matters+0.005
Ol+0.005
Neg logit contributions
rant-0.009
sing-0.008
 horizon-0.008
 enclosure-0.008
iously-0.008
Token hands
Token ID2832
Feature activation+1.11e+00

Pos logit contributions
 band+0.014
P+0.013
 gum+0.013
 basketball+0.013
 matters+0.012
Neg logit contributions
rant-0.017
sing-0.016
 enclosure-0.016
 horizon-0.015
iously-0.015
Token.
Token ID13
Feature activation-3.32e-01

Pos logit contributions
rant+0.034
sing+0.031
 enclosure+0.030
 horizon+0.029
iously+0.028
Neg logit contributions
P-0.033
 band-0.033
 gum-0.031
 matters-0.030
 basketball-0.029
Token
Token ID220
Feature activation+4.47e-01

Pos logit contributions
 band+0.008
P+0.008
 gum+0.008
 basketball+0.007
 matters+0.007
Neg logit contributions
rant-0.012
 enclosure-0.011
sing-0.011
 horizon-0.011
 doorstep-0.010
Token
Token ID198
Feature activation-1.28e-01

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.012
 doorstep+0.011
Neg logit contributions
 band-0.014
P-0.013
 gum-0.012
 basketball-0.012
 mar-0.012
Token
Token ID198
Feature activation+3.66e-01

Pos logit contributions
 band+0.004
 gum+0.003
P+0.003
 basketball+0.003
 pins+0.003
Neg logit contributions
rant-0.004
 enclosure-0.004
 horizon-0.004
sing-0.004
 doorstep-0.003
Token friend
Token ID1545
Feature activation+4.46e-01

Pos logit contributions
 matters+0.007
P+0.007
 band+0.007
aid+0.006
Ol+0.006
Neg logit contributions
 enclosure-0.010
 horizon-0.010
sing-0.010
 doorstep-0.010
rant-0.009
Token,
Token ID11
Feature activation+6.38e-01

Pos logit contributions
 enclosure+0.012
rant+0.012
sing+0.011
 exhibit+0.010
 doorstep+0.010
Neg logit contributions
 band-0.013
P-0.012
aid-0.012
 gum-0.012
 matters-0.012
Token Sue
Token ID26089
Feature activation+1.05e+00

Pos logit contributions
sing+0.018
 enclosure+0.017
rant+0.016
fly+0.013
 entrance+0.013
Neg logit contributions
P-0.018
 penn-0.017
 matters-0.017
 narrative-0.017
 band-0.016
Token.
Token ID13
Feature activation+3.05e-01

Pos logit contributions
rant+0.035
 enclosure+0.032
sing+0.031
 horizon+0.030
asser+0.029
Neg logit contributions
 band-0.029
P-0.027
 gum-0.027
 basketball-0.026
 matters-0.025
Token They
Token ID1119
Feature activation+3.75e-03

Pos logit contributions
rant+0.012
 horizon+0.011
 enclosure+0.010
sing+0.010
 doorstep+0.010
Neg logit contributions
 band-0.007
 basketball-0.006
 gum-0.006
 his-0.006
P-0.005
Token played
Token ID2826
Feature activation-5.30e-01

Pos logit contributions
rant+0.000
 enclosure+0.000
sing+0.000
 horizon+0.000
asser+0.000
Neg logit contributions
 band-0.000
 gum-0.000
P-0.000
 matters-0.000
 basketball-0.000
Token on
Token ID319
Feature activation+6.35e-01

Pos logit contributions
 band+0.011
P+0.011
Ol+0.010
 matters+0.010
 Trevor+0.010
Neg logit contributions
rant-0.017
sing-0.017
asser-0.017
 enclosure-0.016
 entrance-0.014
Token the
Token ID262
Feature activation+1.17e+00

Pos logit contributions
rant+0.025
 enclosure+0.025
 horizon+0.023
sing+0.023
asser+0.023
Neg logit contributions
 band-0.011
P-0.011
aid-0.010
 gum-0.010
 matters-0.010
Token swings
Token ID26728
Feature activation+1.87e+00

Pos logit contributions
rant+0.034
 enclosure+0.032
sing+0.031
 horizon+0.031
 doorstep+0.029
Neg logit contributions
 band-0.035
P-0.033
 matters-0.032
 gum-0.032
 basketball-0.030
Token and
Token ID290
Feature activation+1.61e+00

Pos logit contributions
rant+0.067
 enclosure+0.060
sing+0.060
 horizon+0.060
 doorstep+0.056
Neg logit contributions
 band-0.048
P-0.043
 gum-0.043
 basketball-0.042
 matters-0.039
Token in
Token ID287
Feature activation+7.43e-01

Pos logit contributions
rant+0.059
 enclosure+0.054
sing+0.053
 horizon+0.051
asser+0.048
Neg logit contributions
 band-0.039
P-0.036
 gum-0.036
 basketball-0.033
 matters-0.033
Token to
Token ID284
Feature activation-7.68e-01

Pos logit contributions
rant+0.014
 enclosure+0.012
sing+0.012
 horizon+0.012
asser+0.011
Neg logit contributions
 band-0.010
P-0.009
 gum-0.009
 basketball-0.009
 matters-0.008
Token put
Token ID1234
Feature activation-1.14e+00

Pos logit contributions
 band+0.021
 gum+0.018
P+0.018
 basketball+0.018
 matters+0.017
Neg logit contributions
rant-0.027
 enclosure-0.024
sing-0.024
 horizon-0.024
asser-0.022
Token on
Token ID319
Feature activation+3.71e-01

Pos logit contributions
 band+0.024
P+0.022
 gum+0.020
 basketball+0.020
Ol+0.020
Neg logit contributions
rant-0.046
 enclosure-0.040
 horizon-0.040
sing-0.039
asser-0.038
Token top
Token ID1353
Feature activation+1.80e+00

Pos logit contributions
rant+0.013
 enclosure+0.012
sing+0.012
 horizon+0.012
asser+0.011
Neg logit contributions
 band-0.010
P-0.009
 gum-0.008
 basketball-0.008
 matters-0.008
Token of
Token ID286
Feature activation+4.29e-01

Pos logit contributions
rant+0.071
sing+0.066
 enclosure+0.065
 horizon+0.063
asser+0.061
Neg logit contributions
 band-0.039
 gum-0.035
P-0.034
 basketball-0.033
 matters-0.031
Token her
Token ID607
Feature activation-3.01e-01

Pos logit contributions
rant+0.017
 enclosure+0.016
sing+0.016
 horizon+0.016
asser+0.015
Neg logit contributions
 band-0.008
P-0.008
 gum-0.007
 basketball-0.007
 matters-0.007
Token tower
Token ID10580
Feature activation+4.26e-01

Pos logit contributions
 band+0.009
 gum+0.008
 basketball+0.008
P+0.007
 mar+0.007
Neg logit contributions
rant-0.011
 enclosure-0.009
 horizon-0.009
sing-0.009
asser-0.008
Token.
Token ID13
Feature activation+1.70e-01

Pos logit contributions
rant+0.014
 enclosure+0.013
sing+0.013
 horizon+0.013
 doorstep+0.012
Neg logit contributions
 band-0.012
P-0.011
 gum-0.011
 basketball-0.010
 matters-0.010
Token She
Token ID1375
Feature activation+1.21e-01

Pos logit contributions
rant+0.006
sing+0.005
 enclosure+0.005
 horizon+0.005
 doorstep+0.005
Neg logit contributions
 band-0.005
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
Token reached
Token ID4251
Feature activation+1.68e+00

Pos logit contributions
rant+0.004
 enclosure+0.004
sing+0.004
 horizon+0.003
asser+0.003
Neg logit contributions
 band-0.003
P-0.003
 gum-0.003
 basketball-0.003
 matters-0.003
Token for
Token ID329
Feature activation+1.38e+00

Pos logit contributions
rant+0.055
 enclosure+0.050
 horizon+0.048
sing+0.048
 doorstep+0.045
Neg logit contributions
 band-0.049
 gum-0.044
 basketball-0.043
 his-0.043
 matters-0.042
Token go
Token ID467
Feature activation+3.08e-02

Pos logit contributions
 doorstep+0.006
 enclosure+0.006
rant+0.006
 peak+0.005
sing+0.005
Neg logit contributions
P-0.008
 band-0.008
 gum-0.008
aid-0.007
 basketball-0.007
Token for
Token ID329
Feature activation+1.31e+00

Pos logit contributions
rant+0.001
sing+0.001
 enclosure+0.001
 horizon+0.001
 doorstep+0.001
Neg logit contributions
 band-0.001
P-0.001
 gum-0.001
 basketball-0.001
 matters-0.001
Token a
Token ID257
Feature activation+1.14e+00

Pos logit contributions
rant+0.050
 enclosure+0.047
sing+0.047
 horizon+0.045
asser+0.044
Neg logit contributions
 band-0.025
P-0.025
 basketball-0.023
 gum-0.022
 matters-0.021
Token walk
Token ID2513
Feature activation+1.11e+00

Pos logit contributions
rant+0.040
 enclosure+0.037
sing+0.037
 horizon+0.036
asser+0.033
Neg logit contributions
 band-0.029
P-0.027
 gum-0.026
 basketball-0.025
 matters-0.025
Token in
Token ID287
Feature activation+6.86e-01

Pos logit contributions
rant+0.037
 enclosure+0.034
sing+0.033
 horizon+0.032
asser+0.030
Neg logit contributions
 band-0.030
 gum-0.028
P-0.027
 matters-0.027
 basketball-0.026
Token the
Token ID262
Feature activation-3.00e-01

Pos logit contributions
rant+0.029
 enclosure+0.028
 horizon+0.027
sing+0.027
 peak+0.025
Neg logit contributions
 band-0.012
P-0.011
 gum-0.010
 basketball-0.010
 matters-0.009
Token park
Token ID3952
Feature activation+1.78e+00

Pos logit contributions
 band+0.009
 gum+0.009
P+0.009
 basketball+0.008
 matters+0.008
Neg logit contributions
rant-0.009
 enclosure-0.008
 horizon-0.008
sing-0.008
 doorstep-0.007
Token.
Token ID13
Feature activation+1.76e-01

Pos logit contributions
rant+0.067
sing+0.060
 enclosure+0.059
 horizon+0.059
asser+0.055
Neg logit contributions
 band-0.043
P-0.038
 gum-0.038
 basketball-0.037
 matters-0.035
Token He
Token ID679
Feature activation-1.10e-01

Pos logit contributions
rant+0.007
sing+0.006
 horizon+0.006
 enclosure+0.006
asser+0.006
Neg logit contributions
 band-0.004
 gum-0.004
 basketball-0.003
P-0.003
 his-0.003
Token slowly
Token ID6364
Feature activation+1.29e+00

Pos logit contributions
 band+0.003
 gum+0.003
P+0.003
 basketball+0.003
 matters+0.003
Neg logit contributions
rant-0.003
 enclosure-0.003
sing-0.003
 horizon-0.003
asser-0.003
Token walked
Token ID6807
Feature activation+6.44e-02

Pos logit contributions
 enclosure+0.038
rant+0.038
 horizon+0.034
 peak+0.031
sing+0.031
Neg logit contributions
 gum-0.039
P-0.037
 basketball-0.037
 band-0.036
 matters-0.036
Token no
Token ID645
Feature activation-6.03e-01

Pos logit contributions
P+0.009
 band+0.009
 basketball+0.008
Ol+0.008
aid+0.008
Neg logit contributions
rant-0.007
sing-0.006
iously-0.005
 horizon-0.005
 definitely-0.005
Token because
Token ID780
Feature activation-4.24e-01

Pos logit contributions
 band+0.015
 gum+0.014
P+0.013
 basketball+0.013
 matters+0.013
Neg logit contributions
rant-0.022
 enclosure-0.020
sing-0.020
 horizon-0.019
asser-0.018
Token it
Token ID340
Feature activation-4.87e-01

Pos logit contributions
P+0.010
 band+0.010
Ol+0.009
aid+0.009
 basketball+0.008
Neg logit contributions
rant-0.014
 horizon-0.013
 enclosure-0.013
bell-0.013
sing-0.013
Token was
Token ID373
Feature activation+8.71e-01

Pos logit contributions
P+0.014
 band+0.014
 basketball+0.013
 gum+0.013
aid+0.012
Neg logit contributions
rant-0.014
 enclosure-0.013
sing-0.012
 horizon-0.012
 doorstep-0.012
Token too
Token ID1165
Feature activation+1.11e-01

Pos logit contributions
rant+0.025
sing+0.022
 enclosure+0.021
 horizon+0.021
asser+0.020
Neg logit contributions
 band-0.028
P-0.027
 basketball-0.025
 gum-0.025
 matters-0.024
Token expensive
Token ID5789
Feature activation-2.68e+00

Pos logit contributions
rant+0.003
 enclosure+0.003
sing+0.003
 horizon+0.003
asser+0.003
Neg logit contributions
 band-0.003
P-0.003
 basketball-0.003
 gum-0.003
 matters-0.003
Token.
Token ID13
Feature activation-3.93e-01

Pos logit contributions
 band+0.082
P+0.080
 gum+0.073
 basketball+0.073
Ol+0.071
Neg logit contributions
rant-0.081
 enclosure-0.071
sing-0.070
 horizon-0.069
asser-0.066
Token
Token ID198
Feature activation+2.03e-01

Pos logit contributions
 band+0.010
P+0.009
 gum+0.009
 matters+0.008
 basketball+0.008
Neg logit contributions
rant-0.014
 enclosure-0.012
 horizon-0.012
sing-0.012
asser-0.012
Token
Token ID198
Feature activation+7.00e-01

Pos logit contributions
rant+0.008
 enclosure+0.007
sing+0.007
 horizon+0.007
asser+0.007
Neg logit contributions
 band-0.004
P-0.004
 gum-0.004
 basketball-0.004
 matters-0.004
TokenL
Token ID43
Feature activation-1.24e+00

Pos logit contributions
rant+0.032
 enclosure+0.030
 horizon+0.029
sing+0.029
asser+0.028
Neg logit contributions
 band-0.010
 gum-0.008
 basketball-0.007
 matters-0.007
P-0.007
Tokenily
Token ID813
Feature activation-1.30e+00

Pos logit contributions
 band+0.028
 gum+0.025
P+0.024
 basketball+0.023
 matters+0.022
Neg logit contributions
rant-0.049
 enclosure-0.044
sing-0.043
 horizon-0.043
asser-0.041
Token was
Token ID373
Feature activation+7.33e-01

Pos logit contributions
 band+0.031
P+0.030
 gum+0.029
 basketball+0.028
 mar+0.026
Neg logit contributions
rant-0.037
sing-0.032
 enclosure-0.032
 horizon-0.032
 doorstep-0.029
Token singing
Token ID13777
Feature activation+5.91e-01

Pos logit contributions
rant+0.023
sing+0.021
 horizon+0.020
 enclosure+0.020
 peak+0.018
Neg logit contributions
 band-0.022
P-0.022
 gum-0.020
 basketball-0.019
aid-0.019
Token!
Token ID0
Feature activation+2.44e-01

Pos logit contributions
rant+0.020
sing+0.018
 enclosure+0.018
 horizon+0.017
asser+0.016
Neg logit contributions
 band-0.016
P-0.015
 gum-0.015
 basketball-0.014
 matters-0.014
Token The
Token ID383
Feature activation+1.00e-01

Pos logit contributions
rant+0.007
 enclosure+0.006
 horizon+0.006
sing+0.006
 doorstep+0.005
Neg logit contributions
 band-0.008
P-0.007
 gum-0.007
 mar-0.007
 basketball-0.007
Token little
Token ID1310
Feature activation+3.76e-01

Pos logit contributions
rant+0.003
sing+0.003
 horizon+0.003
 enclosure+0.003
 doorstep+0.003
Neg logit contributions
P-0.003
 band-0.003
 matters-0.003
 gum-0.002
 basketball-0.002
Token people
Token ID661
Feature activation-2.55e+00

Pos logit contributions
 enclosure+0.011
sing+0.011
rant+0.011
 horizon+0.011
 peak+0.010
Neg logit contributions
P-0.011
 matters-0.010
 band-0.009
aid-0.009
Ol-0.009
Token were
Token ID547
Feature activation-9.69e-02

Pos logit contributions
 gum+0.067
 band+0.067
P+0.066
 basketball+0.061
 matters+0.056
Neg logit contributions
rant-0.088
 enclosure-0.078
sing-0.076
 horizon-0.075
 doorstep-0.069
Token so
Token ID523
Feature activation+2.14e-01

Pos logit contributions
 band+0.003
P+0.003
 gum+0.003
 basketball+0.003
 matters+0.003
Neg logit contributions
rant-0.003
sing-0.003
 enclosure-0.003
 horizon-0.003
 doorstep-0.002
Token happy
Token ID3772
Feature activation-4.85e-01

Pos logit contributions
rant+0.007
sing+0.007
 enclosure+0.007
 horizon+0.006
asser+0.006
Neg logit contributions
 band-0.006
P-0.005
 gum-0.005
 basketball-0.005
 matters-0.005
Token to
Token ID284
Feature activation-2.86e-01

Pos logit contributions
 band+0.014
P+0.013
 gum+0.013
 basketball+0.012
 matters+0.012
Neg logit contributions
rant-0.015
 enclosure-0.014
sing-0.014
 horizon-0.014
iously-0.013
Token see
Token ID766
Feature activation-6.84e-01

Pos logit contributions
 band+0.008
 his+0.007
 basketball+0.007
 matters+0.007
P+0.007
Neg logit contributions
rant-0.010
 enclosure-0.009
sing-0.008
 horizon-0.008
asser-0.008
Token of
Token ID286
Feature activation-2.31e+00

Pos logit contributions
P+0.001
 band+0.001
Ol+0.001
 matters+0.001
 gum+0.001
Neg logit contributions
rant-0.001
 enclosure-0.001
 horizon-0.001
asser-0.001
ples-0.001
Token bugs
Token ID11316
Feature activation-1.33e+00

Pos logit contributions
P+0.063
 band+0.061
Ol+0.061
aid+0.058
Johnny+0.057
Neg logit contributions
rant-0.060
 enclosure-0.056
sing-0.052
 horizon-0.050
asser-0.049
Token and
Token ID290
Feature activation-3.76e-01

Pos logit contributions
 band+0.037
P+0.034
 gum+0.033
 basketball+0.032
 matters+0.032
Neg logit contributions
rant-0.044
 enclosure-0.041
sing-0.039
 horizon-0.038
asser-0.036
Token small
Token ID1402
Feature activation-1.13e+00

Pos logit contributions
 band+0.010
P+0.009
 gum+0.009
 matters+0.009
 basketball+0.009
Neg logit contributions
rant-0.012
 enclosure-0.011
sing-0.011
 horizon-0.011
iously-0.010
Token swimming
Token ID14899
Feature activation+2.75e-01

Pos logit contributions
 band+0.028
P+0.025
 gum+0.025
 basketball+0.024
 matters+0.023
Neg logit contributions
rant-0.042
 enclosure-0.038
sing-0.038
 horizon-0.037
asser-0.035
Token things
Token ID1243
Feature activation-2.89e+00

Pos logit contributions
rant+0.009
 enclosure+0.008
sing+0.007
 horizon+0.007
asser+0.007
Neg logit contributions
 band-0.008
 gum-0.008
 matters-0.007
P-0.007
Ol-0.007
Token,
Token ID11
Feature activation+2.94e-01

Pos logit contributions
 band+0.083
P+0.079
 gum+0.075
 basketball+0.072
 matters+0.070
Neg logit contributions
rant-0.095
 enclosure-0.084
sing-0.082
 horizon-0.081
asser-0.076
Token but
Token ID475
Feature activation+8.61e-01

Pos logit contributions
rant+0.008
sing+0.007
 enclosure+0.007
 horizon+0.007
asser+0.006
Neg logit contributions
 band-0.010
P-0.010
 gum-0.009
 matters-0.009
 basketball-0.009
Token he
Token ID339
Feature activation-5.50e-01

Pos logit contributions
rant+0.034
sing+0.032
 enclosure+0.031
 horizon+0.031
 peak+0.029
Neg logit contributions
 band-0.019
P-0.018
 gum-0.016
aid-0.016
Ol-0.014
Token kept
Token ID4030
Feature activation-8.62e-01

Pos logit contributions
 band+0.015
P+0.014
 gum+0.014
 basketball+0.013
 matters+0.013
Neg logit contributions
rant-0.019
 enclosure-0.017
sing-0.017
 horizon-0.016
asser-0.015
Token looking
Token ID2045
Feature activation-6.48e-01

Pos logit contributions
 band+0.028
P+0.027
 gum+0.025
 matters+0.025
aid+0.025
Neg logit contributions
rant-0.024
sing-0.021
 enclosure-0.020
 horizon-0.019
iously-0.018
Token it
Token ID340
Feature activation-2.15e-01

Pos logit contributions
P+0.068
 gol+0.061
 band+0.061
 gum+0.061
 basketball+0.060
Neg logit contributions
rant-0.051
 enclosure-0.044
sing-0.041
 peak-0.039
 horizon-0.036
Token out
Token ID503
Feature activation-3.87e-01

Pos logit contributions
 band+0.006
P+0.006
 gum+0.006
 basketball+0.006
 mar+0.005
Neg logit contributions
rant-0.007
sing-0.006
 enclosure-0.006
 horizon-0.006
asser-0.005
Token.
Token ID13
Feature activation-3.56e-01

Pos logit contributions
 band+0.011
P+0.011
 gum+0.010
 basketball+0.010
 matters+0.009
Neg logit contributions
rant-0.013
 enclosure-0.012
sing-0.011
 horizon-0.011
iously-0.010
Token They
Token ID1119
Feature activation-6.77e-01

Pos logit contributions
 band+0.010
P+0.009
 gum+0.009
 basketball+0.009
 matters+0.008
Neg logit contributions
rant-0.012
 enclosure-0.011
sing-0.011
 horizon-0.011
asser-0.010
Token all
Token ID477
Feature activation-6.79e-01

Pos logit contributions
 band+0.019
 gum+0.019
P+0.018
 basketball+0.018
 matters+0.017
Neg logit contributions
rant-0.022
 enclosure-0.019
sing-0.019
 horizon-0.018
 doorstep-0.017
Token sw
Token ID1509
Feature activation-2.73e+00

Pos logit contributions
 band+0.018
 gum+0.017
P+0.017
 basketball+0.016
 matters+0.015
Neg logit contributions
rant-0.024
 enclosure-0.021
sing-0.021
 horizon-0.020
asser-0.019
Tokenam
Token ID321
Feature activation-4.43e-01

Pos logit contributions
 band+0.077
 gum+0.070
P+0.070
 basketball+0.068
 matters+0.065
Neg logit contributions
rant-0.091
 enclosure-0.082
sing-0.081
 horizon-0.078
asser-0.074
Token over
Token ID625
Feature activation+1.15e+00

Pos logit contributions
 band+0.012
P+0.011
 gum+0.011
 basketball+0.010
 matters+0.010
Neg logit contributions
rant-0.015
 enclosure-0.014
sing-0.014
 horizon-0.013
asser-0.013
Token to
Token ID284
Feature activation+4.24e-01

Pos logit contributions
rant+0.043
 enclosure+0.040
sing+0.039
 horizon+0.039
asser+0.036
Neg logit contributions
 band-0.026
P-0.025
 gum-0.023
 matters-0.023
 basketball-0.022
Token the
Token ID262
Feature activation+1.65e-01

Pos logit contributions
rant+0.020
 enclosure+0.018
sing+0.018
 horizon+0.018
asser+0.017
Neg logit contributions
 band-0.007
P-0.005
 gum-0.005
 basketball-0.005
 matters-0.005
Token flame
Token ID16506
Feature activation+3.65e-01

Pos logit contributions
rant+0.006
iously+0.005
ples+0.005
 enclosure+0.005
sing+0.005
Neg logit contributions
 band-0.005
 gum-0.004
 matters-0.004
 hockey-0.004
 his-0.004
Token and
Token ID290
Feature activation-5.06e-01

Pos logit contributions
P+0.002
 band+0.002
 gum+0.002
 basketball+0.002
Ol+0.002
Neg logit contributions
rant-0.002
 horizon-0.002
 enclosure-0.002
sing-0.002
 peak-0.002
Token the
Token ID262
Feature activation-2.99e-01

Pos logit contributions
P+0.014
 band+0.013
 basketball+0.012
 gum+0.012
Ol+0.012
Neg logit contributions
 enclosure-0.014
rant-0.014
sing-0.014
 horizon-0.013
 doorstep-0.013
Token 3
Token ID513
Feature activation-1.71e-01

Pos logit contributions
P+0.007
 band+0.007
 matters+0.006
 gum+0.006
 basketball+0.006
Neg logit contributions
rant-0.010
 enclosure-0.010
 horizon-0.010
sing-0.009
 doorstep-0.009
Token-
Token ID12
Feature activation-2.96e+00

Pos logit contributions
 band+0.006
P+0.006
 gum+0.006
 basketball+0.006
 matters+0.005
Neg logit contributions
rant-0.004
 enclosure-0.004
sing-0.004
 horizon-0.004
asser-0.003
Tokenyear
Token ID1941
Feature activation-1.46e+00

Pos logit contributions
 band+0.093
 basketball+0.092
 gum+0.090
P+0.087
 jacket+0.085
Neg logit contributions
rant-0.085
 enclosure-0.069
seven-0.065
iously-0.064
ples-0.062
Token-
Token ID12
Feature activation-3.09e+00

Pos logit contributions
 band+0.045
P+0.044
 gum+0.043
 basketball+0.042
 matters+0.041
Neg logit contributions
rant-0.042
 enclosure-0.039
 horizon-0.037
iously-0.037
sing-0.036
Tokenold
Token ID727
Feature activation-2.00e+00

Pos logit contributions
 band+0.094
 basketball+0.092
 gum+0.089
P+0.088
 jacket+0.084
Neg logit contributions
rant-0.088
 enclosure-0.079
iously-0.071
anne-0.069
 horizon-0.068
Token followed
Token ID3940
Feature activation-4.42e-01

Pos logit contributions
P+0.055
 band+0.052
 gum+0.049
 basketball+0.049
 matters+0.046
Neg logit contributions
rant-0.063
 enclosure-0.060
iously-0.055
 horizon-0.055
sing-0.055
Token him
Token ID683
Feature activation-4.30e-01

Pos logit contributions
P+0.012
 band+0.011
aid+0.011
 gum+0.010
Ol+0.010
Neg logit contributions
rant-0.014
 horizon-0.013
 enclosure-0.013
iously-0.012
 peak-0.012
Token,
Token ID11
Feature activation-1.41e-02

Pos logit contributions
 band+0.013
P+0.013
 gum+0.012
 basketball+0.011
Ol+0.011
Neg logit contributions
rant-0.013
 enclosure-0.012
sing-0.012
 horizon-0.011
 doorstep-0.011
Token laughing
Token ID14376
Feature activation+8.76e-01

Pos logit contributions
P+0.000
 band+0.000
 gum+0.000
 matters+0.000
 basketball+0.000
Neg logit contributions
rant-0.000
 enclosure-0.000
sing-0.000
 horizon-0.000
iously-0.000