TOP ACTIVATIONS
MAX = 6.672

more careful when he picked the squash and make sure only
How about we clean your old basketball and make it look
make sure only to pick the ones that were sweet and
was wrong . I wanted the big squash so much .
, " We can make the peanut butter !"
again . <|endoftext|> John loved baseball . He liked to kick
't cry , boys . Acc idents happen . I have
, so she could pick the best ones . But suddenly
so careful when he picked the squash but this time it
very enthusiastic to play with the beans . Tim
!" Lily cried . " The shrimp hurt me !"
was jealous of Johnny 's new basketball . His mom said
and Sam love to play baseball . They have a big
very careful to only pick the big , green beans with
. They like to play baseball in the park . They
saw some other kids playing baseball . He begged his friends
. He saw Sara 's rubber duck and wanted to take
M ira looked at the b erry in her hand
. Every day he would pick one from the bush in
a tiny girl named Freddy . She lived in a big

MIN ACTIVATIONS
MIN = -6.033

. At the bar ber , the bar ber
boots and climbed on the sle igh . Santa gave them
instead . The bar ber didn 't want to
nice ." The bar ber gave Tom a hug
a bar ber . The bar ber was a funny man
scared . The bar ber started to cut Tom
. The bar ber then started cl apping
mom said , " A bar ber is someone who cuts
get his haircut . The bar ber put a cape around
quiet and polite . The bar ber is busy working ."
be back again . The bar ber smiled and waved goodbye
car . The bar ber was a big man
hair ?" The bar ber was surprised and pleased
brave . The bar ber was gentle with his
the bar ber , the bar ber said that he was
off . The bar ber smiled as he watched
in the chair and the bar ber started to cut his
the bar ber . The bar ber smiled and talked to
help . The bar ber said to the little
happy ! The bar ber worked very carefully and

INTERVAL 3.495 - 6.672
CONTAINS 0.016%

, " No , mud isn 't for sharing ."
Tim my loved the idea and he and his
. <|endoftext|> Tom my loved baths . Every day , he
that even though he wanted the mint , it would have
. " You will hit the fence !" But

INTERVAL 0.319 - 3.495
CONTAINS 37.788%

day , they decided to play a game . Joe wanted
. When they arrived at grandma 's house ,
They are wet and dirty . They are sorry . They
follow . As they got closer , Jack could see his
She touched the icy water with her feet , making silly

INTERVAL -2.857 - 0.319
CONTAINS 61.880%

help . She told her friends about the lost dog and
football and Tim my played with the other kids .
. You must take good care of it ."
is a special magazine . She wants to see what is
the freezer to make it cold . Lily asked , "

INTERVAL -6.033 - -2.857
CONTAINS 0.316%

" Mom , are the pizz as ready ?" Lily asked
It was a book about history . Sophie was so proud
They had h oses and l adders . They were firefighters
something magic happened ! The m ugs split . They opened
light of the sunrise streaming into his room .
Token more
Token ID517
Feature activation+6.51e-01

Pos logit contributions
itating+0.047
arre+0.047
melon+0.045
 up+0.045
ling+0.044
Neg logit contributions
year-0.053
ird-0.050
glass-0.049
 hours-0.048
aster-0.047
Token careful
Token ID8161
Feature activation+2.15e+00

Pos logit contributions
itating+0.021
arre+0.021
melon+0.021
 up+0.020
inas+0.020
Neg logit contributions
year-0.013
 hours-0.012
ird-0.012
 minutes-0.012
glass-0.011
Token when
Token ID618
Feature activation+3.70e-01

Pos logit contributions
arre+0.047
itating+0.044
rust+0.044
melon+0.043
ucker+0.040
Neg logit contributions
year-0.066
glass-0.065
 hours-0.063
ird-0.062
 minutes-0.060
Token he
Token ID339
Feature activation+1.36e-01

Pos logit contributions
itating+0.008
arre+0.008
 up+0.008
melon+0.008
ling+0.008
Neg logit contributions
year-0.011
glass-0.011
ird-0.011
 hours-0.010
aster-0.010
Token picked
Token ID6497
Feature activation+2.96e+00

Pos logit contributions
arre+0.003
 up+0.003
melon+0.002
ling+0.002
itating+0.002
Neg logit contributions
year-0.005
glass-0.004
ird-0.004
 hours-0.004
 minutes-0.004
Token the
Token ID262
Feature activation+4.62e+00

Pos logit contributions
itating+0.066
inas+0.063
arre+0.060
ivering+0.057
lings+0.056
Neg logit contributions
year-0.086
 hours-0.085
aster-0.084
 minutes-0.080
 kil-0.079
Token squash
Token ID34613
Feature activation+1.73e+00

Pos logit contributions
inas+0.108
ocked+0.106
ouncing+0.104
ucker+0.103
ivering+0.102
Neg logit contributions
 hours-0.099
year-0.094
 month-0.093
 minutes-0.091
 20-0.091
Token and
Token ID290
Feature activation+1.99e+00

Pos logit contributions
arre+0.037
itating+0.036
ucker+0.032
rust+0.032
 fighting+0.032
Neg logit contributions
year-0.052
ird-0.051
glass-0.051
 hours-0.048
aster-0.048
Token make
Token ID787
Feature activation+1.56e+00

Pos logit contributions
itating+0.042
arre+0.041
 up+0.040
melon+0.039
ling+0.039
Neg logit contributions
year-0.063
glass-0.061
ird-0.061
 hours-0.057
aster-0.056
Token sure
Token ID1654
Feature activation+2.47e-01

Pos logit contributions
itating+0.041
arre+0.039
melon+0.038
flower+0.038
inas+0.037
Neg logit contributions
year-0.041
 hours-0.040
glass-0.039
aster-0.039
ird-0.038
Token only
Token ID691
Feature activation+5.93e-01

Pos logit contributions
arre+0.005
itating+0.005
ling+0.005
melon+0.005
bing+0.005
Neg logit contributions
year-0.008
glass-0.007
ird-0.007
 hours-0.007
aster-0.007
Token How
Token ID1374
Feature activation-2.80e+00

Pos logit contributions
ling+0.021
itating+0.019
 up+0.019
arre+0.018
melon+0.018
Neg logit contributions
year-0.027
ird-0.026
glass-0.025
ris-0.023
 hours-0.023
Token about
Token ID546
Feature activation+1.49e+00

Pos logit contributions
year+0.085
glass+0.081
ird+0.081
 hours+0.077
aster+0.075
Neg logit contributions
arre-0.063
itating-0.063
 up-0.062
melon-0.060
ling-0.059
Token we
Token ID356
Feature activation+2.09e+00

Pos logit contributions
ling+0.041
itating+0.038
flower+0.038
ivering+0.036
arre+0.036
Neg logit contributions
ird-0.037
glass-0.036
lan-0.033
 hours-0.032
 Grand-0.031
Token clean
Token ID3424
Feature activation+2.83e+00

Pos logit contributions
bing+0.067
ling+0.066
fl+0.063
ning+0.063
ping+0.063
Neg logit contributions
ird-0.044
glass-0.034
year-0.033
aster-0.032
ris-0.031
Token your
Token ID534
Feature activation+3.07e+00

Pos logit contributions
itating+0.045
lings+0.040
flower+0.039
melon+0.036
arre+0.035
Neg logit contributions
glass-0.096
year-0.093
 hours-0.092
ird-0.090
 minutes-0.088
Token old
Token ID1468
Feature activation+4.61e+00

Pos logit contributions
flower+0.082
itating+0.082
mill+0.079
ter+0.077
inas+0.073
Neg logit contributions
 hours-0.067
 customers-0.066
 allowance-0.066
 desks-0.065
glass-0.065
Token basketball
Token ID9669
Feature activation+1.62e+00

Pos logit contributions
itating+0.121
ocked+0.109
ucker+0.100
lings+0.100
inated+0.100
Neg logit contributions
ird-0.107
 hours-0.104
glass-0.104
 books-0.102
 desks-0.099
Token and
Token ID290
Feature activation+2.24e+00

Pos logit contributions
itating+0.035
arre+0.032
ling+0.032
rust+0.031
melon+0.031
Neg logit contributions
ird-0.049
glass-0.048
year-0.048
 hours-0.045
aster-0.044
Token make
Token ID787
Feature activation+1.33e+00

Pos logit contributions
itating+0.055
arre+0.052
ling+0.051
melon+0.050
flower+0.050
Neg logit contributions
ird-0.063
year-0.062
glass-0.060
 hours-0.059
 minutes-0.057
Token it
Token ID340
Feature activation+3.44e+00

Pos logit contributions
itating+0.031
melon+0.030
flower+0.029
arre+0.028
ling+0.028
Neg logit contributions
year-0.040
ird-0.038
glass-0.038
 hours-0.037
aster-0.036
Token look
Token ID804
Feature activation+9.34e-01

Pos logit contributions
itating+0.095
ican+0.093
bing+0.092
lings+0.089
flower+0.089
Neg logit contributions
ird-0.074
aster-0.067
year-0.064
 hours-0.063
glass-0.063
Token make
Token ID787
Feature activation+1.56e+00

Pos logit contributions
itating+0.042
arre+0.041
 up+0.040
melon+0.039
ling+0.039
Neg logit contributions
year-0.063
glass-0.061
ird-0.061
 hours-0.057
aster-0.056
Token sure
Token ID1654
Feature activation+2.47e-01

Pos logit contributions
itating+0.041
arre+0.039
melon+0.038
flower+0.038
inas+0.037
Neg logit contributions
year-0.041
 hours-0.040
glass-0.039
aster-0.039
ird-0.038
Token only
Token ID691
Feature activation+5.93e-01

Pos logit contributions
arre+0.005
itating+0.005
ling+0.005
melon+0.005
bing+0.005
Neg logit contributions
year-0.008
glass-0.007
ird-0.007
 hours-0.007
aster-0.007
Token to
Token ID284
Feature activation+1.58e+00

Pos logit contributions
itating+0.012
arre+0.012
ling+0.012
 up+0.011
bing+0.011
Neg logit contributions
year-0.019
ird-0.018
glass-0.018
 hours-0.017
aster-0.017
Token pick
Token ID2298
Feature activation+3.02e+00

Pos logit contributions
arre+0.041
itating+0.040
ling+0.037
bing+0.036
sets+0.036
Neg logit contributions
aster-0.042
ird-0.042
year-0.040
glass-0.038
 hours-0.038
Token the
Token ID262
Feature activation+4.60e+00

Pos logit contributions
itating+0.074
inas+0.071
arre+0.068
flower+0.065
ivering+0.064
Neg logit contributions
year-0.082
 hours-0.079
aster-0.077
 minutes-0.073
 kil-0.072
Token ones
Token ID3392
Feature activation+3.53e-01

Pos logit contributions
inas+0.106
ocked+0.104
ucker+0.098
ouncing+0.097
itating+0.096
Neg logit contributions
 hours-0.106
 minutes-0.103
 month-0.097
year-0.095
 customers-0.095
Token that
Token ID326
Feature activation+1.72e+00

Pos logit contributions
arre+0.008
itating+0.008
inas+0.007
rust+0.007
lings+0.007
Neg logit contributions
year-0.010
ird-0.010
glass-0.010
aster-0.009
 kil-0.009
Token were
Token ID547
Feature activation+2.27e+00

Pos logit contributions
itating+0.039
arre+0.039
ling+0.036
bing+0.036
 up+0.036
Neg logit contributions
year-0.051
ird-0.050
glass-0.049
 hours-0.047
aster-0.047
Token sweet
Token ID6029
Feature activation+1.51e+00

Pos logit contributions
arre+0.049
itating+0.049
ling+0.043
 up+0.043
bing+0.043
Neg logit contributions
year-0.071
ird-0.066
 hours-0.066
glass-0.065
 minutes-0.064
Token and
Token ID290
Feature activation+1.95e+00

Pos logit contributions
itating+0.035
arre+0.032
rust+0.030
bing+0.029
melon+0.029
Neg logit contributions
year-0.045
ird-0.044
 hours-0.043
 minutes-0.043
glass-0.042
Token was
Token ID373
Feature activation+2.18e+00

Pos logit contributions
itating+0.001
bing+0.001
melon+0.001
 up+0.001
sets+0.001
Neg logit contributions
year-0.001
ird-0.001
glass-0.001
ris-0.001
 hours-0.001
Token wrong
Token ID2642
Feature activation+1.66e+00

Pos logit contributions
bing+0.049
itating+0.049
sets+0.049
flower+0.046
ling+0.045
Neg logit contributions
year-0.066
ird-0.065
 minutes-0.061
aster-0.060
glass-0.059
Token.
Token ID13
Feature activation+2.96e-01

Pos logit contributions
itating+0.036
arre+0.036
melon+0.035
 up+0.034
ling+0.033
Neg logit contributions
year-0.052
ird-0.050
glass-0.050
 hours-0.047
 minutes-0.047
Token I
Token ID314
Feature activation-3.25e-01

Pos logit contributions
itating+0.006
arre+0.006
 up+0.005
melon+0.005
ling+0.005
Neg logit contributions
year-0.010
glass-0.010
ird-0.010
 hours-0.009
aster-0.009
Token wanted
Token ID2227
Feature activation+5.80e-01

Pos logit contributions
year+0.011
ird+0.010
glass+0.010
 hours+0.009
ris+0.009
Neg logit contributions
itating-0.007
arre-0.006
 up-0.006
melon-0.006
bing-0.006
Token the
Token ID262
Feature activation+4.47e+00

Pos logit contributions
itating+0.015
arre+0.014
ling+0.014
bing+0.013
melon+0.013
Neg logit contributions
year-0.016
glass-0.015
ird-0.015
 hours-0.015
aster-0.015
Token big
Token ID1263
Feature activation+3.15e+00

Pos logit contributions
bing+0.120
itating+0.112
ocked+0.111
inas+0.111
en+0.105
Neg logit contributions
 passport-0.096
 minutes-0.090
 hours-0.090
 delivery-0.089
 month-0.088
Token squash
Token ID34613
Feature activation+1.93e+00

Pos logit contributions
itating+0.075
ocked+0.066
inas+0.066
bing+0.062
arre+0.059
Neg logit contributions
 minutes-0.093
 hours-0.084
aster-0.084
ird-0.081
 hour-0.080
Token so
Token ID523
Feature activation-2.10e-01

Pos logit contributions
itating+0.042
arre+0.041
ling+0.038
bing+0.038
 up+0.037
Neg logit contributions
year-0.059
ird-0.058
glass-0.057
 hours-0.054
aster-0.053
Token much
Token ID881
Feature activation-7.09e-01

Pos logit contributions
year+0.004
ird+0.004
glass+0.004
 hours+0.004
aster+0.003
Neg logit contributions
arre-0.007
itating-0.007
 up-0.007
melon-0.007
ling-0.007
Token.
Token ID13
Feature activation-5.92e-02

Pos logit contributions
year+0.024
glass+0.023
ird+0.023
 hours+0.022
aster+0.021
Neg logit contributions
itating-0.014
arre-0.014
 up-0.013
melon-0.013
ling-0.013
Token,
Token ID11
Feature activation+6.05e-01

Pos logit contributions
year+0.002
ird+0.002
glass+0.002
 hours+0.002
 minutes+0.002
Neg logit contributions
itating-0.002
arre-0.002
rust-0.001
ling-0.001
bing-0.001
Token "
Token ID366
Feature activation+1.50e+00

Pos logit contributions
arre+0.018
rust+0.017
itating+0.016
ling+0.016
ucker+0.016
Neg logit contributions
year-0.014
glass-0.013
ird-0.012
 hours-0.012
 minutes-0.012
TokenWe
Token ID1135
Feature activation+6.34e-01

Pos logit contributions
 up+0.034
arre+0.031
ling+0.031
 fighting+0.031
melon+0.030
Neg logit contributions
year-0.047
glass-0.045
ird-0.043
Ah-0.042
 hours-0.042
Token can
Token ID460
Feature activation+1.97e+00

Pos logit contributions
itating+0.017
bing+0.016
 up+0.015
ling+0.015
mill+0.015
Neg logit contributions
ird-0.017
year-0.016
glass-0.015
 hours-0.015
ris-0.014
Token make
Token ID787
Feature activation+2.48e+00

Pos logit contributions
arre+0.052
rust+0.050
itating+0.050
ling+0.049
 accidents+0.048
Neg logit contributions
ird-0.049
aster-0.045
lock-0.043
glass-0.042
ris-0.041
Token the
Token ID262
Feature activation+4.45e+00

Pos logit contributions
itating+0.070
flower+0.062
arre+0.061
inas+0.060
mill+0.060
Neg logit contributions
 hours-0.061
glass-0.061
ird-0.061
year-0.060
aster-0.060
Token peanut
Token ID26636
Feature activation+1.67e+00

Pos logit contributions
itating+0.127
ocked+0.115
bing+0.115
mill+0.114
inated+0.107
Neg logit contributions
 hours-0.109
glass-0.108
lan-0.098
ird-0.097
 minutes-0.097
Token butter
Token ID9215
Feature activation+2.01e+00

Pos logit contributions
itating+0.045
arre+0.040
 Sharing+0.039
 Fighting+0.039
Fighting+0.039
Neg logit contributions
ird-0.042
glass-0.040
year-0.040
aster-0.039
 kil-0.039
Token!"
Token ID2474
Feature activation+1.29e+00

Pos logit contributions
itating+0.052
arre+0.050
rust+0.048
 fighting+0.046
ocked+0.046
Neg logit contributions
ird-0.049
glass-0.048
year-0.047
aster-0.046
book-0.044
Token
Token ID220
Feature activation+1.88e-02

Pos logit contributions
ling+0.031
itating+0.030
bing+0.029
arre+0.029
ivering+0.028
Neg logit contributions
glass-0.037
year-0.035
 hours-0.033
ird-0.033
lock-0.032
Token
Token ID198
Feature activation-5.78e-01

Pos logit contributions
ling+0.000
arre+0.000
melon+0.000
itating+0.000
flower+0.000
Neg logit contributions
year-0.001
ird-0.001
glass-0.001
ibr-0.000
aster-0.000
Token again
Token ID757
Feature activation+8.94e-01

Pos logit contributions
itating+0.047
arre+0.046
melon+0.043
rust+0.042
ling+0.041
Neg logit contributions
year-0.086
ird-0.082
glass-0.082
 hours-0.079
 minutes-0.077
Token.
Token ID13
Feature activation+3.07e-01

Pos logit contributions
 up+0.013
ling+0.013
bing+0.012
itating+0.012
arre+0.012
Neg logit contributions
year-0.036
glass-0.033
ird-0.033
ibr-0.032
ibrarian-0.032
Token<|endoftext|>
Token ID50256
Feature activation+5.06e-01

Pos logit contributions
itating+0.006
arre+0.006
 up+0.005
melon+0.005
ling+0.005
Neg logit contributions
year-0.011
ird-0.010
glass-0.010
 hours-0.010
aster-0.010
TokenJohn
Token ID7554
Feature activation+8.66e-01

Pos logit contributions
itating+0.009
arre+0.009
 up+0.009
melon+0.008
ling+0.008
Neg logit contributions
year-0.018
ird-0.017
glass-0.017
 hours-0.016
aster-0.016
Token loved
Token ID6151
Feature activation+3.21e-01

Pos logit contributions
arre+0.015
itating+0.014
 up+0.014
melon+0.014
ling+0.014
Neg logit contributions
year-0.031
glass-0.030
ird-0.030
 hours-0.028
aster-0.028
Token baseball
Token ID9283
Feature activation+4.38e+00

Pos logit contributions
itating+0.008
arre+0.008
ling+0.007
 up+0.007
melon+0.007
Neg logit contributions
year-0.009
glass-0.009
ird-0.009
 hours-0.008
aster-0.008
Token.
Token ID13
Feature activation+2.51e-01

Pos logit contributions
itating+0.132
arre+0.124
ucker+0.119
inated+0.115
aining+0.115
Neg logit contributions
ird-0.092
year-0.091
glass-0.089
 hours-0.088
aster-0.081
Token He
Token ID679
Feature activation+2.15e+00

Pos logit contributions
arre+0.004
itating+0.004
 up+0.004
melon+0.004
ling+0.004
Neg logit contributions
year-0.009
glass-0.009
ird-0.008
 hours-0.008
aster-0.008
Token liked
Token ID8288
Feature activation+2.13e+00

Pos logit contributions
itating+0.053
arre+0.052
 up+0.049
melon+0.048
ling+0.048
Neg logit contributions
year-0.062
glass-0.059
ird-0.059
 hours-0.058
 minutes-0.055
Token to
Token ID284
Feature activation+1.55e+00

Pos logit contributions
itating+0.072
arre+0.068
ling+0.066
ivering+0.066
flower+0.065
Neg logit contributions
 hours-0.041
glass-0.039
year-0.038
aster-0.038
 minutes-0.037
Token kick
Token ID4829
Feature activation+2.73e+00

Pos logit contributions
itating+0.040
arre+0.039
ling+0.036
 accidents+0.035
bing+0.034
Neg logit contributions
aster-0.043
ird-0.043
glass-0.042
year-0.041
 hours-0.041
Token't
Token ID470
Feature activation+1.08e-01

Pos logit contributions
arre+0.017
melon+0.016
itating+0.015
 fighting+0.015
ling+0.015
Neg logit contributions
ird-0.032
year-0.032
glass-0.031
aster-0.030
lock-0.030
Token cry
Token ID3960
Feature activation+1.20e+00

Pos logit contributions
arre+0.003
melon+0.003
itating+0.003
ucker+0.003
rooms+0.003
Neg logit contributions
ird-0.003
aster-0.002
lock-0.002
year-0.002
glass-0.002
Token,
Token ID11
Feature activation+1.02e+00

Pos logit contributions
arre+0.031
itating+0.031
rust+0.031
melon+0.030
rooms+0.027
Neg logit contributions
ird-0.030
year-0.029
 minutes-0.028
glass-0.028
 hours-0.027
Token boys
Token ID6510
Feature activation+2.51e+00

Pos logit contributions
itating+0.019
arre+0.019
 up+0.018
melon+0.018
ling+0.018
Neg logit contributions
year-0.035
ird-0.033
glass-0.033
 hours-0.032
aster-0.031
Token.
Token ID13
Feature activation+1.15e+00

Pos logit contributions
itating+0.061
arre+0.059
rust+0.054
melon+0.054
bing+0.054
Neg logit contributions
ird-0.073
year-0.070
glass-0.068
 minutes-0.066
 hours-0.066
Token Acc
Token ID6366
Feature activation+4.37e+00

Pos logit contributions
arre+0.026
itating+0.026
ling+0.026
 up+0.025
bing+0.024
Neg logit contributions
year-0.034
ird-0.033
glass-0.033
aster-0.030
ibr-0.029
Tokenidents
Token ID3231
Feature activation+1.38e+00

Pos logit contributions
itating+0.077
rooms+0.073
melon+0.073
 Freddy+0.072
urtle+0.070
Neg logit contributions
glass-0.135
ber-0.127
bers-0.127
ird-0.126
lan-0.121
Token happen
Token ID1645
Feature activation+3.38e+00

Pos logit contributions
rust+0.038
flower+0.037
ican+0.035
ishly+0.035
ucker+0.035
Neg logit contributions
ird-0.031
book-0.030
year-0.030
 minutes-0.029
 hour-0.029
Token.
Token ID13
Feature activation+1.28e+00

Pos logit contributions
itating+0.099
arre+0.096
melon+0.094
ican+0.091
rust+0.090
Neg logit contributions
 minutes-0.078
 hours-0.074
year-0.070
glass-0.070
ird-0.069
Token I
Token ID314
Feature activation+4.23e-01

Pos logit contributions
arre+0.030
itating+0.030
ling+0.029
 up+0.029
melon+0.028
Neg logit contributions
year-0.038
ird-0.037
glass-0.036
 minutes-0.033
 hours-0.033
Token have
Token ID423
Feature activation+8.97e-01

Pos logit contributions
itating+0.011
bing+0.011
melon+0.011
 up+0.011
arre+0.011
Neg logit contributions
year-0.011
ird-0.010
glass-0.010
ris-0.009
 hours-0.009
Token,
Token ID11
Feature activation+1.88e+00

Pos logit contributions
arre+0.015
itating+0.015
ling+0.014
 up+0.014
rust+0.014
Neg logit contributions
year-0.025
glass-0.024
ird-0.024
 hours-0.023
 minutes-0.022
Token so
Token ID523
Feature activation-3.61e-01

Pos logit contributions
arre+0.047
rust+0.044
itating+0.044
rooms+0.042
urtle+0.040
Neg logit contributions
 hours-0.052
 minutes-0.051
year-0.050
aster-0.046
glass-0.046
Token she
Token ID673
Feature activation+5.77e-01

Pos logit contributions
year+0.015
glass+0.014
ird+0.014
ris+0.013
 kil+0.013
Neg logit contributions
 up-0.005
itating-0.005
ling-0.004
bing-0.004
arre-0.004
Token could
Token ID714
Feature activation+1.70e+00

Pos logit contributions
itating+0.012
arre+0.012
 up+0.012
melon+0.012
ling+0.012
Neg logit contributions
year-0.018
glass-0.017
ird-0.017
 hours-0.016
aster-0.016
Token pick
Token ID2298
Feature activation+3.51e+00

Pos logit contributions
arre+0.047
ling+0.045
itating+0.045
bing+0.044
mill+0.044
Neg logit contributions
ird-0.042
aster-0.041
year-0.041
glass-0.039
 hours-0.038
Token the
Token ID262
Feature activation+4.34e+00

Pos logit contributions
itating+0.098
arre+0.092
ivering+0.092
bing+0.089
mill+0.087
Neg logit contributions
year-0.085
 hours-0.082
aster-0.077
 minutes-0.074
ird-0.073
Token best
Token ID1266
Feature activation+2.61e+00

Pos logit contributions
mill+0.112
itating+0.111
bing+0.108
arre+0.100
ling+0.099
Neg logit contributions
 hours-0.107
 minutes-0.103
ird-0.102
year-0.097
ns-0.095
Token ones
Token ID3392
Feature activation+7.62e-01

Pos logit contributions
itating+0.079
bing+0.072
mill+0.067
Fighting+0.065
arre+0.065
Neg logit contributions
 hours-0.062
 minutes-0.062
ird-0.058
aster-0.055
year-0.055
Token.
Token ID13
Feature activation+3.93e-01

Pos logit contributions
arre+0.017
itating+0.016
ling+0.015
rust+0.015
bing+0.015
Neg logit contributions
year-0.023
ird-0.023
glass-0.022
 hours-0.021
 minutes-0.021
Token But
Token ID887
Feature activation+9.95e-01

Pos logit contributions
arre+0.009
melon+0.009
itating+0.009
ling+0.009
rust+0.008
Neg logit contributions
year-0.012
ird-0.011
 hours-0.011
glass-0.011
 minutes-0.010
Token suddenly
Token ID6451
Feature activation+1.04e+00

Pos logit contributions
 up+0.019
itating+0.019
arre+0.019
melon+0.018
ling+0.018
Neg logit contributions
year-0.034
glass-0.032
ird-0.032
 hours-0.030
aster-0.030
Token so
Token ID523
Feature activation-1.14e-01

Pos logit contributions
itating+0.007
arre+0.007
 up+0.007
melon+0.007
ling+0.007
Neg logit contributions
year-0.011
glass-0.011
ird-0.011
 hours-0.010
aster-0.010
Token careful
Token ID8161
Feature activation+2.05e+00

Pos logit contributions
year+0.004
glass+0.004
ird+0.004
 hours+0.003
aster+0.003
Neg logit contributions
arre-0.002
itating-0.002
 up-0.002
melon-0.002
ling-0.002
Token when
Token ID618
Feature activation+1.10e-01

Pos logit contributions
arre+0.043
rust+0.042
itating+0.040
melon+0.039
 Bugs+0.037
Neg logit contributions
year-0.064
glass-0.062
ird-0.060
aster-0.058
lock-0.058
Token he
Token ID339
Feature activation+2.55e-01

Pos logit contributions
 up+0.002
itating+0.002
arre+0.002
melon+0.002
 fighting+0.002
Neg logit contributions
year-0.004
ird-0.003
glass-0.003
 hours-0.003
aster-0.003
Token picked
Token ID6497
Feature activation+2.41e+00

Pos logit contributions
 up+0.005
arre+0.005
melon+0.004
ling+0.004
 fighting+0.004
Neg logit contributions
year-0.009
glass-0.009
ird-0.008
 hours-0.008
ibr-0.008
Token the
Token ID262
Feature activation+4.34e+00

Pos logit contributions
itating+0.055
arre+0.052
inas+0.051
lings+0.048
bing+0.047
Neg logit contributions
year-0.070
aster-0.069
 hours-0.068
glass-0.065
 kil-0.064
Token squash
Token ID34613
Feature activation+1.67e+00

Pos logit contributions
inas+0.101
ocked+0.100
ucker+0.099
ouncing+0.098
ivering+0.094
Neg logit contributions
 hours-0.094
year-0.090
 20-0.090
lock-0.087
 month-0.087
Token but
Token ID475
Feature activation+1.04e+00

Pos logit contributions
arre+0.037
itating+0.036
rust+0.034
ucker+0.033
inas+0.033
Neg logit contributions
year-0.048
ird-0.046
glass-0.046
aster-0.045
lock-0.044
Token this
Token ID428
Feature activation+1.94e+00

Pos logit contributions
itating+0.021
arre+0.021
 up+0.020
ling+0.020
melon+0.020
Neg logit contributions
year-0.034
ird-0.032
glass-0.032
 hours-0.031
aster-0.030
Token time
Token ID640
Feature activation-8.48e-01

Pos logit contributions
itating+0.053
arre+0.049
bing+0.048
inas+0.047
rust+0.045
Neg logit contributions
 hours-0.050
year-0.049
ird-0.049
 minutes-0.049
aster-0.048
Token it
Token ID340
Feature activation+1.99e+00

Pos logit contributions
year+0.032
glass+0.031
ird+0.030
aster+0.029
 Bible+0.029
Neg logit contributions
 up-0.014
 fighting-0.012
 Freddy-0.012
melon-0.012
arre-0.011
Token very
Token ID845
Feature activation+9.92e-01

Pos logit contributions
itating+0.042
arre+0.041
mill+0.037
ling+0.036
flower+0.036
Neg logit contributions
year-0.044
 hours-0.041
glass-0.039
 minutes-0.039
ird-0.038
Token enthusiastic
Token ID21848
Feature activation+1.67e+00

Pos logit contributions
itating+0.022
arre+0.022
 up+0.021
melon+0.021
ling+0.020
Neg logit contributions
year-0.031
glass-0.030
ird-0.030
 hours-0.028
aster-0.028
Token to
Token ID284
Feature activation+6.55e-01

Pos logit contributions
itating+0.042
arre+0.039
rust+0.037
melon+0.037
ling+0.036
Neg logit contributions
year-0.049
 hours-0.047
glass-0.047
 minutes-0.045
ird-0.045
Token play
Token ID711
Feature activation+2.19e+00

Pos logit contributions
itating+0.015
arre+0.015
 up+0.014
ling+0.014
melon+0.014
Neg logit contributions
year-0.020
ird-0.019
glass-0.019
 hours-0.018
aster-0.018
Token with
Token ID351
Feature activation+1.83e+00

Pos logit contributions
itating+0.052
arre+0.047
inas+0.046
melon+0.046
ucker+0.046
Neg logit contributions
glass-0.067
 hours-0.066
year-0.066
 minutes-0.062
ird-0.062
Token the
Token ID262
Feature activation+4.33e+00

Pos logit contributions
itating+0.035
arre+0.034
inas+0.029
melon+0.029
flower+0.029
Neg logit contributions
year-0.062
glass-0.061
 hours-0.060
ird-0.060
aster-0.057
Token beans
Token ID16567
Feature activation+1.44e+00

Pos logit contributions
mill+0.137
itating+0.133
ican+0.126
bing+0.125
ocked+0.124
Neg logit contributions
 hours-0.092
 props-0.085
 minutes-0.082
 servants-0.070
 chairs-0.069
Token.
Token ID13
Feature activation+1.55e+00

Pos logit contributions
arre+0.033
itating+0.033
rust+0.030
bing+0.029
ucker+0.029
Neg logit contributions
year-0.041
ird-0.041
glass-0.040
 hours-0.039
 minutes-0.038
Token
Token ID198
Feature activation-3.64e-01

Pos logit contributions
ling+0.048
itating+0.046
arre+0.045
flower+0.045
mill+0.044
Neg logit contributions
ird-0.031
glass-0.030
year-0.029
ris-0.029
 hours-0.029
Token
Token ID198
Feature activation+2.01e+00

Pos logit contributions
year+0.013
glass+0.012
ird+0.012
 hours+0.011
aster+0.011
Neg logit contributions
itating-0.007
arre-0.007
 up-0.006
melon-0.006
ling-0.006
TokenTim
Token ID14967
Feature activation+1.53e+00

Pos logit contributions
melon+0.054
itating+0.054
arre+0.053
sets+0.053
flower+0.052
Neg logit contributions
year-0.051
glass-0.048
 hours-0.047
ird-0.047
ris-0.045
Token!"
Token ID2474
Feature activation-3.59e-01

Pos logit contributions
itating+0.011
arre+0.011
melon+0.011
 up+0.010
ling+0.010
Neg logit contributions
year-0.014
ird-0.014
glass-0.014
 hours-0.013
aster-0.013
Token Lily
Token ID20037
Feature activation-3.79e-01

Pos logit contributions
year+0.016
glass+0.015
ird+0.015
 kil+0.015
ris+0.015
Neg logit contributions
 up-0.004
 fighting-0.004
 Freddy-0.003
 accidents-0.003
 Polly-0.003
Token cried
Token ID16896
Feature activation-1.12e+00

Pos logit contributions
year+0.012
glass+0.011
ird+0.011
 hours+0.011
aster+0.011
Neg logit contributions
 up-0.008
arre-0.008
itating-0.008
 fighting-0.008
ling-0.008
Token.
Token ID13
Feature activation+5.43e-02

Pos logit contributions
year+0.040
glass+0.039
ird+0.039
 hours+0.037
aster+0.036
Neg logit contributions
 up-0.020
itating-0.019
arre-0.019
ling-0.018
 fighting-0.018
Token "
Token ID366
Feature activation+7.80e-01

Pos logit contributions
arre+0.001
ling+0.001
rust+0.001
melon+0.001
itating+0.001
Neg logit contributions
year-0.002
glass-0.002
ird-0.002
 minutes-0.002
lock-0.002
TokenThe
Token ID464
Feature activation+4.29e+00

Pos logit contributions
melon+0.016
arre+0.016
 up+0.015
itating+0.015
ling+0.015
Neg logit contributions
year-0.025
glass-0.025
aster-0.024
ird-0.024
 hours-0.023
Token shrimp
Token ID30060
Feature activation+8.22e-01

Pos logit contributions
itating+0.124
bing+0.117
ning+0.115
mill+0.115
arre+0.114
Neg logit contributions
glass-0.094
 passport-0.091
 minutes-0.091
 hours-0.090
 couple-0.090
Token hurt
Token ID5938
Feature activation+1.26e+00

Pos logit contributions
itating+0.017
arre+0.017
rust+0.016
inas+0.015
bing+0.015
Neg logit contributions
ird-0.026
year-0.025
glass-0.025
 hours-0.024
 minutes-0.024
Token me
Token ID502
Feature activation-5.61e-01

Pos logit contributions
itating+0.021
arre+0.020
melon+0.019
ling+0.018
 up+0.018
Neg logit contributions
year-0.047
ird-0.045
glass-0.045
 hours-0.043
aster-0.043
Token!"
Token ID2474
Feature activation+5.90e-02

Pos logit contributions
year+0.018
glass+0.017
ird+0.017
 hours+0.017
aster+0.016
Neg logit contributions
itating-0.012
arre-0.011
 up-0.011
ling-0.011
melon-0.011
Token
Token ID198
Feature activation-5.31e-01

Pos logit contributions
arre+0.001
itating+0.001
 up+0.001
melon+0.001
ling+0.001
Neg logit contributions
year-0.002
glass-0.002
ird-0.002
 hours-0.002
aster-0.002
Token was
Token ID373
Feature activation+1.70e+00

Pos logit contributions
itating+0.024
bing+0.022
melon+0.022
arre+0.022
mill+0.021
Neg logit contributions
year-0.026
ird-0.025
glass-0.024
 hours-0.022
ris-0.022
Token jealous
Token ID19354
Feature activation+1.72e+00

Pos logit contributions
fl+0.037
itating+0.037
flower+0.036
rooms+0.036
bing+0.036
Neg logit contributions
year-0.052
ird-0.051
 hours-0.044
 minutes-0.044
glass-0.043
Token of
Token ID286
Feature activation-2.46e-01

Pos logit contributions
itating+0.059
bing+0.056
rust+0.056
melon+0.056
flower+0.056
Neg logit contributions
ird-0.031
year-0.031
 minutes-0.029
glass-0.028
 hours-0.027
Token Johnny
Token ID15470
Feature activation+1.34e+00

Pos logit contributions
year+0.008
glass+0.007
ird+0.007
 hours+0.007
aster+0.007
Neg logit contributions
itating-0.005
arre-0.005
ling-0.005
melon-0.005
 up-0.005
Token's
Token ID338
Feature activation+2.87e+00

Pos logit contributions
itating+0.030
arre+0.028
rust+0.027
melon+0.026
flower+0.026
Neg logit contributions
ird-0.041
year-0.039
glass-0.039
aster-0.037
 hours-0.036
Token new
Token ID649
Feature activation+4.27e+00

Pos logit contributions
itating+0.070
flower+0.068
ter+0.065
lings+0.065
mill+0.063
Neg logit contributions
 hours-0.065
 allowance-0.064
ird-0.061
glass-0.059
year-0.059
Token basketball
Token ID9669
Feature activation+1.47e+00

Pos logit contributions
itating+0.118
ping+0.109
ric+0.105
ucker+0.105
lings+0.103
Neg logit contributions
 hours-0.095
ird-0.089
 minutes-0.083
 books-0.081
 customers-0.080
Token.
Token ID13
Feature activation+1.38e+00

Pos logit contributions
itating+0.029
arre+0.028
 up+0.027
melon+0.027
ling+0.027
Neg logit contributions
year-0.049
ird-0.047
glass-0.047
 hours-0.045
aster-0.044
Token His
Token ID2399
Feature activation+1.01e+00

Pos logit contributions
itating+0.035
ling+0.034
arre+0.033
melon+0.032
flower+0.031
Neg logit contributions
year-0.038
ird-0.037
glass-0.036
 hours-0.034
aster-0.033
Token mom
Token ID1995
Feature activation+7.08e-01

Pos logit contributions
itating+0.034
arre+0.032
ling+0.032
melon+0.031
bing+0.031
Neg logit contributions
glass-0.020
ird-0.019
year-0.019
 hours-0.018
lock-0.016
Token said
Token ID531
Feature activation+4.88e-01

Pos logit contributions
itating+0.020
bing+0.019
melon+0.019
arre+0.019
sets+0.018
Neg logit contributions
year-0.018
ird-0.018
glass-0.016
 hours-0.015
ibrarian-0.015
Token and
Token ID290
Feature activation+8.97e-01

Pos logit contributions
itating+0.027
arre+0.025
bing+0.023
ling+0.023
melon+0.022
Neg logit contributions
year-0.045
ird-0.044
glass-0.044
aster-0.042
 hours-0.042
Token Sam
Token ID3409
Feature activation+2.42e+00

Pos logit contributions
itating+0.019
arre+0.019
 up+0.018
melon+0.018
ling+0.018
Neg logit contributions
year-0.029
glass-0.027
ird-0.027
 hours-0.026
aster-0.025
Token love
Token ID1842
Feature activation+5.41e-01

Pos logit contributions
itating+0.067
arre+0.061
bing+0.060
ocked+0.057
 accidents+0.056
Neg logit contributions
ird-0.063
year-0.062
aster-0.060
 hours-0.060
glass-0.060
Token to
Token ID284
Feature activation+7.77e-01

Pos logit contributions
 up+0.011
melon+0.011
arre+0.010
 fighting+0.010
itating+0.010
Neg logit contributions
year-0.018
ird-0.018
glass-0.017
 hours-0.016
 minutes-0.016
Token play
Token ID711
Feature activation+1.76e+00

Pos logit contributions
itating+0.018
arre+0.018
ling+0.018
 up+0.017
bing+0.016
Neg logit contributions
year-0.022
ird-0.022
glass-0.022
aster-0.022
 hours-0.021
Token baseball
Token ID9283
Feature activation+4.27e+00

Pos logit contributions
itating+0.036
arre+0.035
melon+0.033
ling+0.032
inas+0.031
Neg logit contributions
year-0.057
glass-0.057
ird-0.056
 hours-0.054
aster-0.052
Token.
Token ID13
Feature activation+1.22e+00

Pos logit contributions
itating+0.122
arre+0.117
ucker+0.111
inas+0.107
 accidents+0.105
Neg logit contributions
glass-0.102
year-0.102
 hours-0.097
ird-0.095
aster-0.092
Token They
Token ID1119
Feature activation+2.37e+00

Pos logit contributions
itating+0.022
arre+0.021
ling+0.020
 up+0.020
melon+0.019
Neg logit contributions
year-0.043
ird-0.042
glass-0.042
 hours-0.040
aster-0.040
Token have
Token ID423
Feature activation+1.63e+00

Pos logit contributions
itating+0.060
arre+0.059
bing+0.055
ling+0.055
 up+0.055
Neg logit contributions
year-0.065
ird-0.064
glass-0.063
 hours-0.062
aster-0.060
Token a
Token ID257
Feature activation+8.98e-01

Pos logit contributions
arre+0.040
itating+0.039
 up+0.039
melon+0.038
ling+0.037
Neg logit contributions
year-0.047
ird-0.044
glass-0.044
 hours-0.041
aster-0.041
Token big
Token ID1263
Feature activation+1.55e+00

Pos logit contributions
itating+0.022
arre+0.021
bing+0.020
ling+0.020
 up+0.020
Neg logit contributions
year-0.025
ird-0.025
glass-0.024
 hours-0.024
 minutes-0.023
Token very
Token ID845
Feature activation+5.63e-02

Pos logit contributions
arre+0.002
itating+0.002
melon+0.002
ling+0.002
 up+0.002
Neg logit contributions
year-0.002
ird-0.002
glass-0.002
 hours-0.002
 minutes-0.002
Token careful
Token ID8161
Feature activation+2.34e+00

Pos logit contributions
itating+0.001
arre+0.001
 up+0.001
melon+0.001
ling+0.001
Neg logit contributions
year-0.002
ird-0.002
glass-0.002
 hours-0.002
aster-0.002
Token to
Token ID284
Feature activation+1.83e-01

Pos logit contributions
rust+0.053
arre+0.052
melon+0.048
itating+0.048
rooms+0.045
Neg logit contributions
year-0.069
glass-0.068
ird-0.065
 hours-0.065
aster-0.065
Token only
Token ID691
Feature activation+1.04e+00

Pos logit contributions
arre+0.005
itating+0.004
ling+0.004
melon+0.004
mill+0.004
Neg logit contributions
ird-0.005
aster-0.005
year-0.005
glass-0.005
 hours-0.005
Token pick
Token ID2298
Feature activation+3.21e+00

Pos logit contributions
itating+0.023
arre+0.022
ling+0.021
 up+0.021
bing+0.020
Neg logit contributions
year-0.032
ird-0.032
glass-0.031
aster-0.030
 hours-0.030
Token the
Token ID262
Feature activation+4.26e+00

Pos logit contributions
itating+0.074
arre+0.068
mill+0.065
ivering+0.064
inas+0.062
Neg logit contributions
 hours-0.091
year-0.090
 minutes-0.083
 Africa-0.082
 kil-0.081
Token big
Token ID1263
Feature activation+3.34e+00

Pos logit contributions
mill+0.144
itating+0.137
ocked+0.135
ican+0.125
ric+0.125
Neg logit contributions
 hours-0.072
 props-0.071
 minutes-0.068
 clues-0.063
 chairs-0.059
Token,
Token ID11
Feature activation+5.63e-01

Pos logit contributions
itating+0.102
mill+0.088
ocked+0.086
arre+0.085
Fighting+0.083
Neg logit contributions
 minutes-0.082
 hours-0.078
 seconds-0.065
 props-0.065
 kil-0.064
Token green
Token ID4077
Feature activation+3.16e+00

Pos logit contributions
itating+0.011
arre+0.011
 up+0.010
ling+0.010
melon+0.010
Neg logit contributions
year-0.019
ird-0.018
glass-0.018
 hours-0.018
 minutes-0.018
Token beans
Token ID16567
Feature activation+7.03e-01

Pos logit contributions
itating+0.093
arre+0.080
mill+0.076
ucker+0.073
Fighting+0.073
Neg logit contributions
 hours-0.082
 minutes-0.074
glass-0.069
year-0.068
 props-0.067
Token with
Token ID351
Feature activation+1.17e+00

Pos logit contributions
arre+0.016
itating+0.016
rust+0.016
ling+0.014
inas+0.014
Neg logit contributions
year-0.020
ird-0.020
glass-0.020
 hours-0.019
 minutes-0.018
Token.
Token ID13
Feature activation+2.04e+00

Pos logit contributions
arre+0.032
itating+0.032
 up+0.029
melon+0.029
ling+0.029
Neg logit contributions
year-0.038
ird-0.037
glass-0.037
 hours-0.035
aster-0.034
Token They
Token ID1119
Feature activation+5.41e-01

Pos logit contributions
itating+0.034
arre+0.033
 up+0.032
ling+0.031
melon+0.030
Neg logit contributions
year-0.076
ird-0.073
glass-0.072
 hours-0.069
aster-0.068
Token like
Token ID588
Feature activation-3.82e-01

Pos logit contributions
itating+0.013
arre+0.013
 up+0.012
melon+0.012
ling+0.012
Neg logit contributions
year-0.016
glass-0.015
ird-0.015
 hours-0.014
aster-0.014
Token to
Token ID284
Feature activation+2.64e-02

Pos logit contributions
year+0.013
ird+0.012
glass+0.012
 hours+0.011
 minutes+0.011
Neg logit contributions
 up-0.008
arre-0.008
itating-0.007
melon-0.007
 fighting-0.007
Token play
Token ID711
Feature activation+1.50e+00

Pos logit contributions
 up+0.000
arre+0.000
melon+0.000
itating+0.000
 fighting+0.000
Neg logit contributions
year-0.001
glass-0.001
ird-0.001
 hours-0.001
 minutes-0.001
Token baseball
Token ID9283
Feature activation+4.24e+00

Pos logit contributions
itating+0.029
arre+0.028
melon+0.027
 up+0.027
ling+0.026
Neg logit contributions
year-0.051
glass-0.049
ird-0.049
 hours-0.047
aster-0.046
Token in
Token ID287
Feature activation+6.10e-01

Pos logit contributions
itating+0.109
arre+0.102
ucker+0.098
inas+0.094
 accidents+0.093
Neg logit contributions
glass-0.115
 hours-0.110
year-0.110
ird-0.105
aster-0.102
Token the
Token ID262
Feature activation+1.93e+00

Pos logit contributions
arre+0.014
itating+0.014
ling+0.013
melon+0.013
 up+0.012
Neg logit contributions
year-0.019
glass-0.018
ird-0.018
 hours-0.018
aster-0.018
Token park
Token ID3952
Feature activation+1.77e+00

Pos logit contributions
itating+0.052
ling+0.047
bing+0.047
ocked+0.047
arre+0.046
Neg logit contributions
glass-0.048
aster-0.048
 hours-0.046
ird-0.046
 minutes-0.045
Token.
Token ID13
Feature activation+1.11e+00

Pos logit contributions
 up+0.029
itating+0.027
melon+0.027
ling+0.026
 fighting+0.026
Neg logit contributions
year-0.067
ird-0.064
glass-0.063
 hours-0.060
aster-0.060
Token They
Token ID1119
Feature activation+1.91e+00

Pos logit contributions
itating+0.024
arre+0.023
ling+0.023
 up+0.022
bing+0.022
Neg logit contributions
year-0.035
ird-0.034
glass-0.034
 hours-0.032
aster-0.032
Token saw
Token ID2497
Feature activation+4.42e-01

Pos logit contributions
arre+0.010
 up+0.010
itating+0.009
melon+0.009
ling+0.009
Neg logit contributions
year-0.017
glass-0.016
ird-0.016
 hours-0.015
aster-0.015
Token some
Token ID617
Feature activation+6.23e-01

Pos logit contributions
arre+0.010
itating+0.010
 up+0.010
melon+0.010
ling+0.010
Neg logit contributions
year-0.013
ird-0.013
glass-0.013
 hours-0.012
aster-0.012
Token other
Token ID584
Feature activation+7.80e-01

Pos logit contributions
itating+0.014
arre+0.014
melon+0.014
bing+0.013
 up+0.013
Neg logit contributions
year-0.019
ird-0.018
glass-0.018
 hours-0.017
 minutes-0.017
Token kids
Token ID3988
Feature activation+1.64e+00

Pos logit contributions
itating+0.018
arre+0.017
melon+0.017
 up+0.016
ling+0.016
Neg logit contributions
year-0.024
ird-0.023
glass-0.023
 hours-0.022
 minutes-0.021
Token playing
Token ID2712
Feature activation+2.41e+00

Pos logit contributions
itating+0.035
rust+0.035
arre+0.034
aining+0.034
adoes+0.034
Neg logit contributions
ird-0.050
year-0.047
glass-0.046
 hours-0.042
aster-0.042
Token baseball
Token ID9283
Feature activation+4.24e+00

Pos logit contributions
itating+0.054
arre+0.051
melon+0.050
inas+0.048
flower+0.048
Neg logit contributions
year-0.074
glass-0.072
ird-0.071
 hours-0.070
 minutes-0.067
Token.
Token ID13
Feature activation+5.10e-01

Pos logit contributions
itating+0.109
ucker+0.102
arre+0.100
rust+0.094
inas+0.094
Neg logit contributions
glass-0.115
year-0.111
 hours-0.110
ird-0.107
 minutes-0.101
Token He
Token ID679
Feature activation+1.62e+00

Pos logit contributions
itating+0.011
arre+0.011
ling+0.011
 up+0.010
bing+0.010
Neg logit contributions
year-0.016
ird-0.015
glass-0.015
 hours-0.014
aster-0.014
Token begged
Token ID32946
Feature activation-1.76e-01

Pos logit contributions
itating+0.043
bing+0.040
arre+0.040
 accidents+0.039
melon+0.039
Neg logit contributions
year-0.041
 hours-0.040
ird-0.039
glass-0.039
 minutes-0.037
Token his
Token ID465
Feature activation+6.51e-01

Pos logit contributions
year+0.008
ird+0.008
glass+0.007
ris+0.007
ibr+0.007
Neg logit contributions
 up-0.002
 fighting-0.002
 Freddy-0.002
melon-0.002
arre-0.002
Token friends
Token ID2460
Feature activation+1.81e-02

Pos logit contributions
itating+0.014
arre+0.013
ling+0.013
 up+0.013
melon+0.013
Neg logit contributions
year-0.021
glass-0.020
ird-0.020
 hours-0.019
 minutes-0.018
Token.
Token ID13
Feature activation+7.86e-01

Pos logit contributions
itating+0.015
arre+0.014
melon+0.014
rust+0.013
 up+0.013
Neg logit contributions
year-0.021
glass-0.020
ird-0.020
 hours-0.019
 minutes-0.019
Token He
Token ID679
Feature activation+6.88e-01

Pos logit contributions
itating+0.017
ling+0.017
arre+0.017
 up+0.016
bing+0.016
Neg logit contributions
year-0.025
ird-0.024
glass-0.024
aster-0.022
 minutes-0.022
Token saw
Token ID2497
Feature activation+6.07e-01

Pos logit contributions
arre+0.013
 up+0.013
itating+0.013
ling+0.013
melon+0.013
Neg logit contributions
year-0.023
glass-0.022
ird-0.022
 hours-0.021
aster-0.021
Token Sara
Token ID24799
Feature activation+1.06e+00

Pos logit contributions
itating+0.013
arre+0.012
bing+0.011
ling+0.011
melon+0.011
Neg logit contributions
year-0.020
glass-0.019
ird-0.019
 hours-0.018
aster-0.018
Token's
Token ID338
Feature activation+7.28e-01

Pos logit contributions
rust+0.022
itating+0.020
 Freddy+0.019
melon+0.019
arre+0.019
Neg logit contributions
glass-0.033
ird-0.032
year-0.032
aster-0.031
lock-0.029
Token rubber
Token ID14239
Feature activation+4.22e+00

Pos logit contributions
itating+0.017
arre+0.017
melon+0.016
 up+0.016
ling+0.016
Neg logit contributions
year-0.021
glass-0.021
ird-0.021
 hours-0.019
aster-0.019
Token duck
Token ID22045
Feature activation+2.65e+00

Pos logit contributions
itating+0.112
itionally+0.098
rust+0.097
arre+0.095
appointed+0.094
Neg logit contributions
glass-0.100
bers-0.098
ird-0.098
 hours-0.093
 minutes-0.090
Token and
Token ID290
Feature activation+2.02e+00

Pos logit contributions
rust+0.060
itating+0.057
 Sharing+0.054
Being+0.054
arre+0.054
Neg logit contributions
glass-0.066
ird-0.063
e-0.061
ns-0.060
 kil-0.060
Token wanted
Token ID2227
Feature activation+7.00e-01

Pos logit contributions
itating+0.052
arre+0.049
 up+0.045
bing+0.044
ling+0.044
Neg logit contributions
year-0.057
ird-0.057
glass-0.055
 hours-0.055
aster-0.054
Token to
Token ID284
Feature activation+1.88e+00

Pos logit contributions
itating+0.015
arre+0.015
 up+0.014
melon+0.014
ling+0.014
Neg logit contributions
year-0.022
glass-0.021
ird-0.021
 hours-0.020
aster-0.020
Token take
Token ID1011
Feature activation+1.50e-01

Pos logit contributions
itating+0.057
arre+0.057
ling+0.053
sets+0.052
bing+0.052
Neg logit contributions
aster-0.046
ird-0.043
year-0.039
glass-0.038
 hours-0.038
Token
Token ID198
Feature activation-5.19e-01

Pos logit contributions
year+0.043
ird+0.043
glass+0.042
 hours+0.040
ibr+0.039
Neg logit contributions
 up-0.011
itating-0.010
ling-0.009
arre-0.009
 fighting-0.009
TokenM
Token ID44
Feature activation-3.04e+00

Pos logit contributions
ird+0.020
year+0.020
ocket+0.019
glass+0.019
 hours+0.019
Neg logit contributions
itating-0.007
ling-0.007
 up-0.006
bing-0.006
dy-0.006
Tokenira
Token ID8704
Feature activation-8.66e-01

Pos logit contributions
year+0.094
glass+0.090
ird+0.088
 hours+0.088
 ages+0.080
Neg logit contributions
ucker-0.070
ling-0.064
 up-0.064
ies-0.059
melon-0.058
Token looked
Token ID3114
Feature activation+2.17e-01

Pos logit contributions
year+0.029
ird+0.027
glass+0.027
 hours+0.026
aster+0.026
Neg logit contributions
itating-0.017
arre-0.017
 up-0.016
melon-0.016
ling-0.016
Token at
Token ID379
Feature activation-4.04e-01

Pos logit contributions
itating+0.005
arre+0.005
melon+0.005
inas+0.005
rust+0.005
Neg logit contributions
year-0.006
 minutes-0.006
ird-0.006
glass-0.006
 hours-0.006
Token the
Token ID262
Feature activation+4.21e+00

Pos logit contributions
year+0.014
ird+0.013
glass+0.013
 hours+0.012
 minutes+0.012
Neg logit contributions
 up-0.008
itating-0.008
arre-0.007
melon-0.007
ling-0.007
Token b
Token ID275
Feature activation-7.09e-01

Pos logit contributions
itating+0.118
arre+0.110
bing+0.108
mill+0.108
ocked+0.102
Neg logit contributions
 hours-0.093
 minutes-0.091
year-0.090
 kil-0.082
ird-0.082
Tokenerry
Token ID6996
Feature activation+1.79e+00

Pos logit contributions
year+0.015
ird+0.014
glass+0.013
aster+0.012
 hours+0.012
Neg logit contributions
itating-0.023
arre-0.023
 up-0.022
 fighting-0.022
melon-0.021
Token in
Token ID287
Feature activation-7.53e-02

Pos logit contributions
arre+0.040
itating+0.039
rust+0.036
 fighting+0.034
inas+0.034
Neg logit contributions
year-0.054
glass-0.051
ird-0.050
 hours-0.049
aster-0.049
Token her
Token ID607
Feature activation-1.32e-01

Pos logit contributions
year+0.003
glass+0.003
ird+0.003
 hours+0.002
aster+0.002
Neg logit contributions
itating-0.001
arre-0.001
 up-0.001
melon-0.001
 fighting-0.001
Token hand
Token ID1021
Feature activation+3.34e-01

Pos logit contributions
year+0.004
glass+0.004
ird+0.004
 hours+0.003
aster+0.003
Neg logit contributions
itating-0.003
arre-0.003
 up-0.003
melon-0.003
ling-0.003
Token.
Token ID13
Feature activation+8.52e-01

Pos logit contributions
arre+0.022
itating+0.021
ling+0.020
 up+0.019
melon+0.019
Neg logit contributions
year-0.029
ird-0.029
glass-0.028
 hours-0.027
 minutes-0.026
Token Every
Token ID3887
Feature activation+1.27e-01

Pos logit contributions
itating+0.017
arre+0.017
 up+0.016
ling+0.016
melon+0.016
Neg logit contributions
year-0.028
ird-0.027
glass-0.027
 hours-0.026
aster-0.025
Token day
Token ID1110
Feature activation-1.23e-01

Pos logit contributions
itating+0.003
arre+0.003
 up+0.003
ling+0.003
melon+0.003
Neg logit contributions
year-0.003
glass-0.003
ird-0.003
 hours-0.003
aster-0.003
Token he
Token ID339
Feature activation+1.57e+00

Pos logit contributions
year+0.005
ird+0.005
ibr+0.005
glass+0.005
ibrarian+0.005
Neg logit contributions
 up-0.002
 fighting-0.001
 Freddy-0.001
melon-0.001
 accidents-0.001
Token would
Token ID561
Feature activation+1.62e+00

Pos logit contributions
 up+0.032
melon+0.032
arre+0.032
ling+0.031
itating+0.030
Neg logit contributions
year-0.051
glass-0.049
ird-0.048
 hours-0.044
aster-0.044
Token pick
Token ID2298
Feature activation+4.19e+00

Pos logit contributions
itating+0.041
arre+0.041
ling+0.039
bing+0.039
 up+0.037
Neg logit contributions
ird-0.044
year-0.044
aster-0.043
glass-0.042
 hours-0.041
Token one
Token ID530
Feature activation+1.33e+00

Pos logit contributions
itating+0.123
ivering+0.113
arre+0.112
bing+0.107
inas+0.106
Neg logit contributions
year-0.097
 hours-0.090
aster-0.082
 Africa-0.082
book-0.080
Token from
Token ID422
Feature activation+1.88e-01

Pos logit contributions
arre+0.036
itating+0.034
 fighting+0.032
bing+0.032
inas+0.031
Neg logit contributions
year-0.035
ird-0.033
 hours-0.031
aster-0.031
 minutes-0.031
Token the
Token ID262
Feature activation+2.58e+00

Pos logit contributions
 up+0.003
melon+0.003
arre+0.003
itating+0.003
ling+0.003
Neg logit contributions
year-0.007
ird-0.007
glass-0.006
 hours-0.006
 minutes-0.006
Token bush
Token ID24484
Feature activation-3.38e-01

Pos logit contributions
arre+0.059
itating+0.059
ling+0.059
bing+0.058
mill+0.057
Neg logit contributions
year-0.070
ird-0.070
glass-0.069
 hours-0.068
aster-0.065
Token in
Token ID287
Feature activation-6.06e-01

Pos logit contributions
year+0.012
ird+0.012
glass+0.012
 hours+0.011
 minutes+0.011
Neg logit contributions
 up-0.006
itating-0.006
melon-0.006
arre-0.006
ling-0.005
Token a
Token ID257
Feature activation-1.61e+00

Pos logit contributions
year+0.022
ird+0.021
glass+0.021
 kil+0.019
lock+0.019
Neg logit contributions
 up-0.016
arre-0.016
itating-0.015
melon-0.015
 fighting-0.015
Token tiny
Token ID7009
Feature activation-1.39e+00

Pos logit contributions
year+0.063
glass+0.062
ird+0.061
 hours+0.057
ibr+0.057
Neg logit contributions
 up-0.022
ling-0.021
melon-0.021
arre-0.021
 fighting-0.021
Token girl
Token ID2576
Feature activation+8.56e-01

Pos logit contributions
year+0.052
glass+0.051
ird+0.050
 hours+0.047
aster+0.047
Neg logit contributions
melon-0.022
 up-0.022
ling-0.021
arre-0.020
 fighting-0.020
Token named
Token ID3706
Feature activation-9.77e-01

Pos logit contributions
 up+0.023
ling+0.022
 fighting+0.022
itating+0.021
melon+0.021
Neg logit contributions
year-0.022
glass-0.022
ird-0.021
 hours-0.020
ibr-0.020
Token Freddy
Token ID45437
Feature activation+1.60e+00

Pos logit contributions
year+0.040
glass+0.040
ird+0.038
 hours+0.037
aster+0.037
Neg logit contributions
 up-0.012
itating-0.011
arre-0.011
melon-0.010
 fighting-0.010
Token.
Token ID13
Feature activation+4.18e+00

Pos logit contributions
arre+0.036
itating+0.036
melon+0.034
 up+0.033
ling+0.032
Neg logit contributions
year-0.050
ird-0.047
glass-0.047
 hours-0.044
aster-0.044
Token She
Token ID1375
Feature activation+5.12e-01

Pos logit contributions
ning+0.129
mill+0.129
itating+0.126
bing+0.125
izing+0.120
Neg logit contributions
year-0.096
 hours-0.083
book-0.075
 minutes-0.075
 ages-0.068
Token lived
Token ID5615
Feature activation+7.69e-01

Pos logit contributions
itating+0.012
arre+0.012
melon+0.011
 up+0.011
ling+0.011
Neg logit contributions
year-0.016
glass-0.015
ird-0.015
 hours-0.014
aster-0.014
Token in
Token ID287
Feature activation-6.65e-01

Pos logit contributions
itating+0.014
arre+0.014
 up+0.014
melon+0.014
ling+0.013
Neg logit contributions
year-0.026
glass-0.025
ird-0.025
 hours-0.024
aster-0.023
Token a
Token ID257
Feature activation-1.25e+00

Pos logit contributions
year+0.020
ird+0.019
glass+0.019
 hours+0.017
 minutes+0.017
Neg logit contributions
 up-0.015
itating-0.015
arre-0.015
melon-0.015
ling-0.014
Token big
Token ID1263
Feature activation-6.69e-01

Pos logit contributions
year+0.040
ird+0.038
glass+0.038
 hours+0.036
aster+0.035
Neg logit contributions
itating-0.027
arre-0.027
 up-0.026
melon-0.025
ling-0.025
Token.
Token ID13
Feature activation+2.00e-01

Pos logit contributions
year+0.062
ird+0.058
glass+0.058
ibr+0.056
aster+0.056
Neg logit contributions
 up-0.027
itating-0.020
 fighting-0.019
ling-0.019
arre-0.018
Token
Token ID198
Feature activation-8.43e-01

Pos logit contributions
itating+0.004
arre+0.004
ling+0.004
 up+0.004
melon+0.004
Neg logit contributions
year-0.006
ird-0.006
glass-0.006
 hours-0.006
aster-0.006
Token
Token ID198
Feature activation+9.23e-01

Pos logit contributions
year+0.031
ird+0.030
glass+0.030
 hours+0.029
 minutes+0.028
Neg logit contributions
itating-0.013
arre-0.013
 up-0.013
ling-0.012
melon-0.012
TokenAt
Token ID2953
Feature activation-1.33e+00

Pos logit contributions
itating+0.022
 up+0.022
arre+0.021
melon+0.021
ling+0.021
Neg logit contributions
year-0.027
glass-0.026
ird-0.026
 hours-0.024
aster-0.024
Token the
Token ID262
Feature activation-1.80e+00

Pos logit contributions
year+0.059
ird+0.057
glass+0.056
ibr+0.054
 kil+0.053
Neg logit contributions
 up-0.019
 fighting-0.013
itating-0.013
melon-0.012
bing-0.011
Token bar
Token ID2318
Feature activation-6.03e+00

Pos logit contributions
year+0.071
glass+0.065
ocket+0.061
ird+0.061
lan+0.061
Neg logit contributions
 up-0.035
 same-0.029
itating-0.026
 fighting-0.026
 me-0.024
Tokenber
Token ID527
Feature activation-3.53e+00

Pos logit contributions
 hours+0.131
year+0.128
ocket+0.126
 Africa+0.121
glass+0.119
Neg logit contributions
 up-0.171
itating-0.153
 by-0.151
ling-0.148
 fighting-0.147
Token,
Token ID11
Feature activation-5.60e-01

Pos logit contributions
year+0.164
lan+0.144
 hours+0.142
 minutes+0.141
glass+0.140
Neg logit contributions
 up-0.050
's-0.033
my-0.032
 fighting-0.027
ning-0.023
Token the
Token ID262
Feature activation-2.08e+00

Pos logit contributions
year+0.021
ird+0.021
glass+0.020
ris+0.019
 minutes+0.019
Neg logit contributions
 up-0.009
 fighting-0.008
 Freddy-0.008
melon-0.008
itating-0.008
Token bar
Token ID2318
Feature activation-5.63e+00

Pos logit contributions
year+0.085
glass+0.079
ird+0.077
ris+0.077
ocket+0.076
Neg logit contributions
 up-0.033
 same-0.029
 fighting-0.025
 hurting-0.024
 frown-0.023
Tokenber
Token ID527
Feature activation-3.09e+00

Pos logit contributions
 Africa+0.133
 hours+0.125
 takeoff+0.114
ocket+0.114
growth+0.113
Neg logit contributions
 up-0.147
ling-0.145
y-0.143
mill-0.141
ing-0.138
Token boots
Token ID14412
Feature activation-2.31e+00

Pos logit contributions
year+0.076
ris+0.067
ird+0.067
ibr+0.066
ibrarian+0.066
Neg logit contributions
 up-0.034
 qu-0.029
 bl-0.028
 swe-0.025
 Bugs-0.025
Token and
Token ID290
Feature activation-1.82e+00

Pos logit contributions
year+0.098
ird+0.093
glass+0.092
aster+0.091
 hours+0.090
Neg logit contributions
 up-0.028
itating-0.024
melon-0.024
ling-0.023
arre-0.023
Token climbed
Token ID19952
Feature activation-1.17e+00

Pos logit contributions
year+0.073
ae+0.065
ris+0.065
ibr+0.064
ibrarian+0.064
Neg logit contributions
 up-0.030
 qu-0.023
 bl-0.022
 scatter-0.021
 swe-0.021
Token on
Token ID319
Feature activation-1.80e+00

Pos logit contributions
ird+0.048
year+0.047
glass+0.043
ibr+0.043
aster+0.043
Neg logit contributions
 up-0.023
ling-0.014
melon-0.014
bing-0.014
mill-0.014
Token the
Token ID262
Feature activation-1.23e+00

Pos logit contributions
year+0.079
ird+0.076
glass+0.073
ibr+0.071
lock+0.070
Neg logit contributions
 up-0.025
 fighting-0.018
melon-0.017
 Freddy-0.016
bing-0.015
Token sle
Token ID3133
Feature activation-5.97e+00

Pos logit contributions
year+0.040
ird+0.037
aster+0.033
ibr+0.033
glass+0.032
Neg logit contributions
 up-0.033
ling-0.030
melon-0.028
bing-0.027
itating-0.027
Tokenigh
Token ID394
Feature activation-2.67e+00

Pos logit contributions
 hours+0.168
year+0.164
 minutes+0.162
Ah+0.146
ird+0.145
Neg logit contributions
flower-0.132
ling-0.132
ocked-0.129
sets-0.127
my-0.126
Token.
Token ID13
Feature activation-1.40e+00

Pos logit contributions
year+0.112
 hours+0.104
ird+0.103
aster+0.101
 minutes+0.100
Neg logit contributions
 up-0.040
ling-0.034
itating-0.032
bing-0.030
mill-0.028
Token Santa
Token ID8909
Feature activation-2.16e+00

Pos logit contributions
year+0.065
ird+0.065
glass+0.063
ris+0.060
aster+0.060
Neg logit contributions
 up-0.012
itating-0.009
 Sharing-0.008
 fighting-0.008
arre-0.007
Token gave
Token ID2921
Feature activation+3.74e-02

Pos logit contributions
year+0.078
ird+0.074
aster+0.072
 hours+0.070
glass+0.069
Neg logit contributions
 up-0.041
ling-0.039
itating-0.037
 fighting-0.036
melon-0.036
Token them
Token ID606
Feature activation-2.32e+00

Pos logit contributions
 up+0.000
itating+0.000
ling+0.000
arre+0.000
melon+0.000
Neg logit contributions
year-0.002
ird-0.002
glass-0.002
 hours-0.001
lock-0.001
Token instead
Token ID2427
Feature activation-1.37e+00

Pos logit contributions
year+0.045
glass+0.043
ird+0.042
 hours+0.041
 minutes+0.040
Neg logit contributions
 up-0.022
arre-0.021
itating-0.021
ling-0.020
melon-0.020
Token.
Token ID13
Feature activation-6.75e-01

Pos logit contributions
year+0.044
glass+0.043
ird+0.041
 hours+0.039
ibr+0.039
Neg logit contributions
 up-0.029
ling-0.028
itating-0.027
arre-0.026
 fighting-0.026
Token
Token ID198
Feature activation-8.61e-01

Pos logit contributions
year+0.025
glass+0.024
ird+0.024
 hours+0.023
aster+0.022
Neg logit contributions
itating-0.011
arre-0.011
 up-0.011
melon-0.010
 fighting-0.010
Token
Token ID198
Feature activation+1.34e-02

Pos logit contributions
year+0.033
ird+0.032
glass+0.032
 hours+0.031
 minutes+0.030
Neg logit contributions
itating-0.012
arre-0.012
 up-0.012
ling-0.011
melon-0.011
TokenThe
Token ID464
Feature activation-1.93e+00

Pos logit contributions
itating+0.000
arre+0.000
 up+0.000
melon+0.000
ling+0.000
Neg logit contributions
year-0.000
ird-0.000
glass-0.000
 hours-0.000
 minutes-0.000
Token bar
Token ID2318
Feature activation-5.82e+00

Pos logit contributions
ird+0.072
glass+0.072
year+0.071
adders+0.068
lan+0.067
Neg logit contributions
 up-0.033
 fighting-0.028
ling-0.026
 hurting-0.026
 en-0.026
Tokenber
Token ID527
Feature activation-2.96e+00

Pos logit contributions
 Africa+0.126
 hours+0.123
 ages+0.109
aster+0.106
ird+0.103
Neg logit contributions
ling-0.176
mill-0.166
 up-0.161
itating-0.161
 fighting-0.155
Token didn
Token ID1422
Feature activation-2.82e-01

Pos logit contributions
year+0.107
glass+0.102
ird+0.099
 Africa+0.096
 hours+0.096
Neg logit contributions
arre-0.051
 up-0.050
ling-0.047
 fighting-0.046
melon-0.046
Token't
Token ID470
Feature activation-1.78e+00

Pos logit contributions
year+0.010
ird+0.009
aster+0.009
glass+0.009
lock+0.009
Neg logit contributions
ling-0.005
flower-0.005
arre-0.005
lings-0.005
ety-0.005
Token want
Token ID765
Feature activation-1.63e+00

Pos logit contributions
year+0.039
ird+0.039
glass+0.036
aster+0.034
 hours+0.033
Neg logit contributions
itating-0.054
bing-0.054
melon-0.053
arre-0.053
 up-0.053
Token to
Token ID284
Feature activation-1.69e-01

Pos logit contributions
year+0.061
ird+0.059
glass+0.058
lock+0.053
ibr+0.052
Neg logit contributions
 up-0.030
 Freddy-0.027
 fighting-0.026
melon-0.026
arre-0.025
Token nice
Token ID3621
Feature activation-5.83e-01

Pos logit contributions
glass+0.031
 hours+0.029
year+0.029
ird+0.028
aster+0.028
Neg logit contributions
 up-0.017
itating-0.016
melon-0.015
arre-0.015
ling-0.015
Token."
Token ID526
Feature activation-4.52e-01

Pos logit contributions
year+0.024
glass+0.023
aster+0.022
ird+0.022
 hours+0.021
Neg logit contributions
 up-0.008
arre-0.007
 fighting-0.007
mill-0.006
 sharing-0.006
Token
Token ID198
Feature activation-8.67e-01

Pos logit contributions
year+0.017
ird+0.016
glass+0.016
 hours+0.015
aster+0.015
Neg logit contributions
arre-0.007
itating-0.007
 up-0.007
melon-0.007
 fighting-0.007
Token
Token ID198
Feature activation+2.32e-02

Pos logit contributions
year+0.032
ird+0.031
glass+0.030
 hours+0.029
 minutes+0.029
Neg logit contributions
itating-0.014
arre-0.014
 up-0.014
melon-0.013
ling-0.013
TokenThe
Token ID464
Feature activation-1.64e+00

Pos logit contributions
itating+0.000
arre+0.000
 up+0.000
melon+0.000
ling+0.000
Neg logit contributions
year-0.001
ird-0.001
glass-0.001
 hours-0.001
 minutes-0.001
Token bar
Token ID2318
Feature activation-5.81e+00

Pos logit contributions
glass+0.062
ird+0.061
year+0.060
ris+0.057
 hours+0.056
Neg logit contributions
 up-0.028
 fighting-0.022
 hurting-0.021
 same-0.021
ling-0.020
Tokenber
Token ID527
Feature activation-3.01e+00

Pos logit contributions
 Africa+0.124
 hours+0.122
 ages+0.107
aster+0.104
glass+0.102
Neg logit contributions
mill-0.174
ling-0.173
 up-0.167
 fighting-0.162
melon-0.159
Token gave
Token ID2921
Feature activation-1.22e-01

Pos logit contributions
year+0.110
glass+0.102
ird+0.101
 Africa+0.098
 hours+0.098
Neg logit contributions
 up-0.053
arre-0.049
melon-0.048
 fighting-0.047
ling-0.046
Token Tom
Token ID4186
Feature activation-4.37e-01

Pos logit contributions
year+0.004
ird+0.004
glass+0.004
 hours+0.003
 minutes+0.003
Neg logit contributions
 up-0.003
arre-0.003
itating-0.003
melon-0.003
ling-0.003
Token a
Token ID257
Feature activation-2.14e+00

Pos logit contributions
year+0.013
glass+0.012
ird+0.012
 hours+0.012
aster+0.011
Neg logit contributions
itating-0.010
arre-0.010
 up-0.010
melon-0.010
ling-0.010
Token hug
Token ID16225
Feature activation-9.48e-01

Pos logit contributions
year+0.071
glass+0.070
 hours+0.068
ird+0.067
 minutes+0.064
Neg logit contributions
 up-0.044
arre-0.042
melon-0.040
itating-0.040
ling-0.037
Token a
Token ID257
Feature activation-1.89e+00

Pos logit contributions
ird+0.074
glass+0.072
 kil+0.072
year+0.070
ait+0.069
Neg logit contributions
 up-0.031
 Freddy-0.023
 fighting-0.023
 sharing-0.021
 bouncing-0.020
Token bar
Token ID2318
Feature activation-4.50e+00

Pos logit contributions
year+0.073
glass+0.071
ird+0.071
ibr+0.066
 hours+0.066
Neg logit contributions
 up-0.027
 fighting-0.026
itating-0.026
melon-0.026
ling-0.025
Tokenber
Token ID527
Feature activation-1.75e+00

Pos logit contributions
 hours+0.104
aster+0.090
ird+0.089
isc+0.087
 takeoff+0.083
Neg logit contributions
ling-0.126
 up-0.125
 fighting-0.124
y-0.122
melon-0.119
Token.
Token ID13
Feature activation-1.67e+00

Pos logit contributions
year+0.072
 hours+0.068
ibr+0.067
Ah+0.065
 minutes+0.065
Neg logit contributions
 up-0.022
 fighting-0.019
flower-0.018
 sharing-0.018
 Sharing-0.018
Token The
Token ID383
Feature activation-2.73e+00

Pos logit contributions
glass+0.070
year+0.070
ird+0.069
ocket+0.067
izzard+0.066
Neg logit contributions
 Freddy-0.016
 Sharing-0.016
 fighting-0.014
 She-0.013
 up-0.013
Token bar
Token ID2318
Feature activation-5.76e+00

Pos logit contributions
year+0.110
glass+0.104
ird+0.104
ris+0.103
ocket+0.103
Neg logit contributions
 up-0.045
 same-0.039
 fighting-0.032
 sad-0.031
 en-0.031
Tokenber
Token ID527
Feature activation-2.42e+00

Pos logit contributions
 hours+0.133
aster+0.122
 Africa+0.113
ird+0.111
 ages+0.109
Neg logit contributions
ling-0.164
 up-0.157
mill-0.155
y-0.154
 en-0.147
Token was
Token ID373
Feature activation-1.89e+00

Pos logit contributions
year+0.108
aster+0.100
ird+0.098
ait+0.098
 kil+0.097
Neg logit contributions
 up-0.023
 fighting-0.019
my-0.018
 sharing-0.016
melon-0.016
Token a
Token ID257
Feature activation-3.44e+00

Pos logit contributions
ird+0.068
glass+0.067
year+0.065
 hours+0.064
ris+0.063
Neg logit contributions
 up-0.038
 fighting-0.033
melon-0.030
 en-0.030
itating-0.030
Token funny
Token ID8258
Feature activation-3.13e+00

Pos logit contributions
year+0.141
glass+0.139
adders+0.137
ird+0.134
ibr+0.131
Neg logit contributions
 sad-0.040
 en-0.040
 up-0.040
 fighting-0.037
 qu-0.034
Token man
Token ID582
Feature activation-1.69e+00

Pos logit contributions
year+0.123
glass+0.110
ird+0.106
ibr+0.105
adders+0.104
Neg logit contributions
 up-0.057
ling-0.048
 fighting-0.047
arre-0.046
mill-0.043
Token scared
Token ID12008
Feature activation+3.74e-01

Pos logit contributions
glass+0.025
ird+0.025
year+0.024
 hours+0.024
aster+0.023
Neg logit contributions
 up-0.011
itating-0.010
ling-0.010
 fighting-0.010
melon-0.010
Token.
Token ID13
Feature activation-1.06e+00

Pos logit contributions
 up+0.007
itating+0.007
arre+0.007
ling+0.006
 fighting+0.006
Neg logit contributions
year-0.013
ird-0.013
glass-0.013
 hours-0.012
aster-0.012
Token
Token ID198
Feature activation-1.20e+00

Pos logit contributions
year+0.046
ird+0.044
glass+0.044
 hours+0.043
aster+0.042
Neg logit contributions
 Sharing-0.010
 up-0.010
 Freddy-0.010
itating-0.009
 fighting-0.009
Token
Token ID198
Feature activation+4.59e-02

Pos logit contributions
ird+0.049
year+0.049
glass+0.047
 hours+0.046
 minutes+0.045
Neg logit contributions
 up-0.015
itating-0.014
arre-0.013
ling-0.012
 fighting-0.012
TokenThe
Token ID464
Feature activation-1.22e+00

Pos logit contributions
itating+0.001
arre+0.001
 up+0.001
ling+0.001
 fighting+0.001
Neg logit contributions
ird-0.002
year-0.002
glass-0.002
 hours-0.001
ocket-0.001
Token bar
Token ID2318
Feature activation-5.75e+00

Pos logit contributions
year+0.048
ird+0.048
glass+0.045
ris+0.044
ocket+0.043
Neg logit contributions
 up-0.019
 same-0.016
 fighting-0.016
 hurting-0.016
 mess-0.015
Tokenber
Token ID527
Feature activation-2.91e+00

Pos logit contributions
 hours+0.131
 Africa+0.126
 ages+0.118
ird+0.115
prise+0.109
Neg logit contributions
ling-0.163
mill-0.154
 up-0.149
 fighting-0.148
y-0.142
Token started
Token ID2067
Feature activation-9.37e-02

Pos logit contributions
year+0.110
lan+0.096
 Africa+0.096
ird+0.095
 kil+0.095
Neg logit contributions
 up-0.048
 fighting-0.048
arre-0.043
ling-0.043
melon-0.041
Token to
Token ID284
Feature activation-7.57e-01

Pos logit contributions
year+0.003
ird+0.003
glass+0.003
 hours+0.003
aster+0.003
Neg logit contributions
arre-0.002
itating-0.002
 up-0.002
melon-0.002
ling-0.002
Token cut
Token ID2005
Feature activation-2.21e-01

Pos logit contributions
ird+0.022
aster+0.022
year+0.022
glass+0.021
 hours+0.020
Neg logit contributions
itating-0.017
arre-0.017
ling-0.016
melon-0.016
ocked-0.016
Token Tom
Token ID4186
Feature activation+7.91e-01

Pos logit contributions
year+0.008
ird+0.008
glass+0.008
 Bible+0.007
ibr+0.007
Neg logit contributions
 up-0.005
ling-0.004
 fighting-0.004
bing-0.003
melon-0.003
Token.
Token ID13
Feature activation+3.42e-02

Pos logit contributions
year+0.083
glass+0.078
 hours+0.077
aster+0.077
ird+0.076
Neg logit contributions
 up-0.030
 fighting-0.025
itating-0.024
melon-0.023
ling-0.023
Token
Token ID220
Feature activation-5.95e-01

Pos logit contributions
itating+0.001
arre+0.001
 up+0.001
melon+0.001
ling+0.001
Neg logit contributions
year-0.001
glass-0.001
ird-0.001
 hours-0.001
aster-0.001
Token
Token ID198
Feature activation-8.00e-01

Pos logit contributions
year+0.020
ird+0.019
glass+0.019
 hours+0.018
aster+0.018
Neg logit contributions
arre-0.011
itating-0.011
melon-0.011
 up-0.011
ling-0.011
Token
Token ID198
Feature activation+2.46e-01

Pos logit contributions
year+0.030
glass+0.030
ird+0.029
 hours+0.028
 minutes+0.028
Neg logit contributions
itating-0.012
arre-0.012
 up-0.011
ling-0.011
melon-0.010
TokenThe
Token ID464
Feature activation-1.86e+00

Pos logit contributions
itating+0.005
arre+0.005
ling+0.004
 up+0.004
melon+0.004
Neg logit contributions
year-0.009
glass-0.008
ird-0.008
 hours-0.008
 minutes-0.008
Token bar
Token ID2318
Feature activation-5.71e+00

Pos logit contributions
year+0.071
ird+0.071
glass+0.069
adders+0.067
ris+0.067
Neg logit contributions
 up-0.030
 same-0.025
 fighting-0.024
 hurting-0.024
ling-0.023
Tokenber
Token ID527
Feature activation-2.90e+00

Pos logit contributions
 Africa+0.131
 hours+0.129
 ages+0.115
aster+0.114
ird+0.113
Neg logit contributions
ling-0.157
mill-0.151
y-0.144
 up-0.144
 fighting-0.134
Token then
Token ID788
Feature activation-1.28e+00

Pos logit contributions
year+0.110
glass+0.101
 Africa+0.100
 kil+0.099
aster+0.099
Neg logit contributions
 up-0.045
arre-0.042
 fighting-0.041
ling-0.040
melon-0.038
Token started
Token ID2067
Feature activation-6.28e-01

Pos logit contributions
year+0.046
glass+0.044
ird+0.043
 Africa+0.042
 hours+0.042
Neg logit contributions
 up-0.023
arre-0.020
 fighting-0.020
ling-0.019
melon-0.018
Token cl
Token ID537
Feature activation-3.35e-01

Pos logit contributions
year+0.021
ird+0.020
glass+0.020
 hours+0.019
 minutes+0.018
Neg logit contributions
 up-0.013
arre-0.013
itating-0.013
melon-0.012
 fighting-0.012
Tokenapping
Token ID5912
Feature activation+9.19e-01

Pos logit contributions
year+0.011
glass+0.011
ird+0.011
 hours+0.010
aster+0.010
Neg logit contributions
arre-0.007
itating-0.007
 up-0.006
melon-0.006
ling-0.006
Token mom
Token ID1995
Feature activation-4.86e-01

Pos logit contributions
year+0.030
glass+0.027
ird+0.027
aster+0.027
 hours+0.026
Neg logit contributions
 up-0.022
 fighting-0.020
arre-0.019
itating-0.019
melon-0.019
Token said
Token ID531
Feature activation-1.04e+00

Pos logit contributions
year+0.015
ird+0.014
glass+0.014
 hours+0.013
aster+0.013
Neg logit contributions
itating-0.011
arre-0.011
 up-0.011
melon-0.011
ling-0.010
Token,
Token ID11
Feature activation-3.94e-01

Pos logit contributions
year+0.039
glass+0.037
ird+0.036
 hours+0.036
ris+0.035
Neg logit contributions
 up-0.016
melon-0.016
 fighting-0.015
itating-0.015
arre-0.015
Token "
Token ID366
Feature activation-2.90e-01

Pos logit contributions
year+0.015
glass+0.014
ird+0.014
 hours+0.014
aster+0.013
Neg logit contributions
itating-0.006
arre-0.006
 up-0.006
melon-0.006
ling-0.005
TokenA
Token ID32
Feature activation-3.95e+00

Pos logit contributions
year+0.011
ird+0.011
glass+0.010
 hours+0.010
 minutes+0.010
Neg logit contributions
itating-0.005
arre-0.004
 up-0.004
melon-0.004
ling-0.004
Token bar
Token ID2318
Feature activation-5.69e+00

Pos logit contributions
year+0.132
glass+0.128
 hours+0.124
adders+0.121
ibr+0.121
Neg logit contributions
 up-0.079
itating-0.061
 same-0.060
 sharing-0.060
 en-0.060
Tokenber
Token ID527
Feature activation-3.54e+00

Pos logit contributions
 hours+0.132
aster+0.103
year+0.102
 takeoff+0.101
 trek+0.099
Neg logit contributions
mill-0.160
 up-0.159
 fighting-0.157
y-0.156
ter-0.156
Token is
Token ID318
Feature activation-1.80e+00

Pos logit contributions
year+0.141
 hours+0.130
glass+0.127
 minutes+0.122
aster+0.121
Neg logit contributions
 up-0.051
 fighting-0.048
melon-0.044
mill-0.044
 Sharing-0.043
Token someone
Token ID2130
Feature activation-1.95e+00

Pos logit contributions
year+0.066
glass+0.063
ird+0.063
 hours+0.061
ris+0.060
Neg logit contributions
 up-0.035
 fighting-0.031
 sharing-0.027
melon-0.026
 being-0.026
Token who
Token ID508
Feature activation-2.66e+00

Pos logit contributions
year+0.065
glass+0.061
ird+0.059
 hours+0.059
aster+0.058
Neg logit contributions
 up-0.040
itating-0.038
 fighting-0.037
arre-0.037
ling-0.037
Token cuts
Token ID6630
Feature activation-1.65e+00

Pos logit contributions
year+0.088
glass+0.084
 hours+0.082
ird+0.082
 minutes+0.081
Neg logit contributions
 up-0.054
 fighting-0.051
arre-0.051
ling-0.048
itating-0.048
Token get
Token ID651
Feature activation-1.15e+00

Pos logit contributions
year+0.079
 minutes+0.070
glass+0.069
ibr+0.068
lan+0.067
Neg logit contributions
 up-0.032
 fighting-0.025
 bouncing-0.023
 scatter-0.023
 sharing-0.023
Token his
Token ID465
Feature activation-1.76e+00

Pos logit contributions
glass+0.043
ird+0.043
year+0.042
 hours+0.040
ris+0.040
Neg logit contributions
 up-0.024
 fighting-0.017
melon-0.016
ling-0.016
itating-0.015
Token haircut
Token ID45548
Feature activation-2.17e+00

Pos logit contributions
year+0.061
glass+0.053
ris+0.048
 hours+0.047
ird+0.047
Neg logit contributions
 up-0.047
 fighting-0.037
 same-0.037
 bl-0.035
 hurting-0.034
Token.
Token ID13
Feature activation-5.96e-01

Pos logit contributions
year+0.093
glass+0.084
 hours+0.082
aster+0.082
ibr+0.081
Neg logit contributions
 up-0.033
itating-0.024
ling-0.023
bing-0.022
melon-0.021
Token The
Token ID383
Feature activation-1.82e+00

Pos logit contributions
year+0.022
ird+0.022
glass+0.021
 hours+0.020
 minutes+0.020
Neg logit contributions
arre-0.010
 up-0.010
itating-0.010
melon-0.009
 fighting-0.009
Token bar
Token ID2318
Feature activation-5.66e+00

Pos logit contributions
year+0.071
glass+0.068
ris+0.066
ird+0.065
lan+0.065
Neg logit contributions
 up-0.031
 same-0.025
 fighting-0.022
 hurting-0.022
ling-0.022
Tokenber
Token ID527
Feature activation-2.88e+00

Pos logit contributions
 Africa+0.128
 hours+0.116
aster+0.106
 takeoff+0.104
ocket+0.104
Neg logit contributions
ling-0.165
 up-0.161
mill-0.155
y-0.155
itating-0.150
Token put
Token ID1234
Feature activation-4.80e-01

Pos logit contributions
year+0.116
 Africa+0.107
glass+0.107
lan+0.105
aster+0.105
Neg logit contributions
 up-0.043
ling-0.033
 fighting-0.033
arre-0.033
melon-0.030
Token a
Token ID257
Feature activation-2.78e+00

Pos logit contributions
year+0.018
ird+0.018
glass+0.017
ibr+0.016
ris+0.016
Neg logit contributions
 up-0.009
melon-0.007
ling-0.007
 fighting-0.007
 Freddy-0.007
Token cape
Token ID36023
Feature activation-3.00e+00

Pos logit contributions
year+0.105
glass+0.097
ibr+0.093
lan+0.092
 Africa+0.091
Neg logit contributions
 up-0.053
 mess-0.045
 bl-0.044
ling-0.043
itating-0.041
Token around
Token ID1088
Feature activation-7.06e-01

Pos logit contributions
year+0.116
ird+0.112
glass+0.109
 hours+0.107
aster+0.107
Neg logit contributions
 up-0.049
itating-0.041
ling-0.040
arre-0.038
melon-0.037
Token quiet
Token ID5897
Feature activation-8.70e-01

Pos logit contributions
glass+0.012
year+0.012
ird+0.012
ris+0.011
aster+0.011
Neg logit contributions
 up-0.011
arre-0.010
 fighting-0.010
ling-0.010
itating-0.010
Token and
Token ID290
Feature activation-9.94e-01

Pos logit contributions
year+0.034
glass+0.033
ird+0.033
aster+0.033
ris+0.031
Neg logit contributions
 up-0.014
itating-0.011
ling-0.011
 fighting-0.011
arre-0.011
Token polite
Token ID23507
Feature activation-3.45e-01

Pos logit contributions
year+0.036
glass+0.035
ird+0.034
ris+0.033
 hours+0.033
Neg logit contributions
 up-0.019
 fighting-0.017
itating-0.016
arre-0.015
 hurting-0.015
Token.
Token ID13
Feature activation-1.15e+00

Pos logit contributions
year+0.013
ird+0.013
glass+0.013
aster+0.013
 hours+0.012
Neg logit contributions
 up-0.006
 fighting-0.005
ling-0.005
itating-0.005
arre-0.004
Token The
Token ID383
Feature activation-1.21e+00

Pos logit contributions
year+0.042
glass+0.041
 hours+0.040
ird+0.039
aster+0.038
Neg logit contributions
itating-0.019
arre-0.019
 Sharing-0.018
 fighting-0.018
 up-0.018
Token bar
Token ID2318
Feature activation-5.65e+00

Pos logit contributions
year+0.044
glass+0.043
ird+0.041
ris+0.040
ibr+0.039
Neg logit contributions
 up-0.024
 same-0.020
 fighting-0.019
 en-0.018
itating-0.018
Tokenber
Token ID527
Feature activation-2.33e+00

Pos logit contributions
 hours+0.138
aster+0.119
year+0.113
 Africa+0.113
growth+0.111
Neg logit contributions
 up-0.158
mill-0.152
ling-0.150
 Sharing-0.143
y-0.143
Token is
Token ID318
Feature activation-1.38e-01

Pos logit contributions
year+0.089
glass+0.082
 hours+0.079
aster+0.078
ird+0.077
Neg logit contributions
 up-0.041
arre-0.035
 fighting-0.035
ling-0.034
melon-0.034
Token busy
Token ID8179
Feature activation+1.19e-01

Pos logit contributions
glass+0.005
year+0.005
ird+0.005
 hours+0.005
lock+0.005
Neg logit contributions
 up-0.003
 fighting-0.002
itating-0.002
arre-0.002
melon-0.002
Token working
Token ID1762
Feature activation-6.28e-01

Pos logit contributions
 up+0.002
itating+0.002
arre+0.002
 fighting+0.002
ling+0.002
Neg logit contributions
year-0.004
glass-0.004
ird-0.004
 hours-0.004
aster-0.004
Token."
Token ID526
Feature activation+5.05e-01

Pos logit contributions
year+0.023
ird+0.023
glass+0.022
aster+0.021
lock+0.021
Neg logit contributions
 up-0.013
ling-0.010
arre-0.010
itating-0.010
 fighting-0.009
Token be
Token ID307
Feature activation+4.86e-01

Pos logit contributions
itating+0.011
arre+0.011
 up+0.010
melon+0.010
ling+0.010
Neg logit contributions
year-0.018
ird-0.017
glass-0.017
 hours-0.016
aster-0.016
Token back
Token ID736
Feature activation-8.07e-01

Pos logit contributions
 up+0.011
arre+0.010
itating+0.010
ling+0.010
melon+0.009
Neg logit contributions
year-0.015
glass-0.015
ird-0.015
 hours-0.014
aster-0.014
Token again
Token ID757
Feature activation-2.56e-01

Pos logit contributions
year+0.027
glass+0.026
ird+0.026
ibr+0.024
aster+0.024
Neg logit contributions
 up-0.018
itating-0.015
ling-0.014
 sharing-0.014
arre-0.013
Token.
Token ID13
Feature activation-1.28e+00

Pos logit contributions
year+0.009
glass+0.008
ird+0.008
ibrarian+0.008
ibr+0.008
Neg logit contributions
 up-0.005
bing-0.005
ling-0.005
arre-0.005
 sharing-0.005
Token The
Token ID383
Feature activation-1.46e+00

Pos logit contributions
glass+0.048
year+0.047
ird+0.046
 hours+0.044
 minutes+0.044
Neg logit contributions
 up-0.019
arre-0.019
itating-0.019
 Sharing-0.018
 Freddy-0.017
Token bar
Token ID2318
Feature activation-5.65e+00

Pos logit contributions
glass+0.057
year+0.054
ird+0.053
lan+0.051
adders+0.051
Neg logit contributions
 up-0.024
 same-0.019
ling-0.018
 hurting-0.018
 fighting-0.018
Tokenber
Token ID527
Feature activation-3.19e+00

Pos logit contributions
 Africa+0.123
 hours+0.112
glass+0.102
aster+0.102
 takeoff+0.098
Neg logit contributions
ling-0.165
mill-0.163
 up-0.160
itating-0.152
 fighting-0.149
Token smiled
Token ID13541
Feature activation+1.16e+00

Pos logit contributions
year+0.126
glass+0.117
 Africa+0.115
lan+0.112
aster+0.112
Neg logit contributions
 up-0.048
ling-0.039
arre-0.039
melon-0.037
mill-0.036
Token and
Token ID290
Feature activation-1.18e+00

Pos logit contributions
arre+0.020
itating+0.019
melon+0.018
 up+0.018
ling+0.018
Neg logit contributions
year-0.042
glass-0.040
ird-0.040
 hours-0.038
 minutes-0.037
Token waved
Token ID26834
Feature activation-7.37e-01

Pos logit contributions
glass+0.041
year+0.041
ird+0.039
aster+0.038
 kil+0.038
Neg logit contributions
 up-0.023
 Freddy-0.020
melon-0.019
arre-0.019
itating-0.019
Token goodbye
Token ID24829
Feature activation-4.12e-01

Pos logit contributions
year+0.019
ird+0.018
glass+0.018
aster+0.017
ibr+0.016
Neg logit contributions
 up-0.022
ling-0.020
itating-0.019
arre-0.019
 Freddy-0.018
Token car
Token ID1097
Feature activation-1.74e+00

Pos logit contributions
year+0.028
ird+0.026
glass+0.025
ris+0.024
ibr+0.024
Neg logit contributions
 up-0.014
 same-0.011
melon-0.011
 mess-0.011
 fighting-0.010
Token.
Token ID13
Feature activation-7.71e-01

Pos logit contributions
year+0.080
 hours+0.074
Ah+0.073
ibr+0.072
lan+0.071
Neg logit contributions
 up-0.021
itating-0.013
 fighting-0.012
ling-0.011
melon-0.010
Token
Token ID198
Feature activation-9.26e-01

Pos logit contributions
year+0.033
glass+0.032
ird+0.032
 hours+0.031
 minutes+0.030
Neg logit contributions
 up-0.008
itating-0.008
arre-0.008
 fighting-0.007
melon-0.007
Token
Token ID198
Feature activation+3.91e-01

Pos logit contributions
year+0.037
ird+0.036
glass+0.035
 hours+0.035
 minutes+0.034
Neg logit contributions
 up-0.012
itating-0.012
arre-0.011
 fighting-0.010
melon-0.010
TokenThe
Token ID464
Feature activation-1.19e+00

Pos logit contributions
itating+0.008
arre+0.008
 up+0.008
 fighting+0.008
ling+0.008
Neg logit contributions
year-0.012
ird-0.012
glass-0.012
 hours-0.011
 minutes-0.011
Token bar
Token ID2318
Feature activation-5.65e+00

Pos logit contributions
year+0.046
ird+0.044
glass+0.043
ris+0.042
adows+0.041
Neg logit contributions
 up-0.021
 same-0.018
 mess-0.017
 fighting-0.017
 hurting-0.016
Tokenber
Token ID527
Feature activation-3.04e+00

Pos logit contributions
 Africa+0.128
 hours+0.127
 outer+0.115
 ages+0.114
ocket+0.111
Neg logit contributions
ling-0.151
mill-0.145
 up-0.143
 fighting-0.140
y-0.137
Token was
Token ID373
Feature activation-6.00e-01

Pos logit contributions
year+0.127
lan+0.113
 Africa+0.111
aster+0.110
 kil+0.110
Neg logit contributions
 up-0.045
 fighting-0.039
ling-0.032
arre-0.032
 qu-0.030
Token a
Token ID257
Feature activation-2.01e+00

Pos logit contributions
glass+0.023
ird+0.023
year+0.023
 hours+0.022
ris+0.022
Neg logit contributions
 up-0.010
 fighting-0.009
itating-0.008
 hurting-0.008
melon-0.007
Token big
Token ID1263
Feature activation-1.73e+00

Pos logit contributions
year+0.081
glass+0.077
adders+0.075
ird+0.074
ibr+0.073
Neg logit contributions
 mess-0.029
 fighting-0.028
 up-0.026
 sad-0.026
 frown-0.025
Token man
Token ID582
Feature activation-1.74e+00

Pos logit contributions
year+0.072
glass+0.065
ibr+0.063
ird+0.062
lan+0.060
Neg logit contributions
 fighting-0.029
 up-0.028
 mess-0.027
 frown-0.023
 hurting-0.023
Token hair
Token ID4190
Feature activation-1.81e+00

Pos logit contributions
year+0.045
glass+0.042
ird+0.041
 hours+0.040
 minutes+0.039
Neg logit contributions
 up-0.020
arre-0.016
ling-0.015
bing-0.015
itating-0.015
Token?"
Token ID1701
Feature activation-7.59e-01

Pos logit contributions
year+0.065
glass+0.062
ird+0.061
 hours+0.059
 minutes+0.057
Neg logit contributions
 up-0.032
itating-0.031
arre-0.031
melon-0.029
ling-0.029
Token
Token ID198
Feature activation-9.72e-01

Pos logit contributions
year+0.028
glass+0.026
ird+0.026
 hours+0.026
 minutes+0.025
Neg logit contributions
 up-0.013
itating-0.012
arre-0.012
melon-0.012
 fighting-0.012
Token
Token ID198
Feature activation-3.88e-01

Pos logit contributions
year+0.041
ird+0.039
glass+0.039
 hours+0.038
 minutes+0.038
Neg logit contributions
 up-0.011
itating-0.010
arre-0.010
rust-0.009
 fighting-0.008
TokenThe
Token ID464
Feature activation-1.39e+00

Pos logit contributions
year+0.015
ird+0.014
 hours+0.014
glass+0.014
 minutes+0.013
Neg logit contributions
itating-0.006
arre-0.006
 up-0.005
rust-0.005
 Sharing-0.005
Token bar
Token ID2318
Feature activation-5.65e+00

Pos logit contributions
glass+0.051
year+0.051
ird+0.048
ris+0.047
 hours+0.046
Neg logit contributions
 up-0.025
 fighting-0.020
itating-0.020
 en-0.020
 same-0.020
Tokenber
Token ID527
Feature activation-2.96e+00

Pos logit contributions
 hours+0.117
 Africa+0.114
 ages+0.103
glass+0.100
ird+0.099
Neg logit contributions
ling-0.175
mill-0.169
 up-0.167
itating-0.163
 fighting-0.158
Token was
Token ID373
Feature activation-1.12e+00

Pos logit contributions
year+0.104
glass+0.099
ird+0.097
 hours+0.094
 Africa+0.093
Neg logit contributions
 up-0.054
arre-0.052
ling-0.050
melon-0.050
itating-0.049
Token surprised
Token ID6655
Feature activation+1.08e+00

Pos logit contributions
glass+0.038
year+0.038
ird+0.038
 hours+0.036
lan+0.035
Neg logit contributions
 up-0.022
itating-0.020
 fighting-0.019
arre-0.019
melon-0.018
Token and
Token ID290
Feature activation-1.03e+00

Pos logit contributions
arre+0.027
itating+0.026
melon+0.026
flower+0.024
inas+0.024
Neg logit contributions
year-0.031
 minutes-0.030
 hours-0.029
glass-0.028
ris-0.028
Token pleased
Token ID10607
Feature activation+1.60e+00

Pos logit contributions
glass+0.034
ird+0.034
year+0.034
 hours+0.032
aster+0.031
Neg logit contributions
 up-0.021
itating-0.020
melon-0.019
 fighting-0.019
arre-0.019
Token brave
Token ID14802
Feature activation+1.19e-01

Pos logit contributions
 up+0.005
itating+0.004
arre+0.004
 fighting+0.004
ling+0.004
Neg logit contributions
glass-0.009
year-0.008
ird-0.008
 hours-0.008
ibr-0.008
Token.
Token ID13
Feature activation-1.30e-01

Pos logit contributions
arre+0.002
itating+0.002
 up+0.002
ling+0.002
melon+0.002
Neg logit contributions
year-0.004
glass-0.004
ird-0.004
 hours-0.004
aster-0.004
Token
Token ID198
Feature activation-4.42e-01

Pos logit contributions
year+0.004
ird+0.004
glass+0.004
 hours+0.004
aster+0.004
Neg logit contributions
itating-0.003
arre-0.003
ling-0.003
 up-0.003
melon-0.002
Token
Token ID198
Feature activation+1.19e+00

Pos logit contributions
year+0.015
glass+0.014
ird+0.014
 hours+0.014
aster+0.014
Neg logit contributions
itating-0.008
arre-0.008
 up-0.008
melon-0.008
ling-0.008
TokenThe
Token ID464
Feature activation-1.09e+00

Pos logit contributions
 up+0.031
melon+0.029
flower+0.029
itating+0.029
sets+0.029
Neg logit contributions
year-0.033
aster-0.031
ird-0.031
glass-0.031
 hours-0.029
Token bar
Token ID2318
Feature activation-5.64e+00

Pos logit contributions
year+0.042
glass+0.041
ird+0.039
ris+0.039
adders+0.038
Neg logit contributions
 up-0.018
 hurting-0.015
 same-0.015
 fighting-0.015
 mess-0.014
Tokenber
Token ID527
Feature activation-2.73e+00

Pos logit contributions
 Africa+0.132
 hours+0.122
 ages+0.113
 takeoff+0.110
 outer+0.107
Neg logit contributions
ling-0.160
mill-0.150
 up-0.145
y-0.141
 fighting-0.135
Token was
Token ID373
Feature activation-3.49e-01

Pos logit contributions
year+0.103
glass+0.095
 Africa+0.092
 kil+0.092
lan+0.091
Neg logit contributions
 up-0.046
arre-0.041
ling-0.040
 fighting-0.038
mill-0.036
Token gentle
Token ID10296
Feature activation+1.26e+00

Pos logit contributions
year+0.012
glass+0.012
ird+0.012
 hours+0.011
ibr+0.011
Neg logit contributions
 up-0.007
itating-0.006
arre-0.006
 fighting-0.006
ling-0.006
Token with
Token ID351
Feature activation-9.02e-01

Pos logit contributions
itating+0.019
arre+0.019
 up+0.018
melon+0.018
ling+0.017
Neg logit contributions
year-0.048
glass-0.046
ird-0.046
 hours-0.044
aster-0.043
Token his
Token ID465
Feature activation-2.36e+00

Pos logit contributions
year+0.033
glass+0.032
ird+0.032
lock+0.030
lan+0.030
Neg logit contributions
 up-0.016
 fighting-0.014
arre-0.014
 Freddy-0.014
ling-0.014
Token the
Token ID262
Feature activation-1.80e+00

Pos logit contributions
year+0.059
ird+0.057
glass+0.056
ibr+0.054
 kil+0.053
Neg logit contributions
 up-0.019
 fighting-0.013
itating-0.013
melon-0.012
bing-0.011
Token bar
Token ID2318
Feature activation-6.03e+00

Pos logit contributions
year+0.071
glass+0.065
ocket+0.061
ird+0.061
lan+0.061
Neg logit contributions
 up-0.035
 same-0.029
itating-0.026
 fighting-0.026
 me-0.024
Tokenber
Token ID527
Feature activation-3.53e+00

Pos logit contributions
 hours+0.131
year+0.128
ocket+0.126
 Africa+0.121
glass+0.119
Neg logit contributions
 up-0.171
itating-0.153
 by-0.151
ling-0.148
 fighting-0.147
Token,
Token ID11
Feature activation-5.60e-01

Pos logit contributions
year+0.164
lan+0.144
 hours+0.142
 minutes+0.141
glass+0.140
Neg logit contributions
 up-0.050
's-0.033
my-0.032
 fighting-0.027
ning-0.023
Token the
Token ID262
Feature activation-2.08e+00

Pos logit contributions
year+0.021
ird+0.021
glass+0.020
ris+0.019
 minutes+0.019
Neg logit contributions
 up-0.009
 fighting-0.008
 Freddy-0.008
melon-0.008
itating-0.008
Token bar
Token ID2318
Feature activation-5.63e+00

Pos logit contributions
year+0.085
glass+0.079
ird+0.077
ris+0.077
ocket+0.076
Neg logit contributions
 up-0.033
 same-0.029
 fighting-0.025
 hurting-0.024
 frown-0.023
Tokenber
Token ID527
Feature activation-3.09e+00

Pos logit contributions
 Africa+0.133
 hours+0.125
 takeoff+0.114
ocket+0.114
growth+0.113
Neg logit contributions
 up-0.147
ling-0.145
y-0.143
mill-0.141
ing-0.138
Token said
Token ID531
Feature activation-1.82e+00

Pos logit contributions
year+0.128
glass+0.110
lan+0.110
aster+0.109
 Africa+0.109
Neg logit contributions
 up-0.048
 fighting-0.038
ling-0.035
arre-0.034
my-0.034
Token that
Token ID326
Feature activation-1.56e+00

Pos logit contributions
year+0.066
ris+0.063
ird+0.063
glass+0.063
 hours+0.060
Neg logit contributions
 up-0.030
melon-0.029
 Freddy-0.029
 fighting-0.027
 sharing-0.026
Token he
Token ID339
Feature activation-9.45e-01

Pos logit contributions
year+0.058
glass+0.055
ris+0.053
ird+0.053
 minutes+0.052
Neg logit contributions
 up-0.027
 fighting-0.025
 sharing-0.024
melon-0.023
 Freddy-0.022
Token was
Token ID373
Feature activation-5.14e-01

Pos logit contributions
year+0.036
glass+0.034
aster+0.034
 hours+0.033
 minutes+0.033
Neg logit contributions
 up-0.014
arre-0.013
 fighting-0.013
ling-0.012
melon-0.012
Token off
Token ID572
Feature activation-8.86e-01

Pos logit contributions
year+0.075
ird+0.075
glass+0.072
ibr+0.068
 kil+0.068
Neg logit contributions
 up-0.033
arre-0.024
ling-0.024
itating-0.023
melon-0.023
Token.
Token ID13
Feature activation-6.96e-01

Pos logit contributions
year+0.030
ird+0.029
glass+0.029
 hours+0.027
aster+0.027
Neg logit contributions
 up-0.017
itating-0.017
arre-0.016
melon-0.016
 fighting-0.016
Token
Token ID198
Feature activation-1.18e+00

Pos logit contributions
year+0.023
glass+0.023
ird+0.023
 hours+0.021
aster+0.021
Neg logit contributions
arre-0.013
itating-0.013
 up-0.013
melon-0.013
ling-0.012
Token
Token ID198
Feature activation+4.55e-01

Pos logit contributions
year+0.040
ird+0.039
glass+0.038
 hours+0.036
 minutes+0.036
Neg logit contributions
itating-0.022
arre-0.022
 up-0.022
melon-0.021
ling-0.021
TokenThe
Token ID464
Feature activation-1.57e+00

Pos logit contributions
itating+0.011
 up+0.011
arre+0.011
melon+0.010
ling+0.010
Neg logit contributions
year-0.014
glass-0.013
ird-0.013
 hours-0.012
aster-0.012
Token bar
Token ID2318
Feature activation-5.61e+00

Pos logit contributions
glass+0.060
year+0.059
ird+0.058
adders+0.054
lan+0.054
Neg logit contributions
 up-0.026
ling-0.021
 same-0.020
 fighting-0.020
 hurting-0.020
Tokenber
Token ID527
Feature activation-2.22e+00

Pos logit contributions
 Africa+0.127
 hours+0.111
aster+0.102
ocket+0.102
glass+0.101
Neg logit contributions
ling-0.159
mill-0.157
 up-0.150
 fighting-0.146
itating-0.143
Token smiled
Token ID13541
Feature activation+1.09e+00

Pos logit contributions
year+0.087
glass+0.080
 kil+0.078
 Africa+0.076
lan+0.076
Neg logit contributions
 up-0.037
arre-0.031
 fighting-0.031
ling-0.030
flower-0.028
Token as
Token ID355
Feature activation-1.63e+00

Pos logit contributions
arre+0.020
itating+0.019
melon+0.019
 fighting+0.017
rust+0.017
Neg logit contributions
year-0.038
glass-0.036
ird-0.036
 hours-0.035
 minutes-0.034
Token he
Token ID339
Feature activation-1.50e+00

Pos logit contributions
year+0.057
glass+0.056
ird+0.056
 hours+0.051
 minutes+0.051
Neg logit contributions
 up-0.030
itating-0.028
melon-0.027
 fighting-0.027
 Freddy-0.027
Token watched
Token ID7342
Feature activation-5.58e-01

Pos logit contributions
year+0.053
glass+0.051
 kil+0.047
 Africa+0.047
aster+0.047
Neg logit contributions
arre-0.027
 up-0.026
ling-0.025
flower-0.024
 fighting-0.024
Token in
Token ID287
Feature activation-1.71e+00

Pos logit contributions
year+0.002
ird+0.001
glass+0.001
 hours+0.001
aster+0.001
Neg logit contributions
itating-0.001
 up-0.001
arre-0.001
ling-0.001
melon-0.001
Token the
Token ID262
Feature activation-2.21e+00

Pos logit contributions
year+0.073
ird+0.072
ibr+0.071
glass+0.070
lan+0.066
Neg logit contributions
 up-0.023
 fighting-0.018
melon-0.015
itating-0.014
bing-0.014
Token chair
Token ID5118
Feature activation-1.46e+00

Pos logit contributions
year+0.060
glass+0.059
ird+0.055
ris+0.052
ibr+0.051
Neg logit contributions
 up-0.066
ling-0.056
itating-0.054
 same-0.053
that-0.052
Token and
Token ID290
Feature activation-1.02e+00

Pos logit contributions
year+0.066
ird+0.063
glass+0.062
ibr+0.060
aster+0.060
Neg logit contributions
 up-0.015
itating-0.011
arre-0.011
ling-0.011
 fighting-0.011
Token the
Token ID262
Feature activation-1.94e+00

Pos logit contributions
year+0.037
glass+0.035
ird+0.034
ibr+0.032
ris+0.032
Neg logit contributions
 up-0.021
 fighting-0.018
itating-0.017
ling-0.017
arre-0.017
Token bar
Token ID2318
Feature activation-5.59e+00

Pos logit contributions
year+0.074
glass+0.073
ird+0.068
ris+0.068
lan+0.067
Neg logit contributions
 up-0.036
 same-0.028
 hurting-0.027
ling-0.027
itating-0.027
Tokenber
Token ID527
Feature activation-2.63e+00

Pos logit contributions
 Africa+0.128
 hours+0.119
 ages+0.108
aster+0.101
glass+0.101
Neg logit contributions
ling-0.166
 up-0.163
itating-0.151
 fighting-0.151
mill-0.151
Token started
Token ID2067
Feature activation-4.45e-01

Pos logit contributions
year+0.106
 Africa+0.095
glass+0.092
aster+0.091
prise+0.090
Neg logit contributions
 up-0.045
ling-0.035
 fighting-0.033
 bl-0.033
bing-0.030
Token to
Token ID284
Feature activation-1.38e+00

Pos logit contributions
year+0.015
ird+0.014
glass+0.014
 hours+0.013
aster+0.013
Neg logit contributions
 up-0.009
itating-0.009
arre-0.009
melon-0.008
 fighting-0.008
Token cut
Token ID2005
Feature activation-3.59e-01

Pos logit contributions
year+0.047
glass+0.044
ird+0.042
 hours+0.040
 minutes+0.040
Neg logit contributions
 up-0.029
itating-0.028
arre-0.027
melon-0.027
 fighting-0.026
Token his
Token ID465
Feature activation-1.79e+00

Pos logit contributions
year+0.014
glass+0.013
ird+0.013
 Bible+0.012
 hours+0.012
Neg logit contributions
 up-0.007
ling-0.005
arre-0.005
bing-0.005
itating-0.005
Token the
Token ID262
Feature activation-8.98e-01

Pos logit contributions
year+0.068
ird+0.066
glass+0.065
ris+0.062
 hours+0.061
Neg logit contributions
 up-0.032
 fighting-0.027
melon-0.027
bing-0.025
ling-0.025
Token bar
Token ID2318
Feature activation-5.39e+00

Pos logit contributions
year+0.032
ird+0.031
glass+0.030
ris+0.029
 hours+0.028
Neg logit contributions
 up-0.018
ling-0.016
itating-0.015
 fighting-0.015
arre-0.015
Tokenber
Token ID527
Feature activation-2.26e+00

Pos logit contributions
 hours+0.119
 ages+0.105
 Africa+0.100
ird+0.098
glass+0.095
Neg logit contributions
 up-0.176
ling-0.174
itating-0.165
bing-0.155
ing-0.154
Token.
Token ID13
Feature activation-1.56e+00

Pos logit contributions
year+0.086
 hours+0.079
glass+0.079
ird+0.079
 minutes+0.076
Neg logit contributions
 up-0.043
 fighting-0.035
ling-0.034
itating-0.033
bing-0.032
Token The
Token ID383
Feature activation-9.28e-01

Pos logit contributions
year+0.057
ird+0.056
glass+0.056
 hours+0.052
 minutes+0.051
Neg logit contributions
 up-0.026
arre-0.025
melon-0.025
itating-0.024
 fighting-0.023
Token bar
Token ID2318
Feature activation-5.59e+00

Pos logit contributions
year+0.036
glass+0.034
ird+0.033
ris+0.032
ibr+0.031
Neg logit contributions
 up-0.017
 hurting-0.013
 fighting-0.013
 same-0.013
ling-0.013
Tokenber
Token ID527
Feature activation-2.71e+00

Pos logit contributions
 Africa+0.122
 hours+0.116
 ages+0.106
ird+0.101
year+0.100
Neg logit contributions
ling-0.161
mill-0.155
 up-0.153
y-0.148
 fighting-0.146
Token smiled
Token ID13541
Feature activation+1.17e+00

Pos logit contributions
year+0.106
glass+0.094
ird+0.093
 Africa+0.093
lan+0.091
Neg logit contributions
 up-0.044
 fighting-0.039
ling-0.037
melon-0.035
mill-0.035
Token and
Token ID290
Feature activation-1.26e+00

Pos logit contributions
arre+0.021
itating+0.021
melon+0.020
ling+0.019
 up+0.019
Neg logit contributions
year-0.041
glass-0.039
ird-0.039
 hours-0.037
 minutes-0.037
Token talked
Token ID6619
Feature activation-1.33e+00

Pos logit contributions
year+0.045
glass+0.043
ird+0.042
aster+0.040
ibr+0.039
Neg logit contributions
 up-0.023
melon-0.022
 fighting-0.021
 Freddy-0.021
arre-0.020
Token to
Token ID284
Feature activation-5.68e-01

Pos logit contributions
ird+0.053
year+0.051
glass+0.050
aster+0.049
ocket+0.047
Neg logit contributions
 up-0.023
ling-0.017
 fighting-0.017
arre-0.016
itating-0.016
Token help
Token ID1037
Feature activation-4.15e-01

Pos logit contributions
 up+0.003
itating+0.003
arre+0.003
melon+0.002
 fighting+0.002
Neg logit contributions
year-0.003
glass-0.003
ird-0.003
 hours-0.002
 minutes-0.002
Token.
Token ID13
Feature activation-2.61e-01

Pos logit contributions
year+0.016
glass+0.015
ird+0.015
 hours+0.014
 minutes+0.014
Neg logit contributions
 up-0.007
itating-0.006
 fighting-0.006
ling-0.006
bing-0.006
Token
Token ID198
Feature activation-1.04e+00

Pos logit contributions
year+0.009
glass+0.009
ird+0.009
 hours+0.008
aster+0.008
Neg logit contributions
itating-0.005
arre-0.005
 up-0.005
melon-0.005
ling-0.004
Token
Token ID198
Feature activation+4.28e-01

Pos logit contributions
year+0.040
ird+0.039
glass+0.038
 hours+0.037
 minutes+0.036
Neg logit contributions
itating-0.016
 up-0.015
arre-0.015
ling-0.014
 fighting-0.013
TokenThe
Token ID464
Feature activation-2.01e+00

Pos logit contributions
itating+0.008
arre+0.008
 up+0.008
ling+0.008
melon+0.008
Neg logit contributions
year-0.014
ird-0.014
glass-0.014
 hours-0.013
aster-0.013
Token bar
Token ID2318
Feature activation-5.56e+00

Pos logit contributions
ird+0.079
glass+0.079
year+0.078
ocket+0.074
lan+0.072
Neg logit contributions
 up-0.032
 fighting-0.026
ling-0.025
 en-0.023
itating-0.022
Tokenber
Token ID527
Feature activation-2.43e+00

Pos logit contributions
 hours+0.128
 Africa+0.119
 ages+0.113
ird+0.107
aster+0.106
Neg logit contributions
ling-0.165
 up-0.155
mill-0.154
itating-0.150
y-0.147
Token said
Token ID531
Feature activation-1.10e+00

Pos logit contributions
year+0.089
glass+0.084
ird+0.081
 kil+0.081
 hours+0.080
Neg logit contributions
 up-0.041
arre-0.040
 fighting-0.038
ling-0.038
melon-0.037
Token to
Token ID284
Feature activation-1.31e+00

Pos logit contributions
year+0.044
ird+0.043
glass+0.042
ris+0.041
 hours+0.041
Neg logit contributions
 up-0.015
melon-0.014
 fighting-0.013
 Freddy-0.013
ling-0.013
Token the
Token ID262
Feature activation-1.96e+00

Pos logit contributions
year+0.048
ird+0.047
glass+0.046
 hours+0.044
 minutes+0.043
Neg logit contributions
 up-0.022
itating-0.021
arre-0.021
melon-0.021
 fighting-0.020
Token little
Token ID1310
Feature activation-1.41e+00

Pos logit contributions
year+0.079
glass+0.078
ird+0.076
lan+0.073
aster+0.072
Neg logit contributions
 up-0.031
 fighting-0.024
ling-0.023
itating-0.023
 same-0.022
Token happy
Token ID3772
Feature activation+1.76e+00

Pos logit contributions
year+0.027
glass+0.026
ird+0.026
 hours+0.025
aster+0.024
Neg logit contributions
itating-0.019
arre-0.019
 up-0.019
melon-0.018
ling-0.018
Token!
Token ID0
Feature activation-7.72e-01

Pos logit contributions
arre+0.046
itating+0.046
melon+0.043
rust+0.042
ling+0.042
Neg logit contributions
year-0.047
glass-0.046
 hours-0.045
 minutes-0.044
ird-0.044
Token
Token ID198
Feature activation-3.62e-01

Pos logit contributions
year+0.026
glass+0.025
ird+0.025
 hours+0.024
aster+0.023
Neg logit contributions
itating-0.015
arre-0.015
 up-0.014
melon-0.014
ling-0.014
Token
Token ID198
Feature activation+3.09e-01

Pos logit contributions
year+0.013
ird+0.013
glass+0.012
 hours+0.012
 minutes+0.012
Neg logit contributions
itating-0.006
arre-0.006
 up-0.006
ling-0.006
melon-0.006
TokenThe
Token ID464
Feature activation-1.97e+00

Pos logit contributions
arre+0.007
itating+0.007
 up+0.006
melon+0.006
ling+0.006
Neg logit contributions
year-0.010
glass-0.009
ird-0.009
 hours-0.009
aster-0.009
Token bar
Token ID2318
Feature activation-5.52e+00

Pos logit contributions
ird+0.072
year+0.071
glass+0.070
ris+0.065
 hours+0.064
Neg logit contributions
 up-0.037
 fighting-0.030
ling-0.030
itating-0.029
 en-0.028
Tokenber
Token ID527
Feature activation-2.29e+00

Pos logit contributions
 hours+0.120
 Africa+0.119
 ages+0.105
aster+0.104
ird+0.100
Neg logit contributions
ling-0.166
mill-0.159
 up-0.156
itating-0.151
 fighting-0.150
Token worked
Token ID3111
Feature activation+4.04e-01

Pos logit contributions
year+0.086
glass+0.080
ird+0.077
 kil+0.077
 Africa+0.076
Neg logit contributions
 up-0.038
arre-0.036
 fighting-0.034
melon-0.034
ling-0.033
Token very
Token ID845
Feature activation-1.73e+00

Pos logit contributions
 up+0.009
itating+0.009
arre+0.009
ling+0.008
melon+0.008
Neg logit contributions
year-0.013
ird-0.012
glass-0.012
aster-0.011
 hours-0.011
Token carefully
Token ID7773
Feature activation-4.49e-01

Pos logit contributions
year+0.054
ird+0.054
glass+0.053
 hours+0.049
aster+0.049
Neg logit contributions
 up-0.037
itating-0.037
arre-0.036
ling-0.035
melon-0.035
Token and
Token ID290
Feature activation-7.37e-01

Pos logit contributions
year+0.018
ird+0.017
glass+0.017
 hours+0.016
aster+0.016
Neg logit contributions
 up-0.007
arre-0.006
 fighting-0.006
melon-0.006
itating-0.006
Token,
Token ID11
Feature activation-1.23e-01

Pos logit contributions
year+0.014
glass+0.014
ird+0.014
 hours+0.013
aster+0.013
Neg logit contributions
itating-0.008
arre-0.008
 up-0.007
melon-0.007
ling-0.007
Token "
Token ID366
Feature activation+1.01e+00

Pos logit contributions
year+0.004
glass+0.004
ird+0.004
 hours+0.004
 minutes+0.004
Neg logit contributions
itating-0.002
arre-0.002
 up-0.002
ling-0.002
melon-0.002
TokenNo
Token ID2949
Feature activation-4.74e-01

Pos logit contributions
itating+0.022
arre+0.022
melon+0.022
 up+0.022
ling+0.021
Neg logit contributions
year-0.032
glass-0.030
ird-0.030
 hours-0.029
aster-0.028
Token,
Token ID11
Feature activation-3.66e-02

Pos logit contributions
ird+0.013
 hours+0.013
 minutes+0.013
glass+0.012
aster+0.012
Neg logit contributions
itating-0.012
arre-0.011
melon-0.011
ling-0.011
flower-0.011
Token mud
Token ID17492
Feature activation+2.27e+00

Pos logit contributions
ird+0.001
year+0.001
glass+0.001
 hours+0.001
 minutes+0.001
Neg logit contributions
itating-0.001
arre-0.001
ling-0.001
bing-0.001
 up-0.001
Token isn
Token ID2125
Feature activation+3.57e+00

Pos logit contributions
itating+0.063
arre+0.054
rust+0.054
ling+0.050
Being+0.049
Neg logit contributions
ird-0.058
glass-0.056
 minutes-0.056
 hours-0.055
aster-0.053
Token't
Token ID470
Feature activation+1.36e+00

Pos logit contributions
lings+0.103
rust+0.102
liness+0.098
arre+0.098
ling+0.098
Neg logit contributions
ird-0.076
year-0.074
aster-0.073
ibrarian-0.064
lock-0.063
Token for
Token ID329
Feature activation-6.83e-01

Pos logit contributions
itating+0.038
arre+0.035
ishly+0.034
rust+0.034
flower+0.033
Neg logit contributions
 minutes-0.035
ird-0.034
 hours-0.034
year-0.033
aster-0.032
Token sharing
Token ID7373
Feature activation+1.34e+00

Pos logit contributions
year+0.025
glass+0.024
ird+0.024
 hours+0.022
aster+0.022
Neg logit contributions
arre-0.011
itating-0.011
 up-0.011
melon-0.011
ling-0.010
Token."
Token ID526
Feature activation+7.28e-01

Pos logit contributions
itating+0.030
arre+0.030
melon+0.028
ling+0.028
 up+0.027
Neg logit contributions
year-0.040
ird-0.039
glass-0.039
 hours-0.037
 minutes-0.036
Token
Token ID198
Feature activation-3.64e-01

Pos logit contributions
itating+0.017
ling+0.016
flower+0.016
bing+0.016
mill+0.016
Neg logit contributions
year-0.021
glass-0.021
ird-0.021
 hours-0.020
 minutes-0.020
Token
Token ID198
Feature activation+1.21e-01

Pos logit contributions
melon+0.011
ling+0.011
itating+0.010
flower+0.010
arre+0.010
Neg logit contributions
ird-0.012
year-0.011
glass-0.011
ibr-0.010
aster-0.010
Token
Token ID198
Feature activation+1.77e+00

Pos logit contributions
itating+0.003
arre+0.003
melon+0.003
ling+0.003
 up+0.003
Neg logit contributions
year-0.004
ird-0.004
glass-0.004
 hours-0.003
aster-0.003
TokenTim
Token ID14967
Feature activation+5.60e-01

Pos logit contributions
itating+0.047
melon+0.045
 up+0.045
arre+0.044
flower+0.044
Neg logit contributions
year-0.048
ird-0.045
glass-0.045
 hours-0.044
aster-0.043
Tokenmy
Token ID1820
Feature activation+4.49e-01

Pos logit contributions
melon+0.010
itating+0.010
bing+0.009
 fighting+0.009
izing+0.009
Neg logit contributions
year-0.019
ird-0.019
ris-0.018
glass-0.018
 hours-0.017
Token loved
Token ID6151
Feature activation+1.34e+00

Pos logit contributions
melon+0.013
itating+0.013
rooms+0.012
inas+0.012
bing+0.012
Neg logit contributions
ird-0.010
year-0.010
glass-0.010
 hours-0.009
ibrarian-0.009
Token the
Token ID262
Feature activation+3.67e+00

Pos logit contributions
itating+0.038
ling+0.035
flower+0.034
arre+0.033
lings+0.032
Neg logit contributions
 hours-0.036
aster-0.035
year-0.034
glass-0.034
 minutes-0.031
Token idea
Token ID2126
Feature activation+1.23e+00

Pos logit contributions
flies+0.117
ric+0.115
bing+0.114
lings+0.109
inated+0.107
Neg logit contributions
 hours-0.063
book-0.060
year-0.059
 month-0.059
 hour-0.056
Token and
Token ID290
Feature activation+1.16e+00

Pos logit contributions
itating+0.027
arre+0.027
melon+0.026
rust+0.025
ling+0.025
Neg logit contributions
year-0.038
ird-0.037
glass-0.036
 hours-0.034
 minutes-0.034
Token he
Token ID339
Feature activation+2.02e-01

Pos logit contributions
itating+0.031
arre+0.029
ling+0.028
bing+0.028
melon+0.027
Neg logit contributions
year-0.032
ird-0.030
glass-0.030
 hours-0.029
 minutes-0.028
Token and
Token ID290
Feature activation+1.13e+00

Pos logit contributions
itating+0.005
arre+0.005
bing+0.005
melon+0.005
ling+0.005
Neg logit contributions
year-0.006
ird-0.006
glass-0.005
 hours-0.005
 minutes-0.005
Token his
Token ID465
Feature activation+8.82e-01

Pos logit contributions
itating+0.030
arre+0.029
ling+0.027
melon+0.027
 up+0.027
Neg logit contributions
year-0.032
glass-0.031
ird-0.030
 hours-0.029
aster-0.028
Token.
Token ID13
Feature activation+3.74e-01

Pos logit contributions
 up+0.003
melon+0.003
 fighting+0.003
ling+0.003
itating+0.003
Neg logit contributions
year-0.008
glass-0.007
ird-0.007
aster-0.007
ibr-0.007
Token<|endoftext|>
Token ID50256
Feature activation+1.12e+00

Pos logit contributions
ling+0.008
itating+0.008
arre+0.008
mill+0.008
bing+0.008
Neg logit contributions
year-0.011
ird-0.011
glass-0.010
aster-0.010
 hours-0.010
TokenTom
Token ID13787
Feature activation+1.09e+00

Pos logit contributions
itating+0.024
arre+0.023
 up+0.022
melon+0.022
ling+0.022
Neg logit contributions
year-0.036
ird-0.034
glass-0.034
 hours-0.032
aster-0.032
Tokenmy
Token ID1820
Feature activation+1.77e+00

Pos logit contributions
itating+0.013
arre+0.012
melon+0.011
ling+0.011
 up+0.011
Neg logit contributions
year-0.045
ird-0.043
glass-0.043
 hours-0.042
aster-0.041
Token loved
Token ID6151
Feature activation+6.79e-01

Pos logit contributions
itating+0.043
arre+0.040
melon+0.039
ling+0.038
bing+0.037
Neg logit contributions
year-0.051
ird-0.050
glass-0.050
 hours-0.048
aster-0.047
Token baths
Token ID49076
Feature activation+3.69e+00

Pos logit contributions
itating+0.017
arre+0.016
ling+0.015
 up+0.015
melon+0.015
Neg logit contributions
year-0.020
glass-0.019
ird-0.019
 hours-0.018
aster-0.018
Token.
Token ID13
Feature activation+1.21e+00

Pos logit contributions
arre+0.115
itating+0.112
ucker+0.111
rust+0.110
ety+0.105
Neg logit contributions
year-0.072
ird-0.065
 hours-0.063
glass-0.059
book-0.057
Token Every
Token ID3887
Feature activation+3.01e-01

Pos logit contributions
itating+0.024
arre+0.024
 up+0.023
ling+0.023
melon+0.022
Neg logit contributions
year-0.040
ird-0.038
glass-0.038
 hours-0.036
aster-0.036
Token day
Token ID1110
Feature activation-6.15e-02

Pos logit contributions
itating+0.009
arre+0.009
melon+0.008
bing+0.008
rust+0.008
Neg logit contributions
year-0.007
ird-0.007
 hours-0.007
glass-0.007
aster-0.006
Token,
Token ID11
Feature activation+5.05e-02

Pos logit contributions
year+0.002
glass+0.002
ibr+0.002
ibrarian+0.002
ird+0.002
Neg logit contributions
 up-0.001
melon-0.001
 fighting-0.001
 Freddy-0.001
 sharing-0.001
Token he
Token ID339
Feature activation+1.96e+00

Pos logit contributions
 up+0.001
melon+0.001
arre+0.001
 fighting+0.001
itating+0.001
Neg logit contributions
year-0.002
glass-0.002
ird-0.002
ibr-0.002
 hours-0.002
Token that
Token ID326
Feature activation+1.47e+00

Pos logit contributions
arre+0.002
itating+0.002
melon+0.002
ling+0.002
 up+0.001
Neg logit contributions
year-0.003
glass-0.003
ird-0.003
 hours-0.003
aster-0.003
Token even
Token ID772
Feature activation-5.43e-01

Pos logit contributions
itating+0.035
arre+0.033
ling+0.032
melon+0.031
ucker+0.030
Neg logit contributions
year-0.043
glass-0.041
 hours-0.041
ird-0.040
aster-0.039
Token though
Token ID996
Feature activation+6.59e-01

Pos logit contributions
year+0.017
ird+0.016
glass+0.016
 hours+0.015
aster+0.015
Neg logit contributions
itating-0.012
arre-0.012
 up-0.011
melon-0.011
ling-0.011
Token he
Token ID339
Feature activation+3.20e-01

Pos logit contributions
itating+0.016
arre+0.016
ling+0.014
melon+0.014
 up+0.014
Neg logit contributions
year-0.019
ird-0.018
glass-0.018
 hours-0.018
aster-0.017
Token wanted
Token ID2227
Feature activation+3.73e-01

Pos logit contributions
itating+0.006
arre+0.006
melon+0.006
 up+0.006
ling+0.006
Neg logit contributions
year-0.011
ird-0.010
glass-0.010
 hours-0.010
aster-0.009
Token the
Token ID262
Feature activation+3.76e+00

Pos logit contributions
itating+0.009
arre+0.009
ling+0.009
melon+0.008
bing+0.008
Neg logit contributions
year-0.010
glass-0.010
ird-0.010
 hours-0.010
aster-0.009
Token mint
Token ID21043
Feature activation+1.76e+00

Pos logit contributions
itating+0.107
arre+0.105
ocked+0.101
inas+0.100
ning+0.099
Neg logit contributions
 hours-0.080
ird-0.077
 minutes-0.071
year-0.070
 allowance-0.070
Token,
Token ID11
Feature activation+1.19e+00

Pos logit contributions
itating+0.042
arre+0.040
inas+0.035
rust+0.035
bing+0.035
Neg logit contributions
ird-0.048
year-0.048
 hours-0.045
lock-0.043
glass-0.043
Token it
Token ID340
Feature activation+2.18e+00

Pos logit contributions
itating+0.031
arre+0.030
ling+0.028
melon+0.028
 up+0.028
Neg logit contributions
year-0.033
glass-0.031
ird-0.031
 hours-0.030
aster-0.029
Token would
Token ID561
Feature activation+1.78e+00

Pos logit contributions
itating+0.054
bing+0.049
arre+0.049
melon+0.049
ety+0.048
Neg logit contributions
year-0.054
ird-0.053
 hours-0.051
glass-0.050
 kil-0.049
Token have
Token ID423
Feature activation+1.66e+00

Pos logit contributions
arre+0.045
itating+0.042
ling+0.042
bing+0.041
ning+0.040
Neg logit contributions
aster-0.046
ird-0.046
year-0.045
 hours-0.044
glass-0.044
Token.
Token ID13
Feature activation+3.12e-01

Pos logit contributions
arre+0.000
itating+0.000
melon+0.000
rust+0.000
ling+0.000
Neg logit contributions
year-0.001
ird-0.000
glass-0.000
 hours-0.000
aster-0.000
Token "
Token ID366
Feature activation+4.32e-01

Pos logit contributions
rust+0.007
ling+0.007
arre+0.007
itating+0.007
melon+0.007
Neg logit contributions
year-0.009
glass-0.008
ird-0.008
 minutes-0.008
aster-0.008
TokenYou
Token ID1639
Feature activation-7.72e-01

Pos logit contributions
melon+0.008
 up+0.008
itating+0.008
arre+0.008
ling+0.007
Neg logit contributions
year-0.015
glass-0.014
aster-0.014
ird-0.014
 hours-0.013
Token will
Token ID481
Feature activation+1.23e+00

Pos logit contributions
year+0.026
glass+0.024
ird+0.024
 hours+0.023
aster+0.023
Neg logit contributions
arre-0.015
itating-0.015
 up-0.015
melon-0.014
ling-0.014
Token hit
Token ID2277
Feature activation+1.94e+00

Pos logit contributions
arre+0.026
ling+0.026
itating+0.026
melon+0.024
bing+0.024
Neg logit contributions
ird-0.038
glass-0.036
year-0.036
aster-0.036
 hours-0.035
Token the
Token ID262
Feature activation+3.86e+00

Pos logit contributions
itating+0.039
arre+0.038
melon+0.035
inas+0.034
ling+0.034
Neg logit contributions
year-0.061
glass-0.060
 hours-0.060
ird-0.060
aster-0.060
Token fence
Token ID13990
Feature activation+2.73e-01

Pos logit contributions
itating+0.071
inas+0.070
ocked+0.070
ucker+0.069
arre+0.069
Neg logit contributions
glass-0.108
year-0.107
 slot-0.106
 hours-0.105
lock-0.104
Token!"
Token ID2474
Feature activation+3.83e-01

Pos logit contributions
arre+0.005
itating+0.005
melon+0.005
ling+0.005
rust+0.005
Neg logit contributions
year-0.009
ird-0.009
glass-0.009
 hours-0.008
aster-0.008
Token
Token ID198
Feature activation-4.07e-01

Pos logit contributions
arre+0.008
itating+0.008
melon+0.008
ling+0.008
 up+0.008
Neg logit contributions
year-0.012
glass-0.012
ird-0.012
 hours-0.011
aster-0.011
Token
Token ID198
Feature activation+7.04e-01

Pos logit contributions
year+0.014
ird+0.014
glass+0.014
 hours+0.013
aster+0.013
Neg logit contributions
itating-0.007
arre-0.007
 up-0.007
melon-0.007
ling-0.007
TokenBut
Token ID1537
Feature activation+9.57e-01

Pos logit contributions
itating+0.017
arre+0.017
melon+0.017
 up+0.016
ling+0.016
Neg logit contributions
year-0.020
glass-0.020
ird-0.019
aster-0.019
 hours-0.018
Token day
Token ID1110
Feature activation-1.45e-01

Pos logit contributions
year+0.060
ird+0.058
glass+0.058
 hours+0.053
ibr+0.053
Neg logit contributions
 up-0.041
arre-0.040
ling-0.039
itating-0.039
melon-0.039
Token,
Token ID11
Feature activation+2.78e-01

Pos logit contributions
year+0.005
glass+0.005
ird+0.005
 hours+0.005
 minutes+0.004
Neg logit contributions
itating-0.003
arre-0.003
 up-0.003
melon-0.003
ling-0.003
Token they
Token ID484
Feature activation+1.63e+00

Pos logit contributions
arre+0.006
itating+0.006
ling+0.005
melon+0.005
 up+0.005
Neg logit contributions
year-0.009
glass-0.009
ird-0.008
 hours-0.008
 minutes-0.008
Token decided
Token ID3066
Feature activation+1.30e+00

Pos logit contributions
itating+0.040
arre+0.039
bing+0.038
 up+0.038
melon+0.038
Neg logit contributions
year-0.046
ird-0.045
glass-0.044
 hours-0.042
 minutes-0.039
Token to
Token ID284
Feature activation+1.40e+00

Pos logit contributions
arre+0.045
itating+0.044
flower+0.043
ling+0.043
ucker+0.043
Neg logit contributions
glass-0.024
 hours-0.022
year-0.021
 minutes-0.020
lan-0.020
Token play
Token ID711
Feature activation+1.31e+00

Pos logit contributions
itating+0.035
ling+0.034
arre+0.034
bing+0.032
melon+0.032
Neg logit contributions
ird-0.039
glass-0.037
year-0.036
aster-0.035
 hours-0.035
Token a
Token ID257
Feature activation-2.44e-03

Pos logit contributions
 up+0.028
arre+0.028
itating+0.027
 fighting+0.026
ling+0.026
Neg logit contributions
year-0.042
ird-0.040
glass-0.039
aster-0.037
 hours-0.036
Token game
Token ID983
Feature activation+8.29e-01

Pos logit contributions
year+0.000
ird+0.000
glass+0.000
 hours+0.000
aster+0.000
Neg logit contributions
itating-0.000
arre-0.000
 up-0.000
melon-0.000
ling-0.000
Token.
Token ID13
Feature activation+1.06e+00

Pos logit contributions
 up+0.017
itating+0.016
arre+0.016
melon+0.015
 fighting+0.015
Neg logit contributions
year-0.028
ird-0.026
glass-0.026
 hours-0.024
aster-0.024
Token Joe
Token ID5689
Feature activation+1.30e+00

Pos logit contributions
ling+0.025
arre+0.024
itating+0.024
bing+0.023
melon+0.023
Neg logit contributions
year-0.031
glass-0.030
ird-0.029
 hours-0.029
 minutes-0.028
Token wanted
Token ID2227
Feature activation-9.89e-01

Pos logit contributions
itating+0.028
arre+0.027
 up+0.026
melon+0.026
ling+0.026
Neg logit contributions
year-0.041
ird-0.040
glass-0.039
 hours-0.037
aster-0.036
Token.
Token ID13
Feature activation+1.29e+00

Pos logit contributions
 up+0.014
itating+0.013
arre+0.013
ling+0.013
 fighting+0.013
Neg logit contributions
year-0.029
glass-0.027
ird-0.027
aster-0.026
 hours-0.025
Token
Token ID198
Feature activation+1.40e-01

Pos logit contributions
itating+0.025
arre+0.025
ling+0.024
 up+0.024
melon+0.023
Neg logit contributions
year-0.043
ird-0.042
glass-0.041
 hours-0.039
aster-0.039
Token
Token ID198
Feature activation+1.47e+00

Pos logit contributions
itating+0.003
arre+0.003
 up+0.003
melon+0.003
ling+0.003
Neg logit contributions
year-0.005
ird-0.004
glass-0.004
 hours-0.004
aster-0.004
TokenWhen
Token ID2215
Feature activation-1.56e-01

Pos logit contributions
melon+0.035
itating+0.034
arre+0.034
 up+0.034
ling+0.033
Neg logit contributions
year-0.044
ird-0.041
glass-0.041
 hours-0.039
aster-0.039
Token they
Token ID484
Feature activation-7.49e-01

Pos logit contributions
year+0.006
glass+0.006
ird+0.006
 kil+0.006
ibr+0.006
Neg logit contributions
 up-0.002
 fighting-0.002
melon-0.002
arre-0.002
 Freddy-0.002
Token arrived
Token ID5284
Feature activation+7.39e-01

Pos logit contributions
glass+0.026
year+0.026
 kil+0.025
lan+0.025
 Bible+0.024
Neg logit contributions
 up-0.013
arre-0.012
melon-0.012
 fighting-0.012
itating-0.012
Token at
Token ID379
Feature activation+5.53e-02

Pos logit contributions
itating+0.015
arre+0.015
 up+0.014
melon+0.014
ling+0.014
Neg logit contributions
year-0.024
ird-0.023
glass-0.023
 hours-0.022
aster-0.022
Token grandma
Token ID49890
Feature activation+3.54e-01

Pos logit contributions
 up+0.001
 fighting+0.001
melon+0.001
 Freddy+0.001
arre+0.001
Neg logit contributions
year-0.002
glass-0.002
ird-0.002
 kil-0.002
lock-0.002
Token's
Token ID338
Feature activation-5.45e-01

Pos logit contributions
itating+0.004
arre+0.004
 up+0.004
melon+0.004
ling+0.004
Neg logit contributions
year-0.015
glass-0.014
ird-0.014
 hours-0.014
aster-0.013
Token house
Token ID2156
Feature activation-5.37e-01

Pos logit contributions
year+0.019
ird+0.017
glass+0.017
 hours+0.016
aster+0.016
Neg logit contributions
 up-0.011
arre-0.011
itating-0.011
 fighting-0.010
melon-0.010
Token,
Token ID11
Feature activation+8.03e-01

Pos logit contributions
year+0.021
glass+0.020
ird+0.020
 hours+0.019
 kil+0.019
Neg logit contributions
 up-0.008
itating-0.007
 fighting-0.007
arre-0.007
melon-0.007
Token They
Token ID1119
Feature activation+1.31e-01

Pos logit contributions
itating+0.009
arre+0.009
 up+0.008
melon+0.008
ling+0.008
Neg logit contributions
year-0.025
glass-0.024
ird-0.024
 hours-0.023
aster-0.023
Token are
Token ID389
Feature activation+1.43e+00

Pos logit contributions
itating+0.003
arre+0.003
 up+0.003
melon+0.003
ling+0.003
Neg logit contributions
year-0.004
ird-0.004
glass-0.004
 hours-0.004
aster-0.004
Token wet
Token ID9583
Feature activation+1.96e+00

Pos logit contributions
itating+0.028
arre+0.027
melon+0.026
 up+0.026
ling+0.025
Neg logit contributions
year-0.049
ird-0.046
glass-0.046
 hours-0.044
 minutes-0.043
Token and
Token ID290
Feature activation+1.63e+00

Pos logit contributions
itating+0.040
arre+0.038
melon+0.037
rust+0.036
ucker+0.034
Neg logit contributions
year-0.064
glass-0.062
ird-0.061
 hours-0.060
 minutes-0.060
Token dirty
Token ID11841
Feature activation+1.74e+00

Pos logit contributions
itating+0.030
arre+0.029
 up+0.027
melon+0.027
ling+0.027
Neg logit contributions
year-0.057
ird-0.054
glass-0.054
 hours-0.052
 minutes-0.051
Token.
Token ID13
Feature activation+4.12e-01

Pos logit contributions
itating+0.039
arre+0.038
melon+0.036
rust+0.035
 up+0.035
Neg logit contributions
year-0.053
ird-0.051
glass-0.051
 hours-0.048
 minutes-0.048
Token They
Token ID1119
Feature activation+2.32e-01

Pos logit contributions
arre+0.006
itating+0.006
 up+0.005
melon+0.005
ling+0.005
Neg logit contributions
year-0.016
glass-0.016
ird-0.016
 hours-0.015
aster-0.015
Token are
Token ID389
Feature activation+1.58e+00

Pos logit contributions
arre+0.005
 up+0.005
itating+0.005
melon+0.004
 fighting+0.004
Neg logit contributions
year-0.008
glass-0.007
ird-0.007
 hours-0.007
 minutes-0.007
Token sorry
Token ID7926
Feature activation+2.02e+00

Pos logit contributions
itating+0.029
arre+0.028
 up+0.027
melon+0.026
ling+0.026
Neg logit contributions
year-0.055
ird-0.053
glass-0.053
 hours-0.051
 minutes-0.050
Token.
Token ID13
Feature activation+7.40e-01

Pos logit contributions
itating+0.055
arre+0.054
melon+0.052
rust+0.052
bing+0.050
Neg logit contributions
year-0.051
 minutes-0.050
ird-0.049
glass-0.048
 hours-0.048
Token They
Token ID1119
Feature activation+5.19e-01

Pos logit contributions
itating+0.011
arre+0.011
 up+0.011
melon+0.011
ling+0.010
Neg logit contributions
year-0.028
glass-0.027
ird-0.027
 hours-0.026
aster-0.025
Token follow
Token ID1061
Feature activation-1.09e+00

Pos logit contributions
year+0.026
glass+0.024
ird+0.024
 hours+0.023
 minutes+0.022
Neg logit contributions
 up-0.016
itating-0.016
arre-0.016
melon-0.015
 fighting-0.015
Token.
Token ID13
Feature activation-4.80e-01

Pos logit contributions
year+0.043
ird+0.042
glass+0.041
ibr+0.038
 hours+0.038
Neg logit contributions
 up-0.018
ling-0.015
itating-0.014
 fighting-0.014
bing-0.014
Token As
Token ID1081
Feature activation-1.15e+00

Pos logit contributions
year+0.016
glass+0.016
ird+0.016
 hours+0.015
aster+0.015
Neg logit contributions
arre-0.009
itating-0.009
 up-0.009
melon-0.009
ling-0.008
Token they
Token ID484
Feature activation-1.59e+00

Pos logit contributions
year+0.047
ird+0.047
glass+0.045
ibr+0.044
ibrarian+0.043
Neg logit contributions
 up-0.017
 Freddy-0.016
 fighting-0.015
 accidents-0.013
 sharing-0.012
Token got
Token ID1392
Feature activation-6.29e-01

Pos logit contributions
glass+0.054
lan+0.053
aster+0.052
 Bible+0.052
 Africa+0.051
Neg logit contributions
ling-0.026
 up-0.026
arre-0.026
 fighting-0.024
 qu-0.024
Token closer
Token ID5699
Feature activation+6.52e-01

Pos logit contributions
ird+0.020
year+0.020
glass+0.020
aster+0.019
 hours+0.018
Neg logit contributions
 up-0.014
itating-0.013
arre-0.012
melon-0.012
 fighting-0.012
Token,
Token ID11
Feature activation-2.28e-01

Pos logit contributions
arre+0.013
itating+0.013
melon+0.012
 up+0.012
ling+0.012
Neg logit contributions
year-0.021
glass-0.020
ird-0.020
 hours-0.020
 minutes-0.019
Token Jack
Token ID3619
Feature activation-2.67e-01

Pos logit contributions
year+0.008
ird+0.008
glass+0.008
 hours+0.008
ibr+0.008
Neg logit contributions
 up-0.004
itating-0.004
arre-0.004
melon-0.003
 fighting-0.003
Token could
Token ID714
Feature activation-8.83e-01

Pos logit contributions
year+0.009
ird+0.009
glass+0.009
aster+0.008
 hours+0.008
Neg logit contributions
 up-0.005
arre-0.005
itating-0.005
ling-0.005
melon-0.005
Token see
Token ID766
Feature activation-4.22e-01

Pos logit contributions
year+0.022
glass+0.021
ird+0.021
 hours+0.019
aster+0.019
Neg logit contributions
arre-0.025
itating-0.024
melon-0.024
 up-0.024
ling-0.023
Token his
Token ID465
Feature activation-9.44e-01

Pos logit contributions
year+0.015
ird+0.015
glass+0.014
aster+0.013
lock+0.013
Neg logit contributions
 up-0.008
itating-0.007
ling-0.007
arre-0.007
 fighting-0.007
Token She
Token ID1375
Feature activation+7.68e-01

Pos logit contributions
ling+0.017
itating+0.016
arre+0.015
bing+0.015
mill+0.014
Neg logit contributions
year-0.027
ird-0.026
 hours-0.025
glass-0.025
aster-0.025
Token touched
Token ID12615
Feature activation+1.23e+00

Pos logit contributions
 up+0.015
arre+0.014
itating+0.014
ling+0.014
melon+0.014
Neg logit contributions
year-0.026
ird-0.025
glass-0.025
aster-0.023
ibr-0.023
Token the
Token ID262
Feature activation+1.49e+00

Pos logit contributions
itating+0.030
arre+0.030
melon+0.027
rust+0.027
 fighting+0.026
Neg logit contributions
year-0.036
 hours-0.034
ird-0.033
aster-0.033
 minutes-0.032
Token icy
Token ID30284
Feature activation+1.23e+00

Pos logit contributions
arre+0.030
itating+0.029
melon+0.026
rust+0.026
 up+0.026
Neg logit contributions
year-0.048
glass-0.047
ird-0.045
 hours-0.045
 minutes-0.043
Token water
Token ID1660
Feature activation+2.07e+00

Pos logit contributions
itating+0.032
arre+0.030
rust+0.027
 fighting+0.026
inas+0.026
Neg logit contributions
glass-0.034
 hours-0.033
 minutes-0.033
year-0.032
aster-0.032
Token with
Token ID351
Feature activation+6.24e-01

Pos logit contributions
arre+0.050
rust+0.050
itating+0.047
ety+0.043
inas+0.042
Neg logit contributions
year-0.054
glass-0.054
 minutes-0.052
ird-0.052
aster-0.051
Token her
Token ID607
Feature activation+1.43e-01

Pos logit contributions
itating+0.013
arre+0.013
melon+0.012
 up+0.012
ling+0.012
Neg logit contributions
year-0.020
ird-0.019
glass-0.019
 hours-0.019
aster-0.018
Token feet
Token ID3625
Feature activation+1.04e+00

Pos logit contributions
itating+0.004
arre+0.004
 up+0.004
melon+0.003
ling+0.003
Neg logit contributions
year-0.004
glass-0.004
ird-0.004
 hours-0.004
aster-0.003
Token,
Token ID11
Feature activation+1.68e+00

Pos logit contributions
rust+0.026
arre+0.025
itating+0.023
 Freddy+0.023
melon+0.022
Neg logit contributions
ird-0.028
year-0.028
glass-0.027
lock-0.026
 hours-0.025
Token making
Token ID1642
Feature activation+6.32e-01

Pos logit contributions
arre+0.039
itating+0.039
rust+0.037
melon+0.034
rooms+0.033
Neg logit contributions
 hours-0.049
 minutes-0.048
year-0.046
aster-0.045
glass-0.045
Token silly
Token ID14397
Feature activation+2.48e+00

Pos logit contributions
itating+0.015
arre+0.014
melon+0.014
rust+0.013
flower+0.013
Neg logit contributions
year-0.019
 hours-0.018
glass-0.018
ird-0.017
aster-0.017
Token help
Token ID1037
Feature activation+9.69e-02

Pos logit contributions
 up+0.003
itating+0.003
arre+0.003
melon+0.003
 fighting+0.002
Neg logit contributions
year-0.004
glass-0.003
ird-0.003
 hours-0.003
 minutes-0.003
Token.
Token ID13
Feature activation+7.45e-01

Pos logit contributions
 up+0.002
itating+0.002
arre+0.002
melon+0.001
ling+0.001
Neg logit contributions
year-0.004
glass-0.003
ird-0.003
 hours-0.003
 minutes-0.003
Token She
Token ID1375
Feature activation+1.32e-01

Pos logit contributions
itating+0.016
arre+0.015
ling+0.015
melon+0.014
 up+0.014
Neg logit contributions
year-0.024
ird-0.023
glass-0.023
 hours-0.022
 minutes-0.021
Token told
Token ID1297
Feature activation-4.82e-01

Pos logit contributions
arre+0.003
 up+0.003
itating+0.002
melon+0.002
ling+0.002
Neg logit contributions
year-0.004
ird-0.004
glass-0.004
aster-0.004
 hours-0.004
Token her
Token ID607
Feature activation-7.08e-01

Pos logit contributions
year+0.016
glass+0.016
ird+0.016
 hours+0.015
aster+0.015
Neg logit contributions
itating-0.009
arre-0.009
ling-0.009
melon-0.009
 up-0.008
Token friends
Token ID2460
Feature activation+1.86e-01

Pos logit contributions
year+0.020
ird+0.019
glass+0.019
 hours+0.018
 minutes+0.017
Neg logit contributions
arre-0.018
 up-0.018
itating-0.017
melon-0.017
ling-0.017
Token about
Token ID546
Feature activation+6.89e-01

Pos logit contributions
itating+0.005
arre+0.005
bing+0.005
flower+0.005
rust+0.005
Neg logit contributions
year-0.005
ird-0.005
glass-0.005
 hours-0.005
aster-0.004
Token the
Token ID262
Feature activation+1.01e+00

Pos logit contributions
arre+0.015
itating+0.014
ling+0.013
melon+0.013
 up+0.013
Neg logit contributions
year-0.022
glass-0.021
ird-0.021
 hours-0.021
aster-0.020
Token lost
Token ID2626
Feature activation-1.42e-01

Pos logit contributions
itating+0.031
arre+0.030
melon+0.029
ling+0.028
bing+0.028
Neg logit contributions
glass-0.023
year-0.023
ird-0.022
 hours-0.021
aster-0.021
Token dog
Token ID3290
Feature activation+6.67e-01

Pos logit contributions
year+0.004
ird+0.004
glass+0.004
 hours+0.004
 minutes+0.004
Neg logit contributions
arre-0.003
 up-0.003
itating-0.003
melon-0.003
ling-0.003
Token and
Token ID290
Feature activation+1.18e+00

Pos logit contributions
itating+0.012
arre+0.012
melon+0.010
rust+0.010
ling+0.010
Neg logit contributions
year-0.023
ird-0.023
glass-0.023
 hours-0.022
aster-0.021
Token football
Token ID4346
Feature activation+6.36e-02

Pos logit contributions
bing+0.068
itating+0.065
inas+0.065
ucker+0.064
arre+0.063
Neg logit contributions
glass-0.050
 passport-0.049
ird-0.048
 hours-0.047
year-0.047
Token and
Token ID290
Feature activation+3.18e-01

Pos logit contributions
itating+0.001
arre+0.001
 up+0.001
melon+0.001
ling+0.001
Neg logit contributions
year-0.002
glass-0.002
ird-0.002
 hours-0.002
aster-0.002
Token Tim
Token ID5045
Feature activation-5.85e-01

Pos logit contributions
arre+0.006
itating+0.006
 up+0.006
melon+0.006
ling+0.006
Neg logit contributions
year-0.011
glass-0.010
ird-0.010
 hours-0.010
aster-0.010
Tokenmy
Token ID1820
Feature activation+1.36e-01

Pos logit contributions
year+0.027
glass+0.026
ird+0.026
 hours+0.025
aster+0.025
Neg logit contributions
itating-0.004
arre-0.004
 up-0.004
melon-0.004
ling-0.004
Token played
Token ID2826
Feature activation+1.39e+00

Pos logit contributions
itating+0.003
arre+0.003
 up+0.003
melon+0.003
ling+0.003
Neg logit contributions
year-0.004
ird-0.004
glass-0.004
 hours-0.004
aster-0.004
Token with
Token ID351
Feature activation+1.25e-01

Pos logit contributions
itating+0.027
arre+0.026
melon+0.025
ling+0.024
flower+0.023
Neg logit contributions
year-0.048
glass-0.046
ird-0.045
 hours-0.044
aster-0.043
Token the
Token ID262
Feature activation+2.45e+00

Pos logit contributions
 up+0.002
arre+0.002
itating+0.002
melon+0.002
ling+0.002
Neg logit contributions
year-0.005
ird-0.004
glass-0.004
 hours-0.004
aster-0.004
Token other
Token ID584
Feature activation+1.06e+00

Pos logit contributions
bing+0.068
ucker+0.066
itating+0.066
inas+0.064
flower+0.062
Neg logit contributions
 hours-0.057
year-0.055
glass-0.054
 minutes-0.053
ird-0.050
Token kids
Token ID3988
Feature activation+4.55e-01

Pos logit contributions
itating+0.027
melon+0.026
arre+0.025
bing+0.025
ucker+0.025
Neg logit contributions
year-0.028
ird-0.028
 hours-0.028
 minutes-0.027
glass-0.027
Token.
Token ID13
Feature activation+1.19e+00

Pos logit contributions
itating+0.008
arre+0.008
 up+0.008
melon+0.008
ling+0.008
Neg logit contributions
year-0.016
glass-0.015
ird-0.015
 hours-0.014
aster-0.014
Token
Token ID220
Feature activation-1.03e-01

Pos logit contributions
itating+0.033
ling+0.033
arre+0.032
flower+0.031
mill+0.031
Neg logit contributions
year-0.028
glass-0.026
 hours-0.026
ird-0.025
 Grand-0.024
Token.
Token ID13
Feature activation-1.79e+00

Pos logit contributions
itating+0.008
arre+0.008
 up+0.008
melon+0.008
ling+0.008
Neg logit contributions
year-0.018
glass-0.017
ird-0.017
 hours-0.016
aster-0.016
Token You
Token ID921
Feature activation-1.47e+00

Pos logit contributions
year+0.066
glass+0.065
 hours+0.064
 minutes+0.062
aster+0.062
Neg logit contributions
 Sharing-0.027
 up-0.024
itating-0.024
arre-0.024
 Bugs-0.024
Token must
Token ID1276
Feature activation+1.02e+00

Pos logit contributions
year+0.048
glass+0.046
ird+0.044
 hours+0.043
aster+0.043
Neg logit contributions
arre-0.031
itating-0.030
 up-0.030
melon-0.028
 fighting-0.028
Token take
Token ID1011
Feature activation-7.89e-01

Pos logit contributions
itating+0.026
arre+0.025
ling+0.024
melon+0.024
bing+0.024
Neg logit contributions
ird-0.029
year-0.027
glass-0.027
aster-0.027
ris-0.026
Token good
Token ID922
Feature activation-4.63e-01

Pos logit contributions
year+0.028
glass+0.027
ird+0.026
ibr+0.024
aster+0.024
Neg logit contributions
 up-0.015
arre-0.014
itating-0.013
 fighting-0.013
ling-0.013
Token care
Token ID1337
Feature activation-1.95e+00

Pos logit contributions
year+0.011
glass+0.011
ird+0.011
 hours+0.010
aster+0.009
Neg logit contributions
arre-0.013
 up-0.013
itating-0.013
ling-0.013
melon-0.013
Token of
Token ID286
Feature activation+1.89e-01

Pos logit contributions
year+0.051
glass+0.051
ird+0.050
aster+0.048
 hours+0.048
Neg logit contributions
 up-0.051
itating-0.050
arre-0.049
ling-0.048
melon-0.047
Token it
Token ID340
Feature activation-3.24e-01

Pos logit contributions
itating+0.003
arre+0.003
 up+0.003
melon+0.003
ling+0.003
Neg logit contributions
year-0.007
glass-0.007
ird-0.007
 hours-0.006
aster-0.006
Token."
Token ID526
Feature activation-1.20e-02

Pos logit contributions
year+0.012
glass+0.011
ird+0.011
aster+0.011
 hours+0.011
Neg logit contributions
 up-0.006
itating-0.006
arre-0.005
ling-0.005
melon-0.005
Token
Token ID198
Feature activation+4.67e-02

Pos logit contributions
year+0.000
glass+0.000
ird+0.000
 hours+0.000
aster+0.000
Neg logit contributions
arre-0.000
itating-0.000
 up-0.000
melon-0.000
ling-0.000
Token
Token ID198
Feature activation+9.01e-01

Pos logit contributions
itating+0.001
arre+0.001
melon+0.001
 up+0.001
ling+0.001
Neg logit contributions
year-0.002
ird-0.001
glass-0.001
 hours-0.001
aster-0.001
Token is
Token ID318
Feature activation+1.11e+00

Pos logit contributions
itating+0.007
arre+0.007
 up+0.006
melon+0.006
ling+0.006
Neg logit contributions
year-0.018
glass-0.017
ird-0.017
 hours-0.016
aster-0.016
Token a
Token ID257
Feature activation-4.46e-01

Pos logit contributions
itating+0.025
arre+0.025
 up+0.023
melon+0.023
ling+0.023
Neg logit contributions
year-0.034
ird-0.033
glass-0.033
 hours-0.031
 minutes-0.030
Token special
Token ID2041
Feature activation-5.56e-01

Pos logit contributions
year+0.016
ird+0.015
glass+0.014
ibr+0.014
aster+0.013
Neg logit contributions
 up-0.009
melon-0.008
 fighting-0.008
itating-0.008
arre-0.008
Token magazine
Token ID7093
Feature activation-1.64e+00

Pos logit contributions
year+0.018
glass+0.017
ird+0.017
ibr+0.016
 hours+0.016
Neg logit contributions
 up-0.012
melon-0.011
arre-0.011
 fighting-0.011
itating-0.011
Token.
Token ID13
Feature activation-4.40e-01

Pos logit contributions
year+0.068
glass+0.064
aster+0.063
 hours+0.062
ird+0.062
Neg logit contributions
 up-0.022
melon-0.019
 fighting-0.017
 Sharing-0.016
itating-0.016
Token She
Token ID1375
Feature activation-6.24e-01

Pos logit contributions
year+0.017
glass+0.016
 hours+0.016
ird+0.016
aster+0.015
Neg logit contributions
itating-0.006
 up-0.006
arre-0.006
 Sharing-0.006
 fighting-0.006
Token wants
Token ID3382
Feature activation-6.35e-01

Pos logit contributions
year+0.023
glass+0.022
ird+0.021
 hours+0.020
aster+0.020
Neg logit contributions
 up-0.011
arre-0.011
itating-0.010
 fighting-0.010
melon-0.010
Token to
Token ID284
Feature activation-2.13e-01

Pos logit contributions
year+0.023
ird+0.022
glass+0.022
ibr+0.020
 hours+0.020
Neg logit contributions
 up-0.012
 fighting-0.011
melon-0.010
arre-0.010
itating-0.010
Token see
Token ID766
Feature activation-8.69e-01

Pos logit contributions
year+0.008
 minutes+0.007
glass+0.007
 hours+0.007
ibr+0.006
Neg logit contributions
 up-0.004
itating-0.004
 fighting-0.004
melon-0.004
 Freddy-0.004
Token what
Token ID644
Feature activation-6.95e-02

Pos logit contributions
year+0.035
glass+0.033
ird+0.032
 hours+0.031
aster+0.031
Neg logit contributions
 up-0.014
 fighting-0.011
itating-0.011
ling-0.010
arre-0.010
Token is
Token ID318
Feature activation-2.22e-01

Pos logit contributions
year+0.003
glass+0.003
ird+0.003
 hours+0.003
aster+0.003
Neg logit contributions
 up-0.001
itating-0.001
 fighting-0.001
arre-0.001
 Sharing-0.001
Token the
Token ID262
Feature activation-4.19e-01

Pos logit contributions
year+0.008
glass+0.007
ird+0.007
ibr+0.007
ibrarian+0.007
Neg logit contributions
 up-0.003
 fighting-0.003
melon-0.002
itating-0.002
 sharing-0.002
Token freezer
Token ID30967
Feature activation-1.62e+00

Pos logit contributions
year+0.013
glass+0.012
ird+0.012
 hours+0.011
aster+0.011
Neg logit contributions
 up-0.010
arre-0.010
melon-0.010
itating-0.009
ling-0.009
Token to
Token ID284
Feature activation-2.58e+00

Pos logit contributions
year+0.064
 hours+0.061
ird+0.061
glass+0.061
aster+0.060
Neg logit contributions
 up-0.026
melon-0.022
itating-0.021
 fighting-0.020
ling-0.020
Token make
Token ID787
Feature activation-3.99e-01

Pos logit contributions
year+0.099
glass+0.097
ibr+0.096
ibrarian+0.090
 chapters+0.090
Neg logit contributions
 up-0.045
melon-0.036
 fighting-0.030
 scatter-0.029
 Freddy-0.029
Token it
Token ID340
Feature activation+1.21e+00

Pos logit contributions
year+0.014
glass+0.014
ird+0.014
ris+0.013
 hours+0.013
Neg logit contributions
 up-0.008
arre-0.007
ling-0.007
 fighting-0.006
bing-0.006
Token cold
Token ID4692
Feature activation+5.44e-02

Pos logit contributions
itating+0.027
arre+0.027
melon+0.025
ling+0.024
bing+0.024
Neg logit contributions
year-0.037
ird-0.035
glass-0.035
aster-0.033
 hours-0.033
Token.
Token ID13
Feature activation+9.26e-01

Pos logit contributions
itating+0.001
arre+0.001
melon+0.001
 up+0.001
ling+0.001
Neg logit contributions
year-0.002
glass-0.002
ird-0.002
 hours-0.002
aster-0.002
Token Lily
Token ID20037
Feature activation+2.22e-01

Pos logit contributions
itating+0.022
ling+0.022
arre+0.022
melon+0.021
bing+0.021
Neg logit contributions
ird-0.026
year-0.026
glass-0.024
 hours-0.024
 minutes-0.023
Token asked
Token ID1965
Feature activation+2.13e-01

Pos logit contributions
arre+0.004
 up+0.004
itating+0.003
ling+0.003
melon+0.003
Neg logit contributions
year-0.008
glass-0.008
ird-0.008
 hours-0.007
aster-0.007
Token,
Token ID11
Feature activation+1.22e-01

Pos logit contributions
 up+0.002
 fighting+0.001
 sharing+0.001
 Freddy+0.001
melon+0.001
Neg logit contributions
year-0.010
glass-0.010
ris-0.010
 kil-0.010
ird-0.009
Token "
Token ID366
Feature activation-6.84e-02

Pos logit contributions
itating+0.002
arre+0.002
 up+0.002
melon+0.002
ling+0.002
Neg logit contributions
year-0.005
glass-0.004
ird-0.004
 hours-0.004
aster-0.004
Token"
Token ID1
Feature activation-1.26e-01

Pos logit contributions
itating+0.022
arre+0.022
ling+0.021
 up+0.021
bing+0.020
Neg logit contributions
year-0.036
glass-0.035
ird-0.034
 hours-0.033
ibr-0.032
TokenMom
Token ID29252
Feature activation+3.87e-01

Pos logit contributions
year+0.004
glass+0.004
ird+0.004
ibr+0.004
aster+0.004
Neg logit contributions
arre-0.002
itating-0.002
melon-0.002
flower-0.002
 up-0.002
Token,
Token ID11
Feature activation-1.01e+00

Pos logit contributions
arre+0.006
itating+0.006
 up+0.006
melon+0.006
ling+0.006
Neg logit contributions
year-0.014
glass-0.014
ird-0.013
 hours-0.013
aster-0.013
Token are
Token ID389
Feature activation+8.46e-01

Pos logit contributions
year+0.039
glass+0.036
ird+0.034
ris+0.034
 hours+0.034
Neg logit contributions
 up-0.017
 fighting-0.016
melon-0.016
 Freddy-0.014
 accidents-0.014
Token the
Token ID262
Feature activation+1.54e+00

Pos logit contributions
melon+0.010
 up+0.010
 fighting+0.009
 Freddy+0.009
 sharing+0.008
Neg logit contributions
year-0.036
glass-0.036
ird-0.034
ris-0.034
 hours-0.033
Token pizz
Token ID38748
Feature activation-2.89e+00

Pos logit contributions
melon+0.040
 up+0.039
arre+0.038
 fighting+0.038
itating+0.038
Neg logit contributions
year-0.044
ird-0.041
glass-0.040
 hours-0.038
aster-0.036
Tokenas
Token ID292
Feature activation-2.20e-01

Pos logit contributions
 hours+0.067
year+0.063
 minutes+0.061
aster+0.060
glass+0.059
Neg logit contributions
itating-0.086
arre-0.086
melon-0.084
ling-0.080
 up-0.079
Token ready
Token ID3492
Feature activation+1.58e-01

Pos logit contributions
year+0.007
ird+0.007
glass+0.007
aster+0.006
 hours+0.006
Neg logit contributions
itating-0.005
arre-0.005
 up-0.005
melon-0.004
ling-0.004
Token?"
Token ID1701
Feature activation+4.54e-01

Pos logit contributions
 up+0.003
arre+0.003
itating+0.003
melon+0.003
ling+0.003
Neg logit contributions
year-0.006
glass-0.005
ird-0.005
 hours-0.005
ibr-0.005
Token Lily
Token ID20037
Feature activation-9.72e-01

Pos logit contributions
itating+0.007
arre+0.007
 up+0.007
melon+0.007
ling+0.006
Neg logit contributions
year-0.017
glass-0.016
ird-0.016
 hours-0.016
aster-0.015
Token asked
Token ID1965
Feature activation-5.77e-01

Pos logit contributions
year+0.035
glass+0.032
ird+0.031
 hours+0.031
aster+0.031
Neg logit contributions
 up-0.018
arre-0.018
itating-0.018
 fighting-0.017
ling-0.017
Token It
Token ID632
Feature activation+1.63e+00

Pos logit contributions
itating+0.009
arre+0.009
ling+0.009
melon+0.009
 up+0.009
Neg logit contributions
year-0.019
ird-0.018
glass-0.018
 hours-0.018
aster-0.017
Token was
Token ID373
Feature activation-2.49e-01

Pos logit contributions
arre+0.030
itating+0.030
melon+0.027
ling+0.027
bing+0.026
Neg logit contributions
year-0.057
ird-0.055
glass-0.055
 hours-0.052
 minutes-0.050
Token a
Token ID257
Feature activation-1.17e+00

Pos logit contributions
ird+0.009
glass+0.009
year+0.008
ibr+0.008
ris+0.008
Neg logit contributions
 up-0.005
melon-0.004
 fighting-0.004
 Freddy-0.004
 sharing-0.004
Token book
Token ID1492
Feature activation-2.05e+00

Pos logit contributions
year+0.042
ird+0.039
glass+0.039
ibr+0.038
 hours+0.037
Neg logit contributions
 up-0.022
melon-0.022
 fighting-0.020
itating-0.019
arre-0.018
Token about
Token ID546
Feature activation-1.31e+00

Pos logit contributions
year+0.074
glass+0.070
ird+0.070
 hours+0.067
aster+0.067
Neg logit contributions
 up-0.036
itating-0.035
arre-0.033
melon-0.033
ling-0.032
Token history
Token ID2106
Feature activation-2.98e+00

Pos logit contributions
year+0.043
ird+0.043
glass+0.041
aster+0.040
ris+0.039
Neg logit contributions
 up-0.027
 fighting-0.025
 sharing-0.024
melon-0.023
arre-0.023
Token.
Token ID13
Feature activation+6.47e-01

Pos logit contributions
year+0.127
ird+0.120
aster+0.119
glass+0.117
Ah+0.115
Neg logit contributions
 up-0.040
 sharing-0.027
 fighting-0.025
ling-0.025
melon-0.024
Token Sophie
Token ID35331
Feature activation+4.36e-01

Pos logit contributions
itating+0.013
arre+0.013
ling+0.013
melon+0.012
 up+0.012
Neg logit contributions
year-0.021
ird-0.021
glass-0.020
 hours-0.019
 minutes-0.019
Token was
Token ID373
Feature activation-7.47e-01

Pos logit contributions
 up+0.008
arre+0.008
itating+0.008
melon+0.008
ling+0.008
Neg logit contributions
year-0.015
glass-0.014
ird-0.014
 hours-0.013
aster-0.013
Token so
Token ID523
Feature activation-8.60e-01

Pos logit contributions
year+0.025
glass+0.024
ird+0.024
 hours+0.023
aster+0.022
Neg logit contributions
 up-0.015
itating-0.015
arre-0.015
melon-0.014
 fighting-0.014
Token proud
Token ID6613
Feature activation+4.56e-01

Pos logit contributions
glass+0.026
year+0.025
ird+0.025
 hours+0.025
aster+0.024
Neg logit contributions
 up-0.020
itating-0.020
arre-0.019
 fighting-0.019
melon-0.018
Token They
Token ID1119
Feature activation-1.05e+00

Pos logit contributions
year+0.042
ird+0.040
glass+0.039
 hours+0.038
ocket+0.037
Neg logit contributions
 fighting-0.009
arre-0.009
 Sharing-0.009
 Freddy-0.009
 up-0.009
Token had
Token ID550
Feature activation-4.51e-01

Pos logit contributions
year+0.039
ae+0.038
ird+0.037
aster+0.037
glass+0.037
Neg logit contributions
 fighting-0.016
 qu-0.016
melon-0.014
 up-0.014
 en-0.013
Token h
Token ID289
Feature activation-2.13e+00

Pos logit contributions
year+0.016
ird+0.015
glass+0.014
ibr+0.014
aster+0.014
Neg logit contributions
 fighting-0.009
 up-0.008
 qu-0.008
melon-0.008
arre-0.007
Tokenoses
Token ID4629
Feature activation-4.09e-01

Pos logit contributions
year+0.054
ird+0.051
glass+0.051
 hours+0.049
aster+0.048
Neg logit contributions
arre-0.059
itating-0.058
 up-0.056
melon-0.055
 fighting-0.054
Token and
Token ID290
Feature activation-1.01e+00

Pos logit contributions
year+0.016
ird+0.015
glass+0.015
 hours+0.015
aster+0.014
Neg logit contributions
itating-0.006
 up-0.006
arre-0.006
melon-0.005
ling-0.005
Token l
Token ID300
Feature activation-3.20e+00

Pos logit contributions
year+0.036
ird+0.033
ae+0.032
ris+0.031
ibrarian+0.031
Neg logit contributions
 fighting-0.020
 qu-0.018
melon-0.018
 up-0.017
lings-0.017
Tokenadders
Token ID45940
Feature activation-6.56e-01

Pos logit contributions
aster+0.076
 Bible+0.071
 hours+0.065
ae+0.065
ird+0.064
Neg logit contributions
ters-0.091
lings-0.088
arre-0.087
melon-0.087
ling-0.087
Token.
Token ID13
Feature activation-8.01e-01

Pos logit contributions
year+0.024
ird+0.024
glass+0.023
 hours+0.022
aster+0.022
Neg logit contributions
itating-0.010
arre-0.010
 up-0.010
melon-0.010
ling-0.009
Token They
Token ID1119
Feature activation-8.03e-01

Pos logit contributions
year+0.036
ird+0.034
glass+0.034
 hours+0.032
lock+0.032
Neg logit contributions
 Sharing-0.007
 fighting-0.007
itating-0.007
arre-0.007
 Freddy-0.007
Token were
Token ID547
Feature activation-2.09e-01

Pos logit contributions
year+0.031
ae+0.029
glass+0.029
aster+0.029
ird+0.028
Neg logit contributions
 fighting-0.012
 qu-0.011
melon-0.011
 up-0.010
 en-0.009
Token firefighters
Token ID22611
Feature activation-9.17e-02

Pos logit contributions
ird+0.008
year+0.008
glass+0.007
lock+0.007
aster+0.007
Neg logit contributions
 up-0.004
 fighting-0.004
melon-0.003
ling-0.003
 en-0.003
Token something
Token ID1223
Feature activation-6.11e-01

Pos logit contributions
itating+0.024
arre+0.024
 up+0.023
melon+0.023
ling+0.023
Neg logit contributions
year-0.039
ird-0.038
glass-0.038
 hours-0.036
aster-0.035
Token magic
Token ID5536
Feature activation-9.75e-01

Pos logit contributions
year+0.018
glass+0.017
ird+0.017
 hours+0.017
 minutes+0.017
Neg logit contributions
itating-0.014
arre-0.014
bing-0.013
melon-0.013
 up-0.013
Token happened
Token ID3022
Feature activation+2.13e+00

Pos logit contributions
year+0.032
glass+0.031
ird+0.031
 hours+0.029
aster+0.029
Neg logit contributions
itating-0.019
arre-0.019
 up-0.018
melon-0.018
ling-0.018
Token!
Token ID0
Feature activation-8.81e-01

Pos logit contributions
itating+0.049
arre+0.049
melon+0.046
rust+0.046
ling+0.044
Neg logit contributions
year-0.063
glass-0.062
ird-0.059
 hours-0.058
 minutes-0.056
Token The
Token ID383
Feature activation+4.19e-01

Pos logit contributions
year+0.031
glass+0.030
 hours+0.029
ird+0.029
 minutes+0.029
Neg logit contributions
 Freddy-0.015
itating-0.015
arre-0.014
 up-0.014
melon-0.014
Token m
Token ID285
Feature activation-3.81e+00

Pos logit contributions
 up+0.009
melon+0.008
ling+0.008
 fighting+0.008
itating+0.007
Neg logit contributions
year-0.014
ird-0.014
glass-0.013
ris-0.012
aster-0.012
Tokenugs
Token ID10339
Feature activation-1.74e+00

Pos logit contributions
 hours+0.106
year+0.103
 minutes+0.101
glass+0.100
 Africa+0.097
Neg logit contributions
ling-0.092
bing-0.089
flower-0.088
itating-0.087
 up-0.085
Token split
Token ID6626
Feature activation+8.54e-01

Pos logit contributions
year+0.056
glass+0.054
ird+0.053
 hours+0.051
aster+0.050
Neg logit contributions
itating-0.036
arre-0.036
 up-0.035
melon-0.034
ling-0.034
Token.
Token ID13
Feature activation-1.10e+00

Pos logit contributions
itating+0.017
arre+0.017
 up+0.017
melon+0.016
ling+0.016
Neg logit contributions
year-0.028
ird-0.027
glass-0.027
 hours-0.025
aster-0.025
Token They
Token ID1119
Feature activation-5.86e-01

Pos logit contributions
year+0.040
glass+0.039
ird+0.037
 hours+0.037
 minutes+0.036
Neg logit contributions
 Freddy-0.017
arre-0.017
itating-0.017
 up-0.016
melon-0.016
Token opened
Token ID4721
Feature activation+1.10e+00

Pos logit contributions
year+0.021
glass+0.020
ird+0.020
aster+0.019
 minutes+0.019
Neg logit contributions
 up-0.010
 fighting-0.010
melon-0.010
itating-0.010
arre-0.010
Token light
Token ID1657
Feature activation-7.32e-01

Pos logit contributions
 up+0.013
ling+0.012
melon+0.012
arre+0.012
lings+0.012
Neg logit contributions
year-0.033
ird-0.033
glass-0.031
aster-0.030
ris-0.030
Token of
Token ID286
Feature activation-1.19e+00

Pos logit contributions
year+0.026
ird+0.025
glass+0.025
 hours+0.023
aster+0.023
Neg logit contributions
 up-0.013
itating-0.013
arre-0.013
melon-0.013
ling-0.012
Token the
Token ID262
Feature activation-3.92e-01

Pos logit contributions
ird+0.047
year+0.046
glass+0.044
ris+0.042
lock+0.042
Neg logit contributions
 up-0.019
 fighting-0.018
melon-0.017
 Freddy-0.016
 accidents-0.015
Token sunrise
Token ID40048
Feature activation-4.54e-01

Pos logit contributions
ird+0.015
year+0.015
glass+0.014
aster+0.014
ibr+0.013
Neg logit contributions
 up-0.007
ling-0.006
 fighting-0.006
melon-0.006
itating-0.006
Token streaming
Token ID11305
Feature activation-1.73e+00

Pos logit contributions
year+0.018
ird+0.017
glass+0.017
 hours+0.016
aster+0.016
Neg logit contributions
 up-0.007
melon-0.006
ling-0.006
itating-0.006
flower-0.006
Token into
Token ID656
Feature activation-2.90e+00

Pos logit contributions
year+0.070
ird+0.069
glass+0.068
 hours+0.065
aster+0.065
Neg logit contributions
 up-0.022
itating-0.022
arre-0.021
ling-0.020
melon-0.019
Token his
Token ID465
Feature activation-1.43e+00

Pos logit contributions
ird+0.124
year+0.119
glass+0.116
ris+0.115
lock+0.114
Neg logit contributions
 up-0.038
 fighting-0.034
ling-0.032
 accidents-0.030
 en-0.027
Token room
Token ID2119
Feature activation-3.90e-01

Pos logit contributions
year+0.051
ird+0.047
glass+0.045
ris+0.043
 hours+0.043
Neg logit contributions
 up-0.029
ling-0.028
arre-0.027
lings-0.027
 fighting-0.026
Token.
Token ID13
Feature activation+1.65e-01

Pos logit contributions
year+0.015
ird+0.014
glass+0.014
 hours+0.013
aster+0.013
Neg logit contributions
 up-0.006
arre-0.006
ling-0.006
itating-0.006
melon-0.006
Token
Token ID220
Feature activation-7.56e-01

Pos logit contributions
arre+0.003
itating+0.003
 up+0.003
melon+0.003
 fighting+0.002
Neg logit contributions
year-0.006
glass-0.006
ird-0.006
 hours-0.005
aster-0.005
Token
Token ID198
Feature activation-2.02e-01

Pos logit contributions
year+0.025
ird+0.024
glass+0.024
 hours+0.023
aster+0.022
Neg logit contributions
arre-0.015
itating-0.015
melon-0.014
 up-0.014
ling-0.014