TOP ACTIVATIONS
MAX = 0.110

on her neighbor 's windows ill . She climbed down and
and soon his trouble - making days were behind him .
a little girl named G ert ie , stopped by to
for him on her windows ill . He hopped and hopped
otted plant on her windows ill . From here , she
snow man and their snow balls . Mom listens and smiles
their potato dolls and their other dolls . They had fun
mom m ies and d add ies . Lily and Ben
put it on her windows ill . Every day , she
bird sitting on her windows ill ! L ily
a creative chicken named Ch icky . Ch icky liked to
the toy plane and their other toys . They are happy
placed it on her windows ill . Every time she saw
mom m ies and d add ies when they say things
their sc ooters and their other things . They still have
B ub ba and his other friends were hiding for a
leave their sons and their hut alone . They say they
and Pupp y and w add led home . Mama Duck
ight ed on his windows ill and said : " Do
his mom and his sp ade . At the

MIN ACTIVATIONS
MIN = -0.104

would always say ' thank - you ' to his toast
Sarah was so awe - stru ck . It was
John wondered how old this place was . It
. She wrote about how grand baby is always smiling ,
. Tim my wondered how old it was .
ez ing ! The sprinkle - covered pastry flew out of
touched it and wondered how old it was .
old , nobody knew how old it was . Everyone who
He exclaimed , " Pick - a - bo o !
been ?" Pick - a - bo o smiled
forest didn 't like how boss y Billy was . They
so proud of his brand - new suit .
can also learn about how wheels work . She asks them
witch was about to give - up when a little rabbit
Charlie felt so happy - he knew he could play
. He put on extra - special show for his family
where she could still show - off her fashion .
, Mia spotted something special - a glowing egg ! Curious
she needed a little pick - me - up , she
festival . He played tug - of - war and ran

INTERVAL 0.056 - 0.110
CONTAINS 0.084%

, but their cl ums iness often made them laugh too
the truck from Ben and Mia . " Enough
L ily and Tom 's treasure . Do you know how
walking through a forest , Fred saw a big tree .
boy named Tim was very eager to play . He ran

INTERVAL 0.003 - 0.056
CONTAINS 42.544%

safe feeling . <|endoftext|> Tim and Lucy were best friends in
to paint pictures of animals and people . One day ,
smiled . She wasn 't scared anymore and knew that the
had to promise to always be best friends . They made
!" He pulled the balloon hard . The string broke .

INTERVAL -0.051 - 0.003
CONTAINS 57.246%

B enny looked around and saw his friend , Sammy the
Lily said . She hugged her block and did not share
it was a watch he had stolen .
cl apped their hands . Tom and Mia started their show
tightly . From that day on , Lily

INTERVAL -0.104 - -0.051
CONTAINS 0.127%

outside when she noticed how cold the air was . She
the steak . The melt - in - the - mouth
They played hide - and - seek in the park .
stories . Can you read me one ?" The
we can find out how thick it is ."
Token on
Token ID319
Feature activation+2.97e-04

Pos logit contributions
 wounds+0.054
 reins+0.052
 brakes+0.050
 buttons+0.050
 stret+0.050
Neg logit contributions
 named-0.025
ores-0.023
chard-0.022
arium-0.020
apped-0.020
Token her
Token ID607
Feature activation+2.29e-02

Pos logit contributions
 wounds+0.001
 reins+0.001
 buttons+0.001
 stret+0.001
 rules+0.001
Neg logit contributions
ores-0.000
 named-0.000
arium-0.000
ver-0.000
year-0.000
Token neighbor
Token ID4780
Feature activation+1.99e-02

Pos logit contributions
 buttons+0.034
 pounds+0.033
 reins+0.033
Leave+0.033
 brakes+0.033
Neg logit contributions
 named-0.034
chard-0.031
,-0.031
 every-0.030
 night-0.029
Token's
Token ID338
Feature activation+1.74e-03

Pos logit contributions
 wounds+0.031
 reins+0.030
 rules+0.030
 buttons+0.029
 brakes+0.028
Neg logit contributions
 named-0.035
chard-0.028
,-0.028
arium-0.027
 lived-0.027
Token windows
Token ID9168
Feature activation-9.25e-04

Pos logit contributions
 pounds+0.003
 sax+0.003
 brakes+0.003
 buttons+0.003
Leave+0.003
Neg logit contributions
 named-0.002
chard-0.002
 Finally-0.002
,-0.002
ores-0.002
Tokenill
Token ID359
Feature activation+1.01e-01

Pos logit contributions
ores+0.001
 named+0.001
apped+0.001
ver+0.001
year+0.001
Neg logit contributions
 wounds-0.002
 reins-0.002
 stret-0.002
 brakes-0.002
 rules-0.002
Token.
Token ID13
Feature activation+9.21e-03

Pos logit contributions
 wounds+0.220
 brakes+0.209
 rules+0.205
 reins+0.202
 buttons+0.199
Neg logit contributions
 named-0.102
ores-0.089
chard-0.086
arium-0.083
working-0.082
Token She
Token ID1375
Feature activation-1.40e-03

Pos logit contributions
 wounds+0.018
 buttons+0.017
 stret+0.017
 reins+0.017
 rules+0.017
Neg logit contributions
ores-0.010
arium-0.009
ver-0.009
 named-0.009
year-0.009
Token climbed
Token ID19952
Feature activation-3.40e-03

Pos logit contributions
ores+0.002
arium+0.002
ver+0.002
 named+0.002
year+0.002
Neg logit contributions
 wounds-0.002
 buttons-0.002
 reins-0.002
 mess-0.002
 stret-0.002
Token down
Token ID866
Feature activation-1.40e-03

Pos logit contributions
arium+0.005
ver+0.005
ores+0.005
year+0.005
 named+0.005
Neg logit contributions
 wounds-0.005
 buttons-0.005
 reins-0.005
 stret-0.005
 brakes-0.005
Token and
Token ID290
Feature activation-8.41e-03

Pos logit contributions
arium+0.002
ores+0.002
chard+0.002
apped+0.002
ver+0.002
Neg logit contributions
 buttons-0.002
 rules-0.002
 wounds-0.002
 blood-0.002
 mess-0.002
Token and
Token ID290
Feature activation+9.36e-03

Pos logit contributions
 wounds+0.058
 buttons+0.056
 reins+0.055
 rules+0.055
 stret+0.055
Neg logit contributions
ores-0.035
 named-0.033
arium-0.033
year-0.033
ver-0.032
Token soon
Token ID2582
Feature activation+2.49e-02

Pos logit contributions
 wounds+0.015
 rules+0.015
 buttons+0.014
 mess+0.014
 brakes+0.014
Neg logit contributions
ores-0.014
year-0.013
arium-0.013
ver-0.013
apped-0.013
Token his
Token ID465
Feature activation+3.33e-02

Pos logit contributions
 wounds+0.042
 buttons+0.042
 mess+0.041
 rules+0.041
 brakes+0.040
Neg logit contributions
ores-0.032
year-0.031
chard-0.030
arium-0.030
ver-0.030
Token trouble
Token ID5876
Feature activation+6.41e-02

Pos logit contributions
 wounds+0.043
 stret+0.042
 reins+0.041
 buttons+0.041
 pounds+0.039
Neg logit contributions
 named-0.056
ores-0.055
arium-0.053
ver-0.052
 known-0.052
Token-
Token ID12
Feature activation-1.73e-02

Pos logit contributions
 wounds+0.117
 reins+0.113
 buttons+0.109
 stret+0.109
 brakes+0.108
Neg logit contributions
 named-0.080
ores-0.079
ver-0.078
arium-0.076
apped-0.074
Tokenmaking
Token ID8601
Feature activation+9.94e-02

Pos logit contributions
ores+0.028
 named+0.027
arium+0.027
ver+0.027
year+0.026
Neg logit contributions
 wounds-0.025
 buttons-0.024
 reins-0.024
 stret-0.023
 rules-0.023
Token days
Token ID1528
Feature activation+2.26e-02

Pos logit contributions
 wounds+0.141
 reins+0.138
 buttons+0.134
 brakes+0.130
 pounds+0.129
Neg logit contributions
 named-0.158
ores-0.156
arium-0.151
ver-0.149
apped-0.147
Token were
Token ID547
Feature activation+3.69e-02

Pos logit contributions
 wounds+0.040
 buttons+0.039
 stret+0.038
 reins+0.038
 rules+0.038
Neg logit contributions
ores-0.028
arium-0.027
 named-0.026
ver-0.026
year-0.026
Token behind
Token ID2157
Feature activation+2.36e-02

Pos logit contributions
 wounds+0.062
 reins+0.059
 buttons+0.059
 stret+0.058
 rules+0.057
Neg logit contributions
ores-0.051
 named-0.050
ver-0.049
arium-0.048
year-0.047
Token him
Token ID683
Feature activation+1.90e-02

Pos logit contributions
 wounds+0.043
 buttons+0.042
 reins+0.041
 stret+0.041
 rules+0.041
Neg logit contributions
ores-0.028
 named-0.027
arium-0.027
ver-0.027
year-0.026
Token.
Token ID13
Feature activation-1.46e-03

Pos logit contributions
 wounds+0.032
 reins+0.030
 buttons+0.030
 stret+0.030
 rules+0.030
Neg logit contributions
ores-0.026
 named-0.025
arium-0.025
ver-0.025
year-0.024
Token a
Token ID257
Feature activation+2.32e-02

Pos logit contributions
 wounds+0.057
 buttons+0.055
 stret+0.054
 reins+0.054
 rules+0.053
Neg logit contributions
ores-0.042
arium-0.040
ver-0.039
year-0.039
 named-0.038
Token little
Token ID1310
Feature activation+2.32e-02

Pos logit contributions
 wounds+0.029
 reins+0.026
 buttons+0.026
 rules+0.025
 stret+0.025
Neg logit contributions
 named-0.042
ores-0.041
arium-0.041
ver-0.040
chard-0.040
Token girl
Token ID2576
Feature activation+2.29e-02

Pos logit contributions
 wounds+0.030
 stret+0.027
 reins+0.027
 buttons+0.027
 rules+0.027
Neg logit contributions
 named-0.041
ores-0.040
arium-0.040
year-0.039
chard-0.038
Token named
Token ID3706
Feature activation-2.84e-02

Pos logit contributions
 wounds+0.055
 stret+0.053
 buttons+0.053
 reins+0.053
 mess+0.052
Neg logit contributions
ores-0.015
ver-0.013
arium-0.013
year-0.012
apped-0.012
Token G
Token ID402
Feature activation+3.99e-02

Pos logit contributions
ores+0.036
arium+0.034
 named+0.034
ver+0.034
year+0.033
Neg logit contributions
 wounds-0.051
 buttons-0.048
 reins-0.048
 stret-0.048
 rules-0.048
Tokenert
Token ID861
Feature activation+9.77e-02

Pos logit contributions
 wounds+0.061
 buttons+0.058
 reins+0.057
 brakes+0.057
 stret+0.056
Neg logit contributions
ores-0.063
ver-0.061
arium-0.060
 named-0.059
year-0.059
Tokenie
Token ID494
Feature activation+5.56e-02

Pos logit contributions
 wounds+0.163
 reins+0.153
 buttons+0.152
 rules+0.151
 stret+0.150
Neg logit contributions
 named-0.137
ores-0.137
ver-0.133
arium-0.132
year-0.130
Token,
Token ID11
Feature activation+3.73e-02

Pos logit contributions
 wounds+0.120
 stret+0.118
 buttons+0.118
 mess+0.116
 reins+0.115
Neg logit contributions
ores-0.047
arium-0.043
year-0.042
ver-0.041
apped-0.039
Token stopped
Token ID5025
Feature activation+7.15e-03

Pos logit contributions
 wounds+0.066
 buttons+0.064
 reins+0.063
 stret+0.063
 rules+0.062
Neg logit contributions
ores-0.047
arium-0.045
ver-0.044
 named-0.044
year-0.043
Token by
Token ID416
Feature activation-1.22e-02

Pos logit contributions
 wounds+0.012
 reins+0.012
 buttons+0.012
 stret+0.012
 brakes+0.011
Neg logit contributions
ores-0.010
 named-0.010
arium-0.009
year-0.009
 known-0.009
Token to
Token ID284
Feature activation+2.91e-02

Pos logit contributions
ores+0.015
arium+0.015
ver+0.015
 named+0.015
year+0.014
Neg logit contributions
 wounds-0.022
 buttons-0.021
 stret-0.021
 reins-0.021
 rules-0.021
Token for
Token ID329
Feature activation-3.30e-03

Pos logit contributions
ores+0.003
year+0.003
ver+0.003
arium+0.003
apped+0.003
Neg logit contributions
 wounds-0.003
 buttons-0.003
 reins-0.003
 stret-0.003
 rules-0.003
Token him
Token ID683
Feature activation+1.30e-02

Pos logit contributions
ores+0.004
 named+0.004
arium+0.004
ver+0.004
year+0.004
Neg logit contributions
 wounds-0.006
 buttons-0.005
 reins-0.005
 stret-0.005
 rules-0.005
Token on
Token ID319
Feature activation+8.92e-03

Pos logit contributions
 wounds+0.020
 buttons+0.019
 reins+0.019
 rules+0.019
 stret+0.019
Neg logit contributions
ores-0.019
arium-0.019
ver-0.019
year-0.018
 named-0.018
Token her
Token ID607
Feature activation+3.15e-02

Pos logit contributions
 wounds+0.016
 buttons+0.016
 rules+0.016
 mess+0.016
 reins+0.016
Neg logit contributions
ores-0.011
ver-0.010
year-0.010
apped-0.010
arium-0.010
Token windows
Token ID9168
Feature activation-5.10e-03

Pos logit contributions
 buttons+0.045
 wounds+0.044
 reins+0.044
 brakes+0.044
 blood+0.043
Neg logit contributions
 named-0.049
arium-0.046
chard-0.045
 lived-0.043
ores-0.043
Tokenill
Token ID359
Feature activation+9.66e-02

Pos logit contributions
ores+0.007
 named+0.007
arium+0.007
ver+0.007
year+0.007
Neg logit contributions
 wounds-0.008
 buttons-0.008
 reins-0.008
 stret-0.008
 rules-0.008
Token.
Token ID13
Feature activation+6.38e-03

Pos logit contributions
 wounds+0.167
 reins+0.157
 buttons+0.157
 brakes+0.156
 rules+0.154
Neg logit contributions
 named-0.130
ores-0.126
arium-0.125
ver-0.122
chard-0.116
Token He
Token ID679
Feature activation-5.25e-03

Pos logit contributions
 wounds+0.009
 rules+0.009
 reins+0.009
 stret+0.009
 buttons+0.009
Neg logit contributions
ores-0.009
ver-0.009
year-0.009
arium-0.009
 surely-0.009
Token hopped
Token ID45230
Feature activation-1.88e-03

Pos logit contributions
ores+0.008
arium+0.007
ver+0.007
chard+0.007
apped+0.007
Neg logit contributions
 wounds-0.009
 buttons-0.008
 mess-0.008
 reins-0.008
 brakes-0.008
Token and
Token ID290
Feature activation+3.93e-04

Pos logit contributions
arium+0.003
ver+0.003
 known+0.003
ores+0.003
riage+0.003
Neg logit contributions
 reins-0.002
 buttons-0.002
 wounds-0.002
 stret-0.002
 rules-0.002
Token hopped
Token ID45230
Feature activation+8.07e-03

Pos logit contributions
 wounds+0.001
 buttons+0.000
 reins+0.000
 scarf+0.000
 tissue+0.000
Neg logit contributions
ores-0.001
ver-0.001
year-0.001
apped-0.001
 Captain-0.001
Tokenotted
Token ID8426
Feature activation-1.13e-02

Pos logit contributions
 wounds+0.012
 stret+0.010
 buttons+0.010
 mess+0.010
 reins+0.010
Neg logit contributions
ores-0.031
ver-0.031
 named-0.031
arium-0.031
year-0.031
Token plant
Token ID4618
Feature activation+2.58e-02

Pos logit contributions
ores+0.022
 named+0.022
arium+0.022
ver+0.021
year+0.021
Neg logit contributions
 wounds-0.012
 buttons-0.011
 reins-0.011
 stret-0.011
 rules-0.011
Token on
Token ID319
Feature activation-5.72e-04

Pos logit contributions
 wounds+0.040
 buttons+0.039
 reins+0.038
 stret+0.038
 rules+0.038
Neg logit contributions
ores-0.038
ver-0.037
arium-0.036
year-0.036
 named-0.036
Token her
Token ID607
Feature activation+2.49e-02

Pos logit contributions
ver+0.001
ores+0.001
arium+0.001
apped+0.001
year+0.001
Neg logit contributions
 buttons-0.001
 wounds-0.001
 rules-0.001
 mess-0.001
 blood-0.001
Token windows
Token ID9168
Feature activation-2.55e-03

Pos logit contributions
 brakes+0.029
 reins+0.028
 wounds+0.028
 pounds+0.028
 buttons+0.028
Neg logit contributions
 named-0.046
arium-0.042
chard-0.041
ores-0.041
 lived-0.041
Tokenill
Token ID359
Feature activation+9.52e-02

Pos logit contributions
ores+0.004
 named+0.003
arium+0.003
ver+0.003
year+0.003
Neg logit contributions
 wounds-0.004
 reins-0.004
 buttons-0.004
 stret-0.004
 rules-0.004
Token.
Token ID13
Feature activation+3.74e-03

Pos logit contributions
 wounds+0.164
 brakes+0.155
 rules+0.154
 reins+0.153
 buttons+0.151
Neg logit contributions
 named-0.128
ores-0.124
arium-0.121
ver-0.117
apped-0.113
Token From
Token ID3574
Feature activation-7.69e-03

Pos logit contributions
 buttons+0.006
 reins+0.005
 rules+0.005
 pedals+0.005
 brakes+0.005
Neg logit contributions
arium-0.005
chard-0.005
ver-0.005
 named-0.005
ores-0.005
Token here
Token ID994
Feature activation+1.65e-02

Pos logit contributions
ver+0.011
ores+0.011
year+0.011
apped+0.011
arium+0.010
Neg logit contributions
 wounds-0.012
 buttons-0.012
 blood-0.011
 mess-0.011
 zipper-0.011
Token,
Token ID11
Feature activation-1.50e-02

Pos logit contributions
 buttons+0.026
 rules+0.024
 mess+0.024
 wounds+0.024
 zipper+0.024
Neg logit contributions
ores-0.025
arium-0.022
chard-0.022
year-0.022
ver-0.021
Token she
Token ID673
Feature activation+9.99e-03

Pos logit contributions
ores+0.023
arium+0.022
year+0.021
 named+0.021
ver+0.020
Neg logit contributions
 buttons-0.023
 mess-0.022
 rules-0.022
 rip-0.021
 wounds-0.021
Token snow
Token ID6729
Feature activation+2.22e-02

Pos logit contributions
 pounds+0.015
 reins+0.015
 brakes+0.015
 stret+0.014
 wounds+0.014
Neg logit contributions
 named-0.017
ores-0.017
arium-0.016
 lived-0.016
 night-0.016
Tokenman
Token ID805
Feature activation+7.76e-03

Pos logit contributions
 reins+0.034
 wounds+0.034
 brakes+0.033
 rules+0.033
 buttons+0.033
Neg logit contributions
ores-0.033
 named-0.032
arium-0.031
ver-0.030
apped-0.030
Token and
Token ID290
Feature activation+2.55e-02

Pos logit contributions
 wounds+0.012
 stret+0.011
 buttons+0.011
 mess+0.011
 rules+0.011
Neg logit contributions
ores-0.011
year-0.011
arium-0.011
ver-0.011
apped-0.011
Token their
Token ID511
Feature activation+3.74e-02

Pos logit contributions
 wounds+0.046
 buttons+0.044
 reins+0.044
 stret+0.044
 rules+0.044
Neg logit contributions
ores-0.031
 named-0.030
arium-0.030
ver-0.030
year-0.029
Token snow
Token ID6729
Feature activation+5.89e-02

Pos logit contributions
 pounds+0.041
 stret+0.040
 reins+0.040
 wounds+0.040
 brakes+0.039
Neg logit contributions
 named-0.066
ores-0.065
arium-0.062
 Captain-0.061
 lived-0.061
Tokenballs
Token ID21591
Feature activation+9.42e-02

Pos logit contributions
 reins+0.094
 wounds+0.093
 rules+0.092
 brakes+0.091
 buttons+0.091
Neg logit contributions
ores-0.083
 named-0.080
apped-0.076
arium-0.075
 lived-0.074
Token.
Token ID13
Feature activation+1.54e-02

Pos logit contributions
 wounds+0.150
 buttons+0.142
 reins+0.142
 stret+0.141
 rules+0.140
Neg logit contributions
ores-0.136
 named-0.132
arium-0.132
ver-0.130
year-0.128
Token Mom
Token ID11254
Feature activation-8.16e-03

Pos logit contributions
 wounds+0.027
 buttons+0.026
 stret+0.026
 reins+0.026
 rules+0.025
Neg logit contributions
arium-0.019
ver-0.019
year-0.019
ores-0.018
 named-0.018
Token listens
Token ID35019
Feature activation+9.17e-03

Pos logit contributions
ores+0.011
 named+0.011
arium+0.011
ver+0.010
year+0.010
Neg logit contributions
 wounds-0.014
 reins-0.013
 stret-0.013
 rules-0.013
 buttons-0.013
Token and
Token ID290
Feature activation-3.38e-03

Pos logit contributions
 wounds+0.014
 buttons+0.013
 rules+0.013
 reins+0.013
 stret+0.013
Neg logit contributions
ores-0.014
arium-0.014
 named-0.014
ver-0.014
year-0.013
Token smiles
Token ID21845
Feature activation+2.92e-03

Pos logit contributions
ores+0.005
ver+0.005
year+0.005
arium+0.005
 named+0.005
Neg logit contributions
 wounds-0.005
 buttons-0.005
 mess-0.005
 rules-0.005
 reins-0.005
Token their
Token ID511
Feature activation+1.49e-02

Pos logit contributions
ores+0.019
arium+0.018
ver+0.018
year+0.018
 named+0.018
Neg logit contributions
 wounds-0.031
 buttons-0.030
 stret-0.029
 rules-0.029
 blood-0.029
Token potato
Token ID21219
Feature activation+3.98e-02

Pos logit contributions
 stret+0.019
umbling+0.019
 reins+0.019
 pounds+0.018
co+0.018
Neg logit contributions
 named-0.024
arium-0.023
chard-0.022
 Captain-0.022
 lived-0.021
Token dolls
Token ID36062
Feature activation-7.30e-03

Pos logit contributions
 wounds+0.061
 buttons+0.057
 rules+0.056
 reins+0.056
 stret+0.055
Neg logit contributions
ores-0.063
year-0.060
ver-0.059
arium-0.058
 named-0.058
Token and
Token ID290
Feature activation+2.87e-02

Pos logit contributions
ores+0.013
year+0.012
 known+0.012
 Captain+0.012
arium+0.012
Neg logit contributions
 wounds-0.010
 buttons-0.009
 rules-0.009
 stret-0.009
co-0.009
Token their
Token ID511
Feature activation+3.83e-02

Pos logit contributions
 wounds+0.040
 buttons+0.037
 stret+0.036
 mess+0.036
 brakes+0.035
Neg logit contributions
ores-0.053
year-0.051
ver-0.051
arium-0.049
apped-0.048
Token other
Token ID584
Feature activation+9.38e-02

Pos logit contributions
 stret+0.047
 pounds+0.045
 reins+0.045
 wounds+0.044
 brakes+0.043
Neg logit contributions
 named-0.065
arium-0.062
chard-0.061
 Captain-0.060
 lived-0.059
Token dolls
Token ID36062
Feature activation+2.29e-02

Pos logit contributions
 stret+0.119
 wounds+0.114
 reins+0.114
 brakes+0.113
co+0.109
Neg logit contributions
 named-0.160
arium-0.149
 lived-0.144
apped-0.144
 Captain-0.142
Token.
Token ID13
Feature activation+1.78e-02

Pos logit contributions
 wounds+0.031
 stret+0.030
 buttons+0.029
 rules+0.029
 mess+0.029
Neg logit contributions
ores-0.040
year-0.038
arium-0.037
 Captain-0.037
ver-0.036
Token They
Token ID1119
Feature activation+2.91e-03

Pos logit contributions
 wounds+0.028
 buttons+0.028
 rules+0.028
 brakes+0.028
 backs+0.027
Neg logit contributions
year-0.025
ver-0.025
arium-0.025
chard-0.025
 named-0.023
Token had
Token ID550
Feature activation-1.08e-02

Pos logit contributions
 wounds+0.004
 mess+0.004
 buttons+0.004
co+0.004
 swings+0.004
Neg logit contributions
ores-0.005
year-0.005
arium-0.004
 Enjoy-0.004
ver-0.004
Token fun
Token ID1257
Feature activation-8.31e-03

Pos logit contributions
 named+0.014
ores+0.013
 known+0.013
arium+0.013
ver+0.012
Neg logit contributions
 wounds-0.019
 reins-0.019
 stret-0.018
 buttons-0.018
 pounds-0.018
Token mom
Token ID1995
Feature activation-1.83e-02

Pos logit contributions
 wounds+0.059
 reins+0.058
 stret+0.057
 thorn+0.053
 brakes+0.053
Neg logit contributions
 named-0.043
ores-0.042
 known-0.041
arium-0.038
 Captain-0.038
Tokenm
Token ID76
Feature activation+4.12e-02

Pos logit contributions
 named+0.029
ver+0.026
ores+0.026
arium+0.026
 known+0.025
Neg logit contributions
 wounds-0.029
 reins-0.028
 brakes-0.027
 buttons-0.027
 rules-0.026
Tokenies
Token ID444
Feature activation+3.06e-02

Pos logit contributions
 wounds+0.058
 reins+0.055
 buttons+0.055
 stret+0.054
 rules+0.053
Neg logit contributions
ores-0.068
 named-0.066
ver-0.065
arium-0.065
year-0.064
Token and
Token ID290
Feature activation+3.49e-02

Pos logit contributions
 wounds+0.050
 reins+0.048
 buttons+0.048
 stret+0.047
 brakes+0.047
Neg logit contributions
ores-0.043
 named-0.042
arium-0.041
ver-0.041
year-0.039
Token d
Token ID288
Feature activation+3.67e-02

Pos logit contributions
 wounds+0.055
 buttons+0.052
 reins+0.052
 stret+0.052
 rules+0.051
Neg logit contributions
ores-0.051
 named-0.050
arium-0.049
ver-0.049
year-0.048
Tokenadd
Token ID2860
Feature activation+9.25e-02

Pos logit contributions
 wounds+0.054
 rules+0.050
 buttons+0.049
 reins+0.049
 brakes+0.049
Neg logit contributions
ores-0.063
apped-0.058
year-0.058
ver-0.057
 named-0.057
Tokenies
Token ID444
Feature activation+4.75e-02

Pos logit contributions
 wounds+0.128
 buttons+0.124
 stret+0.121
 rules+0.121
 mess+0.120
Neg logit contributions
ores-0.151
arium-0.149
ver-0.148
 named-0.148
year-0.144
Token.
Token ID13
Feature activation+7.33e-03

Pos logit contributions
 wounds+0.079
 buttons+0.076
 reins+0.075
 stret+0.075
 rules+0.074
Neg logit contributions
ores-0.065
 named-0.062
arium-0.062
ver-0.062
year-0.061
Token Lily
Token ID20037
Feature activation+1.06e-04

Pos logit contributions
 wounds+0.011
 rules+0.011
 buttons+0.011
 stret+0.011
 brakes+0.010
Neg logit contributions
year-0.011
ver-0.010
working-0.010
ores-0.010
arium-0.010
Token and
Token ID290
Feature activation+2.88e-02

Pos logit contributions
 wounds+0.000
 mess+0.000
 buttons+0.000
 stret+0.000
 rules+0.000
Neg logit contributions
ores-0.000
year-0.000
arium-0.000
ver-0.000
chard-0.000
Token Ben
Token ID3932
Feature activation+2.18e-02

Pos logit contributions
 wounds+0.040
 buttons+0.038
 blood+0.037
 stret+0.037
 faces+0.037
Neg logit contributions
year-0.052
arium-0.049
ver-0.049
ores-0.048
apped-0.046
Token put
Token ID1234
Feature activation-5.76e-03

Pos logit contributions
 buttons+0.006
 wounds+0.006
 mess+0.006
 scarf+0.006
 swings+0.006
Neg logit contributions
ores-0.007
year-0.007
ver-0.007
arium-0.007
apped-0.007
Token it
Token ID340
Feature activation-2.33e-03

Pos logit contributions
arium+0.010
ver+0.010
ores+0.009
year+0.009
apped+0.009
Neg logit contributions
 buttons-0.008
 wounds-0.008
 blood-0.007
 reins-0.007
 mess-0.007
Token on
Token ID319
Feature activation-1.34e-02

Pos logit contributions
ores+0.003
arium+0.003
 named+0.003
ver+0.003
year+0.003
Neg logit contributions
 wounds-0.004
 buttons-0.004
 reins-0.004
 stret-0.004
 mess-0.004
Token her
Token ID607
Feature activation+1.23e-02

Pos logit contributions
ver+0.021
apped+0.021
year+0.021
arium+0.020
ores+0.020
Neg logit contributions
 buttons-0.019
 rules-0.019
 wounds-0.019
 mess-0.019
 stret-0.019
Token windows
Token ID9168
Feature activation-9.03e-03

Pos logit contributions
 pounds+0.015
 brakes+0.015
 reins+0.015
 buttons+0.014
 blood+0.014
Neg logit contributions
 named-0.022
arium-0.020
ores-0.020
 lived-0.020
chard-0.020
Tokenill
Token ID359
Feature activation+9.13e-02

Pos logit contributions
ores+0.012
 named+0.012
ver+0.012
arium+0.012
year+0.012
Neg logit contributions
 wounds-0.015
 stret-0.015
 reins-0.015
 buttons-0.015
 brakes-0.015
Token.
Token ID13
Feature activation+3.63e-03

Pos logit contributions
 wounds+0.171
 brakes+0.165
 blood+0.161
 rules+0.160
 reins+0.156
Neg logit contributions
 named-0.115
ores-0.104
arium-0.103
ver-0.098
 lived-0.095
Token Every
Token ID3887
Feature activation-2.96e-03

Pos logit contributions
 reins+0.005
 buttons+0.005
 rules+0.005
 backs+0.004
 wounds+0.004
Neg logit contributions
arium-0.006
ver-0.006
year-0.006
 known-0.006
 named-0.006
Token day
Token ID1110
Feature activation-3.57e-02

Pos logit contributions
ores+0.003
ver+0.002
year+0.002
arium+0.002
chard+0.002
Neg logit contributions
 wounds-0.006
 buttons-0.006
 mess-0.006
 rules-0.006
 reins-0.006
Token,
Token ID11
Feature activation-2.36e-02

Pos logit contributions
arium+0.043
year+0.041
ores+0.041
ver+0.038
chard+0.037
Neg logit contributions
 buttons-0.067
 mess-0.063
 rip-0.062
 rules-0.061
 ow-0.061
Token she
Token ID673
Feature activation+1.49e-03

Pos logit contributions
year+0.035
ver+0.034
arium+0.034
ores+0.034
apped+0.033
Neg logit contributions
 wounds-0.036
 buttons-0.036
 mess-0.035
 rules-0.034
 blood-0.034
Token bird
Token ID6512
Feature activation-5.70e-03

Pos logit contributions
 wounds+0.007
 reins+0.006
 stret+0.006
 brakes+0.006
 buttons+0.006
Neg logit contributions
 named-0.009
arium-0.009
ores-0.009
 lived-0.009
apped-0.008
Token sitting
Token ID5586
Feature activation-1.66e-02

Pos logit contributions
 named+0.008
arium+0.008
ores+0.007
 lived+0.007
ver+0.007
Neg logit contributions
 wounds-0.010
 reins-0.010
 stret-0.009
 buttons-0.009
 rules-0.009
Token on
Token ID319
Feature activation+4.23e-03

Pos logit contributions
ores+0.027
ver+0.027
 Captain+0.026
 known+0.026
arium+0.026
Neg logit contributions
 wounds-0.024
 blood-0.022
 bleeding-0.022
 mess-0.021
 reins-0.021
Token her
Token ID607
Feature activation+2.65e-02

Pos logit contributions
 wounds+0.008
 buttons+0.008
 stret+0.008
 rules+0.008
 mess+0.008
Neg logit contributions
ver-0.005
ores-0.005
year-0.005
apped-0.005
arium-0.004
Token windows
Token ID9168
Feature activation-7.39e-03

Pos logit contributions
 pounds+0.036
 buttons+0.033
 wounds+0.033
 brakes+0.032
Sur+0.032
Neg logit contributions
 named-0.044
arium-0.041
 every-0.040
chard-0.040
,-0.040
Tokenill
Token ID359
Feature activation+9.12e-02

Pos logit contributions
ores+0.009
 named+0.009
arium+0.009
ver+0.009
year+0.009
Neg logit contributions
 wounds-0.013
 reins-0.013
 stret-0.013
 buttons-0.013
 rules-0.013
Token!
Token ID0
Feature activation+1.68e-02

Pos logit contributions
 wounds+0.172
 brakes+0.161
 rules+0.158
 reins+0.158
 buttons+0.155
Neg logit contributions
 named-0.114
arium-0.108
ores-0.102
ver-0.098
 lived-0.098
Token
Token ID198
Feature activation+7.77e-03

Pos logit contributions
 wounds+0.026
 stret+0.026
 buttons+0.025
 rules+0.025
 reins+0.025
Neg logit contributions
ores-0.023
arium-0.022
ver-0.022
year-0.022
working-0.021
Token
Token ID198
Feature activation+7.08e-03

Pos logit contributions
 wounds+0.015
 buttons+0.014
 reins+0.014
 stret+0.014
 rules+0.014
Neg logit contributions
ores-0.009
 named-0.008
arium-0.008
ver-0.008
year-0.008
TokenL
Token ID43
Feature activation-8.84e-03

Pos logit contributions
 wounds+0.010
 reins+0.010
 brakes+0.010
 buttons+0.010
 stret+0.009
Neg logit contributions
 named-0.012
ores-0.012
year-0.011
arium-0.011
ver-0.011
Tokenily
Token ID813
Feature activation+1.39e-02

Pos logit contributions
ores+0.011
 named+0.011
year+0.010
arium+0.010
ver+0.010
Neg logit contributions
 wounds-0.017
 reins-0.016
 brakes-0.016
 buttons-0.016
 stret-0.016
Token a
Token ID257
Feature activation+5.92e-03

Pos logit contributions
ores+0.015
ver+0.015
arium+0.015
apped+0.015
year+0.014
Neg logit contributions
 wounds-0.027
 buttons-0.027
 stret-0.026
 rules-0.026
 reins-0.026
Token creative
Token ID7325
Feature activation+2.10e-02

Pos logit contributions
 wounds+0.007
 buttons+0.007
 reins+0.006
 stret+0.006
 rules+0.006
Neg logit contributions
ores-0.011
 named-0.011
arium-0.011
ver-0.011
year-0.010
Token chicken
Token ID9015
Feature activation+5.56e-03

Pos logit contributions
 wounds+0.025
 buttons+0.023
 stret+0.023
 reins+0.023
 rules+0.023
Neg logit contributions
ores-0.039
ver-0.038
arium-0.038
 named-0.038
year-0.038
Token named
Token ID3706
Feature activation-5.00e-02

Pos logit contributions
 wounds+0.010
 buttons+0.009
 stret+0.009
 rules+0.009
 reins+0.009
Neg logit contributions
ores-0.007
ver-0.007
arium-0.007
year-0.007
apped-0.006
Token Ch
Token ID609
Feature activation+3.00e-03

Pos logit contributions
ores+0.063
arium+0.062
year+0.061
ver+0.060
apped+0.060
Neg logit contributions
 buttons-0.083
 reins-0.083
 wounds-0.082
 stret-0.081
 rules-0.081
Tokenicky
Token ID17479
Feature activation+9.03e-02

Pos logit contributions
 wounds+0.006
 buttons+0.005
 reins+0.005
 stret+0.005
 rules+0.005
Neg logit contributions
ores-0.004
 named-0.004
arium-0.003
ver-0.003
year-0.003
Token.
Token ID13
Feature activation+5.67e-03

Pos logit contributions
 wounds+0.144
 buttons+0.137
 reins+0.136
 stret+0.135
 rules+0.134
Neg logit contributions
ores-0.131
 named-0.126
arium-0.126
ver-0.125
year-0.123
Token Ch
Token ID609
Feature activation-1.34e-02

Pos logit contributions
 buttons+0.009
 wounds+0.009
 reins+0.009
 brakes+0.009
 rules+0.009
Neg logit contributions
ores-0.008
arium-0.007
ver-0.007
year-0.007
chard-0.007
Tokenicky
Token ID17479
Feature activation+6.88e-02

Pos logit contributions
ores+0.024
 named+0.024
arium+0.024
chard+0.024
year+0.023
Neg logit contributions
 wounds-0.016
 buttons-0.015
 rules-0.015
 mess-0.015
 reins-0.015
Token liked
Token ID8288
Feature activation+8.42e-04

Pos logit contributions
 wounds+0.123
 buttons+0.120
 mess+0.116
 rules+0.116
 stret+0.115
Neg logit contributions
ores-0.086
ver-0.081
arium-0.080
year-0.079
chard-0.078
Token to
Token ID284
Feature activation+5.51e-03

Pos logit contributions
 wounds+0.002
 stret+0.001
 reins+0.001
 mess+0.001
 buttons+0.001
Neg logit contributions
ores-0.001
 named-0.001
arium-0.001
year-0.001
 Captain-0.001
Token the
Token ID262
Feature activation+2.94e-03

Pos logit contributions
ores+0.025
year+0.025
arium+0.025
ver+0.024
 named+0.023
Neg logit contributions
 wounds-0.041
 buttons-0.039
 rules-0.038
 blood-0.037
 mess-0.037
Token toy
Token ID13373
Feature activation+4.46e-03

Pos logit contributions
 wounds+0.004
 stret+0.004
 pounds+0.004
we+0.004
co+0.004
Neg logit contributions
 known-0.004
 named-0.004
 night-0.004
working-0.004
 surely-0.004
Token plane
Token ID6614
Feature activation-1.09e-02

Pos logit contributions
 wounds+0.005
 reins+0.005
 stret+0.005
 buttons+0.005
 rules+0.005
Neg logit contributions
 named-0.008
arium-0.008
ores-0.008
ver-0.008
apped-0.008
Token and
Token ID290
Feature activation+1.42e-02

Pos logit contributions
ores+0.016
year+0.016
ver+0.015
arium+0.015
chard+0.015
Neg logit contributions
 wounds-0.017
 buttons-0.016
 rules-0.016
 mess-0.016
 stret-0.016
Token their
Token ID511
Feature activation+3.50e-02

Pos logit contributions
 wounds+0.025
 rules+0.024
 mess+0.024
 buttons+0.023
 brakes+0.023
Neg logit contributions
year-0.020
ores-0.020
ver-0.019
arium-0.019
apped-0.018
Token other
Token ID584
Feature activation+8.98e-02

Pos logit contributions
 stret+0.043
 reins+0.043
 wounds+0.043
 pounds+0.042
 buttons+0.041
Neg logit contributions
 named-0.059
arium-0.057
ores-0.056
apped-0.054
 lived-0.054
Token toys
Token ID14958
Feature activation+2.38e-02

Pos logit contributions
 stret+0.115
 reins+0.115
 wounds+0.113
 thorn+0.109
 brakes+0.108
Neg logit contributions
 named-0.152
 lived-0.143
arium-0.135
apped-0.134
working-0.130
Token.
Token ID13
Feature activation+1.46e-02

Pos logit contributions
 wounds+0.035
 buttons+0.033
 rules+0.033
 mess+0.033
 stret+0.033
Neg logit contributions
ores-0.038
year-0.036
arium-0.036
 named-0.036
ver-0.035
Token They
Token ID1119
Feature activation+5.66e-03

Pos logit contributions
 wounds+0.022
 rules+0.022
 buttons+0.021
 blood+0.021
 brakes+0.020
Neg logit contributions
year-0.022
chard-0.021
arium-0.021
ver-0.020
working-0.020
Token are
Token ID389
Feature activation+3.30e-02

Pos logit contributions
 wounds+0.008
 mess+0.008
 rip+0.008
 buttons+0.008
 rules+0.008
Neg logit contributions
ores-0.010
year-0.009
arium-0.009
apped-0.009
chard-0.008
Token happy
Token ID3772
Feature activation+1.63e-02

Pos logit contributions
 wounds+0.054
 mess+0.051
 buttons+0.049
 blood+0.048
 rules+0.048
Neg logit contributions
arium-0.050
year-0.047
ver-0.047
ores-0.046
apped-0.044
Token placed
Token ID4624
Feature activation-5.38e-03

Pos logit contributions
 buttons+0.000
 wounds+0.000
 mess+0.000
 faces+0.000
 blood+0.000
Neg logit contributions
ores-0.000
ver-0.000
year-0.000
arium-0.000
apped-0.000
Token it
Token ID340
Feature activation-1.10e-03

Pos logit contributions
ver+0.009
arium+0.009
ores+0.009
 named+0.009
apped+0.009
Neg logit contributions
 buttons-0.007
 wounds-0.007
 rules-0.007
 reins-0.007
 blood-0.007
Token on
Token ID319
Feature activation-1.12e-02

Pos logit contributions
ores+0.001
arium+0.001
 named+0.001
ver+0.001
 Captain+0.001
Neg logit contributions
 wounds-0.002
 buttons-0.002
 reins-0.002
 mess-0.002
 stret-0.002
Token her
Token ID607
Feature activation+1.65e-02

Pos logit contributions
ver+0.015
ores+0.015
year+0.015
arium+0.015
apped+0.015
Neg logit contributions
 buttons-0.019
 wounds-0.018
 rules-0.018
 mess-0.018
 stret-0.018
Token windows
Token ID9168
Feature activation-1.01e-02

Pos logit contributions
 brakes+0.021
 reins+0.021
 buttons+0.020
 stret+0.020
 blood+0.020
Neg logit contributions
 named-0.027
chard-0.026
arium-0.026
ores-0.025
 surely-0.025
Tokenill
Token ID359
Feature activation+8.89e-02

Pos logit contributions
ores+0.014
 named+0.014
arium+0.014
ver+0.013
year+0.013
Neg logit contributions
 wounds-0.017
 reins-0.016
 buttons-0.016
 stret-0.016
 rules-0.016
Token.
Token ID13
Feature activation+3.34e-03

Pos logit contributions
 wounds+0.159
 brakes+0.153
 rules+0.149
 blood+0.147
 reins+0.147
Neg logit contributions
 named-0.116
ores-0.109
arium-0.108
ver-0.104
chard-0.102
Token Every
Token ID3887
Feature activation-4.39e-04

Pos logit contributions
 buttons+0.005
 rules+0.005
 wounds+0.005
 reins+0.005
 blood+0.005
Neg logit contributions
arium-0.005
ver-0.005
chard-0.005
ores-0.004
year-0.004
Token time
Token ID640
Feature activation-3.95e-02

Pos logit contributions
ores+0.000
ver+0.000
chard+0.000
arium+0.000
apped+0.000
Neg logit contributions
 wounds-0.001
 rules-0.001
 buttons-0.001
 mess-0.001
 reins-0.001
Token she
Token ID673
Feature activation-4.53e-03

Pos logit contributions
ores+0.059
year+0.057
arium+0.057
chard+0.053
ver+0.053
Neg logit contributions
 buttons-0.062
 rules-0.059
 wounds-0.059
 mess-0.059
 blood-0.057
Token saw
Token ID2497
Feature activation-1.95e-02

Pos logit contributions
ores+0.007
arium+0.006
chard+0.006
ver+0.006
apped+0.006
Neg logit contributions
 wounds-0.007
 buttons-0.007
 mess-0.007
 blood-0.007
 brakes-0.007
Token mom
Token ID1995
Feature activation-1.52e-02

Pos logit contributions
ores+0.004
 named+0.004
arium+0.004
ver+0.004
apped+0.004
Neg logit contributions
 stret-0.005
 reins-0.004
 wounds-0.004
 buttons-0.004
 brakes-0.004
Tokenm
Token ID76
Feature activation+3.61e-02

Pos logit contributions
ores+0.029
arium+0.028
year+0.028
apped+0.027
 named+0.027
Neg logit contributions
 wounds-0.017
 buttons-0.017
 mess-0.016
 rules-0.016
 reins-0.016
Tokenies
Token ID444
Feature activation+2.27e-02

Pos logit contributions
 wounds+0.051
 rules+0.048
 buttons+0.047
 mess+0.047
 reins+0.045
Neg logit contributions
arium-0.060
 named-0.059
year-0.058
chard-0.058
ver-0.058
Token and
Token ID290
Feature activation+2.51e-02

Pos logit contributions
 wounds+0.035
 rules+0.034
 buttons+0.033
 mess+0.033
 reins+0.032
Neg logit contributions
ores-0.033
year-0.033
apped-0.032
 named-0.032
arium-0.031
Token d
Token ID288
Feature activation+2.95e-02

Pos logit contributions
 wounds+0.039
 rules+0.037
 buttons+0.036
 mess+0.036
 reins+0.036
Neg logit contributions
ores-0.039
year-0.038
arium-0.038
ver-0.038
apped-0.037
Tokenadd
Token ID2860
Feature activation+8.84e-02

Pos logit contributions
 wounds+0.043
 buttons+0.041
 rules+0.041
 reins+0.040
 brakes+0.040
Neg logit contributions
ores-0.049
ver-0.045
year-0.045
 named-0.045
apped-0.045
Tokenies
Token ID444
Feature activation+3.98e-02

Pos logit contributions
 wounds+0.103
 buttons+0.096
 rules+0.096
 reins+0.093
 mess+0.093
Neg logit contributions
 named-0.166
arium-0.165
ver-0.164
ores-0.163
chard-0.162
Token when
Token ID618
Feature activation+2.55e-02

Pos logit contributions
 wounds+0.067
 rules+0.065
 buttons+0.064
 stret+0.064
 reins+0.063
Neg logit contributions
ores-0.052
year-0.052
 named-0.049
arium-0.049
apped-0.049
Token they
Token ID484
Feature activation+2.13e-02

Pos logit contributions
 wounds+0.041
 rules+0.039
 buttons+0.039
 mess+0.038
 reins+0.037
Neg logit contributions
ores-0.036
year-0.036
ver-0.036
 named-0.036
arium-0.035
Token say
Token ID910
Feature activation+1.29e-02

Pos logit contributions
 mess+0.030
 wounds+0.030
 buttons+0.029
 rules+0.029
 blood+0.028
Neg logit contributions
year-0.035
ores-0.034
apped-0.033
arium-0.033
 named-0.033
Token things
Token ID1243
Feature activation+1.32e-03

Pos logit contributions
 wounds+0.023
 buttons+0.023
 rules+0.022
 reins+0.022
 stret+0.022
Neg logit contributions
arium-0.016
ores-0.016
 named-0.015
year-0.015
ver-0.015
Token their
Token ID511
Feature activation+1.45e-02

Pos logit contributions
 wounds+0.023
 stret+0.022
 reins+0.022
 buttons+0.022
 mess+0.021
Neg logit contributions
 named-0.018
ores-0.018
 known-0.017
arium-0.017
ver-0.017
Token sc
Token ID629
Feature activation+4.31e-02

Pos logit contributions
 pounds+0.018
 wounds+0.018
 stret+0.017
 reins+0.017
 thorn+0.017
Neg logit contributions
 named-0.025
 known-0.024
 lived-0.023
 Captain-0.022
ver-0.022
Tokenooters
Token ID48316
Feature activation+4.45e-03

Pos logit contributions
 wounds+0.044
 buttons+0.041
 stret+0.041
 reins+0.040
 mess+0.040
Neg logit contributions
ores-0.086
 named-0.085
ver-0.085
arium-0.084
year-0.083
Token and
Token ID290
Feature activation+1.56e-02

Pos logit contributions
 wounds+0.007
 buttons+0.007
 reins+0.007
 rules+0.007
 stret+0.007
Neg logit contributions
ores-0.007
arium-0.006
year-0.006
 named-0.006
 known-0.006
Token their
Token ID511
Feature activation+3.71e-02

Pos logit contributions
 wounds+0.029
 buttons+0.027
 brakes+0.027
 rules+0.027
 reins+0.027
Neg logit contributions
ores-0.020
year-0.019
arium-0.019
ver-0.019
 named-0.018
Token other
Token ID584
Feature activation+8.84e-02

Pos logit contributions
 pounds+0.045
 wounds+0.044
 stret+0.043
 reins+0.043
 thorn+0.042
Neg logit contributions
 named-0.066
 known-0.061
 lived-0.060
ores-0.060
arium-0.060
Token things
Token ID1243
Feature activation+3.39e-02

Pos logit contributions
 thorn+0.140
 wounds+0.136
 pounds+0.136
 stret+0.135
 reins+0.132
Neg logit contributions
 named-0.132
 lived-0.115
,-0.112
apped-0.110
 known-0.110
Token.
Token ID13
Feature activation+1.20e-02

Pos logit contributions
 wounds+0.060
 reins+0.058
 stret+0.057
 buttons+0.057
 brakes+0.056
Neg logit contributions
ores-0.043
ver-0.042
 named-0.042
arium-0.041
apped-0.039
Token They
Token ID1119
Feature activation+1.99e-03

Pos logit contributions
 wounds+0.019
 buttons+0.018
 rules+0.018
 brakes+0.018
 blood+0.018
Neg logit contributions
arium-0.017
year-0.017
chard-0.016
ver-0.016
ores-0.016
Token still
Token ID991
Feature activation-1.63e-02

Pos logit contributions
 wounds+0.003
 buttons+0.003
 mess+0.003
 brakes+0.003
 rules+0.003
Neg logit contributions
ores-0.003
year-0.003
arium-0.003
apped-0.003
chard-0.003
Token have
Token ID423
Feature activation+1.57e-02

Pos logit contributions
arium+0.022
ores+0.021
year+0.021
ver+0.020
apped+0.020
Neg logit contributions
 wounds-0.028
 mess-0.027
 brakes-0.027
 buttons-0.027
 rules-0.027
TokenB
Token ID33
Feature activation+4.30e-03

Pos logit contributions
 wounds+0.009
 reins+0.008
 buttons+0.008
 stret+0.008
 brakes+0.008
Neg logit contributions
 named-0.012
ores-0.011
arium-0.011
year-0.011
ver-0.011
Tokenub
Token ID549
Feature activation-1.70e-02

Pos logit contributions
 wounds+0.010
 stret+0.009
 mess+0.009
 brakes+0.009
 buttons+0.009
Neg logit contributions
ver-0.004
 named-0.004
apped-0.004
year-0.004
ores-0.004
Tokenba
Token ID7012
Feature activation+3.38e-03

Pos logit contributions
ores+0.034
arium+0.032
year+0.032
 named+0.032
ver+0.031
Neg logit contributions
 wounds-0.018
 buttons-0.017
 reins-0.016
 stret-0.016
 mess-0.016
Token and
Token ID290
Feature activation+1.64e-02

Pos logit contributions
 wounds+0.005
 buttons+0.005
 mess+0.005
co+0.005
 stret+0.004
Neg logit contributions
ores-0.006
arium-0.005
year-0.005
chard-0.005
ver-0.005
Token his
Token ID465
Feature activation+3.28e-02

Pos logit contributions
 wounds+0.033
 buttons+0.032
 rules+0.031
 mess+0.031
 stret+0.031
Neg logit contributions
ores-0.018
ver-0.017
year-0.016
arium-0.016
 surely-0.015
Token other
Token ID584
Feature activation+8.82e-02

Pos logit contributions
 stret+0.055
 wounds+0.055
 reins+0.054
 buttons+0.052
 pounds+0.052
Neg logit contributions
 named-0.045
 Captain-0.041
arium-0.041
apped-0.041
 known-0.041
Token friends
Token ID2460
Feature activation+3.38e-02

Pos logit contributions
 wounds+0.149
 stret+0.147
 reins+0.142
 brakes+0.138
 buttons+0.137
Neg logit contributions
 named-0.129
apped-0.116
 known-0.112
 lived-0.111
ver-0.111
Token were
Token ID547
Feature activation+4.21e-02

Pos logit contributions
 buttons+0.054
 wounds+0.054
 stret+0.052
 mess+0.052
 brakes+0.051
Neg logit contributions
ores-0.050
arium-0.046
ver-0.043
year-0.043
chard-0.042
Token hiding
Token ID11816
Feature activation+3.39e-02

Pos logit contributions
 wounds+0.067
 mess+0.065
 stret+0.064
 brakes+0.062
 buttons+0.061
Neg logit contributions
arium-0.063
ores-0.060
year-0.058
ver-0.057
apped-0.055
Token for
Token ID329
Feature activation+3.39e-03

Pos logit contributions
 wounds+0.056
 buttons+0.053
 mess+0.052
 rules+0.050
 stret+0.049
Neg logit contributions
ores-0.048
year-0.048
ver-0.046
 known-0.046
 Enjoy-0.044
Token a
Token ID257
Feature activation+4.96e-03

Pos logit contributions
 wounds+0.007
 reins+0.006
 buttons+0.006
 stret+0.006
 rules+0.006
Neg logit contributions
ores-0.004
 named-0.004
arium-0.004
ver-0.004
year-0.003
Token leave
Token ID2666
Feature activation+8.10e-03

Pos logit contributions
 wounds+0.015
 stret+0.014
 buttons+0.014
 rules+0.014
 reins+0.013
Neg logit contributions
ores-0.013
year-0.013
ver-0.013
arium-0.013
apped-0.013
Token their
Token ID511
Feature activation+2.21e-02

Pos logit contributions
 wounds+0.013
 rules+0.012
 buttons+0.012
 blood+0.012
 stret+0.012
Neg logit contributions
ver-0.012
arium-0.012
apped-0.012
ores-0.012
year-0.011
Token sons
Token ID11989
Feature activation+5.54e-02

Pos logit contributions
 reins+0.025
 wounds+0.024
 buttons+0.024
 stret+0.024
 pounds+0.024
Neg logit contributions
 named-0.041
arium-0.040
ores-0.039
 Captain-0.038
chard-0.037
Token and
Token ID290
Feature activation+3.08e-02

Pos logit contributions
 wounds+0.103
 buttons+0.100
 reins+0.099
 stret+0.098
 brakes+0.098
Neg logit contributions
 named-0.064
ores-0.062
arium-0.062
ver-0.060
year-0.057
Token their
Token ID511
Feature activation+3.54e-02

Pos logit contributions
 wounds+0.051
 buttons+0.048
 stret+0.048
 reins+0.048
 rules+0.047
Neg logit contributions
ores-0.044
arium-0.042
ver-0.042
year-0.041
 named-0.041
Token hut
Token ID40812
Feature activation+8.77e-02

Pos logit contributions
 reins+0.039
 pounds+0.039
 wounds+0.038
 buttons+0.038
 stret+0.038
Neg logit contributions
 named-0.065
arium-0.062
ores-0.062
 Captain-0.060
chard-0.059
Token alone
Token ID3436
Feature activation+2.49e-02

Pos logit contributions
 wounds+0.146
 buttons+0.138
 stret+0.137
 reins+0.137
 rules+0.136
Neg logit contributions
ores-0.123
ver-0.117
arium-0.116
 named-0.116
year-0.115
Token.
Token ID13
Feature activation+1.27e-02

Pos logit contributions
 wounds+0.036
 rules+0.034
 buttons+0.034
 mess+0.033
 stret+0.033
Neg logit contributions
ores-0.040
arium-0.039
ver-0.038
year-0.038
 named-0.038
Token They
Token ID1119
Feature activation+6.37e-03

Pos logit contributions
 wounds+0.022
 buttons+0.021
 rules+0.021
 stret+0.021
 brakes+0.021
Neg logit contributions
ver-0.016
ores-0.016
year-0.016
arium-0.016
apped-0.015
Token say
Token ID910
Feature activation+5.37e-03

Pos logit contributions
 wounds+0.011
 stret+0.010
 buttons+0.010
 brakes+0.010
 mess+0.010
Neg logit contributions
ores-0.009
arium-0.009
apped-0.009
ver-0.009
year-0.009
Token they
Token ID484
Feature activation+1.35e-02

Pos logit contributions
 wounds+0.010
 stret+0.010
 buttons+0.010
 rules+0.010
 ow+0.010
Neg logit contributions
arium-0.007
year-0.006
ores-0.006
chard-0.006
ver-0.006
Token and
Token ID290
Feature activation+1.07e-02

Pos logit contributions
 wounds+0.014
 reins+0.013
 stret+0.013
 rules+0.013
 buttons+0.013
Neg logit contributions
ores-0.016
year-0.016
ver-0.015
arium-0.014
 known-0.014
Token Pupp
Token ID20926
Feature activation+2.21e-03

Pos logit contributions
 wounds+0.017
 stret+0.016
 mess+0.015
 scarf+0.015
 reins+0.015
Neg logit contributions
ores-0.016
year-0.016
ver-0.016
arium-0.015
apped-0.015
Tokeny
Token ID88
Feature activation+3.94e-02

Pos logit contributions
 wounds+0.003
 buttons+0.003
 reins+0.003
 stret+0.003
 rules+0.003
Neg logit contributions
ores-0.004
 named-0.004
arium-0.003
ver-0.003
year-0.003
Token and
Token ID290
Feature activation+2.92e-02

Pos logit contributions
 wounds+0.069
 buttons+0.065
 reins+0.065
 stret+0.064
 rules+0.063
Neg logit contributions
ores-0.053
year-0.050
ver-0.049
arium-0.048
chard-0.047
Token w
Token ID266
Feature activation+4.47e-02

Pos logit contributions
 wounds+0.046
 stret+0.043
 reins+0.043
 brakes+0.041
 scarf+0.041
Neg logit contributions
ores-0.045
year-0.044
ver-0.043
arium-0.042
apped-0.039
Tokenadd
Token ID2860
Feature activation+8.75e-02

Pos logit contributions
 wounds+0.073
 buttons+0.070
 reins+0.070
 stret+0.069
 rules+0.069
Neg logit contributions
ores-0.063
 named-0.062
arium-0.061
ver-0.060
year-0.060
Tokenled
Token ID992
Feature activation+2.83e-02

Pos logit contributions
 wounds+0.135
 buttons+0.129
 reins+0.128
 stret+0.127
 rules+0.125
Neg logit contributions
ores-0.129
arium-0.127
ver-0.126
chard-0.123
 named-0.122
Token home
Token ID1363
Feature activation+1.97e-02

Pos logit contributions
 wounds+0.046
 buttons+0.044
 stret+0.043
 reins+0.043
 rules+0.042
Neg logit contributions
ver-0.041
ores-0.040
year-0.039
arium-0.039
chard-0.037
Token.
Token ID13
Feature activation-6.95e-04

Pos logit contributions
 wounds+0.031
 stret+0.029
 rules+0.028
 buttons+0.028
 mess+0.028
Neg logit contributions
ores-0.029
arium-0.028
year-0.027
ver-0.027
 surely-0.027
Token Mama
Token ID40084
Feature activation+5.22e-03

Pos logit contributions
ores+0.001
ver+0.001
arium+0.001
year+0.001
chard+0.001
Neg logit contributions
 wounds-0.001
 reins-0.001
 stret-0.001
 buttons-0.001
 brakes-0.001
Token Duck
Token ID21867
Feature activation+2.52e-02

Pos logit contributions
 wounds+0.009
 buttons+0.008
 reins+0.008
 stret+0.008
 rules+0.008
Neg logit contributions
ores-0.007
 named-0.007
arium-0.007
ver-0.007
year-0.007
Tokenight
Token ID432
Feature activation+1.96e-02

Pos logit contributions
ores+0.026
 named+0.026
 Captain+0.025
arium+0.025
ver+0.024
Neg logit contributions
 wounds-0.026
 reins-0.025
 stret-0.025
 rules-0.025
co-0.025
Tokened
Token ID276
Feature activation+1.01e-02

Pos logit contributions
 wounds+0.034
 buttons+0.033
 reins+0.032
 stret+0.032
 rules+0.032
Neg logit contributions
ores-0.026
 named-0.025
arium-0.025
ver-0.025
year-0.024
Token on
Token ID319
Feature activation+3.13e-03

Pos logit contributions
 buttons+0.016
 wounds+0.016
 reins+0.016
 stret+0.016
 brakes+0.015
Neg logit contributions
ores-0.014
ver-0.014
arium-0.014
year-0.014
apped-0.013
Token his
Token ID465
Feature activation+2.16e-02

Pos logit contributions
 buttons+0.006
 wounds+0.006
 reins+0.006
 brakes+0.006
 blood+0.006
Neg logit contributions
ores-0.004
year-0.004
ver-0.004
arium-0.004
apped-0.004
Token windows
Token ID9168
Feature activation-8.22e-03

Pos logit contributions
 blood+0.032
we+0.031
co+0.031
 pounds+0.031
 stain+0.031
Neg logit contributions
 named-0.031
chard-0.030
arium-0.030
,-0.030
 every-0.029
Tokenill
Token ID359
Feature activation+8.72e-02

Pos logit contributions
ores+0.011
 named+0.011
ver+0.010
arium+0.010
apped+0.010
Neg logit contributions
 wounds-0.015
 stret-0.014
 reins-0.014
 rules-0.014
 buttons-0.014
Token and
Token ID290
Feature activation+2.30e-02

Pos logit contributions
 wounds+0.148
 rules+0.137
 brakes+0.136
 reins+0.134
 stret+0.132
Neg logit contributions
 named-0.129
arium-0.119
ores-0.114
 lived-0.114
chard-0.112
Token said
Token ID531
Feature activation+1.28e-02

Pos logit contributions
 buttons+0.037
 wounds+0.036
 brakes+0.034
 reins+0.034
 blood+0.034
Neg logit contributions
ores-0.036
ver-0.033
year-0.033
arium-0.033
apped-0.032
Token:
Token ID25
Feature activation-1.12e-02

Pos logit contributions
 buttons+0.022
 rules+0.022
 ow+0.022
 wounds+0.021
 brakes+0.021
Neg logit contributions
ores-0.017
arium-0.015
year-0.014
 surely-0.014
 known-0.014
Token "
Token ID366
Feature activation+3.92e-03

Pos logit contributions
ores+0.021
arium+0.021
year+0.021
ver+0.020
apped+0.019
Neg logit contributions
 reins-0.011
 buttons-0.011
 rules-0.011
 brakes-0.011
 stret-0.010
TokenDo
Token ID5211
Feature activation+2.25e-03

Pos logit contributions
 wounds+0.007
 mess+0.007
 stret+0.007
 brakes+0.006
 rules+0.006
Neg logit contributions
 known-0.005
 named-0.005
year-0.005
chard-0.005
 Cr-0.005
Token his
Token ID465
Feature activation+1.32e-02

Pos logit contributions
ores+0.022
arium+0.021
ver+0.021
 named+0.021
year+0.021
Neg logit contributions
 wounds-0.027
 buttons-0.026
 reins-0.025
 stret-0.025
 rules-0.025
Token mom
Token ID1995
Feature activation+4.08e-03

Pos logit contributions
 wounds+0.020
 stret+0.019
 reins+0.019
 rules+0.018
 buttons+0.018
Neg logit contributions
 named-0.021
ores-0.020
arium-0.019
apped-0.019
ver-0.019
Token and
Token ID290
Feature activation+3.21e-02

Pos logit contributions
 wounds+0.006
 buttons+0.006
 reins+0.006
 stret+0.006
 rules+0.006
Neg logit contributions
ores-0.006
arium-0.006
ver-0.006
year-0.006
 named-0.006
Token his
Token ID465
Feature activation+3.22e-02

Pos logit contributions
 wounds+0.041
 stret+0.038
 mess+0.038
 buttons+0.038
 reins+0.037
Neg logit contributions
ores-0.059
year-0.058
arium-0.055
ver-0.055
 surely-0.053
Token sp
Token ID599
Feature activation+7.71e-02

Pos logit contributions
 wounds+0.047
 stret+0.046
 reins+0.045
 pounds+0.044
 buttons+0.044
Neg logit contributions
 named-0.051
ores-0.047
apped-0.047
arium-0.047
 Captain-0.047
Tokenade
Token ID671
Feature activation+8.71e-02

Pos logit contributions
 stret+0.149
 reins+0.149
 wounds+0.147
 brakes+0.145
 thorn+0.139
Neg logit contributions
apped-0.102
ores-0.098
year-0.092
arium-0.087
ver-0.082
Token.
Token ID13
Feature activation+1.37e-02

Pos logit contributions
 wounds+0.145
 reins+0.138
 buttons+0.137
 stret+0.137
 rules+0.135
Neg logit contributions
ores-0.120
 named-0.119
arium-0.117
ver-0.115
apped-0.113
Token
Token ID198
Feature activation+1.31e-02

Pos logit contributions
 wounds+0.023
 buttons+0.023
 reins+0.022
 stret+0.022
 rules+0.022
Neg logit contributions
ores-0.017
arium-0.017
year-0.017
ver-0.017
 named-0.016
Token
Token ID198
Feature activation+1.20e-02

Pos logit contributions
 wounds+0.027
 reins+0.026
 buttons+0.026
 stret+0.026
 brakes+0.026
Neg logit contributions
ores-0.013
 named-0.013
arium-0.013
ver-0.013
year-0.012
TokenAt
Token ID2953
Feature activation-2.48e-03

Pos logit contributions
 wounds+0.021
 reins+0.019
 buttons+0.019
 brakes+0.019
 rules+0.018
Neg logit contributions
 named-0.018
ores-0.017
ver-0.017
arium-0.017
year-0.016
Token the
Token ID262
Feature activation+2.39e-03

Pos logit contributions
year+0.003
ver+0.003
apped+0.003
ores+0.003
 named+0.003
Neg logit contributions
 mess-0.004
 blood-0.004
 wounds-0.004
 rules-0.004
 reins-0.004
Token would
Token ID561
Feature activation-1.31e-02

Pos logit contributions
ores+0.016
arium+0.015
ver+0.015
year+0.015
apped+0.015
Neg logit contributions
 wounds-0.020
 buttons-0.020
 reins-0.019
 stret-0.019
 mess-0.019
Token always
Token ID1464
Feature activation-3.70e-02

Pos logit contributions
ores+0.019
 named+0.018
arium+0.018
ver+0.018
year+0.018
Neg logit contributions
 wounds-0.021
 buttons-0.020
 reins-0.020
 stret-0.020
 rules-0.020
Token say
Token ID910
Feature activation-6.23e-03

Pos logit contributions
ores+0.050
 named+0.049
arium+0.048
ver+0.048
year+0.047
Neg logit contributions
 wounds-0.062
 buttons-0.059
 reins-0.059
 stret-0.059
 rules-0.058
Token '
Token ID705
Feature activation-3.08e-02

Pos logit contributions
ores+0.008
arium+0.008
 named+0.008
year+0.007
ver+0.007
Neg logit contributions
 wounds-0.011
 buttons-0.011
 reins-0.011
 stret-0.010
 rules-0.010
Tokenthank
Token ID40716
Feature activation-5.33e-02

Pos logit contributions
ores+0.051
arium+0.050
 named+0.050
ver+0.049
apped+0.048
Neg logit contributions
 wounds-0.041
 buttons-0.040
 reins-0.040
 stret-0.039
 rules-0.038
Token-
Token ID12
Feature activation-1.04e-01

Pos logit contributions
ores+0.105
arium+0.104
 named+0.102
ver+0.102
apped+0.101
Neg logit contributions
 wounds-0.056
 buttons-0.054
 rules-0.054
 reins-0.053
 stret-0.052
Tokenyou
Token ID5832
Feature activation-3.27e-02

Pos logit contributions
arium+0.174
apped+0.153
ores+0.153
 named+0.150
 lived+0.145
Neg logit contributions
 reins-0.144
 rules-0.140
 buttons-0.138
 pounds-0.135
 instructions-0.134
Token'
Token ID6
Feature activation-2.13e-02

Pos logit contributions
arium+0.050
ores+0.050
 named+0.049
chard+0.048
apped+0.047
Neg logit contributions
 reins-0.047
 buttons-0.046
 rules-0.045
 stret-0.044
 wounds-0.044
Token to
Token ID284
Feature activation-1.00e-02

Pos logit contributions
arium+0.035
ores+0.035
chard+0.033
 named+0.032
ver+0.032
Neg logit contributions
 buttons-0.030
 rules-0.030
 wounds-0.029
 reins-0.029
 stret-0.027
Token his
Token ID465
Feature activation-1.96e-03

Pos logit contributions
ores+0.013
 named+0.013
arium+0.012
ver+0.012
year+0.012
Neg logit contributions
 wounds-0.018
 buttons-0.017
 reins-0.017
 stret-0.017
 rules-0.017
Token toast
Token ID27805
Feature activation-3.37e-03

Pos logit contributions
ores+0.003
 named+0.003
arium+0.003
ver+0.003
year+0.003
Neg logit contributions
 wounds-0.003
 stret-0.003
 reins-0.003
 buttons-0.003
 brakes-0.003
Token
Token ID198
Feature activation-3.67e-03

Pos logit contributions
ores+0.004
 named+0.004
arium+0.004
ver+0.004
year+0.004
Neg logit contributions
 wounds-0.007
 buttons-0.007
 reins-0.007
 stret-0.007
 rules-0.007
TokenSarah
Token ID29284
Feature activation+8.57e-03

Pos logit contributions
ores+0.006
 named+0.006
arium+0.006
ver+0.006
year+0.006
Neg logit contributions
 wounds-0.005
 buttons-0.005
 reins-0.005
 stret-0.005
 brakes-0.005
Token was
Token ID373
Feature activation+1.83e-03

Pos logit contributions
 wounds+0.014
 buttons+0.013
 reins+0.013
 stret+0.013
 rules+0.013
Neg logit contributions
ores-0.012
arium-0.012
 named-0.011
ver-0.011
year-0.011
Token so
Token ID523
Feature activation-4.56e-03

Pos logit contributions
 wounds+0.002
 mess+0.002
 stret+0.002
 buttons+0.002
 brakes+0.002
Neg logit contributions
arium-0.003
ores-0.003
ver-0.003
year-0.003
working-0.003
Token awe
Token ID25030
Feature activation+9.01e-03

Pos logit contributions
arium+0.008
riage+0.007
 Cr+0.007
working+0.007
 Captain+0.007
Neg logit contributions
 mess-0.006
 broken-0.006
 wounds-0.006
 buttons-0.005
 rip-0.005
Token-
Token ID12
Feature activation-1.02e-01

Pos logit contributions
 wounds+0.018
 reins+0.018
 stret+0.017
 brakes+0.017
 rules+0.017
Neg logit contributions
ores-0.010
 named-0.009
arium-0.009
year-0.009
ver-0.009
Tokenstru
Token ID19554
Feature activation-1.39e-02

Pos logit contributions
arium+0.172
ores+0.170
 named+0.169
apped+0.164
 lived+0.160
Neg logit contributions
 buttons-0.130
 reins-0.129
 rules-0.127
 wounds-0.126
 pounds-0.119
Tokenck
Token ID694
Feature activation+6.29e-03

Pos logit contributions
ores+0.023
 named+0.022
arium+0.022
year+0.021
ver+0.021
Neg logit contributions
 wounds-0.018
 reins-0.018
 stret-0.018
 buttons-0.018
 brakes-0.017
Token.
Token ID13
Feature activation-7.34e-03

Pos logit contributions
 wounds+0.010
 stret+0.010
 reins+0.010
 brakes+0.010
 rules+0.009
Neg logit contributions
ores-0.009
 named-0.009
arium-0.008
ver-0.008
year-0.008
Token It
Token ID632
Feature activation-2.11e-02

Pos logit contributions
ores+0.012
arium+0.012
ver+0.011
 named+0.011
year+0.011
Neg logit contributions
 buttons-0.010
 reins-0.010
 rules-0.010
 brakes-0.009
 wounds-0.009
Token was
Token ID373
Feature activation-1.54e-02

Pos logit contributions
ores+0.029
 named+0.028
arium+0.028
ver+0.028
year+0.027
Neg logit contributions
 wounds-0.036
 reins-0.034
 stret-0.034
 buttons-0.033
 rules-0.033
Token
Token ID198
Feature activation-1.77e-03

Pos logit contributions
ores+0.015
arium+0.014
 named+0.014
year+0.014
ver+0.014
Neg logit contributions
 wounds-0.025
 buttons-0.024
 reins-0.024
 stret-0.024
 rules-0.024
Token
Token ID198
Feature activation-1.85e-03

Pos logit contributions
ores+0.002
 named+0.002
ver+0.002
year+0.002
arium+0.002
Neg logit contributions
 wounds-0.004
 buttons-0.004
 mess-0.003
 reins-0.003
 brakes-0.003
TokenJohn
Token ID7554
Feature activation-6.77e-03

Pos logit contributions
year+0.003
 named+0.003
ores+0.003
ver+0.003
arium+0.003
Neg logit contributions
 wounds-0.003
 buttons-0.003
 mess-0.002
 blood-0.002
 ow-0.002
Token wondered
Token ID14028
Feature activation-6.79e-03

Pos logit contributions
 named+0.009
ores+0.009
ver+0.008
arium+0.008
year+0.008
Neg logit contributions
 wounds-0.012
 reins-0.011
 stret-0.011
 buttons-0.011
 rules-0.011
Token how
Token ID703
Feature activation+1.24e-02

Pos logit contributions
ores+0.008
arium+0.008
apped+0.008
year+0.008
 named+0.008
Neg logit contributions
 wounds-0.012
 buttons-0.012
 reins-0.012
 rules-0.012
 stret-0.012
Token old
Token ID1468
Feature activation-9.93e-02

Pos logit contributions
 wounds+0.023
 stret+0.022
 reins+0.022
 buttons+0.022
 mess+0.022
Neg logit contributions
ores-0.015
 named-0.015
ver-0.014
arium-0.014
year-0.014
Token this
Token ID428
Feature activation-2.54e-02

Pos logit contributions
ores+0.145
apped+0.144
year+0.143
ver+0.140
arium+0.138
Neg logit contributions
 buttons-0.158
 wounds-0.155
 rules-0.150
 brakes-0.149
 blood-0.144
Token place
Token ID1295
Feature activation-1.98e-02

Pos logit contributions
ores+0.035
ver+0.033
year+0.033
arium+0.033
 named+0.033
Neg logit contributions
 wounds-0.042
 buttons-0.042
 rules-0.041
 mess-0.040
 reins-0.040
Token was
Token ID373
Feature activation-1.21e-02

Pos logit contributions
ores+0.029
 named+0.027
arium+0.027
year+0.027
ver+0.027
Neg logit contributions
 wounds-0.031
 buttons-0.030
 rules-0.030
 reins-0.029
 mess-0.029
Token.
Token ID13
Feature activation-4.78e-03

Pos logit contributions
ores+0.017
arium+0.016
 named+0.016
year+0.016
ver+0.016
Neg logit contributions
 wounds-0.020
 buttons-0.019
 rules-0.019
 reins-0.019
 stret-0.019
Token It
Token ID632
Feature activation-1.86e-02

Pos logit contributions
ores+0.007
arium+0.007
 named+0.007
ver+0.006
year+0.006
Neg logit contributions
 wounds-0.008
 buttons-0.007
 rules-0.007
 reins-0.007
 brakes-0.007
Token.
Token ID13
Feature activation+4.36e-03

Pos logit contributions
ores+0.020
 named+0.020
arium+0.020
ver+0.019
year+0.019
Neg logit contributions
 wounds-0.024
 buttons-0.023
 reins-0.023
 stret-0.023
 rules-0.023
Token She
Token ID1375
Feature activation-2.62e-03

Pos logit contributions
 wounds+0.009
 reins+0.008
 stret+0.008
 rules+0.008
 buttons+0.008
Neg logit contributions
ores-0.005
 named-0.004
ver-0.004
arium-0.004
year-0.004
Token wrote
Token ID2630
Feature activation-3.18e-02

Pos logit contributions
ores+0.004
arium+0.003
ver+0.003
chard+0.003
year+0.003
Neg logit contributions
 buttons-0.005
 wounds-0.004
 mess-0.004
 reins-0.004
 stret-0.004
Token about
Token ID546
Feature activation-3.31e-02

Pos logit contributions
ores+0.041
arium+0.041
 named+0.039
ver+0.038
year+0.037
Neg logit contributions
 buttons-0.056
 wounds-0.056
 reins-0.054
 rules-0.053
 stret-0.053
Token how
Token ID703
Feature activation+1.45e-02

Pos logit contributions
ores+0.038
arium+0.036
ver+0.036
 named+0.036
year+0.035
Neg logit contributions
 wounds-0.063
 buttons-0.060
 reins-0.060
 rules-0.059
 stret-0.059
Token grand
Token ID4490
Feature activation-9.82e-02

Pos logit contributions
 buttons+0.023
 wounds+0.023
 rules+0.023
 reins+0.022
 mess+0.022
Neg logit contributions
arium-0.020
ver-0.020
ores-0.019
 named-0.019
apped-0.019
Tokenbaby
Token ID40252
Feature activation-5.79e-02

Pos logit contributions
ores+0.128
 named+0.125
arium+0.121
year+0.119
ver+0.119
Neg logit contributions
 wounds-0.174
 reins-0.166
 stret-0.165
 brakes-0.161
 buttons-0.161
Token is
Token ID318
Feature activation-2.46e-03

Pos logit contributions
ores+0.084
arium+0.082
year+0.078
apped+0.077
ver+0.077
Neg logit contributions
 buttons-0.091
 wounds-0.089
 mess-0.086
 rules-0.083
 brakes-0.083
Token always
Token ID1464
Feature activation-2.36e-02

Pos logit contributions
ores+0.004
arium+0.004
apped+0.004
year+0.004
 Finally+0.003
Neg logit contributions
 buttons-0.004
 mess-0.003
 broken-0.003
 rules-0.003
 sticking-0.003
Token smiling
Token ID16755
Feature activation+1.24e-02

Pos logit contributions
ores+0.038
arium+0.036
apped+0.035
year+0.035
ver+0.035
Neg logit contributions
 mess-0.034
 buttons-0.034
 wounds-0.032
 rules-0.032
 bleeding-0.031
Token,
Token ID11
Feature activation+7.17e-03

Pos logit contributions
 buttons+0.021
 wounds+0.021
 reins+0.020
 mess+0.020
 rules+0.020
Neg logit contributions
ores-0.016
arium-0.016
 named-0.016
year-0.015
ver-0.015
Token.
Token ID13
Feature activation-2.05e-03

Pos logit contributions
 wounds+0.001
 stret+0.001
 reins+0.001
 rules+0.001
 brakes+0.001
Neg logit contributions
ores-0.001
 named-0.001
ver-0.001
arium-0.001
apped-0.001
Token Tim
Token ID5045
Feature activation+9.89e-03

Pos logit contributions
ores+0.002
arium+0.002
ver+0.002
year+0.002
 named+0.002
Neg logit contributions
 wounds-0.004
 buttons-0.004
 stret-0.004
 rules-0.004
 brakes-0.004
Tokenmy
Token ID1820
Feature activation-4.79e-03

Pos logit contributions
 stret+0.021
 wounds+0.021
 rules+0.020
 reins+0.020
 brakes+0.020
Neg logit contributions
 named-0.010
ver-0.009
 lived-0.009
 known-0.008
arium-0.008
Token wondered
Token ID14028
Feature activation-1.15e-02

Pos logit contributions
ores+0.007
arium+0.006
ver+0.006
year+0.006
 named+0.006
Neg logit contributions
 wounds-0.008
 buttons-0.008
 reins-0.008
 stret-0.008
 mess-0.008
Token how
Token ID703
Feature activation+1.47e-02

Pos logit contributions
ores+0.016
arium+0.015
apped+0.015
year+0.015
 named+0.014
Neg logit contributions
 wounds-0.020
 buttons-0.019
 reins-0.018
 stret-0.018
 rules-0.018
Token old
Token ID1468
Feature activation-9.79e-02

Pos logit contributions
 wounds+0.026
 buttons+0.025
 reins+0.025
 rules+0.025
 stret+0.024
Neg logit contributions
ores-0.018
arium-0.018
 named-0.018
ver-0.018
year-0.017
Token it
Token ID340
Feature activation-2.97e-02

Pos logit contributions
year+0.155
ores+0.149
apped+0.146
arium+0.142
ver+0.141
Neg logit contributions
 buttons-0.157
 brakes-0.147
 wounds-0.145
 rules-0.142
 blood-0.138
Token was
Token ID373
Feature activation-1.01e-02

Pos logit contributions
ores+0.041
arium+0.039
year+0.039
 named+0.039
apped+0.038
Neg logit contributions
 wounds-0.049
 buttons-0.049
 mess-0.046
 brakes-0.046
 rules-0.046
Token.
Token ID13
Feature activation+1.54e-03

Pos logit contributions
ores+0.014
arium+0.014
 named+0.014
ver+0.014
year+0.014
Neg logit contributions
 wounds-0.016
 buttons-0.016
 reins-0.015
 rules-0.015
 stret-0.015
Token
Token ID220
Feature activation+9.25e-03

Pos logit contributions
 wounds+0.002
 buttons+0.002
 rules+0.002
 stret+0.002
 brakes+0.002
Neg logit contributions
ores-0.002
arium-0.002
ver-0.002
year-0.002
 named-0.002
Token
Token ID198
Feature activation+9.74e-03

Pos logit contributions
 wounds+0.018
 buttons+0.017
 reins+0.017
 stret+0.017
 rules+0.017
Neg logit contributions
ores-0.010
 named-0.010
ver-0.010
arium-0.010
year-0.009
Tokenez
Token ID8471
Feature activation-6.11e-03

Pos logit contributions
arium+0.006
chard+0.006
ver+0.006
ores+0.006
 Captain+0.006
Neg logit contributions
 wounds-0.007
 blood-0.007
 stret-0.007
 buttons-0.007
 thorn-0.007
Tokening
Token ID278
Feature activation+2.61e-02

Pos logit contributions
ores+0.009
ver+0.009
apped+0.009
 named+0.009
year+0.009
Neg logit contributions
 wounds-0.010
 reins-0.009
 stret-0.009
 rules-0.009
 buttons-0.009
Token!
Token ID0
Feature activation+1.00e-02

Pos logit contributions
 wounds+0.050
 reins+0.048
 brakes+0.047
 rules+0.047
 stret+0.046
Neg logit contributions
ores-0.030
 named-0.029
ver-0.026
year-0.026
apped-0.025
Token The
Token ID383
Feature activation-8.45e-03

Pos logit contributions
 wounds+0.018
 buttons+0.017
 stret+0.017
 rules+0.017
 reins+0.017
Neg logit contributions
ores-0.012
arium-0.012
ver-0.012
year-0.012
 named-0.011
Token sprinkle
Token ID45799
Feature activation-1.51e-02

Pos logit contributions
ores+0.012
 named+0.012
chard+0.011
 lived+0.011
 surely+0.010
Neg logit contributions
 wounds-0.014
 pounds-0.013
 brakes-0.013
 reins-0.013
 stret-0.013
Token-
Token ID12
Feature activation-9.79e-02

Pos logit contributions
ores+0.019
 named+0.018
ver+0.018
apped+0.017
year+0.017
Neg logit contributions
 wounds-0.028
 reins-0.027
 stret-0.026
 rules-0.026
 brakes-0.026
Tokencovered
Token ID32111
Feature activation-1.67e-02

Pos logit contributions
arium+0.168
ores+0.167
 named+0.165
ver+0.163
apped+0.158
Neg logit contributions
 wounds-0.127
 buttons-0.123
 reins-0.121
 stret-0.118
 rules-0.118
Token pastry
Token ID41738
Feature activation-2.03e-02

Pos logit contributions
ores+0.027
 named+0.027
arium+0.026
ver+0.026
apped+0.026
Neg logit contributions
 wounds-0.023
 stret-0.022
 reins-0.021
 buttons-0.021
 brakes-0.021
Token flew
Token ID13112
Feature activation-1.97e-02

Pos logit contributions
ores+0.022
 named+0.022
 surely+0.021
ver+0.021
apped+0.020
Neg logit contributions
 wounds-0.040
 reins-0.038
 rules-0.038
 stret-0.038
 brakes-0.036
Token out
Token ID503
Feature activation-2.97e-03

Pos logit contributions
ores+0.025
 named+0.024
arium+0.024
ver+0.024
year+0.023
Neg logit contributions
 wounds-0.035
 buttons-0.033
 reins-0.033
 stret-0.033
 rules-0.033
Token of
Token ID286
Feature activation+3.73e-03

Pos logit contributions
ores+0.004
 named+0.004
year+0.003
arium+0.003
ver+0.003
Neg logit contributions
 wounds-0.006
 stret-0.005
 reins-0.005
 brakes-0.005
 rules-0.005
Token touched
Token ID12615
Feature activation-1.95e-02

Pos logit contributions
 wounds+0.020
 rules+0.019
 stret+0.019
 reins+0.019
 brakes+0.018
Neg logit contributions
 named-0.013
ver-0.012
arium-0.012
 lived-0.012
ores-0.012
Token it
Token ID340
Feature activation-2.06e-02

Pos logit contributions
arium+0.028
ores+0.028
ver+0.027
year+0.026
apped+0.025
Neg logit contributions
 buttons-0.032
 wounds-0.031
 blood-0.031
 thorn-0.031
 reins-0.030
Token and
Token ID290
Feature activation-8.68e-03

Pos logit contributions
ores+0.025
 named+0.024
ver+0.024
arium+0.024
apped+0.023
Neg logit contributions
 wounds-0.038
 stret-0.036
 reins-0.036
 rules-0.035
 buttons-0.035
Token wondered
Token ID14028
Feature activation-4.52e-03

Pos logit contributions
ores+0.013
year+0.013
arium+0.013
ver+0.012
apped+0.012
Neg logit contributions
 buttons-0.014
 wounds-0.014
 mess-0.013
 blood-0.013
 brakes-0.012
Token how
Token ID703
Feature activation+1.30e-02

Pos logit contributions
ores+0.006
arium+0.006
apped+0.006
year+0.005
 named+0.005
Neg logit contributions
 wounds-0.008
 buttons-0.008
 rules-0.007
 reins-0.007
 pounds-0.007
Token old
Token ID1468
Feature activation-9.76e-02

Pos logit contributions
 wounds+0.022
 buttons+0.021
 reins+0.021
 rules+0.021
 stret+0.021
Neg logit contributions
ores-0.017
arium-0.017
 named-0.017
ver-0.017
year-0.016
Token it
Token ID340
Feature activation-2.73e-02

Pos logit contributions
year+0.157
ores+0.154
apped+0.148
ver+0.147
arium+0.145
Neg logit contributions
 buttons-0.149
 wounds-0.143
 rules-0.140
 blood-0.133
 brakes-0.133
Token was
Token ID373
Feature activation-9.27e-03

Pos logit contributions
ores+0.037
arium+0.035
 named+0.035
year+0.035
ver+0.034
Neg logit contributions
 wounds-0.047
 buttons-0.046
 mess-0.044
 rules-0.043
 brakes-0.043
Token.
Token ID13
Feature activation-1.27e-03

Pos logit contributions
ores+0.013
 named+0.012
arium+0.012
ver+0.012
year+0.012
Neg logit contributions
 wounds-0.016
 reins-0.015
 stret-0.015
 buttons-0.015
 rules-0.014
Token
Token ID220
Feature activation+7.50e-03

Pos logit contributions
ores+0.001
arium+0.001
ver+0.001
year+0.001
 named+0.001
Neg logit contributions
 wounds-0.003
 buttons-0.002
 reins-0.002
 rules-0.002
 stret-0.002
Token
Token ID198
Feature activation+7.14e-03

Pos logit contributions
 wounds+0.014
 buttons+0.014
 reins+0.014
 stret+0.014
 rules+0.014
Neg logit contributions
ores-0.008
 named-0.008
arium-0.008
ver-0.008
year-0.008
Token old
Token ID1468
Feature activation-6.97e-02

Pos logit contributions
arium+0.035
ver+0.034
 named+0.033
ores+0.033
apped+0.032
Neg logit contributions
 buttons-0.021
 wounds-0.021
 mess-0.021
 rules-0.020
 brakes-0.019
Token,
Token ID11
Feature activation-9.90e-03

Pos logit contributions
ores+0.094
ver+0.089
arium+0.089
 named+0.089
year+0.088
Neg logit contributions
 wounds-0.118
 buttons-0.114
 stret-0.112
 reins-0.112
 rules-0.111
Token nobody
Token ID8168
Feature activation+1.15e-02

Pos logit contributions
ores+0.012
 named+0.012
arium+0.011
ver+0.011
year+0.011
Neg logit contributions
 wounds-0.018
 reins-0.017
 buttons-0.017
 stret-0.017
 rules-0.017
Token knew
Token ID2993
Feature activation+1.20e-02

Pos logit contributions
 wounds+0.021
 buttons+0.020
 reins+0.020
 stret+0.020
 rules+0.020
Neg logit contributions
ores-0.014
 named-0.013
arium-0.013
ver-0.013
year-0.013
Token how
Token ID703
Feature activation+1.35e-02

Pos logit contributions
 wounds+0.023
 stret+0.022
 reins+0.022
 buttons+0.022
 rules+0.022
Neg logit contributions
ores-0.014
 named-0.013
ver-0.013
arium-0.013
year-0.012
Token old
Token ID1468
Feature activation-9.67e-02

Pos logit contributions
 wounds+0.024
 stret+0.023
 reins+0.023
 buttons+0.022
 brakes+0.022
Neg logit contributions
ores-0.018
 named-0.017
year-0.016
arium-0.016
ver-0.016
Token it
Token ID340
Feature activation-3.51e-02

Pos logit contributions
ores+0.136
year+0.131
ver+0.131
arium+0.129
apped+0.129
Neg logit contributions
 wounds-0.158
 buttons-0.158
 rules-0.152
 mess-0.151
 brakes-0.148
Token was
Token ID373
Feature activation-1.59e-02

Pos logit contributions
ores+0.050
arium+0.047
 named+0.047
year+0.046
ver+0.046
Neg logit contributions
 wounds-0.057
 buttons-0.056
 mess-0.054
 reins-0.054
 rules-0.054
Token.
Token ID13
Feature activation-1.56e-02

Pos logit contributions
ores+0.022
 named+0.021
arium+0.021
ver+0.021
year+0.021
Neg logit contributions
 wounds-0.026
 buttons-0.025
 reins-0.025
 stret-0.025
 rules-0.025
Token Everyone
Token ID11075
Feature activation+9.09e-03

Pos logit contributions
ores+0.019
arium+0.018
ver+0.018
year+0.018
working+0.017
Neg logit contributions
 wounds-0.027
 buttons-0.026
 reins-0.026
 rules-0.026
 stret-0.026
Token who
Token ID508
Feature activation-1.56e-02

Pos logit contributions
 wounds+0.015
 buttons+0.014
 mess+0.014
 rules+0.014
 stret+0.014
Neg logit contributions
ores-0.013
arium-0.012
ver-0.012
chard-0.012
 named-0.012
TokenHe
Token ID1544
Feature activation-6.69e-03

Pos logit contributions
 wounds+0.015
 buttons+0.014
 stret+0.014
 reins+0.013
 mess+0.013
Neg logit contributions
 named-0.016
ores-0.016
ver-0.016
year-0.016
arium-0.016
Token exclaimed
Token ID33552
Feature activation-1.35e-02

Pos logit contributions
ores+0.010
ver+0.009
arium+0.009
year+0.009
apped+0.009
Neg logit contributions
 wounds-0.011
 buttons-0.010
 mess-0.010
 reins-0.010
 stret-0.010
Token,
Token ID11
Feature activation-9.68e-03

Pos logit contributions
ores+0.016
arium+0.016
 known+0.014
chard+0.014
ver+0.014
Neg logit contributions
 rules-0.024
 wounds-0.024
 buttons-0.023
 faces-0.023
 reins-0.023
Token "
Token ID366
Feature activation+4.49e-03

Pos logit contributions
arium+0.015
ores+0.015
ver+0.015
year+0.015
 known+0.014
Neg logit contributions
 wounds-0.013
 stret-0.013
 reins-0.013
 buttons-0.013
 brakes-0.013
TokenPick
Token ID31686
Feature activation-1.96e-02

Pos logit contributions
 wounds+0.008
 mess+0.007
 stret+0.007
 rules+0.007
 brakes+0.007
Neg logit contributions
 known-0.007
 named-0.007
year-0.006
ores-0.006
 surely-0.006
Token-
Token ID12
Feature activation-9.50e-02

Pos logit contributions
ores+0.028
 named+0.027
arium+0.027
ver+0.027
year+0.026
Neg logit contributions
 wounds-0.031
 reins-0.030
 buttons-0.030
 stret-0.030
 rules-0.029
Tokena
Token ID64
Feature activation-5.08e-02

Pos logit contributions
arium+0.149
ores+0.148
ver+0.144
 named+0.140
apped+0.139
Neg logit contributions
 buttons-0.137
 wounds-0.136
 reins-0.136
 rules-0.131
 blood-0.126
Token-
Token ID12
Feature activation-6.92e-02

Pos logit contributions
ores+0.060
arium+0.057
ver+0.056
year+0.056
 named+0.056
Neg logit contributions
 wounds-0.094
 buttons-0.091
 reins-0.090
 rules-0.089
 stret-0.089
Tokenbo
Token ID2127
Feature activation-4.27e-02

Pos logit contributions
ores+0.109
arium+0.107
ver+0.105
 named+0.105
apped+0.102
Neg logit contributions
 wounds-0.101
 buttons-0.097
 reins-0.097
 rules-0.094
 stret-0.093
Tokeno
Token ID78
Feature activation-3.86e-02

Pos logit contributions
ores+0.059
chard+0.056
arium+0.055
 Captain+0.052
 surely+0.052
Neg logit contributions
 reins-0.069
 pounds-0.068
 buttons-0.067
 rules-0.067
 wounds-0.065
Token!
Token ID0
Feature activation+1.68e-02

Pos logit contributions
ores+0.050
arium+0.048
ver+0.047
year+0.047
 named+0.047
Neg logit contributions
 wounds-0.067
 buttons-0.064
 reins-0.064
 rules-0.063
 stret-0.063
Token been
Token ID587
Feature activation+7.38e-03

Pos logit contributions
ores+0.055
year+0.052
ver+0.052
arium+0.051
apped+0.050
Neg logit contributions
 mess-0.067
 wounds-0.067
 buttons-0.065
 rules-0.064
 blood-0.063
Token?"
Token ID1701
Feature activation-4.46e-02

Pos logit contributions
 wounds+0.012
 mess+0.012
 stret+0.012
 buttons+0.012
 rules+0.012
Neg logit contributions
ores-0.010
arium-0.010
ver-0.010
year-0.010
apped-0.009
Token
Token ID198
Feature activation-1.24e-02

Pos logit contributions
ores+0.061
arium+0.059
year+0.058
ver+0.056
 surely+0.052
Neg logit contributions
 wounds-0.075
 stret-0.073
 blood-0.071
 brakes-0.071
 buttons-0.071
Token
Token ID198
Feature activation-8.80e-03

Pos logit contributions
ores+0.014
 named+0.013
ver+0.013
arium+0.013
year+0.013
Neg logit contributions
 wounds-0.024
 buttons-0.023
 reins-0.023
 stret-0.023
 rules-0.023
TokenPick
Token ID31686
Feature activation-5.09e-03

Pos logit contributions
 named+0.015
ores+0.014
ver+0.014
 known+0.014
year+0.014
Neg logit contributions
 wounds-0.013
 buttons-0.012
 mess-0.012
 reins-0.012
 brakes-0.012
Token-
Token ID12
Feature activation-9.04e-02

Pos logit contributions
ores+0.008
arium+0.008
 named+0.008
ver+0.008
year+0.008
Neg logit contributions
 wounds-0.007
 buttons-0.007
 reins-0.007
 stret-0.007
 rules-0.007
Tokena
Token ID64
Feature activation-5.10e-02

Pos logit contributions
ores+0.149
arium+0.144
ver+0.140
 named+0.136
apped+0.136
Neg logit contributions
 wounds-0.125
 buttons-0.122
 reins-0.120
 rules-0.117
co-0.114
Token-
Token ID12
Feature activation-6.53e-02

Pos logit contributions
ores+0.081
arium+0.076
ver+0.075
year+0.074
 named+0.074
Neg logit contributions
 wounds-0.077
 buttons-0.073
 reins-0.071
 stret-0.070
 rules-0.070
Tokenbo
Token ID2127
Feature activation-3.35e-02

Pos logit contributions
ores+0.103
arium+0.099
ver+0.098
 named+0.097
apped+0.095
Neg logit contributions
 wounds-0.097
 buttons-0.093
 reins-0.092
 rules-0.090
 stret-0.089
Tokeno
Token ID78
Feature activation-2.76e-02

Pos logit contributions
ores+0.048
chard+0.046
arium+0.044
 surely+0.043
 Captain+0.042
Neg logit contributions
 wounds-0.052
 buttons-0.051
co-0.050
 blood-0.050
 mess-0.049
Token smiled
Token ID13541
Feature activation-1.93e-02

Pos logit contributions
ores+0.040
arium+0.037
ver+0.036
year+0.036
 named+0.035
Neg logit contributions
 wounds-0.046
 buttons-0.044
 reins-0.042
 stret-0.042
 mess-0.042
Token forest
Token ID8222
Feature activation+8.56e-03

Pos logit contributions
 wounds+0.027
 buttons+0.026
 reins+0.026
 stret+0.026
 rules+0.025
Neg logit contributions
ores-0.025
ver-0.024
 named-0.023
arium-0.023
year-0.023
Token didn
Token ID1422
Feature activation-3.02e-03

Pos logit contributions
 wounds+0.014
 mess+0.013
 rules+0.013
 buttons+0.013
 faces+0.012
Neg logit contributions
ores-0.013
year-0.012
chard-0.011
 Captain-0.011
 Enjoy-0.011
Token't
Token ID470
Feature activation+7.77e-03

Pos logit contributions
ores+0.004
ver+0.004
 surely+0.003
 named+0.003
 Captain+0.003
Neg logit contributions
 wounds-0.006
 reins-0.006
 buttons-0.005
 stret-0.005
 pounds-0.005
Token like
Token ID588
Feature activation-1.46e-02

Pos logit contributions
 wounds+0.016
 buttons+0.015
 reins+0.015
 stret+0.015
 brakes+0.015
Neg logit contributions
ores-0.008
 named-0.008
ver-0.007
arium-0.007
 known-0.007
Token how
Token ID703
Feature activation+1.95e-02

Pos logit contributions
 named+0.015
ores+0.015
 Captain+0.014
arium+0.014
 known+0.013
Neg logit contributions
 reins-0.029
 wounds-0.029
 stret-0.028
 buttons-0.028
 brakes-0.027
Token boss
Token ID6478
Feature activation-9.02e-02

Pos logit contributions
 wounds+0.035
 buttons+0.034
 reins+0.033
 stret+0.033
 rules+0.033
Neg logit contributions
ores-0.024
 named-0.023
arium-0.023
ver-0.023
year-0.023
Tokeny
Token ID88
Feature activation-4.20e-02

Pos logit contributions
ores+0.138
arium+0.137
year+0.129
 named+0.127
ver+0.125
Neg logit contributions
 buttons-0.132
 wounds-0.128
 blood-0.125
 mess-0.124
 brakes-0.124
Token Billy
Token ID15890
Feature activation-2.52e-02

Pos logit contributions
ores+0.048
arium+0.046
 named+0.046
ver+0.045
year+0.044
Neg logit contributions
 wounds-0.080
 buttons-0.077
 reins-0.076
 stret-0.076
 rules-0.075
Token was
Token ID373
Feature activation+1.17e-02

Pos logit contributions
 named+0.031
ver+0.030
 lived+0.029
ores+0.029
arium+0.029
Neg logit contributions
 wounds-0.047
 reins-0.046
 stret-0.044
 buttons-0.044
 rules-0.044
Token.
Token ID13
Feature activation+4.01e-03

Pos logit contributions
 wounds+0.022
 buttons+0.021
 reins+0.021
 stret+0.021
 rules+0.021
Neg logit contributions
ores-0.014
 named-0.013
arium-0.013
ver-0.013
year-0.013
Token They
Token ID1119
Feature activation-5.99e-03

Pos logit contributions
 wounds+0.007
 rules+0.007
 buttons+0.007
 brakes+0.007
 stret+0.006
Neg logit contributions
ores-0.005
year-0.005
arium-0.005
ver-0.005
chard-0.005
Token so
Token ID523
Feature activation-1.34e-02

Pos logit contributions
ores+0.007
arium+0.007
ver+0.006
 named+0.006
year+0.006
Neg logit contributions
 wounds-0.007
 buttons-0.007
 reins-0.007
 mess-0.007
 stret-0.007
Token proud
Token ID6613
Feature activation-1.85e-02

Pos logit contributions
arium+0.018
ores+0.017
ver+0.016
 named+0.015
apped+0.014
Neg logit contributions
 wounds-0.023
 buttons-0.023
 mess-0.022
 rules-0.022
 reins-0.022
Token of
Token ID286
Feature activation-2.29e-02

Pos logit contributions
 named+0.021
ores+0.021
arium+0.020
 Captain+0.019
year+0.019
Neg logit contributions
 wounds-0.037
 stret-0.036
 reins-0.036
 brakes-0.035
 buttons-0.033
Token his
Token ID465
Feature activation-1.46e-02

Pos logit contributions
ores+0.038
ver+0.038
arium+0.037
apped+0.036
 named+0.036
Neg logit contributions
 wounds-0.032
 rules-0.030
 buttons-0.030
 reins-0.029
 brakes-0.028
Token brand
Token ID4508
Feature activation-1.68e-02

Pos logit contributions
ores+0.024
 named+0.023
arium+0.023
ver+0.023
year+0.022
Neg logit contributions
 wounds-0.020
 stret-0.019
 reins-0.019
 buttons-0.019
 brakes-0.019
Token-
Token ID12
Feature activation-9.01e-02

Pos logit contributions
ores+0.034
 named+0.033
arium+0.033
ver+0.033
year+0.033
Neg logit contributions
 wounds-0.017
 stret-0.016
 reins-0.016
 buttons-0.015
 brakes-0.015
Tokennew
Token ID3605
Feature activation-7.11e-03

Pos logit contributions
arium+0.153
ores+0.151
apped+0.145
ver+0.142
 named+0.140
Neg logit contributions
 buttons-0.115
 wounds-0.112
 reins-0.112
 rules-0.111
we-0.104
Token suit
Token ID6050
Feature activation-3.33e-02

Pos logit contributions
 named+0.012
arium+0.011
ores+0.011
 Captain+0.011
 lived+0.011
Neg logit contributions
 wounds-0.010
 stret-0.009
 reins-0.009
 brakes-0.009
 buttons-0.009
Token.
Token ID13
Feature activation-1.32e-02

Pos logit contributions
ores+0.044
 named+0.043
ver+0.043
arium+0.042
year+0.041
Neg logit contributions
 wounds-0.058
 reins-0.055
 stret-0.054
 buttons-0.053
 rules-0.053
Token
Token ID198
Feature activation-2.14e-03

Pos logit contributions
ores+0.017
arium+0.016
ver+0.016
year+0.016
 named+0.016
Neg logit contributions
 wounds-0.022
 reins-0.022
 buttons-0.022
 rules-0.021
 stret-0.021
Token
Token ID198
Feature activation-1.96e-03

Pos logit contributions
ores+0.002
 named+0.002
arium+0.002
ver+0.002
year+0.002
Neg logit contributions
 wounds-0.004
 buttons-0.004
 reins-0.004
 stret-0.004
 rules-0.004
Token can
Token ID460
Feature activation+2.28e-02

Pos logit contributions
 buttons+0.036
 wounds+0.036
 mess+0.035
 rules+0.035
 brakes+0.035
Neg logit contributions
ores-0.030
arium-0.030
year-0.030
chard-0.028
apped-0.028
Token also
Token ID635
Feature activation-1.04e-02

Pos logit contributions
 wounds+0.042
 reins+0.041
 pounds+0.040
 stret+0.040
 rules+0.039
Neg logit contributions
 named-0.028
ores-0.027
ver-0.026
 surely-0.026
 known-0.025
Token learn
Token ID2193
Feature activation+1.76e-02

Pos logit contributions
arium+0.016
ores+0.016
year+0.016
ver+0.016
apped+0.016
Neg logit contributions
 wounds-0.016
 buttons-0.015
 stret-0.015
 rules-0.015
 mess-0.015
Token about
Token ID546
Feature activation-1.65e-02

Pos logit contributions
 wounds+0.022
 reins+0.021
 stret+0.020
 buttons+0.020
 brakes+0.020
Neg logit contributions
ores-0.032
 named-0.031
arium-0.031
ver-0.031
year-0.030
Token how
Token ID703
Feature activation+1.62e-02

Pos logit contributions
 named+0.019
ores+0.019
arium+0.018
 Captain+0.018
 known+0.018
Neg logit contributions
 reins-0.030
 stret-0.030
 wounds-0.029
 brakes-0.028
 buttons-0.028
Token wheels
Token ID13666
Feature activation-8.99e-02

Pos logit contributions
 wounds+0.027
 stret+0.026
 reins+0.026
 buttons+0.026
 brakes+0.025
Neg logit contributions
ores-0.022
 named-0.021
ver-0.021
arium-0.021
year-0.021
Token work
Token ID670
Feature activation-1.10e-02

Pos logit contributions
ores+0.115
 named+0.111
ver+0.111
arium+0.109
year+0.105
Neg logit contributions
 wounds-0.158
 reins-0.151
 stret-0.151
 buttons-0.147
 rules-0.146
Token.
Token ID13
Feature activation-1.94e-03

Pos logit contributions
ores+0.015
 named+0.015
arium+0.015
year+0.015
chard+0.015
Neg logit contributions
 buttons-0.018
 rules-0.018
 wounds-0.017
 mess-0.017
 stret-0.017
Token She
Token ID1375
Feature activation-6.79e-03

Pos logit contributions
arium+0.003
year+0.003
ores+0.003
ver+0.003
 named+0.003
Neg logit contributions
 wounds-0.003
 buttons-0.003
 rules-0.003
 blood-0.003
 reins-0.003
Token asks
Token ID7893
Feature activation-1.42e-02

Pos logit contributions
ores+0.009
 named+0.009
arium+0.009
ver+0.009
apped+0.009
Neg logit contributions
 wounds-0.011
 stret-0.011
 reins-0.011
 rules-0.010
 buttons-0.010
Token them
Token ID606
Feature activation-2.03e-03

Pos logit contributions
arium+0.023
ores+0.023
ver+0.023
 known+0.022
year+0.022
Neg logit contributions
 wounds-0.020
 buttons-0.020
 rules-0.020
 blood-0.018
 faces-0.018
Token witch
Token ID16365
Feature activation+8.42e-03

Pos logit contributions
ores+0.013
ver+0.013
arium+0.013
 named+0.013
year+0.012
Neg logit contributions
 wounds-0.008
 buttons-0.007
 reins-0.007
 rules-0.007
 stret-0.007
Token was
Token ID373
Feature activation-2.18e-03

Pos logit contributions
 wounds+0.014
 buttons+0.013
 stret+0.013
 mess+0.013
 reins+0.013
Neg logit contributions
ores-0.012
arium-0.011
ver-0.011
year-0.011
 named-0.011
Token about
Token ID546
Feature activation-5.73e-03

Pos logit contributions
ores+0.003
arium+0.003
ver+0.003
 named+0.003
year+0.003
Neg logit contributions
 wounds-0.004
 buttons-0.004
 stret-0.004
 reins-0.004
 rules-0.004
Token to
Token ID284
Feature activation+1.34e-02

Pos logit contributions
ores+0.009
ver+0.009
arium+0.009
 named+0.009
year+0.008
Neg logit contributions
 wounds-0.009
 buttons-0.008
 rules-0.008
 reins-0.008
 stret-0.008
Token give
Token ID1577
Feature activation-1.66e-02

Pos logit contributions
 wounds+0.021
 buttons+0.020
 stret+0.020
 rules+0.020
 mess+0.019
Neg logit contributions
ores-0.020
arium-0.020
year-0.019
ver-0.019
apped-0.019
Token-
Token ID12
Feature activation-8.99e-02

Pos logit contributions
ores+0.024
arium+0.023
 named+0.023
year+0.022
ver+0.022
Neg logit contributions
 wounds-0.028
 buttons-0.027
 rules-0.026
 reins-0.025
 mess-0.025
Tokenup
Token ID929
Feature activation-2.28e-02

Pos logit contributions
arium+0.158
ores+0.157
 named+0.155
ver+0.151
apped+0.149
Neg logit contributions
 wounds-0.112
 buttons-0.109
 rules-0.107
 reins-0.106
 blood-0.099
Token when
Token ID618
Feature activation+2.73e-03

Pos logit contributions
ores+0.027
 named+0.026
arium+0.026
ver+0.025
year+0.025
Neg logit contributions
 wounds-0.043
 buttons-0.041
 reins-0.041
 stret-0.040
 rules-0.040
Token a
Token ID257
Feature activation-1.38e-02

Pos logit contributions
 wounds+0.005
 buttons+0.005
 rules+0.004
 blood+0.004
 mess+0.004
Neg logit contributions
ores-0.004
arium-0.004
ver-0.004
year-0.003
apped-0.003
Token little
Token ID1310
Feature activation-1.22e-02

Pos logit contributions
ores+0.024
 named+0.024
arium+0.024
ver+0.024
year+0.023
Neg logit contributions
 wounds-0.018
 reins-0.016
 buttons-0.016
 stret-0.016
 rules-0.016
Token rabbit
Token ID22746
Feature activation+1.39e-02

Pos logit contributions
ores+0.023
ver+0.022
arium+0.022
 named+0.022
year+0.022
Neg logit contributions
 wounds-0.014
 buttons-0.013
 reins-0.013
 stret-0.013
 rules-0.013
Token
Token ID198
Feature activation+7.53e-03

Pos logit contributions
 wounds+0.015
 buttons+0.014
 reins+0.014
 rules+0.014
 stret+0.014
Neg logit contributions
ores-0.010
 named-0.010
arium-0.010
year-0.010
apped-0.010
TokenCharlie
Token ID37136
Feature activation+1.48e-02

Pos logit contributions
 wounds+0.009
 buttons+0.009
 reins+0.009
 stret+0.009
 rules+0.009
Neg logit contributions
ores-0.014
 named-0.013
arium-0.013
ver-0.013
year-0.013
Token felt
Token ID2936
Feature activation-2.76e-03

Pos logit contributions
 wounds+0.022
 buttons+0.022
 mess+0.021
 brakes+0.020
 reins+0.020
Neg logit contributions
ores-0.024
arium-0.022
chard-0.022
ver-0.022
year-0.022
Token so
Token ID523
Feature activation-1.29e-02

Pos logit contributions
arium+0.005
ver+0.005
working+0.005
riage+0.005
apped+0.005
Neg logit contributions
 wounds-0.003
 buttons-0.003
 mess-0.003
 sharp-0.003
 tears-0.003
Token happy
Token ID3772
Feature activation-2.52e-02

Pos logit contributions
arium+0.022
chard+0.020
ver+0.020
 named+0.019
ores+0.019
Neg logit contributions
 mess-0.018
 wounds-0.017
 faces-0.017
 rip-0.017
 rules-0.017
Token-
Token ID12
Feature activation-8.96e-02

Pos logit contributions
ores+0.032
 named+0.031
arium+0.030
ver+0.030
year+0.029
Neg logit contributions
 wounds-0.045
 reins-0.044
 stret-0.044
 buttons-0.043
 brakes-0.042
Token he
Token ID339
Feature activation-7.84e-03

Pos logit contributions
arium+0.168
ores+0.159
 named+0.154
apped+0.150
 Captain+0.147
Neg logit contributions
co-0.097
we-0.095
 rules-0.093
 buttons-0.091
 pounds-0.088
Token knew
Token ID2993
Feature activation+8.17e-03

Pos logit contributions
ores+0.010
arium+0.010
ver+0.009
 named+0.009
year+0.009
Neg logit contributions
 wounds-0.014
 buttons-0.013
 mess-0.013
 reins-0.013
 stret-0.013
Token he
Token ID339
Feature activation+2.87e-03

Pos logit contributions
 wounds+0.015
 buttons+0.014
 rules+0.014
 reins+0.014
 stret+0.014
Neg logit contributions
ores-0.010
arium-0.010
 named-0.009
ver-0.009
year-0.009
Token could
Token ID714
Feature activation-8.39e-03

Pos logit contributions
 wounds+0.005
 buttons+0.005
 mess+0.005
 reins+0.004
 brakes+0.004
Neg logit contributions
ores-0.004
arium-0.004
apped-0.004
 named-0.004
ver-0.004
Token play
Token ID711
Feature activation-1.21e-02

Pos logit contributions
ores+0.012
arium+0.011
 named+0.011
ver+0.011
year+0.011
Neg logit contributions
 wounds-0.014
 buttons-0.013
 reins-0.013
 rules-0.013
 stret-0.013
Token.
Token ID13
Feature activation-9.50e-03

Pos logit contributions
 wounds+0.008
 buttons+0.008
 rules+0.008
 mess+0.008
 stret+0.008
Neg logit contributions
ores-0.008
arium-0.008
 named-0.008
year-0.008
ver-0.008
Token He
Token ID679
Feature activation-1.71e-02

Pos logit contributions
ores+0.015
 named+0.014
arium+0.014
ver+0.014
chard+0.014
Neg logit contributions
 wounds-0.014
 buttons-0.014
 rules-0.014
 reins-0.013
 brakes-0.013
Token put
Token ID1234
Feature activation-2.31e-02

Pos logit contributions
ores+0.022
arium+0.021
ver+0.020
 named+0.020
year+0.020
Neg logit contributions
 wounds-0.031
 buttons-0.030
 reins-0.029
 mess-0.029
 stret-0.028
Token on
Token ID319
Feature activation-3.82e-02

Pos logit contributions
arium+0.033
ores+0.033
ver+0.032
 named+0.032
year+0.031
Neg logit contributions
 wounds-0.038
 buttons-0.038
 reins-0.036
 blood-0.035
 brakes-0.035
Token extra
Token ID3131
Feature activation-3.70e-02

Pos logit contributions
ores+0.054
ver+0.053
arium+0.052
 named+0.052
year+0.051
Neg logit contributions
 wounds-0.064
 buttons-0.062
 reins-0.060
 rules-0.060
 stret-0.059
Token-
Token ID12
Feature activation-8.92e-02

Pos logit contributions
ver+0.064
ores+0.064
chard+0.062
 named+0.062
arium+0.062
Neg logit contributions
 wounds-0.050
 buttons-0.049
 rules-0.047
 reins-0.046
 mess-0.045
Tokenspecial
Token ID20887
Feature activation-1.40e-02

Pos logit contributions
arium+0.179
ores+0.173
 named+0.173
chard+0.171
ver+0.168
Neg logit contributions
 buttons-0.092
 wounds-0.090
 rules-0.088
 reins-0.087
co-0.079
Token show
Token ID905
Feature activation-6.73e-03

Pos logit contributions
ores+0.023
ver+0.023
arium+0.022
 named+0.022
year+0.022
Neg logit contributions
 wounds-0.020
 buttons-0.019
 reins-0.019
 rules-0.018
 stret-0.018
Token for
Token ID329
Feature activation-2.44e-02

Pos logit contributions
ores+0.009
arium+0.009
 named+0.009
ver+0.009
year+0.008
Neg logit contributions
 wounds-0.011
 buttons-0.011
 rules-0.011
 reins-0.011
 stret-0.011
Token his
Token ID465
Feature activation+5.09e-03

Pos logit contributions
ores+0.031
arium+0.030
ver+0.030
 named+0.030
apped+0.029
Neg logit contributions
 wounds-0.044
 buttons-0.042
 rules-0.042
 reins-0.041
 stret-0.041
Token family
Token ID1641
Feature activation+1.84e-03

Pos logit contributions
 wounds+0.008
 reins+0.007
 stret+0.007
 buttons+0.007
 rules+0.007
Neg logit contributions
ores-0.008
 named-0.007
arium-0.007
ver-0.007
year-0.007
Token where
Token ID810
Feature activation+2.70e-02

Pos logit contributions
ores+0.002
arium+0.002
ver+0.002
 named+0.002
year+0.002
Neg logit contributions
 wounds-0.002
 buttons-0.002
 reins-0.002
 stret-0.002
 rules-0.002
Token she
Token ID673
Feature activation-4.11e-03

Pos logit contributions
 wounds+0.045
 buttons+0.045
 rules+0.043
 mess+0.043
 blood+0.042
Neg logit contributions
ores-0.035
arium-0.035
 named-0.034
ver-0.034
apped-0.033
Token could
Token ID714
Feature activation-1.63e-02

Pos logit contributions
ores+0.005
arium+0.005
ver+0.005
 named+0.005
year+0.005
Neg logit contributions
 wounds-0.007
 buttons-0.007
 stret-0.007
 reins-0.007
 mess-0.007
Token still
Token ID991
Feature activation-3.31e-02

Pos logit contributions
 named+0.022
ores+0.022
ver+0.021
arium+0.020
 surely+0.020
Neg logit contributions
 wounds-0.028
 reins-0.027
 pounds-0.027
 buttons-0.027
 stret-0.026
Token show
Token ID905
Feature activation+1.16e-03

Pos logit contributions
ores+0.045
 named+0.044
arium+0.042
ver+0.042
year+0.041
Neg logit contributions
 wounds-0.056
 reins-0.054
 buttons-0.053
 stret-0.053
 rules-0.052
Token-
Token ID12
Feature activation-8.90e-02

Pos logit contributions
 wounds+0.002
 reins+0.002
 stret+0.002
 buttons+0.002
 rules+0.002
Neg logit contributions
ores-0.002
 named-0.002
arium-0.002
ver-0.002
year-0.001
Tokenoff
Token ID2364
Feature activation+8.57e-03

Pos logit contributions
ores+0.195
arium+0.193
 named+0.191
ver+0.189
apped+0.186
Neg logit contributions
 wounds-0.074
 buttons-0.070
 rules-0.068
 reins-0.067
 stret-0.064
Token her
Token ID607
Feature activation-1.10e-02

Pos logit contributions
 wounds+0.014
 stret+0.013
 reins+0.013
 brakes+0.013
 buttons+0.013
Neg logit contributions
ores-0.012
 named-0.012
ver-0.012
arium-0.012
year-0.012
Token fashion
Token ID6977
Feature activation-1.48e-02

Pos logit contributions
ores+0.019
 named+0.019
arium+0.018
ver+0.018
year+0.018
Neg logit contributions
 wounds-0.015
 reins-0.014
 stret-0.014
 buttons-0.013
 brakes-0.013
Token.
Token ID13
Feature activation-3.74e-03

Pos logit contributions
ores+0.023
 named+0.022
arium+0.022
ver+0.022
year+0.021
Neg logit contributions
 wounds-0.022
 reins-0.021
 buttons-0.021
 stret-0.021
 rules-0.021
Token
Token ID198
Feature activation+4.21e-03

Pos logit contributions
ores+0.005
arium+0.005
year+0.005
chard+0.005
ver+0.005
Neg logit contributions
 wounds-0.006
 rules-0.006
 buttons-0.006
 reins-0.006
 stret-0.006
Token,
Token ID11
Feature activation-8.45e-03

Pos logit contributions
 mess+0.034
 buttons+0.034
 rules+0.034
 wounds+0.033
 situation+0.032
Neg logit contributions
ores-0.017
year-0.016
chard-0.015
ver-0.015
 named-0.014
Token Mia
Token ID32189
Feature activation+3.55e-02

Pos logit contributions
ores+0.013
year+0.012
arium+0.012
working+0.012
 Captain+0.012
Neg logit contributions
 buttons-0.012
 wounds-0.012
 ow-0.012
 rules-0.011
 swings-0.011
Token spotted
Token ID13489
Feature activation-1.84e-02

Pos logit contributions
 wounds+0.062
 buttons+0.060
 mess+0.059
 stret+0.058
 reins+0.058
Neg logit contributions
ores-0.048
year-0.044
ver-0.043
arium-0.042
apped-0.042
Token something
Token ID1223
Feature activation-7.33e-04

Pos logit contributions
ores+0.023
ver+0.022
arium+0.021
year+0.021
 surely+0.021
Neg logit contributions
 wounds-0.033
 buttons-0.033
 blood-0.032
 rules-0.032
 ow-0.031
Token special
Token ID2041
Feature activation-2.53e-02

Pos logit contributions
 named+0.001
ores+0.001
arium+0.001
year+0.001
chard+0.001
Neg logit contributions
 wounds-0.001
 reins-0.001
 stret-0.001
 brakes-0.001
 rules-0.001
Token-
Token ID12
Feature activation-8.89e-02

Pos logit contributions
ores+0.036
ver+0.034
arium+0.034
 named+0.034
year+0.033
Neg logit contributions
 wounds-0.041
 buttons-0.040
 rules-0.039
 reins-0.039
 stret-0.038
Tokena
Token ID64
Feature activation-4.80e-02

Pos logit contributions
apped+0.153
ores+0.152
ver+0.148
arium+0.146
 surely+0.143
Neg logit contributions
 buttons-0.119
 rules-0.108
 wounds-0.106
 reins-0.105
co-0.104
Token glowing
Token ID21377
Feature activation-7.18e-03

Pos logit contributions
 named+0.063
ores+0.062
arium+0.061
ver+0.059
year+0.058
Neg logit contributions
 wounds-0.083
 reins-0.079
 stret-0.079
 buttons-0.078
 brakes-0.078
Token egg
Token ID5935
Feature activation-5.15e-03

Pos logit contributions
ores+0.009
ver+0.009
arium+0.009
 named+0.009
year+0.009
Neg logit contributions
 wounds-0.012
 buttons-0.012
 reins-0.012
 rules-0.012
 stret-0.012
Token!
Token ID0
Feature activation+1.40e-02

Pos logit contributions
 named+0.007
ores+0.006
arium+0.006
ver+0.006
apped+0.006
Neg logit contributions
 wounds-0.009
 stret-0.009
 reins-0.009
 brakes-0.009
 rules-0.009
Token Curious
Token ID44269
Feature activation+1.54e-02

Pos logit contributions
 buttons+0.016
 reins+0.015
 rules+0.015
 blood+0.015
 ow+0.015
Neg logit contributions
 known-0.024
 surely-0.024
ores-0.024
ver-0.024
working-0.023
Token she
Token ID673
Feature activation+1.51e-02

Pos logit contributions
 mess+0.032
 wounds+0.031
 rules+0.031
 buttons+0.030
 stain+0.030
Neg logit contributions
year-0.031
ver-0.029
ores-0.029
apped-0.029
arium-0.029
Token needed
Token ID2622
Feature activation-5.59e-03

Pos logit contributions
 wounds+0.023
 buttons+0.023
 mess+0.023
 stret+0.023
 reins+0.022
Neg logit contributions
ores-0.022
chard-0.020
ver-0.020
arium-0.020
year-0.020
Token a
Token ID257
Feature activation-1.32e-02

Pos logit contributions
ores+0.008
year+0.008
ver+0.008
apped+0.008
arium+0.008
Neg logit contributions
 buttons-0.009
 wounds-0.009
 rules-0.009
 reins-0.008
 stret-0.008
Token little
Token ID1310
Feature activation-1.17e-02

Pos logit contributions
arium+0.021
 named+0.021
ver+0.020
chard+0.019
 surely+0.019
Neg logit contributions
 wounds-0.020
 brakes-0.018
 buttons-0.018
 reins-0.018
 rules-0.018
Token pick
Token ID2298
Feature activation+2.14e-02

Pos logit contributions
 named+0.017
ores+0.017
arium+0.017
ver+0.017
year+0.016
Neg logit contributions
 wounds-0.018
 reins-0.017
 buttons-0.017
 stret-0.017
 rules-0.017
Token-
Token ID12
Feature activation-8.79e-02

Pos logit contributions
 wounds+0.042
 buttons+0.041
 rules+0.041
 stret+0.040
 reins+0.040
Neg logit contributions
ores-0.022
arium-0.021
 named-0.021
ver-0.021
year-0.021
Tokenme
Token ID1326
Feature activation-5.46e-02

Pos logit contributions
ores+0.174
 named+0.173
arium+0.171
ver+0.167
apped+0.167
Neg logit contributions
 buttons-0.089
 rules-0.083
 reins-0.082
 wounds-0.081
co-0.076
Token-
Token ID12
Feature activation-8.65e-02

Pos logit contributions
ores+0.087
arium+0.086
chard+0.083
year+0.082
 named+0.082
Neg logit contributions
 wounds-0.079
 buttons-0.078
 rules-0.077
 reins-0.074
 stret-0.072
Tokenup
Token ID929
Feature activation-1.35e-02

Pos logit contributions
ores+0.175
 named+0.174
arium+0.173
ver+0.169
apped+0.169
Neg logit contributions
 buttons-0.083
 rules-0.077
 reins-0.076
 wounds-0.076
co-0.072
Token,
Token ID11
Feature activation-6.88e-03

Pos logit contributions
ores+0.010
arium+0.009
 named+0.009
year+0.009
ver+0.009
Neg logit contributions
 wounds-0.031
 buttons-0.030
 stret-0.030
 reins-0.030
 rules-0.030
Token she
Token ID673
Feature activation+1.27e-02

Pos logit contributions
arium+0.010
ores+0.010
ver+0.009
year+0.009
apped+0.009
Neg logit contributions
 buttons-0.011
 wounds-0.011
 mess-0.010
 rules-0.010
 blood-0.010
Token festival
Token ID10876
Feature activation-8.19e-03

Pos logit contributions
 wounds+0.010
 pounds+0.009
 reins+0.009
 thorn+0.009
 stret+0.009
Neg logit contributions
arium-0.005
ores-0.005
 named-0.005
 Captain-0.005
chard-0.005
Token.
Token ID13
Feature activation-1.03e-03

Pos logit contributions
ores+0.015
year+0.015
ver+0.014
apped+0.013
 known+0.013
Neg logit contributions
 rules-0.010
 mess-0.009
 buttons-0.009
 wounds-0.009
co-0.009
Token He
Token ID679
Feature activation-1.53e-02

Pos logit contributions
ores+0.002
ver+0.002
 named+0.002
arium+0.002
chard+0.002
Neg logit contributions
 buttons-0.001
 rules-0.001
 brakes-0.001
 reins-0.001
 wounds-0.001
Token played
Token ID2826
Feature activation-2.64e-02

Pos logit contributions
ores+0.021
arium+0.019
ver+0.019
year+0.019
apped+0.019
Neg logit contributions
 wounds-0.026
 buttons-0.025
 reins-0.025
 stret-0.025
 mess-0.025
Token tug
Token ID27762
Feature activation-6.21e-03

Pos logit contributions
ores+0.038
ver+0.037
arium+0.037
year+0.036
 named+0.036
Neg logit contributions
 wounds-0.042
 stret-0.041
 buttons-0.041
 rules-0.040
 mess-0.040
Token-
Token ID12
Feature activation-8.74e-02

Pos logit contributions
ores+0.009
 named+0.008
ver+0.008
year+0.008
arium+0.008
Neg logit contributions
 wounds-0.010
 reins-0.010
 stret-0.010
 buttons-0.010
 rules-0.010
Tokenof
Token ID1659
Feature activation-5.69e-03

Pos logit contributions
arium+0.149
ores+0.148
 named+0.148
ver+0.143
apped+0.141
Neg logit contributions
 wounds-0.113
 reins-0.110
 buttons-0.110
 stret-0.108
 rules-0.107
Token-
Token ID12
Feature activation-8.30e-02

Pos logit contributions
ores+0.008
ver+0.008
 named+0.008
arium+0.007
apped+0.007
Neg logit contributions
 wounds-0.009
 brakes-0.009
 rules-0.009
 mess-0.008
 stret-0.008
Tokenwar
Token ID5767
Feature activation+3.55e-03

Pos logit contributions
ores+0.142
arium+0.142
 named+0.141
ver+0.137
apped+0.135
Neg logit contributions
 wounds-0.107
 buttons-0.103
 reins-0.103
 stret-0.101
 rules-0.101
Token and
Token ID290
Feature activation-1.31e-02

Pos logit contributions
 wounds+0.007
 buttons+0.006
 reins+0.006
 stret+0.006
 rules+0.006
Neg logit contributions
ores-0.004
 named-0.004
arium-0.004
ver-0.004
year-0.004
Token ran
Token ID4966
Feature activation+8.04e-03

Pos logit contributions
ores+0.021
arium+0.021
year+0.021
ver+0.020
apped+0.019
Neg logit contributions
 buttons-0.019
 mess-0.019
 brakes-0.018
 wounds-0.018
 rules-0.018
Token,
Token ID11
Feature activation-1.14e-02

Pos logit contributions
 named+0.030
ores+0.030
arium+0.028
ver+0.027
apped+0.027
Neg logit contributions
 wounds-0.050
 reins-0.049
 stret-0.048
 buttons-0.048
 brakes-0.048
Token but
Token ID475
Feature activation+3.90e-03

Pos logit contributions
 named+0.016
ores+0.016
arium+0.015
ver+0.015
year+0.015
Neg logit contributions
 wounds-0.018
 reins-0.018
 stret-0.017
 buttons-0.017
 rules-0.017
Token their
Token ID511
Feature activation+2.43e-02

Pos logit contributions
 wounds+0.005
 bleeding+0.005
 blood+0.005
 mess+0.005
 baby+0.005
Neg logit contributions
ores-0.006
arium-0.006
year-0.006
apped-0.005
ver-0.005
Token cl
Token ID537
Feature activation+1.38e-02

Pos logit contributions
 stret+0.037
 reins+0.034
co+0.034
 buttons+0.033
 wounds+0.033
Neg logit contributions
 named-0.035
 surely-0.034
ores-0.034
arium-0.033
 lived-0.033
Tokenums
Token ID5700
Feature activation+2.81e-02

Pos logit contributions
 wounds+0.024
 buttons+0.023
 reins+0.022
 stret+0.022
 rules+0.022
Neg logit contributions
ores-0.019
 named-0.018
ver-0.018
arium-0.018
apped-0.018
Tokeniness
Token ID1272
Feature activation+5.99e-02

Pos logit contributions
 wounds+0.046
 reins+0.044
 stret+0.044
 buttons+0.044
 rules+0.043
Neg logit contributions
ores-0.039
 named-0.039
ver-0.038
arium-0.038
year-0.037
Token often
Token ID1690
Feature activation-1.21e-02

Pos logit contributions
 wounds+0.122
 reins+0.117
 stret+0.113
 brakes+0.110
 rules+0.109
Neg logit contributions
ver-0.067
ores-0.065
 named-0.064
 lived-0.063
 surely-0.062
Token made
Token ID925
Feature activation-1.51e-02

Pos logit contributions
ores+0.016
year+0.016
arium+0.015
ver+0.015
 named+0.015
Neg logit contributions
 wounds-0.021
 buttons-0.020
 mess-0.020
 rules-0.020
 stret-0.020
Token them
Token ID606
Feature activation+1.04e-02

Pos logit contributions
ores+0.018
 named+0.017
apped+0.016
 known+0.016
 Captain+0.016
Neg logit contributions
 reins-0.029
 wounds-0.028
 brakes-0.028
 stret-0.028
 buttons-0.027
Token laugh
Token ID6487
Feature activation+1.53e-02

Pos logit contributions
 wounds+0.019
 reins+0.019
 buttons+0.019
 stret+0.019
 brakes+0.019
Neg logit contributions
ores-0.012
ver-0.011
 named-0.011
arium-0.011
apped-0.011
Token too
Token ID1165
Feature activation-1.61e-02

Pos logit contributions
 wounds+0.026
 reins+0.025
 buttons+0.025
 stret+0.025
 rules+0.025
Neg logit contributions
ores-0.020
 named-0.019
arium-0.019
ver-0.019
year-0.019
Token the
Token ID262
Feature activation+7.36e-03

Pos logit contributions
 wounds+0.004
 reins+0.003
 buttons+0.003
 stret+0.003
 rules+0.003
Neg logit contributions
ores-0.004
 named-0.004
arium-0.004
ver-0.004
year-0.004
Token truck
Token ID7779
Feature activation-8.13e-04

Pos logit contributions
 pounds+0.009
umbling+0.008
Days+0.008
co+0.008
we+0.007
Neg logit contributions
 named-0.013
chard-0.012
 Captain-0.012
 lived-0.011
apped-0.011
Token from
Token ID422
Feature activation-1.34e-02

Pos logit contributions
ores+0.001
 named+0.001
arium+0.001
ver+0.001
year+0.001
Neg logit contributions
 wounds-0.001
 buttons-0.001
 reins-0.001
 stret-0.001
 rules-0.001
Token Ben
Token ID3932
Feature activation+3.38e-02

Pos logit contributions
ver+0.021
arium+0.018
apped+0.018
year+0.018
 known+0.017
Neg logit contributions
 wounds-0.023
 buttons-0.022
 blood-0.021
 mess-0.021
 rules-0.021
Token and
Token ID290
Feature activation+4.05e-02

Pos logit contributions
 wounds+0.059
 reins+0.056
 buttons+0.056
 stret+0.055
 brakes+0.055
Neg logit contributions
 named-0.045
ores-0.044
arium-0.043
ver-0.042
apped-0.040
Token Mia
Token ID32189
Feature activation+7.99e-02

Pos logit contributions
 wounds+0.056
 buttons+0.053
 reins+0.052
 stret+0.051
 brakes+0.051
Neg logit contributions
ores-0.069
ver-0.068
arium-0.067
year-0.067
 named-0.066
Token.
Token ID13
Feature activation+3.10e-02

Pos logit contributions
 wounds+0.126
 buttons+0.120
 reins+0.119
 stret+0.119
 rules+0.117
Neg logit contributions
ores-0.118
arium-0.114
 named-0.113
ver-0.112
year-0.112
Token
Token ID198
Feature activation+2.26e-02

Pos logit contributions
 wounds+0.039
 blood+0.038
 buttons+0.038
 rules+0.036
 brakes+0.035
Neg logit contributions
ver-0.054
year-0.053
arium-0.053
 surely-0.052
working-0.052
Token
Token ID198
Feature activation+2.06e-02

Pos logit contributions
 wounds+0.043
 buttons+0.041
 reins+0.040
 stret+0.040
 rules+0.040
Neg logit contributions
ores-0.026
arium-0.025
 named-0.025
ver-0.025
year-0.024
Token"
Token ID1
Feature activation+2.60e-02

Pos logit contributions
 wounds+0.030
 buttons+0.028
 rules+0.028
 reins+0.028
 stret+0.027
Neg logit contributions
ores-0.033
ver-0.033
arium-0.032
 named-0.032
year-0.031
TokenEnough
Token ID47323
Feature activation+1.56e-02

Pos logit contributions
 wounds+0.039
 pounds+0.037
 stret+0.037
 rules+0.037
 reins+0.037
Neg logit contributions
ores-0.040
 named-0.039
year-0.038
 known-0.037
chard-0.037
TokenL
Token ID43
Feature activation-4.97e-03

Pos logit contributions
arium+0.022
ores+0.021
 named+0.021
ver+0.021
apped+0.020
Neg logit contributions
 buttons-0.021
 wounds-0.020
 rules-0.020
 reins-0.020
 brakes-0.019
Tokenily
Token ID813
Feature activation+2.19e-02

Pos logit contributions
ores+0.007
 named+0.006
year+0.006
ver+0.006
arium+0.006
Neg logit contributions
 wounds-0.009
 brakes-0.009
 reins-0.009
 buttons-0.009
 stret-0.008
Token and
Token ID290
Feature activation+4.37e-02

Pos logit contributions
 wounds+0.031
 mess+0.031
 buttons+0.030
 stret+0.030
 reins+0.029
Neg logit contributions
year-0.034
ores-0.034
arium-0.033
ver-0.032
chard-0.032
Token Tom
Token ID4186
Feature activation+2.75e-02

Pos logit contributions
 wounds+0.070
 buttons+0.067
 mess+0.065
 blood+0.065
 brakes+0.065
Neg logit contributions
year-0.065
ores-0.064
arium-0.063
ver-0.062
apped-0.060
Token's
Token ID338
Feature activation+3.23e-02

Pos logit contributions
 wounds+0.042
 mess+0.041
 buttons+0.041
 rules+0.040
 stret+0.040
Neg logit contributions
ores-0.040
year-0.039
arium-0.039
ver-0.038
 named-0.037
Token treasure
Token ID14068
Feature activation+6.14e-02

Pos logit contributions
 wounds+0.046
 buttons+0.044
 reins+0.043
 stret+0.043
 rules+0.043
Neg logit contributions
ores-0.052
 named-0.051
arium-0.050
ver-0.050
year-0.049
Token.
Token ID13
Feature activation+2.13e-02

Pos logit contributions
 wounds+0.096
 stret+0.094
 reins+0.093
 brakes+0.089
 rules+0.087
Neg logit contributions
arium-0.089
 named-0.089
ores-0.087
 lived-0.084
ver-0.083
Token Do
Token ID2141
Feature activation+2.26e-04

Pos logit contributions
 wounds+0.030
 buttons+0.030
 reins+0.029
 brakes+0.029
 rules+0.029
Neg logit contributions
arium-0.033
ores-0.032
ver-0.032
year-0.032
chard-0.031
Token you
Token ID345
Feature activation+6.03e-04

Pos logit contributions
 wounds+0.000
 reins+0.000
 stret+0.000
 buttons+0.000
 rules+0.000
Neg logit contributions
ores-0.000
 named-0.000
arium-0.000
ver-0.000
year-0.000
Token know
Token ID760
Feature activation+3.72e-02

Pos logit contributions
 mess+0.001
 rip+0.001
 stain+0.001
 rules+0.001
 boo+0.001
Neg logit contributions
arium-0.001
apped-0.001
year-0.001
ores-0.001
 Finally-0.001
Token how
Token ID703
Feature activation+3.45e-02

Pos logit contributions
 wounds+0.069
 buttons+0.066
 reins+0.066
 rules+0.065
 stret+0.065
Neg logit contributions
ores-0.044
arium-0.042
 named-0.042
ver-0.042
year-0.041
Token walking
Token ID6155
Feature activation+3.06e-03

Pos logit contributions
 blood+0.006
 bleeding+0.006
 sticking+0.006
 ow+0.006
 brakes+0.006
Neg logit contributions
ores-0.009
year-0.009
arium-0.008
ver-0.008
riage-0.008
Token through
Token ID832
Feature activation+5.02e-03

Pos logit contributions
 reins+0.005
 buttons+0.005
 wounds+0.005
 rules+0.005
 brakes+0.005
Neg logit contributions
ores-0.004
ver-0.004
 known-0.004
arium-0.004
year-0.004
Token a
Token ID257
Feature activation-1.66e-03

Pos logit contributions
 wounds+0.010
 buttons+0.010
 rules+0.009
 brakes+0.009
 blood+0.009
Neg logit contributions
ores-0.006
ver-0.005
year-0.005
apped-0.005
arium-0.005
Token forest
Token ID8222
Feature activation+4.77e-03

Pos logit contributions
 named+0.002
arium+0.002
ores+0.002
 lived+0.002
chard+0.002
Neg logit contributions
 wounds-0.003
 reins-0.003
 stret-0.003
 buttons-0.002
 pounds-0.002
Token,
Token ID11
Feature activation+8.30e-04

Pos logit contributions
 blood+0.008
 footprints+0.008
 mess+0.008
 situation+0.008
 buttons+0.008
Neg logit contributions
year-0.006
ores-0.006
ver-0.006
 Enjoy-0.006
 Amen-0.005
Token Fred
Token ID8559
Feature activation+5.81e-02

Pos logit contributions
 blood+0.001
 ow+0.001
 buttons+0.001
 brakes+0.001
 thorn+0.001
Neg logit contributions
year-0.001
ores-0.001
arium-0.001
apped-0.001
riage-0.001
Token saw
Token ID2497
Feature activation-4.28e-04

Pos logit contributions
 wounds+0.109
 buttons+0.104
 reins+0.104
 stret+0.103
 rules+0.102
Neg logit contributions
ores-0.069
arium-0.065
 named-0.065
ver-0.065
year-0.063
Token a
Token ID257
Feature activation-7.98e-03

Pos logit contributions
ver+0.001
ores+0.001
arium+0.001
 surely+0.000
apped+0.000
Neg logit contributions
 wounds-0.001
 blood-0.001
 buttons-0.001
 ow-0.001
 rules-0.001
Token big
Token ID1263
Feature activation-3.95e-03

Pos logit contributions
 named+0.014
arium+0.014
chard+0.014
ores+0.014
 Captain+0.013
Neg logit contributions
 wounds-0.010
 reins-0.009
 buttons-0.009
 stret-0.009
 rules-0.009
Token tree
Token ID5509
Feature activation+1.97e-02

Pos logit contributions
ores+0.007
ver+0.007
arium+0.007
 named+0.006
year+0.006
Neg logit contributions
 wounds-0.005
 buttons-0.005
 reins-0.005
 stret-0.005
 rules-0.005
Token.
Token ID13
Feature activation+9.26e-03

Pos logit contributions
 wounds+0.036
 reins+0.034
 stret+0.034
 buttons+0.034
 rules+0.033
Neg logit contributions
 named-0.025
ores-0.024
arium-0.023
ver-0.023
year-0.023
Token boy
Token ID2933
Feature activation-1.35e-02

Pos logit contributions
ores+0.005
ver+0.005
arium+0.004
year+0.004
 known+0.004
Neg logit contributions
 wounds-0.001
 buttons-0.001
 reins-0.001
 stret-0.001
 mess-0.001
Token named
Token ID3706
Feature activation-4.67e-02

Pos logit contributions
ver+0.012
ores+0.011
 surely+0.010
chard+0.010
arium+0.009
Neg logit contributions
 ow-0.029
 mess-0.028
 scrape-0.028
 blood-0.028
 thorn-0.028
Token Tim
Token ID5045
Feature activation+3.17e-02

Pos logit contributions
ores+0.059
arium+0.058
ver+0.057
year+0.056
 known+0.055
Neg logit contributions
 buttons-0.080
 wounds-0.080
 rules-0.078
 mess-0.078
 brakes-0.077
Token was
Token ID373
Feature activation+2.39e-02

Pos logit contributions
 wounds+0.054
 mess+0.053
 buttons+0.053
 stret+0.052
 blood+0.051
Neg logit contributions
ores-0.041
ver-0.039
arium-0.039
year-0.038
chard-0.037
Token very
Token ID845
Feature activation+1.36e-02

Pos logit contributions
 wounds+0.038
 stret+0.036
 mess+0.036
 buttons+0.036
 blood+0.035
Neg logit contributions
arium-0.036
ores-0.035
ver-0.034
year-0.032
riage-0.031
Token eager
Token ID11069
Feature activation+6.02e-02

Pos logit contributions
 mess+0.021
 boo+0.021
 wounds+0.020
 buttons+0.020
 rip+0.020
Neg logit contributions
arium-0.021
ver-0.020
ores-0.019
riage-0.018
 surely-0.017
Token to
Token ID284
Feature activation+2.14e-02

Pos logit contributions
 wounds+0.131
 stret+0.121
 reins+0.121
 pounds+0.119
 brakes+0.118
Neg logit contributions
 named-0.062
ores-0.056
arium-0.055
working-0.053
apped-0.051
Token play
Token ID711
Feature activation+2.66e-03

Pos logit contributions
 wounds+0.031
 buttons+0.031
 stret+0.030
 reins+0.030
 rules+0.030
Neg logit contributions
ores-0.033
year-0.033
arium-0.032
ver-0.032
apped-0.031
Token.
Token ID13
Feature activation+2.47e-03

Pos logit contributions
 wounds+0.005
 buttons+0.005
 stret+0.005
 reins+0.005
 rules+0.005
Neg logit contributions
ores-0.003
ver-0.003
arium-0.003
 named-0.003
year-0.003
Token He
Token ID679
Feature activation-1.02e-02

Pos logit contributions
 wounds+0.004
 stret+0.004
 reins+0.004
 buttons+0.004
 rules+0.004
Neg logit contributions
ores-0.003
 named-0.003
ver-0.003
arium-0.003
apped-0.003
Token ran
Token ID4966
Feature activation+1.79e-03

Pos logit contributions
ores+0.013
arium+0.012
ver+0.012
 named+0.012
year+0.012
Neg logit contributions
 wounds-0.018
 buttons-0.018
 reins-0.017
 stret-0.017
 rules-0.017
Token safe
Token ID3338
Feature activation+1.95e-02

Pos logit contributions
ores+0.021
 named+0.021
arium+0.021
ver+0.020
year+0.020
Neg logit contributions
 wounds-0.019
 buttons-0.018
 stret-0.018
 reins-0.018
 rules-0.018
Token feeling
Token ID4203
Feature activation+2.31e-02

Pos logit contributions
 wounds+0.030
 buttons+0.028
 reins+0.028
 stret+0.028
 rules+0.028
Neg logit contributions
ores-0.029
 named-0.028
arium-0.028
ver-0.028
year-0.028
Token.
Token ID13
Feature activation-8.13e-03

Pos logit contributions
 wounds+0.031
 rules+0.030
 buttons+0.030
 mess+0.029
 stret+0.029
Neg logit contributions
ores-0.037
arium-0.037
apped-0.036
ver-0.036
chard-0.035
Token<|endoftext|>
Token ID50256
Feature activation+6.47e-03

Pos logit contributions
ores+0.012
chard+0.011
year+0.011
 named+0.010
arium+0.010
Neg logit contributions
 rules-0.013
 wounds-0.013
 reins-0.012
 buttons-0.012
 brakes-0.012
TokenTim
Token ID14967
Feature activation+2.30e-03

Pos logit contributions
 buttons+0.010
 wounds+0.010
 rules+0.010
 blood+0.010
 reins+0.010
Neg logit contributions
ores-0.009
year-0.008
chard-0.008
ver-0.008
arium-0.008
Token and
Token ID290
Feature activation+3.92e-03

Pos logit contributions
 wounds+0.002
 mess+0.002
 buttons+0.002
 reins+0.002
 stret+0.002
Neg logit contributions
ores-0.005
year-0.004
ver-0.004
arium-0.004
apped-0.004
Token Lucy
Token ID22162
Feature activation+2.90e-02

Pos logit contributions
 wounds+0.007
 reins+0.006
 buttons+0.006
 stret+0.006
 brakes+0.006
Neg logit contributions
ores-0.005
arium-0.005
ver-0.005
year-0.005
apped-0.005
Token were
Token ID547
Feature activation+2.96e-02

Pos logit contributions
 buttons+0.046
 stret+0.045
 mess+0.045
 wounds+0.045
 ow+0.044
Neg logit contributions
ores-0.040
arium-0.039
year-0.038
ver-0.037
chard-0.036
Token best
Token ID1266
Feature activation+6.10e-03

Pos logit contributions
 wounds+0.047
 reins+0.045
 mess+0.044
 stret+0.044
 buttons+0.044
Neg logit contributions
ores-0.042
arium-0.041
ver-0.040
year-0.039
chard-0.039
Token friends
Token ID2460
Feature activation+3.17e-03

Pos logit contributions
 wounds+0.010
 stret+0.010
 buttons+0.010
 mess+0.010
 brakes+0.009
Neg logit contributions
 named-0.009
ores-0.008
ver-0.008
apped-0.008
year-0.008
Token in
Token ID287
Feature activation+1.27e-02

Pos logit contributions
 wounds+0.005
 buttons+0.005
 reins+0.005
 stret+0.005
 rules+0.005
Neg logit contributions
ores-0.005
arium-0.005
 named-0.005
ver-0.004
year-0.004
Token to
Token ID284
Feature activation+3.54e-03

Pos logit contributions
ores+0.020
 named+0.019
year+0.018
arium+0.018
chard+0.018
Neg logit contributions
 wounds-0.025
 stret-0.024
 reins-0.024
 brakes-0.023
 buttons-0.023
Token paint
Token ID7521
Feature activation-2.16e-02

Pos logit contributions
 buttons+0.005
 wounds+0.005
 stret+0.005
 mess+0.005
 rules+0.005
Neg logit contributions
ver-0.005
arium-0.005
ores-0.005
year-0.005
apped-0.005
Token pictures
Token ID5986
Feature activation-4.70e-02

Pos logit contributions
arium+0.033
ver+0.033
ores+0.032
year+0.032
 named+0.031
Neg logit contributions
 buttons-0.033
 wounds-0.033
 mess-0.031
 rules-0.031
 stret-0.031
Token of
Token ID286
Feature activation-6.91e-03

Pos logit contributions
ores+0.068
arium+0.067
ver+0.066
 named+0.065
year+0.065
Neg logit contributions
 buttons-0.073
 wounds-0.072
 rules-0.070
 reins-0.069
 mess-0.068
Token animals
Token ID4695
Feature activation-2.03e-02

Pos logit contributions
ores+0.010
 named+0.010
arium+0.010
year+0.010
ver+0.010
Neg logit contributions
 wounds-0.011
 reins-0.010
 stret-0.010
 brakes-0.010
 buttons-0.010
Token and
Token ID290
Feature activation+1.21e-02

Pos logit contributions
ores+0.034
year+0.033
arium+0.033
ver+0.032
apped+0.032
Neg logit contributions
 buttons-0.027
 wounds-0.026
 rules-0.025
 reins-0.025
 mess-0.025
Token people
Token ID661
Feature activation+1.35e-02

Pos logit contributions
 wounds+0.016
 reins+0.015
 buttons+0.015
 stret+0.015
 rules+0.014
Neg logit contributions
ores-0.021
 named-0.021
arium-0.021
ver-0.020
year-0.020
Token.
Token ID13
Feature activation+1.79e-03

Pos logit contributions
 wounds+0.021
 buttons+0.020
 reins+0.020
 stret+0.020
 rules+0.020
Neg logit contributions
ores-0.020
arium-0.020
 named-0.019
ver-0.019
year-0.019
Token One
Token ID1881
Feature activation-5.78e-03

Pos logit contributions
 wounds+0.003
 buttons+0.003
 rules+0.003
 reins+0.003
 stret+0.003
Neg logit contributions
ores-0.002
arium-0.002
ver-0.002
year-0.002
 named-0.002
Token day
Token ID1110
Feature activation-1.65e-02

Pos logit contributions
ver+0.007
ores+0.006
arium+0.006
 Captain+0.006
apped+0.006
Neg logit contributions
 buttons-0.010
 mess-0.010
 ow-0.010
 rules-0.010
 thorn-0.010
Token,
Token ID11
Feature activation-1.04e-02

Pos logit contributions
year+0.023
arium+0.023
ores+0.023
 known+0.021
 surely+0.020
Neg logit contributions
 buttons-0.027
 ow-0.027
 blood-0.026
 sticking-0.026
 button-0.025
Token smiled
Token ID13541
Feature activation-1.70e-02

Pos logit contributions
 wounds+0.029
 buttons+0.028
 reins+0.028
 stret+0.027
 brakes+0.027
Neg logit contributions
ores-0.027
arium-0.026
ver-0.025
year-0.025
 named-0.024
Token.
Token ID13
Feature activation+1.00e-02

Pos logit contributions
ores+0.024
arium+0.023
 named+0.023
ver+0.023
year+0.023
Neg logit contributions
 wounds-0.028
 buttons-0.027
 reins-0.026
 stret-0.026
 rules-0.026
Token She
Token ID1375
Feature activation-3.39e-03

Pos logit contributions
 wounds+0.015
 reins+0.015
 buttons+0.015
 brakes+0.015
 rules+0.014
Neg logit contributions
ores-0.015
arium-0.015
ver-0.015
 named-0.015
year-0.014
Token wasn
Token ID2492
Feature activation-1.08e-02

Pos logit contributions
ores+0.004
arium+0.004
 named+0.004
ver+0.004
year+0.004
Neg logit contributions
 wounds-0.006
 buttons-0.006
 reins-0.006
 stret-0.006
 rules-0.005
Token't
Token ID470
Feature activation+7.59e-03

Pos logit contributions
ver+0.013
ores+0.013
 named+0.012
 surely+0.012
year+0.011
Neg logit contributions
 wounds-0.021
 buttons-0.021
 mess-0.020
 rules-0.020
 pounds-0.020
Token scared
Token ID12008
Feature activation+7.34e-03

Pos logit contributions
 wounds+0.015
 reins+0.014
 buttons+0.014
 rules+0.014
 stret+0.014
Neg logit contributions
 named-0.009
ores-0.009
ver-0.008
 known-0.008
year-0.008
Token anymore
Token ID7471
Feature activation+5.30e-03

Pos logit contributions
 wounds+0.018
 stret+0.017
 reins+0.017
 brakes+0.016
 scarf+0.016
Neg logit contributions
,-0.007
ores-0.005
 named-0.005
ver-0.005
 whenever-0.005
Token and
Token ID290
Feature activation+8.49e-04

Pos logit contributions
 wounds+0.009
 buttons+0.009
 reins+0.009
 stret+0.009
 rules+0.009
Neg logit contributions
ores-0.007
arium-0.007
 named-0.007
ver-0.007
year-0.007
Token knew
Token ID2993
Feature activation+1.69e-02

Pos logit contributions
 wounds+0.001
 buttons+0.001
 brakes+0.001
 mess+0.001
 blood+0.001
Neg logit contributions
ores-0.001
arium-0.001
ver-0.001
apped-0.001
year-0.001
Token that
Token ID326
Feature activation-1.96e-02

Pos logit contributions
 wounds+0.031
 buttons+0.030
 reins+0.029
 rules+0.029
 stret+0.029
Neg logit contributions
ores-0.021
arium-0.020
 named-0.020
ver-0.020
year-0.019
Token the
Token ID262
Feature activation-5.73e-03

Pos logit contributions
ores+0.025
apped+0.024
arium+0.024
ver+0.024
year+0.024
Neg logit contributions
 wounds-0.034
 buttons-0.033
 rules-0.033
 mess-0.032
 blood-0.032
Token had
Token ID550
Feature activation-6.95e-03

Pos logit contributions
 wounds+0.024
 buttons+0.024
 mess+0.024
 zipper+0.023
 stain+0.022
Neg logit contributions
ores-0.028
arium-0.025
year-0.024
 Captain-0.023
 Enjoy-0.023
Token to
Token ID284
Feature activation+2.68e-02

Pos logit contributions
ores+0.009
 named+0.009
arium+0.009
ver+0.009
year+0.008
Neg logit contributions
 wounds-0.012
 reins-0.012
 buttons-0.011
 stret-0.011
 rules-0.011
Token promise
Token ID6991
Feature activation+5.58e-03

Pos logit contributions
 wounds+0.039
 stret+0.037
 rules+0.037
 buttons+0.037
 reins+0.036
Neg logit contributions
ores-0.043
arium-0.042
year-0.042
ver-0.041
 named-0.041
Token to
Token ID284
Feature activation+9.55e-03

Pos logit contributions
 wounds+0.010
 reins+0.010
 buttons+0.009
 stret+0.009
 brakes+0.009
Neg logit contributions
ores-0.007
 named-0.007
arium-0.007
ver-0.007
 known-0.006
Token always
Token ID1464
Feature activation-2.64e-02

Pos logit contributions
 wounds+0.013
 buttons+0.012
 reins+0.012
 stret+0.012
 rules+0.012
Neg logit contributions
ores-0.016
arium-0.016
ver-0.016
year-0.016
 named-0.015
Token be
Token ID307
Feature activation+3.31e-03

Pos logit contributions
ores+0.042
arium+0.041
year+0.040
ver+0.040
 named+0.039
Neg logit contributions
 wounds-0.039
 buttons-0.037
 reins-0.036
 rules-0.036
 stret-0.036
Token best
Token ID1266
Feature activation+1.13e-02

Pos logit contributions
 wounds+0.005
 stret+0.004
 rules+0.004
 buttons+0.004
 reins+0.004
Neg logit contributions
arium-0.005
ores-0.005
ver-0.005
year-0.005
apped-0.005
Token friends
Token ID2460
Feature activation-4.47e-03

Pos logit contributions
 wounds+0.018
 stret+0.017
 reins+0.017
 buttons+0.017
 brakes+0.017
Neg logit contributions
 named-0.016
ores-0.016
ver-0.016
arium-0.015
year-0.015
Token.
Token ID13
Feature activation-8.58e-03

Pos logit contributions
ores+0.007
arium+0.007
year+0.007
apped+0.007
 known+0.007
Neg logit contributions
 wounds-0.006
 rules-0.006
 buttons-0.006
 reins-0.006
 stret-0.006
Token They
Token ID1119
Feature activation-1.12e-02

Pos logit contributions
ores+0.010
year+0.010
ver+0.010
arium+0.010
chard+0.009
Neg logit contributions
 wounds-0.015
 rules-0.015
 stret-0.015
 reins-0.015
 mess-0.014
Token made
Token ID925
Feature activation-2.74e-02

Pos logit contributions
ores+0.018
arium+0.016
year+0.015
ver+0.015
 Captain+0.015
Neg logit contributions
 buttons-0.016
 wounds-0.016
 mess-0.015
co-0.015
 zipper-0.015
Token!"
Token ID2474
Feature activation-1.59e-02

Pos logit contributions
 wounds+0.015
 buttons+0.014
 rules+0.014
 reins+0.014
 mess+0.014
Neg logit contributions
ores-0.020
year-0.019
arium-0.019
 named-0.019
ver-0.019
Token He
Token ID679
Feature activation-1.60e-02

Pos logit contributions
ores+0.017
 named+0.017
arium+0.017
ver+0.016
year+0.016
Neg logit contributions
 wounds-0.031
 reins-0.030
 buttons-0.030
 stret-0.030
 rules-0.029
Token pulled
Token ID5954
Feature activation-3.83e-04

Pos logit contributions
 named+0.022
ver+0.020
ores+0.020
arium+0.020
apped+0.020
Neg logit contributions
 wounds-0.028
 reins-0.027
 stret-0.026
 rules-0.026
 buttons-0.026
Token the
Token ID262
Feature activation-2.15e-02

Pos logit contributions
ores+0.000
 named+0.000
year+0.000
ver+0.000
apped+0.000
Neg logit contributions
 wounds-0.001
 stret-0.001
 reins-0.001
 pounds-0.001
 rules-0.001
Token balloon
Token ID21190
Feature activation-2.60e-02

Pos logit contributions
chard+0.032
apped+0.031
ver+0.031
 surely+0.030
arium+0.030
Neg logit contributions
 wounds-0.031
co-0.031
 stret-0.031
we-0.030
 pounds-0.029
Token hard
Token ID1327
Feature activation+3.96e-03

Pos logit contributions
ores+0.032
ver+0.031
 named+0.031
arium+0.031
 surely+0.029
Neg logit contributions
 wounds-0.047
 reins-0.045
 stret-0.045
 buttons-0.044
 brakes-0.043
Token.
Token ID13
Feature activation-1.99e-03

Pos logit contributions
 wounds+0.007
 reins+0.007
 stret+0.007
 buttons+0.006
 brakes+0.006
Neg logit contributions
ores-0.005
 named-0.005
ver-0.005
arium-0.005
year-0.005
Token The
Token ID383
Feature activation-9.46e-03

Pos logit contributions
ores+0.002
arium+0.002
 named+0.002
ver+0.002
year+0.002
Neg logit contributions
 wounds-0.004
 buttons-0.004
 stret-0.004
 rules-0.004
 reins-0.003
Token string
Token ID4731
Feature activation-1.97e-02

Pos logit contributions
chard+0.015
apped+0.014
 named+0.014
 lived+0.014
arium+0.014
Neg logit contributions
 wounds-0.013
 pounds-0.013
co-0.013
 stret-0.013
we-0.012
Token broke
Token ID6265
Feature activation-2.27e-02

Pos logit contributions
 named+0.025
ores+0.025
ver+0.025
apped+0.024
arium+0.024
Neg logit contributions
 wounds-0.034
 reins-0.033
 stret-0.033
 rules-0.032
 buttons-0.032
Token.
Token ID13
Feature activation-2.86e-04

Pos logit contributions
ores+0.023
apped+0.023
 named+0.022
ver+0.021
year+0.020
Neg logit contributions
 stret-0.047
 wounds-0.046
 reins-0.046
 rules-0.043
 brakes-0.043
TokenB
Token ID33
Feature activation-1.58e-03

Pos logit contributions
 wounds+0.001
 reins+0.001
 buttons+0.001
 stret+0.001
 brakes+0.001
Neg logit contributions
 named-0.001
ores-0.001
year-0.001
ver-0.001
arium-0.001
Tokenenny
Token ID11870
Feature activation+8.40e-03

Pos logit contributions
ver+0.002
apped+0.002
year+0.002
ores+0.002
 named+0.002
Neg logit contributions
 wounds-0.003
 stret-0.003
 buttons-0.003
 brakes-0.003
 reins-0.003
Token looked
Token ID3114
Feature activation-9.61e-03

Pos logit contributions
 wounds+0.012
 buttons+0.012
 mess+0.012
 stret+0.011
 rules+0.011
Neg logit contributions
ores-0.014
ver-0.013
year-0.013
arium-0.013
chard-0.012
Token around
Token ID1088
Feature activation-2.53e-02

Pos logit contributions
ver+0.015
 known+0.015
ores+0.015
riage+0.014
apped+0.014
Neg logit contributions
 mess-0.015
 wounds-0.014
 buttons-0.014
 blood-0.013
 reins-0.013
Token and
Token ID290
Feature activation-1.53e-02

Pos logit contributions
ver+0.042
ores+0.040
apped+0.039
year+0.039
chard+0.038
Neg logit contributions
 wounds-0.036
 buttons-0.035
 mess-0.035
 blood-0.034
 rules-0.034
Token saw
Token ID2497
Feature activation-6.82e-03

Pos logit contributions
ores+0.027
ver+0.027
apped+0.026
year+0.025
 Enjoy+0.025
Neg logit contributions
 wounds-0.019
 mess-0.018
 buttons-0.018
 blood-0.017
 faces-0.017
Token his
Token ID465
Feature activation+1.29e-03

Pos logit contributions
ores+0.009
arium+0.009
ver+0.009
 surely+0.008
year+0.008
Neg logit contributions
 blood-0.012
 wounds-0.012
 buttons-0.012
 ow-0.011
 rules-0.011
Token friend
Token ID1545
Feature activation+5.06e-03

Pos logit contributions
 wounds+0.002
 stret+0.002
 reins+0.002
 buttons+0.002
 brakes+0.002
Neg logit contributions
 named-0.002
arium-0.002
ores-0.002
chard-0.002
ver-0.002
Token,
Token ID11
Feature activation+1.04e-02

Pos logit contributions
 wounds+0.011
 buttons+0.011
 stret+0.011
 mess+0.010
 rules+0.010
Neg logit contributions
ores-0.004
year-0.004
arium-0.004
ver-0.004
apped-0.004
Token Sammy
Token ID44067
Feature activation+3.75e-02

Pos logit contributions
 wounds+0.020
 stret+0.019
 buttons+0.019
 reins+0.019
 brakes+0.019
Neg logit contributions
ores-0.011
arium-0.011
ver-0.011
year-0.011
 named-0.010
Token the
Token ID262
Feature activation+2.09e-02

Pos logit contributions
 wounds+0.072
 buttons+0.069
 reins+0.069
 stret+0.068
 rules+0.068
Neg logit contributions
ores-0.042
 named-0.042
arium-0.041
ver-0.040
year-0.039
Token Lily
Token ID20037
Feature activation+1.17e-02

Pos logit contributions
ores+0.003
ver+0.003
arium+0.003
year+0.003
apped+0.002
Neg logit contributions
 wounds-0.003
 buttons-0.003
 mess-0.003
 rules-0.003
 blood-0.003
Token said
Token ID531
Feature activation+1.57e-02

Pos logit contributions
 wounds+0.020
 buttons+0.020
 mess+0.019
 stret+0.019
 rules+0.019
Neg logit contributions
ores-0.016
arium-0.015
year-0.015
 named-0.014
ver-0.014
Token.
Token ID13
Feature activation+5.07e-03

Pos logit contributions
 wounds+0.025
 buttons+0.024
 reins+0.023
 stret+0.023
 rules+0.023
Neg logit contributions
ores-0.023
arium-0.023
 named-0.022
ver-0.022
year-0.022
Token She
Token ID1375
Feature activation-3.76e-03

Pos logit contributions
 wounds+0.009
 reins+0.009
 buttons+0.009
 stret+0.009
 rules+0.009
Neg logit contributions
ores-0.006
 named-0.006
arium-0.006
ver-0.006
year-0.006
Token hugged
Token ID41130
Feature activation-2.49e-02

Pos logit contributions
ores+0.005
 named+0.005
arium+0.005
ver+0.005
year+0.005
Neg logit contributions
 wounds-0.007
 buttons-0.006
 reins-0.006
 stret-0.006
 rules-0.006
Token her
Token ID607
Feature activation-8.42e-03

Pos logit contributions
ores+0.030
 named+0.029
arium+0.028
ver+0.028
year+0.027
Neg logit contributions
 wounds-0.046
 reins-0.044
 stret-0.044
 buttons-0.044
 rules-0.044
Token block
Token ID2512
Feature activation+1.56e-02

Pos logit contributions
 named+0.015
ores+0.015
ver+0.015
arium+0.015
apped+0.015
Neg logit contributions
 wounds-0.010
 stret-0.010
 reins-0.009
 rules-0.009
 buttons-0.009
Token and
Token ID290
Feature activation-4.42e-03

Pos logit contributions
 wounds+0.023
 buttons+0.022
 reins+0.021
 stret+0.021
 rules+0.021
Neg logit contributions
ores-0.025
arium-0.024
 named-0.024
ver-0.024
year-0.023
Token did
Token ID750
Feature activation+3.01e-02

Pos logit contributions
ores+0.006
arium+0.006
ver+0.006
year+0.006
 named+0.006
Neg logit contributions
 wounds-0.008
 buttons-0.007
 mess-0.007
 brakes-0.007
 blood-0.007
Token not
Token ID407
Feature activation+1.06e-02

Pos logit contributions
 wounds+0.047
 buttons+0.044
 mess+0.044
 rules+0.043
 stret+0.042
Neg logit contributions
arium-0.046
year-0.045
ores-0.045
 named-0.045
apped-0.045
Token share
Token ID2648
Feature activation+6.76e-03

Pos logit contributions
 wounds+0.020
 buttons+0.019
 reins+0.019
 stret+0.019
 rules+0.018
Neg logit contributions
ores-0.013
arium-0.012
 named-0.012
ver-0.012
year-0.012
Token it
Token ID340
Feature activation-1.08e-02

Pos logit contributions
arium+0.019
ores+0.019
year+0.017
chard+0.017
 known+0.017
Neg logit contributions
 wounds-0.026
 buttons-0.025
 blood-0.024
 rules-0.024
 stret-0.024
Token was
Token ID373
Feature activation-5.97e-03

Pos logit contributions
ores+0.018
arium+0.017
 named+0.017
year+0.017
apped+0.016
Neg logit contributions
 buttons-0.015
 wounds-0.015
 rules-0.014
 brakes-0.014
 mess-0.014
Token a
Token ID257
Feature activation-1.00e-03

Pos logit contributions
arium+0.009
apped+0.009
ver+0.009
year+0.009
ores+0.009
Neg logit contributions
 wounds-0.009
 mess-0.009
 buttons-0.009
 blood-0.009
 rules-0.009
Token watch
Token ID2342
Feature activation+2.51e-02

Pos logit contributions
 named+0.002
ores+0.002
arium+0.002
ver+0.002
chard+0.002
Neg logit contributions
 wounds-0.001
 reins-0.001
 buttons-0.001
 stret-0.001
 brakes-0.001
Token he
Token ID339
Feature activation-1.02e-03

Pos logit contributions
 wounds+0.044
 buttons+0.042
 reins+0.042
 stret+0.042
 rules+0.041
Neg logit contributions
ores-0.032
arium-0.031
 named-0.031
ver-0.031
year-0.030
Token had
Token ID550
Feature activation-4.68e-03

Pos logit contributions
ores+0.001
 named+0.001
arium+0.001
ver+0.001
year+0.001
Neg logit contributions
 wounds-0.002
 buttons-0.002
 reins-0.002
 stret-0.002
 rules-0.002
Token stolen
Token ID9909
Feature activation+1.04e-02

Pos logit contributions
ores+0.006
 named+0.006
arium+0.006
ver+0.006
year+0.006
Neg logit contributions
 wounds-0.008
 reins-0.008
 buttons-0.008
 stret-0.008
 rules-0.008
Token.
Token ID13
Feature activation-1.48e-02

Pos logit contributions
 wounds+0.019
 reins+0.018
 stret+0.018
 buttons+0.018
 rules+0.017
Neg logit contributions
ores-0.013
 named-0.012
arium-0.012
ver-0.012
year-0.012
Token
Token ID220
Feature activation+3.23e-03

Pos logit contributions
ores+0.020
arium+0.019
 named+0.019
year+0.019
ver+0.018
Neg logit contributions
 wounds-0.025
 buttons-0.024
 rules-0.024
 brakes-0.024
 stret-0.024
Token
Token ID198
Feature activation-1.61e-03

Pos logit contributions
 wounds+0.006
 buttons+0.006
 stret+0.006
 reins+0.006
 rules+0.006
Neg logit contributions
ores-0.004
arium-0.004
 named-0.004
year-0.004
ver-0.004
Token
Token ID198
Feature activation-9.14e-04

Pos logit contributions
ores+0.002
 named+0.002
ver+0.002
arium+0.002
year+0.002
Neg logit contributions
 wounds-0.003
 reins-0.003
 buttons-0.003
 stret-0.003
 mess-0.003
Token cl
Token ID537
Feature activation+4.96e-03

Pos logit contributions
 buttons+0.003
 wounds+0.003
 mess+0.003
 faces+0.003
 rules+0.003
Neg logit contributions
ores-0.003
ver-0.003
arium-0.003
 Finally-0.003
apped-0.003
Tokenapped
Token ID6320
Feature activation-4.42e-03

Pos logit contributions
 faces+0.008
 wounds+0.008
 stret+0.008
 buttons+0.008
co+0.008
Neg logit contributions
chard-0.007
ver-0.006
 known-0.006
arium-0.006
 surely-0.006
Token their
Token ID511
Feature activation+1.83e-02

Pos logit contributions
ores+0.006
 named+0.005
arium+0.005
ver+0.005
year+0.005
Neg logit contributions
 wounds-0.008
 reins-0.008
 stret-0.008
 buttons-0.008
 brakes-0.007
Token hands
Token ID2832
Feature activation+4.94e-02

Pos logit contributions
 wounds+0.023
 reins+0.022
 stret+0.022
 blood+0.021
 pounds+0.021
Neg logit contributions
ores-0.031
apped-0.031
 named-0.031
arium-0.030
year-0.029
Token.
Token ID13
Feature activation+9.27e-03

Pos logit contributions
 wounds+0.080
 buttons+0.075
 reins+0.075
 stret+0.075
 rules+0.074
Neg logit contributions
ores-0.071
 named-0.069
arium-0.068
ver-0.068
year-0.067
Token Tom
Token ID4186
Feature activation-3.66e-03

Pos logit contributions
 buttons+0.014
 rules+0.014
 brakes+0.013
 wounds+0.013
 mess+0.013
Neg logit contributions
ver-0.015
 surely-0.014
chard-0.014
year-0.014
 known-0.014
Token and
Token ID290
Feature activation+2.25e-02

Pos logit contributions
ores+0.006
year+0.006
arium+0.006
ver+0.006
chard+0.006
Neg logit contributions
 wounds-0.005
 buttons-0.005
 mess-0.005
 stret-0.005
 reins-0.005
Token Mia
Token ID32189
Feature activation+5.74e-02

Pos logit contributions
 wounds+0.039
 buttons+0.038
 stret+0.037
 reins+0.037
 mess+0.037
Neg logit contributions
ores-0.030
ver-0.029
arium-0.029
year-0.028
 named-0.028
Token started
Token ID2067
Feature activation+2.65e-02

Pos logit contributions
 buttons+0.088
 wounds+0.085
 mess+0.083
 reins+0.082
 stret+0.081
Neg logit contributions
ores-0.089
year-0.084
arium-0.084
chard-0.081
 Captain-0.079
Token their
Token ID511
Feature activation+2.25e-02

Pos logit contributions
 wounds+0.036
 buttons+0.036
 rules+0.034
 mess+0.034
 stret+0.034
Neg logit contributions
ver-0.044
ores-0.044
arium-0.043
year-0.042
 named-0.042
Token show
Token ID905
Feature activation+4.71e-02

Pos logit contributions
 wounds+0.032
 reins+0.030
 buttons+0.030
 stret+0.030
 rules+0.029
Neg logit contributions
ores-0.037
 named-0.036
arium-0.036
ver-0.035
year-0.034
Token tightly
Token ID17707
Feature activation+1.80e-02

Pos logit contributions
 wounds+0.031
 reins+0.030
 buttons+0.030
 stret+0.030
 rules+0.029
Neg logit contributions
ores-0.031
arium-0.030
ver-0.029
 named-0.029
year-0.029
Token.
Token ID13
Feature activation+6.67e-03

Pos logit contributions
 wounds+0.032
 reins+0.030
 buttons+0.030
 rules+0.030
 stret+0.030
Neg logit contributions
 named-0.023
ores-0.022
ver-0.022
arium-0.021
apped-0.021
Token
Token ID220
Feature activation+1.66e-02

Pos logit contributions
 wounds+0.012
 reins+0.012
 stret+0.012
 buttons+0.012
 blood+0.011
Neg logit contributions
ores-0.008
year-0.007
arium-0.007
ver-0.007
working-0.007
Token
Token ID198
Feature activation+1.28e-02

Pos logit contributions
 wounds+0.034
 buttons+0.033
 reins+0.033
 rules+0.033
 stret+0.033
Neg logit contributions
ores-0.016
ver-0.016
 named-0.016
apped-0.015
arium-0.015
Token
Token ID198
Feature activation+1.23e-02

Pos logit contributions
 wounds+0.025
 buttons+0.024
 reins+0.024
 stret+0.024
 rules+0.024
Neg logit contributions
ores-0.014
 named-0.013
arium-0.013
ver-0.013
year-0.013
TokenFrom
Token ID4863
Feature activation-6.45e-03

Pos logit contributions
 wounds+0.022
 brakes+0.022
 buttons+0.021
 reins+0.021
 pounds+0.021
Neg logit contributions
 named-0.016
ores-0.015
year-0.015
arium-0.015
ver-0.015
Token that
Token ID326
Feature activation-1.32e-02

Pos logit contributions
ver+0.010
year+0.010
apped+0.010
ores+0.010
arium+0.010
Neg logit contributions
 wounds-0.010
 buttons-0.009
 blood-0.009
 mess-0.009
 ow-0.009
Token day
Token ID1110
Feature activation-8.86e-03

Pos logit contributions
ores+0.016
chard+0.015
year+0.014
ver+0.014
apped+0.014
Neg logit contributions
 rules-0.025
 mess-0.025
 reins-0.024
 buttons-0.024
 situation-0.024
Token on
Token ID319
Feature activation-1.66e-02

Pos logit contributions
ores+0.014
arium+0.014
chard+0.013
year+0.013
 Captain+0.012
Neg logit contributions
 reins-0.012
 wounds-0.012
 brakes-0.012
 buttons-0.012
 stret-0.012
Token,
Token ID11
Feature activation-1.43e-02

Pos logit contributions
ores+0.013
year+0.013
ver+0.011
arium+0.011
chard+0.011
Neg logit contributions
 wounds-0.038
 mess-0.037
 buttons-0.037
 stret-0.037
 rules-0.037
Token Lily
Token ID20037
Feature activation+3.35e-03

Pos logit contributions
year+0.027
ores+0.026
ver+0.025
arium+0.025
apped+0.025
Neg logit contributions
 mess-0.017
 swe-0.016
 rules-0.016
 wounds-0.016
 brace-0.016
Token outside
Token ID2354
Feature activation+1.91e-02

Pos logit contributions
 wounds+0.009
 stret+0.009
 buttons+0.009
 mess+0.009
 rules+0.009
Neg logit contributions
ores-0.007
ver-0.007
arium-0.007
year-0.007
 known-0.006
Token when
Token ID618
Feature activation+9.80e-03

Pos logit contributions
 wounds+0.034
 mess+0.034
 buttons+0.033
 rules+0.033
 blood+0.032
Neg logit contributions
ver-0.023
ores-0.023
arium-0.022
year-0.021
 Captain-0.021
Token she
Token ID673
Feature activation+7.38e-03

Pos logit contributions
 blood+0.014
 wounds+0.014
 buttons+0.014
 ow+0.014
 mess+0.014
Neg logit contributions
year-0.015
ver-0.015
ores-0.015
arium-0.015
apped-0.014
Token noticed
Token ID6810
Feature activation-1.12e-02

Pos logit contributions
 mess+0.011
 blood+0.011
 wounds+0.011
 stain+0.011
 buttons+0.011
Neg logit contributions
ores-0.011
year-0.011
ver-0.010
arium-0.010
 Captain-0.010
Token how
Token ID703
Feature activation+2.21e-02

Pos logit contributions
ores+0.012
arium+0.012
ver+0.012
 named+0.011
year+0.011
Neg logit contributions
 wounds-0.022
 buttons-0.021
 stret-0.021
 reins-0.021
 rules-0.021
Token cold
Token ID4692
Feature activation-6.49e-02

Pos logit contributions
 wounds+0.029
 buttons+0.028
 rules+0.027
 mess+0.027
 reins+0.026
Neg logit contributions
arium-0.038
ores-0.037
ver-0.037
 named-0.037
 Captain-0.036
Token the
Token ID262
Feature activation-2.58e-02

Pos logit contributions
arium+0.082
ores+0.082
ver+0.082
year+0.080
 named+0.080
Neg logit contributions
 wounds-0.112
 buttons-0.110
 rules-0.107
 stret-0.106
 mess-0.106
Token air
Token ID1633
Feature activation-3.52e-02

Pos logit contributions
 named+0.037
 night+0.037
ores+0.036
working+0.036
chard+0.035
Neg logit contributions
 wounds-0.038
 pounds-0.037
 stret-0.035
 reins-0.034
 brakes-0.033
Token was
Token ID373
Feature activation-7.06e-03

Pos logit contributions
ores+0.047
 named+0.045
arium+0.044
ver+0.043
year+0.042
Neg logit contributions
 wounds-0.062
 reins-0.060
 stret-0.058
 buttons-0.057
 rules-0.057
Token.
Token ID13
Feature activation-5.73e-03

Pos logit contributions
ores+0.010
 named+0.009
ver+0.009
arium+0.009
 surely+0.009
Neg logit contributions
 wounds-0.012
 reins-0.012
 stret-0.011
 brakes-0.011
 buttons-0.011
Token She
Token ID1375
Feature activation-8.70e-03

Pos logit contributions
ores+0.008
arium+0.008
ver+0.008
year+0.008
 named+0.008
Neg logit contributions
 wounds-0.009
 buttons-0.009
 reins-0.009
 stret-0.009
 rules-0.009
Token the
Token ID262
Feature activation+1.14e-02

Pos logit contributions
ver+0.059
arium+0.055
apped+0.054
ores+0.053
year+0.052
Neg logit contributions
 wounds-0.064
 buttons-0.061
 blood-0.060
 reins-0.059
 rules-0.058
Token steak
Token ID26320
Feature activation+1.10e-02

Pos logit contributions
 wounds+0.017
 stret+0.016
co+0.016
 reins+0.015
 buttons+0.015
Neg logit contributions
chard-0.017
ores-0.017
arium-0.016
 Captain-0.016
 surely-0.016
Token.
Token ID13
Feature activation-5.47e-03

Pos logit contributions
 wounds+0.019
 reins+0.018
 stret+0.017
 buttons+0.017
 brakes+0.017
Neg logit contributions
ores-0.015
arium-0.015
 named-0.015
ver-0.014
year-0.014
Token The
Token ID383
Feature activation-1.23e-02

Pos logit contributions
arium+0.007
ores+0.006
 named+0.006
chard+0.006
ver+0.006
Neg logit contributions
 wounds-0.009
 buttons-0.009
 reins-0.009
 rules-0.009
 stret-0.009
Token melt
Token ID16867
Feature activation-1.24e-02

Pos logit contributions
ores+0.021
 named+0.020
arium+0.020
year+0.020
chard+0.020
Neg logit contributions
 wounds-0.016
 stret-0.016
 reins-0.015
 buttons-0.015
 brakes-0.015
Token-
Token ID12
Feature activation-8.40e-02

Pos logit contributions
ores+0.016
 named+0.016
arium+0.016
ver+0.015
year+0.015
Neg logit contributions
 wounds-0.022
 buttons-0.021
 reins-0.020
 stret-0.020
 rules-0.020
Tokenin
Token ID259
Feature activation-1.35e-02

Pos logit contributions
arium+0.141
 named+0.137
ores+0.133
 Captain+0.132
ver+0.130
Neg logit contributions
 wounds-0.112
 reins-0.108
 buttons-0.107
 pounds-0.104
 rules-0.104
Token-
Token ID12
Feature activation-8.22e-02

Pos logit contributions
ores+0.022
arium+0.021
 named+0.021
ver+0.021
year+0.021
Neg logit contributions
 wounds-0.019
 buttons-0.018
 reins-0.018
 stret-0.018
 rules-0.018
Tokenthe
Token ID1169
Feature activation-1.91e-03

Pos logit contributions
arium+0.137
 named+0.134
ores+0.130
 Captain+0.128
ver+0.127
Neg logit contributions
 wounds-0.110
 reins-0.106
 buttons-0.105
 pounds-0.102
 rules-0.102
Token-
Token ID12
Feature activation-7.53e-02

Pos logit contributions
ores+0.003
 named+0.003
 surely+0.003
arium+0.003
year+0.003
Neg logit contributions
 wounds-0.003
 stret-0.003
 reins-0.002
 brakes-0.002
 buttons-0.002
Tokenmouth
Token ID14775
Feature activation+8.92e-03

Pos logit contributions
arium+0.125
 named+0.121
ores+0.118
 Captain+0.118
apped+0.116
Neg logit contributions
 wounds-0.098
 reins-0.097
 buttons-0.096
 pounds-0.093
 rules-0.092
Token They
Token ID1119
Feature activation-6.50e-03

Pos logit contributions
 buttons+0.016
 wounds+0.016
 stret+0.015
 reins+0.015
 rules+0.015
Neg logit contributions
ores-0.013
year-0.013
ver-0.012
arium-0.012
 named-0.012
Token played
Token ID2826
Feature activation-1.78e-02

Pos logit contributions
ores+0.010
 surely+0.009
year+0.009
arium+0.009
ver+0.009
Neg logit contributions
 buttons-0.010
 mess-0.009
 swings-0.009
co-0.009
 reins-0.009
Token hide
Token ID7808
Feature activation-1.83e-02

Pos logit contributions
ores+0.026
chard+0.025
year+0.025
ver+0.025
 known+0.025
Neg logit contributions
 wounds-0.028
 buttons-0.027
 mess-0.027
 rules-0.027
co-0.026
Token-
Token ID12
Feature activation-5.78e-02

Pos logit contributions
ores+0.032
 known+0.032
ver+0.031
 surely+0.031
chard+0.031
Neg logit contributions
 wounds-0.024
 buttons-0.023
 mess-0.022
 rules-0.022
co-0.022
Tokenand
Token ID392
Feature activation+2.99e-02

Pos logit contributions
ores+0.132
arium+0.130
ver+0.128
 named+0.128
apped+0.125
Neg logit contributions
 wounds-0.042
 buttons-0.039
 rules-0.039
 reins-0.038
 stret-0.038
Token-
Token ID12
Feature activation-5.55e-02

Pos logit contributions
 wounds+0.036
 buttons+0.034
 stret+0.033
 rules+0.033
 reins+0.033
Neg logit contributions
ores-0.055
arium-0.054
ver-0.054
 named-0.053
year-0.053
Tokenseek
Token ID36163
Feature activation+1.33e-02

Pos logit contributions
ores+0.100
arium+0.099
 surely+0.094
 Captain+0.094
ver+0.092
Neg logit contributions
co-0.072
 rules-0.067
 wounds-0.064
 buttons-0.063
 stret-0.062
Token in
Token ID287
Feature activation-6.98e-03

Pos logit contributions
 wounds+0.019
 buttons+0.019
 rules+0.019
 mess+0.018
co+0.017
Neg logit contributions
ores-0.021
chard-0.020
year-0.020
 surely-0.020
 known-0.020
Token the
Token ID262
Feature activation+1.53e-02

Pos logit contributions
ores+0.006
 named+0.006
arium+0.005
ver+0.005
year+0.005
Neg logit contributions
 wounds-0.016
 reins-0.015
 buttons-0.015
 stret-0.015
 rules-0.015
Token park
Token ID3952
Feature activation+7.11e-03

Pos logit contributions
 wounds+0.022
 reins+0.021
 pounds+0.021
 buttons+0.020
 stret+0.020
Neg logit contributions
ores-0.025
 named-0.025
arium-0.025
 lived-0.023
apped-0.023
Token.
Token ID13
Feature activation-4.03e-04

Pos logit contributions
 rules+0.009
co+0.009
 mess+0.008
 buttons+0.008
 swings+0.008
Neg logit contributions
ores-0.013
year-0.012
 Captain-0.012
 known-0.012
 surely-0.012
Token stories
Token ID3923
Feature activation-2.42e-02

Pos logit contributions
ores+0.022
 named+0.022
arium+0.022
ver+0.021
year+0.021
Neg logit contributions
 wounds-0.024
 stret-0.022
 reins-0.022
 buttons-0.022
 brakes-0.022
Token.
Token ID13
Feature activation+3.42e-03

Pos logit contributions
ores+0.034
 named+0.032
arium+0.032
ver+0.032
year+0.032
Neg logit contributions
 wounds-0.040
 buttons-0.038
 reins-0.038
 stret-0.038
 rules-0.037
Token Can
Token ID1680
Feature activation-3.17e-02

Pos logit contributions
 reins+0.004
 buttons+0.004
 rules+0.004
 brakes+0.004
 wounds+0.004
Neg logit contributions
ores-0.006
arium-0.006
year-0.006
ver-0.006
 known-0.006
Token you
Token ID345
Feature activation-3.60e-02

Pos logit contributions
ores+0.060
year+0.057
arium+0.057
ver+0.057
 named+0.056
Neg logit contributions
 wounds-0.038
 buttons-0.036
 rules-0.035
 reins-0.035
 brakes-0.034
Token read
Token ID1100
Feature activation-1.65e-02

Pos logit contributions
year+0.070
ores+0.066
apped+0.064
ver+0.063
arium+0.062
Neg logit contributions
 stain-0.043
 rip-0.043
 rules-0.042
 mess-0.042
 buttons-0.041
Token me
Token ID502
Feature activation-5.09e-02

Pos logit contributions
ores+0.025
year+0.025
ver+0.024
apped+0.024
 known+0.023
Neg logit contributions
 buttons-0.026
 rules-0.025
 wounds-0.025
 mess-0.024
 reins-0.023
Token one
Token ID530
Feature activation-2.06e-02

Pos logit contributions
ores+0.074
year+0.074
apped+0.073
 known+0.073
chard+0.069
Neg logit contributions
 buttons-0.076
co-0.074
 rules-0.074
 wounds-0.071
 reins-0.069
Token?"
Token ID1701
Feature activation-3.26e-02

Pos logit contributions
ores+0.028
 known+0.027
year+0.027
ver+0.027
apped+0.026
Neg logit contributions
 rules-0.034
 buttons-0.033
 reins-0.032
 mess-0.032
 wounds-0.032
Token
Token ID198
Feature activation-1.30e-03

Pos logit contributions
year+0.052
ores+0.051
arium+0.049
ver+0.049
 surely+0.048
Neg logit contributions
 wounds-0.046
 buttons-0.045
 reins-0.044
 blood-0.044
 rules-0.044
Token
Token ID198
Feature activation+1.50e-03

Pos logit contributions
ores+0.001
 named+0.001
arium+0.001
ver+0.001
year+0.001
Neg logit contributions
 wounds-0.003
 reins-0.003
 buttons-0.003
 stret-0.003
 brakes-0.003
TokenThe
Token ID464
Feature activation-3.78e-03

Pos logit contributions
 wounds+0.002
 buttons+0.002
 reins+0.002
 stret+0.002
 brakes+0.002
Neg logit contributions
 named-0.002
ores-0.002
arium-0.002
ver-0.002
year-0.002
Token we
Token ID356
Feature activation-1.77e-02

Pos logit contributions
 wounds+0.000
 buttons+0.000
 reins+0.000
 rules+0.000
 stret+0.000
Neg logit contributions
ores-0.000
arium-0.000
 named-0.000
ver-0.000
year-0.000
Token can
Token ID460
Feature activation-2.71e-03

Pos logit contributions
ores+0.028
arium+0.027
year+0.026
apped+0.026
chard+0.025
Neg logit contributions
 buttons-0.026
 wounds-0.025
 mess-0.025
 brakes-0.024
 rules-0.024
Token find
Token ID1064
Feature activation+1.81e-02

Pos logit contributions
ores+0.004
 named+0.004
ver+0.003
arium+0.003
year+0.003
Neg logit contributions
 wounds-0.005
 reins-0.004
 stret-0.004
 buttons-0.004
 rules-0.004
Token out
Token ID503
Feature activation-1.02e-03

Pos logit contributions
 wounds+0.035
 buttons+0.034
 reins+0.034
 stret+0.033
 rules+0.033
Neg logit contributions
ores-0.020
 named-0.019
arium-0.019
ver-0.019
year-0.018
Token how
Token ID703
Feature activation+2.17e-02

Pos logit contributions
ores+0.001
arium+0.001
 named+0.001
ver+0.001
year+0.001
Neg logit contributions
 wounds-0.002
 buttons-0.002
 reins-0.002
 rules-0.002
 stret-0.002
Token thick
Token ID6546
Feature activation-5.66e-02

Pos logit contributions
 wounds+0.038
 stret+0.036
 reins+0.036
 buttons+0.036
 mess+0.035
Neg logit contributions
ores-0.029
 named-0.027
arium-0.027
ver-0.027
year-0.027
Token it
Token ID340
Feature activation-2.18e-02

Pos logit contributions
ores+0.076
 named+0.074
arium+0.074
ver+0.073
year+0.072
Neg logit contributions
 wounds-0.096
 buttons-0.092
 reins-0.091
 stret-0.090
 rules-0.090
Token is
Token ID318
Feature activation-6.44e-03

Pos logit contributions
ores+0.033
arium+0.033
 named+0.033
ver+0.032
year+0.031
Neg logit contributions
 wounds-0.033
 reins-0.031
 stret-0.031
 buttons-0.031
 rules-0.030
Token."
Token ID526
Feature activation-2.54e-02

Pos logit contributions
ores+0.010
year+0.010
apped+0.010
arium+0.010
 named+0.010
Neg logit contributions
 wounds-0.009
 buttons-0.009
 mess-0.009
 rules-0.009
 reins-0.009
Token
Token ID198
Feature activation-2.90e-03

Pos logit contributions
ores+0.031
arium+0.030
ver+0.029
year+0.029
 named+0.029
Neg logit contributions
 wounds-0.047
 buttons-0.045
 stret-0.044
 reins-0.044
 rules-0.044
Token
Token ID198
Feature activation-1.69e-03

Pos logit contributions
ores+0.003
 named+0.003
arium+0.002
ver+0.002
year+0.002
Neg logit contributions
 wounds-0.006
 reins-0.006
 buttons-0.006
 brakes-0.006
 stret-0.006