TOP ACTIVATIONS
MAX = 0.155

sailed for days , exploring unch arted waters until he came
different styles like stars or spir als . Everyone was so
. Paul 's impressive football spir alled through the air and
sun fl owers , and tul ips . They see pine
! <|endoftext|> One day , Sil via was getting ready to
river . When Sil via arrived at the river
butterfly , tw irling and spir aling in the air .
the chest were the most wond eful toys the little girl
asleep , dreaming of icy pian os and big piles of
with your new coat ." Sil via smiled and said ,
and you love it , Sil via ! Come now ,
green and a long , spir alling tail . The princess
there was a girl named Sil via . She was 3
same . As Sil via was about to jump
grass and rye grass and al f alf a and cl
forth in the sky , spir alling higher and higher .
looks wonderful on you , Sil via ! You look so
She told him how she se wed , glued , and
own two hands . She se wed it with bright colors
and it fit perfectly ! Sil via was so happy she

MIN ACTIVATIONS
MIN = -0.104

his mistake and gave his sister a turn with the toy
cry . His mom heard him and came to see what
grandma 's house for safe keeping . When
it to me for safe keeping . He told me to
His mom heard him and came to help him
was angry and asked her daughter why she had done something
He tried to push the girl away , but she was
His mom hears him and runs to his room
cry . His mom heard him and ran to the dining
and said sorry to his sister . They started to play
. She sn atches her journal from Tom and pushes him
my 's mom shouted at him to turn it off .
His mom heard him and came to help him
him and came to help him . She saw the broken
Ben said , giving his sister a fruit .
cries . His mom hears him and comes to the sand
screamed and ran to help him . A man
His mom hears him and comes to him .
and Sam run to help him . They are sorry and
his mum came to help him . She was very kind

INTERVAL 0.090 - 0.155
CONTAINS 0.021%

cow . She loved to roam the fields and make friends
and said " Hello , cater pillar ". The cater pillar
see her very first flower blo oming in her garden .
, and they played and spl ashed in the water together
of the bucket . It sw am in a p uddle

INTERVAL 0.025 - 0.090
CONTAINS 14.473%

little girl named Mary . Mary loved to wear her favourite
." Ben che wed the gum and said ,
stayed on and the car moved . Tom was happy .
" Look , Mom ! More spots !" <|endoftext|> Once upon
young , she could still make a difference . <|endoftext|> John

INTERVAL -0.040 - 0.025
CONTAINS 79.538%

together in the yard . The cat was always very curious
it on a table where the dog cannot reach it .
. And that is the moral of the story : sharing
are sorry , m ister ," Lily said . " We
to know what you see ." Mary stepped closer and saw

INTERVAL -0.104 - -0.040
CONTAINS 5.967%

He smiled and squeezed Lily 's hand gently . They entered
up and ran to her mom . She gave
my toy ," said the boy with a sad face .
back to the owl and said , " Thank you !".
heard Sara crying and came to her . She saw Sara
Token sailed
Token ID33835
Feature activation+5.92e-02

Pos logit contributions
ang+0.084
irling+0.078
irl+0.075
ading+0.072
ash+0.069
Neg logit contributions
 ranger-0.048
 sob-0.047
 recovered-0.047
 department-0.045
 aid-0.045
Token for
Token ID329
Feature activation+2.05e-02

Pos logit contributions
ang+0.114
irling+0.102
ee+0.098
irl+0.095
ug+0.092
Neg logit contributions
 ranger-0.104
 bleed-0.101
 bru-0.100
 hinges-0.100
 sob-0.099
Token days
Token ID1528
Feature activation-7.16e-03

Pos logit contributions
ang+0.035
irling+0.030
ee+0.028
irl+0.027
ash+0.026
Neg logit contributions
 ranger-0.042
 hinges-0.041
 bru-0.041
 bleed-0.041
 sob-0.040
Token,
Token ID11
Feature activation+1.09e-02

Pos logit contributions
 ranger+0.009
 hinges+0.008
 aid+0.008
 injuries+0.008
 bru+0.008
Neg logit contributions
ang-0.016
irl-0.016
irling-0.015
ee-0.015
ash-0.015
Token exploring
Token ID13504
Feature activation+4.72e-02

Pos logit contributions
ang+0.023
irling+0.021
ee+0.021
irl+0.020
ash+0.020
Neg logit contributions
 ranger-0.018
 sob-0.016
 bru-0.016
 bleed-0.016
 hinges-0.016
Token unch
Token ID16258
Feature activation+1.54e-01

Pos logit contributions
ang+0.093
irling+0.083
ee+0.080
irl+0.077
ug+0.074
Neg logit contributions
 ranger-0.082
 hinges-0.078
 bru-0.077
 bleed-0.076
 sob-0.076
Tokenarted
Token ID19112
Feature activation+4.28e-02

Pos logit contributions
ang+0.246
irling+0.235
irl+0.226
ee+0.220
ading+0.191
Neg logit contributions
 ranger-0.287
 hinges-0.277
 bleed-0.275
onna-0.274
 bru-0.261
Token waters
Token ID10150
Feature activation+4.63e-02

Pos logit contributions
ang+0.077
irling+0.069
irl+0.065
ee+0.065
ug+0.060
Neg logit contributions
 ranger-0.082
 bleed-0.079
 hinges-0.078
 bru-0.077
 sob-0.076
Token until
Token ID1566
Feature activation+1.82e-03

Pos logit contributions
ang+0.100
irling+0.091
ee+0.089
irl+0.087
ug+0.082
Neg logit contributions
 ranger-0.069
 bleed-0.067
 hinges-0.065
 bru-0.065
 sob-0.064
Token he
Token ID339
Feature activation+1.83e-02

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ug+0.003
Neg logit contributions
 ranger-0.003
 bleed-0.003
 bru-0.003
 hinges-0.003
 sob-0.003
Token came
Token ID1625
Feature activation+1.30e-02

Pos logit contributions
ang+0.042
irling+0.038
ee+0.036
irl+0.036
ug+0.035
Neg logit contributions
 ranger-0.026
 bleed-0.024
 bru-0.024
 sob-0.024
 department-0.024
Token different
Token ID1180
Feature activation+2.37e-02

Pos logit contributions
ang+0.031
irling+0.028
ee+0.027
irl+0.026
ug+0.025
Neg logit contributions
 ranger-0.037
 bleed-0.036
 bru-0.035
 hinges-0.035
 sob-0.035
Token styles
Token ID12186
Feature activation+1.42e-02

Pos logit contributions
ang+0.036
irling+0.032
ee+0.031
irl+0.030
ading+0.029
Neg logit contributions
 ranger-0.049
 bleed-0.049
 bru-0.047
 sob-0.047
 department-0.046
Token like
Token ID588
Feature activation+1.95e-02

Pos logit contributions
ang+0.037
irling+0.034
ee+0.034
irl+0.033
ug+0.032
Neg logit contributions
 ranger-0.015
 bleed-0.014
 hinges-0.014
 bru-0.014
 sob-0.014
Token stars
Token ID5788
Feature activation+3.19e-02

Pos logit contributions
ang+0.034
irling+0.030
ee+0.029
irl+0.028
ug+0.027
Neg logit contributions
 ranger-0.037
 bleed-0.037
 bru-0.037
 hinges-0.036
 sob-0.036
Token or
Token ID393
Feature activation+2.60e-02

Pos logit contributions
ang+0.079
irling+0.073
ee+0.071
irl+0.069
ug+0.067
Neg logit contributions
 ranger-0.038
 bleed-0.037
 bru-0.037
 hinges-0.036
 sob-0.036
Token spir
Token ID9158
Feature activation+1.40e-01

Pos logit contributions
ang+0.042
irling+0.038
ee+0.036
irl+0.035
ug+0.033
Neg logit contributions
 ranger-0.051
 bleed-0.050
 bru-0.049
 hinges-0.049
 sob-0.049
Tokenals
Token ID874
Feature activation+3.48e-02

Pos logit contributions
ang+0.274
irling+0.267
irl+0.245
ee+0.224
ading+0.220
Neg logit contributions
 bleed-0.227
 department-0.214
 hinges-0.212
 bru-0.211
onna-0.210
Token.
Token ID13
Feature activation-5.24e-03

Pos logit contributions
ang+0.075
irling+0.068
ee+0.067
ug+0.063
irl+0.063
Neg logit contributions
 bru-0.052
 ranger-0.052
 sob-0.050
 hinges-0.050
 bleed-0.050
Token Everyone
Token ID11075
Feature activation+4.13e-02

Pos logit contributions
 ranger+0.008
 bleed+0.008
 bru+0.008
 hinges+0.008
 sob+0.008
Neg logit contributions
ang-0.011
irling-0.010
ee-0.010
irl-0.010
ug-0.009
Token was
Token ID373
Feature activation+1.59e-02

Pos logit contributions
ang+0.093
irling+0.087
irl+0.083
ug+0.081
ee+0.080
Neg logit contributions
 sob-0.056
 ranger-0.055
 bleed-0.052
 keeper-0.051
 bru-0.049
Token so
Token ID523
Feature activation+6.49e-03

Pos logit contributions
ang+0.031
ee+0.027
irling+0.026
ash+0.026
irl+0.025
Neg logit contributions
 ranger-0.028
 bleed-0.028
 sob-0.028
 bru-0.027
 keeper-0.026
Token.
Token ID13
Feature activation+1.51e-03

Pos logit contributions
 bleed+0.027
 bru+0.027
 ranger+0.026
 hinges+0.026
 sob+0.026
Neg logit contributions
ang-0.029
irling-0.026
ee-0.026
ug-0.024
irl-0.024
Token Paul
Token ID3362
Feature activation+4.45e-02

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ading+0.003
Neg logit contributions
 ranger-0.002
 hinges-0.002
 bleed-0.002
 bru-0.002
 sob-0.002
Token's
Token ID338
Feature activation-9.87e-03

Pos logit contributions
ang+0.103
irling+0.094
irl+0.092
ading+0.090
ee+0.087
Neg logit contributions
 ranger-0.059
 sob-0.056
 keeper-0.053
 department-0.052
 bru-0.052
Token impressive
Token ID8036
Feature activation-6.38e-04

Pos logit contributions
 ranger+0.015
 bleed+0.015
 bru+0.014
 hinges+0.014
 sob+0.014
Neg logit contributions
ang-0.021
irling-0.019
ee-0.019
irl-0.018
ug-0.018
Token football
Token ID4346
Feature activation+7.73e-03

Pos logit contributions
 ranger+0.001
 bleed+0.001
 bru+0.001
 hinges+0.001
 sob+0.001
Neg logit contributions
ang-0.001
irling-0.001
ee-0.001
irl-0.001
ug-0.001
Token spir
Token ID9158
Feature activation+1.35e-01

Pos logit contributions
ang+0.015
irling+0.013
ee+0.013
irl+0.013
ug+0.012
Neg logit contributions
 bleed-0.013
 bru-0.013
 hinges-0.013
 ranger-0.013
 sob-0.013
Tokenalled
Token ID4262
Feature activation+3.69e-02

Pos logit contributions
irling+0.220
ang+0.206
irl+0.200
icing+0.166
ading+0.162
Neg logit contributions
 department-0.244
 ranger-0.239
 bleed-0.239
driver-0.238
 recovered-0.224
Token through
Token ID832
Feature activation+4.16e-02

Pos logit contributions
ang+0.068
irling+0.061
ee+0.058
irl+0.057
ug+0.054
Neg logit contributions
 ranger-0.067
 bleed-0.066
 bru-0.065
 hinges-0.065
 sob-0.064
Token the
Token ID262
Feature activation+3.67e-02

Pos logit contributions
ang+0.073
irling+0.064
ee+0.061
irl+0.060
ug+0.057
Neg logit contributions
 ranger-0.081
 bleed-0.079
 bru-0.078
 hinges-0.078
 sob-0.077
Token air
Token ID1633
Feature activation+1.48e-03

Pos logit contributions
ang+0.072
irling+0.064
ug+0.060
irl+0.060
ee+0.059
Neg logit contributions
 ranger-0.066
 hinges-0.062
 bru-0.060
 department-0.060
 bleed-0.060
Token and
Token ID290
Feature activation+3.02e-02

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ug+0.003
Neg logit contributions
 ranger-0.002
 bleed-0.002
 bru-0.002
 hinges-0.002
 sob-0.002
Token sun
Token ID4252
Feature activation+3.49e-02

Pos logit contributions
ang+0.029
irling+0.026
ee+0.025
irl+0.025
ug+0.024
Neg logit contributions
 ranger-0.023
 bleed-0.023
 sob-0.023
 bru-0.022
 hinges-0.022
Tokenfl
Token ID2704
Feature activation+8.33e-02

Pos logit contributions
ang+0.075
irling+0.069
ee+0.065
irl+0.065
ug+0.063
Neg logit contributions
 ranger-0.052
 bleed-0.051
 hinges-0.050
 sob-0.049
 bru-0.049
Tokenowers
Token ID3618
Feature activation+4.60e-02

Pos logit contributions
ang+0.150
irling+0.139
irl+0.127
ee+0.126
ug+0.122
Neg logit contributions
 bleed-0.143
 bru-0.137
 ranger-0.137
 sob-0.136
 hinges-0.136
Token,
Token ID11
Feature activation+1.78e-02

Pos logit contributions
ang+0.101
irling+0.092
ee+0.088
irl+0.086
ug+0.086
Neg logit contributions
 ranger-0.066
 bleed-0.066
 sob-0.063
 hinges-0.062
 bru-0.061
Token and
Token ID290
Feature activation+2.50e-02

Pos logit contributions
ang+0.039
irling+0.035
ee+0.034
irl+0.033
ug+0.032
Neg logit contributions
 ranger-0.026
 bleed-0.026
 sob-0.026
 bru-0.026
 hinges-0.026
Token tul
Token ID48373
Feature activation+1.34e-01

Pos logit contributions
ang+0.049
irling+0.044
ee+0.042
irl+0.041
ug+0.040
Neg logit contributions
 bleed-0.043
 ranger-0.043
 bru-0.042
 sob-0.042
 hinges-0.041
Tokenips
Token ID2419
Feature activation+5.94e-02

Pos logit contributions
ang+0.186
irling+0.168
ee+0.156
ug+0.153
irl+0.150
Neg logit contributions
onna-0.273
 recovered-0.257
 bleed-0.256
 bru-0.256
 department-0.250
Token.
Token ID13
Feature activation+6.23e-03

Pos logit contributions
ang+0.118
irling+0.106
ee+0.102
irl+0.099
ug+0.096
Neg logit contributions
 ranger-0.099
 bleed-0.098
 bru-0.096
 hinges-0.096
 sob-0.095
Token They
Token ID1119
Feature activation+2.17e-02

Pos logit contributions
ang+0.014
irling+0.012
ee+0.012
irl+0.012
ug+0.012
Neg logit contributions
 bleed-0.009
 bru-0.009
 ranger-0.009
 hinges-0.009
 sob-0.008
Token see
Token ID766
Feature activation+3.18e-02

Pos logit contributions
ang+0.050
irling+0.045
ee+0.043
ug+0.042
irl+0.041
Neg logit contributions
 bleed-0.030
 ranger-0.030
 sob-0.029
 bru-0.029
 hinges-0.028
Token pine
Token ID20161
Feature activation+5.44e-02

Pos logit contributions
ang+0.058
irling+0.051
ee+0.049
irl+0.048
ug+0.046
Neg logit contributions
 ranger-0.059
 bleed-0.058
 bru-0.057
 sob-0.057
 hinges-0.057
Token!
Token ID0
Feature activation+1.15e-02

Pos logit contributions
 ranger+0.011
 bleed+0.011
 bru+0.011
 hinges+0.011
 sob+0.010
Neg logit contributions
ang-0.009
irling-0.008
ee-0.007
irl-0.007
ug-0.007
Token<|endoftext|>
Token ID50256
Feature activation+1.68e-02

Pos logit contributions
ang+0.028
irling+0.025
ee+0.025
irl+0.024
ug+0.023
Neg logit contributions
 bleed-0.015
 ranger-0.015
 bru-0.014
 hinges-0.014
 sob-0.014
TokenOne
Token ID3198
Feature activation+5.34e-02

Pos logit contributions
ang+0.036
ee+0.032
irling+0.032
irl+0.031
ug+0.030
Neg logit contributions
 bleed-0.026
 hinges-0.026
 ranger-0.025
 bru-0.025
 sob-0.025
Token day
Token ID1110
Feature activation-1.24e-02

Pos logit contributions
ang+0.090
irling+0.084
ug+0.075
angles+0.075
irl+0.069
Neg logit contributions
 ranger-0.105
 aid-0.097
 bru-0.097
 keeper-0.096
 sob-0.094
Token,
Token ID11
Feature activation+3.77e-02

Pos logit contributions
 ranger+0.019
 bleed+0.019
 bru+0.019
 hinges+0.019
 sob+0.019
Neg logit contributions
ang-0.026
irling-0.024
ee-0.023
irl-0.022
ug-0.021
Token Sil
Token ID4243
Feature activation+1.29e-01

Pos logit contributions
ang+0.083
irling+0.076
ee+0.072
irl+0.071
ug+0.069
Neg logit contributions
 ranger-0.055
 bleed-0.054
 sob-0.053
 bru-0.053
 hinges-0.052
Tokenvia
Token ID8869
Feature activation+4.32e-02

Pos logit contributions
ang+0.237
irling+0.215
irl+0.197
ading+0.194
ee+0.189
Neg logit contributions
 aid-0.208
 ranger-0.191
onna-0.190
 sob-0.189
 hinges-0.186
Token was
Token ID373
Feature activation+1.44e-02

Pos logit contributions
ang+0.098
irling+0.089
ee+0.085
irl+0.084
ug+0.082
Neg logit contributions
 ranger-0.059
 bleed-0.058
 sob-0.058
 bru-0.057
 hinges-0.056
Token getting
Token ID1972
Feature activation+1.71e-02

Pos logit contributions
ang+0.029
angles+0.029
ash+0.026
irling+0.026
izing+0.025
Neg logit contributions
 sob-0.024
 bleed-0.022
 aid-0.021
 bru-0.021
 cough-0.021
Token ready
Token ID3492
Feature activation-3.74e-02

Pos logit contributions
ang+0.033
irling+0.030
ee+0.028
irl+0.027
ug+0.027
Neg logit contributions
 bleed-0.029
 ranger-0.029
 sob-0.029
 bru-0.029
 hinges-0.028
Token to
Token ID284
Feature activation-5.11e-02

Pos logit contributions
 ranger+0.045
 bleed+0.045
 bru+0.043
 hinges+0.042
 sob+0.042
Neg logit contributions
ang-0.091
irling-0.085
ee-0.083
irl-0.081
ug-0.078
Token river
Token ID7850
Feature activation-3.02e-03

Pos logit contributions
ang+0.055
irling+0.051
ee+0.049
irl+0.048
ash+0.046
Neg logit contributions
 ranger-0.036
 hinges-0.034
 bleed-0.033
 keeper-0.032
 bru-0.031
Token.
Token ID13
Feature activation-6.63e-03

Pos logit contributions
 ranger+0.004
 bleed+0.004
 bru+0.004
 hinges+0.004
 sob+0.004
Neg logit contributions
ang-0.007
irling-0.007
ee-0.007
irl-0.006
ug-0.006
Token
Token ID198
Feature activation+2.39e-03

Pos logit contributions
 ranger+0.008
 bleed+0.008
 bru+0.008
 hinges+0.008
 sob+0.008
Neg logit contributions
ang-0.016
irling-0.015
ee-0.015
irl-0.014
ug-0.014
Token
Token ID198
Feature activation-1.43e-03

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.005
Neg logit contributions
 bleed-0.003
 ranger-0.003
 bru-0.003
 hinges-0.003
 sob-0.003
TokenWhen
Token ID2215
Feature activation-9.93e-04

Pos logit contributions
 ranger+0.002
 bleed+0.002
 hinges+0.002
 sob+0.002
 bru+0.002
Neg logit contributions
ang-0.003
irling-0.003
ee-0.003
irl-0.002
ash-0.002
Token Sil
Token ID4243
Feature activation+1.29e-01

Pos logit contributions
 ranger+0.001
 bleed+0.001
 hinges+0.001
 bru+0.001
 sob+0.001
Neg logit contributions
ang-0.002
irling-0.002
ee-0.002
irl-0.002
ash-0.002
Tokenvia
Token ID8869
Feature activation+2.58e-02

Pos logit contributions
ang+0.232
irling+0.216
irl+0.202
ading+0.197
ee+0.193
Neg logit contributions
 aid-0.195
 hinges-0.187
 ranger-0.186
ache-0.185
 sob-0.183
Token arrived
Token ID5284
Feature activation+4.18e-03

Pos logit contributions
ang+0.057
irling+0.052
ee+0.049
irl+0.049
ug+0.047
Neg logit contributions
 ranger-0.038
 bleed-0.037
 bru-0.036
 sob-0.036
 hinges-0.036
Token at
Token ID379
Feature activation+2.47e-02

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.004
Neg logit contributions
 ranger-0.010
 bleed-0.009
 bru-0.009
 hinges-0.009
 sob-0.009
Token the
Token ID262
Feature activation+1.25e-02

Pos logit contributions
ang+0.040
irling+0.035
ee+0.033
irl+0.032
ug+0.031
Neg logit contributions
 ranger-0.051
 bleed-0.051
 bru-0.050
 hinges-0.050
 sob-0.050
Token river
Token ID7850
Feature activation+7.01e-03

Pos logit contributions
ang+0.027
irling+0.025
ee+0.023
irl+0.023
ug+0.022
Neg logit contributions
 ranger-0.020
 department-0.018
 hinges-0.018
 bleed-0.018
 bru-0.018
Token butterfly
Token ID35113
Feature activation+2.92e-03

Pos logit contributions
ang+0.001
irling+0.001
ee+0.001
irl+0.001
ug+0.001
Neg logit contributions
 bleed-0.001
 ranger-0.001
 bru-0.001
 hinges-0.001
 sob-0.001
Token,
Token ID11
Feature activation+6.27e-03

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.005
Neg logit contributions
 ranger-0.005
 bleed-0.004
 sob-0.004
 bru-0.004
 bud-0.004
Token tw
Token ID665
Feature activation+6.50e-02

Pos logit contributions
ang+0.013
irling+0.011
ee+0.011
ug+0.011
ash+0.010
Neg logit contributions
 bleed-0.011
 sob-0.010
 ranger-0.010
 bru-0.010
 hinges-0.010
Tokenirling
Token ID24297
Feature activation+5.61e-02

Pos logit contributions
ang+0.066
ash+0.064
 sorts+0.060
icated+0.059
angles+0.058
Neg logit contributions
onna-0.138
 bleed-0.136
 bud-0.134
 aid-0.130
 sob-0.130
Token and
Token ID290
Feature activation+1.99e-02

Pos logit contributions
ang+0.094
irling+0.082
ee+0.077
irl+0.076
ug+0.073
Neg logit contributions
 ranger-0.113
 bleed-0.110
 sob-0.107
 bru-0.107
 hinges-0.106
Token spir
Token ID9158
Feature activation+1.28e-01

Pos logit contributions
ang+0.037
ee+0.032
ug+0.032
ash+0.031
irl+0.030
Neg logit contributions
 ranger-0.036
 sob-0.036
 bleed-0.036
 bru-0.034
 bud-0.034
Tokenaling
Token ID4272
Feature activation+3.11e-02

Pos logit contributions
irling+0.261
ang+0.255
irl+0.241
ading+0.215
icing+0.212
Neg logit contributions
 bleed-0.186
 department-0.171
 ranger-0.169
 recovered-0.169
 bud-0.168
Token in
Token ID287
Feature activation+1.97e-02

Pos logit contributions
ang+0.059
irling+0.053
ee+0.051
irl+0.049
ug+0.048
Neg logit contributions
 ranger-0.055
 bleed-0.055
 bru-0.053
 hinges-0.053
 sob-0.053
Token the
Token ID262
Feature activation+2.81e-02

Pos logit contributions
ang+0.033
irling+0.029
ee+0.027
irl+0.027
ug+0.026
Neg logit contributions
 ranger-0.040
 bleed-0.039
 bru-0.039
 hinges-0.039
 sob-0.038
Token air
Token ID1633
Feature activation-6.42e-03

Pos logit contributions
ang+0.054
irling+0.048
irl+0.045
ee+0.044
ug+0.044
Neg logit contributions
 ranger-0.051
 bru-0.050
 hinges-0.050
 bleed-0.050
 sob-0.049
Token.
Token ID13
Feature activation-1.07e-02

Pos logit contributions
 ranger+0.008
 bleed+0.008
 bru+0.008
 hinges+0.008
 sob+0.008
Neg logit contributions
ang-0.015
irling-0.014
ee-0.014
irl-0.013
ug-0.013
Token the
Token ID262
Feature activation+1.25e-02

Pos logit contributions
ang+0.074
irling+0.067
ee+0.062
irl+0.062
ug+0.060
Neg logit contributions
 ranger-0.067
 bru-0.063
 bleed-0.062
 sob-0.062
 hinges-0.062
Token chest
Token ID7721
Feature activation+1.57e-02

Pos logit contributions
ang+0.027
ee+0.024
irling+0.024
irl+0.023
ading+0.022
Neg logit contributions
 bleed-0.019
 bru-0.018
 ranger-0.017
 department-0.017
 sob-0.017
Token were
Token ID547
Feature activation+3.57e-02

Pos logit contributions
ang+0.031
irling+0.028
ee+0.027
irl+0.026
ug+0.025
Neg logit contributions
 bleed-0.027
 bru-0.026
 ranger-0.025
 hinges-0.025
 sob-0.025
Token the
Token ID262
Feature activation+1.79e-02

Pos logit contributions
ang+0.065
irling+0.058
ee+0.056
irl+0.054
ug+0.052
Neg logit contributions
 ranger-0.066
 bleed-0.065
 hinges-0.064
 bru-0.064
 sob-0.064
Token most
Token ID749
Feature activation+3.31e-02

Pos logit contributions
ang+0.029
irling+0.025
ee+0.025
irl+0.023
ug+0.022
Neg logit contributions
 bleed-0.037
 ranger-0.036
 bru-0.035
 hinges-0.035
 sob-0.035
Token wond
Token ID3135
Feature activation+1.27e-01

Pos logit contributions
ang+0.054
irling+0.047
ee+0.046
irl+0.045
ug+0.043
Neg logit contributions
 ranger-0.068
 hinges-0.065
 bleed-0.065
 sob-0.065
 bru-0.065
Tokeneful
Token ID13839
Feature activation+2.06e-02

Pos logit contributions
ang+0.117
ee+0.097
irling+0.097
irl+0.088
ug+0.086
Neg logit contributions
 ranger-0.296
keeping-0.294
 bru-0.288
 recovered-0.282
 keeper-0.278
Token toys
Token ID14958
Feature activation+2.08e-02

Pos logit contributions
ang+0.045
irling+0.041
ee+0.039
irl+0.038
ug+0.037
Neg logit contributions
 ranger-0.032
 hinges-0.030
 bleed-0.030
 bru-0.030
 sob-0.030
Token the
Token ID262
Feature activation+1.49e-02

Pos logit contributions
ang+0.040
irling+0.036
ee+0.035
irl+0.033
ug+0.033
Neg logit contributions
 bleed-0.036
 bru-0.035
 ranger-0.035
 hinges-0.035
 sob-0.034
Token little
Token ID1310
Feature activation+2.17e-02

Pos logit contributions
ang+0.027
irling+0.024
ee+0.024
irl+0.023
ug+0.022
Neg logit contributions
 bleed-0.027
 ranger-0.027
 bru-0.026
 hinges-0.026
 sob-0.026
Token girl
Token ID2576
Feature activation+2.25e-03

Pos logit contributions
ang+0.043
irling+0.037
ee+0.037
irl+0.035
ug+0.035
Neg logit contributions
 bleed-0.036
 ranger-0.036
 bru-0.035
 hinges-0.035
 sob-0.035
Token asleep
Token ID16039
Feature activation-9.32e-04

Pos logit contributions
ang+0.003
irling+0.003
ug+0.003
oming+0.003
ee+0.003
Neg logit contributions
 Something-0.003
 ranger-0.003
 aid-0.003
 mercy-0.003
 bru-0.003
Token,
Token ID11
Feature activation+2.82e-03

Pos logit contributions
 bleed+0.002
 department+0.002
 ranger+0.002
 hinges+0.002
 bru+0.002
Neg logit contributions
ang-0.002
irling-0.002
ee-0.001
irl-0.001
ug-0.001
Token dreaming
Token ID28368
Feature activation+1.32e-02

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.005
Neg logit contributions
 ranger-0.005
 bleed-0.004
 bru-0.004
 hinges-0.004
 sob-0.004
Token of
Token ID286
Feature activation+2.88e-02

Pos logit contributions
ang+0.014
irling+0.012
ee+0.011
ug+0.010
irl+0.010
Neg logit contributions
 bleed-0.033
 ranger-0.033
 hinges-0.033
 bru-0.032
 sob-0.032
Token icy
Token ID30284
Feature activation+3.68e-02

Pos logit contributions
ang+0.050
irling+0.044
ee+0.042
irl+0.041
ug+0.039
Neg logit contributions
 ranger-0.056
 bleed-0.056
 bru-0.055
 hinges-0.055
 sob-0.054
Token pian
Token ID43923
Feature activation+1.26e-01

Pos logit contributions
ang+0.066
irling+0.058
ee+0.056
irl+0.054
ug+0.052
Neg logit contributions
 ranger-0.070
 bleed-0.069
 bru-0.068
 hinges-0.067
 sob-0.067
Tokenos
Token ID418
Feature activation+4.56e-02

Pos logit contributions
ang+0.239
irling+0.223
irl+0.221
ading+0.205
ee+0.203
Neg logit contributions
onna-0.184
 department-0.182
 bleed-0.179
 recovered-0.175
 ranger-0.174
Token and
Token ID290
Feature activation+4.13e-02

Pos logit contributions
ang+0.091
irling+0.083
ee+0.078
irl+0.078
ading+0.073
Neg logit contributions
 bleed-0.074
 ranger-0.074
 department-0.070
 sob-0.070
 hinges-0.070
Token big
Token ID1263
Feature activation+1.19e-02

Pos logit contributions
ang+0.078
irling+0.070
ee+0.067
irl+0.065
ug+0.062
Neg logit contributions
 ranger-0.073
 bleed-0.073
 bru-0.071
 hinges-0.071
 sob-0.070
Token piles
Token ID30935
Feature activation+3.45e-02

Pos logit contributions
ang+0.024
irling+0.021
ee+0.021
irl+0.020
ug+0.020
Neg logit contributions
 bleed-0.019
 ranger-0.019
 hinges-0.018
 bru-0.018
 sob-0.018
Token of
Token ID286
Feature activation+4.11e-02

Pos logit contributions
ang+0.069
irling+0.061
ee+0.060
ug+0.057
irl+0.057
Neg logit contributions
 bleed-0.057
 bru-0.057
 ranger-0.056
 sob-0.055
 hinges-0.055
Token with
Token ID351
Feature activation+4.82e-02

Pos logit contributions
ang+0.007
irling+0.006
ee+0.006
irl+0.006
ug+0.005
Neg logit contributions
 ranger-0.005
 bleed-0.005
 bru-0.005
 hinges-0.005
 department-0.005
Token your
Token ID534
Feature activation-2.80e-02

Pos logit contributions
ang+0.090
irling+0.077
ee+0.073
ash+0.073
ug+0.071
Neg logit contributions
 ranger-0.082
 bud-0.081
 mercy-0.080
 sob-0.080
 hinges-0.080
Token new
Token ID649
Feature activation-9.03e-03

Pos logit contributions
 ranger+0.050
 bleed+0.050
 bru+0.049
 hinges+0.048
 sob+0.048
Neg logit contributions
ang-0.053
irling-0.047
ee-0.046
irl-0.044
ug-0.042
Token coat
Token ID13209
Feature activation-1.76e-03

Pos logit contributions
 bleed+0.013
 ranger+0.013
 bru+0.012
 hinges+0.012
 sob+0.012
Neg logit contributions
ang-0.021
irling-0.019
ee-0.018
irl-0.018
ug-0.017
Token."
Token ID526
Feature activation-8.99e-03

Pos logit contributions
 ranger+0.003
 bleed+0.003
 bru+0.003
 hinges+0.003
 sob+0.003
Neg logit contributions
ang-0.004
irling-0.003
ee-0.003
irl-0.003
ug-0.003
Token Sil
Token ID4243
Feature activation+1.26e-01

Pos logit contributions
 ranger+0.013
 bleed+0.012
 bru+0.012
 hinges+0.012
 sob+0.012
Neg logit contributions
ang-0.020
irling-0.019
ee-0.018
irl-0.018
ug-0.017
Tokenvia
Token ID8869
Feature activation+2.86e-02

Pos logit contributions
irling+0.208
ang+0.207
irl+0.202
ading+0.186
ee+0.183
Neg logit contributions
 sob-0.198
 ranger-0.194
 aid-0.194
 keeper-0.194
 hinges-0.191
Token smiled
Token ID13541
Feature activation-9.21e-04

Pos logit contributions
ang+0.067
irling+0.061
ee+0.059
irl+0.058
ug+0.057
Neg logit contributions
 ranger-0.038
 sob-0.037
 bleed-0.036
 bru-0.036
 hinges-0.035
Token and
Token ID290
Feature activation+1.01e-02

Pos logit contributions
 bleed+0.002
 bru+0.002
 ranger+0.002
 hinges+0.002
 injuries+0.002
Neg logit contributions
ang-0.001
irling-0.001
ee-0.001
ug-0.001
aw-0.001
Token said
Token ID531
Feature activation-4.08e-02

Pos logit contributions
ang+0.025
irling+0.023
ash+0.022
ug+0.022
ee+0.022
Neg logit contributions
 sob-0.012
 ranger-0.012
 bleed-0.011
 bru-0.011
 recovered-0.011
Token,
Token ID11
Feature activation-3.56e-02

Pos logit contributions
 bleed+0.068
 ranger+0.068
 bru+0.067
 hinges+0.066
 sob+0.065
Neg logit contributions
ang-0.081
irling-0.074
ee-0.071
irl-0.069
ug-0.067
Token and
Token ID290
Feature activation+4.17e-02

Pos logit contributions
ang+0.114
irling+0.093
ug+0.090
ee+0.090
ading+0.088
Neg logit contributions
 sob-0.077
 hinges-0.077
 bleed-0.077
keeping-0.075
 bru-0.074
Token you
Token ID345
Feature activation+1.53e-02

Pos logit contributions
ang+0.092
ee+0.078
irling+0.078
ash+0.077
ug+0.075
Neg logit contributions
 bleed-0.069
 bru-0.065
 sob-0.063
 ranger-0.063
 hinges-0.062
Token love
Token ID1842
Feature activation+1.82e-02

Pos logit contributions
ang+0.037
irling+0.034
ee+0.032
irl+0.031
ug+0.031
Neg logit contributions
 bleed-0.020
 bru-0.019
 ranger-0.018
 sob-0.018
 hinges-0.018
Token it
Token ID340
Feature activation+3.34e-03

Pos logit contributions
ang+0.035
irling+0.030
ee+0.029
ading+0.028
ug+0.028
Neg logit contributions
 bru-0.033
 bleed-0.032
 ranger-0.031
 hinges-0.031
 sob-0.031
Token,
Token ID11
Feature activation+1.88e-02

Pos logit contributions
ang+0.007
irling+0.006
ee+0.006
irl+0.006
ug+0.006
Neg logit contributions
 bleed-0.005
 ranger-0.005
 hinges-0.005
 sob-0.005
 bru-0.005
Token Sil
Token ID4243
Feature activation+1.26e-01

Pos logit contributions
ang+0.042
irling+0.038
ee+0.036
irl+0.035
ug+0.034
Neg logit contributions
 ranger-0.029
 bru-0.028
 sob-0.028
 bleed-0.028
 hinges-0.027
Tokenvia
Token ID8869
Feature activation+3.44e-02

Pos logit contributions
ang+0.217
irling+0.196
irl+0.185
ee+0.178
ading+0.174
Neg logit contributions
onna-0.208
 aid-0.207
ache-0.203
 sob-0.193
 department-0.190
Token!
Token ID0
Feature activation+3.32e-02

Pos logit contributions
ang+0.081
irling+0.074
ee+0.072
irl+0.071
ading+0.069
Neg logit contributions
 ranger-0.043
 bleed-0.043
 bru-0.043
 sob-0.041
 hinges-0.041
Token Come
Token ID7911
Feature activation+2.22e-03

Pos logit contributions
ang+0.071
irling+0.070
irl+0.064
ee+0.063
ading+0.062
Neg logit contributions
 ranger-0.048
 Something-0.048
 bru-0.047
 bleed-0.046
 sob-0.045
Token now
Token ID783
Feature activation-1.70e-02

Pos logit contributions
ang+0.004
irling+0.004
ee+0.004
irl+0.004
ug+0.004
Neg logit contributions
 ranger-0.004
 bru-0.003
 bleed-0.003
 department-0.003
 sob-0.003
Token,
Token ID11
Feature activation+9.01e-03

Pos logit contributions
 ranger+0.030
 bleed+0.030
 bru+0.030
 sob+0.029
 hinges+0.029
Neg logit contributions
ang-0.032
irling-0.028
ee-0.027
irl-0.026
ug-0.026
Token green
Token ID4077
Feature activation+2.82e-02

Pos logit contributions
ang+0.101
irling+0.089
ee+0.085
irl+0.083
ug+0.080
Neg logit contributions
 ranger-0.105
 bleed-0.104
 bru-0.103
 hinges-0.102
 sob-0.101
Token and
Token ID290
Feature activation+3.31e-02

Pos logit contributions
ang+0.051
irling+0.045
ee+0.045
irl+0.042
ug+0.042
Neg logit contributions
 ranger-0.053
 bleed-0.052
 bru-0.050
 sob-0.050
 department-0.048
Token a
Token ID257
Feature activation-1.24e-02

Pos logit contributions
ang+0.064
irling+0.057
ee+0.055
irl+0.053
ug+0.052
Neg logit contributions
 ranger-0.059
 bleed-0.058
 hinges-0.057
 sob-0.056
 bru-0.056
Token long
Token ID890
Feature activation+4.86e-03

Pos logit contributions
 bleed+0.023
 ranger+0.023
 hinges+0.022
 bru+0.022
 sob+0.022
Neg logit contributions
ang-0.022
irling-0.019
ee-0.019
irl-0.018
ug-0.017
Token,
Token ID11
Feature activation+5.79e-03

Pos logit contributions
ang+0.010
irling+0.009
ee+0.009
irl+0.008
ug+0.008
Neg logit contributions
 ranger-0.008
 bleed-0.008
 sob-0.008
 bru-0.008
 hinges-0.008
Token spir
Token ID9158
Feature activation+1.25e-01

Pos logit contributions
ang+0.010
irling+0.009
ee+0.009
irl+0.009
ug+0.008
Neg logit contributions
 ranger-0.011
 bleed-0.011
 bru-0.011
 hinges-0.011
 sob-0.010
Tokenalling
Token ID9221
Feature activation+2.89e-02

Pos logit contributions
ang+0.241
irling+0.237
irl+0.221
ee+0.199
ading+0.194
Neg logit contributions
 bleed-0.205
 ranger-0.195
 department-0.194
 hinges-0.193
 bru-0.192
Token tail
Token ID7894
Feature activation+7.35e-03

Pos logit contributions
ang+0.056
irling+0.049
ee+0.048
irl+0.046
ug+0.046
Neg logit contributions
 bleed-0.050
 bru-0.049
 ranger-0.049
 sob-0.049
 hinges-0.049
Token.
Token ID13
Feature activation+1.93e-03

Pos logit contributions
ang+0.015
irling+0.014
ee+0.014
ug+0.013
irl+0.013
Neg logit contributions
 bru-0.011
 bleed-0.011
 ranger-0.011
 hinges-0.011
 sob-0.011
Token The
Token ID383
Feature activation+1.42e-02

Pos logit contributions
ang+0.004
irling+0.004
ee+0.004
irl+0.004
ug+0.004
Neg logit contributions
 bleed-0.003
 ranger-0.003
 bru-0.003
 hinges-0.003
 sob-0.003
Token princess
Token ID21752
Feature activation+3.62e-02

Pos logit contributions
ang+0.027
irling+0.024
ee+0.024
irl+0.023
ug+0.022
Neg logit contributions
 bleed-0.025
 ranger-0.024
 bru-0.024
 hinges-0.024
 sob-0.023
Token there
Token ID612
Feature activation+4.34e-02

Pos logit contributions
ang+0.101
irling+0.094
ee+0.090
irl+0.088
ug+0.085
Neg logit contributions
 ranger-0.058
 sob-0.055
 bru-0.054
 bleed-0.053
 hinges-0.052
Token was
Token ID373
Feature activation+3.01e-02

Pos logit contributions
ang+0.097
irling+0.089
ee+0.085
irl+0.084
ug+0.081
Neg logit contributions
 ranger-0.062
 bleed-0.061
 bru-0.060
 sob-0.059
 hinges-0.059
Token a
Token ID257
Feature activation+1.02e-02

Pos logit contributions
ang+0.076
irling+0.069
ash+0.064
angles+0.064
ee+0.064
Neg logit contributions
 bleed-0.034
 sob-0.034
 bru-0.032
 ranger-0.031
 hinges-0.030
Token girl
Token ID2576
Feature activation+1.25e-02

Pos logit contributions
ang+0.022
irling+0.022
ash+0.021
ading+0.021
angles+0.020
Neg logit contributions
 ranger-0.015
 bru-0.013
 sob-0.013
 aid-0.013
 cough-0.012
Token named
Token ID3706
Feature activation+4.03e-02

Pos logit contributions
ang+0.023
irling+0.022
irl+0.021
ading+0.021
ee+0.020
Neg logit contributions
 ranger-0.022
 sob-0.022
 bru-0.021
 aid-0.021
 hinges-0.020
Token Sil
Token ID4243
Feature activation+1.25e-01

Pos logit contributions
ang+0.102
irling+0.094
ee+0.090
irl+0.087
ug+0.087
Neg logit contributions
 ranger-0.049
 sob-0.049
 hinges-0.048
 bleed-0.046
 bru-0.045
Tokenvia
Token ID8869
Feature activation+3.46e-02

Pos logit contributions
ang+0.210
irling+0.192
irl+0.180
ee+0.174
ading+0.174
Neg logit contributions
 aid-0.216
onna-0.210
 ranger-0.207
 sob-0.199
 hinges-0.199
Token.
Token ID13
Feature activation+1.76e-02

Pos logit contributions
ang+0.086
irling+0.079
ee+0.077
irl+0.075
ug+0.073
Neg logit contributions
 ranger-0.041
 bleed-0.040
 bru-0.039
 hinges-0.039
 sob-0.039
Token She
Token ID1375
Feature activation+2.22e-02

Pos logit contributions
ang+0.042
irling+0.038
ee+0.037
irl+0.036
ug+0.036
Neg logit contributions
 bleed-0.023
 ranger-0.022
 bru-0.022
 sob-0.022
 hinges-0.021
Token was
Token ID373
Feature activation+3.28e-02

Pos logit contributions
ang+0.040
irling+0.034
ee+0.032
ash+0.032
ug+0.032
Neg logit contributions
 ranger-0.042
 bleed-0.042
 sob-0.042
 bru-0.041
 hinges-0.040
Token 3
Token ID513
Feature activation+4.35e-02

Pos logit contributions
ang+0.082
irling+0.079
ash+0.077
angles+0.077
ee+0.075
Neg logit contributions
 sob-0.038
 bleed-0.037
 cough-0.034
 ranger-0.034
 bru-0.034
Token same
Token ID976
Feature activation-5.42e-03

Pos logit contributions
ang+0.002
irling+0.002
ee+0.002
irl+0.002
ug+0.002
Neg logit contributions
 ranger-0.002
 bleed-0.002
 bru-0.002
 department-0.002
 keeper-0.002
Token.
Token ID13
Feature activation-6.29e-03

Pos logit contributions
 ranger+0.009
 bleed+0.009
 bru+0.009
 aid+0.008
 hinges+0.008
Neg logit contributions
ang-0.011
irling-0.010
ee-0.009
irl-0.009
ash-0.009
Token
Token ID198
Feature activation-2.31e-04

Pos logit contributions
 ranger+0.008
 bleed+0.008
 bru+0.008
 hinges+0.008
 sob+0.008
Neg logit contributions
ang-0.015
irling-0.014
ee-0.014
irl-0.013
ug-0.013
Token
Token ID198
Feature activation-1.03e-03

Pos logit contributions
 bleed+0.000
 ranger+0.000
 bru+0.000
 hinges+0.000
 sob+0.000
Neg logit contributions
ang-0.001
irling-0.001
ee-0.001
irl-0.000
ug-0.000
TokenAs
Token ID1722
Feature activation+9.94e-03

Pos logit contributions
 ranger+0.002
 bleed+0.002
 hinges+0.002
 sob+0.002
 bud+0.002
Neg logit contributions
ang-0.002
irling-0.002
ee-0.002
ash-0.002
irl-0.002
Token Sil
Token ID4243
Feature activation+1.25e-01

Pos logit contributions
ang+0.022
irling+0.021
ash+0.020
ee+0.020
irl+0.020
Neg logit contributions
 ranger-0.014
 bud-0.013
 hinges-0.013
 keeper-0.013
 bleed-0.012
Tokenvia
Token ID8869
Feature activation+2.90e-02

Pos logit contributions
irling+0.201
ang+0.199
irl+0.192
ading+0.179
ee+0.173
Neg logit contributions
 aid-0.192
ache-0.191
 ranger-0.190
 hinges-0.188
 sob-0.188
Token was
Token ID373
Feature activation+3.66e-03

Pos logit contributions
ang+0.064
irling+0.062
irl+0.058
ash+0.055
ee+0.055
Neg logit contributions
 ranger-0.039
 recovered-0.038
 sob-0.035
 aid-0.035
 bru-0.035
Token about
Token ID546
Feature activation-2.90e-02

Pos logit contributions
ang+0.006
ee+0.006
angles+0.005
ash+0.005
irling+0.005
Neg logit contributions
 sob-0.007
 bleed-0.007
 ranger-0.007
 bru-0.007
 aid-0.007
Token to
Token ID284
Feature activation-3.57e-02

Pos logit contributions
 bleed+0.045
ache+0.040
 hinges+0.039
 sob+0.038
 ranger+0.038
Neg logit contributions
ang-0.059
irling-0.056
irl-0.053
ee-0.052
 sorts-0.051
Token jump
Token ID4391
Feature activation+3.48e-02

Pos logit contributions
 ranger+0.052
 bleed+0.052
 bru+0.051
 hinges+0.050
 sob+0.049
Neg logit contributions
ang-0.079
irling-0.072
ee-0.069
irl-0.067
ug-0.066
Token grass
Token ID8701
Feature activation+2.71e-02

Pos logit contributions
ang+0.095
irling+0.084
ee+0.084
irl+0.079
ug+0.078
Neg logit contributions
 ranger-0.080
 bru-0.080
 bleed-0.080
 hinges-0.078
 sob-0.077
Token and
Token ID290
Feature activation+3.86e-02

Pos logit contributions
ang+0.057
irling+0.051
ee+0.050
irl+0.048
ug+0.047
Neg logit contributions
 bru-0.042
 bleed-0.042
 ranger-0.041
 hinges-0.041
 sob-0.040
Token rye
Token ID47553
Feature activation+7.25e-02

Pos logit contributions
ang+0.076
irling+0.067
ee+0.065
irl+0.063
ug+0.061
Neg logit contributions
 ranger-0.066
 bleed-0.066
 bru-0.065
 hinges-0.064
 sob-0.064
Token grass
Token ID8701
Feature activation+3.22e-02

Pos logit contributions
ang+0.148
irling+0.130
ee+0.127
irl+0.122
ug+0.120
Neg logit contributions
 bleed-0.119
 bru-0.119
 hinges-0.118
 ranger-0.118
 sob-0.115
Token and
Token ID290
Feature activation+4.17e-02

Pos logit contributions
ang+0.075
irling+0.067
ee+0.066
irl+0.063
ug+0.062
Neg logit contributions
 bru-0.043
 bleed-0.043
 ranger-0.043
 hinges-0.042
 sob-0.042
Token al
Token ID435
Feature activation+1.25e-01

Pos logit contributions
ang+0.080
irling+0.072
ee+0.069
irl+0.067
ug+0.065
Neg logit contributions
 ranger-0.073
 bleed-0.072
 bru-0.071
 hinges-0.070
 sob-0.070
Tokenf
Token ID69
Feature activation+5.73e-02

Pos logit contributions
ang+0.179
irling+0.153
ee+0.148
irl+0.144
ading+0.135
Neg logit contributions
onna-0.245
 recovered-0.238
 bleed-0.238
 ranger-0.234
ache-0.234
Tokenalf
Token ID1604
Feature activation+2.86e-02

Pos logit contributions
ang+0.096
irling+0.085
irl+0.080
ee+0.078
ug+0.072
Neg logit contributions
 bleed-0.110
 ranger-0.109
 bru-0.107
 hinges-0.107
 department-0.106
Tokena
Token ID64
Feature activation+1.96e-02

Pos logit contributions
ang+0.058
irling+0.052
ee+0.050
irl+0.049
ug+0.047
Neg logit contributions
 ranger-0.046
 bleed-0.045
 bru-0.044
 sob-0.044
 hinges-0.043
Token and
Token ID290
Feature activation+2.78e-02

Pos logit contributions
ang+0.043
irling+0.039
ee+0.038
irl+0.037
ug+0.036
Neg logit contributions
 ranger-0.028
 bleed-0.028
 bru-0.027
 hinges-0.027
 sob-0.027
Token cl
Token ID537
Feature activation+7.21e-02

Pos logit contributions
ang+0.054
irling+0.049
ee+0.047
irl+0.046
ug+0.044
Neg logit contributions
 ranger-0.047
 bleed-0.047
 bru-0.046
 hinges-0.046
 sob-0.045
Token forth
Token ID6071
Feature activation+1.32e-02

Pos logit contributions
ang+0.047
irling+0.041
ee+0.041
ug+0.040
ash+0.039
Neg logit contributions
 sob-0.040
 bleed-0.039
 ranger-0.039
 bru-0.038
 recovered-0.036
Token in
Token ID287
Feature activation+2.18e-02

Pos logit contributions
ang+0.025
irling+0.023
ee+0.022
irl+0.021
ug+0.021
Neg logit contributions
 ranger-0.023
 bleed-0.023
 bru-0.022
 hinges-0.022
 sob-0.022
Token the
Token ID262
Feature activation+3.24e-02

Pos logit contributions
ang+0.033
irling+0.029
ee+0.028
irl+0.027
ug+0.025
Neg logit contributions
 ranger-0.047
 bleed-0.046
 hinges-0.045
 bru-0.045
 sob-0.045
Token sky
Token ID6766
Feature activation-1.50e-03

Pos logit contributions
ang+0.062
irling+0.055
ee+0.051
irl+0.051
ug+0.050
Neg logit contributions
 bru-0.058
 bleed-0.058
 ranger-0.058
 sob-0.057
 hinges-0.056
Token,
Token ID11
Feature activation+8.05e-03

Pos logit contributions
 ranger+0.002
 bleed+0.002
 bru+0.002
 hinges+0.002
 sob+0.002
Neg logit contributions
ang-0.004
irling-0.003
ee-0.003
irl-0.003
ug-0.003
Token spir
Token ID9158
Feature activation+1.24e-01

Pos logit contributions
ang+0.017
ee+0.015
irling+0.014
ug+0.014
ash+0.014
Neg logit contributions
 bleed-0.013
 ranger-0.013
 sob-0.013
 bru-0.013
 hinges-0.012
Tokenalling
Token ID9221
Feature activation+2.85e-02

Pos logit contributions
irling+0.212
ang+0.201
irl+0.192
icing+0.165
ading+0.163
Neg logit contributions
 bleed-0.231
 department-0.216
 recovered-0.214
 bru-0.214
 ranger-0.211
Token higher
Token ID2440
Feature activation+3.00e-02

Pos logit contributions
ang+0.051
irling+0.045
ee+0.043
irl+0.041
ug+0.041
Neg logit contributions
 bleed-0.054
 bru-0.053
 ranger-0.052
 hinges-0.052
 sob-0.052
Token and
Token ID290
Feature activation+3.70e-02

Pos logit contributions
ang+0.068
irling+0.062
ee+0.060
irl+0.059
ug+0.057
Neg logit contributions
 ranger-0.042
 bleed-0.041
 hinges-0.040
 bru-0.040
 sob-0.039
Token higher
Token ID2440
Feature activation+3.25e-02

Pos logit contributions
ang+0.079
ee+0.071
ug+0.070
irling+0.070
irl+0.067
Neg logit contributions
 ranger-0.061
 bleed-0.057
 bru-0.056
 keeper-0.055
 sob-0.055
Token.
Token ID13
Feature activation+1.32e-03

Pos logit contributions
ang+0.079
irling+0.072
ee+0.071
irl+0.069
ug+0.067
Neg logit contributions
 ranger-0.040
 bleed-0.040
 bru-0.039
 hinges-0.038
 sob-0.038
Token looks
Token ID3073
Feature activation+4.80e-02

Pos logit contributions
ang+0.071
irling+0.065
ug+0.059
ading+0.059
irl+0.058
Neg logit contributions
 bleed-0.044
 sob-0.044
 bru-0.042
 ranger-0.042
 Something-0.040
Token wonderful
Token ID7932
Feature activation+1.20e-02

Pos logit contributions
ang+0.107
irling+0.088
ug+0.088
ee+0.086
ites+0.082
Neg logit contributions
 bleed-0.073
 ranger-0.072
 bru-0.072
 hinges-0.071
 sob-0.070
Token on
Token ID319
Feature activation+3.67e-02

Pos logit contributions
ang+0.024
irling+0.021
ee+0.020
irl+0.020
ug+0.019
Neg logit contributions
 ranger-0.020
 bleed-0.019
 hinges-0.019
 bru-0.019
 department-0.019
Token you
Token ID345
Feature activation+8.83e-04

Pos logit contributions
ang+0.062
irling+0.053
ash+0.050
ee+0.048
irl+0.048
Neg logit contributions
 Help-0.065
 ranger-0.065
 bleed-0.063
 hinges-0.063
 bud-0.063
Token,
Token ID11
Feature activation+1.96e-02

Pos logit contributions
ang+0.002
irling+0.002
irl+0.002
ug+0.002
ee+0.002
Neg logit contributions
 ranger-0.001
 bud-0.001
 department-0.001
 hinges-0.001
 sob-0.001
Token Sil
Token ID4243
Feature activation+1.23e-01

Pos logit contributions
ang+0.041
irling+0.039
irl+0.036
ee+0.034
ash+0.034
Neg logit contributions
 ranger-0.033
 bud-0.031
 department-0.031
 keeper-0.030
 hinges-0.029
Tokenvia
Token ID8869
Feature activation+3.56e-02

Pos logit contributions
ang+0.202
irling+0.191
irl+0.186
ee+0.173
ading+0.172
Neg logit contributions
ache-0.199
 aid-0.198
onna-0.193
 ranger-0.185
 keeper-0.183
Token!
Token ID0
Feature activation+3.70e-02

Pos logit contributions
ang+0.076
irling+0.069
irl+0.066
ee+0.065
ading+0.064
Neg logit contributions
 ranger-0.052
 bleed-0.049
 bru-0.049
 keeper-0.048
 sob-0.048
Token You
Token ID921
Feature activation+4.23e-02

Pos logit contributions
irling+0.075
ang+0.075
irl+0.069
ading+0.065
ee+0.064
Neg logit contributions
 Something-0.059
 ranger-0.057
 department-0.055
 Help-0.055
 bru-0.054
Token look
Token ID804
Feature activation+4.29e-02

Pos logit contributions
ang+0.101
irling+0.091
ug+0.083
irl+0.082
ee+0.082
Neg logit contributions
 bleed-0.058
 sob-0.056
 ranger-0.055
 bru-0.055
 bud-0.054
Token so
Token ID523
Feature activation+3.90e-02

Pos logit contributions
ang+0.092
irling+0.079
ug+0.077
irl+0.075
ee+0.075
Neg logit contributions
 bleed-0.064
 ranger-0.064
 hinges-0.063
 sob-0.063
 bru-0.062
Token She
Token ID1375
Feature activation-7.72e-03

Pos logit contributions
 bleed+0.010
 hinges+0.009
 bru+0.009
 injuries+0.009
 recovered+0.009
Neg logit contributions
ang-0.016
irling-0.014
ee-0.014
irl-0.013
ug-0.013
Token told
Token ID1297
Feature activation-6.04e-03

Pos logit contributions
 ranger+0.013
 bleed+0.013
 bru+0.012
 hinges+0.012
 sob+0.012
Neg logit contributions
ang-0.016
irling-0.014
ee-0.014
irl-0.013
ug-0.013
Token him
Token ID683
Feature activation-5.71e-02

Pos logit contributions
 ranger+0.010
 bleed+0.010
 hinges+0.010
 bru+0.010
 department+0.010
Neg logit contributions
ang-0.011
irling-0.010
ee-0.010
irl-0.010
ug-0.009
Token how
Token ID703
Feature activation-3.90e-03

Pos logit contributions
 bleed+0.106
 bru+0.105
 hinges+0.103
 ranger+0.101
 injuries+0.100
Neg logit contributions
ang-0.098
irling-0.090
ee-0.087
irl-0.086
ug-0.081
Token she
Token ID673
Feature activation-1.53e-03

Pos logit contributions
 bleed+0.005
 ranger+0.005
 hinges+0.005
 bru+0.005
 sob+0.005
Neg logit contributions
ang-0.009
irling-0.008
ee-0.008
irl-0.008
ug-0.008
Token se
Token ID384
Feature activation+1.22e-01

Pos logit contributions
 ranger+0.002
 bleed+0.002
 sob+0.002
 bru+0.002
 hinges+0.002
Neg logit contributions
ang-0.003
irling-0.003
ee-0.003
irl-0.003
ug-0.003
Tokenwed
Token ID19103
Feature activation+4.34e-02

Pos logit contributions
ang+0.253
irling+0.227
ee+0.207
ading+0.205
ug+0.199
Neg logit contributions
 bleed-0.190
 bru-0.178
 sob-0.176
 ranger-0.174
 recovered-0.172
Token,
Token ID11
Feature activation-4.64e-03

Pos logit contributions
ang+0.079
irling+0.071
ee+0.065
irl+0.064
ading+0.062
Neg logit contributions
 bleed-0.080
 ranger-0.080
 bru-0.078
 hinges-0.078
 sob-0.077
Token glued
Token ID39783
Feature activation+5.52e-02

Pos logit contributions
 ranger+0.007
 bleed+0.007
 bru+0.007
 sob+0.007
 hinges+0.007
Neg logit contributions
ang-0.010
irling-0.009
ee-0.009
irl-0.008
ug-0.008
Token,
Token ID11
Feature activation+6.34e-03

Pos logit contributions
ang+0.098
irling+0.087
ee+0.082
irl+0.080
ug+0.077
Neg logit contributions
 ranger-0.105
 bleed-0.104
 bru-0.102
 hinges-0.102
 sob-0.101
Token and
Token ID290
Feature activation+1.68e-02

Pos logit contributions
ang+0.013
irling+0.012
ee+0.012
irl+0.011
ug+0.011
Neg logit contributions
 ranger-0.010
 bleed-0.010
 bru-0.010
 hinges-0.010
 sob-0.010
Token own
Token ID898
Feature activation-1.90e-02

Pos logit contributions
 ranger+0.051
 bleed+0.050
 hinges+0.049
 department+0.049
 bru+0.048
Neg logit contributions
ang-0.063
irling-0.056
ee-0.055
irl-0.054
ug-0.051
Token two
Token ID734
Feature activation+5.94e-05

Pos logit contributions
 ranger+0.028
 bleed+0.028
 hinges+0.027
 bru+0.027
 department+0.027
Neg logit contributions
ang-0.041
irling-0.038
ee-0.037
irl-0.036
ug-0.035
Token hands
Token ID2832
Feature activation+7.59e-03

Pos logit contributions
ang+0.000
irling+0.000
ee+0.000
irl+0.000
ug+0.000
Neg logit contributions
 ranger-0.000
 bleed-0.000
 sob-0.000
 bru-0.000
 hinges-0.000
Token.
Token ID13
Feature activation+1.50e-02

Pos logit contributions
ang+0.017
irling+0.015
ee+0.015
irl+0.015
ug+0.014
Neg logit contributions
 ranger-0.011
 bleed-0.011
 bru-0.010
 hinges-0.010
 sob-0.010
Token She
Token ID1375
Feature activation+9.29e-03

Pos logit contributions
ang+0.036
irling+0.033
ee+0.032
irl+0.031
ug+0.030
Neg logit contributions
 ranger-0.019
 bleed-0.019
 bru-0.018
 hinges-0.018
 sob-0.018
Token se
Token ID384
Feature activation+1.21e-01

Pos logit contributions
ang+0.021
irling+0.018
ee+0.018
ug+0.018
ash+0.017
Neg logit contributions
 sob-0.013
 ranger-0.013
 bru-0.013
 bleed-0.013
 aid-0.012
Tokenwed
Token ID19103
Feature activation+5.37e-02

Pos logit contributions
ang+0.243
irling+0.220
irl+0.198
ee+0.197
ading+0.193
Neg logit contributions
 bleed-0.199
 bru-0.192
 sob-0.188
 ranger-0.188
 recovered-0.184
Token it
Token ID340
Feature activation-1.30e-02

Pos logit contributions
ang+0.102
irling+0.095
ee+0.086
irl+0.086
ading+0.082
Neg logit contributions
 ranger-0.092
 bleed-0.092
 bru-0.090
 sob-0.089
 hinges-0.089
Token with
Token ID351
Feature activation+4.69e-02

Pos logit contributions
 ranger+0.026
 bleed+0.026
 bru+0.025
 hinges+0.025
 sob+0.025
Neg logit contributions
ang-0.022
irling-0.019
ee-0.018
irl-0.018
ug-0.017
Token bright
Token ID6016
Feature activation+4.12e-02

Pos logit contributions
ang+0.103
irling+0.093
ee+0.088
irl+0.086
ug+0.085
Neg logit contributions
 bleed-0.071
 ranger-0.070
 bru-0.069
 sob-0.068
 hinges-0.068
Token colors
Token ID7577
Feature activation+3.09e-02

Pos logit contributions
ang+0.079
irling+0.069
ee+0.066
irl+0.064
ug+0.062
Neg logit contributions
 bleed-0.072
 hinges-0.071
 bru-0.071
 ranger-0.071
 sob-0.070
Token and
Token ID290
Feature activation+1.50e-04

Pos logit contributions
 bleed+0.048
 bru+0.048
 ranger+0.047
 hinges+0.046
 sob+0.046
Neg logit contributions
ang-0.057
ee-0.050
irling-0.049
ug-0.047
irl-0.046
Token it
Token ID340
Feature activation+1.18e-02

Pos logit contributions
ang+0.000
irling+0.000
ee+0.000
irl+0.000
ug+0.000
Neg logit contributions
 ranger-0.000
 bleed-0.000
 sob-0.000
 hinges-0.000
 bru-0.000
Token fit
Token ID4197
Feature activation+3.01e-02

Pos logit contributions
ang+0.022
irling+0.020
ee+0.020
irl+0.020
ug+0.018
Neg logit contributions
 ranger-0.020
 bru-0.020
 bleed-0.020
 hinges-0.020
 injuries-0.019
Token perfectly
Token ID7138
Feature activation+1.32e-02

Pos logit contributions
ang+0.047
irling+0.040
ee+0.039
irl+0.037
ug+0.036
Neg logit contributions
 ranger-0.064
 bleed-0.063
 bru-0.063
 hinges-0.062
 sob-0.062
Token!
Token ID0
Feature activation+2.92e-03

Pos logit contributions
ang+0.032
irling+0.030
ee+0.029
irl+0.028
ug+0.028
Neg logit contributions
 ranger-0.016
 bleed-0.016
 bru-0.015
 hinges-0.015
 sob-0.015
Token Sil
Token ID4243
Feature activation+1.21e-01

Pos logit contributions
ang+0.007
irling+0.006
ee+0.006
irl+0.006
ading+0.006
Neg logit contributions
 ranger-0.004
 hinges-0.004
 bleed-0.004
 sob-0.004
 department-0.004
Tokenvia
Token ID8869
Feature activation+2.65e-02

Pos logit contributions
ang+0.213
irling+0.207
irl+0.197
ading+0.188
ee+0.178
Neg logit contributions
 hinges-0.174
 aid-0.174
ache-0.171
 sob-0.169
driver-0.166
Token was
Token ID373
Feature activation+1.68e-02

Pos logit contributions
ang+0.065
irling+0.059
ading+0.056
ug+0.056
ash+0.055
Neg logit contributions
 sob-0.031
 bru-0.029
 ranger-0.029
 bleed-0.028
 recovered-0.028
Token so
Token ID523
Feature activation+1.46e-02

Pos logit contributions
ang+0.033
ee+0.029
irling+0.028
ug+0.028
irl+0.027
Neg logit contributions
 sob-0.029
 bleed-0.029
 ranger-0.029
 bru-0.028
 hinges-0.027
Token happy
Token ID3772
Feature activation-1.96e-02

Pos logit contributions
ang+0.030
ash+0.027
irling+0.027
ug+0.026
ee+0.026
Neg logit contributions
 bleed-0.023
 sob-0.022
 ranger-0.022
 bru-0.021
 keeper-0.020
Token she
Token ID673
Feature activation+5.12e-03

Pos logit contributions
 bleed+0.034
 bru+0.033
 hinges+0.033
 ranger+0.032
 sob+0.031
Neg logit contributions
ang-0.039
irling-0.035
ee-0.034
irl-0.032
ug-0.031
Token his
Token ID465
Feature activation-4.72e-02

Pos logit contributions
 ranger+0.090
 bleed+0.090
 bru+0.089
 hinges+0.088
 sob+0.087
Neg logit contributions
ang-0.075
irling-0.066
ee-0.063
irl-0.061
ug-0.058
Token mistake
Token ID7457
Feature activation-3.77e-02

Pos logit contributions
 ranger+0.082
 bleed+0.081
 recovered+0.078
keeping+0.078
 bru+0.077
Neg logit contributions
irl-0.080
ang-0.076
irling-0.076
ee-0.072
ading-0.067
Token and
Token ID290
Feature activation-9.31e-03

Pos logit contributions
 hinges+0.067
 bleed+0.066
 bru+0.066
 ranger+0.065
 injuries+0.063
Neg logit contributions
ang-0.070
ee-0.061
irling-0.061
ug-0.058
ash-0.055
Token gave
Token ID2921
Feature activation+3.16e-02

Pos logit contributions
 bleed+0.014
 ranger+0.014
 bru+0.014
 hinges+0.014
 department+0.013
Neg logit contributions
ang-0.019
irling-0.018
ee-0.017
irl-0.017
ug-0.016
Token his
Token ID465
Feature activation-5.81e-02

Pos logit contributions
ang+0.069
irling+0.062
ee+0.060
irl+0.059
ug+0.057
Neg logit contributions
 ranger-0.048
 bleed-0.047
 bru-0.046
 hinges-0.046
 sob-0.046
Token sister
Token ID6621
Feature activation-1.04e-01

Pos logit contributions
 bleed+0.116
 ranger+0.110
 bru+0.108
onna+0.108
keeping+0.107
Neg logit contributions
ang-0.085
irl-0.078
ee-0.076
irling-0.075
 these-0.069
Token a
Token ID257
Feature activation-6.26e-02

Pos logit contributions
 ranger+0.183
 bleed+0.182
 hinges+0.181
 bru+0.180
 department+0.179
Neg logit contributions
ang-0.190
ee-0.170
irling-0.169
irl-0.161
ash-0.154
Token turn
Token ID1210
Feature activation-1.69e-02

Pos logit contributions
keeping+0.138
edly+0.130
 hinges+0.127
 ranger+0.122
 injuries+0.121
Neg logit contributions
ee-0.089
ang-0.083
 sorts-0.074
irl-0.072
irling-0.062
Token with
Token ID351
Feature activation+2.31e-02

Pos logit contributions
 hinges+0.027
 bru+0.026
 ranger+0.026
 bleed+0.026
 department+0.025
Neg logit contributions
ang-0.035
ee-0.032
irling-0.030
ug-0.030
ash-0.029
Token the
Token ID262
Feature activation-2.28e-02

Pos logit contributions
ang+0.041
irling+0.037
ee+0.037
irl+0.036
ug+0.034
Neg logit contributions
 ranger-0.042
 hinges-0.041
 department-0.040
 bleed-0.040
 bru-0.040
Token toy
Token ID13373
Feature activation-3.50e-02

Pos logit contributions
 ranger+0.048
 hinges+0.045
 bleed+0.045
 bru+0.045
 mercy+0.044
Neg logit contributions
ang-0.029
ee-0.029
ading-0.027
irl-0.027
irling-0.027
Token cry
Token ID3960
Feature activation-4.92e-02

Pos logit contributions
 department+0.086
 ranger+0.081
 hinges+0.079
edly+0.078
keeper+0.077
Neg logit contributions
irl-0.149
ang-0.146
irling-0.146
ee-0.146
ading-0.135
Token.
Token ID13
Feature activation-4.42e-02

Pos logit contributions
 ranger+0.096
keeping+0.092
onna+0.090
 hinges+0.089
 department+0.088
Neg logit contributions
ang-0.080
ug-0.070
ing-0.065
irling-0.064
ee-0.064
Token His
Token ID2399
Feature activation-4.52e-02

Pos logit contributions
 bleed+0.079
 department+0.075
 injuries+0.073
 recovered+0.073
 mercy+0.071
Neg logit contributions
ang-0.076
aw-0.063
ug-0.060
irling-0.060
 Now-0.058
Token mom
Token ID1995
Feature activation-3.47e-02

Pos logit contributions
 Something+0.082
 recovered+0.079
 bleed+0.078
 opener+0.077
 department+0.077
Neg logit contributions
ang-0.074
irl-0.066
irling-0.064
ee-0.060
utes-0.055
Token heard
Token ID2982
Feature activation-2.41e-02

Pos logit contributions
 bleed+0.055
 hinges+0.055
 ranger+0.054
onna+0.053
 bru+0.052
Neg logit contributions
ang-0.071
ee-0.063
irling-0.063
ug-0.058
irl-0.057
Token him
Token ID683
Feature activation-1.04e-01

Pos logit contributions
 hinges+0.050
 bleed+0.049
 ranger+0.048
 department+0.048
 bru+0.048
Neg logit contributions
ang-0.035
ee-0.031
irling-0.030
aw-0.029
ug-0.028
Token and
Token ID290
Feature activation-4.91e-02

Pos logit contributions
onna+0.210
 ranger+0.207
 department+0.206
 hinges+0.205
 bru+0.204
Neg logit contributions
ang-0.154
ee-0.130
irling-0.126
ug-0.119
ash-0.112
Token came
Token ID1625
Feature activation-2.36e-02

Pos logit contributions
 hinges+0.083
 bleed+0.082
 bru+0.080
onna+0.080
 ranger+0.079
Neg logit contributions
ang-0.092
irling-0.079
ee-0.078
ug-0.075
irl-0.071
Token to
Token ID284
Feature activation-5.57e-02

Pos logit contributions
 bleed+0.032
 bru+0.031
 ranger+0.030
ache+0.030
 Something+0.029
Neg logit contributions
ang-0.055
irling-0.049
ee-0.048
ug-0.047
ading-0.046
Token see
Token ID766
Feature activation-1.91e-02

Pos logit contributions
ache+0.136
edly+0.129
 bleed+0.128
 hinges+0.128
comb+0.126
Neg logit contributions
ang-0.052
 these-0.048
ee-0.047
 us-0.043
 those-0.040
Token what
Token ID644
Feature activation-1.37e-02

Pos logit contributions
 bleed+0.049
 hinges+0.048
 ranger+0.048
onna+0.047
 bru+0.047
Neg logit contributions
ang-0.020
ee-0.015
irling-0.014
aw-0.014
 these-0.012
Token grandma
Token ID49890
Feature activation-3.87e-02

Pos logit contributions
 bleed+0.033
 ranger+0.032
 bru+0.032
 hinges+0.031
 sob+0.031
Neg logit contributions
ang-0.048
irling-0.043
ee-0.042
irl-0.041
ug-0.040
Token's
Token ID338
Feature activation-2.73e-02

Pos logit contributions
 bleed+0.062
 ranger+0.062
 bru+0.060
 hinges+0.060
 sob+0.059
Neg logit contributions
ang-0.080
irling-0.071
ee-0.069
irl-0.067
ug-0.065
Token house
Token ID2156
Feature activation-3.17e-02

Pos logit contributions
 bleed+0.050
onna+0.046
 bru+0.046
 recovered+0.046
 ranger+0.045
Neg logit contributions
ang-0.050
irling-0.044
ee-0.042
ug-0.040
irl-0.040
Token for
Token ID329
Feature activation-4.18e-03

Pos logit contributions
 bleed+0.056
 bru+0.053
 sob+0.051
 hinges+0.051
 ranger+0.049
Neg logit contributions
ang-0.059
irling-0.051
ee-0.050
ug-0.050
ash-0.048
Token safe
Token ID3338
Feature activation-7.41e-03

Pos logit contributions
 ranger+0.007
 bleed+0.007
 department+0.007
 bru+0.007
 hinges+0.007
Neg logit contributions
ang-0.008
irling-0.007
irl-0.007
ee-0.007
ug-0.006
Tokenkeeping
Token ID19934
Feature activation-1.03e-01

Pos logit contributions
 bleed+0.010
 bru+0.010
onna+0.010
 Something+0.010
 department+0.010
Neg logit contributions
ang-0.015
ug-0.014
irling-0.014
ee-0.013
ash-0.013
Token.
Token ID13
Feature activation-8.07e-03

Pos logit contributions
 ranger+0.167
 department+0.166
 bru+0.163
 bleed+0.162
 mercy+0.161
Neg logit contributions
ang-0.185
irling-0.174
ee-0.166
ug-0.161
irl-0.157
Token
Token ID220
Feature activation-3.69e-03

Pos logit contributions
 bleed+0.012
 ranger+0.012
 bru+0.012
 hinges+0.012
 sob+0.012
Neg logit contributions
ang-0.017
irling-0.016
ee-0.015
irl-0.015
ug-0.014
Token
Token ID198
Feature activation-7.85e-03

Pos logit contributions
 ranger+0.005
 bleed+0.005
 bru+0.004
 hinges+0.004
 sob+0.004
Neg logit contributions
ang-0.009
irling-0.008
ee-0.008
irl-0.008
ug-0.008
Token
Token ID198
Feature activation+6.00e-04

Pos logit contributions
 bleed+0.010
 ranger+0.010
 bru+0.009
 hinges+0.009
 sob+0.009
Neg logit contributions
ang-0.019
irling-0.017
ee-0.017
irl-0.017
ug-0.016
TokenWhen
Token ID2215
Feature activation-2.77e-03

Pos logit contributions
ang+0.001
irling+0.001
ee+0.001
irl+0.001
ug+0.001
Neg logit contributions
 ranger-0.001
 bleed-0.001
 bru-0.001
 hinges-0.001
 sob-0.001
Token it
Token ID340
Feature activation-4.53e-02

Pos logit contributions
ang+0.104
irling+0.096
irl+0.086
ee+0.082
ading+0.081
Neg logit contributions
 ranger-0.128
 bleed-0.126
 bru-0.124
 sob-0.123
 hinges-0.123
Token to
Token ID284
Feature activation-3.55e-02

Pos logit contributions
 bru+0.061
 bleed+0.059
 sob+0.058
 injuries+0.054
 ranger+0.053
Neg logit contributions
ang-0.096
ee-0.094
ash-0.087
aw-0.086
ug-0.085
Token me
Token ID502
Feature activation-5.01e-02

Pos logit contributions
 ranger+0.063
 bru+0.061
 bleed+0.061
 sob+0.061
 hinges+0.060
Neg logit contributions
ang-0.069
irling-0.062
ee-0.058
irl-0.057
ug-0.055
Token for
Token ID329
Feature activation+1.87e-03

Pos logit contributions
 bleed+0.076
 ranger+0.075
 bru+0.074
 hinges+0.072
 sob+0.072
Neg logit contributions
ang-0.108
irling-0.096
ee-0.095
irl-0.092
ug-0.090
Token safe
Token ID3338
Feature activation-3.25e-03

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ug+0.003
Neg logit contributions
 ranger-0.004
 bleed-0.004
 bru-0.003
 hinges-0.003
 sob-0.003
Tokenkeeping
Token ID19934
Feature activation-1.02e-01

Pos logit contributions
 bleed+0.004
 bru+0.004
 sob+0.004
 Something+0.004
 department+0.004
Neg logit contributions
ang-0.007
ug-0.006
ee-0.006
irling-0.006
irl-0.006
Token.
Token ID13
Feature activation+3.09e-05

Pos logit contributions
 department+0.165
 hinges+0.165
 Something+0.164
 bleed+0.162
 bru+0.162
Neg logit contributions
ang-0.185
irling-0.172
ee-0.164
irl-0.157
ug-0.153
Token He
Token ID679
Feature activation-8.79e-03

Pos logit contributions
ang+0.000
irling+0.000
ee+0.000
irl+0.000
ug+0.000
Neg logit contributions
 bleed-0.000
 ranger-0.000
 bru-0.000
 hinges-0.000
 sob-0.000
Token told
Token ID1297
Feature activation-1.64e-03

Pos logit contributions
 hinges+0.016
 bleed+0.015
 ranger+0.015
 bru+0.015
 injuries+0.015
Neg logit contributions
ang-0.016
ee-0.014
irling-0.014
irl-0.014
ug-0.013
Token me
Token ID502
Feature activation-4.27e-02

Pos logit contributions
 bleed+0.003
 ranger+0.003
 bru+0.003
 hinges+0.003
 sob+0.003
Neg logit contributions
ang-0.003
irling-0.002
ee-0.002
irl-0.002
ug-0.002
Token to
Token ID284
Feature activation-5.43e-02

Pos logit contributions
 bleed+0.084
 ranger+0.083
 bru+0.082
 hinges+0.081
 sob+0.081
Neg logit contributions
ang-0.070
irling-0.062
ee-0.061
irl-0.058
ug-0.057
Token
Token ID198
Feature activation-8.59e-03

Pos logit contributions
 bleed+0.052
 department+0.052
 bud+0.052
 hinges+0.051
ache+0.050
Neg logit contributions
ang-0.048
ug-0.040
ee-0.039
irling-0.036
aw-0.035
Token
Token ID198
Feature activation-1.02e-02

Pos logit contributions
 bleed+0.013
 hinges+0.012
 ranger+0.012
 bru+0.012
 department+0.012
Neg logit contributions
ang-0.019
irling-0.017
ee-0.017
irl-0.016
ug-0.016
TokenHis
Token ID6653
Feature activation-5.07e-02

Pos logit contributions
 bleed+0.020
 bru+0.020
 hinges+0.020
 department+0.019
 ranger+0.019
Neg logit contributions
ang-0.018
irling-0.015
ee-0.015
ug-0.014
irl-0.014
Token mom
Token ID1995
Feature activation-2.78e-02

Pos logit contributions
 Something+0.086
keeping+0.085
 bleed+0.082
 recovered+0.081
onna+0.081
Neg logit contributions
ang-0.092
irl-0.084
irling-0.081
ee-0.079
ug-0.073
Token heard
Token ID2982
Feature activation-1.96e-02

Pos logit contributions
 bleed+0.040
onna+0.039
 ranger+0.039
 hinges+0.039
 bru+0.038
Neg logit contributions
ang-0.061
ee-0.055
irling-0.053
ug-0.051
aw-0.051
Token him
Token ID683
Feature activation-1.01e-01

Pos logit contributions
 hinges+0.041
 ranger+0.040
 department+0.040
 bleed+0.039
 bru+0.039
Neg logit contributions
ang-0.029
ee-0.026
irling-0.025
ug-0.023
irl-0.023
Token and
Token ID290
Feature activation-4.35e-02

Pos logit contributions
 department+0.186
 ranger+0.184
 hinges+0.180
 bru+0.177
onna+0.176
Neg logit contributions
ang-0.175
ee-0.151
irling-0.145
ug-0.141
irl-0.133
Token came
Token ID1625
Feature activation-2.12e-02

Pos logit contributions
 hinges+0.073
 bleed+0.072
onna+0.071
 bru+0.071
 ranger+0.070
Neg logit contributions
ang-0.083
ee-0.071
irling-0.070
ug-0.068
irl-0.063
Token to
Token ID284
Feature activation-5.50e-02

Pos logit contributions
 bleed+0.026
 bru+0.025
 ranger+0.025
ache+0.024
 Something+0.023
Neg logit contributions
ang-0.053
irling-0.047
ee-0.046
ug-0.045
ading-0.044
Token help
Token ID1037
Feature activation-2.03e-02

Pos logit contributions
ache+0.133
 hinges+0.128
edly+0.127
 department+0.123
onna+0.123
Neg logit contributions
ang-0.054
ee-0.049
 these-0.048
 us-0.044
rous-0.040
Token him
Token ID683
Feature activation-9.76e-02

Pos logit contributions
 department+0.039
 hinges+0.038
ache+0.038
 bru+0.037
 ranger+0.037
Neg logit contributions
ang-0.031
ee-0.030
ug-0.028
ie-0.026
ading-0.025
Token was
Token ID373
Feature activation-1.38e-02

Pos logit contributions
 bleed+0.049
 hinges+0.049
 bru+0.047
 ranger+0.046
 injuries+0.046
Neg logit contributions
ang-0.048
irling-0.044
ee-0.042
irl-0.040
ug-0.038
Token angry
Token ID7954
Feature activation-3.68e-02

Pos logit contributions
 hinges+0.028
 ranger+0.028
 bleed+0.028
 bru+0.028
 department+0.027
Neg logit contributions
ang-0.022
irling-0.020
ee-0.018
irl-0.018
ug-0.017
Token and
Token ID290
Feature activation-2.37e-02

Pos logit contributions
 bleed+0.064
 hinges+0.062
 bru+0.062
 ranger+0.060
 sob+0.059
Neg logit contributions
ang-0.072
irling-0.063
ee-0.061
ug-0.058
irl-0.057
Token asked
Token ID1965
Feature activation-4.98e-02

Pos logit contributions
 hinges+0.036
 bleed+0.036
 ranger+0.036
 bru+0.036
 department+0.035
Neg logit contributions
ang-0.049
irling-0.045
ee-0.042
irl-0.041
ug-0.040
Token her
Token ID607
Feature activation-7.75e-02

Pos logit contributions
 bleed+0.108
 ranger+0.103
 hinges+0.101
 bru+0.100
onna+0.098
Neg logit contributions
ang-0.073
irling-0.064
ee-0.060
ug-0.060
irl-0.058
Token daughter
Token ID4957
Feature activation-1.01e-01

Pos logit contributions
 bleed+0.195
 bru+0.184
 hinges+0.182
 ranger+0.177
 choke+0.175
Neg logit contributions
ang-0.087
irling-0.068
ee-0.067
ug-0.061
ading-0.060
Token why
Token ID1521
Feature activation-4.52e-02

Pos logit contributions
 bleed+0.219
 hinges+0.207
 bru+0.207
 ranger+0.201
 injuries+0.195
Neg logit contributions
ang-0.136
irling-0.116
ee-0.112
ash-0.110
 these-0.109
Token she
Token ID673
Feature activation-4.58e-02

Pos logit contributions
 hinges+0.074
 ranger+0.073
 bleed+0.073
 bru+0.072
 department+0.071
Neg logit contributions
ang-0.091
irling-0.083
ee-0.080
irl-0.079
ug-0.075
Token had
Token ID550
Feature activation-2.59e-03

Pos logit contributions
 ranger+0.075
 hinges+0.075
 bleed+0.074
 bru+0.073
 department+0.071
Neg logit contributions
ang-0.091
irling-0.083
ee-0.082
irl-0.079
ug-0.074
Token done
Token ID1760
Feature activation-3.44e-02

Pos logit contributions
 ranger+0.004
 bleed+0.004
 hinges+0.004
 bru+0.004
 department+0.004
Neg logit contributions
ang-0.005
irling-0.005
ee-0.005
irl-0.005
ug-0.004
Token something
Token ID1223
Feature activation-4.28e-02

Pos logit contributions
 ranger+0.066
 bleed+0.066
 hinges+0.065
 bru+0.065
 sob+0.063
Neg logit contributions
ang-0.059
irling-0.053
ee-0.050
irl-0.049
ug-0.046
Token He
Token ID679
Feature activation-3.41e-02

Pos logit contributions
 department+0.059
 bleed+0.059
 injuries+0.058
comb+0.058
 bud+0.056
Neg logit contributions
ang-0.047
aw-0.037
ug-0.035
 Everyone-0.033
ee-0.033
Token tried
Token ID3088
Feature activation-4.63e-02

Pos logit contributions
 bleed+0.056
 hinges+0.054
 bru+0.054
 ranger+0.054
 department+0.052
Neg logit contributions
ang-0.069
irling-0.062
ee-0.059
irl-0.058
ug-0.055
Token to
Token ID284
Feature activation-7.11e-02

Pos logit contributions
comb+0.076
 hinges+0.074
atively+0.071
edly+0.070
ache+0.070
Neg logit contributions
ang-0.071
 these-0.068
irling-0.066
 all-0.065
irl-0.065
Token push
Token ID4574
Feature activation+1.89e-02

Pos logit contributions
 department+0.111
 ranger+0.110
 hinges+0.110
 bru+0.107
onna+0.106
Neg logit contributions
ang-0.139
irl-0.129
irling-0.129
ee-0.127
ading-0.119
Token the
Token ID262
Feature activation-2.95e-02

Pos logit contributions
ang+0.036
ee+0.031
ug+0.029
irling+0.029
irl+0.028
Neg logit contributions
 bleed-0.033
 ranger-0.033
 bru-0.033
 hinges-0.033
 department-0.032
Token girl
Token ID2576
Feature activation-9.94e-02

Pos logit contributions
 hinges+0.053
 ranger+0.052
onna+0.052
 department+0.052
 bleed+0.051
Neg logit contributions
ang-0.053
ee-0.046
irling-0.043
irl-0.041
is-0.039
Token away
Token ID1497
Feature activation-4.74e-02

Pos logit contributions
 bru+0.183
 hinges+0.180
 department+0.180
onna+0.180
 ranger+0.179
Neg logit contributions
ang-0.174
ee-0.144
irling-0.139
ash-0.134
ug-0.134
Token,
Token ID11
Feature activation-2.95e-02

Pos logit contributions
 bru+0.096
 bleed+0.094
 recovered+0.091
 hinges+0.091
keeping+0.091
Neg logit contributions
ang-0.075
ug-0.060
ee-0.059
irling-0.054
il-0.052
Token but
Token ID475
Feature activation-1.64e-02

Pos logit contributions
 hinges+0.045
 ranger+0.045
 bleed+0.044
 department+0.044
 bru+0.042
Neg logit contributions
ang-0.061
irling-0.055
ee-0.051
irl-0.050
ug-0.049
Token she
Token ID673
Feature activation-2.01e-02

Pos logit contributions
 hinges+0.029
 bleed+0.029
 bru+0.028
 department+0.028
 ranger+0.028
Neg logit contributions
ang-0.030
irling-0.026
ee-0.026
irl-0.025
ug-0.024
Token was
Token ID373
Feature activation-4.41e-03

Pos logit contributions
 hinges+0.034
 bleed+0.033
 bru+0.032
 ranger+0.032
 department+0.031
Neg logit contributions
ang-0.041
ee-0.036
irling-0.035
irl-0.034
ug-0.033
Token
Token ID198
Feature activation-1.21e-02

Pos logit contributions
 injuries+0.075
 bleed+0.075
 hinges+0.074
 recovered+0.073
 department+0.071
Neg logit contributions
ang-0.046
ee-0.035
ug-0.031
ch-0.029
aw-0.029
Token
Token ID198
Feature activation-1.04e-02

Pos logit contributions
 bleed+0.019
 hinges+0.019
 department+0.019
 ranger+0.019
 bru+0.019
Neg logit contributions
ang-0.025
ee-0.022
irling-0.022
irl-0.021
ug-0.021
TokenHis
Token ID6653
Feature activation-4.40e-02

Pos logit contributions
 bleed+0.017
 bru+0.016
 ranger+0.016
 hinges+0.016
 sob+0.016
Neg logit contributions
ang-0.022
irling-0.019
ee-0.019
irl-0.018
ug-0.018
Token mom
Token ID1995
Feature activation-2.21e-02

Pos logit contributions
keeping+0.074
 recovered+0.072
 bleed+0.072
onna+0.070
 Something+0.070
Neg logit contributions
ang-0.081
ee-0.073
irling-0.071
irl-0.070
ug-0.063
Token hears
Token ID23180
Feature activation-1.83e-02

Pos logit contributions
 ranger+0.028
 hinges+0.028
 bleed+0.027
onna+0.026
 bru+0.026
Neg logit contributions
ang-0.051
ee-0.048
irling-0.047
irl-0.044
aw-0.044
Token him
Token ID683
Feature activation-9.90e-02

Pos logit contributions
 hinges+0.037
 bleed+0.037
 ranger+0.037
 bru+0.036
 department+0.036
Neg logit contributions
ang-0.027
ee-0.025
irling-0.023
ug-0.022
aw-0.022
Token and
Token ID290
Feature activation-4.85e-02

Pos logit contributions
 ranger+0.195
 hinges+0.190
 department+0.190
 bru+0.188
onna+0.188
Neg logit contributions
ang-0.156
ee-0.136
irling-0.130
ug-0.121
ash-0.118
Token runs
Token ID4539
Feature activation-1.57e-02

Pos logit contributions
 hinges+0.077
 ranger+0.077
 bru+0.076
 bleed+0.076
 injuries+0.074
Neg logit contributions
ang-0.098
irling-0.086
ee-0.085
irl-0.081
ug-0.078
Token to
Token ID284
Feature activation-4.86e-02

Pos logit contributions
onna+0.025
 bru+0.025
 bleed+0.025
 Something+0.024
ache+0.024
Neg logit contributions
ang-0.029
ee-0.025
ash-0.025
 around-0.024
 these-0.024
Token his
Token ID465
Feature activation-5.67e-02

Pos logit contributions
ache+0.126
 hinges+0.119
 department+0.118
edly+0.117
 bleed+0.115
Neg logit contributions
ang-0.044
ee-0.040
 these-0.036
irl-0.033
ading-0.033
Token room
Token ID2119
Feature activation-6.68e-02

Pos logit contributions
 bleed+0.135
 recovered+0.132
onna+0.132
ache+0.131
 bru+0.127
Neg logit contributions
ang-0.062
irling-0.044
ee-0.040
irl-0.037
ading-0.035
Token cry
Token ID3960
Feature activation-4.54e-02

Pos logit contributions
 ranger+0.136
 department+0.129
 hinges+0.126
 keeper+0.118
 injuries+0.117
Neg logit contributions
ang-0.166
ee-0.160
irl-0.146
irling-0.144
ug-0.138
Token.
Token ID13
Feature activation-2.53e-02

Pos logit contributions
 ranger+0.071
 bru+0.070
 hinges+0.067
 bleed+0.066
onna+0.064
Neg logit contributions
ang-0.092
ee-0.082
ug-0.081
irling-0.080
ash-0.076
Token His
Token ID2399
Feature activation-5.09e-02

Pos logit contributions
 bleed+0.041
 hinges+0.040
 department+0.039
 ranger+0.039
 recovered+0.038
Neg logit contributions
ang-0.052
ee-0.046
irling-0.044
irl-0.043
ug-0.042
Token mom
Token ID1995
Feature activation-2.63e-02

Pos logit contributions
 bleed+0.084
 recovered+0.079
 ranger+0.079
onna+0.078
 hinges+0.077
Neg logit contributions
ang-0.090
ee-0.087
irl-0.085
irling-0.083
ug-0.072
Token heard
Token ID2982
Feature activation-1.71e-02

Pos logit contributions
 ranger+0.039
 bleed+0.039
 hinges+0.039
 bru+0.038
onna+0.037
Neg logit contributions
ang-0.056
ee-0.051
irling-0.050
irl-0.046
ug-0.046
Token him
Token ID683
Feature activation-9.90e-02

Pos logit contributions
 ranger+0.033
 hinges+0.033
 bleed+0.033
 bru+0.032
 department+0.032
Neg logit contributions
ang-0.028
ee-0.026
irling-0.025
irl-0.024
ug-0.023
Token and
Token ID290
Feature activation-4.10e-02

Pos logit contributions
 ranger+0.195
 hinges+0.190
 bru+0.189
 department+0.188
onna+0.187
Neg logit contributions
ang-0.155
ee-0.138
irling-0.129
ug-0.124
ading-0.116
Token ran
Token ID4966
Feature activation-2.18e-02

Pos logit contributions
 bleed+0.066
 hinges+0.065
 bru+0.065
 ranger+0.064
 injuries+0.062
Neg logit contributions
ang-0.083
ee-0.072
irling-0.072
ug-0.068
irl-0.066
Token to
Token ID284
Feature activation-5.74e-02

Pos logit contributions
 bru+0.035
 bleed+0.035
ache+0.034
onna+0.032
 Something+0.032
Neg logit contributions
ang-0.039
ash-0.036
ee-0.036
 these-0.035
 around-0.034
Token the
Token ID262
Feature activation-2.53e-02

Pos logit contributions
ache+0.143
 hinges+0.135
 department+0.135
edly+0.133
 bleed+0.130
Neg logit contributions
ang-0.059
ee-0.056
irl-0.049
 these-0.049
irling-0.045
Token dining
Token ID17423
Feature activation-4.03e-02

Pos logit contributions
 bleed+0.052
 recovered+0.050
ache+0.049
 bru+0.049
onna+0.047
Neg logit contributions
ang-0.029
ee-0.026
irling-0.025
 these-0.024
ading-0.024
Token and
Token ID290
Feature activation+3.07e-03

Pos logit contributions
 hinges+0.098
 bleed+0.098
 bru+0.096
 aid+0.094
onna+0.094
Neg logit contributions
ang-0.073
ee-0.060
ug-0.058
irling-0.057
ash-0.056
Token said
Token ID531
Feature activation-5.51e-02

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.005
Neg logit contributions
 hinges-0.005
 bru-0.005
 bleed-0.005
 injuries-0.005
 department-0.005
Token sorry
Token ID7926
Feature activation-6.10e-02

Pos logit contributions
 hinges+0.108
keeping+0.107
 bleed+0.103
 ranger+0.102
 aid+0.102
Neg logit contributions
ang-0.085
irling-0.084
irl-0.077
ug-0.075
ee-0.071
Token to
Token ID284
Feature activation-4.12e-02

Pos logit contributions
 bru+0.122
keeping+0.120
 bleed+0.120
 distress+0.120
 injuries+0.119
Neg logit contributions
ing-0.083
ang-0.077
ash-0.074
 at-0.071
ee-0.068
Token his
Token ID465
Feature activation-5.10e-02

Pos logit contributions
 ranger+0.070
 bleed+0.070
 hinges+0.069
 bru+0.068
 sob+0.067
Neg logit contributions
ang-0.079
irling-0.071
ee-0.068
irl-0.068
ug-0.065
Token sister
Token ID6621
Feature activation-9.83e-02

Pos logit contributions
 bleed+0.097
keeping+0.094
 recovered+0.094
 bru+0.093
onna+0.093
Neg logit contributions
ang-0.082
irl-0.073
irling-0.071
ee-0.069
ading-0.065
Token.
Token ID13
Feature activation-1.52e-02

Pos logit contributions
 bleed+0.190
 hinges+0.184
 bru+0.184
onna+0.178
 injuries+0.167
Neg logit contributions
ang-0.142
ash-0.126
 these-0.124
aw-0.119
irling-0.118
Token They
Token ID1119
Feature activation-2.19e-04

Pos logit contributions
 bleed+0.025
 injuries+0.025
 bru+0.024
 hinges+0.023
ache+0.023
Neg logit contributions
ang-0.031
irling-0.026
ee-0.026
ug-0.025
ash-0.025
Token started
Token ID2067
Feature activation-2.01e-02

Pos logit contributions
 bleed+0.000
 ranger+0.000
 bru+0.000
 hinges+0.000
 aid+0.000
Neg logit contributions
ang-0.000
irling-0.000
ee-0.000
irl-0.000
ug-0.000
Token to
Token ID284
Feature activation-4.60e-02

Pos logit contributions
 hinges+0.024
 ranger+0.023
 bleed+0.023
 bud+0.022
ache+0.021
Neg logit contributions
ang-0.045
irling-0.044
irl-0.041
ee-0.040
ading-0.039
Token play
Token ID711
Feature activation+1.85e-02

Pos logit contributions
 ranger+0.069
edly+0.069
 hinges+0.067
onna+0.067
 department+0.066
Neg logit contributions
irl-0.083
irling-0.081
ang-0.079
ee-0.076
 sorts-0.073
Token.
Token ID13
Feature activation-3.15e-02

Pos logit contributions
 hinges+0.061
keeping+0.061
 ranger+0.061
 severe+0.061
 aid+0.060
Neg logit contributions
ang-0.054
ee-0.045
aw-0.045
irling-0.045
ash-0.044
Token She
Token ID1375
Feature activation-4.56e-02

Pos logit contributions
 bleed+0.048
 department+0.046
 injuries+0.046
 recovered+0.046
 ranger+0.045
Neg logit contributions
ang-0.064
irling-0.057
aw-0.057
ee-0.056
irl-0.054
Token sn
Token ID3013
Feature activation+5.07e-02

Pos logit contributions
 ranger+0.073
 bleed+0.073
 bru+0.072
 sob+0.071
 hinges+0.071
Neg logit contributions
ang-0.094
irling-0.085
ee-0.082
irl-0.079
ug-0.078
Tokenatches
Token ID20981
Feature activation+1.66e-02

Pos logit contributions
ang+0.084
irling+0.074
ee+0.071
irl+0.068
ug+0.065
Neg logit contributions
 ranger-0.104
 bleed-0.104
 hinges-0.102
 bru-0.102
 sob-0.100
Token her
Token ID607
Feature activation-5.88e-02

Pos logit contributions
ang+0.034
irling+0.032
irl+0.029
ee+0.029
ug+0.028
Neg logit contributions
 bleed-0.027
 ranger-0.027
 bru-0.027
 sob-0.027
 hinges-0.026
Token journal
Token ID3989
Feature activation-9.79e-02

Pos logit contributions
 ranger+0.097
 bleed+0.092
 bru+0.091
 department+0.089
onna+0.089
Neg logit contributions
ang-0.115
ee-0.107
irl-0.104
irling-0.103
angles-0.096
Token from
Token ID422
Feature activation-1.43e-02

Pos logit contributions
 bru+0.202
 mercy+0.197
 injuries+0.195
 department+0.194
 Something+0.193
Neg logit contributions
ang-0.129
ee-0.114
irl-0.102
 at-0.099
irling-0.097
Token Tom
Token ID4186
Feature activation-6.07e-02

Pos logit contributions
 ranger+0.025
 bleed+0.025
 hinges+0.025
 bru+0.025
 department+0.024
Neg logit contributions
ang-0.027
irling-0.024
ee-0.023
irl-0.023
ug-0.021
Token and
Token ID290
Feature activation-2.42e-02

Pos logit contributions
 ranger+0.113
 bleed+0.111
 bru+0.110
 hinges+0.110
 department+0.108
Neg logit contributions
ang-0.109
ee-0.095
irling-0.094
irl-0.088
ug-0.087
Token pushes
Token ID20070
Feature activation+1.60e-02

Pos logit contributions
 ranger+0.039
 bleed+0.039
 bru+0.038
 hinges+0.038
 sob+0.037
Neg logit contributions
ang-0.049
irling-0.045
ee-0.043
irl-0.042
ug-0.040
Token him
Token ID683
Feature activation-8.92e-02

Pos logit contributions
ang+0.032
irling+0.029
ee+0.028
irl+0.027
ug+0.026
Neg logit contributions
 ranger-0.027
 bleed-0.026
 bru-0.026
 hinges-0.026
 sob-0.025
Tokenmy
Token ID1820
Feature activation-2.90e-02

Pos logit contributions
 ranger+0.033
 bru+0.031
 hinges+0.031
 bleed+0.031
 Something+0.030
Neg logit contributions
ang-0.026
ee-0.023
irling-0.022
ug-0.020
irl-0.020
Token's
Token ID338
Feature activation-2.77e-02

Pos logit contributions
 ranger+0.047
 bleed+0.046
 bru+0.045
 hinges+0.045
 sob+0.044
Neg logit contributions
ang-0.060
irling-0.054
ee-0.053
irl-0.051
ug-0.049
Token mom
Token ID1995
Feature activation-3.22e-02

Pos logit contributions
 ranger+0.042
 bleed+0.042
 bru+0.041
 hinges+0.040
 department+0.040
Neg logit contributions
ang-0.058
irling-0.053
ee-0.052
irl-0.050
ug-0.049
Token shouted
Token ID17293
Feature activation-3.67e-02

Pos logit contributions
 ranger+0.052
 bleed+0.051
 bru+0.050
 hinges+0.050
 sob+0.048
Neg logit contributions
ang-0.067
irling-0.060
ee-0.059
irl-0.056
ug-0.055
Token at
Token ID379
Feature activation-2.28e-02

Pos logit contributions
 ranger+0.076
 bleed+0.075
 hinges+0.074
 bru+0.074
 department+0.071
Neg logit contributions
ang-0.058
irling-0.052
ee-0.049
irl-0.047
ug-0.046
Token him
Token ID683
Feature activation-9.79e-02

Pos logit contributions
 ranger+0.041
 bleed+0.040
 bru+0.040
 hinges+0.040
 sob+0.039
Neg logit contributions
ang-0.041
irling-0.037
ee-0.037
irl-0.035
ug-0.035
Token to
Token ID284
Feature activation-6.50e-02

Pos logit contributions
 bru+0.168
 ranger+0.156
 hinges+0.156
 bleed+0.155
 Something+0.152
Neg logit contributions
ang-0.167
ee-0.148
ug-0.148
irling-0.147
ash-0.146
Token turn
Token ID1210
Feature activation+1.71e-03

Pos logit contributions
 ranger+0.109
 department+0.109
keeper+0.107
 hinges+0.105
onna+0.103
Neg logit contributions
ang-0.108
irling-0.106
irl-0.104
ee-0.102
ading-0.098
Token it
Token ID340
Feature activation-4.83e-02

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ug+0.003
Neg logit contributions
 ranger-0.003
 hinges-0.003
 bru-0.003
 bleed-0.003
 department-0.003
Token off
Token ID572
Feature activation-3.34e-02

Pos logit contributions
 ranger+0.100
 bru+0.100
 Something+0.095
 mercy+0.094
 sob+0.094
Neg logit contributions
ang-0.071
ug-0.063
ee-0.063
irling-0.061
ash-0.061
Token.
Token ID13
Feature activation-2.14e-02

Pos logit contributions
 sob+0.052
 bleed+0.052
 ranger+0.052
 bru+0.051
 hinges+0.050
Neg logit contributions
ang-0.065
ee-0.059
irling-0.057
ug-0.056
irl-0.056
Token
Token ID198
Feature activation-8.33e-03

Pos logit contributions
 department+0.065
 bud+0.061
 recovered+0.061
 mercy+0.060
 injuries+0.060
Neg logit contributions
ang-0.052
ug-0.041
ee-0.040
aw-0.039
irling-0.039
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
 bleed+0.014
 department+0.013
 hinges+0.013
 ranger+0.013
 bru+0.013
Neg logit contributions
ang-0.017
ee-0.015
irling-0.015
irl-0.014
ug-0.014
TokenHis
Token ID6653
Feature activation-5.24e-02

Pos logit contributions
 bleed+0.018
 bru+0.018
 department+0.018
 hinges+0.017
 ranger+0.017
Neg logit contributions
ang-0.021
irling-0.019
ee-0.018
irl-0.017
ug-0.017
Token mom
Token ID1995
Feature activation-3.12e-02

Pos logit contributions
keeping+0.101
 recovered+0.097
 Something+0.097
 department+0.094
onna+0.093
Neg logit contributions
ang-0.084
irl-0.077
irling-0.072
ee-0.071
utes-0.061
Token heard
Token ID2982
Feature activation-2.12e-02

Pos logit contributions
onna+0.046
 bleed+0.046
 hinges+0.046
 ranger+0.046
 bru+0.043
Neg logit contributions
ang-0.066
ee-0.060
irling-0.058
ug-0.056
aw-0.056
Token him
Token ID683
Feature activation-9.78e-02

Pos logit contributions
 hinges+0.043
 ranger+0.042
 bleed+0.042
 department+0.042
 bru+0.042
Neg logit contributions
ang-0.032
ee-0.028
irling-0.027
aw-0.026
ug-0.026
Token and
Token ID290
Feature activation-4.48e-02

Pos logit contributions
 department+0.184
 ranger+0.184
 hinges+0.183
onna+0.180
 bru+0.180
Neg logit contributions
ang-0.165
ee-0.140
irling-0.136
ug-0.130
irl-0.124
Token came
Token ID1625
Feature activation-2.06e-02

Pos logit contributions
 hinges+0.073
 bleed+0.072
 bru+0.071
onna+0.071
 ranger+0.071
Neg logit contributions
ang-0.088
ee-0.076
irling-0.075
ug-0.072
irl-0.068
Token to
Token ID284
Feature activation-5.40e-02

Pos logit contributions
 bleed+0.025
 bru+0.025
 ranger+0.024
ache+0.023
 Something+0.023
Neg logit contributions
ang-0.051
irling-0.046
ee-0.045
ug-0.043
ading-0.043
Token help
Token ID1037
Feature activation-2.28e-02

Pos logit contributions
ache+0.127
 hinges+0.121
edly+0.120
 bleed+0.118
 department+0.116
Neg logit contributions
ang-0.056
ee-0.053
 these-0.051
 us-0.050
 your-0.043
Token him
Token ID683
Feature activation-9.73e-02

Pos logit contributions
 department+0.045
 bru+0.044
 hinges+0.043
 bleed+0.043
ache+0.043
Neg logit contributions
ang-0.032
ee-0.030
ug-0.028
ie-0.027
 us-0.026
Token him
Token ID683
Feature activation-1.01e-01

Pos logit contributions
 hinges+0.041
 ranger+0.040
 department+0.040
 bleed+0.039
 bru+0.039
Neg logit contributions
ang-0.029
ee-0.026
irling-0.025
ug-0.023
irl-0.023
Token and
Token ID290
Feature activation-4.35e-02

Pos logit contributions
 department+0.186
 ranger+0.184
 hinges+0.180
 bru+0.177
onna+0.176
Neg logit contributions
ang-0.175
ee-0.151
irling-0.145
ug-0.141
irl-0.133
Token came
Token ID1625
Feature activation-2.12e-02

Pos logit contributions
 hinges+0.073
 bleed+0.072
onna+0.071
 bru+0.071
 ranger+0.070
Neg logit contributions
ang-0.083
ee-0.071
irling-0.070
ug-0.068
irl-0.063
Token to
Token ID284
Feature activation-5.50e-02

Pos logit contributions
 bleed+0.026
 bru+0.025
 ranger+0.025
ache+0.024
 Something+0.023
Neg logit contributions
ang-0.053
irling-0.047
ee-0.046
ug-0.045
ading-0.044
Token help
Token ID1037
Feature activation-2.03e-02

Pos logit contributions
ache+0.133
 hinges+0.128
edly+0.127
 department+0.123
onna+0.123
Neg logit contributions
ang-0.054
ee-0.049
 these-0.048
 us-0.044
rous-0.040
Token him
Token ID683
Feature activation-9.76e-02

Pos logit contributions
 department+0.039
 hinges+0.038
ache+0.038
 bru+0.037
 ranger+0.037
Neg logit contributions
ang-0.031
ee-0.030
ug-0.028
ie-0.026
ading-0.025
Token.
Token ID13
Feature activation-3.15e-02

Pos logit contributions
 department+0.180
 ranger+0.179
 injuries+0.175
 bru+0.175
 Something+0.168
Neg logit contributions
ang-0.159
irling-0.135
irl-0.133
ading-0.132
ee-0.130
Token She
Token ID1375
Feature activation-2.96e-02

Pos logit contributions
 injuries+0.058
ache+0.055
 mercy+0.055
 bleed+0.054
 recovered+0.054
Neg logit contributions
ang-0.052
aw-0.041
ee-0.038
ug-0.038
irling-0.037
Token saw
Token ID2497
Feature activation+1.21e-02

Pos logit contributions
 bleed+0.054
 hinges+0.052
 bru+0.050
onna+0.049
 ranger+0.049
Neg logit contributions
ang-0.056
irling-0.047
ee-0.047
irl-0.043
ug-0.043
Token the
Token ID262
Feature activation-1.72e-02

Pos logit contributions
ang+0.018
irling+0.015
ee+0.015
irl+0.014
ug+0.013
Neg logit contributions
 ranger-0.026
 bleed-0.026
 hinges-0.026
 bru-0.026
 sob-0.025
Token broken
Token ID5445
Feature activation-2.07e-02

Pos logit contributions
 hinges+0.027
onna+0.026
 ranger+0.026
 department+0.026
 bleed+0.026
Neg logit contributions
ang-0.035
ee-0.030
irling-0.030
irl-0.029
aw-0.027
Token Ben
Token ID3932
Feature activation-1.09e-02

Pos logit contributions
 ranger+0.047
 sob+0.046
 bleed+0.046
 bru+0.045
 hinges+0.044
Neg logit contributions
ang-0.057
irling-0.052
ee-0.050
irl-0.048
ug-0.047
Token said
Token ID531
Feature activation-5.17e-02

Pos logit contributions
 bleed+0.022
 injuries+0.022
 bru+0.021
 hinges+0.021
 department+0.021
Neg logit contributions
ang-0.019
ee-0.015
irling-0.015
ug-0.014
irl-0.014
Token,
Token ID11
Feature activation-4.76e-02

Pos logit contributions
 injuries+0.103
 bru+0.103
 hinges+0.101
 bleed+0.100
 department+0.099
Neg logit contributions
ang-0.076
ug-0.067
irling-0.064
ee-0.061
ap-0.059
Token giving
Token ID3501
Feature activation+3.83e-02

Pos logit contributions
 ranger+0.086
 bleed+0.086
 bru+0.084
 hinges+0.084
 department+0.082
Neg logit contributions
ang-0.088
irling-0.079
ee-0.074
irl-0.073
ug-0.070
Token his
Token ID465
Feature activation-5.77e-02

Pos logit contributions
ang+0.070
irling+0.062
ee+0.059
irl+0.057
ug+0.056
Neg logit contributions
 ranger-0.071
 bleed-0.071
 bru-0.069
 sob-0.069
 hinges-0.069
Token sister
Token ID6621
Feature activation-9.76e-02

Pos logit contributions
 bleed+0.119
 ranger+0.112
 hinges+0.111
 bru+0.110
 sob+0.110
Neg logit contributions
ang-0.090
ee-0.081
il-0.073
irl-0.073
irling-0.072
Token a
Token ID257
Feature activation-6.14e-02

Pos logit contributions
 bleed+0.162
 ranger+0.161
 bru+0.159
 hinges+0.158
 department+0.155
Neg logit contributions
ang-0.193
irling-0.172
ee-0.170
irl-0.162
ug-0.156
Token fruit
Token ID8234
Feature activation-3.05e-02

Pos logit contributions
keeping+0.119
 ranger+0.116
 bleed+0.115
 hinges+0.114
 injuries+0.112
Neg logit contributions
ang-0.102
ee-0.100
irl-0.089
irling-0.083
 sorts-0.082
Token.
Token ID13
Feature activation-2.11e-02

Pos logit contributions
 bru+0.052
 bleed+0.051
 sob+0.050
 injuries+0.049
 Something+0.048
Neg logit contributions
ang-0.061
ee-0.053
ash-0.048
aw-0.048
il-0.047
Token
Token ID198
Feature activation-8.18e-03

Pos logit contributions
 bleed+0.033
 hinges+0.031
 injuries+0.031
 mercy+0.030
 bru+0.030
Neg logit contributions
ang-0.046
irling-0.041
ee-0.040
irl-0.039
ug-0.038
Token
Token ID198
Feature activation-5.20e-03

Pos logit contributions
 ranger+0.009
 bleed+0.009
 bru+0.009
 hinges+0.009
 sob+0.009
Neg logit contributions
ang-0.021
irling-0.019
ee-0.019
irl-0.018
ug-0.018
Token cries
Token ID24691
Feature activation-2.69e-02

Pos logit contributions
 ranger+0.065
 hinges+0.063
 bleed+0.062
 bru+0.061
 department+0.061
Neg logit contributions
ang-0.100
irling-0.090
ee-0.089
irl-0.087
ug-0.082
Token.
Token ID13
Feature activation-2.59e-02

Pos logit contributions
 ranger+0.053
keeping+0.052
 hinges+0.052
onna+0.052
 bru+0.052
Neg logit contributions
ang-0.044
ug-0.041
ee-0.037
aw-0.035
ing-0.034
Token His
Token ID2399
Feature activation-5.05e-02

Pos logit contributions
 bleed+0.046
 hinges+0.044
 recovered+0.044
 injuries+0.043
ache+0.043
Neg logit contributions
ang-0.050
ug-0.042
ee-0.042
irling-0.039
irl-0.038
Token mom
Token ID1995
Feature activation-4.51e-02

Pos logit contributions
 recovered+0.083
onna+0.082
ache+0.080
 bleed+0.080
 department+0.077
Neg logit contributions
ang-0.089
ee-0.083
irl-0.081
irling-0.077
ug-0.075
Token hears
Token ID23180
Feature activation-2.23e-02

Pos logit contributions
 hinges+0.069
 ranger+0.068
 bleed+0.066
 bru+0.065
onna+0.064
Neg logit contributions
ang-0.097
ee-0.087
irling-0.085
irl-0.080
ug-0.078
Token him
Token ID683
Feature activation-9.71e-02

Pos logit contributions
 hinges+0.050
 bru+0.050
 bleed+0.050
 ranger+0.050
 department+0.048
Neg logit contributions
ang-0.029
irling-0.025
ee-0.025
ug-0.023
irl-0.023
Token and
Token ID290
Feature activation-5.26e-02

Pos logit contributions
 hinges+0.171
 department+0.170
 ranger+0.169
 bru+0.167
 Something+0.165
Neg logit contributions
ang-0.175
ee-0.154
irling-0.145
ug-0.142
ash-0.136
Token comes
Token ID2058
Feature activation-3.12e-02

Pos logit contributions
 hinges+0.083
 ranger+0.082
 bru+0.082
 bleed+0.081
 injuries+0.079
Neg logit contributions
ang-0.108
irling-0.095
ee-0.093
irl-0.090
ug-0.086
Token to
Token ID284
Feature activation-5.08e-02

Pos logit contributions
 bru+0.039
 bleed+0.039
ache+0.038
 ranger+0.037
 Something+0.035
Neg logit contributions
ang-0.075
ee-0.067
irling-0.066
ug-0.066
ading-0.064
Token the
Token ID262
Feature activation-1.69e-02

Pos logit contributions
ache+0.118
 hinges+0.110
edly+0.109
 bleed+0.107
 department+0.106
Neg logit contributions
ang-0.059
ee-0.055
 these-0.052
 us-0.050
irl-0.046
Token sand
Token ID6450
Feature activation-2.61e-02

Pos logit contributions
 bleed+0.034
 sob+0.032
 recovered+0.032
onna+0.032
ache+0.031
Neg logit contributions
ang-0.025
ee-0.022
irling-0.021
ading-0.019
aw-0.019
Token screamed
Token ID25421
Feature activation-4.94e-02

Pos logit contributions
 hinges+0.043
 department+0.040
 bleed+0.040
onna+0.040
 injuries+0.039
Neg logit contributions
ang-0.038
ee-0.034
irling-0.033
aw-0.031
irl-0.030
Token and
Token ID290
Feature activation-2.44e-02

Pos logit contributions
 ranger+0.090
 hinges+0.087
 bru+0.084
 department+0.084
 bleed+0.083
Neg logit contributions
ang-0.088
irling-0.081
ee-0.076
ug-0.075
irl-0.074
Token ran
Token ID4966
Feature activation-1.73e-02

Pos logit contributions
 hinges+0.036
 department+0.036
 ranger+0.036
 aid+0.034
 bru+0.034
Neg logit contributions
ang-0.052
irling-0.047
ee-0.046
irl-0.044
ug-0.043
Token to
Token ID284
Feature activation-5.88e-02

Pos logit contributions
 bru+0.033
onna+0.033
 Something+0.032
 sob+0.032
ache+0.032
Neg logit contributions
ang-0.027
ug-0.022
 around-0.021
aw-0.021
ee-0.021
Token help
Token ID1037
Feature activation-2.14e-02

Pos logit contributions
ache+0.114
 hinges+0.114
 department+0.112
 ranger+0.109
 bru+0.108
Neg logit contributions
ang-0.092
irl-0.085
ee-0.084
irling-0.082
ading-0.077
Token him
Token ID683
Feature activation-9.71e-02

Pos logit contributions
 hinges+0.042
 department+0.042
 bru+0.039
ache+0.039
 bleed+0.039
Neg logit contributions
ang-0.033
ee-0.032
irl-0.031
irling-0.029
ug-0.029
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
 bru+0.172
 department+0.171
 hinges+0.168
 ranger+0.166
 injuries+0.164
Neg logit contributions
ang-0.173
irling-0.148
irl-0.147
ee-0.142
ading-0.140
Token
Token ID198
Feature activation-6.99e-03

Pos logit contributions
 hinges+0.038
 bleed+0.038
 injuries+0.037
ache+0.037
 mercy+0.037
Neg logit contributions
ang-0.041
ee-0.035
irling-0.033
irl-0.032
ug-0.031
Token
Token ID198
Feature activation-9.18e-03

Pos logit contributions
 bleed+0.010
 hinges+0.010
 bru+0.010
 ranger+0.010
 department+0.010
Neg logit contributions
ang-0.015
irling-0.014
ee-0.014
irl-0.013
ug-0.013
TokenA
Token ID32
Feature activation-2.53e-02

Pos logit contributions
 ranger+0.014
 bleed+0.014
 sob+0.014
 hinges+0.013
 bru+0.013
Neg logit contributions
ang-0.020
irling-0.017
ee-0.017
ash-0.016
ug-0.016
Token man
Token ID582
Feature activation-1.14e-02

Pos logit contributions
 bleed+0.040
 hinges+0.037
 bru+0.037
 sob+0.036
keeping+0.035
Neg logit contributions
ang-0.054
ee-0.046
irling-0.046
irl-0.044
ug-0.044
Token
Token ID198
Feature activation-2.24e-02

Pos logit contributions
keeping+0.107
 recovered+0.103
 injuries+0.101
 bleed+0.100
 mercy+0.098
Neg logit contributions
ang-0.044
ug-0.029
 everyone-0.028
 at-0.028
ee-0.028
Token
Token ID198
Feature activation-1.58e-02

Pos logit contributions
 department+0.042
 bleed+0.041
 hinges+0.041
 ranger+0.040
 bru+0.039
Neg logit contributions
ang-0.041
ee-0.035
irling-0.035
irl-0.035
ug-0.034
TokenHis
Token ID6653
Feature activation-4.64e-02

Pos logit contributions
 bleed+0.029
 bru+0.029
 department+0.028
 hinges+0.028
onna+0.027
Neg logit contributions
ang-0.030
irling-0.026
ee-0.025
irl-0.024
ug-0.024
Token mom
Token ID1995
Feature activation-3.42e-02

Pos logit contributions
 bleed+0.083
keeping+0.082
 recovered+0.081
onna+0.078
ache+0.076
Neg logit contributions
ang-0.080
irl-0.072
ee-0.072
irling-0.068
utes-0.061
Token hears
Token ID23180
Feature activation-1.21e-02

Pos logit contributions
 ranger+0.045
 hinges+0.044
 bleed+0.044
keeping+0.042
onna+0.042
Neg logit contributions
ang-0.078
irling-0.073
ee-0.072
irl-0.069
aw-0.066
Token him
Token ID683
Feature activation-9.67e-02

Pos logit contributions
 hinges+0.024
 bleed+0.024
 bru+0.024
 ranger+0.024
 department+0.024
Neg logit contributions
ang-0.018
ee-0.017
irling-0.016
ug-0.016
irl-0.015
Token and
Token ID290
Feature activation-4.79e-02

Pos logit contributions
 bru+0.179
 department+0.179
 ranger+0.179
 hinges+0.176
 injuries+0.175
Neg logit contributions
ang-0.161
ee-0.140
irling-0.131
ug-0.128
irl-0.120
Token comes
Token ID2058
Feature activation-2.59e-02

Pos logit contributions
 hinges+0.078
 ranger+0.077
 bru+0.076
 bleed+0.076
 injuries+0.074
Neg logit contributions
ang-0.095
irling-0.084
ee-0.083
irl-0.080
ug-0.075
Token to
Token ID284
Feature activation-4.42e-02

Pos logit contributions
 bleed+0.035
 bru+0.034
ache+0.034
 ranger+0.032
 cough+0.030
Neg logit contributions
ang-0.060
ee-0.053
irling-0.053
ug-0.052
ading-0.051
Token him
Token ID683
Feature activation-8.30e-02

Pos logit contributions
ache+0.116
 hinges+0.108
 bleed+0.107
 department+0.106
edly+0.106
Neg logit contributions
ang-0.037
ee-0.036
 these-0.032
 us-0.030
 sorts-0.028
Token.
Token ID13
Feature activation-2.78e-02

Pos logit contributions
 bru+0.151
 department+0.145
 injuries+0.143
ache+0.141
 recovered+0.140
Neg logit contributions
ang-0.149
irling-0.117
ee-0.116
irl-0.111
 sorts-0.111
Token and
Token ID290
Feature activation+1.70e-03

Pos logit contributions
 ranger+0.049
 hinges+0.049
 bleed+0.048
 bru+0.047
 injuries+0.046
Neg logit contributions
ang-0.052
irling-0.046
ee-0.045
irl-0.042
ug-0.040
Token Sam
Token ID3409
Feature activation-2.67e-02

Pos logit contributions
ang+0.003
ee+0.002
irling+0.002
irl+0.002
ug+0.002
Neg logit contributions
 bleed-0.003
 ranger-0.003
 hinges-0.003
 bru-0.003
 department-0.003
Token run
Token ID1057
Feature activation-2.29e-02

Pos logit contributions
 ranger+0.045
 hinges+0.044
 department+0.043
onna+0.041
 aid+0.041
Neg logit contributions
ang-0.053
ee-0.050
irling-0.047
irl-0.045
ug-0.043
Token to
Token ID284
Feature activation-6.11e-02

Pos logit contributions
 bru+0.039
onna+0.038
ache+0.037
 Something+0.037
 sob+0.036
Neg logit contributions
ang-0.042
ee-0.036
ug-0.034
aw-0.034
 around-0.032
Token help
Token ID1037
Feature activation-3.20e-02

Pos logit contributions
ache+0.146
 hinges+0.141
 bleed+0.140
 ranger+0.138
 department+0.137
Neg logit contributions
ang-0.070
ee-0.064
irling-0.060
irl-0.059
ading-0.056
Token him
Token ID683
Feature activation-9.66e-02

Pos logit contributions
 ranger+0.052
 bleed+0.052
 bru+0.051
 hinges+0.050
 department+0.049
Neg logit contributions
ang-0.063
ee-0.059
irling-0.058
irl-0.056
ading-0.054
Token.
Token ID13
Feature activation-3.22e-02

Pos logit contributions
 ranger+0.157
 bru+0.156
 injuries+0.150
 department+0.150
 hinges+0.145
Neg logit contributions
ang-0.191
irling-0.169
ee-0.166
irl-0.164
ading-0.161
Token They
Token ID1119
Feature activation-2.95e-02

Pos logit contributions
ache+0.069
 mercy+0.068
 bleed+0.068
 injuries+0.067
 recovered+0.066
Neg logit contributions
ang-0.046
ee-0.039
irling-0.038
irl-0.035
ading-0.034
Token are
Token ID389
Feature activation+1.94e-03

Pos logit contributions
 ranger+0.045
 hinges+0.045
 department+0.043
 bru+0.042
 sob+0.041
Neg logit contributions
ang-0.061
ee-0.056
irl-0.055
irling-0.054
ug-0.049
Token sorry
Token ID7926
Feature activation-4.50e-02

Pos logit contributions
ang+0.004
irling+0.004
ee+0.004
irl+0.003
ug+0.003
Neg logit contributions
 ranger-0.003
 bleed-0.003
 bru-0.003
 hinges-0.003
 sob-0.003
Token and
Token ID290
Feature activation-3.59e-02

Pos logit contributions
 ranger+0.073
 bleed+0.073
 bru+0.072
 hinges+0.071
 aid+0.068
Neg logit contributions
ang-0.089
irling-0.079
ee-0.079
ug-0.075
irl-0.074
Token his
Token ID465
Feature activation-4.55e-02

Pos logit contributions
 department+0.042
onna+0.042
 hinges+0.042
 ranger+0.042
 bleed+0.041
Neg logit contributions
ang-0.046
irling-0.041
irl-0.040
ee-0.040
ug-0.038
Token mum
Token ID25682
Feature activation-3.83e-02

Pos logit contributions
 Something+0.077
onna+0.077
 ranger+0.076
 recovered+0.075
 department+0.075
Neg logit contributions
ang-0.078
irl-0.072
irling-0.070
ee-0.070
ug-0.064
Token came
Token ID1625
Feature activation-1.96e-02

Pos logit contributions
 bleed+0.062
 hinges+0.062
 ranger+0.061
 bru+0.060
onna+0.058
Neg logit contributions
ang-0.077
irling-0.068
ee-0.068
ug-0.065
irl-0.062
Token to
Token ID284
Feature activation-5.51e-02

Pos logit contributions
 ranger+0.035
 bleed+0.035
 bru+0.034
 hinges+0.033
 department+0.033
Neg logit contributions
ang-0.037
irling-0.033
ee-0.031
irl-0.030
ug-0.030
Token help
Token ID1037
Feature activation-2.49e-02

Pos logit contributions
 hinges+0.097
ache+0.097
onna+0.094
edly+0.093
 department+0.093
Neg logit contributions
ee-0.083
ang-0.082
irl-0.076
 us-0.076
 these-0.075
Token him
Token ID683
Feature activation-9.60e-02

Pos logit contributions
 ranger+0.050
 department+0.049
 bru+0.048
 hinges+0.047
 bleed+0.047
Neg logit contributions
ang-0.036
ee-0.036
irling-0.032
ug-0.032
ading-0.031
Token.
Token ID13
Feature activation-3.62e-02

Pos logit contributions
 ranger+0.171
 department+0.162
 injuries+0.156
 bru+0.155
onna+0.153
Neg logit contributions
ang-0.167
irling-0.155
irl-0.153
ee-0.149
ading-0.147
Token She
Token ID1375
Feature activation-2.28e-02

Pos logit contributions
 bleed+0.057
 injuries+0.055
 bru+0.052
 mercy+0.052
 recovered+0.051
Neg logit contributions
ang-0.070
irling-0.064
ee-0.062
 sorts-0.059
aw-0.059
Token was
Token ID373
Feature activation+1.53e-03

Pos logit contributions
 ranger+0.034
 bleed+0.034
 bru+0.033
 sob+0.033
 hinges+0.033
Neg logit contributions
ang-0.050
irling-0.045
ee-0.044
irl-0.043
ug-0.041
Token very
Token ID845
Feature activation-7.09e-03

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.002
ug+0.002
Neg logit contributions
 ranger-0.003
 bleed-0.003
 bru-0.003
 hinges-0.003
 sob-0.003
Token kind
Token ID1611
Feature activation-1.52e-02

Pos logit contributions
 ranger+0.011
 hinges+0.011
 bleed+0.011
 bru+0.011
 department+0.011
Neg logit contributions
ang-0.015
irling-0.014
ee-0.013
irl-0.013
ug-0.012
Token cow
Token ID9875
Feature activation+6.19e-03

Pos logit contributions
 bleed+0.037
 ranger+0.036
 hinges+0.036
 bru+0.035
 sob+0.035
Neg logit contributions
ang-0.038
irling-0.034
ee-0.033
irl-0.032
ug-0.030
Token.
Token ID13
Feature activation+7.00e-03

Pos logit contributions
ang+0.011
ee+0.010
irling+0.010
ug+0.009
irl+0.009
Neg logit contributions
 bleed-0.011
 hinges-0.011
 ranger-0.011
 bru-0.011
 sob-0.011
Token She
Token ID1375
Feature activation+6.60e-03

Pos logit contributions
ang+0.017
irling+0.015
ee+0.015
irl+0.015
ug+0.014
Neg logit contributions
 ranger-0.009
 bleed-0.009
 bru-0.008
 hinges-0.008
 sob-0.008
Token loved
Token ID6151
Feature activation+1.14e-02

Pos logit contributions
ang+0.014
irling+0.013
ee+0.012
irl+0.012
ug+0.012
Neg logit contributions
 ranger-0.010
 bleed-0.010
 sob-0.010
 bru-0.010
 hinges-0.010
Token to
Token ID284
Feature activation-1.59e-02

Pos logit contributions
ang+0.028
irling+0.026
ee+0.025
irl+0.025
ug+0.024
Neg logit contributions
 ranger-0.014
 bleed-0.013
 bru-0.013
 sob-0.013
 hinges-0.013
Token roam
Token ID35563
Feature activation+9.03e-02

Pos logit contributions
 ranger+0.025
 bleed+0.024
 bru+0.024
 hinges+0.024
 sob+0.023
Neg logit contributions
ang-0.033
irling-0.030
ee-0.029
irl-0.029
ug-0.028
Token the
Token ID262
Feature activation+4.63e-02

Pos logit contributions
ang+0.156
irling+0.153
irl+0.136
ee+0.136
ading+0.127
Neg logit contributions
 ranger-0.172
 bleed-0.164
 hinges-0.158
 sob-0.156
 keeper-0.154
Token fields
Token ID7032
Feature activation+3.21e-02

Pos logit contributions
ang+0.091
irling+0.088
irl+0.082
angles+0.081
ug+0.081
Neg logit contributions
 ranger-0.073
 keeper-0.066
 bleed-0.065
 aid-0.062
 sob-0.061
Token and
Token ID290
Feature activation+3.74e-02

Pos logit contributions
ang+0.070
irling+0.064
ee+0.062
irl+0.061
ug+0.058
Neg logit contributions
 ranger-0.048
 bleed-0.046
 hinges-0.045
 bru-0.045
 sob-0.044
Token make
Token ID787
Feature activation+4.09e-02

Pos logit contributions
ang+0.086
ee+0.078
irling+0.077
ug+0.074
ash+0.074
Neg logit contributions
 ranger-0.054
 bleed-0.053
 bru-0.051
 sob-0.050
 hinges-0.050
Token friends
Token ID2460
Feature activation-4.35e-03

Pos logit contributions
ang+0.082
irling+0.071
ee+0.069
irl+0.065
ash+0.065
Neg logit contributions
 bleed-0.072
 sob-0.068
 ranger-0.068
 bru-0.065
 hinges-0.064
Token and
Token ID290
Feature activation+2.30e-02

Pos logit contributions
 bleed+0.034
 ranger+0.034
 bru+0.033
 hinges+0.033
 sob+0.032
Neg logit contributions
ang-0.053
irling-0.048
ee-0.046
irl-0.045
ug-0.044
Token said
Token ID531
Feature activation-2.60e-02

Pos logit contributions
ang+0.050
irling+0.045
ee+0.044
irl+0.043
ug+0.042
Neg logit contributions
 ranger-0.034
 bleed-0.034
 sob-0.034
 bru-0.033
 hinges-0.031
Token "
Token ID366
Feature activation+2.70e-02

Pos logit contributions
 bleed+0.045
 ranger+0.045
 hinges+0.045
 bru+0.044
 department+0.042
Neg logit contributions
ang-0.050
irling-0.045
ee-0.044
irl-0.042
ug-0.042
TokenHello
Token ID15496
Feature activation+1.20e-03

Pos logit contributions
ang+0.057
irling+0.052
ee+0.050
irl+0.050
ug+0.048
Neg logit contributions
 ranger-0.041
 bleed-0.041
 bru-0.040
 sob-0.039
 hinges-0.039
Token,
Token ID11
Feature activation+2.69e-02

Pos logit contributions
ang+0.002
irling+0.002
irl+0.002
ug+0.002
ading+0.002
Neg logit contributions
 ranger-0.002
 department-0.002
 bru-0.002
 sob-0.002
 keeper-0.002
Token cater
Token ID20825
Feature activation+9.37e-02

Pos logit contributions
ang+0.056
irling+0.054
irl+0.049
ug+0.049
angles+0.047
Neg logit contributions
 ranger-0.043
 bru-0.041
 sob-0.040
 Something-0.040
 keeper-0.038
Tokenpillar
Token ID41643
Feature activation+4.01e-02

Pos logit contributions
angles+0.188
irling+0.187
irl+0.186
ang+0.182
ading+0.182
Neg logit contributions
 ranger-0.116
ache-0.115
onna-0.115
 bleed-0.110
 bru-0.109
Token".
Token ID1911
Feature activation+3.06e-03

Pos logit contributions
ang+0.084
irling+0.077
irl+0.073
ee+0.073
ading+0.071
Neg logit contributions
 ranger-0.062
 bru-0.058
 bleed-0.058
 sob-0.058
 department-0.056
Token The
Token ID383
Feature activation+1.44e-02

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.005
Neg logit contributions
 bleed-0.005
 hinges-0.005
 ranger-0.005
 bru-0.005
 department-0.005
Token cater
Token ID20825
Feature activation+8.85e-02

Pos logit contributions
ang+0.030
irling+0.027
ee+0.026
irl+0.025
ug+0.025
Neg logit contributions
 ranger-0.023
 bleed-0.023
 bru-0.022
 sob-0.022
 hinges-0.022
Tokenpillar
Token ID41643
Feature activation+2.80e-02

Pos logit contributions
ang+0.169
irling+0.166
irl+0.162
angles+0.161
ading+0.158
Neg logit contributions
 ranger-0.129
 bleed-0.128
 sob-0.126
 bru-0.124
 recovered-0.119
Token see
Token ID766
Feature activation+1.99e-02

Pos logit contributions
 ranger+0.038
 bleed+0.038
 hinges+0.037
 bru+0.037
 sob+0.036
Neg logit contributions
ang-0.067
irling-0.059
ee-0.058
ug-0.055
irl-0.055
Token her
Token ID607
Feature activation-2.43e-02

Pos logit contributions
ang+0.039
irling+0.036
ee+0.033
irl+0.033
ug+0.032
Neg logit contributions
 ranger-0.033
 bleed-0.033
 sob-0.033
 bru-0.032
 hinges-0.031
Token very
Token ID845
Feature activation-2.54e-03

Pos logit contributions
 bleed+0.040
 bru+0.039
 ranger+0.039
 hinges+0.039
 sob+0.037
Neg logit contributions
ang-0.049
irling-0.044
ee-0.043
irl-0.042
ading-0.039
Token first
Token ID717
Feature activation-1.29e-02

Pos logit contributions
 ranger+0.004
 bleed+0.004
 hinges+0.004
 sob+0.004
 bru+0.004
Neg logit contributions
ang-0.005
irling-0.004
ee-0.004
ug-0.004
irl-0.004
Token flower
Token ID15061
Feature activation+6.38e-03

Pos logit contributions
 bleed+0.021
 ranger+0.021
 bru+0.020
 hinges+0.020
 sob+0.020
Neg logit contributions
ang-0.026
irling-0.024
ee-0.023
irl-0.022
ug-0.022
Token blo
Token ID24924
Feature activation+9.05e-02

Pos logit contributions
ang+0.013
irling+0.012
ee+0.012
irl+0.012
ug+0.011
Neg logit contributions
 bru-0.010
 bleed-0.010
 ranger-0.009
 hinges-0.009
 sob-0.009
Tokenoming
Token ID3383
Feature activation+4.65e-02

Pos logit contributions
ang+0.084
irling+0.072
ee+0.069
irl+0.068
ug+0.064
Neg logit contributions
onna-0.219
 ranger-0.216
 department-0.215
 bud-0.215
 bleed-0.213
Token in
Token ID287
Feature activation+3.45e-02

Pos logit contributions
ang+0.094
irling+0.085
ee+0.082
irl+0.080
ug+0.077
Neg logit contributions
 bleed-0.076
 ranger-0.075
 bru-0.075
 hinges-0.073
 sob-0.073
Token her
Token ID607
Feature activation+6.05e-03

Pos logit contributions
ang+0.059
irling+0.052
ee+0.050
irl+0.049
ug+0.046
Neg logit contributions
 ranger-0.067
 bleed-0.067
 bru-0.066
 hinges-0.065
 sob-0.065
Token garden
Token ID11376
Feature activation-1.15e-03

Pos logit contributions
ang+0.012
irling+0.011
ee+0.011
irl+0.010
ug+0.010
Neg logit contributions
 ranger-0.010
 bru-0.010
 bleed-0.010
 hinges-0.010
 sob-0.010
Token.
Token ID13
Feature activation+3.27e-03

Pos logit contributions
 bleed+0.002
 ranger+0.001
 bru+0.001
 hinges+0.001
 sob+0.001
Neg logit contributions
ang-0.003
irling-0.002
ee-0.002
irl-0.002
ug-0.002
Token,
Token ID11
Feature activation+8.16e-03

Pos logit contributions
 bleed+0.026
 hinges+0.026
 bru+0.026
 ranger+0.025
 sob+0.024
Neg logit contributions
ang-0.039
irling-0.034
ee-0.034
irl-0.032
ug-0.032
Token and
Token ID290
Feature activation+2.97e-02

Pos logit contributions
ang+0.016
irling+0.014
ee+0.014
irl+0.014
ug+0.013
Neg logit contributions
 bleed-0.014
 ranger-0.013
 bru-0.013
 hinges-0.013
 sob-0.013
Token they
Token ID484
Feature activation+3.54e-02

Pos logit contributions
ang+0.059
irling+0.053
ee+0.051
irl+0.050
ug+0.048
Neg logit contributions
 ranger-0.051
 bleed-0.050
 bru-0.049
 sob-0.048
 hinges-0.048
Token played
Token ID2826
Feature activation+5.34e-02

Pos logit contributions
ang+0.061
irling+0.054
ee+0.052
irl+0.050
ug+0.048
Neg logit contributions
 ranger-0.068
 bleed-0.068
 bru-0.067
 hinges-0.067
 sob-0.066
Token and
Token ID290
Feature activation+4.47e-02

Pos logit contributions
ang+0.085
irling+0.072
ee+0.070
irl+0.064
ash+0.063
Neg logit contributions
 ranger-0.113
 bleed-0.109
 sob-0.109
 bru-0.108
 hinges-0.105
Token spl
Token ID4328
Feature activation+1.03e-01

Pos logit contributions
ang+0.094
irling+0.085
ee+0.083
irl+0.079
ading+0.077
Neg logit contributions
 ranger-0.073
 bleed-0.071
 bru-0.069
 sob-0.069
 hinges-0.067
Tokenashed
Token ID5263
Feature activation+7.45e-02

Pos logit contributions
ang+0.200
irling+0.185
ee+0.179
irl+0.175
 sorts+0.164
Neg logit contributions
 ranger-0.155
 bleed-0.154
 recovered-0.148
 department-0.146
 sob-0.143
Token in
Token ID287
Feature activation+3.39e-02

Pos logit contributions
ang+0.124
irling+0.113
irl+0.102
ee+0.099
ug+0.093
Neg logit contributions
 ranger-0.148
 bleed-0.143
 sob-0.141
 bru-0.140
 hinges-0.136
Token the
Token ID262
Feature activation+3.50e-02

Pos logit contributions
ang+0.060
irling+0.053
ee+0.050
irl+0.048
ug+0.047
Neg logit contributions
 ranger-0.066
 bleed-0.064
 bru-0.064
 hinges-0.063
 sob-0.063
Token water
Token ID1660
Feature activation+2.22e-02

Pos logit contributions
ang+0.068
irling+0.061
ug+0.056
ee+0.056
irl+0.056
Neg logit contributions
 ranger-0.063
 bleed-0.061
 bru-0.060
 sob-0.059
 hinges-0.059
Token together
Token ID1978
Feature activation+9.03e-03

Pos logit contributions
ang+0.057
irling+0.052
ee+0.051
irl+0.050
ug+0.048
Neg logit contributions
 ranger-0.025
 bleed-0.024
 bru-0.023
 hinges-0.023
 sob-0.023
Token of
Token ID286
Feature activation+3.58e-02

Pos logit contributions
 bleed+0.026
 bru+0.026
 ranger+0.026
 hinges+0.026
ache+0.025
Neg logit contributions
ang-0.019
irling-0.016
ug-0.015
ee-0.015
ading-0.015
Token the
Token ID262
Feature activation+1.65e-02

Pos logit contributions
ang+0.062
irling+0.054
ee+0.052
irl+0.050
ug+0.048
Neg logit contributions
 ranger-0.070
 bleed-0.070
 bru-0.069
 hinges-0.068
 sob-0.068
Token bucket
Token ID19236
Feature activation+2.20e-03

Pos logit contributions
ang+0.031
irling+0.028
ee+0.027
irl+0.026
ug+0.025
Neg logit contributions
 ranger-0.030
 bleed-0.030
 bru-0.029
 hinges-0.029
 sob-0.029
Token.
Token ID13
Feature activation+8.71e-05

Pos logit contributions
ang+0.004
irling+0.004
ee+0.004
ug+0.004
irl+0.004
Neg logit contributions
 bleed-0.004
 bru-0.004
 ranger-0.004
 hinges-0.003
 sob-0.003
Token It
Token ID632
Feature activation+1.69e-02

Pos logit contributions
ang+0.000
irling+0.000
ee+0.000
irl+0.000
ug+0.000
Neg logit contributions
 bleed-0.000
 bru-0.000
 hinges-0.000
 ranger-0.000
 sob-0.000
Token sw
Token ID1509
Feature activation+9.14e-02

Pos logit contributions
ang+0.037
irling+0.033
ee+0.032
irl+0.031
ug+0.031
Neg logit contributions
 ranger-0.026
 bleed-0.025
 sob-0.025
 bru-0.024
 hinges-0.024
Tokenam
Token ID321
Feature activation+5.07e-02

Pos logit contributions
ang+0.171
irling+0.149
ee+0.145
irl+0.140
ug+0.132
Neg logit contributions
 ranger-0.158
 hinges-0.156
 bleed-0.156
 department-0.150
 bru-0.149
Token in
Token ID287
Feature activation+1.87e-02

Pos logit contributions
ang+0.099
irling+0.088
ee+0.086
irl+0.082
ug+0.080
Neg logit contributions
 bleed-0.088
 bru-0.087
 ranger-0.086
 hinges-0.085
 sob-0.084
Token a
Token ID257
Feature activation-9.72e-04

Pos logit contributions
ang+0.026
irling+0.022
ee+0.022
irl+0.021
ading+0.020
Neg logit contributions
 bleed-0.042
 ranger-0.040
ache-0.040
 bru-0.040
 sob-0.040
Token p
Token ID279
Feature activation+5.84e-02

Pos logit contributions
 ranger+0.002
 bru+0.002
 bleed+0.002
 sob+0.001
 hinges+0.001
Neg logit contributions
ang-0.002
irling-0.002
ee-0.002
ug-0.002
irl-0.002
Tokenuddle
Token ID24500
Feature activation+1.38e-02

Pos logit contributions
ang+0.103
irling+0.098
ee+0.098
irl+0.096
ug+0.091
Neg logit contributions
 ranger-0.100
 hinges-0.092
 bleed-0.090
 keeper-0.089
 bru-0.089
Token little
Token ID1310
Feature activation+1.11e-02

Pos logit contributions
 ranger+0.001
 bru+0.001
 sob+0.001
 department+0.001
 keeper+0.001
Neg logit contributions
ang-0.002
irling-0.002
ash-0.002
ee-0.001
ug-0.001
Token girl
Token ID2576
Feature activation+4.22e-03

Pos logit contributions
irling+0.028
ang+0.027
irl+0.027
ading+0.026
ash+0.025
Neg logit contributions
 ranger-0.013
 bru-0.012
 bleed-0.011
 sob-0.011
 cough-0.011
Token named
Token ID3706
Feature activation+2.51e-02

Pos logit contributions
ang+0.008
irling+0.007
irl+0.007
ading+0.007
ee+0.007
Neg logit contributions
 ranger-0.008
 sob-0.007
 bru-0.007
 hinges-0.007
 aid-0.007
Token Mary
Token ID5335
Feature activation+2.40e-02

Pos logit contributions
ang+0.059
irling+0.054
ee+0.052
irl+0.050
ug+0.049
Neg logit contributions
 ranger-0.035
 sob-0.034
 hinges-0.034
 bleed-0.034
 bru-0.033
Token.
Token ID13
Feature activation+2.71e-03

Pos logit contributions
ang+0.052
ee+0.048
irling+0.047
irl+0.044
ug+0.043
Neg logit contributions
 ranger-0.036
 hinges-0.036
 bleed-0.035
 department-0.035
 bru-0.034
Token Mary
Token ID5335
Feature activation+4.18e-02

Pos logit contributions
ang+0.006
irling+0.006
ee+0.006
irl+0.005
ug+0.005
Neg logit contributions
 bleed-0.004
 ranger-0.004
 bru-0.004
 sob-0.003
 hinges-0.003
Token loved
Token ID6151
Feature activation-6.00e-03

Pos logit contributions
ang+0.080
irling+0.072
ee+0.069
irl+0.067
ug+0.065
Neg logit contributions
 ranger-0.073
 bleed-0.073
 bru-0.071
 hinges-0.071
 sob-0.070
Token to
Token ID284
Feature activation-3.96e-02

Pos logit contributions
 ranger+0.005
 bleed+0.005
 bru+0.004
 hinges+0.004
 sob+0.004
Neg logit contributions
ang-0.017
irling-0.016
ee-0.016
irl-0.015
ug-0.015
Token wear
Token ID5806
Feature activation+2.30e-02

Pos logit contributions
 ranger+0.067
 bleed+0.065
 department+0.064
 bru+0.064
 hinges+0.063
Neg logit contributions
ang-0.076
irling-0.071
irl-0.070
ee-0.069
ug-0.064
Token her
Token ID607
Feature activation-3.07e-02

Pos logit contributions
ang+0.060
ee+0.051
irling+0.050
ug+0.050
ash+0.048
Neg logit contributions
 sob-0.027
 bleed-0.027
 hinges-0.026
 bru-0.025
 ranger-0.025
Token favourite
Token ID12507
Feature activation-8.55e-03

Pos logit contributions
 ranger+0.062
 bleed+0.061
 bru+0.060
 hinges+0.060
 sob+0.060
Neg logit contributions
ang-0.051
irling-0.045
ee-0.043
irl-0.041
ug-0.040
Token."
Token ID526
Feature activation-8.72e-03

Pos logit contributions
 ranger+0.010
 bleed+0.010
 sob+0.009
 bru+0.009
 hinges+0.009
Neg logit contributions
ang-0.016
irling-0.015
ee-0.014
irl-0.014
ug-0.013
Token
Token ID198
Feature activation+1.04e-02

Pos logit contributions
 bleed+0.015
 hinges+0.014
 department+0.014
 bru+0.013
 injuries+0.013
Neg logit contributions
ang-0.019
irling-0.016
ee-0.016
irl-0.015
aw-0.014
Token
Token ID198
Feature activation+1.43e-02

Pos logit contributions
ang+0.026
irling+0.024
ee+0.023
irl+0.022
ug+0.022
Neg logit contributions
 ranger-0.013
 bleed-0.012
 bru-0.012
 hinges-0.012
 sob-0.012
TokenBen
Token ID11696
Feature activation+1.52e-02

Pos logit contributions
ang+0.034
irling+0.031
ee+0.030
ug+0.029
irl+0.029
Neg logit contributions
 ranger-0.019
 bru-0.018
 bleed-0.017
 sob-0.017
 hinges-0.017
Token che
Token ID1125
Feature activation+8.45e-02

Pos logit contributions
ang+0.034
irling+0.031
ee+0.030
irl+0.029
ug+0.028
Neg logit contributions
 ranger-0.022
 bleed-0.022
 bru-0.021
 hinges-0.021
 sob-0.021
Tokenwed
Token ID19103
Feature activation+2.86e-02

Pos logit contributions
ang+0.163
irling+0.145
ee+0.138
irl+0.135
ug+0.131
Neg logit contributions
 bleed-0.145
 ranger-0.144
 bru-0.144
 sob-0.141
 hinges-0.141
Token the
Token ID262
Feature activation-1.46e-02

Pos logit contributions
ang+0.053
irling+0.048
ee+0.045
irl+0.045
ug+0.043
Neg logit contributions
 ranger-0.051
 bleed-0.051
 bru-0.050
 sob-0.049
 hinges-0.049
Token gum
Token ID27142
Feature activation-2.07e-02

Pos logit contributions
 ranger+0.024
 bleed+0.023
 hinges+0.023
 bru+0.023
 sob+0.023
Neg logit contributions
ang-0.030
irling-0.027
ee-0.026
irl-0.025
ug-0.024
Token and
Token ID290
Feature activation+1.31e-02

Pos logit contributions
 bru+0.040
 bleed+0.039
 hinges+0.039
 ranger+0.039
 injuries+0.038
Neg logit contributions
ang-0.035
ee-0.031
irling-0.030
irl-0.029
ug-0.027
Token said
Token ID531
Feature activation-3.37e-02

Pos logit contributions
ang+0.027
irling+0.025
ee+0.024
irl+0.023
ug+0.022
Neg logit contributions
 ranger-0.021
 bleed-0.021
 bru-0.020
 hinges-0.020
 sob-0.020
Token,
Token ID11
Feature activation-3.65e-02

Pos logit contributions
 bleed+0.069
 hinges+0.068
 ranger+0.067
 injuries+0.065
 bru+0.065
Neg logit contributions
ang-0.053
irling-0.049
ee-0.047
irl-0.046
ug-0.045
Token stayed
Token ID9658
Feature activation+3.36e-02

Pos logit contributions
ang+0.073
irling+0.065
ee+0.064
irl+0.061
ug+0.061
Neg logit contributions
 bleed-0.056
 bru-0.055
 ranger-0.055
 hinges-0.055
 sob-0.054
Token on
Token ID319
Feature activation+1.07e-02

Pos logit contributions
ang+0.068
ee+0.061
ash+0.058
irling+0.058
irl+0.054
Neg logit contributions
 bleed-0.055
 ranger-0.055
 hinges-0.054
 bru-0.053
 sob-0.053
Token and
Token ID290
Feature activation+2.73e-02

Pos logit contributions
ang+0.018
irling+0.016
ee+0.015
irl+0.015
ug+0.014
Neg logit contributions
 ranger-0.021
 bleed-0.021
 bru-0.021
 hinges-0.021
 sob-0.020
Token the
Token ID262
Feature activation+2.12e-03

Pos logit contributions
ang+0.053
irling+0.047
ee+0.045
irl+0.044
ug+0.042
Neg logit contributions
 ranger-0.048
 bleed-0.047
 bru-0.046
 hinges-0.046
 sob-0.046
Token car
Token ID1097
Feature activation+1.59e-02

Pos logit contributions
ang+0.004
irling+0.004
ee+0.004
irl+0.003
ug+0.003
Neg logit contributions
 bleed-0.004
 ranger-0.004
 sob-0.004
 bru-0.004
 hinges-0.004
Token moved
Token ID3888
Feature activation+4.06e-02

Pos logit contributions
ang+0.030
irling+0.027
ee+0.027
irl+0.026
ug+0.026
Neg logit contributions
 bleed-0.027
 bru-0.027
 hinges-0.026
 ranger-0.026
 sob-0.026
Token.
Token ID13
Feature activation-4.73e-03

Pos logit contributions
ang+0.082
irling+0.073
ee+0.072
irl+0.068
ug+0.068
Neg logit contributions
 bleed-0.067
 ranger-0.067
 bru-0.066
 sob-0.065
 hinges-0.065
Token Tom
Token ID4186
Feature activation+1.31e-02

Pos logit contributions
 bleed+0.007
 hinges+0.007
 ranger+0.007
 bru+0.007
 sob+0.007
Neg logit contributions
ang-0.010
irling-0.009
ee-0.009
irl-0.008
ug-0.008
Token was
Token ID373
Feature activation+2.94e-02

Pos logit contributions
ang+0.029
irling+0.027
ee+0.026
irl+0.025
ading+0.025
Neg logit contributions
 ranger-0.019
 sob-0.018
 bleed-0.018
 bru-0.018
 hinges-0.017
Token happy
Token ID3772
Feature activation-1.76e-02

Pos logit contributions
ang+0.062
ee+0.057
ash+0.054
irling+0.054
angles+0.053
Neg logit contributions
 bleed-0.048
 sob-0.047
 ranger-0.047
 bru-0.045
 hinges-0.042
Token.
Token ID13
Feature activation-2.12e-04

Pos logit contributions
 bleed+0.034
 hinges+0.034
 bru+0.031
 severe+0.031
 limp+0.031
Neg logit contributions
ang-0.031
irling-0.026
ee-0.026
aw-0.024
ug-0.023
Token "
Token ID366
Feature activation+2.89e-02

Pos logit contributions
 bleed+0.049
 ranger+0.048
 bru+0.047
 hinges+0.047
 sob+0.046
Neg logit contributions
ang-0.078
irling-0.071
ee-0.069
irl-0.067
ug-0.065
TokenLook
Token ID8567
Feature activation+4.60e-03

Pos logit contributions
ang+0.067
irling+0.062
ug+0.060
irl+0.059
ading+0.058
Neg logit contributions
 ranger-0.037
 Something-0.035
 bleed-0.034
 bru-0.034
 sob-0.033
Token,
Token ID11
Feature activation+1.28e-03

Pos logit contributions
ang+0.011
irling+0.010
 sorts+0.010
ading+0.010
ug+0.010
Neg logit contributions
 sob-0.005
 bud-0.005
 ranger-0.005
 bru-0.005
 keeper-0.004
Token Mom
Token ID11254
Feature activation+1.21e-02

Pos logit contributions
ang+0.003
irling+0.003
ee+0.002
ug+0.002
irl+0.002
Neg logit contributions
 ranger-0.002
 bru-0.002
 sob-0.002
 bleed-0.002
 department-0.002
Token!
Token ID0
Feature activation+1.59e-02

Pos logit contributions
ang+0.026
irling+0.024
irl+0.022
ee+0.022
ading+0.022
Neg logit contributions
 ranger-0.019
 bru-0.018
 department-0.018
 bleed-0.018
 sob-0.018
Token More
Token ID3125
Feature activation+2.70e-02

Pos logit contributions
ang+0.031
irling+0.028
ee+0.027
irl+0.026
ug+0.025
Neg logit contributions
 ranger-0.027
 bleed-0.027
 bru-0.027
 hinges-0.026
 sob-0.026
Token spots
Token ID10222
Feature activation+4.90e-02

Pos logit contributions
ang+0.055
irling+0.052
ee+0.049
irl+0.048
ash+0.048
Neg logit contributions
 bud-0.040
 hinges-0.040
 bru-0.040
 bleed-0.040
 ranger-0.039
Token!"
Token ID2474
Feature activation-1.24e-02

Pos logit contributions
ang+0.094
irling+0.083
ee+0.076
irl+0.075
ug+0.074
Neg logit contributions
 bleed-0.087
 ranger-0.085
 bru-0.082
 bud-0.082
 hinges-0.081
Token<|endoftext|>
Token ID50256
Feature activation+1.05e-02

Pos logit contributions
 ranger+0.018
 bleed+0.018
 hinges+0.018
 bru+0.018
 department+0.018
Neg logit contributions
ang-0.027
irling-0.024
ee-0.023
irl-0.023
ug-0.022
TokenOnce
Token ID7454
Feature activation+2.43e-02

Pos logit contributions
ang+0.021
ee+0.020
irling+0.018
irl+0.018
ash+0.017
Neg logit contributions
 bleed-0.018
 hinges-0.018
 bru-0.017
 department-0.016
 ranger-0.016
Token upon
Token ID2402
Feature activation+2.04e-02

Pos logit contributions
ang+0.055
irling+0.051
ee+0.048
irl+0.048
ug+0.046
Neg logit contributions
 ranger-0.033
 bru-0.032
 sob-0.032
 bleed-0.032
 hinges-0.031
Token young
Token ID1862
Feature activation+1.26e-03

Pos logit contributions
ang+0.049
irling+0.043
ee+0.043
irl+0.041
ug+0.040
Neg logit contributions
 ranger-0.039
 bleed-0.038
 hinges-0.037
 department-0.036
 bud-0.036
Token,
Token ID11
Feature activation+5.10e-03

Pos logit contributions
ang+0.003
irling+0.002
ee+0.002
irl+0.002
ug+0.002
Neg logit contributions
 bleed-0.002
 ranger-0.002
 hinges-0.002
 bru-0.002
 sob-0.002
Token she
Token ID673
Feature activation+8.51e-03

Pos logit contributions
ang+0.011
irling+0.010
ee+0.009
irl+0.009
ug+0.009
Neg logit contributions
 ranger-0.008
 bleed-0.008
 bru-0.008
 sob-0.008
 hinges-0.008
Token could
Token ID714
Feature activation-2.88e-02

Pos logit contributions
ang+0.019
irling+0.017
ee+0.017
irl+0.016
ug+0.016
Neg logit contributions
 ranger-0.012
 bleed-0.012
 bru-0.012
 sob-0.012
 hinges-0.011
Token still
Token ID991
Feature activation-7.72e-03

Pos logit contributions
 ranger+0.050
 hinges+0.049
 bru+0.048
 bleed+0.048
 sob+0.048
Neg logit contributions
ang-0.054
irling-0.050
irl-0.049
ee-0.048
ading-0.045
Token make
Token ID787
Feature activation+3.06e-02

Pos logit contributions
 ranger+0.012
 bleed+0.012
 bru+0.012
 hinges+0.012
 sob+0.012
Neg logit contributions
ang-0.016
irling-0.014
ee-0.014
irl-0.014
ading-0.013
Token a
Token ID257
Feature activation-1.95e-02

Pos logit contributions
ang+0.061
irling+0.053
ee+0.051
irl+0.049
ug+0.048
Neg logit contributions
 bleed-0.053
 ranger-0.052
 sob-0.051
 bru-0.051
 hinges-0.050
Token difference
Token ID3580
Feature activation+3.64e-02

Pos logit contributions
 bleed+0.036
 ranger+0.035
 hinges+0.035
 sob+0.035
 bru+0.034
Neg logit contributions
ang-0.035
irling-0.031
ee-0.030
irl-0.029
ug-0.027
Token.
Token ID13
Feature activation-3.72e-03

Pos logit contributions
ang+0.077
irling+0.070
ee+0.068
irl+0.066
ug+0.064
Neg logit contributions
 ranger-0.056
 bleed-0.055
 bru-0.054
 hinges-0.054
 sob-0.053
Token<|endoftext|>
Token ID50256
Feature activation+1.18e-02

Pos logit contributions
 bleed+0.006
 ranger+0.005
 bru+0.005
 hinges+0.005
 department+0.005
Neg logit contributions
ang-0.008
irling-0.007
ee-0.007
irl-0.007
ug-0.006
TokenJohn
Token ID7554
Feature activation+2.74e-02

Pos logit contributions
ang+0.025
ee+0.022
irling+0.022
irl+0.021
ug+0.020
Neg logit contributions
 bleed-0.019
 hinges-0.018
 ranger-0.018
 bru-0.018
 sob-0.017
Token together
Token ID1978
Feature activation-4.19e-03

Pos logit contributions
ang+0.048
irling+0.042
ee+0.041
irl+0.038
ug+0.037
Neg logit contributions
 ranger-0.054
 bleed-0.053
 bru-0.053
 sob-0.052
 hinges-0.052
Token in
Token ID287
Feature activation+2.61e-02

Pos logit contributions
 ranger+0.005
 bleed+0.005
 bru+0.005
 sob+0.005
 hinges+0.005
Neg logit contributions
ang-0.010
irling-0.009
ee-0.009
irl-0.009
ug-0.008
Token the
Token ID262
Feature activation+3.77e-02

Pos logit contributions
ang+0.050
irling+0.043
ee+0.042
ug+0.040
irl+0.040
Neg logit contributions
 ranger-0.049
 bru-0.046
 hinges-0.045
 bleed-0.045
 aid-0.045
Token yard
Token ID12699
Feature activation-1.46e-02

Pos logit contributions
ang+0.091
irling+0.081
ug+0.080
irl+0.075
ee+0.074
Neg logit contributions
 ranger-0.055
 aid-0.050
 bru-0.048
 keeper-0.047
 hinges-0.047
Token.
Token ID13
Feature activation+3.53e-03

Pos logit contributions
 bleed+0.017
 ranger+0.017
 bru+0.017
 hinges+0.016
 sob+0.016
Neg logit contributions
ang-0.036
irling-0.033
ee-0.032
irl-0.031
ug-0.031
Token The
Token ID383
Feature activation+6.11e-03

Pos logit contributions
ang+0.009
irling+0.009
ee+0.008
irl+0.008
ug+0.008
Neg logit contributions
 ranger-0.004
 bru-0.004
 hinges-0.004
 Something-0.004
 department-0.004
Token cat
Token ID3797
Feature activation+1.18e-02

Pos logit contributions
ang+0.012
irling+0.010
ee+0.010
irl+0.010
ug+0.009
Neg logit contributions
 ranger-0.011
 bleed-0.011
 bru-0.011
 hinges-0.011
 sob-0.011
Token was
Token ID373
Feature activation+2.59e-02

Pos logit contributions
ang+0.025
irling+0.022
ee+0.022
irl+0.021
ug+0.021
Neg logit contributions
 ranger-0.018
 bleed-0.018
 bru-0.018
 sob-0.018
 hinges-0.018
Token always
Token ID1464
Feature activation+2.44e-02

Pos logit contributions
ang+0.055
ee+0.049
irling+0.047
ug+0.047
ash+0.047
Neg logit contributions
 ranger-0.042
 sob-0.039
 bleed-0.039
 bru-0.039
 keeper-0.037
Token very
Token ID845
Feature activation+1.90e-02

Pos logit contributions
ang+0.048
ee+0.044
irling+0.042
irl+0.042
ug+0.041
Neg logit contributions
 ranger-0.042
 sob-0.040
 bleed-0.039
 bru-0.039
 keeper-0.038
Token curious
Token ID11040
Feature activation-1.94e-02

Pos logit contributions
ang+0.037
ee+0.034
irling+0.034
irl+0.033
angles+0.033
Neg logit contributions
 ranger-0.032
 bleed-0.030
 sob-0.029
 bru-0.028
 keeper-0.028
Token it
Token ID340
Feature activation-1.90e-02

Pos logit contributions
 bleed+0.014
 ranger+0.014
 bru+0.013
 sob+0.013
 hinges+0.013
Neg logit contributions
ang-0.018
ee-0.015
irling-0.015
ug-0.014
ash-0.014
Token on
Token ID319
Feature activation+3.74e-03

Pos logit contributions
 ranger+0.034
 bleed+0.033
 bru+0.033
 hinges+0.033
 sob+0.033
Neg logit contributions
ang-0.036
irling-0.032
ee-0.031
irl-0.030
ug-0.029
Token a
Token ID257
Feature activation-7.64e-03

Pos logit contributions
ang+0.007
irling+0.006
ee+0.006
irl+0.006
ug+0.006
Neg logit contributions
 ranger-0.007
 bleed-0.007
 bru-0.007
 hinges-0.007
 sob-0.007
Token table
Token ID3084
Feature activation-5.02e-03

Pos logit contributions
 bleed+0.015
 ranger+0.014
 bru+0.014
 hinges+0.014
 sob+0.014
Neg logit contributions
ang-0.013
irling-0.012
ee-0.011
irl-0.011
ug-0.011
Token where
Token ID810
Feature activation+2.36e-02

Pos logit contributions
 bleed+0.008
 bru+0.007
 sob+0.007
 hinges+0.007
 ranger+0.007
Neg logit contributions
ang-0.011
irling-0.010
ee-0.010
irl-0.009
ug-0.009
Token the
Token ID262
Feature activation+8.28e-03

Pos logit contributions
ang+0.045
irling+0.040
ee+0.038
irl+0.036
ug+0.035
Neg logit contributions
 ranger-0.043
 bleed-0.042
 bru-0.041
 hinges-0.041
 sob-0.041
Token dog
Token ID3290
Feature activation+1.51e-03

Pos logit contributions
ang+0.014
ee+0.012
irling+0.012
irl+0.012
ading+0.011
Neg logit contributions
 bleed-0.016
 sob-0.015
 bru-0.015
 ranger-0.015
 hinges-0.015
Token cannot
Token ID2314
Feature activation-2.50e-02

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ug+0.002
Neg logit contributions
 ranger-0.002
 bleed-0.002
 bru-0.002
 hinges-0.002
 sob-0.002
Token reach
Token ID3151
Feature activation+2.59e-02

Pos logit contributions
 sob+0.042
 bru+0.042
 ranger+0.042
 bleed+0.042
 hinges+0.041
Neg logit contributions
ang-0.044
ee-0.042
irl-0.041
irling-0.041
ug-0.039
Token it
Token ID340
Feature activation-3.37e-02

Pos logit contributions
ang+0.055
irling+0.050
ee+0.048
irl+0.047
ug+0.045
Neg logit contributions
 ranger-0.040
 bleed-0.040
 bru-0.039
 hinges-0.038
 sob-0.038
Token.
Token ID13
Feature activation-4.87e-03

Pos logit contributions
 bru+0.052
 ranger+0.050
 sob+0.050
 bleed+0.049
 injuries+0.048
Neg logit contributions
ang-0.069
ee-0.062
irl-0.062
irling-0.061
ash-0.060
Token.
Token ID13
Feature activation+1.56e-02

Pos logit contributions
ang+0.086
irling+0.078
ee+0.075
irl+0.074
ug+0.069
Neg logit contributions
 bleed-0.075
 bru-0.073
 hinges-0.073
 ranger-0.072
 sob-0.070
Token And
Token ID843
Feature activation+2.04e-02

Pos logit contributions
ang+0.042
irling+0.040
ee+0.039
irl+0.039
ading+0.037
Neg logit contributions
 ranger-0.016
 sob-0.013
 bleed-0.013
 bru-0.013
 keeper-0.013
Token that
Token ID326
Feature activation+2.73e-02

Pos logit contributions
ang+0.040
irling+0.037
ee+0.035
irl+0.034
ash+0.032
Neg logit contributions
 ranger-0.037
 bleed-0.034
 sob-0.034
 keeper-0.034
 bru-0.034
Token is
Token ID318
Feature activation+1.55e-02

Pos logit contributions
ang+0.066
irling+0.059
ug+0.057
ash+0.055
ading+0.055
Neg logit contributions
 ranger-0.033
 keeper-0.031
 sob-0.030
 truth-0.027
 cough-0.027
Token the
Token ID262
Feature activation+9.53e-03

Pos logit contributions
ang+0.029
irling+0.026
ee+0.025
ug+0.023
irl+0.023
Neg logit contributions
 ranger-0.029
 keeper-0.027
 department-0.027
 bru-0.027
 bleed-0.027
Token moral
Token ID6573
Feature activation+2.23e-03

Pos logit contributions
ang+0.012
irling+0.010
ee+0.010
irl+0.010
ading+0.009
Neg logit contributions
 bleed-0.023
 hinges-0.022
 bru-0.022
 sob-0.022
 ranger-0.022
Token of
Token ID286
Feature activation+3.45e-02

Pos logit contributions
ang+0.003
ee+0.002
irling+0.002
irl+0.002
ug+0.002
Neg logit contributions
 bleed-0.006
 bru-0.005
 hinges-0.005
 sob-0.005
 ranger-0.005
Token the
Token ID262
Feature activation+1.14e-02

Pos logit contributions
ang+0.065
irling+0.061
irl+0.056
ee+0.056
ug+0.052
Neg logit contributions
 ranger-0.064
 bru-0.059
 sob-0.059
 keeper-0.058
 mercy-0.058
Token story
Token ID1621
Feature activation-2.56e-02

Pos logit contributions
ang+0.028
irling+0.026
ee+0.025
irl+0.024
ug+0.024
Neg logit contributions
 ranger-0.015
 bleed-0.014
 bru-0.014
 sob-0.013
 hinges-0.013
Token:
Token ID25
Feature activation-1.41e-02

Pos logit contributions
 bleed+0.041
 bru+0.040
 hinges+0.039
 ranger+0.038
 sob+0.038
Neg logit contributions
ang-0.054
ee-0.048
irling-0.048
irl-0.046
ug-0.045
Token sharing
Token ID7373
Feature activation+4.83e-02

Pos logit contributions
 ranger+0.023
 bru+0.022
 bleed+0.022
 sob+0.021
 hinges+0.021
Neg logit contributions
ang-0.030
irling-0.027
ee-0.026
irl-0.025
ug-0.024
Token are
Token ID389
Feature activation+3.40e-02

Pos logit contributions
keeping+0.100
edly+0.093
 ranger+0.091
 hinges+0.090
 recovered+0.087
Neg logit contributions
ang-0.098
ee-0.085
irling-0.084
irl-0.083
ch-0.077
Token sorry
Token ID7926
Feature activation-3.01e-02

Pos logit contributions
ang+0.066
irling+0.057
ee+0.056
irl+0.054
ug+0.052
Neg logit contributions
 ranger-0.061
 bleed-0.060
 bru-0.059
 sob-0.059
 hinges-0.057
Token,
Token ID11
Feature activation-1.42e-03

Pos logit contributions
 hinges+0.050
 bleed+0.050
 ranger+0.050
 bru+0.049
 department+0.047
Neg logit contributions
ang-0.057
ee-0.051
irling-0.050
irl-0.048
ug-0.048
Token m
Token ID285
Feature activation+4.15e-02

Pos logit contributions
 hinges+0.002
 bleed+0.002
 recovered+0.002
 bru+0.002
 department+0.002
Neg logit contributions
ang-0.003
irling-0.002
ee-0.002
irl-0.002
ug-0.002
Tokenister
Token ID1694
Feature activation+8.03e-03

Pos logit contributions
ang+0.073
irling+0.063
ee+0.062
irl+0.058
ug+0.057
Neg logit contributions
 ranger-0.081
 bleed-0.081
 hinges-0.080
 bru-0.079
 sob-0.078
Token,"
Token ID553
Feature activation-1.10e-02

Pos logit contributions
ang+0.018
ee+0.016
irling+0.015
ug+0.015
irl+0.014
Neg logit contributions
 hinges-0.012
 bleed-0.012
 bru-0.011
 sob-0.011
 ranger-0.011
Token Lily
Token ID20037
Feature activation-3.97e-03

Pos logit contributions
 hinges+0.021
 bleed+0.021
 department+0.020
 injuries+0.019
 ranger+0.019
Neg logit contributions
ang-0.020
ee-0.017
irling-0.016
irl-0.015
ug-0.015
Token said
Token ID531
Feature activation-4.31e-02

Pos logit contributions
 hinges+0.009
 injuries+0.009
 bleed+0.008
 ranger+0.008
keeping+0.008
Neg logit contributions
ang-0.005
ee-0.005
irling-0.005
irl-0.005
aw-0.004
Token.
Token ID13
Feature activation-2.57e-02

Pos logit contributions
 hinges+0.077
 ranger+0.076
 department+0.073
 bleed+0.072
 bru+0.071
Neg logit contributions
ang-0.076
irling-0.073
ug-0.068
irl-0.065
ee-0.064
Token "
Token ID366
Feature activation+1.99e-03

Pos logit contributions
 bleed+0.048
 hinges+0.046
 mercy+0.046
 recovered+0.046
 injuries+0.045
Neg logit contributions
ang-0.048
irling-0.040
ee-0.037
ug-0.037
irl-0.037
TokenWe
Token ID1135
Feature activation+6.15e-03

Pos logit contributions
ang+0.004
irling+0.003
ee+0.003
irl+0.003
ug+0.003
Neg logit contributions
 hinges-0.004
 ranger-0.004
 department-0.003
 bru-0.003
 bleed-0.003
Token to
Token ID284
Feature activation-5.39e-02

Pos logit contributions
ang+0.012
irling+0.012
ee+0.011
irl+0.011
ug+0.011
Neg logit contributions
 ranger-0.004
 bleed-0.004
 bru-0.004
 hinges-0.004
 sob-0.004
Token know
Token ID760
Feature activation-1.03e-02

Pos logit contributions
 ranger+0.091
 department+0.089
 bru+0.087
 sob+0.087
 bleed+0.086
Neg logit contributions
ang-0.094
irl-0.093
ee-0.092
irling-0.089
ading-0.081
Token what
Token ID644
Feature activation+1.35e-03

Pos logit contributions
 ranger+0.018
 hinges+0.017
 department+0.017
 mercy+0.017
 sob+0.017
Neg logit contributions
ang-0.021
irling-0.018
ee-0.017
irl-0.017
ug-0.016
Token you
Token ID345
Feature activation+9.24e-03

Pos logit contributions
ang+0.003
irling+0.002
ee+0.002
irl+0.002
ug+0.002
Neg logit contributions
 ranger-0.002
 bleed-0.002
 bru-0.002
 sob-0.002
 hinges-0.002
Token see
Token ID766
Feature activation+2.60e-02

Pos logit contributions
ang+0.019
irling+0.017
ee+0.017
irl+0.016
ug+0.015
Neg logit contributions
 ranger-0.015
 bleed-0.015
 bru-0.015
 hinges-0.014
 sob-0.014
Token."
Token ID526
Feature activation-1.51e-02

Pos logit contributions
ang+0.061
irling+0.057
irl+0.054
ee+0.053
ug+0.051
Neg logit contributions
 ranger-0.033
 hinges-0.032
 sob-0.031
 bleed-0.031
 bru-0.031
Token Mary
Token ID5335
Feature activation+1.84e-02

Pos logit contributions
 bleed+0.028
 injuries+0.026
 bru+0.025
 department+0.025
 hinges+0.025
Neg logit contributions
ang-0.029
ee-0.026
irling-0.024
aw-0.024
 sorts-0.023
Token stepped
Token ID10764
Feature activation+3.38e-02

Pos logit contributions
ang+0.038
ee+0.034
irling+0.034
irl+0.032
ug+0.030
Neg logit contributions
 bleed-0.030
 hinges-0.029
 bru-0.028
 ranger-0.028
 injuries-0.028
Token closer
Token ID5699
Feature activation-3.85e-02

Pos logit contributions
ang+0.068
irling+0.059
ee+0.058
ading+0.055
irl+0.054
Neg logit contributions
 bleed-0.060
 bru-0.057
 ranger-0.057
 hinges-0.056
onna-0.054
Token and
Token ID290
Feature activation+6.49e-03

Pos logit contributions
 bleed+0.082
 bru+0.076
 injuries+0.073
 mercy+0.073
 choke+0.073
Neg logit contributions
ang-0.061
irling-0.051
ee-0.050
aw-0.047
ing-0.045
Token saw
Token ID2497
Feature activation+1.06e-02

Pos logit contributions
ang+0.014
irling+0.013
ee+0.012
irl+0.012
ug+0.011
Neg logit contributions
 ranger-0.010
 bleed-0.010
 bru-0.010
 hinges-0.010
 sob-0.009
Token He
Token ID679
Feature activation-6.46e-03

Pos logit contributions
 bleed+0.040
 injuries+0.037
ache+0.037
 recovered+0.036
 mercy+0.036
Neg logit contributions
ang-0.036
ee-0.027
ug-0.027
aw-0.026
irling-0.026
Token smiled
Token ID13541
Feature activation+6.75e-03

Pos logit contributions
 bleed+0.011
 hinges+0.010
 bru+0.010
 ranger+0.010
 injuries+0.010
Neg logit contributions
ang-0.013
irling-0.012
ee-0.012
irl-0.011
ug-0.011
Token and
Token ID290
Feature activation+8.41e-03

Pos logit contributions
ang+0.010
irling+0.008
aw+0.008
ee+0.008
ug+0.008
Neg logit contributions
 bleed-0.015
 bru-0.014
 hinges-0.014
keeping-0.014
onna-0.014
Token squeezed
Token ID29650
Feature activation+4.48e-02

Pos logit contributions
ang+0.019
irling+0.017
ee+0.017
irl+0.016
ug+0.016
Neg logit contributions
 ranger-0.012
 bleed-0.012
 sob-0.012
 bru-0.011
 hinges-0.011
Token Lily
Token ID20037
Feature activation-5.94e-02

Pos logit contributions
ang+0.088
irling+0.078
ee+0.076
irl+0.073
ug+0.071
Neg logit contributions
 ranger-0.078
 bleed-0.077
 bru-0.076
 hinges-0.075
 sob-0.075
Token's
Token ID338
Feature activation-5.16e-02

Pos logit contributions
 ranger+0.116
 bru+0.115
 hinges+0.115
 department+0.113
 bleed+0.112
Neg logit contributions
ang-0.101
ee-0.087
irling-0.086
irl-0.080
ug-0.077
Token hand
Token ID1021
Feature activation-2.80e-02

Pos logit contributions
 department+0.104
 ranger+0.096
 recovered+0.095
onna+0.094
ache+0.090
Neg logit contributions
ang-0.080
ee-0.076
irling-0.064
irl-0.063
ading-0.062
Token gently
Token ID15165
Feature activation-4.57e-02

Pos logit contributions
 ranger+0.041
 bru+0.041
 hinges+0.040
 bleed+0.040
 department+0.039
Neg logit contributions
ang-0.060
ee-0.053
irling-0.052
ug-0.050
ash-0.049
Token.
Token ID13
Feature activation-1.20e-02

Pos logit contributions
 department+0.081
 ranger+0.080
 hinges+0.076
 mercy+0.076
 bru+0.074
Neg logit contributions
ang-0.083
irling-0.071
ee-0.071
ug-0.066
irl-0.065
Token They
Token ID1119
Feature activation+1.00e-02

Pos logit contributions
 bleed+0.022
ache+0.020
 mercy+0.020
 hinges+0.020
 injuries+0.020
Neg logit contributions
ang-0.023
irling-0.019
irl-0.018
aw-0.018
ee-0.018
Token entered
Token ID5982
Feature activation+3.05e-02

Pos logit contributions
ang+0.020
irling+0.017
ee+0.017
irl+0.017
ug+0.016
Neg logit contributions
 ranger-0.017
 bleed-0.017
 hinges-0.017
 bru-0.017
 department-0.016
Token up
Token ID510
Feature activation+2.16e-04

Pos logit contributions
 ranger+0.095
 bleed+0.095
 bru+0.095
 hinges+0.093
 sob+0.093
Neg logit contributions
ang-0.051
irling-0.042
ee-0.041
irl-0.039
ug-0.036
Token and
Token ID290
Feature activation+2.74e-03

Pos logit contributions
ang+0.000
ee+0.000
irling+0.000
irl+0.000
aw+0.000
Neg logit contributions
 bleed-0.000
 department-0.000
 bru-0.000
 hinges-0.000
 ranger-0.000
Token ran
Token ID4966
Feature activation+5.73e-03

Pos logit contributions
ang+0.006
irling+0.005
ee+0.005
irl+0.005
ug+0.005
Neg logit contributions
 ranger-0.004
 bleed-0.004
 bru-0.004
 sob-0.004
 hinges-0.004
Token to
Token ID284
Feature activation-3.64e-02

Pos logit contributions
ang+0.012
ee+0.010
irling+0.010
ash+0.010
irl+0.010
Neg logit contributions
 bru-0.009
 bleed-0.009
 sob-0.009
ache-0.008
 mercy-0.008
Token her
Token ID607
Feature activation-5.13e-02

Pos logit contributions
 ranger+0.077
 bleed+0.075
 hinges+0.075
 bru+0.074
 department+0.074
Neg logit contributions
ang-0.056
irling-0.049
ee-0.048
irl-0.047
ug-0.042
Token mom
Token ID1995
Feature activation-6.83e-02

Pos logit contributions
 bleed+0.111
 recovered+0.106
 mercy+0.104
 bru+0.103
ache+0.102
Neg logit contributions
ang-0.078
irling-0.064
irl-0.062
ee-0.059
utes-0.052
Token.
Token ID13
Feature activation-2.01e-02

Pos logit contributions
 bleed+0.096
 bru+0.094
 ranger+0.093
 hinges+0.090
 sob+0.088
Neg logit contributions
ang-0.156
irling-0.138
ee-0.137
irl-0.131
ash-0.130
Token
Token ID198
Feature activation+1.36e-03

Pos logit contributions
 bleed+0.036
 injuries+0.035
 mercy+0.035
ache+0.034
 bru+0.034
Neg logit contributions
ang-0.037
ee-0.031
irling-0.030
irl-0.030
ug-0.029
Token
Token ID198
Feature activation-2.20e-03

Pos logit contributions
ang+0.003
irling+0.003
ee+0.003
irl+0.003
ug+0.003
Neg logit contributions
 bleed-0.002
 ranger-0.002
 bru-0.002
 hinges-0.002
 department-0.002
TokenShe
Token ID3347
Feature activation-1.51e-02

Pos logit contributions
 ranger+0.003
 bleed+0.003
 sob+0.003
 hinges+0.003
 bru+0.003
Neg logit contributions
ang-0.005
irling-0.005
ee-0.005
irl-0.004
ug-0.004
Token gave
Token ID2921
Feature activation+4.17e-02

Pos logit contributions
 ranger+0.023
 bleed+0.023
 bru+0.022
 sob+0.022
 hinges+0.022
Neg logit contributions
ang-0.032
irling-0.029
ee-0.028
irl-0.028
ug-0.027
Token my
Token ID616
Feature activation-3.48e-02

Pos logit contributions
ang+0.013
irling+0.011
ee+0.010
irl+0.010
ug+0.009
Neg logit contributions
 ranger-0.028
 bleed-0.028
 bru-0.028
 hinges-0.027
 sob-0.027
Token toy
Token ID13373
Feature activation-2.03e-02

Pos logit contributions
 bleed+0.066
 recovered+0.065
 ranger+0.064
 department+0.063
 Something+0.062
Neg logit contributions
ang-0.056
irling-0.048
ee-0.048
irl-0.048
ading-0.041
Token,"
Token ID553
Feature activation-2.79e-02

Pos logit contributions
 bleed+0.037
 bru+0.036
 recovered+0.036
 ranger+0.035
 Something+0.035
Neg logit contributions
ang-0.037
ug-0.030
ee-0.030
irling-0.028
irl-0.027
Token said
Token ID531
Feature activation-3.32e-02

Pos logit contributions
 hinges+0.050
 bleed+0.050
 ranger+0.049
 department+0.049
 bru+0.049
Neg logit contributions
ang-0.050
irling-0.044
ee-0.043
irl-0.042
ug-0.040
Token the
Token ID262
Feature activation-2.03e-02

Pos logit contributions
 hinges+0.071
 bleed+0.069
 ranger+0.067
onna+0.067
 bru+0.066
Neg logit contributions
ang-0.048
ee-0.041
ug-0.039
irling-0.039
irl-0.038
Token boy
Token ID2933
Feature activation-5.91e-02

Pos logit contributions
onna+0.042
 bleed+0.042
 injuries+0.041
 hinges+0.041
 recovered+0.040
Neg logit contributions
ang-0.030
ee-0.026
irling-0.023
aw-0.022
irl-0.022
Token with
Token ID351
Feature activation+2.53e-02

Pos logit contributions
 hinges+0.081
 department+0.080
 bru+0.080
 ranger+0.079
 bleed+0.079
Neg logit contributions
ang-0.136
irling-0.118
ee-0.113
ug-0.113
irl-0.110
Token a
Token ID257
Feature activation-5.13e-02

Pos logit contributions
ang+0.045
irling+0.040
ee+0.040
irl+0.039
ug+0.036
Neg logit contributions
 ranger-0.046
 department-0.045
 hinges-0.044
 bru-0.044
onna-0.044
Token sad
Token ID6507
Feature activation-5.71e-02

Pos logit contributions
keeping+0.109
 department+0.098
 hinges+0.098
 injuries+0.097
edly+0.097
Neg logit contributions
ang-0.078
irling-0.071
ee-0.070
irl-0.070
 sorts-0.061
Token face
Token ID1986
Feature activation-5.60e-02

Pos logit contributions
 department+0.105
 hinges+0.096
 injuries+0.095
keeping+0.095
 ranger+0.088
Neg logit contributions
ang-0.094
ee-0.083
ug-0.080
irl-0.080
irling-0.079
Token.
Token ID13
Feature activation-3.63e-02

Pos logit contributions
 ranger+0.075
 bleed+0.073
 bru+0.072
 hinges+0.072
 department+0.070
Neg logit contributions
ang-0.131
irling-0.117
ee-0.113
ug-0.112
irl-0.109
Token back
Token ID736
Feature activation-3.27e-02

Pos logit contributions
 bru+0.073
 bleed+0.071
 department+0.070
 ranger+0.070
 hinges+0.069
Neg logit contributions
ang-0.090
ee-0.081
irling-0.076
ug-0.073
irl-0.072
Token to
Token ID284
Feature activation-2.96e-02

Pos logit contributions
 bleed+0.039
 ranger+0.039
 bru+0.039
 sob+0.037
 hinges+0.037
Neg logit contributions
ang-0.082
irling-0.073
ee-0.073
irl-0.070
ug-0.069
Token the
Token ID262
Feature activation-3.08e-03

Pos logit contributions
 sob+0.049
 bru+0.049
 bleed+0.048
 ranger+0.048
 injuries+0.047
Neg logit contributions
ang-0.060
irling-0.053
ee-0.050
ug-0.049
ash-0.049
Token owl
Token ID39610
Feature activation-4.89e-02

Pos logit contributions
 bleed+0.004
 department+0.004
 ranger+0.004
 bru+0.004
 sob+0.004
Neg logit contributions
ang-0.007
irling-0.006
ee-0.006
irl-0.006
ug-0.006
Token and
Token ID290
Feature activation-5.44e-03

Pos logit contributions
 bleed+0.067
 bru+0.065
 ranger+0.065
 hinges+0.064
 department+0.063
Neg logit contributions
ang-0.114
irling-0.101
ee-0.098
irl-0.094
ug-0.094
Token said
Token ID531
Feature activation-4.34e-02

Pos logit contributions
 bleed+0.009
 hinges+0.009
 bru+0.009
 ranger+0.009
 department+0.009
Neg logit contributions
ang-0.010
irling-0.009
ee-0.009
irl-0.009
ug-0.008
Token,
Token ID11
Feature activation-4.68e-02

Pos logit contributions
 bleed+0.090
 hinges+0.089
 ranger+0.088
 injuries+0.086
 bru+0.085
Neg logit contributions
ang-0.067
irling-0.061
ee-0.057
ug-0.056
irl-0.056
Token "
Token ID366
Feature activation+1.10e-02

Pos logit contributions
 bleed+0.078
 hinges+0.075
 ranger+0.075
 bru+0.074
 injuries+0.071
Neg logit contributions
ang-0.094
irling-0.085
ee-0.080
irl-0.078
ug-0.077
TokenThank
Token ID10449
Feature activation+8.12e-03

Pos logit contributions
ang+0.024
irling+0.022
ee+0.021
irl+0.021
ug+0.020
Neg logit contributions
 ranger-0.016
 bleed-0.016
 bru-0.016
 sob-0.016
 hinges-0.015
Token you
Token ID345
Feature activation-3.07e-02

Pos logit contributions
ang+0.012
irling+0.011
irl+0.010
ash+0.010
angles+0.010
Neg logit contributions
 ranger-0.017
 mercy-0.016
 bleed-0.016
 sob-0.016
 bru-0.016
Token!".
Token ID48220
Feature activation-1.62e-02

Pos logit contributions
 ranger+0.052
 bleed+0.051
 bru+0.050
 hinges+0.050
 sob+0.049
Neg logit contributions
ang-0.062
irling-0.056
ee-0.053
irl-0.052
ug-0.050
Token heard
Token ID2982
Feature activation-3.78e-03

Pos logit contributions
 bleed+0.044
 hinges+0.043
onna+0.041
keeping+0.040
 bru+0.040
Neg logit contributions
ang-0.041
ee-0.038
irling-0.036
aw-0.036
ug-0.032
Token Sara
Token ID24799
Feature activation-6.47e-02

Pos logit contributions
 ranger+0.007
 bleed+0.007
 hinges+0.007
 bru+0.007
 department+0.007
Neg logit contributions
ang-0.006
irling-0.006
ee-0.005
irl-0.005
ug-0.005
Token crying
Token ID13774
Feature activation-2.81e-02

Pos logit contributions
 ranger+0.113
 hinges+0.112
 bru+0.110
 department+0.108
onna+0.107
Neg logit contributions
ang-0.121
ee-0.106
irling-0.105
ug-0.097
irl-0.097
Token and
Token ID290
Feature activation-4.01e-02

Pos logit contributions
onna+0.059
 department+0.057
 ranger+0.056
 hinges+0.056
 bru+0.054
Neg logit contributions
ang-0.044
aw-0.037
ee-0.036
ash-0.034
irling-0.034
Token came
Token ID1625
Feature activation-1.96e-02

Pos logit contributions
 hinges+0.063
 bleed+0.062
 ranger+0.062
 bru+0.060
 department+0.059
Neg logit contributions
ang-0.082
irling-0.072
ee-0.071
irl-0.066
ug-0.066
Token to
Token ID284
Feature activation-4.83e-02

Pos logit contributions
 bleed+0.025
 bru+0.024
 ranger+0.024
ache+0.023
 hinges+0.022
Neg logit contributions
ang-0.048
irling-0.043
ee-0.042
ug-0.040
irl-0.040
Token her
Token ID607
Feature activation-6.51e-02

Pos logit contributions
ache+0.115
 bleed+0.109
 hinges+0.109
driver+0.107
 department+0.105
Neg logit contributions
ang-0.054
ee-0.050
 these-0.046
 us-0.043
rous-0.043
Token.
Token ID13
Feature activation-3.11e-02

Pos logit contributions
 bleed+0.125
 bru+0.123
 recovered+0.123
onna+0.122
ache+0.121
Neg logit contributions
ang-0.107
irling-0.089
irl-0.083
ee-0.081
 sorts-0.077
Token She
Token ID1375
Feature activation-3.28e-02

Pos logit contributions
 bleed+0.051
 injuries+0.051
 mercy+0.050
 recovered+0.049
 department+0.049
Neg logit contributions
ang-0.060
ee-0.049
aw-0.046
irling-0.046
uff-0.042
Token saw
Token ID2497
Feature activation+1.36e-02

Pos logit contributions
 bleed+0.057
 hinges+0.054
 bru+0.054
 ranger+0.053
onna+0.052
Neg logit contributions
ang-0.065
irling-0.057
ee-0.055
irl-0.052
ug-0.050
Token Sara
Token ID24799
Feature activation-4.52e-02

Pos logit contributions
ang+0.025
irling+0.023
ee+0.022
irl+0.021
ug+0.020
Neg logit contributions
 ranger-0.025
 bleed-0.024
 bru-0.024
 sob-0.024
 hinges-0.024