TOP ACTIVATIONS
MAX = 0.333

bird can be happy in heaven ." Her mom smiled and
to go to do ggy heaven . Sarah was so sad
like all the angels in heaven were singing .
now . You are in heaven . You are an angel
knew that Teddy was in heaven and that he would always
have to go back to heaven ," said the angel .
had to go back to heaven again . Lily waved goodbye
was dancing and drawing in heaven . <|endoftext|> Once upon a
He is with God in heaven . He still loves us
had to go to squirrel heaven . His friends were very
it was happy in bug heaven . <|endoftext|> One day ,
knew she was now in heaven , dancing with the angels
thanked the big hand in heaven for sending it such a
it . It tasted like heaven ! R alph
said he had a rare disease that made him allergic to
cure for a very bad disease . You will be heroes
at the sign and the speaker . She felt silly and
hear your voice on the speaker ." Tom my
to the shore and c ork ed the bottle tightly .
a gift , not a curse . They learned that they

MIN ACTIVATIONS
MIN = -0.296

some ac orns and pine cones on them . They made
of ac orns and pine cones on the shelf . They
ips . They see pine cones , ac orns , nuts
. The bear takes the cones . The bear leaves .
They reached the cones and the boss y pony
balls of ice cream on cones for everyone ! Derek was
, they threw some pine cones on top to make it
Then they put some pine cones on the top to make
nods and gives her two cones . Anna pays with her
have a ball and two cones for the goal . Sara
Tom and handed them the cones . " Here
them ice - cream in cones . He is nice .
mom pulled two ice cream cones out of her bag .
tree and look for the cones . They find some on
gift . He took two cones and put a small scoop
disappeared , and the two cones were happy . Whenever the
a special places that sold cones . Alex looked at all
scream . They drop the cones . They try to spring
cones . They picked some cones and smelled them . They
looked at all of the cones to find the perfect one

INTERVAL 0.176 - 0.333
CONTAINS 0.322%

's daddy gave her a lecture about taking care of her
too late . The bad thing had already vanished , taking
smiled . She knew the wish was safe inside until she
creature . The woman took her furry friend to
to go fishing again , despite her husband 's warnings .

INTERVAL 0.019 - 0.176
CONTAINS 38.810%

They smiled . Their moms and dads smiled , too .
It wanted to be free and explore , so it started
just an ordinary pitch . You can go and play here
They lit them with a match . They thought it would
learned at school . He hum med one of them as

INTERVAL -0.139 - 0.019
CONTAINS 58.572%

show it to her friends . <|endoftext|> Once upon a time
hair is boring and plain ." Anna says ,
basketball . He had a big , hard basketball that he
medicine . He will feel better soon ." L
're too small to get the food from the tree !"

INTERVAL -0.296 - -0.139
CONTAINS 2.297%

asked her mom if she could stay close to the shore
They have a lot of fun and laugh a lot .
soon had a great big pile of sand . He started
you are and where you go . They go
a time , there were two best friends named Paul and
Token bird
Token ID6512
Feature activation+3.54e-02

Pos logit contributions
 wound+0.010
 tears+0.010
 creature+0.010
eros+0.009
 hospital+0.009
Neg logit contributions
First-0.025
 topp-0.025
 forks-0.024
 apr-0.023
avored-0.023
Token can
Token ID460
Feature activation-1.28e-01

Pos logit contributions
ez+0.067
eros+0.065
 wound+0.064
-+0.063
 heaven+0.063
Neg logit contributions
First-0.036
 topp-0.036
 ju-0.034
 flour-0.031
 flavor-0.031
Token be
Token ID307
Feature activation+3.75e-03

Pos logit contributions
 topp+0.198
First+0.187
 ju+0.184
 forks+0.176
 flour+0.174
Neg logit contributions
ez-0.177
 wound-0.176
eros-0.173
--0.171
 its-0.169
Token happy
Token ID3772
Feature activation+4.02e-02

Pos logit contributions
ez+0.005
eros+0.004
 wound+0.004
 tears+0.004
 heaven+0.004
Neg logit contributions
 topp-0.006
First-0.006
 ju-0.006
 flavor-0.006
 flour-0.006
Token in
Token ID287
Feature activation-5.06e-02

Pos logit contributions
 its+0.034
-+0.034
ez+0.033
 heaven+0.033
 knowing+0.032
Neg logit contributions
 topp-0.082
First-0.082
 ju-0.078
 flour-0.074
 apr-0.074
Token heaven
Token ID9538
Feature activation+3.33e-01

Pos logit contributions
 topp+0.106
First+0.105
 ju+0.104
 flavor+0.098
 forks+0.097
Neg logit contributions
ez-0.042
eros-0.039
 heaven-0.039
 its-0.038
 wound-0.037
Token."
Token ID526
Feature activation-2.12e-02

Pos logit contributions
ez+0.354
eros+0.346
 heaven+0.343
-+0.338
 wound+0.338
Neg logit contributions
 topp-0.600
First-0.594
 ju-0.579
 flour-0.548
 flavor-0.546
Token Her
Token ID2332
Feature activation+5.75e-02

Pos logit contributions
First+0.033
 topp+0.030
 ju+0.030
 flour+0.029
 First+0.029
Neg logit contributions
eros-0.030
ez-0.030
 creature-0.029
 terrified-0.029
crow-0.028
Token mom
Token ID1995
Feature activation-4.22e-02

Pos logit contributions
 tears+0.067
 wound+0.066
 crying+0.061
eros+0.060
-+0.057
Neg logit contributions
First-0.113
 topp-0.107
 ju-0.106
avored-0.094
teen-0.094
Token smiled
Token ID13541
Feature activation+2.16e-02

Pos logit contributions
First+0.074
 topp+0.074
 ju+0.073
 flour+0.068
 forks+0.068
Neg logit contributions
ez-0.049
eros-0.046
 wound-0.045
--0.045
 tears-0.044
Token and
Token ID290
Feature activation+2.72e-02

Pos logit contributions
ez+0.027
eros+0.027
crow+0.025
lves+0.023
 speaker+0.023
Neg logit contributions
 ju-0.036
First-0.035
 would-0.034
 topp-0.033
 yes-0.033
Token to
Token ID284
Feature activation-4.25e-02

Pos logit contributions
 topp+0.075
First+0.074
 ju+0.073
 forks+0.068
 flavor+0.068
Neg logit contributions
ez-0.056
 tears-0.053
 wound-0.053
 crying-0.051
 its-0.050
Token go
Token ID467
Feature activation-1.50e-01

Pos logit contributions
 topp+0.060
 ju+0.060
First+0.057
 forks+0.053
 flour+0.051
Neg logit contributions
 wound-0.065
ez-0.063
 heaven-0.063
 tears-0.063
 its-0.062
Token to
Token ID284
Feature activation-4.16e-02

Pos logit contributions
First+0.180
 usual+0.164
 stack+0.164
 would+0.161
 count+0.159
Neg logit contributions
ez-0.273
eros-0.251
oc-0.238
ment-0.237
 realizes-0.229
Token do
Token ID466
Feature activation-3.64e-02

Pos logit contributions
First+0.069
 topp+0.067
 ju+0.065
 flavor+0.065
 flour+0.064
Neg logit contributions
ez-0.051
eros-0.049
oring-0.046
les-0.045
crow-0.045
Tokenggy
Token ID19970
Feature activation+1.33e-01

Pos logit contributions
 topp+0.061
First+0.060
 flavor+0.057
 flour+0.056
 ju+0.055
Neg logit contributions
eros-0.044
ez-0.041
crow-0.040
oring-0.038
an-0.038
Token heaven
Token ID9538
Feature activation+3.33e-01

Pos logit contributions
ez+0.138
eros+0.130
 wound+0.126
 tears+0.122
 heaven+0.122
Neg logit contributions
 topp-0.247
First-0.244
 ju-0.240
 flavor-0.227
 flour-0.226
Token.
Token ID13
Feature activation-5.29e-03

Pos logit contributions
ez+0.365
eros+0.347
 wound+0.326
 tears+0.324
 terrified+0.317
Neg logit contributions
 topp-0.592
First-0.590
 ju-0.576
 flavor-0.552
 forks-0.541
Token Sarah
Token ID10490
Feature activation+2.25e-02

Pos logit contributions
First+0.009
 ju+0.008
 topp+0.008
 First+0.008
 stack+0.007
Neg logit contributions
ez-0.007
eros-0.007
--0.006
ment-0.006
les-0.006
Token was
Token ID373
Feature activation+3.31e-02

Pos logit contributions
eros+0.030
ez+0.029
ment+0.029
 Start+0.028
crow+0.027
Neg logit contributions
 would-0.033
 topp-0.032
First-0.032
 ju-0.032
'd-0.032
Token so
Token ID523
Feature activation-1.64e-01

Pos logit contributions
eros+0.047
oc+0.047
ment+0.046
rows+0.045
les+0.044
Neg logit contributions
 ju-0.044
First-0.043
 flour-0.043
 ten-0.042
 topp-0.041
Token sad
Token ID6507
Feature activation+1.51e-01

Pos logit contributions
First+0.307
 topp+0.304
 ju+0.303
 flour+0.289
 flavor+0.287
Neg logit contributions
eros-0.157
ez-0.152
crow-0.137
 speaker-0.133
rows-0.133
Token like
Token ID588
Feature activation+2.44e-02

Pos logit contributions
 its+0.127
 knowing+0.124
 tears+0.123
-+0.121
ez+0.120
Neg logit contributions
First-0.168
 topp-0.166
 forks-0.152
 ju-0.151
 flour-0.149
Token all
Token ID477
Feature activation-8.75e-02

Pos logit contributions
 its+0.022
 tears+0.020
 knowing+0.019
 heaven+0.019
ez+0.019
Neg logit contributions
First-0.050
 topp-0.048
 ju-0.047
 forks-0.044
 stack-0.044
Token the
Token ID262
Feature activation+8.81e-03

Pos logit contributions
 topp+0.154
First+0.153
 ju+0.149
 flavor+0.141
 flour+0.140
Neg logit contributions
ez-0.099
eros-0.094
 wound-0.091
 tears-0.089
 heaven-0.089
Token angels
Token ID21981
Feature activation-8.95e-02

Pos logit contributions
 tears+0.012
ez+0.012
 speaker+0.012
 wound+0.012
 creature+0.012
Neg logit contributions
First-0.013
 topp-0.013
 ju-0.012
 flour-0.011
 forks-0.011
Token in
Token ID287
Feature activation-5.11e-02

Pos logit contributions
First+0.144
 topp+0.142
 ju+0.139
 flour+0.130
 flavor+0.127
Neg logit contributions
ez-0.117
eros-0.111
--0.107
 wound-0.107
 crying-0.105
Token heaven
Token ID9538
Feature activation+3.33e-01

Pos logit contributions
 topp+0.093
First+0.092
 ju+0.090
 flavor+0.087
 flour+0.085
Neg logit contributions
ez-0.054
eros-0.052
 wound-0.050
oring-0.048
 tears-0.048
Token were
Token ID547
Feature activation-9.68e-02

Pos logit contributions
ez+0.364
eros+0.347
 wound+0.336
 tears+0.326
 heaven+0.326
Neg logit contributions
 topp-0.595
First-0.588
 ju-0.578
 flavor-0.546
 flour-0.543
Token singing
Token ID13777
Feature activation+1.12e-01

Pos logit contributions
 topp+0.169
First+0.168
 ju+0.164
 flavor+0.154
 flour+0.153
Neg logit contributions
ez-0.111
eros-0.105
 wound-0.103
 tears-0.101
 heaven-0.100
Token.
Token ID13
Feature activation-4.03e-04

Pos logit contributions
ez+0.113
eros+0.107
 wound+0.103
 heaven+0.101
 tears+0.100
Neg logit contributions
 topp-0.209
First-0.207
 ju-0.203
 flavor-0.192
 flour-0.192
Token
Token ID198
Feature activation+1.64e-02

Pos logit contributions
First+0.001
 topp+0.001
 ju+0.001
 flavor+0.001
 forks+0.001
Neg logit contributions
ez-0.000
eros-0.000
 wound-0.000
--0.000
 terrified-0.000
Token
Token ID198
Feature activation+1.52e-02

Pos logit contributions
eros+0.019
ez+0.018
 wound+0.017
 heaven+0.017
 tears+0.016
Neg logit contributions
First-0.031
 topp-0.029
 ju-0.028
 flour-0.027
 flavor-0.027
Token now
Token ID783
Feature activation-7.23e-03

Pos logit contributions
 topp+0.023
 ju+0.023
First+0.022
 flavor+0.021
 flour+0.021
Neg logit contributions
--0.009
ez-0.009
 heaven-0.008
eros-0.008
 wound-0.008
Token.
Token ID13
Feature activation+8.56e-03

Pos logit contributions
 topp+0.013
First+0.012
 ju+0.012
 flavor+0.012
 flour+0.011
Neg logit contributions
ez-0.008
eros-0.008
 wound-0.008
 tears-0.007
 terrified-0.007
Token You
Token ID921
Feature activation+3.45e-02

Pos logit contributions
ez+0.011
les+0.010
rows+0.010
 wound+0.010
-+0.010
Neg logit contributions
First-0.014
 ju-0.013
 topp-0.013
 flour-0.012
 First-0.012
Token are
Token ID389
Feature activation-6.37e-02

Pos logit contributions
ez+0.061
eros+0.059
 wound+0.059
 heaven+0.058
 tears+0.058
Neg logit contributions
 topp-0.039
First-0.038
 ju-0.037
 flavor-0.033
 flour-0.033
Token in
Token ID287
Feature activation-5.14e-02

Pos logit contributions
 topp+0.104
 ju+0.102
First+0.099
 flour+0.097
 forks+0.096
Neg logit contributions
ez-0.079
eros-0.077
ment-0.075
oc-0.074
oring-0.074
Token heaven
Token ID9538
Feature activation+3.32e-01

Pos logit contributions
 topp+0.095
First+0.094
 ju+0.093
 flavor+0.088
 flour+0.088
Neg logit contributions
ez-0.053
eros-0.051
 wound-0.049
 tears-0.047
 heaven-0.047
Token.
Token ID13
Feature activation+7.08e-03

Pos logit contributions
ez+0.438
eros+0.417
 wound+0.396
 tears+0.395
 Start+0.392
Neg logit contributions
 topp-0.515
First-0.509
 ju-0.493
 flavor-0.476
 forks-0.470
Token You
Token ID921
Feature activation+3.42e-02

Pos logit contributions
les+0.009
 wound+0.009
rows+0.009
ez+0.009
ling+0.009
Neg logit contributions
First-0.011
 topp-0.011
 ju-0.011
 flour-0.010
 First-0.010
Token are
Token ID389
Feature activation-6.35e-02

Pos logit contributions
ez+0.063
eros+0.062
 wound+0.060
 tears+0.059
 heaven+0.059
Neg logit contributions
 topp-0.035
First-0.035
 ju-0.033
 flavor-0.030
 flour-0.030
Token an
Token ID281
Feature activation+4.86e-03

Pos logit contributions
 topp+0.099
 ju+0.098
First+0.095
 flour+0.093
 forks+0.092
Neg logit contributions
ez-0.083
eros-0.080
ment-0.079
oc-0.078
oring-0.077
Token angel
Token ID18304
Feature activation+1.08e-01

Pos logit contributions
ez+0.006
oring+0.006
ment+0.006
eros+0.006
 Start+0.006
Neg logit contributions
First-0.007
 ju-0.007
 flour-0.007
 topp-0.007
 flavor-0.007
Token knew
Token ID2993
Feature activation+1.12e-01

Pos logit contributions
ez+0.066
eros+0.066
crow+0.062
 heaven+0.061
 speaker+0.060
Neg logit contributions
 topp-0.089
First-0.087
 ju-0.085
 flavor-0.081
 forks-0.080
Token that
Token ID326
Feature activation+8.44e-02

Pos logit contributions
ez+0.126
eros+0.120
oring+0.116
oc+0.116
 terrified+0.115
Neg logit contributions
 topp-0.193
 ju-0.192
First-0.192
 flour-0.188
 flavor-0.186
Token Teddy
Token ID29345
Feature activation-5.84e-04

Pos logit contributions
ez+0.097
eros+0.095
 wound+0.089
oring+0.087
crow+0.087
Neg logit contributions
 topp-0.147
First-0.146
 ju-0.144
 flavor-0.136
 flour-0.135
Token was
Token ID373
Feature activation+3.30e-02

Pos logit contributions
 topp+0.001
First+0.001
 ju+0.001
 flavor+0.001
 flour+0.001
Neg logit contributions
ez-0.001
eros-0.001
 wound-0.001
 tears-0.001
 heaven-0.001
Token in
Token ID287
Feature activation-5.38e-02

Pos logit contributions
eros+0.051
oring+0.051
oc+0.050
ez+0.050
ment+0.048
Neg logit contributions
 ju-0.046
 topp-0.044
 flour-0.042
First-0.042
 flavor-0.040
Token heaven
Token ID9538
Feature activation+3.32e-01

Pos logit contributions
 topp+0.093
First+0.092
 ju+0.090
 flour+0.088
 flavor+0.087
Neg logit contributions
eros-0.061
oring-0.061
ez-0.061
rows-0.057
izes-0.056
Token and
Token ID290
Feature activation+3.89e-02

Pos logit contributions
ez+0.446
eros+0.431
 Start+0.404
 tears+0.399
crow+0.398
Neg logit contributions
 topp-0.493
First-0.491
 would-0.482
 ju-0.481
 flavor-0.469
Token that
Token ID326
Feature activation+8.69e-02

Pos logit contributions
ez+0.046
eros+0.046
oring+0.043
crow+0.042
 speaker+0.041
Neg logit contributions
 topp-0.066
 ju-0.064
First-0.064
 flavor-0.061
 forks-0.060
Token he
Token ID339
Feature activation+1.11e-02

Pos logit contributions
ez+0.092
eros+0.090
 wound+0.084
oring+0.082
crow+0.081
Neg logit contributions
 topp-0.159
First-0.157
 ju-0.155
 flavor-0.148
 flour-0.146
Token would
Token ID561
Feature activation-1.26e-01

Pos logit contributions
eros+0.019
ez+0.018
 speaker+0.018
ment+0.018
crow+0.018
Neg logit contributions
 topp-0.012
 would-0.012
 ju-0.012
First-0.012
'd-0.012
Token always
Token ID1464
Feature activation-6.49e-04

Pos logit contributions
 topp+0.186
First+0.182
 ju+0.178
 flavor+0.166
 flour+0.165
Neg logit contributions
ez-0.180
eros-0.173
 wound-0.171
 heaven-0.168
 tears-0.167
Token have
Token ID423
Feature activation+4.07e-02

Pos logit contributions
 topp+0.110
First+0.107
 ju+0.103
 would+0.100
 flavor+0.098
Neg logit contributions
eros-0.222
ez-0.210
ment-0.210
 speaker-0.204
 tears-0.203
Token to
Token ID284
Feature activation-4.36e-02

Pos logit contributions
ez+0.045
 tears+0.045
 heaven+0.043
 its+0.043
 wound+0.043
Neg logit contributions
First-0.071
 topp-0.071
 ju-0.069
 flavor-0.065
 forks-0.063
Token go
Token ID467
Feature activation-1.49e-01

Pos logit contributions
 topp+0.065
 ju+0.063
First+0.062
 forks+0.057
 flour+0.056
Neg logit contributions
 wound-0.063
 heaven-0.063
 tears-0.062
ez-0.061
 its-0.060
Token back
Token ID736
Feature activation-2.58e-02

Pos logit contributions
First+0.194
 would+0.181
 ju+0.175
 stack+0.175
 usual+0.173
Neg logit contributions
ez-0.253
eros-0.230
ment-0.230
 tears-0.222
oc-0.216
Token to
Token ID284
Feature activation-4.33e-02

Pos logit contributions
First+0.041
 topp+0.040
 ju+0.039
 flavor+0.037
 flour+0.037
Neg logit contributions
ez-0.034
eros-0.032
 wound-0.031
 tears-0.031
 terrified-0.030
Token heaven
Token ID9538
Feature activation+3.32e-01

Pos logit contributions
 topp+0.072
First+0.072
 ju+0.070
 forks+0.069
 flavor+0.069
Neg logit contributions
ez-0.052
eros-0.048
oring-0.046
ment-0.046
 terrified-0.045
Token,"
Token ID553
Feature activation-3.59e-02

Pos logit contributions
ez+0.440
eros+0.418
 wound+0.398
 tears+0.397
 terrified+0.391
Neg logit contributions
First-0.520
 topp-0.518
 ju-0.502
 flavor-0.479
 flour-0.471
Token said
Token ID531
Feature activation+4.27e-02

Pos logit contributions
 topp+0.053
 flavor+0.052
 ju+0.051
First+0.050
 flour+0.049
Neg logit contributions
ez-0.049
oring-0.049
rows-0.048
eros-0.047
 terrified-0.046
Token the
Token ID262
Feature activation+9.97e-03

Pos logit contributions
ez+0.048
eros+0.045
 wound+0.044
oring+0.043
 terrified+0.043
Neg logit contributions
 topp-0.074
 ju-0.073
First-0.072
 flavor-0.070
 flour-0.069
Token angel
Token ID18304
Feature activation+1.05e-01

Pos logit contributions
ez+0.009
eros+0.008
 wound+0.008
oring+0.008
 heaven+0.007
Neg logit contributions
 topp-0.020
First-0.020
 ju-0.020
 flavor-0.019
 flour-0.019
Token.
Token ID13
Feature activation+1.70e-02

Pos logit contributions
ez+0.122
eros+0.119
 wound+0.113
 tears+0.111
 heaven+0.110
Neg logit contributions
 topp-0.177
 ju-0.172
First-0.171
 flavor-0.163
 forks-0.159
Token had
Token ID550
Feature activation-5.78e-02

Pos logit contributions
ez+0.151
eros+0.149
ment+0.141
 heaven+0.140
crow+0.138
Neg logit contributions
 topp-0.143
 ju-0.141
 would-0.135
First-0.135
 flavor-0.133
Token to
Token ID284
Feature activation-4.42e-02

Pos logit contributions
 topp+0.088
First+0.087
 ju+0.085
 flavor+0.079
 flour+0.079
Neg logit contributions
ez-0.079
eros-0.074
 wound-0.074
 tears-0.074
 heaven-0.072
Token go
Token ID467
Feature activation-1.46e-01

Pos logit contributions
 topp+0.053
First+0.051
 ju+0.049
 flour+0.044
 flavor+0.044
Neg logit contributions
ez-0.076
 wound-0.074
eros-0.074
 heaven-0.074
 tears-0.073
Token back
Token ID736
Feature activation-2.54e-02

Pos logit contributions
First+0.193
 picnic+0.192
 would+0.191
 usual+0.189
 stack+0.184
Neg logit contributions
ez-0.234
oc-0.212
eros-0.205
ment-0.201
 aids-0.194
Token to
Token ID284
Feature activation-4.41e-02

Pos logit contributions
First+0.038
 topp+0.037
 ju+0.036
 flavor+0.035
 usual+0.035
Neg logit contributions
ez-0.035
eros-0.033
crow-0.031
 terrified-0.031
 wound-0.031
Token heaven
Token ID9538
Feature activation+3.31e-01

Pos logit contributions
 usual+0.064
 ten+0.063
 topp+0.063
 picnic+0.062
 stack+0.062
Neg logit contributions
eros-0.060
ez-0.060
rows-0.057
crow-0.056
oring-0.056
Token again
Token ID757
Feature activation+5.80e-02

Pos logit contributions
ez+0.529
eros+0.506
crow+0.480
lves+0.464
 Start+0.459
Neg logit contributions
 would-0.420
First-0.398
 topp-0.397
 flavor-0.392
'd-0.392
Token.
Token ID13
Feature activation-1.05e-02

Pos logit contributions
ez+0.073
eros+0.072
crow+0.066
 tears+0.065
 wound+0.065
Neg logit contributions
First-0.093
 topp-0.093
 ju-0.091
 flavor-0.088
 flour-0.085
Token Lily
Token ID20037
Feature activation+4.02e-02

Pos logit contributions
First+0.015
 ju+0.015
 First+0.013
 flour+0.013
 topp+0.012
Neg logit contributions
eros-0.017
ez-0.017
an-0.016
ging-0.016
ling-0.016
Token waved
Token ID26834
Feature activation+4.22e-02

Pos logit contributions
ez+0.051
eros+0.051
ment+0.049
 heaven+0.048
crow+0.048
Neg logit contributions
 ju-0.060
 would-0.060
First-0.060
 topp-0.059
'd-0.058
Token goodbye
Token ID24829
Feature activation+8.69e-02

Pos logit contributions
oc+0.074
ez+0.073
eros+0.070
web+0.070
coon+0.067
Neg logit contributions
 yes-0.053
 ju-0.047
 would-0.047
 ten-0.045
 picnic-0.044
Token was
Token ID373
Feature activation+3.93e-02

Pos logit contributions
ez+0.122
 wound+0.120
-+0.119
eros+0.117
 tears+0.117
Neg logit contributions
First-0.053
 topp-0.052
 ju-0.048
 flour-0.044
 forks-0.041
Token dancing
Token ID15360
Feature activation+3.07e-02

Pos logit contributions
ez+0.054
eros+0.052
oc+0.052
crow+0.049
oring+0.049
Neg logit contributions
 ju-0.060
 topp-0.058
First-0.055
 flour-0.055
 forks-0.054
Token and
Token ID290
Feature activation+5.90e-02

Pos logit contributions
ez+0.033
eros+0.032
 wound+0.030
 tears+0.029
 terrified+0.029
Neg logit contributions
 topp-0.055
First-0.054
 ju-0.054
 flavor-0.051
 forks-0.050
Token drawing
Token ID8263
Feature activation+1.15e-01

Pos logit contributions
ez+0.070
eros+0.068
 wound+0.064
crow+0.063
oring+0.063
Neg logit contributions
 topp-0.100
First-0.098
 ju-0.097
 flavor-0.092
 forks-0.091
Token in
Token ID287
Feature activation-5.70e-02

Pos logit contributions
ez+0.125
eros+0.120
 wound+0.116
 tears+0.112
 heaven+0.112
Neg logit contributions
 topp-0.207
First-0.204
 ju-0.200
 flavor-0.190
 flour-0.188
Token heaven
Token ID9538
Feature activation+3.29e-01

Pos logit contributions
 topp+0.105
First+0.102
 ju+0.100
 flavor+0.098
 flour+0.097
Neg logit contributions
ez-0.061
eros-0.060
oring-0.056
 wound-0.055
crow-0.054
Token.
Token ID13
Feature activation+1.39e-02

Pos logit contributions
ez+0.396
eros+0.379
 wound+0.363
 tears+0.351
 terrified+0.347
Neg logit contributions
 topp-0.554
First-0.545
 ju-0.536
 flavor-0.511
 forks-0.503
Token<|endoftext|>
Token ID50256
Feature activation+3.76e-02

Pos logit contributions
ez+0.020
eros+0.020
-+0.019
 wound+0.018
rows+0.018
Neg logit contributions
First-0.023
 ju-0.021
 First-0.019
 topp-0.019
 flour-0.018
TokenOnce
Token ID7454
Feature activation-1.24e-01

Pos logit contributions
ez+0.031
eros+0.030
 wound+0.028
 tears+0.027
 heaven+0.027
Neg logit contributions
 topp-0.077
First-0.076
 ju-0.075
 flavor-0.071
 flour-0.071
Token upon
Token ID2402
Feature activation+6.08e-02

Pos logit contributions
First+0.151
 topp+0.145
 ju+0.141
 forks+0.139
 usual+0.139
Neg logit contributions
ez-0.208
eros-0.194
 wound-0.190
 speaker-0.188
 heaven-0.188
Token a
Token ID257
Feature activation-7.68e-02

Pos logit contributions
eros+0.085
ez+0.084
crow+0.080
oring+0.079
 wound+0.078
Neg logit contributions
First-0.091
 topp-0.091
 ju-0.088
 flavor-0.085
 flour-0.083
Token He
Token ID679
Feature activation-3.07e-02

Pos logit contributions
ez+0.009
eros+0.009
 wound+0.009
-+0.009
 tears+0.009
Neg logit contributions
First-0.015
 topp-0.015
 ju-0.015
 forks-0.014
 flavor-0.014
Token is
Token ID318
Feature activation+7.36e-02

Pos logit contributions
 topp+0.045
 ju+0.043
First+0.042
 would+0.041
 flavor+0.041
Neg logit contributions
ez-0.042
eros-0.041
ment-0.039
oring-0.039
crow-0.039
Token with
Token ID351
Feature activation+1.29e-01

Pos logit contributions
eros+0.095
ez+0.095
oring+0.092
oc+0.092
crow+0.089
Neg logit contributions
 topp-0.113
 ju-0.112
First-0.107
 flour-0.105
 flavor-0.103
Token God
Token ID1793
Feature activation+1.11e-01

Pos logit contributions
ez+0.133
eros+0.124
 tears+0.122
 heaven+0.121
 wound+0.119
Neg logit contributions
First-0.240
 topp-0.239
 ju-0.234
 flavor-0.220
 flour-0.218
Token in
Token ID287
Feature activation-5.51e-02

Pos logit contributions
ez+0.100
eros+0.097
-+0.093
 heaven+0.092
 wound+0.087
Neg logit contributions
First-0.228
 ju-0.222
 topp-0.221
 flour-0.206
 flavor-0.205
Token heaven
Token ID9538
Feature activation+3.29e-01

Pos logit contributions
 topp+0.101
First+0.098
 ju+0.095
 flavor+0.094
 flour+0.092
Neg logit contributions
eros-0.057
ez-0.057
oring-0.055
 wound-0.054
rows-0.052
Token.
Token ID13
Feature activation+6.16e-03

Pos logit contributions
ez+0.431
eros+0.407
 wound+0.391
 Start+0.387
 tears+0.386
Neg logit contributions
 topp-0.514
First-0.492
 ju-0.482
 flavor-0.470
 forks-0.468
Token He
Token ID679
Feature activation-3.22e-02

Pos logit contributions
 wound+0.007
ez+0.007
eros+0.007
-+0.007
 tears+0.007
Neg logit contributions
First-0.011
 topp-0.010
 ju-0.010
 flavor-0.009
 forks-0.009
Token still
Token ID991
Feature activation+2.31e-02

Pos logit contributions
 topp+0.045
 would+0.044
'd+0.042
 ju+0.041
 flavor+0.040
Neg logit contributions
eros-0.045
ez-0.045
ment-0.044
oring-0.044
crow-0.043
Token loves
Token ID10408
Feature activation+9.23e-02

Pos logit contributions
eros+0.035
ez+0.032
crow+0.032
oring+0.032
rows+0.031
Neg logit contributions
 topp-0.031
 would-0.030
'd-0.029
 flavor-0.029
 ju-0.029
Token us
Token ID514
Feature activation+1.13e-02

Pos logit contributions
ez+0.089
eros+0.086
 heaven+0.084
 tears+0.081
 wound+0.081
Neg logit contributions
First-0.177
 topp-0.176
 ju-0.175
 flavor-0.162
 flour-0.161
Token had
Token ID550
Feature activation-5.67e-02

Pos logit contributions
eros+0.012
ez+0.012
 heaven+0.012
ment+0.012
 speaker+0.012
Neg logit contributions
 topp-0.016
 ju-0.016
First-0.016
 would-0.016
'd-0.015
Token to
Token ID284
Feature activation-4.42e-02

Pos logit contributions
 topp+0.092
First+0.092
 ju+0.089
 forks+0.084
 flavor+0.084
Neg logit contributions
ez-0.072
 wound-0.067
 tears-0.067
eros-0.067
 heaven-0.065
Token go
Token ID467
Feature activation-1.47e-01

Pos logit contributions
 topp+0.063
First+0.060
 ju+0.059
 forks+0.055
 flour+0.052
Neg logit contributions
 wound-0.071
 heaven-0.071
 hospital-0.069
 tears-0.068
 its-0.068
Token to
Token ID284
Feature activation-4.44e-02

Pos logit contributions
First+0.208
 ju+0.204
 stack+0.196
 shovel+0.191
 would+0.191
Neg logit contributions
ez-0.219
eros-0.195
ment-0.186
oma-0.185
ller-0.185
Token squirrel
Token ID33039
Feature activation-7.13e-03

Pos logit contributions
 usual+0.067
 picnic+0.065
 flavor+0.065
First+0.064
 topp+0.064
Neg logit contributions
ez-0.060
les-0.057
oring-0.056
eros-0.055
rows-0.055
Token heaven
Token ID9538
Feature activation+3.28e-01

Pos logit contributions
First+0.012
 topp+0.012
 ju+0.012
 flavor+0.011
 flour+0.011
Neg logit contributions
ez-0.008
eros-0.008
 tears-0.007
 speaker-0.007
crow-0.007
Token.
Token ID13
Feature activation-7.77e-03

Pos logit contributions
ez+0.416
eros+0.393
 Start+0.370
 tears+0.364
Start+0.362
Neg logit contributions
 would-0.508
First-0.495
 topp-0.491
 ju-0.486
 usual-0.484
Token His
Token ID2399
Feature activation+1.94e-02

Pos logit contributions
 ju+0.011
First+0.010
 First+0.010
're+0.009
 topp+0.009
Neg logit contributions
web-0.013
ez-0.012
izes-0.012
les-0.012
 speaker-0.011
Token friends
Token ID2460
Feature activation-7.91e-02

Pos logit contributions
eros+0.022
 wound+0.021
ez+0.021
 tears+0.021
-+0.020
Neg logit contributions
First-0.036
 topp-0.035
 ju-0.034
 flour-0.032
 forks-0.032
Token were
Token ID547
Feature activation-1.04e-01

Pos logit contributions
First+0.074
 topp+0.070
 would+0.069
 ju+0.068
'd+0.067
Neg logit contributions
eros-0.149
ez-0.146
ment-0.145
 heaven-0.140
 wound-0.139
Token very
Token ID845
Feature activation-4.12e-02

Pos logit contributions
 ju+0.132
 flour+0.126
First+0.126
 would+0.126
 separate+0.124
Neg logit contributions
ment-0.158
oc-0.146
eros-0.142
rows-0.142
les-0.142
Token it
Token ID340
Feature activation+1.04e-01

Pos logit contributions
ez+0.079
eros+0.075
 wound+0.070
oring+0.069
 terrified+0.069
Neg logit contributions
 topp-0.138
First-0.136
 ju-0.134
 flavor-0.129
 flour-0.128
Token was
Token ID373
Feature activation+3.13e-02

Pos logit contributions
ez+0.200
eros+0.195
 wound+0.190
 heaven+0.187
 tears+0.187
Neg logit contributions
 topp-0.101
First-0.098
 ju-0.095
 flavor-0.086
 flour-0.084
Token happy
Token ID3772
Feature activation+3.62e-02

Pos logit contributions
ez+0.047
eros+0.046
 wound+0.045
 tears+0.044
 heaven+0.044
Neg logit contributions
 topp-0.043
First-0.043
 ju-0.042
 flavor-0.039
 flour-0.038
Token in
Token ID287
Feature activation-5.55e-02

Pos logit contributions
ez+0.028
 wound+0.027
 its+0.027
 crying+0.027
 tears+0.027
Neg logit contributions
First-0.077
 topp-0.076
 ju-0.074
 flour-0.069
 flavor-0.069
Token bug
Token ID5434
Feature activation+4.88e-02

Pos logit contributions
 topp+0.090
 flour+0.085
 forks+0.084
First+0.083
 ju+0.083
Neg logit contributions
oring-0.072
ez-0.070
eros-0.068
les-0.065
crow-0.064
Token heaven
Token ID9538
Feature activation+3.28e-01

Pos logit contributions
ez+0.047
 wound+0.043
 heaven+0.043
eros+0.042
-+0.042
Neg logit contributions
First-0.097
 topp-0.096
 ju-0.093
 flavor-0.089
 flour-0.088
Token.
Token ID13
Feature activation+1.00e-02

Pos logit contributions
ez+0.405
eros+0.385
 wound+0.360
 Start+0.358
crow+0.355
Neg logit contributions
 topp-0.540
First-0.529
 ju-0.523
 flavor-0.498
 forks-0.495
Token<|endoftext|>
Token ID50256
Feature activation+4.11e-02

Pos logit contributions
ez+0.015
eros+0.015
les+0.014
ling+0.013
-+0.013
Neg logit contributions
First-0.015
 ju-0.014
 topp-0.014
 flour-0.013
 flavor-0.013
TokenOne
Token ID3198
Feature activation-8.50e-02

Pos logit contributions
ez+0.024
eros+0.022
crow+0.021
-+0.020
 wound+0.019
Neg logit contributions
 topp-0.096
 ju-0.095
First-0.094
 flavor-0.088
 forks-0.088
Token day
Token ID1110
Feature activation+1.37e-02

Pos logit contributions
 topp+0.136
First+0.135
 ju+0.134
 forks+0.122
nuts+0.120
Neg logit contributions
--0.115
 its-0.113
ez-0.111
 heaven-0.107
 wound-0.107
Token,
Token ID11
Feature activation-3.12e-02

Pos logit contributions
-+0.011
 its+0.009
 crying+0.009
ment+0.008
eros+0.008
Neg logit contributions
 ju-0.032
 topp-0.030
First-0.030
 redd-0.028
 stack-0.027
Token knew
Token ID2993
Feature activation+1.11e-01

Pos logit contributions
 topp+0.075
First+0.072
 ju+0.071
 flour+0.066
 flavor+0.066
Neg logit contributions
ez-0.068
 wound-0.064
 tears-0.063
 heaven-0.062
eros-0.062
Token she
Token ID673
Feature activation+4.75e-02

Pos logit contributions
ez+0.144
eros+0.137
oring+0.132
oc+0.131
 terrified+0.129
Neg logit contributions
 topp-0.178
First-0.175
 ju-0.174
 flour-0.170
 flavor-0.169
Token was
Token ID373
Feature activation+2.26e-02

Pos logit contributions
ez+0.077
eros+0.075
 wound+0.073
 tears+0.071
 heaven+0.071
Neg logit contributions
 topp-0.061
First-0.059
 ju-0.058
 flavor-0.054
 flour-0.053
Token now
Token ID783
Feature activation-5.20e-03

Pos logit contributions
oc+0.038
eros+0.036
ez+0.035
ment+0.035
oring+0.033
Neg logit contributions
 ju-0.029
 topp-0.027
First-0.027
 usual-0.027
 flour-0.026
Token in
Token ID287
Feature activation-5.12e-02

Pos logit contributions
 topp+0.008
First+0.008
 ju+0.008
 flavor+0.008
 flour+0.008
Neg logit contributions
ez-0.007
eros-0.006
 wound-0.006
crow-0.006
oring-0.006
Token heaven
Token ID9538
Feature activation+3.28e-01

Pos logit contributions
 topp+0.083
First+0.082
 ju+0.080
 flavor+0.075
 flour+0.075
Neg logit contributions
ez-0.065
eros-0.063
 wound-0.061
 tears-0.060
 heaven-0.060
Token,
Token ID11
Feature activation-3.21e-02

Pos logit contributions
ez+0.404
eros+0.388
 wound+0.367
 tears+0.360
crow+0.358
Neg logit contributions
 topp-0.536
First-0.531
 ju-0.520
 flavor-0.498
 forks-0.488
Token dancing
Token ID15360
Feature activation+3.29e-02

Pos logit contributions
 ju+0.053
 topp+0.052
First+0.051
 flavor+0.049
 flour+0.049
Neg logit contributions
eros-0.041
ez-0.040
crow-0.036
oring-0.036
oc-0.036
Token with
Token ID351
Feature activation+1.32e-01

Pos logit contributions
ez+0.034
eros+0.033
 wound+0.031
 tears+0.030
 terrified+0.030
Neg logit contributions
 topp-0.060
First-0.060
 ju-0.059
 flavor-0.056
 flour-0.055
Token the
Token ID262
Feature activation+6.06e-03

Pos logit contributions
ez+0.129
eros+0.122
 wound+0.118
 tears+0.114
 heaven+0.114
Neg logit contributions
 topp-0.251
First-0.248
 ju-0.244
 flavor-0.232
 flour-0.230
Token angels
Token ID21981
Feature activation-8.81e-02

Pos logit contributions
ez+0.006
 wound+0.006
 tears+0.006
eros+0.006
 crying+0.005
Neg logit contributions
 topp-0.012
First-0.012
 ju-0.011
 flavor-0.010
 forks-0.010
Token thanked
Token ID26280
Feature activation-2.90e-02

Pos logit contributions
ez+0.018
eros+0.018
 heaven+0.017
 wound+0.016
crow+0.016
Neg logit contributions
 topp-0.022
First-0.022
 ju-0.022
 flavor-0.020
 forks-0.020
Token the
Token ID262
Feature activation+5.17e-03

Pos logit contributions
 topp+0.044
 ju+0.043
 flavor+0.042
 flour+0.041
First+0.041
Neg logit contributions
ez-0.039
oring-0.036
 terrified-0.035
 wound-0.035
rows-0.034
Token big
Token ID1263
Feature activation-1.03e-01

Pos logit contributions
 creature+0.005
 wound+0.004
 tears+0.004
 voice+0.004
 crying+0.004
Neg logit contributions
First-0.011
 topp-0.011
avored-0.010
 ju-0.010
 flavor-0.010
Token hand
Token ID1021
Feature activation-1.41e-02

Pos logit contributions
First+0.215
 topp+0.205
 ju+0.198
 flour+0.187
avored+0.187
Neg logit contributions
eros-0.096
 creature-0.095
--0.092
 wound-0.092
 tears-0.091
Token in
Token ID287
Feature activation-5.97e-02

Pos logit contributions
 topp+0.028
First+0.028
 ju+0.027
 flavor+0.026
 flour+0.026
Neg logit contributions
ez-0.013
eros-0.012
 wound-0.011
 tears-0.011
 heaven-0.011
Token heaven
Token ID9538
Feature activation+3.27e-01

Pos logit contributions
 topp+0.125
First+0.122
 ju+0.121
 flavor+0.120
 flour+0.116
Neg logit contributions
oring-0.042
eros-0.042
rows-0.040
ez-0.040
 wound-0.037
Token for
Token ID329
Feature activation+8.20e-02

Pos logit contributions
ez+0.421
eros+0.393
crow+0.372
 wound+0.372
 heaven+0.361
Neg logit contributions
 topp-0.522
 ju-0.510
First-0.506
 flavor-0.484
 forks-0.476
Token sending
Token ID7216
Feature activation+9.26e-02

Pos logit contributions
eros+0.096
oring+0.095
ez+0.092
crow+0.090
les+0.089
Neg logit contributions
 topp-0.137
First-0.132
 ju-0.128
 flavor-0.127
 flour-0.127
Token it
Token ID340
Feature activation+1.00e-01

Pos logit contributions
ez+0.097
eros+0.095
 wound+0.090
oring+0.089
 terrified+0.087
Neg logit contributions
 topp-0.170
First-0.164
 ju-0.164
 flour-0.157
 flavor-0.156
Token such
Token ID884
Feature activation+5.43e-02

Pos logit contributions
ez+0.113
eros+0.112
 wound+0.103
crow+0.100
oring+0.099
Neg logit contributions
 topp-0.173
First-0.172
 ju-0.170
 flavor-0.161
 flour-0.159
Token a
Token ID257
Feature activation-7.37e-02

Pos logit contributions
ez+0.063
eros+0.059
 wound+0.059
 tears+0.058
 heaven+0.058
Neg logit contributions
 topp-0.093
First-0.093
 ju-0.090
 forks-0.085
 flavor-0.085
Token it
Token ID340
Feature activation+1.01e-01

Pos logit contributions
 its+0.003
 tears+0.003
 knowing+0.002
 crying+0.002
ez+0.002
Neg logit contributions
First-0.021
 ju-0.020
 topp-0.019
avored-0.019
 First-0.019
Token.
Token ID13
Feature activation-7.95e-03

Pos logit contributions
-+0.074
 its+0.071
 knowing+0.069
ez+0.065
 crying+0.063
Neg logit contributions
First-0.232
 topp-0.223
 ju-0.213
 flour-0.209
 apr-0.207
Token It
Token ID632
Feature activation+2.60e-02

Pos logit contributions
First+0.015
 topp+0.015
 ju+0.015
 flavor+0.014
 flour+0.014
Neg logit contributions
ez-0.008
eros-0.007
 speaker-0.007
 wound-0.007
 heaven-0.007
Token tasted
Token ID29187
Feature activation-5.55e-02

Pos logit contributions
ez+0.025
 wound+0.023
 tears+0.022
eros+0.022
 terrified+0.021
Neg logit contributions
First-0.051
 topp-0.050
 ju-0.049
 flour-0.047
 forks-0.046
Token like
Token ID588
Feature activation+1.92e-02

Pos logit contributions
First+0.097
 topp+0.090
 stack+0.086
 forks+0.085
 shovel+0.084
Neg logit contributions
 its-0.069
 tears-0.063
--0.062
ez-0.060
 knowing-0.058
Token heaven
Token ID9538
Feature activation+3.26e-01

Pos logit contributions
 its+0.018
 tears+0.015
 his+0.015
 knowing+0.015
 someone+0.014
Neg logit contributions
First-0.042
Deal-0.036
 topp-0.036
 ju-0.036
teen-0.036
Token!
Token ID0
Feature activation-2.60e-02

Pos logit contributions
ez+0.373
eros+0.355
 wound+0.344
 tears+0.336
 heaven+0.331
Neg logit contributions
 topp-0.569
First-0.556
 ju-0.551
 flavor-0.526
 flour-0.514
Token
Token ID198
Feature activation+1.51e-02

Pos logit contributions
First+0.045
 topp+0.043
 ju+0.043
 flavor+0.041
 flour+0.039
Neg logit contributions
ez-0.029
 speaker-0.029
--0.028
 wound-0.028
eros-0.028
Token
Token ID198
Feature activation+1.44e-02

Pos logit contributions
ez+0.015
 heaven+0.015
eros+0.015
 speaker+0.015
 wound+0.014
Neg logit contributions
First-0.028
 topp-0.026
 ju-0.025
 flavor-0.025
 flour-0.025
TokenR
Token ID49
Feature activation+2.39e-02

Pos logit contributions
ez+0.019
eros+0.019
 heaven+0.018
 tears+0.018
 wound+0.018
Neg logit contributions
First-0.023
 topp-0.022
 ju-0.022
 flavor-0.020
 flour-0.020
Tokenalph
Token ID17307
Feature activation-9.36e-02

Pos logit contributions
ez+0.038
eros+0.037
-+0.036
ment+0.035
 wound+0.035
Neg logit contributions
 ju-0.032
 topp-0.032
First-0.032
 flour-0.029
 forks-0.028
Token said
Token ID531
Feature activation+2.87e-02

Pos logit contributions
eros+0.078
ez+0.078
 wound+0.073
crow+0.073
 speaker+0.073
Neg logit contributions
 topp-0.092
 ju-0.089
First-0.088
 flavor-0.085
 flour-0.083
Token he
Token ID339
Feature activation+3.16e-03

Pos logit contributions
ez+0.022
eros+0.020
 wound+0.020
 its+0.019
 tears+0.019
Neg logit contributions
First-0.062
 topp-0.061
 ju-0.059
 flavor-0.056
 forks-0.055
Token had
Token ID550
Feature activation-5.68e-02

Pos logit contributions
ez+0.004
 wound+0.004
eros+0.004
-+0.004
 tears+0.004
Neg logit contributions
First-0.005
 topp-0.005
 ju-0.005
 flour-0.004
 flavor-0.004
Token a
Token ID257
Feature activation-7.21e-02

Pos logit contributions
First+0.099
 ju+0.092
 topp+0.091
teen+0.087
avored+0.087
Neg logit contributions
ez-0.068
 tears-0.066
 wound-0.065
 its-0.063
 heaven-0.063
Token rare
Token ID4071
Feature activation+6.09e-02

Pos logit contributions
First+0.115
 topp+0.110
 ju+0.105
 forks+0.099
teen+0.098
Neg logit contributions
 wound-0.096
 tears-0.094
 crying-0.093
 heaven-0.092
ez-0.091
Token disease
Token ID4369
Feature activation+3.14e-01

Pos logit contributions
ez+0.097
eros+0.094
 wound+0.091
 tears+0.089
 heaven+0.089
Neg logit contributions
 topp-0.079
First-0.077
 ju-0.076
 flavor-0.071
 flour-0.070
Token that
Token ID326
Feature activation+8.26e-02

Pos logit contributions
ez+0.311
eros+0.299
 wound+0.285
 tears+0.276
 heaven+0.275
Neg logit contributions
 topp-0.593
First-0.583
 ju-0.573
 flavor-0.549
 flour-0.543
Token made
Token ID925
Feature activation+3.16e-03

Pos logit contributions
ez+0.091
 wound+0.086
 tears+0.086
 crying+0.084
 its+0.084
Neg logit contributions
First-0.147
 topp-0.144
 ju-0.139
 forks-0.133
 flour-0.131
Token him
Token ID683
Feature activation-1.33e-02

Pos logit contributions
ez+0.003
eros+0.003
 wound+0.002
 tears+0.002
 its+0.002
Neg logit contributions
 topp-0.006
First-0.006
 ju-0.006
 flavor-0.006
 forks-0.006
Token allergic
Token ID31856
Feature activation+6.83e-02

Pos logit contributions
 topp+0.023
First+0.023
 ju+0.023
 flavor+0.021
 flour+0.021
Neg logit contributions
ez-0.015
eros-0.014
 wound-0.014
 tears-0.014
 heaven-0.014
Token to
Token ID284
Feature activation-4.26e-02

Pos logit contributions
ez+0.080
eros+0.076
 wound+0.074
-+0.073
 tears+0.073
Neg logit contributions
 topp-0.118
First-0.118
 ju-0.117
 flavor-0.107
 flour-0.107
Token cure
Token ID13996
Feature activation+1.93e-01

Pos logit contributions
First+0.106
 topp+0.104
 ju+0.098
 forks+0.094
teen+0.093
Neg logit contributions
 wound-0.090
 crying-0.085
--0.084
 tears-0.083
 speaker-0.083
Token for
Token ID329
Feature activation+8.46e-02

Pos logit contributions
ez+0.222
eros+0.213
 wound+0.205
 tears+0.201
 terrified+0.198
Neg logit contributions
 topp-0.336
First-0.331
 ju-0.324
 flavor-0.309
 flour-0.306
Token a
Token ID257
Feature activation-6.74e-02

Pos logit contributions
ez+0.090
eros+0.087
 wound+0.086
 tears+0.085
 heaven+0.085
Neg logit contributions
First-0.155
 topp-0.153
 ju-0.152
 flavor-0.140
 flour-0.137
Token very
Token ID845
Feature activation-3.55e-02

Pos logit contributions
First+0.108
 topp+0.102
 ju+0.097
teen+0.093
avored+0.093
Neg logit contributions
 wound-0.089
 crying-0.088
 speaker-0.086
--0.086
 heaven-0.086
Token bad
Token ID2089
Feature activation+9.59e-02

Pos logit contributions
First+0.054
 topp+0.054
 ju+0.052
 forks+0.048
 flavor+0.048
Neg logit contributions
 wound-0.048
--0.046
ez-0.045
 tears-0.045
eros-0.045
Token disease
Token ID4369
Feature activation+3.13e-01

Pos logit contributions
 wound+0.103
ez+0.103
eros+0.101
-+0.100
 tears+0.098
Neg logit contributions
First-0.176
 topp-0.171
 ju-0.169
 forks-0.155
 flour-0.155
Token.
Token ID13
Feature activation+1.46e-02

Pos logit contributions
ez+0.361
eros+0.346
 wound+0.335
 tears+0.326
 heaven+0.325
Neg logit contributions
 topp-0.542
First-0.535
 ju-0.525
 flavor-0.496
 flour-0.492
Token You
Token ID921
Feature activation+3.46e-02

Pos logit contributions
ez+0.016
eros+0.015
 wound+0.015
 tears+0.015
-+0.014
Neg logit contributions
 topp-0.026
First-0.026
 ju-0.025
 flavor-0.024
 flour-0.024
Token will
Token ID481
Feature activation-7.43e-02

Pos logit contributions
ez+0.063
eros+0.061
 wound+0.061
 heaven+0.060
 tears+0.060
Neg logit contributions
 topp-0.037
First-0.036
 ju-0.035
 flavor-0.032
 flour-0.031
Token be
Token ID307
Feature activation+3.88e-03

Pos logit contributions
 topp+0.108
First+0.107
 ju+0.104
 flavor+0.097
 flour+0.096
Neg logit contributions
ez-0.107
eros-0.104
 wound-0.100
 tears-0.098
 heaven-0.098
Token heroes
Token ID10281
Feature activation-2.11e-02

Pos logit contributions
eros+0.006
oring+0.006
ez+0.006
oc+0.006
ment+0.006
Neg logit contributions
 ju-0.005
 flour-0.005
 topp-0.005
First-0.005
 ten-0.005
Token at
Token ID379
Feature activation+4.88e-03

Pos logit contributions
 ju+0.035
 topp+0.033
 forks+0.033
 flour+0.032
First+0.032
Neg logit contributions
ez-0.032
eros-0.032
 speaker-0.030
ment-0.030
crow-0.030
Token the
Token ID262
Feature activation+1.42e-02

Pos logit contributions
ez+0.006
eros+0.005
oring+0.005
 wound+0.005
oc+0.005
Neg logit contributions
 topp-0.008
First-0.008
 ju-0.008
 forks-0.008
 flour-0.008
Token sign
Token ID1051
Feature activation+1.34e-01

Pos logit contributions
 wound+0.019
ez+0.019
 tears+0.018
eros+0.018
 crying+0.018
Neg logit contributions
First-0.023
 topp-0.021
 ju-0.021
 flavor-0.019
 forks-0.019
Token and
Token ID290
Feature activation+5.33e-02

Pos logit contributions
ez+0.089
eros+0.082
 wound+0.082
 tears+0.081
-+0.078
Neg logit contributions
First-0.296
 topp-0.296
 ju-0.290
 flavor-0.275
 forks-0.273
Token the
Token ID262
Feature activation+1.66e-02

Pos logit contributions
ez+0.081
eros+0.079
oring+0.077
crow+0.073
oc+0.071
Neg logit contributions
 topp-0.077
 ju-0.074
First-0.073
 forks-0.071
 flavor-0.071
Token speaker
Token ID10834
Feature activation+3.06e-01

Pos logit contributions
ez+0.020
eros+0.019
 wound+0.018
 tears+0.018
 heaven+0.018
Neg logit contributions
 topp-0.028
First-0.028
 ju-0.027
 flavor-0.026
 flour-0.025
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
ez+0.329
eros+0.312
 wound+0.311
 tears+0.308
-+0.300
Neg logit contributions
First-0.551
 topp-0.549
 ju-0.539
 flavor-0.501
 forks-0.500
Token She
Token ID1375
Feature activation+1.56e-02

Pos logit contributions
ez+0.010
eros+0.009
 wound+0.009
 heaven+0.009
 tears+0.009
Neg logit contributions
 topp-0.025
First-0.024
 ju-0.024
 flavor-0.023
 forks-0.022
Token felt
Token ID2936
Feature activation+6.75e-02

Pos logit contributions
ez+0.023
eros+0.023
oring+0.022
 heaven+0.022
crow+0.022
Neg logit contributions
 topp-0.021
First-0.020
 ju-0.020
 would-0.020
 flavor-0.020
Token silly
Token ID14397
Feature activation-2.70e-02

Pos logit contributions
eros+0.087
oring+0.087
ez+0.082
oc+0.082
rows+0.081
Neg logit contributions
 ju-0.110
 flour-0.104
 topp-0.100
 flavor-0.099
First-0.098
Token and
Token ID290
Feature activation+4.94e-02

Pos logit contributions
 topp+0.050
First+0.046
 forks+0.046
 ju+0.046
 flavor+0.045
Neg logit contributions
eros-0.029
ez-0.028
crow-0.026
 devastated-0.025
Start-0.025
Token hear
Token ID3285
Feature activation+1.44e-01

Pos logit contributions
 topp+0.188
 ju+0.182
First+0.181
 flour+0.165
 forks+0.164
Neg logit contributions
ez-0.189
eros-0.182
 wound-0.181
 heaven-0.177
 speaker-0.176
Token your
Token ID534
Feature activation+1.22e-01

Pos logit contributions
ez+0.133
eros+0.126
 wound+0.120
 tears+0.115
crow+0.114
Neg logit contributions
 topp-0.285
First-0.280
 ju-0.276
 flavor-0.264
 flour-0.263
Token voice
Token ID3809
Feature activation+1.76e-01

Pos logit contributions
ez+0.161
eros+0.152
crow+0.146
 wound+0.145
oring+0.144
Neg logit contributions
 topp-0.194
First-0.188
 ju-0.186
 flavor-0.178
 flour-0.176
Token on
Token ID319
Feature activation-3.79e-02

Pos logit contributions
ez+0.194
eros+0.186
 wound+0.178
 tears+0.174
 heaven+0.172
Neg logit contributions
 topp-0.314
First-0.309
 ju-0.304
 flavor-0.289
 flour-0.286
Token the
Token ID262
Feature activation+1.29e-02

Pos logit contributions
 topp+0.070
First+0.068
 ju+0.067
 flour+0.066
 flavor+0.065
Neg logit contributions
ez-0.039
eros-0.036
 wound-0.035
oring-0.035
 terrified-0.035
Token speaker
Token ID10834
Feature activation+3.03e-01

Pos logit contributions
ez+0.016
eros+0.016
 wound+0.015
 heaven+0.015
 tears+0.015
Neg logit contributions
 topp-0.021
First-0.021
 ju-0.020
 flavor-0.019
 flour-0.019
Token."
Token ID526
Feature activation-2.27e-02

Pos logit contributions
ez+0.304
eros+0.284
 wound+0.280
 heaven+0.275
-+0.273
Neg logit contributions
 topp-0.569
First-0.562
 ju-0.557
 flavor-0.522
 flour-0.520
Token
Token ID198
Feature activation+2.57e-02

Pos logit contributions
First+0.040
 topp+0.038
 ju+0.037
 flavor+0.036
 flour+0.036
Neg logit contributions
ez-0.028
eros-0.026
 terrified-0.025
 wound-0.024
 creature-0.024
Token
Token ID198
Feature activation+2.45e-02

Pos logit contributions
ez+0.032
eros+0.031
 terrified+0.029
ment+0.028
 creature+0.028
Neg logit contributions
First-0.045
 topp-0.040
 flavor-0.039
 flour-0.039
 ju-0.038
TokenTom
Token ID13787
Feature activation-3.17e-02

Pos logit contributions
ez+0.042
eros+0.041
 tears+0.039
 terrified+0.039
 wound+0.039
Neg logit contributions
First-0.031
 topp-0.029
 ju-0.028
 flavor-0.027
 flour-0.026
Tokenmy
Token ID1820
Feature activation-8.94e-02

Pos logit contributions
 ju+0.042
 topp+0.041
First+0.041
 flour+0.037
 flavor+0.036
Neg logit contributions
ez-0.052
 wound-0.049
--0.049
eros-0.048
oring-0.047
Token to
Token ID284
Feature activation-4.41e-02

Pos logit contributions
 topp+0.055
First+0.054
 ju+0.053
 flavor+0.050
 forks+0.050
Neg logit contributions
ez-0.038
eros-0.036
 wound-0.035
 tears-0.034
 heaven-0.033
Token the
Token ID262
Feature activation+9.25e-03

Pos logit contributions
 topp+0.077
First+0.074
 ju+0.072
 flavor+0.072
 forks+0.072
Neg logit contributions
ez-0.051
eros-0.048
oring-0.046
rows-0.045
 terrified-0.044
Token shore
Token ID15191
Feature activation+6.45e-02

Pos logit contributions
 wound+0.012
eros+0.012
 tears+0.011
 hospital+0.011
 creature+0.011
Neg logit contributions
First-0.015
 topp-0.015
 ju-0.014
 flavor-0.013
teen-0.013
Token and
Token ID290
Feature activation+3.62e-02

Pos logit contributions
ez+0.058
eros+0.055
 wound+0.052
 tears+0.051
 heaven+0.050
Neg logit contributions
 topp-0.128
First-0.126
 ju-0.124
 flavor-0.119
 flour-0.118
Token c
Token ID269
Feature activation+1.06e-01

Pos logit contributions
eros+0.059
ez+0.058
 heaven+0.053
oc+0.053
oring+0.053
Neg logit contributions
 ju-0.047
 topp-0.046
First-0.043
 would-0.043
 shovel-0.042
Tokenork
Token ID967
Feature activation+3.03e-01

Pos logit contributions
 speaker+0.156
ez+0.153
 heaven+0.149
oc+0.142
 Start+0.141
Neg logit contributions
 topp-0.154
ipp-0.151
 ju-0.143
First-0.138
 flavor-0.137
Tokened
Token ID276
Feature activation+2.73e-02

Pos logit contributions
ez+0.297
eros+0.286
 wound+0.274
 heaven+0.266
 tears+0.264
Neg logit contributions
 topp-0.573
First-0.563
 ju-0.555
 flavor-0.529
 flour-0.524
Token the
Token ID262
Feature activation+8.78e-03

Pos logit contributions
ez+0.026
eros+0.025
 wound+0.024
oring+0.023
 heaven+0.023
Neg logit contributions
 topp-0.052
 ju-0.051
First-0.051
 flavor-0.048
 forks-0.048
Token bottle
Token ID9294
Feature activation+2.59e-02

Pos logit contributions
 wound+0.011
eros+0.011
 tears+0.011
ez+0.011
 creature+0.011
Neg logit contributions
First-0.014
 topp-0.014
 ju-0.013
 flavor-0.012
teen-0.012
Token tightly
Token ID17707
Feature activation+1.29e-02

Pos logit contributions
ez+0.024
 wound+0.023
-+0.023
 tears+0.022
 its+0.022
Neg logit contributions
First-0.051
 topp-0.051
 ju-0.049
 flavor-0.046
 flour-0.046
Token.
Token ID13
Feature activation-8.23e-03

Pos logit contributions
ez+0.012
eros+0.011
 wound+0.011
-+0.011
 tears+0.011
Neg logit contributions
First-0.026
 topp-0.025
 ju-0.025
 flour-0.023
 flavor-0.023
Token a
Token ID257
Feature activation-7.16e-02

Pos logit contributions
ez+0.049
eros+0.048
oring+0.046
oc+0.044
 wound+0.044
Neg logit contributions
 ju-0.048
 topp-0.048
First-0.048
 flour-0.045
 flavor-0.044
Token gift
Token ID6979
Feature activation+8.99e-02

Pos logit contributions
 topp+0.096
First+0.094
 ju+0.093
 flavor+0.087
 flour+0.085
Neg logit contributions
ez-0.114
eros-0.109
oring-0.103
 wound-0.103
crow-0.102
Token,
Token ID11
Feature activation-2.32e-02

Pos logit contributions
ez+0.077
eros+0.071
 wound+0.067
 tears+0.065
 heaven+0.064
Neg logit contributions
 topp-0.183
First-0.181
 ju-0.179
 flavor-0.171
 flour-0.170
Token not
Token ID407
Feature activation+6.77e-02

Pos logit contributions
 topp+0.039
 ju+0.039
First+0.038
 flavor+0.037
 flour+0.036
Neg logit contributions
ez-0.029
eros-0.028
oring-0.026
crow-0.026
 wound-0.025
Token a
Token ID257
Feature activation-6.95e-02

Pos logit contributions
 its+0.069
 heaven+0.067
 tears+0.066
 crying+0.065
ez+0.065
Neg logit contributions
First-0.130
 topp-0.123
 ju-0.122
avored-0.114
 apr-0.114
Token curse
Token ID17328
Feature activation+3.02e-01

Pos logit contributions
 topp+0.108
First+0.106
 ju+0.104
 flavor+0.098
 flour+0.097
Neg logit contributions
ez-0.094
eros-0.090
 wound-0.086
oring-0.084
 tears-0.084
Token.
Token ID13
Feature activation+1.20e-02

Pos logit contributions
ez+0.309
eros+0.293
 wound+0.284
-+0.281
 heaven+0.279
Neg logit contributions
First-0.565
 topp-0.563
 ju-0.555
 flavor-0.515
 flour-0.513
Token They
Token ID1119
Feature activation-2.78e-02

Pos logit contributions
-+0.017
ez+0.017
 wound+0.017
eros+0.016
ling+0.016
Neg logit contributions
First-0.018
 topp-0.017
 ju-0.017
 flour-0.016
 First-0.015
Token learned
Token ID4499
Feature activation+1.11e-01

Pos logit contributions
First+0.037
 topp+0.037
 ju+0.036
 flavor+0.034
 forks+0.033
Neg logit contributions
ez-0.043
eros-0.042
 wound-0.039
oring-0.039
 heaven-0.039
Token that
Token ID326
Feature activation+8.52e-02

Pos logit contributions
ez+0.167
oring+0.151
oc+0.151
eros+0.151
ivering+0.145
Neg logit contributions
 flour-0.163
First-0.161
 topp-0.161
 flavor-0.155
 would-0.154
Token they
Token ID484
Feature activation+3.22e-03

Pos logit contributions
ez+0.099
eros+0.096
oring+0.090
 wound+0.089
crow+0.088
Neg logit contributions
 topp-0.148
First-0.145
 ju-0.143
 flavor-0.137
 flour-0.136
Token some
Token ID617
Feature activation-5.31e-02

Pos logit contributions
First+0.105
 ju+0.101
 topp+0.101
Deal+0.095
 flavor+0.094
Neg logit contributions
 its-0.036
ez-0.036
 tears-0.036
--0.034
 wound-0.033
Token ac
Token ID936
Feature activation-8.56e-02

Pos logit contributions
First+0.073
 ju+0.067
Deal+0.065
 topp+0.065
'd+0.062
Neg logit contributions
--0.082
 tears-0.081
 its-0.079
ez-0.078
les-0.077
Tokenorns
Token ID19942
Feature activation+5.03e-04

Pos logit contributions
 topp+0.157
First+0.155
 ju+0.152
 flavor+0.144
 flour+0.144
Neg logit contributions
ez-0.090
eros-0.087
 wound-0.085
 tears-0.084
 heaven-0.084
Token and
Token ID290
Feature activation+4.50e-02

Pos logit contributions
-+0.000
ez+0.000
 its+0.000
eros+0.000
 wound+0.000
Neg logit contributions
 topp-0.001
First-0.001
 ju-0.001
 flavor-0.001
 forks-0.001
Token pine
Token ID20161
Feature activation-7.55e-02

Pos logit contributions
ez+0.066
 tears+0.065
 wound+0.064
 its+0.063
ment+0.061
Neg logit contributions
First-0.064
 ju-0.061
 topp-0.059
Deal-0.056
avored-0.055
Token cones
Token ID47314
Feature activation-2.96e-01

Pos logit contributions
First+0.121
 ju+0.110
 topp+0.108
 redd+0.100
 flavor+0.098
Neg logit contributions
--0.107
ez-0.103
eros-0.103
ment-0.098
les-0.097
Token on
Token ID319
Feature activation-4.38e-02

Pos logit contributions
First+0.638
 topp+0.629
 ju+0.616
 forks+0.576
Deal+0.573
Neg logit contributions
--0.233
 its-0.217
ez-0.200
 wound-0.188
ment-0.187
Token them
Token ID606
Feature activation+2.38e-02

Pos logit contributions
First+0.092
 ju+0.091
 topp+0.089
Deal+0.084
'd+0.083
Neg logit contributions
 its-0.036
 tears-0.033
ez-0.032
 wound-0.032
--0.031
Token.
Token ID13
Feature activation+1.02e-02

Pos logit contributions
-+0.017
 its+0.016
ez+0.016
 wound+0.015
 tears+0.015
Neg logit contributions
First-0.053
 topp-0.052
 ju-0.052
 flour-0.048
 redd-0.047
Token They
Token ID1119
Feature activation-3.89e-02

Pos logit contributions
ez+0.010
eros+0.009
 wound+0.009
 heaven+0.009
 its+0.009
Neg logit contributions
 topp-0.019
 ju-0.019
First-0.019
 flavor-0.018
 forks-0.017
Token made
Token ID925
Feature activation-2.56e-03

Pos logit contributions
First+0.059
 topp+0.057
 ju+0.057
avored+0.051
 flour+0.051
Neg logit contributions
ez-0.052
 wound-0.052
 tears-0.051
 its-0.049
 speaker-0.047
Token of
Token ID286
Feature activation-3.79e-02

Pos logit contributions
First+0.449
 topp+0.442
 ju+0.416
teen+0.408
Deal+0.407
Neg logit contributions
--0.159
 wound-0.138
 its-0.129
ment-0.127
ez-0.121
Token ac
Token ID936
Feature activation-8.34e-02

Pos logit contributions
First+0.050
 topp+0.045
 ju+0.044
'd+0.042
Deal+0.041
Neg logit contributions
 tears-0.060
 its-0.060
 wound-0.060
ez-0.058
--0.057
Tokenorns
Token ID19942
Feature activation+2.49e-04

Pos logit contributions
 topp+0.163
First+0.161
 ju+0.158
 flavor+0.150
 flour+0.149
Neg logit contributions
ez-0.079
eros-0.075
 wound-0.072
 tears-0.070
 heaven-0.069
Token and
Token ID290
Feature activation+4.85e-02

Pos logit contributions
ez+0.000
eros+0.000
 wound+0.000
 heaven+0.000
 tears+0.000
Neg logit contributions
 topp-0.000
First-0.000
 ju-0.000
 flavor-0.000
 flour-0.000
Token pine
Token ID20161
Feature activation-7.16e-02

Pos logit contributions
ez+0.073
 wound+0.073
 tears+0.072
 its+0.070
 heaven+0.069
Neg logit contributions
First-0.067
 topp-0.064
 ju-0.063
 flavor-0.058
teen-0.055
Token cones
Token ID47314
Feature activation-2.96e-01

Pos logit contributions
First+0.093
 ju+0.082
 topp+0.082
 flavor+0.074
 redd+0.072
Neg logit contributions
eros-0.124
--0.123
ez-0.121
 wound-0.115
ment-0.115
Token on
Token ID319
Feature activation-3.98e-02

Pos logit contributions
 topp+0.605
First+0.605
 ju+0.590
 flavor+0.558
 flour+0.547
Neg logit contributions
ez-0.242
--0.240
eros-0.228
 wound-0.227
 its-0.224
Token the
Token ID262
Feature activation+1.23e-02

Pos logit contributions
First+0.085
 topp+0.084
 ju+0.084
Deal+0.078
 flavor+0.077
Neg logit contributions
 its-0.030
ez-0.029
 tears-0.028
 heaven-0.027
 wound-0.027
Token shelf
Token ID18316
Feature activation+7.16e-03

Pos logit contributions
 wound+0.015
 speaker+0.015
 tears+0.015
eros+0.015
 hospital+0.014
Neg logit contributions
First-0.022
 topp-0.020
 ju-0.019
 flavor-0.018
avored-0.018
Token.
Token ID13
Feature activation+9.02e-03

Pos logit contributions
-+0.005
 wound+0.005
 heaven+0.005
 its+0.004
ez+0.004
Neg logit contributions
First-0.016
 ju-0.016
 topp-0.015
 flavor-0.014
 fiber-0.014
Token They
Token ID1119
Feature activation-3.26e-02

Pos logit contributions
ez+0.011
eros+0.011
 wound+0.011
 tears+0.010
 heaven+0.010
Neg logit contributions
 topp-0.015
First-0.015
 ju-0.014
 flavor-0.014
 flour-0.013
Tokenips
Token ID2419
Feature activation-4.46e-02

Pos logit contributions
First+0.120
 topp+0.117
 ju+0.113
 flavor+0.111
 stack+0.106
Neg logit contributions
ez-0.103
oring-0.098
les-0.097
eros-0.096
--0.094
Token.
Token ID13
Feature activation-4.65e-06

Pos logit contributions
First+0.092
 topp+0.090
 ju+0.089
Deal+0.082
 stack+0.082
Neg logit contributions
--0.039
 its-0.038
ment-0.035
ez-0.035
les-0.033
Token They
Token ID1119
Feature activation-4.35e-02

Feature not active on prev token;
no predictions were affected.
Token see
Token ID766
Feature activation+8.17e-02

Pos logit contributions
 topp+0.066
First+0.065
 ju+0.060
avored+0.058
 flavor+0.057
Neg logit contributions
 tears-0.059
 its-0.057
 wound-0.057
 heaven-0.056
ez-0.056
Token pine
Token ID20161
Feature activation-7.77e-02

Pos logit contributions
 its+0.084
 tears+0.081
ez+0.081
-+0.078
eros+0.078
Neg logit contributions
First-0.155
 topp-0.150
 ju-0.142
 flavor-0.134
Deal-0.133
Token cones
Token ID47314
Feature activation-2.95e-01

Pos logit contributions
First+0.133
 topp+0.124
 ju+0.119
 flavor+0.111
 forks+0.110
Neg logit contributions
eros-0.102
--0.099
ez-0.095
crow-0.093
ment-0.092
Token,
Token ID11
Feature activation-3.20e-02

Pos logit contributions
 topp+0.658
First+0.653
 ju+0.640
 apr+0.605
Deal+0.591
Neg logit contributions
--0.208
 its-0.197
oring-0.176
ment-0.167
 wound-0.165
Token ac
Token ID936
Feature activation-8.59e-02

Pos logit contributions
First+0.049
 topp+0.046
Deal+0.043
 ju+0.042
avored+0.042
Neg logit contributions
 its-0.044
 tears-0.042
 wound-0.042
ment-0.040
ez-0.040
Tokenorns
Token ID19942
Feature activation-1.59e-03

Pos logit contributions
 topp+0.174
First+0.172
 ju+0.169
 flavor+0.160
 forks+0.160
Neg logit contributions
ez-0.076
eros-0.073
--0.067
 wound-0.066
crow-0.065
Token,
Token ID11
Feature activation-3.24e-02

Pos logit contributions
First+0.003
 topp+0.003
 ju+0.003
 flavor+0.003
 flour+0.003
Neg logit contributions
ez-0.001
eros-0.001
--0.001
 wound-0.001
 its-0.001
Token nuts
Token ID14380
Feature activation-1.97e-01

Pos logit contributions
First+0.070
 topp+0.067
 ju+0.063
Deal+0.062
 flavor+0.062
Neg logit contributions
 its-0.026
 wound-0.023
 tears-0.023
--0.022
ment-0.022
Token.
Token ID13
Feature activation+8.40e-04

Pos logit contributions
First+0.025
 ju+0.024
 topp+0.024
teen+0.022
 flavor+0.022
Neg logit contributions
ez-0.008
 tears-0.008
 wound-0.008
--0.008
 its-0.008
Token The
Token ID383
Feature activation+4.93e-02

Pos logit contributions
ez+0.001
eros+0.001
 wound+0.001
-+0.001
oring+0.001
Neg logit contributions
 topp-0.002
First-0.002
 ju-0.001
 flour-0.001
 flavor-0.001
Token bear
Token ID6842
Feature activation+2.60e-02

Pos logit contributions
 wound+0.030
eros+0.029
 tears+0.029
ez+0.026
 crying+0.026
Neg logit contributions
First-0.118
 topp-0.113
 ju-0.110
 flavor-0.104
ipp-0.104
Token takes
Token ID2753
Feature activation-1.44e-02

Pos logit contributions
ez+0.033
eros+0.033
crow+0.032
 speaker+0.031
 Start+0.031
Neg logit contributions
 topp-0.041
 ju-0.038
 flour-0.037
 forks-0.037
First-0.036
Token the
Token ID262
Feature activation+7.57e-03

Pos logit contributions
 topp+0.028
First+0.028
 ju+0.028
 flavor+0.026
 flour+0.026
Neg logit contributions
ez-0.013
eros-0.013
 wound-0.012
 heaven-0.012
 tears-0.012
Token cones
Token ID47314
Feature activation-2.95e-01

Pos logit contributions
 wound+0.013
 tears+0.013
-+0.012
eros+0.012
 blood+0.012
Neg logit contributions
First-0.010
raction-0.008
aney-0.008
 First-0.008
 topp-0.008
Token.
Token ID13
Feature activation-2.17e-03

Pos logit contributions
 topp+0.640
First+0.625
 ju+0.616
 flavor+0.587
teen+0.579
Neg logit contributions
ez-0.211
--0.205
 wound-0.205
 its-0.197
eros-0.194
Token The
Token ID383
Feature activation+4.77e-02

Pos logit contributions
 topp+0.004
First+0.004
 ju+0.004
 flour+0.004
 flavor+0.004
Neg logit contributions
ez-0.002
eros-0.002
 wound-0.002
--0.002
 heaven-0.002
Token bear
Token ID6842
Feature activation+2.48e-02

Pos logit contributions
 wound+0.029
eros+0.027
 tears+0.026
 lim+0.023
 hospital+0.022
Neg logit contributions
First-0.118
 topp-0.110
 ju-0.107
aney-0.107
ipp-0.105
Token leaves
Token ID5667
Feature activation-1.82e-01

Pos logit contributions
ez+0.033
eros+0.033
crow+0.031
 speaker+0.031
 knowing+0.031
Neg logit contributions
 topp-0.038
 ju-0.036
 flour-0.035
First-0.034
 forks-0.034
Token.
Token ID13
Feature activation-2.58e-03

Pos logit contributions
 topp+0.302
First+0.296
 ju+0.291
 flavor+0.275
 flour+0.274
Neg logit contributions
ez-0.223
eros-0.215
 wound-0.206
 heaven-0.201
 tears-0.200
Token
Token ID198
Feature activation+2.42e-02

Pos logit contributions
First+0.000
 ju+0.000
 topp+0.000
 flour+0.000
 flavor+0.000
Neg logit contributions
ez-0.000
eros-0.000
 wound-0.000
 heaven-0.000
 tears-0.000
Token
Token ID198
Feature activation+2.26e-02

Pos logit contributions
ez+0.022
eros+0.021
 wound+0.019
 heaven+0.019
 tears+0.019
Neg logit contributions
First-0.049
 topp-0.048
 ju-0.047
 flavor-0.045
 flour-0.045
TokenThey
Token ID2990
Feature activation-7.89e-02

Pos logit contributions
ez+0.028
eros+0.027
 wound+0.026
 tears+0.025
 heaven+0.025
Neg logit contributions
 topp-0.038
First-0.037
 ju-0.037
 flavor-0.034
 flour-0.034
Token reached
Token ID4251
Feature activation-1.03e-01

Pos logit contributions
 topp+0.131
First+0.123
 ju+0.121
 flour+0.116
 flavor+0.116
Neg logit contributions
ez-0.101
 tears-0.097
 wound-0.096
 its-0.095
 terrified-0.093
Token the
Token ID262
Feature activation+3.13e-03

Pos logit contributions
 topp+0.204
First+0.201
 ju+0.198
 flavor+0.188
 flour+0.187
Neg logit contributions
ez-0.095
eros-0.090
 wound-0.087
 tears-0.084
 heaven-0.083
Token cones
Token ID47314
Feature activation-2.95e-01

Pos logit contributions
 tears+0.004
 wound+0.004
 hospital+0.004
eros+0.003
 terrified+0.003
Neg logit contributions
 topp-0.005
First-0.005
 flour-0.005
 forks-0.005
 ju-0.005
Token and
Token ID290
Feature activation+3.85e-02

Pos logit contributions
 topp+0.642
 ju+0.613
First+0.610
 flavor+0.586
 flour+0.580
Neg logit contributions
oring-0.205
 wound-0.202
--0.200
 its-0.195
eros-0.188
Token the
Token ID262
Feature activation+3.77e-03

Pos logit contributions
ez+0.050
eros+0.047
 wound+0.047
 tears+0.046
 terrified+0.045
Neg logit contributions
 topp-0.061
First-0.061
 ju-0.058
 flavor-0.055
 flour-0.055
Token boss
Token ID6478
Feature activation+1.84e-02

Pos logit contributions
 tears+0.005
 wound+0.004
 woman+0.004
 crying+0.004
 hospital+0.004
Neg logit contributions
First-0.007
 topp-0.007
 forks-0.006
 flour-0.006
 shovel-0.006
Tokeny
Token ID88
Feature activation+3.23e-02

Pos logit contributions
ez+0.024
eros+0.022
crow+0.022
-+0.022
 wound+0.021
Neg logit contributions
 topp-0.030
First-0.030
 ju-0.029
 forks-0.027
 flour-0.027
Token pony
Token ID26902
Feature activation+4.01e-02

Pos logit contributions
ez+0.035
eros+0.034
 wound+0.033
 tears+0.032
-+0.031
Neg logit contributions
 topp-0.059
First-0.058
 ju-0.056
 flour-0.053
 flavor-0.053
Token balls
Token ID11333
Feature activation-1.98e-01

Pos logit contributions
First+0.170
Deal+0.155
 topp+0.154
 ju+0.147
teen+0.142
Neg logit contributions
 tears-0.126
 its-0.122
ez-0.116
 knowing-0.113
 crying-0.113
Token of
Token ID286
Feature activation-3.60e-02

Pos logit contributions
First+0.409
 topp+0.404
 ju+0.389
 flavor+0.370
 apr+0.369
Neg logit contributions
ez-0.170
 crying-0.165
--0.165
oring-0.156
 wound-0.156
Token ice
Token ID4771
Feature activation-1.24e-01

Pos logit contributions
First+0.054
Deal+0.047
 ju+0.046
 topp+0.045
ven+0.044
Neg logit contributions
 tears-0.053
 its-0.051
 crying-0.050
ez-0.050
 wound-0.049
Tokencream
Token ID36277
Feature activation-7.49e-02

Pos logit contributions
First+0.247
 ju+0.226
 topp+0.217
 First+0.214
teen+0.208
Neg logit contributions
--0.147
les-0.119
eros-0.116
ez-0.115
ment-0.110
Token on
Token ID319
Feature activation-3.82e-02

Pos logit contributions
First+0.158
 topp+0.149
 ju+0.145
 apr+0.139
 fiber+0.138
Neg logit contributions
--0.064
eros-0.057
oring-0.056
ez-0.055
 its-0.053
Token cones
Token ID47314
Feature activation-2.94e-01

Pos logit contributions
First+0.081
 topp+0.081
 ju+0.080
 flavor+0.075
 flour+0.075
Neg logit contributions
ez-0.028
eros-0.027
 wound-0.026
 tears-0.025
 its-0.025
Token for
Token ID329
Feature activation+8.60e-02

Pos logit contributions
First+0.604
 topp+0.597
 ju+0.590
 flour+0.556
 apr+0.550
Neg logit contributions
--0.240
ez-0.232
oring-0.228
eros-0.216
 wound-0.214
Token everyone
Token ID2506
Feature activation+6.58e-02

Pos logit contributions
ez+0.094
eros+0.086
 heaven+0.086
 tears+0.086
 its+0.085
Neg logit contributions
First-0.157
 ju-0.156
 topp-0.155
 flour-0.141
 flavor-0.140
Token!
Token ID0
Feature activation-2.20e-02

Pos logit contributions
ez+0.065
-+0.063
 crying+0.060
 its+0.060
 wound+0.059
Neg logit contributions
First-0.127
 topp-0.124
 ju-0.121
 flour-0.116
 flavor-0.112
Token Derek
Token ID20893
Feature activation+2.61e-02

Pos logit contributions
First+0.037
 ju+0.034
 topp+0.032
 First+0.032
ipp+0.031
Neg logit contributions
 heaven-0.029
ez-0.028
eros-0.028
web-0.027
 wound-0.026
Token was
Token ID373
Feature activation+2.37e-02

Pos logit contributions
ez+0.029
 wound+0.028
eros+0.028
 tears+0.027
-+0.027
Neg logit contributions
First-0.046
 topp-0.046
 ju-0.045
 flour-0.042
 flavor-0.042
Token,
Token ID11
Feature activation-3.97e-02

Pos logit contributions
 topp+0.216
First+0.213
 ju+0.209
 flavor+0.197
 flour+0.195
Neg logit contributions
ez-0.171
eros-0.165
 wound-0.160
 tears-0.156
 heaven-0.156
Token they
Token ID484
Feature activation-7.25e-03

Pos logit contributions
 topp+0.074
First+0.073
 ju+0.071
 flavor+0.068
 flour+0.067
Neg logit contributions
ez-0.041
eros-0.039
 wound-0.038
 tears-0.037
 heaven-0.037
Token threw
Token ID9617
Feature activation-5.70e-02

Pos logit contributions
 topp+0.014
First+0.013
 ju+0.013
 flavor+0.012
avored+0.012
Neg logit contributions
ez-0.008
 wound-0.008
 tears-0.007
 its-0.007
 lim-0.007
Token some
Token ID617
Feature activation-5.10e-02

Pos logit contributions
First+0.117
 topp+0.115
 ju+0.113
 flavor+0.107
 forks+0.105
Neg logit contributions
ez-0.048
 tears-0.045
eros-0.044
 wound-0.044
 its-0.044
Token pine
Token ID20161
Feature activation-7.28e-02

Pos logit contributions
First+0.081
 ju+0.076
 topp+0.075
Deal+0.073
'd+0.072
Neg logit contributions
 tears-0.068
ez-0.068
--0.067
 wound-0.064
 its-0.063
Token cones
Token ID47314
Feature activation-2.94e-01

Pos logit contributions
First+0.093
 topp+0.085
 ju+0.081
 flavor+0.075
 apr+0.072
Neg logit contributions
--0.127
eros-0.124
ez-0.121
crow-0.118
ment-0.117
Token on
Token ID319
Feature activation-4.38e-02

Pos logit contributions
 topp+0.651
First+0.631
 ju+0.622
 flavor+0.597
 forks+0.589
Neg logit contributions
--0.217
 its-0.209
ez-0.195
oring-0.190
 crying-0.181
Token top
Token ID1353
Feature activation-5.34e-02

Pos logit contributions
First+0.092
 topp+0.091
 ju+0.091
 flavor+0.084
 forks+0.083
Neg logit contributions
 its-0.034
ez-0.033
eros-0.033
 tears-0.032
 wound-0.031
Token to
Token ID284
Feature activation-4.47e-02

Pos logit contributions
 ju+0.114
 topp+0.114
First+0.114
Deal+0.109
 apr+0.106
Neg logit contributions
--0.041
 its-0.037
ez-0.036
crow-0.036
ment-0.035
Token make
Token ID787
Feature activation+1.14e-03

Pos logit contributions
 topp+0.075
First+0.075
 ju+0.072
 flour+0.067
 forks+0.067
Neg logit contributions
ez-0.053
 wound-0.052
eros-0.052
 tears-0.051
 its-0.050
Token it
Token ID340
Feature activation+1.04e-01

Pos logit contributions
ez+0.001
eros+0.001
 wound+0.001
 tears+0.001
 heaven+0.001
Neg logit contributions
First-0.002
 topp-0.002
 ju-0.002
 flavor-0.002
 forks-0.002
Token Then
Token ID3244
Feature activation-1.01e-01

Pos logit contributions
ez+0.010
eros+0.010
 tears+0.010
 wound+0.010
 heaven+0.010
Neg logit contributions
 topp-0.019
First-0.018
 ju-0.018
 flavor-0.017
 forks-0.017
Token they
Token ID484
Feature activation-9.64e-04

Pos logit contributions
 topp+0.195
First+0.190
 ju+0.188
 flavor+0.173
 forks+0.170
Neg logit contributions
 its-0.096
 tears-0.094
eros-0.092
--0.091
 wound-0.091
Token put
Token ID1234
Feature activation-2.65e-02

Pos logit contributions
avored+0.002
First+0.002
 ju+0.002
 fiber+0.002
umm+0.002
Neg logit contributions
 tears-0.001
 cried-0.001
 wound-0.001
 realized-0.001
ez-0.001
Token some
Token ID617
Feature activation-4.60e-02

Pos logit contributions
First+0.058
 topp+0.056
 ju+0.055
 flavor+0.052
'd+0.052
Neg logit contributions
 tears-0.019
 its-0.019
ez-0.018
eros-0.017
 wound-0.017
Token pine
Token ID20161
Feature activation-6.83e-02

Pos logit contributions
First+0.065
 ju+0.057
'd+0.057
Deal+0.056
 topp+0.053
Neg logit contributions
--0.071
 tears-0.070
 wound-0.067
 its-0.067
les-0.064
Token cones
Token ID47314
Feature activation-2.94e-01

Pos logit contributions
First+0.068
 topp+0.058
 ju+0.058
 flavor+0.052
 usual+0.049
Neg logit contributions
--0.139
eros-0.137
ez-0.129
 wound-0.128
ment-0.128
Token on
Token ID319
Feature activation-3.58e-02

Pos logit contributions
 topp+0.656
First+0.645
 ju+0.635
 flavor+0.608
 fiber+0.604
Neg logit contributions
--0.214
 its-0.184
oring-0.174
 wound-0.171
eros-0.166
Token the
Token ID262
Feature activation+1.41e-02

Pos logit contributions
 topp+0.066
First+0.066
 ju+0.066
 flavor+0.060
 forks+0.060
Neg logit contributions
 its-0.037
 tears-0.035
ez-0.034
eros-0.034
 wound-0.034
Token top
Token ID1353
Feature activation-4.79e-02

Pos logit contributions
 wound+0.018
 tears+0.018
eros+0.016
 speaker+0.016
 hospital+0.016
Neg logit contributions
First-0.025
 topp-0.024
 ju-0.022
 flavor-0.021
Deal-0.021
Token to
Token ID284
Feature activation-3.97e-02

Pos logit contributions
 ju+0.110
First+0.109
 topp+0.108
Deal+0.106
 fiber+0.101
Neg logit contributions
--0.032
er-0.027
 its-0.027
 wound-0.024
crow-0.024
Token make
Token ID787
Feature activation+5.29e-03

Pos logit contributions
First+0.059
 topp+0.056
 ju+0.055
 forks+0.049
avored+0.049
Neg logit contributions
 its-0.057
 wound-0.056
 tears-0.054
 heaven-0.053
ez-0.053
Token nods
Token ID34833
Feature activation-1.49e-03

Pos logit contributions
ez+0.192
eros+0.184
 wound+0.181
 tears+0.178
 terrified+0.172
Neg logit contributions
First-0.270
 topp-0.263
 ju-0.261
 flavor-0.243
 flour-0.241
Token and
Token ID290
Feature activation+3.91e-02

Pos logit contributions
 topp+0.002
First+0.002
 flavor+0.002
 would+0.002
 forks+0.002
Neg logit contributions
eros-0.002
lves-0.002
ez-0.002
 terrified-0.002
 hospital-0.002
Token gives
Token ID3607
Feature activation+5.09e-02

Pos logit contributions
ez+0.047
 wound+0.046
 tears+0.046
eros+0.046
 heaven+0.043
Neg logit contributions
First-0.066
 topp-0.063
 ju-0.062
 flavor-0.059
 flour-0.058
Token her
Token ID607
Feature activation+8.43e-02

Pos logit contributions
ez+0.043
 tears+0.039
eros+0.039
 heaven+0.039
 its+0.038
Neg logit contributions
 topp-0.103
First-0.103
 ju-0.102
 flavor-0.096
 stack-0.094
Token two
Token ID734
Feature activation-2.02e-01

Pos logit contributions
ez+0.073
-+0.070
 tears+0.069
 its+0.069
 wound+0.068
Neg logit contributions
First-0.171
 topp-0.162
 ju-0.157
'd-0.150
teen-0.149
Token cones
Token ID47314
Feature activation-2.94e-01

Pos logit contributions
First+0.261
Deal+0.236
'd+0.228
 ju+0.228
 topp+0.227
Neg logit contributions
 tears-0.329
--0.325
 wound-0.306
 relieved-0.290
ez-0.290
Token.
Token ID13
Feature activation+5.60e-03

Pos logit contributions
First+0.612
 topp+0.596
 ju+0.593
teen+0.552
 flavor+0.545
Neg logit contributions
--0.247
ez-0.229
 its-0.221
ment-0.221
eros-0.220
Token Anna
Token ID11735
Feature activation+1.97e-02

Pos logit contributions
ez+0.007
eros+0.007
-+0.006
 wound+0.006
rows+0.006
Neg logit contributions
 topp-0.010
First-0.010
 ju-0.009
 forks-0.009
 flour-0.009
Token pays
Token ID13831
Feature activation+3.51e-02

Pos logit contributions
ez+0.022
 wound+0.021
 tears+0.021
eros+0.020
oring+0.019
Neg logit contributions
First-0.036
 ju-0.035
 topp-0.034
 flavor-0.032
 usual-0.032
Token with
Token ID351
Feature activation+1.31e-01

Pos logit contributions
ez+0.030
eros+0.028
-+0.027
 tears+0.027
 wound+0.027
Neg logit contributions
 topp-0.072
First-0.071
 ju-0.071
 flavor-0.067
 flour-0.066
Token her
Token ID607
Feature activation+8.44e-02

Pos logit contributions
 tears+0.123
 its+0.114
ez+0.105
 heaven+0.103
 crying+0.097
Neg logit contributions
First-0.280
 topp-0.268
 ju-0.264
'd-0.255
Deal-0.251
Token have
Token ID423
Feature activation+4.94e-02

Pos logit contributions
 topp+0.082
 ju+0.077
 flour+0.077
 flavor+0.074
avored+0.074
Neg logit contributions
 its-0.050
 wound-0.047
 tears-0.047
 alone-0.046
 speaker-0.046
Token a
Token ID257
Feature activation-7.14e-02

Pos logit contributions
 its+0.039
 tears+0.030
 his+0.029
 someone+0.027
-+0.026
Neg logit contributions
umm-0.110
 topp-0.107
avored-0.106
 ju-0.106
teen-0.101
Token ball
Token ID2613
Feature activation-1.37e-01

Pos logit contributions
First+0.117
 topp+0.115
 ju+0.114
nuts+0.109
avored+0.105
Neg logit contributions
 wound-0.089
--0.088
 creature-0.085
 tears-0.084
 speaker-0.082
Token and
Token ID290
Feature activation+3.44e-02

Pos logit contributions
 topp+0.326
First+0.326
 ju+0.322
 flavor+0.302
 flour+0.298
Neg logit contributions
--0.073
ez-0.069
eros-0.069
 its-0.066
 wound-0.066
Token two
Token ID734
Feature activation-1.98e-01

Pos logit contributions
 its+0.027
 tears+0.024
 someone+0.021
 his+0.021
 wound+0.020
Neg logit contributions
 ju-0.072
First-0.070
 topp-0.070
umm-0.069
avored-0.068
Token cones
Token ID47314
Feature activation-2.94e-01

Pos logit contributions
 ju+0.373
First+0.357
avored+0.356
 topp+0.355
umm+0.343
Neg logit contributions
 tears-0.235
--0.213
les-0.207
 cries-0.194
 and-0.194
Token for
Token ID329
Feature activation+8.76e-02

Pos logit contributions
 ju+0.589
First+0.585
 topp+0.583
 flour+0.539
 flavor+0.538
Neg logit contributions
ez-0.257
eros-0.245
--0.242
 wound-0.239
 its-0.235
Token the
Token ID262
Feature activation+8.18e-03

Pos logit contributions
 tears+0.091
 its+0.082
 crying+0.076
 his+0.075
eros+0.066
Neg logit contributions
First-0.185
 ju-0.184
'd-0.175
Deal-0.167
 redd-0.166
Token goal
Token ID3061
Feature activation-6.07e-03

Pos logit contributions
 tears+0.009
 creature+0.009
 noise+0.009
 hospital+0.009
 wound+0.009
Neg logit contributions
First-0.014
 topp-0.014
aney-0.013
ven-0.013
teen-0.013
Token.
Token ID13
Feature activation-8.42e-03

Pos logit contributions
 ju+0.016
 topp+0.015
First+0.014
 redd+0.014
 apr+0.014
Neg logit contributions
--0.004
 its-0.003
er-0.003
crow-0.002
 wound-0.002
Token Sara
Token ID24799
Feature activation+3.46e-02

Pos logit contributions
 topp+0.017
 ju+0.017
First+0.016
nuts+0.016
'd+0.016
Neg logit contributions
 its-0.007
 Start-0.007
 tears-0.006
 heaven-0.006
 wound-0.006
Token Tom
Token ID4186
Feature activation-7.63e-02

Pos logit contributions
ez+0.084
eros+0.081
rows+0.080
crow+0.080
oring+0.079
Neg logit contributions
 topp-0.047
 ju-0.045
First-0.043
 forks-0.041
 flavor-0.041
Token and
Token ID290
Feature activation+4.51e-02

Pos logit contributions
 topp+0.162
 ju+0.162
First+0.161
 flavor+0.152
 flour+0.150
Neg logit contributions
ez-0.061
 wound-0.056
 crying-0.055
 heaven-0.055
 tears-0.053
Token handed
Token ID10158
Feature activation+3.30e-03

Pos logit contributions
ez+0.071
eros+0.070
oring+0.067
rows+0.067
crow+0.066
Neg logit contributions
 topp-0.061
 ju-0.059
First-0.057
 flavor-0.055
 forks-0.052
Token them
Token ID606
Feature activation+2.41e-02

Pos logit contributions
ez+0.003
eros+0.003
 wound+0.003
 terrified+0.003
oring+0.003
Neg logit contributions
 topp-0.006
First-0.006
 ju-0.006
 flavor-0.006
 flour-0.006
Token the
Token ID262
Feature activation+7.18e-03

Pos logit contributions
ez+0.025
eros+0.024
 wound+0.024
-+0.023
 tears+0.023
Neg logit contributions
 topp-0.043
First-0.043
 ju-0.042
 flavor-0.040
 flour-0.039
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
 tears+0.011
 wound+0.011
-+0.010
 crying+0.010
 blood+0.010
Neg logit contributions
First-0.010
teen-0.009
 topp-0.009
glers-0.008
 First-0.008
Token.
Token ID13
Feature activation+1.64e-02

Pos logit contributions
First+0.608
 topp+0.602
 ju+0.589
teen+0.558
 flour+0.547
Neg logit contributions
--0.235
 wound-0.219
ez-0.218
eros-0.215
 its-0.211
Token
Token ID198
Feature activation+2.64e-02

Pos logit contributions
ez+0.018
eros+0.017
 wound+0.016
 terrified+0.016
 tears+0.015
Neg logit contributions
First-0.031
 topp-0.030
 ju-0.030
 flour-0.028
 flavor-0.028
Token
Token ID198
Feature activation+2.43e-02

Pos logit contributions
ez+0.035
eros+0.034
 creature+0.031
 terrified+0.031
 heaven+0.031
Neg logit contributions
First-0.045
 topp-0.039
 flour-0.038
 flavor-0.038
 ju-0.038
Token"
Token ID1
Feature activation-4.46e-02

Pos logit contributions
ez+0.036
eros+0.035
 tears+0.034
 crying+0.033
 creature+0.033
Neg logit contributions
First-0.038
 topp-0.035
 ju-0.035
 flour-0.032
 flavor-0.032
TokenHere
Token ID4342
Feature activation-5.91e-02

Pos logit contributions
 topp+0.073
First+0.070
 ju+0.070
 stack+0.069
 forks+0.069
Neg logit contributions
ez-0.051
Start-0.047
crow-0.047
 wound-0.047
 heaven-0.046
Token them
Token ID606
Feature activation+2.47e-02

Pos logit contributions
ez+0.036
eros+0.034
 its+0.032
 tears+0.032
-+0.032
Neg logit contributions
First-0.102
 topp-0.100
 ju-0.099
 flavor-0.093
 flour-0.091
Token ice
Token ID4771
Feature activation-1.23e-01

Pos logit contributions
-+0.026
ez+0.026
eros+0.025
 its+0.025
 heaven+0.024
Neg logit contributions
First-0.044
 topp-0.043
 ju-0.042
 flavor-0.039
'd-0.038
Token-
Token ID12
Feature activation+1.64e-01

Pos logit contributions
First+0.313
 ju+0.301
 topp+0.299
 forks+0.281
teen+0.281
Neg logit contributions
--0.054
eros-0.044
ez-0.044
 wound-0.038
ment-0.037
Tokencream
Token ID36277
Feature activation-7.86e-02

Pos logit contributions
ez+0.220
eros+0.211
 wound+0.203
-+0.200
crow+0.198
Neg logit contributions
 topp-0.253
First-0.252
 ju-0.245
 flavor-0.229
 forks-0.228
Token in
Token ID287
Feature activation-5.63e-02

Pos logit contributions
First+0.171
 topp+0.163
 ju+0.161
 forks+0.150
teen+0.150
Neg logit contributions
--0.061
eros-0.055
ment-0.051
ez-0.051
 its-0.051
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
 topp+0.116
First+0.116
 ju+0.115
 flavor+0.107
 flour+0.107
Neg logit contributions
ez-0.045
eros-0.042
 wound-0.041
 tears-0.040
 heaven-0.040
Token.
Token ID13
Feature activation+2.09e-03

Pos logit contributions
 topp+0.597
First+0.593
 ju+0.592
 flour+0.547
 forks+0.545
Neg logit contributions
--0.233
ez-0.232
eros-0.223
 wound-0.220
 its-0.217
Token He
Token ID679
Feature activation-3.48e-02

Pos logit contributions
ez+0.002
eros+0.002
 wound+0.002
-+0.002
oring+0.002
Neg logit contributions
First-0.004
 topp-0.004
 ju-0.004
 flour-0.003
 forks-0.003
Token is
Token ID318
Feature activation+6.75e-02

Pos logit contributions
First+0.061
 ju+0.059
 topp+0.058
 flavor+0.054
 flour+0.053
Neg logit contributions
ez-0.040
 wound-0.039
 tears-0.038
eros-0.038
 heaven-0.036
Token nice
Token ID3621
Feature activation-6.12e-02

Pos logit contributions
ez+0.072
 wound+0.069
eros+0.068
 tears+0.067
 terrified+0.067
Neg logit contributions
First-0.123
 topp-0.122
 ju-0.118
 flavor-0.111
 forks-0.110
Token.
Token ID13
Feature activation+2.27e-03

Pos logit contributions
First+0.140
 ju+0.136
 topp+0.135
 flour+0.125
 apr+0.124
Neg logit contributions
--0.044
eros-0.037
 knowing-0.036
ez-0.036
 heaven-0.035
Token mom
Token ID1995
Feature activation-4.12e-02

Pos logit contributions
 tears+0.070
-+0.067
 crying+0.066
 wound+0.065
 voice+0.059
Neg logit contributions
First-0.156
 ju-0.143
 topp-0.142
nuts-0.130
Deal-0.130
Token pulled
Token ID5954
Feature activation-9.09e-02

Pos logit contributions
 topp+0.078
First+0.076
 ju+0.076
 flavor+0.071
 flour+0.070
Neg logit contributions
ez-0.043
 wound-0.040
--0.040
 crying-0.040
eros-0.039
Token two
Token ID734
Feature activation-2.01e-01

Pos logit contributions
 topp+0.189
First+0.187
 ju+0.185
 flavor+0.175
 flour+0.172
Neg logit contributions
ez-0.076
eros-0.069
 wound-0.067
 tears-0.067
--0.065
Token ice
Token ID4771
Feature activation-1.23e-01

Pos logit contributions
First+0.334
 topp+0.331
Deal+0.322
 ju+0.306
 flavor+0.290
Neg logit contributions
--0.264
 tears-0.263
 crying-0.257
 terrified-0.251
ez-0.251
Token cream
Token ID8566
Feature activation-1.27e-01

Pos logit contributions
First+0.243
 ju+0.227
 topp+0.220
 redd+0.209
'd+0.206
Neg logit contributions
--0.150
ment-0.126
les-0.119
ez-0.119
rows-0.115
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
First+0.295
 topp+0.276
 ju+0.269
Deal+0.259
 redd+0.258
Neg logit contributions
--0.095
 its-0.075
ez-0.070
er-0.069
eros-0.069
Token out
Token ID503
Feature activation+3.30e-02

Pos logit contributions
 topp+0.654
First+0.653
 ju+0.630
Deal+0.606
 apr+0.604
Neg logit contributions
--0.212
 its-0.187
oring-0.183
ez-0.174
 crying-0.172
Token of
Token ID286
Feature activation-3.37e-02

Pos logit contributions
 its+0.027
-+0.026
ez+0.025
 tears+0.024
ment+0.024
Neg logit contributions
First-0.070
 topp-0.069
 ju-0.069
Deal-0.064
 flavor-0.064
Token her
Token ID607
Feature activation+8.10e-02

Pos logit contributions
First+0.064
Deal+0.060
 ju+0.059
'd+0.058
 topp+0.056
Neg logit contributions
 its-0.037
 tears-0.036
ez-0.033
--0.032
 heaven-0.031
Token bag
Token ID6131
Feature activation-3.85e-02

Pos logit contributions
-+0.094
 wound+0.091
 tears+0.089
eros+0.086
ez+0.085
Neg logit contributions
First-0.157
 ju-0.143
 topp-0.142
Deal-0.132
 redd-0.129
Token.
Token ID13
Feature activation+1.84e-03

Pos logit contributions
 topp+0.082
First+0.081
 ju+0.079
 apr+0.074
 forks+0.074
Neg logit contributions
--0.031
ez-0.027
 wound-0.027
 crying-0.027
 its-0.027
Token tree
Token ID5509
Feature activation-3.47e-02

Pos logit contributions
 topp+0.089
First+0.087
 ju+0.085
 flavor+0.079
 flour+0.078
Neg logit contributions
ez-0.101
eros-0.098
 wound-0.096
 tears-0.093
 heaven-0.093
Token and
Token ID290
Feature activation+3.04e-02

Pos logit contributions
 topp+0.075
First+0.074
 ju+0.074
 flavor+0.070
 flour+0.069
Neg logit contributions
 wound-0.025
ez-0.024
 heaven-0.023
eros-0.023
--0.023
Token look
Token ID804
Feature activation+2.77e-02

Pos logit contributions
ez+0.048
eros+0.047
 wound+0.046
 tears+0.045
 heaven+0.045
Neg logit contributions
 topp-0.040
First-0.039
 ju-0.038
 flavor-0.036
 flour-0.035
Token for
Token ID329
Feature activation+8.62e-02

Pos logit contributions
ez+0.038
eros+0.037
 speaker+0.036
crow+0.035
age+0.035
Neg logit contributions
 ju-0.041
 forks-0.041
 topp-0.040
 flour-0.039
 shovel-0.038
Token the
Token ID262
Feature activation+7.90e-03

Pos logit contributions
ez+0.082
eros+0.080
 tears+0.078
 wound+0.077
 its+0.077
Neg logit contributions
First-0.165
 topp-0.164
 ju-0.160
 flavor-0.149
 forks-0.148
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
 wound+0.010
 tears+0.010
 lim+0.009
 hospital+0.009
-+0.009
Neg logit contributions
 topp-0.013
First-0.012
 forks-0.011
 apr-0.011
avored-0.011
Token.
Token ID13
Feature activation-5.09e-03

Pos logit contributions
 topp+0.618
First+0.610
 ju+0.597
teen+0.566
 flavor+0.561
Neg logit contributions
--0.225
 wound-0.220
ez-0.216
eros-0.215
 its-0.214
Token They
Token ID1119
Feature activation-4.73e-02

Pos logit contributions
 topp+0.009
First+0.009
 ju+0.009
 flavor+0.009
 flour+0.008
Neg logit contributions
eros-0.005
 tears-0.005
 wound-0.005
 heaven-0.005
ez-0.005
Token find
Token ID1064
Feature activation+1.29e-01

Pos logit contributions
 topp+0.096
First+0.091
 ju+0.087
 flavor+0.086
 flour+0.086
Neg logit contributions
 wound-0.046
 tears-0.046
 terrified-0.042
 heaven-0.041
 relieved-0.041
Token some
Token ID617
Feature activation-4.63e-02

Pos logit contributions
ez+0.149
eros+0.142
 wound+0.138
 tears+0.134
 heaven+0.134
Neg logit contributions
 topp-0.223
First-0.221
 ju-0.216
 flavor-0.204
 flour-0.203
Token on
Token ID319
Feature activation-3.50e-02

Pos logit contributions
First+0.066
 topp+0.063
ipp+0.059
ven+0.058
aney+0.058
Neg logit contributions
--0.067
 wound-0.064
 tears-0.064
 its-0.062
 blood-0.061
Token gift
Token ID6979
Feature activation+8.63e-02

Pos logit contributions
First+0.105
 topp+0.098
 ju+0.096
teen+0.092
 forks+0.092
Neg logit contributions
--0.097
 wound-0.097
 Start-0.095
 tears-0.094
 crying-0.093
Token.
Token ID13
Feature activation+7.57e-03

Pos logit contributions
ez+0.078
eros+0.073
 wound+0.070
 tears+0.068
 heaven+0.067
Neg logit contributions
 topp-0.171
First-0.169
 ju-0.167
 flavor-0.159
 flour-0.158
Token He
Token ID679
Feature activation-3.01e-02

Pos logit contributions
ez+0.010
eros+0.009
ling+0.009
les+0.009
 crying+0.008
Neg logit contributions
First-0.013
 ju-0.013
 topp-0.012
 flour-0.012
 flavor-0.012
Token took
Token ID1718
Feature activation-4.22e-02

Pos logit contributions
 topp+0.044
First+0.042
 ju+0.042
 flavor+0.040
 flour+0.039
Neg logit contributions
ez-0.044
eros-0.042
 heaven-0.040
oring-0.040
crow-0.040
Token two
Token ID734
Feature activation-2.04e-01

Pos logit contributions
 topp+0.071
 flour+0.067
First+0.067
 ju+0.067
 forks+0.065
Neg logit contributions
ez-0.052
eros-0.050
oring-0.050
ller-0.049
 knowing-0.049
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
First+0.299
 topp+0.295
 ju+0.287
 flavor+0.265
 flour+0.256
Neg logit contributions
ez-0.282
 tears-0.282
 wound-0.279
--0.273
eros-0.270
Token and
Token ID290
Feature activation+4.05e-02

Pos logit contributions
 topp+0.598
First+0.591
 ju+0.582
 flavor+0.554
 flour+0.549
Neg logit contributions
ez-0.249
eros-0.236
 wound-0.226
 tears-0.216
 heaven-0.215
Token put
Token ID1234
Feature activation-3.41e-02

Pos logit contributions
 tears+0.052
 wound+0.050
ez+0.048
-+0.048
 its+0.048
Neg logit contributions
First-0.066
 ju-0.061
 topp-0.060
Deal-0.058
avored-0.057
Token a
Token ID257
Feature activation-7.05e-02

Pos logit contributions
First+0.071
 topp+0.071
 ju+0.069
 flavor+0.066
 forks+0.065
Neg logit contributions
ez-0.027
eros-0.025
 tears-0.024
 wound-0.024
--0.023
Token small
Token ID1402
Feature activation-6.91e-02

Pos logit contributions
First+0.096
 topp+0.092
 ju+0.090
 forks+0.082
 flavor+0.080
Neg logit contributions
 wound-0.106
ez-0.104
--0.103
eros-0.102
 tears-0.102
Token scoop
Token ID23529
Feature activation-5.58e-02

Pos logit contributions
First+0.090
 topp+0.087
 ju+0.086
 flavor+0.076
 forks+0.076
Neg logit contributions
ez-0.108
 wound-0.106
eros-0.106
--0.104
 tears-0.103
Token disappeared
Token ID12120
Feature activation+8.92e-02

Pos logit contributions
eros+0.194
ucks+0.189
 wound+0.184
rows+0.181
 speaker+0.180
Neg logit contributions
 would-0.190
 ju-0.186
'd-0.183
 topp-0.176
First-0.169
Token,
Token ID11
Feature activation-3.56e-02

Pos logit contributions
ez+0.093
eros+0.088
 wound+0.084
 heaven+0.081
 tears+0.081
Neg logit contributions
 topp-0.165
First-0.162
 ju-0.162
 flavor-0.153
 flour-0.151
Token and
Token ID290
Feature activation+3.26e-02

Pos logit contributions
 ju+0.058
 topp+0.055
 would+0.054
'd+0.052
 usual+0.052
Neg logit contributions
eros-0.047
ez-0.045
crow-0.043
rows-0.042
iren-0.040
Token the
Token ID262
Feature activation+3.69e-03

Pos logit contributions
ez+0.039
eros+0.038
 wound+0.036
oring+0.036
 speaker+0.035
Neg logit contributions
 topp-0.055
 ju-0.055
First-0.053
 flavor-0.051
 forks-0.050
Token two
Token ID734
Feature activation-2.05e-01

Pos logit contributions
 tears+0.006
 wound+0.005
 relieved+0.005
 crying+0.005
 hospital+0.005
Neg logit contributions
First-0.006
glers-0.005
 topp-0.005
aney-0.005
 First-0.005
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
First+0.377
 topp+0.352
 ju+0.345
 First+0.343
 apr+0.325
Neg logit contributions
 tears-0.250
--0.240
 wound-0.215
les-0.214
 relieved-0.214
Token were
Token ID547
Feature activation-9.91e-02

Pos logit contributions
First+0.375
 topp+0.374
 ju+0.364
 flour+0.330
 flavor+0.327
Neg logit contributions
ez-0.472
eros-0.456
 wound-0.451
 tears-0.440
--0.438
Token happy
Token ID3772
Feature activation+3.29e-02

Pos logit contributions
 topp+0.168
First+0.166
 ju+0.162
 flavor+0.153
 flour+0.152
Neg logit contributions
ez-0.120
eros-0.114
 wound-0.111
 tears-0.110
 heaven-0.108
Token.
Token ID13
Feature activation-7.06e-03

Pos logit contributions
ez+0.035
eros+0.033
 wound+0.032
 heaven+0.030
crow+0.030
Neg logit contributions
 topp-0.061
First-0.060
 ju-0.059
 flavor-0.057
 flour-0.056
Token Whenever
Token ID21326
Feature activation-6.83e-02

Pos logit contributions
First+0.012
 topp+0.012
 ju+0.012
 flavor+0.011
 flour+0.011
Neg logit contributions
ez-0.008
eros-0.008
 wound-0.008
--0.007
 heaven-0.007
Token the
Token ID262
Feature activation+2.17e-03

Pos logit contributions
 topp+0.130
First+0.129
 ju+0.127
 flour+0.119
 flavor+0.119
Neg logit contributions
ez-0.066
eros-0.064
 wound-0.061
 tears-0.061
 heaven-0.061
Token a
Token ID257
Feature activation-6.99e-02

Pos logit contributions
 topp+0.055
First+0.054
 ju+0.052
 flavor+0.049
 forks+0.049
Neg logit contributions
ez-0.061
eros-0.059
 wound-0.057
crow-0.056
oring-0.056
Token special
Token ID2041
Feature activation+4.50e-02

Pos logit contributions
 topp+0.108
First+0.106
 ju+0.100
 forks+0.095
 flavor+0.094
Neg logit contributions
 wound-0.091
 heaven-0.091
 hospital-0.090
--0.089
 crying-0.089
Token places
Token ID4113
Feature activation+7.92e-02

Pos logit contributions
eros+0.054
ez+0.054
 wound+0.054
-+0.053
 heaven+0.052
Neg logit contributions
 topp-0.074
First-0.074
 ju-0.071
 forks-0.066
 flour-0.066
Token that
Token ID326
Feature activation+8.31e-02

Pos logit contributions
ez+0.083
eros+0.079
 wound+0.077
 heaven+0.076
 tears+0.075
Neg logit contributions
 topp-0.146
First-0.144
 ju-0.141
 flavor-0.133
 flour-0.132
Token sold
Token ID2702
Feature activation+6.66e-03

Pos logit contributions
ez+0.102
eros+0.098
 wound+0.096
 tears+0.094
 heaven+0.094
Neg logit contributions
 topp-0.137
First-0.135
 ju-0.132
 flavor-0.124
 flour-0.123
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
 its+0.008
-+0.007
 tears+0.007
crow+0.007
 creature+0.007
Neg logit contributions
First-0.012
 topp-0.011
 ju-0.011
'd-0.011
teen-0.010
Token.
Token ID13
Feature activation-4.96e-03

Pos logit contributions
 topp+0.609
First+0.607
 ju+0.592
teen+0.554
 apr+0.553
Neg logit contributions
--0.232
 its-0.219
ez-0.211
eros-0.209
 wound-0.207
Token Alex
Token ID4422
Feature activation-5.56e-02

Pos logit contributions
 topp+0.009
 ju+0.009
First+0.008
 flavor+0.008
 flour+0.008
Neg logit contributions
ez-0.005
eros-0.005
 Start-0.005
 heaven-0.005
 tears-0.005
Token looked
Token ID3114
Feature activation-3.27e-02

Pos logit contributions
 topp+0.092
First+0.091
 ju+0.089
 flavor+0.084
 flour+0.083
Neg logit contributions
ez-0.068
eros-0.066
 wound-0.064
 tears-0.062
 heaven-0.062
Token at
Token ID379
Feature activation+2.23e-03

Pos logit contributions
 ju+0.055
 topp+0.054
First+0.051
 forks+0.050
 flour+0.050
Neg logit contributions
ez-0.043
eros-0.041
crow-0.038
 wound-0.038
ment-0.038
Token all
Token ID477
Feature activation-8.74e-02

Pos logit contributions
ez+0.002
eros+0.002
 wound+0.002
 heaven+0.002
oring+0.002
Neg logit contributions
 topp-0.004
First-0.004
 ju-0.004
 flavor-0.004
 flour-0.004
Token scream
Token ID8196
Feature activation+9.66e-02

Pos logit contributions
 topp+0.097
 ju+0.094
First+0.093
 flavor+0.088
 flour+0.087
Neg logit contributions
ez-0.064
 wound-0.061
 tears-0.060
eros-0.060
 terrified-0.059
Token.
Token ID13
Feature activation+6.06e-03

Pos logit contributions
ez+0.106
eros+0.102
 wound+0.097
 heaven+0.095
 tears+0.094
Neg logit contributions
 topp-0.172
First-0.170
 ju-0.167
 flavor-0.158
 flour-0.157
Token They
Token ID1119
Feature activation-4.33e-02

Pos logit contributions
ez+0.008
eros+0.007
 heaven+0.007
 wound+0.007
oring+0.007
Neg logit contributions
First-0.010
 topp-0.010
 ju-0.009
 flour-0.009
 flavor-0.009
Token drop
Token ID4268
Feature activation+5.77e-02

Pos logit contributions
 topp+0.051
First+0.049
 ju+0.048
 flavor+0.044
 flour+0.043
Neg logit contributions
ez-0.075
 wound-0.073
eros-0.073
 tears-0.072
 terrified-0.071
Token the
Token ID262
Feature activation+1.20e-02

Pos logit contributions
ez+0.062
eros+0.059
oring+0.056
 wound+0.055
 heaven+0.054
Neg logit contributions
 topp-0.106
First-0.103
 ju-0.103
 flour-0.100
 flavor-0.098
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
 wound+0.023
 tears+0.022
-+0.022
eros+0.022
les+0.021
Neg logit contributions
First-0.013
 topp-0.011
 ju-0.010
'd-0.010
Deal-0.010
Token.
Token ID13
Feature activation+6.97e-03

Pos logit contributions
 topp+0.602
First+0.590
 ju+0.589
 flavor+0.555
 flour+0.546
Neg logit contributions
ez-0.243
 wound-0.235
--0.228
eros-0.227
 tears-0.222
Token They
Token ID1119
Feature activation-4.30e-02

Pos logit contributions
ez+0.008
eros+0.007
 wound+0.007
 heaven+0.007
oring+0.007
Neg logit contributions
 topp-0.012
First-0.012
 ju-0.012
 flour-0.011
 flavor-0.011
Token try
Token ID1949
Feature activation-6.13e-02

Pos logit contributions
 topp+0.056
First+0.055
 ju+0.053
 flavor+0.050
 flour+0.049
Neg logit contributions
ez-0.069
 wound-0.067
 tears-0.067
eros-0.066
 terrified-0.065
Token to
Token ID284
Feature activation-4.30e-02

Pos logit contributions
 topp+0.074
First+0.072
 ju+0.068
 flavor+0.068
 flour+0.066
Neg logit contributions
ez-0.104
eros-0.103
crow-0.097
oc-0.097
ment-0.096
Token spring
Token ID6076
Feature activation+2.64e-02

Pos logit contributions
 topp+0.062
First+0.061
 ju+0.059
 flour+0.055
 flavor+0.054
Neg logit contributions
 wound-0.064
ez-0.062
 tears-0.062
eros-0.061
 heaven-0.060
Token cones
Token ID47314
Feature activation-2.92e-01

Pos logit contributions
First+0.126
 ju+0.124
 topp+0.117
Deal+0.115
'd+0.110
Neg logit contributions
--0.114
ez-0.110
 tears-0.107
eros-0.106
 its-0.105
Token.
Token ID13
Feature activation-2.86e-03

Pos logit contributions
 ju+0.652
 topp+0.651
First+0.647
 apr+0.610
teen+0.605
Neg logit contributions
--0.196
 its-0.177
 wound-0.172
 crying-0.171
eros-0.165
Token They
Token ID1119
Feature activation-4.18e-02

Pos logit contributions
 topp+0.005
First+0.005
 ju+0.005
 flavor+0.005
 forks+0.005
Neg logit contributions
ez-0.003
eros-0.003
 wound-0.003
 tears-0.003
 Start-0.003
Token picked
Token ID6497
Feature activation-4.05e-02

Pos logit contributions
 topp+0.080
avored+0.079
umm+0.078
 ju+0.075
First+0.074
Neg logit contributions
 tears-0.042
 realized-0.041
 its-0.041
 cried-0.040
 wound-0.039
Token some
Token ID617
Feature activation-4.59e-02

Pos logit contributions
First+0.080
 topp+0.078
 ju+0.075
 forks+0.073
 flavor+0.071
Neg logit contributions
ez-0.035
 its-0.035
--0.034
 tears-0.034
 wound-0.034
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
First+0.069
 topp+0.067
'd+0.062
Deal+0.062
 ju+0.062
Neg logit contributions
--0.063
 wound-0.056
 tears-0.056
les-0.055
 its-0.055
Token and
Token ID290
Feature activation+3.55e-02

Pos logit contributions
 topp+0.628
First+0.618
 ju+0.609
 flavor+0.577
 flour+0.573
Neg logit contributions
ez-0.207
eros-0.202
 wound-0.196
--0.194
 its-0.187
Token smelled
Token ID37609
Feature activation-1.85e-02

Pos logit contributions
 tears+0.044
 its+0.042
 wound+0.042
-+0.039
ez+0.038
Neg logit contributions
First-0.056
avored-0.056
teen-0.054
ipp-0.053
 topp-0.053
Token them
Token ID606
Feature activation+2.59e-02

Pos logit contributions
First+0.040
 topp+0.038
 ju+0.036
Deal+0.035
 forks+0.035
Neg logit contributions
 its-0.014
 tears-0.014
--0.013
ez-0.013
 wound-0.013
Token.
Token ID13
Feature activation-3.36e-03

Pos logit contributions
 its+0.016
 tears+0.016
-+0.014
 wound+0.014
 crying+0.013
Neg logit contributions
First-0.060
 topp-0.059
 ju-0.057
teen-0.055
 apr-0.055
Token They
Token ID1119
Feature activation-4.26e-02

Pos logit contributions
 topp+0.006
First+0.006
 ju+0.006
 flavor+0.006
 forks+0.006
Neg logit contributions
 wound-0.003
ez-0.003
eros-0.003
 tears-0.003
 Start-0.003
Token looked
Token ID3114
Feature activation-3.27e-02

Pos logit contributions
 topp+0.092
First+0.091
 ju+0.089
 flavor+0.084
 flour+0.083
Neg logit contributions
ez-0.068
eros-0.066
 wound-0.064
 tears-0.062
 heaven-0.062
Token at
Token ID379
Feature activation+2.23e-03

Pos logit contributions
 ju+0.055
 topp+0.054
First+0.051
 forks+0.050
 flour+0.050
Neg logit contributions
ez-0.043
eros-0.041
crow-0.038
 wound-0.038
ment-0.038
Token all
Token ID477
Feature activation-8.74e-02

Pos logit contributions
ez+0.002
eros+0.002
 wound+0.002
 heaven+0.002
oring+0.002
Neg logit contributions
 topp-0.004
First-0.004
 ju-0.004
 flavor-0.004
 flour-0.004
Token of
Token ID286
Feature activation-3.38e-02

Pos logit contributions
 topp+0.145
First+0.143
 ju+0.139
 flour+0.133
 flavor+0.131
Neg logit contributions
ez-0.111
eros-0.106
 wound-0.103
 heaven-0.099
 speaker-0.098
Token the
Token ID262
Feature activation+5.86e-03

Pos logit contributions
First+0.066
 ju+0.065
 topp+0.065
 flavor+0.061
 flour+0.059
Neg logit contributions
ez-0.030
 tears-0.029
 its-0.029
eros-0.028
 wound-0.028
Token cones
Token ID47314
Feature activation-2.93e-01

Pos logit contributions
 tears+0.010
 wound+0.010
-+0.010
oc+0.009
les+0.009
Neg logit contributions
First-0.006
 topp-0.006
 ju-0.006
 shovel-0.006
 First-0.006
Token to
Token ID284
Feature activation-4.00e-02

Pos logit contributions
First+0.621
 topp+0.621
 ju+0.611
 flavor+0.569
 flour+0.566
Neg logit contributions
ez-0.201
 wound-0.200
--0.200
eros-0.198
 crying-0.192
Token find
Token ID1064
Feature activation+1.30e-01

Pos logit contributions
 topp+0.052
First+0.051
 ju+0.050
 flour+0.046
 forks+0.045
Neg logit contributions
ez-0.063
eros-0.061
 wound-0.061
 tears-0.060
 heaven-0.060
Token the
Token ID262
Feature activation+6.97e-03

Pos logit contributions
ez+0.121
-+0.121
 its+0.118
 tears+0.117
eros+0.117
Neg logit contributions
First-0.248
 topp-0.242
 ju-0.235
 flavor-0.222
 forks-0.218
Token perfect
Token ID2818
Feature activation-9.64e-02

Pos logit contributions
 tears+0.010
 wound+0.010
 hospital+0.009
 man+0.008
 knowing+0.008
Neg logit contributions
First-0.010
 shovel-0.009
glers-0.009
 topp-0.009
 forks-0.009
Token one
Token ID530
Feature activation+7.07e-02

Pos logit contributions
First+0.143
 ju+0.126
 topp+0.117
glers+0.117
 forks+0.116
Neg logit contributions
--0.147
 wound-0.144
 tears-0.138
eros-0.129
oc-0.128
Token's
Token ID338
Feature activation-2.61e-02

Pos logit contributions
eros+0.055
ez+0.052
crow+0.050
ment+0.049
 Start+0.049
Neg logit contributions
First-0.063
 topp-0.062
 ju-0.061
 would-0.059
 flavor-0.058
Token daddy
Token ID35822
Feature activation-3.33e-02

Pos logit contributions
First+0.055
 topp+0.053
 ju+0.051
teen+0.047
ven+0.046
Neg logit contributions
 tears-0.028
 wound-0.026
 crying-0.026
eros-0.026
 creature-0.024
Token gave
Token ID2921
Feature activation+2.78e-02

Pos logit contributions
 topp+0.060
First+0.060
 ju+0.059
 flavor+0.055
 flour+0.055
Neg logit contributions
ez-0.037
 wound-0.034
eros-0.034
 crying-0.033
--0.033
Token her
Token ID607
Feature activation+7.90e-02

Pos logit contributions
eros+0.029
ez+0.028
 terrified+0.027
 wound+0.026
 speaker+0.025
Neg logit contributions
 topp-0.051
First-0.051
 ju-0.050
 flour-0.048
 flavor-0.048
Token a
Token ID257
Feature activation-7.20e-02

Pos logit contributions
ez+0.094
eros+0.090
 wound+0.087
 tears+0.085
 heaven+0.085
Neg logit contributions
 topp-0.135
First-0.133
 ju-0.131
 flavor-0.123
 flour-0.122
Token lecture
Token ID19143
Feature activation+1.89e-01

Pos logit contributions
First+0.111
 topp+0.111
 ju+0.104
 forks+0.097
 flour+0.097
Neg logit contributions
 wound-0.094
ez-0.094
 tears-0.092
 crying-0.091
eros-0.090
Token about
Token ID546
Feature activation+8.70e-02

Pos logit contributions
ez+0.179
eros+0.176
 wound+0.164
 terrified+0.158
 tears+0.158
Neg logit contributions
 topp-0.365
First-0.361
 ju-0.349
 flavor-0.341
 flour-0.335
Token taking
Token ID2263
Feature activation+3.49e-03

Pos logit contributions
ez+0.093
eros+0.089
 wound+0.085
 tears+0.082
crow+0.082
Neg logit contributions
 topp-0.159
First-0.156
 ju-0.153
 flavor-0.146
 flour-0.145
Token care
Token ID1337
Feature activation+1.15e-01

Pos logit contributions
eros+0.005
 wound+0.005
oring+0.005
 terrified+0.005
ez+0.004
Neg logit contributions
First-0.005
 topp-0.005
 flour-0.005
 ju-0.005
 flavor-0.005
Token of
Token ID286
Feature activation-3.75e-02

Pos logit contributions
eros+0.185
ez+0.179
 wound+0.175
 terrified+0.174
 tears+0.173
Neg logit contributions
First-0.149
 topp-0.144
 forks-0.136
 flour-0.136
 would-0.135
Token her
Token ID607
Feature activation+8.04e-02

Pos logit contributions
 topp+0.074
First+0.072
 ju+0.071
 flavor+0.069
 flour+0.069
Neg logit contributions
ez-0.035
eros-0.033
oring-0.031
 wound-0.031
 terrified-0.031
Token too
Token ID1165
Feature activation-3.29e-02

Pos logit contributions
eros+0.027
ez+0.027
oring+0.025
 wound+0.024
crow+0.024
Neg logit contributions
 topp-0.039
 ju-0.038
First-0.038
 flavor-0.036
 flour-0.036
Token late
Token ID2739
Feature activation+1.60e-01

Pos logit contributions
First+0.067
 ju+0.065
 topp+0.064
 flour+0.063
 flavor+0.062
Neg logit contributions
ez-0.029
eros-0.029
 wound-0.025
 tears-0.024
 speaker-0.024
Token.
Token ID13
Feature activation-5.38e-03

Pos logit contributions
ez+0.199
eros+0.187
 wound+0.183
crow+0.179
 tears+0.177
Neg logit contributions
 topp-0.261
First-0.253
 ju-0.245
 flavor-0.238
 forks-0.235
Token The
Token ID383
Feature activation+5.31e-02

Pos logit contributions
 topp+0.009
 ju+0.009
First+0.009
 flour+0.009
 First+0.008
Neg logit contributions
ez-0.006
les-0.006
--0.005
 wound-0.005
rows-0.005
Token bad
Token ID2089
Feature activation+9.12e-02

Pos logit contributions
ez+0.061
eros+0.059
 wound+0.058
 tears+0.056
 heaven+0.056
Neg logit contributions
 topp-0.093
First-0.093
 ju-0.090
 flavor-0.085
 flour-0.084
Token thing
Token ID1517
Feature activation+1.84e-01

Pos logit contributions
ez+0.129
oring+0.116
 terrified+0.114
 Start+0.112
 unconscious+0.112
Neg logit contributions
 flavor-0.136
 topp-0.136
 usual-0.128
 flour-0.127
 picnic-0.127
Token had
Token ID550
Feature activation-5.68e-02

Pos logit contributions
ez+0.221
eros+0.212
 heaven+0.203
 wound+0.203
oring+0.201
Neg logit contributions
 topp-0.303
First-0.293
 ju-0.293
 flavor-0.281
 flour-0.276
Token already
Token ID1541
Feature activation+2.40e-02

Pos logit contributions
 topp+0.084
First+0.080
 ju+0.079
 flavor+0.076
 flour+0.074
Neg logit contributions
ez-0.081
eros-0.080
oring-0.077
oc-0.075
 wound-0.075
Token vanished
Token ID23717
Feature activation+8.26e-02

Pos logit contributions
eros+0.037
ez+0.036
oring+0.034
oc+0.034
 heaven+0.033
Neg logit contributions
 topp-0.031
First-0.030
 flavor-0.029
 would-0.029
'd-0.029
Token,
Token ID11
Feature activation-3.45e-02

Pos logit contributions
ez+0.092
eros+0.087
 wound+0.083
 heaven+0.080
crow+0.080
Neg logit contributions
 topp-0.147
First-0.144
 ju-0.144
 flavor-0.136
 flour-0.135
Token taking
Token ID2263
Feature activation+4.75e-03

Pos logit contributions
 ju+0.051
 topp+0.050
 would+0.049
'd+0.047
 flour+0.046
Neg logit contributions
eros-0.049
ez-0.048
ment-0.044
crow-0.044
oring-0.044
Token smiled
Token ID13541
Feature activation+2.24e-02

Pos logit contributions
ez+0.071
eros+0.069
oring+0.066
crow+0.063
 wound+0.062
Neg logit contributions
 topp-0.079
 ju-0.077
First-0.074
 flavor-0.073
 forks-0.072
Token.
Token ID13
Feature activation+5.67e-03

Pos logit contributions
ez+0.030
eros+0.028
crow+0.027
 wound+0.026
 speaker+0.026
Neg logit contributions
 ju-0.036
 topp-0.035
First-0.034
 flavor-0.033
 forks-0.032
Token She
Token ID1375
Feature activation+6.59e-03

Pos logit contributions
ez+0.006
eros+0.005
 wound+0.005
 tears+0.005
 heaven+0.005
Neg logit contributions
 topp-0.011
First-0.011
 ju-0.010
 flavor-0.010
 flour-0.010
Token knew
Token ID2993
Feature activation+1.11e-01

Pos logit contributions
ez+0.009
eros+0.009
 heaven+0.008
crow+0.008
oring+0.008
Neg logit contributions
 topp-0.010
 ju-0.010
First-0.010
 flavor-0.009
 flour-0.009
Token the
Token ID262
Feature activation+1.09e-02

Pos logit contributions
ez+0.096
eros+0.089
 wound+0.085
oring+0.082
 heaven+0.081
Neg logit contributions
 topp-0.225
First-0.222
 ju-0.220
 flavor-0.211
 flour-0.210
Token wish
Token ID4601
Feature activation+1.96e-01

Pos logit contributions
ez+0.013
eros+0.013
 wound+0.013
 tears+0.013
 heaven+0.012
Neg logit contributions
First-0.018
 topp-0.018
 ju-0.017
 flavor-0.016
 flour-0.016
Token was
Token ID373
Feature activation+3.38e-02

Pos logit contributions
ez+0.334
eros+0.325
 wound+0.316
 heaven+0.310
 tears+0.310
Neg logit contributions
 topp-0.230
First-0.223
 ju-0.218
 flavor-0.203
 flour-0.199
Token safe
Token ID3338
Feature activation+4.66e-02

Pos logit contributions
ez+0.040
eros+0.038
 wound+0.037
oring+0.035
 tears+0.035
Neg logit contributions
 topp-0.058
First-0.057
 ju-0.056
 flavor-0.053
 flour-0.053
Token inside
Token ID2641
Feature activation+6.53e-02

Pos logit contributions
ez+0.035
eros+0.032
 wound+0.032
 heaven+0.030
 tears+0.030
Neg logit contributions
 topp-0.100
First-0.099
 ju-0.097
 flavor-0.092
 flour-0.092
Token until
Token ID1566
Feature activation-2.47e-02

Pos logit contributions
ez+0.061
eros+0.057
 wound+0.056
 heaven+0.055
 tears+0.055
Neg logit contributions
 topp-0.127
First-0.126
 ju-0.123
 flavor-0.116
 flour-0.116
Token she
Token ID673
Feature activation+4.91e-02

Pos logit contributions
 topp+0.049
First+0.048
 ju+0.048
 flavor+0.045
 flour+0.045
Neg logit contributions
ez-0.023
eros-0.022
 wound-0.021
 tears-0.020
 heaven-0.020
Token creature
Token ID7185
Feature activation+1.95e-01

Pos logit contributions
ez+0.033
 terrified+0.030
eros+0.030
oring+0.029
 devastated+0.029
Neg logit contributions
 topp-0.052
 flavor-0.050
 ju-0.050
First-0.048
 flour-0.048
Token.
Token ID13
Feature activation-4.92e-03

Pos logit contributions
ez+0.196
eros+0.186
 wound+0.179
 tears+0.175
 heaven+0.173
Neg logit contributions
 topp-0.365
First-0.360
 ju-0.355
 flavor-0.338
 flour-0.335
Token
Token ID198
Feature activation+1.90e-02

Pos logit contributions
 topp+0.010
First+0.010
 ju+0.010
 flavor+0.010
 flour+0.010
Neg logit contributions
ez-0.004
eros-0.004
 wound-0.003
 tears-0.003
 terrified-0.003
Token
Token ID198
Feature activation+1.82e-02

Pos logit contributions
ez+0.015
eros+0.014
 wound+0.014
 tears+0.013
 heaven+0.013
Neg logit contributions
 topp-0.039
First-0.039
 ju-0.038
 flavor-0.037
 flour-0.036
TokenThe
Token ID464
Feature activation+2.69e-02

Pos logit contributions
Start+0.022
crow+0.022
 Start+0.021
 speaker+0.021
 heaven+0.021
Neg logit contributions
 topp-0.032
 forks-0.030
 ju-0.028
 shovel-0.028
 flavor-0.027
Token woman
Token ID2415
Feature activation+2.46e-01

Pos logit contributions
ez+0.035
ulk+0.030
oring+0.029
ment+0.028
rows+0.027
Neg logit contributions
 flour-0.047
 flavor-0.046
 picnic-0.046
 would-0.045
 topp-0.044
Token took
Token ID1718
Feature activation-4.04e-02

Pos logit contributions
ez+0.286
eros+0.272
 wound+0.270
-+0.260
 tears+0.258
Neg logit contributions
 topp-0.424
First-0.423
 ju-0.414
 flavor-0.387
 flour-0.386
Token her
Token ID607
Feature activation+7.82e-02

Pos logit contributions
First+0.060
 topp+0.060
 ju+0.060
 flour+0.059
 flavor+0.057
Neg logit contributions
eros-0.053
oring-0.052
 wound-0.052
ez-0.051
rows-0.050
Token furry
Token ID40857
Feature activation+2.79e-02

Pos logit contributions
ez+0.096
eros+0.094
 wound+0.093
 tears+0.090
-+0.090
Neg logit contributions
 topp-0.130
First-0.128
 ju-0.123
 flavor-0.117
 flour-0.116
Token friend
Token ID1545
Feature activation+2.95e-02

Pos logit contributions
ez+0.035
eros+0.032
 wound+0.031
oring+0.031
 terrified+0.031
Neg logit contributions
 topp-0.047
First-0.045
 ju-0.045
 flavor-0.043
 forks-0.042
Token to
Token ID284
Feature activation-4.21e-02

Pos logit contributions
ez+0.028
eros+0.026
-+0.026
 heaven+0.026
 crying+0.026
Neg logit contributions
 topp-0.058
First-0.057
 ju-0.056
 flavor-0.053
 flour-0.052
Token to
Token ID284
Feature activation-4.18e-02

Pos logit contributions
First+0.083
 topp+0.082
 ju+0.080
 flavor+0.080
 flour+0.079
Neg logit contributions
eros-0.076
ez-0.072
oc-0.070
crow-0.068
oring-0.066
Token go
Token ID467
Feature activation-1.44e-01

Pos logit contributions
 topp+0.061
First+0.059
 ju+0.059
 flour+0.054
 flavor+0.053
Neg logit contributions
 wound-0.061
ez-0.060
 tears-0.059
 heaven-0.058
eros-0.058
Token fishing
Token ID12478
Feature activation-5.29e-02

Pos logit contributions
First+0.241
 ju+0.238
 picnic+0.228
 stack+0.228
 shovel+0.226
Neg logit contributions
ez-0.183
eros-0.155
oc-0.150
ller-0.148
 tears-0.147
Token again
Token ID757
Feature activation+5.85e-02

Pos logit contributions
 topp+0.105
First+0.103
 ju+0.102
 flavor+0.097
 flour+0.096
Neg logit contributions
ez-0.048
eros-0.045
 wound-0.044
 tears-0.043
 heaven-0.042
Token,
Token ID11
Feature activation-3.02e-02

Pos logit contributions
ez+0.052
-+0.051
 wound+0.050
 heaven+0.049
 its+0.049
Neg logit contributions
 topp-0.118
First-0.114
 ju-0.113
 flour-0.106
 forks-0.106
Token despite
Token ID3805
Feature activation+1.81e-01

Pos logit contributions
 ju+0.038
First+0.036
 topp+0.035
 forks+0.033
 stack+0.032
Neg logit contributions
eros-0.052
ez-0.051
crow-0.047
oc-0.047
rows-0.047
Token her
Token ID607
Feature activation+8.01e-02

Pos logit contributions
ez+0.189
eros+0.180
oring+0.176
oc+0.172
 wound+0.170
Neg logit contributions
 topp-0.332
First-0.323
 ju-0.323
 flavor-0.311
 flour-0.311
Token husband
Token ID5229
Feature activation+1.14e-01

Pos logit contributions
ez+0.115
eros+0.111
 wound+0.107
 heaven+0.105
 tears+0.104
Neg logit contributions
 topp-0.118
First-0.115
 ju-0.113
 flavor-0.106
 flour-0.105
Token's
Token ID338
Feature activation-2.81e-02

Pos logit contributions
ez+0.124
eros+0.118
 wound+0.114
 tears+0.111
 heaven+0.111
Neg logit contributions
 topp-0.207
First-0.205
 ju-0.201
 flavor-0.190
 flour-0.189
Token warnings
Token ID14601
Feature activation+1.24e-01

Pos logit contributions
 topp+0.050
First+0.049
 ju+0.048
 flavor+0.046
 flour+0.046
Neg logit contributions
ez-0.031
eros-0.030
 heaven-0.027
oring-0.027
 wound-0.027
Token.
Token ID13
Feature activation+4.44e-04

Pos logit contributions
ez+0.111
 wound+0.102
eros+0.102
-+0.101
 tears+0.101
Neg logit contributions
 topp-0.246
First-0.244
 ju-0.241
 flavor-0.227
 flour-0.226
Token They
Token ID1119
Feature activation-4.02e-02

Pos logit contributions
ez+0.008
eros+0.008
 wound+0.008
 tears+0.008
 heaven+0.008
Neg logit contributions
 topp-0.011
First-0.011
 ju-0.011
 flavor-0.010
 flour-0.010
Token smiled
Token ID13541
Feature activation+2.34e-02

Pos logit contributions
 topp+0.064
First+0.062
 ju+0.061
 flour+0.057
 flavor+0.057
Neg logit contributions
ez-0.053
 wound-0.051
 tears-0.050
eros-0.049
 heaven-0.048
Token.
Token ID13
Feature activation+4.24e-03

Pos logit contributions
ez+0.031
eros+0.030
crow+0.029
 speaker+0.027
 wound+0.026
Neg logit contributions
 ju-0.036
First-0.035
 topp-0.035
 would-0.034
 forks-0.033
Token Their
Token ID5334
Feature activation+4.87e-02

Pos logit contributions
ez+0.005
eros+0.005
 wound+0.005
 tears+0.005
 heaven+0.005
Neg logit contributions
 topp-0.007
First-0.007
 ju-0.007
 flavor-0.007
 flour-0.006
Token moms
Token ID34015
Feature activation-1.23e-01

Pos logit contributions
 tears+0.061
 wound+0.059
eros+0.059
 crying+0.059
ez+0.058
Neg logit contributions
First-0.084
 ju-0.081
 topp-0.081
 flavor-0.073
 flour-0.072
Token and
Token ID290
Feature activation+4.28e-02

Pos logit contributions
 topp+0.217
First+0.215
 ju+0.211
 flavor+0.197
 flour+0.197
Neg logit contributions
ez-0.141
eros-0.132
 wound-0.132
 tears-0.129
 heaven-0.127
Token dads
Token ID47713
Feature activation-1.45e-01

Pos logit contributions
ez+0.062
eros+0.060
 wound+0.059
 tears+0.058
 heaven+0.057
Neg logit contributions
 topp-0.062
First-0.061
 ju-0.059
 flavor-0.055
 flour-0.055
Token smiled
Token ID13541
Feature activation+2.29e-02

Pos logit contributions
 topp+0.233
First+0.230
 ju+0.226
 flour+0.212
 flavor+0.209
Neg logit contributions
ez-0.186
 wound-0.176
 tears-0.174
eros-0.173
 crying-0.172
Token,
Token ID11
Feature activation-2.51e-02

Pos logit contributions
ez+0.030
eros+0.029
crow+0.027
 wound+0.027
 speaker+0.026
Neg logit contributions
 topp-0.035
 ju-0.035
First-0.035
 flavor-0.033
 forks-0.033
Token too
Token ID1165
Feature activation-3.08e-02

Pos logit contributions
 ju+0.036
 would+0.033
First+0.032
 yes+0.032
 topp+0.032
Neg logit contributions
eros-0.042
ez-0.039
crow-0.038
oc-0.037
ore-0.036
Token.
Token ID13
Feature activation+3.58e-03

Pos logit contributions
 topp+0.047
First+0.045
 ju+0.044
 flavor+0.043
 flour+0.042
Neg logit contributions
ez-0.043
eros-0.040
crow-0.038
 wound-0.038
 heaven-0.037
Token It
Token ID632
Feature activation+1.96e-02

Pos logit contributions
First+0.020
 topp+0.020
 ju+0.020
 flour+0.018
 forks+0.018
Neg logit contributions
ez-0.011
eros-0.010
 wound-0.010
--0.010
 tears-0.009
Token wanted
Token ID2227
Feature activation-5.32e-02

Pos logit contributions
eros+0.034
oring+0.031
rows+0.031
ment+0.030
ez+0.030
Neg logit contributions
 would-0.025
 ju-0.024
'd-0.024
 topp-0.022
First-0.022
Token to
Token ID284
Feature activation-4.49e-02

Pos logit contributions
 topp+0.084
First+0.084
 ju+0.082
 flavor+0.076
 flour+0.076
Neg logit contributions
ez-0.069
eros-0.066
 wound-0.065
 tears-0.064
 heaven-0.063
Token be
Token ID307
Feature activation+2.61e-04

Pos logit contributions
 topp+0.060
First+0.059
 ju+0.058
 flour+0.053
 forks+0.052
Neg logit contributions
ez-0.070
 wound-0.068
eros-0.068
 heaven-0.067
 tears-0.067
Token free
Token ID1479
Feature activation-2.36e-02

Pos logit contributions
oring+0.000
eros+0.000
ez+0.000
crow+0.000
oc+0.000
Neg logit contributions
 ju-0.000
 topp-0.000
 flavor-0.000
 shovel-0.000
 flour-0.000
Token and
Token ID290
Feature activation+3.33e-02

Pos logit contributions
 topp+0.053
First+0.052
 ju+0.050
 flavor+0.049
 apr+0.048
Neg logit contributions
--0.016
 its-0.016
ez-0.016
 heaven-0.015
 crying-0.015
Token explore
Token ID7301
Feature activation+6.56e-02

Pos logit contributions
ez+0.046
eros+0.045
 wound+0.044
 tears+0.043
 heaven+0.043
Neg logit contributions
 topp-0.050
First-0.049
 ju-0.048
 flavor-0.045
 flour-0.044
Token,
Token ID11
Feature activation-3.55e-02

Pos logit contributions
ez+0.072
eros+0.069
 wound+0.064
oring+0.063
 terrified+0.062
Neg logit contributions
 topp-0.118
First-0.116
 ju-0.116
 flavor-0.111
 flour-0.110
Token so
Token ID523
Feature activation-1.70e-01

Pos logit contributions
 ju+0.053
First+0.052
 topp+0.052
 flavor+0.048
 flour+0.047
Neg logit contributions
ez-0.051
eros-0.050
crow-0.046
oring-0.045
 wound-0.045
Token it
Token ID340
Feature activation+9.81e-02

Pos logit contributions
 topp+0.350
First+0.347
 ju+0.342
 flavor+0.328
 flour+0.325
Neg logit contributions
ez-0.141
eros-0.131
 wound-0.124
 heaven-0.119
 tears-0.119
Token started
Token ID2067
Feature activation-5.56e-02

Pos logit contributions
eros+0.164
ez+0.160
 speaker+0.151
oring+0.150
 heaven+0.150
Neg logit contributions
 ju-0.124
 topp-0.117
First-0.116
 would-0.114
'd-0.111
Token just
Token ID655
Feature activation+6.57e-02

Pos logit contributions
ez+0.098
 wound+0.097
 tears+0.096
 crying+0.094
 its+0.092
Neg logit contributions
First-0.135
 topp-0.132
 ju-0.129
teen-0.118
 forks-0.117
Token an
Token ID281
Feature activation-4.65e-03

Pos logit contributions
ez+0.085
 tears+0.079
 wound+0.078
 its+0.078
 heaven+0.078
Neg logit contributions
First-0.108
 topp-0.106
 ju-0.103
 flour-0.097
 flavor-0.096
Token ordinary
Token ID8850
Feature activation+1.50e-02

Pos logit contributions
First+0.009
 topp+0.008
 ju+0.008
 forks+0.008
avored+0.007
Neg logit contributions
--0.005
ez-0.005
 hospital-0.005
 creature-0.005
 wound-0.005
Token pitch
Token ID7078
Feature activation-1.72e-02

Pos logit contributions
ez+0.019
 wound+0.019
eros+0.019
-+0.019
 creature+0.019
Neg logit contributions
First-0.024
 topp-0.023
 ju-0.023
 forks-0.021
 flour-0.021
Token.
Token ID13
Feature activation-3.84e-03

Pos logit contributions
First+0.034
 topp+0.033
 ju+0.032
 forks+0.030
 flour+0.030
Neg logit contributions
ez-0.016
--0.015
eros-0.015
 wound-0.015
crow-0.015
Token You
Token ID921
Feature activation+2.95e-02

Pos logit contributions
 topp+0.008
 ju+0.007
First+0.007
 flavor+0.007
 flour+0.007
Neg logit contributions
ez-0.004
eros-0.003
 heaven-0.003
 Start-0.003
crow-0.003
Token can
Token ID460
Feature activation-1.35e-01

Pos logit contributions
ez+0.053
eros+0.051
 wound+0.051
 tears+0.050
 heaven+0.050
Neg logit contributions
 topp-0.032
First-0.031
 ju-0.030
 flour-0.028
 flavor-0.028
Token go
Token ID467
Feature activation-1.53e-01

Pos logit contributions
 topp+0.204
 ju+0.194
First+0.191
 forks+0.181
 flour+0.180
Neg logit contributions
ez-0.187
 wound-0.184
eros-0.181
 crying-0.178
 its-0.177
Token and
Token ID290
Feature activation+3.45e-02

Pos logit contributions
First+0.251
 topp+0.246
 ju+0.243
 flavor+0.232
 flour+0.228
Neg logit contributions
ez-0.192
eros-0.181
 tears-0.176
 wound-0.172
 heaven-0.169
Token play
Token ID711
Feature activation-3.05e-02

Pos logit contributions
 wound+0.052
ez+0.052
 heaven+0.051
 crying+0.050
eros+0.050
Neg logit contributions
 topp-0.047
First-0.045
 ju-0.042
 flour-0.041
 flavor-0.040
Token here
Token ID994
Feature activation-5.54e-02

Pos logit contributions
 topp+0.055
First+0.054
 ju+0.053
 flavor+0.051
 flour+0.050
Neg logit contributions
ez-0.033
eros-0.032
 tears-0.030
 wound-0.030
ment-0.030
Token They
Token ID1119
Feature activation-3.92e-02

Pos logit contributions
 topp+0.001
 ju+0.001
First+0.001
 flavor+0.001
'd+0.001
Neg logit contributions
 tears-0.001
ez-0.001
eros-0.001
 Start-0.001
 wound-0.001
Token lit
Token ID6578
Feature activation-4.75e-04

Pos logit contributions
First+0.067
 topp+0.066
 ju+0.066
avored+0.064
umm+0.061
Neg logit contributions
 wound-0.046
 tears-0.044
 lim-0.041
 crying-0.041
 speaker-0.041
Token them
Token ID606
Feature activation+2.81e-02

Pos logit contributions
 topp+0.001
First+0.001
 ju+0.001
 flour+0.001
 flavor+0.001
Neg logit contributions
ez-0.000
eros-0.000
 wound-0.000
oring-0.000
 heaven-0.000
Token with
Token ID351
Feature activation+1.35e-01

Pos logit contributions
 wound+0.014
-+0.013
 tears+0.013
ez+0.013
 its+0.012
Neg logit contributions
First-0.067
 topp-0.066
 ju-0.066
 flavor-0.062
 forks-0.061
Token a
Token ID257
Feature activation-6.84e-02

Pos logit contributions
 tears+0.178
 its+0.174
ez+0.160
 speaker+0.157
 crying+0.155
Neg logit contributions
First-0.233
 ju-0.224
Deal-0.217
ipp-0.211
 topp-0.210
Token match
Token ID2872
Feature activation+3.27e-02

Pos logit contributions
First+0.100
 topp+0.095
 ju+0.093
 flavor+0.085
 forks+0.084
Neg logit contributions
 wound-0.099
 speaker-0.098
 tears-0.097
ez-0.096
 crying-0.095
Token.
Token ID13
Feature activation-8.90e-04

Pos logit contributions
-+0.029
eros+0.029
 wound+0.029
ez+0.029
 speaker+0.027
Neg logit contributions
First-0.067
 ju-0.065
 topp-0.063
 flavor-0.058
 forks-0.058
Token They
Token ID1119
Feature activation-4.02e-02

Pos logit contributions
 topp+0.002
 ju+0.002
First+0.001
 flavor+0.001
 forks+0.001
Neg logit contributions
ez-0.001
 tears-0.001
eros-0.001
 wound-0.001
 Start-0.001
Token thought
Token ID1807
Feature activation+5.14e-02

Pos logit contributions
First+0.065
 ju+0.064
 topp+0.063
avored+0.058
 flavor+0.057
Neg logit contributions
 wound-0.050
 tears-0.049
ez-0.048
 crying-0.046
 its-0.046
Token it
Token ID340
Feature activation+1.06e-01

Pos logit contributions
 its+0.033
 tears+0.031
-+0.028
ez+0.027
 wound+0.027
Neg logit contributions
First-0.121
 ju-0.118
 topp-0.114
Deal-0.108
avored-0.107
Token would
Token ID561
Feature activation-1.29e-01

Pos logit contributions
ez+0.082
 wound+0.082
-+0.081
 tears+0.076
 creature+0.072
Neg logit contributions
First-0.224
 ju-0.218
 topp-0.218
 forks-0.211
 flour-0.205
Token learned
Token ID4499
Feature activation+1.14e-01

Pos logit contributions
ez+0.006
eros+0.005
 wound+0.005
 tears+0.005
 heaven+0.005
Neg logit contributions
 topp-0.006
First-0.005
 ju-0.005
 flour-0.005
 flavor-0.005
Token at
Token ID379
Feature activation-4.56e-03

Pos logit contributions
ez+0.123
eros+0.118
 wound+0.111
 terrified+0.108
 tears+0.108
Neg logit contributions
 topp-0.209
First-0.206
 ju-0.202
 flavor-0.194
 flour-0.191
Token school
Token ID1524
Feature activation+1.57e-01

Pos logit contributions
 topp+0.008
First+0.007
 ju+0.007
 flavor+0.007
 flour+0.007
Neg logit contributions
ez-0.006
eros-0.005
 wound-0.005
 tears-0.005
 heaven-0.005
Token.
Token ID13
Feature activation-3.90e-03

Pos logit contributions
ez+0.164
eros+0.155
 wound+0.150
 heaven+0.147
 tears+0.146
Neg logit contributions
 topp-0.289
First-0.285
 ju-0.282
 flavor-0.266
 flour-0.265
Token He
Token ID679
Feature activation-3.05e-02

Pos logit contributions
 ju+0.006
 topp+0.006
First+0.006
 flour+0.006
 flavor+0.006
Neg logit contributions
ez-0.005
 Start-0.005
 heaven-0.004
 its-0.004
 tears-0.004
Token hum
Token ID1311
Feature activation+1.03e-01

Pos logit contributions
 topp+0.040
First+0.039
 ju+0.039
 flour+0.035
 flavor+0.035
Neg logit contributions
ez-0.049
 wound-0.046
eros-0.046
 tears-0.045
--0.045
Tokenmed
Token ID1150
Feature activation-6.89e-02

Pos logit contributions
eros+0.093
 wound+0.092
ez+0.090
 heaven+0.089
 tears+0.087
Neg logit contributions
First-0.200
 topp-0.199
 ju-0.190
 flavor-0.185
 flour-0.182
Token one
Token ID530
Feature activation+6.76e-02

Pos logit contributions
 topp+0.122
First+0.120
 ju+0.116
 flavor+0.112
 forks+0.111
Neg logit contributions
ez-0.075
eros-0.074
 wound-0.071
crow-0.068
 heaven-0.068
Token of
Token ID286
Feature activation-3.59e-02

Pos logit contributions
ez+0.089
eros+0.086
 wound+0.083
 heaven+0.082
 tears+0.081
Neg logit contributions
 topp-0.106
First-0.103
 ju-0.102
 flavor-0.097
 flour-0.095
Token them
Token ID606
Feature activation+2.38e-02

Pos logit contributions
 topp+0.072
First+0.068
 flavor+0.066
 ju+0.066
 flour+0.065
Neg logit contributions
oring-0.033
eros-0.033
ez-0.031
 wound-0.030
 terrified-0.030
Token as
Token ID355
Feature activation-1.71e-02

Pos logit contributions
ez+0.027
eros+0.025
 wound+0.025
 tears+0.024
 heaven+0.024
Neg logit contributions
 topp-0.042
First-0.042
 ju-0.041
 flour-0.038
 flavor-0.038
Token show
Token ID905
Feature activation+1.48e-01

Pos logit contributions
First+0.062
 ju+0.060
 topp+0.059
Deal+0.054
 flour+0.053
Neg logit contributions
 its-0.042
 tears-0.042
 wound-0.041
 crying-0.041
 heaven-0.041
Token it
Token ID340
Feature activation+1.10e-01

Pos logit contributions
ez+0.152
eros+0.148
oc+0.139
oring+0.138
 wound+0.137
Neg logit contributions
 topp-0.267
First-0.266
 ju-0.258
 flavor-0.253
 flour-0.252
Token to
Token ID284
Feature activation-3.47e-02

Pos logit contributions
ez+0.141
eros+0.138
 wound+0.129
crow+0.126
 terrified+0.125
Neg logit contributions
 topp-0.178
First-0.177
 ju-0.173
 flavor-0.164
 flour-0.162
Token her
Token ID607
Feature activation+9.27e-02

Pos logit contributions
 topp+0.065
First+0.064
 ju+0.063
 flavor+0.061
 forks+0.060
Neg logit contributions
ez-0.035
eros-0.033
 wound-0.031
oring-0.031
 terrified-0.030
Token friends
Token ID2460
Feature activation-7.40e-02

Pos logit contributions
 wound+0.104
eros+0.104
ez+0.102
 tears+0.102
 heaven+0.100
Neg logit contributions
First-0.165
 topp-0.162
 ju-0.157
 flour-0.145
 flavor-0.143
Token.
Token ID13
Feature activation-3.73e-03

Pos logit contributions
 topp+0.152
First+0.148
 ju+0.147
 flour+0.139
 flavor+0.138
Neg logit contributions
ez-0.062
 heaven-0.059
 wound-0.058
 crying-0.057
--0.057
Token<|endoftext|>
Token ID50256
Feature activation+4.26e-02

Pos logit contributions
First+0.006
 ju+0.006
 topp+0.006
 flavor+0.005
 First+0.005
Neg logit contributions
ez-0.005
eros-0.005
 wound-0.004
 tears-0.004
--0.004
TokenOnce
Token ID7454
Feature activation-1.17e-01

Pos logit contributions
ez+0.036
eros+0.033
 wound+0.032
 tears+0.031
 heaven+0.031
Neg logit contributions
 topp-0.087
First-0.086
 ju-0.085
 flavor-0.081
 flour-0.080
Token upon
Token ID2402
Feature activation+6.82e-02

Pos logit contributions
First+0.140
 topp+0.135
 ju+0.133
 forks+0.129
 usual+0.128
Neg logit contributions
ez-0.200
eros-0.186
 wound-0.182
 heaven-0.181
 speaker-0.181
Token a
Token ID257
Feature activation-7.72e-02

Pos logit contributions
ez+0.084
eros+0.082
 wound+0.078
 tears+0.076
crow+0.076
Neg logit contributions
 topp-0.113
First-0.112
 ju-0.109
 flavor-0.103
 flour-0.102
Token time
Token ID640
Feature activation-6.97e-02

Pos logit contributions
 ju+0.130
 topp+0.127
nuts+0.126
First+0.119
 forks+0.118
Neg logit contributions
--0.100
ez-0.098
oc-0.091
 heaven-0.090
 tears-0.089
Token hair
Token ID4190
Feature activation+4.25e-03

Pos logit contributions
ez+0.035
eros+0.033
-+0.032
 wound+0.032
 heaven+0.032
Neg logit contributions
First-0.039
 topp-0.039
 ju-0.039
 flavor-0.035
 flour-0.034
Token is
Token ID318
Feature activation+7.23e-02

Pos logit contributions
ez+0.005
eros+0.005
 wound+0.005
-+0.005
 heaven+0.005
Neg logit contributions
 topp-0.007
First-0.007
 ju-0.007
 flavor-0.006
 flour-0.006
Token boring
Token ID14262
Feature activation+2.52e-03

Pos logit contributions
ez+0.092
 its+0.091
 wound+0.090
 tears+0.090
 heaven+0.088
Neg logit contributions
First-0.115
 topp-0.113
 ju-0.107
 forks-0.101
 flavor-0.101
Token and
Token ID290
Feature activation+4.66e-02

Pos logit contributions
-+0.002
ez+0.002
 wound+0.002
 heaven+0.002
eros+0.001
Neg logit contributions
 topp-0.006
First-0.005
 ju-0.005
 flavor-0.005
 flour-0.005
Token plain
Token ID8631
Feature activation+5.14e-02

Pos logit contributions
ez+0.067
 wound+0.064
 tears+0.063
eros+0.063
 heaven+0.063
Neg logit contributions
First-0.067
 topp-0.066
 ju-0.063
 flavor-0.060
 flour-0.059
Token."
Token ID526
Feature activation-3.36e-02

Pos logit contributions
-+0.044
ez+0.043
 wound+0.041
 heaven+0.040
 tears+0.040
Neg logit contributions
First-0.104
 topp-0.103
 ju-0.102
 flour-0.094
 forks-0.093
Token
Token ID198
Feature activation+2.47e-02

Pos logit contributions
First+0.068
 topp+0.067
 ju+0.066
 flavor+0.063
 flour+0.063
Neg logit contributions
ez-0.030
eros-0.028
 wound-0.027
 tears-0.026
 terrified-0.025
Token
Token ID198
Feature activation+2.39e-02

Pos logit contributions
ez+0.028
eros+0.028
 terrified+0.026
crow+0.026
 creature+0.025
Neg logit contributions
First-0.043
 topp-0.040
 flour-0.039
 flavor-0.038
 ju-0.038
TokenAnna
Token ID31160
Feature activation+7.68e-02

Pos logit contributions
ez+0.032
eros+0.031
 wound+0.030
 tears+0.029
 heaven+0.029
Neg logit contributions
 topp-0.038
First-0.037
 ju-0.036
 flavor-0.034
 flour-0.034
Token says
Token ID1139
Feature activation+6.66e-02

Pos logit contributions
ez+0.093
eros+0.090
 wound+0.086
 heaven+0.084
 tears+0.083
Neg logit contributions
 topp-0.129
First-0.127
 ju-0.124
 flavor-0.118
 flour-0.117
Token,
Token ID11
Feature activation-2.49e-02

Pos logit contributions
-+0.057
 its+0.053
 crying+0.050
ment+0.046
ez+0.045
Neg logit contributions
First-0.151
 ju-0.147
 topp-0.147
 forks-0.133
 flavor-0.132
Token basketball
Token ID9669
Feature activation-7.98e-02

Pos logit contributions
 topp+0.057
First+0.055
 ju+0.054
 flavor+0.052
 flour+0.052
Neg logit contributions
ez-0.040
ment-0.037
eros-0.037
 terrified-0.036
crow-0.036
Token.
Token ID13
Feature activation-8.39e-03

Pos logit contributions
 topp+0.177
 ju+0.173
First+0.173
 flavor+0.164
 flour+0.164
Neg logit contributions
ez-0.053
--0.052
 wound-0.047
eros-0.047
oring-0.045
Token He
Token ID679
Feature activation-4.26e-02

Pos logit contributions
 ju+0.022
First+0.021
 topp+0.021
nuts+0.020
 forks+0.020
Neg logit contributions
 its-0.002
 It-0.002
 his-0.002
 Start-0.002
 The-0.001
Token had
Token ID550
Feature activation-5.48e-02

Pos logit contributions
 topp+0.071
 ju+0.068
First+0.066
 flour+0.064
nuts+0.063
Neg logit contributions
--0.054
 wound-0.052
ez-0.051
 its-0.050
 crying-0.048
Token a
Token ID257
Feature activation-7.40e-02

Pos logit contributions
 topp+0.117
teen+0.112
avored+0.112
 ju+0.110
First+0.109
Neg logit contributions
 its-0.044
 tears-0.040
 his-0.038
 wound-0.035
 someone-0.033
Token big
Token ID1263
Feature activation-9.95e-02

Pos logit contributions
First+0.122
 topp+0.119
 ju+0.110
nuts+0.108
teen+0.106
Neg logit contributions
 wound-0.098
 tears-0.092
 crying-0.092
--0.090
 creature-0.088
Token,
Token ID11
Feature activation-3.73e-02

Pos logit contributions
First+0.141
 topp+0.138
 ju+0.131
 flour+0.119
 forks+0.116
Neg logit contributions
 wound-0.151
 tears-0.150
 crying-0.144
--0.144
eros-0.141
Token hard
Token ID1327
Feature activation-6.29e-02

Pos logit contributions
First+0.051
 topp+0.051
teen+0.044
 forks+0.044
 flavor+0.044
Neg logit contributions
 wound-0.055
 tears-0.055
 its-0.053
 crying-0.053
--0.052
Token basketball
Token ID9669
Feature activation-8.01e-02

Pos logit contributions
 topp+0.076
First+0.076
 ju+0.074
 flour+0.067
 forks+0.065
Neg logit contributions
 wound-0.103
ez-0.102
--0.102
 tears-0.101
eros-0.100
Token that
Token ID326
Feature activation+8.08e-02

Pos logit contributions
First+0.172
 topp+0.169
 ju+0.168
 flour+0.158
 flavor+0.155
Neg logit contributions
--0.063
eros-0.058
 wound-0.057
ez-0.057
crow-0.054
Token he
Token ID339
Feature activation-1.97e-03

Pos logit contributions
 its+0.052
 his+0.049
 wound+0.046
-+0.045
ez+0.043
Neg logit contributions
First-0.181
 topp-0.176
 ju-0.168
teen-0.166
 flour-0.164
Token medicine
Token ID9007
Feature activation+1.12e-01

Pos logit contributions
First+0.074
 topp+0.070
 ju+0.070
 flavor+0.062
teen+0.062
Neg logit contributions
ez-0.061
--0.060
 heaven-0.060
 wound-0.059
 tears-0.059
Token.
Token ID13
Feature activation+1.76e-03

Pos logit contributions
ez+0.102
-+0.098
eros+0.097
 heaven+0.096
 wound+0.096
Neg logit contributions
 topp-0.217
First-0.216
 ju-0.210
 flavor-0.199
 flour-0.197
Token He
Token ID679
Feature activation-3.61e-02

Pos logit contributions
ez+0.002
eros+0.002
 wound+0.002
 heaven+0.002
 tears+0.002
Neg logit contributions
 topp-0.003
First-0.003
 ju-0.003
 flavor-0.003
 flour-0.003
Token will
Token ID481
Feature activation-7.22e-02

Pos logit contributions
 topp+0.053
First+0.052
 ju+0.051
 flavor+0.048
 flour+0.047
Neg logit contributions
ez-0.051
eros-0.049
 wound-0.048
 tears-0.047
 heaven-0.047
Token feel
Token ID1254
Feature activation+1.19e-01

Pos logit contributions
 topp+0.100
First+0.098
 ju+0.096
 flavor+0.089
 flour+0.088
Neg logit contributions
ez-0.109
eros-0.105
 wound-0.102
 tears-0.100
 heaven-0.100
Token better
Token ID1365
Feature activation+1.31e-02

Pos logit contributions
eros+0.166
ez+0.163
oring+0.153
 speaker+0.151
 wound+0.151
Neg logit contributions
 topp-0.177
 ju-0.176
First-0.172
 flour-0.163
 flavor-0.161
Token soon
Token ID2582
Feature activation+2.63e-02

Pos logit contributions
ez+0.016
eros+0.015
 tears+0.014
crow+0.014
 wound+0.014
Neg logit contributions
First-0.022
 topp-0.022
 ju-0.022
 flavor-0.021
 flour-0.020
Token."
Token ID526
Feature activation-5.27e-02

Pos logit contributions
ez+0.026
eros+0.024
 wound+0.024
 heaven+0.024
-+0.024
Neg logit contributions
 topp-0.050
First-0.050
 ju-0.049
 flavor-0.046
 flour-0.045
Token
Token ID198
Feature activation+1.32e-02

Pos logit contributions
First+0.089
 topp+0.085
 ju+0.084
 flavor+0.081
 forks+0.081
Neg logit contributions
ez-0.063
eros-0.059
rows-0.057
 wound-0.056
oring-0.055
Token
Token ID198
Feature activation+1.37e-02

Pos logit contributions
ez+0.014
eros+0.014
 wound+0.013
 tears+0.013
oring+0.013
Neg logit contributions
First-0.024
 topp-0.022
 ju-0.022
 flavor-0.021
 flour-0.021
TokenL
Token ID43
Feature activation+9.91e-02

Pos logit contributions
ez+0.015
eros+0.014
 wound+0.014
 heaven+0.013
 tears+0.013
Neg logit contributions
 topp-0.025
First-0.024
 ju-0.024
 flavor-0.023
 flour-0.022
Token're
Token ID821
Feature activation-1.15e-01

Pos logit contributions
 Start+0.002
 terrified+0.002
ez+0.002
 tears+0.002
Start+0.002
Neg logit contributions
First-0.001
 topp-0.001
 ju-0.001
 apr-0.001
 forks-0.001
Token too
Token ID1165
Feature activation-2.75e-02

Pos logit contributions
 topp+0.202
First+0.200
 ju+0.195
 flavor+0.185
 flour+0.183
Neg logit contributions
ez-0.132
eros-0.125
 wound-0.122
 tears-0.120
 heaven-0.119
Token small
Token ID1402
Feature activation-6.42e-02

Pos logit contributions
 topp+0.049
First+0.048
 ju+0.047
 forks+0.045
 apr+0.044
Neg logit contributions
 terrified-0.031
 crying-0.031
--0.030
ez-0.029
 Start-0.029
Token to
Token ID284
Feature activation-4.09e-02

Pos logit contributions
First+0.127
 ju+0.126
 topp+0.125
Deal+0.117
 flavor+0.116
Neg logit contributions
--0.063
 crying-0.058
ez-0.057
 heaven-0.054
 knowing-0.054
Token get
Token ID651
Feature activation+5.26e-02

Pos logit contributions
 topp+0.061
First+0.058
 ju+0.057
 forks+0.055
 flour+0.054
Neg logit contributions
ez-0.059
 heaven-0.057
 wound-0.057
 crying-0.057
 its-0.056
Token the
Token ID262
Feature activation+9.41e-03

Pos logit contributions
ez+0.054
eros+0.052
 wound+0.050
 tears+0.048
 heaven+0.048
Neg logit contributions
 topp-0.098
First-0.096
 ju-0.095
 flavor-0.090
 flour-0.089
Token food
Token ID2057
Feature activation-2.73e-02

Pos logit contributions
 creature+0.011
 tears+0.011
 wound+0.011
 hospital+0.011
 heaven+0.011
Neg logit contributions
First-0.016
 topp-0.014
ipp-0.014
teen-0.013
bars-0.013
Token from
Token ID422
Feature activation+1.22e-01

Pos logit contributions
 topp+0.051
First+0.051
 ju+0.047
 flavor+0.045
 apr+0.044
Neg logit contributions
--0.028
 heaven-0.026
 wound-0.026
 crying-0.025
eros-0.025
Token the
Token ID262
Feature activation+1.09e-02

Pos logit contributions
ez+0.113
eros+0.109
 heaven+0.104
 wound+0.104
 tears+0.102
Neg logit contributions
 topp-0.240
First-0.240
 ju-0.235
 flavor-0.222
 flour-0.218
Token tree
Token ID5509
Feature activation-3.54e-02

Pos logit contributions
 hospital+0.013
 tears+0.012
 wound+0.012
 creature+0.012
eros+0.011
Neg logit contributions
First-0.020
 topp-0.020
ipp-0.018
 flavor-0.017
aney-0.017
Token!"
Token ID2474
Feature activation-7.94e-02

Pos logit contributions
 topp+0.060
First+0.059
 ju+0.058
 forks+0.055
 flavor+0.054
Neg logit contributions
--0.040
ez-0.040
 wound-0.040
 heaven-0.040
eros-0.040
Token asked
Token ID1965
Feature activation-5.23e-03

Pos logit contributions
 topp+0.002
First+0.002
 ju+0.002
 flavor+0.002
 flour+0.002
Neg logit contributions
ez-0.002
 wound-0.002
eros-0.002
 tears-0.002
 heaven-0.002
Token her
Token ID607
Feature activation+8.25e-02

Pos logit contributions
 topp+0.011
First+0.011
 ju+0.011
 flavor+0.010
 flour+0.010
Neg logit contributions
ez-0.004
 its-0.004
 tears-0.004
--0.004
 crying-0.004
Token mom
Token ID1995
Feature activation-4.09e-02

Pos logit contributions
 crying+0.083
 wound+0.082
 tears+0.080
-+0.080
eros+0.077
Neg logit contributions
First-0.158
 topp-0.158
 ju-0.148
avored-0.139
 flavor-0.139
Token if
Token ID611
Feature activation+1.67e-02

Pos logit contributions
 topp+0.082
First+0.082
 ju+0.079
 flavor+0.073
 apr+0.073
Neg logit contributions
--0.036
 crying-0.035
 knowing-0.033
 its-0.033
 heaven-0.032
Token she
Token ID673
Feature activation+6.13e-02

Pos logit contributions
ez+0.016
eros+0.016
 wound+0.015
 heaven+0.015
 tears+0.015
Neg logit contributions
 topp-0.032
First-0.031
 ju-0.030
 flavor-0.029
 flour-0.029
Token could
Token ID714
Feature activation-1.62e-01

Pos logit contributions
-+0.079
 wound+0.078
ez+0.076
 relieved+0.074
 tears+0.073
Neg logit contributions
First-0.101
 topp-0.100
 flour-0.093
 ju-0.092
 apr-0.091
Token stay
Token ID2652
Feature activation-1.74e-02

Pos logit contributions
 topp+0.283
 ju+0.274
First+0.262
 flour+0.250
 forks+0.249
Neg logit contributions
 wound-0.210
--0.206
 crying-0.200
 its-0.198
 heaven-0.195
Token close
Token ID1969
Feature activation+1.84e-02

Pos logit contributions
First+0.026
 ju+0.026
 topp+0.025
 flour+0.025
 flavor+0.025
Neg logit contributions
ez-0.023
eros-0.023
crow-0.021
oc-0.020
 Start-0.020
Token to
Token ID284
Feature activation-4.34e-02

Pos logit contributions
 its+0.014
-+0.014
 heaven+0.014
ez+0.014
 crying+0.013
Neg logit contributions
 topp-0.039
First-0.037
 ju-0.037
 flavor-0.035
 stack-0.035
Token the
Token ID262
Feature activation+1.13e-02

Pos logit contributions
 topp+0.086
First+0.085
 ju+0.084
 flavor+0.080
 flour+0.080
Neg logit contributions
ez-0.040
eros-0.037
 wound-0.036
 tears-0.034
 heaven-0.034
Token shore
Token ID15191
Feature activation+6.12e-02

Pos logit contributions
 wound+0.014
 hospital+0.014
 tears+0.014
 creature+0.013
 speaker+0.013
Neg logit contributions
First-0.020
 topp-0.019
 ju-0.018
 flavor-0.017
ipp-0.016
Token They
Token ID1119
Feature activation-3.38e-02

Pos logit contributions
ez+0.008
eros+0.008
 wound+0.007
 tears+0.007
 heaven+0.007
Neg logit contributions
 topp-0.010
First-0.009
 ju-0.009
 flavor-0.009
 flour-0.009
Token have
Token ID423
Feature activation+4.89e-02

Pos logit contributions
 topp+0.049
 ju+0.047
First+0.046
 flavor+0.045
 flour+0.044
Neg logit contributions
 wound-0.049
 tears-0.048
ez-0.048
 speaker-0.048
 terrified-0.046
Token a
Token ID257
Feature activation-6.80e-02

Pos logit contributions
 its+0.064
 tears+0.063
 speaker+0.059
 wound+0.058
ez+0.057
Neg logit contributions
avored-0.079
 ju-0.077
 topp-0.077
umm-0.073
Deal-0.072
Token lot
Token ID1256
Feature activation+4.76e-02

Pos logit contributions
First+0.117
 topp+0.114
 ju+0.112
 forks+0.104
nuts+0.104
Neg logit contributions
 speaker-0.084
 wound-0.082
 tears-0.078
 hospital-0.078
 crying-0.077
Token of
Token ID286
Feature activation-3.43e-02

Pos logit contributions
ez+0.054
eros+0.052
 wound+0.050
 tears+0.049
 heaven+0.049
Neg logit contributions
 topp-0.084
First-0.083
 ju-0.081
 flavor-0.077
 flour-0.077
Token fun
Token ID1257
Feature activation-1.40e-01

Pos logit contributions
 ju+0.031
 topp+0.030
First+0.029
 flavor+0.025
 forks+0.025
Neg logit contributions
ez-0.069
 tears-0.067
 wound-0.067
 speaker-0.066
eros-0.066
Token and
Token ID290
Feature activation+4.41e-02

Pos logit contributions
 topp+0.223
First+0.214
 stack+0.207
 flour+0.205
 flavor+0.202
Neg logit contributions
ez-0.184
eros-0.173
crow-0.170
 heaven-0.168
 Start-0.167
Token laugh
Token ID6487
Feature activation+2.66e-02

Pos logit contributions
ez+0.065
eros+0.063
 wound+0.062
 tears+0.060
 heaven+0.060
Neg logit contributions
 topp-0.062
First-0.061
 ju-0.060
 flavor-0.056
 flour-0.055
Token a
Token ID257
Feature activation-6.93e-02

Pos logit contributions
ez+0.027
eros+0.026
 wound+0.025
 tears+0.024
crow+0.024
Neg logit contributions
 topp-0.049
First-0.049
 ju-0.048
 flavor-0.046
 flour-0.045
Token lot
Token ID1256
Feature activation+4.71e-02

Pos logit contributions
 topp+0.072
First+0.068
 ju+0.066
 flavor+0.064
 stack+0.060
Neg logit contributions
ez-0.132
eros-0.130
crow-0.127
 wound-0.123
 unconscious-0.122
Token.
Token ID13
Feature activation+1.49e-03

Pos logit contributions
ez+0.058
eros+0.054
 wound+0.053
 tears+0.052
 terrified+0.050
Neg logit contributions
 topp-0.079
First-0.077
 ju-0.076
 flavor-0.072
 flour-0.072
Token soon
Token ID2582
Feature activation+3.01e-02

Pos logit contributions
 ju+0.068
First+0.067
 topp+0.065
avored+0.064
 redd+0.061
Neg logit contributions
--0.036
ez-0.036
 wound-0.035
les-0.031
oc-0.030
Token had
Token ID550
Feature activation-5.47e-02

Pos logit contributions
eros+0.039
ez+0.037
 heaven+0.035
 wound+0.034
 speaker+0.034
Neg logit contributions
 ju-0.049
 topp-0.048
First-0.047
 flavor-0.046
 forks-0.046
Token a
Token ID257
Feature activation-6.99e-02

Pos logit contributions
First+0.099
 topp+0.097
 ju+0.094
Deal+0.089
 flavor+0.088
Neg logit contributions
ez-0.059
 tears-0.057
 its-0.056
 wound-0.056
 heaven-0.052
Token great
Token ID1049
Feature activation-7.48e-02

Pos logit contributions
First+0.121
 topp+0.109
nuts+0.101
 ju+0.101
avored+0.099
Neg logit contributions
 wound-0.089
 tears-0.088
 crying-0.087
 creature-0.087
 hospital-0.084
Token big
Token ID1263
Feature activation-9.83e-02

Pos logit contributions
First+0.108
 ju+0.102
 topp+0.099
 forks+0.091
nuts+0.090
Neg logit contributions
--0.115
 wound-0.112
 tears-0.110
 creature-0.106
 its-0.104
Token pile
Token ID14540
Feature activation-1.51e-01

Pos logit contributions
First+0.158
 ju+0.138
 topp+0.131
avored+0.123
nuts+0.122
Neg logit contributions
 wound-0.147
 tears-0.147
--0.146
 creature-0.145
eros-0.140
Token of
Token ID286
Feature activation-3.70e-02

Pos logit contributions
First+0.348
 topp+0.338
 ju+0.327
 apr+0.312
 fiber+0.309
Neg logit contributions
--0.109
 its-0.100
 wound-0.093
ez-0.092
eros-0.091
Token sand
Token ID6450
Feature activation-9.54e-02

Pos logit contributions
First+0.064
ver+0.057
'd+0.057
Deal+0.057
avored+0.056
Neg logit contributions
 tears-0.048
 wound-0.047
 its-0.045
 creature-0.043
--0.042
Token.
Token ID13
Feature activation-3.49e-03

Pos logit contributions
First+0.183
 topp+0.163
 ju+0.163
 flavor+0.151
 First+0.150
Neg logit contributions
--0.110
eros-0.103
 wound-0.103
ment-0.100
ez-0.095
Token He
Token ID679
Feature activation-3.32e-02

Pos logit contributions
 topp+0.008
First+0.008
 ju+0.008
 flavor+0.007
 flour+0.007
Neg logit contributions
ez-0.002
eros-0.002
 tears-0.002
 wound-0.002
 Start-0.002
Token started
Token ID2067
Feature activation-5.66e-02

Pos logit contributions
First+0.060
 ju+0.057
 topp+0.056
avored+0.054
teen+0.052
Neg logit contributions
 wound-0.038
ez-0.037
 tears-0.036
--0.036
 its-0.034
Token you
Token ID345
Feature activation+1.21e-03

Pos logit contributions
ez+0.091
eros+0.088
 heaven+0.087
 wound+0.086
-+0.085
Neg logit contributions
First-0.089
 topp-0.089
 ju-0.088
 flavor-0.079
 flour-0.079
Token are
Token ID389
Feature activation-5.56e-02

Pos logit contributions
ez+0.002
 heaven+0.002
 wound+0.002
 its+0.002
-+0.002
Neg logit contributions
 topp-0.002
 ju-0.002
First-0.001
 flour-0.001
 flavor-0.001
Token and
Token ID290
Feature activation+3.25e-02

Pos logit contributions
 topp+0.098
First+0.098
 ju+0.096
 flavor+0.089
 flour+0.087
Neg logit contributions
ez-0.060
eros-0.059
 heaven-0.059
--0.059
 its-0.058
Token where
Token ID810
Feature activation+5.62e-02

Pos logit contributions
 heaven+0.039
 its+0.039
 tears+0.038
 crying+0.037
 wound+0.036
Neg logit contributions
First-0.055
 ju-0.052
 topp-0.051
avored-0.048
 flavor-0.047
Token you
Token ID345
Feature activation+2.56e-03

Pos logit contributions
 its+0.053
 crying+0.051
 tears+0.051
 heaven+0.050
eros+0.049
Neg logit contributions
First-0.107
 topp-0.106
 ju-0.103
 flavor-0.095
Deal-0.094
Token go
Token ID467
Feature activation-1.42e-01

Pos logit contributions
ez+0.004
 heaven+0.004
 wound+0.004
-+0.004
 speaker+0.004
Neg logit contributions
 topp-0.004
 ju-0.003
First-0.003
 flour-0.003
 flavor-0.003
Token.
Token ID13
Feature activation-8.06e-03

Pos logit contributions
First+0.232
 topp+0.229
 ju+0.221
 flavor+0.215
 flour+0.212
Neg logit contributions
ez-0.180
eros-0.169
 tears-0.163
 wound-0.163
 terrified-0.157
Token
Token ID198
Feature activation+9.57e-03

Pos logit contributions
 topp+0.015
 ju+0.014
First+0.014
 flavor+0.014
 forks+0.013
Neg logit contributions
ez-0.008
eros-0.008
 heaven-0.008
 wound-0.008
crow-0.008
Token
Token ID198
Feature activation+1.04e-02

Pos logit contributions
ez+0.008
eros+0.007
 wound+0.007
 tears+0.007
 heaven+0.007
Neg logit contributions
 topp-0.020
First-0.020
 ju-0.019
 flavor-0.018
 flour-0.018
TokenThey
Token ID2990
Feature activation-7.74e-02

Pos logit contributions
ez+0.016
crow+0.016
-+0.015
 wound+0.015
 heaven+0.015
Neg logit contributions
 topp-0.013
 ju-0.013
 forks-0.012
First-0.011
 shovel-0.011
Token go
Token ID467
Feature activation-1.43e-01

Pos logit contributions
 topp+0.117
 ju+0.112
First+0.108
 flour+0.103
 flavor+0.102
Neg logit contributions
 heaven-0.107
ez-0.107
 speaker-0.106
 wound-0.105
 tears-0.105
Token a
Token ID257
Feature activation-7.85e-02

Pos logit contributions
ez+0.080
eros+0.078
 wound+0.074
crow+0.073
oring+0.073
Neg logit contributions
 topp-0.103
First-0.102
 ju-0.099
 flavor-0.094
 flour-0.093
Token time
Token ID640
Feature activation-7.39e-02

Pos logit contributions
 ju+0.125
 topp+0.123
nuts+0.119
First+0.116
 forks+0.114
Neg logit contributions
ez-0.107
--0.107
 heaven-0.102
oc-0.099
 tears-0.097
Token,
Token ID11
Feature activation-3.96e-02

Pos logit contributions
 topp+0.135
First+0.131
 ju+0.130
 flavor+0.121
 flour+0.121
Neg logit contributions
ez-0.080
eros-0.079
--0.075
 crying-0.075
 its-0.073
Token there
Token ID612
Feature activation-5.21e-02

Pos logit contributions
 topp+0.063
First+0.062
 ju+0.061
 flavor+0.057
 flour+0.057
Neg logit contributions
ez-0.051
eros-0.049
 wound-0.048
 heaven-0.047
 tears-0.047
Token were
Token ID547
Feature activation-9.65e-02

Pos logit contributions
 topp+0.103
First+0.102
 ju+0.101
 flavor+0.095
 flour+0.095
Neg logit contributions
ez-0.048
eros-0.045
 wound-0.043
 tears-0.042
 heaven-0.041
Token two
Token ID734
Feature activation-2.03e-01

Pos logit contributions
 topp+0.160
Deal+0.155
'd+0.152
 stack+0.149
First+0.147
Neg logit contributions
ez-0.116
 its-0.112
 crying-0.108
 tears-0.107
--0.105
Token best
Token ID1266
Feature activation-1.54e-01

Pos logit contributions
First+0.441
 ju+0.415
 topp+0.414
 First+0.383
 flavor+0.375
Neg logit contributions
 tears-0.188
--0.186
 crying-0.173
eros-0.164
 wound-0.163
Token friends
Token ID2460
Feature activation-7.80e-02

Pos logit contributions
 ju+0.373
First+0.363
 topp+0.348
'd+0.335
 redd+0.329
Neg logit contributions
--0.119
eros-0.098
 tears-0.086
ez-0.082
 wound-0.081
Token named
Token ID3706
Feature activation+9.94e-02

Pos logit contributions
First+0.094
 shovel+0.091
 topp+0.090
 flavor+0.089
 flour+0.088
Neg logit contributions
oring-0.117
ez-0.116
izes-0.115
Start-0.114
eros-0.114
Token Paul
Token ID3362
Feature activation+3.48e-02

Pos logit contributions
 its+0.062
 Lily+0.057
 Tinker+0.056
-+0.055
 heaven+0.054
Neg logit contributions
 topp-0.211
Deal-0.210
 ju-0.210
'd-0.209
First-0.204
Token and
Token ID290
Feature activation+2.40e-02

Pos logit contributions
ez+0.033
 terrified+0.032
eros+0.032
 tears+0.032
 wound+0.032
Neg logit contributions
 topp-0.066
First-0.063
 flavor-0.061
 flour-0.060
 forks-0.060