TOP ACTIVATIONS
MAX = 0.191

felt a chill down his spine . He suddenly remembered the
and ran off in the opposite direction . He was free
a chill run down his spine . He could not believe
sh iver go down her spine . It was very cold
sh iver go down her spine . It tasted different than
sh iver go down his spine . He looked around but
a chill going down their spine . This was the wisdom
to Ell a before hopping off to find his family .
time , the mug flew off the path , just like
the air and it soared off under the rainbow .
breeze blew her in the opposite direction of the cat .
, leaving it to its former state .
last time , before running off to explore something else .
the village back to its former glory !"
The plane flew off , feeling very proud of
kissed him goodbye before running off with a big smile .
night . As she drifted off to sleep , she thought
kind animal before scam pering off to explore the forest again
move again and she ran off in search of her next
the duck as it walked off , and Kate went back

MIN ACTIVATIONS
MIN = -0.181

Lily said . " It 's okay , my
eat them . " Maybe we can fly with
Ben said . " No , I 'm not
Ben shouts . " No , it 's mine
the sand . " Look , Sam , I
Ben says . " You are very wise ,
Tom says . " That is a good idea
pretty hair ." " Hi , Max ," Tom
were bored . " Let 's organize our toys
high things . " Come on , Kim ,
blamed Mia . " It 's your fault !"
Lily said . " Hey , cat . Go
find treasure . " Look , Tom , there
Ben said . " Yes , I do .
Ben said . " Ben took my car .
a feather . " Thank you , bird ,
Ben said . " Then you are very lucky
Lily said . " L ily , the pictures
is pretty . " Can I have your star
for Ben . " Ben , you need to

INTERVAL 0.098 - 0.191
CONTAINS 1.336%

bush nearby . He flew over and called out , "
didn 't try to hang from the truck .
to be with his coat again . <|endoftext|> Once there was
sticks . The bird flew out of the nest and attacked
. The snake slit hered away . Mom and

INTERVAL 0.005 - 0.098
CONTAINS 43.353%

. " Da isy , come inside now ,"
." Lucy nodded and went outside to look for the mystery
!" The little girl jumped with excitement . She
â„¢ s pick some flowers to take home with us .
we go inside ?" Lily asked . " No

INTERVAL -0.088 - 0.005
CONTAINS 53.816%

. <|endoftext|> Once upon a time , there was a little
the sand , and the water . She dreamed of the
on the table . It had a label that said "
. The box is tight and hard to open . They
proud . It thinks it has won the game .

INTERVAL -0.181 - -0.088
CONTAINS 1.495%

a fire truck . They wanted to see it up close
L ily replied , " I fell and got a
says to Anna , " Why don 't we share our
" Hello , Lily . What a nice surprise . Are
looked very excited . " Can I come with you ?"
Token felt
Token ID2936
Feature activation+3.12e-02

Pos logit contributions
 ju+0.020
 golf+0.019
 makeup+0.018
 butter+0.018
 contest+0.017
Neg logit contributions
rum-0.028
ously-0.027
 despair-0.026
 deeply-0.026
 horror-0.025
Token a
Token ID257
Feature activation+3.50e-04

Pos logit contributions
ously+0.063
rum+0.062
 despair+0.055
aced+0.054
 horror+0.053
Neg logit contributions
 ju-0.043
 golf-0.041
 makeup-0.039
 butter-0.037
 gum-0.036
Token chill
Token ID20493
Feature activation+6.01e-02

Pos logit contributions
rum+0.001
ously+0.001
uzz+0.001
aced+0.001
ment+0.001
Neg logit contributions
 ju-0.000
 butter-0.000
 golf-0.000
 gum-0.000
 lemon-0.000
Token down
Token ID866
Feature activation+8.88e-02

Pos logit contributions
ously+0.107
rum+0.107
 despair+0.100
 deeply+0.099
 horror+0.096
Neg logit contributions
 ju-0.090
 golf-0.086
 makeup-0.084
 butter-0.082
mons-0.080
Token his
Token ID465
Feature activation+7.22e-02

Pos logit contributions
rum+0.193
ously+0.178
 despair+0.177
 horror+0.170
 heartbeat+0.168
Neg logit contributions
 ju-0.104
 golf-0.102
 contest-0.093
 butter-0.091
 makeup-0.090
Token spine
Token ID19656
Feature activation+1.91e-01

Pos logit contributions
rum+0.097
ously+0.095
 despair+0.081
 deeply+0.081
 horror+0.076
Neg logit contributions
 golf-0.145
 makeup-0.144
 ju-0.142
 swing-0.137
 file-0.135
Token.
Token ID13
Feature activation+3.78e-02

Pos logit contributions
rum+0.401
 horror+0.387
 despair+0.386
Flash+0.371
 Deal+0.370
Neg logit contributions
 golf-0.186
 butter-0.174
 makeup-0.172
 swing-0.168
 ju-0.164
Token He
Token ID679
Feature activation+3.96e-02

Pos logit contributions
rum+0.063
ously+0.061
 despair+0.057
urbed+0.056
 deeply+0.055
Neg logit contributions
 golf-0.065
 ju-0.065
mons-0.061
 gum-0.059
 makeup-0.058
Token suddenly
Token ID6451
Feature activation+3.15e-02

Pos logit contributions
rum+0.095
ously+0.086
uzzle+0.080
aced+0.077
 horror+0.077
Neg logit contributions
 pin-0.042
 ju-0.040
 butter-0.039
 swing-0.034
 makeup-0.034
Token remembered
Token ID12086
Feature activation+4.63e-02

Pos logit contributions
rum+0.068
ously+0.061
 horror+0.061
 despair+0.060
 deeply+0.059
Neg logit contributions
 ju-0.036
 butter-0.035
 golf-0.035
 makeup-0.033
 gum-0.032
Token the
Token ID262
Feature activation-1.13e-02

Pos logit contributions
rum+0.090
ously+0.088
 despair+0.083
 deeply+0.083
 horror+0.080
Neg logit contributions
 golf-0.063
 ju-0.063
 makeup-0.060
 butter-0.056
 grapes-0.056
Token and
Token ID290
Feature activation+2.00e-02

Pos logit contributions
rum+0.101
 despair+0.096
ously+0.096
 horror+0.093
 heartbeat+0.091
Neg logit contributions
 golf-0.071
 ju-0.069
 contest-0.065
 competition-0.064
 makeup-0.064
Token ran
Token ID4966
Feature activation-5.89e-03

Pos logit contributions
rum+0.042
ously+0.040
 deeply+0.038
 despair+0.037
 horror+0.037
Neg logit contributions
 ju-0.028
 golf-0.026
 butter-0.024
 makeup-0.023
 swing-0.023
Token off
Token ID572
Feature activation+1.73e-01

Pos logit contributions
 golf+0.009
 ju+0.008
 makeup+0.008
 contest+0.008
 competition+0.007
Neg logit contributions
rum-0.011
ously-0.011
 despair-0.011
 horror-0.010
 heartbeat-0.010
Token in
Token ID287
Feature activation+1.14e-01

Pos logit contributions
rum+0.440
ously+0.431
 despair+0.426
aced+0.421
 heartbeat+0.410
Neg logit contributions
 golf-0.156
 contest-0.127
 swing-0.124
 competition-0.117
 ju-0.111
Token the
Token ID262
Feature activation+2.53e-02

Pos logit contributions
ously+0.219
aned+0.204
rum+0.204
atively+0.198
edly+0.197
Neg logit contributions
 golf-0.157
 makeup-0.144
 ju-0.138
 gum-0.137
 competition-0.133
Token opposite
Token ID6697
Feature activation+1.87e-01

Pos logit contributions
ously+0.049
rum+0.047
 despair+0.045
 deeply+0.045
 horror+0.043
Neg logit contributions
 ju-0.036
 golf-0.034
 makeup-0.033
mons-0.033
 contest-0.032
Token direction
Token ID4571
Feature activation+9.86e-02

Pos logit contributions
rum+0.281
ously+0.277
uzz+0.253
 heartbeat+0.250
 silently+0.244
Neg logit contributions
 golf-0.305
 ju-0.303
 swing-0.292
 grapes-0.283
 competition-0.282
Token.
Token ID13
Feature activation+5.29e-02

Pos logit contributions
rum+0.178
ously+0.172
 despair+0.168
 deeply+0.160
 horror+0.159
Neg logit contributions
 golf-0.142
 ju-0.139
 makeup-0.131
 contest-0.123
 gum-0.122
Token He
Token ID679
Feature activation+5.33e-02

Pos logit contributions
ously+0.095
rum+0.093
 despair+0.084
aced+0.084
 deeply+0.081
Neg logit contributions
 golf-0.084
 ju-0.083
mons-0.072
 gum-0.071
 basketball-0.070
Token was
Token ID373
Feature activation+2.70e-02

Pos logit contributions
rum+0.112
ously+0.105
 despair+0.097
 deeply+0.096
 heartbeat+0.096
Neg logit contributions
 ju-0.071
 golf-0.062
 swing-0.060
 butter-0.060
 makeup-0.058
Token free
Token ID1479
Feature activation+3.08e-02

Pos logit contributions
ously+0.063
rum+0.062
 despair+0.055
urbed+0.054
atively+0.054
Neg logit contributions
 ju-0.028
 golf-0.025
 makeup-0.023
 butter-0.023
 competition-0.022
Token a
Token ID257
Feature activation-3.29e-03

Pos logit contributions
ously+0.064
rum+0.063
 despair+0.059
 deeply+0.058
 horror+0.057
Neg logit contributions
 ju-0.045
 golf-0.043
 makeup-0.042
mons-0.040
 contest-0.040
Token chill
Token ID20493
Feature activation+5.46e-02

Pos logit contributions
 ju+0.005
 butter+0.004
 makeup+0.004
 golf+0.004
 gum+0.004
Neg logit contributions
rum-0.007
ously-0.007
uzz-0.006
aced-0.006
 despair-0.006
Token run
Token ID1057
Feature activation-4.26e-03

Pos logit contributions
ously+0.096
rum+0.093
 despair+0.088
 deeply+0.087
 horror+0.084
Neg logit contributions
 ju-0.086
 golf-0.082
mons-0.080
 makeup-0.080
 contest-0.078
Token down
Token ID866
Feature activation+8.19e-02

Pos logit contributions
 ju+0.008
 golf+0.007
mons+0.007
 makeup+0.007
atmeal+0.007
Neg logit contributions
ously-0.007
rum-0.006
 despair-0.006
 deeply-0.006
 horror-0.006
Token his
Token ID465
Feature activation+6.55e-02

Pos logit contributions
rum+0.157
ously+0.149
 despair+0.140
 horror+0.134
 deeply+0.134
Neg logit contributions
 golf-0.119
 ju-0.116
 makeup-0.110
 contest-0.110
 competition-0.105
Token spine
Token ID19656
Feature activation+1.86e-01

Pos logit contributions
rum+0.086
ously+0.085
 despair+0.072
 deeply+0.070
 horror+0.068
Neg logit contributions
 makeup-0.134
 ju-0.133
 golf-0.133
 file-0.127
 quiz-0.126
Token.
Token ID13
Feature activation+4.40e-02

Pos logit contributions
rum+0.394
 despair+0.374
 horror+0.373
Flash+0.364
ously+0.360
Neg logit contributions
 golf-0.187
 makeup-0.178
 butter-0.167
 swing-0.167
 paints-0.160
Token He
Token ID679
Feature activation+3.87e-02

Pos logit contributions
ously+0.065
rum+0.064
 despair+0.058
 deeply+0.055
urbed+0.054
Neg logit contributions
 golf-0.085
 ju-0.082
 makeup-0.080
mons-0.079
 gum-0.078
Token could
Token ID714
Feature activation-2.42e-03

Pos logit contributions
rum+0.083
ously+0.079
 despair+0.071
 horror+0.070
 deeply+0.070
Neg logit contributions
 ju-0.047
 makeup-0.044
 paints-0.043
 butter-0.043
 golf-0.042
Token not
Token ID407
Feature activation+2.60e-02

Pos logit contributions
 ju+0.003
 golf+0.003
 makeup+0.003
 swing+0.003
 quiz+0.003
Neg logit contributions
ously-0.005
rum-0.005
 despair-0.004
 deeply-0.004
aced-0.004
Token believe
Token ID1975
Feature activation+5.78e-02

Pos logit contributions
ously+0.050
rum+0.049
 deeply+0.045
 despair+0.045
 horror+0.043
Neg logit contributions
 ju-0.036
 golf-0.035
 makeup-0.035
 contest-0.033
 quiz-0.033
Token sh
Token ID427
Feature activation-1.24e-02

Pos logit contributions
 ju+0.018
 golf+0.017
 makeup+0.017
 butter+0.016
 quiz+0.016
Neg logit contributions
rum-0.024
ously-0.024
 despair-0.021
 deeply-0.021
aced-0.021
Tokeniver
Token ID1428
Feature activation+4.85e-02

Pos logit contributions
 ju+0.023
 golf+0.022
 makeup+0.021
mons+0.021
 grapes+0.021
Neg logit contributions
ously-0.019
rum-0.019
 deeply-0.017
 despair-0.017
 horror-0.016
Token go
Token ID467
Feature activation-9.87e-03

Pos logit contributions
ously+0.080
rum+0.078
 despair+0.074
 deeply+0.073
 horror+0.070
Neg logit contributions
 ju-0.081
 golf-0.078
mons-0.076
 makeup-0.076
 contest-0.073
Token down
Token ID866
Feature activation+7.69e-02

Pos logit contributions
 ju+0.015
mons+0.015
 grapes+0.014
atmeal+0.014
 golf+0.013
Neg logit contributions
ously-0.017
 deeply-0.017
rum-0.016
 despair-0.016
 silently-0.016
Token her
Token ID607
Feature activation+6.69e-02

Pos logit contributions
ously+0.140
rum+0.138
 despair+0.129
 deeply+0.124
 horror+0.123
Neg logit contributions
 ju-0.117
 golf-0.116
 makeup-0.112
 contest-0.110
 grapes-0.107
Token spine
Token ID19656
Feature activation+1.85e-01

Pos logit contributions
rum+0.106
ously+0.106
 despair+0.096
 deeply+0.094
 horror+0.092
Neg logit contributions
 ju-0.119
 golf-0.118
 makeup-0.116
 swing-0.112
mons-0.112
Token.
Token ID13
Feature activation+3.87e-02

Pos logit contributions
rum+0.419
 despair+0.412
 horror+0.408
ously+0.397
Flash+0.396
Neg logit contributions
 makeup-0.155
 golf-0.150
 butter-0.146
 swing-0.142
 paints-0.136
Token It
Token ID632
Feature activation-1.13e-02

Pos logit contributions
ously+0.055
rum+0.054
 despair+0.050
 deeply+0.049
 horror+0.047
Neg logit contributions
 ju-0.075
 golf-0.072
 makeup-0.071
mons-0.070
 grapes-0.068
Token was
Token ID373
Feature activation+2.10e-02

Pos logit contributions
 ju+0.017
 makeup+0.015
 golf+0.015
 butter+0.014
 contest+0.014
Neg logit contributions
rum-0.022
ously-0.021
 despair-0.019
 deeply-0.019
 horror-0.019
Token very
Token ID845
Feature activation+3.00e-02

Pos logit contributions
ously+0.044
rum+0.044
 despair+0.039
 deeply+0.038
 horror+0.038
Neg logit contributions
 ju-0.027
 makeup-0.026
 golf-0.026
 butter-0.025
 grapes-0.024
Token cold
Token ID4692
Feature activation-6.62e-03

Pos logit contributions
ously+0.081
rum+0.077
urbed+0.071
 despair+0.069
 horror+0.069
Neg logit contributions
 ju-0.025
 makeup-0.023
 golf-0.022
 mean-0.019
 butter-0.018
Token sh
Token ID427
Feature activation-1.08e-02

Pos logit contributions
 ju+0.016
 butter+0.014
 makeup+0.014
 golf+0.014
 gum+0.013
Neg logit contributions
rum-0.024
ously-0.023
uzz-0.020
 despair-0.020
 deeply-0.020
Tokeniver
Token ID1428
Feature activation+4.86e-02

Pos logit contributions
 ju+0.019
 golf+0.019
 makeup+0.018
mons+0.018
 grapes+0.018
Neg logit contributions
ously-0.017
rum-0.017
 despair-0.015
 deeply-0.015
 horror-0.014
Token go
Token ID467
Feature activation-6.65e-03

Pos logit contributions
ously+0.080
rum+0.076
 despair+0.073
 deeply+0.073
 horror+0.069
Neg logit contributions
 ju-0.083
 golf-0.079
mons-0.079
 makeup-0.076
 contest-0.075
Token down
Token ID866
Feature activation+7.79e-02

Pos logit contributions
 ju+0.011
mons+0.011
atmeal+0.011
 grapes+0.011
 golf+0.011
Neg logit contributions
ously-0.010
rum-0.010
 despair-0.010
 deeply-0.010
 silently-0.009
Token her
Token ID607
Feature activation+7.16e-02

Pos logit contributions
rum+0.146
ously+0.144
 despair+0.134
 deeply+0.127
 horror+0.127
Neg logit contributions
 golf-0.114
 ju-0.113
 makeup-0.111
 contest-0.109
 grapes-0.106
Token spine
Token ID19656
Feature activation+1.85e-01

Pos logit contributions
ously+0.121
rum+0.118
 despair+0.111
 deeply+0.109
 horror+0.105
Neg logit contributions
 ju-0.122
 golf-0.119
 makeup-0.117
mons-0.114
 grapes-0.112
Token.
Token ID13
Feature activation+4.06e-02

Pos logit contributions
rum+0.414
 despair+0.407
 horror+0.401
Flash+0.396
Ar+0.389
Neg logit contributions
 makeup-0.155
 golf-0.155
 butter-0.150
 swing-0.145
mons-0.136
Token It
Token ID632
Feature activation-1.13e-02

Pos logit contributions
ously+0.060
rum+0.058
 despair+0.054
 deeply+0.053
 horror+0.051
Neg logit contributions
 ju-0.076
 golf-0.073
 makeup-0.072
mons-0.071
 contest-0.069
Token tasted
Token ID29187
Feature activation+5.98e-03

Pos logit contributions
 ju+0.016
 makeup+0.015
 golf+0.014
 butter+0.014
 contest+0.014
Neg logit contributions
ously-0.022
rum-0.022
 despair-0.020
 deeply-0.019
 horror-0.019
Token different
Token ID1180
Feature activation+2.35e-02

Pos logit contributions
ously+0.013
rum+0.012
 despair+0.012
 deeply+0.012
 horror+0.011
Neg logit contributions
 ju-0.007
 golf-0.007
 makeup-0.007
mons-0.007
 grapes-0.006
Token than
Token ID621
Feature activation-2.56e-02

Pos logit contributions
ously+0.037
rum+0.036
 despair+0.034
 deeply+0.033
 horror+0.032
Neg logit contributions
 ju-0.042
 makeup-0.040
 golf-0.040
mons-0.039
 grapes-0.038
Token sh
Token ID427
Feature activation-1.20e-02

Pos logit contributions
 ju+0.012
 golf+0.011
 butter+0.010
 makeup+0.010
 gum+0.010
Neg logit contributions
rum-0.016
ously-0.016
 despair-0.014
 deeply-0.014
aced-0.014
Tokeniver
Token ID1428
Feature activation+4.97e-02

Pos logit contributions
 ju+0.021
 golf+0.021
 makeup+0.020
mons+0.020
 grapes+0.020
Neg logit contributions
ously-0.019
rum-0.019
 despair-0.017
 deeply-0.017
 horror-0.016
Token go
Token ID467
Feature activation-7.83e-03

Pos logit contributions
ously+0.080
rum+0.078
 despair+0.073
 deeply+0.072
 horror+0.070
Neg logit contributions
 ju-0.085
 golf-0.082
 makeup-0.080
mons-0.079
 contest-0.077
Token down
Token ID866
Feature activation+7.78e-02

Pos logit contributions
 ju+0.013
 golf+0.012
mons+0.012
 makeup+0.012
 grapes+0.011
Neg logit contributions
ously-0.013
rum-0.013
 despair-0.013
 deeply-0.012
 horror-0.012
Token his
Token ID465
Feature activation+5.65e-02

Pos logit contributions
rum+0.161
ously+0.160
 despair+0.148
urbed+0.143
 horror+0.143
Neg logit contributions
 golf-0.103
 ju-0.096
 contest-0.095
 makeup-0.092
 quiz-0.088
Token spine
Token ID19656
Feature activation+1.85e-01

Pos logit contributions
rum+0.108
ously+0.105
 despair+0.095
 horror+0.093
urbed+0.091
Neg logit contributions
 swing-0.090
 golf-0.087
 quiz-0.083
 makeup-0.082
coat-0.081
Token.
Token ID13
Feature activation+3.67e-02

Pos logit contributions
rum+0.412
 horror+0.406
 despair+0.405
ously+0.392
Flash+0.392
Neg logit contributions
 golf-0.163
 makeup-0.151
 swing-0.148
 butter-0.141
 which-0.138
Token He
Token ID679
Feature activation+2.87e-02

Pos logit contributions
ously+0.045
rum+0.043
 despair+0.039
 deeply+0.038
 horror+0.036
Neg logit contributions
 ju-0.080
 golf-0.079
 makeup-0.076
mons-0.075
 grapes-0.073
Token looked
Token ID3114
Feature activation-3.83e-03

Pos logit contributions
rum+0.066
ously+0.063
 despair+0.058
 horror+0.057
 deeply+0.057
Neg logit contributions
 ju-0.031
 golf-0.030
 makeup-0.030
 butter-0.028
 paints-0.027
Token around
Token ID1088
Feature activation+5.56e-02

Pos logit contributions
 ju+0.005
 golf+0.004
 makeup+0.004
 paints+0.004
 butter+0.004
Neg logit contributions
rum-0.008
ously-0.008
 despair-0.008
 horror-0.007
 plea-0.007
Token but
Token ID475
Feature activation+1.67e-02

Pos logit contributions
rum+0.125
ously+0.117
 despair+0.110
aced+0.109
 jaws+0.107
Neg logit contributions
 golf-0.067
 makeup-0.060
 ju-0.059
 quiz-0.058
 contest-0.057
Token a
Token ID257
Feature activation+5.39e-03

Pos logit contributions
ously+0.071
rum+0.070
 despair+0.062
 horror+0.060
urbed+0.060
Neg logit contributions
 ju-0.047
 golf-0.043
 makeup-0.040
 butter-0.040
 contest-0.039
Token chill
Token ID20493
Feature activation+6.12e-02

Pos logit contributions
rum+0.012
ously+0.011
uzz+0.010
aced+0.010
sey+0.010
Neg logit contributions
 ju-0.007
 butter-0.006
 golf-0.006
 contest-0.006
 competition-0.006
Token going
Token ID1016
Feature activation-6.27e-03

Pos logit contributions
ously+0.103
rum+0.100
 despair+0.094
 deeply+0.093
 horror+0.090
Neg logit contributions
 ju-0.101
 golf-0.097
 makeup-0.095
mons-0.094
 contest-0.091
Token down
Token ID866
Feature activation+8.54e-02

Pos logit contributions
mons+0.011
 ju+0.011
 makeup+0.011
atmeal+0.011
 grapes+0.011
Neg logit contributions
ously-0.009
 deeply-0.009
rum-0.009
 silently-0.008
 despair-0.008
Token their
Token ID511
Feature activation+6.54e-02

Pos logit contributions
rum+0.159
ously+0.154
 despair+0.149
 horror+0.143
 heartbeat+0.140
Neg logit contributions
 ju-0.124
 golf-0.123
 contest-0.115
 makeup-0.110
 competition-0.110
Token spine
Token ID19656
Feature activation+1.84e-01

Pos logit contributions
ously+0.120
rum+0.118
 despair+0.110
 deeply+0.109
 horror+0.106
Neg logit contributions
 ju-0.099
 golf-0.094
 makeup-0.091
mons-0.089
 contest-0.088
Token.
Token ID13
Feature activation+4.55e-02

Pos logit contributions
rum+0.390
 horror+0.369
ously+0.368
 despair+0.366
Flash+0.345
Neg logit contributions
 golf-0.200
 ju-0.188
 butter-0.177
 swing-0.169
 makeup-0.168
Token This
Token ID770
Feature activation-5.09e-02

Pos logit contributions
rum+0.080
ously+0.079
 despair+0.073
 horror+0.068
 jaws+0.067
Neg logit contributions
 golf-0.074
 ju-0.072
 contest-0.068
mons-0.067
 grapes-0.066
Token was
Token ID373
Feature activation+3.41e-02

Pos logit contributions
 ju+0.076
 golf+0.073
 makeup+0.071
mons+0.071
 grapes+0.068
Neg logit contributions
ously-0.094
rum-0.092
 despair-0.088
 deeply-0.087
 horror-0.083
Token the
Token ID262
Feature activation-6.40e-03

Pos logit contributions
ously+0.066
rum+0.064
 despair+0.060
 deeply+0.059
 horror+0.058
Neg logit contributions
 ju-0.049
 golf-0.047
 makeup-0.045
mons-0.044
 contest-0.043
Token wisdom
Token ID11501
Feature activation+7.27e-02

Pos logit contributions
 ju+0.009
 golf+0.008
 makeup+0.008
mons+0.008
 contest+0.007
Neg logit contributions
ously-0.013
rum-0.013
 despair-0.012
 deeply-0.012
 horror-0.012
Token to
Token ID284
Feature activation+3.90e-02

Pos logit contributions
rum+0.264
 despair+0.248
ously+0.244
 horror+0.241
 deeply+0.230
Neg logit contributions
 ju-0.126
 golf-0.121
mons-0.111
 makeup-0.110
 contest-0.103
Token Ell
Token ID7122
Feature activation+6.16e-02

Pos logit contributions
ously+0.089
aced+0.088
rum+0.083
 jaws+0.081
aned+0.076
Neg logit contributions
 makeup-0.046
 bake-0.040
 store-0.040
 that-0.039
 ju-0.039
Tokena
Token ID64
Feature activation+4.15e-02

Pos logit contributions
ously+0.111
rum+0.110
 despair+0.108
 horror+0.104
 deeply+0.104
Neg logit contributions
 ju-0.090
mons-0.083
 golf-0.083
 butter-0.083
 makeup-0.083
Token before
Token ID878
Feature activation+5.66e-02

Pos logit contributions
ously+0.077
rum+0.076
 despair+0.072
 horror+0.068
 deeply+0.068
Neg logit contributions
 golf-0.058
 ju-0.057
 makeup-0.055
 butter-0.054
 quiz-0.053
Token hopping
Token ID47153
Feature activation-2.10e-03

Pos logit contributions
rum+0.114
ously+0.108
 despair+0.101
aced+0.099
 horror+0.098
Neg logit contributions
 golf-0.081
 makeup-0.074
 ju-0.074
 butter-0.072
 gum-0.071
Token off
Token ID572
Feature activation+1.84e-01

Pos logit contributions
 golf+0.003
 ju+0.003
 makeup+0.003
 competition+0.003
 contest+0.003
Neg logit contributions
rum-0.004
 horror-0.004
 despair-0.004
ously-0.004
 heartbeat-0.004
Token to
Token ID284
Feature activation+5.43e-02

Pos logit contributions
rum+0.405
ously+0.384
 despair+0.370
aced+0.356
 heartbeat+0.356
Neg logit contributions
 golf-0.225
 contest-0.212
 ju-0.208
 makeup-0.207
 competition-0.203
Token find
Token ID1064
Feature activation-1.94e-02

Pos logit contributions
ously+0.117
rum+0.112
aced+0.105
aned+0.104
lessly+0.100
Neg logit contributions
 golf-0.071
 swing-0.067
 makeup-0.066
 quiz-0.062
 competition-0.061
Token his
Token ID465
Feature activation+3.02e-02

Pos logit contributions
 ju+0.026
 makeup+0.024
 golf+0.024
 butter+0.024
 grapes+0.023
Neg logit contributions
rum-0.040
ously-0.039
 despair-0.036
aned-0.035
urbed-0.035
Token family
Token ID1641
Feature activation+4.28e-02

Pos logit contributions
ously+0.045
rum+0.043
 despair+0.043
 deeply+0.042
 horror+0.041
Neg logit contributions
 ju-0.054
mons-0.051
 golf-0.050
atmeal-0.050
 makeup-0.049
Token.
Token ID13
Feature activation+3.81e-02

Pos logit contributions
ously+0.063
rum+0.059
 deeply+0.058
 despair+0.057
 horror+0.053
Neg logit contributions
 ju-0.079
mons-0.077
 golf-0.075
 makeup-0.073
atmeal-0.072
Token time
Token ID640
Feature activation-2.41e-03

Pos logit contributions
mons+0.087
 ju+0.085
 golf+0.084
atmeal+0.082
 makeup+0.079
Neg logit contributions
ously-0.076
 despair-0.072
 deeply-0.072
 horror-0.067
 drifted-0.065
Token,
Token ID11
Feature activation-4.84e-03

Pos logit contributions
 ju+0.003
 makeup+0.003
 golf+0.002
 butter+0.002
 contest+0.002
Neg logit contributions
rum-0.005
ously-0.005
 despair-0.005
 deeply-0.005
 horror-0.005
Token the
Token ID262
Feature activation-2.28e-03

Pos logit contributions
 ju+0.005
 makeup+0.005
 golf+0.005
 butter+0.005
 grapes+0.004
Neg logit contributions
rum-0.011
ously-0.011
 despair-0.010
 jaws-0.010
 horror-0.010
Token mug
Token ID25152
Feature activation-9.69e-03

Pos logit contributions
atmeal+0.003
mons+0.003
 golf+0.003
 ju+0.003
paper+0.003
Neg logit contributions
 despair-0.004
 horror-0.004
 deeply-0.004
 fear-0.004
ously-0.004
Token flew
Token ID13112
Feature activation+2.21e-02

Pos logit contributions
 ju+0.015
 golf+0.014
 makeup+0.014
 contest+0.013
 butter+0.013
Neg logit contributions
rum-0.018
ously-0.017
 despair-0.016
 deeply-0.016
 horror-0.015
Token off
Token ID572
Feature activation+1.84e-01

Pos logit contributions
rum+0.042
ously+0.038
 despair+0.037
 horror+0.036
 heartbeat+0.034
Neg logit contributions
 ju-0.035
 golf-0.034
 makeup-0.031
 contest-0.031
 competition-0.031
Token the
Token ID262
Feature activation+2.32e-02

Pos logit contributions
rum+0.403
ously+0.369
 despair+0.355
 heartbeat+0.343
 jaws+0.337
Neg logit contributions
 golf-0.229
 contest-0.228
 ju-0.222
 competition-0.219
 makeup-0.218
Token path
Token ID3108
Feature activation+4.02e-02

Pos logit contributions
ously+0.039
rum+0.038
 despair+0.036
 deeply+0.035
 horror+0.034
Neg logit contributions
 ju-0.039
 golf-0.037
 makeup-0.036
mons-0.036
 contest-0.035
Token,
Token ID11
Feature activation+9.10e-03

Pos logit contributions
rum+0.078
ously+0.072
 despair+0.071
 heartbeat+0.068
 horror+0.067
Neg logit contributions
 golf-0.056
 ju-0.055
 contest-0.054
 competition-0.053
 makeup-0.052
Token just
Token ID655
Feature activation+1.29e-01

Pos logit contributions
rum+0.016
ously+0.016
 despair+0.015
 deeply+0.014
 horror+0.014
Neg logit contributions
 ju-0.014
 golf-0.014
 makeup-0.013
 butter-0.013
mons-0.013
Token like
Token ID588
Feature activation-3.82e-02

Pos logit contributions
rum+0.260
ously+0.257
 despair+0.238
 deeply+0.235
 horror+0.231
Neg logit contributions
 ju-0.171
 golf-0.161
 makeup-0.159
 butter-0.149
mons-0.148
Token the
Token ID262
Feature activation+1.36e-02

Pos logit contributions
ously+0.250
rum+0.248
aced+0.220
aned+0.219
 deeply+0.218
Neg logit contributions
 golf-0.139
 makeup-0.129
 ju-0.121
 grapes-0.121
 paints-0.118
Token air
Token ID1633
Feature activation+5.27e-02

Pos logit contributions
ously+0.025
rum+0.025
 despair+0.024
 deeply+0.024
 horror+0.023
Neg logit contributions
 ju-0.019
 golf-0.018
mons-0.018
 makeup-0.018
atmeal-0.017
Token and
Token ID290
Feature activation+1.08e-02

Pos logit contributions
rum+0.096
ously+0.091
 despair+0.088
 horror+0.084
 deeply+0.084
Neg logit contributions
 golf-0.078
 ju-0.078
 makeup-0.074
 contest-0.072
 butter-0.072
Token it
Token ID340
Feature activation+3.54e-03

Pos logit contributions
rum+0.021
ously+0.020
 deeply+0.018
 despair+0.018
 horror+0.018
Neg logit contributions
 ju-0.016
 golf-0.016
 makeup-0.015
 butter-0.014
mons-0.014
Token soared
Token ID35292
Feature activation+4.82e-02

Pos logit contributions
rum+0.007
ously+0.007
 despair+0.006
 deeply+0.006
 horror+0.006
Neg logit contributions
 ju-0.005
 golf-0.005
 makeup-0.005
 contest-0.005
 butter-0.005
Token off
Token ID572
Feature activation+1.84e-01

Pos logit contributions
rum+0.102
ously+0.095
 despair+0.093
 horror+0.092
 heartbeat+0.086
Neg logit contributions
 ju-0.067
 golf-0.065
 contest-0.059
 competition-0.058
 makeup-0.056
Token under
Token ID739
Feature activation+1.11e-01

Pos logit contributions
rum+0.423
ously+0.406
 despair+0.376
 heartbeat+0.366
aced+0.358
Neg logit contributions
 golf-0.223
 contest-0.219
 competition-0.204
 ju-0.203
 grapes-0.185
Token the
Token ID262
Feature activation+1.33e-02

Pos logit contributions
rum+0.253
ously+0.250
 deeply+0.219
aced+0.218
 despair+0.217
Neg logit contributions
 golf-0.135
 makeup-0.123
 grapes-0.118
 ju-0.117
 contest-0.117
Token rainbow
Token ID27223
Feature activation+1.70e-02

Pos logit contributions
ously+0.029
rum+0.028
 despair+0.027
 deeply+0.026
 horror+0.026
Neg logit contributions
 ju-0.016
 golf-0.015
 makeup-0.014
mons-0.014
 contest-0.014
Token.
Token ID13
Feature activation+4.17e-02

Pos logit contributions
ously+0.027
rum+0.027
 despair+0.025
 deeply+0.025
 horror+0.024
Neg logit contributions
 ju-0.029
 golf-0.028
 makeup-0.027
 contest-0.027
mons-0.027
Token
Token ID220
Feature activation+2.55e-02

Pos logit contributions
rum+0.082
ously+0.082
 despair+0.075
aced+0.074
 heartbeat+0.073
Neg logit contributions
 golf-0.058
 ju-0.054
mons-0.053
 makeup-0.051
Draw-0.051
Token breeze
Token ID28633
Feature activation+6.64e-02

Pos logit contributions
 despair+0.005
 horror+0.005
 heartbeat+0.005
 fear+0.005
ously+0.005
Neg logit contributions
atmeal-0.005
 ju-0.005
mons-0.005
 golf-0.004
 makeup-0.004
Token blew
Token ID17948
Feature activation+1.96e-02

Pos logit contributions
ously+0.117
rum+0.114
 despair+0.108
 deeply+0.107
 horror+0.103
Neg logit contributions
 ju-0.103
 golf-0.099
 makeup-0.097
mons-0.096
 contest-0.093
Token her
Token ID607
Feature activation+6.07e-02

Pos logit contributions
rum+0.037
ously+0.036
 despair+0.032
 horror+0.031
 deeply+0.031
Neg logit contributions
 ju-0.030
 golf-0.028
 makeup-0.028
 grapes-0.027
 butter-0.027
Token in
Token ID287
Feature activation+1.10e-01

Pos logit contributions
ously+0.105
rum+0.103
 despair+0.095
 deeply+0.095
 horror+0.091
Neg logit contributions
 ju-0.099
 golf-0.096
 makeup-0.096
mons-0.092
 butter-0.091
Token the
Token ID262
Feature activation+2.04e-02

Pos logit contributions
ously+0.237
rum+0.237
aned+0.212
 deeply+0.211
urbed+0.208
Neg logit contributions
 makeup-0.137
 golf-0.129
 grapes-0.124
 ju-0.123
 butter-0.121
Token opposite
Token ID6697
Feature activation+1.82e-01

Pos logit contributions
ously+0.039
rum+0.038
 despair+0.036
 deeply+0.036
 horror+0.034
Neg logit contributions
 ju-0.030
 golf-0.028
 makeup-0.028
mons-0.027
 contest-0.026
Token direction
Token ID4571
Feature activation+9.13e-02

Pos logit contributions
ously+0.288
rum+0.285
 deeply+0.252
 silently+0.247
urbed+0.239
Neg logit contributions
 golf-0.304
 ju-0.286
 swing-0.284
 contest-0.282
 makeup-0.281
Token of
Token ID286
Feature activation+7.63e-02

Pos logit contributions
ously+0.151
rum+0.150
 despair+0.140
 deeply+0.137
 horror+0.133
Neg logit contributions
 ju-0.149
 golf-0.145
 makeup-0.142
mons-0.137
 contest-0.136
Token the
Token ID262
Feature activation+1.80e-02

Pos logit contributions
ously+0.141
rum+0.140
 deeply+0.128
 despair+0.128
 horror+0.123
Neg logit contributions
 ju-0.113
 golf-0.111
 makeup-0.109
 grapes-0.105
mons-0.104
Token cat
Token ID3797
Feature activation+2.13e-02

Pos logit contributions
 despair+0.029
 fear+0.028
 horror+0.028
 screams+0.027
 heartbeat+0.027
Neg logit contributions
 ju-0.030
atmeal-0.028
mons-0.028
cakes-0.025
 redd-0.025
Token.
Token ID13
Feature activation+4.75e-02

Pos logit contributions
ously+0.041
rum+0.040
 despair+0.038
 deeply+0.038
 horror+0.036
Neg logit contributions
 ju-0.030
 golf-0.029
 makeup-0.028
mons-0.027
 contest-0.027
Token,
Token ID11
Feature activation-1.23e-02

Pos logit contributions
 ju+0.007
 golf+0.006
 contest+0.006
 makeup+0.006
mons+0.006
Neg logit contributions
rum-0.007
ously-0.007
 despair-0.007
 deeply-0.007
 horror-0.007
Token leaving
Token ID4305
Feature activation+1.88e-02

Pos logit contributions
 ju+0.019
 golf+0.018
 makeup+0.017
 butter+0.017
 grapes+0.016
Neg logit contributions
rum-0.022
ously-0.022
 despair-0.020
 deeply-0.020
 horror-0.019
Token it
Token ID340
Feature activation+1.04e-02

Pos logit contributions
rum+0.040
ously+0.037
 despair+0.034
urbed+0.033
 heartbeat+0.033
Neg logit contributions
 golf-0.024
 ju-0.023
 makeup-0.022
 butter-0.022
 grapes-0.022
Token to
Token ID284
Feature activation+4.67e-02

Pos logit contributions
ously+0.019
rum+0.019
 despair+0.017
 deeply+0.017
 horror+0.017
Neg logit contributions
 ju-0.016
 golf-0.015
 makeup-0.014
 contest-0.014
mons-0.014
Token its
Token ID663
Feature activation+8.16e-02

Pos logit contributions
ously+0.096
rum+0.093
aced+0.083
aned+0.083
urbed+0.082
Neg logit contributions
 golf-0.073
 ju-0.064
 swing-0.062
 contest-0.059
 competition-0.059
Token former
Token ID1966
Feature activation+1.82e-01

Pos logit contributions
ously+0.156
rum+0.153
aced+0.128
urbed+0.127
lessly+0.126
Neg logit contributions
 golf-0.133
 ju-0.129
 makeup-0.115
 competition-0.112
 swing-0.111
Token state
Token ID1181
Feature activation+6.58e-02

Pos logit contributions
ously+0.309
rum+0.273
 Flash+0.258
Flash+0.256
 despair+0.248
Neg logit contributions
 golf-0.298
 ju-0.297
 makeup-0.281
 grapes-0.272
 paints-0.267
Token.
Token ID13
Feature activation+4.57e-02

Pos logit contributions
rum+0.110
ously+0.108
 despair+0.103
 horror+0.099
 deeply+0.099
Neg logit contributions
 ju-0.106
 golf-0.103
 makeup-0.098
 butter-0.095
mons-0.094
Token
Token ID220
Feature activation+2.94e-02

Pos logit contributions
ously+0.091
rum+0.091
 heartbeat+0.084
aced+0.083
 despair+0.081
Neg logit contributions
 golf-0.062
 ju-0.059
Draw-0.053
mons-0.053
 contest-0.052
Token
Token ID198
Feature activation+4.26e-02

Pos logit contributions
ously+0.053
 despair+0.051
rum+0.050
 horror+0.048
 heartbeat+0.047
Neg logit contributions
 golf-0.044
mons-0.043
 ju-0.042
 makeup-0.040
 paints-0.039
Token
Token ID198
Feature activation+4.06e-02

Pos logit contributions
ously+0.077
rum+0.074
 despair+0.073
 heartbeat+0.071
 jaws+0.071
Neg logit contributions
 golf-0.064
 ju-0.062
Draw-0.059
mons-0.057
 contest-0.057
Token last
Token ID938
Feature activation+1.11e-01

Pos logit contributions
ously+0.081
rum+0.078
 despair+0.075
 deeply+0.075
 horror+0.071
Neg logit contributions
 ju-0.074
 golf-0.071
mons-0.070
 makeup-0.069
 grapes-0.067
Token time
Token ID640
Feature activation+1.88e-02

Pos logit contributions
rum+0.209
ously+0.208
 despair+0.192
 deeply+0.191
 horror+0.187
Neg logit contributions
 ju-0.157
 golf-0.153
 makeup-0.148
 contest-0.145
 competition-0.142
Token,
Token ID11
Feature activation+1.78e-03

Pos logit contributions
rum+0.032
ously+0.032
 despair+0.030
 horror+0.029
 deeply+0.029
Neg logit contributions
 golf-0.030
 ju-0.030
 makeup-0.028
 contest-0.027
mons-0.027
Token before
Token ID878
Feature activation+5.72e-02

Pos logit contributions
rum+0.003
ously+0.003
 despair+0.003
 deeply+0.003
 horror+0.003
Neg logit contributions
 ju-0.003
 golf-0.003
 makeup-0.002
 butter-0.002
 swing-0.002
Token running
Token ID2491
Feature activation+2.00e-02

Pos logit contributions
rum+0.110
ously+0.107
 despair+0.100
aced+0.095
 horror+0.095
Neg logit contributions
 golf-0.087
 ju-0.083
 makeup-0.079
 butter-0.078
 gum-0.078
Token off
Token ID572
Feature activation+1.81e-01

Pos logit contributions
rum+0.035
ously+0.034
 despair+0.033
 horror+0.032
 deeply+0.031
Neg logit contributions
 golf-0.032
 ju-0.030
 makeup-0.030
 contest-0.029
 competition-0.029
Token to
Token ID284
Feature activation+5.44e-02

Pos logit contributions
rum+0.384
ously+0.371
 despair+0.361
aced+0.344
 horror+0.340
Neg logit contributions
 golf-0.233
 contest-0.213
 ju-0.210
 makeup-0.207
 competition-0.203
Token explore
Token ID7301
Feature activation+2.17e-02

Pos logit contributions
ously+0.120
rum+0.114
aced+0.107
aned+0.104
 jaws+0.104
Neg logit contributions
 swing-0.065
 golf-0.063
 makeup-0.062
 contest-0.061
 competition-0.060
Token something
Token ID1223
Feature activation+1.80e-02

Pos logit contributions
ously+0.039
rum+0.039
 despair+0.036
 horror+0.034
 deeply+0.034
Neg logit contributions
 ju-0.033
 golf-0.033
 makeup-0.032
 contest-0.030
 butter-0.030
Token else
Token ID2073
Feature activation+3.96e-02

Pos logit contributions
ously+0.036
rum+0.035
 deeply+0.034
 despair+0.034
 horror+0.032
Neg logit contributions
 ju-0.023
mons-0.022
 golf-0.021
 makeup-0.021
atmeal-0.020
Token.
Token ID13
Feature activation+4.75e-02

Pos logit contributions
ously+0.059
rum+0.056
 despair+0.054
 deeply+0.054
 horror+0.050
Neg logit contributions
 ju-0.072
 golf-0.069
mons-0.069
 makeup-0.068
 grapes-0.066
Token the
Token ID262
Feature activation+1.89e-03

Pos logit contributions
ously+0.060
rum+0.058
 despair+0.056
 deeply+0.055
 horror+0.053
Neg logit contributions
 ju-0.042
 golf-0.040
 makeup-0.039
mons-0.038
 contest-0.037
Token village
Token ID7404
Feature activation+4.00e-02

Pos logit contributions
 despair+0.003
 fear+0.003
 horror+0.003
 plea+0.003
 deeply+0.003
Neg logit contributions
atmeal-0.003
mons-0.003
 ju-0.003
 gloss-0.003
coat-0.003
Token back
Token ID736
Feature activation+1.08e-01

Pos logit contributions
ously+0.069
rum+0.067
 despair+0.064
 deeply+0.063
 horror+0.060
Neg logit contributions
 ju-0.064
 golf-0.061
 makeup-0.059
mons-0.059
 contest-0.057
Token to
Token ID284
Feature activation+3.71e-02

Pos logit contributions
ously+0.198
rum+0.192
 despair+0.183
 deeply+0.179
 horror+0.177
Neg logit contributions
 ju-0.162
 golf-0.156
 makeup-0.151
mons-0.147
 contest-0.146
Token its
Token ID663
Feature activation+7.64e-02

Pos logit contributions
ously+0.078
rum+0.074
 despair+0.067
 deeply+0.066
 horror+0.066
Neg logit contributions
 golf-0.048
 ju-0.048
 makeup-0.046
 contest-0.045
 butter-0.044
Token former
Token ID1966
Feature activation+1.80e-01

Pos logit contributions
ously+0.138
rum+0.132
 despair+0.114
 deeply+0.113
 horror+0.112
Neg logit contributions
 ju-0.128
 golf-0.124
 makeup-0.118
 grapes-0.117
 contest-0.116
Token glory
Token ID13476
Feature activation+9.12e-02

Pos logit contributions
ously+0.296
rum+0.260
 horror+0.240
 despair+0.238
 silently+0.232
Neg logit contributions
 ju-0.314
 golf-0.295
 makeup-0.294
 grapes-0.289
 butter-0.280
Token!"
Token ID2474
Feature activation+1.52e-02

Pos logit contributions
ously+0.143
rum+0.140
 deeply+0.132
 despair+0.130
 horror+0.123
Neg logit contributions
 ju-0.158
 golf-0.153
mons-0.149
 makeup-0.149
 contest-0.144
Token
Token ID220
Feature activation+3.24e-02

Pos logit contributions
rum+0.024
ously+0.024
 despair+0.023
 horror+0.021
 heartbeat+0.021
Neg logit contributions
 ju-0.027
 golf-0.026
 makeup-0.025
mons-0.024
 grapes-0.024
Token
Token ID198
Feature activation+4.02e-02

Pos logit contributions
ously+0.059
 despair+0.057
rum+0.055
 horror+0.052
 heartbeat+0.052
Neg logit contributions
 golf-0.049
 ju-0.048
mons-0.047
 contest-0.043
 gum-0.043
Token
Token ID198
Feature activation+3.77e-02

Pos logit contributions
ously+0.066
rum+0.064
 despair+0.061
 horror+0.058
 deeply+0.058
Neg logit contributions
 ju-0.069
 golf-0.069
 makeup-0.063
 contest-0.063
mons-0.063
Token
Token ID198
Feature activation+3.66e-02

Pos logit contributions
ously+0.060
rum+0.059
 despair+0.056
 deeply+0.052
 heartbeat+0.051
Neg logit contributions
 golf-0.067
 ju-0.066
mons-0.063
 makeup-0.063
 grapes-0.062
Token
Token ID198
Feature activation+3.52e-02

Pos logit contributions
ously+0.059
 despair+0.057
rum+0.056
 horror+0.054
 jaws+0.053
Neg logit contributions
 golf-0.061
 ju-0.060
 makeup-0.057
 contest-0.057
mons-0.057
TokenThe
Token ID464
Feature activation+7.81e-03

Pos logit contributions
rum+0.045
ously+0.044
 deeply+0.041
 despair+0.041
 silently+0.038
Neg logit contributions
 ju-0.070
 makeup-0.069
mons-0.068
atmeal-0.068
 butter-0.066
Token plane
Token ID6614
Feature activation+1.44e-02

Pos logit contributions
 despair+0.010
ously+0.010
 deeply+0.009
 horror+0.009
rum+0.009
Neg logit contributions
 ju-0.016
mons-0.016
atmeal-0.015
 golf-0.015
 makeup-0.015
Token flew
Token ID13112
Feature activation+1.84e-02

Pos logit contributions
rum+0.025
ously+0.025
 despair+0.023
 deeply+0.023
 horror+0.022
Neg logit contributions
 ju-0.023
 golf-0.022
 contest-0.021
 makeup-0.021
mons-0.021
Token off
Token ID572
Feature activation+1.80e-01

Pos logit contributions
rum+0.037
ously+0.036
 despair+0.034
 horror+0.034
 heartbeat+0.032
Neg logit contributions
 ju-0.027
 golf-0.025
 contest-0.023
 competition-0.023
 grapes-0.023
Token,
Token ID11
Feature activation+7.23e-03

Pos logit contributions
rum+0.448
ously+0.426
 despair+0.411
 heartbeat+0.395
 plea+0.388
Neg logit contributions
 contest-0.196
 golf-0.187
 competition-0.173
 ju-0.152
 grapes-0.151
Token feeling
Token ID4203
Feature activation+5.53e-02

Pos logit contributions
rum+0.013
ously+0.013
 despair+0.012
 deeply+0.012
 horror+0.012
Neg logit contributions
 ju-0.011
 golf-0.011
 makeup-0.010
 contest-0.010
 grapes-0.010
Token very
Token ID845
Feature activation+5.67e-02

Pos logit contributions
ously+0.092
rum+0.090
 despair+0.085
 deeply+0.085
 horror+0.081
Neg logit contributions
 ju-0.091
 golf-0.088
 makeup-0.086
mons-0.086
 contest-0.083
Token proud
Token ID6613
Feature activation+4.50e-02

Pos logit contributions
ously+0.127
rum+0.124
 despair+0.113
 horror+0.112
sey+0.111
Neg logit contributions
 ju-0.073
 golf-0.066
 butter-0.058
 makeup-0.058
 contest-0.057
Token of
Token ID286
Feature activation+4.94e-02

Pos logit contributions
ously+0.069
rum+0.067
 despair+0.063
 deeply+0.062
 horror+0.060
Neg logit contributions
 ju-0.081
 golf-0.078
 makeup-0.076
mons-0.076
 contest-0.074
Token kissed
Token ID27946
Feature activation+8.78e-02

Pos logit contributions
rum+0.038
ously+0.037
 despair+0.036
 deeply+0.035
 horror+0.035
Neg logit contributions
 ju-0.026
 golf-0.025
 makeup-0.024
 butter-0.023
mons-0.023
Token him
Token ID683
Feature activation+4.27e-02

Pos logit contributions
ously+0.128
rum+0.125
 despair+0.116
 deeply+0.115
 horror+0.110
Neg logit contributions
 ju-0.164
 golf-0.158
 makeup-0.155
mons-0.153
 contest-0.150
Token goodbye
Token ID24829
Feature activation+1.21e-01

Pos logit contributions
ously+0.076
rum+0.073
 despair+0.070
 horror+0.067
 deeply+0.064
Neg logit contributions
 ju-0.066
 golf-0.065
 makeup-0.063
 butter-0.060
 quiz-0.059
Token before
Token ID878
Feature activation+6.29e-02

Pos logit contributions
rum+0.238
 despair+0.227
ously+0.227
 horror+0.219
 jaws+0.204
Neg logit contributions
 ju-0.163
 golf-0.162
 makeup-0.148
mons-0.141
 butter-0.139
Token running
Token ID2491
Feature activation+2.17e-02

Pos logit contributions
rum+0.119
ously+0.116
 despair+0.109
 deeply+0.107
 horror+0.104
Neg logit contributions
 golf-0.093
 ju-0.089
 makeup-0.086
 grapes-0.084
 butter-0.082
Token off
Token ID572
Feature activation+1.80e-01

Pos logit contributions
rum+0.036
ously+0.035
 despair+0.033
 horror+0.032
 deeply+0.032
Neg logit contributions
 golf-0.037
 makeup-0.035
 ju-0.035
 contest-0.033
 competition-0.033
Token with
Token ID351
Feature activation+5.69e-02

Pos logit contributions
rum+0.377
ously+0.373
 despair+0.359
aced+0.349
 heartbeat+0.343
Neg logit contributions
 golf-0.233
 ju-0.210
 makeup-0.206
 contest-0.205
 competition-0.196
Token a
Token ID257
Feature activation-6.10e-03

Pos logit contributions
ously+0.094
rum+0.093
 despair+0.086
 deeply+0.085
 horror+0.082
Neg logit contributions
 ju-0.095
 golf-0.093
 makeup-0.090
mons-0.087
 grapes-0.087
Token big
Token ID1263
Feature activation-1.52e-02

Pos logit contributions
 ju+0.009
 golf+0.009
 makeup+0.008
 contest+0.008
 butter+0.008
Neg logit contributions
ously-0.011
rum-0.011
 despair-0.010
 deeply-0.010
 horror-0.010
Token smile
Token ID8212
Feature activation+1.13e-01

Pos logit contributions
 ju+0.031
 golf+0.031
 makeup+0.030
mons+0.029
 contest+0.029
Neg logit contributions
ously-0.019
rum-0.019
 despair-0.017
 deeply-0.017
 horror-0.016
Token.
Token ID13
Feature activation+4.40e-02

Pos logit contributions
ously+0.196
rum+0.193
 despair+0.186
 deeply+0.181
 horror+0.177
Neg logit contributions
 ju-0.173
 golf-0.168
 makeup-0.164
mons-0.159
 butter-0.157
Token night
Token ID1755
Feature activation+6.47e-02

Pos logit contributions
ously+0.194
rum+0.193
 jaws+0.177
aned+0.174
 heartbeat+0.172
Neg logit contributions
 ju-0.113
 contest-0.113
 competition-0.111
 golf-0.108
 makeup-0.105
Token.
Token ID13
Feature activation+4.47e-02

Pos logit contributions
rum+0.118
ously+0.117
 despair+0.108
 horror+0.105
 deeply+0.103
Neg logit contributions
 ju-0.097
 golf-0.094
 makeup-0.090
 contest-0.088
 butter-0.088
Token As
Token ID1081
Feature activation+4.77e-02

Pos logit contributions
ously+0.077
rum+0.077
 heartbeat+0.072
 jaws+0.070
 despair+0.070
Neg logit contributions
 golf-0.064
 ju-0.060
 basketball-0.058
 makeup-0.057
 baseball-0.057
Token she
Token ID673
Feature activation+5.28e-02

Pos logit contributions
rum+0.111
ously+0.108
aced+0.098
aned+0.096
 heartbeat+0.096
Neg logit contributions
 golf-0.053
 ju-0.048
 contest-0.048
 competition-0.047
 gum-0.046
Token drifted
Token ID38648
Feature activation+1.19e-01

Pos logit contributions
rum+0.112
ously+0.103
 horror+0.096
 despair+0.096
lessly+0.094
Neg logit contributions
 butter-0.062
 baked-0.061
atmeal-0.060
 ju-0.060
 swing-0.059
Token off
Token ID572
Feature activation+1.80e-01

Pos logit contributions
rum+0.209
ously+0.186
 horror+0.169
 despair+0.167
 disbelief+0.161
Neg logit contributions
 makeup-0.173
 golf-0.170
 competition-0.168
 contest-0.167
 quiz-0.156
Token to
Token ID284
Feature activation+5.99e-02

Pos logit contributions
rum+0.467
ously+0.427
 despair+0.410
 horror+0.399
aced+0.395
Neg logit contributions
 golf-0.160
 contest-0.157
 competition-0.149
 ju-0.146
 makeup-0.145
Token sleep
Token ID3993
Feature activation+6.47e-02

Pos logit contributions
ously+0.125
rum+0.122
aned+0.118
aced+0.110
umbled+0.109
Neg logit contributions
 swing-0.076
 makeup-0.075
 quiz-0.073
 golf-0.071
 party-0.070
Token,
Token ID11
Feature activation+1.19e-02

Pos logit contributions
rum+0.140
ously+0.132
 despair+0.123
 jaws+0.121
 horror+0.121
Neg logit contributions
 contest-0.075
 golf-0.074
 competition-0.074
 makeup-0.073
 quiz-0.072
Token she
Token ID673
Feature activation+3.40e-02

Pos logit contributions
rum+0.026
ously+0.025
 jaws+0.023
aced+0.023
 heartbeat+0.022
Neg logit contributions
 ju-0.014
 golf-0.013
 makeup-0.013
 competition-0.013
 contest-0.013
Token thought
Token ID1807
Feature activation-1.38e-03

Pos logit contributions
rum+0.070
ously+0.065
 jaws+0.059
 despair+0.059
 horror+0.058
Neg logit contributions
 ju-0.043
 butter-0.040
 paints-0.040
 makeup-0.038
 competition-0.038
Token kind
Token ID1611
Feature activation+3.78e-02

Pos logit contributions
ously+0.013
rum+0.013
 despair+0.013
 deeply+0.012
 horror+0.012
Neg logit contributions
 ju-0.012
mons-0.012
 golf-0.012
 makeup-0.012
atmeal-0.011
Token animal
Token ID5044
Feature activation+3.77e-02

Pos logit contributions
rum+0.077
ously+0.075
 despair+0.071
 drifted+0.068
 deeply+0.068
Neg logit contributions
 golf-0.047
 ju-0.043
 contest-0.043
 file-0.042
 grapes-0.042
Token before
Token ID878
Feature activation+5.84e-02

Pos logit contributions
rum+0.077
ously+0.074
 despair+0.070
 heartbeat+0.069
 horror+0.066
Neg logit contributions
 golf-0.045
 ju-0.041
 competition-0.039
 butter-0.039
 swing-0.039
Token scam
Token ID19126
Feature activation+5.95e-02

Pos logit contributions
rum+0.115
ously+0.109
 despair+0.100
aced+0.098
 deeply+0.097
Neg logit contributions
 golf-0.088
 ju-0.081
 butter-0.078
 contest-0.077
 makeup-0.076
Tokenpering
Token ID21255
Feature activation+2.13e-02

Pos logit contributions
ously+0.075
rum+0.074
 despair+0.066
 heartbeat+0.064
 deeply+0.064
Neg logit contributions
 golf-0.125
 contest-0.116
 ju-0.115
 competition-0.114
 makeup-0.114
Token off
Token ID572
Feature activation+1.80e-01

Pos logit contributions
rum+0.040
ously+0.038
 despair+0.036
 deeply+0.035
 heartbeat+0.035
Neg logit contributions
 golf-0.031
 competition-0.029
 contest-0.029
 makeup-0.027
 ju-0.026
Token to
Token ID284
Feature activation+4.92e-02

Pos logit contributions
rum+0.421
ously+0.394
 despair+0.385
 heartbeat+0.374
aced+0.372
Neg logit contributions
 golf-0.206
 contest-0.180
 competition-0.172
 ju-0.163
 makeup-0.157
Token explore
Token ID7301
Feature activation+2.17e-02

Pos logit contributions
ously+0.110
rum+0.107
aned+0.101
aced+0.100
urbed+0.097
Neg logit contributions
 golf-0.062
 swing-0.056
 plant-0.053
 makeup-0.052
 contest-0.052
Token the
Token ID262
Feature activation+7.19e-03

Pos logit contributions
rum+0.043
ously+0.042
 despair+0.039
 heartbeat+0.038
 jaws+0.038
Neg logit contributions
 golf-0.030
 ju-0.029
 makeup-0.027
 grapes-0.026
 contest-0.026
Token forest
Token ID8222
Feature activation+3.17e-02

Pos logit contributions
ously+0.011
 despair+0.011
rum+0.011
 deeply+0.011
 horror+0.010
Neg logit contributions
 ju-0.012
mons-0.012
 makeup-0.012
atmeal-0.011
 golf-0.011
Token again
Token ID757
Feature activation+1.36e-01

Pos logit contributions
rum+0.054
ously+0.054
 despair+0.050
 horror+0.048
 deeply+0.048
Neg logit contributions
 ju-0.050
 golf-0.049
 makeup-0.046
 contest-0.046
 competition-0.045
Token move
Token ID1445
Feature activation+3.61e-02

Pos logit contributions
rum+0.125
ously+0.120
 jaws+0.110
 deeply+0.109
 despair+0.108
Neg logit contributions
 golf-0.068
 ju-0.065
 swing-0.065
 competition-0.061
 contest-0.060
Token again
Token ID757
Feature activation+1.37e-01

Pos logit contributions
rum+0.065
ously+0.062
 despair+0.059
 horror+0.057
 deeply+0.056
Neg logit contributions
 ju-0.055
 golf-0.054
 makeup-0.052
 contest-0.050
 competition-0.049
Token and
Token ID290
Feature activation+2.70e-02

Pos logit contributions
rum+0.253
ously+0.249
 despair+0.233
 horror+0.228
 deeply+0.225
Neg logit contributions
 golf-0.200
 ju-0.198
 makeup-0.188
 contest-0.183
 competition-0.180
Token she
Token ID673
Feature activation+4.68e-02

Pos logit contributions
rum+0.054
ously+0.050
 deeply+0.046
 despair+0.044
 jaws+0.044
Neg logit contributions
 ju-0.040
 golf-0.039
 contest-0.037
 competition-0.037
 makeup-0.036
Token ran
Token ID4966
Feature activation-4.65e-05

Pos logit contributions
rum+0.099
ously+0.087
 despair+0.086
 deeply+0.084
 heartbeat+0.084
Neg logit contributions
 ju-0.059
 swing-0.057
 golf-0.057
 competition-0.055
 contest-0.053
Token off
Token ID572
Feature activation+1.80e-01

Pos logit contributions
 golf+0.000
 makeup+0.000
 competition+0.000
 contest+0.000
 ju+0.000
Neg logit contributions
rum-0.000
ously-0.000
 despair-0.000
 horror-0.000
 heartbeat-0.000
Token in
Token ID287
Feature activation+1.20e-01

Pos logit contributions
rum+0.440
ously+0.406
 despair+0.397
aced+0.389
 jaws+0.386
Neg logit contributions
 golf-0.199
 contest-0.183
 competition-0.176
 swing-0.171
 makeup-0.162
Token search
Token ID2989
Feature activation-1.88e-02

Pos logit contributions
rum+0.246
ously+0.241
aned+0.233
aced+0.222
urbed+0.221
Neg logit contributions
 makeup-0.146
 golf-0.146
 competition-0.135
 contest-0.133
 swing-0.132
Token of
Token ID286
Feature activation+7.04e-02

Pos logit contributions
 ju+0.026
 golf+0.025
 makeup+0.025
 contest+0.024
 competition+0.024
Neg logit contributions
rum-0.037
ously-0.034
 despair-0.033
 jaws-0.032
 heartbeat-0.031
Token her
Token ID607
Feature activation+6.45e-02

Pos logit contributions
ously+0.136
rum+0.135
 despair+0.123
 deeply+0.121
urbed+0.117
Neg logit contributions
 ju-0.101
 golf-0.098
 makeup-0.096
 butter-0.091
 grapes-0.091
Token next
Token ID1306
Feature activation+2.19e-03

Pos logit contributions
ously+0.119
rum+0.116
 despair+0.111
 deeply+0.110
 horror+0.106
Neg logit contributions
 ju-0.096
 golf-0.090
mons-0.088
 makeup-0.088
 grapes-0.085
Token the
Token ID262
Feature activation+8.14e-03

Pos logit contributions
rum+0.089
ously+0.089
 despair+0.081
aced+0.080
 deeply+0.079
Neg logit contributions
 ju-0.054
 golf-0.053
 makeup-0.053
 butter-0.051
 grapes-0.050
Token duck
Token ID22045
Feature activation+1.84e-02

Pos logit contributions
 despair+0.012
 deeply+0.012
ously+0.012
 horror+0.012
rum+0.012
Neg logit contributions
 ju-0.013
atmeal-0.013
mons-0.013
 newspapers-0.013
 quiz-0.013
Token as
Token ID355
Feature activation+2.94e-02

Pos logit contributions
rum+0.035
ously+0.035
 despair+0.033
 horror+0.032
 deeply+0.030
Neg logit contributions
 golf-0.025
 ju-0.025
 butter-0.023
 makeup-0.023
 contest-0.023
Token it
Token ID340
Feature activation+1.14e-02

Pos logit contributions
ously+0.066
rum+0.065
 despair+0.062
 deeply+0.060
 horror+0.060
Neg logit contributions
 ju-0.032
 golf-0.031
 makeup-0.030
mons-0.029
 contest-0.028
Token walked
Token ID6807
Feature activation+1.88e-02

Pos logit contributions
rum+0.022
ously+0.021
 despair+0.020
 horror+0.019
 deeply+0.019
Neg logit contributions
 ju-0.017
 golf-0.016
 butter-0.015
 contest-0.015
mons-0.015
Token off
Token ID572
Feature activation+1.80e-01

Pos logit contributions
rum+0.035
ously+0.032
 despair+0.032
 horror+0.031
 heartbeat+0.029
Neg logit contributions
 golf-0.030
 ju-0.029
 makeup-0.027
 competition-0.027
 contest-0.027
Token,
Token ID11
Feature activation+5.58e-04

Pos logit contributions
rum+0.454
ously+0.424
 despair+0.416
aced+0.405
 jaws+0.397
Neg logit contributions
 golf-0.177
 contest-0.168
 competition-0.159
 makeup-0.151
 ju-0.139
Token and
Token ID290
Feature activation+1.44e-02

Pos logit contributions
rum+0.001
ously+0.001
 despair+0.001
 deeply+0.001
 jaws+0.001
Neg logit contributions
 ju-0.001
 golf-0.001
 makeup-0.001
 butter-0.001
 swing-0.001
Token Kate
Token ID16693
Feature activation+5.13e-02

Pos logit contributions
rum+0.031
ously+0.029
 jaws+0.026
 deeply+0.026
 despair+0.026
Neg logit contributions
 ju-0.019
 golf-0.017
 makeup-0.017
 competition-0.016
 contest-0.016
Token went
Token ID1816
Feature activation-2.32e-04

Pos logit contributions
rum+0.109
ously+0.095
 despair+0.095
 jaws+0.095
 horror+0.093
Neg logit contributions
 ju-0.061
 butter-0.060
 swing-0.057
 baked-0.056
 competition-0.055
Token back
Token ID736
Feature activation+1.03e-01

Pos logit contributions
 golf+0.000
 makeup+0.000
 ju+0.000
 contest+0.000
 quiz+0.000
Neg logit contributions
rum-0.000
ously-0.000
 despair-0.000
 horror-0.000
aced-0.000
Token Lily
Token ID20037
Feature activation-3.53e-02

Pos logit contributions
 ju+0.064
 makeup+0.064
 golf+0.063
 butter+0.061
 mixer+0.061
Neg logit contributions
ously-0.081
rum-0.080
aced-0.076
 deeply-0.072
 despair-0.072
Token said
Token ID531
Feature activation+3.42e-03

Pos logit contributions
 ju+0.046
 makeup+0.045
 golf+0.043
 butter+0.042
 grapes+0.041
Neg logit contributions
rum-0.072
ously-0.069
 despair-0.065
 deeply-0.064
urbed-0.062
Token.
Token ID13
Feature activation-1.77e-02

Pos logit contributions
rum+0.006
 despair+0.006
ously+0.006
 jaws+0.006
 horror+0.006
Neg logit contributions
 makeup-0.005
 golf-0.005
 butter-0.005
 ju-0.004
 gum-0.004
Token
Token ID198
Feature activation-1.19e-02

Pos logit contributions
 golf+0.027
 makeup+0.026
 ju+0.025
 grapes+0.025
 gum+0.024
Neg logit contributions
rum-0.032
ously-0.032
 despair-0.031
aced-0.029
 deeply-0.028
Token
Token ID198
Feature activation-1.36e-02

Pos logit contributions
 golf+0.018
 makeup+0.017
 gum+0.016
Draw+0.016
 paints+0.015
Neg logit contributions
ously-0.022
 despair-0.021
rum-0.021
urbed-0.020
 jaws-0.020
Token"
Token ID1
Feature activation-1.81e-01

Pos logit contributions
 golf+0.017
 ju+0.016
 makeup+0.016
mons+0.015
 contest+0.015
Neg logit contributions
ously-0.029
rum-0.028
 despair-0.027
 deeply-0.027
 horror-0.026
TokenIt
Token ID1026
Feature activation-3.00e-02

Pos logit contributions
 ju+0.258
atmeal+0.254
 golf+0.237
 makeup+0.230
 quiz+0.230
Neg logit contributions
ously-0.334
 despair-0.308
 deeply-0.299
rum-0.299
 silently-0.295
Token's
Token ID338
Feature activation-1.71e-02

Pos logit contributions
 ju+0.037
 golf+0.036
mons+0.034
 makeup+0.034
 grapes+0.032
Neg logit contributions
ously-0.062
rum-0.062
 despair-0.059
 deeply-0.058
 horror-0.056
Token okay
Token ID8788
Feature activation+2.05e-02

Pos logit contributions
 ju+0.023
 golf+0.023
 makeup+0.022
 butter+0.021
 grapes+0.021
Neg logit contributions
ously-0.035
rum-0.033
 despair-0.032
 deeply-0.031
 horror-0.031
Token,
Token ID11
Feature activation-7.14e-02

Pos logit contributions
ously+0.034
rum+0.033
 despair+0.031
 deeply+0.031
 horror+0.030
Neg logit contributions
 ju-0.034
 golf-0.033
 makeup-0.032
mons-0.032
 contest-0.031
Token my
Token ID616
Feature activation-1.77e-02

Pos logit contributions
 ju+0.110
mons+0.108
atmeal+0.106
 golf+0.105
 contest+0.103
Neg logit contributions
ously-0.123
rum-0.119
 despair-0.118
 horror-0.116
 deeply-0.116
Token eat
Token ID4483
Feature activation-2.86e-02

Pos logit contributions
ously+0.067
rum+0.066
 deeply+0.061
 despair+0.061
 horror+0.061
Neg logit contributions
 ju-0.048
 golf-0.047
 makeup-0.044
 contest-0.044
 quiz-0.043
Token them
Token ID606
Feature activation+1.44e-02

Pos logit contributions
 ju+0.043
 golf+0.041
 makeup+0.040
mons+0.040
 contest+0.039
Neg logit contributions
ously-0.051
rum-0.050
 despair-0.048
 deeply-0.047
 horror-0.046
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
ously+0.023
rum+0.022
 despair+0.021
 deeply+0.020
 horror+0.020
Neg logit contributions
 ju-0.025
 golf-0.024
 makeup-0.024
mons-0.024
 contest-0.023
Token
Token ID198
Feature activation-9.89e-03

Pos logit contributions
 golf+0.036
 ju+0.035
 makeup+0.034
mons+0.032
 grapes+0.032
Neg logit contributions
ously-0.036
rum-0.034
 despair-0.031
 deeply-0.031
aced-0.031
Token
Token ID198
Feature activation-1.05e-02

Pos logit contributions
 golf+0.014
 makeup+0.013
Draw+0.013
 ju+0.013
 grapes+0.013
Neg logit contributions
ously-0.019
rum-0.017
 despair-0.017
urbed-0.017
 jaws-0.016
Token"
Token ID1
Feature activation-1.81e-01

Pos logit contributions
 golf+0.012
 ju+0.011
 makeup+0.011
mons+0.010
 contest+0.010
Neg logit contributions
ously-0.023
rum-0.022
 deeply-0.021
 despair-0.021
 horror-0.021
TokenMaybe
Token ID13300
Feature activation-2.45e-02

Pos logit contributions
 ju+0.244
 golf+0.229
 makeup+0.225
atmeal+0.221
 quiz+0.218
Neg logit contributions
ously-0.354
rum-0.340
 despair-0.334
 horror-0.318
 deeply-0.315
Token we
Token ID356
Feature activation-5.84e-02

Pos logit contributions
 ju+0.034
mons+0.031
atmeal+0.031
 makeup+0.031
 golf+0.031
Neg logit contributions
ously-0.046
rum-0.045
 despair-0.043
 deeply-0.042
 horror-0.042
Token can
Token ID460
Feature activation-5.78e-02

Pos logit contributions
 ju+0.072
 golf+0.068
 makeup+0.067
mons+0.066
atmeal+0.063
Neg logit contributions
ously-0.121
rum-0.118
 despair-0.115
 deeply-0.113
 horror-0.111
Token fly
Token ID6129
Feature activation-1.01e-02

Pos logit contributions
 ju+0.082
 golf+0.078
 makeup+0.076
mons+0.075
 contest+0.073
Neg logit contributions
ously-0.111
rum-0.108
 despair-0.103
 deeply-0.102
 horror-0.098
Token with
Token ID351
Feature activation+2.06e-02

Pos logit contributions
 ju+0.016
 golf+0.015
mons+0.015
 makeup+0.015
atmeal+0.014
Neg logit contributions
ously-0.017
rum-0.017
 despair-0.016
 deeply-0.016
 horror-0.015
Token Ben
Token ID3932
Feature activation-4.50e-02

Pos logit contributions
 makeup+0.010
 golf+0.010
mons+0.009
 ju+0.009
 grapes+0.009
Neg logit contributions
rum-0.011
ously-0.011
aced-0.010
 deeply-0.010
 despair-0.010
Token said
Token ID531
Feature activation-1.86e-03

Pos logit contributions
 ju+0.060
 makeup+0.059
 golf+0.057
 butter+0.055
 grapes+0.055
Neg logit contributions
rum-0.087
ously-0.085
 despair-0.078
 deeply-0.076
urbed-0.074
Token.
Token ID13
Feature activation-2.04e-02

Pos logit contributions
 golf+0.002
 makeup+0.002
 ju+0.002
 contest+0.002
mons+0.002
Neg logit contributions
rum-0.003
ously-0.003
 despair-0.003
 horror-0.003
 jaws-0.003
Token
Token ID198
Feature activation-9.20e-03

Pos logit contributions
 golf+0.030
 ju+0.028
 gum+0.027
 makeup+0.027
 baseball+0.027
Neg logit contributions
rum-0.037
ously-0.037
 despair-0.036
aced-0.035
urbed-0.034
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
 golf+0.012
 makeup+0.011
Draw+0.011
 gum+0.011
How+0.010
Neg logit contributions
ously-0.018
 despair-0.017
rum-0.017
 jaws-0.017
iously-0.017
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.012
 ju+0.012
 makeup+0.011
mons+0.011
 contest+0.011
Neg logit contributions
ously-0.023
rum-0.022
 deeply-0.022
 despair-0.021
 horror-0.021
TokenNo
Token ID2949
Feature activation-4.87e-02

Pos logit contributions
 ju+0.235
 golf+0.224
 makeup+0.219
atmeal+0.219
 quiz+0.210
Neg logit contributions
ously-0.361
rum-0.345
 despair-0.337
 deeply-0.325
 horror-0.320
Token,
Token ID11
Feature activation-8.40e-02

Pos logit contributions
 golf+0.065
 makeup+0.064
 ju+0.062
 butter+0.061
 gum+0.061
Neg logit contributions
ously-0.094
rum-0.091
 despair-0.089
 deeply-0.089
 silently-0.085
Token I
Token ID314
Feature activation-8.16e-02

Pos logit contributions
 ju+0.122
mons+0.119
 golf+0.117
atmeal+0.116
 makeup+0.111
Neg logit contributions
ously-0.152
rum-0.149
 despair-0.148
 horror-0.147
 deeply-0.144
Token'm
Token ID1101
Feature activation-3.42e-02

Pos logit contributions
 ju+0.121
 golf+0.115
 makeup+0.112
mons+0.111
 grapes+0.110
Neg logit contributions
ously-0.151
rum-0.149
 despair-0.141
 deeply-0.137
 horror-0.135
Token not
Token ID407
Feature activation-6.00e-03

Pos logit contributions
 ju+0.051
 golf+0.049
 makeup+0.048
mons+0.046
 contest+0.046
Neg logit contributions
ously-0.065
rum-0.062
 despair-0.059
 deeply-0.057
 horror-0.056
Token Ben
Token ID3932
Feature activation-3.90e-02

Pos logit contributions
 golf+0.041
 makeup+0.040
 ju+0.038
 grapes+0.036
 contest+0.036
Neg logit contributions
rum-0.089
ously-0.086
aced-0.081
 despair-0.079
 deeply-0.078
Token shouts
Token ID34757
Feature activation+1.91e-02

Pos logit contributions
 ju+0.048
 makeup+0.047
 golf+0.047
 paints+0.046
 grapes+0.046
Neg logit contributions
rum-0.074
ously-0.074
 despair-0.068
 deeply-0.068
ment-0.067
Token.
Token ID13
Feature activation-1.97e-02

Pos logit contributions
ously+0.034
rum+0.034
 despair+0.033
 deeply+0.031
 horror+0.030
Neg logit contributions
 golf-0.027
 makeup-0.025
 ju-0.025
 contest-0.024
mons-0.024
Token
Token ID198
Feature activation-1.23e-02

Pos logit contributions
 golf+0.026
 baseball+0.023
 gum+0.023
 contest+0.022
 ju+0.022
Neg logit contributions
rum-0.040
 despair-0.038
ously-0.038
aced-0.036
urbed-0.036
Token
Token ID198
Feature activation-1.40e-02

Pos logit contributions
 golf+0.014
How+0.011
 gum+0.011
Draw+0.011
 baseball+0.011
Neg logit contributions
ously-0.028
 despair-0.026
rum-0.026
 jaws-0.026
urbed-0.026
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.016
 ju+0.013
 makeup+0.013
mons+0.013
Draw+0.013
Neg logit contributions
ously-0.032
rum-0.030
 deeply-0.030
 despair-0.029
 horror-0.029
TokenNo
Token ID2949
Feature activation-5.03e-02

Pos logit contributions
 ju+0.254
 makeup+0.236
atmeal+0.236
 golf+0.234
mons+0.228
Neg logit contributions
ously-0.340
rum-0.331
 despair-0.319
 deeply-0.304
 horror-0.303
Token,
Token ID11
Feature activation-8.65e-02

Pos logit contributions
 golf+0.075
 ju+0.074
 makeup+0.072
 contest+0.069
 butter+0.069
Neg logit contributions
ously-0.091
rum-0.089
 despair-0.085
 deeply-0.085
 horror-0.081
Token it
Token ID340
Feature activation-4.55e-02

Pos logit contributions
 ju+0.122
mons+0.118
atmeal+0.115
 golf+0.112
 makeup+0.111
Neg logit contributions
ously-0.161
rum-0.159
 despair-0.155
 horror-0.153
 deeply-0.151
Token's
Token ID338
Feature activation-3.45e-02

Pos logit contributions
 ju+0.072
 golf+0.069
mons+0.067
 makeup+0.066
 grapes+0.065
Neg logit contributions
rum-0.080
ously-0.076
 deeply-0.072
 despair-0.072
 horror-0.070
Token mine
Token ID6164
Feature activation-7.14e-03

Pos logit contributions
 ju+0.044
 golf+0.041
mons+0.040
 makeup+0.040
atmeal+0.039
Neg logit contributions
rum-0.070
ously-0.069
 despair-0.066
 deeply-0.066
 horror-0.064
Token the
Token ID262
Feature activation-4.13e-02

Pos logit contributions
ously+0.022
rum+0.021
 despair+0.020
 deeply+0.020
 horror+0.019
Neg logit contributions
 ju-0.028
 golf-0.027
mons-0.027
atmeal-0.026
 makeup-0.026
Token sand
Token ID6450
Feature activation-4.95e-02

Pos logit contributions
 ju+0.048
 golf+0.045
mons+0.044
 makeup+0.043
atmeal+0.041
Neg logit contributions
ously-0.088
rum-0.086
 despair-0.083
 deeply-0.082
 horror-0.080
Token.
Token ID13
Feature activation-1.82e-02

Pos logit contributions
 ju+0.081
 golf+0.077
 makeup+0.075
mons+0.074
 contest+0.072
Neg logit contributions
ously-0.084
rum-0.080
 deeply-0.075
 despair-0.074
 horror-0.072
Token
Token ID198
Feature activation-1.09e-02

Pos logit contributions
 ju+0.035
 golf+0.035
 makeup+0.034
mons+0.033
 grapes+0.032
Neg logit contributions
ously-0.026
rum-0.025
 despair-0.023
 deeply-0.023
 horror-0.021
Token
Token ID198
Feature activation-1.14e-02

Pos logit contributions
 golf+0.018
 ju+0.017
Draw+0.016
 makeup+0.016
mons+0.016
Neg logit contributions
ously-0.019
 despair-0.017
rum-0.017
 jaws-0.016
 horror-0.016
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 ju+0.015
 golf+0.014
 makeup+0.014
mons+0.013
 contest+0.013
Neg logit contributions
ously-0.023
rum-0.023
 despair-0.022
 deeply-0.022
 horror-0.021
TokenLook
Token ID8567
Feature activation-4.58e-02

Pos logit contributions
 ju+0.310
 makeup+0.302
 golf+0.298
atmeal+0.298
 quiz+0.293
Neg logit contributions
ously-0.258
rum-0.246
 despair-0.238
 horror-0.229
 drifted-0.218
Token,
Token ID11
Feature activation-9.07e-02

Pos logit contributions
 ju+0.094
mons+0.088
 golf+0.086
Draw+0.085
atmeal+0.084
Neg logit contributions
rum-0.057
ously-0.057
 deeply-0.053
 sadly-0.053
 horror-0.050
Token Sam
Token ID3409
Feature activation-6.71e-02

Pos logit contributions
atmeal+0.165
mons+0.165
 ju+0.165
organized+0.162
Draw+0.157
Neg logit contributions
 horror-0.115
 despair-0.112
rum-0.110
ously-0.109
 deeply-0.103
Token,
Token ID11
Feature activation-9.45e-02

Pos logit contributions
 ju+0.125
 golf+0.119
mons+0.117
 makeup+0.116
 contest+0.112
Neg logit contributions
ously-0.100
rum-0.098
 despair-0.089
 deeply-0.088
 horror-0.085
Token I
Token ID314
Feature activation-8.60e-02

Pos logit contributions
mons+0.168
atmeal+0.166
organized+0.165
coat+0.164
 ju+0.156
Neg logit contributions
 his-0.122
 its-0.116
 fear-0.116
 horror-0.115
 despair-0.111
Token Ben
Token ID3932
Feature activation-6.00e-02

Pos logit contributions
 golf+0.027
 makeup+0.026
 ju+0.026
 grapes+0.025
mons+0.024
Neg logit contributions
rum-0.040
ously-0.039
 despair-0.037
 deeply-0.037
aced-0.036
Token says
Token ID1139
Feature activation+4.21e-03

Pos logit contributions
 ju+0.062
 makeup+0.061
 butter+0.060
 paints+0.057
 golf+0.056
Neg logit contributions
rum-0.133
ously-0.125
 despair-0.123
 horror-0.120
 deeply-0.118
Token.
Token ID13
Feature activation-1.79e-02

Pos logit contributions
rum+0.008
 despair+0.008
ously+0.007
 horror+0.007
 jaws+0.007
Neg logit contributions
 golf-0.006
 makeup-0.005
 ju-0.005
 contest-0.005
 butter-0.005
Token
Token ID198
Feature activation-1.09e-02

Pos logit contributions
 golf+0.029
 ju+0.027
 makeup+0.025
 gum+0.025
 grapes+0.025
Neg logit contributions
rum-0.031
ously-0.030
 despair-0.030
aced-0.028
urbed-0.027
Token
Token ID198
Feature activation-1.35e-02

Pos logit contributions
 golf+0.016
 makeup+0.014
 gum+0.013
Draw+0.013
 ju+0.013
Neg logit contributions
ously-0.021
 despair-0.020
rum-0.020
urbed-0.018
 horror-0.018
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.016
 ju+0.015
 makeup+0.014
mons+0.013
 contest+0.013
Neg logit contributions
ously-0.029
rum-0.028
 despair-0.027
 deeply-0.027
 horror-0.027
TokenYou
Token ID1639
Feature activation-3.46e-02

Pos logit contributions
 ju+0.241
atmeal+0.231
 makeup+0.231
 golf+0.227
 butter+0.220
Neg logit contributions
ously-0.345
rum-0.326
 despair-0.318
 deeply-0.309
 silently-0.303
Token are
Token ID389
Feature activation-2.96e-02

Pos logit contributions
 ju+0.051
 golf+0.050
mons+0.048
 makeup+0.048
 contest+0.046
Neg logit contributions
ously-0.063
rum-0.062
 despair-0.060
 deeply-0.059
 horror-0.057
Token very
Token ID845
Feature activation-9.22e-03

Pos logit contributions
 ju+0.042
 golf+0.040
 makeup+0.039
mons+0.038
 contest+0.037
Neg logit contributions
ously-0.057
rum-0.056
 despair-0.053
 deeply-0.052
 horror-0.051
Token wise
Token ID10787
Feature activation+5.12e-04

Pos logit contributions
 ju+0.012
 golf+0.011
 makeup+0.011
mons+0.011
 contest+0.010
Neg logit contributions
ously-0.019
rum-0.019
 despair-0.018
 deeply-0.018
 horror-0.017
Token,
Token ID11
Feature activation-6.66e-02

Pos logit contributions
ously+0.001
rum+0.001
 despair+0.001
 deeply+0.001
 horror+0.001
Neg logit contributions
 ju-0.001
 golf-0.001
 makeup-0.001
mons-0.001
 contest-0.001
Token Tom
Token ID4186
Feature activation-2.78e-02

Pos logit contributions
 golf+0.013
 makeup+0.013
 ju+0.012
 grapes+0.012
 gum+0.012
Neg logit contributions
ously-0.016
rum-0.016
aced-0.016
 despair-0.015
 deeply-0.015
Token says
Token ID1139
Feature activation-5.80e-03

Pos logit contributions
 ju+0.041
 makeup+0.040
 golf+0.039
 paints+0.038
 butter+0.038
Neg logit contributions
rum-0.050
ously-0.050
 despair-0.046
 deeply-0.045
 horror-0.044
Token.
Token ID13
Feature activation-1.38e-02

Pos logit contributions
 golf+0.008
 makeup+0.008
 ju+0.008
 contest+0.008
 quiz+0.007
Neg logit contributions
ously-0.010
rum-0.010
 despair-0.010
 horror-0.009
 jaws-0.009
Token
Token ID198
Feature activation-7.65e-03

Pos logit contributions
 golf+0.024
 ju+0.023
 gum+0.022
 grapes+0.022
 makeup+0.022
Neg logit contributions
ously-0.022
rum-0.021
 despair-0.021
urbed-0.020
aced-0.020
Token
Token ID198
Feature activation-9.43e-03

Pos logit contributions
 golf+0.010
Draw+0.009
 gum+0.009
 ju+0.009
 makeup+0.009
Neg logit contributions
ously-0.015
 despair-0.014
rum-0.014
 jaws-0.014
urbed-0.013
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.010
 ju+0.010
 makeup+0.009
 contest+0.009
mons+0.009
Neg logit contributions
ously-0.021
rum-0.020
 despair-0.020
 deeply-0.020
 horror-0.019
TokenThat
Token ID2504
Feature activation-5.44e-02

Pos logit contributions
 ju+0.251
 golf+0.238
 makeup+0.235
atmeal+0.232
mons+0.226
Neg logit contributions
ously-0.340
rum-0.326
 despair-0.314
 deeply-0.304
 horror-0.298
Token is
Token ID318
Feature activation-2.91e-02

Pos logit contributions
 ju+0.084
 golf+0.080
mons+0.080
 makeup+0.076
 grapes+0.075
Neg logit contributions
ously-0.096
rum-0.094
 deeply-0.090
 despair-0.089
 horror-0.086
Token a
Token ID257
Feature activation-5.05e-02

Pos logit contributions
 ju+0.044
mons+0.043
atmeal+0.042
 golf+0.041
 makeup+0.041
Neg logit contributions
rum-0.050
 deeply-0.049
ously-0.048
 despair-0.047
 horror-0.046
Token good
Token ID922
Feature activation-1.71e-02

Pos logit contributions
mons+0.078
 ju+0.077
atmeal+0.072
 grapes+0.067
 golf+0.067
Neg logit contributions
 deeply-0.083
rum-0.082
ously-0.081
 despair-0.081
 horror-0.080
Token idea
Token ID2126
Feature activation-1.66e-02

Pos logit contributions
 ju+0.028
mons+0.026
 golf+0.026
atmeal+0.025
 makeup+0.025
Neg logit contributions
ously-0.029
rum-0.028
 despair-0.027
 deeply-0.027
 horror-0.026
Token pretty
Token ID2495
Feature activation-6.03e-02

Pos logit contributions
 ju+0.089
mons+0.083
atmeal+0.082
 makeup+0.079
 golf+0.079
Neg logit contributions
ously-0.104
rum-0.099
 despair-0.096
 deeply-0.093
 horror-0.091
Token hair
Token ID4190
Feature activation-1.76e-02

Pos logit contributions
 ju+0.078
 golf+0.071
mons+0.070
atmeal+0.069
 contest+0.067
Neg logit contributions
ously-0.122
rum-0.117
 deeply-0.113
 despair-0.112
 horror-0.107
Token."
Token ID526
Feature activation-1.81e-02

Pos logit contributions
 golf+0.032
 ju+0.032
Draw+0.031
 competition+0.030
 contest+0.029
Neg logit contributions
ously-0.022
 deeply-0.022
rum-0.021
 sadly-0.021
 silently-0.020
Token
Token ID198
Feature activation-1.03e-02

Pos logit contributions
 golf+0.028
 makeup+0.027
 ju+0.027
 grapes+0.026
mons+0.025
Neg logit contributions
ously-0.032
rum-0.031
 despair-0.030
aced-0.030
 deeply-0.029
Token
Token ID198
Feature activation-1.10e-02

Pos logit contributions
 golf+0.016
 makeup+0.015
 ju+0.014
Draw+0.014
 gum+0.014
Neg logit contributions
ously-0.018
 despair-0.017
rum-0.017
 jaws-0.017
 horror-0.017
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 ju+0.013
 golf+0.013
 makeup+0.013
mons+0.012
 contest+0.012
Neg logit contributions
ously-0.023
rum-0.022
 deeply-0.022
 despair-0.022
 horror-0.021
TokenHi
Token ID17250
Feature activation-3.26e-02

Pos logit contributions
atmeal+0.249
 ju+0.244
 golf+0.242
 quiz+0.239
 file+0.239
Neg logit contributions
ously-0.303
 despair-0.280
rum-0.270
 deeply-0.266
 silently-0.266
Token,
Token ID11
Feature activation-8.45e-02

Pos logit contributions
 ju+0.064
mons+0.063
atmeal+0.062
Draw+0.060
 golf+0.059
Neg logit contributions
rum-0.039
ously-0.039
 despair-0.037
 horror-0.036
 deeply-0.036
Token Max
Token ID5436
Feature activation-2.52e-02

Pos logit contributions
 ju+0.127
mons+0.125
atmeal+0.120
 golf+0.119
 makeup+0.113
Neg logit contributions
ously-0.143
rum-0.141
 despair-0.136
 deeply-0.135
 horror-0.133
Token,"
Token ID553
Feature activation-3.27e-02

Pos logit contributions
 ju+0.042
 golf+0.040
 makeup+0.039
mons+0.039
 contest+0.038
Neg logit contributions
ously-0.042
rum-0.041
 despair-0.039
 deeply-0.038
 horror-0.037
Token Tom
Token ID4186
Feature activation-4.13e-02

Pos logit contributions
 ju+0.046
 golf+0.044
 makeup+0.043
 grapes+0.041
mons+0.041
Neg logit contributions
ously-0.064
rum-0.063
 despair-0.059
 deeply-0.058
 horror-0.056
Token were
Token ID547
Feature activation-1.64e-02

Pos logit contributions
 ju+0.039
 makeup+0.036
 golf+0.036
 paints+0.035
 butter+0.034
Neg logit contributions
rum-0.066
ously-0.064
 deeply-0.059
 despair-0.058
 heartbeat-0.056
Token bored
Token ID17533
Feature activation-3.20e-02

Pos logit contributions
 ju+0.024
 golf+0.023
 makeup+0.023
 grapes+0.022
mons+0.022
Neg logit contributions
ously-0.031
rum-0.030
 despair-0.028
 deeply-0.028
 horror-0.027
Token.
Token ID13
Feature activation-1.91e-02

Pos logit contributions
 ju+0.064
mons+0.061
atmeal+0.059
 golf+0.059
 makeup+0.056
Neg logit contributions
ously-0.045
rum-0.042
 despair-0.041
 deeply-0.041
 horror-0.040
Token
Token ID198
Feature activation-1.10e-02

Pos logit contributions
 golf+0.037
 ju+0.037
 makeup+0.036
mons+0.035
 grapes+0.034
Neg logit contributions
ously-0.027
rum-0.026
 deeply-0.023
 despair-0.023
urbed-0.022
Token
Token ID198
Feature activation-1.14e-02

Pos logit contributions
 golf+0.019
 makeup+0.017
mons+0.017
 ju+0.017
Draw+0.017
Neg logit contributions
ously-0.019
rum-0.017
 despair-0.017
 deeply-0.016
urbed-0.016
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 ju+0.016
 golf+0.016
 makeup+0.015
mons+0.015
 contest+0.015
Neg logit contributions
ously-0.022
rum-0.021
 despair-0.020
 deeply-0.020
 horror-0.019
TokenLet
Token ID5756
Feature activation-8.55e-02

Pos logit contributions
 ju+0.254
atmeal+0.249
 quiz+0.230
 golf+0.230
 makeup+0.223
Neg logit contributions
ously-0.334
 despair-0.320
rum-0.310
 horror-0.309
 silently-0.298
Token's
Token ID338
Feature activation-2.72e-02

Pos logit contributions
 ju+0.123
mons+0.119
 golf+0.112
atmeal+0.110
 makeup+0.106
Neg logit contributions
ously-0.160
rum-0.157
 despair-0.155
 horror-0.151
 sadly-0.146
Token organize
Token ID16481
Feature activation-2.49e-02

Pos logit contributions
 ju+0.039
mons+0.037
 golf+0.035
 makeup+0.034
 grapes+0.034
Neg logit contributions
ously-0.050
rum-0.050
 despair-0.049
 deeply-0.047
 horror-0.046
Token our
Token ID674
Feature activation-4.18e-03

Pos logit contributions
 ju+0.044
mons+0.041
 golf+0.041
atmeal+0.040
 makeup+0.039
Neg logit contributions
rum-0.038
ously-0.038
 deeply-0.037
 despair-0.036
 horror-0.034
Token toys
Token ID14958
Feature activation-3.28e-03

Pos logit contributions
 ju+0.006
mons+0.006
atmeal+0.005
 golf+0.005
 butter+0.005
Neg logit contributions
rum-0.007
ously-0.007
 despair-0.007
 deeply-0.007
 horror-0.007
Token high
Token ID1029
Feature activation-1.61e-03

Pos logit contributions
 ju+0.130
mons+0.122
atmeal+0.119
 golf+0.118
 contest+0.112
Neg logit contributions
ously-0.127
rum-0.123
 despair-0.121
 deeply-0.118
 horror-0.116
Token things
Token ID1243
Feature activation-3.77e-02

Pos logit contributions
 ju+0.003
 golf+0.002
mons+0.002
 makeup+0.002
atmeal+0.002
Neg logit contributions
ously-0.003
rum-0.003
 despair-0.003
 deeply-0.003
 horror-0.002
Token.
Token ID13
Feature activation-1.66e-02

Pos logit contributions
 ju+0.071
mons+0.068
 golf+0.068
atmeal+0.067
 makeup+0.065
Neg logit contributions
ously-0.054
rum-0.051
 deeply-0.050
 despair-0.048
 horror-0.045
Token
Token ID198
Feature activation-7.40e-03

Pos logit contributions
 ju+0.027
 golf+0.027
 makeup+0.027
mons+0.026
 grapes+0.025
Neg logit contributions
ously-0.028
rum-0.027
 despair-0.025
 deeply-0.025
 horror-0.023
Token
Token ID198
Feature activation-7.99e-03

Pos logit contributions
 golf+0.012
 makeup+0.011
Draw+0.011
mons+0.010
 ju+0.010
Neg logit contributions
ously-0.013
rum-0.012
 despair-0.012
 jaws-0.011
 horror-0.011
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 ju+0.011
 golf+0.010
 makeup+0.010
mons+0.010
 contest+0.009
Neg logit contributions
ously-0.016
rum-0.015
 despair-0.015
 deeply-0.015
 horror-0.014
TokenCome
Token ID16773
Feature activation-8.01e-02

Pos logit contributions
 ju+0.271
atmeal+0.264
 quiz+0.251
 golf+0.249
 makeup+0.238
Neg logit contributions
ously-0.312
 despair-0.303
rum-0.289
 horror-0.281
 silently-0.274
Token on
Token ID319
Feature activation+2.25e-02

Pos logit contributions
 ju+0.151
mons+0.141
 redd+0.134
 grapes+0.134
 makeup+0.131
Neg logit contributions
rum-0.110
 sadly-0.106
ously-0.106
 silently-0.105
 deeply-0.103
Token,
Token ID11
Feature activation-8.24e-02

Pos logit contributions
ously+0.035
rum+0.035
 despair+0.033
 deeply+0.032
 horror+0.031
Neg logit contributions
 ju-0.039
mons-0.037
 golf-0.037
 makeup-0.036
atmeal-0.035
Token Kim
Token ID6502
Feature activation-4.24e-02

Pos logit contributions
 ju+0.135
mons+0.130
 golf+0.128
atmeal+0.125
 makeup+0.124
Neg logit contributions
ously-0.135
rum-0.131
 despair-0.129
 horror-0.124
 deeply-0.124
Token,
Token ID11
Feature activation-8.59e-02

Pos logit contributions
 ju+0.074
 golf+0.071
 makeup+0.069
mons+0.069
 contest+0.067
Neg logit contributions
ously-0.068
rum-0.066
 despair-0.061
 deeply-0.061
 horror-0.059
Token blamed
Token ID13772
Feature activation+5.71e-02

Pos logit contributions
 makeup+0.041
 paints+0.040
 ju+0.036
 swing+0.035
 golf+0.035
Neg logit contributions
rum-0.076
ously-0.074
 despair-0.070
ment-0.070
 heartbeat-0.069
Token Mia
Token ID32189
Feature activation-1.59e-02

Pos logit contributions
ously+0.106
rum+0.103
 deeply+0.099
 despair+0.096
 drifted+0.096
Neg logit contributions
 makeup-0.085
 golf-0.085
 grapes-0.077
mons-0.073
 lipstick-0.073
Token.
Token ID13
Feature activation-1.61e-02

Pos logit contributions
 golf+0.022
 makeup+0.022
 ju+0.021
 quiz+0.020
 butter+0.020
Neg logit contributions
ously-0.028
rum-0.028
 despair-0.027
 deeply-0.025
 horror-0.025
Token
Token ID198
Feature activation-7.08e-03

Pos logit contributions
 golf+0.023
 ju+0.021
 makeup+0.021
 grapes+0.021
 contest+0.020
Neg logit contributions
rum-0.031
ously-0.030
 despair-0.030
urbed-0.028
aced-0.028
Token
Token ID198
Feature activation-8.91e-03

Pos logit contributions
 golf+0.010
 makeup+0.009
Draw+0.009
 paints+0.008
 contest+0.008
Neg logit contributions
ously-0.013
 despair-0.013
rum-0.013
 jaws-0.013
 horror-0.012
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.010
 ju+0.010
 makeup+0.010
mons+0.009
 contest+0.009
Neg logit contributions
ously-0.019
rum-0.018
 deeply-0.018
 despair-0.018
 horror-0.017
TokenIt
Token ID1026
Feature activation-3.11e-02

Pos logit contributions
 ju+0.246
atmeal+0.237
 butter+0.224
 quiz+0.219
 golf+0.217
Neg logit contributions
ously-0.339
 despair-0.319
rum-0.318
 horror-0.299
 deeply-0.297
Token's
Token ID338
Feature activation-1.89e-02

Pos logit contributions
 ju+0.041
 golf+0.040
mons+0.039
 makeup+0.037
atmeal+0.037
Neg logit contributions
rum-0.061
ously-0.060
 despair-0.058
 deeply-0.057
 horror-0.055
Token your
Token ID534
Feature activation-3.27e-03

Pos logit contributions
 ju+0.027
 golf+0.026
mons+0.025
 makeup+0.025
atmeal+0.025
Neg logit contributions
ously-0.035
rum-0.035
 despair-0.033
 deeply-0.033
 horror-0.032
Token fault
Token ID8046
Feature activation+3.10e-02

Pos logit contributions
 ju+0.005
 golf+0.004
 makeup+0.004
mons+0.004
 contest+0.004
Neg logit contributions
ously-0.006
rum-0.006
 despair-0.006
 deeply-0.006
 horror-0.006
Token!"
Token ID2474
Feature activation-3.52e-02

Pos logit contributions
ously+0.056
rum+0.055
 despair+0.055
 horror+0.053
 deeply+0.052
Neg logit contributions
 ju-0.043
 golf-0.042
 makeup-0.042
mons-0.041
 butter-0.041
Token Lily
Token ID20037
Feature activation-3.35e-02

Pos logit contributions
 golf+0.027
 makeup+0.026
 ju+0.025
 grapes+0.023
mons+0.023
Neg logit contributions
ously-0.030
rum-0.030
aced-0.028
 deeply-0.028
 despair-0.027
Token said
Token ID531
Feature activation-4.08e-04

Pos logit contributions
 ju+0.047
 makeup+0.046
 golf+0.045
 paints+0.043
 grapes+0.042
Neg logit contributions
rum-0.064
ously-0.062
 despair-0.058
 deeply-0.058
 horror-0.056
Token.
Token ID13
Feature activation-2.32e-02

Pos logit contributions
 ju+0.001
 golf+0.001
 makeup+0.001
mons+0.001
 contest+0.001
Neg logit contributions
rum-0.001
ously-0.001
 despair-0.001
 horror-0.001
 deeply-0.001
Token
Token ID198
Feature activation-7.18e-03

Pos logit contributions
 golf+0.039
 ju+0.038
 makeup+0.037
mons+0.036
 grapes+0.035
Neg logit contributions
ously-0.038
rum-0.038
 despair-0.036
 deeply-0.035
aced-0.033
Token
Token ID198
Feature activation-9.24e-03

Pos logit contributions
 golf+0.011
 makeup+0.010
Draw+0.010
 ju+0.010
 gum+0.010
Neg logit contributions
ously-0.013
rum-0.013
 despair-0.013
 jaws-0.012
 deeply-0.012
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 ju+0.012
 golf+0.012
 makeup+0.011
mons+0.011
 contest+0.011
Neg logit contributions
ously-0.019
rum-0.018
 despair-0.017
 deeply-0.017
 horror-0.017
TokenHey
Token ID10814
Feature activation-1.52e-02

Pos logit contributions
atmeal+0.271
 ju+0.263
 quiz+0.253
 makeup+0.250
 golf+0.247
Neg logit contributions
ously-0.318
 despair-0.291
rum-0.286
 silently-0.274
 horror-0.274
Token,
Token ID11
Feature activation-8.92e-02

Pos logit contributions
 ju+0.036
atmeal+0.035
mons+0.035
cakes+0.033
Draw+0.033
Neg logit contributions
rum-0.013
 horror-0.013
ously-0.013
 despair-0.012
 sadly-0.012
Token cat
Token ID3797
Feature activation-5.26e-02

Pos logit contributions
 ju+0.133
mons+0.126
atmeal+0.125
organized+0.116
Draw+0.114
Neg logit contributions
ously-0.159
 horror-0.156
 despair-0.154
rum-0.153
 fear-0.145
Token.
Token ID13
Feature activation-3.32e-02

Pos logit contributions
 ju+0.113
atmeal+0.103
mons+0.102
 golf+0.101
 makeup+0.101
Neg logit contributions
ously-0.067
rum-0.066
 deeply-0.058
 despair-0.056
 horror-0.056
Token Go
Token ID1514
Feature activation-6.41e-02

Pos logit contributions
 ju+0.057
atmeal+0.053
 golf+0.052
 butter+0.051
 competition+0.051
Neg logit contributions
rum-0.053
ously-0.051
 horror-0.050
 despair-0.050
 deeply-0.047
Token find
Token ID1064
Feature activation-3.49e-02

Pos logit contributions
ously+0.029
rum+0.028
 deeply+0.026
 despair+0.025
aced+0.025
Neg logit contributions
 golf-0.015
 ju-0.014
 contest-0.013
 swing-0.012
 competition-0.012
Token treasure
Token ID14068
Feature activation-6.07e-03

Pos logit contributions
 ju+0.047
 golf+0.046
 makeup+0.044
 butter+0.043
 grapes+0.042
Neg logit contributions
ously-0.070
rum-0.069
 despair-0.064
 deeply-0.063
 horror-0.061
Token.
Token ID13
Feature activation-1.51e-02

Pos logit contributions
 ju+0.011
 golf+0.010
 makeup+0.010
mons+0.010
 contest+0.010
Neg logit contributions
ously-0.009
rum-0.009
 deeply-0.008
 despair-0.008
 horror-0.008
Token
Token ID198
Feature activation-6.14e-03

Pos logit contributions
 golf+0.027
 ju+0.027
 makeup+0.025
mons+0.025
 contest+0.024
Neg logit contributions
ously-0.025
rum-0.024
 despair-0.022
 deeply-0.021
aced-0.021
Token
Token ID198
Feature activation-8.55e-03

Pos logit contributions
 golf+0.010
 ju+0.009
 makeup+0.009
Draw+0.009
mons+0.009
Neg logit contributions
ously-0.010
rum-0.010
 despair-0.010
 horror-0.009
 jaws-0.009
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 ju+0.011
 golf+0.011
 makeup+0.010
mons+0.010
 contest+0.010
Neg logit contributions
ously-0.017
rum-0.017
 despair-0.016
 deeply-0.016
 horror-0.016
TokenLook
Token ID8567
Feature activation-4.31e-02

Pos logit contributions
 ju+0.274
 golf+0.262
 makeup+0.258
mons+0.252
atmeal+0.248
Neg logit contributions
ously-0.320
rum-0.308
 despair-0.297
 deeply-0.286
 horror-0.283
Token,
Token ID11
Feature activation-8.81e-02

Pos logit contributions
 ju+0.085
mons+0.080
 golf+0.078
atmeal+0.076
Draw+0.076
Neg logit contributions
ously-0.057
rum-0.057
 deeply-0.053
 sadly-0.051
 despair-0.051
Token Tom
Token ID4186
Feature activation-5.39e-02

Pos logit contributions
 ju+0.149
mons+0.149
atmeal+0.145
Draw+0.141
organized+0.140
Neg logit contributions
ously-0.125
rum-0.124
 despair-0.122
 horror-0.122
 deeply-0.116
Token,
Token ID11
Feature activation-9.16e-02

Pos logit contributions
 ju+0.090
 golf+0.087
 makeup+0.085
mons+0.084
 contest+0.082
Neg logit contributions
ously-0.089
rum-0.087
 despair-0.082
 deeply-0.081
 horror-0.078
Token there
Token ID612
Feature activation-3.76e-02

Pos logit contributions
mons+0.150
atmeal+0.144
organized+0.142
 ju+0.136
coat+0.136
Neg logit contributions
 Flash-0.134
rum-0.133
 despair-0.133
 horror-0.133
ously-0.131
Token Ben
Token ID3932
Feature activation-3.97e-02

Pos logit contributions
ously+0.012
rum+0.012
 despair+0.011
 deeply+0.011
aced+0.011
Neg logit contributions
 golf-0.012
 makeup-0.011
 ju-0.011
 grapes-0.011
 gum-0.011
Token said
Token ID531
Feature activation+9.85e-03

Pos logit contributions
 ju+0.054
 makeup+0.054
 paints+0.053
 golf+0.052
 butter+0.051
Neg logit contributions
ously-0.074
rum-0.073
 despair-0.070
 horror-0.066
 deeply-0.065
Token.
Token ID13
Feature activation-1.34e-02

Pos logit contributions
ously+0.017
rum+0.016
 despair+0.016
 horror+0.015
 deeply+0.015
Neg logit contributions
 ju-0.015
 golf-0.015
 makeup-0.014
 contest-0.014
 quiz-0.014
Token
Token ID198
Feature activation-5.50e-03

Pos logit contributions
 golf+0.021
 ju+0.021
 makeup+0.019
 grapes+0.019
 gum+0.019
Neg logit contributions
rum-0.023
ously-0.023
 despair-0.022
aced-0.021
urbed-0.020
Token
Token ID198
Feature activation-9.01e-03

Pos logit contributions
 golf+0.007
 gum+0.006
 makeup+0.006
Draw+0.006
 contest+0.006
Neg logit contributions
ously-0.011
 despair-0.011
rum-0.011
iously-0.010
 jaws-0.010
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.009
 ju+0.009
 makeup+0.008
 contest+0.008
Draw+0.008
Neg logit contributions
ously-0.020
rum-0.020
 despair-0.019
 deeply-0.019
 horror-0.019
TokenYes
Token ID5297
Feature activation-6.88e-03

Pos logit contributions
 ju+0.250
atmeal+0.240
 makeup+0.239
 golf+0.236
 butter+0.231
Neg logit contributions
ously-0.330
rum-0.313
 despair-0.300
 deeply-0.297
 silently-0.288
Token,
Token ID11
Feature activation-9.33e-02

Pos logit contributions
 ju+0.009
 golf+0.009
 makeup+0.009
 butter+0.009
 contest+0.009
Neg logit contributions
ously-0.013
rum-0.013
 despair-0.013
 deeply-0.012
 horror-0.012
Token I
Token ID314
Feature activation-9.27e-02

Pos logit contributions
mons+0.142
atmeal+0.142
organized+0.136
 ju+0.135
Draw+0.133
Neg logit contributions
ously-0.159
rum-0.150
 horror-0.149
 deeply-0.149
 despair-0.148
Token do
Token ID466
Feature activation-7.05e-02

Pos logit contributions
 ju+0.138
 golf+0.130
mons+0.129
 makeup+0.124
 grapes+0.124
Neg logit contributions
ously-0.170
rum-0.166
 despair-0.157
 deeply-0.154
 horror-0.154
Token.
Token ID13
Feature activation-4.08e-02

Pos logit contributions
 ju+0.132
 golf+0.124
mons+0.122
 makeup+0.120
atmeal+0.118
Neg logit contributions
rum-0.104
ously-0.103
 deeply-0.096
 despair-0.093
 horror-0.089
Token Ben
Token ID3932
Feature activation-5.58e-02

Pos logit contributions
 golf+0.039
 makeup+0.039
 ju+0.037
 contest+0.036
 grapes+0.036
Neg logit contributions
rum-0.065
ously-0.063
aced-0.059
 deeply-0.058
 despair-0.058
Token said
Token ID531
Feature activation+9.30e-03

Pos logit contributions
 ju+0.069
 makeup+0.068
 golf+0.066
 butter+0.065
 quiz+0.064
Neg logit contributions
rum-0.114
ously-0.109
 despair-0.102
 deeply-0.099
urbed-0.098
Token.
Token ID13
Feature activation-2.08e-02

Pos logit contributions
 despair+0.018
rum+0.017
ously+0.017
 horror+0.017
 jaws+0.016
Neg logit contributions
 golf-0.012
 makeup-0.011
 ju-0.011
 contest-0.011
 butter-0.010
Token
Token ID198
Feature activation-1.25e-02

Pos logit contributions
 golf+0.030
 ju+0.026
 makeup+0.025
 contest+0.025
 gum+0.025
Neg logit contributions
rum-0.041
ously-0.040
 despair-0.039
aced-0.038
urbed-0.036
Token
Token ID198
Feature activation-1.49e-02

Pos logit contributions
 golf+0.018
 makeup+0.015
 gum+0.015
Draw+0.015
 baseball+0.015
Neg logit contributions
ously-0.025
rum-0.023
 despair-0.023
urbed-0.022
iously-0.022
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.018
 ju+0.017
 makeup+0.016
 contest+0.015
mons+0.015
Neg logit contributions
ously-0.031
rum-0.030
 deeply-0.030
 despair-0.029
 horror-0.029
TokenBen
Token ID11696
Feature activation-8.80e-02

Pos logit contributions
 ju+0.224
 golf+0.210
 makeup+0.207
mons+0.201
atmeal+0.199
Neg logit contributions
ously-0.371
rum-0.359
 despair-0.346
 deeply-0.341
 horror-0.330
Token took
Token ID1718
Feature activation-4.88e-02

Pos logit contributions
 ju+0.140
 golf+0.133
mons+0.129
 makeup+0.128
atmeal+0.124
Neg logit contributions
ously-0.153
rum-0.148
 despair-0.141
 deeply-0.139
 horror-0.135
Token my
Token ID616
Feature activation-1.48e-02

Pos logit contributions
 ju+0.071
 golf+0.071
 makeup+0.069
 grapes+0.067
 butter+0.065
Neg logit contributions
ously-0.093
rum-0.088
 despair-0.082
 deeply-0.082
 horror-0.079
Token car
Token ID1097
Feature activation-4.01e-02

Pos logit contributions
 ju+0.020
 golf+0.019
mons+0.018
 makeup+0.018
atmeal+0.018
Neg logit contributions
ously-0.029
rum-0.028
 despair-0.027
 deeply-0.027
 horror-0.026
Token.
Token ID13
Feature activation-2.84e-02

Pos logit contributions
 ju+0.067
 golf+0.064
 makeup+0.063
mons+0.062
 contest+0.061
Neg logit contributions
ously-0.066
rum-0.065
 despair-0.061
 deeply-0.060
 horror-0.058
Token a
Token ID257
Feature activation-5.47e-02

Pos logit contributions
 ju+0.021
 golf+0.020
mons+0.020
 makeup+0.019
 contest+0.019
Neg logit contributions
ously-0.027
rum-0.027
 deeply-0.025
 despair-0.025
 horror-0.024
Token feather
Token ID17972
Feature activation-4.18e-02

Pos logit contributions
 ju+0.073
mons+0.072
atmeal+0.068
 golf+0.067
 makeup+0.066
Neg logit contributions
ously-0.102
rum-0.100
 deeply-0.098
 despair-0.096
 horror-0.093
Token.
Token ID13
Feature activation-1.62e-02

Pos logit contributions
 ju+0.071
 golf+0.068
 makeup+0.067
mons+0.066
 contest+0.064
Neg logit contributions
ously-0.067
rum-0.065
 deeply-0.061
 despair-0.061
 horror-0.058
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
 ju+0.030
 golf+0.029
 makeup+0.028
mons+0.028
 grapes+0.027
Neg logit contributions
ously-0.024
rum-0.023
 despair-0.022
 deeply-0.021
 horror-0.021
Token
Token ID198
Feature activation-1.05e-02

Pos logit contributions
 golf+0.017
 ju+0.015
 makeup+0.015
Draw+0.015
 grapes+0.015
Neg logit contributions
ously-0.019
rum-0.017
 despair-0.017
 horror-0.016
 deeply-0.016
Token"
Token ID1
Feature activation-1.79e-01

Pos logit contributions
 golf+0.013
 ju+0.013
 makeup+0.012
mons+0.012
 contest+0.011
Neg logit contributions
ously-0.022
rum-0.021
 despair-0.020
 deeply-0.020
 horror-0.020
TokenThank
Token ID10449
Feature activation-4.51e-02

Pos logit contributions
 ju+0.257
atmeal+0.257
 quiz+0.250
 makeup+0.249
 golf+0.247
Neg logit contributions
ously-0.307
 despair-0.287
rum-0.287
 deeply-0.270
 horror-0.268
Token you
Token ID345
Feature activation-3.80e-02

Pos logit contributions
 ju+0.046
 golf+0.042
 makeup+0.040
mons+0.039
atmeal+0.038
Neg logit contributions
ously-0.104
rum-0.103
 deeply-0.098
 despair-0.098
 horror-0.094
Token,
Token ID11
Feature activation-7.97e-02

Pos logit contributions
 ju+0.043
 golf+0.042
 grapes+0.041
 butter+0.040
 makeup+0.039
Neg logit contributions
ously-0.082
rum-0.080
 despair-0.077
 drifted-0.073
 heartbeat-0.072
Token bird
Token ID6512
Feature activation-3.45e-02

Pos logit contributions
 ju+0.129
 golf+0.125
 makeup+0.121
mons+0.118
 grapes+0.117
Neg logit contributions
ously-0.138
rum-0.135
 despair-0.126
 deeply-0.124
 horror-0.119
Token,
Token ID11
Feature activation-8.27e-02

Pos logit contributions
 ju+0.055
 golf+0.053
 makeup+0.052
mons+0.051
 contest+0.050
Neg logit contributions
ously-0.059
rum-0.057
 despair-0.055
 deeply-0.054
 horror-0.053
Token Ben
Token ID3932
Feature activation-4.57e-02

Pos logit contributions
 golf+0.045
 makeup+0.044
 ju+0.043
 grapes+0.041
mons+0.041
Neg logit contributions
rum-0.059
ously-0.058
 despair-0.054
 deeply-0.053
aced-0.053
Token said
Token ID531
Feature activation+1.31e-02

Pos logit contributions
 ju+0.056
 makeup+0.053
 golf+0.052
 butter+0.051
 grapes+0.049
Neg logit contributions
rum-0.095
ously-0.092
 despair-0.088
 horror-0.084
 deeply-0.084
Token.
Token ID13
Feature activation-1.80e-02

Pos logit contributions
 despair+0.025
rum+0.025
ously+0.024
 horror+0.024
 jaws+0.024
Neg logit contributions
 golf-0.016
 makeup-0.015
 ju-0.015
 butter-0.014
 contest-0.014
Token
Token ID198
Feature activation-1.20e-02

Pos logit contributions
 golf+0.026
 ju+0.022
 baseball+0.022
 gum+0.022
 grapes+0.022
Neg logit contributions
rum-0.035
ously-0.034
 despair-0.034
aced-0.033
urbed-0.032
Token
Token ID198
Feature activation-1.38e-02

Pos logit contributions
 golf+0.016
 gum+0.014
Draw+0.014
 ju+0.014
 makeup+0.013
Neg logit contributions
ously-0.024
 despair-0.023
rum-0.023
iously-0.022
urbed-0.022
Token"
Token ID1
Feature activation-1.78e-01

Pos logit contributions
 golf+0.016
 ju+0.015
 makeup+0.014
 contest+0.014
mons+0.014
Neg logit contributions
ously-0.030
rum-0.029
 despair-0.028
 deeply-0.028
 horror-0.027
TokenThen
Token ID6423
Feature activation-2.40e-02

Pos logit contributions
 ju+0.257
atmeal+0.253
 makeup+0.246
 golf+0.241
 butter+0.235
Neg logit contributions
ously-0.325
rum-0.298
 despair-0.295
 deeply-0.289
 silently-0.284
Token you
Token ID345
Feature activation-1.86e-02

Pos logit contributions
 ju+0.036
mons+0.034
atmeal+0.034
 golf+0.033
Draw+0.032
Neg logit contributions
ously-0.041
 despair-0.040
rum-0.040
 deeply-0.039
 horror-0.038
Token are
Token ID389
Feature activation-7.69e-03

Pos logit contributions
 ju+0.026
mons+0.025
 golf+0.025
 makeup+0.024
atmeal+0.024
Neg logit contributions
ously-0.034
rum-0.033
 despair-0.033
 deeply-0.033
 horror-0.032
Token very
Token ID845
Feature activation+7.81e-03

Pos logit contributions
 ju+0.011
 golf+0.011
 makeup+0.011
 butter+0.010
 contest+0.010
Neg logit contributions
ously-0.015
rum-0.014
 despair-0.013
 deeply-0.013
 horror-0.013
Token lucky
Token ID9670
Feature activation-1.93e-02

Pos logit contributions
ously+0.017
rum+0.016
 despair+0.015
 deeply+0.015
 horror+0.015
Neg logit contributions
 ju-0.010
 golf-0.009
 makeup-0.009
mons-0.009
 contest-0.009
Token Lily
Token ID20037
Feature activation-2.67e-02

Pos logit contributions
 golf+0.017
 makeup+0.017
 gum+0.015
 grapes+0.015
 ju+0.015
Neg logit contributions
rum-0.023
ously-0.022
aced-0.022
 despair-0.021
 deeply-0.021
Token said
Token ID531
Feature activation-4.47e-03

Pos logit contributions
 ju+0.040
 makeup+0.038
 golf+0.038
 paints+0.036
 grapes+0.036
Neg logit contributions
rum-0.048
ously-0.048
 despair-0.045
 deeply-0.044
 horror-0.042
Token.
Token ID13
Feature activation-1.23e-02

Pos logit contributions
 ju+0.007
 golf+0.007
 makeup+0.006
mons+0.006
 contest+0.006
Neg logit contributions
rum-0.008
ously-0.008
 despair-0.008
 deeply-0.007
 horror-0.007
Token
Token ID198
Feature activation-7.08e-03

Pos logit contributions
 golf+0.020
 makeup+0.020
 ju+0.019
mons+0.019
 grapes+0.019
Neg logit contributions
rum-0.021
ously-0.021
 despair-0.020
 deeply-0.019
aced-0.019
Token
Token ID198
Feature activation-8.51e-03

Pos logit contributions
 golf+0.010
 makeup+0.009
Draw+0.009
 gum+0.009
 paints+0.009
Neg logit contributions
ously-0.013
 despair-0.013
rum-0.013
 jaws-0.012
 horror-0.012
Token"
Token ID1
Feature activation-1.78e-01

Pos logit contributions
 ju+0.011
 golf+0.011
 makeup+0.010
mons+0.010
 contest+0.010
Neg logit contributions
ously-0.017
rum-0.017
 despair-0.016
 deeply-0.016
 horror-0.015
TokenL
Token ID43
Feature activation-5.43e-02

Pos logit contributions
atmeal+0.266
 ju+0.263
 golf+0.248
 quiz+0.241
 grapes+0.236
Neg logit contributions
ously-0.320
 despair-0.292
rum-0.284
 deeply-0.276
 silently-0.276
Tokenily
Token ID813
Feature activation-4.88e-02

Pos logit contributions
 ju+0.095
 golf+0.082
mons+0.077
 makeup+0.077
 butter+0.076
Neg logit contributions
ously-0.096
rum-0.095
 deeply-0.083
 despair-0.082
aced-0.081
Token,
Token ID11
Feature activation-7.78e-02

Pos logit contributions
 ju+0.093
 golf+0.088
mons+0.085
atmeal+0.084
 makeup+0.083
Neg logit contributions
ously-0.072
rum-0.069
 despair-0.064
 deeply-0.063
 horror-0.061
Token the
Token ID262
Feature activation-5.54e-02

Pos logit contributions
 ju+0.124
mons+0.122
atmeal+0.118
organized+0.114
 golf+0.107
Neg logit contributions
 horror-0.130
ously-0.128
 despair-0.127
rum-0.124
 deeply-0.122
Token pictures
Token ID5986
Feature activation-4.53e-02

Pos logit contributions
atmeal+0.079
 ju+0.079
mons+0.070
 golf+0.065
organized+0.064
Neg logit contributions
ously-0.097
 despair-0.096
 horror-0.095
rum-0.090
 heartbeat-0.090
Token is
Token ID318
Feature activation-1.94e-02

Pos logit contributions
 ju+0.037
 golf+0.036
mons+0.035
 makeup+0.034
atmeal+0.033
Neg logit contributions
ously-0.047
rum-0.046
 deeply-0.044
 despair-0.043
 silently-0.041
Token pretty
Token ID2495
Feature activation-4.11e-02

Pos logit contributions
 ju+0.026
 golf+0.024
mons+0.024
 makeup+0.024
atmeal+0.023
Neg logit contributions
ously-0.037
rum-0.037
 deeply-0.035
 despair-0.035
 horror-0.034
Token.
Token ID13
Feature activation-1.96e-02

Pos logit contributions
 ju+0.083
mons+0.081
atmeal+0.080
 golf+0.080
Draw+0.078
Neg logit contributions
ously-0.050
rum-0.049
 deeply-0.047
 despair-0.043
 sadly-0.042
Token
Token ID198
Feature activation-1.06e-02

Pos logit contributions
 ju+0.033
 golf+0.033
 makeup+0.032
mons+0.031
 grapes+0.030
Neg logit contributions
ously-0.032
rum-0.031
 despair-0.029
 deeply-0.029
 horror-0.027
Token
Token ID198
Feature activation-1.04e-02

Pos logit contributions
 golf+0.017
 makeup+0.015
 ju+0.015
Draw+0.015
 grapes+0.015
Neg logit contributions
ously-0.019
rum-0.018
 despair-0.017
urbed-0.017
 horror-0.016
Token"
Token ID1
Feature activation-1.78e-01

Pos logit contributions
 golf+0.013
 ju+0.012
 makeup+0.012
mons+0.011
 contest+0.011
Neg logit contributions
ously-0.022
rum-0.021
 despair-0.020
 deeply-0.020
 horror-0.020
TokenCan
Token ID6090
Feature activation-1.33e-01

Pos logit contributions
atmeal+0.274
 ju+0.269
 quiz+0.259
 golf+0.253
 makeup+0.248
Neg logit contributions
ously-0.298
 despair-0.282
rum-0.275
 horror-0.265
 silently-0.261
Token I
Token ID314
Feature activation-7.88e-02

Pos logit contributions
 ju+0.188
atmeal+0.184
mons+0.171
 newspapers+0.169
 golf+0.166
Neg logit contributions
 despair-0.222
rum-0.221
 sadly-0.219
ously-0.217
 deeply-0.213
Token have
Token ID423
Feature activation-6.29e-02

Pos logit contributions
 ju+0.134
mons+0.130
atmeal+0.124
 newspapers+0.123
 grapes+0.120
Neg logit contributions
ously-0.113
rum-0.113
 despair-0.112
 deeply-0.108
 sadly-0.107
Token your
Token ID534
Feature activation-2.36e-02

Pos logit contributions
 ju+0.108
atmeal+0.106
mons+0.105
coat+0.095
 quiz+0.094
Neg logit contributions
 despair-0.088
rum-0.088
 deeply-0.087
ously-0.085
 its-0.083
Token star
Token ID3491
Feature activation+1.52e-03

Pos logit contributions
 ju+0.032
atmeal+0.029
mons+0.028
Draw+0.027
organized+0.027
Neg logit contributions
 despair-0.042
rum-0.042
 jaws-0.041
 horror-0.041
 deeply-0.041
Token for
Token ID329
Feature activation-1.39e-02

Pos logit contributions
 ju+0.044
 golf+0.042
 makeup+0.041
mons+0.041
 contest+0.039
Neg logit contributions
ously-0.047
rum-0.046
 despair-0.043
 deeply-0.043
 horror-0.041
Token Ben
Token ID3932
Feature activation-5.14e-02

Pos logit contributions
 ju+0.019
 golf+0.018
 makeup+0.018
mons+0.017
 grapes+0.017
Neg logit contributions
ously-0.028
rum-0.027
 despair-0.026
 deeply-0.025
 horror-0.024
Token.
Token ID13
Feature activation-2.45e-02

Pos logit contributions
 ju+0.092
 golf+0.088
mons+0.086
 makeup+0.086
 contest+0.083
Neg logit contributions
ously-0.080
rum-0.078
 despair-0.072
 deeply-0.071
 horror-0.069
Token
Token ID198
Feature activation-1.42e-02

Pos logit contributions
 golf+0.045
 ju+0.044
 makeup+0.043
mons+0.042
 grapes+0.041
Neg logit contributions
ously-0.038
rum-0.035
 despair-0.033
 deeply-0.032
urbed-0.032
Token
Token ID198
Feature activation-1.35e-02

Pos logit contributions
 golf+0.022
 makeup+0.020
Draw+0.020
 ju+0.020
 gum+0.020
Neg logit contributions
ously-0.026
 despair-0.023
rum-0.023
urbed-0.022
 jaws-0.022
Token"
Token ID1
Feature activation-1.78e-01

Pos logit contributions
 golf+0.018
 ju+0.018
 makeup+0.017
mons+0.017
 contest+0.016
Neg logit contributions
ously-0.027
rum-0.026
 despair-0.025
 deeply-0.025
 horror-0.024
TokenBen
Token ID11696
Feature activation-8.70e-02

Pos logit contributions
 ju+0.227
 golf+0.213
atmeal+0.213
 makeup+0.211
 quiz+0.209
Neg logit contributions
ously-0.357
rum-0.343
 despair-0.337
 horror-0.323
 deeply-0.318
Token,
Token ID11
Feature activation-8.42e-02

Pos logit contributions
 ju+0.183
 golf+0.172
 makeup+0.167
atmeal+0.166
mons+0.166
Neg logit contributions
ously-0.106
rum-0.106
 despair-0.096
 deeply-0.093
 horror-0.093
Token you
Token ID345
Feature activation-3.53e-02

Pos logit contributions
 ju+0.128
mons+0.122
 golf+0.118
atmeal+0.114
 makeup+0.113
Neg logit contributions
ously-0.149
rum-0.146
 despair-0.144
 horror-0.142
 deeply-0.138
Token need
Token ID761
Feature activation-3.24e-02

Pos logit contributions
 ju+0.062
 golf+0.059
mons+0.059
atmeal+0.057
 makeup+0.057
Neg logit contributions
 despair-0.053
rum-0.053
ously-0.052
 horror-0.050
 deeply-0.048
Token to
Token ID284
Feature activation-5.83e-03

Pos logit contributions
 ju+0.042
 golf+0.040
 makeup+0.038
mons+0.038
 contest+0.037
Neg logit contributions
ously-0.066
rum-0.064
 despair-0.061
 deeply-0.061
 horror-0.059
Token bush
Token ID24484
Feature activation-2.75e-02

Pos logit contributions
 ju+0.008
 golf+0.007
 makeup+0.007
mons+0.007
 contest+0.007
Neg logit contributions
ously-0.010
rum-0.010
 despair-0.009
 deeply-0.009
 horror-0.009
Token nearby
Token ID6716
Feature activation-1.36e-02

Pos logit contributions
 ju+0.050
 golf+0.048
 makeup+0.047
mons+0.047
atmeal+0.045
Neg logit contributions
ously-0.041
rum-0.039
 deeply-0.037
 despair-0.037
 horror-0.034
Token.
Token ID13
Feature activation+2.07e-02

Pos logit contributions
 ju+0.022
 golf+0.021
 makeup+0.021
mons+0.021
 contest+0.020
Neg logit contributions
ously-0.023
rum-0.022
 despair-0.021
 deeply-0.021
 horror-0.020
Token He
Token ID679
Feature activation+1.93e-02

Pos logit contributions
ously+0.033
rum+0.032
 despair+0.030
 deeply+0.029
aced+0.028
Neg logit contributions
 ju-0.036
 golf-0.036
 makeup-0.035
 grapes-0.033
mons-0.033
Token flew
Token ID13112
Feature activation-1.13e-03

Pos logit contributions
rum+0.037
ously+0.036
 despair+0.033
 deeply+0.032
 horror+0.032
Neg logit contributions
 ju-0.027
 makeup-0.026
 golf-0.026
 paints-0.026
 butter-0.025
Token over
Token ID625
Feature activation+9.79e-02

Pos logit contributions
 ju+0.002
 golf+0.002
 makeup+0.001
 contest+0.001
 grapes+0.001
Neg logit contributions
rum-0.002
ously-0.002
 despair-0.002
 horror-0.002
 deeply-0.002
Token and
Token ID290
Feature activation+1.08e-02

Pos logit contributions
ously+0.193
rum+0.189
 despair+0.180
 deeply+0.178
 horror+0.173
Neg logit contributions
 ju-0.133
 golf-0.126
 makeup-0.122
mons-0.121
 contest-0.117
Token called
Token ID1444
Feature activation-7.91e-03

Pos logit contributions
ously+0.020
rum+0.020
 despair+0.018
 deeply+0.018
 horror+0.018
Neg logit contributions
 ju-0.016
 golf-0.015
 makeup-0.015
mons-0.015
 contest-0.014
Token out
Token ID503
Feature activation+1.07e-01

Pos logit contributions
 golf+0.010
 ju+0.010
 makeup+0.010
 quiz+0.010
 contest+0.010
Neg logit contributions
ously-0.015
rum-0.015
 despair-0.014
aced-0.014
 horror-0.013
Token,
Token ID11
Feature activation-1.96e-02

Pos logit contributions
ously+0.230
rum+0.227
 despair+0.220
 drifted+0.208
aced+0.207
Neg logit contributions
 golf-0.110
 quiz-0.106
 ju-0.103
 contest-0.103
 makeup-0.102
Token "
Token ID366
Feature activation-1.16e-01

Pos logit contributions
 ju+0.025
 golf+0.024
 makeup+0.023
 swing+0.021
 butter+0.021
Neg logit contributions
rum-0.040
ously-0.039
 despair-0.037
aced-0.036
 horror-0.035
Token didn
Token ID1422
Feature activation-2.48e-02

Pos logit contributions
 ju+0.024
 golf+0.024
 makeup+0.023
 butter+0.022
mons+0.022
Neg logit contributions
rum-0.037
 deeply-0.034
ously-0.034
 despair-0.033
 jaws-0.033
Token't
Token ID470
Feature activation+1.74e-02

Pos logit contributions
 ju+0.036
 golf+0.034
 makeup+0.033
mons+0.033
 contest+0.032
Neg logit contributions
ously-0.046
rum-0.045
 despair-0.043
 deeply-0.043
 horror-0.041
Token try
Token ID1949
Feature activation-1.90e-03

Pos logit contributions
ously+0.030
rum+0.030
 despair+0.028
 deeply+0.027
 horror+0.026
Neg logit contributions
 ju-0.028
mons-0.026
 golf-0.026
 makeup-0.026
 grapes-0.025
Token to
Token ID284
Feature activation+2.20e-02

Pos logit contributions
 ju+0.003
mons+0.003
atmeal+0.003
 grapes+0.003
 contest+0.003
Neg logit contributions
ously-0.003
rum-0.003
 deeply-0.003
 despair-0.003
 silently-0.003
Token hang
Token ID8181
Feature activation-2.23e-03

Pos logit contributions
ously+0.040
rum+0.039
 despair+0.038
 deeply+0.037
 horror+0.036
Neg logit contributions
 ju-0.033
 golf-0.031
mons-0.031
 makeup-0.030
 grapes-0.030
Token from
Token ID422
Feature activation+1.29e-01

Pos logit contributions
 ju+0.004
atmeal+0.004
mons+0.004
Draw+0.004
 golf+0.004
Neg logit contributions
 deeply-0.003
rum-0.003
ously-0.003
 despair-0.003
 sadly-0.003
Token the
Token ID262
Feature activation-1.10e-02

Pos logit contributions
ously+0.210
rum+0.207
 deeply+0.199
 despair+0.198
 horror+0.192
Neg logit contributions
 ju-0.217
 golf-0.200
mons-0.200
atmeal-0.196
 makeup-0.195
Token truck
Token ID7779
Feature activation-1.30e-02

Pos logit contributions
 ju+0.016
 golf+0.015
mons+0.015
 makeup+0.015
atmeal+0.014
Neg logit contributions
ously-0.021
rum-0.021
 despair-0.020
 deeply-0.019
 horror-0.019
Token.
Token ID13
Feature activation+2.59e-02

Pos logit contributions
 ju+0.025
mons+0.024
 makeup+0.024
 golf+0.024
 grapes+0.023
Neg logit contributions
ously-0.018
rum-0.017
 despair-0.016
 deeply-0.016
 horror-0.016
Token
Token ID198
Feature activation+3.04e-02

Pos logit contributions
rum+0.042
ously+0.041
 despair+0.038
 deeply+0.037
 jaws+0.035
Neg logit contributions
 ju-0.045
 makeup-0.045
 golf-0.044
mons-0.044
 grapes-0.042
Token
Token ID198
Feature activation+2.90e-02

Pos logit contributions
ously+0.054
 despair+0.053
rum+0.053
 jaws+0.052
 horror+0.051
Neg logit contributions
 golf-0.044
 makeup-0.043
Draw-0.042
 ju-0.041
mons-0.040
Token to
Token ID284
Feature activation+1.90e-02

Pos logit contributions
rum+0.022
ously+0.021
 despair+0.021
 horror+0.020
aced+0.020
Neg logit contributions
 golf-0.006
 makeup-0.005
 ju-0.005
 quiz-0.005
 contest-0.004
Token be
Token ID307
Feature activation-1.15e-02

Pos logit contributions
ously+0.042
rum+0.039
aced+0.038
 deeply+0.038
urbed+0.037
Neg logit contributions
 golf-0.022
 makeup-0.021
 swing-0.021
 contest-0.020
 quiz-0.020
Token with
Token ID351
Feature activation+3.41e-02

Pos logit contributions
 ju+0.020
mons+0.019
 golf+0.018
 makeup+0.018
 grapes+0.018
Neg logit contributions
ously-0.019
rum-0.018
 despair-0.017
 deeply-0.017
 horror-0.016
Token his
Token ID465
Feature activation+2.26e-02

Pos logit contributions
ously+0.041
rum+0.037
aced+0.032
 despair+0.031
 deeply+0.031
Neg logit contributions
 makeup-0.073
 golf-0.073
 ju-0.073
 butter-0.069
 grapes-0.069
Token coat
Token ID13209
Feature activation-2.62e-02

Pos logit contributions
ously+0.045
rum+0.045
 despair+0.043
 deeply+0.042
 horror+0.041
Neg logit contributions
 ju-0.030
 golf-0.029
 makeup-0.028
mons-0.027
 butter-0.026
Token again
Token ID757
Feature activation+1.13e-01

Pos logit contributions
 ju+0.045
 golf+0.043
 makeup+0.042
mons+0.042
 contest+0.040
Neg logit contributions
ously-0.043
rum-0.041
 despair-0.039
 deeply-0.039
 horror-0.037
Token.
Token ID13
Feature activation+3.61e-02

Pos logit contributions
ously+0.215
rum+0.211
 despair+0.198
 horror+0.197
 deeply+0.195
Neg logit contributions
 golf-0.163
 ju-0.155
 makeup-0.148
 butter-0.143
 gum-0.142
Token<|endoftext|>
Token ID50256
Feature activation-3.70e-02

Pos logit contributions
ously+0.076
 plea+0.074
 despair+0.073
rum+0.073
 jaws+0.073
Neg logit contributions
 golf-0.035
 ju-0.035
 gum-0.033
Draw-0.032
 that-0.032
TokenOnce
Token ID7454
Feature activation-1.75e-02

Pos logit contributions
 ju+0.059
 golf+0.059
 makeup+0.057
 contest+0.056
 gum+0.054
Neg logit contributions
ously-0.062
rum-0.061
 despair-0.058
 deeply-0.055
 horror-0.055
Token there
Token ID612
Feature activation-3.14e-02

Pos logit contributions
 contest+0.031
 ju+0.031
 golf+0.030
 makeup+0.030
 paints+0.029
Neg logit contributions
rum-0.025
ously-0.025
 despair-0.024
 heartbeat-0.023
 horror-0.022
Token was
Token ID373
Feature activation-1.12e-04

Pos logit contributions
 ju+0.043
 golf+0.041
 makeup+0.041
 contest+0.039
mons+0.038
Neg logit contributions
ously-0.061
rum-0.060
 despair-0.056
 deeply-0.056
 heartbeat-0.055
Token sticks
Token ID16461
Feature activation-3.03e-03

Pos logit contributions
 ju+0.036
 golf+0.034
 makeup+0.033
mons+0.033
 grapes+0.032
Neg logit contributions
ously-0.046
rum-0.046
 despair-0.042
 deeply-0.042
 horror-0.040
Token.
Token ID13
Feature activation-9.25e-03

Pos logit contributions
 golf+0.004
 ju+0.004
 contest+0.004
 makeup+0.004
 butter+0.004
Neg logit contributions
rum-0.005
ously-0.005
 despair-0.005
 horror-0.005
 deeply-0.005
Token The
Token ID383
Feature activation-2.58e-02

Pos logit contributions
 golf+0.016
 ju+0.015
 makeup+0.014
 grapes+0.014
mons+0.014
Neg logit contributions
ously-0.016
rum-0.015
 despair-0.014
 deeply-0.014
aced-0.013
Token bird
Token ID6512
Feature activation-1.97e-03

Pos logit contributions
 ju+0.033
mons+0.031
atmeal+0.031
 makeup+0.030
 golf+0.030
Neg logit contributions
ously-0.050
 despair-0.048
rum-0.047
 horror-0.046
 deeply-0.046
Token flew
Token ID13112
Feature activation-3.20e-03

Pos logit contributions
 ju+0.003
 golf+0.003
 makeup+0.003
 contest+0.003
mons+0.003
Neg logit contributions
ously-0.004
rum-0.004
 despair-0.003
 deeply-0.003
 horror-0.003
Token out
Token ID503
Feature activation+1.33e-01

Pos logit contributions
 ju+0.005
 golf+0.004
 contest+0.004
 makeup+0.004
 grapes+0.004
Neg logit contributions
rum-0.006
ously-0.006
 despair-0.006
 horror-0.006
 heartbeat-0.005
Token of
Token ID286
Feature activation+7.30e-02

Pos logit contributions
rum+0.274
ously+0.266
 despair+0.256
 heartbeat+0.243
 plea+0.237
Neg logit contributions
 golf-0.157
 contest-0.151
 ju-0.148
 grapes-0.147
 butter-0.146
Token the
Token ID262
Feature activation+1.29e-02

Pos logit contributions
ously+0.150
rum+0.148
 deeply+0.134
 despair+0.131
aced+0.129
Neg logit contributions
 golf-0.100
 ju-0.095
 grapes-0.095
 makeup-0.095
 butter-0.092
Token nest
Token ID16343
Feature activation+2.13e-02

Pos logit contributions
ously+0.020
 despair+0.020
 deeply+0.019
 horror+0.019
 jaws+0.019
Neg logit contributions
 ju-0.021
mons-0.021
atmeal-0.020
 makeup-0.019
 golf-0.019
Token and
Token ID290
Feature activation-1.13e-03

Pos logit contributions
rum+0.042
ously+0.041
 despair+0.041
 horror+0.039
 heartbeat+0.037
Neg logit contributions
 ju-0.028
 golf-0.027
 contest-0.026
 butter-0.026
mons-0.026
Token attacked
Token ID7384
Feature activation+1.54e-02

Pos logit contributions
 ju+0.002
 golf+0.001
 butter+0.001
mons+0.001
 makeup+0.001
Neg logit contributions
rum-0.002
ously-0.002
 despair-0.002
 deeply-0.002
 horror-0.002
Token.
Token ID13
Feature activation+1.62e-03

Pos logit contributions
 ju+0.034
 golf+0.033
 makeup+0.032
 contest+0.032
 butter+0.032
Neg logit contributions
ously-0.038
rum-0.037
 despair-0.035
 deeply-0.034
 heartbeat-0.034
Token The
Token ID383
Feature activation-1.81e-02

Pos logit contributions
ously+0.003
rum+0.003
 despair+0.003
 deeply+0.003
aced+0.002
Neg logit contributions
 ju-0.002
 golf-0.002
 makeup-0.002
mons-0.002
 grapes-0.002
Token snake
Token ID17522
Feature activation+4.88e-03

Pos logit contributions
 ju+0.028
 golf+0.027
 makeup+0.026
mons+0.026
atmeal+0.025
Neg logit contributions
ously-0.032
rum-0.032
 despair-0.030
 deeply-0.030
 horror-0.029
Token slit
Token ID40724
Feature activation-1.21e-02

Pos logit contributions
ously+0.009
rum+0.009
 despair+0.008
 deeply+0.008
 horror+0.008
Neg logit contributions
 ju-0.007
 golf-0.007
 makeup-0.007
mons-0.007
 butter-0.006
Tokenhered
Token ID6083
Feature activation+5.16e-02

Pos logit contributions
 golf+0.018
 ju+0.018
mons+0.017
 makeup+0.017
 contest+0.017
Neg logit contributions
ously-0.021
rum-0.021
 deeply-0.020
 despair-0.020
 horror-0.019
Token away
Token ID1497
Feature activation+1.25e-01

Pos logit contributions
rum+0.090
ously+0.088
 deeply+0.086
 despair+0.086
 horror+0.081
Neg logit contributions
 golf-0.073
 makeup-0.071
 ju-0.071
 contest-0.068
 grapes-0.068
Token.
Token ID13
Feature activation+4.96e-03

Pos logit contributions
ously+0.285
 despair+0.280
rum+0.278
obby+0.275
 tal+0.275
Neg logit contributions
 contest-0.095
 golf-0.094
 swing-0.092
 that-0.090
 makeup-0.088
Token
Token ID198
Feature activation+8.99e-03

Pos logit contributions
ously+0.010
rum+0.010
aced+0.009
 despair+0.009
urbed+0.009
Neg logit contributions
 golf-0.007
 makeup-0.007
 ju-0.007
 gum-0.006
 grapes-0.006
Token
Token ID198
Feature activation+5.66e-03

Pos logit contributions
ously+0.019
rum+0.018
 despair+0.018
urbed+0.017
 jaws+0.017
Neg logit contributions
 golf-0.011
 makeup-0.010
Draw-0.010
 gum-0.010
mons-0.009
TokenMom
Token ID29252
Feature activation-1.42e-02

Pos logit contributions
ously+0.012
rum+0.012
 despair+0.011
 deeply+0.011
 horror+0.011
Neg logit contributions
 ju-0.007
 golf-0.007
 makeup-0.006
mons-0.006
 contest-0.006
Token and
Token ID290
Feature activation-3.13e-02

Pos logit contributions
 ju+0.020
 golf+0.019
 makeup+0.019
 butter+0.018
mons+0.018
Neg logit contributions
rum-0.026
ously-0.026
 despair-0.025
 deeply-0.025
 horror-0.023
Token.
Token ID13
Feature activation+3.61e-02

Pos logit contributions
ously+0.024
rum+0.023
 despair+0.022
 deeply+0.022
 horror+0.021
Neg logit contributions
 ju-0.020
 golf-0.019
 makeup-0.018
mons-0.018
 contest-0.018
Token
Token ID198
Feature activation+3.67e-02

Pos logit contributions
rum+0.079
ously+0.075
aced+0.069
 despair+0.069
urbed+0.068
Neg logit contributions
 makeup-0.041
 ju-0.041
 golf-0.040
 grapes-0.039
 contest-0.038
Token
Token ID198
Feature activation+3.56e-02

Pos logit contributions
ously+0.057
rum+0.055
 despair+0.053
 horror+0.051
 deeply+0.051
Neg logit contributions
 ju-0.063
 golf-0.062
 makeup-0.060
 contest-0.059
 grapes-0.058
Token"
Token ID1
Feature activation-1.42e-01

Pos logit contributions
ously+0.060
rum+0.058
 despair+0.056
 deeply+0.055
 silently+0.052
Neg logit contributions
 ju-0.058
mons-0.055
atmeal-0.055
 golf-0.055
 makeup-0.054
TokenDa
Token ID26531
Feature activation+3.80e-02

Pos logit contributions
 golf+0.251
mons+0.247
atmeal+0.246
 ju+0.245
 quiz+0.240
Neg logit contributions
ously-0.214
 despair-0.205
rum-0.195
 horror-0.181
 deeply-0.178
Tokenisy
Token ID13560
Feature activation+3.06e-02

Pos logit contributions
ously+0.062
rum+0.061
 despair+0.057
 deeply+0.057
 horror+0.052
Neg logit contributions
 ju-0.065
 golf-0.064
 makeup-0.062
mons-0.061
 paints-0.060
Token,
Token ID11
Feature activation-4.28e-02

Pos logit contributions
ously+0.049
rum+0.048
 despair+0.045
 deeply+0.045
 horror+0.043
Neg logit contributions
 ju-0.053
 golf-0.051
mons-0.049
 makeup-0.049
 grapes-0.048
Token come
Token ID1282
Feature activation-2.30e-02

Pos logit contributions
mons+0.068
 ju+0.066
 golf+0.065
atmeal+0.062
Draw+0.059
Neg logit contributions
ously-0.072
 despair-0.072
rum-0.070
 horror-0.068
 deeply-0.068
Token inside
Token ID2641
Feature activation+7.08e-02

Pos logit contributions
 ju+0.035
 golf+0.033
mons+0.032
 makeup+0.032
 grapes+0.031
Neg logit contributions
ously-0.042
rum-0.041
 despair-0.039
 deeply-0.038
 horror-0.037
Token now
Token ID783
Feature activation+1.57e-02

Pos logit contributions
ously+0.143
rum+0.136
 despair+0.129
 deeply+0.125
 horror+0.124
Neg logit contributions
 ju-0.093
 golf-0.091
 makeup-0.088
 butter-0.085
 contest-0.085
Token,"
Token ID553
Feature activation+2.30e-03

Pos logit contributions
ously+0.028
rum+0.027
 despair+0.026
 deeply+0.025
 horror+0.025
Neg logit contributions
 ju-0.024
 golf-0.023
 makeup-0.023
mons-0.022
 contest-0.022
Token."
Token ID526
Feature activation+2.32e-02

Pos logit contributions
ously+0.034
rum+0.034
 deeply+0.031
 despair+0.031
 horror+0.029
Neg logit contributions
 ju-0.046
 golf-0.044
 makeup-0.043
mons-0.042
 grapes-0.041
Token Lucy
Token ID22162
Feature activation+3.62e-02

Pos logit contributions
rum+0.040
ously+0.040
 despair+0.037
 jaws+0.035
 horror+0.035
Neg logit contributions
 makeup-0.038
 ju-0.036
mons-0.034
 grapes-0.034
 golf-0.034
Token nodded
Token ID14464
Feature activation+9.32e-02

Pos logit contributions
rum+0.070
ously+0.069
 despair+0.065
 deeply+0.064
 horror+0.063
Neg logit contributions
 ju-0.048
 golf-0.045
 makeup-0.044
mons-0.044
 butter-0.044
Token and
Token ID290
Feature activation-1.07e-02

Pos logit contributions
rum+0.183
ously+0.178
 despair+0.168
 jaws+0.164
 horror+0.163
Neg logit contributions
 makeup-0.122
 golf-0.116
 ju-0.113
 butter-0.110
 paints-0.106
Token went
Token ID1816
Feature activation-1.88e-02

Pos logit contributions
 ju+0.013
 golf+0.012
 butter+0.012
 makeup+0.012
mons+0.012
Neg logit contributions
rum-0.023
ously-0.020
 deeply-0.020
 jaws-0.020
opher-0.019
Token outside
Token ID2354
Feature activation+6.62e-02

Pos logit contributions
 ju+0.024
 golf+0.024
 makeup+0.023
mons+0.022
 contest+0.022
Neg logit contributions
ously-0.038
rum-0.037
 despair-0.036
 deeply-0.035
 horror-0.034
Token to
Token ID284
Feature activation+2.69e-02

Pos logit contributions
rum+0.154
ously+0.152
 despair+0.144
 jaws+0.144
aced+0.142
Neg logit contributions
 golf-0.062
 contest-0.060
 makeup-0.059
 quiz-0.058
 competition-0.056
Token look
Token ID804
Feature activation+8.07e-03

Pos logit contributions
ously+0.059
rum+0.058
 deeply+0.055
 despair+0.054
 horror+0.052
Neg logit contributions
 ju-0.031
 golf-0.030
 makeup-0.030
 contest-0.028
 quiz-0.028
Token for
Token ID329
Feature activation+1.08e-02

Pos logit contributions
ously+0.016
rum+0.016
 despair+0.015
 plea+0.014
 horror+0.014
Neg logit contributions
 ju-0.011
 golf-0.011
 makeup-0.011
 grapes-0.010
 quiz-0.010
Token the
Token ID262
Feature activation-2.39e-02

Pos logit contributions
ously+0.024
rum+0.024
 deeply+0.021
aced+0.021
urbed+0.021
Neg logit contributions
 makeup-0.013
 ju-0.012
 golf-0.012
 butter-0.012
 grapes-0.012
Token mystery
Token ID10715
Feature activation+7.17e-03

Pos logit contributions
 ju+0.035
 golf+0.033
 makeup+0.032
mons+0.032
 contest+0.031
Neg logit contributions
ously-0.045
rum-0.044
 despair-0.042
 deeply-0.041
 horror-0.040
Token!"
Token ID2474
Feature activation-3.97e-04

Pos logit contributions
 ju+0.069
 golf+0.065
 makeup+0.064
mons+0.064
 grapes+0.062
Neg logit contributions
ously-0.068
rum-0.067
 deeply-0.063
 despair-0.062
 horror-0.059
Token
Token ID198
Feature activation+2.79e-02

Pos logit contributions
 ju+0.001
 golf+0.001
 mixer+0.001
mons+0.001
 makeup+0.001
Neg logit contributions
rum-0.001
 despair-0.001
ously-0.001
 horror-0.001
aced-0.001
Token
Token ID198
Feature activation+2.50e-02

Pos logit contributions
 despair+0.049
rum+0.049
ously+0.049
 jaws+0.048
 horror+0.048
Neg logit contributions
 ju-0.043
 golf-0.040
 makeup-0.038
 mixer-0.038
 contest-0.038
TokenThe
Token ID464
Feature activation-7.75e-03

Pos logit contributions
ously+0.047
rum+0.046
 despair+0.044
 deeply+0.043
 horror+0.042
Neg logit contributions
 ju-0.037
 golf-0.035
 makeup-0.034
mons-0.033
 contest-0.032
Token little
Token ID1310
Feature activation+1.26e-02

Pos logit contributions
 ju+0.012
mons+0.012
 golf+0.012
atmeal+0.012
 makeup+0.012
Neg logit contributions
ously-0.013
rum-0.012
 despair-0.012
 deeply-0.012
 horror-0.012
Token girl
Token ID2576
Feature activation+5.58e-02

Pos logit contributions
ously+0.020
 despair+0.018
 horror+0.017
rum+0.017
 deeply+0.017
Neg logit contributions
 ju-0.021
atmeal-0.021
mons-0.021
 golf-0.020
 contest-0.020
Token jumped
Token ID11687
Feature activation-3.51e-02

Pos logit contributions
rum+0.102
ously+0.101
 despair+0.095
 deeply+0.093
 horror+0.091
Neg logit contributions
 ju-0.082
 golf-0.081
 makeup-0.077
mons-0.075
 paints-0.075
Token with
Token ID351
Feature activation+5.20e-02

Pos logit contributions
 ju+0.052
 golf+0.052
 makeup+0.048
 contest+0.047
 competition+0.046
Neg logit contributions
ously-0.066
rum-0.065
 despair-0.061
 horror-0.058
 deeply-0.058
Token excitement
Token ID14067
Feature activation+5.50e-02

Pos logit contributions
rum+0.071
ously+0.071
 deeply+0.062
 despair+0.061
urbed+0.057
Neg logit contributions
 ju-0.104
 golf-0.101
 makeup-0.100
 grapes-0.096
 butter-0.096
Token.
Token ID13
Feature activation+3.37e-02

Pos logit contributions
rum+0.092
ously+0.091
 despair+0.088
 horror+0.083
 deeply+0.082
Neg logit contributions
 ju-0.088
 golf-0.087
 makeup-0.085
mons-0.081
 butter-0.081
Token She
Token ID1375
Feature activation+3.30e-02

Pos logit contributions
rum+0.061
ously+0.059
 despair+0.058
 deeply+0.056
 horror+0.053
Neg logit contributions
 ju-0.053
 golf-0.050
mons-0.049
 makeup-0.048
 grapes-0.047
Tokenâ„¢
Token ID8151
Feature activation-1.27e-01

Pos logit contributions
rum+0.019
ously+0.019
 despair+0.017
 deeply+0.017
 horror+0.016
Neg logit contributions
 ju-0.022
 makeup-0.021
 golf-0.020
 grapes-0.020
 butter-0.020
Tokens
Token ID82
Feature activation+4.11e-03

Pos logit contributions
 ju+0.172
mons+0.167
 golf+0.163
 contest+0.160
 makeup+0.156
Neg logit contributions
 despair-0.243
ously-0.241
rum-0.236
 horror-0.232
 deeply-0.229
Token pick
Token ID2298
Feature activation-5.85e-02

Pos logit contributions
ously+0.008
rum+0.008
 despair+0.007
 deeply+0.007
 horror+0.007
Neg logit contributions
 ju-0.006
 golf-0.005
 makeup-0.005
mons-0.005
 contest-0.005
Token some
Token ID617
Feature activation-2.27e-02

Pos logit contributions
 ju+0.090
atmeal+0.085
 contest+0.085
 quiz+0.085
mons+0.084
Neg logit contributions
ously-0.096
rum-0.096
 deeply-0.089
 sadly-0.086
 silently-0.085
Token flowers
Token ID12734
Feature activation+9.32e-03

Pos logit contributions
 ju+0.026
atmeal+0.025
mons+0.024
 contest+0.023
 golf+0.023
Neg logit contributions
rum-0.045
ously-0.044
 deeply-0.044
 despair-0.042
 jaws-0.041
Token to
Token ID284
Feature activation+1.73e-02

Pos logit contributions
ously+0.014
rum+0.013
 deeply+0.013
 despair+0.012
 silently+0.012
Neg logit contributions
 ju-0.016
 golf-0.016
 makeup-0.015
mons-0.015
atmeal-0.015
Token take
Token ID1011
Feature activation-2.51e-02

Pos logit contributions
ously+0.031
rum+0.031
 despair+0.030
 deeply+0.030
 silently+0.029
Neg logit contributions
 ju-0.023
mons-0.023
atmeal-0.021
 golf-0.021
 makeup-0.020
Token home
Token ID1363
Feature activation+6.74e-02

Pos logit contributions
 ju+0.039
mons+0.036
 golf+0.036
atmeal+0.036
 contest+0.035
Neg logit contributions
ously-0.042
rum-0.042
 deeply-0.040
 despair-0.039
 silently-0.037
Token with
Token ID351
Feature activation+2.20e-02

Pos logit contributions
ously+0.112
rum+0.110
 despair+0.103
 deeply+0.102
 horror+0.098
Neg logit contributions
 ju-0.111
 golf-0.107
 makeup-0.104
mons-0.103
 contest-0.101
Token us
Token ID514
Feature activation+1.06e-02

Pos logit contributions
ously+0.038
rum+0.038
 despair+0.036
 deeply+0.035
 horror+0.034
Neg logit contributions
 ju-0.035
 golf-0.033
 makeup-0.032
mons-0.032
 contest-0.031
Token.
Token ID13
Feature activation+1.97e-02

Pos logit contributions
ously+0.018
rum+0.018
 despair+0.017
 deeply+0.016
 horror+0.016
Neg logit contributions
 ju-0.017
 golf-0.016
 makeup-0.016
mons-0.016
 contest-0.015
Token we
Token ID356
Feature activation-6.94e-02

Pos logit contributions
 ju+0.076
 golf+0.071
mons+0.070
 makeup+0.067
 grapes+0.065
Neg logit contributions
rum-0.124
ously-0.122
 despair-0.121
 deeply-0.119
 horror-0.115
Token go
Token ID467
Feature activation-4.40e-02

Pos logit contributions
 ju+0.097
mons+0.094
 golf+0.091
 makeup+0.090
 grapes+0.089
Neg logit contributions
ously-0.125
rum-0.123
 despair-0.121
 deeply-0.120
 silently-0.115
Token inside
Token ID2641
Feature activation+5.84e-02

Pos logit contributions
 ju+0.075
mons+0.073
 grapes+0.069
 makeup+0.067
 golf+0.066
Neg logit contributions
ously-0.067
 deeply-0.066
rum-0.065
 silently-0.064
 sadly-0.063
Token?"
Token ID1701
Feature activation+5.22e-04

Pos logit contributions
ously+0.120
rum+0.117
 despair+0.111
 deeply+0.109
 horror+0.107
Neg logit contributions
 ju-0.075
 golf-0.072
 makeup-0.069
 contest-0.066
mons-0.066
Token Lily
Token ID20037
Feature activation-3.21e-03

Pos logit contributions
ously+0.001
rum+0.001
 deeply+0.001
aced+0.001
 despair+0.001
Neg logit contributions
 golf-0.001
 ju-0.001
 makeup-0.001
 grapes-0.001
 paints-0.001
Token asked
Token ID1965
Feature activation+2.90e-02

Pos logit contributions
 ju+0.005
 makeup+0.004
 golf+0.004
 butter+0.004
 paints+0.004
Neg logit contributions
ously-0.006
rum-0.006
 despair-0.006
 deeply-0.005
 horror-0.005
Token.
Token ID13
Feature activation+8.64e-03

Pos logit contributions
rum+0.058
ously+0.057
 despair+0.054
 jaws+0.052
urbed+0.051
Neg logit contributions
 makeup-0.038
 golf-0.036
 butter-0.035
 ju-0.034
 paints-0.033
Token
Token ID198
Feature activation+1.22e-02

Pos logit contributions
rum+0.015
ously+0.015
 despair+0.014
aced+0.014
urbed+0.014
Neg logit contributions
 golf-0.014
 ju-0.013
 makeup-0.013
mons-0.013
 paints-0.013
Token
Token ID198
Feature activation+9.09e-03

Pos logit contributions
ously+0.026
 despair+0.026
 jaws+0.025
rum+0.025
urbed+0.024
Neg logit contributions
 golf-0.014
 makeup-0.013
 paints-0.013
Draw-0.013
 mixer-0.013
Token"
Token ID1
Feature activation-1.56e-01

Pos logit contributions
ously+0.018
rum+0.018
 despair+0.017
 deeply+0.017
 horror+0.016
Neg logit contributions
 ju-0.012
 golf-0.012
 makeup-0.011
mons-0.011
 contest-0.011
TokenNo
Token ID2949
Feature activation-3.45e-02

Pos logit contributions
atmeal+0.231
 ju+0.219
 makeup+0.218
 golf+0.218
 butter+0.215
Neg logit contributions
ously-0.272
 despair-0.242
 deeply-0.242
rum-0.240
 silently-0.238
Token.
Token ID13
Feature activation+3.67e-02

Pos logit contributions
rum+0.177
ously+0.172
 despair+0.162
 horror+0.159
 jaws+0.156
Neg logit contributions
 ju-0.106
 golf-0.104
 butter-0.093
 makeup-0.093
 quiz-0.089
Token<|endoftext|>
Token ID50256
Feature activation-3.95e-02

Pos logit contributions
ously+0.083
rum+0.080
aced+0.077
 plea+0.075
 despair+0.075
Neg logit contributions
 ju-0.038
 golf-0.037
 gum-0.032
 How-0.032
Draw-0.031
TokenOnce
Token ID7454
Feature activation-2.09e-02

Pos logit contributions
 golf+0.060
 ju+0.060
 makeup+0.058
 contest+0.058
 gum+0.056
Neg logit contributions
ously-0.070
rum-0.069
 despair-0.066
 horror-0.063
 deeply-0.062
Token upon
Token ID2402
Feature activation-1.66e-02

Pos logit contributions
 contest+0.034
 golf+0.034
 ju+0.034
 makeup+0.034
 mixer+0.032
Neg logit contributions
rum-0.033
ously-0.031
 despair-0.031
 heartbeat-0.030
edly-0.029
Token a
Token ID257
Feature activation-5.05e-02

Pos logit contributions
 ju+0.022
 golf+0.021
 makeup+0.020
mons+0.019
 grapes+0.019
Neg logit contributions
ously-0.034
rum-0.033
 despair-0.031
 deeply-0.031
 horror-0.030
Token time
Token ID640
Feature activation-3.84e-02

Pos logit contributions
 ju+0.076
 golf+0.074
 makeup+0.071
mons+0.070
 contest+0.069
Neg logit contributions
ously-0.091
rum-0.089
 despair-0.084
 deeply-0.083
 horror-0.080
Token,
Token ID11
Feature activation-5.57e-02

Pos logit contributions
 golf+0.055
 ju+0.054
 makeup+0.053
 contest+0.052
mons+0.051
Neg logit contributions
ously-0.071
rum-0.070
 despair-0.067
 deeply-0.065
 horror-0.064
Token there
Token ID612
Feature activation-3.12e-02

Pos logit contributions
 ju+0.036
 golf+0.033
 makeup+0.031
mons+0.030
 contest+0.028
Neg logit contributions
ously-0.149
rum-0.147
 despair-0.142
 deeply-0.140
 horror-0.137
Token was
Token ID373
Feature activation-3.10e-03

Pos logit contributions
 ju+0.040
 golf+0.040
 makeup+0.039
 contest+0.037
 gum+0.036
Neg logit contributions
ously-0.062
rum-0.061
 despair-0.057
 deeply-0.057
 heartbeat-0.055
Token a
Token ID257
Feature activation-3.95e-02

Pos logit contributions
 ju+0.005
mons+0.005
 golf+0.004
 makeup+0.004
atmeal+0.004
Neg logit contributions
ously-0.005
rum-0.005
 deeply-0.005
 despair-0.005
 horror-0.005
Token little
Token ID1310
Feature activation-1.08e-02

Pos logit contributions
mons+0.077
 ju+0.075
 makeup+0.072
atmeal+0.072
 newspapers+0.070
Neg logit contributions
 deeply-0.047
ously-0.045
 horror-0.045
 despair-0.045
rum-0.043
Token the
Token ID262
Feature activation-1.83e-02

Pos logit contributions
 ju+0.059
 golf+0.056
 makeup+0.054
mons+0.054
 contest+0.052
Neg logit contributions
ously-0.073
rum-0.072
 despair-0.068
 deeply-0.067
 horror-0.065
Token sand
Token ID6450
Feature activation-4.39e-02

Pos logit contributions
 ju+0.026
 golf+0.024
mons+0.024
 makeup+0.023
atmeal+0.023
Neg logit contributions
ously-0.034
rum-0.033
 despair-0.032
 deeply-0.032
 horror-0.031
Token,
Token ID11
Feature activation-3.40e-02

Pos logit contributions
 ju+0.059
 golf+0.056
 makeup+0.054
mons+0.053
 contest+0.051
Neg logit contributions
ously-0.087
rum-0.085
 despair-0.080
 deeply-0.080
 horror-0.077
Token and
Token ID290
Feature activation-2.77e-02

Pos logit contributions
 ju+0.052
mons+0.051
 golf+0.050
 makeup+0.049
atmeal+0.049
Neg logit contributions
ously-0.059
rum-0.056
 despair-0.055
 deeply-0.054
 horror-0.052
Token the
Token ID262
Feature activation-2.41e-02

Pos logit contributions
 ju+0.042
 golf+0.041
 makeup+0.040
mons+0.039
 contest+0.038
Neg logit contributions
ously-0.049
rum-0.048
 despair-0.046
 deeply-0.045
 horror-0.043
Token water
Token ID1660
Feature activation-2.52e-02

Pos logit contributions
 ju+0.032
 golf+0.030
mons+0.029
 makeup+0.029
 contest+0.028
Neg logit contributions
ously-0.047
rum-0.046
 despair-0.044
 deeply-0.044
 horror-0.043
Token.
Token ID13
Feature activation+6.72e-03

Pos logit contributions
 ju+0.038
 golf+0.037
 makeup+0.036
mons+0.035
 contest+0.034
Neg logit contributions
ously-0.045
rum-0.044
 despair-0.042
 deeply-0.041
 horror-0.040
Token She
Token ID1375
Feature activation+1.35e-02

Pos logit contributions
ously+0.011
rum+0.010
 despair+0.010
 deeply+0.009
aced+0.009
Neg logit contributions
 ju-0.012
 golf-0.012
 makeup-0.011
mons-0.011
 grapes-0.011
Token dreamed
Token ID27947
Feature activation-2.81e-02

Pos logit contributions
ously+0.026
rum+0.025
 despair+0.024
 deeply+0.023
 horror+0.023
Neg logit contributions
 ju-0.019
 golf-0.018
 makeup-0.018
mons-0.017
 contest-0.017
Token of
Token ID286
Feature activation+2.04e-02

Pos logit contributions
 ju+0.040
 golf+0.039
 makeup+0.038
 contest+0.037
 grapes+0.037
Neg logit contributions
rum-0.053
ously-0.051
 despair-0.050
 horror-0.048
 deeply-0.046
Token the
Token ID262
Feature activation-2.08e-02

Pos logit contributions
ously+0.038
rum+0.037
 despair+0.036
 deeply+0.035
 horror+0.034
Neg logit contributions
 ju-0.030
 golf-0.028
mons-0.027
 makeup-0.027
atmeal-0.026
Token on
Token ID319
Feature activation+4.85e-02

Pos logit contributions
 ju+0.029
mons+0.026
atmeal+0.026
 grapes+0.026
 makeup+0.025
Neg logit contributions
ously-0.027
 deeply-0.025
rum-0.024
 silently-0.023
 sadly-0.023
Token the
Token ID262
Feature activation-1.97e-02

Pos logit contributions
ously+0.093
rum+0.090
 despair+0.084
 deeply+0.083
 horror+0.080
Neg logit contributions
 ju-0.070
 golf-0.069
 makeup-0.067
mons-0.064
 grapes-0.063
Token table
Token ID3084
Feature activation-4.03e-02

Pos logit contributions
 ju+0.028
 golf+0.027
mons+0.026
 makeup+0.026
atmeal+0.025
Neg logit contributions
ously-0.036
rum-0.036
 despair-0.034
 deeply-0.034
 horror-0.033
Token.
Token ID13
Feature activation+2.18e-02

Pos logit contributions
 ju+0.081
 golf+0.076
 makeup+0.075
mons+0.074
atmeal+0.073
Neg logit contributions
ously-0.055
rum-0.052
 deeply-0.048
 despair-0.048
 horror-0.046
Token It
Token ID632
Feature activation-2.90e-02

Pos logit contributions
ously+0.027
rum+0.025
 despair+0.024
 deeply+0.024
 horror+0.023
Neg logit contributions
 ju-0.044
atmeal-0.043
 golf-0.042
 makeup-0.042
 quiz-0.042
Token had
Token ID550
Feature activation-3.50e-02

Pos logit contributions
 ju+0.054
 golf+0.052
mons+0.051
atmeal+0.050
 makeup+0.050
Neg logit contributions
ously-0.041
rum-0.040
 deeply-0.038
 despair-0.038
 horror-0.036
Token a
Token ID257
Feature activation-3.64e-02

Pos logit contributions
 ju+0.045
 golf+0.042
 makeup+0.042
 butter+0.040
 grapes+0.040
Neg logit contributions
rum-0.073
ously-0.072
 despair-0.068
 deeply-0.067
 horror-0.065
Token label
Token ID6167
Feature activation-3.61e-02

Pos logit contributions
 ju+0.045
 golf+0.043
 makeup+0.041
mons+0.040
 contest+0.039
Neg logit contributions
ously-0.077
rum-0.075
 despair-0.072
 deeply-0.070
 horror-0.069
Token that
Token ID326
Feature activation-5.54e-02

Pos logit contributions
 ju+0.058
atmeal+0.054
mons+0.048
 golf+0.047
 grapes+0.047
Neg logit contributions
ously-0.063
 deeply-0.060
rum-0.060
 sadly-0.056
 silently-0.055
Token said
Token ID531
Feature activation-5.29e-03

Pos logit contributions
 ju+0.090
 golf+0.087
mons+0.084
 makeup+0.083
atmeal+0.082
Neg logit contributions
ously-0.091
rum-0.088
 deeply-0.085
 despair-0.085
 horror-0.081
Token "
Token ID366
Feature activation-1.39e-01

Pos logit contributions
 ju+0.010
atmeal+0.010
 golf+0.009
mons+0.009
 grapes+0.009
Neg logit contributions
ously-0.008
rum-0.007
 deeply-0.007
 despair-0.007
 horror-0.007
Token.
Token ID13
Feature activation+1.60e-02

Pos logit contributions
ously+0.023
rum+0.022
 despair+0.020
 deeply+0.020
 horror+0.019
Neg logit contributions
 ju-0.031
 golf-0.029
 makeup-0.029
mons-0.028
 grapes-0.028
Token The
Token ID383
Feature activation-2.07e-02

Pos logit contributions
ously+0.023
rum+0.022
 despair+0.021
 deeply+0.020
 horror+0.020
Neg logit contributions
 ju-0.030
 golf-0.029
 makeup-0.029
mons-0.028
 contest-0.028
Token box
Token ID3091
Feature activation-1.12e-02

Pos logit contributions
atmeal+0.037
mons+0.033
 ju+0.033
iable+0.032
please+0.030
Neg logit contributions
 horror-0.032
 despair-0.032
 deeply-0.031
ously-0.030
 heart-0.029
Token is
Token ID318
Feature activation-1.27e-02

Pos logit contributions
 ju+0.018
 golf+0.017
mons+0.016
 makeup+0.016
atmeal+0.016
Neg logit contributions
ously-0.020
rum-0.019
 deeply-0.018
 despair-0.018
 horror-0.018
Token tight
Token ID5381
Feature activation+4.87e-02

Pos logit contributions
atmeal+0.019
 ju+0.018
mons+0.018
 golf+0.017
 contest+0.016
Neg logit contributions
ously-0.022
 deeply-0.022
rum-0.022
 despair-0.022
 horror-0.021
Token and
Token ID290
Feature activation-2.40e-02

Pos logit contributions
ously+0.069
rum+0.066
 deeply+0.064
 despair+0.060
 silently+0.058
Neg logit contributions
 ju-0.094
 golf-0.087
mons-0.087
atmeal-0.086
 makeup-0.085
Token hard
Token ID1327
Feature activation-1.25e-03

Pos logit contributions
 ju+0.033
 golf+0.032
atmeal+0.031
mons+0.031
 makeup+0.030
Neg logit contributions
ously-0.044
rum-0.044
 deeply-0.044
 despair-0.043
 horror-0.042
Token to
Token ID284
Feature activation+1.55e-02

Pos logit contributions
 ju+0.002
 golf+0.002
 makeup+0.002
mons+0.002
 contest+0.002
Neg logit contributions
ously-0.002
rum-0.002
 despair-0.002
 deeply-0.002
 horror-0.002
Token open
Token ID1280
Feature activation+4.30e-02

Pos logit contributions
ously+0.029
rum+0.029
 despair+0.028
 deeply+0.027
 horror+0.026
Neg logit contributions
 ju-0.022
 golf-0.021
mons-0.020
 makeup-0.020
 grapes-0.019
Token.
Token ID13
Feature activation+1.92e-02

Pos logit contributions
ously+0.085
rum+0.083
 despair+0.079
 deeply+0.078
 horror+0.075
Neg logit contributions
 ju-0.058
 golf-0.056
 makeup-0.054
mons-0.053
 contest-0.051
Token They
Token ID1119
Feature activation-4.72e-03

Pos logit contributions
ously+0.031
rum+0.030
 despair+0.028
 deeply+0.028
urbed+0.026
Neg logit contributions
 ju-0.034
 golf-0.032
mons-0.031
 makeup-0.031
 grapes-0.031
Token proud
Token ID6613
Feature activation+3.78e-02

Pos logit contributions
 ju+0.018
 golf+0.017
 makeup+0.016
mons+0.016
 swing+0.016
Neg logit contributions
ously-0.026
rum-0.025
 despair-0.022
urbed-0.022
 deeply-0.022
Token.
Token ID13
Feature activation+8.32e-03

Pos logit contributions
ously+0.065
rum+0.064
 despair+0.060
 deeply+0.057
 horror+0.056
Neg logit contributions
 golf-0.060
 ju-0.058
 makeup-0.058
 grapes-0.056
 contest-0.056
Token It
Token ID632
Feature activation-2.65e-02

Pos logit contributions
ously+0.019
rum+0.019
urbed+0.017
 despair+0.017
aced+0.017
Neg logit contributions
 golf-0.009
 grapes-0.008
 ju-0.008
 gum-0.007
mons-0.007
Token thinks
Token ID6834
Feature activation-1.25e-02

Pos logit contributions
 ju+0.036
 golf+0.033
 makeup+0.032
 contest+0.032
 quiz+0.032
Neg logit contributions
ously-0.054
rum-0.052
 despair-0.048
 deeply-0.047
aced-0.045
Token it
Token ID340
Feature activation-3.54e-02

Pos logit contributions
 ju+0.017
 golf+0.016
 makeup+0.016
 contest+0.015
mons+0.015
Neg logit contributions
ously-0.025
rum-0.024
 despair-0.023
 deeply-0.022
 horror-0.022
Token has
Token ID468
Feature activation-5.17e-02

Pos logit contributions
 ju+0.061
 golf+0.061
mons+0.058
 makeup+0.057
 grapes+0.056
Neg logit contributions
rum-0.056
ously-0.055
 deeply-0.053
 despair-0.052
 silently-0.050
Token won
Token ID1839
Feature activation+2.23e-02

Pos logit contributions
 ju+0.065
 golf+0.064
 makeup+0.062
 butter+0.060
 grapes+0.060
Neg logit contributions
ously-0.110
rum-0.109
 deeply-0.102
 despair-0.100
 horror-0.097
Token the
Token ID262
Feature activation-3.49e-02

Pos logit contributions
ously+0.043
rum+0.040
 despair+0.039
 horror+0.038
 deeply+0.038
Neg logit contributions
 golf-0.031
 makeup-0.030
 contest-0.030
 ju-0.029
mons-0.029
Token game
Token ID983
Feature activation-3.49e-02

Pos logit contributions
atmeal+0.056
 butter+0.048
 ju+0.046
mons+0.045
eating+0.044
Neg logit contributions
 horror-0.059
 despair-0.058
 fear-0.057
 plea-0.054
 pleas-0.052
Token.
Token ID13
Feature activation+3.67e-03

Pos logit contributions
 ju+0.065
 golf+0.062
 makeup+0.061
mons+0.061
atmeal+0.059
Neg logit contributions
ously-0.051
rum-0.049
 despair-0.046
 deeply-0.046
 horror-0.043
Token
Token ID198
Feature activation+2.66e-03

Pos logit contributions
ously+0.008
rum+0.008
 despair+0.007
urbed+0.007
 deeply+0.007
Neg logit contributions
 golf-0.005
 ju-0.004
 grapes-0.004
mons-0.004
 gum-0.004
Token a
Token ID257
Feature activation-3.32e-02

Pos logit contributions
ously+0.012
rum+0.012
 despair+0.011
 deeply+0.011
 horror+0.011
Neg logit contributions
 ju-0.008
 golf-0.008
 makeup-0.008
mons-0.008
 contest-0.007
Token fire
Token ID2046
Feature activation-5.52e-03

Pos logit contributions
 ju+0.046
mons+0.044
 golf+0.043
 makeup+0.043
atmeal+0.041
Neg logit contributions
ously-0.062
rum-0.061
 despair-0.059
 deeply-0.058
 horror-0.057
Token truck
Token ID7779
Feature activation-4.45e-02

Pos logit contributions
 ju+0.007
 golf+0.007
mons+0.007
 makeup+0.007
 contest+0.007
Neg logit contributions
ously-0.011
rum-0.010
 despair-0.010
 deeply-0.010
 horror-0.009
Token.
Token ID13
Feature activation+1.85e-02

Pos logit contributions
 ju+0.080
 makeup+0.073
mons+0.073
 golf+0.073
 grapes+0.072
Neg logit contributions
ously-0.068
rum-0.067
 deeply-0.063
 despair-0.062
 horror-0.060
Token They
Token ID1119
Feature activation-1.12e-02

Pos logit contributions
ously+0.028
rum+0.028
 despair+0.026
 deeply+0.025
 horror+0.024
Neg logit contributions
 ju-0.034
 golf-0.033
 makeup-0.032
mons-0.032
 grapes-0.031
Token wanted
Token ID2227
Feature activation-8.82e-02

Pos logit contributions
 ju+0.013
 paints+0.013
 golf+0.013
 makeup+0.012
 swing+0.012
Neg logit contributions
rum-0.024
ously-0.023
 deeply-0.022
 despair-0.021
 horror-0.021
Token to
Token ID284
Feature activation-4.57e-03

Pos logit contributions
 ju+0.126
 golf+0.117
mons+0.115
 makeup+0.114
atmeal+0.113
Neg logit contributions
ously-0.168
rum-0.163
 despair-0.157
 deeply-0.155
 horror-0.151
Token see
Token ID766
Feature activation-6.50e-02

Pos logit contributions
 golf+0.005
 ju+0.005
 swing+0.005
 contest+0.005
 makeup+0.004
Neg logit contributions
ously-0.011
rum-0.010
 deeply-0.010
 despair-0.009
urbed-0.009
Token it
Token ID340
Feature activation-1.56e-02

Pos logit contributions
 ju+0.100
 golf+0.095
mons+0.094
 makeup+0.093
 contest+0.091
Neg logit contributions
ously-0.115
rum-0.113
 despair-0.107
 deeply-0.106
 horror-0.103
Token up
Token ID510
Feature activation+7.68e-02

Pos logit contributions
 ju+0.024
 golf+0.023
 makeup+0.022
mons+0.022
 contest+0.021
Neg logit contributions
ously-0.028
rum-0.027
 despair-0.026
 deeply-0.026
 horror-0.025
Token close
Token ID1969
Feature activation+1.19e-01

Pos logit contributions
ously+0.143
rum+0.141
 despair+0.128
 horror+0.126
 deeply+0.126
Neg logit contributions
 golf-0.116
 ju-0.114
 contest-0.107
 makeup-0.105
 competition-0.104
Token
Token ID198
Feature activation+2.91e-02

Pos logit contributions
ously+0.053
rum+0.051
 despair+0.050
 horror+0.048
 deeply+0.048
Neg logit contributions
 golf-0.055
 ju-0.055
 makeup-0.054
mons-0.051
 gum-0.050
TokenL
Token ID43
Feature activation-2.90e-02

Pos logit contributions
ously+0.057
rum+0.056
 despair+0.053
 deeply+0.052
 horror+0.051
Neg logit contributions
 ju-0.040
 golf-0.037
mons-0.037
 makeup-0.036
 contest-0.035
Tokenily
Token ID813
Feature activation-2.02e-02

Pos logit contributions
 makeup+0.031
 ju+0.030
 golf+0.030
mons+0.029
 grapes+0.028
Neg logit contributions
ously-0.064
rum-0.062
 despair-0.062
 deeply-0.060
 horror-0.059
Token replied
Token ID8712
Feature activation+2.25e-02

Pos logit contributions
 ju+0.035
 golf+0.033
 makeup+0.032
mons+0.032
 contest+0.031
Neg logit contributions
ously-0.033
rum-0.032
 despair-0.030
 deeply-0.030
 horror-0.028
Token,
Token ID11
Feature activation-3.49e-02

Pos logit contributions
ously+0.039
rum+0.039
 despair+0.037
 deeply+0.036
 horror+0.035
Neg logit contributions
 ju-0.034
 golf-0.033
 makeup-0.033
mons-0.032
 contest-0.031
Token "
Token ID366
Feature activation-1.36e-01

Pos logit contributions
 ju+0.050
 golf+0.048
 makeup+0.047
mons+0.045
 grapes+0.044
Neg logit contributions
rum-0.066
ously-0.065
 despair-0.063
 horror-0.060
 deeply-0.059
TokenI
Token ID40
Feature activation-4.98e-02

Pos logit contributions
atmeal+0.194
 contest+0.190
 golf+0.189
 quiz+0.188
 grapes+0.180
Neg logit contributions
ously-0.250
 despair-0.231
rum-0.226
 deeply-0.215
 silently-0.214
Token fell
Token ID3214
Feature activation+1.38e-02

Pos logit contributions
 ju+0.083
 golf+0.080
 makeup+0.078
mons+0.078
 grapes+0.076
Neg logit contributions
ously-0.083
rum-0.081
 despair-0.076
 deeply-0.075
 horror-0.072
Token and
Token ID290
Feature activation-5.36e-03

Pos logit contributions
ously+0.025
rum+0.024
 despair+0.024
 horror+0.023
 deeply+0.023
Neg logit contributions
 makeup-0.020
 ju-0.020
 golf-0.020
 contest-0.019
 butter-0.018
Token got
Token ID1392
Feature activation-5.11e-02

Pos logit contributions
 ju+0.007
 golf+0.007
mons+0.007
 makeup+0.007
 grapes+0.006
Neg logit contributions
ously-0.010
rum-0.010
 despair-0.010
 deeply-0.010
 horror-0.009
Token a
Token ID257
Feature activation-2.79e-02

Pos logit contributions
 ju+0.073
 golf+0.070
 makeup+0.068
mons+0.067
 contest+0.065
Neg logit contributions
ously-0.097
rum-0.095
 despair-0.090
 deeply-0.089
 horror-0.086
Token says
Token ID1139
Feature activation-1.87e-02

Pos logit contributions
ously+0.006
rum+0.006
 deeply+0.005
 despair+0.005
 horror+0.005
Neg logit contributions
 ju-0.006
 golf-0.006
mons-0.006
 makeup-0.005
atmeal-0.005
Token to
Token ID284
Feature activation-1.19e-02

Pos logit contributions
 ju+0.035
mons+0.032
 golf+0.032
atmeal+0.032
 makeup+0.031
Neg logit contributions
ously-0.029
rum-0.027
 deeply-0.026
 despair-0.025
 horror-0.024
Token Anna
Token ID11735
Feature activation-1.00e-03

Pos logit contributions
 makeup+0.015
 golf+0.015
 quiz+0.015
 contest+0.014
 grapes+0.014
Neg logit contributions
ously-0.025
rum-0.024
aced-0.022
 despair-0.022
 deeply-0.022
Token,
Token ID11
Feature activation-5.88e-02

Pos logit contributions
 ju+0.002
 golf+0.002
 makeup+0.002
mons+0.001
 contest+0.001
Neg logit contributions
ously-0.002
rum-0.002
 despair-0.002
 deeply-0.002
 horror-0.002
Token "
Token ID366
Feature activation-1.49e-01

Pos logit contributions
 ju+0.089
 golf+0.087
 makeup+0.084
mons+0.082
 contest+0.081
Neg logit contributions
ously-0.105
rum-0.103
 despair-0.098
 deeply-0.096
 horror-0.093
TokenWhy
Token ID5195
Feature activation-9.48e-02

Pos logit contributions
 ju+0.182
 golf+0.176
 makeup+0.170
 grapes+0.165
atmeal+0.163
Neg logit contributions
ously-0.311
rum-0.302
 despair-0.291
 silently-0.272
 horror-0.271
Token don
Token ID836
Feature activation-6.27e-02

Pos logit contributions
 ju+0.135
 golf+0.123
mons+0.121
 makeup+0.118
 grapes+0.116
Neg logit contributions
ously-0.185
rum-0.184
 deeply-0.171
 despair-0.171
 horror-0.165
Token't
Token ID470
Feature activation+1.15e-02

Pos logit contributions
 ju+0.134
mons+0.124
 makeup+0.121
 grapes+0.118
atmeal+0.118
Neg logit contributions
rum-0.079
 deeply-0.077
ously-0.071
 despair-0.069
 horror-0.066
Token we
Token ID356
Feature activation-5.72e-02

Pos logit contributions
rum+0.025
ously+0.025
 despair+0.024
 deeply+0.023
 horror+0.023
Neg logit contributions
 ju-0.013
mons-0.012
 golf-0.011
 makeup-0.011
 grapes-0.011
Token share
Token ID2648
Feature activation-2.31e-03

Pos logit contributions
 ju+0.083
mons+0.077
 golf+0.076
 makeup+0.074
atmeal+0.073
Neg logit contributions
ously-0.105
rum-0.103
 despair-0.100
 deeply-0.098
 horror-0.095
Token our
Token ID674
Feature activation-5.99e-03

Pos logit contributions
 ju+0.004
atmeal+0.004
mons+0.004
 contest+0.004
 golf+0.004
Neg logit contributions
rum-0.003
ously-0.003
 deeply-0.003
 despair-0.003
 sadly-0.003
Token "
Token ID366
Feature activation-1.23e-01

Pos logit contributions
rum+0.013
ously+0.013
 despair+0.012
aced+0.012
 deeply+0.012
Neg logit contributions
 ju-0.009
 golf-0.009
 makeup-0.008
 grapes-0.008
 newspapers-0.008
TokenHello
Token ID15496
Feature activation-3.54e-03

Pos logit contributions
 ju+0.153
 golf+0.149
mons+0.147
 makeup+0.144
 contest+0.140
Neg logit contributions
ously-0.250
rum-0.240
 despair-0.233
 deeply-0.227
 horror-0.219
Token,
Token ID11
Feature activation-6.77e-02

Pos logit contributions
 ju+0.008
mons+0.008
atmeal+0.007
 golf+0.007
Draw+0.007
Neg logit contributions
rum-0.003
ously-0.003
 sadly-0.003
 deeply-0.003
 despair-0.003
Token Lily
Token ID20037
Feature activation-3.74e-02

Pos logit contributions
mons+0.117
 golf+0.110
 ju+0.110
atmeal+0.110
 paints+0.104
Neg logit contributions
ously-0.096
 deeply-0.095
 despair-0.095
rum-0.094
 horror-0.093
Token.
Token ID13
Feature activation-1.64e-02

Pos logit contributions
 ju+0.075
 golf+0.072
mons+0.071
 makeup+0.070
 contest+0.070
Neg logit contributions
ously-0.050
rum-0.049
 deeply-0.044
 despair-0.043
 horror-0.041
Token What
Token ID1867
Feature activation-9.43e-02

Pos logit contributions
 ju+0.023
 golf+0.022
mons+0.021
 makeup+0.021
 butter+0.021
Neg logit contributions
ously-0.031
rum-0.031
 despair-0.029
 deeply-0.028
 horror-0.028
Token a
Token ID257
Feature activation-3.46e-02

Pos logit contributions
 ju+0.172
mons+0.167
 golf+0.162
atmeal+0.154
coat+0.150
Neg logit contributions
rum-0.128
ously-0.126
 despair-0.117
aced-0.113
 deeply-0.113
Token nice
Token ID3621
Feature activation-2.19e-02

Pos logit contributions
mons+0.060
 ju+0.053
atmeal+0.052
 paints+0.050
 pens+0.048
Neg logit contributions
 horror-0.051
 despair-0.051
 deeply-0.051
rum-0.048
 heartbeat-0.048
Token surprise
Token ID5975
Feature activation-2.34e-02

Pos logit contributions
 ju+0.032
mons+0.030
 golf+0.030
 makeup+0.029
atmeal+0.029
Neg logit contributions
ously-0.040
rum-0.039
 despair-0.037
 deeply-0.037
 horror-0.036
Token.
Token ID13
Feature activation-4.43e-03

Pos logit contributions
 ju+0.039
 golf+0.038
mons+0.036
 makeup+0.036
 grapes+0.035
Neg logit contributions
ously-0.038
rum-0.038
 deeply-0.035
 despair-0.035
 horror-0.033
Token Are
Token ID4231
Feature activation-6.02e-02

Pos logit contributions
 ju+0.006
 golf+0.006
 makeup+0.005
mons+0.005
 contest+0.005
Neg logit contributions
ously-0.009
rum-0.009
 despair-0.008
 deeply-0.008
 horror-0.008
Token looked
Token ID3114
Feature activation+1.08e-02

Pos logit contributions
rum+0.027
ously+0.025
 despair+0.025
 horror+0.025
 heartbeat+0.024
Neg logit contributions
 ju-0.017
 golf-0.017
 competition-0.015
 contest-0.015
 swing-0.014
Token very
Token ID845
Feature activation+4.81e-02

Pos logit contributions
rum+0.025
ously+0.024
 despair+0.022
 horror+0.021
 heartbeat+0.021
Neg logit contributions
 ju-0.013
 golf-0.012
 makeup-0.011
 competition-0.010
 contest-0.010
Token excited
Token ID6568
Feature activation+1.85e-03

Pos logit contributions
ously+0.109
rum+0.105
 despair+0.094
aced+0.093
urbed+0.092
Neg logit contributions
 ju-0.057
 golf-0.057
 makeup-0.048
 contest-0.046
 competition-0.046
Token.
Token ID13
Feature activation+3.81e-02

Pos logit contributions
ously+0.003
rum+0.003
 despair+0.003
 deeply+0.003
 horror+0.003
Neg logit contributions
 ju-0.003
 golf-0.003
 makeup-0.003
mons-0.003
 contest-0.003
Token "
Token ID366
Feature activation-1.13e-01

Pos logit contributions
rum+0.070
ously+0.069
 despair+0.065
 horror+0.062
 deeply+0.061
Neg logit contributions
 ju-0.059
 golf-0.058
 makeup-0.053
mons-0.052
 gum-0.051
TokenCan
Token ID6090
Feature activation-1.02e-01

Pos logit contributions
 makeup+0.139
 golf+0.135
mons+0.135
 ju+0.134
 grapes+0.134
Neg logit contributions
ously-0.229
rum-0.219
 despair-0.214
 deeply-0.202
 silently-0.200
Token I
Token ID314
Feature activation-6.50e-02

Pos logit contributions
 ju+0.133
mons+0.128
 makeup+0.128
atmeal+0.124
 grapes+0.123
Neg logit contributions
ously-0.187
rum-0.182
 despair-0.182
 deeply-0.178
 sadly-0.175
Token come
Token ID1282
Feature activation-2.68e-02

Pos logit contributions
 ju+0.092
 golf+0.087
 makeup+0.086
mons+0.086
 grapes+0.083
Neg logit contributions
ously-0.122
rum-0.120
 despair-0.114
 deeply-0.113
 horror-0.109
Token with
Token ID351
Feature activation+2.14e-02

Pos logit contributions
 ju+0.044
mons+0.041
 makeup+0.040
 grapes+0.040
 golf+0.039
Neg logit contributions
rum-0.044
ously-0.044
 deeply-0.041
 despair-0.041
 silently-0.040
Token you
Token ID345
Feature activation-1.88e-02

Pos logit contributions
ously+0.044
rum+0.044
 despair+0.042
 deeply+0.042
urbed+0.040
Neg logit contributions
 ju-0.026
mons-0.024
 golf-0.023
atmeal-0.023
 makeup-0.023
Token?"
Token ID1701
Feature activation+3.14e-02

Pos logit contributions
 ju+0.021
mons+0.021
 makeup+0.020
 grapes+0.019
 golf+0.019
Neg logit contributions
ously-0.039
rum-0.038
 deeply-0.038
 despair-0.037
 silently-0.037