TOP ACTIVATIONS
MAX = 0.133

a swing and pumped his legs back and forth . He
this egg ?" he said , holding it up .
your trip !" Dad said , picking her up and spinning
I made !" James said , folding his arms in front
so he stomp ed his feet and shouted . Everyone in
, Mama !" Drake said , amazed . M
swinging . He pumped his legs higher and higher as he
Mom ," she said quietly , her eyes full of tears
than yours !" Ben said , z ooming his red boat
star fish !" Sam said , pointing to a red and
to the bird and whispered , " I will miss you
it is !" Lily whispered , pointing to the doll .
this one !" she said , picking up the bunny .
to sleep , she whispered , " I love you ,
is gone !" he said , between so bs .
L ila ?" she asked , smiling . L
do it !" Sarah said , laughing .
your princess !" Tom said , making his wire w iggle
you go ," she said , pat ting the little girl
something together !" said Emma , adm iring their work .

MIN ACTIVATIONS
MIN = -0.141

stickers . " King Tom , I have a
! You are the best tutor ever !" Tom smiled and
, " Yes , my Queen . The shield will protect
fly and thought , " Hmm , maybe I can bite
sick . " Here , Tim , let me
Let 's go and see what is inside ," Bee said
mom . " Here , you can have my
on her chest . " Hmm , it sounds good .
. He thought , " Hmm , maybe this hole will
behind the couch . " Here I am !" she shouted
Lily 's dress . That boy was very bad and he
" Let 's see what is in the sack ,"
her mom said " That glove has been sitting here for
him feel better . " Here , take this medicine .
. " Let 's see what 's inside !" said Bun
Alice . Every day , Princess Alice liked to look nice
looked and said , " Hmm , it seems like a
Benny and asked , " What 's wrong , Benny ?"
finds the car . " Here it is ! It is
" Hello , children . Who are you ? What do

INTERVAL 0.064 - 0.133
CONTAINS 1.432%

." J enny waited for the rain to stop
. They make a big train . They put a kettle
so soft that she could barely hear it , but it
them . The lion watched , but he didn 't do
. Billy jumped from his bed and told his friends to

INTERVAL -0.004 - 0.064
CONTAINS 54.802%

When he was finished , the trophy gle amed and spark
and wanted to fly away . She asked her bird friends
gym . He tried out the slides , and he swung
herself . The sun was shining and Mia was happy .
very important mission to do . The cat ran

INTERVAL -0.072 - -0.004
CONTAINS 42.836%

. She was always poking her big nose around , looking
<|endoftext|> Once upon a time , there was a rabbit .
upon a time , there was a little girl named Lily
to clean his room before he could go outside to play
smiled and said , " You are welcome . I am

INTERVAL -0.141 - -0.072
CONTAINS 0.931%

. <|endoftext|> Once upon a time , there was a little
, " That was very naughty , Lily . You wasted
" You 're the best friend ever !" Tim my hugged
pot and tell me if it starts to bubble ."
Oh , children , what have you done ?" Mrs .
Token a
Token ID257
Feature activation+5.59e-02

Pos logit contributions
eyed+0.137
Unfortunately+0.135
Finally+0.132
Meanwhile+0.132
Days+0.130
Neg logit contributions
 is-0.064
 happens-0.051
 doesn-0.050
 isn-0.049
 guys-0.049
Token swing
Token ID9628
Feature activation+5.09e-02

Pos logit contributions
eyed+0.118
Unfortunately+0.116
Meanwhile+0.115
Just+0.108
har+0.106
Neg logit contributions
 is-0.097
 belongs-0.087
 can-0.087
 isn-0.086
 happens-0.084
Token and
Token ID290
Feature activation+1.30e-02

Pos logit contributions
Unfortunately+0.135
eyed+0.132
Meanwhile+0.129
har+0.129
Finally+0.129
Neg logit contributions
 is-0.065
 isn-0.054
 doesn-0.054
 can-0.050
 belongs-0.049
Token pumped
Token ID29104
Feature activation+6.20e-02

Pos logit contributions
Unfortunately+0.024
eyed+0.023
Meanwhile+0.023
 widening+0.022
Just+0.022
Neg logit contributions
 happens-0.025
 is-0.024
 doesn-0.024
 isn-0.023
 belongs-0.023
Token his
Token ID465
Feature activation+1.58e-02

Pos logit contributions
Unfortunately+0.158
Meanwhile+0.155
Finally+0.151
eyed+0.150
Days+0.150
Neg logit contributions
 is-0.087
 happens-0.074
 isn-0.073
 doesn-0.071
 guys-0.066
Token legs
Token ID7405
Feature activation+1.33e-01

Pos logit contributions
Unfortunately+0.032
Meanwhile+0.031
eyed+0.030
 widening+0.029
However+0.028
Neg logit contributions
 is-0.028
 happens-0.028
 doesn-0.027
 guys-0.027
 belongs-0.027
Token back
Token ID736
Feature activation+5.21e-02

Pos logit contributions
 sidelines+0.358
har+0.354
Unfortunately+0.346
Days+0.336
Sadly+0.334
Neg logit contributions
,-0.142
 is-0.131
 can-0.124
 doesn-0.087
 isn-0.083
Token and
Token ID290
Feature activation+3.57e-02

Pos logit contributions
Unfortunately+0.128
Meanwhile+0.125
 sidelines+0.121
Days+0.119
Finally+0.118
Neg logit contributions
 is-0.076
 doesn-0.064
 belongs-0.064
 isn-0.063
 happens-0.063
Token forth
Token ID6071
Feature activation+7.18e-02

Pos logit contributions
Unfortunately+0.071
eyed+0.068
Meanwhile+0.066
Days+0.066
Finally+0.065
Neg logit contributions
 happens-0.066
 is-0.065
 doesn-0.063
 isn-0.061
 belongs-0.059
Token.
Token ID13
Feature activation-9.94e-03

Pos logit contributions
Unfortunately+0.215
Meanwhile+0.210
 sidelines+0.201
har+0.199
eyed+0.198
Neg logit contributions
 is-0.068
 can-0.056
 isn-0.055
 guys-0.053
 doesn-0.052
Token He
Token ID679
Feature activation+8.98e-03

Pos logit contributions
 happens+0.023
 belongs+0.022
 guys+0.022
 is+0.022
 doesn+0.022
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.013
Just-0.013
 widening-0.012
Token this
Token ID428
Feature activation-2.27e-02

Pos logit contributions
 happens+0.058
 guys+0.052
 belongs+0.050
 doesn+0.050
 isn+0.048
Neg logit contributions
Unfortunately-0.105
Meanwhile-0.104
eyed-0.103
On-0.098
 pounding-0.097
Token egg
Token ID5935
Feature activation-4.32e-02

Pos logit contributions
 happens+0.032
 guys+0.031
 doesn+0.030
 belongs+0.029
 isn+0.029
Neg logit contributions
Unfortunately-0.050
Meanwhile-0.049
eyed-0.047
Just-0.047
 widening-0.046
Token?"
Token ID1701
Feature activation+1.99e-02

Pos logit contributions
 happens+0.031
 guys+0.029
 isn+0.025
 doesn+0.024
 belongs+0.024
Neg logit contributions
Meanwhile-0.121
Unfortunately-0.121
eyed-0.120
 pounding-0.117
Just-0.116
Token he
Token ID339
Feature activation+5.01e-02

Pos logit contributions
Unfortunately+0.038
Meanwhile+0.038
eyed+0.037
 widening+0.037
 pounding+0.036
Neg logit contributions
 happens-0.035
 is-0.034
 doesn-0.033
 isn-0.032
 belongs-0.032
Token said
Token ID531
Feature activation+6.11e-02

Pos logit contributions
 sidelines+0.115
Unfortunately+0.110
har+0.107
 crashing+0.107
eyed+0.106
Neg logit contributions
 happens-0.075
 is-0.072
 means-0.069
 doesn-0.068
 guys-0.066
Token,
Token ID11
Feature activation+1.32e-01

Pos logit contributions
Unfortunately+0.180
 sidelines+0.178
Meanwhile+0.176
Finally+0.176
Eventually+0.174
Neg logit contributions
 is-0.044
 happens-0.043
 doesn-0.037
 means-0.036
 brother-0.035
Token holding
Token ID4769
Feature activation-3.09e-03

Pos logit contributions
Unfortunately+0.282
Eventually+0.280
Finally+0.279
Days+0.278
Years+0.272
Neg logit contributions
,-0.181
 brother-0.162
?"-0.159
 I-0.157
 is-0.157
Token it
Token ID340
Feature activation+6.26e-02

Pos logit contributions
 happens+0.006
 is+0.006
 doesn+0.006
 belongs+0.006
 isn+0.006
Neg logit contributions
Unfortunately-0.006
Meanwhile-0.005
eyed-0.005
Just-0.005
 widening-0.005
Token up
Token ID510
Feature activation+7.46e-02

Pos logit contributions
 sidelines+0.140
Unfortunately+0.139
Meanwhile+0.133
Finally+0.131
Eventually+0.129
Neg logit contributions
 is-0.098
 doesn-0.092
 belongs-0.089
 needs-0.088
 isn-0.087
Token.
Token ID13
Feature activation+4.26e-02

Pos logit contributions
 sidelines+0.215
Unfortunately+0.208
Eventually+0.199
Days+0.199
Meanwhile+0.198
Neg logit contributions
 is-0.065
,-0.058
 happened-0.052
 doesn-0.052
?-0.050
Token
Token ID220
Feature activation+1.61e-02

Pos logit contributions
eyed+0.085
Unfortunately+0.084
har+0.083
 pounding+0.083
Meanwhile+0.083
Neg logit contributions
 is-0.068
 happens-0.067
 isn-0.064
 doesn-0.063
 means-0.059
Token your
Token ID534
Feature activation-2.34e-02

Pos logit contributions
 happens+0.008
 is+0.008
 belongs+0.008
 doesn+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.016
Meanwhile-0.016
eyed-0.015
Just-0.015
 pounding-0.015
Token trip
Token ID5296
Feature activation-1.11e-02

Pos logit contributions
 happens+0.042
 belongs+0.040
 doesn+0.039
 is+0.039
 isn+0.038
Neg logit contributions
Unfortunately-0.042
eyed-0.041
Meanwhile-0.040
 pounding-0.040
Just-0.039
Token!"
Token ID2474
Feature activation+2.52e-02

Pos logit contributions
 happens+0.016
 is+0.015
 belongs+0.015
 doesn+0.015
 isn+0.015
Neg logit contributions
Unfortunately-0.025
Meanwhile-0.024
eyed-0.024
Just-0.023
 pounding-0.023
Token Dad
Token ID17415
Feature activation+4.96e-02

Pos logit contributions
Unfortunately+0.055
Meanwhile+0.054
eyed+0.054
 widening+0.053
 pounding+0.053
Neg logit contributions
 is-0.037
 happens-0.035
 doesn-0.035
 isn-0.034
 belongs-0.033
Token said
Token ID531
Feature activation+5.97e-02

Pos logit contributions
Unfortunately+0.125
 sidelines+0.124
Days+0.119
Meanwhile+0.117
Hours+0.117
Neg logit contributions
 is-0.060
 doesn-0.053
 isn-0.053
 happens-0.052
 means-0.051
Token,
Token ID11
Feature activation+1.31e-01

Pos logit contributions
Unfortunately+0.175
Meanwhile+0.174
Finally+0.171
 sidelines+0.170
Eventually+0.168
Neg logit contributions
 is-0.042
 happens-0.038
 doesn-0.035
 isn-0.035
 guys-0.033
Token picking
Token ID10868
Feature activation+1.23e-02

Pos logit contributions
Days+0.254
Unfortunately+0.248
Finally+0.246
Eventually+0.245
Years+0.241
Neg logit contributions
,-0.187
 I-0.174
?"-0.164
 is-0.164
 brother-0.164
Token her
Token ID607
Feature activation-1.18e-03

Pos logit contributions
Unfortunately+0.023
Meanwhile+0.023
eyed+0.022
Just+0.022
 pounding+0.021
Neg logit contributions
 is-0.022
 happens-0.022
 doesn-0.022
 isn-0.021
 belongs-0.021
Token up
Token ID510
Feature activation+2.42e-02

Pos logit contributions
 happens+0.002
 is+0.002
 doesn+0.002
 belongs+0.002
 isn+0.002
Neg logit contributions
Unfortunately-0.002
Meanwhile-0.002
eyed-0.002
Just-0.002
 widening-0.002
Token and
Token ID290
Feature activation+6.01e-02

Pos logit contributions
Unfortunately+0.052
Meanwhile+0.051
eyed+0.050
Just+0.049
Finally+0.048
Neg logit contributions
 is-0.038
 doesn-0.037
 happens-0.036
 isn-0.036
 belongs-0.035
Token spinning
Token ID19493
Feature activation+2.34e-02

Pos logit contributions
Unfortunately+0.110
Days+0.108
Finally+0.105
Just+0.104
eyed+0.102
Neg logit contributions
 is-0.114
 isn-0.107
 doesn-0.107
 happens-0.103
 needs-0.102
Token I
Token ID314
Feature activation+1.98e-03

Pos logit contributions
unn+0.105
uffy+0.102
ishes+0.098
 belongs+0.098
ens+0.097
Neg logit contributions
His-0.152
Unfortunately-0.152
 impressed-0.149
 determined-0.148
 victorious-0.146
Token made
Token ID925
Feature activation-1.25e-02

Pos logit contributions
Unfortunately+0.004
Meanwhile+0.004
eyed+0.004
 widening+0.004
Just+0.004
Neg logit contributions
 happens-0.003
 is-0.003
 doesn-0.003
 belongs-0.003
 isn-0.003
Token!"
Token ID2474
Feature activation+2.33e-02

Pos logit contributions
 happens+0.016
 belongs+0.016
 doesn+0.016
 is+0.015
 guys+0.015
Neg logit contributions
Unfortunately-0.029
eyed-0.029
Meanwhile-0.029
Just-0.027
 widening-0.027
Token James
Token ID3700
Feature activation+5.44e-02

Pos logit contributions
Unfortunately+0.050
Meanwhile+0.050
eyed+0.049
 pounding+0.048
 widening+0.048
Neg logit contributions
 is-0.035
 happens-0.035
 doesn-0.034
 isn-0.033
 belongs-0.032
Token said
Token ID531
Feature activation+7.88e-02

Pos logit contributions
Unfortunately+0.121
 sidelines+0.121
Days+0.118
Years+0.114
 widening+0.114
Neg logit contributions
 is-0.081
 happens-0.074
 doesn-0.072
 isn-0.072
 needs-0.067
Token,
Token ID11
Feature activation+1.31e-01

Pos logit contributions
 sidelines+0.219
Unfortunately+0.219
Meanwhile+0.218
Finally+0.215
Days+0.213
Neg logit contributions
 is-0.066
,-0.062
 happens-0.061
 brother-0.059
 sister-0.055
Token folding
Token ID29909
Feature activation+3.51e-02

Pos logit contributions
Days+0.282
Unfortunately+0.270
Finally+0.266
Years+0.265
Hours+0.264
Neg logit contributions
,-0.174
 I-0.170
 is-0.154
?"-0.149
 we-0.143
Token his
Token ID465
Feature activation+2.93e-03

Pos logit contributions
Unfortunately+0.078
Meanwhile+0.077
Just+0.074
eyed+0.074
rolling+0.073
Neg logit contributions
 is-0.058
 isn-0.054
 happens-0.054
 doesn-0.054
 belongs-0.051
Token arms
Token ID5101
Feature activation+9.71e-02

Pos logit contributions
Unfortunately+0.006
Meanwhile+0.006
eyed+0.005
Just+0.005
 widening+0.005
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 belongs-0.005
 isn-0.005
Token in
Token ID287
Feature activation+5.58e-02

Pos logit contributions
Unfortunately+0.253
Days+0.242
Meanwhile+0.240
har+0.239
 sidelines+0.239
Neg logit contributions
 is-0.116
 doesn-0.102
 isn-0.101
 can-0.100
 happens-0.090
Token front
Token ID2166
Feature activation+5.83e-02

Pos logit contributions
rolling+0.120
ooting+0.119
Just+0.119
Meanwhile+0.119
Unfortunately+0.118
Neg logit contributions
 is-0.093
 isn-0.086
 can-0.079
 doesn-0.078
 needs-0.076
Token so
Token ID523
Feature activation+1.26e-02

Pos logit contributions
eyed+0.169
rolling+0.164
Finally+0.162
Days+0.160
har+0.159
Neg logit contributions
 doesn-0.108
 is-0.106
 isn-0.104
 can-0.090
 needs-0.090
Token he
Token ID339
Feature activation+1.53e-02

Pos logit contributions
Unfortunately+0.027
eyed+0.026
Meanwhile+0.026
 widening+0.026
Just+0.025
Neg logit contributions
 happens-0.020
 is-0.019
 doesn-0.019
 isn-0.018
 belongs-0.018
Token stomp
Token ID32221
Feature activation+1.03e-02

Pos logit contributions
Unfortunately+0.027
eyed+0.027
Meanwhile+0.027
 widening+0.026
 sidelines+0.025
Neg logit contributions
 happens-0.029
 is-0.029
 doesn-0.029
 guys-0.028
 isn-0.028
Tokened
Token ID276
Feature activation+4.28e-02

Pos logit contributions
Unfortunately+0.026
Meanwhile+0.026
 widening+0.025
eyed+0.025
 sidelines+0.025
Neg logit contributions
 happens-0.012
 is-0.012
 doesn-0.011
 guys-0.011
 isn-0.011
Token his
Token ID465
Feature activation+2.51e-02

Pos logit contributions
Meanwhile+0.089
Unfortunately+0.088
Finally+0.086
 widening+0.084
rolling+0.083
Neg logit contributions
 is-0.075
 isn-0.071
 doesn-0.071
 happens-0.069
 guys-0.068
Token feet
Token ID3625
Feature activation+1.30e-01

Pos logit contributions
Unfortunately+0.056
Meanwhile+0.055
eyed+0.053
 widening+0.052
His+0.051
Neg logit contributions
 is-0.038
 belongs-0.036
 doesn-0.036
 needs-0.036
 isn-0.036
Token and
Token ID290
Feature activation+3.85e-02

Pos logit contributions
Finally+0.358
Unfortunately+0.352
Eventually+0.345
Meanwhile+0.344
har+0.338
Neg logit contributions
 is-0.106
,-0.100
 doesn-0.083
 can-0.083
 isn-0.066
Token shouted
Token ID17293
Feature activation+2.06e-02

Pos logit contributions
Unfortunately+0.078
eyed+0.078
Finally+0.076
Meanwhile+0.076
rolling+0.074
Neg logit contributions
 doesn-0.064
 happens-0.064
 is-0.063
 guys-0.061
 isn-0.060
Token.
Token ID13
Feature activation+1.99e-02

Pos logit contributions
Unfortunately+0.053
Meanwhile+0.053
eyed+0.051
 widening+0.050
Finally+0.049
Neg logit contributions
 is-0.024
 happens-0.024
 isn-0.023
 doesn-0.023
 guys-0.022
Token Everyone
Token ID11075
Feature activation+1.66e-02

Pos logit contributions
Unfortunately+0.038
eyed+0.038
Meanwhile+0.038
 pounding+0.037
 widening+0.036
Neg logit contributions
 is-0.034
 happens-0.033
 isn-0.033
 doesn-0.033
 belongs-0.032
Token in
Token ID287
Feature activation-7.68e-04

Pos logit contributions
Unfortunately+0.038
Meanwhile+0.037
 widening+0.037
eyed+0.036
Finally+0.036
Neg logit contributions
 happens-0.024
 is-0.024
 doesn-0.023
 isn-0.022
 belongs-0.022
Token,
Token ID11
Feature activation-2.17e-02

Pos logit contributions
 happens+0.053
 guys+0.050
 belongs+0.049
 doesn+0.047
 needs+0.045
Neg logit contributions
Unfortunately-0.101
eyed-0.099
Meanwhile-0.098
Just-0.095
 pounding-0.094
Token Mama
Token ID40084
Feature activation-4.70e-02

Pos logit contributions
 happens+0.035
 belongs+0.034
 is+0.032
 guys+0.032
 needs+0.031
Neg logit contributions
Unfortunately-0.045
Meanwhile-0.044
eyed-0.042
 pounding-0.042
Just-0.042
Token!"
Token ID2474
Feature activation+3.80e-02

Pos logit contributions
 happens+0.063
 belongs+0.058
 guys+0.056
 doesn+0.054
 isn+0.053
Neg logit contributions
eyed-0.108
Meanwhile-0.106
Unfortunately-0.104
 pounding-0.104
Just-0.100
Token Drake
Token ID18515
Feature activation+2.02e-02

Pos logit contributions
eyed+0.086
 widening+0.086
Unfortunately+0.086
Days+0.085
 sidelines+0.085
Neg logit contributions
 happens-0.050
 is-0.050
 doesn-0.048
 isn-0.047
 means-0.045
Token said
Token ID531
Feature activation+5.86e-02

Pos logit contributions
Unfortunately+0.043
Meanwhile+0.041
eyed+0.040
 sidelines+0.040
 widening+0.040
Neg logit contributions
 happens-0.032
 is-0.032
 doesn-0.031
 belongs-0.030
 isn-0.030
Token,
Token ID11
Feature activation+1.29e-01

Pos logit contributions
Unfortunately+0.149
 sidelines+0.146
Meanwhile+0.144
Finally+0.142
Days+0.141
Neg logit contributions
 happens-0.069
 is-0.068
 happened-0.063
 brother-0.059
 doesn-0.059
Token amazed
Token ID24562
Feature activation+1.40e-02

Pos logit contributions
Days+0.265
Unfortunately+0.264
Years+0.260
Finally+0.250
Eventually+0.249
Neg logit contributions
,-0.177
 I-0.176
 is-0.172
?"-0.170
 brother-0.159
Token.
Token ID13
Feature activation+5.78e-02

Pos logit contributions
Unfortunately+0.034
Meanwhile+0.033
eyed+0.032
Finally+0.031
Just+0.031
Neg logit contributions
 happens-0.018
 is-0.018
 doesn-0.018
 isn-0.017
 guys-0.017
Token
Token ID198
Feature activation+5.64e-02

Pos logit contributions
 sidelines+0.129
 widening+0.128
har+0.128
Days+0.124
 pounding+0.123
Neg logit contributions
 is-0.085
 isn-0.082
 doesn-0.079
 happens-0.079
 can-0.076
Token
Token ID198
Feature activation+6.80e-02

Pos logit contributions
 widening+0.141
 sidelines+0.140
 pounding+0.130
side+0.128
 winding+0.126
Neg logit contributions
 is-0.077
 happens-0.066
?"-0.065
 can-0.064
 isn-0.063
TokenM
Token ID44
Feature activation+1.62e-02

Pos logit contributions
 widening+0.143
 sidelines+0.140
 contain+0.127
 ledge+0.125
side+0.123
Neg logit contributions
 is-0.112
 can-0.107
 happens-0.102
 isn-0.097
 doesn-0.096
Token swinging
Token ID25635
Feature activation+6.76e-02

Pos logit contributions
Unfortunately+0.205
Finally+0.202
Days+0.195
Meanwhile+0.193
Eventually+0.193
Neg logit contributions
 is-0.092
 happens-0.075
 doesn-0.071
 isn-0.071
 things-0.071
Token.
Token ID13
Feature activation+3.07e-02

Pos logit contributions
Unfortunately+0.161
Meanwhile+0.159
eyed+0.154
Finally+0.153
rolling+0.153
Neg logit contributions
 is-0.104
 isn-0.087
 can-0.086
 doesn-0.086
 happens-0.085
Token He
Token ID679
Feature activation+2.71e-02

Pos logit contributions
 pounding+0.061
Meanwhile+0.060
eyed+0.060
Unfortunately+0.060
 widening+0.059
Neg logit contributions
 isn-0.050
 is-0.050
 doesn-0.048
 happens-0.046
 belongs-0.045
Token pumped
Token ID29104
Feature activation+6.70e-02

Pos logit contributions
Unfortunately+0.056
Meanwhile+0.054
eyed+0.054
 widening+0.052
 sidelines+0.052
Neg logit contributions
 is-0.046
 happens-0.045
 doesn-0.044
 isn-0.044
 belongs-0.042
Token his
Token ID465
Feature activation+2.59e-02

Pos logit contributions
Unfortunately+0.158
eyed+0.154
har+0.151
Meanwhile+0.150
Finally+0.149
Neg logit contributions
 is-0.102
 isn-0.088
 doesn-0.086
 happens-0.084
 guys-0.084
Token legs
Token ID7405
Feature activation+1.29e-01

Pos logit contributions
Unfortunately+0.054
Meanwhile+0.053
eyed+0.050
 widening+0.049
However+0.048
Neg logit contributions
 is-0.044
 happens-0.043
 doesn-0.043
 isn-0.043
 guys-0.042
Token higher
Token ID2440
Feature activation+5.35e-02

Pos logit contributions
har+0.353
 sidelines+0.350
Unfortunately+0.346
Days+0.336
Sadly+0.331
Neg logit contributions
 is-0.123
,-0.122
 can-0.106
 are-0.074
 doesn-0.072
Token and
Token ID290
Feature activation+4.28e-02

Pos logit contributions
Unfortunately+0.138
 sidelines+0.131
Meanwhile+0.131
eyed+0.131
rolling+0.129
Neg logit contributions
 is-0.069
 doesn-0.057
 isn-0.054
 happens-0.053
 guys-0.050
Token higher
Token ID2440
Feature activation+3.59e-02

Pos logit contributions
eyed+0.052
Unfortunately+0.052
Meanwhile+0.048
Days+0.047
Their+0.045
Neg logit contributions
 is-0.109
 happens-0.108
 doesn-0.107
 isn-0.105
 can-0.101
Token as
Token ID355
Feature activation+7.91e-03

Pos logit contributions
Unfortunately+0.085
Meanwhile+0.080
eyed+0.079
rolling+0.077
Finally+0.076
Neg logit contributions
 is-0.054
 happens-0.049
 doesn-0.049
 isn-0.048
 guys-0.046
Token he
Token ID339
Feature activation+1.39e-02

Pos logit contributions
Unfortunately+0.016
Meanwhile+0.016
eyed+0.016
Just+0.015
 widening+0.015
Neg logit contributions
 happens-0.013
 is-0.013
 doesn-0.012
 isn-0.012
 belongs-0.012
Token Mom
Token ID11254
Feature activation-3.82e-02

Pos logit contributions
 happens+0.037
 belongs+0.036
 guys+0.036
 doesn+0.034
 is+0.033
Neg logit contributions
Unfortunately-0.047
Meanwhile-0.045
Just-0.043
 pounding-0.043
However-0.042
Token,"
Token ID553
Feature activation-1.93e-03

Pos logit contributions
 happens+0.052
 guys+0.049
 belongs+0.048
 isn+0.046
 doesn+0.045
Neg logit contributions
eyed-0.088
Meanwhile-0.086
Unfortunately-0.085
 pounding-0.084
Just-0.082
Token she
Token ID673
Feature activation+3.73e-02

Pos logit contributions
 happens+0.003
 is+0.003
 doesn+0.003
 belongs+0.003
 isn+0.003
Neg logit contributions
Unfortunately-0.004
Meanwhile-0.004
eyed-0.004
Just-0.004
 widening-0.004
Token said
Token ID531
Feature activation+5.68e-02

Pos logit contributions
 sidelines+0.099
Unfortunately+0.095
eyed+0.095
side+0.093
Finally+0.092
Neg logit contributions
 happens-0.043
 is-0.041
 doesn-0.040
 isn-0.037
 belongs-0.036
Token quietly
Token ID12703
Feature activation+8.02e-02

Pos logit contributions
Unfortunately+0.171
Meanwhile+0.170
Finally+0.168
 sidelines+0.167
Eventually+0.165
Neg logit contributions
 is-0.037
 happens-0.033
 doesn-0.032
 brother-0.029
 happened-0.028
Token,
Token ID11
Feature activation+1.26e-01

Pos logit contributions
Unfortunately+0.268
Finally+0.261
Eventually+0.257
Days+0.253
Meanwhile+0.252
Neg logit contributions
,-0.032
 is-0.029
 brother-0.028
 happens-0.026
 doesn-0.024
Token her
Token ID607
Feature activation+1.35e-02

Pos logit contributions
Finally+0.280
Unfortunately+0.279
Days+0.279
Years+0.276
Hours+0.272
Neg logit contributions
 doesn-0.148
 is-0.147
,-0.146
 I-0.140
 isn-0.136
Token eyes
Token ID2951
Feature activation+1.13e-01

Pos logit contributions
Unfortunately+0.034
Meanwhile+0.034
eyed+0.033
Just+0.032
However+0.031
Neg logit contributions
 is-0.017
 happens-0.016
 doesn-0.016
 isn-0.016
 needs-0.015
Token full
Token ID1336
Feature activation+4.00e-02

Pos logit contributions
Unfortunately+0.268
 sidelines+0.254
Meanwhile+0.253
Finally+0.246
However+0.244
Neg logit contributions
 is-0.154
 happened-0.149
 can-0.145
 happens-0.142
 doesn-0.138
Token of
Token ID286
Feature activation+1.13e-03

Pos logit contributions
Unfortunately+0.102
Meanwhile+0.101
rolling+0.098
eyed+0.097
shadow+0.097
Neg logit contributions
 is-0.049
 happens-0.046
 isn-0.046
 doesn-0.045
 needs-0.042
Token tears
Token ID10953
Feature activation+3.91e-02

Pos logit contributions
Unfortunately+0.002
Meanwhile+0.002
eyed+0.002
Just+0.002
 widening+0.002
Neg logit contributions
 happens-0.002
 is-0.002
 doesn-0.002
 belongs-0.002
 isn-0.002
Token than
Token ID621
Feature activation-2.55e-02

Pos logit contributions
 happens+0.045
 belongs+0.042
 doesn+0.040
 isn+0.040
 needs+0.040
Neg logit contributions
Unfortunately-0.081
eyed-0.080
Meanwhile-0.079
Just-0.078
 pounding-0.077
Token yours
Token ID12431
Feature activation-4.32e-02

Pos logit contributions
 happens+0.032
 belongs+0.030
 doesn+0.030
 is+0.030
 isn+0.029
Neg logit contributions
Unfortunately-0.060
Meanwhile-0.059
eyed-0.058
 sidelines-0.057
Just-0.057
Token!"
Token ID2474
Feature activation-1.64e-02

Pos logit contributions
 happens+0.052
 doesn+0.049
 belongs+0.047
 guys+0.047
 isn+0.047
Neg logit contributions
eyed-0.100
Unfortunately-0.098
Meanwhile-0.096
 pounding-0.095
Just-0.095
Token Ben
Token ID3932
Feature activation+2.68e-02

Pos logit contributions
 belongs+0.024
 happens+0.024
 doesn+0.023
 isn+0.023
 guys+0.023
Neg logit contributions
Unfortunately-0.036
Meanwhile-0.036
eyed-0.035
Just-0.034
 widening-0.034
Token said
Token ID531
Feature activation+4.37e-02

Pos logit contributions
Unfortunately+0.061
Meanwhile+0.059
 sidelines+0.059
eyed+0.057
Days+0.057
Neg logit contributions
 is-0.038
 doesn-0.037
 happens-0.037
 isn-0.037
 belongs-0.035
Token,
Token ID11
Feature activation+1.26e-01

Pos logit contributions
Unfortunately+0.123
Meanwhile+0.122
Finally+0.118
Eventually+0.116
 sidelines+0.116
Neg logit contributions
 happens-0.039
 is-0.038
 guys-0.036
 doesn-0.035
 isn-0.034
Token z
Token ID1976
Feature activation+1.06e-02

Pos logit contributions
Unfortunately+0.272
Finally+0.267
Eventually+0.264
Days+0.259
Years+0.253
Neg logit contributions
,-0.163
 I-0.160
 is-0.154
 brother-0.147
?"-0.144
Tokenooming
Token ID30602
Feature activation+4.73e-02

Pos logit contributions
Unfortunately+0.019
Meanwhile+0.019
eyed+0.018
 widening+0.017
 pounding+0.017
Neg logit contributions
 happens-0.020
 is-0.019
 doesn-0.019
 guys-0.019
 belongs-0.019
Token his
Token ID465
Feature activation+2.54e-02

Pos logit contributions
Unfortunately+0.102
Finally+0.098
Meanwhile+0.097
rolling+0.094
Eventually+0.094
Neg logit contributions
 is-0.078
 guys-0.076
 happens-0.074
 isn-0.072
 doesn-0.071
Token red
Token ID2266
Feature activation+1.45e-02

Pos logit contributions
Unfortunately+0.060
Meanwhile+0.058
eyed+0.056
Finally+0.054
 widening+0.054
Neg logit contributions
 is-0.037
 guys-0.037
 doesn-0.035
 happens-0.035
 isn-0.035
Token boat
Token ID8848
Feature activation+9.66e-02

Pos logit contributions
Unfortunately+0.030
Meanwhile+0.029
eyed+0.029
 widening+0.028
Just+0.028
Neg logit contributions
 is-0.024
 happens-0.024
 guys-0.024
 belongs-0.023
 isn-0.023
Token star
Token ID3491
Feature activation-6.08e-02

Pos logit contributions
 happens+0.034
 belongs+0.033
 doesn+0.033
 is+0.032
 isn+0.031
Neg logit contributions
Unfortunately-0.035
Meanwhile-0.034
eyed-0.034
 pounding-0.033
 widening-0.033
Tokenfish
Token ID11084
Feature activation-5.49e-02

Pos logit contributions
 guys+0.080
 happens+0.079
 doesn+0.076
 belongs+0.071
 isn+0.070
Neg logit contributions
eyed-0.126
 pounding-0.122
Meanwhile-0.121
 widening-0.120
 crashing-0.120
Token!"
Token ID2474
Feature activation-1.25e-02

Pos logit contributions
 happens+0.075
 guys+0.072
 doesn+0.070
 isn+0.067
 belongs+0.065
Neg logit contributions
eyed-0.114
 widening-0.110
 pounding-0.109
Unfortunately-0.109
Meanwhile-0.109
Token Sam
Token ID3409
Feature activation+2.75e-02

Pos logit contributions
 happens+0.017
 isn+0.017
 is+0.017
 doesn+0.017
 belongs+0.017
Neg logit contributions
Unfortunately-0.029
Meanwhile-0.028
eyed-0.027
Just-0.027
 widening-0.026
Token said
Token ID531
Feature activation+1.94e-02

Pos logit contributions
Unfortunately+0.069
 sidelines+0.067
Meanwhile+0.066
Finally+0.066
eyed+0.065
Neg logit contributions
 is-0.034
 happens-0.032
 doesn-0.032
 belongs-0.031
 isn-0.030
Token,
Token ID11
Feature activation+1.26e-01

Pos logit contributions
Unfortunately+0.051
Meanwhile+0.051
Finally+0.048
 sidelines+0.048
Just+0.047
Neg logit contributions
 happens-0.022
 is-0.022
 doesn-0.020
 belongs-0.020
 isn-0.020
Token pointing
Token ID10609
Feature activation+1.86e-02

Pos logit contributions
Unfortunately+0.274
Finally+0.268
Eventually+0.264
Days+0.257
Years+0.256
Neg logit contributions
,-0.167
 is-0.162
 I-0.157
 brother-0.151
?"-0.148
Token to
Token ID284
Feature activation+1.43e-02

Pos logit contributions
Unfortunately+0.044
Meanwhile+0.042
 sidelines+0.041
Finally+0.040
 widening+0.040
Neg logit contributions
 is-0.028
 happens-0.027
 isn-0.026
 doesn-0.026
 belongs-0.025
Token a
Token ID257
Feature activation+3.63e-02

Pos logit contributions
Unfortunately+0.032
Meanwhile+0.031
eyed+0.030
Just+0.030
 widening+0.029
Neg logit contributions
 is-0.022
 happens-0.022
 guys-0.021
 doesn-0.021
 isn-0.021
Token red
Token ID2266
Feature activation+5.77e-03

Pos logit contributions
Unfortunately+0.079
Meanwhile+0.076
eyed+0.076
Just+0.074
Days+0.071
Neg logit contributions
 is-0.061
 isn-0.058
 happens-0.057
 doesn-0.055
 can-0.055
Token and
Token ID290
Feature activation+8.07e-03

Pos logit contributions
Unfortunately+0.012
Meanwhile+0.012
eyed+0.011
Just+0.011
 pounding+0.011
Neg logit contributions
 happens-0.009
 is-0.009
 doesn-0.009
 isn-0.009
 belongs-0.009
Token to
Token ID284
Feature activation+5.91e-03

Pos logit contributions
Unfortunately+0.090
 sidelines+0.089
Meanwhile+0.088
Finally+0.086
 widening+0.085
Neg logit contributions
 is-0.050
 doesn-0.049
 isn-0.045
 guys-0.044
 happens-0.044
Token the
Token ID262
Feature activation+8.09e-03

Pos logit contributions
Unfortunately+0.013
Meanwhile+0.012
eyed+0.012
Just+0.012
 widening+0.012
Neg logit contributions
 is-0.009
 happens-0.009
 doesn-0.009
 belongs-0.009
 isn-0.009
Token bird
Token ID6512
Feature activation+2.86e-02

Pos logit contributions
Unfortunately+0.020
eyed+0.019
Meanwhile+0.019
Just+0.019
 widening+0.019
Neg logit contributions
 happens-0.010
 is-0.010
 doesn-0.010
 belongs-0.010
 guys-0.010
Token and
Token ID290
Feature activation+1.63e-02

Pos logit contributions
Unfortunately+0.070
Meanwhile+0.068
 widening+0.065
eyed+0.065
Finally+0.065
Neg logit contributions
 is-0.039
 doesn-0.038
 belongs-0.036
 needs-0.036
 isn-0.036
Token whispered
Token ID25029
Feature activation+3.20e-02

Pos logit contributions
Unfortunately+0.032
Meanwhile+0.031
eyed+0.031
 widening+0.030
Just+0.030
Neg logit contributions
 happens-0.029
 is-0.028
 doesn-0.028
 belongs-0.028
 isn-0.027
Token,
Token ID11
Feature activation+1.25e-01

Pos logit contributions
Unfortunately+0.068
Meanwhile+0.066
 sidelines+0.062
eyed+0.062
Finally+0.062
Neg logit contributions
 happens-0.052
 is-0.052
 guys-0.050
 doesn-0.050
 isn-0.049
Token "
Token ID366
Feature activation+4.34e-02

Pos logit contributions
Days+0.344
har+0.343
amed+0.340
inge+0.339
 sidelines+0.337
Neg logit contributions
 I-0.089
 is-0.077
 doesn-0.074
 isn-0.072
 we-0.071
TokenI
Token ID40
Feature activation-6.14e-03

Pos logit contributions
rolling+0.075
 sidelines+0.075
 pounding+0.075
Unfortunately+0.074
 widening+0.074
Neg logit contributions
 doesn-0.076
 is-0.076
 isn-0.075
 happens-0.074
 belongs-0.070
Token will
Token ID481
Feature activation+3.41e-02

Pos logit contributions
 happens+0.008
 belongs+0.007
 doesn+0.007
 is+0.007
 guys+0.007
Neg logit contributions
Unfortunately-0.015
Meanwhile-0.015
eyed-0.015
Just-0.014
 widening-0.014
Token miss
Token ID2051
Feature activation-1.93e-02

Pos logit contributions
Meanwhile+0.067
Unfortunately+0.065
eyed+0.061
 widening+0.061
 sidelines+0.061
Neg logit contributions
 is-0.064
 doesn-0.061
 happens-0.059
 isn-0.059
 needs-0.059
Token you
Token ID345
Feature activation-1.40e-02

Pos logit contributions
 happens+0.019
 belongs+0.018
 is+0.017
 doesn+0.017
 isn+0.017
Neg logit contributions
Unfortunately-0.051
eyed-0.050
Meanwhile-0.049
Just-0.049
 pounding-0.048
Token it
Token ID340
Feature activation-6.82e-02

Pos logit contributions
unn+0.108
 guys+0.094
ishes+0.089
 happens+0.086
affles+0.083
Neg logit contributions
Unfortunately-0.186
Meanwhile-0.185
 ledge-0.185
 pounding-0.184
 lower-0.183
Token is
Token ID318
Feature activation-4.89e-02

Pos logit contributions
unn+0.094
 guys+0.093
mm+0.087
ishes+0.083
 guy+0.080
Neg logit contributions
Meanwhile-0.135
Unfortunately-0.130
Just-0.129
 pounding-0.127
On-0.126
Token!"
Token ID2474
Feature activation+4.48e-05

Pos logit contributions
unn+0.092
 happens+0.089
 belongs+0.086
 guys+0.085
 doesn+0.085
Neg logit contributions
Unfortunately-0.079
Meanwhile-0.076
 pounding-0.074
On-0.072
Just-0.070
Token Lily
Token ID20037
Feature activation+2.87e-02

Pos logit contributions
Unfortunately+0.000
Meanwhile+0.000
eyed+0.000
Just+0.000
 widening+0.000
Neg logit contributions
 happens-0.000
 is-0.000
 doesn-0.000
 belongs-0.000
 isn-0.000
Token whispered
Token ID25029
Feature activation+4.39e-02

Pos logit contributions
 sidelines+0.063
Unfortunately+0.063
eyed+0.061
Finally+0.061
Meanwhile+0.060
Neg logit contributions
 is-0.042
 doesn-0.041
 happens-0.040
 isn-0.039
 belongs-0.037
Token,
Token ID11
Feature activation+1.25e-01

Pos logit contributions
Unfortunately+0.118
Meanwhile+0.114
 sidelines+0.110
Finally+0.110
Eventually+0.109
Neg logit contributions
 guys-0.045
 is-0.043
 happens-0.043
 doesn-0.039
 brother-0.037
Token pointing
Token ID10609
Feature activation+1.32e-02

Pos logit contributions
Unfortunately+0.274
Finally+0.272
Eventually+0.269
Years+0.258
tted+0.250
Neg logit contributions
,-0.164
 brother-0.147
 is-0.145
 I-0.142
?"-0.134
Token to
Token ID284
Feature activation+9.36e-03

Pos logit contributions
Unfortunately+0.028
Meanwhile+0.027
 sidelines+0.025
 widening+0.025
eyed+0.025
Neg logit contributions
 happens-0.022
 is-0.022
 doesn-0.021
 isn-0.021
 belongs-0.021
Token the
Token ID262
Feature activation+3.84e-03

Pos logit contributions
Unfortunately+0.020
Meanwhile+0.020
eyed+0.020
Just+0.019
 widening+0.019
Neg logit contributions
 is-0.015
 happens-0.015
 guys-0.014
 doesn-0.014
 isn-0.014
Token doll
Token ID3654
Feature activation+1.52e-02

Pos logit contributions
Unfortunately+0.009
Meanwhile+0.009
eyed+0.008
Just+0.008
 pounding+0.008
Neg logit contributions
 happens-0.005
 is-0.005
 belongs-0.005
 doesn-0.005
 isn-0.005
Token.
Token ID13
Feature activation+3.13e-02

Pos logit contributions
Unfortunately+0.035
Meanwhile+0.034
eyed+0.034
Just+0.033
 widening+0.032
Neg logit contributions
 is-0.022
 happens-0.021
 belongs-0.021
 doesn-0.021
 isn-0.020
Token this
Token ID428
Feature activation-1.15e-02

Pos logit contributions
 happens+0.034
 belongs+0.034
 doesn+0.034
 isn+0.033
 is+0.033
Neg logit contributions
Unfortunately-0.036
eyed-0.035
Meanwhile-0.034
Just-0.033
However-0.032
Token one
Token ID530
Feature activation-1.49e-02

Pos logit contributions
 happens+0.022
 doesn+0.021
 belongs+0.021
 is+0.020
 isn+0.020
Neg logit contributions
Unfortunately-0.020
Meanwhile-0.020
eyed-0.020
 pounding-0.019
Just-0.019
Token!"
Token ID2474
Feature activation+3.05e-02

Pos logit contributions
 happens+0.024
 doesn+0.023
 belongs+0.023
 guys+0.023
 isn+0.022
Neg logit contributions
Unfortunately-0.031
Meanwhile-0.030
eyed-0.030
Just-0.029
 pounding-0.028
Token she
Token ID673
Feature activation+4.38e-02

Pos logit contributions
Unfortunately+0.070
Meanwhile+0.069
Days+0.069
 pounding+0.068
 sidelines+0.068
Neg logit contributions
 is-0.041
 happens-0.040
 isn-0.038
 doesn-0.037
 belongs-0.037
Token said
Token ID531
Feature activation+6.29e-02

Pos logit contributions
 sidelines+0.111
Unfortunately+0.109
 crashing+0.105
eyed+0.105
Meanwhile+0.104
Neg logit contributions
 happens-0.052
 is-0.050
 doesn-0.048
 isn-0.046
 belongs-0.046
Token,
Token ID11
Feature activation+1.24e-01

Pos logit contributions
Unfortunately+0.170
Meanwhile+0.169
 sidelines+0.165
Finally+0.164
Days+0.163
Neg logit contributions
 is-0.059
 happens-0.054
 doesn-0.052
 brother-0.050
 happened-0.049
Token picking
Token ID10868
Feature activation+2.40e-02

Pos logit contributions
Days+0.273
Unfortunately+0.268
Finally+0.257
Years+0.257
Hours+0.255
Neg logit contributions
,-0.153
 I-0.148
 is-0.142
?"-0.133
 doesn-0.133
Token up
Token ID510
Feature activation+2.77e-03

Pos logit contributions
Unfortunately+0.050
Meanwhile+0.049
Just+0.047
eyed+0.047
 pounding+0.046
Neg logit contributions
 is-0.041
 happens-0.039
 isn-0.038
 doesn-0.038
 belongs-0.037
Token the
Token ID262
Feature activation-9.16e-03

Pos logit contributions
Unfortunately+0.005
Meanwhile+0.005
eyed+0.005
Just+0.005
 widening+0.005
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 belongs-0.005
 isn-0.005
Token bunny
Token ID44915
Feature activation+3.53e-02

Pos logit contributions
 happens+0.016
 doesn+0.015
 belongs+0.015
 is+0.015
 isn+0.015
Neg logit contributions
Unfortunately-0.018
Meanwhile-0.018
eyed-0.017
 widening-0.017
 pounding-0.017
Token.
Token ID13
Feature activation+3.66e-02

Pos logit contributions
Unfortunately+0.083
Meanwhile+0.080
Finally+0.076
eyed+0.075
 sidelines+0.075
Neg logit contributions
 is-0.051
 belongs-0.048
 doesn-0.048
 isn-0.047
 needs-0.045
Token to
Token ID284
Feature activation+1.92e-02

Pos logit contributions
Unfortunately+0.118
eyed+0.115
Meanwhile+0.112
Finally+0.112
Just+0.110
Neg logit contributions
 is-0.051
 happens-0.046
 guys-0.044
 needs-0.042
 can-0.042
Token sleep
Token ID3993
Feature activation+2.32e-02

Pos logit contributions
Unfortunately+0.039
Meanwhile+0.039
eyed+0.037
Just+0.037
 widening+0.036
Neg logit contributions
 is-0.034
 happens-0.033
 doesn-0.032
 needs-0.031
 belongs-0.031
Token,
Token ID11
Feature activation+2.64e-02

Pos logit contributions
eyed+0.069
Unfortunately+0.068
Meanwhile+0.067
Just+0.065
har+0.065
Neg logit contributions
 is-0.017
 happens-0.017
 guys-0.016
 doesn-0.015
 isn-0.015
Token she
Token ID673
Feature activation+2.21e-02

Pos logit contributions
eyed+0.064
Unfortunately+0.061
Meanwhile+0.060
har+0.058
 widening+0.058
Neg logit contributions
 is-0.034
 happens-0.034
 doesn-0.034
 guys-0.033
 isn-0.033
Token whispered
Token ID25029
Feature activation+3.60e-02

Pos logit contributions
Unfortunately+0.048
eyed+0.047
Meanwhile+0.045
 sidelines+0.045
 widening+0.045
Neg logit contributions
 is-0.035
 happens-0.034
 doesn-0.034
 isn-0.032
 needs-0.032
Token,
Token ID11
Feature activation+1.24e-01

Pos logit contributions
Unfortunately+0.078
Meanwhile+0.074
 sidelines+0.072
eyed+0.072
Finally+0.071
Neg logit contributions
 happens-0.056
 is-0.055
 guys-0.054
 doesn-0.053
 isn-0.052
Token "
Token ID366
Feature activation+4.50e-02

Pos logit contributions
Days+0.336
inge+0.333
har+0.331
 sidelines+0.331
Unfortunately+0.329
Neg logit contributions
 is-0.089
 doesn-0.085
 I-0.081
 isn-0.080
 has-0.072
TokenI
Token ID40
Feature activation-2.02e-02

Pos logit contributions
Unfortunately+0.076
eyed+0.074
 victorious+0.074
 widening+0.073
 sidelines+0.073
Neg logit contributions
 doesn-0.078
 isn-0.078
 is-0.078
 happens-0.076
 guys-0.075
Token love
Token ID1842
Feature activation-3.21e-02

Pos logit contributions
 happens+0.033
 guys+0.032
 belongs+0.031
 isn+0.031
 doesn+0.030
Neg logit contributions
Unfortunately-0.040
eyed-0.040
Meanwhile-0.039
Just-0.038
 pounding-0.038
Token you
Token ID345
Feature activation-8.63e-03

Pos logit contributions
 happens+0.042
 belongs+0.040
 guys+0.037
 doesn+0.036
 is+0.035
Neg logit contributions
Unfortunately-0.075
Meanwhile-0.072
 pounding-0.072
Just-0.071
 crashing-0.071
Token,
Token ID11
Feature activation-7.31e-03

Pos logit contributions
 happens+0.013
 belongs+0.013
 doesn+0.013
 is+0.012
 isn+0.012
Neg logit contributions
Unfortunately-0.019
Meanwhile-0.018
eyed-0.018
Just-0.017
 pounding-0.017
Token is
Token ID318
Feature activation-5.20e-03

Pos logit contributions
 guys+0.057
 happens+0.054
unn+0.048
 doesn+0.041
 lives+0.041
Neg logit contributions
Unfortunately-0.168
Meanwhile-0.165
 pounding-0.164
eyed-0.163
Just-0.161
Token gone
Token ID3750
Feature activation-3.67e-02

Pos logit contributions
 happens+0.010
 is+0.010
 doesn+0.010
 belongs+0.010
 isn+0.010
Neg logit contributions
Unfortunately-0.009
Meanwhile-0.009
eyed-0.008
 pounding-0.008
Just-0.008
Token!"
Token ID2474
Feature activation+3.48e-02

Pos logit contributions
 happens+0.045
 belongs+0.041
 guys+0.040
 doesn+0.040
 isn+0.040
Neg logit contributions
Unfortunately-0.087
eyed-0.085
Meanwhile-0.085
 pounding-0.083
Just-0.082
Token he
Token ID339
Feature activation+3.80e-02

Pos logit contributions
Unfortunately+0.075
Meanwhile+0.075
Days+0.074
har+0.074
 pounding+0.074
Neg logit contributions
 is-0.049
 happens-0.049
 doesn-0.047
 isn-0.046
 belongs-0.044
Token said
Token ID531
Feature activation+6.43e-02

Pos logit contributions
 sidelines+0.093
Unfortunately+0.089
eyed+0.088
side+0.088
Meanwhile+0.086
Neg logit contributions
 is-0.050
 happens-0.049
 doesn-0.048
 isn-0.046
 means-0.044
Token,
Token ID11
Feature activation+1.24e-01

Pos logit contributions
Meanwhile+0.173
Unfortunately+0.171
 sidelines+0.168
Finally+0.167
Eventually+0.165
Neg logit contributions
 is-0.065
 doesn-0.059
 happens-0.057
 brother-0.056
 isn-0.055
Token between
Token ID1022
Feature activation+1.97e-02

Pos logit contributions
Finally+0.252
Eventually+0.249
Unfortunately+0.246
Days+0.246
Years+0.241
Neg logit contributions
 I-0.166
 is-0.164
,-0.159
 doesn-0.157
?"-0.154
Token so
Token ID523
Feature activation+2.88e-02

Pos logit contributions
Unfortunately+0.041
eyed+0.040
Meanwhile+0.040
Just+0.039
har+0.038
Neg logit contributions
 is-0.031
 happens-0.031
 guys-0.030
 isn-0.029
 doesn-0.029
Tokenbs
Token ID1443
Feature activation+3.41e-02

Pos logit contributions
Unfortunately+0.064
eyed+0.063
Meanwhile+0.063
 widening+0.061
rolling+0.061
Neg logit contributions
 happens-0.041
 is-0.039
 doesn-0.036
 needs-0.036
 guys-0.035
Token.
Token ID13
Feature activation+3.78e-02

Pos logit contributions
Unfortunately+0.081
Meanwhile+0.076
eyed+0.076
Finally+0.075
Just+0.074
Neg logit contributions
 is-0.047
 happens-0.046
 doesn-0.045
 isn-0.044
 guys-0.043
Token
Token ID198
Feature activation+4.56e-02

Pos logit contributions
eyed+0.074
 pounding+0.073
 widening+0.071
har+0.071
Meanwhile+0.071
Neg logit contributions
 is-0.062
 isn-0.061
 happens-0.060
 doesn-0.059
 belongs-0.055
Token L
Token ID406
Feature activation-3.82e-02

Pos logit contributions
 happens+0.018
 belongs+0.017
 guys+0.017
 doesn+0.016
 is+0.016
Neg logit contributions
Unfortunately-0.025
Meanwhile-0.025
eyed-0.024
Just-0.024
 pounding-0.024
Tokenila
Token ID10102
Feature activation-2.39e-02

Pos logit contributions
 guys+0.067
 needs+0.064
 happens+0.064
 belongs+0.063
 lives+0.062
Neg logit contributions
Just-0.072
Meanwhile-0.071
Unfortunately-0.070
eyed-0.068
Her-0.065
Token?"
Token ID1701
Feature activation+1.85e-02

Pos logit contributions
 happens+0.014
 guys+0.012
 isn+0.012
 belongs+0.011
 is+0.010
Neg logit contributions
Unfortunately-0.072
eyed-0.072
Meanwhile-0.072
Just-0.070
 pounding-0.069
Token she
Token ID673
Feature activation+3.00e-02

Pos logit contributions
Unfortunately+0.043
eyed+0.041
har+0.041
Meanwhile+0.041
 widening+0.040
Neg logit contributions
 is-0.027
 happens-0.026
 doesn-0.025
 belongs-0.024
 isn-0.024
Token asked
Token ID1965
Feature activation+5.48e-03

Pos logit contributions
Unfortunately+0.060
eyed+0.059
 sidelines+0.057
har+0.057
Finally+0.056
Neg logit contributions
 happens-0.053
 is-0.051
 doesn-0.050
 isn-0.047
 guys-0.047
Token,
Token ID11
Feature activation+1.24e-01

Pos logit contributions
Unfortunately+0.014
Meanwhile+0.014
eyed+0.013
Just+0.013
 widening+0.013
Neg logit contributions
 happens-0.006
 is-0.006
 doesn-0.006
 belongs-0.006
 isn-0.006
Token smiling
Token ID16755
Feature activation+3.40e-03

Pos logit contributions
Unfortunately+0.274
Finally+0.269
Days+0.265
Eventually+0.265
Hours+0.258
Neg logit contributions
 brother-0.169
?"-0.157
,-0.152
 sister-0.149
 is-0.144
Token.
Token ID13
Feature activation+3.18e-02

Pos logit contributions
Unfortunately+0.008
Meanwhile+0.008
eyed+0.007
Just+0.007
 widening+0.007
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 belongs-0.005
 isn-0.005
Token
Token ID198
Feature activation+1.87e-02

Pos logit contributions
 sidelines+0.066
eyed+0.065
Unfortunately+0.065
har+0.065
Meanwhile+0.064
Neg logit contributions
 is-0.053
 happens-0.050
 isn-0.048
 doesn-0.048
 can-0.046
Token
Token ID198
Feature activation+3.48e-02

Pos logit contributions
Unfortunately+0.035
Meanwhile+0.034
 widening+0.034
eyed+0.034
 pounding+0.033
Neg logit contributions
 is-0.035
 happens-0.034
 isn-0.032
 doesn-0.032
 belongs-0.032
TokenL
Token ID43
Feature activation-1.88e-02

Pos logit contributions
 widening+0.074
 sidelines+0.073
eyed+0.072
 pounding+0.071
 victorious+0.071
Neg logit contributions
 is-0.057
 happens-0.049
 belongs-0.048
 doesn-0.048
 isn-0.047
Token do
Token ID466
Feature activation-1.95e-04

Pos logit contributions
Unfortunately+0.002
Meanwhile+0.002
eyed+0.002
Just+0.001
 widening+0.001
Neg logit contributions
 happens-0.001
 is-0.001
 doesn-0.001
 belongs-0.001
 isn-0.001
Token it
Token ID340
Feature activation-3.12e-02

Pos logit contributions
 happens+0.000
 is+0.000
 doesn+0.000
 belongs+0.000
 isn+0.000
Neg logit contributions
Unfortunately-0.000
Meanwhile-0.000
eyed-0.000
Just-0.000
 widening-0.000
Token!"
Token ID2474
Feature activation+2.44e-02

Pos logit contributions
 happens+0.040
 belongs+0.038
 guys+0.037
 doesn+0.037
 isn+0.037
Neg logit contributions
Unfortunately-0.070
Meanwhile-0.069
eyed-0.069
Just-0.068
 pounding-0.067
Token Sarah
Token ID10490
Feature activation+4.30e-02

Pos logit contributions
Unfortunately+0.056
Days+0.055
Meanwhile+0.055
eyed+0.055
 widening+0.054
Neg logit contributions
 is-0.034
 happens-0.033
 doesn-0.033
 isn-0.032
 belongs-0.031
Token said
Token ID531
Feature activation+6.69e-02

Pos logit contributions
Unfortunately+0.097
 sidelines+0.095
Days+0.094
Meanwhile+0.093
Finally+0.092
Neg logit contributions
 is-0.062
 doesn-0.060
 isn-0.058
 happens-0.058
 belongs-0.054
Token,
Token ID11
Feature activation+1.24e-01

Pos logit contributions
Meanwhile+0.185
Unfortunately+0.184
 sidelines+0.180
Finally+0.180
Days+0.180
Neg logit contributions
 is-0.059
 doesn-0.053
 happens-0.051
 brother-0.049
 isn-0.048
Token laughing
Token ID14376
Feature activation+3.33e-02

Pos logit contributions
Days+0.253
Unfortunately+0.243
Finally+0.240
Years+0.239
Hours+0.238
Neg logit contributions
 I-0.177
,-0.169
 is-0.165
 doesn-0.162
?"-0.161
Token.
Token ID13
Feature activation+2.58e-02

Pos logit contributions
Unfortunately+0.077
Meanwhile+0.076
Finally+0.072
eyed+0.072
rolling+0.071
Neg logit contributions
 is-0.047
 doesn-0.046
 isn-0.046
 happens-0.045
 guys-0.044
Token
Token ID220
Feature activation+2.66e-02

Pos logit contributions
Unfortunately+0.042
eyed+0.042
Meanwhile+0.042
 pounding+0.040
 widening+0.039
Neg logit contributions
 is-0.052
 isn-0.052
 doesn-0.051
 happens-0.050
 belongs-0.048
Token
Token ID198
Feature activation+5.18e-02

Pos logit contributions
Unfortunately+0.049
Meanwhile+0.048
eyed+0.047
 widening+0.047
 sidelines+0.047
Neg logit contributions
 is-0.048
 happens-0.047
 isn-0.046
 doesn-0.046
 belongs-0.045
Token
Token ID198
Feature activation+6.21e-02

Pos logit contributions
 widening+0.112
 pounding+0.110
 ledge+0.107
 sidelines+0.107
Meanwhile+0.107
Neg logit contributions
 is-0.077
 isn-0.068
 doesn-0.064
 happens-0.062
?"-0.062
Token your
Token ID534
Feature activation-1.90e-02

Pos logit contributions
 happens+0.083
 belongs+0.081
 needs+0.080
 cares+0.078
 doesn+0.078
Neg logit contributions
Unfortunately-0.130
Meanwhile-0.126
eyed-0.122
On-0.122
However-0.121
Token princess
Token ID21752
Feature activation-4.54e-02

Pos logit contributions
 happens+0.033
 doesn+0.031
 isn+0.030
 belongs+0.030
 is+0.030
Neg logit contributions
Unfortunately-0.035
eyed-0.035
Meanwhile-0.034
 pounding-0.034
 widening-0.034
Token!"
Token ID2474
Feature activation-4.10e-03

Pos logit contributions
 happens+0.063
 isn+0.058
 doesn+0.057
 guys+0.056
 belongs+0.056
Neg logit contributions
eyed-0.093
 pounding-0.092
Unfortunately-0.092
Just-0.090
Meanwhile-0.089
Token Tom
Token ID4186
Feature activation+1.29e-02

Pos logit contributions
 happens+0.005
 is+0.004
 belongs+0.004
 doesn+0.004
 isn+0.004
Neg logit contributions
Unfortunately-0.011
Meanwhile-0.010
eyed-0.010
Just-0.010
 widening-0.010
Token said
Token ID531
Feature activation+2.67e-02

Pos logit contributions
Unfortunately+0.029
Meanwhile+0.028
eyed+0.027
 sidelines+0.027
Finally+0.027
Neg logit contributions
 is-0.019
 happens-0.019
 doesn-0.019
 belongs-0.018
 isn-0.018
Token,
Token ID11
Feature activation+1.24e-01

Pos logit contributions
Unfortunately+0.075
Meanwhile+0.074
Finally+0.071
Eventually+0.070
 sidelines+0.069
Neg logit contributions
 is-0.026
 happens-0.026
 doesn-0.024
 isn-0.023
 belongs-0.023
Token making
Token ID1642
Feature activation+2.05e-02

Pos logit contributions
Unfortunately+0.260
Finally+0.254
Eventually+0.254
Days+0.241
Years+0.237
Neg logit contributions
 is-0.166
 I-0.166
,-0.163
 wants-0.158
 likes-0.153
Token his
Token ID465
Feature activation-6.38e-03

Pos logit contributions
Unfortunately+0.045
Meanwhile+0.044
eyed+0.044
Just+0.043
 widening+0.042
Neg logit contributions
 is-0.031
 happens-0.031
 guys-0.030
 doesn-0.030
 isn-0.029
Token wire
Token ID6503
Feature activation+4.86e-02

Pos logit contributions
 happens+0.011
 doesn+0.010
 is+0.010
 belongs+0.010
 isn+0.010
Neg logit contributions
Unfortunately-0.013
Meanwhile-0.013
eyed-0.013
 widening-0.012
 pounding-0.012
Token w
Token ID266
Feature activation+1.65e-02

Pos logit contributions
Unfortunately+0.098
Meanwhile+0.095
Just+0.090
Finally+0.090
eyed+0.089
Neg logit contributions
 is-0.095
 doesn-0.088
 belongs-0.086
 happens-0.084
 isn-0.084
Tokeniggle
Token ID24082
Feature activation+4.39e-02

Pos logit contributions
Unfortunately+0.030
Meanwhile+0.030
eyed+0.029
 pounding+0.028
rolling+0.028
Neg logit contributions
 happens-0.030
 is-0.030
 guys-0.029
 belongs-0.029
 doesn-0.029
Token you
Token ID345
Feature activation-9.70e-03

Pos logit contributions
 happens+0.079
unn+0.075
 guys+0.074
 belongs+0.073
 needs+0.070
Neg logit contributions
Unfortunately-0.139
Finally-0.134
Meanwhile-0.131
Just-0.131
 widening-0.128
Token go
Token ID467
Feature activation-9.96e-03

Pos logit contributions
 happens+0.017
 doesn+0.016
 belongs+0.016
 is+0.016
 isn+0.016
Neg logit contributions
Unfortunately-0.019
Meanwhile-0.018
eyed-0.018
Just-0.017
 widening-0.017
Token,"
Token ID553
Feature activation+2.53e-02

Pos logit contributions
 happens+0.015
 doesn+0.014
 belongs+0.014
 is+0.014
 guys+0.014
Neg logit contributions
Unfortunately-0.021
Meanwhile-0.021
eyed-0.021
Just-0.020
 pounding-0.020
Token she
Token ID673
Feature activation+3.61e-02

Pos logit contributions
eyed+0.058
Unfortunately+0.058
Meanwhile+0.057
rolling+0.055
 widening+0.055
Neg logit contributions
 is-0.036
 happens-0.035
 doesn-0.034
 isn-0.034
 belongs-0.033
Token said
Token ID531
Feature activation+6.42e-02

Pos logit contributions
 sidelines+0.088
Unfortunately+0.086
eyed+0.085
 crashing+0.084
Meanwhile+0.084
Neg logit contributions
 happens-0.047
 is-0.045
 doesn-0.044
 isn-0.043
 belongs-0.042
Token,
Token ID11
Feature activation+1.23e-01

Pos logit contributions
Meanwhile+0.179
Unfortunately+0.178
 sidelines+0.174
Finally+0.172
Eventually+0.170
Neg logit contributions
 is-0.055
 brother-0.050
 doesn-0.049
 happens-0.047
 sister-0.046
Token pat
Token ID1458
Feature activation+4.58e-02

Pos logit contributions
Days+0.255
Unfortunately+0.247
Finally+0.241
Years+0.240
Hours+0.238
Neg logit contributions
 is-0.156
,-0.155
 I-0.155
 doesn-0.152
 isn-0.144
Tokenting
Token ID889
Feature activation+6.50e-02

Pos logit contributions
Unfortunately+0.096
Meanwhile+0.092
eyed+0.088
Days+0.087
Just+0.085
Neg logit contributions
 is-0.073
 guys-0.071
 doesn-0.071
 happens-0.069
 isn-0.069
Token the
Token ID262
Feature activation+1.62e-02

Pos logit contributions
Unfortunately+0.157
Meanwhile+0.157
Just+0.152
Finally+0.151
Eventually+0.147
Neg logit contributions
 is-0.087
 doesn-0.080
 guys-0.078
 isn-0.077
 happens-0.076
Token little
Token ID1310
Feature activation-1.01e-02

Pos logit contributions
Unfortunately+0.034
eyed+0.034
Meanwhile+0.033
His+0.031
rolling+0.031
Neg logit contributions
 is-0.027
 happens-0.026
 doesn-0.025
 isn-0.025
 needs-0.025
Token girl
Token ID2576
Feature activation+7.28e-02

Pos logit contributions
 happens+0.014
 belongs+0.013
 doesn+0.013
 is+0.013
 isn+0.013
Neg logit contributions
Unfortunately-0.024
Meanwhile-0.023
eyed-0.023
 widening-0.022
 pounding-0.022
Token something
Token ID1223
Feature activation-1.92e-02

Pos logit contributions
 happens+0.017
 belongs+0.016
 doesn+0.016
 is+0.016
 isn+0.016
Neg logit contributions
Unfortunately-0.018
Meanwhile-0.017
eyed-0.017
Just-0.016
 pounding-0.016
Token together
Token ID1978
Feature activation-2.69e-02

Pos logit contributions
 happens+0.038
 guys+0.036
 doesn+0.036
 belongs+0.036
 is+0.036
Neg logit contributions
Unfortunately-0.032
eyed-0.031
Meanwhile-0.031
 pounding-0.029
Just-0.029
Token!"
Token ID2474
Feature activation+6.78e-03

Pos logit contributions
 happens+0.030
 guys+0.029
 belongs+0.029
 doesn+0.028
 is+0.027
Neg logit contributions
Unfortunately-0.066
eyed-0.065
Meanwhile-0.065
 pounding-0.063
 widening-0.063
Token said
Token ID531
Feature activation-1.59e-02

Pos logit contributions
Unfortunately+0.014
Meanwhile+0.013
eyed+0.013
 pounding+0.013
 widening+0.013
Neg logit contributions
 happens-0.011
 is-0.011
 doesn-0.011
 belongs-0.011
 isn-0.011
Token Emma
Token ID18966
Feature activation+6.12e-02

Pos logit contributions
 happens+0.022
 belongs+0.022
 doesn+0.021
 is+0.021
 isn+0.021
Neg logit contributions
Unfortunately-0.036
Meanwhile-0.035
eyed-0.035
 pounding-0.034
 widening-0.034
Token,
Token ID11
Feature activation+1.23e-01

Pos logit contributions
Unfortunately+0.176
 sidelines+0.170
Finally+0.170
Meanwhile+0.168
Days+0.168
Neg logit contributions
 is-0.052
 doesn-0.046
 isn-0.044
 happens-0.043
 brother-0.041
Token adm
Token ID6178
Feature activation+3.75e-02

Pos logit contributions
Unfortunately+0.245
Days+0.241
Finally+0.234
Years+0.232
Eventually+0.229
Neg logit contributions
,-0.172
 I-0.164
 brother-0.163
?"-0.156
 is-0.149
Tokeniring
Token ID3428
Feature activation+2.97e-02

Pos logit contributions
Unfortunately+0.085
 sidelines+0.079
Meanwhile+0.078
shadow+0.076
Finally+0.076
Neg logit contributions
 happens-0.052
 is-0.052
 belongs-0.048
?"-0.047
 doesn-0.047
Token their
Token ID511
Feature activation+2.70e-02

Pos logit contributions
Unfortunately+0.067
Meanwhile+0.066
eyed+0.065
Just+0.064
Finally+0.062
Neg logit contributions
 is-0.045
 happens-0.044
 doesn-0.043
 isn-0.043
 belongs-0.041
Token work
Token ID670
Feature activation+6.93e-02

Pos logit contributions
Unfortunately+0.058
Meanwhile+0.056
eyed+0.054
Just+0.053
However+0.053
Neg logit contributions
 is-0.046
 happens-0.044
 isn-0.044
 belongs-0.044
 lives-0.044
Token.
Token ID13
Feature activation+2.65e-02

Pos logit contributions
Unfortunately+0.178
Meanwhile+0.175
Just+0.168
Finally+0.167
 sidelines+0.164
Neg logit contributions
 is-0.089
 happens-0.081
 doesn-0.078
 belongs-0.077
 isn-0.076
Token stickers
Token ID28568
Feature activation+1.19e-02

Pos logit contributions
Unfortunately+0.013
Meanwhile+0.013
eyed+0.013
Just+0.012
 widening+0.012
Neg logit contributions
 happens-0.009
 is-0.009
 doesn-0.009
 isn-0.008
 belongs-0.008
Token.
Token ID13
Feature activation-8.86e-03

Pos logit contributions
Unfortunately+0.028
Meanwhile+0.027
eyed+0.026
Just+0.026
 widening+0.025
Neg logit contributions
 is-0.017
 happens-0.017
 doesn-0.016
 belongs-0.016
 guys-0.016
Token
Token ID198
Feature activation+2.78e-02

Pos logit contributions
 happens+0.019
 belongs+0.018
 guys+0.018
 doesn+0.018
 is+0.018
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.013
Just-0.013
 widening-0.013
Token
Token ID198
Feature activation+3.71e-02

Pos logit contributions
 widening+0.054
 sidelines+0.054
 pounding+0.054
eyed+0.053
Unfortunately+0.053
Neg logit contributions
 is-0.049
 happens-0.045
 isn-0.044
 doesn-0.043
 belongs-0.043
Token"
Token ID1
Feature activation+2.30e-02

Pos logit contributions
 pounding+0.058
 victorious+0.057
 widening+0.057
eyed+0.056
har+0.055
Neg logit contributions
 is-0.077
 happens-0.072
 isn-0.071
 belongs-0.071
 doesn-0.070
TokenKing
Token ID15708
Feature activation-1.41e-01

Pos logit contributions
rolling+0.041
eyed+0.040
 pounding+0.040
Unfortunately+0.040
 victorious+0.039
Neg logit contributions
 happens-0.043
 is-0.042
 guys-0.040
 isn-0.040
 doesn-0.039
Token Tom
Token ID4186
Feature activation-8.49e-02

Pos logit contributions
 needs+0.248
 belongs+0.238
 happens+0.234
 doesn+0.224
mm+0.217
Neg logit contributions
 ï¿½-0.170
shadow-0.166
king-0.163
eyed-0.161
On-0.160
Token,
Token ID11
Feature activation-5.85e-02

Pos logit contributions
 guys+0.143
 happens+0.142
unn+0.142
mm+0.128
 belongs+0.126
Neg logit contributions
 pounding-0.133
eyed-0.131
 widening-0.124
 victorious-0.120
Meanwhile-0.120
Token I
Token ID314
Feature activation-2.09e-02

Pos logit contributions
unn+0.084
 happens+0.083
 belongs+0.080
ishes+0.077
uffy+0.075
Neg logit contributions
Unfortunately-0.112
On-0.106
However-0.105
Meanwhile-0.103
Just-0.102
Token have
Token ID423
Feature activation-3.09e-02

Pos logit contributions
 happens+0.035
 belongs+0.033
 doesn+0.032
 isn+0.032
 guys+0.032
Neg logit contributions
eyed-0.041
Unfortunately-0.041
Meanwhile-0.039
Just-0.038
 pounding-0.038
Token a
Token ID257
Feature activation-7.48e-03

Pos logit contributions
 belongs+0.061
 happens+0.061
 doesn+0.061
 isn+0.059
 is+0.058
Neg logit contributions
Unfortunately-0.049
Meanwhile-0.048
Just-0.046
eyed-0.045
 pounding-0.043
Token!
Token ID0
Feature activation-6.10e-02

Pos logit contributions
 happens+0.169
 guys+0.166
unn+0.164
 belongs+0.156
 lives+0.151
Neg logit contributions
But-0.081
Just-0.081
eyed-0.079
 pounding-0.079
His-0.077
Token You
Token ID921
Feature activation-3.64e-02

Pos logit contributions
 happens+0.085
 guys+0.084
 belongs+0.083
ishes+0.078
 tastes+0.074
Neg logit contributions
Unfortunately-0.134
Just-0.127
eyed-0.127
Meanwhile-0.125
However-0.124
Token are
Token ID389
Feature activation-1.07e-02

Pos logit contributions
 belongs+0.037
 happens+0.036
 doesn+0.033
 isn+0.032
 needs+0.032
Neg logit contributions
Unfortunately-0.090
eyed-0.087
 victorious-0.087
 widening-0.086
Meanwhile-0.084
Token the
Token ID262
Feature activation-6.78e-03

Pos logit contributions
 happens+0.021
 belongs+0.020
 doesn+0.020
 is+0.020
 isn+0.019
Neg logit contributions
Unfortunately-0.019
Meanwhile-0.018
eyed-0.018
 pounding-0.017
 widening-0.017
Token best
Token ID1266
Feature activation-2.95e-02

Pos logit contributions
 happens+0.008
 belongs+0.007
 doesn+0.007
 is+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.017
Meanwhile-0.017
eyed-0.017
Just-0.017
 widening-0.016
Token tutor
Token ID48680
Feature activation-1.32e-01

Pos logit contributions
 belongs+0.050
 happens+0.049
 doesn+0.049
 isn+0.046
 needs+0.045
Neg logit contributions
eyed-0.056
Unfortunately-0.056
 crashing-0.054
Meanwhile-0.054
 pounding-0.054
Token ever
Token ID1683
Feature activation-5.14e-02

Pos logit contributions
 guys+0.269
mm+0.257
 happens+0.257
unn+0.253
ishes+0.252
Neg logit contributions
eyed-0.192
 victorious-0.163
Just-0.154
 free-0.148
But-0.147
Token!"
Token ID2474
Feature activation-4.19e-03

Pos logit contributions
uffy+0.079
ishes+0.079
 belongs+0.077
 happens+0.077
 guys+0.076
Neg logit contributions
eyed-0.095
 victorious-0.089
shadow-0.088
 free-0.088
Just-0.087
Token Tom
Token ID4186
Feature activation-8.89e-03

Pos logit contributions
 happens+0.006
 belongs+0.006
 is+0.006
 doesn+0.006
 isn+0.006
Neg logit contributions
Unfortunately-0.010
Meanwhile-0.009
eyed-0.009
Just-0.009
 widening-0.009
Token smiled
Token ID13541
Feature activation+2.37e-02

Pos logit contributions
 happens+0.017
 belongs+0.016
 doesn+0.016
 guys+0.016
 is+0.016
Neg logit contributions
Unfortunately-0.016
Meanwhile-0.015
eyed-0.015
Just-0.015
 widening-0.015
Token and
Token ID290
Feature activation-2.12e-04

Pos logit contributions
Unfortunately+0.057
Meanwhile+0.055
 sidelines+0.055
eyed+0.053
Finally+0.053
Neg logit contributions
 is-0.033
 happens-0.031
 doesn-0.031
 guys-0.030
 isn-0.030
Token,
Token ID11
Feature activation+7.90e-02

Pos logit contributions
 belongs+0.027
 doesn+0.026
 happens+0.026
 isn+0.025
 needs+0.025
Neg logit contributions
Unfortunately-0.052
eyed-0.050
Meanwhile-0.050
His-0.050
Just-0.050
Token "
Token ID366
Feature activation+2.01e-02

Pos logit contributions
har+0.219
eyed+0.217
inge+0.212
 sidelines+0.211
Unfortunately+0.209
Neg logit contributions
 is-0.074
 can-0.066
 isn-0.065
 I-0.065
 doesn-0.062
TokenYes
Token ID5297
Feature activation-4.46e-02

Pos logit contributions
Unfortunately+0.038
Meanwhile+0.038
 pounding+0.037
rolling+0.037
eyed+0.037
Neg logit contributions
 is-0.034
 happens-0.033
 isn-0.033
 doesn-0.033
 belongs-0.032
Token,
Token ID11
Feature activation-2.48e-02

Pos logit contributions
 belongs+0.072
 happens+0.072
 lives+0.068
 needs+0.067
ishes+0.066
Neg logit contributions
Unfortunately-0.085
Just-0.082
 crashing-0.082
His-0.081
 widening-0.080
Token my
Token ID616
Feature activation-3.30e-02

Pos logit contributions
 happens+0.026
 belongs+0.025
 is+0.023
 doesn+0.023
 needs+0.022
Neg logit contributions
Unfortunately-0.064
Meanwhile-0.062
Just-0.061
However-0.060
eyed-0.060
Token Queen
Token ID7542
Feature activation-1.32e-01

Pos logit contributions
 happens+0.063
 belongs+0.058
 doesn+0.056
 isn+0.055
 is+0.054
Neg logit contributions
 pounding-0.054
Just-0.054
Unfortunately-0.054
 widening-0.054
eyed-0.053
Token.
Token ID13
Feature activation-5.38e-02

Pos logit contributions
mm+0.208
 guys+0.205
 happens+0.201
 doesn+0.197
 belongs+0.192
Neg logit contributions
eyed-0.195
 victorious-0.181
 pat-0.171
 grate-0.170
 pounding-0.165
Token The
Token ID383
Feature activation-4.61e-03

Pos logit contributions
 happens+0.069
 guys+0.067
 belongs+0.061
 happened+0.060
 doesn+0.057
Neg logit contributions
Unfortunately-0.130
eyed-0.125
Just-0.124
Meanwhile-0.123
His-0.119
Token shield
Token ID7614
Feature activation-5.52e-02

Pos logit contributions
 happens+0.008
 is+0.007
 belongs+0.007
 doesn+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.009
Meanwhile-0.009
eyed-0.009
Just-0.009
 widening-0.009
Token will
Token ID481
Feature activation+5.51e-02

Pos logit contributions
 guys+0.038
 happens+0.038
 doesn+0.028
 isn+0.028
unn+0.026
Neg logit contributions
eyed-0.152
Unfortunately-0.152
Meanwhile-0.149
Just-0.147
 pounding-0.147
Token protect
Token ID1805
Feature activation-9.65e-03

Pos logit contributions
Meanwhile+0.115
Unfortunately+0.113
eyed+0.108
rolling+0.106
Just+0.105
Neg logit contributions
 is-0.099
 can-0.090
 needs-0.088
 belongs-0.088
 doesn-0.086
Token fly
Token ID6129
Feature activation+1.94e-02

Pos logit contributions
 happens+0.000
 is+0.000
 doesn+0.000
 belongs+0.000
 isn+0.000
Neg logit contributions
Unfortunately-0.000
Meanwhile-0.000
eyed-0.000
Just-0.000
 widening-0.000
Token and
Token ID290
Feature activation-1.23e-02

Pos logit contributions
Unfortunately+0.044
Meanwhile+0.043
eyed+0.042
Just+0.041
Finally+0.041
Neg logit contributions
 is-0.029
 happens-0.028
 doesn-0.028
 belongs-0.028
 isn-0.027
Token thought
Token ID1807
Feature activation-2.64e-02

Pos logit contributions
 happens+0.022
 belongs+0.021
 is+0.021
 isn+0.021
 doesn+0.021
Neg logit contributions
Unfortunately-0.023
Meanwhile-0.023
eyed-0.022
 widening-0.021
Just-0.021
Token,
Token ID11
Feature activation+3.23e-02

Pos logit contributions
 belongs+0.041
 happens+0.040
 doesn+0.039
 guys+0.038
 needs+0.038
Neg logit contributions
Unfortunately-0.054
Meanwhile-0.051
 pounding-0.051
 widening-0.051
Just-0.050
Token "
Token ID366
Feature activation+1.84e-02

Pos logit contributions
eyed+0.069
har+0.067
Unfortunately+0.066
Meanwhile+0.065
rolling+0.063
Neg logit contributions
 doesn-0.054
 is-0.053
 isn-0.052
 happens-0.052
 guys-0.050
TokenHmm
Token ID44217
Feature activation-1.31e-01

Pos logit contributions
Unfortunately+0.037
 sidelines+0.037
 pounding+0.037
rolling+0.037
eyed+0.037
Neg logit contributions
 happens-0.029
 is-0.029
 doesn-0.028
 belongs-0.028
 isn-0.028
Token,
Token ID11
Feature activation-4.43e-02

Pos logit contributions
unn+0.255
 cares+0.244
ishes+0.236
 needs+0.235
 lives+0.229
Neg logit contributions
Suddenly-0.133
 low-0.129
 firmly-0.125
 cautiously-0.125
 carefully-0.123
Token maybe
Token ID3863
Feature activation-7.27e-02

Pos logit contributions
 happens+0.075
 belongs+0.074
unn+0.070
 needs+0.067
 lives+0.067
Neg logit contributions
Unfortunately-0.080
Meanwhile-0.075
Just-0.073
 crashing-0.073
However-0.073
Token I
Token ID314
Feature activation-3.32e-02

Pos logit contributions
unn+0.111
 happens+0.108
uffy+0.108
 belongs+0.107
 isn+0.104
Neg logit contributions
But-0.132
Suddenly-0.128
Just-0.126
His-0.125
On-0.125
Token can
Token ID460
Feature activation+1.34e-02

Pos logit contributions
 happens+0.030
 guys+0.027
 belongs+0.026
mm+0.026
 doesn+0.025
Neg logit contributions
eyed-0.087
Unfortunately-0.087
Meanwhile-0.084
On-0.083
Just-0.082
Token bite
Token ID13197
Feature activation-4.12e-02

Pos logit contributions
Unfortunately+0.023
Meanwhile+0.023
eyed+0.022
 pounding+0.022
 widening+0.021
Neg logit contributions
 is-0.027
 belongs-0.026
 happens-0.026
 guys-0.026
 doesn-0.025
Token sick
Token ID6639
Feature activation-2.90e-02

Pos logit contributions
Unfortunately+0.053
Meanwhile+0.051
Finally+0.049
eyed+0.048
 sidelines+0.048
Neg logit contributions
 is-0.046
 happens-0.045
 doesn-0.044
?"-0.044
 needs-0.044
Token.
Token ID13
Feature activation-4.50e-03

Pos logit contributions
 happens+0.053
 belongs+0.053
 isn+0.051
 doesn+0.050
 lives+0.050
Neg logit contributions
Unfortunately-0.051
eyed-0.049
Meanwhile-0.048
 pounding-0.048
Just-0.047
Token
Token ID198
Feature activation+3.30e-03

Pos logit contributions
 happens+0.009
 is+0.009
 doesn+0.009
 belongs+0.009
 guys+0.009
Neg logit contributions
Unfortunately-0.008
Meanwhile-0.007
eyed-0.007
Just-0.007
 widening-0.007
Token
Token ID198
Feature activation+2.28e-02

Pos logit contributions
Unfortunately+0.005
Meanwhile+0.005
eyed+0.005
 widening+0.005
 pounding+0.005
Neg logit contributions
 happens-0.007
 is-0.007
 doesn-0.006
 belongs-0.006
 isn-0.006
Token"
Token ID1
Feature activation+3.21e-03

Pos logit contributions
 victorious+0.041
 sidelines+0.041
Unfortunately+0.040
 widening+0.040
 pounding+0.040
Neg logit contributions
 is-0.041
 belongs-0.038
 isn-0.037
 happens-0.037
 lives-0.036
TokenHere
Token ID4342
Feature activation-1.29e-01

Pos logit contributions
Unfortunately+0.005
eyed+0.005
Meanwhile+0.005
 pounding+0.004
 widening+0.004
Neg logit contributions
 happens-0.007
 is-0.007
 doesn-0.007
 belongs-0.007
 isn-0.007
Token,
Token ID11
Feature activation-4.43e-02

Pos logit contributions
unn+0.259
elly+0.257
ays+0.250
oline+0.240
 favorites+0.239
Neg logit contributions
 free-0.138
His-0.133
 pounding-0.126
The-0.125
 calm-0.120
Token Tim
Token ID5045
Feature activation-6.95e-02

Pos logit contributions
 happens+0.076
unn+0.070
 belongs+0.068
 doesn+0.066
ishes+0.066
Neg logit contributions
Unfortunately-0.082
Meanwhile-0.077
His-0.076
 crashing-0.073
However-0.072
Token,
Token ID11
Feature activation-3.31e-02

Pos logit contributions
 happens+0.123
unn+0.118
mm+0.117
 guys+0.116
 isn+0.114
Neg logit contributions
 pounding-0.105
eyed-0.105
 free-0.103
Just-0.095
Meanwhile-0.094
Token let
Token ID1309
Feature activation-5.52e-02

Pos logit contributions
 happens+0.048
 belongs+0.045
unn+0.044
 guys+0.043
 doesn+0.042
Neg logit contributions
Unfortunately-0.070
Meanwhile-0.066
eyed-0.065
 crashing-0.064
His-0.063
Token me
Token ID502
Feature activation+4.39e-03

Pos logit contributions
 cares+0.103
unn+0.098
 happens+0.097
 needs+0.097
 belongs+0.096
Neg logit contributions
Unfortunately-0.086
Meanwhile-0.082
Hours-0.079
The-0.078
 free-0.078
TokenLet
Token ID5756
Feature activation-5.35e-02

Pos logit contributions
Unfortunately+0.016
eyed+0.016
Meanwhile+0.016
 pounding+0.015
rolling+0.015
Neg logit contributions
 happens-0.019
 is-0.018
 guys-0.018
 doesn-0.017
 isn-0.017
Token's
Token ID338
Feature activation+2.52e-03

Pos logit contributions
unn+0.104
mm+0.097
 cares+0.096
 dinosaurs+0.096
ishes+0.096
Neg logit contributions
Unfortunately-0.082
Her-0.077
Meanwhile-0.077
The-0.077
Just-0.076
Token go
Token ID467
Feature activation+7.74e-04

Pos logit contributions
Unfortunately+0.004
Meanwhile+0.004
eyed+0.004
Just+0.004
 widening+0.004
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 belongs-0.005
 guys-0.005
Token and
Token ID290
Feature activation+6.08e-05

Pos logit contributions
Unfortunately+0.001
Meanwhile+0.001
eyed+0.001
 widening+0.001
Just+0.001
Neg logit contributions
 happens-0.001
 is-0.001
 doesn-0.001
 belongs-0.001
 isn-0.001
Token see
Token ID766
Feature activation-5.60e-02

Pos logit contributions
Unfortunately+0.000
Meanwhile+0.000
eyed+0.000
Just+0.000
 widening+0.000
Neg logit contributions
 happens-0.000
 is-0.000
 doesn-0.000
 belongs-0.000
 isn-0.000
Token what
Token ID644
Feature activation-1.28e-01

Pos logit contributions
 happens+0.103
 belongs+0.102
 needs+0.099
 guys+0.098
 cares+0.097
Neg logit contributions
Unfortunately-0.085
Suddenly-0.084
eyed-0.084
shadow-0.084
But-0.083
Token is
Token ID318
Feature activation-3.11e-02

Pos logit contributions
unn+0.197
mm+0.177
ishes+0.176
 guys+0.172
ems+0.171
Neg logit contributions
The-0.225
 low-0.205
On-0.200
 slowly-0.198
 nearby-0.198
Token inside
Token ID2641
Feature activation-6.21e-02

Pos logit contributions
 doesn+0.061
 isn+0.061
 happens+0.061
 needs+0.060
 belongs+0.060
Neg logit contributions
Unfortunately-0.046
eyed-0.044
Meanwhile-0.044
 widening-0.042
 pounding-0.042
Token,"
Token ID553
Feature activation-3.12e-02

Pos logit contributions
 happens+0.094
 guys+0.088
 cares+0.085
 belongs+0.084
 doesn+0.082
Neg logit contributions
eyed-0.123
Unfortunately-0.121
Meanwhile-0.118
 lower-0.115
On-0.114
Token Bee
Token ID24719
Feature activation+1.74e-03

Pos logit contributions
 isn+0.054
 doesn+0.053
 belongs+0.053
 guys+0.052
 happens+0.052
Neg logit contributions
Meanwhile-0.058
Unfortunately-0.058
eyed-0.056
His-0.055
Just-0.055
Token said
Token ID531
Feature activation+1.18e-02

Pos logit contributions
Unfortunately+0.004
Meanwhile+0.004
eyed+0.003
 widening+0.003
Just+0.003
Neg logit contributions
 happens-0.003
 is-0.003
 doesn-0.003
 belongs-0.003
 isn-0.003
Token mom
Token ID1995
Feature activation-1.80e-02

Pos logit contributions
 happens+0.014
 belongs+0.014
 isn+0.013
 doesn+0.013
 is+0.013
Neg logit contributions
Unfortunately-0.013
Meanwhile-0.013
eyed-0.013
Just-0.012
 pounding-0.012
Token.
Token ID13
Feature activation+1.44e-03

Pos logit contributions
 happens+0.033
 belongs+0.032
 isn+0.031
 guys+0.031
 doesn+0.031
Neg logit contributions
eyed-0.034
Unfortunately-0.033
Meanwhile-0.032
Just-0.031
 pounding-0.031
Token
Token ID198
Feature activation+2.25e-02

Pos logit contributions
Unfortunately+0.002
Meanwhile+0.002
eyed+0.002
 widening+0.002
 pounding+0.002
Neg logit contributions
 happens-0.003
 is-0.003
 doesn-0.003
 belongs-0.003
 isn-0.003
Token
Token ID198
Feature activation+3.63e-02

Pos logit contributions
Unfortunately+0.042
 widening+0.041
Meanwhile+0.041
 pounding+0.041
eyed+0.041
Neg logit contributions
 is-0.041
 happens-0.039
 doesn-0.038
 belongs-0.037
 isn-0.037
Token"
Token ID1
Feature activation+2.38e-02

Pos logit contributions
 pounding+0.076
 widening+0.076
rolling+0.076
 crashing+0.075
 victorious+0.074
Neg logit contributions
 is-0.058
 happens-0.049
?"-0.047
 belongs-0.047
 needs-0.047
TokenHere
Token ID4342
Feature activation-1.27e-01

Pos logit contributions
rolling+0.044
eyed+0.043
 pounding+0.042
Unfortunately+0.042
har+0.041
Neg logit contributions
 happens-0.043
 is-0.043
 doesn-0.040
 isn-0.039
 needs-0.038
Token,
Token ID11
Feature activation-3.76e-02

Pos logit contributions
unn+0.279
akes+0.274
ishes+0.274
 guys+0.272
eer+0.261
Neg logit contributions
Her-0.133
But-0.121
 cheer-0.121
The-0.119
 firm-0.113
Token you
Token ID345
Feature activation-3.10e-02

Pos logit contributions
 happens+0.059
 belongs+0.056
 guys+0.056
unn+0.053
ishes+0.052
Neg logit contributions
Unfortunately-0.074
Meanwhile-0.073
Just-0.068
However-0.068
Her-0.068
Token can
Token ID460
Feature activation-1.55e-02

Pos logit contributions
 belongs+0.037
 happens+0.036
 doesn+0.034
mm+0.034
 isn+0.033
Neg logit contributions
eyed-0.069
Unfortunately-0.069
Meanwhile-0.066
rolling-0.065
Just-0.065
Token have
Token ID423
Feature activation-3.82e-02

Pos logit contributions
 happens+0.030
 doesn+0.028
 belongs+0.028
 isn+0.028
 guys+0.028
Neg logit contributions
eyed-0.027
Unfortunately-0.027
Meanwhile-0.026
Just-0.025
 widening-0.025
Token my
Token ID616
Feature activation-4.25e-02

Pos logit contributions
 belongs+0.070
 doesn+0.068
 isn+0.068
 needs+0.067
 happens+0.067
Neg logit contributions
Unfortunately-0.067
Just-0.063
Meanwhile-0.062
However-0.061
Their-0.060
Token on
Token ID319
Feature activation+2.42e-02

Pos logit contributions
Unfortunately+0.108
Meanwhile+0.105
 sidelines+0.105
rolling+0.101
 widening+0.101
Neg logit contributions
 is-0.068
 happens-0.065
 doesn-0.062
 guys-0.062
 belongs-0.062
Token her
Token ID607
Feature activation+1.40e-02

Pos logit contributions
Unfortunately+0.053
Meanwhile+0.053
eyed+0.053
rolling+0.051
 widening+0.051
Neg logit contributions
 is-0.038
 isn-0.035
 doesn-0.035
 happens-0.035
 guys-0.034
Token chest
Token ID7721
Feature activation+5.97e-02

Pos logit contributions
Unfortunately+0.027
Meanwhile+0.026
eyed+0.025
 widening+0.024
Finally+0.024
Neg logit contributions
 is-0.026
 doesn-0.025
 belongs-0.025
 happens-0.025
 guys-0.025
Token.
Token ID13
Feature activation+3.24e-02

Pos logit contributions
Unfortunately+0.170
 sidelines+0.169
Meanwhile+0.163
Eventually+0.162
Finally+0.161
Neg logit contributions
 is-0.058
 belongs-0.052
 doesn-0.051
 needs-0.047
 happens-0.046
Token "
Token ID366
Feature activation+5.91e-02

Pos logit contributions
eyed+0.073
Unfortunately+0.068
 pounding+0.068
Meanwhile+0.068
 widening+0.067
Neg logit contributions
 is-0.049
 isn-0.047
 happens-0.047
 doesn-0.045
 belongs-0.042
TokenHmm
Token ID44217
Feature activation-1.27e-01

Pos logit contributions
rolling+0.119
 sidelines+0.113
 pounding+0.111
 widening+0.110
 victorious+0.108
Neg logit contributions
 is-0.098
 happens-0.095
 can-0.093
 doesn-0.092
 likes-0.090
Token,
Token ID11
Feature activation-3.98e-02

Pos logit contributions
unn+0.215
 guys+0.207
 cares+0.197
ishes+0.193
cers+0.192
Neg logit contributions
Suddenly-0.171
 carefully-0.167
shadow-0.158
Just-0.156
 pounding-0.156
Token it
Token ID340
Feature activation-5.36e-02

Pos logit contributions
 happens+0.058
 belongs+0.056
unn+0.052
 guys+0.052
 needs+0.051
Neg logit contributions
Unfortunately-0.084
Meanwhile-0.079
Just-0.078
 pounding-0.077
However-0.077
Token sounds
Token ID5238
Feature activation-3.18e-02

Pos logit contributions
 guys+0.050
mm+0.042
 happens+0.040
unn+0.039
 guy+0.037
Neg logit contributions
Unfortunately-0.130
 pounding-0.129
Meanwhile-0.128
eyed-0.128
Just-0.127
Token good
Token ID922
Feature activation-6.78e-02

Pos logit contributions
 belongs+0.062
 happens+0.062
unn+0.061
 guys+0.059
 doesn+0.059
Neg logit contributions
Unfortunately-0.049
 widening-0.045
 pounding-0.045
eyed-0.045
Just-0.044
Token.
Token ID13
Feature activation-2.30e-02

Pos logit contributions
 isn+0.127
uffy+0.126
 belongs+0.125
mm+0.125
 doesn+0.124
Neg logit contributions
eyed-0.088
Just-0.083
 pounding-0.080
Meanwhile-0.078
 realizing-0.078
Token.
Token ID13
Feature activation-1.56e-02

Pos logit contributions
Unfortunately+0.044
Meanwhile+0.043
eyed+0.042
Just+0.041
 widening+0.041
Neg logit contributions
 is-0.024
 belongs-0.022
 happens-0.022
 doesn-0.022
 isn-0.021
Token He
Token ID679
Feature activation+6.02e-03

Pos logit contributions
 happens+0.031
 belongs+0.030
 doesn+0.029
 is+0.029
 guys+0.029
Neg logit contributions
Unfortunately-0.027
Meanwhile-0.026
Just-0.025
eyed-0.025
 widening-0.024
Token thought
Token ID1807
Feature activation-3.44e-02

Pos logit contributions
Unfortunately+0.013
eyed+0.013
Meanwhile+0.013
 widening+0.013
 sidelines+0.012
Neg logit contributions
 happens-0.009
 is-0.009
 doesn-0.008
 belongs-0.008
 isn-0.008
Token,
Token ID11
Feature activation+3.35e-02

Pos logit contributions
 belongs+0.062
 happens+0.060
 doesn+0.060
ishes+0.059
 isn+0.059
Neg logit contributions
Unfortunately-0.060
 pounding-0.058
Meanwhile-0.056
His-0.056
 crashing-0.055
Token "
Token ID366
Feature activation+1.98e-02

Pos logit contributions
eyed+0.071
har+0.070
Unfortunately+0.068
Meanwhile+0.067
 sidelines+0.066
Neg logit contributions
 is-0.056
 doesn-0.055
 isn-0.053
 happens-0.053
 guys-0.050
TokenHmm
Token ID44217
Feature activation-1.27e-01

Pos logit contributions
 sidelines+0.037
Unfortunately+0.036
 widening+0.036
Meanwhile+0.036
rolling+0.036
Neg logit contributions
 is-0.035
 doesn-0.034
 isn-0.034
 happens-0.034
 belongs-0.034
Token,
Token ID11
Feature activation-4.06e-02

Pos logit contributions
 cares+0.258
unn+0.251
ishes+0.246
 needs+0.243
 acts+0.242
Neg logit contributions
Suddenly-0.122
shadow-0.118
His-0.115
 firmly-0.110
 low-0.110
Token maybe
Token ID3863
Feature activation-7.24e-02

Pos logit contributions
 happens+0.079
 belongs+0.075
 needs+0.071
unn+0.071
 guys+0.069
Neg logit contributions
Unfortunately-0.068
Meanwhile-0.062
Just-0.062
However-0.062
 pounding-0.060
Token this
Token ID428
Feature activation-6.12e-02

Pos logit contributions
unn+0.115
 happens+0.114
 belongs+0.112
uffy+0.109
akes+0.108
Neg logit contributions
But-0.131
His-0.126
Suddenly-0.125
Just-0.124
shadow-0.121
Token hole
Token ID7604
Feature activation-1.01e-01

Pos logit contributions
unn+0.093
mm+0.092
ishes+0.088
 guys+0.079
uffy+0.075
Neg logit contributions
Unfortunately-0.119
 lower-0.113
Meanwhile-0.112
 ledge-0.112
On-0.111
Token will
Token ID481
Feature activation+2.69e-02

Pos logit contributions
 guys+0.075
mm+0.073
unn+0.070
 guy+0.056
 dreams+0.052
Neg logit contributions
eyed-0.236
 lower-0.236
 aside-0.234
Meanwhile-0.230
 victorious-0.229
Token behind
Token ID2157
Feature activation-6.28e-03

Pos logit contributions
 happens+0.013
 doesn+0.012
 is+0.012
 isn+0.012
 belongs+0.012
Neg logit contributions
Unfortunately-0.010
Meanwhile-0.009
eyed-0.009
 widening-0.009
Just-0.009
Token the
Token ID262
Feature activation-1.09e-02

Pos logit contributions
 happens+0.012
 belongs+0.012
 doesn+0.012
 isn+0.011
 is+0.011
Neg logit contributions
Unfortunately-0.011
Meanwhile-0.011
eyed-0.011
 widening-0.010
Just-0.010
Token couch
Token ID18507
Feature activation+4.50e-02

Pos logit contributions
 happens+0.020
 doesn+0.019
 belongs+0.019
 guys+0.019
 isn+0.019
Neg logit contributions
Unfortunately-0.020
Meanwhile-0.020
eyed-0.019
 widening-0.019
 pounding-0.019
Token.
Token ID13
Feature activation+2.84e-03

Pos logit contributions
Unfortunately+0.125
Meanwhile+0.120
Finally+0.119
Eventually+0.118
eyed+0.118
Neg logit contributions
 is-0.042
 guys-0.039
 doesn-0.036
 happens-0.036
 belongs-0.035
Token "
Token ID366
Feature activation+2.61e-02

Pos logit contributions
Unfortunately+0.006
Meanwhile+0.006
eyed+0.006
Just+0.005
 widening+0.005
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 isn-0.005
 belongs-0.005
TokenHere
Token ID4342
Feature activation-1.27e-01

Pos logit contributions
rolling+0.043
eyed+0.042
 widening+0.042
Unfortunately+0.041
 pounding+0.041
Neg logit contributions
 happens-0.050
 is-0.049
 doesn-0.048
 isn-0.048
 guys-0.047
Token I
Token ID314
Feature activation-4.73e-02

Pos logit contributions
unn+0.274
ishes+0.272
akes+0.270
ays+0.259
elly+0.256
Neg logit contributions
 free-0.136
 pounding-0.119
Her-0.119
 cheer-0.115
 unt-0.113
Token am
Token ID716
Feature activation-3.61e-02

Pos logit contributions
 happens+0.094
 doesn+0.093
mm+0.090
 isn+0.088
 guys+0.088
Neg logit contributions
eyed-0.070
Unfortunately-0.065
On-0.065
His-0.064
Meanwhile-0.064
Token!"
Token ID2474
Feature activation-5.27e-03

Pos logit contributions
 happens+0.053
 doesn+0.050
 belongs+0.047
 isn+0.047
 is+0.046
Neg logit contributions
Unfortunately-0.079
eyed-0.078
Meanwhile-0.077
Just-0.076
rolling-0.076
Token she
Token ID673
Feature activation+7.42e-03

Pos logit contributions
 happens+0.008
 is+0.007
 belongs+0.007
 doesn+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.012
Meanwhile-0.011
eyed-0.011
Just-0.011
 widening-0.011
Token shouted
Token ID17293
Feature activation+2.47e-02

Pos logit contributions
Unfortunately+0.016
Meanwhile+0.015
eyed+0.015
 widening+0.015
Just+0.014
Neg logit contributions
 happens-0.012
 is-0.012
 doesn-0.011
 belongs-0.011
 isn-0.011
Token Lily
Token ID20037
Feature activation-4.51e-02

Pos logit contributions
 happens+0.007
 is+0.007
 doesn+0.006
 belongs+0.006
 isn+0.006
Neg logit contributions
Unfortunately-0.006
Meanwhile-0.006
eyed-0.006
Just-0.005
 pounding-0.005
Token's
Token ID338
Feature activation-3.98e-03

Pos logit contributions
 happens+0.076
 guys+0.072
 isn+0.067
 belongs+0.067
 doesn+0.065
Neg logit contributions
eyed-0.086
Unfortunately-0.082
Just-0.081
 pounding-0.080
Meanwhile-0.080
Token dress
Token ID6576
Feature activation-3.16e-02

Pos logit contributions
 happens+0.008
 is+0.007
 belongs+0.007
 doesn+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.007
Meanwhile-0.007
eyed-0.007
Just-0.006
 pounding-0.006
Token.
Token ID13
Feature activation-4.82e-02

Pos logit contributions
 happens+0.054
 guys+0.050
 belongs+0.050
 isn+0.049
 doesn+0.049
Neg logit contributions
Unfortunately-0.059
eyed-0.058
Meanwhile-0.057
Just-0.057
 pounding-0.056
Token That
Token ID1320
Feature activation-5.30e-02

Pos logit contributions
 guys+0.085
 happens+0.082
 belongs+0.078
 happened+0.077
 is+0.074
Neg logit contributions
Unfortunately-0.091
eyed-0.087
Just-0.086
Meanwhile-0.085
Her-0.080
Token boy
Token ID2933
Feature activation-1.26e-01

Pos logit contributions
unn+0.081
mm+0.079
 guys+0.079
ishes+0.077
 owns+0.071
Neg logit contributions
Unfortunately-0.094
Meanwhile-0.094
 pounding-0.092
 lower-0.091
 distance-0.088
Token was
Token ID373
Feature activation-2.49e-02

Pos logit contributions
 guys+0.174
 mushrooms+0.145
mm+0.140
unn+0.136
berries+0.136
Neg logit contributions
eyed-0.240
 unt-0.238
 determined-0.237
 free-0.233
 proud-0.231
Token very
Token ID845
Feature activation-3.15e-02

Pos logit contributions
 happens+0.055
 guys+0.054
unn+0.054
 belongs+0.052
 isn+0.051
Neg logit contributions
Unfortunately-0.035
Meanwhile-0.033
Just-0.032
eyed-0.032
 pounding-0.032
Token bad
Token ID2089
Feature activation-8.04e-02

Pos logit contributions
 guys+0.069
 happens+0.066
 isn+0.064
 doesn+0.063
 is+0.062
Neg logit contributions
eyed-0.046
Unfortunately-0.045
His-0.044
Just-0.044
Meanwhile-0.042
Token and
Token ID290
Feature activation-5.03e-02

Pos logit contributions
 belongs+0.169
ishes+0.161
 owns+0.157
uffy+0.156
 lives+0.154
Neg logit contributions
But-0.098
Just-0.096
 realizing-0.090
 faster-0.090
 louder-0.090
Token he
Token ID339
Feature activation-8.19e-02

Pos logit contributions
 guys+0.076
 happens+0.070
unn+0.066
 isn+0.066
 belongs+0.065
Neg logit contributions
Unfortunately-0.098
eyed-0.097
Meanwhile-0.095
 pounding-0.094
Just-0.091
Token
Token ID198
Feature activation+2.30e-02

Pos logit contributions
Unfortunately+0.036
Meanwhile+0.036
 widening+0.036
 sidelines+0.036
eyed+0.035
Neg logit contributions
 is-0.037
 happens-0.035
 isn-0.034
 belongs-0.033
 doesn-0.033
Token"
Token ID1
Feature activation+7.70e-03

Pos logit contributions
Unfortunately+0.040
eyed+0.039
 pounding+0.039
 sidelines+0.039
Meanwhile+0.038
Neg logit contributions
 is-0.043
 happens-0.041
 belongs-0.040
 isn-0.040
 doesn-0.039
TokenLet
Token ID5756
Feature activation-4.68e-02

Pos logit contributions
Unfortunately+0.012
Meanwhile+0.012
eyed+0.012
 pounding+0.012
 widening+0.011
Neg logit contributions
 happens-0.016
 is-0.016
 isn-0.015
 doesn-0.015
 guys-0.015
Token's
Token ID338
Feature activation+5.61e-03

Pos logit contributions
unn+0.080
 needs+0.078
 cares+0.076
 dinosaurs+0.075
 happens+0.074
Neg logit contributions
Unfortunately-0.083
Meanwhile-0.081
Just-0.079
Her-0.078
eyed-0.078
Token see
Token ID766
Feature activation-5.41e-02

Pos logit contributions
Unfortunately+0.010
Meanwhile+0.010
eyed+0.009
Just+0.009
 widening+0.009
Neg logit contributions
 happens-0.011
 is-0.011
 guys-0.011
 doesn-0.011
 belongs-0.011
Token what
Token ID644
Feature activation-1.26e-01

Pos logit contributions
 happens+0.096
 needs+0.093
 belongs+0.093
 guys+0.090
unn+0.089
Neg logit contributions
Unfortunately-0.088
eyed-0.087
shadow-0.087
Suddenly-0.084
On-0.083
Token is
Token ID318
Feature activation-3.28e-02

Pos logit contributions
unn+0.230
 guys+0.184
affles+0.184
ishes+0.183
akes+0.181
Neg logit contributions
The-0.213
On-0.198
 nearby-0.190
 low-0.187
Unfortunately-0.185
Token in
Token ID287
Feature activation-4.08e-02

Pos logit contributions
 happens+0.066
 doesn+0.066
 isn+0.065
 needs+0.064
 is+0.063
Neg logit contributions
Unfortunately-0.048
eyed-0.047
Meanwhile-0.045
Just-0.044
 pounding-0.043
Token the
Token ID262
Feature activation-1.68e-02

Pos logit contributions
 happens+0.080
 guys+0.078
unn+0.078
 isn+0.077
 doesn+0.077
Neg logit contributions
Unfortunately-0.063
eyed-0.061
Meanwhile-0.061
On-0.060
 victorious-0.057
Token sack
Token ID23704
Feature activation-3.50e-02

Pos logit contributions
 happens+0.030
 doesn+0.027
 isn+0.027
 belongs+0.027
 is+0.027
Neg logit contributions
Unfortunately-0.033
Meanwhile-0.032
eyed-0.031
 sidelines-0.030
 widening-0.030
Token,"
Token ID553
Feature activation-1.97e-02

Pos logit contributions
 happens+0.040
 isn+0.037
 doesn+0.037
 belongs+0.035
 guys+0.035
Neg logit contributions
Unfortunately-0.086
Meanwhile-0.085
eyed-0.084
 widening-0.081
On-0.080
Token her
Token ID607
Feature activation-2.35e-02

Pos logit contributions
 belongs+0.089
unn+0.087
 needs+0.081
 lives+0.079
ishes+0.079
Neg logit contributions
Unfortunately-0.092
Her-0.089
Meanwhile-0.089
Suddenly-0.087
However-0.087
Token mom
Token ID1995
Feature activation-1.66e-02

Pos logit contributions
 belongs+0.043
 happens+0.042
 doesn+0.042
 isn+0.041
 is+0.040
Neg logit contributions
Meanwhile-0.043
Unfortunately-0.042
 pounding-0.042
eyed-0.041
Just-0.040
Token said
Token ID531
Feature activation-1.51e-02

Pos logit contributions
 happens+0.028
 belongs+0.027
 doesn+0.027
 isn+0.027
 is+0.026
Neg logit contributions
Meanwhile-0.032
Unfortunately-0.032
eyed-0.031
 pounding-0.031
Just-0.030
Token "
Token ID366
Feature activation+3.50e-02

Pos logit contributions
 belongs+0.028
 happens+0.028
 doesn+0.028
 guys+0.027
 is+0.027
Neg logit contributions
Unfortunately-0.026
Meanwhile-0.026
eyed-0.025
Just-0.025
 pounding-0.025
TokenThat
Token ID2504
Feature activation-5.43e-02

Pos logit contributions
rolling+0.061
 victorious+0.061
Unfortunately+0.059
eyed+0.059
 pounding+0.059
Neg logit contributions
 doesn-0.065
 happens-0.064
 is-0.064
 isn-0.063
 guys-0.059
Token glove
Token ID29144
Feature activation-1.26e-01

Pos logit contributions
mm+0.086
unn+0.085
ishes+0.081
 belongs+0.078
 happens+0.077
Neg logit contributions
Meanwhile-0.097
Unfortunately-0.097
 pounding-0.095
Just-0.091
 ledge-0.089
Token has
Token ID468
Feature activation-3.48e-02

Pos logit contributions
unn+0.108
 guys+0.095
 happens+0.093
mm+0.091
 dreams+0.086
Neg logit contributions
 pounding-0.280
 free-0.279
 barely-0.273
Unfortunately-0.267
 lower-0.266
Token been
Token ID587
Feature activation-2.59e-02

Pos logit contributions
 happens+0.062
 belongs+0.061
 doesn+0.060
 is+0.059
 isn+0.059
Neg logit contributions
Unfortunately-0.060
Meanwhile-0.059
 pounding-0.057
Just-0.056
 widening-0.055
Token sitting
Token ID5586
Feature activation-2.92e-02

Pos logit contributions
 happens+0.051
 belongs+0.049
 doesn+0.049
 is+0.049
 isn+0.048
Neg logit contributions
Unfortunately-0.042
Meanwhile-0.041
 pounding-0.040
Just-0.040
eyed-0.039
Token here
Token ID994
Feature activation-2.23e-02

Pos logit contributions
 happens+0.059
 belongs+0.056
 doesn+0.055
 isn+0.054
 is+0.053
Neg logit contributions
Unfortunately-0.048
Meanwhile-0.046
eyed-0.046
Just-0.045
 victorious-0.042
Token for
Token ID329
Feature activation+1.06e-02

Pos logit contributions
 happens+0.035
 belongs+0.033
 doesn+0.033
 is+0.032
 guys+0.032
Neg logit contributions
Unfortunately-0.046
Meanwhile-0.045
eyed-0.045
Just-0.043
 widening-0.043
Token him
Token ID683
Feature activation+2.44e-02

Pos logit contributions
 happens+0.019
 belongs+0.019
 doesn+0.018
 isn+0.018
 is+0.018
Neg logit contributions
Unfortunately-0.022
Meanwhile-0.021
eyed-0.021
 pounding-0.020
Just-0.020
Token feel
Token ID1254
Feature activation+6.14e-03

Pos logit contributions
Unfortunately+0.045
Meanwhile+0.045
eyed+0.042
Just+0.042
 widening+0.041
Neg logit contributions
 is-0.046
 doesn-0.045
 happens-0.045
 guys-0.044
 belongs-0.044
Token better
Token ID1365
Feature activation+1.56e-03

Pos logit contributions
Unfortunately+0.011
Meanwhile+0.011
eyed+0.011
Just+0.010
 widening+0.010
Neg logit contributions
 happens-0.012
 is-0.011
 doesn-0.011
 isn-0.011
 belongs-0.011
Token.
Token ID13
Feature activation-5.18e-03

Pos logit contributions
Unfortunately+0.004
Meanwhile+0.003
eyed+0.003
 widening+0.003
Just+0.003
Neg logit contributions
 happens-0.002
 is-0.002
 doesn-0.002
 belongs-0.002
 isn-0.002
Token "
Token ID366
Feature activation+3.60e-02

Pos logit contributions
 happens+0.008
 is+0.008
 belongs+0.008
 doesn+0.008
 isn+0.008
Neg logit contributions
Unfortunately-0.011
Meanwhile-0.011
eyed-0.011
Just-0.010
 widening-0.010
TokenHere
Token ID4342
Feature activation-1.26e-01

Pos logit contributions
rolling+0.068
 pounding+0.063
 widening+0.063
eyed+0.061
 victorious+0.061
Neg logit contributions
 doesn-0.064
 happens-0.063
 is-0.063
 isn-0.061
 needs-0.060
Token,
Token ID11
Feature activation-4.39e-02

Pos logit contributions
elly+0.217
ays+0.217
oline+0.214
unn+0.206
akes+0.204
Neg logit contributions
His-0.165
 determined-0.156
 free-0.156
But-0.152
 pounding-0.149
Token take
Token ID1011
Feature activation-3.75e-02

Pos logit contributions
 happens+0.073
 belongs+0.065
 guys+0.065
ishes+0.063
unn+0.063
Neg logit contributions
Unfortunately-0.087
Meanwhile-0.082
His-0.080
However-0.079
On-0.077
Token this
Token ID428
Feature activation-5.94e-03

Pos logit contributions
 belongs+0.065
 happens+0.060
 doesn+0.060
 isn+0.059
 needs+0.059
Neg logit contributions
Unfortunately-0.073
Meanwhile-0.068
eyed-0.066
Just-0.066
Eventually-0.064
Token medicine
Token ID9007
Feature activation-3.03e-02

Pos logit contributions
 happens+0.010
 is+0.009
 belongs+0.009
 doesn+0.009
 isn+0.009
Neg logit contributions
Unfortunately-0.012
Meanwhile-0.012
eyed-0.012
 pounding-0.011
Just-0.011
Token.
Token ID13
Feature activation-3.44e-02

Pos logit contributions
 happens+0.042
 isn+0.040
 belongs+0.039
 doesn+0.039
 guys+0.039
Neg logit contributions
Unfortunately-0.067
Meanwhile-0.066
eyed-0.065
Just-0.063
 widening-0.062
Token.
Token ID13
Feature activation-4.57e-03

Pos logit contributions
Unfortunately+0.079
Meanwhile+0.076
eyed+0.075
Finally+0.074
rolling+0.074
Neg logit contributions
 is-0.037
 doesn-0.033
 belongs-0.032
 happens-0.032
 isn-0.031
Token "
Token ID366
Feature activation+3.12e-02

Pos logit contributions
 happens+0.007
 is+0.007
 doesn+0.007
 belongs+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.010
Meanwhile-0.009
eyed-0.009
Just-0.009
 widening-0.009
TokenLet
Token ID5756
Feature activation-4.54e-02

Pos logit contributions
 pounding+0.053
rolling+0.053
 widening+0.052
eyed+0.052
 sidelines+0.052
Neg logit contributions
 guys-0.056
 is-0.056
 happens-0.055
 isn-0.055
 doesn-0.054
Token's
Token ID338
Feature activation+1.09e-03

Pos logit contributions
 cares+0.075
unn+0.073
 needs+0.073
 happens+0.072
 belongs+0.070
Neg logit contributions
Unfortunately-0.084
Meanwhile-0.079
Just-0.079
However-0.076
Finally-0.074
Token see
Token ID766
Feature activation-4.72e-02

Pos logit contributions
Unfortunately+0.002
Meanwhile+0.002
eyed+0.002
Just+0.002
 widening+0.002
Neg logit contributions
 happens-0.002
 is-0.002
 doesn-0.002
 belongs-0.002
 isn-0.002
Token what
Token ID644
Feature activation-1.26e-01

Pos logit contributions
 happens+0.096
 belongs+0.095
 needs+0.091
 cares+0.088
 doesn+0.088
Neg logit contributions
Unfortunately-0.070
shadow-0.069
eyed-0.067
Suddenly-0.066
 crashing-0.066
Token's
Token ID338
Feature activation-3.17e-02

Pos logit contributions
unn+0.229
mm+0.224
ishes+0.223
ems+0.221
unes+0.220
Neg logit contributions
The-0.177
Unfortunately-0.149
 free-0.148
 slowly-0.146
 low-0.145
Token inside
Token ID2641
Feature activation-6.73e-02

Pos logit contributions
 isn+0.065
 doesn+0.064
 guys+0.064
 happens+0.064
unn+0.063
Neg logit contributions
Unfortunately-0.044
Meanwhile-0.042
eyed-0.042
 pounding-0.042
 widening-0.040
Token!"
Token ID2474
Feature activation+5.06e-03

Pos logit contributions
 happens+0.105
 cares+0.099
 guys+0.099
unn+0.095
 belongs+0.094
Neg logit contributions
Unfortunately-0.123
Meanwhile-0.119
 lower-0.118
 victorious-0.116
eyed-0.116
Token said
Token ID531
Feature activation-2.41e-04

Pos logit contributions
Unfortunately+0.010
Meanwhile+0.010
eyed+0.010
 widening+0.010
Just+0.010
Neg logit contributions
 happens-0.008
 is-0.008
 doesn-0.008
 belongs-0.008
 isn-0.008
Token Bun
Token ID28515
Feature activation+2.14e-03

Pos logit contributions
 happens+0.000
 is+0.000
 doesn+0.000
 belongs+0.000
 isn+0.000
Neg logit contributions
Unfortunately-0.000
Meanwhile-0.000
eyed-0.000
Just-0.000
 widening-0.000
Token Alice
Token ID14862
Feature activation-5.22e-02

Pos logit contributions
 happens+0.116
 belongs+0.110
 needs+0.108
 doesn+0.105
 isn+0.105
Neg logit contributions
 pounding-0.099
eyed-0.098
Her-0.098
Just-0.097
 widening-0.094
Token.
Token ID13
Feature activation-2.46e-02

Pos logit contributions
 happens+0.119
 guys+0.116
unn+0.109
 belongs+0.106
 doesn+0.105
Neg logit contributions
eyed-0.077
Just-0.070
Her-0.069
His-0.068
But-0.067
Token Every
Token ID3887
Feature activation-4.18e-02

Pos logit contributions
 happens+0.051
 doesn+0.050
 belongs+0.050
 guys+0.049
 isn+0.049
Neg logit contributions
Unfortunately-0.039
Meanwhile-0.038
eyed-0.037
 widening-0.037
On-0.036
Token day
Token ID1110
Feature activation-8.54e-02

Pos logit contributions
 doesn+0.075
 isn+0.074
ishes+0.073
 belongs+0.072
 happens+0.070
Neg logit contributions
Meanwhile-0.069
eyed-0.069
Unfortunately-0.068
Just-0.067
On-0.067
Token,
Token ID11
Feature activation-4.99e-02

Pos logit contributions
ishes+0.187
 belongs+0.172
iles+0.171
unn+0.169
 doesn+0.165
Neg logit contributions
Her-0.120
On-0.113
 low-0.112
side-0.112
But-0.111
Token Princess
Token ID8449
Feature activation-1.25e-01

Pos logit contributions
 belongs+0.111
 happens+0.109
unn+0.107
 doesn+0.106
 needs+0.105
Neg logit contributions
Unfortunately-0.069
On-0.068
Her-0.068
Meanwhile-0.066
However-0.064
Token Alice
Token ID14862
Feature activation-9.22e-02

Pos logit contributions
akes+0.254
icer+0.238
mm+0.235
unn+0.233
 aren+0.231
Neg logit contributions
Her-0.162
On-0.150
 carefully-0.146
eyed-0.145
 proudly-0.144
Token liked
Token ID8288
Feature activation-2.89e-02

Pos logit contributions
mm+0.170
 isn+0.161
 guys+0.158
 happens+0.156
 belongs+0.150
Neg logit contributions
eyed-0.152
On-0.136
Just-0.134
 pounding-0.134
But-0.129
Token to
Token ID284
Feature activation+1.00e-02

Pos logit contributions
 belongs+0.066
 happens+0.065
 doesn+0.064
 isn+0.063
 needs+0.063
Neg logit contributions
Unfortunately-0.037
Meanwhile-0.037
 pounding-0.037
eyed-0.036
Just-0.035
Token look
Token ID804
Feature activation-2.96e-02

Pos logit contributions
Unfortunately+0.018
Meanwhile+0.018
eyed+0.017
Just+0.017
 widening+0.016
Neg logit contributions
 happens-0.019
 is-0.019
 guys-0.019
 belongs-0.018
 doesn-0.018
Token nice
Token ID3621
Feature activation-4.97e-02

Pos logit contributions
 happens+0.057
 belongs+0.053
 doesn+0.053
 isn+0.053
 is+0.052
Neg logit contributions
eyed-0.050
Unfortunately-0.048
 int-0.048
Meanwhile-0.047
 widening-0.046
Token looked
Token ID3114
Feature activation-1.37e-03

Pos logit contributions
Unfortunately+0.032
Meanwhile+0.031
eyed+0.030
 sidelines+0.030
 widening+0.029
Neg logit contributions
 is-0.022
 happens-0.022
 doesn-0.022
 belongs-0.021
 isn-0.021
Token and
Token ID290
Feature activation+9.34e-03

Pos logit contributions
 happens+0.002
 is+0.002
 doesn+0.002
 belongs+0.002
 isn+0.002
Neg logit contributions
Unfortunately-0.003
Meanwhile-0.003
eyed-0.002
Just-0.002
 widening-0.002
Token said
Token ID531
Feature activation-3.14e-02

Pos logit contributions
Unfortunately+0.021
eyed+0.021
Meanwhile+0.020
 pounding+0.019
Just+0.019
Neg logit contributions
 happens-0.014
 is-0.014
 doesn-0.013
 isn-0.013
 belongs-0.013
Token,
Token ID11
Feature activation+4.61e-02

Pos logit contributions
 belongs+0.041
 needs+0.040
 doesn+0.040
 happens+0.040
 guys+0.040
Neg logit contributions
Unfortunately-0.069
His-0.069
eyed-0.068
 pounding-0.067
Just-0.067
Token "
Token ID366
Feature activation+1.84e-02

Pos logit contributions
eyed+0.112
Unfortunately+0.111
har+0.109
Meanwhile+0.108
 sidelines+0.104
Neg logit contributions
 is-0.062
 isn-0.059
 doesn-0.059
 happens-0.055
 can-0.055
TokenHmm
Token ID44217
Feature activation-1.24e-01

Pos logit contributions
Unfortunately+0.029
 pounding+0.029
 widening+0.028
eyed+0.028
Meanwhile+0.028
Neg logit contributions
 is-0.037
 happens-0.037
 isn-0.037
 doesn-0.036
 belongs-0.036
Token,
Token ID11
Feature activation-4.43e-02

Pos logit contributions
unn+0.271
ishes+0.248
 cares+0.245
 lives+0.234
eer+0.234
Neg logit contributions
 ï¿½-0.120
Suddenly-0.117
 carefully-0.115
 revealing-0.115
 pounding-0.110
Token it
Token ID340
Feature activation-6.79e-02

Pos logit contributions
unn+0.081
 happens+0.078
 belongs+0.078
ishes+0.074
 needs+0.073
Neg logit contributions
Unfortunately-0.077
Meanwhile-0.070
However-0.070
Just-0.069
 pounding-0.067
Token seems
Token ID2331
Feature activation-1.13e-02

Pos logit contributions
mm+0.100
unn+0.099
 guys+0.098
 guy+0.087
ens+0.085
Neg logit contributions
Unfortunately-0.122
 pounding-0.121
Meanwhile-0.119
Just-0.118
 lower-0.117
Token like
Token ID588
Feature activation-3.37e-02

Pos logit contributions
 happens+0.022
 belongs+0.021
 is+0.021
 doesn+0.021
 guys+0.021
Neg logit contributions
Unfortunately-0.019
Meanwhile-0.019
eyed-0.019
 pounding-0.018
Just-0.018
Token a
Token ID257
Feature activation+5.10e-03

Pos logit contributions
 belongs+0.063
 happens+0.062
 doesn+0.061
 is+0.061
 isn+0.060
Neg logit contributions
Unfortunately-0.057
Meanwhile-0.054
 pounding-0.052
Just-0.051
However-0.050
Token Benny
Token ID44275
Feature activation+2.08e-02

Pos logit contributions
 belongs+0.093
 happens+0.090
 isn+0.089
 guys+0.088
 needs+0.088
Neg logit contributions
Meanwhile-0.067
His-0.065
Unfortunately-0.065
Her-0.064
Their-0.061
Token and
Token ID290
Feature activation-1.94e-02

Pos logit contributions
Unfortunately+0.044
Meanwhile+0.042
eyed+0.041
Finally+0.041
Just+0.041
Neg logit contributions
 is-0.035
 doesn-0.033
 belongs-0.033
 needs-0.032
 happens-0.032
Token asked
Token ID1965
Feature activation-5.10e-02

Pos logit contributions
 happens+0.036
 guys+0.036
 belongs+0.035
 isn+0.035
 is+0.035
Neg logit contributions
Meanwhile-0.034
Unfortunately-0.033
eyed-0.032
 pounding-0.032
Just-0.032
Token,
Token ID11
Feature activation+3.75e-02

Pos logit contributions
 happens+0.094
 belongs+0.093
 needs+0.088
 doesn+0.086
 cares+0.085
Neg logit contributions
Her-0.085
His-0.085
 realizing-0.080
 vigorously-0.080
Unfortunately-0.080
Token "
Token ID366
Feature activation+2.70e-03

Pos logit contributions
Unfortunately+0.085
eyed+0.083
Meanwhile+0.082
 sidelines+0.079
har+0.079
Neg logit contributions
 is-0.055
 isn-0.052
 can-0.051
 doesn-0.051
 happens-0.051
TokenWhat
Token ID2061
Feature activation-1.24e-01

Pos logit contributions
Unfortunately+0.004
eyed+0.003
Meanwhile+0.003
 widening+0.003
 pounding+0.003
Neg logit contributions
 happens-0.006
 is-0.006
 doesn-0.006
 belongs-0.006
 isn-0.006
Token's
Token ID338
Feature activation-3.10e-02

Pos logit contributions
unn+0.228
ishes+0.223
mm+0.210
ley+0.204
ebra+0.203
Neg logit contributions
 overcome-0.170
 understand-0.167
 unt-0.147
On-0.143
The-0.142
Token wrong
Token ID2642
Feature activation-6.72e-02

Pos logit contributions
 isn+0.056
 happens+0.055
unn+0.054
 doesn+0.054
 guys+0.053
Neg logit contributions
Unfortunately-0.052
eyed-0.050
Meanwhile-0.049
Just-0.049
His-0.049
Token,
Token ID11
Feature activation-3.84e-02

Pos logit contributions
 belongs+0.107
 happens+0.105
 cares+0.103
 doesn+0.097
 lives+0.096
Neg logit contributions
 pounding-0.124
His-0.116
Unfortunately-0.114
rolling-0.114
Meanwhile-0.113
Token Benny
Token ID44275
Feature activation-8.47e-02

Pos logit contributions
 happens+0.064
 belongs+0.062
 lives+0.057
 cares+0.055
 needs+0.055
Neg logit contributions
Unfortunately-0.076
Meanwhile-0.074
However-0.072
His-0.070
 crashing-0.070
Token?"
Token ID1701
Feature activation-1.29e-02

Pos logit contributions
mm+0.108
 belongs+0.095
 happens+0.094
 cares+0.094
 isn+0.094
Neg logit contributions
 nearby-0.171
eyed-0.169
 pounding-0.169
On-0.168
broken-0.167
Token finds
Token ID7228
Feature activation+1.11e-02

Pos logit contributions
 happens+0.011
 belongs+0.011
 is+0.011
 isn+0.011
 doesn+0.011
Neg logit contributions
Unfortunately-0.015
Meanwhile-0.015
eyed-0.014
 widening-0.014
 pounding-0.014
Token the
Token ID262
Feature activation+1.35e-03

Pos logit contributions
Unfortunately+0.023
Meanwhile+0.022
eyed+0.022
Just+0.022
rolling+0.021
Neg logit contributions
 happens-0.019
 is-0.018
 doesn-0.017
 belongs-0.017
 isn-0.017
Token car
Token ID1097
Feature activation+6.61e-03

Pos logit contributions
Unfortunately+0.003
Meanwhile+0.003
eyed+0.003
Just+0.003
 widening+0.003
Neg logit contributions
 happens-0.002
 is-0.002
 doesn-0.002
 belongs-0.002
 isn-0.002
Token.
Token ID13
Feature activation+5.66e-03

Pos logit contributions
Unfortunately+0.015
Meanwhile+0.015
eyed+0.014
Just+0.014
 widening+0.014
Neg logit contributions
 is-0.010
 happens-0.009
 belongs-0.009
 doesn-0.009
 isn-0.009
Token "
Token ID366
Feature activation+3.21e-02

Pos logit contributions
Unfortunately+0.012
Meanwhile+0.012
eyed+0.012
 pounding+0.011
Just+0.011
Neg logit contributions
 happens-0.009
 is-0.009
 doesn-0.009
 isn-0.009
 belongs-0.009
TokenHere
Token ID4342
Feature activation-1.24e-01

Pos logit contributions
rolling+0.058
 pounding+0.057
Meanwhile+0.054
 sidelines+0.053
ooting+0.053
Neg logit contributions
 happens-0.061
 is-0.061
 doesn-0.057
 belongs-0.057
 guys-0.057
Token it
Token ID340
Feature activation-7.06e-02

Pos logit contributions
unn+0.265
akes+0.258
elly+0.248
ancer+0.244
 guys+0.242
Neg logit contributions
 free-0.131
Her-0.127
But-0.122
The-0.119
 unt-0.114
Token is
Token ID318
Feature activation-2.81e-02

Pos logit contributions
 guys+0.054
unn+0.049
 cares+0.044
 happens+0.039
ishes+0.039
Neg logit contributions
 aside-0.173
 lower-0.172
 carefully-0.168
eyed-0.168
Just-0.168
Token!
Token ID0
Feature activation-4.28e-02

Pos logit contributions
 happens+0.040
 doesn+0.037
 guys+0.037
 isn+0.037
unn+0.036
Neg logit contributions
Unfortunately-0.059
Just-0.057
eyed-0.057
Meanwhile-0.056
 widening-0.056
Token It
Token ID632
Feature activation-6.07e-02

Pos logit contributions
 guys+0.069
 happens+0.068
 happened+0.064
 belongs+0.064
 doesn+0.063
Neg logit contributions
Unfortunately-0.084
eyed-0.079
Just-0.078
His-0.077
Meanwhile-0.076
Token is
Token ID318
Feature activation-1.23e-03

Pos logit contributions
 guys+0.052
mm+0.049
unn+0.046
 happens+0.043
 happened+0.041
Neg logit contributions
Unfortunately-0.145
 lower-0.144
 unt-0.142
Meanwhile-0.141
 victorious-0.141
Token"
Token ID1
Feature activation+1.07e-02

Pos logit contributions
 pounding+0.066
rolling+0.065
 widening+0.064
Unfortunately+0.063
 sidelines+0.062
Neg logit contributions
 is-0.052
 happens-0.046
 belongs-0.045
?"-0.043
 lives-0.043
TokenHello
Token ID15496
Feature activation-1.02e-01

Pos logit contributions
Unfortunately+0.019
eyed+0.018
rolling+0.018
 pounding+0.018
Meanwhile+0.018
Neg logit contributions
 happens-0.021
 is-0.021
 doesn-0.020
 isn-0.020
 belongs-0.020
Token,
Token ID11
Feature activation-3.66e-02

Pos logit contributions
 belongs+0.219
 doesn+0.208
 needs+0.203
 vitamins+0.202
 happens+0.199
Neg logit contributions
Suddenly-0.114
But-0.112
 closer-0.111
On-0.109
 slowly-0.107
Token children
Token ID1751
Feature activation-5.05e-02

Pos logit contributions
 happens+0.055
 belongs+0.052
 doesn+0.051
 guys+0.050
 isn+0.047
Neg logit contributions
Meanwhile-0.076
Unfortunately-0.075
 crashing-0.072
eyed-0.070
 sidelines-0.070
Token.
Token ID13
Feature activation-7.90e-02

Pos logit contributions
 happens+0.104
 guys+0.098
 doesn+0.094
 isn+0.092
 belongs+0.092
Neg logit contributions
eyed-0.087
Unfortunately-0.076
har-0.076
 widening-0.075
Just-0.075
Token Who
Token ID5338
Feature activation-1.24e-01

Pos logit contributions
 happens+0.135
 guys+0.134
ishes+0.127
 doesn+0.124
 belongs+0.122
Neg logit contributions
eyed-0.147
Unfortunately-0.134
Just-0.132
Meanwhile-0.128
But-0.128
Token are
Token ID389
Feature activation-5.26e-02

Pos logit contributions
unn+0.175
ummy+0.172
ishes+0.169
umm+0.165
 guys+0.161
Neg logit contributions
eyed-0.196
 unt-0.195
 overcome-0.187
On-0.185
The-0.179
Token you
Token ID345
Feature activation-2.75e-02

Pos logit contributions
 doesn+0.071
 happens+0.068
 isn+0.068
 belongs+0.064
 needs+0.063
Neg logit contributions
Unfortunately-0.116
eyed-0.115
Just-0.112
Her-0.112
Meanwhile-0.111
Token?
Token ID30
Feature activation-5.35e-02

Pos logit contributions
 happens+0.022
 doesn+0.020
 isn+0.020
 belongs+0.020
 guys+0.020
Neg logit contributions
eyed-0.077
Meanwhile-0.076
Just-0.076
Unfortunately-0.075
On-0.073
Token What
Token ID1867
Feature activation-1.06e-01

Pos logit contributions
 guys+0.083
 happens+0.081
 belongs+0.075
 doesn+0.075
 isn+0.071
Neg logit contributions
eyed-0.116
Unfortunately-0.110
Just-0.108
Meanwhile-0.107
His-0.106
Token do
Token ID466
Feature activation-6.81e-02

Pos logit contributions
ishes+0.193
unn+0.177
ish+0.162
ley+0.159
uffy+0.159
Neg logit contributions
 understand-0.168
On-0.168
The-0.159
His-0.157
 overcome-0.157
Token."
Token ID526
Feature activation+1.51e-02

Pos logit contributions
 happens+0.022
 belongs+0.020
 doesn+0.020
 guys+0.020
 is+0.020
Neg logit contributions
Unfortunately-0.049
Meanwhile-0.048
eyed-0.047
Just-0.046
 pounding-0.046
Token
Token ID198
Feature activation+4.62e-02

Pos logit contributions
Unfortunately+0.027
Meanwhile+0.027
eyed+0.027
 pounding+0.027
 widening+0.026
Neg logit contributions
 is-0.028
 happens-0.027
 doesn-0.027
 isn-0.027
 belongs-0.026
Token
Token ID198
Feature activation+5.54e-02

Pos logit contributions
 widening+0.101
 pounding+0.100
 sidelines+0.098
 crashing+0.097
 winding+0.096
Neg logit contributions
 is-0.068
 happens-0.062
 isn-0.059
?"-0.059
 doesn-0.057
TokenJ
Token ID41
Feature activation-3.34e-02

Pos logit contributions
 crashing+0.089
 widening+0.088
 pounding+0.085
 steering+0.084
 afloat+0.083
Neg logit contributions
 is-0.108
 happens-0.102
 isn-0.098
 doesn-0.096
ara-0.096
Tokenenny
Token ID11870
Feature activation+1.20e-02

Pos logit contributions
 belongs+0.059
 happens+0.059
 guys+0.059
 lives+0.058
 needs+0.058
Neg logit contributions
Unfortunately-0.061
eyed-0.061
Meanwhile-0.060
Just-0.059
 pounding-0.055
Token waited
Token ID13488
Feature activation+6.98e-02

Pos logit contributions
Unfortunately+0.026
Meanwhile+0.026
eyed+0.025
 widening+0.025
 sidelines+0.024
Neg logit contributions
 happens-0.018
 is-0.018
 doesn-0.017
 isn-0.017
 belongs-0.017
Token for
Token ID329
Feature activation+1.70e-02

Pos logit contributions
Unfortunately+0.178
har+0.174
eyed+0.174
Meanwhile+0.172
Days+0.165
Neg logit contributions
 is-0.083
 doesn-0.074
 isn-0.071
?"-0.069
 can-0.069
Token the
Token ID262
Feature activation+1.95e-03

Pos logit contributions
eyed+0.038
Unfortunately+0.037
Meanwhile+0.037
rolling+0.036
Just+0.036
Neg logit contributions
 happens-0.025
 is-0.024
 guys-0.024
 doesn-0.024
 isn-0.023
Token rain
Token ID6290
Feature activation-9.74e-04

Pos logit contributions
Unfortunately+0.004
Meanwhile+0.004
eyed+0.004
Just+0.004
 widening+0.004
Neg logit contributions
 happens-0.003
 is-0.003
 doesn-0.003
 belongs-0.003
 isn-0.003
Token to
Token ID284
Feature activation+4.91e-02

Pos logit contributions
 happens+0.002
 is+0.002
 doesn+0.002
 belongs+0.002
 isn+0.002
Neg logit contributions
Unfortunately-0.002
Meanwhile-0.002
eyed-0.002
Just-0.002
 widening-0.002
Token stop
Token ID2245
Feature activation+3.36e-02

Pos logit contributions
Meanwhile+0.100
Unfortunately+0.099
eyed+0.096
Just+0.091
His+0.091
Neg logit contributions
 is-0.087
 guys-0.081
 doesn-0.080
?"-0.080
 can-0.080
Token.
Token ID13
Feature activation-6.97e-03

Pos logit contributions
Unfortunately+0.073
Meanwhile+0.070
Just+0.068
Finally+0.068
 widening+0.067
Neg logit contributions
 is-0.044
 happens-0.044
 doesn-0.040
 belongs-0.040
 guys-0.039
Token They
Token ID1119
Feature activation+2.32e-02

Pos logit contributions
 happens+0.011
 belongs+0.011
 doesn+0.011
 is+0.011
 guys+0.011
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.014
Just-0.013
 widening-0.013
Token make
Token ID787
Feature activation+4.65e-02

Pos logit contributions
Unfortunately+0.045
Meanwhile+0.044
Just+0.043
eyed+0.041
 widening+0.041
Neg logit contributions
 is-0.044
 happens-0.043
 guys-0.042
 doesn-0.042
 belongs-0.041
Token a
Token ID257
Feature activation+7.82e-02

Pos logit contributions
Unfortunately+0.106
Meanwhile+0.103
eyed+0.103
 widening+0.101
Just+0.101
Neg logit contributions
 is-0.075
 happens-0.072
 guys-0.069
 doesn-0.067
 isn-0.064
Token big
Token ID1263
Feature activation+3.78e-02

Pos logit contributions
eyed+0.186
Just+0.177
Unfortunately+0.176
Meanwhile+0.175
but+0.170
Neg logit contributions
 is-0.125
 guys-0.115
 can-0.111
 lives-0.110
 needs-0.105
Token train
Token ID4512
Feature activation+6.57e-02

Pos logit contributions
eyed+0.080
Unfortunately+0.080
Meanwhile+0.079
Just+0.078
His+0.075
Neg logit contributions
 is-0.067
 guys-0.062
 isn-0.061
 doesn-0.061
 needs-0.059
Token.
Token ID13
Feature activation-3.71e-03

Pos logit contributions
Unfortunately+0.158
Meanwhile+0.154
Just+0.147
 widening+0.145
eyed+0.144
Neg logit contributions
 is-0.106
 guys-0.089
 belongs-0.087
 doesn-0.085
 can-0.084
Token They
Token ID1119
Feature activation+2.60e-02

Pos logit contributions
 happens+0.006
 is+0.006
 belongs+0.006
 doesn+0.006
 isn+0.006
Neg logit contributions
Unfortunately-0.008
Meanwhile-0.008
eyed-0.007
Just-0.007
 widening-0.007
Token put
Token ID1234
Feature activation+9.37e-03

Pos logit contributions
Unfortunately+0.053
Meanwhile+0.052
Just+0.050
eyed+0.048
 widening+0.048
Neg logit contributions
 is-0.047
 happens-0.045
 guys-0.045
 doesn-0.044
 can-0.044
Token a
Token ID257
Feature activation+6.57e-02

Pos logit contributions
Unfortunately+0.019
Meanwhile+0.018
eyed+0.018
Just+0.018
 widening+0.017
Neg logit contributions
 happens-0.016
 is-0.016
 doesn-0.016
 belongs-0.016
 isn-0.015
Token kettle
Token ID40231
Feature activation+4.39e-02

Pos logit contributions
eyed+0.161
Unfortunately+0.160
Meanwhile+0.155
Just+0.154
ivated+0.153
Neg logit contributions
 is-0.092
 can-0.087
 guys-0.083
 happens-0.079
 needs-0.076
Token so
Token ID523
Feature activation+1.45e-02

Pos logit contributions
 sidelines+0.126
Unfortunately+0.119
Just+0.118
Finally+0.117
Meanwhile+0.114
Neg logit contributions
 is-0.110
?"-0.104
 isn-0.097
 happens-0.094
 happened-0.094
Token soft
Token ID2705
Feature activation-1.21e-02

Pos logit contributions
Unfortunately+0.019
Meanwhile+0.019
Just+0.018
eyed+0.018
 sidelines+0.018
Neg logit contributions
 is-0.035
 happens-0.035
 isn-0.034
 belongs-0.033
 doesn-0.033
Token that
Token ID326
Feature activation-2.92e-02

Pos logit contributions
 happens+0.022
 belongs+0.022
 doesn+0.021
 is+0.021
 guys+0.021
Neg logit contributions
Unfortunately-0.022
Meanwhile-0.021
eyed-0.021
 widening-0.021
Just-0.021
Token she
Token ID673
Feature activation-1.14e-03

Pos logit contributions
 belongs+0.048
 happens+0.048
 guys+0.047
 doesn+0.046
 needs+0.046
Neg logit contributions
Unfortunately-0.055
Meanwhile-0.054
 pounding-0.053
eyed-0.051
 widening-0.051
Token could
Token ID714
Feature activation+5.50e-02

Pos logit contributions
 happens+0.002
 is+0.002
 doesn+0.002
 belongs+0.002
 isn+0.002
Neg logit contributions
Unfortunately-0.002
Meanwhile-0.002
eyed-0.002
Just-0.002
 widening-0.002
Token barely
Token ID8523
Feature activation+6.94e-02

Pos logit contributions
Meanwhile+0.112
Unfortunately+0.110
shadow+0.106
side+0.104
 sidelines+0.102
Neg logit contributions
 is-0.097
?"-0.093
 can-0.090
 needs-0.084
 doesn-0.084
Token hear
Token ID3285
Feature activation-7.31e-03

Pos logit contributions
Meanwhile+0.137
Unfortunately+0.135
rolling+0.124
Just+0.122
 pounding+0.122
Neg logit contributions
 can-0.117
 guys-0.116
 is-0.113
 happens-0.112
?"-0.111
Token it
Token ID340
Feature activation+5.84e-02

Pos logit contributions
 happens+0.013
 belongs+0.012
 is+0.012
 doesn+0.012
 isn+0.012
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.013
 pounding-0.013
 widening-0.013
Token,
Token ID11
Feature activation+5.68e-02

Pos logit contributions
Unfortunately+0.158
Meanwhile+0.156
Just+0.147
eyed+0.147
har+0.146
Neg logit contributions
 is-0.073
 doesn-0.063
 happens-0.062
 guys-0.061
 isn-0.061
Token but
Token ID475
Feature activation+3.20e-03

Pos logit contributions
eyed+0.126
Unfortunately+0.122
Meanwhile+0.121
Finally+0.119
rolling+0.119
Neg logit contributions
 doesn-0.091
 isn-0.089
 is-0.087
 can-0.078
 needs-0.075
Token it
Token ID340
Feature activation+8.49e-03

Pos logit contributions
Unfortunately+0.007
Meanwhile+0.007
eyed+0.007
Just+0.006
 widening+0.006
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 belongs-0.005
 isn-0.005
Token them
Token ID606
Feature activation+3.03e-02

Pos logit contributions
 happens+0.016
 belongs+0.015
 is+0.015
 doesn+0.015
 isn+0.015
Neg logit contributions
Unfortunately-0.017
Meanwhile-0.016
eyed-0.016
 widening-0.016
Just-0.016
Token.
Token ID13
Feature activation+2.60e-02

Pos logit contributions
Unfortunately+0.075
Meanwhile+0.075
eyed+0.072
Just+0.071
 sidelines+0.070
Neg logit contributions
 is-0.038
 doesn-0.036
 isn-0.035
 belongs-0.035
 happens-0.035
Token The
Token ID383
Feature activation-1.35e-03

Pos logit contributions
eyed+0.049
 pounding+0.048
Meanwhile+0.048
Unfortunately+0.048
 widening+0.047
Neg logit contributions
 is-0.045
 isn-0.045
 doesn-0.044
 happens-0.044
 belongs-0.042
Token lion
Token ID18744
Feature activation+1.47e-02

Pos logit contributions
 happens+0.002
 is+0.002
 doesn+0.002
 belongs+0.002
 isn+0.002
Neg logit contributions
Unfortunately-0.003
Meanwhile-0.003
eyed-0.003
Just-0.003
 widening-0.003
Token watched
Token ID7342
Feature activation+4.19e-02

Pos logit contributions
Unfortunately+0.033
eyed+0.032
Meanwhile+0.032
 widening+0.031
 sidelines+0.031
Neg logit contributions
 is-0.021
 happens-0.021
 doesn-0.020
 isn-0.020
 belongs-0.020
Token,
Token ID11
Feature activation+7.40e-02

Pos logit contributions
Unfortunately+0.096
Meanwhile+0.093
eyed+0.091
rolling+0.088
Just+0.088
Neg logit contributions
 guys-0.062
 is-0.060
 isn-0.058
 doesn-0.057
 happens-0.057
Token but
Token ID475
Feature activation+1.30e-02

Pos logit contributions
Unfortunately+0.157
eyed+0.157
rolling+0.153
Finally+0.151
har+0.148
Neg logit contributions
 is-0.117
 doesn-0.116
 isn-0.109
 can-0.106
 wants-0.103
Token he
Token ID339
Feature activation+9.81e-03

Pos logit contributions
Unfortunately+0.025
eyed+0.025
Meanwhile+0.024
 widening+0.024
Just+0.023
Neg logit contributions
 happens-0.023
 is-0.023
 doesn-0.022
 isn-0.022
 belongs-0.022
Token didn
Token ID1422
Feature activation+2.88e-02

Pos logit contributions
Unfortunately+0.022
eyed+0.022
Meanwhile+0.022
 widening+0.021
 sidelines+0.021
Neg logit contributions
 happens-0.014
 is-0.014
 doesn-0.014
 isn-0.013
 belongs-0.013
Token't
Token ID470
Feature activation+2.19e-02

Pos logit contributions
eyed+0.053
Meanwhile+0.052
Unfortunately+0.051
 widening+0.051
har+0.050
Neg logit contributions
 is-0.054
 happens-0.053
 doesn-0.050
 isn-0.049
 belongs-0.049
Token do
Token ID466
Feature activation+7.00e-03

Pos logit contributions
Unfortunately+0.045
Meanwhile+0.045
rolling+0.041
Finally+0.041
 widening+0.041
Neg logit contributions
 happens-0.039
 guys-0.038
 is-0.038
 doesn-0.036
 needs-0.036
Token.
Token ID13
Feature activation+3.20e-02

Pos logit contributions
Unfortunately+0.111
Meanwhile+0.108
Finally+0.107
Days+0.106
eyed+0.106
Neg logit contributions
 is-0.056
 isn-0.049
 doesn-0.049
 happens-0.049
 needs-0.047
Token Billy
Token ID15890
Feature activation+2.56e-02

Pos logit contributions
eyed+0.076
Meanwhile+0.075
Unfortunately+0.074
 pounding+0.072
 widening+0.072
Neg logit contributions
 isn-0.041
 doesn-0.041
 happens-0.040
 is-0.040
 guys-0.038
Token jumped
Token ID11687
Feature activation+5.57e-02

Pos logit contributions
Unfortunately+0.059
eyed+0.056
Meanwhile+0.056
 widening+0.056
 sidelines+0.055
Neg logit contributions
 is-0.037
 doesn-0.035
 happens-0.035
 isn-0.034
 belongs-0.033
Token from
Token ID422
Feature activation+1.97e-02

Pos logit contributions
Unfortunately+0.120
eyed+0.118
Meanwhile+0.118
Finally+0.116
Days+0.115
Neg logit contributions
 is-0.087
 happens-0.079
 isn-0.077
 can-0.076
 guys-0.076
Token his
Token ID465
Feature activation-4.42e-05

Pos logit contributions
Unfortunately+0.039
Meanwhile+0.038
eyed+0.037
Just+0.036
Finally+0.036
Neg logit contributions
 is-0.035
 happens-0.034
 belongs-0.033
 doesn-0.033
 needs-0.033
Token bed
Token ID3996
Feature activation+6.48e-02

Pos logit contributions
 happens+0.000
 is+0.000
 doesn+0.000
 belongs+0.000
 isn+0.000
Neg logit contributions
Unfortunately-0.000
Meanwhile-0.000
eyed-0.000
Just-0.000
 widening-0.000
Token and
Token ID290
Feature activation+3.94e-02

Pos logit contributions
eyed+0.171
Unfortunately+0.168
Meanwhile+0.162
Finally+0.161
shadow+0.161
Neg logit contributions
 is-0.071
 doesn-0.063
 happens-0.062
 isn-0.062
 guys-0.060
Token told
Token ID1297
Feature activation+2.51e-02

Pos logit contributions
Unfortunately+0.080
Finally+0.077
eyed+0.077
 widening+0.076
Meanwhile+0.076
Neg logit contributions
 happens-0.067
 is-0.066
 doesn-0.065
 isn-0.062
 belongs-0.061
Token his
Token ID465
Feature activation-1.82e-02

Pos logit contributions
Meanwhile+0.050
Unfortunately+0.049
Just+0.048
 widening+0.047
eyed+0.047
Neg logit contributions
 happens-0.046
 guys-0.045
 is-0.044
 isn-0.042
 doesn-0.042
Token friends
Token ID2460
Feature activation+9.51e-03

Pos logit contributions
 happens+0.026
 belongs+0.026
 doesn+0.026
 isn+0.025
 is+0.025
Neg logit contributions
Unfortunately-0.039
Meanwhile-0.039
eyed-0.038
 pounding-0.038
Just-0.037
Token to
Token ID284
Feature activation+5.48e-02

Pos logit contributions
Unfortunately+0.021
Meanwhile+0.020
eyed+0.020
Just+0.020
 widening+0.020
Neg logit contributions
 happens-0.014
 is-0.014
 doesn-0.014
 isn-0.014
 belongs-0.014
TokenWhen
Token ID2215
Feature activation+1.21e-02

Pos logit contributions
 widening+0.113
 pounding+0.102
 sidelines+0.098
 aside+0.097
king+0.095
Neg logit contributions
 is-0.163
 isn-0.155
 can-0.147
 I-0.146
 happens-0.144
Token he
Token ID339
Feature activation+1.10e-02

Pos logit contributions
eyed+0.026
Unfortunately+0.025
Meanwhile+0.025
Just+0.024
 widening+0.024
Neg logit contributions
 happens-0.019
 is-0.019
 guys-0.018
 doesn-0.018
 isn-0.018
Token was
Token ID373
Feature activation+6.12e-02

Pos logit contributions
Unfortunately+0.022
eyed+0.022
Meanwhile+0.022
Just+0.020
 sidelines+0.020
Neg logit contributions
 happens-0.019
 is-0.019
 isn-0.018
 doesn-0.018
 belongs-0.018
Token finished
Token ID5201
Feature activation+2.75e-02

Pos logit contributions
eyed+0.144
har+0.139
Unfortunately+0.138
Meanwhile+0.137
Dog+0.137
Neg logit contributions
 happens-0.075
?"-0.074
 is-0.071
 happened-0.069
 isn-0.064
Token,
Token ID11
Feature activation+6.12e-03

Pos logit contributions
Unfortunately+0.081
eyed+0.080
Meanwhile+0.079
Just+0.077
Days+0.076
Neg logit contributions
 is-0.023
 happens-0.021
 isn-0.020
 doesn-0.020
 guys-0.019
Token the
Token ID262
Feature activation-1.69e-03

Pos logit contributions
Unfortunately+0.012
Meanwhile+0.012
eyed+0.012
Just+0.012
 widening+0.012
Neg logit contributions
 happens-0.010
 is-0.010
 doesn-0.010
 isn-0.010
 guys-0.010
Token trophy
Token ID16383
Feature activation+3.04e-02

Pos logit contributions
 happens+0.003
 is+0.003
 doesn+0.003
 belongs+0.003
 isn+0.003
Neg logit contributions
Unfortunately-0.003
Meanwhile-0.003
eyed-0.003
Just-0.003
 widening-0.003
Token gle
Token ID26852
Feature activation+1.89e-02

Pos logit contributions
Unfortunately+0.071
eyed+0.070
Meanwhile+0.068
 sidelines+0.066
Just+0.066
Neg logit contributions
 is-0.043
 happens-0.042
 doesn-0.040
 belongs-0.040
 isn-0.040
Tokenamed
Token ID2434
Feature activation+3.17e-02

Pos logit contributions
Unfortunately+0.021
Meanwhile+0.019
 sidelines+0.018
 pounding+0.018
eyed+0.018
Neg logit contributions
 is-0.048
 happens-0.048
 belongs-0.047
 isn-0.046
 doesn-0.046
Token and
Token ID290
Feature activation+4.60e-02

Pos logit contributions
Unfortunately+0.068
Meanwhile+0.065
Finally+0.064
Days+0.062
Eventually+0.061
Neg logit contributions
 is-0.051
 isn-0.050
 doesn-0.049
 happens-0.048
 guys-0.046
Token spark
Token ID9009
Feature activation-1.63e-02

Pos logit contributions
Unfortunately+0.095
eyed+0.094
Just+0.090
Meanwhile+0.090
Days+0.090
Neg logit contributions
 happens-0.075
 is-0.074
 isn-0.072
 doesn-0.071
 happened-0.067
Token and
Token ID290
Feature activation-2.93e-02

Pos logit contributions
 happens+0.045
 belongs+0.043
 isn+0.042
 guys+0.041
 doesn+0.041
Neg logit contributions
Unfortunately-0.039
Meanwhile-0.038
eyed-0.037
Just-0.037
 pounding-0.037
Token wanted
Token ID2227
Feature activation-4.73e-03

Pos logit contributions
 happens+0.053
 isn+0.051
 is+0.050
 guys+0.050
 belongs+0.049
Neg logit contributions
Unfortunately-0.050
Meanwhile-0.050
eyed-0.048
Just-0.047
 pounding-0.047
Token to
Token ID284
Feature activation-5.57e-03

Pos logit contributions
 happens+0.009
 is+0.009
 doesn+0.009
 belongs+0.009
 isn+0.009
Neg logit contributions
Unfortunately-0.008
Meanwhile-0.008
eyed-0.008
Just-0.008
 pounding-0.007
Token fly
Token ID6129
Feature activation+5.36e-03

Pos logit contributions
 happens+0.012
 doesn+0.011
 is+0.011
 belongs+0.011
 isn+0.011
Neg logit contributions
Unfortunately-0.009
Meanwhile-0.008
eyed-0.008
Just-0.008
 widening-0.008
Token away
Token ID1497
Feature activation-1.75e-03

Pos logit contributions
Unfortunately+0.010
Meanwhile+0.010
eyed+0.009
 widening+0.009
Just+0.009
Neg logit contributions
 happens-0.010
 is-0.010
 doesn-0.010
 belongs-0.010
 isn-0.010
Token.
Token ID13
Feature activation+6.87e-03

Pos logit contributions
 happens+0.003
 is+0.003
 doesn+0.003
 belongs+0.003
 isn+0.003
Neg logit contributions
Unfortunately-0.004
Meanwhile-0.004
eyed-0.004
Just-0.003
 widening-0.003
Token She
Token ID1375
Feature activation-2.23e-02

Pos logit contributions
Unfortunately+0.011
eyed+0.011
Meanwhile+0.011
 pounding+0.011
 widening+0.011
Neg logit contributions
 is-0.014
 happens-0.014
 doesn-0.013
 isn-0.013
 belongs-0.013
Token asked
Token ID1965
Feature activation-2.14e-02

Pos logit contributions
 happens+0.043
 isn+0.042
 belongs+0.041
 doesn+0.041
 is+0.041
Neg logit contributions
Meanwhile-0.037
Unfortunately-0.037
Just-0.036
eyed-0.035
 pounding-0.034
Token her
Token ID607
Feature activation-1.98e-02

Pos logit contributions
 happens+0.047
 doesn+0.046
 belongs+0.046
 isn+0.046
 needs+0.045
Neg logit contributions
Unfortunately-0.030
Meanwhile-0.030
His-0.029
eyed-0.028
 pounding-0.028
Token bird
Token ID6512
Feature activation-2.10e-02

Pos logit contributions
 happens+0.034
 doesn+0.034
 belongs+0.034
 isn+0.034
 needs+0.032
Neg logit contributions
Meanwhile-0.037
Unfortunately-0.037
eyed-0.036
 pounding-0.036
Just-0.035
Token friends
Token ID2460
Feature activation-8.81e-03

Pos logit contributions
 happens+0.028
 doesn+0.026
 isn+0.026
 belongs+0.026
 guys+0.025
Neg logit contributions
Meanwhile-0.049
eyed-0.049
Unfortunately-0.049
 pounding-0.047
Just-0.047
Token gym
Token ID11550
Feature activation+2.46e-02

Pos logit contributions
Unfortunately+0.060
Meanwhile+0.059
eyed+0.058
Just+0.056
 widening+0.055
Neg logit contributions
 is-0.039
 doesn-0.037
 happens-0.037
 isn-0.037
 belongs-0.037
Token.
Token ID13
Feature activation-3.61e-03

Pos logit contributions
Unfortunately+0.055
Meanwhile+0.054
eyed+0.053
Just+0.052
 widening+0.051
Neg logit contributions
 is-0.036
 doesn-0.035
 happens-0.035
 belongs-0.034
 isn-0.034
Token He
Token ID679
Feature activation+1.20e-02

Pos logit contributions
 happens+0.007
 is+0.007
 belongs+0.007
 doesn+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.006
Meanwhile-0.006
eyed-0.006
Just-0.006
 widening-0.006
Token tried
Token ID3088
Feature activation+4.58e-02

Pos logit contributions
Unfortunately+0.024
Meanwhile+0.024
eyed+0.023
 widening+0.023
Just+0.022
Neg logit contributions
 is-0.020
 happens-0.020
 doesn-0.020
 isn-0.020
 belongs-0.019
Token out
Token ID503
Feature activation+1.13e-02

Pos logit contributions
 widening+0.105
Meanwhile+0.105
Unfortunately+0.105
eyed+0.103
har+0.103
Neg logit contributions
 is-0.071
 happens-0.065
 isn-0.064
 doesn-0.064
 guys-0.061
Token the
Token ID262
Feature activation+3.73e-03

Pos logit contributions
Unfortunately+0.023
Meanwhile+0.023
eyed+0.022
 widening+0.022
Just+0.021
Neg logit contributions
 is-0.019
 happens-0.019
 doesn-0.018
 isn-0.018
 belongs-0.018
Token slides
Token ID19392
Feature activation+3.77e-02

Pos logit contributions
Unfortunately+0.007
Meanwhile+0.007
eyed+0.007
Just+0.007
 pounding+0.007
Neg logit contributions
 happens-0.006
 is-0.006
 doesn-0.006
 isn-0.006
 belongs-0.006
Token,
Token ID11
Feature activation+3.81e-02

Pos logit contributions
Unfortunately+0.100
Meanwhile+0.097
eyed+0.096
Just+0.094
Finally+0.093
Neg logit contributions
 is-0.046
 isn-0.042
 doesn-0.042
 happens-0.041
 belongs-0.040
Token and
Token ID290
Feature activation+2.69e-02

Pos logit contributions
eyed+0.080
Unfortunately+0.080
Meanwhile+0.078
Finally+0.077
 widening+0.074
Neg logit contributions
 is-0.066
 doesn-0.065
 isn-0.064
 can-0.060
 happens-0.059
Token he
Token ID339
Feature activation+2.19e-02

Pos logit contributions
Unfortunately+0.053
eyed+0.053
Meanwhile+0.051
 widening+0.050
Just+0.050
Neg logit contributions
 happens-0.046
 is-0.046
 doesn-0.045
 isn-0.044
 guys-0.044
Token swung
Token ID28507
Feature activation+4.56e-02

Pos logit contributions
Unfortunately+0.047
Meanwhile+0.045
eyed+0.045
 widening+0.044
Just+0.043
Neg logit contributions
 is-0.035
 happens-0.035
 doesn-0.034
 isn-0.033
 belongs-0.033
Token herself
Token ID5223
Feature activation+1.19e-02

Pos logit contributions
Meanwhile+0.032
Unfortunately+0.032
eyed+0.030
 widening+0.030
rolling+0.030
Neg logit contributions
 is-0.035
 happens-0.033
 doesn-0.033
 isn-0.033
 belongs-0.032
Token.
Token ID13
Feature activation-8.86e-03

Pos logit contributions
Unfortunately+0.028
Meanwhile+0.027
eyed+0.027
 widening+0.026
Just+0.026
Neg logit contributions
 is-0.016
 happens-0.016
 doesn-0.016
 belongs-0.016
 isn-0.016
Token The
Token ID383
Feature activation+8.12e-03

Pos logit contributions
 happens+0.019
 guys+0.018
 belongs+0.018
 is+0.018
 doesn+0.018
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.014
Just-0.013
 widening-0.013
Token sun
Token ID4252
Feature activation-6.44e-03

Pos logit contributions
Unfortunately+0.016
eyed+0.016
Meanwhile+0.016
Just+0.015
 widening+0.015
Neg logit contributions
 happens-0.014
 is-0.014
 doesn-0.013
 belongs-0.013
 isn-0.013
Token was
Token ID373
Feature activation+6.62e-02

Pos logit contributions
 happens+0.008
 doesn+0.008
 belongs+0.008
 guys+0.008
 is+0.008
Neg logit contributions
Unfortunately-0.015
Meanwhile-0.015
eyed-0.015
 pounding-0.015
 widening-0.015
Token shining
Token ID22751
Feature activation+4.55e-03

Pos logit contributions
Unfortunately+0.136
eyed+0.134
Just+0.129
 sidelines+0.127
Meanwhile+0.125
Neg logit contributions
 is-0.110
 isn-0.103
?"-0.102
 happened-0.101
 happens-0.098
Token and
Token ID290
Feature activation-1.03e-02

Pos logit contributions
Unfortunately+0.010
Meanwhile+0.010
eyed+0.010
Just+0.009
 pounding+0.009
Neg logit contributions
 happens-0.007
 is-0.007
 doesn-0.007
 isn-0.007
 belongs-0.007
Token Mia
Token ID32189
Feature activation-1.50e-02

Pos logit contributions
 happens+0.018
 belongs+0.018
 doesn+0.017
 is+0.017
 guys+0.017
Neg logit contributions
Unfortunately-0.020
Meanwhile-0.019
eyed-0.019
 pounding-0.018
 widening-0.018
Token was
Token ID373
Feature activation+5.03e-02

Pos logit contributions
 happens+0.023
 guys+0.022
 belongs+0.022
 doesn+0.021
 isn+0.021
Neg logit contributions
Unfortunately-0.032
eyed-0.031
Meanwhile-0.031
Just-0.030
 pounding-0.030
Token happy
Token ID3772
Feature activation-2.52e-03

Pos logit contributions
Unfortunately+0.091
Meanwhile+0.084
 sidelines+0.083
Finally+0.082
Just+0.082
Neg logit contributions
 is-0.098
 isn-0.091
?"-0.090
 happens-0.088
 doesn-0.086
Token.
Token ID13
Feature activation-2.15e-02

Pos logit contributions
 happens+0.004
 is+0.004
 belongs+0.004
 doesn+0.004
 isn+0.004
Neg logit contributions
Unfortunately-0.005
Meanwhile-0.005
eyed-0.005
Just-0.005
 widening-0.005
Token very
Token ID845
Feature activation+1.32e-03

Pos logit contributions
Unfortunately+0.029
eyed+0.029
Meanwhile+0.028
Just+0.027
 widening+0.027
Neg logit contributions
 is-0.025
 happens-0.025
 doesn-0.024
 belongs-0.024
 guys-0.024
Token important
Token ID1593
Feature activation-1.02e-02

Pos logit contributions
Unfortunately+0.002
Meanwhile+0.002
eyed+0.002
 widening+0.002
Just+0.002
Neg logit contributions
 happens-0.003
 is-0.003
 belongs-0.003
 doesn-0.003
 isn-0.002
Token mission
Token ID4365
Feature activation-6.02e-04

Pos logit contributions
 happens+0.016
 doesn+0.015
 belongs+0.015
 is+0.015
 isn+0.015
Neg logit contributions
Unfortunately-0.021
Meanwhile-0.021
eyed-0.021
 pounding-0.020
Just-0.020
Token to
Token ID284
Feature activation+3.89e-02

Pos logit contributions
 happens+0.001
 is+0.001
 doesn+0.001
 belongs+0.001
 isn+0.001
Neg logit contributions
Unfortunately-0.001
Meanwhile-0.001
eyed-0.001
Just-0.001
 widening-0.001
Token do
Token ID466
Feature activation-1.10e-02

Pos logit contributions
Meanwhile+0.074
 widening+0.074
eyed+0.073
Unfortunately+0.073
Just+0.073
Neg logit contributions
 is-0.072
 guys-0.070
 belongs-0.067
 happens-0.066
 lives-0.065
Token.
Token ID13
Feature activation+4.90e-03

Pos logit contributions
 happens+0.017
 belongs+0.016
 doesn+0.016
 is+0.016
 isn+0.016
Neg logit contributions
Unfortunately-0.023
Meanwhile-0.023
eyed-0.022
Just-0.022
 pounding-0.022
Token
Token ID198
Feature activation+3.78e-02

Pos logit contributions
Unfortunately+0.009
Meanwhile+0.009
eyed+0.009
 widening+0.008
 pounding+0.008
Neg logit contributions
 happens-0.009
 is-0.009
 doesn-0.009
 isn-0.009
 belongs-0.009
Token
Token ID198
Feature activation+4.23e-02

Pos logit contributions
 sidelines+0.083
 widening+0.081
ured+0.080
har+0.079
 pounding+0.078
Neg logit contributions
 is-0.060
 isn-0.051
 happens-0.050
?"-0.050
 can-0.049
TokenThe
Token ID464
Feature activation+6.41e-03

Pos logit contributions
eyed+0.071
 sidelines+0.070
har+0.070
 widening+0.070
ured+0.069
Neg logit contributions
 is-0.082
 guys-0.074
 isn-0.074
 can-0.072
 happens-0.072
Token cat
Token ID3797
Feature activation-2.24e-02

Pos logit contributions
Unfortunately+0.018
eyed+0.018
Meanwhile+0.018
 widening+0.017
Just+0.017
Neg logit contributions
 is-0.006
 happens-0.006
 doesn-0.005
 guys-0.005
 belongs-0.005
Token ran
Token ID4966
Feature activation+3.84e-02

Pos logit contributions
 happens+0.038
 isn+0.036
 belongs+0.036
 doesn+0.036
 is+0.036
Neg logit contributions
Meanwhile-0.043
Unfortunately-0.043
eyed-0.042
Just-0.041
 pounding-0.040
Token.
Token ID13
Feature activation-7.75e-03

Pos logit contributions
Unfortunately+0.025
Meanwhile+0.024
eyed+0.023
 widening+0.023
Just+0.023
Neg logit contributions
 is-0.015
 happens-0.015
 belongs-0.015
 doesn-0.015
 needs-0.014
Token She
Token ID1375
Feature activation-2.72e-02

Pos logit contributions
 happens+0.015
 doesn+0.014
 belongs+0.014
 is+0.014
 isn+0.014
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.013
 widening-0.013
Just-0.013
Token was
Token ID373
Feature activation+1.50e-02

Pos logit contributions
 happens+0.035
 guys+0.032
 doesn+0.032
 isn+0.032
 belongs+0.031
Neg logit contributions
Just-0.063
eyed-0.063
Meanwhile-0.062
Unfortunately-0.062
His-0.061
Token always
Token ID1464
Feature activation+1.64e-02

Pos logit contributions
Unfortunately+0.025
Meanwhile+0.024
eyed+0.024
 widening+0.023
Just+0.023
Neg logit contributions
 is-0.031
 happens-0.030
 belongs-0.029
 needs-0.029
 doesn-0.029
Token poking
Token ID39048
Feature activation-3.48e-02

Pos logit contributions
Unfortunately+0.029
Meanwhile+0.028
eyed+0.027
Finally+0.027
 widening+0.027
Neg logit contributions
 is-0.032
 happens-0.031
 isn-0.031
 doesn-0.031
 belongs-0.031
Token her
Token ID607
Feature activation-1.62e-02

Pos logit contributions
 belongs+0.067
 happens+0.067
 needs+0.065
 doesn+0.064
mm+0.062
Neg logit contributions
Unfortunately-0.056
eyed-0.054
Meanwhile-0.054
Just-0.052
 pounding-0.051
Token big
Token ID1263
Feature activation-4.24e-02

Pos logit contributions
 happens+0.030
 belongs+0.028
 doesn+0.028
 isn+0.027
 is+0.027
Neg logit contributions
eyed-0.029
Unfortunately-0.029
Meanwhile-0.029
 pounding-0.028
Just-0.027
Token nose
Token ID9686
Feature activation+4.94e-02

Pos logit contributions
 happens+0.081
 belongs+0.076
 doesn+0.074
mm+0.072
 is+0.071
Neg logit contributions
 pounding-0.069
 crashing-0.067
 ledge-0.067
Unfortunately-0.067
 winding-0.067
Token around
Token ID1088
Feature activation-6.52e-03

Pos logit contributions
Unfortunately+0.125
har+0.120
Meanwhile+0.117
 sidelines+0.116
eyed+0.116
Neg logit contributions
 is-0.064
 guys-0.058
 doesn-0.057
 isn-0.055
 happens-0.055
Token,
Token ID11
Feature activation+3.12e-02

Pos logit contributions
 happens+0.012
 belongs+0.011
 doesn+0.011
 is+0.011
 isn+0.011
Neg logit contributions
Unfortunately-0.012
Meanwhile-0.012
eyed-0.012
 widening-0.012
Just-0.012
Token looking
Token ID2045
Feature activation-1.54e-02

Pos logit contributions
Unfortunately+0.061
eyed+0.061
Meanwhile+0.059
Finally+0.057
rolling+0.057
Neg logit contributions
 is-0.057
 doesn-0.055
 isn-0.055
 belongs-0.051
 needs-0.051
Token<|endoftext|>
Token ID50256
Feature activation-5.06e-02

Pos logit contributions
Unfortunately+0.004
Meanwhile+0.004
eyed+0.004
 widening+0.004
 pounding+0.004
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 isn-0.005
 belongs-0.005
TokenOnce
Token ID7454
Feature activation-6.96e-02

Pos logit contributions
 guys+0.113
 belongs+0.112
 happens+0.112
 doesn+0.110
 needs+0.109
Neg logit contributions
Unfortunately-0.069
Meanwhile-0.068
On-0.068
eyed-0.065
Just-0.064
Token upon
Token ID2402
Feature activation-8.21e-03

Pos logit contributions
 guys+0.172
 happens+0.166
 doesn+0.159
 tastes+0.158
 belongs+0.158
Neg logit contributions
Meanwhile-0.080
On-0.073
Unfortunately-0.072
rolling-0.072
eyed-0.071
Token a
Token ID257
Feature activation-1.84e-02

Pos logit contributions
 happens+0.016
 belongs+0.015
 doesn+0.015
 isn+0.015
 is+0.015
Neg logit contributions
Unfortunately-0.015
Meanwhile-0.014
eyed-0.014
Just-0.014
 widening-0.014
Token time
Token ID640
Feature activation-7.75e-02

Pos logit contributions
 happens+0.027
 belongs+0.025
 doesn+0.025
 guys+0.025
 isn+0.025
Neg logit contributions
Unfortunately-0.042
Meanwhile-0.040
eyed-0.040
Just-0.040
 widening-0.039
Token,
Token ID11
Feature activation-4.36e-02

Pos logit contributions
 happens+0.153
ishes+0.148
 tastes+0.146
 needs+0.146
 belongs+0.144
Neg logit contributions
rolling-0.122
On-0.119
Unfortunately-0.117
Meanwhile-0.116
His-0.115
Token there
Token ID612
Feature activation-6.68e-02

Pos logit contributions
 happens+0.091
 guys+0.087
 doesn+0.085
unn+0.085
 belongs+0.085
Neg logit contributions
On-0.070
Meanwhile-0.069
Unfortunately-0.068
Just-0.065
His-0.065
Token was
Token ID373
Feature activation-2.19e-02

Pos logit contributions
 guys+0.138
unn+0.135
ishes+0.135
 happens+0.133
 cares+0.128
Neg logit contributions
His-0.098
Meanwhile-0.097
Unfortunately-0.097
On-0.095
 pounding-0.093
Token a
Token ID257
Feature activation-5.43e-03

Pos logit contributions
 happens+0.047
 belongs+0.046
 doesn+0.046
 isn+0.045
 needs+0.044
Neg logit contributions
Unfortunately-0.035
Meanwhile-0.034
eyed-0.032
Just-0.032
 widening-0.032
Token rabbit
Token ID22746
Feature activation-3.87e-03

Pos logit contributions
 happens+0.009
 is+0.009
 doesn+0.009
 belongs+0.009
 isn+0.009
Neg logit contributions
Unfortunately-0.011
Meanwhile-0.011
eyed-0.011
Just-0.010
 widening-0.010
Token.
Token ID13
Feature activation-2.26e-02

Pos logit contributions
 happens+0.007
 is+0.006
 doesn+0.006
 belongs+0.006
 isn+0.006
Neg logit contributions
Unfortunately-0.008
Meanwhile-0.007
eyed-0.007
Just-0.007
 pounding-0.007
Token upon
Token ID2402
Feature activation+6.16e-04

Pos logit contributions
 happens+0.069
 guys+0.067
 doesn+0.066
 belongs+0.066
 isn+0.066
Neg logit contributions
Meanwhile-0.049
Unfortunately-0.048
eyed-0.047
His-0.045
On-0.044
Token a
Token ID257
Feature activation-2.01e-02

Pos logit contributions
Unfortunately+0.001
Meanwhile+0.001
eyed+0.001
 widening+0.001
Just+0.001
Neg logit contributions
 happens-0.001
 is-0.001
 doesn-0.001
 belongs-0.001
 isn-0.001
Token time
Token ID640
Feature activation-5.89e-02

Pos logit contributions
 happens+0.030
 belongs+0.029
 doesn+0.028
 isn+0.028
 guys+0.027
Neg logit contributions
Unfortunately-0.044
Meanwhile-0.043
Just-0.043
eyed-0.042
Finally-0.041
Token,
Token ID11
Feature activation-9.29e-03

Pos logit contributions
 happens+0.105
 doesn+0.100
 belongs+0.099
 guys+0.099
 needs+0.099
Neg logit contributions
Unfortunately-0.106
Meanwhile-0.105
His-0.105
Just-0.104
On-0.103
Token there
Token ID612
Feature activation-5.99e-02

Pos logit contributions
 happens+0.015
 doesn+0.014
 belongs+0.014
 isn+0.014
 guys+0.014
Neg logit contributions
Unfortunately-0.020
Meanwhile-0.020
eyed-0.019
Just-0.019
 pounding-0.018
Token was
Token ID373
Feature activation-1.46e-02

Pos logit contributions
 guys+0.122
ishes+0.119
unn+0.117
 happens+0.116
 cares+0.110
Neg logit contributions
His-0.093
Unfortunately-0.091
Meanwhile-0.091
On-0.090
 pounding-0.087
Token a
Token ID257
Feature activation-1.62e-02

Pos logit contributions
 happens+0.030
 doesn+0.029
 belongs+0.029
 isn+0.029
 is+0.028
Neg logit contributions
Unfortunately-0.025
Meanwhile-0.024
eyed-0.023
Just-0.023
 widening-0.023
Token little
Token ID1310
Feature activation-5.74e-02

Pos logit contributions
 happens+0.029
 belongs+0.028
 doesn+0.028
 isn+0.027
 is+0.027
Neg logit contributions
Unfortunately-0.030
Meanwhile-0.030
eyed-0.029
Just-0.029
 widening-0.029
Token girl
Token ID2576
Feature activation-2.30e-02

Pos logit contributions
 happens+0.104
 belongs+0.098
 doesn+0.097
 isn+0.096
 is+0.086
Neg logit contributions
 sidelines-0.100
 crashing-0.100
 lower-0.099
eyed-0.099
 low-0.098
Token named
Token ID3706
Feature activation-2.50e-02

Pos logit contributions
 happens+0.049
 isn+0.047
 guys+0.047
 doesn+0.047
 belongs+0.046
Neg logit contributions
eyed-0.037
Unfortunately-0.035
Meanwhile-0.035
Just-0.035
 pounding-0.034
Token Lily
Token ID20037
Feature activation-2.04e-02

Pos logit contributions
 happens+0.037
 doesn+0.035
 isn+0.034
 guys+0.033
 belongs+0.033
Neg logit contributions
Unfortunately-0.056
Meanwhile-0.055
eyed-0.054
Just-0.054
Her-0.054
Token to
Token ID284
Feature activation+2.99e-02

Pos logit contributions
 happens+0.022
 belongs+0.022
 doesn+0.021
 is+0.021
 isn+0.021
Neg logit contributions
Unfortunately-0.019
Meanwhile-0.019
eyed-0.018
Just-0.018
 pounding-0.018
Token clean
Token ID3424
Feature activation-2.46e-02

Pos logit contributions
Meanwhile+0.061
Unfortunately+0.060
eyed+0.058
Just+0.057
 widening+0.056
Neg logit contributions
 is-0.051
 guys-0.051
 belongs-0.048
 doesn-0.048
 happens-0.048
Token his
Token ID465
Feature activation-9.86e-03

Pos logit contributions
 happens+0.058
 belongs+0.055
 needs+0.055
 doesn+0.054
 guys+0.054
Neg logit contributions
Unfortunately-0.031
Meanwhile-0.030
eyed-0.029
 pounding-0.028
His-0.028
Token room
Token ID2119
Feature activation+1.34e-02

Pos logit contributions
 happens+0.016
 doesn+0.015
 isn+0.015
 is+0.015
 belongs+0.015
Neg logit contributions
Unfortunately-0.020
Meanwhile-0.020
eyed-0.020
Just-0.019
 pounding-0.019
Token before
Token ID878
Feature activation+4.46e-03

Pos logit contributions
Unfortunately+0.028
Meanwhile+0.028
eyed+0.028
 widening+0.027
Just+0.027
Neg logit contributions
 is-0.021
 doesn-0.020
 happens-0.020
 belongs-0.020
 isn-0.020
Token he
Token ID339
Feature activation-3.80e-02

Pos logit contributions
Unfortunately+0.009
eyed+0.009
Meanwhile+0.009
Just+0.008
 widening+0.008
Neg logit contributions
 happens-0.008
 is-0.007
 doesn-0.007
 belongs-0.007
 isn-0.007
Token could
Token ID714
Feature activation+2.29e-02

Pos logit contributions
 happens+0.063
 guys+0.063
 isn+0.062
 belongs+0.061
unn+0.059
Neg logit contributions
Unfortunately-0.067
Just-0.066
eyed-0.066
Meanwhile-0.066
 pounding-0.066
Token go
Token ID467
Feature activation+1.03e-02

Pos logit contributions
Meanwhile+0.043
Unfortunately+0.042
eyed+0.041
 victorious+0.040
 widening+0.040
Neg logit contributions
 is-0.044
 happens-0.041
 doesn-0.041
 isn-0.041
 belongs-0.040
Token outside
Token ID2354
Feature activation+1.67e-02

Pos logit contributions
Unfortunately+0.020
Meanwhile+0.019
eyed+0.019
 widening+0.018
Just+0.018
Neg logit contributions
 is-0.019
 happens-0.018
 belongs-0.018
 doesn-0.018
 guys-0.018
Token to
Token ID284
Feature activation+1.35e-02

Pos logit contributions
Unfortunately+0.039
Meanwhile+0.038
eyed+0.037
 widening+0.037
Just+0.037
Neg logit contributions
 is-0.023
 happens-0.022
 doesn-0.022
 guys-0.022
 isn-0.021
Token play
Token ID711
Feature activation+3.64e-03

Pos logit contributions
Unfortunately+0.025
Meanwhile+0.025
eyed+0.024
Just+0.023
 widening+0.023
Neg logit contributions
 is-0.025
 happens-0.025
 belongs-0.024
 doesn-0.024
 isn-0.024
Token smiled
Token ID13541
Feature activation+1.56e-02

Pos logit contributions
 happens+0.021
 belongs+0.020
 guys+0.020
 isn+0.020
 doesn+0.020
Neg logit contributions
Unfortunately-0.020
Meanwhile-0.019
eyed-0.019
Just-0.018
 widening-0.018
Token and
Token ID290
Feature activation-3.48e-06

Pos logit contributions
Unfortunately+0.035
Meanwhile+0.034
 sidelines+0.033
eyed+0.033
rolling+0.033
Neg logit contributions
 is-0.024
 happens-0.023
 doesn-0.022
 isn-0.021
 guys-0.021
Token said
Token ID531
Feature activation-4.22e-02

Feature not active on prev token;
no predictions were affected.
Token,
Token ID11
Feature activation+5.43e-02

Pos logit contributions
 belongs+0.065
 doesn+0.062
 isn+0.062
 needs+0.061
 guys+0.061
Neg logit contributions
His-0.083
eyed-0.083
Her-0.083
Unfortunately-0.082
Just-0.080
Token "
Token ID366
Feature activation-3.44e-03

Pos logit contributions
Unfortunately+0.132
eyed+0.131
Meanwhile+0.130
har+0.128
 sidelines+0.121
Neg logit contributions
 is-0.070
 doesn-0.064
 isn-0.063
 can-0.063
 happens-0.062
TokenYou
Token ID1639
Feature activation-4.35e-02

Pos logit contributions
 happens+0.007
 is+0.007
 belongs+0.007
 doesn+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.006
Meanwhile-0.006
eyed-0.006
Just-0.005
 widening-0.005
Token are
Token ID389
Feature activation-1.88e-02

Pos logit contributions
 belongs+0.056
 happens+0.055
 doesn+0.051
 isn+0.050
 needs+0.049
Neg logit contributions
eyed-0.092
Unfortunately-0.091
 victorious-0.090
His-0.088
 widening-0.086
Token welcome
Token ID7062
Feature activation-5.48e-02

Pos logit contributions
 happens+0.037
 doesn+0.035
 belongs+0.034
 guys+0.034
 isn+0.034
Neg logit contributions
Unfortunately-0.033
eyed-0.033
Meanwhile-0.032
 victorious-0.031
His-0.031
Token.
Token ID13
Feature activation-9.76e-02

Pos logit contributions
 happens+0.100
 belongs+0.099
 doesn+0.093
 needs+0.092
 isn+0.088
Neg logit contributions
Just-0.094
Her-0.092
Unfortunately-0.091
 victorious-0.091
eyed-0.090
Token I
Token ID314
Feature activation-1.33e-02

Pos logit contributions
 guys+0.172
 happens+0.154
 happened+0.150
ishes+0.150
ls+0.142
Neg logit contributions
eyed-0.174
Unfortunately-0.168
But-0.168
Just-0.167
His-0.161
Token am
Token ID716
Feature activation+1.64e-02

Pos logit contributions
 happens+0.020
 belongs+0.019
 doesn+0.019
 guys+0.018
 isn+0.018
Neg logit contributions
Unfortunately-0.029
eyed-0.028
Meanwhile-0.028
Just-0.027
 widening-0.027
Token.
Token ID13
Feature activation-1.04e-02

Pos logit contributions
Unfortunately+0.048
Meanwhile+0.048
eyed+0.046
rolling+0.044
Just+0.043
Neg logit contributions
 is-0.037
 happens-0.037
 guys-0.036
 doesn-0.035
 isn-0.035
Token<|endoftext|>
Token ID50256
Feature activation-3.33e-02

Pos logit contributions
 happens+0.023
 guys+0.022
 belongs+0.022
 is+0.022
 doesn+0.022
Neg logit contributions
Unfortunately-0.016
Meanwhile-0.015
eyed-0.015
Just-0.014
 widening-0.014
TokenOnce
Token ID7454
Feature activation-6.27e-02

Pos logit contributions
 happens+0.081
 guys+0.081
 belongs+0.080
 doesn+0.079
 needs+0.078
Neg logit contributions
Unfortunately-0.041
eyed-0.038
Meanwhile-0.038
On-0.036
Just-0.036
Token upon
Token ID2402
Feature activation-7.67e-03

Pos logit contributions
 guys+0.146
 happens+0.142
 doesn+0.140
 isn+0.139
 belongs+0.139
Neg logit contributions
Meanwhile-0.079
Unfortunately-0.074
On-0.074
eyed-0.072
His-0.071
Token a
Token ID257
Feature activation-3.27e-02

Pos logit contributions
 happens+0.014
 belongs+0.014
 doesn+0.014
 is+0.014
 isn+0.014
Neg logit contributions
Unfortunately-0.014
Meanwhile-0.014
eyed-0.013
Just-0.013
 widening-0.013
Token time
Token ID640
Feature activation-8.10e-02

Pos logit contributions
 happens+0.052
 belongs+0.050
 guys+0.049
 doesn+0.048
 isn+0.048
Neg logit contributions
Unfortunately-0.069
Just-0.066
Meanwhile-0.066
On-0.065
Finally-0.065
Token,
Token ID11
Feature activation-3.87e-02

Pos logit contributions
ishes+0.166
 happens+0.164
 guys+0.161
 belongs+0.158
 needs+0.158
Neg logit contributions
His-0.117
On-0.113
Just-0.112
Her-0.110
Unfortunately-0.109
Token there
Token ID612
Feature activation-7.07e-02

Pos logit contributions
 happens+0.077
 guys+0.076
 doesn+0.075
 isn+0.074
 belongs+0.073
Neg logit contributions
On-0.064
Meanwhile-0.063
Unfortunately-0.062
His-0.061
Just-0.060
Token was
Token ID373
Feature activation-3.36e-02

Pos logit contributions
unn+0.153
ishes+0.151
 guys+0.150
 happens+0.142
 cares+0.137
Neg logit contributions
His-0.101
On-0.095
But-0.094
Unfortunately-0.093
 pounding-0.092
Token a
Token ID257
Feature activation-1.44e-02

Pos logit contributions
 happens+0.077
 belongs+0.076
 doesn+0.076
 isn+0.075
 needs+0.074
Neg logit contributions
Unfortunately-0.049
Meanwhile-0.047
On-0.045
Just-0.045
 widening-0.043
Token little
Token ID1310
Feature activation-5.74e-02

Pos logit contributions
 happens+0.026
 belongs+0.025
 doesn+0.025
 isn+0.025
 is+0.024
Neg logit contributions
Unfortunately-0.027
Meanwhile-0.027
eyed-0.026
Just-0.025
 widening-0.025
Token,
Token ID11
Feature activation+3.35e-02

Pos logit contributions
 belongs+0.084
 needs+0.079
 lives+0.078
 doesn+0.076
 isn+0.074
Neg logit contributions
Unfortunately-0.092
Her-0.091
Just-0.089
Meanwhile-0.089
 realizing-0.088
Token "
Token ID366
Feature activation+2.20e-02

Pos logit contributions
eyed+0.075
Unfortunately+0.074
Meanwhile+0.072
har+0.072
 sidelines+0.070
Neg logit contributions
 is-0.051
 isn-0.049
 happens-0.049
 doesn-0.049
 can-0.046
TokenThat
Token ID2504
Feature activation-7.00e-02

Pos logit contributions
 widening+0.039
rolling+0.038
eyed+0.038
 victorious+0.037
 pounding+0.037
Neg logit contributions
 is-0.041
 happens-0.041
 doesn-0.041
 isn-0.039
 belongs-0.039
Token was
Token ID373
Feature activation-3.53e-02

Pos logit contributions
unn+0.105
mm+0.100
 guys+0.092
ishes+0.091
 cares+0.087
Neg logit contributions
Meanwhile-0.126
Unfortunately-0.122
Just-0.116
eyed-0.116
On-0.113
Token very
Token ID845
Feature activation-1.82e-02

Pos logit contributions
 happens+0.083
unn+0.081
 belongs+0.079
 guys+0.079
 is+0.078
Neg logit contributions
Unfortunately-0.043
Meanwhile-0.039
Just-0.038
 pounding-0.037
On-0.037
Token naughty
Token ID45128
Feature activation-7.90e-02

Pos logit contributions
 happens+0.031
 doesn+0.031
 guys+0.031
 isn+0.031
 is+0.030
Neg logit contributions
Unfortunately-0.034
Meanwhile-0.033
eyed-0.033
Just-0.032
 widening-0.031
Token,
Token ID11
Feature activation-2.39e-02

Pos logit contributions
 belongs+0.170
ishes+0.162
mm+0.160
 happens+0.158
 lives+0.158
Neg logit contributions
 louder-0.092
 faster-0.091
 crashing-0.090
Meanwhile-0.089
Unfortunately-0.088
Token Lily
Token ID20037
Feature activation-7.45e-02

Pos logit contributions
 happens+0.032
 belongs+0.030
 lives+0.028
 needs+0.028
 is+0.028
Neg logit contributions
Unfortunately-0.057
Meanwhile-0.054
However-0.053
Just-0.053
 widening-0.052
Token.
Token ID13
Feature activation-5.62e-02

Pos logit contributions
 guys+0.169
unn+0.167
 happens+0.162
mm+0.161
 belongs+0.152
Neg logit contributions
eyed-0.090
 pounding-0.084
Just-0.080
But-0.078
broken-0.075
Token You
Token ID921
Feature activation-6.32e-02

Pos logit contributions
 happens+0.084
 happened+0.079
 guys+0.078
 belongs+0.075
 doesn+0.073
Neg logit contributions
Unfortunately-0.124
eyed-0.121
Meanwhile-0.118
Just-0.116
His-0.112
Token wasted
Token ID18359
Feature activation-1.57e-02

Pos logit contributions
 happens+0.095
 belongs+0.087
mm+0.087
 isn+0.085
 doesn+0.083
Neg logit contributions
eyed-0.121
Unfortunately-0.121
On-0.114
 unt-0.111
har-0.111
Token "
Token ID366
Feature activation+3.10e-02

Pos logit contributions
Unfortunately+0.200
eyed+0.199
Meanwhile+0.199
har+0.199
Days+0.193
Neg logit contributions
 is-0.077
 doesn-0.077
 isn-0.075
 can-0.071
 I-0.070
TokenYou
Token ID1639
Feature activation-3.67e-02

Pos logit contributions
rolling+0.053
eyed+0.050
 victorious+0.050
 pounding+0.049
Unfortunately+0.049
Neg logit contributions
 doesn-0.059
 isn-0.058
 is-0.058
 happens-0.057
 belongs-0.054
Token're
Token ID821
Feature activation-3.39e-03

Pos logit contributions
 belongs+0.048
 happens+0.048
 lives+0.044
 is+0.043
unn+0.043
Neg logit contributions
Unfortunately-0.080
eyed-0.076
Meanwhile-0.074
Just-0.073
Finally-0.073
Token the
Token ID262
Feature activation+5.49e-03

Pos logit contributions
 happens+0.007
 is+0.007
 doesn+0.007
 belongs+0.007
 isn+0.007
Neg logit contributions
Unfortunately-0.006
Meanwhile-0.006
eyed-0.005
Just-0.005
 widening-0.005
Token best
Token ID1266
Feature activation-2.61e-02

Pos logit contributions
Unfortunately+0.015
Meanwhile+0.015
eyed+0.015
 widening+0.014
Just+0.014
Neg logit contributions
 happens-0.005
 is-0.005
 doesn-0.005
 guys-0.005
 isn-0.005
Token friend
Token ID1545
Feature activation-9.84e-02

Pos logit contributions
 belongs+0.043
 doesn+0.041
 happens+0.041
 isn+0.039
 is+0.038
Neg logit contributions
Unfortunately-0.052
eyed-0.050
 pounding-0.049
Meanwhile-0.049
Just-0.049
Token ever
Token ID1683
Feature activation-4.89e-02

Pos logit contributions
unn+0.196
 happens+0.196
ishes+0.194
 guys+0.193
wic+0.193
Neg logit contributions
eyed-0.123
But-0.113
 faster-0.110
Just-0.108
 free-0.108
Token!"
Token ID2474
Feature activation+6.64e-03

Pos logit contributions
 happens+0.091
 guys+0.088
 belongs+0.087
 owns+0.084
uffy+0.083
Neg logit contributions
eyed-0.079
Just-0.069
 realizing-0.068
shadow-0.067
But-0.066
Token Tim
Token ID5045
Feature activation-1.38e-02

Pos logit contributions
Unfortunately+0.012
Meanwhile+0.012
eyed+0.012
 widening+0.012
 pounding+0.012
Neg logit contributions
 is-0.012
 happens-0.012
 doesn-0.012
 isn-0.012
 belongs-0.011
Tokenmy
Token ID1820
Feature activation+1.97e-02

Pos logit contributions
 happens+0.022
 belongs+0.021
 guys+0.020
 isn+0.020
 doesn+0.020
Neg logit contributions
Unfortunately-0.029
Meanwhile-0.029
eyed-0.029
Just-0.027
 pounding-0.027
Token hugged
Token ID41130
Feature activation+1.24e-02

Pos logit contributions
Unfortunately+0.043
Meanwhile+0.042
eyed+0.042
 sidelines+0.041
 widening+0.041
Neg logit contributions
 is-0.030
 doesn-0.029
 happens-0.029
 isn-0.028
 belongs-0.028
Token pot
Token ID1787
Feature activation-3.15e-03

Pos logit contributions
 happens+0.022
 belongs+0.022
 doesn+0.021
 is+0.020
 isn+0.020
Neg logit contributions
Unfortunately-0.037
Meanwhile-0.037
eyed-0.035
Just-0.035
 widening-0.035
Token and
Token ID290
Feature activation-1.19e-02

Pos logit contributions
 happens+0.005
 is+0.004
 doesn+0.004
 belongs+0.004
 isn+0.004
Neg logit contributions
Unfortunately-0.007
Meanwhile-0.007
eyed-0.007
Just-0.007
 widening-0.007
Token tell
Token ID1560
Feature activation-2.95e-02

Pos logit contributions
 happens+0.017
 belongs+0.017
 is+0.016
 isn+0.016
 doesn+0.016
Neg logit contributions
Unfortunately-0.026
Meanwhile-0.026
eyed-0.025
Just-0.024
 pounding-0.024
Token me
Token ID502
Feature activation-2.43e-02

Pos logit contributions
 belongs+0.050
 happens+0.049
 needs+0.047
 doesn+0.047
 isn+0.047
Neg logit contributions
Unfortunately-0.056
Meanwhile-0.053
eyed-0.052
Just-0.051
 sidelines-0.050
Token if
Token ID611
Feature activation-8.75e-02

Pos logit contributions
 happens+0.042
 belongs+0.042
 doesn+0.041
 isn+0.040
 guys+0.040
Neg logit contributions
Unfortunately-0.046
Meanwhile-0.044
eyed-0.044
 pounding-0.042
Just-0.042
Token it
Token ID340
Feature activation-8.02e-02

Pos logit contributions
 belongs+0.114
unn+0.113
ishes+0.113
 isn+0.112
 guys+0.104
Neg logit contributions
Unfortunately-0.163
Meanwhile-0.163
Just-0.158
Suddenly-0.155
But-0.154
Token starts
Token ID4940
Feature activation-1.02e-02

Pos logit contributions
mm+0.074
 guys+0.071
unn+0.065
 guy+0.063
 isn+0.058
Neg logit contributions
eyed-0.173
Just-0.170
 pounding-0.170
 barely-0.168
 lower-0.168
Token to
Token ID284
Feature activation+1.77e-02

Pos logit contributions
 happens+0.019
 belongs+0.018
 is+0.018
 doesn+0.018
 guys+0.018
Neg logit contributions
Unfortunately-0.019
Meanwhile-0.018
eyed-0.018
Just-0.017
 pounding-0.017
Token bubble
Token ID14310
Feature activation-2.68e-02

Pos logit contributions
Unfortunately+0.033
eyed+0.033
Meanwhile+0.032
Just+0.031
rolling+0.030
Neg logit contributions
 is-0.033
 happens-0.033
 doesn-0.032
 guys-0.032
 belongs-0.032
Token."
Token ID526
Feature activation-2.25e-02

Pos logit contributions
 belongs+0.042
 happens+0.041
 isn+0.041
 doesn+0.041
 is+0.040
Neg logit contributions
Unfortunately-0.052
Meanwhile-0.050
eyed-0.050
Just-0.049
 pounding-0.049
Token
Token ID198
Feature activation+5.08e-03

Pos logit contributions
 happens+0.032
 belongs+0.030
 guys+0.030
 needs+0.030
 doesn+0.029
Neg logit contributions
Unfortunately-0.051
Meanwhile-0.049
Just-0.048
 widening-0.048
eyed-0.047
TokenOh
Token ID5812
Feature activation-6.13e-02

Pos logit contributions
Unfortunately+0.012
eyed+0.012
Meanwhile+0.012
 pounding+0.012
rolling+0.012
Neg logit contributions
 happens-0.015
 is-0.014
 doesn-0.014
 isn-0.014
 belongs-0.014
Token,
Token ID11
Feature activation-6.88e-02

Pos logit contributions
 belongs+0.116
unn+0.115
 happens+0.112
ishes+0.111
 doesn+0.108
Neg logit contributions
Her-0.100
Unfortunately-0.099
His-0.096
Their-0.093
Just-0.092
Token children
Token ID1751
Feature activation-4.88e-02

Pos logit contributions
unn+0.120
 belongs+0.120
ishes+0.115
 happens+0.115
 lives+0.110
Neg logit contributions
Unfortunately-0.117
Meanwhile-0.110
 sidelines-0.110
However-0.107
 understanding-0.106
Token,
Token ID11
Feature activation-6.75e-02

Pos logit contributions
 happens+0.083
 guys+0.078
 belongs+0.074
 doesn+0.073
 isn+0.072
Neg logit contributions
eyed-0.092
 widening-0.088
Unfortunately-0.088
 pounding-0.087
har-0.086
Token what
Token ID644
Feature activation-9.68e-02

Pos logit contributions
unn+0.114
 belongs+0.110
ishes+0.109
 happens+0.104
 lives+0.098
Neg logit contributions
Unfortunately-0.119
 understanding-0.114
 realizing-0.110
However-0.109
Meanwhile-0.107
Token have
Token ID423
Feature activation-9.64e-02

Pos logit contributions
ishes+0.176
unn+0.163
ley+0.151
mm+0.146
eros+0.143
Neg logit contributions
 understand-0.156
On-0.151
His-0.149
 overcome-0.146
eyed-0.145
Token you
Token ID345
Feature activation-2.27e-02

Pos logit contributions
ishes+0.169
unn+0.164
ish+0.158
eer+0.150
 needs+0.147
Neg logit contributions
Just-0.151
Their-0.145
Unfortunately-0.143
Her-0.139
The-0.136
Token done
Token ID1760
Feature activation-6.35e-02

Pos logit contributions
 happens+0.042
 doesn+0.040
 isn+0.040
 belongs+0.040
 guys+0.039
Neg logit contributions
eyed-0.039
Unfortunately-0.039
Meanwhile-0.038
Just-0.037
 widening-0.037
Token?"
Token ID1701
Feature activation-1.01e-02

Pos logit contributions
unn+0.082
ishes+0.078
 needs+0.073
 happens+0.072
 guys+0.072
Neg logit contributions
But-0.138
Unfortunately-0.137
eyed-0.135
On-0.134
Just-0.133
Token Mrs
Token ID9074
Feature activation-5.78e-02

Pos logit contributions
 happens+0.015
 belongs+0.014
 doesn+0.014
 isn+0.014
 guys+0.014
Neg logit contributions
Unfortunately-0.022
Meanwhile-0.022
eyed-0.021
 widening-0.021
Just-0.021
Token.
Token ID13
Feature activation+5.14e-04

Pos logit contributions
 isn+0.118
 happens+0.114
 doesn+0.113
 belongs+0.112
 guys+0.110
Neg logit contributions
eyed-0.093
Her-0.084
His-0.084
 widening-0.083
Meanwhile-0.082