TOP ACTIVATIONS
MAX = 0.258

, " You have to blow on the sail and steer
sounds . He likes to blow the trumpet in the morning
The fan started to blow gently and soon Molly felt
happy . He wants to blow a bigger bubble . He
and the wind started to blow very hard . The waves
your imagination . Let 's blow some more bubbles and have
ute . You have to blow hard and soft , and
whistle that he loved to blow . It was his favorite
'll puff and I 'll blow your house down !" said
" Make a wish and blow out the candles !"
holes with your fingers and blow gently . Do you want
horn . He liked to blow the horn and make loud
used her mouth again to blow and blow and blow until
used her mouth again to blow and blow and blow until
coloured stream ers that would blow in the wind .
park . Bubble wanted to blow some bubbles , so he
to blow and blow and blow until she was back on
to blow and blow and blow until she was back on
, " You have to blow in the pipe really fast
mouth again to blow and blow and blow until she was

MIN ACTIVATIONS
MIN = -0.343

listen . She threw a tant rum in the store .
them and never threw a tant rum again . <|endoftext|> Once
to cry and threw a tant rum . She didn 't
sad . She threw a tant rum and cried loudly .
her feet and threw a tant rum . She grabbed the
his feet and threw a tant rum . "
. He kept throwing a tant rum . His mom was
stubborn . She throws a tant rum . She kicks and
kept crying and throwing a tant rum . Finally
She cried and threw a tant rum . Her mom tried
surprise . He threw a tant rum . He kicked and
and started to throw a tant rum . Suddenly , she
and so she threw a tant rum when her parents said
temper and often threw temper tant rums . One day ,
The girl was too pet rified to move and just stared
it , she was pet rified . One
moms were mad . No matter how much they tried ,
His mom was taken aback . She said calmly ,
care of him . No matter what happened , Grand pa
its message - that no matter what , their friendship would

INTERVAL 0.108 - 0.258
CONTAINS 7.438%

truck loved to drive around and explore the world .
Pam . Everyday they would play together . One
towards it and began to slide down . Her eyes spark
They liked to ride fast and see the view of the
help him . He saw Tom 's hand was bleeding and

INTERVAL -0.043 - 0.108
CONTAINS 66.449%

there was a small baby called Johnny . He was very
flowers made the superhero feel a bit dizz y , but
oranges with oranges . She liked to sort things . When
selling . Bobby never stopped being brave and helping his mom
Once there was a mom my and a little girl .

INTERVAL -0.193 - -0.043
CONTAINS 25.719%

hospital , but it was too late . Lily had lost
erupt ing . He sh ivered in terror , thankful that
was reminded that no matter what , she had a family
could help stir the soup too . Her mom
So Sam grabbed his bag and started walking .

INTERVAL -0.343 - -0.193
CONTAINS 0.394%

with his teeth and ran away to play with it .
it . The frog jumped away , but the boy kept
won 't be sm elly anymore !" Tim my
. Lily said , " Wow ! Look at the big
magnet wouldn 't stick to anything anymore . But
Token,
Token ID11
Feature activation-3.05e-02

Pos logit contributions
 next+0.067
 put+0.065
 waves+0.064
 bat+0.063
 blow+0.063
Neg logit contributions
 unlike-0.135
 tactic-0.135
 matter-0.130
rified-0.130
 Tears-0.127
Token "
Token ID366
Feature activation-8.84e-02

Pos logit contributions
 unlike+0.037
 tactic+0.033
 Where+0.033
 Tears+0.033
rified+0.031
Neg logit contributions
ans-0.054
 waves-0.052
 bat-0.051
 next-0.051
 blow-0.050
TokenYou
Token ID1639
Feature activation+2.60e-03

Pos logit contributions
 unlike+0.146
 tactic+0.141
 matter+0.140
 Where+0.138
rified+0.137
Neg logit contributions
ans-0.116
 waves-0.111
 put-0.108
 bat-0.108
 blow-0.107
Token have
Token ID423
Feature activation+1.38e-01

Pos logit contributions
ans+0.003
 next+0.003
 waves+0.003
 blow+0.003
 bat+0.003
Neg logit contributions
 unlike-0.004
 tactic-0.004
 matter-0.004
rified-0.004
 Where-0.004
Token to
Token ID284
Feature activation-5.21e-02

Pos logit contributions
ans+0.180
 next+0.166
 blow+0.165
 waves+0.164
ck+0.163
Neg logit contributions
 unlike-0.230
 tactic-0.218
 matter-0.217
 Where-0.212
 Tears-0.212
Token blow
Token ID6611
Feature activation+2.58e-01

Pos logit contributions
 matter+0.097
mate+0.087
 avail+0.086
 tactic+0.086
 superpower+0.086
Neg logit contributions
ites-0.060
 pige-0.057
ans-0.056
go-0.054
sn-0.054
Token on
Token ID319
Feature activation-4.11e-02

Pos logit contributions
ans+0.270
 blow+0.255
 next+0.253
 put+0.249
 waves+0.247
Neg logit contributions
 unlike-0.474
 tactic-0.464
 matter-0.454
rified-0.448
 Tears-0.446
Token the
Token ID262
Feature activation+7.42e-02

Pos logit contributions
 unlike+0.081
 tactic+0.079
 matter+0.078
 Tears+0.077
rified+0.077
Neg logit contributions
ans-0.037
 next-0.036
 blow-0.035
 waves-0.035
 put-0.034
Token sail
Token ID14936
Feature activation+1.58e-01

Pos logit contributions
ck+0.085
ila+0.080
ans+0.079
ites+0.079
irts+0.079
Neg logit contributions
 unlike-0.124
 matter-0.124
ifying-0.121
 spec-0.120
 tactic-0.118
Token and
Token ID290
Feature activation+1.20e-01

Pos logit contributions
ans+0.158
 waves+0.152
 blow+0.150
ck+0.150
 put+0.148
Neg logit contributions
 unlike-0.297
 tactic-0.293
mate-0.284
rified-0.284
 matter-0.283
Token steer
Token ID27401
Feature activation+1.76e-01

Pos logit contributions
ans+0.122
is+0.103
irts+0.100
ites+0.098
 next+0.095
Neg logit contributions
 unlike-0.234
 matter-0.230
 Where-0.224
 tactic-0.222
 alas-0.221
Token sounds
Token ID5238
Feature activation-3.61e-02

Pos logit contributions
 unlike+0.061
 tactic+0.060
 matter+0.057
rified+0.057
 Where+0.057
Neg logit contributions
ans-0.042
 next-0.042
 blow-0.040
 put-0.040
 bat-0.040
Token.
Token ID13
Feature activation-9.45e-03

Pos logit contributions
 unlike+0.074
 tactic+0.071
 matter+0.069
 Where+0.069
rified+0.069
Neg logit contributions
ans-0.032
 next-0.031
 waves-0.030
 blow-0.030
 bat-0.030
Token He
Token ID679
Feature activation-8.05e-02

Pos logit contributions
 unlike+0.014
 Where+0.013
 tactic+0.013
 Tears+0.013
rified+0.013
Neg logit contributions
ans-0.014
 next-0.014
 blow-0.014
ck-0.013
 put-0.013
Token likes
Token ID7832
Feature activation+1.16e-02

Pos logit contributions
 unlike+0.129
 matter+0.124
 blonde+0.122
 spec+0.122
 kinda+0.121
Neg logit contributions
ans-0.102
 next-0.097
line-0.090
is-0.089
ites-0.088
Token to
Token ID284
Feature activation-4.55e-02

Pos logit contributions
ans+0.014
 next+0.013
 blow+0.013
 bat+0.013
 put+0.013
Neg logit contributions
 unlike-0.020
 tactic-0.019
 matter-0.019
 Where-0.019
 Tears-0.018
Token blow
Token ID6611
Feature activation+2.58e-01

Pos logit contributions
 matter+0.084
 tactic+0.081
 blonde+0.079
 unlike+0.078
 superpower+0.077
Neg logit contributions
ans-0.049
 next-0.048
ites-0.046
irts-0.045
 pige-0.044
Token the
Token ID262
Feature activation+8.53e-02

Pos logit contributions
ans+0.252
 next+0.244
 blow+0.237
 put+0.235
 bat+0.232
Neg logit contributions
 unlike-0.496
 tactic-0.488
 matter-0.470
rified-0.470
 Tears-0.467
Token trumpet
Token ID40334
Feature activation-7.52e-02

Pos logit contributions
ans+0.132
 next+0.130
 put+0.128
 blow+0.127
 bat+0.126
Neg logit contributions
 unlike-0.116
 tactic-0.114
rified-0.108
 matter-0.108
 Tears-0.106
Token in
Token ID287
Feature activation-3.40e-02

Pos logit contributions
 unlike+0.150
 tactic+0.147
 matter+0.141
rified+0.139
 Tears+0.139
Neg logit contributions
ans-0.069
 waves-0.065
 next-0.065
 put-0.064
 bat-0.063
Token the
Token ID262
Feature activation+8.54e-02

Pos logit contributions
 unlike+0.067
 tactic+0.065
rified+0.063
 matter+0.063
 Where+0.063
Neg logit contributions
 next-0.033
ans-0.033
 waves-0.032
 blow-0.032
 put-0.032
Token morning
Token ID3329
Feature activation-1.63e-02

Pos logit contributions
ans+0.090
 next+0.086
ck+0.086
 blow+0.084
 put+0.083
Neg logit contributions
 unlike-0.153
 tactic-0.153
 Tears-0.150
rified-0.149
 matter-0.148
Token
Token ID198
Feature activation+2.86e-02

Pos logit contributions
ans+0.037
 waves+0.037
 goes+0.036
 blow+0.035
irts+0.034
Neg logit contributions
 tactic-0.049
 Tears-0.048
 Where-0.048
 unlike-0.047
mate-0.047
TokenThe
Token ID464
Feature activation+7.33e-02

Pos logit contributions
ans+0.034
 waves+0.034
 goes+0.032
 blow+0.032
 put+0.031
Neg logit contributions
 tactic-0.049
 unlike-0.049
mate-0.047
 Tears-0.046
 Where-0.046
Token fan
Token ID4336
Feature activation-5.94e-02

Pos logit contributions
 next+0.079
ans+0.077
 waves+0.076
 blow+0.076
 put+0.075
Neg logit contributions
 unlike-0.136
 tactic-0.133
rified-0.128
 matter-0.127
 Where-0.127
Token started
Token ID2067
Feature activation-4.35e-02

Pos logit contributions
 unlike+0.104
 tactic+0.100
 matter+0.097
rified+0.096
 Tears+0.095
Neg logit contributions
ans-0.069
 bat-0.065
 next-0.065
 waves-0.064
 blow-0.063
Token to
Token ID284
Feature activation-5.50e-02

Pos logit contributions
 unlike+0.081
 tactic+0.080
rified+0.079
 Where+0.078
 matter+0.077
Neg logit contributions
 next-0.046
 blow-0.045
 put-0.045
 waves-0.044
 bat-0.044
Token blow
Token ID6611
Feature activation+2.57e-01

Pos logit contributions
 tactic+0.101
 unlike+0.100
 matter+0.099
 disgust+0.093
 Tears+0.093
Neg logit contributions
ans-0.061
 next-0.056
ites-0.055
 waves-0.054
irts-0.052
Token gently
Token ID15165
Feature activation+2.92e-02

Pos logit contributions
 next+0.196
 waves+0.189
 put+0.186
 bat+0.184
ans+0.182
Neg logit contributions
 unlike-0.564
 tactic-0.552
 Where-0.538
rified-0.533
 matter-0.532
Token and
Token ID290
Feature activation+1.08e-01

Pos logit contributions
ans+0.026
 next+0.024
 waves+0.023
 bat+0.022
ites+0.022
Neg logit contributions
 unlike-0.059
 tactic-0.059
 Tears-0.056
 matter-0.056
rified-0.055
Token soon
Token ID2582
Feature activation+4.64e-02

Pos logit contributions
ans+0.099
 next+0.096
 waves+0.093
 blow+0.092
 bat+0.092
Neg logit contributions
 unlike-0.216
 tactic-0.210
 matter-0.204
rified-0.203
 Where-0.203
Token Molly
Token ID30236
Feature activation+6.14e-03

Pos logit contributions
ans+0.042
 next+0.040
 blow+0.038
 bat+0.038
 waves+0.038
Neg logit contributions
 unlike-0.095
 tactic-0.091
rified-0.090
 matter-0.089
 Tears-0.089
Token felt
Token ID2936
Feature activation+2.50e-02

Pos logit contributions
ans+0.008
 next+0.007
 dads+0.007
irts+0.007
ites+0.007
Neg logit contributions
 unlike-0.010
rified-0.010
 blonde-0.010
 Tears-0.010
 spec-0.010
Token happy
Token ID3772
Feature activation-5.49e-03

Pos logit contributions
 next+0.045
ans+0.045
 put+0.044
 blow+0.044
 waves+0.044
Neg logit contributions
 unlike-0.063
 tactic-0.062
rified-0.060
 matter-0.059
 Where-0.059
Token.
Token ID13
Feature activation-6.45e-03

Pos logit contributions
ifying+0.014
 unlike+0.014
rified+0.014
 tactic+0.014
 alas+0.013
Neg logit contributions
 next-0.004
 and-0.003
 put-0.003
 goes-0.003
 waves-0.002
Token He
Token ID679
Feature activation-9.15e-02

Pos logit contributions
 matter+0.012
 perspective+0.011
 unlike+0.011
 tactic+0.011
rified+0.011
Neg logit contributions
 bat-0.008
 next-0.008
 put-0.007
 and-0.007
 goes-0.007
Token wants
Token ID3382
Feature activation+8.28e-02

Pos logit contributions
 unlike+0.155
 spec+0.150
 kinda+0.150
 matter+0.149
 must+0.146
Neg logit contributions
line-0.101
 next-0.099
ans-0.097
ites-0.093
pins-0.092
Token to
Token ID284
Feature activation-5.10e-02

Pos logit contributions
 next+0.082
 put+0.079
 bat+0.079
 waves+0.079
 blow+0.077
Neg logit contributions
 unlike-0.165
 tactic-0.163
rified-0.158
 Tears-0.154
 matter-0.154
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
 matter+0.096
 tactic+0.093
 unlike+0.090
mate+0.089
 superpower+0.087
Neg logit contributions
ans-0.053
 next-0.049
ites-0.049
irts-0.047
 goes-0.046
Token a
Token ID257
Feature activation-3.00e-03

Pos logit contributions
ans+0.252
 next+0.240
 blow+0.236
 put+0.235
 waves+0.234
Neg logit contributions
 unlike-0.493
 tactic-0.486
rified-0.469
 matter-0.467
 Tears-0.466
Token bigger
Token ID5749
Feature activation-3.95e-02

Pos logit contributions
 unlike+0.007
 tactic+0.006
rified+0.006
 Where+0.006
 Tears+0.006
Neg logit contributions
 bat-0.002
 blow-0.002
 next-0.002
 waves-0.002
 put-0.002
Token bubble
Token ID14310
Feature activation+7.67e-02

Pos logit contributions
 tactic+0.077
 unlike+0.076
 matter+0.075
rified+0.072
mate+0.072
Neg logit contributions
ans-0.038
 next-0.035
 waves-0.034
 put-0.034
 blow-0.033
Token.
Token ID13
Feature activation-8.06e-03

Pos logit contributions
ans+0.079
 waves+0.074
 next+0.071
 put+0.071
 blow+0.069
Neg logit contributions
 unlike-0.144
 tactic-0.141
mate-0.138
 matter-0.136
rified-0.135
Token He
Token ID679
Feature activation-9.02e-02

Pos logit contributions
 unlike+0.017
 perspective+0.017
 matter+0.017
 tactic+0.017
eway+0.016
Neg logit contributions
 next-0.008
 bat-0.007
 put-0.007
 and-0.007
 says-0.006
Token and
Token ID290
Feature activation+1.18e-01

Pos logit contributions
ck+0.154
irts+0.150
 blow+0.150
is+0.147
 climbs+0.145
Neg logit contributions
 unlike-0.232
 tactic-0.224
eway-0.217
 chamber-0.216
 exactly-0.216
Token the
Token ID262
Feature activation+7.82e-02

Pos logit contributions
ans+0.113
 next+0.105
is+0.101
 blow+0.098
 waves+0.098
Neg logit contributions
 unlike-0.230
 tactic-0.223
 Where-0.218
rified-0.218
 matter-0.217
Token wind
Token ID2344
Feature activation+7.11e-02

Pos logit contributions
irts+0.074
ck+0.073
ans+0.072
ites+0.070
 goes+0.067
Neg logit contributions
 matter-0.145
 unlike-0.143
 spec-0.142
rified-0.142
 tactic-0.139
Token started
Token ID2067
Feature activation-3.73e-02

Pos logit contributions
ans+0.081
irts+0.076
 dads+0.072
ites+0.072
 next+0.071
Neg logit contributions
 unlike-0.126
 tactic-0.116
 matter-0.116
 spec-0.114
rified-0.114
Token to
Token ID284
Feature activation-5.07e-02

Pos logit contributions
 unlike+0.059
 tactic+0.053
 Tears+0.052
 matter+0.051
 terribly+0.051
Neg logit contributions
ans-0.052
irts-0.048
 bat-0.047
ck-0.047
ites-0.047
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
 matter+0.088
 tactic+0.081
 source+0.079
 unlike+0.079
 disaster+0.079
Neg logit contributions
ites-0.067
ip-0.063
ila-0.060
 moms-0.059
 dads-0.057
Token very
Token ID845
Feature activation-1.18e-01

Pos logit contributions
ans+0.249
 next+0.239
 blow+0.237
 put+0.233
 bat+0.230
Neg logit contributions
 unlike-0.495
 tactic-0.483
rified-0.470
 matter-0.468
 Tears-0.464
Token hard
Token ID1327
Feature activation-1.59e-02

Pos logit contributions
rier+0.151
 slightly+0.150
 impossible+0.145
 kinda+0.145
 terribly+0.143
Neg logit contributions
ki-0.182
ans-0.182
 pige-0.176
Kids-0.171
 rides-0.171
Token.
Token ID13
Feature activation-1.51e-02

Pos logit contributions
 unlike+0.029
 tactic+0.028
 matter+0.027
 terribly+0.027
 Tears+0.026
Neg logit contributions
ans-0.018
 dads-0.016
 next-0.016
 bat-0.016
ck-0.016
Token The
Token ID383
Feature activation+6.47e-02

Pos logit contributions
 unlike+0.022
 tactic+0.022
 matter+0.021
rified+0.020
mate+0.020
Neg logit contributions
 next-0.022
ans-0.022
 put-0.022
 waves-0.021
 blow-0.021
Token waves
Token ID9813
Feature activation+1.61e-01

Pos logit contributions
irts+0.073
ck+0.071
ites+0.071
ans+0.070
sn+0.066
Neg logit contributions
 matter-0.112
 disaster-0.107
 spec-0.106
rified-0.105
 tactic-0.101
Token your
Token ID534
Feature activation+5.58e-03

Pos logit contributions
 tactic+0.193
 unlike+0.192
rified+0.184
 matter+0.182
mate+0.181
Neg logit contributions
 next-0.106
 put-0.103
 waves-0.103
 blow-0.101
ans-0.101
Token imagination
Token ID13843
Feature activation-2.83e-02

Pos logit contributions
 next+0.008
 waves+0.007
 put+0.007
 and+0.007
 bat+0.007
Neg logit contributions
 unlike-0.010
rified-0.010
 tactic-0.009
eway-0.009
 Tears-0.009
Token.
Token ID13
Feature activation-1.91e-02

Pos logit contributions
 unlike+0.050
 tactic+0.048
 matter+0.047
 terribly+0.046
 must+0.045
Neg logit contributions
ans-0.033
irts-0.031
 waves-0.030
ck-0.030
 dads-0.029
Token Let
Token ID3914
Feature activation+6.55e-03

Pos logit contributions
 unlike+0.027
 tactic+0.026
 matter+0.025
rified+0.025
 Tears+0.024
Neg logit contributions
 next-0.028
ans-0.028
 put-0.028
 waves-0.028
 bat-0.028
Token's
Token ID338
Feature activation+2.09e-02

Pos logit contributions
ites+0.009
 surf+0.008
ans+0.008
orse+0.008
 dads+0.008
Neg logit contributions
 matter-0.010
rified-0.010
 alright-0.010
 must-0.010
 ok-0.010
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
ans+0.014
 next+0.013
 waves+0.012
 blow+0.012
 bat+0.011
Neg logit contributions
 unlike-0.047
 tactic-0.046
 matter-0.046
 Where-0.045
 Tears-0.045
Token some
Token ID617
Feature activation+3.19e-02

Pos logit contributions
ans+0.300
 blow+0.288
 put+0.284
 dads+0.283
 next+0.281
Neg logit contributions
 tactic-0.420
 unlike-0.420
rified-0.409
mate-0.398
 Tears-0.395
Token more
Token ID517
Feature activation-3.60e-02

Pos logit contributions
ans+0.034
 next+0.031
 put+0.030
 blow+0.030
ck+0.029
Neg logit contributions
 unlike-0.059
 tactic-0.058
 matter-0.057
ifying-0.055
mate-0.054
Token bubbles
Token ID25037
Feature activation+9.92e-03

Pos logit contributions
 unlike+0.067
 tactic+0.065
 matter+0.064
eway+0.061
 superpower+0.060
Neg logit contributions
ans-0.037
 next-0.034
 dads-0.034
irts-0.033
 put-0.033
Token and
Token ID290
Feature activation+1.08e-01

Pos logit contributions
ans+0.012
ck+0.012
 dads+0.012
 climbs+0.012
 moms+0.012
Neg logit contributions
 unlike-0.015
mate-0.014
ifying-0.014
eway-0.013
 emanating-0.013
Token have
Token ID423
Feature activation+1.39e-01

Pos logit contributions
ans+0.110
is+0.090
 next+0.090
irts+0.090
ites+0.089
Neg logit contributions
 unlike-0.209
 matter-0.208
 tactic-0.197
 Where-0.196
 spec-0.193
Tokenute
Token ID1133
Feature activation+2.78e-02

Pos logit contributions
 put+0.106
 waves+0.105
 goes+0.104
ans+0.103
 next+0.103
Neg logit contributions
mate-0.069
 unlike-0.067
 tactic-0.063
ifying-0.063
 superpower-0.060
Token.
Token ID13
Feature activation-1.07e-02

Pos logit contributions
ans+0.027
 waves+0.026
 blow+0.025
 put+0.025
 bat+0.025
Neg logit contributions
 unlike-0.054
 tactic-0.052
mate-0.051
 matter-0.050
 Where-0.050
Token You
Token ID921
Feature activation-4.42e-02

Pos logit contributions
 unlike+0.012
 Where+0.012
 Tears+0.011
 tactic+0.011
rified+0.011
Neg logit contributions
ans-0.020
 next-0.019
 waves-0.019
 blow-0.019
 put-0.018
Token have
Token ID423
Feature activation+1.41e-01

Pos logit contributions
 unlike+0.061
 matter+0.058
 tactic+0.058
rified+0.056
 Where+0.056
Neg logit contributions
ans-0.068
 next-0.066
 waves-0.065
 bat-0.062
 blow-0.062
Token to
Token ID284
Feature activation-5.28e-02

Pos logit contributions
ans+0.165
 next+0.162
 waves+0.159
 blow+0.159
 bat+0.156
Neg logit contributions
 unlike-0.248
 tactic-0.240
 matter-0.232
 Where-0.231
 Tears-0.230
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
 matter+0.103
 tactic+0.092
 disgust+0.091
 blonde+0.090
 avail+0.089
Neg logit contributions
ites-0.058
ans-0.055
 waves-0.052
 pige-0.051
 next-0.050
Token hard
Token ID1327
Feature activation-2.08e-02

Pos logit contributions
ans+0.281
 blow+0.267
 next+0.262
 put+0.259
 waves+0.258
Neg logit contributions
 unlike-0.453
 tactic-0.449
 matter-0.433
rified-0.431
 Tears-0.430
Token and
Token ID290
Feature activation+1.15e-01

Pos logit contributions
 unlike+0.041
 tactic+0.040
 matter+0.040
 Where+0.039
rified+0.038
Neg logit contributions
ans-0.019
ck-0.017
 waves-0.017
 next-0.016
 bat-0.016
Token soft
Token ID2705
Feature activation+1.64e-02

Pos logit contributions
ans+0.135
ites+0.113
 waves+0.110
is+0.108
irts+0.108
Neg logit contributions
 matter-0.209
 unlike-0.203
 tactic-0.195
 Where-0.194
 kinda-0.193
Token,
Token ID11
Feature activation-2.36e-02

Pos logit contributions
ans+0.017
 next+0.017
 blow+0.016
 put+0.016
 waves+0.016
Neg logit contributions
 unlike-0.029
 tactic-0.029
 matter-0.028
rified-0.028
 Tears-0.028
Token and
Token ID290
Feature activation+1.15e-01

Pos logit contributions
 unlike+0.039
 matter+0.036
 tactic+0.035
 kinda+0.035
 Tears+0.035
Neg logit contributions
ans-0.031
 waves-0.028
 next-0.028
ites-0.028
ck-0.028
Token whistle
Token ID16121
Feature activation+6.42e-02

Pos logit contributions
rified+0.002
eway+0.002
 unlike+0.002
 Tears+0.002
 tactic+0.002
Neg logit contributions
 bat-0.001
 put-0.001
 next-0.001
 swing-0.001
 slide-0.001
Token that
Token ID326
Feature activation+1.23e-01

Pos logit contributions
 next+0.043
 put+0.041
 blow+0.039
 bat+0.039
 and+0.038
Neg logit contributions
 tactic-0.149
 unlike-0.146
 disgust-0.144
rified-0.143
 Tears-0.143
Token he
Token ID339
Feature activation+1.69e-02

Pos logit contributions
 next+0.135
 put+0.133
 waves+0.131
 bat+0.131
 blow+0.131
Neg logit contributions
 unlike-0.224
 tactic-0.223
rified-0.217
 Tears-0.212
ifying-0.211
Token loved
Token ID6151
Feature activation+3.95e-02

Pos logit contributions
ans+0.022
 next+0.022
 waves+0.021
 blow+0.021
 bat+0.021
Neg logit contributions
 unlike-0.027
 tactic-0.026
 matter-0.026
rified-0.025
 Where-0.025
Token to
Token ID284
Feature activation-4.96e-02

Pos logit contributions
 next+0.034
 and+0.032
 each+0.031
 put+0.030
 bat+0.030
Neg logit contributions
 tactic-0.087
rified-0.086
ifying-0.085
 unlike-0.084
 disgust-0.084
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
 matter+0.092
 tactic+0.089
 unlike+0.088
mate+0.086
 blonde+0.086
Neg logit contributions
ans-0.053
ites-0.050
irts-0.050
 next-0.048
 goes-0.045
Token.
Token ID13
Feature activation-1.53e-03

Pos logit contributions
ans+0.266
 next+0.264
 blow+0.257
 waves+0.256
 put+0.256
Neg logit contributions
 unlike-0.480
 tactic-0.469
rified-0.452
 matter-0.451
 Tears-0.448
Token It
Token ID632
Feature activation-4.86e-02

Pos logit contributions
 unlike+0.003
 tactic+0.003
rified+0.003
 matter+0.003
ifying+0.003
Neg logit contributions
 next-0.001
ans-0.001
 bat-0.001
 put-0.001
 waves-0.001
Token was
Token ID373
Feature activation-6.85e-02

Pos logit contributions
 unlike+0.074
 tactic+0.069
 matter+0.069
 Where+0.065
rified+0.064
Neg logit contributions
ans-0.072
 next-0.066
 waves-0.065
ites-0.065
irts-0.065
Token his
Token ID465
Feature activation-6.47e-02

Pos logit contributions
rified+0.139
 tactic+0.137
 Where+0.132
 unlike+0.131
 perspective+0.130
Neg logit contributions
 next-0.068
 put-0.068
 blow-0.066
 bat-0.065
 waves-0.063
Token favorite
Token ID4004
Feature activation-7.22e-02

Pos logit contributions
rified+0.105
 unlike+0.105
 Where+0.102
 tactic+0.102
eway+0.100
Neg logit contributions
 next-0.091
 bat-0.087
 waves-0.086
 blow-0.085
 put-0.084
Token'll
Token ID1183
Feature activation-1.24e-01

Pos logit contributions
 next+0.162
ans+0.162
 put+0.160
 blow+0.160
 bat+0.159
Neg logit contributions
 unlike-0.180
 tactic-0.175
rified-0.168
 matter-0.166
 Tears-0.166
Token puff
Token ID33679
Feature activation+3.26e-02

Pos logit contributions
 unlike+0.228
 matter+0.227
mate+0.213
 tactic+0.212
 superpower+0.212
Neg logit contributions
ans-0.137
ites-0.128
irts-0.116
 waves-0.115
 next-0.114
Token and
Token ID290
Feature activation+1.15e-01

Pos logit contributions
ans+0.038
 bat+0.036
 next+0.036
 put+0.036
 waves+0.036
Neg logit contributions
 unlike-0.055
rified-0.052
 tactic-0.052
 matter-0.052
 Tears-0.051
Token I
Token ID314
Feature activation+1.17e-01

Pos logit contributions
ans+0.118
 next+0.106
is+0.099
ites+0.098
 waves+0.098
Neg logit contributions
 unlike-0.220
 matter-0.212
 Where-0.207
 tactic-0.207
 Tears-0.205
Token'll
Token ID1183
Feature activation-1.24e-01

Pos logit contributions
ans+0.175
 next+0.172
 waves+0.169
 blow+0.164
 bat+0.163
Neg logit contributions
 unlike-0.167
 tactic-0.161
 matter-0.156
 Where-0.154
rified-0.152
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
 unlike+0.224
 matter+0.220
mate+0.206
 terribly+0.205
 tactic+0.204
Neg logit contributions
ans-0.137
ites-0.130
irts-0.122
 next-0.120
 waves-0.119
Token your
Token ID534
Feature activation+8.52e-03

Pos logit contributions
ans+0.278
 next+0.258
 blow+0.258
 put+0.256
 waves+0.252
Neg logit contributions
 unlike-0.459
 tactic-0.444
rified-0.436
 matter-0.435
 Tears-0.428
Token house
Token ID2156
Feature activation-8.87e-03

Pos logit contributions
ans+0.012
 next+0.011
 put+0.011
 blow+0.011
 bat+0.011
Neg logit contributions
 unlike-0.013
 tactic-0.013
 blonde-0.012
 matter-0.012
 Tears-0.012
Token down
Token ID866
Feature activation-1.08e-01

Pos logit contributions
 unlike+0.016
 matter+0.015
 tactic+0.015
ifying+0.015
 terribly+0.014
Neg logit contributions
ans-0.010
 waves-0.009
 bat-0.009
 dove-0.009
 dads-0.009
Token!"
Token ID2474
Feature activation+7.92e-02

Pos logit contributions
 unlike+0.214
 tactic+0.204
 matter+0.202
rified+0.199
ifying+0.198
Neg logit contributions
ans-0.105
 blow-0.095
 waves-0.095
 bat-0.094
 put-0.092
Token said
Token ID531
Feature activation+8.47e-02

Pos logit contributions
ans+0.054
 next+0.052
 waves+0.051
 blow+0.050
 bat+0.049
Neg logit contributions
 unlike-0.177
 tactic-0.174
rified-0.169
 Where-0.168
 Tears-0.168
Token "
Token ID366
Feature activation-8.44e-02

Pos logit contributions
 unlike+0.032
 Where+0.029
 Tears+0.028
 tactic+0.027
 Was+0.026
Neg logit contributions
ans-0.058
ck-0.054
 waves-0.054
ites-0.053
irts-0.053
TokenMake
Token ID12050
Feature activation+3.31e-02

Pos logit contributions
 unlike+0.114
 Where+0.112
 matter+0.111
ifying+0.109
 total+0.107
Neg logit contributions
ans-0.142
 waves-0.131
 bat-0.128
 put-0.124
 boats-0.123
Token a
Token ID257
Feature activation-9.89e-03

Pos logit contributions
 swing+0.044
 slide+0.044
ites+0.043
 next+0.043
 bat+0.043
Neg logit contributions
 unlike-0.047
 matter-0.047
 tactic-0.047
 Where-0.045
 Tears-0.045
Token wish
Token ID4601
Feature activation-5.90e-02

Pos logit contributions
 unlike+0.017
 tactic+0.016
rified+0.016
 Where+0.016
 Tears+0.015
Neg logit contributions
 next-0.012
 blow-0.012
 put-0.012
 waves-0.012
 bat-0.012
Token and
Token ID290
Feature activation+1.17e-01

Pos logit contributions
 unlike+0.111
 matter+0.103
 tactic+0.103
 Where+0.102
mate+0.101
Neg logit contributions
ans-0.065
 bat-0.058
 waves-0.058
ck-0.057
irts-0.057
Token blow
Token ID6611
Feature activation+2.56e-01

Pos logit contributions
ans+0.124
 next+0.113
 waves+0.112
 bat+0.111
 blow+0.110
Neg logit contributions
 unlike-0.224
 tactic-0.217
 matter-0.213
 Where-0.212
 Tears-0.208
Token out
Token ID503
Feature activation-1.29e-01

Pos logit contributions
ans+0.279
 next+0.265
 blow+0.264
 bat+0.262
 put+0.258
Neg logit contributions
 unlike-0.457
 tactic-0.445
 matter-0.433
rified-0.429
 Tears-0.427
Token the
Token ID262
Feature activation+6.84e-02

Pos logit contributions
 unlike+0.242
 tactic+0.229
 matter+0.227
ifying+0.226
 Tears+0.226
Neg logit contributions
ans-0.135
 blow-0.123
 waves-0.122
 next-0.121
 bat-0.121
Token candles
Token ID32268
Feature activation-2.76e-02

Pos logit contributions
ans+0.075
 next+0.074
 blow+0.073
 waves+0.072
 put+0.072
Neg logit contributions
 unlike-0.124
 tactic-0.120
 matter-0.116
rified-0.116
 Where-0.115
Token!"
Token ID2474
Feature activation+8.54e-02

Pos logit contributions
 unlike+0.050
ifying+0.049
 tactic+0.048
 matter+0.047
rified+0.046
Neg logit contributions
ans-0.031
ck-0.028
 bat-0.026
 waves-0.026
 dads-0.026
Token
Token ID198
Feature activation+2.71e-02

Pos logit contributions
ans+0.097
 blow+0.092
ck+0.092
 next+0.090
 waves+0.088
Neg logit contributions
 unlike-0.153
 Tears-0.151
 tactic-0.150
 Where-0.150
rified-0.144
Token holes
Token ID10421
Feature activation-5.73e-02

Pos logit contributions
ans+0.092
 next+0.087
 put+0.086
 bat+0.086
 blow+0.085
Neg logit contributions
 unlike-0.125
 tactic-0.124
 matter-0.122
 Tears-0.116
rified-0.116
Token with
Token ID351
Feature activation-7.70e-03

Pos logit contributions
 unlike+0.110
 tactic+0.103
rified+0.100
ifying+0.100
 matter+0.100
Neg logit contributions
ans-0.060
 waves-0.055
 blow-0.055
 bat-0.054
ck-0.053
Token your
Token ID534
Feature activation+1.08e-02

Pos logit contributions
 unlike+0.015
 matter+0.014
 tactic+0.014
ifying+0.014
 blonde+0.014
Neg logit contributions
ans-0.008
 next-0.007
 waves-0.007
 blow-0.007
 bat-0.007
Token fingers
Token ID9353
Feature activation-1.53e-02

Pos logit contributions
ans+0.013
 next+0.013
 blow+0.013
 waves+0.013
 put+0.013
Neg logit contributions
 unlike-0.018
 tactic-0.017
 matter-0.017
rified-0.017
 Tears-0.016
Token and
Token ID290
Feature activation+1.16e-01

Pos logit contributions
 unlike+0.030
 tactic+0.029
ifying+0.028
 matter+0.028
 Tears+0.028
Neg logit contributions
ans-0.015
 waves-0.014
 blow-0.013
 bat-0.013
 goes-0.013
Token blow
Token ID6611
Feature activation+2.55e-01

Pos logit contributions
ans+0.127
 next+0.117
 waves+0.113
is+0.113
 bat+0.111
Neg logit contributions
 unlike-0.212
 tactic-0.205
 matter-0.203
 Tears-0.198
 Where-0.197
Token gently
Token ID15165
Feature activation+3.00e-02

Pos logit contributions
ans+0.254
 next+0.243
 blow+0.242
 waves+0.237
 put+0.236
Neg logit contributions
 unlike-0.480
 tactic-0.468
 matter-0.454
rified-0.450
 Tears-0.449
Token.
Token ID13
Feature activation-1.29e-02

Pos logit contributions
ans+0.032
 waves+0.029
 next+0.029
 dads+0.029
ites+0.028
Neg logit contributions
 unlike-0.054
 tactic-0.053
 matter-0.052
ifying-0.052
 Tears-0.051
Token Do
Token ID2141
Feature activation-1.02e-01

Pos logit contributions
 unlike+0.017
 tactic+0.016
rified+0.015
 matter+0.015
 Tears+0.015
Neg logit contributions
 next-0.021
ans-0.021
 waves-0.021
 blow-0.021
 put-0.020
Token you
Token ID345
Feature activation+1.09e-01

Pos logit contributions
 matter+0.205
 Where+0.194
 unlike+0.194
 Tears+0.191
 tactic+0.190
Neg logit contributions
ans-0.092
 waves-0.087
irts-0.082
 bat-0.080
 blow-0.078
Token want
Token ID765
Feature activation+8.07e-02

Pos logit contributions
ans+0.142
 waves+0.136
 next+0.134
ites+0.129
 goes+0.125
Neg logit contributions
 matter-0.180
 unlike-0.177
 tactic-0.176
 disgust-0.164
 kinda-0.163
Token horn
Token ID12718
Feature activation-6.61e-02

Pos logit contributions
 unlike+0.029
 tactic+0.029
rified+0.028
 Tears+0.028
 Where+0.027
Neg logit contributions
 next-0.013
 bat-0.013
 blow-0.013
 waves-0.013
ans-0.013
Token.
Token ID13
Feature activation-1.02e-02

Pos logit contributions
 tactic+0.149
 unlike+0.147
rified+0.141
 matter+0.141
 Tears+0.141
Neg logit contributions
 next-0.048
 blow-0.046
 put-0.045
 bat-0.044
 waves-0.043
Token He
Token ID679
Feature activation-8.55e-02

Pos logit contributions
 unlike+0.016
 tactic+0.015
rified+0.015
 matter+0.015
ifying+0.014
Neg logit contributions
ans-0.014
 next-0.014
 bat-0.014
 waves-0.014
 put-0.014
Token liked
Token ID8288
Feature activation-1.62e-02

Pos logit contributions
 unlike+0.133
 matter+0.127
 tactic+0.124
 blonde+0.121
 terribly+0.119
Neg logit contributions
ans-0.119
 next-0.111
irts-0.109
 waves-0.109
ites-0.106
Token to
Token ID284
Feature activation-5.09e-02

Pos logit contributions
 unlike+0.030
 tactic+0.029
rified+0.029
 matter+0.028
 Tears+0.028
Neg logit contributions
 next-0.017
ans-0.017
 waves-0.017
 blow-0.017
 put-0.017
Token blow
Token ID6611
Feature activation+2.55e-01

Pos logit contributions
 matter+0.097
 tactic+0.093
 unlike+0.091
 blonde+0.090
 superpower+0.088
Neg logit contributions
ans-0.052
 next-0.052
irts-0.049
ites-0.048
is-0.045
Token the
Token ID262
Feature activation+8.14e-02

Pos logit contributions
ans+0.257
 next+0.254
 blow+0.247
 put+0.245
 waves+0.244
Neg logit contributions
 unlike-0.485
 tactic-0.473
rified-0.457
 matter-0.456
 Tears-0.452
Token horn
Token ID12718
Feature activation-6.78e-02

Pos logit contributions
ans+0.126
 next+0.126
 waves+0.124
 blow+0.124
 put+0.123
Neg logit contributions
 unlike-0.112
 tactic-0.108
rified-0.103
 matter-0.102
 Where-0.102
Token and
Token ID290
Feature activation+1.14e-01

Pos logit contributions
 unlike+0.148
 tactic+0.145
rified+0.141
 matter+0.141
 Tears+0.140
Neg logit contributions
ans-0.049
 next-0.048
 waves-0.047
 blow-0.046
 put-0.046
Token make
Token ID787
Feature activation+8.89e-02

Pos logit contributions
ans+0.088
 next+0.087
 waves+0.083
 blow+0.082
 bat+0.082
Neg logit contributions
 unlike-0.245
 tactic-0.240
 matter-0.233
rified-0.232
 Where-0.231
Token loud
Token ID7812
Feature activation-4.18e-02

Pos logit contributions
ans+0.112
 next+0.112
 blow+0.109
 waves+0.109
 bat+0.109
Neg logit contributions
 unlike-0.146
 tactic-0.142
 matter-0.136
rified-0.135
 Where-0.135
Token used
Token ID973
Feature activation+6.82e-02

Pos logit contributions
ans+0.031
irts+0.028
ites+0.027
 dads+0.026
ands+0.026
Neg logit contributions
 unlike-0.026
 must-0.025
 matter-0.024
 terribly-0.024
rified-0.024
Token her
Token ID607
Feature activation-3.61e-02

Pos logit contributions
 put+0.059
 next+0.057
 blow+0.055
 bat+0.054
 waves+0.053
Neg logit contributions
 unlike-0.143
 tactic-0.143
rified-0.141
 Where-0.138
 Thompson-0.135
Token mouth
Token ID5422
Feature activation-1.29e-01

Pos logit contributions
 unlike+0.060
 tactic+0.059
rified+0.057
 Where+0.056
 matter+0.056
Neg logit contributions
 next-0.045
ans-0.045
 waves-0.044
 blow-0.044
 put-0.044
Token again
Token ID757
Feature activation-5.88e-02

Pos logit contributions
 tactic+0.247
 unlike+0.246
 matter+0.238
 Where+0.236
rified+0.236
Neg logit contributions
 next-0.132
 put-0.126
 blow-0.123
 bat-0.122
 waves-0.122
Token to
Token ID284
Feature activation-5.85e-02

Pos logit contributions
 unlike+0.131
 tactic+0.126
 matter+0.124
 Tears+0.123
 blonde+0.122
Neg logit contributions
ans-0.046
 waves-0.039
 blow-0.037
 bat-0.036
ck-0.035
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.110
 unlike+0.108
 tactic+0.105
mate+0.104
 disgust+0.103
Neg logit contributions
ans-0.063
ites-0.059
 next-0.055
irts-0.054
 dads-0.052
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.240
 waves+0.235
 put+0.232
 bat+0.229
 blow+0.228
Neg logit contributions
 unlike-0.506
 tactic-0.497
 Where-0.484
rified-0.479
 matter-0.478
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.104
 next+0.091
 waves+0.086
 bat+0.086
ites+0.085
Neg logit contributions
 unlike-0.197
 tactic-0.188
 matter-0.187
 Tears-0.186
 Where-0.182
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.243
 waves+0.238
 put+0.237
 blow+0.235
ans+0.234
Neg logit contributions
 unlike-0.500
 tactic-0.492
 Where-0.475
 matter-0.472
rified-0.472
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.109
 next+0.096
 waves+0.091
 bat+0.090
ites+0.089
Neg logit contributions
 unlike-0.193
 matter-0.183
 tactic-0.182
 Tears-0.181
 Where-0.176
Token until
Token ID1566
Feature activation-3.86e-02

Pos logit contributions
 next+0.235
 waves+0.230
 put+0.229
 blow+0.227
ans+0.225
Neg logit contributions
 unlike-0.508
 tactic-0.501
 Where-0.483
rified-0.480
 matter-0.480
Token used
Token ID973
Feature activation+6.82e-02

Pos logit contributions
ans+0.031
irts+0.028
ites+0.027
 dads+0.026
ands+0.026
Neg logit contributions
 unlike-0.026
 must-0.025
 matter-0.024
 terribly-0.024
rified-0.024
Token her
Token ID607
Feature activation-3.61e-02

Pos logit contributions
 put+0.059
 next+0.057
 blow+0.055
 bat+0.054
 waves+0.053
Neg logit contributions
 unlike-0.143
 tactic-0.143
rified-0.141
 Where-0.138
 Thompson-0.135
Token mouth
Token ID5422
Feature activation-1.29e-01

Pos logit contributions
 unlike+0.060
 tactic+0.059
rified+0.057
 Where+0.056
 matter+0.056
Neg logit contributions
 next-0.045
ans-0.045
 waves-0.044
 blow-0.044
 put-0.044
Token again
Token ID757
Feature activation-5.88e-02

Pos logit contributions
 tactic+0.247
 unlike+0.246
 matter+0.238
 Where+0.236
rified+0.236
Neg logit contributions
 next-0.132
 put-0.126
 blow-0.123
 bat-0.122
 waves-0.122
Token to
Token ID284
Feature activation-5.85e-02

Pos logit contributions
 unlike+0.131
 tactic+0.126
 matter+0.124
 Tears+0.123
 blonde+0.122
Neg logit contributions
ans-0.046
 waves-0.039
 blow-0.037
 bat-0.036
ck-0.035
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.110
 unlike+0.108
 tactic+0.105
mate+0.104
 disgust+0.103
Neg logit contributions
ans-0.063
ites-0.059
 next-0.055
irts-0.054
 dads-0.052
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.240
 waves+0.235
 put+0.232
 bat+0.229
 blow+0.228
Neg logit contributions
 unlike-0.506
 tactic-0.497
 Where-0.484
rified-0.479
 matter-0.478
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.104
 next+0.091
 waves+0.086
 bat+0.086
ites+0.085
Neg logit contributions
 unlike-0.197
 tactic-0.188
 matter-0.187
 Tears-0.186
 Where-0.182
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.243
 waves+0.238
 put+0.237
 blow+0.235
ans+0.234
Neg logit contributions
 unlike-0.500
 tactic-0.492
 Where-0.475
 matter-0.472
rified-0.472
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.109
 next+0.096
 waves+0.091
 bat+0.090
ites+0.089
Neg logit contributions
 unlike-0.193
 matter-0.183
 tactic-0.182
 Tears-0.181
 Where-0.176
Token until
Token ID1566
Feature activation-3.86e-02

Pos logit contributions
 next+0.235
 waves+0.230
 put+0.229
 blow+0.227
ans+0.225
Neg logit contributions
 unlike-0.508
 tactic-0.501
 Where-0.483
rified-0.480
 matter-0.480
Token coloured
Token ID34746
Feature activation+2.40e-03

Pos logit contributions
 unlike+0.281
 tactic+0.268
 matter+0.258
 Tears+0.256
 blonde+0.254
Neg logit contributions
ans-0.241
 next-0.237
 bat-0.234
 blow-0.233
 waves-0.232
Token stream
Token ID4269
Feature activation+1.04e-01

Pos logit contributions
 next+0.002
ans+0.002
 waves+0.002
 blow+0.002
 put+0.002
Neg logit contributions
 unlike-0.005
 tactic-0.005
rified-0.004
 matter-0.004
 Where-0.004
Tokeners
Token ID364
Feature activation-1.48e-02

Pos logit contributions
ck+0.134
 dads+0.134
 blow+0.132
ans+0.131
irts+0.128
Neg logit contributions
 unlike-0.169
 terribly-0.152
 blonde-0.146
eway-0.145
ifying-0.145
Token that
Token ID326
Feature activation+1.18e-01

Pos logit contributions
 unlike+0.026
ifying+0.024
 blonde+0.024
 terribly+0.023
 slightly+0.023
Neg logit contributions
ans-0.018
 dads-0.016
irts-0.016
 waves-0.016
 bat-0.016
Token would
Token ID561
Feature activation-8.05e-02

Pos logit contributions
ans+0.135
 next+0.128
 waves+0.126
 blow+0.124
 bat+0.123
Neg logit contributions
 unlike-0.216
 tactic-0.205
 matter-0.201
 Where-0.199
rified-0.199
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.152
 unlike+0.152
 tactic+0.136
 total+0.135
 terribly+0.133
Neg logit contributions
ans-0.096
ites-0.085
irts-0.084
 waves-0.082
 dads-0.078
Token in
Token ID287
Feature activation-4.27e-02

Pos logit contributions
ans+0.284
 next+0.278
 blow+0.271
 waves+0.270
 put+0.270
Neg logit contributions
 unlike-0.457
 tactic-0.441
 matter-0.429
rified-0.427
 Tears-0.420
Token the
Token ID262
Feature activation+7.24e-02

Pos logit contributions
 unlike+0.091
 tactic+0.089
 matter+0.086
rified+0.086
 Where+0.085
Neg logit contributions
ans-0.034
 next-0.033
 waves-0.032
 blow-0.032
 put-0.032
Token wind
Token ID2344
Feature activation+6.87e-02

Pos logit contributions
ans+0.079
ck+0.076
 blow+0.070
 put+0.069
 next+0.068
Neg logit contributions
 unlike-0.129
ifying-0.129
 Tears-0.127
 matter-0.126
 terribly-0.126
Token.
Token ID13
Feature activation-8.73e-03

Pos logit contributions
ans+0.076
 waves+0.072
 goes+0.070
 bat+0.069
ck+0.068
Neg logit contributions
 unlike-0.124
 tactic-0.117
 matter-0.113
ifying-0.113
 Tears-0.111
Token
Token ID220
Feature activation-1.01e-01

Pos logit contributions
 unlike+0.017
 tactic+0.016
 matter+0.016
rified+0.016
mate+0.015
Neg logit contributions
 next-0.009
 put-0.009
 bat-0.009
ans-0.009
 waves-0.008
Token park
Token ID3952
Feature activation+7.84e-02

Pos logit contributions
ans+0.084
ck+0.082
 blow+0.078
 put+0.076
 next+0.075
Neg logit contributions
 unlike-0.120
 matter-0.117
 tactic-0.117
 Tears-0.115
 terribly-0.114
Token.
Token ID13
Feature activation-2.29e-02

Pos logit contributions
ans+0.072
 blow+0.070
 next+0.070
 waves+0.069
 bat+0.068
Neg logit contributions
 unlike-0.156
 tactic-0.152
rified-0.148
 matter-0.146
 Tears-0.146
Token Bubble
Token ID33691
Feature activation+2.62e-02

Pos logit contributions
 unlike+0.037
 tactic+0.036
 matter+0.035
rified+0.034
 Tears+0.034
Neg logit contributions
ans-0.030
 next-0.030
 bat-0.029
 waves-0.029
 put-0.029
Token wanted
Token ID2227
Feature activation+4.62e-02

Pos logit contributions
irts+0.038
ans+0.038
aster+0.037
 dads+0.036
 blow+0.036
Neg logit contributions
 unlike-0.036
rified-0.035
 must-0.034
 terribly-0.034
 tactic-0.033
Token to
Token ID284
Feature activation-5.56e-02

Pos logit contributions
 next+0.047
 put+0.046
 bat+0.046
 waves+0.046
ans+0.045
Neg logit contributions
 unlike-0.089
 tactic-0.088
rified-0.085
 matter-0.084
 Tears-0.083
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 tactic+0.107
 matter+0.106
 unlike+0.102
mate+0.101
 Where+0.098
Neg logit contributions
ans-0.057
 next-0.052
irts-0.052
ites-0.051
is-0.048
Token some
Token ID617
Feature activation+3.17e-02

Pos logit contributions
ans+0.262
 next+0.261
 waves+0.254
 blow+0.254
 put+0.252
Neg logit contributions
 unlike-0.478
 tactic-0.468
rified-0.451
 matter-0.449
 Where-0.447
Token bubbles
Token ID25037
Feature activation+9.00e-03

Pos logit contributions
ans+0.031
 next+0.029
 put+0.027
 blow+0.027
 waves+0.027
Neg logit contributions
 unlike-0.061
 tactic-0.060
 matter-0.059
mate-0.057
 Tears-0.057
Token,
Token ID11
Feature activation-3.32e-02

Pos logit contributions
ans+0.010
 waves+0.009
ck+0.009
irts+0.009
 goes+0.009
Neg logit contributions
 unlike-0.016
 tactic-0.015
mate-0.015
rified-0.015
 terribly-0.015
Token so
Token ID523
Feature activation-9.36e-02

Pos logit contributions
 unlike+0.044
 tactic+0.041
 Tears+0.038
 matter+0.038
 kinda+0.038
Neg logit contributions
ans-0.055
 next-0.053
 blow-0.052
 waves-0.052
 goes-0.051
Token he
Token ID339
Feature activation+6.61e-03

Pos logit contributions
 unlike+0.172
 tactic+0.166
rified+0.161
 Where+0.160
 matter+0.160
Neg logit contributions
ans-0.103
 next-0.100
 waves-0.098
 blow-0.098
 put-0.097
Token to
Token ID284
Feature activation-5.85e-02

Pos logit contributions
 unlike+0.131
 tactic+0.126
 matter+0.124
 Tears+0.123
 blonde+0.122
Neg logit contributions
ans-0.046
 waves-0.039
 blow-0.037
 bat-0.036
ck-0.035
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.110
 unlike+0.108
 tactic+0.105
mate+0.104
 disgust+0.103
Neg logit contributions
ans-0.063
ites-0.059
 next-0.055
irts-0.054
 dads-0.052
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.240
 waves+0.235
 put+0.232
 bat+0.229
 blow+0.228
Neg logit contributions
 unlike-0.506
 tactic-0.497
 Where-0.484
rified-0.479
 matter-0.478
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.104
 next+0.091
 waves+0.086
 bat+0.086
ites+0.085
Neg logit contributions
 unlike-0.197
 tactic-0.188
 matter-0.187
 Tears-0.186
 Where-0.182
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.243
 waves+0.238
 put+0.237
 blow+0.235
ans+0.234
Neg logit contributions
 unlike-0.500
 tactic-0.492
 Where-0.475
 matter-0.472
rified-0.472
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.109
 next+0.096
 waves+0.091
 bat+0.090
ites+0.089
Neg logit contributions
 unlike-0.193
 matter-0.183
 tactic-0.182
 Tears-0.181
 Where-0.176
Token until
Token ID1566
Feature activation-3.86e-02

Pos logit contributions
 next+0.235
 waves+0.230
 put+0.229
 blow+0.227
ans+0.225
Neg logit contributions
 unlike-0.508
 tactic-0.501
 Where-0.483
rified-0.480
 matter-0.480
Token she
Token ID673
Feature activation+2.38e-02

Pos logit contributions
 unlike+0.078
 tactic+0.078
rified+0.075
 matter+0.074
 Where+0.073
Neg logit contributions
 next-0.035
 put-0.034
 blow-0.034
 waves-0.033
 bat-0.033
Token was
Token ID373
Feature activation-7.77e-02

Pos logit contributions
ans+0.037
 dads+0.033
 next+0.033
ites+0.032
irts+0.032
Neg logit contributions
 unlike-0.035
rified-0.031
 spec-0.031
 matter-0.031
 terribly-0.031
Token back
Token ID736
Feature activation-1.03e-02

Pos logit contributions
 unlike+0.123
 Tears+0.115
 terribly+0.114
 kinda+0.111
 blonde+0.111
Neg logit contributions
ans-0.109
 goes-0.100
 waves-0.099
 next-0.099
ites-0.097
Token on
Token ID319
Feature activation-4.01e-02

Pos logit contributions
 unlike+0.019
 tactic+0.018
 Tears+0.017
 matter+0.017
rified+0.017
Neg logit contributions
ans-0.012
 waves-0.011
 blow-0.011
 next-0.011
 bat-0.011
Token to
Token ID284
Feature activation-5.85e-02

Pos logit contributions
 unlike+0.131
 tactic+0.126
 matter+0.124
 Tears+0.123
 blonde+0.122
Neg logit contributions
ans-0.046
 waves-0.039
 blow-0.037
 bat-0.036
ck-0.035
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.110
 unlike+0.108
 tactic+0.105
mate+0.104
 disgust+0.103
Neg logit contributions
ans-0.063
ites-0.059
 next-0.055
irts-0.054
 dads-0.052
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.240
 waves+0.235
 put+0.232
 bat+0.229
 blow+0.228
Neg logit contributions
 unlike-0.506
 tactic-0.497
 Where-0.484
rified-0.479
 matter-0.478
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.104
 next+0.091
 waves+0.086
 bat+0.086
ites+0.085
Neg logit contributions
 unlike-0.197
 tactic-0.188
 matter-0.187
 Tears-0.186
 Where-0.182
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.243
 waves+0.238
 put+0.237
 blow+0.235
ans+0.234
Neg logit contributions
 unlike-0.500
 tactic-0.492
 Where-0.475
 matter-0.472
rified-0.472
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.109
 next+0.096
 waves+0.091
 bat+0.090
ites+0.089
Neg logit contributions
 unlike-0.193
 matter-0.183
 tactic-0.182
 Tears-0.181
 Where-0.176
Token until
Token ID1566
Feature activation-3.86e-02

Pos logit contributions
 next+0.235
 waves+0.230
 put+0.229
 blow+0.227
ans+0.225
Neg logit contributions
 unlike-0.508
 tactic-0.501
 Where-0.483
rified-0.480
 matter-0.480
Token she
Token ID673
Feature activation+2.38e-02

Pos logit contributions
 unlike+0.078
 tactic+0.078
rified+0.075
 matter+0.074
 Where+0.073
Neg logit contributions
 next-0.035
 put-0.034
 blow-0.034
 waves-0.033
 bat-0.033
Token was
Token ID373
Feature activation-7.77e-02

Pos logit contributions
ans+0.037
 dads+0.033
 next+0.033
ites+0.032
irts+0.032
Neg logit contributions
 unlike-0.035
rified-0.031
 spec-0.031
 matter-0.031
 terribly-0.031
Token back
Token ID736
Feature activation-1.03e-02

Pos logit contributions
 unlike+0.123
 Tears+0.115
 terribly+0.114
 kinda+0.111
 blonde+0.111
Neg logit contributions
ans-0.109
 goes-0.100
 waves-0.099
 next-0.099
ites-0.097
Token on
Token ID319
Feature activation-4.01e-02

Pos logit contributions
 unlike+0.019
 tactic+0.018
 Tears+0.017
 matter+0.017
rified+0.017
Neg logit contributions
ans-0.012
 waves-0.011
 blow-0.011
 next-0.011
 bat-0.011
Token,
Token ID11
Feature activation-4.07e-02

Pos logit contributions
 next+0.068
 put+0.065
 blow+0.063
 waves+0.062
 bat+0.062
Neg logit contributions
 unlike-0.141
 tactic-0.141
 matter-0.137
rified-0.136
 Where-0.133
Token "
Token ID366
Feature activation-9.55e-02

Pos logit contributions
 unlike+0.047
 Where+0.041
 Tears+0.040
 tactic+0.040
 alas+0.038
Neg logit contributions
ans-0.078
 waves-0.073
ites-0.072
ck-0.072
 bat-0.072
TokenYou
Token ID1639
Feature activation-1.14e-03

Pos logit contributions
 unlike+0.160
 tactic+0.155
 matter+0.150
rified+0.149
 Where+0.148
Neg logit contributions
ans-0.120
 next-0.118
 waves-0.116
 bat-0.115
 blow-0.115
Token have
Token ID423
Feature activation+1.36e-01

Pos logit contributions
 unlike+0.002
 tactic+0.002
rified+0.002
 Tears+0.002
 Where+0.002
Neg logit contributions
 put-0.002
 next-0.002
ans-0.002
 blow-0.002
 waves-0.002
Token to
Token ID284
Feature activation-5.89e-02

Pos logit contributions
ans+0.149
 next+0.148
 waves+0.144
 blow+0.144
 put+0.143
Neg logit contributions
 unlike-0.249
 tactic-0.243
rified-0.234
 matter-0.234
 Where-0.232
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.114
 tactic+0.111
 unlike+0.109
 disgust+0.106
mate+0.105
Neg logit contributions
ans-0.057
ites-0.056
 waves-0.054
 next-0.052
 goes-0.051
Token in
Token ID287
Feature activation-4.76e-02

Pos logit contributions
ans+0.264
 next+0.258
 blow+0.254
 waves+0.251
 put+0.251
Neg logit contributions
 unlike-0.471
 tactic-0.460
 matter-0.445
rified-0.444
 Tears-0.442
Token the
Token ID262
Feature activation+6.11e-02

Pos logit contributions
 unlike+0.097
 tactic+0.095
 matter+0.093
 Tears+0.091
 Where+0.090
Neg logit contributions
ans-0.043
 waves-0.040
 blow-0.039
 next-0.039
 bat-0.038
Token pipe
Token ID12656
Feature activation-3.50e-02

Pos logit contributions
 next+0.082
 waves+0.081
 blow+0.079
ans+0.079
 bat+0.078
Neg logit contributions
 unlike-0.100
 tactic-0.098
rified-0.093
mate-0.092
 matter-0.091
Token really
Token ID1107
Feature activation-7.62e-02

Pos logit contributions
 unlike+0.066
 tactic+0.063
rified+0.062
 Tears+0.061
 matter+0.061
Neg logit contributions
ans-0.036
ck-0.034
 waves-0.034
 blow-0.033
 next-0.033
Token fast
Token ID3049
Feature activation+1.00e-01

Pos logit contributions
 unlike+0.122
 terribly+0.113
 matter+0.113
 alas+0.113
 slightly+0.112
Neg logit contributions
ans-0.100
 waves-0.091
 dads-0.089
 pige-0.089
 bat-0.088
Token mouth
Token ID5422
Feature activation-1.29e-01

Pos logit contributions
 unlike+0.060
 tactic+0.059
rified+0.057
 Where+0.056
 matter+0.056
Neg logit contributions
 next-0.045
ans-0.045
 waves-0.044
 blow-0.044
 put-0.044
Token again
Token ID757
Feature activation-5.88e-02

Pos logit contributions
 tactic+0.247
 unlike+0.246
 matter+0.238
 Where+0.236
rified+0.236
Neg logit contributions
 next-0.132
 put-0.126
 blow-0.123
 bat-0.122
 waves-0.122
Token to
Token ID284
Feature activation-5.85e-02

Pos logit contributions
 unlike+0.131
 tactic+0.126
 matter+0.124
 Tears+0.123
 blonde+0.122
Neg logit contributions
ans-0.046
 waves-0.039
 blow-0.037
 bat-0.036
ck-0.035
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
 matter+0.110
 unlike+0.108
 tactic+0.105
mate+0.104
 disgust+0.103
Neg logit contributions
ans-0.063
ites-0.059
 next-0.055
irts-0.054
 dads-0.052
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.240
 waves+0.235
 put+0.232
 bat+0.229
 blow+0.228
Neg logit contributions
 unlike-0.506
 tactic-0.497
 Where-0.484
rified-0.479
 matter-0.478
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.104
 next+0.091
 waves+0.086
 bat+0.086
ites+0.085
Neg logit contributions
 unlike-0.197
 tactic-0.188
 matter-0.187
 Tears-0.186
 Where-0.182
Token and
Token ID290
Feature activation+1.00e-01

Pos logit contributions
 next+0.243
 waves+0.238
 put+0.237
 blow+0.235
ans+0.234
Neg logit contributions
 unlike-0.500
 tactic-0.492
 Where-0.475
 matter-0.472
rified-0.472
Token blow
Token ID6611
Feature activation+2.54e-01

Pos logit contributions
ans+0.109
 next+0.096
 waves+0.091
 bat+0.090
ites+0.089
Neg logit contributions
 unlike-0.193
 matter-0.183
 tactic-0.182
 Tears-0.181
 Where-0.176
Token until
Token ID1566
Feature activation-3.86e-02

Pos logit contributions
 next+0.235
 waves+0.230
 put+0.229
 blow+0.227
ans+0.225
Neg logit contributions
 unlike-0.508
 tactic-0.501
 Where-0.483
rified-0.480
 matter-0.480
Token she
Token ID673
Feature activation+2.38e-02

Pos logit contributions
 unlike+0.078
 tactic+0.078
rified+0.075
 matter+0.074
 Where+0.073
Neg logit contributions
 next-0.035
 put-0.034
 blow-0.034
 waves-0.033
 bat-0.033
Token was
Token ID373
Feature activation-7.77e-02

Pos logit contributions
ans+0.037
 dads+0.033
 next+0.033
ites+0.032
irts+0.032
Neg logit contributions
 unlike-0.035
rified-0.031
 spec-0.031
 matter-0.031
 terribly-0.031
Token listen
Token ID6004
Feature activation+1.09e-01

Pos logit contributions
ans+0.027
 next+0.026
 waves+0.026
ites+0.024
 blow+0.024
Neg logit contributions
 matter-0.046
 tactic-0.045
 unlike-0.044
 Where-0.043
rified-0.042
Token.
Token ID13
Feature activation-2.20e-02

Pos logit contributions
ans+0.116
 waves+0.112
 blow+0.111
 bat+0.110
 next+0.110
Neg logit contributions
 unlike-0.202
 tactic-0.196
 matter-0.189
rified-0.189
 Tears-0.189
Token She
Token ID1375
Feature activation-3.99e-02

Pos logit contributions
 unlike+0.035
 tactic+0.035
 matter+0.033
rified+0.033
mate+0.032
Neg logit contributions
 next-0.030
 bat-0.029
 put-0.029
 waves-0.028
ans-0.028
Token threw
Token ID9617
Feature activation+1.53e-01

Pos logit contributions
 unlike+0.063
 matter+0.062
rified+0.060
 tactic+0.060
 terribly+0.059
Neg logit contributions
ans-0.051
ites-0.050
 next-0.048
 blow-0.046
irts-0.045
Token a
Token ID257
Feature activation-1.02e-02

Pos logit contributions
 next+0.116
 put+0.112
 waves+0.110
 bat+0.109
 blow+0.108
Neg logit contributions
 unlike-0.339
 tactic-0.331
 Where-0.316
rified-0.316
 Tears-0.315
Token tant
Token ID24246
Feature activation-3.43e-01

Pos logit contributions
 unlike+0.012
 tactic+0.011
rified+0.010
 Where+0.010
 Tears+0.010
Neg logit contributions
 bat-0.020
 blow-0.019
 put-0.019
 waves-0.019
 next-0.019
Tokenrum
Token ID6582
Feature activation-1.71e-01

Pos logit contributions
ifying+0.319
 unlike+0.319
 Tears+0.303
rified+0.297
 Where+0.296
Neg logit contributions
ck-0.694
ans-0.684
 next-0.660
is-0.653
 blow-0.648
Token in
Token ID287
Feature activation-5.14e-02

Pos logit contributions
 unlike+0.393
 tactic+0.380
rified+0.376
 Tears+0.374
 Where+0.372
Neg logit contributions
ans-0.110
 blow-0.101
 waves-0.099
 bat-0.099
 put-0.097
Token the
Token ID262
Feature activation+6.84e-02

Pos logit contributions
 unlike+0.114
 tactic+0.111
rified+0.108
 Where+0.107
 matter+0.107
Neg logit contributions
 next-0.038
 put-0.036
ans-0.036
 bat-0.036
 blow-0.035
Token store
Token ID3650
Feature activation+2.81e-02

Pos logit contributions
ans+0.053
 blow+0.050
 next+0.050
 put+0.049
ck+0.048
Neg logit contributions
 tactic-0.143
 unlike-0.142
rified-0.139
 Where-0.139
 matter-0.138
Token.
Token ID13
Feature activation-2.22e-02

Pos logit contributions
ans+0.022
 blow+0.021
 waves+0.020
 bat+0.020
ck+0.019
Neg logit contributions
 unlike-0.060
 tactic-0.058
rified-0.057
 Tears-0.057
 matter-0.056
Token them
Token ID606
Feature activation+2.10e-02

Pos logit contributions
ans+0.042
 waves+0.041
 blow+0.041
 bat+0.040
 next+0.040
Neg logit contributions
 unlike-0.064
 tactic-0.061
 matter-0.060
 Tears-0.059
 Where-0.058
Token and
Token ID290
Feature activation+1.10e-01

Pos logit contributions
ans+0.022
 waves+0.021
 blow+0.021
 next+0.020
 bat+0.020
Neg logit contributions
 unlike-0.040
 tactic-0.038
 matter-0.037
 Tears-0.037
rified-0.037
Token never
Token ID1239
Feature activation+9.05e-02

Pos logit contributions
ans+0.110
 next+0.104
 waves+0.103
 blow+0.102
 bat+0.101
Neg logit contributions
 unlike-0.213
 tactic-0.206
 matter-0.200
 Where-0.198
 Tears-0.198
Token threw
Token ID9617
Feature activation+1.59e-01

Pos logit contributions
 next+0.079
 put+0.078
 blow+0.077
ans+0.077
 waves+0.077
Neg logit contributions
 unlike-0.186
 tactic-0.182
rified-0.176
 Where-0.174
 Tears-0.173
Token a
Token ID257
Feature activation-1.79e-02

Pos logit contributions
 next+0.140
 put+0.133
ans+0.133
 waves+0.132
 bat+0.128
Neg logit contributions
 tactic-0.330
 unlike-0.329
rified-0.315
 Tears-0.312
 Where-0.310
Token tant
Token ID24246
Feature activation-3.42e-01

Pos logit contributions
 unlike+0.031
 tactic+0.029
rified+0.029
 Where+0.029
 Tears+0.029
Neg logit contributions
 bat-0.023
 put-0.022
 blow-0.022
 next-0.022
 waves-0.022
Tokenrum
Token ID6582
Feature activation-1.60e-01

Pos logit contributions
 unlike+0.298
 Where+0.279
ifying+0.278
rified+0.271
 Tears+0.270
Neg logit contributions
ans-0.719
ck-0.699
 next-0.687
 blow-0.669
is-0.669
Token again
Token ID757
Feature activation-5.70e-02

Pos logit contributions
 unlike+0.324
 tactic+0.315
rified+0.307
 Where+0.305
 Tears+0.305
Neg logit contributions
ans-0.144
 next-0.138
 blow-0.138
 waves-0.137
 bat-0.136
Token.
Token ID13
Feature activation-1.85e-02

Pos logit contributions
 unlike+0.112
 tactic+0.108
 matter+0.106
 Tears+0.105
 Where+0.104
Neg logit contributions
ans-0.056
 waves-0.053
 blow-0.052
 next-0.051
 bat-0.051
Token<|endoftext|>
Token ID50256
Feature activation+4.39e-02

Pos logit contributions
 unlike+0.035
 tactic+0.034
rified+0.033
 Where+0.032
 Tears+0.032
Neg logit contributions
ans-0.020
 next-0.019
 waves-0.019
 blow-0.019
 put-0.019
TokenOnce
Token ID7454
Feature activation-8.87e-02

Pos logit contributions
 next+0.035
ans+0.035
 waves+0.034
 put+0.034
 blow+0.033
Neg logit contributions
 unlike-0.093
 tactic-0.091
rified-0.088
 matter-0.088
 Where-0.087
Token to
Token ID284
Feature activation-6.57e-02

Pos logit contributions
 unlike+0.097
 tactic+0.094
rified+0.094
 matter+0.091
 Where+0.090
Neg logit contributions
 next-0.054
 put-0.053
 blow-0.052
 bat-0.052
 waves-0.051
Token cry
Token ID3960
Feature activation+1.48e-02

Pos logit contributions
 unlike+0.118
 tactic+0.114
rified+0.111
 Where+0.110
 Tears+0.109
Neg logit contributions
 next-0.075
ans-0.074
 put-0.074
 blow-0.074
 bat-0.073
Token and
Token ID290
Feature activation+1.03e-01

Pos logit contributions
 next+0.008
ans+0.008
 blow+0.008
 put+0.008
 bat+0.008
Neg logit contributions
 unlike-0.035
 tactic-0.035
rified-0.034
 matter-0.033
 Where-0.033
Token threw
Token ID9617
Feature activation+1.52e-01

Pos logit contributions
 put+0.096
 next+0.088
 blow+0.085
 bat+0.085
 waves+0.081
Neg logit contributions
 unlike-0.224
ifying-0.215
 tactic-0.214
rified-0.211
eway-0.210
Token a
Token ID257
Feature activation-1.29e-02

Pos logit contributions
 next+0.123
 put+0.122
 bat+0.113
 each+0.109
 and+0.108
Neg logit contributions
 unlike-0.341
 tactic-0.327
eway-0.320
ifying-0.320
rified-0.316
Token tant
Token ID24246
Feature activation-3.41e-01

Pos logit contributions
 unlike+0.023
 avail+0.022
rified+0.022
eway+0.022
 alas+0.022
Neg logit contributions
 bat-0.018
 se-0.017
 put-0.017
 swing-0.017
 blow-0.017
Tokenrum
Token ID6582
Feature activation-1.70e-01

Pos logit contributions
ifying+0.317
 unlike+0.309
 Where+0.294
 Tears+0.294
rified+0.293
Neg logit contributions
ck-0.690
ans-0.672
 next-0.655
 blow-0.650
 put-0.645
Token.
Token ID13
Feature activation-2.96e-02

Pos logit contributions
 unlike+0.373
 tactic+0.373
 matter+0.361
rified+0.352
 Where+0.351
Neg logit contributions
 next-0.133
 put-0.120
 waves-0.117
 blow-0.117
 bat-0.115
Token She
Token ID1375
Feature activation-4.75e-02

Pos logit contributions
 tactic+0.050
mate+0.048
 unlike+0.048
 matter+0.048
eway+0.047
Neg logit contributions
 next-0.039
 put-0.038
 goes-0.036
 and-0.036
 bat-0.036
Token didn
Token ID1422
Feature activation-9.02e-02

Pos logit contributions
 unlike+0.082
 tactic+0.080
 matter+0.078
rified+0.078
 Tears+0.077
Neg logit contributions
ans-0.057
 next-0.055
 waves-0.054
 blow-0.054
 bat-0.053
Token't
Token ID470
Feature activation-2.73e-03

Pos logit contributions
 unlike+0.084
 terribly+0.076
 Where+0.076
 Tears+0.075
 tactic+0.073
Neg logit contributions
ans-0.192
 waves-0.182
 dads-0.177
ites-0.172
 dove-0.172
Token sad
Token ID6507
Feature activation-1.14e-01

Pos logit contributions
ans+0.138
 next+0.132
 waves+0.131
 blow+0.130
 bat+0.129
Neg logit contributions
 unlike-0.168
 tactic-0.162
 Tears-0.158
 matter-0.157
rified-0.157
Token.
Token ID13
Feature activation-2.45e-02

Pos logit contributions
 unlike+0.273
 tactic+0.273
ifying+0.264
rified+0.263
eway+0.262
Neg logit contributions
 next-0.075
 and-0.066
 put-0.063
 each-0.056
 blow-0.054
Token She
Token ID1375
Feature activation-4.34e-02

Pos logit contributions
 unlike+0.044
 tactic+0.043
 matter+0.041
rified+0.041
ifying+0.041
Neg logit contributions
 next-0.029
 put-0.027
ans-0.027
 waves-0.027
 bat-0.026
Token threw
Token ID9617
Feature activation+1.53e-01

Pos logit contributions
 unlike+0.083
 tactic+0.082
rified+0.078
eway+0.078
 Where+0.077
Neg logit contributions
 put-0.047
 next-0.045
 blow-0.043
 waves-0.043
ans-0.043
Token a
Token ID257
Feature activation-1.29e-02

Pos logit contributions
 next+0.086
 put+0.086
 bat+0.079
 waves+0.078
 blow+0.076
Neg logit contributions
 unlike-0.372
 tactic-0.361
 Where-0.349
eway-0.347
rified-0.346
Token tant
Token ID24246
Feature activation-3.41e-01

Pos logit contributions
 unlike+0.017
 Where+0.016
 avail+0.016
rified+0.016
 tactic+0.016
Neg logit contributions
 bat-0.022
 put-0.022
 blow-0.022
 se-0.021
 slide-0.021
Tokenrum
Token ID6582
Feature activation-1.69e-01

Pos logit contributions
ifying+0.316
 unlike+0.302
 Where+0.294
 Tears+0.290
rified+0.287
Neg logit contributions
ck-0.697
ans-0.677
 next-0.651
is-0.646
 blow-0.641
Token and
Token ID290
Feature activation+1.03e-01

Pos logit contributions
 unlike+0.413
 tactic+0.406
rified+0.395
 matter+0.394
 Where+0.392
Neg logit contributions
 next-0.079
ans-0.079
 waves-0.074
 blow-0.073
 put-0.073
Token cried
Token ID16896
Feature activation+6.27e-02

Pos logit contributions
 next+0.110
ans+0.109
 put+0.109
 blow+0.108
 waves+0.107
Neg logit contributions
 unlike-0.189
 tactic-0.186
rified-0.178
 Where-0.177
 matter-0.177
Token loudly
Token ID23112
Feature activation-4.99e-02

Pos logit contributions
 next+0.048
 and+0.044
 put+0.043
 bat+0.042
 each+0.041
Neg logit contributions
 tactic-0.144
 unlike-0.142
 matter-0.137
eway-0.137
 Where-0.134
Token.
Token ID13
Feature activation-2.61e-02

Pos logit contributions
 tactic+0.113
 unlike+0.111
 matter+0.109
 kinda+0.105
 perspective+0.105
Neg logit contributions
 next-0.039
 put-0.035
 and-0.033
 blow-0.033
 waves-0.032
Token her
Token ID607
Feature activation-2.97e-02

Pos logit contributions
 unlike+0.063
 tactic+0.061
ifying+0.060
 matter+0.059
eway+0.059
Neg logit contributions
 put-0.023
 next-0.023
 and-0.022
 each-0.022
 waves-0.022
Token feet
Token ID3625
Feature activation+5.46e-02

Pos logit contributions
 unlike+0.061
 tactic+0.059
 Where+0.058
rified+0.057
eway+0.057
Neg logit contributions
 next-0.030
 put-0.028
 bat-0.027
 waves-0.027
 blow-0.027
Token and
Token ID290
Feature activation+1.03e-01

Pos logit contributions
ans+0.027
 waves+0.025
 blow+0.023
 next+0.023
 bat+0.022
Neg logit contributions
 unlike-0.133
 tactic-0.130
rified-0.127
 Tears-0.127
 matter-0.126
Token threw
Token ID9617
Feature activation+1.53e-01

Pos logit contributions
 put+0.079
 blow+0.073
 next+0.073
 waves+0.072
 bat+0.071
Neg logit contributions
 unlike-0.228
 tactic-0.228
 Where-0.218
eway-0.218
 matter-0.216
Token a
Token ID257
Feature activation-1.16e-02

Pos logit contributions
 put+0.119
 next+0.118
 waves+0.113
 bat+0.110
 blow+0.108
Neg logit contributions
 unlike-0.337
 tactic-0.331
rified-0.316
 Where-0.315
eway-0.314
Token tant
Token ID24246
Feature activation-3.41e-01

Pos logit contributions
 unlike+0.010
 tactic+0.009
 Where+0.009
rified+0.009
 Tears+0.009
Neg logit contributions
 bat-0.025
 put-0.025
 blow-0.024
 se-0.024
 swing-0.024
Tokenrum
Token ID6582
Feature activation-1.69e-01

Pos logit contributions
ifying+0.295
 unlike+0.293
 Tears+0.281
 Where+0.278
rified+0.275
Neg logit contributions
ck-0.710
ans-0.697
 next-0.671
 put-0.660
 blow-0.660
Token.
Token ID13
Feature activation-2.40e-02

Pos logit contributions
 unlike+0.363
 tactic+0.362
 matter+0.351
rified+0.343
 Where+0.342
Neg logit contributions
 next-0.134
 waves-0.123
 put-0.122
ans-0.122
 blow-0.120
Token She
Token ID1375
Feature activation-4.24e-02

Pos logit contributions
 tactic+0.038
 unlike+0.038
 matter+0.037
mate+0.035
rified+0.035
Neg logit contributions
 next-0.032
 put-0.032
 waves-0.031
ans-0.031
 bat-0.031
Token grabbed
Token ID13176
Feature activation+9.34e-02

Pos logit contributions
 unlike+0.085
 tactic+0.083
rified+0.080
 matter+0.080
 Where+0.080
Neg logit contributions
ans-0.039
 next-0.038
 waves-0.037
 blow-0.037
 put-0.037
Token the
Token ID262
Feature activation+6.90e-02

Pos logit contributions
 next+0.077
 put+0.076
ans+0.075
 blow+0.075
 waves+0.074
Neg logit contributions
 unlike-0.196
 tactic-0.193
rified-0.185
 matter-0.185
 Where-0.184
Token his
Token ID465
Feature activation-7.65e-02

Pos logit contributions
 next+0.014
 put+0.014
 bat+0.014
 waves+0.014
ans+0.014
Neg logit contributions
 unlike-0.020
 tactic-0.019
 matter-0.019
 Tears-0.019
 perspective-0.018
Token feet
Token ID3625
Feature activation+5.46e-02

Pos logit contributions
 tactic+0.118
 unlike+0.117
 Tears+0.111
 blonde+0.110
 matter+0.109
Neg logit contributions
ans-0.100
 blow-0.096
 next-0.095
 waves-0.094
is-0.094
Token and
Token ID290
Feature activation+1.06e-01

Pos logit contributions
ans+0.043
 waves+0.038
 dads+0.035
ites+0.034
irts+0.034
Neg logit contributions
 unlike-0.118
 terribly-0.114
 slightly-0.114
 tactic-0.113
rified-0.112
Token threw
Token ID9617
Feature activation+1.58e-01

Pos logit contributions
ans+0.103
 next+0.092
is+0.089
 blow+0.089
ites+0.088
Neg logit contributions
 unlike-0.209
 Tears-0.200
 tactic-0.199
rified-0.198
 blonde-0.195
Token a
Token ID257
Feature activation-1.48e-02

Pos logit contributions
 next+0.160
 waves+0.157
 put+0.156
ans+0.155
 bat+0.155
Neg logit contributions
 unlike-0.306
 tactic-0.300
rified-0.285
 matter-0.285
 Tears-0.284
Token tant
Token ID24246
Feature activation-3.40e-01

Pos logit contributions
 unlike+0.028
 tactic+0.027
rified+0.027
 Where+0.027
 Tears+0.026
Neg logit contributions
 bat-0.016
 put-0.016
 blow-0.016
 waves-0.015
 se-0.015
Tokenrum
Token ID6582
Feature activation-1.63e-01

Pos logit contributions
 unlike+0.270
ifying+0.250
 Where+0.249
 Tears+0.242
rified+0.234
Neg logit contributions
ans-0.745
ck-0.728
 next-0.714
 blow-0.698
 put-0.692
Token.
Token ID13
Feature activation-2.24e-02

Pos logit contributions
 unlike+0.331
 tactic+0.322
rified+0.315
 Tears+0.312
 Where+0.312
Neg logit contributions
ans-0.144
 blow-0.138
 next-0.137
 waves-0.136
 put-0.136
Token
Token ID198
Feature activation+2.10e-02

Pos logit contributions
 unlike+0.045
 tactic+0.045
 matter+0.045
rified+0.043
 perspective+0.042
Neg logit contributions
 next-0.021
 bat-0.020
 put-0.020
 waves-0.019
ans-0.019
Token
Token ID198
Feature activation+2.04e-02

Pos logit contributions
 waves+0.027
ans+0.027
 blow+0.026
 swing+0.024
 goes+0.024
Neg logit contributions
 Where-0.035
 Tears-0.033
 tactic-0.033
 unlike-0.033
 Was-0.032
Token"
Token ID1
Feature activation+6.18e-02

Pos logit contributions
 waves+0.021
ans+0.020
 blow+0.019
 goes+0.018
 put+0.018
Neg logit contributions
 Where-0.038
mate-0.037
 Tears-0.037
 unlike-0.037
ifying-0.037
Token.
Token ID13
Feature activation-2.15e-02

Pos logit contributions
 next+0.086
 put+0.079
ans+0.078
 bat+0.077
 blow+0.076
Neg logit contributions
 unlike-0.236
 tactic-0.230
rified-0.225
 matter-0.223
 Where-0.221
Token He
Token ID679
Feature activation-1.02e-01

Pos logit contributions
 unlike+0.036
 tactic+0.035
 matter+0.035
ifying+0.034
rified+0.033
Neg logit contributions
 next-0.029
 goes-0.028
 put-0.028
 and-0.028
 bat-0.027
Token kept
Token ID4030
Feature activation+3.60e-02

Pos logit contributions
 unlike+0.173
 tactic+0.165
rified+0.162
 matter+0.162
 Tears+0.159
Neg logit contributions
ans-0.128
 next-0.123
 blow-0.120
 waves-0.120
 bat-0.119
Token throwing
Token ID9644
Feature activation+1.41e-01

Pos logit contributions
ans+0.050
 waves+0.046
ites+0.045
 next+0.045
 blow+0.045
Neg logit contributions
 unlike-0.055
 Tears-0.053
 tactic-0.053
 matter-0.051
 Where-0.051
Token a
Token ID257
Feature activation-1.09e-02

Pos logit contributions
 next+0.129
ans+0.126
 put+0.126
 waves+0.126
 bat+0.124
Neg logit contributions
 unlike-0.286
 tactic-0.279
rified-0.269
 Where-0.268
 matter-0.267
Token tant
Token ID24246
Feature activation-3.40e-01

Pos logit contributions
 unlike+0.021
 tactic+0.019
rified+0.019
 Tears+0.019
 Where+0.019
Neg logit contributions
 bat-0.013
 put-0.012
 next-0.012
 blow-0.012
 slide-0.012
Tokenrum
Token ID6582
Feature activation-1.69e-01

Pos logit contributions
 unlike+0.269
ifying+0.252
 Tears+0.246
rified+0.245
 Where+0.244
Neg logit contributions
ans-0.734
ck-0.721
 next-0.709
 blow-0.699
 put-0.693
Token.
Token ID13
Feature activation-1.84e-02

Pos logit contributions
 unlike+0.361
 tactic+0.355
 matter+0.344
rified+0.343
 Where+0.340
Neg logit contributions
 next-0.134
ans-0.130
 waves-0.127
 put-0.126
 blow-0.125
Token His
Token ID2399
Feature activation-1.20e-01

Pos logit contributions
 tactic+0.029
 unlike+0.029
 matter+0.028
rified+0.028
 spec+0.027
Neg logit contributions
 next-0.026
 put-0.025
 bat-0.025
 goes-0.024
 waves-0.024
Token mom
Token ID1995
Feature activation+2.74e-02

Pos logit contributions
 unlike+0.284
 tactic+0.278
rified+0.273
 matter+0.266
eway+0.265
Neg logit contributions
 next-0.086
 bat-0.080
 waves-0.077
 put-0.076
 blow-0.071
Token was
Token ID373
Feature activation-8.11e-02

Pos logit contributions
ans+0.035
 waves+0.031
irts+0.030
ites+0.030
 bat+0.030
Neg logit contributions
 unlike-0.045
rified-0.045
 matter-0.044
 tactic-0.044
 Tears-0.043
Token stubborn
Token ID24444
Feature activation-3.25e-02

Pos logit contributions
 unlike+0.031
 tactic+0.030
rified+0.029
 matter+0.028
 Where+0.028
Neg logit contributions
ans-0.025
 next-0.024
 waves-0.024
 blow-0.024
 put-0.024
Token.
Token ID13
Feature activation-2.16e-02

Pos logit contributions
 unlike+0.066
 tactic+0.065
rified+0.064
 matter+0.063
 Where+0.063
Neg logit contributions
ans-0.029
 blow-0.027
 waves-0.026
 bat-0.026
 next-0.026
Token She
Token ID1375
Feature activation-3.23e-02

Pos logit contributions
 unlike+0.037
 tactic+0.037
 matter+0.036
rified+0.035
mate+0.035
Neg logit contributions
 next-0.027
 put-0.025
 bat-0.025
 goes-0.025
 waves-0.025
Token throws
Token ID12542
Feature activation+1.55e-01

Pos logit contributions
 unlike+0.051
 matter+0.051
rified+0.049
 Tears+0.049
 kinda+0.049
Neg logit contributions
ans-0.040
 next-0.039
 bat-0.038
ites-0.037
line-0.037
Token a
Token ID257
Feature activation-8.38e-03

Pos logit contributions
 next+0.119
ans+0.118
 waves+0.115
 put+0.115
 blow+0.114
Neg logit contributions
 unlike-0.333
 tactic-0.326
rified-0.315
 matter-0.314
 Where-0.314
Token tant
Token ID24246
Feature activation-3.39e-01

Pos logit contributions
 unlike+0.009
 tactic+0.009
rified+0.008
 Where+0.008
 Tears+0.008
Neg logit contributions
 next-0.015
 bat-0.015
 blow-0.015
ans-0.015
 put-0.015
Tokenrum
Token ID6582
Feature activation-1.64e-01

Pos logit contributions
 unlike+0.282
ifying+0.280
 Where+0.262
 Tears+0.258
rified+0.257
Neg logit contributions
ans-0.722
ck-0.713
 next-0.692
is-0.674
 blow-0.673
Token.
Token ID13
Feature activation-2.22e-02

Pos logit contributions
 unlike+0.323
rified+0.309
 Tears+0.309
 tactic+0.309
 Where+0.306
Neg logit contributions
ans-0.158
 blow-0.148
 bat-0.147
 waves-0.145
 put-0.144
Token She
Token ID1375
Feature activation-3.34e-02

Pos logit contributions
 unlike+0.038
 tactic+0.038
 matter+0.037
rified+0.036
mate+0.035
Neg logit contributions
 next-0.028
 put-0.026
 bat-0.026
 blow-0.026
 waves-0.026
Token kicks
Token ID17364
Feature activation+1.08e-01

Pos logit contributions
 unlike+0.057
 spec+0.056
 matter+0.056
 Tears+0.056
 kinda+0.056
Neg logit contributions
 next-0.035
ans-0.035
line-0.033
is-0.032
ites-0.032
Token and
Token ID290
Feature activation+1.09e-01

Pos logit contributions
 next+0.059
 put+0.054
 waves+0.051
 blow+0.050
 bat+0.050
Neg logit contributions
 unlike-0.264
 tactic-0.257
 Where-0.248
rified-0.248
 matter-0.247
Token kept
Token ID4030
Feature activation+4.04e-02

Pos logit contributions
 unlike+0.065
 matter+0.063
 tactic+0.062
rified+0.062
 Tears+0.062
Neg logit contributions
ans-0.048
 waves-0.045
 next-0.045
ites-0.044
 blow-0.043
Token crying
Token ID13774
Feature activation-1.52e-02

Pos logit contributions
ans+0.056
 next+0.054
 waves+0.054
 blow+0.053
 bat+0.053
Neg logit contributions
 unlike-0.063
 tactic-0.061
 Tears-0.058
 matter-0.058
 Where-0.058
Token and
Token ID290
Feature activation+1.08e-01

Pos logit contributions
 unlike+0.039
 tactic+0.039
 matter+0.038
 Where+0.038
eway+0.037
Neg logit contributions
 next-0.006
 put-0.005
 bat-0.004
 blow-0.004
 waves-0.004
Token throwing
Token ID9644
Feature activation+1.44e-01

Pos logit contributions
 next+0.132
 put+0.129
 blow+0.125
 bat+0.124
 waves+0.120
Neg logit contributions
 unlike-0.191
 tactic-0.187
rified-0.179
ifying-0.178
 Where-0.178
Token a
Token ID257
Feature activation-1.09e-02

Pos logit contributions
 put+0.114
 next+0.110
 bat+0.106
 and+0.104
 each+0.102
Neg logit contributions
 unlike-0.323
 tactic-0.315
eway-0.308
 Where-0.304
rified-0.304
Token tant
Token ID24246
Feature activation-3.39e-01

Pos logit contributions
 unlike+0.015
rified+0.014
 Where+0.013
 tactic+0.013
eway+0.013
Neg logit contributions
 bat-0.019
 put-0.018
 blow-0.018
 next-0.018
 se-0.017
Tokenrum
Token ID6582
Feature activation-1.66e-01

Pos logit contributions
 unlike+0.302
ifying+0.302
 Where+0.284
rified+0.283
 Tears+0.281
Neg logit contributions
ck-0.691
ans-0.675
 next-0.658
 blow-0.652
 put-0.650
Token.
Token ID13
Feature activation-1.80e-02

Pos logit contributions
 tactic+0.377
 unlike+0.371
 matter+0.365
 Where+0.350
 disgust+0.350
Neg logit contributions
 next-0.126
 put-0.108
 and-0.101
 waves-0.101
 blow-0.100
Token
Token ID198
Feature activation+2.59e-02

Pos logit contributions
 matter+0.039
 tactic+0.038
 unlike+0.038
ifying+0.037
mate+0.037
Neg logit contributions
 next-0.017
 put-0.015
 and-0.014
 bat-0.013
 then-0.012
Token
Token ID198
Feature activation+2.40e-02

Pos logit contributions
ans+0.022
 next+0.021
 waves+0.021
 blow+0.021
 put+0.021
Neg logit contributions
 unlike-0.054
 tactic-0.053
rified-0.051
 matter-0.051
 Where-0.051
TokenFinally
Token ID11158
Feature activation-1.28e-01

Pos logit contributions
 waves+0.032
ans+0.032
 blow+0.030
 bat+0.029
 put+0.029
Neg logit contributions
 unlike-0.039
 tactic-0.038
 Tears-0.037
 Where-0.037
 matter-0.037
Token She
Token ID1375
Feature activation-5.20e-02

Pos logit contributions
 unlike+0.050
 tactic+0.049
 matter+0.048
rified+0.046
mate+0.046
Neg logit contributions
 next-0.045
 put-0.043
ans-0.042
 waves-0.042
 goes-0.042
Token cried
Token ID16896
Feature activation+6.23e-02

Pos logit contributions
 matter+0.085
 unlike+0.084
 terribly+0.082
 tactic+0.081
rified+0.080
Neg logit contributions
ans-0.065
ites-0.064
irts-0.058
 waves-0.058
line-0.057
Token and
Token ID290
Feature activation+1.03e-01

Pos logit contributions
 next+0.020
 put+0.017
 bat+0.016
 blow+0.016
ans+0.014
Neg logit contributions
 unlike-0.165
 tactic-0.164
 matter-0.158
 Where-0.158
rified-0.157
Token threw
Token ID9617
Feature activation+1.54e-01

Pos logit contributions
ans+0.075
 next+0.073
 waves+0.071
 blow+0.070
 bat+0.069
Neg logit contributions
 unlike-0.227
 tactic-0.223
rified-0.216
 matter-0.216
 Tears-0.215
Token a
Token ID257
Feature activation-1.10e-02

Pos logit contributions
 put+0.130
 next+0.123
 and+0.123
 each+0.120
 bat+0.116
Neg logit contributions
 unlike-0.342
 tactic-0.330
eway-0.324
ifying-0.321
rified-0.318
Token tant
Token ID24246
Feature activation-3.39e-01

Pos logit contributions
 unlike+0.016
 avail+0.015
rified+0.015
 Where+0.014
 tactic+0.014
Neg logit contributions
 bat-0.019
 se-0.018
 swing-0.018
 put-0.018
 slide-0.018
Tokenrum
Token ID6582
Feature activation-1.69e-01

Pos logit contributions
ifying+0.317
 unlike+0.309
 Tears+0.298
rified+0.294
 Where+0.292
Neg logit contributions
ck-0.680
ans-0.673
 next-0.647
is-0.641
 blow-0.641
Token.
Token ID13
Feature activation-2.90e-02

Pos logit contributions
 tactic+0.386
 unlike+0.382
 matter+0.375
rified+0.360
 disgust+0.359
Neg logit contributions
 next-0.124
 put-0.107
 waves-0.103
 bat-0.101
 blow-0.099
Token Her
Token ID2332
Feature activation-6.85e-02

Pos logit contributions
 tactic+0.047
 unlike+0.046
 matter+0.046
rified+0.044
ifying+0.043
Neg logit contributions
 next-0.040
 put-0.039
 bat-0.038
 blow-0.037
 goes-0.037
Token mom
Token ID1995
Feature activation+3.43e-02

Pos logit contributions
 unlike+0.165
 tactic+0.163
rified+0.163
eway+0.158
 Where+0.157
Neg logit contributions
 next-0.048
 bat-0.043
 put-0.041
 waves-0.040
 se-0.037
Token tried
Token ID3088
Feature activation-4.88e-03

Pos logit contributions
ans+0.043
 waves+0.039
irts+0.038
ites+0.037
 next+0.037
Neg logit contributions
 unlike-0.057
rified-0.057
 matter-0.056
 tactic-0.056
 Tears-0.055
Token surprise
Token ID5975
Feature activation-9.72e-02

Pos logit contributions
 next+0.094
 waves+0.085
 slide+0.082
 bat+0.081
 put+0.080
Neg logit contributions
 unlike-0.124
 tactic-0.120
rified-0.120
 Tears-0.119
eway-0.115
Token.
Token ID13
Feature activation-1.50e-02

Pos logit contributions
 unlike+0.212
 tactic+0.207
rified+0.203
 Tears+0.203
 blonde+0.200
Neg logit contributions
 next-0.078
 put-0.072
 waves-0.070
ans-0.069
 blow-0.068
Token He
Token ID679
Feature activation-9.41e-02

Pos logit contributions
 unlike+0.026
 tactic+0.025
 matter+0.025
 emanating+0.024
 perspective+0.024
Neg logit contributions
 next-0.020
 bat-0.019
 put-0.019
 goes-0.018
 and-0.018
Token threw
Token ID9617
Feature activation+1.59e-01

Pos logit contributions
 unlike+0.172
 tactic+0.167
 matter+0.163
rified+0.163
 Tears+0.160
Neg logit contributions
ans-0.102
 next-0.100
 blow-0.098
 waves-0.097
 bat-0.097
Token a
Token ID257
Feature activation-9.67e-03

Pos logit contributions
 next+0.125
 put+0.124
 waves+0.122
 bat+0.120
ans+0.116
Neg logit contributions
 unlike-0.347
 tactic-0.343
rified-0.326
 matter-0.324
 Tears-0.323
Token tant
Token ID24246
Feature activation-3.39e-01

Pos logit contributions
 unlike+0.018
rified+0.017
 tactic+0.017
 Where+0.017
 Tears+0.017
Neg logit contributions
 bat-0.012
 put-0.011
 blow-0.011
 se-0.011
 slide-0.011
Tokenrum
Token ID6582
Feature activation-1.66e-01

Pos logit contributions
 unlike+0.279
ifying+0.271
 Tears+0.260
 Where+0.257
 blonde+0.255
Neg logit contributions
ans-0.723
ck-0.715
 next-0.697
 blow-0.676
is-0.676
Token.
Token ID13
Feature activation-1.65e-02

Pos logit contributions
 unlike+0.349
 tactic+0.340
rified+0.332
 Where+0.329
 Tears+0.329
Neg logit contributions
ans-0.134
 next-0.129
 blow-0.128
 waves-0.127
 put-0.126
Token He
Token ID679
Feature activation-9.72e-02

Pos logit contributions
 tactic+0.027
 unlike+0.027
 matter+0.026
rified+0.025
 perspective+0.025
Neg logit contributions
 next-0.023
 bat-0.022
 put-0.022
 waves-0.021
 goes-0.021
Token kicked
Token ID12165
Feature activation+1.03e-01

Pos logit contributions
 unlike+0.178
rified+0.170
 matter+0.169
 tactic+0.169
 Tears+0.166
Neg logit contributions
ans-0.103
 next-0.099
 blow-0.095
 waves-0.094
 bat-0.092
Token and
Token ID290
Feature activation+1.11e-01

Pos logit contributions
 next+0.063
 put+0.060
ans+0.059
 waves+0.059
 bat+0.058
Neg logit contributions
 unlike-0.244
 tactic-0.239
rified-0.230
 matter-0.229
 Where-0.229
Token and
Token ID290
Feature activation+1.08e-01

Pos logit contributions
 unlike+0.151
 tactic+0.147
 Tears+0.145
rified+0.145
 Where+0.143
Neg logit contributions
ans-0.034
 blow-0.030
 waves-0.028
 bat-0.027
ck-0.027
Token started
Token ID2067
Feature activation-4.68e-02

Pos logit contributions
ans+0.122
 next+0.115
 waves+0.111
ites+0.109
 bat+0.109
Neg logit contributions
 unlike-0.194
 tactic-0.187
 matter-0.187
 Tears-0.185
rified-0.183
Token to
Token ID284
Feature activation-6.11e-02

Pos logit contributions
 unlike+0.091
 tactic+0.088
rified+0.088
 matter+0.086
ifying+0.084
Neg logit contributions
 next-0.048
 put-0.047
 bat-0.047
 blow-0.047
ans-0.045
Token throw
Token ID3714
Feature activation+2.07e-01

Pos logit contributions
 unlike+0.107
 tactic+0.101
rified+0.100
 Where+0.098
ifying+0.097
Neg logit contributions
 put-0.076
 blow-0.075
 bat-0.075
 next-0.074
 waves-0.073
Token a
Token ID257
Feature activation-8.88e-03

Pos logit contributions
 put+0.157
 next+0.154
 bat+0.150
ans+0.148
 waves+0.148
Neg logit contributions
 unlike-0.465
 tactic-0.451
eway-0.436
 Where-0.432
 Tears-0.430
Token tant
Token ID24246
Feature activation-3.39e-01

Pos logit contributions
 unlike+0.016
 Tears+0.015
rified+0.015
 Where+0.015
 tactic+0.015
Neg logit contributions
 bat-0.011
 blow-0.011
 put-0.011
 se-0.011
 dove-0.010
Tokenrum
Token ID6582
Feature activation-1.66e-01

Pos logit contributions
 unlike+0.289
ifying+0.280
 Where+0.265
rified+0.264
 blonde+0.263
Neg logit contributions
ans-0.703
ck-0.699
 next-0.682
 blow-0.673
 put-0.667
Token.
Token ID13
Feature activation-2.47e-02

Pos logit contributions
 unlike+0.355
 tactic+0.350
 matter+0.339
rified+0.336
 Where+0.334
Neg logit contributions
 next-0.133
ans-0.125
 put-0.124
 waves-0.123
 blow-0.123
Token Suddenly
Token ID24975
Feature activation-1.24e-01

Pos logit contributions
 unlike+0.040
 tactic+0.040
 matter+0.040
mate+0.039
ifying+0.039
Neg logit contributions
 next-0.034
 put-0.032
 goes-0.031
 bat-0.031
 and-0.030
Token,
Token ID11
Feature activation-3.05e-02

Pos logit contributions
 tactic+0.261
 unlike+0.261
rified+0.247
 matter+0.246
ifying+0.244
Neg logit contributions
 next-0.105
 put-0.104
 bat-0.098
 blow-0.098
 waves-0.096
Token she
Token ID673
Feature activation+2.17e-02

Pos logit contributions
 tactic+0.078
 unlike+0.075
eway+0.074
 perspective+0.073
 phrase+0.071
Neg logit contributions
 next-0.020
 put-0.019
 blow-0.017
 then-0.017
 bat-0.017
Token and
Token ID290
Feature activation+1.11e-01

Pos logit contributions
 unlike+0.091
 tactic+0.089
rified+0.087
 matter+0.086
 Where+0.086
Neg logit contributions
 next-0.017
ans-0.016
 put-0.016
 waves-0.016
 blow-0.015
Token so
Token ID523
Feature activation-9.34e-02

Pos logit contributions
 next+0.153
 put+0.150
ans+0.150
 blow+0.149
 bat+0.148
Neg logit contributions
 unlike-0.171
 tactic-0.168
rified-0.159
 Where-0.158
 matter-0.157
Token she
Token ID673
Feature activation+2.42e-02

Pos logit contributions
 unlike+0.181
 tactic+0.177
rified+0.171
 matter+0.171
 Where+0.170
Neg logit contributions
ans-0.093
 next-0.091
 waves-0.090
 blow-0.089
 bat-0.088
Token threw
Token ID9617
Feature activation+1.54e-01

Pos logit contributions
ans+0.026
 next+0.024
 waves+0.024
 blow+0.023
ites+0.023
Neg logit contributions
 unlike-0.044
 tactic-0.043
 matter-0.043
rified-0.042
 terribly-0.042
Token a
Token ID257
Feature activation-1.03e-02

Pos logit contributions
ans+0.181
 next+0.172
 blow+0.171
 waves+0.169
 bat+0.167
Neg logit contributions
 unlike-0.270
 tactic-0.265
 matter-0.257
rified-0.255
 Tears-0.253
Token tant
Token ID24246
Feature activation-3.38e-01

Pos logit contributions
 unlike+0.021
 Where+0.020
 tactic+0.020
rified+0.020
 Tears+0.019
Neg logit contributions
 bat-0.011
 blow-0.010
 waves-0.010
 se-0.010
 put-0.010
Tokenrum
Token ID6582
Feature activation-1.67e-01

Pos logit contributions
ifying+0.310
 unlike+0.295
rified+0.287
 Tears+0.285
 Where+0.285
Neg logit contributions
ck-0.685
ans-0.678
 next-0.659
 blow-0.644
 put-0.642
Token when
Token ID618
Feature activation-7.98e-02

Pos logit contributions
 unlike+0.358
 tactic+0.357
 matter+0.348
rified+0.339
 Where+0.337
Neg logit contributions
 next-0.136
 waves-0.124
 put-0.123
ans-0.123
 blow-0.122
Token her
Token ID607
Feature activation-3.09e-02

Pos logit contributions
 unlike+0.153
 tactic+0.149
rified+0.144
 matter+0.144
 Where+0.143
Neg logit contributions
ans-0.081
 next-0.078
 blow-0.077
 waves-0.077
 put-0.076
Token parents
Token ID3397
Feature activation+2.33e-02

Pos logit contributions
 unlike+0.063
 tactic+0.062
rified+0.060
 matter+0.060
 Where+0.059
Neg logit contributions
ans-0.027
 next-0.027
 blow-0.026
 waves-0.026
 put-0.026
Token said
Token ID531
Feature activation+6.96e-02

Pos logit contributions
ans+0.027
 next+0.025
irts+0.025
 waves+0.025
ites+0.023
Neg logit contributions
rified-0.040
 unlike-0.039
 matter-0.039
 tactic-0.039
 terribly-0.038
Token temper
Token ID4124
Feature activation-1.09e-01

Pos logit contributions
 unlike+0.131
 tactic+0.128
rified+0.126
ifying+0.125
 Where+0.124
Neg logit contributions
 waves-0.103
ans-0.103
 put-0.102
 bat-0.100
 next-0.100
Token and
Token ID290
Feature activation+1.14e-01

Pos logit contributions
 unlike+0.251
 tactic+0.249
rified+0.244
 Where+0.242
 matter+0.241
Neg logit contributions
 next-0.064
 put-0.063
ans-0.062
 waves-0.062
 bat-0.060
Token often
Token ID1690
Feature activation-1.91e-02

Pos logit contributions
 put+0.125
 next+0.121
 waves+0.118
 bat+0.118
 blow+0.114
Neg logit contributions
 tactic-0.209
 unlike-0.208
rified-0.206
ifying-0.198
mate-0.198
Token threw
Token ID9617
Feature activation+1.60e-01

Pos logit contributions
 unlike+0.036
 matter+0.035
 tactic+0.035
 blonde+0.034
 Tears+0.034
Neg logit contributions
ans-0.020
 next-0.019
 waves-0.018
 blow-0.018
 bat-0.018
Token temper
Token ID4124
Feature activation-1.08e-01

Pos logit contributions
 waves+0.166
 put+0.164
 next+0.162
 bat+0.159
ans+0.155
Neg logit contributions
 unlike-0.309
 tactic-0.305
rified-0.294
 Thompson-0.291
 Where-0.290
Token tant
Token ID24246
Feature activation-3.32e-01

Pos logit contributions
 unlike+0.195
 tactic+0.192
 matter+0.185
 Where+0.183
 Tears+0.183
Neg logit contributions
ans-0.123
 waves-0.122
 next-0.122
 put-0.119
 bat-0.118
Tokenrums
Token ID45241
Feature activation-9.83e-02

Pos logit contributions
ifying+0.392
 Where+0.360
 unlike+0.359
rified+0.351
 kinda+0.347
Neg logit contributions
ans-0.625
ck-0.625
is-0.572
 next-0.569
ites-0.554
Token.
Token ID13
Feature activation-2.15e-02

Pos logit contributions
 unlike+0.192
 tactic+0.190
rified+0.183
 matter+0.182
 Where+0.181
Neg logit contributions
 next-0.097
 put-0.093
ans-0.091
 blow-0.090
 bat-0.090
Token One
Token ID1881
Feature activation-1.43e-02

Pos logit contributions
mate+0.035
rified+0.035
 unlike+0.035
ifying+0.034
 emanating+0.034
Neg logit contributions
 next-0.029
 put-0.027
 waves-0.026
 and-0.026
 goes-0.025
Token day
Token ID1110
Feature activation+1.80e-02

Pos logit contributions
 unlike+0.028
 tactic+0.027
 matter+0.026
rified+0.026
 Tears+0.026
Neg logit contributions
ans-0.015
 waves-0.014
 next-0.014
 blow-0.014
 put-0.013
Token,
Token ID11
Feature activation-3.37e-02

Pos logit contributions
 next+0.013
 put+0.012
 and+0.012
 then+0.010
 they+0.010
Neg logit contributions
 tactic-0.043
ifying-0.042
 unlike-0.041
rified-0.041
 emanating-0.041
Token The
Token ID383
Feature activation+5.04e-02

Pos logit contributions
 unlike+0.037
 tactic+0.036
rified+0.034
 matter+0.034
 Where+0.034
Neg logit contributions
 next-0.030
ans-0.030
 waves-0.029
 put-0.029
 blow-0.029
Token girl
Token ID2576
Feature activation-1.65e-01

Pos logit contributions
 next+0.081
 waves+0.074
 put+0.074
 bat+0.072
 blow+0.072
Neg logit contributions
 unlike-0.077
 tactic-0.074
eway-0.071
 Tears-0.070
ifying-0.070
Token was
Token ID373
Feature activation-7.95e-02

Pos logit contributions
 unlike+0.276
 tactic+0.265
 Tears+0.261
rified+0.260
 matter+0.257
Neg logit contributions
ans-0.207
irts-0.191
 waves-0.191
 bat-0.191
 next-0.189
Token too
Token ID1165
Feature activation-7.87e-02

Pos logit contributions
 unlike+0.117
 tactic+0.116
rified+0.111
 matter+0.111
 Where+0.110
Neg logit contributions
 next-0.116
 put-0.115
ans-0.112
 blow-0.112
 bat-0.112
Token pet
Token ID4273
Feature activation+7.89e-02

Pos logit contributions
 unlike+0.094
 tactic+0.086
 Tears+0.085
 terribly+0.083
rified+0.082
Neg logit contributions
ans-0.137
 waves-0.135
 next-0.132
 blow-0.131
 goes-0.131
Tokenrified
Token ID41301
Feature activation-2.95e-01

Pos logit contributions
 next+0.088
 waves+0.081
 put+0.081
 and+0.080
 goes+0.079
Neg logit contributions
 tactic-0.151
 unlike-0.147
 matter-0.141
eway-0.138
 superpower-0.138
Token to
Token ID284
Feature activation-6.58e-02

Pos logit contributions
 unlike+0.678
 tactic+0.654
rified+0.642
 Tears+0.639
 Where+0.638
Neg logit contributions
ans-0.189
 waves-0.174
 blow-0.173
 next-0.171
 bat-0.171
Token move
Token ID1445
Feature activation+8.68e-02

Pos logit contributions
 matter+0.130
 unlike+0.126
 tactic+0.126
mate+0.120
 disgust+0.120
Neg logit contributions
ans-0.061
ites-0.058
 next-0.057
 waves-0.056
irts-0.054
Token and
Token ID290
Feature activation+1.04e-01

Pos logit contributions
 next+0.069
 and+0.069
 put+0.059
 each+0.056
 bat+0.055
Neg logit contributions
 tactic-0.197
 unlike-0.192
 matter-0.185
 Where-0.184
 superpower-0.183
Token just
Token ID655
Feature activation-1.13e-01

Pos logit contributions
 next+0.102
 put+0.101
 blow+0.097
 waves+0.095
 bat+0.094
Neg logit contributions
 tactic-0.206
 unlike-0.205
 matter-0.193
rified-0.192
ifying-0.191
Token stared
Token ID18484
Feature activation+5.56e-02

Pos logit contributions
 unlike+0.216
 tactic+0.213
rified+0.204
 matter+0.203
 Where+0.203
Neg logit contributions
 next-0.113
ans-0.111
 put-0.111
 blow-0.110
 waves-0.110
Token it
Token ID340
Feature activation+2.03e-02

Pos logit contributions
ifying+0.004
 tactic+0.004
 unlike+0.004
rified+0.004
mate+0.004
Neg logit contributions
 next-0.001
 put-0.001
 waves-0.001
 each-0.001
 blow-0.001
Token,
Token ID11
Feature activation-3.26e-02

Pos logit contributions
 next+0.020
ans+0.020
 put+0.019
 blow+0.019
 waves+0.019
Neg logit contributions
 unlike-0.039
 tactic-0.038
rified-0.037
 matter-0.037
 Where-0.037
Token she
Token ID673
Feature activation+2.14e-02

Pos logit contributions
 tactic+0.068
ifying+0.065
 unlike+0.065
 perspective+0.064
mate+0.064
Neg logit contributions
 next-0.031
 put-0.028
 bat-0.027
 waves-0.026
 blow-0.026
Token was
Token ID373
Feature activation-7.64e-02

Pos logit contributions
ans+0.029
 next+0.028
 blow+0.028
 waves+0.027
 bat+0.027
Neg logit contributions
 unlike-0.033
 tactic-0.031
rified-0.031
 matter-0.031
 Tears-0.030
Token pet
Token ID4273
Feature activation+8.05e-02

Pos logit contributions
 unlike+0.116
 tactic+0.115
rified+0.110
 matter+0.109
 Where+0.109
Neg logit contributions
 next-0.108
 put-0.107
 blow-0.105
ans-0.105
 bat-0.104
Tokenrified
Token ID41301
Feature activation-2.95e-01

Pos logit contributions
 next+0.083
 and+0.078
 waves+0.076
 each+0.075
 put+0.074
Neg logit contributions
 tactic-0.166
 unlike-0.160
eway-0.156
 matter-0.155
 superpower-0.153
Token.
Token ID13
Feature activation-2.07e-02

Pos logit contributions
 tactic+0.656
 unlike+0.647
 matter+0.636
rified+0.623
 Tears+0.616
Neg logit contributions
 next-0.221
 put-0.197
ans-0.195
 waves-0.195
 blow-0.193
Token
Token ID220
Feature activation-1.05e-01

Pos logit contributions
 matter+0.047
ifying+0.046
 unlike+0.046
mate+0.045
 tactic+0.044
Neg logit contributions
 next-0.016
 and-0.014
 each-0.013
 put-0.013
 bat-0.012
Token
Token ID198
Feature activation+2.59e-02

Pos logit contributions
 unlike+0.235
 tactic+0.230
 matter+0.227
rified+0.225
ifying+0.219
Neg logit contributions
 next-0.074
ans-0.069
 put-0.069
 bat-0.066
 blow-0.065
Token
Token ID198
Feature activation+2.48e-02

Pos logit contributions
ans+0.023
 waves+0.022
 next+0.022
 blow+0.022
 put+0.021
Neg logit contributions
 unlike-0.053
 tactic-0.052
 Where-0.050
rified-0.050
 Tears-0.050
TokenOne
Token ID3198
Feature activation+1.18e-02

Pos logit contributions
ans+0.033
 next+0.032
 waves+0.032
 blow+0.032
 put+0.032
Neg logit contributions
 unlike-0.040
 tactic-0.038
rified-0.037
 matter-0.037
 Where-0.037
Token moms
Token ID34015
Feature activation+9.44e-02

Pos logit contributions
 unlike+0.017
 tactic+0.016
 matter+0.016
 Tears+0.016
rified+0.016
Neg logit contributions
ans-0.006
 blow-0.005
 next-0.005
 waves-0.005
 bat-0.005
Token were
Token ID547
Feature activation-4.02e-02

Pos logit contributions
ans+0.108
 next+0.107
 waves+0.105
 blow+0.104
 bat+0.103
Neg logit contributions
 unlike-0.167
 tactic-0.163
rified-0.157
 matter-0.156
 Tears-0.156
Token mad
Token ID8805
Feature activation-9.58e-02

Pos logit contributions
 unlike+0.063
 terribly+0.060
 Tears+0.058
 tactic+0.057
 matter+0.057
Neg logit contributions
ans-0.057
 waves-0.054
 goes-0.051
 blow-0.051
 next-0.051
Token.
Token ID13
Feature activation-3.17e-02

Pos logit contributions
 unlike+0.189
 tactic+0.179
rified+0.179
 Tears+0.176
 Where+0.176
Neg logit contributions
ans-0.093
 blow-0.087
 waves-0.086
 next-0.085
 bat-0.085
Token No
Token ID1400
Feature activation-8.01e-02

Pos logit contributions
 tactic+0.064
 unlike+0.064
 matter+0.063
mate+0.061
 avail+0.060
Neg logit contributions
 next-0.028
ans-0.027
 put-0.027
 bat-0.027
 waves-0.026
Token matter
Token ID2300
Feature activation-2.81e-01

Pos logit contributions
 unlike+0.129
 tactic+0.121
 Tears+0.119
 Where+0.119
rified+0.118
Neg logit contributions
 next-0.112
 put-0.110
 blow-0.108
 waves-0.106
 bat-0.104
Token how
Token ID703
Feature activation-4.70e-02

Pos logit contributions
 unlike+0.472
 tactic+0.461
rified+0.444
 blonde+0.435
 Where+0.432
Neg logit contributions
 next-0.358
 put-0.348
 blow-0.342
ans-0.340
 bat-0.339
Token much
Token ID881
Feature activation-4.39e-02

Pos logit contributions
 unlike+0.063
 Tears+0.061
 tactic+0.060
 terribly+0.060
 matter+0.059
Neg logit contributions
ans-0.076
 waves-0.072
 blow-0.071
 dove-0.070
 put-0.069
Token they
Token ID484
Feature activation+8.78e-02

Pos logit contributions
 unlike+0.084
 tactic+0.082
 matter+0.080
 Tears+0.078
rified+0.078
Neg logit contributions
ans-0.045
 next-0.043
 waves-0.042
 blow-0.042
 bat-0.041
Token tried
Token ID3088
Feature activation-9.71e-03

Pos logit contributions
ans+0.125
 waves+0.114
 next+0.113
irts+0.109
 bat+0.108
Neg logit contributions
 matter-0.135
 unlike-0.135
 tactic-0.131
rified-0.130
 terribly-0.126
Token,
Token ID11
Feature activation-3.62e-02

Pos logit contributions
 unlike+0.020
 tactic+0.020
rified+0.020
 Where+0.019
 matter+0.019
Neg logit contributions
 next-0.009
 put-0.008
 bat-0.008
 blow-0.008
 waves-0.008
Token
Token ID198
Feature activation+2.01e-02

Pos logit contributions
ans+0.022
 waves+0.022
 blow+0.021
 next+0.020
 put+0.020
Neg logit contributions
 tactic-0.039
 unlike-0.039
 Where-0.039
 Tears-0.038
mate-0.037
TokenHis
Token ID6653
Feature activation-1.17e-01

Pos logit contributions
ans+0.037
 waves+0.037
 blow+0.035
 boats+0.034
 goes+0.034
Neg logit contributions
 unlike-0.022
 tactic-0.021
 Tears-0.020
 Where-0.020
 matter-0.020
Token mom
Token ID1995
Feature activation+3.64e-02

Pos logit contributions
 unlike+0.283
rified+0.279
 tactic+0.271
eway+0.267
ifying+0.266
Neg logit contributions
 next-0.083
 put-0.073
 bat-0.072
 waves-0.066
 blow-0.063
Token was
Token ID373
Feature activation-8.12e-02

Pos logit contributions
ans+0.033
 waves+0.030
 bat+0.030
 next+0.030
 blow+0.030
Neg logit contributions
 unlike-0.074
 tactic-0.072
rified-0.071
 Tears-0.071
 matter-0.071
Token taken
Token ID2077
Feature activation+1.10e-01

Pos logit contributions
 unlike+0.132
 tactic+0.131
rified+0.129
 matter+0.125
 Where+0.125
Neg logit contributions
 next-0.109
 put-0.106
 blow-0.102
 bat-0.102
ans-0.101
Token aback
Token ID50147
Feature activation-2.80e-01

Pos logit contributions
 put+0.092
 next+0.085
 and+0.080
 bat+0.079
 waves+0.075
Neg logit contributions
 unlike-0.252
 tactic-0.245
eway-0.245
 perspective-0.237
ifying-0.236
Token.
Token ID13
Feature activation-2.17e-02

Pos logit contributions
 unlike+0.644
 tactic+0.636
rified+0.621
 matter+0.617
 Tears+0.614
Neg logit contributions
ans-0.165
 blow-0.159
 waves-0.159
 next-0.158
 bat-0.158
Token She
Token ID1375
Feature activation-4.96e-02

Pos logit contributions
 tactic+0.034
 unlike+0.034
 matter+0.033
mate+0.032
rified+0.032
Neg logit contributions
 next-0.030
 put-0.029
 bat-0.028
 waves-0.028
 blow-0.028
Token said
Token ID531
Feature activation+6.70e-02

Pos logit contributions
 unlike+0.096
 matter+0.094
 terribly+0.091
 must+0.090
 exactly+0.090
Neg logit contributions
ans-0.051
ites-0.044
 waves-0.044
 bat-0.043
irts-0.043
Token calmly
Token ID30180
Feature activation-8.35e-02

Pos logit contributions
 next+0.065
ans+0.064
 put+0.063
 blow+0.063
 waves+0.063
Neg logit contributions
 unlike-0.130
 tactic-0.128
 matter-0.124
rified-0.123
 Tears-0.122
Token,
Token ID11
Feature activation-3.53e-02

Pos logit contributions
 unlike+0.163
 tactic+0.159
rified+0.154
 matter+0.153
 Where+0.153
Neg logit contributions
ans-0.080
 next-0.079
 waves-0.078
 blow-0.077
 put-0.077
Token care
Token ID1337
Feature activation-6.61e-03

Pos logit contributions
 next+0.215
ans+0.212
 put+0.209
 waves+0.209
 blow+0.206
Neg logit contributions
 unlike-0.303
 tactic-0.295
rified-0.282
 Tears-0.280
 matter-0.279
Token of
Token ID286
Feature activation-5.30e-02

Pos logit contributions
 unlike+0.013
 tactic+0.013
rified+0.012
 Tears+0.012
 matter+0.012
Neg logit contributions
ans-0.006
 waves-0.006
 blow-0.006
 bat-0.006
 next-0.006
Token him
Token ID683
Feature activation-4.58e-02

Pos logit contributions
 unlike+0.096
 tactic+0.094
rified+0.090
 Tears+0.089
 Where+0.089
Neg logit contributions
 next-0.061
 put-0.058
ans-0.058
 waves-0.058
 blow-0.057
Token.
Token ID13
Feature activation-2.86e-02

Pos logit contributions
 unlike+0.096
 tactic+0.094
rified+0.091
 matter+0.091
 Tears+0.090
Neg logit contributions
ans-0.037
 blow-0.035
 waves-0.035
 next-0.035
 bat-0.034
Token No
Token ID1400
Feature activation-8.01e-02

Pos logit contributions
 unlike+0.042
 tactic+0.042
 matter+0.040
mate+0.040
rified+0.039
Neg logit contributions
 next-0.041
 put-0.040
 goes-0.040
 waves-0.040
 bat-0.039
Token matter
Token ID2300
Feature activation-2.80e-01

Pos logit contributions
 unlike+0.054
 Tears+0.045
eway+0.045
ifying+0.044
 Where+0.044
Neg logit contributions
 next-0.193
 put-0.189
 blow-0.185
 waves-0.181
 goes-0.179
Token what
Token ID644
Feature activation-8.98e-02

Pos logit contributions
 unlike+0.513
 tactic+0.497
rified+0.485
 blonde+0.483
 Terr+0.479
Neg logit contributions
 next-0.331
 put-0.310
 each-0.297
 and-0.295
 blow-0.287
Token happened
Token ID3022
Feature activation-7.43e-02

Pos logit contributions
 tactic+0.183
 unlike+0.183
rified+0.177
 Where+0.173
 Tears+0.172
Neg logit contributions
 next-0.083
 put-0.081
 waves-0.077
 bat-0.076
 goes-0.074
Token,
Token ID11
Feature activation-3.54e-02

Pos logit contributions
 unlike+0.137
 tactic+0.132
 Where+0.128
rified+0.128
 Tears+0.127
Neg logit contributions
ans-0.081
 blow-0.078
 waves-0.077
 next-0.076
 bat-0.076
Token Grand
Token ID5675
Feature activation+7.47e-02

Pos logit contributions
 unlike+0.057
 tactic+0.056
rified+0.053
 matter+0.052
 Where+0.052
Neg logit contributions
 next-0.047
ans-0.045
 waves-0.045
 put-0.045
 blow-0.045
Tokenpa
Token ID8957
Feature activation+2.04e-02

Pos logit contributions
 bat+0.104
ans+0.101
 blow+0.100
 waves+0.098
 swing+0.097
Neg logit contributions
 tactic-0.114
 matter-0.113
 unlike-0.112
rified-0.106
 Thompson-0.105
Token its
Token ID663
Feature activation-3.84e-02

Pos logit contributions
 unlike+0.129
 tactic+0.127
rified+0.123
 matter+0.122
 Where+0.122
Neg logit contributions
ans-0.043
 next-0.043
 waves-0.041
 blow-0.041
 put-0.041
Token message
Token ID3275
Feature activation-8.09e-02

Pos logit contributions
 unlike+0.058
rified+0.057
 tactic+0.055
 Where+0.054
ifying+0.054
Neg logit contributions
 next-0.057
 put-0.054
 waves-0.054
ans-0.054
 bat-0.053
Token -
Token ID532
Feature activation-9.86e-02

Pos logit contributions
 unlike+0.181
 tactic+0.176
 matter+0.171
rified+0.171
 Tears+0.170
Neg logit contributions
ans-0.057
 waves-0.053
 blow-0.052
 next-0.052
 bat-0.052
Token that
Token ID326
Feature activation+1.00e-01

Pos logit contributions
 unlike+0.188
 tactic+0.177
 matter+0.174
 Tears+0.173
 Where+0.172
Neg logit contributions
ans-0.109
 waves-0.099
 blow-0.097
 bat-0.097
 next-0.096
Token no
Token ID645
Feature activation-4.22e-02

Pos logit contributions
 next+0.104
 put+0.100
 blow+0.099
 waves+0.098
ans+0.098
Neg logit contributions
 unlike-0.188
 tactic-0.186
rified-0.181
 Where-0.178
 Tears-0.177
Token matter
Token ID2300
Feature activation-2.80e-01

Pos logit contributions
 unlike+0.037
rified+0.035
 Where+0.034
 Tears+0.033
 Thompson+0.033
Neg logit contributions
 put-0.089
 next-0.089
 blow-0.086
 waves-0.084
 goes-0.084
Token what
Token ID644
Feature activation-8.93e-02

Pos logit contributions
 unlike+0.475
 tactic+0.462
rified+0.451
 Where+0.439
 blonde+0.435
Neg logit contributions
 next-0.351
 put-0.341
 blow-0.330
ans-0.327
 bat-0.323
Token,
Token ID11
Feature activation-3.96e-02

Pos logit contributions
 unlike+0.169
 tactic+0.169
rified+0.164
 Where+0.160
 Tears+0.159
Neg logit contributions
 next-0.093
 put-0.090
 waves-0.087
 bat-0.086
 blow-0.086
Token their
Token ID511
Feature activation-8.04e-03

Pos logit contributions
 tactic+0.079
 unlike+0.078
rified+0.077
 Thompson+0.075
 Tears+0.074
Neg logit contributions
 next-0.038
 put-0.036
 blow-0.034
 waves-0.034
 bat-0.034
Token friendship
Token ID14738
Feature activation-6.21e-02

Pos logit contributions
 unlike+0.013
 tactic+0.012
rified+0.012
 Where+0.012
 Tears+0.012
Neg logit contributions
 next-0.011
ans-0.011
 waves-0.011
 put-0.011
 bat-0.011
Token would
Token ID561
Feature activation-9.60e-02

Pos logit contributions
 unlike+0.101
 tactic+0.096
 matter+0.095
 Tears+0.092
rified+0.092
Neg logit contributions
ans-0.082
 next-0.077
 waves-0.077
 bat-0.076
 blow-0.076
Token truck
Token ID7779
Feature activation-4.73e-02

Pos logit contributions
ans+0.070
ck+0.066
 blow+0.064
 waves+0.064
 next+0.064
Neg logit contributions
 unlike-0.096
 tactic-0.094
 matter-0.093
 Where-0.091
 Tears-0.089
Token loved
Token ID6151
Feature activation+2.94e-02

Pos logit contributions
 unlike+0.076
 tactic+0.070
 Where+0.069
 matter+0.069
 Tears+0.069
Neg logit contributions
ans-0.064
irts-0.060
 blow-0.059
 bat-0.059
ck-0.059
Token to
Token ID284
Feature activation-6.06e-02

Pos logit contributions
ans+0.032
 next+0.032
 waves+0.031
 blow+0.031
 put+0.031
Neg logit contributions
 unlike-0.054
 tactic-0.053
rified-0.051
 matter-0.051
 Where-0.050
Token drive
Token ID3708
Feature activation+1.79e-01

Pos logit contributions
 matter+0.114
 blonde+0.104
 tactic+0.104
 superpower+0.103
 unlike+0.102
Neg logit contributions
ans-0.065
irts-0.063
 next-0.061
ites-0.059
is-0.059
Token around
Token ID1088
Feature activation-3.97e-02

Pos logit contributions
 next+0.146
 put+0.136
ans+0.133
 waves+0.133
 bat+0.129
Neg logit contributions
 unlike-0.388
 tactic-0.385
rified-0.371
 Tears-0.368
 Where-0.367
Token and
Token ID290
Feature activation+1.11e-01

Pos logit contributions
 unlike+0.094
 tactic+0.093
rified+0.090
 matter+0.089
 Tears+0.089
Neg logit contributions
 next-0.023
ans-0.021
 put-0.021
 waves-0.021
 blow-0.021
Token explore
Token ID7301
Feature activation+9.00e-02

Pos logit contributions
ans+0.107
 next+0.094
 waves+0.092
 bat+0.091
irts+0.090
Neg logit contributions
 unlike-0.223
 tactic-0.216
 matter-0.215
 Where-0.212
 Tears-0.208
Token the
Token ID262
Feature activation+7.61e-02

Pos logit contributions
ans+0.068
 blow+0.065
 next+0.064
 bat+0.063
 waves+0.063
Neg logit contributions
 unlike-0.195
 tactic-0.191
 matter-0.186
rified-0.185
 Where-0.184
Token world
Token ID995
Feature activation-8.36e-02

Pos logit contributions
ck+0.116
ans+0.115
 blow+0.114
 bat+0.110
 put+0.110
Neg logit contributions
 unlike-0.100
 matter-0.097
 tactic-0.096
 Where-0.095
 total-0.091
Token.
Token ID13
Feature activation-1.45e-02

Pos logit contributions
 unlike+0.141
 tactic+0.135
 Where+0.130
 terribly+0.129
eway+0.129
Neg logit contributions
ans-0.105
 bat-0.099
irts-0.098
 blow-0.098
 dads-0.096
Token
Token ID198
Feature activation+2.80e-02

Pos logit contributions
 unlike+0.028
 tactic+0.028
 Where+0.027
 Tears+0.027
rified+0.027
Neg logit contributions
ans-0.014
 next-0.014
 waves-0.014
 blow-0.014
 put-0.014
Token Pam
Token ID24497
Feature activation+7.52e-03

Pos logit contributions
ans+0.086
 next+0.084
 waves+0.084
 blow+0.083
 put+0.082
Neg logit contributions
 unlike-0.105
 tactic-0.102
 Where-0.098
 matter-0.098
 Tears-0.098
Token.
Token ID13
Feature activation-7.90e-03

Pos logit contributions
 waves+0.009
ans+0.008
 blow+0.008
 goes+0.008
irts+0.008
Neg logit contributions
 unlike-0.013
 tactic-0.012
 terribly-0.012
 Thompson-0.012
 Tears-0.012
Token Everyday
Token ID48154
Feature activation-2.34e-02

Pos logit contributions
 unlike+0.015
 tactic+0.015
 matter+0.014
rified+0.014
mate+0.014
Neg logit contributions
 next-0.008
ans-0.008
 put-0.008
 bat-0.008
 waves-0.008
Token they
Token ID484
Feature activation+1.02e-01

Pos logit contributions
 unlike+0.052
 tactic+0.051
rified+0.049
 matter+0.049
 Where+0.049
Neg logit contributions
ans-0.016
 next-0.016
 waves-0.016
 blow-0.016
 put-0.016
Token would
Token ID561
Feature activation-8.03e-02

Pos logit contributions
ans+0.096
 next+0.092
is+0.086
 waves+0.085
irts+0.083
Neg logit contributions
 matter-0.197
 unlike-0.196
rified-0.192
 blonde-0.190
 total-0.189
Token play
Token ID711
Feature activation+1.41e-01

Pos logit contributions
 matter+0.161
 unlike+0.159
 blonde+0.158
 tactic+0.154
 total+0.153
Neg logit contributions
ans-0.070
ites-0.065
is-0.065
 next-0.063
irts-0.063
Token together
Token ID1978
Feature activation-7.79e-02

Pos logit contributions
 next+0.139
 put+0.132
 bat+0.131
ans+0.127
 blow+0.127
Neg logit contributions
 unlike-0.282
 tactic-0.277
rified-0.272
 matter-0.270
 Tears-0.270
Token.
Token ID13
Feature activation-1.03e-02

Pos logit contributions
 unlike+0.163
 tactic+0.159
 matter+0.154
rified+0.153
 Where+0.152
Neg logit contributions
ans-0.065
 waves-0.062
 next-0.061
 blow-0.061
 bat-0.060
Token
Token ID198
Feature activation+3.22e-02

Pos logit contributions
 unlike+0.020
 tactic+0.020
rified+0.019
 matter+0.019
 Tears+0.019
Neg logit contributions
ans-0.010
 next-0.010
 waves-0.010
 blow-0.010
 put-0.010
Token
Token ID198
Feature activation+3.00e-02

Pos logit contributions
ans+0.030
 waves+0.029
 blow+0.028
 next+0.028
 bat+0.028
Neg logit contributions
 unlike-0.064
 tactic-0.063
 Where-0.061
 Tears-0.061
 matter-0.060
TokenOne
Token ID3198
Feature activation+1.74e-02

Pos logit contributions
 waves+0.039
ans+0.039
 blow+0.038
 next+0.037
 put+0.037
Neg logit contributions
 unlike-0.048
 tactic-0.047
 Where-0.046
 matter-0.046
mate-0.044
Token towards
Token ID3371
Feature activation-3.52e-02

Pos logit contributions
ans+0.242
 waves+0.226
 blow+0.224
 bat+0.222
 next+0.220
Neg logit contributions
 unlike-0.321
 tactic-0.312
 Where-0.299
 matter-0.298
 Tears-0.296
Token it
Token ID340
Feature activation+2.64e-02

Pos logit contributions
 Where+0.052
 unlike+0.052
 tactic+0.051
 matter+0.051
 Tears+0.050
Neg logit contributions
ans-0.051
irts-0.046
 bat-0.044
ars-0.044
 blow-0.043
Token and
Token ID290
Feature activation+1.15e-01

Pos logit contributions
ans+0.018
 waves+0.015
 blow+0.015
ck+0.014
irts+0.014
Neg logit contributions
 unlike-0.061
 tactic-0.060
 Tears-0.058
 matter-0.057
 terribly-0.057
Token began
Token ID2540
Feature activation-7.95e-02

Pos logit contributions
ans+0.113
 next+0.107
 waves+0.104
 blow+0.103
 bat+0.103
Neg logit contributions
 unlike-0.227
 tactic-0.222
rified-0.216
 matter-0.215
 Tears-0.214
Token to
Token ID284
Feature activation-5.52e-02

Pos logit contributions
 unlike+0.131
 tactic+0.128
 Tears+0.123
 Where+0.122
 matter+0.121
Neg logit contributions
ans-0.103
 waves-0.095
 next-0.094
 blow-0.094
 bat-0.093
Token slide
Token ID10649
Feature activation+1.77e-01

Pos logit contributions
 tactic+0.133
 unlike+0.132
 matter+0.130
 Tears+0.127
 Where+0.127
Neg logit contributions
ans-0.030
 next-0.026
 waves-0.023
 bat-0.023
 blow-0.022
Token down
Token ID866
Feature activation-1.10e-01

Pos logit contributions
ans+0.226
 blow+0.218
 next+0.217
 waves+0.216
 bat+0.215
Neg logit contributions
 unlike-0.291
 tactic-0.284
 matter-0.272
rified-0.272
 Where-0.271
Token.
Token ID13
Feature activation-1.43e-02

Pos logit contributions
 unlike+0.247
 tactic+0.243
rified+0.235
 matter+0.235
 Where+0.233
Neg logit contributions
 next-0.079
 put-0.072
 waves-0.072
 blow-0.070
 bat-0.070
Token Her
Token ID2332
Feature activation-5.63e-02

Pos logit contributions
 unlike+0.026
 tactic+0.025
 matter+0.024
rified+0.024
 Tears+0.024
Neg logit contributions
 next-0.016
ans-0.016
 put-0.016
 waves-0.016
 blow-0.016
Token eyes
Token ID2951
Feature activation-1.36e-01

Pos logit contributions
 unlike+0.122
 tactic+0.119
rified+0.117
 Where+0.115
 matter+0.114
Neg logit contributions
 next-0.050
 waves-0.047
 put-0.046
 bat-0.045
 blow-0.045
Token spark
Token ID9009
Feature activation+1.84e-02

Pos logit contributions
 unlike+0.221
 blonde+0.201
 matter+0.200
 tactic+0.199
 Tears+0.197
Neg logit contributions
ans-0.195
 dads-0.175
 next-0.174
 bat-0.171
 waves-0.169
Token They
Token ID1119
Feature activation+3.85e-02

Pos logit contributions
 unlike+0.026
 tactic+0.026
 Where+0.025
rified+0.025
 Tears+0.025
Neg logit contributions
ans-0.017
 next-0.016
 blow-0.016
 waves-0.016
 put-0.016
Token liked
Token ID8288
Feature activation-2.55e-02

Pos logit contributions
ans+0.063
irts+0.057
 next+0.056
is+0.055
ck+0.053
Neg logit contributions
 unlike-0.053
 blonde-0.052
 matter-0.049
 total-0.048
 Where-0.047
Token to
Token ID284
Feature activation-6.10e-02

Pos logit contributions
 unlike+0.039
 matter+0.036
 tactic+0.036
 Where+0.035
 blonde+0.034
Neg logit contributions
ans-0.037
irts-0.035
ck-0.034
is-0.033
 blow-0.033
Token ride
Token ID6594
Feature activation+1.87e-01

Pos logit contributions
 matter+0.118
 tactic+0.114
 blonde+0.110
 total+0.105
 avail+0.104
Neg logit contributions
ans-0.063
is-0.061
ugg-0.060
udd-0.060
ip-0.059
Token fast
Token ID3049
Feature activation+9.94e-02

Pos logit contributions
ans+0.145
 next+0.143
 blow+0.139
 waves+0.139
 put+0.137
Neg logit contributions
 unlike-0.400
 tactic-0.392
rified-0.380
 matter-0.379
 Where-0.377
Token and
Token ID290
Feature activation+1.11e-01

Pos logit contributions
ans+0.061
 blow+0.057
 bat+0.055
 next+0.054
 put+0.053
Neg logit contributions
 unlike-0.226
 tactic-0.222
rified-0.220
 matter-0.218
 Where-0.215
Token see
Token ID766
Feature activation+1.32e-01

Pos logit contributions
ans+0.085
 next+0.082
 waves+0.079
 blow+0.078
 bat+0.077
Neg logit contributions
 unlike-0.239
 tactic-0.235
 matter-0.228
 Where-0.226
rified-0.225
Token the
Token ID262
Feature activation+7.49e-02

Pos logit contributions
ans+0.130
 next+0.129
 blow+0.126
 bat+0.124
 waves+0.123
Neg logit contributions
 unlike-0.254
 tactic-0.250
 matter-0.240
rified-0.240
 Where-0.239
Token view
Token ID1570
Feature activation-1.06e-02

Pos logit contributions
ans+0.073
 next+0.071
 blow+0.071
 put+0.070
 bat+0.070
Neg logit contributions
 unlike-0.145
 tactic-0.141
 matter-0.137
rified-0.136
 Where-0.135
Token of
Token ID286
Feature activation-4.37e-02

Pos logit contributions
 unlike+0.019
 matter+0.018
 tactic+0.018
 emanating+0.017
eway+0.017
Neg logit contributions
ans-0.013
 bat-0.012
irts-0.011
 dads-0.011
 waves-0.011
Token the
Token ID262
Feature activation+7.44e-02

Pos logit contributions
 unlike+0.089
 matter+0.086
 tactic+0.085
 blonde+0.083
 Tears+0.083
Neg logit contributions
ans-0.039
 next-0.034
 goes-0.033
is-0.033
 blow-0.033
Token help
Token ID1037
Feature activation+2.84e-02

Pos logit contributions
 matter+0.106
 tactic+0.105
mate+0.100
 unlike+0.100
 disgust+0.098
Neg logit contributions
ans-0.065
ites-0.062
 waves-0.059
irts-0.057
 next-0.056
Token him
Token ID683
Feature activation-3.35e-02

Pos logit contributions
ans+0.037
 waves+0.036
 next+0.036
 blow+0.036
 bat+0.035
Neg logit contributions
 unlike-0.045
 tactic-0.044
 matter-0.043
 Where-0.042
rified-0.042
Token.
Token ID13
Feature activation-1.95e-02

Pos logit contributions
 unlike+0.071
 tactic+0.069
rified+0.067
 matter+0.067
 Where+0.066
Neg logit contributions
 next-0.028
ans-0.027
 put-0.027
 blow-0.027
 bat-0.026
Token He
Token ID679
Feature activation-9.62e-02

Pos logit contributions
 unlike+0.029
 tactic+0.028
 matter+0.028
ifying+0.028
rified+0.027
Neg logit contributions
 next-0.029
 put-0.029
 bat-0.028
 waves-0.028
 blow-0.027
Token saw
Token ID2497
Feature activation-2.12e-03

Pos logit contributions
 unlike+0.173
rified+0.165
 tactic+0.164
 matter+0.163
 terribly+0.161
Neg logit contributions
ans-0.106
 waves-0.094
ites-0.093
irts-0.092
 next-0.091
Token Tom
Token ID4186
Feature activation+1.28e-01

Pos logit contributions
 unlike+0.004
 tactic+0.004
rified+0.004
 Where+0.004
 Tears+0.004
Neg logit contributions
ans-0.002
 next-0.002
 blow-0.002
 waves-0.002
 bat-0.002
Token's
Token ID338
Feature activation+1.73e-02

Pos logit contributions
ans+0.136
 waves+0.128
ites+0.121
 dads+0.119
 bat+0.116
Neg logit contributions
 unlike-0.237
 tactic-0.234
 Tears-0.232
rified-0.226
 terribly-0.225
Token hand
Token ID1021
Feature activation+1.67e-02

Pos logit contributions
 next+0.022
ans+0.021
 put+0.021
 waves+0.021
 blow+0.021
Neg logit contributions
 unlike-0.029
 tactic-0.028
rified-0.027
 matter-0.027
 Where-0.027
Token was
Token ID373
Feature activation-7.80e-02

Pos logit contributions
ans+0.011
 waves+0.010
 next+0.010
 bat+0.009
 blow+0.009
Neg logit contributions
 unlike-0.038
 tactic-0.037
 Tears-0.036
rified-0.036
 matter-0.036
Token bleeding
Token ID16832
Feature activation-1.06e-01

Pos logit contributions
 unlike+0.130
 tactic+0.128
rified+0.123
 matter+0.122
 Where+0.121
Neg logit contributions
 next-0.099
ans-0.097
 put-0.096
 blow-0.096
 bat-0.096
Token and
Token ID290
Feature activation+1.12e-01

Pos logit contributions
 unlike+0.239
 tactic+0.235
rified+0.228
 matter+0.227
 Where+0.226
Neg logit contributions
 next-0.071
ans-0.071
 blow-0.068
 waves-0.068
 put-0.067
Token there
Token ID612
Feature activation-8.75e-02

Pos logit contributions
 unlike+0.159
 tactic+0.156
 matter+0.153
rified+0.152
ifying+0.151
Neg logit contributions
 next-0.063
 put-0.056
 bat-0.056
 goes-0.053
 blow-0.053
Token was
Token ID373
Feature activation-7.20e-02

Pos logit contributions
 unlike+0.181
ifying+0.181
 tactic+0.175
rified+0.171
 superpower+0.169
Neg logit contributions
 put-0.084
 next-0.081
 goes-0.076
 bat-0.076
 and-0.074
Token a
Token ID257
Feature activation-6.90e-03

Pos logit contributions
 unlike+0.128
 tactic+0.125
 matter+0.120
rified+0.120
 Where+0.119
Neg logit contributions
ans-0.082
 next-0.079
 waves-0.079
 blow-0.078
 put-0.078
Token small
Token ID1402
Feature activation-1.57e-02

Pos logit contributions
 unlike+0.010
 matter+0.010
 tactic+0.010
 Where+0.010
 blonde+0.010
Neg logit contributions
ans-0.010
 waves-0.009
 next-0.009
ck-0.009
irts-0.009
Token baby
Token ID5156
Feature activation-1.03e-01

Pos logit contributions
 unlike+0.024
 tactic+0.024
 matter+0.023
rified+0.023
 Where+0.023
Neg logit contributions
ans-0.021
 next-0.020
ck-0.020
 blow-0.020
 waves-0.020
Token called
Token ID1444
Feature activation+5.20e-02

Pos logit contributions
 unlike+0.177
 Where+0.168
rified+0.168
 Thompson+0.166
 tactic+0.165
Neg logit contributions
ans-0.122
 blow-0.112
irts-0.111
 waves-0.110
ck-0.110
Token Johnny
Token ID15470
Feature activation-4.25e-02

Pos logit contributions
 put+0.061
 next+0.061
 bat+0.060
 blow+0.059
ans+0.057
Neg logit contributions
 unlike-0.091
 tactic-0.091
rified-0.085
 matter-0.084
 Tears-0.083
Token.
Token ID13
Feature activation-1.93e-02

Pos logit contributions
 unlike+0.069
 tactic+0.068
 matter+0.067
 terribly+0.065
 Thompson+0.065
Neg logit contributions
 waves-0.051
ans-0.051
irts-0.049
ites-0.049
 goes-0.047
Token He
Token ID679
Feature activation-9.36e-02

Pos logit contributions
 unlike+0.024
 tactic+0.023
rified+0.022
 matter+0.021
 Where+0.021
Neg logit contributions
ans-0.033
 next-0.032
 waves-0.032
 blow-0.032
 put-0.032
Token was
Token ID373
Feature activation-7.30e-02

Pos logit contributions
 unlike+0.154
 tactic+0.145
 matter+0.142
 Where+0.141
rified+0.140
Neg logit contributions
ans-0.123
 next-0.117
 waves-0.117
 blow-0.115
 bat-0.113
Token very
Token ID845
Feature activation-1.26e-01

Pos logit contributions
 tactic+0.121
 unlike+0.118
rified+0.116
eway+0.113
 matter+0.113
Neg logit contributions
 next-0.096
 put-0.096
 blow-0.094
 bat-0.093
 waves-0.091
Token flowers
Token ID12734
Feature activation+2.04e-02

Pos logit contributions
 next+0.081
 waves+0.079
 blow+0.078
 put+0.078
ans+0.078
Neg logit contributions
 unlike-0.108
 tactic-0.106
rified-0.101
 Where-0.100
 Tears-0.100
Token made
Token ID925
Feature activation+4.30e-02

Pos logit contributions
ans+0.028
irts+0.025
 dads+0.025
 bat+0.025
 waves+0.024
Neg logit contributions
 unlike-0.034
rified-0.032
 Tears-0.031
 tactic-0.031
 alas-0.030
Token the
Token ID262
Feature activation+6.38e-02

Pos logit contributions
ans+0.052
 next+0.050
 blow+0.049
 waves+0.049
 bat+0.049
Neg logit contributions
 unlike-0.074
 tactic-0.072
rified-0.070
 matter-0.070
 Tears-0.069
Token superhero
Token ID17343
Feature activation-1.31e-01

Pos logit contributions
ans+0.083
 next+0.080
 blow+0.079
 waves+0.079
 bat+0.079
Neg logit contributions
 unlike-0.104
 tactic-0.101
rified-0.098
 matter-0.098
 Where-0.097
Token feel
Token ID1254
Feature activation+1.14e-02

Pos logit contributions
 unlike+0.231
 tactic+0.229
 Tears+0.218
 terribly+0.213
 matter+0.212
Neg logit contributions
ans-0.153
irts-0.141
 next-0.141
 waves-0.139
 bat-0.137
Token a
Token ID257
Feature activation-1.67e-02

Pos logit contributions
ans+0.013
 next+0.013
 waves+0.013
 blow+0.013
 put+0.013
Neg logit contributions
 unlike-0.020
 tactic-0.019
rified-0.019
 matter-0.019
 Where-0.019
Token bit
Token ID1643
Feature activation-9.85e-02

Pos logit contributions
 tactic+0.029
 unlike+0.029
 Where+0.028
rified+0.028
eway+0.028
Neg logit contributions
 put-0.020
 blow-0.020
 next-0.020
 bat-0.019
 waves-0.019
Token dizz
Token ID29527
Feature activation-5.07e-02

Pos logit contributions
 unlike+0.166
 tactic+0.162
rified+0.155
 Where+0.155
 Tears+0.154
Neg logit contributions
 next-0.122
 put-0.120
 blow-0.119
ans-0.119
 waves-0.119
Tokeny
Token ID88
Feature activation-8.26e-02

Pos logit contributions
 unlike+0.069
 tactic+0.068
 matter+0.064
 Where+0.064
 Tears+0.064
Neg logit contributions
 next-0.078
ans-0.078
 blow-0.077
 put-0.076
 waves-0.076
Token,
Token ID11
Feature activation-3.67e-02

Pos logit contributions
 unlike+0.170
 tactic+0.167
rified+0.161
 matter+0.161
 Where+0.160
Neg logit contributions
ans-0.070
 next-0.070
 waves-0.067
 blow-0.067
 put-0.067
Token but
Token ID475
Feature activation-1.05e-01

Pos logit contributions
 unlike+0.061
 tactic+0.057
 matter+0.056
rified+0.055
 Tears+0.055
Neg logit contributions
ans-0.048
 next-0.046
 waves-0.046
 blow-0.046
 bat-0.045
Token oranges
Token ID48389
Feature activation+1.17e-02

Pos logit contributions
ans+0.104
 bat+0.088
 waves+0.087
 blow+0.085
 next+0.085
Neg logit contributions
 unlike-0.208
 matter-0.198
 tactic-0.196
 blonde-0.192
 Where-0.192
Token with
Token ID351
Feature activation-8.46e-03

Pos logit contributions
ans+0.015
 waves+0.013
ck+0.013
 blow+0.013
orse+0.013
Neg logit contributions
 unlike-0.020
 tactic-0.019
 terribly-0.019
 Where-0.018
mate-0.018
Token oranges
Token ID48389
Feature activation+1.22e-02

Pos logit contributions
 unlike+0.017
 tactic+0.016
 matter+0.016
 blonde+0.016
 Where+0.015
Neg logit contributions
ans-0.008
 next-0.008
 blow-0.008
 waves-0.008
 bat-0.008
Token.
Token ID13
Feature activation-3.35e-02

Pos logit contributions
ans+0.015
ck+0.014
 blow+0.014
 waves+0.013
 bat+0.013
Neg logit contributions
 unlike-0.021
 tactic-0.019
ifying-0.019
 Where-0.019
 terribly-0.019
Token She
Token ID1375
Feature activation-4.31e-02

Pos logit contributions
 unlike+0.054
 matter+0.053
 tactic+0.052
rified+0.050
ifying+0.050
Neg logit contributions
 next-0.044
 put-0.043
ans-0.043
 waves-0.043
 bat-0.042
Token liked
Token ID8288
Feature activation-2.49e-02

Pos logit contributions
 unlike+0.065
 tactic+0.060
 matter+0.057
 Tears+0.057
 Was+0.057
Neg logit contributions
ans-0.066
irts-0.058
ites-0.057
 waves-0.057
 blow-0.056
Token to
Token ID284
Feature activation-5.90e-02

Pos logit contributions
rified+0.052
 tactic+0.052
 unlike+0.052
ifying+0.050
 Tears+0.050
Neg logit contributions
 next-0.021
 put-0.021
 blow-0.020
 waves-0.019
 bat-0.019
Token sort
Token ID3297
Feature activation+3.33e-02

Pos logit contributions
 unlike+0.112
 tactic+0.111
 matter+0.109
 blonde+0.105
 Where+0.105
Neg logit contributions
ans-0.061
 next-0.057
 waves-0.055
ites-0.053
 blow-0.053
Token things
Token ID1243
Feature activation-1.60e-01

Pos logit contributions
 next+0.022
 put+0.022
 waves+0.020
ans+0.020
 each+0.019
Neg logit contributions
 unlike-0.076
 tactic-0.075
rified-0.074
 Tears-0.073
 Where-0.072
Token.
Token ID13
Feature activation-2.91e-02

Pos logit contributions
 unlike+0.321
 tactic+0.315
rified+0.306
 matter+0.303
 Where+0.303
Neg logit contributions
 next-0.146
ans-0.142
 put-0.140
 blow-0.139
 waves-0.138
Token When
Token ID1649
Feature activation-1.27e-01

Pos logit contributions
 unlike+0.053
 matter+0.051
 tactic+0.051
rified+0.049
ifying+0.049
Neg logit contributions
 next-0.033
ans-0.032
 put-0.032
 waves-0.031
 bat-0.031
Token selling
Token ID6301
Feature activation+2.88e-02

Pos logit contributions
 unlike+0.061
 tactic+0.060
 matter+0.058
 Where+0.057
 Tears+0.057
Neg logit contributions
ans-0.039
 blow-0.037
 next-0.037
 bat-0.037
 waves-0.037
Token.
Token ID13
Feature activation-3.27e-02

Pos logit contributions
 next+0.031
ans+0.031
 waves+0.030
 put+0.030
 blow+0.029
Neg logit contributions
 unlike-0.053
 tactic-0.052
rified-0.050
 Where-0.050
 Tears-0.050
Token Bobby
Token ID17155
Feature activation-6.71e-02

Pos logit contributions
 unlike+0.052
 tactic+0.051
 matter+0.049
rified+0.048
 Tears+0.047
Neg logit contributions
 next-0.043
ans-0.043
 put-0.042
 waves-0.042
 bat-0.042
Token never
Token ID1239
Feature activation+8.87e-02

Pos logit contributions
 unlike+0.101
 Tears+0.097
rified+0.097
 terribly+0.096
 tactic+0.096
Neg logit contributions
ans-0.093
 blow-0.087
irts-0.087
 bat-0.086
 next-0.086
Token stopped
Token ID5025
Feature activation-4.22e-02

Pos logit contributions
ans+0.101
 next+0.093
 waves+0.092
 blow+0.091
 bat+0.091
Neg logit contributions
 unlike-0.161
 matter-0.158
 tactic-0.157
 Where-0.153
rified-0.151
Token being
Token ID852
Feature activation+4.91e-02

Pos logit contributions
 unlike+0.076
 tactic+0.075
rified+0.073
 matter+0.071
 Where+0.071
Neg logit contributions
 next-0.048
ans-0.047
 put-0.046
 waves-0.046
 blow-0.046
Token brave
Token ID14802
Feature activation-1.43e-02

Pos logit contributions
 next+0.063
 put+0.062
ans+0.061
 waves+0.061
 bat+0.059
Neg logit contributions
 tactic-0.081
 unlike-0.081
rified-0.077
 Where-0.076
 Tears-0.076
Token and
Token ID290
Feature activation+9.40e-02

Pos logit contributions
 unlike+0.029
 tactic+0.029
 Tears+0.028
 Where+0.028
rified+0.027
Neg logit contributions
 next-0.013
 waves-0.012
 put-0.012
 bat-0.012
 blow-0.012
Token helping
Token ID5742
Feature activation-4.62e-02

Pos logit contributions
 next+0.103
ans+0.101
 put+0.101
 waves+0.100
 blow+0.100
Neg logit contributions
 unlike-0.171
 tactic-0.167
rified-0.161
 matter-0.160
 Where-0.159
Token his
Token ID465
Feature activation-8.40e-02

Pos logit contributions
 unlike+0.088
 tactic+0.087
rified+0.086
 blonde+0.085
 Where+0.083
Neg logit contributions
 next-0.049
 put-0.047
 bat-0.046
 blow-0.045
 each-0.044
Token mom
Token ID1995
Feature activation+2.59e-02

Pos logit contributions
 unlike+0.157
 tactic+0.154
 matter+0.148
rified+0.147
 Where+0.147
Neg logit contributions
ans-0.088
 next-0.086
 blow-0.085
 waves-0.084
 put-0.084
TokenOnce
Token ID7454
Feature activation-8.17e-02

Pos logit contributions
 next+0.060
ans+0.058
 waves+0.058
 put+0.058
 bat+0.058
Neg logit contributions
 unlike-0.100
 tactic-0.097
 matter-0.094
rified-0.093
 Tears-0.091
Token there
Token ID612
Feature activation-8.76e-02

Pos logit contributions
 unlike+0.087
 tactic+0.082
rified+0.078
 matter+0.077
 Where+0.075
Neg logit contributions
 next-0.154
 put-0.150
ans-0.149
 bat-0.148
 blow-0.147
Token was
Token ID373
Feature activation-7.25e-02

Pos logit contributions
ifying+0.195
 unlike+0.190
 tactic+0.183
rier+0.180
 superpower+0.180
Neg logit contributions
 put-0.077
 next-0.072
 and-0.071
 each-0.062
 slide-0.062
Token a
Token ID257
Feature activation-7.77e-03

Pos logit contributions
 unlike+0.126
 tactic+0.122
 matter+0.117
 Where+0.117
rified+0.117
Neg logit contributions
ans-0.087
 next-0.083
 waves-0.082
 blow-0.082
 put-0.080
Token mom
Token ID1995
Feature activation+3.90e-02

Pos logit contributions
 unlike+0.012
 matter+0.011
 tactic+0.011
 Where+0.011
 blonde+0.011
Neg logit contributions
ans-0.011
 waves-0.010
 next-0.010
 goes-0.010
ck-0.010
Tokenmy
Token ID1820
Feature activation-2.50e-02

Pos logit contributions
ans+0.071
 waves+0.067
ars+0.067
irts+0.065
ck+0.064
Neg logit contributions
 matter-0.042
 unlike-0.040
 tactic-0.038
 Where-0.037
 Thompson-0.035
Token and
Token ID290
Feature activation+1.11e-01

Pos logit contributions
 unlike+0.036
 Tears+0.034
 perspective+0.034
 terribly+0.033
 matter+0.033
Neg logit contributions
ans-0.036
 waves-0.034
irts-0.034
ars-0.033
ites-0.032
Token a
Token ID257
Feature activation-8.46e-03

Pos logit contributions
ans+0.140
 waves+0.126
 next+0.125
ites+0.120
 goes+0.119
Neg logit contributions
 unlike-0.192
 Where-0.182
 tactic-0.181
 matter-0.181
 blonde-0.180
Token little
Token ID1310
Feature activation-6.03e-02

Pos logit contributions
 unlike+0.016
 tactic+0.016
eway+0.016
rified+0.016
 Tears+0.015
Neg logit contributions
 bat-0.009
 put-0.009
 blow-0.009
 next-0.009
 se-0.009
Token girl
Token ID2576
Feature activation-1.62e-01

Pos logit contributions
 unlike+0.107
 tactic+0.104
eway+0.100
rified+0.099
 matter+0.099
Neg logit contributions
 next-0.072
 bat-0.071
 blow-0.071
 put-0.071
 waves-0.070
Token.
Token ID13
Feature activation-2.08e-02

Pos logit contributions
 unlike+0.305
 tactic+0.296
 Where+0.291
 matter+0.287
 Tears+0.287
Neg logit contributions
ans-0.168
 waves-0.160
 next-0.156
 blow-0.156
 bat-0.154
Token hospital
Token ID4436
Feature activation-4.87e-02

Pos logit contributions
ans+0.079
 blow+0.076
 next+0.076
 put+0.075
 waves+0.074
Neg logit contributions
 unlike-0.117
 tactic-0.113
 Where-0.111
rified-0.110
 matter-0.109
Token,
Token ID11
Feature activation-3.28e-02

Pos logit contributions
 unlike+0.099
 tactic+0.098
 Tears+0.095
 Thompson+0.093
 terribly+0.093
Neg logit contributions
ans-0.044
 waves-0.043
ites-0.039
 blow-0.039
 bat-0.039
Token but
Token ID475
Feature activation-1.03e-01

Pos logit contributions
 unlike+0.046
 tactic+0.043
 Tears+0.042
rified+0.041
 Where+0.041
Neg logit contributions
ans-0.052
 waves-0.050
 next-0.048
 blow-0.048
 goes-0.048
Token it
Token ID340
Feature activation+2.40e-02

Pos logit contributions
 tactic+0.231
 unlike+0.224
 perspective+0.219
 blonde+0.216
rified+0.216
Neg logit contributions
 next-0.087
 put-0.082
 bat-0.073
 blow-0.072
 then-0.072
Token was
Token ID373
Feature activation-7.85e-02

Pos logit contributions
 next+0.027
 put+0.026
 goes+0.024
 blow+0.024
 bat+0.023
Neg logit contributions
 tactic-0.047
 perspective-0.046
 unlike-0.045
 Where-0.045
 blonde-0.044
Token too
Token ID1165
Feature activation-7.56e-02

Pos logit contributions
 tactic+0.147
rified+0.139
 unlike+0.138
 Where+0.138
 perspective+0.133
Neg logit contributions
 next-0.096
 put-0.091
 bat-0.088
 blow-0.086
 goes-0.081
Token late
Token ID2739
Feature activation-3.06e-02

Pos logit contributions
 unlike+0.085
 terribly+0.080
 slightly+0.079
 matter+0.077
 total+0.075
Neg logit contributions
ans-0.140
 waves-0.134
orse-0.127
ites-0.126
 blow-0.126
Token.
Token ID13
Feature activation-2.29e-02

Pos logit contributions
 unlike+0.058
 tactic+0.056
 Tears+0.054
 Where+0.054
rified+0.054
Neg logit contributions
ans-0.033
 waves-0.031
 bat-0.030
 blow-0.030
 put-0.029
Token Lily
Token ID20037
Feature activation+2.94e-02

Pos logit contributions
 tactic+0.049
 unlike+0.049
 matter+0.048
 perspective+0.046
mate+0.046
Neg logit contributions
 next-0.019
 put-0.018
 bat-0.018
 blow-0.016
ans-0.015
Token had
Token ID550
Feature activation+4.25e-02

Pos logit contributions
ans+0.039
irts+0.035
ites+0.035
 waves+0.035
 blow+0.035
Neg logit contributions
 terribly-0.048
 Tears-0.048
rified-0.048
 unlike-0.047
 must-0.046
Token lost
Token ID2626
Feature activation-4.46e-02

Pos logit contributions
ans+0.062
 waves+0.057
 blow+0.056
 next+0.056
ck+0.054
Neg logit contributions
 unlike-0.066
 matter-0.061
 Tears-0.061
 tactic-0.060
 Where-0.060
Token erupt
Token ID17866
Feature activation+5.39e-02

Pos logit contributions
 unlike+0.125
 tactic+0.122
rified+0.118
 matter+0.117
 Where+0.117
Neg logit contributions
 next-0.063
ans-0.063
 waves-0.061
 blow-0.061
 put-0.061
Tokening
Token ID278
Feature activation-4.49e-02

Pos logit contributions
ans+0.069
 bat+0.068
 next+0.067
 put+0.067
 blow+0.067
Neg logit contributions
 unlike-0.087
 tactic-0.082
rified-0.082
 matter-0.081
 Tears-0.080
Token.
Token ID13
Feature activation-2.99e-02

Pos logit contributions
 unlike+0.088
 tactic+0.087
rified+0.084
 matter+0.083
 Where+0.083
Neg logit contributions
 next-0.042
ans-0.042
 waves-0.041
 blow-0.041
 put-0.041
Token He
Token ID679
Feature activation-1.01e-01

Pos logit contributions
 matter+0.049
 tactic+0.048
 unlike+0.046
mate+0.046
 avail+0.045
Neg logit contributions
 next-0.040
 waves-0.038
 put-0.038
 bat-0.038
 goes-0.037
Token sh
Token ID427
Feature activation+6.18e-02

Pos logit contributions
 unlike+0.165
 Tears+0.153
 terribly+0.152
 tactic+0.151
 matter+0.149
Neg logit contributions
ans-0.136
irts-0.127
ites-0.126
 blow-0.120
 next-0.118
Tokenivered
Token ID6396
Feature activation-1.16e-01

Pos logit contributions
ans+0.118
 next+0.114
ck+0.112
 put+0.112
 blow+0.111
Neg logit contributions
 unlike-0.064
 tactic-0.063
 matter-0.062
 Where-0.061
rified-0.057
Token in
Token ID287
Feature activation-4.75e-02

Pos logit contributions
 tactic+0.256
 unlike+0.255
 Where+0.247
 matter+0.246
rified+0.245
Neg logit contributions
 next-0.085
 put-0.079
 waves-0.077
 blow-0.077
 bat-0.076
Token terror
Token ID3246
Feature activation-1.16e-01

Pos logit contributions
 unlike+0.104
 tactic+0.103
rified+0.099
 Where+0.097
 Thompson+0.097
Neg logit contributions
 next-0.039
 put-0.038
 bat-0.036
 waves-0.036
 blow-0.035
Token,
Token ID11
Feature activation-3.48e-02

Pos logit contributions
 unlike+0.231
rified+0.220
 tactic+0.215
 terribly+0.215
 matter+0.213
Neg logit contributions
ans-0.112
 bat-0.106
 goes-0.104
 dads-0.103
 waves-0.103
Token thankful
Token ID25535
Feature activation+1.38e-02

Pos logit contributions
 unlike+0.062
 Tears+0.057
 tactic+0.057
rified+0.056
 terribly+0.055
Neg logit contributions
ans-0.043
 waves-0.040
ites-0.039
 goes-0.039
 bat-0.039
Token that
Token ID326
Feature activation+1.11e-01

Pos logit contributions
ans+0.012
 next+0.012
 waves+0.011
 blow+0.011
 bat+0.011
Neg logit contributions
 unlike-0.029
 tactic-0.028
rified-0.027
 matter-0.027
 Tears-0.027
Token was
Token ID373
Feature activation-8.41e-02

Pos logit contributions
 unlike+0.063
 must+0.062
 terribly+0.060
 matter+0.060
 shouldn+0.058
Neg logit contributions
ans-0.070
ites-0.063
irts-0.062
line-0.059
 pige-0.058
Token reminded
Token ID14516
Feature activation-8.45e-02

Pos logit contributions
 unlike+0.139
 terribly+0.129
 tactic+0.128
 Tears+0.127
 kinda+0.124
Neg logit contributions
ans-0.115
 waves-0.108
 blow-0.104
 goes-0.103
 bat-0.102
Token that
Token ID326
Feature activation+1.04e-01

Pos logit contributions
 unlike+0.169
 tactic+0.162
mate+0.161
 Tears+0.161
 terribly+0.160
Neg logit contributions
ans-0.084
 waves-0.079
 blow-0.076
 bat-0.075
 goes-0.072
Token no
Token ID645
Feature activation-4.12e-02

Pos logit contributions
ans+0.128
 waves+0.111
 blow+0.110
irts+0.106
ck+0.106
Neg logit contributions
 unlike-0.193
 tactic-0.181
 matter-0.181
 Where-0.176
 Tears-0.176
Token matter
Token ID2300
Feature activation-2.79e-01

Pos logit contributions
 unlike+0.060
ifying+0.058
 Terr+0.058
rified+0.057
eway+0.056
Neg logit contributions
 next-0.067
 put-0.063
 waves-0.060
 and-0.060
 blow-0.060
Token what
Token ID644
Feature activation-8.69e-02

Pos logit contributions
 unlike+0.504
 tactic+0.490
rified+0.482
 blonde+0.473
 Terr+0.473
Neg logit contributions
 next-0.332
 put-0.312
 each-0.300
 and-0.297
 blow-0.294
Token,
Token ID11
Feature activation-3.50e-02

Pos logit contributions
 tactic+0.147
 unlike+0.147
rified+0.143
 Where+0.139
ifying+0.139
Neg logit contributions
 next-0.109
 put-0.105
 goes-0.102
 bat-0.102
 waves-0.102
Token she
Token ID673
Feature activation+1.97e-02

Pos logit contributions
 tactic+0.072
 unlike+0.072
rified+0.070
 Where+0.068
eway+0.068
Neg logit contributions
 next-0.032
 put-0.029
 bat-0.028
 blow-0.028
 waves-0.028
Token had
Token ID550
Feature activation+3.77e-02

Pos logit contributions
ans+0.029
 bat+0.026
 waves+0.026
 blow+0.025
irts+0.025
Neg logit contributions
 unlike-0.030
 matter-0.028
rified-0.028
 terribly-0.027
 tactic-0.027
Token a
Token ID257
Feature activation-1.63e-02

Pos logit contributions
 next+0.048
ans+0.048
 put+0.048
 waves+0.047
 bat+0.047
Neg logit contributions
 unlike-0.062
 tactic-0.060
rified-0.058
 matter-0.057
 Where-0.057
Token family
Token ID1641
Feature activation+3.01e-02

Pos logit contributions
eway+0.029
 unlike+0.029
rified+0.029
 tactic+0.028
 Tears+0.027
Neg logit contributions
 bat-0.020
 next-0.020
 put-0.020
 se-0.019
 blow-0.018
Token could
Token ID714
Feature activation-9.11e-02

Pos logit contributions
 next+0.025
 put+0.025
ans+0.024
 blow+0.024
 waves+0.024
Neg logit contributions
 unlike-0.029
 tactic-0.029
rified-0.028
 Tears-0.027
 matter-0.027
Token help
Token ID1037
Feature activation+2.14e-02

Pos logit contributions
 unlike+0.173
 Where+0.165
 matter+0.163
 tactic+0.163
mate+0.160
Neg logit contributions
ans-0.100
 waves-0.087
ites-0.085
 next-0.083
irts-0.082
Token stir
Token ID11240
Feature activation+9.99e-02

Pos logit contributions
ans+0.020
 waves+0.020
 next+0.019
 bat+0.019
 blow+0.019
Neg logit contributions
 unlike-0.042
 tactic-0.041
 matter-0.040
 Where-0.040
 Tears-0.039
Token the
Token ID262
Feature activation+7.00e-02

Pos logit contributions
ans+0.085
 waves+0.081
 next+0.081
 bat+0.080
 blow+0.080
Neg logit contributions
 unlike-0.204
 tactic-0.201
 matter-0.193
rified-0.192
 Tears-0.192
Token soup
Token ID17141
Feature activation-3.46e-02

Pos logit contributions
 next+0.072
 waves+0.070
ans+0.070
 blow+0.069
 put+0.069
Neg logit contributions
 unlike-0.135
 tactic-0.132
rified-0.129
 Where-0.127
 matter-0.127
Token too
Token ID1165
Feature activation-7.73e-02

Pos logit contributions
 unlike+0.065
 tactic+0.063
 matter+0.059
 Tears+0.059
 terribly+0.059
Neg logit contributions
ans-0.036
 waves-0.034
 blow-0.033
 bat-0.032
ck-0.032
Token.
Token ID13
Feature activation-3.41e-02

Pos logit contributions
 unlike+0.140
 tactic+0.131
 terribly+0.131
 Tears+0.129
 matter+0.129
Neg logit contributions
ans-0.089
 waves-0.084
 blow-0.081
 goes-0.078
 bat-0.078
Token
Token ID198
Feature activation+1.46e-02

Pos logit contributions
 unlike+0.053
 matter+0.051
 tactic+0.050
rified+0.050
ifying+0.049
Neg logit contributions
 next-0.049
 put-0.048
 bat-0.047
 blow-0.047
 goes-0.047
Token
Token ID198
Feature activation+1.39e-02

Pos logit contributions
ans+0.015
 waves+0.015
 blow+0.015
 next+0.014
 bat+0.014
Neg logit contributions
 tactic-0.027
 unlike-0.027
 Where-0.026
 Tears-0.026
rified-0.025
TokenHer
Token ID9360
Feature activation-6.64e-02

Pos logit contributions
ans+0.018
 waves+0.017
 blow+0.016
 next+0.016
 bat+0.016
Neg logit contributions
 unlike-0.023
 tactic-0.023
 matter-0.022
 Where-0.022
 Tears-0.022
Token mom
Token ID1995
Feature activation+3.23e-02

Pos logit contributions
rified+0.170
 unlike+0.169
eway+0.164
ifying+0.163
 tactic+0.162
Neg logit contributions
 next-0.044
 put-0.037
 bat-0.034
 blow-0.030
 and-0.029
Token
Token ID198
Feature activation+2.92e-02

Pos logit contributions
 unlike+0.031
 matter+0.030
 tactic+0.030
rified+0.029
mate+0.029
Neg logit contributions
 next-0.018
 put-0.018
 bat-0.018
ans-0.018
 waves-0.018
Token
Token ID198
Feature activation+2.73e-02

Pos logit contributions
ans+0.029
 waves+0.029
 blow+0.028
 next+0.027
 put+0.027
Neg logit contributions
 unlike-0.056
 tactic-0.055
 Where-0.054
 Tears-0.053
rified-0.052
TokenSo
Token ID2396
Feature activation-8.09e-02

Pos logit contributions
ans+0.035
 waves+0.035
 blow+0.034
 next+0.034
 put+0.034
Neg logit contributions
 unlike-0.044
 tactic-0.043
 Where-0.042
 matter-0.041
rified-0.041
Token Sam
Token ID3409
Feature activation+7.15e-02

Pos logit contributions
 unlike+0.166
 tactic+0.164
rified+0.156
 emanating+0.155
 blonde+0.155
Neg logit contributions
 next-0.076
 put-0.073
 bat-0.070
 waves-0.069
 blow-0.069
Token grabbed
Token ID13176
Feature activation+1.02e-01

Pos logit contributions
ans+0.085
irts+0.080
ites+0.076
 pige+0.073
 bread+0.073
Neg logit contributions
 must-0.119
rified-0.117
 unlike-0.116
 terribly-0.113
 matter-0.113
Token his
Token ID465
Feature activation-6.97e-02

Pos logit contributions
ans+0.130
 next+0.127
 blow+0.125
 waves+0.124
 bat+0.123
Neg logit contributions
 unlike-0.167
 tactic-0.162
rified-0.157
 matter-0.156
 Tears-0.155
Token bag
Token ID6131
Feature activation-1.68e-02

Pos logit contributions
 unlike+0.128
 tactic+0.125
 matter+0.120
 Where+0.119
rified+0.119
Neg logit contributions
ans-0.074
 next-0.072
 blow-0.072
 put-0.071
 bat-0.070
Token and
Token ID290
Feature activation+1.08e-01

Pos logit contributions
 unlike+0.036
 terribly+0.035
 tactic+0.035
 Tears+0.034
 slightly+0.034
Neg logit contributions
ans-0.013
 blow-0.013
 waves-0.013
ck-0.012
irts-0.012
Token started
Token ID2067
Feature activation-4.58e-02

Pos logit contributions
ans+0.109
is+0.096
 next+0.090
irts+0.089
ites+0.087
Neg logit contributions
 unlike-0.211
 Tears-0.202
 tactic-0.202
rified-0.201
 blonde-0.199
Token walking
Token ID6155
Feature activation+9.65e-02

Pos logit contributions
 unlike+0.073
 tactic+0.070
 Tears+0.069
 Where+0.067
 blonde+0.067
Neg logit contributions
ans-0.062
 next-0.056
 waves-0.056
 blow-0.055
ites-0.054
Token.
Token ID13
Feature activation-2.29e-02

Pos logit contributions
ans+0.097
 next+0.094
 blow+0.094
 waves+0.093
 put+0.093
Neg logit contributions
 unlike-0.184
 tactic-0.181
rified-0.174
 matter-0.173
 Where-0.172
Token with
Token ID351
Feature activation-8.78e-03

Pos logit contributions
 unlike+0.253
 tactic+0.252
 matter+0.247
 blonde+0.245
 Tears+0.243
Neg logit contributions
 next-0.055
 bat-0.048
 put-0.047
 blow-0.046
 waves-0.046
Token his
Token ID465
Feature activation-7.21e-02

Pos logit contributions
 unlike+0.015
 tactic+0.014
rified+0.014
 matter+0.014
 Where+0.014
Neg logit contributions
ans-0.011
 next-0.011
 blow-0.011
 waves-0.011
 put-0.011
Token teeth
Token ID9941
Feature activation-1.48e-01

Pos logit contributions
eway+0.148
rified+0.146
 unlike+0.141
 tactic+0.139
 Where+0.139
Neg logit contributions
 put-0.078
 bat-0.076
 next-0.075
 and-0.073
 blow-0.070
Token and
Token ID290
Feature activation+1.08e-01

Pos logit contributions
 unlike+0.355
 tactic+0.342
 Tears+0.334
rified+0.333
 terribly+0.332
Neg logit contributions
ans-0.085
 waves-0.071
 blow-0.069
 bat-0.065
 goes-0.064
Token ran
Token ID4966
Feature activation+1.84e-01

Pos logit contributions
ans+0.090
 next+0.085
 blow+0.078
 waves+0.077
 bat+0.077
Neg logit contributions
 unlike-0.231
 tactic-0.222
 matter-0.217
 Tears-0.216
 blonde-0.214
Token away
Token ID1497
Feature activation-2.01e-01

Pos logit contributions
 next+0.197
ans+0.193
 put+0.192
 waves+0.190
 bat+0.190
Neg logit contributions
 unlike-0.346
 tactic-0.338
rified-0.327
 matter-0.326
 Tears-0.323
Token to
Token ID284
Feature activation-6.26e-02

Pos logit contributions
 unlike+0.453
 tactic+0.448
rified+0.447
 matter+0.447
ifying+0.440
Neg logit contributions
 next-0.151
 put-0.141
 bat-0.130
 and-0.127
 waves-0.126
Token play
Token ID711
Feature activation+1.32e-01

Pos logit contributions
 tactic+0.124
 unlike+0.124
 matter+0.121
 Where+0.120
 Tears+0.119
Neg logit contributions
ans-0.059
 next-0.055
ites-0.052
 waves-0.052
 blow-0.052
Token with
Token ID351
Feature activation-8.59e-03

Pos logit contributions
 next+0.093
 put+0.085
 bat+0.084
 and+0.083
 waves+0.069
Neg logit contributions
 unlike-0.315
ifying-0.310
 matter-0.306
rified-0.303
 tactic-0.301
Token it
Token ID340
Feature activation+2.14e-02

Pos logit contributions
 unlike+0.018
 tactic+0.017
 matter+0.017
rified+0.017
 Tears+0.017
Neg logit contributions
 next-0.008
ans-0.008
 waves-0.008
 put-0.007
 bat-0.007
Token.
Token ID13
Feature activation-1.83e-02

Pos logit contributions
 next+0.015
 put+0.013
 bat+0.012
 and+0.012
 waves+0.012
Neg logit contributions
 unlike-0.050
 tactic-0.049
rified-0.048
 Where-0.048
ifying-0.048
Token it
Token ID340
Feature activation+2.20e-02

Pos logit contributions
 tactic+0.021
 unlike+0.021
 Thompson+0.020
 Where+0.020
rified+0.020
Neg logit contributions
 put-0.010
 bat-0.010
 next-0.010
 waves-0.010
 blow-0.010
Token.
Token ID13
Feature activation-2.33e-02

Pos logit contributions
 next+0.017
 put+0.016
 and+0.015
 blow+0.014
 bat+0.014
Neg logit contributions
 tactic-0.049
 unlike-0.048
 Where-0.048
rified-0.047
ifying-0.047
Token The
Token ID383
Feature activation+5.31e-02

Pos logit contributions
 unlike+0.043
 tactic+0.043
 matter+0.042
rified+0.041
 perspective+0.040
Neg logit contributions
 next-0.025
 put-0.024
 bat-0.024
 waves-0.024
 blow-0.024
Token frog
Token ID21264
Feature activation-3.65e-02

Pos logit contributions
 next+0.058
 waves+0.057
 bat+0.053
 put+0.052
 dove+0.052
Neg logit contributions
 unlike-0.111
ifying-0.101
eway-0.101
 tactic-0.099
 terribly-0.098
Token jumped
Token ID11687
Feature activation+1.68e-01

Pos logit contributions
 unlike+0.067
 tactic+0.065
rified+0.062
 matter+0.062
 Tears+0.062
Neg logit contributions
ans-0.040
 next-0.039
 waves-0.038
 blow-0.038
 bat-0.038
Token away
Token ID1497
Feature activation-1.98e-01

Pos logit contributions
 next+0.154
 bat+0.143
 waves+0.142
 put+0.142
 blow+0.139
Neg logit contributions
 tactic-0.347
 unlike-0.345
 matter-0.337
 Where-0.330
 blonde-0.326
Token,
Token ID11
Feature activation-3.14e-02

Pos logit contributions
 tactic+0.469
 unlike+0.467
 Where+0.461
 matter+0.457
rified+0.455
Neg logit contributions
 next-0.117
 put-0.108
 blow-0.102
 bat-0.101
 waves-0.101
Token but
Token ID475
Feature activation-1.03e-01

Pos logit contributions
 unlike+0.056
 tactic+0.053
 Tears+0.052
rified+0.052
 matter+0.051
Neg logit contributions
ans-0.037
 next-0.035
 waves-0.035
 bat-0.034
 blow-0.034
Token the
Token ID262
Feature activation+6.99e-02

Pos logit contributions
 unlike+0.218
 tactic+0.217
 matter+0.208
 Where+0.207
rified+0.207
Neg logit contributions
 next-0.083
 put-0.079
 blow-0.077
 waves-0.077
ans-0.076
Token boy
Token ID2933
Feature activation-1.13e-01

Pos logit contributions
 next+0.080
 waves+0.076
 bat+0.074
 put+0.073
 blow+0.071
Neg logit contributions
 unlike-0.140
 tactic-0.129
ifying-0.128
eway-0.128
 Tears-0.126
Token kept
Token ID4030
Feature activation+4.15e-02

Pos logit contributions
 unlike+0.200
 tactic+0.199
 Where+0.191
 matter+0.188
rified+0.187
Neg logit contributions
 put-0.131
 next-0.130
 blow-0.125
 waves-0.125
ans-0.124
Token won
Token ID1839
Feature activation-5.90e-02

Pos logit contributions
ans+0.021
 pige+0.018
 dads+0.018
ites+0.018
 next+0.018
Neg logit contributions
 unlike-0.024
 tactic-0.023
 matter-0.022
rified-0.022
 Tears-0.022
Token't
Token ID470
Feature activation+6.89e-03

Pos logit contributions
 unlike+0.109
 Tears+0.103
 tactic+0.103
 matter+0.103
 terribly+0.101
Neg logit contributions
ans-0.067
 waves-0.061
 dads-0.059
ites-0.058
 dove-0.058
Token be
Token ID307
Feature activation+1.22e-01

Pos logit contributions
ans+0.008
ites+0.007
 waves+0.007
irts+0.007
 pige+0.006
Neg logit contributions
 matter-0.013
 unlike-0.013
 tactic-0.013
 terribly-0.012
mate-0.012
Token sm
Token ID895
Feature activation+3.77e-03

Pos logit contributions
ans+0.162
 next+0.160
 waves+0.158
 blow+0.155
 bat+0.154
Neg logit contributions
 unlike-0.196
 tactic-0.188
 matter-0.183
 Tears-0.180
rified-0.180
Tokenelly
Token ID6148
Feature activation-2.49e-02

Pos logit contributions
ans+0.006
 next+0.006
 blow+0.005
 waves+0.005
 put+0.005
Neg logit contributions
 unlike-0.005
 tactic-0.005
rified-0.005
 matter-0.005
 Where-0.005
Token anymore
Token ID7471
Feature activation-2.10e-01

Pos logit contributions
 unlike+0.047
 tactic+0.046
 Tears+0.044
rified+0.044
 Where+0.044
Neg logit contributions
 next-0.026
 waves-0.025
 put-0.025
ans-0.025
 bat-0.025
Token!"
Token ID2474
Feature activation+4.60e-02

Pos logit contributions
 unlike+0.447
 tactic+0.444
rified+0.433
 matter+0.427
 Where+0.425
Neg logit contributions
 next-0.168
 put-0.158
ans-0.155
 bat-0.155
 blow-0.154
Token
Token ID198
Feature activation+1.70e-02

Pos logit contributions
ans+0.047
 next+0.047
 waves+0.045
 blow+0.045
 put+0.045
Neg logit contributions
 unlike-0.087
 tactic-0.085
rified-0.082
 matter-0.082
 Where-0.081
Token
Token ID198
Feature activation+1.69e-02

Pos logit contributions
ans+0.017
 waves+0.017
 blow+0.017
 next+0.016
 put+0.016
Neg logit contributions
 unlike-0.032
 tactic-0.032
 Where-0.031
 Tears-0.031
rified-0.030
TokenTim
Token ID14967
Feature activation+4.22e-02

Pos logit contributions
ans+0.020
 waves+0.019
 blow+0.019
 next+0.019
 put+0.018
Neg logit contributions
 unlike-0.029
 tactic-0.029
 Where-0.028
 Tears-0.028
 matter-0.028
Tokenmy
Token ID1820
Feature activation-4.71e-02

Pos logit contributions
ans+0.048
 next+0.045
 bat+0.045
 blow+0.045
 waves+0.044
Neg logit contributions
 tactic-0.073
 unlike-0.072
 Tears-0.072
rified-0.072
 matter-0.071
Token.
Token ID13
Feature activation-1.70e-02

Pos logit contributions
ans+0.034
 blow+0.032
 waves+0.032
 bat+0.031
 put+0.030
Neg logit contributions
 unlike-0.064
 tactic-0.061
 Tears-0.060
rified-0.059
 matter-0.058
Token Lily
Token ID20037
Feature activation+3.29e-02

Pos logit contributions
 unlike+0.036
 tactic+0.035
 matter+0.035
rified+0.034
 perspective+0.034
Neg logit contributions
 put-0.015
 next-0.014
 bat-0.014
 blow-0.013
 waves-0.013
Token said
Token ID531
Feature activation+7.07e-02

Pos logit contributions
ans+0.045
 waves+0.042
ites+0.041
irts+0.041
 next+0.041
Neg logit contributions
 unlike-0.052
 Tears-0.049
 tactic-0.049
 matter-0.047
 terribly-0.047
Token,
Token ID11
Feature activation-3.14e-02

Pos logit contributions
 next+0.048
 put+0.046
 bat+0.044
 dove+0.042
 and+0.042
Neg logit contributions
 tactic-0.164
 matter-0.163
 unlike-0.161
ifying-0.160
 perspective-0.157
Token "
Token ID366
Feature activation-8.77e-02

Pos logit contributions
 unlike+0.045
 tactic+0.045
 matter+0.043
rified+0.042
 Tears+0.041
Neg logit contributions
 next-0.047
 blow-0.045
 put-0.045
 bat-0.045
ans-0.045
TokenWow
Token ID22017
Feature activation-2.11e-01

Pos logit contributions
 unlike+0.133
 tactic+0.130
rified+0.123
 matter+0.122
 Tears+0.122
Neg logit contributions
 next-0.124
ans-0.124
 blow-0.122
 waves-0.121
 put-0.121
Token!
Token ID0
Feature activation-9.79e-03

Pos logit contributions
 unlike+0.381
 tactic+0.372
rified+0.362
 matter+0.359
 Where+0.358
Neg logit contributions
ans-0.232
 next-0.225
 waves-0.222
 blow-0.220
 put-0.218
Token Look
Token ID6803
Feature activation-8.50e-02

Pos logit contributions
 tactic+0.015
 unlike+0.014
 matter+0.014
 perspective+0.013
eway+0.013
Neg logit contributions
 bat-0.015
 next-0.014
 blow-0.014
 put-0.014
 waves-0.014
Token at
Token ID379
Feature activation-3.80e-02

Pos logit contributions
 unlike+0.172
 tactic+0.170
 matter+0.161
rified+0.159
 Tears+0.159
Neg logit contributions
 next-0.080
 put-0.079
ans-0.079
 waves-0.077
 bat-0.076
Token the
Token ID262
Feature activation+7.37e-02

Pos logit contributions
 unlike+0.077
 Where+0.073
 Tears+0.072
rified+0.071
 matter+0.071
Neg logit contributions
ans-0.032
ars-0.031
ites-0.030
 goes-0.030
ck-0.029
Token big
Token ID1263
Feature activation+4.33e-02

Pos logit contributions
 next+0.101
 waves+0.095
 blow+0.094
 water+0.093
 and+0.091
Neg logit contributions
eway-0.128
 tactic-0.124
ifying-0.123
spe-0.123
rified-0.123
Token magnet
Token ID19972
Feature activation-1.77e-01

Pos logit contributions
rified+0.090
eway+0.090
 unlike+0.085
ifying+0.084
 tactic+0.084
Neg logit contributions
 next-0.137
 bat-0.128
 put-0.127
 waves-0.125
 and-0.122
Token wouldn
Token ID3636
Feature activation-1.46e-01

Pos logit contributions
 tactic+0.322
 unlike+0.321
 Where+0.315
eway+0.314
rified+0.312
Neg logit contributions
 put-0.202
 next-0.195
 bat-0.191
 blow-0.189
 waves-0.185
Token't
Token ID470
Feature activation-1.62e-04

Pos logit contributions
 unlike+0.120
 terribly+0.113
 Where+0.113
mate+0.112
 Tears+0.099
Neg logit contributions
ans-0.330
 dads-0.308
 waves-0.302
 pige-0.297
ites-0.296
Token stick
Token ID4859
Feature activation+5.77e-02

Pos logit contributions
 matter+0.000
 unlike+0.000
 tactic+0.000
 terribly+0.000
mate+0.000
Neg logit contributions
ans-0.000
 waves-0.000
 next-0.000
ites-0.000
irts-0.000
Token to
Token ID284
Feature activation-6.15e-02

Pos logit contributions
ans+0.075
 next+0.074
 waves+0.073
 blow+0.073
 put+0.072
Neg logit contributions
 unlike-0.093
 tactic-0.089
 matter-0.087
rified-0.086
 Where-0.085
Token anything
Token ID1997
Feature activation-1.94e-01

Pos logit contributions
 unlike+0.113
 tactic+0.110
 matter+0.110
 Where+0.105
 Tears+0.105
Neg logit contributions
ans-0.068
 waves-0.064
 next-0.064
 blow-0.061
 goes-0.060
Token anymore
Token ID7471
Feature activation-2.12e-01

Pos logit contributions
 unlike+0.373
 tactic+0.368
rified+0.358
 Tears+0.353
 Where+0.352
Neg logit contributions
 next-0.193
 put-0.187
 blow-0.185
ans-0.185
 bat-0.184
Token.
Token ID13
Feature activation-1.87e-02

Pos logit contributions
 unlike+0.450
 tactic+0.442
rified+0.434
 Tears+0.425
 matter+0.424
Neg logit contributions
 next-0.172
 put-0.165
ans-0.162
 bat-0.162
 blow-0.162
Token
Token ID198
Feature activation+2.75e-02

Pos logit contributions
ifying+0.042
 unlike+0.042
mate+0.041
 matter+0.041
rified+0.041
Neg logit contributions
 next-0.014
 put-0.013
 bat-0.012
 and-0.011
ans-0.011
Token
Token ID198
Feature activation+2.56e-02

Pos logit contributions
ans+0.025
 waves+0.025
 blow+0.024
 next+0.024
 bat+0.024
Neg logit contributions
 unlike-0.054
 tactic-0.054
 Where-0.052
 Tears-0.052
rified-0.051
TokenBut
Token ID1537
Feature activation-6.65e-02

Pos logit contributions
 next+0.025
ans+0.025
 put+0.024
 blow+0.024
 bat+0.024
Neg logit contributions
 unlike-0.050
 tactic-0.049
rified-0.048
 matter-0.047
 Where-0.047