TOP ACTIVATIONS
MAX = 15.985

silly joke . Tam my was very pleased ,
the hive . Tam my went right up to
the toys ." Miss Lee looked at them .
loud noises . Miss Lee says , " Thank
their dreams . Miss Lee nodded . She hugged
hug . Tam my was so proud of
were inside . Tam my started counting . One
mistake . " Miss , we have your photo
sunshine . Tam my was so lucky that
tooth ! Tam my looked around the house
protect it . Miss y thought about this for
that bush !" Miss y rushed over and sure
cry . Tam my felt really bad for
sorry again . Miss Lee smiled . She gave
says proudly . Miss Lee smiles and cl aps
had fun . Miss Lily saw them and smiled
to do . Miss Lee saw what happened .
like you ." Miss Lee says , " You
many things . Miss Lily says , " Do
. She hugs Tom and asks him what happened . Tom

MIN ACTIVATIONS
MIN = -17.834

mom my gave him a brand new pair of shoes to
her - there was a brand new box of cr ay
wall , there was a brand new clock . It was
eyes . Inside was a brand new dress !
a popular toy , a brand new doll .
confidently back home with a brand new haircut . He was
's toys and they looked brand new . Lily was so
then start over with a brand new adventure .
was dry , it looked brand - new again ! L
the freezer until it looked brand new .
" That 's a famous brand , Lily !"
her mom bought her a brand new pair of boots .
Billy . Billy had a brand new car that could speed
he discovered it was a brand new bike !
up as she saw a brand new game on the screen
the sack and found a brand new bright yellow ball !
under the counter ; a brand new doll ! Ab bie
see that it was a brand new puzzle ! He put
joy - it was a brand new doll ! All of
happy ! Inside was a brand new toy that she could

INTERVAL 7.530 - 15.985
CONTAINS 4.243%

did not want Tom to run with her basket . She
with a dent . She says , " What happened ?
He was so happy that Sara was safe .
One day , Tim my went to the park to draw
long bl ouse back and thanked her friend for helping her

INTERVAL -0.924 - 7.530
CONTAINS 52.616%

have to ask before you use them . Now let 's
come in and play with you ?" Rex w
colourful , with many buildings painted in bright colours .
, Lily and Rose went to pick some berries in the
Suddenly , she heard laughter coming from a nearby park .

INTERVAL -9.379 - -0.924
CONTAINS 41.060%

He had had such a wonderful time . He was proud
keep that huge fork as a reminder of their day of
a boy . The boy did not like the bed .
Lily said . " This is the best birthday ever !"
happy , and it hopped all the way home . <|endoftext|>

INTERVAL -17.834 - -9.379
CONTAINS 2.081%

the sea . He was so excited , he cl apped
outside , and Billy was so excited . He wanted to
She felt so helpless and disappointed in herself
" That was so much fun , thanks for
can go to grandma ?" Her mom my
Token silly
Token ID14397
Feature activation-1.78e+00

Pos logit contributions
 brand+0.225
 brightly+0.173
 lot+0.156
 truly+0.154
 valuable+0.140
Neg logit contributions
ases-0.459
per-0.408
pered-0.397
ena-0.394
ws-0.390
Token joke
Token ID9707
Feature activation-1.46e+00

Pos logit contributions
 brand+0.047
 brightly+0.040
 lot+0.040
 truly+0.036
 wonderful+0.036
Neg logit contributions
ases-0.066
per-0.055
ws-0.055
wed-0.055
pered-0.055
Token.
Token ID13
Feature activation+1.88e+00

Pos logit contributions
 brand+0.046
 brightly+0.041
 lot+0.039
 truly+0.038
 very+0.037
Neg logit contributions
ases-0.043
ws-0.036
per-0.036
pered-0.036
wed-0.036
Token
Token ID198
Feature activation-2.76e+00

Pos logit contributions
ases+0.069
wed+0.060
ws+0.060
per+0.059
pered+0.057
Neg logit contributions
 brand-0.047
 brightly-0.041
 lot-0.038
 truly-0.037
-0.035
Token
Token ID198
Feature activation-2.88e+00

Pos logit contributions
 brand+0.051
 lot+0.041
 brightly+0.041
 truly+0.038
 much+0.035
Neg logit contributions
ases-0.120
per-0.107
ws-0.104
wed-0.104
pered-0.103
TokenTam
Token ID42061
Feature activation+1.54e+01

Pos logit contributions
 brand+0.057
 brightly+0.049
 lot+0.047
 truly+0.044
 valuable+0.042
Neg logit contributions
ases-0.120
ws-0.105
per-0.104
wed-0.103
pered-0.102
Tokenmy
Token ID1820
Feature activation+4.58e+00

Pos logit contributions
ases+0.533
pered+0.471
per+0.449
ws+0.434
ucks+0.419
Neg logit contributions
 brand-0.392
 truly-0.349
 brightly-0.332
 wonderful-0.325
 very-0.319
Token was
Token ID373
Feature activation-4.32e+00

Pos logit contributions
ases+0.189
pered+0.156
ws+0.154
per+0.149
race+0.147
Neg logit contributions
 brand-0.101
 brightly-0.084
 truly-0.084
 very-0.082
 lot-0.079
Token very
Token ID845
Feature activation-1.18e+01

Pos logit contributions
 brand+0.073
 truly+0.051
 brightly+0.051
 lot+0.047
 very+0.047
Neg logit contributions
ases-0.201
per-0.177
ws-0.176
wed-0.172
pered-0.168
Token pleased
Token ID10607
Feature activation-3.49e+00

Pos logit contributions
 brand+0.258
 brightly+0.214
 lot+0.201
 truly+0.199
 valuable+0.190
Neg logit contributions
ases-0.477
per-0.420
ws-0.411
wed-0.405
pered-0.401
Token,
Token ID11
Feature activation-3.31e+00

Pos logit contributions
 brand+0.114
 brightly+0.100
 lot+0.093
 truly+0.092
 valuable+0.089
Neg logit contributions
ases-0.107
per-0.088
ws-0.088
wed-0.086
pered-0.085
Token the
Token ID262
Feature activation+1.90e-01

Pos logit contributions
ases+0.122
ws+0.103
per+0.102
wed+0.098
pered+0.097
Neg logit contributions
 brand-0.132
 brightly-0.118
 lot-0.114
 truly-0.108
 very-0.107
Token hive
Token ID35881
Feature activation-4.83e+00

Pos logit contributions
ases+0.006
per+0.005
pered+0.005
ws+0.004
ena+0.004
Neg logit contributions
 brand-0.006
 brightly-0.005
 truly-0.004
 valuable-0.004
-0.004
Token.
Token ID13
Feature activation+1.61e+00

Pos logit contributions
 brand+0.164
 brightly+0.144
 truly+0.138
 lot+0.135
 very+0.131
Neg logit contributions
ases-0.143
ws-0.114
per-0.111
wed-0.111
ena-0.110
Token
Token ID198
Feature activation-3.24e+00

Pos logit contributions
ases+0.049
per+0.040
pered+0.039
ena+0.039
ws+0.038
Neg logit contributions
 brand-0.052
 brightly-0.045
 lot-0.041
 truly-0.041
 valuable-0.040
Token
Token ID198
Feature activation-3.30e+00

Pos logit contributions
 brand+0.080
 lot+0.064
 brightly+0.063
 truly+0.061
 valuable+0.060
Neg logit contributions
ases-0.123
per-0.109
ws-0.103
wed-0.102
pered-0.100
TokenTam
Token ID42061
Feature activation+1.52e+01

Pos logit contributions
 brand+0.091
 brightly+0.076
 lot+0.073
 truly+0.072
 valuable+0.069
Neg logit contributions
ases-0.115
per-0.099
ws-0.095
pered-0.094
ena-0.093
Tokenmy
Token ID1820
Feature activation+4.07e+00

Pos logit contributions
ases+0.499
pered+0.411
per+0.407
ws+0.394
ucks+0.358
Neg logit contributions
 brand-0.446
 truly-0.382
 brightly-0.365
 wonderful-0.362
 lot-0.354
Token went
Token ID1816
Feature activation+7.90e+00

Pos logit contributions
ases+0.167
ws+0.140
pered+0.136
per+0.135
wed+0.131
Neg logit contributions
 brand-0.093
 truly-0.075
 brightly-0.074
 lot-0.070
 very-0.070
Token right
Token ID826
Feature activation+4.89e+00

Pos logit contributions
ases+0.287
ws+0.236
wed+0.235
ena+0.231
per+0.227
Neg logit contributions
 brand-0.223
 very-0.188
 lot-0.179
 truly-0.172
 brightly-0.169
Token up
Token ID510
Feature activation+2.12e+00

Pos logit contributions
ases+0.146
per+0.119
ws+0.114
ena+0.113
pered+0.112
Neg logit contributions
 brand-0.162
 brightly-0.143
 lot-0.133
 truly-0.131
 valuable-0.127
Token to
Token ID284
Feature activation+6.40e+00

Pos logit contributions
ases+0.073
per+0.061
ws+0.061
wed+0.060
ena+0.059
Neg logit contributions
 brand-0.060
 brightly-0.052
 lot-0.049
 truly-0.048
 valuable-0.045
Token the
Token ID262
Feature activation+9.49e-01

Pos logit contributions
ases+0.002
per+0.002
ena+0.002
pered+0.002
ws+0.002
Neg logit contributions
 brand-0.002
 brightly-0.002
 lot-0.001
 truly-0.001
 valuable-0.001
Token toys
Token ID14958
Feature activation-1.48e+00

Pos logit contributions
ases+0.034
per+0.029
pered+0.028
ena+0.028
ws+0.028
Neg logit contributions
 brand-0.025
 brightly-0.021
 lot-0.019
 truly-0.019
 valuable-0.018
Token."
Token ID526
Feature activation+3.48e+00

Pos logit contributions
 brand+0.049
 brightly+0.042
 lot+0.041
 truly+0.040
 valuable+0.039
Neg logit contributions
ases-0.043
per-0.036
pered-0.035
ena-0.034
ws-0.034
Token
Token ID198
Feature activation-2.07e+00

Pos logit contributions
ases+0.157
wed+0.144
ws+0.143
per+0.139
ps+0.136
Neg logit contributions
 brand-0.055
 brightly-0.048
 lot-0.045
 much-0.039
 very-0.039
Token
Token ID198
Feature activation-2.25e+00

Pos logit contributions
 brand+0.030
 lot+0.024
 brightly+0.023
 much+0.021
 truly+0.019
Neg logit contributions
ases-0.096
wed-0.087
per-0.087
ws-0.086
pered-0.084
TokenMiss
Token ID17140
Feature activation+1.50e+01

Pos logit contributions
 brand+0.051
 brightly+0.047
 lot+0.046
 sudden+0.041
 truly+0.041
Neg logit contributions
ases-0.084
ws-0.074
wed-0.073
pered-0.072
per-0.071
Token Lee
Token ID5741
Feature activation+6.98e+00

Pos logit contributions
ases+0.384
ws+0.325
pered+0.320
per+0.312
wed+0.307
Neg logit contributions
 brand-0.515
 brightly-0.462
 lot-0.430
 truly-0.420
 most-0.417
Token looked
Token ID3114
Feature activation-5.49e+00

Pos logit contributions
ases+0.240
pered+0.204
race+0.194
ws+0.192
per+0.190
Neg logit contributions
 brand-0.195
 brightly-0.167
 truly-0.157
 lot-0.157
 very-0.151
Token at
Token ID379
Feature activation+6.09e+00

Pos logit contributions
 brand+0.207
 brightly+0.182
 lot+0.174
 truly+0.171
 valuable+0.168
Neg logit contributions
ases-0.139
per-0.112
pered-0.105
ena-0.105
ws-0.103
Token them
Token ID606
Feature activation+6.56e-01

Pos logit contributions
ases+0.172
per+0.144
pered+0.141
ws+0.139
wed+0.138
Neg logit contributions
 brand-0.204
 brightly-0.182
 lot-0.182
 truly-0.165
 wonderful-0.165
Token.
Token ID13
Feature activation+2.37e+00

Pos logit contributions
ases+0.019
ws+0.015
per+0.015
ena+0.015
wed+0.015
Neg logit contributions
 brand-0.022
 brightly-0.020
 lot-0.019
 truly-0.018
 very-0.018
Token loud
Token ID7812
Feature activation+1.98e+00

Pos logit contributions
ases+0.071
per+0.063
pered+0.061
wed+0.059
ws+0.058
Neg logit contributions
 brand-0.064
 brightly-0.054
 lot-0.054
 wonderful-0.050
 fun-0.049
Token noises
Token ID26782
Feature activation-4.38e+00

Pos logit contributions
ases+0.075
per+0.067
pered+0.065
wed+0.064
ws+0.063
Neg logit contributions
 brand-0.048
 brightly-0.039
 lot-0.037
 truly-0.033
 wonderful-0.033
Token.
Token ID13
Feature activation+2.65e+00

Pos logit contributions
 brand+0.140
 brightly+0.125
 lot+0.119
 very+0.116
 truly+0.115
Neg logit contributions
ases-0.130
pered-0.107
ws-0.105
per-0.104
wed-0.104
Token
Token ID198
Feature activation-2.01e+00

Pos logit contributions
ases+0.082
ena+0.067
per+0.067
pered+0.065
ws+0.063
Neg logit contributions
 brand-0.085
 brightly-0.070
 truly-0.068
 lot-0.066
 fantastic-0.065
Token
Token ID198
Feature activation-2.14e+00

Pos logit contributions
 brand+0.044
 brightly+0.035
 lot+0.034
 truly+0.032
 valuable+0.031
Neg logit contributions
ases-0.081
per-0.072
pered-0.069
wed-0.069
ws-0.069
TokenMiss
Token ID17140
Feature activation+1.50e+01

Pos logit contributions
 brand+0.052
 brightly+0.045
 lot+0.045
 valuable+0.041
 truly+0.040
Neg logit contributions
ases-0.078
per-0.069
ws-0.067
wed-0.067
pered-0.067
Token Lee
Token ID5741
Feature activation+6.83e+00

Pos logit contributions
ases+0.360
ws+0.318
pered+0.308
per+0.304
wed+0.298
Neg logit contributions
 brand-0.499
 brightly-0.445
 lot-0.430
 most-0.415
 truly-0.409
Token says
Token ID1139
Feature activation+6.98e+00

Pos logit contributions
ases+0.221
pered+0.200
race+0.190
per+0.186
wed+0.182
Neg logit contributions
 brand-0.190
 lot-0.161
 brightly-0.156
 very-0.153
 wonderful-0.152
Token,
Token ID11
Feature activation-2.21e+00

Pos logit contributions
ases+0.231
per+0.197
pered+0.190
ena+0.189
ws+0.184
Neg logit contributions
 brand-0.205
 brightly-0.171
 lot-0.162
 truly-0.162
 valuable-0.154
Token "
Token ID366
Feature activation-1.65e+00

Pos logit contributions
 brand+0.060
 brightly+0.051
 lot+0.047
 truly+0.047
 valuable+0.045
Neg logit contributions
ases-0.076
per-0.066
pered-0.064
ena-0.063
ws-0.061
TokenThank
Token ID10449
Feature activation+4.56e+00

Pos logit contributions
 brand+0.048
 brightly+0.037
 lot+0.036
 truly+0.035
 fantastic+0.035
Neg logit contributions
ases-0.057
per-0.049
ena-0.048
pered-0.045
race-0.045
Token their
Token ID511
Feature activation+2.88e+00

Pos logit contributions
ases+0.172
per+0.148
pered+0.144
ws+0.142
wed+0.141
Neg logit contributions
 brand-0.120
 brightly-0.097
 lot-0.094
 truly-0.091
 valuable-0.090
Token dreams
Token ID10625
Feature activation-6.46e+00

Pos logit contributions
ases+0.085
per+0.070
ena+0.068
pered+0.066
ws+0.066
Neg logit contributions
 brand-0.094
 brightly-0.082
 lot-0.077
 truly-0.077
 valuable-0.073
Token.
Token ID13
Feature activation+2.65e+00

Pos logit contributions
 brand+0.224
 brightly+0.196
 truly+0.190
 lot+0.188
 valuable+0.182
Neg logit contributions
ases-0.180
per-0.144
pered-0.142
ws-0.139
wed-0.134
Token
Token ID198
Feature activation-2.20e+00

Pos logit contributions
ases+0.091
per+0.078
ws+0.075
pered+0.074
wed+0.074
Neg logit contributions
 brand-0.075
 brightly-0.064
 lot-0.061
 truly-0.058
 valuable-0.057
Token
Token ID198
Feature activation-2.31e+00

Pos logit contributions
 brand+0.043
 lot+0.033
 brightly+0.033
 truly+0.030
 valuable+0.029
Neg logit contributions
ases-0.094
per-0.084
pered-0.081
wed-0.081
ws-0.081
TokenMiss
Token ID17140
Feature activation+1.50e+01

Pos logit contributions
 brand+0.064
 brightly+0.057
 lot+0.055
 truly+0.051
 valuable+0.050
Neg logit contributions
ases-0.078
per-0.066
ws-0.065
pered-0.065
wed-0.064
Token Lee
Token ID5741
Feature activation+7.04e+00

Pos logit contributions
ases+0.384
ws+0.317
pered+0.316
per+0.308
wed+0.296
Neg logit contributions
 brand-0.520
 brightly-0.465
 lot-0.435
 truly-0.431
 most-0.421
Token nodded
Token ID14464
Feature activation+3.12e+00

Pos logit contributions
ases+0.256
pered+0.214
ws+0.208
per+0.203
race+0.202
Neg logit contributions
 brand-0.193
 brightly-0.163
 truly-0.154
 lot-0.149
 wonderful-0.143
Token.
Token ID13
Feature activation+2.41e+00

Pos logit contributions
ases+0.099
ws+0.080
per+0.078
wed+0.077
race+0.077
Neg logit contributions
 brand-0.100
 brightly-0.093
 very-0.084
 lot-0.084
 truly-0.080
Token She
Token ID1375
Feature activation+1.21e+00

Pos logit contributions
ases+0.080
per+0.068
pered+0.065
ena+0.065
ws+0.065
Neg logit contributions
 brand-0.071
 brightly-0.060
 lot-0.057
 truly-0.056
 valuable-0.053
Token hugged
Token ID41130
Feature activation+8.48e+00

Pos logit contributions
ases+0.044
per+0.037
pered+0.037
ws+0.036
wed+0.035
Neg logit contributions
 brand-0.033
 brightly-0.027
 truly-0.026
 lot-0.026
 valuable-0.025
Token hug
Token ID16225
Feature activation+5.74e+00

Pos logit contributions
 brand+0.056
 brightly+0.048
 lot+0.046
 truly+0.045
 valuable+0.043
Neg logit contributions
ases-0.058
per-0.049
pered-0.047
ena-0.046
ws-0.046
Token.
Token ID13
Feature activation+1.45e+00

Pos logit contributions
ases+0.172
ws+0.136
per+0.133
pered+0.132
wed+0.129
Neg logit contributions
 brand-0.197
 brightly-0.175
 lot-0.163
 truly-0.163
 very-0.156
Token
Token ID220
Feature activation-1.04e+01

Pos logit contributions
ases+0.050
per+0.043
pered+0.041
ena+0.041
ws+0.041
Neg logit contributions
 brand-0.041
 brightly-0.034
 lot-0.032
 truly-0.032
 valuable-0.030
Token
Token ID198
Feature activation-3.51e+00

Pos logit contributions
 brand+0.189
 truly+0.130
 lot+0.129
+0.128
 brightly+0.127
Neg logit contributions
ases-0.449
per-0.402
wed-0.394
ws-0.391
pered-0.377
Token
Token ID198
Feature activation-3.57e+00

Pos logit contributions
 brand+0.070
 lot+0.055
 brightly+0.051
 truly+0.050
 valuable+0.049
Neg logit contributions
ases-0.148
per-0.133
ws-0.127
wed-0.127
pered-0.124
TokenTam
Token ID42061
Feature activation+1.50e+01

Pos logit contributions
 brand+0.090
 brightly+0.074
 lot+0.072
 truly+0.069
 valuable+0.067
Neg logit contributions
ases-0.133
per-0.116
ws-0.111
pered-0.110
wed-0.109
Tokenmy
Token ID1820
Feature activation+3.61e+00

Pos logit contributions
ases+0.456
pered+0.377
per+0.366
ws+0.347
race+0.321
Neg logit contributions
 brand-0.495
 truly-0.422
 brightly-0.412
 lot-0.409
 wonderful-0.405
Token was
Token ID373
Feature activation-4.05e+00

Pos logit contributions
ases+0.141
per+0.118
ws+0.117
pered+0.117
ena+0.111
Neg logit contributions
 brand-0.091
 brightly-0.072
 truly-0.071
 lot-0.070
 valuable-0.065
Token so
Token ID523
Feature activation-1.24e+01

Pos logit contributions
 brand+0.070
 brightly+0.053
 lot+0.048
 truly+0.044
 valuable+0.042
Neg logit contributions
ases-0.181
per-0.162
pered-0.158
ena-0.157
ws-0.155
Token proud
Token ID6613
Feature activation-2.85e+00

Pos logit contributions
 brand+0.339
 brightly+0.277
 lot+0.267
+0.257
 sudden+0.242
Neg logit contributions
ases-0.403
ena-0.341
per-0.338
pered-0.333
e-0.329
Token of
Token ID286
Feature activation+5.74e+00

Pos logit contributions
 brand+0.090
 brightly+0.078
 truly+0.073
 lot+0.073
 valuable+0.071
Neg logit contributions
ases-0.090
per-0.076
ws-0.075
pered-0.074
wed-0.072
Token were
Token ID547
Feature activation-4.83e+00

Pos logit contributions
ases+0.003
per+0.003
pered+0.003
ena+0.003
ws+0.003
Neg logit contributions
 brand-0.003
 brightly-0.002
 lot-0.002
 truly-0.002
 valuable-0.002
Token inside
Token ID2641
Feature activation+3.72e+00

Pos logit contributions
 brand+0.145
 truly+0.125
 brightly+0.124
 very+0.123
 lot+0.120
Neg logit contributions
ases-0.156
ws-0.125
pered-0.124
per-0.124
ucks-0.123
Token.
Token ID13
Feature activation+1.60e+00

Pos logit contributions
ases+0.104
per+0.085
pered+0.083
ws+0.082
ena+0.081
Neg logit contributions
 brand-0.129
 brightly-0.115
 lot-0.110
 truly-0.107
 wonderful-0.103
Token
Token ID198
Feature activation-3.17e+00

Pos logit contributions
ases+0.056
per+0.047
ena+0.047
pered+0.047
ws+0.045
Neg logit contributions
 brand-0.044
 brightly-0.036
 truly-0.034
 lot-0.033
 valuable-0.032
Token
Token ID198
Feature activation-3.26e+00

Pos logit contributions
 brand+0.076
 brightly+0.062
 lot+0.059
 truly+0.057
 valuable+0.054
Neg logit contributions
ases-0.121
per-0.106
pered-0.102
ws-0.101
ena-0.101
TokenTam
Token ID42061
Feature activation+1.49e+01

Pos logit contributions
 brand+0.094
 brightly+0.079
 lot+0.076
 truly+0.074
 valuable+0.071
Neg logit contributions
ases-0.110
per-0.094
pered-0.090
ws-0.089
ena-0.089
Tokenmy
Token ID1820
Feature activation+4.14e+00

Pos logit contributions
ases+0.504
pered+0.427
per+0.419
ws+0.391
icks+0.372
Neg logit contributions
 brand-0.406
 truly-0.338
 wonderful-0.337
 lot-0.330
 brightly-0.328
Token started
Token ID2067
Feature activation+1.23e+00

Pos logit contributions
ases+0.161
ws+0.130
pered+0.126
ags+0.122
per+0.120
Neg logit contributions
 brand-0.104
 very-0.090
 brightly-0.089
 truly-0.087
 lot-0.084
Token counting
Token ID14143
Feature activation+3.65e+00

Pos logit contributions
ases+0.044
per+0.039
ws+0.037
pered+0.037
wed+0.036
Neg logit contributions
 brand-0.032
 brightly-0.027
 lot-0.026
 truly-0.025
 very-0.024
Token.
Token ID13
Feature activation+1.25e+00

Pos logit contributions
ases+0.102
per+0.085
pered+0.080
ena+0.080
ws+0.078
Neg logit contributions
 brand-0.126
 brightly-0.105
 truly-0.103
 lot-0.101
 valuable-0.100
Token One
Token ID1881
Feature activation-2.61e+00

Pos logit contributions
ases+0.045
per+0.039
ws+0.038
wed+0.038
ena+0.037
Neg logit contributions
 brand-0.032
 brightly-0.028
 lot-0.026
 valuable-0.024
 truly-0.024
Token mistake
Token ID7457
Feature activation+1.45e+00

Pos logit contributions
 brand+0.166
+0.149
 brightly+0.148
 brilliantly+0.138
 truly+0.137
Neg logit contributions
ases-0.095
ena-0.077
per-0.075
 aid-0.073
ws-0.072
Token.
Token ID13
Feature activation+2.74e+00

Pos logit contributions
ases+0.046
ws+0.039
wed+0.038
pered+0.036
aled+0.036
Neg logit contributions
 brand-0.044
 brightly-0.039
 very-0.037
 valuable-0.036
 lot-0.036
Token
Token ID198
Feature activation-2.30e+00

Pos logit contributions
ases+0.093
per+0.079
pered+0.076
ena+0.076
ws+0.075
Neg logit contributions
 brand-0.079
 brightly-0.066
 lot-0.062
 truly-0.062
 valuable-0.059
Token
Token ID198
Feature activation-2.36e+00

Pos logit contributions
 brand+0.052
 brightly+0.041
 lot+0.039
 truly+0.038
 valuable+0.036
Neg logit contributions
ases-0.092
per-0.081
pered-0.078
ws-0.077
wed-0.077
Token"
Token ID1
Feature activation+2.09e+00

Pos logit contributions
 brand+0.069
 brightly+0.060
 lot+0.057
 truly+0.055
 valuable+0.054
Neg logit contributions
ases-0.078
per-0.066
ws-0.064
pered-0.064
wed-0.064
TokenMiss
Token ID17140
Feature activation+1.47e+01

Pos logit contributions
ases+0.069
ena+0.060
per+0.059
wed+0.055
pered+0.054
Neg logit contributions
 brand-0.064
 lot-0.047
 brightly-0.046
 fantastic-0.045
 truly-0.045
Token,
Token ID11
Feature activation-1.84e+00

Pos logit contributions
ases+0.386
pered+0.333
ws+0.332
wed+0.326
per+0.318
Neg logit contributions
 brand-0.485
 brightly-0.434
 lot-0.411
 truly-0.406
 most-0.396
Token we
Token ID356
Feature activation+4.36e+00

Pos logit contributions
 brand+0.053
 brightly+0.044
 lot+0.042
 truly+0.041
 valuable+0.040
Neg logit contributions
ases-0.063
per-0.053
pered-0.051
ws-0.051
ena-0.051
Token have
Token ID423
Feature activation+1.92e+00

Pos logit contributions
ases+0.152
per+0.133
pered+0.130
ena+0.125
ws+0.123
Neg logit contributions
 brand-0.117
 lot-0.099
 brightly-0.098
 truly-0.097
 wonderful-0.091
Token your
Token ID534
Feature activation+3.54e+00

Pos logit contributions
ases+0.063
per+0.054
pered+0.051
ena+0.051
ws+0.051
Neg logit contributions
 brand-0.057
 brightly-0.048
 lot-0.046
 truly-0.045
 valuable-0.043
Token photo
Token ID4590
Feature activation-4.60e+00

Pos logit contributions
ases+0.114
per+0.093
ena+0.092
pered+0.089
ws+0.089
Neg logit contributions
 brand-0.105
 brightly-0.092
 truly-0.084
 brilliantly-0.082
 lot-0.081
Token sunshine
Token ID34488
Feature activation-6.82e+00

Pos logit contributions
 brand+0.004
 brightly+0.003
 truly+0.003
 lot+0.003
 valuable+0.003
Neg logit contributions
ases-0.007
per-0.006
pered-0.006
ena-0.006
ws-0.006
Token.
Token ID13
Feature activation+9.27e-01

Pos logit contributions
 brand+0.246
 brightly+0.216
 lot+0.207
 truly+0.204
 valuable+0.198
Neg logit contributions
ases-0.180
per-0.146
pered-0.138
ena-0.137
ws-0.136
Token
Token ID220
Feature activation-1.08e+01

Pos logit contributions
ases+0.033
per+0.028
pered+0.028
ena+0.028
ws+0.027
Neg logit contributions
 brand-0.025
 brightly-0.020
 lot-0.019
 truly-0.019
 valuable-0.018
Token
Token ID198
Feature activation-4.06e+00

Pos logit contributions
 brand+0.201
 lot+0.143
 brightly+0.143
 truly+0.141
+0.138
Neg logit contributions
ases-0.465
per-0.418
ws-0.406
wed-0.405
pered-0.390
Token
Token ID198
Feature activation-4.13e+00

Pos logit contributions
 brand+0.086
 lot+0.067
 brightly+0.066
 truly+0.064
 valuable+0.060
Neg logit contributions
ases-0.166
per-0.148
ws-0.142
wed-0.141
pered-0.139
TokenTam
Token ID42061
Feature activation+1.47e+01

Pos logit contributions
 brand+0.108
 brightly+0.089
 lot+0.085
 truly+0.084
 valuable+0.080
Neg logit contributions
ases-0.150
per-0.131
ws-0.125
pered-0.124
ena-0.123
Tokenmy
Token ID1820
Feature activation+3.26e+00

Pos logit contributions
ases+0.443
per+0.371
ws+0.359
pered+0.358
ucks+0.310
Neg logit contributions
 brand-0.473
 truly-0.406
 brightly-0.398
 lot-0.390
 wonderful-0.388
Token was
Token ID373
Feature activation-4.50e+00

Pos logit contributions
ases+0.138
ws+0.117
per+0.113
pered+0.113
wed+0.108
Neg logit contributions
 brand-0.072
 truly-0.058
 brightly-0.057
 lot-0.055
 very-0.054
Token so
Token ID523
Feature activation-1.28e+01

Pos logit contributions
 brand+0.078
 brightly+0.055
 truly+0.053
 lot+0.050
 wonderful+0.045
Neg logit contributions
ases-0.210
per-0.185
ws-0.182
wed-0.178
pered-0.177
Token lucky
Token ID9670
Feature activation-4.22e+00

Pos logit contributions
 brand+0.305
 brightly+0.244
 lot+0.227
 truly+0.217
 valuable+0.210
Neg logit contributions
ases-0.493
per-0.428
pered-0.420
ena-0.416
ws-0.405
Token that
Token ID326
Feature activation-2.43e+00

Pos logit contributions
 brand+0.125
 brightly+0.105
 lot+0.100
 truly+0.098
 valuable+0.095
Neg logit contributions
ases-0.139
per-0.118
pered-0.114
ena-0.113
ws-0.112
Token tooth
Token ID16162
Feature activation-4.67e+00

Pos logit contributions
ases+0.125
per+0.106
ena+0.104
pered+0.103
wed+0.100
Neg logit contributions
 brand-0.099
 brightly-0.085
 lot-0.079
 truly-0.079
 valuable-0.076
Token!
Token ID0
Feature activation+1.10e+00

Pos logit contributions
 brand+0.153
 brightly+0.131
 truly+0.123
 lot+0.123
 valuable+0.118
Neg logit contributions
ases-0.145
per-0.116
ws-0.116
pered-0.115
ena-0.112
Token
Token ID220
Feature activation-1.07e+01

Pos logit contributions
ases+0.033
per+0.028
ena+0.027
pered+0.027
ws+0.026
Neg logit contributions
 brand-0.035
 brightly-0.030
 lot-0.028
 truly-0.028
 valuable-0.027
Token
Token ID198
Feature activation-3.53e+00

Pos logit contributions
 brand+0.225
 brightly+0.174
 lot+0.168
 truly+0.162
 valuable+0.151
Neg logit contributions
ases-0.442
per-0.393
ws-0.379
wed-0.377
pered-0.375
Token
Token ID198
Feature activation-3.59e+00

Pos logit contributions
 brand+0.080
 brightly+0.064
 lot+0.062
 truly+0.059
 valuable+0.056
Neg logit contributions
ases-0.140
per-0.123
ws-0.118
pered-0.118
wed-0.117
TokenTam
Token ID42061
Feature activation+1.47e+01

Pos logit contributions
 brand+0.122
 brightly+0.107
 lot+0.104
 truly+0.101
 valuable+0.099
Neg logit contributions
ases-0.101
per-0.084
ws-0.080
pered-0.079
wed-0.078
Tokenmy
Token ID1820
Feature activation+3.56e+00

Pos logit contributions
ases+0.486
pered+0.408
per+0.405
ws+0.396
ucks+0.350
Neg logit contributions
 brand-0.422
 truly-0.349
 brightly-0.336
 lot-0.336
 wonderful-0.332
Token looked
Token ID3114
Feature activation-6.01e+00

Pos logit contributions
ases+0.134
ws+0.110
pered+0.104
per+0.103
wed+0.101
Neg logit contributions
 brand-0.096
 brightly-0.081
 truly-0.077
 lot-0.076
 very-0.074
Token around
Token ID1088
Feature activation+1.90e+00

Pos logit contributions
 brand+0.219
 brightly+0.192
 lot+0.185
 truly+0.184
+0.179
Neg logit contributions
ases-0.148
per-0.121
pered-0.118
ws-0.112
wed-0.110
Token the
Token ID262
Feature activation-1.67e-01

Pos logit contributions
ases+0.058
per+0.047
ws+0.046
wed+0.045
ena+0.045
Neg logit contributions
 brand-0.063
 brightly-0.056
 lot-0.053
 truly-0.051
 wonderful-0.050
Token house
Token ID2156
Feature activation-2.85e+00

Pos logit contributions
 brand+0.005
 brightly+0.004
 lot+0.004
 truly+0.004
 valuable+0.004
Neg logit contributions
ases-0.005
per-0.004
ws-0.004
pered-0.004
ena-0.004
Token protect
Token ID1805
Feature activation+5.73e+00

Pos logit contributions
ases+0.179
per+0.147
pered+0.145
ena+0.142
ws+0.138
Neg logit contributions
 brand-0.212
 brightly-0.186
 lot-0.183
 truly-0.182
 valuable-0.171
Token it
Token ID340
Feature activation-7.32e+00

Pos logit contributions
ases+0.190
per+0.161
pered+0.157
ws+0.155
wed+0.154
Neg logit contributions
 brand-0.162
 brightly-0.145
 lot-0.141
 truly-0.133
 valuable-0.130
Token.
Token ID13
Feature activation+1.68e+00

Pos logit contributions
 brand+0.256
 brightly+0.226
 lot+0.217
 truly+0.216
 valuable+0.211
Neg logit contributions
ases-0.204
ena-0.162
per-0.161
pered-0.159
ws-0.155
Token
Token ID198
Feature activation-3.22e+00

Pos logit contributions
ases+0.062
per+0.053
pered+0.051
ena+0.051
ws+0.051
Neg logit contributions
 brand-0.043
 brightly-0.036
 lot-0.033
 truly-0.033
 valuable-0.031
Token
Token ID198
Feature activation-3.26e+00

Pos logit contributions
 brand+0.071
 brightly+0.056
 lot+0.055
 truly+0.053
 valuable+0.050
Neg logit contributions
ases-0.130
per-0.114
ws-0.111
pered-0.110
wed-0.109
TokenMiss
Token ID17140
Feature activation+1.46e+01

Pos logit contributions
 brand+0.102
 brightly+0.088
 lot+0.085
 truly+0.083
 valuable+0.081
Neg logit contributions
ases-0.102
per-0.085
ws-0.082
pered-0.082
wed-0.080
Tokeny
Token ID88
Feature activation-3.83e-01

Pos logit contributions
ases+0.483
ws+0.421
pered+0.402
per+0.396
wed+0.383
Neg logit contributions
 brand-0.413
 brightly-0.362
 truly-0.327
 lot-0.324
 most-0.311
Token thought
Token ID1807
Feature activation-1.57e+00

Pos logit contributions
 brand+0.010
 brightly+0.008
 truly+0.008
 lot+0.008
 valuable+0.007
Neg logit contributions
ases-0.014
ws-0.012
pered-0.012
per-0.011
race-0.011
Token about
Token ID546
Feature activation+1.06e+00

Pos logit contributions
 brand+0.044
 brightly+0.036
 lot+0.035
 truly+0.035
 valuable+0.034
Neg logit contributions
ases-0.057
per-0.048
pered-0.047
ws-0.046
wed-0.045
Token this
Token ID428
Feature activation-4.17e+00

Pos logit contributions
ases+0.036
per+0.031
ena+0.030
pered+0.029
ws+0.029
Neg logit contributions
 brand-0.030
 brightly-0.025
 lot-0.024
 truly-0.023
 valuable-0.022
Token for
Token ID329
Feature activation+5.63e+00

Pos logit contributions
 brand+0.127
 brightly+0.108
 lot+0.101
 truly+0.100
+0.095
Neg logit contributions
ases-0.130
per-0.111
ena-0.107
pered-0.106
ws-0.106
Token that
Token ID326
Feature activation-2.03e+00

Pos logit contributions
ases+0.272
per+0.230
pered+0.224
ws+0.223
ena+0.222
Neg logit contributions
 brand-0.244
 brightly-0.210
 lot-0.203
 truly-0.194
 valuable-0.185
Token bush
Token ID24484
Feature activation-6.10e+00

Pos logit contributions
 brand+0.061
 brightly+0.051
 lot+0.048
 truly+0.047
 valuable+0.046
Neg logit contributions
ases-0.065
per-0.056
ena-0.052
ws-0.052
pered-0.052
Token!"
Token ID2474
Feature activation+2.30e+00

Pos logit contributions
 brand+0.163
 brightly+0.138
 lot+0.133
 truly+0.131
 valuable+0.121
Neg logit contributions
ases-0.215
per-0.183
pered-0.182
ws-0.179
wed-0.177
Token
Token ID198
Feature activation-3.43e+00

Pos logit contributions
ases+0.081
per+0.070
ena+0.068
pered+0.067
ws+0.065
Neg logit contributions
 brand-0.061
 brightly-0.051
 truly-0.048
 lot-0.046
 valuable-0.046
Token
Token ID198
Feature activation-3.48e+00

Pos logit contributions
 brand+0.077
 brightly+0.061
 lot+0.059
 truly+0.055
 valuable+0.053
Neg logit contributions
ases-0.139
per-0.123
ws-0.119
pered-0.118
wed-0.118
TokenMiss
Token ID17140
Feature activation+1.46e+01

Pos logit contributions
 brand+0.120
 brightly+0.105
 lot+0.102
 truly+0.099
 valuable+0.098
Neg logit contributions
ases-0.099
per-0.081
ws-0.079
pered-0.077
wed-0.076
Tokeny
Token ID88
Feature activation-2.14e-01

Pos logit contributions
ases+0.465
per+0.401
ws+0.401
pered+0.392
wed+0.375
Neg logit contributions
 brand-0.424
 brightly-0.369
 lot-0.343
 truly-0.326
 most-0.319
Token rushed
Token ID13999
Feature activation+1.17e+01

Pos logit contributions
 brand+0.006
 brightly+0.005
 truly+0.005
 lot+0.005
 valuable+0.005
Neg logit contributions
ases-0.008
ws-0.006
per-0.006
pered-0.006
wed-0.006
Token over
Token ID625
Feature activation+4.51e+00

Pos logit contributions
ases+0.452
ws+0.383
per+0.381
wed+0.362
ena+0.358
Neg logit contributions
 brand-0.288
 brightly-0.243
 very-0.239
 lot-0.237
 much-0.203
Token and
Token ID290
Feature activation-1.04e+00

Pos logit contributions
ases+0.180
ws+0.157
per+0.153
wed+0.153
ena+0.146
Neg logit contributions
 brand-0.104
 brightly-0.093
 lot-0.087
 very-0.081
 wonderful-0.076
Token sure
Token ID1654
Feature activation-1.22e+00

Pos logit contributions
 brand+0.035
 brightly+0.030
 lot+0.028
 truly+0.028
 valuable+0.027
Neg logit contributions
ases-0.031
per-0.025
ws-0.024
pered-0.024
ena-0.024
Token cry
Token ID3960
Feature activation+2.39e+00

Pos logit contributions
ases+0.185
per+0.154
ena+0.145
pered+0.143
ws+0.141
Neg logit contributions
 brand-0.198
 lot-0.169
 truly-0.163
 brightly-0.162
 wonderful-0.160
Token.
Token ID13
Feature activation+1.17e+00

Pos logit contributions
ases+0.062
per+0.049
ws+0.047
ena+0.046
pered+0.046
Neg logit contributions
 brand-0.089
 brightly-0.079
 lot-0.075
 truly-0.074
 valuable-0.072
Token
Token ID220
Feature activation-1.07e+01

Pos logit contributions
ases+0.039
per+0.032
pered+0.032
ena+0.032
ws+0.030
Neg logit contributions
 brand-0.034
 brightly-0.029
 lot-0.027
 truly-0.027
 valuable-0.026
Token
Token ID198
Feature activation-3.72e+00

Pos logit contributions
 brand+0.214
 brightly+0.161
 lot+0.160
 truly+0.153
 valuable+0.141
Neg logit contributions
ases-0.446
per-0.398
ws-0.385
wed-0.382
pered-0.376
Token
Token ID198
Feature activation-3.79e+00

Pos logit contributions
 brand+0.083
 brightly+0.066
 lot+0.064
 truly+0.062
 valuable+0.058
Neg logit contributions
ases-0.149
per-0.131
ws-0.126
wed-0.125
pered-0.124
TokenTam
Token ID42061
Feature activation+1.46e+01

Pos logit contributions
 brand+0.096
 brightly+0.080
 lot+0.076
 truly+0.074
 valuable+0.071
Neg logit contributions
ases-0.141
per-0.122
ws-0.118
pered-0.116
wed-0.115
Tokenmy
Token ID1820
Feature activation+3.51e+00

Pos logit contributions
ases+0.496
ws+0.413
per+0.409
pered+0.408
ucks+0.370
Neg logit contributions
 brand-0.408
 truly-0.348
 brightly-0.335
 wonderful-0.328
 lot-0.325
Token felt
Token ID2936
Feature activation-3.84e+00

Pos logit contributions
ases+0.145
ws+0.121
pered+0.113
per+0.111
wed+0.110
Neg logit contributions
 brand-0.083
 truly-0.068
 very-0.068
 brightly-0.067
 lot-0.065
Token really
Token ID1107
Feature activation-8.38e+00

Pos logit contributions
 brand+0.083
 brightly+0.068
+0.067
 lifetime+0.062
ww+0.061
Neg logit contributions
ases-0.138
pered-0.129
per-0.124
ily-0.121
e-0.119
Token bad
Token ID2089
Feature activation-2.56e+00

Pos logit contributions
 brand+0.206
 brightly+0.167
 lot+0.155
 truly+0.150
+0.145
Neg logit contributions
ases-0.311
per-0.273
pered-0.269
ena-0.259
ws-0.258
Token for
Token ID329
Feature activation+4.77e+00

Pos logit contributions
 brand+0.087
 brightly+0.077
 lot+0.074
 truly+0.071
 valuable+0.070
Neg logit contributions
ases-0.073
per-0.060
ws-0.059
wed-0.057
pered-0.057
Token sorry
Token ID7926
Feature activation+1.71e+00

Pos logit contributions
 brand+0.094
 brightly+0.075
 truly+0.073
 lot+0.071
 wonderful+0.071
Neg logit contributions
ases-0.165
per-0.139
race-0.138
ws-0.137
wed-0.137
Token again
Token ID757
Feature activation-1.99e+00

Pos logit contributions
ases+0.046
per+0.038
pered+0.036
ena+0.036
ws+0.035
Neg logit contributions
 brand-0.061
 brightly-0.053
 lot-0.051
 truly-0.050
 valuable-0.049
Token.
Token ID13
Feature activation+2.45e+00

Pos logit contributions
 brand+0.061
 brightly+0.054
 lot+0.053
 valuable+0.052
 truly+0.052
Neg logit contributions
ases-0.064
ws-0.051
aled-0.051
pered-0.050
per-0.050
Token
Token ID198
Feature activation-2.52e+00

Pos logit contributions
ases+0.076
per+0.065
ena+0.063
pered+0.062
ws+0.060
Neg logit contributions
 brand-0.076
 brightly-0.064
 truly-0.061
 lot-0.060
 fantastic-0.058
Token
Token ID198
Feature activation-2.59e+00

Pos logit contributions
 brand+0.055
 brightly+0.043
 lot+0.042
 truly+0.039
 valuable+0.038
Neg logit contributions
ases-0.102
per-0.090
pered-0.087
ws-0.086
wed-0.086
TokenMiss
Token ID17140
Feature activation+1.46e+01

Pos logit contributions
 brand+0.071
 brightly+0.060
 lot+0.058
 truly+0.054
 valuable+0.054
Neg logit contributions
ases-0.091
per-0.077
pered-0.076
ws-0.076
wed-0.075
Token Lee
Token ID5741
Feature activation+6.45e+00

Pos logit contributions
ases+0.334
pered+0.272
ws+0.271
per+0.266
wed+0.249
Neg logit contributions
 brand-0.540
 brightly-0.487
 lot-0.465
 truly-0.451
 most-0.445
Token smiled
Token ID13541
Feature activation+1.88e+00

Pos logit contributions
ases+0.218
pered+0.178
per+0.176
ws+0.174
wed+0.166
Neg logit contributions
 brand-0.193
 brightly-0.163
 truly-0.154
 lot-0.153
 wonderful-0.145
Token.
Token ID13
Feature activation+2.39e+00

Pos logit contributions
ases+0.048
per+0.039
pered+0.037
ena+0.036
ws+0.036
Neg logit contributions
 brand-0.070
 brightly-0.062
 lot-0.059
 truly-0.058
 valuable-0.057
Token She
Token ID1375
Feature activation+1.34e+00

Pos logit contributions
ases+0.075
per+0.064
pered+0.061
ena+0.061
ws+0.060
Neg logit contributions
 brand-0.073
 brightly-0.062
 lot-0.059
 truly-0.059
 valuable-0.056
Token gave
Token ID2921
Feature activation+7.58e+00

Pos logit contributions
ases+0.043
per+0.036
pered+0.035
ws+0.034
ena+0.034
Neg logit contributions
 brand-0.041
 brightly-0.035
 lot-0.033
 truly-0.033
 valuable-0.031
Token says
Token ID1139
Feature activation+6.58e+00

Pos logit contributions
 brand+0.016
 lot+0.014
 brightly+0.014
 truly+0.013
 wonderful+0.013
Neg logit contributions
ases-0.012
pered-0.010
per-0.010
wed-0.009
ena-0.009
Token proudly
Token ID23017
Feature activation+4.21e+00

Pos logit contributions
ases+0.206
ws+0.171
wed+0.167
per+0.161
ena+0.160
Neg logit contributions
 brand-0.209
 brightly-0.187
 very-0.175
 much-0.167
 lot-0.165
Token.
Token ID13
Feature activation+1.97e+00

Pos logit contributions
ases+0.138
ws+0.114
wed+0.113
pered+0.107
ena+0.106
Neg logit contributions
 brand-0.118
 brightly-0.104
 very-0.098
 lot-0.092
 valuable-0.090
Token
Token ID198
Feature activation-2.95e+00

Pos logit contributions
ases+0.077
per+0.067
ws+0.065
wed+0.065
pered+0.065
Neg logit contributions
 brand-0.047
 brightly-0.038
 lot-0.036
 truly-0.034
 valuable-0.033
Token
Token ID198
Feature activation-3.00e+00

Pos logit contributions
 brand+0.065
 brightly+0.052
 lot+0.050
 truly+0.047
 valuable+0.045
Neg logit contributions
ases-0.118
per-0.105
pered-0.101
ws-0.101
wed-0.100
TokenMiss
Token ID17140
Feature activation+1.45e+01

Pos logit contributions
 brand+0.077
 brightly+0.069
 lot+0.069
 valuable+0.062
 truly+0.060
Neg logit contributions
ases-0.104
ws-0.090
wed-0.088
pered-0.086
ps-0.083
Token Lee
Token ID5741
Feature activation+6.72e+00

Pos logit contributions
ases+0.343
pered+0.285
ws+0.281
per+0.277
wed+0.274
Neg logit contributions
 brand-0.518
 brightly-0.470
 lot-0.447
 truly-0.434
 most-0.429
Token smiles
Token ID21845
Feature activation+1.17e+00

Pos logit contributions
ases+0.212
pered+0.186
per+0.176
race+0.171
wed+0.171
Neg logit contributions
 brand-0.194
 lot-0.164
 brightly-0.164
 wonderful-0.158
 truly-0.157
Token and
Token ID290
Feature activation-8.48e-01

Pos logit contributions
ases+0.029
per+0.023
pered+0.022
ena+0.021
ws+0.021
Neg logit contributions
 brand-0.044
 brightly-0.039
 lot-0.037
 truly-0.037
 valuable-0.036
Token cl
Token ID537
Feature activation+3.38e+00

Pos logit contributions
 brand+0.032
 brightly+0.028
 lot+0.027
 truly+0.026
 wonderful+0.026
Neg logit contributions
ases-0.021
per-0.017
pered-0.017
wed-0.016
ena-0.016
Tokenaps
Token ID1686
Feature activation+3.76e+00

Pos logit contributions
ases+0.069
per+0.062
wed+0.059
ena+0.058
pered+0.057
Neg logit contributions
 brand-0.123
 brightly-0.111
 truly-0.103
 lot-0.101
 brilliantly-0.100
Token had
Token ID550
Feature activation+3.67e-01

Pos logit contributions
 brand+0.024
 brightly+0.021
 lot+0.020
 truly+0.019
 valuable+0.019
Neg logit contributions
ases-0.025
per-0.021
pered-0.020
ws-0.020
ena-0.020
Token fun
Token ID1257
Feature activation-6.45e+00

Pos logit contributions
ases+0.016
per+0.015
ena+0.014
ws+0.014
wed+0.014
Neg logit contributions
 brand-0.006
 brightly-0.004
-0.004
 lot-0.003
 truly-0.003
Token.
Token ID13
Feature activation+2.43e+00

Pos logit contributions
 brand+0.232
 brightly+0.207
 lot+0.203
 truly+0.192
 valuable+0.186
Neg logit contributions
ases-0.170
per-0.138
ws-0.133
pered-0.132
ena-0.131
Token
Token ID198
Feature activation-2.47e+00

Pos logit contributions
ases+0.074
per+0.063
pered+0.060
ena+0.060
ws+0.059
Neg logit contributions
 brand-0.080
 brightly-0.066
 truly-0.063
 lot-0.060
 fantastic-0.060
Token
Token ID198
Feature activation-2.54e+00

Pos logit contributions
 brand+0.056
 brightly+0.044
 lot+0.043
 truly+0.041
 valuable+0.039
Neg logit contributions
ases-0.099
per-0.087
pered-0.084
ws-0.082
ena-0.082
TokenMiss
Token ID17140
Feature activation+1.45e+01

Pos logit contributions
 brand+0.063
 brightly+0.051
 lot+0.048
 truly+0.047
 valuable+0.044
Neg logit contributions
ases-0.096
per-0.084
pered-0.081
ena-0.081
ws-0.080
Token Lily
Token ID20037
Feature activation+7.73e+00

Pos logit contributions
ases+0.372
pered+0.313
ws+0.307
per+0.303
wed+0.288
Neg logit contributions
 brand-0.474
 brightly-0.444
 lot-0.418
 truly-0.406
 most-0.399
Token saw
Token ID2497
Feature activation+5.82e+00

Pos logit contributions
ases+0.308
pered+0.262
ws+0.253
rum+0.247
race+0.246
Neg logit contributions
 brand-0.179
 truly-0.149
 very-0.145
 lot-0.138
 fun-0.134
Token them
Token ID606
Feature activation+9.00e-01

Pos logit contributions
ases+0.167
per+0.139
pered+0.134
ena+0.130
ws+0.130
Neg logit contributions
 brand-0.198
 brightly-0.171
 lot-0.164
 truly-0.160
 valuable-0.154
Token and
Token ID290
Feature activation-7.89e-01

Pos logit contributions
ases+0.028
ws+0.023
per+0.023
pered+0.023
wed+0.023
Neg logit contributions
 brand-0.028
 brightly-0.024
 lot-0.024
 very-0.024
 truly-0.023
Token smiled
Token ID13541
Feature activation+1.88e+00

Pos logit contributions
 brand+0.023
 brightly+0.020
 lot+0.019
 truly+0.018
 wonderful+0.017
Neg logit contributions
ases-0.027
per-0.023
pered-0.022
ws-0.022
wed-0.021
Token to
Token ID284
Feature activation+6.33e+00

Pos logit contributions
 brand+0.138
 brightly+0.115
 lot+0.108
 truly+0.106
 valuable+0.102
Neg logit contributions
ases-0.178
per-0.153
pered-0.147
ena-0.147
ws-0.146
Token do
Token ID466
Feature activation-7.64e-01

Pos logit contributions
ases+0.203
per+0.174
pered+0.171
ws+0.164
ena+0.161
Neg logit contributions
 brand-0.192
 brightly-0.164
 lot-0.158
 truly-0.153
 wonderful-0.148
Token.
Token ID13
Feature activation+2.11e+00

Pos logit contributions
 brand+0.024
 brightly+0.021
 lot+0.020
 truly+0.020
 very+0.020
Neg logit contributions
ases-0.022
pered-0.019
per-0.018
ws-0.018
wed-0.018
Token
Token ID198
Feature activation-2.53e+00

Pos logit contributions
ases+0.069
per+0.058
ena+0.057
pered+0.057
ws+0.054
Neg logit contributions
 brand-0.064
 brightly-0.053
 truly-0.050
 lot-0.049
 fantastic-0.048
Token
Token ID198
Feature activation-2.62e+00

Pos logit contributions
 brand+0.062
 brightly+0.050
 lot+0.048
 truly+0.046
 valuable+0.044
Neg logit contributions
ases-0.096
per-0.084
pered-0.081
ws-0.080
ena-0.080
TokenMiss
Token ID17140
Feature activation+1.45e+01

Pos logit contributions
 brand+0.070
 brightly+0.060
 lot+0.058
 truly+0.055
 valuable+0.054
Neg logit contributions
ases-0.093
per-0.080
pered-0.077
ws-0.077
wed-0.075
Token Lee
Token ID5741
Feature activation+6.46e+00

Pos logit contributions
ases+0.357
ws+0.301
pered+0.294
per+0.292
wed+0.280
Neg logit contributions
 brand-0.500
 brightly-0.445
 lot-0.421
 truly-0.408
 most-0.407
Token saw
Token ID2497
Feature activation+5.51e+00

Pos logit contributions
ases+0.246
pered+0.206
ws+0.203
per+0.199
race+0.195
Neg logit contributions
 brand-0.164
 brightly-0.136
 lot-0.128
 truly-0.127
 very-0.119
Token what
Token ID644
Feature activation-5.20e+00

Pos logit contributions
ases+0.169
per+0.145
pered+0.141
wed+0.136
ws+0.136
Neg logit contributions
 brand-0.175
 brightly-0.146
 lot-0.146
 truly-0.134
 wonderful-0.132
Token happened
Token ID3022
Feature activation+3.22e+00

Pos logit contributions
 brand+0.131
 brightly+0.106
 lot+0.099
 truly+0.097
 valuable+0.092
Neg logit contributions
ases-0.194
per-0.166
ena-0.164
pered-0.160
ws-0.159
Token.
Token ID13
Feature activation+1.96e+00

Pos logit contributions
ases+0.093
per+0.074
ws+0.073
pered+0.072
wed+0.072
Neg logit contributions
 brand-0.108
 brightly-0.094
 lot-0.092
 very-0.090
 truly-0.089
Token like
Token ID588
Feature activation-1.80e+00

Pos logit contributions
ases+0.093
per+0.080
pered+0.077
ena+0.076
ws+0.076
Neg logit contributions
 brand-0.075
 brightly-0.062
 lot-0.059
 truly-0.058
 valuable-0.056
Token you
Token ID345
Feature activation+2.78e+00

Pos logit contributions
 brand+0.050
 brightly+0.042
 lot+0.040
 truly+0.038
 valuable+0.038
Neg logit contributions
ases-0.062
per-0.056
pered-0.054
wed-0.053
ws-0.053
Token."
Token ID526
Feature activation+3.28e+00

Pos logit contributions
ases+0.079
per+0.065
pered+0.062
ena+0.061
ws+0.061
Neg logit contributions
 brand-0.095
 brightly-0.082
 lot-0.080
 truly-0.078
 valuable-0.075
Token
Token ID198
Feature activation-2.36e+00

Pos logit contributions
ases+0.143
wed+0.127
ws+0.126
per+0.125
pered+0.122
Neg logit contributions
 brand-0.063
 brightly-0.049
 lot-0.048
 truly-0.041
 valuable-0.040
Token
Token ID198
Feature activation-2.47e+00

Pos logit contributions
 brand+0.040
 lot+0.030
 brightly+0.029
 truly+0.025
 valuable+0.025
Neg logit contributions
ases-0.106
per-0.096
wed-0.095
ws-0.093
pered-0.093
TokenMiss
Token ID17140
Feature activation+1.44e+01

Pos logit contributions
 brand+0.054
 brightly+0.047
 lot+0.046
 valuable+0.042
 truly+0.041
Neg logit contributions
ases-0.096
ws-0.085
wed-0.085
pered-0.083
ps-0.082
Token Lee
Token ID5741
Feature activation+6.31e+00

Pos logit contributions
ases+0.334
pered+0.287
per+0.279
ws+0.278
wed+0.274
Neg logit contributions
 brand-0.506
 brightly-0.446
 lot-0.434
-0.418
 most-0.415
Token says
Token ID1139
Feature activation+6.49e+00

Pos logit contributions
ases+0.187
pered+0.161
per+0.151
wed+0.149
race+0.147
Neg logit contributions
 brand-0.202
 brightly-0.171
 lot-0.169
 truly-0.165
 wonderful-0.163
Token,
Token ID11
Feature activation-2.38e+00

Pos logit contributions
ases+0.221
per+0.188
pered+0.181
ws+0.181
ena+0.180
Neg logit contributions
 brand-0.187
 brightly-0.158
 lot-0.149
 truly-0.145
 valuable-0.140
Token "
Token ID366
Feature activation-1.81e+00

Pos logit contributions
 brand+0.063
 brightly+0.052
 lot+0.049
 truly+0.049
 valuable+0.047
Neg logit contributions
ases-0.084
per-0.074
pered-0.071
ena-0.070
ws-0.069
TokenYou
Token ID1639
Feature activation+1.90e+00

Pos logit contributions
 brand+0.052
 brightly+0.042
 lot+0.040
 truly+0.039
 fantastic+0.038
Neg logit contributions
ases-0.062
per-0.054
ena-0.052
pered-0.050
ws-0.049
Token many
Token ID867
Feature activation-2.77e+00

Pos logit contributions
ases+0.004
per+0.004
pered+0.004
ws+0.003
ena+0.003
Neg logit contributions
 brand-0.003
 brightly-0.003
 lot-0.003
 truly-0.003
 valuable-0.002
Token things
Token ID1243
Feature activation-8.64e+00

Pos logit contributions
 brand+0.083
 brightly+0.071
 lot+0.069
 truly+0.066
 valuable+0.064
Neg logit contributions
ases-0.089
per-0.077
pered-0.074
ws-0.072
ena-0.072
Token.
Token ID13
Feature activation+2.11e+00

Pos logit contributions
 brand+0.282
 brightly+0.243
 lot+0.233
 truly+0.229
 valuable+0.222
Neg logit contributions
ases-0.256
per-0.214
pered-0.204
ena-0.202
ws-0.201
Token
Token ID198
Feature activation-2.60e+00

Pos logit contributions
ases+0.066
per+0.055
ena+0.054
pered+0.053
ws+0.052
Neg logit contributions
 brand-0.066
 brightly-0.056
 lot-0.053
 truly-0.053
 valuable-0.051
Token
Token ID198
Feature activation-2.68e+00

Pos logit contributions
 brand+0.062
 brightly+0.050
 lot+0.048
 truly+0.046
 valuable+0.044
Neg logit contributions
ases-0.100
per-0.088
pered-0.085
ws-0.084
ena-0.083
TokenMiss
Token ID17140
Feature activation+1.44e+01

Pos logit contributions
 brand+0.085
 brightly+0.076
 lot+0.074
 truly+0.071
 valuable+0.069
Neg logit contributions
ases-0.081
per-0.067
pered-0.066
ws-0.066
wed-0.064
Token Lily
Token ID20037
Feature activation+7.75e+00

Pos logit contributions
ases+0.372
per+0.309
ws+0.304
pered+0.303
wed+0.296
Neg logit contributions
 brand-0.499
 brightly-0.443
 lot-0.426
 truly-0.416
 wonderful-0.408
Token says
Token ID1139
Feature activation+6.62e+00

Pos logit contributions
ases+0.258
pered+0.227
rum+0.222
wed+0.209
race+0.207
Neg logit contributions
 brand-0.191
 very-0.165
 fun-0.163
 truly-0.162
 wonderful-0.160
Token,
Token ID11
Feature activation-2.19e+00

Pos logit contributions
ases+0.230
per+0.197
ws+0.190
pered+0.190
ena+0.187
Neg logit contributions
 brand-0.185
 brightly-0.156
 lot-0.148
 truly-0.143
 valuable-0.140
Token "
Token ID366
Feature activation-1.75e+00

Pos logit contributions
 brand+0.057
 brightly+0.047
 lot+0.044
 truly+0.043
 valuable+0.041
Neg logit contributions
ases-0.080
per-0.069
pered-0.067
ena-0.066
ws-0.066
TokenDo
Token ID5211
Feature activation+3.31e+00

Pos logit contributions
 brand+0.046
 brightly+0.039
 lot+0.037
 truly+0.036
 valuable+0.035
Neg logit contributions
ases-0.063
per-0.054
pered-0.053
ws-0.052
wed-0.051
Token.
Token ID13
Feature activation+2.54e+00

Pos logit contributions
ases+0.040
per+0.036
pered+0.034
ena+0.033
ws+0.033
Neg logit contributions
 brand-0.027
 brightly-0.022
 lot-0.021
 truly-0.020
 wonderful-0.020
Token She
Token ID1375
Feature activation+3.16e+00

Pos logit contributions
ases+0.098
ws+0.085
per+0.085
wed+0.085
pered+0.081
Neg logit contributions
 brand-0.061
 brightly-0.051
 lot-0.048
 truly-0.046
-0.044
Token hugs
Token ID40657
Feature activation+8.04e+00

Pos logit contributions
ases+0.115
pered+0.099
per+0.096
race+0.095
ws+0.090
Neg logit contributions
 brand-0.082
 truly-0.069
 brightly-0.068
 lot-0.068
 very-0.064
Token Tom
Token ID4186
Feature activation+1.15e+01

Pos logit contributions
ases+0.242
ws+0.203
wed+0.199
per+0.195
pered+0.191
Neg logit contributions
 brand-0.257
 brightly-0.235
 lot-0.226
 very-0.207
 truly-0.200
Token and
Token ID290
Feature activation-5.39e-01

Pos logit contributions
ases+0.399
pered+0.319
per+0.318
ws+0.308
wed+0.307
Neg logit contributions
 brand-0.334
 very-0.291
 lot-0.277
 brightly-0.275
 most-0.258
Token asks
Token ID7893
Feature activation+1.44e+01

Pos logit contributions
 brand+0.019
 brightly+0.016
 lot+0.016
 wonderful+0.015
 fun+0.015
Neg logit contributions
ases-0.015
per-0.013
pered-0.012
wed-0.011
ws-0.011
Token him
Token ID683
Feature activation+4.19e+00

Pos logit contributions
ases+0.446
per+0.388
pered+0.377
wed+0.372
aled+0.369
Neg logit contributions
 brand-0.404
 lot-0.376
 brightly-0.371
 very-0.367
 much-0.347
Token what
Token ID644
Feature activation-4.01e+00

Pos logit contributions
ases+0.147
per+0.124
pered+0.121
wed+0.120
race+0.118
Neg logit contributions
 brand-0.116
 brightly-0.095
 lot-0.094
 very-0.091
 wonderful-0.089
Token happened
Token ID3022
Feature activation+4.39e+00

Pos logit contributions
 brand+0.105
 brightly+0.086
 lot+0.081
 truly+0.079
 valuable+0.075
Neg logit contributions
ases-0.146
per-0.126
ena-0.122
pered-0.121
ws-0.120
Token.
Token ID13
Feature activation+3.35e+00

Pos logit contributions
ases+0.138
per+0.113
ws+0.113
wed+0.113
aled+0.109
Neg logit contributions
 brand-0.132
 brightly-0.118
 very-0.115
 truly-0.109
 valuable-0.107
Token Tom
Token ID4186
Feature activation+1.05e+01

Pos logit contributions
ases+0.114
per+0.098
wed+0.094
ws+0.094
pered+0.093
Neg logit contributions
 brand-0.096
 brightly-0.080
 lot-0.077
 truly-0.074
 valuable-0.071
Token mom
Token ID1995
Feature activation+7.73e+00

Pos logit contributions
ases+0.084
per+0.071
ws+0.068
pered+0.068
wed+0.066
Neg logit contributions
 brand-0.079
 brightly-0.067
 lot-0.064
 truly-0.062
 valuable-0.060
Tokenmy
Token ID1820
Feature activation+4.42e+00

Pos logit contributions
ases+0.292
pered+0.243
ws+0.234
per+0.231
ags+0.223
Neg logit contributions
 brand-0.217
 truly-0.181
 very-0.171
 brightly-0.170
 delicious-0.165
Token gave
Token ID2921
Feature activation+7.37e+00

Pos logit contributions
ases+0.177
pered+0.144
ws+0.142
ags+0.140
per+0.135
Neg logit contributions
 brand-0.113
 truly-0.093
 very-0.087
 brightly-0.086
 lot-0.084
Token him
Token ID683
Feature activation+2.54e+00

Pos logit contributions
ases+0.210
per+0.180
pered+0.169
ena+0.166
ws+0.161
Neg logit contributions
 brand-0.244
 brightly-0.219
 lot-0.212
 very-0.200
 wonderful-0.199
Token a
Token ID257
Feature activation-5.41e+00

Pos logit contributions
ases+0.103
per+0.091
pered+0.088
ws+0.086
ena+0.085
Neg logit contributions
 brand-0.058
 brightly-0.045
 lot-0.044
 truly-0.041
 wonderful-0.041
Token brand
Token ID4508
Feature activation-1.78e+01

Pos logit contributions
 brand+0.119
 brightly+0.115
 truly+0.101
 brilliantly+0.100
+0.098
Neg logit contributions
ases-0.168
ena-0.146
ws-0.143
 aid-0.141
race-0.140
Token new
Token ID649
Feature activation-3.08e+00

Pos logit contributions
 brand+0.313
+0.285
 sudden+0.283
 truly+0.272
 brightly+0.269
Neg logit contributions
ases-0.656
ucks-0.564
ws-0.554
wed-0.553
ena-0.547
Token pair
Token ID5166
Feature activation+1.48e+00

Pos logit contributions
 brand+0.088
 brightly+0.074
 lot+0.070
 truly+0.069
 valuable+0.067
Neg logit contributions
ases-0.103
per-0.088
pered-0.085
ena-0.084
ws-0.083
Token of
Token ID286
Feature activation+4.89e+00

Pos logit contributions
ases+0.050
ws+0.042
per+0.042
pered+0.041
ena+0.040
Neg logit contributions
 brand-0.044
 brightly-0.037
 truly-0.036
 very-0.035
 lot-0.034
Token shoes
Token ID10012
Feature activation-5.12e-01

Pos logit contributions
ases+0.169
per+0.147
pered+0.142
ws+0.139
ena+0.137
Neg logit contributions
 brand-0.142
 brightly-0.119
 lot-0.116
 truly-0.109
 wonderful-0.108
Token to
Token ID284
Feature activation+5.84e+00

Pos logit contributions
 brand+0.018
 brightly+0.015
 lot+0.015
 truly+0.014
 valuable+0.014
Neg logit contributions
ases-0.014
per-0.012
pered-0.011
ena-0.011
ws-0.011
Token her
Token ID607
Feature activation+3.66e+00

Pos logit contributions
ases+0.127
per+0.101
ws+0.098
pered+0.097
wed+0.094
Neg logit contributions
 brand-0.178
 brightly-0.159
 lot-0.152
 truly-0.149
 wonderful-0.143
Token-
Token ID12
Feature activation+9.21e-02

Pos logit contributions
ases+0.118
per+0.097
ws+0.097
pered+0.094
wed+0.091
Neg logit contributions
 brand-0.122
 brightly-0.103
 truly-0.097
 lot-0.097
 valuable-0.096
Token there
Token ID612
Feature activation-2.49e+00

Pos logit contributions
ases+0.003
per+0.003
pered+0.002
ws+0.002
wed+0.002
Neg logit contributions
 brand-0.003
 brightly-0.002
 lot-0.002
 truly-0.002
 valuable-0.002
Token was
Token ID373
Feature activation-4.67e+00

Pos logit contributions
 brand+0.086
 brightly+0.077
 truly+0.074
 lot+0.073
 wonderful+0.069
Neg logit contributions
ases-0.072
per-0.057
ws-0.057
pered-0.057
wed-0.054
Token a
Token ID257
Feature activation-5.50e+00

Pos logit contributions
 brand+0.124
 brightly+0.102
 lot+0.096
 truly+0.095
 valuable+0.090
Neg logit contributions
ases-0.172
per-0.149
pered-0.144
ws-0.142
ena-0.139
Token brand
Token ID4508
Feature activation-1.77e+01

Pos logit contributions
 brand+0.114
 brightly+0.099
+0.092
 truly+0.092
 brilliantly+0.090
Neg logit contributions
ases-0.162
ena-0.152
 aid-0.150
per-0.150
pered-0.141
Token new
Token ID649
Feature activation-3.22e+00

Pos logit contributions
 brand+0.306
+0.270
 sudden+0.256
 truly+0.246
 brightly+0.245
Neg logit contributions
ases-0.666
per-0.574
ena-0.564
ws-0.561
wed-0.559
Token box
Token ID3091
Feature activation-6.62e+00

Pos logit contributions
 brand+0.090
 brightly+0.075
 lot+0.071
 truly+0.069
 valuable+0.067
Neg logit contributions
ases-0.114
per-0.098
pered-0.095
ws-0.093
ena-0.092
Token of
Token ID286
Feature activation+4.83e+00

Pos logit contributions
 brand+0.209
 brightly+0.176
 truly+0.169
 very+0.159
 lot+0.155
Neg logit contributions
ases-0.228
ws-0.193
pered-0.182
wed-0.178
per-0.173
Token cr
Token ID1067
Feature activation-3.09e+00

Pos logit contributions
ases+0.177
per+0.155
pered+0.149
ws+0.147
ena+0.146
Neg logit contributions
 brand-0.128
 brightly-0.106
 lot-0.100
 truly-0.096
 valuable-0.093
Tokenay
Token ID323
Feature activation+3.21e-01

Pos logit contributions
 brand+0.074
 brightly+0.060
 truly+0.054
 lot+0.054
 valuable+0.050
Neg logit contributions
ases-0.118
per-0.103
pered-0.102
ws-0.101
ena-0.098
Token wall
Token ID3355
Feature activation-6.93e+00

Pos logit contributions
 brand+0.005
 brightly+0.004
 truly+0.004
 lot+0.004
 valuable+0.004
Neg logit contributions
ases-0.006
per-0.005
ena-0.005
pered-0.005
ws-0.005
Token,
Token ID11
Feature activation-3.27e+00

Pos logit contributions
 brand+0.177
 brightly+0.147
 lot+0.139
 truly+0.136
 most+0.127
Neg logit contributions
ases-0.256
per-0.218
ws-0.216
pered-0.215
wed-0.208
Token there
Token ID612
Feature activation-2.37e+00

Pos logit contributions
 brand+0.099
 brightly+0.084
 lot+0.080
 truly+0.079
 valuable+0.076
Neg logit contributions
ases-0.105
per-0.089
pered-0.085
ena-0.084
ws-0.084
Token was
Token ID373
Feature activation-4.42e+00

Pos logit contributions
 brand+0.047
 brightly+0.039
 lot+0.036
 truly+0.035
 wonderful+0.031
Neg logit contributions
ases-0.100
per-0.089
pered-0.087
ws-0.086
ena-0.084
Token a
Token ID257
Feature activation-5.41e+00

Pos logit contributions
 brand+0.121
 brightly+0.100
 lot+0.095
 truly+0.094
 valuable+0.093
Neg logit contributions
ases-0.145
ena-0.127
per-0.124
pered-0.120
ws-0.118
Token brand
Token ID4508
Feature activation-1.77e+01

Pos logit contributions
 brand+0.112
 brightly+0.091
 truly+0.089
+0.088
 brilliantly+0.082
Neg logit contributions
ases-0.170
per-0.160
ena-0.158
 aid-0.151
pered-0.151
Token new
Token ID649
Feature activation-3.08e+00

Pos logit contributions
 brand+0.307
+0.271
 truly+0.239
 sudden+0.234
 brightly+0.232
Neg logit contributions
ases-0.699
per-0.597
ws-0.592
wed-0.589
ena-0.587
Token clock
Token ID8801
Feature activation-3.15e+00

Pos logit contributions
 brand+0.090
 brightly+0.076
 lot+0.073
 truly+0.071
 valuable+0.068
Neg logit contributions
ases-0.104
per-0.089
pered-0.085
ws-0.085
ena-0.084
Token.
Token ID13
Feature activation+1.22e+00

Pos logit contributions
 brand+0.093
 brightly+0.082
 truly+0.076
 very+0.073
 lot+0.071
Neg logit contributions
ases-0.108
ws-0.092
pered-0.088
wed-0.087
per-0.086
Token It
Token ID632
Feature activation-9.42e+00

Pos logit contributions
ases+0.036
per+0.030
pered+0.028
ws+0.028
ena+0.028
Neg logit contributions
 brand-0.041
 brightly-0.035
 lot-0.034
 truly-0.033
 valuable-0.032
Token was
Token ID373
Feature activation-4.56e+00

Pos logit contributions
 brand+0.244
 brightly+0.198
 lot+0.188
 truly+0.179
 valuable+0.176
Neg logit contributions
ases-0.338
per-0.294
ena-0.286
ws-0.279
pered-0.279
Token eyes
Token ID2951
Feature activation-5.65e+00

Pos logit contributions
ases+0.118
per+0.102
pered+0.098
ena+0.097
ws+0.096
Neg logit contributions
 brand-0.097
 brightly-0.081
 lot-0.077
 truly-0.076
 valuable-0.072
Token.
Token ID13
Feature activation+9.10e-01

Pos logit contributions
 brand+0.194
 brightly+0.168
 lot+0.162
 truly+0.159
 valuable+0.155
Neg logit contributions
ases-0.156
per-0.130
pered-0.122
ena-0.121
ws-0.120
Token Inside
Token ID14384
Feature activation-6.90e-01

Pos logit contributions
ases+0.031
per+0.027
ws+0.025
pered+0.025
ena+0.025
Neg logit contributions
 brand-0.026
 brightly-0.022
 lot-0.021
 truly-0.020
 valuable-0.019
Token was
Token ID373
Feature activation-4.89e+00

Pos logit contributions
 brand+0.018
 brightly+0.015
 lot+0.014
 truly+0.014
 valuable+0.013
Neg logit contributions
ases-0.025
per-0.021
pered-0.021
ena-0.020
ws-0.020
Token a
Token ID257
Feature activation-5.65e+00

Pos logit contributions
 brand+0.130
 brightly+0.108
 lot+0.103
 truly+0.100
 valuable+0.097
Neg logit contributions
ases-0.170
per-0.147
ena-0.143
pered-0.142
ws-0.139
Token brand
Token ID4508
Feature activation-1.77e+01

Pos logit contributions
 brand+0.090
+0.078
 brightly+0.077
 brilliantly+0.069
 truly+0.069
Neg logit contributions
ases-0.186
ena-0.177
per-0.172
 aid-0.170
 ear-0.163
Token new
Token ID649
Feature activation-3.30e+00

Pos logit contributions
 brand+0.294
+0.284
 sudden+0.258
 truly+0.249
 brightly+0.243
Neg logit contributions
ases-0.648
old-0.546
ws-0.545
wed-0.545
ena-0.544
Token dress
Token ID6576
Feature activation-1.49e+00

Pos logit contributions
 brand+0.087
 brightly+0.073
 truly+0.067
 lot+0.067
 valuable+0.064
Neg logit contributions
ases-0.115
per-0.099
ena-0.096
pered-0.095
ws-0.093
Token!
Token ID0
Feature activation+5.44e-01

Pos logit contributions
 brand+0.046
 brightly+0.039
 truly+0.036
 lot+0.036
 most+0.033
Neg logit contributions
ases-0.051
ws-0.042
pered-0.041
per-0.041
wed-0.040
Token
Token ID220
Feature activation-1.10e+01

Pos logit contributions
ases+0.019
per+0.016
ws+0.015
pered+0.015
ena+0.015
Neg logit contributions
 brand-0.015
 brightly-0.013
 lot-0.012
 truly-0.012
 valuable-0.011
Token
Token ID198
Feature activation-4.21e+00

Pos logit contributions
 brand+0.212
 brightly+0.149
 truly+0.147
 lot+0.145
+0.133
Neg logit contributions
ases-0.477
per-0.417
ws-0.410
wed-0.409
pered-0.398
Token a
Token ID257
Feature activation-5.58e+00

Pos logit contributions
 brand+0.116
 brightly+0.094
 lot+0.087
 truly+0.087
 valuable+0.081
Neg logit contributions
ases-0.186
per-0.161
ws-0.155
pered-0.155
wed-0.152
Token popular
Token ID2968
Feature activation-1.66e+00

Pos logit contributions
 brand+0.110
 brightly+0.093
+0.088
 truly+0.086
 brilliantly+0.084
Neg logit contributions
ases-0.168
per-0.164
ena-0.164
 aid-0.158
e-0.152
Token toy
Token ID13373
Feature activation-3.84e+00

Pos logit contributions
 brand+0.045
 brightly+0.037
 lot+0.036
 truly+0.034
 valuable+0.033
Neg logit contributions
ases-0.061
per-0.052
pered-0.051
ws-0.051
wed-0.050
Token,
Token ID11
Feature activation-3.43e+00

Pos logit contributions
 brand+0.119
 brightly+0.102
 lot+0.097
 truly+0.095
 very+0.090
Neg logit contributions
ases-0.124
ws-0.106
wed-0.103
per-0.101
pered-0.101
Token a
Token ID257
Feature activation-5.65e+00

Pos logit contributions
 brand+0.092
 brightly+0.076
 lot+0.072
 truly+0.071
 valuable+0.068
Neg logit contributions
ases-0.118
per-0.103
ena-0.099
pered-0.098
ws-0.096
Token brand
Token ID4508
Feature activation-1.77e+01

Pos logit contributions
 brand+0.121
 brightly+0.102
 truly+0.098
+0.093
 brilliantly+0.090
Neg logit contributions
ases-0.185
per-0.171
ena-0.169
pered-0.160
e-0.156
Token new
Token ID649
Feature activation-3.23e+00

Pos logit contributions
 brand+0.288
+0.271
 sudden+0.242
 truly+0.239
 brightly+0.237
Neg logit contributions
ases-0.671
per-0.574
ena-0.574
ws-0.568
wed-0.561
Token doll
Token ID3654
Feature activation+1.41e-01

Pos logit contributions
 brand+0.090
 brightly+0.075
 lot+0.072
 truly+0.069
 valuable+0.067
Neg logit contributions
ases-0.113
per-0.097
pered-0.094
ws-0.093
ena-0.092
Token.
Token ID13
Feature activation+9.09e-01

Pos logit contributions
ases+0.005
ws+0.004
pered+0.004
wed+0.004
per+0.004
Neg logit contributions
 brand-0.004
 brightly-0.004
 truly-0.004
 lot-0.003
 very-0.003
Token
Token ID198
Feature activation-3.99e+00

Pos logit contributions
ases+0.028
per+0.024
pered+0.024
ena+0.023
ws+0.022
Neg logit contributions
 brand-0.027
 brightly-0.022
 truly-0.021
 lot-0.021
 valuable-0.021
Token
Token ID198
Feature activation-4.06e+00

Pos logit contributions
 brand+0.101
 brightly+0.083
 lot+0.077
 truly+0.076
 valuable+0.073
Neg logit contributions
ases-0.148
per-0.129
pered-0.124
ena-0.123
ws-0.122
Token confidently
Token ID37891
Feature activation+3.40e+00

Pos logit contributions
ases+0.211
ws+0.176
wed+0.170
per+0.164
ena+0.162
Neg logit contributions
 brand-0.195
 brightly-0.173
 lot-0.163
 very-0.157
 truly-0.154
Token back
Token ID736
Feature activation+3.73e+00

Pos logit contributions
ases+0.115
ws+0.097
wed+0.094
ena+0.091
adies+0.090
Neg logit contributions
 brand-0.098
 brightly-0.082
 lot-0.081
 very-0.080
 truly-0.078
Token home
Token ID1363
Feature activation-7.77e-01

Pos logit contributions
ases+0.138
wed+0.121
ws+0.120
per+0.117
pered+0.114
Neg logit contributions
 brand-0.093
 lot-0.079
 brightly-0.077
 very-0.075
 truly-0.070
Token with
Token ID351
Feature activation+2.49e+00

Pos logit contributions
 brand+0.023
 brightly+0.020
 lot+0.019
 very+0.019
 truly+0.019
Neg logit contributions
ases-0.026
ws-0.022
wed-0.021
pered-0.020
per-0.020
Token a
Token ID257
Feature activation-5.65e+00

Pos logit contributions
ases+0.085
per+0.073
pered+0.070
ena+0.069
ws+0.069
Neg logit contributions
 brand-0.070
 brightly-0.059
 lot-0.056
 truly-0.055
 valuable-0.053
Token brand
Token ID4508
Feature activation-1.77e+01

Pos logit contributions
 brand+0.126
 brightly+0.122
 brilliantly+0.109
+0.107
 truly+0.105
Neg logit contributions
ases-0.151
 aid-0.142
ena-0.138
per-0.135
race-0.130
Token new
Token ID649
Feature activation-3.26e+00

Pos logit contributions
 brand+0.318
+0.263
 truly+0.262
 brightly+0.260
 sudden+0.253
Neg logit contributions
ases-0.671
ws-0.565
wed-0.565
per-0.563
ucks-0.561
Token haircut
Token ID45548
Feature activation+1.20e-01

Pos logit contributions
 brand+0.098
 brightly+0.083
 lot+0.079
 truly+0.078
 valuable+0.075
Neg logit contributions
ases-0.105
per-0.089
pered-0.085
ena-0.085
ws-0.084
Token.
Token ID13
Feature activation+9.26e-01

Pos logit contributions
ases+0.003
pered+0.003
ws+0.003
per+0.003
wed+0.003
Neg logit contributions
 brand-0.004
 brightly-0.004
 lot-0.003
 truly-0.003
 most-0.003
Token He
Token ID679
Feature activation-1.10e+00

Pos logit contributions
ases+0.035
per+0.030
wed+0.029
ws+0.029
pered+0.028
Neg logit contributions
 brand-0.025
 brightly-0.020
 lot-0.019
 truly-0.019
 valuable-0.018
Token was
Token ID373
Feature activation-4.73e+00

Pos logit contributions
 brand+0.029
 brightly+0.024
 truly+0.024
 lot+0.023
 wonderful+0.021
Neg logit contributions
ases-0.042
pered-0.035
per-0.035
ws-0.034
adies-0.032
Token's
Token ID338
Feature activation-1.17e+00

Pos logit contributions
ases+0.265
pered+0.215
ws+0.214
per+0.213
ena+0.212
Neg logit contributions
 brand-0.212
 brightly-0.172
 truly-0.168
 lot-0.164
 very-0.160
Token toys
Token ID14958
Feature activation-2.04e+00

Pos logit contributions
 brand+0.033
 brightly+0.028
 lot+0.026
 truly+0.026
 valuable+0.025
Neg logit contributions
ases-0.040
per-0.034
pered-0.033
ena-0.032
ws-0.032
Token and
Token ID290
Feature activation-2.20e+00

Pos logit contributions
 brand+0.074
 brightly+0.064
 lot+0.062
 truly+0.061
 valuable+0.059
Neg logit contributions
ases-0.053
per-0.044
pered-0.041
ena-0.040
ws-0.040
Token they
Token ID484
Feature activation-9.54e-02

Pos logit contributions
 brand+0.063
 brightly+0.053
 lot+0.051
 truly+0.050
 wonderful+0.048
Neg logit contributions
ases-0.077
per-0.066
pered-0.065
ws-0.062
ena-0.061
Token looked
Token ID3114
Feature activation-6.16e+00

Pos logit contributions
 brand+0.002
 brightly+0.002
 truly+0.002
 lot+0.002
 wonderful+0.002
Neg logit contributions
ases-0.004
per-0.003
pered-0.003
ena-0.003
ws-0.003
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.138
 brightly+0.113
+0.107
 lot+0.106
 truly+0.105
Neg logit contributions
ases-0.227
per-0.199
pered-0.195
ws-0.191
wed-0.191
Token new
Token ID649
Feature activation-3.04e+00

Pos logit contributions
+0.374
 brand+0.320
 sudden+0.319
Uh+0.312
Imagine+0.306
Neg logit contributions
ases-0.566
ucks-0.480
wed-0.471
old-0.468
ws-0.467
Token.
Token ID13
Feature activation+1.12e+00

Pos logit contributions
 brand+0.117
 brightly+0.102
 lot+0.099
 truly+0.098
 valuable+0.096
Neg logit contributions
ases-0.070
per-0.055
ena-0.051
pered-0.050
wed-0.049
Token Lily
Token ID20037
Feature activation+7.15e+00

Pos logit contributions
ases+0.039
per+0.033
pered+0.032
ena+0.032
ws+0.032
Neg logit contributions
 brand-0.031
 brightly-0.026
 lot-0.025
 truly-0.024
 valuable-0.023
Token was
Token ID373
Feature activation-4.69e+00

Pos logit contributions
ases+0.305
pered+0.260
ws+0.256
per+0.254
adies+0.245
Neg logit contributions
 brand-0.155
 truly-0.120
 brightly-0.117
 lot-0.111
 very-0.106
Token so
Token ID523
Feature activation-1.30e+01

Pos logit contributions
 brand+0.078
 brightly+0.061
 lot+0.056
+0.048
 valuable+0.047
Neg logit contributions
ases-0.204
per-0.185
ena-0.182
pered-0.181
wed-0.174
Token then
Token ID788
Feature activation-2.76e+00

Pos logit contributions
 brand+0.058
 brightly+0.049
 lot+0.047
 truly+0.046
 wonderful+0.043
Neg logit contributions
ases-0.072
per-0.062
pered-0.060
ws-0.058
ena-0.058
Token start
Token ID923
Feature activation+4.73e-01

Pos logit contributions
 brand+0.071
 brightly+0.062
 lot+0.058
 truly+0.057
 wonderful+0.055
Neg logit contributions
ases-0.101
pered-0.088
per-0.087
ws-0.083
ena-0.082
Token over
Token ID625
Feature activation+4.00e+00

Pos logit contributions
ases+0.015
per+0.013
pered+0.013
ena+0.012
ws+0.012
Neg logit contributions
 brand-0.015
 brightly-0.012
 lot-0.012
 truly-0.011
 valuable-0.011
Token with
Token ID351
Feature activation+2.75e+00

Pos logit contributions
ases+0.127
per+0.110
pered+0.106
ws+0.106
wed+0.104
Neg logit contributions
 brand-0.126
 brightly-0.109
 lot-0.103
 truly-0.099
 wonderful-0.096
Token a
Token ID257
Feature activation-5.46e+00

Pos logit contributions
ases+0.078
per+0.064
pered+0.061
ena+0.060
ws+0.060
Neg logit contributions
 brand-0.093
 brightly-0.081
 lot-0.077
 truly-0.076
 valuable-0.074
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.148
 brightly+0.129
 truly+0.120
 lot+0.115
 valuable+0.114
Neg logit contributions
ases-0.172
per-0.150
pered-0.144
ws-0.141
ena-0.140
Token new
Token ID649
Feature activation-3.09e+00

Pos logit contributions
 brand+0.312
 truly+0.261
+0.257
 brightly+0.254
 sudden+0.249
Neg logit contributions
ases-0.681
ws-0.581
wed-0.577
per-0.576
ucks-0.570
Token adventure
Token ID8855
Feature activation-2.04e+00

Pos logit contributions
 brand+0.086
 brightly+0.072
 lot+0.069
 truly+0.066
 valuable+0.064
Neg logit contributions
ases-0.109
per-0.093
pered-0.090
ena-0.089
ws-0.089
Token.
Token ID13
Feature activation+1.22e+00

Pos logit contributions
 brand+0.070
 brightly+0.062
 truly+0.058
 lot+0.057
 valuable+0.056
Neg logit contributions
ases-0.059
per-0.047
ws-0.047
pered-0.047
wed-0.045
Token
Token ID220
Feature activation-1.08e+01

Pos logit contributions
ases+0.047
per+0.041
pered+0.039
ws+0.039
ena+0.039
Neg logit contributions
 brand-0.030
 brightly-0.024
 lot-0.023
 truly-0.022
 valuable-0.021
Token
Token ID198
Feature activation-3.86e+00

Pos logit contributions
 brand+0.198
 lot+0.140
 brightly+0.139
 truly+0.138
+0.130
Neg logit contributions
ases-0.464
per-0.414
wed-0.406
ws-0.401
pered-0.396
Token was
Token ID373
Feature activation-4.67e+00

Pos logit contributions
 brand+0.213
 brightly+0.173
 lot+0.164
 truly+0.159
 valuable+0.159
Neg logit contributions
ases-0.284
per-0.248
ena-0.238
pered-0.238
wed-0.236
Token dry
Token ID5894
Feature activation+6.12e-01

Pos logit contributions
 brand+0.135
 brightly+0.115
 lot+0.109
 truly+0.107
 valuable+0.105
Neg logit contributions
ases-0.151
per-0.129
pered-0.126
wed-0.124
ws-0.124
Token,
Token ID11
Feature activation-3.58e+00

Pos logit contributions
ases+0.018
per+0.015
pered+0.015
ws+0.014
ena+0.014
Neg logit contributions
 brand-0.020
 brightly-0.017
 lot-0.016
 truly-0.016
 valuable-0.016
Token it
Token ID340
Feature activation-8.07e+00

Pos logit contributions
 brand+0.116
 brightly+0.097
 truly+0.093
 valuable+0.092
 lot+0.092
Neg logit contributions
ases-0.101
per-0.085
ena-0.083
pered-0.079
ws-0.076
Token looked
Token ID3114
Feature activation-6.32e+00

Pos logit contributions
 brand+0.234
 brightly+0.196
 lot+0.188
 truly+0.181
 valuable+0.177
Neg logit contributions
ases-0.264
per-0.229
ena-0.218
pered-0.217
wed-0.216
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.124
+0.117
ww+0.099
 brightly+0.098
 lot+0.098
Neg logit contributions
ases-0.224
per-0.201
pered-0.200
wed-0.198
ws-0.197
Token-
Token ID12
Feature activation+2.40e-01

Pos logit contributions
+0.370
 sudden+0.321
 brand+0.315
Imagine+0.295
 lifetime+0.283
Neg logit contributions
ases-0.539
ily-0.475
wed-0.462
ws-0.456
ucks-0.449
Tokennew
Token ID3605
Feature activation-2.62e+00

Pos logit contributions
ases+0.006
per+0.005
wed+0.005
pered+0.005
ws+0.005
Neg logit contributions
 brand-0.008
 brightly-0.007
 lot-0.007
 truly-0.007
 valuable-0.007
Token again
Token ID757
Feature activation-2.65e+00

Pos logit contributions
 brand+0.094
 brightly+0.082
 lot+0.077
 truly+0.077
 valuable+0.073
Neg logit contributions
ases-0.073
per-0.058
pered-0.056
ws-0.056
ena-0.056
Token!
Token ID0
Feature activation+7.25e-01

Pos logit contributions
 brand+0.085
 brightly+0.075
 truly+0.071
 lot+0.069
 wonderful+0.066
Neg logit contributions
ases-0.084
pered-0.068
per-0.067
ws-0.067
wed-0.064
Token L
Token ID406
Feature activation+6.91e+00

Pos logit contributions
ases+0.024
per+0.020
pered+0.019
ena+0.019
ws+0.019
Neg logit contributions
 brand-0.022
 brightly-0.019
 lot-0.018
 truly-0.017
 valuable-0.017
Token the
Token ID262
Feature activation-4.10e-01

Pos logit contributions
ases+0.166
per+0.139
pered+0.133
ena+0.131
ws+0.130
Neg logit contributions
 brand-0.180
 brightly-0.156
 lot-0.150
 truly-0.145
 valuable-0.140
Token freezer
Token ID30967
Feature activation-7.13e+00

Pos logit contributions
 brand+0.012
 brightly+0.010
 lot+0.009
 truly+0.009
 valuable+0.009
Neg logit contributions
ases-0.014
per-0.012
pered-0.012
ena-0.012
ws-0.012
Token until
Token ID1566
Feature activation-4.31e+00

Pos logit contributions
 brand+0.245
 brightly+0.213
 lot+0.203
 truly+0.200
 valuable+0.194
Neg logit contributions
ases-0.203
per-0.166
pered-0.159
ena-0.156
ws-0.156
Token it
Token ID340
Feature activation-8.01e+00

Pos logit contributions
 brand+0.127
 brightly+0.102
 valuable+0.101
 truly+0.098
 lot+0.096
Neg logit contributions
ases-0.127
per-0.108
wed-0.106
pered-0.105
ena-0.105
Token looked
Token ID3114
Feature activation-6.07e+00

Pos logit contributions
 brand+0.189
 brightly+0.152
 lot+0.142
 truly+0.138
 valuable+0.132
Neg logit contributions
ases-0.312
per-0.273
pered-0.264
ena-0.262
ws-0.261
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.111
+0.091
 brightly+0.085
 lot+0.081
 truly+0.080
Neg logit contributions
ases-0.234
per-0.213
pered-0.208
wed-0.207
ws-0.207
Token new
Token ID649
Feature activation-3.17e+00

Pos logit contributions
+0.346
 sudden+0.327
 brand+0.319
Imagine+0.286
Uh+0.278
Neg logit contributions
ases-0.572
ws-0.488
ucks-0.487
wed-0.476
rap-0.472
Token.
Token ID13
Feature activation+8.74e-01

Pos logit contributions
 brand+0.121
 brightly+0.106
 lot+0.103
 truly+0.101
 valuable+0.098
Neg logit contributions
ases-0.076
per-0.061
pered-0.057
ena-0.056
ws-0.056
Token
Token ID220
Feature activation-1.08e+01

Pos logit contributions
ases+0.026
per+0.022
pered+0.022
ena+0.022
ws+0.021
Neg logit contributions
 brand-0.027
 brightly-0.022
 truly-0.021
 lot-0.021
 valuable-0.021
Token
Token ID198
Feature activation-4.06e+00

Pos logit contributions
 brand+0.211
 brightly+0.154
 lot+0.152
 truly+0.148
+0.140
Neg logit contributions
ases-0.459
per-0.407
wed-0.394
ws-0.390
pered-0.387
Token
Token ID198
Feature activation-4.11e+00

Pos logit contributions
 brand+0.100
 brightly+0.082
 lot+0.077
 truly+0.075
 valuable+0.071
Neg logit contributions
ases-0.154
per-0.134
pered-0.129
ena-0.128
ws-0.128
Token "
Token ID366
Feature activation-2.39e+00

Pos logit contributions
 brand+0.083
 brightly+0.069
 lot+0.065
 truly+0.065
 fantastic+0.062
Neg logit contributions
ases-0.100
per-0.089
pered-0.086
ena-0.085
ws-0.082
TokenThat
Token ID2504
Feature activation-6.81e+00

Pos logit contributions
 brand+0.059
 brightly+0.046
 lot+0.044
 truly+0.043
 valuable+0.041
Neg logit contributions
ases-0.092
per-0.080
ena-0.077
pered-0.076
ws-0.076
Token's
Token ID338
Feature activation-1.29e+00

Pos logit contributions
 brand+0.230
 brightly+0.201
 lot+0.188
 truly+0.187
+0.187
Neg logit contributions
ases-0.179
per-0.153
ena-0.148
ws-0.141
wed-0.138
Token a
Token ID257
Feature activation-5.44e+00

Pos logit contributions
 brand+0.036
+0.031
 brightly+0.031
 sudden+0.031
 lifetime+0.028
Neg logit contributions
ases-0.039
ena-0.036
per-0.035
pered-0.034
ily-0.033
Token famous
Token ID5863
Feature activation-1.04e+00

Pos logit contributions
 brand+0.127
 brightly+0.114
+0.110
 brilliantly+0.103
 truly+0.099
Neg logit contributions
ases-0.156
per-0.147
e-0.141
 ear-0.138
ena-0.137
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.029
 brightly+0.024
 lot+0.023
 truly+0.022
 valuable+0.022
Neg logit contributions
ases-0.037
per-0.031
pered-0.031
ws-0.031
ena-0.030
Token,
Token ID11
Feature activation-3.28e+00

Pos logit contributions
 brand+0.312
+0.288
 sudden+0.280
 brightly+0.267
 truly+0.256
Neg logit contributions
ases-0.665
per-0.562
ucks-0.559
ena-0.559
ws-0.556
Token Lily
Token ID20037
Feature activation+7.08e+00

Pos logit contributions
 brand+0.090
 brightly+0.074
 lot+0.070
 truly+0.069
 valuable+0.066
Neg logit contributions
ases-0.113
per-0.098
ena-0.094
pered-0.094
ws-0.092
Token!"
Token ID2474
Feature activation+2.23e+00

Pos logit contributions
ases+0.237
pered+0.197
aled+0.191
ws+0.186
race+0.186
Neg logit contributions
 brand-0.209
 very-0.176
 brightly-0.176
 valuable-0.176
 truly-0.175
Token
Token ID198
Feature activation-3.61e+00

Pos logit contributions
ases+0.085
per+0.073
pered+0.071
ws+0.071
ena+0.070
Neg logit contributions
 brand-0.056
 brightly-0.045
 lot-0.043
 truly-0.041
 valuable-0.039
Token
Token ID198
Feature activation-3.66e+00

Pos logit contributions
 brand+0.078
 brightly+0.061
 lot+0.060
 truly+0.056
 valuable+0.054
Neg logit contributions
ases-0.149
per-0.131
pered-0.127
ws-0.126
wed-0.125
Token her
Token ID607
Feature activation+4.09e+00

Pos logit contributions
 brand+0.093
 truly+0.078
 valuable+0.077
 brightly+0.075
 fantastic+0.073
Neg logit contributions
ases-0.075
ena-0.064
per-0.063
ily-0.059
pered-0.058
Token mom
Token ID1995
Feature activation+7.99e+00

Pos logit contributions
ases+0.125
per+0.105
pered+0.100
ena+0.100
ws+0.098
Neg logit contributions
 brand-0.131
 brightly-0.113
 lot-0.107
 truly-0.106
 valuable-0.102
Token bought
Token ID5839
Feature activation+4.93e+00

Pos logit contributions
ases+0.278
pered+0.238
per+0.231
ws+0.218
ps+0.206
Neg logit contributions
 brand-0.234
 truly-0.198
 lot-0.194
 brightly-0.193
 very-0.188
Token her
Token ID607
Feature activation+3.99e+00

Pos logit contributions
ases+0.145
per+0.121
ena+0.115
pered+0.115
ws+0.114
Neg logit contributions
 brand-0.164
 brightly-0.141
 lot-0.134
 truly-0.133
 valuable-0.128
Token a
Token ID257
Feature activation-5.19e+00

Pos logit contributions
ases+0.155
per+0.136
pered+0.131
ws+0.129
ena+0.128
Neg logit contributions
 brand-0.097
 brightly-0.077
 lot-0.074
 truly-0.071
 valuable-0.068
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brightly+0.107
 brand+0.105
 truly+0.098
 brilliantly+0.098
+0.097
Neg logit contributions
ases-0.143
 aid-0.131
ena-0.128
per-0.123
race-0.123
Token new
Token ID649
Feature activation-2.93e+00

Pos logit contributions
 brand+0.320
 sudden+0.283
 truly+0.282
+0.282
 brightly+0.272
Neg logit contributions
ases-0.649
ucks-0.559
ws-0.544
wed-0.543
ena-0.538
Token pair
Token ID5166
Feature activation+1.77e+00

Pos logit contributions
 brand+0.085
 brightly+0.072
 truly+0.068
 lot+0.068
 valuable+0.065
Neg logit contributions
ases-0.095
per-0.081
pered-0.077
ena-0.077
ws-0.076
Token of
Token ID286
Feature activation+5.06e+00

Pos logit contributions
ases+0.060
ws+0.052
per+0.051
pered+0.049
ena+0.049
Neg logit contributions
 brand-0.051
 brightly-0.043
 very-0.042
 truly-0.042
 lot-0.041
Token boots
Token ID14412
Feature activation+8.67e-01

Pos logit contributions
ases+0.167
per+0.143
pered+0.138
ws+0.135
ena+0.135
Neg logit contributions
 brand-0.152
 brightly-0.129
 lot-0.125
 truly-0.119
 wonderful-0.116
Token.
Token ID13
Feature activation+1.39e+00

Pos logit contributions
ases+0.026
per+0.021
pered+0.021
ws+0.021
ena+0.021
Neg logit contributions
 brand-0.029
 brightly-0.024
 lot-0.024
 truly-0.023
-0.022
Token Billy
Token ID15890
Feature activation+8.88e+00

Pos logit contributions
ases+0.154
ws+0.140
wed+0.132
per+0.127
pered+0.123
Neg logit contributions
 brand-0.114
 brightly-0.089
 very-0.088
 truly-0.086
 lot-0.084
Token.
Token ID13
Feature activation+1.25e+00

Pos logit contributions
ases+0.321
ws+0.261
ena+0.253
wed+0.249
pered+0.248
Neg logit contributions
 brand-0.258
 very-0.211
 truly-0.202
 brightly-0.196
 shining-0.195
Token Billy
Token ID15890
Feature activation+8.75e+00

Pos logit contributions
ases+0.032
per+0.027
pered+0.027
ena+0.025
race+0.025
Neg logit contributions
 brand-0.042
 brightly-0.038
 truly-0.036
 valuable-0.035
 lot-0.035
Token had
Token ID550
Feature activation-2.34e-01

Pos logit contributions
ases+0.299
per+0.256
pered+0.246
ena+0.244
ws+0.243
Neg logit contributions
 brand-0.248
 brightly-0.208
 lot-0.197
 truly-0.194
 valuable-0.186
Token a
Token ID257
Feature activation-5.23e+00

Pos logit contributions
 brand+0.006
 brightly+0.005
 lot+0.005
 truly+0.005
 valuable+0.005
Neg logit contributions
ases-0.008
per-0.007
ena-0.007
pered-0.007
wed-0.007
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.112
 brightly+0.111
 brilliantly+0.101
 truly+0.101
izing+0.096
Neg logit contributions
ases-0.153
 aid-0.135
pered-0.135
per-0.132
ena-0.132
Token new
Token ID649
Feature activation-2.82e+00

Pos logit contributions
 brand+0.301
+0.276
 brightly+0.274
 truly+0.271
 sudden+0.263
Neg logit contributions
ases-0.688
ws-0.591
wed-0.582
ucks-0.577
ena-0.562
Token car
Token ID1097
Feature activation-2.03e+00

Pos logit contributions
 brand+0.081
 brightly+0.068
 lot+0.065
 truly+0.063
 valuable+0.061
Neg logit contributions
ases-0.096
per-0.082
pered-0.079
ena-0.078
ws-0.078
Token that
Token ID326
Feature activation-2.31e+00

Pos logit contributions
 brand+0.045
 brightly+0.033
 lot+0.031
 truly+0.031
 very+0.029
Neg logit contributions
ases-0.085
ws-0.074
per-0.074
pered-0.073
wed-0.072
Token could
Token ID714
Feature activation-2.80e+00

Pos logit contributions
 brand+0.065
 brightly+0.055
 lot+0.053
+0.052
 fantastic+0.050
Neg logit contributions
ases-0.071
ena-0.059
per-0.059
wed-0.058
race-0.057
Token speed
Token ID2866
Feature activation+1.97e+00

Pos logit contributions
 brand+0.093
 brightly+0.080
 lot+0.077
 truly+0.076
 wonderful+0.071
Neg logit contributions
ases-0.083
per-0.071
ws-0.068
pered-0.068
ena-0.067
Token he
Token ID339
Feature activation-1.25e+00

Pos logit contributions
 brand+0.257
 brightly+0.221
 lot+0.214
 truly+0.209
 valuable+0.206
Neg logit contributions
ases-0.235
per-0.204
pered-0.191
ena-0.190
ws-0.184
Token discovered
Token ID5071
Feature activation+2.64e+00

Pos logit contributions
 brand+0.035
 truly+0.028
 brightly+0.027
 lot+0.025
 wonderful+0.025
Neg logit contributions
ases-0.047
pered-0.039
per-0.038
ws-0.038
race-0.038
Token it
Token ID340
Feature activation-8.12e+00

Pos logit contributions
ases+0.103
per+0.089
pered+0.086
ws+0.086
ena+0.085
Neg logit contributions
 brand-0.064
 brightly-0.052
 lot-0.048
 truly-0.047
 valuable-0.044
Token was
Token ID373
Feature activation-4.83e+00

Pos logit contributions
 brand+0.235
 brightly+0.196
 truly+0.186
 lot+0.184
 valuable+0.175
Neg logit contributions
ases-0.281
per-0.236
pered-0.230
ws-0.228
ena-0.224
Token a
Token ID257
Feature activation-5.60e+00

Pos logit contributions
 brand+0.128
 brightly+0.104
 truly+0.096
 lot+0.096
 valuable+0.091
Neg logit contributions
ases-0.183
per-0.156
pered-0.151
ws-0.150
ena-0.149
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.104
+0.096
 brightly+0.091
 truly+0.089
 brilliantly+0.084
Neg logit contributions
ases-0.161
per-0.157
ena-0.157
 aid-0.155
 ear-0.152
Token new
Token ID649
Feature activation-3.20e+00

Pos logit contributions
+0.293
 brand+0.286
 sudden+0.266
 truly+0.256
 brightly+0.237
Neg logit contributions
ases-0.630
per-0.552
ws-0.536
rap-0.533
ena-0.533
Token bike
Token ID7161
Feature activation-6.18e-01

Pos logit contributions
 brand+0.087
 brightly+0.073
 lot+0.069
 truly+0.068
 valuable+0.065
Neg logit contributions
ases-0.109
per-0.094
ena-0.090
pered-0.090
ws-0.089
Token!
Token ID0
Feature activation+7.00e-01

Pos logit contributions
 brand+0.019
 brightly+0.015
 truly+0.015
 lot+0.014
 very+0.014
Neg logit contributions
ases-0.021
ws-0.018
pered-0.017
wed-0.017
per-0.017
Token
Token ID198
Feature activation-4.15e+00

Pos logit contributions
ases+0.026
per+0.022
ws+0.022
wed+0.022
pered+0.021
Neg logit contributions
 brand-0.018
 brightly-0.015
 lot-0.014
 truly-0.014
 valuable-0.013
Token
Token ID198
Feature activation-4.22e+00

Pos logit contributions
 brand+0.079
 lot+0.058
 brightly+0.058
 truly+0.056
 valuable+0.053
Neg logit contributions
ases-0.183
per-0.161
wed-0.160
ws-0.158
pered-0.154
Token up
Token ID510
Feature activation+1.73e+00

Pos logit contributions
ases+0.008
per+0.007
pered+0.007
ena+0.006
ws+0.006
Neg logit contributions
 brand-0.011
 lot-0.009
 brightly-0.009
 truly-0.009
-0.009
Token as
Token ID355
Feature activation-9.12e+00

Pos logit contributions
ases+0.051
per+0.041
ws+0.040
pered+0.040
ena+0.039
Neg logit contributions
 brand-0.058
 brightly-0.052
 lot-0.048
 truly-0.048
 valuable-0.046
Token she
Token ID673
Feature activation-3.66e-02

Pos logit contributions
 brand+0.206
 brightly+0.168
 lot+0.154
 truly+0.151
 valuable+0.143
Neg logit contributions
ases-0.365
per-0.320
pered-0.309
ws-0.307
ena-0.305
Token saw
Token ID2497
Feature activation+5.23e+00

Pos logit contributions
 brand+0.001
 brightly+0.001
 lot+0.001
 truly+0.001
 valuable+0.001
Neg logit contributions
ases-0.001
per-0.001
pered-0.001
ws-0.001
ena-0.001
Token a
Token ID257
Feature activation-5.30e+00

Pos logit contributions
ases+0.192
per+0.166
pered+0.160
ws+0.159
ena+0.158
Neg logit contributions
 brand-0.136
 brightly-0.113
 lot-0.106
 truly-0.103
 valuable-0.098
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.100
+0.082
 brightly+0.076
 truly+0.073
 valuable+0.068
Neg logit contributions
ases-0.176
per-0.174
ena-0.163
pered-0.156
 aid-0.154
Token new
Token ID649
Feature activation-3.07e+00

Pos logit contributions
 brand+0.302
+0.296
 sudden+0.250
 truly+0.238
 brightly+0.231
Neg logit contributions
ases-0.655
per-0.576
ws-0.556
ena-0.551
old-0.546
Token game
Token ID983
Feature activation-6.74e+00

Pos logit contributions
 brand+0.082
 brightly+0.068
 lot+0.064
 truly+0.063
 valuable+0.060
Neg logit contributions
ases-0.109
per-0.094
pered-0.090
ena-0.090
ws-0.089
Token on
Token ID319
Feature activation+3.95e+00

Pos logit contributions
 brand+0.204
 brightly+0.177
 truly+0.168
 lot+0.163
 wonderful+0.156
Neg logit contributions
ases-0.219
ws-0.190
wed-0.184
per-0.178
pered-0.176
Token the
Token ID262
Feature activation-3.83e-01

Pos logit contributions
ases+0.118
per+0.098
pered+0.095
ws+0.094
wed+0.093
Neg logit contributions
 brand-0.129
 brightly-0.113
 lot-0.109
 truly-0.105
 valuable-0.101
Token screen
Token ID3159
Feature activation-6.49e+00

Pos logit contributions
 brand+0.011
 brightly+0.009
 lot+0.008
 truly+0.008
 valuable+0.008
Neg logit contributions
ases-0.013
per-0.011
pered-0.011
ena-0.011
ws-0.011
Token the
Token ID262
Feature activation-2.54e-01

Pos logit contributions
ases+0.198
per+0.161
ws+0.156
ena+0.155
pered+0.155
Neg logit contributions
 brand-0.209
 brightly-0.182
 lot-0.174
 truly-0.166
 wonderful-0.161
Token sack
Token ID23704
Feature activation-1.86e+00

Pos logit contributions
 brand+0.008
 brightly+0.007
 truly+0.007
 brilliantly+0.007
+0.007
Neg logit contributions
ases-0.006
per-0.006
pered-0.006
ws-0.005
race-0.005
Token and
Token ID290
Feature activation-2.09e+00

Pos logit contributions
 brand+0.062
 brightly+0.055
 lot+0.051
 truly+0.051
 valuable+0.049
Neg logit contributions
ases-0.056
ws-0.044
ena-0.043
wed-0.042
per-0.042
Token found
Token ID1043
Feature activation+4.50e+00

Pos logit contributions
 brand+0.069
 brightly+0.061
 truly+0.056
 lot+0.056
 wonderful+0.054
Neg logit contributions
ases-0.066
per-0.051
pered-0.051
ws-0.051
ena-0.050
Token a
Token ID257
Feature activation-5.36e+00

Pos logit contributions
ases+0.163
per+0.141
pered+0.136
ena+0.135
ws+0.134
Neg logit contributions
 brand-0.118
 brightly-0.097
 lot-0.092
 truly-0.090
 valuable-0.086
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.102
+0.093
 brightly+0.091
 brilliantly+0.085
 truly+0.085
Neg logit contributions
per-0.157
ases-0.152
 aid-0.150
pered-0.142
ena-0.140
Token new
Token ID649
Feature activation-3.22e+00

Pos logit contributions
 brand+0.301
+0.289
 sudden+0.260
 truly+0.256
 brightly+0.255
Neg logit contributions
ases-0.652
ws-0.563
per-0.563
wed-0.559
ucks-0.548
Token bright
Token ID6016
Feature activation-4.50e+00

Pos logit contributions
 brand+0.089
 brightly+0.075
 truly+0.070
 lot+0.070
 fantastic+0.066
Neg logit contributions
ases-0.107
per-0.094
pered-0.089
ena-0.087
ws-0.087
Token yellow
Token ID7872
Feature activation-2.38e+00

Pos logit contributions
 brand+0.114
 brightly+0.094
 truly+0.089
 lot+0.088
 valuable+0.084
Neg logit contributions
ases-0.157
per-0.139
pered-0.132
ws-0.130
ena-0.129
Token ball
Token ID2613
Feature activation-3.32e+00

Pos logit contributions
 brand+0.063
 brightly+0.052
 lot+0.049
 truly+0.049
 valuable+0.047
Neg logit contributions
ases-0.084
per-0.073
pered-0.070
ena-0.069
ws-0.069
Token!
Token ID0
Feature activation+7.01e-01

Pos logit contributions
 brand+0.106
 brightly+0.092
 very+0.085
 truly+0.085
 lot+0.083
Neg logit contributions
ases-0.103
ws-0.088
ena-0.085
wed-0.082
per-0.081
Token under
Token ID739
Feature activation+6.65e+00

Pos logit contributions
 brand+0.129
 brightly+0.113
 truly+0.107
 lot+0.107
 valuable+0.103
Neg logit contributions
ases-0.103
per-0.084
pered-0.081
ws-0.080
ena-0.078
Token the
Token ID262
Feature activation-1.97e-01

Pos logit contributions
ases+0.218
ws+0.182
per+0.181
pered+0.180
wed+0.179
Neg logit contributions
 brand-0.201
 brightly-0.177
 lot-0.173
 truly-0.158
 wonderful-0.152
Token counter
Token ID3753
Feature activation-1.96e+00

Pos logit contributions
 brand+0.006
 brightly+0.005
 lot+0.004
 truly+0.004
 valuable+0.004
Neg logit contributions
ases-0.007
per-0.006
pered-0.006
ena-0.005
ws-0.005
Token;
Token ID26
Feature activation-4.15e+00

Pos logit contributions
 brand+0.065
 brightly+0.055
 truly+0.054
 lot+0.052
 very+0.051
Neg logit contributions
ases-0.060
ws-0.049
pered-0.048
wed-0.047
per-0.045
Token a
Token ID257
Feature activation-5.49e+00

Pos logit contributions
 brand+0.108
 brightly+0.087
 lot+0.083
 truly+0.083
 valuable+0.079
Neg logit contributions
ases-0.142
per-0.123
ena-0.121
pered-0.117
ws-0.115
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.114
 brightly+0.095
 truly+0.095
+0.090
 brilliantly+0.087
Neg logit contributions
ases-0.173
per-0.163
ena-0.160
pered-0.148
e-0.147
Token new
Token ID649
Feature activation-3.12e+00

Pos logit contributions
 brand+0.295
+0.277
 sudden+0.250
 truly+0.246
 brightly+0.244
Neg logit contributions
ases-0.661
per-0.564
ws-0.563
ena-0.563
wed-0.555
Token doll
Token ID3654
Feature activation+1.88e-01

Pos logit contributions
 brand+0.085
 brightly+0.071
 lot+0.067
 truly+0.066
 valuable+0.063
Neg logit contributions
ases-0.110
per-0.094
pered-0.091
ena-0.090
ws-0.090
Token!
Token ID0
Feature activation+5.50e-01

Pos logit contributions
ases+0.007
ws+0.006
pered+0.006
per+0.005
wed+0.005
Neg logit contributions
 brand-0.006
 brightly-0.004
 truly-0.004
 lot-0.004
 very-0.004
Token Ab
Token ID2275
Feature activation+1.02e+01

Pos logit contributions
ases+0.017
per+0.015
pered+0.014
ws+0.014
ena+0.014
Neg logit contributions
 brand-0.017
 brightly-0.015
 lot-0.014
 truly-0.014
 valuable-0.013
Tokenbie
Token ID12590
Feature activation+1.76e+00

Pos logit contributions
ases+0.348
pered+0.304
per+0.293
race+0.280
ws+0.271
Neg logit contributions
 brand-0.270
 brightly-0.232
 truly-0.230
 lot-0.221
 wonderful-0.208
Token see
Token ID766
Feature activation+4.72e+00

Pos logit contributions
ases+0.185
per+0.153
wed+0.152
ws+0.151
pered+0.151
Neg logit contributions
 brand-0.185
 truly-0.160
 lot-0.159
 brightly-0.155
 wonderful-0.148
Token that
Token ID326
Feature activation-2.37e+00

Pos logit contributions
ases+0.161
per+0.138
pered+0.134
ws+0.131
wed+0.130
Neg logit contributions
 brand-0.138
 brightly-0.115
 lot-0.110
 truly-0.107
 valuable-0.103
Token it
Token ID340
Feature activation-7.93e+00

Pos logit contributions
 brand+0.061
 brightly+0.050
 lot+0.047
 truly+0.046
 valuable+0.044
Neg logit contributions
ases-0.089
per-0.077
pered-0.075
ws-0.073
ena-0.073
Token was
Token ID373
Feature activation-4.53e+00

Pos logit contributions
 brand+0.191
 brightly+0.156
 lot+0.146
 truly+0.140
 valuable+0.135
Neg logit contributions
ases-0.302
per-0.265
ena-0.254
pered-0.254
ws-0.252
Token a
Token ID257
Feature activation-5.35e+00

Pos logit contributions
 brand+0.112
 brightly+0.090
 lot+0.084
 truly+0.083
 valuable+0.079
Neg logit contributions
ases-0.178
per-0.154
pered-0.148
ws-0.147
ena-0.146
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.109
 brightly+0.097
+0.092
 brilliantly+0.086
 truly+0.085
Neg logit contributions
ases-0.156
per-0.150
ena-0.148
 aid-0.144
 ear-0.140
Token new
Token ID649
Feature activation-3.09e+00

Pos logit contributions
 brand+0.291
+0.279
 sudden+0.268
 truly+0.251
 brightly+0.246
Neg logit contributions
ases-0.652
per-0.566
ws-0.555
rap-0.550
ena-0.546
Token puzzle
Token ID15027
Feature activation-7.53e+00

Pos logit contributions
 brand+0.076
 brightly+0.062
 lot+0.059
 truly+0.057
 valuable+0.054
Neg logit contributions
ases-0.117
per-0.102
pered-0.098
ena-0.097
ws-0.097
Token!
Token ID0
Feature activation+1.24e+00

Pos logit contributions
 brand+0.265
 brightly+0.229
 truly+0.220
 lot+0.218
 valuable+0.213
Neg logit contributions
ases-0.213
pered-0.169
per-0.169
ws-0.168
wed-0.164
Token He
Token ID679
Feature activation-7.88e-01

Pos logit contributions
ases+0.044
per+0.038
pered+0.036
ena+0.036
ws+0.036
Neg logit contributions
 brand-0.034
 brightly-0.028
 lot-0.026
 truly-0.026
 valuable-0.025
Token put
Token ID1234
Feature activation+5.80e+00

Pos logit contributions
 brand+0.023
 brightly+0.019
 truly+0.019
 lot+0.019
 valuable+0.017
Neg logit contributions
ases-0.027
per-0.023
pered-0.023
ws-0.022
ena-0.021
Token joy
Token ID8716
Feature activation-3.12e+00

Pos logit contributions
ases+0.093
per+0.080
pered+0.077
ena+0.077
ws+0.076
Neg logit contributions
 brand-0.075
 brightly-0.062
 lot-0.059
 truly-0.057
 valuable-0.055
Token -
Token ID532
Feature activation-4.03e+00

Pos logit contributions
 brand+0.114
 brightly+0.102
 lot+0.094
 truly+0.094
 valuable+0.090
Neg logit contributions
ases-0.084
per-0.067
pered-0.065
ws-0.065
ena-0.062
Token it
Token ID340
Feature activation-8.02e+00

Pos logit contributions
 brand+0.102
 brightly+0.084
 lot+0.078
 truly+0.075
 valuable+0.072
Neg logit contributions
ases-0.153
per-0.134
ws-0.129
pered-0.129
wed-0.128
Token was
Token ID373
Feature activation-4.57e+00

Pos logit contributions
 brand+0.211
 brightly+0.173
 lot+0.167
 valuable+0.159
+0.157
Neg logit contributions
ases-0.273
per-0.241
ena-0.235
pered-0.225
wed-0.224
Token a
Token ID257
Feature activation-5.55e+00

Pos logit contributions
 brand+0.126
 brightly+0.105
 lot+0.099
 truly+0.098
 valuable+0.093
Neg logit contributions
ases-0.161
per-0.138
pered-0.133
ws-0.132
ena-0.131
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.123
 brightly+0.105
 truly+0.104
+0.102
 brilliantly+0.098
Neg logit contributions
ases-0.163
ena-0.148
per-0.148
 aid-0.143
 ear-0.139
Token new
Token ID649
Feature activation-3.18e+00

Pos logit contributions
+0.341
 sudden+0.298
 brand+0.297
 truly+0.280
Imagine+0.279
Neg logit contributions
ases-0.572
ucks-0.490
ena-0.475
ws-0.474
per-0.473
Token doll
Token ID3654
Feature activation+1.58e-01

Pos logit contributions
 brand+0.089
 brightly+0.075
 truly+0.071
 lot+0.070
 valuable+0.067
Neg logit contributions
ases-0.105
per-0.088
ena-0.087
pered-0.084
ws-0.084
Token!
Token ID0
Feature activation+7.55e-01

Pos logit contributions
ases+0.006
ws+0.005
wed+0.005
pered+0.005
per+0.005
Neg logit contributions
 brand-0.004
 brightly-0.004
 truly-0.003
 lot-0.003
 very-0.003
Token All
Token ID1439
Feature activation-5.88e+00

Pos logit contributions
ases+0.024
per+0.021
pered+0.020
ws+0.019
ena+0.019
Neg logit contributions
 brand-0.023
 brightly-0.020
 lot-0.019
 truly-0.018
 valuable-0.018
Token of
Token ID286
Feature activation+5.05e+00

Pos logit contributions
 brand+0.205
 brightly+0.179
 lot+0.171
 truly+0.168
 valuable+0.163
Neg logit contributions
ases-0.163
per-0.136
pered-0.129
ws-0.127
ena-0.126
Token happy
Token ID3772
Feature activation-5.26e+00

Pos logit contributions
 brand+0.336
 brightly+0.277
 lot+0.269
 truly+0.250
+0.243
Neg logit contributions
ases-0.460
per-0.395
pered-0.388
ena-0.385
ws-0.371
Token!
Token ID0
Feature activation+9.04e-01

Pos logit contributions
 brand+0.168
 brightly+0.146
 truly+0.136
 lot+0.136
 valuable+0.129
Neg logit contributions
ases-0.166
per-0.138
ws-0.136
pered-0.134
wed-0.132
Token Inside
Token ID14384
Feature activation-7.00e-01

Pos logit contributions
ases+0.032
per+0.028
ws+0.027
wed+0.027
ena+0.026
Neg logit contributions
 brand-0.025
 brightly-0.021
 lot-0.020
 truly-0.019
 valuable-0.019
Token was
Token ID373
Feature activation-4.72e+00

Pos logit contributions
 brand+0.019
 brightly+0.016
 lot+0.015
 truly+0.015
 valuable+0.014
Neg logit contributions
ases-0.025
per-0.021
pered-0.021
ws-0.020
ena-0.020
Token a
Token ID257
Feature activation-5.59e+00

Pos logit contributions
 brand+0.128
 brightly+0.107
 lot+0.102
 truly+0.099
 valuable+0.096
Neg logit contributions
ases-0.161
per-0.138
ena-0.136
pered-0.134
ws-0.130
Token brand
Token ID4508
Feature activation-1.76e+01

Pos logit contributions
 brand+0.094
 brightly+0.081
+0.081
 brilliantly+0.070
 truly+0.070
Neg logit contributions
ases-0.177
ena-0.173
per-0.169
 aid-0.168
 ear-0.162
Token new
Token ID649
Feature activation-3.16e+00

Pos logit contributions
 brand+0.295
+0.261
 sudden+0.242
 truly+0.240
 brightly+0.237
Neg logit contributions
ases-0.673
ws-0.576
per-0.574
ena-0.566
wed-0.565
Token toy
Token ID13373
Feature activation-3.84e+00

Pos logit contributions
 brand+0.088
 brightly+0.074
 lot+0.070
 truly+0.069
 valuable+0.066
Neg logit contributions
ases-0.107
per-0.092
pered-0.089
ena-0.088
ws-0.087
Token that
Token ID326
Feature activation-2.62e+00

Pos logit contributions
 brand+0.111
 brightly+0.093
 lot+0.087
 truly+0.086
 valuable+0.080
Neg logit contributions
ases-0.133
ws-0.114
wed-0.112
per-0.112
pered-0.110
Token she
Token ID673
Feature activation-3.04e-01

Pos logit contributions
 brand+0.070
 brightly+0.056
 lot+0.055
 truly+0.051
+0.050
Neg logit contributions
ases-0.090
per-0.077
ena-0.077
wed-0.074
pered-0.073
Token could
Token ID714
Feature activation-3.00e+00

Pos logit contributions
 brand+0.007
 truly+0.006
 brightly+0.006
 lot+0.005
 valuable+0.005
Neg logit contributions
ases-0.012
ws-0.010
per-0.010
pered-0.010
ena-0.010
Token did
Token ID750
Feature activation-3.82e+00

Pos logit contributions
ases+0.358
ags+0.310
ws+0.303
pered+0.299
ucks+0.297
Neg logit contributions
 brand-0.159
 very-0.148
 truly-0.137
 seemed-0.129
 really-0.127
Token not
Token ID407
Feature activation-2.30e+00

Pos logit contributions
 brand+0.086
 brightly+0.068
 lot+0.065
 truly+0.063
 valuable+0.059
Neg logit contributions
ases-0.155
per-0.136
pered-0.131
ws-0.131
wed-0.130
Token want
Token ID765
Feature activation+2.63e+00

Pos logit contributions
 truly+0.058
 brand+0.057
 very+0.056
 fun+0.055
 lot+0.055
Neg logit contributions
ases-0.075
ucks-0.063
pered-0.062
wed-0.060
ws-0.059
Token Tom
Token ID4186
Feature activation+9.19e+00

Pos logit contributions
ases+0.089
per+0.077
pered+0.074
ws+0.073
ena+0.072
Neg logit contributions
 brand-0.076
 brightly-0.064
 lot-0.061
 truly-0.060
 valuable-0.058
Token to
Token ID284
Feature activation+6.24e+00

Pos logit contributions
ases+0.343
per+0.298
pered+0.293
ws+0.289
ena+0.279
Neg logit contributions
 brand-0.223
 truly-0.179
 very-0.177
 lot-0.174
 valuable-0.172
Token run
Token ID1057
Feature activation+8.76e+00

Pos logit contributions
ases+0.204
ws+0.170
pered+0.166
ucks+0.165
ena+0.163
Neg logit contributions
 brand-0.172
 lot-0.167
 truly-0.161
 fun-0.157
 very-0.156
Token with
Token ID351
Feature activation+2.87e+00

Pos logit contributions
ases+0.267
ena+0.221
ws+0.219
per+0.218
wed+0.211
Neg logit contributions
 brand-0.276
 lot-0.241
 brightly-0.240
 very-0.224
 truly-0.222
Token her
Token ID607
Feature activation+3.95e+00

Pos logit contributions
ases+0.081
per+0.071
ws+0.068
wed+0.066
pered+0.065
Neg logit contributions
 brand-0.096
 lot-0.084
 brightly-0.083
 fun-0.077
 wonderful-0.077
Token basket
Token ID7988
Feature activation-3.33e+00

Pos logit contributions
ases+0.122
per+0.099
ws+0.098
ena+0.097
pered+0.096
Neg logit contributions
 brand-0.127
 brightly-0.109
 lot-0.107
 truly-0.102
 valuable-0.099
Token.
Token ID13
Feature activation+1.51e+00

Pos logit contributions
 brand+0.106
 brightly+0.092
 very+0.088
 truly+0.087
 lot+0.087
Neg logit contributions
ases-0.104
ws-0.086
ena-0.082
per-0.081
pered-0.080
Token She
Token ID1375
Feature activation+7.82e-01

Pos logit contributions
ases+0.053
per+0.047
ws+0.045
wed+0.045
pered+0.044
Neg logit contributions
 brand-0.040
 brightly-0.033
 lot-0.032
 truly-0.030
-0.030
Token with
Token ID351
Feature activation+4.33e+00

Pos logit contributions
 brand+0.008
 brightly+0.007
 truly+0.006
 lot+0.006
 most+0.006
Neg logit contributions
ases-0.008
per-0.007
pered-0.007
ws-0.007
ena-0.007
Token a
Token ID257
Feature activation-4.07e+00

Pos logit contributions
ases+0.134
per+0.117
pered+0.111
ws+0.109
ena+0.109
Neg logit contributions
 brand-0.137
 brightly-0.118
 lot-0.113
 truly-0.107
 wonderful-0.105
Token dent
Token ID18794
Feature activation-3.26e+00

Pos logit contributions
 brand+0.093
 brightly+0.077
 truly+0.072
+0.071
 brilliantly+0.066
Neg logit contributions
ases-0.141
per-0.126
pered-0.121
ena-0.120
ws-0.118
Token.
Token ID13
Feature activation+2.43e+00

Pos logit contributions
 brand+0.097
 brightly+0.086
 very+0.081
 lot+0.079
 truly+0.075
Neg logit contributions
ases-0.102
pered-0.084
race-0.083
per-0.082
ws-0.081
Token She
Token ID1375
Feature activation+2.41e+00

Pos logit contributions
ases+0.091
wed+0.080
ws+0.078
per+0.078
pered+0.074
Neg logit contributions
 brand-0.060
 brightly-0.051
 lot-0.047
 truly-0.045
-0.044
Token says
Token ID1139
Feature activation+7.99e+00

Pos logit contributions
ases+0.080
pered+0.071
per+0.067
race+0.065
adies+0.061
Neg logit contributions
 brand-0.070
 lot-0.058
 brightly-0.057
 truly-0.056
 very-0.055
Token,
Token ID11
Feature activation-1.28e+00

Pos logit contributions
ases+0.292
per+0.252
ws+0.244
pered+0.243
wed+0.240
Neg logit contributions
 brand-0.211
 brightly-0.178
 lot-0.167
 truly-0.160
 valuable-0.155
Token "
Token ID366
Feature activation-1.06e+00

Pos logit contributions
 brand+0.028
 brightly+0.022
 lot+0.021
 truly+0.019
 valuable+0.018
Neg logit contributions
ases-0.054
wed-0.046
per-0.046
ws-0.046
pered-0.044
TokenWhat
Token ID2061
Feature activation-3.29e+00

Pos logit contributions
 brand+0.026
 brightly+0.022
 lot+0.020
 truly+0.020
 valuable+0.019
Neg logit contributions
ases-0.041
per-0.035
pered-0.034
ena-0.034
ws-0.034
Token happened
Token ID3022
Feature activation+4.17e+00

Pos logit contributions
 brand+0.095
 brightly+0.080
 lot+0.076
 truly+0.074
 valuable+0.071
Neg logit contributions
ases-0.111
per-0.095
pered-0.091
ena-0.090
ws-0.090
Token?
Token ID30
Feature activation+2.47e+00

Pos logit contributions
ases+0.153
wed+0.134
ws+0.132
per+0.132
pered+0.129
Neg logit contributions
 brand-0.109
 brightly-0.090
 truly-0.084
 very-0.083
 lot-0.083
Token He
Token ID679
Feature activation+2.22e-02

Pos logit contributions
ases+0.084
per+0.072
wed+0.072
ws+0.072
pered+0.068
Neg logit contributions
 brand-0.050
 brightly-0.040
 lot-0.039
 truly-0.036
-0.035
Token was
Token ID373
Feature activation-3.99e+00

Pos logit contributions
ases+0.001
per+0.001
pered+0.001
ws+0.001
ena+0.001
Neg logit contributions
 brand-0.001
 brightly-0.001
 lot-0.001
 truly-0.000
 valuable-0.000
Token so
Token ID523
Feature activation-1.24e+01

Pos logit contributions
 brand+0.077
 brightly+0.058
 lot+0.053
 truly+0.053
 valuable+0.048
Neg logit contributions
ases-0.175
per-0.154
ws-0.149
pered-0.149
ena-0.148
Token happy
Token ID3772
Feature activation-4.90e+00

Pos logit contributions
 brand+0.300
 brightly+0.243
 lot+0.230
 truly+0.220
 valuable+0.210
Neg logit contributions
ases-0.474
per-0.412
pered-0.401
ena-0.399
ws-0.392
Token that
Token ID326
Feature activation-2.06e+00

Pos logit contributions
 brand+0.128
 brightly+0.108
 lot+0.098
 truly+0.098
 very+0.091
Neg logit contributions
ases-0.183
ws-0.156
per-0.155
wed-0.153
pered-0.153
Token Sara
Token ID24799
Feature activation+8.90e+00

Pos logit contributions
 brand+0.043
 brightly+0.035
 lot+0.032
 truly+0.032
 valuable+0.030
Neg logit contributions
ases-0.087
per-0.077
pered-0.075
ws-0.075
wed-0.073
Token was
Token ID373
Feature activation-4.31e+00

Pos logit contributions
ases+0.391
pered+0.329
ws+0.324
race+0.320
ucks+0.311
Neg logit contributions
 brand-0.176
 truly-0.153
 brightly-0.139
 very-0.132
 wonderful-0.126
Token safe
Token ID3338
Feature activation-2.69e+00

Pos logit contributions
 brand+0.103
 truly+0.082
 brightly+0.080
 very+0.075
 wonderful+0.073
Neg logit contributions
ases-0.177
ws-0.152
per-0.146
wed-0.145
ena-0.142
Token.
Token ID13
Feature activation+1.95e+00

Pos logit contributions
 brand+0.087
 brightly+0.078
 truly+0.073
 lot+0.071
 very+0.070
Neg logit contributions
ases-0.084
ws-0.069
wed-0.066
per-0.065
pered-0.064
Token
Token ID198
Feature activation-2.95e+00

Pos logit contributions
ases+0.075
per+0.064
wed+0.063
ws+0.063
pered+0.061
Neg logit contributions
 brand-0.048
 brightly-0.039
 lot-0.038
 truly-0.036
 valuable-0.035
Token
Token ID198
Feature activation-3.10e+00

Pos logit contributions
 brand+0.049
 lot+0.040
 much+0.035
 brightly+0.034
 truly+0.034
Neg logit contributions
ases-0.133
per-0.118
wed-0.115
ws-0.114
pered-0.112
Token One
Token ID1881
Feature activation-2.43e+00

Pos logit contributions
ases+0.051
per+0.043
pered+0.041
ena+0.041
ws+0.041
Neg logit contributions
 brand-0.045
 brightly-0.039
 lot-0.036
 truly-0.036
 valuable-0.035
Token day
Token ID1110
Feature activation-7.88e+00

Pos logit contributions
 brand+0.057
 brightly+0.045
 truly+0.041
 lot+0.040
+0.039
Neg logit contributions
ases-0.093
ena-0.082
per-0.081
wed-0.079
ws-0.077
Token,
Token ID11
Feature activation-2.78e+00

Pos logit contributions
 brand+0.293
 truly+0.256
 brightly+0.250
 valuable+0.244
 fantastic+0.236
Neg logit contributions
ases-0.192
pered-0.142
per-0.141
wed-0.139
ena-0.138
Token Tim
Token ID5045
Feature activation+9.11e+00

Pos logit contributions
 brand+0.120
 valuable+0.112
 truly+0.111
 lifetime+0.107
 marvelous+0.106
Neg logit contributions
ases-0.038
 Tom-0.030
 Jen-0.026
 Sam-0.024
ena-0.024
Tokenmy
Token ID1820
Feature activation+4.26e+00

Pos logit contributions
ases+0.286
pered+0.244
per+0.239
ws+0.213
ps+0.210
Neg logit contributions
 brand-0.275
 lot-0.232
 very-0.230
 truly-0.229
 brightly-0.223
Token went
Token ID1816
Feature activation+7.64e+00

Pos logit contributions
ases+0.165
pered+0.143
per+0.136
ws+0.135
ps+0.133
Neg logit contributions
 brand-0.106
 lot-0.085
 brightly-0.084
 truly-0.082
 very-0.081
Token to
Token ID284
Feature activation+6.09e+00

Pos logit contributions
ases+0.267
per+0.228
ena+0.219
ws+0.218
pered+0.218
Neg logit contributions
 brand-0.217
 brightly-0.175
 lot-0.172
 truly-0.164
 valuable-0.160
Token the
Token ID262
Feature activation+1.87e-01

Pos logit contributions
ases+0.205
per+0.176
pered+0.175
wed+0.170
ws+0.164
Neg logit contributions
 brand-0.183
 lot-0.156
 brightly-0.143
 truly-0.143
 wonderful-0.141
Token park
Token ID3952
Feature activation-4.32e+00

Pos logit contributions
ases+0.005
per+0.004
ena+0.004
race+0.003
ws+0.003
Neg logit contributions
 brand-0.006
 brightly-0.006
 truly-0.005
 valuable-0.005
 brilliantly-0.005
Token to
Token ID284
Feature activation+6.09e+00

Pos logit contributions
 brand+0.114
 brightly+0.093
 lot+0.091
 very+0.089
 truly+0.088
Neg logit contributions
ases-0.156
ws-0.134
per-0.132
pered-0.131
wed-0.129
Token draw
Token ID3197
Feature activation+5.20e+00

Pos logit contributions
ases+0.183
per+0.156
pered+0.152
ena+0.146
wed+0.144
Neg logit contributions
 brand-0.203
 lot-0.169
 truly-0.165
 brightly-0.164
 wonderful-0.158
Token long
Token ID890
Feature activation-4.84e-01

Pos logit contributions
ases+0.109
per+0.094
wed+0.087
pered+0.086
ena+0.086
Neg logit contributions
 brand-0.106
 brightly-0.098
 brilliantly-0.088
 truly-0.088
-0.086
Token bl
Token ID698
Feature activation-2.91e+00

Pos logit contributions
 brand+0.014
 brightly+0.012
 lot+0.012
 truly+0.011
 valuable+0.011
Neg logit contributions
ases-0.016
per-0.013
pered-0.013
ena-0.013
ws-0.013
Tokenouse
Token ID1076
Feature activation-5.60e+00

Pos logit contributions
 brand+0.077
 brightly+0.065
 lot+0.061
 truly+0.060
 valuable+0.057
Neg logit contributions
ases-0.103
per-0.090
pered-0.086
ena-0.085
ws-0.084
Token back
Token ID736
Feature activation+3.66e+00

Pos logit contributions
 brand+0.178
 brightly+0.145
 truly+0.142
 lot+0.136
 valuable+0.133
Neg logit contributions
ases-0.184
ws-0.151
per-0.149
pered-0.145
ena-0.143
Token and
Token ID290
Feature activation-1.76e+00

Pos logit contributions
ases+0.116
ws+0.094
per+0.094
pered+0.092
ena+0.091
Neg logit contributions
 brand-0.119
 brightly-0.100
 truly-0.097
 lot-0.094
 very-0.091
Token thanked
Token ID26280
Feature activation+8.58e+00

Pos logit contributions
 brand+0.055
 brightly+0.045
 lot+0.043
 truly+0.043
 valuable+0.041
Neg logit contributions
ases-0.058
per-0.049
pered-0.047
ena-0.046
ws-0.046
Token her
Token ID607
Feature activation+3.73e+00

Pos logit contributions
ases+0.249
per+0.206
pered+0.203
ws+0.199
wed+0.198
Neg logit contributions
 brand-0.294
 brightly-0.262
 lot-0.254
 truly-0.242
 wonderful-0.234
Token friend
Token ID1545
Feature activation+3.89e+00

Pos logit contributions
ases+0.140
ws+0.116
per+0.115
wed+0.115
pered+0.111
Neg logit contributions
 brand-0.103
 lot-0.086
 brightly-0.083
 wonderful-0.082
 valuable-0.082
Token for
Token ID329
Feature activation+5.14e+00

Pos logit contributions
ases+0.141
ws+0.116
pered+0.115
per+0.113
wed+0.112
Neg logit contributions
 brand-0.109
 very-0.094
 lot-0.089
 wonderful-0.086
 truly-0.085
Token helping
Token ID5742
Feature activation+4.88e+00

Pos logit contributions
ases+0.158
per+0.130
ws+0.125
pered+0.124
ena+0.124
Neg logit contributions
 brand-0.169
 brightly-0.142
 lot-0.137
 truly-0.136
 wonderful-0.130
Token her
Token ID607
Feature activation+3.70e+00

Pos logit contributions
ases+0.166
ws+0.135
pered+0.134
aled+0.131
wed+0.128
Neg logit contributions
 brand-0.138
 lot-0.127
 very-0.124
 brightly-0.123
 truly-0.115
Token have
Token ID423
Feature activation+1.52e+00

Pos logit contributions
ases+0.081
per+0.066
pered+0.063
ena+0.062
ws+0.061
Neg logit contributions
 brand-0.107
 brightly-0.094
 lot-0.090
 truly-0.089
 valuable-0.086
Token to
Token ID284
Feature activation+6.30e+00

Pos logit contributions
ases+0.047
per+0.039
ena+0.037
wed+0.037
ws+0.037
Neg logit contributions
 brand-0.049
 brightly-0.042
 lot-0.039
 truly-0.039
 fantastic-0.037
Token ask
Token ID1265
Feature activation+1.17e+01

Pos logit contributions
ases+0.197
pered+0.174
adies+0.161
per+0.158
ena+0.155
Neg logit contributions
 brand-0.171
 lot-0.170
 fun-0.160
 wonderful-0.156
 truly-0.152
Token before
Token ID878
Feature activation-2.24e+00

Pos logit contributions
ases+0.412
pered+0.341
per+0.336
race+0.331
ws+0.325
Neg logit contributions
 brand-0.313
 brightly-0.272
 lot-0.270
 very-0.257
 wonderful-0.246
Token you
Token ID345
Feature activation+3.06e+00

Pos logit contributions
 brand+0.054
 brightly+0.045
 lot+0.044
 truly+0.040
 valuable+0.039
Neg logit contributions
ases-0.085
per-0.074
pered-0.073
ws-0.072
ena-0.072
Token use
Token ID779
Feature activation+6.08e+00

Pos logit contributions
ases+0.086
pered+0.074
per+0.072
ena+0.070
ws+0.065
Neg logit contributions
 brand-0.101
 lot-0.088
 brightly-0.088
 truly-0.085
 very-0.080
Token them
Token ID606
Feature activation+7.32e-01

Pos logit contributions
ases+0.179
per+0.148
ws+0.141
pered+0.139
ena+0.139
Neg logit contributions
 brand-0.198
 brightly-0.173
 truly-0.164
 lot-0.161
 valuable-0.155
Token.
Token ID13
Feature activation+1.88e+00

Pos logit contributions
ases+0.022
per+0.018
pered+0.018
ena+0.017
ws+0.017
Neg logit contributions
 brand-0.024
 brightly-0.020
 lot-0.020
 truly-0.019
 valuable-0.019
Token Now
Token ID2735
Feature activation-8.41e-01

Pos logit contributions
ases+0.062
per+0.052
wed+0.052
ws+0.051
pered+0.050
Neg logit contributions
 brand-0.053
 brightly-0.045
 lot-0.043
 truly-0.042
 valuable-0.041
Token let
Token ID1309
Feature activation+8.61e+00

Pos logit contributions
 brand+0.020
 brightly+0.017
 lot+0.017
 truly+0.015
 wonderful+0.015
Neg logit contributions
ases-0.033
pered-0.028
per-0.027
ena-0.027
ws-0.027
Token's
Token ID338
Feature activation-5.11e-01

Pos logit contributions
ases+0.302
per+0.257
pered+0.257
ws+0.243
ena+0.240
Neg logit contributions
 brand-0.233
 lot-0.202
 brightly-0.195
 truly-0.187
 wonderful-0.185
Token come
Token ID1282
Feature activation+5.99e+00

Pos logit contributions
ases+0.079
per+0.065
pered+0.065
ws+0.062
ena+0.062
Neg logit contributions
 brand-0.092
 lot-0.082
 brightly-0.080
 truly-0.080
-0.074
Token in
Token ID287
Feature activation+5.00e+00

Pos logit contributions
ases+0.205
wed+0.178
ws+0.176
per+0.173
ena+0.167
Neg logit contributions
 brand-0.171
 brightly-0.142
 very-0.140
 truly-0.133
 lot-0.133
Token and
Token ID290
Feature activation-1.31e+00

Pos logit contributions
ases+0.165
pered+0.141
wed+0.138
per+0.137
ws+0.136
Neg logit contributions
 brand-0.150
 brightly-0.128
 lot-0.123
 truly-0.118
 very-0.117
Token play
Token ID711
Feature activation+4.06e+00

Pos logit contributions
 brand+0.048
 brightly+0.042
 lot+0.041
 truly+0.040
 wonderful+0.038
Neg logit contributions
ases-0.035
per-0.028
pered-0.027
ena-0.026
ws-0.026
Token with
Token ID351
Feature activation+3.40e+00

Pos logit contributions
ases+0.129
pered+0.109
wed+0.109
per+0.108
ws+0.108
Neg logit contributions
 brand-0.124
 brightly-0.110
 lot-0.103
 truly-0.097
 very-0.095
Token you
Token ID345
Feature activation+2.90e+00

Pos logit contributions
ases+0.133
pered+0.119
per+0.117
ws+0.115
wed+0.114
Neg logit contributions
 brand-0.078
 brightly-0.066
 lot-0.062
 truly-0.056
 wonderful-0.055
Token?"
Token ID1701
Feature activation+2.09e+00

Pos logit contributions
ases+0.083
per+0.068
pered+0.065
ena+0.064
ws+0.064
Neg logit contributions
 brand-0.099
 brightly-0.086
 lot-0.082
 truly-0.081
 valuable-0.078
Token
Token ID198
Feature activation-2.97e+00

Pos logit contributions
ases+0.078
wed+0.070
ws+0.069
per+0.068
pered+0.066
Neg logit contributions
 brand-0.050
 lot-0.044
 brightly-0.042
-0.040
 very-0.039
Token
Token ID198
Feature activation-3.06e+00

Pos logit contributions
 brand+0.052
 lot+0.043
 brightly+0.041
 truly+0.035
 wonderful+0.033
Neg logit contributions
ases-0.131
per-0.116
wed-0.116
ws-0.115
pered-0.113
TokenRex
Token ID47389
Feature activation+6.70e+00

Pos logit contributions
 brand+0.074
 brightly+0.064
 lot+0.063
 truly+0.057
 sudden+0.055
Neg logit contributions
ases-0.114
ws-0.100
wed-0.098
per-0.097
pered-0.096
Token w
Token ID266
Feature activation+3.13e+00

Pos logit contributions
ases+0.255
per+0.214
pered+0.211
ws+0.210
wed+0.205
Neg logit contributions
 brand-0.169
 brightly-0.139
 lot-0.131
 truly-0.130
 wonderful-0.121
Token colourful
Token ID44799
Feature activation-3.33e+00

Pos logit contributions
 brand+0.297
+0.242
 brightly+0.220
 lot+0.216
 truly+0.207
Neg logit contributions
ases-0.410
per-0.367
pered-0.361
ena-0.348
ws-0.335
Token,
Token ID11
Feature activation-3.02e+00

Pos logit contributions
 brand+0.134
 brightly+0.117
 lot+0.115
 truly+0.114
 valuable+0.112
Neg logit contributions
ases-0.067
per-0.057
pered-0.050
ena-0.048
ws-0.043
Token with
Token ID351
Feature activation+2.82e+00

Pos logit contributions
 brand+0.100
 brightly+0.085
 lot+0.084
+0.082
 truly+0.081
Neg logit contributions
ases-0.080
per-0.069
pered-0.065
ena-0.065
ws-0.061
Token many
Token ID867
Feature activation-3.25e+00

Pos logit contributions
ases+0.105
per+0.089
pered+0.088
ena+0.087
ws+0.084
Neg logit contributions
 brand-0.068
-0.054
 brightly-0.053
 truly-0.051
 lot-0.049
Token buildings
Token ID6832
Feature activation-8.17e-01

Pos logit contributions
 brand+0.088
 brightly+0.071
 truly+0.068
+0.068
 lot+0.065
Neg logit contributions
ases-0.112
per-0.095
pered-0.093
ena-0.090
ws-0.090
Token painted
Token ID13055
Feature activation+1.87e+00

Pos logit contributions
 brand+0.029
 brightly+0.025
 lot+0.024
 truly+0.024
 valuable+0.023
Neg logit contributions
ases-0.022
per-0.018
pered-0.017
ena-0.017
ws-0.017
Token in
Token ID287
Feature activation+4.78e+00

Pos logit contributions
ases+0.060
per+0.050
ws+0.049
pered+0.048
ena+0.048
Neg logit contributions
 brand-0.058
 brightly-0.050
 lot-0.047
 truly-0.046
 valuable-0.044
Token bright
Token ID6016
Feature activation-4.32e+00

Pos logit contributions
ases+0.166
per+0.143
pered+0.137
ena+0.136
ws+0.135
Neg logit contributions
 brand-0.132
 brightly-0.110
 lot-0.103
 truly-0.103
 valuable-0.098
Token colours
Token ID18915
Feature activation-7.09e+00

Pos logit contributions
 brand+0.116
 brightly+0.095
 truly+0.090
 lot+0.090
 valuable+0.086
Neg logit contributions
ases-0.153
per-0.131
pered-0.126
ena-0.124
ws-0.124
Token.
Token ID13
Feature activation+1.39e+00

Pos logit contributions
 brand+0.266
 brightly+0.235
 lot+0.225
 truly+0.222
 valuable+0.216
Neg logit contributions
ases-0.176
per-0.141
pered-0.133
ena-0.132
ws-0.131
Token
Token ID198
Feature activation-3.44e+00

Pos logit contributions
ases+0.047
per+0.040
pered+0.039
ena+0.038
ws+0.038
Neg logit contributions
 brand-0.040
 brightly-0.034
 lot-0.031
 truly-0.031
 valuable-0.030
Token,
Token ID11
Feature activation-2.80e+00

Pos logit contributions
 brand+0.287
 brightly+0.246
 truly+0.245
 valuable+0.234
 fantastic+0.228
Neg logit contributions
ases-0.207
per-0.162
ena-0.159
pered-0.154
e-0.148
Token Lily
Token ID20037
Feature activation+7.56e+00

Pos logit contributions
 brand+0.109
 brightly+0.095
 truly+0.095
 valuable+0.094
 fantastic+0.091
Neg logit contributions
ases-0.057
ena-0.047
per-0.045
pered-0.040
ily-0.040
Token and
Token ID290
Feature activation-1.67e+00

Pos logit contributions
ases+0.283
pered+0.240
ws+0.236
per+0.232
ps+0.225
Neg logit contributions
 brand-0.187
 brightly-0.155
 lot-0.154
 truly-0.152
 very-0.149
Token Rose
Token ID8049
Feature activation+6.19e+00

Pos logit contributions
 brand+0.063
 brightly+0.056
 lot+0.053
 truly+0.053
 valuable+0.051
Neg logit contributions
ases-0.041
per-0.033
ena-0.031
pered-0.031
ws-0.030
Token went
Token ID1816
Feature activation+7.54e+00

Pos logit contributions
ases+0.200
per+0.167
pered+0.163
ws+0.160
ena+0.158
Neg logit contributions
 brand-0.187
 brightly-0.162
 lot-0.157
 truly-0.153
 wonderful-0.145
Token to
Token ID284
Feature activation+6.06e+00

Pos logit contributions
ases+0.260
per+0.225
ena+0.217
ws+0.214
wed+0.213
Neg logit contributions
 brand-0.211
 lot-0.173
 brightly-0.173
 truly-0.163
 valuable-0.158
Token pick
Token ID2298
Feature activation+6.54e+00

Pos logit contributions
ases+0.179
per+0.151
pered+0.145
ws+0.141
wed+0.141
Neg logit contributions
 brand-0.201
 brightly-0.171
 lot-0.170
 truly-0.164
 wonderful-0.159
Token some
Token ID617
Feature activation-3.08e+00

Pos logit contributions
ases+0.217
per+0.183
ena+0.175
pered+0.174
ws+0.174
Neg logit contributions
 brand-0.193
 brightly-0.162
 truly-0.153
 lot-0.152
 valuable-0.145
Token berries
Token ID36322
Feature activation-2.57e+00

Pos logit contributions
 brand+0.077
 brightly+0.063
 truly+0.059
 lot+0.058
 valuable+0.055
Neg logit contributions
ases-0.114
per-0.099
pered-0.095
ena-0.095
ws-0.094
Token in
Token ID287
Feature activation+4.93e+00

Pos logit contributions
 brand+0.100
 brightly+0.088
 lot+0.084
 truly+0.083
 valuable+0.081
Neg logit contributions
ases-0.061
per-0.049
pered-0.046
ws-0.045
ena-0.045
Token the
Token ID262
Feature activation-2.91e-02

Pos logit contributions
ases+0.139
per+0.115
pered+0.110
ws+0.108
ena+0.107
Neg logit contributions
 brand-0.170
 brightly-0.148
 lot-0.143
 truly-0.139
 valuable-0.135
TokenSuddenly
Token ID38582
Feature activation-3.00e+00

Pos logit contributions
 brand+0.070
 brightly+0.059
 lot+0.057
 truly+0.056
 valuable+0.052
Neg logit contributions
ases-0.150
per-0.131
ws-0.131
wed-0.127
pered-0.127
Token,
Token ID11
Feature activation-3.19e+00

Pos logit contributions
 brand+0.077
 brightly+0.068
 lot+0.064
 truly+0.060
 wonderful+0.058
Neg logit contributions
ases-0.112
per-0.094
ws-0.094
wed-0.092
pered-0.092
Token she
Token ID673
Feature activation+6.46e-02

Pos logit contributions
 brand+0.090
 brightly+0.077
 lot+0.073
 truly+0.071
 valuable+0.067
Neg logit contributions
ases-0.110
per-0.094
pered-0.090
ws-0.090
wed-0.089
Token heard
Token ID2982
Feature activation+3.57e+00

Pos logit contributions
ases+0.002
ws+0.002
pered+0.002
per+0.002
wed+0.002
Neg logit contributions
 brand-0.002
 brightly-0.002
 truly-0.002
 lot-0.002
 sudden-0.001
Token laughter
Token ID20263
Feature activation-2.85e+00

Pos logit contributions
ases+0.138
per+0.120
ws+0.116
pered+0.115
wed+0.114
Neg logit contributions
 brand-0.088
 brightly-0.073
 lot-0.067
 truly-0.065
 wonderful-0.062
Token coming
Token ID2406
Feature activation+2.78e+00

Pos logit contributions
 brand+0.099
 brightly+0.088
 lot+0.082
 truly+0.082
 valuable+0.078
Neg logit contributions
ases-0.080
per-0.064
ws-0.062
pered-0.061
ena-0.061
Token from
Token ID422
Feature activation+6.65e+00

Pos logit contributions
ases+0.100
ws+0.088
wed+0.088
per+0.078
ena+0.078
Neg logit contributions
 brand-0.073
 brightly-0.073
 very-0.067
 lot-0.061
 truly-0.060
Token a
Token ID257
Feature activation-5.26e+00

Pos logit contributions
ases+0.220
wed+0.194
per+0.190
ws+0.188
ena+0.178
Neg logit contributions
 brand-0.191
 brightly-0.175
 lot-0.160
 very-0.156
 wonderful-0.154
Token nearby
Token ID6716
Feature activation-8.23e-01

Pos logit contributions
 brand+0.130
 brightly+0.113
+0.111
 truly+0.104
 brilliantly+0.102
Neg logit contributions
ases-0.144
per-0.141
 aid-0.131
ena-0.129
pered-0.126
Token park
Token ID3952
Feature activation-4.50e+00

Pos logit contributions
 brand+0.022
 brightly+0.018
 lot+0.017
 truly+0.017
 valuable+0.016
Neg logit contributions
ases-0.029
per-0.025
pered-0.024
ena-0.024
ws-0.024
Token.
Token ID13
Feature activation+1.26e+00

Pos logit contributions
 brand+0.140
 brightly+0.127
 truly+0.118
 very+0.115
 lot+0.114
Neg logit contributions
ases-0.146
ws-0.129
wed-0.122
pered-0.113
per-0.112
Token He
Token ID679
Feature activation-8.81e-01

Pos logit contributions
ases+0.047
per+0.040
ena+0.038
ws+0.038
pered+0.038
Neg logit contributions
 brand-0.038
 brightly-0.031
 lot-0.030
 truly-0.029
 valuable-0.028
Token had
Token ID550
Feature activation-4.09e-01

Pos logit contributions
 brand+0.025
 brightly+0.021
 truly+0.020
 lot+0.020
 valuable+0.019
Neg logit contributions
ases-0.031
per-0.026
pered-0.025
ws-0.025
ena-0.024
Token had
Token ID550
Feature activation-3.95e-01

Pos logit contributions
 brand+0.010
 brightly+0.008
 lot+0.007
 truly+0.007
 valuable+0.007
Neg logit contributions
ases-0.016
per-0.014
pered-0.013
ws-0.013
ena-0.013
Token such
Token ID884
Feature activation-5.21e+00

Pos logit contributions
 brand+0.010
 brightly+0.008
 lot+0.008
 truly+0.007
 valuable+0.007
Neg logit contributions
ases-0.015
per-0.013
pered-0.013
ws-0.012
ena-0.012
Token a
Token ID257
Feature activation-5.56e+00

Pos logit contributions
 brand+0.160
 brightly+0.130
+0.130
 lot+0.118
 truly+0.117
Neg logit contributions
ases-0.151
ena-0.128
per-0.128
pered-0.127
e-0.120
Token wonderful
Token ID7932
Feature activation-4.45e+00

Pos logit contributions
 brand+0.088
 brightly+0.075
+0.060
 brilliantly+0.059
 truly+0.053
Neg logit contributions
ases-0.210
per-0.192
ena-0.186
pered-0.185
e-0.183
Token time
Token ID640
Feature activation-7.06e+00

Pos logit contributions
 brand+0.108
 brightly+0.087
 lot+0.084
 truly+0.080
 valuable+0.076
Neg logit contributions
ases-0.172
per-0.150
pered-0.145
ena-0.144
ws-0.143
Token.
Token ID13
Feature activation+1.39e+00

Pos logit contributions
 brand+0.218
 brightly+0.191
 truly+0.184
 lot+0.176
 very+0.171
Neg logit contributions
ases-0.223
ws-0.193
wed-0.185
per-0.181
pered-0.181
Token He
Token ID679
Feature activation-8.98e-01

Pos logit contributions
ases+0.045
per+0.038
pered+0.037
ws+0.036
ena+0.036
Neg logit contributions
 brand-0.042
 brightly-0.036
 lot-0.034
 truly-0.033
 valuable-0.032
Token was
Token ID373
Feature activation-4.59e+00

Pos logit contributions
 brand+0.023
 brightly+0.019
 truly+0.019
 lot+0.019
 valuable+0.017
Neg logit contributions
ases-0.034
per-0.028
pered-0.028
ws-0.028
ena-0.027
Token proud
Token ID6613
Feature activation-3.36e+00

Pos logit contributions
 brand+0.089
 truly+0.066
 brightly+0.065
 lot+0.060
 very+0.058
Neg logit contributions
ases-0.204
per-0.178
ws-0.176
wed-0.172
pered-0.170
Token keep
Token ID1394
Feature activation+3.13e+00

Pos logit contributions
ases+0.025
pered+0.021
per+0.021
ws+0.021
ena+0.020
Neg logit contributions
 brand-0.013
 brightly-0.011
 truly-0.011
 lot-0.010
 very-0.010
Token that
Token ID326
Feature activation-2.30e+00

Pos logit contributions
ases+0.090
per+0.075
pered+0.071
ena+0.070
ws+0.070
Neg logit contributions
 brand-0.105
 brightly-0.091
 lot-0.087
 truly-0.086
 valuable-0.083
Token huge
Token ID3236
Feature activation-9.20e-01

Pos logit contributions
 brand+0.060
 brightly+0.049
 lot+0.046
 truly+0.046
+0.045
Neg logit contributions
ases-0.078
per-0.069
ws-0.066
ena-0.065
pered-0.065
Token fork
Token ID15563
Feature activation-4.13e+00

Pos logit contributions
 brand+0.026
 brightly+0.022
 lot+0.021
 truly+0.020
 wonderful+0.020
Neg logit contributions
ases-0.032
per-0.027
pered-0.026
ena-0.026
ws-0.026
Token as
Token ID355
Feature activation-8.97e+00

Pos logit contributions
 brand+0.118
 brightly+0.099
 lot+0.094
 truly+0.092
 valuable+0.089
Neg logit contributions
ases-0.140
per-0.119
pered-0.115
ena-0.114
ws-0.113
Token a
Token ID257
Feature activation-5.52e+00

Pos logit contributions
 brand+0.224
 brightly+0.182
 lot+0.172
 truly+0.168
 valuable+0.160
Neg logit contributions
ases-0.334
per-0.290
pered-0.279
ena-0.279
ws-0.277
Token reminder
Token ID15438
Feature activation-4.47e+00

Pos logit contributions
 brand+0.130
 brightly+0.105
 truly+0.101
+0.099
 lot+0.094
Neg logit contributions
ases-0.188
per-0.169
ena-0.161
pered-0.158
ws-0.157
Token of
Token ID286
Feature activation+4.99e+00

Pos logit contributions
 brand+0.147
 brightly+0.126
 lot+0.121
 truly+0.120
 valuable+0.116
Neg logit contributions
ases-0.137
per-0.112
pered-0.110
ws-0.107
ena-0.106
Token their
Token ID511
Feature activation+1.45e+00

Pos logit contributions
ases+0.137
per+0.113
ena+0.110
ws+0.106
pered+0.105
Neg logit contributions
 brand-0.172
 brightly-0.149
 lot-0.140
 truly-0.139
 valuable-0.136
Token day
Token ID1110
Feature activation-7.99e+00

Pos logit contributions
ases+0.056
per+0.049
pered+0.047
ena+0.047
ws+0.047
Neg logit contributions
 brand-0.034
 brightly-0.027
 lot-0.025
 truly-0.025
 valuable-0.024
Token of
Token ID286
Feature activation+4.99e+00

Pos logit contributions
 brand+0.255
 brightly+0.233
 truly+0.217
 lot+0.214
 very+0.202
Neg logit contributions
ases-0.245
ws-0.206
pered-0.194
per-0.192
wed-0.192
Token a
Token ID257
Feature activation-4.91e+00

Pos logit contributions
 brand+0.112
 brightly+0.092
 lot+0.088
 truly+0.088
+0.085
Neg logit contributions
ases-0.138
ena-0.120
per-0.119
pered-0.115
wed-0.115
Token boy
Token ID2933
Feature activation+7.82e-01

Pos logit contributions
 brand+0.108
+0.097
 truly+0.089
 brightly+0.086
t+0.082
Neg logit contributions
ena-0.129
per-0.126
 ear-0.126
ases-0.124
 aid-0.124
Token.
Token ID13
Feature activation+1.77e+00

Pos logit contributions
ases+0.024
ws+0.022
pered+0.021
ps+0.020
aled+0.020
Neg logit contributions
 brand-0.024
 very-0.023
 much-0.021
 brightly-0.021
 lot-0.021
Token The
Token ID383
Feature activation+3.53e-01

Pos logit contributions
ases+0.053
per+0.045
pered+0.043
ws+0.042
ena+0.042
Neg logit contributions
 brand-0.057
 brightly-0.050
 lot-0.047
 truly-0.046
 valuable-0.045
Token boy
Token ID2933
Feature activation+8.00e-01

Pos logit contributions
ases+0.010
ena+0.009
per+0.008
ws+0.008
wed+0.008
Neg logit contributions
 brand-0.011
 brightly-0.009
 truly-0.009
-0.009
 valuable-0.009
Token did
Token ID750
Feature activation-3.82e+00

Pos logit contributions
ases+0.034
pered+0.030
per+0.029
ws+0.028
race+0.028
Neg logit contributions
 brand-0.017
 brightly-0.014
 truly-0.013
 very-0.013
 lot-0.013
Token not
Token ID407
Feature activation-2.18e+00

Pos logit contributions
 brand+0.093
 brightly+0.076
 lot+0.070
 truly+0.069
 valuable+0.066
Neg logit contributions
ases-0.146
per-0.127
pered-0.123
ena-0.122
ws-0.122
Token like
Token ID588
Feature activation-1.85e+00

Pos logit contributions
 lot+0.055
 brand+0.053
 very+0.053
 truly+0.052
 much+0.051
Neg logit contributions
ases-0.074
pered-0.064
per-0.060
ucks-0.058
ena-0.056
Token the
Token ID262
Feature activation+2.87e-01

Pos logit contributions
 brand+0.061
 brightly+0.052
 lot+0.050
 truly+0.049
 valuable+0.048
Neg logit contributions
ases-0.055
per-0.046
pered-0.044
ena-0.043
ws-0.043
Token bed
Token ID3996
Feature activation-3.40e+00

Pos logit contributions
ases+0.010
ena+0.008
per+0.008
ws+0.008
wed+0.008
Neg logit contributions
 brand-0.008
 brightly-0.006
 truly-0.006
-0.006
 valuable-0.006
Token.
Token ID13
Feature activation+1.79e+00

Pos logit contributions
 brand+0.116
 brightly+0.100
 lot+0.096
 truly+0.094
 valuable+0.091
Neg logit contributions
ases-0.095
per-0.079
ena-0.075
pered-0.075
ws-0.073
Token Lily
Token ID20037
Feature activation+7.26e+00

Pos logit contributions
ases+0.044
per+0.038
pered+0.037
ws+0.037
wed+0.037
Neg logit contributions
 brand-0.047
 lot-0.043
 brightly-0.041
 very-0.039
 fun-0.039
Token said
Token ID531
Feature activation+5.77e+00

Pos logit contributions
ases+0.249
pered+0.207
per+0.206
ws+0.201
wed+0.198
Neg logit contributions
 brand-0.212
 brightly-0.173
 lot-0.169
 truly-0.169
 valuable-0.161
Token.
Token ID13
Feature activation+1.60e+00

Pos logit contributions
ases+0.202
ws+0.176
wed+0.166
ena+0.161
aled+0.154
Neg logit contributions
 brand-0.159
 very-0.145
 brightly-0.137
 much-0.134
 lot-0.124
Token "
Token ID366
Feature activation-2.06e+00

Pos logit contributions
ases+0.062
per+0.054
ws+0.052
ena+0.052
wed+0.052
Neg logit contributions
 brand-0.038
 brightly-0.031
 lot-0.029
 truly-0.028
 valuable-0.027
TokenThis
Token ID1212
Feature activation-1.66e+00

Pos logit contributions
 brand+0.063
 brightly+0.052
 lot+0.050
 truly+0.049
 valuable+0.047
Neg logit contributions
ases-0.066
per-0.057
ena-0.054
pered-0.054
ws-0.053
Token is
Token ID318
Feature activation-5.76e+00

Pos logit contributions
 brand+0.053
 brightly+0.046
 sudden+0.042
 daunting+0.042
 brilliantly+0.042
Neg logit contributions
ases-0.044
ena-0.039
ws-0.036
e-0.036
wed-0.035
Token the
Token ID262
Feature activation-7.42e-02

Pos logit contributions
 brand+0.188
 brightly+0.164
+0.162
 sudden+0.158
 lot+0.153
Neg logit contributions
ases-0.153
ena-0.132
ws-0.125
per-0.123
wed-0.119
Token best
Token ID1266
Feature activation-4.98e+00

Pos logit contributions
 brand+0.002
 brightly+0.001
+0.001
 sudden+0.001
 brilliantly+0.001
Neg logit contributions
ases-0.003
ena-0.002
per-0.002
ws-0.002
ps-0.002
Token birthday
Token ID10955
Feature activation-5.05e+00

Pos logit contributions
 brand+0.118
 brightly+0.095
 lot+0.088
 truly+0.087
 valuable+0.083
Neg logit contributions
ases-0.193
per-0.167
ena-0.162
pered-0.161
ws-0.161
Token ever
Token ID1683
Feature activation+1.40e+00

Pos logit contributions
 brand+0.157
 brightly+0.135
 lot+0.128
 truly+0.127
 valuable+0.122
Neg logit contributions
ases-0.158
per-0.135
pered-0.130
ws-0.127
ena-0.125
Token!"
Token ID2474
Feature activation+2.07e+00

Pos logit contributions
ases+0.038
per+0.031
pered+0.030
ws+0.029
ena+0.029
Neg logit contributions
 brand-0.050
 brightly-0.043
 lot-0.042
 truly-0.041
 valuable-0.040
Token happy
Token ID3772
Feature activation-5.25e+00

Pos logit contributions
 brand+0.302
 brightly+0.245
 lot+0.228
 truly+0.219
+0.212
Neg logit contributions
ases-0.448
per-0.393
pered-0.387
ena-0.371
wed-0.368
Token,
Token ID11
Feature activation-3.59e+00

Pos logit contributions
 brand+0.156
 brightly+0.136
 truly+0.127
 lot+0.126
 valuable+0.120
Neg logit contributions
ases-0.175
per-0.146
ws-0.145
pered-0.142
wed-0.141
Token and
Token ID290
Feature activation-2.27e+00

Pos logit contributions
 brand+0.117
 brightly+0.101
 lot+0.097
 truly+0.095
 valuable+0.092
Neg logit contributions
ases-0.111
per-0.090
ws-0.087
ena-0.087
pered-0.087
Token it
Token ID340
Feature activation-7.97e+00

Pos logit contributions
 brand+0.060
 brightly+0.050
 lot+0.047
 truly+0.047
 valuable+0.045
Neg logit contributions
ases-0.083
per-0.071
ena-0.068
ws-0.068
pered-0.068
Token hopped
Token ID45230
Feature activation+7.08e+00

Pos logit contributions
 brand+0.224
 brightly+0.190
 truly+0.185
 lot+0.179
 valuable+0.173
Neg logit contributions
ases-0.279
per-0.231
pered-0.226
ws-0.224
ena-0.223
Token all
Token ID477
Feature activation-3.03e+00

Pos logit contributions
ases+0.221
ena+0.185
per+0.176
ws+0.175
wed+0.170
Neg logit contributions
 brand-0.230
 brightly-0.203
 lot-0.196
 truly-0.187
 very-0.182
Token the
Token ID262
Feature activation-4.96e-01

Pos logit contributions
 brand+0.078
 brightly+0.068
 lot+0.067
 truly+0.065
 very+0.062
Neg logit contributions
ases-0.106
per-0.095
ena-0.090
pered-0.089
ws-0.089
Token way
Token ID835
Feature activation-3.89e+00

Pos logit contributions
 brand+0.013
 brightly+0.011
 truly+0.010
 lot+0.010
+0.009
Neg logit contributions
ases-0.018
per-0.016
pered-0.015
ws-0.015
wed-0.015
Token home
Token ID1363
Feature activation-1.08e+00

Pos logit contributions
 brand+0.106
 brightly+0.089
 truly+0.087
 very+0.086
 lot+0.084
Neg logit contributions
ases-0.143
ws-0.122
ena-0.120
wed-0.114
per-0.114
Token.
Token ID13
Feature activation+6.69e-01

Pos logit contributions
 brand+0.036
 brightly+0.031
 truly+0.030
 very+0.029
 lot+0.029
Neg logit contributions
ases-0.033
ws-0.027
per-0.026
ena-0.025
wed-0.025
Token<|endoftext|>
Token ID50256
Feature activation+3.28e+00

Pos logit contributions
ases+0.020
per+0.017
pered+0.016
ena+0.016
ws+0.016
Neg logit contributions
 brand-0.022
 brightly-0.018
 lot-0.018
 truly-0.017
 valuable-0.017
Token the
Token ID262
Feature activation+1.81e-02

Pos logit contributions
ases+0.150
per+0.124
pered+0.122
ws+0.118
wed+0.118
Neg logit contributions
 brand-0.158
 brightly-0.137
 lot-0.134
 truly-0.127
 valuable-0.123
Token sea
Token ID5417
Feature activation-7.10e+00

Pos logit contributions
ases+0.001
per+0.001
ws+0.000
pered+0.000
ena+0.000
Neg logit contributions
 brand-0.001
 brightly-0.000
-0.000
 truly-0.000
 lot-0.000
Token.
Token ID13
Feature activation+1.48e+00

Pos logit contributions
 brand+0.235
 brightly+0.205
 lot+0.200
 truly+0.193
 very+0.190
Neg logit contributions
ases-0.212
pered-0.173
ws-0.169
wed-0.164
ena-0.162
Token He
Token ID679
Feature activation-6.06e-01

Pos logit contributions
ases+0.048
per+0.041
pered+0.039
ena+0.039
ws+0.039
Neg logit contributions
 brand-0.045
 brightly-0.038
 lot-0.036
 truly-0.036
 valuable-0.034
Token was
Token ID373
Feature activation-4.38e+00

Pos logit contributions
 brand+0.017
 brightly+0.015
 lot+0.014
 truly+0.014
 fantastic+0.013
Neg logit contributions
ases-0.020
per-0.017
ena-0.017
pered-0.016
wed-0.016
Token so
Token ID523
Feature activation-1.26e+01

Pos logit contributions
 brand+0.078
 brightly+0.060
+0.058
 lot+0.056
ww+0.049
Neg logit contributions
ases-0.184
per-0.172
pered-0.169
ena-0.163
ws-0.156
Token excited
Token ID6568
Feature activation-2.30e+00

Pos logit contributions
 brand+0.353
 brightly+0.281
+0.277
 lot+0.277
 truly+0.265
Neg logit contributions
ases-0.422
per-0.372
pered-0.357
ena-0.346
ws-0.341
Token,
Token ID11
Feature activation-2.82e+00

Pos logit contributions
 brand+0.073
 brightly+0.062
 lot+0.059
 truly+0.058
 valuable+0.056
Neg logit contributions
ases-0.071
per-0.060
pered-0.057
ws-0.057
ena-0.056
Token he
Token ID339
Feature activation-1.00e+00

Pos logit contributions
 brand+0.064
 brightly+0.051
 lot+0.047
 truly+0.046
 valuable+0.044
Neg logit contributions
ases-0.111
per-0.099
pered-0.095
ena-0.094
ws-0.093
Token cl
Token ID537
Feature activation+2.63e+00

Pos logit contributions
 brand+0.033
 brightly+0.028
 lot+0.027
 truly+0.027
 valuable+0.026
Neg logit contributions
ases-0.029
per-0.025
pered-0.023
ena-0.023
ws-0.023
Tokenapped
Token ID6320
Feature activation-3.54e-01

Pos logit contributions
ases+0.074
ws+0.063
per+0.063
pered+0.060
ena+0.060
Neg logit contributions
 brand-0.085
 brightly-0.076
 truly-0.069
 lot-0.067
 brilliantly-0.066
Token outside
Token ID2354
Feature activation+1.27e+00

Pos logit contributions
 brand+0.211
 lot+0.178
 valuable+0.177
 truly+0.177
 brightly+0.176
Neg logit contributions
ases-0.130
per-0.110
pered-0.100
ena-0.098
wed-0.097
Token,
Token ID11
Feature activation-2.83e+00

Pos logit contributions
ases+0.041
ws+0.035
per+0.034
wed+0.034
pered+0.033
Neg logit contributions
 brand-0.039
 brightly-0.036
 lot-0.033
 truly-0.031
 most-0.031
Token and
Token ID290
Feature activation-1.59e+00

Pos logit contributions
 brand+0.094
 brightly+0.081
 lot+0.077
 truly+0.076
 valuable+0.073
Neg logit contributions
ases-0.083
per-0.069
pered-0.066
ws-0.065
ena-0.065
Token Billy
Token ID15890
Feature activation+8.90e+00

Pos logit contributions
 brand+0.046
 brightly+0.039
 lot+0.037
 truly+0.036
 valuable+0.034
Neg logit contributions
ases-0.054
per-0.046
pered-0.044
ws-0.043
ena-0.043
Token was
Token ID373
Feature activation-4.38e+00

Pos logit contributions
ases+0.387
pered+0.323
ws+0.312
per+0.307
ucks+0.301
Neg logit contributions
 brand-0.189
 truly-0.159
 brightly-0.154
 very-0.149
 lot-0.145
Token so
Token ID523
Feature activation-1.25e+01

Pos logit contributions
 brand+0.104
 brightly+0.084
 truly+0.078
 lot+0.077
 valuable+0.071
Neg logit contributions
ases-0.174
per-0.149
ws-0.146
pered-0.144
wed-0.143
Token excited
Token ID6568
Feature activation-2.24e+00

Pos logit contributions
 brand+0.345
 lot+0.276
+0.275
 brightly+0.273
 truly+0.259
Neg logit contributions
ases-0.392
per-0.333
pered-0.324
ena-0.323
wed-0.308
Token.
Token ID13
Feature activation+1.39e+00

Pos logit contributions
 brand+0.074
 brightly+0.064
 lot+0.060
 truly+0.060
 valuable+0.058
Neg logit contributions
ases-0.067
per-0.056
pered-0.053
ws-0.053
ena-0.053
Token He
Token ID679
Feature activation-4.64e-01

Pos logit contributions
ases+0.041
per+0.035
pered+0.034
ena+0.033
ws+0.033
Neg logit contributions
 brand-0.044
 brightly-0.038
 lot-0.036
 truly-0.036
 valuable-0.034
Token wanted
Token ID2227
Feature activation+3.63e+00

Pos logit contributions
 brand+0.013
 brightly+0.011
 truly+0.011
 lot+0.011
 wonderful+0.010
Neg logit contributions
ases-0.017
per-0.014
pered-0.014
ws-0.014
ena-0.013
Token to
Token ID284
Feature activation+6.02e+00

Pos logit contributions
ases+0.098
per+0.080
ena+0.078
ws+0.075
wed+0.075
Neg logit contributions
 brand-0.122
 brightly-0.105
-0.103
 lot-0.101
 truly-0.098
Token
Token ID220
Feature activation-1.07e+01

Pos logit contributions
ases+0.052
per+0.045
pered+0.044
ena+0.044
ws+0.043
Neg logit contributions
 brand-0.035
 brightly-0.028
 lot-0.026
 truly-0.026
 valuable-0.025
Token
Token ID198
Feature activation-3.60e+00

Pos logit contributions
 brand+0.222
 brightly+0.171
 lot+0.165
 truly+0.162
 valuable+0.148
Neg logit contributions
ases-0.445
per-0.393
ws-0.379
wed-0.377
pered-0.376
Token
Token ID198
Feature activation-3.70e+00

Pos logit contributions
 brand+0.093
 brightly+0.076
 lot+0.071
 truly+0.069
 valuable+0.066
Neg logit contributions
ases-0.131
per-0.114
ena-0.110
pered-0.110
ws-0.108
TokenShe
Token ID3347
Feature activation+3.77e-01

Pos logit contributions
 brand+0.089
 brightly+0.070
 lot+0.064
 truly+0.063
 valuable+0.059
Neg logit contributions
ases-0.142
per-0.126
ena-0.122
pered-0.121
wed-0.118
Token felt
Token ID2936
Feature activation-3.96e+00

Pos logit contributions
ases+0.013
per+0.011
pered+0.011
ws+0.011
ena+0.011
Neg logit contributions
 brand-0.010
 brightly-0.009
 truly-0.008
 lot-0.008
 valuable-0.008
Token so
Token ID523
Feature activation-1.28e+01

Pos logit contributions
 brand+0.091
 brightly+0.076
ww+0.073
+0.073
 lifetime+0.068
Neg logit contributions
ases-0.137
per-0.126
pered-0.123
e-0.118
ws-0.114
Token helpless
Token ID21144
Feature activation-3.45e+00

Pos logit contributions
 brand+0.308
 brightly+0.242
 lot+0.224
 truly+0.215
+0.213
Neg logit contributions
ases-0.474
per-0.414
pered-0.407
ws-0.396
wed-0.394
Token and
Token ID290
Feature activation-1.93e+00

Pos logit contributions
 brand+0.113
 brightly+0.098
 lot+0.093
 truly+0.092
 valuable+0.088
Neg logit contributions
ases-0.104
per-0.086
pered-0.082
ws-0.082
ena-0.081
Token disappointed
Token ID11679
Feature activation-4.08e+00

Pos logit contributions
 brand+0.051
 brightly+0.041
 lot+0.039
 truly+0.038
+0.037
Neg logit contributions
ases-0.067
per-0.059
pered-0.057
ena-0.056
wed-0.056
Token in
Token ID287
Feature activation+4.73e+00

Pos logit contributions
 brand+0.156
 brightly+0.137
 lot+0.132
 truly+0.130
 valuable+0.127
Neg logit contributions
ases-0.098
per-0.079
pered-0.074
ena-0.073
ws-0.072
Token herself
Token ID5223
Feature activation+6.59e-01

Pos logit contributions
ases+0.147
per+0.128
ws+0.122
pered+0.121
wed+0.121
Neg logit contributions
 brand-0.145
 lot-0.129
 brightly-0.128
 wonderful-0.121
 truly-0.120
Token
Token ID198
Feature activation-2.57e+00

Pos logit contributions
ases+0.086
per+0.074
ws+0.074
wed+0.074
pered+0.071
Neg logit contributions
 brand-0.049
 brightly-0.039
 lot-0.037
 truly-0.036
 valuable-0.034
Token
Token ID198
Feature activation-2.75e+00

Pos logit contributions
 brand+0.050
 lot+0.038
 brightly+0.037
 truly+0.036
 much+0.035
Neg logit contributions
ases-0.111
per-0.099
ws-0.096
wed-0.096
pered-0.093
Token"
Token ID1
Feature activation+1.83e+00

Pos logit contributions
 brand+0.064
 brightly+0.055
 lot+0.053
 truly+0.052
 valuable+0.050
Neg logit contributions
ases-0.106
ws-0.092
per-0.091
wed-0.089
ps-0.088
TokenThat
Token ID2504
Feature activation-6.08e+00

Pos logit contributions
ases+0.066
ena+0.058
per+0.058
pered+0.054
wed+0.054
Neg logit contributions
 brand-0.048
 lot-0.036
 brightly-0.036
 valuable-0.034
 truly-0.033
Token was
Token ID373
Feature activation-4.15e+00

Pos logit contributions
 brand+0.194
 brightly+0.168
+0.162
 lot+0.159
 sudden+0.154
Neg logit contributions
ases-0.162
ena-0.138
per-0.137
ws-0.130
wed-0.128
Token so
Token ID523
Feature activation-1.28e+01

Pos logit contributions
 brand+0.082
+0.072
 brightly+0.067
 sudden+0.065
 lot+0.061
Neg logit contributions
ases-0.154
ena-0.138
per-0.136
ws-0.133
pered-0.131
Token much
Token ID881
Feature activation-1.07e+01

Pos logit contributions
 brand+0.204
 brightly+0.142
+0.130
 lot+0.125
 sudden+0.118
Neg logit contributions
ases-0.562
per-0.493
ena-0.483
pered-0.483
ws-0.480
Token fun
Token ID1257
Feature activation-6.72e+00

Pos logit contributions
 brand+0.152
+0.107
 brightly+0.096
 lot+0.092
 truly+0.079
Neg logit contributions
ases-0.469
per-0.409
ws-0.406
ena-0.403
race-0.397
Token,
Token ID11
Feature activation-3.03e+00

Pos logit contributions
 brand+0.215
 brightly+0.188
 lot+0.176
 truly+0.175
 valuable+0.170
Neg logit contributions
ases-0.210
per-0.175
ws-0.170
pered-0.170
wed-0.168
Token thanks
Token ID5176
Feature activation+5.69e+00

Pos logit contributions
 brand+0.093
 brightly+0.081
 lot+0.077
 truly+0.075
 wonderful+0.073
Neg logit contributions
ases-0.101
per-0.083
ws-0.082
pered-0.082
wed-0.081
Token for
Token ID329
Feature activation+5.17e+00

Pos logit contributions
ases+0.211
per+0.186
ws+0.178
pered+0.177
wed+0.176
Neg logit contributions
 brand-0.147
 brightly-0.128
 lot-0.120
 wonderful-0.115
 fun-0.114
Token can
Token ID460
Feature activation-5.11e+00

Pos logit contributions
 brand+0.204
 brightly+0.170
 lot+0.159
 truly+0.149
 valuable+0.147
Neg logit contributions
ases-0.270
per-0.236
ws-0.231
wed-0.227
pered-0.222
Token go
Token ID467
Feature activation+5.29e+00

Pos logit contributions
 brand+0.169
 brightly+0.147
 lot+0.141
 truly+0.139
 valuable+0.132
Neg logit contributions
ases-0.151
per-0.124
pered-0.121
ena-0.120
ws-0.118
Token to
Token ID284
Feature activation+6.19e+00

Pos logit contributions
ases+0.152
per+0.127
pered+0.121
ws+0.119
ena+0.119
Neg logit contributions
 brand-0.176
 brightly-0.152
 lot-0.145
 truly-0.144
 valuable-0.140
Token grandma
Token ID49890
Feature activation+7.07e+00

Pos logit contributions
ases+0.200
per+0.165
pered+0.164
wed+0.161
ena+0.159
Neg logit contributions
 brand-0.193
 brightly-0.164
 lot-0.160
 truly-0.155
 wonderful-0.147
Token?"
Token ID1701
Feature activation+1.73e+00

Pos logit contributions
ases+0.230
pered+0.190
ws+0.182
per+0.179
wed+0.179
Neg logit contributions
 brand-0.222
 brightly-0.182
 truly-0.182
 lot-0.172
 very-0.169
Token
Token ID220
Feature activation-1.06e+01

Pos logit contributions
ases+0.066
per+0.057
ws+0.055
pered+0.055
wed+0.054
Neg logit contributions
 brand-0.044
 brightly-0.036
 lot-0.034
 truly-0.033
 valuable-0.031
Token
Token ID198
Feature activation-3.63e+00

Pos logit contributions
 brand+0.197
 lot+0.146
 brightly+0.146
 truly+0.137
 valuable+0.127
Neg logit contributions
ases-0.458
per-0.407
ws-0.393
pered-0.392
wed-0.390
Token
Token ID198
Feature activation-3.70e+00

Pos logit contributions
 brand+0.076
 brightly+0.060
 lot+0.058
 truly+0.055
 valuable+0.051
Neg logit contributions
ases-0.153
per-0.134
ws-0.132
pered-0.131
wed-0.128
TokenHer
Token ID9360
Feature activation+3.19e+00

Pos logit contributions
 brand+0.116
 brightly+0.100
 lot+0.095
 truly+0.093
 valuable+0.090
Neg logit contributions
ases-0.117
per-0.098
pered-0.095
ws-0.094
wed-0.092
Token mom
Token ID1995
Feature activation+7.58e+00

Pos logit contributions
ases+0.083
ena+0.070
per+0.068
pered+0.065
wed+0.063
Neg logit contributions
 brand-0.109
 brightly-0.097
 truly-0.093
 lot-0.091
 brilliantly-0.088
Tokenmy
Token ID1820
Feature activation+4.10e+00

Pos logit contributions
ases+0.203
per+0.163
pered+0.160
ws+0.151
wed+0.138
Neg logit contributions
 brand-0.279
 brightly-0.242
 truly-0.232
 lot-0.231
 wonderful-0.222