TOP ACTIVATIONS
MAX = 9.559

She ate popcorn and roasted marsh m allows .
camp fire . They roasted marsh m allows and told stories
and Mia enjoyed ro asting marsh m allows and telling stories
made a fire and roasted marsh m allows . They gave
. They pretended to roast marsh m allows and sing songs
him tight while they roasted marsh m allows over the camp
around the fireplace and roasted marsh m allows . Suddenly ,
Daisy enjoyed making popcorn with marsh m allows , because it
it was delicious ! The marsh m allows had made the
They shared some cookies and marsh m allows . They were
cheered and began to asting marsh m allows . All of
ugs with hot cocoa and marsh m allows . They smiled
and ate hot dogs and marsh m allows . They liked
While they were ro asting marsh m allows , they heard
ate some hot dogs and marsh m allows . They sang
They were going to roast marsh m allows in the fireplace
because she loved to roast marsh m allows over the fire
fireplace so they could roast marsh m allows . They were
also see some rocks . Some are small and shiny .
They see many flowers . Some are red , some are

MIN ACTIVATIONS
MIN = -7.408

The bad ending was fore shadow ed by Maggie 's
lived a princess called R yle e . Every day ,
mum said , " Oh look , he likes you .
he didn 't have an invitation . He asked his friend
world and vowed to never look back . <|endoftext|> Once upon
day , Benny received an invitation to a party . He
her why she didn 't look happy . Lucy said :
it around . " Oh look , Amy , your heart
ot ter who didn 't look before he leapt . He
The girl had been fore shadow ed . Even though
yard . He had been fore shadow ed about this bad
said ' You don 't look very fine '. Tina started
The old man thanked the messenger and said � � €
. " Oh look at you !" said Jane
day , she received an invitation to a party from her
know , her mum was fore shadow ing ... the next
day , Lily received an invitation in the mail . It
. But he didn 't look where he was going and
fish said , " Hi Gab ri ella , would you
ran away and didn 't look back . He never returned

INTERVAL 5.317 - 9.559
CONTAINS 0.100%

filled with treasures from long ago ! The adventurers
goose played together all day long . They ran , jumped
! He continued r aking the leaves until all
fresh air made him feel more awake . When he went
about their snow man . Mom and dad said they were

INTERVAL 1.075 - 5.317
CONTAINS 21.785%

ies and toys . They were very happy and shouted :
with many green needles . Lily liked the pine tree .
her some new poppy seeds as a thank you present .
arms up high and his mom would wrap him up in
, I will have to squeeze you !" Mary was very

INTERVAL -3.166 - 1.075
CONTAINS 76.463%

George felt so sad . He wanted to fix it ,
the end , Lily 's grandma came and they had a
did it !" he exclaimed . His mother smiled
little girl named Daisy . Daisy was playing in her front
her mom my and daddy said , " L ily ,

INTERVAL -7.408 - -3.166
CONTAINS 1.651%

. It 's important to look nice for our family ."
Her mom said , " Of course , let 's go
On the day of the trip , Tim my 's mom
amazed at all the weird looking leaves she found . But
's enough for today . How about we have a snack
Token She
Token ID1375
Feature activation+3.10e-01

Pos logit contributions
First+0.010
 brushes+0.009
 sizes+0.008
ups+0.008
 seasons+0.008
Neg logit contributions
 victorious-0.008
 hero-0.008
 CA-0.007
 sorry-0.007
 devastated-0.007
Token ate
Token ID15063
Feature activation+2.00e+00

Pos logit contributions
First+0.011
 brushes+0.011
 sizes+0.010
 flour+0.009
ups+0.009
Neg logit contributions
 victorious-0.010
 hero-0.010
 CA-0.010
 sorry-0.009
 devastated-0.009
Token popcorn
Token ID38914
Feature activation+1.88e+00

Pos logit contributions
First+0.077
 brushes+0.073
ups+0.070
 sizes+0.068
vals+0.065
Neg logit contributions
 victorious-0.060
 hero-0.058
 sorry-0.056
 CA-0.055
 hi-0.054
Token and
Token ID290
Feature activation+1.06e+00

Pos logit contributions
First+0.070
 brushes+0.067
 sizes+0.061
ups+0.059
 seasons+0.059
Neg logit contributions
 victorious-0.061
 hero-0.060
 CA-0.054
 sorry-0.054
 goodbye-0.051
Token roasted
Token ID30490
Feature activation+4.98e+00

Pos logit contributions
First+0.046
 brushes+0.042
ups+0.041
akes+0.040
vals+0.038
Neg logit contributions
 goodbye-0.025
 hero-0.024
 victorious-0.023
 CA-0.023
 sorry-0.023
Token marsh
Token ID22397
Feature activation+8.81e+00

Pos logit contributions
First+0.124
 brushes+0.118
ups+0.097
 sizes+0.097
 styles+0.091
Neg logit contributions
 victorious-0.217
 hero-0.212
 CA-0.207
 sorry-0.202
 hi-0.196
Tokenm
Token ID76
Feature activation+5.02e+00

Pos logit contributions
 brushes+0.336
First+0.322
 sizes+0.301
 styles+0.295
 seasons+0.293
Neg logit contributions
 victorious-0.234
 hero-0.225
eros-0.218
 sorry-0.214
 hi-0.201
Tokenallows
Token ID47205
Feature activation+5.22e+00

Pos logit contributions
First+0.157
 brushes+0.142
 sizes+0.113
 styles+0.111
ups+0.111
Neg logit contributions
 victorious-0.188
 sorry-0.187
 hero-0.174
 CA-0.169
 relieved-0.169
Token.
Token ID13
Feature activation+7.18e-01

Pos logit contributions
First+0.198
 brushes+0.194
 sizes+0.172
ups+0.171
 seasons+0.168
Neg logit contributions
 victorious-0.161
 hero-0.156
 CA-0.145
 sorry-0.143
 goodbye-0.132
Token
Token ID198
Feature activation-3.07e-01

Pos logit contributions
First+0.029
 brushes+0.028
 sizes+0.026
ups+0.025
 colors+0.025
Neg logit contributions
 victorious-0.021
 hero-0.020
 CA-0.019
 sorry-0.018
 devastated-0.018
Token
Token ID198
Feature activation-1.13e+00

Pos logit contributions
 victorious+0.009
 hero+0.009
 CA+0.009
 sorry+0.008
 devastated+0.008
Neg logit contributions
First-0.012
 brushes-0.012
 sizes-0.011
ups-0.010
 seasons-0.010
Token camp
Token ID1413
Feature activation+1.86e+00

Pos logit contributions
First+0.016
 brushes+0.016
 sizes+0.015
ups+0.014
 styles+0.014
Neg logit contributions
 hero-0.010
 victorious-0.010
 CA-0.009
 sorry-0.008
 devastated-0.008
Tokenfire
Token ID6495
Feature activation+1.33e+00

Pos logit contributions
First+0.071
 brushes+0.070
 sizes+0.061
 flour+0.060
ups+0.059
Neg logit contributions
 victorious-0.058
 hero-0.057
 CA-0.051
 sorry-0.050
 devastated-0.049
Token.
Token ID13
Feature activation+3.23e-01

Pos logit contributions
First+0.052
 brushes+0.050
ups+0.043
 sizes+0.043
 seasons+0.042
Neg logit contributions
 victorious-0.041
 hero-0.039
 sorry-0.035
 goodbye-0.034
 CA-0.034
Token They
Token ID1119
Feature activation+2.09e+00

Pos logit contributions
First+0.011
 brushes+0.010
 sizes+0.009
ups+0.009
 styles+0.009
Neg logit contributions
 victorious-0.011
 hero-0.011
 CA-0.011
 sorry-0.010
 devastated-0.010
Token roasted
Token ID30490
Feature activation+4.83e+00

Pos logit contributions
First+0.070
 brushes+0.068
ups+0.062
 sizes+0.061
 styles+0.059
Neg logit contributions
 victorious-0.072
 hero-0.071
 CA-0.067
 devastated-0.066
 sorry-0.065
Token marsh
Token ID22397
Feature activation+8.72e+00

Pos logit contributions
First+0.143
 brushes+0.130
ups+0.116
 sizes+0.113
 styles+0.110
Neg logit contributions
 victorious-0.190
 hero-0.183
 CA-0.175
 sorry-0.174
 goodbye-0.166
Tokenm
Token ID76
Feature activation+4.85e+00

Pos logit contributions
 brushes+0.320
First+0.315
 styles+0.295
 sizes+0.291
 seasons+0.284
Neg logit contributions
 victorious-0.236
 hero-0.228
eros-0.225
 sorry-0.219
 reunited-0.204
Tokenallows
Token ID47205
Feature activation+5.18e+00

Pos logit contributions
First+0.170
 brushes+0.149
ups+0.126
 styles+0.126
vals+0.125
Neg logit contributions
 sorry-0.158
 victorious-0.151
 relieved-0.148
 lucky-0.138
 hero-0.138
Token and
Token ID290
Feature activation+1.13e+00

Pos logit contributions
First+0.197
 brushes+0.191
ups+0.173
 sizes+0.171
 styles+0.168
Neg logit contributions
 victorious-0.157
 hero-0.153
 CA-0.142
 sorry-0.140
 goodbye-0.130
Token told
Token ID1297
Feature activation+5.45e-01

Pos logit contributions
First+0.042
ups+0.040
 brushes+0.037
akes+0.034
vals+0.033
Neg logit contributions
 victorious-0.031
 sorry-0.031
 relieved-0.030
 goodbye-0.030
 hero-0.029
Token stories
Token ID3923
Feature activation-4.22e-01

Pos logit contributions
First+0.020
 brushes+0.019
 sizes+0.017
ups+0.017
 styles+0.016
Neg logit contributions
 victorious-0.018
 hero-0.018
 CA-0.017
 sorry-0.017
 devastated-0.016
Token and
Token ID290
Feature activation-5.38e-01

Pos logit contributions
First+0.029
 brushes+0.028
 sizes+0.025
ups+0.024
 colors+0.023
Neg logit contributions
 victorious-0.031
 hero-0.030
 CA-0.029
 sorry-0.028
 devastated-0.027
Token Mia
Token ID32189
Feature activation+2.59e+00

Pos logit contributions
 victorious+0.019
 hero+0.019
 sorry+0.017
 goodbye+0.016
 CA+0.016
Neg logit contributions
First-0.018
 brushes-0.017
 sizes-0.016
 seasons-0.015
 styles-0.015
Token enjoyed
Token ID8359
Feature activation+3.58e+00

Pos logit contributions
First+0.083
 brushes+0.081
 sizes+0.072
ups+0.070
 colors+0.068
Neg logit contributions
 victorious-0.094
 hero-0.092
 CA-0.090
 sorry-0.086
 devastated-0.083
Token ro
Token ID686
Feature activation+1.11e+00

Pos logit contributions
First+0.143
 brushes+0.137
ups+0.125
 sizes+0.124
 flour+0.116
Neg logit contributions
 victorious-0.108
 hero-0.103
 sorry-0.099
 CA-0.096
 hi-0.089
Tokenasting
Token ID9222
Feature activation+2.69e+00

Pos logit contributions
First+0.034
 brushes+0.032
 sizes+0.029
ups+0.029
 colors+0.027
Neg logit contributions
 CA-0.041
 hero-0.041
 victorious-0.041
 sorry-0.039
 devastated-0.037
Token marsh
Token ID22397
Feature activation+8.49e+00

Pos logit contributions
First+0.076
 brushes+0.069
ups+0.060
 sizes+0.060
 styles+0.056
Neg logit contributions
 victorious-0.110
 hero-0.107
 CA-0.104
 sorry-0.104
 hi-0.097
Tokenm
Token ID76
Feature activation+4.72e+00

Pos logit contributions
 brushes+0.325
First+0.315
 styles+0.285
 seasons+0.281
 sizes+0.280
Neg logit contributions
 victorious-0.238
 hero-0.224
 sorry-0.222
eros-0.213
smart-0.200
Tokenallows
Token ID47205
Feature activation+5.02e+00

Pos logit contributions
First+0.149
 brushes+0.135
 sizes+0.106
ups+0.106
vals+0.105
Neg logit contributions
 sorry-0.179
 victorious-0.178
 hero-0.164
 relieved-0.159
 devastated-0.158
Token and
Token ID290
Feature activation+6.94e-01

Pos logit contributions
First+0.191
 brushes+0.188
 sizes+0.166
ups+0.164
 seasons+0.162
Neg logit contributions
 victorious-0.155
 hero-0.150
 sorry-0.138
 CA-0.135
 devastated-0.128
Token telling
Token ID5149
Feature activation-1.18e+00

Pos logit contributions
First+0.029
 brushes+0.025
ups+0.023
 flour+0.022
oons+0.022
Neg logit contributions
 victorious-0.020
 sorry-0.019
 hero-0.018
 relieved-0.017
 heroic-0.017
Token stories
Token ID3923
Feature activation-5.69e-01

Pos logit contributions
 victorious+0.039
 hero+0.038
 sorry+0.037
 CA+0.035
 hi+0.035
Neg logit contributions
First-0.042
 brushes-0.039
 sizes-0.035
ups-0.034
vals-0.033
Token made
Token ID925
Feature activation+2.89e+00

Pos logit contributions
First+0.120
 brushes+0.116
 sizes+0.104
ups+0.100
 colors+0.097
Neg logit contributions
 victorious-0.135
 hero-0.132
 CA-0.129
 sorry-0.123
 devastated-0.119
Token a
Token ID257
Feature activation+1.01e+00

Pos logit contributions
First+0.103
 brushes+0.099
 sizes+0.092
ups+0.087
 styles+0.085
Neg logit contributions
 victorious-0.096
 hero-0.092
 CA-0.090
 sorry-0.090
 devastated-0.084
Token fire
Token ID2046
Feature activation+1.72e+00

Pos logit contributions
 brushes+0.036
First+0.036
 sizes+0.031
 flour+0.031
 colors+0.031
Neg logit contributions
 victorious-0.033
 hero-0.032
 CA-0.031
 sorry-0.029
 devastated-0.028
Token and
Token ID290
Feature activation-1.14e-01

Pos logit contributions
First+0.068
 brushes+0.065
 sizes+0.058
ups+0.057
 styles+0.056
Neg logit contributions
 hero-0.052
 victorious-0.052
 CA-0.049
 sorry-0.047
 devastated-0.044
Token roasted
Token ID30490
Feature activation+4.71e+00

Pos logit contributions
 victorious+0.003
 hero+0.003
 sorry+0.003
 CA+0.003
 goodbye+0.003
Neg logit contributions
First-0.004
 brushes-0.004
 sizes-0.004
ups-0.004
 styles-0.004
Token marsh
Token ID22397
Feature activation+8.48e+00

Pos logit contributions
First+0.126
 brushes+0.115
 sizes+0.099
ups+0.098
 styles+0.096
Neg logit contributions
 victorious-0.201
 hero-0.194
 CA-0.192
 sorry-0.190
 hi-0.176
Tokenm
Token ID76
Feature activation+4.86e+00

Pos logit contributions
 brushes+0.317
First+0.312
 styles+0.285
 sizes+0.284
 seasons+0.272
Neg logit contributions
 victorious-0.247
 sorry-0.236
 hero-0.229
 CA-0.219
eros-0.218
Tokenallows
Token ID47205
Feature activation+5.03e+00

Pos logit contributions
First+0.158
 brushes+0.141
 sizes+0.117
 styles+0.116
ups+0.110
Neg logit contributions
 sorry-0.178
 victorious-0.175
 CA-0.161
 hero-0.161
 relieved-0.159
Token.
Token ID13
Feature activation+6.89e-01

Pos logit contributions
First+0.185
 brushes+0.180
 sizes+0.161
ups+0.158
 styles+0.156
Neg logit contributions
 victorious-0.160
 hero-0.155
 CA-0.150
 sorry-0.145
 devastated-0.136
Token They
Token ID1119
Feature activation+1.96e+00

Pos logit contributions
First+0.025
 brushes+0.024
 sizes+0.023
ups+0.022
 styles+0.021
Neg logit contributions
 victorious-0.022
 CA-0.021
 sorry-0.021
 hero-0.021
 devastated-0.019
Token gave
Token ID2921
Feature activation-8.16e-02

Pos logit contributions
First+0.068
 brushes+0.064
ups+0.060
 sizes+0.058
 styles+0.058
Neg logit contributions
 victorious-0.065
 hero-0.063
 CA-0.061
 sorry-0.059
 devastated-0.058
Token.
Token ID13
Feature activation+7.19e-01

Pos logit contributions
First+0.088
 brushes+0.084
ups+0.076
 sizes+0.076
vals+0.073
Neg logit contributions
 victorious-0.060
 hero-0.057
 sorry-0.054
 CA-0.052
 goodbye-0.050
Token They
Token ID1119
Feature activation+1.96e+00

Pos logit contributions
First+0.025
 brushes+0.024
 sizes+0.022
ups+0.022
 styles+0.021
Neg logit contributions
 victorious-0.024
 hero-0.024
 CA-0.023
 sorry-0.022
 devastated-0.021
Token pretended
Token ID36766
Feature activation+1.52e+00

Pos logit contributions
First+0.068
 brushes+0.065
ups+0.062
 sizes+0.059
 seasons+0.059
Neg logit contributions
 hero-0.063
 victorious-0.062
 CA-0.057
 sorry-0.056
 devastated-0.055
Token to
Token ID284
Feature activation+2.37e-01

Pos logit contributions
First+0.055
 brushes+0.050
 sizes+0.045
 styles+0.044
 seasons+0.043
Neg logit contributions
 victorious-0.051
 hero-0.050
 CA-0.049
 sorry-0.047
 hi-0.044
Token roast
Token ID32595
Feature activation+4.94e+00

Pos logit contributions
First+0.010
 brushes+0.009
ups+0.009
 sizes+0.009
 flour+0.008
Neg logit contributions
 hero-0.007
 victorious-0.006
 CA-0.006
 sorry-0.006
 devastated-0.005
Token marsh
Token ID22397
Feature activation+8.48e+00

Pos logit contributions
First+0.155
 brushes+0.137
ups+0.123
 sizes+0.121
 styles+0.116
Neg logit contributions
 victorious-0.186
 hero-0.182
 CA-0.179
 sorry-0.177
 goodbye-0.164
Tokenm
Token ID76
Feature activation+4.84e+00

Pos logit contributions
 brushes+0.302
First+0.300
 styles+0.269
 sizes+0.266
 seasons+0.259
Neg logit contributions
 hero-0.248
 victorious-0.246
eros-0.231
 sorry-0.231
 CA-0.219
Tokenallows
Token ID47205
Feature activation+5.05e+00

Pos logit contributions
First+0.163
 brushes+0.139
 sizes+0.117
 styles+0.115
 flavors+0.113
Neg logit contributions
 sorry-0.168
 victorious-0.162
 hero-0.156
 CA-0.150
 relieved-0.148
Token and
Token ID290
Feature activation+8.78e-01

Pos logit contributions
First+0.189
 brushes+0.183
 sizes+0.164
ups+0.162
 styles+0.160
Neg logit contributions
 victorious-0.158
 hero-0.155
 CA-0.145
 sorry-0.142
 devastated-0.131
Token sing
Token ID1702
Feature activation+3.32e-01

Pos logit contributions
First+0.036
ups+0.032
 brushes+0.031
akes+0.030
w+0.029
Neg logit contributions
 sorry-0.023
 hero-0.022
 victorious-0.022
 goodbye-0.021
 CA-0.020
Token songs
Token ID7259
Feature activation+1.03e+00

Pos logit contributions
First+0.012
 brushes+0.012
ups+0.011
 sizes+0.010
 flavors+0.010
Neg logit contributions
 victorious-0.010
 hero-0.010
 sorry-0.009
 CA-0.009
 hi-0.009
Token him
Token ID683
Feature activation-8.04e-01

Pos logit contributions
 victorious+0.035
 hero+0.035
 CA+0.034
 sorry+0.032
 hi+0.031
Neg logit contributions
First-0.036
 brushes-0.035
 sizes-0.032
ups-0.031
 seasons-0.030
Token tight
Token ID5381
Feature activation+7.97e-01

Pos logit contributions
 CA+0.035
 victorious+0.035
 sorry+0.032
 devastated+0.032
 hero+0.031
Neg logit contributions
First-0.020
 brushes-0.020
 colors-0.018
 sizes-0.017
 shapes-0.016
Token while
Token ID981
Feature activation+3.54e+00

Pos logit contributions
First+0.030
 brushes+0.028
 sizes+0.025
ups+0.025
 seasons+0.025
Neg logit contributions
 hero-0.025
 victorious-0.025
 sorry-0.022
 CA-0.022
 goodbye-0.022
Token they
Token ID484
Feature activation+1.05e+00

Pos logit contributions
First+0.110
 brushes+0.106
 sizes+0.094
ups+0.091
 styles+0.088
Neg logit contributions
 victorious-0.134
 hero-0.133
 CA-0.128
 sorry-0.126
 devastated-0.120
Token roasted
Token ID30490
Feature activation+4.65e+00

Pos logit contributions
First+0.033
 brushes+0.032
 sizes+0.028
ups+0.027
 colors+0.026
Neg logit contributions
 victorious-0.039
 hero-0.039
 CA-0.038
 sorry-0.036
 devastated-0.035
Token marsh
Token ID22397
Feature activation+8.40e+00

Pos logit contributions
First+0.123
 brushes+0.113
ups+0.098
 sizes+0.095
 styles+0.089
Neg logit contributions
 victorious-0.198
 hero-0.195
 CA-0.191
 sorry-0.183
 hi-0.177
Tokenm
Token ID76
Feature activation+4.55e+00

Pos logit contributions
 brushes+0.294
First+0.290
 sizes+0.266
 seasons+0.261
 styles+0.260
Neg logit contributions
 victorious-0.239
 hero-0.238
eros-0.230
 sorry-0.224
 CA-0.222
Tokenallows
Token ID47205
Feature activation+4.86e+00

Pos logit contributions
First+0.137
 brushes+0.123
 sizes+0.099
 styles+0.096
ups+0.096
Neg logit contributions
 sorry-0.170
 victorious-0.170
 hero-0.166
 CA-0.161
 hi-0.155
Token over
Token ID625
Feature activation-9.45e-02

Pos logit contributions
First+0.176
 brushes+0.172
 sizes+0.154
ups+0.152
 styles+0.148
Neg logit contributions
 victorious-0.155
 hero-0.152
 CA-0.146
 sorry-0.139
 devastated-0.132
Token the
Token ID262
Feature activation+1.09e+00

Pos logit contributions
 victorious+0.002
 hero+0.002
 CA+0.002
 sorry+0.002
 goodbye+0.002
Neg logit contributions
First-0.005
 brushes-0.004
vals-0.004
ups-0.004
 sizes-0.004
Token camp
Token ID1413
Feature activation+2.18e+00

Pos logit contributions
First+0.036
 brushes+0.036
ups+0.031
 sizes+0.030
 styles+0.030
Neg logit contributions
 hero-0.040
 victorious-0.039
 CA-0.038
 sorry-0.035
 hunter-0.034
Token around
Token ID1088
Feature activation-4.04e-02

Pos logit contributions
First+0.098
 brushes+0.097
 sizes+0.087
ups+0.086
 colors+0.085
Neg logit contributions
 victorious-0.072
 hero-0.070
 CA-0.069
 sorry-0.064
 devastated-0.061
Token the
Token ID262
Feature activation+1.09e-01

Pos logit contributions
 victorious+0.001
 hero+0.001
 CA+0.001
 sorry+0.001
 devastated+0.001
Neg logit contributions
First-0.001
 brushes-0.001
 colors-0.001
 sizes-0.001
 flour-0.001
Token fireplace
Token ID44250
Feature activation+2.30e+00

Pos logit contributions
First+0.004
 brushes+0.004
 sizes+0.004
ups+0.004
 styles+0.003
Neg logit contributions
 hero-0.003
 victorious-0.003
 CA-0.003
 sorry-0.003
 devastated-0.003
Token and
Token ID290
Feature activation+1.16e-01

Pos logit contributions
First+0.090
 brushes+0.084
 sizes+0.076
ups+0.075
 seasons+0.074
Neg logit contributions
 victorious-0.070
 hero-0.069
 sorry-0.064
 devastated-0.062
 CA-0.061
Token roasted
Token ID30490
Feature activation+4.31e+00

Pos logit contributions
First+0.005
 brushes+0.004
 sizes+0.004
ups+0.004
 seasons+0.004
Neg logit contributions
 sorry-0.003
 victorious-0.003
 hero-0.003
 devastated-0.003
 CA-0.003
Token marsh
Token ID22397
Feature activation+8.21e+00

Pos logit contributions
First+0.123
 brushes+0.108
 sizes+0.098
ups+0.096
 styles+0.094
Neg logit contributions
 victorious-0.173
 hero-0.168
 CA-0.165
 sorry-0.164
 goodbye-0.154
Tokenm
Token ID76
Feature activation+4.29e+00

Pos logit contributions
 brushes+0.316
First+0.308
 sizes+0.300
 styles+0.296
 seasons+0.289
Neg logit contributions
 victorious-0.196
 sorry-0.194
 hero-0.185
eros-0.183
 devastated-0.173
Tokenallows
Token ID47205
Feature activation+4.53e+00

Pos logit contributions
First+0.146
 brushes+0.127
 sizes+0.111
vals+0.110
 flavors+0.108
Neg logit contributions
 sorry-0.153
 victorious-0.140
 devastated-0.136
 relieved-0.135
 unlucky-0.134
Token.
Token ID13
Feature activation+3.08e-01

Pos logit contributions
First+0.174
 brushes+0.171
 sizes+0.155
ups+0.152
 styles+0.151
Neg logit contributions
 victorious-0.133
 hero-0.129
 CA-0.121
 sorry-0.120
 devastated-0.112
Token Suddenly
Token ID24975
Feature activation+3.11e+00

Pos logit contributions
First+0.011
 brushes+0.011
 sizes+0.010
ups+0.010
 colors+0.009
Neg logit contributions
 victorious-0.010
 hero-0.010
 CA-0.009
 sorry-0.009
 devastated-0.009
Token,
Token ID11
Feature activation+1.06e+00

Pos logit contributions
First+0.137
 sizes+0.128
 brushes+0.128
 styles+0.122
 flavors+0.121
Neg logit contributions
 hero-0.081
 sorry-0.080
 victorious-0.077
 CA-0.072
 goodbye-0.069
Token Daisy
Token ID40355
Feature activation-1.84e-01

Pos logit contributions
 victorious+0.002
 hero+0.002
 sorry+0.002
 CA+0.001
 devastated+0.001
Neg logit contributions
First-0.002
 brushes-0.002
ups-0.002
 sizes-0.002
 styles-0.002
Token enjoyed
Token ID8359
Feature activation+3.18e+00

Pos logit contributions
 hero+0.006
 victorious+0.006
 CA+0.005
 devastated+0.005
 sorry+0.005
Neg logit contributions
First-0.007
 brushes-0.007
 sizes-0.006
 flour-0.006
 styles-0.006
Token making
Token ID1642
Feature activation+1.58e+00

Pos logit contributions
First+0.129
 brushes+0.125
 sizes+0.117
ups+0.114
 seasons+0.108
Neg logit contributions
 victorious-0.089
 sorry-0.086
 hero-0.084
 CA-0.077
 goodbye-0.074
Token popcorn
Token ID38914
Feature activation+2.66e+00

Pos logit contributions
First+0.056
 brushes+0.053
 sizes+0.047
vals+0.044
ups+0.044
Neg logit contributions
 victorious-0.052
 sorry-0.051
 hero-0.050
 hi-0.045
 CA-0.045
Token with
Token ID351
Feature activation-8.39e-01

Pos logit contributions
First+0.104
 brushes+0.100
 sizes+0.087
 seasons+0.086
ups+0.085
Neg logit contributions
 victorious-0.082
 hero-0.080
 sorry-0.072
 CA-0.067
 heroic-0.066
Token marsh
Token ID22397
Feature activation+8.21e+00

Pos logit contributions
 victorious+0.021
 sorry+0.021
 hero+0.020
 goodbye+0.019
 hi+0.018
Neg logit contributions
First-0.035
 brushes-0.032
ups-0.030
vals-0.030
 sizes-0.029
Tokenm
Token ID76
Feature activation+4.23e+00

Pos logit contributions
 brushes+0.326
First+0.312
 styles+0.286
 seasons+0.280
 sizes+0.273
Neg logit contributions
 victorious-0.210
 hero-0.200
 sorry-0.193
eros-0.190
run-0.178
Tokenallows
Token ID47205
Feature activation+4.48e+00

Pos logit contributions
First+0.144
 brushes+0.133
vals+0.105
 mats+0.104
 sizes+0.103
Neg logit contributions
 sorry-0.146
 victorious-0.137
 relieved-0.129
 lucky-0.129
 devastated-0.124
Token,
Token ID11
Feature activation+2.16e+00

Pos logit contributions
First+0.173
 brushes+0.168
 sizes+0.148
ups+0.146
 seasons+0.144
Neg logit contributions
 victorious-0.136
 hero-0.132
 sorry-0.121
 CA-0.119
 devastated-0.111
Token because
Token ID780
Feature activation+1.35e+00

Pos logit contributions
First+0.084
 brushes+0.076
ups+0.072
vals+0.066
 seasons+0.066
Neg logit contributions
 victorious-0.065
 sorry-0.062
 hero-0.060
 relieved-0.057
 goodbye-0.057
Token it
Token ID340
Feature activation+1.25e-01

Pos logit contributions
First+0.058
vals+0.053
 brushes+0.053
 necks+0.053
ups+0.051
Neg logit contributions
 sorry-0.032
 hero-0.031
 victorious-0.028
 lucky-0.028
 relieved-0.026
Token it
Token ID340
Feature activation-6.17e-01

Pos logit contributions
First+0.011
 brushes+0.011
 sizes+0.010
ups+0.009
 styles+0.009
Neg logit contributions
 victorious-0.012
 hero-0.012
 sorry-0.011
 CA-0.011
 devastated-0.010
Token was
Token ID373
Feature activation+5.67e-01

Pos logit contributions
 victorious+0.021
 hero+0.021
 CA+0.020
 sorry+0.019
 devastated+0.019
Neg logit contributions
First-0.022
 brushes-0.020
 sizes-0.018
 styles-0.018
 seasons-0.017
Token delicious
Token ID12625
Feature activation+1.93e+00

Pos logit contributions
First+0.023
 brushes+0.022
ups+0.021
 sizes+0.020
 styles+0.020
Neg logit contributions
 victorious-0.017
 sorry-0.016
 hero-0.015
 CA-0.015
 devastated-0.014
Token!
Token ID0
Feature activation-1.08e+00

Pos logit contributions
First+0.095
 brushes+0.090
ups+0.080
 seasons+0.078
 sizes+0.077
Neg logit contributions
 hero-0.042
 victorious-0.039
 sorry-0.035
 goodbye-0.031
 heroic-0.030
Token The
Token ID383
Feature activation+8.20e-01

Pos logit contributions
 victorious+0.034
 hero+0.034
 CA+0.032
 sorry+0.030
 devastated+0.029
Neg logit contributions
First-0.040
 brushes-0.039
 sizes-0.035
ups-0.033
 seasons-0.033
Token marsh
Token ID22397
Feature activation+8.17e+00

Pos logit contributions
First+0.030
 brushes+0.028
 sizes+0.026
ups+0.025
 styles+0.024
Neg logit contributions
 hero-0.028
 victorious-0.027
 sorry-0.026
 CA-0.025
 devastated-0.024
Tokenm
Token ID76
Feature activation+4.28e+00

Pos logit contributions
First+0.317
 brushes+0.309
 styles+0.283
 sizes+0.282
ups+0.279
Neg logit contributions
 victorious-0.233
 hero-0.232
 sorry-0.220
eros-0.219
 CA-0.217
Tokenallows
Token ID47205
Feature activation+4.78e+00

Pos logit contributions
First+0.117
 brushes+0.107
 sizes+0.090
 styles+0.086
ups+0.085
Neg logit contributions
 victorious-0.176
 hero-0.175
 sorry-0.171
 CA-0.170
 devastated-0.163
Token had
Token ID550
Feature activation+9.96e-01

Pos logit contributions
First+0.142
 brushes+0.136
 sizes+0.120
ups+0.117
 styles+0.114
Neg logit contributions
 victorious-0.186
 hero-0.183
 CA-0.178
 sorry-0.171
 devastated-0.168
Token made
Token ID925
Feature activation+2.64e+00

Pos logit contributions
First+0.036
 brushes+0.033
 styles+0.030
 sizes+0.029
 seasons+0.029
Neg logit contributions
 victorious-0.033
 CA-0.031
 hero-0.029
 sorry-0.028
 devastated-0.028
Token the
Token ID262
Feature activation+5.31e-01

Pos logit contributions
First+0.101
 brushes+0.096
 sizes+0.088
ups+0.086
 styles+0.084
Neg logit contributions
 victorious-0.082
 hero-0.079
 sorry-0.075
 CA-0.074
 devastated-0.069
Token They
Token ID1119
Feature activation+1.85e+00

Pos logit contributions
First+0.019
 brushes+0.018
ups+0.017
 sizes+0.017
 styles+0.016
Neg logit contributions
 victorious-0.018
 CA-0.017
 sorry-0.017
 hero-0.017
 hi-0.015
Token shared
Token ID4888
Feature activation+1.03e+00

Pos logit contributions
First+0.071
ups+0.066
 brushes+0.064
vals+0.061
 seasons+0.061
Neg logit contributions
 victorious-0.049
 hero-0.048
 CA-0.046
 relieved-0.046
 devastated-0.046
Token some
Token ID617
Feature activation+3.65e+00

Pos logit contributions
First+0.043
 brushes+0.037
 sizes+0.036
ups+0.036
vals+0.035
Neg logit contributions
 sorry-0.027
 victorious-0.027
 CA-0.026
 hi-0.026
 hero-0.026
Token cookies
Token ID14746
Feature activation+3.12e+00

Pos logit contributions
First+0.131
 brushes+0.118
 sizes+0.111
vals+0.109
ups+0.104
Neg logit contributions
 hero-0.117
 victorious-0.109
 sorry-0.103
 goodbye-0.102
 hi-0.100
Token and
Token ID290
Feature activation+7.83e-01

Pos logit contributions
First+0.128
 brushes+0.124
 sizes+0.111
 seasons+0.110
ups+0.106
Neg logit contributions
 victorious-0.090
 hero-0.089
 sorry-0.078
 goodbye-0.074
 CA-0.073
Token marsh
Token ID22397
Feature activation+8.16e+00

Pos logit contributions
First+0.029
akes+0.025
ups+0.023
ooks+0.023
vals+0.023
Neg logit contributions
 goodbye-0.022
 sorry-0.021
 CA-0.021
 hero-0.020
 victorious-0.020
Tokenm
Token ID76
Feature activation+4.57e+00

Pos logit contributions
First+0.336
 brushes+0.333
 seasons+0.302
 sizes+0.296
 styles+0.295
Neg logit contributions
 hero-0.191
 sorry-0.190
 victorious-0.190
eros-0.182
 reunited-0.165
Tokenallows
Token ID47205
Feature activation+4.92e+00

Pos logit contributions
First+0.160
 brushes+0.134
vals+0.116
 mats+0.115
 strips+0.112
Neg logit contributions
 sorry-0.166
 relieved-0.147
 victorious-0.144
 lucky-0.144
 glad-0.139
Token.
Token ID13
Feature activation+9.83e-01

Pos logit contributions
First+0.192
 brushes+0.183
 sizes+0.164
ups+0.161
 seasons+0.159
Neg logit contributions
 victorious-0.151
 hero-0.146
 sorry-0.136
 CA-0.135
 goodbye-0.124
Token They
Token ID1119
Feature activation+2.04e+00

Pos logit contributions
First+0.035
 brushes+0.035
ups+0.033
 sizes+0.032
 styles+0.031
Neg logit contributions
 victorious-0.030
 CA-0.030
 sorry-0.029
 hero-0.029
 hi-0.026
Token were
Token ID547
Feature activation+2.08e+00

Pos logit contributions
First+0.075
 brushes+0.069
ups+0.069
 seasons+0.064
 sizes+0.064
Neg logit contributions
 victorious-0.059
 hero-0.059
 CA-0.055
 sorry-0.054
 devastated-0.053
Token cheered
Token ID34767
Feature activation-8.11e-01

Pos logit contributions
First+0.053
 brushes+0.052
 sizes+0.047
ups+0.045
 colors+0.045
Neg logit contributions
 victorious-0.048
 hero-0.047
 CA-0.046
 sorry-0.043
 devastated-0.041
Token and
Token ID290
Feature activation-1.47e+00

Pos logit contributions
 victorious+0.029
 hero+0.028
 CA+0.028
 sorry+0.026
 devastated+0.025
Neg logit contributions
First-0.027
 brushes-0.026
 sizes-0.023
ups-0.023
 colors-0.022
Token began
Token ID2540
Feature activation+3.11e+00

Pos logit contributions
 hero+0.051
 victorious+0.051
 CA+0.050
 sorry+0.046
 devastated+0.046
Neg logit contributions
First-0.049
 brushes-0.048
 sizes-0.043
ups-0.041
 colors-0.041
Token to
Token ID284
Feature activation-1.31e-01

Pos logit contributions
First+0.107
 brushes+0.102
 sizes+0.093
ups+0.089
 seasons+0.088
Neg logit contributions
 victorious-0.109
 hero-0.104
 sorry-0.101
 CA-0.098
 devastated-0.095
Tokenasting
Token ID9222
Feature activation+8.39e-01

Pos logit contributions
 victorious+0.004
 hero+0.004
 sorry+0.003
 CA+0.003
 devastated+0.003
Neg logit contributions
First-0.005
 brushes-0.005
ups-0.005
 sizes-0.005
 flour-0.005
Token marsh
Token ID22397
Feature activation+8.13e+00

Pos logit contributions
First+0.032
 brushes+0.028
ups+0.026
 sizes+0.025
 seasons+0.024
Neg logit contributions
 victorious-0.028
 sorry-0.026
 hero-0.026
 CA-0.025
 goodbye-0.023
Tokenm
Token ID76
Feature activation+4.53e+00

Pos logit contributions
 brushes+0.308
First+0.299
 styles+0.277
 sizes+0.273
 seasons+0.272
Neg logit contributions
 victorious-0.223
 hero-0.202
 sorry-0.201
eros-0.198
 devastated-0.185
Tokenallows
Token ID47205
Feature activation+4.72e+00

Pos logit contributions
First+0.153
 brushes+0.130
 styles+0.109
 sizes+0.108
 flavors+0.108
Neg logit contributions
 sorry-0.161
 victorious-0.156
 hero-0.144
 devastated-0.142
 relieved-0.142
Token.
Token ID13
Feature activation+7.99e-02

Pos logit contributions
First+0.180
 brushes+0.175
 sizes+0.156
ups+0.153
 styles+0.152
Neg logit contributions
 victorious-0.145
 hero-0.139
 sorry-0.130
 CA-0.130
 devastated-0.122
Token All
Token ID1439
Feature activation+1.35e+00

Pos logit contributions
First+0.003
 brushes+0.003
 sizes+0.002
ups+0.002
 styles+0.002
Neg logit contributions
 victorious-0.003
 hero-0.003
 CA-0.003
 sorry-0.002
 devastated-0.002
Token of
Token ID286
Feature activation+3.71e+00

Pos logit contributions
First+0.053
 brushes+0.051
 sizes+0.046
ups+0.045
 styles+0.045
Neg logit contributions
 victorious-0.040
 hero-0.039
 sorry-0.035
 CA-0.035
 devastated-0.035
Tokenugs
Token ID10339
Feature activation+3.29e+00

Pos logit contributions
First+0.114
 sizes+0.108
 brushes+0.102
 colors+0.092
 seasons+0.092
Neg logit contributions
 CA-0.125
 victorious-0.123
 hero-0.123
eros-0.123
 sorry-0.122
Token with
Token ID351
Feature activation-5.69e-01

Pos logit contributions
First+0.138
 brushes+0.131
 sizes+0.125
 seasons+0.123
 styles+0.117
Neg logit contributions
 victorious-0.089
 hero-0.082
 sorry-0.075
 relieved-0.071
 devastated-0.070
Token hot
Token ID3024
Feature activation+4.78e+00

Pos logit contributions
 sorry+0.018
 victorious+0.016
 goodbye+0.016
 hero+0.016
 sad+0.015
Neg logit contributions
First-0.021
vals-0.019
ups-0.019
Lisa-0.016
akes-0.016
Token cocoa
Token ID35845
Feature activation+4.44e+00

Pos logit contributions
First+0.160
ups+0.128
 sizes+0.126
vals+0.122
Pract+0.120
Neg logit contributions
 sorry-0.142
 goodbye-0.130
 CA-0.129
 hello-0.129
 hi-0.128
Token and
Token ID290
Feature activation+4.46e-01

Pos logit contributions
First+0.213
 brushes+0.191
 seasons+0.180
 sizes+0.176
ups+0.176
Neg logit contributions
 hero-0.079
 back-0.075
 goodbye-0.072
 victorious-0.072
 sorry-0.069
Token marsh
Token ID22397
Feature activation+8.12e+00

Pos logit contributions
First+0.017
akes+0.016
ups+0.015
aters+0.015
vals+0.015
Neg logit contributions
 sorry-0.013
 relieved-0.012
 goodbye-0.011
 victorious-0.010
 sad-0.010
Tokenm
Token ID76
Feature activation+4.60e+00

Pos logit contributions
First+0.315
 brushes+0.313
 seasons+0.288
 sizes+0.287
 styles+0.285
Neg logit contributions
 sorry-0.205
 hero-0.200
 victorious-0.200
eros-0.185
 CA-0.176
Tokenallows
Token ID47205
Feature activation+5.06e+00

Pos logit contributions
First+0.172
 brushes+0.144
vals+0.131
 sizes+0.126
Pract+0.125
Neg logit contributions
 sorry-0.160
 relieved-0.138
 sad-0.135
 lucky-0.134
 unlucky-0.130
Token.
Token ID13
Feature activation+7.87e-01

Pos logit contributions
First+0.203
 brushes+0.194
 sizes+0.177
ups+0.174
 seasons+0.173
Neg logit contributions
 victorious-0.144
 hero-0.135
 sorry-0.131
 CA-0.128
 goodbye-0.119
Token They
Token ID1119
Feature activation+2.10e+00

Pos logit contributions
First+0.025
 brushes+0.024
 sizes+0.022
ups+0.022
 styles+0.021
Neg logit contributions
 victorious-0.028
 hero-0.028
 CA-0.027
 sorry-0.027
 devastated-0.025
Token smiled
Token ID13541
Feature activation+6.94e-01

Pos logit contributions
First+0.080
ups+0.075
 sizes+0.072
 brushes+0.072
 styles+0.071
Neg logit contributions
 victorious-0.055
 hero-0.054
 sorry-0.053
 relieved-0.052
 devastated-0.050
Token and
Token ID290
Feature activation-3.33e-01

Pos logit contributions
 victorious+0.036
 hero+0.036
 CA+0.033
 sorry+0.032
 devastated+0.030
Neg logit contributions
First-0.040
 brushes-0.039
 sizes-0.035
ups-0.034
 styles-0.033
Token ate
Token ID15063
Feature activation+1.88e+00

Pos logit contributions
 victorious+0.009
 sorry+0.009
 hero+0.009
 CA+0.009
 goodbye+0.009
Neg logit contributions
First-0.013
 brushes-0.012
ups-0.012
 sizes-0.011
 styles-0.011
Token hot
Token ID3024
Feature activation+4.63e+00

Pos logit contributions
First+0.069
 brushes+0.063
ups+0.060
vals+0.058
 sizes+0.056
Neg logit contributions
 sorry-0.059
 hero-0.058
 victorious-0.058
 CA-0.057
 goodbye-0.054
Token dogs
Token ID6844
Feature activation+1.36e+00

Pos logit contributions
First+0.156
 brushes+0.138
ups+0.127
 sizes+0.119
vals+0.114
Neg logit contributions
 hero-0.160
 CA-0.150
eros-0.147
 goodbye-0.145
 sorry-0.144
Token and
Token ID290
Feature activation+6.59e-01

Pos logit contributions
First+0.051
 brushes+0.051
 sizes+0.044
ups+0.043
 seasons+0.043
Neg logit contributions
 hero-0.042
 victorious-0.042
 CA-0.038
 sorry-0.038
 devastated-0.035
Token marsh
Token ID22397
Feature activation+8.08e+00

Pos logit contributions
First+0.026
ups+0.023
 brushes+0.022
akes+0.022
vals+0.021
Neg logit contributions
 sorry-0.019
 hero-0.018
 goodbye-0.017
 victorious-0.017
 CA-0.016
Tokenm
Token ID76
Feature activation+4.67e+00

Pos logit contributions
 brushes+0.296
First+0.290
 sizes+0.267
 styles+0.256
 seasons+0.250
Neg logit contributions
 hero-0.234
 sorry-0.231
 victorious-0.229
eros-0.211
 CA-0.209
Tokenallows
Token ID47205
Feature activation+5.03e+00

Pos logit contributions
First+0.141
 brushes+0.125
 sizes+0.099
vals+0.099
 flavors+0.098
Neg logit contributions
 sorry-0.186
 victorious-0.175
 hero-0.171
 unlucky-0.165
 relieved-0.165
Token.
Token ID13
Feature activation+9.45e-01

Pos logit contributions
First+0.193
 brushes+0.189
 sizes+0.168
ups+0.167
 styles+0.163
Neg logit contributions
 victorious-0.149
 hero-0.147
 CA-0.137
 sorry-0.135
 goodbye-0.123
Token They
Token ID1119
Feature activation+2.18e+00

Pos logit contributions
First+0.034
 brushes+0.034
ups+0.033
 sizes+0.033
 styles+0.031
Neg logit contributions
 sorry-0.027
 CA-0.026
 victorious-0.026
 hero-0.025
 hi-0.024
Token liked
Token ID8288
Feature activation+3.36e+00

Pos logit contributions
First+0.080
 brushes+0.077
ups+0.075
 styles+0.068
 sizes+0.068
Neg logit contributions
 hero-0.066
 victorious-0.063
 sorry-0.062
 CA-0.061
 devastated-0.059
TokenWhile
Token ID3633
Feature activation+3.70e+00

Pos logit contributions
 CA+0.043
 victorious+0.043
 hero+0.042
 devastated+0.040
 sorry+0.039
Neg logit contributions
 brushes-0.040
 sizes-0.038
 flavors-0.036
 styles-0.036
First-0.036
Token they
Token ID484
Feature activation+1.44e+00

Pos logit contributions
First+0.109
 brushes+0.106
 sizes+0.094
ups+0.091
 styles+0.088
Neg logit contributions
 victorious-0.145
 hero-0.143
 CA-0.140
 sorry-0.136
 devastated-0.130
Token were
Token ID547
Feature activation+1.76e+00

Pos logit contributions
First+0.037
 brushes+0.036
 sizes+0.031
ups+0.029
 colors+0.029
Neg logit contributions
 victorious-0.062
 hero-0.061
 CA-0.060
 sorry-0.058
 devastated-0.056
Token ro
Token ID686
Feature activation+9.74e-01

Pos logit contributions
First+0.070
 brushes+0.068
 sizes+0.062
ups+0.060
 styles+0.060
Neg logit contributions
 victorious-0.051
 hero-0.051
 CA-0.048
 sorry-0.046
 devastated-0.044
Tokenasting
Token ID9222
Feature activation+2.61e+00

Pos logit contributions
First+0.031
 brushes+0.030
 sizes+0.027
ups+0.026
 seasons+0.025
Neg logit contributions
 victorious-0.036
 hero-0.035
 CA-0.035
 sorry-0.033
 devastated-0.032
Token marsh
Token ID22397
Feature activation+7.99e+00

Pos logit contributions
First+0.077
 brushes+0.070
ups+0.064
 sizes+0.062
 styles+0.060
Neg logit contributions
 victorious-0.103
 hero-0.102
 CA-0.100
 sorry-0.099
 hi-0.091
Tokenm
Token ID76
Feature activation+4.12e+00

Pos logit contributions
 brushes+0.293
First+0.292
 styles+0.268
 sizes+0.266
 seasons+0.256
Neg logit contributions
 victorious-0.230
 sorry-0.228
 hero-0.227
eros-0.219
 CA-0.204
Tokenallows
Token ID47205
Feature activation+4.85e+00

Pos logit contributions
First+0.131
 brushes+0.123
 styles+0.105
 sizes+0.102
 flavors+0.098
Neg logit contributions
 sorry-0.151
 victorious-0.145
 hero-0.143
 unlucky-0.133
 relieved-0.133
Token,
Token ID11
Feature activation+1.71e+00

Pos logit contributions
First+0.155
 brushes+0.152
 sizes+0.135
ups+0.130
 colors+0.128
Neg logit contributions
 victorious-0.176
 hero-0.172
 CA-0.168
 sorry-0.160
 devastated-0.156
Token they
Token ID484
Feature activation+1.63e+00

Pos logit contributions
First+0.063
 brushes+0.060
 sizes+0.056
ups+0.056
 styles+0.054
Neg logit contributions
 victorious-0.053
 hero-0.053
 sorry-0.051
 CA-0.051
 devastated-0.047
Token heard
Token ID2982
Feature activation+1.02e+00

Pos logit contributions
First+0.063
 brushes+0.063
 sizes+0.058
 styles+0.058
ups+0.056
Neg logit contributions
 victorious-0.047
 hero-0.046
 CA-0.044
 sorry-0.043
 devastated-0.043
Token ate
Token ID15063
Feature activation+2.14e+00

Pos logit contributions
First+0.052
 brushes+0.051
 sizes+0.045
ups+0.044
 colors+0.043
Neg logit contributions
 victorious-0.057
 hero-0.056
 CA-0.054
 sorry-0.052
 devastated-0.050
Token some
Token ID617
Feature activation+3.75e+00

Pos logit contributions
First+0.081
 brushes+0.076
ups+0.072
 sizes+0.070
vals+0.066
Neg logit contributions
 victorious-0.067
 hero-0.065
 sorry-0.065
 CA-0.063
 hi-0.058
Token hot
Token ID3024
Feature activation+4.91e+00

Pos logit contributions
First+0.140
 brushes+0.128
 sizes+0.118
ups+0.115
vals+0.110
Neg logit contributions
 hero-0.117
 victorious-0.113
 CA-0.106
 sorry-0.102
 hi-0.099
Token dogs
Token ID6844
Feature activation+1.56e+00

Pos logit contributions
First+0.186
ups+0.150
 brushes+0.150
vals+0.135
 sizes+0.133
Neg logit contributions
eros-0.146
 hero-0.145
 CA-0.141
 goodbye-0.136
 hi-0.134
Token and
Token ID290
Feature activation+5.26e-01

Pos logit contributions
First+0.051
 brushes+0.050
 sizes+0.044
ups+0.043
 styles+0.042
Neg logit contributions
 victorious-0.056
 hero-0.055
 CA-0.053
 sorry-0.051
 devastated-0.049
Token marsh
Token ID22397
Feature activation+7.91e+00

Pos logit contributions
First+0.022
akes+0.020
ups+0.018
vals+0.018
 brushes+0.018
Neg logit contributions
 goodbye-0.013
 victorious-0.013
 sorry-0.013
 hero-0.012
 relieved-0.012
Tokenm
Token ID76
Feature activation+4.26e+00

Pos logit contributions
First+0.300
 brushes+0.297
 styles+0.267
 sizes+0.263
 seasons+0.260
Neg logit contributions
 victorious-0.222
 hero-0.222
 sorry-0.216
eros-0.204
 reunited-0.194
Tokenallows
Token ID47205
Feature activation+4.85e+00

Pos logit contributions
First+0.129
 brushes+0.112
vals+0.090
 styles+0.090
 sizes+0.090
Neg logit contributions
 sorry-0.174
 victorious-0.164
 hero-0.157
 relieved-0.157
 devastated-0.154
Token.
Token ID13
Feature activation+7.50e-01

Pos logit contributions
First+0.169
 brushes+0.164
 sizes+0.147
ups+0.143
 styles+0.140
Neg logit contributions
 victorious-0.164
 hero-0.161
 CA-0.155
 sorry-0.149
 devastated-0.143
Token They
Token ID1119
Feature activation+1.96e+00

Pos logit contributions
First+0.024
 brushes+0.023
ups+0.021
 sizes+0.021
 styles+0.020
Neg logit contributions
 victorious-0.027
 CA-0.027
 hero-0.026
 sorry-0.026
 hi-0.024
Token sang
Token ID25889
Feature activation+4.01e-02

Pos logit contributions
First+0.069
 brushes+0.066
 sizes+0.060
ups+0.060
 styles+0.058
Neg logit contributions
 victorious-0.065
 hero-0.064
 CA-0.061
 sorry-0.060
 devastated-0.059
Token They
Token ID1119
Feature activation+1.94e+00

Pos logit contributions
First+0.001
 brushes+0.001
 sizes+0.001
ups+0.001
 styles+0.001
Neg logit contributions
 victorious-0.000
 hero-0.000
 CA-0.000
 sorry-0.000
 devastated-0.000
Token were
Token ID547
Feature activation+2.21e+00

Pos logit contributions
First+0.065
 brushes+0.064
 sizes+0.058
ups+0.057
 styles+0.055
Neg logit contributions
 victorious-0.067
 hero-0.066
 CA-0.064
 sorry-0.061
 devastated-0.059
Token going
Token ID1016
Feature activation-1.18e+00

Pos logit contributions
First+0.080
 brushes+0.079
 sizes+0.071
ups+0.069
 colors+0.068
Neg logit contributions
 victorious-0.071
 hero-0.070
 CA-0.068
 sorry-0.063
 devastated-0.061
Token to
Token ID284
Feature activation-1.32e-01

Pos logit contributions
 victorious+0.039
 hero+0.039
 CA+0.035
 sorry+0.035
 devastated+0.033
Neg logit contributions
First-0.041
 brushes-0.041
 sizes-0.039
ups-0.036
 styles-0.036
Token roast
Token ID32595
Feature activation+4.56e+00

Pos logit contributions
 victorious+0.005
 hero+0.004
 CA+0.004
 sorry+0.004
 devastated+0.004
Neg logit contributions
First-0.005
 brushes-0.004
 sizes-0.004
 seasons-0.004
 flour-0.004
Token marsh
Token ID22397
Feature activation+7.87e+00

Pos logit contributions
First+0.140
 brushes+0.127
 sizes+0.115
ups+0.114
 styles+0.107
Neg logit contributions
 victorious-0.167
 CA-0.165
 hero-0.164
 sorry-0.160
 goodbye-0.151
Tokenm
Token ID76
Feature activation+4.21e+00

Pos logit contributions
 brushes+0.298
 sizes+0.283
First+0.282
 styles+0.273
 seasons+0.268
Neg logit contributions
 hero-0.202
 victorious-0.200
 sorry-0.197
eros-0.193
 CA-0.179
Tokenallows
Token ID47205
Feature activation+4.64e+00

Pos logit contributions
First+0.138
 brushes+0.125
 sizes+0.115
 styles+0.107
 flavors+0.106
Neg logit contributions
 sorry-0.144
 victorious-0.129
 relieved-0.126
 hero-0.125
 hi-0.124
Token in
Token ID287
Feature activation+6.36e-01

Pos logit contributions
First+0.179
 brushes+0.178
 sizes+0.164
ups+0.161
 styles+0.159
Neg logit contributions
 victorious-0.130
 hero-0.128
 CA-0.117
 sorry-0.117
 goodbye-0.107
Token the
Token ID262
Feature activation+1.22e+00

Pos logit contributions
First+0.029
 brushes+0.029
ups+0.028
vals+0.027
 sizes+0.026
Neg logit contributions
 victorious-0.013
 hero-0.013
 CA-0.012
 sorry-0.012
 devastated-0.010
Token fireplace
Token ID44250
Feature activation+2.45e+00

Pos logit contributions
First+0.040
 brushes+0.039
 sizes+0.036
ups+0.035
 styles+0.033
Neg logit contributions
 hero-0.043
 CA-0.042
 victorious-0.041
 sorry-0.039
 goodbye-0.037
Token because
Token ID780
Feature activation+6.17e-02

Pos logit contributions
 hero+0.008
 victorious+0.008
 sorry+0.007
 CA+0.007
 devastated+0.007
Neg logit contributions
First-0.009
 brushes-0.009
 sizes-0.008
 flour-0.008
ups-0.008
Token she
Token ID673
Feature activation+7.47e-01

Pos logit contributions
First+0.002
vals+0.002
 brushes+0.002
 sizes+0.002
ups+0.002
Neg logit contributions
 hero-0.002
 sorry-0.002
 victorious-0.002
 CA-0.002
 devastated-0.002
Token loved
Token ID6151
Feature activation+2.15e+00

Pos logit contributions
First+0.028
 brushes+0.026
 sizes+0.024
ups+0.024
 flour+0.023
Neg logit contributions
 hero-0.023
 victorious-0.023
 CA-0.023
 sorry-0.021
 devastated-0.021
Token to
Token ID284
Feature activation-3.72e-01

Pos logit contributions
First+0.085
 brushes+0.073
ups+0.070
 sizes+0.069
vals+0.068
Neg logit contributions
 hero-0.062
 CA-0.061
 sorry-0.060
 victorious-0.059
 hi-0.055
Token roast
Token ID32595
Feature activation+4.68e+00

Pos logit contributions
 hero+0.010
 victorious+0.010
 CA+0.010
 sorry+0.010
 hi+0.009
Neg logit contributions
First-0.015
 brushes-0.014
 sizes-0.014
ups-0.014
 flour-0.013
Token marsh
Token ID22397
Feature activation+7.87e+00

Pos logit contributions
First+0.140
 brushes+0.122
ups+0.108
 sizes+0.107
 styles+0.100
Neg logit contributions
 victorious-0.180
 sorry-0.176
 hero-0.176
 CA-0.175
 goodbye-0.162
Tokenm
Token ID76
Feature activation+4.23e+00

Pos logit contributions
 brushes+0.306
First+0.293
 sizes+0.281
 styles+0.270
 seasons+0.270
Neg logit contributions
 hero-0.200
 sorry-0.199
 victorious-0.196
eros-0.186
 CA-0.177
Tokenallows
Token ID47205
Feature activation+4.68e+00

Pos logit contributions
First+0.140
 brushes+0.126
 sizes+0.110
 styles+0.104
vals+0.104
Neg logit contributions
 sorry-0.149
 victorious-0.138
 hero-0.135
 CA-0.129
 relieved-0.129
Token over
Token ID625
Feature activation-1.66e-01

Pos logit contributions
First+0.178
 brushes+0.176
 sizes+0.158
ups+0.155
 styles+0.152
Neg logit contributions
 victorious-0.140
 hero-0.137
 CA-0.127
 sorry-0.126
 devastated-0.117
Token the
Token ID262
Feature activation+9.05e-01

Pos logit contributions
 victorious+0.003
 hero+0.003
 sorry+0.003
 alone+0.002
 CA+0.002
Neg logit contributions
First-0.008
 brushes-0.008
vals-0.008
 sizes-0.007
ups-0.007
Token fire
Token ID2046
Feature activation+2.81e+00

Pos logit contributions
First+0.033
 brushes+0.030
ups+0.027
 sizes+0.027
 styles+0.026
Neg logit contributions
 hero-0.031
 victorious-0.029
 CA-0.029
 sorry-0.027
 hunter-0.026
Token fireplace
Token ID44250
Feature activation+1.37e+00

Pos logit contributions
 victorious+0.027
 hero+0.027
 CA+0.026
 sorry+0.025
 devastated+0.024
Neg logit contributions
First-0.029
 brushes-0.028
 sizes-0.025
ups-0.024
 styles-0.024
Token so
Token ID523
Feature activation+1.03e+00

Pos logit contributions
First+0.055
 brushes+0.051
 sizes+0.047
 seasons+0.046
ups+0.045
Neg logit contributions
 victorious-0.041
 hero-0.040
 CA-0.036
 sorry-0.035
 devastated-0.035
Token they
Token ID484
Feature activation+1.63e+00

Pos logit contributions
First+0.039
 brushes+0.036
 sizes+0.034
 styles+0.032
 seasons+0.032
Neg logit contributions
 hero-0.033
 victorious-0.033
 sorry-0.031
 CA-0.031
 devastated-0.028
Token could
Token ID714
Feature activation+4.51e+00

Pos logit contributions
First+0.053
 brushes+0.051
 sizes+0.046
ups+0.044
 seasons+0.043
Neg logit contributions
 victorious-0.059
 hero-0.058
 CA-0.057
 sorry-0.055
 devastated-0.053
Token roast
Token ID32595
Feature activation+4.52e+00

Pos logit contributions
First+0.195
 brushes+0.181
 sizes+0.178
ups+0.173
 styles+0.167
Neg logit contributions
 victorious-0.119
 hero-0.114
 CA-0.109
 sorry-0.108
 devastated-0.099
Token marsh
Token ID22397
Feature activation+7.75e+00

Pos logit contributions
First+0.161
 brushes+0.137
ups+0.129
 sizes+0.127
 styles+0.121
Neg logit contributions
 victorious-0.152
 CA-0.144
 sorry-0.142
 hero-0.141
 goodbye-0.131
Tokenm
Token ID76
Feature activation+3.92e+00

Pos logit contributions
 brushes+0.288
First+0.288
 sizes+0.275
 styles+0.271
 seasons+0.271
Neg logit contributions
 victorious-0.194
 sorry-0.186
 hero-0.185
 CA-0.180
eros-0.178
Tokenallows
Token ID47205
Feature activation+4.44e+00

Pos logit contributions
First+0.133
 brushes+0.115
 sizes+0.099
 styles+0.096
 flavors+0.095
Neg logit contributions
 sorry-0.136
 victorious-0.129
 relieved-0.123
 CA-0.118
 hero-0.118
Token.
Token ID13
Feature activation+7.75e-01

Pos logit contributions
First+0.176
 brushes+0.173
 sizes+0.157
ups+0.153
 styles+0.153
Neg logit contributions
 victorious-0.124
 hero-0.120
 CA-0.115
 sorry-0.110
 devastated-0.101
Token They
Token ID1119
Feature activation+2.23e+00

Pos logit contributions
First+0.028
 brushes+0.028
 sizes+0.025
ups+0.025
 styles+0.024
Neg logit contributions
 victorious-0.024
 hero-0.024
 CA-0.024
 sorry-0.023
 devastated-0.021
Token were
Token ID547
Feature activation+2.54e+00

Pos logit contributions
First+0.085
 brushes+0.079
ups+0.077
 styles+0.075
 sizes+0.075
Neg logit contributions
 hero-0.061
 victorious-0.061
 devastated-0.058
 CA-0.057
 sorry-0.057
Token also
Token ID635
Feature activation+2.94e+00

Pos logit contributions
First+0.076
 brushes+0.072
 sizes+0.066
ups+0.065
 flour+0.064
Neg logit contributions
 victorious-0.075
 hero-0.074
 CA-0.070
 sorry-0.069
 devastated-0.066
Token see
Token ID766
Feature activation-7.27e-01

Pos logit contributions
First+0.101
 brushes+0.098
 sizes+0.088
ups+0.086
 flour+0.084
Neg logit contributions
 victorious-0.100
 hero-0.098
 CA-0.095
 sorry-0.092
 devastated-0.088
Token some
Token ID617
Feature activation+3.48e+00

Pos logit contributions
 victorious+0.021
 hero+0.021
 sorry+0.020
 CA+0.020
 devastated+0.018
Neg logit contributions
First-0.029
 brushes-0.027
 sizes-0.025
ups-0.024
 seasons-0.024
Token rocks
Token ID12586
Feature activation+2.55e+00

Pos logit contributions
First+0.115
 brushes+0.099
 sizes+0.090
ups+0.087
 styles+0.084
Neg logit contributions
 hero-0.130
 victorious-0.128
 sorry-0.124
 CA-0.119
 hi-0.113
Token.
Token ID13
Feature activation+7.39e-01

Pos logit contributions
First+0.099
 brushes+0.093
 sizes+0.084
ups+0.082
 styles+0.082
Neg logit contributions
 victorious-0.078
 hero-0.075
 CA-0.071
 sorry-0.070
 devastated-0.065
Token Some
Token ID2773
Feature activation+7.69e+00

Pos logit contributions
First+0.024
 brushes+0.024
 sizes+0.021
ups+0.021
 styles+0.020
Neg logit contributions
 victorious-0.026
 hero-0.026
 CA-0.025
 sorry-0.024
 devastated-0.023
Token are
Token ID389
Feature activation+3.22e+00

Pos logit contributions
First+0.244
 brushes+0.203
ups+0.184
vals+0.177
 styles+0.176
Neg logit contributions
 hero-0.274
 victorious-0.262
 CA-0.253
 sorry-0.253
 hi-0.240
Token small
Token ID1402
Feature activation+2.98e+00

Pos logit contributions
First+0.144
 brushes+0.122
akes+0.122
 styles+0.116
 occasions+0.114
Neg logit contributions
 sorry-0.082
 victorious-0.081
 hero-0.076
 sad-0.068
 unlucky-0.065
Token and
Token ID290
Feature activation-1.20e+00

Pos logit contributions
First+0.132
 brushes+0.114
ups+0.107
Mom+0.102
 styles+0.101
Neg logit contributions
 hero-0.077
 victorious-0.072
 sorry-0.069
 CA-0.066
smart-0.060
Token shiny
Token ID22441
Feature activation+1.68e+00

Pos logit contributions
 sorry+0.038
 victorious+0.037
 hero+0.036
 CA+0.035
 unlucky+0.032
Neg logit contributions
First-0.043
 brushes-0.036
ups-0.036
akes-0.036
 styles-0.034
Token.
Token ID13
Feature activation+5.32e-01

Pos logit contributions
First+0.075
 brushes+0.062
ups+0.060
vals+0.055
arge+0.055
Neg logit contributions
 hero-0.045
 victorious-0.042
 sorry-0.041
 CA-0.040
eros-0.038
Token They
Token ID1119
Feature activation+2.14e+00

Pos logit contributions
First+0.076
 brushes+0.073
 sizes+0.066
ups+0.064
 styles+0.062
Neg logit contributions
 victorious-0.076
 hero-0.075
 CA-0.073
 sorry-0.070
 devastated-0.067
Token see
Token ID766
Feature activation-6.24e-01

Pos logit contributions
First+0.075
 brushes+0.071
 sizes+0.065
ups+0.064
 flour+0.063
Neg logit contributions
 victorious-0.072
 hero-0.071
 CA-0.067
 sorry-0.066
 devastated-0.063
Token many
Token ID867
Feature activation+2.56e+00

Pos logit contributions
 victorious+0.018
 hero+0.017
 CA+0.017
 sorry+0.016
 hi+0.015
Neg logit contributions
First-0.026
 brushes-0.024
 sizes-0.022
ups-0.022
 seasons-0.021
Token flowers
Token ID12734
Feature activation+2.40e+00

Pos logit contributions
First+0.087
 brushes+0.070
vals+0.066
ups+0.065
akes+0.062
Neg logit contributions
 hero-0.094
 victorious-0.090
 sorry-0.085
 CA-0.084
 hi-0.082
Token.
Token ID13
Feature activation+7.86e-01

Pos logit contributions
First+0.097
 brushes+0.091
 sizes+0.082
ups+0.082
 styles+0.080
Neg logit contributions
 victorious-0.070
 hero-0.068
 sorry-0.064
 CA-0.064
 hi-0.058
Token Some
Token ID2773
Feature activation+7.65e+00

Pos logit contributions
First+0.024
 brushes+0.024
 sizes+0.021
ups+0.021
 styles+0.020
Neg logit contributions
 victorious-0.029
 hero-0.029
 CA-0.028
 sorry-0.027
 devastated-0.026
Token are
Token ID389
Feature activation+3.27e+00

Pos logit contributions
First+0.237
 brushes+0.191
ups+0.179
 sizes+0.175
vals+0.164
Neg logit contributions
 hero-0.279
 victorious-0.275
 CA-0.274
 sorry-0.270
 hi-0.257
Token red
Token ID2266
Feature activation+2.67e+00

Pos logit contributions
First+0.148
akes+0.125
 brushes+0.122
vals+0.118
ups+0.117
Neg logit contributions
 sorry-0.084
 victorious-0.079
 hero-0.074
 sad-0.070
 lucky-0.066
Token,
Token ID11
Feature activation+1.50e+00

Pos logit contributions
First+0.130
ups+0.104
Matt+0.099
Jake+0.098
 brushes+0.098
Neg logit contributions
 CA-0.059
 hero-0.057
 hello-0.054
eros-0.053
 sorry-0.053
Token some
Token ID617
Feature activation+4.49e+00

Pos logit contributions
First+0.062
ups+0.056
 brushes+0.052
akes+0.051
 occasions+0.050
Neg logit contributions
 sorry-0.042
 victorious-0.037
 hello-0.036
 hi-0.035
 hero-0.035
Token are
Token ID389
Feature activation+2.63e+00

Pos logit contributions
First+0.173
 brushes+0.147
ups+0.142
 sizes+0.134
 styles+0.131
Neg logit contributions
 victorious-0.134
 hero-0.133
 sorry-0.132
 CA-0.129
 hi-0.118
Token
Token ID198
Feature activation-2.02e+00

Pos logit contributions
 victorious+0.041
 hero+0.040
 CA+0.039
 sorry+0.036
 devastated+0.035
Neg logit contributions
First-0.056
 brushes-0.054
 sizes-0.049
ups-0.048
 flour-0.047
TokenThe
Token ID464
Feature activation-4.24e-01

Pos logit contributions
 victorious+0.084
 hero+0.081
 CA+0.079
 devastated+0.077
 sorry+0.077
Neg logit contributions
 brushes-0.055
First-0.051
 sizes-0.050
 styles-0.046
 colors-0.045
Token bad
Token ID2089
Feature activation-1.72e+00

Pos logit contributions
 victorious+0.014
 hero+0.014
 CA+0.013
 sorry+0.013
 devastated+0.012
Neg logit contributions
First-0.015
 brushes-0.015
 sizes-0.013
ups-0.013
 styles-0.013
Token ending
Token ID7464
Feature activation-2.10e+00

Pos logit contributions
 hero+0.050
 victorious+0.050
 CA+0.048
 sorry+0.047
 devastated+0.043
Neg logit contributions
First-0.070
 brushes-0.065
 sizes-0.060
ups-0.058
 flour-0.057
Token was
Token ID373
Feature activation-1.60e+00

Pos logit contributions
 victorious+0.076
 hero+0.075
 CA+0.073
 sorry+0.070
 devastated+0.068
Neg logit contributions
First-0.066
 brushes-0.065
 sizes-0.058
ups-0.057
 colors-0.055
Token fore
Token ID1674
Feature activation-7.41e+00

Pos logit contributions
 CA+0.056
 hero+0.056
 victorious+0.054
eros+0.050
 sorry+0.048
Neg logit contributions
First-0.050
 brushes-0.048
ups-0.045
 sizes-0.044
 flour-0.044
Tokenshadow
Token ID19106
Feature activation-2.12e+00

Pos logit contributions
 CA+0.175
 hero+0.168
 victorious+0.161
 sorry+0.152
eros+0.140
Neg logit contributions
First-0.339
ups-0.319
 brushes-0.311
 sizes-0.285
 colors-0.278
Tokened
Token ID276
Feature activation-3.94e+00

Pos logit contributions
 victorious+0.070
 hero+0.069
 CA+0.067
 sorry+0.064
 devastated+0.061
Neg logit contributions
First-0.075
 brushes-0.073
 sizes-0.066
ups-0.064
 styles-0.063
Token by
Token ID416
Feature activation-2.20e+00

Pos logit contributions
 victorious+0.126
 hero+0.123
 CA+0.117
 sorry+0.114
 devastated+0.108
Neg logit contributions
First-0.147
 brushes-0.142
 sizes-0.128
ups-0.123
 styles-0.122
Token Maggie
Token ID33014
Feature activation-1.15e+00

Pos logit contributions
 victorious+0.075
 hero+0.074
 CA+0.072
 sorry+0.068
 devastated+0.066
Neg logit contributions
First-0.075
 brushes-0.074
 sizes-0.066
ups-0.064
 colors-0.063
Token's
Token ID338
Feature activation+4.47e-01

Pos logit contributions
 victorious+0.040
 hero+0.040
 CA+0.039
 sorry+0.037
 devastated+0.035
Neg logit contributions
First-0.038
 brushes-0.037
 sizes-0.033
ups-0.033
 colors-0.031
Token lived
Token ID5615
Feature activation-5.00e-01

Pos logit contributions
 victorious+0.026
 hero+0.026
 CA+0.025
 sorry+0.024
 devastated+0.023
Neg logit contributions
First-0.030
 brushes-0.029
 sizes-0.026
ups-0.026
 styles-0.025
Token a
Token ID257
Feature activation-9.41e-01

Pos logit contributions
 hero+0.018
 CA+0.018
 victorious+0.018
 sorry+0.017
eros+0.017
Neg logit contributions
First-0.016
 brushes-0.014
 flour-0.013
ups-0.012
 colors-0.012
Token princess
Token ID21752
Feature activation-1.85e+00

Pos logit contributions
 victorious+0.024
 CA+0.024
 hero+0.021
 devastated+0.021
eros+0.020
Neg logit contributions
 brushes-0.037
First-0.037
 seasons-0.034
 flour-0.034
 colors-0.034
Token called
Token ID1444
Feature activation-5.11e-01

Pos logit contributions
 victorious+0.050
 hero+0.048
 CA+0.047
 sorry+0.045
 devastated+0.042
Neg logit contributions
First-0.076
 brushes-0.075
 sizes-0.069
ups-0.067
 colors-0.066
Token R
Token ID371
Feature activation-5.64e-01

Pos logit contributions
 victorious+0.017
 hero+0.017
 CA+0.016
 devastated+0.015
 sorry+0.015
Neg logit contributions
First-0.018
 brushes-0.017
 sizes-0.015
ups-0.015
 flour-0.015
Tokenyle
Token ID2349
Feature activation-6.82e+00

Pos logit contributions
 victorious+0.022
 hero+0.021
 sorry+0.021
 CA+0.020
 devastated+0.020
Neg logit contributions
First-0.016
 brushes-0.015
ups-0.014
 seasons-0.014
 flour-0.013
Tokene
Token ID68
Feature activation-1.54e+00

Pos logit contributions
 victorious+0.261
 hero+0.248
 CA+0.246
 sorry+0.246
 devastated+0.233
Neg logit contributions
First-0.203
 brushes-0.192
ups-0.176
 sizes-0.172
 colors-0.163
Token.
Token ID13
Feature activation-9.32e-01

Pos logit contributions
 victorious+0.058
 hero+0.056
 CA+0.055
 sorry+0.053
 devastated+0.051
Neg logit contributions
First-0.046
 brushes-0.046
 sizes-0.041
ups-0.040
 colors-0.039
Token Every
Token ID3887
Feature activation+1.44e+00

Pos logit contributions
 victorious+0.033
 hero+0.032
 CA+0.031
 sorry+0.030
 devastated+0.029
Neg logit contributions
First-0.032
 brushes-0.031
 sizes-0.027
ups-0.026
 seasons-0.026
Token day
Token ID1110
Feature activation+6.57e-01

Pos logit contributions
First+0.032
 brushes+0.030
 seasons+0.025
 sizes+0.025
 colors+0.024
Neg logit contributions
 victorious-0.067
 hero-0.064
 CA-0.064
 sorry-0.063
 devastated-0.061
Token,
Token ID11
Feature activation+8.43e-01

Pos logit contributions
 brushes+0.026
First+0.025
 sizes+0.023
ups+0.022
 styles+0.022
Neg logit contributions
 hero-0.019
 victorious-0.019
 CA-0.018
 sorry-0.018
 hi-0.017
Token mum
Token ID25682
Feature activation+2.93e+00

Pos logit contributions
First+0.096
 brushes+0.092
 sizes+0.084
ups+0.081
 styles+0.078
Neg logit contributions
 victorious-0.087
 hero-0.087
 CA-0.082
 sorry-0.081
 devastated-0.077
Token said
Token ID531
Feature activation-9.71e-01

Pos logit contributions
First+0.101
 brushes+0.098
 sizes+0.088
ups+0.084
 seasons+0.084
Neg logit contributions
 victorious-0.100
 hero-0.098
 CA-0.094
 sorry-0.091
 devastated-0.089
Token,
Token ID11
Feature activation+1.17e-01

Pos logit contributions
 victorious+0.034
 hero+0.034
 CA+0.033
 sorry+0.030
 devastated+0.030
Neg logit contributions
First-0.032
 brushes-0.031
 sizes-0.028
ups-0.027
 flour-0.027
Token "
Token ID366
Feature activation+5.89e-01

Pos logit contributions
First+0.005
 brushes+0.005
 sizes+0.004
ups+0.004
 flour+0.004
Neg logit contributions
 victorious-0.003
 hero-0.003
 CA-0.003
 devastated-0.003
 sorry-0.003
TokenOh
Token ID5812
Feature activation-2.38e+00

Pos logit contributions
 brushes+0.021
 sizes+0.018
First+0.018
 styles+0.017
 sheets+0.017
Neg logit contributions
 victorious-0.019
 CA-0.018
 hero-0.018
 devastated-0.017
 Start-0.016
Token look
Token ID804
Feature activation-6.71e+00

Pos logit contributions
 victorious+0.091
 hero+0.087
 CA+0.087
 devastated+0.081
 sorry+0.078
Neg logit contributions
First-0.072
 brushes-0.072
 sizes-0.062
 flour-0.062
 colors-0.060
Token,
Token ID11
Feature activation-9.35e-01

Pos logit contributions
 victorious+0.245
 hero+0.243
 CA+0.242
 devastated+0.222
 sorry+0.212
Neg logit contributions
 brushes-0.206
First-0.203
 sizes-0.180
 flour-0.178
 colors-0.177
Token he
Token ID339
Feature activation-8.38e-01

Pos logit contributions
 victorious+0.032
 hero+0.031
 CA+0.030
 devastated+0.028
 sorry+0.028
Neg logit contributions
First-0.032
 brushes-0.032
 sizes-0.028
 flour-0.028
 seasons-0.027
Token likes
Token ID7832
Feature activation+1.68e+00

Pos logit contributions
 victorious+0.030
 hero+0.029
 CA+0.029
 sorry+0.027
 devastated+0.027
Neg logit contributions
 brushes-0.027
First-0.026
 sizes-0.023
ups-0.023
 colors-0.023
Token you
Token ID345
Feature activation-1.35e+00

Pos logit contributions
First+0.067
 brushes+0.062
 sizes+0.055
ups+0.054
 styles+0.052
Neg logit contributions
 hero-0.049
 victorious-0.049
 CA-0.047
 sorry-0.046
 hi-0.043
Token.
Token ID13
Feature activation-1.22e+00

Pos logit contributions
 victorious+0.044
 hero+0.043
 sorry+0.041
 CA+0.041
 devastated+0.038
Neg logit contributions
First-0.049
 brushes-0.048
 sizes-0.043
ups-0.043
 flour-0.041
Token he
Token ID339
Feature activation+5.08e-01

Pos logit contributions
First+0.022
 brushes+0.022
ups+0.020
 sizes+0.019
 flour+0.019
Neg logit contributions
 hero-0.019
 victorious-0.019
 sorry-0.018
 CA-0.018
 devastated-0.016
Token didn
Token ID1422
Feature activation+6.49e-01

Pos logit contributions
First+0.015
 brushes+0.015
 sizes+0.013
ups+0.013
 colors+0.012
Neg logit contributions
 victorious-0.019
 hero-0.019
 CA-0.019
 sorry-0.018
 devastated-0.017
Token't
Token ID470
Feature activation+3.09e-01

Pos logit contributions
First+0.019
 brushes+0.018
ups+0.016
 sizes+0.015
 flour+0.015
Neg logit contributions
 hero-0.025
 victorious-0.025
 CA-0.024
 sorry-0.023
 devastated-0.022
Token have
Token ID423
Feature activation-1.74e+00

Pos logit contributions
First+0.010
 brushes+0.010
 sizes+0.009
ups+0.009
 seasons+0.008
Neg logit contributions
 victorious-0.011
 CA-0.011
 hero-0.010
 sorry-0.010
 devastated-0.010
Token an
Token ID281
Feature activation-1.47e+00

Pos logit contributions
 victorious+0.061
 hero+0.059
 CA+0.057
 sorry+0.055
 devastated+0.054
Neg logit contributions
First-0.058
 brushes-0.057
 sizes-0.051
ups-0.051
 colors-0.050
Token invitation
Token ID17023
Feature activation-6.67e+00

Pos logit contributions
 victorious+0.060
 CA+0.058
 hero+0.055
 devastated+0.051
 Terr+0.051
Neg logit contributions
 brushes-0.038
First-0.038
 colors-0.036
 styles-0.033
 flour-0.033
Token.
Token ID13
Feature activation-6.19e-01

Pos logit contributions
 victorious+0.281
 CA+0.277
 hero+0.272
 sorry+0.263
 devastated+0.262
Neg logit contributions
First-0.167
 brushes-0.161
ups-0.147
 sizes-0.135
 colors-0.133
Token He
Token ID679
Feature activation-2.63e-01

Pos logit contributions
 victorious+0.020
 hero+0.020
 CA+0.019
 sorry+0.019
 devastated+0.018
Neg logit contributions
First-0.023
 brushes-0.021
 sizes-0.019
 colors-0.018
ups-0.018
Token asked
Token ID1965
Feature activation+5.44e-01

Pos logit contributions
 victorious+0.010
 hero+0.010
 CA+0.009
 sorry+0.009
 devastated+0.009
Neg logit contributions
First-0.008
 brushes-0.008
 sizes-0.007
ups-0.007
 colors-0.007
Token his
Token ID465
Feature activation-2.47e+00

Pos logit contributions
First+0.018
 brushes+0.017
 colors+0.016
 flour+0.016
 seasons+0.015
Neg logit contributions
 victorious-0.019
 CA-0.019
 devastated-0.018
 hero-0.018
eros-0.017
Token friend
Token ID1545
Feature activation-1.33e+00

Pos logit contributions
 CA+0.094
 victorious+0.093
 devastated+0.086
 hero+0.082
 sorry+0.081
Neg logit contributions
 brushes-0.073
First-0.070
 colors-0.067
 shapes-0.064
 sizes-0.063
Token world
Token ID995
Feature activation-2.28e+00

Pos logit contributions
First+0.012
 brushes+0.011
 sizes+0.010
ups+0.010
 styles+0.010
Neg logit contributions
 hero-0.010
 victorious-0.010
 CA-0.009
 sorry-0.009
 devastated-0.008
Token and
Token ID290
Feature activation-2.66e-01

Pos logit contributions
 victorious+0.081
 CA+0.080
 hero+0.080
 sorry+0.077
 devastated+0.074
Neg logit contributions
First-0.071
 brushes-0.070
 sizes-0.065
ups-0.062
 colors-0.060
Token vowed
Token ID19982
Feature activation-1.89e+00

Pos logit contributions
 victorious+0.009
 hero+0.008
 CA+0.008
 sorry+0.008
 devastated+0.007
Neg logit contributions
First-0.010
 brushes-0.010
 sizes-0.008
 flour-0.008
ups-0.008
Token to
Token ID284
Feature activation-1.57e+00

Pos logit contributions
 victorious+0.070
 hero+0.069
 CA+0.066
 sorry+0.064
 devastated+0.061
Neg logit contributions
First-0.060
 brushes-0.059
 sizes-0.053
 flour-0.050
 seasons-0.049
Token never
Token ID1239
Feature activation-2.60e+00

Pos logit contributions
 victorious+0.054
 hero+0.053
 CA+0.050
 sorry+0.049
 devastated+0.047
Neg logit contributions
First-0.055
 brushes-0.053
 sizes-0.048
ups-0.047
 flour-0.046
Token look
Token ID804
Feature activation-6.63e+00

Pos logit contributions
 victorious+0.087
 hero+0.085
 CA+0.085
eros+0.078
 sorry+0.078
Neg logit contributions
First-0.088
 brushes-0.087
 sizes-0.079
ups-0.076
 colors-0.076
Token back
Token ID736
Feature activation-3.68e+00

Pos logit contributions
 CA+0.214
 hero+0.193
 victorious+0.188
eros+0.186
 devastated+0.178
Neg logit contributions
First-0.235
 brushes-0.232
ups-0.218
 sizes-0.215
 colors-0.212
Token.
Token ID13
Feature activation-7.30e-01

Pos logit contributions
 victorious+0.110
 hero+0.108
 CA+0.104
 sorry+0.099
 devastated+0.095
Neg logit contributions
First-0.143
 brushes-0.139
 sizes-0.127
ups-0.123
 flour-0.121
Token<|endoftext|>
Token ID50256
Feature activation+1.80e+00

Pos logit contributions
 hero+0.027
 victorious+0.026
 sorry+0.026
 CA+0.026
 devastated+0.025
Neg logit contributions
First-0.023
 brushes-0.019
 sizes-0.018
ups-0.016
 seasons-0.016
TokenOnce
Token ID7454
Feature activation-7.75e-01

Pos logit contributions
First+0.066
 brushes+0.062
 sizes+0.056
ups+0.054
 seasons+0.054
Neg logit contributions
 victorious-0.058
 hero-0.057
 CA-0.054
 sorry-0.054
 devastated-0.051
Token upon
Token ID2402
Feature activation+6.87e-01

Pos logit contributions
 victorious+0.035
 CA+0.034
 hero+0.034
 sorry+0.031
 devastated+0.031
Neg logit contributions
 brushes-0.018
First-0.018
 colors-0.014
 sizes-0.014
ups-0.014
Token day
Token ID1110
Feature activation-4.46e-01

Pos logit contributions
First+0.082
 brushes+0.073
 seasons+0.068
 colors+0.063
 sizes+0.062
Neg logit contributions
 victorious-0.209
 CA-0.201
 hero-0.194
 devastated-0.192
 sorry-0.191
Token,
Token ID11
Feature activation-1.62e-01

Pos logit contributions
 victorious+0.016
 hero+0.016
 CA+0.016
 sorry+0.014
 devastated+0.014
Neg logit contributions
First-0.014
 brushes-0.014
 sizes-0.012
ups-0.012
 colors-0.012
Token Benny
Token ID44275
Feature activation+7.40e-01

Pos logit contributions
 victorious+0.005
 hero+0.004
 CA+0.004
 devastated+0.004
 sorry+0.004
Neg logit contributions
First-0.007
 brushes-0.006
 sizes-0.006
 colors-0.006
 flour-0.006
Token received
Token ID2722
Feature activation-2.74e+00

Pos logit contributions
First+0.027
 brushes+0.027
 sizes+0.025
 styles+0.024
 flour+0.023
Neg logit contributions
 victorious-0.023
 hero-0.023
 CA-0.022
 sorry-0.020
 devastated-0.020
Token an
Token ID281
Feature activation-1.58e+00

Pos logit contributions
 victorious+0.086
 hero+0.084
 CA+0.081
 sorry+0.077
 devastated+0.074
Neg logit contributions
First-0.103
 brushes-0.100
 sizes-0.091
ups-0.088
 colors-0.086
Token invitation
Token ID17023
Feature activation-6.61e+00

Pos logit contributions
 victorious+0.056
 CA+0.053
 hero+0.049
 devastated+0.048
 sorry+0.045
Neg logit contributions
First-0.050
 brushes-0.049
 colors-0.046
 flour-0.046
 sizes-0.045
Token to
Token ID284
Feature activation-1.15e+00

Pos logit contributions
 victorious+0.296
 CA+0.286
 hero+0.284
 sorry+0.276
 devastated+0.275
Neg logit contributions
First-0.144
 brushes-0.140
ups-0.120
 colors-0.110
 flour-0.109
Token a
Token ID257
Feature activation-1.16e+00

Pos logit contributions
 victorious+0.042
 devastated+0.042
 CA+0.041
 hero+0.041
eros+0.039
Neg logit contributions
First-0.032
 brushes-0.032
 color-0.029
 colors-0.029
 shapes-0.027
Token party
Token ID2151
Feature activation-3.89e+00

Pos logit contributions
 victorious+0.034
 CA+0.033
eros+0.032
 hero+0.030
 devastated+0.030
Neg logit contributions
First-0.041
 brushes-0.040
 flour-0.038
 colors-0.036
 seasons-0.036
Token.
Token ID13
Feature activation-5.34e-01

Pos logit contributions
 victorious+0.156
 CA+0.152
 hero+0.151
 sorry+0.145
 devastated+0.144
Neg logit contributions
First-0.108
 brushes-0.106
ups-0.093
 sizes-0.090
 colors-0.089
Token He
Token ID679
Feature activation+1.59e-01

Pos logit contributions
 victorious+0.019
 hero+0.018
 CA+0.018
 devastated+0.017
eros+0.017
Neg logit contributions
First-0.018
 brushes-0.016
 colors-0.014
 sizes-0.014
 shapes-0.014
Token her
Token ID607
Feature activation-2.88e+00

Pos logit contributions
First+0.004
 brushes+0.004
 sizes+0.004
ups+0.004
 colors+0.004
Neg logit contributions
 hero-0.004
 victorious-0.003
 CA-0.003
 sorry-0.003
 hi-0.003
Token why
Token ID1521
Feature activation-1.14e+00

Pos logit contributions
 victorious+0.100
 CA+0.098
 hero+0.095
 devastated+0.089
 sorry+0.087
Neg logit contributions
First-0.097
 brushes-0.096
 sizes-0.085
ups-0.085
 flour-0.082
Token she
Token ID673
Feature activation-6.35e-01

Pos logit contributions
 hero+0.039
 victorious+0.039
 CA+0.038
 sorry+0.038
 devastated+0.034
Neg logit contributions
First-0.040
 brushes-0.037
 sizes-0.033
ups-0.032
 flour-0.031
Token didn
Token ID1422
Feature activation+1.62e-01

Pos logit contributions
 victorious+0.027
 hero+0.026
 CA+0.025
 sorry+0.024
 devastated+0.024
Neg logit contributions
 brushes-0.016
First-0.016
 sizes-0.014
 colors-0.014
 flavors-0.014
Token't
Token ID470
Feature activation-1.29e-01

Pos logit contributions
First+0.005
 brushes+0.005
 sizes+0.005
ups+0.004
 colors+0.004
Neg logit contributions
 victorious-0.006
 hero-0.006
 CA-0.005
 sorry-0.005
 devastated-0.005
Token look
Token ID804
Feature activation-6.60e+00

Pos logit contributions
 victorious+0.004
 hero+0.004
 CA+0.004
 sorry+0.004
 devastated+0.004
Neg logit contributions
First-0.004
 brushes-0.004
 sizes-0.004
ups-0.004
 colors-0.004
Token happy
Token ID3772
Feature activation-2.34e+00

Pos logit contributions
 CA+0.231
 victorious+0.221
 hero+0.221
eros+0.206
 devastated+0.201
Neg logit contributions
 brushes-0.210
First-0.203
 colors-0.194
 sizes-0.193
 styles-0.184
Token.
Token ID13
Feature activation-5.66e-01

Pos logit contributions
 hero+0.070
 victorious+0.070
 sorry+0.067
 CA+0.065
 hi+0.060
Neg logit contributions
First-0.093
 brushes-0.086
 sizes-0.075
 flour-0.075
ups-0.075
Token Lucy
Token ID22162
Feature activation+1.04e+00

Pos logit contributions
 victorious+0.016
 hero+0.016
 CA+0.015
 sorry+0.014
 devastated+0.014
Neg logit contributions
First-0.023
 brushes-0.022
 sizes-0.020
ups-0.020
 colors-0.019
Token said
Token ID531
Feature activation-5.55e-01

Pos logit contributions
First+0.031
 brushes+0.030
 sizes+0.027
ups+0.026
 colors+0.025
Neg logit contributions
 victorious-0.040
 hero-0.039
 CA-0.038
 sorry-0.036
 devastated-0.035
Token:
Token ID25
Feature activation-3.03e-01

Pos logit contributions
 victorious+0.019
 hero+0.019
 CA+0.019
 sorry+0.017
 devastated+0.017
Neg logit contributions
First-0.018
 brushes-0.018
 sizes-0.016
ups-0.016
 flour-0.015
Token it
Token ID340
Feature activation-1.22e+00

Pos logit contributions
First+0.134
 brushes+0.118
 seasons+0.111
vals+0.110
 sizes+0.109
Neg logit contributions
 victorious-0.070
 hero-0.065
 sorry-0.057
 hi-0.054
 Start-0.053
Token around
Token ID1088
Feature activation+4.01e-01

Pos logit contributions
 victorious+0.038
 hero+0.036
 sorry+0.034
 CA+0.034
 devastated+0.032
Neg logit contributions
First-0.046
 brushes-0.044
 sizes-0.040
 seasons-0.039
ups-0.039
Token.
Token ID13
Feature activation+3.46e-03

Pos logit contributions
First+0.017
 brushes+0.015
ups+0.014
 sizes+0.013
 seasons+0.013
Neg logit contributions
 victorious-0.012
 hero-0.011
 sorry-0.011
 CA-0.010
 devastated-0.010
Token "
Token ID366
Feature activation+2.18e+00

Pos logit contributions
First+0.000
 brushes+0.000
 sizes+0.000
ups+0.000
 styles+0.000
Neg logit contributions
 victorious-0.000
 hero-0.000
 CA-0.000
 sorry-0.000
 devastated-0.000
TokenOh
Token ID5812
Feature activation-1.78e+00

Pos logit contributions
 brushes+0.077
ups+0.072
 pots+0.071
 sizes+0.071
 sheets+0.070
Neg logit contributions
 CA-0.063
 victorious-0.061
 hero-0.058
 Start-0.054
 Looking-0.053
Token look
Token ID804
Feature activation-6.59e+00

Pos logit contributions
 victorious+0.058
 hero+0.057
 CA+0.056
 sorry+0.052
 devastated+0.051
Neg logit contributions
First-0.063
 brushes-0.062
 sizes-0.056
ups-0.054
 colors-0.053
Token,
Token ID11
Feature activation-9.21e-01

Pos logit contributions
 victorious+0.242
 hero+0.240
 CA+0.238
 devastated+0.220
 sorry+0.212
Neg logit contributions
 brushes-0.205
First-0.202
 sizes-0.177
 flour-0.175
 colors-0.174
Token Amy
Token ID14235
Feature activation-4.62e-01

Pos logit contributions
 hero+0.028
 victorious+0.028
 sorry+0.026
 CA+0.026
 hi+0.024
Neg logit contributions
First-0.036
 brushes-0.034
 sizes-0.031
ups-0.031
 styles-0.030
Token,
Token ID11
Feature activation-6.27e-02

Pos logit contributions
 victorious+0.014
 hero+0.014
 sorry+0.014
 CA+0.013
eros+0.012
Neg logit contributions
First-0.018
 brushes-0.017
 sizes-0.015
 seasons-0.015
ups-0.015
Token your
Token ID534
Feature activation+3.10e-01

Pos logit contributions
 hero+0.002
 victorious+0.002
 CA+0.001
 sorry+0.001
 hi+0.001
Neg logit contributions
First-0.003
 brushes-0.003
 sizes-0.002
ups-0.002
 seasons-0.002
Token heart
Token ID2612
Feature activation-1.32e+00

Pos logit contributions
First+0.011
 brushes+0.011
 sizes+0.010
ups+0.009
 seasons+0.009
Neg logit contributions
 victorious-0.010
 hero-0.010
 CA-0.010
 sorry-0.010
 devastated-0.009
Token ot
Token ID30972
Feature activation-1.50e+00

Pos logit contributions
 victorious+0.024
 CA+0.022
 devastated+0.021
 hero+0.020
 sorry+0.019
Neg logit contributions
 brushes-0.035
First-0.034
 colors-0.032
 sizes-0.032
 seasons-0.032
Tokenter
Token ID353
Feature activation-3.02e+00

Pos logit contributions
 victorious+0.045
 CA+0.042
 hero+0.040
 devastated+0.039
 sorry+0.039
Neg logit contributions
First-0.057
 brushes-0.056
ups-0.052
 sizes-0.052
 colors-0.050
Token who
Token ID508
Feature activation-3.35e+00

Pos logit contributions
 hero+0.063
 victorious+0.060
 CA+0.056
 sorry+0.054
 devastated+0.050
Neg logit contributions
First-0.145
 brushes-0.141
 sizes-0.128
 flour-0.126
ups-0.126
Token didn
Token ID1422
Feature activation+2.11e-02

Pos logit contributions
 victorious+0.112
 hero+0.110
 CA+0.107
 sorry+0.102
 devastated+0.098
Neg logit contributions
First-0.117
 brushes-0.115
 sizes-0.104
ups-0.100
 colors-0.098
Token't
Token ID470
Feature activation-3.93e-01

Pos logit contributions
First+0.001
 brushes+0.001
ups+0.000
 sizes+0.000
 seasons+0.000
Neg logit contributions
 hero-0.001
 victorious-0.001
 CA-0.001
 sorry-0.001
eros-0.001
Token look
Token ID804
Feature activation-6.59e+00

Pos logit contributions
 victorious+0.016
 hero+0.016
 CA+0.015
 sorry+0.015
 devastated+0.014
Neg logit contributions
First-0.011
 brushes-0.011
 sizes-0.009
 seasons-0.009
 colors-0.009
Token before
Token ID878
Feature activation+2.01e+00

Pos logit contributions
 CA+0.229
 victorious+0.208
 hero+0.195
 devastated+0.194
eros+0.189
Neg logit contributions
 brushes-0.216
 sizes-0.205
First-0.204
 colors-0.204
 shapes-0.201
Token he
Token ID339
Feature activation-1.14e+00

Pos logit contributions
First+0.077
 brushes+0.075
ups+0.065
 sizes+0.065
 styles+0.064
Neg logit contributions
 victorious-0.062
 hero-0.060
 CA-0.058
 sorry-0.058
 devastated-0.053
Token leapt
Token ID43713
Feature activation-5.72e-01

Pos logit contributions
 victorious+0.044
 hero+0.043
 CA+0.042
 sorry+0.041
 devastated+0.039
Neg logit contributions
First-0.033
 brushes-0.032
 sizes-0.029
ups-0.028
 colors-0.028
Token.
Token ID13
Feature activation-1.18e+00

Pos logit contributions
 victorious+0.018
 hero+0.018
 CA+0.018
 sorry+0.017
 devastated+0.017
Neg logit contributions
First-0.020
 brushes-0.019
 sizes-0.018
ups-0.017
 colors-0.017
Token He
Token ID679
Feature activation-9.96e-02

Pos logit contributions
 victorious+0.034
 hero+0.033
 CA+0.032
 sorry+0.030
 devastated+0.029
Neg logit contributions
First-0.047
 brushes-0.046
 sizes-0.042
ups-0.041
 seasons-0.040
Token
Token ID198
Feature activation-1.71e+00

Pos logit contributions
 hero+0.035
 victorious+0.034
 CA+0.033
 sorry+0.032
eros+0.030
Neg logit contributions
First-0.038
 brushes-0.034
ups-0.030
 flour-0.029
 sizes-0.029
TokenThe
Token ID464
Feature activation-1.91e-02

Pos logit contributions
 victorious+0.065
 hero+0.064
 CA+0.062
 sorry+0.059
 devastated+0.058
Neg logit contributions
First-0.052
 brushes-0.051
 sizes-0.045
ups-0.044
 colors-0.043
Token girl
Token ID2576
Feature activation-2.27e+00

Pos logit contributions
 CA+0.001
 victorious+0.001
 devastated+0.001
 hero+0.001
 sorry+0.001
Neg logit contributions
 brushes-0.001
First-0.001
 colors-0.001
 shapes-0.001
 flour-0.001
Token had
Token ID550
Feature activation+2.35e-01

Pos logit contributions
 victorious+0.082
 hero+0.081
 CA+0.079
 sorry+0.075
 devastated+0.072
Neg logit contributions
First-0.073
 brushes-0.071
 sizes-0.063
ups-0.061
 colors-0.060
Token been
Token ID587
Feature activation-1.29e+00

Pos logit contributions
First+0.008
 brushes+0.008
 sizes+0.007
ups+0.007
 seasons+0.007
Neg logit contributions
 victorious-0.008
 hero-0.008
 CA-0.007
 sorry-0.007
 devastated-0.007
Token fore
Token ID1674
Feature activation-6.59e+00

Pos logit contributions
 CA+0.043
 hero+0.041
 victorious+0.039
eros+0.037
 sorry+0.035
Neg logit contributions
First-0.044
 brushes-0.043
ups-0.040
 colors-0.039
 flour-0.038
Tokenshadow
Token ID19106
Feature activation-1.52e+00

Pos logit contributions
 CA+0.160
 hero+0.154
 victorious+0.144
 sorry+0.139
eros+0.127
Neg logit contributions
First-0.301
ups-0.280
 brushes-0.278
 sizes-0.255
 colors-0.252
Tokened
Token ID276
Feature activation-3.43e+00

Pos logit contributions
 victorious+0.051
 hero+0.051
 CA+0.051
 sorry+0.048
 devastated+0.046
Neg logit contributions
First-0.051
 brushes-0.050
ups-0.045
 sizes-0.044
 colors-0.043
Token.
Token ID13
Feature activation-7.66e-01

Pos logit contributions
 victorious+0.115
 hero+0.113
 CA+0.111
 sorry+0.104
 devastated+0.101
Neg logit contributions
First-0.119
 brushes-0.117
 sizes-0.105
ups-0.102
 colors-0.100
Token Even
Token ID3412
Feature activation+1.75e+00

Pos logit contributions
 hero+0.030
 victorious+0.030
 CA+0.029
 sorry+0.028
 devastated+0.027
Neg logit contributions
First-0.022
 brushes-0.021
 sizes-0.017
ups-0.017
 seasons-0.016
Token though
Token ID996
Feature activation+2.48e+00

Pos logit contributions
First+0.056
 brushes+0.054
 sizes+0.047
ups+0.046
 seasons+0.045
Neg logit contributions
 victorious-0.065
 hero-0.064
 sorry-0.060
 CA-0.059
 devastated-0.057
Token yard
Token ID12699
Feature activation-8.86e-02

Pos logit contributions
First+0.081
 brushes+0.073
ups+0.066
 sizes+0.064
vals+0.064
Neg logit contributions
 hero-0.066
 victorious-0.061
 CA-0.059
eros-0.055
 sorry-0.054
Token.
Token ID13
Feature activation-1.39e+00

Pos logit contributions
 victorious+0.003
 hero+0.003
 sorry+0.003
 CA+0.003
 devastated+0.003
Neg logit contributions
First-0.003
 brushes-0.003
 sizes-0.003
ups-0.003
 styles-0.003
Token He
Token ID679
Feature activation-5.57e-01

Pos logit contributions
 victorious+0.046
 hero+0.045
 CA+0.045
 sorry+0.041
 messenger+0.040
Neg logit contributions
First-0.047
 brushes-0.044
 sizes-0.041
 seasons-0.039
 colors-0.038
Token had
Token ID550
Feature activation+2.05e-01

Pos logit contributions
 victorious+0.020
 hero+0.020
 CA+0.019
 sorry+0.019
 devastated+0.018
Neg logit contributions
First-0.018
 brushes-0.017
 sizes-0.015
ups-0.015
 colors-0.015
Token been
Token ID587
Feature activation-1.20e+00

Pos logit contributions
First+0.008
 brushes+0.007
 sizes+0.007
 seasons+0.006
 styles+0.006
Neg logit contributions
 victorious-0.007
 hero-0.006
 CA-0.006
 sorry-0.006
 devastated-0.006
Token fore
Token ID1674
Feature activation-6.57e+00

Pos logit contributions
 CA+0.040
 hero+0.037
 victorious+0.037
eros+0.034
 hi+0.033
Neg logit contributions
First-0.040
 brushes-0.040
ups-0.038
 flour-0.036
 colors-0.036
Tokenshadow
Token ID19106
Feature activation-1.48e+00

Pos logit contributions
 CA+0.153
 hero+0.148
 victorious+0.143
 sorry+0.133
eros+0.121
Neg logit contributions
First-0.304
ups-0.282
 brushes-0.282
 sizes-0.258
 colors-0.254
Tokened
Token ID276
Feature activation-3.47e+00

Pos logit contributions
 victorious+0.049
 hero+0.048
 CA+0.047
 sorry+0.045
 devastated+0.043
Neg logit contributions
First-0.052
 brushes-0.051
 sizes-0.045
ups-0.045
 colors-0.043
Token about
Token ID546
Feature activation-1.35e+00

Pos logit contributions
 victorious+0.120
 hero+0.118
 CA+0.117
 sorry+0.109
 devastated+0.107
Neg logit contributions
First-0.115
 brushes-0.112
 sizes-0.101
ups-0.099
 colors-0.096
Token this
Token ID428
Feature activation-1.22e+00

Pos logit contributions
 victorious+0.043
 CA+0.042
 hero+0.042
 sorry+0.038
 devastated+0.038
Neg logit contributions
First-0.048
 brushes-0.047
 sizes-0.043
 colors-0.042
 flour-0.042
Token bad
Token ID2089
Feature activation-1.82e+00

Pos logit contributions
 victorious+0.046
 hero+0.045
 CA+0.045
 sorry+0.043
 devastated+0.042
Neg logit contributions
First-0.037
 brushes-0.036
 sizes-0.032
 colors-0.031
ups-0.030
Token said
Token ID531
Feature activation-8.70e-01

Pos logit contributions
 victorious+0.012
 hero+0.012
 sorry+0.012
 CA+0.012
 devastated+0.011
Neg logit contributions
First-0.013
 brushes-0.012
 sizes-0.011
 seasons-0.011
 styles-0.011
Token '
Token ID705
Feature activation-6.67e-02

Pos logit contributions
 victorious+0.026
 hero+0.026
 CA+0.025
 sorry+0.024
 devastated+0.023
Neg logit contributions
First-0.033
 brushes-0.032
 sizes-0.029
ups-0.029
 styles-0.028
TokenYou
Token ID1639
Feature activation-3.64e+00

Pos logit contributions
 victorious+0.002
 CA+0.002
 hero+0.002
 devastated+0.002
 sorry+0.002
Neg logit contributions
 brushes-0.002
First-0.002
 sizes-0.002
 seasons-0.002
 styles-0.002
Token don
Token ID836
Feature activation+1.37e+00

Pos logit contributions
 victorious+0.127
 hero+0.125
 CA+0.121
 sorry+0.112
eros+0.110
Neg logit contributions
First-0.122
 brushes-0.119
 sizes-0.102
ups-0.102
 flour-0.102
Token't
Token ID470
Feature activation-1.51e-01

Pos logit contributions
First+0.042
 brushes+0.039
ups+0.034
 flour+0.033
 sizes+0.033
Neg logit contributions
 hero-0.052
 victorious-0.052
 CA-0.050
 sorry-0.047
eros-0.046
Token look
Token ID804
Feature activation-6.55e+00

Pos logit contributions
 hero+0.004
 victorious+0.004
 sorry+0.004
 CA+0.003
 devastated+0.003
Neg logit contributions
 brushes-0.007
First-0.007
ups-0.006
 sizes-0.006
 styles-0.006
Token very
Token ID845
Feature activation-1.41e+00

Pos logit contributions
 CA+0.203
 victorious+0.203
 hero+0.202
 devastated+0.179
eros+0.176
Neg logit contributions
First-0.234
 brushes-0.234
 sizes-0.210
ups-0.207
 flour-0.207
Token fine
Token ID3734
Feature activation+3.64e-01

Pos logit contributions
 victorious+0.037
 hero+0.037
 CA+0.036
 devastated+0.032
 sorry+0.031
Neg logit contributions
First-0.058
 brushes-0.058
 sizes-0.052
ups-0.052
 flour-0.051
Token'.
Token ID4458
Feature activation-7.24e-02

Pos logit contributions
First+0.015
 brushes+0.015
 sizes+0.013
ups+0.013
 styles+0.012
Neg logit contributions
 victorious-0.010
 hero-0.010
 sorry-0.009
 CA-0.009
 devastated-0.009
Token Tina
Token ID35903
Feature activation-1.88e-01

Pos logit contributions
 victorious+0.002
 hero+0.002
 CA+0.002
 sorry+0.002
 devastated+0.002
Neg logit contributions
First-0.003
 brushes-0.002
 sizes-0.002
ups-0.002
 styles-0.002
Token started
Token ID2067
Feature activation+3.70e+00

Pos logit contributions
 victorious+0.006
 hero+0.006
 CA+0.006
 sorry+0.006
 devastated+0.005
Neg logit contributions
First-0.007
 brushes-0.007
 sizes-0.006
 styles-0.006
 flour-0.006
TokenThe
Token ID464
Feature activation-3.14e-01

Pos logit contributions
 victorious+0.058
 hero+0.056
 CA+0.055
 devastated+0.052
 sorry+0.052
Neg logit contributions
 brushes-0.049
First-0.045
 sizes-0.044
 colors-0.041
 styles-0.041
Token old
Token ID1468
Feature activation-1.49e+00

Pos logit contributions
 victorious+0.010
 CA+0.009
 hero+0.009
 sorry+0.008
 devastated+0.008
Neg logit contributions
First-0.012
 brushes-0.012
 sizes-0.010
ups-0.010
 colors-0.010
Token man
Token ID582
Feature activation-1.87e+00

Pos logit contributions
 victorious+0.044
 hero+0.043
 CA+0.041
 sorry+0.039
 devastated+0.037
Neg logit contributions
First-0.058
 brushes-0.057
 sizes-0.052
ups-0.051
 colors-0.050
Token thanked
Token ID26280
Feature activation-2.54e+00

Pos logit contributions
 victorious+0.069
 CA+0.068
 hero+0.068
 sorry+0.064
 devastated+0.061
Neg logit contributions
 brushes-0.055
First-0.054
ups-0.048
 sizes-0.048
 colors-0.047
Token the
Token ID262
Feature activation-2.65e+00

Pos logit contributions
 victorious+0.095
 hero+0.093
 CA+0.093
 devastated+0.088
 sorry+0.085
Neg logit contributions
First-0.075
 brushes-0.074
 sizes-0.064
 flour-0.064
 styles-0.064
Token messenger
Token ID31228
Feature activation-6.52e+00

Pos logit contributions
 victorious+0.075
 CA+0.074
 hero+0.067
 devastated+0.066
 sorry+0.065
Neg logit contributions
 brushes-0.100
First-0.098
 colors-0.092
 seasons-0.090
 sizes-0.090
Token and
Token ID290
Feature activation-1.85e+00

Pos logit contributions
 CA+0.225
 victorious+0.224
 hero+0.216
 sorry+0.208
 devastated+0.202
Neg logit contributions
First-0.215
 brushes-0.211
 sizes-0.191
ups-0.190
 colors-0.180
Token said
Token ID531
Feature activation-1.33e+00

Pos logit contributions
 CA+0.071
 hero+0.070
 victorious+0.070
 sorry+0.066
 devastated+0.066
Neg logit contributions
 brushes-0.051
First-0.051
ups-0.045
 sizes-0.043
 colors-0.042
Token �
Token ID6184
Feature activation-6.93e-01

Pos logit contributions
 CA+0.040
 victorious+0.039
 hero+0.038
 devastated+0.036
 unlucky+0.034
Neg logit contributions
First-0.047
 brushes-0.046
ups-0.042
 flour-0.042
 sizes-0.040
Token�
Token ID95
Feature activation-3.85e-01

Pos logit contributions
 CA+0.021
 devastated+0.020
 sorry+0.020
 victorious+0.020
 hero+0.019
Neg logit contributions
 brushes-0.021
First-0.021
ups-0.019
 styles-0.019
 flavors-0.018
Token€
Token ID26391
Feature activation-2.06e+00

Pos logit contributions
 sorry+0.012
 hero+0.012
 CA+0.011
 hi+0.011
 devastated+0.011
Neg logit contributions
First-0.013
 brushes-0.012
ups-0.011
 sizes-0.011
vals-0.011
Token.
Token ID13
Feature activation-1.12e-01

Pos logit contributions
 victorious+0.067
 hero+0.067
 CA+0.063
 sorry+0.061
 devastated+0.058
Neg logit contributions
First-0.075
 brushes-0.073
 sizes-0.065
ups-0.064
 styles-0.062
Token
Token ID198
Feature activation-5.17e-01

Pos logit contributions
 victorious+0.003
 hero+0.003
 CA+0.003
 sorry+0.003
 devastated+0.003
Neg logit contributions
First-0.004
 brushes-0.004
 sizes-0.004
ups-0.004
 flour-0.004
Token
Token ID198
Feature activation-1.23e+00

Pos logit contributions
 victorious+0.013
 hero+0.013
 CA+0.012
 sorry+0.012
 devastated+0.011
Neg logit contributions
 brushes-0.022
First-0.022
 sizes-0.020
 styles-0.019
ups-0.019
Token"
Token ID1
Feature activation+1.52e+00

Pos logit contributions
 victorious+0.043
 hero+0.042
 CA+0.041
 devastated+0.039
 sorry+0.038
Neg logit contributions
 brushes-0.041
 sizes-0.038
First-0.036
 colors-0.035
 styles-0.035
TokenOh
Token ID5812
Feature activation-2.19e+00

Pos logit contributions
 brushes+0.051
First+0.050
ups+0.046
 sizes+0.046
 styles+0.043
Neg logit contributions
 victorious-0.051
 CA-0.051
 hero-0.050
 sorry-0.045
 devastated-0.045
Token look
Token ID804
Feature activation-6.52e+00

Pos logit contributions
 victorious+0.079
 hero+0.075
 CA+0.074
 devastated+0.071
 sorry+0.066
Neg logit contributions
 brushes-0.071
First-0.070
 sizes-0.062
 flour-0.061
 colors-0.060
Token at
Token ID379
Feature activation-3.25e-01

Pos logit contributions
 victorious+0.240
 CA+0.238
 hero+0.237
 devastated+0.218
 sorry+0.207
Neg logit contributions
 brushes-0.198
First-0.196
 flour-0.172
 sizes-0.171
 colors-0.170
Token you
Token ID345
Feature activation-2.66e+00

Pos logit contributions
 victorious+0.011
 hero+0.010
 CA+0.010
 devastated+0.009
 sorry+0.009
Neg logit contributions
First-0.011
 brushes-0.011
 sizes-0.010
 flour-0.010
 colors-0.010
Token!"
Token ID2474
Feature activation+9.78e-01

Pos logit contributions
 victorious+0.097
 hero+0.096
 CA+0.093
 sorry+0.089
 devastated+0.086
Neg logit contributions
First-0.085
 brushes-0.083
 sizes-0.074
ups-0.071
 colors-0.069
Token said
Token ID531
Feature activation-5.21e-01

Pos logit contributions
First+0.030
 brushes+0.029
 sizes+0.026
ups+0.025
 styles+0.024
Neg logit contributions
 victorious-0.037
 hero-0.036
 CA-0.035
 sorry-0.034
 devastated-0.033
Token Jane
Token ID12091
Feature activation-6.60e-01

Pos logit contributions
 victorious+0.019
 hero+0.019
 CA+0.019
 devastated+0.017
 sorry+0.017
Neg logit contributions
 brushes-0.015
First-0.015
 flour-0.013
 colors-0.013
ups-0.012
Token day
Token ID1110
Feature activation+4.59e-02

Pos logit contributions
First+0.102
 brushes+0.098
 sizes+0.082
 colors+0.080
 seasons+0.080
Neg logit contributions
 victorious-0.204
 hero-0.195
 CA-0.191
 sorry-0.186
 devastated-0.183
Token,
Token ID11
Feature activation+5.06e-01

Pos logit contributions
First+0.002
 brushes+0.002
 sizes+0.002
ups+0.002
 styles+0.002
Neg logit contributions
 hero-0.001
 CA-0.001
 sorry-0.001
 victorious-0.001
 hi-0.001
Token she
Token ID673
Feature activation-2.71e-01

Pos logit contributions
First+0.019
ups+0.018
 brushes+0.018
 sizes+0.018
 seasons+0.017
Neg logit contributions
 sorry-0.015
 hero-0.015
 CA-0.014
 victorious-0.014
 hi-0.013
Token received
Token ID2722
Feature activation-2.65e+00

Pos logit contributions
 CA+0.007
 hero+0.007
 victorious+0.007
 sorry+0.007
 devastated+0.007
Neg logit contributions
First-0.011
 brushes-0.011
 sizes-0.010
 styles-0.010
 seasons-0.009
Token an
Token ID281
Feature activation-1.43e+00

Pos logit contributions
 victorious+0.078
 hero+0.077
 CA+0.075
 sorry+0.071
 devastated+0.067
Neg logit contributions
First-0.104
 brushes-0.101
 sizes-0.094
ups-0.089
 seasons-0.089
Token invitation
Token ID17023
Feature activation-6.51e+00

Pos logit contributions
 victorious+0.049
 CA+0.044
 hero+0.041
 devastated+0.038
smart+0.038
Neg logit contributions
 brushes-0.052
 colors-0.048
First-0.048
 flour-0.046
 styles-0.044
Token to
Token ID284
Feature activation-1.12e+00

Pos logit contributions
 victorious+0.291
 hero+0.282
 CA+0.275
 sorry+0.269
eros+0.268
Neg logit contributions
 brushes-0.152
First-0.151
ups-0.124
 colors-0.118
 flour-0.116
Token a
Token ID257
Feature activation-1.01e+00

Pos logit contributions
 victorious+0.048
 hero+0.047
eros+0.045
 devastated+0.045
 CA+0.045
Neg logit contributions
 brushes-0.028
First-0.027
 colors-0.023
 color-0.022
 styles-0.022
Token party
Token ID2151
Feature activation-3.75e+00

Pos logit contributions
 victorious+0.032
eros+0.030
 CA+0.029
 hero+0.028
 devastated+0.027
Neg logit contributions
 brushes-0.034
First-0.033
 flour-0.032
 colors-0.031
 color-0.030
Token from
Token ID422
Feature activation-3.80e+00

Pos logit contributions
 victorious+0.150
 hero+0.146
 CA+0.145
 sorry+0.138
 devastated+0.137
Neg logit contributions
First-0.106
 brushes-0.106
ups-0.092
 colors-0.090
 sizes-0.089
Token her
Token ID607
Feature activation-2.02e+00

Pos logit contributions
 victorious+0.153
 hero+0.151
 CA+0.147
 sorry+0.140
 devastated+0.137
Neg logit contributions
 brushes-0.108
First-0.105
 colors-0.091
 styles-0.087
 flour-0.086
Token know
Token ID760
Feature activation-1.26e+00

Pos logit contributions
 victorious+0.021
 hero+0.021
 CA+0.020
 sorry+0.019
 devastated+0.018
Neg logit contributions
First-0.021
 brushes-0.020
 sizes-0.018
ups-0.018
 flour-0.017
Token,
Token ID11
Feature activation+7.43e-01

Pos logit contributions
 victorious+0.031
 hero+0.030
 sorry+0.029
 CA+0.027
 devastated+0.026
Neg logit contributions
First-0.056
 brushes-0.054
 sizes-0.050
ups-0.048
 seasons-0.048
Token her
Token ID607
Feature activation-1.13e+00

Pos logit contributions
First+0.024
 brushes+0.024
 sizes+0.021
ups+0.021
 colors+0.020
Neg logit contributions
 victorious-0.026
 hero-0.026
 CA-0.025
 sorry-0.024
 devastated-0.023
Token mum
Token ID25682
Feature activation+2.64e+00

Pos logit contributions
 victorious+0.040
 hero+0.039
 CA+0.038
 sorry+0.036
 devastated+0.035
Neg logit contributions
First-0.037
 brushes-0.037
 sizes-0.033
ups-0.032
 colors-0.031
Token was
Token ID373
Feature activation+5.11e-02

Pos logit contributions
First+0.090
 brushes+0.086
 sizes+0.077
ups+0.075
 flour+0.074
Neg logit contributions
 victorious-0.091
 hero-0.090
 CA-0.086
 sorry-0.083
 devastated-0.081
Token fore
Token ID1674
Feature activation-6.50e+00

Pos logit contributions
First+0.002
 brushes+0.002
 sizes+0.002
ups+0.002
 colors+0.002
Neg logit contributions
 victorious-0.002
 hero-0.002
 CA-0.002
 sorry-0.001
 devastated-0.001
Tokenshadow
Token ID19106
Feature activation-1.45e+00

Pos logit contributions
 CA+0.160
 hero+0.151
 victorious+0.145
 sorry+0.135
eros+0.126
Neg logit contributions
First-0.298
 brushes-0.274
ups-0.272
 sizes-0.250
 colors-0.246
Tokening
Token ID278
Feature activation-1.99e+00

Pos logit contributions
 victorious+0.053
 hero+0.052
 CA+0.050
 sorry+0.048
 devastated+0.047
Neg logit contributions
First-0.047
 brushes-0.045
 sizes-0.040
ups-0.038
 styles-0.038
Token...
Token ID986
Feature activation+1.05e+00

Pos logit contributions
 victorious+0.063
 hero+0.062
 CA+0.059
 sorry+0.058
 devastated+0.055
Neg logit contributions
First-0.073
 brushes-0.071
 sizes-0.064
ups-0.062
 flour-0.061
Tokenthe
Token ID1169
Feature activation+4.65e-01

Pos logit contributions
First+0.034
 brushes+0.034
 sizes+0.030
ups+0.029
 styles+0.028
Neg logit contributions
 victorious-0.037
 hero-0.036
 CA-0.035
 sorry-0.034
 devastated-0.033
Token next
Token ID1306
Feature activation-6.53e-02

Pos logit contributions
First+0.017
 brushes+0.016
 sizes+0.015
ups+0.014
 styles+0.014
Neg logit contributions
 hero-0.015
 victorious-0.015
 CA-0.014
 sorry-0.014
 devastated-0.013
Token day
Token ID1110
Feature activation-7.16e-02

Pos logit contributions
First+0.096
 brushes+0.093
 sizes+0.078
 seasons+0.076
 colors+0.076
Neg logit contributions
 victorious-0.191
 CA-0.183
 hero-0.182
 sorry-0.176
 devastated-0.173
Token,
Token ID11
Feature activation+4.29e-01

Pos logit contributions
 hero+0.002
 victorious+0.002
 CA+0.002
 sorry+0.002
 hi+0.002
Neg logit contributions
First-0.003
 brushes-0.003
 sizes-0.003
ups-0.002
 styles-0.002
Token Lily
Token ID20037
Feature activation+1.36e+00

Pos logit contributions
First+0.015
 brushes+0.014
 sizes+0.014
ups+0.013
 seasons+0.013
Neg logit contributions
 hero-0.014
 victorious-0.014
 CA-0.014
 sorry-0.014
 devastated-0.012
Token received
Token ID2722
Feature activation-2.61e+00

Pos logit contributions
First+0.056
 brushes+0.054
 sizes+0.052
 styles+0.049
 flour+0.048
Neg logit contributions
 CA-0.036
 hero-0.035
 victorious-0.034
 devastated-0.032
 sorry-0.031
Token an
Token ID281
Feature activation-1.33e+00

Pos logit contributions
 victorious+0.075
 hero+0.074
 CA+0.071
 sorry+0.068
 devastated+0.064
Neg logit contributions
First-0.104
 brushes-0.100
 sizes-0.095
 seasons-0.090
ups-0.090
Token invitation
Token ID17023
Feature activation-6.49e+00

Pos logit contributions
 victorious+0.042
 CA+0.040
 hero+0.037
 devastated+0.034
eros+0.033
Neg logit contributions
 brushes-0.050
First-0.047
 colors-0.046
 styles-0.044
 flour-0.044
Token in
Token ID287
Feature activation-1.43e+00

Pos logit contributions
 victorious+0.283
 hero+0.274
 CA+0.272
 sorry+0.264
 devastated+0.259
Neg logit contributions
First-0.157
 brushes-0.156
ups-0.130
 colors-0.121
 flour-0.119
Token the
Token ID262
Feature activation-5.44e-01

Pos logit contributions
 victorious+0.060
 hero+0.059
 CA+0.057
 sorry+0.055
 devastated+0.053
Neg logit contributions
 brushes-0.039
First-0.039
 colors-0.033
 sizes-0.032
 styles-0.031
Token mail
Token ID6920
Feature activation-1.92e+00

Pos logit contributions
 victorious+0.022
 hero+0.021
 CA+0.020
eros+0.019
 heroic+0.019
Neg logit contributions
 brushes-0.015
First-0.015
 flour-0.013
 colors-0.013
ups-0.013
Token.
Token ID13
Feature activation+3.69e-02

Pos logit contributions
 victorious+0.067
 hero+0.066
 CA+0.064
 sorry+0.061
 devastated+0.059
Neg logit contributions
First-0.065
 brushes-0.063
 sizes-0.057
ups-0.055
 styles-0.053
Token It
Token ID632
Feature activation-4.27e-01

Pos logit contributions
First+0.001
 brushes+0.001
 sizes+0.001
 colors+0.001
 styles+0.001
Neg logit contributions
 victorious-0.002
 hero-0.001
 CA-0.001
 sorry-0.001
 devastated-0.001
Token.
Token ID13
Feature activation-2.99e-01

Pos logit contributions
 victorious+0.055
 hero+0.054
 CA+0.053
 sorry+0.050
 devastated+0.048
Neg logit contributions
First-0.059
 brushes-0.057
 sizes-0.052
ups-0.050
 colors-0.049
Token But
Token ID887
Feature activation+1.11e+00

Pos logit contributions
 hero+0.009
 victorious+0.009
 CA+0.008
 sorry+0.008
 devastated+0.008
Neg logit contributions
First-0.012
 brushes-0.011
 sizes-0.010
ups-0.010
 flour-0.010
Token he
Token ID339
Feature activation-1.47e-01

Pos logit contributions
First+0.029
 brushes+0.028
 sizes+0.024
 colors+0.023
 flour+0.023
Neg logit contributions
 CA-0.046
 victorious-0.045
 hero-0.045
 devastated-0.041
 sorry-0.041
Token didn
Token ID1422
Feature activation+3.53e-01

Pos logit contributions
 victorious+0.005
 CA+0.005
 hero+0.005
 sorry+0.005
 devastated+0.005
Neg logit contributions
First-0.005
 brushes-0.004
ups-0.004
 sizes-0.004
 colors-0.004
Token't
Token ID470
Feature activation-1.71e-01

Pos logit contributions
First+0.009
 brushes+0.008
ups+0.007
 flour+0.007
 seasons+0.007
Neg logit contributions
 hero-0.015
 victorious-0.014
 CA-0.014
 sorry-0.013
 messenger-0.013
Token look
Token ID804
Feature activation-6.48e+00

Pos logit contributions
 CA+0.006
 victorious+0.006
eros+0.006
 devastated+0.006
 hero+0.006
Neg logit contributions
First-0.005
 brushes-0.005
 sizes-0.004
 colors-0.004
 seasons-0.004
Token where
Token ID810
Feature activation-4.45e+00

Pos logit contributions
 CA+0.245
eros+0.217
 hero+0.203
 victorious+0.202
 devastated+0.196
Neg logit contributions
 brushes-0.187
 colors-0.186
First-0.183
 sizes-0.182
 shapes-0.181
Token he
Token ID339
Feature activation-6.57e-01

Pos logit contributions
 victorious+0.145
 hero+0.142
 CA+0.138
 sorry+0.130
 devastated+0.127
Neg logit contributions
First-0.159
 brushes-0.156
 sizes-0.140
ups-0.137
 colors-0.134
Token was
Token ID373
Feature activation-1.55e+00

Pos logit contributions
 victorious+0.031
 hero+0.030
 sorry+0.029
 goodbye+0.028
 devastated+0.028
Neg logit contributions
 brushes-0.013
First-0.012
 colors-0.012
 sizes-0.011
 flavors-0.011
Token going
Token ID1016
Feature activation-3.02e+00

Pos logit contributions
 CA+0.060
 hero+0.059
 victorious+0.059
eros+0.058
 devastated+0.054
Neg logit contributions
First-0.039
 brushes-0.035
 sizes-0.033
 colors-0.033
ups-0.032
Token and
Token ID290
Feature activation-1.47e+00

Pos logit contributions
 victorious+0.110
 hero+0.109
 CA+0.106
 sorry+0.101
 devastated+0.098
Neg logit contributions
First-0.097
 brushes-0.093
 sizes-0.084
ups-0.081
 colors-0.079
Token fish
Token ID5916
Feature activation+1.69e+00

Pos logit contributions
First+0.047
 brushes+0.046
 sizes+0.042
ups+0.041
 colors+0.041
Neg logit contributions
 victorious-0.031
 hero-0.030
 CA-0.029
 sorry-0.027
 devastated-0.026
Token said
Token ID531
Feature activation-1.34e+00

Pos logit contributions
First+0.054
 brushes+0.053
 sizes+0.047
ups+0.046
 colors+0.045
Neg logit contributions
 victorious-0.061
 hero-0.060
 CA-0.059
 sorry-0.056
 devastated-0.054
Token,
Token ID11
Feature activation-2.22e-01

Pos logit contributions
 victorious+0.047
 hero+0.046
 CA+0.045
 devastated+0.042
 sorry+0.041
Neg logit contributions
First-0.043
 brushes-0.042
 sizes-0.038
ups-0.037
 flour-0.037
Token "
Token ID366
Feature activation+6.30e-01

Pos logit contributions
 victorious+0.006
 hero+0.006
 CA+0.006
 devastated+0.005
 sorry+0.005
Neg logit contributions
First-0.009
 brushes-0.009
 sizes-0.008
 flour-0.008
 colors-0.008
TokenHi
Token ID17250
Feature activation-1.66e+00

Pos logit contributions
 brushes+0.021
First+0.020
 sizes+0.018
 flour+0.018
ups+0.017
Neg logit contributions
 victorious-0.022
 CA-0.022
 hero-0.022
 devastated-0.019
 sorry-0.019
Token Gab
Token ID12300
Feature activation-6.47e+00

Pos logit contributions
 victorious+0.070
 devastated+0.068
 CA+0.067
 hero+0.063
 sorry+0.062
Neg logit contributions
First-0.040
 brushes-0.039
 colors-0.037
 flour-0.036
 seasons-0.036
Tokenri
Token ID380
Feature activation-1.24e+00

Pos logit contributions
 victorious+0.261
 hero+0.261
 CA+0.250
 sorry+0.239
 devastated+0.231
Neg logit contributions
First-0.184
 brushes-0.177
 sizes-0.155
ups-0.146
 seasons-0.143
Tokenella
Token ID12627
Feature activation-1.45e+00

Pos logit contributions
 victorious+0.043
 CA+0.042
 devastated+0.041
 hero+0.041
 sorry+0.040
Neg logit contributions
 brushes-0.038
First-0.037
 sizes-0.036
 colors-0.034
 flavors-0.033
Token,
Token ID11
Feature activation-3.80e-01

Pos logit contributions
 victorious+0.051
 hero+0.050
 CA+0.049
 sorry+0.047
 devastated+0.045
Neg logit contributions
First-0.048
 brushes-0.047
 sizes-0.042
ups-0.040
 colors-0.039
Token would
Token ID561
Feature activation+2.26e+00

Pos logit contributions
 victorious+0.012
 hero+0.011
 CA+0.011
 devastated+0.010
eros+0.010
Neg logit contributions
 brushes-0.014
First-0.013
 sizes-0.012
 flour-0.012
 seasons-0.012
Token you
Token ID345
Feature activation-1.40e+00

Pos logit contributions
First+0.075
 brushes+0.074
 sizes+0.066
 colors+0.063
ups+0.063
Neg logit contributions
 victorious-0.079
 hero-0.077
 CA-0.076
 devastated-0.070
 sorry-0.070
Token ran
Token ID4966
Feature activation-1.42e+00

Pos logit contributions
 victorious+0.012
 hero+0.012
 CA+0.012
 sorry+0.011
 devastated+0.011
Neg logit contributions
First-0.010
 brushes-0.010
 sizes-0.009
ups-0.009
 colors-0.008
Token away
Token ID1497
Feature activation-2.84e+00

Pos logit contributions
 victorious+0.033
 hero+0.032
 CA+0.029
 sorry+0.028
 devastated+0.026
Neg logit contributions
First-0.066
 brushes-0.065
 sizes-0.060
ups-0.058
 flour-0.058
Token and
Token ID290
Feature activation-1.14e+00

Pos logit contributions
 CA+0.103
 hero+0.100
 victorious+0.099
 sorry+0.092
 devastated+0.090
Neg logit contributions
First-0.088
 brushes-0.086
 sizes-0.078
ups-0.076
 colors-0.074
Token didn
Token ID1422
Feature activation+2.75e-01

Pos logit contributions
 victorious+0.040
 hero+0.040
 CA+0.039
 sorry+0.036
 devastated+0.035
Neg logit contributions
First-0.036
 brushes-0.036
 sizes-0.032
ups-0.031
 colors-0.030
Token't
Token ID470
Feature activation+8.75e-02

Pos logit contributions
First+0.008
 brushes+0.007
ups+0.006
 flour+0.006
 seasons+0.006
Neg logit contributions
 hero-0.011
 victorious-0.011
 CA-0.010
 sorry-0.010
 devastated-0.010
Token look
Token ID804
Feature activation-6.46e+00

Pos logit contributions
First+0.003
 brushes+0.003
 sizes+0.003
ups+0.003
 seasons+0.003
Neg logit contributions
 victorious-0.003
 CA-0.003
 hero-0.003
 sorry-0.002
 devastated-0.002
Token back
Token ID736
Feature activation-3.56e+00

Pos logit contributions
 CA+0.196
 hero+0.172
 victorious+0.169
eros+0.165
 devastated+0.158
Neg logit contributions
 brushes-0.240
First-0.234
 sizes-0.224
ups-0.223
 colors-0.222
Token.
Token ID13
Feature activation-7.95e-01

Pos logit contributions
 victorious+0.114
 hero+0.111
 CA+0.105
 sorry+0.103
 devastated+0.098
Neg logit contributions
First-0.132
 brushes-0.129
 sizes-0.117
ups-0.112
 flour-0.111
Token He
Token ID679
Feature activation-8.33e-01

Pos logit contributions
 hero+0.026
 victorious+0.026
 CA+0.026
 sorry+0.024
 devastated+0.023
Neg logit contributions
First-0.028
 brushes-0.026
 sizes-0.023
ups-0.023
 seasons-0.022
Token never
Token ID1239
Feature activation-1.51e+00

Pos logit contributions
 victorious+0.032
 hero+0.031
 CA+0.031
 sorry+0.030
 devastated+0.028
Neg logit contributions
First-0.024
 brushes-0.024
 sizes-0.021
ups-0.021
 colors-0.020
Token returned
Token ID4504
Feature activation-2.71e+00

Pos logit contributions
 victorious+0.052
 hero+0.051
 CA+0.050
 sorry+0.048
 devastated+0.046
Neg logit contributions
First-0.051
 brushes-0.050
 sizes-0.045
ups-0.043
 colors-0.042
Token filled
Token ID5901
Feature activation+2.06e+00

Pos logit contributions
 victorious+0.028
 hero+0.027
 CA+0.026
 sorry+0.025
 devastated+0.024
Neg logit contributions
First-0.039
 brushes-0.036
 sizes-0.034
ups-0.033
 seasons-0.033
Token with
Token ID351
Feature activation-1.06e+00

Pos logit contributions
First+0.077
 brushes+0.076
 sizes+0.068
ups+0.067
 styles+0.064
Neg logit contributions
 victorious-0.065
 hero-0.062
 CA-0.061
 sorry-0.056
 devastated-0.055
Token treasures
Token ID29561
Feature activation-3.72e-01

Pos logit contributions
 victorious+0.035
 hero+0.034
 sorry+0.032
 devastated+0.030
 hi+0.029
Neg logit contributions
First-0.038
 brushes-0.034
ups-0.034
vals-0.034
akes-0.031
Token from
Token ID422
Feature activation-2.37e+00

Pos logit contributions
 victorious+0.011
 hero+0.011
 CA+0.010
 sorry+0.010
 devastated+0.010
Neg logit contributions
First-0.014
 brushes-0.014
 seasons-0.012
ups-0.012
 sizes-0.012
Token long
Token ID890
Feature activation+3.16e+00

Pos logit contributions
 victorious+0.067
 hero+0.066
 CA+0.061
 sorry+0.059
 devastated+0.056
Neg logit contributions
First-0.098
 brushes-0.094
ups-0.087
 sizes-0.085
 styles-0.082
Token ago
Token ID2084
Feature activation+6.03e+00

Pos logit contributions
First+0.110
 brushes+0.099
ups+0.094
vals+0.087
akes+0.086
Neg logit contributions
 hero-0.105
 victorious-0.103
 CA-0.094
 goodbye-0.093
 sorry-0.089
Token!
Token ID0
Feature activation-4.15e-01

Pos logit contributions
First+0.259
 brushes+0.241
 seasons+0.215
 sizes+0.214
ups+0.213
Neg logit contributions
 victorious-0.154
 hero-0.145
 sorry-0.137
 CA-0.135
 goodbye-0.129
Token
Token ID198
Feature activation-5.94e-01

Pos logit contributions
 victorious+0.013
 hero+0.013
 CA+0.012
 sorry+0.012
 devastated+0.011
Neg logit contributions
First-0.016
 brushes-0.015
 sizes-0.014
 flour-0.013
ups-0.013
Token
Token ID198
Feature activation-1.39e+00

Pos logit contributions
 victorious+0.016
 hero+0.016
 CA+0.015
 sorry+0.014
 devastated+0.014
Neg logit contributions
First-0.024
 brushes-0.024
 sizes-0.022
ups-0.021
 styles-0.021
TokenThe
Token ID464
Feature activation+7.58e-02

Pos logit contributions
 victorious+0.050
 hero+0.048
 CA+0.047
 devastated+0.044
 sorry+0.042
Neg logit contributions
 brushes-0.046
 sizes-0.042
First-0.041
 styles-0.040
 colors-0.039
Token adventurers
Token ID36843
Feature activation-1.03e+00

Pos logit contributions
First+0.003
 brushes+0.003
 colors+0.003
 seasons+0.003
 sizes+0.003
Neg logit contributions
 CA-0.002
 victorious-0.002
 devastated-0.002
eros-0.002
 hero-0.002
Token goose
Token ID37246
Feature activation-1.13e+00

Pos logit contributions
First+0.014
 brushes+0.014
 colors+0.013
 seasons+0.013
 flour+0.013
Neg logit contributions
 victorious-0.013
 hero-0.012
 CA-0.012
 devastated-0.011
 sorry-0.011
Token played
Token ID2826
Feature activation+1.07e+00

Pos logit contributions
 victorious+0.045
 CA+0.044
 hero+0.044
 sorry+0.041
 devastated+0.040
Neg logit contributions
First-0.032
 brushes-0.032
 sizes-0.028
 colors-0.027
ups-0.026
Token together
Token ID1978
Feature activation+1.13e+00

Pos logit contributions
First+0.030
 brushes+0.028
 sizes+0.025
 colors+0.024
 shapes+0.024
Neg logit contributions
 CA-0.043
 victorious-0.042
 hero-0.042
 devastated-0.040
 sorry-0.039
Token all
Token ID477
Feature activation+2.06e+00

Pos logit contributions
First+0.040
 brushes+0.039
 sizes+0.035
ups+0.035
 flour+0.033
Neg logit contributions
 victorious-0.037
 hero-0.037
 CA-0.035
 sorry-0.034
 devastated-0.032
Token day
Token ID1110
Feature activation+2.39e+00

Pos logit contributions
First+0.063
 brushes+0.061
ups+0.054
 flour+0.052
 sizes+0.052
Neg logit contributions
 victorious-0.076
 hero-0.076
 CA-0.073
 sorry-0.069
 devastated-0.067
Token long
Token ID890
Feature activation+5.48e+00

Pos logit contributions
 brushes+0.094
First+0.093
 seasons+0.084
ups+0.084
 flavors+0.084
Neg logit contributions
 victorious-0.068
 hero-0.067
 sorry-0.061
 devastated-0.057
 CA-0.055
Token.
Token ID13
Feature activation+2.87e-02

Pos logit contributions
First+0.223
 brushes+0.216
 sizes+0.191
ups+0.185
 flavors+0.184
Neg logit contributions
 victorious-0.155
 hero-0.150
 CA-0.139
 sorry-0.138
eros-0.127
Token They
Token ID1119
Feature activation+2.26e+00

Pos logit contributions
First+0.001
 brushes+0.001
 sizes+0.001
ups+0.001
 flour+0.001
Neg logit contributions
 victorious-0.001
 hero-0.001
 CA-0.001
 sorry-0.001
 devastated-0.001
Token ran
Token ID4966
Feature activation-4.33e-01

Pos logit contributions
First+0.081
 brushes+0.079
ups+0.072
 sizes+0.070
 flour+0.069
Neg logit contributions
 victorious-0.073
 hero-0.073
 CA-0.070
 sorry-0.066
 devastated-0.064
Token,
Token ID11
Feature activation+2.60e+00

Pos logit contributions
 victorious+0.015
 hero+0.015
 CA+0.015
 sorry+0.014
 devastated+0.014
Neg logit contributions
First-0.014
 brushes-0.014
 sizes-0.012
ups-0.012
 colors-0.012
Token jumped
Token ID11687
Feature activation+1.47e+00

Pos logit contributions
First+0.089
 brushes+0.087
 sizes+0.078
ups+0.075
 colors+0.074
Neg logit contributions
 victorious-0.090
 hero-0.088
 CA-0.086
 sorry-0.082
 devastated-0.079
Token!
Token ID0
Feature activation-1.18e+00

Pos logit contributions
 victorious+0.066
 CA+0.066
 hero+0.065
 devastated+0.060
 sorry+0.060
Neg logit contributions
First-0.051
 brushes-0.051
 sizes-0.045
ups-0.045
 colors-0.044
Token
Token ID198
Feature activation-8.70e-01

Pos logit contributions
 hero+0.042
 victorious+0.041
 sorry+0.038
 devastated+0.038
 CA+0.038
Neg logit contributions
First-0.038
 brushes-0.036
 sizes-0.031
ups-0.031
 seasons-0.030
Token
Token ID198
Feature activation-1.59e+00

Pos logit contributions
 hero+0.027
 victorious+0.025
 CA+0.024
 sorry+0.024
 devastated+0.022
Neg logit contributions
First-0.034
 brushes-0.031
ups-0.029
 flour-0.028
 seasons-0.027
TokenHe
Token ID1544
Feature activation-4.17e-01

Pos logit contributions
 victorious+0.055
 hero+0.052
 CA+0.051
 devastated+0.047
 sorry+0.047
Neg logit contributions
 brushes-0.056
First-0.053
 sizes-0.051
 styles-0.048
 flour-0.047
Token continued
Token ID3767
Feature activation+1.24e+00

Pos logit contributions
 hero+0.016
 victorious+0.016
 sorry+0.016
 CA+0.016
eros+0.015
Neg logit contributions
 brushes-0.011
First-0.011
ups-0.009
 colors-0.009
 sizes-0.009
Token r
Token ID374
Feature activation+5.48e+00

Pos logit contributions
First+0.030
 brushes+0.030
ups+0.025
 colors+0.024
 flour+0.024
Neg logit contributions
 victorious-0.052
 CA-0.052
 hero-0.052
 devastated-0.047
 sorry-0.047
Tokenaking
Token ID868
Feature activation+6.69e-01

Pos logit contributions
 brushes+0.223
 sizes+0.213
First+0.204
 flavors+0.201
 sheets+0.198
Neg logit contributions
 victorious-0.139
 hero-0.126
eros-0.112
smart-0.109
 devastated-0.108
Token the
Token ID262
Feature activation+4.30e-01

Pos logit contributions
First+0.027
 brushes+0.026
 sizes+0.022
ups+0.022
 flavors+0.022
Neg logit contributions
 victorious-0.021
 hero-0.020
 sorry-0.018
 CA-0.017
 devastated-0.016
Token leaves
Token ID5667
Feature activation+3.58e+00

Pos logit contributions
First+0.014
 brushes+0.013
 sizes+0.012
 styles+0.012
ups+0.012
Neg logit contributions
 hero-0.016
 victorious-0.015
 CA-0.014
 sorry-0.014
 unlucky-0.013
Token until
Token ID1566
Feature activation+3.41e+00

Pos logit contributions
First+0.120
 brushes+0.117
 sizes+0.104
ups+0.101
 styles+0.099
Neg logit contributions
 victorious-0.126
 hero-0.123
 CA-0.118
 sorry-0.114
 devastated-0.110
Token all
Token ID477
Feature activation+1.69e+00

Pos logit contributions
First+0.126
 brushes+0.122
 sizes+0.112
ups+0.107
 styles+0.106
Neg logit contributions
 victorious-0.112
 hero-0.109
 CA-0.105
 sorry-0.101
 devastated-0.097
Token fresh
Token ID4713
Feature activation+2.45e+00

Pos logit contributions
First+0.017
 brushes+0.016
 sizes+0.015
ups+0.015
 styles+0.014
Neg logit contributions
 hero-0.011
 victorious-0.011
 CA-0.010
 sorry-0.010
 devastated-0.009
Token air
Token ID1633
Feature activation+1.70e+00

Pos logit contributions
First+0.080
 brushes+0.071
ups+0.067
 sizes+0.065
 styles+0.060
Neg logit contributions
 hero-0.091
 victorious-0.089
 CA-0.087
 sorry-0.083
eros-0.079
Token made
Token ID925
Feature activation+3.56e+00

Pos logit contributions
First+0.058
 brushes+0.057
 sizes+0.050
ups+0.049
 styles+0.048
Neg logit contributions
 victorious-0.059
 hero-0.057
 CA-0.056
 sorry-0.053
 devastated-0.051
Token him
Token ID683
Feature activation+9.31e-01

Pos logit contributions
First+0.124
 brushes+0.121
 sizes+0.109
ups+0.106
 colors+0.104
Neg logit contributions
 victorious-0.119
 hero-0.117
 CA-0.114
 sorry-0.108
 devastated-0.105
Token feel
Token ID1254
Feature activation-5.74e-01

Pos logit contributions
 brushes+0.034
First+0.033
 sizes+0.030
ups+0.029
 colors+0.029
Neg logit contributions
 CA-0.029
 hero-0.028
 victorious-0.028
eros-0.026
 sorry-0.025
Token more
Token ID517
Feature activation+5.56e+00

Pos logit contributions
 hero+0.017
 CA+0.017
eros+0.016
 victorious+0.016
 devastated+0.014
Neg logit contributions
First-0.021
 brushes-0.020
 sizes-0.019
 colors-0.019
ups-0.018
Token awake
Token ID21693
Feature activation-1.09e-01

Pos logit contributions
First+0.219
 brushes+0.214
 sizes+0.196
ups+0.190
 colors+0.187
Neg logit contributions
 victorious-0.162
 hero-0.159
 CA-0.155
 sorry-0.144
 devastated-0.139
Token.
Token ID13
Feature activation-1.13e-02

Pos logit contributions
 victorious+0.003
 hero+0.003
 sorry+0.003
 CA+0.003
 devastated+0.003
Neg logit contributions
First-0.004
 brushes-0.004
 sizes-0.003
ups-0.003
 seasons-0.003
Token When
Token ID1649
Feature activation+2.46e+00

Pos logit contributions
 hero+0.000
 victorious+0.000
 CA+0.000
 sorry+0.000
 devastated+0.000
Neg logit contributions
First-0.000
 brushes-0.000
 sizes-0.000
ups-0.000
 flour-0.000
Token he
Token ID339
Feature activation+3.98e-02

Pos logit contributions
First+0.101
 brushes+0.100
 sizes+0.093
ups+0.089
 styles+0.089
Neg logit contributions
 victorious-0.072
 hero-0.071
 sorry-0.065
 CA-0.065
 devastated-0.059
Token went
Token ID1816
Feature activation-1.29e+00

Pos logit contributions
First+0.001
 brushes+0.001
 soon+0.001
ups+0.001
 batter+0.001
Neg logit contributions
 sorry-0.002
 hero-0.002
 victorious-0.002
eros-0.002
 CA-0.002
Token about
Token ID546
Feature activation+2.83e-01

Pos logit contributions
 hero+0.072
 sorry+0.070
 victorious+0.070
 CA+0.066
 hi+0.064
Neg logit contributions
First-0.105
 brushes-0.097
 sizes-0.092
 styles-0.087
 seasons-0.087
Token their
Token ID511
Feature activation+2.31e-01

Pos logit contributions
First+0.011
 brushes+0.010
 sizes+0.009
ups+0.009
 styles+0.008
Neg logit contributions
 victorious-0.009
 hero-0.009
 CA-0.009
 sorry-0.009
 hi-0.008
Token snow
Token ID6729
Feature activation+3.52e+00

Pos logit contributions
First+0.006
 brushes+0.006
 sizes+0.005
ups+0.005
 styles+0.005
Neg logit contributions
 hero-0.010
 victorious-0.010
 CA-0.009
 sorry-0.009
 devastated-0.009
Tokenman
Token ID805
Feature activation-8.65e-01

Pos logit contributions
First+0.177
 brushes+0.159
 sizes+0.143
 styles+0.140
ups+0.138
Neg logit contributions
 hero-0.082
 victorious-0.072
eros-0.072
 CA-0.067
 sorry-0.067
Token.
Token ID13
Feature activation+2.39e-01

Pos logit contributions
 victorious+0.033
 CA+0.032
 hero+0.032
 sorry+0.030
 devastated+0.030
Neg logit contributions
First-0.025
 brushes-0.025
 sizes-0.022
ups-0.022
 flour-0.021
Token Mom
Token ID11254
Feature activation+5.38e+00

Pos logit contributions
First+0.008
 brushes+0.008
 sizes+0.007
 flour+0.007
ups+0.007
Neg logit contributions
 victorious-0.008
 hero-0.008
 CA-0.008
 devastated-0.007
 sorry-0.007
Token and
Token ID290
Feature activation-1.17e+00

Pos logit contributions
First+0.199
 brushes+0.184
 sizes+0.175
 styles+0.166
 seasons+0.166
Neg logit contributions
 victorious-0.168
 hero-0.168
 CA-0.159
 sorry-0.156
 devastated-0.152
Token dad
Token ID9955
Feature activation+1.77e+00

Pos logit contributions
 hero+0.033
 victorious+0.032
 sorry+0.031
 CA+0.031
 hi+0.028
Neg logit contributions
First-0.048
 brushes-0.045
 sizes-0.043
ups-0.041
 seasons-0.041
Token said
Token ID531
Feature activation-2.08e-01

Pos logit contributions
First+0.051
 brushes+0.050
 sizes+0.043
ups+0.042
 colors+0.041
Neg logit contributions
 victorious-0.070
 hero-0.069
 CA-0.069
 sorry-0.064
 devastated-0.062
Token they
Token ID484
Feature activation-8.70e-01

Pos logit contributions
 sorry+0.007
 victorious+0.007
 hero+0.007
 CA+0.007
 hi+0.006
Neg logit contributions
First-0.008
 brushes-0.007
 sizes-0.007
ups-0.006
 styles-0.006
Token were
Token ID547
Feature activation-4.75e-01

Pos logit contributions
 hero+0.029
 victorious+0.029
 CA+0.028
 sorry+0.028
 devastated+0.027
Neg logit contributions
First-0.030
 brushes-0.028
 seasons-0.025
ups-0.025
 sizes-0.025
Tokenies
Token ID444
Feature activation+2.15e+00

Pos logit contributions
First+0.091
 brushes+0.069
 seasons+0.068
 sizes+0.065
 styles+0.062
Neg logit contributions
 hero-0.086
 victorious-0.084
 sorry-0.080
eros-0.078
 reunited-0.078
Token and
Token ID290
Feature activation+1.35e+00

Pos logit contributions
First+0.084
 brushes+0.078
 sizes+0.070
ups+0.070
 seasons+0.069
Neg logit contributions
 victorious-0.066
 hero-0.065
 sorry-0.061
 CA-0.061
 devastated-0.056
Token toys
Token ID14958
Feature activation+3.51e+00

Pos logit contributions
First+0.044
akes+0.038
ups+0.035
 brushes+0.034
vals+0.033
Neg logit contributions
 sorry-0.046
 victorious-0.045
 hero-0.045
 CA-0.045
 goodbye-0.043
Token.
Token ID13
Feature activation+8.01e-01

Pos logit contributions
First+0.140
 brushes+0.129
 seasons+0.119
 sizes+0.117
ups+0.115
Neg logit contributions
 victorious-0.103
 hero-0.101
 CA-0.093
 sorry-0.092
 devastated-0.084
Token They
Token ID1119
Feature activation+2.09e+00

Pos logit contributions
First+0.024
 brushes+0.023
 sizes+0.021
ups+0.020
 styles+0.019
Neg logit contributions
 victorious-0.031
 hero-0.031
 CA-0.030
 sorry-0.029
 devastated-0.028
Token were
Token ID547
Feature activation+2.34e+00

Pos logit contributions
First+0.080
 brushes+0.071
ups+0.071
 seasons+0.069
akes+0.067
Neg logit contributions
 victorious-0.062
 hero-0.060
 CA-0.057
 sorry-0.057
 devastated-0.055
Token very
Token ID845
Feature activation+1.69e-01

Pos logit contributions
First+0.116
 brushes+0.108
ups+0.101
 sizes+0.099
 seasons+0.098
Neg logit contributions
 victorious-0.050
 sorry-0.049
 hero-0.045
 CA-0.040
 relieved-0.039
Token happy
Token ID3772
Feature activation-9.16e-01

Pos logit contributions
First+0.009
 brushes+0.008
 sizes+0.008
ups+0.008
 seasons+0.008
Neg logit contributions
 victorious-0.003
 hero-0.003
 sorry-0.003
 CA-0.003
 devastated-0.002
Token and
Token ID290
Feature activation-1.07e+00

Pos logit contributions
 hero+0.025
 victorious+0.024
 sorry+0.022
 CA+0.020
eros+0.020
Neg logit contributions
First-0.041
 brushes-0.037
 sizes-0.034
vals-0.033
ups-0.033
Token shouted
Token ID17293
Feature activation+4.20e-01

Pos logit contributions
 victorious+0.034
 hero+0.034
 CA+0.033
 sorry+0.031
 devastated+0.030
Neg logit contributions
First-0.039
 brushes-0.038
 sizes-0.034
ups-0.034
 seasons-0.033
Token:
Token ID25
Feature activation+3.16e-01

Pos logit contributions
 brushes+0.012
First+0.012
 shapes+0.011
 colors+0.010
ups+0.010
Neg logit contributions
 CA-0.016
 hero-0.015
 victorious-0.015
 devastated-0.015
eros-0.015
Token with
Token ID351
Feature activation-1.61e+00

Pos logit contributions
 victorious+0.035
 hero+0.035
 CA+0.033
 sorry+0.031
 devastated+0.030
Neg logit contributions
First-0.051
 brushes-0.048
 sizes-0.044
ups-0.044
 styles-0.043
Token many
Token ID867
Feature activation+2.38e+00

Pos logit contributions
 CA+0.036
 victorious+0.036
 hero+0.034
 hi+0.033
 hello+0.032
Neg logit contributions
First-0.066
akes-0.064
ups-0.063
vals-0.062
 brushes-0.060
Token green
Token ID4077
Feature activation+3.07e+00

Pos logit contributions
First+0.076
vals+0.070
 brushes+0.069
akes+0.067
ups+0.061
Neg logit contributions
 hero-0.078
 victorious-0.074
 CA-0.070
 lucky-0.069
 hi-0.069
Token needles
Token ID25209
Feature activation+2.86e+00

Pos logit contributions
First+0.109
ups+0.093
 brushes+0.087
vals+0.087
Lisa+0.085
Neg logit contributions
 CA-0.086
 hello-0.081
 hero-0.081
eros-0.079
 hi-0.077
Token.
Token ID13
Feature activation+7.73e-01

Pos logit contributions
First+0.124
 brushes+0.116
 seasons+0.107
 styles+0.105
 flavors+0.104
Neg logit contributions
 victorious-0.070
 hero-0.066
 CA-0.064
 sorry-0.064
 devastated-0.058
Token Lily
Token ID20037
Feature activation+2.40e+00

Pos logit contributions
First+0.026
 brushes+0.025
 sizes+0.023
ups+0.022
 styles+0.022
Neg logit contributions
 victorious-0.027
 hero-0.026
 CA-0.026
 sorry-0.025
 devastated-0.023
Token liked
Token ID8288
Feature activation+2.84e+00

Pos logit contributions
First+0.097
 brushes+0.090
 sizes+0.089
akes+0.088
vals+0.088
Neg logit contributions
 CA-0.056
 hero-0.054
 victorious-0.052
 relieved-0.051
 devastated-0.049
Token the
Token ID262
Feature activation+6.25e-01

Pos logit contributions
First+0.100
 brushes+0.095
 sizes+0.087
ups+0.085
 styles+0.082
Neg logit contributions
 hero-0.091
 victorious-0.091
 CA-0.090
 sorry-0.083
 hi-0.080
Token pine
Token ID20161
Feature activation+1.92e+00

Pos logit contributions
First+0.024
 brushes+0.023
ups+0.022
 sizes+0.021
vals+0.020
Neg logit contributions
 hero-0.018
 CA-0.016
 victorious-0.016
 hi-0.015
 sorry-0.015
Token tree
Token ID5509
Feature activation-3.35e-01

Pos logit contributions
First+0.083
 brushes+0.073
 styles+0.066
ups+0.065
 sizes+0.065
Neg logit contributions
 hero-0.053
 CA-0.049
eros-0.047
 victorious-0.047
 hunter-0.043
Token.
Token ID13
Feature activation+6.75e-01

Pos logit contributions
 victorious+0.009
 hero+0.009
 sorry+0.008
 CA+0.008
 devastated+0.008
Neg logit contributions
First-0.014
 brushes-0.013
ups-0.012
 seasons-0.011
 styles-0.011
Token her
Token ID607
Feature activation-1.30e+00

Pos logit contributions
 victorious+0.045
 hero+0.045
 CA+0.044
 sorry+0.042
 devastated+0.041
Neg logit contributions
First-0.036
 brushes-0.035
 sizes-0.031
ups-0.030
 flour-0.030
Token some
Token ID617
Feature activation+2.84e+00

Pos logit contributions
 CA+0.048
 victorious+0.048
 hero+0.048
 sorry+0.044
eros+0.044
Neg logit contributions
 brushes-0.041
First-0.036
 colors-0.036
 flour-0.035
ups-0.034
Token new
Token ID649
Feature activation-4.65e-03

Pos logit contributions
First+0.092
 brushes+0.090
 sizes+0.080
ups+0.078
 colors+0.076
Neg logit contributions
 victorious-0.102
 hero-0.100
 CA-0.098
 sorry-0.093
 devastated-0.090
Token poppy
Token ID47867
Feature activation-2.03e+00

Pos logit contributions
 victorious+0.000
 hero+0.000
 CA+0.000
 sorry+0.000
 devastated+0.000
Neg logit contributions
First-0.000
 brushes-0.000
 sizes-0.000
ups-0.000
 seasons-0.000
Token seeds
Token ID11904
Feature activation+2.61e+00

Pos logit contributions
 victorious+0.063
 hero+0.061
 CA+0.060
 sorry+0.056
 devastated+0.054
Neg logit contributions
First-0.076
 brushes-0.075
 sizes-0.068
ups-0.066
 colors-0.065
Token as
Token ID355
Feature activation+2.24e+00

Pos logit contributions
First+0.088
 brushes+0.086
 sizes+0.077
ups+0.074
 seasons+0.073
Neg logit contributions
 victorious-0.091
 hero-0.089
 CA-0.086
 sorry-0.083
 devastated-0.080
Token a
Token ID257
Feature activation+7.09e-01

Pos logit contributions
First+0.086
 brushes+0.080
 sizes+0.074
ups+0.071
 seasons+0.071
Neg logit contributions
 victorious-0.070
 hero-0.069
 CA-0.064
 sorry-0.064
 hi-0.059
Token thank
Token ID5875
Feature activation-1.76e+00

Pos logit contributions
First+0.027
 brushes+0.027
 flour+0.024
 colors+0.024
ups+0.023
Neg logit contributions
 victorious-0.020
 CA-0.020
 hero-0.018
eros-0.018
 devastated-0.018
Token you
Token ID345
Feature activation-2.95e+00

Pos logit contributions
 victorious+0.065
 hero+0.064
 CA+0.062
 sorry+0.060
 devastated+0.057
Neg logit contributions
First-0.056
 brushes-0.054
 sizes-0.048
ups-0.046
 seasons-0.045
Token present
Token ID1944
Feature activation-4.06e+00

Pos logit contributions
 victorious+0.100
 hero+0.098
 CA+0.096
 sorry+0.091
 devastated+0.089
Neg logit contributions
First-0.101
 brushes-0.098
 sizes-0.089
ups-0.086
 styles-0.084
Token.
Token ID13
Feature activation-2.90e-01

Pos logit contributions
 victorious+0.136
 hero+0.135
 CA+0.128
 sorry+0.124
 devastated+0.118
Neg logit contributions
First-0.143
 brushes-0.139
 sizes-0.124
 seasons-0.119
ups-0.118
Token arms
Token ID5101
Feature activation+8.09e-01

Pos logit contributions
 victorious+0.031
 CA+0.030
 hero+0.028
 devastated+0.027
 sorry+0.027
Neg logit contributions
 brushes-0.020
 colors-0.018
First-0.018
 shapes-0.017
 styles-0.017
Token up
Token ID510
Feature activation-7.54e-01

Pos logit contributions
First+0.032
 brushes+0.031
 flavors+0.029
 seasons+0.028
 sizes+0.028
Neg logit contributions
 victorious-0.024
 hero-0.023
 sorry-0.021
 devastated-0.020
 CA-0.020
Token high
Token ID1029
Feature activation+7.08e-01

Pos logit contributions
 victorious+0.025
 hero+0.024
 CA+0.023
 sorry+0.022
 devastated+0.021
Neg logit contributions
First-0.028
 brushes-0.027
 sizes-0.024
ups-0.024
 styles-0.023
Token and
Token ID290
Feature activation-7.50e-01

Pos logit contributions
First+0.031
 brushes+0.028
 sizes+0.025
ups+0.025
 flour+0.024
Neg logit contributions
 hero-0.019
 victorious-0.019
 CA-0.016
eros-0.016
 sorry-0.015
Token his
Token ID465
Feature activation-1.85e-01

Pos logit contributions
 victorious+0.022
 hero+0.021
 sorry+0.020
 CA+0.019
 hi+0.019
Neg logit contributions
First-0.030
 brushes-0.028
ups-0.026
 sizes-0.026
 flour-0.026
Token mom
Token ID1995
Feature activation+3.31e+00

Pos logit contributions
 victorious+0.007
 CA+0.007
 hero+0.006
 sorry+0.006
 devastated+0.006
Neg logit contributions
 brushes-0.006
First-0.006
 sizes-0.005
 colors-0.005
 styles-0.005
Token would
Token ID561
Feature activation+3.35e+00

Pos logit contributions
First+0.112
 brushes+0.107
 sizes+0.093
 seasons+0.092
 flour+0.092
Neg logit contributions
 victorious-0.116
 hero-0.115
 CA-0.105
 sorry-0.104
 devastated-0.102
Token wrap
Token ID14441
Feature activation-4.57e-01

Pos logit contributions
First+0.128
 brushes+0.125
ups+0.116
 sizes+0.114
 flour+0.111
Neg logit contributions
 victorious-0.102
 hero-0.101
 sorry-0.094
 CA-0.093
 devastated-0.087
Token him
Token ID683
Feature activation+7.61e-01

Pos logit contributions
 victorious+0.012
 hero+0.012
 sorry+0.010
 hi+0.010
 goodbye+0.010
Neg logit contributions
First-0.020
 brushes-0.019
 sizes-0.017
 seasons-0.017
vals-0.016
Token up
Token ID510
Feature activation-1.78e+00

Pos logit contributions
First+0.027
 brushes+0.026
 sizes+0.024
ups+0.023
 styles+0.023
Neg logit contributions
 victorious-0.025
 CA-0.025
 hero-0.024
 sorry-0.023
 devastated-0.022
Token in
Token ID287
Feature activation-4.44e-01

Pos logit contributions
 victorious+0.060
 hero+0.059
 CA+0.056
 sorry+0.055
 devastated+0.052
Neg logit contributions
First-0.062
 brushes-0.060
 sizes-0.054
ups-0.053
 seasons-0.051
Token,
Token ID11
Feature activation+6.81e-02

Pos logit contributions
First+0.030
 brushes+0.029
 sizes+0.025
 seasons+0.024
vals+0.024
Neg logit contributions
 hero-0.023
 victorious-0.023
 sorry-0.022
 CA-0.019
 unlucky-0.019
Token I
Token ID314
Feature activation-3.91e-01

Pos logit contributions
First+0.003
 brushes+0.002
ups+0.002
 sizes+0.002
 styles+0.002
Neg logit contributions
 victorious-0.002
 hero-0.002
 sorry-0.002
 CA-0.002
 hi-0.002
Token will
Token ID481
Feature activation+2.80e+00

Pos logit contributions
 victorious+0.015
 hero+0.014
 CA+0.014
 sorry+0.013
 devastated+0.013
Neg logit contributions
First-0.012
 brushes-0.012
 sizes-0.011
ups-0.010
 styles-0.010
Token have
Token ID423
Feature activation-4.99e-01

Pos logit contributions
First+0.117
 brushes+0.115
 sizes+0.107
ups+0.104
 flour+0.101
Neg logit contributions
 victorious-0.075
 hero-0.074
 sorry-0.072
 CA-0.066
 devastated-0.063
Token to
Token ID284
Feature activation-8.70e-01

Pos logit contributions
 victorious+0.015
 hero+0.014
 sorry+0.014
 CA+0.014
 devastated+0.013
Neg logit contributions
First-0.019
 brushes-0.018
 sizes-0.017
 seasons-0.016
 styles-0.016
Token squeeze
Token ID21229
Feature activation+1.59e+00

Pos logit contributions
 victorious+0.024
 hero+0.024
 sorry+0.023
 CA+0.021
 devastated+0.020
Neg logit contributions
First-0.036
 brushes-0.035
 sizes-0.033
ups-0.032
 seasons-0.030
Token you
Token ID345
Feature activation-1.40e+00

Pos logit contributions
First+0.064
 brushes+0.061
 sizes+0.056
vals+0.053
ups+0.053
Neg logit contributions
 hero-0.044
 victorious-0.043
 sorry-0.041
 CA-0.038
 hi-0.037
Token!"
Token ID2474
Feature activation+2.08e+00

Pos logit contributions
 victorious+0.043
 hero+0.042
 sorry+0.039
 CA+0.039
 devastated+0.037
Neg logit contributions
First-0.053
 brushes-0.052
 sizes-0.047
ups-0.045
 seasons-0.045
Token Mary
Token ID5335
Feature activation-2.70e-01

Pos logit contributions
First+0.080
 brushes+0.078
 sizes+0.071
ups+0.069
 colors+0.067
Neg logit contributions
 victorious-0.062
 hero-0.061
 CA-0.059
 sorry-0.056
 devastated-0.054
Token was
Token ID373
Feature activation+8.53e-02

Pos logit contributions
 victorious+0.009
 hero+0.009
 CA+0.009
 sorry+0.008
 devastated+0.008
Neg logit contributions
First-0.010
 brushes-0.009
 sizes-0.008
ups-0.008
 flour-0.008
Token very
Token ID845
Feature activation-1.59e+00

Pos logit contributions
First+0.003
 brushes+0.003
 sizes+0.003
ups+0.003
 colors+0.003
Neg logit contributions
 hero-0.002
 CA-0.002
 victorious-0.002
eros-0.002
 hi-0.002
Token George
Token ID4502
Feature activation+1.74e+00

Pos logit contributions
 victorious+0.012
 hero+0.012
 CA+0.012
 sorry+0.011
 devastated+0.010
Neg logit contributions
First-0.018
 brushes-0.018
 sizes-0.016
ups-0.016
 seasons-0.016
Token felt
Token ID2936
Feature activation-1.11e+00

Pos logit contributions
First+0.059
 brushes+0.057
 sizes+0.051
ups+0.049
 styles+0.049
Neg logit contributions
 victorious-0.061
 hero-0.060
 CA-0.058
 sorry-0.055
 devastated-0.054
Token so
Token ID523
Feature activation+6.42e-01

Pos logit contributions
 CA+0.024
 hero+0.024
 victorious+0.024
eros+0.023
 hi+0.020
Neg logit contributions
First-0.047
 brushes-0.047
 colors-0.044
 sizes-0.044
 flour-0.043
Token sad
Token ID6507
Feature activation-7.77e-01

Pos logit contributions
 brushes+0.029
First+0.029
 colors+0.027
 shapes+0.026
ups+0.025
Neg logit contributions
 hero-0.014
 victorious-0.013
 CA-0.013
eros-0.013
 hi-0.011
Token.
Token ID13
Feature activation-2.78e-01

Pos logit contributions
 hero+0.025
 victorious+0.025
 CA+0.023
 sorry+0.023
 devastated+0.021
Neg logit contributions
First-0.029
 brushes-0.028
 sizes-0.025
ups-0.024
 flour-0.024
Token He
Token ID679
Feature activation-5.54e-01

Pos logit contributions
 victorious+0.008
 hero+0.008
 CA+0.007
 sorry+0.007
 devastated+0.007
Neg logit contributions
First-0.011
 brushes-0.011
 sizes-0.010
ups-0.010
 seasons-0.010
Token wanted
Token ID2227
Feature activation-6.29e-01

Pos logit contributions
 victorious+0.019
 hero+0.019
 CA+0.019
 sorry+0.018
 devastated+0.017
Neg logit contributions
First-0.019
 brushes-0.018
 sizes-0.016
ups-0.016
 flour-0.015
Token to
Token ID284
Feature activation-1.48e+00

Pos logit contributions
 hero+0.019
 victorious+0.019
 sorry+0.018
 CA+0.017
 devastated+0.016
Neg logit contributions
First-0.025
 brushes-0.023
 sizes-0.022
akes-0.021
vals-0.021
Token fix
Token ID4259
Feature activation+2.83e-01

Pos logit contributions
 victorious+0.046
 hero+0.045
 CA+0.043
 sorry+0.041
 devastated+0.039
Neg logit contributions
First-0.056
 brushes-0.055
 sizes-0.050
ups-0.048
 colors-0.047
Token it
Token ID340
Feature activation-1.60e+00

Pos logit contributions
First+0.012
 brushes+0.011
ups+0.010
 sizes+0.010
 seasons+0.010
Neg logit contributions
 victorious-0.008
 hero-0.008
 sorry-0.007
 CA-0.007
 hi-0.007
Token,
Token ID11
Feature activation+1.37e+00

Pos logit contributions
 CA+0.061
 victorious+0.061
 hero+0.060
eros+0.056
 devastated+0.056
Neg logit contributions
First-0.048
 brushes-0.045
 sizes-0.041
 colors-0.039
 styles-0.038
Token the
Token ID262
Feature activation-7.23e-01

Pos logit contributions
 brushes+0.018
First+0.018
 flour+0.015
 colors+0.015
 sizes+0.015
Neg logit contributions
 victorious-0.020
 CA-0.019
 hero-0.019
 hi-0.017
 sorry-0.017
Token end
Token ID886
Feature activation+2.18e+00

Pos logit contributions
 hero+0.023
 victorious+0.023
 CA+0.022
 sorry+0.021
 devastated+0.020
Neg logit contributions
First-0.027
 brushes-0.026
 sizes-0.024
ups-0.023
 styles-0.023
Token,
Token ID11
Feature activation+2.98e-01

Pos logit contributions
First+0.092
 brushes+0.086
 sizes+0.085
 flavors+0.081
 styles+0.079
Neg logit contributions
 sorry-0.058
 victorious-0.056
 hero-0.055
 CA-0.050
 devastated-0.047
Token Lily
Token ID20037
Feature activation+1.39e+00

Pos logit contributions
First+0.013
 brushes+0.012
 sizes+0.012
ups+0.011
 flavors+0.011
Neg logit contributions
 hero-0.008
 sorry-0.008
 victorious-0.008
 CA-0.007
 hi-0.006
Token's
Token ID338
Feature activation+2.28e+00

Pos logit contributions
First+0.052
 brushes+0.050
 sizes+0.048
 styles+0.045
 seasons+0.045
Neg logit contributions
 victorious-0.043
 hero-0.042
 CA-0.041
 devastated-0.039
 sorry-0.039
Token grandma
Token ID49890
Feature activation+8.72e-01

Pos logit contributions
 brushes+0.062
 colors+0.054
First+0.054
 flour+0.052
 shapes+0.049
Neg logit contributions
 victorious-0.097
 CA-0.095
 hero-0.088
eros-0.085
 hi-0.084
Token came
Token ID1625
Feature activation-2.02e+00

Pos logit contributions
First+0.030
 brushes+0.029
 sizes+0.026
ups+0.025
 flour+0.024
Neg logit contributions
 victorious-0.030
 hero-0.030
 CA-0.029
 sorry-0.027
 devastated-0.027
Token and
Token ID290
Feature activation-1.10e+00

Pos logit contributions
 victorious+0.066
 hero+0.065
 CA+0.062
 sorry+0.060
 devastated+0.058
Neg logit contributions
First-0.073
 brushes-0.072
 sizes-0.065
ups-0.062
 seasons-0.061
Token they
Token ID484
Feature activation+7.42e-01

Pos logit contributions
 victorious+0.038
 hero+0.037
 CA+0.036
 sorry+0.035
 devastated+0.033
Neg logit contributions
First-0.038
 brushes-0.037
 sizes-0.033
ups-0.032
 flour-0.031
Token had
Token ID550
Feature activation+7.02e-01

Pos logit contributions
First+0.024
 brushes+0.023
 sizes+0.021
ups+0.020
 colors+0.020
Neg logit contributions
 victorious-0.027
 hero-0.026
 CA-0.026
 sorry-0.025
 devastated-0.024
Token a
Token ID257
Feature activation+3.03e-01

Pos logit contributions
First+0.026
 brushes+0.025
 sizes+0.025
 seasons+0.023
vals+0.023
Neg logit contributions
 victorious-0.021
 hero-0.020
 CA-0.019
 sorry-0.019
 devastated-0.019
Token did
Token ID750
Feature activation+1.56e+00

Pos logit contributions
 victorious+0.018
 hero+0.018
 CA+0.017
 sorry+0.017
 devastated+0.016
Neg logit contributions
First-0.017
 brushes-0.016
 sizes-0.015
ups-0.015
 styles-0.014
Token it
Token ID340
Feature activation-1.81e+00

Pos logit contributions
First+0.065
 brushes+0.060
 sizes+0.055
ups+0.055
 seasons+0.054
Neg logit contributions
 victorious-0.047
 sorry-0.044
 hero-0.043
 CA-0.039
 hi-0.038
Token!"
Token ID2474
Feature activation+1.41e+00

Pos logit contributions
 victorious+0.056
 hero+0.056
 sorry+0.050
 CA+0.048
 hi+0.045
Neg logit contributions
 brushes-0.070
First-0.069
 sizes-0.061
 flour-0.060
 seasons-0.059
Token he
Token ID339
Feature activation+4.28e-01

Pos logit contributions
First+0.047
 brushes+0.045
 sizes+0.041
ups+0.040
 styles+0.039
Neg logit contributions
 victorious-0.049
 hero-0.048
 CA-0.047
 sorry-0.045
 devastated-0.044
Token exclaimed
Token ID33552
Feature activation+3.84e-02

Pos logit contributions
First+0.014
 brushes+0.013
 sizes+0.012
 seasons+0.012
 flour+0.012
Neg logit contributions
 victorious-0.015
 hero-0.015
 CA-0.014
 sorry-0.014
 devastated-0.013
Token.
Token ID13
Feature activation-1.24e+00

Pos logit contributions
First+0.001
 brushes+0.001
ups+0.001
 sizes+0.001
 colors+0.001
Neg logit contributions
 CA-0.002
 hero-0.002
 victorious-0.002
 devastated-0.001
eros-0.001
Token
Token ID198
Feature activation-1.05e+00

Pos logit contributions
 hero+0.039
 victorious+0.039
 CA+0.037
 devastated+0.035
 sorry+0.035
Neg logit contributions
First-0.046
 brushes-0.044
 sizes-0.039
ups-0.038
 flour-0.037
Token
Token ID198
Feature activation-1.58e+00

Pos logit contributions
 hero+0.033
 victorious+0.032
 CA+0.031
 sorry+0.030
 devastated+0.029
Neg logit contributions
First-0.040
 brushes-0.038
ups-0.034
 sizes-0.033
 flour-0.032
TokenHis
Token ID6653
Feature activation+3.18e-01

Pos logit contributions
 victorious+0.061
 hero+0.060
 CA+0.058
 sorry+0.055
 devastated+0.054
Neg logit contributions
 brushes-0.047
First-0.046
 sizes-0.042
 colors-0.039
 flour-0.039
Token mother
Token ID2802
Feature activation+1.91e+00

Pos logit contributions
 brushes+0.011
First+0.011
 sizes+0.009
 colors+0.009
ups+0.009
Neg logit contributions
 victorious-0.011
 CA-0.011
 hero-0.011
 sorry-0.010
 devastated-0.010
Token smiled
Token ID13541
Feature activation-3.38e-01

Pos logit contributions
First+0.065
 brushes+0.063
 sizes+0.057
ups+0.055
 colors+0.054
Neg logit contributions
 victorious-0.066
 hero-0.064
 CA-0.063
 sorry-0.060
 devastated-0.058
Token little
Token ID1310
Feature activation-2.18e+00

Pos logit contributions
 victorious+0.082
 CA+0.072
 devastated+0.071
 hero+0.064
 sorry+0.064
Neg logit contributions
 brushes-0.114
First-0.108
 colors-0.106
 seasons-0.104
 sizes-0.104
Token girl
Token ID2576
Feature activation-4.09e+00

Pos logit contributions
 hero+0.054
 victorious+0.054
 CA+0.051
 sorry+0.048
 devastated+0.044
Neg logit contributions
First-0.096
 brushes-0.093
 sizes-0.086
ups-0.085
 styles-0.082
Token named
Token ID3706
Feature activation-1.28e+00

Pos logit contributions
 hero+0.056
 victorious+0.050
 CA+0.049
 sorry+0.048
 devastated+0.039
Neg logit contributions
First-0.226
 brushes-0.214
 sizes-0.204
 flavors-0.198
 styles-0.197
Token Daisy
Token ID40355
Feature activation-2.07e+00

Pos logit contributions
 victorious+0.038
 hero+0.038
 CA+0.037
 sorry+0.035
 devastated+0.033
Neg logit contributions
First-0.049
 brushes-0.048
 sizes-0.045
ups-0.043
 flavors-0.042
Token.
Token ID13
Feature activation-1.39e+00

Pos logit contributions
 victorious+0.069
 hero+0.069
 CA+0.067
 sorry+0.063
 devastated+0.061
Neg logit contributions
First-0.072
 brushes-0.070
 sizes-0.062
ups-0.061
 flour-0.059
Token Daisy
Token ID40355
Feature activation+3.11e-01

Pos logit contributions
 victorious+0.042
 hero+0.042
 CA+0.041
 sorry+0.039
 hi+0.036
Neg logit contributions
 brushes-0.051
First-0.051
ups-0.048
 sizes-0.046
 styles-0.045
Token was
Token ID373
Feature activation+2.83e-02

Pos logit contributions
First+0.013
 brushes+0.012
ups+0.011
 sizes+0.011
 flour+0.011
Neg logit contributions
 CA-0.009
 hero-0.009
 victorious-0.008
 sorry-0.008
 devastated-0.007
Token playing
Token ID2712
Feature activation-1.05e-01

Pos logit contributions
First+0.001
 brushes+0.001
 sizes+0.001
 flour+0.001
 colors+0.001
Neg logit contributions
 victorious-0.001
 CA-0.001
 hero-0.001
 devastated-0.001
eros-0.001
Token in
Token ID287
Feature activation+4.62e-02

Pos logit contributions
 victorious+0.003
 hero+0.003
 CA+0.003
 sorry+0.003
 devastated+0.003
Neg logit contributions
First-0.004
 brushes-0.004
 sizes-0.004
ups-0.003
 styles-0.003
Token her
Token ID607
Feature activation-3.74e-01

Pos logit contributions
First+0.002
 brushes+0.002
ups+0.002
 sizes+0.002
 flavors+0.002
Neg logit contributions
 victorious-0.001
 hero-0.001
 CA-0.001
 sorry-0.001
 devastated-0.001
Token front
Token ID2166
Feature activation+8.97e-01

Pos logit contributions
 victorious+0.016
 hero+0.015
 CA+0.015
 sorry+0.014
 devastated+0.014
Neg logit contributions
First-0.010
 brushes-0.010
 sizes-0.008
ups-0.008
 colors-0.008
Token her
Token ID607
Feature activation-9.07e-01

Pos logit contributions
First+0.067
 brushes+0.063
 sizes+0.062
ups+0.060
 flavors+0.056
Neg logit contributions
 sorry-0.062
 hero-0.061
 victorious-0.058
 goodbye-0.057
 CA-0.056
Token mom
Token ID1995
Feature activation+2.71e+00

Pos logit contributions
 victorious+0.033
 CA+0.032
 hero+0.031
 sorry+0.029
 devastated+0.029
Neg logit contributions
First-0.029
 brushes-0.029
 sizes-0.025
ups-0.025
 colors-0.025
Tokenmy
Token ID1820
Feature activation+2.53e+00

Pos logit contributions
First+0.097
 brushes+0.090
 sizes+0.089
 styles+0.086
vals+0.085
Neg logit contributions
 hero-0.077
 sorry-0.076
 devastated-0.074
 sad-0.071
 relieved-0.071
Token and
Token ID290
Feature activation-7.52e-01

Pos logit contributions
First+0.092
 brushes+0.089
 sizes+0.084
 styles+0.080
 seasons+0.078
Neg logit contributions
 hero-0.081
 victorious-0.079
 sorry-0.075
 CA-0.075
 devastated-0.073
Token daddy
Token ID35822
Feature activation+1.44e+00

Pos logit contributions
 hero+0.019
 sorry+0.019
 CA+0.017
 victorious+0.017
 devastated+0.016
Neg logit contributions
First-0.031
 brushes-0.029
 sizes-0.028
vals-0.028
ups-0.027
Token said
Token ID531
Feature activation-6.53e-01

Pos logit contributions
 brushes+0.049
First+0.049
 sizes+0.045
 seasons+0.043
 styles+0.043
Neg logit contributions
 hero-0.048
 victorious-0.047
 sorry-0.046
 CA-0.045
 devastated-0.045
Token,
Token ID11
Feature activation+5.74e-01

Pos logit contributions
 sorry+0.021
 hero+0.020
 CA+0.019
 victorious+0.019
 hi+0.018
Neg logit contributions
First-0.026
 brushes-0.024
 sizes-0.023
 styles-0.021
ups-0.021
Token "
Token ID366
Feature activation+6.63e-01

Pos logit contributions
First+0.028
 brushes+0.027
 sizes+0.025
ups+0.025
 styles+0.024
Neg logit contributions
 hero-0.011
 victorious-0.011
 CA-0.011
 sorry-0.011
 devastated-0.009
TokenL
Token ID43
Feature activation+8.91e-01

Pos logit contributions
 brushes+0.025
 sizes+0.023
 styles+0.022
 flavors+0.022
vals+0.021
Neg logit contributions
 CA-0.018
 devastated-0.017
 hero-0.017
 Start-0.016
 victorious-0.016
Tokenily
Token ID813
Feature activation+1.31e+00

Pos logit contributions
First+0.027
 brushes+0.026
ups+0.022
 seasons+0.022
 sizes+0.022
Neg logit contributions
 victorious-0.035
 hero-0.034
 CA-0.033
 sorry-0.032
 devastated-0.032
Token,
Token ID11
Feature activation+5.69e-01

Pos logit contributions
First+0.056
 brushes+0.052
 sizes+0.048
 styles+0.047
ups+0.046
Neg logit contributions
 hero-0.036
 sorry-0.035
 victorious-0.034
 CA-0.033
 devastated-0.030
Token.
Token ID13
Feature activation-5.88e-01

Pos logit contributions
 hero+0.076
 victorious+0.076
 sorry+0.071
 CA+0.070
 devastated+0.065
Neg logit contributions
First-0.085
 brushes-0.082
 sizes-0.076
 seasons-0.072
ups-0.071
Token It
Token ID632
Feature activation-3.76e-01

Pos logit contributions
 victorious+0.014
 hero+0.013
 CA+0.013
 sorry+0.012
 devastated+0.011
Neg logit contributions
First-0.027
 brushes-0.026
 sizes-0.024
ups-0.024
 colors-0.023
Token's
Token ID338
Feature activation+4.23e+00

Pos logit contributions
 victorious+0.012
 hero+0.012
 CA+0.011
 sorry+0.011
 devastated+0.010
Neg logit contributions
First-0.014
 brushes-0.013
 sizes-0.012
ups-0.012
 styles-0.012
Token important
Token ID1593
Feature activation+2.42e+00

Pos logit contributions
First+0.169
 brushes+0.160
ups+0.149
 sizes+0.147
 flavors+0.142
Neg logit contributions
 sorry-0.119
 victorious-0.116
 hero-0.114
 CA-0.112
 unlucky-0.103
Token to
Token ID284
Feature activation-6.51e-01

Pos logit contributions
First+0.100
 brushes+0.091
 sizes+0.084
vals+0.083
 flour+0.083
Neg logit contributions
 hero-0.064
 goodbye-0.060
 sorry-0.059
 CA-0.057
 victorious-0.056
Token look
Token ID804
Feature activation-4.79e+00

Pos logit contributions
 hero+0.019
 victorious+0.019
 CA+0.018
 sorry+0.018
 devastated+0.016
Neg logit contributions
First-0.025
 brushes-0.024
 sizes-0.023
ups-0.022
 flour-0.021
Token nice
Token ID3621
Feature activation+7.52e-01

Pos logit contributions
 victorious+0.145
 CA+0.145
 hero+0.143
 devastated+0.127
eros+0.127
Neg logit contributions
First-0.178
 brushes-0.177
 sizes-0.158
ups-0.158
 colors-0.156
Token for
Token ID329
Feature activation+2.26e+00

Pos logit contributions
First+0.032
 brushes+0.030
vals+0.028
 seasons+0.027
 sizes+0.027
Neg logit contributions
 hero-0.019
 goodbye-0.017
 victorious-0.017
 sorry-0.017
 hi-0.015
Token our
Token ID674
Feature activation-3.81e-02

Pos logit contributions
First+0.080
 brushes+0.076
 sizes+0.068
ups+0.066
 seasons+0.064
Neg logit contributions
 victorious-0.076
 hero-0.076
 sorry-0.072
 CA-0.072
 hi-0.066
Token family
Token ID1641
Feature activation-7.01e-01

Pos logit contributions
 hero+0.001
 victorious+0.001
 sorry+0.001
 CA+0.001
 goodbye+0.001
Neg logit contributions
First-0.001
 brushes-0.001
 sizes-0.001
ups-0.001
vals-0.001
Token."
Token ID526
Feature activation+1.35e+00

Pos logit contributions
 hero+0.023
 victorious+0.022
 sorry+0.020
 CA+0.020
 devastated+0.019
Neg logit contributions
First-0.025
 brushes-0.025
 sizes-0.021
 seasons-0.021
ups-0.021
Token Her
Token ID2332
Feature activation+3.50e-01

Pos logit contributions
 victorious+0.017
 hero+0.017
 CA+0.016
 devastated+0.015
 sorry+0.015
Neg logit contributions
First-0.021
 brushes-0.021
 sizes-0.018
ups-0.018
 flour-0.018
Token mom
Token ID1995
Feature activation+2.90e+00

Pos logit contributions
First+0.012
 brushes+0.011
 sizes+0.010
ups+0.010
 colors+0.010
Neg logit contributions
 victorious-0.012
 hero-0.012
 CA-0.012
 sorry-0.011
 devastated-0.011
Token said
Token ID531
Feature activation-1.94e-01

Pos logit contributions
First+0.098
 brushes+0.097
 sizes+0.085
ups+0.084
 colors+0.082
Neg logit contributions
 victorious-0.100
 hero-0.097
 CA-0.097
 sorry-0.090
 devastated-0.088
Token,
Token ID11
Feature activation+6.66e-01

Pos logit contributions
 victorious+0.006
 hero+0.006
 CA+0.006
 devastated+0.006
eros+0.005
Neg logit contributions
First-0.007
 brushes-0.007
 flour-0.006
 sizes-0.006
ups-0.006
Token "
Token ID366
Feature activation+1.13e+00

Pos logit contributions
First+0.025
 brushes+0.025
 flour+0.022
 sizes+0.022
 colors+0.021
Neg logit contributions
 victorious-0.021
 hero-0.020
 CA-0.020
 devastated-0.018
eros-0.017
TokenOf
Token ID5189
Feature activation-3.28e+00

Pos logit contributions
 brushes+0.042
 sizes+0.042
 seasons+0.039
 flour+0.038
 styles+0.038
Neg logit contributions
 hero-0.031
 victorious-0.028
 CA-0.027
 devastated-0.026
 Start-0.026
Token course
Token ID1781
Feature activation-2.67e+00

Pos logit contributions
 victorious+0.099
 hero+0.095
 CA+0.093
 sorry+0.087
 devastated+0.085
Neg logit contributions
First-0.127
 brushes-0.124
 sizes-0.112
ups-0.111
 colors-0.109
Token,
Token ID11
Feature activation-3.57e-02

Pos logit contributions
 victorious+0.103
 CA+0.100
 hero+0.098
 devastated+0.094
 sorry+0.090
Neg logit contributions
 brushes-0.078
First-0.077
 sizes-0.068
ups-0.067
 colors-0.066
Token let
Token ID1309
Feature activation-1.04e+00

Pos logit contributions
 hero+0.001
 victorious+0.001
 sorry+0.001
 CA+0.001
 hi+0.001
Neg logit contributions
First-0.001
 brushes-0.001
 sizes-0.001
ups-0.001
 seasons-0.001
Token's
Token ID338
Feature activation+1.60e+00

Pos logit contributions
 victorious+0.037
 hero+0.035
 CA+0.034
 sorry+0.032
 devastated+0.032
Neg logit contributions
First-0.036
 brushes-0.035
ups-0.032
 flour-0.031
 colors-0.030
Token go
Token ID467
Feature activation-1.84e+00

Pos logit contributions
First+0.062
 brushes+0.060
 sizes+0.055
ups+0.053
 styles+0.052
Neg logit contributions
 victorious-0.048
 hero-0.048
 CA-0.046
 sorry-0.044
 devastated-0.042
TokenOn
Token ID2202
Feature activation+1.10e-01

Pos logit contributions
 victorious+0.045
 CA+0.045
 hero+0.044
 devastated+0.041
 sorry+0.039
Neg logit contributions
 brushes-0.039
 sizes-0.034
First-0.034
 styles-0.033
 dots-0.032
Token the
Token ID262
Feature activation-3.63e-01

Pos logit contributions
First+0.004
 brushes+0.004
 sizes+0.004
ups+0.004
 styles+0.003
Neg logit contributions
 victorious-0.003
 hero-0.003
 CA-0.003
 sorry-0.003
 devastated-0.003
Token day
Token ID1110
Feature activation-8.38e-02

Pos logit contributions
 victorious+0.013
 hero+0.013
 sorry+0.012
 CA+0.012
 devastated+0.012
Neg logit contributions
First-0.012
 brushes-0.011
 sizes-0.010
 colors-0.010
 seasons-0.010
Token of
Token ID286
Feature activation+9.61e-01

Pos logit contributions
 victorious+0.003
 hero+0.003
 CA+0.003
 sorry+0.002
 devastated+0.002
Neg logit contributions
First-0.003
 brushes-0.003
 sizes-0.003
ups-0.003
 styles-0.003
Token the
Token ID262
Feature activation-3.11e-01

Pos logit contributions
First+0.025
 brushes+0.024
 colors+0.021
 flour+0.021
 seasons+0.020
Neg logit contributions
 victorious-0.040
 hero-0.040
 CA-0.038
 devastated-0.038
onel-0.036
Token trip
Token ID5296
Feature activation-3.23e+00

Pos logit contributions
 victorious+0.012
 hero+0.011
 sorry+0.011
 CA+0.010
 devastated+0.010
Neg logit contributions
First-0.010
 brushes-0.009
 colors-0.009
 seasons-0.008
ups-0.008
Token,
Token ID11
Feature activation+6.69e-01

Pos logit contributions
 victorious+0.111
 hero+0.109
 CA+0.106
 sorry+0.101
 devastated+0.098
Neg logit contributions
First-0.110
 brushes-0.107
 sizes-0.096
ups-0.093
 colors-0.091
Token Tim
Token ID5045
Feature activation+2.20e+00

Pos logit contributions
First+0.025
 brushes+0.025
 sizes+0.023
ups+0.023
 styles+0.022
Neg logit contributions
 victorious-0.020
 hero-0.020
 CA-0.019
 sorry-0.019
 devastated-0.017
Tokenmy
Token ID1820
Feature activation+2.14e+00

Pos logit contributions
First+0.049
 brushes+0.049
 sizes+0.041
 styles+0.038
ups+0.038
Neg logit contributions
 hero-0.099
 victorious-0.098
 CA-0.096
 sorry-0.091
 devastated-0.091
Token's
Token ID338
Feature activation+2.39e+00

Pos logit contributions
First+0.075
 brushes+0.073
 sizes+0.065
ups+0.064
 colors+0.062
Neg logit contributions
 victorious-0.072
 hero-0.071
 CA-0.068
 sorry-0.066
 devastated-0.063
Token mom
Token ID1995
Feature activation+2.86e+00

Pos logit contributions
 brushes+0.060
First+0.059
 colors+0.054
ups+0.051
 flour+0.050
Neg logit contributions
 victorious-0.101
 CA-0.099
smart-0.092
 hero-0.088
 devastated-0.087
Token amazed
Token ID24562
Feature activation-1.25e+00

Pos logit contributions
First+0.043
 brushes+0.041
ups+0.039
oons+0.038
akes+0.038
Neg logit contributions
 victorious-0.021
 sorry-0.020
 relieved-0.018
 hero-0.017
 devastated-0.017
Token at
Token ID379
Feature activation-1.71e-01

Pos logit contributions
 hero+0.032
 victorious+0.032
 sorry+0.028
 hi+0.026
 CA+0.026
Neg logit contributions
First-0.055
 brushes-0.050
 sizes-0.046
 flour-0.046
vals-0.045
Token all
Token ID477
Feature activation+8.26e-01

Pos logit contributions
 victorious+0.004
 hero+0.004
 CA+0.004
 sorry+0.004
 devastated+0.004
Neg logit contributions
First-0.007
 brushes-0.007
 sizes-0.007
ups-0.006
 colors-0.006
Token the
Token ID262
Feature activation+9.90e-01

Pos logit contributions
First+0.033
 brushes+0.032
 sizes+0.029
ups+0.029
 flavors+0.028
Neg logit contributions
 victorious-0.022
 hero-0.022
 CA-0.021
 sorry-0.020
 devastated-0.019
Token weird
Token ID7650
Feature activation+1.79e+00

Pos logit contributions
First+0.032
vals+0.031
w+0.029
 yeast+0.029
ups+0.028
Neg logit contributions
 hero-0.029
 lucky-0.027
 threat-0.026
 hello-0.026
 little-0.026
Token looking
Token ID2045
Feature activation-3.82e+00

Pos logit contributions
First+0.065
vals+0.050
Mom+0.049
 brushes+0.048
w+0.047
Neg logit contributions
 hero-0.055
 hello-0.052
 hi-0.051
 threat-0.050
 sorry-0.049
Token leaves
Token ID5667
Feature activation+4.34e+00

Pos logit contributions
 hero+0.129
 victorious+0.127
 CA+0.122
 sorry+0.120
 hi+0.111
Neg logit contributions
First-0.136
 brushes-0.125
 sizes-0.113
ups-0.110
 styles-0.107
Token she
Token ID673
Feature activation+1.01e+00

Pos logit contributions
First+0.177
 brushes+0.163
ups+0.146
 seasons+0.146
 styles+0.145
Neg logit contributions
 victorious-0.125
 hero-0.117
 sorry-0.111
 CA-0.110
 devastated-0.103
Token found
Token ID1043
Feature activation-5.84e-01

Pos logit contributions
First+0.034
 brushes+0.033
 sizes+0.029
ups+0.028
 seasons+0.028
Neg logit contributions
 victorious-0.036
 hero-0.035
 CA-0.034
 sorry-0.032
 devastated-0.031
Token.
Token ID13
Feature activation-1.06e-01

Pos logit contributions
 victorious+0.017
 hero+0.017
 CA+0.015
 sorry+0.015
 devastated+0.014
Neg logit contributions
First-0.023
 brushes-0.022
 sizes-0.020
 seasons-0.019
 styles-0.019
Token But
Token ID887
Feature activation+1.57e+00

Pos logit contributions
 victorious+0.004
 hero+0.004
 CA+0.003
 sorry+0.003
 devastated+0.003
Neg logit contributions
First-0.004
 brushes-0.004
 sizes-0.003
ups-0.003
 seasons-0.003
Token's
Token ID338
Feature activation+2.26e+00

Pos logit contributions
 hero+0.055
 CA+0.054
 victorious+0.052
 sorry+0.051
 hi+0.047
Neg logit contributions
First-0.062
 brushes-0.056
 sizes-0.050
ups-0.049
 styles-0.048
Token enough
Token ID1576
Feature activation+6.09e-01

Pos logit contributions
 brushes+0.084
First+0.083
 sizes+0.076
 flour+0.074
 colors+0.074
Neg logit contributions
 victorious-0.069
 hero-0.067
 CA-0.067
 devastated-0.060
 sorry-0.059
Token for
Token ID329
Feature activation+1.50e+00

Pos logit contributions
First+0.025
 brushes+0.022
 sizes+0.020
ups+0.020
vals+0.019
Neg logit contributions
 hero-0.017
 sorry-0.017
 victorious-0.017
 CA-0.015
 hi-0.015
Token today
Token ID1909
Feature activation+1.67e+00

Pos logit contributions
First+0.054
 brushes+0.051
 sizes+0.047
ups+0.045
 styles+0.043
Neg logit contributions
 victorious-0.050
 hero-0.049
 sorry-0.047
 CA-0.047
 devastated-0.043
Token.
Token ID13
Feature activation-6.97e-01

Pos logit contributions
First+0.069
 brushes+0.066
 sizes+0.059
vals+0.057
 styles+0.056
Neg logit contributions
 hero-0.044
 victorious-0.043
 sorry-0.041
 CA-0.040
 goodbye-0.036
Token How
Token ID1374
Feature activation-3.27e+00

Pos logit contributions
 hero+0.017
 victorious+0.017
 CA+0.016
 sorry+0.016
 hi+0.014
Neg logit contributions
First-0.030
 brushes-0.030
 sizes-0.027
ups-0.027
 colors-0.027
Token about
Token ID546
Feature activation-1.14e+00

Pos logit contributions
 hero+0.084
 victorious+0.083
 CA+0.081
 sorry+0.081
 hi+0.071
Neg logit contributions
First-0.137
 brushes-0.128
 sizes-0.125
ups-0.119
 styles-0.118
Token we
Token ID356
Feature activation-5.65e-01

Pos logit contributions
 victorious+0.035
 hero+0.035
 CA+0.034
 sorry+0.033
 hi+0.030
Neg logit contributions
First-0.043
 brushes-0.041
 sizes-0.038
ups-0.036
 styles-0.036
Token have
Token ID423
Feature activation-7.62e-01

Pos logit contributions
 hero+0.017
 victorious+0.017
 sorry+0.016
 CA+0.015
 devastated+0.014
Neg logit contributions
First-0.022
 brushes-0.021
 sizes-0.020
ups-0.019
 styles-0.019
Token a
Token ID257
Feature activation-2.58e-01

Pos logit contributions
 victorious+0.023
 hero+0.023
 CA+0.022
 sorry+0.022
 devastated+0.020
Neg logit contributions
First-0.029
 brushes-0.027
 sizes-0.026
ups-0.024
 seasons-0.024
Token snack
Token ID26906
Feature activation+1.27e+00

Pos logit contributions
 victorious+0.008
 CA+0.008
 hero+0.007
 devastated+0.007
eros+0.007
Neg logit contributions
 brushes-0.009
First-0.009
 flour-0.008
 colors-0.008
 seasons-0.008