TOP ACTIVATIONS
MAX = 0.300

not a monster , my de ars . That is a
was a dream , my de ars ," she said .
He was a mis chie vous little baby and always got
happy , it was a brand new ice cream machine .
" That , my de ars , is a cave
" Hello , my de ars !" Grand ma says
a special gift - a brand new bath robe . It
. " Hello , my de ars ! What a nice
re welcome , my de ars . Enjoy your cake
That 's good , my de ars . But you still
. She had designed a de vious test ! The fairy
bright yellow colour and a fr illy style . She tried
novel ! It was a sweet novel with a bright pink
amazing . It was a brand new toy . They both
am not mad , my de ars . I am glad
tigers , but he was most excited to see the elephants
are not flowers , my de ars . Those are caul
" Hello , my de ars . What do you
" Hello , my de ars ," she said .
Peter was a mis chie vous 3 year old who liked

MIN ACTIVATIONS
MIN = -0.294

on her face . Ready to greet her friend , Molly
apped and shouted " ready to go !" She was ready
if you fall . Ready to try again ?"
in her bag . Ready to go , she ran back
. " Time to bat he , L ila
. You should go back to your own country and play
. " Time to get up , my loves
. " Time to go , kids !" she
the park . " Ready to make the k ite work
will be able to climb to the top of the tallest
big zip ! " Ready to go !" she shouted .
. We should go back to the park . Mom and
he had managed to climb to the top . And he
. " Time to go , Tommy ," she
. " Ready to catch some fish ?" asked
had been able to climb to the top . After a
use it to climb up to the balloon .
. " Time to go home , kids ,"
and said , " Time to go , kids . The
to the hill and climbed to the top . They looked

INTERVAL 0.151 - 0.300
CONTAINS 4.372%

Once upon a time , a little girl named Lily went
, the daughter saw a pretty dress . She asked her
a time , there was a little boy named Tim my
right away . Lucy was scared and began to cry ,
One day , they find a shiny coin under the bed

INTERVAL 0.003 - 0.151
CONTAINS 40.788%

the mess and the blood . She was very angry and
she wouldn 't get hurt . When they got
an important report for her boss . She did not like
still sing and be happy ." Lily smiled and hugged her
But the tree was too tall and Twe ety couldn 't

INTERVAL -0.146 - 0.003
CONTAINS 49.996%

not want to share her bear . She thought he was
pretend to eat their stew and say , " M mm
's bed . Finally , they found the wand behind the
asked many questions . The man patiently answered them all and
. From that day on , Lily and Tim my learned

INTERVAL -0.294 - -0.146
CONTAINS 4.844%

So they both walked up to the old lady , and
one day , Jane went to the park to practice with
dog . He goes back to his ball and plays some
!" Jim pointed to the sky and said ,
alone . From that day on , the animals had a
Token not
Token ID407
Feature activation+8.96e-02

Pos logit contributions
 idea+0.148
 adorable+0.143
 prized+0.143
 pleased+0.136
 best+0.135
Neg logit contributions
oes-0.205
rost-0.199
isk-0.199
ted-0.197
stack-0.196
Token a
Token ID257
Feature activation+1.78e-01

Pos logit contributions
 prized+0.127
 thrilled+0.116
 adorable+0.115
 precious+0.115
ement+0.114
Neg logit contributions
oes-0.166
isk-0.161
rost-0.158
talk-0.156
 south-0.154
Token monster
Token ID9234
Feature activation+1.77e-02

Pos logit contributions
 thrilled+0.310
 adorable+0.306
 pleased+0.303
 delighted+0.297
 impressed+0.293
Neg logit contributions
oes-0.266
use-0.261
rost-0.257
ricane-0.248
talk-0.245
Token,
Token ID11
Feature activation-8.22e-03

Pos logit contributions
 adorable+0.024
 best+0.023
 important+0.023
 idea+0.022
 precious+0.022
Neg logit contributions
oes-0.039
talk-0.039
ted-0.038
stack-0.036
rost-0.036
Token my
Token ID616
Feature activation+1.25e-01

Pos logit contributions
oes+0.012
talk+0.012
ted+0.012
stack+0.011
rost+0.011
Neg logit contributions
 adorable-0.017
 best-0.016
 important-0.016
 idea-0.016
 precious-0.016
Token de
Token ID390
Feature activation+2.58e-01

Pos logit contributions
 adorable+0.180
 important+0.175
 pleased+0.171
 thrilled+0.170
 impressed+0.168
Neg logit contributions
oes-0.241
ted-0.230
rost-0.222
talk-0.220
use-0.216
Tokenars
Token ID945
Feature activation-1.60e-02

Pos logit contributions
 adorable+0.552
 best+0.534
 idea+0.530
 important+0.530
 precious+0.514
Neg logit contributions
oes-0.387
talk-0.386
ted-0.367
rost-0.344
stack-0.343
Token.
Token ID13
Feature activation+1.96e-02

Pos logit contributions
oes+0.035
talk+0.035
ted+0.034
stack+0.033
rost+0.032
Neg logit contributions
 adorable-0.022
 important-0.021
 best-0.021
 precious-0.020
 idea-0.020
Token That
Token ID1320
Feature activation+4.61e-02

Pos logit contributions
 adorable+0.035
 best+0.034
 important+0.033
 idea+0.033
 precious+0.032
Neg logit contributions
talk-0.036
oes-0.036
ted-0.034
stack-0.033
isk-0.032
Token is
Token ID318
Feature activation+1.15e-01

Pos logit contributions
 pleased+0.094
 adorable+0.090
 impressed+0.090
 thrilled+0.089
 delighted+0.086
Neg logit contributions
oes-0.070
ted-0.064
talk-0.064
stack-0.060
use-0.059
Token a
Token ID257
Feature activation+1.97e-01

Pos logit contributions
 adorable+0.159
 idea+0.153
 best+0.153
 important+0.150
 precious+0.144
Neg logit contributions
oes-0.253
talk-0.249
ted-0.244
stack-0.234
rost-0.234
Token was
Token ID373
Feature activation+1.13e-01

Pos logit contributions
oes+0.084
ted+0.081
talk+0.080
isk+0.073
ting+0.072
Neg logit contributions
 adorable-0.107
 pleased-0.105
 prized-0.104
 precious-0.104
 idea-0.103
Token a
Token ID257
Feature activation+1.88e-01

Pos logit contributions
 idea+0.149
 prized+0.146
 adorable+0.145
 thrilled+0.142
 pleased+0.141
Neg logit contributions
rost-0.216
oes-0.206
 south-0.204
isk-0.203
 Enter-0.196
Token dream
Token ID4320
Feature activation+1.05e-01

Pos logit contributions
 adorable+0.198
 idea+0.186
 impressed+0.183
 thrilled+0.182
 prized+0.177
Neg logit contributions
rost-0.407
oes-0.395
stack-0.394
talk-0.387
use-0.383
Token,
Token ID11
Feature activation-6.41e-03

Pos logit contributions
 adorable+0.159
 best+0.152
 important+0.150
 idea+0.149
 precious+0.145
Neg logit contributions
oes-0.223
talk-0.220
ted-0.212
stack-0.204
rost-0.203
Token my
Token ID616
Feature activation+1.24e-01

Pos logit contributions
oes+0.009
rost+0.008
ted+0.008
stack+0.008
isk+0.008
Neg logit contributions
 adorable-0.013
 idea-0.013
 best-0.013
 important-0.013
 prized-0.012
Token de
Token ID390
Feature activation+2.57e-01

Pos logit contributions
 impressed+0.153
 pleased+0.145
 thrilled+0.138
 delighted+0.133
 adorable+0.131
Neg logit contributions
rost-0.230
ted-0.216
oes-0.213
 Rex-0.212
isk-0.203
Tokenars
Token ID945
Feature activation-2.52e-02

Pos logit contributions
 adorable+0.537
 best+0.524
 important+0.519
 idea+0.518
 precious+0.504
Neg logit contributions
oes-0.395
talk-0.390
ted-0.375
stack-0.354
rost-0.349
Token,"
Token ID553
Feature activation-3.29e-03

Pos logit contributions
oes+0.051
talk+0.051
ted+0.049
stack+0.047
rost+0.046
Neg logit contributions
 adorable-0.040
 best-0.038
 important-0.038
 idea-0.037
 precious-0.037
Token she
Token ID673
Feature activation-2.72e-02

Pos logit contributions
oes+0.005
talk+0.005
ted+0.005
stack+0.004
rost+0.004
Neg logit contributions
 adorable-0.007
 best-0.007
 important-0.007
 idea-0.007
 precious-0.007
Token said
Token ID531
Feature activation+1.09e-01

Pos logit contributions
talk+0.042
ted+0.041
oes+0.036
isk+0.035
Dog+0.035
Neg logit contributions
 best-0.054
 adorable-0.053
 important-0.052
 prized-0.051
 precious-0.051
Token.
Token ID13
Feature activation+1.76e-02

Pos logit contributions
 adorable+0.180
 best+0.174
 important+0.173
 idea+0.172
 precious+0.167
Neg logit contributions
oes-0.213
talk-0.212
ted-0.205
stack-0.195
rost-0.193
Token He
Token ID679
Feature activation-4.55e-02

Pos logit contributions
talk+0.004
oes+0.004
ted+0.004
use+0.004
ting+0.004
Neg logit contributions
 adorable-0.005
 idea-0.005
 important-0.004
 best-0.004
 beautiful-0.004
Token was
Token ID373
Feature activation+1.34e-01

Pos logit contributions
ted+0.077
talk+0.075
oes+0.074
rost+0.070
isk+0.070
Neg logit contributions
 best-0.081
 adorable-0.079
 idea-0.079
 precious-0.078
 important-0.077
Token a
Token ID257
Feature activation+2.07e-01

Pos logit contributions
 idea+0.174
 precious+0.171
 important+0.163
 adorable+0.161
 valuable+0.159
Neg logit contributions
oes-0.279
 gra-0.263
ted-0.263
isk-0.261
talk-0.255
Token mis
Token ID2984
Feature activation+1.52e-01

Pos logit contributions
 idea+0.226
 mistakes+0.208
 best+0.203
 important+0.197
 adorable+0.191
Neg logit contributions
oes-0.446
rost-0.444
Dog-0.442
 gra-0.437
stack-0.437
Tokenchie
Token ID3043
Feature activation+1.69e-01

Pos logit contributions
 best+0.336
 important+0.325
 idea+0.324
 adorable+0.321
 precious+0.312
Neg logit contributions
talk-0.240
oes-0.228
isk-0.199
Dog-0.195
ting-0.191
Tokenvous
Token ID31222
Feature activation+2.56e-01

Pos logit contributions
 best+0.281
 idea+0.271
 adorable+0.269
 precious+0.269
 important+0.258
Neg logit contributions
oes-0.328
talk-0.324
ted-0.299
Dog-0.285
 gra-0.283
Token little
Token ID1310
Feature activation+1.48e-01

Pos logit contributions
 important+0.404
 precious+0.397
 best+0.394
 adorable+0.390
 impressed+0.389
Neg logit contributions
oes-0.446
ted-0.441
rost-0.429
ting-0.427
talk-0.425
Token baby
Token ID5156
Feature activation+9.21e-02

Pos logit contributions
 important+0.232
 pleased+0.226
 impressed+0.225
 prized+0.224
 idea+0.223
Neg logit contributions
oes-0.275
ted-0.254
ting-0.253
stack-0.244
talk-0.241
Token and
Token ID290
Feature activation-2.64e-02

Pos logit contributions
 idea+0.167
 adorable+0.167
 important+0.164
 best+0.163
 precious+0.160
Neg logit contributions
oes-0.158
ted-0.156
talk-0.154
stack-0.148
ting-0.147
Token always
Token ID1464
Feature activation+8.50e-03

Pos logit contributions
oes+0.046
ted+0.045
talk+0.044
stack+0.044
rost+0.043
Neg logit contributions
 best-0.044
 adorable-0.044
 idea-0.044
 precious-0.043
 important-0.042
Token got
Token ID1392
Feature activation+3.82e-02

Pos logit contributions
 adorable+0.015
 best+0.014
 idea+0.014
 important+0.014
 precious+0.014
Neg logit contributions
oes-0.016
talk-0.015
ted-0.015
stack-0.014
isk-0.014
Token happy
Token ID3772
Feature activation+1.93e-01

Pos logit contributions
 best+0.162
 adorable+0.162
 idea+0.161
 prized+0.151
 precious+0.142
Neg logit contributions
oes-0.435
rost-0.416
 gra-0.412
talk-0.411
 south-0.403
Token,
Token ID11
Feature activation+2.02e-02

Pos logit contributions
 adorable+0.335
 best+0.324
 important+0.321
 idea+0.318
 precious+0.314
Neg logit contributions
oes-0.359
talk-0.352
ted-0.345
stack-0.330
rost-0.321
Token it
Token ID340
Feature activation-2.01e-02

Pos logit contributions
 adorable+0.035
 idea+0.035
 best+0.035
 important+0.035
 precious+0.034
Neg logit contributions
oes-0.034
ted-0.033
talk-0.033
stack-0.032
rost-0.031
Token was
Token ID373
Feature activation+1.41e-01

Pos logit contributions
ted+0.032
talk+0.032
oes+0.031
stack+0.030
ting+0.029
Neg logit contributions
 adorable-0.038
 best-0.037
 idea-0.037
 important-0.037
 precious-0.036
Token a
Token ID257
Feature activation+2.20e-01

Pos logit contributions
 adorable+0.223
 idea+0.222
 precious+0.219
 prized+0.214
 best+0.208
Neg logit contributions
oes-0.241
stack-0.230
ted-0.227
 gra-0.224
 south-0.224
Token brand
Token ID4508
Feature activation+2.55e-01

Pos logit contributions
 idea+0.225
 adorable+0.224
 mistakes+0.207
 best+0.203
 important+0.200
Neg logit contributions
stack-0.464
 gra-0.453
rost-0.447
oes-0.445
 south-0.442
Token new
Token ID649
Feature activation+2.07e-01

Pos logit contributions
 adorable+0.263
 idea+0.252
 best+0.249
 thrilled+0.243
 loyal+0.226
Neg logit contributions
oes-0.622
talk-0.584
ted-0.583
iffs-0.565
rost-0.562
Token ice
Token ID4771
Feature activation+9.06e-02

Pos logit contributions
 adorable+0.359
 impressed+0.345
 pleased+0.344
 important+0.338
 precious+0.335
Neg logit contributions
oes-0.358
stack-0.344
talk-0.331
ted-0.317
 south-0.300
Token cream
Token ID8566
Feature activation+6.06e-02

Pos logit contributions
 adorable+0.159
 idea+0.154
 important+0.151
 best+0.149
 precious+0.148
Neg logit contributions
oes-0.169
talk-0.157
ted-0.153
stack-0.151
rost-0.149
Token machine
Token ID4572
Feature activation-2.80e-03

Pos logit contributions
 adorable+0.101
 important+0.097
 idea+0.097
 best+0.096
 precious+0.094
Neg logit contributions
oes-0.118
talk-0.115
ted-0.113
stack-0.110
rost-0.106
Token.
Token ID13
Feature activation+2.27e-02

Pos logit contributions
oes+0.006
talk+0.005
ted+0.005
stack+0.005
rost+0.005
Neg logit contributions
 adorable-0.005
 best-0.004
 important-0.004
 idea-0.004
 precious-0.004
Token
Token ID198
Feature activation-2.37e-02

Pos logit contributions
oes+0.045
talk+0.045
ted+0.043
stack+0.041
rost+0.040
Neg logit contributions
 adorable-0.054
 best-0.053
 important-0.052
 idea-0.052
 precious-0.051
Token"
Token ID1
Feature activation-2.32e-02

Pos logit contributions
oes+0.050
talk+0.050
ted+0.048
stack+0.046
rost+0.046
Neg logit contributions
 adorable-0.036
 best-0.035
 idea-0.035
 important-0.034
 precious-0.033
TokenThat
Token ID2504
Feature activation+6.32e-02

Pos logit contributions
oes+0.039
isk+0.036
talk+0.036
stack+0.036
ted+0.035
Neg logit contributions
 adorable-0.046
 best-0.044
 precious-0.044
 important-0.044
 beautiful-0.043
Token,
Token ID11
Feature activation-1.84e-02

Pos logit contributions
 pleased+0.126
 impressed+0.123
 importantly+0.121
 adorable+0.121
 precious+0.120
Neg logit contributions
oes-0.099
talk-0.095
ted-0.094
stack-0.087
 gra-0.086
Token my
Token ID616
Feature activation+1.16e-01

Pos logit contributions
ted+0.034
oes+0.033
talk+0.031
ting+0.031
rost+0.031
Neg logit contributions
 important-0.030
 adorable-0.029
 idea-0.029
 prized-0.029
 best-0.028
Token de
Token ID390
Feature activation+2.53e-01

Pos logit contributions
 impressed+0.147
 important+0.145
 pleased+0.142
 adorable+0.139
 prized+0.137
Neg logit contributions
ted-0.230
oes-0.221
ricanes-0.214
rost-0.212
talk-0.210
Tokenars
Token ID945
Feature activation-3.29e-02

Pos logit contributions
 adorable+0.541
 best+0.527
 important+0.522
 idea+0.521
 precious+0.507
Neg logit contributions
oes-0.377
talk-0.373
ted-0.357
stack-0.335
rost-0.332
Token,
Token ID11
Feature activation-8.23e-03

Pos logit contributions
oes+0.070
talk+0.070
ted+0.068
stack+0.065
rost+0.064
Neg logit contributions
 adorable-0.048
 best-0.046
 important-0.046
 idea-0.045
 precious-0.044
Token is
Token ID318
Feature activation+1.17e-01

Pos logit contributions
oes+0.015
talk+0.015
ted+0.014
stack+0.013
rost+0.013
Neg logit contributions
 adorable-0.015
 best-0.015
 important-0.015
 idea-0.014
 precious-0.014
Token a
Token ID257
Feature activation+2.00e-01

Pos logit contributions
 adorable+0.222
 best+0.219
 important+0.215
 precious+0.206
 beautiful+0.204
Neg logit contributions
talk-0.221
oes-0.219
ted-0.204
stack-0.194
isk-0.191
Token cave
Token ID11527
Feature activation-4.15e-02

Pos logit contributions
 adorable+0.315
 idea+0.304
 important+0.300
 best+0.298
 precious+0.285
Neg logit contributions
oes-0.368
talk-0.361
stack-0.357
rost-0.351
ted-0.349
Token
Token ID198
Feature activation-1.90e-02

Pos logit contributions
oes+0.039
ted+0.038
talk+0.038
ting+0.036
stack+0.035
Neg logit contributions
 adorable-0.035
 important-0.034
 idea-0.033
 best-0.033
 beautiful-0.033
Token"
Token ID1
Feature activation-2.26e-02

Pos logit contributions
oes+0.035
ted+0.034
talk+0.033
stack+0.032
ting+0.031
Neg logit contributions
 adorable-0.034
 important-0.034
 beautiful-0.032
 best-0.032
 loyal-0.032
TokenHello
Token ID15496
Feature activation+4.81e-02

Pos logit contributions
oes+0.033
talk+0.030
ted+0.030
stack+0.030
isk+0.028
Neg logit contributions
 adorable-0.050
 best-0.048
 important-0.047
 beautiful-0.047
 loyal-0.046
Token,
Token ID11
Feature activation-3.71e-02

Pos logit contributions
 adorable+0.069
 important+0.069
 idea+0.069
 best+0.066
 precious+0.064
Neg logit contributions
oes-0.100
ted-0.098
talk-0.096
rost-0.092
ting-0.091
Token my
Token ID616
Feature activation+1.15e-01

Pos logit contributions
oes+0.074
talk+0.072
ted+0.072
stack+0.069
rost+0.069
Neg logit contributions
 adorable-0.056
 idea-0.055
 best-0.055
 important-0.055
 precious-0.052
Token de
Token ID390
Feature activation+2.51e-01

Pos logit contributions
 impressed+0.144
 adorable+0.143
 pleased+0.143
 idea+0.140
 important+0.138
Neg logit contributions
oes-0.233
ted-0.232
talk-0.229
isk-0.228
rost-0.225
Tokenars
Token ID945
Feature activation-3.64e-02

Pos logit contributions
 adorable+0.546
 best+0.532
 idea+0.528
 important+0.526
 precious+0.510
Neg logit contributions
oes-0.367
talk-0.364
ted-0.343
stack-0.324
rost-0.324
Token!"
Token ID2474
Feature activation-2.53e-02

Pos logit contributions
oes+0.080
talk+0.079
ted+0.076
stack+0.075
use+0.073
Neg logit contributions
 adorable-0.050
 important-0.049
 best-0.048
 precious-0.047
 idea-0.046
Token Grand
Token ID5675
Feature activation+1.12e-01

Pos logit contributions
oes+0.046
talk+0.044
ted+0.043
isk+0.043
stack+0.042
Neg logit contributions
 adorable-0.047
 best-0.045
 idea-0.045
 important-0.044
 beautiful-0.043
Tokenma
Token ID2611
Feature activation+2.69e-02

Pos logit contributions
 adorable+0.178
 best+0.172
 important+0.170
 idea+0.169
 precious+0.164
Neg logit contributions
oes-0.228
talk-0.226
ted-0.219
stack-0.210
rost-0.208
Token says
Token ID1139
Feature activation+1.11e-01

Pos logit contributions
 adorable+0.039
 idea+0.038
 best+0.038
 important+0.036
 precious+0.035
Neg logit contributions
oes-0.057
talk-0.056
ted-0.055
stack-0.054
rost-0.054
Token a
Token ID257
Feature activation+2.14e-01

Pos logit contributions
 adorable+0.150
 important+0.144
 idea+0.141
 best+0.140
 precious+0.139
Neg logit contributions
oes-0.222
talk-0.216
ted-0.213
stack-0.211
rost-0.205
Token special
Token ID2041
Feature activation+1.90e-01

Pos logit contributions
 adorable+0.256
 idea+0.242
 best+0.239
 important+0.237
 precious+0.227
Neg logit contributions
stack-0.446
oes-0.444
 gra-0.441
talk-0.432
use-0.432
Token gift
Token ID6979
Feature activation+3.96e-02

Pos logit contributions
 adorable+0.314
 best+0.301
 important+0.295
 idea+0.289
 precious+0.287
Neg logit contributions
oes-0.362
talk-0.356
ted-0.344
stack-0.340
rost-0.330
Token -
Token ID532
Feature activation+2.81e-02

Pos logit contributions
 adorable+0.070
 best+0.068
 important+0.067
 idea+0.067
 precious+0.065
Neg logit contributions
oes-0.072
talk-0.071
ted-0.069
stack-0.067
rost-0.065
Token a
Token ID257
Feature activation+2.08e-01

Pos logit contributions
 adorable+0.046
 idea+0.046
 best+0.045
 important+0.044
 precious+0.043
Neg logit contributions
oes-0.051
ted-0.050
talk-0.049
stack-0.048
 gra-0.046
Token brand
Token ID4508
Feature activation+2.51e-01

Pos logit contributions
 adorable+0.202
 idea+0.197
 best+0.184
 impressed+0.183
 thrilled+0.176
Neg logit contributions
stack-0.453
 gra-0.443
use-0.427
oes-0.426
 south-0.423
Token new
Token ID649
Feature activation+1.98e-01

Pos logit contributions
 adorable+0.266
 best+0.254
 thrilled+0.250
 idea+0.250
 loyal+0.236
Neg logit contributions
oes-0.604
talk-0.574
ted-0.564
ricane-0.544
rost-0.543
Token bath
Token ID7837
Feature activation-7.04e-04

Pos logit contributions
 impressed+0.312
 adorable+0.306
 pleased+0.303
 thrilled+0.294
 intelligent+0.284
Neg logit contributions
stack-0.356
oes-0.345
talk-0.331
ted-0.325
 gra-0.312
Tokenrobe
Token ID25481
Feature activation+6.72e-02

Pos logit contributions
oes+0.001
ted+0.001
talk+0.001
stack+0.001
isk+0.001
Neg logit contributions
 adorable-0.001
 important-0.001
 precious-0.001
 idea-0.001
 best-0.001
Token.
Token ID13
Feature activation+2.63e-02

Pos logit contributions
 adorable+0.116
 idea+0.114
 important+0.111
 best+0.111
 precious+0.109
Neg logit contributions
ted-0.121
oes-0.120
talk-0.119
stack-0.115
ting-0.110
Token It
Token ID632
Feature activation-2.75e-02

Pos logit contributions
 adorable+0.053
 best+0.051
 important+0.051
 idea+0.051
 precious+0.049
Neg logit contributions
oes-0.042
talk-0.042
ted-0.040
stack-0.037
rost-0.037
Token.
Token ID13
Feature activation+2.06e-02

Pos logit contributions
 important+0.297
 adorable+0.292
 idea+0.290
 precious+0.288
 best+0.288
Neg logit contributions
ted-0.233
oes-0.231
talk-0.224
ting-0.215
 south-0.214
Token "
Token ID366
Feature activation-2.86e-02

Pos logit contributions
 adorable+0.033
 best+0.030
 beautiful+0.030
 important+0.030
 idea+0.029
Neg logit contributions
oes-0.044
talk-0.044
ted-0.041
ales-0.041
stack-0.040
TokenHello
Token ID15496
Feature activation+4.92e-02

Pos logit contributions
oes+0.045
stack+0.045
ted+0.044
isk+0.043
talk+0.043
Neg logit contributions
 adorable-0.058
 beautiful-0.053
 loyal-0.053
 precious-0.052
 important-0.052
Token,
Token ID11
Feature activation-3.56e-02

Pos logit contributions
 adorable+0.070
 important+0.070
 idea+0.069
 best+0.067
 precious+0.066
Neg logit contributions
oes-0.104
ted-0.101
talk-0.100
rost-0.095
use-0.095
Token my
Token ID616
Feature activation+1.19e-01

Pos logit contributions
oes+0.079
talk+0.077
ted+0.076
rost+0.073
stack+0.073
Neg logit contributions
 adorable-0.048
 idea-0.047
 best-0.047
 important-0.046
 precious-0.044
Token de
Token ID390
Feature activation+2.51e-01

Pos logit contributions
 impressed+0.137
 pleased+0.125
 thrilled+0.117
 important+0.112
 idea+0.112
Neg logit contributions
ted-0.246
oes-0.246
ricanes-0.243
isk-0.243
rost-0.241
Tokenars
Token ID945
Feature activation-3.38e-02

Pos logit contributions
 adorable+0.539
 best+0.526
 idea+0.521
 important+0.520
 precious+0.505
Neg logit contributions
oes-0.371
talk-0.368
ted-0.350
stack-0.329
rost-0.326
Token!
Token ID0
Feature activation+4.29e-03

Pos logit contributions
oes+0.073
talk+0.072
ted+0.070
stack+0.068
rost+0.067
Neg logit contributions
 adorable-0.048
 important-0.046
 best-0.046
 idea-0.045
 precious-0.045
Token What
Token ID1867
Feature activation-1.77e-02

Pos logit contributions
 adorable+0.008
 best+0.007
 loyal+0.007
 important+0.007
 precious+0.007
Neg logit contributions
talk-0.008
oes-0.008
ales-0.007
ted-0.007
isk-0.007
Token a
Token ID257
Feature activation+1.77e-01

Pos logit contributions
oes+0.020
talk+0.018
ted+0.017
 gra+0.017
rost+0.016
Neg logit contributions
 adorable-0.040
 pleased-0.039
 idea-0.039
 precious-0.039
 important-0.039
Token nice
Token ID3621
Feature activation+1.88e-01

Pos logit contributions
 adorable+0.177
 idea+0.173
 impressed+0.169
 pleased+0.165
 best+0.165
Neg logit contributions
talk-0.399
 gra-0.395
Dog-0.389
oes-0.389
use-0.386
Token
Token ID8151
Feature activation-7.49e-03

Pos logit contributions
 adorable+0.218
 important+0.212
 idea+0.211
 best+0.211
 precious+0.205
Neg logit contributions
oes-0.190
talk-0.189
ted-0.182
isk-0.172
stack-0.172
Tokenre
Token ID260
Feature activation+9.05e-02

Pos logit contributions
oes+0.013
talk+0.013
ted+0.013
stack+0.012
isk+0.012
Neg logit contributions
 adorable-0.014
 idea-0.013
 important-0.013
 best-0.013
 precious-0.013
Token welcome
Token ID7062
Feature activation+1.13e-01

Pos logit contributions
 adorable+0.130
 idea+0.127
 precious+0.125
 prized+0.124
 important+0.121
Neg logit contributions
oes-0.169
isk-0.166
stack-0.164
talk-0.164
 gra-0.164
Token,
Token ID11
Feature activation-1.27e-02

Pos logit contributions
 adorable+0.155
 idea+0.149
 important+0.149
 best+0.147
 precious+0.145
Neg logit contributions
oes-0.246
talk-0.246
ted-0.243
stack-0.230
ting-0.227
Token my
Token ID616
Feature activation+1.21e-01

Pos logit contributions
oes+0.022
ted+0.022
talk+0.022
rost+0.020
stack+0.020
Neg logit contributions
 adorable-0.022
 idea-0.022
 best-0.021
 important-0.021
 precious-0.021
Token de
Token ID390
Feature activation+2.50e-01

Pos logit contributions
 adorable+0.115
 impressed+0.115
 idea+0.114
 important+0.113
 best+0.108
Neg logit contributions
ted-0.265
rost-0.261
oes-0.261
 gra-0.254
talk-0.254
Tokenars
Token ID945
Feature activation-4.51e-02

Pos logit contributions
 adorable+0.559
 idea+0.542
 best+0.541
 important+0.536
 precious+0.523
Neg logit contributions
talk-0.351
oes-0.344
ted-0.330
rost-0.305
use-0.304
Token.
Token ID13
Feature activation-2.88e-04

Pos logit contributions
oes+0.100
talk+0.099
ted+0.097
stack+0.093
rost+0.092
Neg logit contributions
 adorable-0.062
 best-0.060
 important-0.059
 idea-0.059
 precious-0.057
Token Enjoy
Token ID18179
Feature activation-1.44e-02

Pos logit contributions
oes+0.000
talk+0.000
ted+0.000
stack+0.000
rost+0.000
Neg logit contributions
 adorable-0.001
 best-0.001
 important-0.001
 idea-0.001
 precious-0.001
Token your
Token ID534
Feature activation+1.17e-02

Pos logit contributions
oes+0.029
talk+0.029
ted+0.028
stack+0.027
rost+0.026
Neg logit contributions
 adorable-0.023
 best-0.022
 important-0.022
 idea-0.022
 precious-0.021
Token cake
Token ID12187
Feature activation-8.09e-02

Pos logit contributions
 adorable+0.018
 best+0.017
 important+0.017
 impressed+0.017
 precious+0.017
Neg logit contributions
oes-0.021
ted-0.020
 gra-0.020
talk-0.020
use-0.019
TokenThat
Token ID2504
Feature activation+6.38e-02

Pos logit contributions
oes+0.028
talk+0.028
ted+0.027
stack+0.025
rost+0.025
Neg logit contributions
 adorable-0.038
 idea-0.038
 best-0.038
 important-0.037
 precious-0.036
Token's
Token ID338
Feature activation+6.59e-02

Pos logit contributions
 pleased+0.133
 impressed+0.128
 importantly+0.127
 prized+0.124
 precious+0.124
Neg logit contributions
oes-0.099
talk-0.091
ted-0.090
stack-0.082
 gra-0.082
Token good
Token ID922
Feature activation+1.54e-01

Pos logit contributions
 prized+0.077
 adorable+0.076
 idea+0.075
 pleased+0.074
 thrilled+0.073
Neg logit contributions
oes-0.141
rost-0.139
isk-0.139
 south-0.134
talk-0.132
Token,
Token ID11
Feature activation-1.59e-02

Pos logit contributions
 prized+0.290
 pleased+0.289
 impressed+0.288
 adorable+0.286
 thrilled+0.283
Neg logit contributions
oes-0.269
talk-0.240
ted-0.239
rost-0.215
use-0.214
Token my
Token ID616
Feature activation+1.14e-01

Pos logit contributions
oes+0.018
rost+0.017
 Rex+0.016
stack+0.016
ted+0.015
Neg logit contributions
 prized-0.036
 important-0.034
 valuable-0.034
 adorable-0.034
 precious-0.033
Token de
Token ID390
Feature activation+2.50e-01

Pos logit contributions
 prized+0.167
 adorable+0.162
 impressed+0.162
 thrilled+0.160
 pleased+0.157
Neg logit contributions
oes-0.209
rost-0.196
ricanes-0.195
ted-0.187
isk-0.181
Tokenars
Token ID945
Feature activation-3.23e-02

Pos logit contributions
 adorable+0.537
 best+0.522
 idea+0.518
 important+0.517
 precious+0.503
Neg logit contributions
oes-0.369
talk-0.366
ted-0.349
stack-0.327
rost-0.326
Token.
Token ID13
Feature activation+1.55e-02

Pos logit contributions
oes+0.059
talk+0.058
ted+0.055
stack+0.053
rost+0.053
Neg logit contributions
 adorable-0.057
 important-0.055
 best-0.054
 precious-0.054
 idea-0.054
Token But
Token ID887
Feature activation-5.66e-03

Pos logit contributions
 adorable+0.026
 best+0.025
 important+0.024
 idea+0.024
 precious+0.023
Neg logit contributions
talk-0.030
oes-0.030
ted-0.029
isk-0.028
stack-0.027
Token you
Token ID345
Feature activation+2.93e-02

Pos logit contributions
talk+0.011
oes+0.011
ted+0.010
stack+0.010
isk+0.010
Neg logit contributions
 adorable-0.010
 best-0.010
 important-0.009
 idea-0.009
 precious-0.009
Token still
Token ID991
Feature activation+5.04e-02

Pos logit contributions
 adorable+0.055
 best+0.054
 important+0.053
 idea+0.053
 precious+0.051
Neg logit contributions
oes-0.051
talk-0.050
ted-0.048
rost-0.046
stack-0.046
Token.
Token ID13
Feature activation+1.77e-02

Pos logit contributions
 adorable+0.103
 best+0.097
 important+0.097
 idea+0.095
 precious+0.094
Neg logit contributions
oes-0.137
talk-0.134
ted-0.130
stack-0.127
use-0.123
Token She
Token ID1375
Feature activation-2.72e-02

Pos logit contributions
 adorable+0.035
 important+0.033
 best+0.033
 idea+0.033
 precious+0.032
Neg logit contributions
oes-0.029
talk-0.029
ted-0.029
rost-0.027
stack-0.027
Token had
Token ID550
Feature activation+2.17e-02

Pos logit contributions
talk+0.044
ted+0.042
oes+0.042
stack+0.039
isk+0.039
Neg logit contributions
 adorable-0.055
 best-0.054
 important-0.052
 idea-0.052
 precious-0.051
Token designed
Token ID3562
Feature activation-7.71e-04

Pos logit contributions
 adorable+0.039
 best+0.038
 idea+0.037
 important+0.037
 precious+0.036
Neg logit contributions
oes-0.039
talk-0.039
ted-0.038
stack-0.036
rost-0.035
Token a
Token ID257
Feature activation+1.52e-01

Pos logit contributions
talk+0.002
ted+0.002
oes+0.002
stack+0.001
isk+0.001
Neg logit contributions
 adorable-0.001
 idea-0.001
 best-0.001
 important-0.001
 precious-0.001
Token de
Token ID390
Feature activation+2.49e-01

Pos logit contributions
 adorable+0.170
 idea+0.153
 best+0.151
 important+0.149
 precious+0.140
Neg logit contributions
oes-0.356
talk-0.348
stack-0.344
 gra-0.335
 south-0.331
Tokenvious
Token ID1442
Feature activation+2.15e-01

Pos logit contributions
 adorable+0.456
 best+0.443
 idea+0.438
 important+0.437
 precious+0.423
Neg logit contributions
oes-0.448
talk-0.446
ted-0.430
stack-0.408
rost-0.404
Token test
Token ID1332
Feature activation-9.20e-03

Pos logit contributions
 adorable+0.332
 precious+0.308
 important+0.307
 best+0.306
 impressed+0.298
Neg logit contributions
oes-0.416
talk-0.406
ted-0.390
stack-0.386
rost-0.373
Token!
Token ID0
Feature activation+1.53e-02

Pos logit contributions
oes+0.017
talk+0.017
ted+0.017
stack+0.016
rost+0.016
Neg logit contributions
 adorable-0.016
 best-0.015
 important-0.015
 idea-0.015
 precious-0.015
Token The
Token ID383
Feature activation+5.41e-02

Pos logit contributions
 adorable+0.030
 best+0.029
 important+0.029
 idea+0.029
 precious+0.028
Neg logit contributions
oes-0.025
talk-0.025
ted-0.024
stack-0.023
rost-0.023
Token fairy
Token ID25607
Feature activation+4.17e-02

Pos logit contributions
 adorable+0.077
 important+0.071
 best+0.070
 idea+0.070
 impressed+0.069
Neg logit contributions
oes-0.110
talk-0.110
ted-0.106
stack-0.105
use-0.101
Token bright
Token ID6016
Feature activation+1.64e-01

Pos logit contributions
 important+0.221
 adorable+0.221
 idea+0.218
 best+0.213
 intelligent+0.204
Neg logit contributions
stack-0.321
 gra-0.317
 south-0.313
oes-0.310
talk-0.309
Token yellow
Token ID7872
Feature activation+1.64e-01

Pos logit contributions
 best+0.266
 important+0.261
 precious+0.249
 loyal+0.245
 intelligent+0.244
Neg logit contributions
ted-0.283
oes-0.275
stack-0.265
 south-0.264
talk-0.263
Token colour
Token ID9568
Feature activation+9.28e-02

Pos logit contributions
 important+0.299
 adorable+0.298
 best+0.293
 precious+0.284
 impressed+0.284
Neg logit contributions
oes-0.283
ted-0.283
talk-0.276
stack-0.266
ting-0.246
Token and
Token ID290
Feature activation-3.08e-02

Pos logit contributions
 adorable+0.157
 important+0.152
 best+0.152
 idea+0.150
 precious+0.147
Neg logit contributions
oes-0.177
talk-0.171
ted-0.170
stack-0.160
ting-0.157
Token a
Token ID257
Feature activation+1.85e-01

Pos logit contributions
ted+0.065
 benches+0.058
talk+0.058
oes+0.058
 gra+0.057
Neg logit contributions
 idea-0.041
 important-0.038
 best-0.038
 adorable-0.037
 prized-0.037
Token fr
Token ID1216
Feature activation+2.48e-01

Pos logit contributions
 important+0.235
 adorable+0.233
 idea+0.228
 best+0.222
 intelligent+0.212
Neg logit contributions
stack-0.361
 gra-0.360
oes-0.359
talk-0.355
 south-0.355
Tokenilly
Token ID6548
Feature activation+1.04e-01

Pos logit contributions
 best+0.393
 adorable+0.385
 idea+0.383
 important+0.375
 loyal+0.372
Neg logit contributions
oes-0.515
talk-0.501
ted-0.496
rost-0.459
ales-0.456
Token style
Token ID3918
Feature activation+1.15e-01

Pos logit contributions
 adorable+0.171
 important+0.169
 best+0.169
 idea+0.167
 precious+0.162
Neg logit contributions
oes-0.192
talk-0.190
ted-0.189
stack-0.180
rost-0.174
Token.
Token ID13
Feature activation+1.70e-02

Pos logit contributions
 adorable+0.216
 important+0.211
 best+0.209
 loyal+0.208
 precious+0.206
Neg logit contributions
oes-0.190
ted-0.183
talk-0.183
stack-0.167
ting-0.163
Token She
Token ID1375
Feature activation-3.02e-02

Pos logit contributions
 adorable+0.029
 best+0.028
 important+0.027
 idea+0.027
 beautiful+0.027
Neg logit contributions
oes-0.033
talk-0.032
rost-0.031
isk-0.030
ales-0.029
Token tried
Token ID3088
Feature activation+3.46e-02

Pos logit contributions
ted+0.043
talk+0.041
oes+0.038
 walked+0.036
 approached+0.035
Neg logit contributions
 idea-0.062
 best-0.061
 important-0.061
 adorable-0.061
 precious-0.059
Token novel
Token ID5337
Feature activation-8.23e-03

Pos logit contributions
 idea+0.267
 adorable+0.255
 important+0.255
 best+0.253
 precious+0.230
Neg logit contributions
stack-0.393
oes-0.382
 gra-0.381
ted-0.378
ricane-0.375
Token!
Token ID0
Feature activation+2.18e-02

Pos logit contributions
oes+0.018
ted+0.018
talk+0.017
stack+0.017
ting+0.017
Neg logit contributions
 adorable-0.011
 best-0.011
 important-0.011
 idea-0.011
 precious-0.010
Token It
Token ID632
Feature activation-3.27e-02

Pos logit contributions
 adorable+0.041
 best+0.039
 idea+0.039
 important+0.038
 beautiful+0.038
Neg logit contributions
talk-0.039
oes-0.039
stack-0.037
rost-0.036
ted-0.036
Token was
Token ID373
Feature activation+1.40e-01

Pos logit contributions
ted+0.053
talk+0.051
oes+0.051
stack+0.048
isk+0.047
Neg logit contributions
 best-0.061
 idea-0.061
 adorable-0.060
 important-0.059
 precious-0.058
Token a
Token ID257
Feature activation+2.13e-01

Pos logit contributions
 idea+0.189
 adorable+0.186
 best+0.185
 important+0.178
 precious+0.176
Neg logit contributions
oes-0.282
ted-0.282
stack-0.275
talk-0.271
rost-0.268
Token sweet
Token ID6029
Feature activation+2.47e-01

Pos logit contributions
 idea+0.240
 best+0.231
 adorable+0.228
 important+0.219
 mistakes+0.206
Neg logit contributions
stack-0.443
oes-0.437
rost-0.430
 gra-0.427
ricane-0.425
Token novel
Token ID5337
Feature activation+1.39e-03

Pos logit contributions
 best+0.426
 impressed+0.423
 important+0.408
vern+0.404
 precious+0.401
Neg logit contributions
ted-0.401
oes-0.394
rost-0.376
talk-0.370
ting-0.360
Token with
Token ID351
Feature activation+3.18e-02

Pos logit contributions
 adorable+0.002
 idea+0.002
 important+0.002
 best+0.002
 precious+0.002
Neg logit contributions
oes-0.003
stack-0.003
ted-0.003
talk-0.003
ting-0.003
Token a
Token ID257
Feature activation+1.69e-01

Pos logit contributions
 idea+0.045
 important+0.042
 adorable+0.042
 best+0.042
 precious+0.040
Neg logit contributions
ted-0.068
talk-0.066
oes-0.065
ting-0.062
stack-0.060
Token bright
Token ID6016
Feature activation+1.64e-01

Pos logit contributions
 best+0.226
 adorable+0.226
 idea+0.224
 important+0.220
 intelligent+0.202
Neg logit contributions
talk-0.317
oes-0.310
stack-0.310
ted-0.307
 gra-0.304
Token pink
Token ID11398
Feature activation+1.71e-01

Pos logit contributions
 best+0.264
 important+0.254
 precious+0.247
 adorable+0.240
 impressed+0.239
Neg logit contributions
ted-0.296
oes-0.289
talk-0.286
stack-0.269
 towards-0.265
Token amazing
Token ID4998
Feature activation+1.36e-01

Pos logit contributions
 adorable+0.099
 best+0.096
 idea+0.095
 important+0.092
 precious+0.089
Neg logit contributions
oes-0.179
ted-0.176
talk-0.175
stack-0.168
rost-0.165
Token.
Token ID13
Feature activation+1.21e-02

Pos logit contributions
 adorable+0.193
 best+0.186
 important+0.183
 idea+0.182
 precious+0.175
Neg logit contributions
oes-0.300
talk-0.297
ted-0.291
stack-0.279
rost-0.274
Token It
Token ID632
Feature activation-3.48e-02

Pos logit contributions
 adorable+0.022
 best+0.021
 idea+0.021
 important+0.021
 beautiful+0.020
Neg logit contributions
talk-0.021
oes-0.021
ted-0.020
isk-0.019
use-0.019
Token was
Token ID373
Feature activation+1.38e-01

Pos logit contributions
oes+0.061
ted+0.060
talk+0.059
stack+0.057
isk+0.055
Neg logit contributions
 adorable-0.061
 best-0.059
 idea-0.059
 important-0.059
 precious-0.057
Token a
Token ID257
Feature activation+2.15e-01

Pos logit contributions
 adorable+0.189
 idea+0.183
 best+0.181
 important+0.178
 precious+0.174
Neg logit contributions
oes-0.291
ted-0.281
stack-0.278
talk-0.277
rost-0.271
Token brand
Token ID4508
Feature activation+2.47e-01

Pos logit contributions
 adorable+0.272
 idea+0.265
 best+0.256
 important+0.254
 intelligent+0.236
Neg logit contributions
stack-0.447
oes-0.439
 gra-0.422
rost-0.416
 south-0.416
Token new
Token ID649
Feature activation+1.93e-01

Pos logit contributions
 adorable+0.261
 best+0.243
 thrilled+0.233
 idea+0.230
 loyal+0.223
Neg logit contributions
oes-0.605
talk-0.569
ted-0.557
rost-0.555
iffs-0.551
Token toy
Token ID13373
Feature activation+8.10e-02

Pos logit contributions
 adorable+0.333
 pleased+0.329
 impressed+0.328
 important+0.313
 thrilled+0.310
Neg logit contributions
stack-0.335
oes-0.327
talk-0.300
 hunt-0.286
 towards-0.283
Token.
Token ID13
Feature activation+1.79e-02

Pos logit contributions
 adorable+0.133
 important+0.128
 idea+0.127
 best+0.125
 precious+0.124
Neg logit contributions
oes-0.155
ted-0.152
stack-0.150
talk-0.149
ting-0.141
Token They
Token ID1119
Feature activation-4.26e-02

Pos logit contributions
 adorable+0.034
 best+0.033
 idea+0.033
 important+0.032
 precious+0.031
Neg logit contributions
talk-0.030
oes-0.030
ted-0.029
isk-0.027
rost-0.027
Token both
Token ID1111
Feature activation+3.23e-02

Pos logit contributions
oes+0.075
talk+0.074
ted+0.071
stack+0.067
rost+0.067
Neg logit contributions
 adorable-0.080
 best-0.078
 important-0.077
 idea-0.077
 precious-0.075
Token am
Token ID716
Feature activation+8.76e-02

Pos logit contributions
 adorable+0.028
 best+0.026
 important+0.026
 idea+0.026
 precious+0.025
Neg logit contributions
talk-0.027
oes-0.027
ted-0.026
rost-0.025
isk-0.024
Token not
Token ID407
Feature activation+9.52e-02

Pos logit contributions
 idea+0.109
 adorable+0.102
 prized+0.099
 precious+0.097
 best+0.095
Neg logit contributions
oes-0.171
stack-0.170
 gra-0.170
talk-0.170
 Enter-0.170
Token mad
Token ID8805
Feature activation+1.64e-01

Pos logit contributions
 adorable+0.112
 prized+0.110
 precious+0.107
 idea+0.106
 best+0.105
Neg logit contributions
oes-0.206
talk-0.205
use-0.193
rost-0.190
ting-0.190
Token,
Token ID11
Feature activation-2.07e-02

Pos logit contributions
 adorable+0.283
 precious+0.272
 idea+0.271
 important+0.271
 best+0.263
Neg logit contributions
oes-0.305
ted-0.290
talk-0.290
ting-0.281
rost-0.266
Token my
Token ID616
Feature activation+1.12e-01

Pos logit contributions
ted+0.029
oes+0.028
talk+0.027
ting+0.026
rost+0.026
Neg logit contributions
 adorable-0.043
 important-0.043
 idea-0.043
 best-0.042
 prized-0.042
Token de
Token ID390
Feature activation+2.47e-01

Pos logit contributions
 adorable+0.138
 prized+0.131
 important+0.129
 impressed+0.129
 best+0.129
Neg logit contributions
oes-0.242
talk-0.236
rost-0.233
ted-0.231
isk-0.219
Tokenars
Token ID945
Feature activation-3.57e-02

Pos logit contributions
 adorable+0.516
 best+0.504
 important+0.499
 idea+0.498
 precious+0.485
Neg logit contributions
oes-0.379
talk-0.374
ted-0.360
stack-0.340
rost-0.334
Token.
Token ID13
Feature activation+1.09e-02

Pos logit contributions
oes+0.077
talk+0.077
ted+0.074
stack+0.071
rost+0.071
Neg logit contributions
 adorable-0.052
 best-0.052
 idea-0.050
 important-0.050
 precious-0.047
Token I
Token ID314
Feature activation-1.62e-02

Pos logit contributions
 adorable+0.018
 best+0.017
 important+0.017
 precious+0.016
 loyal+0.016
Neg logit contributions
talk-0.022
oes-0.022
ted-0.021
isk-0.020
stack-0.020
Token am
Token ID716
Feature activation+9.32e-02

Pos logit contributions
talk+0.028
oes+0.028
ted+0.027
rost+0.025
stack+0.025
Neg logit contributions
 adorable-0.030
 best-0.030
 important-0.029
 idea-0.029
 precious-0.028
Token glad
Token ID9675
Feature activation+1.40e-01

Pos logit contributions
 idea+0.107
 adorable+0.101
 best+0.097
 important+0.092
 precious+0.091
Neg logit contributions
oes-0.208
talk-0.207
ted-0.200
 gra-0.200
isk-0.199
Token tigers
Token ID43114
Feature activation-3.14e-02

Pos logit contributions
oes+0.078
talk+0.077
ted+0.073
stack+0.070
rost+0.069
Neg logit contributions
 adorable-0.110
 best-0.106
 important-0.105
 idea-0.104
 precious-0.102
Token,
Token ID11
Feature activation-2.56e-02

Pos logit contributions
talk+0.077
oes+0.077
rost+0.073
ricanes+0.072
ted+0.072
Neg logit contributions
 adorable-0.037
 best-0.036
 idea-0.035
 important-0.033
 beautiful-0.033
Token but
Token ID475
Feature activation-3.55e-02

Pos logit contributions
oes+0.043
talk+0.042
ted+0.040
stack+0.038
rost+0.038
Neg logit contributions
 adorable-0.051
 best-0.050
 important-0.049
 idea-0.049
 precious-0.047
Token he
Token ID339
Feature activation-2.69e-02

Pos logit contributions
talk+0.062
oes+0.061
ted+0.057
iffs+0.055
isk+0.054
Neg logit contributions
 adorable-0.071
 best-0.070
 important-0.068
 idea-0.066
 loyal-0.065
Token was
Token ID373
Feature activation+1.47e-01

Pos logit contributions
oes+0.051
talk+0.051
ted+0.050
stack+0.048
isk+0.047
Neg logit contributions
 adorable-0.045
 best-0.044
 idea-0.044
 important-0.043
 precious-0.042
Token most
Token ID749
Feature activation+2.46e-01

Pos logit contributions
 idea+0.201
 adorable+0.197
 best+0.191
 important+0.187
 precious+0.185
Neg logit contributions
oes-0.299
talk-0.293
stack-0.288
ted-0.288
isk-0.285
Token excited
Token ID6568
Feature activation+2.08e-01

Pos logit contributions
 idea+0.237
 best+0.212
 darling+0.173
kin+0.160
 adorable+0.156
Neg logit contributions
ted-0.528
ting-0.526
 towards-0.508
oes-0.503
talk-0.501
Token to
Token ID284
Feature activation-1.21e-01

Pos logit contributions
 adorable+0.395
 best+0.385
 important+0.381
 idea+0.379
 precious+0.368
Neg logit contributions
oes-0.360
talk-0.356
ted-0.343
stack-0.327
rost-0.322
Token see
Token ID766
Feature activation-5.20e-03

Pos logit contributions
ted+0.207
oes+0.198
talk+0.198
ricanes+0.186
ting+0.183
Neg logit contributions
 idea-0.241
 adorable-0.231
 best-0.230
 important-0.222
 beautiful-0.217
Token the
Token ID262
Feature activation+2.62e-02

Pos logit contributions
oes+0.009
talk+0.009
rost+0.008
stack+0.008
ted+0.008
Neg logit contributions
 adorable-0.010
 best-0.010
 important-0.010
 beautiful-0.009
 idea-0.009
Token elephants
Token ID25368
Feature activation-2.68e-02

Pos logit contributions
 adorable+0.051
 idea+0.050
 important+0.050
 best+0.049
 precious+0.048
Neg logit contributions
oes-0.041
talk-0.040
ted-0.039
isk-0.037
stack-0.036
Token are
Token ID389
Feature activation+8.26e-02

Pos logit contributions
 adorable+0.067
 best+0.065
 important+0.065
 idea+0.064
 precious+0.063
Neg logit contributions
oes-0.053
talk-0.052
ted-0.050
stack-0.046
rost-0.046
Token not
Token ID407
Feature activation+8.83e-02

Pos logit contributions
 idea+0.110
 adorable+0.109
 best+0.103
 important+0.101
 loyal+0.097
Neg logit contributions
oes-0.181
ted-0.173
talk-0.171
stack-0.169
rost-0.167
Token flowers
Token ID12734
Feature activation-9.30e-02

Pos logit contributions
 adorable+0.131
 idea+0.127
 best+0.127
 prized+0.122
 precious+0.120
Neg logit contributions
oes-0.172
talk-0.164
ted-0.159
stack-0.158
isk-0.158
Token,
Token ID11
Feature activation-3.08e-02

Pos logit contributions
oes+0.193
talk+0.188
ted+0.185
stack+0.177
rost+0.174
Neg logit contributions
 adorable-0.141
 best-0.135
 important-0.135
 idea-0.134
 precious-0.129
Token my
Token ID616
Feature activation+1.07e-01

Pos logit contributions
oes+0.053
ted+0.053
rost+0.052
ting+0.050
stack+0.048
Neg logit contributions
 prized-0.049
 adorable-0.049
 idea-0.048
 best-0.047
 important-0.047
Token de
Token ID390
Feature activation+2.46e-01

Pos logit contributions
 adorable+0.175
 best+0.170
 important+0.169
 idea+0.168
 loyal+0.162
Neg logit contributions
oes-0.195
ted-0.189
talk-0.184
isk-0.178
stack-0.177
Tokenars
Token ID945
Feature activation-3.72e-02

Pos logit contributions
 adorable+0.512
 best+0.501
 important+0.496
 idea+0.494
 precious+0.482
Neg logit contributions
oes-0.380
talk-0.375
ted-0.361
stack-0.341
rost-0.335
Token.
Token ID13
Feature activation+5.24e-03

Pos logit contributions
oes+0.076
talk+0.074
ted+0.072
stack+0.069
rost+0.068
Neg logit contributions
 adorable-0.058
 best-0.056
 important-0.056
 idea-0.055
 precious-0.054
Token Those
Token ID5845
Feature activation+3.46e-02

Pos logit contributions
 adorable+0.010
 best+0.010
 important+0.010
 precious+0.010
 idea+0.010
Neg logit contributions
talk-0.009
oes-0.008
ted-0.008
isk-0.008
rost-0.007
Token are
Token ID389
Feature activation+9.15e-02

Pos logit contributions
 adorable+0.068
 best+0.066
 important+0.065
 idea+0.065
 precious+0.063
Neg logit contributions
oes-0.058
talk-0.057
ted-0.055
stack-0.053
rost-0.052
Token caul
Token ID46087
Feature activation+3.45e-02

Pos logit contributions
 adorable+0.142
 best+0.138
 important+0.137
 idea+0.132
 precious+0.132
Neg logit contributions
talk-0.192
oes-0.191
ted-0.185
stack-0.177
rost-0.174
Token
Token ID198
Feature activation-2.59e-02

Pos logit contributions
oes+0.055
talk+0.054
ted+0.054
stack+0.051
ting+0.050
Neg logit contributions
 adorable-0.052
 important-0.050
 best-0.049
 idea-0.049
 beautiful-0.048
Token"
Token ID1
Feature activation-2.20e-02

Pos logit contributions
oes+0.063
talk+0.061
ted+0.061
stack+0.058
isk+0.057
Neg logit contributions
 adorable-0.032
 important-0.030
 best-0.029
 precious-0.029
 beautiful-0.029
TokenHello
Token ID15496
Feature activation+4.54e-02

Pos logit contributions
oes+0.034
stack+0.032
talk+0.031
isk+0.031
ted+0.031
Neg logit contributions
 adorable-0.047
 best-0.045
 important-0.044
 beautiful-0.044
 precious-0.044
Token,
Token ID11
Feature activation-3.68e-02

Pos logit contributions
 important+0.069
 adorable+0.069
 idea+0.067
 precious+0.065
 best+0.064
Neg logit contributions
oes-0.090
ted-0.089
talk-0.086
rost-0.084
use-0.083
Token my
Token ID616
Feature activation+1.13e-01

Pos logit contributions
oes+0.078
talk+0.076
ted+0.076
rost+0.073
stack+0.072
Neg logit contributions
 adorable-0.052
 important-0.051
 idea-0.051
 best-0.051
 precious-0.048
Token de
Token ID390
Feature activation+2.46e-01

Pos logit contributions
 impressed+0.142
 pleased+0.133
 prized+0.130
 important+0.130
 thrilled+0.127
Neg logit contributions
ted-0.224
oes-0.223
rost-0.221
isk-0.220
iffs-0.210
Tokenars
Token ID945
Feature activation-3.85e-02

Pos logit contributions
 adorable+0.511
 best+0.500
 important+0.496
 idea+0.492
 precious+0.482
Neg logit contributions
oes-0.380
talk-0.374
ted-0.361
stack-0.342
rost-0.334
Token.
Token ID13
Feature activation+7.69e-03

Pos logit contributions
oes+0.077
talk+0.076
ted+0.073
stack+0.071
use+0.070
Neg logit contributions
 adorable-0.061
 important-0.059
 precious-0.058
 best-0.058
 idea-0.057
Token What
Token ID1867
Feature activation-2.23e-02

Pos logit contributions
 adorable+0.013
 best+0.012
 loyal+0.011
 important+0.011
 precious+0.011
Neg logit contributions
talk-0.015
oes-0.015
ted-0.014
ales-0.014
isk-0.014
Token do
Token ID466
Feature activation+1.92e-02

Pos logit contributions
oes+0.027
talk+0.023
 gra+0.022
ted+0.022
rost+0.022
Neg logit contributions
 pleased-0.048
 adorable-0.048
 idea-0.047
 prized-0.047
 precious-0.047
Token you
Token ID345
Feature activation+7.03e-03

Pos logit contributions
 adorable+0.031
 best+0.030
 important+0.030
 idea+0.029
 precious+0.028
Neg logit contributions
talk-0.039
oes-0.039
ted-0.038
stack-0.037
isk-0.036
Token
Token ID198
Feature activation-2.79e-02

Pos logit contributions
oes+0.059
talk+0.058
ted+0.057
stack+0.054
ting+0.053
Neg logit contributions
 adorable-0.060
 important-0.058
 best-0.058
 idea-0.057
 precious-0.056
Token"
Token ID1
Feature activation-2.21e-02

Pos logit contributions
oes+0.067
talk+0.065
ted+0.065
stack+0.061
isk+0.061
Neg logit contributions
 adorable-0.035
 important-0.034
 best-0.033
 precious-0.032
 beautiful-0.032
TokenHello
Token ID15496
Feature activation+4.49e-02

Pos logit contributions
oes+0.037
stack+0.033
isk+0.032
ted+0.032
ales+0.031
Neg logit contributions
 adorable-0.045
 beautiful-0.043
 best-0.042
 precious-0.041
 loyal-0.041
Token,
Token ID11
Feature activation-3.74e-02

Pos logit contributions
 important+0.073
 adorable+0.071
 idea+0.069
 valuable+0.069
 precious+0.068
Neg logit contributions
oes-0.085
ted-0.082
talk-0.080
rost-0.078
use-0.077
Token my
Token ID616
Feature activation+1.10e-01

Pos logit contributions
oes+0.075
talk+0.073
ted+0.072
rost+0.070
stack+0.069
Neg logit contributions
 adorable-0.056
 idea-0.056
 important-0.055
 best-0.055
 precious-0.052
Token de
Token ID390
Feature activation+2.45e-01

Pos logit contributions
 impressed+0.150
 pleased+0.145
 prized+0.136
 thrilled+0.134
 motivated+0.132
Neg logit contributions
 Shelter-0.200
ted-0.199
isk-0.195
ricanes-0.194
oes-0.193
Tokenars
Token ID945
Feature activation-3.65e-02

Pos logit contributions
 adorable+0.518
 best+0.506
 important+0.501
 idea+0.499
 precious+0.487
Neg logit contributions
oes-0.372
talk-0.367
ted-0.353
stack-0.333
rost-0.327
Token,"
Token ID553
Feature activation-8.86e-03

Pos logit contributions
talk+0.081
oes+0.081
ted+0.077
stack+0.076
use+0.075
Neg logit contributions
 adorable-0.049
 important-0.048
 precious-0.046
 best-0.046
 idea-0.044
Token she
Token ID673
Feature activation-2.83e-02

Pos logit contributions
oes+0.013
talk+0.013
ted+0.012
stack+0.011
rost+0.011
Neg logit contributions
 adorable-0.019
 best-0.019
 important-0.019
 idea-0.019
 precious-0.018
Token said
Token ID531
Feature activation+1.07e-01

Pos logit contributions
talk+0.051
ted+0.050
oes+0.048
Dog+0.045
isk+0.044
Neg logit contributions
 best-0.050
 adorable-0.049
 important-0.049
 precious-0.048
 idea-0.048
Token.
Token ID13
Feature activation+1.41e-02

Pos logit contributions
 adorable+0.161
 best+0.157
 idea+0.153
 important+0.152
 precious+0.145
Neg logit contributions
oes-0.225
talk-0.222
ted-0.215
stack-0.210
isk-0.208
TokenPeter
Token ID19727
Feature activation-6.76e-02

Pos logit contributions
oes+0.222
talk+0.218
ted+0.199
rost+0.198
isk+0.197
Neg logit contributions
 best-0.268
 adorable-0.264
 idea-0.257
 important-0.252
 beautiful-0.248
Token was
Token ID373
Feature activation+1.13e-01

Pos logit contributions
oes+0.128
talk+0.127
ted+0.123
stack+0.117
rost+0.116
Neg logit contributions
 adorable-0.115
 best-0.112
 important-0.111
 idea-0.110
 precious-0.107
Token a
Token ID257
Feature activation+1.93e-01

Pos logit contributions
 precious+0.189
 idea+0.182
 adorable+0.181
 valuable+0.179
 important+0.176
Neg logit contributions
ted-0.184
oes-0.182
stack-0.182
 gra-0.180
talk-0.180
Token mis
Token ID2984
Feature activation+1.42e-01

Pos logit contributions
 idea+0.217
 mistakes+0.199
 important+0.196
 best+0.191
vern+0.189
Neg logit contributions
oes-0.409
Dog-0.401
 gra-0.401
rost-0.393
stack-0.389
Tokenchie
Token ID3043
Feature activation+1.52e-01

Pos logit contributions
 best+0.315
 idea+0.303
 adorable+0.301
 important+0.300
 precious+0.291
Neg logit contributions
talk-0.229
oes-0.216
Dog-0.182
isk-0.182
ting-0.177
Tokenvous
Token ID31222
Feature activation+2.45e-01

Pos logit contributions
 best+0.247
 idea+0.243
 precious+0.242
 adorable+0.242
 important+0.230
Neg logit contributions
oes-0.297
talk-0.291
ted-0.268
 gra-0.254
ting-0.252
Token 3
Token ID513
Feature activation+1.65e-01

Pos logit contributions
 important+0.358
 precious+0.349
 adorable+0.344
 best+0.343
 idea+0.335
Neg logit contributions
oes-0.481
ted-0.466
talk-0.461
ting-0.458
stack-0.446
Token year
Token ID614
Feature activation+8.30e-02

Pos logit contributions
 precious+0.253
 best+0.252
 adorable+0.248
 important+0.239
 idea+0.235
Neg logit contributions
talk-0.325
ted-0.313
oes-0.303
ting-0.299
rost-0.288
Token old
Token ID1468
Feature activation+1.59e-01

Pos logit contributions
 best+0.071
 adorable+0.070
 idea+0.068
 important+0.062
 precious+0.061
Neg logit contributions
oes-0.224
talk-0.220
ted-0.215
stack-0.207
rost-0.206
Token who
Token ID508
Feature activation-2.44e-02

Pos logit contributions
 adorable+0.276
 important+0.272
 best+0.269
 idea+0.268
 precious+0.266
Neg logit contributions
oes-0.285
talk-0.285
ted-0.283
stack-0.269
ting-0.266
Token liked
Token ID8288
Feature activation+1.14e-01

Pos logit contributions
oes+0.048
talk+0.047
ted+0.046
stack+0.044
rost+0.043
Neg logit contributions
 adorable-0.040
 best-0.039
 idea-0.038
 important-0.038
 precious-0.037
Token on
Token ID319
Feature activation-1.36e-01

Pos logit contributions
 adorable+0.083
 best+0.080
 idea+0.079
 important+0.078
 precious+0.075
Neg logit contributions
oes-0.107
talk-0.107
ted-0.101
rost-0.099
isk-0.098
Token her
Token ID607
Feature activation+1.72e-02

Pos logit contributions
oes+0.241
talk+0.237
ted+0.225
isk+0.215
rost+0.214
Neg logit contributions
 adorable-0.259
 best-0.250
 idea-0.249
 important-0.247
 beautiful-0.239
Token face
Token ID1986
Feature activation-4.05e-02

Pos logit contributions
 adorable+0.028
 important+0.027
 best+0.027
 idea+0.026
 precious+0.026
Neg logit contributions
oes-0.032
talk-0.031
ted-0.030
stack-0.030
rost-0.029
Token.
Token ID13
Feature activation+3.47e-02

Pos logit contributions
oes+0.086
talk+0.086
ted+0.082
rost+0.080
stack+0.079
Neg logit contributions
 adorable-0.061
 best-0.059
 important-0.058
 idea-0.057
 precious-0.055
Token Ready
Token ID23432
Feature activation-6.97e-02

Pos logit contributions
 adorable+0.068
 best+0.065
 important+0.064
 idea+0.064
 loyal+0.062
Neg logit contributions
talk-0.059
oes-0.058
ted-0.055
rost-0.052
isk-0.052
Token to
Token ID284
Feature activation-2.71e-01

Pos logit contributions
talk+0.157
Dog+0.146
ted+0.146
oes+0.145
stack+0.143
Neg logit contributions
 best-0.095
 adorable-0.091
 idea-0.089
 important-0.085
 beautiful-0.081
Token greet
Token ID12589
Feature activation-7.30e-02

Pos logit contributions
talk+0.461
ted+0.444
isk+0.443
oes+0.439
rost+0.432
Neg logit contributions
 idea-0.529
 beautiful-0.510
 adorable-0.509
 best-0.506
 amazing-0.488
Token her
Token ID607
Feature activation-3.42e-05

Pos logit contributions
oes+0.137
talk+0.137
ted+0.126
stack+0.125
rost+0.123
Neg logit contributions
 adorable-0.131
 best-0.127
 important-0.123
 idea-0.121
 beautiful-0.121
Token friend
Token ID1545
Feature activation-7.36e-02

Pos logit contributions
oes+0.000
talk+0.000
stack+0.000
ted+0.000
 gra+0.000
Neg logit contributions
 adorable-0.000
 important-0.000
 precious-0.000
 idea-0.000
 best-0.000
Token,
Token ID11
Feature activation-6.57e-02

Pos logit contributions
oes+0.154
talk+0.153
ted+0.148
stack+0.143
rost+0.142
Neg logit contributions
 adorable-0.111
 best-0.109
 important-0.106
 idea-0.106
 precious-0.101
Token Molly
Token ID30236
Feature activation+3.53e-02

Pos logit contributions
oes+0.120
talk+0.120
ted+0.115
rost+0.108
stack+0.108
Neg logit contributions
 adorable-0.119
 best-0.116
 idea-0.114
 important-0.114
 precious-0.109
Tokenapped
Token ID6320
Feature activation+5.41e-02

Pos logit contributions
 important+0.248
 adorable+0.236
 idea+0.235
 precious+0.233
 best+0.233
Neg logit contributions
ted-0.194
talk-0.189
oes-0.183
iffs-0.168
aled-0.161
Token and
Token ID290
Feature activation-5.67e-02

Pos logit contributions
 adorable+0.113
 best+0.112
 precious+0.111
 important+0.111
 idea+0.110
Neg logit contributions
oes-0.070
ted-0.069
talk-0.068
 south-0.063
ting-0.061
Token shouted
Token ID17293
Feature activation+7.47e-02

Pos logit contributions
ted+0.097
talk+0.091
ting+0.085
oes+0.085
stack+0.081
Neg logit contributions
 adorable-0.104
 important-0.103
 idea-0.102
 prized-0.100
 precious-0.100
Token "
Token ID366
Feature activation-4.53e-02

Pos logit contributions
 adorable+0.126
 idea+0.122
 best+0.122
 important+0.121
 precious+0.120
Neg logit contributions
ted-0.134
oes-0.133
talk-0.132
ting-0.127
stack-0.123
Tokenready
Token ID1493
Feature activation-8.85e-02

Pos logit contributions
oes+0.071
talk+0.070
ted+0.068
stack+0.062
rost+0.061
Neg logit contributions
 adorable-0.093
 best-0.091
 idea-0.091
 important-0.091
 precious-0.088
Token to
Token ID284
Feature activation-2.69e-01

Pos logit contributions
oes+0.204
talk+0.204
ted+0.197
stack+0.190
rost+0.189
Neg logit contributions
 adorable-0.114
 best-0.112
 idea-0.109
 important-0.108
 precious-0.103
Token go
Token ID467
Feature activation-1.23e-01

Pos logit contributions
talk+0.425
oes+0.420
ted+0.414
isk+0.393
stack+0.388
Neg logit contributions
 idea-0.546
 adorable-0.543
 best-0.529
 beautiful-0.524
 important-0.513
Token!"
Token ID2474
Feature activation-2.61e-02

Pos logit contributions
talk+0.314
oes+0.311
ted+0.305
rost+0.300
isk+0.295
Neg logit contributions
 adorable-0.126
 best-0.125
 idea-0.123
 important-0.118
 beautiful-0.112
Token She
Token ID1375
Feature activation-5.94e-02

Pos logit contributions
oes+0.048
talk+0.048
use+0.046
rost+0.046
ted+0.045
Neg logit contributions
 adorable-0.045
 best-0.045
 idea-0.044
 important-0.042
 beautiful-0.041
Token was
Token ID373
Feature activation+1.19e-01

Pos logit contributions
talk+0.085
ted+0.083
oes+0.080
stack+0.073
isk+0.070
Neg logit contributions
 adorable-0.129
 important-0.127
 best-0.126
 idea-0.125
 precious-0.122
Token ready
Token ID3492
Feature activation+4.32e-02

Pos logit contributions
 precious+0.156
 idea+0.156
 adorable+0.154
 important+0.154
 prized+0.145
Neg logit contributions
oes-0.226
talk-0.225
ted-0.225
ting-0.223
aled-0.218
Token if
Token ID611
Feature activation-1.75e-02

Pos logit contributions
oes+0.075
talk+0.074
ted+0.071
stack+0.068
rost+0.067
Neg logit contributions
 adorable-0.063
 best-0.061
 important-0.060
 idea-0.060
 precious-0.058
Token you
Token ID345
Feature activation-2.42e-02

Pos logit contributions
talk+0.035
oes+0.034
ted+0.033
stack+0.032
isk+0.032
Neg logit contributions
 adorable-0.029
 best-0.029
 important-0.028
 idea-0.027
 precious-0.027
Token fall
Token ID2121
Feature activation-4.73e-02

Pos logit contributions
talk+0.042
oes+0.042
ted+0.041
stack+0.040
rost+0.038
Neg logit contributions
 adorable-0.044
 idea-0.044
 best-0.044
 important-0.043
 precious-0.042
Token.
Token ID13
Feature activation-8.23e-03

Pos logit contributions
talk+0.079
oes+0.076
rost+0.072
ted+0.072
stack+0.071
Neg logit contributions
 best-0.094
 adorable-0.093
 important-0.091
 idea-0.089
 beautiful-0.087
Token Ready
Token ID23432
Feature activation-6.05e-02

Pos logit contributions
talk+0.014
oes+0.014
ted+0.014
stack+0.013
rost+0.013
Neg logit contributions
 adorable-0.015
 best-0.015
 important-0.015
 idea-0.014
 precious-0.014
Token to
Token ID284
Feature activation-2.65e-01

Pos logit contributions
talk+0.138
ted+0.126
Dog+0.125
oes+0.124
stack+0.124
Neg logit contributions
 best-0.085
 adorable-0.079
 idea-0.076
 important-0.075
 precious-0.071
Token try
Token ID1949
Feature activation-5.90e-02

Pos logit contributions
talk+0.461
ted+0.441
oes+0.440
rost+0.421
isk+0.420
Neg logit contributions
 idea-0.520
 adorable-0.504
 best-0.502
 beautiful-0.493
 important-0.484
Token again
Token ID757
Feature activation-1.25e-01

Pos logit contributions
talk+0.129
oes+0.127
ted+0.123
stack+0.118
rost+0.118
Neg logit contributions
 adorable-0.086
 best-0.085
 important-0.083
 idea-0.081
 precious-0.079
Token?"
Token ID1701
Feature activation-9.65e-03

Pos logit contributions
talk+0.277
oes+0.276
ted+0.269
stack+0.258
rost+0.256
Neg logit contributions
 adorable-0.175
 best-0.171
 important-0.169
 idea-0.166
 precious-0.159
Token
Token ID198
Feature activation-2.73e-02

Pos logit contributions
oes+0.017
talk+0.017
ted+0.016
stack+0.015
rost+0.015
Neg logit contributions
 adorable-0.018
 best-0.018
 idea-0.018
 important-0.018
 precious-0.017
Token
Token ID198
Feature activation-2.31e-02

Pos logit contributions
oes+0.043
talk+0.043
ted+0.041
stack+0.039
rost+0.038
Neg logit contributions
 adorable-0.056
 best-0.054
 important-0.054
 idea-0.054
 precious-0.052
Token in
Token ID287
Feature activation-1.76e-01

Pos logit contributions
oes+0.152
talk+0.149
ted+0.139
stack+0.137
rost+0.136
Neg logit contributions
 best-0.177
 adorable-0.176
 important-0.170
 beautiful-0.165
 idea-0.165
Token her
Token ID607
Feature activation-2.12e-02

Pos logit contributions
stack+0.224
talk+0.223
ted+0.214
oes+0.213
isk+0.210
Neg logit contributions
 adorable-0.380
 important-0.370
 best-0.363
 precious-0.358
 idea-0.356
Token bag
Token ID6131
Feature activation-7.07e-02

Pos logit contributions
oes+0.039
talk+0.039
ted+0.038
stack+0.037
rost+0.036
Neg logit contributions
 adorable-0.036
 important-0.035
 idea-0.035
 best-0.035
 precious-0.034
Token.
Token ID13
Feature activation+9.20e-03

Pos logit contributions
oes+0.134
talk+0.132
ted+0.129
stack+0.124
isk+0.120
Neg logit contributions
 adorable-0.121
 best-0.116
 important-0.116
 idea-0.115
 precious-0.113
Token Ready
Token ID23432
Feature activation-6.18e-02

Pos logit contributions
 adorable+0.018
 best+0.017
 important+0.017
 idea+0.017
 beautiful+0.016
Neg logit contributions
talk-0.016
oes-0.015
rost-0.014
ted-0.014
use-0.014
Token to
Token ID284
Feature activation-2.61e-01

Pos logit contributions
talk+0.137
oes+0.127
ted+0.126
Dog+0.125
stack+0.121
Neg logit contributions
 best-0.089
 adorable-0.087
 idea-0.084
 important-0.082
 precious-0.079
Token go
Token ID467
Feature activation-1.23e-01

Pos logit contributions
talk+0.463
oes+0.458
ted+0.440
isk+0.432
stack+0.431
Neg logit contributions
 idea-0.508
 adorable-0.489
 best-0.486
 beautiful-0.475
 amazing-0.464
Token,
Token ID11
Feature activation-1.04e-01

Pos logit contributions
talk+0.299
oes+0.293
ted+0.289
rost+0.283
isk+0.282
Neg logit contributions
 best-0.150
 adorable-0.145
 idea-0.141
 important-0.138
 beautiful-0.128
Token she
Token ID673
Feature activation-8.58e-02

Pos logit contributions
oes+0.183
talk+0.181
ted+0.174
stack+0.164
rost+0.164
Neg logit contributions
 adorable-0.196
 best-0.192
 idea-0.189
 important-0.189
 precious-0.182
Token ran
Token ID4966
Feature activation-5.89e-02

Pos logit contributions
oes+0.143
talk+0.141
ted+0.136
stack+0.130
rost+0.129
Neg logit contributions
 adorable-0.166
 best-0.164
 important-0.161
 idea-0.161
 precious-0.156
Token back
Token ID736
Feature activation-1.39e-01

Pos logit contributions
oes+0.110
talk+0.108
ted+0.104
stack+0.100
rost+0.099
Neg logit contributions
 adorable-0.104
 best-0.102
 important-0.101
 idea-0.100
 precious-0.097
Token.
Token ID13
Feature activation+1.32e-03

Pos logit contributions
 adorable+0.162
 important+0.155
 precious+0.154
 best+0.153
 idea+0.152
Neg logit contributions
oes-0.172
talk-0.169
ted-0.168
ting-0.159
stack-0.155
Token
Token ID198
Feature activation-4.21e-02

Pos logit contributions
 best+0.002
 adorable+0.002
 idea+0.002
 important+0.002
 beautiful+0.002
Neg logit contributions
talk-0.003
ted-0.003
isk-0.003
stack-0.003
ting-0.003
Token
Token ID198
Feature activation-3.84e-02

Pos logit contributions
talk+0.078
ted+0.078
oes+0.077
stack+0.075
ting+0.074
Neg logit contributions
 adorable-0.073
 important-0.072
 best-0.072
 idea-0.069
 beautiful-0.068
Token"
Token ID1
Feature activation-2.73e-02

Pos logit contributions
oes+0.062
talk+0.062
ted+0.060
stack+0.057
isk+0.056
Neg logit contributions
 adorable-0.077
 best-0.075
 important-0.075
 idea-0.073
 precious-0.072
TokenTime
Token ID7575
Feature activation-5.67e-02

Pos logit contributions
oes+0.046
talk+0.044
ted+0.044
stack+0.043
isk+0.042
Neg logit contributions
 adorable-0.052
 best-0.051
 important-0.049
 beautiful-0.048
 idea-0.047
Token to
Token ID284
Feature activation-2.57e-01

Pos logit contributions
talk+0.076
oes+0.073
ted+0.069
stack+0.068
isk+0.064
Neg logit contributions
 best-0.131
 adorable-0.129
 important-0.128
 idea-0.128
 beautiful-0.122
Token bat
Token ID7365
Feature activation-2.56e-02

Pos logit contributions
talk+0.426
oes+0.426
ted+0.412
stack+0.387
rost+0.378
Neg logit contributions
 adorable-0.508
 best-0.495
 idea-0.495
 important-0.488
 beautiful-0.470
Tokenhe
Token ID258
Feature activation-7.18e-02

Pos logit contributions
oes+0.040
talk+0.039
ted+0.038
stack+0.036
rost+0.035
Neg logit contributions
 adorable-0.053
 best-0.052
 important-0.051
 idea-0.051
 precious-0.050
Token,
Token ID11
Feature activation-8.54e-02

Pos logit contributions
talk+0.130
oes+0.130
stack+0.128
ted+0.128
rost+0.122
Neg logit contributions
 adorable-0.116
 precious-0.115
 prized-0.112
 important-0.110
 valuable-0.108
Token L
Token ID406
Feature activation+1.54e-01

Pos logit contributions
oes+0.122
talk+0.121
ted+0.116
stack+0.109
rost+0.107
Neg logit contributions
 adorable-0.188
 best-0.184
 important-0.182
 idea-0.181
 precious-0.177
Tokenila
Token ID10102
Feature activation+6.80e-02

Pos logit contributions
 adorable+0.298
 best+0.290
 important+0.288
 idea+0.286
 precious+0.280
Neg logit contributions
oes-0.261
talk-0.256
ted-0.247
stack-0.233
rost-0.233
Token.
Token ID13
Feature activation-8.07e-03

Pos logit contributions
talk+0.145
oes+0.142
ted+0.142
stack+0.135
isk+0.134
Neg logit contributions
 adorable-0.087
 best-0.087
 important-0.086
 idea-0.082
 precious-0.080
Token You
Token ID921
Feature activation+3.68e-02

Pos logit contributions
oes+0.014
talk+0.014
ted+0.013
stack+0.013
rost+0.013
Neg logit contributions
 adorable-0.015
 idea-0.015
 best-0.015
 important-0.015
 precious-0.014
Token should
Token ID815
Feature activation-6.71e-02

Pos logit contributions
 adorable+0.064
 best+0.062
 important+0.061
 idea+0.061
 precious+0.059
Neg logit contributions
oes-0.069
talk-0.069
ted-0.066
stack-0.063
rost-0.062
Token go
Token ID467
Feature activation-9.31e-02

Pos logit contributions
ted+0.124
oes+0.122
talk+0.120
ting+0.112
stack+0.111
Neg logit contributions
 adorable-0.121
 best-0.119
 idea-0.119
 important-0.119
 precious-0.111
Token back
Token ID736
Feature activation-1.67e-01

Pos logit contributions
ted+0.155
rost+0.153
stack+0.152
oes+0.147
 gra+0.147
Neg logit contributions
 best-0.185
 important-0.177
 adorable-0.175
 loyal-0.169
 precious-0.166
Token to
Token ID284
Feature activation-2.57e-01

Pos logit contributions
oes+0.264
ted+0.255
talk+0.254
stack+0.253
rost+0.244
Neg logit contributions
 best-0.341
 adorable-0.340
 important-0.340
 precious-0.326
 beautiful-0.322
Token your
Token ID534
Feature activation-3.29e-02

Pos logit contributions
talk+0.545
oes+0.542
ted+0.533
stack+0.500
rost+0.498
Neg logit contributions
 adorable-0.412
 best-0.405
 important-0.403
 idea-0.389
 beautiful-0.384
Token own
Token ID898
Feature activation+1.54e-02

Pos logit contributions
oes+0.067
talk+0.066
ted+0.063
stack+0.062
rost+0.061
Neg logit contributions
 adorable-0.051
 best-0.048
 idea-0.048
 important-0.048
 precious-0.046
Token country
Token ID1499
Feature activation-1.38e-01

Pos logit contributions
 adorable+0.027
 best+0.026
 important+0.026
 idea+0.026
 precious+0.025
Neg logit contributions
oes-0.029
talk-0.028
ted-0.027
stack-0.026
rost-0.026
Token and
Token ID290
Feature activation-8.28e-02

Pos logit contributions
oes+0.287
talk+0.284
ted+0.278
stack+0.265
rost+0.262
Neg logit contributions
 adorable-0.210
 best-0.204
 important-0.201
 idea-0.200
 precious-0.193
Token play
Token ID711
Feature activation-4.44e-02

Pos logit contributions
ted+0.159
oes+0.158
talk+0.157
rost+0.148
ting+0.145
Neg logit contributions
 best-0.147
 adorable-0.146
 important-0.145
 precious-0.140
 loyal-0.140
Token.
Token ID13
Feature activation+5.44e-03

Pos logit contributions
oes+0.127
talk+0.122
ted+0.114
rost+0.111
iffs+0.108
Neg logit contributions
 adorable-0.148
 best-0.142
 idea-0.141
 precious-0.140
 important-0.139
Token
Token ID198
Feature activation-2.95e-02

Pos logit contributions
 adorable+0.010
 important+0.010
 idea+0.010
 best+0.010
 beautiful+0.009
Neg logit contributions
talk-0.010
oes-0.010
ted-0.009
stack-0.009
isk-0.009
Token
Token ID198
Feature activation-2.47e-02

Pos logit contributions
oes+0.056
talk+0.054
ted+0.054
stack+0.051
ting+0.050
Neg logit contributions
 adorable-0.052
 important-0.051
 best-0.049
 idea-0.049
 beautiful-0.048
Token"
Token ID1
Feature activation-2.33e-02

Pos logit contributions
oes+0.063
talk+0.060
ted+0.060
stack+0.058
isk+0.056
Neg logit contributions
 adorable-0.029
 important-0.028
 precious-0.026
 loyal-0.026
 best-0.026
TokenTime
Token ID7575
Feature activation-5.14e-02

Pos logit contributions
oes+0.039
talk+0.036
stack+0.034
ted+0.033
ales+0.033
Neg logit contributions
 adorable-0.049
 best-0.046
 important-0.045
 precious-0.045
 beautiful-0.045
Token to
Token ID284
Feature activation-2.56e-01

Pos logit contributions
talk+0.066
oes+0.065
ted+0.059
stack+0.059
rost+0.056
Neg logit contributions
 best-0.121
 adorable-0.119
 important-0.118
 idea-0.117
 precious-0.113
Token get
Token ID651
Feature activation-1.67e-02

Pos logit contributions
oes+0.377
talk+0.373
ted+0.357
stack+0.333
ricanes+0.327
Neg logit contributions
 adorable-0.557
 idea-0.548
 best-0.547
 important-0.527
 beautiful-0.525
Token up
Token ID510
Feature activation-9.47e-02

Pos logit contributions
oes+0.028
talk+0.028
stack+0.027
ted+0.026
isk+0.026
Neg logit contributions
 adorable-0.033
 best-0.032
 important-0.032
 idea-0.030
 precious-0.030
Token,
Token ID11
Feature activation-6.91e-02

Pos logit contributions
oes+0.178
talk+0.172
ted+0.172
use+0.165
stack+0.165
Neg logit contributions
 adorable-0.152
 idea-0.144
 best-0.142
 precious-0.142
 important-0.141
Token my
Token ID616
Feature activation+7.94e-02

Pos logit contributions
oes+0.121
rost+0.118
ted+0.118
ting+0.113
 south+0.113
Neg logit contributions
 prized-0.115
 important-0.113
 valuable-0.109
 precious-0.108
 adorable-0.107
Token loves
Token ID10408
Feature activation+8.37e-02

Pos logit contributions
 impressed+0.096
vern+0.091
 prized+0.090
 important+0.089
 valuable+0.088
Neg logit contributions
rost-0.157
oes-0.156
ted-0.153
 Rex-0.145
 hike-0.143
Token.
Token ID13
Feature activation+6.50e-03

Pos logit contributions
oes+0.083
talk+0.083
ted+0.080
stack+0.077
rost+0.076
Neg logit contributions
 adorable-0.067
 best-0.064
 important-0.064
 idea-0.063
 precious-0.062
Token
Token ID198
Feature activation-3.28e-02

Pos logit contributions
 adorable+0.011
 best+0.010
 important+0.010
 idea+0.010
 beautiful+0.010
Neg logit contributions
oes-0.014
talk-0.013
isk-0.012
ales-0.012
stack-0.012
Token
Token ID198
Feature activation-2.85e-02

Pos logit contributions
oes+0.061
talk+0.059
ted+0.057
stack+0.055
isk+0.054
Neg logit contributions
 adorable-0.060
 important-0.058
 best-0.057
 idea-0.057
 precious-0.055
Token"
Token ID1
Feature activation-2.14e-02

Pos logit contributions
oes+0.067
talk+0.065
ted+0.064
stack+0.062
isk+0.061
Neg logit contributions
 adorable-0.038
 important-0.036
 best-0.036
 idea-0.034
 precious-0.034
TokenTime
Token ID7575
Feature activation-5.63e-02

Pos logit contributions
oes+0.042
stack+0.036
ales+0.035
talk+0.035
isk+0.034
Neg logit contributions
 adorable-0.039
 best-0.038
 beautiful-0.037
 precious-0.037
 loyal-0.037
Token to
Token ID284
Feature activation-2.55e-01

Pos logit contributions
talk+0.074
oes+0.074
stack+0.066
ted+0.066
rost+0.064
Neg logit contributions
 best-0.130
 adorable-0.129
 important-0.127
 idea-0.126
 precious-0.122
Token go
Token ID467
Feature activation-1.14e-01

Pos logit contributions
oes+0.365
talk+0.360
ted+0.350
stack+0.324
rost+0.316
Neg logit contributions
 adorable-0.559
 idea-0.554
 best-0.548
 important-0.537
 beautiful-0.526
Token,
Token ID11
Feature activation-9.58e-02

Pos logit contributions
talk+0.219
oes+0.217
ted+0.214
rost+0.208
isk+0.206
Neg logit contributions
 best-0.189
 adorable-0.187
 idea-0.185
 important-0.181
 beautiful-0.175
Token kids
Token ID3988
Feature activation-6.49e-02

Pos logit contributions
oes+0.170
talk+0.168
ted+0.167
rost+0.162
ting+0.158
Neg logit contributions
 adorable-0.161
 important-0.161
 best-0.157
 idea-0.155
 prized-0.155
Token!"
Token ID2474
Feature activation-2.21e-02

Pos logit contributions
oes+0.163
talk+0.161
ted+0.158
stack+0.152
rost+0.150
Neg logit contributions
 adorable-0.072
 best-0.070
 idea-0.068
 important-0.068
 precious-0.064
Token she
Token ID673
Feature activation-3.12e-02

Pos logit contributions
oes+0.036
talk+0.035
ted+0.033
isk+0.033
stack+0.032
Neg logit contributions
 adorable-0.045
 best-0.044
 idea-0.044
 important-0.043
 precious-0.042
Token the
Token ID262
Feature activation-3.74e-02

Pos logit contributions
oes+0.242
talk+0.239
ted+0.230
stack+0.218
rost+0.214
Neg logit contributions
 adorable-0.339
 best-0.331
 important-0.328
 idea-0.326
 precious-0.318
Token park
Token ID3952
Feature activation-1.15e-01

Pos logit contributions
oes+0.069
talk+0.068
ted+0.066
isk+0.063
rost+0.061
Neg logit contributions
 adorable-0.070
 best-0.069
 important-0.067
 beautiful-0.066
 precious-0.065
Token.
Token ID13
Feature activation+1.77e-02

Pos logit contributions
oes+0.245
talk+0.244
ted+0.238
rost+0.227
stack+0.224
Neg logit contributions
 adorable-0.171
 best-0.168
 idea-0.164
 important-0.164
 precious-0.155
Token "
Token ID366
Feature activation-3.20e-02

Pos logit contributions
 adorable+0.033
 idea+0.033
 important+0.031
 best+0.030
 loyal+0.030
Neg logit contributions
talk-0.031
oes-0.031
use-0.030
ted-0.029
ting-0.029
TokenReady
Token ID35474
Feature activation-5.38e-02

Pos logit contributions
oes+0.053
ted+0.048
stack+0.047
talk+0.046
isk+0.045
Neg logit contributions
 adorable-0.063
 best-0.059
 important-0.059
 loyal-0.058
 precious-0.058
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
talk+0.129
oes+0.128
ted+0.125
stack+0.121
isk+0.119
Neg logit contributions
 best-0.064
 adorable-0.063
 idea-0.061
 important-0.059
 precious-0.056
Token make
Token ID787
Feature activation-5.50e-02

Pos logit contributions
talk+0.419
ted+0.403
oes+0.401
isk+0.387
stack+0.385
Neg logit contributions
 idea-0.505
 adorable-0.500
 best-0.496
 beautiful-0.489
 important-0.468
Token the
Token ID262
Feature activation-1.35e-02

Pos logit contributions
ted+0.093
oes+0.093
talk+0.092
stack+0.084
ting+0.084
Neg logit contributions
 adorable-0.101
 important-0.099
 idea-0.098
 best-0.097
 precious-0.097
Token k
Token ID479
Feature activation+1.54e-02

Pos logit contributions
oes+0.021
stack+0.020
talk+0.019
ted+0.019
 gra+0.019
Neg logit contributions
 adorable-0.026
 important-0.026
 precious-0.025
 idea-0.025
 impressed-0.025
Tokenite
Token ID578
Feature activation-9.74e-02

Pos logit contributions
 adorable+0.031
 idea+0.031
 important+0.031
 best+0.030
 precious+0.030
Neg logit contributions
oes-0.025
talk-0.025
ted-0.024
stack-0.023
iffs-0.022
Token work
Token ID670
Feature activation-2.15e-02

Pos logit contributions
oes+0.167
talk+0.156
stack+0.146
ted+0.146
isk+0.142
Neg logit contributions
 adorable-0.184
 important-0.180
 precious-0.176
 best-0.175
 idea-0.171
Token will
Token ID481
Feature activation-4.12e-02

Pos logit contributions
 adorable+0.033
 best+0.033
 important+0.032
 idea+0.032
 precious+0.031
Neg logit contributions
oes-0.028
talk-0.027
ted-0.026
stack-0.025
rost-0.024
Token be
Token ID307
Feature activation+5.41e-02

Pos logit contributions
oes+0.072
talk+0.071
ted+0.071
stack+0.065
rost+0.064
Neg logit contributions
 adorable-0.077
 best-0.076
 idea-0.075
 important-0.075
 precious-0.072
Token able
Token ID1498
Feature activation+3.73e-02

Pos logit contributions
 idea+0.071
ummy+0.069
 adorable+0.068
 prized+0.064
 darling+0.064
Neg logit contributions
 south-0.103
talk-0.102
ting-0.099
 away-0.099
Why-0.097
Token to
Token ID284
Feature activation-1.98e-01

Pos logit contributions
 adorable+0.101
 idea+0.100
 best+0.099
 important+0.098
 precious+0.096
Neg logit contributions
oes-0.033
talk-0.031
ted-0.029
stack-0.026
rost-0.025
Token climb
Token ID12080
Feature activation-1.72e-01

Pos logit contributions
ted+0.322
oes+0.319
talk+0.311
ricanes+0.287
stack+0.287
Neg logit contributions
 adorable-0.400
 best-0.400
 idea-0.399
 important-0.388
 beautiful-0.380
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
talk+0.324
oes+0.322
ted+0.300
rost+0.296
isk+0.293
Neg logit contributions
 adorable-0.306
 best-0.302
 important-0.297
 idea-0.287
 precious-0.284
Token the
Token ID262
Feature activation-6.14e-02

Pos logit contributions
oes+0.442
talk+0.442
ted+0.429
stack+0.400
rost+0.398
Neg logit contributions
 adorable-0.482
 best-0.471
 important-0.465
 idea-0.464
 precious-0.448
Token top
Token ID1353
Feature activation-4.27e-02

Pos logit contributions
oes+0.109
talk+0.107
ted+0.103
rost+0.098
isk+0.097
Neg logit contributions
 adorable-0.117
 best-0.115
 important-0.114
 idea-0.111
 precious-0.110
Token of
Token ID286
Feature activation-1.34e-01

Pos logit contributions
talk+0.097
oes+0.096
ted+0.095
stack+0.091
rost+0.091
Neg logit contributions
 adorable-0.056
 best-0.055
 idea-0.054
 important-0.054
 precious-0.050
Token the
Token ID262
Feature activation-3.73e-02

Pos logit contributions
oes+0.211
talk+0.208
ted+0.201
stack+0.191
rost+0.187
Neg logit contributions
 adorable-0.273
 best-0.266
 important-0.264
 idea-0.263
 precious-0.256
Token tallest
Token ID38760
Feature activation+1.24e-01

Pos logit contributions
oes+0.057
talk+0.056
ted+0.054
stack+0.051
rost+0.050
Neg logit contributions
 adorable-0.079
 best-0.077
 important-0.076
 idea-0.076
 precious-0.074
Token big
Token ID1263
Feature activation+1.24e-01

Pos logit contributions
 adorable+0.241
 best+0.227
 important+0.224
 idea+0.210
 precious+0.209
Neg logit contributions
oes-0.304
 gra-0.294
talk-0.292
rost-0.287
stack-0.284
Token zip
Token ID19974
Feature activation-8.32e-03

Pos logit contributions
 adorable+0.273
 best+0.268
 important+0.266
 precious+0.256
 impressed+0.254
Neg logit contributions
oes-0.158
talk-0.150
stack-0.149
ted-0.149
rost-0.149
Token!
Token ID0
Feature activation+1.66e-02

Pos logit contributions
oes+0.016
talk+0.015
ted+0.015
stack+0.014
rost+0.014
Neg logit contributions
 adorable-0.014
 best-0.014
 idea-0.014
 important-0.014
 precious-0.013
Token "
Token ID366
Feature activation-3.21e-02

Pos logit contributions
 adorable+0.037
 best+0.037
 important+0.036
 idea+0.036
 precious+0.035
Neg logit contributions
oes-0.023
talk-0.023
ted-0.022
stack-0.020
rost-0.020
TokenReady
Token ID35474
Feature activation-4.71e-02

Pos logit contributions
oes+0.052
talk+0.052
ted+0.050
stack+0.047
rost+0.047
Neg logit contributions
 adorable-0.064
 best-0.062
 important-0.061
 idea-0.061
 precious-0.060
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
talk+0.107
oes+0.107
ted+0.104
stack+0.100
rost+0.098
Neg logit contributions
 adorable-0.062
 best-0.062
 idea-0.060
 important-0.059
 precious-0.056
Token go
Token ID467
Feature activation-1.22e-01

Pos logit contributions
oes+0.361
talk+0.358
ted+0.340
stack+0.322
rost+0.316
Neg logit contributions
 adorable-0.563
 idea-0.554
 best-0.554
 important-0.545
 beautiful-0.527
Token!"
Token ID2474
Feature activation-1.95e-02

Pos logit contributions
talk+0.298
oes+0.297
ted+0.289
rost+0.284
stack+0.280
Neg logit contributions
 adorable-0.141
 best-0.139
 idea-0.137
 important-0.134
 beautiful-0.127
Token she
Token ID673
Feature activation-4.24e-02

Pos logit contributions
oes+0.032
talk+0.032
ted+0.031
stack+0.028
rost+0.028
Neg logit contributions
 adorable-0.038
 best-0.038
 important-0.037
 idea-0.036
 precious-0.036
Token shouted
Token ID17293
Feature activation+7.97e-02

Pos logit contributions
talk+0.064
ted+0.059
oes+0.057
ricane+0.050
 towards+0.049
Neg logit contributions
 adorable-0.088
 precious-0.087
 best-0.087
 important-0.086
 loyal-0.081
Token.
Token ID13
Feature activation+9.25e-03

Pos logit contributions
 best+0.152
 precious+0.150
 adorable+0.149
 important+0.147
 idea+0.143
Neg logit contributions
talk-0.123
ted-0.120
ting-0.115
 towards-0.115
oes-0.110
Token.
Token ID13
Feature activation-5.55e-04

Pos logit contributions
 adorable+0.022
 idea+0.022
 best+0.021
 important+0.021
 precious+0.020
Neg logit contributions
oes-0.024
talk-0.024
ted-0.023
rost-0.022
use-0.022
Token We
Token ID775
Feature activation-4.17e-02

Pos logit contributions
oes+0.001
talk+0.001
ted+0.001
stack+0.001
isk+0.001
Neg logit contributions
 adorable-0.001
 best-0.001
 important-0.001
 idea-0.001
 precious-0.001
Token should
Token ID815
Feature activation-6.33e-02

Pos logit contributions
oes+0.085
talk+0.081
ted+0.080
stack+0.077
ting+0.076
Neg logit contributions
 adorable-0.068
 best-0.067
 idea-0.066
 important-0.066
 precious-0.064
Token go
Token ID467
Feature activation-9.11e-02

Pos logit contributions
oes+0.127
talk+0.126
ted+0.125
stack+0.117
rost+0.115
Neg logit contributions
 adorable-0.103
 best-0.100
 important-0.099
 idea-0.099
 precious-0.095
Token back
Token ID736
Feature activation-1.68e-01

Pos logit contributions
ted+0.164
oes+0.162
stack+0.159
rost+0.157
talk+0.156
Neg logit contributions
 best-0.165
 adorable-0.160
 important-0.158
 loyal-0.152
 precious-0.151
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
oes+0.290
stack+0.280
ted+0.278
talk+0.276
isk+0.263
Neg logit contributions
 adorable-0.320
 best-0.316
 important-0.315
 precious-0.306
 beautiful-0.301
Token the
Token ID262
Feature activation-5.43e-02

Pos logit contributions
oes+0.470
talk+0.466
ted+0.457
stack+0.430
ricanes+0.423
Neg logit contributions
 adorable-0.478
 best-0.460
 important-0.459
 idea-0.449
 beautiful-0.446
Token park
Token ID3952
Feature activation-1.28e-01

Pos logit contributions
oes+0.091
talk+0.090
ted+0.088
stack+0.081
rost+0.081
Neg logit contributions
 adorable-0.107
 best-0.106
 important-0.104
 idea-0.102
 precious-0.101
Token.
Token ID13
Feature activation-1.69e-02

Pos logit contributions
oes+0.256
talk+0.254
ted+0.248
stack+0.236
rost+0.233
Neg logit contributions
 adorable-0.206
 best-0.201
 important-0.198
 idea-0.197
 precious-0.190
Token Mom
Token ID11254
Feature activation+4.57e-02

Pos logit contributions
oes+0.032
talk+0.032
ted+0.031
isk+0.030
stack+0.029
Neg logit contributions
 adorable-0.029
 best-0.028
 important-0.028
 idea-0.027
 precious-0.027
Token and
Token ID290
Feature activation-4.97e-02

Pos logit contributions
 adorable+0.075
 best+0.074
 important+0.072
 idea+0.072
 precious+0.070
Neg logit contributions
oes-0.090
talk-0.089
ted-0.086
stack-0.084
rost-0.083
Token he
Token ID339
Feature activation-2.78e-02

Pos logit contributions
oes+0.025
talk+0.024
ted+0.023
stack+0.022
rost+0.021
Neg logit contributions
 adorable-0.044
 idea-0.043
 best-0.042
 important-0.042
 precious-0.042
Token had
Token ID550
Feature activation+1.64e-02

Pos logit contributions
ted+0.040
talk+0.039
oes+0.037
ting+0.036
stack+0.035
Neg logit contributions
 idea-0.057
 best-0.056
 adorable-0.056
 important-0.053
 precious-0.053
Token managed
Token ID5257
Feature activation-9.14e-03

Pos logit contributions
 adorable+0.030
 idea+0.029
 best+0.029
 important+0.028
 precious+0.028
Neg logit contributions
talk-0.029
oes-0.029
ted-0.028
stack-0.026
rost-0.026
Token to
Token ID284
Feature activation-2.03e-01

Pos logit contributions
talk+0.012
oes+0.011
stack+0.011
ted+0.011
rost+0.010
Neg logit contributions
 important-0.022
 adorable-0.022
 best-0.022
 idea-0.021
 beautiful-0.021
Token climb
Token ID12080
Feature activation-1.70e-01

Pos logit contributions
oes+0.300
talk+0.297
ted+0.282
stack+0.268
rost+0.263
Neg logit contributions
 adorable-0.433
 best-0.423
 important-0.419
 idea-0.415
 precious-0.408
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
talk+0.319
oes+0.316
ted+0.297
isk+0.290
rost+0.289
Neg logit contributions
 adorable-0.300
 best-0.299
 important-0.290
 idea-0.286
 beautiful-0.281
Token the
Token ID262
Feature activation-5.39e-02

Pos logit contributions
oes+0.438
talk+0.437
ted+0.429
rost+0.394
stack+0.394
Neg logit contributions
 adorable-0.484
 best-0.477
 idea-0.471
 important-0.467
 beautiful-0.454
Token top
Token ID1353
Feature activation-3.73e-02

Pos logit contributions
oes+0.091
talk+0.090
ted+0.087
stack+0.082
rost+0.081
Neg logit contributions
 adorable-0.105
 best-0.102
 important-0.101
 idea-0.100
 precious-0.098
Token.
Token ID13
Feature activation+1.10e-02

Pos logit contributions
talk+0.088
ted+0.086
oes+0.086
stack+0.082
isk+0.082
Neg logit contributions
 adorable-0.047
 best-0.046
 idea-0.045
 important-0.044
 beautiful-0.042
Token And
Token ID843
Feature activation+3.46e-02

Pos logit contributions
 adorable+0.021
 best+0.020
 important+0.020
 idea+0.020
 loyal+0.019
Neg logit contributions
talk-0.019
oes-0.019
ted-0.018
use-0.017
rost-0.017
Token he
Token ID339
Feature activation-3.02e-02

Pos logit contributions
 adorable+0.072
 best+0.070
 important+0.069
 idea+0.069
 precious+0.067
Neg logit contributions
oes-0.054
talk-0.053
ted-0.051
stack-0.048
rost-0.048
Token.
Token ID13
Feature activation+8.98e-03

Pos logit contributions
ted+0.009
oes+0.009
talk+0.008
stack+0.008
ricanes+0.008
Neg logit contributions
 adorable-0.005
 best-0.004
 idea-0.004
 important-0.004
 loyal-0.004
Token
Token ID198
Feature activation-3.46e-02

Pos logit contributions
 adorable+0.014
 idea+0.013
 beautiful+0.013
 important+0.013
 best+0.013
Neg logit contributions
talk-0.019
oes-0.018
ales-0.018
isk-0.017
ted-0.017
Token
Token ID198
Feature activation-3.01e-02

Pos logit contributions
oes+0.067
talk+0.065
ted+0.065
stack+0.062
ting+0.061
Neg logit contributions
 adorable-0.058
 important-0.057
 best-0.056
 idea-0.056
 beautiful-0.054
Token"
Token ID1
Feature activation-2.31e-02

Pos logit contributions
oes+0.074
talk+0.072
ted+0.071
stack+0.069
rost+0.068
Neg logit contributions
 adorable-0.037
 important-0.035
 best-0.033
 precious-0.033
 loyal-0.032
TokenTime
Token ID7575
Feature activation-5.69e-02

Pos logit contributions
oes+0.039
stack+0.034
isk+0.033
talk+0.033
ales+0.032
Neg logit contributions
 adorable-0.048
 precious-0.046
 beautiful-0.045
 important-0.044
 best-0.044
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
talk+0.074
oes+0.074
stack+0.068
ted+0.066
rost+0.064
Neg logit contributions
 best-0.131
 adorable-0.129
 important-0.129
 idea-0.127
 precious-0.123
Token go
Token ID467
Feature activation-1.12e-01

Pos logit contributions
oes+0.411
talk+0.402
ted+0.395
stack+0.373
ricanes+0.368
Neg logit contributions
 idea-0.519
 adorable-0.515
 best-0.499
 beautiful-0.495
 important-0.482
Token,
Token ID11
Feature activation-8.58e-02

Pos logit contributions
oes+0.253
talk+0.253
ted+0.249
stack+0.245
rost+0.244
Neg logit contributions
 best-0.143
 adorable-0.142
 idea-0.139
 important-0.135
 beautiful-0.129
Token Tommy
Token ID19919
Feature activation+1.75e-02

Pos logit contributions
oes+0.167
ted+0.160
rost+0.156
talk+0.156
ting+0.155
Neg logit contributions
 important-0.133
 adorable-0.130
 prized-0.130
 precious-0.128
 best-0.128
Token,"
Token ID553
Feature activation-7.17e-03

Pos logit contributions
 adorable+0.028
 best+0.027
 precious+0.026
 idea+0.026
 prized+0.026
Neg logit contributions
talk-0.034
oes-0.033
ted-0.033
stack-0.032
isk-0.031
Token she
Token ID673
Feature activation-3.01e-02

Pos logit contributions
oes+0.011
talk+0.011
ted+0.011
stack+0.010
rost+0.010
Neg logit contributions
 adorable-0.014
 best-0.014
 important-0.014
 idea-0.014
 precious-0.013
Token.
Token ID13
Feature activation-4.67e-03

Pos logit contributions
talk+0.221
oes+0.220
ted+0.209
rost+0.205
isk+0.202
Neg logit contributions
 best-0.206
 adorable-0.205
 idea-0.197
 important-0.197
 precious-0.190
Token
Token ID198
Feature activation-3.44e-02

Pos logit contributions
talk+0.008
oes+0.008
ted+0.007
rost+0.007
isk+0.007
Neg logit contributions
 adorable-0.009
 best-0.009
 important-0.009
 idea-0.008
 loyal-0.008
Token
Token ID198
Feature activation-2.87e-02

Pos logit contributions
oes+0.060
talk+0.059
ted+0.057
stack+0.054
rost+0.053
Neg logit contributions
 adorable-0.065
 best-0.063
 important-0.063
 idea-0.062
 precious-0.060
Token"
Token ID1
Feature activation-2.27e-02

Pos logit contributions
oes+0.050
talk+0.049
ted+0.047
stack+0.045
rost+0.044
Neg logit contributions
 adorable-0.054
 best-0.053
 important-0.053
 idea-0.052
 precious-0.051
TokenReady
Token ID35474
Feature activation-4.49e-02

Pos logit contributions
oes+0.037
talk+0.036
ted+0.034
stack+0.033
isk+0.032
Neg logit contributions
 adorable-0.046
 best-0.044
 important-0.044
 precious-0.043
 idea-0.042
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
oes+0.104
talk+0.104
ted+0.101
stack+0.098
rost+0.096
Neg logit contributions
 adorable-0.057
 best-0.056
 idea-0.054
 important-0.054
 precious-0.051
Token catch
Token ID4929
Feature activation-6.98e-02

Pos logit contributions
talk+0.386
oes+0.385
ted+0.372
stack+0.352
isk+0.344
Neg logit contributions
 idea-0.533
 adorable-0.531
 best-0.531
 important-0.515
 beautiful-0.504
Token some
Token ID617
Feature activation-9.41e-03

Pos logit contributions
oes+0.124
talk+0.123
ted+0.117
stack+0.112
rost+0.111
Neg logit contributions
 adorable-0.130
 best-0.127
 important-0.125
 idea-0.124
 precious-0.121
Token fish
Token ID5916
Feature activation-7.90e-02

Pos logit contributions
oes+0.018
talk+0.017
ted+0.017
rost+0.016
use+0.016
Neg logit contributions
 adorable-0.016
 best-0.015
 idea-0.015
 important-0.014
 precious-0.014
Token?"
Token ID1701
Feature activation-2.90e-02

Pos logit contributions
oes+0.181
talk+0.180
ted+0.176
stack+0.169
rost+0.168
Neg logit contributions
 adorable-0.103
 best-0.100
 idea-0.098
 important-0.098
 precious-0.094
Token asked
Token ID1965
Feature activation+1.10e-02

Pos logit contributions
oes+0.052
talk+0.052
ted+0.050
stack+0.048
rost+0.047
Neg logit contributions
 adorable-0.053
 best-0.051
 important-0.051
 idea-0.051
 precious-0.049
Token had
Token ID550
Feature activation+1.63e-02

Pos logit contributions
talk+0.062
oes+0.061
ted+0.060
stack+0.056
rost+0.056
Neg logit contributions
 adorable-0.070
 best-0.068
 idea-0.066
 important-0.066
 precious-0.065
Token been
Token ID587
Feature activation+9.68e-02

Pos logit contributions
 adorable+0.031
 best+0.030
 idea+0.029
 important+0.029
 precious+0.028
Neg logit contributions
oes-0.028
talk-0.028
ted-0.027
stack-0.025
rost-0.025
Token able
Token ID1498
Feature activation+3.48e-02

Pos logit contributions
 adorable+0.146
 idea+0.146
 best+0.135
 important+0.133
 precious+0.133
Neg logit contributions
talk-0.190
oes-0.188
ted-0.186
ting-0.178
isk-0.174
Token to
Token ID284
Feature activation-1.98e-01

Pos logit contributions
 adorable+0.092
 best+0.090
 idea+0.089
 important+0.089
 precious+0.088
Neg logit contributions
oes-0.034
talk-0.033
ted-0.031
stack-0.028
rost-0.027
Token climb
Token ID12080
Feature activation-1.75e-01

Pos logit contributions
oes+0.298
talk+0.292
ted+0.287
rost+0.262
stack+0.259
Neg logit contributions
 adorable-0.420
 best-0.413
 idea-0.412
 important-0.408
 precious-0.393
Token to
Token ID284
Feature activation-2.54e-01

Pos logit contributions
talk+0.324
oes+0.322
ted+0.302
rost+0.295
isk+0.295
Neg logit contributions
 best-0.313
 adorable-0.312
 important-0.303
 idea-0.300
 beautiful-0.292
Token the
Token ID262
Feature activation-4.83e-02

Pos logit contributions
oes+0.452
talk+0.450
ted+0.443
rost+0.406
ricanes+0.398
Neg logit contributions
 adorable-0.470
 best-0.465
 idea-0.461
 important-0.454
 beautiful-0.442
Token top
Token ID1353
Feature activation-3.44e-02

Pos logit contributions
oes+0.081
talk+0.079
ted+0.076
stack+0.074
rost+0.072
Neg logit contributions
 adorable-0.093
 best-0.090
 idea-0.090
 important-0.089
 precious-0.087
Token.
Token ID13
Feature activation+8.87e-03

Pos logit contributions
talk+0.082
oes+0.080
ted+0.080
rost+0.077
stack+0.076
Neg logit contributions
 adorable-0.042
 best-0.042
 idea-0.041
 important-0.040
 beautiful-0.038
Token After
Token ID2293
Feature activation-1.96e-02

Pos logit contributions
 adorable+0.017
 best+0.017
 idea+0.016
 important+0.016
 loyal+0.016
Neg logit contributions
talk-0.015
oes-0.015
ted-0.014
rost-0.013
use-0.013
Token a
Token ID257
Feature activation+1.37e-01

Pos logit contributions
oes+0.038
talk+0.037
ted+0.035
rost+0.033
stack+0.033
Neg logit contributions
 adorable-0.035
 best-0.034
 important-0.033
 idea-0.033
 precious-0.032
Token use
Token ID779
Feature activation-2.94e-03

Pos logit contributions
oes+0.092
talk+0.091
ted+0.091
stack+0.082
rost+0.082
Neg logit contributions
 adorable-0.120
 best-0.118
 idea-0.118
 important-0.116
 precious-0.112
Token it
Token ID340
Feature activation-9.54e-02

Pos logit contributions
oes+0.005
talk+0.005
ted+0.004
stack+0.004
rost+0.004
Neg logit contributions
 adorable-0.006
 best-0.006
 important-0.006
 idea-0.006
 precious-0.006
Token to
Token ID284
Feature activation-2.06e-01

Pos logit contributions
talk+0.177
oes+0.174
ted+0.170
stack+0.163
rost+0.163
Neg logit contributions
 best-0.178
 important-0.171
 adorable-0.169
 idea-0.168
 precious-0.159
Token climb
Token ID12080
Feature activation-1.67e-01

Pos logit contributions
ted+0.331
oes+0.327
talk+0.325
ting+0.290
rost+0.290
Neg logit contributions
 idea-0.424
 adorable-0.420
 best-0.414
 important-0.406
 beautiful-0.402
Token up
Token ID510
Feature activation-1.24e-01

Pos logit contributions
talk+0.317
oes+0.313
isk+0.297
ted+0.296
rost+0.286
Neg logit contributions
 best-0.296
 adorable-0.291
 important-0.284
 idea-0.281
 beautiful-0.276
Token to
Token ID284
Feature activation-2.53e-01

Pos logit contributions
oes+0.240
talk+0.237
ted+0.230
stack+0.218
rost+0.217
Neg logit contributions
 adorable-0.210
 best-0.205
 idea-0.203
 important-0.202
 precious-0.195
Token the
Token ID262
Feature activation-6.42e-02

Pos logit contributions
talk+0.503
oes+0.501
ted+0.496
isk+0.455
ricanes+0.453
Neg logit contributions
 best-0.420
 adorable-0.420
 idea-0.414
 beautiful-0.408
 important-0.393
Token balloon
Token ID21190
Feature activation-9.62e-02

Pos logit contributions
oes+0.101
talk+0.099
ted+0.094
stack+0.093
rost+0.088
Neg logit contributions
 adorable-0.131
 important-0.126
 best-0.126
 idea-0.125
 precious-0.122
Token.
Token ID13
Feature activation+3.16e-03

Pos logit contributions
talk+0.215
ted+0.214
oes+0.210
rost+0.204
isk+0.202
Neg logit contributions
 best-0.131
 idea-0.129
 adorable-0.127
 important-0.119
 beautiful-0.116
Token
Token ID198
Feature activation-3.41e-02

Pos logit contributions
 adorable+0.005
 best+0.005
 important+0.004
 idea+0.004
 loyal+0.004
Neg logit contributions
oes-0.007
talk-0.006
isk-0.006
ted-0.006
ting-0.006
Token
Token ID198
Feature activation-2.82e-02

Pos logit contributions
oes+0.065
talk+0.063
ted+0.062
stack+0.059
ting+0.058
Neg logit contributions
 adorable-0.060
 important-0.058
 best-0.057
 idea-0.057
 beautiful-0.055
Token.
Token ID13
Feature activation+4.76e-03

Pos logit contributions
 best+0.178
 adorable+0.174
 idea+0.168
 important+0.163
 beautiful+0.157
Neg logit contributions
talk-0.320
oes-0.317
stack-0.308
ted-0.307
isk-0.303
Token
Token ID198
Feature activation-3.22e-02

Pos logit contributions
 adorable+0.007
 best+0.007
 important+0.007
 loyal+0.007
 beautiful+0.007
Neg logit contributions
talk-0.011
oes-0.010
isk-0.010
ted-0.010
ales-0.010
Token
Token ID198
Feature activation-2.73e-02

Pos logit contributions
talk+0.060
oes+0.060
ted+0.059
stack+0.056
ting+0.055
Neg logit contributions
 adorable-0.057
 important-0.055
 best-0.055
 idea-0.053
 precious-0.052
Token"
Token ID1
Feature activation-2.04e-02

Pos logit contributions
oes+0.063
talk+0.062
ted+0.061
stack+0.058
isk+0.057
Neg logit contributions
 adorable-0.037
 best-0.036
 important-0.036
 precious-0.034
 loyal-0.034
TokenTime
Token ID7575
Feature activation-5.36e-02

Pos logit contributions
oes+0.031
talk+0.031
ted+0.029
stack+0.028
isk+0.027
Neg logit contributions
 adorable-0.043
 best-0.042
 important-0.041
 idea-0.041
 precious-0.040
Token to
Token ID284
Feature activation-2.53e-01

Pos logit contributions
talk+0.070
oes+0.069
stack+0.063
ted+0.063
isk+0.060
Neg logit contributions
 best-0.124
 adorable-0.123
 important-0.121
 idea-0.121
 precious-0.117
Token go
Token ID467
Feature activation-1.13e-01

Pos logit contributions
oes+0.395
talk+0.394
ted+0.387
stack+0.356
rost+0.349
Neg logit contributions
 idea-0.523
 adorable-0.520
 best-0.511
 beautiful-0.499
 important-0.491
Token home
Token ID1363
Feature activation-1.51e-01

Pos logit contributions
talk+0.267
oes+0.264
ted+0.260
isk+0.255
rost+0.255
Neg logit contributions
 best-0.139
 adorable-0.135
 idea-0.132
 important-0.126
 beautiful-0.124
Token,
Token ID11
Feature activation-7.63e-02

Pos logit contributions
oes+0.285
talk+0.282
stack+0.264
ted+0.262
rost+0.259
Neg logit contributions
 adorable-0.249
 important-0.235
 idea-0.232
 precious-0.230
 best-0.227
Token kids
Token ID3988
Feature activation-4.90e-02

Pos logit contributions
oes+0.127
ted+0.123
talk+0.121
rost+0.120
stack+0.117
Neg logit contributions
 important-0.140
 adorable-0.138
 idea-0.136
 prized-0.133
 best-0.133
Token,"
Token ID553
Feature activation-8.10e-03

Pos logit contributions
oes+0.118
talk+0.117
ted+0.114
stack+0.110
rost+0.109
Neg logit contributions
 adorable-0.060
 best-0.057
 important-0.056
 idea-0.056
 precious-0.053
Token and
Token ID290
Feature activation-6.69e-02

Pos logit contributions
talk+0.066
oes+0.066
ted+0.063
stack+0.062
rost+0.061
Neg logit contributions
 adorable-0.076
 best-0.074
 important-0.073
 idea-0.072
 loyal-0.070
Token said
Token ID531
Feature activation+1.02e-01

Pos logit contributions
oes+0.131
stack+0.124
talk+0.117
rost+0.116
isk+0.112
Neg logit contributions
 adorable-0.120
 precious-0.113
 beautiful-0.113
 loyal-0.112
 intelligent-0.111
Token,
Token ID11
Feature activation-1.09e-02

Pos logit contributions
 adorable+0.126
 best+0.125
 important+0.120
 idea+0.117
 precious+0.111
Neg logit contributions
oes-0.249
talk-0.246
ted-0.237
stack-0.234
isk-0.232
Token "
Token ID366
Feature activation-3.38e-02

Pos logit contributions
talk+0.028
oes+0.027
ales+0.026
ted+0.025
aled+0.025
Neg logit contributions
 adorable-0.013
 loyal-0.013
 important-0.012
 best-0.012
 generous-0.012
TokenTime
Token ID7575
Feature activation-5.47e-02

Pos logit contributions
stack+0.060
oes+0.057
rost+0.050
 gra+0.050
talk+0.049
Neg logit contributions
 adorable-0.061
 loyal-0.058
 popular-0.057
 important-0.057
 beautiful-0.057
Token to
Token ID284
Feature activation-2.52e-01

Pos logit contributions
talk+0.070
oes+0.070
stack+0.063
ted+0.063
rost+0.061
Neg logit contributions
 best-0.128
 adorable-0.127
 important-0.126
 idea-0.125
 precious-0.121
Token go
Token ID467
Feature activation-1.10e-01

Pos logit contributions
oes+0.377
talk+0.372
ted+0.362
stack+0.332
rost+0.327
Neg logit contributions
 adorable-0.545
 idea-0.540
 best-0.532
 important-0.523
 beautiful-0.513
Token,
Token ID11
Feature activation-9.97e-02

Pos logit contributions
talk+0.243
ted+0.237
oes+0.237
rost+0.235
stack+0.231
Neg logit contributions
 best-0.153
 adorable-0.151
 idea-0.147
 important-0.145
 beautiful-0.139
Token kids
Token ID3988
Feature activation-6.68e-02

Pos logit contributions
isk+0.192
ted+0.191
ting+0.191
oes+0.191
talk+0.189
Neg logit contributions
 important-0.144
 prized-0.143
 adorable-0.141
 best-0.141
 precious-0.139
Token.
Token ID13
Feature activation-1.43e-02

Pos logit contributions
oes+0.167
talk+0.165
ted+0.161
stack+0.157
rost+0.155
Neg logit contributions
 adorable-0.074
 best-0.073
 important-0.070
 idea-0.070
 precious-0.065
Token The
Token ID383
Feature activation+2.85e-02

Pos logit contributions
oes+0.031
talk+0.031
stack+0.029
isk+0.029
ted+0.029
Neg logit contributions
 adorable-0.019
 best-0.018
 important-0.017
 loyal-0.017
 precious-0.016
Token to
Token ID284
Feature activation-2.18e-01

Pos logit contributions
oes+0.053
talk+0.053
ted+0.050
isk+0.049
stack+0.049
Neg logit contributions
 adorable-0.090
 best-0.089
 important-0.088
 idea-0.087
 beautiful-0.085
Token the
Token ID262
Feature activation-4.46e-02

Pos logit contributions
oes+0.373
talk+0.371
ted+0.355
isk+0.337
stack+0.335
Neg logit contributions
 adorable-0.426
 best-0.416
 idea-0.416
 important-0.403
 beautiful-0.393
Token hill
Token ID12788
Feature activation-1.46e-01

Pos logit contributions
oes+0.059
talk+0.056
ted+0.055
stack+0.054
rost+0.050
Neg logit contributions
 adorable-0.101
 important-0.098
 idea-0.098
 best-0.097
 precious-0.096
Token and
Token ID290
Feature activation-8.98e-02

Pos logit contributions
oes+0.312
talk+0.308
ted+0.299
stack+0.288
rost+0.284
Neg logit contributions
 adorable-0.217
 best-0.209
 important-0.207
 idea-0.206
 precious-0.199
Token climbed
Token ID19952
Feature activation-1.46e-01

Pos logit contributions
oes+0.160
talk+0.157
rost+0.155
isk+0.155
use+0.147
Neg logit contributions
 adorable-0.166
 best-0.159
 beautiful-0.155
 loyal-0.152
 idea-0.151
Token to
Token ID284
Feature activation-2.52e-01

Pos logit contributions
talk+0.256
oes+0.254
ted+0.241
isk+0.236
rost+0.233
Neg logit contributions
 adorable-0.272
 best-0.271
 idea-0.264
 important-0.262
 precious-0.253
Token the
Token ID262
Feature activation-7.03e-02

Pos logit contributions
talk+0.427
oes+0.423
ted+0.413
isk+0.388
rost+0.387
Neg logit contributions
 adorable-0.492
 best-0.485
 idea-0.482
 important-0.466
 beautiful-0.465
Token top
Token ID1353
Feature activation-4.74e-02

Pos logit contributions
oes+0.110
talk+0.105
ted+0.103
stack+0.102
rost+0.095
Neg logit contributions
 adorable-0.143
 important-0.138
 idea-0.137
 best-0.136
 precious-0.134
Token.
Token ID13
Feature activation-4.62e-03

Pos logit contributions
oes+0.099
talk+0.098
ted+0.095
stack+0.091
rost+0.090
Neg logit contributions
 adorable-0.072
 best-0.070
 important-0.069
 idea-0.068
 precious-0.067
Token They
Token ID1119
Feature activation-7.48e-02

Pos logit contributions
talk+0.009
use+0.009
isk+0.009
oes+0.008
x+0.008
Neg logit contributions
 idea-0.008
 adorable-0.008
 best-0.008
 thrilled-0.007
 excited-0.007
Token looked
Token ID3114
Feature activation+1.02e-01

Pos logit contributions
oes+0.115
talk+0.113
ted+0.108
rost+0.102
stack+0.102
Neg logit contributions
 adorable-0.157
 best-0.154
 idea-0.151
 important-0.150
 precious-0.146
TokenOnce
Token ID7454
Feature activation-7.09e-02

Pos logit contributions
oes+0.265
talk+0.259
ted+0.244
stack+0.242
ricanes+0.239
Neg logit contributions
 adorable-0.216
 best-0.215
 idea-0.205
 important-0.204
 beautiful-0.200
Token upon
Token ID2402
Feature activation-4.51e-02

Pos logit contributions
oes+0.111
talk+0.106
ted+0.103
rost+0.098
stack+0.097
Neg logit contributions
 adorable-0.145
 best-0.141
 important-0.140
 idea-0.139
 precious-0.137
Token a
Token ID257
Feature activation+1.15e-01

Pos logit contributions
oes+0.089
talk+0.089
ted+0.086
rost+0.084
 gra+0.083
Neg logit contributions
 best-0.066
 idea-0.064
 precious-0.064
 adorable-0.064
 important-0.064
Token time
Token ID640
Feature activation-6.64e-02

Pos logit contributions
 adorable+0.181
 best+0.175
 precious+0.173
 important+0.171
 idea+0.168
Neg logit contributions
oes-0.218
talk-0.215
rost-0.213
isk-0.210
ted-0.200
Token,
Token ID11
Feature activation-5.75e-02

Pos logit contributions
oes+0.132
ted+0.126
talk+0.124
rost+0.122
stack+0.121
Neg logit contributions
 adorable-0.102
 important-0.101
 idea-0.099
 precious-0.099
 best-0.095
Token a
Token ID257
Feature activation+1.57e-01

Pos logit contributions
oes+0.087
ted+0.083
talk+0.082
stack+0.080
rost+0.079
Neg logit contributions
 adorable-0.114
 important-0.112
 idea-0.112
 best-0.111
 precious-0.110
Token little
Token ID1310
Feature activation+1.17e-01

Pos logit contributions
 idea+0.140
 best+0.134
 adorable+0.134
 important+0.132
 precious+0.127
Neg logit contributions
oes-0.389
rost-0.382
Dog-0.375
talk-0.374
isk-0.370
Token girl
Token ID2576
Feature activation+5.78e-02

Pos logit contributions
 important+0.174
 idea+0.172
 adorable+0.170
 best+0.167
 precious+0.165
Neg logit contributions
oes-0.242
ted-0.233
talk-0.230
Dog-0.220
rost-0.219
Token named
Token ID3706
Feature activation+9.92e-03

Pos logit contributions
 adorable+0.113
 best+0.110
 important+0.108
 idea+0.107
 precious+0.104
Neg logit contributions
oes-0.097
talk-0.096
ted-0.092
stack-0.088
rost-0.087
Token Lily
Token ID20037
Feature activation+7.91e-02

Pos logit contributions
 adorable+0.018
 best+0.018
 important+0.017
 idea+0.017
 beautiful+0.017
Neg logit contributions
talk-0.018
oes-0.018
ted-0.017
stack-0.017
rost-0.016
Token went
Token ID1816
Feature activation-6.48e-02

Pos logit contributions
 adorable+0.166
 best+0.162
 important+0.160
 idea+0.160
 precious+0.155
Neg logit contributions
oes-0.121
talk-0.119
ted-0.115
stack-0.109
rost-0.107
Token,
Token ID11
Feature activation-4.02e-02

Pos logit contributions
oes+0.144
talk+0.140
ted+0.137
stack+0.133
isk+0.130
Neg logit contributions
 adorable-0.103
 important-0.096
 precious-0.096
 idea-0.095
 best-0.093
Token the
Token ID262
Feature activation+3.94e-02

Pos logit contributions
oes+0.064
talk+0.063
ted+0.059
stack+0.056
rost+0.056
Neg logit contributions
 adorable-0.083
 best-0.082
 important-0.080
 idea-0.080
 precious-0.078
Token daughter
Token ID4957
Feature activation+5.12e-02

Pos logit contributions
 adorable+0.049
 idea+0.048
 impressed+0.048
 prized+0.047
 important+0.046
Neg logit contributions
ted-0.080
oes-0.078
talk-0.077
ting-0.077
stack-0.075
Token saw
Token ID2497
Feature activation+1.94e-02

Pos logit contributions
 adorable+0.099
 best+0.096
 idea+0.096
 important+0.096
 precious+0.094
Neg logit contributions
talk-0.084
oes-0.082
ted-0.082
stack-0.076
isk-0.075
Token a
Token ID257
Feature activation+1.78e-01

Pos logit contributions
 adorable+0.031
 best+0.030
 important+0.030
 idea+0.030
 precious+0.029
Neg logit contributions
oes-0.039
talk-0.039
ted-0.038
stack-0.036
rost-0.036
Token pretty
Token ID2495
Feature activation+1.96e-01

Pos logit contributions
 adorable+0.238
 idea+0.225
 best+0.224
 important+0.221
 precious+0.208
Neg logit contributions
stack-0.333
 gra-0.329
oes-0.326
ricane-0.325
 hike-0.321
Token dress
Token ID6576
Feature activation+3.50e-02

Pos logit contributions
 best+0.337
 important+0.328
 adorable+0.326
 pleased+0.320
 precious+0.319
Neg logit contributions
ted-0.309
stack-0.308
oes-0.294
 gra-0.289
 benches-0.287
Token.
Token ID13
Feature activation+2.28e-02

Pos logit contributions
 adorable+0.060
 best+0.058
 important+0.058
 idea+0.057
 precious+0.056
Neg logit contributions
oes-0.066
talk-0.065
ted-0.063
stack-0.060
isk-0.059
Token She
Token ID1375
Feature activation-2.35e-02

Pos logit contributions
 adorable+0.044
 idea+0.041
 best+0.041
 important+0.041
 beautiful+0.040
Neg logit contributions
talk-0.041
oes-0.040
ted-0.037
rost-0.036
use-0.036
Token asked
Token ID1965
Feature activation-3.65e-03

Pos logit contributions
ted+0.034
talk+0.033
oes+0.032
isk+0.030
stack+0.029
Neg logit contributions
 best-0.049
 idea-0.048
 adorable-0.047
 important-0.047
 precious-0.045
Token her
Token ID607
Feature activation+3.64e-02

Pos logit contributions
oes+0.006
talk+0.006
ted+0.006
stack+0.006
rost+0.006
Neg logit contributions
 adorable-0.007
 best-0.007
 important-0.007
 idea-0.007
 precious-0.007
Token a
Token ID257
Feature activation+1.14e-01

Pos logit contributions
talk+0.060
oes+0.060
ted+0.059
rost+0.057
iffs+0.056
Neg logit contributions
 best-0.061
 idea-0.060
 precious-0.059
 important-0.059
 prized-0.058
Token time
Token ID640
Feature activation-6.39e-02

Pos logit contributions
 adorable+0.185
 precious+0.180
 important+0.177
 best+0.175
 idea+0.169
Neg logit contributions
oes-0.208
rost-0.203
talk-0.202
isk-0.199
use-0.190
Token,
Token ID11
Feature activation-5.56e-02

Pos logit contributions
oes+0.118
ted+0.110
iffs+0.107
talk+0.107
rost+0.106
Neg logit contributions
 important-0.106
 adorable-0.106
 precious-0.104
 idea-0.103
 valuable-0.100
Token there
Token ID612
Feature activation-1.45e-01

Pos logit contributions
oes+0.085
talk+0.079
ted+0.079
rost+0.077
stack+0.075
Neg logit contributions
 adorable-0.110
 idea-0.108
 important-0.107
 best-0.106
 precious-0.106
Token was
Token ID373
Feature activation+1.04e-01

Pos logit contributions
oes+0.261
ted+0.254
talk+0.250
rost+0.241
stack+0.239
Neg logit contributions
 adorable-0.250
 important-0.243
 precious-0.241
 best-0.238
 idea-0.238
Token a
Token ID257
Feature activation+1.76e-01

Pos logit contributions
 precious+0.175
 idea+0.172
 pleased+0.171
 prized+0.170
 important+0.167
Neg logit contributions
ted-0.168
oes-0.153
rost-0.152
 gra-0.151
talk-0.150
Token little
Token ID1310
Feature activation+1.32e-01

Pos logit contributions
 idea+0.198
 best+0.185
 important+0.185
 mistakes+0.179
 prized+0.177
Neg logit contributions
Dog-0.380
 gra-0.368
rost-0.367
oes-0.364
isk-0.352
Token boy
Token ID2933
Feature activation+5.99e-02

Pos logit contributions
 important+0.217
 idea+0.212
 prized+0.210
 thrilled+0.210
 pleased+0.208
Neg logit contributions
oes-0.234
ted-0.226
Dog-0.221
ting-0.216
talk-0.214
Token named
Token ID3706
Feature activation+2.19e-02

Pos logit contributions
 adorable+0.121
 best+0.118
 important+0.117
 idea+0.116
 precious+0.113
Neg logit contributions
oes-0.096
talk-0.095
ted-0.091
stack-0.087
rost-0.085
Token Tim
Token ID5045
Feature activation+4.96e-02

Pos logit contributions
 adorable+0.038
 best+0.035
 important+0.034
 beautiful+0.034
 loyal+0.033
Neg logit contributions
talk-0.047
ted-0.043
stack-0.043
ting-0.041
oes-0.041
Tokenmy
Token ID1820
Feature activation+4.72e-02

Pos logit contributions
 adorable+0.100
 best+0.097
 important+0.096
 idea+0.096
 precious+0.093
Neg logit contributions
oes-0.080
talk-0.079
ted-0.076
stack-0.072
rost-0.071
Token right
Token ID826
Feature activation+4.95e-02

Pos logit contributions
ted+0.089
oes+0.088
talk+0.086
rost+0.084
stack+0.081
Neg logit contributions
 best-0.089
 important-0.088
 idea-0.087
 adorable-0.084
 precious-0.080
Token away
Token ID1497
Feature activation-5.34e-02

Pos logit contributions
 important+0.098
 idea+0.097
 best+0.097
 adorable+0.090
 precious+0.087
Neg logit contributions
 gra-0.082
ted-0.082
Dog-0.082
oes-0.082
talk-0.080
Token.
Token ID13
Feature activation+6.88e-03

Pos logit contributions
talk+0.120
oes+0.120
ted+0.116
rost+0.111
stack+0.110
Neg logit contributions
 best-0.074
 adorable-0.074
 important-0.073
 idea-0.070
 precious-0.067
Token Lucy
Token ID22162
Feature activation+4.83e-02

Pos logit contributions
 adorable+0.013
 important+0.012
 idea+0.012
 best+0.012
 loyal+0.012
Neg logit contributions
oes-0.013
talk-0.012
ted-0.012
ricanes-0.011
use-0.011
Token was
Token ID373
Feature activation+1.38e-01

Pos logit contributions
 adorable+0.080
 best+0.078
 idea+0.077
 important+0.077
 precious+0.074
Neg logit contributions
talk-0.093
oes-0.093
ted-0.091
stack-0.086
isk-0.085
Token scared
Token ID12008
Feature activation+1.58e-01

Pos logit contributions
 idea+0.149
 adorable+0.147
 best+0.143
 precious+0.138
 important+0.135
Neg logit contributions
talk-0.317
ted-0.310
oes-0.309
isk-0.306
stack-0.305
Token and
Token ID290
Feature activation-2.40e-02

Pos logit contributions
 adorable+0.227
 best+0.219
 important+0.216
 idea+0.215
 precious+0.207
Neg logit contributions
oes-0.346
talk-0.343
ted-0.333
stack-0.321
rost-0.317
Token began
Token ID2540
Feature activation+1.07e-02

Pos logit contributions
oes+0.055
talk+0.053
iffs+0.050
rost+0.049
ted+0.048
Neg logit contributions
 adorable-0.037
 important-0.036
 best-0.034
 idea-0.034
 loyal-0.033
Token to
Token ID284
Feature activation-1.90e-01

Pos logit contributions
 important+0.023
 best+0.023
 adorable+0.023
 loyal+0.022
 idea+0.022
Neg logit contributions
oes-0.017
talk-0.016
rost-0.015
ted-0.015
stack-0.015
Token cry
Token ID3960
Feature activation+5.37e-02

Pos logit contributions
oes+0.402
talk+0.397
ted+0.393
rost+0.369
iffs+0.368
Neg logit contributions
 idea-0.299
 adorable-0.299
 best-0.287
 important-0.284
 beautiful-0.271
Token,
Token ID11
Feature activation-2.86e-02

Pos logit contributions
 adorable+0.083
 best+0.081
 important+0.080
 idea+0.079
 precious+0.076
Neg logit contributions
oes-0.112
talk-0.111
ted-0.108
stack-0.104
rost-0.102
Token One
Token ID1881
Feature activation+1.44e-02

Pos logit contributions
talk+0.012
oes+0.011
ted+0.011
stack+0.011
isk+0.011
Neg logit contributions
 adorable-0.013
 best-0.013
 idea-0.012
 important-0.012
 beautiful-0.011
Token day
Token ID1110
Feature activation-6.93e-02

Pos logit contributions
 adorable+0.026
 precious+0.026
 best+0.025
 prized+0.025
 important+0.025
Neg logit contributions
oes-0.026
talk-0.023
rost-0.022
ted-0.022
stack-0.021
Token,
Token ID11
Feature activation-4.50e-02

Pos logit contributions
oes+0.146
ted+0.145
talk+0.143
rost+0.135
iffs+0.133
Neg logit contributions
 adorable-0.102
 important-0.097
 idea-0.097
 precious-0.095
 best-0.094
Token they
Token ID484
Feature activation-8.01e-02

Pos logit contributions
oes+0.056
talk+0.055
ted+0.054
stack+0.048
rost+0.048
Neg logit contributions
 adorable-0.106
 best-0.103
 important-0.103
 idea-0.103
 precious-0.101
Token find
Token ID1064
Feature activation-1.12e-02

Pos logit contributions
oes+0.138
ted+0.136
talk+0.134
stack+0.127
isk+0.125
Neg logit contributions
 adorable-0.151
 idea-0.150
 best-0.149
 important-0.145
 precious-0.140
Token a
Token ID257
Feature activation+1.63e-01

Pos logit contributions
talk+0.024
ted+0.024
oes+0.024
rost+0.022
stack+0.022
Neg logit contributions
 adorable-0.016
 idea-0.015
 important-0.015
 precious-0.014
 best-0.014
Token shiny
Token ID22441
Feature activation+1.83e-01

Pos logit contributions
 adorable+0.185
 idea+0.173
 best+0.169
 important+0.169
 precious+0.159
Neg logit contributions
oes-0.387
talk-0.367
stack-0.364
rost-0.359
ted-0.358
Token coin
Token ID10752
Feature activation+2.21e-02

Pos logit contributions
 pleased+0.312
 impressed+0.310
 thrilled+0.308
 adorable+0.305
 delighted+0.292
Neg logit contributions
oes-0.321
ted-0.304
stack-0.302
talk-0.291
rost-0.290
Token under
Token ID739
Feature activation-1.07e-01

Pos logit contributions
 adorable+0.044
 best+0.043
 important+0.043
 idea+0.043
 precious+0.041
Neg logit contributions
oes-0.036
talk-0.035
ted-0.034
stack-0.032
rost-0.032
Token the
Token ID262
Feature activation-4.97e-02

Pos logit contributions
oes+0.229
talk+0.227
ted+0.220
stack+0.211
rost+0.209
Neg logit contributions
 adorable-0.162
 best-0.156
 important-0.154
 idea-0.153
 precious-0.148
Token bed
Token ID3996
Feature activation-1.06e-01

Pos logit contributions
talk+0.094
oes+0.092
ted+0.091
isk+0.088
rost+0.086
Neg logit contributions
 adorable-0.088
 best-0.086
 beautiful-0.083
 idea-0.082
 important-0.082
Token the
Token ID262
Feature activation+2.32e-02

Pos logit contributions
oes+0.102
talk+0.097
ted+0.090
stack+0.090
rost+0.089
Neg logit contributions
 adorable-0.100
 best-0.094
 important-0.094
 idea-0.094
 precious-0.092
Token mess
Token ID2085
Feature activation-8.67e-03

Pos logit contributions
 adorable+0.040
 best+0.039
 important+0.038
 idea+0.038
 precious+0.037
Neg logit contributions
oes-0.043
talk-0.043
ted-0.041
stack-0.040
rost-0.039
Token and
Token ID290
Feature activation-6.25e-02

Pos logit contributions
oes+0.017
talk+0.017
ted+0.016
stack+0.016
rost+0.015
Neg logit contributions
 adorable-0.014
 best-0.014
 important-0.014
 idea-0.014
 precious-0.013
Token the
Token ID262
Feature activation+1.50e-02

Pos logit contributions
oes+0.128
talk+0.124
use+0.114
rost+0.113
iffs+0.111
Neg logit contributions
 adorable-0.107
 important-0.101
 best-0.101
 idea-0.100
 beautiful-0.099
Token blood
Token ID2910
Feature activation-3.49e-02

Pos logit contributions
 adorable+0.025
 best+0.024
 important+0.024
 idea+0.024
 precious+0.023
Neg logit contributions
oes-0.029
talk-0.028
ted-0.027
stack-0.027
rost-0.026
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
oes+0.077
talk+0.076
stack+0.071
rost+0.071
ricanes+0.069
Neg logit contributions
 adorable-0.052
 best-0.052
 important-0.050
 idea-0.050
 beautiful-0.047
Token She
Token ID1375
Feature activation-3.97e-02

Pos logit contributions
 adorable+0.023
 important+0.022
 idea+0.022
 best+0.022
 precious+0.022
Neg logit contributions
oes-0.021
talk-0.020
ted-0.019
isk-0.019
rost-0.018
Token was
Token ID373
Feature activation+1.35e-01

Pos logit contributions
talk+0.074
oes+0.074
ted+0.073
isk+0.068
stack+0.068
Neg logit contributions
 adorable-0.068
 best-0.067
 idea-0.066
 important-0.066
 precious-0.063
Token very
Token ID845
Feature activation+1.69e-01

Pos logit contributions
 idea+0.158
 adorable+0.156
 best+0.153
 important+0.148
 precious+0.143
Neg logit contributions
talk-0.307
oes-0.304
ted-0.302
stack-0.295
isk-0.294
Token angry
Token ID7954
Feature activation+1.61e-01

Pos logit contributions
 best+0.200
 adorable+0.198
 idea+0.197
 prized+0.191
 important+0.181
Neg logit contributions
talk-0.347
oes-0.342
ted-0.334
rost-0.332
iffs-0.332
Token and
Token ID290
Feature activation-2.70e-02

Pos logit contributions
 adorable+0.237
 best+0.229
 important+0.226
 idea+0.225
 precious+0.217
Neg logit contributions
oes-0.346
talk-0.343
ted-0.334
stack-0.318
rost-0.316
Token she
Token ID673
Feature activation-3.75e-02

Pos logit contributions
 adorable+0.115
 best+0.111
 idea+0.111
 important+0.110
 precious+0.107
Neg logit contributions
oes-0.103
talk-0.101
ted-0.099
stack-0.093
rost-0.092
Token wouldn
Token ID3636
Feature activation-1.64e-02

Pos logit contributions
oes+0.044
talk+0.044
ted+0.042
stack+0.038
rost+0.037
Neg logit contributions
 adorable-0.090
 idea-0.088
 best-0.088
 important-0.087
 precious-0.085
Token't
Token ID470
Feature activation+2.58e-02

Pos logit contributions
oes+0.024
talk+0.024
ted+0.023
stack+0.021
rost+0.021
Neg logit contributions
 adorable-0.035
 best-0.034
 important-0.034
 idea-0.034
 precious-0.033
Token get
Token ID651
Feature activation+1.54e-02

Pos logit contributions
 adorable+0.052
 best+0.051
 idea+0.051
 important+0.051
 beautiful+0.048
Neg logit contributions
ted-0.042
talk-0.041
oes-0.041
stack-0.038
iffs-0.038
Token hurt
Token ID5938
Feature activation+5.07e-02

Pos logit contributions
 adorable+0.031
 best+0.030
 important+0.030
 precious+0.029
 idea+0.029
Neg logit contributions
oes-0.026
talk-0.025
ted-0.024
stack-0.024
isk-0.023
Token.
Token ID13
Feature activation+8.57e-03

Pos logit contributions
 adorable+0.086
 best+0.083
 idea+0.083
 important+0.083
 precious+0.080
Neg logit contributions
oes-0.097
talk-0.095
ted-0.093
stack-0.088
rost-0.087
Token
Token ID198
Feature activation-2.58e-02

Pos logit contributions
 adorable+0.019
 best+0.019
 important+0.018
 idea+0.018
 precious+0.018
Neg logit contributions
oes-0.012
talk-0.011
ted-0.011
stack-0.010
rost-0.010
Token
Token ID198
Feature activation-2.11e-02

Pos logit contributions
talk+0.037
oes+0.037
ted+0.035
rost+0.033
stack+0.033
Neg logit contributions
 adorable-0.055
 best-0.054
 idea-0.054
 important-0.053
 precious-0.052
TokenWhen
Token ID2215
Feature activation-5.11e-02

Pos logit contributions
talk+0.031
oes+0.030
ted+0.028
rost+0.027
Dog+0.026
Neg logit contributions
 adorable-0.044
 best-0.044
 idea-0.043
 precious-0.041
 important-0.041
Token they
Token ID484
Feature activation-1.07e-01

Pos logit contributions
talk+0.075
oes+0.075
ted+0.074
stack+0.068
rost+0.066
Neg logit contributions
 adorable-0.106
 idea-0.101
 important-0.101
 best-0.101
 precious-0.101
Token got
Token ID1392
Feature activation+4.24e-02

Pos logit contributions
oes+0.202
ricanes+0.179
stack+0.178
talk+0.177
iffs+0.174
Neg logit contributions
 best-0.196
 adorable-0.193
 idea-0.192
 beautiful-0.183
 amazing-0.182
Token an
Token ID281
Feature activation+1.27e-01

Pos logit contributions
oes+0.073
talk+0.073
ted+0.070
stack+0.067
rost+0.066
Neg logit contributions
 adorable-0.076
 best-0.074
 important-0.073
 idea-0.073
 precious-0.070
Token important
Token ID1593
Feature activation+1.34e-01

Pos logit contributions
 best+0.244
 adorable+0.244
 precious+0.237
 pleased+0.235
 impressed+0.230
Neg logit contributions
oes-0.197
talk-0.189
rost-0.178
isk-0.171
 hike-0.166
Token report
Token ID989
Feature activation-4.11e-02

Pos logit contributions
 adorable+0.175
 best+0.164
 precious+0.157
 important+0.156
 idea+0.154
Neg logit contributions
oes-0.310
talk-0.309
ted-0.291
rost-0.287
stack-0.284
Token for
Token ID329
Feature activation-5.29e-02

Pos logit contributions
oes+0.083
talk+0.082
ted+0.080
stack+0.078
rost+0.077
Neg logit contributions
 adorable-0.066
 best-0.066
 important-0.065
 idea-0.064
 precious-0.060
Token her
Token ID607
Feature activation+1.68e-02

Pos logit contributions
oes+0.105
talk+0.104
ted+0.099
stack+0.096
iffs+0.094
Neg logit contributions
 adorable-0.092
 important-0.091
 best-0.090
 idea-0.086
 beautiful-0.085
Token boss
Token ID6478
Feature activation+4.98e-02

Pos logit contributions
 adorable+0.026
 best+0.025
 important+0.025
 idea+0.025
 precious+0.024
Neg logit contributions
oes-0.033
talk-0.032
ted-0.031
rost-0.030
isk-0.030
Token.
Token ID13
Feature activation+6.76e-03

Pos logit contributions
 adorable+0.077
 best+0.076
 idea+0.075
 important+0.075
 precious+0.070
Neg logit contributions
oes-0.100
talk-0.099
ted-0.097
stack-0.096
rost-0.093
Token She
Token ID1375
Feature activation-4.20e-02

Pos logit contributions
 adorable+0.013
 important+0.013
 best+0.013
 idea+0.013
 beautiful+0.012
Neg logit contributions
talk-0.012
oes-0.011
stack-0.011
ted-0.011
rost-0.010
Token did
Token ID750
Feature activation+4.29e-02

Pos logit contributions
talk+0.073
oes+0.073
ted+0.071
stack+0.066
rost+0.066
Neg logit contributions
 adorable-0.078
 best-0.076
 important-0.075
 idea-0.075
 precious-0.072
Token not
Token ID407
Feature activation+5.51e-02

Pos logit contributions
 adorable+0.064
 best+0.062
 important+0.061
 idea+0.061
 precious+0.058
Neg logit contributions
oes-0.092
talk-0.092
ted-0.089
stack-0.086
rost-0.084
Token like
Token ID588
Feature activation+5.45e-02

Pos logit contributions
 adorable+0.103
 idea+0.101
 best+0.100
 important+0.100
 precious+0.095
Neg logit contributions
oes-0.097
ted-0.095
talk-0.095
stack-0.091
rost-0.087
Token still
Token ID991
Feature activation+1.88e-02

Pos logit contributions
oes+0.080
talk+0.079
ted+0.077
stack+0.072
rost+0.072
Neg logit contributions
 adorable-0.089
 best-0.087
 important-0.086
 idea-0.086
 precious-0.083
Token sing
Token ID1702
Feature activation-3.05e-02

Pos logit contributions
 adorable+0.041
 best+0.040
 important+0.039
 idea+0.039
 precious+0.038
Neg logit contributions
oes-0.027
talk-0.027
ted-0.026
stack-0.024
rost-0.024
Token and
Token ID290
Feature activation-8.06e-02

Pos logit contributions
oes+0.057
talk+0.057
ted+0.056
rost+0.052
stack+0.052
Neg logit contributions
 adorable-0.051
 idea-0.049
 best-0.049
 important-0.048
 precious-0.047
Token be
Token ID307
Feature activation+5.54e-02

Pos logit contributions
oes+0.125
talk+0.124
ted+0.120
stack+0.113
rost+0.112
Neg logit contributions
 adorable-0.167
 best-0.164
 important-0.163
 idea-0.161
 precious-0.157
Token happy
Token ID3772
Feature activation+1.63e-01

Pos logit contributions
ummy+0.070
 idea+0.070
 prized+0.065
ement+0.062
 darling+0.060
Neg logit contributions
 away-0.107
 south-0.107
talk-0.105
isk-0.103
 them-0.103
Token."
Token ID526
Feature activation+1.46e-02

Pos logit contributions
 adorable+0.314
 idea+0.304
 best+0.303
 important+0.300
 precious+0.294
Neg logit contributions
oes-0.272
talk-0.265
ted-0.259
stack-0.243
rost-0.240
Token Lily
Token ID20037
Feature activation+1.10e-01

Pos logit contributions
 adorable+0.027
 best+0.026
 important+0.026
 idea+0.026
 precious+0.025
Neg logit contributions
oes-0.026
talk-0.026
ted-0.025
stack-0.024
rost-0.023
Token smiled
Token ID13541
Feature activation+1.64e-01

Pos logit contributions
 adorable+0.230
 important+0.226
 precious+0.225
 best+0.220
 idea+0.219
Neg logit contributions
ted-0.156
talk-0.153
isk-0.138
 walked-0.131
oes-0.128
Token and
Token ID290
Feature activation-2.79e-02

Pos logit contributions
 important+0.320
 idea+0.318
 adorable+0.315
 precious+0.311
 prized+0.309
Neg logit contributions
ted-0.231
talk-0.209
 towards-0.206
ting-0.205
oes-0.200
Token hugged
Token ID41130
Feature activation+4.63e-02

Pos logit contributions
ted+0.045
talk+0.042
oes+0.040
ting+0.038
stack+0.038
Neg logit contributions
 adorable-0.055
 idea-0.054
 important-0.054
 best-0.054
 precious-0.052
Token her
Token ID607
Feature activation+5.42e-02

Pos logit contributions
 best+0.074
 adorable+0.074
 idea+0.071
 important+0.071
 beautiful+0.067
Neg logit contributions
oes-0.099
talk-0.097
stack-0.092
use-0.089
ricanes-0.088
Token But
Token ID887
Feature activation-1.47e-03

Pos logit contributions
 adorable+0.028
 best+0.026
 important+0.026
 idea+0.026
 beautiful+0.025
Neg logit contributions
talk-0.030
oes-0.030
ted-0.028
rost-0.028
use-0.028
Token the
Token ID262
Feature activation+4.81e-02

Pos logit contributions
talk+0.003
oes+0.003
ted+0.003
isk+0.003
iffs+0.003
Neg logit contributions
 adorable-0.002
 best-0.002
 important-0.002
 idea-0.002
 precious-0.002
Token tree
Token ID5509
Feature activation-9.48e-02

Pos logit contributions
 adorable+0.089
 best+0.087
 important+0.086
 idea+0.086
 precious+0.083
Neg logit contributions
oes-0.085
talk-0.084
ted-0.081
stack-0.078
rost-0.076
Token was
Token ID373
Feature activation+1.41e-01

Pos logit contributions
oes+0.187
talk+0.182
ted+0.180
stack+0.173
isk+0.169
Neg logit contributions
 adorable-0.154
 best-0.148
 idea-0.147
 important-0.146
 precious-0.142
Token too
Token ID1165
Feature activation+9.70e-02

Pos logit contributions
 adorable+0.184
 idea+0.182
 best+0.179
 important+0.170
 precious+0.166
Neg logit contributions
oes-0.312
talk-0.298
ted-0.297
stack-0.292
rost-0.290
Token tall
Token ID7331
Feature activation+1.15e-01

Pos logit contributions
 best+0.161
 idea+0.158
 adorable+0.155
 important+0.144
 prized+0.140
Neg logit contributions
oes-0.184
talk-0.169
iffs-0.168
rost-0.167
ted-0.167
Token and
Token ID290
Feature activation-2.61e-02

Pos logit contributions
 adorable+0.191
 best+0.185
 idea+0.183
 important+0.182
 precious+0.177
Neg logit contributions
oes-0.224
talk-0.221
ted-0.218
stack-0.206
isk-0.203
Token Twe
Token ID24205
Feature activation+1.04e-01

Pos logit contributions
talk+0.050
oes+0.050
ted+0.048
stack+0.046
rost+0.045
Neg logit contributions
 adorable-0.045
 best-0.044
 important-0.044
 idea-0.042
 precious-0.042
Tokenety
Token ID2963
Feature activation+2.85e-02

Pos logit contributions
 adorable+0.258
 best+0.251
 important+0.250
 idea+0.246
 precious+0.245
Neg logit contributions
talk-0.115
oes-0.113
ted-0.110
stack-0.108
rost-0.100
Token couldn
Token ID3521
Feature activation+2.39e-02

Pos logit contributions
 adorable+0.052
 best+0.051
 idea+0.050
 important+0.050
 precious+0.049
Neg logit contributions
oes-0.050
talk-0.049
ted-0.049
stack-0.046
isk-0.045
Token't
Token ID470
Feature activation+3.75e-02

Pos logit contributions
 adorable+0.056
 best+0.055
 idea+0.055
 important+0.054
 precious+0.054
Neg logit contributions
oes-0.030
talk-0.029
ted-0.028
stack-0.027
isk-0.026
Token not
Token ID407
Feature activation+5.18e-02

Pos logit contributions
 adorable+0.062
 best+0.060
 idea+0.059
 important+0.059
 precious+0.058
Neg logit contributions
oes-0.072
talk-0.072
ted-0.069
rost-0.066
isk-0.066
Token want
Token ID765
Feature activation+6.34e-02

Pos logit contributions
 adorable+0.104
 best+0.102
 idea+0.101
 important+0.101
 precious+0.098
Neg logit contributions
oes-0.084
talk-0.083
ted-0.081
stack-0.077
rost-0.074
Token to
Token ID284
Feature activation-1.88e-01

Pos logit contributions
 important+0.134
 best+0.134
 adorable+0.133
 beautiful+0.128
 precious+0.126
Neg logit contributions
stack-0.104
talk-0.095
oes-0.093
isk-0.093
Dog-0.089
Token share
Token ID2648
Feature activation-5.68e-02

Pos logit contributions
oes+0.284
talk+0.280
ted+0.276
stack+0.264
rost+0.253
Neg logit contributions
 adorable-0.399
 best-0.389
 idea-0.386
 important-0.385
 precious-0.372
Token her
Token ID607
Feature activation-2.27e-03

Pos logit contributions
oes+0.109
talk+0.108
ted+0.104
stack+0.101
rost+0.098
Neg logit contributions
 adorable-0.098
 best-0.096
 important-0.094
 idea-0.094
 precious-0.091
Token bear
Token ID6842
Feature activation-8.79e-02

Pos logit contributions
oes+0.005
talk+0.005
ted+0.005
rost+0.004
isk+0.004
Neg logit contributions
 adorable-0.003
 important-0.003
 best-0.003
 idea-0.003
 impressed-0.003
Token.
Token ID13
Feature activation+8.28e-03

Pos logit contributions
talk+0.161
oes+0.159
ted+0.150
isk+0.150
rost+0.145
Neg logit contributions
 adorable-0.153
 best-0.148
 important-0.148
 idea-0.147
 precious-0.146
Token She
Token ID1375
Feature activation-4.34e-02

Pos logit contributions
 adorable+0.015
 important+0.014
 best+0.014
 idea+0.014
 loyal+0.014
Neg logit contributions
talk-0.016
oes-0.015
ted-0.015
stack-0.014
use-0.014
Token thought
Token ID1807
Feature activation+1.37e-01

Pos logit contributions
talk+0.076
ted+0.074
oes+0.073
isk+0.068
rost+0.067
Neg logit contributions
 adorable-0.081
 best-0.079
 idea-0.078
 important-0.078
 precious-0.075
Token he
Token ID339
Feature activation-3.16e-02

Pos logit contributions
 adorable+0.284
 best+0.278
 important+0.276
 idea+0.273
 precious+0.267
Neg logit contributions
oes-0.212
talk-0.212
ted-0.201
stack-0.192
isk-0.187
Token was
Token ID373
Feature activation+1.41e-01

Pos logit contributions
talk+0.053
ted+0.052
oes+0.052
isk+0.050
rost+0.047
Neg logit contributions
 adorable-0.059
 idea-0.058
 best-0.057
 important-0.057
 precious-0.055
Token pretend
Token ID16614
Feature activation+1.01e-01

Pos logit contributions
oes+0.098
talk+0.096
ted+0.092
stack+0.087
rost+0.086
Neg logit contributions
 adorable-0.130
 best-0.127
 idea-0.126
 important-0.125
 precious-0.122
Token to
Token ID284
Feature activation-1.58e-01

Pos logit contributions
 adorable+0.224
 important+0.221
 best+0.219
 beautiful+0.214
 idea+0.213
Neg logit contributions
talk-0.155
oes-0.150
ted-0.140
isk-0.136
stack-0.134
Token eat
Token ID4483
Feature activation-2.61e-02

Pos logit contributions
oes+0.266
talk+0.263
ted+0.263
stack+0.239
ting+0.239
Neg logit contributions
 adorable-0.309
 idea-0.301
 best-0.298
 important-0.298
 precious-0.288
Token their
Token ID511
Feature activation+3.78e-03

Pos logit contributions
oes+0.048
talk+0.048
ted+0.046
stack+0.044
rost+0.043
Neg logit contributions
 adorable-0.047
 best-0.045
 important-0.045
 idea-0.045
 precious-0.043
Token stew
Token ID20798
Feature activation-1.04e-01

Pos logit contributions
 adorable+0.006
 important+0.006
 idea+0.006
 best+0.006
 precious+0.006
Neg logit contributions
oes-0.007
talk-0.007
ted-0.007
rost-0.006
 gra-0.006
Token and
Token ID290
Feature activation-6.58e-02

Pos logit contributions
oes+0.229
talk+0.228
ted+0.220
rost+0.211
stack+0.210
Neg logit contributions
 adorable-0.147
 best-0.147
 idea-0.143
 important-0.141
 precious-0.134
Token say
Token ID910
Feature activation+9.30e-02

Pos logit contributions
oes+0.114
talk+0.112
ted+0.108
isk+0.101
iffs+0.100
Neg logit contributions
 adorable-0.128
 best-0.125
 idea-0.125
 important-0.123
 precious-0.120
Token,
Token ID11
Feature activation-2.01e-02

Pos logit contributions
 adorable+0.133
 best+0.129
 important+0.126
 idea+0.126
 precious+0.121
Neg logit contributions
oes-0.205
talk-0.204
ted-0.198
stack-0.190
rost-0.188
Token "
Token ID366
Feature activation-3.61e-02

Pos logit contributions
talk+0.049
oes+0.047
ricane+0.043
use+0.043
ales+0.043
Neg logit contributions
 adorable-0.029
 impressed-0.027
 loyal-0.027
 pleased-0.027
 happy-0.027
TokenM
Token ID44
Feature activation+8.36e-02

Pos logit contributions
talk+0.061
oes+0.060
ted+0.058
rost+0.054
stack+0.054
Neg logit contributions
 adorable-0.071
 best-0.069
 idea-0.069
 important-0.069
 precious-0.067
Tokenmm
Token ID3020
Feature activation+1.23e-01

Pos logit contributions
 adorable+0.110
 important+0.100
 best+0.100
 idea+0.099
 precious+0.099
Neg logit contributions
oes-0.200
talk-0.193
rost-0.186
ted-0.186
isk-0.181
Token's
Token ID338
Feature activation-8.59e-03

Pos logit contributions
 adorable+0.012
 best+0.012
 important+0.012
 idea+0.011
 beautiful+0.011
Neg logit contributions
oes-0.022
talk-0.022
ted-0.021
rost-0.021
stack-0.021
Token bed
Token ID3996
Feature activation-1.09e-01

Pos logit contributions
oes+0.015
talk+0.014
ted+0.014
stack+0.013
rost+0.013
Neg logit contributions
 adorable-0.016
 important-0.016
 idea-0.015
 best-0.015
 precious-0.015
Token.
Token ID13
Feature activation+5.53e-03

Pos logit contributions
talk+0.216
oes+0.215
ted+0.211
stack+0.201
isk+0.198
Neg logit contributions
 adorable-0.174
 best-0.165
 important-0.165
 idea-0.164
 precious-0.161
Token Finally
Token ID9461
Feature activation-1.94e-02

Pos logit contributions
 adorable+0.011
 best+0.011
 important+0.011
 idea+0.011
 beautiful+0.010
Neg logit contributions
talk-0.009
oes-0.009
ted-0.008
stack-0.008
rost-0.008
Token,
Token ID11
Feature activation-5.83e-02

Pos logit contributions
ted+0.038
talk+0.037
oes+0.037
stack+0.035
rost+0.035
Neg logit contributions
 adorable-0.031
 important-0.029
 precious-0.029
 idea-0.029
 prized-0.028
Token they
Token ID484
Feature activation-9.78e-02

Pos logit contributions
talk+0.110
oes+0.109
rost+0.096
use+0.096
isk+0.094
Neg logit contributions
 best-0.115
 adorable-0.110
 beautiful-0.107
 important-0.106
 idea-0.104
Token found
Token ID1043
Feature activation+1.84e-02

Pos logit contributions
oes+0.190
talk+0.180
ted+0.173
stack+0.169
ricanes+0.167
Neg logit contributions
 adorable-0.171
 best-0.170
 idea-0.164
 important-0.162
 beautiful-0.162
Token the
Token ID262
Feature activation+2.81e-02

Pos logit contributions
 adorable+0.036
 important+0.035
 idea+0.035
 best+0.034
 precious+0.034
Neg logit contributions
ted-0.030
oes-0.029
talk-0.029
stack-0.027
ting-0.026
Token wand
Token ID11569
Feature activation-5.61e-02

Pos logit contributions
 adorable+0.033
 impressed+0.032
 pleased+0.032
 thrilled+0.032
 important+0.031
Neg logit contributions
oes-0.057
stack-0.057
 south-0.056
talk-0.056
ted-0.053
Token behind
Token ID2157
Feature activation-1.38e-01

Pos logit contributions
oes+0.112
talk+0.111
ted+0.109
stack+0.105
isk+0.101
Neg logit contributions
 adorable-0.090
 best-0.086
 important-0.086
 idea-0.085
 precious-0.082
Token the
Token ID262
Feature activation-4.03e-02

Pos logit contributions
oes+0.291
talk+0.286
ted+0.273
stack+0.265
rost+0.264
Neg logit contributions
 adorable-0.213
 best-0.207
 idea-0.202
 important-0.202
 precious-0.194
Token asked
Token ID1965
Feature activation-9.96e-03

Pos logit contributions
 adorable+0.083
 best+0.080
 important+0.079
 idea+0.079
 precious+0.077
Neg logit contributions
oes-0.077
talk-0.077
ted-0.075
stack-0.071
rost-0.069
Token many
Token ID867
Feature activation+2.72e-02

Pos logit contributions
oes+0.021
talk+0.020
ted+0.020
stack+0.020
ricanes+0.019
Neg logit contributions
 best-0.016
 idea-0.015
 important-0.015
 adorable-0.015
 loyal-0.014
Token questions
Token ID2683
Feature activation-7.29e-02

Pos logit contributions
 adorable+0.052
 best+0.050
 idea+0.049
 important+0.049
 precious+0.048
Neg logit contributions
oes-0.045
talk-0.044
ted-0.044
rost-0.041
ting-0.040
Token.
Token ID13
Feature activation+1.31e-03

Pos logit contributions
talk+0.166
oes+0.164
ted+0.160
stack+0.153
isk+0.153
Neg logit contributions
 adorable-0.099
 best-0.098
 idea-0.096
 important-0.095
 precious-0.088
Token The
Token ID383
Feature activation+4.10e-02

Pos logit contributions
 idea+0.002
 adorable+0.002
 best+0.002
 important+0.002
 loyal+0.002
Neg logit contributions
talk-0.002
use-0.002
ted-0.002
oes-0.002
ting-0.002
Token man
Token ID582
Feature activation-6.07e-03

Pos logit contributions
 adorable+0.068
 best+0.066
 important+0.065
 idea+0.065
 precious+0.063
Neg logit contributions
oes-0.081
talk-0.080
ted-0.078
stack-0.075
rost-0.074
Token patiently
Token ID37220
Feature activation+5.79e-02

Pos logit contributions
talk+0.011
oes+0.011
ted+0.011
stack+0.010
rost+0.010
Neg logit contributions
 adorable-0.010
 best-0.010
 important-0.010
 idea-0.010
 precious-0.009
Token answered
Token ID9373
Feature activation+6.85e-02

Pos logit contributions
 adorable+0.092
 idea+0.090
 best+0.090
 important+0.089
 precious+0.085
Neg logit contributions
oes-0.118
talk-0.114
ted-0.110
stack-0.109
rost-0.107
Token them
Token ID606
Feature activation-6.28e-02

Pos logit contributions
 adorable+0.101
 best+0.101
 important+0.097
 idea+0.094
 beautiful+0.092
Neg logit contributions
oes-0.153
talk-0.152
use-0.139
stack-0.139
ted-0.138
Token all
Token ID477
Feature activation-7.30e-02

Pos logit contributions
talk+0.135
oes+0.134
ted+0.127
stack+0.125
rost+0.124
Neg logit contributions
 adorable-0.095
 best-0.094
 important-0.092
 idea-0.091
 precious-0.086
Token and
Token ID290
Feature activation-5.88e-02

Pos logit contributions
talk+0.150
oes+0.147
ted+0.141
stack+0.140
isk+0.140
Neg logit contributions
 idea-0.123
 adorable-0.122
 important-0.121
 best-0.118
 beautiful-0.111
Token.
Token ID13
Feature activation+1.04e-02

Pos logit contributions
oes+0.091
talk+0.088
ted+0.088
ting+0.085
stack+0.084
Neg logit contributions
 adorable-0.104
 idea-0.103
 best-0.098
 important-0.098
 precious-0.097
Token From
Token ID3574
Feature activation+3.99e-02

Pos logit contributions
 adorable+0.020
 best+0.020
 important+0.020
 idea+0.019
 precious+0.019
Neg logit contributions
oes-0.017
talk-0.017
ted-0.016
rost-0.015
stack-0.015
Token that
Token ID326
Feature activation-7.14e-02

Pos logit contributions
 adorable+0.081
 idea+0.081
 best+0.080
 important+0.079
 precious+0.077
Neg logit contributions
oes-0.060
talk-0.060
ted-0.060
stack-0.055
rost-0.053
Token day
Token ID1110
Feature activation-6.35e-02

Pos logit contributions
oes+0.108
stack+0.099
ted+0.097
talk+0.096
rost+0.094
Neg logit contributions
 adorable-0.142
 precious-0.135
 idea-0.134
 prized-0.131
 darling-0.130
Token on
Token ID319
Feature activation-1.61e-01

Pos logit contributions
oes+0.102
talk+0.101
ted+0.095
stack+0.091
rost+0.091
Neg logit contributions
 adorable-0.129
 best-0.128
 important-0.125
 idea-0.124
 precious-0.120
Token,
Token ID11
Feature activation-4.33e-02

Pos logit contributions
oes+0.323
ted+0.318
stack+0.314
talk+0.313
rost+0.299
Neg logit contributions
 adorable-0.243
 idea-0.229
 important-0.226
 precious-0.226
 best-0.219
Token Lily
Token ID20037
Feature activation+8.93e-02

Pos logit contributions
oes+0.061
talk+0.060
ted+0.058
stack+0.054
rost+0.053
Neg logit contributions
 adorable-0.095
 best-0.093
 important-0.092
 idea-0.092
 precious-0.090
Token and
Token ID290
Feature activation-3.72e-02

Pos logit contributions
 adorable+0.161
 idea+0.156
 precious+0.152
 important+0.151
 best+0.149
Neg logit contributions
oes-0.155
ted-0.155
talk-0.154
ting-0.142
isk-0.142
Token Tim
Token ID5045
Feature activation+4.83e-02

Pos logit contributions
talk+0.067
oes+0.067
ted+0.062
iffs+0.062
rost+0.060
Neg logit contributions
 adorable-0.071
 best-0.070
 idea-0.067
 important-0.067
 loyal-0.065
Tokenmy
Token ID1820
Feature activation+3.97e-02

Pos logit contributions
 best+0.089
 adorable+0.087
 important+0.083
 idea+0.083
 beautiful+0.079
Neg logit contributions
talk-0.087
oes-0.086
ted-0.084
stack-0.083
iffs-0.081
Token learned
Token ID4499
Feature activation+6.77e-02

Pos logit contributions
 adorable+0.080
 best+0.079
 important+0.078
 idea+0.077
 precious+0.075
Neg logit contributions
oes-0.063
talk-0.063
ted-0.060
stack-0.057
rost-0.056
Token So
Token ID1406
Feature activation+3.49e-02

Pos logit contributions
 adorable+0.199
 important+0.193
 best+0.193
 idea+0.191
 precious+0.187
Neg logit contributions
oes-0.175
talk-0.174
ted-0.168
stack-0.160
isk-0.158
Token they
Token ID484
Feature activation-9.14e-02

Pos logit contributions
 adorable+0.062
 important+0.059
 best+0.059
 idea+0.059
 precious+0.057
Neg logit contributions
oes-0.064
talk-0.063
ted-0.061
stack-0.058
rost-0.057
Token both
Token ID1111
Feature activation-2.68e-03

Pos logit contributions
oes+0.158
talk+0.149
ted+0.145
iffs+0.142
rost+0.142
Neg logit contributions
 adorable-0.172
 best-0.171
 idea-0.170
 important-0.166
 precious-0.162
Token walked
Token ID6807
Feature activation-8.96e-02

Pos logit contributions
oes+0.004
talk+0.004
ted+0.004
stack+0.004
rost+0.004
Neg logit contributions
 adorable-0.005
 best-0.005
 important-0.005
 idea-0.005
 precious-0.005
Token up
Token ID510
Feature activation-9.55e-02

Pos logit contributions
oes+0.125
isk+0.121
rost+0.121
talk+0.115
stack+0.113
Neg logit contributions
 best-0.201
 adorable-0.197
 idea-0.192
 important-0.192
 beautiful-0.188
Token to
Token ID284
Feature activation-2.33e-01

Pos logit contributions
oes+0.145
talk+0.141
isk+0.133
rost+0.133
ted+0.131
Neg logit contributions
 best-0.201
 adorable-0.200
 idea-0.197
 important-0.194
 beautiful-0.194
Token the
Token ID262
Feature activation-4.80e-02

Pos logit contributions
oes+0.453
talk+0.442
ted+0.405
stack+0.398
rost+0.396
Neg logit contributions
 adorable-0.416
 best-0.400
 idea-0.395
 beautiful-0.392
 important-0.384
Token old
Token ID1468
Feature activation+5.48e-02

Pos logit contributions
oes+0.085
talk+0.083
ted+0.081
stack+0.080
rost+0.077
Neg logit contributions
 adorable-0.084
 idea-0.083
 best-0.082
 important-0.082
 precious-0.078
Token lady
Token ID10846
Feature activation-6.76e-02

Pos logit contributions
 adorable+0.104
 best+0.103
 important+0.101
 idea+0.098
 precious+0.098
Neg logit contributions
oes-0.091
talk-0.091
stack-0.088
ted-0.086
rost-0.081
Token,
Token ID11
Feature activation-3.38e-02

Pos logit contributions
oes+0.152
talk+0.151
ted+0.147
rost+0.142
stack+0.140
Neg logit contributions
 adorable-0.092
 best-0.088
 idea-0.086
 important-0.085
 precious-0.082
Token and
Token ID290
Feature activation-5.00e-02

Pos logit contributions
oes+0.067
talk+0.066
ted+0.064
stack+0.061
rost+0.060
Neg logit contributions
 adorable-0.056
 best-0.054
 important-0.053
 idea-0.053
 precious-0.052
Token one
Token ID530
Feature activation+9.11e-02

Pos logit contributions
 adorable+0.066
 best+0.064
 important+0.063
 idea+0.063
 precious+0.060
Neg logit contributions
oes-0.078
talk-0.077
ted-0.075
stack-0.072
rost-0.071
Token day
Token ID1110
Feature activation-4.38e-02

Pos logit contributions
 adorable+0.151
 best+0.147
 idea+0.145
 important+0.145
 precious+0.142
Neg logit contributions
oes-0.177
talk-0.172
ted-0.168
rost-0.159
stack-0.158
Token,
Token ID11
Feature activation-2.01e-02

Pos logit contributions
oes+0.092
talk+0.090
ted+0.085
stack+0.084
rost+0.084
Neg logit contributions
 adorable-0.069
 best-0.068
 important-0.066
 idea-0.065
 beautiful-0.062
Token Jane
Token ID12091
Feature activation+7.89e-02

Pos logit contributions
oes+0.036
talk+0.035
ted+0.033
rost+0.033
stack+0.032
Neg logit contributions
 adorable-0.037
 best-0.037
 important-0.036
 idea-0.036
 precious-0.035
Token went
Token ID1816
Feature activation-5.71e-02

Pos logit contributions
 adorable+0.149
 best+0.145
 important+0.143
 idea+0.142
 precious+0.139
Neg logit contributions
oes-0.136
talk-0.136
ted-0.132
stack-0.123
isk-0.122
Token to
Token ID284
Feature activation-2.15e-01

Pos logit contributions
rost+0.086
oes+0.084
talk+0.082
ted+0.081
stack+0.077
Neg logit contributions
 best-0.124
 important-0.121
 adorable-0.120
 idea-0.119
 beautiful-0.117
Token the
Token ID262
Feature activation-2.88e-02

Pos logit contributions
oes+0.414
talk+0.413
ted+0.406
rost+0.385
iffs+0.374
Neg logit contributions
 adorable-0.373
 idea-0.369
 best-0.361
 beautiful-0.358
 important-0.354
Token park
Token ID3952
Feature activation-1.11e-01

Pos logit contributions
oes+0.048
talk+0.048
ted+0.047
rost+0.044
stack+0.043
Neg logit contributions
 adorable-0.057
 best-0.056
 important-0.055
 idea-0.054
 precious-0.053
Token to
Token ID284
Feature activation-2.23e-01

Pos logit contributions
oes+0.223
talk+0.222
ted+0.219
rost+0.213
use+0.207
Neg logit contributions
 adorable-0.175
 best-0.170
 idea-0.168
 important-0.165
 beautiful-0.159
Token practice
Token ID3357
Feature activation-2.76e-02

Pos logit contributions
ted+0.354
oes+0.354
talk+0.350
rost+0.330
iffs+0.311
Neg logit contributions
 idea-0.459
 adorable-0.459
 beautiful-0.449
 best-0.441
 important-0.430
Token with
Token ID351
Feature activation-2.66e-02

Pos logit contributions
oes+0.050
talk+0.049
ted+0.048
stack+0.046
isk+0.045
Neg logit contributions
 adorable-0.048
 best-0.046
 important-0.045
 precious-0.045
 idea-0.045
Token dog
Token ID3290
Feature activation-4.64e-02

Pos logit contributions
 adorable+0.187
 best+0.182
 important+0.179
 idea+0.179
 precious+0.174
Neg logit contributions
oes-0.187
talk-0.184
ted-0.179
stack-0.170
rost-0.169
Token.
Token ID13
Feature activation+9.94e-03

Pos logit contributions
oes+0.114
talk+0.112
ricanes+0.111
ted+0.109
rost+0.108
Neg logit contributions
 best-0.055
 adorable-0.052
 idea-0.050
 important-0.049
 loyal-0.047
Token He
Token ID679
Feature activation-4.74e-02

Pos logit contributions
 adorable+0.019
 idea+0.018
 best+0.018
 important+0.018
 loyal+0.017
Neg logit contributions
talk-0.018
oes-0.018
use-0.017
ted-0.016
stack-0.016
Token goes
Token ID2925
Feature activation-4.38e-02

Pos logit contributions
oes+0.080
talk+0.078
ted+0.074
stack+0.072
rost+0.071
Neg logit contributions
 adorable-0.093
 best-0.091
 important-0.090
 idea-0.090
 precious-0.088
Token back
Token ID736
Feature activation-1.29e-01

Pos logit contributions
 gra+0.080
oes+0.078
talk+0.077
ricanes+0.077
ted+0.076
Neg logit contributions
 best-0.078
 loyal-0.074
 adorable-0.073
 happy-0.071
 important-0.071
Token to
Token ID284
Feature activation-2.29e-01

Pos logit contributions
oes+0.224
ricanes+0.223
talk+0.223
rost+0.222
 gra+0.219
Neg logit contributions
 best-0.224
 loyal-0.221
 adorable-0.220
 friends-0.217
 important-0.217
Token his
Token ID465
Feature activation-1.43e-02

Pos logit contributions
talk+0.537
oes+0.508
ricanes+0.498
ted+0.494
ricane+0.477
Neg logit contributions
 adorable-0.339
 idea-0.339
 best-0.335
 loyal-0.331
 happy-0.330
Token ball
Token ID2613
Feature activation-4.97e-02

Pos logit contributions
oes+0.025
talk+0.024
stack+0.024
ted+0.023
isk+0.022
Neg logit contributions
 adorable-0.026
 idea-0.024
 important-0.024
 best-0.024
 precious-0.023
Token and
Token ID290
Feature activation-8.36e-02

Pos logit contributions
oes+0.124
talk+0.123
ted+0.120
ricanes+0.118
rost+0.117
Neg logit contributions
 best-0.055
 idea-0.053
 adorable-0.053
 important-0.050
 loyal-0.048
Token plays
Token ID5341
Feature activation-2.43e-02

Pos logit contributions
oes+0.164
talk+0.161
rost+0.154
ales+0.144
isk+0.144
Neg logit contributions
 best-0.143
 adorable-0.142
 loyal-0.141
 precious-0.135
 idea-0.134
Token some
Token ID617
Feature activation+3.77e-03

Pos logit contributions
oes+0.054
talk+0.053
ted+0.051
stack+0.049
use+0.048
Neg logit contributions
 adorable-0.036
 best-0.035
 idea-0.035
 important-0.034
 beautiful-0.033
Token!"
Token ID2474
Feature activation+3.11e-03

Pos logit contributions
oes+0.131
talk+0.130
ted+0.124
stack+0.122
 gra+0.121
Neg logit contributions
 adorable-0.070
 important-0.068
 best-0.066
 precious-0.066
 idea-0.064
Token
Token ID198
Feature activation-3.10e-02

Pos logit contributions
 adorable+0.006
 idea+0.006
 best+0.006
 important+0.005
 beautiful+0.005
Neg logit contributions
oes-0.006
talk-0.006
rost-0.005
isk-0.005
ted-0.005
Token
Token ID198
Feature activation-2.84e-02

Pos logit contributions
oes+0.056
talk+0.055
ted+0.054
stack+0.052
isk+0.052
Neg logit contributions
 adorable-0.055
 best-0.054
 important-0.053
 idea-0.053
 beautiful-0.051
TokenJim
Token ID18050
Feature activation+9.19e-03

Pos logit contributions
oes+0.049
talk+0.048
ted+0.047
stack+0.045
isk+0.044
Neg logit contributions
 adorable-0.054
 best-0.053
 important-0.052
 idea-0.052
 precious-0.050
Token pointed
Token ID6235
Feature activation+7.15e-02

Pos logit contributions
 adorable+0.016
 best+0.016
 important+0.015
 idea+0.015
 precious+0.015
Neg logit contributions
oes-0.017
talk-0.017
ted-0.017
stack-0.016
rost-0.015
Token to
Token ID284
Feature activation-1.61e-01

Pos logit contributions
 adorable+0.156
 best+0.151
 idea+0.150
 important+0.149
 precious+0.145
Neg logit contributions
oes-0.106
talk-0.105
ted-0.098
stack-0.097
isk-0.096
Token the
Token ID262
Feature activation+5.09e-03

Pos logit contributions
oes+0.323
talk+0.321
ted+0.308
stack+0.296
rost+0.292
Neg logit contributions
 adorable-0.267
 best-0.256
 idea-0.253
 important-0.252
 precious-0.243
Token sky
Token ID6766
Feature activation-1.01e-01

Pos logit contributions
 adorable+0.010
 best+0.010
 important+0.010
 idea+0.010
 precious+0.010
Neg logit contributions
oes-0.008
talk-0.008
ted-0.008
stack-0.007
rost-0.007
Token and
Token ID290
Feature activation-6.04e-02

Pos logit contributions
talk+0.228
oes+0.227
ted+0.220
rost+0.212
stack+0.212
Neg logit contributions
 adorable-0.138
 best-0.134
 idea-0.131
 important-0.130
 precious-0.124
Token said
Token ID531
Feature activation+1.05e-01

Pos logit contributions
oes+0.126
talk+0.124
ted+0.118
isk+0.116
rost+0.115
Neg logit contributions
 adorable-0.095
 best-0.090
 important-0.089
 idea-0.089
 beautiful-0.086
Token,
Token ID11
Feature activation-4.10e-03

Pos logit contributions
 adorable+0.141
 best+0.135
 important+0.134
 idea+0.133
 precious+0.129
Neg logit contributions
oes-0.237
talk-0.235
ted-0.230
stack-0.221
rost-0.219
Token alone
Token ID3436
Feature activation+1.30e-03

Pos logit contributions
talk+0.122
oes+0.118
ted+0.117
rost+0.111
stack+0.111
Neg logit contributions
 adorable-0.100
 important-0.099
 best-0.097
 idea-0.093
 precious-0.093
Token.
Token ID13
Feature activation+8.82e-03

Pos logit contributions
 adorable+0.002
 best+0.002
 important+0.002
 idea+0.002
 precious+0.002
Neg logit contributions
oes-0.003
talk-0.003
ted-0.003
stack-0.002
rost-0.002
Token From
Token ID3574
Feature activation+3.03e-02

Pos logit contributions
 adorable+0.017
 best+0.017
 important+0.017
 idea+0.017
 precious+0.016
Neg logit contributions
oes-0.015
talk-0.015
ted-0.014
stack-0.013
rost-0.013
Token that
Token ID326
Feature activation-6.82e-02

Pos logit contributions
 adorable+0.061
 best+0.060
 important+0.059
 idea+0.059
 precious+0.057
Neg logit contributions
oes-0.048
talk-0.048
ted-0.046
stack-0.044
rost-0.043
Token day
Token ID1110
Feature activation-6.60e-02

Pos logit contributions
oes+0.108
talk+0.099
ted+0.098
stack+0.095
use+0.094
Neg logit contributions
 adorable-0.133
 precious-0.127
 idea-0.125
 prized-0.123
 darling-0.122
Token on
Token ID319
Feature activation-1.59e-01

Pos logit contributions
oes+0.108
talk+0.106
ted+0.100
stack+0.097
rost+0.097
Neg logit contributions
 adorable-0.132
 best-0.132
 important-0.129
 idea-0.127
 precious-0.123
Token,
Token ID11
Feature activation-4.39e-02

Pos logit contributions
oes+0.327
ted+0.323
talk+0.313
stack+0.311
rost+0.299
Neg logit contributions
 adorable-0.238
 idea-0.224
 precious-0.221
 important-0.220
 best-0.215
Token the
Token ID262
Feature activation+2.09e-02

Pos logit contributions
ted+0.059
oes+0.059
ting+0.055
stack+0.052
isk+0.052
Neg logit contributions
 adorable-0.091
 idea-0.091
 precious-0.089
 best-0.088
 important-0.087
Token animals
Token ID4695
Feature activation-7.98e-02

Pos logit contributions
 adorable+0.042
 pleased+0.041
 impressed+0.041
 thrilled+0.041
 idea+0.041
Neg logit contributions
oes-0.028
ted-0.027
talk-0.027
stack-0.026
isk-0.026
Token had
Token ID550
Feature activation+3.53e-03

Pos logit contributions
oes+0.155
talk+0.154
rost+0.146
ted+0.146
iffs+0.142
Neg logit contributions
 best-0.139
 adorable-0.136
 important-0.132
 idea-0.132
 beautiful-0.125
Token a
Token ID257
Feature activation+1.55e-01

Pos logit contributions
 adorable+0.006
 idea+0.006
 important+0.005
 best+0.005
 precious+0.005
Neg logit contributions
talk-0.007
oes-0.007
ted-0.007
isk-0.006
stack-0.006