TOP ACTIVATIONS
MAX = 10.626

Suddenly , the rh in oc eros heard his friends laughing
The rh in oc eros walked to the fence
," said the rh in oc eros , feeling discouraged .
replied the rh in oc eros sadly . It had
The rh in oc eros looked up , hopeful
here ?" The rh in oc eros replied , " I
doing ?" The rh in oc eros replied , " I
?" The refrigerator replied , " Yes ,
Billy . The rh in oc eros looked at Billy with
. The av oc ados were shared with everybody
The rh in oc eros nodded its head .
The rh in oc eros couldn 't believe his
happy ." The rh in oc eros did what the bird
The rh in oc eros was so delighted with
help . The rh in oc eros was very happy and
," said the rh in oc eros . " I 'm
Suddenly , the rh in oc eros saw something in the
Finally , the rh in oc eros remembered that animals often
name ?" The rh in oc eros was surprised that the
The rh in oc eros was so excited that

MIN ACTIVATIONS
MIN = -9.057

y ummy spaghetti with meat balls ." Lily was so happy
Lily loved spaghetti with meat balls . She watched as her
food , spaghetti with meat balls . It smelled so delicious
food - spaghetti with meat balls . L ily
make snow men and snow balls ." L ily
of all the fun things they did . Ch oo -
every day and shared everything they had . <|endoftext|> Once upon
menu with pictures of food they liked , such as pizza
snow men and throwing snow balls . " Wow
people use to buy things they want or need .
and enjoyed the beautiful sights they found . At the end
of all the wonderful things they had done . It was
. They share their snow balls and snow men . They
They had fun throwing snow balls at each other . But
good friends and share everything they found in the park from
snow men , threw snow balls , and slid down hills
on , they shared everything they found in the forest and
ty and Sammy shared everything they found and were the happiest
liked to examine different things they found , like leaves ,
He cooked spaghetti with meat balls . His friends all said

INTERVAL 5.705 - 10.626
CONTAINS 0.126%

under the couch . The nos y person was determined to
snap '. The egg looked around and saw a
water was gone , the town was very happy . They
?" The man smiled and answered , "
hello . The stranger smiled at the bird ,

INTERVAL 0.785 - 5.705
CONTAINS 32.104%

is that ?" " That is a bird 's
The referee is calm and kind . He smiles at Ben
When they were done , John said , " Look Daddy
it . Sam was curious but she was also
gloomy . She was getting worried that the party might be

INTERVAL -4.136 - 0.785
CONTAINS 66.167%

some big balloons , some paint and some fuzzy things .
I will help you . â
very careful with his cars and never let anyone else touch
friends , Max and Sally . They lived by the beach
her toys off the shelf . Lily was very sad and

INTERVAL -9.057 - -4.136
CONTAINS 1.603%

. The pigeon had many friends . They played together every
Tim and the squirrel played with the cone , rolling
to run and jump and throw balls . He was very
day , Lily wanted to wear her favorite dress to school
will come back tomorrow and play with you ," they say
TokenSuddenly
Token ID38582
Feature activation+2.49e+00

Pos logit contributions
 beast+0.023
 gut+0.023
 mistake+0.022
 swollen+0.021
 therm+0.021
Neg logit contributions
 parties-0.024
 tea-0.021
 celebrate-0.018
 recess-0.018
 picnic-0.017
Token,
Token ID11
Feature activation+1.18e+00

Pos logit contributions
 beast+0.100
 gut+0.100
 mistake+0.094
inct+0.093
 therm+0.093
Neg logit contributions
 parties-0.066
 tea-0.057
 recess-0.046
 celebrate-0.043
 party-0.042
Token the
Token ID262
Feature activation+4.39e+00

Pos logit contributions
 gut+0.049
 beast+0.048
 mistake+0.048
 swollen+0.046
inct+0.046
Neg logit contributions
 parties-0.030
 tea-0.027
 recess-0.021
 celebrate-0.020
 picnic-0.019
Token rh
Token ID9529
Feature activation+5.52e+00

Pos logit contributions
 gut+0.157
rum+0.152
 mistake+0.136
 outage+0.136
 severe+0.135
Neg logit contributions
 parties-0.122
 tea-0.119
 picnic-0.111
 party-0.107
 celebrate-0.106
Tokenin
Token ID259
Feature activation+5.62e+00

Pos logit contributions
 gut+0.240
 plea+0.220
 severe+0.215
iously+0.214
 swollen+0.214
Neg logit contributions
 parties-0.106
 tea-0.088
acky-0.082
 together-0.076
P-0.075
Tokenoc
Token ID420
Feature activation+9.07e+00

Pos logit contributions
 gut+0.215
 swollen+0.188
 mistake+0.185
rum+0.184
 vain+0.184
Neg logit contributions
 played-0.125
 picnic-0.118
 tea-0.118
 play-0.111
acky-0.105
Tokeneros
Token ID27498
Feature activation+4.81e+00

Pos logit contributions
 severe+0.304
 outage+0.294
 swollen+0.283
 gut+0.280
 reckless+0.274
Neg logit contributions
 parties-0.255
 tea-0.218
ils-0.196
 sandwiches-0.193
 recess-0.186
Token heard
Token ID2982
Feature activation+3.52e+00

Pos logit contributions
 gut+0.192
rum+0.178
 swollen+0.176
 therm+0.173
 mistake+0.173
Neg logit contributions
 parties-0.116
 tea-0.109
 played-0.098
 party-0.092
 picnic-0.092
Token his
Token ID465
Feature activation+1.94e+00

Pos logit contributions
 gut+0.153
 beast+0.147
 swollen+0.146
cut+0.143
 mistake+0.143
Neg logit contributions
 parties-0.084
 tea-0.075
 celebrate-0.054
 recess-0.052
 party-0.051
Token friends
Token ID2460
Feature activation-9.98e-01

Pos logit contributions
 gut+0.054
 beast+0.053
cut+0.052
 swollen+0.051
inct+0.050
Neg logit contributions
 parties-0.071
 tea-0.066
 recess-0.055
 party-0.055
 celebrate-0.055
Token laughing
Token ID14376
Feature activation-8.10e-01

Pos logit contributions
 parties+0.033
 tea+0.029
 recess+0.025
 celebrate+0.024
 party+0.024
Neg logit contributions
 beast-0.033
 gut-0.033
inct-0.031
 swollen-0.031
 mistake-0.031
Token
Token ID198
Feature activation+2.27e-01

Pos logit contributions
 gut+0.013
 beast+0.013
 mistake+0.012
 therm+0.012
inct+0.012
Neg logit contributions
 parties-0.014
 tea-0.013
 recess-0.011
 celebrate-0.011
 party-0.010
Token
Token ID198
Feature activation+6.13e-01

Pos logit contributions
 gut+0.007
 beast+0.007
 mistake+0.007
inct+0.007
 therm+0.007
Neg logit contributions
 parties-0.008
 tea-0.007
 recess-0.006
 party-0.006
 celebrate-0.006
TokenThe
Token ID464
Feature activation+4.18e+00

Pos logit contributions
 gut+0.021
 beast+0.021
 mistake+0.019
inct+0.019
 therm+0.019
Neg logit contributions
 parties-0.020
 tea-0.018
 recess-0.015
 celebrate-0.015
 party-0.014
Token rh
Token ID9529
Feature activation+4.82e+00

Pos logit contributions
 gut+0.136
rum+0.123
 beast+0.119
inct+0.119
 severe+0.118
Neg logit contributions
 parties-0.137
 tea-0.129
 party-0.112
 picnic-0.112
 celebrate-0.110
Tokenin
Token ID259
Feature activation+5.15e+00

Pos logit contributions
 gut+0.195
iously+0.183
inct+0.182
 plea+0.182
 severe+0.180
Neg logit contributions
 parties-0.116
 tea-0.101
 recess-0.080
 party-0.078
acky-0.076
Tokenoc
Token ID420
Feature activation+9.01e+00

Pos logit contributions
 gut+0.162
 severe+0.149
 swollen+0.144
 reckless+0.140
rum+0.138
Neg logit contributions
 tea-0.154
 parties-0.146
 played-0.144
 picnic-0.138
 party-0.135
Tokeneros
Token ID27498
Feature activation+4.57e+00

Pos logit contributions
 severe+0.303
 swollen+0.286
 outage+0.276
 reckless+0.272
 gut+0.268
Neg logit contributions
 parties-0.252
 tea-0.219
 sandwiches-0.192
 recess-0.191
ils-0.191
Token walked
Token ID6807
Feature activation-1.53e+00

Pos logit contributions
 gut+0.178
 severe+0.166
 beast+0.164
 swollen+0.163
 therm+0.162
Neg logit contributions
 parties-0.122
 tea-0.116
 played-0.098
 party-0.093
 picnic-0.091
Token to
Token ID284
Feature activation+3.58e+00

Pos logit contributions
 parties+0.053
 tea+0.047
 recess+0.040
 celebrate+0.038
 party+0.037
Neg logit contributions
 beast-0.050
 gut-0.050
 mistake-0.047
 swollen-0.046
 therm-0.046
Token the
Token ID262
Feature activation+6.34e-01

Pos logit contributions
 beast+0.131
 gut+0.130
 mistake+0.126
 therm+0.121
 passer+0.121
Neg logit contributions
 parties-0.106
 tea-0.090
 recess-0.075
ets-0.072
ils-0.069
Token fence
Token ID13990
Feature activation+4.07e+00

Pos logit contributions
 gut+0.020
rum+0.020
inct+0.019
 beast+0.018
 swollen+0.018
Neg logit contributions
 parties-0.020
 tea-0.020
 picnic-0.018
 party-0.018
 celebrate-0.017
Token,"
Token ID553
Feature activation+1.13e+00

Pos logit contributions
 gut+0.064
 beast+0.064
inct+0.061
 mistake+0.060
 therm+0.060
Neg logit contributions
 parties-0.041
 tea-0.036
 recess-0.028
 celebrate-0.027
 party-0.026
Token said
Token ID531
Feature activation-7.05e-01

Pos logit contributions
 gut+0.043
 beast+0.042
 mistake+0.040
inct+0.039
 severe+0.038
Neg logit contributions
 parties-0.033
 tea-0.030
 celebrate-0.024
 recess-0.023
 party-0.023
Token the
Token ID262
Feature activation+3.31e+00

Pos logit contributions
 parties+0.020
 tea+0.018
 recess+0.014
 celebrate+0.014
 party+0.013
Neg logit contributions
 gut-0.027
 beast-0.027
 mistake-0.026
inct-0.025
 swollen-0.025
Token rh
Token ID9529
Feature activation+5.03e+00

Pos logit contributions
 gut+0.104
rum+0.093
 beast+0.093
inct+0.091
 mistake+0.089
Neg logit contributions
 parties-0.113
 tea-0.103
 celebrate-0.091
 picnic-0.090
 party-0.090
Tokenin
Token ID259
Feature activation+5.27e+00

Pos logit contributions
 gut+0.204
inct+0.189
 plea+0.186
 swollen+0.186
 mistake+0.185
Neg logit contributions
 parties-0.117
 tea-0.097
acky-0.082
 celebrate-0.080
 together-0.079
Tokenoc
Token ID420
Feature activation+9.01e+00

Pos logit contributions
 swollen+0.214
 gut+0.210
 distress+0.200
 severe+0.199
 beast+0.198
Neg logit contributions
 tea-0.090
 parties-0.086
 together-0.085
 played-0.069
 play-0.068
Tokeneros
Token ID27498
Feature activation+4.90e+00

Pos logit contributions
 outage+0.276
 severe+0.273
 swollen+0.264
 gut+0.263
 confusion+0.260
Neg logit contributions
 parties-0.259
 tea-0.220
ils-0.212
 sandwiches-0.198
 recess-0.188
Token,
Token ID11
Feature activation+3.29e-01

Pos logit contributions
 gut+0.218
 beast+0.216
 swollen+0.208
inct+0.208
 mistake+0.207
Neg logit contributions
 parties-0.094
 tea-0.087
 party-0.057
 celebrate-0.056
 played-0.055
Token feeling
Token ID4203
Feature activation+2.16e+00

Pos logit contributions
 gut+0.012
 beast+0.011
 severe+0.011
 mistake+0.011
inct+0.011
Neg logit contributions
 tea-0.010
 parties-0.009
 played-0.008
 Lily-0.008
 together-0.008
Token discouraged
Token ID30170
Feature activation+3.61e+00

Pos logit contributions
 beast+0.063
inct+0.063
 mistake+0.062
 gut+0.061
 adult+0.059
Neg logit contributions
 parties-0.073
 tea-0.072
 celebrate-0.057
 played-0.056
 wear-0.056
Token.
Token ID13
Feature activation+1.60e+00

Pos logit contributions
 beast+0.178
 gut+0.178
inct+0.177
 swollen+0.170
 passer+0.169
Neg logit contributions
 parties-0.056
 tea-0.048
 together-0.033
 wear-0.023
 party-0.023
Token
Token ID26391
Feature activation-6.98e-01

Pos logit contributions
 gut+0.039
 beast+0.038
 swollen+0.037
 mistake+0.036
 therm+0.036
Neg logit contributions
 parties-0.028
 tea-0.023
 recess-0.020
 celebrate-0.020
 wear-0.018
Token replied
Token ID8712
Feature activation-1.70e+00

Pos logit contributions
 parties+0.022
 tea+0.019
 recess+0.016
 celebrate+0.016
 picnic+0.015
Neg logit contributions
 beast-0.024
 gut-0.024
 mistake-0.023
 therm-0.023
 swollen-0.023
Token the
Token ID262
Feature activation+3.41e+00

Pos logit contributions
 parties+0.040
 tea+0.036
 celebrate+0.026
 recess+0.025
 wear+0.024
Neg logit contributions
 beast-0.073
 gut-0.070
 mistake-0.069
inct-0.068
 swollen-0.067
Token rh
Token ID9529
Feature activation+5.04e+00

Pos logit contributions
 gut+0.109
 beast+0.094
inct+0.093
 severe+0.093
 swollen+0.093
Neg logit contributions
 parties-0.116
 tea-0.113
 party-0.096
 picnic-0.096
 celebrate-0.094
Tokenin
Token ID259
Feature activation+5.49e+00

Pos logit contributions
 gut+0.204
inct+0.194
 plea+0.193
 mistake+0.190
 swollen+0.188
Neg logit contributions
 parties-0.123
 tea-0.106
 recess-0.085
ils-0.081
acky-0.080
Tokenoc
Token ID420
Feature activation+8.99e+00

Pos logit contributions
 gut+0.171
 swollen+0.165
 severe+0.155
 distress+0.155
 injured+0.154
Neg logit contributions
 tea-0.164
 parties-0.151
 played-0.134
 play-0.129
 celebrate-0.129
Tokeneros
Token ID27498
Feature activation+4.60e+00

Pos logit contributions
 severe+0.291
 swollen+0.272
 outage+0.268
 terrified+0.267
 frightening+0.260
Neg logit contributions
 parties-0.262
 tea-0.230
ils-0.217
 recess-0.204
 sandwiches-0.193
Token sadly
Token ID21098
Feature activation+5.55e+00

Pos logit contributions
 gut+0.160
 beast+0.156
inct+0.152
 mistake+0.150
 swollen+0.149
Neg logit contributions
 parties-0.134
 tea-0.128
 party-0.097
 celebrate-0.097
 played-0.096
Token.
Token ID13
Feature activation+5.09e-01

Pos logit contributions
 beast+0.246
 gut+0.238
inct+0.227
 passer+0.226
 swollen+0.223
Neg logit contributions
 parties-0.119
 tea-0.107
 wear-0.080
 celebrate-0.079
 together-0.078
Token It
Token ID632
Feature activation+4.02e+00

Pos logit contributions
 beast+0.021
 gut+0.020
 mistake+0.020
 swollen+0.019
cut+0.019
Neg logit contributions
 parties-0.013
 tea-0.011
 celebrate-0.009
 recess-0.009
 Lilly-0.009
Token had
Token ID550
Feature activation-1.43e+00

Pos logit contributions
 gut+0.157
 beast+0.152
 passer+0.141
 adult+0.141
 mistake+0.141
Neg logit contributions
 tea-0.098
 parties-0.092
 played-0.084
 wear-0.073
 together-0.072
Token
Token ID198
Feature activation+5.72e-01

Pos logit contributions
 gut+0.037
 beast+0.036
 mistake+0.036
 swollen+0.035
 severe+0.034
Neg logit contributions
 parties-0.021
 tea-0.018
 celebrate-0.015
 together-0.014
 recess-0.014
Token
Token ID198
Feature activation+1.13e+00

Pos logit contributions
 beast+0.020
 gut+0.020
 mistake+0.019
inct+0.018
 therm+0.018
Neg logit contributions
 parties-0.018
 tea-0.016
 recess-0.014
 celebrate-0.013
 party-0.013
TokenThe
Token ID464
Feature activation+4.53e+00

Pos logit contributions
 beast+0.041
 gut+0.041
 swollen+0.040
 therm+0.040
 mistake+0.039
Neg logit contributions
 parties-0.033
 tea-0.026
 celebrate-0.025
 games-0.023
 recess-0.023
Token rh
Token ID9529
Feature activation+5.34e+00

Pos logit contributions
 beast+0.113
 gut+0.112
 mistake+0.102
inct+0.101
 therm+0.100
Neg logit contributions
 parties-0.190
 tea-0.173
 recess-0.153
 celebrate-0.149
 party-0.146
Tokenin
Token ID259
Feature activation+5.56e+00

Pos logit contributions
 gut+0.218
 plea+0.203
inct+0.201
iously+0.200
 swollen+0.197
Neg logit contributions
 parties-0.127
 tea-0.105
acky-0.087
 together-0.086
 celebrate-0.085
Tokenoc
Token ID420
Feature activation+8.98e+00

Pos logit contributions
 gut+0.178
 swollen+0.159
 reckless+0.156
 mistake+0.154
 severe+0.152
Neg logit contributions
 parties-0.157
 tea-0.155
 played-0.151
 celebrate-0.143
 play-0.140
Tokeneros
Token ID27498
Feature activation+4.83e+00

Pos logit contributions
 severe+0.270
 swollen+0.256
 outage+0.254
 gut+0.253
 reckless+0.247
Neg logit contributions
 parties-0.287
 tea-0.238
ils-0.229
 recess-0.221
 party-0.205
Token looked
Token ID3114
Feature activation-4.51e-01

Pos logit contributions
 gut+0.193
 swollen+0.188
 beast+0.184
 therm+0.181
rum+0.179
Neg logit contributions
 parties-0.112
 tea-0.105
 played-0.102
 party-0.089
 picnic-0.086
Token up
Token ID510
Feature activation+3.99e+00

Pos logit contributions
 parties+0.014
 tea+0.012
 recess+0.011
 celebrate+0.010
 party+0.009
Neg logit contributions
 gut-0.017
 beast-0.016
 mistake-0.015
inct-0.015
 therm-0.015
Token,
Token ID11
Feature activation-3.29e-01

Pos logit contributions
 beast+0.155
 gut+0.152
 mistake+0.144
 swollen+0.143
inct+0.143
Neg logit contributions
 parties-0.111
 tea-0.097
 recess-0.079
 celebrate-0.075
 picnic-0.073
Token hopeful
Token ID17836
Feature activation+2.31e+00

Pos logit contributions
 parties+0.007
 tea+0.006
 together+0.006
 played+0.005
 wear+0.005
Neg logit contributions
 beast-0.014
 gut-0.014
rum-0.014
 severe-0.013
inct-0.013
Token here
Token ID994
Feature activation+1.12e+00

Pos logit contributions
 parties+0.030
 tea+0.026
 celebrate+0.021
 recess+0.020
 party+0.020
Neg logit contributions
 gut-0.042
 beast-0.041
 swollen-0.041
inct-0.040
 mistake-0.040
Token?"
Token ID1701
Feature activation+2.61e+00

Pos logit contributions
 beast+0.044
 gut+0.043
inct+0.041
 mistake+0.041
 swollen+0.041
Neg logit contributions
 parties-0.032
 tea-0.027
 recess-0.022
 celebrate-0.022
 party-0.021
Token The
Token ID383
Feature activation+4.37e+00

Pos logit contributions
 swollen+0.109
 gut+0.109
 mistake+0.106
 severe+0.105
 therm+0.105
Neg logit contributions
 parties-0.061
 tea-0.053
 celebrate-0.044
 played-0.042
 recess-0.041
Token rh
Token ID9529
Feature activation+5.51e+00

Pos logit contributions
 gut+0.146
inct+0.130
rum+0.130
 beast+0.125
 swollen+0.121
Neg logit contributions
 parties-0.140
 tea-0.127
 celebrate-0.117
 picnic-0.114
 party-0.113
Tokenin
Token ID259
Feature activation+5.96e+00

Pos logit contributions
 gut+0.237
inct+0.216
iously+0.215
 plea+0.215
 therm+0.212
Neg logit contributions
 parties-0.116
 tea-0.091
 together-0.083
acky-0.080
 celebrate-0.077
Tokenoc
Token ID420
Feature activation+8.97e+00

Pos logit contributions
 gut+0.221
 swollen+0.191
 outage+0.190
 severe+0.189
rier+0.187
Neg logit contributions
 played-0.135
 named-0.126
 together-0.124
 play-0.123
 tea-0.122
Tokeneros
Token ID27498
Feature activation+5.45e+00

Pos logit contributions
 severe+0.286
 outage+0.284
 gut+0.275
 swollen+0.270
 therm+0.260
Neg logit contributions
 parties-0.267
 tea-0.217
 recess-0.203
ils-0.202
 sandwiches-0.200
Token replied
Token ID8712
Feature activation-4.89e-01

Pos logit contributions
 gut+0.241
 severe+0.223
rum+0.223
 beast+0.219
 outage+0.218
Neg logit contributions
 parties-0.101
 tea-0.093
 played-0.090
 party-0.078
 together-0.077
Token,
Token ID11
Feature activation+3.50e-01

Pos logit contributions
 parties+0.010
 tea+0.009
 together+0.006
 celebrate+0.006
 recess+0.006
Neg logit contributions
 beast-0.022
 gut-0.021
inct-0.021
 mistake-0.021
 swollen-0.021
Token "
Token ID366
Feature activation+3.72e+00

Pos logit contributions
 beast+0.014
 gut+0.013
 severe+0.013
 outage+0.013
 mistake+0.013
Neg logit contributions
 parties-0.008
 tea-0.008
 together-0.007
 them-0.007
 played-0.007
TokenI
Token ID40
Feature activation-2.13e+00

Pos logit contributions
 gut+0.157
 mistake+0.155
 beast+0.153
 therm+0.144
 swollen+0.143
Neg logit contributions
 parties-0.086
 tea-0.071
 celebrate-0.057
 recess-0.056
 wear-0.056
Token doing
Token ID1804
Feature activation-9.61e-01

Pos logit contributions
 parties+0.086
 tea+0.076
 party+0.063
 celebrate+0.063
 recess+0.061
Neg logit contributions
 gut-0.097
 beast-0.092
 swollen-0.090
inct-0.089
 severe-0.088
Token?"
Token ID1701
Feature activation+1.88e+00

Pos logit contributions
 parties+0.027
 tea+0.023
 recess+0.019
 celebrate+0.018
 party+0.017
Neg logit contributions
 beast-0.037
 gut-0.037
 mistake-0.035
inct-0.035
 swollen-0.035
Token The
Token ID383
Feature activation+4.10e+00

Pos logit contributions
 gut+0.075
 beast+0.074
 mistake+0.070
 swollen+0.070
 severe+0.068
Neg logit contributions
 parties-0.052
 tea-0.045
 celebrate-0.037
 recess-0.036
 party-0.034
Token rh
Token ID9529
Feature activation+5.44e+00

Pos logit contributions
 gut+0.121
 beast+0.112
inct+0.107
rum+0.107
 severe+0.102
Neg logit contributions
 parties-0.152
 tea-0.143
 party-0.123
 picnic-0.122
 celebrate-0.122
Tokenin
Token ID259
Feature activation+5.97e+00

Pos logit contributions
 gut+0.216
inct+0.195
 beast+0.193
 plea+0.193
 therm+0.193
Neg logit contributions
 parties-0.143
 tea-0.123
 celebrate-0.101
 together-0.098
acky-0.098
Tokenoc
Token ID420
Feature activation+8.94e+00

Pos logit contributions
 gut+0.196
 reckless+0.163
 severe+0.163
 beast+0.162
 adult+0.160
Neg logit contributions
 tea-0.172
 played-0.170
 parties-0.167
 play-0.155
 together-0.154
Tokeneros
Token ID27498
Feature activation+5.47e+00

Pos logit contributions
 severe+0.276
 gut+0.270
 outage+0.263
 swollen+0.256
 therm+0.251
Neg logit contributions
 parties-0.274
 tea-0.235
ils-0.213
 sandwiches-0.212
 recess-0.203
Token replied
Token ID8712
Feature activation-7.51e-01

Pos logit contributions
 gut+0.232
 beast+0.218
 severe+0.212
 therm+0.207
 swollen+0.207
Neg logit contributions
 parties-0.132
 tea-0.117
 played-0.100
 party-0.094
 celebrate-0.091
Token,
Token ID11
Feature activation+2.89e-01

Pos logit contributions
 parties+0.019
 tea+0.017
 recess+0.013
 celebrate+0.013
 party+0.012
Neg logit contributions
 beast-0.031
 gut-0.031
 mistake-0.029
inct-0.029
 therm-0.029
Token "
Token ID366
Feature activation+3.58e+00

Pos logit contributions
 gut+0.010
 beast+0.010
 severe+0.009
inct+0.009
 mistake+0.009
Neg logit contributions
 parties-0.009
 tea-0.009
 together-0.007
 played-0.007
 celebrate-0.007
TokenI
Token ID40
Feature activation-2.27e+00

Pos logit contributions
 gut+0.142
 beast+0.139
 mistake+0.134
inct+0.128
 therm+0.128
Neg logit contributions
 parties-0.097
 tea-0.085
 celebrate-0.069
 recess-0.068
 wear-0.064
Token?"
Token ID1701
Feature activation+2.07e+00

Pos logit contributions
 parties+0.023
 tea+0.020
 recess+0.016
 celebrate+0.016
 party+0.015
Neg logit contributions
 gut-0.032
 beast-0.031
inct-0.030
 mistake-0.029
 swollen-0.029
Token
Token ID220
Feature activation+1.64e+00

Pos logit contributions
 swollen+0.088
 gut+0.088
 mistake+0.086
 severe+0.086
 beast+0.086
Neg logit contributions
 parties-0.049
 tea-0.040
 recess-0.033
 celebrate-0.032
 wear-0.029
Token
Token ID198
Feature activation+6.75e-01

Pos logit contributions
 beast+0.061
 gut+0.061
 mistake+0.057
 therm+0.057
 swollen+0.057
Neg logit contributions
 parties-0.048
 tea-0.042
 recess-0.035
 celebrate-0.033
 party-0.032
Token
Token ID198
Feature activation+1.15e+00

Pos logit contributions
 gut+0.029
 beast+0.029
 mistake+0.028
 swollen+0.028
 therm+0.028
Neg logit contributions
 parties-0.015
 tea-0.012
 recess-0.010
 celebrate-0.010
 wear-0.009
TokenThe
Token ID464
Feature activation+4.62e+00

Pos logit contributions
 mistake+0.044
 therm+0.043
 gut+0.043
 swollen+0.043
 beast+0.042
Neg logit contributions
 parties-0.028
P-0.023
 tea-0.022
 celebrate-0.022
 recess-0.022
Token refrigerator
Token ID30500
Feature activation+8.92e+00

Pos logit contributions
 gut+0.144
rum+0.137
 beast+0.130
inct+0.129
shadow+0.126
Neg logit contributions
 tea-0.165
 parties-0.161
 party-0.145
 birthday-0.132
 picnic-0.129
Token replied
Token ID8712
Feature activation+9.07e-02

Pos logit contributions
 gut+0.342
rum+0.316
 beast+0.313
 severe+0.310
 mistake+0.300
Neg logit contributions
 parties-0.242
 tea-0.231
 party-0.200
 celebrate-0.186
 picnic-0.184
Token,
Token ID11
Feature activation+9.91e-01

Pos logit contributions
 beast+0.004
 gut+0.004
 mistake+0.004
 swollen+0.004
inct+0.004
Neg logit contributions
 parties-0.002
 tea-0.002
 recess-0.001
 celebrate-0.001
 party-0.001
Token "
Token ID366
Feature activation+3.89e+00

Pos logit contributions
 beast+0.036
 gut+0.036
 severe+0.035
 mistake+0.034
 swollen+0.033
Neg logit contributions
 parties-0.027
 tea-0.027
 together-0.023
 them-0.021
 celebrate-0.021
TokenYes
Token ID5297
Feature activation-2.75e+00

Pos logit contributions
 gut+0.132
 beast+0.132
 mistake+0.126
 therm+0.122
inct+0.121
Neg logit contributions
 parties-0.127
 tea-0.112
 recess-0.096
 celebrate-0.094
 party-0.089
Token,
Token ID11
Feature activation-2.14e+00

Pos logit contributions
 parties+0.085
 tea+0.072
 recess+0.065
 celebrate+0.060
 picnic+0.057
Neg logit contributions
 beast-0.099
 gut-0.098
 mistake-0.093
 therm-0.091
inct-0.090
Token Billy
Token ID15890
Feature activation+8.00e-01

Pos logit contributions
 parties+0.044
 tea+0.039
 celebrate+0.031
 recess+0.030
 wear+0.029
Neg logit contributions
 beast-0.061
 gut-0.059
 mistake-0.057
inct-0.057
 swollen-0.056
Token.
Token ID13
Feature activation+6.09e-01

Pos logit contributions
 gut+0.029
 beast+0.029
inct+0.028
 mistake+0.027
 swollen+0.027
Neg logit contributions
 parties-0.023
 tea-0.021
 celebrate-0.016
 wear-0.016
 recess-0.016
Token The
Token ID383
Feature activation+3.42e+00

Pos logit contributions
 beast+0.022
 gut+0.022
 mistake+0.021
 swollen+0.020
inct+0.020
Neg logit contributions
 parties-0.018
 tea-0.016
 celebrate-0.013
 recess-0.013
 wear-0.012
Token rh
Token ID9529
Feature activation+4.78e+00

Pos logit contributions
 gut+0.117
rum+0.106
inct+0.103
 outage+0.101
 severe+0.100
Neg logit contributions
 parties-0.106
 tea-0.103
 picnic-0.092
 party-0.090
 celebrate-0.087
Tokenin
Token ID259
Feature activation+5.05e+00

Pos logit contributions
 gut+0.205
 plea+0.190
inct+0.188
 severe+0.184
iously+0.182
Neg logit contributions
 parties-0.107
 tea-0.086
 recess-0.070
acky-0.069
 together-0.069
Tokenoc
Token ID420
Feature activation+8.88e+00

Pos logit contributions
 gut+0.171
 reckless+0.148
 severe+0.147
 vain+0.145
 mistake+0.145
Neg logit contributions
 tea-0.134
 played-0.132
 parties-0.126
 play-0.121
 picnic-0.120
Tokeneros
Token ID27498
Feature activation+4.43e+00

Pos logit contributions
 severe+0.302
 outage+0.292
 swollen+0.275
 gut+0.274
 reckless+0.266
Neg logit contributions
 parties-0.253
 tea-0.213
ils-0.207
 recess-0.188
 sandwiches-0.182
Token looked
Token ID3114
Feature activation-9.23e-01

Pos logit contributions
 gut+0.168
 beast+0.160
inct+0.153
 mistake+0.150
 severe+0.150
Neg logit contributions
 parties-0.123
 tea-0.115
 played-0.096
 party-0.092
 picnic-0.090
Token at
Token ID379
Feature activation+3.13e+00

Pos logit contributions
 parties+0.029
 tea+0.025
 recess+0.022
 celebrate+0.020
 party+0.019
Neg logit contributions
 gut-0.033
 beast-0.032
 mistake-0.031
 therm-0.030
 swollen-0.030
Token Billy
Token ID15890
Feature activation-1.13e+00

Pos logit contributions
 beast+0.128
 gut+0.127
 mistake+0.121
 therm+0.119
inct+0.119
Neg logit contributions
 parties-0.081
 tea-0.069
 recess-0.056
 celebrate-0.053
 party-0.050
Token with
Token ID351
Feature activation-1.46e+00

Pos logit contributions
 parties+0.029
 tea+0.026
 celebrate+0.021
 picnic+0.020
 wear+0.020
Neg logit contributions
 gut-0.046
 beast-0.045
inct-0.043
 swollen-0.042
 severe-0.041
Token.
Token ID13
Feature activation-7.72e-01

Pos logit contributions
 gut+0.067
 beast+0.066
inct+0.062
cut+0.062
 passer+0.062
Neg logit contributions
 parties-0.034
 tea-0.028
 celebrate-0.020
 recess-0.020
 party-0.020
Token
Token ID198
Feature activation-8.48e-01

Pos logit contributions
 parties+0.021
 tea+0.018
 recess+0.015
 celebrate+0.015
 party+0.013
Neg logit contributions
 gut-0.030
 beast-0.030
 mistake-0.029
 swollen-0.029
 therm-0.028
Token
Token ID198
Feature activation+4.73e-02

Pos logit contributions
 parties+0.028
 tea+0.024
 recess+0.021
 party+0.019
 celebrate+0.019
Neg logit contributions
 beast-0.029
 gut-0.029
 mistake-0.027
inct-0.027
 therm-0.027
TokenThe
Token ID464
Feature activation+3.72e+00

Pos logit contributions
 beast+0.001
 gut+0.001
 mistake+0.001
 therm+0.001
inct+0.001
Neg logit contributions
 parties-0.002
 tea-0.002
 recess-0.001
 celebrate-0.001
 party-0.001
Token av
Token ID1196
Feature activation+5.59e+00

Pos logit contributions
 gut+0.117
rum+0.114
inct+0.113
 beast+0.112
 outage+0.110
Neg logit contributions
 parties-0.127
 tea-0.116
 party-0.105
 picnic-0.100
 celebrate-0.099
Tokenoc
Token ID420
Feature activation+8.88e+00

Pos logit contributions
 gut+0.153
 mistake+0.137
 beast+0.137
 therm+0.129
 reckless+0.126
Neg logit contributions
 parties-0.214
 tea-0.195
 celebrate-0.179
ils-0.176
ets-0.174
Tokenados
Token ID22484
Feature activation+3.73e+00

Pos logit contributions
 outage+0.380
 gut+0.358
 severe+0.358
 swollen+0.353
 therm+0.352
Neg logit contributions
 parties-0.196
 tea-0.144
ils-0.143
 recess-0.121
 party-0.120
Token were
Token ID547
Feature activation+5.69e-01

Pos logit contributions
 gut+0.166
 beast+0.163
 mistake+0.156
inct+0.154
 therm+0.153
Neg logit contributions
 parties-0.086
 tea-0.070
 celebrate-0.060
 party-0.056
 picnic-0.052
Token shared
Token ID4888
Feature activation-3.24e+00

Pos logit contributions
 beast+0.022
 gut+0.021
 mistake+0.020
inct+0.020
 therm+0.020
Neg logit contributions
 parties-0.017
 tea-0.015
 celebrate-0.012
 recess-0.012
 party-0.011
Token with
Token ID351
Feature activation-2.74e+00

Pos logit contributions
 parties+0.096
 tea+0.084
 recess+0.072
 celebrate+0.066
 party+0.064
Neg logit contributions
 beast-0.118
 gut-0.118
 mistake-0.112
inct-0.109
 therm-0.109
Token everybody
Token ID7288
Feature activation-2.78e+00

Pos logit contributions
 parties+0.071
 tea+0.062
 celebrate+0.050
 recess+0.049
 party+0.044
Neg logit contributions
 beast-0.112
 gut-0.111
 therm-0.106
inct-0.105
 mistake-0.104
Token
Token ID198
Feature activation+2.52e-01

Pos logit contributions
 beast+0.011
 gut+0.011
 mistake+0.010
 therm+0.010
inct+0.010
Neg logit contributions
 parties-0.013
 tea-0.012
 recess-0.010
 celebrate-0.010
 party-0.010
Token
Token ID198
Feature activation+8.35e-01

Pos logit contributions
 beast+0.009
 gut+0.009
 mistake+0.009
inct+0.008
 therm+0.008
Neg logit contributions
 parties-0.008
 tea-0.007
 recess-0.006
 celebrate-0.005
 party-0.005
TokenThe
Token ID464
Feature activation+4.09e+00

Pos logit contributions
 beast+0.027
 gut+0.027
 mistake+0.026
 therm+0.025
inct+0.025
Neg logit contributions
 parties-0.028
 tea-0.025
 recess-0.022
 celebrate-0.021
 party-0.020
Token rh
Token ID9529
Feature activation+4.56e+00

Pos logit contributions
 gut+0.131
rum+0.117
inct+0.117
 beast+0.117
 severe+0.110
Neg logit contributions
 parties-0.140
 tea-0.131
 picnic-0.112
 celebrate-0.112
 party-0.112
Tokenin
Token ID259
Feature activation+4.92e+00

Pos logit contributions
 gut+0.188
iously+0.181
 plea+0.177
inct+0.175
 severe+0.174
Neg logit contributions
 parties-0.104
 tea-0.089
 recess-0.073
 together-0.069
acky-0.068
Tokenoc
Token ID420
Feature activation+8.87e+00

Pos logit contributions
 gut+0.165
 severe+0.144
 reckless+0.141
 adult+0.140
rum+0.139
Neg logit contributions
 tea-0.144
 parties-0.133
 played-0.132
 picnic-0.127
 play-0.124
Tokeneros
Token ID27498
Feature activation+4.28e+00

Pos logit contributions
 severe+0.299
 swollen+0.274
 outage+0.273
 gut+0.269
 reckless+0.269
Neg logit contributions
 parties-0.244
 tea-0.212
 recess-0.189
ils-0.186
 sandwiches-0.184
Token nodded
Token ID14464
Feature activation+1.95e+00

Pos logit contributions
 gut+0.165
 beast+0.153
 severe+0.151
 swollen+0.150
rum+0.150
Neg logit contributions
 parties-0.119
 tea-0.112
 played-0.091
 party-0.091
 celebrate-0.088
Token its
Token ID663
Feature activation+2.85e+00

Pos logit contributions
 gut+0.063
 beast+0.063
 mistake+0.060
 therm+0.059
 swollen+0.058
Neg logit contributions
 parties-0.069
 tea-0.058
 recess-0.053
 celebrate-0.048
 picnic-0.048
Token head
Token ID1182
Feature activation+3.32e+00

Pos logit contributions
 beast+0.089
cut+0.084
 gut+0.083
 passer+0.083
 therm+0.081
Neg logit contributions
 tea-0.081
 parties-0.078
 birthday-0.069
 play-0.069
 pretend-0.068
Token.
Token ID13
Feature activation+1.01e-01

Pos logit contributions
 beast+0.134
 gut+0.133
 mistake+0.126
inct+0.125
 therm+0.124
Neg logit contributions
 parties-0.086
 tea-0.075
 recess-0.059
 celebrate-0.057
 party-0.054
Token
Token ID198
Feature activation+1.89e-01

Pos logit contributions
 beast+0.044
 gut+0.043
 mistake+0.041
 swollen+0.041
 therm+0.041
Neg logit contributions
 parties-0.029
 tea-0.024
 recess-0.019
 celebrate-0.019
 picnic-0.018
Token
Token ID198
Feature activation+6.85e-01

Pos logit contributions
 beast+0.007
 gut+0.007
 mistake+0.006
inct+0.006
 therm+0.006
Neg logit contributions
 parties-0.006
 tea-0.005
 recess-0.004
 celebrate-0.004
 party-0.004
TokenThe
Token ID464
Feature activation+4.33e+00

Pos logit contributions
 beast+0.025
 gut+0.024
 mistake+0.023
 therm+0.023
inct+0.023
Neg logit contributions
 parties-0.021
 tea-0.017
 celebrate-0.016
 recess-0.015
 picnic-0.014
Token rh
Token ID9529
Feature activation+5.02e+00

Pos logit contributions
 gut+0.137
rum+0.129
inct+0.122
 outage+0.121
 beast+0.119
Neg logit contributions
 parties-0.140
 tea-0.140
 picnic-0.123
 party-0.121
 celebrate-0.115
Tokenin
Token ID259
Feature activation+5.29e+00

Pos logit contributions
 gut+0.213
 plea+0.196
inct+0.196
 therm+0.194
 severe+0.193
Neg logit contributions
 parties-0.100
 tea-0.087
acky-0.078
 together-0.075
P-0.074
Tokenoc
Token ID420
Feature activation+8.86e+00

Pos logit contributions
 gut+0.181
 outage+0.162
 reckless+0.162
 swollen+0.160
 vain+0.158
Neg logit contributions
 played-0.133
 tea-0.133
 together-0.123
acky-0.122
 picnic-0.121
Tokeneros
Token ID27498
Feature activation+4.80e+00

Pos logit contributions
 outage+0.297
 severe+0.296
 swollen+0.271
 therm+0.267
 gut+0.266
Neg logit contributions
 parties-0.230
ils-0.204
 tea-0.201
 sandwiches-0.180
 recess-0.170
Token couldn
Token ID3521
Feature activation+3.50e+00

Pos logit contributions
 gut+0.180
 beast+0.174
 swollen+0.171
 therm+0.169
 severe+0.165
Neg logit contributions
 parties-0.136
 tea-0.122
 played-0.105
 party-0.101
 celebrate-0.100
Token't
Token ID470
Feature activation+2.72e+00

Pos logit contributions
 gut+0.103
 beast+0.100
 swollen+0.093
 mistake+0.093
inct+0.091
Neg logit contributions
 parties-0.130
 tea-0.118
 celebrate-0.102
 recess-0.100
 wear-0.096
Token believe
Token ID1975
Feature activation-3.93e-01

Pos logit contributions
 beast+0.099
 gut+0.098
 mistake+0.093
 therm+0.090
cut+0.089
Neg logit contributions
 parties-0.082
 tea-0.073
 recess-0.061
 picnic-0.055
 celebrate-0.054
Token his
Token ID465
Feature activation+9.97e-01

Pos logit contributions
 parties+0.013
 tea+0.011
 recess+0.009
 celebrate+0.009
 party+0.009
Neg logit contributions
 beast-0.014
 gut-0.013
inct-0.013
 mistake-0.013
 therm-0.013
Token happy
Token ID3772
Feature activation+3.01e-01

Pos logit contributions
 beast+0.013
 gut+0.013
 mistake+0.012
inct+0.012
 therm+0.012
Neg logit contributions
 parties-0.010
 tea-0.009
 recess-0.007
 celebrate-0.007
 party-0.007
Token."
Token ID526
Feature activation+4.97e-01

Pos logit contributions
 beast+0.012
 gut+0.012
 mistake+0.011
 therm+0.011
inct+0.011
Neg logit contributions
 parties-0.008
 tea-0.007
 recess-0.006
 celebrate-0.006
 party-0.006
Token The
Token ID383
Feature activation+4.04e+00

Pos logit contributions
 gut+0.018
 beast+0.018
 mistake+0.017
 swollen+0.017
 therm+0.017
Neg logit contributions
 parties-0.015
 tea-0.013
 celebrate-0.011
 recess-0.011
 picnic-0.010
Token rh
Token ID9529
Feature activation+5.21e+00

Pos logit contributions
 gut+0.125
rum+0.114
 beast+0.113
inct+0.109
 severe+0.104
Neg logit contributions
 parties-0.142
 tea-0.137
 party-0.119
 picnic-0.119
 celebrate-0.117
Tokenin
Token ID259
Feature activation+5.57e+00

Pos logit contributions
 gut+0.213
inct+0.196
 plea+0.195
 therm+0.194
 beast+0.192
Neg logit contributions
 parties-0.119
 tea-0.102
 together-0.086
acky-0.085
 celebrate-0.083
Tokenoc
Token ID420
Feature activation+8.85e+00

Pos logit contributions
 gut+0.189
 vain+0.172
 reckless+0.168
 mistake+0.164
 swollen+0.164
Neg logit contributions
 played-0.145
 tea-0.143
 together-0.130
 celebrate-0.128
 play-0.127
Tokeneros
Token ID27498
Feature activation+4.98e+00

Pos logit contributions
 severe+0.280
 outage+0.267
 gut+0.258
 swollen+0.257
 therm+0.255
Neg logit contributions
 parties-0.268
 tea-0.228
ils-0.209
 sandwiches-0.202
 recess-0.196
Token did
Token ID750
Feature activation+2.55e+00

Pos logit contributions
 gut+0.189
 beast+0.183
 swollen+0.182
 therm+0.178
 mistake+0.174
Neg logit contributions
 parties-0.129
 tea-0.121
 played-0.110
 party-0.099
 celebrate-0.098
Token what
Token ID644
Feature activation-2.31e-01

Pos logit contributions
 beast+0.086
 gut+0.083
 swollen+0.081
 therm+0.080
cut+0.080
Neg logit contributions
 parties-0.084
 tea-0.075
 celebrate-0.063
 recess-0.061
 wear-0.060
Token the
Token ID262
Feature activation+1.54e+00

Pos logit contributions
 parties+0.006
 tea+0.005
 celebrate+0.004
 recess+0.004
 played+0.004
Neg logit contributions
 gut-0.009
 beast-0.009
inct-0.009
 swollen-0.009
 plea-0.009
Token bird
Token ID6512
Feature activation+3.55e+00

Pos logit contributions
 gut+0.046
 beast+0.042
rum+0.041
inct+0.039
 swollen+0.039
Neg logit contributions
 parties-0.056
 tea-0.054
 picnic-0.047
 party-0.047
 celebrate-0.046
Token
Token ID198
Feature activation-5.18e-01

Pos logit contributions
 parties+0.006
 tea+0.005
 recess+0.004
 celebrate+0.004
 played+0.004
Neg logit contributions
 beast-0.007
 gut-0.007
 mistake-0.007
 swollen-0.007
 severe-0.007
Token
Token ID198
Feature activation+2.82e-01

Pos logit contributions
 parties+0.014
 tea+0.012
 recess+0.009
 celebrate+0.009
 wear+0.008
Neg logit contributions
 beast-0.021
 gut-0.020
 mistake-0.019
 therm-0.019
 swollen-0.019
TokenThe
Token ID464
Feature activation+3.98e+00

Pos logit contributions
 beast+0.010
 mistake+0.010
 gut+0.010
 therm+0.010
 swollen+0.010
Neg logit contributions
 parties-0.008
 tea-0.007
 celebrate-0.006
P-0.006
 recess-0.006
Token rh
Token ID9529
Feature activation+4.83e+00

Pos logit contributions
 gut+0.120
rum+0.111
inct+0.108
 beast+0.107
 severe+0.101
Neg logit contributions
 parties-0.139
 tea-0.132
 picnic-0.119
 party-0.116
 celebrate-0.116
Tokenin
Token ID259
Feature activation+5.04e+00

Pos logit contributions
 gut+0.198
inct+0.183
 plea+0.182
 therm+0.180
iously+0.177
Neg logit contributions
 parties-0.107
 tea-0.090
acky-0.078
 together-0.076
 celebrate-0.075
Tokenoc
Token ID420
Feature activation+8.85e+00

Pos logit contributions
 gut+0.172
 outage+0.157
 reckless+0.153
 vain+0.151
rier+0.149
Neg logit contributions
 played-0.125
 tea-0.121
 picnic-0.117
 play-0.113
 parties-0.113
Tokeneros
Token ID27498
Feature activation+4.26e+00

Pos logit contributions
 outage+0.277
 severe+0.276
 swollen+0.258
 therm+0.257
 gut+0.257
Neg logit contributions
 parties-0.260
 tea-0.220
ils-0.211
 sandwiches-0.199
 recess-0.197
Token was
Token ID373
Feature activation+1.91e+00

Pos logit contributions
 gut+0.163
 beast+0.156
 therm+0.153
 swollen+0.153
inct+0.149
Neg logit contributions
 parties-0.112
 tea-0.102
 played-0.091
 celebrate-0.086
 party-0.085
Token so
Token ID523
Feature activation+1.45e+00

Pos logit contributions
 beast+0.074
 gut+0.073
inct+0.071
 therm+0.068
 mistake+0.068
Neg logit contributions
 parties-0.053
 tea-0.048
 celebrate-0.039
 party-0.036
 picnic-0.036
Token delighted
Token ID20707
Feature activation+2.57e+00

Pos logit contributions
 beast+0.059
 gut+0.057
inct+0.055
 mistake+0.054
 therm+0.054
Neg logit contributions
 parties-0.039
 tea-0.034
 celebrate-0.029
 wear-0.026
 picnic-0.026
Token with
Token ID351
Feature activation+6.15e-01

Pos logit contributions
 gut+0.097
 beast+0.096
 swollen+0.092
 mistake+0.092
inct+0.090
Neg logit contributions
 parties-0.074
 tea-0.064
 recess-0.051
 celebrate-0.050
 party-0.050
Token help
Token ID1037
Feature activation+6.02e-01

Pos logit contributions
 gut+0.163
 beast+0.161
inct+0.155
 swollen+0.150
 mistake+0.149
Neg logit contributions
 parties-0.117
 tea-0.102
 celebrate-0.086
 recess-0.081
 party-0.080
Token.
Token ID13
Feature activation+1.12e-01

Pos logit contributions
 gut+0.028
 beast+0.028
inct+0.028
 swollen+0.027
 guardian+0.026
Neg logit contributions
 parties-0.012
 tea-0.009
 celebrate-0.007
 wear-0.006
 recess-0.006
Token The
Token ID383
Feature activation+3.64e+00

Pos logit contributions
 gut+0.005
 beast+0.005
 mistake+0.005
 swollen+0.005
 guardian+0.005
Neg logit contributions
 parties-0.002
 tea-0.002
 played-0.001
 celebrate-0.001
 together-0.001
Token rh
Token ID9529
Feature activation+5.15e+00

Pos logit contributions
 gut+0.104
 beast+0.096
inct+0.095
rum+0.090
 therm+0.090
Neg logit contributions
 parties-0.139
 tea-0.128
 party-0.111
 recess-0.110
 picnic-0.109
Tokenin
Token ID259
Feature activation+5.22e+00

Pos logit contributions
 gut+0.203
 therm+0.186
inct+0.183
 beast+0.183
 plea+0.182
Neg logit contributions
 parties-0.137
 tea-0.115
 together-0.097
 celebrate-0.095
 party-0.095
Tokenoc
Token ID420
Feature activation+8.84e+00

Pos logit contributions
 gut+0.183
 reckless+0.157
 swollen+0.155
 outage+0.152
 mistake+0.152
Neg logit contributions
 tea-0.136
 parties-0.133
 played-0.130
acky-0.120
 picnic-0.119
Tokeneros
Token ID27498
Feature activation+4.77e+00

Pos logit contributions
 severe+0.255
 gut+0.243
 swollen+0.242
 therm+0.239
 outage+0.238
Neg logit contributions
 parties-0.309
 tea-0.270
ils-0.249
 recess-0.247
 sandwiches-0.232
Token was
Token ID373
Feature activation+2.24e+00

Pos logit contributions
 gut+0.205
 outage+0.191
 beast+0.190
rum+0.187
 swollen+0.183
Neg logit contributions
 played-0.099
 tea-0.092
 parties-0.089
 shared-0.080
 together-0.079
Token very
Token ID845
Feature activation+3.48e+00

Pos logit contributions
 beast+0.088
 gut+0.088
inct+0.086
 mistake+0.080
 adult+0.079
Neg logit contributions
 parties-0.056
 tea-0.055
 together-0.044
 celebrate-0.043
 played-0.043
Token happy
Token ID3772
Feature activation+8.70e-01

Pos logit contributions
 beast+0.143
 gut+0.133
inct+0.132
der+0.124
 therm+0.123
Neg logit contributions
 parties-0.084
 tea-0.079
 wear-0.067
 celebrate-0.066
 together-0.066
Token and
Token ID290
Feature activation+7.68e-01

Pos logit contributions
 gut+0.035
 beast+0.035
inct+0.034
 swollen+0.034
 severe+0.033
Neg logit contributions
 parties-0.023
 tea-0.020
 party-0.015
 together-0.015
 recess-0.015
Token,"
Token ID553
Feature activation-3.38e-02

Pos logit contributions
 gut+0.115
 beast+0.115
inct+0.110
 therm+0.108
 mistake+0.108
Neg logit contributions
 parties-0.074
 tea-0.063
 recess-0.049
 celebrate-0.048
 party-0.046
Token said
Token ID531
Feature activation-7.64e-01

Pos logit contributions
 parties+0.001
 tea+0.001
 recess+0.001
 celebrate+0.001
 party+0.001
Neg logit contributions
 beast-0.001
 gut-0.001
 mistake-0.001
inct-0.001
 therm-0.001
Token the
Token ID262
Feature activation+3.56e+00

Pos logit contributions
 parties+0.021
 tea+0.018
 recess+0.015
 celebrate+0.014
 party+0.014
Neg logit contributions
 beast-0.030
 gut-0.030
 mistake-0.028
 therm-0.028
inct-0.028
Token rh
Token ID9529
Feature activation+4.77e+00

Pos logit contributions
 gut+0.125
rum+0.116
 severe+0.107
 outage+0.107
inct+0.107
Neg logit contributions
 parties-0.100
 tea-0.094
 celebrate-0.085
 party-0.084
 picnic-0.082
Tokenin
Token ID259
Feature activation+5.39e+00

Pos logit contributions
 gut+0.195
inct+0.183
 plea+0.181
 mistake+0.178
 swollen+0.177
Neg logit contributions
 parties-0.107
 tea-0.091
acky-0.075
 recess-0.073
 together-0.073
Tokenoc
Token ID420
Feature activation+8.84e+00

Pos logit contributions
 gut+0.185
 swollen+0.174
 shattering+0.161
 refusing+0.160
 severe+0.159
Neg logit contributions
 together-0.124
 tea-0.124
 parties-0.119
 played-0.114
 named-0.114
Tokeneros
Token ID27498
Feature activation+4.79e+00

Pos logit contributions
 severe+0.289
 outage+0.278
 swollen+0.270
 gut+0.267
 confusion+0.263
Neg logit contributions
 parties-0.229
 tea-0.193
ils-0.190
 sandwiches-0.171
 recess-0.170
Token.
Token ID13
Feature activation+3.73e-01

Pos logit contributions
 gut+0.216
 beast+0.214
inct+0.206
 swollen+0.205
 mistake+0.204
Neg logit contributions
 parties-0.084
 tea-0.079
 together-0.053
 played-0.051
 celebrate-0.050
Token "
Token ID366
Feature activation+3.55e+00

Pos logit contributions
 beast+0.011
 gut+0.011
 mistake+0.010
 swollen+0.010
cut+0.010
Neg logit contributions
 parties-0.014
 tea-0.013
 celebrate-0.011
 recess-0.011
 party-0.010
TokenI
Token ID40
Feature activation-2.26e+00

Pos logit contributions
 beast+0.129
 gut+0.127
 mistake+0.123
 therm+0.118
 swollen+0.116
Neg logit contributions
 parties-0.104
 tea-0.092
 celebrate-0.080
 recess-0.078
 wear-0.073
Token'm
Token ID1101
Feature activation+4.57e-01

Pos logit contributions
 parties+0.064
 tea+0.060
 celebrate+0.051
 wear+0.050
 picnic+0.047
Neg logit contributions
 beast-0.081
 gut-0.081
 swollen-0.077
 passer-0.076
 mistake-0.075
TokenSuddenly
Token ID38582
Feature activation+2.24e+00

Pos logit contributions
 beast+0.007
 gut+0.007
 therm+0.007
 swollen+0.007
 mistake+0.007
Neg logit contributions
 parties-0.009
 tea-0.007
 recess-0.007
 celebrate-0.007
 party-0.006
Token,
Token ID11
Feature activation+1.06e+00

Pos logit contributions
 gut+0.094
 beast+0.093
 mistake+0.090
 swollen+0.089
inct+0.088
Neg logit contributions
 parties-0.054
 tea-0.047
 recess-0.037
 celebrate-0.034
 picnic-0.033
Token the
Token ID262
Feature activation+4.31e+00

Pos logit contributions
 gut+0.042
 mistake+0.041
 beast+0.040
 swollen+0.040
rum+0.039
Neg logit contributions
 parties-0.029
 tea-0.026
 recess-0.021
 celebrate-0.020
 picnic-0.019
Token rh
Token ID9529
Feature activation+5.31e+00

Pos logit contributions
 gut+0.133
rum+0.124
 mistake+0.120
 beast+0.119
inct+0.116
Neg logit contributions
 parties-0.152
 tea-0.141
 picnic-0.126
 party-0.124
 celebrate-0.124
Tokenin
Token ID259
Feature activation+5.39e+00

Pos logit contributions
 gut+0.224
 plea+0.208
 mistake+0.204
iously+0.203
inct+0.203
Neg logit contributions
 parties-0.117
 tea-0.097
acky-0.083
 together-0.080
 celebrate-0.078
Tokenoc
Token ID420
Feature activation+8.84e+00

Pos logit contributions
 gut+0.190
rum+0.175
rier+0.172
 mistake+0.172
 vain+0.171
Neg logit contributions
 played-0.136
 tea-0.134
 picnic-0.125
 party-0.120
 play-0.120
Tokeneros
Token ID27498
Feature activation+4.66e+00

Pos logit contributions
 severe+0.279
 swollen+0.266
 outage+0.263
 gut+0.253
 reckless+0.253
Neg logit contributions
 parties-0.276
 tea-0.231
ils-0.211
 recess-0.205
 celebrate-0.201
Token saw
Token ID2497
Feature activation+2.71e+00

Pos logit contributions
 gut+0.180
rum+0.170
 swollen+0.166
 therm+0.165
cut+0.165
Neg logit contributions
 parties-0.120
 tea-0.112
 played-0.097
 party-0.095
 picnic-0.092
Token something
Token ID1223
Feature activation+2.33e+00

Pos logit contributions
 gut+0.117
inct+0.112
 beast+0.111
 swollen+0.110
 mistake+0.110
Neg logit contributions
 parties-0.064
 tea-0.058
 recess-0.042
 celebrate-0.039
 party-0.039
Token in
Token ID287
Feature activation+2.49e+00

Pos logit contributions
 beast+0.075
 gut+0.075
 mistake+0.070
 therm+0.069
inct+0.069
Neg logit contributions
 parties-0.080
 tea-0.071
 recess-0.061
 celebrate-0.059
 party-0.057
Token the
Token ID262
Feature activation-9.66e-01

Pos logit contributions
 gut+0.114
inct+0.106
iously+0.105
ache+0.105
ometer+0.104
Neg logit contributions
 parties-0.051
 tea-0.046
 recess-0.038
 picnic-0.032
 celebrate-0.032
TokenFinally
Token ID11158
Feature activation+2.85e+00

Pos logit contributions
 beast+0.028
 gut+0.027
 mistake+0.025
 therm+0.025
inct+0.025
Neg logit contributions
 parties-0.028
 tea-0.025
 recess-0.021
 celebrate-0.021
 party-0.020
Token,
Token ID11
Feature activation+5.75e-01

Pos logit contributions
 beast+0.077
 gut+0.076
 mistake+0.073
 therm+0.071
 passer+0.069
Neg logit contributions
 parties-0.105
 tea-0.086
ets-0.085
 outfits-0.082
 recess-0.082
Token the
Token ID262
Feature activation+3.78e+00

Pos logit contributions
 beast+0.023
 gut+0.023
rum+0.021
 mistake+0.021
inct+0.021
Neg logit contributions
 parties-0.016
 tea-0.015
 recess-0.011
 celebrate-0.011
 party-0.011
Token rh
Token ID9529
Feature activation+5.12e+00

Pos logit contributions
 gut+0.153
 beast+0.143
rum+0.143
 severe+0.140
 mistake+0.139
Neg logit contributions
 parties-0.099
 tea-0.094
 picnic-0.080
 celebrate-0.078
 party-0.077
Tokenin
Token ID259
Feature activation+5.10e+00

Pos logit contributions
 gut+0.210
 plea+0.192
inct+0.191
 mistake+0.190
 swollen+0.189
Neg logit contributions
 parties-0.126
 tea-0.107
 celebrate-0.086
acky-0.086
 party-0.085
Tokenoc
Token ID420
Feature activation+8.83e+00

Pos logit contributions
 gut+0.164
 swollen+0.148
 mistake+0.145
 vain+0.143
 reckless+0.142
Neg logit contributions
 tea-0.163
 parties-0.150
 played-0.150
 picnic-0.145
 celebrate-0.141
Tokeneros
Token ID27498
Feature activation+4.36e+00

Pos logit contributions
 severe+0.276
 swollen+0.263
 reckless+0.248
 terrified+0.246
 outage+0.244
Neg logit contributions
 parties-0.285
 tea-0.250
ils-0.220
 recess-0.219
 birthday-0.216
Token remembered
Token ID12086
Feature activation+2.19e+00

Pos logit contributions
 gut+0.166
 beast+0.160
 swollen+0.158
 therm+0.156
inct+0.153
Neg logit contributions
 parties-0.123
 tea-0.109
 party-0.090
 celebrate-0.087
 picnic-0.087
Token that
Token ID326
Feature activation-1.34e+00

Pos logit contributions
 beast+0.097
inct+0.095
 gut+0.092
cut+0.092
 passer+0.089
Neg logit contributions
 parties-0.057
 tea-0.049
 party-0.036
 celebrate-0.033
 recess-0.033
Token animals
Token ID4695
Feature activation-4.68e-01

Pos logit contributions
 parties+0.032
 tea+0.029
 celebrate+0.022
 party+0.020
 together+0.019
Neg logit contributions
 gut-0.059
 beast-0.058
inct-0.057
 therm-0.054
 severe-0.054
Token often
Token ID1690
Feature activation+1.30e+00

Pos logit contributions
 parties+0.012
 tea+0.010
 celebrate+0.008
 party+0.007
 recess+0.007
Neg logit contributions
 gut-0.020
 beast-0.020
inct-0.019
 therm-0.019
 swollen-0.018
Token name
Token ID1438
Feature activation+2.83e-02

Pos logit contributions
 parties+0.008
 tea+0.007
 picnic+0.005
 birthday+0.005
 party+0.005
Neg logit contributions
 gut-0.010
 swollen-0.010
 severe-0.010
 beast-0.010
inct-0.010
Token?"
Token ID1701
Feature activation+2.01e+00

Pos logit contributions
 beast+0.001
 gut+0.001
 mistake+0.001
inct+0.001
 therm+0.001
Neg logit contributions
 parties-0.001
 tea-0.001
 recess-0.001
 celebrate-0.001
 party-0.001
Token The
Token ID383
Feature activation+3.98e+00

Pos logit contributions
 gut+0.081
 beast+0.078
 mistake+0.077
 swollen+0.076
inct+0.074
Neg logit contributions
 parties-0.055
 tea-0.048
 recess-0.039
 celebrate-0.039
 party-0.036
Token rh
Token ID9529
Feature activation+4.92e+00

Pos logit contributions
 gut+0.140
rum+0.127
inct+0.126
 severe+0.121
iously+0.119
Neg logit contributions
 parties-0.127
 tea-0.120
 party-0.104
 celebrate-0.102
 picnic-0.102
Tokenin
Token ID259
Feature activation+5.61e+00

Pos logit contributions
 gut+0.213
 plea+0.196
inct+0.195
iously+0.193
 severe+0.192
Neg logit contributions
 parties-0.106
 tea-0.088
acky-0.074
 together-0.073
 celebrate-0.068
Tokenoc
Token ID420
Feature activation+8.81e+00

Pos logit contributions
 gut+0.204
 swollen+0.177
 reckless+0.175
 severe+0.175
 outage+0.171
Neg logit contributions
 played-0.140
 tea-0.138
 parties-0.134
 play-0.130
 together-0.125
Tokeneros
Token ID27498
Feature activation+5.14e+00

Pos logit contributions
 severe+0.294
 outage+0.279
 gut+0.273
 swollen+0.272
 confusion+0.265
Neg logit contributions
 parties-0.253
 tea-0.211
ils-0.199
 sandwiches-0.190
 recess-0.181
Token was
Token ID373
Feature activation+2.50e+00

Pos logit contributions
 gut+0.221
 beast+0.201
 severe+0.200
 swollen+0.198
rum+0.198
Neg logit contributions
 parties-0.118
 tea-0.105
 played-0.097
 party-0.088
 picnic-0.083
Token surprised
Token ID6655
Feature activation+3.34e+00

Pos logit contributions
 gut+0.085
 beast+0.084
inct+0.080
 mistake+0.078
 therm+0.077
Neg logit contributions
 parties-0.081
 tea-0.074
 recess-0.059
 celebrate-0.058
 party-0.058
Token that
Token ID326
Feature activation-1.17e-01

Pos logit contributions
 beast+0.108
 gut+0.108
 mistake+0.101
 therm+0.098
inct+0.098
Neg logit contributions
 parties-0.115
 tea-0.103
 recess-0.090
 celebrate-0.086
 picnic-0.083
Token the
Token ID262
Feature activation+2.76e+00

Pos logit contributions
 parties+0.003
 tea+0.003
 celebrate+0.002
 recess+0.002
 party+0.002
Neg logit contributions
 gut-0.005
 swollen-0.005
inct-0.005
 beast-0.005
 mistake-0.005
Token
Token ID198
Feature activation-4.68e-01

Pos logit contributions
 parties+0.006
 tea+0.005
 recess+0.004
 celebrate+0.004
 picnic+0.004
Neg logit contributions
 gut-0.010
 beast-0.010
 mistake-0.010
 swollen-0.010
 therm-0.009
Token
Token ID198
Feature activation+3.52e-01

Pos logit contributions
 parties+0.016
 tea+0.014
 recess+0.012
 party+0.011
 celebrate+0.011
Neg logit contributions
 beast-0.016
 gut-0.015
 mistake-0.014
 therm-0.014
inct-0.014
TokenThe
Token ID464
Feature activation+3.88e+00

Pos logit contributions
 gut+0.011
 beast+0.011
 mistake+0.011
 therm+0.010
inct+0.010
Neg logit contributions
 parties-0.012
 tea-0.011
 recess-0.009
 celebrate-0.009
 picnic-0.009
Token rh
Token ID9529
Feature activation+4.82e+00

Pos logit contributions
 gut+0.122
rum+0.116
inct+0.111
 outage+0.110
 beast+0.107
Neg logit contributions
 parties-0.125
 tea-0.120
 party-0.108
 picnic-0.106
 celebrate-0.106
Tokenin
Token ID259
Feature activation+5.14e+00

Pos logit contributions
 gut+0.213
 plea+0.199
iously+0.197
inct+0.194
 swollen+0.193
Neg logit contributions
 parties-0.083
 tea-0.065
 together-0.065
acky-0.062
P-0.061
Tokenoc
Token ID420
Feature activation+8.80e+00

Pos logit contributions
 gut+0.184
 swollen+0.176
 outage+0.174
 unstable+0.169
 vain+0.168
Neg logit contributions
 played-0.106
 celebrate-0.094
 play-0.093
 together-0.092
 parties-0.091
Tokeneros
Token ID27498
Feature activation+4.39e+00

Pos logit contributions
 outage+0.297
 severe+0.297
 swollen+0.285
 therm+0.269
 gut+0.265
Neg logit contributions
 parties-0.228
ils-0.191
 tea-0.181
 recess-0.168
 celebrate-0.161
Token was
Token ID373
Feature activation+1.63e+00

Pos logit contributions
 gut+0.165
 swollen+0.162
 therm+0.159
 beast+0.158
rum+0.153
Neg logit contributions
 parties-0.116
 tea-0.101
 played-0.095
 party-0.089
 celebrate-0.088
Token so
Token ID523
Feature activation+1.40e+00

Pos logit contributions
 beast+0.065
 gut+0.064
inct+0.062
 therm+0.062
cut+0.060
Neg logit contributions
 parties-0.041
 tea-0.037
 celebrate-0.032
 together-0.029
 party-0.028
Token excited
Token ID6568
Feature activation+2.80e+00

Pos logit contributions
 beast+0.062
 gut+0.059
 therm+0.058
 mistake+0.058
inct+0.058
Neg logit contributions
 parties-0.029
 celebrate-0.024
 tea-0.024
 together-0.022
 played-0.021
Token that
Token ID326
Feature activation+4.55e-02

Pos logit contributions
 beast+0.102
 gut+0.102
 mistake+0.096
inct+0.095
 therm+0.095
Neg logit contributions
 parties-0.085
 tea-0.074
 recess-0.062
 celebrate-0.059
 party-0.057
Token y
Token ID331
Feature activation-3.05e-01

Pos logit contributions
 parties+0.087
 tea+0.076
 recess+0.064
 celebrate+0.061
 party+0.059
Neg logit contributions
 beast-0.105
 gut-0.104
 mistake-0.098
 therm-0.097
inct-0.097
Tokenummy
Token ID13513
Feature activation-3.16e+00

Pos logit contributions
 parties+0.011
 tea+0.010
 recess+0.009
 celebrate+0.009
 costumes+0.009
Neg logit contributions
 therm-0.008
 gut-0.008
 beast-0.008
 mistake-0.008
 severe-0.007
Token spaghetti
Token ID48241
Feature activation-1.69e+00

Pos logit contributions
 parties+0.101
 tea+0.086
 recess+0.082
 celebrate+0.073
 wear+0.070
Neg logit contributions
 beast-0.104
 gut-0.101
 therm-0.099
 mistake-0.097
 passer-0.095
Token with
Token ID351
Feature activation-3.13e+00

Pos logit contributions
 parties+0.050
 tea+0.043
 recess+0.040
 celebrate+0.036
 wear+0.034
Neg logit contributions
 beast-0.057
 gut-0.056
 mistake-0.054
 therm-0.053
cut-0.053
Token meat
Token ID6174
Feature activation-1.47e+00

Pos logit contributions
 parties+0.093
 tea+0.083
 recess+0.065
 celebrate+0.065
 party+0.062
Neg logit contributions
 beast-0.119
 gut-0.119
inct-0.112
 mistake-0.111
 swollen-0.108
Tokenballs
Token ID21591
Feature activation-9.06e+00

Pos logit contributions
 parties+0.041
 tea+0.036
 recess+0.030
 celebrate+0.028
 party+0.027
Neg logit contributions
 beast-0.056
 gut-0.056
 mistake-0.053
 therm-0.052
inct-0.052
Token."
Token ID526
Feature activation+1.97e-01

Pos logit contributions
 parties+0.273
 tea+0.241
 recess+0.208
 celebrate+0.196
 picnic+0.189
Neg logit contributions
 beast-0.315
 gut-0.313
 therm-0.296
 mistake-0.295
cut-0.290
Token Lily
Token ID20037
Feature activation-8.13e-01

Pos logit contributions
 mistake+0.004
inct+0.003
 therm+0.003
Start+0.003
 Fear+0.003
Neg logit contributions
 parties-0.008
 recess-0.007
ils-0.007
 seas-0.007
 camping-0.007
Token was
Token ID373
Feature activation+1.78e+00

Pos logit contributions
 parties+0.029
 tea+0.026
 recess+0.022
 celebrate+0.021
 party+0.021
Neg logit contributions
 beast-0.025
 gut-0.025
 mistake-0.023
 therm-0.023
inct-0.023
Token so
Token ID523
Feature activation+1.19e+00

Pos logit contributions
 beast+0.066
 gut+0.064
inct+0.061
 mistake+0.060
 therm+0.059
Neg logit contributions
 parties-0.054
 tea-0.048
 celebrate-0.039
 recess-0.038
 party-0.037
Token happy
Token ID3772
Feature activation+1.30e+00

Pos logit contributions
 gut+0.038
 beast+0.037
 mistake+0.036
 therm+0.036
 swollen+0.035
Neg logit contributions
 parties-0.040
 tea-0.035
 recess-0.033
 outfits-0.028
 celebrate-0.028
Token Lily
Token ID20037
Feature activation-2.02e+00

Pos logit contributions
 parties+0.022
ils+0.019
 recess+0.019
 tea+0.018
ets+0.017
Neg logit contributions
 therm-0.014
 mistake-0.014
 gut-0.014
 passer-0.013
inct-0.013
Token loved
Token ID6151
Feature activation-2.99e-01

Pos logit contributions
 parties+0.065
 tea+0.056
 recess+0.054
 celebrate+0.045
 picnic+0.043
Neg logit contributions
 gut-0.067
 beast-0.066
 mistake-0.064
 therm-0.064
cut-0.062
Token spaghetti
Token ID48241
Feature activation-1.86e+00

Pos logit contributions
 parties+0.009
 tea+0.007
 recess+0.006
 celebrate+0.006
 party+0.006
Neg logit contributions
 beast-0.011
 gut-0.011
 mistake-0.010
 therm-0.010
 swollen-0.010
Token with
Token ID351
Feature activation-3.18e+00

Pos logit contributions
 parties+0.060
 recess+0.054
 tea+0.051
 celebrate+0.048
 wear+0.046
Neg logit contributions
 beast-0.053
 therm-0.052
 mistake-0.051
 gut-0.051
 passer-0.048
Token meat
Token ID6174
Feature activation-1.54e+00

Pos logit contributions
 parties+0.084
 tea+0.069
 recess+0.060
 celebrate+0.056
 party+0.053
Neg logit contributions
 beast-0.125
 gut-0.124
 therm-0.121
 swollen-0.119
 mistake-0.118
Tokenballs
Token ID21591
Feature activation-8.65e+00

Pos logit contributions
 parties+0.046
 tea+0.040
 recess+0.034
 celebrate+0.033
 party+0.031
Neg logit contributions
 beast-0.056
 gut-0.055
 mistake-0.052
 therm-0.052
 swollen-0.051
Token.
Token ID13
Feature activation-1.22e+00

Pos logit contributions
 parties+0.282
 tea+0.248
 recess+0.220
 celebrate+0.207
 picnic+0.200
Neg logit contributions
 beast-0.282
 gut-0.280
 therm-0.271
 mistake-0.263
cut-0.260
Token She
Token ID1375
Feature activation-1.75e-01

Pos logit contributions
 parties+0.047
 tea+0.042
 recess+0.040
ils+0.039
 celebrate+0.037
Neg logit contributions
 gut-0.030
 therm-0.030
 mistake-0.028
inct-0.028
 Owl-0.027
Token watched
Token ID7342
Feature activation-1.13e+00

Pos logit contributions
 parties+0.006
 recess+0.005
 tea+0.005
 celebrate+0.004
dates+0.004
Neg logit contributions
 gut-0.006
 beast-0.006
 therm-0.006
 mistake-0.006
cut-0.005
Token as
Token ID355
Feature activation-3.93e-01

Pos logit contributions
 parties+0.033
 tea+0.027
 recess+0.025
 picnic+0.023
 party+0.022
Neg logit contributions
 gut-0.040
 beast-0.040
 therm-0.039
 mistake-0.038
cut-0.037
Token her
Token ID607
Feature activation+3.36e-02

Pos logit contributions
 parties+0.009
 tea+0.008
 recess+0.006
 celebrate+0.006
 party+0.006
Neg logit contributions
 beast-0.017
 gut-0.017
 mistake-0.016
 therm-0.016
 swollen-0.016
Token food
Token ID2057
Feature activation+7.44e-01

Pos logit contributions
 parties+0.050
 tea+0.043
 recess+0.035
 celebrate+0.033
 party+0.032
Neg logit contributions
 beast-0.073
 gut-0.072
 mistake-0.068
inct-0.068
 therm-0.068
Token,
Token ID11
Feature activation-2.21e+00

Pos logit contributions
 beast+0.027
 gut+0.027
 therm+0.026
 mistake+0.026
cut+0.025
Neg logit contributions
 parties-0.022
 tea-0.018
 recess-0.017
 celebrate-0.015
 picnic-0.014
Token spaghetti
Token ID48241
Feature activation-6.29e-01

Pos logit contributions
 parties+0.069
 tea+0.062
 recess+0.050
 celebrate+0.049
 party+0.048
Neg logit contributions
 beast-0.079
 gut-0.078
inct-0.073
 mistake-0.073
 swollen-0.072
Token with
Token ID351
Feature activation-2.88e+00

Pos logit contributions
 parties+0.019
 tea+0.016
 recess+0.015
 celebrate+0.014
 outfits+0.013
Neg logit contributions
 beast-0.021
 therm-0.021
 mistake-0.021
 gut-0.020
cut-0.020
Token meat
Token ID6174
Feature activation-1.20e+00

Pos logit contributions
 parties+0.087
 tea+0.075
 recess+0.065
 celebrate+0.061
 party+0.059
Neg logit contributions
 beast-0.104
 gut-0.103
 therm-0.098
 mistake-0.097
 swollen-0.096
Tokenballs
Token ID21591
Feature activation-8.54e+00

Pos logit contributions
 parties+0.034
 tea+0.029
 recess+0.025
 celebrate+0.023
 party+0.022
Neg logit contributions
 beast-0.045
 gut-0.045
 mistake-0.043
 therm-0.043
inct-0.042
Token.
Token ID13
Feature activation-7.22e-01

Pos logit contributions
 parties+0.261
 tea+0.227
 recess+0.202
 celebrate+0.192
 picnic+0.180
Neg logit contributions
 beast-0.285
 gut-0.281
 therm-0.278
cut-0.269
 mistake-0.269
Token It
Token ID632
Feature activation+3.32e+00

Pos logit contributions
 parties+0.021
 tea+0.016
 recess+0.016
ils+0.014
 celebrate+0.014
Neg logit contributions
 therm-0.026
 gut-0.025
 mistake-0.025
inct-0.025
 swollen-0.025
Token smelled
Token ID37609
Feature activation+2.40e+00

Pos logit contributions
 gut+0.109
 beast+0.107
 mistake+0.099
inct+0.098
 therm+0.095
Neg logit contributions
 parties-0.114
 tea-0.106
 celebrate-0.086
 recess-0.085
 party-0.084
Token so
Token ID523
Feature activation+6.26e-01

Pos logit contributions
 therm+0.064
 beast+0.064
 mistake+0.062
 gut+0.061
 severe+0.058
Neg logit contributions
 parties-0.087
 recess-0.080
 tea-0.072
 outfits-0.068
dates-0.068
Token delicious
Token ID12625
Feature activation+1.25e+00

Pos logit contributions
 gut+0.019
 mistake+0.019
 therm+0.018
 swollen+0.018
 beast+0.018
Neg logit contributions
 parties-0.021
 recess-0.018
 tea-0.017
 outfits-0.015
 celebrate-0.015
Token food
Token ID2057
Feature activation+8.07e-01

Pos logit contributions
 parties+0.035
 tea+0.029
 recess+0.025
 celebrate+0.024
 party+0.022
Neg logit contributions
 beast-0.045
 gut-0.045
 mistake-0.043
 therm-0.043
 swollen-0.042
Token -
Token ID532
Feature activation+2.64e-01

Pos logit contributions
 beast+0.031
 gut+0.031
 mistake+0.029
 therm+0.029
 swollen+0.028
Neg logit contributions
 parties-0.023
 tea-0.019
 recess-0.017
 celebrate-0.015
 picnic-0.014
Token spaghetti
Token ID48241
Feature activation-7.95e-01

Pos logit contributions
 beast+0.010
 gut+0.010
inct+0.010
 mistake+0.010
 swollen+0.009
Neg logit contributions
 parties-0.008
 tea-0.007
 recess-0.005
 celebrate-0.005
 picnic-0.005
Token with
Token ID351
Feature activation-2.57e+00

Pos logit contributions
 parties+0.025
 tea+0.021
 recess+0.020
 celebrate+0.018
 outfits+0.017
Neg logit contributions
 beast-0.027
 therm-0.025
 gut-0.025
 mistake-0.025
cut-0.024
Token meat
Token ID6174
Feature activation-1.33e+00

Pos logit contributions
 parties+0.081
 tea+0.068
 recess+0.064
 celebrate+0.058
 party+0.056
Neg logit contributions
 beast-0.086
 gut-0.084
 therm-0.083
 swollen-0.082
 mistake-0.080
Tokenballs
Token ID21591
Feature activation-8.46e+00

Pos logit contributions
 parties+0.038
 tea+0.033
 recess+0.028
 celebrate+0.027
 party+0.025
Neg logit contributions
 beast-0.049
 gut-0.049
 mistake-0.046
 therm-0.046
inct-0.045
Token.
Token ID13
Feature activation-6.21e-01

Pos logit contributions
 parties+0.247
 tea+0.213
 recess+0.183
 celebrate+0.173
 picnic+0.163
Neg logit contributions
 beast-0.308
 gut-0.304
 therm-0.294
 mistake-0.290
cut-0.284
Token
Token ID198
Feature activation-6.39e-01

Pos logit contributions
 parties+0.020
 tea+0.016
 recess+0.016
ils+0.015
 celebrate+0.014
Neg logit contributions
 therm-0.019
 gut-0.019
inct-0.019
 mistake-0.019
 swollen-0.019
Token
Token ID198
Feature activation+2.13e-01

Pos logit contributions
 parties+0.027
 tea+0.024
 recess+0.023
ils+0.022
 party+0.021
Neg logit contributions
 mistake-0.013
 therm-0.013
 gut-0.013
 beast-0.012
 severe-0.012
TokenL
Token ID43
Feature activation-7.36e-01

Pos logit contributions
 mistake+0.006
 gut+0.006
 therm+0.006
 swollen+0.006
 severe+0.006
Neg logit contributions
 parties-0.007
 tea-0.006
 recess-0.006
 celebrate-0.005
 party-0.005
Tokenily
Token ID813
Feature activation-8.11e-01

Pos logit contributions
 parties+0.019
 tea+0.016
 celebrate+0.014
 recess+0.014
 picnic+0.014
Neg logit contributions
 mistake-0.028
 adult-0.027
 beast-0.026
 gut-0.026
rum-0.025
Token make
Token ID787
Feature activation-2.00e+00

Pos logit contributions
 parties+0.058
 tea+0.053
 celebrate+0.041
 wear+0.039
 picnic+0.039
Neg logit contributions
 beast-0.086
 gut-0.084
inct-0.084
 passer-0.081
 mistake-0.080
Token snow
Token ID6729
Feature activation-2.48e+00

Pos logit contributions
 parties+0.065
 tea+0.055
 recess+0.053
 celebrate+0.047
 party+0.046
Neg logit contributions
 beast-0.066
 gut-0.066
 mistake-0.063
 therm-0.061
inct-0.060
Tokenmen
Token ID3653
Feature activation-1.57e+00

Pos logit contributions
 parties+0.089
 recess+0.077
 tea+0.076
 celebrate+0.067
 wear+0.066
Neg logit contributions
 beast-0.070
 gut-0.069
cut-0.066
 therm-0.066
 mistake-0.066
Token and
Token ID290
Feature activation-2.64e+00

Pos logit contributions
 parties+0.047
 tea+0.041
 recess+0.037
 celebrate+0.032
 picnic+0.031
Neg logit contributions
 gut-0.055
 beast-0.055
 mistake-0.052
 therm-0.052
cut-0.051
Token snow
Token ID6729
Feature activation-3.31e+00

Pos logit contributions
 parties+0.076
 tea+0.070
 celebrate+0.054
 picnic+0.051
 party+0.051
Neg logit contributions
 beast-0.104
 gut-0.101
inct-0.101
 passer-0.097
 mistake-0.095
Tokenballs
Token ID21591
Feature activation-8.32e+00

Pos logit contributions
 parties+0.119
 tea+0.106
 recess+0.098
 celebrate+0.091
 wear+0.088
Neg logit contributions
 beast-0.096
 gut-0.095
 mistake-0.089
 therm-0.089
cut-0.089
Token."
Token ID526
Feature activation-6.23e-01

Pos logit contributions
 parties+0.253
 tea+0.223
 recess+0.199
 celebrate+0.179
 picnic+0.175
Neg logit contributions
 beast-0.282
 gut-0.280
 mistake-0.269
 therm-0.268
cut-0.264
Token
Token ID198
Feature activation-5.39e-01

Pos logit contributions
 parties+0.021
 tea+0.018
 recess+0.016
 celebrate+0.015
 party+0.015
Neg logit contributions
 gut-0.020
 beast-0.019
 mistake-0.019
 therm-0.018
inct-0.018
Token
Token ID198
Feature activation+1.02e-01

Pos logit contributions
 parties+0.019
 tea+0.017
 recess+0.015
 celebrate+0.014
 party+0.014
Neg logit contributions
 gut-0.017
 beast-0.016
 mistake-0.016
 therm-0.015
cut-0.015
TokenL
Token ID43
Feature activation-5.27e-01

Pos logit contributions
 beast+0.004
 gut+0.004
 mistake+0.004
inct+0.003
 therm+0.003
Neg logit contributions
 parties-0.003
 tea-0.003
 recess-0.002
 celebrate-0.002
 party-0.002
Tokenily
Token ID813
Feature activation-4.37e-01

Pos logit contributions
 parties+0.010
 tea+0.008
 recess+0.006
 celebrate+0.006
 picnic+0.005
Neg logit contributions
 beast-0.025
 gut-0.025
 mistake-0.024
 therm-0.024
inct-0.023
Token of
Token ID286
Feature activation-1.83e+00

Pos logit contributions
 parties+0.098
 tea+0.088
 recess+0.079
ackers+0.074
 picnic+0.073
Neg logit contributions
 beast-0.089
 gut-0.087
cut-0.085
 ringing-0.084
 therm-0.084
Token all
Token ID477
Feature activation-3.63e+00

Pos logit contributions
 parties+0.048
 tea+0.042
 recess+0.033
 celebrate+0.032
 party+0.030
Neg logit contributions
 beast-0.075
 gut-0.075
inct-0.071
 mistake-0.070
 therm-0.070
Token the
Token ID262
Feature activation-2.33e+00

Pos logit contributions
 parties+0.111
 tea+0.096
 recess+0.084
 picnic+0.081
 party+0.079
Neg logit contributions
 beast-0.124
 gut-0.124
 mistake-0.117
cut-0.117
 therm-0.116
Token fun
Token ID1257
Feature activation-4.85e+00

Pos logit contributions
 parties+0.076
 tea+0.068
 recess+0.061
 celebrate+0.057
 picnic+0.055
Neg logit contributions
 beast-0.075
 gut-0.073
 mistake-0.071
 therm-0.069
inct-0.068
Token things
Token ID1243
Feature activation-1.79e+00

Pos logit contributions
 parties+0.142
 tea+0.127
ackers+0.117
 recess+0.116
 celebrate+0.113
Neg logit contributions
 beast-0.147
 mistake-0.142
 gut-0.141
 ringing-0.135
 therm-0.134
Token they
Token ID484
Feature activation-8.31e+00

Pos logit contributions
 parties+0.044
 tea+0.038
 recess+0.033
 picnic+0.028
 celebrate+0.028
Neg logit contributions
 gut-0.069
 beast-0.069
 therm-0.067
cut-0.066
 swollen-0.064
Token did
Token ID750
Feature activation-2.78e+00

Pos logit contributions
 parties+0.267
 tea+0.253
 recess+0.209
 sandwiches+0.195
 picnic+0.193
Neg logit contributions
 beast-0.257
 gut-0.249
 mistake-0.234
 injured-0.234
 reckless-0.232
Token.
Token ID13
Feature activation-1.47e+00

Pos logit contributions
 parties+0.086
 tea+0.081
 recess+0.070
 picnic+0.065
 celebrate+0.062
Neg logit contributions
 gut-0.089
 mistake-0.089
 beast-0.086
inct-0.080
 plea-0.079
Token Ch
Token ID609
Feature activation-1.97e+00

Pos logit contributions
 parties+0.049
 recess+0.041
 tea+0.040
 party+0.037
 celebrate+0.036
Neg logit contributions
 gut-0.046
inct-0.045
 therm-0.045
 beast-0.043
 Owl-0.043
Tokenoo
Token ID2238
Feature activation+8.36e-01

Pos logit contributions
 parties+0.068
 tea+0.061
 recess+0.053
 celebrate+0.051
 party+0.049
Neg logit contributions
 beast-0.063
 gut-0.063
 mistake-0.059
 therm-0.058
inct-0.058
Token-
Token ID12
Feature activation+5.36e+00

Pos logit contributions
 beast+0.028
 gut+0.028
 mistake+0.026
 therm+0.025
inct+0.025
Neg logit contributions
 parties-0.028
 tea-0.025
 recess-0.021
 celebrate-0.020
 party-0.020
Token every
Token ID790
Feature activation-3.90e+00

Pos logit contributions
 parties+0.121
 tea+0.105
 recess+0.087
 celebrate+0.081
 picnic+0.078
Neg logit contributions
 beast-0.178
 gut-0.178
 mistake-0.170
 therm-0.164
inct-0.164
Token day
Token ID1110
Feature activation-3.22e+00

Pos logit contributions
 parties+0.090
 tea+0.074
 recess+0.057
 celebrate+0.053
 party+0.049
Neg logit contributions
 beast-0.171
 gut-0.169
 mistake-0.163
 therm-0.160
inct-0.159
Token and
Token ID290
Feature activation-1.70e+00

Pos logit contributions
 parties+0.103
 tea+0.091
 recess+0.077
 celebrate+0.075
 picnic+0.073
Neg logit contributions
 gut-0.107
 beast-0.107
 mistake-0.100
 injured-0.097
 therm-0.096
Token shared
Token ID4888
Feature activation-4.37e+00

Pos logit contributions
 parties+0.046
 tea+0.040
 recess+0.033
 celebrate+0.031
 party+0.030
Neg logit contributions
 beast-0.067
 gut-0.066
 mistake-0.063
 therm-0.062
inct-0.062
Token everything
Token ID2279
Feature activation-1.83e+00

Pos logit contributions
 parties+0.113
 recess+0.093
 tea+0.093
acky+0.082
dates+0.081
Neg logit contributions
 mistake-0.157
 gut-0.150
 beast-0.145
 severe-0.136
 swollen-0.136
Token they
Token ID484
Feature activation-8.30e+00

Pos logit contributions
 parties+0.054
 tea+0.045
 recess+0.044
 picnic+0.039
ashed+0.038
Neg logit contributions
 gut-0.060
 mistake-0.058
 beast-0.056
 guardian-0.055
 therm-0.054
Token had
Token ID550
Feature activation-6.27e+00

Pos logit contributions
 parties+0.257
 tea+0.255
 sandwiches+0.209
 outfits+0.209
text+0.203
Neg logit contributions
 reckless-0.221
 beast-0.219
 disob-0.215
 Owl-0.212
 injured-0.212
Token.
Token ID13
Feature activation-2.15e+00

Pos logit contributions
 parties+0.191
 tea+0.188
 recess+0.162
 celebrate+0.150
acky+0.146
Neg logit contributions
 mistake-0.184
 gut-0.170
 injured-0.163
 beast-0.160
 reckless-0.158
Token<|endoftext|>
Token ID50256
Feature activation-8.00e-01

Pos logit contributions
 parties+0.085
 tea+0.071
 recess+0.068
 outfits+0.067
 costumes+0.065
Neg logit contributions
 beast-0.053
 mistake-0.053
 gut-0.052
 Owl-0.051
 severe-0.050
TokenOnce
Token ID7454
Feature activation-3.14e+00

Pos logit contributions
 parties+0.032
 tea+0.029
 recess+0.025
 outfits+0.025
ets+0.025
Neg logit contributions
 beast-0.020
 gut-0.018
inct-0.018
 mistake-0.017
 therm-0.017
Token upon
Token ID2402
Feature activation+2.64e+00

Pos logit contributions
 parties+0.135
 tea+0.124
 celebrate+0.111
 recess+0.109
ils+0.108
Neg logit contributions
 gut-0.069
 mistake-0.068
 beast-0.066
 therm-0.061
 adult-0.060
Token menu
Token ID6859
Feature activation+2.20e+00

Pos logit contributions
 parties+0.046
 tea+0.041
 recess+0.034
 celebrate+0.033
 party+0.032
Neg logit contributions
 beast-0.052
 gut-0.052
 mistake-0.048
inct-0.048
 therm-0.048
Token with
Token ID351
Feature activation-2.30e+00

Pos logit contributions
 beast+0.074
 mistake+0.072
 gut+0.071
 therm+0.071
cut+0.069
Neg logit contributions
 parties-0.066
 tea-0.057
 recess-0.053
 celebrate-0.047
dates-0.045
Token pictures
Token ID5986
Feature activation-2.57e+00

Pos logit contributions
 parties+0.070
 tea+0.059
 recess+0.054
 celebrate+0.050
 party+0.048
Neg logit contributions
 beast-0.079
 gut-0.077
 therm-0.076
 mistake-0.073
 passer-0.073
Token of
Token ID286
Feature activation-1.59e+00

Pos logit contributions
 parties+0.083
 tea+0.074
 recess+0.068
 celebrate+0.062
 picnic+0.060
Neg logit contributions
 beast-0.079
cut-0.074
 gut-0.074
 therm-0.074
 passer-0.070
Token food
Token ID2057
Feature activation-2.11e+00

Pos logit contributions
 parties+0.047
 tea+0.041
 recess+0.035
 celebrate+0.033
 party+0.032
Neg logit contributions
 beast-0.058
 gut-0.057
 mistake-0.054
 therm-0.054
 swollen-0.053
Token they
Token ID484
Feature activation-8.30e+00

Pos logit contributions
 parties+0.062
 tea+0.054
 recess+0.047
 celebrate+0.044
 picnic+0.041
Neg logit contributions
 beast-0.076
 gut-0.075
 mistake-0.071
 therm-0.071
cut-0.069
Token liked
Token ID8288
Feature activation-2.25e+00

Pos logit contributions
 parties+0.255
 tea+0.231
 recess+0.195
 celebrate+0.176
 picnic+0.175
Neg logit contributions
 beast-0.287
 gut-0.278
 therm-0.264
 mistake-0.263
cut-0.259
Token,
Token ID11
Feature activation-3.64e+00

Pos logit contributions
 parties+0.067
 tea+0.060
 recess+0.055
 celebrate+0.051
 wear+0.049
Neg logit contributions
 beast-0.075
 gut-0.072
 mistake-0.070
 therm-0.069
cut-0.067
Token such
Token ID884
Feature activation-5.27e-01

Pos logit contributions
 parties+0.114
 tea+0.101
 recess+0.081
 celebrate+0.080
 party+0.078
Neg logit contributions
 beast-0.131
 gut-0.131
inct-0.123
 mistake-0.122
 swollen-0.121
Token as
Token ID355
Feature activation-6.73e-01

Pos logit contributions
 parties+0.015
 tea+0.013
 recess+0.011
 celebrate+0.010
 party+0.010
Neg logit contributions
 beast-0.020
 gut-0.020
 mistake-0.019
 therm-0.019
inct-0.019
Token pizza
Token ID14256
Feature activation-1.36e+00

Pos logit contributions
 parties+0.020
 tea+0.017
 recess+0.014
 celebrate+0.014
 party+0.013
Neg logit contributions
 beast-0.025
 gut-0.025
 mistake-0.023
 therm-0.023
 swollen-0.023
Token snow
Token ID6729
Feature activation-1.53e+00

Pos logit contributions
 parties+0.016
 tea+0.013
 recess+0.013
 celebrate+0.011
ets+0.011
Neg logit contributions
 beast-0.020
 gut-0.019
 mistake-0.019
 therm-0.019
 swollen-0.018
Tokenmen
Token ID3653
Feature activation-7.31e-01

Pos logit contributions
 parties+0.059
 tea+0.053
 recess+0.048
 celebrate+0.046
 party+0.044
Neg logit contributions
 beast-0.042
 gut-0.041
inct-0.038
 mistake-0.038
 therm-0.038
Token and
Token ID290
Feature activation-2.15e+00

Pos logit contributions
 parties+0.023
 tea+0.020
 recess+0.019
 celebrate+0.017
 picnic+0.016
Neg logit contributions
 beast-0.024
 gut-0.024
 therm-0.023
 mistake-0.023
cut-0.023
Token throwing
Token ID9644
Feature activation-4.27e+00

Pos logit contributions
 parties+0.066
 tea+0.062
 recess+0.047
 celebrate+0.046
 costumes+0.046
Neg logit contributions
 gut-0.080
 beast-0.079
inct-0.078
 passer-0.075
 mistake-0.074
Token snow
Token ID6729
Feature activation-3.21e+00

Pos logit contributions
 parties+0.137
 tea+0.113
 recess+0.107
 celebrate+0.101
 party+0.098
Neg logit contributions
 beast-0.141
 gut-0.136
 therm-0.133
 mistake-0.130
inct-0.129
Tokenballs
Token ID21591
Feature activation-8.16e+00

Pos logit contributions
 parties+0.096
 tea+0.083
 recess+0.073
 celebrate+0.070
 party+0.066
Neg logit contributions
 beast-0.115
 gut-0.113
 therm-0.108
inct-0.107
cut-0.107
Token.
Token ID13
Feature activation-1.73e+00

Pos logit contributions
 parties+0.251
 tea+0.217
 recess+0.193
 celebrate+0.184
 picnic+0.170
Neg logit contributions
 beast-0.278
 gut-0.274
 therm-0.264
cut-0.263
 mistake-0.262
Token
Token ID198
Feature activation-1.25e+00

Pos logit contributions
 parties+0.049
 tea+0.042
 recess+0.034
 celebrate+0.034
ils+0.034
Neg logit contributions
 gut-0.065
 beast-0.064
inct-0.061
 therm-0.060
 Owl-0.060
Token
Token ID198
Feature activation-2.92e-01

Pos logit contributions
 parties+0.036
 tea+0.032
 recess+0.026
 celebrate+0.025
 party+0.024
Neg logit contributions
 beast-0.047
 gut-0.047
 mistake-0.044
 therm-0.044
inct-0.044
Token"
Token ID1
Feature activation+2.12e+00

Pos logit contributions
 parties+0.010
 tea+0.009
 recess+0.008
 celebrate+0.007
 picnic+0.007
Neg logit contributions
 beast-0.010
 gut-0.010
 mistake-0.009
 therm-0.009
inct-0.009
TokenWow
Token ID22017
Feature activation-1.73e+00

Pos logit contributions
 beast+0.063
 gut+0.059
inct+0.058
 passer+0.057
 therm+0.057
Neg logit contributions
 parties-0.078
 tea-0.069
 recess-0.060
 celebrate-0.057
 party-0.057
Token people
Token ID661
Feature activation+2.57e-01

Pos logit contributions
 parties+0.034
 tea+0.029
 recess+0.022
 celebrate+0.022
 party+0.021
Neg logit contributions
 gut-0.058
 beast-0.057
inct-0.056
 swollen-0.055
 therm-0.054
Token use
Token ID779
Feature activation-2.70e+00

Pos logit contributions
 gut+0.011
 beast+0.011
inct+0.011
 swollen+0.011
 therm+0.011
Neg logit contributions
 parties-0.006
 tea-0.005
 celebrate-0.004
 wear-0.004
 party-0.004
Token to
Token ID284
Feature activation+4.27e-01

Pos logit contributions
 parties+0.072
 tea+0.060
 celebrate+0.048
 recess+0.047
 party+0.045
Neg logit contributions
 beast-0.111
 gut-0.110
inct-0.106
 swollen-0.105
 passer-0.102
Token buy
Token ID2822
Feature activation-1.97e+00

Pos logit contributions
 beast+0.017
 gut+0.017
inct+0.016
 swollen+0.016
 therm+0.015
Neg logit contributions
 parties-0.012
 tea-0.010
 celebrate-0.009
 recess-0.008
 wear-0.008
Token things
Token ID1243
Feature activation-2.40e+00

Pos logit contributions
 parties+0.060
 tea+0.052
 recess+0.043
 celebrate+0.042
 party+0.040
Neg logit contributions
 beast-0.072
 gut-0.072
inct-0.067
 mistake-0.067
 swollen-0.067
Token they
Token ID484
Feature activation-8.15e+00

Pos logit contributions
 parties+0.062
 tea+0.052
 celebrate+0.040
 recess+0.040
 party+0.038
Neg logit contributions
 beast-0.101
 gut-0.099
inct-0.094
 swollen-0.094
 mistake-0.094
Token want
Token ID765
Feature activation-3.51e+00

Pos logit contributions
 parties+0.236
 tea+0.197
 celebrate+0.168
 recess+0.159
 wear+0.158
Neg logit contributions
 beast-0.317
 gut-0.314
 swollen-0.297
inct-0.295
 therm-0.292
Token or
Token ID393
Feature activation-1.02e+00

Pos logit contributions
 parties+0.097
 tea+0.085
 recess+0.070
 celebrate+0.066
 party+0.063
Neg logit contributions
 beast-0.135
 gut-0.134
 mistake-0.127
 therm-0.125
inct-0.125
Token need
Token ID761
Feature activation-2.04e+00

Pos logit contributions
 parties+0.032
 tea+0.028
 celebrate+0.024
 wear+0.023
 party+0.023
Neg logit contributions
 beast-0.038
inct-0.037
 gut-0.037
 swollen-0.034
 therm-0.034
Token.
Token ID13
Feature activation-1.35e+00

Pos logit contributions
 parties+0.053
 tea+0.045
 recess+0.036
 celebrate+0.035
 party+0.033
Neg logit contributions
 beast-0.083
 gut-0.083
inct-0.079
 swollen-0.078
 mistake-0.078
Token
Token ID198
Feature activation-1.28e+00

Pos logit contributions
 parties+0.039
 tea+0.034
 recess+0.027
 celebrate+0.026
 party+0.025
Neg logit contributions
 beast-0.052
 gut-0.052
 mistake-0.049
 swollen-0.048
inct-0.048
Token and
Token ID290
Feature activation-6.38e-01

Pos logit contributions
 parties+0.008
 tea+0.007
 recess+0.006
 celebrate+0.005
 party+0.005
Neg logit contributions
 gut-0.009
 beast-0.009
 mistake-0.009
 therm-0.008
inct-0.008
Token enjoyed
Token ID8359
Feature activation-1.76e+00

Pos logit contributions
 parties+0.017
 tea+0.015
 recess+0.012
 celebrate+0.011
 party+0.011
Neg logit contributions
 beast-0.026
 gut-0.026
 therm-0.024
 mistake-0.024
inct-0.024
Token the
Token ID262
Feature activation-8.51e-01

Pos logit contributions
 parties+0.054
 tea+0.047
 recess+0.040
 celebrate+0.038
 party+0.037
Neg logit contributions
 gut-0.063
 beast-0.063
 mistake-0.059
 therm-0.058
inct-0.057
Token beautiful
Token ID4950
Feature activation-1.42e+00

Pos logit contributions
 parties+0.027
 tea+0.024
 recess+0.020
 celebrate+0.019
 party+0.018
Neg logit contributions
 beast-0.030
 gut-0.029
 mistake-0.028
 therm-0.027
inct-0.027
Token sights
Token ID21343
Feature activation-8.79e-01

Pos logit contributions
 parties+0.048
 tea+0.042
 recess+0.036
 celebrate+0.035
 picnic+0.034
Neg logit contributions
 gut-0.048
 beast-0.047
 swollen-0.045
 therm-0.044
 passer-0.044
Token they
Token ID484
Feature activation-8.14e+00

Pos logit contributions
 parties+0.028
 tea+0.025
 recess+0.022
 celebrate+0.021
 outfits+0.020
Neg logit contributions
 beast-0.029
 gut-0.027
 therm-0.026
cut-0.026
 ringing-0.026
Token found
Token ID1043
Feature activation-3.34e+00

Pos logit contributions
 parties+0.265
 tea+0.252
 recess+0.196
 sandwiches+0.189
 picnic+0.188
Neg logit contributions
 beast-0.265
 gut-0.252
 therm-0.234
 Owl-0.233
 injured-0.233
Token.
Token ID13
Feature activation-1.65e+00

Pos logit contributions
 parties+0.095
 tea+0.086
 recess+0.071
 celebrate+0.068
 picnic+0.064
Neg logit contributions
 beast-0.123
 gut-0.121
 mistake-0.115
cut-0.113
 therm-0.112
Token At
Token ID1629
Feature activation+1.96e+00

Pos logit contributions
 parties+0.061
 tea+0.054
 recess+0.050
 skirts+0.046
 outfits+0.046
Neg logit contributions
 beast-0.045
 gut-0.044
inct-0.043
 mistake-0.042
 therm-0.040
Token the
Token ID262
Feature activation+6.18e-01

Pos logit contributions
 beast+0.064
 gut+0.063
 mistake+0.060
 therm+0.058
inct+0.058
Neg logit contributions
 parties-0.066
 tea-0.059
 recess-0.049
 celebrate-0.048
 party-0.046
Token end
Token ID886
Feature activation+3.18e+00

Pos logit contributions
 beast+0.020
 gut+0.020
 mistake+0.019
 swollen+0.019
 therm+0.018
Neg logit contributions
 parties-0.021
 tea-0.019
 recess-0.016
 celebrate-0.016
 party-0.015
Token of
Token ID286
Feature activation-2.39e-01

Pos logit contributions
 beast+0.054
 gut+0.053
inct+0.050
 mistake+0.050
 swollen+0.050
Neg logit contributions
 parties-0.041
 tea-0.036
 recess-0.029
 celebrate-0.028
 party-0.027
Token all
Token ID477
Feature activation-2.24e+00

Pos logit contributions
 parties+0.005
 tea+0.004
 recess+0.003
 celebrate+0.003
 party+0.003
Neg logit contributions
 beast-0.011
 gut-0.011
inct-0.011
 swollen-0.011
 therm-0.011
Token the
Token ID262
Feature activation-1.18e+00

Pos logit contributions
 parties+0.057
 tea+0.049
 celebrate+0.037
 recess+0.035
 party+0.034
Neg logit contributions
 beast-0.095
 gut-0.094
inct-0.090
 swollen-0.090
 therm-0.089
Token wonderful
Token ID7932
Feature activation-2.50e+00

Pos logit contributions
 parties+0.038
 tea+0.033
 recess+0.029
 celebrate+0.027
 party+0.026
Neg logit contributions
 beast-0.041
 gut-0.041
 mistake-0.039
 therm-0.038
inct-0.037
Token things
Token ID1243
Feature activation-1.16e+00

Pos logit contributions
 parties+0.086
 tea+0.077
 recess+0.069
 celebrate+0.065
 picnic+0.061
Neg logit contributions
 beast-0.078
 gut-0.076
 mistake-0.074
 therm-0.070
inct-0.069
Token they
Token ID484
Feature activation-8.13e+00

Pos logit contributions
 parties+0.030
 tea+0.026
 recess+0.021
 celebrate+0.019
 picnic+0.018
Neg logit contributions
 beast-0.046
 gut-0.046
 mistake-0.044
 therm-0.043
inct-0.043
Token had
Token ID550
Feature activation-6.70e+00

Pos logit contributions
 parties+0.244
 tea+0.226
 recess+0.185
 picnic+0.169
 outfits+0.164
Neg logit contributions
 beast-0.281
 gut-0.277
 mistake-0.261
cut-0.255
 guardian-0.253
Token done
Token ID1760
Feature activation-2.95e+00

Pos logit contributions
 parties+0.198
 tea+0.189
 recess+0.156
 picnic+0.144
 celebrate+0.142
Neg logit contributions
 gut-0.214
 beast-0.213
 mistake-0.211
inct-0.196
 injured-0.196
Token.
Token ID13
Feature activation-1.27e+00

Pos logit contributions
 parties+0.085
 tea+0.076
 recess+0.064
 picnic+0.060
 celebrate+0.059
Neg logit contributions
 gut-0.107
 beast-0.105
 mistake-0.103
inct-0.098
 therm-0.097
Token It
Token ID632
Feature activation+3.05e+00

Pos logit contributions
 parties+0.040
 tea+0.035
 recess+0.030
 celebrate+0.028
 party+0.028
Neg logit contributions
 beast-0.044
 gut-0.044
 mistake-0.041
inct-0.041
 therm-0.041
Token was
Token ID373
Feature activation+1.43e+00

Pos logit contributions
 gut+0.114
 beast+0.111
rum+0.104
 therm+0.103
 swollen+0.102
Neg logit contributions
 parties-0.090
 tea-0.082
 celebrate-0.069
 party-0.065
 together-0.063
Token.
Token ID13
Feature activation-1.68e+00

Pos logit contributions
 parties+0.061
 tea+0.054
 recess+0.046
 celebrate+0.043
 picnic+0.041
Neg logit contributions
 beast-0.078
 gut-0.078
 mistake-0.075
cut-0.074
 therm-0.073
Token They
Token ID1119
Feature activation-3.34e+00

Pos logit contributions
 parties+0.048
 tea+0.041
 recess+0.035
ils+0.033
 party+0.033
Neg logit contributions
 gut-0.062
 beast-0.061
 therm-0.059
 Owl-0.058
inct-0.058
Token share
Token ID2648
Feature activation-3.05e+00

Pos logit contributions
 parties+0.090
 tea+0.077
 recess+0.064
 celebrate+0.058
 party+0.057
Neg logit contributions
 beast-0.132
 gut-0.131
 mistake-0.124
 therm-0.123
 swollen-0.122
Token their
Token ID511
Feature activation-2.74e+00

Pos logit contributions
 parties+0.065
 tea+0.051
 recess+0.044
 celebrate+0.038
 party+0.036
Neg logit contributions
 gut-0.134
 beast-0.133
 mistake-0.131
 therm-0.128
 swollen-0.127
Token snow
Token ID6729
Feature activation-3.22e+00

Pos logit contributions
 parties+0.087
 tea+0.075
 recess+0.066
 celebrate+0.063
 party+0.059
Neg logit contributions
 gut-0.094
 beast-0.094
 mistake-0.090
 therm-0.088
 swollen-0.086
Tokenballs
Token ID21591
Feature activation-8.12e+00

Pos logit contributions
 parties+0.143
 tea+0.129
 recess+0.123
 celebrate+0.116
 wear+0.112
Neg logit contributions
 beast-0.067
 gut-0.067
cut-0.062
 mistake-0.061
 therm-0.059
Token and
Token ID290
Feature activation-3.00e+00

Pos logit contributions
 parties+0.270
 tea+0.240
 recess+0.208
 celebrate+0.197
 picnic+0.190
Neg logit contributions
 beast-0.266
 gut-0.265
 mistake-0.250
 therm-0.245
cut-0.243
Token snow
Token ID6729
Feature activation-3.25e+00

Pos logit contributions
 parties+0.078
 tea+0.068
 recess+0.054
 celebrate+0.051
 party+0.049
Neg logit contributions
 beast-0.122
 gut-0.122
 mistake-0.115
inct-0.115
 therm-0.114
Tokenmen
Token ID3653
Feature activation-1.52e+00

Pos logit contributions
 parties+0.122
 tea+0.109
 recess+0.100
 celebrate+0.095
 wear+0.091
Neg logit contributions
 beast-0.091
 gut-0.090
cut-0.085
 mistake-0.084
 therm-0.083
Token.
Token ID13
Feature activation-2.06e+00

Pos logit contributions
 parties+0.042
 tea+0.036
 recess+0.030
 celebrate+0.028
 party+0.027
Neg logit contributions
 beast-0.059
 gut-0.059
 mistake-0.056
inct-0.055
 therm-0.055
Token They
Token ID1119
Feature activation-3.43e+00

Pos logit contributions
 parties+0.063
 tea+0.053
ils+0.049
 recess+0.047
 party+0.046
Neg logit contributions
 gut-0.070
 Owl-0.067
 therm-0.066
 beast-0.066
inct-0.065
Token They
Token ID1119
Feature activation-3.16e+00

Pos logit contributions
 parties+0.032
 tea+0.026
ils+0.024
 recess+0.024
 celebrate+0.022
Neg logit contributions
 gut-0.038
 therm-0.036
 beast-0.036
 mistake-0.036
inct-0.036
Token had
Token ID550
Feature activation-3.07e+00

Pos logit contributions
 parties+0.097
 tea+0.085
 recess+0.074
 celebrate+0.067
 party+0.065
Neg logit contributions
 beast-0.111
 gut-0.109
 therm-0.105
cut-0.105
 mistake-0.103
Token fun
Token ID1257
Feature activation-5.86e+00

Pos logit contributions
 parties+0.098
 tea+0.082
 recess+0.078
 celebrate+0.071
 party+0.068
Neg logit contributions
 beast-0.103
 gut-0.099
 therm-0.098
 swollen-0.096
 mistake-0.095
Token throwing
Token ID9644
Feature activation-3.83e+00

Pos logit contributions
 parties+0.189
 tea+0.164
 recess+0.143
 celebrate+0.137
 party+0.132
Neg logit contributions
 beast-0.201
 gut-0.199
 mistake-0.190
 therm-0.187
inct-0.185
Token snow
Token ID6729
Feature activation-2.94e+00

Pos logit contributions
 parties+0.133
 tea+0.115
 recess+0.103
 celebrate+0.099
 party+0.096
Neg logit contributions
 beast-0.122
 gut-0.119
 therm-0.112
 mistake-0.112
inct-0.111
Tokenballs
Token ID21591
Feature activation-8.11e+00

Pos logit contributions
 parties+0.101
 tea+0.088
 recess+0.084
 celebrate+0.080
 wear+0.075
Neg logit contributions
 beast-0.090
cut-0.086
 gut-0.086
inct-0.083
 therm-0.082
Token at
Token ID379
Feature activation-7.10e-01

Pos logit contributions
 parties+0.263
 tea+0.228
 recess+0.206
 celebrate+0.202
 picnic+0.185
Neg logit contributions
 beast-0.255
 gut-0.248
cut-0.243
 mistake-0.239
 therm-0.236
Token each
Token ID1123
Feature activation-3.73e+00

Pos logit contributions
 parties+0.022
 tea+0.020
 recess+0.016
 celebrate+0.016
ets+0.016
Neg logit contributions
 beast-0.025
 gut-0.024
 mistake-0.023
 therm-0.023
inct-0.022
Token other
Token ID584
Feature activation-5.48e-02

Pos logit contributions
 parties+0.135
 tea+0.120
 recess+0.105
 celebrate+0.101
 party+0.097
Neg logit contributions
 beast-0.115
 gut-0.112
 mistake-0.107
 therm-0.105
inct-0.102
Token.
Token ID13
Feature activation-1.38e+00

Pos logit contributions
 parties+0.001
 tea+0.001
 recess+0.001
 celebrate+0.001
 party+0.001
Neg logit contributions
 beast-0.002
 gut-0.002
 mistake-0.002
 therm-0.002
inct-0.002
Token But
Token ID887
Feature activation+1.90e+00

Pos logit contributions
 parties+0.051
 tea+0.043
 recess+0.039
ils+0.038
 celebrate+0.038
Neg logit contributions
 gut-0.039
 beast-0.037
 mistake-0.037
inct-0.037
 therm-0.035
Token good
Token ID922
Feature activation-1.35e+00

Pos logit contributions
 parties+0.064
 tea+0.055
 recess+0.049
 picnic+0.047
ets+0.047
Neg logit contributions
 gut-0.056
 beast-0.053
 mistake-0.052
 guardian-0.050
 reckless-0.050
Token friends
Token ID2460
Feature activation-4.89e+00

Pos logit contributions
 parties+0.043
 tea+0.039
 recess+0.033
 celebrate+0.033
 picnic+0.032
Neg logit contributions
 gut-0.046
 beast-0.044
 mistake-0.043
 reckless-0.041
 adult-0.040
Token and
Token ID290
Feature activation-1.97e+00

Pos logit contributions
 parties+0.168
 tea+0.153
 recess+0.137
 celebrate+0.125
 picnic+0.123
Neg logit contributions
 gut-0.152
 beast-0.151
 mistake-0.141
cut-0.134
 therm-0.133
Token share
Token ID2648
Feature activation-3.35e+00

Pos logit contributions
 parties+0.049
 tea+0.042
 recess+0.033
 celebrate+0.032
 party+0.030
Neg logit contributions
 beast-0.083
 gut-0.082
inct-0.078
 mistake-0.078
 therm-0.078
Token everything
Token ID2279
Feature activation-1.65e+00

Pos logit contributions
 parties+0.091
 tea+0.078
 recess+0.069
 celebrate+0.061
 party+0.060
Neg logit contributions
 mistake-0.125
 gut-0.124
 beast-0.123
 swollen-0.116
 therm-0.115
Token they
Token ID484
Feature activation-8.09e+00

Pos logit contributions
 parties+0.049
 tea+0.041
 recess+0.040
 picnic+0.035
 celebrate+0.033
Neg logit contributions
 gut-0.054
 mistake-0.053
cut-0.051
 beast-0.051
 swollen-0.050
Token found
Token ID1043
Feature activation-3.39e+00

Pos logit contributions
 parties+0.260
 tea+0.252
 recess+0.209
text+0.198
 outfits+0.198
Neg logit contributions
 beast-0.229
 reckless-0.224
 gut-0.223
 injured-0.219
 mistake-0.216
Token in
Token ID287
Feature activation+9.92e-02

Pos logit contributions
 parties+0.097
 tea+0.086
 recess+0.073
 celebrate+0.068
 picnic+0.065
Neg logit contributions
 beast-0.124
 gut-0.123
 mistake-0.119
cut-0.115
 therm-0.114
Token the
Token ID262
Feature activation-2.58e+00

Pos logit contributions
 beast+0.004
 gut+0.004
inct+0.004
 therm+0.004
cut+0.004
Neg logit contributions
 parties-0.003
 tea-0.002
 recess-0.002
 celebrate-0.002
 party-0.002
Token park
Token ID3952
Feature activation-1.97e+00

Pos logit contributions
 parties+0.078
 tea+0.072
 recess+0.059
 party+0.058
 picnic+0.057
Neg logit contributions
 gut-0.094
 beast-0.093
rum-0.090
 swollen-0.089
inct-0.088
Token from
Token ID422
Feature activation+6.44e-01

Pos logit contributions
 parties+0.053
 tea+0.046
 recess+0.038
 celebrate+0.036
 picnic+0.034
Neg logit contributions
 beast-0.076
 gut-0.076
 mistake-0.072
 therm-0.070
inct-0.070
Token snow
Token ID6729
Feature activation-1.47e+00

Pos logit contributions
 parties+0.068
 recess+0.057
 tea+0.056
dates+0.050
 party+0.050
Neg logit contributions
 gut-0.069
 beast-0.068
 mistake-0.066
 therm-0.065
 injured-0.062
Tokenmen
Token ID3653
Feature activation-5.42e-01

Pos logit contributions
 parties+0.053
 tea+0.047
 recess+0.045
 celebrate+0.040
 wear+0.039
Neg logit contributions
 beast-0.042
 gut-0.042
 mistake-0.040
cut-0.039
inct-0.039
Token,
Token ID11
Feature activation-3.04e+00

Pos logit contributions
 parties+0.018
 tea+0.016
 recess+0.015
 celebrate+0.013
 picnic+0.013
Neg logit contributions
 gut-0.017
 beast-0.017
 mistake-0.016
 therm-0.016
cut-0.016
Token threw
Token ID9617
Feature activation-3.90e+00

Pos logit contributions
 parties+0.095
 tea+0.088
 celebrate+0.068
 recess+0.067
 party+0.066
Neg logit contributions
 beast-0.111
 gut-0.110
inct-0.107
 mistake-0.103
 passer-0.101
Token snow
Token ID6729
Feature activation-2.76e+00

Pos logit contributions
 parties+0.127
 tea+0.108
 recess+0.097
 celebrate+0.092
 party+0.090
Neg logit contributions
 beast-0.131
 gut-0.128
 therm-0.123
 mistake-0.121
 swollen-0.119
Tokenballs
Token ID21591
Feature activation-8.09e+00

Pos logit contributions
 parties+0.086
 tea+0.075
 recess+0.067
 celebrate+0.063
 party+0.059
Neg logit contributions
 beast-0.095
 gut-0.095
 therm-0.089
cut-0.088
 mistake-0.088
Token,
Token ID11
Feature activation-3.42e+00

Pos logit contributions
 parties+0.279
 tea+0.246
 recess+0.220
 celebrate+0.214
 picnic+0.199
Neg logit contributions
 beast-0.244
 gut-0.244
 mistake-0.230
 therm-0.228
cut-0.227
Token and
Token ID290
Feature activation-1.65e+00

Pos logit contributions
 parties+0.105
 tea+0.095
 recess+0.076
 celebrate+0.074
 party+0.072
Neg logit contributions
 beast-0.125
 gut-0.124
inct-0.117
 mistake-0.116
 therm-0.113
Token slid
Token ID27803
Feature activation-3.14e+00

Pos logit contributions
 parties+0.051
 tea+0.045
 recess+0.038
 celebrate+0.036
 party+0.035
Neg logit contributions
 beast-0.059
 gut-0.059
 mistake-0.055
inct-0.055
 therm-0.055
Token down
Token ID866
Feature activation+5.82e-01

Pos logit contributions
 parties+0.099
 tea+0.085
 recess+0.074
 celebrate+0.072
 party+0.069
Neg logit contributions
 beast-0.109
 gut-0.108
 therm-0.102
inct-0.101
 swollen-0.100
Token hills
Token ID18639
Feature activation-1.23e+00

Pos logit contributions
 gut+0.024
 beast+0.024
inct+0.023
 mistake+0.023
 swollen+0.022
Neg logit contributions
 parties-0.016
 tea-0.014
 recess-0.011
 together-0.010
 celebrate-0.010
Token on
Token ID319
Feature activation+3.15e-01

Pos logit contributions
 parties+0.057
 tea+0.049
 recess+0.037
 celebrate+0.037
 picnic+0.037
Neg logit contributions
 beast-0.083
 gut-0.082
inct-0.078
 therm-0.077
 severe-0.076
Token,
Token ID11
Feature activation-2.08e+00

Pos logit contributions
 beast+0.012
 gut+0.012
 mistake+0.011
 therm+0.011
inct+0.011
Neg logit contributions
 parties-0.009
 tea-0.008
 recess-0.007
 celebrate-0.006
 party-0.006
Token they
Token ID484
Feature activation-4.01e+00

Pos logit contributions
 parties+0.056
 tea+0.049
 recess+0.039
 celebrate+0.038
 party+0.036
Neg logit contributions
 beast-0.083
 gut-0.083
 mistake-0.079
inct-0.078
 therm-0.078
Token shared
Token ID4888
Feature activation-4.08e+00

Pos logit contributions
 parties+0.111
 tea+0.097
 recess+0.079
 celebrate+0.074
 party+0.071
Neg logit contributions
 beast-0.156
 gut-0.155
 mistake-0.146
 therm-0.145
inct-0.145
Token everything
Token ID2279
Feature activation-1.16e+00

Pos logit contributions
 parties+0.112
 tea+0.094
 recess+0.085
 celebrate+0.074
 party+0.073
Neg logit contributions
 beast-0.153
 gut-0.153
 mistake-0.149
 swollen-0.142
 therm-0.142
Token they
Token ID484
Feature activation-8.07e+00

Pos logit contributions
 parties+0.029
 tea+0.024
 recess+0.022
 celebrate+0.018
 picnic+0.018
Neg logit contributions
 gut-0.047
 beast-0.046
 mistake-0.044
cut-0.044
 therm-0.043
Token found
Token ID1043
Feature activation-3.40e+00

Pos logit contributions
 parties+0.247
 tea+0.232
 recess+0.184
 outfits+0.177
 birthday+0.171
Neg logit contributions
 beast-0.267
 gut-0.258
 reckless-0.253
 injured-0.252
 Owl-0.251
Token in
Token ID287
Feature activation+1.08e-01

Pos logit contributions
 parties+0.098
 tea+0.086
 recess+0.071
 celebrate+0.068
 party+0.065
Neg logit contributions
 beast-0.128
 gut-0.127
 mistake-0.120
 therm-0.119
inct-0.119
Token the
Token ID262
Feature activation-2.12e+00

Pos logit contributions
 beast+0.004
 gut+0.004
inct+0.004
 therm+0.004
cut+0.004
Neg logit contributions
 parties-0.003
 tea-0.003
 recess-0.002
 celebrate-0.002
 party-0.002
Token forest
Token ID8222
Feature activation-1.52e-01

Pos logit contributions
 parties+0.066
 tea+0.064
 picnic+0.053
 party+0.053
 recess+0.050
Neg logit contributions
 gut-0.072
rum-0.071
 beast-0.068
cut-0.067
iously-0.067
Token and
Token ID290
Feature activation-1.99e+00

Pos logit contributions
 parties+0.004
 tea+0.003
 celebrate+0.002
 party+0.002
 together+0.002
Neg logit contributions
 beast-0.007
 gut-0.007
inct-0.006
 therm-0.006
 swollen-0.006
Tokenty
Token ID774
Feature activation-5.88e-01

Pos logit contributions
 gut+0.090
 beast+0.088
inct+0.085
 therm+0.084
 mistake+0.084
Neg logit contributions
 parties-0.058
 tea-0.051
 celebrate-0.037
 party-0.037
 played-0.037
Token and
Token ID290
Feature activation+4.29e-01

Pos logit contributions
 parties+0.017
 tea+0.015
 played+0.012
 celebrate+0.012
 party+0.011
Neg logit contributions
 beast-0.023
 gut-0.022
inct-0.021
 therm-0.021
 swollen-0.021
Token Sammy
Token ID44067
Feature activation-2.09e+00

Pos logit contributions
 beast+0.015
 gut+0.015
 mistake+0.014
 therm+0.014
inct+0.014
Neg logit contributions
 parties-0.014
 tea-0.012
 recess-0.010
 celebrate-0.010
 party-0.010
Token shared
Token ID4888
Feature activation-3.88e+00

Pos logit contributions
 parties+0.054
 tea+0.048
 recess+0.041
 picnic+0.035
 birthday+0.034
Neg logit contributions
 beast-0.078
 gut-0.077
cut-0.075
 mistake-0.074
 Owl-0.073
Token everything
Token ID2279
Feature activation-1.28e+00

Pos logit contributions
 parties+0.103
 tea+0.088
 recess+0.087
ets+0.079
dates+0.077
Neg logit contributions
 mistake-0.132
 gut-0.129
 beast-0.123
 therm-0.122
 severe-0.122
Token they
Token ID484
Feature activation-8.05e+00

Pos logit contributions
 parties+0.030
 tea+0.025
 recess+0.024
 picnic+0.019
 celebrate+0.018
Neg logit contributions
 gut-0.049
 beast-0.047
cut-0.047
 mistake-0.046
 guardian-0.046
Token found
Token ID1043
Feature activation-3.17e+00

Pos logit contributions
 tea+0.232
 parties+0.232
 sandwiches+0.181
 recess+0.176
 outfits+0.176
Neg logit contributions
 Owl-0.241
 beast-0.240
 reckless-0.231
 Eagle-0.231
 gut-0.228
Token and
Token ID290
Feature activation-1.95e+00

Pos logit contributions
 parties+0.097
 tea+0.086
 recess+0.075
 celebrate+0.069
 picnic+0.067
Neg logit contributions
 beast-0.111
 gut-0.111
 mistake-0.105
cut-0.102
 therm-0.102
Token were
Token ID547
Feature activation-1.47e+00

Pos logit contributions
 parties+0.048
 tea+0.041
 recess+0.033
 celebrate+0.030
 party+0.029
Neg logit contributions
 gut-0.080
 beast-0.079
 mistake-0.075
 swollen-0.074
cut-0.074
Token the
Token ID262
Feature activation-4.48e-02

Pos logit contributions
 parties+0.048
 tea+0.042
 recess+0.037
 picnic+0.034
 celebrate+0.034
Neg logit contributions
 gut-0.049
 beast-0.048
 mistake-0.046
 swollen-0.045
 therm-0.045
Token happiest
Token ID49414
Feature activation-9.10e-01

Pos logit contributions
 parties+0.001
 tea+0.001
 recess+0.001
 celebrate+0.001
 party+0.001
Neg logit contributions
 beast-0.002
 gut-0.002
 mistake-0.002
 therm-0.002
inct-0.002
Token liked
Token ID8288
Feature activation-2.62e-01

Pos logit contributions
 parties+0.126
 tea+0.117
 recess+0.103
dates+0.101
 sandwiches+0.101
Neg logit contributions
 claw-0.061
 beast-0.057
 gut-0.056
 panicked-0.055
 shaking-0.055
Token to
Token ID284
Feature activation+1.58e+00

Pos logit contributions
 parties+0.010
 tea+0.009
 recess+0.008
 celebrate+0.007
 party+0.007
Neg logit contributions
 beast-0.007
 gut-0.007
 mistake-0.007
 therm-0.007
 swollen-0.007
Token examine
Token ID10716
Feature activation-7.39e-01

Pos logit contributions
 claw+0.043
 gut+0.043
 beast+0.042
 mistake+0.041
 spike+0.039
Neg logit contributions
 parties-0.053
 tea-0.047
text-0.044
ils-0.043
ouses-0.043
Token different
Token ID1180
Feature activation-5.59e-02

Pos logit contributions
 parties+0.025
 tea+0.021
 recess+0.020
 celebrate+0.019
 party+0.018
Neg logit contributions
 beast-0.024
 gut-0.023
 therm-0.022
 mistake-0.022
 swollen-0.021
Token things
Token ID1243
Feature activation-2.48e+00

Pos logit contributions
 parties+0.002
 tea+0.002
 recess+0.001
 celebrate+0.001
 party+0.001
Neg logit contributions
 beast-0.002
 gut-0.002
 mistake-0.002
 swollen-0.002
 therm-0.002
Token they
Token ID484
Feature activation-8.02e+00

Pos logit contributions
 parties+0.075
 tea+0.066
 recess+0.057
 celebrate+0.053
 party+0.051
Neg logit contributions
 beast-0.089
 gut-0.089
 mistake-0.083
 therm-0.083
cut-0.082
Token found
Token ID1043
Feature activation-3.51e+00

Pos logit contributions
 parties+0.258
 tea+0.239
 recess+0.197
 celebrate+0.184
 picnic+0.183
Neg logit contributions
 beast-0.260
 gut-0.258
 mistake-0.239
 therm-0.239
 swollen-0.237
Token,
Token ID11
Feature activation-3.02e+00

Pos logit contributions
 parties+0.101
 tea+0.089
 recess+0.078
 celebrate+0.073
 picnic+0.069
Neg logit contributions
 beast-0.127
 gut-0.126
 mistake-0.120
cut-0.117
 therm-0.116
Token like
Token ID588
Feature activation-2.58e+00

Pos logit contributions
 parties+0.080
 tea+0.069
 recess+0.056
 celebrate+0.052
 party+0.050
Neg logit contributions
 beast-0.121
 gut-0.121
 mistake-0.114
 therm-0.113
inct-0.113
Token leaves
Token ID5667
Feature activation-1.92e+00

Pos logit contributions
 parties+0.080
 tea+0.070
 recess+0.062
 celebrate+0.058
 party+0.056
Neg logit contributions
 beast-0.089
 gut-0.087
 mistake-0.083
 therm-0.082
 swollen-0.081
Token,
Token ID11
Feature activation-3.38e+00

Pos logit contributions
 parties+0.052
 tea+0.045
 recess+0.034
 celebrate+0.033
 party+0.033
Neg logit contributions
 beast-0.078
 gut-0.078
inct-0.074
 mistake-0.073
 swollen-0.072
Token He
Token ID679
Feature activation+1.03e+00

Pos logit contributions
 beast+0.001
 gut+0.001
 mistake+0.001
 swollen+0.001
 therm+0.001
Neg logit contributions
 parties-0.001
 tea-0.001
 recess-0.000
 celebrate-0.000
 party-0.000
Token cooked
Token ID15847
Feature activation-9.51e-01

Pos logit contributions
 gut+0.037
 beast+0.036
inct+0.034
 mistake+0.033
 swollen+0.033
Neg logit contributions
 parties-0.032
 tea-0.029
 celebrate-0.024
 party-0.023
 recess-0.023
Token spaghetti
Token ID48241
Feature activation-5.36e-01

Pos logit contributions
 parties+0.028
 tea+0.024
 recess+0.021
 celebrate+0.019
 party+0.018
Neg logit contributions
 beast-0.035
 gut-0.035
 mistake-0.033
 therm-0.033
inct-0.032
Token with
Token ID351
Feature activation-2.09e+00

Pos logit contributions
 parties+0.016
 tea+0.013
 recess+0.012
 celebrate+0.011
 picnic+0.010
Neg logit contributions
 beast-0.019
 gut-0.019
 mistake-0.018
 therm-0.018
cut-0.018
Token meat
Token ID6174
Feature activation-9.14e-01

Pos logit contributions
 parties+0.060
 tea+0.051
 recess+0.044
 celebrate+0.041
 party+0.040
Neg logit contributions
 beast-0.078
 gut-0.077
 therm-0.073
 mistake-0.073
 swollen-0.072
Tokenballs
Token ID21591
Feature activation-7.99e+00

Pos logit contributions
 parties+0.026
 tea+0.023
 recess+0.018
 celebrate+0.018
 party+0.017
Neg logit contributions
 beast-0.035
 gut-0.035
 mistake-0.033
inct-0.033
 swollen-0.032
Token.
Token ID13
Feature activation-3.30e-01

Pos logit contributions
 parties+0.246
 tea+0.216
 recess+0.184
 celebrate+0.174
 party+0.167
Neg logit contributions
 beast-0.283
 gut-0.281
 mistake-0.265
 therm-0.263
inct-0.260
Token His
Token ID2399
Feature activation+1.88e+00

Pos logit contributions
 parties+0.011
 tea+0.009
 recess+0.008
 celebrate+0.008
 party+0.007
Neg logit contributions
 beast-0.012
 gut-0.011
 swollen-0.011
 mistake-0.011
 severe-0.011
Token friends
Token ID2460
Feature activation-1.61e+00

Pos logit contributions
 beast+0.071
 gut+0.070
inct+0.067
 swollen+0.066
cut+0.066
Neg logit contributions
 parties-0.054
 tea-0.049
 recess-0.038
 picnic-0.037
 celebrate-0.037
Token all
Token ID477
Feature activation-8.74e-01

Pos logit contributions
 parties+0.043
 tea+0.037
 celebrate+0.029
 recess+0.029
 party+0.028
Neg logit contributions
 beast-0.065
 gut-0.065
 mistake-0.060
inct-0.060
 therm-0.060
Token said
Token ID531
Feature activation-3.01e+00

Pos logit contributions
 parties+0.026
 tea+0.022
 celebrate+0.018
 recess+0.018
 party+0.017
Neg logit contributions
 gut-0.033
 beast-0.033
inct-0.031
 mistake-0.031
 therm-0.030
Token under
Token ID739
Feature activation+1.82e-01

Pos logit contributions
 parties+0.046
 tea+0.040
 recess+0.033
 celebrate+0.031
 party+0.029
Neg logit contributions
 beast-0.068
 gut-0.067
 therm-0.064
 mistake-0.064
inct-0.064
Token the
Token ID262
Feature activation-1.22e+00

Pos logit contributions
 gut+0.008
 beast+0.008
cut+0.008
inct+0.008
 mistake+0.007
Neg logit contributions
 parties-0.004
 tea-0.004
 recess-0.003
 celebrate-0.003
 picnic-0.003
Token couch
Token ID18507
Feature activation+1.43e+00

Pos logit contributions
 parties+0.043
 tea+0.040
 party+0.033
 recess+0.033
 picnic+0.033
Neg logit contributions
 gut-0.037
 beast-0.037
inct-0.036
 mistake-0.033
cut-0.033
Token.
Token ID13
Feature activation+1.51e-01

Pos logit contributions
 beast+0.064
 gut+0.064
inct+0.061
 swollen+0.061
 mistake+0.060
Neg logit contributions
 parties-0.032
 tea-0.027
 party-0.019
 recess-0.019
 celebrate-0.018
Token The
Token ID383
Feature activation+3.24e+00

Pos logit contributions
 beast+0.007
 swollen+0.006
 gut+0.006
 mistake+0.006
ache+0.006
Neg logit contributions
 parties-0.003
 tea-0.003
 celebrate-0.002
 recess-0.002
 played-0.002
Token nos
Token ID43630
Feature activation+6.40e+00

Pos logit contributions
rum+0.101
inct+0.095
 beast+0.093
 gut+0.091
iously+0.091
Neg logit contributions
 parties-0.112
 tea-0.110
 party-0.096
 picnic-0.093
 recess-0.092
Tokeny
Token ID88
Feature activation+3.00e+00

Pos logit contributions
 gut+0.267
 beast+0.256
 therm+0.253
 swollen+0.253
inct+0.250
Neg logit contributions
 parties-0.152
 tea-0.124
 recess-0.102
 celebrate-0.094
ashed-0.091
Token person
Token ID1048
Feature activation+4.85e+00

Pos logit contributions
 swollen+0.087
 gut+0.086
rum+0.085
inct+0.084
 beast+0.080
Neg logit contributions
 tea-0.097
 parties-0.097
 party-0.089
 picnic-0.085
 play-0.078
Token was
Token ID373
Feature activation+1.50e+00

Pos logit contributions
 gut+0.180
 beast+0.177
inct+0.172
 therm+0.168
 swollen+0.168
Neg logit contributions
 parties-0.136
 tea-0.122
 played-0.104
 party-0.101
 celebrate-0.100
Token determined
Token ID5295
Feature activation+4.00e+00

Pos logit contributions
 beast+0.055
 gut+0.053
inct+0.053
 therm+0.050
cut+0.050
Neg logit contributions
 parties-0.044
 tea-0.041
 celebrate-0.034
 party-0.032
 together-0.032
Token to
Token ID284
Feature activation+3.20e+00

Pos logit contributions
 beast+0.165
inct+0.163
 swollen+0.161
 gut+0.160
 therm+0.159
Neg logit contributions
 parties-0.102
 tea-0.082
 recess-0.066
 together-0.066
 party-0.065
Tokensnap
Token ID45380
Feature activation+8.89e-03

Pos logit contributions
 gut+0.066
 beast+0.062
 mistake+0.060
inct+0.056
 swollen+0.055
Neg logit contributions
 parties-0.057
 tea-0.051
 celebrate-0.045
 recess-0.045
 wear-0.043
Token'.
Token ID4458
Feature activation+1.56e+00

Pos logit contributions
 gut+0.000
 beast+0.000
 mistake+0.000
 swollen+0.000
inct+0.000
Neg logit contributions
 parties-0.000
 tea-0.000
 celebrate-0.000
 recess-0.000
 party-0.000
Token
Token ID198
Feature activation-1.56e-01

Pos logit contributions
 gut+0.061
 beast+0.060
 mistake+0.058
 swollen+0.057
 therm+0.056
Neg logit contributions
 parties-0.041
 tea-0.037
 celebrate-0.030
 recess-0.030
 together-0.027
Token
Token ID198
Feature activation+6.00e-01

Pos logit contributions
 parties+0.004
 tea+0.003
 recess+0.003
 celebrate+0.003
 wear+0.002
Neg logit contributions
 gut-0.006
 beast-0.006
 mistake-0.006
 therm-0.006
 swollen-0.006
TokenThe
Token ID464
Feature activation+4.01e+00

Pos logit contributions
 gut+0.021
 beast+0.021
 mistake+0.020
 therm+0.020
 swollen+0.020
Neg logit contributions
 parties-0.017
 tea-0.015
 celebrate-0.013
 recess-0.013
P-0.013
Token egg
Token ID5935
Feature activation+6.29e+00

Pos logit contributions
rum+0.132
 gut+0.124
shadow+0.112
inct+0.111
aned+0.109
Neg logit contributions
 tea-0.134
 party-0.115
 parties-0.114
 picnic-0.111
 wedding-0.108
Token looked
Token ID3114
Feature activation+4.14e-01

Pos logit contributions
 gut+0.253
 beast+0.236
 passer+0.233
rum+0.232
 severe+0.228
Neg logit contributions
 tea-0.145
 parties-0.138
 played-0.125
 party-0.117
 together-0.112
Token around
Token ID1088
Feature activation+3.14e+00

Pos logit contributions
 beast+0.016
 gut+0.016
 mistake+0.015
inct+0.015
 therm+0.015
Neg logit contributions
 parties-0.012
 tea-0.010
 recess-0.009
 celebrate-0.008
 party-0.008
Token and
Token ID290
Feature activation+8.70e-01

Pos logit contributions
 beast+0.115
 gut+0.115
 mistake+0.108
inct+0.107
 therm+0.107
Neg logit contributions
 parties-0.094
 tea-0.083
 recess-0.069
 celebrate-0.066
 party-0.064
Token saw
Token ID2497
Feature activation+2.60e+00

Pos logit contributions
 gut+0.034
 beast+0.034
 therm+0.032
 mistake+0.032
inct+0.032
Neg logit contributions
 parties-0.023
 tea-0.021
 recess-0.017
 celebrate-0.016
 party-0.016
Token a
Token ID257
Feature activation-2.72e-01

Pos logit contributions
 gut+0.115
 beast+0.110
inct+0.108
 mistake+0.108
 passer+0.107
Neg logit contributions
 parties-0.054
 tea-0.052
 celebrate-0.035
 together-0.035
 recess-0.033
Token water
Token ID1660
Feature activation+1.45e+00

Pos logit contributions
 gut+0.039
 beast+0.039
inct+0.037
 swollen+0.036
 severe+0.036
Neg logit contributions
 parties-0.039
 tea-0.035
 recess-0.030
 party-0.028
 celebrate-0.028
Token was
Token ID373
Feature activation+2.70e+00

Pos logit contributions
 gut+0.052
 beast+0.051
 swollen+0.048
rum+0.048
inct+0.047
Neg logit contributions
 parties-0.045
 tea-0.042
 party-0.034
 celebrate-0.033
 picnic-0.033
Token gone
Token ID3750
Feature activation+1.33e+00

Pos logit contributions
 beast+0.110
 gut+0.108
inct+0.099
 passer+0.099
 mistake+0.097
Neg logit contributions
 parties-0.076
 tea-0.068
 celebrate-0.054
 party-0.053
 recess-0.052
Token,
Token ID11
Feature activation+2.21e-01

Pos logit contributions
 beast+0.051
 gut+0.051
 mistake+0.048
 therm+0.048
inct+0.048
Neg logit contributions
 parties-0.037
 tea-0.032
 recess-0.027
 celebrate-0.025
 party-0.024
Token the
Token ID262
Feature activation+3.05e+00

Pos logit contributions
 beast+0.009
 gut+0.009
 mistake+0.009
inct+0.009
 therm+0.009
Neg logit contributions
 parties-0.006
 tea-0.005
 recess-0.004
 celebrate-0.004
 party-0.003
Token town
Token ID3240
Feature activation+6.45e+00

Pos logit contributions
 beast+0.097
 gut+0.097
rum+0.096
inct+0.094
 swollen+0.091
Neg logit contributions
 parties-0.103
 tea-0.099
 recess-0.082
 party-0.081
 picnic-0.079
Token was
Token ID373
Feature activation+2.31e+00

Pos logit contributions
 gut+0.234
 beast+0.233
 swollen+0.228
rum+0.224
 therm+0.224
Neg logit contributions
 parties-0.185
 tea-0.164
 party-0.145
 celebrate-0.143
 played-0.141
Token very
Token ID845
Feature activation+3.68e+00

Pos logit contributions
 beast+0.090
 gut+0.087
inct+0.083
 passer+0.081
 Eagle+0.079
Neg logit contributions
 parties-0.066
 tea-0.062
 celebrate-0.050
 party-0.049
 picnic-0.047
Token happy
Token ID3772
Feature activation+1.02e+00

Pos logit contributions
 beast+0.147
 gut+0.131
 Eagle+0.131
 Owl+0.131
inct+0.130
Neg logit contributions
 parties-0.103
 tea-0.101
 celebrate-0.081
 party-0.078
 together-0.077
Token.
Token ID13
Feature activation+1.02e-01

Pos logit contributions
 beast+0.038
 gut+0.038
 mistake+0.036
 therm+0.035
inct+0.035
Neg logit contributions
 parties-0.030
 tea-0.026
 recess-0.022
 celebrate-0.021
 party-0.020
Token They
Token ID1119
Feature activation-2.57e+00

Pos logit contributions
 beast+0.004
 gut+0.004
 mistake+0.004
 swollen+0.004
 therm+0.004
Neg logit contributions
 parties-0.003
 tea-0.002
 recess-0.002
 celebrate-0.002
 party-0.002
Token?"
Token ID1701
Feature activation+1.77e+00

Pos logit contributions
 parties+0.058
 tea+0.051
 recess+0.043
 celebrate+0.040
 picnic+0.039
Neg logit contributions
 beast-0.072
 gut-0.071
 mistake-0.067
 therm-0.066
inct-0.066
Token
Token ID220
Feature activation+1.11e+00

Pos logit contributions
 beast+0.069
 gut+0.068
 swollen+0.065
 mistake+0.065
inct+0.064
Neg logit contributions
 parties-0.052
 tea-0.045
 celebrate-0.037
 recess-0.037
 party-0.035
Token
Token ID198
Feature activation+2.99e-01

Pos logit contributions
 gut+0.035
 beast+0.035
inct+0.033
 mistake+0.033
 therm+0.033
Neg logit contributions
 parties-0.038
 tea-0.034
 recess-0.030
 party-0.028
 celebrate-0.028
Token
Token ID198
Feature activation+8.86e-01

Pos logit contributions
 gut+0.009
 beast+0.009
 therm+0.009
 mistake+0.009
inct+0.009
Neg logit contributions
 parties-0.010
 tea-0.009
 recess-0.008
 picnic-0.007
 party-0.007
TokenThe
Token ID464
Feature activation+4.37e+00

Pos logit contributions
 beast+0.032
 gut+0.031
 mistake+0.029
 swollen+0.028
 therm+0.028
Neg logit contributions
 parties-0.028
 tea-0.025
 celebrate-0.022
 recess-0.021
 party-0.020
Token man
Token ID582
Feature activation+6.20e+00

Pos logit contributions
 gut+0.102
 beast+0.100
 swollen+0.097
inct+0.093
 therm+0.092
Neg logit contributions
 parties-0.192
 tea-0.173
 recess-0.158
 celebrate-0.152
 party-0.151
Token smiled
Token ID13541
Feature activation+1.30e+00

Pos logit contributions
 gut+0.245
 beast+0.236
 severe+0.226
 swollen+0.224
 therm+0.221
Neg logit contributions
 parties-0.166
 tea-0.148
 played-0.127
 celebrate-0.124
 party-0.123
Token and
Token ID290
Feature activation+9.15e-02

Pos logit contributions
 gut+0.043
 beast+0.043
 mistake+0.040
 therm+0.040
inct+0.040
Neg logit contributions
 parties-0.044
 tea-0.039
 recess-0.034
 celebrate-0.032
 party-0.031
Token answered
Token ID9373
Feature activation+5.15e-01

Pos logit contributions
 gut+0.003
 beast+0.003
 mistake+0.003
inct+0.003
 therm+0.003
Neg logit contributions
 parties-0.003
 tea-0.003
 recess-0.002
 celebrate-0.002
 party-0.002
Token,
Token ID11
Feature activation-4.53e-01

Pos logit contributions
 gut+0.019
 beast+0.019
 mistake+0.018
 therm+0.018
 swollen+0.018
Neg logit contributions
 parties-0.015
 tea-0.013
 recess-0.011
 celebrate-0.010
 picnic-0.010
Token "
Token ID366
Feature activation+3.50e+00

Pos logit contributions
 parties+0.012
 tea+0.012
 together+0.010
 Lily+0.010
 celebrate+0.010
Neg logit contributions
 beast-0.017
 gut-0.017
 severe-0.016
inct-0.016
rum-0.015
Token hello
Token ID23748
Feature activation+1.29e+00

Pos logit contributions
 parties+0.106
 tea+0.094
 celebrate+0.076
 recess+0.072
 party+0.072
Neg logit contributions
 beast-0.129
 gut-0.129
 swollen-0.124
 mistake-0.124
inct-0.124
Token.
Token ID13
Feature activation+1.66e-01

Pos logit contributions
 beast+0.050
 gut+0.049
 swollen+0.049
 mistake+0.048
 therm+0.047
Neg logit contributions
 parties-0.035
 tea-0.030
 celebrate-0.023
 party-0.023
 recess-0.022
Token
Token ID198
Feature activation-2.41e-01

Pos logit contributions
 mistake+0.007
 gut+0.007
 beast+0.007
 swollen+0.007
 severe+0.007
Neg logit contributions
 parties-0.004
 tea-0.003
 together-0.002
 celebrate-0.002
 recess-0.002
Token
Token ID198
Feature activation+5.41e-01

Pos logit contributions
 parties+0.007
 tea+0.006
 recess+0.005
 celebrate+0.005
 picnic+0.004
Neg logit contributions
 beast-0.009
 gut-0.009
 mistake-0.009
 therm-0.009
inct-0.009
TokenThe
Token ID464
Feature activation+4.38e+00

Pos logit contributions
 beast+0.019
 gut+0.019
 mistake+0.018
 therm+0.018
inct+0.017
Neg logit contributions
 parties-0.016
 tea-0.014
 celebrate-0.013
 recess-0.012
 picnic-0.012
Token stranger
Token ID16195
Feature activation+5.86e+00

Pos logit contributions
 gut+0.129
rum+0.126
inct+0.122
 mistake+0.118
 beast+0.117
Neg logit contributions
 parties-0.160
 tea-0.149
 party-0.134
 picnic-0.134
 recess-0.126
Token smiled
Token ID13541
Feature activation+1.69e+00

Pos logit contributions
 gut+0.223
 beast+0.215
 swollen+0.208
 mistake+0.207
 severe+0.207
Neg logit contributions
 parties-0.165
 tea-0.148
 played-0.124
 celebrate-0.123
 party-0.122
Token at
Token ID379
Feature activation+1.92e+00

Pos logit contributions
 beast+0.065
 gut+0.064
 mistake+0.062
 swollen+0.061
 therm+0.061
Neg logit contributions
 parties-0.046
 tea-0.041
 recess-0.031
 celebrate-0.031
 party-0.030
Token the
Token ID262
Feature activation+7.18e-01

Pos logit contributions
 beast+0.069
 gut+0.068
 mistake+0.064
 therm+0.063
inct+0.063
Neg logit contributions
 parties-0.060
 tea-0.053
 recess-0.044
 celebrate-0.042
 party-0.041
Token bird
Token ID6512
Feature activation+2.43e+00

Pos logit contributions
 gut+0.019
inct+0.018
rum+0.018
 beast+0.017
 swollen+0.017
Neg logit contributions
 parties-0.029
 tea-0.026
 party-0.024
 picnic-0.023
 celebrate-0.023
Token,
Token ID11
Feature activation-9.62e-01

Pos logit contributions
 gut+0.098
 beast+0.098
inct+0.093
 swollen+0.092
 mistake+0.091
Neg logit contributions
 parties-0.062
 tea-0.056
 celebrate-0.043
 party-0.041
 wear-0.040
Token is
Token ID318
Feature activation+1.28e+00

Pos logit contributions
 beast+0.003
 gut+0.003
 mistake+0.003
 therm+0.003
inct+0.003
Neg logit contributions
 parties-0.003
 tea-0.003
 recess-0.002
 celebrate-0.002
 party-0.002
Token that
Token ID326
Feature activation-2.95e+00

Pos logit contributions
 gut+0.046
 beast+0.046
 mistake+0.043
inct+0.043
 swollen+0.042
Neg logit contributions
 parties-0.039
 tea-0.035
 recess-0.029
 celebrate-0.028
 party-0.027
Token?"
Token ID1701
Feature activation+1.39e+00

Pos logit contributions
 tea+0.064
 parties+0.063
 picnic+0.053
 party+0.053
 celebrate+0.051
Neg logit contributions
 gut-0.125
 swollen-0.120
rum-0.117
 passer-0.116
 vain-0.114
Token
Token ID198
Feature activation-1.43e-02

Pos logit contributions
 beast+0.058
 gut+0.057
 mistake+0.053
inct+0.053
 swollen+0.053
Neg logit contributions
 parties-0.037
 tea-0.032
 recess-0.025
 celebrate-0.025
 party-0.023
Token
Token ID198
Feature activation+5.89e-01

Pos logit contributions
 parties+0.000
 tea+0.000
 recess+0.000
 celebrate+0.000
 party+0.000
Neg logit contributions
 beast-0.001
 gut-0.001
 mistake-0.000
inct-0.000
 therm-0.000
Token"
Token ID1
Feature activation+2.28e+00

Pos logit contributions
 beast+0.023
 gut+0.022
 mistake+0.021
 swollen+0.021
inct+0.021
Neg logit contributions
 parties-0.017
 tea-0.015
 celebrate-0.012
 recess-0.012
 party-0.012
TokenThat
Token ID2504
Feature activation+2.03e+00

Pos logit contributions
 gut+0.089
 beast+0.085
 mistake+0.085
rum+0.080
iously+0.078
Neg logit contributions
 parties-0.062
 tea-0.057
 celebrate-0.048
 recess-0.045
 picnic-0.044
Token is
Token ID318
Feature activation+2.71e+00

Pos logit contributions
 gut+0.095
rum+0.092
 swollen+0.090
 passer+0.089
 reckless+0.088
Neg logit contributions
 tea-0.037
 parties-0.034
 picnic-0.029
 party-0.024
 celebrate-0.023
Token a
Token ID257
Feature activation-1.26e+00

Pos logit contributions
 gut+0.128
rum+0.123
 beast+0.122
 passer+0.118
cut+0.118
Neg logit contributions
 tea-0.057
 parties-0.054
 picnic-0.035
 celebrate-0.034
 party-0.033
Token bird
Token ID6512
Feature activation+1.36e+00

Pos logit contributions
 tea+0.042
 parties+0.041
 picnic+0.036
 party+0.036
 birthday+0.034
Neg logit contributions
 gut-0.041
rum-0.041
iously-0.037
inct-0.037
 beast-0.037
Token's
Token ID338
Feature activation-1.00e+00

Pos logit contributions
 gut+0.061
 beast+0.059
inct+0.057
 mistake+0.056
 passer+0.056
Neg logit contributions
 parties-0.030
 tea-0.027
 celebrate-0.019
 party-0.018
 wear-0.018
TokenThe
Token ID464
Feature activation+3.73e+00

Pos logit contributions
 parties+0.006
 tea+0.005
 recess+0.004
 celebrate+0.004
 party+0.004
Neg logit contributions
 beast-0.009
 gut-0.009
 mistake-0.009
inct-0.009
 therm-0.009
Token referee
Token ID22957
Feature activation+4.01e+00

Pos logit contributions
 gut+0.126
 beast+0.124
inct+0.119
 swollen+0.116
 therm+0.116
Neg logit contributions
 parties-0.124
 tea-0.112
 recess-0.097
 party-0.091
 celebrate-0.090
Token is
Token ID318
Feature activation+2.01e+00

Pos logit contributions
 gut+0.131
 beast+0.130
 therm+0.120
 swollen+0.120
 mistake+0.120
Neg logit contributions
 parties-0.137
 tea-0.122
 recess-0.104
 celebrate-0.101
 party-0.099
Token calm
Token ID9480
Feature activation+5.41e-01

Pos logit contributions
 beast+0.066
 gut+0.065
inct+0.064
 passer+0.060
 therm+0.059
Neg logit contributions
 parties-0.066
 tea-0.062
 celebrate-0.052
 recess-0.051
 together-0.050
Token and
Token ID290
Feature activation+7.68e-01

Pos logit contributions
 beast+0.020
 gut+0.020
inct+0.019
 swollen+0.019
 therm+0.019
Neg logit contributions
 parties-0.016
 tea-0.014
 recess-0.011
 party-0.011
 celebrate-0.011
Token kind
Token ID1611
Feature activation+2.71e+00

Pos logit contributions
 beast+0.027
inct+0.026
 gut+0.026
rum+0.026
 Owl+0.025
Neg logit contributions
 parties-0.023
 tea-0.021
 play-0.019
 played-0.018
 wear-0.018
Token.
Token ID13
Feature activation+1.12e-02

Pos logit contributions
 gut+0.117
 beast+0.117
inct+0.111
 swollen+0.111
 therm+0.110
Neg logit contributions
 parties-0.064
 tea-0.054
 recess-0.042
 party-0.038
 celebrate-0.038
Token He
Token ID679
Feature activation+1.22e+00

Pos logit contributions
 beast+0.000
cut+0.000
 gut+0.000
 swollen+0.000
 therm+0.000
Neg logit contributions
 parties-0.000
 tea-0.000
 recess-0.000
 celebrate-0.000
 party-0.000
Token smiles
Token ID21845
Feature activation+3.32e-01

Pos logit contributions
 gut+0.042
 beast+0.042
inct+0.039
 swollen+0.039
 severe+0.039
Neg logit contributions
 parties-0.040
 tea-0.036
 party-0.028
 recess-0.028
 picnic-0.028
Token at
Token ID379
Feature activation+1.69e+00

Pos logit contributions
 beast+0.013
 gut+0.013
cut+0.013
inct+0.013
 mistake+0.013
Neg logit contributions
 parties-0.009
 tea-0.007
 recess-0.006
 together-0.006
 celebrate-0.006
Token Ben
Token ID3932
Feature activation-3.17e+00

Pos logit contributions
 beast+0.073
 gut+0.070
 mistake+0.069
 adult+0.066
 therm+0.065
Neg logit contributions
 parties-0.041
 tea-0.034
 celebrate-0.027
 picnic-0.026
 party-0.025
TokenWhen
Token ID2215
Feature activation+3.30e+00

Pos logit contributions
 beast+0.013
 swollen+0.013
 gut+0.013
 thorn+0.013
 mistake+0.013
Neg logit contributions
 parties-0.007
 celebrate-0.006
 tea-0.006
 together-0.006
 recess-0.005
Token they
Token ID484
Feature activation-1.86e+00

Pos logit contributions
 gut+0.139
 beast+0.135
 swollen+0.134
 mistake+0.130
cut+0.129
Neg logit contributions
 parties-0.084
 tea-0.072
 recess-0.060
 celebrate-0.059
 picnic-0.055
Token were
Token ID547
Feature activation+9.34e-01

Pos logit contributions
 parties+0.043
 tea+0.037
 celebrate+0.029
 recess+0.028
 party+0.026
Neg logit contributions
 gut-0.083
 beast-0.082
 mistake-0.078
inct-0.077
 swollen-0.077
Token done
Token ID1760
Feature activation+1.32e+00

Pos logit contributions
 beast+0.035
 gut+0.035
 mistake+0.033
inct+0.033
 therm+0.033
Neg logit contributions
 parties-0.028
 tea-0.024
 recess-0.020
 celebrate-0.019
 party-0.018
Token,
Token ID11
Feature activation+3.23e-02

Pos logit contributions
 beast+0.050
 gut+0.050
 mistake+0.047
inct+0.047
 therm+0.047
Neg logit contributions
 parties-0.038
 tea-0.033
 recess-0.027
 celebrate-0.026
 party-0.025
Token John
Token ID1757
Feature activation+9.90e-01

Pos logit contributions
 gut+0.001
 beast+0.001
 swollen+0.001
 thorn+0.001
 mistake+0.001
Neg logit contributions
 parties-0.001
 tea-0.001
 celebrate-0.001
 together-0.001
 recess-0.000
Token said
Token ID531
Feature activation-1.63e+00

Pos logit contributions
 gut+0.040
 beast+0.039
inct+0.036
 swollen+0.036
 mistake+0.036
Neg logit contributions
 parties-0.025
 tea-0.023
 played-0.020
 celebrate-0.019
 party-0.018
Token,
Token ID11
Feature activation+4.46e-01

Pos logit contributions
 parties+0.045
 tea+0.039
 recess+0.031
 celebrate+0.030
 party+0.029
Neg logit contributions
 beast-0.064
 gut-0.064
 mistake-0.060
 therm-0.060
inct-0.060
Token "
Token ID366
Feature activation+3.62e+00

Pos logit contributions
 beast+0.017
 gut+0.017
 swollen+0.016
 mistake+0.016
 distress+0.016
Neg logit contributions
 parties-0.011
 tea-0.010
 together-0.010
 they-0.010
 celebrate-0.010
TokenLook
Token ID8567
Feature activation-1.89e+00

Pos logit contributions
 beast+0.121
 gut+0.120
 mistake+0.112
 therm+0.111
inct+0.111
Neg logit contributions
 parties-0.121
 tea-0.108
 recess-0.092
 celebrate-0.088
 party-0.085
Token Daddy
Token ID26926
Feature activation-6.49e-01

Pos logit contributions
 parties+0.065
 tea+0.057
 recess+0.054
ets+0.047
dates+0.046
Neg logit contributions
 beast-0.059
 gut-0.058
 mistake-0.055
inct-0.052
 therm-0.051
Token it
Token ID340
Feature activation+6.61e-01

Pos logit contributions
 parties+0.044
 tea+0.038
 recess+0.031
 celebrate+0.030
 party+0.029
Neg logit contributions
 beast-0.058
 gut-0.058
 mistake-0.054
 therm-0.054
inct-0.054
Token.
Token ID13
Feature activation-6.84e-01

Pos logit contributions
 beast+0.023
 gut+0.023
 mistake+0.022
inct+0.022
 therm+0.022
Neg logit contributions
 parties-0.020
 tea-0.017
 recess-0.015
 celebrate-0.014
 picnic-0.014
Token
Token ID198
Feature activation-7.38e-01

Pos logit contributions
 parties+0.019
 tea+0.016
 recess+0.013
 celebrate+0.012
 picnic+0.012
Neg logit contributions
 beast-0.027
 gut-0.027
 mistake-0.026
 swollen-0.025
 therm-0.025
Token
Token ID198
Feature activation+1.84e-01

Pos logit contributions
 parties+0.029
 tea+0.026
 recess+0.023
 party+0.022
 celebrate+0.022
Neg logit contributions
 beast-0.020
 gut-0.020
 mistake-0.019
inct-0.018
 therm-0.018
TokenSam
Token ID16305
Feature activation+5.47e-02

Pos logit contributions
 beast+0.006
 gut+0.006
 mistake+0.005
inct+0.005
 therm+0.005
Neg logit contributions
 parties-0.006
 tea-0.006
 recess-0.005
 celebrate-0.005
 party-0.005
Token was
Token ID373
Feature activation+1.66e+00

Pos logit contributions
 gut+0.002
 beast+0.002
 mistake+0.002
inct+0.002
 therm+0.002
Neg logit contributions
 parties-0.002
 tea-0.001
 recess-0.001
 celebrate-0.001
 party-0.001
Token curious
Token ID11040
Feature activation+3.45e+00

Pos logit contributions
 beast+0.061
 gut+0.060
inct+0.059
cut+0.057
 therm+0.056
Neg logit contributions
 parties-0.052
 tea-0.047
 celebrate-0.038
 played-0.037
 party-0.036
Token but
Token ID475
Feature activation+2.04e+00

Pos logit contributions
 beast+0.124
 gut+0.123
 mistake+0.116
 therm+0.114
inct+0.112
Neg logit contributions
 parties-0.106
 tea-0.093
 recess-0.080
 celebrate-0.076
 picnic-0.072
Token she
Token ID673
Feature activation-1.10e+00

Pos logit contributions
inct+0.086
 gut+0.085
 beast+0.084
 therm+0.081
 mistake+0.081
Neg logit contributions
 parties-0.049
 tea-0.041
 played-0.036
 wear-0.035
 recess-0.033
Token was
Token ID373
Feature activation+1.47e+00

Pos logit contributions
 parties+0.030
 played+0.026
 tea+0.025
 drew+0.022
 wear+0.022
Neg logit contributions
 gut-0.041
rum-0.040
 beast-0.039
 swollen-0.039
 therm-0.039
Token also
Token ID635
Feature activation+1.03e+00

Pos logit contributions
inct+0.052
 beast+0.050
cut+0.049
 gut+0.049
 therm+0.047
Neg logit contributions
 parties-0.046
 tea-0.042
 played-0.035
 wear-0.034
 together-0.034
Token gloomy
Token ID46400
Feature activation+2.51e+00

Pos logit contributions
 gut+0.110
 beast+0.109
inct+0.105
 passer+0.104
 mistake+0.103
Neg logit contributions
 tea-0.062
 parties-0.062
 played-0.045
 recess-0.043
 play-0.042
Token.
Token ID13
Feature activation+4.18e-01

Pos logit contributions
 beast+0.103
 gut+0.103
 mistake+0.097
inct+0.096
 therm+0.096
Neg logit contributions
 parties-0.065
 tea-0.056
 recess-0.044
 celebrate-0.042
 party-0.040
Token She
Token ID1375
Feature activation+1.29e+00

Pos logit contributions
 beast+0.017
 gut+0.017
 mistake+0.016
cut+0.016
 severe+0.016
Neg logit contributions
 parties-0.010
 tea-0.009
 recess-0.007
 wear-0.007
 celebrate-0.007
Token was
Token ID373
Feature activation+1.57e+00

Pos logit contributions
 beast+0.051
 gut+0.050
inct+0.048
rum+0.048
cut+0.048
Neg logit contributions
 parties-0.031
 tea-0.029
 played-0.028
 wear-0.025
 danced-0.024
Token getting
Token ID1972
Feature activation+2.69e+00

Pos logit contributions
 beast+0.058
inct+0.055
 gut+0.054
 passer+0.053
rum+0.052
Neg logit contributions
 parties-0.045
 tea-0.045
 played-0.035
 together-0.035
 outside-0.035
Token worried
Token ID7960
Feature activation+4.26e+00

Pos logit contributions
 beast+0.103
 gut+0.096
rum+0.095
 mistake+0.094
inct+0.090
Neg logit contributions
 parties-0.079
 tea-0.078
 together-0.063
 recess-0.057
 them-0.057
Token that
Token ID326
Feature activation+3.95e-01

Pos logit contributions
 beast+0.161
 gut+0.160
inct+0.151
 mistake+0.150
 therm+0.149
Neg logit contributions
 parties-0.124
 tea-0.108
 recess-0.088
 celebrate-0.085
 party-0.082
Token the
Token ID262
Feature activation+3.26e+00

Pos logit contributions
 beast+0.016
 gut+0.016
inct+0.015
cut+0.015
 mistake+0.015
Neg logit contributions
 parties-0.011
 tea-0.009
 recess-0.007
 celebrate-0.007
 party-0.007
Token party
Token ID2151
Feature activation+5.13e-01

Pos logit contributions
rum+0.122
 gut+0.115
 beast+0.108
 passer+0.107
 severe+0.107
Neg logit contributions
 tea-0.099
 parties-0.098
 party-0.080
 picnic-0.079
 celebrate-0.073
Token might
Token ID1244
Feature activation+4.95e+00

Pos logit contributions
 beast+0.018
 gut+0.018
inct+0.017
 swollen+0.017
 therm+0.017
Neg logit contributions
 parties-0.016
 tea-0.014
 party-0.012
 celebrate-0.012
 recess-0.012
Token be
Token ID307
Feature activation+5.33e-01

Pos logit contributions
 beast+0.199
 gut+0.196
inct+0.188
 passer+0.188
 Eagle+0.183
Neg logit contributions
 parties-0.134
 tea-0.123
 celebrate-0.102
 wear-0.101
 play-0.101
Token some
Token ID617
Feature activation-1.57e+00

Pos logit contributions
 beast+0.042
 gut+0.042
 mistake+0.039
inct+0.039
rum+0.038
Neg logit contributions
 parties-0.031
 tea-0.029
 recess-0.022
 celebrate-0.021
 together-0.021
Token big
Token ID1263
Feature activation-5.94e-01

Pos logit contributions
 parties+0.054
 tea+0.051
 recess+0.041
 celebrate+0.040
 picnic+0.039
Neg logit contributions
 beast-0.051
 gut-0.051
 mistake-0.047
inct-0.047
 swollen-0.046
Token balloons
Token ID35485
Feature activation-1.51e+00

Pos logit contributions
 parties+0.020
 tea+0.019
 recess+0.015
 party+0.015
 picnic+0.015
Neg logit contributions
 gut-0.020
 beast-0.019
inct-0.018
 mistake-0.018
 severe-0.018
Token,
Token ID11
Feature activation-1.71e+00

Pos logit contributions
 parties+0.047
 tea+0.041
 recess+0.035
 celebrate+0.033
 party+0.032
Neg logit contributions
 beast-0.054
 gut-0.053
 mistake-0.050
 therm-0.050
inct-0.049
Token some
Token ID617
Feature activation-1.64e+00

Pos logit contributions
 parties+0.052
 tea+0.047
 recess+0.036
 celebrate+0.036
 picnic+0.035
Neg logit contributions
 beast-0.064
 gut-0.064
 mistake-0.060
inct-0.060
 therm-0.057
Token paint
Token ID7521
Feature activation-6.42e-02

Pos logit contributions
 parties+0.052
 tea+0.046
 recess+0.039
 celebrate+0.037
 party+0.036
Neg logit contributions
 beast-0.057
 gut-0.057
 mistake-0.053
inct-0.052
 therm-0.052
Token and
Token ID290
Feature activation-8.77e-01

Pos logit contributions
 parties+0.002
 tea+0.002
 recess+0.001
 celebrate+0.001
 party+0.001
Neg logit contributions
 beast-0.002
 gut-0.002
 therm-0.002
 mistake-0.002
inct-0.002
Token some
Token ID617
Feature activation-2.20e+00

Pos logit contributions
 parties+0.025
 tea+0.022
 recess+0.017
 celebrate+0.017
 party+0.016
Neg logit contributions
 gut-0.034
 beast-0.034
 mistake-0.032
inct-0.032
 therm-0.031
Token fuzzy
Token ID34669
Feature activation-1.97e+00

Pos logit contributions
 parties+0.069
 tea+0.062
 recess+0.051
 celebrate+0.050
 party+0.048
Neg logit contributions
 beast-0.077
 gut-0.077
 mistake-0.072
inct-0.072
 therm-0.071
Token things
Token ID1243
Feature activation-7.11e-01

Pos logit contributions
 parties+0.064
 tea+0.057
 recess+0.048
 celebrate+0.046
 party+0.045
Neg logit contributions
 beast-0.068
 gut-0.067
 mistake-0.063
inct-0.062
 therm-0.062
Token.
Token ID13
Feature activation-1.66e-01

Pos logit contributions
 parties+0.020
 tea+0.018
 recess+0.015
 celebrate+0.014
 party+0.013
Neg logit contributions
 beast-0.026
 gut-0.026
 therm-0.025
 mistake-0.025
cut-0.024
Token
Token ID6184
Feature activation+3.67e-01

Pos logit contributions
 parties+0.038
 tea+0.033
 recess+0.025
 celebrate+0.024
 party+0.023
Neg logit contributions
 beast-0.063
 gut-0.062
 mistake-0.060
inct-0.059
 swollen-0.058
Token
Token ID95
Feature activation+5.92e-03

Pos logit contributions
 gut+0.011
 beast+0.011
inct+0.010
 therm+0.010
 mistake+0.010
Neg logit contributions
 parties-0.013
 tea-0.012
 recess-0.010
 celebrate-0.010
 picnic-0.010
Token
Token ID26391
Feature activation-4.87e-01

Pos logit contributions
 beast+0.000
 gut+0.000
 mistake+0.000
inct+0.000
 therm+0.000
Neg logit contributions
 parties-0.000
 tea-0.000
 recess-0.000
 celebrate-0.000
 party-0.000
Token
Token ID129
Feature activation+8.89e-01

Pos logit contributions
 parties+0.019
 tea+0.018
 recess+0.015
 celebrate+0.015
 party+0.014
Neg logit contributions
 beast-0.013
 gut-0.013
inct-0.012
 mistake-0.012
 therm-0.012
Token
Token ID241
Feature activation+2.86e+00

Pos logit contributions
 beast+0.027
 gut+0.027
 swollen+0.026
 therm+0.025
 mistake+0.025
Neg logit contributions
 parties-0.031
 tea-0.028
 recess-0.024
 celebrate-0.024
 wear-0.023
TokenI
Token ID40
Feature activation-1.02e+00

Pos logit contributions
 beast+0.091
 gut+0.090
 mistake+0.084
 therm+0.083
inct+0.082
Neg logit contributions
 parties-0.100
 tea-0.089
 recess-0.076
 celebrate-0.074
 party-0.072
Token will
Token ID481
Feature activation+2.41e+00

Pos logit contributions
 parties+0.034
 tea+0.030
 recess+0.025
 celebrate+0.025
 party+0.024
Neg logit contributions
 beast-0.035
 gut-0.034
 mistake-0.032
 swollen-0.032
inct-0.032
Token help
Token ID1037
Feature activation+9.10e-01

Pos logit contributions
 beast+0.096
 gut+0.093
inct+0.092
 swollen+0.091
 mistake+0.090
Neg logit contributions
 parties-0.064
 tea-0.053
 wear-0.049
 celebrate-0.048
 play-0.043
Token you
Token ID345
Feature activation-1.63e+00

Pos logit contributions
 beast+0.040
 gut+0.039
inct+0.038
 swollen+0.037
 mistake+0.037
Neg logit contributions
 parties-0.021
 tea-0.017
 recess-0.013
 celebrate-0.013
 wear-0.012
Token.
Token ID13
Feature activation-2.82e-01

Pos logit contributions
 parties+0.042
 tea+0.035
 celebrate+0.030
 wear+0.029
 party+0.027
Neg logit contributions
 beast-0.066
 gut-0.065
 swollen-0.064
 passer-0.063
inct-0.062
Tokenâ
Token ID22940
Feature activation+6.09e-01

Pos logit contributions
 parties+0.009
 tea+0.008
 recess+0.007
 celebrate+0.006
 party+0.006
Neg logit contributions
 beast-0.010
 gut-0.010
 mistake-0.009
 swollen-0.009
inct-0.009
Token very
Token ID845
Feature activation+2.12e+00

Pos logit contributions
 beast+0.014
 gut+0.014
inct+0.013
 mistake+0.013
 therm+0.013
Neg logit contributions
 parties-0.012
 tea-0.011
 recess-0.009
 celebrate-0.009
 party-0.008
Token careful
Token ID8161
Feature activation+2.36e+00

Pos logit contributions
 beast+0.074
 gut+0.072
inct+0.069
 mistake+0.068
 therm+0.067
Neg logit contributions
 parties-0.069
 tea-0.062
 recess-0.052
 celebrate-0.050
 wear-0.048
Token with
Token ID351
Feature activation-6.57e-01

Pos logit contributions
 beast+0.084
 gut+0.081
inct+0.081
 swollen+0.079
 passer+0.078
Neg logit contributions
 parties-0.075
 tea-0.065
 celebrate-0.052
 wear-0.052
 together-0.051
Token his
Token ID465
Feature activation-1.22e+00

Pos logit contributions
 parties+0.017
 tea+0.015
 recess+0.011
 costumes+0.010
 them+0.010
Neg logit contributions
inct-0.028
 beast-0.028
 gut-0.027
 passer-0.026
 plea-0.026
Token cars
Token ID5006
Feature activation-1.99e+00

Pos logit contributions
 parties+0.035
 tea+0.032
 recess+0.026
 party+0.025
 costumes+0.025
Neg logit contributions
 beast-0.046
 gut-0.045
inct-0.044
 passer-0.044
 severe-0.043
Token and
Token ID290
Feature activation-1.39e+00

Pos logit contributions
 parties+0.049
 tea+0.039
 recess+0.030
 party+0.029
 celebrate+0.029
Neg logit contributions
 beast-0.087
 gut-0.085
inct-0.085
 swollen-0.082
 passer-0.082
Token never
Token ID1239
Feature activation+1.71e+00

Pos logit contributions
 parties+0.044
 tea+0.037
 recess+0.031
 costumes+0.030
 played+0.030
Neg logit contributions
inct-0.054
 beast-0.053
 gut-0.052
 passer-0.050
 therm-0.049
Token let
Token ID1309
Feature activation-2.19e+00

Pos logit contributions
 gut+0.067
 beast+0.066
inct+0.063
 therm+0.062
 mistake+0.061
Neg logit contributions
 parties-0.050
 tea-0.043
 recess-0.036
 celebrate-0.034
 party-0.034
Token anyone
Token ID2687
Feature activation+2.34e-01

Pos logit contributions
 parties+0.056
 tea+0.048
 recess+0.037
 celebrate+0.036
 party+0.034
Neg logit contributions
 gut-0.090
 beast-0.090
inct-0.085
 swollen-0.085
 mistake-0.085
Token else
Token ID2073
Feature activation-8.29e-01

Pos logit contributions
 beast+0.008
 gut+0.008
 mistake+0.008
 therm+0.008
 swollen+0.008
Neg logit contributions
 parties-0.007
 tea-0.006
 recess-0.005
 celebrate-0.005
 party-0.005
Token touch
Token ID3638
Feature activation-6.10e-01

Pos logit contributions
 parties+0.023
 tea+0.020
 recess+0.016
 celebrate+0.015
 party+0.015
Neg logit contributions
 beast-0.032
 gut-0.032
 mistake-0.030
inct-0.030
 therm-0.030
Token friends
Token ID2460
Feature activation-3.93e+00

Pos logit contributions
 parties+0.101
 tea+0.086
 recess+0.085
 celebrate+0.080
 wear+0.074
Neg logit contributions
 beast-0.095
 gut-0.095
 mistake-0.093
 therm-0.090
 swollen-0.089
Token,
Token ID11
Feature activation-2.84e+00

Pos logit contributions
 parties+0.134
 tea+0.113
 recess+0.106
 picnic+0.101
ets+0.099
Neg logit contributions
 gut-0.125
 mistake-0.124
 beast-0.123
 Owl-0.120
 adult-0.120
Token Max
Token ID5436
Feature activation-2.74e+00

Pos logit contributions
 parties+0.099
ils+0.083
ets+0.080
 tea+0.079
 outfits+0.075
Neg logit contributions
 Owl-0.084
 beast-0.082
 Eagle-0.080
 Wise-0.079
 therm-0.076
Token and
Token ID290
Feature activation-8.34e-01

Pos logit contributions
 parties+0.108
 tea+0.094
 recess+0.090
 celebrate+0.082
 picnic+0.078
Neg logit contributions
 beast-0.074
 gut-0.073
 mistake-0.072
 therm-0.069
 Owl-0.068
Token Sally
Token ID25737
Feature activation-2.81e+00

Pos logit contributions
 parties+0.027
 tea+0.023
ets+0.023
irl+0.022
ils+0.022
Neg logit contributions
 Owl-0.025
 passer-0.024
 beast-0.024
 Eagle-0.024
 Wise-0.024
Token.
Token ID13
Feature activation-8.40e-01

Pos logit contributions
 parties+0.098
 tea+0.083
 recess+0.079
 celebrate+0.071
 picnic+0.070
Neg logit contributions
 gut-0.085
 mistake-0.084
cut-0.084
 beast-0.084
 therm-0.083
Token They
Token ID1119
Feature activation-2.81e+00

Pos logit contributions
 parties+0.025
 tea+0.023
 seas+0.021
ils+0.020
 celebrate+0.019
Neg logit contributions
 Owl-0.025
 Wise-0.024
 gut-0.024
 beast-0.024
 mistake-0.024
Token lived
Token ID5615
Feature activation-7.79e-01

Pos logit contributions
 parties+0.101
 recess+0.091
 tea+0.091
 pic+0.086
 seas+0.082
Neg logit contributions
cut-0.063
 injured-0.060
 panicked-0.060
 reckless-0.060
 claw-0.060
Token by
Token ID416
Feature activation+4.01e+00

Pos logit contributions
 parties+0.021
 tea+0.018
 recess+0.015
 celebrate+0.014
 picnic+0.014
Neg logit contributions
 gut-0.030
 beast-0.029
 mistake-0.028
inct-0.028
 therm-0.028
Token the
Token ID262
Feature activation-8.61e-01

Pos logit contributions
 gut+0.174
 beast+0.171
inct+0.162
cut+0.161
 swollen+0.160
Neg logit contributions
 parties-0.103
 tea-0.090
 recess-0.069
 celebrate-0.064
 picnic-0.062
Token beach
Token ID10481
Feature activation-1.49e+00

Pos logit contributions
 parties+0.028
 tea+0.025
 party+0.021
 recess+0.021
 picnic+0.020
Neg logit contributions
 gut-0.031
 beast-0.030
rum-0.030
inct-0.029
 swollen-0.029
Token her
Token ID607
Feature activation+3.52e-01

Pos logit contributions
 beast+0.038
 therm+0.038
 gut+0.037
 swollen+0.037
cut+0.036
Neg logit contributions
 parties-0.044
 tea-0.038
 recess-0.034
 picnic-0.033
 celebrate-0.032
Token toys
Token ID14958
Feature activation-4.34e-01

Pos logit contributions
 beast+0.013
 gut+0.013
 therm+0.012
 mistake+0.012
 swollen+0.012
Neg logit contributions
 parties-0.011
 tea-0.009
 recess-0.008
 celebrate-0.008
 party-0.007
Token off
Token ID572
Feature activation+5.86e-01

Pos logit contributions
 parties+0.012
 tea+0.010
 recess+0.008
 celebrate+0.008
 party+0.008
Neg logit contributions
 beast-0.017
 gut-0.017
 mistake-0.016
inct-0.016
 therm-0.016
Token the
Token ID262
Feature activation-6.12e-01

Pos logit contributions
 beast+0.021
 gut+0.021
 mistake+0.020
 therm+0.020
 swollen+0.019
Neg logit contributions
 parties-0.017
 tea-0.015
 recess-0.013
 celebrate-0.012
 picnic-0.012
Token shelf
Token ID18316
Feature activation+2.58e+00

Pos logit contributions
 parties+0.026
 tea+0.023
 recess+0.021
 celebrate+0.020
 party+0.020
Neg logit contributions
 beast-0.015
 gut-0.015
inct-0.014
 mistake-0.014
 swollen-0.014
Token.
Token ID13
Feature activation-2.04e-01

Pos logit contributions
 beast+0.109
 gut+0.109
inct+0.102
 mistake+0.102
 swollen+0.101
Neg logit contributions
 parties-0.064
 tea-0.054
 recess-0.043
 celebrate-0.040
 party-0.039
Token Lily
Token ID20037
Feature activation-1.62e+00

Pos logit contributions
 parties+0.005
 tea+0.005
 recess+0.004
 celebrate+0.004
 party+0.003
Neg logit contributions
 beast-0.008
 gut-0.008
 mistake-0.008
 therm-0.008
inct-0.008
Token was
Token ID373
Feature activation+1.41e+00

Pos logit contributions
 parties+0.039
 played+0.036
 tea+0.035
 drew+0.028
 danced+0.028
Neg logit contributions
 beast-0.068
 gut-0.068
 guardian-0.065
inct-0.064
cut-0.062
Token very
Token ID845
Feature activation+2.63e+00

Pos logit contributions
 beast+0.053
inct+0.051
 gut+0.049
cut+0.048
rum+0.045
Neg logit contributions
 parties-0.043
 tea-0.041
 played-0.034
 together-0.032
 celebrate-0.032
Token sad
Token ID6507
Feature activation+3.78e+00

Pos logit contributions
 beast+0.086
inct+0.078
 gut+0.078
cut+0.073
 Eagle+0.073
Neg logit contributions
 parties-0.095
 tea-0.088
 wear-0.074
 celebrate-0.073
 recess-0.072
Token and
Token ID290
Feature activation+1.75e+00

Pos logit contributions
 beast+0.146
 gut+0.144
inct+0.138
cut+0.135
 swollen+0.135
Neg logit contributions
 parties-0.110
 tea-0.095
 recess-0.075
 party-0.073
 celebrate-0.072
Token.
Token ID13
Feature activation-8.12e-01

Pos logit contributions
 beast+0.060
 gut+0.057
cut+0.057
 therm+0.056
 Owl+0.055
Neg logit contributions
 parties-0.048
 tea-0.039
 recess-0.036
 celebrate-0.032
 picnic-0.032
Token The
Token ID383
Feature activation+2.50e+00

Pos logit contributions
 parties+0.025
 tea+0.022
 recess+0.018
 celebrate+0.018
 picnic+0.017
Neg logit contributions
 gut-0.029
 beast-0.029
 mistake-0.028
inct-0.027
 swollen-0.027
Token pigeon
Token ID46387
Feature activation+4.07e+00

Pos logit contributions
 beast+0.061
 mistake+0.057
 gut+0.056
 injured+0.055
 crow+0.054
Neg logit contributions
 parties-0.102
 tea-0.092
 recess-0.083
ets-0.080
ils-0.080
Token had
Token ID550
Feature activation-2.19e+00

Pos logit contributions
 beast+0.133
 gut+0.127
cut+0.125
 mistake+0.121
 therm+0.119
Neg logit contributions
 parties-0.136
 tea-0.118
 recess-0.112
 celebrate-0.097
 picnic-0.095
Token many
Token ID867
Feature activation-1.01e+00

Pos logit contributions
 parties+0.064
 tea+0.056
 recess+0.048
 celebrate+0.044
 party+0.043
Neg logit contributions
 beast-0.081
 gut-0.081
 mistake-0.076
 therm-0.075
 swollen-0.075
Token friends
Token ID2460
Feature activation-4.32e+00

Pos logit contributions
 parties+0.029
 tea+0.026
 recess+0.022
 celebrate+0.021
 party+0.020
Neg logit contributions
 beast-0.037
 gut-0.037
 mistake-0.035
 therm-0.035
 swollen-0.035
Token.
Token ID13
Feature activation-1.12e+00

Pos logit contributions
 parties+0.153
 tea+0.139
 recess+0.130
 picnic+0.119
 sandwiches+0.113
Neg logit contributions
 gut-0.118
 beast-0.118
cut-0.113
 reckless-0.111
 guardian-0.111
Token They
Token ID1119
Feature activation-2.79e+00

Pos logit contributions
 parties+0.033
 tea+0.030
 recess+0.025
ils+0.024
 party+0.023
Neg logit contributions
 beast-0.040
 gut-0.038
 Owl-0.037
inct-0.037
 therm-0.036
Token played
Token ID2826
Feature activation-5.35e+00

Pos logit contributions
 parties+0.081
 tea+0.076
 recess+0.068
 outfits+0.060
dates+0.060
Neg logit contributions
cut-0.087
 beast-0.086
 gut-0.086
 Owl-0.084
 Eagle-0.082
Token together
Token ID1978
Feature activation-3.84e+00

Pos logit contributions
 parties+0.147
 tea+0.125
ouses+0.125
 recess+0.124
ackers+0.122
Neg logit contributions
 beast-0.160
 warning-0.156
 gut-0.155
 mistake-0.153
inct-0.147
Token every
Token ID790
Feature activation-3.55e+00

Pos logit contributions
 parties+0.107
 tea+0.093
 recess+0.083
 celebrate+0.076
 picnic+0.073
Neg logit contributions
 beast-0.137
 gut-0.134
 mistake-0.132
cut-0.128
inct-0.123
Token
Token ID198
Feature activation-2.60e-02

Pos logit contributions
 parties+0.032
 tea+0.028
 recess+0.024
 celebrate+0.023
 party+0.023
Neg logit contributions
 beast-0.031
 gut-0.030
 mistake-0.029
inct-0.028
 therm-0.028
TokenTim
Token ID14967
Feature activation-7.64e-02

Pos logit contributions
 parties+0.001
 tea+0.001
 celebrate+0.001
 recess+0.001
 picnic+0.001
Neg logit contributions
 beast-0.001
 gut-0.001
 mistake-0.001
 therm-0.001
inct-0.001
Token and
Token ID290
Feature activation+1.00e+00

Pos logit contributions
 parties+0.002
 tea+0.002
 played+0.002
 celebrate+0.002
 party+0.002
Neg logit contributions
 gut-0.003
 beast-0.003
inct-0.003
 therm-0.003
 swollen-0.003
Token the
Token ID262
Feature activation+2.02e+00

Pos logit contributions
 gut+0.046
inct+0.046
 beast+0.045
 plea+0.044
umbling+0.043
Neg logit contributions
 parties-0.019
 tea-0.017
 played-0.014
 play-0.014
 recess-0.013
Token squirrel
Token ID33039
Feature activation+2.27e+00

Pos logit contributions
rum+0.075
 gut+0.073
inct+0.073
iously+0.069
 outage+0.068
Neg logit contributions
 parties-0.052
 tea-0.050
 picnic-0.046
 party-0.043
 play-0.043
Token played
Token ID2826
Feature activation-5.18e+00

Pos logit contributions
 beast+0.084
 gut+0.082
cut+0.082
 therm+0.079
 injured+0.078
Neg logit contributions
 parties-0.063
 tea-0.053
 recess-0.047
 celebrate-0.041
 costumes-0.040
Token with
Token ID351
Feature activation-2.51e+00

Pos logit contributions
 parties+0.140
 tea+0.118
 recess+0.109
 outfits+0.097
 notebooks+0.096
Neg logit contributions
 beast-0.186
 gut-0.186
 mistake-0.182
cut-0.173
 warning-0.171
Token the
Token ID262
Feature activation-1.51e+00

Pos logit contributions
 parties+0.071
 tea+0.064
 recess+0.050
 celebrate+0.049
 picnic+0.047
Neg logit contributions
 beast-0.096
 gut-0.096
inct-0.092
 mistake-0.091
 therm-0.090
Token cone
Token ID27763
Feature activation+1.52e+00

Pos logit contributions
 parties+0.046
 tea+0.043
 picnic+0.042
 sandwiches+0.038
 party+0.036
Neg logit contributions
 gut-0.052
inct-0.050
 beast-0.046
 guardian-0.045
rum-0.045
Token,
Token ID11
Feature activation-2.20e+00

Pos logit contributions
 beast+0.065
 gut+0.065
 mistake+0.062
cut+0.062
 therm+0.061
Neg logit contributions
 parties-0.034
 tea-0.028
 recess-0.023
 celebrate-0.021
 picnic-0.019
Token rolling
Token ID10708
Feature activation-2.68e+00

Pos logit contributions
 parties+0.068
 tea+0.061
 celebrate+0.048
 recess+0.048
 party+0.048
Neg logit contributions
 gut-0.080
 beast-0.080
inct-0.077
 therm-0.074
 severe-0.073
Token to
Token ID284
Feature activation+1.54e+00

Pos logit contributions
 parties+0.013
 tea+0.012
 recess+0.011
 celebrate+0.010
ets+0.010
Neg logit contributions
 beast-0.009
 gut-0.009
 mistake-0.008
 therm-0.008
 adult-0.008
Token run
Token ID1057
Feature activation-3.28e+00

Pos logit contributions
 gut+0.050
 mistake+0.050
 adult+0.047
 beast+0.047
 warning+0.046
Neg logit contributions
 parties-0.041
 tea-0.038
text-0.034
ils-0.032
ouses-0.032
Token and
Token ID290
Feature activation-2.69e+00

Pos logit contributions
 parties+0.111
 tea+0.105
 celebrate+0.093
 birthday+0.089
 recess+0.089
Neg logit contributions
 beast-0.084
 gut-0.080
 mistake-0.077
 adult-0.072
 law-0.071
Token jump
Token ID4391
Feature activation-3.81e+00

Pos logit contributions
 parties+0.056
 tea+0.046
 recess+0.034
 celebrate+0.031
 party+0.029
Neg logit contributions
 beast-0.119
 gut-0.119
 mistake-0.114
 therm-0.113
 swollen-0.113
Token and
Token ID290
Feature activation-2.31e+00

Pos logit contributions
 parties+0.125
 tea+0.110
 recess+0.106
 celebrate+0.103
 wedding+0.102
Neg logit contributions
 beast-0.098
cut-0.098
inct-0.094
 gut-0.093
 therm-0.092
Token throw
Token ID3714
Feature activation-4.52e+00

Pos logit contributions
 parties+0.051
 tea+0.042
 recess+0.033
 celebrate+0.029
 party+0.028
Neg logit contributions
 beast-0.101
 gut-0.100
 mistake-0.096
 therm-0.094
 swollen-0.094
Token balls
Token ID11333
Feature activation-4.77e+00

Pos logit contributions
 parties+0.145
 tea+0.122
 celebrate+0.115
 recess+0.114
 wedding+0.111
Neg logit contributions
 beast-0.132
 mistake-0.124
 therm-0.121
 gut-0.119
 ang-0.118
Token.
Token ID13
Feature activation-9.47e-01

Pos logit contributions
 parties+0.158
 tea+0.143
 celebrate+0.128
 pic+0.125
 picnic+0.124
Neg logit contributions
cut-0.134
 gut-0.126
 beast-0.126
 mistake-0.120
 claw-0.120
Token He
Token ID679
Feature activation+6.26e-01

Pos logit contributions
 parties+0.031
 tea+0.027
 recess+0.024
ils+0.022
 celebrate+0.022
Neg logit contributions
 beast-0.031
 gut-0.031
 therm-0.030
 mistake-0.029
inct-0.029
Token was
Token ID373
Feature activation+6.62e-01

Pos logit contributions
 beast+0.020
 gut+0.020
 mistake+0.019
cut+0.018
 therm+0.018
Neg logit contributions
 parties-0.021
 tea-0.018
 recess-0.016
 celebrate-0.015
 birthday-0.015
Token very
Token ID845
Feature activation+2.18e+00

Pos logit contributions
 gut+0.020
 beast+0.020
 mistake+0.019
 therm+0.019
 swollen+0.019
Neg logit contributions
 parties-0.023
 tea-0.020
 recess-0.018
 celebrate-0.017
 picnic-0.017
Token day
Token ID1110
Feature activation+6.13e-01

Pos logit contributions
 beast+0.020
 passer+0.019
 mistake+0.019
 adult+0.019
 therm+0.019
Neg logit contributions
 parties-0.018
 tea-0.015
 skirts-0.013
 outfits-0.013
ils-0.013
Token,
Token ID11
Feature activation+2.18e-01

Pos logit contributions
 gut+0.018
 beast+0.018
 therm+0.017
 mistake+0.017
 passer+0.017
Neg logit contributions
 parties-0.021
 tea-0.018
 recess-0.016
 celebrate-0.016
ets-0.015
Token Lily
Token ID20037
Feature activation-1.72e+00

Pos logit contributions
 therm+0.008
 mistake+0.007
 gut+0.007
 swollen+0.007
 beast+0.007
Neg logit contributions
 parties-0.006
ets-0.005
ils-0.005
 recess-0.005
ag-0.004
Token wanted
Token ID2227
Feature activation-5.20e-02

Pos logit contributions
 parties+0.058
 tea+0.053
 recess+0.049
 celebrate+0.044
dates+0.041
Neg logit contributions
 beast-0.052
 gut-0.050
 mistake-0.050
 therm-0.048
 injured-0.047
Token to
Token ID284
Feature activation+3.37e+00

Pos logit contributions
 parties+0.002
 tea+0.002
 recess+0.001
 celebrate+0.001
 party+0.001
Neg logit contributions
 beast-0.002
 gut-0.002
inct-0.002
cut-0.002
 swollen-0.002
Token wear
Token ID5806
Feature activation-4.80e+00

Pos logit contributions
 gut+0.131
 beast+0.130
 mistake+0.128
 therm+0.124
 adult+0.123
Neg logit contributions
 parties-0.080
 tea-0.070
 recess-0.059
ets-0.057
ashed-0.056
Token her
Token ID607
Feature activation-2.60e+00

Pos logit contributions
 parties+0.149
 tea+0.129
 recess+0.114
 celebrate+0.109
ashed+0.107
Neg logit contributions
 gut-0.159
 beast-0.155
 therm-0.151
 swollen-0.148
 mistake-0.146
Token favorite
Token ID4004
Feature activation-3.12e+00

Pos logit contributions
 parties+0.069
 tea+0.058
 recess+0.048
 celebrate+0.045
 party+0.042
Neg logit contributions
 beast-0.104
 gut-0.104
 mistake-0.098
 therm-0.098
inct-0.097
Token dress
Token ID6576
Feature activation-2.24e+00

Pos logit contributions
 parties+0.104
 tea+0.087
 recess+0.082
ets+0.077
 celebrate+0.075
Neg logit contributions
 beast-0.099
 therm-0.098
 gut-0.096
 mistake-0.094
 swollen-0.091
Token to
Token ID284
Feature activation+1.63e+00

Pos logit contributions
 parties+0.068
 tea+0.061
 recess+0.054
 celebrate+0.050
 picnic+0.048
Neg logit contributions
 gut-0.077
 beast-0.077
 mistake-0.073
 therm-0.073
cut-0.070
Token school
Token ID1524
Feature activation-1.65e+00

Pos logit contributions
 mistake+0.058
 gut+0.058
 beast+0.056
 therm+0.054
 adult+0.053
Neg logit contributions
 parties-0.042
 tea-0.037
ashed-0.033
 recess-0.032
ets-0.030
Token will
Token ID481
Feature activation+8.18e-01

Pos logit contributions
 parties+0.135
 tea+0.116
 recess+0.099
 celebrate+0.083
 party+0.082
Neg logit contributions
 beast-0.197
 gut-0.194
 mistake-0.184
 therm-0.183
cut-0.183
Token come
Token ID1282
Feature activation-1.42e+00

Pos logit contributions
 gut+0.028
 beast+0.028
 mistake+0.026
 therm+0.026
inct+0.025
Neg logit contributions
 parties-0.026
 tea-0.023
 recess-0.020
 celebrate-0.018
 party-0.018
Token back
Token ID736
Feature activation+1.97e+00

Pos logit contributions
 parties+0.051
 tea+0.046
 recess+0.041
ils+0.039
ets+0.036
Neg logit contributions
 gut-0.044
 beast-0.041
 guardian-0.039
 mistake-0.039
 Owl-0.039
Token tomorrow
Token ID9439
Feature activation-2.93e+00

Pos logit contributions
 beast+0.071
 gut+0.071
 mistake+0.067
 therm+0.066
inct+0.066
Neg logit contributions
 parties-0.060
 tea-0.053
 recess-0.044
 celebrate-0.042
 party-0.040
Token and
Token ID290
Feature activation-1.42e+00

Pos logit contributions
 parties+0.086
 tea+0.076
 recess+0.065
 celebrate+0.060
 party+0.057
Neg logit contributions
 beast-0.105
 gut-0.105
 mistake-0.100
 therm-0.098
inct-0.096
Token play
Token ID711
Feature activation-6.01e+00

Pos logit contributions
 parties+0.028
 tea+0.023
 celebrate+0.018
 recess+0.016
 picnic+0.016
Neg logit contributions
 beast-0.067
 gut-0.066
inct-0.064
 mistake-0.063
 therm-0.063
Token with
Token ID351
Feature activation-3.36e+00

Pos logit contributions
 parties+0.202
 tea+0.172
 recess+0.162
ashed+0.160
 notebooks+0.153
Neg logit contributions
 beast-0.179
 gut-0.173
 mistake-0.167
cut-0.161
inct-0.154
Token you
Token ID345
Feature activation-5.21e+00

Pos logit contributions
 parties+0.095
 tea+0.081
 recess+0.068
 celebrate+0.063
 party+0.061
Neg logit contributions
 beast-0.130
 gut-0.129
 mistake-0.122
 therm-0.120
 swollen-0.119
Token,"
Token ID553
Feature activation-4.64e-01

Pos logit contributions
 parties+0.162
 tea+0.144
 recess+0.126
 celebrate+0.115
 picnic+0.111
Neg logit contributions
 beast-0.178
 gut-0.178
 mistake-0.170
 therm-0.165
 swollen-0.163
Token they
Token ID484
Feature activation-1.43e+00

Pos logit contributions
 parties+0.012
 tea+0.010
 recess+0.008
 celebrate+0.008
 party+0.007
Neg logit contributions
 beast-0.019
 gut-0.019
 mistake-0.018
inct-0.018
 therm-0.018
Token say
Token ID910
Feature activation-3.23e+00

Pos logit contributions
 parties+0.045
 tea+0.041
 recess+0.038
dates+0.036
ils+0.034
Neg logit contributions
 gut-0.044
 beast-0.044
 mistake-0.042
 therm-0.042
 plea-0.041