TOP ACTIVATIONS
MAX = 0.598

park . The k ang aroo is big and brown .
they see a k ang aroo in the park . The
find the big k ang aroo and asked for his help
The k ang aroo nods . It hops towards
runs after the k ang aroo , but it is too
boys ," the k ang aroo says . " I am
go catch the k ang aroo !" Ben says . He
The baby k ang aroo made a soft noise and
saw a big k ang aroo hopping near the trees .
play with the k ang aroo until it is time to
them to the k ang aroo , he struggled to eat
play with the k ang aroo . They have a lot
. " The k ang aroo is not from the park
The k ang aroo felt embarrassed , and quickly
The k ang aroo was so happy with the
aroo !" The k ang aroo sque aked in response ,
there was a k ang aroo . He loved to paint
aroo . The k ang aroo is hopping and eating grass
careful . The k ang aroo is not a toy .
" Hello , k ang aroo ," Anna says . "

MIN ACTIVATIONS
MIN = -0.609

get hurt or make a mess . And you 're making
they also made a big mess . They threw their toys
too . After all the mess was made , Sophia used
dad came and saw the mess . They were not happy
it looked like a big mess . The little boy thought
ceiling . The mess was so big that it
she had made a big mess ! Leaves were all over
odge began to spread his mess again . He jumped and
week and making a huge mess . It was so sm
. They only had a mess . <|endoftext|> S ara and
microwave and saw a black mess . Smoke came out .
loud noise and a big mess . Lily was scared and
. They look at the mess of blocks .
he would make a huge mess in his house , but
he couldn 't clear the mess by himself .
, the garden was a mess . Polly and her friends
throwing blocks and making a mess ? Why are you being
woke up to a big mess . The rh in oc
it was making a big mess . So , he grabbed
went he made a huge mess . One day his mom

INTERVAL 0.296 - 0.598
CONTAINS 1.434%

. That is very kind of you ." Tim my carried
art . It is made of paper and paint . It
day , a little boy named Tim went for a walk
home . L ily was very brave and smart
the ladder to the top of the tree and looked around

INTERVAL -0.005 - 0.296
CONTAINS 51.091%

a lot of trees and flowers along the way . As
But you have to be patient . Good things take time
a time , there was a little girl named Lily .
friends or playing games . She would always do her best
okay . He does not see the rock . The rock

INTERVAL -0.307 - -0.005
CONTAINS 47.019%

voice calling his name . ' Set o
called Ell a . She was three and liked to do
<|endoftext|> Once upon a time , there was a little girl
little girl named Lily . She loved to eat candy and
, " My name is Tom . And this is Spot

INTERVAL -0.609 - -0.307
CONTAINS 0.457%

. They had completed their fun . They smiled at each
. They had lots of fun and laughter and never stopped
a big bite . It tasted so delicious . She ate
make the puzzle on the floor of their room .
come home late , and not pay any attention to his
Token park
Token ID3952
Feature activation-3.29e-02

Pos logit contributions
aroo+0.189
ter+0.171
apped+0.160
ovable+0.157
itte+0.156
Neg logit contributions
 mess-0.115
 wra-0.091
aj-0.091
 pic-0.088
 slippery-0.087
Token.
Token ID13
Feature activation+5.98e-02

Pos logit contributions
 mess+0.060
aj+0.050
 wra+0.049
onna+0.047
 slippery+0.047
Neg logit contributions
aroo-0.071
ter-0.064
itte-0.057
ovable-0.056
crow-0.056
Token The
Token ID383
Feature activation+2.24e-01

Pos logit contributions
aroo+0.102
ter+0.089
crow+0.077
apped+0.076
itte+0.073
Neg logit contributions
 mess-0.136
aj-0.123
 wra-0.117
onna-0.116
 pic-0.114
Token k
Token ID479
Feature activation+1.17e-01

Pos logit contributions
aroo+0.465
ovable+0.460
onel+0.455
itte+0.442
apped+0.436
Neg logit contributions
 mess-0.370
 slippery-0.305
 wra-0.283
 fun-0.280
 steep-0.276
Tokenang
Token ID648
Feature activation+3.09e-01

Pos logit contributions
aroo+0.140
ter+0.132
ovable+0.119
itte+0.118
 citizens+0.117
Neg logit contributions
 mess-0.272
aj-0.252
 wra-0.249
liv-0.247
 slippery-0.243
Tokenaroo
Token ID38049
Feature activation+5.97e-01

Pos logit contributions
onel+0.114
ter+0.105
 crew+0.096
ovable+0.093
 citizens+0.092
Neg logit contributions
 mess-0.895
 slippery-0.856
liv-0.855
 steep-0.838
 wra-0.837
Token is
Token ID318
Feature activation-1.22e-01

Pos logit contributions
aroo+1.436
ter+1.286
crow+1.195
apped+1.164
itte+1.149
Neg logit contributions
 mess-0.896
aj-0.810
 wra-0.737
onna-0.729
 fault-0.717
Token big
Token ID1263
Feature activation-6.90e-02

Pos logit contributions
 mess+0.220
aj+0.197
 wra+0.184
onna+0.181
 pic+0.177
Neg logit contributions
aroo-0.266
ter-0.240
crow-0.212
apped-0.211
itte-0.206
Token and
Token ID290
Feature activation-5.49e-02

Pos logit contributions
 mess+0.124
aj+0.118
liv+0.112
onna+0.111
aman+0.107
Neg logit contributions
aroo-0.141
ter-0.131
crow-0.115
apped-0.110
 member-0.110
Token brown
Token ID7586
Feature activation-5.01e-02

Pos logit contributions
 mess+0.097
aj+0.090
aman+0.083
 fault+0.083
onna+0.080
Neg logit contributions
aroo-0.121
ter-0.106
crow-0.096
itte-0.092
apped-0.091
Token.
Token ID13
Feature activation+6.10e-02

Pos logit contributions
 mess+0.089
aj+0.076
 wra+0.076
onna+0.072
 fault+0.071
Neg logit contributions
aroo-0.108
ter-0.095
itte-0.087
crow-0.085
apped-0.085
Token they
Token ID484
Feature activation+6.69e-03

Pos logit contributions
aroo+0.051
ter+0.046
apped+0.040
crow+0.038
 member+0.037
Neg logit contributions
 mess-0.057
aj-0.052
onna-0.050
 fault-0.048
icol-0.047
Token see
Token ID766
Feature activation+1.19e-01

Pos logit contributions
aroo+0.015
ter+0.013
apped+0.011
crow+0.011
ovable+0.011
Neg logit contributions
 mess-0.013
 pic-0.011
 slippery-0.010
 wra-0.010
 pays-0.010
Token a
Token ID257
Feature activation+3.56e-02

Pos logit contributions
aroo+0.251
ter+0.224
crow+0.200
apped+0.197
itte+0.194
Neg logit contributions
 mess-0.222
aj-0.197
 wra-0.187
onna-0.182
 pic-0.179
Token k
Token ID479
Feature activation+1.15e-01

Pos logit contributions
aroo+0.079
ter+0.072
crow+0.064
apped+0.062
itte+0.062
Neg logit contributions
 mess-0.060
aj-0.055
onna-0.050
 wra-0.050
 fault-0.049
Tokenang
Token ID648
Feature activation+3.08e-01

Pos logit contributions
aroo+0.142
ter+0.130
ovable+0.122
 citizens+0.120
 sailors+0.120
Neg logit contributions
 mess-0.268
 wra-0.246
aj-0.241
liv-0.237
 slippery-0.233
Tokenaroo
Token ID38049
Feature activation+5.96e-01

Pos logit contributions
 citizens+0.197
onel+0.171
 crew+0.155
 sailors+0.150
ovable+0.148
Neg logit contributions
 mess-0.765
 snowball-0.761
 slippery-0.760
liv-0.742
gh-0.731
Token in
Token ID287
Feature activation+2.87e-01

Pos logit contributions
aroo+1.302
ter+1.158
itte+1.128
ovable+1.118
 citizens+1.099
Neg logit contributions
 mess-1.012
 wra-0.773
 slippery-0.763
aj-0.736
onna-0.731
Token the
Token ID262
Feature activation+7.58e-02

Pos logit contributions
aroo+0.496
ter+0.425
crow+0.383
 member+0.349
apped+0.346
Neg logit contributions
 mess-0.647
aj-0.622
onna-0.567
 wra-0.564
icol-0.551
Token park
Token ID3952
Feature activation-3.29e-02

Pos logit contributions
aroo+0.189
ter+0.171
apped+0.160
ovable+0.157
itte+0.156
Neg logit contributions
 mess-0.115
 wra-0.091
aj-0.091
 pic-0.088
 slippery-0.087
Token.
Token ID13
Feature activation+5.98e-02

Pos logit contributions
 mess+0.060
aj+0.050
 wra+0.049
onna+0.047
 slippery+0.047
Neg logit contributions
aroo-0.071
ter-0.064
itte-0.057
ovable-0.056
crow-0.056
Token The
Token ID383
Feature activation+2.24e-01

Pos logit contributions
aroo+0.102
ter+0.089
crow+0.077
apped+0.076
itte+0.073
Neg logit contributions
 mess-0.136
aj-0.123
 wra-0.117
onna-0.116
 pic-0.114
Token find
Token ID1064
Feature activation+1.58e-01

Pos logit contributions
 mess+0.109
aj+0.101
 wra+0.095
onna+0.092
liv+0.091
Neg logit contributions
aroo-0.097
ter-0.086
crow-0.074
apped-0.072
 member-0.072
Token the
Token ID262
Feature activation+7.99e-02

Pos logit contributions
aroo+0.373
ter+0.345
apped+0.326
ive+0.310
itte+0.309
Neg logit contributions
 mess-0.220
 slippery-0.188
 wra-0.178
 fun-0.177
 scar-0.177
Token big
Token ID1263
Feature activation-6.48e-02

Pos logit contributions
aroo+0.173
ter+0.153
apped+0.141
crow+0.138
itte+0.136
Neg logit contributions
 mess-0.149
aj-0.127
 wra-0.122
onna-0.119
 fault-0.118
Token k
Token ID479
Feature activation+1.17e-01

Pos logit contributions
 mess+0.113
aj+0.106
 wra+0.096
liv+0.096
onna+0.095
Neg logit contributions
aroo-0.139
ter-0.127
crow-0.113
apped-0.108
 member-0.107
Tokenang
Token ID648
Feature activation+3.09e-01

Pos logit contributions
aroo+0.146
 sailors+0.139
 citizens+0.136
ter+0.136
 Queen+0.135
Neg logit contributions
 mess-0.226
aj-0.224
zan-0.218
 wra-0.216
onna-0.216
Tokenaroo
Token ID38049
Feature activation+5.96e-01

Pos logit contributions
 citizens+0.136
apped+0.133
 sailors+0.124
 married+0.116
ovable+0.108
Neg logit contributions
 slippery-0.782
onna-0.764
 steep-0.759
 mess-0.759
gh-0.752
Token and
Token ID290
Feature activation-5.31e-02

Pos logit contributions
aroo+1.450
ter+1.297
apped+1.222
crow+1.220
itte+1.203
Neg logit contributions
 mess-0.923
aj-0.737
 wra-0.720
 slippery-0.702
onna-0.701
Token asked
Token ID1965
Feature activation+1.21e-01

Pos logit contributions
 mess+0.094
aj+0.080
 wra+0.078
onna+0.077
 slippery+0.077
Neg logit contributions
aroo-0.115
ter-0.104
crow-0.095
apped-0.094
itte-0.092
Token for
Token ID329
Feature activation+6.38e-02

Pos logit contributions
aroo+0.288
ter+0.255
apped+0.238
crow+0.236
itte+0.234
Neg logit contributions
 mess-0.179
 wra-0.144
onna-0.143
 mistake-0.142
 pays-0.141
Token his
Token ID465
Feature activation-2.84e-02

Pos logit contributions
aroo+0.109
ter+0.098
apped+0.092
itte+0.089
crow+0.088
Neg logit contributions
 mess-0.135
 wra-0.119
 slippery-0.118
 pic-0.118
 fun-0.117
Token help
Token ID1037
Feature activation+3.18e-01

Pos logit contributions
 mess+0.065
aj+0.059
 fault+0.057
 wra+0.056
onna+0.056
Neg logit contributions
aroo-0.048
ter-0.043
apped-0.038
crow-0.038
itte-0.036
Token
Token ID198
Feature activation+2.71e-02

Pos logit contributions
aroo+0.087
ter+0.080
apped+0.071
crow+0.070
itte+0.070
Neg logit contributions
 mess-0.058
aj-0.050
onna-0.049
 wra-0.048
 pic-0.048
Token
Token ID198
Feature activation+2.70e-02

Pos logit contributions
aroo+0.056
ter+0.050
crow+0.045
apped+0.044
itte+0.043
Neg logit contributions
 mess-0.052
aj-0.046
 wra-0.043
onna-0.043
 pic-0.042
TokenThe
Token ID464
Feature activation+1.00e-01

Pos logit contributions
aroo+0.055
ter+0.051
apped+0.046
ovable+0.045
itte+0.045
Neg logit contributions
 mess-0.049
aj-0.043
 wra-0.042
 slippery-0.041
onna-0.041
Token k
Token ID479
Feature activation+1.13e-01

Pos logit contributions
aroo+0.203
ter+0.184
apped+0.179
ovable+0.178
onel+0.175
Neg logit contributions
 mess-0.188
 slippery-0.155
 wra-0.146
 steep-0.144
aj-0.143
Tokenang
Token ID648
Feature activation+3.09e-01

Pos logit contributions
ter+0.146
aroo+0.145
 citizens+0.142
ovable+0.141
itte+0.139
Neg logit contributions
 mess-0.227
aj-0.210
 wra-0.210
 snowball-0.208
 slippery-0.206
Tokenaroo
Token ID38049
Feature activation+5.95e-01

Pos logit contributions
onel+0.154
 citizens+0.144
ovable+0.135
 crew+0.126
 sailors+0.119
Neg logit contributions
 mess-0.817
 slippery-0.803
 wob-0.799
 snowball-0.797
liv-0.779
Token nods
Token ID34833
Feature activation+3.24e-03

Pos logit contributions
aroo+1.432
ter+1.286
crow+1.206
apped+1.172
itte+1.157
Neg logit contributions
 mess-0.854
aj-0.779
 wra-0.710
liv-0.702
onna-0.699
Token.
Token ID13
Feature activation+5.47e-02

Pos logit contributions
aroo+0.007
ter+0.006
crow+0.005
apped+0.005
itte+0.005
Neg logit contributions
 mess-0.006
aj-0.006
 wra-0.005
onna-0.005
 pic-0.005
Token It
Token ID632
Feature activation+9.18e-02

Pos logit contributions
aroo+0.120
ter+0.112
 member+0.099
crow+0.096
 pilot+0.095
Neg logit contributions
 mess-0.090
onna-0.083
aj-0.083
 wra-0.076
 pic-0.076
Token hops
Token ID29438
Feature activation+1.05e-02

Pos logit contributions
aroo+0.215
ter+0.193
crow+0.176
apped+0.173
itte+0.172
Neg logit contributions
 mess-0.145
aj-0.129
 wra-0.119
aman-0.116
onna-0.116
Token towards
Token ID3371
Feature activation+1.21e-01

Pos logit contributions
aroo+0.020
ter+0.018
itte+0.015
crow+0.015
apped+0.015
Neg logit contributions
 mess-0.021
aj-0.019
 wra-0.018
onna-0.018
 slippery-0.018
Token runs
Token ID4539
Feature activation-9.24e-02

Pos logit contributions
aroo+0.094
ter+0.082
apped+0.076
itte+0.076
crow+0.074
Neg logit contributions
 mess-0.066
aj-0.059
onna-0.056
liv-0.055
 fault-0.054
Token after
Token ID706
Feature activation+7.75e-02

Pos logit contributions
 mess+0.187
aj+0.172
onna+0.157
 wra+0.157
 fault+0.156
Neg logit contributions
aroo-0.184
ter-0.162
crow-0.147
apped-0.139
 member-0.138
Token the
Token ID262
Feature activation+7.36e-02

Pos logit contributions
aroo+0.178
ter+0.164
itte+0.153
apped+0.152
ovable+0.151
Neg logit contributions
 mess-0.121
 wra-0.098
onna-0.096
aj-0.096
 pic-0.096
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
aroo+0.146
ter+0.135
apped+0.132
ovable+0.131
itte+0.130
Neg logit contributions
 mess-0.136
 slippery-0.114
 wra-0.110
 steep-0.105
 fuss-0.104
Tokenang
Token ID648
Feature activation+3.10e-01

Pos logit contributions
ter+0.142
aroo+0.141
 Queen+0.138
itte+0.136
ovable+0.135
Neg logit contributions
 mess-0.231
aj-0.225
liv-0.220
 wra-0.219
 slippery-0.214
Tokenaroo
Token ID38049
Feature activation+5.95e-01

Pos logit contributions
onel+0.150
 citizens+0.149
ovable+0.140
 crew+0.121
 sailors+0.118
Neg logit contributions
 slippery-0.798
 mess-0.789
liv-0.784
 snowball-0.778
 wob-0.775
Token,
Token ID11
Feature activation+2.88e-02

Pos logit contributions
aroo+1.425
ter+1.294
crow+1.175
itte+1.175
apped+1.173
Neg logit contributions
 mess-0.921
aj-0.776
 wra-0.746
onna-0.731
 pic-0.709
Token but
Token ID475
Feature activation-1.64e-01

Pos logit contributions
aroo+0.068
ter+0.062
crow+0.054
apped+0.052
 member+0.052
Neg logit contributions
 mess-0.046
aj-0.043
icol-0.038
 wra-0.037
onna-0.037
Token it
Token ID340
Feature activation-6.68e-02

Pos logit contributions
 mess+0.283
aj+0.247
 wra+0.236
 slippery+0.233
onna+0.229
Neg logit contributions
aroo-0.354
ter-0.318
crow-0.289
apped-0.288
itte-0.286
Token is
Token ID318
Feature activation-1.24e-01

Pos logit contributions
 mess+0.096
aj+0.094
aman+0.084
icol+0.083
onna+0.083
Neg logit contributions
aroo-0.160
ter-0.141
crow-0.132
itte-0.126
apped-0.125
Token too
Token ID1165
Feature activation-4.57e-02

Pos logit contributions
icol+0.269
aman+0.258
aj+0.256
 mess+0.252
array+0.243
Neg logit contributions
aroo-0.215
ter-0.198
 protector-0.154
 hero-0.145
apped-0.145
Token boys
Token ID6510
Feature activation+1.71e-01

Pos logit contributions
aroo+0.209
ter+0.178
itte+0.176
apped+0.175
ovable+0.174
Neg logit contributions
 mess-0.235
 wra-0.201
aj-0.200
 slippery-0.195
aman-0.194
Token,"
Token ID553
Feature activation+6.03e-02

Pos logit contributions
aroo+0.425
apped+0.375
ter+0.373
itte+0.372
ovable+0.354
Neg logit contributions
 mess-0.266
 wra-0.192
onna-0.188
 fuss-0.182
aj-0.182
Token the
Token ID262
Feature activation+6.94e-02

Pos logit contributions
aroo+0.136
ter+0.123
apped+0.114
crow+0.111
itte+0.111
Neg logit contributions
 mess-0.101
aj-0.084
 wra-0.083
onna-0.083
 pic-0.082
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
aroo+0.135
ter+0.124
apped+0.120
ovable+0.117
ive+0.117
Neg logit contributions
 mess-0.129
 slippery-0.109
aj-0.105
 wra-0.104
aman-0.102
Tokenang
Token ID648
Feature activation+3.06e-01

Pos logit contributions
ter+0.157
 citizens+0.153
aroo+0.149
itte+0.146
 crew+0.145
Neg logit contributions
 mess-0.211
aj-0.203
 snowball-0.203
liv-0.200
 slippery-0.200
Tokenaroo
Token ID38049
Feature activation+5.93e-01

Pos logit contributions
onel+0.145
 citizens+0.138
ive+0.124
 alien+0.117
ter+0.113
Neg logit contributions
 mess-0.786
 slippery-0.771
 snowball-0.766
 wob-0.759
liv-0.754
Token says
Token ID1139
Feature activation-1.32e-01

Pos logit contributions
aroo+1.403
ter+1.262
crow+1.194
apped+1.131
itte+1.128
Neg logit contributions
 mess-0.906
aj-0.819
 wra-0.754
onna-0.740
liv-0.728
Token.
Token ID13
Feature activation+5.83e-02

Pos logit contributions
 mess+0.210
aj+0.179
onna+0.173
 wra+0.170
 fault+0.162
Neg logit contributions
aroo-0.315
ter-0.277
crow-0.259
itte-0.254
 member-0.245
Token "
Token ID366
Feature activation+5.23e-02

Pos logit contributions
aroo+0.117
ter+0.105
crow+0.093
apped+0.092
itte+0.089
Neg logit contributions
 mess-0.115
aj-0.102
 wra-0.097
onna-0.095
 pic-0.093
TokenI
Token ID40
Feature activation+2.23e-01

Pos logit contributions
aroo+0.105
ter+0.092
ovable+0.092
itte+0.089
apped+0.088
Neg logit contributions
 mess-0.088
 wra-0.078
onna-0.075
 slippery-0.072
aj-0.070
Token am
Token ID716
Feature activation+1.03e-01

Pos logit contributions
aroo+0.455
ter+0.406
apped+0.358
crow+0.355
itte+0.344
Neg logit contributions
 mess-0.441
aj-0.382
onna-0.371
 wra-0.362
 fault-0.356
Token go
Token ID467
Feature activation+9.96e-02

Pos logit contributions
 mess+0.052
 pic+0.044
 slippery+0.043
aj+0.042
onna+0.041
Neg logit contributions
aroo-0.057
ter-0.052
apped-0.049
itte-0.046
ovable-0.046
Token catch
Token ID4929
Feature activation+9.13e-02

Pos logit contributions
aroo+0.216
ter+0.197
apped+0.181
crow+0.179
itte+0.178
Neg logit contributions
 mess-0.170
aj-0.147
onna-0.139
 slippery-0.136
 pic-0.134
Token the
Token ID262
Feature activation+7.21e-02

Pos logit contributions
aroo+0.198
ter+0.177
apped+0.164
itte+0.164
ovable+0.161
Neg logit contributions
 mess-0.163
aj-0.135
 wra-0.133
 pic-0.129
 slippery-0.128
Token k
Token ID479
Feature activation+1.12e-01

Pos logit contributions
aroo+0.152
apped+0.138
itte+0.137
ovable+0.137
ive+0.132
Neg logit contributions
 mess-0.128
 slippery-0.104
 wra-0.102
 steep-0.099
 fun-0.098
Tokenang
Token ID648
Feature activation+3.08e-01

Pos logit contributions
ovable+0.146
 citizens+0.143
itte+0.142
 Queen+0.140
 crew+0.139
Neg logit contributions
 mess-0.212
 wra-0.207
aj-0.204
liv-0.201
corn-0.197
Tokenaroo
Token ID38049
Feature activation+5.93e-01

Pos logit contributions
onel+0.202
ovable+0.195
 citizens+0.183
itte+0.156
 sailors+0.151
Neg logit contributions
 mess-0.750
 slippery-0.742
 snowball-0.737
 wra-0.717
liv-0.716
Token!"
Token ID2474
Feature activation-1.29e-01

Pos logit contributions
aroo+1.567
ter+1.414
itte+1.364
apped+1.348
crow+1.344
Neg logit contributions
 mess-0.786
aj-0.614
 wra-0.569
onna-0.560
 slippery-0.547
Token Ben
Token ID3932
Feature activation-6.36e-02

Pos logit contributions
 mess+0.201
aj+0.176
 wra+0.162
onna+0.160
 fault+0.155
Neg logit contributions
aroo-0.309
ter-0.280
crow-0.255
apped-0.252
itte-0.247
Token says
Token ID1139
Feature activation-1.34e-01

Pos logit contributions
 mess+0.109
aj+0.099
icol+0.095
onna+0.092
aman+0.092
Neg logit contributions
aroo-0.139
ter-0.122
crow-0.118
 member-0.112
apped-0.111
Token.
Token ID13
Feature activation+5.93e-02

Pos logit contributions
 mess+0.230
aj+0.201
onna+0.195
 wra+0.189
 fault+0.183
Neg logit contributions
aroo-0.296
ter-0.268
crow-0.242
itte-0.234
 pilot-0.228
Token He
Token ID679
Feature activation+4.09e-02

Pos logit contributions
aroo+0.132
ter+0.124
ovable+0.108
itte+0.106
crow+0.106
Neg logit contributions
 mess-0.097
onna-0.086
aj-0.086
 wra-0.082
 pic-0.079
Token
Token ID198
Feature activation+1.84e-02

Pos logit contributions
aroo+0.036
ter+0.032
crow+0.029
apped+0.028
itte+0.028
Neg logit contributions
 mess-0.034
aj-0.031
 wra-0.029
onna-0.028
 pic-0.028
TokenThe
Token ID464
Feature activation+1.05e-01

Pos logit contributions
aroo+0.035
ter+0.032
crow+0.027
apped+0.027
itte+0.027
Neg logit contributions
 mess-0.038
aj-0.034
 wra-0.033
onna-0.032
 pic-0.031
Token baby
Token ID5156
Feature activation+2.13e-01

Pos logit contributions
aroo+0.199
apped+0.189
ter+0.184
ive+0.181
onel+0.180
Neg logit contributions
 mess-0.205
 mistake-0.168
 slippery-0.166
 wra-0.161
 fun-0.159
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
aroo+0.348
ovable+0.327
onel+0.321
apped+0.318
itte+0.317
Neg logit contributions
 mess-0.426
 wra-0.390
 scar-0.378
aj-0.370
 wob-0.362
Tokenang
Token ID648
Feature activation+3.06e-01

Pos logit contributions
 sailors+0.160
 citizens+0.159
 crew+0.158
rest+0.156
ter+0.155
Neg logit contributions
aj-0.203
liv-0.194
 wra-0.189
 mess-0.188
 snowball-0.180
Tokenaroo
Token ID38049
Feature activation+5.93e-01

Pos logit contributions
 citizens+0.125
onel+0.122
 sailors+0.119
 crew+0.112
ovable+0.099
Neg logit contributions
liv-0.823
 wob-0.804
 mess-0.788
 wra-0.777
 slippery-0.775
Token made
Token ID925
Feature activation-4.31e-02

Pos logit contributions
aroo+1.414
ter+1.289
crow+1.176
apped+1.152
itte+1.146
Neg logit contributions
 mess-0.931
aj-0.797
 wra-0.764
onna-0.731
 pic-0.715
Token a
Token ID257
Feature activation+3.22e-02

Pos logit contributions
 mess+0.082
aj+0.074
 wra+0.069
onna+0.069
 pic+0.067
Neg logit contributions
aroo-0.088
ter-0.079
crow-0.071
apped-0.070
itte-0.068
Token soft
Token ID2705
Feature activation-1.15e-01

Pos logit contributions
aroo+0.068
ter+0.060
apped+0.054
 pilot+0.054
 hero+0.054
Neg logit contributions
aj-0.053
 mess-0.051
 pic-0.048
liv-0.048
 fault-0.048
Token noise
Token ID7838
Feature activation-1.35e-01

Pos logit contributions
 mess+0.226
aj+0.216
 wra+0.203
 pic+0.199
 fault+0.198
Neg logit contributions
aroo-0.225
ter-0.200
apped-0.175
crow-0.173
 member-0.162
Token and
Token ID290
Feature activation-6.79e-02

Pos logit contributions
 mess+0.216
 wra+0.168
onna+0.167
aj+0.166
 slippery+0.160
Neg logit contributions
aroo-0.328
ter-0.288
crow-0.269
itte-0.268
apped-0.262
Token saw
Token ID2497
Feature activation+4.84e-02

Pos logit contributions
aroo+0.033
ter+0.029
crow+0.027
ive+0.026
ovable+0.025
Neg logit contributions
 mess-0.026
 wra-0.022
aj-0.021
 slippery-0.021
 pic-0.021
Token a
Token ID257
Feature activation+3.86e-02

Pos logit contributions
aroo+0.106
ter+0.094
crow+0.084
apped+0.084
itte+0.083
Neg logit contributions
 mess-0.086
aj-0.073
 wra-0.072
 pic-0.069
 slippery-0.068
Token big
Token ID1263
Feature activation-6.63e-02

Pos logit contributions
aroo+0.084
ter+0.075
crow+0.066
 member+0.065
 pilot+0.064
Neg logit contributions
 mess-0.066
aj-0.063
 fault-0.058
onna-0.058
 wra-0.056
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
 mess+0.124
aj+0.117
onna+0.108
 fault+0.107
 wra+0.106
Neg logit contributions
aroo-0.135
ter-0.121
crow-0.107
apped-0.103
 member-0.102
Tokenang
Token ID648
Feature activation+3.08e-01

Pos logit contributions
aroo+0.136
ter+0.125
ovable+0.117
itte+0.113
 citizens+0.111
Neg logit contributions
 mess-0.273
aj-0.254
 wra-0.252
liv-0.242
 slippery-0.239
Tokenaroo
Token ID38049
Feature activation+5.93e-01

Pos logit contributions
onel+0.130
 citizens+0.125
ovable+0.115
 crew+0.091
 sailors+0.085
Neg logit contributions
 mess-0.858
liv-0.828
 slippery-0.821
 snowball-0.815
 wra-0.803
Token hopping
Token ID47153
Feature activation+8.20e-02

Pos logit contributions
aroo+1.233
ter+1.103
ovable+1.049
itte+1.048
crow+1.040
Neg logit contributions
 mess-1.075
 wra-0.870
aj-0.867
 slippery-0.835
onna-0.827
Token near
Token ID1474
Feature activation+2.33e-01

Pos logit contributions
aroo+0.136
ter+0.120
ovable+0.112
 pilot+0.110
itte+0.109
Neg logit contributions
 mess-0.169
 slippery-0.151
 steep-0.148
 wra-0.147
aj-0.140
Token the
Token ID262
Feature activation+7.79e-02

Pos logit contributions
aroo+0.429
ter+0.383
crow+0.337
apped+0.323
 member+0.315
Neg logit contributions
 mess-0.488
aj-0.461
onna-0.427
 wra-0.421
liv-0.412
Token trees
Token ID7150
Feature activation+5.21e-02

Pos logit contributions
aroo+0.175
apped+0.151
ter+0.148
ovable+0.147
itte+0.145
Neg logit contributions
 mess-0.139
 slippery-0.111
 wra-0.111
 steep-0.106
aj-0.102
Token.
Token ID13
Feature activation+6.75e-02

Pos logit contributions
aroo+0.109
ter+0.097
crow+0.088
apped+0.086
itte+0.086
Neg logit contributions
 mess-0.097
aj-0.084
 wra-0.081
onna-0.080
 pic-0.077
Token play
Token ID711
Feature activation+1.42e-01

Pos logit contributions
 mess+0.005
onna+0.004
 pic+0.004
aj+0.004
 wra+0.004
Neg logit contributions
aroo-0.005
ter-0.005
apped-0.004
crow-0.004
itte-0.004
Token with
Token ID351
Feature activation+1.26e-01

Pos logit contributions
aroo+0.300
ter+0.260
itte+0.250
crow+0.246
ovable+0.239
Neg logit contributions
 mess-0.259
 pic-0.216
 flips-0.211
 fun-0.210
 wra-0.206
Token the
Token ID262
Feature activation+6.54e-02

Pos logit contributions
aroo+0.276
ter+0.248
itte+0.226
apped+0.225
crow+0.221
Neg logit contributions
 mess-0.224
 wra-0.189
aj-0.188
 pic-0.183
 slippery-0.179
Token k
Token ID479
Feature activation+1.11e-01

Pos logit contributions
aroo+0.132
ter+0.115
apped+0.114
itte+0.112
ovable+0.111
Neg logit contributions
 mess-0.131
 slippery-0.106
 wra-0.105
 fault-0.103
 steep-0.101
Tokenang
Token ID648
Feature activation+3.06e-01

Pos logit contributions
aroo+0.140
ter+0.134
itte+0.132
 Queen+0.130
 citizens+0.127
Neg logit contributions
 mess-0.234
aj-0.226
 wra-0.219
 slippery-0.215
liv-0.215
Tokenaroo
Token ID38049
Feature activation+5.92e-01

Pos logit contributions
onel+0.154
 citizens+0.147
ovable+0.130
ive+0.123
itte+0.117
Neg logit contributions
 snowball-0.776
 slippery-0.776
 mess-0.769
liv-0.768
onna-0.741
Token until
Token ID1566
Feature activation-9.09e-02

Pos logit contributions
aroo+1.449
ter+1.275
itte+1.257
crow+1.221
apped+1.173
Neg logit contributions
 mess-0.920
aj-0.703
 wra-0.695
 pic-0.695
 fault-0.692
Token it
Token ID340
Feature activation-7.00e-02

Pos logit contributions
 mess+0.127
 wra+0.097
 pic+0.092
 slippery+0.092
onna+0.089
Neg logit contributions
aroo-0.236
ter-0.212
itte-0.204
ovable-0.203
crow-0.198
Token is
Token ID318
Feature activation-1.31e-01

Pos logit contributions
 mess+0.108
aj+0.101
 wra+0.093
aman+0.092
liv+0.091
Neg logit contributions
aroo-0.158
ter-0.143
crow-0.135
apped-0.133
ive-0.126
Token time
Token ID640
Feature activation-2.34e-01

Pos logit contributions
 mess+0.200
 wra+0.192
liv+0.187
aj+0.186
icol+0.184
Neg logit contributions
ter-0.279
aroo-0.279
apped-0.239
 sailors-0.233
 pilot-0.226
Token to
Token ID284
Feature activation-5.35e-02

Pos logit contributions
 mess+0.455
aj+0.406
 wra+0.400
liv+0.395
aman+0.390
Neg logit contributions
aroo-0.445
ter-0.398
apped-0.368
 member-0.347
crow-0.346
Token them
Token ID606
Feature activation-4.49e-02

Pos logit contributions
aroo+0.436
ter+0.394
crow+0.353
apped+0.346
itte+0.339
Neg logit contributions
 mess-0.362
aj-0.322
 wra-0.300
onna-0.298
 fault-0.288
Token to
Token ID284
Feature activation-5.55e-02

Pos logit contributions
 mess+0.076
 wra+0.065
aj+0.063
 slippery+0.063
 pic+0.061
Neg logit contributions
aroo-0.103
ter-0.090
apped-0.084
 pilot-0.082
itte-0.081
Token the
Token ID262
Feature activation+7.88e-02

Pos logit contributions
 mess+0.103
aj+0.087
 wra+0.085
onna+0.083
 pic+0.083
Neg logit contributions
aroo-0.120
ter-0.105
apped-0.098
itte-0.096
crow-0.096
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
aroo+0.157
apped+0.137
itte+0.131
ovable+0.130
ter+0.129
Neg logit contributions
 mess-0.162
 slippery-0.132
 wra-0.130
aman-0.127
 fun-0.124
Tokenang
Token ID648
Feature activation+3.05e-01

Pos logit contributions
aroo+0.141
ter+0.134
itte+0.122
 citizens+0.121
ovable+0.118
Neg logit contributions
aj-0.256
 mess-0.256
liv-0.242
 wra-0.236
aman-0.234
Tokenaroo
Token ID38049
Feature activation+5.92e-01

Pos logit contributions
 citizens+0.145
onel+0.141
 sailors+0.138
ovable+0.112
 astronaut+0.099
Neg logit contributions
liv-0.816
 slippery-0.791
 mess-0.785
aman-0.778
 snowball-0.769
Token,
Token ID11
Feature activation+3.12e-02

Pos logit contributions
aroo+1.475
ter+1.327
apped+1.251
itte+1.243
crow+1.219
Neg logit contributions
 mess-0.867
aj-0.716
onna-0.682
 wra-0.652
 slippery-0.640
Token he
Token ID339
Feature activation+3.90e-02

Pos logit contributions
aroo+0.074
ter+0.067
apped+0.065
ive+0.063
crow+0.063
Neg logit contributions
 mess-0.048
 wra-0.042
 slippery-0.038
 pic-0.036
aj-0.036
Token struggled
Token ID11615
Feature activation-1.08e-01

Pos logit contributions
aroo+0.090
ter+0.081
crow+0.074
apped+0.072
itte+0.071
Neg logit contributions
 mess-0.066
aj-0.057
 wra-0.054
onna-0.052
 pic-0.050
Token to
Token ID284
Feature activation-5.37e-02

Pos logit contributions
 mess+0.217
aj+0.205
 pic+0.195
icol+0.193
aman+0.191
Neg logit contributions
aroo-0.197
ter-0.183
apped-0.157
 member-0.154
crow-0.150
Token eat
Token ID4483
Feature activation+2.66e-02

Pos logit contributions
 mess+0.109
aj+0.098
 wra+0.093
onna+0.092
 pic+0.089
Neg logit contributions
aroo-0.104
ter-0.093
crow-0.082
apped-0.080
itte-0.079
Token play
Token ID711
Feature activation+1.45e-01

Pos logit contributions
 mess+0.123
onna+0.102
 wra+0.100
 slippery+0.099
aj+0.098
Neg logit contributions
aroo-0.159
ter-0.146
crow-0.131
itte-0.131
apped-0.129
Token with
Token ID351
Feature activation+1.26e-01

Pos logit contributions
aroo+0.314
itte+0.267
ter+0.265
ovable+0.259
 pilot+0.253
Neg logit contributions
 mess-0.271
 pic-0.219
 fun-0.218
 wra-0.211
 flips-0.208
Token the
Token ID262
Feature activation+7.01e-02

Pos logit contributions
aroo+0.314
ter+0.287
onel+0.271
itte+0.270
apped+0.266
Neg logit contributions
 mess-0.184
 slippery-0.153
 pic-0.146
 fun-0.142
 wra-0.141
Token k
Token ID479
Feature activation+1.15e-01

Pos logit contributions
aroo+0.136
ter+0.119
apped+0.114
itte+0.112
ovable+0.111
Neg logit contributions
 mess-0.145
 slippery-0.119
 wra-0.118
aj-0.117
 fault-0.113
Tokenang
Token ID648
Feature activation+3.05e-01

Pos logit contributions
 Queen+0.149
itte+0.148
aroo+0.147
ovable+0.144
 citizens+0.143
Neg logit contributions
 mess-0.223
aj-0.219
liv-0.214
 slippery-0.214
 wra-0.213
Tokenaroo
Token ID38049
Feature activation+5.92e-01

Pos logit contributions
onel+0.179
 citizens+0.166
ovable+0.164
ive+0.155
 alien+0.147
Neg logit contributions
 slippery-0.753
 snowball-0.742
liv-0.740
 mess-0.740
onna-0.727
Token.
Token ID13
Feature activation+4.85e-02

Pos logit contributions
aroo+1.478
itte+1.306
ter+1.300
crow+1.261
ovable+1.201
Neg logit contributions
 mess-0.892
 wra-0.679
 slippery-0.665
 fault-0.662
 pic-0.648
Token They
Token ID1119
Feature activation+1.18e-03

Pos logit contributions
aroo+0.100
ter+0.089
crow+0.079
apped+0.078
itte+0.077
Neg logit contributions
 mess-0.093
aj-0.083
 wra-0.078
onna-0.078
 pic-0.076
Token have
Token ID423
Feature activation+1.43e-01

Pos logit contributions
aroo+0.002
ter+0.002
apped+0.002
crow+0.002
itte+0.002
Neg logit contributions
 mess-0.002
onna-0.002
 wra-0.002
 slippery-0.002
aj-0.002
Token a
Token ID257
Feature activation+3.10e-02

Pos logit contributions
aroo+0.272
ter+0.263
 sailors+0.243
apped+0.229
 citizens+0.229
Neg logit contributions
 mess-0.237
aj-0.233
aman-0.225
icol-0.214
liv-0.208
Token lot
Token ID1256
Feature activation-1.20e-01

Pos logit contributions
aroo+0.064
ter+0.055
 hero+0.053
 pilot+0.053
 member+0.053
Neg logit contributions
aj-0.052
 mess-0.051
liv-0.048
array-0.048
onna-0.046
Token.
Token ID13
Feature activation+4.31e-02

Pos logit contributions
 mess+0.245
aj+0.216
 wra+0.209
onna+0.204
 fault+0.198
Neg logit contributions
aroo-0.304
ter-0.268
crow-0.247
itte-0.244
ovable-0.234
Token "
Token ID366
Feature activation+4.47e-02

Pos logit contributions
aroo+0.098
ter+0.089
itte+0.080
crow+0.080
 member+0.080
Neg logit contributions
 mess-0.068
onna-0.058
aj-0.058
 wra-0.057
 fault-0.053
TokenThe
Token ID464
Feature activation+1.02e-01

Pos logit contributions
aroo+0.085
ter+0.073
apped+0.072
ovable+0.071
itte+0.070
Neg logit contributions
 mess-0.085
 wra-0.075
 slippery-0.072
onna-0.067
aj-0.067
Token k
Token ID479
Feature activation+1.13e-01

Pos logit contributions
aroo+0.200
apped+0.176
itte+0.175
ovable+0.174
ter+0.172
Neg logit contributions
 mess-0.210
 slippery-0.166
 wra-0.165
 steep-0.163
 mistake-0.163
Tokenang
Token ID648
Feature activation+3.05e-01

Pos logit contributions
aroo+0.147
itte+0.146
ovable+0.143
 citizens+0.143
 Queen+0.142
Neg logit contributions
 mess-0.217
aj-0.208
 wra-0.205
 slippery-0.200
liv-0.199
Tokenaroo
Token ID38049
Feature activation+5.92e-01

Pos logit contributions
onel+0.148
ovable+0.137
apped+0.133
 citizens+0.129
ive+0.117
Neg logit contributions
 mess-0.804
 slippery-0.782
 snowball-0.756
 steep-0.753
liv-0.750
Token is
Token ID318
Feature activation-1.32e-01

Pos logit contributions
aroo+1.598
ter+1.477
apped+1.363
itte+1.350
crow+1.346
Neg logit contributions
 mess-0.772
aj-0.597
 wra-0.558
onna-0.543
 slippery-0.528
Token not
Token ID407
Feature activation-2.52e-01

Pos logit contributions
 mess+0.217
aj+0.195
 wra+0.181
onna+0.177
aman+0.173
Neg logit contributions
aroo-0.302
ter-0.275
crow-0.246
apped-0.241
itte-0.238
Token from
Token ID422
Feature activation+1.80e-01

Pos logit contributions
 mess+0.430
aj+0.360
 wra+0.341
onna+0.338
 fault+0.330
Neg logit contributions
aroo-0.578
ter-0.516
apped-0.478
crow-0.474
itte-0.468
Token the
Token ID262
Feature activation+6.80e-02

Pos logit contributions
aroo+0.398
ter+0.351
apped+0.341
itte+0.338
crow+0.330
Neg logit contributions
 mess-0.317
aj-0.255
 pic-0.250
 slippery-0.248
 wra-0.246
Token park
Token ID3952
Feature activation-3.60e-02

Pos logit contributions
aroo+0.137
apped+0.119
ter+0.117
itte+0.117
ovable+0.113
Neg logit contributions
 mess-0.137
 steep-0.110
 slippery-0.110
 wra-0.110
 fault-0.108
Token
Token ID198
Feature activation+2.94e-02

Pos logit contributions
aroo+0.113
ter+0.102
crow+0.090
apped+0.088
itte+0.087
Neg logit contributions
 mess-0.107
aj-0.097
onna-0.091
 wra-0.091
 pic-0.088
Token
Token ID198
Feature activation+2.98e-02

Pos logit contributions
aroo+0.051
ter+0.042
crow+0.039
apped+0.038
itte+0.037
Neg logit contributions
 mess-0.066
aj-0.059
 wra-0.058
 fault-0.057
icol-0.054
TokenThe
Token ID464
Feature activation+1.11e-01

Pos logit contributions
aroo+0.062
ter+0.058
apped+0.053
ovable+0.050
crow+0.049
Neg logit contributions
 mess-0.054
aj-0.047
 wra-0.045
onna-0.044
 slippery-0.044
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
aroo+0.229
ovable+0.213
apped+0.211
ter+0.207
onel+0.207
Neg logit contributions
 mess-0.213
 slippery-0.175
 mistake-0.158
 steep-0.157
 wra-0.157
Tokenang
Token ID648
Feature activation+3.03e-01

Pos logit contributions
aroo+0.150
ter+0.146
ovable+0.146
 Queen+0.144
 citizens+0.144
Neg logit contributions
 mess-0.229
aj-0.217
 wra-0.211
liv-0.209
 slippery-0.209
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
ovable+0.123
onel+0.120
 citizens+0.116
 sailors+0.106
 married+0.095
Neg logit contributions
 mess-0.822
 slippery-0.817
 wob-0.808
liv-0.799
 snowball-0.792
Token felt
Token ID2936
Feature activation+2.20e-02

Pos logit contributions
aroo+1.355
ter+1.226
crow+1.109
apped+1.097
ovable+1.081
Neg logit contributions
 mess-0.992
aj-0.850
 wra-0.800
onna-0.780
 pic-0.760
Token embarrassed
Token ID21100
Feature activation+7.27e-02

Pos logit contributions
aroo+0.052
ter+0.046
apped+0.042
 member+0.041
ive+0.041
Neg logit contributions
 mess-0.037
 slippery-0.030
 fun-0.029
 wra-0.028
onna-0.027
Token,
Token ID11
Feature activation+3.27e-02

Pos logit contributions
aroo+0.147
ter+0.134
crow+0.118
apped+0.117
itte+0.113
Neg logit contributions
 mess-0.140
aj-0.128
 wra-0.119
onna-0.118
 pic-0.116
Token and
Token ID290
Feature activation-5.58e-02

Pos logit contributions
aroo+0.071
ter+0.067
crow+0.059
apped+0.055
 pilot+0.055
Neg logit contributions
aj-0.055
 mess-0.053
onna-0.048
liv-0.047
 fault-0.047
Token quickly
Token ID2952
Feature activation-9.09e-02

Pos logit contributions
 mess+0.093
 slippery+0.074
 wra+0.074
aj+0.072
 fuss+0.069
Neg logit contributions
aroo-0.127
ter-0.112
apped-0.105
itte-0.104
crow-0.104
Token
Token ID198
Feature activation+2.90e-02

Pos logit contributions
aroo+0.131
ter+0.119
crow+0.105
apped+0.104
itte+0.103
Neg logit contributions
 mess-0.097
aj-0.087
onna-0.081
 wra-0.080
 pic-0.079
Token
Token ID198
Feature activation+2.97e-02

Pos logit contributions
aroo+0.051
ter+0.041
crow+0.038
apped+0.037
 member+0.037
Neg logit contributions
 mess-0.066
aj-0.060
 wra-0.059
 fault-0.056
icol-0.056
TokenThe
Token ID464
Feature activation+1.13e-01

Pos logit contributions
aroo+0.074
ter+0.068
crow+0.062
apped+0.061
itte+0.060
Neg logit contributions
 mess-0.044
aj-0.038
 wra-0.035
onna-0.034
 pic-0.033
Token k
Token ID479
Feature activation+1.16e-01

Pos logit contributions
aroo+0.255
ovable+0.246
apped+0.245
onel+0.238
itte+0.233
Neg logit contributions
 mess-0.189
 slippery-0.153
 fun-0.145
 scar-0.140
 wra-0.139
Tokenang
Token ID648
Feature activation+3.03e-01

Pos logit contributions
aroo+0.155
ovable+0.154
ter+0.151
 messenger+0.150
 Queen+0.149
Neg logit contributions
 mess-0.218
 wra-0.203
aj-0.202
 slippery-0.200
liv-0.200
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
ovable+0.182
onel+0.171
 alien+0.150
 sailors+0.147
ative+0.146
Neg logit contributions
 slippery-0.755
 mess-0.749
 wob-0.735
liv-0.731
 snowball-0.719
Token was
Token ID373
Feature activation-9.67e-03

Pos logit contributions
aroo+1.395
ter+1.250
crow+1.142
ovable+1.106
apped+1.095
Neg logit contributions
 mess-0.992
aj-0.823
 wra-0.765
onna-0.755
 slippery-0.751
Token so
Token ID523
Feature activation-5.78e-02

Pos logit contributions
 mess+0.017
aj+0.015
 wra+0.014
onna+0.014
 slippery+0.013
Neg logit contributions
aroo-0.022
ter-0.019
crow-0.018
apped-0.017
itte-0.017
Token happy
Token ID3772
Feature activation+1.04e-01

Pos logit contributions
 mess+0.111
aj+0.096
onna+0.093
 slippery+0.091
 wra+0.090
Neg logit contributions
aroo-0.122
ter-0.110
crow-0.105
ive-0.099
 member-0.093
Token with
Token ID351
Feature activation+1.27e-01

Pos logit contributions
aroo+0.215
ter+0.195
apped+0.169
crow+0.168
 member+0.164
Neg logit contributions
 mess-0.199
aj-0.182
 wra-0.170
onna-0.167
liv-0.163
Token the
Token ID262
Feature activation+7.99e-02

Pos logit contributions
aroo+0.282
ter+0.255
crow+0.229
apped+0.226
itte+0.222
Neg logit contributions
 mess-0.223
aj-0.197
 wra-0.184
onna-0.182
 pic-0.177
Tokenaroo
Token ID38049
Feature activation+5.90e-01

Pos logit contributions
 citizens+0.109
 pilot+0.108
itte+0.107
 crew+0.106
 sailors+0.104
Neg logit contributions
 mess-0.815
gh-0.800
 slippery-0.795
liv-0.793
 wra-0.776
Token!"
Token ID2474
Feature activation-1.16e-01

Pos logit contributions
aroo+1.581
ter+1.441
apped+1.372
itte+1.369
crow+1.359
Neg logit contributions
 mess-0.747
aj-0.602
 wra-0.542
onna-0.535
 pic-0.524
Token The
Token ID383
Feature activation+2.22e-01

Pos logit contributions
 mess+0.178
aj+0.166
icol+0.149
aman+0.148
 pic+0.147
Neg logit contributions
aroo-0.274
ter-0.244
crow-0.225
itte-0.223
ovable-0.221
Token k
Token ID479
Feature activation+1.15e-01

Pos logit contributions
aroo+0.474
ter+0.431
ovable+0.430
apped+0.430
ive+0.428
Neg logit contributions
 mess-0.382
 slippery-0.318
 fun-0.294
 wra-0.290
 steep-0.284
Tokenang
Token ID648
Feature activation+3.04e-01

Pos logit contributions
 citizens+0.157
 crew+0.156
 sailors+0.152
ter+0.149
 messenger+0.149
Neg logit contributions
 mess-0.215
liv-0.209
 wra-0.206
 slippery-0.205
aj-0.205
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
 sailors+0.171
 crew+0.168
 citizens+0.166
onel+0.153
ovable+0.140
Neg logit contributions
liv-0.780
 slippery-0.767
 wob-0.751
 mess-0.750
 snowball-0.742
Token sque
Token ID35776
Feature activation-3.02e-02

Pos logit contributions
aroo+1.357
ter+1.237
crow+1.121
apped+1.095
itte+1.083
Neg logit contributions
 mess-0.975
aj-0.855
 wra-0.793
onna-0.788
 pic-0.763
Tokenaked
Token ID4335
Feature activation+1.40e-01

Pos logit contributions
 mess+0.064
aj+0.059
 wra+0.055
onna+0.054
liv+0.053
Neg logit contributions
aroo-0.055
ter-0.048
 pilot-0.041
crow-0.040
 member-0.039
Token in
Token ID287
Feature activation+2.89e-01

Pos logit contributions
aroo+0.298
ter+0.267
crow+0.239
apped+0.235
itte+0.232
Neg logit contributions
 mess-0.256
aj-0.226
 wra-0.214
onna-0.210
 pic-0.205
Token response
Token ID2882
Feature activation+8.80e-02

Pos logit contributions
aroo+0.531
ter+0.470
crow+0.410
apped+0.404
 member+0.392
Neg logit contributions
 mess-0.616
aj-0.566
 wra-0.532
onna-0.528
 fault-0.516
Token,
Token ID11
Feature activation+3.47e-02

Pos logit contributions
aroo+0.209
ter+0.186
crow+0.172
itte+0.171
apped+0.167
Neg logit contributions
 mess-0.138
aj-0.118
 wra-0.112
onna-0.112
 pic-0.106
Token there
Token ID612
Feature activation+1.13e-01

Pos logit contributions
aroo+0.055
ter+0.049
crow+0.046
 member+0.044
 citizens+0.044
Neg logit contributions
 mess-0.050
aj-0.043
 slippery-0.041
 wra-0.041
onna-0.040
Token was
Token ID373
Feature activation-1.34e-02

Pos logit contributions
aroo+0.235
ter+0.210
crow+0.188
apped+0.187
itte+0.180
Neg logit contributions
 mess-0.214
aj-0.198
 wra-0.181
onna-0.179
 fault-0.175
Token a
Token ID257
Feature activation+3.62e-02

Pos logit contributions
 mess+0.029
aj+0.026
onna+0.025
 wra+0.024
 fuss+0.024
Neg logit contributions
aroo-0.024
ter-0.022
apped-0.019
crow-0.018
itte-0.018
Token k
Token ID479
Feature activation+1.14e-01

Pos logit contributions
aroo+0.075
ter+0.067
crow+0.064
apped+0.063
ovable+0.061
Neg logit contributions
 mess-0.072
aj-0.058
 wra-0.056
 slippery-0.056
 pic-0.055
Tokenang
Token ID648
Feature activation+3.05e-01

Pos logit contributions
 citizens+0.153
 sailors+0.151
rest+0.150
onel+0.150
ter+0.149
Neg logit contributions
 mess-0.213
 wra-0.194
 slippery-0.191
 snowball-0.190
 steep-0.188
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
 citizens+0.184
 sailors+0.182
onel+0.170
apped+0.160
 married+0.146
Neg logit contributions
 slippery-0.747
 snowball-0.729
liv-0.725
gh-0.717
 mess-0.714
Token.
Token ID13
Feature activation+5.60e-02

Pos logit contributions
aroo+0.926
onel+0.898
folk+0.895
ter+0.879
itte+0.859
Neg logit contributions
 mess-1.136
onna-0.980
ine-0.978
 slippery-0.971
 snowball-0.965
Token He
Token ID679
Feature activation+4.15e-02

Pos logit contributions
aroo+0.112
ter+0.104
onel+0.093
 member+0.093
itte+0.092
Neg logit contributions
 mess-0.097
onna-0.091
aj-0.088
 steep-0.087
 pic-0.087
Token loved
Token ID6151
Feature activation+1.73e-01

Pos logit contributions
aroo+0.077
ter+0.071
crow+0.063
apped+0.062
 member+0.061
Neg logit contributions
 mess-0.082
aj-0.072
 wra-0.072
 pic-0.070
 slippery-0.070
Token to
Token ID284
Feature activation-5.22e-02

Pos logit contributions
aroo+0.384
ter+0.346
crow+0.309
apped+0.302
 member+0.296
Neg logit contributions
 mess-0.300
aj-0.272
 wra-0.248
onna-0.244
liv-0.243
Token paint
Token ID7521
Feature activation+7.61e-02

Pos logit contributions
 mess+0.095
 pic+0.082
onna+0.082
 slippery+0.079
aj+0.077
Neg logit contributions
aroo-0.109
ter-0.102
apped-0.096
itte-0.093
ovable-0.090
Tokenaroo
Token ID38049
Feature activation+5.90e-01

Pos logit contributions
 citizens+0.168
onel+0.133
ovable+0.123
 crew+0.119
 sailors+0.116
Neg logit contributions
 mess-0.814
 snowball-0.783
 slippery-0.776
liv-0.774
onna-0.774
Token.
Token ID13
Feature activation+5.47e-02

Pos logit contributions
aroo+1.285
ter+1.144
itte+1.112
ovable+1.104
 citizens+1.085
Neg logit contributions
 mess-1.000
 wra-0.785
 slippery-0.756
aj-0.754
onna-0.733
Token The
Token ID383
Feature activation+2.17e-01

Pos logit contributions
aroo+0.106
ter+0.096
ovable+0.082
itte+0.081
onel+0.080
Neg logit contributions
 mess-0.108
aj-0.099
onna-0.097
 pic-0.094
 wra-0.094
Token k
Token ID479
Feature activation+1.10e-01

Pos logit contributions
aroo+0.438
onel+0.423
ovable+0.422
itte+0.410
ter+0.410
Neg logit contributions
 mess-0.374
 slippery-0.318
 steep-0.293
 fun-0.293
 wra-0.292
Tokenang
Token ID648
Feature activation+3.05e-01

Pos logit contributions
aroo+0.130
ter+0.123
 citizens+0.115
ovable+0.115
itte+0.114
Neg logit contributions
 mess-0.251
 wra-0.231
aj-0.230
liv-0.227
 slippery-0.227
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
onel+0.110
 citizens+0.098
ter+0.092
 crew+0.088
ovable+0.086
Neg logit contributions
 mess-0.877
liv-0.848
 slippery-0.843
 snowball-0.829
 wob-0.821
Token is
Token ID318
Feature activation-1.28e-01

Pos logit contributions
aroo+1.418
ter+1.269
crow+1.180
apped+1.153
itte+1.139
Neg logit contributions
 mess-0.884
aj-0.800
 wra-0.729
onna-0.718
 fault-0.710
Token hopping
Token ID47153
Feature activation+7.90e-02

Pos logit contributions
 mess+0.229
aj+0.205
 wra+0.192
onna+0.188
 pic+0.184
Neg logit contributions
aroo-0.281
ter-0.254
crow-0.225
apped-0.224
itte-0.218
Token and
Token ID290
Feature activation-6.02e-02

Pos logit contributions
aroo+0.128
ter+0.110
itte+0.093
apped+0.093
crow+0.092
Neg logit contributions
 mess-0.181
aj-0.159
 wra-0.158
 slippery-0.156
 steep-0.153
Token eating
Token ID6600
Feature activation-8.21e-02

Pos logit contributions
 mess+0.101
aj+0.090
 wra+0.083
onna+0.082
 fault+0.080
Neg logit contributions
aroo-0.138
ter-0.125
crow-0.112
apped-0.111
itte-0.108
Token grass
Token ID8701
Feature activation-1.24e-01

Pos logit contributions
 mess+0.162
aj+0.152
onna+0.138
liv+0.136
 fault+0.135
Neg logit contributions
aroo-0.163
ter-0.147
crow-0.128
apped-0.125
 member-0.120
Token careful
Token ID8161
Feature activation+2.75e-02

Pos logit contributions
aroo+0.003
ter+0.003
crow+0.002
apped+0.002
itte+0.002
Neg logit contributions
 mess-0.002
aj-0.002
 slippery-0.002
 wra-0.002
 fault-0.002
Token.
Token ID13
Feature activation+3.91e-02

Pos logit contributions
aroo+0.054
ter+0.048
crow+0.043
apped+0.042
itte+0.041
Neg logit contributions
 mess-0.055
aj-0.049
 wra-0.047
onna-0.046
 pic-0.045
Token The
Token ID383
Feature activation+2.08e-01

Pos logit contributions
aroo+0.077
ter+0.067
 member+0.062
apped+0.058
itte+0.057
Neg logit contributions
 mess-0.072
aj-0.071
onna-0.065
 fault-0.064
 pic-0.063
Token k
Token ID479
Feature activation+1.07e-01

Pos logit contributions
aroo+0.420
ter+0.377
apped+0.369
ovable+0.358
itte+0.355
Neg logit contributions
 mess-0.420
 slippery-0.338
 fault-0.326
 steep-0.325
 mistake-0.325
Tokenang
Token ID648
Feature activation+3.04e-01

Pos logit contributions
aroo+0.143
ter+0.141
 Queen+0.137
itte+0.134
ovable+0.134
Neg logit contributions
 mess-0.223
aj-0.207
 slippery-0.197
liv-0.196
 wra-0.195
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
onel+0.125
ovable+0.114
 citizens+0.110
 crew+0.101
 sailors+0.100
Neg logit contributions
 mess-0.845
 slippery-0.800
liv-0.781
 mistake-0.767
 wob-0.761
Token is
Token ID318
Feature activation-1.37e-01

Pos logit contributions
aroo+1.618
ter+1.502
apped+1.380
itte+1.360
crow+1.347
Neg logit contributions
 mess-0.755
aj-0.586
 wra-0.541
onna-0.515
 pays-0.515
Token not
Token ID407
Feature activation-2.56e-01

Pos logit contributions
 mess+0.194
aj+0.166
 wra+0.153
onna+0.150
 pic+0.144
Neg logit contributions
aroo-0.349
ter-0.319
crow-0.291
apped-0.288
itte-0.284
Token a
Token ID257
Feature activation+2.32e-02

Pos logit contributions
 mess+0.455
aj+0.368
 slippery+0.358
onna+0.356
 fault+0.355
Neg logit contributions
aroo-0.577
ter-0.507
apped-0.482
itte-0.474
crow-0.474
Token toy
Token ID13373
Feature activation+3.56e-02

Pos logit contributions
aroo+0.047
ter+0.042
apped+0.040
itte+0.038
crow+0.037
Neg logit contributions
 mess-0.047
 fault-0.038
onna-0.037
 wra-0.037
aj-0.037
Token.
Token ID13
Feature activation+4.18e-02

Pos logit contributions
aroo+0.086
itte+0.083
apped+0.081
year+0.078
appers+0.075
Neg logit contributions
 mess-0.046
 wra-0.037
 fault-0.035
 mistake-0.031
onna-0.030
Token"
Token ID1
Feature activation+9.24e-02

Pos logit contributions
aroo+0.043
ter+0.039
apped+0.034
crow+0.033
ovable+0.033
Neg logit contributions
 mess-0.044
aj-0.040
 wra-0.038
onna-0.037
 pic-0.036
TokenHello
Token ID15496
Feature activation+2.19e-01

Pos logit contributions
aroo+0.180
ter+0.161
apped+0.148
ovable+0.145
itte+0.144
Neg logit contributions
 mess-0.180
 wra-0.157
aj-0.152
 slippery-0.150
onna-0.149
Token,
Token ID11
Feature activation+1.67e-02

Pos logit contributions
aroo+0.495
ter+0.442
apped+0.412
crow+0.411
itte+0.409
Neg logit contributions
 mess-0.370
aj-0.324
 wra-0.299
 pic-0.291
onna-0.286
Token k
Token ID479
Feature activation+1.09e-01

Pos logit contributions
aroo+0.030
apped+0.030
 citizens+0.027
ovable+0.027
ive+0.026
Neg logit contributions
 mess-0.032
 slippery-0.027
 wra-0.027
 scar-0.026
aman-0.025
Tokenang
Token ID648
Feature activation+3.05e-01

Pos logit contributions
aroo+0.132
itte+0.127
 citizens+0.127
ter+0.126
ovable+0.126
Neg logit contributions
 mess-0.223
 wra-0.213
aj-0.209
liv-0.208
 slippery-0.206
Tokenaroo
Token ID38049
Feature activation+5.91e-01

Pos logit contributions
apped+0.132
onel+0.123
 citizens+0.122
ovable+0.109
 protector+0.098
Neg logit contributions
 mess-0.852
 slippery-0.819
liv-0.811
gh-0.794
 wra-0.792
Token,"
Token ID553
Feature activation+5.51e-02

Pos logit contributions
aroo+1.433
ter+1.306
apped+1.299
itte+1.272
ovable+1.229
Neg logit contributions
 mess-0.879
onna-0.664
aj-0.661
 slippery-0.629
 wra-0.626
Token Anna
Token ID11735
Feature activation-1.16e-01

Pos logit contributions
aroo+0.149
ter+0.141
apped+0.129
onel+0.128
ovable+0.128
Neg logit contributions
 mess-0.068
 wra-0.052
 pic-0.050
aj-0.049
onna-0.049
Token says
Token ID1139
Feature activation-1.38e-01

Pos logit contributions
 mess+0.187
aj+0.166
 wra+0.154
liv+0.154
onna+0.151
Neg logit contributions
aroo-0.265
ter-0.241
crow-0.225
apped-0.218
 member-0.215
Token.
Token ID13
Feature activation+5.04e-02

Pos logit contributions
 mess+0.216
aj+0.188
onna+0.184
 wra+0.181
 fault+0.172
Neg logit contributions
aroo-0.320
ter-0.282
crow-0.266
itte-0.260
ovable-0.247
Token "
Token ID366
Feature activation+4.69e-02

Pos logit contributions
aroo+0.116
ter+0.108
 member+0.096
crow+0.096
ovable+0.096
Neg logit contributions
 mess-0.073
onna-0.066
 wra-0.063
aj-0.062
 pic-0.059
Token get
Token ID651
Feature activation-3.64e-02

Pos logit contributions
aroo+0.156
ter+0.141
crow+0.127
apped+0.126
itte+0.123
Neg logit contributions
 mess-0.135
aj-0.114
onna-0.111
 wra-0.107
 fault-0.105
Token hurt
Token ID5938
Feature activation-8.35e-02

Pos logit contributions
 mess+0.067
aj+0.055
 wra+0.054
 slippery+0.053
onna+0.052
Neg logit contributions
aroo-0.082
ter-0.071
itte-0.066
crow-0.066
apped-0.065
Token or
Token ID393
Feature activation-1.44e-01

Pos logit contributions
 mess+0.155
aj+0.131
 wra+0.127
onna+0.125
 fault+0.122
Neg logit contributions
aroo-0.180
ter-0.157
apped-0.146
itte-0.146
crow-0.142
Token make
Token ID787
Feature activation-2.67e-02

Pos logit contributions
 mess+0.259
aj+0.211
 wra+0.205
onna+0.202
 slippery+0.200
Neg logit contributions
aroo-0.324
ter-0.287
apped-0.269
itte-0.266
crow-0.266
Token a
Token ID257
Feature activation-6.46e-02

Pos logit contributions
 mess+0.048
aj+0.043
 wra+0.040
onna+0.040
 pic+0.039
Neg logit contributions
aroo-0.057
ter-0.051
crow-0.046
apped-0.045
itte-0.044
Token mess
Token ID2085
Feature activation-6.09e-01

Pos logit contributions
 mess+0.072
aj+0.070
liv+0.061
 wra+0.061
 pic+0.059
Neg logit contributions
aroo-0.172
ter-0.159
crow-0.146
 pilot-0.144
 member-0.142
Token.
Token ID13
Feature activation-4.04e-02

Pos logit contributions
 mess+1.187
aj+0.963
 wra+0.963
onna+0.915
 slippery+0.890
Neg logit contributions
aroo-1.294
ter-1.097
crow-1.026
apped-1.021
itte-1.012
Token And
Token ID843
Feature activation+4.92e-02

Pos logit contributions
 mess+0.071
aj+0.065
onna+0.061
 wra+0.059
 pic+0.058
Neg logit contributions
aroo-0.087
ter-0.080
crow-0.069
apped-0.069
 member-0.068
Token you
Token ID345
Feature activation-1.92e-02

Pos logit contributions
aroo+0.106
ter+0.093
crow+0.086
apped+0.085
itte+0.085
Neg logit contributions
 mess-0.089
aj-0.074
onna-0.073
 wra-0.072
 pic-0.071
Token're
Token ID821
Feature activation+7.82e-03

Pos logit contributions
 mess+0.037
 wra+0.029
onna+0.029
aj+0.029
 slippery+0.029
Neg logit contributions
aroo-0.041
ter-0.036
apped-0.033
crow-0.032
itte-0.031
Token making
Token ID1642
Feature activation-1.64e-01

Pos logit contributions
aroo+0.019
ter+0.016
crow+0.016
ive+0.016
apped+0.016
Neg logit contributions
 mess-0.013
 slippery-0.010
aj-0.010
 fault-0.009
 fun-0.009
Token they
Token ID484
Feature activation-6.76e-02

Pos logit contributions
aroo+0.101
ter+0.092
apped+0.084
crow+0.084
itte+0.081
Neg logit contributions
 mess-0.056
aj-0.046
 wra-0.043
 pays-0.042
onna-0.040
Token also
Token ID635
Feature activation-1.18e-01

Pos logit contributions
 mess+0.108
 slippery+0.081
onna+0.081
aj+0.079
 pic+0.079
Neg logit contributions
aroo-0.162
ter-0.146
crow-0.130
eros-0.126
ative-0.126
Token made
Token ID925
Feature activation-1.53e-01

Pos logit contributions
 mess+0.219
aj+0.182
onna+0.170
 wra+0.167
 fault+0.167
Neg logit contributions
aroo-0.267
ter-0.238
crow-0.210
itte-0.208
apped-0.204
Token a
Token ID257
Feature activation-7.05e-02

Pos logit contributions
 mess+0.282
aj+0.248
 wra+0.233
onna+0.229
 pic+0.225
Neg logit contributions
aroo-0.329
ter-0.294
crow-0.263
apped-0.259
itte-0.256
Token big
Token ID1263
Feature activation-1.74e-01

Pos logit contributions
 mess+0.103
aj+0.103
liv+0.095
 wra+0.092
onna+0.090
Neg logit contributions
aroo-0.159
ter-0.143
 pilot-0.131
 member-0.130
crow-0.127
Token mess
Token ID2085
Feature activation-6.07e-01

Pos logit contributions
 mess+0.286
aj+0.257
 wra+0.237
onna+0.235
 pic+0.227
Neg logit contributions
aroo-0.400
ter-0.362
crow-0.326
apped-0.322
 member-0.315
Token.
Token ID13
Feature activation-4.95e-02

Pos logit contributions
 mess+1.142
 wra+0.932
aj+0.913
onna+0.867
 slippery+0.846
Neg logit contributions
aroo-1.321
ter-1.096
crow-1.069
itte-1.043
apped-1.029
Token They
Token ID1119
Feature activation-8.45e-02

Pos logit contributions
 mess+0.086
aj+0.079
onna+0.074
 wra+0.072
 pic+0.071
Neg logit contributions
aroo-0.108
ter-0.098
crow-0.087
apped-0.085
 member-0.084
Token threw
Token ID9617
Feature activation-1.26e-01

Pos logit contributions
 mess+0.148
aj+0.121
 wra+0.120
onna+0.114
 slippery+0.113
Neg logit contributions
aroo-0.197
ter-0.176
crow-0.160
apped-0.156
ive-0.152
Token their
Token ID511
Feature activation-1.82e-01

Pos logit contributions
 mess+0.252
aj+0.247
onna+0.229
liv+0.224
aman+0.220
Neg logit contributions
aroo-0.233
ter-0.211
crow-0.188
apped-0.181
 member-0.173
Token toys
Token ID14958
Feature activation-8.10e-02

Pos logit contributions
 mess+0.306
aj+0.264
 wra+0.251
onna+0.239
 fault+0.235
Neg logit contributions
aroo-0.421
ter-0.383
crow-0.348
apped-0.344
itte-0.339
Token too
Token ID1165
Feature activation-1.43e-01

Pos logit contributions
 mess+0.226
aj+0.203
 wra+0.192
onna+0.190
 pic+0.184
Neg logit contributions
aroo-0.241
ter-0.219
crow-0.192
apped-0.190
itte-0.184
Token.
Token ID13
Feature activation-4.07e-02

Pos logit contributions
 mess+0.316
aj+0.297
 wra+0.286
liv+0.280
icol+0.278
Neg logit contributions
aroo-0.236
ter-0.208
apped-0.176
crow-0.176
 member-0.171
Token After
Token ID2293
Feature activation-1.44e-01

Pos logit contributions
 mess+0.064
aj+0.062
onna+0.057
 pic+0.055
 wra+0.054
Neg logit contributions
aroo-0.092
ter-0.084
 member-0.073
crow-0.073
itte-0.072
Token all
Token ID477
Feature activation+4.64e-02

Pos logit contributions
 mess+0.264
aj+0.227
 wra+0.220
onna+0.214
 pic+0.212
Neg logit contributions
aroo-0.310
ter-0.278
crow-0.252
apped-0.246
itte-0.246
Token the
Token ID262
Feature activation-5.70e-03

Pos logit contributions
aroo+0.101
ter+0.091
apped+0.083
itte+0.081
ovable+0.081
Neg logit contributions
 mess-0.087
aj-0.068
 wra-0.067
 slippery-0.067
 pic-0.066
Token mess
Token ID2085
Feature activation-6.07e-01

Pos logit contributions
 mess+0.010
aj+0.008
 wra+0.008
 pic+0.008
onna+0.008
Neg logit contributions
aroo-0.013
ter-0.012
apped-0.011
crow-0.011
itte-0.010
Token was
Token ID373
Feature activation-9.62e-02

Pos logit contributions
 mess+1.046
aj+0.893
 wra+0.861
onna+0.825
 fault+0.792
Neg logit contributions
aroo-1.373
ter-1.210
crow-1.120
itte-1.090
apped-1.088
Token made
Token ID925
Feature activation-1.35e-01

Pos logit contributions
 mess+0.156
aj+0.145
 wra+0.138
icol+0.137
 pic+0.134
Neg logit contributions
aroo-0.206
ter-0.191
apped-0.169
 member-0.166
crow-0.166
Token,
Token ID11
Feature activation-4.53e-02

Pos logit contributions
 mess+0.264
aj+0.238
 wra+0.224
onna+0.221
 pic+0.216
Neg logit contributions
aroo-0.271
ter-0.242
crow-0.214
apped-0.212
itte-0.207
Token Sophia
Token ID39953
Feature activation-2.12e-01

Pos logit contributions
 mess+0.081
aj+0.074
onna+0.068
 wra+0.068
 fault+0.066
Neg logit contributions
aroo-0.096
ter-0.088
crow-0.077
apped-0.076
itte-0.074
Token used
Token ID973
Feature activation-4.13e-02

Pos logit contributions
 mess+0.361
aj+0.308
 wra+0.285
onna+0.284
 fault+0.269
Neg logit contributions
aroo-0.497
ter-0.448
crow-0.405
apped-0.397
itte-0.390
Token dad
Token ID9955
Feature activation+2.19e-02

Pos logit contributions
 mess+0.291
aj+0.251
 wra+0.239
onna+0.233
 pic+0.230
Neg logit contributions
aroo-0.343
ter-0.308
crow-0.276
apped-0.274
itte-0.272
Token came
Token ID1625
Feature activation-8.78e-03

Pos logit contributions
aroo+0.053
ter+0.046
crow+0.043
itte+0.042
ive+0.041
Neg logit contributions
 mess-0.036
aj-0.030
 pays-0.028
 wra-0.028
onna-0.027
Token and
Token ID290
Feature activation-1.68e-01

Pos logit contributions
 mess+0.016
aj+0.014
 wra+0.013
onna+0.013
 slippery+0.013
Neg logit contributions
aroo-0.019
ter-0.016
crow-0.015
itte-0.015
apped-0.015
Token saw
Token ID2497
Feature activation-5.50e-02

Pos logit contributions
 mess+0.289
aj+0.247
 wra+0.241
onna+0.226
 slippery+0.220
Neg logit contributions
aroo-0.381
ter-0.336
crow-0.317
ovable-0.312
apped-0.308
Token the
Token ID262
Feature activation-2.89e-02

Pos logit contributions
 mess+0.101
aj+0.085
 wra+0.084
onna+0.079
 pic+0.079
Neg logit contributions
aroo-0.119
ter-0.106
apped-0.096
crow-0.095
itte-0.094
Token mess
Token ID2085
Feature activation-6.06e-01

Pos logit contributions
 mess+0.057
aj+0.047
 wra+0.046
 fault+0.044
 slippery+0.043
Neg logit contributions
aroo-0.062
ter-0.054
apped-0.050
crow-0.050
itte-0.048
Token.
Token ID13
Feature activation-4.52e-02

Pos logit contributions
 mess+1.152
 wra+0.962
aj+0.937
onna+0.881
 slippery+0.878
Neg logit contributions
aroo-1.300
ter-1.108
crow-1.039
itte-1.017
apped-1.005
Token They
Token ID1119
Feature activation-9.01e-02

Pos logit contributions
 mess+0.085
aj+0.076
 wra+0.072
onna+0.072
 pic+0.069
Neg logit contributions
aroo-0.094
ter-0.084
crow-0.074
apped-0.073
itte-0.072
Token were
Token ID547
Feature activation+3.66e-02

Pos logit contributions
 mess+0.174
 wra+0.144
aj+0.140
 slippery+0.134
onna+0.132
Neg logit contributions
aroo-0.197
ter-0.172
crow-0.159
apped-0.153
ive-0.152
Token not
Token ID407
Feature activation-3.40e-01

Pos logit contributions
aroo+0.084
ter+0.075
crow+0.069
apped+0.067
itte+0.065
Neg logit contributions
 mess-0.065
aj-0.055
 wra-0.052
onna-0.050
 slippery-0.049
Token happy
Token ID3772
Feature activation+1.99e-02

Pos logit contributions
 mess+0.701
aj+0.602
 wra+0.576
onna+0.556
 fault+0.550
Neg logit contributions
aroo-0.698
ter-0.596
crow-0.557
itte-0.528
apped-0.528
Token it
Token ID340
Feature activation-1.59e-01

Pos logit contributions
 mess+0.291
aj+0.252
 wra+0.240
onna+0.235
 pic+0.234
Neg logit contributions
aroo-0.314
ter-0.279
crow-0.249
apped-0.247
itte-0.242
Token looked
Token ID3114
Feature activation-2.15e-01

Pos logit contributions
 mess+0.267
aj+0.227
 wra+0.215
onna+0.212
 pic+0.205
Neg logit contributions
aroo-0.369
ter-0.332
crow-0.301
apped-0.298
itte-0.293
Token like
Token ID588
Feature activation-3.17e-02

Pos logit contributions
 mess+0.384
aj+0.347
 wra+0.326
onna+0.323
liv+0.314
Neg logit contributions
aroo-0.452
ter-0.413
crow-0.366
apped-0.365
itte-0.357
Token a
Token ID257
Feature activation-6.68e-02

Pos logit contributions
 mess+0.056
aj+0.050
 wra+0.047
onna+0.046
 pic+0.045
Neg logit contributions
aroo-0.069
ter-0.062
crow-0.055
apped-0.055
itte-0.054
Token big
Token ID1263
Feature activation-1.69e-01

Pos logit contributions
 mess+0.122
aj+0.110
 wra+0.102
onna+0.101
liv+0.098
Neg logit contributions
aroo-0.141
ter-0.127
crow-0.113
apped-0.111
itte-0.109
Token mess
Token ID2085
Feature activation-6.06e-01

Pos logit contributions
 mess+0.318
aj+0.276
 wra+0.263
onna+0.259
 pic+0.253
Neg logit contributions
aroo-0.354
ter-0.317
crow-0.284
apped-0.282
itte-0.277
Token.
Token ID13
Feature activation-4.63e-02

Pos logit contributions
 mess+1.130
 wra+0.936
aj+0.910
onna+0.876
 slippery+0.843
Neg logit contributions
aroo-1.319
ter-1.111
crow-1.048
itte-1.043
apped-1.042
Token The
Token ID383
Feature activation+1.31e-01

Pos logit contributions
 mess+0.081
aj+0.074
onna+0.069
 wra+0.068
 pic+0.067
Neg logit contributions
aroo-0.101
ter-0.091
crow-0.080
apped-0.080
itte-0.078
Token little
Token ID1310
Feature activation+2.80e-02

Pos logit contributions
aroo+0.258
ter+0.225
apped+0.222
ovable+0.209
crow+0.206
Neg logit contributions
 mess-0.282
 wra-0.231
aj-0.223
 slippery-0.217
 mistake-0.215
Token boy
Token ID2933
Feature activation+1.06e-01

Pos logit contributions
aroo+0.042
ovable+0.035
apped+0.034
ter+0.034
itte+0.034
Neg logit contributions
 mess-0.069
 wra-0.059
aj-0.058
 slippery-0.056
 scar-0.056
Token thought
Token ID1807
Feature activation-1.20e-01

Pos logit contributions
aroo+0.253
ter+0.232
crow+0.209
apped+0.204
itte+0.201
Neg logit contributions
 mess-0.174
aj-0.144
 wra-0.138
onna-0.134
 slippery-0.129
Token ceiling
Token ID13387
Feature activation-1.12e-01

Pos logit contributions
 mess+0.016
aj+0.013
 wra+0.013
 fault+0.012
 slippery+0.012
Neg logit contributions
aroo-0.016
ter-0.014
apped-0.013
ovable-0.012
crow-0.012
Token.
Token ID13
Feature activation-3.43e-02

Pos logit contributions
 mess+0.212
aj+0.185
onna+0.175
 wra+0.174
 fault+0.167
Neg logit contributions
aroo-0.236
ter-0.209
crow-0.189
apped-0.186
itte-0.184
Token
Token ID198
Feature activation-5.61e-02

Pos logit contributions
 mess+0.049
aj+0.049
onna+0.045
 pic+0.044
aman+0.043
Neg logit contributions
aroo-0.079
ter-0.072
ovable-0.063
crow-0.063
 member-0.063
Token
Token ID198
Feature activation-5.85e-02

Pos logit contributions
 mess+0.102
aj+0.093
onna+0.086
 pic+0.084
liv+0.084
Neg logit contributions
aroo-0.118
ter-0.110
apped-0.093
crow-0.093
itte-0.091
TokenThe
Token ID464
Feature activation+2.30e-02

Pos logit contributions
 mess+0.082
aj+0.075
liv+0.071
 pic+0.068
aman+0.067
Neg logit contributions
aroo-0.139
ter-0.136
apped-0.118
crow-0.115
ovable-0.114
Token mess
Token ID2085
Feature activation-6.05e-01

Pos logit contributions
aroo+0.044
apped+0.039
ter+0.039
crow+0.038
ovable+0.036
Neg logit contributions
 mess-0.049
 wra-0.039
 mistake-0.038
 slippery-0.038
aj-0.038
Token was
Token ID373
Feature activation-1.02e-01

Pos logit contributions
 mess+0.996
aj+0.755
 wra+0.751
 pays+0.718
 slippery+0.704
Neg logit contributions
aroo-1.451
ter-1.280
crow-1.200
itte-1.163
ovable-1.154
Token so
Token ID523
Feature activation-1.49e-01

Pos logit contributions
 mess+0.158
aj+0.150
 wra+0.136
 pic+0.134
onna+0.134
Neg logit contributions
aroo-0.233
ter-0.217
apped-0.191
crow-0.189
 member-0.187
Token big
Token ID1263
Feature activation-1.51e-01

Pos logit contributions
 mess+0.215
aj+0.180
 wra+0.166
onna+0.164
 slippery+0.155
Neg logit contributions
aroo-0.386
ter-0.353
crow-0.322
apped-0.316
itte-0.312
Token that
Token ID326
Feature activation+6.49e-02

Pos logit contributions
 mess+0.306
aj+0.291
onna+0.271
liv+0.267
 wra+0.266
Neg logit contributions
aroo-0.276
ter-0.251
crow-0.216
apped-0.208
 member-0.204
Token it
Token ID340
Feature activation-1.49e-01

Pos logit contributions
aroo+0.143
ter+0.128
crow+0.116
apped+0.115
itte+0.114
Neg logit contributions
 mess-0.116
aj-0.096
 wra-0.094
onna-0.092
 pic-0.092
Token she
Token ID673
Feature activation-5.45e-02

Pos logit contributions
 mess+0.110
aj+0.099
 wra+0.094
onna+0.093
 pic+0.091
Neg logit contributions
aroo-0.090
ter-0.079
crow-0.069
apped-0.069
itte-0.067
Token had
Token ID550
Feature activation-9.52e-02

Pos logit contributions
 mess+0.092
aj+0.075
 wra+0.072
onna+0.070
 fault+0.068
Neg logit contributions
aroo-0.130
ter-0.115
crow-0.105
apped-0.102
itte-0.100
Token made
Token ID925
Feature activation-1.46e-01

Pos logit contributions
 mess+0.159
aj+0.149
 wra+0.135
onna+0.133
aman+0.133
Neg logit contributions
aroo-0.203
ter-0.189
crow-0.169
apped-0.168
 member-0.166
Token a
Token ID257
Feature activation-5.75e-02

Pos logit contributions
 mess+0.289
aj+0.270
 wra+0.252
onna+0.251
 pic+0.243
Neg logit contributions
aroo-0.276
ter-0.251
crow-0.220
apped-0.212
 member-0.210
Token big
Token ID1263
Feature activation-1.67e-01

Pos logit contributions
 mess+0.074
aj+0.073
liv+0.064
 wra+0.063
onna+0.062
Neg logit contributions
aroo-0.142
ter-0.129
 member-0.119
crow-0.118
 pilot-0.118
Token mess
Token ID2085
Feature activation-6.05e-01

Pos logit contributions
 mess+0.250
aj+0.226
 wra+0.206
onna+0.203
liv+0.197
Neg logit contributions
aroo-0.403
ter-0.367
crow-0.333
apped-0.327
 member-0.324
Token!
Token ID0
Feature activation-2.51e-01

Pos logit contributions
 mess+1.235
 wra+1.035
aj+1.020
onna+0.987
 slippery+0.957
Neg logit contributions
aroo-1.206
ter-1.008
crow-0.946
apped-0.930
itte-0.919
Token Leaves
Token ID46597
Feature activation-3.53e-02

Pos logit contributions
 mess+0.407
aj+0.373
onna+0.351
 wra+0.336
 pic+0.333
Neg logit contributions
aroo-0.575
ter-0.523
apped-0.454
crow-0.452
 pilot-0.450
Token were
Token ID547
Feature activation+3.99e-02

Pos logit contributions
 mess+0.070
aj+0.064
 wra+0.059
onna+0.059
 pic+0.058
Neg logit contributions
aroo-0.069
ter-0.061
crow-0.054
apped-0.053
itte-0.052
Token all
Token ID477
Feature activation+4.03e-02

Pos logit contributions
aroo+0.093
ter+0.088
crow+0.076
apped+0.075
itte+0.074
Neg logit contributions
 mess-0.058
aj-0.055
liv-0.050
icol-0.049
 pic-0.049
Token over
Token ID625
Feature activation-4.92e-02

Pos logit contributions
aroo+0.094
ter+0.089
crow+0.078
apped+0.078
 member+0.077
Neg logit contributions
 mess-0.058
aj-0.056
 pic-0.051
icol-0.050
 wra-0.049
Tokenodge
Token ID9728
Feature activation-1.07e-01

Pos logit contributions
aroo+0.045
ter+0.037
 pilot+0.037
 member+0.036
 messenger+0.035
Neg logit contributions
 mess-0.039
aj-0.038
onna-0.034
 wra-0.034
liv-0.033
Token began
Token ID2540
Feature activation+2.06e-03

Pos logit contributions
 mess+0.189
 slippery+0.153
 wra+0.149
 flips+0.149
onna+0.148
Neg logit contributions
aroo-0.211
ter-0.193
ive-0.177
itte-0.176
apped-0.174
Token to
Token ID284
Feature activation-1.44e-01

Pos logit contributions
aroo+0.005
ter+0.004
crow+0.004
apped+0.004
itte+0.004
Neg logit contributions
 mess-0.003
aj-0.003
 wra-0.003
onna-0.003
 pic-0.003
Token spread
Token ID4104
Feature activation-4.32e-02

Pos logit contributions
 mess+0.278
aj+0.229
onna+0.223
 wra+0.219
 slippery+0.218
Neg logit contributions
aroo-0.303
ter-0.269
apped-0.246
crow-0.239
itte-0.238
Token his
Token ID465
Feature activation-1.28e-01

Pos logit contributions
 mess+0.079
aj+0.068
 wra+0.066
onna+0.064
 slippery+0.064
Neg logit contributions
aroo-0.092
ter-0.081
itte-0.073
apped-0.072
crow-0.072
Token mess
Token ID2085
Feature activation-6.05e-01

Pos logit contributions
 mess+0.216
aj+0.192
 wra+0.174
 slippery+0.169
 fault+0.168
Neg logit contributions
aroo-0.291
ter-0.269
crow-0.242
apped-0.240
itte-0.238
Token again
Token ID757
Feature activation-1.26e-01

Pos logit contributions
 mess+1.198
aj+1.067
 wra+1.013
onna+0.996
 pic+0.966
Neg logit contributions
aroo-1.201
ter-1.061
crow-0.948
apped-0.931
itte-0.915
Token.
Token ID13
Feature activation-3.32e-02

Pos logit contributions
 mess+0.238
aj+0.210
 wra+0.198
onna+0.196
 pic+0.191
Neg logit contributions
aroo-0.262
ter-0.234
crow-0.209
apped-0.207
itte-0.203
Token He
Token ID679
Feature activation-4.91e-02

Pos logit contributions
 mess+0.054
aj+0.051
onna+0.048
 wra+0.046
 pic+0.046
Neg logit contributions
aroo-0.073
ter-0.067
crow-0.059
apped-0.059
 member-0.058
Token jumped
Token ID11687
Feature activation-1.45e-01

Pos logit contributions
 mess+0.085
aj+0.071
 wra+0.068
onna+0.067
 slippery+0.065
Neg logit contributions
aroo-0.112
ter-0.101
crow-0.092
apped-0.090
itte-0.088
Token and
Token ID290
Feature activation-1.45e-01

Pos logit contributions
 mess+0.313
aj+0.293
 pic+0.272
 wra+0.270
liv+0.268
Neg logit contributions
aroo-0.258
ter-0.232
apped-0.196
crow-0.196
itte-0.188
Token week
Token ID1285
Feature activation-1.62e-01

Pos logit contributions
aroo+0.022
ter+0.020
apped+0.018
itte+0.017
crow+0.017
Neg logit contributions
 mess-0.020
aj-0.017
onna-0.016
 pic-0.016
 wra-0.016
Token and
Token ID290
Feature activation-1.71e-01

Pos logit contributions
 mess+0.296
aj+0.256
 wra+0.244
onna+0.238
 pic+0.236
Neg logit contributions
aroo-0.356
ter-0.316
crow-0.286
itte-0.281
apped-0.280
Token making
Token ID1642
Feature activation-1.79e-01

Pos logit contributions
 mess+0.320
aj+0.276
 wra+0.261
onna+0.256
 pic+0.254
Neg logit contributions
aroo-0.366
ter-0.325
crow-0.292
apped-0.284
itte-0.283
Token a
Token ID257
Feature activation-6.60e-02

Pos logit contributions
 mess+0.327
aj+0.301
 wra+0.281
onna+0.279
liv+0.269
Neg logit contributions
aroo-0.368
ter-0.333
crow-0.295
apped-0.291
 member-0.284
Token huge
Token ID3236
Feature activation-1.55e-01

Pos logit contributions
 mess+0.107
aj+0.095
 wra+0.088
onna+0.087
 pic+0.084
Neg logit contributions
aroo-0.152
ter-0.138
crow-0.125
apped-0.123
itte-0.121
Token mess
Token ID2085
Feature activation-6.04e-01

Pos logit contributions
 mess+0.277
aj+0.227
 wra+0.217
onna+0.214
 pic+0.210
Neg logit contributions
aroo-0.351
ter-0.316
crow-0.287
apped-0.284
itte-0.282
Token.
Token ID13
Feature activation-5.23e-02

Pos logit contributions
 mess+1.199
 wra+0.989
aj+0.981
onna+0.943
 slippery+0.911
Neg logit contributions
aroo-1.248
ter-1.041
crow-0.986
apped-0.965
itte-0.963
Token It
Token ID632
Feature activation+1.24e-03

Pos logit contributions
 mess+0.089
aj+0.081
onna+0.076
 wra+0.074
 pic+0.074
Neg logit contributions
aroo-0.116
ter-0.106
crow-0.093
apped-0.091
itte-0.091
Token was
Token ID373
Feature activation-1.08e-01

Pos logit contributions
aroo+0.003
ter+0.002
crow+0.002
itte+0.002
apped+0.002
Neg logit contributions
 mess-0.002
aj-0.002
 wra-0.002
aman-0.002
onna-0.002
Token so
Token ID523
Feature activation-1.45e-01

Pos logit contributions
 mess+0.177
aj+0.172
icol+0.157
 wra+0.155
liv+0.155
Neg logit contributions
aroo-0.226
ter-0.213
apped-0.183
crow-0.179
itte-0.178
Token sm
Token ID895
Feature activation-2.23e-01

Pos logit contributions
 mess+0.206
aj+0.181
 wra+0.166
onna+0.162
 pic+0.160
Neg logit contributions
aroo-0.364
ter-0.333
apped-0.303
crow-0.303
itte-0.298
Token.
Token ID13
Feature activation-4.92e-02

Pos logit contributions
 mess+0.343
aj+0.297
onna+0.286
 fault+0.273
 wra+0.269
Neg logit contributions
aroo-0.496
ter-0.444
itte-0.411
crow-0.403
apped-0.402
Token They
Token ID1119
Feature activation-9.42e-02

Pos logit contributions
 mess+0.081
aj+0.077
onna+0.074
 pic+0.070
 wra+0.070
Neg logit contributions
aroo-0.107
ter-0.098
 member-0.085
crow-0.085
apped-0.084
Token only
Token ID691
Feature activation-6.21e-02

Pos logit contributions
 mess+0.178
 wra+0.147
aj+0.144
onna+0.143
 slippery+0.139
Neg logit contributions
aroo-0.205
ter-0.184
apped-0.164
crow-0.164
ive-0.157
Token had
Token ID550
Feature activation-9.29e-02

Pos logit contributions
 mess+0.115
onna+0.091
 wra+0.091
 pays+0.088
 fuss+0.088
Neg logit contributions
aroo-0.146
ter-0.131
crow-0.114
apped-0.114
itte-0.111
Token a
Token ID257
Feature activation-6.33e-02

Pos logit contributions
 mess+0.170
aj+0.152
 wra+0.141
onna+0.140
liv+0.137
Neg logit contributions
aroo-0.195
ter-0.176
crow-0.157
apped-0.155
 member-0.151
Token mess
Token ID2085
Feature activation-6.04e-01

Pos logit contributions
 mess+0.116
aj+0.109
liv+0.101
 wra+0.098
onna+0.098
Neg logit contributions
aroo-0.129
ter-0.114
crow-0.100
 member-0.099
 pilot-0.099
Token.
Token ID13
Feature activation-3.66e-02

Pos logit contributions
 mess+1.131
 wra+0.967
aj+0.929
onna+0.876
 slippery+0.852
Neg logit contributions
aroo-1.279
ter-1.040
crow-1.038
itte-0.998
apped-0.986
Token<|endoftext|>
Token ID50256
Feature activation+2.07e-02

Pos logit contributions
 mess+0.065
aj+0.059
onna+0.055
 wra+0.055
 pic+0.053
Neg logit contributions
aroo-0.079
ter-0.072
crow-0.063
apped-0.062
itte-0.062
TokenS
Token ID50
Feature activation-1.22e-01

Pos logit contributions
aroo+0.043
ter+0.039
crow+0.034
itte+0.034
apped+0.033
Neg logit contributions
 mess-0.038
aj-0.034
 wra-0.032
onna-0.032
 pic-0.032
Tokenara
Token ID3301
Feature activation-2.92e-01

Pos logit contributions
 mess+0.176
 wra+0.138
 slippery+0.136
 fault+0.135
aj+0.133
Neg logit contributions
aroo-0.289
ter-0.268
 member-0.251
crow-0.245
itte-0.243
Token and
Token ID290
Feature activation-1.43e-01

Pos logit contributions
 mess+0.450
aj+0.385
 wra+0.357
onna+0.351
 pic+0.336
Neg logit contributions
aroo-0.714
ter-0.646
crow-0.587
apped-0.576
itte-0.572
Token microwave
Token ID27000
Feature activation-2.79e-01

Pos logit contributions
 mess+0.102
aj+0.092
onna+0.087
 fault+0.086
 wra+0.085
Neg logit contributions
aroo-0.101
ter-0.091
apped-0.079
crow-0.079
itte-0.076
Token and
Token ID290
Feature activation-1.54e-01

Pos logit contributions
 mess+0.516
aj+0.467
 pic+0.438
 wra+0.434
onna+0.428
Neg logit contributions
aroo-0.570
ter-0.519
apped-0.460
crow-0.459
 member-0.436
Token saw
Token ID2497
Feature activation-5.22e-02

Pos logit contributions
 mess+0.283
aj+0.239
 wra+0.237
onna+0.225
 slippery+0.219
Neg logit contributions
aroo-0.335
ter-0.299
crow-0.274
apped-0.271
ovable-0.268
Token a
Token ID257
Feature activation-7.05e-02

Pos logit contributions
 mess+0.099
aj+0.088
 wra+0.083
onna+0.082
 pic+0.080
Neg logit contributions
aroo-0.108
ter-0.097
crow-0.086
apped-0.085
itte-0.083
Token black
Token ID2042
Feature activation-2.29e-01

Pos logit contributions
 mess+0.110
aj+0.110
liv+0.103
 fault+0.103
onna+0.101
Neg logit contributions
aroo-0.150
ter-0.134
 pilot-0.124
 member-0.120
crow-0.119
Token mess
Token ID2085
Feature activation-6.04e-01

Pos logit contributions
 mess+0.415
 wra+0.341
aj+0.338
onna+0.316
 pic+0.304
Neg logit contributions
aroo-0.504
ter-0.447
apped-0.414
ovable-0.413
itte-0.411
Token.
Token ID13
Feature activation-4.33e-02

Pos logit contributions
 mess+1.267
 wra+1.075
aj+1.062
onna+1.004
 slippery+0.987
Neg logit contributions
aroo-1.176
ter-0.976
crow-0.924
itte-0.894
apped-0.890
Token Smoke
Token ID25416
Feature activation-7.84e-02

Pos logit contributions
 mess+0.081
aj+0.072
 wra+0.068
onna+0.068
 pic+0.065
Neg logit contributions
aroo-0.091
ter-0.082
crow-0.072
apped-0.071
 member-0.071
Token came
Token ID1625
Feature activation+2.74e-03

Pos logit contributions
 mess+0.138
aj+0.124
 wra+0.115
onna+0.114
 pic+0.112
Neg logit contributions
aroo-0.170
ter-0.155
crow-0.138
apped-0.136
itte-0.132
Token out
Token ID503
Feature activation+2.21e-02

Pos logit contributions
aroo+0.005
ter+0.004
crow+0.003
apped+0.003
itte+0.003
Neg logit contributions
 mess-0.006
aj-0.006
 wra-0.005
onna-0.005
 pic-0.005
Token.
Token ID13
Feature activation-3.31e-02

Pos logit contributions
aroo+0.031
ter+0.025
crow+0.022
itte+0.021
apped+0.020
Neg logit contributions
 mess-0.060
aj-0.051
onna-0.051
 wra-0.051
 slippery-0.050
Token loud
Token ID7812
Feature activation-2.48e-01

Pos logit contributions
aroo+0.304
ter+0.274
crow+0.250
apped+0.250
itte+0.247
Neg logit contributions
 mess-0.200
aj-0.170
 wra-0.161
onna-0.160
 pic-0.154
Token noise
Token ID7838
Feature activation-2.17e-01

Pos logit contributions
 mess+0.403
aj+0.359
 wra+0.333
onna+0.328
 pic+0.318
Neg logit contributions
aroo-0.573
ter-0.520
crow-0.472
apped-0.462
itte-0.452
Token and
Token ID290
Feature activation-1.59e-01

Pos logit contributions
 mess+0.412
aj+0.353
 wra+0.337
onna+0.330
 fault+0.319
Neg logit contributions
aroo-0.458
ter-0.404
crow-0.364
apped-0.360
itte-0.359
Token a
Token ID257
Feature activation-6.40e-02

Pos logit contributions
 mess+0.272
aj+0.239
 wra+0.223
onna+0.220
 pic+0.213
Neg logit contributions
aroo-0.359
ter-0.325
crow-0.292
apped-0.289
itte-0.284
Token big
Token ID1263
Feature activation-1.67e-01

Pos logit contributions
 mess+0.102
aj+0.094
 wra+0.085
onna+0.084
 pic+0.083
Neg logit contributions
aroo-0.147
ter-0.132
crow-0.120
apped-0.117
 member-0.116
Token mess
Token ID2085
Feature activation-6.03e-01

Pos logit contributions
 mess+0.320
aj+0.274
 wra+0.263
onna+0.257
 pic+0.249
Neg logit contributions
aroo-0.348
ter-0.313
apped-0.278
crow-0.277
itte-0.273
Token.
Token ID13
Feature activation-4.31e-02

Pos logit contributions
 mess+1.197
 wra+0.996
aj+0.971
onna+0.919
 slippery+0.908
Neg logit contributions
aroo-1.233
ter-1.041
crow-0.980
apped-0.968
itte-0.967
Token Lily
Token ID20037
Feature activation-1.11e-01

Pos logit contributions
 mess+0.079
aj+0.072
onna+0.067
 wra+0.067
 pic+0.065
Neg logit contributions
aroo-0.091
ter-0.082
crow-0.071
apped-0.071
 member-0.070
Token was
Token ID373
Feature activation-1.12e-01

Pos logit contributions
 mess+0.195
aj+0.190
 pic+0.180
onna+0.180
aman+0.178
Neg logit contributions
aroo-0.216
ter-0.199
crow-0.179
apped-0.177
itte-0.176
Token scared
Token ID12008
Feature activation+1.08e-02

Pos logit contributions
 mess+0.209
aj+0.170
 slippery+0.162
 wra+0.161
onna+0.159
Neg logit contributions
aroo-0.247
ter-0.219
crow-0.201
ive-0.195
apped-0.195
Token and
Token ID290
Feature activation-1.62e-01

Pos logit contributions
aroo+0.022
ter+0.020
crow+0.018
apped+0.017
 member+0.017
Neg logit contributions
 mess-0.020
aj-0.019
onna-0.017
 wra-0.017
liv-0.017
Token.
Token ID13
Feature activation-2.10e-02

Pos logit contributions
 mess+0.218
aj+0.213
onna+0.203
liv+0.199
aman+0.199
Neg logit contributions
aroo-0.193
ter-0.176
apped-0.153
crow-0.152
itte-0.150
Token They
Token ID1119
Feature activation-7.20e-02

Pos logit contributions
 mess+0.032
aj+0.031
onna+0.029
 pic+0.028
 wra+0.027
Neg logit contributions
aroo-0.047
ter-0.044
itte-0.038
ovable-0.038
 member-0.038
Token look
Token ID804
Feature activation-2.01e-01

Pos logit contributions
 mess+0.129
 wra+0.101
 slippery+0.100
aj+0.098
onna+0.096
Neg logit contributions
aroo-0.166
ter-0.150
apped-0.134
crow-0.130
ive-0.129
Token at
Token ID379
Feature activation+8.03e-02

Pos logit contributions
 mess+0.462
aj+0.414
 wra+0.395
onna+0.387
 pic+0.382
Neg logit contributions
aroo-0.345
ter-0.298
crow-0.258
apped-0.251
itte-0.244
Token the
Token ID262
Feature activation-3.27e-02

Pos logit contributions
aroo+0.157
ter+0.139
crow+0.123
apped+0.121
itte+0.119
Neg logit contributions
 mess-0.163
aj-0.145
 wra-0.138
onna-0.136
 pic-0.133
Token mess
Token ID2085
Feature activation-6.03e-01

Pos logit contributions
 mess+0.062
aj+0.052
 wra+0.051
onna+0.048
 pic+0.047
Neg logit contributions
aroo-0.071
ter-0.063
apped-0.057
crow-0.057
itte-0.056
Token of
Token ID286
Feature activation+2.66e-01

Pos logit contributions
 mess+1.130
aj+0.995
 wra+0.945
onna+0.926
 pic+0.896
Neg logit contributions
aroo-1.264
ter-1.124
crow-1.010
apped-0.992
itte-0.979
Token blocks
Token ID7021
Feature activation-1.76e-01

Pos logit contributions
aroo+0.540
ter+0.488
crow+0.426
apped+0.421
 member+0.411
Neg logit contributions
 mess-0.498
aj-0.460
onna-0.425
 wra-0.421
liv-0.413
Token.
Token ID13
Feature activation-5.53e-02

Pos logit contributions
 mess+0.340
aj+0.316
 pic+0.290
icol+0.290
onna+0.290
Neg logit contributions
aroo-0.343
ter-0.312
crow-0.270
apped-0.269
 member-0.263
Token
Token ID198
Feature activation-7.78e-02

Pos logit contributions
 mess+0.100
aj+0.090
onna+0.084
 wra+0.084
 pic+0.082
Neg logit contributions
aroo-0.119
ter-0.107
crow-0.094
apped-0.093
itte-0.092
Token
Token ID198
Feature activation-7.55e-02

Pos logit contributions
 mess+0.170
aj+0.151
 wra+0.147
 fault+0.144
icol+0.143
Neg logit contributions
aroo-0.139
ter-0.114
crow-0.108
apped-0.104
itte-0.102
Token he
Token ID339
Feature activation-3.73e-02

Pos logit contributions
aroo+0.188
ovable+0.173
onel+0.173
rest+0.167
apped+0.166
Neg logit contributions
 mess-0.272
 scar-0.233
 wra-0.231
 snowball-0.225
 slippery-0.224
Token would
Token ID561
Feature activation-1.47e-01

Pos logit contributions
 mess+0.066
aj+0.055
onna+0.052
 wra+0.051
 slippery+0.050
Neg logit contributions
aroo-0.084
ter-0.076
crow-0.067
apped-0.066
itte-0.065
Token make
Token ID787
Feature activation-2.06e-02

Pos logit contributions
 mess+0.284
onna+0.234
aj+0.234
 pic+0.224
 slippery+0.222
Neg logit contributions
aroo-0.306
ter-0.276
crow-0.249
apped-0.246
itte-0.240
Token a
Token ID257
Feature activation-5.75e-02

Pos logit contributions
 mess+0.038
aj+0.035
 wra+0.032
onna+0.032
 pic+0.031
Neg logit contributions
aroo-0.043
ter-0.038
crow-0.034
apped-0.034
itte-0.033
Token huge
Token ID3236
Feature activation-1.56e-01

Pos logit contributions
 mess+0.094
aj+0.086
 wra+0.080
onna+0.078
liv+0.077
Neg logit contributions
aroo-0.129
ter-0.116
crow-0.104
apped-0.103
 member-0.102
Token mess
Token ID2085
Feature activation-6.02e-01

Pos logit contributions
 mess+0.285
aj+0.232
 wra+0.221
onna+0.219
 pic+0.214
Neg logit contributions
aroo-0.350
ter-0.315
crow-0.287
apped-0.283
itte-0.282
Token in
Token ID287
Feature activation+1.90e-01

Pos logit contributions
 mess+1.183
aj+0.987
 wra+0.981
onna+0.952
 slippery+0.916
Neg logit contributions
aroo-1.233
ter-1.050
crow-0.981
apped-0.965
itte-0.957
Token his
Token ID465
Feature activation-1.27e-01

Pos logit contributions
aroo+0.326
ter+0.294
crow+0.252
 member+0.247
 person+0.235
Neg logit contributions
aj-0.395
 mess-0.378
zan-0.357
icol-0.356
liv-0.354
Token house
Token ID2156
Feature activation+2.46e-02

Pos logit contributions
 mess+0.215
aj+0.188
 wra+0.173
onna+0.167
 pic+0.167
Neg logit contributions
aroo-0.295
ter-0.269
crow-0.242
apped-0.238
itte-0.235
Token,
Token ID11
Feature activation-7.06e-02

Pos logit contributions
aroo+0.049
ter+0.044
crow+0.039
apped+0.038
itte+0.038
Neg logit contributions
 mess-0.048
aj-0.043
 wra-0.041
onna-0.040
 pic-0.039
Token but
Token ID475
Feature activation-2.61e-01

Pos logit contributions
 mess+0.118
aj+0.105
 wra+0.097
onna+0.096
liv+0.093
Neg logit contributions
aroo-0.159
ter-0.145
crow-0.129
apped-0.128
itte-0.125
Token he
Token ID339
Feature activation-6.26e-02

Pos logit contributions
aroo+0.124
ter+0.111
apped+0.103
crow+0.101
itte+0.101
Neg logit contributions
 mess-0.099
 wra-0.081
aj-0.081
onna-0.078
 pic-0.078
Token couldn
Token ID3521
Feature activation-8.89e-02

Pos logit contributions
 mess+0.107
aj+0.085
 wra+0.084
onna+0.081
 slippery+0.081
Neg logit contributions
aroo-0.146
ter-0.132
apped-0.121
crow-0.120
itte-0.112
Token't
Token ID470
Feature activation-1.12e-01

Pos logit contributions
 mess+0.128
onna+0.103
aj+0.099
 wra+0.099
liv+0.095
Neg logit contributions
aroo-0.220
ter-0.204
apped-0.188
crow-0.183
 sailors-0.181
Token clear
Token ID1598
Feature activation-1.69e-01

Pos logit contributions
 mess+0.193
onna+0.155
 wob+0.148
 unw+0.146
 slippery+0.141
Neg logit contributions
aroo-0.248
apped-0.239
ter-0.226
itte-0.209
crow-0.205
Token the
Token ID262
Feature activation-1.37e-02

Pos logit contributions
 mess+0.307
aj+0.292
liv+0.269
 wra+0.267
onna+0.266
Neg logit contributions
aroo-0.344
ter-0.315
crow-0.281
 member-0.262
apped-0.261
Token mess
Token ID2085
Feature activation-6.02e-01

Pos logit contributions
 mess+0.024
aj+0.020
 wra+0.020
onna+0.019
 fault+0.019
Neg logit contributions
aroo-0.031
ter-0.028
apped-0.026
crow-0.025
itte-0.025
Token by
Token ID416
Feature activation+2.86e-02

Pos logit contributions
 mess+1.176
 wra+1.000
aj+0.990
onna+0.946
 slippery+0.926
Neg logit contributions
aroo-1.224
ter-1.041
crow-0.969
apped-0.966
itte-0.938
Token himself
Token ID2241
Feature activation-4.63e-02

Pos logit contributions
aroo+0.059
ter+0.053
crow+0.047
apped+0.045
itte+0.045
Neg logit contributions
 mess-0.053
aj-0.049
 wra-0.046
onna-0.045
 pic-0.044
Token.
Token ID13
Feature activation-2.73e-02

Pos logit contributions
 mess+0.089
aj+0.079
 wra+0.074
onna+0.073
 fault+0.071
Neg logit contributions
aroo-0.096
ter-0.085
crow-0.076
apped-0.075
itte-0.073
Token
Token ID198
Feature activation-4.94e-02

Pos logit contributions
 mess+0.050
aj+0.044
 wra+0.041
onna+0.041
 pic+0.040
Neg logit contributions
aroo-0.059
ter-0.053
crow-0.047
apped-0.046
itte-0.046
Token
Token ID198
Feature activation-4.81e-02

Pos logit contributions
 mess+0.094
aj+0.085
onna+0.079
 wra+0.079
 pic+0.077
Neg logit contributions
aroo-0.101
ter-0.091
crow-0.080
apped-0.080
itte-0.077
Token,
Token ID11
Feature activation-5.24e-02

Pos logit contributions
aroo+0.031
ter+0.027
apped+0.025
crow+0.024
itte+0.024
Neg logit contributions
 mess-0.025
 fun-0.021
 slippery-0.021
aj-0.020
die-0.019
Token the
Token ID262
Feature activation-1.17e-02

Pos logit contributions
 mess+0.092
aj+0.079
 wra+0.074
onna+0.073
 pic+0.072
Neg logit contributions
aroo-0.116
ter-0.103
apped-0.096
crow-0.095
itte-0.094
Token garden
Token ID11376
Feature activation-9.85e-02

Pos logit contributions
 mess+0.026
aj+0.021
 wra+0.021
 slippery+0.020
 mistake+0.020
Neg logit contributions
aroo-0.022
ter-0.019
apped-0.018
ovable-0.017
itte-0.017
Token was
Token ID373
Feature activation-1.11e-01

Pos logit contributions
 mess+0.162
aj+0.137
 wra+0.131
onna+0.127
 pic+0.124
Neg logit contributions
aroo-0.232
ter-0.208
crow-0.189
apped-0.186
itte-0.185
Token a
Token ID257
Feature activation-6.02e-02

Pos logit contributions
 mess+0.213
aj+0.202
 wra+0.188
icol+0.185
 fault+0.182
Neg logit contributions
aroo-0.212
ter-0.194
crow-0.166
apped-0.164
 member-0.161
Token mess
Token ID2085
Feature activation-6.01e-01

Pos logit contributions
 mess+0.104
aj+0.092
 wra+0.086
onna+0.085
 pic+0.082
Neg logit contributions
aroo-0.134
ter-0.121
crow-0.109
apped-0.108
itte-0.105
Token.
Token ID13
Feature activation-4.25e-02

Pos logit contributions
 mess+1.105
aj+0.910
 wra+0.909
onna+0.863
 slippery+0.838
Neg logit contributions
aroo-1.317
ter-1.141
crow-1.065
apped-1.052
itte-1.039
Token Polly
Token ID36898
Feature activation-1.41e-01

Pos logit contributions
 mess+0.070
aj+0.066
onna+0.062
 pic+0.060
 wra+0.059
Neg logit contributions
aroo-0.094
ter-0.085
 member-0.074
crow-0.074
apped-0.074
Token and
Token ID290
Feature activation-1.48e-01

Pos logit contributions
 mess+0.259
aj+0.230
 wra+0.216
onna+0.213
 pic+0.207
Neg logit contributions
aroo-0.301
ter-0.270
crow-0.241
apped-0.238
itte-0.234
Token her
Token ID607
Feature activation-1.18e-01

Pos logit contributions
 mess+0.216
aj+0.174
 wra+0.171
onna+0.164
 pic+0.163
Neg logit contributions
aroo-0.376
ter-0.339
apped-0.311
crow-0.308
itte-0.303
Token friends
Token ID2460
Feature activation+2.37e-01

Pos logit contributions
 mess+0.271
aj+0.237
 wra+0.226
 fault+0.223
 pic+0.211
Neg logit contributions
aroo-0.213
ter-0.187
crow-0.173
itte-0.158
apped-0.156
Token throwing
Token ID9644
Feature activation-1.95e-01

Pos logit contributions
aj+0.129
 mess+0.126
onna+0.126
aman+0.121
liv+0.121
Neg logit contributions
aroo-0.122
ter-0.120
crow-0.095
 sailors-0.091
al-0.090
Token blocks
Token ID7021
Feature activation-1.60e-01

Pos logit contributions
 mess+0.355
aj+0.324
onna+0.298
 wra+0.297
 pic+0.288
Neg logit contributions
aroo-0.413
ter-0.372
crow-0.335
apped-0.326
itte-0.316
Token and
Token ID290
Feature activation-1.67e-01

Pos logit contributions
 mess+0.334
aj+0.299
 wra+0.285
onna+0.281
 pic+0.273
Neg logit contributions
aroo-0.298
ter-0.263
crow-0.232
apped-0.228
itte-0.224
Token making
Token ID1642
Feature activation-1.69e-01

Pos logit contributions
 mess+0.261
aj+0.247
aman+0.224
liv+0.223
onna+0.220
Neg logit contributions
aroo-0.382
ter-0.354
crow-0.313
apped-0.312
 member-0.301
Token a
Token ID257
Feature activation-6.88e-02

Pos logit contributions
aj+0.299
 mess+0.298
liv+0.274
onna+0.270
aman+0.268
Neg logit contributions
aroo-0.332
ter-0.306
crow-0.266
apped-0.265
 sailors-0.257
Token mess
Token ID2085
Feature activation-6.00e-01

Pos logit contributions
aj+0.086
 mess+0.081
liv+0.076
 wra+0.072
aman+0.072
Neg logit contributions
aroo-0.171
ter-0.159
 pilot-0.146
crow-0.144
apped-0.143
Token?
Token ID30
Feature activation-1.44e-01

Pos logit contributions
 mess+1.161
 wra+0.941
aj+0.939
onna+0.897
 slippery+0.864
Neg logit contributions
aroo-1.292
ter-1.092
crow-1.026
itte-1.000
apped-0.996
Token Why
Token ID4162
Feature activation+1.09e-01

Pos logit contributions
 mess+0.241
aj+0.219
onna+0.210
 wra+0.204
 fault+0.195
Neg logit contributions
aroo-0.327
ter-0.288
crow-0.257
 pilot-0.254
 member-0.250
Token are
Token ID389
Feature activation+9.95e-03

Pos logit contributions
aroo+0.212
ter+0.194
apped+0.165
itte+0.165
crow+0.165
Neg logit contributions
 mess-0.202
aj-0.193
icol-0.180
aman-0.179
 wra-0.178
Token you
Token ID345
Feature activation-1.41e-02

Pos logit contributions
aroo+0.021
ter+0.018
crow+0.017
apped+0.016
ive+0.016
Neg logit contributions
 mess-0.019
aj-0.017
 wra-0.016
onna-0.015
 pic-0.015
Token being
Token ID852
Feature activation-7.40e-02

Pos logit contributions
 mess+0.023
aj+0.021
 wra+0.020
 pic+0.020
onna+0.020
Neg logit contributions
aroo-0.031
ter-0.029
apped-0.025
itte-0.025
 member-0.025
Token woke
Token ID19092
Feature activation-9.16e-02

Pos logit contributions
aroo+0.390
ter+0.343
crow+0.310
ative+0.302
itte+0.298
Neg logit contributions
 mess-0.379
aj-0.336
onna-0.310
 wra-0.304
 pic-0.298
Token up
Token ID510
Feature activation-1.63e-01

Pos logit contributions
 mess+0.155
aj+0.140
 wra+0.133
onna+0.130
 pic+0.127
Neg logit contributions
aroo-0.203
ter-0.187
crow-0.163
apped-0.163
itte-0.159
Token to
Token ID284
Feature activation-1.47e-01

Pos logit contributions
 mess+0.281
aj+0.245
 wra+0.230
onna+0.227
 pic+0.220
Neg logit contributions
aroo-0.367
ter-0.330
crow-0.298
apped-0.293
itte-0.290
Token a
Token ID257
Feature activation-6.27e-02

Pos logit contributions
 mess+0.270
 pic+0.214
 fuss+0.213
onna+0.211
aj+0.208
Neg logit contributions
aroo-0.327
ter-0.287
apped-0.269
itte-0.262
ovable-0.257
Token big
Token ID1263
Feature activation-1.68e-01

Pos logit contributions
 mess+0.116
aj+0.097
 wra+0.094
onna+0.091
 slippery+0.089
Neg logit contributions
aroo-0.137
ter-0.124
apped-0.114
crow-0.111
itte-0.109
Token mess
Token ID2085
Feature activation-6.00e-01

Pos logit contributions
 mess+0.331
aj+0.263
 wra+0.259
onna+0.251
 fuss+0.250
Neg logit contributions
aroo-0.355
ter-0.322
apped-0.299
ovable-0.289
itte-0.288
Token.
Token ID13
Feature activation-3.71e-02

Pos logit contributions
 mess+1.206
 wra+1.007
aj+0.987
onna+0.938
 slippery+0.920
Neg logit contributions
aroo-1.202
ter-1.016
apped-0.950
itte-0.943
crow-0.942
Token The
Token ID383
Feature activation+1.27e-01

Pos logit contributions
 mess+0.063
aj+0.058
onna+0.054
 wra+0.052
aman+0.052
Neg logit contributions
aroo-0.084
ter-0.075
apped-0.065
crow-0.065
 member-0.065
Token rh
Token ID9529
Feature activation+1.29e-01

Pos logit contributions
aroo+0.265
ter+0.227
apped+0.226
ovable+0.211
crow+0.210
Neg logit contributions
 mess-0.263
 wra-0.212
aj-0.207
 slippery-0.199
 mistake-0.197
Tokenin
Token ID259
Feature activation+2.03e-01

Pos logit contributions
aroo+0.183
ter+0.167
ovable+0.154
apped+0.153
crow+0.151
Neg logit contributions
 mess-0.304
 wra-0.254
 mistake-0.238
corn-0.236
aj-0.232
Tokenoc
Token ID420
Feature activation+5.37e-02

Pos logit contributions
aroo+0.391
ter+0.359
apped+0.322
crow+0.315
ovable+0.313
Neg logit contributions
 mess-0.421
aj-0.361
 wra-0.347
onna-0.332
 slippery-0.325
Token it
Token ID340
Feature activation-1.65e-01

Pos logit contributions
 mess+0.277
aj+0.243
 wra+0.225
onna+0.222
 pic+0.215
Neg logit contributions
aroo-0.403
ter-0.365
crow-0.330
apped-0.326
itte-0.321
Token was
Token ID373
Feature activation-1.13e-01

Pos logit contributions
 mess+0.260
aj+0.225
 wra+0.210
onna+0.206
 pic+0.199
Neg logit contributions
aroo-0.392
ter-0.356
crow-0.323
apped-0.319
itte-0.314
Token making
Token ID1642
Feature activation-1.71e-01

Pos logit contributions
 mess+0.192
aj+0.183
 wra+0.167
icol+0.167
liv+0.166
Neg logit contributions
aroo-0.236
ter-0.217
crow-0.184
apped-0.181
 member-0.181
Token a
Token ID257
Feature activation-6.15e-02

Pos logit contributions
 mess+0.306
aj+0.293
onna+0.270
 wra+0.265
liv+0.263
Neg logit contributions
aroo-0.344
ter-0.318
crow-0.280
apped-0.272
 member-0.263
Token big
Token ID1263
Feature activation-1.67e-01

Pos logit contributions
aj+0.086
 mess+0.081
liv+0.076
 wra+0.073
onna+0.072
Neg logit contributions
aroo-0.146
ter-0.136
 pilot-0.123
crow-0.119
 hero-0.119
Token mess
Token ID2085
Feature activation-6.00e-01

Pos logit contributions
aj+0.207
 mess+0.192
liv+0.178
onna+0.172
 wra+0.172
Neg logit contributions
aroo-0.429
ter-0.401
crow-0.356
 pilot-0.354
apped-0.352
Token.
Token ID13
Feature activation-4.43e-02

Pos logit contributions
 mess+1.147
 wra+0.942
aj+0.933
onna+0.903
 slippery+0.873
Neg logit contributions
aroo-1.267
ter-1.074
crow-1.014
apped-1.009
itte-0.994
Token So
Token ID1406
Feature activation-9.88e-02

Pos logit contributions
 mess+0.074
aj+0.069
onna+0.064
 pic+0.062
 wra+0.062
Neg logit contributions
aroo-0.098
ter-0.090
crow-0.078
apped-0.078
 member-0.078
Token,
Token ID11
Feature activation-6.02e-02

Pos logit contributions
 mess+0.162
aj+0.138
 wra+0.131
 pic+0.129
onna+0.127
Neg logit contributions
aroo-0.226
ter-0.202
itte-0.186
crow-0.186
apped-0.185
Token he
Token ID339
Feature activation-4.20e-02

Pos logit contributions
 mess+0.084
 pic+0.071
 wra+0.069
 slippery+0.065
 scar+0.064
Neg logit contributions
aroo-0.144
apped-0.129
ter-0.129
itte-0.128
ovable-0.128
Token grabbed
Token ID13176
Feature activation-9.92e-02

Pos logit contributions
 mess+0.080
aj+0.069
 wra+0.066
onna+0.065
 pic+0.063
Neg logit contributions
aroo-0.089
ter-0.079
crow-0.071
apped-0.070
ovable-0.068
Token went
Token ID1816
Feature activation+8.62e-02

Pos logit contributions
 mess+0.066
aj+0.055
onna+0.052
 wra+0.051
 slippery+0.050
Neg logit contributions
aroo-0.084
ter-0.076
crow-0.067
apped-0.066
itte-0.065
Token he
Token ID339
Feature activation-5.24e-02

Pos logit contributions
aroo+0.170
ter+0.151
crow+0.134
apped+0.132
itte+0.129
Neg logit contributions
 mess-0.172
aj-0.154
 wra-0.146
onna-0.143
 pic-0.141
Token made
Token ID925
Feature activation-1.42e-01

Pos logit contributions
 mess+0.091
aj+0.078
onna+0.073
 slippery+0.073
 wra+0.072
Neg logit contributions
aroo-0.118
ter-0.105
crow-0.096
apped-0.095
itte-0.094
Token a
Token ID257
Feature activation-5.93e-02

Pos logit contributions
 mess+0.255
aj+0.225
 wra+0.212
onna+0.209
 pic+0.203
Neg logit contributions
aroo-0.306
ter-0.276
crow-0.247
apped-0.244
itte-0.239
Token huge
Token ID3236
Feature activation-1.51e-01

Pos logit contributions
 mess+0.092
aj+0.089
liv+0.081
 wra+0.080
onna+0.079
Neg logit contributions
aroo-0.132
ter-0.119
 pilot-0.106
 member-0.106
crow-0.106
Token mess
Token ID2085
Feature activation-6.00e-01

Pos logit contributions
 mess+0.273
aj+0.232
 wra+0.220
onna+0.217
 pic+0.211
Neg logit contributions
aroo-0.333
ter-0.300
crow-0.271
apped-0.268
itte-0.264
Token.
Token ID13
Feature activation-3.90e-02

Pos logit contributions
 mess+1.139
 wra+0.944
aj+0.925
onna+0.904
 slippery+0.870
Neg logit contributions
aroo-1.265
ter-1.077
crow-1.027
apped-1.006
itte-1.002
Token One
Token ID1881
Feature activation-1.33e-01

Pos logit contributions
 mess+0.063
aj+0.060
onna+0.056
 pic+0.054
 wra+0.053
Neg logit contributions
aroo-0.088
ter-0.080
crow-0.070
ovable-0.069
apped-0.069
Token day
Token ID1110
Feature activation+5.33e-02

Pos logit contributions
 mess+0.278
 slippery+0.234
aj+0.225
 steep+0.222
onna+0.217
Neg logit contributions
aroo-0.255
ter-0.235
apped-0.209
ovable-0.203
crow-0.198
Token his
Token ID465
Feature activation-1.14e-01

Pos logit contributions
aroo+0.122
ter+0.111
ovable+0.100
crow+0.100
apped+0.100
Neg logit contributions
 mess-0.084
 wra-0.071
aj-0.071
onna-0.068
 pic-0.067
Token mom
Token ID1995
Feature activation+4.83e-02

Pos logit contributions
 mess+0.234
aj+0.211
 fault+0.197
 wra+0.193
 pic+0.192
Neg logit contributions
aroo-0.221
ter-0.199
crow-0.189
apped-0.173
itte-0.168
Token.
Token ID13
Feature activation+5.71e-02

Pos logit contributions
 mess+0.271
aj+0.241
 wra+0.227
onna+0.224
 fault+0.217
Neg logit contributions
aroo-0.291
ter-0.258
crow-0.231
apped-0.228
itte-0.224
Token That
Token ID1320
Feature activation+8.31e-02

Pos logit contributions
aroo+0.117
ter+0.100
 member+0.098
 pilot+0.095
crow+0.094
Neg logit contributions
aj-0.089
onna-0.085
 mess-0.083
 pays-0.083
icol-0.082
Token is
Token ID318
Feature activation-1.32e-01

Pos logit contributions
aroo+0.178
ter+0.167
crow+0.151
 member+0.149
 pilot+0.146
Neg logit contributions
 mess-0.138
aj-0.137
onna-0.132
 wra-0.120
icol-0.117
Token very
Token ID845
Feature activation-1.77e-02

Pos logit contributions
 mess+0.249
aj+0.224
 wra+0.209
onna+0.207
 fault+0.199
Neg logit contributions
aroo-0.276
ter-0.250
crow-0.219
apped-0.216
itte-0.213
Token kind
Token ID1611
Feature activation+1.92e-01

Pos logit contributions
 mess+0.030
aj+0.026
 wra+0.024
onna+0.024
 pic+0.024
Neg logit contributions
aroo-0.041
ter-0.037
crow-0.033
apped-0.033
 member-0.032
Token of
Token ID286
Feature activation+3.59e-01

Pos logit contributions
aroo+0.275
ter+0.222
itte+0.200
ovable+0.184
apped+0.184
Neg logit contributions
 mess-0.490
aj-0.436
 wra-0.426
onna-0.423
 fault-0.422
Token you
Token ID345
Feature activation+6.24e-02

Pos logit contributions
aroo+0.795
apped+0.695
ter+0.692
itte+0.663
ive+0.642
Neg logit contributions
 mess-0.630
 wra-0.518
 pic-0.512
aj-0.505
 slippery-0.504
Token."
Token ID526
Feature activation+7.20e-02

Pos logit contributions
aroo+0.156
ter+0.130
crow+0.130
itte+0.123
apped+0.122
Neg logit contributions
 mess-0.097
aj-0.082
 pays-0.077
 wra-0.073
 fault-0.072
Token Tim
Token ID5045
Feature activation-1.45e-02

Pos logit contributions
aroo+0.162
ter+0.147
apped+0.132
crow+0.132
itte+0.127
Neg logit contributions
 mess-0.124
aj-0.108
 wra-0.101
onna-0.100
 fault-0.097
Tokenmy
Token ID1820
Feature activation-1.38e-01

Pos logit contributions
 mess+0.025
 pic+0.022
 wra+0.022
onna+0.022
icol+0.022
Neg logit contributions
aroo-0.030
ter-0.029
 member-0.026
apped-0.025
crow-0.024
Token carried
Token ID5281
Feature activation+9.24e-02

Pos logit contributions
 mess+0.262
aj+0.233
 wra+0.221
onna+0.219
 fault+0.214
Neg logit contributions
aroo-0.284
ter-0.254
apped-0.225
crow-0.224
itte-0.219
Token art
Token ID1242
Feature activation-1.41e-01

Pos logit contributions
 mess+0.247
aj+0.205
 wra+0.202
onna+0.201
 fault+0.198
Neg logit contributions
aroo-0.218
ter-0.190
apped-0.177
itte-0.176
ovable-0.173
Token.
Token ID13
Feature activation+2.94e-03

Pos logit contributions
 mess+0.261
 wra+0.206
aj+0.200
onna+0.194
 pic+0.192
Neg logit contributions
aroo-0.311
ter-0.277
apped-0.258
itte-0.252
ovable-0.251
Token It
Token ID632
Feature activation+5.36e-02

Pos logit contributions
aroo+0.006
ter+0.006
 member+0.005
 citizens+0.005
 pilot+0.005
Neg logit contributions
 mess-0.004
aj-0.004
onna-0.004
 wra-0.004
 fault-0.004
Token is
Token ID318
Feature activation-1.72e-01

Pos logit contributions
aroo+0.135
ter+0.123
apped+0.112
crow+0.112
itte+0.109
Neg logit contributions
 mess-0.079
aj-0.066
 wra-0.062
onna-0.061
 fault-0.058
Token made
Token ID925
Feature activation-8.36e-02

Pos logit contributions
 mess+0.279
aj+0.254
 wra+0.232
onna+0.228
liv+0.227
Neg logit contributions
aroo-0.389
ter-0.358
crow-0.317
apped-0.311
itte-0.307
Token of
Token ID286
Feature activation+3.29e-01

Pos logit contributions
 mess+0.219
aj+0.190
 wra+0.184
onna+0.184
 pays+0.183
Neg logit contributions
aroo-0.120
ter-0.096
itte-0.083
crow-0.083
ive-0.078
Token paper
Token ID3348
Feature activation-2.54e-01

Pos logit contributions
aroo+0.762
ter+0.696
crow+0.626
apped+0.617
itte+0.605
Neg logit contributions
 mess-0.529
aj-0.470
 wra-0.432
onna-0.429
liv-0.414
Token and
Token ID290
Feature activation-1.08e-01

Pos logit contributions
 mess+0.446
aj+0.383
 wra+0.361
onna+0.350
 pic+0.340
Neg logit contributions
aroo-0.581
ter-0.514
crow-0.468
apped-0.467
itte-0.464
Token paint
Token ID7521
Feature activation+3.84e-02

Pos logit contributions
 mess+0.164
aj+0.144
 wra+0.131
onna+0.131
liv+0.126
Neg logit contributions
aroo-0.260
ter-0.238
crow-0.215
apped-0.213
itte-0.208
Token.
Token ID13
Feature activation+5.20e-03

Pos logit contributions
aroo+0.076
ter+0.068
crow+0.060
apped+0.059
itte+0.058
Neg logit contributions
 mess-0.076
aj-0.068
 wra-0.065
onna-0.064
 pic-0.062
Token It
Token ID632
Feature activation+5.77e-02

Pos logit contributions
aroo+0.011
ter+0.010
 member+0.009
 pilot+0.009
itte+0.009
Neg logit contributions
 mess-0.008
aj-0.008
onna-0.008
 pic-0.008
 wra-0.007
Token day
Token ID1110
Feature activation+1.08e-01

Pos logit contributions
 mess+0.329
aj+0.279
 slippery+0.278
 pic+0.268
 fun+0.267
Neg logit contributions
aroo-0.345
ter-0.326
crow-0.277
apped-0.274
itte-0.268
Token,
Token ID11
Feature activation-5.90e-03

Pos logit contributions
aroo+0.220
ter+0.195
apped+0.173
crow+0.173
itte+0.168
Neg logit contributions
 mess-0.212
aj-0.188
 wra-0.178
onna-0.176
 pic-0.172
Token a
Token ID257
Feature activation+6.37e-03

Pos logit contributions
 mess+0.010
aj+0.009
 wra+0.008
onna+0.008
 slippery+0.008
Neg logit contributions
aroo-0.013
ter-0.012
crow-0.011
ovable-0.010
 member-0.010
Token little
Token ID1310
Feature activation+8.68e-02

Pos logit contributions
aroo+0.014
ter+0.013
apped+0.012
crow+0.012
onel+0.012
Neg logit contributions
 mess-0.011
aj-0.009
 fun-0.009
 wra-0.009
 slippery-0.009
Token boy
Token ID2933
Feature activation+1.64e-01

Pos logit contributions
aroo+0.137
onel+0.133
ovable+0.133
rest+0.123
ter+0.119
Neg logit contributions
 mess-0.182
 wra-0.152
 snowball-0.151
 slippery-0.151
 steep-0.149
Token named
Token ID3706
Feature activation+3.42e-01

Pos logit contributions
aroo+0.237
ter+0.215
ative+0.185
onel+0.184
ovable+0.180
Neg logit contributions
 mess-0.389
 pic-0.347
 wra-0.332
array-0.319
icol-0.317
Token Tim
Token ID5045
Feature activation-3.21e-02

Pos logit contributions
aroo+0.763
ter+0.689
crow+0.619
apped+0.611
itte+0.600
Neg logit contributions
 mess-0.601
aj-0.530
 wra-0.496
onna-0.489
 pic-0.475
Token went
Token ID1816
Feature activation+1.40e-01

Pos logit contributions
 mess+0.069
aj+0.060
 wra+0.058
onna+0.056
 pic+0.056
Neg logit contributions
aroo-0.060
ter-0.052
crow-0.046
apped-0.045
itte-0.044
Token for
Token ID329
Feature activation+3.04e-02

Pos logit contributions
aroo+0.354
ter+0.324
crow+0.298
itte+0.290
apped+0.289
Neg logit contributions
 mess-0.204
aj-0.168
onna-0.157
 wra-0.152
 pic-0.151
Token a
Token ID257
Feature activation+8.57e-03

Pos logit contributions
aroo+0.053
ter+0.048
crow+0.040
apped+0.039
itte+0.039
Neg logit contributions
 mess-0.065
aj-0.060
onna-0.057
aman-0.056
 wra-0.056
Token walk
Token ID2513
Feature activation-2.34e-02

Pos logit contributions
aroo+0.017
ter+0.015
crow+0.014
 member+0.013
itte+0.013
Neg logit contributions
 mess-0.016
aj-0.015
 wra-0.014
onna-0.014
 fault-0.013
Token home
Token ID1363
Feature activation+1.30e-01

Pos logit contributions
 mess+0.128
aj+0.108
 wra+0.100
onna+0.099
 fault+0.099
Neg logit contributions
aroo-0.158
ter-0.143
crow-0.135
itte-0.129
apped-0.129
Token.
Token ID13
Feature activation+1.03e-02

Pos logit contributions
aroo+0.280
ter+0.247
crow+0.226
itte+0.220
apped+0.218
Neg logit contributions
 mess-0.245
aj-0.210
 wra-0.201
onna-0.200
 slippery-0.191
Token
Token ID198
Feature activation-1.45e-02

Pos logit contributions
aroo+0.022
ter+0.020
crow+0.017
apped+0.017
itte+0.017
Neg logit contributions
 mess-0.019
aj-0.017
onna-0.016
 wra-0.016
 pic-0.015
Token
Token ID198
Feature activation-1.41e-02

Pos logit contributions
 mess+0.029
aj+0.026
 wra+0.024
onna+0.024
 fault+0.023
Neg logit contributions
aroo-0.029
ter-0.026
crow-0.023
apped-0.023
itte-0.022
TokenL
Token ID43
Feature activation-7.16e-02

Pos logit contributions
 mess+0.024
aj+0.021
 wra+0.020
onna+0.019
liv+0.019
Neg logit contributions
aroo-0.032
ter-0.030
apped-0.026
crow-0.026
itte-0.025
Tokenily
Token ID813
Feature activation+3.11e-01

Pos logit contributions
 mess+0.212
aj+0.196
 wra+0.189
onna+0.187
 fault+0.185
Neg logit contributions
aroo-0.075
ter-0.057
crow-0.043
apped-0.041
 member-0.038
Token was
Token ID373
Feature activation-5.25e-02

Pos logit contributions
aroo+0.689
ter+0.622
crow+0.559
apped+0.550
itte+0.542
Neg logit contributions
 mess-0.545
aj-0.479
 wra-0.449
onna-0.441
 pic-0.429
Token very
Token ID845
Feature activation-3.68e-02

Pos logit contributions
 mess+0.095
aj+0.080
 slippery+0.076
 wra+0.075
onna+0.074
Neg logit contributions
aroo-0.117
ter-0.103
crow-0.095
apped-0.092
itte-0.091
Token brave
Token ID14802
Feature activation+1.98e-01

Pos logit contributions
 mess+0.069
 slippery+0.057
aj+0.057
onna+0.054
 steep+0.052
Neg logit contributions
aroo-0.080
ter-0.074
crow-0.067
ive-0.064
 member-0.063
Token and
Token ID290
Feature activation-1.02e-01

Pos logit contributions
aroo+0.385
ter+0.356
crow+0.314
 member+0.307
apped+0.304
Neg logit contributions
 mess-0.372
aj-0.354
liv-0.328
onna-0.326
 wra-0.321
Token smart
Token ID4451
Feature activation+2.63e-02

Pos logit contributions
 mess+0.199
aj+0.171
 slippery+0.170
 wra+0.170
onna+0.166
Neg logit contributions
aroo-0.201
ter-0.180
crow-0.162
apped-0.155
 member-0.154
Token the
Token ID262
Feature activation+4.72e-03

Pos logit contributions
 mess+0.134
aj+0.123
 fault+0.112
 wra+0.112
onna+0.111
Neg logit contributions
aroo-0.117
ter-0.106
crow-0.089
apped-0.086
itte-0.081
Token ladder
Token ID18002
Feature activation-1.55e-01

Pos logit contributions
aroo+0.010
ter+0.009
crow+0.008
apped+0.008
itte+0.008
Neg logit contributions
 mess-0.009
aj-0.008
 wra-0.007
onna-0.007
 pic-0.007
Token to
Token ID284
Feature activation-1.23e-01

Pos logit contributions
 mess+0.302
aj+0.275
 wra+0.259
 pic+0.252
onna+0.250
Neg logit contributions
aroo-0.311
ter-0.281
apped-0.253
 member-0.246
crow-0.246
Token the
Token ID262
Feature activation+5.66e-03

Pos logit contributions
 mess+0.234
onna+0.194
aj+0.194
 slippery+0.190
 wra+0.189
Neg logit contributions
aroo-0.254
ter-0.226
apped-0.213
crow-0.207
itte-0.203
Token top
Token ID1353
Feature activation-1.45e-01

Pos logit contributions
aroo+0.013
apped+0.012
ter+0.012
ovable+0.011
crow+0.011
Neg logit contributions
 mess-0.010
 slippery-0.008
 wra-0.007
 steep-0.007
aj-0.007
Token of
Token ID286
Feature activation+3.02e-01

Pos logit contributions
 mess+0.440
aj+0.414
 wra+0.399
onna+0.393
 fault+0.389
Neg logit contributions
aroo-0.134
ter-0.102
crow-0.075
apped-0.069
 member-0.065
Token the
Token ID262
Feature activation+7.82e-03

Pos logit contributions
aroo+0.572
ter+0.507
crow+0.442
apped+0.432
itte+0.422
Neg logit contributions
 mess-0.629
aj-0.575
 wra-0.537
onna-0.533
 fault-0.519
Token tree
Token ID5509
Feature activation+1.61e-02

Pos logit contributions
aroo+0.018
ter+0.017
apped+0.016
ovable+0.015
crow+0.015
Neg logit contributions
 mess-0.013
 wra-0.010
 slippery-0.010
aj-0.010
 steep-0.010
Token and
Token ID290
Feature activation-1.24e-01

Pos logit contributions
aroo+0.035
ter+0.031
crow+0.028
apped+0.027
itte+0.027
Neg logit contributions
 mess-0.029
aj-0.026
 wra-0.024
onna-0.024
 pic-0.023
Token looked
Token ID3114
Feature activation-1.73e-01

Pos logit contributions
 mess+0.223
aj+0.190
 wra+0.184
onna+0.180
 slippery+0.178
Neg logit contributions
aroo-0.268
ter-0.239
crow-0.220
apped-0.216
ovable-0.214
Token around
Token ID1088
Feature activation+1.52e-01

Pos logit contributions
 mess+0.377
aj+0.338
 wra+0.319
onna+0.315
 pic+0.308
Neg logit contributions
aroo-0.318
ter-0.277
crow-0.245
apped-0.237
itte-0.231
Token a
Token ID257
Feature activation+1.57e-02

Pos logit contributions
aroo+0.137
ter+0.125
apped+0.111
crow+0.111
 member+0.105
Neg logit contributions
 mess-0.182
aj-0.171
onna-0.164
 fault-0.155
liv-0.155
Token lot
Token ID1256
Feature activation-1.43e-01

Pos logit contributions
aroo+0.034
ter+0.031
crow+0.028
 member+0.027
 pilot+0.027
Neg logit contributions
 mess-0.026
aj-0.025
 wra-0.023
onna-0.023
 fault-0.022
Token of
Token ID286
Feature activation+3.46e-01

Pos logit contributions
 mess+0.407
aj+0.378
 wra+0.364
onna+0.361
 pic+0.355
Neg logit contributions
aroo-0.160
ter-0.129
crow-0.100
apped-0.096
itte-0.092
Token trees
Token ID7150
Feature activation+2.92e-02

Pos logit contributions
aroo+0.787
ter+0.713
crow+0.670
 member+0.630
apped+0.628
Neg logit contributions
 mess-0.540
aj-0.515
onna-0.491
liv-0.460
 fault-0.453
Token and
Token ID290
Feature activation-8.00e-02

Pos logit contributions
aroo+0.061
ter+0.056
crow+0.049
apped+0.049
 member+0.048
Neg logit contributions
 mess-0.054
aj-0.050
 wra-0.045
onna-0.045
liv-0.045
Token flowers
Token ID12734
Feature activation+9.45e-02

Pos logit contributions
 mess+0.161
aj+0.140
 wra+0.138
 pic+0.134
 slippery+0.132
Neg logit contributions
aroo-0.159
ter-0.139
apped-0.123
crow-0.123
itte-0.120
Token along
Token ID1863
Feature activation+2.67e-01

Pos logit contributions
aroo+0.203
ter+0.179
crow+0.164
itte+0.162
apped+0.159
Neg logit contributions
 mess-0.173
aj-0.147
 wra-0.145
onna-0.142
 pic-0.138
Token the
Token ID262
Feature activation+5.91e-02

Pos logit contributions
aroo+0.591
ter+0.522
itte+0.470
crow+0.457
apped+0.455
Neg logit contributions
 mess-0.487
 pic-0.402
aj-0.392
 wra-0.389
 slippery-0.386
Token way
Token ID835
Feature activation-2.90e-01

Pos logit contributions
aroo+0.127
ter+0.116
crow+0.103
 member+0.099
 pilot+0.099
Neg logit contributions
 mess-0.103
aj-0.097
onna-0.089
 wra-0.088
liv-0.086
Token.
Token ID13
Feature activation+3.71e-02

Pos logit contributions
 mess+0.562
aj+0.497
 wra+0.470
onna+0.468
 fault+0.456
Neg logit contributions
aroo-0.586
ter-0.519
crow-0.462
apped-0.455
itte-0.451
Token As
Token ID1081
Feature activation-1.42e-02

Pos logit contributions
aroo+0.068
ter+0.061
apped+0.060
crow+0.057
 Everyone+0.057
Neg logit contributions
 mess-0.076
aj-0.066
 wra-0.065
 fault-0.062
onna-0.061
Token But
Token ID887
Feature activation-3.57e-04

Pos logit contributions
aroo+0.083
ter+0.070
 member+0.066
itte+0.064
 pilot+0.063
Neg logit contributions
 mess-0.063
aj-0.062
onna-0.058
 fault-0.058
icol-0.057
Token you
Token ID345
Feature activation+6.58e-02

Pos logit contributions
 mess+0.001
aj+0.001
 wra+0.001
onna+0.001
 pic+0.001
Neg logit contributions
aroo-0.001
ter-0.001
crow-0.001
apped-0.001
itte-0.001
Token have
Token ID423
Feature activation+1.41e-01

Pos logit contributions
aroo+0.110
ter+0.096
crow+0.083
apped+0.081
itte+0.079
Neg logit contributions
 mess-0.150
aj-0.138
 wra-0.131
onna-0.129
 pic-0.126
Token to
Token ID284
Feature activation-6.00e-02

Pos logit contributions
aroo+0.309
ter+0.282
crow+0.253
apped+0.251
 member+0.246
Neg logit contributions
 mess-0.242
aj-0.214
 wra-0.200
icol-0.195
liv-0.194
Token be
Token ID307
Feature activation+3.50e-03

Pos logit contributions
 mess+0.126
aj+0.117
 wra+0.110
onna+0.107
 pic+0.106
Neg logit contributions
aroo-0.108
ter-0.097
crow-0.085
apped-0.081
 member-0.081
Token patient
Token ID5827
Feature activation+1.09e-01

Pos logit contributions
aroo+0.008
ter+0.007
crow+0.006
apped+0.006
itte+0.006
Neg logit contributions
 mess-0.007
aj-0.006
 wra-0.005
onna-0.005
 slippery-0.005
Token.
Token ID13
Feature activation+4.11e-02

Pos logit contributions
aroo+0.228
ter+0.206
crow+0.183
apped+0.181
itte+0.176
Neg logit contributions
 mess-0.200
aj-0.178
 wra-0.168
onna-0.166
 pic-0.162
Token Good
Token ID4599
Feature activation+1.46e-02

Pos logit contributions
aroo+0.087
ter+0.074
 member+0.068
 pilot+0.066
apped+0.065
Neg logit contributions
 mess-0.072
aj-0.071
onna-0.066
 fault-0.064
 pic-0.064
Token things
Token ID1243
Feature activation+1.43e-02

Pos logit contributions
aroo+0.032
ter+0.027
itte+0.025
ovable+0.025
crow+0.024
Neg logit contributions
 mess-0.029
aj-0.025
 fuss-0.023
 wra-0.022
 mistake-0.022
Token take
Token ID1011
Feature activation+9.74e-02

Pos logit contributions
aroo+0.030
ter+0.026
crow+0.024
apped+0.024
itte+0.023
Neg logit contributions
 mess-0.029
 wra-0.024
aj-0.024
onna-0.023
 slippery-0.022
Token time
Token ID640
Feature activation-2.36e-01

Pos logit contributions
aroo+0.204
ter+0.183
crow+0.163
apped+0.161
itte+0.158
Neg logit contributions
 mess-0.183
aj-0.162
 wra-0.153
onna-0.151
 pic-0.147
Token a
Token ID257
Feature activation-1.69e-03

Pos logit contributions
 mess+0.407
aj+0.385
aman+0.362
onna+0.360
icol+0.358
Neg logit contributions
aroo-0.302
ter-0.271
apped-0.230
itte-0.219
 member-0.217
Token time
Token ID640
Feature activation-2.76e-01

Pos logit contributions
 wra+0.003
 mess+0.003
 equations+0.003
drivers+0.003
onna+0.003
Neg logit contributions
aroo-0.003
 of-0.003
 member-0.002
 named-0.002
itte-0.002
Token,
Token ID11
Feature activation-1.02e-02

Pos logit contributions
 mess+0.625
aj+0.561
 wra+0.531
onna+0.530
icol+0.523
Neg logit contributions
aroo-0.489
ter-0.421
apped-0.369
itte-0.354
 member-0.347
Token there
Token ID612
Feature activation+7.50e-02

Pos logit contributions
 mess+0.020
aj+0.018
 wra+0.017
onna+0.017
 slippery+0.016
Neg logit contributions
aroo-0.020
ter-0.017
crow-0.016
apped-0.015
 member-0.015
Token was
Token ID373
Feature activation-5.04e-02

Pos logit contributions
aroo+0.143
ter+0.126
apped+0.112
crow+0.111
 member+0.108
Neg logit contributions
 mess-0.152
aj-0.147
onna-0.132
 wra-0.132
icol-0.131
Token a
Token ID257
Feature activation+3.59e-04

Pos logit contributions
 mess+0.106
aj+0.097
onna+0.092
 wra+0.090
 fuss+0.087
Neg logit contributions
aroo-0.094
ter-0.084
apped-0.073
itte-0.071
crow-0.071
Token little
Token ID1310
Feature activation+7.90e-02

Pos logit contributions
aroo+0.001
ter+0.001
crow+0.001
apped+0.001
ovable+0.001
Neg logit contributions
 mess-0.001
aj-0.001
 wra-0.001
 pic-0.001
 slippery-0.001
Token girl
Token ID2576
Feature activation+2.48e-01

Pos logit contributions
aroo+0.120
ovable+0.102
onel+0.100
apped+0.099
ter+0.099
Neg logit contributions
 mess-0.189
 wra-0.160
 scar-0.157
 slippery-0.155
aj-0.150
Token named
Token ID3706
Feature activation+3.33e-01

Pos logit contributions
aroo+0.353
ter+0.306
apped+0.259
onel+0.258
folk+0.256
Neg logit contributions
 mess-0.614
aj-0.549
 pic-0.524
onna-0.516
 wra-0.515
Token Lily
Token ID20037
Feature activation-4.82e-02

Pos logit contributions
aroo+0.688
ter+0.630
apped+0.554
crow+0.549
itte+0.545
Neg logit contributions
 mess-0.609
aj-0.550
 wra-0.521
 fault-0.508
onna-0.508
Token.
Token ID13
Feature activation+2.88e-02

Pos logit contributions
 mess+0.094
aj+0.082
 wra+0.078
onna+0.077
 pic+0.075
Neg logit contributions
aroo-0.099
ter-0.088
crow-0.079
apped-0.078
itte-0.076
Token friends
Token ID2460
Feature activation+3.10e-01

Pos logit contributions
 mess+0.075
aj+0.066
 fault+0.061
onna+0.060
 wra+0.060
Neg logit contributions
aroo-0.085
ter-0.075
crow-0.069
itte-0.067
apped-0.066
Token or
Token ID393
Feature activation-5.04e-02

Pos logit contributions
aroo+0.705
ter+0.608
crow+0.569
apped+0.563
itte+0.558
Neg logit contributions
 mess-0.561
aj-0.453
onna-0.433
 wra-0.432
 fuss-0.422
Token playing
Token ID2712
Feature activation+6.13e-02

Pos logit contributions
 mess+0.095
aj+0.081
 wra+0.077
onna+0.076
 pic+0.075
Neg logit contributions
aroo-0.108
ter-0.095
apped-0.086
crow-0.086
itte-0.084
Token games
Token ID1830
Feature activation-2.77e-02

Pos logit contributions
aroo+0.125
ter+0.109
apped+0.098
itte+0.098
crow+0.097
Neg logit contributions
 mess-0.121
aj-0.101
 wra-0.100
 pic-0.097
 slippery-0.094
Token.
Token ID13
Feature activation+5.02e-02

Pos logit contributions
 mess+0.052
aj+0.045
 wra+0.043
onna+0.042
 pic+0.041
Neg logit contributions
aroo-0.059
ter-0.052
crow-0.047
apped-0.047
itte-0.046
Token She
Token ID1375
Feature activation+2.37e-02

Pos logit contributions
aroo+0.109
ter+0.098
crow+0.087
apped+0.086
itte+0.084
Neg logit contributions
 mess-0.090
aj-0.081
 wra-0.075
onna-0.075
 pic-0.073
Token would
Token ID561
Feature activation-7.65e-02

Pos logit contributions
aroo+0.053
ter+0.048
crow+0.043
apped+0.042
itte+0.041
Neg logit contributions
 mess-0.042
aj-0.036
 wra-0.034
onna-0.033
 pic-0.033
Token always
Token ID1464
Feature activation+5.17e-02

Pos logit contributions
 mess+0.147
aj+0.126
onna+0.123
 wra+0.120
 pic+0.118
Neg logit contributions
aroo-0.158
ter-0.142
crow-0.126
apped-0.124
itte-0.122
Token do
Token ID466
Feature activation+9.28e-02

Pos logit contributions
aroo+0.102
ter+0.091
crow+0.080
apped+0.079
itte+0.078
Neg logit contributions
 mess-0.104
aj-0.092
 wra-0.087
onna-0.086
 pic-0.084
Token her
Token ID607
Feature activation-3.64e-02

Pos logit contributions
aroo+0.184
ter+0.164
crow+0.152
 member+0.147
apped+0.146
Neg logit contributions
 mess-0.165
aj-0.160
 wra-0.147
onna-0.146
icol-0.142
Token best
Token ID1266
Feature activation+8.35e-02

Pos logit contributions
 mess+0.075
aj+0.068
 wra+0.065
onna+0.064
aman+0.062
Neg logit contributions
aroo-0.069
ter-0.061
crow-0.053
apped-0.053
 member-0.052
Token okay
Token ID8788
Feature activation+1.49e-02

Pos logit contributions
aroo+0.017
ter+0.015
crow+0.014
apped+0.014
itte+0.013
Neg logit contributions
 mess-0.017
aj-0.015
 wra-0.014
 slippery-0.013
onna-0.013
Token.
Token ID13
Feature activation+4.72e-02

Pos logit contributions
aroo+0.031
ter+0.027
crow+0.025
apped+0.024
itte+0.024
Neg logit contributions
 mess-0.029
aj-0.025
onna-0.024
 wra-0.024
 fault-0.023
Token He
Token ID679
Feature activation+3.46e-02

Pos logit contributions
aroo+0.085
ter+0.076
crow+0.068
apped+0.067
itte+0.064
Neg logit contributions
 mess-0.101
aj-0.090
 wra-0.087
onna-0.084
aman-0.084
Token does
Token ID857
Feature activation-2.15e-01

Pos logit contributions
aroo+0.075
ter+0.067
crow+0.063
itte+0.061
ovable+0.060
Neg logit contributions
 mess-0.056
aj-0.051
onna-0.048
aman-0.048
liv-0.048
Token not
Token ID407
Feature activation-2.50e-01

Pos logit contributions
 mess+0.269
aj+0.263
aman+0.255
 wra+0.253
 pic+0.239
Neg logit contributions
aroo-0.538
ter-0.487
 pilot-0.439
 hero-0.429
crow-0.428
Token see
Token ID766
Feature activation+1.27e-01

Pos logit contributions
 mess+0.486
onna+0.387
 slippery+0.381
 fuss+0.380
aj+0.371
Neg logit contributions
aroo-0.541
ter-0.480
apped-0.461
itte-0.433
ovable-0.417
Token the
Token ID262
Feature activation+7.09e-02

Pos logit contributions
aroo+0.274
ter+0.242
apped+0.219
crow+0.217
itte+0.216
Neg logit contributions
 mess-0.238
aj-0.204
 wra-0.195
onna-0.190
 fault-0.189
Token rock
Token ID3881
Feature activation+4.63e-02

Pos logit contributions
aroo+0.158
ter+0.143
crow+0.128
apped+0.126
itte+0.124
Neg logit contributions
 mess-0.123
aj-0.109
 wra-0.102
onna-0.101
 pic-0.098
Token.
Token ID13
Feature activation+4.90e-02

Pos logit contributions
aroo+0.097
ter+0.086
apped+0.080
itte+0.079
crow+0.078
Neg logit contributions
 mess-0.089
aj-0.072
 wra-0.072
onna-0.070
 slippery-0.070
Token The
Token ID383
Feature activation+2.18e-01

Pos logit contributions
aroo+0.090
ter+0.081
crow+0.072
apped+0.071
itte+0.068
Neg logit contributions
 mess-0.104
aj-0.092
 wra-0.089
onna-0.086
aman-0.085
Token rock
Token ID3881
Feature activation+4.65e-02

Pos logit contributions
aroo+0.481
ter+0.435
crow+0.388
apped+0.388
itte+0.382
Neg logit contributions
 mess-0.396
aj-0.333
 wra-0.321
onna-0.309
 slippery-0.309
Token voice
Token ID3809
Feature activation+1.61e-01

Pos logit contributions
 mess+0.065
 wra+0.053
onna+0.052
aj+0.052
 fuss+0.052
Neg logit contributions
aroo-0.046
ter-0.040
apped-0.036
itte-0.035
ovable-0.035
Token calling
Token ID4585
Feature activation-1.42e-01

Pos logit contributions
aroo+0.365
ter+0.327
apped+0.312
itte+0.300
ovable+0.293
Neg logit contributions
 mess-0.268
aj-0.213
 wra-0.206
 fuss-0.203
aman-0.202
Token his
Token ID465
Feature activation-6.57e-02

Pos logit contributions
 mess+0.249
onna+0.208
 wra+0.206
aj+0.205
icol+0.198
Neg logit contributions
aroo-0.304
ter-0.269
itte-0.243
apped-0.240
 member-0.235
Token name
Token ID1438
Feature activation+1.21e-01

Pos logit contributions
 mess+0.139
aj+0.126
 fault+0.116
onna+0.115
 wra+0.114
Neg logit contributions
aroo-0.126
ter-0.110
crow-0.098
apped-0.094
itte-0.091
Token.
Token ID13
Feature activation+1.85e-02

Pos logit contributions
aroo+0.271
ter+0.238
itte+0.218
crow+0.216
apped+0.215
Neg logit contributions
 mess-0.200
 wra-0.167
aj-0.163
aman-0.157
onna-0.156
Token
Token ID220
Feature activation-1.36e-01

Pos logit contributions
aroo+0.037
ter+0.033
crow+0.029
apped+0.029
itte+0.028
Neg logit contributions
 mess-0.037
aj-0.033
 wra-0.031
onna-0.031
 pic-0.030
Token
Token ID198
Feature activation-4.12e-04

Pos logit contributions
 mess+0.273
aj+0.242
 wra+0.232
onna+0.225
 pic+0.223
Neg logit contributions
aroo-0.266
ter-0.236
crow-0.210
apped-0.209
itte-0.204
Token
Token ID198
Feature activation+6.23e-04

Pos logit contributions
 mess+0.001
aj+0.001
 wra+0.001
 fault+0.001
onna+0.001
Neg logit contributions
aroo-0.001
ter-0.001
crow-0.001
apped-0.001
itte-0.001
Token'
Token ID6
Feature activation+2.44e-01

Pos logit contributions
aroo+0.001
ter+0.001
apped+0.001
crow+0.001
ovable+0.001
Neg logit contributions
 mess-0.001
aj-0.001
 wra-0.001
onna-0.001
liv-0.001
TokenSet
Token ID7248
Feature activation+3.41e-02

Pos logit contributions
aroo+0.506
ovable+0.485
onel+0.477
ter+0.471
apped+0.460
Neg logit contributions
aman-0.286
liv-0.284
Lisa-0.284
aj-0.272
 mess-0.257
Tokeno
Token ID78
Feature activation-1.05e-02

Pos logit contributions
aroo+0.071
ter+0.064
crow+0.059
apped+0.059
ovable+0.057
Neg logit contributions
 mess-0.057
aj-0.055
onna-0.050
 pays-0.048
die-0.048
Token called
Token ID1444
Feature activation+4.25e-02

Pos logit contributions
aroo+0.337
ter+0.294
folk+0.264
onel+0.262
apped+0.256
Neg logit contributions
 mess-0.519
aj-0.455
ine-0.443
 pic-0.436
onna-0.430
Token Ell
Token ID7122
Feature activation+2.46e-02

Pos logit contributions
aroo+0.106
ter+0.097
crow+0.088
apped+0.087
 pilot+0.086
Neg logit contributions
 mess-0.060
aj-0.053
onna-0.050
 wra-0.048
 pic-0.047
Tokena
Token ID64
Feature activation+4.50e-02

Pos logit contributions
aroo+0.056
crow+0.048
ter+0.047
 Everyone+0.044
ive+0.042
Neg logit contributions
 mess-0.041
aj-0.038
onna-0.034
 fault-0.033
 wra-0.033
Token.
Token ID13
Feature activation-1.53e-02

Pos logit contributions
aroo+0.118
ter+0.106
crow+0.103
folk+0.102
onel+0.102
Neg logit contributions
 mess-0.055
aj-0.046
 mistake-0.042
ine-0.040
 wra-0.038
Token She
Token ID1375
Feature activation-2.81e-02

Pos logit contributions
 mess+0.024
onna+0.022
aj+0.022
 wra+0.021
 pic+0.019
Neg logit contributions
aroo-0.036
ter-0.031
crow-0.029
 member-0.028
 sailors-0.028
Token was
Token ID373
Feature activation-7.00e-02

Pos logit contributions
 mess+0.059
aj+0.052
 wra+0.050
onna+0.050
 pic+0.049
Neg logit contributions
aroo-0.052
ter-0.047
crow-0.041
apped-0.040
ive-0.039
Token three
Token ID1115
Feature activation+2.40e-01

Pos logit contributions
 mess+0.102
 slippery+0.081
aj+0.079
 pic+0.079
 wra+0.078
Neg logit contributions
aroo-0.172
ter-0.149
crow-0.148
ive-0.145
 member-0.139
Token and
Token ID290
Feature activation-1.22e-01

Pos logit contributions
aroo+0.586
ter+0.537
apped+0.479
crow+0.477
itte+0.463
Neg logit contributions
 mess-0.389
aj-0.330
 wra-0.310
 pic-0.294
 slippery-0.285
Token liked
Token ID8288
Feature activation+8.40e-02

Pos logit contributions
 mess+0.237
aj+0.207
 wra+0.203
liv+0.202
 slippery+0.198
Neg logit contributions
aroo-0.242
ter-0.211
crow-0.203
ive-0.199
 member-0.185
Token to
Token ID284
Feature activation-1.08e-01

Pos logit contributions
aroo+0.184
ter+0.167
crow+0.146
apped+0.146
 member+0.143
Neg logit contributions
 mess-0.146
aj-0.133
 wra-0.121
onna-0.120
liv-0.119
Token do
Token ID466
Feature activation+4.39e-02

Pos logit contributions
 mess+0.200
onna+0.163
 pic+0.163
aj+0.160
 slippery+0.156
Neg logit contributions
aroo-0.238
ter-0.218
apped-0.196
crow-0.193
itte-0.188
Token<|endoftext|>
Token ID50256
Feature activation+6.82e-02

Pos logit contributions
 mess+0.254
aj+0.245
onna+0.231
aman+0.227
liv+0.224
Neg logit contributions
aroo-0.505
ter-0.469
 pilot-0.412
 sailors-0.409
crow-0.407
TokenOnce
Token ID7454
Feature activation-1.29e-01

Pos logit contributions
aroo+0.145
ter+0.129
itte+0.112
crow+0.112
apped+0.112
Neg logit contributions
 mess-0.122
aj-0.109
 wra-0.103
 pic-0.102
onna-0.102
Token upon
Token ID2402
Feature activation-1.85e-01

Pos logit contributions
 mess+0.170
aj+0.143
 wra+0.131
onna+0.128
 pic+0.123
Neg logit contributions
aroo-0.340
ter-0.312
crow-0.286
apped-0.283
itte-0.279
Token a
Token ID257
Feature activation-6.85e-03

Pos logit contributions
 mess+0.407
aj+0.382
onna+0.355
 wra+0.353
aman+0.352
Neg logit contributions
aroo-0.316
ter-0.282
apped-0.236
itte-0.231
crow-0.231
Token time
Token ID640
Feature activation-2.79e-01

Pos logit contributions
 mess+0.011
 wra+0.011
onna+0.010
aj+0.010
idents+0.010
Neg logit contributions
aroo-0.013
 member-0.011
itte-0.011
 named-0.011
 hero-0.011
Token,
Token ID11
Feature activation-2.05e-02

Pos logit contributions
 mess+0.616
aj+0.550
 wra+0.523
onna+0.512
icol+0.507
Neg logit contributions
aroo-0.518
ter-0.448
apped-0.389
crow-0.381
itte-0.377
Token there
Token ID612
Feature activation+7.15e-02

Pos logit contributions
 mess+0.038
aj+0.034
 slippery+0.032
onna+0.032
 wra+0.032
Neg logit contributions
aroo-0.041
ter-0.036
crow-0.034
 member-0.032
ive-0.032
Token was
Token ID373
Feature activation-5.64e-02

Pos logit contributions
aroo+0.147
ter+0.131
crow+0.117
apped+0.116
itte+0.112
Neg logit contributions
 mess-0.136
aj-0.128
 wra-0.116
onna-0.115
icol-0.113
Token a
Token ID257
Feature activation-4.16e-03

Pos logit contributions
 mess+0.113
aj+0.102
onna+0.095
 wra+0.095
 fault+0.092
Neg logit contributions
aroo-0.111
ter-0.100
crow-0.087
apped-0.087
itte-0.085
Token little
Token ID1310
Feature activation+7.24e-02

Pos logit contributions
 mess+0.008
aj+0.006
 wra+0.006
 slippery+0.006
 pic+0.006
Neg logit contributions
aroo-0.009
ter-0.008
crow-0.007
apped-0.007
ovable-0.007
Token girl
Token ID2576
Feature activation+2.43e-01

Pos logit contributions
aroo+0.111
ovable+0.098
onel+0.097
apped+0.096
rest+0.092
Neg logit contributions
 mess-0.169
 wra-0.142
 slippery-0.140
 scar-0.140
 snowball-0.135
Token little
Token ID1310
Feature activation+3.07e-02

Pos logit contributions
 mess+0.092
aj+0.074
 wra+0.072
 slippery+0.071
 pic+0.070
Neg logit contributions
aroo-0.102
ter-0.091
crow-0.088
apped-0.086
ovable-0.082
Token girl
Token ID2576
Feature activation+2.01e-01

Pos logit contributions
aroo+0.047
ovable+0.041
onel+0.041
apped+0.040
rest+0.039
Neg logit contributions
 mess-0.071
 wra-0.060
 scar-0.059
 slippery-0.059
 snowball-0.057
Token named
Token ID3706
Feature activation+2.89e-01

Pos logit contributions
aroo+0.306
ter+0.267
onel+0.238
folk+0.237
apped+0.232
Neg logit contributions
 mess-0.465
aj-0.411
ine-0.404
 pic-0.399
onna-0.390
Token Lily
Token ID20037
Feature activation-9.56e-02

Pos logit contributions
aroo+0.627
ter+0.568
crow+0.506
apped+0.503
itte+0.497
Neg logit contributions
 mess-0.518
aj-0.460
 wra-0.435
onna-0.423
 fault-0.417
Token.
Token ID13
Feature activation-3.61e-02

Pos logit contributions
 mess+0.182
aj+0.157
 wra+0.150
onna+0.148
 pic+0.143
Neg logit contributions
aroo-0.201
ter-0.177
crow-0.159
apped-0.158
itte-0.154
Token She
Token ID1375
Feature activation-4.21e-02

Pos logit contributions
 mess+0.073
aj+0.065
 wra+0.062
onna+0.060
 fault+0.060
Neg logit contributions
aroo-0.070
ter-0.063
apped-0.056
crow-0.055
itte-0.053
Token loved
Token ID6151
Feature activation+9.84e-02

Pos logit contributions
 mess+0.091
aj+0.080
 wra+0.077
onna+0.076
 pic+0.074
Neg logit contributions
aroo-0.077
ter-0.069
crow-0.061
apped-0.059
ive-0.057
Token to
Token ID284
Feature activation-1.30e-01

Pos logit contributions
aroo+0.214
ter+0.193
crow+0.170
apped+0.168
 member+0.166
Neg logit contributions
 mess-0.172
aj-0.157
 wra-0.143
onna-0.142
liv-0.141
Token eat
Token ID4483
Feature activation-4.75e-02

Pos logit contributions
 mess+0.247
aj+0.212
onna+0.210
 pic+0.209
 wra+0.199
Neg logit contributions
aroo-0.273
ter-0.253
crow-0.226
apped-0.223
itte-0.217
Token candy
Token ID18550
Feature activation-2.08e-01

Pos logit contributions
aj+0.081
 fault+0.081
 mess+0.081
liv+0.079
onna+0.074
Neg logit contributions
aroo-0.094
ter-0.089
crow-0.075
 ot-0.071
 of-0.071
Token and
Token ID290
Feature activation-1.36e-01

Pos logit contributions
 mess+0.458
aj+0.439
 fault+0.415
icol+0.413
onna+0.404
Neg logit contributions
aroo-0.346
ter-0.317
 member-0.261
apped-0.256
crow-0.256
Token,
Token ID11
Feature activation-2.39e-02

Pos logit contributions
 mess+0.216
onna+0.214
aj+0.193
gh+0.189
 wra+0.183
Neg logit contributions
aroo-0.434
ter-0.415
itte-0.376
crow-0.373
 member-0.371
Token "
Token ID366
Feature activation+4.28e-03

Pos logit contributions
 mess+0.032
onna+0.030
aj+0.028
 wra+0.027
 pays+0.024
Neg logit contributions
aroo-0.058
ter-0.053
crow-0.051
ovable-0.051
itte-0.050
TokenMy
Token ID3666
Feature activation+1.66e-01

Pos logit contributions
aroo+0.008
ter+0.007
ovable+0.006
itte+0.006
onel+0.006
Neg logit contributions
 mess-0.008
 wra-0.007
aj-0.007
liv-0.007
die-0.007
Token name
Token ID1438
Feature activation+1.25e-01

Pos logit contributions
aroo+0.314
ter+0.275
itte+0.249
crow+0.247
ovable+0.236
Neg logit contributions
 mess-0.347
aj-0.316
 fault-0.298
 wra-0.287
 pic-0.280
Token is
Token ID318
Feature activation-1.72e-01

Pos logit contributions
aroo+0.356
ter+0.327
itte+0.308
crow+0.304
apped+0.303
Neg logit contributions
 mess-0.135
 wra-0.100
aj-0.098
onna-0.088
 pays-0.088
Token Tom
Token ID4186
Feature activation-1.98e-01

Pos logit contributions
 mess+0.211
aj+0.188
 pic+0.177
 wra+0.165
liv+0.165
Neg logit contributions
aroo-0.439
ter-0.399
apped-0.377
crow-0.370
ive-0.369
Token.
Token ID13
Feature activation+4.10e-03

Pos logit contributions
 mess+0.362
aj+0.321
 wra+0.300
onna+0.294
 fault+0.289
Neg logit contributions
aroo-0.426
ter-0.374
crow-0.342
apped-0.337
itte-0.335
Token And
Token ID843
Feature activation+1.01e-01

Pos logit contributions
aroo+0.009
ter+0.008
 member+0.007
itte+0.007
onel+0.007
Neg logit contributions
onna-0.006
aj-0.006
 mess-0.006
icol-0.006
 fault-0.006
Token this
Token ID428
Feature activation-6.01e-02

Pos logit contributions
aroo+0.208
ter+0.189
apped+0.178
crow+0.174
itte+0.171
Neg logit contributions
 mess-0.178
aj-0.159
 wra-0.154
 pic-0.151
onna-0.150
Token is
Token ID318
Feature activation-1.70e-01

Pos logit contributions
 mess+0.085
aj+0.074
 wra+0.068
onna+0.067
 pic+0.063
Neg logit contributions
aroo-0.153
ter-0.139
crow-0.128
apped-0.125
 member-0.123
Token Spot
Token ID15899
Feature activation-3.60e-02

Pos logit contributions
 mess+0.288
aj+0.254
 wra+0.238
onna+0.233
 pic+0.225
Neg logit contributions
aroo-0.384
ter-0.349
crow-0.312
apped-0.307
itte-0.303
Token.
Token ID13
Feature activation+1.29e-02

Pos logit contributions
 mess+0.219
aj+0.208
icol+0.196
 wra+0.194
liv+0.192
Neg logit contributions
aroo-0.179
ter-0.159
apped-0.139
 member-0.138
crow-0.136
Token They
Token ID1119
Feature activation-3.19e-02

Pos logit contributions
aroo+0.029
ter+0.026
crow+0.023
itte+0.022
 member+0.022
Neg logit contributions
 mess-0.023
aj-0.021
onna-0.019
 wra-0.019
 pic-0.019
Token had
Token ID550
Feature activation-2.95e-02

Pos logit contributions
 mess+0.055
 wra+0.046
onna+0.043
aj+0.042
 slippery+0.041
Neg logit contributions
aroo-0.077
ter-0.068
crow-0.065
ive-0.061
apped-0.060
Token completed
Token ID5668
Feature activation-8.57e-02

Pos logit contributions
 mess+0.052
aj+0.049
aman+0.045
icol+0.045
liv+0.044
Neg logit contributions
aroo-0.058
ter-0.055
apped-0.049
 member-0.048
 sailors-0.047
Token their
Token ID511
Feature activation-1.25e-01

Pos logit contributions
aj+0.176
 mess+0.173
liv+0.165
icol+0.163
array+0.158
Neg logit contributions
aroo-0.139
ter-0.130
apped-0.109
 sailors-0.108
 member-0.107
Token fun
Token ID1257
Feature activation-3.66e-01

Pos logit contributions
 mess+0.188
aj+0.175
liv+0.163
onna+0.162
aman+0.160
Neg logit contributions
aroo-0.281
ter-0.249
 member-0.230
apped-0.227
crow-0.224
Token.
Token ID13
Feature activation+1.65e-02

Pos logit contributions
 mess+0.786
aj+0.746
 wra+0.696
onna+0.695
liv+0.692
Neg logit contributions
aroo-0.627
ter-0.541
apped-0.476
 member-0.471
crow-0.466
Token They
Token ID1119
Feature activation-2.86e-02

Pos logit contributions
aroo+0.036
ter+0.032
crow+0.029
apped+0.028
itte+0.028
Neg logit contributions
 mess-0.030
aj-0.027
onna-0.025
 wra-0.025
 pic-0.025
Token smiled
Token ID13541
Feature activation-2.56e-02

Pos logit contributions
 mess+0.050
 wra+0.041
onna+0.039
aj+0.038
 slippery+0.037
Neg logit contributions
aroo-0.069
ter-0.061
crow-0.058
ive-0.054
apped-0.054
Token at
Token ID379
Feature activation+1.45e-01

Pos logit contributions
 mess+0.039
aj+0.034
 fault+0.030
 fun+0.029
 wra+0.028
Neg logit contributions
aroo-0.067
crow-0.056
ter-0.055
itte-0.052
aro-0.052
Token each
Token ID1123
Feature activation+7.95e-02

Pos logit contributions
aroo+0.329
ter+0.288
ive+0.274
crow+0.271
itte+0.264
Neg logit contributions
 mess-0.252
aj-0.211
 wra-0.196
 pic-0.195
 mistake-0.194
Token.
Token ID13
Feature activation+5.33e-02

Pos logit contributions
aroo+0.140
ter+0.126
crow+0.111
apped+0.110
itte+0.107
Neg logit contributions
 mess-0.132
aj-0.118
 wra-0.111
onna-0.110
 pic-0.106
Token They
Token ID1119
Feature activation+7.95e-03

Pos logit contributions
aroo+0.105
ter+0.093
crow+0.081
itte+0.078
apped+0.078
Neg logit contributions
 mess-0.107
aj-0.097
onna-0.091
 wra-0.091
 pic-0.091
Token had
Token ID550
Feature activation-2.45e-03

Pos logit contributions
aroo+0.018
ter+0.016
crow+0.015
ive+0.014
itte+0.014
Neg logit contributions
 mess-0.014
 wra-0.011
 scar-0.011
 slippery-0.011
 pic-0.011
Token lots
Token ID6041
Feature activation-6.21e-02

Pos logit contributions
 mess+0.005
aj+0.005
aman+0.005
onna+0.005
 wra+0.005
Neg logit contributions
aroo-0.004
ter-0.004
apped-0.003
 sailors-0.003
 member-0.003
Token of
Token ID286
Feature activation+3.60e-01

Pos logit contributions
 mess+0.188
aj+0.173
 pic+0.169
 wra+0.168
onna+0.165
Neg logit contributions
aroo-0.060
ter-0.049
crow-0.033
apped-0.032
itte-0.031
Token fun
Token ID1257
Feature activation-3.43e-01

Pos logit contributions
aroo+1.039
ter+0.946
 sailors+0.920
 member+0.906
ovable+0.901
Neg logit contributions
 mess-0.308
onna-0.272
aj-0.264
aman-0.256
liv-0.240
Token and
Token ID290
Feature activation-6.19e-02

Pos logit contributions
 mess+0.676
aj+0.620
onna+0.583
 wra+0.580
aman+0.561
Neg logit contributions
aroo-0.679
ter-0.602
apped-0.532
crow-0.523
 member-0.514
Token laughter
Token ID20263
Feature activation-1.02e-03

Pos logit contributions
 mess+0.125
aj+0.114
 wra+0.106
onna+0.106
aman+0.102
Neg logit contributions
aroo-0.120
ter-0.107
crow-0.094
apped-0.094
itte-0.090
Token and
Token ID290
Feature activation-6.09e-02

Pos logit contributions
 mess+0.002
 wra+0.001
aj+0.001
onna+0.001
 pic+0.001
Neg logit contributions
aroo-0.002
ter-0.002
crow-0.002
itte-0.002
ovable-0.002
Token never
Token ID1239
Feature activation-8.59e-02

Pos logit contributions
 mess+0.120
aj+0.106
 wra+0.102
 pic+0.099
onna+0.099
Neg logit contributions
aroo-0.122
ter-0.108
crow-0.097
apped-0.095
itte-0.094
Token stopped
Token ID5025
Feature activation+2.23e-02

Pos logit contributions
 mess+0.141
aj+0.114
onna+0.110
 wra+0.108
 slippery+0.104
Neg logit contributions
aroo-0.211
ter-0.179
crow-0.172
itte-0.170
apped-0.165
Token a
Token ID257
Feature activation-5.08e-02

Pos logit contributions
 mess+0.038
aj+0.031
 wra+0.031
 pic+0.030
onna+0.030
Neg logit contributions
aroo-0.055
ter-0.051
crow-0.045
apped-0.045
itte-0.045
Token big
Token ID1263
Feature activation-1.55e-01

Pos logit contributions
 mess+0.078
aj+0.073
 fault+0.065
liv+0.065
 wra+0.064
Neg logit contributions
aroo-0.122
ter-0.105
crow-0.097
 member-0.096
itte-0.094
Token bite
Token ID13197
Feature activation-2.56e-01

Pos logit contributions
 mess+0.201
aj+0.173
 wra+0.156
onna+0.152
 pic+0.147
Neg logit contributions
aroo-0.411
ter-0.376
crow-0.345
apped-0.341
itte-0.336
Token.
Token ID13
Feature activation-1.83e-02

Pos logit contributions
 mess+0.514
 wra+0.444
onna+0.436
aj+0.434
 slippery+0.421
Neg logit contributions
aroo-0.497
ter-0.432
itte-0.390
 pilot-0.386
apped-0.381
Token It
Token ID632
Feature activation+1.32e-02

Pos logit contributions
 mess+0.034
aj+0.030
onna+0.028
 wra+0.028
 pic+0.028
Neg logit contributions
aroo-0.038
ter-0.035
crow-0.031
apped-0.030
itte-0.030
Token tasted
Token ID29187
Feature activation-3.14e-01

Pos logit contributions
aroo+0.030
ter+0.027
crow+0.024
apped+0.024
itte+0.024
Neg logit contributions
 mess-0.022
aj-0.020
 fault-0.018
 wra-0.018
onna-0.018
Token so
Token ID523
Feature activation-1.37e-01

Pos logit contributions
aj+0.657
 mess+0.558
liv+0.545
icol+0.533
 fault+0.531
Neg logit contributions
ter-0.579
aroo-0.561
 of-0.472
crow-0.462
apped-0.434
Token delicious
Token ID12625
Feature activation-2.68e-01

Pos logit contributions
 mess+0.216
aj+0.207
 wra+0.180
aman+0.179
icol+0.178
Neg logit contributions
aroo-0.315
ter-0.287
apped-0.257
crow-0.256
itte-0.254
Token.
Token ID13
Feature activation-1.08e-02

Pos logit contributions
 mess+0.486
aj+0.477
 fault+0.409
onna+0.404
liv+0.401
Neg logit contributions
aroo-0.572
ter-0.538
crow-0.453
apped-0.432
ive-0.428
Token She
Token ID1375
Feature activation-3.37e-02

Pos logit contributions
 mess+0.019
aj+0.017
 wra+0.016
onna+0.016
 pic+0.015
Neg logit contributions
aroo-0.024
ter-0.021
crow-0.019
apped-0.019
itte-0.019
Token ate
Token ID15063
Feature activation-3.32e-02

Pos logit contributions
 mess+0.062
aj+0.055
 wra+0.051
onna+0.051
 pic+0.049
Neg logit contributions
aroo-0.072
ter-0.065
crow-0.058
apped-0.057
itte-0.056
Token make
Token ID787
Feature activation+7.43e-02

Pos logit contributions
 mess+0.109
aj+0.097
 wra+0.092
onna+0.091
 pic+0.089
Neg logit contributions
aroo-0.109
ter-0.097
crow-0.086
apped-0.084
itte-0.083
Token the
Token ID262
Feature activation+7.27e-02

Pos logit contributions
aroo+0.140
ter+0.131
apped+0.121
 pilot+0.114
 sailors+0.113
Neg logit contributions
aj-0.134
 mess-0.127
liv-0.124
onna-0.123
aman-0.120
Token puzzle
Token ID15027
Feature activation-1.48e-01

Pos logit contributions
aroo+0.184
ter+0.169
crow+0.154
apped+0.153
itte+0.150
Neg logit contributions
 mess-0.102
aj-0.089
 wra-0.081
onna-0.080
 pic-0.077
Token on
Token ID319
Feature activation+1.51e-01

Pos logit contributions
aj+0.281
 mess+0.266
liv+0.259
aman+0.257
icol+0.252
Neg logit contributions
aroo-0.272
ter-0.247
 member-0.223
apped-0.223
crow-0.215
Token the
Token ID262
Feature activation+7.29e-02

Pos logit contributions
aroo+0.206
ter+0.192
 person+0.167
 member+0.157
 pilot+0.152
Neg logit contributions
aj-0.379
liv-0.329
 mess-0.322
icol-0.317
onna-0.306
Token floor
Token ID4314
Feature activation-3.12e-01

Pos logit contributions
aroo+0.189
ter+0.176
 member+0.160
crow+0.158
apped+0.157
Neg logit contributions
 mess-0.088
aj-0.085
onna-0.074
liv-0.071
aman-0.070
Token of
Token ID286
Feature activation+3.64e-01

Pos logit contributions
 mess+0.579
aj+0.510
 wra+0.481
onna+0.476
 fault+0.461
Neg logit contributions
aroo-0.661
ter-0.590
crow-0.529
apped-0.518
itte-0.512
Token their
Token ID511
Feature activation-8.13e-02

Pos logit contributions
aroo+0.696
ter+0.636
apped+0.542
crow+0.541
 member+0.537
Neg logit contributions
 mess-0.697
aj-0.695
onna-0.616
icol-0.614
liv-0.606
Token room
Token ID2119
Feature activation-1.11e-01

Pos logit contributions
 mess+0.154
aj+0.135
 wra+0.128
onna+0.125
 pic+0.123
Neg logit contributions
aroo-0.172
ter-0.155
crow-0.138
apped-0.134
itte-0.133
Token.
Token ID13
Feature activation+5.72e-02

Pos logit contributions
 mess+0.217
aj+0.193
 wra+0.183
onna+0.180
 pic+0.175
Neg logit contributions
aroo-0.225
ter-0.201
crow-0.178
apped-0.175
itte-0.172
Token
Token ID198
Feature activation+3.15e-02

Pos logit contributions
aroo+0.086
apped+0.074
 Everyone+0.074
crow+0.070
ter+0.068
Neg logit contributions
 mess-0.136
aj-0.120
 wra-0.120
icol-0.116
onna-0.115
Token come
Token ID1282
Feature activation-1.19e-01

Pos logit contributions
 mess+0.257
onna+0.208
aj+0.202
 wra+0.202
 slippery+0.201
Neg logit contributions
aroo-0.289
ter-0.262
apped-0.240
crow-0.231
itte-0.226
Token home
Token ID1363
Feature activation+9.17e-02

Pos logit contributions
 mess+0.192
aj+0.166
onna+0.151
 wra+0.148
 slippery+0.145
Neg logit contributions
aroo-0.286
ter-0.252
crow-0.239
itte-0.228
apped-0.226
Token late
Token ID2739
Feature activation-2.19e-01

Pos logit contributions
aroo+0.194
ter+0.176
crow+0.155
apped+0.155
 member+0.151
Neg logit contributions
 mess-0.167
aj-0.149
 wra-0.141
onna-0.138
 pic-0.135
Token,
Token ID11
Feature activation-3.69e-02

Pos logit contributions
 mess+0.419
aj+0.374
 wra+0.352
onna+0.348
 pic+0.339
Neg logit contributions
aroo-0.450
ter-0.403
crow-0.358
apped-0.353
itte-0.345
Token and
Token ID290
Feature activation-1.26e-01

Pos logit contributions
 mess+0.064
 wra+0.052
aj+0.051
 pic+0.050
 slippery+0.049
Neg logit contributions
aroo-0.085
ter-0.075
crow-0.070
apped-0.070
itte-0.068
Token not
Token ID407
Feature activation-3.09e-01

Pos logit contributions
 mess+0.215
 wra+0.169
aj+0.167
onna+0.165
 slippery+0.163
Neg logit contributions
aroo-0.292
ter-0.260
crow-0.239
apped-0.237
itte-0.232
Token pay
Token ID1414
Feature activation-3.87e-02

Pos logit contributions
 mess+0.581
onna+0.460
 slippery+0.441
 fuss+0.441
 wra+0.438
Neg logit contributions
aroo-0.699
ter-0.592
crow-0.576
apped-0.572
itte-0.565
Token any
Token ID597
Feature activation-5.05e-02

Pos logit contributions
 mess+0.075
aj+0.068
onna+0.064
liv+0.064
 wra+0.063
Neg logit contributions
aroo-0.075
ter-0.068
crow-0.059
apped-0.058
 member-0.057
Token attention
Token ID3241
Feature activation-1.28e-01

Pos logit contributions
 mess+0.084
aj+0.067
 wra+0.066
 fault+0.063
 slippery+0.062
Neg logit contributions
aroo-0.124
ter-0.112
apped-0.104
itte-0.100
crow-0.099
Token to
Token ID284
Feature activation-1.15e-01

Pos logit contributions
 mess+0.212
aj+0.177
 wra+0.173
onna+0.171
 fault+0.168
Neg logit contributions
aroo-0.297
ter-0.266
crow-0.244
apped-0.241
itte-0.240
Token his
Token ID465
Feature activation-8.25e-02

Pos logit contributions
 mess+0.216
aj+0.175
 wra+0.175
onna+0.171
 pic+0.168
Neg logit contributions
aroo-0.249
ter-0.222
apped-0.209
crow-0.201
itte-0.201