TOP ACTIVATIONS
MAX = 11.934

of snow and watched it disinteg rate in his hands .
harder to push the wheel bar row . But her backyard
placed it onto the wheel bar row . John
light en up the wheel bar row . <|endoftext|> Once there
showed him how to hang it safely . Tommy vowed to
find a place to hang it . From then on ,
he drop everything to try it . <|endoftext|> Once upon a
together to examine it more closely . <|endoftext|> Once upon a
Lily for helping to wrap it . Lily smiled and felt
of her strength to hang it onto the jail .
cereal and set off mailing it to all its special destinations
quickly started to m unch it all up . He was
properly instead of just giving it to him . She decided
would be fun to try it too . She found some
he was determined to try it . So he began to
fire and started to pour it on the flames . It
looked and everyone gob bled it up . All agreed that
and then how to hang it . Jimmy
couldn 't wait to try it out .
you again how to balance it ." L ily

MIN ACTIVATIONS
MIN = -11.406

time , there were two cats . They were friends and
time , there were two cats that lived very far away
a time there were two cats who used to travel a
Once upon a time , a little boy named Pete and
time , there were two brothers who were very excited .
<|endoftext|> One day , a 3 year old girl named Sally
time , there were twin cats named Tom and Tim .
time , there was a 3 year old named Anna .
upon a time , two brothers named Jack and Daniel had
<|endoftext|> Ben and Sam were brothers who liked to play with
<|endoftext|> Tom and Sam were brothers who liked to gaze at
time , there were two brothers . The brothers were always
<|endoftext|> Tom and Sam were brothers who liked to play with
a Mom my and a 3 year old girl went for
One day , a little 3 year old girl arrived at
upon a time , two brothers Jack and Tom were walking
a time there were two brothers . One was called Paul
upon a time there lived a little girl and her daddy
upon a time there lived a brave boy called Dave .
<|endoftext|> Once there was a three - year - old who

INTERVAL 6.099 - 11.934
CONTAINS 0.611%

might break it or swallow it . You can play with
hurt it , or take it away from her . She
so excited and he opened it quickly . Inside the box
. She wiped his tears away and said , " Don
with you when you get better ." Mia says .

INTERVAL 0.264 - 6.099
CONTAINS 49.860%

asks his dad , " When will mom come home ?"
is rubbish !" Ben said . He felt sad and silly
the porch , jumping high and running low . She hopped
. They were excited to eat their sweet things later .
They laughed and smiled and enjoyed their time together . It

INTERVAL -5.571 - 0.264
CONTAINS 46.116%

Tim my . Tim my loved to draw with his pen
big box of clothes that Lily could use to make different
, it had a nut in its mouth . The squirrel
, there was a girl who lived in a big town
" Hey kid , do you want to sell that pill

INTERVAL -11.406 - -5.571
CONTAINS 3.413%

Once upon a time there was a family of wild animals
a time , there was a little girl named Lily .
She sees a book with a big lion on the cover
a farm when he heard a little voice say , "
upon a time there was a white tray . It was
Token of
Token ID286
Feature activation-3.62e+00

Pos logit contributions
orting+0.032
 Later+0.031
uing+0.030
 nicely+0.029
 Finally+0.028
Neg logit contributions
e-0.015
 seen-0.014
 coloured-0.013
 spe-0.013
 kinds-0.013
Token snow
Token ID6729
Feature activation+2.33e+00

Pos logit contributions
 seen+0.123
 supposed+0.122
e+0.120
 flown+0.116
vous+0.111
Neg logit contributions
 nicely-0.146
orting-0.145
 Later-0.137
uing-0.131
 patiently-0.122
Token and
Token ID290
Feature activation+2.40e-01

Pos logit contributions
orting+0.098
 Later+0.096
uing+0.091
 nicely+0.087
 Finally+0.082
Neg logit contributions
e-0.087
 seen-0.083
 coloured-0.080
 spe-0.079
 kinds-0.077
Token watched
Token ID7342
Feature activation+3.52e+00

Pos logit contributions
 Later+0.010
orting+0.009
 nicely+0.009
uing+0.009
 patiently+0.008
Neg logit contributions
e-0.009
 seen-0.008
 supposed-0.008
ymes-0.008
vous-0.008
Token it
Token ID340
Feature activation+6.59e+00

Pos logit contributions
orting+0.161
 Later+0.159
uing+0.149
 nicely+0.139
 Finally+0.135
Neg logit contributions
e-0.120
 seen-0.111
 coloured-0.110
 spe-0.108
 comet-0.104
Token disinteg
Token ID29930
Feature activation+9.99e+00

Pos logit contributions
orting+0.330
 Later+0.321
uing+0.302
 nicely+0.290
 Finally+0.280
Neg logit contributions
e-0.198
 seen-0.180
 coloured-0.179
 spe-0.172
 sorts-0.170
Tokenrate
Token ID4873
Feature activation+3.77e+00

Pos logit contributions
orting+0.447
 Later+0.390
uing+0.382
 Finally+0.373
der+0.371
Neg logit contributions
 coloured-0.336
e-0.327
 colored-0.313
 seen-0.289
 supposed-0.271
Token in
Token ID287
Feature activation+3.55e-01

Pos logit contributions
orting+0.165
 Later+0.163
uing+0.149
 Finally+0.141
 nicely+0.141
Neg logit contributions
e-0.137
 seen-0.130
 coloured-0.124
 ever-0.121
 wearing-0.117
Token his
Token ID465
Feature activation+4.87e+00

Pos logit contributions
 nicely+0.015
 Later+0.014
orting+0.013
uing+0.013
 patiently+0.013
Neg logit contributions
leaf-0.012
vous-0.012
e-0.012
 seen-0.012
ymes-0.011
Token hands
Token ID2832
Feature activation-6.55e-01

Pos logit contributions
 nicely+0.191
orting+0.189
 Later+0.188
uing+0.183
 Finally+0.166
Neg logit contributions
e-0.178
 seen-0.177
vous-0.169
leaf-0.168
 sorts-0.162
Token.
Token ID13
Feature activation+8.23e-01

Pos logit contributions
e+0.026
 seen+0.025
 coloured+0.024
 spe+0.023
 kinds+0.023
Neg logit contributions
orting-0.026
 Later-0.025
uing-0.024
 nicely-0.023
 Finally-0.021
Token harder
Token ID7069
Feature activation+4.56e+00

Pos logit contributions
e+0.052
 seen+0.047
 spe+0.047
vous+0.046
leaf+0.046
Neg logit contributions
orting-0.056
 Later-0.056
 nicely-0.055
uing-0.052
 Finally-0.049
Token to
Token ID284
Feature activation-6.16e-01

Pos logit contributions
orting+0.191
 Later+0.188
uing+0.177
 nicely+0.167
 Finally+0.159
Neg logit contributions
e-0.167
 seen-0.166
 coloured-0.158
 spe-0.154
 kinds-0.153
Token push
Token ID4574
Feature activation+1.68e+00

Pos logit contributions
e+0.020
 seen+0.020
 kinds+0.020
 sorts+0.019
leaf+0.019
Neg logit contributions
orting-0.026
 Later-0.025
 nicely-0.025
uing-0.025
 patiently-0.022
Token the
Token ID262
Feature activation+3.52e+00

Pos logit contributions
orting+0.075
 Later+0.074
uing+0.070
 nicely+0.065
 Finally+0.063
Neg logit contributions
e-0.060
 seen-0.056
 coloured-0.053
 spe-0.052
 kinds-0.051
Token wheel
Token ID7825
Feature activation+7.59e-01

Pos logit contributions
orting+0.144
 nicely+0.142
 Later+0.135
uing+0.132
arl+0.124
Neg logit contributions
e-0.128
 seen-0.124
 sorts-0.117
 kinds-0.113
 coloured-0.111
Tokenbar
Token ID5657
Feature activation+9.75e+00

Pos logit contributions
 Later+0.032
orting+0.032
uing+0.029
 nicely+0.027
 Finally+0.027
Neg logit contributions
e-0.028
 seen-0.027
 coloured-0.025
 ever-0.025
 spe-0.024
Tokenrow
Token ID808
Feature activation+3.19e+00

Pos logit contributions
 Later+0.347
orting+0.334
uing+0.303
 nicely+0.291
 Finally+0.284
Neg logit contributions
e-0.424
 seen-0.407
 coloured-0.385
 kinds-0.383
 ever-0.376
Token.
Token ID13
Feature activation+5.84e-02

Pos logit contributions
orting+0.143
 Later+0.142
uing+0.134
 nicely+0.120
 Finally+0.119
Neg logit contributions
e-0.102
 seen-0.101
 ever-0.096
 coloured-0.095
 shaped-0.092
Token But
Token ID887
Feature activation+1.85e+00

Pos logit contributions
orting+0.002
uing+0.002
 Later+0.002
 nicely+0.002
lated+0.001
Neg logit contributions
e-0.003
 seen-0.003
 wearing-0.002
 sorts-0.002
 spe-0.002
Token her
Token ID607
Feature activation+4.45e-01

Pos logit contributions
orting+0.078
 Later+0.075
uing+0.072
 nicely+0.066
 Finally+0.063
Neg logit contributions
e-0.070
 seen-0.067
 coloured-0.063
 spe-0.063
 wearing-0.061
Token backyard
Token ID24296
Feature activation-1.31e+00

Pos logit contributions
 Later+0.011
orting+0.011
 nicely+0.011
uing+0.011
 Finally+0.009
Neg logit contributions
e-0.023
 sorts-0.022
 kinds-0.022
leaf-0.022
 seen-0.022
Token placed
Token ID4624
Feature activation+1.80e+00

Pos logit contributions
 Later+0.008
orting+0.008
 nicely+0.007
uing+0.007
 patiently+0.007
Neg logit contributions
e-0.008
 supposed-0.007
vous-0.007
 spe-0.007
ymes-0.007
Token it
Token ID340
Feature activation+6.65e+00

Pos logit contributions
orting+0.082
 Later+0.081
uing+0.077
 nicely+0.073
 Finally+0.069
Neg logit contributions
e-0.061
 seen-0.059
 coloured-0.056
 spe-0.055
 kinds-0.054
Token onto
Token ID4291
Feature activation-2.84e-01

Pos logit contributions
orting+0.370
 Later+0.363
uing+0.350
 nicely+0.338
 Finally+0.322
Neg logit contributions
e-0.156
 seen-0.147
 coloured-0.136
 spe-0.133
 kinds-0.128
Token the
Token ID262
Feature activation+3.09e+00

Pos logit contributions
e+0.009
 seen+0.009
 spe+0.008
 coloured+0.008
leaf+0.008
Neg logit contributions
orting-0.013
 Later-0.013
 nicely-0.012
uing-0.012
 Finally-0.012
Token wheel
Token ID7825
Feature activation+1.18e+00

Pos logit contributions
orting+0.130
 Later+0.128
 nicely+0.124
uing+0.120
 Finally+0.111
Neg logit contributions
e-0.114
 seen-0.107
 sorts-0.103
 kinds-0.102
 coloured-0.100
Tokenbar
Token ID5657
Feature activation+9.47e+00

Pos logit contributions
orting+0.048
 Later+0.047
uing+0.044
 nicely+0.042
 Finally+0.040
Neg logit contributions
e-0.046
 seen-0.044
 coloured-0.042
 kinds-0.040
 spe-0.040
Tokenrow
Token ID808
Feature activation+3.21e+00

Pos logit contributions
 Later+0.431
orting+0.431
uing+0.403
 nicely+0.391
 Finally+0.375
Neg logit contributions
e-0.313
 seen-0.299
 coloured-0.284
 kinds-0.280
 colored-0.261
Token.
Token ID13
Feature activation+7.55e-01

Pos logit contributions
orting+0.145
 Later+0.142
uing+0.135
 nicely+0.123
 Finally+0.120
Neg logit contributions
e-0.106
 seen-0.105
 coloured-0.101
 shaped-0.097
 ever-0.094
Token
Token ID198
Feature activation+1.78e+00

Pos logit contributions
orting+0.028
uing+0.027
izes+0.022
 nicely+0.021
lated+0.021
Neg logit contributions
 seen-0.031
 wearing-0.031
e-0.031
 sorts-0.030
 kinds-0.029
Token
Token ID198
Feature activation+1.09e+00

Pos logit contributions
orting+0.064
uing+0.059
 Later+0.055
der+0.051
rier+0.049
Neg logit contributions
e-0.081
 seen-0.076
 sorts-0.072
 wearing-0.071
 coloured-0.070
TokenJohn
Token ID7554
Feature activation-4.29e-01

Pos logit contributions
orting+0.041
uing+0.037
 Later+0.036
 nicely+0.032
der+0.031
Neg logit contributions
e-0.047
 seen-0.044
 coloured-0.042
 sorts-0.042
vous-0.042
Token light
Token ID1657
Feature activation+3.91e+00

Pos logit contributions
orting+0.022
 Later+0.021
uing+0.020
 nicely+0.020
 Finally+0.018
Neg logit contributions
e-0.016
 seen-0.015
 spe-0.015
 coloured-0.015
 kinds-0.014
Tokenen
Token ID268
Feature activation+3.49e+00

Pos logit contributions
orting+0.167
 Later+0.162
uing+0.156
 nicely+0.150
 Finally+0.138
Neg logit contributions
e-0.142
 seen-0.138
 coloured-0.130
 spe-0.129
 kinds-0.126
Token up
Token ID510
Feature activation+4.62e+00

Pos logit contributions
orting+0.154
 Later+0.150
uing+0.144
 nicely+0.138
 Finally+0.128
Neg logit contributions
e-0.122
 seen-0.118
 coloured-0.111
 spe-0.110
 kinds-0.108
Token the
Token ID262
Feature activation+4.28e+00

Pos logit contributions
orting+0.179
 Later+0.174
 nicely+0.168
uing+0.165
 Finally+0.147
Neg logit contributions
e-0.184
 seen-0.173
leaf-0.169
vous-0.166
 spe-0.165
Token wheel
Token ID7825
Feature activation+1.98e+00

Pos logit contributions
orting+0.198
 Later+0.193
uing+0.183
 nicely+0.180
 Finally+0.167
Neg logit contributions
e-0.144
 seen-0.133
 coloured-0.125
 sorts-0.124
 kinds-0.123
Tokenbar
Token ID5657
Feature activation+9.41e+00

Pos logit contributions
orting+0.077
 Later+0.077
uing+0.072
 nicely+0.068
 Finally+0.065
Neg logit contributions
e-0.080
 seen-0.076
 coloured-0.073
 spe-0.072
 kinds-0.070
Tokenrow
Token ID808
Feature activation+3.40e+00

Pos logit contributions
 Later+0.448
orting+0.423
 nicely+0.399
uing+0.399
 Finally+0.397
Neg logit contributions
e-0.300
 coloured-0.277
 seen-0.276
 kinds-0.271
 colored-0.252
Token.
Token ID13
Feature activation+1.13e+00

Pos logit contributions
orting+0.146
 Later+0.146
uing+0.137
 nicely+0.127
 Finally+0.124
Neg logit contributions
e-0.120
 seen-0.115
 coloured-0.110
 spe-0.108
 kinds-0.106
Token<|endoftext|>
Token ID50256
Feature activation-3.61e+00

Pos logit contributions
orting+0.046
uing+0.045
izes+0.038
edy+0.036
rier+0.036
Neg logit contributions
e-0.046
 sorts-0.044
 wearing-0.044
 seen-0.042
 kinds-0.041
TokenOnce
Token ID7454
Feature activation-5.77e-01

Pos logit contributions
e+0.183
 seen+0.181
 spe+0.176
 coloured+0.176
 kinds+0.173
Neg logit contributions
orting-0.100
 Later-0.099
 nicely-0.085
uing-0.080
 Finally-0.079
Token there
Token ID612
Feature activation-8.96e-01

Pos logit contributions
 seen+0.018
 wearing+0.016
 ever+0.015
 happened+0.015
 sorts+0.015
Neg logit contributions
orting-0.030
uing-0.027
 Later-0.027
rier-0.025
der-0.024
Token showed
Token ID3751
Feature activation+3.09e+00

Pos logit contributions
orting+0.055
 Later+0.053
uing+0.051
 nicely+0.050
 Finally+0.046
Neg logit contributions
e-0.052
 seen-0.049
 spe-0.047
 coloured-0.047
 kinds-0.046
Token him
Token ID683
Feature activation+4.87e+00

Pos logit contributions
orting+0.124
 Later+0.123
uing+0.116
 nicely+0.105
 Finally+0.102
Neg logit contributions
e-0.121
 seen-0.118
 coloured-0.112
 kinds-0.109
 sorts-0.108
Token how
Token ID703
Feature activation+2.75e+00

Pos logit contributions
orting+0.191
 Later+0.186
uing+0.178
 nicely+0.172
 Finally+0.156
Neg logit contributions
e-0.193
 seen-0.187
 spe-0.179
 coloured-0.178
 comet-0.173
Token to
Token ID284
Feature activation+4.73e-01

Pos logit contributions
orting+0.125
 Later+0.123
uing+0.117
 nicely+0.112
 Finally+0.105
Neg logit contributions
e-0.093
 seen-0.090
 coloured-0.085
 spe-0.084
 kinds-0.082
Token hang
Token ID8181
Feature activation+5.24e+00

Pos logit contributions
orting+0.023
 Later+0.022
uing+0.022
 nicely+0.021
 Finally+0.019
Neg logit contributions
e-0.014
 seen-0.014
 coloured-0.013
 spe-0.013
 kinds-0.013
Token it
Token ID340
Feature activation+9.30e+00

Pos logit contributions
orting+0.235
 Later+0.229
uing+0.220
 nicely+0.212
 Finally+0.197
Neg logit contributions
e-0.179
 seen-0.173
 coloured-0.162
 spe-0.162
 kinds-0.158
Token safely
Token ID11512
Feature activation+6.21e+00

Pos logit contributions
orting+0.436
 Later+0.427
uing+0.408
 nicely+0.391
 Finally+0.370
Neg logit contributions
e-0.299
 seen-0.286
 coloured-0.271
 spe-0.267
 kinds-0.261
Token.
Token ID13
Feature activation+1.52e+00

Pos logit contributions
orting+0.238
 Later+0.229
uing+0.220
 nicely+0.215
 Finally+0.192
Neg logit contributions
e-0.253
 seen-0.244
 spe-0.234
 coloured-0.232
leaf-0.231
Token Tommy
Token ID19919
Feature activation+7.47e-01

Pos logit contributions
uing+0.077
orting+0.070
izes+0.066
ail+0.065
 nicely+0.062
Neg logit contributions
 wearing-0.043
 blonde-0.041
 seen-0.039
 riding-0.037
 sorts-0.037
Token vowed
Token ID19982
Feature activation+4.66e+00

Pos logit contributions
 Later+0.030
orting+0.029
uing+0.028
 nicely+0.027
 Finally+0.025
Neg logit contributions
e-0.028
 seen-0.028
 spe-0.027
 comet-0.026
leaf-0.026
Token to
Token ID284
Feature activation+3.88e-01

Pos logit contributions
orting+0.266
uing+0.265
 Later+0.265
arl+0.238
lated+0.238
Neg logit contributions
 seen-0.094
 ever-0.093
 wearing-0.084
 supposed-0.081
 flying-0.075
Token find
Token ID1064
Feature activation+2.70e+00

Pos logit contributions
orting+0.105
 Later+0.103
uing+0.100
 nicely+0.097
 Finally+0.088
Neg logit contributions
e-0.084
 seen-0.082
 spe-0.081
 coloured-0.079
 comet-0.077
Token a
Token ID257
Feature activation-3.30e+00

Pos logit contributions
orting+0.112
 Later+0.108
uing+0.104
 nicely+0.103
 Finally+0.094
Neg logit contributions
e-0.099
 seen-0.093
 spe-0.092
 coloured-0.090
vous-0.088
Token place
Token ID1295
Feature activation+7.44e-01

Pos logit contributions
e+0.130
 seen+0.122
 coloured+0.116
 kinds+0.114
 spe+0.113
Neg logit contributions
orting-0.132
 Later-0.129
uing-0.125
 nicely-0.123
 Finally-0.112
Token to
Token ID284
Feature activation-1.27e+00

Pos logit contributions
orting+0.027
 nicely+0.025
 Later+0.025
uing+0.024
 Finally+0.021
Neg logit contributions
e-0.032
 spe-0.030
 comet-0.029
ymes-0.029
vous-0.029
Token hang
Token ID8181
Feature activation+4.86e+00

Pos logit contributions
e+0.041
 seen+0.040
 spe+0.039
 coloured+0.038
 kinds+0.037
Neg logit contributions
orting-0.057
 Later-0.056
 nicely-0.054
uing-0.054
 Finally-0.048
Token it
Token ID340
Feature activation+9.12e+00

Pos logit contributions
orting+0.205
 Later+0.193
uing+0.186
 nicely+0.186
 Finally+0.165
Neg logit contributions
e-0.178
 seen-0.172
 spe-0.164
 coloured-0.160
vous-0.160
Token.
Token ID13
Feature activation+1.33e+00

Pos logit contributions
orting+0.414
 Later+0.404
uing+0.386
 nicely+0.367
 Finally+0.348
Neg logit contributions
e-0.307
 seen-0.296
 coloured-0.283
 spe-0.276
 kinds-0.272
Token From
Token ID3574
Feature activation+1.98e+00

Pos logit contributions
uing+0.059
orting+0.055
rier+0.050
izes+0.048
 nicely+0.048
Neg logit contributions
 wearing-0.045
 sorts-0.043
e-0.043
 blonde-0.041
 musical-0.039
Token then
Token ID788
Feature activation+1.18e+00

Pos logit contributions
orting+0.066
 Later+0.064
uing+0.061
 nicely+0.057
 Finally+0.052
Neg logit contributions
e-0.091
 seen-0.089
 coloured-0.085
 spe-0.084
 kinds-0.083
Token on
Token ID319
Feature activation-2.78e+00

Pos logit contributions
orting+0.074
 Later+0.073
uing+0.071
 nicely+0.069
 Finally+0.065
Neg logit contributions
e-0.020
 seen-0.017
 coloured-0.015
 spe-0.014
 sorts-0.014
Token,
Token ID11
Feature activation+3.29e+00

Pos logit contributions
e+0.108
 seen+0.099
 coloured+0.096
 wearing+0.095
 spe+0.095
Neg logit contributions
orting-0.114
 Later-0.111
uing-0.107
 nicely-0.099
arl-0.092
Token he
Token ID339
Feature activation+5.22e+00

Pos logit contributions
orting+0.090
 Later+0.087
 nicely+0.083
uing+0.080
 Finally+0.072
Neg logit contributions
e-0.112
 seen-0.105
 spe-0.103
 coloured-0.102
leaf-0.101
Token drop
Token ID4268
Feature activation+2.08e+00

Pos logit contributions
 nicely+0.184
 Later+0.178
orting+0.177
uing+0.169
 patiently+0.158
Neg logit contributions
 spe-0.188
 kinds-0.187
 comet-0.185
 universe-0.182
 sorts-0.182
Token everything
Token ID2279
Feature activation+5.33e+00

Pos logit contributions
orting+0.079
 nicely+0.074
 Later+0.074
uing+0.070
 Finally+0.063
Neg logit contributions
e-0.082
 spe-0.080
 seen-0.078
 coloured-0.077
 comet-0.076
Token to
Token ID284
Feature activation+2.96e-01

Pos logit contributions
orting+0.188
 Later+0.180
uing+0.172
 nicely+0.169
 Finally+0.148
Neg logit contributions
e-0.231
 seen-0.221
 spe-0.215
 coloured-0.213
 comet-0.208
Token try
Token ID1949
Feature activation+5.49e+00

Pos logit contributions
orting+0.014
 Later+0.014
uing+0.013
 nicely+0.013
 Finally+0.012
Neg logit contributions
e-0.010
 seen-0.009
 coloured-0.008
 spe-0.008
 kinds-0.008
Token it
Token ID340
Feature activation+9.07e+00

Pos logit contributions
orting+0.230
 Later+0.225
uing+0.213
 nicely+0.204
 Finally+0.191
Neg logit contributions
e-0.204
 seen-0.197
 coloured-0.188
 spe-0.185
 kinds-0.182
Token.
Token ID13
Feature activation+6.85e-01

Pos logit contributions
orting+0.366
 Later+0.359
uing+0.339
 nicely+0.321
 Finally+0.303
Neg logit contributions
e-0.352
 seen-0.339
 coloured-0.324
 spe-0.316
 kinds-0.315
Token<|endoftext|>
Token ID50256
Feature activation-3.05e+00

Pos logit contributions
uing+0.026
orting+0.024
rier+0.020
lated+0.020
izes+0.020
Neg logit contributions
e-0.029
 sorts-0.029
 wearing-0.028
 seen-0.028
 kinds-0.027
TokenOnce
Token ID7454
Feature activation-2.37e-01

Pos logit contributions
e+0.168
 seen+0.167
 spe+0.165
 coloured+0.164
 comet+0.160
Neg logit contributions
orting-0.068
 Later-0.068
 nicely-0.056
uing-0.052
 Finally-0.051
Token upon
Token ID2402
Feature activation-3.37e+00

Pos logit contributions
 seen+0.007
 wearing+0.006
 happened+0.006
 ever+0.006
 universe+0.006
Neg logit contributions
orting-0.013
uing-0.012
 Later-0.011
der-0.011
rier-0.010
Token a
Token ID257
Feature activation-8.33e+00

Pos logit contributions
 spe+0.133
 sorts+0.119
leaf+0.116
 kinds+0.115
 comet+0.113
Neg logit contributions
 nicely-0.126
 Later-0.119
He-0.106
His-0.106
 Finally-0.105
Token together
Token ID1978
Feature activation+5.66e+00

Pos logit contributions
e+0.002
 seen+0.002
 coloured+0.002
 spe+0.002
 comet+0.002
Neg logit contributions
orting-0.002
 Later-0.002
uing-0.002
 nicely-0.002
 Finally-0.002
Token to
Token ID284
Feature activation-5.11e-01

Pos logit contributions
orting+0.200
 Later+0.194
 nicely+0.186
uing+0.185
 Finally+0.163
Neg logit contributions
e-0.238
 seen-0.231
 coloured-0.227
leaf-0.226
 spe-0.225
Token examine
Token ID10716
Feature activation+4.75e+00

Pos logit contributions
e+0.017
 seen+0.016
 coloured+0.016
 spe+0.016
 kinds+0.015
Neg logit contributions
orting-0.023
 Later-0.023
uing-0.022
 nicely-0.021
 Finally-0.020
Token it
Token ID340
Feature activation+7.85e+00

Pos logit contributions
orting+0.194
 nicely+0.183
 Later+0.181
uing+0.176
 Finally+0.155
Neg logit contributions
e-0.174
 seen-0.172
 supposed-0.164
vous-0.164
leaf-0.161
Token more
Token ID517
Feature activation+4.19e+00

Pos logit contributions
orting+0.320
 Later+0.311
uing+0.291
 nicely+0.269
 Finally+0.262
Neg logit contributions
e-0.302
 seen-0.290
 coloured-0.283
 kinds-0.275
 spe-0.273
Token closely
Token ID7173
Feature activation+9.02e+00

Pos logit contributions
orting+0.151
 Later+0.147
uing+0.140
 nicely+0.139
 Finally+0.120
Neg logit contributions
e-0.181
 seen-0.171
 spe-0.165
 coloured-0.163
leaf-0.163
Token.
Token ID13
Feature activation+1.51e+00

Pos logit contributions
orting+0.294
 Later+0.277
 nicely+0.272
uing+0.263
 Finally+0.232
Neg logit contributions
e-0.407
leaf-0.390
 seen-0.389
vous-0.387
 comet-0.382
Token<|endoftext|>
Token ID50256
Feature activation-2.88e+00

Pos logit contributions
uing+0.076
orting+0.069
rier+0.069
izes+0.063
outing+0.063
Neg logit contributions
 wearing-0.044
 sorts-0.044
 a-0.040
 seen-0.039
e-0.037
TokenOnce
Token ID7454
Feature activation+2.37e-01

Pos logit contributions
e+0.170
 seen+0.169
 spe+0.167
 coloured+0.166
 kinds+0.164
Neg logit contributions
 Later-0.054
orting-0.053
 nicely-0.043
uing-0.039
 Finally-0.038
Token upon
Token ID2402
Feature activation-3.27e+00

Pos logit contributions
orting+0.012
uing+0.011
 Later+0.011
rier+0.010
der+0.010
Neg logit contributions
 seen-0.007
 wearing-0.007
 happened-0.006
 ever-0.006
 kinds-0.006
Token a
Token ID257
Feature activation-8.31e+00

Pos logit contributions
 spe+0.118
leaf+0.105
 sorts+0.103
 kinds+0.101
 comet+0.098
Neg logit contributions
 nicely-0.136
 Later-0.134
orting-0.119
 Finally-0.118
Finally-0.115
Token Lily
Token ID20037
Feature activation+2.14e+00

Pos logit contributions
orting+0.143
 Later+0.141
uing+0.134
 nicely+0.122
 Finally+0.121
Neg logit contributions
e-0.106
 seen-0.096
 coloured-0.095
 spe-0.093
 wearing-0.091
Token for
Token ID329
Feature activation-1.06e+00

Pos logit contributions
orting+0.077
 Later+0.075
uing+0.070
Finally+0.063
arl+0.062
Neg logit contributions
e-0.091
 seen-0.084
 kinds-0.083
 wearing-0.083
 sorts-0.083
Token helping
Token ID5742
Feature activation+3.37e+00

Pos logit contributions
e+0.032
 seen+0.031
 coloured+0.029
 spe+0.029
 kinds+0.028
Neg logit contributions
orting-0.051
 Later-0.050
uing-0.048
 nicely-0.047
 Finally-0.043
Token to
Token ID284
Feature activation+1.12e-01

Pos logit contributions
orting+0.131
 Later+0.128
uing+0.123
 nicely+0.121
 Finally+0.108
Neg logit contributions
e-0.132
 seen-0.128
 spe-0.125
 coloured-0.124
leaf-0.121
Token wrap
Token ID14441
Feature activation+4.52e+00

Pos logit contributions
orting+0.005
 Later+0.005
uing+0.005
 nicely+0.005
 Finally+0.004
Neg logit contributions
e-0.004
 seen-0.004
 coloured-0.003
 spe-0.003
 kinds-0.003
Token it
Token ID340
Feature activation+8.98e+00

Pos logit contributions
orting+0.210
 Later+0.207
uing+0.197
 nicely+0.187
 Finally+0.178
Neg logit contributions
e-0.149
 seen-0.142
 coloured-0.135
 spe-0.132
 kinds-0.131
Token.
Token ID13
Feature activation+1.68e+00

Pos logit contributions
orting+0.398
 Later+0.386
uing+0.364
 nicely+0.337
 Finally+0.330
Neg logit contributions
e-0.316
 seen-0.303
 coloured-0.293
 colored-0.274
 sorts-0.274
Token Lily
Token ID20037
Feature activation-2.44e-01

Pos logit contributions
uing+0.082
orting+0.075
izes+0.071
rier+0.070
ealing+0.066
Neg logit contributions
 wearing-0.049
 sorts-0.047
 a-0.043
 blonde-0.040
 kinds-0.040
Token smiled
Token ID13541
Feature activation+2.58e+00

Pos logit contributions
e+0.009
 seen+0.008
 spe+0.008
 coloured+0.008
 kinds+0.008
Neg logit contributions
orting-0.010
 Later-0.010
uing-0.010
 nicely-0.009
 Finally-0.009
Token and
Token ID290
Feature activation+1.81e+00

Pos logit contributions
orting+0.141
uing+0.131
 Later+0.126
hew+0.120
arl+0.120
Neg logit contributions
 wearing-0.062
 coloured-0.062
e-0.061
 seen-0.061
 comet-0.058
Token felt
Token ID2936
Feature activation-6.87e-01

Pos logit contributions
orting+0.077
 Later+0.074
uing+0.072
 nicely+0.068
 Finally+0.063
Neg logit contributions
e-0.067
 seen-0.065
 coloured-0.062
 spe-0.061
 kinds-0.060
Token of
Token ID286
Feature activation-1.72e+00

Pos logit contributions
orting+0.075
 Later+0.074
 nicely+0.069
uing+0.068
 Finally+0.062
Neg logit contributions
 spe-0.083
e-0.081
 comet-0.078
vous-0.076
 seen-0.075
Token her
Token ID607
Feature activation+3.59e+00

Pos logit contributions
e+0.065
 seen+0.063
 spe+0.060
 coloured+0.060
vous+0.060
Neg logit contributions
orting-0.069
 Later-0.067
uing-0.065
 nicely-0.063
 Finally-0.057
Token strength
Token ID4202
Feature activation+1.21e+00

Pos logit contributions
orting+0.146
uing+0.143
 Later+0.140
 nicely+0.138
 Finally+0.129
Neg logit contributions
e-0.122
 seen-0.117
 sorts-0.117
vous-0.117
 kinds-0.114
Token to
Token ID284
Feature activation-3.08e-01

Pos logit contributions
orting+0.052
 Later+0.052
uing+0.049
 nicely+0.046
 Finally+0.044
Neg logit contributions
e-0.043
 seen-0.042
 coloured-0.040
 spe-0.039
 kinds-0.038
Token hang
Token ID8181
Feature activation+4.68e+00

Pos logit contributions
e+0.009
 seen+0.009
 spe+0.009
 kinds+0.009
 coloured+0.009
Neg logit contributions
orting-0.014
 Later-0.014
uing-0.014
 nicely-0.013
 Finally-0.012
Token it
Token ID340
Feature activation+8.90e+00

Pos logit contributions
orting+0.217
 Later+0.211
uing+0.203
 nicely+0.196
 Finally+0.183
Neg logit contributions
e-0.154
 seen-0.148
 coloured-0.139
 spe-0.138
 kinds-0.135
Token onto
Token ID4291
Feature activation+1.37e+00

Pos logit contributions
orting+0.407
 Later+0.398
uing+0.378
 nicely+0.358
 Finally+0.343
Neg logit contributions
e-0.297
 seen-0.283
 coloured-0.272
 spe-0.264
 sorts-0.260
Token the
Token ID262
Feature activation+4.58e+00

Pos logit contributions
orting+0.058
 Later+0.057
 nicely+0.055
uing+0.054
 Finally+0.051
Neg logit contributions
 seen-0.048
e-0.047
leaf-0.045
vous-0.044
 coloured-0.044
Token jail
Token ID7356
Feature activation+2.69e+00

Pos logit contributions
 nicely+0.188
orting+0.184
 Later+0.179
 Finally+0.162
uing+0.152
Neg logit contributions
 sorts-0.168
 kinds-0.167
e-0.161
 coloured-0.160
 seen-0.159
Token.
Token ID13
Feature activation+1.33e+00

Pos logit contributions
orting+0.106
 Later+0.103
uing+0.098
 nicely+0.095
 Finally+0.087
Neg logit contributions
e-0.106
 seen-0.102
 coloured-0.098
 spe-0.097
 kinds-0.095
Token
Token ID220
Feature activation+5.49e+00

Pos logit contributions
uing+0.042
orting+0.042
 nicely+0.035
 Later+0.033
edy+0.031
Neg logit contributions
e-0.064
 wearing-0.059
 seen-0.059
 sorts-0.059
leaf-0.057
Token cereal
Token ID33158
Feature activation+8.60e-01

Pos logit contributions
orting+0.104
 Later+0.100
 nicely+0.100
uing+0.093
 Finally+0.092
Neg logit contributions
e-0.089
 seen-0.082
 sorts-0.078
 coloured-0.077
 kinds-0.075
Token and
Token ID290
Feature activation+1.05e+00

Pos logit contributions
orting+0.034
 Later+0.033
uing+0.032
 nicely+0.031
 Finally+0.028
Neg logit contributions
e-0.034
 seen-0.033
 coloured-0.031
 spe-0.031
 kinds-0.030
Token set
Token ID900
Feature activation+7.13e-01

Pos logit contributions
 Later+0.041
orting+0.040
 nicely+0.038
uing+0.037
 Finally+0.035
Neg logit contributions
e-0.042
 seen-0.040
 spe-0.038
vous-0.038
 comet-0.037
Token off
Token ID572
Feature activation+2.18e+00

Pos logit contributions
orting+0.034
 Later+0.033
uing+0.030
outing+0.028
der+0.028
Neg logit contributions
e-0.024
 seen-0.021
 wearing-0.021
 coloured-0.020
 ever-0.020
Token mailing
Token ID21898
Feature activation+4.75e+00

Pos logit contributions
orting+0.097
 Later+0.095
uing+0.090
 nicely+0.088
 Finally+0.082
Neg logit contributions
e-0.075
 seen-0.072
 coloured-0.069
 spe-0.068
 kinds-0.067
Token it
Token ID340
Feature activation+8.90e+00

Pos logit contributions
orting+0.220
 Later+0.217
uing+0.206
 nicely+0.195
 Finally+0.185
Neg logit contributions
e-0.157
 seen-0.150
 coloured-0.143
 spe-0.140
 kinds-0.139
Token to
Token ID284
Feature activation+1.49e+00

Pos logit contributions
orting+0.374
 Later+0.366
uing+0.347
 nicely+0.325
 Finally+0.309
Neg logit contributions
e-0.329
 seen-0.319
 coloured-0.302
 sorts-0.293
 kinds-0.293
Token all
Token ID477
Feature activation+3.28e+00

Pos logit contributions
orting+0.069
 Later+0.066
uing+0.064
 nicely+0.058
 Finally+0.056
Neg logit contributions
e-0.052
 seen-0.049
 coloured-0.045
 sorts-0.045
 kinds-0.045
Token its
Token ID663
Feature activation+6.75e+00

Pos logit contributions
orting+0.139
 Later+0.136
uing+0.129
 nicely+0.126
 Finally+0.115
Neg logit contributions
e-0.119
 seen-0.115
 spe-0.111
 coloured-0.110
 comet-0.107
Token special
Token ID2041
Feature activation+5.17e+00

Pos logit contributions
orting+0.304
 Later+0.302
 nicely+0.284
uing+0.284
 Finally+0.267
Neg logit contributions
e-0.226
 seen-0.214
 spe-0.201
 coloured-0.200
 comet-0.199
Token destinations
Token ID23982
Feature activation+1.88e+00

Pos logit contributions
orting+0.224
 nicely+0.224
 Later+0.217
uing+0.211
 Finally+0.205
Neg logit contributions
e-0.168
leaf-0.157
 seen-0.155
 spe-0.151
 supposed-0.148
Token quickly
Token ID2952
Feature activation+1.15e+00

Pos logit contributions
 Later+0.045
orting+0.042
 nicely+0.040
uing+0.039
 Finally+0.038
Neg logit contributions
e-0.043
 universe-0.042
ymes-0.041
 comet-0.041
vous-0.041
Token started
Token ID2067
Feature activation+8.40e-01

Pos logit contributions
orting+0.053
 Later+0.053
uing+0.050
 nicely+0.049
 Finally+0.046
Neg logit contributions
e-0.037
 seen-0.035
 spe-0.033
 comet-0.033
 coloured-0.033
Token to
Token ID284
Feature activation-9.69e-01

Pos logit contributions
 Later+0.038
orting+0.037
uing+0.036
 nicely+0.036
 patiently+0.034
Neg logit contributions
e-0.025
 comet-0.025
vous-0.024
ymes-0.024
 universe-0.024
Token m
Token ID285
Feature activation+1.96e+00

Pos logit contributions
ymes+0.038
e+0.035
 seen+0.034
 spe+0.034
leaf+0.033
Neg logit contributions
 Later-0.038
orting-0.035
uing-0.034
 nicely-0.034
 patiently-0.032
Tokenunch
Token ID3316
Feature activation+4.10e+00

Pos logit contributions
orting+0.088
 Later+0.085
 nicely+0.083
uing+0.081
 Finally+0.074
Neg logit contributions
e-0.072
 seen-0.064
vous-0.057
 supposed-0.056
 coloured-0.056
Token it
Token ID340
Feature activation+8.86e+00

Pos logit contributions
 Later+0.219
orting+0.209
uing+0.205
 Finally+0.190
 nicely+0.189
Neg logit contributions
e-0.112
 seen-0.095
 coloured-0.094
 ever-0.092
 kinds-0.088
Token all
Token ID477
Feature activation+4.37e+00

Pos logit contributions
orting+0.491
 Later+0.478
uing+0.465
 nicely+0.443
 Finally+0.423
Neg logit contributions
e-0.215
 seen-0.202
 coloured-0.186
 spe-0.177
 ever-0.171
Token up
Token ID510
Feature activation+5.53e+00

Pos logit contributions
orting+0.183
 Later+0.180
uing+0.178
 nicely+0.162
 Finally+0.154
Neg logit contributions
e-0.160
 seen-0.159
 coloured-0.156
 sorts-0.150
 kinds-0.149
Token.
Token ID13
Feature activation+1.38e+00

Pos logit contributions
orting+0.254
 Later+0.248
uing+0.241
 nicely+0.222
 Finally+0.212
Neg logit contributions
e-0.182
 seen-0.180
 coloured-0.171
 spe-0.168
 ever-0.160
Token He
Token ID679
Feature activation+3.60e+00

Pos logit contributions
uing+0.051
orting+0.050
izes+0.042
lated+0.039
 nicely+0.038
Neg logit contributions
e-0.057
 seen-0.056
 wearing-0.056
 sorts-0.055
leaf-0.052
Token was
Token ID373
Feature activation-3.50e+00

Pos logit contributions
 Later+0.144
orting+0.135
 nicely+0.134
uing+0.133
 patiently+0.124
Neg logit contributions
e-0.129
 seen-0.125
 comet-0.124
 universe-0.124
leaf-0.123
Token properly
Token ID6105
Feature activation+4.25e+00

Pos logit contributions
orting+0.025
 Later+0.024
uing+0.023
 nicely+0.023
 Finally+0.020
Neg logit contributions
e-0.026
 seen-0.025
 spe-0.024
 coloured-0.024
 comet-0.023
Token instead
Token ID2427
Feature activation+5.14e+00

Pos logit contributions
orting+0.165
 Later+0.160
uing+0.152
 nicely+0.147
 Finally+0.134
Neg logit contributions
e-0.171
 seen-0.165
 spe-0.157
 coloured-0.157
 kinds-0.154
Token of
Token ID286
Feature activation-1.46e+00

Pos logit contributions
orting+0.194
 Later+0.188
 nicely+0.177
uing+0.176
 Finally+0.156
Neg logit contributions
e-0.205
 spe-0.200
 seen-0.198
leaf-0.195
 coloured-0.194
Token just
Token ID655
Feature activation+4.63e+00

Pos logit contributions
e+0.047
vous+0.046
 seen+0.045
leaf+0.044
 sorts+0.043
Neg logit contributions
orting-0.065
 nicely-0.065
uing-0.062
 Later-0.062
 patiently-0.057
Token giving
Token ID3501
Feature activation+3.71e+00

Pos logit contributions
orting+0.195
 nicely+0.187
 Later+0.186
uing+0.177
 Finally+0.163
Neg logit contributions
e-0.162
 comet-0.154
 spe-0.153
leaf-0.150
 sorts-0.150
Token it
Token ID340
Feature activation+8.85e+00

Pos logit contributions
orting+0.147
 Later+0.144
uing+0.137
 nicely+0.128
 Finally+0.120
Neg logit contributions
e-0.148
 seen-0.143
 coloured-0.136
 spe-0.133
 kinds-0.132
Token to
Token ID284
Feature activation+1.17e+00

Pos logit contributions
orting+0.347
 Later+0.329
 nicely+0.320
uing+0.317
 Finally+0.277
Neg logit contributions
e-0.353
 spe-0.337
 seen-0.333
 coloured-0.325
 comet-0.322
Token him
Token ID683
Feature activation+5.76e+00

Pos logit contributions
orting+0.050
 Later+0.049
uing+0.046
 nicely+0.043
 Finally+0.041
Neg logit contributions
e-0.044
 seen-0.042
 coloured-0.040
 kinds-0.039
 sorts-0.038
Token.
Token ID13
Feature activation+1.08e+00

Pos logit contributions
orting+0.219
 Later+0.213
uing+0.203
 nicely+0.197
 Finally+0.178
Neg logit contributions
e-0.234
 seen-0.228
 spe-0.219
 coloured-0.218
 kinds-0.212
Token She
Token ID1375
Feature activation+2.55e+00

Pos logit contributions
uing+0.054
rier+0.047
orting+0.046
izes+0.042
lated+0.042
Neg logit contributions
 sorts-0.034
 wearing-0.031
 seen-0.030
 kinds-0.029
e-0.027
Token decided
Token ID3066
Feature activation+7.52e-01

Pos logit contributions
orting+0.114
 Later+0.107
uing+0.104
 nicely+0.097
arl+0.095
Neg logit contributions
e-0.091
 seen-0.088
 coloured-0.086
 spe-0.082
 colored-0.082
Token would
Token ID561
Feature activation-1.17e-01

Pos logit contributions
orting+0.260
 Later+0.257
uing+0.247
 nicely+0.238
 Finally+0.221
Neg logit contributions
e-0.236
 seen-0.229
 spe-0.216
 coloured-0.213
leaf-0.212
Token be
Token ID307
Feature activation+1.67e+00

Pos logit contributions
e+0.005
 seen+0.004
 coloured+0.004
 spe+0.004
 kinds+0.004
Neg logit contributions
orting-0.005
 Later-0.004
uing-0.004
 nicely-0.004
 Finally-0.004
Token fun
Token ID1257
Feature activation+2.60e+00

Pos logit contributions
orting+0.074
 Later+0.073
 nicely+0.072
uing+0.070
 Finally+0.065
Neg logit contributions
e-0.055
leaf-0.052
 comet-0.051
 seen-0.050
 spe-0.049
Token to
Token ID284
Feature activation-2.34e-01

Pos logit contributions
orting+0.087
 nicely+0.087
uing+0.084
 Later+0.080
 patiently+0.073
Neg logit contributions
e-0.113
 spe-0.110
leaf-0.110
 comet-0.107
vous-0.105
Token try
Token ID1949
Feature activation+5.11e+00

Pos logit contributions
e+0.007
 seen+0.007
 coloured+0.007
 spe+0.007
 kinds+0.006
Neg logit contributions
orting-0.011
 Later-0.011
uing-0.010
 nicely-0.010
 Finally-0.009
Token it
Token ID340
Feature activation+8.84e+00

Pos logit contributions
orting+0.203
 Later+0.199
uing+0.187
 nicely+0.178
 Finally+0.168
Neg logit contributions
e-0.201
 seen-0.195
 coloured-0.187
 spe-0.183
 kinds-0.181
Token too
Token ID1165
Feature activation+2.37e+00

Pos logit contributions
orting+0.366
 Later+0.360
uing+0.338
 nicely+0.321
 Finally+0.305
Neg logit contributions
e-0.334
 seen-0.322
 coloured-0.308
 spe-0.300
 kinds-0.298
Token.
Token ID13
Feature activation+6.67e-01

Pos logit contributions
orting+0.095
 Later+0.092
uing+0.088
 nicely+0.084
 Finally+0.078
Neg logit contributions
e-0.094
 seen-0.091
 coloured-0.087
 spe-0.086
 kinds-0.084
Token She
Token ID1375
Feature activation+2.27e+00

Pos logit contributions
uing+0.028
orting+0.026
rier+0.022
lated+0.021
izes+0.021
Neg logit contributions
e-0.026
 seen-0.025
 sorts-0.025
 wearing-0.024
 blonde-0.023
Token found
Token ID1043
Feature activation+9.82e-01

Pos logit contributions
 Later+0.092
orting+0.087
 nicely+0.087
uing+0.087
 patiently+0.079
Neg logit contributions
e-0.083
 seen-0.079
 spe-0.076
 comet-0.076
 supposed-0.076
Token some
Token ID617
Feature activation-2.16e+00

Pos logit contributions
orting+0.048
 Later+0.047
uing+0.045
 nicely+0.044
 Finally+0.042
Neg logit contributions
e-0.029
 seen-0.028
 spe-0.027
 coloured-0.026
 kinds-0.025
Token he
Token ID339
Feature activation+3.47e+00

Pos logit contributions
orting+0.035
 Later+0.034
uing+0.031
 nicely+0.030
 Finally+0.027
Neg logit contributions
e-0.050
 seen-0.048
 coloured-0.046
 spe-0.046
 kinds-0.045
Token was
Token ID373
Feature activation-3.53e+00

Pos logit contributions
 Later+0.128
orting+0.122
 nicely+0.119
uing+0.118
 Finally+0.108
Neg logit contributions
e-0.137
 seen-0.132
 spe-0.131
 coloured-0.129
leaf-0.128
Token determined
Token ID5295
Feature activation+2.89e+00

Pos logit contributions
e+0.131
 seen+0.123
 coloured+0.122
 spe+0.121
 kinds+0.118
Neg logit contributions
 Later-0.144
orting-0.143
 nicely-0.134
uing-0.134
 Finally-0.121
Token to
Token ID284
Feature activation+3.40e-01

Pos logit contributions
orting+0.137
 Later+0.135
uing+0.128
 nicely+0.121
 Finally+0.117
Neg logit contributions
 seen-0.092
e-0.091
 coloured-0.083
 kinds-0.082
 sorts-0.080
Token try
Token ID1949
Feature activation+5.05e+00

Pos logit contributions
orting+0.015
 Later+0.014
uing+0.014
 nicely+0.013
 Finally+0.012
Neg logit contributions
e-0.012
 seen-0.012
 spe-0.011
 coloured-0.011
 kinds-0.011
Token it
Token ID340
Feature activation+8.84e+00

Pos logit contributions
orting+0.228
 Later+0.224
uing+0.212
 nicely+0.203
 Finally+0.194
Neg logit contributions
e-0.173
 seen-0.167
 coloured-0.157
 kinds-0.152
 spe-0.152
Token.
Token ID13
Feature activation+1.52e+00

Pos logit contributions
orting+0.390
 Later+0.385
uing+0.363
 nicely+0.345
 Finally+0.330
Neg logit contributions
e-0.307
 seen-0.296
 coloured-0.280
 kinds-0.273
 sorts-0.272
Token So
Token ID1406
Feature activation+1.09e+00

Pos logit contributions
uing+0.056
orting+0.053
izes+0.043
arl+0.041
lated+0.041
Neg logit contributions
e-0.066
 sorts-0.065
 seen-0.065
 kinds-0.062
 wearing-0.061
Token he
Token ID339
Feature activation+3.94e+00

Pos logit contributions
orting+0.037
 Later+0.035
uing+0.034
 nicely+0.031
arl+0.028
Neg logit contributions
e-0.050
 seen-0.048
 coloured-0.046
 spe-0.046
 kinds-0.045
Token began
Token ID2540
Feature activation+4.72e-01

Pos logit contributions
 Later+0.164
orting+0.160
uing+0.154
 nicely+0.149
 Finally+0.140
Neg logit contributions
e-0.144
 seen-0.137
 spe-0.131
 comet-0.129
 kinds-0.129
Token to
Token ID284
Feature activation-1.18e+00

Pos logit contributions
 Later+0.020
orting+0.019
uing+0.019
 nicely+0.018
 patiently+0.017
Neg logit contributions
e-0.017
 seen-0.016
 spe-0.016
vous-0.016
 coloured-0.015
Token fire
Token ID2046
Feature activation+3.70e+00

Pos logit contributions
 Later+0.149
orting+0.149
 nicely+0.140
uing+0.136
 Finally+0.129
Neg logit contributions
e-0.113
 seen-0.107
 coloured-0.100
 spe-0.098
leaf-0.098
Token and
Token ID290
Feature activation+1.60e+00

Pos logit contributions
orting+0.158
 Later+0.154
uing+0.148
 nicely+0.139
 Finally+0.130
Neg logit contributions
e-0.136
 seen-0.132
 coloured-0.126
 spe-0.125
 kinds-0.122
Token started
Token ID2067
Feature activation+9.72e-01

Pos logit contributions
orting+0.070
 Later+0.069
 nicely+0.065
uing+0.065
 Finally+0.060
Neg logit contributions
e-0.055
 seen-0.052
 coloured-0.049
 spe-0.049
 comet-0.049
Token to
Token ID284
Feature activation-7.16e-01

Pos logit contributions
orting+0.040
 Later+0.040
uing+0.038
 nicely+0.038
 Finally+0.035
Neg logit contributions
e-0.035
 seen-0.033
vous-0.032
 comet-0.032
 coloured-0.031
Token pour
Token ID12797
Feature activation+4.22e+00

Pos logit contributions
e+0.024
 seen+0.023
 coloured+0.021
leaf+0.021
 spe+0.021
Neg logit contributions
orting-0.032
 Later-0.032
uing-0.030
 nicely-0.030
 Finally-0.027
Token it
Token ID340
Feature activation+8.83e+00

Pos logit contributions
orting+0.188
 Later+0.184
uing+0.176
 nicely+0.169
 Finally+0.158
Neg logit contributions
e-0.146
 seen-0.140
 coloured-0.132
 spe-0.131
 kinds-0.128
Token on
Token ID319
Feature activation+7.82e-01

Pos logit contributions
orting+0.414
 Later+0.404
uing+0.386
 nicely+0.362
 Finally+0.348
Neg logit contributions
e-0.286
 seen-0.275
 coloured-0.265
 spe-0.256
 kinds-0.249
Token the
Token ID262
Feature activation+4.61e+00

Pos logit contributions
orting+0.033
 Later+0.033
 nicely+0.031
uing+0.031
 Finally+0.029
Neg logit contributions
e-0.028
 seen-0.027
leaf-0.025
 coloured-0.025
vous-0.025
Token flames
Token ID16209
Feature activation+1.19e+00

Pos logit contributions
orting+0.177
 Later+0.177
 nicely+0.166
uing+0.159
 Finally+0.149
Neg logit contributions
e-0.185
 seen-0.179
 sorts-0.167
 coloured-0.166
 kinds-0.166
Token.
Token ID13
Feature activation+1.39e+00

Pos logit contributions
orting+0.045
 Later+0.043
uing+0.041
 nicely+0.040
 Finally+0.036
Neg logit contributions
e-0.049
 seen-0.047
 coloured-0.045
 spe-0.045
leaf-0.044
Token It
Token ID632
Feature activation+2.99e+00

Pos logit contributions
uing+0.050
orting+0.050
 nicely+0.040
 Later+0.040
arl+0.038
Neg logit contributions
e-0.062
 seen-0.057
 sorts-0.057
 spe-0.057
leaf-0.056
Token looked
Token ID3114
Feature activation+4.90e+00

Pos logit contributions
orting+0.266
 Later+0.258
 nicely+0.250
uing+0.250
 Finally+0.223
Neg logit contributions
e-0.225
 seen-0.218
 spe-0.218
leaf-0.210
 comet-0.209
Token and
Token ID290
Feature activation+2.68e+00

Pos logit contributions
orting+0.227
 Later+0.221
uing+0.214
 nicely+0.209
 Finally+0.193
Neg logit contributions
e-0.159
 seen-0.149
 spe-0.141
 coloured-0.141
 comet-0.138
Token everyone
Token ID2506
Feature activation+8.88e-01

Pos logit contributions
orting+0.107
 Later+0.107
uing+0.100
 nicely+0.100
 Finally+0.092
Neg logit contributions
e-0.102
 seen-0.096
 spe-0.091
 coloured-0.090
 comet-0.090
Token gob
Token ID48484
Feature activation+6.21e+00

Pos logit contributions
orting+0.036
 Later+0.036
uing+0.035
 nicely+0.035
 Finally+0.031
Neg logit contributions
e-0.031
 seen-0.030
 spe-0.029
 coloured-0.028
vous-0.028
Tokenbled
Token ID9342
Feature activation+7.21e+00

Pos logit contributions
orting+0.172
 Later+0.172
uing+0.133
 nicely+0.131
 Finally+0.129
Neg logit contributions
e-0.322
 coloured-0.292
 ever-0.279
 colored-0.276
ymes-0.275
Token it
Token ID340
Feature activation+8.80e+00

Pos logit contributions
orting+0.272
 Later+0.266
uing+0.251
 nicely+0.238
 Finally+0.220
Neg logit contributions
e-0.299
 seen-0.288
 coloured-0.276
 spe-0.273
 kinds-0.268
Token up
Token ID510
Feature activation+5.70e+00

Pos logit contributions
orting+0.383
 Later+0.370
uing+0.343
 nicely+0.319
 Finally+0.311
Neg logit contributions
e-0.324
 seen-0.300
 coloured-0.294
 ever-0.283
 sorts-0.278
Token.
Token ID13
Feature activation+1.68e+00

Pos logit contributions
orting+0.219
 Later+0.213
uing+0.203
 nicely+0.199
 Finally+0.180
Neg logit contributions
e-0.231
 seen-0.222
 spe-0.212
 coloured-0.211
leaf-0.207
Token All
Token ID1439
Feature activation+1.57e+00

Pos logit contributions
orting+0.081
uing+0.078
 nicely+0.069
 Later+0.069
arl+0.066
Neg logit contributions
e-0.053
 seen-0.051
 sorts-0.050
 kinds-0.048
 coloured-0.048
Token agreed
Token ID4987
Feature activation+4.15e+00

Pos logit contributions
orting+0.069
 Later+0.068
uing+0.065
 nicely+0.064
 Finally+0.059
Neg logit contributions
e-0.051
 seen-0.049
 spe-0.048
 coloured-0.046
leaf-0.046
Token that
Token ID326
Feature activation+5.19e+00

Pos logit contributions
orting+0.176
 Later+0.174
uing+0.162
 nicely+0.151
 Finally+0.145
Neg logit contributions
e-0.153
 seen-0.147
 coloured-0.140
 ever-0.137
 wearing-0.137
Token and
Token ID290
Feature activation+1.11e-01

Pos logit contributions
orting+0.120
 Later+0.117
uing+0.111
 nicely+0.109
 Finally+0.100
Neg logit contributions
e-0.128
 seen-0.121
 coloured-0.116
 spe-0.116
 comet-0.114
Token then
Token ID788
Feature activation+1.28e+00

Pos logit contributions
orting+0.004
 Later+0.004
uing+0.003
 nicely+0.003
 Finally+0.003
Neg logit contributions
e-0.005
 seen-0.005
 spe-0.005
 coloured-0.005
 kinds-0.005
Token how
Token ID703
Feature activation+1.25e+00

Pos logit contributions
orting+0.040
 Later+0.039
uing+0.037
 nicely+0.036
 Finally+0.032
Neg logit contributions
e-0.061
 seen-0.059
 spe-0.057
 coloured-0.056
vous-0.056
Token to
Token ID284
Feature activation-5.25e-01

Pos logit contributions
orting+0.052
 Later+0.050
uing+0.048
 nicely+0.045
 Finally+0.042
Neg logit contributions
e-0.048
 seen-0.046
 coloured-0.044
 spe-0.044
 kinds-0.043
Token hang
Token ID8181
Feature activation+4.48e+00

Pos logit contributions
e+0.016
 seen+0.015
 coloured+0.015
 spe+0.014
 sorts+0.014
Neg logit contributions
orting-0.026
 Later-0.026
uing-0.024
 nicely-0.023
 Finally-0.022
Token it
Token ID340
Feature activation+8.78e+00

Pos logit contributions
orting+0.216
 Later+0.214
uing+0.204
 nicely+0.191
 Finally+0.184
Neg logit contributions
e-0.139
 seen-0.131
 coloured-0.128
 spe-0.122
 kinds-0.121
Token.
Token ID13
Feature activation+1.05e+00

Pos logit contributions
orting+0.422
 Later+0.412
uing+0.394
 nicely+0.371
 Finally+0.357
Neg logit contributions
e-0.270
 seen-0.258
 coloured-0.249
 spe-0.236
 sorts-0.235
Token
Token ID220
Feature activation+5.80e+00

Pos logit contributions
orting+0.041
uing+0.039
izes+0.033
lated+0.032
arl+0.032
Neg logit contributions
e-0.041
 wearing-0.041
 sorts-0.039
 seen-0.037
 kinds-0.037
Token
Token ID198
Feature activation+2.42e+00

Pos logit contributions
orting+0.253
uing+0.240
lated+0.221
izes+0.218
der+0.217
Neg logit contributions
e-0.182
 wearing-0.172
 seen-0.168
 sorts-0.157
 supposed-0.155
Token
Token ID198
Feature activation+1.64e+00

Pos logit contributions
orting+0.099
uing+0.090
 Later+0.082
der+0.081
rier+0.079
Neg logit contributions
e-0.098
 wearing-0.088
 seen-0.084
 sorts-0.082
leaf-0.082
TokenJimmy
Token ID40335
Feature activation+2.05e-01

Pos logit contributions
orting+0.072
 Later+0.065
uing+0.064
ail+0.058
arl+0.057
Neg logit contributions
e-0.062
 seen-0.054
 coloured-0.053
 wearing-0.052
leaf-0.052
Token couldn
Token ID3521
Feature activation-2.40e+00

Pos logit contributions
 Later+0.170
orting+0.168
uing+0.160
 nicely+0.157
 Finally+0.146
Neg logit contributions
e-0.142
 seen-0.137
 coloured-0.129
 spe-0.128
 kinds-0.128
Token't
Token ID470
Feature activation-1.32e+00

Pos logit contributions
e+0.074
 ever+0.064
 seen+0.063
 spe+0.061
 wearing+0.060
Neg logit contributions
orting-0.120
uing-0.114
 Later-0.114
 nicely-0.106
 Finally-0.098
Token wait
Token ID4043
Feature activation+3.75e+00

Pos logit contributions
e+0.052
 seen+0.050
 coloured+0.049
 kinds+0.048
 spe+0.048
Neg logit contributions
orting-0.051
 Later-0.049
uing-0.047
 nicely-0.046
 Finally-0.041
Token to
Token ID284
Feature activation+4.78e-01

Pos logit contributions
orting+0.173
 Later+0.169
uing+0.160
 nicely+0.153
 Finally+0.145
Neg logit contributions
e-0.125
 seen-0.123
 coloured-0.115
 spe-0.112
 sorts-0.109
Token try
Token ID1949
Feature activation+5.05e+00

Pos logit contributions
 Later+0.016
orting+0.016
 nicely+0.016
uing+0.015
 patiently+0.014
Neg logit contributions
ymes-0.021
 kinds-0.019
vous-0.019
 spe-0.019
e-0.019
Token it
Token ID340
Feature activation+8.77e+00

Pos logit contributions
orting+0.230
 Later+0.225
uing+0.215
 nicely+0.206
 Finally+0.194
Neg logit contributions
e-0.171
 seen-0.164
 coloured-0.155
 spe-0.153
 kinds-0.150
Token out
Token ID503
Feature activation+4.14e+00

Pos logit contributions
orting+0.409
 Later+0.404
uing+0.382
 nicely+0.364
 Finally+0.351
Neg logit contributions
e-0.285
 seen-0.275
 coloured-0.260
 sorts-0.247
 kinds-0.247
Token.
Token ID13
Feature activation+7.88e-01

Pos logit contributions
orting+0.165
 Later+0.160
uing+0.153
 nicely+0.148
 Finally+0.135
Neg logit contributions
e-0.163
 seen-0.156
 spe-0.150
 coloured-0.150
 comet-0.146
Token
Token ID220
Feature activation+5.55e+00

Pos logit contributions
uing+0.033
orting+0.032
izes+0.026
 nicely+0.026
lated+0.025
Neg logit contributions
 wearing-0.030
 seen-0.029
e-0.028
 sorts-0.027
 blonde-0.026
Token
Token ID198
Feature activation+2.25e+00

Pos logit contributions
orting+0.252
uing+0.242
lated+0.216
der+0.214
rier+0.210
Neg logit contributions
 wearing-0.167
e-0.165
 seen-0.162
 supposed-0.149
 a-0.146
Token
Token ID198
Feature activation+1.52e+00

Pos logit contributions
orting+0.095
uing+0.091
der+0.081
rier+0.080
idents+0.078
Neg logit contributions
e-0.085
 seen-0.081
 wearing-0.079
 ever-0.074
 sorts-0.071
Token you
Token ID345
Feature activation+7.59e-01

Pos logit contributions
orting+0.072
 Later+0.071
uing+0.067
 nicely+0.064
 Finally+0.061
Neg logit contributions
e-0.058
 seen-0.055
 coloured-0.052
 spe-0.051
 kinds-0.051
Token again
Token ID757
Feature activation+5.10e+00

Pos logit contributions
orting+0.030
 Later+0.029
uing+0.028
 nicely+0.027
 Finally+0.024
Neg logit contributions
e-0.030
 seen-0.029
 spe-0.028
 coloured-0.028
 sorts-0.027
Token how
Token ID703
Feature activation+2.79e+00

Pos logit contributions
orting+0.201
 Later+0.197
uing+0.187
 nicely+0.181
 Finally+0.165
Neg logit contributions
e-0.202
 seen-0.194
 coloured-0.185
 spe-0.185
 sorts-0.180
Token to
Token ID284
Feature activation+3.87e-01

Pos logit contributions
orting+0.109
 Later+0.108
 nicely+0.107
uing+0.102
 Finally+0.092
Neg logit contributions
e-0.111
 seen-0.103
 coloured-0.099
 spe-0.099
 sorts-0.096
Token balance
Token ID5236
Feature activation+4.06e+00

Pos logit contributions
orting+0.019
 Later+0.019
uing+0.018
 nicely+0.017
 Finally+0.016
Neg logit contributions
e-0.012
 seen-0.011
 coloured-0.010
 spe-0.010
 kinds-0.010
Token it
Token ID340
Feature activation+8.75e+00

Pos logit contributions
orting+0.187
 Later+0.184
uing+0.175
 nicely+0.168
 Finally+0.159
Neg logit contributions
e-0.134
 seen-0.128
 coloured-0.122
 spe-0.120
 kinds-0.117
Token."
Token ID526
Feature activation-3.19e-01

Pos logit contributions
orting+0.372
 Later+0.364
uing+0.347
 nicely+0.330
 Finally+0.310
Neg logit contributions
e-0.318
 seen-0.306
 coloured-0.291
 spe-0.288
 kinds-0.282
Token
Token ID198
Feature activation+2.04e+00

Pos logit contributions
e+0.012
 wearing+0.011
 seen+0.011
 pencil+0.010
 sorts+0.010
Neg logit contributions
uing-0.014
orting-0.013
 Later-0.012
lated-0.012
Days-0.011
Token
Token ID198
Feature activation+1.38e+00

Pos logit contributions
uing+0.084
orting+0.082
 Later+0.077
rier+0.073
der+0.069
Neg logit contributions
e-0.080
 seen-0.074
 wearing-0.068
 sorts-0.066
 coloured-0.066
TokenL
Token ID43
Feature activation-2.63e+00

Pos logit contributions
 Later+0.062
orting+0.061
uing+0.061
 Eventually+0.054
rier+0.054
Neg logit contributions
e-0.048
 seen-0.042
 seek-0.042
 coloured-0.040
ymes-0.040
Tokenily
Token ID813
Feature activation-1.95e+00

Pos logit contributions
e+0.111
 seen+0.107
 coloured+0.103
 spe+0.102
 kinds+0.100
Neg logit contributions
orting-0.099
 Later-0.096
uing-0.091
 nicely-0.087
 Finally-0.080
Token time
Token ID640
Feature activation-3.67e+00

Pos logit contributions
e+0.169
 seen+0.156
 coloured+0.137
 spe+0.132
 sorts+0.126
Neg logit contributions
orting-0.562
 Later-0.550
uing-0.536
 nicely-0.517
 Finally-0.492
Token,
Token ID11
Feature activation-1.84e+00

Pos logit contributions
e+0.136
 seen+0.130
 wearing+0.120
 ever+0.120
 spe+0.119
Neg logit contributions
orting-0.159
 Later-0.154
uing-0.149
 nicely-0.134
 Finally-0.129
Token there
Token ID612
Feature activation-1.36e+00

Pos logit contributions
e+0.046
 seen+0.043
 coloured+0.040
 spe+0.040
leaf+0.040
Neg logit contributions
 Later-0.098
orting-0.098
uing-0.093
 nicely-0.093
 Finally-0.087
Token were
Token ID547
Feature activation-7.52e+00

Pos logit contributions
e+0.046
 seen+0.044
 coloured+0.042
 spe+0.041
 kinds+0.040
Neg logit contributions
orting-0.062
 Later-0.060
uing-0.058
 nicely-0.055
 Finally-0.052
Token two
Token ID734
Feature activation-7.06e+00

Pos logit contributions
e+0.257
 spe+0.257
 seen+0.255
 coloured+0.249
leaf+0.248
Neg logit contributions
 Later-0.323
orting-0.304
 nicely-0.292
 Finally-0.275
 Eventually-0.264
Token cats
Token ID11875
Feature activation-1.11e+01

Pos logit contributions
e+0.224
 seen+0.215
 coloured+0.202
 spe+0.201
 kinds+0.195
Neg logit contributions
orting-0.334
 Later-0.327
uing-0.313
 nicely-0.302
 Finally-0.283
Token.
Token ID13
Feature activation-3.39e+00

Pos logit contributions
e+0.289
 seen+0.276
 coloured+0.257
 spe+0.256
 kinds+0.247
Neg logit contributions
orting-0.590
 Later-0.578
uing-0.553
 nicely-0.541
 Finally-0.509
Token They
Token ID1119
Feature activation-2.19e+00

Pos logit contributions
 seen+0.181
e+0.180
 coloured+0.173
 comet+0.173
leaf+0.171
Neg logit contributions
 Later-0.086
orting-0.081
 nicely-0.069
 Finally-0.066
uing-0.064
Token were
Token ID547
Feature activation-6.28e+00

Pos logit contributions
 seen+0.077
e+0.077
 comet+0.075
 spe+0.074
leaf+0.073
Neg logit contributions
 Later-0.089
orting-0.089
 nicely-0.085
uing-0.081
 Finally-0.077
Token friends
Token ID2460
Feature activation-8.71e+00

Pos logit contributions
 seen+0.240
 comet+0.235
 coloured+0.228
leaf+0.225
ymes+0.224
Neg logit contributions
 Later-0.258
orting-0.238
 nicely-0.230
 Finally-0.219
 Eventually-0.205
Token and
Token ID290
Feature activation-2.83e+00

Pos logit contributions
e+0.335
 seen+0.333
 spe+0.325
 comet+0.320
leaf+0.320
Neg logit contributions
orting-0.342
 Later-0.332
 nicely-0.312
uing-0.306
 Finally-0.275
Token time
Token ID640
Feature activation-3.33e+00

Pos logit contributions
e+0.179
 seen+0.168
 coloured+0.154
 spe+0.152
 kinds+0.146
Neg logit contributions
orting-0.517
 Later-0.511
uing-0.491
 nicely-0.479
 Finally-0.457
Token,
Token ID11
Feature activation-1.40e+00

Pos logit contributions
e+0.131
 seen+0.126
 spe+0.118
 coloured+0.118
 wearing+0.116
Neg logit contributions
orting-0.134
 Later-0.130
uing-0.125
 nicely-0.115
 Finally-0.108
Token there
Token ID612
Feature activation-9.81e-01

Pos logit contributions
e+0.036
 seen+0.035
leaf+0.034
 kinds+0.033
 spe+0.033
Neg logit contributions
 Later-0.072
orting-0.071
 nicely-0.069
uing-0.068
 Finally-0.064
Token were
Token ID547
Feature activation-7.17e+00

Pos logit contributions
e+0.036
 seen+0.036
leaf+0.034
 spe+0.034
 coloured+0.034
Neg logit contributions
 Later-0.040
orting-0.039
uing-0.037
 nicely-0.037
 Finally-0.034
Token two
Token ID734
Feature activation-6.83e+00

Pos logit contributions
e+0.246
 sorts+0.242
leaf+0.241
 seen+0.241
 kinds+0.240
Neg logit contributions
 Later-0.303
orting-0.281
 nicely-0.276
 Finally-0.259
 Eventually-0.248
Token cats
Token ID11875
Feature activation-1.09e+01

Pos logit contributions
e+0.215
 seen+0.206
 coloured+0.194
 spe+0.192
 kinds+0.187
Neg logit contributions
orting-0.326
 Later-0.319
uing-0.306
 nicely-0.294
 Finally-0.278
Token that
Token ID326
Feature activation+3.18e-01

Pos logit contributions
e+0.293
 seen+0.287
 spe+0.273
 coloured+0.268
 comet+0.266
Neg logit contributions
orting-0.556
 Later-0.545
 nicely-0.515
uing-0.514
 Finally-0.476
Token lived
Token ID5615
Feature activation-3.72e+00

Pos logit contributions
uing+0.017
orting+0.017
 Later+0.016
rier+0.015
arl+0.015
Neg logit contributions
e-0.009
 seen-0.008
 sorts-0.007
 coloured-0.007
 kinds-0.007
Token very
Token ID845
Feature activation-4.52e+00

Pos logit contributions
 comet+0.119
 seen+0.119
 spe+0.114
e+0.113
 kinds+0.113
Neg logit contributions
 nicely-0.169
 Later-0.167
orting-0.165
uing-0.153
 Finally-0.144
Token far
Token ID1290
Feature activation+5.62e-01

Pos logit contributions
e+0.139
 seen+0.133
 coloured+0.125
 spe+0.124
 kinds+0.120
Neg logit contributions
orting-0.220
 Later-0.215
uing-0.206
 nicely-0.199
 Finally-0.187
Token away
Token ID1497
Feature activation+2.98e+00

Pos logit contributions
orting+0.020
 nicely+0.018
 Later+0.017
uing+0.016
arl+0.015
Neg logit contributions
e-0.025
 seen-0.024
odic-0.023
leaf-0.023
 supposed-0.023
Token a
Token ID257
Feature activation-9.29e+00

Pos logit contributions
 spe+0.165
 sorts+0.154
leaf+0.149
 kinds+0.148
 supposed+0.139
Neg logit contributions
 nicely-0.158
 Later-0.153
orting-0.132
 Finally-0.131
His-0.130
Token time
Token ID640
Feature activation-4.04e+00

Pos logit contributions
e+0.173
 seen+0.156
 coloured+0.138
 spe+0.137
 kinds+0.127
Neg logit contributions
orting-0.569
 Later-0.556
uing-0.542
 nicely-0.522
 Finally-0.499
Token there
Token ID612
Feature activation-1.56e+00

Pos logit contributions
e+0.138
 seen+0.128
 wearing+0.125
 ever+0.123
 spe+0.120
Neg logit contributions
orting-0.184
 Later-0.180
uing-0.175
 nicely-0.155
 Finally-0.154
Token were
Token ID547
Feature activation-7.52e+00

Pos logit contributions
e+0.048
 seen+0.045
 coloured+0.042
 spe+0.040
 kinds+0.040
Neg logit contributions
orting-0.079
 Later-0.076
uing-0.074
 nicely-0.069
arl-0.065
Token two
Token ID734
Feature activation-7.34e+00

Pos logit contributions
e+0.251
 sorts+0.246
 kinds+0.242
 seen+0.242
leaf+0.240
Neg logit contributions
 Later-0.327
orting-0.319
 nicely-0.297
 Finally-0.278
uing-0.272
Token cats
Token ID11875
Feature activation-1.08e+01

Pos logit contributions
e+0.229
 seen+0.218
 coloured+0.206
 spe+0.204
 kinds+0.199
Neg logit contributions
orting-0.353
 Later-0.347
uing-0.332
 nicely-0.318
 Finally-0.302
Token who
Token ID508
Feature activation-3.84e+00

Pos logit contributions
e+0.332
 seen+0.329
 spe+0.320
 comet+0.313
leaf+0.312
Neg logit contributions
orting-0.508
 Later-0.495
 nicely-0.473
uing-0.460
 Finally-0.428
Token used
Token ID973
Feature activation-1.19e+00

Pos logit contributions
 seen+0.125
e+0.124
 comet+0.124
 coloured+0.118
leaf+0.118
Neg logit contributions
 Later-0.166
orting-0.163
 nicely-0.155
uing-0.152
 Finally-0.147
Token to
Token ID284
Feature activation-3.09e+00

Pos logit contributions
 comet+0.046
 universe+0.044
e+0.044
vous+0.041
 reader+0.040
Neg logit contributions
 nicely-0.044
orting-0.044
 Later-0.043
uing-0.040
 Finally-0.039
Token travel
Token ID3067
Feature activation+2.85e-01

Pos logit contributions
 seen+0.102
leaf+0.099
 comet+0.099
 supposed+0.098
 spe+0.098
Neg logit contributions
orting-0.129
 nicely-0.128
 Later-0.127
uing-0.122
 Finally-0.110
Token a
Token ID257
Feature activation-5.90e+00

Pos logit contributions
orting+0.012
 Later+0.011
 nicely+0.011
uing+0.010
 Finally+0.010
Neg logit contributions
e-0.011
 seen-0.010
 spe-0.010
 comet-0.010
 coloured-0.010
TokenOnce
Token ID7454
Feature activation-1.89e+00

Pos logit contributions
 spe+0.210
 kinds+0.201
 seen+0.200
 comet+0.200
 coloured+0.200
Neg logit contributions
 Later-0.087
orting-0.085
 nicely-0.072
 Finally-0.063
 Eventually-0.058
Token upon
Token ID2402
Feature activation-4.95e+00

Pos logit contributions
 seen+0.054
 wearing+0.047
 nowhere+0.043
e+0.042
 ever+0.041
Neg logit contributions
orting-0.105
uing-0.096
 Later-0.090
rier-0.083
lated-0.083
Token a
Token ID257
Feature activation-1.01e+01

Pos logit contributions
 spe+0.192
leaf+0.176
 sorts+0.175
 kinds+0.169
ymes+0.162
Neg logit contributions
 Later-0.185
 nicely-0.184
orting-0.156
 Finally-0.154
 He-0.153
Token time
Token ID640
Feature activation-4.65e+00

Pos logit contributions
e+0.181
 seen+0.162
 spe+0.141
 coloured+0.141
 comet+0.129
Neg logit contributions
orting-0.628
 Later-0.614
uing-0.601
 nicely-0.575
 Finally-0.551
Token,
Token ID11
Feature activation-2.82e+00

Pos logit contributions
e+0.182
 seen+0.174
 coloured+0.163
 spe+0.162
 wearing+0.161
Neg logit contributions
orting-0.189
 Later-0.185
uing-0.178
 nicely-0.163
 Finally-0.156
Token a
Token ID257
Feature activation-1.08e+01

Pos logit contributions
e+0.061
 seen+0.057
 coloured+0.053
 spe+0.051
 kinds+0.049
Neg logit contributions
orting-0.163
 Later-0.160
uing-0.154
 nicely-0.149
 Finally-0.142
Token little
Token ID1310
Feature activation-8.07e+00

Pos logit contributions
e+0.337
 seen+0.317
 spe+0.302
 coloured+0.302
 comet+0.293
Neg logit contributions
orting-0.525
 Later-0.513
uing-0.494
 nicely-0.468
 Finally-0.445
Token boy
Token ID2933
Feature activation-9.08e+00

Pos logit contributions
 seen+0.336
 supposed+0.320
e+0.306
 flown+0.293
leaf+0.291
Neg logit contributions
orting-0.310
 nicely-0.287
 Later-0.269
arl-0.253
uing-0.247
Token named
Token ID3706
Feature activation-9.94e+00

Pos logit contributions
leaf+0.318
 comet+0.293
ymes+0.293
 sorts+0.277
 universe+0.276
Neg logit contributions
 nicely-0.373
 Later-0.372
orting-0.367
 Finally-0.321
uing-0.313
Token Pete
Token ID17542
Feature activation-3.50e+00

Pos logit contributions
e+0.348
 seen+0.341
 ever+0.304
 comet+0.303
 coloured+0.301
Neg logit contributions
orting-0.449
uing-0.430
 Later-0.408
der-0.368
 nicely-0.363
Token and
Token ID290
Feature activation-4.74e+00

Pos logit contributions
e+0.119
 seen+0.116
 spe+0.113
leaf+0.110
 coloured+0.109
Neg logit contributions
orting-0.151
 Later-0.149
uing-0.140
 nicely-0.139
 Finally-0.127
Token time
Token ID640
Feature activation-4.29e+00

Pos logit contributions
e+0.174
 seen+0.156
 coloured+0.135
 spe+0.134
 kinds+0.123
Neg logit contributions
orting-0.593
 Later-0.580
uing-0.566
 nicely-0.543
 Finally-0.520
Token,
Token ID11
Feature activation-2.65e+00

Pos logit contributions
e+0.140
 seen+0.131
 wearing+0.128
 ever+0.127
 spe+0.120
Neg logit contributions
orting-0.202
 Later-0.196
uing-0.193
 nicely-0.169
 Finally-0.168
Token there
Token ID612
Feature activation-1.81e+00

Pos logit contributions
e+0.049
 seen+0.045
 coloured+0.041
 spe+0.038
 wearing+0.038
Neg logit contributions
orting-0.164
 Later-0.158
uing-0.155
 nicely-0.146
 Finally-0.142
Token were
Token ID547
Feature activation-7.65e+00

Pos logit contributions
e+0.060
 seen+0.058
 coloured+0.054
 spe+0.053
 kinds+0.053
Neg logit contributions
orting-0.084
 Later-0.082
uing-0.079
 nicely-0.075
 Finally-0.070
Token two
Token ID734
Feature activation-7.45e+00

Pos logit contributions
e+0.257
 sorts+0.254
 kinds+0.251
 seen+0.250
 spe+0.249
Neg logit contributions
 Later-0.331
orting-0.317
 nicely-0.300
 Finally-0.281
uing-0.270
Token brothers
Token ID9397
Feature activation-1.07e+01

Pos logit contributions
e+0.231
 seen+0.220
 coloured+0.208
 spe+0.205
 kinds+0.201
Neg logit contributions
orting-0.360
 Later-0.355
uing-0.339
 nicely-0.324
 Finally-0.309
Token who
Token ID508
Feature activation-3.84e+00

Pos logit contributions
e+0.310
 seen+0.293
 coloured+0.274
 spe+0.268
 kinds+0.262
Neg logit contributions
orting-0.545
 Later-0.535
uing-0.516
 nicely-0.493
 Finally-0.468
Token were
Token ID547
Feature activation-6.48e+00

Pos logit contributions
 comet+0.126
e+0.123
 seen+0.122
 spe+0.120
leaf+0.118
Neg logit contributions
 Later-0.164
orting-0.160
 nicely-0.156
uing-0.152
 Finally-0.147
Token very
Token ID845
Feature activation-5.19e+00

Pos logit contributions
e+0.184
 seen+0.184
 coloured+0.174
 comet+0.173
 spe+0.172
Neg logit contributions
 Later-0.314
orting-0.312
 nicely-0.299
uing-0.292
 Finally-0.278
Token excited
Token ID6568
Feature activation-1.73e+00

Pos logit contributions
e+0.170
 seen+0.166
 coloured+0.155
 spe+0.153
 comet+0.152
Neg logit contributions
orting-0.234
 Later-0.233
uing-0.222
 nicely-0.220
 Finally-0.203
Token.
Token ID13
Feature activation-2.91e+00

Pos logit contributions
e+0.088
 spe+0.086
ymes+0.082
 seen+0.082
 flown+0.082
Neg logit contributions
orting-0.047
 Later-0.046
 nicely-0.044
uing-0.040
 patiently-0.039
Token<|endoftext|>
Token ID50256
Feature activation-4.15e+00

Pos logit contributions
e+0.055
 seen+0.053
 sorts+0.050
 coloured+0.050
 wearing+0.050
Neg logit contributions
orting-0.034
uing-0.032
 Later-0.030
 nicely-0.027
lated-0.024
TokenOne
Token ID3198
Feature activation-2.35e+00

Pos logit contributions
e+0.230
 seen+0.229
 spe+0.225
 coloured+0.225
 kinds+0.220
Neg logit contributions
orting-0.092
 Later-0.092
 nicely-0.076
uing-0.072
 Finally-0.068
Token day
Token ID1110
Feature activation-2.74e+00

Pos logit contributions
leaf+0.088
ymes+0.086
 spe+0.086
keeping+0.084
e+0.084
Neg logit contributions
 Later-0.091
 nicely-0.087
orting-0.086
 Finally-0.074
uing-0.072
Token,
Token ID11
Feature activation-2.39e+00

Pos logit contributions
e+0.122
 seen+0.120
 spe+0.116
 coloured+0.115
 kinds+0.114
Neg logit contributions
orting-0.092
 Later-0.091
uing-0.085
 nicely-0.082
 Finally-0.074
Token a
Token ID257
Feature activation-9.79e+00

Pos logit contributions
e+0.052
 seen+0.051
 kinds+0.051
 sorts+0.050
 spe+0.050
Neg logit contributions
 Later-0.133
 nicely-0.128
orting-0.126
 Finally-0.120
uing-0.119
Token 3
Token ID513
Feature activation-1.06e+01

Pos logit contributions
 seen+0.359
e+0.354
 kinds+0.329
 sorts+0.326
 coloured+0.324
Neg logit contributions
orting-0.410
 Later-0.402
 nicely-0.387
uing-0.379
 Finally-0.347
Token year
Token ID614
Feature activation-8.39e+00

Pos logit contributions
 seen+0.313
leaf+0.301
 supposed+0.300
 spe+0.298
 flown+0.297
Neg logit contributions
 nicely-0.470
 Later-0.466
orting-0.458
uing-0.433
ail-0.409
Token old
Token ID1468
Feature activation-4.37e+00

Pos logit contributions
e+0.274
 seen+0.269
 coloured+0.261
 kinds+0.257
 sorts+0.255
Neg logit contributions
orting-0.381
uing-0.367
 Later-0.366
 nicely-0.330
 Finally-0.325
Token girl
Token ID2576
Feature activation-6.93e+00

Pos logit contributions
e+0.187
 seen+0.185
 supposed+0.182
leaf+0.174
 flown+0.174
Neg logit contributions
 nicely-0.143
 Later-0.140
orting-0.137
:"-0.116
uing-0.114
Token named
Token ID3706
Feature activation-8.97e+00

Pos logit contributions
leaf+0.331
ymes+0.316
 comet+0.299
 prey+0.288
odic+0.288
Neg logit contributions
 Later-0.204
 nicely-0.194
 Finally-0.171
:"-0.165
orting-0.157
Token Sally
Token ID25737
Feature activation-3.17e+00

Pos logit contributions
e+0.354
 seen+0.349
 sorts+0.319
 supposed+0.318
 coloured+0.316
Neg logit contributions
orting-0.386
uing-0.349
 Later-0.326
der-0.293
 nicely-0.289
Token time
Token ID640
Feature activation-3.33e+00

Pos logit contributions
e+0.166
 seen+0.154
 coloured+0.138
 spe+0.135
 kinds+0.128
Neg logit contributions
orting-0.526
 Later-0.516
uing-0.501
 nicely-0.485
 Finally-0.462
Token,
Token ID11
Feature activation-1.63e+00

Pos logit contributions
e+0.128
 seen+0.122
 coloured+0.113
 spe+0.113
 wearing+0.112
Neg logit contributions
orting-0.138
 Later-0.134
uing-0.130
 nicely-0.118
 Finally-0.113
Token there
Token ID612
Feature activation-9.33e-01

Pos logit contributions
e+0.040
 seen+0.038
 spe+0.035
 coloured+0.035
leaf+0.035
Neg logit contributions
orting-0.088
 Later-0.087
uing-0.083
 nicely-0.082
 Finally-0.077
Token were
Token ID547
Feature activation-7.20e+00

Pos logit contributions
e+0.034
 seen+0.033
 spe+0.031
 coloured+0.031
leaf+0.031
Neg logit contributions
orting-0.039
 Later-0.039
uing-0.036
 nicely-0.036
 Finally-0.033
Token twin
Token ID15203
Feature activation-5.63e+00

Pos logit contributions
e+0.242
 spe+0.239
 seen+0.238
leaf+0.235
 coloured+0.233
Neg logit contributions
 Later-0.313
orting-0.298
 nicely-0.286
 Finally-0.267
 Eventually-0.256
Token cats
Token ID11875
Feature activation-1.06e+01

Pos logit contributions
 seen+0.205
e+0.203
 spe+0.196
leaf+0.193
 coloured+0.192
Neg logit contributions
orting-0.233
 Later-0.224
 nicely-0.215
uing-0.214
 Finally-0.185
Token named
Token ID3706
Feature activation-8.16e+00

Pos logit contributions
 seen+0.376
 spe+0.371
e+0.369
 comet+0.364
 coloured+0.363
Neg logit contributions
orting-0.449
 Later-0.431
 nicely-0.419
uing-0.391
 Finally-0.362
Token Tom
Token ID4186
Feature activation-2.87e+00

Pos logit contributions
e+0.359
 seen+0.352
 coloured+0.334
 spe+0.333
 kinds+0.327
Neg logit contributions
orting-0.285
 Later-0.280
uing-0.259
 nicely-0.251
 Finally-0.229
Token and
Token ID290
Feature activation-3.21e+00

Pos logit contributions
e+0.103
 seen+0.098
 coloured+0.091
 kinds+0.088
 wearing+0.088
Neg logit contributions
orting-0.128
 Later-0.126
uing-0.122
 nicely-0.113
 Finally-0.110
Token Tim
Token ID5045
Feature activation-2.62e+00

Pos logit contributions
e+0.167
 seen+0.162
 coloured+0.160
ymes+0.158
 comet+0.157
Neg logit contributions
 nicely-0.069
orting-0.069
 Later-0.066
uing-0.056
 Finally-0.053
Token.
Token ID13
Feature activation-2.80e+00

Pos logit contributions
e+0.105
 seen+0.102
 coloured+0.097
 spe+0.096
 kinds+0.094
Neg logit contributions
orting-0.103
 Later-0.100
uing-0.095
 nicely-0.091
 Finally-0.084
Token time
Token ID640
Feature activation-4.38e+00

Pos logit contributions
e+0.184
 seen+0.168
 coloured+0.150
 spe+0.148
 kinds+0.139
Neg logit contributions
orting-0.588
 Later-0.576
uing-0.561
 nicely-0.541
 Finally-0.517
Token,
Token ID11
Feature activation-2.56e+00

Pos logit contributions
e+0.160
 seen+0.151
 wearing+0.143
 ever+0.141
 spe+0.140
Neg logit contributions
orting-0.190
 Later-0.185
uing-0.180
 nicely-0.162
 Finally-0.158
Token there
Token ID612
Feature activation-1.81e+00

Pos logit contributions
e+0.053
 seen+0.050
 coloured+0.045
 spe+0.044
 comet+0.042
Neg logit contributions
orting-0.151
 Later-0.147
uing-0.143
 nicely-0.137
 Finally-0.132
Token was
Token ID373
Feature activation-7.80e+00

Pos logit contributions
e+0.062
 seen+0.060
 coloured+0.056
 spe+0.056
 kinds+0.055
Neg logit contributions
orting-0.081
 Later-0.079
uing-0.076
 nicely-0.073
 Finally-0.068
Token a
Token ID257
Feature activation-9.83e+00

Pos logit contributions
 kinds+0.275
ymes+0.270
 sorts+0.265
 spe+0.264
leaf+0.256
Neg logit contributions
 Later-0.332
 nicely-0.300
 Finally-0.293
Their-0.274
 Eventually-0.272
Token 3
Token ID513
Feature activation-1.06e+01

Pos logit contributions
 seen+0.374
e+0.371
 kinds+0.342
 sorts+0.337
 coloured+0.336
Neg logit contributions
orting-0.394
 Later-0.391
 nicely-0.382
uing-0.367
 Finally-0.332
Token year
Token ID614
Feature activation-8.18e+00

Pos logit contributions
 seen+0.292
 spe+0.278
leaf+0.273
 flown+0.273
 supposed+0.271
Neg logit contributions
 Later-0.491
 nicely-0.489
orting-0.488
uing-0.462
ail-0.427
Token old
Token ID1468
Feature activation-3.98e+00

Pos logit contributions
e+0.266
 seen+0.257
 coloured+0.253
 spe+0.250
 kinds+0.247
Neg logit contributions
orting-0.374
 Later-0.360
uing-0.357
 nicely-0.322
 Finally-0.322
Token named
Token ID3706
Feature activation-8.86e+00

Pos logit contributions
e+0.156
 seen+0.155
 supposed+0.149
 flown+0.146
leaf+0.142
Neg logit contributions
 nicely-0.145
orting-0.144
 Later-0.144
 patiently-0.122
uing-0.122
Token Anna
Token ID11735
Feature activation-3.58e+00

Pos logit contributions
e+0.313
 seen+0.302
 wearing+0.283
 coloured+0.280
 supposed+0.280
Neg logit contributions
orting-0.400
uing-0.377
 Later-0.351
izes-0.315
 nicely-0.312
Token.
Token ID13
Feature activation-3.68e+00

Pos logit contributions
leaf+0.173
 spe+0.170
 seen+0.162
 comet+0.162
vous+0.161
Neg logit contributions
 Later-0.099
 nicely-0.094
orting-0.092
uing-0.083
 Finally-0.072
Token upon
Token ID2402
Feature activation-3.37e+00

Pos logit contributions
orting+0.018
 Later+0.017
uing+0.016
rier+0.015
der+0.015
Neg logit contributions
 seen-0.010
 wearing-0.009
 happened-0.009
 ever-0.009
 kinds-0.009
Token a
Token ID257
Feature activation-8.45e+00

Pos logit contributions
 spe+0.130
leaf+0.115
 kinds+0.111
 sorts+0.110
 comet+0.109
Neg logit contributions
 nicely-0.132
 Later-0.129
 Finally-0.114
His-0.110
 He-0.110
Token time
Token ID640
Feature activation-2.94e+00

Pos logit contributions
e+0.169
 seen+0.158
 coloured+0.144
 spe+0.142
 kinds+0.136
Neg logit contributions
orting-0.498
 Later-0.491
uing-0.473
 nicely-0.461
 Finally-0.439
Token,
Token ID11
Feature activation-1.10e+00

Pos logit contributions
e+0.107
 seen+0.101
 wearing+0.096
 spe+0.094
 coloured+0.094
Neg logit contributions
orting-0.127
 Later-0.125
uing-0.120
 nicely-0.108
 Finally-0.105
Token two
Token ID734
Feature activation-7.03e+00

Pos logit contributions
e+0.027
 seen+0.026
 spe+0.024
 coloured+0.024
leaf+0.024
Neg logit contributions
orting-0.059
 Later-0.059
 nicely-0.056
uing-0.056
 Finally-0.052
Token brothers
Token ID9397
Feature activation-1.06e+01

Pos logit contributions
e+0.228
 seen+0.221
 coloured+0.209
 spe+0.207
leaf+0.203
Neg logit contributions
orting-0.325
 Later-0.317
uing-0.303
 nicely-0.295
 Finally-0.274
Token named
Token ID3706
Feature activation-7.89e+00

Pos logit contributions
 seen+0.366
 spe+0.364
e+0.362
 comet+0.362
leaf+0.360
Neg logit contributions
orting-0.440
 Later-0.435
 nicely-0.403
uing-0.400
 Finally-0.373
Token Jack
Token ID3619
Feature activation-2.30e+00

Pos logit contributions
e+0.316
 seen+0.303
 coloured+0.290
 spe+0.285
 kinds+0.283
Neg logit contributions
orting-0.311
 Later-0.300
uing-0.288
 nicely-0.272
 Finally-0.252
Token and
Token ID290
Feature activation-3.26e+00

Pos logit contributions
 spe+0.098
 comet+0.096
e+0.095
leaf+0.095
ymes+0.093
Neg logit contributions
 Later-0.077
orting-0.076
uing-0.070
 nicely-0.068
 Finally-0.063
Token Daniel
Token ID7806
Feature activation-1.85e+00

Pos logit contributions
e+0.146
 spe+0.141
 seen+0.141
ymes+0.138
 coloured+0.137
Neg logit contributions
 Later-0.097
orting-0.097
 nicely-0.095
uing-0.090
 Finally-0.083
Token had
Token ID550
Feature activation-5.36e+00

Pos logit contributions
e+0.075
 seen+0.073
 spe+0.072
 comet+0.070
 coloured+0.070
Neg logit contributions
orting-0.069
 Later-0.067
uing-0.064
 nicely-0.062
 Finally-0.057
Token<|endoftext|>
Token ID50256
Feature activation-4.74e+00

Pos logit contributions
e+0.109
 seen+0.105
 coloured+0.100
 spe+0.100
 sorts+0.100
Neg logit contributions
orting-0.062
uing-0.057
 Later-0.056
 nicely-0.050
 Finally-0.043
TokenBen
Token ID11696
Feature activation-4.69e+00

Pos logit contributions
 seen+0.267
 spe+0.265
e+0.264
 coloured+0.264
leaf+0.260
Neg logit contributions
 Later-0.100
orting-0.098
 nicely-0.081
 Finally-0.071
uing-0.069
Token and
Token ID290
Feature activation-5.41e+00

Pos logit contributions
e+0.183
 seen+0.177
 spe+0.174
 coloured+0.170
leaf+0.167
Neg logit contributions
orting-0.184
 Later-0.182
uing-0.168
 nicely-0.165
 Finally-0.152
Token Sam
Token ID3409
Feature activation-4.33e+00

Pos logit contributions
 coloured+0.237
e+0.234
 kinds+0.233
 sorts+0.231
ymes+0.230
Neg logit contributions
 Later-0.172
orting-0.164
 nicely-0.161
 Finally-0.141
uing-0.136
Token were
Token ID547
Feature activation-7.88e+00

Pos logit contributions
e+0.145
 seen+0.140
 coloured+0.132
 spe+0.131
 kinds+0.128
Neg logit contributions
orting-0.196
 Later-0.192
uing-0.183
 nicely-0.177
 Finally-0.165
Token brothers
Token ID9397
Feature activation-1.05e+01

Pos logit contributions
e+0.243
 seen+0.233
 spe+0.233
 coloured+0.230
 kinds+0.222
Neg logit contributions
 Later-0.375
orting-0.371
 nicely-0.344
uing-0.342
 Finally-0.325
Token who
Token ID508
Feature activation-3.80e+00

Pos logit contributions
e+0.351
 seen+0.328
 coloured+0.304
 wearing+0.301
 spe+0.297
Neg logit contributions
orting-0.499
 Later-0.486
uing-0.476
 nicely-0.432
 Finally-0.421
Token liked
Token ID8288
Feature activation-5.02e+00

Pos logit contributions
e+0.117
 seen+0.111
 coloured+0.104
 spe+0.102
 kinds+0.100
Neg logit contributions
orting-0.186
 Later-0.181
uing-0.174
 nicely-0.166
 Finally-0.157
Token to
Token ID284
Feature activation-4.55e+00

Pos logit contributions
e+0.290
vous+0.259
 seen+0.252
odic+0.249
 spe+0.246
Neg logit contributions
 nicely-0.123
orting-0.097
 Later-0.096
 patiently-0.090
uing-0.080
Token play
Token ID711
Feature activation-3.00e+00

Pos logit contributions
ymes+0.185
 seen+0.181
 supposed+0.175
 spe+0.174
leaf+0.172
Neg logit contributions
 nicely-0.159
 Later-0.157
orting-0.153
uing-0.153
 Finally-0.132
Token with
Token ID351
Feature activation-2.20e+00

Pos logit contributions
 seen+0.094
e+0.093
 spe+0.088
 kinds+0.085
 supposed+0.084
Neg logit contributions
orting-0.144
 Later-0.138
 nicely-0.135
uing-0.135
 Finally-0.121
Token<|endoftext|>
Token ID50256
Feature activation-4.75e+00

Pos logit contributions
e+0.081
 seen+0.076
 sorts+0.076
 universe+0.076
 wearing+0.075
Neg logit contributions
uing-0.060
orting-0.058
 Later-0.047
 nicely-0.046
lated-0.045
TokenTom
Token ID13787
Feature activation-4.29e+00

Pos logit contributions
 seen+0.269
 spe+0.267
e+0.266
 coloured+0.265
leaf+0.260
Neg logit contributions
orting-0.098
 Later-0.097
 nicely-0.081
 Finally-0.070
uing-0.067
Token and
Token ID290
Feature activation-5.18e+00

Pos logit contributions
e+0.183
 seen+0.177
 coloured+0.168
 kinds+0.164
 spe+0.163
Neg logit contributions
orting-0.160
 Later-0.154
uing-0.149
 nicely-0.139
 Finally-0.129
Token Sam
Token ID3409
Feature activation-4.30e+00

Pos logit contributions
 coloured+0.223
e+0.221
 spe+0.218
 kinds+0.217
ymes+0.216
Neg logit contributions
 Later-0.171
orting-0.164
 nicely-0.157
 Finally-0.141
uing-0.138
Token were
Token ID547
Feature activation-7.89e+00

Pos logit contributions
e+0.144
 seen+0.138
 coloured+0.131
 spe+0.129
 kinds+0.126
Neg logit contributions
orting-0.196
 Later-0.192
uing-0.184
 nicely-0.176
 Finally-0.166
Token brothers
Token ID9397
Feature activation-1.05e+01

Pos logit contributions
e+0.256
 seen+0.244
 spe+0.236
 coloured+0.236
 kinds+0.228
Neg logit contributions
orting-0.367
 Later-0.363
uing-0.341
 nicely-0.334
 Finally-0.314
Token who
Token ID508
Feature activation-3.66e+00

Pos logit contributions
e+0.360
 seen+0.335
 wearing+0.310
 coloured+0.307
 spe+0.301
Neg logit contributions
orting-0.494
 Later-0.482
uing-0.471
 nicely-0.421
 Finally-0.415
Token liked
Token ID8288
Feature activation-4.91e+00

Pos logit contributions
e+0.114
 seen+0.109
 coloured+0.103
 spe+0.102
 kinds+0.099
Neg logit contributions
orting-0.176
 Later-0.172
uing-0.165
 nicely-0.159
 Finally-0.149
Token to
Token ID284
Feature activation-4.40e+00

Pos logit contributions
e+0.288
vous+0.259
 seen+0.249
odic+0.248
ymes+0.246
Neg logit contributions
 nicely-0.117
 Later-0.091
orting-0.090
 patiently-0.085
 "-0.072
Token gaze
Token ID17841
Feature activation+1.34e+00

Pos logit contributions
ymes+0.187
 seen+0.179
 supposed+0.177
 spe+0.173
leaf+0.171
Neg logit contributions
 nicely-0.148
 Later-0.143
orting-0.139
uing-0.139
 Finally-0.121
Token at
Token ID379
Feature activation+1.19e+00

Pos logit contributions
orting+0.053
 Later+0.051
uing+0.049
 nicely+0.046
 Finally+0.043
Neg logit contributions
e-0.054
 seen-0.052
 coloured-0.050
 spe-0.049
 kinds-0.048
Token time
Token ID640
Feature activation-3.36e+00

Pos logit contributions
e+0.176
 seen+0.165
 coloured+0.150
 spe+0.148
 kinds+0.142
Neg logit contributions
orting-0.520
 Later-0.513
uing-0.494
 nicely-0.481
 Finally-0.459
Token,
Token ID11
Feature activation-1.58e+00

Pos logit contributions
e+0.131
 seen+0.126
 spe+0.119
 coloured+0.119
 wearing+0.116
Neg logit contributions
orting-0.136
 Later-0.132
uing-0.127
 nicely-0.116
 Finally-0.110
Token there
Token ID612
Feature activation-9.79e-01

Pos logit contributions
e+0.043
leaf+0.043
 seen+0.042
 kinds+0.041
 sorts+0.040
Neg logit contributions
 Later-0.079
orting-0.077
 nicely-0.076
uing-0.073
 Finally-0.070
Token were
Token ID547
Feature activation-7.20e+00

Pos logit contributions
e+0.036
 seen+0.035
leaf+0.034
 spe+0.034
 coloured+0.033
Neg logit contributions
 Later-0.040
orting-0.040
uing-0.037
 nicely-0.037
 Finally-0.034
Token two
Token ID734
Feature activation-6.81e+00

Pos logit contributions
e+0.254
leaf+0.248
 sorts+0.247
 spe+0.247
 seen+0.246
Neg logit contributions
 Later-0.297
orting-0.278
 nicely-0.275
 Finally-0.253
 Eventually-0.242
Token brothers
Token ID9397
Feature activation-1.05e+01

Pos logit contributions
e+0.223
 seen+0.215
 coloured+0.202
 spe+0.200
 sorts+0.195
Neg logit contributions
orting-0.315
 Later-0.308
uing-0.295
 nicely-0.286
 Finally-0.266
Token.
Token ID13
Feature activation-3.04e+00

Pos logit contributions
e+0.294
 seen+0.267
 coloured+0.249
 spe+0.240
 kinds+0.236
Neg logit contributions
orting-0.550
 Later-0.541
uing-0.529
 nicely-0.492
 Finally-0.476
Token The
Token ID383
Feature activation-7.26e-01

Pos logit contributions
e+0.164
 seen+0.162
 spe+0.156
 coloured+0.156
 comet+0.156
Neg logit contributions
 Later-0.074
orting-0.072
 nicely-0.061
uing-0.058
 Finally-0.056
Token brothers
Token ID9397
Feature activation-1.03e+01

Pos logit contributions
e+0.024
 seen+0.023
 coloured+0.021
 spe+0.021
 kinds+0.020
Neg logit contributions
orting-0.034
 Later-0.033
uing-0.032
 nicely-0.031
 Finally-0.029
Token were
Token ID547
Feature activation-6.43e+00

Pos logit contributions
e+0.317
 seen+0.305
 spe+0.286
 coloured+0.285
 kinds+0.276
Neg logit contributions
orting-0.498
 Later-0.488
uing-0.467
 nicely-0.453
 Finally-0.426
Token always
Token ID1464
Feature activation+2.45e-02

Pos logit contributions
 seen+0.219
e+0.210
 comet+0.204
leaf+0.202
ymes+0.201
Neg logit contributions
 Later-0.283
orting-0.272
 nicely-0.269
uing-0.250
 Finally-0.248
Token<|endoftext|>
Token ID50256
Feature activation-4.42e+00

Pos logit contributions
e+0.101
 seen+0.096
 sorts+0.095
 wearing+0.095
 kinds+0.093
Neg logit contributions
uing-0.057
orting-0.057
 Later-0.047
 nicely-0.043
ail-0.042
TokenTom
Token ID13787
Feature activation-4.01e+00

Pos logit contributions
e+0.253
 seen+0.253
 spe+0.250
 coloured+0.249
 comet+0.243
Neg logit contributions
orting-0.089
 Later-0.088
 nicely-0.072
uing-0.065
 Finally-0.062
Token and
Token ID290
Feature activation-4.92e+00

Pos logit contributions
e+0.169
 seen+0.164
 coloured+0.155
 kinds+0.152
 sorts+0.150
Neg logit contributions
orting-0.152
 Later-0.146
uing-0.142
 nicely-0.132
 Finally-0.122
Token Sam
Token ID3409
Feature activation-4.17e+00

Pos logit contributions
e+0.217
 coloured+0.215
 spe+0.212
 seen+0.211
ymes+0.210
Neg logit contributions
 Later-0.158
orting-0.152
 nicely-0.144
 Finally-0.130
uing-0.127
Token were
Token ID547
Feature activation-7.64e+00

Pos logit contributions
e+0.136
 seen+0.131
 coloured+0.123
 spe+0.122
 kinds+0.119
Neg logit contributions
orting-0.193
 Later-0.189
uing-0.181
 nicely-0.174
 Finally-0.163
Token brothers
Token ID9397
Feature activation-1.05e+01

Pos logit contributions
e+0.237
 seen+0.227
 spe+0.224
 coloured+0.222
 comet+0.214
Neg logit contributions
 Later-0.363
orting-0.361
uing-0.334
 nicely-0.333
 Finally-0.315
Token who
Token ID508
Feature activation-3.59e+00

Pos logit contributions
e+0.344
 seen+0.317
 wearing+0.293
 coloured+0.290
 ever+0.280
Neg logit contributions
orting-0.505
 Later-0.493
uing-0.486
arl-0.429
 nicely-0.428
Token liked
Token ID8288
Feature activation-4.77e+00

Pos logit contributions
e+0.108
 seen+0.102
 coloured+0.096
 spe+0.093
 kinds+0.092
Neg logit contributions
orting-0.179
 Later-0.174
uing-0.169
 nicely-0.159
 Finally-0.151
Token to
Token ID284
Feature activation-4.33e+00

Pos logit contributions
e+0.279
vous+0.249
 seen+0.241
odic+0.241
 spe+0.238
Neg logit contributions
 nicely-0.115
 Later-0.090
orting-0.090
 patiently-0.083
uing-0.072
Token play
Token ID711
Feature activation-2.72e+00

Pos logit contributions
ymes+0.178
 seen+0.174
 supposed+0.170
 spe+0.168
leaf+0.166
Neg logit contributions
 nicely-0.150
 Later-0.146
orting-0.143
uing-0.143
 Finally-0.124
Token with
Token ID351
Feature activation-1.97e+00

Pos logit contributions
 seen+0.087
e+0.086
 spe+0.082
 kinds+0.078
 supposed+0.078
Neg logit contributions
orting-0.129
 Later-0.124
 nicely-0.121
uing-0.120
 Finally-0.108
Token a
Token ID257
Feature activation-9.91e+00

Pos logit contributions
e+0.050
 seen+0.047
 spe+0.043
 coloured+0.042
 kinds+0.042
Neg logit contributions
 Later-0.146
orting-0.146
uing-0.139
 nicely-0.138
 Finally-0.131
Token Mom
Token ID11254
Feature activation-5.64e+00

Pos logit contributions
e+0.334
 seen+0.330
 coloured+0.305
 kinds+0.302
 sorts+0.300
Neg logit contributions
orting-0.445
 Later-0.436
uing-0.414
 nicely-0.410
 Finally-0.377
Tokenmy
Token ID1820
Feature activation-4.66e+00

Pos logit contributions
e+0.253
 spe+0.252
leaf+0.251
 seen+0.250
ymes+0.249
Neg logit contributions
 Later-0.173
orting-0.170
 nicely-0.147
uing-0.138
 Finally-0.133
Token and
Token ID290
Feature activation-5.16e+00

Pos logit contributions
leaf+0.199
ymes+0.197
 musical+0.194
 spe+0.194
 comet+0.191
Neg logit contributions
 Later-0.146
orting-0.141
 nicely-0.127
uing-0.116
 Finally-0.115
Token a
Token ID257
Feature activation-9.27e+00

Pos logit contributions
e+0.197
 seen+0.179
 coloured+0.176
 spe+0.176
 sorts+0.172
Neg logit contributions
 Later-0.203
 nicely-0.191
orting-0.189
uing-0.173
 Finally-0.170
Token 3
Token ID513
Feature activation-1.05e+01

Pos logit contributions
e+0.359
 seen+0.356
 kinds+0.328
 coloured+0.326
 sorts+0.325
Neg logit contributions
orting-0.360
 Later-0.358
 nicely-0.353
uing-0.332
 Finally-0.315
Token year
Token ID614
Feature activation-7.92e+00

Pos logit contributions
 seen+0.260
 comet+0.253
 spe+0.252
leaf+0.247
 universe+0.246
Neg logit contributions
 Later-0.504
orting-0.500
 nicely-0.497
uing-0.483
 Finally-0.446
Token old
Token ID1468
Feature activation-3.96e+00

Pos logit contributions
 seen+0.229
 kinds+0.225
e+0.224
 coloured+0.222
 sorts+0.220
Neg logit contributions
orting-0.379
uing-0.369
 Later-0.368
 Finally-0.332
 nicely-0.325
Token girl
Token ID2576
Feature activation-6.47e+00

Pos logit contributions
e+0.171
 seen+0.165
 supposed+0.161
 flown+0.157
leaf+0.154
Neg logit contributions
 nicely-0.128
 Later-0.126
orting-0.126
:"-0.108
 Finally-0.107
Token went
Token ID1816
Feature activation-3.87e+00

Pos logit contributions
leaf+0.254
 comet+0.249
ymes+0.242
odic+0.233
 seek+0.222
Neg logit contributions
 nicely-0.233
 Later-0.232
orting-0.214
 Finally-0.212
:"-0.204
Token for
Token ID329
Feature activation-3.52e+00

Pos logit contributions
e+0.134
 seen+0.129
 coloured+0.122
 spe+0.120
 kinds+0.118
Neg logit contributions
orting-0.173
 Later-0.169
uing-0.162
 nicely-0.155
 Finally-0.145
TokenOne
Token ID3198
Feature activation-1.43e+00

Pos logit contributions
e+0.005
 seen+0.005
 sorts+0.005
 coloured+0.005
 wearing+0.005
Neg logit contributions
orting-0.004
uing-0.003
 Later-0.003
 nicely-0.003
der-0.003
Token day
Token ID1110
Feature activation-2.49e+00

Pos logit contributions
e+0.052
 seen+0.049
leaf+0.049
ymes+0.048
keeping+0.048
Neg logit contributions
 Later-0.059
 nicely-0.058
orting-0.057
 Finally-0.050
uing-0.050
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
e+0.120
 seen+0.120
 kinds+0.119
 spe+0.119
vous+0.119
Neg logit contributions
orting-0.068
 Later-0.066
 nicely-0.064
uing-0.057
 Finally-0.056
Token a
Token ID257
Feature activation-9.15e+00

Pos logit contributions
e+0.042
 seen+0.041
 spe+0.040
 kinds+0.040
 coloured+0.040
Neg logit contributions
 Later-0.035
orting-0.033
 nicely-0.033
uing-0.030
 Finally-0.030
Token little
Token ID1310
Feature activation-7.10e+00

Pos logit contributions
e+0.389
 seen+0.383
 kinds+0.366
 sorts+0.355
 supposed+0.349
Neg logit contributions
 nicely-0.340
orting-0.321
 Later-0.313
uing-0.296
 Finally-0.284
Token 3
Token ID513
Feature activation-1.05e+01

Pos logit contributions
 supposed+0.339
e+0.332
 seen+0.324
keeping+0.300
leaf+0.296
Neg logit contributions
orting-0.217
 nicely-0.213
 Later-0.181
 patiently-0.180
rier-0.173
Token year
Token ID614
Feature activation-7.92e+00

Pos logit contributions
 seen+0.293
 spe+0.291
leaf+0.288
 universe+0.288
 comet+0.287
Neg logit contributions
 Later-0.472
 nicely-0.465
orting-0.463
uing-0.452
:"-0.419
Token old
Token ID1468
Feature activation-3.89e+00

Pos logit contributions
 seen+0.235
 coloured+0.232
e+0.230
 kinds+0.222
 spe+0.222
Neg logit contributions
orting-0.371
uing-0.362
 Later-0.358
 nicely-0.324
 Finally-0.324
Token girl
Token ID2576
Feature activation-6.23e+00

Pos logit contributions
 supposed+0.160
e+0.150
leaf+0.145
vous+0.143
 seen+0.143
Neg logit contributions
 nicely-0.133
 Later-0.132
orting-0.130
:"-0.121
 patiently-0.116
Token arrived
Token ID5284
Feature activation-7.36e-01

Pos logit contributions
leaf+0.248
ymes+0.246
 comet+0.238
 kinds+0.237
 sorts+0.232
Neg logit contributions
 Later-0.217
 nicely-0.204
:"-0.197
orting-0.190
 Finally-0.188
Token at
Token ID379
Feature activation-1.04e+00

Pos logit contributions
e+0.030
 seen+0.028
 spe+0.028
 coloured+0.027
 kinds+0.027
Neg logit contributions
orting-0.029
 Later-0.028
uing-0.026
 nicely-0.026
 Finally-0.024
Token upon
Token ID2402
Feature activation-3.55e+00

Pos logit contributions
 seen+0.002
 wearing+0.001
 ever+0.001
 happened+0.001
 kinds+0.001
Neg logit contributions
orting-0.003
uing-0.002
 Later-0.002
der-0.002
rier-0.002
Token a
Token ID257
Feature activation-8.52e+00

Pos logit contributions
 spe+0.128
leaf+0.114
 sorts+0.112
 comet+0.111
 kinds+0.110
Neg logit contributions
 nicely-0.145
 Later-0.143
 Finally-0.127
orting-0.126
His-0.126
Token time
Token ID640
Feature activation-3.08e+00

Pos logit contributions
e+0.165
 seen+0.153
 coloured+0.138
 spe+0.135
 kinds+0.129
Neg logit contributions
orting-0.510
 Later-0.500
uing-0.485
 nicely-0.470
 Finally-0.448
Token,
Token ID11
Feature activation-1.30e+00

Pos logit contributions
e+0.117
 seen+0.112
 spe+0.105
 coloured+0.104
 wearing+0.103
Neg logit contributions
orting-0.129
 Later-0.124
uing-0.121
 nicely-0.109
 Finally-0.104
Token two
Token ID734
Feature activation-6.90e+00

Pos logit contributions
e+0.033
 seen+0.031
leaf+0.030
 kinds+0.030
 spe+0.029
Neg logit contributions
 Later-0.069
orting-0.068
 nicely-0.065
uing-0.065
 Finally-0.061
Token brothers
Token ID9397
Feature activation-1.04e+01

Pos logit contributions
e+0.228
 seen+0.221
 coloured+0.210
 spe+0.208
leaf+0.206
Neg logit contributions
orting-0.314
 Later-0.306
uing-0.292
 nicely-0.285
 Finally-0.264
Token Jack
Token ID3619
Feature activation-2.16e+00

Pos logit contributions
 spe+0.361
 comet+0.358
 seen+0.357
leaf+0.357
e+0.350
Neg logit contributions
orting-0.434
 Later-0.431
 nicely-0.400
uing-0.393
 Finally-0.371
Token and
Token ID290
Feature activation-3.24e+00

Pos logit contributions
 spe+0.101
 comet+0.101
ymes+0.100
 ghost+0.099
leaf+0.098
Neg logit contributions
 Later-0.062
orting-0.060
uing-0.053
 nicely-0.053
 Finally-0.049
Token Tom
Token ID4186
Feature activation-2.20e+00

Pos logit contributions
e+0.145
 spe+0.140
 seen+0.139
 comet+0.137
 coloured+0.135
Neg logit contributions
 Later-0.097
orting-0.095
 nicely-0.094
uing-0.090
 Finally-0.082
Token were
Token ID547
Feature activation-5.95e+00

Pos logit contributions
 comet+0.078
 spe+0.078
e+0.076
leaf+0.074
ymes+0.073
Neg logit contributions
orting-0.090
 Later-0.090
 nicely-0.083
uing-0.082
 Finally-0.078
Token walking
Token ID6155
Feature activation-1.79e+00

Pos logit contributions
 spe+0.147
 seen+0.147
 coloured+0.145
 comet+0.144
leaf+0.139
Neg logit contributions
 Later-0.312
orting-0.308
 nicely-0.295
uing-0.285
 Finally-0.282
Token a
Token ID257
Feature activation-8.89e+00

Pos logit contributions
 spe+0.141
leaf+0.126
 sorts+0.124
 kinds+0.123
 coloured+0.118
Neg logit contributions
 Later-0.163
 nicely-0.158
 Finally-0.142
orting-0.140
Finally-0.135
Token time
Token ID640
Feature activation-3.41e+00

Pos logit contributions
e+0.198
 seen+0.190
 coloured+0.177
 spe+0.177
 kinds+0.174
Neg logit contributions
orting-0.493
 Later-0.491
uing-0.464
 nicely-0.460
 Finally-0.439
Token there
Token ID612
Feature activation-1.03e+00

Pos logit contributions
e+0.130
 seen+0.125
 coloured+0.117
 spe+0.116
 wearing+0.115
Neg logit contributions
orting-0.141
 Later-0.138
uing-0.132
 nicely-0.121
 Finally-0.116
Token were
Token ID547
Feature activation-7.20e+00

Pos logit contributions
e+0.038
 seen+0.038
leaf+0.036
 spe+0.036
 coloured+0.035
Neg logit contributions
 Later-0.042
orting-0.041
uing-0.038
 nicely-0.038
 Finally-0.035
Token two
Token ID734
Feature activation-6.78e+00

Pos logit contributions
e+0.257
 seen+0.243
leaf+0.241
 spe+0.237
 sorts+0.233
Neg logit contributions
 Later-0.306
orting-0.294
 nicely-0.280
 Finally-0.259
uing-0.251
Token brothers
Token ID9397
Feature activation-1.04e+01

Pos logit contributions
e+0.220
 seen+0.211
 coloured+0.198
 spe+0.196
 sorts+0.191
Neg logit contributions
orting-0.317
 Later-0.309
uing-0.296
 nicely-0.287
 Finally-0.268
Token.
Token ID13
Feature activation-3.06e+00

Pos logit contributions
e+0.334
 seen+0.306
 coloured+0.284
 wearing+0.278
 spe+0.273
Neg logit contributions
orting-0.506
 Later-0.502
uing-0.496
 nicely-0.447
 Finally-0.438
Token One
Token ID1881
Feature activation-4.48e-01

Pos logit contributions
e+0.172
 seen+0.168
 spe+0.163
 coloured+0.162
 comet+0.161
Neg logit contributions
orting-0.069
 Later-0.068
uing-0.058
 nicely-0.057
 Finally-0.050
Token was
Token ID373
Feature activation-6.87e+00

Pos logit contributions
e+0.015
 seen+0.015
 spe+0.014
leaf+0.014
ymes+0.014
Neg logit contributions
 Later-0.020
orting-0.019
 nicely-0.019
uing-0.018
 Finally-0.017
Token called
Token ID1444
Feature activation-7.59e+00

Pos logit contributions
ymes+0.196
 seen+0.195
keeping+0.193
 comet+0.189
e+0.188
Neg logit contributions
 Later-0.330
 nicely-0.319
orting-0.312
 Finally-0.293
 patiently-0.286
Token Paul
Token ID3362
Feature activation-3.11e+00

Pos logit contributions
 seen+0.331
e+0.327
 coloured+0.312
 spe+0.311
 kinds+0.309
Neg logit contributions
orting-0.262
 Later-0.257
 nicely-0.246
uing-0.238
 Finally-0.215
Token upon
Token ID2402
Feature activation-4.95e+00

Pos logit contributions
 seen+0.054
 wearing+0.047
 nowhere+0.043
e+0.042
 ever+0.041
Neg logit contributions
orting-0.105
uing-0.096
 Later-0.090
rier-0.083
lated-0.083
Token a
Token ID257
Feature activation-1.01e+01

Pos logit contributions
 spe+0.192
leaf+0.176
 sorts+0.175
 kinds+0.169
ymes+0.162
Neg logit contributions
 Later-0.185
 nicely-0.184
orting-0.156
 Finally-0.154
 He-0.153
Token time
Token ID640
Feature activation-4.65e+00

Pos logit contributions
e+0.181
 seen+0.162
 spe+0.141
 coloured+0.141
 comet+0.129
Neg logit contributions
orting-0.628
 Later-0.614
uing-0.601
 nicely-0.575
 Finally-0.551
Token there
Token ID612
Feature activation-2.31e+00

Pos logit contributions
e+0.182
 seen+0.174
 coloured+0.163
 spe+0.162
 wearing+0.161
Neg logit contributions
orting-0.189
 Later-0.185
uing-0.178
 nicely-0.163
 Finally-0.156
Token lived
Token ID5615
Feature activation-4.87e+00

Pos logit contributions
e+0.076
 seen+0.072
 coloured+0.068
 spe+0.066
 kinds+0.065
Neg logit contributions
orting-0.110
 Later-0.106
uing-0.103
 nicely-0.097
 Finally-0.092
Token a
Token ID257
Feature activation-1.04e+01

Pos logit contributions
 kinds+0.116
 spe+0.114
 sorts+0.113
 seen+0.112
 comet+0.105
Neg logit contributions
 Later-0.265
 nicely-0.261
orting-0.260
 Finally-0.229
uing-0.229
Token little
Token ID1310
Feature activation-7.21e+00

Pos logit contributions
 seen+0.383
e+0.382
 kinds+0.356
 sorts+0.349
 coloured+0.348
Neg logit contributions
orting-0.425
 Later-0.424
 nicely-0.410
uing-0.392
 Finally-0.364
Token girl
Token ID2576
Feature activation-7.67e+00

Pos logit contributions
 seen+0.311
 supposed+0.297
e+0.293
leaf+0.278
 flown+0.272
Neg logit contributions
orting-0.261
 nicely-0.250
 Later-0.234
uing-0.210
arl-0.207
Token and
Token ID290
Feature activation-4.98e+00

Pos logit contributions
leaf+0.319
ymes+0.281
 kinds+0.281
 comet+0.280
 supposed+0.278
Neg logit contributions
 nicely-0.278
 Later-0.269
orting-0.258
 Finally-0.218
 patiently-0.216
Token her
Token ID607
Feature activation-1.59e+00

Pos logit contributions
e+0.191
leaf+0.185
ymes+0.179
 seen+0.178
 spe+0.174
Neg logit contributions
 Later-0.184
 nicely-0.182
orting-0.174
 Finally-0.157
 patiently-0.156
Token daddy
Token ID35822
Feature activation-4.57e+00

Pos logit contributions
 seen+0.059
e+0.059
 sorts+0.055
 kinds+0.055
 coloured+0.055
Neg logit contributions
orting-0.064
 nicely-0.063
 Later-0.063
uing-0.058
 Finally-0.056
Token upon
Token ID2402
Feature activation-4.95e+00

Pos logit contributions
 seen+0.054
 wearing+0.047
 nowhere+0.043
e+0.042
 ever+0.041
Neg logit contributions
orting-0.105
uing-0.096
 Later-0.090
rier-0.083
lated-0.083
Token a
Token ID257
Feature activation-1.01e+01

Pos logit contributions
 spe+0.192
leaf+0.176
 sorts+0.175
 kinds+0.169
ymes+0.162
Neg logit contributions
 Later-0.185
 nicely-0.184
orting-0.156
 Finally-0.154
 He-0.153
Token time
Token ID640
Feature activation-4.65e+00

Pos logit contributions
e+0.181
 seen+0.162
 spe+0.141
 coloured+0.141
 comet+0.129
Neg logit contributions
orting-0.628
 Later-0.614
uing-0.601
 nicely-0.575
 Finally-0.551
Token there
Token ID612
Feature activation-2.31e+00

Pos logit contributions
e+0.182
 seen+0.174
 coloured+0.163
 spe+0.162
 wearing+0.161
Neg logit contributions
orting-0.189
 Later-0.185
uing-0.178
 nicely-0.163
 Finally-0.156
Token lived
Token ID5615
Feature activation-4.87e+00

Pos logit contributions
e+0.076
 seen+0.072
 coloured+0.068
 spe+0.066
 kinds+0.065
Neg logit contributions
orting-0.110
 Later-0.106
uing-0.103
 nicely-0.097
 Finally-0.092
Token a
Token ID257
Feature activation-1.04e+01

Pos logit contributions
 kinds+0.116
 spe+0.114
 sorts+0.113
 seen+0.112
 comet+0.105
Neg logit contributions
 Later-0.265
 nicely-0.261
orting-0.260
 Finally-0.229
uing-0.229
Token brave
Token ID14802
Feature activation-5.51e+00

Pos logit contributions
 seen+0.383
e+0.382
 kinds+0.356
 sorts+0.349
 coloured+0.348
Neg logit contributions
orting-0.425
 Later-0.424
 nicely-0.410
uing-0.392
 Finally-0.364
Token boy
Token ID2933
Feature activation-7.93e+00

Pos logit contributions
 seen+0.219
e+0.210
 supposed+0.207
leaf+0.195
 flown+0.194
Neg logit contributions
orting-0.210
 nicely-0.206
 Later-0.201
uing-0.179
 patiently-0.176
Token called
Token ID1444
Feature activation-8.57e+00

Pos logit contributions
leaf+0.319
 comet+0.315
ymes+0.305
 universe+0.292
 supposed+0.286
Neg logit contributions
 nicely-0.284
orting-0.278
 Later-0.270
 Finally-0.220
 until-0.217
Token Dave
Token ID9935
Feature activation-1.69e+00

Pos logit contributions
e+0.376
 seen+0.372
 coloured+0.371
 spe+0.361
 kinds+0.360
Neg logit contributions
orting-0.290
 Later-0.283
 nicely-0.266
uing-0.245
 Finally-0.240
Token.
Token ID13
Feature activation-3.76e+00

Pos logit contributions
e+0.072
 spe+0.072
 seen+0.071
leaf+0.070
 coloured+0.069
Neg logit contributions
orting-0.056
 Later-0.055
 nicely-0.053
uing-0.051
 Finally-0.044
Token<|endoftext|>
Token ID50256
Feature activation-3.76e+00

Pos logit contributions
orting+0.060
 Later+0.057
 nicely+0.056
uing+0.055
 Finally+0.050
Neg logit contributions
e-0.048
 spe-0.047
 seen-0.044
 coloured-0.043
 comet-0.043
TokenOnce
Token ID7454
Feature activation-1.89e+00

Pos logit contributions
 spe+0.210
 kinds+0.201
 seen+0.200
 comet+0.200
 coloured+0.200
Neg logit contributions
 Later-0.087
orting-0.085
 nicely-0.072
 Finally-0.063
 Eventually-0.058
Token there
Token ID612
Feature activation-2.58e+00

Pos logit contributions
 seen+0.054
 wearing+0.047
 nowhere+0.043
e+0.042
 ever+0.041
Neg logit contributions
orting-0.105
uing-0.096
 Later-0.090
rier-0.083
lated-0.083
Token was
Token ID373
Feature activation-8.41e+00

Pos logit contributions
e+0.085
 seen+0.081
 wearing+0.071
 coloured+0.071
 ever+0.070
Neg logit contributions
orting-0.129
uing-0.122
 Later-0.121
arl-0.108
 nicely-0.107
Token a
Token ID257
Feature activation-1.05e+01

Pos logit contributions
ymes+0.281
 kinds+0.275
 sorts+0.267
 spe+0.265
leaf+0.248
Neg logit contributions
 Later-0.368
 nicely-0.348
 Finally-0.338
orting-0.334
 Eventually-0.319
Token three
Token ID1115
Feature activation-1.04e+01

Pos logit contributions
e+0.359
 seen+0.349
 coloured+0.325
 kinds+0.319
 spe+0.318
Neg logit contributions
orting-0.468
 Later-0.460
uing-0.438
 nicely-0.429
 Finally-0.396
Token-
Token ID12
Feature activation-7.19e+00

Pos logit contributions
e+0.305
 seen+0.302
 spe+0.276
 coloured+0.274
leaf+0.268
Neg logit contributions
orting-0.499
 Later-0.495
uing-0.469
 nicely-0.469
 Finally-0.429
Tokenyear
Token ID1941
Feature activation-5.41e+00

Pos logit contributions
e+0.222
 seen+0.215
 spe+0.213
 comet+0.208
 coloured+0.204
Neg logit contributions
orting-0.325
 Later-0.323
 nicely-0.308
uing-0.304
 Finally-0.281
Token-
Token ID12
Feature activation-4.63e+00

Pos logit contributions
e+0.189
 seen+0.182
 coloured+0.171
 spe+0.170
 sorts+0.165
Neg logit contributions
orting-0.238
 Later-0.233
uing-0.222
 nicely-0.214
 Finally-0.199
Tokenold
Token ID727
Feature activation-4.77e+00

Pos logit contributions
e+0.167
 seen+0.161
 spe+0.157
 coloured+0.153
 comet+0.152
Neg logit contributions
orting-0.195
 Later-0.191
uing-0.181
 nicely-0.178
 Finally-0.163
Token who
Token ID508
Feature activation-3.63e+00

Pos logit contributions
 comet+0.177
 supposed+0.174
 spe+0.174
 universe+0.171
e+0.171
Neg logit contributions
 nicely-0.173
orting-0.172
 Later-0.168
uing-0.154
 Finally-0.147
Token might
Token ID1244
Feature activation+1.87e-01

Pos logit contributions
e+0.177
 seen+0.173
 coloured+0.170
 spe+0.168
leaf+0.165
Neg logit contributions
orting-0.170
 Later-0.161
uing-0.157
 nicely-0.154
 Finally-0.136
Token break
Token ID2270
Feature activation+3.44e+00

Pos logit contributions
orting+0.008
 Later+0.008
uing+0.008
 nicely+0.008
 Finally+0.007
Neg logit contributions
e-0.006
 seen-0.006
 coloured-0.006
 kinds-0.006
 spe-0.006
Token it
Token ID340
Feature activation+7.14e+00

Pos logit contributions
orting+0.137
 Later+0.133
uing+0.128
 nicely+0.125
 Finally+0.112
Neg logit contributions
e-0.132
 seen-0.129
 spe-0.123
 coloured-0.122
 sorts-0.120
Token or
Token ID393
Feature activation+1.26e+00

Pos logit contributions
orting+0.311
 Later+0.304
uing+0.288
 nicely+0.266
 Finally+0.260
Neg logit contributions
e-0.256
 seen-0.243
 coloured-0.228
 spe-0.225
 kinds-0.224
Token swallow
Token ID26633
Feature activation+4.07e+00

Pos logit contributions
orting+0.052
 Later+0.051
 nicely+0.051
uing+0.049
 Finally+0.044
Neg logit contributions
e-0.046
 seen-0.042
 coloured-0.042
 sorts-0.042
 kinds-0.042
Token it
Token ID340
Feature activation+7.03e+00

Pos logit contributions
orting+0.181
 Later+0.178
uing+0.170
 nicely+0.160
 Finally+0.152
Neg logit contributions
e-0.142
 seen-0.135
 coloured-0.128
 spe-0.126
 kinds-0.124
Token.
Token ID13
Feature activation+1.28e+00

Pos logit contributions
orting+0.309
 Later+0.301
uing+0.285
 nicely+0.262
 Finally+0.259
Neg logit contributions
e-0.252
 seen-0.236
 coloured-0.225
 spe-0.221
 kinds-0.218
Token You
Token ID921
Feature activation-1.95e+00

Pos logit contributions
orting+0.061
uing+0.060
 Later+0.054
ail+0.052
 nicely+0.052
Neg logit contributions
e-0.042
 seen-0.037
 spe-0.037
 universe-0.035
 wearing-0.035
Token can
Token ID460
Feature activation-2.14e+00

Pos logit contributions
 coloured+0.077
e+0.074
 sorts+0.074
 kinds+0.073
 musical+0.073
Neg logit contributions
orting-0.069
 Later-0.067
 nicely-0.065
uing-0.063
 Finally-0.058
Token play
Token ID711
Feature activation-9.41e-01

Pos logit contributions
e+0.069
 seen+0.061
 coloured+0.057
 spe+0.057
 ever+0.056
Neg logit contributions
orting-0.106
 Later-0.103
uing-0.101
 nicely-0.093
 Finally-0.090
Token with
Token ID351
Feature activation+4.03e-01

Pos logit contributions
e+0.031
 seen+0.029
 coloured+0.028
 spe+0.027
 kinds+0.026
Neg logit contributions
orting-0.044
 Later-0.043
uing-0.041
 nicely-0.039
 Finally-0.037
Token hurt
Token ID5938
Feature activation+2.11e+00

Pos logit contributions
e+0.053
 seen+0.051
 coloured+0.051
leaf+0.051
 kinds+0.050
Neg logit contributions
orting-0.060
 Later-0.059
uing-0.057
 nicely-0.057
 Finally-0.050
Token it
Token ID340
Feature activation+6.89e+00

Pos logit contributions
orting+0.081
 nicely+0.076
 Later+0.075
uing+0.075
 Finally+0.063
Neg logit contributions
e-0.080
 seen-0.079
 coloured-0.077
leaf-0.077
 spe-0.077
Token,
Token ID11
Feature activation+3.99e+00

Pos logit contributions
orting+0.261
 Later+0.251
uing+0.243
 nicely+0.236
 Finally+0.210
Neg logit contributions
e-0.282
 seen-0.270
 spe-0.261
 coloured-0.261
leaf-0.258
Token or
Token ID393
Feature activation+2.75e-01

Pos logit contributions
orting+0.165
 Later+0.160
uing+0.153
 nicely+0.148
 Finally+0.137
Neg logit contributions
e-0.150
 seen-0.143
 coloured-0.138
 spe-0.137
 kinds-0.134
Token take
Token ID1011
Feature activation+1.29e+00

Pos logit contributions
 nicely+0.010
orting+0.010
 Later+0.009
uing+0.009
 Finally+0.008
Neg logit contributions
e-0.011
 coloured-0.011
 seen-0.010
vous-0.010
 spe-0.010
Token it
Token ID340
Feature activation+7.43e+00

Pos logit contributions
orting+0.057
 Later+0.056
uing+0.054
 nicely+0.054
 Finally+0.049
Neg logit contributions
e-0.042
 seen-0.042
 coloured-0.041
 spe-0.040
 comet-0.040
Token away
Token ID1497
Feature activation+6.83e+00

Pos logit contributions
orting+0.267
 Later+0.258
uing+0.245
 nicely+0.237
 Finally+0.213
Neg logit contributions
e-0.321
 seen-0.308
 coloured-0.298
 spe-0.298
 kinds-0.291
Token from
Token ID422
Feature activation+1.03e+00

Pos logit contributions
orting+0.273
 Later+0.265
uing+0.253
 nicely+0.246
 Finally+0.224
Neg logit contributions
e-0.266
 seen-0.256
 coloured-0.244
 spe-0.244
 kinds-0.237
Token her
Token ID607
Feature activation+3.73e+00

Pos logit contributions
orting+0.040
 Later+0.039
 nicely+0.038
uing+0.036
 Finally+0.033
Neg logit contributions
e-0.039
 seen-0.038
 spe-0.038
vous-0.037
 coloured-0.037
Token.
Token ID13
Feature activation+1.23e+00

Pos logit contributions
 Later+0.141
orting+0.141
 nicely+0.140
uing+0.132
 Finally+0.122
Neg logit contributions
e-0.144
 seen-0.139
 coloured-0.138
leaf-0.138
vous-0.135
Token She
Token ID1375
Feature activation+2.89e+00

Pos logit contributions
uing+0.045
orting+0.044
 nicely+0.036
izes+0.035
 Later+0.035
Neg logit contributions
e-0.052
 seen-0.050
 wearing-0.050
 sorts-0.048
 blonde-0.047
Token so
Token ID523
Feature activation-2.86e+00

Pos logit contributions
leaf+0.199
ymes+0.198
 comet+0.194
 coloured+0.191
 kinds+0.191
Neg logit contributions
 Later-0.183
 nicely-0.177
 Finally-0.152
 patiently-0.151
uing-0.148
Token excited
Token ID6568
Feature activation-8.02e-01

Pos logit contributions
e+0.091
 seen+0.088
 coloured+0.082
 spe+0.082
 kinds+0.079
Neg logit contributions
orting-0.138
 Later-0.134
uing-0.129
 nicely-0.122
 Finally-0.116
Token and
Token ID290
Feature activation-2.78e-01

Pos logit contributions
e+0.036
 spe+0.034
 seen+0.032
 comet+0.032
leaf+0.031
Neg logit contributions
orting-0.029
 Later-0.028
 nicely-0.026
uing-0.025
 Finally-0.023
Token he
Token ID339
Feature activation+3.76e+00

Pos logit contributions
e+0.010
 seen+0.010
 coloured+0.009
 spe+0.009
 kinds+0.009
Neg logit contributions
orting-0.012
 Later-0.012
uing-0.011
 nicely-0.011
 Finally-0.010
Token opened
Token ID4721
Feature activation+2.48e+00

Pos logit contributions
 Later+0.160
 nicely+0.147
orting+0.145
uing+0.142
 Finally+0.136
Neg logit contributions
e-0.136
 seen-0.127
 kinds-0.124
leaf-0.123
 spe-0.123
Token it
Token ID340
Feature activation+6.55e+00

Pos logit contributions
orting+0.129
 Later+0.121
uing+0.120
 nicely+0.113
arl+0.107
Neg logit contributions
e-0.071
 seen-0.070
 coloured-0.067
 spe-0.063
 kinds-0.063
Token quickly
Token ID2952
Feature activation+2.51e+00

Pos logit contributions
orting+0.307
uing+0.283
 Later+0.281
arl+0.263
 nicely+0.255
Neg logit contributions
 coloured-0.203
e-0.203
 seen-0.202
 sorts-0.195
 kinds-0.194
Token.
Token ID13
Feature activation+4.64e-01

Pos logit contributions
orting+0.100
 Later+0.098
uing+0.093
 nicely+0.089
 Finally+0.082
Neg logit contributions
e-0.099
 seen-0.095
 coloured-0.091
 spe-0.090
 kinds-0.088
Token Inside
Token ID14384
Feature activation-1.63e-02

Pos logit contributions
orting+0.018
uing+0.017
 nicely+0.014
 Later+0.014
arl+0.013
Neg logit contributions
 seen-0.019
e-0.019
 sorts-0.019
 kinds-0.019
 coloured-0.018
Token the
Token ID262
Feature activation+1.50e+00

Pos logit contributions
e+0.001
 seen+0.001
 coloured+0.001
 spe+0.000
 kinds+0.000
Neg logit contributions
orting-0.001
 Later-0.001
uing-0.001
 nicely-0.001
 Finally-0.001
Token box
Token ID3091
Feature activation-1.58e+00

Pos logit contributions
orting+0.066
 Later+0.066
 nicely+0.063
uing+0.061
 Finally+0.058
Neg logit contributions
e-0.053
 seen-0.049
leaf-0.048
 spe-0.047
 coloured-0.046
Token.
Token ID13
Feature activation+5.11e-02

Pos logit contributions
orting+0.001
 Later+0.000
uing+0.000
 nicely+0.000
 Finally+0.000
Neg logit contributions
 seen-0.001
e-0.001
leaf-0.001
 spe-0.001
 coloured-0.001
Token She
Token ID1375
Feature activation+2.00e+00

Pos logit contributions
uing+0.002
orting+0.002
izes+0.002
lated+0.002
rier+0.002
Neg logit contributions
 wearing-0.002
 seen-0.002
 sorts-0.002
e-0.002
 blonde-0.002
Token wiped
Token ID21122
Feature activation+4.98e+00

Pos logit contributions
orting+0.097
 Later+0.090
uing+0.088
 nicely+0.083
arl+0.079
Neg logit contributions
e-0.065
 seen-0.064
 coloured-0.062
 spe-0.059
 colored-0.058
Token his
Token ID465
Feature activation+4.85e+00

Pos logit contributions
orting+0.249
uing+0.242
 Later+0.239
arl+0.220
 nicely+0.216
Neg logit contributions
e-0.159
 coloured-0.146
 seen-0.140
 sorts-0.138
 colored-0.137
Token tears
Token ID10953
Feature activation+1.07e+00

Pos logit contributions
 Later+0.164
orting+0.160
uing+0.149
 nicely+0.148
 Finally+0.134
Neg logit contributions
e-0.213
 seen-0.208
 sorts-0.201
 spe-0.200
 coloured-0.199
Token away
Token ID1497
Feature activation+6.38e+00

Pos logit contributions
orting+0.045
 Later+0.044
uing+0.042
 nicely+0.041
 Finally+0.037
Neg logit contributions
e-0.039
 seen-0.038
 spe-0.036
 kinds-0.036
 coloured-0.036
Token and
Token ID290
Feature activation+1.67e+00

Pos logit contributions
orting+0.308
 Later+0.296
uing+0.285
 Finally+0.258
 nicely+0.257
Neg logit contributions
e-0.193
 coloured-0.190
 seen-0.185
 spe-0.185
 shaped-0.182
Token said
Token ID531
Feature activation+3.40e+00

Pos logit contributions
orting+0.072
 Later+0.070
uing+0.068
 nicely+0.064
 Finally+0.060
Neg logit contributions
e-0.060
 seen-0.060
 coloured-0.057
 spe-0.055
 sorts-0.054
Token,
Token ID11
Feature activation+3.78e+00

Pos logit contributions
 Later+0.170
orting+0.164
uing+0.153
lated+0.140
Finally+0.140
Neg logit contributions
e-0.100
 ever-0.094
 spe-0.094
 wearing-0.090
 seen-0.089
Token "
Token ID366
Feature activation+7.63e-01

Pos logit contributions
orting+0.156
uing+0.145
lated+0.136
udd+0.133
 Later+0.131
Neg logit contributions
 wearing-0.152
 flying-0.138
 singing-0.133
 running-0.132
 spe-0.132
TokenDon
Token ID3987
Feature activation+2.37e+00

Pos logit contributions
orting+0.035
 Later+0.034
uing+0.034
 nicely+0.032
 Finally+0.029
Neg logit contributions
e-0.026
 seen-0.025
 coloured-0.024
vous-0.023
 sorts-0.023
Token with
Token ID351
Feature activation+6.23e-01

Pos logit contributions
 seen+0.029
 kinds+0.028
 supposed+0.028
 sorts+0.027
 spe+0.027
Neg logit contributions
orting-0.036
 nicely-0.035
uing-0.033
 Later-0.033
 Finally-0.029
Token you
Token ID345
Feature activation+6.35e-02

Pos logit contributions
orting+0.026
 Later+0.026
uing+0.025
 nicely+0.024
 Finally+0.022
Neg logit contributions
e-0.023
 seen-0.022
 coloured-0.021
 spe-0.021
 kinds-0.020
Token when
Token ID618
Feature activation+2.19e+00

Pos logit contributions
orting+0.002
 Later+0.002
uing+0.002
 nicely+0.002
 Finally+0.002
Neg logit contributions
e-0.003
 seen-0.003
 coloured-0.003
 spe-0.003
 sorts-0.003
Token you
Token ID345
Feature activation-3.72e-01

Pos logit contributions
orting+0.077
 Later+0.077
 nicely+0.074
uing+0.069
 Finally+0.065
Neg logit contributions
e-0.091
 seen-0.088
 sorts-0.087
leaf-0.087
 coloured-0.087
Token get
Token ID651
Feature activation+1.96e+00

Pos logit contributions
e+0.013
 seen+0.012
 spe+0.012
 coloured+0.012
 comet+0.012
Neg logit contributions
orting-0.017
 Later-0.016
uing-0.015
 nicely-0.015
 Finally-0.014
Token better
Token ID1365
Feature activation+6.22e+00

Pos logit contributions
orting+0.081
 Later+0.078
 nicely+0.076
uing+0.074
 Finally+0.067
Neg logit contributions
e-0.072
 comet-0.068
 sorts-0.068
 spe-0.068
 kinds-0.067
Token."
Token ID526
Feature activation-1.35e+00

Pos logit contributions
orting+0.215
 Later+0.204
uing+0.199
 nicely+0.198
 Finally+0.167
Neg logit contributions
e-0.272
leaf-0.261
 sorts-0.254
vous-0.254
 spe-0.253
Token Mia
Token ID32189
Feature activation-1.41e+00

Pos logit contributions
e+0.062
 seen+0.062
 sorts+0.058
 kinds+0.058
 coloured+0.058
Neg logit contributions
orting-0.046
uing-0.044
 Later-0.041
 nicely-0.037
 Finally-0.032
Token says
Token ID1139
Feature activation+1.95e+00

Pos logit contributions
e+0.058
 seen+0.055
 coloured+0.054
 kinds+0.053
 sorts+0.053
Neg logit contributions
orting-0.053
 Later-0.051
uing-0.050
 nicely-0.046
 Finally-0.043
Token.
Token ID13
Feature activation+9.65e-01

Pos logit contributions
 Later+0.101
orting+0.096
uing+0.090
 Finally+0.086
Finally+0.085
Neg logit contributions
e-0.056
 ever-0.051
 wearing-0.050
 seen-0.047
 spe-0.045
Token
Token ID198
Feature activation+2.73e+00

Pos logit contributions
uing+0.039
orting+0.036
 Later+0.031
lated+0.030
ail+0.029
Neg logit contributions
 wearing-0.038
 seen-0.038
e-0.038
 spe-0.036
 cloud-0.036
Token asks
Token ID7893
Feature activation+1.57e+00

Pos logit contributions
orting+0.119
 Later+0.118
uing+0.110
 nicely+0.107
 Finally+0.099
Neg logit contributions
e-0.120
 seen-0.116
 coloured-0.111
 spe-0.110
 kinds-0.108
Token his
Token ID465
Feature activation+3.28e+00

Pos logit contributions
orting+0.075
 Later+0.074
uing+0.070
ealing+0.065
hew+0.064
Neg logit contributions
e-0.049
 seen-0.046
 ever-0.046
 ghosts-0.044
 wearing-0.044
Token dad
Token ID9955
Feature activation-1.20e+00

Pos logit contributions
orting+0.071
 Later+0.070
 nicely+0.065
uing+0.064
 Finally+0.050
Neg logit contributions
e-0.188
 seen-0.183
 coloured-0.177
 spe-0.176
 sorts-0.174
Token,
Token ID11
Feature activation+2.50e+00

Pos logit contributions
e+0.036
 seen+0.035
 ever+0.034
 wearing+0.034
 kinds+0.032
Neg logit contributions
 Later-0.058
orting-0.055
uing-0.050
Finally-0.050
 Finally-0.048
Token "
Token ID366
Feature activation+4.89e-01

Pos logit contributions
orting+0.093
 Later+0.088
uing+0.087
udd+0.081
hew+0.081
Neg logit contributions
 wearing-0.112
e-0.100
 seen-0.098
 a-0.096
 ever-0.096
TokenWhen
Token ID2215
Feature activation+1.65e+00

Pos logit contributions
orting+0.021
 Later+0.021
uing+0.020
 nicely+0.019
 Finally+0.018
Neg logit contributions
e-0.018
 seen-0.017
 coloured-0.016
 comet-0.016
 spe-0.016
Token will
Token ID481
Feature activation-2.94e-01

Pos logit contributions
orting+0.060
 Later+0.058
 nicely+0.057
uing+0.055
 Finally+0.050
Neg logit contributions
e-0.067
 coloured-0.065
 seen-0.064
 sorts-0.064
 kinds-0.063
Token mom
Token ID1995
Feature activation-1.04e+00

Pos logit contributions
e+0.011
 seen+0.011
 coloured+0.010
 spe+0.010
 kinds+0.010
Neg logit contributions
orting-0.012
 Later-0.012
uing-0.011
 nicely-0.011
 Finally-0.010
Token come
Token ID1282
Feature activation+1.44e+00

Pos logit contributions
e+0.037
 seen+0.036
 coloured+0.034
 spe+0.034
 kinds+0.033
Neg logit contributions
orting-0.045
 Later-0.044
uing-0.042
 nicely-0.041
 Finally-0.038
Token home
Token ID1363
Feature activation+4.19e+00

Pos logit contributions
orting+0.058
 Later+0.057
uing+0.054
 nicely+0.053
 Finally+0.048
Neg logit contributions
e-0.055
 seen-0.053
 coloured-0.051
 spe-0.051
 kinds-0.051
Token?"
Token ID1701
Feature activation-1.39e+00

Pos logit contributions
orting+0.183
 Later+0.177
uing+0.170
 nicely+0.160
 Finally+0.153
Neg logit contributions
e-0.150
 seen-0.145
 coloured-0.136
 spe-0.133
 kinds-0.132
Token is
Token ID318
Feature activation-2.41e+00

Pos logit contributions
orting+0.081
 nicely+0.075
 Later+0.074
uing+0.072
 Finally+0.064
Neg logit contributions
e-0.112
 seen-0.108
leaf-0.106
 spe-0.104
vous-0.102
Token rubbish
Token ID35022
Feature activation+2.01e-01

Pos logit contributions
e+0.091
 seen+0.086
 coloured+0.083
 sorts+0.082
 kinds+0.081
Neg logit contributions
orting-0.099
 Later-0.098
 nicely-0.093
uing-0.092
 Finally-0.083
Token!"
Token ID2474
Feature activation-2.78e+00

Pos logit contributions
orting+0.008
 nicely+0.007
uing+0.007
 Later+0.006
der+0.006
Neg logit contributions
e-0.008
leaf-0.008
 seen-0.008
vous-0.008
 blonde-0.007
Token Ben
Token ID3932
Feature activation-8.16e-01

Pos logit contributions
e+0.127
 seen+0.124
 coloured+0.118
 spe+0.117
 kinds+0.114
Neg logit contributions
orting-0.095
 Later-0.090
uing-0.089
 nicely-0.079
 Finally-0.072
Token said
Token ID531
Feature activation+1.37e+00

Pos logit contributions
e+0.035
 spe+0.034
 comet+0.034
 seen+0.033
 universe+0.033
Neg logit contributions
orting-0.028
 Later-0.028
 nicely-0.026
uing-0.025
 Finally-0.023
Token.
Token ID13
Feature activation+5.09e-01

Pos logit contributions
 Later+0.072
orting+0.070
uing+0.065
Finally+0.062
 Finally+0.061
Neg logit contributions
e-0.039
 seen-0.035
 ever-0.035
 wearing-0.034
 spe-0.032
Token He
Token ID679
Feature activation+2.81e+00

Pos logit contributions
orting+0.021
uing+0.021
hew+0.017
 Later+0.017
udd+0.016
Neg logit contributions
 wearing-0.020
 spe-0.019
e-0.019
 a-0.019
 ghost-0.018
Token felt
Token ID2936
Feature activation-2.39e+00

Pos logit contributions
orting+0.122
 Later+0.121
uing+0.115
 nicely+0.112
 Finally+0.103
Neg logit contributions
e-0.099
 seen-0.095
 coloured-0.089
 spe-0.089
 sorts-0.089
Token sad
Token ID6507
Feature activation+2.31e+00

Pos logit contributions
e+0.101
 seen+0.098
 coloured+0.093
 spe+0.092
 kinds+0.091
Neg logit contributions
orting-0.090
 Later-0.087
uing-0.083
 nicely-0.077
 Finally-0.072
Token and
Token ID290
Feature activation+6.35e-01

Pos logit contributions
 Later+0.132
uing+0.128
orting+0.125
 Finally+0.120
ail+0.118
Neg logit contributions
 wearing-0.045
 seen-0.044
 ever-0.044
 ghosts-0.042
e-0.039
Token silly
Token ID14397
Feature activation+3.23e-01

Pos logit contributions
orting+0.027
 Later+0.026
uing+0.026
 nicely+0.024
 Finally+0.023
Neg logit contributions
e-0.023
 seen-0.023
 coloured-0.021
 spe-0.021
 kinds-0.021
Token the
Token ID262
Feature activation+1.54e+00

Pos logit contributions
orting+0.017
 Later+0.017
 nicely+0.017
uing+0.015
 patiently+0.014
Neg logit contributions
 seen-0.018
e-0.017
leaf-0.017
vous-0.017
 coloured-0.017
Token porch
Token ID33179
Feature activation-1.99e+00

Pos logit contributions
orting+0.081
uing+0.079
 Later+0.076
 nicely+0.072
 Finally+0.069
Neg logit contributions
e-0.042
 seen-0.041
 sorts-0.039
 coloured-0.039
 kinds-0.038
Token,
Token ID11
Feature activation+2.56e+00

Pos logit contributions
e+0.070
 seen+0.067
 sorts+0.065
 wearing+0.065
 kinds+0.064
Neg logit contributions
orting-0.087
uing-0.084
 Later-0.082
arl-0.075
 nicely-0.073
Token jumping
Token ID14284
Feature activation+1.94e+00

Pos logit contributions
orting+0.133
 Later+0.130
uing+0.126
 nicely+0.120
 Finally+0.114
Neg logit contributions
e-0.071
 seen-0.068
 coloured-0.063
 spe-0.062
 wearing-0.061
Token high
Token ID1029
Feature activation+3.28e+00

Pos logit contributions
orting+0.086
 Later+0.084
uing+0.082
 nicely+0.075
 Finally+0.072
Neg logit contributions
e-0.067
 seen-0.062
 coloured-0.062
 spe-0.060
 shaped-0.058
Token and
Token ID290
Feature activation+1.37e+00

Pos logit contributions
orting+0.137
 Later+0.135
uing+0.126
 nicely+0.125
 Finally+0.114
Neg logit contributions
e-0.123
 seen-0.120
 spe-0.110
 coloured-0.110
 sorts-0.110
Token running
Token ID2491
Feature activation+3.41e-01

Pos logit contributions
orting+0.065
 Later+0.063
uing+0.062
 nicely+0.057
 Finally+0.054
Neg logit contributions
e-0.044
 seen-0.043
 coloured-0.040
 spe-0.040
 wearing-0.039
Token low
Token ID1877
Feature activation+2.61e+00

Pos logit contributions
orting+0.014
 Later+0.013
uing+0.013
 nicely+0.012
 Finally+0.011
Neg logit contributions
e-0.014
 seen-0.013
 coloured-0.012
 spe-0.012
 kinds-0.012
Token.
Token ID13
Feature activation+1.47e-01

Pos logit contributions
orting+0.100
 Later+0.097
uing+0.092
 nicely+0.088
 Finally+0.081
Neg logit contributions
e-0.106
 seen-0.103
 coloured-0.098
 spe-0.098
 kinds-0.096
Token She
Token ID1375
Feature activation+2.43e+00

Pos logit contributions
uing+0.005
orting+0.005
 nicely+0.004
rier+0.004
izes+0.004
Neg logit contributions
e-0.007
 seen-0.006
 wearing-0.006
 sorts-0.006
 coloured-0.006
Token hopped
Token ID45230
Feature activation-4.81e-01

Pos logit contributions
orting+0.100
 Later+0.100
uing+0.095
 nicely+0.094
 Finally+0.086
Neg logit contributions
e-0.086
 seen-0.084
 coloured-0.080
leaf-0.080
 spe-0.080
Token.
Token ID13
Feature activation+9.56e-01

Pos logit contributions
orting+0.059
 Later+0.058
uing+0.055
 nicely+0.051
 Finally+0.049
Neg logit contributions
e-0.053
 seen-0.051
 coloured-0.048
 kinds-0.047
 sorts-0.047
Token They
Token ID1119
Feature activation+6.71e-01

Pos logit contributions
orting+0.035
uing+0.035
ail+0.028
 Later+0.028
izes+0.028
Neg logit contributions
e-0.039
 seen-0.037
 wearing-0.037
 spe-0.037
 sorts-0.036
Token were
Token ID547
Feature activation-3.61e+00

Pos logit contributions
 Later+0.028
orting+0.027
 nicely+0.026
uing+0.026
 Finally+0.024
Neg logit contributions
e-0.023
 seen-0.022
 comet-0.022
 supposed-0.022
ymes-0.021
Token excited
Token ID6568
Feature activation-3.13e-01

Pos logit contributions
e+0.148
 seen+0.141
 comet+0.139
 coloured+0.138
 sorts+0.138
Neg logit contributions
 Later-0.133
orting-0.126
 nicely-0.125
uing-0.120
 Finally-0.110
Token to
Token ID284
Feature activation-2.05e+00

Pos logit contributions
e+0.013
 seen+0.012
 spe+0.012
 coloured+0.012
 comet+0.011
Neg logit contributions
orting-0.012
 Later-0.012
uing-0.011
 nicely-0.011
 Finally-0.010
Token eat
Token ID4483
Feature activation+1.55e+00

Pos logit contributions
 seen+0.075
e+0.075
 coloured+0.073
ymes+0.072
 kinds+0.072
Neg logit contributions
 Later-0.083
orting-0.081
uing-0.077
 nicely-0.077
 Finally-0.070
Token their
Token ID511
Feature activation+5.45e+00

Pos logit contributions
orting+0.066
 Later+0.063
uing+0.061
 nicely+0.060
 Finally+0.054
Neg logit contributions
e-0.058
 seen-0.056
 spe-0.053
 coloured-0.052
vous-0.052
Token sweet
Token ID6029
Feature activation+2.43e+00

Pos logit contributions
orting+0.245
 Later+0.232
 nicely+0.231
uing+0.226
 Finally+0.204
Neg logit contributions
e-0.184
 seen-0.173
vous-0.168
leaf-0.167
 sorts-0.163
Token things
Token ID1243
Feature activation+5.65e-02

Pos logit contributions
orting+0.094
uing+0.081
 nicely+0.081
 Later+0.074
uddle+0.070
Neg logit contributions
 supposed-0.094
vous-0.093
leaf-0.093
e-0.093
 seen-0.090
Token later
Token ID1568
Feature activation+4.86e+00

Pos logit contributions
 Later+0.002
orting+0.002
uing+0.002
 nicely+0.002
 Finally+0.002
Neg logit contributions
e-0.002
 seen-0.002
 coloured-0.002
 spe-0.002
 kinds-0.002
Token.
Token ID13
Feature activation+1.27e+00

Pos logit contributions
orting+0.179
 Later+0.173
uing+0.164
 nicely+0.160
 Finally+0.145
Neg logit contributions
e-0.206
 seen-0.199
 coloured-0.192
 spe-0.191
 sorts-0.188
Token They
Token ID1119
Feature activation+8.18e-01

Pos logit contributions
orting+0.025
uing+0.025
 nicely+0.020
 Later+0.020
izes+0.019
Neg logit contributions
e-0.039
 seen-0.037
 sorts-0.037
 wearing-0.036
 kinds-0.035
Token laughed
Token ID13818
Feature activation+2.48e+00

Pos logit contributions
orting+0.035
 Later+0.034
uing+0.033
 nicely+0.032
 Finally+0.029
Neg logit contributions
e-0.028
 seen-0.027
 coloured-0.026
 spe-0.026
 kinds-0.025
Token and
Token ID290
Feature activation+1.89e+00

Pos logit contributions
orting+0.110
 Later+0.108
uing+0.103
 nicely+0.095
 Finally+0.091
Neg logit contributions
e-0.085
 seen-0.080
 coloured-0.079
 spe-0.076
 wearing-0.075
Token smiled
Token ID13541
Feature activation+2.99e+00

Pos logit contributions
orting+0.084
 Later+0.082
 nicely+0.077
uing+0.076
 Finally+0.073
Neg logit contributions
e-0.060
 seen-0.059
 comet-0.056
 coloured-0.056
vous-0.056
Token and
Token ID290
Feature activation+2.01e+00

Pos logit contributions
orting+0.136
 Later+0.129
uing+0.124
arl+0.112
 nicely+0.112
Neg logit contributions
e-0.098
 coloured-0.095
 seen-0.093
 spe-0.091
 wearing-0.090
Token enjoyed
Token ID8359
Feature activation+1.65e+00

Pos logit contributions
orting+0.088
 Later+0.086
 nicely+0.081
uing+0.081
 Finally+0.076
Neg logit contributions
e-0.067
 seen-0.066
 coloured-0.062
 spe-0.061
 comet-0.061
Token their
Token ID511
Feature activation+5.08e+00

Pos logit contributions
orting+0.066
 Later+0.063
 nicely+0.062
uing+0.061
 Finally+0.055
Neg logit contributions
e-0.063
 seen-0.062
vous-0.059
 coloured-0.058
 spe-0.058
Token time
Token ID640
Feature activation+2.02e+00

Pos logit contributions
orting+0.211
 nicely+0.206
uing+0.193
 Later+0.189
 patiently+0.175
Neg logit contributions
 seen-0.186
e-0.167
 spe-0.166
 supposed-0.163
 sorts-0.159
Token together
Token ID1978
Feature activation+4.81e+00

Pos logit contributions
orting+0.075
uing+0.071
 Later+0.070
 nicely+0.070
 patiently+0.060
Neg logit contributions
e-0.081
 seen-0.081
 spe-0.077
vous-0.077
 sorts-0.076
Token.
Token ID13
Feature activation+1.24e+00

Pos logit contributions
orting+0.188
 Later+0.183
uing+0.174
 nicely+0.166
 Finally+0.153
Neg logit contributions
e-0.192
 seen-0.186
 coloured-0.178
 spe-0.176
 kinds-0.173
Token It
Token ID632
Feature activation+3.15e+00

Pos logit contributions
uing+0.042
orting+0.041
rier+0.033
izes+0.033
 nicely+0.033
Neg logit contributions
e-0.058
 sorts-0.054
 wearing-0.053
 seen-0.053
 kinds-0.051
Token Tim
Token ID5045
Feature activation-2.80e+00

Pos logit contributions
e+0.297
 seen+0.288
 coloured+0.266
 wearing+0.264
 ever+0.258
Neg logit contributions
orting-0.351
uing-0.341
 Later-0.320
 nicely-0.288
der-0.277
Tokenmy
Token ID1820
Feature activation-3.33e+00

Pos logit contributions
e+0.125
 seen+0.121
 spe+0.118
 coloured+0.117
 comet+0.114
Neg logit contributions
orting-0.096
 Later-0.093
uing-0.087
 nicely-0.084
 Finally-0.075
Token.
Token ID13
Feature activation-2.74e+00

Pos logit contributions
e+0.140
 seen+0.137
 spe+0.134
 coloured+0.131
leaf+0.131
Neg logit contributions
orting-0.116
 Later-0.116
 nicely-0.106
uing-0.105
 Finally-0.093
Token Tim
Token ID5045
Feature activation-2.24e+00

Pos logit contributions
e+0.107
 seen+0.103
 coloured+0.102
 sorts+0.101
 kinds+0.101
Neg logit contributions
uing-0.109
orting-0.109
 Later-0.099
 nicely-0.095
izes-0.088
Tokenmy
Token ID1820
Feature activation-2.93e+00

Pos logit contributions
 spe+0.117
leaf+0.115
e+0.110
 comet+0.109
 seen+0.108
Neg logit contributions
 Later-0.059
orting-0.055
 nicely-0.050
uing-0.044
 Finally-0.041
Token loved
Token ID6151
Feature activation-4.01e+00

Pos logit contributions
leaf+0.112
 spe+0.111
 seen+0.110
ymes+0.107
e+0.107
Neg logit contributions
 Later-0.115
orting-0.106
 nicely-0.105
 Finally-0.095
uing-0.094
Token to
Token ID284
Feature activation-3.85e+00

Pos logit contributions
e+0.198
vous+0.183
 seen+0.181
ymes+0.176
 spe+0.170
Neg logit contributions
 Later-0.118
orting-0.111
 nicely-0.107
 patiently-0.094
 Finally-0.087
Token draw
Token ID3197
Feature activation+1.90e-01

Pos logit contributions
ymes+0.175
 seen+0.167
 supposed+0.162
 spe+0.159
vous+0.157
Neg logit contributions
 nicely-0.116
 Later-0.116
orting-0.116
uing-0.110
 His-0.100
Token with
Token ID351
Feature activation-6.28e-01

Pos logit contributions
orting+0.009
 Later+0.008
 nicely+0.008
uing+0.008
 Finally+0.007
Neg logit contributions
 seen-0.006
e-0.006
 spe-0.006
leaf-0.006
 comet-0.006
Token his
Token ID465
Feature activation+2.39e+00

Pos logit contributions
e+0.026
 seen+0.026
 spe+0.025
vous+0.025
leaf+0.024
Neg logit contributions
 Later-0.022
orting-0.022
 nicely-0.021
uing-0.020
 Finally-0.018
Token pen
Token ID3112
Feature activation-1.96e+00

Pos logit contributions
 Later+0.120
orting+0.119
 nicely+0.118
uing+0.112
 Finally+0.110
Neg logit contributions
e-0.062
 seen-0.060
vous-0.057
leaf-0.056
 spe-0.055
Token big
Token ID1263
Feature activation-4.42e+00

Pos logit contributions
e+0.274
 seen+0.250
keeping+0.246
vous+0.242
 kinds+0.233
Neg logit contributions
 nicely-0.267
uing-0.264
orting-0.263
 Later-0.262
 patiently-0.247
Token box
Token ID3091
Feature activation-3.87e+00

Pos logit contributions
vous+0.182
e+0.181
 seen+0.176
leaf+0.175
 supposed+0.174
Neg logit contributions
orting-0.158
 nicely-0.145
uing-0.145
 Later-0.139
 patiently-0.136
Token of
Token ID286
Feature activation-5.86e+00

Pos logit contributions
e+0.138
ymes+0.129
 seen+0.128
leaf+0.126
 spe+0.125
Neg logit contributions
orting-0.163
 Later-0.159
 nicely-0.153
uing-0.152
 patiently-0.138
Token clothes
Token ID8242
Feature activation-4.51e+00

Pos logit contributions
e+0.209
vous+0.195
 flown+0.186
 seen+0.186
 supposed+0.183
Neg logit contributions
orting-0.247
 nicely-0.239
 Later-0.230
uing-0.224
 patiently-0.207
Token that
Token ID326
Feature activation+2.10e+00

Pos logit contributions
e+0.173
 seen+0.167
 coloured+0.159
 spe+0.157
 kinds+0.154
Neg logit contributions
orting-0.183
 Later-0.180
uing-0.170
 nicely-0.163
 Finally-0.152
Token Lily
Token ID20037
Feature activation-8.45e-01

Pos logit contributions
orting+0.076
 Later+0.075
uing+0.070
 nicely+0.069
 Finally+0.063
Neg logit contributions
e-0.089
 seen-0.085
 spe-0.081
 coloured-0.081
leaf-0.080
Token could
Token ID714
Feature activation-3.51e+00

Pos logit contributions
e+0.030
 seen+0.029
 spe+0.028
leaf+0.028
 comet+0.027
Neg logit contributions
 Later-0.035
orting-0.034
uing-0.034
 nicely-0.033
 Finally-0.030
Token use
Token ID779
Feature activation+1.19e+00

Pos logit contributions
e+0.105
 seen+0.101
 spe+0.095
 coloured+0.093
vous+0.093
Neg logit contributions
orting-0.168
 Later-0.167
uing-0.162
 nicely-0.157
 Finally-0.145
Token to
Token ID284
Feature activation-1.58e+00

Pos logit contributions
orting+0.042
 Later+0.039
uing+0.039
 nicely+0.038
 Finally+0.033
Neg logit contributions
e-0.052
vous-0.049
 seen-0.049
 spe-0.049
 comet-0.049
Token make
Token ID787
Feature activation-1.05e-01

Pos logit contributions
e+0.046
 seen+0.044
 coloured+0.041
 spe+0.041
 kinds+0.039
Neg logit contributions
orting-0.079
 Later-0.077
uing-0.074
 nicely-0.072
 Finally-0.067
Token different
Token ID1180
Feature activation+3.97e+00

Pos logit contributions
e+0.004
 seen+0.004
 spe+0.004
vous+0.004
 comet+0.004
Neg logit contributions
orting-0.004
uing-0.004
 Later-0.004
 nicely-0.004
 Finally-0.003
Token,
Token ID11
Feature activation+6.65e-01

Pos logit contributions
e+0.044
 seen+0.043
 spe+0.041
 coloured+0.041
leaf+0.040
Neg logit contributions
orting-0.037
 Later-0.036
uing-0.034
 nicely-0.034
 Finally-0.030
Token it
Token ID340
Feature activation+2.91e+00

Pos logit contributions
orting+0.024
 Later+0.024
 nicely+0.023
uing+0.022
 Finally+0.020
Neg logit contributions
e-0.028
 seen-0.027
 coloured-0.026
 spe-0.026
leaf-0.025
Token had
Token ID550
Feature activation-3.99e+00

Pos logit contributions
orting+0.104
 Later+0.104
uing+0.098
 nicely+0.096
 Finally+0.086
Neg logit contributions
e-0.118
 seen-0.118
leaf-0.112
 spe-0.111
 coloured-0.110
Token a
Token ID257
Feature activation-6.65e+00

Pos logit contributions
e+0.126
ymes+0.111
 comet+0.107
 sorts+0.104
vous+0.103
Neg logit contributions
 Later-0.176
 nicely-0.170
orting-0.163
 Finally-0.159
 Eventually-0.155
Token nut
Token ID6701
Feature activation-2.61e+00

Pos logit contributions
e+0.251
 seen+0.236
 kinds+0.223
leaf+0.220
 coloured+0.216
Neg logit contributions
 nicely-0.267
 Later-0.263
orting-0.256
uing-0.252
 Finally-0.245
Token in
Token ID287
Feature activation-3.97e+00

Pos logit contributions
e+0.109
leaf+0.109
 sorts+0.105
vous+0.103
ymes+0.103
Neg logit contributions
orting-0.090
 nicely-0.088
uing-0.080
 Later-0.078
der-0.075
Token its
Token ID663
Feature activation+3.08e+00

Pos logit contributions
leaf+0.158
e+0.153
vous+0.150
ymes+0.150
 seen+0.150
Neg logit contributions
 Later-0.142
 nicely-0.130
orting-0.124
 Finally-0.119
uing-0.118
Token mouth
Token ID5422
Feature activation-1.04e-01

Pos logit contributions
orting+0.133
 Later+0.128
uing+0.124
 nicely+0.117
 Finally+0.110
Neg logit contributions
e-0.112
 seen-0.108
 coloured-0.102
 spe-0.101
 kinds-0.099
Token.
Token ID13
Feature activation-6.70e-01

Pos logit contributions
e+0.004
 seen+0.004
 spe+0.004
 coloured+0.004
 kinds+0.004
Neg logit contributions
orting-0.004
 Later-0.004
uing-0.004
 nicely-0.004
 Finally-0.003
Token The
Token ID383
Feature activation+8.43e-01

Pos logit contributions
e+0.030
 seen+0.029
 spe+0.028
 coloured+0.027
 sorts+0.027
Neg logit contributions
orting-0.023
uing-0.022
 Later-0.021
 nicely-0.019
 Finally-0.017
Token squirrel
Token ID33039
Feature activation-5.97e+00

Pos logit contributions
orting+0.039
 Later+0.039
 nicely+0.038
uing+0.035
 Finally+0.034
Neg logit contributions
e-0.027
 seen-0.026
 sorts-0.024
leaf-0.024
 supposed-0.023
Token,
Token ID11
Feature activation-9.16e-01

Pos logit contributions
e+0.099
 seen+0.095
 wearing+0.088
 ever+0.087
 spe+0.087
Neg logit contributions
orting-0.117
 Later-0.114
uing-0.109
 nicely-0.099
 Finally-0.096
Token there
Token ID612
Feature activation-2.09e-01

Pos logit contributions
e+0.021
 seen+0.019
 coloured+0.018
 spe+0.018
 kinds+0.017
Neg logit contributions
orting-0.052
 Later-0.051
uing-0.049
 nicely-0.048
 Finally-0.045
Token was
Token ID373
Feature activation-6.55e+00

Pos logit contributions
e+0.008
 seen+0.008
leaf+0.007
 spe+0.007
 coloured+0.007
Neg logit contributions
 Later-0.008
orting-0.008
uing-0.008
 nicely-0.008
 Finally-0.007
Token a
Token ID257
Feature activation-8.42e+00

Pos logit contributions
ymes+0.249
 spe+0.246
keeping+0.242
leaf+0.236
 kinds+0.234
Neg logit contributions
 Later-0.255
 nicely-0.235
 Finally-0.229
Their-0.209
 Eventually-0.206
Token girl
Token ID2576
Feature activation-6.18e+00

Pos logit contributions
 seen+0.318
e+0.317
 kinds+0.292
 coloured+0.288
 sorts+0.288
Neg logit contributions
orting-0.340
 Later-0.336
 nicely-0.327
uing-0.315
 Finally-0.286
Token who
Token ID508
Feature activation-3.28e+00

Pos logit contributions
leaf+0.225
 comet+0.207
 kinds+0.207
ymes+0.204
 seen+0.201
Neg logit contributions
 nicely-0.253
 Later-0.246
orting-0.243
uing-0.208
 patiently-0.203
Token lived
Token ID5615
Feature activation-3.04e+00

Pos logit contributions
 comet+0.140
leaf+0.139
 seen+0.134
 conf+0.134
ymes+0.134
Neg logit contributions
 Later-0.109
 nicely-0.104
 Finally-0.094
orting-0.092
 Eventually-0.083
Token in
Token ID287
Feature activation-4.56e+00

Pos logit contributions
 spe+0.096
 comet+0.095
 seen+0.094
 kinds+0.093
 sorts+0.091
Neg logit contributions
 nicely-0.141
 Later-0.138
orting-0.135
uing-0.122
 patiently-0.119
Token a
Token ID257
Feature activation-7.24e+00

Pos logit contributions
leaf+0.169
ymes+0.165
 supposed+0.158
vous+0.156
iest+0.155
Neg logit contributions
 Later-0.154
 nicely-0.150
 -0.130
Their-0.130
 Finally-0.129
Token big
Token ID1263
Feature activation-4.43e+00

Pos logit contributions
e+0.199
 seen+0.187
 spe+0.179
 comet+0.175
 coloured+0.175
Neg logit contributions
orting-0.382
 Later-0.364
uing-0.361
 nicely-0.332
 Finally-0.323
Token town
Token ID3240
Feature activation-4.01e+00

Pos logit contributions
e+0.171
 seen+0.166
leaf+0.163
 coloured+0.157
 supposed+0.156
Neg logit contributions
 Later-0.174
orting-0.174
 nicely-0.162
uing-0.154
 Finally-0.143
Token "
Token ID366
Feature activation+1.41e-01

Pos logit contributions
orting+0.089
uing+0.083
 Later+0.079
udd+0.069
lated+0.069
Neg logit contributions
 wearing-0.116
 spe-0.111
e-0.108
 flying-0.108
 coloured-0.104
TokenHey
Token ID10814
Feature activation-4.22e+00

Pos logit contributions
orting+0.007
 Later+0.007
uing+0.006
 nicely+0.006
 Finally+0.006
Neg logit contributions
e-0.005
 seen-0.004
 coloured-0.004
 spe-0.004
 kinds-0.004
Token kid
Token ID5141
Feature activation-5.91e+00

Pos logit contributions
e+0.185
leaf+0.184
 sorts+0.182
 seen+0.178
 kinds+0.178
Neg logit contributions
orting-0.139
 nicely-0.129
uing-0.125
 Later-0.121
:"-0.106
Token,
Token ID11
Feature activation+8.62e-01

Pos logit contributions
e+0.263
 seen+0.262
 coloured+0.256
 comet+0.254
 spe+0.253
Neg logit contributions
orting-0.195
 Later-0.180
uing-0.174
 nicely-0.171
 Finally-0.146
Token do
Token ID466
Feature activation-3.91e+00

Pos logit contributions
orting+0.030
 Later+0.028
uing+0.027
 nicely+0.027
 Finally+0.023
Neg logit contributions
e-0.038
 seen-0.037
 coloured-0.036
 spe-0.036
 kinds-0.035
Token you
Token ID345
Feature activation-2.00e+00

Pos logit contributions
 seen+0.164
 coloured+0.163
 sorts+0.161
 kinds+0.160
 spe+0.160
Neg logit contributions
orting-0.137
 Later-0.126
 nicely-0.125
uing-0.122
 Finally-0.105
Token want
Token ID765
Feature activation-2.71e+00

Pos logit contributions
ymes+0.083
 spe+0.082
 comet+0.080
 ghosts+0.076
 musical+0.076
Neg logit contributions
 nicely-0.065
orting-0.064
uing-0.063
 Later-0.061
hew-0.055
Token to
Token ID284
Feature activation-1.51e+00

Pos logit contributions
e+0.130
 spe+0.128
vous+0.126
ymes+0.125
 sorts+0.120
Neg logit contributions
 nicely-0.076
orting-0.074
 Later-0.070
uing-0.057
 patiently-0.057
Token sell
Token ID3677
Feature activation+1.88e+00

Pos logit contributions
ymes+0.057
 seen+0.054
leaf+0.054
 kinds+0.054
e+0.054
Neg logit contributions
orting-0.057
 Later-0.056
 nicely-0.054
uing-0.053
 Finally-0.046
Token that
Token ID326
Feature activation+4.89e+00

Pos logit contributions
orting+0.075
 Later+0.071
 nicely+0.071
uing+0.067
 Finally+0.059
Neg logit contributions
e-0.072
 spe-0.069
 seen-0.069
vous-0.068
 supposed-0.067
Token pill
Token ID9582
Feature activation+1.46e+00

Pos logit contributions
orting+0.213
 Later+0.206
 nicely+0.199
uing+0.190
 Finally+0.179
Neg logit contributions
e-0.161
leaf-0.157
 coloured-0.150
 seen-0.150
 supposed-0.145
TokenOnce
Token ID7454
Feature activation+1.95e-01

Pos logit contributions
 seen+0.150
e+0.150
 spe+0.148
 coloured+0.146
 comet+0.144
Neg logit contributions
 Later-0.077
orting-0.076
 nicely-0.063
 Finally-0.061
uing-0.058
Token upon
Token ID2402
Feature activation-3.25e+00

Pos logit contributions
orting+0.010
 Later+0.009
uing+0.009
der+0.008
rier+0.008
Neg logit contributions
 seen-0.006
 wearing-0.006
 happened-0.005
 kinds-0.005
 sorts-0.005
Token a
Token ID257
Feature activation-8.16e+00

Pos logit contributions
 spe+0.119
leaf+0.107
 sorts+0.106
 kinds+0.104
 comet+0.100
Neg logit contributions
 nicely-0.132
 Later-0.132
 Finally-0.116
orting-0.112
Finally-0.111
Token time
Token ID640
Feature activation-2.77e+00

Pos logit contributions
e+0.180
 seen+0.172
 coloured+0.161
 spe+0.159
 kinds+0.157
Neg logit contributions
orting-0.454
 Later-0.452
uing-0.430
 nicely-0.424
 Finally-0.404
Token there
Token ID612
Feature activation-2.40e-01

Pos logit contributions
e+0.107
 seen+0.104
 coloured+0.097
 spe+0.097
 wearing+0.095
Neg logit contributions
orting-0.112
 Later-0.110
uing-0.105
 nicely-0.097
 Finally-0.092
Token was
Token ID373
Feature activation-6.63e+00

Pos logit contributions
e+0.009
 seen+0.009
leaf+0.009
 spe+0.009
 coloured+0.009
Neg logit contributions
 Later-0.009
orting-0.009
 nicely-0.009
uing-0.008
 Finally-0.008
Token a
Token ID257
Feature activation-8.37e+00

Pos logit contributions
ymes+0.257
keeping+0.253
leaf+0.249
 spe+0.248
 kinds+0.234
Neg logit contributions
 Later-0.248
 nicely-0.231
 Finally-0.220
Their-0.202
 until-0.200
Token family
Token ID1641
Feature activation-6.26e+00

Pos logit contributions
e+0.307
 seen+0.303
 coloured+0.277
 kinds+0.277
 sorts+0.272
Neg logit contributions
orting-0.352
 Later-0.348
 nicely-0.333
uing-0.326
 Finally-0.297
Token of
Token ID286
Feature activation-6.68e+00

Pos logit contributions
 seen+0.245
e+0.245
 spe+0.242
ymes+0.238
 flown+0.232
Neg logit contributions
orting-0.232
 Later-0.226
 nicely-0.222
uing-0.205
 patiently-0.185
Token wild
Token ID4295
Feature activation-1.16e+00

Pos logit contributions
 seen+0.225
e+0.222
 sorts+0.217
 kinds+0.211
vous+0.209
Neg logit contributions
 nicely-0.290
orting-0.285
 patiently-0.272
 Later-0.270
arl-0.243
Token animals
Token ID4695
Feature activation-7.45e+00

Pos logit contributions
e+0.042
 seen+0.039
 supposed+0.038
leaf+0.037
vous+0.037
Neg logit contributions
orting-0.049
 Later-0.047
 nicely-0.045
udd-0.040
der-0.040
Token a
Token ID257
Feature activation-8.35e+00

Pos logit contributions
 spe+0.118
leaf+0.109
 sorts+0.106
 kinds+0.105
 comet+0.103
Neg logit contributions
 nicely-0.146
 Later-0.145
orting-0.131
 Finally-0.129
Finally-0.123
Token time
Token ID640
Feature activation-2.93e+00

Pos logit contributions
e+0.157
 seen+0.145
 coloured+0.128
 spe+0.126
 kinds+0.118
Neg logit contributions
orting-0.506
 Later-0.496
uing-0.482
 nicely-0.466
 Finally-0.444
Token,
Token ID11
Feature activation-1.27e+00

Pos logit contributions
e+0.112
 seen+0.107
 spe+0.099
 wearing+0.099
 coloured+0.098
Neg logit contributions
orting-0.122
 Later-0.119
uing-0.115
 nicely-0.104
 Finally-0.100
Token there
Token ID612
Feature activation-4.76e-01

Pos logit contributions
e+0.027
 seen+0.026
 coloured+0.023
 spe+0.023
 kinds+0.022
Neg logit contributions
orting-0.074
 Later-0.072
uing-0.070
 nicely-0.067
 Finally-0.064
Token was
Token ID373
Feature activation-6.86e+00

Pos logit contributions
e+0.018
 seen+0.017
leaf+0.017
 spe+0.017
 coloured+0.016
Neg logit contributions
 Later-0.019
orting-0.019
uing-0.018
 nicely-0.018
 Finally-0.016
Token a
Token ID257
Feature activation-8.65e+00

Pos logit contributions
 spe+0.251
ymes+0.251
keeping+0.246
leaf+0.244
 kinds+0.243
Neg logit contributions
 Later-0.278
 nicely-0.257
 Finally-0.250
 Eventually-0.224
Their-0.224
Token little
Token ID1310
Feature activation-5.94e+00

Pos logit contributions
e+0.311
 seen+0.308
 coloured+0.284
 kinds+0.284
 sorts+0.280
Neg logit contributions
orting-0.366
 Later-0.360
 nicely-0.345
uing-0.340
 Finally-0.309
Token girl
Token ID2576
Feature activation-6.37e+00

Pos logit contributions
 seen+0.262
e+0.250
 supposed+0.247
 flown+0.233
leaf+0.229
Neg logit contributions
orting-0.214
 nicely-0.201
 Later-0.186
uing-0.170
arl-0.168
Token named
Token ID3706
Feature activation-8.15e+00

Pos logit contributions
leaf+0.214
 kinds+0.199
 comet+0.198
 seen+0.195
 supposed+0.192
Neg logit contributions
 nicely-0.274
 Later-0.270
orting-0.269
uing-0.235
 patiently-0.223
Token Lily
Token ID20037
Feature activation-3.39e+00

Pos logit contributions
e+0.368
 seen+0.362
 coloured+0.338
 spe+0.337
 wearing+0.332
Neg logit contributions
orting-0.287
uing-0.267
 Later-0.262
 nicely-0.228
 Finally-0.214
Token.
Token ID13
Feature activation-2.89e+00

Pos logit contributions
e+0.145
 spe+0.141
 seen+0.141
leaf+0.140
 coloured+0.138
Neg logit contributions
orting-0.113
 Later-0.113
 nicely-0.105
uing-0.100
 Finally-0.089
Token She
Token ID1375
Feature activation+1.58e+00

Pos logit contributions
e+0.022
 seen+0.021
 kinds+0.021
 spe+0.021
 coloured+0.021
Neg logit contributions
orting-0.016
uing-0.015
 Later-0.014
 nicely-0.013
ail-0.012
Token sees
Token ID7224
Feature activation-7.22e-01

Pos logit contributions
orting+0.065
 Later+0.065
uing+0.060
 nicely+0.059
 Finally+0.055
Neg logit contributions
e-0.059
 seen-0.057
 coloured-0.054
 spe-0.054
 sorts-0.052
Token a
Token ID257
Feature activation-6.43e+00

Pos logit contributions
e+0.024
vous+0.022
 supposed+0.021
 seen+0.021
 spe+0.021
Neg logit contributions
orting-0.032
 nicely-0.032
 Later-0.031
uing-0.028
der-0.028
Token book
Token ID1492
Feature activation-3.87e+00

Pos logit contributions
e+0.250
 seen+0.231
leaf+0.216
vous+0.210
 coloured+0.209
Neg logit contributions
 nicely-0.261
 Later-0.249
orting-0.246
uing-0.234
 Finally-0.223
Token with
Token ID351
Feature activation-2.18e+00

Pos logit contributions
e+0.141
 seen+0.136
 spe+0.130
 coloured+0.128
 kinds+0.126
Neg logit contributions
orting-0.167
 Later-0.163
uing-0.153
 nicely-0.153
 Finally-0.138
Token a
Token ID257
Feature activation-6.69e+00

Pos logit contributions
e+0.076
vous+0.074
 comet+0.073
leaf+0.069
 seen+0.069
Neg logit contributions
 Later-0.083
 nicely-0.081
 Finally-0.074
orting-0.071
 Eventually-0.069
Token big
Token ID1263
Feature activation-4.13e+00

Pos logit contributions
e+0.189
 seen+0.177
 kinds+0.166
 coloured+0.165
 sorts+0.164
Neg logit contributions
 Later-0.327
orting-0.319
 nicely-0.317
uing-0.306
 Finally-0.296
Token lion
Token ID18744
Feature activation-5.58e+00

Pos logit contributions
vous+0.132
e+0.131
leaf+0.130
 supposed+0.127
 seen+0.127
Neg logit contributions
orting-0.173
 Later-0.170
 nicely-0.167
uing-0.157
 Finally-0.154
Token on
Token ID319
Feature activation-4.49e+00

Pos logit contributions
 sorts+0.240
leaf+0.240
vous+0.239
 kinds+0.236
 seen+0.232
Neg logit contributions
orting-0.185
 nicely-0.174
 Later-0.171
uing-0.151
der-0.149
Token the
Token ID262
Feature activation+6.64e-01

Pos logit contributions
 seen+0.167
e+0.161
leaf+0.157
vous+0.156
 supposed+0.153
Neg logit contributions
 Later-0.182
orting-0.172
 nicely-0.167
uing-0.160
 Finally-0.155
Token cover
Token ID3002
Feature activation+7.83e-01

Pos logit contributions
orting+0.032
 Later+0.032
 nicely+0.030
uing+0.030
 Finally+0.028
Neg logit contributions
e-0.019
 seen-0.019
 coloured-0.018
 spe-0.018
 sorts-0.017
Token a
Token ID257
Feature activation-7.94e+00

Pos logit contributions
e+0.081
leaf+0.080
 seen+0.080
 spe+0.078
vous+0.077
Neg logit contributions
 Later-0.078
orting-0.074
 nicely-0.074
uing-0.069
 Finally-0.063
Token farm
Token ID5318
Feature activation-3.75e+00

Pos logit contributions
e+0.231
 seen+0.219
 spe+0.211
 comet+0.208
 coloured+0.208
Neg logit contributions
orting-0.402
 Later-0.387
uing-0.380
 nicely-0.352
 Finally-0.341
Token when
Token ID618
Feature activation-1.24e+00

Pos logit contributions
 spe+0.162
leaf+0.161
e+0.159
ymes+0.158
 seen+0.157
Neg logit contributions
orting-0.129
 Later-0.128
 nicely-0.122
uing-0.102
 Finally-0.101
Token he
Token ID339
Feature activation+1.80e+00

Pos logit contributions
e+0.064
leaf+0.062
 seen+0.061
 kinds+0.061
vous+0.060
Neg logit contributions
 Later-0.034
orting-0.030
 nicely-0.030
 Finally-0.028
 Eventually-0.025
Token heard
Token ID2982
Feature activation-3.09e+00

Pos logit contributions
orting+0.078
 Later+0.077
uing+0.071
 nicely+0.071
 Finally+0.066
Neg logit contributions
e-0.060
leaf-0.058
 kinds-0.058
 sorts-0.057
 coloured-0.057
Token a
Token ID257
Feature activation-6.83e+00

Pos logit contributions
e+0.076
 seen+0.067
 spe+0.067
 supposed+0.063
 sorts+0.062
Neg logit contributions
 Later-0.166
orting-0.165
 nicely-0.155
uing-0.147
 Finally-0.144
Token little
Token ID1310
Feature activation-4.90e+00

Pos logit contributions
e+0.195
 seen+0.185
 spe+0.177
 coloured+0.176
 comet+0.170
Neg logit contributions
orting-0.344
 Later-0.337
uing-0.324
 nicely-0.311
 Finally-0.295
Token voice
Token ID3809
Feature activation-4.17e+00

Pos logit contributions
e+0.148
 seen+0.142
 coloured+0.132
 spe+0.128
 kinds+0.126
Neg logit contributions
orting-0.240
 Later-0.234
uing-0.224
 nicely-0.219
 Finally-0.203
Token say
Token ID910
Feature activation-3.39e-01

Pos logit contributions
e+0.160
 seen+0.154
 spe+0.150
 coloured+0.146
 kinds+0.146
Neg logit contributions
orting-0.169
 Later-0.161
uing-0.153
 nicely-0.149
 Finally-0.132
Token,
Token ID11
Feature activation+1.38e+00

Pos logit contributions
e+0.012
 seen+0.011
 wearing+0.011
 comet+0.011
 spe+0.011
Neg logit contributions
 Later-0.015
orting-0.015
uing-0.014
 Finally-0.013
 nicely-0.013
Token "
Token ID366
Feature activation-1.14e+00

Pos logit contributions
orting+0.043
uing+0.042
 Later+0.038
 nicely+0.034
lated+0.030
Neg logit contributions
 wearing-0.067
e-0.065
 seen-0.065
 spe-0.064
 flying-0.064
Token upon
Token ID2402
Feature activation-3.40e+00

Pos logit contributions
orting+0.005
 Later+0.004
uing+0.004
der+0.004
rier+0.004
Neg logit contributions
 seen-0.003
 wearing-0.003
 happened-0.002
 ever-0.002
 sorts-0.002
Token a
Token ID257
Feature activation-8.47e+00

Pos logit contributions
 spe+0.130
 sorts+0.113
leaf+0.113
 kinds+0.112
 comet+0.110
Neg logit contributions
 nicely-0.131
 Later-0.129
 Finally-0.115
His-0.112
 He-0.112
Token time
Token ID640
Feature activation-2.95e+00

Pos logit contributions
e+0.185
 seen+0.177
 coloured+0.165
 spe+0.165
 kinds+0.162
Neg logit contributions
orting-0.473
 Later-0.472
uing-0.447
 nicely-0.443
 Finally-0.422
Token there
Token ID612
Feature activation-5.54e-01

Pos logit contributions
e+0.113
 seen+0.109
 coloured+0.101
 spe+0.101
 wearing+0.100
Neg logit contributions
orting-0.122
 Later-0.119
uing-0.114
 nicely-0.104
 Finally-0.099
Token was
Token ID373
Feature activation-6.87e+00

Pos logit contributions
e+0.021
 seen+0.020
leaf+0.020
 spe+0.020
 coloured+0.019
Neg logit contributions
 Later-0.022
orting-0.022
 nicely-0.020
uing-0.020
 Finally-0.019
Token a
Token ID257
Feature activation-8.71e+00

Pos logit contributions
ymes+0.269
keeping+0.261
 spe+0.261
leaf+0.253
 kinds+0.245
Neg logit contributions
 Later-0.264
 nicely-0.248
 Finally-0.237
Their-0.217
 Then-0.216
Token white
Token ID2330
Feature activation-3.19e+00

Pos logit contributions
e+0.319
 seen+0.316
 kinds+0.288
 coloured+0.287
 sorts+0.283
Neg logit contributions
orting-0.366
 Later-0.361
 nicely-0.347
uing-0.339
 Finally-0.309
Token tray
Token ID26473
Feature activation-5.61e+00

Pos logit contributions
 seen+0.117
e+0.107
 supposed+0.107
leaf+0.100
vous+0.099
Neg logit contributions
orting-0.140
 Later-0.137
 nicely-0.133
uing-0.118
 patiently-0.118
Token.
Token ID13
Feature activation-2.74e+00

Pos logit contributions
 seen+0.267
ymes+0.264
 spe+0.263
leaf+0.257
e+0.255
Neg logit contributions
orting-0.162
 Later-0.160
 nicely-0.158
:"-0.128
 Finally-0.119
Token It
Token ID632
Feature activation+7.07e-01

Pos logit contributions
 supposed+0.165
 seen+0.164
ymes+0.163
e+0.161
leaf+0.160
Neg logit contributions
 Later-0.059
 Finally-0.042
orting-0.036
 nicely-0.035
 Eventually-0.033
Token was
Token ID373
Feature activation-5.59e+00

Pos logit contributions
 Later+0.028
orting+0.026
 nicely+0.025
uing+0.024
 Finally+0.024
Neg logit contributions
e-0.027
 seen-0.027
leaf-0.025
 comet-0.025
ymes-0.025