TOP ACTIVATIONS
MAX = 0.537

a pot , some fire wood and some water . Scare
- shiny rocks , drift wood and sea foam . But
log and make some fire wood . The dove
collect funny pieces of drift wood . Then they would take
was known as the mail man of the mighty lake .
Once there was a fire man named Sam . He was
decided to build a snow man . He grabbed some big
there was a humble fire man . He worked very hard
, there was a fire man who was so nice and
said : " The mail man always has the best stories
. <|endoftext|> A famous post man named Bob was out delivering
there was an attractive sun flower growing in a garden .
He thanked the repair - man and promised to never forget
smiled and thanked the post man . Comfort ably eating his
was a very brave fire man . Every day , he
<|endoftext|> Jim was a fire man . He was very strong
. It was a trades man and he was carrying a
excited and asked the fire man if he could help .
The bald fire man laughed and said , "
After the mail man left , Tim ran back

MIN ACTIVATIONS
MIN = -0.387

can hurt you or someone else . You could have been
light . She wondered what else she could learn by guessing
bear . It is someone else 's bear . And it
or a smoke or something else . The fire man are
Maybe you should search somewhere else ." L ily
They wanted to play somewhere else , where they could have
he didn 't need anyone else 's help . From that
Maybe you left it somewhere else . Maybe you should search
your diary . Maybe someone else did . Maybe it was
decide to play with something else . Something safe and fun
. It looks for something else to eat .
. He looks for something else to eat . He hopes
But you can play something else inside . How about you
" Can I have something else ?" Her mom
ask before you use someone else 's things . You should
She wants to paint something else . She goes
hears them . No one else comes . The dog is
wanted to shoot at something else . She saw a butterfly
have killed yourself or someone else . You caused a lot
now . Go play somewhere else , please ," dad said

INTERVAL 0.306 - 0.537
CONTAINS 0.179%

cater pillar and the sun flower smiled . The
her penny in the pig gy bank ever again . She
to be rude and hurt ful to your friend . You
, grabbing the leaves for Joey to eat .
. They see their snow man smiling at them . But

INTERVAL 0.075 - 0.306
CONTAINS 17.996%

went to her mom and said , " Mom , can
One day , they see a very big boat
L ily says , " Yes , Tom , I want
ers loved to sleep all day long and never wanted to
. The little girl was so proud of herself

INTERVAL -0.156 - 0.075
CONTAINS 76.738%

didn 't have enough money . She stared at the taxi
Tom and Lily ran to their mom , who was
did with Fl uffy . He hopes it will hatch and
The dog took the ball back to her toy , and
show , and he said he wanted to dance and make

INTERVAL -0.387 - -0.156
CONTAINS 5.087%

. The bear learned how to climb trees and
drink coffee every morning , but one day he spilled it
. I 'm sorry , Lily . I 'm sorry ,
soap box and smiled . But when he turned around ,
want to join the team at first , but the other
Token a
Token ID257
Feature activation+1.34e-01

Pos logit contributions
fters+0.198
ets+0.188
 mats+0.182
 quizz+0.181
 else+0.181
Neg logit contributions
wood-0.217
flower-0.207
tha-0.206
fly-0.205
man-0.201
Token pot
Token ID1787
Feature activation+1.40e-01

Pos logit contributions
wood+0.242
tha+0.227
flower+0.221
fly+0.217
 extingu+0.216
Neg logit contributions
fters-0.215
 else-0.197
ets-0.197
 quizz-0.195
 mats-0.193
Token,
Token ID11
Feature activation+2.23e-02

Pos logit contributions
wood+0.279
tha+0.267
fly+0.264
flower+0.261
 extingu+0.258
Neg logit contributions
fters-0.197
ets-0.188
 mats-0.175
 else-0.173
 quizz-0.172
Token some
Token ID617
Feature activation-6.32e-02

Pos logit contributions
wood+0.043
tha+0.040
flower+0.040
fly+0.040
man+0.040
Neg logit contributions
fters-0.033
ets-0.030
 mats-0.030
 quizz-0.029
 else-0.029
Token fire
Token ID2046
Feature activation+1.09e-01

Pos logit contributions
fters+0.108
ets+0.101
 mats+0.098
 quizz+0.097
 scores+0.097
Neg logit contributions
wood-0.108
fly-0.101
flower-0.100
man-0.100
 extingu-0.099
Tokenwood
Token ID3822
Feature activation+5.37e-01

Pos logit contributions
 Amen+0.189
tha+0.170
 Frankie+0.163
 Martha+0.161
 Mill+0.159
Neg logit contributions
ets-0.175
 anymore-0.166
 mats-0.163
 corners-0.158
pler-0.158
Token and
Token ID290
Feature activation+1.17e-02

Pos logit contributions
wood+1.055
tha+1.016
fly+1.012
flower+1.002
man+0.988
Neg logit contributions
fters-0.738
ets-0.706
 mats-0.672
 else-0.654
 scores-0.650
Token some
Token ID617
Feature activation-6.66e-02

Pos logit contributions
wood+0.021
tha+0.020
flower+0.019
fly+0.019
man+0.019
Neg logit contributions
fters-0.019
ets-0.018
 mats-0.018
 else-0.018
 quizz-0.017
Token water
Token ID1660
Feature activation+5.72e-02

Pos logit contributions
fters+0.112
ets+0.104
 mats+0.101
 quizz+0.100
 scores+0.099
Neg logit contributions
wood-0.117
fly-0.110
flower-0.108
man-0.108
 extingu-0.107
Token.
Token ID13
Feature activation-4.91e-03

Pos logit contributions
wood+0.114
tha+0.112
 Amen+0.111
fly+0.109
flower+0.107
Neg logit contributions
fters-0.074
ets-0.073
 mats-0.069
 anymore-0.067
 else-0.067
Token Scare
Token ID47605
Feature activation+1.55e-01

Pos logit contributions
fters+0.009
ets+0.008
 scores+0.008
 quizz+0.008
 mats+0.008
Neg logit contributions
wood-0.008
 extingu-0.007
fly-0.007
man-0.007
 Mill-0.007
Token -
Token ID532
Feature activation+1.58e-01

Pos logit contributions
fters+0.330
 mats+0.290
ets+0.290
 quizz+0.283
 scores+0.283
Neg logit contributions
wood-0.349
 extingu-0.319
fly-0.318
man-0.318
work-0.312
Token shiny
Token ID22441
Feature activation+3.60e-02

Pos logit contributions
wood+0.298
fly+0.277
 extingu+0.274
flower+0.273
man+0.273
Neg logit contributions
fters-0.246
ets-0.228
 mats-0.215
 scores-0.212
 quizz-0.210
Token rocks
Token ID12586
Feature activation-1.86e-01

Pos logit contributions
wood+0.068
fly+0.063
 extingu+0.062
flower+0.062
man+0.061
Neg logit contributions
fters-0.058
ets-0.050
itionally-0.049
 quizz-0.049
and-0.049
Token,
Token ID11
Feature activation+1.54e-02

Pos logit contributions
fters+0.263
ets+0.254
 mats+0.237
 else+0.236
 scores+0.233
Neg logit contributions
wood-0.365
tha-0.347
fly-0.345
flower-0.344
 extingu-0.342
Token drift
Token ID24260
Feature activation+7.36e-02

Pos logit contributions
wood+0.030
flower+0.028
fly+0.028
 extingu+0.027
man+0.027
Neg logit contributions
fters-0.023
ets-0.021
 scores-0.021
 quizz-0.020
itionally-0.020
Tokenwood
Token ID3822
Feature activation+5.37e-01

Pos logit contributions
 Amen+0.130
flower+0.127
fly+0.126
tha+0.126
man+0.122
Neg logit contributions
ets-0.118
fters-0.111
 corners-0.107
 mats-0.103
 anymore-0.102
Token and
Token ID290
Feature activation+1.27e-02

Pos logit contributions
wood+1.062
fly+1.033
flower+1.029
tha+1.029
man+1.004
Neg logit contributions
fters-0.728
ets-0.713
 mats-0.666
 quizz-0.650
 scores-0.648
Token sea
Token ID5417
Feature activation+3.01e-02

Pos logit contributions
wood+0.024
fly+0.023
flower+0.022
man+0.022
 extingu+0.022
Neg logit contributions
fters-0.020
ets-0.018
 mats-0.017
 quizz-0.017
 scores-0.017
Token foam
Token ID19828
Feature activation+2.25e-02

Pos logit contributions
wood+0.056
fly+0.054
flower+0.053
tha+0.053
 extingu+0.053
Neg logit contributions
fters-0.045
ets-0.043
 mats-0.041
 quizz-0.041
 scores-0.041
Token.
Token ID13
Feature activation-7.38e-03

Pos logit contributions
wood+0.043
fly+0.040
flower+0.040
man+0.040
 extingu+0.039
Neg logit contributions
fters-0.035
ets-0.032
 mats-0.031
 quizz-0.031
 scores-0.030
Token But
Token ID887
Feature activation-1.75e-01

Pos logit contributions
fters+0.012
ets+0.011
 mats+0.010
 scores+0.010
 quizz+0.010
Neg logit contributions
wood-0.014
fly-0.013
man-0.013
flower-0.013
 extingu-0.013
Token log
Token ID2604
Feature activation+7.39e-02

Pos logit contributions
wood+0.006
work+0.006
fly+0.006
tha+0.006
 extingu+0.006
Neg logit contributions
fters-0.007
 corners-0.006
ets-0.006
 tant-0.006
 else-0.006
Token and
Token ID290
Feature activation+1.30e-02

Pos logit contributions
wood+0.150
tha+0.142
fly+0.142
flower+0.141
 extingu+0.140
Neg logit contributions
fters-0.100
ets-0.094
 mats-0.090
 else-0.089
 anymore-0.089
Token make
Token ID787
Feature activation-1.26e-02

Pos logit contributions
wood+0.026
fly+0.024
flower+0.024
man+0.024
tha+0.024
Neg logit contributions
fters-0.019
ets-0.018
 mats-0.017
 quizz-0.017
 scores-0.017
Token some
Token ID617
Feature activation-7.75e-02

Pos logit contributions
fters+0.020
ets+0.018
 quizz+0.018
 else+0.018
 mats+0.018
Neg logit contributions
wood-0.023
fly-0.022
flower-0.022
tha-0.021
 extingu-0.021
Token fire
Token ID2046
Feature activation+1.01e-01

Pos logit contributions
fters+0.119
ets+0.110
 mats+0.107
 quizz+0.106
 scores+0.105
Neg logit contributions
wood-0.147
fly-0.139
flower-0.138
man-0.137
tha-0.137
Tokenwood
Token ID3822
Feature activation+5.36e-01

Pos logit contributions
tha+0.195
 Amen+0.193
 Martha+0.177
rey+0.177
Mill+0.175
Neg logit contributions
 anymore-0.136
ets-0.127
 else-0.122
 scores-0.117
fters-0.115
Token.
Token ID13
Feature activation-1.96e-03

Pos logit contributions
wood+1.073
fly+1.043
tha+1.041
flower+1.029
man+1.014
Neg logit contributions
fters-0.735
ets-0.688
 mats-0.657
 else-0.653
 scores-0.650
Token
Token ID198
Feature activation-1.73e-02

Pos logit contributions
fters+0.003
ets+0.003
 mats+0.003
 scores+0.003
 quizz+0.003
Neg logit contributions
wood-0.004
man-0.003
fly-0.003
tha-0.003
flower-0.003
Token
Token ID198
Feature activation-1.74e-02

Pos logit contributions
fters+0.030
ets+0.029
 mats+0.027
 scores+0.027
 quizz+0.027
Neg logit contributions
wood-0.029
man-0.028
fly-0.027
tha-0.027
 extingu-0.026
TokenThe
Token ID464
Feature activation-3.58e-02

Pos logit contributions
fters+0.028
ets+0.028
 anymore+0.027
 scores+0.027
 quizz+0.026
Neg logit contributions
wood-0.030
man-0.029
flower-0.028
fly-0.028
tha-0.027
Token dove
Token ID38862
Feature activation+3.93e-02

Pos logit contributions
fters+0.063
 else+0.054
 mats+0.054
 rides+0.053
 corners+0.053
Neg logit contributions
wood-0.064
work-0.060
fly-0.059
 extingu-0.058
 Mill-0.055
Token collect
Token ID2824
Feature activation-1.05e-01

Pos logit contributions
wood+0.023
fly+0.022
flower+0.021
tha+0.021
man+0.021
Neg logit contributions
fters-0.013
ets-0.013
 mats-0.012
 quizz-0.012
 scores-0.012
Token funny
Token ID8258
Feature activation-1.05e-01

Pos logit contributions
fters+0.149
ets+0.128
 calculations+0.123
itionally+0.122
 quizz+0.122
Neg logit contributions
wood-0.220
 extingu-0.205
flower-0.199
fly-0.196
man-0.195
Token pieces
Token ID5207
Feature activation-7.65e-02

Pos logit contributions
fters+0.155
itionally+0.129
ets+0.125
 anymore+0.121
 scored+0.120
Neg logit contributions
wood-0.219
flower-0.199
 extingu-0.197
work-0.193
fly-0.193
Token of
Token ID286
Feature activation-3.53e-02

Pos logit contributions
fters+0.124
 mats+0.110
ets+0.110
 quizz+0.110
 calculations+0.108
Neg logit contributions
wood-0.140
fly-0.131
flower-0.130
man-0.129
work-0.128
Token drift
Token ID24260
Feature activation+7.47e-02

Pos logit contributions
fters+0.057
ets+0.053
itionally+0.050
and+0.049
 mats+0.048
Neg logit contributions
wood-0.064
 extingu-0.061
fly-0.058
work-0.058
flower-0.058
Tokenwood
Token ID3822
Feature activation+5.36e-01

Pos logit contributions
flower+0.076
tha+0.075
 Amen+0.075
fly+0.074
 Tit+0.069
Neg logit contributions
ets-0.179
ag-0.168
 corners-0.163
and-0.161
fters-0.160
Token.
Token ID13
Feature activation-2.43e-03

Pos logit contributions
wood+1.059
fly+1.026
flower+1.018
tha+1.017
man+1.000
Neg logit contributions
fters-0.747
ets-0.714
 mats-0.679
 quizz-0.667
 scores-0.660
Token Then
Token ID3244
Feature activation-1.21e-02

Pos logit contributions
fters+0.005
ets+0.004
 else+0.004
 tant+0.004
 loses+0.004
Neg logit contributions
wood-0.004
work-0.003
flower-0.003
fly-0.003
man-0.003
Token they
Token ID484
Feature activation+1.10e-01

Pos logit contributions
 quizz+0.016
fters+0.016
 scores+0.015
 mats+0.015
 else+0.015
Neg logit contributions
wood-0.025
 extingu-0.024
fly-0.024
man-0.023
tha-0.023
Token would
Token ID561
Feature activation+4.71e-02

Pos logit contributions
wood+0.212
tha+0.200
 extingu+0.199
fly+0.199
flower+0.197
Neg logit contributions
fters-0.160
ets-0.156
 else-0.147
 quizz-0.146
 mats-0.145
Token take
Token ID1011
Feature activation-9.68e-03

Pos logit contributions
wood+0.098
tha+0.095
flower+0.093
fly+0.093
man+0.092
Neg logit contributions
fters-0.059
ets-0.056
 quizz-0.054
 scores-0.053
 rides-0.053
Token was
Token ID373
Feature activation+3.43e-02

Pos logit contributions
wood+0.180
fly+0.169
flower+0.167
man+0.166
tha+0.166
Neg logit contributions
fters-0.153
ets-0.142
 mats-0.138
 quizz-0.136
 scores-0.135
Token known
Token ID1900
Feature activation+2.38e-01

Pos logit contributions
wood+0.060
fly+0.056
 extingu+0.056
flower+0.055
tha+0.055
Neg logit contributions
fters-0.058
ets-0.054
 quizz-0.052
 mats-0.051
 scores-0.051
Token as
Token ID355
Feature activation-2.38e-02

Pos logit contributions
wood+0.440
fly+0.411
tha+0.406
 extingu+0.404
man+0.400
Neg logit contributions
fters-0.377
ets-0.344
 mats-0.343
 scores-0.326
 quizz-0.326
Token the
Token ID262
Feature activation+2.61e-02

Pos logit contributions
fters+0.035
 quizz+0.032
ets+0.032
 else+0.031
 scores+0.031
Neg logit contributions
wood-0.046
flower-0.044
fly-0.044
 extingu-0.044
man-0.043
Token mail
Token ID6920
Feature activation+1.51e-01

Pos logit contributions
wood+0.049
fly+0.047
tha+0.046
 extingu+0.046
flower+0.046
Neg logit contributions
fters-0.040
ets-0.037
 quizz-0.036
 mats-0.036
 scores-0.036
Tokenman
Token ID805
Feature activation+5.21e-01

Pos logit contributions
wood+0.214
 extingu+0.201
tha+0.199
fly+0.194
flower+0.192
Neg logit contributions
fters-0.284
ets-0.277
 else-0.268
 scores-0.267
 quizz-0.267
Token of
Token ID286
Feature activation-1.30e-02

Pos logit contributions
wood+0.994
fly+0.983
man+0.962
flower+0.953
work+0.941
Neg logit contributions
fters-0.788
 mats-0.728
ets-0.701
 tant-0.692
 quizz-0.680
Token the
Token ID262
Feature activation+2.43e-02

Pos logit contributions
fters+0.019
ets+0.018
 quizz+0.018
 scores+0.018
 mats+0.017
Neg logit contributions
wood-0.025
fly-0.024
flower-0.024
tha-0.023
 extingu-0.023
Token mighty
Token ID18680
Feature activation+5.92e-02

Pos logit contributions
wood+0.042
fly+0.040
flower+0.039
man+0.039
 extingu+0.039
Neg logit contributions
fters-0.041
ets-0.039
 mats-0.037
 quizz-0.037
 scores-0.037
Token lake
Token ID13546
Feature activation+1.71e-01

Pos logit contributions
wood+0.101
fly+0.094
flower+0.093
man+0.093
 extingu+0.092
Neg logit contributions
fters-0.103
ets-0.096
 mats-0.093
 quizz-0.092
 scores-0.091
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
wood+0.331
 extingu+0.311
fly+0.310
tha+0.310
flower+0.308
Neg logit contributions
fters-0.247
ets-0.233
 mats-0.220
 scores-0.220
 quizz-0.220
TokenOnce
Token ID7454
Feature activation-1.20e-01

Pos logit contributions
fters+0.096
ets+0.089
 mats+0.086
 quizz+0.085
 scores+0.084
Neg logit contributions
wood-0.096
fly-0.090
flower-0.089
man-0.089
tha-0.087
Token there
Token ID612
Feature activation-1.56e-02

Pos logit contributions
fters+0.198
ets+0.182
 mats+0.179
 quizz+0.174
 calculations+0.170
Neg logit contributions
wood-0.212
tha-0.207
fly-0.206
man-0.205
flower-0.204
Token was
Token ID373
Feature activation+2.77e-02

Pos logit contributions
fters+0.027
ets+0.026
 mats+0.025
 quizz+0.025
 scores+0.024
Neg logit contributions
wood-0.026
fly-0.025
flower-0.025
man-0.025
tha-0.025
Token a
Token ID257
Feature activation+1.35e-01

Pos logit contributions
wood+0.048
fly+0.046
man+0.045
tha+0.045
flower+0.043
Neg logit contributions
fters-0.047
 else-0.045
ets-0.045
 mats-0.045
 quizz-0.044
Token fire
Token ID2046
Feature activation+1.20e-01

Pos logit contributions
wood+0.294
fly+0.279
flower+0.276
man+0.275
 extingu+0.274
Neg logit contributions
fters-0.168
ets-0.153
 mats-0.146
 quizz-0.145
 scores-0.143
Tokenman
Token ID805
Feature activation+5.17e-01

Pos logit contributions
wood+0.155
tha+0.141
 Frankie+0.133
 extingu+0.132
 Martha+0.129
Neg logit contributions
 anymore-0.234
 else-0.221
 nicely-0.208
ets-0.208
 dolls-0.207
Token named
Token ID3706
Feature activation+2.90e-03

Pos logit contributions
wood+1.024
fly+0.966
flower+0.955
man+0.949
 extingu+0.948
Neg logit contributions
fters-0.752
ets-0.694
 mats-0.669
 quizz-0.663
 scores-0.656
Token Sam
Token ID3409
Feature activation+1.08e-01

Pos logit contributions
wood+0.005
flower+0.005
fly+0.005
work+0.005
tha+0.005
Neg logit contributions
fters-0.005
 quizz-0.005
ets-0.004
 else-0.004
 mats-0.004
Token.
Token ID13
Feature activation+9.93e-03

Pos logit contributions
wood+0.214
fly+0.200
flower+0.198
tha+0.198
 extingu+0.197
Neg logit contributions
fters-0.156
ets-0.146
 mats-0.139
 quizz-0.139
 else-0.138
Token He
Token ID679
Feature activation+4.00e-02

Pos logit contributions
tha+0.019
wood+0.018
work+0.018
 extingu+0.018
fly+0.018
Neg logit contributions
 quizz-0.013
ets-0.013
 mats-0.013
 scored-0.013
fters-0.013
Token was
Token ID373
Feature activation+2.59e-02

Pos logit contributions
wood+0.094
tha+0.088
 extingu+0.087
flower+0.087
work+0.085
Neg logit contributions
fters-0.042
 else-0.040
ets-0.039
 scored-0.038
 rides-0.038
Token decided
Token ID3066
Feature activation+1.21e-01

Pos logit contributions
wood+0.130
fly+0.122
tha+0.121
work+0.121
 extingu+0.120
Neg logit contributions
ets-0.078
fters-0.076
 else-0.075
 rides-0.074
 scored-0.072
Token to
Token ID284
Feature activation+3.03e-04

Pos logit contributions
wood+0.191
man+0.181
tha+0.172
fly+0.171
work+0.169
Neg logit contributions
fters-0.226
ets-0.208
 tant-0.203
 mats-0.199
pler-0.198
Token build
Token ID1382
Feature activation-6.82e-02

Pos logit contributions
fly+0.001
wood+0.001
tha+0.001
flower+0.001
 extingu+0.001
Neg logit contributions
fters-0.000
 mats-0.000
 rides-0.000
 quizz-0.000
 else-0.000
Token a
Token ID257
Feature activation+1.30e-01

Pos logit contributions
fters+0.154
ets+0.143
itionally+0.137
 tant+0.137
 mats+0.135
Neg logit contributions
wood-0.084
 extingu-0.076
man-0.074
fly-0.070
work-0.069
Token snow
Token ID6729
Feature activation+1.55e-01

Pos logit contributions
wood+0.241
fly+0.226
man+0.224
flower+0.223
 extingu+0.222
Neg logit contributions
fters-0.207
ets-0.193
 mats-0.184
 quizz-0.182
 scores-0.181
Tokenman
Token ID805
Feature activation+5.13e-01

Pos logit contributions
wood+0.149
tha+0.149
 Mill+0.132
 Frankie+0.129
fly+0.129
Neg logit contributions
fters-0.355
ets-0.349
 mats-0.348
 anymore-0.347
 quizz-0.338
Token.
Token ID13
Feature activation+1.86e-03

Pos logit contributions
wood+0.916
man+0.888
fly+0.872
flower+0.868
work+0.860
Neg logit contributions
fters-0.845
 mats-0.780
 tant-0.764
ets-0.760
 quizz-0.733
Token He
Token ID679
Feature activation+3.63e-02

Pos logit contributions
wood+0.003
man+0.003
 extingu+0.003
flower+0.003
fly+0.003
Neg logit contributions
fters-0.004
ets-0.003
 else-0.003
pler-0.003
 mats-0.003
Token grabbed
Token ID13176
Feature activation-5.60e-02

Pos logit contributions
wood+0.073
fly+0.069
flower+0.068
tha+0.068
 extingu+0.068
Neg logit contributions
fters-0.050
ets-0.048
 else-0.047
 mats-0.046
 scored-0.045
Token some
Token ID617
Feature activation-7.17e-02

Pos logit contributions
fters+0.091
ets+0.086
 quizz+0.085
 mats+0.084
 scores+0.083
Neg logit contributions
wood-0.099
fly-0.095
flower-0.094
tha-0.093
man-0.092
Token big
Token ID1263
Feature activation+5.44e-02

Pos logit contributions
fters+0.118
ets+0.109
 mats+0.105
 quizz+0.105
 scores+0.105
Neg logit contributions
wood-0.129
 extingu-0.120
fly-0.119
flower-0.119
man-0.118
Token there
Token ID612
Feature activation-1.73e-02

Pos logit contributions
fters+0.191
ets+0.177
 mats+0.171
 quizz+0.170
 scores+0.168
Neg logit contributions
wood-0.214
fly-0.202
flower-0.199
man-0.198
tha-0.198
Token was
Token ID373
Feature activation+2.49e-02

Pos logit contributions
fters+0.026
ets+0.024
 else+0.023
 mats+0.023
 quizz+0.023
Neg logit contributions
wood-0.033
fly-0.031
 extingu-0.031
flower-0.031
tha-0.031
Token a
Token ID257
Feature activation+1.33e-01

Pos logit contributions
wood+0.048
fly+0.045
tha+0.044
man+0.044
 extingu+0.043
Neg logit contributions
fters-0.038
 mats-0.037
 else-0.036
 quizz-0.036
 rides-0.036
Token humble
Token ID19695
Feature activation-4.82e-02

Pos logit contributions
wood+0.293
fly+0.276
flower+0.273
 extingu+0.272
tha+0.271
Neg logit contributions
fters-0.165
ets-0.149
 mats-0.145
 quizz-0.143
 else-0.141
Token fire
Token ID2046
Feature activation+1.14e-01

Pos logit contributions
fters+0.061
 mats+0.059
 quizz+0.058
ets+0.058
 scores+0.056
Neg logit contributions
wood-0.101
fly-0.093
 extingu-0.092
tha-0.091
flower-0.090
Tokenman
Token ID805
Feature activation+5.12e-01

Pos logit contributions
wood+0.151
tha+0.147
 Frankie+0.134
 extingu+0.133
 Mill+0.133
Neg logit contributions
 anymore-0.211
ets-0.198
 else-0.196
 quizz-0.191
 scores-0.191
Token.
Token ID13
Feature activation+6.64e-05

Pos logit contributions
wood+0.993
fly+0.935
flower+0.925
man+0.919
tha+0.919
Neg logit contributions
fters-0.765
ets-0.708
 mats-0.683
 quizz-0.677
 scores-0.670
Token He
Token ID679
Feature activation+4.03e-02

Pos logit contributions
wood+0.000
tha+0.000
fly+0.000
flower+0.000
flake+0.000
Neg logit contributions
fters-0.000
 quizz-0.000
 mats-0.000
ets-0.000
 Many-0.000
Token worked
Token ID3111
Feature activation+1.01e-01

Pos logit contributions
wood+0.089
tha+0.084
 extingu+0.084
flower+0.084
fly+0.081
Neg logit contributions
fters-0.047
ets-0.045
 else-0.045
 scored-0.043
 rides-0.043
Token very
Token ID845
Feature activation+1.21e-01

Pos logit contributions
wood+0.153
fly+0.140
man+0.139
flower+0.137
 extingu+0.136
Neg logit contributions
fters-0.194
ets-0.180
 mats-0.177
 scores-0.174
 quizz-0.174
Token hard
Token ID1327
Feature activation+1.22e-01

Pos logit contributions
wood+0.145
man+0.133
fly+0.130
tha+0.125
work+0.123
Neg logit contributions
fters-0.268
 mats-0.244
ets-0.243
pler-0.243
pieces-0.243
Token,
Token ID11
Feature activation+1.95e-02

Pos logit contributions
fters+0.193
ets+0.180
 mats+0.173
 quizz+0.172
 scores+0.170
Neg logit contributions
wood-0.218
fly-0.205
flower-0.202
man-0.201
tha-0.201
Token there
Token ID612
Feature activation-1.94e-02

Pos logit contributions
wood+0.041
fly+0.038
 extingu+0.038
flower+0.037
tha+0.037
Neg logit contributions
fters-0.026
ets-0.024
 else-0.024
 mats-0.023
 quizz-0.023
Token was
Token ID373
Feature activation+2.35e-02

Pos logit contributions
fters+0.031
ets+0.029
 mats+0.028
 quizz+0.028
 scores+0.028
Neg logit contributions
wood-0.036
fly-0.033
flower-0.033
 extingu-0.033
man-0.033
Token a
Token ID257
Feature activation+1.31e-01

Pos logit contributions
wood+0.045
fly+0.042
tha+0.042
man+0.041
flower+0.040
Neg logit contributions
fters-0.036
 mats-0.035
 else-0.034
 rides-0.034
 quizz-0.034
Token fire
Token ID2046
Feature activation+1.10e-01

Pos logit contributions
wood+0.301
fly+0.281
flower+0.276
 extingu+0.276
tha+0.276
Neg logit contributions
fters-0.152
ets-0.134
 mats-0.132
 quizz-0.132
 else-0.132
Tokenman
Token ID805
Feature activation+5.12e-01

Pos logit contributions
wood+0.154
tha+0.146
 Frankie+0.137
 Mill+0.136
 Martha+0.134
Neg logit contributions
 anymore-0.200
 else-0.185
ets-0.180
 corners-0.179
 mats-0.178
Token who
Token ID508
Feature activation-1.33e-01

Pos logit contributions
wood+1.004
fly+0.947
flower+0.937
man+0.932
 extingu+0.929
Neg logit contributions
fters-0.757
ets-0.698
 mats-0.674
 quizz-0.667
 scores-0.660
Token was
Token ID373
Feature activation+2.30e-02

Pos logit contributions
 else+0.123
ets+0.117
 scored+0.115
 rides+0.115
 meets+0.106
Neg logit contributions
tha-0.315
wood-0.310
flower-0.304
 extingu-0.301
burn-0.297
Token so
Token ID523
Feature activation+2.27e-02

Pos logit contributions
wood+0.046
tha+0.046
flower+0.045
flake+0.045
fly+0.044
Neg logit contributions
 else-0.028
fters-0.028
 quizz-0.027
 rides-0.026
 corners-0.026
Token nice
Token ID3621
Feature activation-2.66e-02

Pos logit contributions
flower+0.050
fly+0.050
flake+0.050
wood+0.049
 extingu+0.047
Neg logit contributions
 else-0.028
fters-0.026
 Many-0.023
 quizz-0.023
 laps-0.023
Token and
Token ID290
Feature activation+1.57e-02

Pos logit contributions
 anymore+0.025
ets+0.025
 mats+0.023
 else+0.023
 quizz+0.023
Neg logit contributions
tha-0.063
fly-0.060
wood-0.059
flower-0.059
 Amen-0.059
Token said
Token ID531
Feature activation+1.20e-01

Pos logit contributions
wood+0.067
 extingu+0.064
fly+0.062
man+0.062
flower+0.061
Neg logit contributions
fters-0.046
 else-0.044
ets-0.044
 scores-0.043
 quizz-0.043
Token:
Token ID25
Feature activation+9.21e-02

Pos logit contributions
wood+0.239
fly+0.225
flower+0.223
 extingu+0.222
tha+0.219
Neg logit contributions
fters-0.175
ets-0.161
 quizz-0.158
 mats-0.156
 else-0.156
Token "
Token ID366
Feature activation-4.22e-02

Pos logit contributions
wood+0.201
fly+0.188
 extingu+0.188
tha+0.187
flower+0.186
Neg logit contributions
fters-0.115
 quizz-0.102
ets-0.102
 mats-0.100
 scores-0.100
TokenThe
Token ID464
Feature activation-1.51e-02

Pos logit contributions
fters+0.059
and+0.054
ets+0.053
 else+0.053
 scored+0.051
Neg logit contributions
wood-0.084
 extingu-0.082
tha-0.078
fly-0.077
man-0.075
Token mail
Token ID6920
Feature activation+1.50e-01

Pos logit contributions
fters+0.022
 mats+0.021
 corners+0.021
 quizz+0.020
 else+0.020
Neg logit contributions
wood-0.027
fly-0.026
tha-0.025
work-0.025
 extingu-0.024
Tokenman
Token ID805
Feature activation+5.12e-01

Pos logit contributions
wood+0.265
 Tit+0.247
 extingu+0.245
tha+0.235
burn+0.234
Neg logit contributions
ets-0.217
 else-0.215
 Find-0.211
pler-0.204
fters-0.202
Token always
Token ID1464
Feature activation+6.74e-02

Pos logit contributions
wood+0.974
 extingu+0.914
tha+0.891
fly+0.884
flower+0.881
Neg logit contributions
fters-0.764
ets-0.741
 else-0.723
 scores-0.712
 quizz-0.698
Token has
Token ID468
Feature activation+2.03e-02

Pos logit contributions
wood+0.126
flower+0.118
 extingu+0.118
fly+0.118
tha+0.117
Neg logit contributions
fters-0.103
ets-0.100
 scores-0.094
 else-0.094
 quizz-0.094
Token the
Token ID262
Feature activation+1.93e-02

Pos logit contributions
wood+0.034
fly+0.032
flower+0.031
man+0.031
tha+0.031
Neg logit contributions
fters-0.036
ets-0.033
 mats-0.032
 quizz-0.032
 scores-0.032
Token best
Token ID1266
Feature activation+7.09e-02

Pos logit contributions
wood+0.033
fly+0.032
tha+0.031
flower+0.031
 extingu+0.030
Neg logit contributions
fters-0.031
ets-0.030
 quizz-0.029
 mats-0.028
 scores-0.028
Token stories
Token ID3923
Feature activation-8.14e-02

Pos logit contributions
wood+0.133
fly+0.128
tha+0.126
flower+0.125
 extingu+0.122
Neg logit contributions
fters-0.106
ets-0.103
 mats-0.097
 quizz-0.097
 scores-0.096
Token.
Token ID13
Feature activation+9.71e-03

Pos logit contributions
wood+0.363
tha+0.339
 extingu+0.336
fly+0.334
flower+0.333
Neg logit contributions
fters-0.269
ets-0.255
 mats-0.244
 else-0.243
 scores-0.242
Token<|endoftext|>
Token ID50256
Feature activation-6.40e-02

Pos logit contributions
wood+0.019
fly+0.018
flower+0.017
man+0.017
tha+0.017
Neg logit contributions
fters-0.015
ets-0.014
 mats-0.013
 quizz-0.013
 scores-0.013
TokenA
Token ID32
Feature activation+6.60e-02

Pos logit contributions
fters+0.097
ets+0.090
 mats+0.086
 quizz+0.086
 scores+0.085
Neg logit contributions
wood-0.123
fly-0.116
flower-0.114
man-0.114
tha-0.113
Token famous
Token ID5863
Feature activation-2.47e-03

Pos logit contributions
wood+0.153
fly+0.144
flower+0.143
 extingu+0.143
tha+0.142
Neg logit contributions
fters-0.075
ets-0.068
 mats-0.065
 quizz-0.065
 else-0.064
Token post
Token ID1281
Feature activation+1.01e-01

Pos logit contributions
fters+0.004
ets+0.004
 mats+0.004
 quizz+0.004
 scores+0.004
Neg logit contributions
wood-0.004
fly-0.004
 extingu-0.004
flower-0.004
tha-0.004
Tokenman
Token ID805
Feature activation+5.11e-01

Pos logit contributions
wood+0.195
 extingu+0.187
tha+0.177
fly+0.176
flower+0.172
Neg logit contributions
fters-0.138
ets-0.133
 else-0.131
 scores-0.128
 mats-0.127
Token named
Token ID3706
Feature activation-3.07e-03

Pos logit contributions
wood+0.951
fly+0.900
flower+0.890
man+0.886
tha+0.876
Neg logit contributions
fters-0.807
ets-0.734
 mats-0.719
 quizz-0.709
 scores-0.699
Token Bob
Token ID5811
Feature activation+2.91e-01

Pos logit contributions
fters+0.006
ets+0.006
 mats+0.005
 quizz+0.005
 scores+0.005
Neg logit contributions
wood-0.005
fly-0.004
flower-0.004
man-0.004
 extingu-0.004
Token was
Token ID373
Feature activation+2.76e-02

Pos logit contributions
wood+0.465
 extingu+0.446
tha+0.440
work+0.434
 Amen+0.415
Neg logit contributions
ets-0.486
 else-0.470
 meets-0.465
 scored-0.459
fters-0.456
Token out
Token ID503
Feature activation+5.39e-02

Pos logit contributions
wood+0.051
fly+0.049
flower+0.048
 extingu+0.048
tha+0.047
Neg logit contributions
fters-0.044
ets-0.040
 quizz-0.040
 scores-0.039
 mats-0.038
Token delivering
Token ID13630
Feature activation+1.49e-01

Pos logit contributions
wood+0.101
fly+0.095
flower+0.094
 extingu+0.093
man+0.093
Neg logit contributions
fters-0.086
ets-0.079
 mats-0.076
 scores-0.072
 quizz-0.072
Token there
Token ID612
Feature activation-1.83e-02

Pos logit contributions
wood+0.033
fly+0.031
 extingu+0.030
flower+0.030
tha+0.030
Neg logit contributions
fters-0.023
ets-0.021
 mats-0.020
 else-0.020
 quizz-0.020
Token was
Token ID373
Feature activation+2.18e-02

Pos logit contributions
fters+0.030
ets+0.028
 mats+0.027
 quizz+0.027
 scores+0.027
Neg logit contributions
wood-0.033
fly-0.031
flower-0.030
man-0.030
tha-0.030
Token an
Token ID281
Feature activation+1.04e-01

Pos logit contributions
wood+0.039
fly+0.036
tha+0.036
man+0.035
flower+0.035
Neg logit contributions
fters-0.036
 mats-0.035
 quizz-0.034
ets-0.034
 else-0.034
Token attractive
Token ID10966
Feature activation-1.46e-02

Pos logit contributions
wood+0.196
fly+0.186
man+0.179
flower+0.179
 extingu+0.178
Neg logit contributions
fters-0.162
ets-0.152
 quizz-0.148
 mats-0.147
 else-0.144
Token sun
Token ID4252
Feature activation+5.62e-02

Pos logit contributions
fters+0.021
 mats+0.020
 quizz+0.020
ets+0.019
 scores+0.019
Neg logit contributions
wood-0.029
fly-0.026
 extingu-0.026
tha-0.026
man-0.025
Tokenflower
Token ID25547
Feature activation+5.11e-01

Pos logit contributions
wood+0.083
 extingu+0.072
tha+0.070
 Amen+0.068
 Frankie+0.067
Neg logit contributions
 anymore-0.100
fters-0.098
ets-0.096
 dresses-0.096
 mats-0.096
Token growing
Token ID3957
Feature activation+1.40e-01

Pos logit contributions
wood+1.059
 extingu+0.995
tha+0.990
fly+0.975
man+0.950
Neg logit contributions
fters-0.674
ets-0.638
 else-0.608
 mats-0.605
 quizz-0.604
Token in
Token ID287
Feature activation-3.01e-02

Pos logit contributions
 extingu+0.350
tha+0.340
 Amen+0.332
cle+0.319
wood+0.319
Neg logit contributions
ets-0.128
 corners-0.111
ag-0.108
fters-0.108
les-0.107
Token a
Token ID257
Feature activation+1.29e-01

Pos logit contributions
 corners+0.030
 quizz+0.027
fters+0.027
 rides+0.026
 holes+0.025
Neg logit contributions
wood-0.073
fly-0.070
tha-0.070
man-0.069
 extingu-0.069
Token garden
Token ID11376
Feature activation-4.28e-02

Pos logit contributions
wood+0.214
flower+0.210
man+0.208
fly+0.208
tha+0.199
Neg logit contributions
fters-0.222
ets-0.214
 mats-0.201
 scores-0.198
 quizz-0.197
Token.
Token ID13
Feature activation-3.43e-03

Pos logit contributions
fters+0.063
ets+0.058
 mats+0.056
 quizz+0.056
 scores+0.055
Neg logit contributions
wood-0.084
fly-0.079
flower-0.078
tha-0.078
 extingu-0.078
Token He
Token ID679
Feature activation+4.24e-02

Pos logit contributions
wood+0.305
fly+0.281
man+0.281
flower+0.279
tha+0.277
Neg logit contributions
fters-0.277
ets-0.254
 mats-0.244
 quizz-0.241
 scores-0.238
Token thanked
Token ID26280
Feature activation+4.14e-02

Pos logit contributions
wood+0.079
fly+0.074
flower+0.073
 extingu+0.073
man+0.072
Neg logit contributions
fters-0.067
ets-0.063
 mats-0.060
 quizz-0.060
 scores-0.059
Token the
Token ID262
Feature activation+2.48e-02

Pos logit contributions
wood+0.083
fly+0.079
flower+0.078
 extingu+0.076
man+0.075
Neg logit contributions
fters-0.060
 quizz-0.055
ets-0.054
 mats-0.054
 scores-0.053
Token repair
Token ID9185
Feature activation+1.20e-02

Pos logit contributions
wood+0.047
fly+0.044
flower+0.043
tha+0.043
 extingu+0.043
Neg logit contributions
fters-0.039
ets-0.035
 mats-0.035
 quizz-0.035
 else-0.034
Token-
Token ID12
Feature activation+6.61e-02

Pos logit contributions
wood+0.019
tha+0.018
fly+0.017
flower+0.017
 extingu+0.017
Neg logit contributions
fters-0.020
ets-0.020
 quizz-0.020
 tant-0.019
 else-0.019
Tokenman
Token ID805
Feature activation+5.10e-01

Pos logit contributions
wood+0.124
fly+0.117
 extingu+0.116
flower+0.115
tha+0.115
Neg logit contributions
fters-0.102
ets-0.095
 quizz-0.091
 mats-0.091
 else-0.091
Token and
Token ID290
Feature activation+2.85e-02

Pos logit contributions
wood+0.956
fly+0.904
man+0.895
flower+0.891
 extingu+0.881
Neg logit contributions
fters-0.808
ets-0.729
 mats-0.716
 quizz-0.704
 scores-0.697
Token promised
Token ID8072
Feature activation+1.16e-01

Pos logit contributions
wood+0.056
fly+0.054
flower+0.054
 extingu+0.053
flake+0.051
Neg logit contributions
 else-0.038
 rides-0.036
 scored-0.036
fters-0.036
ets-0.036
Token to
Token ID284
Feature activation+6.30e-04

Pos logit contributions
wood+0.234
fly+0.226
flower+0.226
 extingu+0.221
man+0.218
Neg logit contributions
fters-0.162
ets-0.148
 mats-0.143
 quizz-0.143
 scores-0.141
Token never
Token ID1239
Feature activation+1.33e-01

Pos logit contributions
wood+0.001
 extingu+0.001
fly+0.001
flower+0.001
tha+0.001
Neg logit contributions
fters-0.001
ets-0.001
 rides-0.001
 else-0.001
 quizz-0.001
Token forget
Token ID6044
Feature activation-8.45e-03

Pos logit contributions
wood+0.210
fly+0.202
flower+0.194
work+0.188
 extingu+0.179
Neg logit contributions
fters-0.254
 mats-0.239
ets-0.238
 tant-0.230
pler-0.224
Token smiled
Token ID13541
Feature activation+2.98e-02

Pos logit contributions
wood+0.058
flower+0.055
tha+0.055
fly+0.055
 extingu+0.055
Neg logit contributions
fters-0.055
ets-0.051
 mats-0.049
 quizz-0.049
 scores-0.048
Token and
Token ID290
Feature activation+3.21e-02

Pos logit contributions
wood+0.069
tha+0.066
fly+0.066
 Amen+0.065
 extingu+0.065
Neg logit contributions
 anymore-0.027
ets-0.025
 corners-0.025
 else-0.024
 laps-0.024
Token thanked
Token ID26280
Feature activation+3.68e-02

Pos logit contributions
wood+0.066
fly+0.063
flower+0.062
 extingu+0.062
man+0.061
Neg logit contributions
 else-0.042
fters-0.041
ets-0.039
 scored-0.039
 scores-0.039
Token the
Token ID262
Feature activation+1.67e-02

Pos logit contributions
wood+0.073
fly+0.069
flower+0.069
 extingu+0.067
man+0.066
Neg logit contributions
fters-0.053
 quizz-0.050
 mats-0.048
ets-0.048
 else-0.048
Token post
Token ID1281
Feature activation+1.02e-01

Pos logit contributions
wood+0.029
man+0.028
flower+0.027
fly+0.027
tha+0.027
Neg logit contributions
fters-0.028
ets-0.027
 mats-0.025
 quizz-0.025
 scores-0.025
Tokenman
Token ID805
Feature activation+5.10e-01

Pos logit contributions
wood+0.128
 extingu+0.111
fly+0.109
tha+0.106
flower+0.105
Neg logit contributions
fters-0.210
 else-0.204
ets-0.201
 quizz-0.199
 anymore-0.197
Token.
Token ID13
Feature activation+1.56e-02

Pos logit contributions
wood+0.979
fly+0.932
man+0.925
flower+0.914
 extingu+0.903
Neg logit contributions
fters-0.793
ets-0.706
 mats-0.693
 quizz-0.677
 tant-0.666
Token Comfort
Token ID45769
Feature activation+5.75e-02

Pos logit contributions
wood+0.035
flower+0.035
fly+0.035
 extingu+0.034
burn+0.033
Neg logit contributions
fters-0.015
 quizz-0.015
 scores-0.015
and-0.015
 calculations-0.014
Tokenably
Token ID1346
Feature activation+9.68e-02

Pos logit contributions
wood+0.109
 extingu+0.107
 Amen+0.103
fly+0.102
flower+0.102
Neg logit contributions
fters-0.082
ets-0.082
 else-0.077
 scores-0.074
and-0.073
Token eating
Token ID6600
Feature activation+2.41e-02

Pos logit contributions
wood+0.174
fly+0.164
flower+0.161
man+0.161
tha+0.160
Neg logit contributions
fters-0.159
ets-0.147
 mats-0.143
 quizz-0.141
 scores-0.140
Token his
Token ID465
Feature activation-9.42e-03

Pos logit contributions
wood+0.044
fly+0.041
man+0.041
flower+0.040
 extingu+0.040
Neg logit contributions
fters-0.039
ets-0.036
 mats-0.035
 quizz-0.034
 scores-0.034
Token was
Token ID373
Feature activation+2.59e-02

Pos logit contributions
wood+0.094
tha+0.088
 extingu+0.087
flower+0.087
work+0.085
Neg logit contributions
fters-0.042
 else-0.040
ets-0.039
 scored-0.038
 rides-0.038
Token a
Token ID257
Feature activation+1.31e-01

Pos logit contributions
wood+0.053
tha+0.051
flake+0.051
fly+0.051
flower+0.051
Neg logit contributions
 quizz-0.032
fters-0.032
 else-0.031
 rides-0.030
 Many-0.029
Token very
Token ID845
Feature activation+1.22e-01

Pos logit contributions
wood+0.280
tha+0.250
flower+0.249
fly+0.249
work+0.243
Neg logit contributions
 else-0.169
fters-0.168
 quizz-0.160
 tant-0.153
ets-0.152
Token brave
Token ID14802
Feature activation+1.49e-02

Pos logit contributions
wood+0.241
fly+0.227
flower+0.226
 extingu+0.223
tha+0.220
Neg logit contributions
fters-0.177
ets-0.164
 else-0.161
 quizz-0.159
 mats-0.155
Token fire
Token ID2046
Feature activation+1.13e-01

Pos logit contributions
wood+0.027
fly+0.025
flower+0.025
tha+0.025
work+0.024
Neg logit contributions
fters-0.023
ets-0.023
 quizz-0.022
 mats-0.022
 anymore-0.021
Tokenman
Token ID805
Feature activation+5.09e-01

Pos logit contributions
wood+0.063
tha+0.059
 Mill+0.051
 Frankie+0.048
 Martha+0.045
Neg logit contributions
 anymore-0.297
ets-0.292
 else-0.289
 quizz-0.289
 mats-0.283
Token.
Token ID13
Feature activation+7.89e-03

Pos logit contributions
wood+0.932
man+0.906
fly+0.904
flower+0.886
work+0.864
Neg logit contributions
fters-0.835
ets-0.748
 mats-0.745
 tant-0.730
 scores-0.713
Token Every
Token ID3887
Feature activation+1.04e-02

Pos logit contributions
wood+0.015
fly+0.014
man+0.014
flower+0.013
 extingu+0.013
Neg logit contributions
fters-0.013
ets-0.012
 mats-0.011
 else-0.011
 scores-0.011
Token day
Token ID1110
Feature activation+1.25e-01

Pos logit contributions
tha+0.022
 extingu+0.022
wood+0.022
flower+0.021
fly+0.020
Neg logit contributions
ets-0.012
 quizz-0.011
fters-0.011
 rides-0.011
ag-0.011
Token,
Token ID11
Feature activation+1.60e-02

Pos logit contributions
wood+0.219
man+0.205
fly+0.205
flower+0.202
tha+0.197
Neg logit contributions
fters-0.214
ets-0.196
 mats-0.189
 tant-0.185
 scores-0.184
Token he
Token ID339
Feature activation-6.40e-02

Pos logit contributions
wood+0.034
fly+0.032
flower+0.032
 extingu+0.031
tha+0.031
Neg logit contributions
fters-0.022
ets-0.020
 mats-0.019
 quizz-0.019
 scores-0.019
Token<|endoftext|>
Token ID50256
Feature activation-6.29e-02

Pos logit contributions
wood+0.017
fly+0.016
flower+0.016
man+0.016
tha+0.016
Neg logit contributions
fters-0.014
ets-0.013
 mats-0.013
 scores-0.013
 quizz-0.013
TokenJim
Token ID18050
Feature activation+9.37e-02

Pos logit contributions
fters+0.089
ets+0.082
 mats+0.080
 quizz+0.079
 scores+0.079
Neg logit contributions
wood-0.125
fly-0.118
flower-0.117
 extingu-0.117
tha-0.117
Token was
Token ID373
Feature activation+2.48e-02

Pos logit contributions
wood+0.196
 extingu+0.189
tha+0.184
work+0.183
fly+0.183
Neg logit contributions
ets-0.117
fters-0.110
 else-0.108
 scores-0.105
 meets-0.104
Token a
Token ID257
Feature activation+1.32e-01

Pos logit contributions
wood+0.052
fly+0.050
 extingu+0.048
flower+0.048
tha+0.048
Neg logit contributions
fters-0.034
ets-0.031
 quizz-0.030
 mats-0.029
 scores-0.029
Token fire
Token ID2046
Feature activation+1.16e-01

Pos logit contributions
wood+0.291
fly+0.266
 extingu+0.265
tha+0.259
man+0.259
Neg logit contributions
fters-0.163
 quizz-0.154
 else-0.152
ets-0.148
 mats-0.147
Tokenman
Token ID805
Feature activation+5.09e-01

Pos logit contributions
wood+0.135
tha+0.133
 Amen+0.130
 Frankie+0.126
 Mill+0.124
Neg logit contributions
 anymore-0.230
ets-0.208
 corners-0.205
 mats-0.203
 else-0.203
Token.
Token ID13
Feature activation-3.82e-03

Pos logit contributions
wood+0.984
fly+0.931
flower+0.921
man+0.917
 extingu+0.910
Neg logit contributions
fters-0.764
ets-0.703
 mats-0.682
 quizz-0.673
 scores-0.667
Token He
Token ID679
Feature activation+3.45e-02

Pos logit contributions
 Many+0.004
 corners+0.003
 Find+0.003
ets+0.003
 scores+0.003
Neg logit contributions
burn-0.009
tha-0.009
wood-0.008
flake-0.008
flower-0.008
Token was
Token ID373
Feature activation+2.13e-02

Pos logit contributions
wood+0.070
flower+0.065
 extingu+0.065
tha+0.065
fly+0.064
Neg logit contributions
fters-0.048
ets-0.046
 else-0.044
 mats-0.044
 quizz-0.043
Token very
Token ID845
Feature activation+1.18e-01

Pos logit contributions
wood+0.042
flower+0.040
 extingu+0.040
fly+0.040
tha+0.039
Neg logit contributions
fters-0.029
 else-0.027
 quizz-0.027
 rides-0.026
 corners-0.026
Token strong
Token ID1913
Feature activation+1.41e-01

Pos logit contributions
wood+0.262
flower+0.256
fly+0.254
 extingu+0.249
flake+0.249
Neg logit contributions
fters-0.144
 else-0.142
 quizz-0.129
ets-0.128
 tant-0.121
Token.
Token ID13
Feature activation+9.74e-03

Pos logit contributions
fters+0.126
ets+0.120
 mats+0.118
 quizz+0.116
 else+0.115
Neg logit contributions
wood-0.193
tha-0.185
fly-0.180
flower-0.178
 extingu-0.177
Token It
Token ID632
Feature activation-4.02e-02

Pos logit contributions
wood+0.016
man+0.014
fly+0.014
 extingu+0.014
 Mill+0.014
Neg logit contributions
fters-0.019
ets-0.018
 tant-0.016
 else-0.016
 scores-0.016
Token was
Token ID373
Feature activation+2.37e-02

Pos logit contributions
fters+0.052
 else+0.051
ets+0.048
 mats+0.047
 scored+0.047
Neg logit contributions
wood-0.085
tha-0.080
 extingu-0.079
flower-0.077
work-0.076
Token a
Token ID257
Feature activation+1.31e-01

Pos logit contributions
wood+0.045
tha+0.044
 Amen+0.042
fly+0.041
flower+0.041
Neg logit contributions
 mats-0.034
 quizz-0.033
 else-0.033
fters-0.033
 corners-0.032
Token trades
Token ID17674
Feature activation-2.10e-01

Pos logit contributions
wood+0.276
tha+0.250
 extingu+0.242
fly+0.239
flower+0.239
Neg logit contributions
fters-0.172
 else-0.169
 mats-0.164
 quizz-0.164
 scores-0.161
Tokenman
Token ID805
Feature activation+5.09e-01

Pos logit contributions
 scores+0.252
 else+0.248
 anymore+0.243
 scored+0.242
ets+0.241
Neg logit contributions
wood-0.443
tha-0.423
 Amen-0.399
fly-0.395
flower-0.395
Token and
Token ID290
Feature activation+2.24e-02

Pos logit contributions
wood+0.987
fly+0.916
tha+0.909
flower+0.909
 extingu+0.908
Neg logit contributions
fters-0.759
ets-0.706
 else-0.676
 scores-0.675
 quizz-0.674
Token he
Token ID339
Feature activation-6.21e-02

Pos logit contributions
wood+0.049
 extingu+0.046
flower+0.046
tha+0.046
fly+0.046
Neg logit contributions
fters-0.027
 else-0.025
ets-0.025
 mats-0.025
 quizz-0.024
Token was
Token ID373
Feature activation+2.36e-02

Pos logit contributions
fters+0.094
ets+0.087
 mats+0.084
 quizz+0.084
 scores+0.084
Neg logit contributions
wood-0.119
fly-0.111
 extingu-0.111
flower-0.111
tha-0.110
Token carrying
Token ID6872
Feature activation+7.89e-02

Pos logit contributions
wood+0.046
tha+0.044
 extingu+0.044
flower+0.043
fly+0.043
Neg logit contributions
fters-0.034
 quizz-0.032
 else-0.031
 scores-0.031
ets-0.030
Token a
Token ID257
Feature activation+1.31e-01

Pos logit contributions
wood+0.110
 extingu+0.098
man+0.097
fly+0.097
work+0.095
Neg logit contributions
fters-0.165
ets-0.154
 mats-0.147
 scores-0.145
 quizz-0.145
Token excited
Token ID6568
Feature activation+6.75e-02

Pos logit contributions
wood+0.233
fly+0.219
man+0.216
flower+0.216
tha+0.216
Neg logit contributions
fters-0.178
ets-0.166
 mats-0.160
 quizz-0.155
 scores-0.154
Token and
Token ID290
Feature activation+2.81e-02

Pos logit contributions
wood+0.125
fly+0.117
flower+0.116
man+0.116
 extingu+0.115
Neg logit contributions
fters-0.108
ets-0.099
 mats-0.097
 quizz-0.096
 scores-0.095
Token asked
Token ID1965
Feature activation-1.38e-02

Pos logit contributions
wood+0.057
fly+0.055
 extingu+0.054
flower+0.054
tha+0.053
Neg logit contributions
fters-0.039
ets-0.036
 else-0.035
 quizz-0.034
 mats-0.034
Token the
Token ID262
Feature activation+1.84e-02

Pos logit contributions
fters+0.025
ets+0.023
 mats+0.023
 scores+0.022
 loses+0.022
Neg logit contributions
wood-0.023
man-0.022
tha-0.021
fly-0.021
 extingu-0.021
Token fire
Token ID2046
Feature activation+1.15e-01

Pos logit contributions
wood+0.033
fly+0.031
flower+0.031
tha+0.030
 extingu+0.030
Neg logit contributions
fters-0.030
 else-0.027
 mats-0.027
ets-0.027
 quizz-0.027
Tokenman
Token ID805
Feature activation+5.09e-01

Pos logit contributions
tha+0.103
burn+0.095
 Amen+0.093
wood+0.092
 Frankie+0.086
Neg logit contributions
 else-0.257
 anymore-0.255
 nicely-0.245
ets-0.245
pler-0.241
Token if
Token ID611
Feature activation+1.40e-02

Pos logit contributions
wood+0.919
fly+0.867
man+0.858
flower+0.842
work+0.837
Neg logit contributions
fters-0.871
ets-0.788
 mats-0.774
 tant-0.741
 scores-0.739
Token he
Token ID339
Feature activation-6.32e-02

Pos logit contributions
wood+0.032
flower+0.031
 extingu+0.030
fly+0.030
tha+0.029
Neg logit contributions
fters-0.016
 else-0.015
 quizz-0.014
ets-0.014
 scores-0.014
Token could
Token ID714
Feature activation-8.49e-02

Pos logit contributions
fters+0.074
ets+0.070
 else+0.070
 scores+0.066
 quizz+0.065
Neg logit contributions
wood-0.141
flower-0.132
 extingu-0.132
fly-0.132
tha-0.131
Token help
Token ID1037
Feature activation+1.88e-02

Pos logit contributions
fters+0.112
 else+0.101
ets+0.101
 scores+0.097
 anymore+0.097
Neg logit contributions
wood-0.178
fly-0.172
tha-0.170
flower-0.168
man-0.167
Token.
Token ID13
Feature activation+9.43e-03

Pos logit contributions
wood+0.034
fly+0.032
man+0.032
 extingu+0.032
tha+0.031
Neg logit contributions
fters-0.031
ets-0.029
 mats-0.028
 scores-0.027
 quizz-0.027
Token
Token ID198
Feature activation-2.94e-03

Pos logit contributions
wood+0.013
fly+0.013
flower+0.012
man+0.012
 extingu+0.012
Neg logit contributions
fters-0.008
ets-0.008
 mats-0.008
 quizz-0.007
 scores-0.007
Token
Token ID198
Feature activation-9.08e-04

Pos logit contributions
fters+0.004
ets+0.004
 mats+0.004
 quizz+0.004
 scores+0.004
Neg logit contributions
wood-0.006
fly-0.005
man-0.005
tha-0.005
flower-0.005
TokenThe
Token ID464
Feature activation-1.78e-02

Pos logit contributions
fters+0.001
ets+0.001
 quizz+0.001
 mats+0.001
 scores+0.001
Neg logit contributions
wood-0.002
fly-0.002
flower-0.002
man-0.002
tha-0.002
Token bald
Token ID28273
Feature activation-4.56e-02

Pos logit contributions
fters+0.028
ets+0.026
 mats+0.025
 quizz+0.025
 scores+0.025
Neg logit contributions
wood-0.033
fly-0.031
 extingu-0.031
flower-0.031
tha-0.031
Token fire
Token ID2046
Feature activation+1.12e-01

Pos logit contributions
fters+0.089
ets+0.083
 else+0.080
 mats+0.080
 quizz+0.080
Neg logit contributions
wood-0.067
man-0.066
flower-0.063
fly-0.063
tha-0.062
Tokenman
Token ID805
Feature activation+5.09e-01

Pos logit contributions
wood+0.178
 Amen+0.169
 extingu+0.165
tha+0.164
burn+0.158
Neg logit contributions
ets-0.178
 else-0.176
fters-0.173
 scores-0.173
 scored-0.172
Token laughed
Token ID13818
Feature activation-5.98e-02

Pos logit contributions
wood+0.870
fly+0.835
man+0.821
flower+0.820
tha+0.806
Neg logit contributions
fters-0.879
ets-0.804
 mats-0.790
 quizz-0.783
 scores-0.763
Token and
Token ID290
Feature activation+2.68e-02

Pos logit contributions
fters+0.072
ets+0.071
 anymore+0.068
 mats+0.066
 scores+0.066
Neg logit contributions
wood-0.128
tha-0.122
fly-0.121
flower-0.119
 extingu-0.118
Token said
Token ID531
Feature activation+1.16e-01

Pos logit contributions
wood+0.051
flower+0.048
 extingu+0.048
fly+0.048
man+0.047
Neg logit contributions
fters-0.038
 else-0.037
ets-0.037
 scores-0.036
 scored-0.036
Token,
Token ID11
Feature activation+2.81e-02

Pos logit contributions
wood+0.212
fly+0.200
man+0.197
flower+0.196
 extingu+0.195
Neg logit contributions
fters-0.187
ets-0.173
 mats-0.167
 quizz-0.164
 scores-0.163
Token "
Token ID366
Feature activation-4.74e-02

Pos logit contributions
wood+0.063
fly+0.059
flower+0.059
 extingu+0.058
tha+0.058
Neg logit contributions
fters-0.033
ets-0.030
 quizz-0.029
 mats-0.029
 else-0.028
Token
Token ID198
Feature activation+3.34e-03

Pos logit contributions
fters+0.106
ets+0.098
 mats+0.095
 quizz+0.093
 scores+0.093
Neg logit contributions
wood-0.144
fly-0.136
flower-0.135
man-0.134
 extingu-0.134
Token
Token ID198
Feature activation+3.90e-03

Pos logit contributions
wood+0.006
man+0.006
fly+0.006
flower+0.006
tha+0.006
Neg logit contributions
fters-0.005
ets-0.005
 mats-0.005
 quizz-0.005
 scores-0.005
TokenAfter
Token ID3260
Feature activation-1.28e-03

Pos logit contributions
wood+0.007
man+0.006
fly+0.006
flower+0.006
tha+0.006
Neg logit contributions
fters-0.006
ets-0.006
 quizz-0.006
 mats-0.006
 scores-0.006
Token the
Token ID262
Feature activation+1.86e-02

Pos logit contributions
fters+0.002
 quizz+0.002
 mats+0.002
 scores+0.002
ets+0.002
Neg logit contributions
wood-0.002
flower-0.002
fly-0.002
man-0.002
tha-0.002
Token mail
Token ID6920
Feature activation+1.47e-01

Pos logit contributions
wood+0.034
fly+0.031
tha+0.031
flower+0.031
 extingu+0.031
Neg logit contributions
fters-0.030
ets-0.028
 mats-0.028
 scores-0.027
 quizz-0.027
Tokenman
Token ID805
Feature activation+5.09e-01

Pos logit contributions
wood+0.178
 extingu+0.170
 Amen+0.167
 Tit+0.161
burn+0.161
Neg logit contributions
fters-0.280
 anymore-0.278
 else-0.276
 scored-0.275
ets-0.274
Token left
Token ID1364
Feature activation-1.16e-02

Pos logit contributions
wood+0.867
fly+0.833
man+0.827
tha+0.810
work+0.809
Neg logit contributions
fters-0.887
 mats-0.806
ets-0.799
 quizz-0.780
 tant-0.776
Token,
Token ID11
Feature activation+3.11e-02

Pos logit contributions
fters+0.018
ets+0.017
 mats+0.016
 quizz+0.016
 scores+0.016
Neg logit contributions
wood-0.022
fly-0.021
flower-0.020
 extingu-0.020
man-0.020
Token Tim
Token ID5045
Feature activation+8.41e-02

Pos logit contributions
wood+0.054
fly+0.051
flower+0.050
 extingu+0.050
man+0.049
Neg logit contributions
fters-0.052
 mats-0.048
ets-0.047
 quizz-0.047
 else-0.047
Token ran
Token ID4966
Feature activation+8.80e-03

Pos logit contributions
wood+0.138
 extingu+0.129
fly+0.128
tha+0.126
flower+0.125
Neg logit contributions
fters-0.148
ets-0.142
 scores-0.135
 quizz-0.135
 mats-0.134
Token back
Token ID736
Feature activation-3.52e-02

Pos logit contributions
wood+0.014
man+0.013
 extingu+0.013
tha+0.012
work+0.012
Neg logit contributions
fters-0.017
ets-0.015
 calculations-0.014
 quizz-0.014
 else-0.014
Token can
Token ID460
Feature activation-8.37e-02

Pos logit contributions
fters+0.104
ets+0.096
 mats+0.094
 quizz+0.093
 scores+0.091
Neg logit contributions
wood-0.123
fly-0.116
flower-0.114
man-0.114
 extingu-0.113
Token hurt
Token ID5938
Feature activation+1.00e-02

Pos logit contributions
fters+0.129
 else+0.119
 mats+0.112
ets+0.112
 loses+0.111
Neg logit contributions
wood-0.162
tha-0.152
flower-0.152
 extingu-0.150
fly-0.147
Token you
Token ID345
Feature activation-8.12e-02

Pos logit contributions
wood+0.021
fly+0.020
flower+0.019
man+0.019
 extingu+0.019
Neg logit contributions
fters-0.014
ets-0.013
 mats-0.012
 quizz-0.012
 scores-0.012
Token or
Token ID393
Feature activation-1.44e-01

Pos logit contributions
fters+0.135
ets+0.123
 mats+0.123
 quizz+0.121
 scores+0.119
Neg logit contributions
wood-0.146
fly-0.139
man-0.136
flower-0.135
 extingu-0.133
Token someone
Token ID2130
Feature activation-1.59e-01

Pos logit contributions
fters+0.222
ets+0.205
 mats+0.197
 quizz+0.195
 scores+0.193
Neg logit contributions
wood-0.274
fly-0.260
flower-0.255
man-0.254
 extingu-0.253
Token else
Token ID2073
Feature activation-3.87e-01

Pos logit contributions
fters+0.234
 mats+0.206
 calculations+0.198
 quizz+0.197
 tant+0.194
Neg logit contributions
fly-0.303
wood-0.295
man-0.289
flower-0.282
work-0.280
Token.
Token ID13
Feature activation-3.76e-02

Pos logit contributions
fters+0.631
ets+0.569
 mats+0.562
 quizz+0.556
 scores+0.540
Neg logit contributions
wood-0.708
fly-0.670
man-0.657
flower-0.655
tha-0.642
Token You
Token ID921
Feature activation-3.71e-02

Pos logit contributions
fters+0.051
ets+0.047
 mats+0.045
 quizz+0.045
 scores+0.045
Neg logit contributions
wood-0.077
fly-0.073
flower-0.073
man-0.072
tha-0.072
Token could
Token ID714
Feature activation-1.07e-01

Pos logit contributions
fters+0.057
ets+0.053
 else+0.051
 scores+0.050
 mats+0.050
Neg logit contributions
wood-0.071
flower-0.066
fly-0.066
tha-0.066
 extingu-0.066
Token have
Token ID423
Feature activation-2.70e-02

Pos logit contributions
fters+0.161
ets+0.148
 mats+0.143
 quizz+0.142
 scores+0.141
Neg logit contributions
wood-0.208
fly-0.195
flower-0.193
tha-0.192
man-0.192
Token been
Token ID587
Feature activation+1.50e-01

Pos logit contributions
fters+0.046
ets+0.043
 mats+0.041
 quizz+0.041
 scores+0.040
Neg logit contributions
wood-0.048
fly-0.045
flower-0.044
man-0.044
tha-0.044
Token light
Token ID1657
Feature activation+1.08e-01

Pos logit contributions
wood+0.241
fly+0.226
flower+0.223
tha+0.222
 extingu+0.222
Neg logit contributions
fters-0.211
ets-0.196
 mats-0.190
 quizz-0.188
 scores-0.187
Token.
Token ID13
Feature activation-4.10e-02

Pos logit contributions
wood+0.218
tha+0.208
 extingu+0.205
 Amen+0.202
flower+0.199
Neg logit contributions
ets-0.143
fters-0.137
 anymore-0.137
 else-0.131
 scores-0.130
Token She
Token ID1375
Feature activation+5.17e-02

Pos logit contributions
fters+0.063
ets+0.058
 mats+0.057
 quizz+0.056
 scores+0.055
Neg logit contributions
wood-0.078
fly-0.073
flower-0.072
man-0.072
tha-0.072
Token wondered
Token ID14028
Feature activation+6.15e-02

Pos logit contributions
wood+0.094
fly+0.090
flower+0.088
man+0.088
 extingu+0.087
Neg logit contributions
fters-0.084
ets-0.075
 mats-0.073
 quizz-0.072
 scores-0.071
Token what
Token ID644
Feature activation-8.10e-02

Pos logit contributions
wood+0.121
fly+0.115
flower+0.113
tha+0.113
 extingu+0.113
Neg logit contributions
fters-0.090
ets-0.083
 mats-0.080
 scores-0.079
 quizz-0.078
Token else
Token ID2073
Feature activation-3.85e-01

Pos logit contributions
fters+0.126
ets+0.117
 mats+0.112
 quizz+0.111
 scores+0.110
Neg logit contributions
wood-0.153
fly-0.144
flower-0.142
man-0.142
tha-0.141
Token she
Token ID673
Feature activation-1.84e-02

Pos logit contributions
fters+0.560
ets+0.517
 mats+0.499
 quizz+0.495
 else+0.490
Neg logit contributions
wood-0.761
fly-0.716
flower-0.710
 extingu-0.708
man-0.706
Token could
Token ID714
Feature activation-9.93e-02

Pos logit contributions
fters+0.027
ets+0.024
 mats+0.024
 quizz+0.024
 scores+0.023
Neg logit contributions
wood-0.037
fly-0.035
flower-0.034
man-0.034
tha-0.034
Token learn
Token ID2193
Feature activation-1.31e-01

Pos logit contributions
fters+0.117
ets+0.106
 else+0.101
 mats+0.099
 quizz+0.098
Neg logit contributions
wood-0.223
tha-0.211
fly-0.211
 extingu-0.210
flower-0.209
Token by
Token ID416
Feature activation-5.09e-02

Pos logit contributions
fters+0.216
ets+0.198
 mats+0.191
 scores+0.189
 quizz+0.188
Neg logit contributions
wood-0.234
fly-0.220
man-0.218
flower-0.216
 extingu-0.215
Token guessing
Token ID25260
Feature activation-1.45e-01

Pos logit contributions
fters+0.079
ets+0.073
 mats+0.071
 quizz+0.070
 scores+0.070
Neg logit contributions
wood-0.096
fly-0.090
flower-0.089
man-0.089
 extingu-0.088
Token bear
Token ID6842
Feature activation+6.28e-02

Pos logit contributions
fters+0.069
ets+0.064
 mats+0.062
 quizz+0.061
 scores+0.061
Neg logit contributions
wood-0.082
fly-0.077
flower-0.076
man-0.076
 extingu-0.075
Token.
Token ID13
Feature activation-4.07e-02

Pos logit contributions
wood+0.128
fly+0.121
flower+0.119
 extingu+0.119
tha+0.119
Neg logit contributions
fters-0.088
ets-0.081
 mats-0.078
 quizz-0.077
 scores-0.076
Token It
Token ID632
Feature activation-6.53e-02

Pos logit contributions
fters+0.049
 else+0.045
ets+0.045
 quizz+0.044
 mats+0.044
Neg logit contributions
wood-0.089
flower-0.083
tha-0.083
fly-0.082
 extingu-0.082
Token is
Token ID318
Feature activation-9.90e-03

Pos logit contributions
fters+0.104
ets+0.096
 mats+0.093
 quizz+0.093
 scores+0.091
Neg logit contributions
wood-0.120
fly-0.114
man-0.112
flower-0.112
 extingu-0.111
Token someone
Token ID2130
Feature activation-1.56e-01

Pos logit contributions
fters+0.016
ets+0.015
 mats+0.014
 quizz+0.014
 scores+0.014
Neg logit contributions
wood-0.018
flower-0.017
fly-0.017
tha-0.017
 extingu-0.017
Token else
Token ID2073
Feature activation-3.84e-01

Pos logit contributions
fters+0.218
 mats+0.188
 calculations+0.181
 quizz+0.180
ets+0.179
Neg logit contributions
fly-0.320
wood-0.312
man-0.310
work-0.306
flower-0.304
Token's
Token ID338
Feature activation-2.32e-01

Pos logit contributions
fters+0.599
ets+0.549
 mats+0.535
 quizz+0.529
 scores+0.521
Neg logit contributions
wood-0.723
fly-0.683
flower-0.673
man-0.671
tha-0.665
Token bear
Token ID6842
Feature activation+6.74e-02

Pos logit contributions
fters+0.392
ets+0.362
 mats+0.351
 quizz+0.347
 scores+0.344
Neg logit contributions
wood-0.409
fly-0.385
flower-0.379
man-0.378
 extingu-0.374
Token.
Token ID13
Feature activation-3.16e-02

Pos logit contributions
wood+0.132
fly+0.123
tha+0.123
flower+0.122
 extingu+0.122
Neg logit contributions
fters-0.098
ets-0.092
 mats-0.088
 quizz-0.087
 scores-0.087
Token And
Token ID843
Feature activation-5.45e-02

Pos logit contributions
fters+0.041
ets+0.038
 else+0.037
 mats+0.037
 quizz+0.037
Neg logit contributions
wood-0.066
flower-0.062
tha-0.062
fly-0.062
 extingu-0.061
Token it
Token ID340
Feature activation-3.84e-02

Pos logit contributions
fters+0.080
ets+0.074
 mats+0.071
 quizz+0.071
 scores+0.070
Neg logit contributions
wood-0.107
fly-0.101
flower-0.100
man-0.099
 extingu-0.099
Token or
Token ID393
Feature activation-1.34e-01

Pos logit contributions
tha+0.124
wood+0.122
 Mill+0.121
 Tit+0.117
 Frankie+0.116
Neg logit contributions
 anymore-0.170
ets-0.160
 corners-0.154
fters-0.152
pler-0.151
Token a
Token ID257
Feature activation+1.08e-01

Pos logit contributions
fters+0.233
 mats+0.223
 else+0.220
ets+0.219
 quizz+0.219
Neg logit contributions
wood-0.217
flower-0.209
tha-0.207
fly-0.204
man-0.202
Token smoke
Token ID7523
Feature activation+1.79e-01

Pos logit contributions
wood+0.179
fly+0.166
flower+0.164
man+0.163
tha+0.163
Neg logit contributions
fters-0.191
ets-0.178
 mats-0.174
 quizz-0.172
 scores-0.171
Token or
Token ID393
Feature activation-1.35e-01

Pos logit contributions
wood+0.359
fly+0.337
tha+0.335
flower+0.335
 extingu+0.331
Neg logit contributions
fters-0.252
ets-0.235
 mats-0.224
 quizz-0.221
 scores-0.220
Token something
Token ID1223
Feature activation-1.04e-01

Pos logit contributions
fters+0.230
 mats+0.224
 else+0.219
 quizz+0.219
ets+0.219
Neg logit contributions
wood-0.218
flower-0.210
tha-0.209
man-0.203
fly-0.202
Token else
Token ID2073
Feature activation-3.82e-01

Pos logit contributions
fters+0.169
ets+0.152
 mats+0.149
 quizz+0.147
 scores+0.145
Neg logit contributions
wood-0.190
fly-0.179
man-0.178
 extingu-0.176
work-0.173
Token.
Token ID13
Feature activation-2.18e-02

Pos logit contributions
fters+0.639
ets+0.578
 mats+0.568
 quizz+0.563
 tant+0.557
Neg logit contributions
wood-0.686
fly-0.642
man-0.628
 extingu-0.622
flower-0.622
Token The
Token ID383
Feature activation-8.78e-02

Pos logit contributions
fters+0.033
ets+0.030
 mats+0.029
 quizz+0.029
 scores+0.029
Neg logit contributions
wood-0.042
fly-0.039
flower-0.039
man-0.039
tha-0.039
Token fire
Token ID2046
Feature activation+9.06e-02

Pos logit contributions
fters+0.157
ets+0.147
 mats+0.145
 quizz+0.142
 scores+0.141
Neg logit contributions
wood-0.144
fly-0.135
flower-0.132
tha-0.132
 extingu-0.129
Tokenman
Token ID805
Feature activation+4.81e-01

Pos logit contributions
wood+0.164
 Mill+0.164
tha+0.163
 Frankie+0.159
Mill+0.158
Neg logit contributions
ets-0.124
 anymore-0.118
 scores-0.114
 meets-0.112
pler-0.112
Token are
Token ID389
Feature activation+2.71e-03

Pos logit contributions
wood+0.852
fly+0.836
man+0.815
flower+0.803
work+0.793
Neg logit contributions
fters-0.791
 mats-0.720
 quizz-0.703
ets-0.698
 tant-0.683
Token Maybe
Token ID6674
Feature activation+6.81e-03

Pos logit contributions
fters+0.032
ets+0.029
 mats+0.028
 scores+0.028
 quizz+0.028
Neg logit contributions
wood-0.040
fly-0.038
 extingu-0.037
man-0.037
flower-0.037
Token you
Token ID345
Feature activation-7.88e-02

Pos logit contributions
wood+0.012
man+0.011
fly+0.011
flower+0.011
tha+0.011
Neg logit contributions
fters-0.012
ets-0.011
 tant-0.011
 mats-0.010
 quizz-0.010
Token should
Token ID815
Feature activation+7.27e-02

Pos logit contributions
fters+0.136
 mats+0.125
ets+0.122
 quizz+0.122
 laps+0.119
Neg logit contributions
wood-0.135
fly-0.130
man-0.124
flower-0.124
work-0.124
Token search
Token ID2989
Feature activation-3.01e-02

Pos logit contributions
wood+0.148
fly+0.140
flower+0.139
 extingu+0.137
man+0.137
Neg logit contributions
fters-0.104
ets-0.095
 mats-0.090
 scores-0.090
 else-0.090
Token somewhere
Token ID7382
Feature activation-1.14e-01

Pos logit contributions
fters+0.057
itionally+0.051
 mats+0.049
pler+0.049
 scores+0.047
Neg logit contributions
wood-0.046
man-0.044
work-0.043
 extingu-0.041
fly-0.041
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.167
 mats+0.151
pler+0.141
 calculations+0.139
boards+0.138
Neg logit contributions
wood-0.205
fly-0.204
man-0.202
work-0.196
 possible-0.194
Token."
Token ID526
Feature activation-6.10e-03

Pos logit contributions
fters+0.764
 tant+0.685
 mats+0.682
ets+0.673
 laps+0.664
Neg logit contributions
wood-0.582
fly-0.544
man-0.531
work-0.515
flower-0.489
Token
Token ID198
Feature activation-3.63e-02

Pos logit contributions
fters+0.014
ets+0.012
 laps+0.012
 scores+0.012
 mats+0.011
Neg logit contributions
wood-0.008
 Mill-0.007
man-0.007
fly-0.007
flower-0.007
Token
Token ID198
Feature activation-3.59e-02

Pos logit contributions
fters+0.070
ets+0.065
 quizz+0.062
 mats+0.062
 scores+0.062
Neg logit contributions
man-0.052
wood-0.052
fly-0.050
 extingu-0.050
 Amen-0.049
TokenL
Token ID43
Feature activation-2.08e-01

Pos logit contributions
fters+0.055
 quizz+0.053
ets+0.052
 scores+0.051
 anymore+0.049
Neg logit contributions
man-0.060
 Amen-0.059
flower-0.059
 extingu-0.059
wood-0.057
Tokenily
Token ID813
Feature activation-2.26e-01

Pos logit contributions
fters+0.284
 quizz+0.247
itionally+0.246
 laps+0.246
 calculations+0.246
Neg logit contributions
fly-0.397
flower-0.394
man-0.393
wood-0.392
 extingu-0.391
Token They
Token ID1119
Feature activation+1.20e-01

Pos logit contributions
fters+0.052
ets+0.048
 mats+0.046
 quizz+0.046
 scores+0.046
Neg logit contributions
wood-0.040
fly-0.037
 extingu-0.036
man-0.036
tha-0.036
Token wanted
Token ID2227
Feature activation-2.59e-02

Pos logit contributions
wood+0.210
fly+0.198
tha+0.196
flower+0.195
man+0.194
Neg logit contributions
fters-0.202
ets-0.185
 mats-0.181
 quizz-0.181
 scores-0.178
Token to
Token ID284
Feature activation-2.23e-02

Pos logit contributions
fters+0.046
ets+0.042
 mats+0.039
 scores+0.039
 tant+0.039
Neg logit contributions
wood-0.046
man-0.042
 extingu-0.041
fly-0.041
work-0.041
Token play
Token ID711
Feature activation-1.45e-01

Pos logit contributions
fters+0.024
 else+0.023
 rides+0.023
 anymore+0.022
 scores+0.022
Neg logit contributions
wood-0.052
flower-0.049
fly-0.049
tha-0.048
man-0.048
Token somewhere
Token ID7382
Feature activation-1.14e-01

Pos logit contributions
fters+0.276
ets+0.258
itionally+0.248
 tant+0.244
 scores+0.243
Neg logit contributions
man-0.224
 extingu-0.209
wood-0.203
work-0.199
fly-0.195
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.143
 calculations+0.123
 mats+0.120
 quizz+0.117
pler+0.116
Neg logit contributions
fly-0.234
man-0.234
 extingu-0.233
work-0.228
wood-0.228
Token,
Token ID11
Feature activation-5.59e-03

Pos logit contributions
fters+0.751
ets+0.680
 tant+0.650
 mats+0.649
 quizz+0.634
Neg logit contributions
wood-0.577
fly-0.560
man-0.548
work-0.543
flower-0.527
Token where
Token ID810
Feature activation-1.39e-01

Pos logit contributions
fters+0.007
ets+0.006
 quizz+0.006
 mats+0.006
 scores+0.006
Neg logit contributions
wood-0.012
fly-0.012
flower-0.011
man-0.011
 extingu-0.011
Token they
Token ID484
Feature activation+8.56e-02

Pos logit contributions
fters+0.225
 else+0.211
 mats+0.209
ets+0.209
 quizz+0.206
Neg logit contributions
wood-0.247
fly-0.232
flower-0.232
 extingu-0.232
tha-0.226
Token could
Token ID714
Feature activation-1.06e-01

Pos logit contributions
wood+0.145
fly+0.142
flower+0.139
man+0.137
tha+0.136
Neg logit contributions
fters-0.148
 mats-0.134
 quizz-0.133
ets-0.132
 calculations-0.129
Token have
Token ID423
Feature activation-3.86e-02

Pos logit contributions
fters+0.141
ets+0.125
 else+0.124
 mats+0.122
 scores+0.122
Neg logit contributions
wood-0.225
fly-0.209
flower-0.209
 extingu-0.208
tha-0.208
Token he
Token ID339
Feature activation-7.82e-02

Pos logit contributions
ets+0.055
fters+0.054
 mats+0.049
 tant+0.048
pler+0.047
Neg logit contributions
wood-0.055
fly-0.052
flower-0.051
man-0.051
work-0.049
Token didn
Token ID1422
Feature activation-7.06e-02

Pos logit contributions
fters+0.123
ets+0.112
 mats+0.111
 quizz+0.109
 tant+0.106
Neg logit contributions
wood-0.146
fly-0.140
flower-0.138
man-0.138
tha-0.136
Token't
Token ID470
Feature activation+7.18e-02

Pos logit contributions
fters+0.108
 quizz+0.094
 anymore+0.093
 scores+0.093
 else+0.093
Neg logit contributions
wood-0.140
flower-0.133
tha-0.133
fly-0.132
 extingu-0.131
Token need
Token ID761
Feature activation+2.61e-02

Pos logit contributions
wood+0.123
fly+0.120
flower+0.115
work+0.112
man+0.111
Neg logit contributions
fters-0.120
 mats-0.116
ets-0.110
 laps-0.107
 quizz-0.107
Token anyone
Token ID2687
Feature activation-2.52e-01

Pos logit contributions
wood+0.039
fly+0.036
 Mill+0.034
 Frankie+0.033
 Tit+0.033
Neg logit contributions
fters-0.052
pler-0.048
ets-0.047
itionally-0.046
 tant-0.046
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.507
 mats+0.472
 tant+0.447
 calculations+0.441
 laps+0.441
Neg logit contributions
fly-0.372
wood-0.359
flower-0.340
work-0.332
 extingu-0.317
Token's
Token ID338
Feature activation-2.31e-01

Pos logit contributions
fters+0.748
 mats+0.668
 tant+0.667
ets+0.644
 calculations+0.632
Neg logit contributions
wood-0.599
fly-0.577
work-0.545
flower-0.545
man-0.514
Token help
Token ID1037
Feature activation+8.02e-03

Pos logit contributions
fters+0.383
ets+0.354
 mats+0.344
 quizz+0.337
 tant+0.331
Neg logit contributions
wood-0.411
fly-0.393
flower-0.386
man-0.384
tha-0.373
Token.
Token ID13
Feature activation-7.20e-03

Pos logit contributions
wood+0.012
fly+0.011
man+0.011
flower+0.011
 Frankie+0.010
Neg logit contributions
fters-0.015
 mats-0.015
ets-0.014
pler-0.014
 tant-0.014
Token From
Token ID3574
Feature activation-5.87e-02

Pos logit contributions
fters+0.011
ets+0.010
 mats+0.010
 quizz+0.009
 scores+0.009
Neg logit contributions
wood-0.014
fly-0.013
man-0.013
flower-0.013
tha-0.013
Token that
Token ID326
Feature activation-3.14e-02

Pos logit contributions
fters+0.065
 else+0.063
 rides+0.056
ets+0.056
 scores+0.055
Neg logit contributions
wood-0.135
 extingu-0.130
flower-0.125
fly-0.125
man-0.123
Token Maybe
Token ID6674
Feature activation+6.64e-03

Pos logit contributions
fters+0.031
ets+0.028
 mats+0.027
 scores+0.027
 quizz+0.027
Neg logit contributions
wood-0.034
fly-0.031
 extingu-0.031
man-0.031
flower-0.030
Token you
Token ID345
Feature activation-7.80e-02

Pos logit contributions
wood+0.011
man+0.011
fly+0.011
flower+0.011
tha+0.011
Neg logit contributions
fters-0.011
ets-0.011
 tant-0.010
 mats-0.010
 quizz-0.010
Token left
Token ID1364
Feature activation-2.89e-02

Pos logit contributions
fters+0.129
 mats+0.117
ets+0.115
 quizz+0.114
 laps+0.111
Neg logit contributions
wood-0.140
fly-0.134
flower-0.130
man-0.129
work-0.129
Token it
Token ID340
Feature activation-3.98e-02

Pos logit contributions
fters+0.054
ets+0.050
itionally+0.049
 mats+0.049
 quizz+0.047
Neg logit contributions
wood-0.043
man-0.042
tha-0.040
work-0.040
fly-0.040
Token somewhere
Token ID7382
Feature activation-1.14e-01

Pos logit contributions
fters+0.068
 calculations+0.060
 quizz+0.060
 mats+0.060
itionally+0.059
Neg logit contributions
wood-0.068
man-0.067
fly-0.067
flower-0.065
work-0.065
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.154
 calculations+0.135
 mats+0.133
 quizz+0.130
itionally+0.130
Neg logit contributions
wood-0.213
work-0.209
fly-0.206
man-0.202
 extingu-0.200
Token.
Token ID13
Feature activation-2.07e-02

Pos logit contributions
fters+0.773
 mats+0.667
 tant+0.665
 quizz+0.658
 calculations+0.652
Neg logit contributions
wood-0.569
fly-0.522
work-0.518
man-0.508
flower-0.494
Token Maybe
Token ID6674
Feature activation+6.81e-03

Pos logit contributions
fters+0.032
ets+0.029
 mats+0.028
 scores+0.028
 quizz+0.028
Neg logit contributions
wood-0.040
fly-0.038
 extingu-0.037
man-0.037
flower-0.037
Token you
Token ID345
Feature activation-7.88e-02

Pos logit contributions
wood+0.012
man+0.011
fly+0.011
flower+0.011
tha+0.011
Neg logit contributions
fters-0.012
ets-0.011
 tant-0.011
 mats-0.010
 quizz-0.010
Token should
Token ID815
Feature activation+7.27e-02

Pos logit contributions
fters+0.136
 mats+0.125
ets+0.122
 quizz+0.122
 laps+0.119
Neg logit contributions
wood-0.135
fly-0.130
man-0.124
flower-0.124
work-0.124
Token search
Token ID2989
Feature activation-3.01e-02

Pos logit contributions
wood+0.148
fly+0.140
flower+0.139
 extingu+0.137
man+0.137
Neg logit contributions
fters-0.104
ets-0.095
 mats-0.090
 scores-0.090
 else-0.090
Token your
Token ID534
Feature activation-4.55e-02

Pos logit contributions
wood+0.032
fly+0.030
flower+0.030
work+0.030
man+0.030
Neg logit contributions
fters-0.036
ets-0.032
itionally-0.031
 mats-0.030
 laps-0.030
Token diary
Token ID26339
Feature activation-1.17e-01

Pos logit contributions
fters+0.078
ets+0.064
 laps+0.063
and+0.063
 mats+0.063
Neg logit contributions
flower-0.079
man-0.078
fly-0.078
wood-0.077
work-0.075
Token.
Token ID13
Feature activation-3.19e-02

Pos logit contributions
fters+0.189
ets+0.167
 mats+0.166
 quizz+0.165
 scores+0.162
Neg logit contributions
wood-0.213
man-0.204
fly-0.203
flower-0.199
 extingu-0.196
Token Maybe
Token ID6674
Feature activation+4.19e-03

Pos logit contributions
fters+0.047
ets+0.043
 mats+0.042
 quizz+0.041
 scores+0.041
Neg logit contributions
wood-0.063
fly-0.059
flower-0.058
man-0.058
tha-0.058
Token someone
Token ID2130
Feature activation-1.54e-01

Pos logit contributions
wood+0.007
fly+0.007
man+0.007
flower+0.007
tha+0.007
Neg logit contributions
fters-0.007
ets-0.007
 tant-0.006
 mats-0.006
 scores-0.006
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.217
 mats+0.189
 laps+0.180
 calculations+0.178
 tant+0.178
Neg logit contributions
fly-0.308
man-0.304
wood-0.304
work-0.296
flower-0.293
Token did
Token ID750
Feature activation-5.94e-02

Pos logit contributions
fters+0.636
 mats+0.563
ets+0.560
 quizz+0.552
 tant+0.547
Neg logit contributions
wood-0.684
fly-0.650
man-0.643
flower-0.631
work-0.627
Token.
Token ID13
Feature activation-3.41e-02

Pos logit contributions
fters+0.115
 mats+0.111
ets+0.110
 tant+0.110
 laps+0.105
Neg logit contributions
man-0.081
wood-0.079
fly-0.078
work-0.074
tha-0.073
Token Maybe
Token ID6674
Feature activation+3.65e-03

Pos logit contributions
fters+0.054
ets+0.049
 mats+0.047
 quizz+0.047
 scores+0.047
Neg logit contributions
wood-0.065
fly-0.061
man-0.060
 extingu-0.060
flower-0.059
Token it
Token ID340
Feature activation-4.17e-02

Pos logit contributions
wood+0.006
fly+0.006
man+0.006
flower+0.006
tha+0.006
Neg logit contributions
fters-0.006
ets-0.006
 tant-0.005
 mats-0.005
 scores-0.005
Token was
Token ID373
Feature activation-8.75e-03

Pos logit contributions
fters+0.070
ets+0.064
 mats+0.063
 quizz+0.063
 scores+0.061
Neg logit contributions
wood-0.074
fly-0.071
man-0.070
flower-0.069
 extingu-0.068
Token decide
Token ID5409
Feature activation-1.09e-01

Pos logit contributions
wood+0.213
fly+0.203
tha+0.201
 extingu+0.201
flower+0.200
Neg logit contributions
fters-0.218
ets-0.201
and-0.197
 mats-0.194
 quizz-0.193
Token to
Token ID284
Feature activation-2.34e-02

Pos logit contributions
fters+0.165
ets+0.152
 mats+0.147
 quizz+0.146
 scores+0.144
Neg logit contributions
wood-0.212
fly-0.200
flower-0.198
man-0.196
tha-0.196
Token play
Token ID711
Feature activation-1.43e-01

Pos logit contributions
fters+0.025
ets+0.022
 rides+0.022
 scores+0.021
 else+0.021
Neg logit contributions
wood-0.054
fly-0.051
flower-0.051
tha-0.051
work-0.050
Token with
Token ID351
Feature activation-9.05e-02

Pos logit contributions
fters+0.282
ets+0.264
itionally+0.252
anna+0.248
 tant+0.248
Neg logit contributions
man-0.209
 extingu-0.202
wood-0.193
fly-0.189
work-0.187
Token something
Token ID1223
Feature activation-1.03e-01

Pos logit contributions
fters+0.169
ets+0.160
 quizz+0.149
 mats+0.149
 scores+0.149
Neg logit contributions
wood-0.143
 extingu-0.135
tha-0.133
fly-0.132
man-0.131
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.105
 tant+0.092
 calculations+0.089
 Many+0.088
ets+0.086
Neg logit contributions
man-0.243
fly-0.228
 extingu-0.228
wood-0.227
work-0.222
Token.
Token ID13
Feature activation-2.00e-02

Pos logit contributions
fters+0.719
ets+0.637
 tant+0.634
 mats+0.623
 quizz+0.623
Neg logit contributions
wood-0.610
fly-0.580
work-0.563
man-0.561
flower-0.556
Token Something
Token ID13742
Feature activation-5.69e-02

Pos logit contributions
fters+0.041
ets+0.038
 quizz+0.036
 mats+0.036
 scores+0.036
Neg logit contributions
wood-0.029
fly-0.027
 extingu-0.026
man-0.026
flower-0.026
Token safe
Token ID3338
Feature activation-4.73e-03

Pos logit contributions
fters+0.097
ets+0.085
 tant+0.085
and+0.082
 calculations+0.081
Neg logit contributions
fly-0.096
man-0.096
wood-0.096
 extingu-0.094
work-0.092
Token and
Token ID290
Feature activation-6.49e-03

Pos logit contributions
ets+0.006
 anymore+0.006
fters+0.006
 mats+0.006
 else+0.006
Neg logit contributions
wood-0.010
 Amen-0.009
tha-0.009
 extingu-0.009
flower-0.009
Token fun
Token ID1257
Feature activation-9.24e-02

Pos logit contributions
fters+0.010
ets+0.009
 quizz+0.009
 mats+0.009
 tant+0.009
Neg logit contributions
wood-0.012
man-0.011
fly-0.011
tha-0.011
 extingu-0.011
Token.
Token ID13
Feature activation-1.56e-02

Pos logit contributions
fters+0.337
ets+0.314
 mats+0.303
 quizz+0.300
 scores+0.299
Neg logit contributions
wood-0.466
fly-0.439
flower-0.436
tha-0.435
man-0.433
Token It
Token ID632
Feature activation-6.00e-02

Pos logit contributions
fters+0.026
ets+0.024
 mats+0.023
 quizz+0.022
 scores+0.022
Neg logit contributions
wood-0.028
fly-0.026
 extingu-0.026
man-0.026
tha-0.025
Token looks
Token ID3073
Feature activation+6.38e-02

Pos logit contributions
fters+0.100
and+0.089
ets+0.084
pieces+0.081
 calculations+0.081
Neg logit contributions
fly-0.109
wood-0.106
man-0.103
 extingu-0.102
flower-0.099
Token for
Token ID329
Feature activation-2.55e-02

Pos logit contributions
wood+0.093
 extingu+0.092
fly+0.089
 Mill+0.086
man+0.086
Neg logit contributions
fters-0.125
 mats-0.107
and-0.106
 calculations-0.102
 Find-0.100
Token something
Token ID1223
Feature activation-1.03e-01

Pos logit contributions
fters+0.044
ets+0.042
 tant+0.039
 mats+0.039
 scores+0.039
Neg logit contributions
wood-0.043
fly-0.041
 extingu-0.041
tha-0.040
man-0.040
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.145
 tant+0.128
 calculations+0.122
 quizz+0.120
pieces+0.119
Neg logit contributions
wood-0.203
fly-0.202
man-0.200
 extingu-0.198
tha-0.186
Token to
Token ID284
Feature activation-2.25e-02

Pos logit contributions
fters+0.651
ets+0.581
 mats+0.570
 tant+0.570
 quizz+0.565
Neg logit contributions
wood-0.677
fly-0.643
man-0.623
flower-0.619
 extingu-0.613
Token eat
Token ID4483
Feature activation-5.34e-02

Pos logit contributions
fters+0.031
ets+0.029
 mats+0.027
 quizz+0.027
 scores+0.027
Neg logit contributions
wood-0.046
fly-0.044
 extingu-0.043
man-0.043
flower-0.043
Token.
Token ID13
Feature activation-1.62e-02

Pos logit contributions
fters+0.084
ets+0.077
 mats+0.074
 quizz+0.074
 scores+0.073
Neg logit contributions
wood-0.100
fly-0.094
flower-0.093
man-0.093
 extingu-0.092
Token
Token ID198
Feature activation-3.68e-02

Pos logit contributions
fters+0.027
ets+0.025
 mats+0.024
 quizz+0.023
 scores+0.023
Neg logit contributions
wood-0.029
fly-0.027
 extingu-0.027
man-0.027
tha-0.026
Token
Token ID198
Feature activation-3.65e-02

Pos logit contributions
fters+0.057
ets+0.053
 mats+0.051
 quizz+0.050
 scores+0.050
Neg logit contributions
wood-0.070
fly-0.066
man-0.065
 extingu-0.064
flower-0.064
Token.
Token ID13
Feature activation-1.93e-02

Pos logit contributions
fters+0.033
 mats+0.029
 tant+0.029
 calculations+0.029
itionally+0.028
Neg logit contributions
fly-0.015
wood-0.015
man-0.015
work-0.014
flower-0.014
Token He
Token ID679
Feature activation+6.83e-03

Pos logit contributions
fters+0.035
ets+0.032
 mats+0.031
 else+0.031
 quizz+0.031
Neg logit contributions
wood-0.032
 extingu-0.029
fly-0.029
man-0.029
flower-0.029
Token looks
Token ID3073
Feature activation+6.36e-02

Pos logit contributions
fly+0.012
wood+0.011
man+0.011
 extingu+0.011
flower+0.011
Neg logit contributions
fters-0.012
and-0.011
ets-0.010
pieces-0.010
itionally-0.010
Token for
Token ID329
Feature activation-2.57e-02

Pos logit contributions
man+0.068
 extingu+0.065
wood+0.065
 Frank+0.064
fly+0.064
Neg logit contributions
fters-0.144
 calculations-0.121
and-0.119
 mats-0.119
 basis-0.118
Token something
Token ID1223
Feature activation-1.02e-01

Pos logit contributions
fters+0.048
ets+0.046
 scores+0.043
 tant+0.042
 calculations+0.042
Neg logit contributions
wood-0.040
 extingu-0.039
tha-0.037
 Tit-0.037
 Mill-0.036
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.172
pieces+0.150
 tant+0.150
 calculations+0.150
 quizz+0.143
Neg logit contributions
man-0.174
wood-0.170
 extingu-0.164
fly-0.162
tha-0.151
Token to
Token ID284
Feature activation-2.38e-02

Pos logit contributions
fters+0.749
 tant+0.656
ets+0.643
 quizz+0.639
 calculations+0.638
Neg logit contributions
wood-0.606
fly-0.574
man-0.557
flower-0.539
work-0.539
Token eat
Token ID4483
Feature activation-5.42e-02

Pos logit contributions
fters+0.036
ets+0.035
 tant+0.032
 mats+0.032
 calculations+0.032
Neg logit contributions
wood-0.044
fly-0.043
man-0.042
 extingu-0.042
flower-0.042
Token.
Token ID13
Feature activation-1.97e-02

Pos logit contributions
fters+0.090
ets+0.082
 quizz+0.079
 mats+0.079
 scores+0.078
Neg logit contributions
wood-0.097
fly-0.091
man-0.090
flower-0.089
 extingu-0.089
Token He
Token ID679
Feature activation+6.21e-03

Pos logit contributions
fters+0.045
ets+0.040
pieces+0.038
brate+0.037
 scores+0.037
Neg logit contributions
wood-0.027
 extingu-0.024
man-0.023
 Tit-0.023
fly-0.022
Token hopes
Token ID7176
Feature activation+1.79e-02

Pos logit contributions
fly+0.011
wood+0.010
man+0.010
flower+0.009
 extingu+0.009
Neg logit contributions
fters-0.011
and-0.011
ets-0.009
pieces-0.009
 tant-0.009
Token But
Token ID887
Feature activation-1.86e-01

Pos logit contributions
fters+0.029
ets+0.027
 mats+0.026
 quizz+0.026
 scores+0.025
Neg logit contributions
wood-0.043
fly-0.040
flower-0.040
man-0.040
tha-0.040
Token you
Token ID345
Feature activation-7.89e-02

Pos logit contributions
fters+0.282
ets+0.257
 mats+0.254
 else+0.252
 quizz+0.250
Neg logit contributions
wood-0.358
fly-0.335
flower-0.333
man-0.331
 extingu-0.330
Token can
Token ID460
Feature activation-8.36e-02

Pos logit contributions
fters+0.109
ets+0.103
 else+0.096
 scores+0.095
 mats+0.094
Neg logit contributions
wood-0.164
flower-0.155
man-0.155
fly-0.154
 extingu-0.153
Token play
Token ID711
Feature activation-1.42e-01

Pos logit contributions
fters+0.109
 else+0.099
ets+0.094
 mats+0.094
 anymore+0.094
Neg logit contributions
wood-0.181
fly-0.170
flower-0.168
 extingu-0.168
man-0.166
Token something
Token ID1223
Feature activation-1.02e-01

Pos logit contributions
ets+0.253
fters+0.252
anna+0.241
and+0.231
 tant+0.227
Neg logit contributions
 extingu-0.197
man-0.193
work-0.186
wood-0.180
fly-0.179
Token else
Token ID2073
Feature activation-3.81e-01

Pos logit contributions
fters+0.085
 calculations+0.064
 quizz+0.062
ets+0.062
 scores+0.061
Neg logit contributions
wood-0.255
man-0.253
 extingu-0.249
fly-0.248
work-0.245
Token inside
Token ID2641
Feature activation+1.87e-02

Pos logit contributions
fters+0.621
ets+0.572
 mats+0.552
 quizz+0.550
 scores+0.544
Neg logit contributions
wood-0.690
fly-0.647
flower-0.636
man-0.634
 extingu-0.632
Token.
Token ID13
Feature activation-2.11e-02

Pos logit contributions
wood+0.034
fly+0.032
flower+0.032
man+0.031
tha+0.031
Neg logit contributions
fters-0.030
ets-0.028
 mats-0.027
 quizz-0.027
 scores-0.026
Token How
Token ID1374
Feature activation-7.57e-02

Pos logit contributions
fters+0.031
ets+0.028
 mats+0.027
 scores+0.027
 quizz+0.027
Neg logit contributions
wood-0.042
fly-0.039
 extingu-0.039
flower-0.039
man-0.038
Token about
Token ID546
Feature activation-1.93e-01

Pos logit contributions
fters+0.110
 mats+0.092
pler+0.091
 tant+0.090
ets+0.090
Neg logit contributions
wood-0.143
man-0.139
fly-0.138
work-0.136
tha-0.131
Token you
Token ID345
Feature activation-7.84e-02

Pos logit contributions
fters+0.345
ets+0.330
 tant+0.315
 mats+0.305
 quizz+0.304
Neg logit contributions
wood-0.322
fly-0.299
 extingu-0.296
work-0.294
flower-0.292
Token "
Token ID366
Feature activation-7.83e-02

Pos logit contributions
fters+0.019
ets+0.018
 scores+0.018
 tant+0.017
 loses+0.017
Neg logit contributions
wood-0.018
man-0.018
flower-0.018
 extingu-0.018
fly-0.017
TokenCan
Token ID6090
Feature activation-3.76e-02

Pos logit contributions
fters+0.137
ets+0.133
 quizz+0.127
 scores+0.126
 laps+0.124
Neg logit contributions
wood-0.129
flower-0.129
man-0.126
fly-0.123
tha-0.122
Token I
Token ID314
Feature activation-1.01e-02

Pos logit contributions
ets+0.078
fters+0.075
 tant+0.072
and+0.070
 quizz+0.070
Neg logit contributions
tha-0.049
man-0.049
fly-0.048
wood-0.048
flower-0.047
Token have
Token ID423
Feature activation-3.81e-02

Pos logit contributions
fters+0.014
ets+0.013
 mats+0.012
 quizz+0.012
 else+0.012
Neg logit contributions
wood-0.021
fly-0.020
tha-0.019
flower-0.019
man-0.019
Token something
Token ID1223
Feature activation-1.01e-01

Pos logit contributions
fters+0.065
ets+0.062
 mats+0.059
and+0.058
 scores+0.057
Neg logit contributions
wood-0.065
 extingu-0.060
fly-0.059
flower-0.058
work-0.058
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.153
ets+0.138
 mats+0.134
 quizz+0.131
 calculations+0.130
Neg logit contributions
wood-0.195
fly-0.185
man-0.184
 extingu-0.183
flower-0.181
Token?"
Token ID1701
Feature activation+1.20e-01

Pos logit contributions
fters+0.727
ets+0.674
 mats+0.657
 quizz+0.645
 scores+0.641
Neg logit contributions
wood-0.593
fly-0.551
flower-0.542
man-0.539
 extingu-0.531
Token
Token ID198
Feature activation-4.39e-02

Pos logit contributions
wood+0.168
man+0.167
flower+0.163
 Mill+0.160
tha+0.160
Neg logit contributions
fters-0.243
ets-0.233
 laps-0.213
 scores-0.212
 mats-0.209
Token
Token ID198
Feature activation-4.34e-02

Pos logit contributions
fters+0.074
ets+0.071
 mats+0.067
 scores+0.066
 quizz+0.065
Neg logit contributions
wood-0.074
man-0.073
tha-0.072
fly-0.071
 extingu-0.070
TokenHer
Token ID9360
Feature activation-9.68e-02

Pos logit contributions
fters+0.061
ets+0.059
 mats+0.054
 quizz+0.054
 scores+0.054
Neg logit contributions
wood-0.086
man-0.083
fly-0.083
tha-0.082
flower-0.082
Token mom
Token ID1995
Feature activation+7.82e-02

Pos logit contributions
fters+0.183
ets+0.173
 mats+0.171
 quizz+0.169
 else+0.168
Neg logit contributions
wood-0.151
fly-0.138
flower-0.135
tha-0.134
 extingu-0.134
Token ask
Token ID1265
Feature activation-6.12e-02

Pos logit contributions
wood+0.078
fly+0.074
flower+0.072
man+0.072
 extingu+0.072
Neg logit contributions
fters-0.073
ets-0.068
 mats-0.066
 quizz-0.065
 scores-0.064
Token before
Token ID878
Feature activation-1.84e-01

Pos logit contributions
fters+0.106
ets+0.094
 scored+0.094
 tant+0.094
pler+0.092
Neg logit contributions
wood-0.100
tha-0.093
work-0.092
man-0.092
fly-0.091
Token you
Token ID345
Feature activation-7.82e-02

Pos logit contributions
fters+0.250
ets+0.231
 mats+0.219
 quizz+0.217
 scores+0.215
Neg logit contributions
wood-0.382
fly-0.361
flower-0.357
man-0.356
tha-0.356
Token use
Token ID779
Feature activation+1.39e-03

Pos logit contributions
fters+0.129
 mats+0.115
ets+0.114
 quizz+0.111
 laps+0.111
Neg logit contributions
wood-0.142
fly-0.135
flower-0.131
work-0.130
 extingu-0.129
Token someone
Token ID2130
Feature activation-1.51e-01

Pos logit contributions
wood+0.002
work+0.002
fly+0.002
 extingu+0.002
man+0.002
Neg logit contributions
fters-0.003
ets-0.002
 mats-0.002
itionally-0.002
 laps-0.002
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.232
 tant+0.183
 mats+0.181
 laps+0.180
 sidelines+0.180
Neg logit contributions
wood-0.289
work-0.278
fly-0.276
 extingu-0.267
tha-0.258
Token's
Token ID338
Feature activation-2.39e-01

Pos logit contributions
fters+0.684
 mats+0.570
 tant+0.563
 laps+0.562
ets+0.555
Neg logit contributions
wood-0.666
work-0.605
fly-0.602
 extingu-0.580
man-0.574
Token things
Token ID1243
Feature activation-1.63e-01

Pos logit contributions
fters+0.363
ets+0.318
 mats+0.316
 quizz+0.306
 scores+0.305
Neg logit contributions
wood-0.467
fly-0.442
man-0.440
flower-0.434
work-0.434
Token.
Token ID13
Feature activation-2.82e-02

Pos logit contributions
fters+0.276
 mats+0.246
 quizz+0.243
ets+0.243
 tant+0.241
Neg logit contributions
wood-0.287
fly-0.267
man-0.263
work-0.262
 extingu-0.261
Token You
Token ID921
Feature activation-3.04e-02

Pos logit contributions
fters+0.041
ets+0.038
 mats+0.037
 quizz+0.036
 scores+0.036
Neg logit contributions
wood-0.056
fly-0.053
flower-0.052
 extingu-0.052
man-0.052
Token should
Token ID815
Feature activation+7.11e-02

Pos logit contributions
fters+0.047
ets+0.043
 mats+0.042
 quizz+0.042
 scores+0.041
Neg logit contributions
wood-0.058
fly-0.054
flower-0.053
man-0.053
 extingu-0.053
Token She
Token ID1375
Feature activation+7.30e-02

Pos logit contributions
fters+0.009
ets+0.008
 scores+0.008
 laps+0.008
 tant+0.007
Neg logit contributions
wood-0.006
man-0.005
fly-0.005
flower-0.005
tha-0.005
Token wants
Token ID3382
Feature activation-1.13e-01

Pos logit contributions
wood+0.115
fly+0.110
flower+0.107
 Amen+0.106
work+0.106
Neg logit contributions
fters-0.128
and-0.121
 anymore-0.107
 tant-0.106
ets-0.105
Token to
Token ID284
Feature activation-2.57e-02

Pos logit contributions
fters+0.195
ets+0.179
 mats+0.168
 scores+0.165
 laps+0.163
Neg logit contributions
wood-0.202
man-0.187
fly-0.186
flower-0.184
work-0.184
Token paint
Token ID7521
Feature activation+8.45e-03

Pos logit contributions
fters+0.036
ets+0.032
 quizz+0.031
 else+0.031
 scores+0.031
Neg logit contributions
wood-0.054
fly-0.051
man-0.050
 extingu-0.050
tha-0.050
Token something
Token ID1223
Feature activation-1.02e-01

Pos logit contributions
wood+0.015
fly+0.014
flower+0.014
tha+0.014
man+0.014
Neg logit contributions
fters-0.014
ets-0.013
 mats-0.012
 quizz-0.012
 scores-0.012
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.143
ets+0.126
 mats+0.121
 calculations+0.118
 scores+0.117
Neg logit contributions
wood-0.203
man-0.196
work-0.195
flower-0.194
fly-0.194
Token.
Token ID13
Feature activation-4.02e-03

Pos logit contributions
fters+0.562
ets+0.522
 mats+0.504
 quizz+0.500
 scores+0.496
Neg logit contributions
wood-0.741
fly-0.698
flower-0.691
tha-0.688
 extingu-0.687
Token
Token ID198
Feature activation-2.49e-02

Pos logit contributions
fters+0.008
ets+0.007
 scores+0.006
 mats+0.006
 laps+0.006
Neg logit contributions
wood-0.007
man-0.006
fly-0.006
flower-0.006
tha-0.006
Token
Token ID198
Feature activation-2.42e-02

Pos logit contributions
fters+0.042
ets+0.040
 mats+0.038
 scores+0.038
 quizz+0.037
Neg logit contributions
wood-0.042
man-0.040
fly-0.040
tha-0.040
 extingu-0.039
TokenShe
Token ID3347
Feature activation+4.91e-02

Pos logit contributions
fters+0.040
ets+0.039
 scores+0.038
 anymore+0.037
 quizz+0.037
Neg logit contributions
wood-0.040
man-0.040
tha-0.038
flower-0.038
fly-0.037
Token goes
Token ID2925
Feature activation+6.53e-02

Pos logit contributions
wood+0.080
flower+0.077
 extingu+0.076
work+0.076
fly+0.076
Neg logit contributions
fters-0.081
and-0.076
itionally-0.070
 anymore-0.069
 tant-0.069
Token hears
Token ID23180
Feature activation-1.22e-01

Pos logit contributions
fters+0.543
ets+0.510
 scores+0.492
 else+0.490
 quizz+0.489
Neg logit contributions
wood-0.760
flower-0.713
fly-0.710
 extingu-0.705
tha-0.701
Token them
Token ID606
Feature activation-1.13e-01

Pos logit contributions
fters+0.174
ets+0.162
 mats+0.156
 quizz+0.156
 scores+0.155
Neg logit contributions
wood-0.245
fly-0.232
flower-0.231
 extingu-0.226
tha-0.225
Token.
Token ID13
Feature activation-5.17e-03

Pos logit contributions
fters+0.144
 anymore+0.136
ets+0.136
 else+0.132
 scores+0.128
Neg logit contributions
wood-0.241
tha-0.231
flower-0.225
fly-0.225
 extingu-0.223
Token No
Token ID1400
Feature activation-1.48e-02

Pos logit contributions
fters+0.008
ets+0.007
 mats+0.007
 quizz+0.007
 scores+0.007
Neg logit contributions
wood-0.010
fly-0.009
 extingu-0.009
man-0.009
tha-0.009
Token one
Token ID530
Feature activation-1.84e-02

Pos logit contributions
fters+0.025
ets+0.024
 mats+0.023
 quizz+0.022
and+0.022
Neg logit contributions
wood-0.025
man-0.024
fly-0.024
 extingu-0.024
flower-0.023
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.027
ets+0.024
 mats+0.023
 quizz+0.022
 calculations+0.022
Neg logit contributions
wood-0.036
fly-0.036
man-0.035
flower-0.034
 extingu-0.034
Token comes
Token ID2058
Feature activation+6.68e-03

Pos logit contributions
fters+0.600
ets+0.547
 mats+0.535
 quizz+0.525
 scores+0.515
Neg logit contributions
wood-0.717
fly-0.680
man-0.672
tha-0.665
flower-0.663
Token.
Token ID13
Feature activation-7.52e-03

Pos logit contributions
wood+0.014
work+0.014
tha+0.014
 Amen+0.014
flower+0.014
Neg logit contributions
 anymore-0.008
fters-0.007
 else-0.007
and-0.007
ets-0.007
Token The
Token ID383
Feature activation-7.81e-02

Pos logit contributions
fters+0.011
ets+0.010
 mats+0.010
 quizz+0.010
 scores+0.010
Neg logit contributions
wood-0.015
fly-0.014
 extingu-0.014
man-0.014
flower-0.014
Token dog
Token ID3290
Feature activation+1.27e-02

Pos logit contributions
fters+0.148
ets+0.140
 quizz+0.132
and+0.132
 scores+0.130
Neg logit contributions
wood-0.121
man-0.117
 extingu-0.114
flower-0.114
fly-0.113
Token is
Token ID318
Feature activation+1.37e-03

Pos logit contributions
fly+0.019
man+0.018
flower+0.017
 extingu+0.016
wood+0.016
Neg logit contributions
fters-0.025
pieces-0.023
and-0.023
 calculations-0.022
itionally-0.022
Token wanted
Token ID2227
Feature activation-2.43e-02

Pos logit contributions
wood+0.105
fly+0.104
man+0.101
tha+0.100
flower+0.100
Neg logit contributions
fters-0.128
and-0.114
 quizz-0.113
ets-0.112
 mats-0.111
Token to
Token ID284
Feature activation-2.25e-02

Pos logit contributions
fters+0.042
ets+0.040
 mats+0.036
 scores+0.036
 tant+0.036
Neg logit contributions
wood-0.043
fly-0.040
man-0.039
 extingu-0.039
flower-0.039
Token shoot
Token ID2686
Feature activation-7.41e-02

Pos logit contributions
fters+0.028
 else+0.025
 scores+0.025
ets+0.024
 quizz+0.024
Neg logit contributions
wood-0.050
tha-0.047
flower-0.046
 extingu-0.046
fly-0.046
Token at
Token ID379
Feature activation-1.73e-01

Pos logit contributions
fters+0.133
ets+0.120
 mats+0.116
 calculations+0.114
 quizz+0.112
Neg logit contributions
wood-0.124
fly-0.118
flower-0.116
 extingu-0.114
man-0.114
Token something
Token ID1223
Feature activation-1.01e-01

Pos logit contributions
fters+0.250
 quizz+0.229
 mats+0.229
ets+0.227
 else+0.225
Neg logit contributions
wood-0.338
 extingu-0.316
fly-0.316
man-0.313
flower-0.313
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.114
ets+0.097
 calculations+0.094
 quizz+0.092
 mats+0.091
Neg logit contributions
fly-0.232
man-0.228
flower-0.219
wood-0.218
work-0.217
Token.
Token ID13
Feature activation-1.06e-02

Pos logit contributions
fters+0.633
ets+0.577
 mats+0.559
 quizz+0.553
 scores+0.544
Neg logit contributions
wood-0.681
fly-0.652
flower-0.638
man-0.634
work-0.625
Token She
Token ID1375
Feature activation+6.80e-02

Pos logit contributions
fters+0.023
ets+0.021
 laps+0.020
 tant+0.020
 scores+0.020
Neg logit contributions
wood-0.015
flower-0.013
 Mill-0.013
 Tit-0.012
man-0.012
Token saw
Token ID2497
Feature activation-1.49e-01

Pos logit contributions
wood+0.112
fly+0.110
man+0.108
flower+0.106
tha+0.105
Neg logit contributions
fters-0.122
ets-0.107
 quizz-0.107
and-0.106
 anymore-0.105
Token a
Token ID257
Feature activation+1.13e-01

Pos logit contributions
fters+0.282
ets+0.265
 mats+0.260
 quizz+0.258
 scores+0.255
Neg logit contributions
wood-0.232
fly-0.215
flower-0.212
tha-0.210
man-0.210
Token butterfly
Token ID35113
Feature activation+8.69e-03

Pos logit contributions
wood+0.223
tha+0.203
 extingu+0.201
work+0.198
flower+0.197
Neg logit contributions
fters-0.169
 else-0.154
 scores-0.152
 mats-0.151
 quizz-0.151
Token have
Token ID423
Feature activation-3.53e-02

Pos logit contributions
fters+0.154
ets+0.140
 mats+0.136
 scores+0.136
 quizz+0.135
Neg logit contributions
wood-0.214
flower-0.200
fly-0.199
tha-0.198
man-0.197
Token killed
Token ID2923
Feature activation-3.58e-02

Pos logit contributions
fters+0.060
ets+0.056
 mats+0.054
 quizz+0.053
 scores+0.053
Neg logit contributions
wood-0.062
fly-0.058
flower-0.057
man-0.057
 extingu-0.057
Token yourself
Token ID3511
Feature activation-6.65e-03

Pos logit contributions
fters+0.051
ets+0.047
 mats+0.046
 quizz+0.046
 anymore+0.045
Neg logit contributions
wood-0.071
fly-0.067
flower-0.067
man-0.067
 extingu-0.066
Token or
Token ID393
Feature activation-1.33e-01

Pos logit contributions
fters+0.010
ets+0.009
 mats+0.009
 quizz+0.009
 scores+0.009
Neg logit contributions
wood-0.013
fly-0.012
flower-0.012
tha-0.012
man-0.012
Token someone
Token ID2130
Feature activation-1.52e-01

Pos logit contributions
fters+0.219
ets+0.201
 mats+0.188
 quizz+0.184
 calculations+0.183
Neg logit contributions
wood-0.245
fly-0.233
 extingu-0.227
work-0.225
man-0.224
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.190
 mats+0.162
 calculations+0.156
 quizz+0.152
ets+0.151
Neg logit contributions
fly-0.320
wood-0.316
man-0.315
flower-0.312
work-0.309
Token.
Token ID13
Feature activation-2.34e-02

Pos logit contributions
fters+0.604
ets+0.555
 mats+0.540
 quizz+0.535
 scores+0.527
Neg logit contributions
wood-0.705
fly-0.663
flower-0.654
man-0.652
 extingu-0.647
Token You
Token ID921
Feature activation-2.96e-02

Pos logit contributions
fters+0.034
ets+0.031
 mats+0.030
 quizz+0.030
 scores+0.030
Neg logit contributions
wood-0.047
fly-0.044
flower-0.043
man-0.043
 extingu-0.043
Token caused
Token ID4073
Feature activation-1.92e-02

Pos logit contributions
fters+0.045
ets+0.042
 else+0.041
 scores+0.040
 mats+0.040
Neg logit contributions
wood-0.057
tha-0.053
flower-0.053
fly-0.053
 extingu-0.053
Token a
Token ID257
Feature activation+1.08e-01

Pos logit contributions
fters+0.031
ets+0.029
 quizz+0.028
 mats+0.028
 scores+0.028
Neg logit contributions
wood-0.035
flower-0.033
fly-0.033
man-0.033
tha-0.033
Token lot
Token ID1256
Feature activation-8.72e-02

Pos logit contributions
wood+0.237
fly+0.223
flower+0.223
tha+0.221
 extingu+0.219
Neg logit contributions
fters-0.135
ets-0.124
 mats-0.118
 quizz-0.118
 scores-0.117
Token now
Token ID783
Feature activation+1.70e-01

Pos logit contributions
wood+0.108
tha+0.105
fly+0.103
flower+0.103
man+0.103
Neg logit contributions
fters-0.079
 else-0.076
 anymore-0.074
ets-0.074
 quizz-0.074
Token.
Token ID13
Feature activation-9.55e-03

Pos logit contributions
wood+0.290
fly+0.271
flower+0.266
tha+0.265
man+0.265
Neg logit contributions
fters-0.294
ets-0.273
 mats-0.264
 quizz-0.262
 scores-0.259
Token Go
Token ID1514
Feature activation-1.19e-01

Pos logit contributions
fters+0.014
ets+0.013
 mats+0.013
 quizz+0.013
 scores+0.013
Neg logit contributions
wood-0.018
fly-0.017
flower-0.017
man-0.017
tha-0.017
Token play
Token ID711
Feature activation-1.44e-01

Pos logit contributions
fters+0.155
ets+0.146
 mats+0.134
 scores+0.134
 else+0.134
Neg logit contributions
wood-0.244
fly-0.235
tha-0.233
 extingu-0.233
flower-0.232
Token somewhere
Token ID7382
Feature activation-1.14e-01

Pos logit contributions
fters+0.227
ets+0.219
 calculations+0.203
 mats+0.201
 scores+0.198
Neg logit contributions
man-0.250
wood-0.247
 extingu-0.242
fly-0.238
work-0.231
Token else
Token ID2073
Feature activation-3.80e-01

Pos logit contributions
fters+0.094
 calculations+0.081
 quizz+0.078
 mats+0.077
 scores+0.075
Neg logit contributions
wood-0.281
 extingu-0.271
fly-0.271
man-0.270
tha-0.268
Token,
Token ID11
Feature activation+1.87e-03

Pos logit contributions
fters+0.774
ets+0.728
 mats+0.707
 quizz+0.705
 scores+0.696
Neg logit contributions
wood-0.536
fly-0.491
man-0.481
flower-0.479
 extingu-0.474
Token please
Token ID3387
Feature activation+5.06e-02

Pos logit contributions
wood+0.003
fly+0.003
man+0.003
tha+0.003
flower+0.003
Neg logit contributions
fters-0.003
ets-0.003
 quizz-0.003
 mats-0.003
 scores-0.003
Token,"
Token ID553
Feature activation+1.17e-01

Pos logit contributions
wood+0.083
flower+0.079
fly+0.077
work+0.077
tha+0.075
Neg logit contributions
fters-0.090
ets-0.085
 else-0.084
 anymore-0.081
 mats-0.081
Token dad
Token ID9955
Feature activation+1.47e-01

Pos logit contributions
wood+0.203
fly+0.189
flower+0.185
 extingu+0.182
tha+0.181
Neg logit contributions
fters-0.200
 else-0.187
 mats-0.187
 quizz-0.186
 tant-0.184
Token said
Token ID531
Feature activation+9.59e-02

Pos logit contributions
wood+0.235
fly+0.218
flower+0.216
tha+0.211
 extingu+0.208
Neg logit contributions
fters-0.271
ets-0.256
 else-0.252
 mats-0.247
 quizz-0.247
Token cater
Token ID20825
Feature activation+7.40e-02

Pos logit contributions
wood+0.277
tha+0.252
 extingu+0.250
work+0.248
fly+0.247
Neg logit contributions
fters-0.162
 else-0.153
 mats-0.147
 quizz-0.146
 scores-0.142
Tokenpillar
Token ID41643
Feature activation-1.72e-01

Pos logit contributions
fly+0.161
flower+0.160
wood+0.153
man+0.149
flake+0.149
Neg logit contributions
fters-0.087
 mats-0.080
 calculations-0.079
 laps-0.077
 scores-0.076
Token and
Token ID290
Feature activation+2.00e-02

Pos logit contributions
fters+0.308
 tant+0.283
ets+0.277
 mats+0.276
 quizz+0.273
Neg logit contributions
fly-0.290
flower-0.284
wood-0.275
man-0.270
work-0.259
Token the
Token ID262
Feature activation+1.45e-02

Pos logit contributions
wood+0.038
 extingu+0.035
tha+0.035
flower+0.035
fly+0.035
Neg logit contributions
fters-0.031
ets-0.029
 else-0.029
 mats-0.028
 quizz-0.028
Token sun
Token ID4252
Feature activation+5.04e-02

Pos logit contributions
wood+0.027
fly+0.025
 extingu+0.025
tha+0.025
man+0.025
Neg logit contributions
fters-0.023
ets-0.021
 mats-0.021
 quizz-0.020
 scores-0.020
Tokenflower
Token ID25547
Feature activation+5.00e-01

Pos logit contributions
wood+0.035
 extingu+0.032
 Amen+0.027
tha+0.026
 Frankie+0.024
Neg logit contributions
fters-0.128
 anymore-0.126
ets-0.125
 else-0.125
pler-0.124
Token smiled
Token ID13541
Feature activation+1.98e-02

Pos logit contributions
wood+0.824
flower+0.817
fly+0.817
man+0.794
work+0.753
Neg logit contributions
fters-0.892
 mats-0.810
ets-0.810
 quizz-0.806
 tant-0.788
Token.
Token ID13
Feature activation+9.08e-03

Pos logit contributions
tha+0.048
wood+0.047
 extingu+0.045
fly+0.044
man+0.044
Neg logit contributions
 anymore-0.019
ets-0.018
 corners-0.016
 goals-0.016
 else-0.016
Token
Token ID198
Feature activation-9.30e-03

Pos logit contributions
wood+0.018
fly+0.017
man+0.017
flower+0.016
 extingu+0.016
Neg logit contributions
fters-0.014
ets-0.012
 mats-0.012
 quizz-0.012
 scores-0.012
Token
Token ID198
Feature activation-9.15e-03

Pos logit contributions
fters+0.015
ets+0.014
 mats+0.014
 quizz+0.014
 scores+0.014
Neg logit contributions
wood-0.016
man-0.016
fly-0.015
tha-0.015
flower-0.015
TokenThe
Token ID464
Feature activation-2.68e-02

Pos logit contributions
fters+0.015
ets+0.014
 tant+0.014
 quizz+0.014
 anymore+0.014
Neg logit contributions
wood-0.015
man-0.015
fly-0.015
flower-0.014
tha-0.014
Token her
Token ID607
Feature activation+2.13e-02

Pos logit contributions
fters+0.071
ets+0.066
 mats+0.064
 quizz+0.062
 scores+0.061
Neg logit contributions
wood-0.071
tha-0.065
flower-0.065
fly-0.064
 extingu-0.064
Token penny
Token ID26369
Feature activation+8.56e-02

Pos logit contributions
wood+0.033
flower+0.031
tha+0.031
 extingu+0.030
fly+0.030
Neg logit contributions
fters-0.041
ets-0.040
 quizz-0.037
 mats-0.037
 calculations-0.036
Token in
Token ID287
Feature activation-3.06e-02

Pos logit contributions
wood+0.152
flower+0.148
man+0.144
fly+0.142
tha+0.139
Neg logit contributions
fters-0.140
 mats-0.131
 quizz-0.131
ets-0.131
 calculations-0.128
Token the
Token ID262
Feature activation+1.38e-02

Pos logit contributions
fters+0.047
 quizz+0.043
 mats+0.043
ets+0.043
 scores+0.042
Neg logit contributions
wood-0.058
fly-0.054
man-0.053
flower-0.053
tha-0.053
Token pig
Token ID12967
Feature activation+1.63e-02

Pos logit contributions
wood+0.026
man+0.024
 extingu+0.023
flower+0.023
fly+0.023
Neg logit contributions
fters-0.023
ets-0.021
and-0.020
 quizz-0.020
 mats-0.020
Tokengy
Token ID1360
Feature activation+3.58e-01

Pos logit contributions
wood+0.017
tha+0.016
 extingu+0.016
fly+0.015
flower+0.015
Neg logit contributions
fters-0.037
ets-0.036
 else-0.034
 scores-0.034
 quizz-0.034
Token bank
Token ID3331
Feature activation-1.93e-01

Pos logit contributions
wood+0.845
man+0.797
flower+0.796
fly+0.792
 extingu+0.777
Neg logit contributions
fters-0.415
ets-0.365
 quizz-0.345
 mats-0.326
 calculations-0.326
Token ever
Token ID1683
Feature activation+2.53e-01

Pos logit contributions
fters+0.320
 mats+0.291
 quizz+0.290
ets+0.285
 scores+0.284
Neg logit contributions
wood-0.342
man-0.319
flower-0.313
work-0.306
fly-0.305
Token again
Token ID757
Feature activation-2.19e-01

Pos logit contributions
wood+0.503
flower+0.487
fly+0.473
work+0.471
man+0.437
Neg logit contributions
fters-0.368
 mats-0.351
 laps-0.326
pler-0.326
 Many-0.322
Token.
Token ID13
Feature activation+9.67e-03

Pos logit contributions
fters+0.359
ets+0.332
 mats+0.325
 quizz+0.318
 scores+0.314
Neg logit contributions
wood-0.395
fly-0.370
flower-0.368
man-0.362
tha-0.361
Token She
Token ID1375
Feature activation+8.77e-02

Pos logit contributions
wood+0.016
 Mill+0.014
man+0.014
flower+0.014
fly+0.014
Neg logit contributions
fters-0.018
ets-0.017
 mats-0.016
 laps-0.016
 scores-0.016
Token to
Token ID284
Feature activation-1.73e-02

Pos logit contributions
wood+0.005
fly+0.005
flower+0.005
 extingu+0.005
tha+0.005
Neg logit contributions
fters-0.004
ets-0.004
 mats-0.004
 quizz-0.004
 scores-0.003
Token be
Token ID307
Feature activation+7.33e-02

Pos logit contributions
fters+0.027
ets+0.025
 mats+0.025
 quizz+0.024
 scores+0.024
Neg logit contributions
wood-0.032
fly-0.031
flower-0.030
man-0.030
 extingu-0.030
Token rude
Token ID22066
Feature activation-9.63e-02

Pos logit contributions
wood+0.132
fly+0.128
flower+0.123
work+0.122
man+0.121
Neg logit contributions
fters-0.123
 mats-0.110
ets-0.104
 calculations-0.103
boards-0.101
Token and
Token ID290
Feature activation+6.24e-04

Pos logit contributions
fters+0.182
 quizz+0.153
ets+0.152
 mats+0.152
 calculations+0.149
Neg logit contributions
wood-0.165
fly-0.148
flower-0.147
man-0.147
work-0.145
Token hurt
Token ID5938
Feature activation+1.91e-02

Pos logit contributions
wood+0.001
fly+0.001
work+0.001
flower+0.001
man+0.001
Neg logit contributions
fters-0.001
ets-0.001
 mats-0.001
 quizz-0.001
 calculations-0.001
Tokenful
Token ID913
Feature activation+3.66e-01

Pos logit contributions
wood+0.037
fly+0.035
flower+0.034
man+0.034
tha+0.033
Neg logit contributions
fters-0.030
ets-0.027
 mats-0.026
 quizz-0.026
 scores-0.025
Token to
Token ID284
Feature activation-1.85e-02

Pos logit contributions
wood+0.600
man+0.557
fly+0.549
flower+0.540
work+0.527
Neg logit contributions
fters-0.678
ets-0.586
 quizz-0.582
 mats-0.578
 scores-0.578
Token your
Token ID534
Feature activation-4.05e-02

Pos logit contributions
fters+0.029
ets+0.027
 mats+0.026
 calculations+0.025
boards+0.025
Neg logit contributions
wood-0.034
fly-0.033
tha-0.032
man-0.032
flower-0.032
Token friend
Token ID1545
Feature activation+3.14e-02

Pos logit contributions
fters+0.069
ets+0.062
 mats+0.057
 quizz+0.057
 else+0.056
Neg logit contributions
wood-0.074
flower-0.071
man-0.069
fly-0.069
tha-0.069
Token.
Token ID13
Feature activation-1.86e-02

Pos logit contributions
wood+0.050
flower+0.049
fly+0.048
man+0.047
work+0.046
Neg logit contributions
fters-0.058
ets-0.051
 mats-0.050
 tant-0.050
 laps-0.050
Token You
Token ID921
Feature activation-2.43e-02

Pos logit contributions
fters+0.025
ets+0.024
 mats+0.023
 quizz+0.023
 scores+0.022
Neg logit contributions
wood-0.038
fly-0.036
flower-0.036
tha-0.036
man-0.035
Token,
Token ID11
Feature activation+1.51e-02

Pos logit contributions
fters+0.005
ets+0.005
 mats+0.005
 quizz+0.005
 scores+0.005
Neg logit contributions
wood-0.007
fly-0.007
flower-0.007
tha-0.007
 extingu-0.007
Token grabbing
Token ID23256
Feature activation+1.91e-02

Pos logit contributions
wood+0.030
fly+0.029
flower+0.029
tha+0.028
man+0.028
Neg logit contributions
fters-0.022
ets-0.020
 mats-0.019
 quizz-0.019
 scores-0.019
Token the
Token ID262
Feature activation+8.78e-03

Pos logit contributions
wood+0.035
fly+0.034
flower+0.034
man+0.033
tha+0.033
Neg logit contributions
fters-0.030
ets-0.028
 mats-0.028
 quizz-0.027
 scores-0.027
Token leaves
Token ID5667
Feature activation-8.32e-02

Pos logit contributions
wood+0.015
fly+0.014
tha+0.014
man+0.014
 extingu+0.014
Neg logit contributions
fters-0.015
ets-0.014
 mats-0.014
 scores-0.013
 quizz-0.013
Token for
Token ID329
Feature activation-1.20e-02

Pos logit contributions
fters+0.121
ets+0.116
 anymore+0.114
 else+0.112
 mats+0.111
Neg logit contributions
wood-0.156
 Amen-0.155
tha-0.153
man-0.147
fly-0.147
Token Joey
Token ID26154
Feature activation+3.69e-01

Pos logit contributions
fters+0.019
 mats+0.018
 quizz+0.018
ets+0.018
 scores+0.017
Neg logit contributions
wood-0.022
flower-0.021
fly-0.021
man-0.021
work-0.020
Token to
Token ID284
Feature activation-5.65e-03

Pos logit contributions
wood+0.755
fly+0.716
tha+0.715
flower+0.710
 extingu+0.710
Neg logit contributions
fters-0.495
ets-0.465
 else-0.444
 mats-0.441
 quizz-0.438
Token eat
Token ID4483
Feature activation-4.23e-02

Pos logit contributions
fters+0.008
ets+0.007
 mats+0.007
 quizz+0.007
 scores+0.007
Neg logit contributions
wood-0.012
fly-0.011
flower-0.011
man-0.011
 extingu-0.011
Token.
Token ID13
Feature activation+1.57e-02

Pos logit contributions
fters+0.064
ets+0.059
 mats+0.057
 quizz+0.057
 scores+0.056
Neg logit contributions
wood-0.081
fly-0.076
flower-0.076
man-0.075
 extingu-0.075
Token
Token ID198
Feature activation-7.31e-03

Pos logit contributions
wood+0.030
fly+0.029
flower+0.029
man+0.028
 extingu+0.028
Neg logit contributions
fters-0.023
ets-0.022
 mats-0.021
 quizz-0.021
 scores-0.020
Token
Token ID198
Feature activation-6.56e-03

Pos logit contributions
fters+0.010
ets+0.009
 mats+0.009
 quizz+0.009
 scores+0.009
Neg logit contributions
wood-0.015
fly-0.014
flower-0.014
 extingu-0.014
tha-0.014
Token.
Token ID13
Feature activation-2.86e-02

Pos logit contributions
wood+0.077
tha+0.074
fly+0.073
 Amen+0.072
flower+0.072
Neg logit contributions
ets-0.043
 anymore-0.042
fters-0.041
 mats-0.040
 corners-0.039
Token They
Token ID1119
Feature activation+1.27e-01

Pos logit contributions
fters+0.055
ets+0.052
 mats+0.051
 quizz+0.051
 scores+0.050
Neg logit contributions
wood-0.043
fly-0.040
flower-0.040
tha-0.039
man-0.039
Token see
Token ID766
Feature activation-5.98e-02

Pos logit contributions
wood+0.246
fly+0.234
flower+0.230
 extingu+0.229
tha+0.229
Neg logit contributions
fters-0.188
ets-0.171
 mats-0.167
 quizz-0.165
 scores-0.162
Token their
Token ID511
Feature activation+1.30e-01

Pos logit contributions
 mats+0.096
 else+0.094
fters+0.094
 quizz+0.092
 corners+0.091
Neg logit contributions
wood-0.105
tha-0.099
flower-0.098
fly-0.097
man-0.095
Token snow
Token ID6729
Feature activation+1.32e-01

Pos logit contributions
wood+0.221
fly+0.206
flower+0.203
tha+0.202
man+0.201
Neg logit contributions
fters-0.225
ets-0.210
 mats-0.207
 quizz-0.204
 scores-0.203
Tokenman
Token ID805
Feature activation+4.81e-01

Pos logit contributions
 Mill+0.100
wood+0.094
 Frankie+0.091
 Amen+0.090
tha+0.087
Neg logit contributions
 anymore-0.323
 mats-0.316
 towers-0.315
 corners-0.314
ets-0.311
Token smiling
Token ID16755
Feature activation+1.02e-01

Pos logit contributions
wood+0.939
fly+0.890
flower+0.880
man+0.875
 extingu+0.871
Neg logit contributions
fters-0.712
ets-0.657
 mats-0.635
 quizz-0.628
 scores-0.619
Token at
Token ID379
Feature activation-1.66e-01

Pos logit contributions
wood+0.234
tha+0.228
fly+0.220
 extingu+0.219
flower+0.215
Neg logit contributions
ets-0.105
 anymore-0.102
fters-0.099
 mats-0.096
 scores-0.096
Token them
Token ID606
Feature activation-1.18e-01

Pos logit contributions
fters+0.229
ets+0.211
 mats+0.208
 quizz+0.205
 scores+0.203
Neg logit contributions
wood-0.341
fly-0.321
flower-0.319
 extingu-0.317
tha-0.315
Token.
Token ID13
Feature activation-2.92e-02

Pos logit contributions
fters+0.164
ets+0.153
 mats+0.148
 quizz+0.146
 scores+0.145
Neg logit contributions
wood-0.238
tha-0.224
fly-0.223
flower-0.221
 extingu-0.221
Token But
Token ID887
Feature activation-1.84e-01

Pos logit contributions
fters+0.055
ets+0.052
 mats+0.051
 quizz+0.050
 scores+0.050
Neg logit contributions
wood-0.045
fly-0.042
flower-0.041
tha-0.041
man-0.041
Token went
Token ID1816
Feature activation+9.77e-02

Pos logit contributions
wood+0.141
fly+0.133
flower+0.132
 extingu+0.131
tha+0.130
Neg logit contributions
fters-0.112
ets-0.104
 mats-0.100
 quizz-0.099
 else-0.098
Token to
Token ID284
Feature activation-2.14e-02

Pos logit contributions
wood+0.167
man+0.153
 extingu+0.150
fly+0.150
flower+0.150
Neg logit contributions
fters-0.175
ets-0.163
 scores-0.155
 mats-0.154
 quizz-0.153
Token her
Token ID607
Feature activation+5.20e-03

Pos logit contributions
fters+0.031
 quizz+0.028
 mats+0.027
 scores+0.027
ets+0.027
Neg logit contributions
wood-0.043
fly-0.041
flower-0.040
man-0.040
 extingu-0.039
Token mom
Token ID1995
Feature activation+9.22e-02

Pos logit contributions
wood+0.008
fly+0.007
flower+0.007
man+0.007
tha+0.007
Neg logit contributions
fters-0.010
ets-0.010
 mats-0.009
 quizz-0.009
 scores-0.009
Token and
Token ID290
Feature activation+9.34e-03

Pos logit contributions
wood+0.152
flower+0.142
man+0.141
 extingu+0.140
fly+0.139
Neg logit contributions
fters-0.173
ets-0.156
 scores-0.155
 mats-0.154
 tant-0.154
Token said
Token ID531
Feature activation+1.01e-01

Pos logit contributions
wood+0.015
fly+0.014
flower+0.014
man+0.014
 extingu+0.014
Neg logit contributions
fters-0.017
ets-0.016
 mats-0.015
 quizz-0.015
 scores-0.015
Token,
Token ID11
Feature activation+4.95e-03

Pos logit contributions
wood+0.150
fly+0.146
man+0.145
tha+0.144
flower+0.140
Neg logit contributions
fters-0.188
ets-0.184
 mats-0.173
 scores-0.172
 tant-0.170
Token "
Token ID366
Feature activation-6.71e-02

Pos logit contributions
wood+0.010
fly+0.009
man+0.009
flower+0.009
 extingu+0.009
Neg logit contributions
fters-0.007
ets-0.007
 mats-0.006
 scores-0.006
 quizz-0.006
TokenMom
Token ID29252
Feature activation+2.56e-01

Pos logit contributions
fters+0.145
 else+0.133
 mats+0.132
ets+0.131
 tant+0.129
Neg logit contributions
wood-0.089
fly-0.081
work-0.076
 extingu-0.075
tha-0.074
Token,
Token ID11
Feature activation+5.01e-03

Pos logit contributions
wood+0.576
 extingu+0.563
fly+0.544
tha+0.538
 Amen+0.537
Neg logit contributions
 anymore-0.280
ets-0.251
 else-0.239
fters-0.236
 scores-0.231
Token can
Token ID460
Feature activation-8.21e-02

Pos logit contributions
wood+0.010
fly+0.010
flower+0.010
man+0.010
tha+0.010
Neg logit contributions
fters-0.007
ets-0.006
 mats-0.006
 quizz-0.006
 else-0.006
Token
Token ID198
Feature activation-1.14e-02

Pos logit contributions
fters+0.025
ets+0.023
 mats+0.022
 quizz+0.022
 scores+0.022
Neg logit contributions
wood-0.030
fly-0.029
flower-0.028
man-0.028
 extingu-0.028
Token
Token ID198
Feature activation-1.40e-02

Pos logit contributions
fters+0.017
ets+0.016
 mats+0.015
 quizz+0.015
 scores+0.015
Neg logit contributions
wood-0.022
fly-0.021
man-0.020
flower-0.020
tha-0.020
TokenOne
Token ID3198
Feature activation-7.22e-02

Pos logit contributions
fters+0.022
ets+0.020
 mats+0.020
 quizz+0.019
 scores+0.019
Neg logit contributions
wood-0.026
fly-0.025
flower-0.024
man-0.024
tha-0.024
Token day
Token ID1110
Feature activation+1.25e-01

Pos logit contributions
 else+0.105
fters+0.103
ets+0.101
 mats+0.101
 quizz+0.098
Neg logit contributions
wood-0.140
 extingu-0.137
tha-0.131
fly-0.129
flower-0.124
Token,
Token ID11
Feature activation+8.88e-03

Pos logit contributions
wood+0.243
fly+0.229
flower+0.227
 extingu+0.224
man+0.224
Neg logit contributions
fters-0.187
ets-0.173
 mats-0.171
 quizz-0.170
 scores-0.167
Token they
Token ID484
Feature activation+1.07e-01

Pos logit contributions
wood+0.018
fly+0.017
 extingu+0.017
flower+0.016
man+0.016
Neg logit contributions
fters-0.012
 mats-0.012
 quizz-0.012
 else-0.012
 scores-0.011
Token see
Token ID766
Feature activation-5.98e-02

Pos logit contributions
wood+0.232
 extingu+0.220
tha+0.218
man+0.218
fly+0.215
Neg logit contributions
fters-0.123
ets-0.123
 else-0.121
 scores-0.117
 meets-0.116
Token a
Token ID257
Feature activation+1.24e-01

Pos logit contributions
fters+0.102
 else+0.101
 mats+0.099
 quizz+0.099
ets+0.099
Neg logit contributions
wood-0.099
flower-0.093
fly-0.092
tha-0.091
man-0.090
Token very
Token ID845
Feature activation+1.16e-01

Pos logit contributions
wood+0.249
tha+0.225
flower+0.225
 extingu+0.225
fly+0.224
Neg logit contributions
fters-0.177
 else-0.165
 quizz-0.163
 mats-0.161
 scores-0.161
Token big
Token ID1263
Feature activation+5.12e-02

Pos logit contributions
wood+0.233
fly+0.219
flower+0.218
 extingu+0.218
tha+0.213
Neg logit contributions
fters-0.165
ets-0.152
 else-0.149
 mats-0.149
 quizz-0.148
Token boat
Token ID8848
Feature activation+5.54e-02

Pos logit contributions
wood+0.105
tha+0.094
fly+0.094
 extingu+0.093
man+0.092
Neg logit contributions
fters-0.068
 scores-0.067
 quizz-0.067
 mats-0.067
 anymore-0.065
TokenL
Token ID43
Feature activation-1.99e-01

Pos logit contributions
fters+0.018
ets+0.016
 quizz+0.015
 scores+0.015
 mats+0.015
Neg logit contributions
wood-0.033
man-0.033
flower-0.032
tha-0.031
fly-0.031
Tokenily
Token ID813
Feature activation-2.19e-01

Pos logit contributions
fters+0.280
itionally+0.262
 tant+0.254
 calculations+0.252
 quizz+0.248
Neg logit contributions
man-0.355
wood-0.346
tha-0.340
flower-0.337
 extingu-0.337
Token says
Token ID1139
Feature activation+5.22e-02

Pos logit contributions
fters+0.396
ets+0.349
 tant+0.338
 quizz+0.336
 else+0.335
Neg logit contributions
wood-0.366
fly-0.343
flower-0.341
man-0.339
 extingu-0.337
Token,
Token ID11
Feature activation+6.13e-03

Pos logit contributions
wood+0.094
fly+0.089
man+0.087
flower+0.087
tha+0.087
Neg logit contributions
fters-0.086
ets-0.080
 mats-0.077
 quizz-0.076
 scores-0.076
Token "
Token ID366
Feature activation-6.13e-02

Pos logit contributions
wood+0.013
fly+0.013
flower+0.013
tha+0.013
 extingu+0.013
Neg logit contributions
fters-0.008
ets-0.007
 mats-0.007
 quizz-0.007
 scores-0.007
TokenYes
Token ID5297
Feature activation+1.95e-01

Pos logit contributions
fters+0.129
ets+0.123
 quizz+0.118
 mats+0.117
 scores+0.117
Neg logit contributions
wood-0.083
flower-0.077
fly-0.077
man-0.076
tha-0.074
Token,
Token ID11
Feature activation+5.93e-03

Pos logit contributions
wood+0.360
fly+0.339
flower+0.335
man+0.333
 extingu+0.332
Neg logit contributions
fters-0.311
ets-0.288
 mats-0.279
 quizz-0.277
 scores-0.274
Token Tom
Token ID4186
Feature activation+6.01e-03

Pos logit contributions
tha+0.008
 extingu+0.008
 Mill+0.008
work+0.008
man+0.008
Neg logit contributions
fters-0.011
ets-0.011
 tant-0.010
 corners-0.010
 meets-0.010
Token,
Token ID11
Feature activation+5.04e-03

Pos logit contributions
wood+0.015
work+0.014
tha+0.014
 extingu+0.014
fly+0.014
Neg logit contributions
ets-0.006
 anymore-0.006
 else-0.005
 scores-0.005
 quizz-0.005
Token I
Token ID314
Feature activation-2.28e-03

Pos logit contributions
fly+0.006
work+0.006
 Mill+0.005
 extingu+0.005
man+0.005
Neg logit contributions
 tant-0.010
ets-0.010
fters-0.010
pen-0.009
les-0.009
Token want
Token ID765
Feature activation-2.23e-02

Pos logit contributions
fters+0.004
ets+0.003
 mats+0.003
 quizz+0.003
 scores+0.003
Neg logit contributions
wood-0.004
fly-0.004
man-0.004
flower-0.004
 extingu-0.004
Tokeners
Token ID364
Feature activation-9.60e-02

Pos logit contributions
fters+0.064
ets+0.058
 mats+0.055
 quizz+0.055
 scores+0.054
Neg logit contributions
wood-0.124
fly-0.118
flower-0.117
man-0.117
tha-0.116
Token loved
Token ID6151
Feature activation-5.73e-02

Pos logit contributions
fters+0.150
ets+0.137
 mats+0.133
 quizz+0.131
 scores+0.131
Neg logit contributions
wood-0.180
fly-0.172
man-0.172
flower-0.171
tha-0.168
Token to
Token ID284
Feature activation-1.15e-02

Pos logit contributions
fters+0.096
ets+0.089
 scores+0.084
 mats+0.082
pler+0.082
Neg logit contributions
wood-0.104
fly-0.095
man-0.094
flower-0.094
 extingu-0.093
Token sleep
Token ID3993
Feature activation-6.51e-02

Pos logit contributions
fters+0.014
ets+0.013
 mats+0.012
 quizz+0.012
 scores+0.012
Neg logit contributions
wood-0.025
fly-0.024
flower-0.024
man-0.024
 extingu-0.024
Token all
Token ID477
Feature activation-4.84e-02

Pos logit contributions
fters+0.112
 scores+0.106
ets+0.103
 else+0.100
pieces+0.100
Neg logit contributions
man-0.107
wood-0.104
fly-0.096
flake-0.096
 extingu-0.096
Token day
Token ID1110
Feature activation+1.22e-01

Pos logit contributions
fters+0.090
ets+0.086
 mats+0.083
 quizz+0.083
 scores+0.082
Neg logit contributions
wood-0.076
fly-0.071
 extingu-0.070
tha-0.069
flower-0.069
Token long
Token ID890
Feature activation-3.22e-03

Pos logit contributions
wood+0.178
man+0.170
work+0.161
fly+0.160
tha+0.157
Neg logit contributions
fters-0.231
 scores-0.210
 meets-0.206
 mats-0.205
 calculations-0.205
Token and
Token ID290
Feature activation+1.13e-02

Pos logit contributions
fters+0.007
itionally+0.007
pieces+0.006
 scores+0.006
 meets+0.006
Neg logit contributions
wood-0.003
man-0.003
fly-0.003
 Frankie-0.003
 Mill-0.003
Token never
Token ID1239
Feature activation+1.24e-01

Pos logit contributions
wood+0.019
man+0.018
fly+0.018
flower+0.017
tha+0.017
Neg logit contributions
fters-0.019
ets-0.018
 scores-0.018
 mats-0.017
 quizz-0.017
Token wanted
Token ID2227
Feature activation-1.46e-02

Pos logit contributions
wood+0.207
fly+0.199
flower+0.195
man+0.186
work+0.184
Neg logit contributions
fters-0.231
 mats-0.202
ets-0.198
pler-0.198
pieces-0.197
Token to
Token ID284
Feature activation-1.24e-02

Pos logit contributions
fters+0.034
ets+0.029
itionally+0.029
pler+0.029
 tant+0.028
Neg logit contributions
wood-0.020
 Mill-0.016
fly-0.015
work-0.015
 Tit-0.015
Token.
Token ID13
Feature activation-5.18e-03

Pos logit contributions
fters+0.227
ets+0.203
 tant+0.200
 mats+0.200
itionally+0.198
Neg logit contributions
wood-0.217
flower-0.206
fly-0.204
 extingu-0.202
work-0.201
Token
Token ID198
Feature activation-1.79e-02

Pos logit contributions
fters+0.008
ets+0.007
 mats+0.007
 quizz+0.007
 scores+0.007
Neg logit contributions
wood-0.010
fly-0.009
flower-0.009
man-0.009
 extingu-0.009
Token
Token ID198
Feature activation-1.79e-02

Pos logit contributions
fters+0.028
ets+0.026
 mats+0.025
 scores+0.025
 quizz+0.025
Neg logit contributions
wood-0.033
man-0.031
fly-0.031
tha-0.030
flower-0.030
TokenThe
Token ID464
Feature activation-3.35e-02

Pos logit contributions
fters+0.026
ets+0.024
 scores+0.023
 quizz+0.023
 mats+0.023
Neg logit contributions
wood-0.035
flower-0.033
fly-0.033
man-0.033
tha-0.033
Token little
Token ID1310
Feature activation-1.19e-01

Pos logit contributions
fters+0.052
ets+0.048
 scores+0.046
 loses+0.046
 quizz+0.045
Neg logit contributions
wood-0.061
flower-0.061
man-0.060
 extingu-0.059
tha-0.058
Token girl
Token ID2576
Feature activation+8.17e-02

Pos logit contributions
fters+0.176
 mats+0.172
ets+0.172
 quizz+0.162
 tant+0.162
Neg logit contributions
wood-0.221
fly-0.213
man-0.208
tha-0.207
 extingu-0.204
Token was
Token ID373
Feature activation+1.33e-02

Pos logit contributions
wood+0.130
flower+0.127
man+0.125
fly+0.125
tha+0.123
Neg logit contributions
fters-0.150
ets-0.137
 mats-0.133
 quizz-0.131
 tant-0.130
Token so
Token ID523
Feature activation+1.65e-02

Pos logit contributions
wood+0.024
fly+0.022
flower+0.022
man+0.022
tha+0.022
Neg logit contributions
fters-0.022
ets-0.021
 mats-0.020
 scores-0.019
 quizz-0.019
Token proud
Token ID6613
Feature activation+2.31e-02

Pos logit contributions
wood+0.036
fly+0.034
 extingu+0.034
man+0.034
flower+0.033
Neg logit contributions
fters-0.022
 else-0.020
ets-0.020
 quizz-0.019
 scores-0.019
Token of
Token ID286
Feature activation-3.23e-02

Pos logit contributions
wood+0.038
flower+0.036
fly+0.035
man+0.035
 extingu+0.035
Neg logit contributions
fters-0.043
ets-0.037
itionally-0.037
 tant-0.037
 mats-0.036
Token herself
Token ID5223
Feature activation-1.79e-02

Pos logit contributions
fters+0.055
ets+0.053
 mats+0.050
 tant+0.050
 scores+0.050
Neg logit contributions
wood-0.055
tha-0.052
fly-0.052
flower-0.052
 extingu-0.050
Token didn
Token ID1422
Feature activation-6.31e-02

Pos logit contributions
fters+0.009
ets+0.008
 mats+0.008
 quizz+0.008
 else+0.008
Neg logit contributions
wood-0.011
fly-0.010
 extingu-0.010
flower-0.010
man-0.010
Token't
Token ID470
Feature activation+7.89e-02

Pos logit contributions
fters+0.086
 anymore+0.073
 else+0.072
 rides+0.071
 quizz+0.070
Neg logit contributions
wood-0.140
flower-0.136
fly-0.134
 extingu-0.131
tha-0.130
Token have
Token ID423
Feature activation-2.25e-02

Pos logit contributions
wood+0.157
fly+0.151
 extingu+0.149
flower+0.149
tha+0.148
Neg logit contributions
fters-0.116
ets-0.110
 quizz-0.103
 else-0.102
 scores-0.101
Token enough
Token ID1576
Feature activation+1.07e-01

Pos logit contributions
fters+0.043
ets+0.039
 mats+0.038
 tant+0.037
 calculations+0.037
Neg logit contributions
wood-0.035
tha-0.031
fly-0.031
man-0.031
flower-0.030
Token money
Token ID1637
Feature activation-1.00e-01

Pos logit contributions
wood+0.228
 extingu+0.213
man+0.212
tha+0.212
fly+0.212
Neg logit contributions
fters-0.139
ets-0.125
 mats-0.121
 quizz-0.120
 scores-0.119
Token.
Token ID13
Feature activation+1.23e-03

Pos logit contributions
fters+0.139
ets+0.133
 mats+0.126
 else+0.124
 quizz+0.124
Neg logit contributions
wood-0.200
fly-0.193
flower-0.190
 extingu-0.188
man-0.187
Token She
Token ID1375
Feature activation+8.37e-02

Pos logit contributions
 Mill+0.002
 Frankie+0.001
wood+0.001
 Martha+0.001
 Amen+0.001
Neg logit contributions
fters-0.003
ets-0.003
 scores-0.002
 laps-0.002
 loses-0.002
Token stared
Token ID18484
Feature activation-1.95e-02

Pos logit contributions
wood+0.150
fly+0.140
man+0.139
flower+0.139
tha+0.139
Neg logit contributions
fters-0.139
ets-0.129
 mats-0.125
 quizz-0.124
 scores-0.123
Token at
Token ID379
Feature activation-1.70e-01

Pos logit contributions
fters+0.023
ets+0.020
 mats+0.019
 scores+0.019
 quizz+0.019
Neg logit contributions
wood-0.045
flower-0.042
fly-0.042
man-0.042
 extingu-0.042
Token the
Token ID262
Feature activation+7.14e-03

Pos logit contributions
fters+0.304
ets+0.283
 mats+0.271
 scores+0.269
 quizz+0.269
Neg logit contributions
wood-0.281
fly-0.262
tha-0.261
man-0.259
flower-0.259
Token taxi
Token ID17536
Feature activation+1.35e-01

Pos logit contributions
wood+0.011
fly+0.011
flower+0.010
 extingu+0.010
tha+0.010
Neg logit contributions
fters-0.013
ets-0.012
 mats-0.012
 quizz-0.012
 scores-0.012
Token
Token ID198
Feature activation-3.27e-02

Pos logit contributions
fters+0.055
ets+0.051
 mats+0.049
 quizz+0.049
 scores+0.049
Neg logit contributions
wood-0.058
man-0.054
fly-0.054
 extingu-0.053
tha-0.053
TokenTom
Token ID13787
Feature activation-9.40e-02

Pos logit contributions
fters+0.051
 quizz+0.048
ets+0.047
 scores+0.046
 anymore+0.046
Neg logit contributions
wood-0.059
man-0.059
tha-0.055
 extingu-0.055
flower-0.055
Token and
Token ID290
Feature activation-2.55e-03

Pos logit contributions
fters+0.175
 quizz+0.156
pieces+0.155
 calculations+0.154
ets+0.153
Neg logit contributions
man-0.145
fly-0.142
wood-0.142
flower-0.139
tha-0.138
Token Lily
Token ID20037
Feature activation-2.16e-01

Pos logit contributions
fters+0.004
ets+0.003
 calculations+0.003
 scores+0.003
 quizz+0.003
Neg logit contributions
wood-0.005
fly-0.005
man-0.005
 extingu-0.005
tha-0.005
Token ran
Token ID4966
Feature activation-1.66e-02

Pos logit contributions
fters+0.412
 quizz+0.377
ets+0.374
 calculations+0.368
 mats+0.363
Neg logit contributions
wood-0.311
fly-0.310
flower-0.307
tha-0.304
 extingu-0.299
Token to
Token ID284
Feature activation-1.86e-02

Pos logit contributions
fters+0.030
 calculations+0.026
 quizz+0.026
pieces+0.025
ets+0.025
Neg logit contributions
wood-0.029
 extingu-0.026
man-0.026
fly-0.025
tha-0.025
Token their
Token ID511
Feature activation+1.36e-01

Pos logit contributions
fters+0.032
ets+0.030
 mats+0.029
 quizz+0.029
 scores+0.029
Neg logit contributions
wood-0.032
fly-0.030
flower-0.030
 extingu-0.029
man-0.029
Token mom
Token ID1995
Feature activation+8.55e-02

Pos logit contributions
wood+0.202
man+0.184
 extingu+0.183
flower+0.182
fly+0.180
Neg logit contributions
fters-0.282
ets-0.264
 calculations-0.243
 quizz-0.241
 tant-0.237
Token,
Token ID11
Feature activation-8.93e-04

Pos logit contributions
wood+0.156
fly+0.146
 extingu+0.145
man+0.145
flower+0.145
Neg logit contributions
fters-0.140
ets-0.128
 mats-0.124
 quizz-0.123
 scores-0.123
Token who
Token ID508
Feature activation-1.54e-01

Pos logit contributions
fters+0.001
ets+0.001
 mats+0.001
 quizz+0.001
 scores+0.001
Neg logit contributions
wood-0.002
fly-0.002
flower-0.002
man-0.002
 extingu-0.002
Token was
Token ID373
Feature activation-2.42e-03

Pos logit contributions
fters+0.208
 else+0.198
 mats+0.185
 rides+0.185
 quizz+0.184
Neg logit contributions
wood-0.325
fly-0.297
work-0.293
 extingu-0.292
flower-0.291
Token did
Token ID750
Feature activation-5.48e-02

Pos logit contributions
fters+0.140
ets+0.125
 mats+0.125
 quizz+0.122
 tant+0.121
Neg logit contributions
wood-0.169
fly-0.163
flower-0.159
man-0.159
 extingu-0.156
Token with
Token ID351
Feature activation-8.91e-02

Pos logit contributions
fters+0.105
 tant+0.104
itionally+0.100
 mats+0.098
 quizz+0.096
Neg logit contributions
man-0.076
wood-0.074
fly-0.070
work-0.069
flower-0.067
Token Fl
Token ID1610
Feature activation-5.50e-03

Pos logit contributions
fters+0.165
ets+0.158
 mats+0.148
and+0.145
 scores+0.143
Neg logit contributions
wood-0.139
man-0.131
tha-0.128
fly-0.126
 extingu-0.126
Tokenuffy
Token ID15352
Feature activation+8.42e-02

Pos logit contributions
fters+0.008
ets+0.008
 mats+0.008
 scores+0.008
 else+0.008
Neg logit contributions
wood-0.010
 extingu-0.010
fly-0.009
man-0.009
flower-0.009
Token.
Token ID13
Feature activation-1.61e-02

Pos logit contributions
wood+0.141
fly+0.136
man+0.133
flower+0.132
work+0.130
Neg logit contributions
fters-0.151
ets-0.136
 mats-0.132
 tant-0.131
 calculations-0.130
Token He
Token ID679
Feature activation+1.06e-02

Pos logit contributions
fters+0.028
ets+0.026
 scores+0.024
 mats+0.024
 quizz+0.024
Neg logit contributions
wood-0.028
fly-0.026
man-0.026
 extingu-0.026
flower-0.025
Token hopes
Token ID7176
Feature activation+1.99e-02

Pos logit contributions
wood+0.020
fly+0.019
 extingu+0.019
flower+0.019
man+0.019
Neg logit contributions
fters-0.016
ets-0.015
 mats-0.014
 quizz-0.014
 scores-0.014
Token it
Token ID340
Feature activation-3.84e-02

Pos logit contributions
wood+0.037
fly+0.034
man+0.034
tha+0.034
 extingu+0.034
Neg logit contributions
fters-0.032
ets-0.030
 mats-0.028
 scores-0.028
 quizz-0.028
Token will
Token ID481
Feature activation+1.48e-01

Pos logit contributions
fters+0.075
ets+0.071
 mats+0.069
 quizz+0.068
 scores+0.068
Neg logit contributions
wood-0.057
fly-0.052
flower-0.052
tha-0.052
 extingu-0.051
Token hatch
Token ID25834
Feature activation+1.89e-01

Pos logit contributions
wood+0.304
tha+0.287
flower+0.282
 extingu+0.277
work+0.273
Neg logit contributions
fters-0.207
ets-0.191
 anymore-0.184
 else-0.183
 loses-0.175
Token and
Token ID290
Feature activation-4.19e-03

Pos logit contributions
wood+0.346
 extingu+0.344
tha+0.335
 Amen+0.324
fly+0.319
Neg logit contributions
ets-0.267
 anymore-0.265
fters-0.258
 else-0.254
and-0.250
TokenThe
Token ID464
Feature activation-3.53e-02

Pos logit contributions
ets+0.021
fters+0.021
 anymore+0.021
 scores+0.021
 quizz+0.020
Neg logit contributions
wood-0.024
flower-0.023
man-0.022
tha-0.022
fly-0.022
Token dog
Token ID3290
Feature activation+1.39e-02

Pos logit contributions
fters+0.058
ets+0.054
 quizz+0.052
 mats+0.052
 scores+0.051
Neg logit contributions
wood-0.064
fly-0.060
flower-0.060
man-0.060
 extingu-0.059
Token took
Token ID1718
Feature activation+2.29e-03

Pos logit contributions
wood+0.022
fly+0.021
man+0.021
flower+0.020
tha+0.019
Neg logit contributions
fters-0.027
ets-0.023
 tant-0.023
 quizz-0.023
 conclusions-0.023
Token the
Token ID262
Feature activation+8.32e-03

Pos logit contributions
wood+0.004
fly+0.004
flower+0.004
man+0.004
tha+0.004
Neg logit contributions
fters-0.004
ets-0.004
 mats-0.003
 quizz-0.003
 scores-0.003
Token ball
Token ID2613
Feature activation-6.40e-02

Pos logit contributions
wood+0.015
flower+0.015
man+0.015
 extingu+0.014
fly+0.014
Neg logit contributions
fters-0.013
ets-0.012
 quizz-0.012
 calculations-0.011
and-0.011
Token back
Token ID736
Feature activation-4.51e-02

Pos logit contributions
fters+0.086
ets+0.085
 mats+0.082
 anymore+0.082
and+0.081
Neg logit contributions
wood-0.125
tha-0.123
fly-0.119
 Amen-0.119
flower-0.118
Token to
Token ID284
Feature activation-1.44e-02

Pos logit contributions
and+0.046
ets+0.045
 mats+0.045
 corners+0.044
 anymore+0.043
Neg logit contributions
 Amen-0.104
fly-0.098
tha-0.098
 extingu-0.097
wood-0.096
Token her
Token ID607
Feature activation+1.41e-02

Pos logit contributions
fters+0.026
ets+0.025
 mats+0.024
 tant+0.023
 scores+0.023
Neg logit contributions
wood-0.023
fly-0.022
tha-0.022
man-0.021
 extingu-0.021
Token toy
Token ID13373
Feature activation-6.27e-02

Pos logit contributions
wood+0.025
fly+0.023
flower+0.023
man+0.023
 extingu+0.023
Neg logit contributions
fters-0.025
ets-0.023
 quizz-0.022
 mats-0.022
 calculations-0.021
Token,
Token ID11
Feature activation+1.46e-02

Pos logit contributions
fters+0.099
ets+0.090
 quizz+0.085
 mats+0.083
 scores+0.083
Neg logit contributions
wood-0.120
fly-0.114
man-0.114
flower-0.114
 extingu-0.110
Token and
Token ID290
Feature activation+1.52e-02

Pos logit contributions
wood+0.027
fly+0.026
flower+0.026
 extingu+0.026
man+0.026
Neg logit contributions
fters-0.023
ets-0.021
 quizz-0.020
 mats-0.020
 scores-0.019
Token show
Token ID905
Feature activation-1.08e-01

Pos logit contributions
work+0.024
man+0.024
 Amen+0.022
wood+0.022
tha+0.022
Neg logit contributions
fters-0.020
and-0.019
ets-0.018
 tant-0.018
itionally-0.017
Token,
Token ID11
Feature activation+1.97e-02

Pos logit contributions
fters+0.173
ets+0.156
 mats+0.153
 scores+0.150
 quizz+0.150
Neg logit contributions
wood-0.200
man-0.187
fly-0.185
flower-0.183
work-0.182
Token and
Token ID290
Feature activation+1.77e-02

Pos logit contributions
wood+0.043
 extingu+0.041
man+0.041
fly+0.041
flower+0.040
Neg logit contributions
fters-0.024
ets-0.022
 scores-0.021
 mats-0.020
 quizz-0.020
Token he
Token ID339
Feature activation-6.76e-02

Pos logit contributions
wood+0.032
fly+0.030
flower+0.030
 extingu+0.029
tha+0.029
Neg logit contributions
fters-0.029
ets-0.027
 mats-0.026
 quizz-0.026
 else-0.026
Token said
Token ID531
Feature activation+1.10e-01

Pos logit contributions
 else+0.098
fters+0.091
 scored+0.090
ets+0.089
 quizz+0.088
Neg logit contributions
wood-0.133
 extingu-0.126
fly-0.123
flower-0.121
tha-0.119
Token he
Token ID339
Feature activation-6.76e-02

Pos logit contributions
wood+0.193
man+0.185
fly+0.184
tha+0.182
work+0.180
Neg logit contributions
fters-0.180
ets-0.170
 mats-0.162
 scores-0.160
 quizz-0.158
Token wanted
Token ID2227
Feature activation-1.15e-02

Pos logit contributions
 else+0.077
ets+0.071
fters+0.070
 scored+0.069
 scores+0.065
Neg logit contributions
wood-0.151
 extingu-0.149
flower-0.143
tha-0.141
fly-0.139
Token to
Token ID284
Feature activation-1.00e-02

Pos logit contributions
fters+0.021
ets+0.019
 mats+0.018
 tant+0.018
 scores+0.018
Neg logit contributions
wood-0.020
man-0.018
work-0.018
fly-0.017
 extingu-0.017
Token dance
Token ID9280
Feature activation-4.11e-02

Pos logit contributions
fters+0.013
 else+0.011
 quizz+0.011
ets+0.011
 scores+0.011
Neg logit contributions
wood-0.022
fly-0.021
flower-0.021
 extingu-0.021
man-0.021
Token and
Token ID290
Feature activation+1.78e-02

Pos logit contributions
fters+0.072
ets+0.063
pler+0.062
 scores+0.061
 mats+0.061
Neg logit contributions
wood-0.070
man-0.066
work-0.064
fly-0.064
flower-0.062
Token make
Token ID787
Feature activation-7.87e-03

Pos logit contributions
wood+0.034
fly+0.032
work+0.031
flower+0.031
man+0.031
Neg logit contributions
fters-0.027
ets-0.025
 tant-0.024
 mats-0.024
 quizz-0.023
Token.
Token ID13
Feature activation+3.47e-03

Pos logit contributions
fters+0.258
ets+0.234
 mats+0.230
 quizz+0.228
 scores+0.225
Neg logit contributions
wood-0.284
fly-0.264
 extingu-0.258
man-0.258
flower-0.257
Token
Token ID198
Feature activation-1.16e-02

Pos logit contributions
wood+0.006
man+0.006
fly+0.006
 Mill+0.006
flower+0.006
Neg logit contributions
fters-0.006
ets-0.006
 loses-0.005
 scores-0.005
 tant-0.005
Token
Token ID198
Feature activation-1.11e-02

Pos logit contributions
fters+0.020
ets+0.020
 mats+0.019
 scores+0.018
 quizz+0.018
Neg logit contributions
wood-0.019
man-0.018
fly-0.018
tha-0.017
flower-0.017
TokenThe
Token ID464
Feature activation-2.71e-02

Pos logit contributions
ets+0.019
fters+0.018
 loses+0.018
 scores+0.018
 tant+0.017
Neg logit contributions
man-0.018
wood-0.018
fly-0.017
 Amen-0.017
flower-0.017
Token bear
Token ID6842
Feature activation+8.54e-02

Pos logit contributions
fters+0.059
ets+0.059
 scores+0.056
 loses+0.056
 quizz+0.055
Neg logit contributions
man-0.035
wood-0.035
flower-0.033
fly-0.032
 extingu-0.030
Token learned
Token ID4499
Feature activation-1.90e-01

Pos logit contributions
wood+0.149
man+0.145
fly+0.144
flower+0.143
tha+0.134
Neg logit contributions
fters-0.151
ets-0.131
 mats-0.131
 tant-0.130
 scores-0.129
Token how
Token ID703
Feature activation-1.05e-02

Pos logit contributions
fters+0.335
ets+0.311
pler+0.298
 scores+0.295
 mats+0.292
Neg logit contributions
wood-0.328
fly-0.298
 extingu-0.296
man-0.296
work-0.288
Token to
Token ID284
Feature activation-7.04e-03

Pos logit contributions
fters+0.019
ets+0.017
 mats+0.017
 tant+0.016
 scores+0.016
Neg logit contributions
wood-0.018
fly-0.016
man-0.016
flower-0.016
tha-0.015
Token climb
Token ID12080
Feature activation-3.67e-02

Pos logit contributions
fters+0.009
ets+0.008
 mats+0.008
 quizz+0.008
 else+0.008
Neg logit contributions
wood-0.015
fly-0.014
flower-0.014
man-0.014
tha-0.014
Token trees
Token ID7150
Feature activation-1.50e-01

Pos logit contributions
fters+0.055
ets+0.051
 mats+0.049
 quizz+0.048
 scores+0.048
Neg logit contributions
wood-0.071
fly-0.067
flower-0.066
man-0.066
 extingu-0.066
Token and
Token ID290
Feature activation+1.83e-02

Pos logit contributions
fters+0.233
ets+0.215
 mats+0.208
 quizz+0.206
 scores+0.205
Neg logit contributions
wood-0.284
fly-0.268
flower-0.264
man-0.263
 extingu-0.262
Token drink
Token ID4144
Feature activation-4.73e-02

Pos logit contributions
fters+0.004
 mats+0.004
 quizz+0.004
 rides+0.004
 dresses+0.004
Neg logit contributions
wood-0.015
tha-0.014
fly-0.014
flower-0.014
man-0.014
Token coffee
Token ID6891
Feature activation-2.96e-02

Pos logit contributions
fters+0.071
ets+0.066
 mats+0.065
 quizz+0.064
 else+0.063
Neg logit contributions
wood-0.091
fly-0.087
tha-0.086
flower-0.085
man-0.084
Token every
Token ID790
Feature activation-7.79e-02

Pos logit contributions
fters+0.041
ets+0.038
 mats+0.038
 quizz+0.037
 else+0.036
Neg logit contributions
wood-0.061
fly-0.058
tha-0.057
flower-0.056
 extingu-0.056
Token morning
Token ID3329
Feature activation+9.77e-02

Pos logit contributions
fters+0.122
 mats+0.117
 else+0.115
 quizz+0.114
ets+0.114
Neg logit contributions
wood-0.142
 extingu-0.134
tha-0.133
fly-0.133
flower-0.127
Token,
Token ID11
Feature activation+1.74e-02

Pos logit contributions
wood+0.175
man+0.162
flower+0.162
 extingu+0.161
fly+0.161
Neg logit contributions
fters-0.162
ets-0.146
 scores-0.143
 mats-0.142
 quizz-0.141
Token but
Token ID475
Feature activation-1.74e-01

Pos logit contributions
wood+0.035
flower+0.034
 extingu+0.033
man+0.032
fly+0.032
Neg logit contributions
fters-0.023
ets-0.022
 scores-0.022
 quizz-0.020
 calculations-0.020
Token one
Token ID530
Feature activation-1.32e-02

Pos logit contributions
fters+0.246
ets+0.220
 mats+0.220
 quizz+0.216
 else+0.215
Neg logit contributions
wood-0.352
fly-0.332
 extingu-0.329
flower-0.326
tha-0.325
Token day
Token ID1110
Feature activation+1.26e-01

Pos logit contributions
fters+0.023
 else+0.022
ets+0.022
 quizz+0.021
 mats+0.021
Neg logit contributions
wood-0.022
fly-0.020
tha-0.020
 extingu-0.020
work-0.020
Token he
Token ID339
Feature activation-6.53e-02

Pos logit contributions
wood+0.242
fly+0.228
man+0.225
tha+0.225
flower+0.225
Neg logit contributions
fters-0.193
ets-0.179
 mats-0.168
 scores-0.165
 quizz-0.164
Token spilled
Token ID31896
Feature activation-5.11e-02

Pos logit contributions
fters+0.101
ets+0.092
 mats+0.088
 quizz+0.088
 calculations+0.087
Neg logit contributions
wood-0.122
flower-0.118
fly-0.117
man-0.116
tha-0.115
Token it
Token ID340
Feature activation-2.06e-02

Pos logit contributions
fters+0.067
 mats+0.063
ets+0.063
 tant+0.062
 anymore+0.061
Neg logit contributions
wood-0.105
tha-0.105
fly-0.104
flower-0.100
man-0.099
Token.
Token ID13
Feature activation-2.03e-02

Pos logit contributions
fters+0.361
ets+0.327
 mats+0.314
 quizz+0.314
 scores+0.311
Neg logit contributions
wood-0.346
flower-0.320
man-0.320
fly-0.320
 extingu-0.317
Token I
Token ID314
Feature activation-3.12e-03

Pos logit contributions
fters+0.032
ets+0.030
 mats+0.029
 scores+0.028
 quizz+0.028
Neg logit contributions
wood-0.038
fly-0.035
flower-0.035
 extingu-0.035
man-0.035
Token'm
Token ID1101
Feature activation+1.86e-01

Pos logit contributions
fters+0.005
ets+0.005
 mats+0.005
 quizz+0.005
 scores+0.005
Neg logit contributions
wood-0.005
fly-0.005
flower-0.005
man-0.005
 extingu-0.005
Token sorry
Token ID7926
Feature activation-5.29e-02

Pos logit contributions
wood+0.363
fly+0.345
tha+0.341
flower+0.340
man+0.339
Neg logit contributions
fters-0.273
ets-0.254
 mats-0.247
 scores-0.239
 quizz-0.239
Token,
Token ID11
Feature activation-1.09e-03

Pos logit contributions
fters+0.098
ets+0.090
 mats+0.086
 tant+0.086
 quizz+0.086
Neg logit contributions
wood-0.085
man-0.081
fly-0.080
flower-0.080
tha-0.076
Token Lily
Token ID20037
Feature activation-2.15e-01

Pos logit contributions
fters+0.002
ets+0.002
 mats+0.002
 quizz+0.002
 scores+0.002
Neg logit contributions
wood-0.002
flower-0.002
fly-0.002
man-0.002
tha-0.002
Token.
Token ID13
Feature activation-2.20e-02

Pos logit contributions
fters+0.287
ets+0.273
 else+0.258
 anymore+0.257
 scores+0.256
Neg logit contributions
wood-0.449
fly-0.420
 extingu-0.419
tha-0.415
flower-0.410
Token I
Token ID314
Feature activation-3.66e-03

Pos logit contributions
fters+0.033
ets+0.031
 mats+0.030
 quizz+0.029
 scores+0.029
Neg logit contributions
wood-0.043
fly-0.040
flower-0.040
man-0.039
tha-0.039
Token'm
Token ID1101
Feature activation+1.86e-01

Pos logit contributions
fters+0.006
ets+0.005
 mats+0.005
 quizz+0.005
 scores+0.005
Neg logit contributions
wood-0.007
fly-0.006
flower-0.006
tha-0.006
man-0.006
Token sorry
Token ID7926
Feature activation-5.25e-02

Pos logit contributions
wood+0.355
fly+0.340
tha+0.336
flower+0.334
man+0.334
Neg logit contributions
fters-0.278
ets-0.260
 mats-0.253
 scores-0.244
 quizz-0.244
Token,
Token ID11
Feature activation-2.45e-03

Pos logit contributions
fters+0.098
ets+0.090
 tant+0.087
 mats+0.087
 quizz+0.086
Neg logit contributions
wood-0.083
man-0.080
fly-0.079
flower-0.078
work-0.074
Token soap
Token ID19533
Feature activation-2.17e-02

Pos logit contributions
wood+0.093
fly+0.087
tha+0.086
flower+0.085
man+0.085
Neg logit contributions
fters-0.071
ets-0.066
 mats-0.065
 quizz-0.064
 scores-0.064
Token box
Token ID3091
Feature activation+6.45e-02

Pos logit contributions
fters+0.029
ets+0.027
 mats+0.027
 scores+0.026
 quizz+0.026
Neg logit contributions
wood-0.044
tha-0.042
fly-0.041
flower-0.040
 extingu-0.040
Token and
Token ID290
Feature activation+1.09e-02

Pos logit contributions
wood+0.138
fly+0.132
tha+0.131
flower+0.129
 extingu+0.128
Neg logit contributions
ets-0.077
fters-0.077
 mats-0.071
 else-0.071
 anymore-0.069
Token smiled
Token ID13541
Feature activation+1.64e-02

Pos logit contributions
wood+0.022
fly+0.021
flower+0.021
man+0.021
tha+0.020
Neg logit contributions
fters-0.015
ets-0.014
 mats-0.014
 quizz-0.014
 else-0.014
Token.
Token ID13
Feature activation-1.48e-03

Pos logit contributions
wood+0.036
fly+0.035
tha+0.035
flower+0.034
 extingu+0.034
Neg logit contributions
fters-0.017
 anymore-0.017
ets-0.017
 else-0.017
 quizz-0.017
Token But
Token ID887
Feature activation-1.82e-01

Pos logit contributions
fters+0.003
ets+0.002
 else+0.002
 scores+0.002
 tant+0.002
Neg logit contributions
wood-0.003
man-0.002
 extingu-0.002
fly-0.002
work-0.002
Token when
Token ID618
Feature activation-5.73e-02

Pos logit contributions
fters+0.320
ets+0.310
 mats+0.292
 quizz+0.289
 scores+0.287
Neg logit contributions
wood-0.313
tha-0.289
fly-0.289
flower-0.288
man-0.288
Token he
Token ID339
Feature activation-7.40e-02

Pos logit contributions
fters+0.059
 quizz+0.052
 mats+0.051
 else+0.051
 scores+0.051
Neg logit contributions
wood-0.136
 extingu-0.133
fly-0.131
flower-0.130
work-0.127
Token turned
Token ID2900
Feature activation+1.11e-01

Pos logit contributions
fters+0.122
ets+0.115
 mats+0.113
 quizz+0.109
 scores+0.107
Neg logit contributions
wood-0.130
fly-0.126
flower-0.126
tha-0.124
man-0.123
Token around
Token ID1088
Feature activation-2.11e-01

Pos logit contributions
wood+0.178
man+0.160
 extingu+0.158
tha+0.157
flower+0.156
Neg logit contributions
fters-0.206
ets-0.198
 mats-0.187
itionally-0.186
 quizz-0.182
Token,
Token ID11
Feature activation+1.31e-02

Pos logit contributions
fters+0.401
ets+0.368
 mats+0.351
 scores+0.348
itionally+0.346
Neg logit contributions
wood-0.339
man-0.311
tha-0.306
 extingu-0.306
fly-0.305
Token want
Token ID765
Feature activation-3.12e-02

Pos logit contributions
wood+0.140
fly+0.131
flower+0.129
man+0.129
 extingu+0.128
Neg logit contributions
fters-0.123
ets-0.114
 mats-0.111
 quizz-0.110
 scores-0.108
Token to
Token ID284
Feature activation-1.43e-02

Pos logit contributions
fters+0.061
ets+0.053
pler+0.052
 tant+0.051
 mats+0.051
Neg logit contributions
wood-0.051
fly-0.045
man-0.045
work-0.045
 extingu-0.043
Token join
Token ID4654
Feature activation-8.34e-02

Pos logit contributions
fters+0.019
ets+0.017
 scores+0.017
 quizz+0.016
 mats+0.016
Neg logit contributions
wood-0.030
fly-0.029
flower-0.029
 extingu-0.029
tha-0.028
Token the
Token ID262
Feature activation+5.01e-03

Pos logit contributions
fters+0.143
ets+0.131
 mats+0.126
 scores+0.122
pler+0.122
Neg logit contributions
wood-0.146
fly-0.139
work-0.133
man-0.133
flower-0.132
Token team
Token ID1074
Feature activation-8.35e-02

Pos logit contributions
wood+0.010
man+0.009
fly+0.009
flower+0.009
work+0.009
Neg logit contributions
fters-0.007
ets-0.007
 else-0.006
 tant-0.006
 mats-0.006
Token at
Token ID379
Feature activation-1.71e-01

Pos logit contributions
fters+0.149
ets+0.129
 mats+0.129
 tant+0.127
 calculations+0.125
Neg logit contributions
wood-0.142
fly-0.136
man-0.132
work-0.130
flower-0.130
Token first
Token ID717
Feature activation+8.18e-02

Pos logit contributions
fters+0.261
 quizz+0.241
ets+0.240
 scores+0.238
 mats+0.236
Neg logit contributions
wood-0.325
flower-0.308
fly-0.308
 extingu-0.306
tha-0.303
Token,
Token ID11
Feature activation+8.88e-03

Pos logit contributions
wood+0.130
fly+0.122
man+0.118
work+0.116
flower+0.115
Neg logit contributions
fters-0.156
ets-0.136
 mats-0.130
itionally-0.130
 tant-0.129
Token but
Token ID475
Feature activation-1.80e-01

Pos logit contributions
wood+0.017
fly+0.016
flower+0.016
man+0.016
 extingu+0.016
Neg logit contributions
fters-0.014
ets-0.012
 quizz-0.012
 mats-0.012
 scores-0.012
Token the
Token ID262
Feature activation+4.48e-03

Pos logit contributions
fters+0.266
ets+0.245
 mats+0.238
 quizz+0.236
 else+0.236
Neg logit contributions
wood-0.355
fly-0.333
 extingu-0.333
flower-0.332
man-0.329
Token other
Token ID584
Feature activation-2.55e-01

Pos logit contributions
wood+0.010
fly+0.009
flower+0.009
man+0.009
 extingu+0.009
Neg logit contributions
fters-0.006
ets-0.005
 mats-0.005
 quizz-0.005
 scores-0.005