TOP ACTIVATIONS
MAX = 9.570

big , round cookie with spec ks of sugar on the
high they were just tiny spec ks in the sky .
the leaf there were little spec ks of gold ! She
with lots of food . Them ouse 's mouth watered ,
away , a bright blue spe ck in the sky .
play chess . He showed Mate o how to move the
of the frog . He spe ared the lazy frog and
baby thought about the night  ´ s adventure as she
was ready , the b unn ies hopped over to the
The eggs were blue and spe ck led . They looked
. Grand dad also showed Mate o how to make careful
So , Grand dad taught Mate o how to play chess
After a while , Mate o was getting very good
on top . The b unn ies wanted to have a
watched , she realized that    the ground was alive
floor . " Â M ira mi
soon nothing but a small spe ck in the sky .
. They were blue and spe ck led . L ila
You are mean . B unn ies are nice and they
dad was so proud of Mate o 's skills . He

MIN ACTIVATIONS
MIN = -12.603

two squirrel s kept qu arre ling until finally Sammy said
together . They never qu arre led again . <|endoftext|> Once
on , they never qu arre led again over clothes .
and their children , qu arre ling .
mum and understood that qu arre lling was not a good
friends didn 't stop qu arre ling and were getting louder
walking by saw them qu arre ling . She was very
he found a rabbit qu arre lling with a squirrel .
but today they were qu arre ling . Jenna wanted to
little birds that were qu arre ling . The
on , they never qu arre led again and played together
â t qu arre ling , the furniture stayed
and saw two birds qu arre ling ! They were squ
Finally , after much qu arre ling , one of the
Lily and Jack never qu arre led again . They learned
the other kn obs qu arre led over who should be
but today they were qu arre ling . They were arguing
, why are you qu arre ling ?" But the birds
other . They starting qu arre ling again and their voices
never happy . He qu arre led and yelled and sc

INTERVAL 4.026 - 9.570
CONTAINS 1.692%

. But they were sad when they saw the prices .
. Maybe one day she could come back and tie a
a Tuesday morning , the sun was glowing brightly . Sally
on , they knew they could trust each other no matter
on , they knew they could always help each other when

INTERVAL -1.517 - 4.026
CONTAINS 82.672%

take in oxygen for all the animals to breathe . <|endoftext|>
mixed some ingredients together and made a y ummy shrimp dish
You have to stay hidden until I find you !" He
and smile . Then they start to pick up the blocks
caught . The puppy 's owners laughed and

INTERVAL -7.060 - -1.517
CONTAINS 15.584%

she heard a loud bark from her dog , Max .
a walk in the grass , he found a pen and
proud ! When they reached the entrance of the
and sang as loud as they could , but Billy was
faster . The dog ch ases them and tries to bite

INTERVAL -12.603 - -7.060
CONTAINS 0.051%

help . The moral of the story is that
in different rooms and makes stories for her dolls .
favorite thing to write was stories about sailing . Every day
happy . The moral of the story is :
away . The moral of the story is that
Token big
Token ID1263
Feature activation-2.70e-01

Pos logit contributions
nuts+0.029
cases+0.028
 Sale+0.028
 puzzles+0.028
 invitations+0.028
Neg logit contributions
avis-0.038
 sur-0.036
 rely-0.036
heart-0.034
 bas-0.034
Token,
Token ID11
Feature activation-1.38e+00

Pos logit contributions
cases+0.009
nuts+0.008
ils+0.008
 Sale+0.008
 invitations+0.008
Neg logit contributions
 sur-0.008
avis-0.008
 rely-0.008
please-0.008
 listener-0.007
Token round
Token ID2835
Feature activation-2.03e-01

Pos logit contributions
cases+0.053
ils+0.052
nuts+0.050
 Sale+0.050
 invitations+0.048
Neg logit contributions
 sur-0.033
avis-0.032
 rely-0.031
please-0.030
 polar-0.030
Token cookie
Token ID19751
Feature activation-2.49e+00

Pos logit contributions
cases+0.008
ils+0.007
nuts+0.007
 Sale+0.007
 invitations+0.007
Neg logit contributions
 sur-0.005
avis-0.005
 fro-0.005
 polar-0.005
 clim-0.005
Token with
Token ID351
Feature activation-3.85e-01

Pos logit contributions
cases+0.089
 Sale+0.084
ports+0.082
ils+0.081
 invitations+0.080
Neg logit contributions
 sur-0.065
 rely-0.063
avis-0.062
 clim-0.061
 fro-0.061
Token spec
Token ID1020
Feature activation+9.57e+00

Pos logit contributions
cases+0.011
ils+0.010
nuts+0.009
driver+0.009
 Sale+0.009
Neg logit contributions
avis-0.013
please-0.013
 polar-0.013
 sur-0.013
 fro-0.012
Tokenks
Token ID591
Feature activation-1.73e+00

Pos logit contributions
 sur+0.324
avis+0.319
 rely+0.310
please+0.298
 listener+0.298
Neg logit contributions
cases-0.283
nuts-0.263
ils-0.262
 Sale-0.253
 invitations-0.247
Token of
Token ID286
Feature activation+2.28e+00

Pos logit contributions
 Sale+0.069
cases+0.065
ports+0.064
ils+0.063
nuts+0.062
Neg logit contributions
 sur-0.036
 bra-0.035
 clim-0.033
 polar-0.033
 bas-0.032
Token sugar
Token ID7543
Feature activation+1.47e+00

Pos logit contributions
avis+0.073
 sur+0.073
 polar+0.070
 rely+0.070
 fro+0.069
Neg logit contributions
cases-0.074
ils-0.067
nuts-0.063
driver-0.060
 Sale-0.060
Token on
Token ID319
Feature activation-3.79e+00

Pos logit contributions
please+0.030
 rely+0.029
avis+0.029
 day+0.029
 bra+0.028
Neg logit contributions
 Sale-0.057
 invitations-0.055
cases-0.055
ports-0.054
ils-0.052
Token the
Token ID262
Feature activation+2.43e-01

Pos logit contributions
cases+0.149
ils+0.138
 invitations+0.132
nuts+0.129
ports+0.121
Neg logit contributions
avis-0.090
 But-0.089
 Sk-0.089
 ask-0.088
 Butter-0.088
Token high
Token ID1029
Feature activation-1.63e+00

Pos logit contributions
 sur+0.043
avis+0.042
 rely+0.041
please+0.039
 listener+0.039
Neg logit contributions
cases-0.057
ils-0.054
nuts-0.053
 Sale-0.052
 invitations-0.051
Token they
Token ID484
Feature activation-2.61e+00

Pos logit contributions
cases+0.056
 Sale+0.056
 invitations+0.054
ils+0.053
nuts+0.053
Neg logit contributions
 sur-0.046
avis-0.043
 rely-0.042
 clim-0.041
 bas-0.041
Token were
Token ID547
Feature activation-1.16e+00

Pos logit contributions
cases+0.103
ils+0.099
nuts+0.098
 Sale+0.096
 invitations+0.095
Neg logit contributions
 sur-0.062
avis-0.060
 rely-0.058
please-0.055
 listener-0.054
Token just
Token ID655
Feature activation+3.14e+00

Pos logit contributions
cases+0.041
ils+0.039
nuts+0.038
 Sale+0.038
 invitations+0.037
Neg logit contributions
 sur-0.033
avis-0.032
 rely-0.031
please-0.030
 listener-0.030
Token tiny
Token ID7009
Feature activation+3.77e+00

Pos logit contributions
avis+0.094
 sur+0.092
 rely+0.088
please+0.086
 listener+0.085
Neg logit contributions
cases-0.101
nuts-0.098
ils-0.097
 invitations-0.094
 Sale-0.090
Token spec
Token ID1020
Feature activation+9.51e+00

Pos logit contributions
 sur+0.111
avis+0.107
 rely+0.104
 listener+0.101
 fro+0.099
Neg logit contributions
cases-0.128
ils-0.121
 Sale-0.119
nuts-0.119
 invitations-0.114
Tokenks
Token ID591
Feature activation-1.65e+00

Pos logit contributions
 sur+0.352
avis+0.349
 rely+0.341
please+0.329
 listener+0.326
Neg logit contributions
cases-0.244
ils-0.234
nuts-0.227
 Sale-0.213
 invitations-0.213
Token in
Token ID287
Feature activation+1.25e-01

Pos logit contributions
 Sale+0.069
cases+0.066
ports+0.065
ils+0.063
 files+0.063
Neg logit contributions
 sur-0.034
 bra-0.033
 clim-0.032
 fro-0.031
 bas-0.030
Token the
Token ID262
Feature activation+4.61e-01

Pos logit contributions
 sur+0.003
avis+0.003
 matter+0.003
 rely+0.003
 polar+0.003
Neg logit contributions
cases-0.005
ils-0.004
nuts-0.004
driver-0.004
 Sale-0.004
Token sky
Token ID6766
Feature activation+1.31e+00

Pos logit contributions
 sur+0.010
avis+0.010
 rely+0.010
please+0.009
 listener+0.009
Neg logit contributions
cases-0.019
ils-0.018
nuts-0.018
 Sale-0.018
 invitations-0.017
Token.
Token ID13
Feature activation-3.35e-01

Pos logit contributions
 sur+0.039
avis+0.038
 rely+0.037
please+0.035
 listener+0.035
Neg logit contributions
cases-0.043
ils-0.041
nuts-0.041
 Sale-0.040
 invitations-0.039
Token the
Token ID262
Feature activation+5.58e-01

Pos logit contributions
cases+0.059
ils+0.056
nuts+0.055
 Sale+0.054
 invitations+0.053
Neg logit contributions
 sur-0.046
avis-0.046
 rely-0.044
 fro-0.042
 listener-0.042
Token leaf
Token ID12835
Feature activation+2.86e+00

Pos logit contributions
 sur+0.017
avis+0.017
please+0.016
 listener+0.015
 rely+0.015
Neg logit contributions
nuts-0.018
cases-0.018
ils-0.017
 Sale-0.017
 puzzles-0.017
Token there
Token ID612
Feature activation+6.00e-01

Pos logit contributions
 sur+0.076
avis+0.074
 rely+0.072
 fro+0.068
please+0.068
Neg logit contributions
cases-0.104
ils-0.099
nuts-0.098
 Sale-0.096
 invitations-0.094
Token were
Token ID547
Feature activation-1.05e+00

Pos logit contributions
 sur+0.022
avis+0.022
 rely+0.021
please+0.020
 listener+0.020
Neg logit contributions
cases-0.016
ils-0.015
nuts-0.015
 Sale-0.014
es-0.014
Token little
Token ID1310
Feature activation+4.49e+00

Pos logit contributions
cases+0.029
ils+0.028
driver+0.026
 Sale+0.025
fs+0.024
Neg logit contributions
 Arctic-0.035
 polar-0.035
 crit-0.034
avis-0.034
 fro-0.034
Token spec
Token ID1020
Feature activation+9.50e+00

Pos logit contributions
 sur+0.145
avis+0.144
 rely+0.140
please+0.135
 listener+0.133
Neg logit contributions
cases-0.136
ils-0.129
nuts-0.129
 Sale-0.123
 invitations-0.122
Tokenks
Token ID591
Feature activation-1.57e+00

Pos logit contributions
 sur+0.321
avis+0.315
 rely+0.307
 listener+0.297
please+0.295
Neg logit contributions
cases-0.282
nuts-0.260
ils-0.257
 Sale-0.256
 invitations-0.246
Token of
Token ID286
Feature activation+2.55e+00

Pos logit contributions
 Sale+0.064
ports+0.061
cases+0.059
ils+0.057
nuts+0.056
Neg logit contributions
 bra-0.035
 sur-0.035
 clim-0.032
 fro-0.032
 polar-0.031
Token gold
Token ID3869
Feature activation+1.13e+00

Pos logit contributions
 sur+0.067
 pleasure+0.063
avis+0.063
 polar+0.062
 rely+0.062
Neg logit contributions
cases-0.104
ils-0.093
driver-0.089
nuts-0.088
 Sale-0.086
Token!
Token ID0
Feature activation-1.07e+00

Pos logit contributions
 sur+0.030
 fro+0.028
 clim+0.028
 polar+0.028
 rely+0.028
Neg logit contributions
 Sale-0.040
cases-0.040
ils-0.037
ports-0.037
 invitations-0.036
Token She
Token ID1375
Feature activation-1.32e-01

Pos logit contributions
cases+0.042
ils+0.041
nuts+0.041
 invitations+0.039
 Sale+0.039
Neg logit contributions
 sur-0.024
avis-0.024
 rely-0.023
 listener-0.022
please-0.022
Token with
Token ID351
Feature activation-2.95e-01

Pos logit contributions
cases+0.002
ils+0.002
nuts+0.002
 Sale+0.002
 invitations+0.002
Neg logit contributions
 sur-0.002
avis-0.002
 rely-0.002
please-0.002
 listener-0.002
Token lots
Token ID6041
Feature activation-5.19e+00

Pos logit contributions
cases+0.009
ils+0.008
nuts+0.008
 Sale+0.008
driver+0.007
Neg logit contributions
 sur-0.010
avis-0.010
 polar-0.010
 rely-0.010
 fro-0.009
Token of
Token ID286
Feature activation+2.31e+00

Pos logit contributions
cases+0.191
ils+0.182
nuts+0.180
 Sale+0.175
 invitations+0.172
Neg logit contributions
 sur-0.137
avis-0.135
 rely-0.130
please-0.124
 listener-0.123
Token food
Token ID2057
Feature activation-4.58e-01

Pos logit contributions
 sur+0.071
avis+0.070
 rely+0.068
please+0.065
 listener+0.064
Neg logit contributions
cases-0.075
ils-0.072
nuts-0.070
 invitations-0.069
 Sale-0.068
Token.
Token ID13
Feature activation-4.58e-01

Pos logit contributions
cases+0.017
 Sale+0.017
ils+0.017
ports+0.016
nuts+0.016
Neg logit contributions
 sur-0.011
 rely-0.010
 polar-0.010
 bas-0.010
avis-0.010
Token Them
Token ID26460
Feature activation+7.57e+00

Pos logit contributions
cases+0.016
nuts+0.015
 Sale+0.015
ils+0.015
 invitations+0.014
Neg logit contributions
 sur-0.013
avis-0.013
 rely-0.013
 listener-0.012
please-0.012
Tokenouse
Token ID1076
Feature activation-8.93e-01

Pos logit contributions
 sur+0.179
avis+0.168
 rely+0.165
 listener+0.161
 fro+0.158
Neg logit contributions
ils-0.291
cases-0.288
nuts-0.284
 Sale-0.277
 invitations-0.275
Token's
Token ID338
Feature activation+1.88e+00

Pos logit contributions
 invitations+0.042
 Sale+0.042
cases+0.042
ils+0.041
nuts+0.041
Neg logit contributions
 sur-0.011
 polar-0.011
 fro-0.011
 bas-0.010
avis-0.010
Token mouth
Token ID5422
Feature activation-2.14e-01

Pos logit contributions
 sur+0.054
avis+0.053
 rely+0.049
please+0.049
 listener+0.047
Neg logit contributions
cases-0.066
nuts-0.064
ils-0.063
 invitations-0.061
 Sale-0.061
Token watered
Token ID48024
Feature activation-2.48e+00

Pos logit contributions
ils+0.008
cases+0.008
 Sale+0.008
 invitations+0.008
ports+0.008
Neg logit contributions
 polar-0.005
 Arctic-0.005
avis-0.005
 rely-0.004
 fro-0.004
Token,
Token ID11
Feature activation-1.18e+00

Pos logit contributions
cases+0.082
ils+0.078
nuts+0.077
 Sale+0.074
 invitations+0.073
Neg logit contributions
 sur-0.074
avis-0.073
 rely-0.071
please-0.068
 listener-0.068
Token away
Token ID1497
Feature activation-3.86e+00

Pos logit contributions
 sur+0.063
avis+0.061
 rely+0.059
 fro+0.059
 listener+0.057
Neg logit contributions
cases-0.059
ils-0.056
 Sale-0.054
nuts-0.054
 invitations-0.052
Token,
Token ID11
Feature activation-1.01e+00

Pos logit contributions
cases+0.138
ils+0.137
 Sale+0.136
nuts+0.129
ports+0.129
Neg logit contributions
 sur-0.106
 rely-0.097
 bas-0.095
avis-0.094
 fro-0.094
Token a
Token ID257
Feature activation-1.02e+00

Pos logit contributions
cases+0.035
nuts+0.032
ils+0.031
 invitations+0.030
 Sale+0.030
Neg logit contributions
avis-0.029
 sur-0.028
 rely-0.027
please-0.026
 listener-0.026
Token bright
Token ID6016
Feature activation+6.37e-01

Pos logit contributions
cases+0.032
nuts+0.031
 invitations+0.030
 Sale+0.030
 sandwiches+0.029
Neg logit contributions
avis-0.030
 rely-0.028
 sur-0.028
please-0.027
 listener-0.026
Token blue
Token ID4171
Feature activation+2.38e+00

Pos logit contributions
avis+0.018
please+0.017
 rely+0.017
 sur+0.017
 listener+0.016
Neg logit contributions
 invitations-0.020
cases-0.020
nuts-0.020
ils-0.020
 pages-0.019
Token spe
Token ID693
Feature activation+7.40e+00

Pos logit contributions
 sur+0.061
avis+0.060
 rely+0.058
please+0.055
 listener+0.055
Neg logit contributions
cases-0.088
ils-0.084
nuts-0.084
 Sale-0.081
 invitations-0.080
Tokenck
Token ID694
Feature activation+1.93e+00

Pos logit contributions
 sur+0.241
 Are+0.232
avis+0.225
 rely+0.222
ancer+0.219
Neg logit contributions
ils-0.226
nuts-0.215
cases-0.213
 Sale-0.199
 invitations-0.198
Token in
Token ID287
Feature activation+5.45e-02

Pos logit contributions
 sur+0.061
avis+0.058
 fro+0.056
 rely+0.056
 listener+0.056
Neg logit contributions
cases-0.061
 Sale-0.057
ils-0.055
nuts-0.055
 invitations-0.053
Token the
Token ID262
Feature activation+3.73e-01

Pos logit contributions
 sur+0.002
avis+0.002
 rely+0.002
 polar+0.002
 fro+0.002
Neg logit contributions
cases-0.002
ils-0.002
nuts-0.002
driver-0.002
 Sale-0.002
Token sky
Token ID6766
Feature activation+1.20e+00

Pos logit contributions
 sur+0.010
avis+0.010
 rely+0.009
please+0.009
 listener+0.009
Neg logit contributions
cases-0.013
nuts-0.013
ils-0.013
 invitations-0.013
 Sale-0.013
Token.
Token ID13
Feature activation-6.25e-03

Pos logit contributions
 sur+0.040
avis+0.040
 rely+0.038
please+0.037
 listener+0.037
Neg logit contributions
cases-0.036
nuts-0.034
ils-0.034
 Sale-0.032
 invitations-0.031
Token play
Token ID711
Feature activation+1.93e+00

Pos logit contributions
cases+0.002
ils+0.002
nuts+0.002
 Sale+0.002
 invitations+0.002
Neg logit contributions
 sur-0.001
avis-0.001
 rely-0.001
please-0.001
 listener-0.001
Token chess
Token ID19780
Feature activation-6.13e+00

Pos logit contributions
 sur+0.049
 fro+0.049
 rely+0.047
 place+0.047
 bra+0.044
Neg logit contributions
cases-0.076
 Sale-0.068
nuts-0.068
ils-0.067
ifying-0.065
Token.
Token ID13
Feature activation+3.04e-01

Pos logit contributions
cases+0.232
 Sale+0.226
nuts+0.215
ils+0.212
ports+0.208
Neg logit contributions
 sur-0.149
 rely-0.144
 day-0.137
 fro-0.133
 place-0.132
Token He
Token ID679
Feature activation-1.42e-01

Pos logit contributions
 rely+0.009
avis+0.009
 sur+0.009
please+0.008
 matter+0.008
Neg logit contributions
cases-0.010
 Sale-0.010
nuts-0.009
driver-0.009
ports-0.009
Token showed
Token ID3751
Feature activation-1.01e+00

Pos logit contributions
cases+0.006
ils+0.006
nuts+0.006
 Sale+0.005
 invitations+0.005
Neg logit contributions
 sur-0.003
avis-0.003
 rely-0.003
please-0.003
 listener-0.003
Token Mate
Token ID24787
Feature activation+7.36e+00

Pos logit contributions
cases+0.032
ils+0.031
 Sale+0.030
 invitations+0.030
nuts+0.030
Neg logit contributions
 rely-0.032
 sur-0.032
avis-0.031
please-0.030
 listener-0.030
Tokeno
Token ID78
Feature activation+2.71e+00

Pos logit contributions
 sur+0.132
 rely+0.128
avis+0.124
 beacon+0.116
 listener+0.110
Neg logit contributions
cases-0.331
nuts-0.318
ils-0.315
es-0.313
driver-0.304
Token how
Token ID703
Feature activation+3.47e+00

Pos logit contributions
 fro+0.068
 sur+0.066
 pap+0.063
avis+0.062
 day+0.061
Neg logit contributions
cases-0.098
 Sale-0.091
nuts-0.090
ils-0.090
ifying-0.089
Token to
Token ID284
Feature activation+2.85e-02

Pos logit contributions
 sur+0.102
avis+0.092
 listener+0.089
 rely+0.087
please+0.087
Neg logit contributions
cases-0.125
ils-0.120
nuts-0.118
 Sale-0.116
 invitations-0.112
Token move
Token ID1445
Feature activation+1.10e+00

Pos logit contributions
 sur+0.001
avis+0.001
 rely+0.001
please+0.001
 listener+0.001
Neg logit contributions
cases-0.001
ils-0.001
nuts-0.001
 Sale-0.001
 invitations-0.001
Token the
Token ID262
Feature activation+7.46e-01

Pos logit contributions
 sur+0.031
avis+0.029
 rely+0.029
 fro+0.028
 bra+0.028
Neg logit contributions
cases-0.039
ils-0.037
 Sale-0.036
nuts-0.036
 invitations-0.034
Token of
Token ID286
Feature activation+2.34e+00

Pos logit contributions
 sur+0.052
avis+0.051
 rely+0.049
 fro+0.047
 polar+0.047
Neg logit contributions
cases-0.067
ils-0.066
 Sale-0.063
nuts-0.063
 invitations-0.061
Token the
Token ID262
Feature activation+5.09e-01

Pos logit contributions
 sur+0.076
avis+0.075
 rely+0.072
please+0.070
 listener+0.070
Neg logit contributions
cases-0.071
ils-0.068
nuts-0.068
 Sale-0.065
 invitations-0.064
Token frog
Token ID21264
Feature activation+1.18e+00

Pos logit contributions
 sur+0.014
avis+0.014
 rely+0.013
please+0.013
 listener+0.013
Neg logit contributions
cases-0.018
ils-0.017
nuts-0.017
 Sale-0.017
 invitations-0.016
Token.
Token ID13
Feature activation+3.97e-01

Pos logit contributions
 sur+0.035
 rely+0.034
avis+0.034
 fro+0.032
 polar+0.032
Neg logit contributions
cases-0.039
ils-0.036
 Sale-0.036
nuts-0.036
 invitations-0.035
Token He
Token ID679
Feature activation-1.60e-01

Pos logit contributions
avis+0.012
 sur+0.011
 listener+0.011
please+0.011
 rely+0.011
Neg logit contributions
cases-0.013
 Sale-0.013
nuts-0.012
driver-0.011
es-0.011
Token spe
Token ID693
Feature activation+7.32e+00

Pos logit contributions
cases+0.006
ils+0.006
nuts+0.006
es+0.005
 Sale+0.005
Neg logit contributions
avis-0.004
 sur-0.004
 rely-0.004
 listener-0.004
please-0.004
Tokenared
Token ID1144
Feature activation-2.92e+00

Pos logit contributions
 sur+0.238
avis+0.227
 rely+0.225
 Are+0.221
 listener+0.218
Neg logit contributions
ils-0.222
nuts-0.216
cases-0.214
 Sale-0.203
 invitations-0.194
Token the
Token ID262
Feature activation+4.85e-01

Pos logit contributions
cases+0.100
ils+0.096
 Sale+0.096
nuts+0.094
 invitations+0.093
Neg logit contributions
 sur-0.083
avis-0.077
 rely-0.076
 fro-0.075
 bas-0.075
Token lazy
Token ID16931
Feature activation-1.68e+00

Pos logit contributions
 sur+0.013
avis+0.013
 rely+0.012
 fro+0.012
please+0.012
Neg logit contributions
cases-0.018
ils-0.017
nuts-0.016
 Sale-0.016
 invitations-0.016
Token frog
Token ID21264
Feature activation+1.20e+00

Pos logit contributions
cases+0.069
nuts+0.068
ils+0.067
 Sale+0.065
 invitations+0.064
Neg logit contributions
avis-0.035
 sur-0.034
 rely-0.033
please-0.031
 listener-0.030
Token and
Token ID290
Feature activation-6.40e-01

Pos logit contributions
 sur+0.035
avis+0.034
 rely+0.033
 fro+0.032
please+0.031
Neg logit contributions
cases-0.041
ils-0.039
nuts-0.038
 Sale-0.038
 invitations-0.037
Token baby
Token ID5156
Feature activation+2.17e+00

Pos logit contributions
 sur+0.029
avis+0.028
 rely+0.028
 polar+0.027
 listener+0.027
Neg logit contributions
cases-0.043
ils-0.041
nuts-0.041
 Sale-0.039
driver-0.038
Token thought
Token ID1807
Feature activation-3.17e+00

Pos logit contributions
 sur+0.054
avis+0.053
 rely+0.052
please+0.050
 listener+0.048
Neg logit contributions
cases-0.083
ils-0.079
nuts-0.078
es-0.075
 invitations-0.074
Token about
Token ID546
Feature activation-2.72e-01

Pos logit contributions
cases+0.106
nuts+0.102
ils+0.101
 Sale+0.099
 invitations+0.095
Neg logit contributions
 sur-0.093
avis-0.091
 rely-0.089
 listener-0.085
 fro-0.084
Token the
Token ID262
Feature activation+8.02e-01

Pos logit contributions
cases+0.008
ils+0.007
nuts+0.007
 Sale+0.007
 invitations+0.007
Neg logit contributions
 sur-0.009
avis-0.009
 rely-0.009
 fro-0.009
 listener-0.008
Token night
Token ID1755
Feature activation+1.37e+00

Pos logit contributions
 sur+0.021
avis+0.020
 rely+0.020
 listener+0.019
 polar+0.019
Neg logit contributions
cases-0.030
ils-0.029
nuts-0.029
 Sale-0.028
es-0.027
TokenÂ
Token ID5523
Feature activation+7.29e+00

Pos logit contributions
 sur+0.041
avis+0.041
 rely+0.039
please+0.039
 Kang+0.037
Neg logit contributions
cases-0.046
ils-0.043
nuts-0.042
 invitations-0.041
 puzzles-0.040
Token´
Token ID18265
Feature activation+1.36e+00

Pos logit contributions
 sur+0.195
avis+0.181
please+0.177
 rely+0.175
 Arctic+0.169
Neg logit contributions
cases-0.258
 invitations-0.252
nuts-0.246
ils-0.246
 Sale-0.243
Tokens
Token ID82
Feature activation+1.39e+00

Pos logit contributions
 sur+0.049
avis+0.048
 rely+0.047
 listener+0.045
please+0.045
Neg logit contributions
cases-0.037
ils-0.034
nuts-0.034
 Sale-0.032
 invitations-0.031
Token adventure
Token ID8855
Feature activation-4.07e-01

Pos logit contributions
 sur+0.038
avis+0.037
 rely+0.036
 listener+0.034
 fro+0.033
Neg logit contributions
cases-0.050
ils-0.048
nuts-0.047
 Sale-0.046
 invitations-0.045
Token as
Token ID355
Feature activation+4.54e+00

Pos logit contributions
cases+0.014
ils+0.013
nuts+0.013
 Sale+0.012
 invitations+0.012
Neg logit contributions
 sur-0.012
avis-0.012
 rely-0.011
please-0.011
 listener-0.011
Token she
Token ID673
Feature activation+2.23e-01

Pos logit contributions
 sur+0.166
avis+0.159
please+0.158
 rely+0.155
 Arctic+0.153
Neg logit contributions
 puzzles-0.111
cases-0.111
es-0.104
ils-0.103
nuts-0.100
Token was
Token ID373
Feature activation+8.37e-01

Pos logit contributions
cases+0.059
nuts+0.056
ils+0.056
 Sale+0.054
 invitations+0.053
Neg logit contributions
 sur-0.041
avis-0.039
 rely-0.038
 fro-0.037
 listener-0.036
Token ready
Token ID3492
Feature activation-6.56e-01

Pos logit contributions
 sur+0.018
avis+0.017
 rely+0.017
 polar+0.016
please+0.016
Neg logit contributions
cases-0.035
ils-0.034
nuts-0.033
 Sale-0.032
es-0.032
Token,
Token ID11
Feature activation-8.20e-01

Pos logit contributions
cases+0.025
ils+0.025
nuts+0.024
 Sale+0.023
 invitations+0.023
Neg logit contributions
 sur-0.016
avis-0.015
 rely-0.014
 fro-0.014
 listener-0.014
Token the
Token ID262
Feature activation+6.39e-01

Pos logit contributions
cases+0.030
ils+0.029
nuts+0.029
 Sale+0.028
driver+0.027
Neg logit contributions
 sur-0.021
avis-0.021
 rely-0.020
 fro-0.020
 polar-0.019
Token b
Token ID275
Feature activation+1.16e+00

Pos logit contributions
 sur+0.015
avis+0.015
 rely+0.014
please+0.014
 listener+0.013
Neg logit contributions
cases-0.025
ils-0.024
nuts-0.024
 invitations-0.023
 Sale-0.023
Tokenunn
Token ID20935
Feature activation+7.27e+00

Pos logit contributions
 sur+0.026
 rely+0.024
avis+0.023
please+0.022
 fro+0.022
Neg logit contributions
cases-0.049
ils-0.048
nuts-0.047
driver-0.044
es-0.044
Tokenies
Token ID444
Feature activation-2.65e-01

Pos logit contributions
 sur+0.196
avis+0.193
 rely+0.185
please+0.173
 fro+0.172
Neg logit contributions
cases-0.265
ils-0.257
nuts-0.250
 invitations-0.239
es-0.237
Token hopped
Token ID45230
Feature activation-3.59e-01

Pos logit contributions
cases+0.011
 Sale+0.011
nuts+0.011
ils+0.010
ados+0.010
Neg logit contributions
 sur-0.006
 rely-0.005
 bas-0.005
 ask-0.005
 fro-0.005
Token over
Token ID625
Feature activation-1.10e+00

Pos logit contributions
 Sale+0.013
cases+0.013
ils+0.013
 files+0.013
ports+0.012
Neg logit contributions
 sur-0.009
 fro-0.009
 morning-0.009
 day-0.009
 bra-0.009
Token to
Token ID284
Feature activation-8.18e-02

Pos logit contributions
cases+0.039
ils+0.036
nuts+0.035
 Sale+0.034
ports+0.033
Neg logit contributions
 sur-0.032
avis-0.031
 fro-0.030
 rely-0.030
 polar-0.029
Token the
Token ID262
Feature activation+4.44e-01

Pos logit contributions
cases+0.002
ils+0.002
nuts+0.002
 Sale+0.002
 invitations+0.002
Neg logit contributions
 sur-0.003
avis-0.003
 rely-0.002
 listener-0.002
please-0.002
Token The
Token ID383
Feature activation+9.79e-01

Pos logit contributions
 sur+0.005
 rely+0.005
avis+0.005
please+0.005
 listener+0.005
Neg logit contributions
cases-0.006
 Sale-0.006
nuts-0.005
ils-0.005
 invitations-0.005
Token eggs
Token ID9653
Feature activation+1.82e+00

Pos logit contributions
 sur+0.024
avis+0.023
 rely+0.022
 listener+0.021
please+0.021
Neg logit contributions
nuts-0.036
cases-0.036
ils-0.035
 invitations-0.035
 Sale-0.035
Token were
Token ID547
Feature activation-1.17e+00

Pos logit contributions
 sur+0.051
avis+0.050
 rely+0.049
 fro+0.046
please+0.046
Neg logit contributions
cases-0.063
ils-0.060
nuts-0.059
 Sale-0.059
 invitations-0.058
Token blue
Token ID4171
Feature activation+2.66e+00

Pos logit contributions
cases+0.045
ils+0.042
 Sale+0.040
nuts+0.040
driver+0.039
Neg logit contributions
 rely-0.029
 sur-0.029
avis-0.029
 polar-0.028
please-0.028
Token and
Token ID290
Feature activation-4.68e-01

Pos logit contributions
 fro+0.051
 sur+0.050
 polar+0.049
 Arctic+0.049
 clim+0.049
Neg logit contributions
cases-0.109
ils-0.108
 Sale-0.108
ports-0.101
 invitations-0.100
Token spe
Token ID693
Feature activation+7.26e+00

Pos logit contributions
cases+0.019
ils+0.018
nuts+0.017
 invitations+0.017
 Sale+0.017
Neg logit contributions
avis-0.010
 rely-0.010
 sur-0.010
 polar-0.010
please-0.010
Tokenck
Token ID694
Feature activation+1.91e+00

Pos logit contributions
 sur+0.229
avis+0.217
 rely+0.213
 listener+0.205
 Are+0.202
Neg logit contributions
ils-0.230
cases-0.228
nuts-0.223
 Sale-0.214
 invitations-0.206
Tokenled
Token ID992
Feature activation-2.34e+00

Pos logit contributions
 fro+0.062
 sur+0.061
 bas+0.061
 crit+0.060
 polar+0.060
Neg logit contributions
cases-0.053
 Sale-0.052
 invitations-0.048
nuts-0.046
 puzzles-0.046
Token.
Token ID13
Feature activation+9.49e-02

Pos logit contributions
cases+0.089
 Sale+0.085
ils+0.081
driver+0.078
ports+0.077
Neg logit contributions
 sur-0.060
 bra-0.059
 fro-0.058
 bas-0.057
 polar-0.056
Token They
Token ID1119
Feature activation-1.66e+00

Pos logit contributions
 sur+0.003
avis+0.003
 rely+0.003
please+0.002
 listener+0.002
Neg logit contributions
cases-0.003
nuts-0.003
ils-0.003
 Sale-0.003
 invitations-0.003
Token looked
Token ID3114
Feature activation-1.62e+00

Pos logit contributions
cases+0.066
ils+0.062
 Sale+0.062
nuts+0.062
 invitations+0.061
Neg logit contributions
 sur-0.039
 rely-0.037
avis-0.036
 fro-0.035
 bas-0.035
Token.
Token ID13
Feature activation+7.36e-02

Pos logit contributions
 sur+0.170
avis+0.167
 rely+0.163
 listener+0.155
please+0.154
Neg logit contributions
cases-0.187
ils-0.178
nuts-0.175
 Sale-0.170
 invitations-0.167
Token Grand
Token ID5675
Feature activation+1.10e+00

Pos logit contributions
 rely+0.002
 matter+0.002
please+0.002
 sur+0.002
 listener+0.002
Neg logit contributions
cases-0.003
 Sale-0.002
ports-0.002
nuts-0.002
driver-0.002
Tokendad
Token ID47984
Feature activation+7.03e-01

Pos logit contributions
 sur+0.031
 Kang+0.030
 bra+0.030
 rely+0.030
 listener+0.030
Neg logit contributions
ils-0.036
 Sale-0.035
nuts-0.035
cases-0.034
 invitations-0.033
Token also
Token ID635
Feature activation+3.12e+00

Pos logit contributions
 sur+0.016
avis+0.015
 rely+0.015
please+0.014
 listener+0.014
Neg logit contributions
cases-0.029
ils-0.027
nuts-0.027
 Sale-0.026
 invitations-0.026
Token showed
Token ID3751
Feature activation-1.05e+00

Pos logit contributions
 sur+0.070
avis+0.067
 rely+0.065
please+0.061
 listener+0.060
Neg logit contributions
cases-0.129
ils-0.123
nuts-0.123
 Sale-0.119
 invitations-0.117
Token Mate
Token ID24787
Feature activation+7.24e+00

Pos logit contributions
cases+0.037
ils+0.036
nuts+0.034
 invitations+0.034
 Sale+0.034
Neg logit contributions
 rely-0.030
 sur-0.030
avis-0.028
please-0.028
 listener-0.027
Tokeno
Token ID78
Feature activation+2.55e+00

Pos logit contributions
 sur+0.133
 rely+0.128
avis+0.124
 beacon+0.117
 crit+0.112
Neg logit contributions
cases-0.318
nuts-0.305
ils-0.305
es-0.304
driver-0.293
Token how
Token ID703
Feature activation+3.47e+00

Pos logit contributions
 sur+0.068
 fro+0.067
 rely+0.064
avis+0.064
 pap+0.063
Neg logit contributions
cases-0.088
 Sale-0.082
ils-0.081
nuts-0.079
 invitations-0.077
Token to
Token ID284
Feature activation-2.11e-04

Pos logit contributions
 sur+0.098
avis+0.086
 listener+0.085
 rely+0.084
please+0.082
Neg logit contributions
cases-0.129
ils-0.123
 Sale-0.120
nuts-0.120
 invitations-0.116
Token make
Token ID787
Feature activation+2.27e+00

Pos logit contributions
cases+0.000
nuts+0.000
ils+0.000
 Sale+0.000
 invitations+0.000
Neg logit contributions
avis-0.000
 sur-0.000
please-0.000
 rely-0.000
 listener-0.000
Token careful
Token ID8161
Feature activation+1.81e+00

Pos logit contributions
 sur+0.073
avis+0.072
 rely+0.070
 fro+0.067
 bra+0.065
Neg logit contributions
cases-0.071
ils-0.067
nuts-0.066
 Sale-0.065
driver-0.060
TokenSo
Token ID2396
Feature activation+1.53e+00

Pos logit contributions
 sur+0.058
avis+0.058
 rely+0.057
 listener+0.054
please+0.054
Neg logit contributions
cases-0.046
ils-0.044
nuts-0.044
 Sale-0.042
 invitations-0.041
Token,
Token ID11
Feature activation-8.69e-01

Pos logit contributions
 sur+0.047
avis+0.046
 rely+0.045
please+0.043
 listener+0.043
Neg logit contributions
cases-0.049
ils-0.047
nuts-0.046
 Sale-0.045
 invitations-0.044
Token Grand
Token ID5675
Feature activation+8.76e-01

Pos logit contributions
cases+0.028
nuts+0.027
ils+0.027
 invitations+0.026
 Sale+0.026
Neg logit contributions
avis-0.025
 rely-0.025
 sur-0.025
please-0.024
 listener-0.024
Tokendad
Token ID47984
Feature activation+2.26e-01

Pos logit contributions
 sur+0.026
 rely+0.024
avis+0.024
 listener+0.024
please+0.024
Neg logit contributions
cases-0.029
ils-0.029
nuts-0.029
 Sale-0.028
 invitations-0.027
Token taught
Token ID7817
Feature activation-1.76e+00

Pos logit contributions
 sur+0.006
avis+0.006
 rely+0.005
please+0.005
 listener+0.005
Neg logit contributions
cases-0.009
ils-0.008
nuts-0.008
 Sale-0.008
 invitations-0.008
Token Mate
Token ID24787
Feature activation+7.22e+00

Pos logit contributions
cases+0.062
ils+0.060
nuts+0.059
 Sale+0.059
 invitations+0.058
Neg logit contributions
 sur-0.048
 rely-0.047
avis-0.045
 listener-0.044
please-0.044
Tokeno
Token ID78
Feature activation+2.69e+00

Pos logit contributions
 sur+0.127
avis+0.125
 rely+0.125
 beacon+0.116
 Kang+0.112
Neg logit contributions
cases-0.324
ils-0.310
nuts-0.309
es-0.303
driver-0.295
Token how
Token ID703
Feature activation+3.35e+00

Pos logit contributions
 fro+0.075
 rely+0.074
 sur+0.073
avis+0.073
 ask+0.072
Neg logit contributions
cases-0.089
 Sale-0.081
nuts-0.081
ils-0.080
ifying-0.079
Token to
Token ID284
Feature activation-4.93e-02

Pos logit contributions
 sur+0.099
avis+0.095
 rely+0.092
please+0.090
 listener+0.090
Neg logit contributions
cases-0.114
ils-0.111
nuts-0.109
 Sale-0.104
 invitations-0.102
Token play
Token ID711
Feature activation+1.93e+00

Pos logit contributions
cases+0.002
ils+0.002
nuts+0.002
 Sale+0.002
 invitations+0.002
Neg logit contributions
 sur-0.001
avis-0.001
 rely-0.001
please-0.001
 listener-0.001
Token chess
Token ID19780
Feature activation-6.13e+00

Pos logit contributions
 sur+0.049
 fro+0.049
 rely+0.047
 place+0.047
 bra+0.044
Neg logit contributions
cases-0.076
 Sale-0.068
nuts-0.068
ils-0.067
ifying-0.065
Token
Token ID198
Feature activation+1.09e+00

Pos logit contributions
 sur+0.027
 listener+0.027
 matter+0.026
 rely+0.026
avis+0.026
Neg logit contributions
cases-0.036
 Sale-0.035
es-0.034
ils-0.032
ades-0.032
TokenAfter
Token ID3260
Feature activation-1.65e+00

Pos logit contributions
 sur+0.034
avis+0.033
 rely+0.033
 listener+0.031
please+0.031
Neg logit contributions
cases-0.034
ils-0.033
nuts-0.032
 Sale-0.032
 invitations-0.030
Token a
Token ID257
Feature activation-8.67e-01

Pos logit contributions
cases+0.054
ils+0.051
nuts+0.050
 invitations+0.046
 Sale+0.046
Neg logit contributions
 sur-0.053
avis-0.051
 rely-0.050
 polar-0.049
 fro-0.048
Token while
Token ID981
Feature activation+2.09e+00

Pos logit contributions
cases+0.023
 Sale+0.022
nuts+0.021
 puzzles+0.021
ils+0.021
Neg logit contributions
avis-0.031
 rely-0.029
 Arctic-0.029
 sur-0.029
 bas-0.028
Token,
Token ID11
Feature activation-1.18e+00

Pos logit contributions
 rely+0.076
please+0.076
avis+0.075
 sur+0.074
 listener+0.071
Neg logit contributions
cases-0.051
 Sale-0.047
nuts-0.046
es-0.046
ils-0.044
Token Mate
Token ID24787
Feature activation+7.21e+00

Pos logit contributions
cases+0.039
ils+0.036
nuts+0.035
 Sale+0.035
 invitations+0.034
Neg logit contributions
 rely-0.035
 sur-0.035
avis-0.035
please-0.033
 listener-0.033
Tokeno
Token ID78
Feature activation+2.54e+00

Pos logit contributions
 rely+0.171
 crit+0.167
 regret+0.166
 beacon+0.165
avis+0.160
Neg logit contributions
cases-0.272
es-0.262
driver-0.248
ports-0.243
nuts-0.240
Token was
Token ID373
Feature activation+8.29e-01

Pos logit contributions
 sur+0.061
avis+0.059
 rely+0.056
please+0.053
 fro+0.053
Neg logit contributions
cases-0.099
ils-0.096
nuts-0.095
 Sale-0.093
 invitations-0.092
Token getting
Token ID1972
Feature activation+1.31e+00

Pos logit contributions
 sur+0.024
avis+0.024
 rely+0.023
please+0.022
 listener+0.022
Neg logit contributions
cases-0.028
ils-0.027
nuts-0.027
 Sale-0.026
 invitations-0.025
Token very
Token ID845
Feature activation-7.34e-01

Pos logit contributions
 sur+0.033
avis+0.032
 rely+0.031
 listener+0.029
please+0.029
Neg logit contributions
cases-0.050
ils-0.047
nuts-0.047
 Sale-0.046
 invitations-0.045
Token good
Token ID922
Feature activation-1.53e+00

Pos logit contributions
cases+0.030
ils+0.029
nuts+0.028
 Sale+0.028
 invitations+0.027
Neg logit contributions
 sur-0.017
avis-0.016
 rely-0.016
please-0.015
 listener-0.015
Token on
Token ID319
Feature activation-2.58e+00

Pos logit contributions
 sur+0.073
avis+0.071
 rely+0.070
 listener+0.067
please+0.067
Neg logit contributions
cases-0.069
ils-0.066
nuts-0.064
 invitations-0.063
 Sale-0.062
Token top
Token ID1353
Feature activation-2.63e+00

Pos logit contributions
cases+0.102
ils+0.098
nuts+0.095
ports+0.091
 Sale+0.088
Neg logit contributions
 place-0.062
 rely-0.062
 day-0.061
 ask-0.058
 sur-0.058
Token.
Token ID13
Feature activation+9.13e-01

Pos logit contributions
cases+0.087
ils+0.083
nuts+0.082
 Sale+0.079
 invitations+0.078
Neg logit contributions
 sur-0.078
avis-0.077
 rely-0.074
please-0.071
 listener-0.071
Token The
Token ID383
Feature activation+8.92e-01

Pos logit contributions
 sur+0.024
avis+0.023
 rely+0.022
 listener+0.022
please+0.021
Neg logit contributions
cases-0.034
nuts-0.032
 Sale-0.031
ils-0.031
 invitations-0.030
Token b
Token ID275
Feature activation+1.30e+00

Pos logit contributions
 sur+0.023
avis+0.023
 rely+0.022
please+0.021
 listener+0.021
Neg logit contributions
cases-0.033
ils-0.032
nuts-0.031
 Sale-0.030
 invitations-0.030
Tokenunn
Token ID20935
Feature activation+7.19e+00

Pos logit contributions
 sur+0.031
avis+0.029
 rely+0.029
please+0.027
 Are+0.027
Neg logit contributions
ils-0.051
cases-0.050
nuts-0.049
 Sale-0.046
driver-0.046
Tokenies
Token ID444
Feature activation-3.14e-01

Pos logit contributions
 sur+0.177
avis+0.174
 rely+0.166
please+0.157
 listener+0.156
Neg logit contributions
cases-0.276
ils-0.267
nuts-0.263
 Sale-0.253
 invitations-0.251
Token wanted
Token ID2227
Feature activation+3.87e-02

Pos logit contributions
cases+0.011
 Sale+0.011
nuts+0.010
ils+0.010
driver+0.010
Neg logit contributions
 sur-0.009
 rely-0.008
 ask-0.008
 fro-0.008
 bas-0.008
Token to
Token ID284
Feature activation-8.57e-02

Pos logit contributions
 sur+0.001
avis+0.001
 rely+0.001
please+0.001
 fro+0.001
Neg logit contributions
cases-0.001
ils-0.001
nuts-0.001
 Sale-0.001
 invitations-0.001
Token have
Token ID423
Feature activation-1.70e+00

Pos logit contributions
cases+0.004
ils+0.003
nuts+0.003
 Sale+0.003
 invitations+0.003
Neg logit contributions
 sur-0.002
avis-0.002
 rely-0.002
please-0.002
 listener-0.002
Token a
Token ID257
Feature activation-1.08e+00

Pos logit contributions
cases+0.063
ils+0.056
driver+0.056
nuts+0.055
es+0.052
Neg logit contributions
 rely-0.045
 sur-0.044
 fro-0.044
please-0.044
 pleasure-0.043
Token watched
Token ID7342
Feature activation+9.96e-01

Pos logit contributions
 rely+0.007
avis+0.007
 pleasure+0.007
 sur+0.007
 listener+0.007
Neg logit contributions
ils-0.008
cases-0.008
nuts-0.007
es-0.007
 sandwiches-0.007
Token,
Token ID11
Feature activation-1.17e+00

Pos logit contributions
 sur+0.026
avis+0.026
 fro+0.025
 rely+0.025
 ask+0.024
Neg logit contributions
cases-0.038
ils-0.036
nuts-0.035
 Sale-0.034
ports-0.033
Token she
Token ID673
Feature activation+2.02e-01

Pos logit contributions
cases+0.049
ils+0.048
nuts+0.047
 Sale+0.046
 invitations+0.044
Neg logit contributions
 sur-0.026
avis-0.025
 fro-0.024
 rely-0.023
 crit-0.023
Token realized
Token ID6939
Feature activation-4.57e+00

Pos logit contributions
avis+0.005
 rely+0.005
 sur+0.005
 pleasure+0.005
 listener+0.005
Neg logit contributions
cases-0.007
ils-0.007
nuts-0.007
es-0.007
 invitations-0.006
Token that
Token ID326
Feature activation+2.19e+00

Pos logit contributions
cases+0.155
ils+0.149
nuts+0.147
 Sale+0.145
 invitations+0.137
Neg logit contributions
avis-0.134
 sur-0.134
 rely-0.126
 bra-0.124
 bas-0.122
TokenÂ
Token ID5523
Feature activation+7.18e+00

Pos logit contributions
 sur+0.073
avis+0.072
 rely+0.070
please+0.068
 listener+0.067
Neg logit contributions
cases-0.065
ils-0.062
nuts-0.061
 Sale-0.059
 invitations-0.058
Token 
Token ID1849
Feature activation+2.27e+00

Pos logit contributions
 sur+0.258
avis+0.248
 rely+0.243
please+0.236
 listener+0.233
Neg logit contributions
cases-0.193
ils-0.188
nuts-0.183
 Sale-0.180
 invitations-0.179
Tokenthe
Token ID1169
Feature activation+6.47e-01

Pos logit contributions
 sur+0.057
avis+0.053
 rely+0.052
 polar+0.050
please+0.049
Neg logit contributions
cases-0.086
ils-0.083
nuts-0.083
 Sale-0.081
 invitations-0.079
Token ground
Token ID2323
Feature activation+1.58e+00

Pos logit contributions
avis+0.017
 sur+0.017
 rely+0.016
please+0.015
 polar+0.015
Neg logit contributions
cases-0.024
ils-0.023
nuts-0.023
 invitations-0.022
 Sale-0.022
Token was
Token ID373
Feature activation+7.96e-01

Pos logit contributions
 sur+0.020
avis+0.019
 rely+0.019
 bas+0.018
 fro+0.017
Neg logit contributions
cases-0.075
ils-0.073
 invitations-0.073
nuts-0.072
 Sale-0.072
Token alive
Token ID6776
Feature activation-1.32e+00

Pos logit contributions
 rely+0.021
 sur+0.021
avis+0.021
 bra+0.019
 polar+0.019
Neg logit contributions
cases-0.030
ils-0.028
nuts-0.027
 Sale-0.026
ports-0.026
Token floor
Token ID4314
Feature activation-8.18e-01

Pos logit contributions
 sur+0.020
avis+0.019
 rely+0.018
 place+0.018
 polar+0.018
Neg logit contributions
cases-0.022
ils-0.021
nuts-0.020
 Sale-0.020
es-0.019
Token.
Token ID13
Feature activation-6.89e-02

Pos logit contributions
ils+0.032
cases+0.032
 Sale+0.031
nuts+0.030
ports+0.030
Neg logit contributions
 sur-0.019
avis-0.018
 rely-0.018
 clim-0.017
 listener-0.017
Token
Token ID198
Feature activation+5.76e-01

Pos logit contributions
cases+0.002
nuts+0.002
driver+0.002
 invitations+0.002
 Sale+0.002
Neg logit contributions
 rely-0.002
avis-0.002
 sur-0.002
 bas-0.002
please-0.002
Token
Token ID198
Feature activation+6.11e-01

Pos logit contributions
 sur+0.019
 rely+0.019
avis+0.018
 clim+0.018
 listener+0.018
Neg logit contributions
cases-0.016
Their-0.016
es-0.016
ork-0.016
 Sale-0.015
Token"
Token ID1
Feature activation-2.87e+00

Pos logit contributions
 sur+0.019
avis+0.019
 rely+0.018
please+0.018
 listener+0.017
Neg logit contributions
cases-0.020
ils-0.019
nuts-0.019
 Sale-0.018
 invitations-0.018
TokenÂ
Token ID5523
Feature activation+7.18e+00

Pos logit contributions
cases+0.082
nuts+0.081
 Sale+0.078
ports+0.076
 puzzles+0.075
Neg logit contributions
avis-0.090
 polar-0.089
 rely-0.089
 Arctic-0.088
heart-0.083
Token
Token ID126
Feature activation-1.41e+00

Pos logit contributions
 sur+0.216
avis+0.214
 rely+0.204
 listener+0.196
 fro+0.195
Neg logit contributions
cases-0.243
ils-0.228
nuts-0.227
 Sale-0.212
 invitations-0.209
Token
Token ID94
Feature activation-1.24e+00

Pos logit contributions
cases+0.047
nuts+0.045
ils+0.045
 invitations+0.043
 Sale+0.043
Neg logit contributions
 sur-0.042
avis-0.041
 rely-0.040
please-0.038
 listener-0.037
TokenM
Token ID44
Feature activation-1.16e+00

Pos logit contributions
cases+0.045
nuts+0.043
ils+0.043
 invitations+0.042
 sandwiches+0.041
Neg logit contributions
 sur-0.034
avis-0.032
please-0.030
 rely-0.029
 listener-0.029
Tokenira
Token ID8704
Feature activation+2.88e+00

Pos logit contributions
cases+0.035
ils+0.032
nuts+0.032
 Sale+0.032
 invitations+0.030
Neg logit contributions
 sur-0.037
 rely-0.037
 bas-0.036
avis-0.036
please-0.036
Token mi
Token ID21504
Feature activation+2.66e+00

Pos logit contributions
avis+0.100
 sur+0.099
please+0.095
 listener+0.093
 Arctic+0.093
Neg logit contributions
cases-0.079
ils-0.076
es-0.074
nuts-0.073
driver-0.070
Token soon
Token ID2582
Feature activation+9.73e-01

Pos logit contributions
 sur+0.033
avis+0.032
 rely+0.030
 listener+0.029
please+0.029
Neg logit contributions
cases-0.032
nuts-0.031
ils-0.030
 invitations-0.029
 Sale-0.029
Token nothing
Token ID2147
Feature activation-2.23e+00

Pos logit contributions
 sur+0.029
avis+0.028
 rely+0.028
 listener+0.026
please+0.026
Neg logit contributions
cases-0.032
ils-0.031
nuts-0.030
 Sale-0.030
 invitations-0.029
Token but
Token ID475
Feature activation+2.12e+00

Pos logit contributions
cases+0.066
ils+0.066
nuts+0.061
ls+0.058
 invitations+0.057
Neg logit contributions
avis-0.075
 sur-0.071
 Arctic-0.069
 rely-0.069
 Kang-0.068
Token a
Token ID257
Feature activation-7.93e-01

Pos logit contributions
avis+0.068
 sur+0.068
 rely+0.065
please+0.064
 polar+0.062
Neg logit contributions
cases-0.065
ils-0.062
nuts-0.061
 invitations-0.059
 Sale-0.058
Token small
Token ID1402
Feature activation+1.72e+00

Pos logit contributions
 invitations+0.023
 sandwiches+0.022
nuts+0.022
 puzzles+0.022
 Sale+0.021
Neg logit contributions
avis-0.024
 rely-0.023
 sur-0.022
ancer-0.022
please-0.022
Token spe
Token ID693
Feature activation+7.16e+00

Pos logit contributions
avis+0.045
please+0.042
 sur+0.042
 rely+0.041
 Arctic+0.039
Neg logit contributions
 invitations-0.057
nuts-0.057
ils-0.057
cases-0.056
 Sale-0.055
Tokenck
Token ID694
Feature activation+2.17e+00

Pos logit contributions
 sur+0.253
 Are+0.247
avis+0.239
ancer+0.238
 rely+0.237
Neg logit contributions
ils-0.196
nuts-0.176
cases-0.172
 Sale-0.170
ork-0.160
Token in
Token ID287
Feature activation+2.12e-01

Pos logit contributions
 sur+0.058
avis+0.056
 rely+0.054
 listener+0.052
 fro+0.051
Neg logit contributions
cases-0.080
nuts-0.075
ils-0.074
 Sale-0.073
 invitations-0.071
Token the
Token ID262
Feature activation+6.65e-01

Pos logit contributions
 sur+0.006
avis+0.006
 rely+0.006
 polar+0.006
 matter+0.006
Neg logit contributions
cases-0.008
ils-0.007
nuts-0.007
driver-0.007
 Sale-0.006
Token sky
Token ID6766
Feature activation+1.47e+00

Pos logit contributions
 sur+0.016
avis+0.015
 rely+0.015
please+0.014
 listener+0.014
Neg logit contributions
cases-0.026
ils-0.025
nuts-0.025
 Sale-0.024
 invitations-0.024
Token.
Token ID13
Feature activation-9.73e-02

Pos logit contributions
 sur+0.049
avis+0.049
 rely+0.047
please+0.046
 listener+0.046
Neg logit contributions
cases-0.043
ils-0.041
nuts-0.040
 invitations-0.037
 Sale-0.037
Token.
Token ID13
Feature activation-2.05e-01

Pos logit contributions
 Sale+0.057
cases+0.057
ils+0.056
 invitations+0.055
nuts+0.054
Neg logit contributions
 rely-0.029
 day-0.027
 bas-0.027
avis-0.026
 ask-0.026
Token They
Token ID1119
Feature activation-1.56e+00

Pos logit contributions
cases+0.009
ils+0.008
nuts+0.008
 Sale+0.008
 invitations+0.008
Neg logit contributions
 sur-0.004
avis-0.004
 rely-0.004
please-0.004
 listener-0.004
Token were
Token ID547
Feature activation-1.02e+00

Pos logit contributions
cases+0.076
ils+0.072
 Sale+0.072
 invitations+0.072
nuts+0.071
Neg logit contributions
 ask-0.022
 sur-0.021
 asked-0.020
 bas-0.019
 wonder-0.019
Token blue
Token ID4171
Feature activation+2.91e+00

Pos logit contributions
cases+0.040
ils+0.038
nuts+0.037
 Sale+0.036
driver+0.036
Neg logit contributions
 sur-0.024
 rely-0.024
avis-0.023
please-0.023
 polar-0.023
Token and
Token ID290
Feature activation-1.93e-01

Pos logit contributions
avis+0.069
 polar+0.068
 fro+0.068
 Arctic+0.067
 sur+0.066
Neg logit contributions
ils-0.108
cases-0.108
 Sale-0.104
ports-0.098
 invitations-0.098
Token spe
Token ID693
Feature activation+7.15e+00

Pos logit contributions
cases+0.008
ils+0.008
nuts+0.007
 Sale+0.007
 invitations+0.007
Neg logit contributions
avis-0.004
 rely-0.004
 sur-0.004
 polar-0.004
 cos-0.004
Tokenck
Token ID694
Feature activation+2.19e+00

Pos logit contributions
 sur+0.220
avis+0.210
 rely+0.207
 listener+0.197
 fro+0.194
Neg logit contributions
ils-0.229
cases-0.229
nuts-0.221
 Sale-0.214
 invitations-0.207
Tokenled
Token ID992
Feature activation-2.09e+00

Pos logit contributions
 crit+0.069
 fro+0.069
 sur+0.068
avis+0.068
 bas+0.068
Neg logit contributions
 Sale-0.063
cases-0.061
 invitations-0.057
nuts-0.056
 puzzles-0.054
Token.
Token ID13
Feature activation-1.71e-01

Pos logit contributions
cases+0.077
 Sale+0.074
driver+0.070
ils+0.069
fs+0.067
Neg logit contributions
 sur-0.055
 fro-0.054
 bas-0.054
 bra-0.053
 polar-0.053
Token L
Token ID406
Feature activation+1.60e+00

Pos logit contributions
cases+0.007
ils+0.007
nuts+0.007
 Sale+0.007
 invitations+0.007
Neg logit contributions
 sur-0.004
avis-0.004
 rely-0.004
please-0.003
 listener-0.003
Tokenila
Token ID10102
Feature activation+2.39e+00

Pos logit contributions
 rely+0.039
 sur+0.039
avis+0.038
please+0.037
 polar+0.036
Neg logit contributions
cases-0.061
ils-0.059
nuts-0.057
 invitations-0.055
 Sale-0.055
Token You
Token ID921
Feature activation-1.43e+00

Pos logit contributions
 sur+0.027
avis+0.027
 rely+0.025
 listener+0.025
please+0.025
Neg logit contributions
cases-0.032
 Sale-0.031
nuts-0.030
driver-0.029
 invitations-0.028
Token are
Token ID389
Feature activation-6.67e-01

Pos logit contributions
cases+0.046
ils+0.044
nuts+0.043
 Sale+0.042
 invitations+0.042
Neg logit contributions
 sur-0.044
 rely-0.043
avis-0.043
please-0.040
 ask-0.040
Token mean
Token ID1612
Feature activation-2.60e+00

Pos logit contributions
cases+0.022
ils+0.020
nuts+0.020
 Sale+0.019
driver+0.018
Neg logit contributions
 sur-0.021
avis-0.021
 rely-0.021
please-0.020
 polar-0.019
Token.
Token ID13
Feature activation+1.60e+00

Pos logit contributions
cases+0.100
ils+0.098
nuts+0.095
 Sale+0.093
 invitations+0.092
Neg logit contributions
 sur-0.063
avis-0.058
 listener-0.057
please-0.057
 rely-0.056
Token B
Token ID347
Feature activation+2.74e+00

Pos logit contributions
 sur+0.044
avis+0.043
 rely+0.041
please+0.040
 listener+0.040
Neg logit contributions
cases-0.056
ils-0.054
 Sale-0.054
nuts-0.054
 invitations-0.051
Tokenunn
Token ID20935
Feature activation+7.15e+00

Pos logit contributions
 sur+0.075
avis+0.074
 rely+0.072
please+0.070
 listener+0.069
Neg logit contributions
cases-0.099
nuts-0.093
ils-0.092
 Sale-0.090
driver-0.086
Tokenies
Token ID444
Feature activation+1.17e-02

Pos logit contributions
 sur+0.165
avis+0.161
 rely+0.156
 listener+0.147
please+0.147
Neg logit contributions
cases-0.285
ils-0.270
nuts-0.270
 Sale-0.265
 invitations-0.259
Token are
Token ID389
Feature activation-5.95e-01

Pos logit contributions
 sur+0.000
 rely+0.000
avis+0.000
please+0.000
 ask+0.000
Neg logit contributions
cases-0.000
nuts-0.000
ils-0.000
 Sale-0.000
 invitations-0.000
Token nice
Token ID3621
Feature activation+4.23e-01

Pos logit contributions
cases+0.019
ils+0.017
nuts+0.017
 Sale+0.017
driver+0.016
Neg logit contributions
 sur-0.019
avis-0.019
 rely-0.018
please-0.018
 polar-0.018
Token and
Token ID290
Feature activation+1.56e-01

Pos logit contributions
 sur+0.010
 rely+0.010
avis+0.010
 listener+0.010
please+0.009
Neg logit contributions
cases-0.017
ils-0.015
nuts-0.015
 Sale-0.015
 invitations-0.014
Token they
Token ID484
Feature activation-1.95e+00

Pos logit contributions
 sur+0.004
avis+0.004
 rely+0.004
 listener+0.004
please+0.004
Neg logit contributions
cases-0.006
ils-0.005
nuts-0.005
 Sale-0.005
 invitations-0.005
Tokendad
Token ID47984
Feature activation+9.56e-01

Pos logit contributions
cases+0.000
ils+0.000
nuts+0.000
 Sale+0.000
 invitations+0.000
Neg logit contributions
 sur-0.000
avis-0.000
 rely-0.000
 Kang-0.000
 listener-0.000
Token was
Token ID373
Feature activation+1.09e+00

Pos logit contributions
 sur+0.022
avis+0.021
 rely+0.020
please+0.019
 listener+0.019
Neg logit contributions
cases-0.039
ils-0.037
nuts-0.037
 Sale-0.036
 invitations-0.035
Token so
Token ID523
Feature activation+1.62e+00

Pos logit contributions
 sur+0.033
avis+0.032
 rely+0.031
please+0.030
 listener+0.030
Neg logit contributions
cases-0.036
ils-0.034
nuts-0.034
 Sale-0.033
 invitations-0.032
Token proud
Token ID6613
Feature activation-1.83e+00

Pos logit contributions
avis+0.039
 sur+0.037
 rely+0.036
please+0.035
 Arctic+0.035
Neg logit contributions
cases-0.063
nuts-0.059
ils-0.059
 Sale-0.058
 invitations-0.057
Token of
Token ID286
Feature activation+2.57e+00

Pos logit contributions
cases+0.043
nuts+0.039
ils+0.039
es+0.039
 invitations+0.037
Neg logit contributions
avis-0.069
 rely-0.068
please-0.068
 Arctic-0.066
 Kang-0.065
Token Mate
Token ID24787
Feature activation+7.14e+00

Pos logit contributions
 sur+0.081
please+0.079
avis+0.077
 listener+0.077
 rely+0.077
Neg logit contributions
ils-0.076
cases-0.076
nuts-0.073
 Sale-0.073
 invitations-0.070
Tokeno
Token ID78
Feature activation+2.54e+00

Pos logit contributions
 rely+0.149
avis+0.140
 sur+0.135
 beacon+0.135
 Arctic+0.130
Neg logit contributions
cases-0.296
es-0.282
nuts-0.278
ils-0.277
ports-0.274
Token's
Token ID338
Feature activation+2.03e+00

Pos logit contributions
 sur+0.075
avis+0.073
 rely+0.071
 polar+0.067
 listener+0.067
Neg logit contributions
cases-0.085
 Sale-0.081
nuts-0.080
ils-0.080
 invitations-0.077
Token skills
Token ID4678
Feature activation-1.83e+00

Pos logit contributions
 sur+0.054
avis+0.051
 rely+0.051
 listener+0.048
please+0.047
Neg logit contributions
cases-0.075
ils-0.073
nuts-0.071
 Sale-0.070
 invitations-0.068
Token.
Token ID13
Feature activation+4.36e-01

Pos logit contributions
cases+0.064
ils+0.064
nuts+0.062
 Sale+0.062
 invitations+0.062
Neg logit contributions
 sur-0.050
 rely-0.048
avis-0.046
 listener-0.043
 clim-0.043
Token He
Token ID679
Feature activation-1.28e-01

Pos logit contributions
 rely+0.013
please+0.012
avis+0.012
 matter+0.012
 bas+0.012
Neg logit contributions
cases-0.015
 Sale-0.014
ports-0.013
driver-0.013
nuts-0.013
Token two
Token ID734
Feature activation-1.69e+00

Pos logit contributions
avis+0.005
 sur+0.005
 rely+0.005
please+0.005
 bas+0.005
Neg logit contributions
cases-0.007
 Sale-0.007
nuts-0.007
ils-0.007
 invitations-0.007
Token squirrel
Token ID33039
Feature activation+1.66e-01

Pos logit contributions
cases+0.064
ils+0.061
nuts+0.060
 Sale+0.059
 invitations+0.058
Neg logit contributions
 sur-0.043
avis-0.042
 rely-0.041
please-0.038
 listener-0.038
Tokens
Token ID82
Feature activation+7.84e-01

Pos logit contributions
 sur+0.005
avis+0.005
 rely+0.005
please+0.004
 listener+0.004
Neg logit contributions
cases-0.006
ils-0.006
nuts-0.005
 Sale-0.005
 invitations-0.005
Token kept
Token ID4030
Feature activation+3.22e+00

Pos logit contributions
 sur+0.021
avis+0.021
 rely+0.020
please+0.019
 listener+0.019
Neg logit contributions
cases-0.029
ils-0.027
nuts-0.027
 Sale-0.026
 invitations-0.026
Token qu
Token ID627
Feature activation-3.31e+00

Pos logit contributions
 sur+0.088
avis+0.085
 rely+0.083
please+0.079
 polar+0.077
Neg logit contributions
cases-0.118
ils-0.112
nuts-0.111
 Sale-0.107
 invitations-0.105
Tokenarre
Token ID9624
Feature activation-1.26e+01

Pos logit contributions
cases+0.101
ils+0.095
nuts+0.094
 Sale+0.092
 invitations+0.090
Neg logit contributions
 sur-0.109
avis-0.108
 rely-0.103
please-0.100
 fro-0.099
Tokenling
Token ID1359
Feature activation-7.04e-01

Pos logit contributions
cases+0.441
 Sale+0.431
nuts+0.426
ils+0.409
 invitations+0.406
Neg logit contributions
 sur-0.345
avis-0.331
 rely-0.326
please-0.311
 listener-0.307
Token until
Token ID1566
Feature activation+6.40e-01

Pos logit contributions
cases+0.025
 Sale+0.023
ils+0.023
nuts+0.023
 invitations+0.022
Neg logit contributions
 sur-0.019
 fro-0.018
avis-0.018
 crit-0.017
 rely-0.017
Token finally
Token ID3443
Feature activation+1.43e+00

Pos logit contributions
 sur+0.018
avis+0.018
 rely+0.017
 fro+0.016
 polar+0.016
Neg logit contributions
cases-0.022
ils-0.022
nuts-0.021
 invitations-0.020
 Sale-0.020
Token Sammy
Token ID44067
Feature activation-1.72e+00

Pos logit contributions
 sur+0.038
avis+0.034
 ask+0.033
 fro+0.033
 Sk+0.033
Neg logit contributions
nuts-0.053
cases-0.051
ils-0.050
driver-0.049
 pages-0.049
Token said
Token ID531
Feature activation+7.28e-01

Pos logit contributions
cases+0.065
ils+0.061
nuts+0.060
 Sale+0.058
 invitations+0.056
Neg logit contributions
avis-0.044
 sur-0.043
 rely-0.042
 listener-0.040
please-0.039
Token together
Token ID1978
Feature activation-2.91e+00

Pos logit contributions
 sur+0.054
 place+0.053
 rely+0.053
 fro+0.051
 day+0.051
Neg logit contributions
cases-0.066
nuts-0.059
 Sale-0.059
ils-0.055
 invitations-0.054
Token.
Token ID13
Feature activation+3.46e-02

Pos logit contributions
cases+0.100
nuts+0.092
ils+0.092
 Sale+0.091
 invitations+0.089
Neg logit contributions
 sur-0.085
 rely-0.083
avis-0.079
 listener-0.077
 fro-0.076
Token They
Token ID1119
Feature activation-1.88e+00

Pos logit contributions
 sur+0.001
avis+0.001
 rely+0.001
please+0.001
 listener+0.001
Neg logit contributions
cases-0.001
ils-0.001
nuts-0.001
 Sale-0.001
 invitations-0.001
Token never
Token ID1239
Feature activation-7.39e-01

Pos logit contributions
cases+0.076
nuts+0.071
 Sale+0.070
ados+0.070
 invitations+0.069
Neg logit contributions
 sur-0.044
 rely-0.041
avis-0.040
 fro-0.040
 bas-0.039
Token qu
Token ID627
Feature activation-2.94e+00

Pos logit contributions
cases+0.025
ils+0.025
nuts+0.024
 Sale+0.023
 invitations+0.022
Neg logit contributions
avis-0.022
 sur-0.020
 Kang-0.019
 Arctic-0.019
 rely-0.019
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.080
 Sale+0.073
nuts+0.071
 invitations+0.070
ils+0.068
Neg logit contributions
 fro-0.099
 sur-0.098
 Arctic-0.097
avis-0.096
please-0.093
Tokenled
Token ID992
Feature activation-2.46e+00

Pos logit contributions
 Sale+0.329
cases+0.325
nuts+0.321
 sandwiches+0.288
 invitations+0.286
Neg logit contributions
 sur-0.419
 matter-0.419
heart-0.413
 listener-0.404
 rely-0.403
Token again
Token ID757
Feature activation-2.50e+00

Pos logit contributions
cases+0.089
 Sale+0.085
ils+0.083
nuts+0.082
ports+0.078
Neg logit contributions
 sur-0.070
 rely-0.064
 listener-0.063
avis-0.063
 fro-0.061
Token.
Token ID13
Feature activation+3.64e-02

Pos logit contributions
cases+0.083
ils+0.079
nuts+0.078
 Sale+0.075
 invitations+0.074
Neg logit contributions
 sur-0.075
avis-0.074
 rely-0.072
please-0.068
 listener-0.068
Token<|endoftext|>
Token ID50256
Feature activation-5.37e+00

Pos logit contributions
 sur+0.001
avis+0.001
 rely+0.001
please+0.001
 listener+0.001
Neg logit contributions
cases-0.001
 Sale-0.001
ils-0.001
nuts-0.001
 invitations-0.001
TokenOnce
Token ID7454
Feature activation-1.08e+00

Pos logit contributions
cases+0.218
ils+0.210
nuts+0.207
 Sale+0.205
es+0.200
Neg logit contributions
 sur-0.117
avis-0.116
 rely-0.111
please-0.106
 listener-0.104
Token on
Token ID319
Feature activation-3.67e+00

Pos logit contributions
 day+0.039
 rely+0.039
 sur+0.039
avis+0.039
 morning+0.038
Neg logit contributions
cases-0.034
nuts-0.033
 invitations-0.031
ils-0.030
 sandwiches-0.029
Token,
Token ID11
Feature activation-8.19e-01

Pos logit contributions
cases+0.152
nuts+0.146
ils+0.142
 invitations+0.138
 Sale+0.136
Neg logit contributions
 rely-0.079
avis-0.078
 sur-0.078
 day-0.076
 fro-0.071
Token they
Token ID484
Feature activation-2.45e+00

Pos logit contributions
cases+0.023
ils+0.022
nuts+0.021
 Sale+0.021
 invitations+0.021
Neg logit contributions
 sur-0.028
avis-0.028
 rely-0.027
please-0.026
 listener-0.026
Token never
Token ID1239
Feature activation-6.63e-01

Pos logit contributions
cases+0.098
nuts+0.091
ils+0.090
 Sale+0.088
 invitations+0.088
Neg logit contributions
 sur-0.059
avis-0.058
 rely-0.056
please-0.054
 listener-0.053
Token qu
Token ID627
Feature activation-2.86e+00

Pos logit contributions
ils+0.020
cases+0.019
nuts+0.019
 puzzles+0.018
 Sale+0.018
Neg logit contributions
avis-0.021
 Arctic-0.020
 Kang-0.019
 sur-0.019
ang-0.019
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.073
 Sale+0.066
nuts+0.064
ils+0.062
driver+0.062
Neg logit contributions
avis-0.103
 sur-0.102
 fro-0.101
please-0.100
 Arctic-0.099
Tokenled
Token ID992
Feature activation-2.43e+00

Pos logit contributions
cases+0.303
 Sale+0.294
nuts+0.290
 invitations+0.263
 sandwiches+0.259
Neg logit contributions
 sur-0.461
 rely-0.444
 listener-0.441
please-0.440
avis-0.437
Token again
Token ID757
Feature activation-2.44e+00

Pos logit contributions
cases+0.085
ils+0.079
 Sale+0.079
nuts+0.079
 invitations+0.074
Neg logit contributions
 sur-0.071
avis-0.067
 rely-0.067
 listener-0.065
please-0.064
Token over
Token ID625
Feature activation-8.61e-01

Pos logit contributions
cases+0.079
ils+0.075
nuts+0.074
 Sale+0.071
 invitations+0.070
Neg logit contributions
 sur-0.075
avis-0.074
 rely-0.071
please-0.068
 listener-0.068
Token clothes
Token ID8242
Feature activation+6.61e-01

Pos logit contributions
cases+0.029
ils+0.027
nuts+0.027
 Sale+0.026
 invitations+0.025
Neg logit contributions
 sur-0.027
avis-0.026
 rely-0.025
 listener-0.024
please-0.024
Token.
Token ID13
Feature activation+1.18e-01

Pos logit contributions
 sur+0.018
avis+0.017
 rely+0.016
 listener+0.016
please+0.016
Neg logit contributions
 Sale-0.023
ils-0.023
cases-0.023
nuts-0.022
ports-0.022
Token and
Token ID290
Feature activation-6.40e-01

Pos logit contributions
 rely+0.001
 sur+0.001
avis+0.001
 ask+0.001
 polar+0.001
Neg logit contributions
cases-0.001
ils-0.001
 Sale-0.001
nuts-0.001
 invitations-0.001
Token their
Token ID511
Feature activation-2.17e+00

Pos logit contributions
cases+0.022
ils+0.021
nuts+0.021
 Sale+0.020
driver+0.019
Neg logit contributions
 sur-0.018
avis-0.017
 rely-0.017
 polar-0.017
please-0.017
Token children
Token ID1751
Feature activation+4.80e-01

Pos logit contributions
cases+0.075
ils+0.071
nuts+0.070
 Sale+0.068
driver+0.066
Neg logit contributions
 sur-0.061
avis-0.059
 rely-0.058
 listener-0.056
 polar-0.056
Token,
Token ID11
Feature activation-1.06e+00

Pos logit contributions
 rely+0.011
 sur+0.011
avis+0.011
 ask+0.011
 bas+0.010
Neg logit contributions
cases-0.018
ils-0.017
 Sale-0.017
nuts-0.017
 invitations-0.017
Token qu
Token ID627
Feature activation-2.91e+00

Pos logit contributions
cases+0.039
ils+0.038
nuts+0.037
 Sale+0.036
 invitations+0.035
Neg logit contributions
 sur-0.028
avis-0.027
 rely-0.026
 polar-0.025
 fro-0.025
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.104
driver+0.097
 Sale+0.096
nuts+0.095
ils+0.093
Neg logit contributions
avis-0.080
 sur-0.075
 Arctic-0.075
please-0.074
heart-0.073
Tokenling
Token ID1359
Feature activation-5.23e-01

Pos logit contributions
 Sale+0.531
cases+0.509
nuts+0.508
ifying+0.494
driver+0.492
Neg logit contributions
heart-0.246
 sur-0.222
 matter-0.214
avis-0.212
ie-0.207
Token.
Token ID13
Feature activation-3.19e-01

Pos logit contributions
cases+0.021
driver+0.020
 Sale+0.019
nuts+0.019
ils+0.019
Neg logit contributions
 crit-0.011
 fro-0.011
heart-0.011
 day-0.011
 sur-0.011
Token
Token ID220
Feature activation-4.82e-01

Pos logit contributions
cases+0.012
nuts+0.011
ils+0.011
 Sale+0.011
 invitations+0.011
Neg logit contributions
 sur-0.008
avis-0.008
 rely-0.008
 listener-0.008
please-0.008
Token
Token ID198
Feature activation+3.88e-01

Pos logit contributions
cases+0.022
ils+0.021
nuts+0.021
 Sale+0.021
 invitations+0.020
Neg logit contributions
 sur-0.008
avis-0.008
 rely-0.008
please-0.007
 listener-0.007
Token
Token ID198
Feature activation+4.63e-01

Pos logit contributions
 sur+0.008
avis+0.008
 rely+0.008
 listener+0.007
 fro+0.007
Neg logit contributions
cases-0.016
nuts-0.015
 Sale-0.015
ils-0.015
 invitations-0.015
Token mum
Token ID25682
Feature activation+8.45e-01

Pos logit contributions
cases+0.082
ils+0.079
nuts+0.078
driver+0.075
es+0.074
Neg logit contributions
 sur-0.052
avis-0.049
 rely-0.048
 listener-0.046
 fro-0.044
Token and
Token ID290
Feature activation-7.05e-01

Pos logit contributions
 sur+0.026
avis+0.025
 rely+0.025
 listener+0.023
please+0.023
Neg logit contributions
cases-0.028
ils-0.026
nuts-0.026
 Sale-0.025
 invitations-0.025
Token understood
Token ID7247
Feature activation-2.65e+00

Pos logit contributions
cases+0.025
ils+0.024
nuts+0.023
 Sale+0.023
 invitations+0.022
Neg logit contributions
 sur-0.019
avis-0.019
 rely-0.019
please-0.018
 listener-0.018
Token that
Token ID326
Feature activation+2.13e+00

Pos logit contributions
cases+0.096
ils+0.087
nuts+0.087
 Sale+0.084
driver+0.084
Neg logit contributions
 rely-0.074
 sur-0.073
avis-0.072
 ask-0.070
 check-0.068
Token qu
Token ID627
Feature activation-2.94e+00

Pos logit contributions
 sur+0.070
avis+0.068
 rely+0.068
 listener+0.064
please+0.063
Neg logit contributions
cases-0.065
ils-0.061
nuts-0.060
 Sale-0.058
 invitations-0.057
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.084
 invitations+0.072
 Sale+0.072
driver+0.071
ils+0.071
Neg logit contributions
avis-0.099
 sur-0.097
 Arctic-0.097
 rely-0.096
ini-0.096
Tokenlling
Token ID2680
Feature activation-4.34e+00

Pos logit contributions
cases+0.405
 Sale+0.398
nuts+0.390
driver+0.364
 invitations+0.360
Neg logit contributions
 sur-0.346
heart-0.341
 rely-0.340
 matter-0.339
avis-0.330
Token was
Token ID373
Feature activation+7.85e-01

Pos logit contributions
cases+0.167
 Sale+0.163
nuts+0.162
ils+0.159
 invitations+0.155
Neg logit contributions
 sur-0.093
 rely-0.093
 day-0.087
 ask-0.086
 matter-0.084
Token not
Token ID407
Feature activation-2.41e+00

Pos logit contributions
 sur+0.024
 rely+0.024
avis+0.023
please+0.022
 listener+0.022
Neg logit contributions
cases-0.026
ils-0.024
nuts-0.024
 Sale-0.023
driver-0.022
Token a
Token ID257
Feature activation-9.77e-01

Pos logit contributions
cases+0.087
ils+0.081
nuts+0.080
 Sale+0.078
 invitations+0.075
Neg logit contributions
 sur-0.067
 rely-0.066
avis-0.063
 listener-0.062
please-0.061
Token good
Token ID922
Feature activation-1.48e+00

Pos logit contributions
cases+0.033
ils+0.031
nuts+0.031
 Sale+0.030
 invitations+0.030
Neg logit contributions
avis-0.028
 sur-0.028
 rely-0.027
please-0.026
 listener-0.026
Token friends
Token ID2460
Feature activation+5.61e-01

Pos logit contributions
cases+0.050
ils+0.048
 Sale+0.048
nuts+0.047
driver+0.045
Neg logit contributions
 sur-0.025
avis-0.024
 rely-0.023
 fro-0.022
 polar-0.022
Token didn
Token ID1422
Feature activation-3.38e+00

Pos logit contributions
 sur+0.014
 ask+0.012
 rely+0.012
 regret+0.012
 bas+0.011
Neg logit contributions
cases-0.022
nuts-0.021
 Sale-0.021
ados-0.021
 invitations-0.021
Token't
Token ID470
Feature activation-1.77e+00

Pos logit contributions
ils+0.162
 invitations+0.148
ports+0.143
 pages+0.142
ados+0.139
Neg logit contributions
 rely-0.066
 sur-0.064
 But-0.063
'm-0.061
avis-0.058
Token stop
Token ID2245
Feature activation-1.10e+00

Pos logit contributions
cases+0.075
ils+0.072
nuts+0.071
 Sale+0.069
 invitations+0.069
Neg logit contributions
 sur-0.037
avis-0.036
 rely-0.034
please-0.032
 listener-0.032
Token qu
Token ID627
Feature activation-2.90e+00

Pos logit contributions
cases+0.033
ils+0.031
nuts+0.031
 Sale+0.029
 invitations+0.029
Neg logit contributions
 sur-0.036
avis-0.036
 rely-0.035
please-0.033
 listener-0.033
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.082
 Sale+0.078
ils+0.074
 invitations+0.074
driver+0.073
Neg logit contributions
avis-0.098
 fro-0.097
 sur-0.096
 Arctic-0.092
heart-0.091
Tokenling
Token ID1359
Feature activation-3.58e-01

Pos logit contributions
 Sale+0.482
nuts+0.455
cases+0.447
ifying+0.441
 invitations+0.436
Neg logit contributions
heart-0.288
 sur-0.286
 matter-0.277
 day-0.261
 listener-0.256
Token and
Token ID290
Feature activation-8.79e-01

Pos logit contributions
cases+0.014
 Sale+0.014
ils+0.014
nuts+0.013
ports+0.013
Neg logit contributions
 sur-0.008
 fro-0.007
 bra-0.007
 rely-0.007
 crit-0.007
Token were
Token ID547
Feature activation-1.29e+00

Pos logit contributions
cases+0.031
ils+0.029
nuts+0.029
 Sale+0.028
 invitations+0.027
Neg logit contributions
 sur-0.025
avis-0.025
 rely-0.024
please-0.023
 listener-0.023
Token getting
Token ID1972
Feature activation+1.29e+00

Pos logit contributions
cases+0.051
ils+0.048
 Sale+0.047
nuts+0.047
ports+0.046
Neg logit contributions
 sur-0.032
 rely-0.031
avis-0.030
 listener-0.029
 bra-0.029
Token louder
Token ID27089
Feature activation-1.12e+00

Pos logit contributions
 sur+0.031
avis+0.030
 rely+0.029
please+0.028
 polar+0.027
Neg logit contributions
cases-0.052
ils-0.050
 Sale-0.048
nuts-0.048
 invitations-0.046
Token walking
Token ID6155
Feature activation+1.70e+00

Pos logit contributions
 sur+0.027
avis+0.026
 rely+0.026
please+0.024
 listener+0.024
Neg logit contributions
cases-0.039
ils-0.037
nuts-0.037
 Sale-0.036
driver-0.034
Token by
Token ID416
Feature activation+1.25e+00

Pos logit contributions
 sur+0.042
 fro+0.040
 listener+0.040
 girl+0.040
avis+0.040
Neg logit contributions
cases-0.063
 Sale-0.061
nuts-0.061
ils-0.060
driver-0.057
Token saw
Token ID2497
Feature activation-7.87e-01

Pos logit contributions
 sur+0.034
avis+0.033
 fro+0.032
 rely+0.032
 ask+0.031
Neg logit contributions
cases-0.048
nuts-0.043
ils-0.043
 Sale-0.041
 invitations-0.040
Token them
Token ID606
Feature activation-5.80e-01

Pos logit contributions
cases+0.027
nuts+0.023
ils+0.022
driver+0.022
 Sale+0.022
Neg logit contributions
avis-0.023
 fro-0.022
 sur-0.022
 ask-0.022
 crit-0.022
Token qu
Token ID627
Feature activation-2.94e+00

Pos logit contributions
ils+0.022
cases+0.021
 Sale+0.021
nuts+0.021
 invitations+0.020
Neg logit contributions
 rely-0.014
 sur-0.014
avis-0.013
 ask-0.013
 fro-0.013
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.092
 Sale+0.084
driver+0.083
nuts+0.081
ils+0.080
Neg logit contributions
avis-0.091
 Arctic-0.089
 sur-0.087
 fro-0.087
please-0.086
Tokenling
Token ID1359
Feature activation-4.25e-01

Pos logit contributions
 Sale+0.538
cases+0.529
nuts+0.514
ifying+0.491
driver+0.490
Neg logit contributions
heart-0.227
 sur-0.222
 matter-0.209
avis-0.202
 rely-0.201
Token.
Token ID13
Feature activation+3.57e-01

Pos logit contributions
cases+0.019
 Sale+0.018
ils+0.018
nuts+0.017
ports+0.017
Neg logit contributions
 bra-0.007
 fro-0.007
heart-0.007
 day-0.007
 girl-0.007
Token She
Token ID1375
Feature activation+1.27e-01

Pos logit contributions
 sur+0.010
avis+0.010
 rely+0.009
please+0.009
 listener+0.009
Neg logit contributions
cases-0.013
ils-0.012
nuts-0.012
 Sale-0.012
 invitations-0.012
Token was
Token ID373
Feature activation+9.97e-01

Pos logit contributions
 sur+0.003
avis+0.003
 rely+0.003
 listener+0.002
please+0.002
Neg logit contributions
cases-0.005
ils-0.005
nuts-0.005
 Sale-0.005
 invitations-0.005
Token very
Token ID845
Feature activation-5.59e-01

Pos logit contributions
 sur+0.029
avis+0.029
 rely+0.028
please+0.027
 polar+0.026
Neg logit contributions
cases-0.034
ils-0.033
nuts-0.032
 Sale-0.031
 invitations-0.030
Token he
Token ID339
Feature activation-8.52e-02

Pos logit contributions
cases+0.045
es+0.042
ls+0.041
nuts+0.041
ils+0.040
Neg logit contributions
avis-0.038
please-0.037
 sur-0.036
 rely-0.036
 listener-0.035
Token found
Token ID1043
Feature activation-1.37e+00

Pos logit contributions
cases+0.003
ils+0.003
es+0.003
nuts+0.003
 puzzles+0.003
Neg logit contributions
avis-0.002
 rely-0.002
 sur-0.002
 pleasure-0.002
 listener-0.002
Token a
Token ID257
Feature activation-9.34e-01

Pos logit contributions
cases+0.024
nuts+0.024
ils+0.024
 Sale+0.021
 invitations+0.021
Neg logit contributions
 sur-0.061
avis-0.060
 rely-0.059
please-0.058
 listener-0.058
Token rabbit
Token ID22746
Feature activation+1.47e+00

Pos logit contributions
cases+0.028
nuts+0.027
ils+0.026
 invitations+0.026
 Sale+0.026
Neg logit contributions
avis-0.030
 sur-0.029
 rely-0.028
please-0.027
 listener-0.027
Token qu
Token ID627
Feature activation-3.09e+00

Pos logit contributions
 sur+0.040
avis+0.039
 rely+0.038
please+0.036
 listener+0.036
Neg logit contributions
cases-0.053
ils-0.050
nuts-0.049
 Sale-0.048
 invitations-0.047
Tokenarre
Token ID9624
Feature activation-1.24e+01

Pos logit contributions
cases+0.114
nuts+0.105
 Sale+0.104
ils+0.104
driver+0.100
Neg logit contributions
avis-0.084
 sur-0.082
 rely-0.077
please-0.076
 fro-0.076
Tokenlling
Token ID2680
Feature activation-4.32e+00

Pos logit contributions
cases+0.455
 Sale+0.435
nuts+0.429
ils+0.412
 invitations+0.401
Neg logit contributions
 sur-0.323
avis-0.312
 rely-0.305
 listener-0.293
please-0.291
Token with
Token ID351
Feature activation-3.24e-01

Pos logit contributions
cases+0.159
 Sale+0.154
ports+0.147
ils+0.140
nuts+0.138
Neg logit contributions
 sur-0.113
 bas-0.109
 day-0.108
 clim-0.108
 place-0.105
Token a
Token ID257
Feature activation-9.87e-01

Pos logit contributions
cases+0.009
ils+0.008
nuts+0.008
 Sale+0.008
 invitations+0.007
Neg logit contributions
 sur-0.012
avis-0.012
 rely-0.011
please-0.011
 fro-0.011
Token squirrel
Token ID33039
Feature activation+4.70e-01

Pos logit contributions
cases+0.035
ils+0.034
nuts+0.032
 Sale+0.031
 invitations+0.031
Neg logit contributions
 sur-0.028
avis-0.027
 rely-0.027
please-0.026
 fro-0.026
Token.
Token ID13
Feature activation-4.11e-01

Pos logit contributions
 sur+0.012
avis+0.012
 rely+0.012
 polar+0.011
 fro+0.011
Neg logit contributions
cases-0.017
 Sale-0.016
ils-0.016
 invitations-0.016
nuts-0.016
Token but
Token ID475
Feature activation+1.97e+00

Pos logit contributions
cases+0.045
nuts+0.042
ils+0.042
 invitations+0.040
 Sale+0.040
Neg logit contributions
 sur-0.033
avis-0.033
 rely-0.032
please-0.030
 listener-0.030
Token today
Token ID1909
Feature activation-2.80e+00

Pos logit contributions
 sur+0.057
avis+0.056
 rely+0.054
please+0.051
 listener+0.051
Neg logit contributions
cases-0.068
ils-0.065
nuts-0.064
 Sale-0.062
 invitations-0.061
Token they
Token ID484
Feature activation-2.63e+00

Pos logit contributions
cases+0.080
ils+0.076
nuts+0.076
 Sale+0.072
 invitations+0.070
Neg logit contributions
 sur-0.096
avis-0.094
 rely-0.091
 listener-0.088
 fro-0.088
Token were
Token ID547
Feature activation-1.31e+00

Pos logit contributions
cases+0.099
nuts+0.096
driver+0.095
 Sale+0.095
 invitations+0.094
Neg logit contributions
 sur-0.066
 rely-0.060
avis-0.059
 ask-0.058
please-0.058
Token qu
Token ID627
Feature activation-2.99e+00

Pos logit contributions
cases+0.045
ils+0.043
nuts+0.043
 Sale+0.042
driver+0.041
Neg logit contributions
 sur-0.038
avis-0.036
 rely-0.036
please-0.034
 listener-0.034
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.074
 Sale+0.070
ils+0.068
nuts+0.068
driver+0.066
Neg logit contributions
avis-0.112
 sur-0.112
please-0.107
 Arctic-0.106
 rely-0.106
Tokenling
Token ID1359
Feature activation-6.14e-01

Pos logit contributions
 Sale+0.436
cases+0.415
nuts+0.410
driver+0.398
 sandwiches+0.381
Neg logit contributions
 sur-0.335
heart-0.326
 rely-0.318
please-0.312
 listener-0.309
Token.
Token ID13
Feature activation-2.69e-01

Pos logit contributions
cases+0.023
 Sale+0.023
driver+0.022
nuts+0.022
ils+0.022
Neg logit contributions
 day-0.014
heart-0.014
 sur-0.014
 rely-0.013
 morning-0.013
Token Jenna
Token ID45377
Feature activation-2.75e+00

Pos logit contributions
cases+0.010
nuts+0.010
 Sale+0.009
ils+0.009
 invitations+0.009
Neg logit contributions
 sur-0.007
avis-0.006
 rely-0.006
 listener-0.006
please-0.006
Token wanted
Token ID2227
Feature activation-3.28e-02

Pos logit contributions
cases+0.120
ils+0.115
nuts+0.115
 Sale+0.113
 invitations+0.111
Neg logit contributions
 sur-0.053
avis-0.051
 rely-0.049
please-0.046
 listener-0.045
Token to
Token ID284
Feature activation-1.86e-01

Pos logit contributions
cases+0.001
ils+0.001
nuts+0.001
 Sale+0.001
driver+0.001
Neg logit contributions
 sur-0.001
avis-0.001
 rely-0.001
please-0.001
 fro-0.001
Token little
Token ID1310
Feature activation+4.18e+00

Pos logit contributions
cases+0.053
ils+0.051
 Sale+0.050
nuts+0.050
driver+0.049
Neg logit contributions
 sur-0.033
avis-0.033
 rely-0.031
 polar-0.031
 fro-0.030
Token birds
Token ID10087
Feature activation+1.96e+00

Pos logit contributions
 sur+0.114
avis+0.112
 rely+0.110
please+0.104
 listener+0.102
Neg logit contributions
cases-0.147
ils-0.142
nuts-0.141
 invitations-0.136
 Sale-0.133
Token that
Token ID326
Feature activation+2.19e+00

Pos logit contributions
 sur+0.049
avis+0.047
 rely+0.046
 fro+0.044
please+0.043
Neg logit contributions
cases-0.075
ils-0.071
nuts-0.070
 Sale-0.069
 invitations-0.068
Token were
Token ID547
Feature activation-1.19e+00

Pos logit contributions
avis+0.073
 sur+0.072
 rely+0.070
 listener+0.067
please+0.066
Neg logit contributions
cases-0.067
ils-0.064
nuts-0.062
es-0.060
 Sale-0.058
Token qu
Token ID627
Feature activation-3.08e+00

Pos logit contributions
cases+0.050
ils+0.047
 Sale+0.046
nuts+0.046
driver+0.045
Neg logit contributions
 sur-0.025
avis-0.025
 rely-0.025
please-0.023
 polar-0.023
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.107
 Sale+0.100
nuts+0.097
driver+0.096
ils+0.096
Neg logit contributions
avis-0.089
 sur-0.084
please-0.082
 fro-0.081
 Arctic-0.081
Tokenling
Token ID1359
Feature activation-2.94e-01

Pos logit contributions
 Sale+0.489
cases+0.482
nuts+0.460
driver+0.439
 invitations+0.438
Neg logit contributions
 sur-0.277
heart-0.273
avis-0.265
 matter-0.257
 rely-0.257
Token.
Token ID13
Feature activation-3.06e-01

Pos logit contributions
cases+0.013
 Sale+0.013
ports+0.012
ils+0.012
driver+0.012
Neg logit contributions
 fro-0.005
 bra-0.005
heart-0.005
 clim-0.005
 day-0.005
Token
Token ID198
Feature activation+5.25e-01

Pos logit contributions
cases+0.011
ils+0.011
nuts+0.011
 invitations+0.010
 Sale+0.010
Neg logit contributions
 sur-0.007
avis-0.007
 ask-0.007
 crit-0.007
 Kang-0.007
Token
Token ID198
Feature activation+5.50e-01

Pos logit contributions
 sur+0.013
avis+0.012
 rely+0.012
 listener+0.012
 polar+0.011
Neg logit contributions
cases-0.020
 Sale-0.019
ils-0.019
es-0.019
nuts-0.019
TokenThe
Token ID464
Feature activation+8.63e-01

Pos logit contributions
 sur+0.014
 Are+0.013
 ask+0.013
please+0.013
avis+0.013
Neg logit contributions
cases-0.021
nuts-0.020
 invitations-0.020
ils-0.019
 sandwiches-0.019
Token on
Token ID319
Feature activation-3.55e+00

Pos logit contributions
 day+0.040
avis+0.039
 sur+0.038
 morning+0.038
 rely+0.038
Neg logit contributions
cases-0.039
nuts-0.038
ils-0.036
 invitations-0.036
 puzzles-0.033
Token,
Token ID11
Feature activation-7.62e-01

Pos logit contributions
cases+0.151
ils+0.147
nuts+0.147
 invitations+0.140
 Sale+0.135
Neg logit contributions
 sur-0.073
avis-0.071
 rely-0.069
 day-0.069
 polar-0.066
Token they
Token ID484
Feature activation-2.42e+00

Pos logit contributions
cases+0.026
ils+0.024
nuts+0.024
 Sale+0.023
 invitations+0.023
Neg logit contributions
 sur-0.023
avis-0.022
 rely-0.022
please-0.021
 listener-0.021
Token never
Token ID1239
Feature activation-6.01e-01

Pos logit contributions
cases+0.092
nuts+0.085
ils+0.083
 invitations+0.082
 Sale+0.082
Neg logit contributions
 sur-0.064
avis-0.062
 rely-0.061
please-0.059
 listener-0.058
Token qu
Token ID627
Feature activation-2.93e+00

Pos logit contributions
ils+0.019
cases+0.019
nuts+0.018
 Sale+0.017
 puzzles+0.017
Neg logit contributions
avis-0.019
 sur-0.017
 Arctic-0.017
 Kang-0.017
 rely-0.017
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.076
 Sale+0.069
nuts+0.068
ils+0.065
 invitations+0.065
Neg logit contributions
avis-0.105
 fro-0.103
 sur-0.103
 Arctic-0.101
please-0.101
Tokenled
Token ID992
Feature activation-2.22e+00

Pos logit contributions
cases+0.286
nuts+0.272
 Sale+0.271
 invitations+0.244
ils+0.240
Neg logit contributions
 sur-0.475
avis-0.462
 rely-0.458
 listener-0.456
please-0.456
Token again
Token ID757
Feature activation-2.36e+00

Pos logit contributions
cases+0.075
ils+0.070
nuts+0.070
 Sale+0.069
 invitations+0.066
Neg logit contributions
 sur-0.067
avis-0.065
 rely-0.063
 listener-0.061
please-0.061
Token and
Token ID290
Feature activation-5.36e-01

Pos logit contributions
cases+0.076
ils+0.073
nuts+0.071
 Sale+0.068
 invitations+0.068
Neg logit contributions
 sur-0.073
avis-0.072
 rely-0.069
 listener-0.066
please-0.066
Token played
Token ID2826
Feature activation+1.02e+00

Pos logit contributions
ils+0.014
cases+0.014
nuts+0.013
 Sale+0.013
 puzzles+0.013
Neg logit contributions
avis-0.018
 sur-0.018
 rely-0.018
 Kang-0.018
 Arctic-0.018
Token together
Token ID1978
Feature activation-2.80e+00

Pos logit contributions
 sur+0.028
 rely+0.028
 fro+0.028
 place+0.028
 ask+0.027
Neg logit contributions
cases-0.037
nuts-0.034
 Sale-0.034
ifying-0.032
driver-0.032
Tokenâ
Token ID22940
Feature activation-1.60e+00

Pos logit contributions
ils+0.074
ados+0.068
ports+0.067
 files+0.066
 puzzles+0.065
Neg logit contributions
 sur-0.042
avis-0.041
 rely-0.040
 check-0.039
'm-0.038
Token
Token ID26391
Feature activation+1.75e-01

Pos logit contributions
cases+0.058
ils+0.058
 Sale+0.055
nuts+0.054
 invitations+0.051
Neg logit contributions
 sur-0.041
 polar-0.041
avis-0.040
 rely-0.039
please-0.038
Token
Token ID8151
Feature activation+2.09e+00

Pos logit contributions
 sur+0.005
avis+0.005
 rely+0.005
 listener+0.004
 fro+0.004
Neg logit contributions
nuts-0.006
ils-0.006
cases-0.006
 invitations-0.006
 Sale-0.006
Tokent
Token ID83
Feature activation-1.35e+00

Pos logit contributions
 sur+0.057
avis+0.057
ancer+0.052
 fro+0.052
 listener+0.052
Neg logit contributions
ils-0.073
 puzzles-0.068
 invitations-0.068
 Sale-0.066
nuts-0.066
Token qu
Token ID627
Feature activation-3.20e+00

Pos logit contributions
cases+0.050
ils+0.046
nuts+0.046
 invitations+0.045
es+0.044
Neg logit contributions
avis-0.034
 rely-0.033
 sur-0.033
 fro-0.032
 Arctic-0.031
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.084
 Sale+0.079
ils+0.078
nuts+0.077
driver+0.076
Neg logit contributions
avis-0.116
 sur-0.114
 fro-0.111
heart-0.109
 Arctic-0.108
Tokenling
Token ID1359
Feature activation-1.84e-01

Pos logit contributions
 Sale+0.443
nuts+0.427
cases+0.425
 invitations+0.408
 sandwiches+0.403
Neg logit contributions
 sur-0.326
heart-0.322
avis-0.304
 matter-0.301
 rely-0.298
Token,
Token ID11
Feature activation-1.04e+00

Pos logit contributions
cases+0.007
ils+0.007
nuts+0.007
 Sale+0.007
 invitations+0.006
Neg logit contributions
 sur-0.004
avis-0.004
 rely-0.004
 fro-0.004
 polar-0.004
Token the
Token ID262
Feature activation+5.52e-01

Pos logit contributions
cases+0.032
ils+0.030
nuts+0.030
 Sale+0.029
 invitations+0.028
Neg logit contributions
 sur-0.034
avis-0.033
 rely-0.032
 listener-0.031
please-0.031
Token furniture
Token ID13091
Feature activation-4.76e-01

Pos logit contributions
 sur+0.014
avis+0.014
 rely+0.013
 listener+0.013
 polar+0.012
Neg logit contributions
cases-0.021
nuts-0.020
ils-0.020
 Sale-0.019
 invitations-0.019
Token stayed
Token ID9658
Feature activation+2.13e+00

Pos logit contributions
cases+0.019
ils+0.018
nuts+0.018
 Sale+0.018
 invitations+0.018
Neg logit contributions
 sur-0.011
avis-0.010
 rely-0.010
 fro-0.009
 polar-0.009
Token and
Token ID290
Feature activation-8.73e-01

Pos logit contributions
cases+0.015
ils+0.014
 Sale+0.014
nuts+0.014
 invitations+0.013
Neg logit contributions
 sur-0.012
avis-0.011
 rely-0.011
 listener-0.011
 fro-0.011
Token saw
Token ID2497
Feature activation-1.19e+00

Pos logit contributions
cases+0.034
nuts+0.031
ils+0.030
es+0.029
ports+0.029
Neg logit contributions
avis-0.022
 rely-0.021
 sur-0.021
 listener-0.021
 Kang-0.020
Token two
Token ID734
Feature activation-1.38e+00

Pos logit contributions
cases+0.031
ils+0.029
nuts+0.028
 Sale+0.027
 invitations+0.026
Neg logit contributions
 sur-0.045
avis-0.044
 rely-0.043
 polar-0.041
 fro-0.041
Token birds
Token ID10087
Feature activation+2.05e+00

Pos logit contributions
cases+0.051
ils+0.050
 Sale+0.049
nuts+0.047
driver+0.046
Neg logit contributions
 sur-0.034
avis-0.033
 rely-0.033
 polar-0.033
 crit-0.032
Token qu
Token ID627
Feature activation-2.96e+00

Pos logit contributions
 sur+0.050
avis+0.048
 rely+0.048
 fro+0.045
 polar+0.045
Neg logit contributions
cases-0.078
ils-0.075
 Sale-0.074
nuts-0.073
 invitations-0.072
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.120
 Sale+0.112
driver+0.111
ils+0.110
 puzzles+0.109
Neg logit contributions
avis-0.066
 sur-0.063
 Arctic-0.062
please-0.061
 fro-0.061
Tokenling
Token ID1359
Feature activation-4.89e-01

Pos logit contributions
 Sale+0.498
cases+0.479
nuts+0.462
driver+0.456
ifying+0.449
Neg logit contributions
heart-0.265
 sur-0.257
 crit-0.241
 matter-0.239
 rely-0.238
Token!
Token ID0
Feature activation-1.08e+00

Pos logit contributions
cases+0.019
 Sale+0.018
ils+0.017
ports+0.017
driver+0.017
Neg logit contributions
 fro-0.012
 sur-0.011
 crit-0.011
 polar-0.011
 bas-0.011
Token They
Token ID1119
Feature activation-1.78e+00

Pos logit contributions
cases+0.042
ils+0.042
nuts+0.040
 invitations+0.039
 Sale+0.039
Neg logit contributions
 sur-0.026
avis-0.025
 rely-0.024
 fro-0.023
please-0.022
Token were
Token ID547
Feature activation-1.24e+00

Pos logit contributions
cases+0.062
ils+0.059
 Sale+0.058
nuts+0.058
 invitations+0.057
Neg logit contributions
 sur-0.050
 rely-0.047
avis-0.047
 bas-0.046
 fro-0.046
Token squ
Token ID2809
Feature activation+6.55e-01

Pos logit contributions
cases+0.042
ils+0.040
nuts+0.039
 Sale+0.038
 invitations+0.037
Neg logit contributions
 sur-0.037
avis-0.036
 rely-0.035
please-0.034
 listener-0.033
Token Finally
Token ID9461
Feature activation+4.69e-01

Pos logit contributions
 sur+0.005
avis+0.005
 rely+0.004
please+0.004
 listener+0.004
Neg logit contributions
cases-0.007
ils-0.007
nuts-0.006
 Sale-0.006
 invitations-0.006
Token,
Token ID11
Feature activation-9.07e-01

Pos logit contributions
 sur+0.007
 day+0.007
 morning+0.006
 fro+0.006
 ask+0.006
Neg logit contributions
cases-0.023
nuts-0.022
ils-0.022
opes-0.021
driver-0.021
Token after
Token ID706
Feature activation-2.70e+00

Pos logit contributions
ils+0.032
cases+0.032
nuts+0.030
driver+0.029
 Sale+0.029
Neg logit contributions
 sur-0.026
 fro-0.024
avis-0.023
 day-0.023
 rely-0.023
Token much
Token ID881
Feature activation-1.29e+00

Pos logit contributions
cases+0.098
ils+0.094
nuts+0.088
driver+0.087
ifying+0.087
Neg logit contributions
 day-0.077
 polar-0.069
 sur-0.068
 fro-0.068
 morning-0.068
Token qu
Token ID627
Feature activation-2.89e+00

Pos logit contributions
cases+0.050
ils+0.046
nuts+0.045
ports+0.045
 Sale+0.044
Neg logit contributions
 sur-0.033
 rely-0.032
 ask-0.029
 bra-0.029
avis-0.029
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.117
driver+0.109
 puzzles+0.103
 Sale+0.102
 invitations+0.102
Neg logit contributions
avis-0.066
 fro-0.066
heart-0.064
 Arctic-0.063
 sur-0.061
Tokenling
Token ID1359
Feature activation-3.53e-01

Pos logit contributions
 Sale+0.494
cases+0.493
nuts+0.484
driver+0.470
ifying+0.463
Neg logit contributions
heart-0.265
 sur-0.253
 matter-0.243
 day-0.238
 crit-0.227
Token,
Token ID11
Feature activation-9.84e-01

Pos logit contributions
cases+0.014
nuts+0.013
 Sale+0.013
ils+0.013
 invitations+0.012
Neg logit contributions
 sur-0.008
 fro-0.008
 crit-0.008
 polar-0.007
 rely-0.007
Token one
Token ID530
Feature activation+5.71e-01

Pos logit contributions
ils+0.035
cases+0.035
nuts+0.033
driver+0.032
 Sale+0.031
Neg logit contributions
 sur-0.028
 fro-0.026
avis-0.025
 crit-0.025
 rely-0.025
Token of
Token ID286
Feature activation+2.35e+00

Pos logit contributions
avis+0.019
please+0.018
 Kang+0.017
 rely+0.017
 Arctic+0.017
Neg logit contributions
ils-0.017
 invitations-0.016
cases-0.016
nuts-0.015
 sandwiches-0.015
Token the
Token ID262
Feature activation+5.18e-01

Pos logit contributions
 sur+0.072
avis+0.070
 rely+0.067
 fro+0.066
 polar+0.066
Neg logit contributions
cases-0.084
ils-0.074
nuts-0.073
 Sale-0.070
 invitations-0.069
Token Lily
Token ID20037
Feature activation+1.07e-01

Pos logit contributions
cases+0.039
ils+0.037
nuts+0.037
 Sale+0.036
 invitations+0.035
Neg logit contributions
 sur-0.023
avis-0.023
 rely-0.022
 listener-0.021
 fro-0.021
Token and
Token ID290
Feature activation-6.57e-01

Pos logit contributions
avis+0.004
 sur+0.004
 rely+0.004
 polar+0.003
 Arctic+0.003
Neg logit contributions
ils-0.003
cases-0.003
es-0.003
nuts-0.003
 puzzles-0.002
Token Jack
Token ID3619
Feature activation-2.40e+00

Pos logit contributions
cases+0.027
nuts+0.025
ils+0.024
 Sale+0.024
driver+0.024
Neg logit contributions
 sur-0.016
avis-0.016
 fro-0.015
 ask-0.015
 listener-0.015
Token never
Token ID1239
Feature activation-7.25e-01

Pos logit contributions
cases+0.097
nuts+0.090
ils+0.089
 invitations+0.088
 Sale+0.088
Neg logit contributions
 sur-0.056
 rely-0.054
avis-0.054
please-0.051
 listener-0.050
Token qu
Token ID627
Feature activation-2.99e+00

Pos logit contributions
ils+0.023
cases+0.022
nuts+0.022
 Sale+0.021
 puzzles+0.020
Neg logit contributions
avis-0.022
 Arctic-0.020
 rely-0.020
 sur-0.020
 Kang-0.020
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.078
 Sale+0.071
nuts+0.068
ils+0.066
driver+0.066
Neg logit contributions
avis-0.107
 sur-0.105
 fro-0.105
 listener-0.103
please-0.102
Tokenled
Token ID992
Feature activation-2.38e+00

Pos logit contributions
cases+0.291
 Sale+0.276
nuts+0.275
 invitations+0.246
driver+0.244
Neg logit contributions
 sur-0.472
avis-0.455
 listener-0.453
please-0.451
 rely-0.450
Token again
Token ID757
Feature activation-2.40e+00

Pos logit contributions
cases+0.082
 Sale+0.077
ils+0.076
nuts+0.076
 invitations+0.071
Neg logit contributions
 sur-0.071
avis-0.067
 rely-0.066
 listener-0.066
please-0.064
Token.
Token ID13
Feature activation+5.03e-02

Pos logit contributions
cases+0.081
ils+0.077
nuts+0.076
 Sale+0.074
 invitations+0.073
Neg logit contributions
 sur-0.071
avis-0.070
 rely-0.068
please-0.065
 listener-0.065
Token They
Token ID1119
Feature activation-1.93e+00

Pos logit contributions
 sur+0.002
avis+0.002
 rely+0.002
please+0.002
 fro+0.002
Neg logit contributions
cases-0.001
ils-0.001
nuts-0.001
 Sale-0.001
 invitations-0.001
Token learned
Token ID4499
Feature activation-5.31e+00

Pos logit contributions
cases+0.072
nuts+0.067
ils+0.066
 Sale+0.065
 invitations+0.065
Neg logit contributions
 sur-0.052
avis-0.050
 rely-0.050
please-0.047
 listener-0.046
Token the
Token ID262
Feature activation+3.29e-01

Pos logit contributions
ils+0.018
 puzzles+0.016
cases+0.016
 invitations+0.016
nuts+0.016
Neg logit contributions
 sur-0.025
 rely-0.024
avis-0.024
please-0.023
 listener-0.022
Token other
Token ID584
Feature activation+2.94e+00

Pos logit contributions
 sur+0.013
 rely+0.012
 fro+0.012
 Fro+0.012
 pleasure+0.012
Neg logit contributions
 papers-0.008
 puzzles-0.008
 pages-0.008
ils-0.007
 files-0.007
Token kn
Token ID638
Feature activation+1.61e+00

Pos logit contributions
 sur+0.095
 rely+0.090
avis+0.087
 fro+0.087
 polar+0.085
Neg logit contributions
cases-0.093
ils-0.090
 Sale-0.084
nuts-0.084
 invitations-0.083
Tokenobs
Token ID8158
Feature activation+3.30e+00

Pos logit contributions
 fro+0.056
 sur+0.056
 rely+0.054
 matter+0.054
 polar+0.054
Neg logit contributions
cases-0.047
driver-0.042
nuts-0.041
ils-0.039
es-0.038
Token qu
Token ID627
Feature activation-3.22e+00

Pos logit contributions
 sur+0.110
avis+0.108
 rely+0.106
please+0.100
 listener+0.100
Neg logit contributions
cases-0.097
ils-0.091
nuts-0.091
 Sale-0.089
 invitations-0.086
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.123
 Sale+0.116
driver+0.113
nuts+0.113
 invitations+0.110
Neg logit contributions
avis-0.081
 sur-0.077
 Arctic-0.074
please-0.074
 fro-0.072
Tokenled
Token ID992
Feature activation-2.23e+00

Pos logit contributions
cases+0.460
 Sale+0.453
nuts+0.442
 invitations+0.419
ils+0.418
Neg logit contributions
 sur-0.300
avis-0.288
 rely-0.279
heart-0.275
please-0.270
Token over
Token ID625
Feature activation-9.32e-01

Pos logit contributions
cases+0.091
 Sale+0.086
ils+0.083
nuts+0.082
driver+0.082
Neg logit contributions
 sur-0.051
 day-0.047
 fro-0.047
 rely-0.047
avis-0.046
Token who
Token ID508
Feature activation+4.33e+00

Pos logit contributions
cases+0.035
ils+0.031
nuts+0.031
 Sale+0.030
driver+0.029
Neg logit contributions
 sur-0.027
avis-0.026
 rely-0.024
please-0.024
 listener-0.024
Token should
Token ID815
Feature activation-1.08e-01

Pos logit contributions
 sur+0.115
avis+0.114
 rely+0.108
 listener+0.104
 polar+0.102
Neg logit contributions
cases-0.157
ils-0.154
nuts-0.147
 Sale-0.145
es-0.140
Token be
Token ID307
Feature activation+2.45e+00

Pos logit contributions
cases+0.005
ils+0.004
nuts+0.004
 Sale+0.004
 invitations+0.004
Neg logit contributions
 sur-0.002
avis-0.002
 rely-0.002
please-0.002
 listener-0.002
Token but
Token ID475
Feature activation+2.01e+00

Pos logit contributions
cases+0.036
nuts+0.034
ils+0.034
 puzzles+0.033
es+0.032
Neg logit contributions
 sur-0.037
avis-0.036
 rely-0.034
please-0.033
 listener-0.032
Token today
Token ID1909
Feature activation-2.72e+00

Pos logit contributions
 sur+0.070
avis+0.067
please+0.063
 rely+0.063
 Arctic+0.061
Neg logit contributions
cases-0.056
ils-0.055
nuts-0.055
 Sale-0.054
 puzzles-0.052
Token they
Token ID484
Feature activation-2.67e+00

Pos logit contributions
cases+0.096
ils+0.092
nuts+0.090
 Sale+0.088
 invitations+0.086
Neg logit contributions
 sur-0.076
avis-0.074
 rely-0.072
please-0.069
 listener-0.068
Token were
Token ID547
Feature activation-1.28e+00

Pos logit contributions
cases+0.111
 invitations+0.108
driver+0.108
nuts+0.107
ados+0.106
Neg logit contributions
 sur-0.055
 rely-0.053
 ask-0.052
 regret-0.052
 asked-0.051
Token qu
Token ID627
Feature activation-3.08e+00

Pos logit contributions
cases+0.051
nuts+0.048
ils+0.048
 Sale+0.048
driver+0.047
Neg logit contributions
 sur-0.029
 rely-0.029
please-0.028
 listener-0.028
avis-0.028
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.085
 Sale+0.078
nuts+0.077
ils+0.076
driver+0.075
Neg logit contributions
avis-0.110
 sur-0.108
please-0.104
 Arctic-0.103
 listener-0.102
Tokenling
Token ID1359
Feature activation-1.75e-01

Pos logit contributions
cases+0.455
 Sale+0.448
nuts+0.439
 invitations+0.416
ils+0.411
Neg logit contributions
 sur-0.305
avis-0.298
 rely-0.292
please-0.286
 listener-0.285
Token.
Token ID13
Feature activation-6.12e-01

Pos logit contributions
cases+0.007
 Sale+0.006
nuts+0.006
ils+0.006
driver+0.006
Neg logit contributions
 day-0.004
 morning-0.004
heart-0.004
 rely-0.004
 listener-0.004
Token They
Token ID1119
Feature activation-2.03e+00

Pos logit contributions
cases+0.023
ils+0.022
nuts+0.022
 Sale+0.021
 invitations+0.021
Neg logit contributions
 sur-0.016
avis-0.015
 rely-0.015
please-0.014
 listener-0.014
Token were
Token ID547
Feature activation-1.26e+00

Pos logit contributions
cases+0.084
nuts+0.080
driver+0.079
 invitations+0.078
ils+0.078
Neg logit contributions
 sur-0.043
avis-0.041
 rely-0.041
please-0.040
 ask-0.040
Token arguing
Token ID11810
Feature activation-5.00e+00

Pos logit contributions
cases+0.048
ils+0.045
nuts+0.045
 Sale+0.044
driver+0.043
Neg logit contributions
 sur-0.031
 rely-0.031
avis-0.031
please-0.030
 listener-0.029
Token,
Token ID11
Feature activation-9.40e-01

Pos logit contributions
cases+0.040
ils+0.039
nuts+0.038
 Sale+0.037
 invitations+0.036
Neg logit contributions
 sur-0.031
avis-0.030
 rely-0.029
 listener-0.028
please-0.027
Token why
Token ID1521
Feature activation+2.74e+00

Pos logit contributions
cases+0.037
ils+0.035
nuts+0.035
 Sale+0.034
 invitations+0.033
Neg logit contributions
 sur-0.023
avis-0.022
 rely-0.022
please-0.020
 listener-0.020
Token are
Token ID389
Feature activation-8.39e-01

Pos logit contributions
 sur+0.068
 rely+0.065
avis+0.063
 ask+0.061
 fro+0.061
Neg logit contributions
cases-0.105
ils-0.102
nuts-0.100
 invitations-0.099
 Sale-0.095
Token you
Token ID345
Feature activation-4.46e-01

Pos logit contributions
cases+0.037
ils+0.035
driver+0.034
es+0.033
 pages+0.033
Neg logit contributions
 sur-0.016
avis-0.015
 fro-0.015
 Kang-0.014
 ask-0.014
Token qu
Token ID627
Feature activation-3.03e+00

Pos logit contributions
cases+0.018
ils+0.017
nuts+0.017
 Sale+0.016
 invitations+0.016
Neg logit contributions
 sur-0.011
 rely-0.010
avis-0.010
please-0.010
 fro-0.010
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.082
ils+0.072
 Sale+0.071
driver+0.070
nuts+0.069
Neg logit contributions
avis-0.111
 sur-0.110
please-0.107
 fro-0.106
 rely-0.105
Tokenling
Token ID1359
Feature activation-3.38e-01

Pos logit contributions
cases+0.471
 Sale+0.462
nuts+0.441
driver+0.423
 invitations+0.422
Neg logit contributions
 sur-0.291
heart-0.288
 rely-0.278
 matter-0.269
avis-0.268
Token?"
Token ID1701
Feature activation-9.89e-01

Pos logit contributions
cases+0.015
ils+0.014
nuts+0.013
 Sale+0.013
 invitations+0.013
Neg logit contributions
 sur-0.007
 rely-0.007
avis-0.006
 polar-0.006
 fro-0.006
Token But
Token ID887
Feature activation+2.88e+00

Pos logit contributions
cases+0.033
nuts+0.031
 Sale+0.031
ils+0.030
es+0.029
Neg logit contributions
 sur-0.028
avis-0.028
 listener-0.027
 rely-0.027
please-0.026
Token the
Token ID262
Feature activation+4.83e-01

Pos logit contributions
 sur+0.091
avis+0.089
 rely+0.087
 listener+0.083
please+0.083
Neg logit contributions
cases-0.091
ils-0.087
nuts-0.085
 Sale-0.083
 invitations-0.082
Token birds
Token ID10087
Feature activation+2.04e+00

Pos logit contributions
 sur+0.011
avis+0.011
 rely+0.011
please+0.010
 listener+0.010
Neg logit contributions
cases-0.019
ils-0.019
nuts-0.018
 Sale-0.018
 invitations-0.018
Token other
Token ID584
Feature activation+3.12e+00

Pos logit contributions
 sur+0.009
avis+0.009
 rely+0.008
 listener+0.008
please+0.008
Neg logit contributions
cases-0.013
nuts-0.013
ils-0.013
 Sale-0.012
 invitations-0.012
Token.
Token ID13
Feature activation-3.00e-01

Pos logit contributions
 sur+0.076
 rely+0.074
avis+0.073
 listener+0.069
 fro+0.069
Neg logit contributions
cases-0.115
ils-0.111
 Sale-0.110
nuts-0.109
 invitations-0.107
Token They
Token ID1119
Feature activation-2.03e+00

Pos logit contributions
cases+0.011
nuts+0.011
ils+0.011
 Sale+0.011
 invitations+0.010
Neg logit contributions
 sur-0.008
avis-0.008
 rely-0.007
please-0.007
 listener-0.007
Token starting
Token ID3599
Feature activation+4.13e+00

Pos logit contributions
ados+0.082
cases+0.082
nuts+0.081
 invitations+0.080
 sandwiches+0.080
Neg logit contributions
 ask-0.043
 asked-0.042
 regret-0.042
 sur-0.040
 check-0.038
Token qu
Token ID627
Feature activation-2.98e+00

Pos logit contributions
 sur+0.109
 rely+0.103
 check+0.097
avis+0.097
please+0.096
Neg logit contributions
cases-0.161
ils-0.158
nuts-0.149
ports-0.145
 Sale-0.143
Tokenarre
Token ID9624
Feature activation-1.23e+01

Pos logit contributions
cases+0.081
 Sale+0.075
nuts+0.073
driver+0.070
 invitations+0.070
Neg logit contributions
avis-0.104
 Arctic-0.099
heart-0.099
 sur-0.098
 fro-0.097
Tokenling
Token ID1359
Feature activation-7.63e-02

Pos logit contributions
 Sale+0.428
nuts+0.412
cases+0.409
ifying+0.385
 invitations+0.381
Neg logit contributions
heart-0.327
 sur-0.319
 matter-0.312
avis-0.300
 rely-0.299
Token again
Token ID757
Feature activation-2.39e+00

Pos logit contributions
cases+0.003
 Sale+0.003
ils+0.003
nuts+0.003
ports+0.003
Neg logit contributions
 sur-0.002
 fro-0.002
 day-0.002
 bra-0.002
 crit-0.002
Token and
Token ID290
Feature activation-6.09e-01

Pos logit contributions
cases+0.086
ils+0.082
nuts+0.080
 Sale+0.079
 invitations+0.077
Neg logit contributions
 sur-0.067
 rely-0.064
avis-0.063
 listener-0.060
please-0.060
Token their
Token ID511
Feature activation-2.13e+00

Pos logit contributions
cases+0.019
ils+0.018
nuts+0.018
 Sale+0.017
 invitations+0.017
Neg logit contributions
 sur-0.019
avis-0.019
 rely-0.018
please-0.018
 listener-0.017
Token voices
Token ID10839
Feature activation-6.98e-01

Pos logit contributions
cases+0.080
ils+0.076
nuts+0.076
 Sale+0.073
 invitations+0.073
Neg logit contributions
 sur-0.055
avis-0.054
 rely-0.052
please-0.049
 listener-0.049
Token never
Token ID1239
Feature activation-7.03e-01

Pos logit contributions
 sur+0.028
avis+0.027
 rely+0.027
please+0.026
 listener+0.025
Neg logit contributions
cases-0.033
ils-0.031
nuts-0.031
 Sale-0.030
 invitations-0.029
Token happy
Token ID3772
Feature activation-7.10e-01

Pos logit contributions
cases+0.026
ils+0.025
nuts+0.024
 invitations+0.024
 Sale+0.023
Neg logit contributions
 sur-0.018
avis-0.018
 rely-0.017
 listener-0.016
 Kang-0.016
Token.
Token ID13
Feature activation+6.03e-02

Pos logit contributions
cases+0.021
ils+0.020
nuts+0.020
 invitations+0.019
 Sale+0.019
Neg logit contributions
avis-0.024
 sur-0.024
 rely-0.023
please-0.022
 listener-0.021
Token He
Token ID679
Feature activation-3.42e-01

Pos logit contributions
 sur+0.002
 bas+0.002
please+0.002
 fro+0.002
 matter+0.002
Neg logit contributions
 Sale-0.002
cases-0.002
es-0.002
ports-0.002
icks-0.002
Token qu
Token ID627
Feature activation-2.99e+00

Pos logit contributions
ils+0.012
cases+0.012
nuts+0.011
es+0.011
 Sale+0.011
Neg logit contributions
avis-0.010
 sur-0.009
 rely-0.009
 listener-0.009
please-0.009
Tokenarre
Token ID9624
Feature activation-1.22e+01

Pos logit contributions
cases+0.110
nuts+0.100
 Sale+0.099
driver+0.099
 invitations+0.096
Neg logit contributions
avis-0.077
 Arctic-0.074
 fro-0.073
 sur-0.072
heart-0.071
Tokenled
Token ID992
Feature activation-2.21e+00

Pos logit contributions
cases+0.310
 Sale+0.297
nuts+0.295
 invitations+0.265
driver+0.261
Neg logit contributions
 sur-0.445
avis-0.434
 rely-0.428
heart-0.427
 listener-0.420
Token and
Token ID290
Feature activation-3.19e-01

Pos logit contributions
cases+0.071
ils+0.068
nuts+0.067
 Sale+0.064
 invitations+0.064
Neg logit contributions
 sur-0.069
avis-0.068
 rely-0.066
please-0.063
 listener-0.062
Token yelled
Token ID22760
Feature activation-2.61e-01

Pos logit contributions
cases+0.012
ils+0.012
nuts+0.011
 Sale+0.011
 invitations+0.011
Neg logit contributions
avis-0.008
 sur-0.008
 rely-0.008
 listener-0.007
please-0.007
Token and
Token ID290
Feature activation-3.47e-01

Pos logit contributions
cases+0.011
nuts+0.011
 Sale+0.010
ils+0.010
ports+0.010
Neg logit contributions
 sur-0.005
 bra-0.005
 Are-0.005
 listener-0.005
 polar-0.005
Token sc
Token ID629
Feature activation+1.57e+00

Pos logit contributions
cases+0.013
ils+0.013
nuts+0.013
 Sale+0.012
 invitations+0.012
Neg logit contributions
avis-0.009
 sur-0.009
 rely-0.008
 listener-0.008
 Arctic-0.008
Token.
Token ID13
Feature activation+2.59e-01

Pos logit contributions
 Sale+0.140
icks+0.131
ils+0.130
 invitations+0.128
ports+0.127
Neg logit contributions
 sur-0.057
 rely-0.055
 day-0.054
 place-0.052
 ask-0.052
Token But
Token ID887
Feature activation+2.92e+00

Pos logit contributions
 sur+0.008
avis+0.008
 rely+0.008
please+0.007
 listener+0.007
Neg logit contributions
cases-0.008
ils-0.008
nuts-0.008
 Sale-0.008
 invitations-0.008
Token they
Token ID484
Feature activation-2.32e+00

Pos logit contributions
 sur+0.080
 fro+0.078
avis+0.077
 rely+0.076
 ask+0.075
Neg logit contributions
cases-0.101
ils-0.098
nuts-0.095
 invitations-0.092
 Sale-0.092
Token were
Token ID547
Feature activation-1.05e+00

Pos logit contributions
cases+0.093
 invitations+0.090
ils+0.090
 Sale+0.089
nuts+0.089
Neg logit contributions
 sur-0.053
avis-0.050
 rely-0.048
please-0.046
 ask-0.046
Token sad
Token ID6507
Feature activation-1.22e+00

Pos logit contributions
cases+0.043
 Sale+0.040
ils+0.039
ports+0.039
nuts+0.039
Neg logit contributions
 bra-0.023
 rely-0.023
 sur-0.023
 listener-0.022
 cos-0.022
Token when
Token ID618
Feature activation+4.67e+00

Pos logit contributions
cases+0.042
ils+0.040
nuts+0.040
 Sale+0.039
 invitations+0.038
Neg logit contributions
 sur-0.035
avis-0.035
 rely-0.033
 listener-0.032
please-0.031
Token they
Token ID484
Feature activation-2.35e+00

Pos logit contributions
 sur+0.141
avis+0.139
 fro+0.133
 rely+0.132
 bas+0.132
Neg logit contributions
cases-0.153
ils-0.151
nuts-0.145
driver-0.141
 Sale-0.135
Token saw
Token ID2497
Feature activation-7.25e-01

Pos logit contributions
cases+0.101
ils+0.097
nuts+0.096
 Sale+0.094
 invitations+0.093
Neg logit contributions
 sur-0.048
avis-0.046
 rely-0.044
please-0.042
 fro-0.041
Token the
Token ID262
Feature activation+5.87e-01

Pos logit contributions
cases+0.024
nuts+0.021
driver+0.021
ils+0.021
 Sale+0.020
Neg logit contributions
avis-0.021
 rely-0.020
 sur-0.020
 Sk-0.020
 polar-0.020
Token prices
Token ID4536
Feature activation-4.10e+00

Pos logit contributions
 sur+0.019
avis+0.018
 rely+0.018
please+0.017
 listener+0.017
Neg logit contributions
cases-0.018
ils-0.018
nuts-0.017
 Sale-0.017
 invitations-0.016
Token.
Token ID13
Feature activation+3.38e-01

Pos logit contributions
ils+0.164
 Sale+0.158
opes+0.154
 invitations+0.153
ados+0.153
Neg logit contributions
 sur-0.089
 day-0.086
 boy-0.078
avis-0.078
 place-0.076
Token.
Token ID13
Feature activation-7.28e-02

Pos logit contributions
 sur+0.000
avis+0.000
 rely+0.000
 fro+0.000
 Arctic+0.000
Neg logit contributions
cases-0.001
ils-0.001
 Sale-0.001
nuts-0.001
driver-0.001
Token Maybe
Token ID6674
Feature activation+1.10e+00

Pos logit contributions
cases+0.003
ils+0.002
nuts+0.002
 Sale+0.002
 invitations+0.002
Neg logit contributions
 sur-0.002
avis-0.002
 rely-0.002
please-0.002
 listener-0.002
Token one
Token ID530
Feature activation+4.99e-01

Pos logit contributions
 sur+0.035
avis+0.034
 rely+0.033
please+0.032
 listener+0.032
Neg logit contributions
cases-0.034
ils-0.033
nuts-0.032
 Sale-0.031
 invitations-0.031
Token day
Token ID1110
Feature activation+9.86e-01

Pos logit contributions
avis+0.011
please+0.010
 rely+0.010
 sur+0.009
ancer+0.009
Neg logit contributions
ils-0.020
 Sale-0.018
 invitations-0.018
 all-0.018
 puzzles-0.018
Token she
Token ID673
Feature activation+1.36e-01

Pos logit contributions
 sur+0.025
avis+0.023
 rely+0.023
 fro+0.022
 ask+0.022
Neg logit contributions
cases-0.037
nuts-0.035
ils-0.035
 invitations-0.034
driver-0.034
Token could
Token ID714
Feature activation+5.33e+00

Pos logit contributions
avis+0.003
 sur+0.003
 rely+0.003
 listener+0.003
 Kang+0.003
Neg logit contributions
cases-0.005
ils-0.005
nuts-0.005
 Sale-0.005
es-0.005
Token come
Token ID1282
Feature activation+3.30e+00

Pos logit contributions
avis+0.120
 sur+0.119
 listener+0.107
 rely+0.106
please+0.106
Neg logit contributions
cases-0.213
nuts-0.206
ils-0.204
 Sale-0.199
 invitations-0.197
Token back
Token ID736
Feature activation-1.15e+00

Pos logit contributions
 sur+0.100
 rely+0.098
avis+0.096
please+0.094
 fro+0.093
Neg logit contributions
cases-0.110
ils-0.109
 Sale-0.102
nuts-0.100
 invitations-0.095
Token and
Token ID290
Feature activation-6.66e-01

Pos logit contributions
cases+0.040
ils+0.038
nuts+0.038
 Sale+0.037
 invitations+0.036
Neg logit contributions
 sur-0.033
avis-0.032
 rely-0.031
please-0.030
 listener-0.030
Token tie
Token ID9839
Feature activation+8.62e-01

Pos logit contributions
cases+0.026
nuts+0.025
ils+0.025
 invitations+0.024
 Sale+0.024
Neg logit contributions
avis-0.016
 sur-0.015
 listener-0.014
 Arctic-0.013
 Kang-0.013
Token a
Token ID257
Feature activation-9.47e-01

Pos logit contributions
 fro+0.027
avis+0.027
 sur+0.027
 rely+0.026
 bra+0.026
Neg logit contributions
cases-0.027
ils-0.024
 Sale-0.024
nuts-0.023
driver-0.022
Token a
Token ID257
Feature activation-1.16e+00

Pos logit contributions
cases+0.101
ils+0.095
nuts+0.095
 Sale+0.090
 invitations+0.088
Neg logit contributions
avis-0.079
 sur-0.079
 rely-0.077
 fro-0.072
 Arctic-0.072
Token Tuesday
Token ID3431
Feature activation-8.57e-01

Pos logit contributions
cases+0.044
ils+0.042
nuts+0.041
 puzzles+0.041
 Sale+0.040
Neg logit contributions
 sur-0.029
avis-0.028
 rely-0.026
 Kang-0.025
 listener-0.025
Token morning
Token ID3329
Feature activation+1.31e+00

Pos logit contributions
cases+0.024
 Sale+0.021
nuts+0.020
ils+0.020
 invitations+0.020
Neg logit contributions
 rely-0.031
avis-0.031
 sur-0.030
please-0.030
 listener-0.029
Token,
Token ID11
Feature activation-1.48e+00

Pos logit contributions
 sur+0.037
avis+0.036
 rely+0.034
 fro+0.033
 listener+0.033
Neg logit contributions
cases-0.046
ils-0.044
nuts-0.044
 Sale-0.042
 invitations-0.041
Token the
Token ID262
Feature activation+3.28e-01

Pos logit contributions
cases+0.048
ils+0.046
nuts+0.046
 Sale+0.044
 invitations+0.043
Neg logit contributions
 sur-0.045
avis-0.044
 rely-0.043
 listener-0.041
 fro-0.041
Token sun
Token ID4252
Feature activation+4.16e+00

Pos logit contributions
 rely+0.008
avis+0.008
please+0.008
 sur+0.007
 listener+0.007
Neg logit contributions
cases-0.012
nuts-0.011
 invitations-0.011
 sandwiches-0.011
 Sale-0.011
Token was
Token ID373
Feature activation+6.54e-01

Pos logit contributions
 sur+0.119
 rely+0.117
avis+0.115
please+0.110
 pleasure+0.108
Neg logit contributions
cases-0.140
ils-0.132
es-0.131
nuts-0.125
 invitations-0.124
Token glowing
Token ID21377
Feature activation-1.34e-01

Pos logit contributions
 sur+0.020
avis+0.019
 rely+0.019
 listener+0.018
 Kang+0.018
Neg logit contributions
cases-0.021
nuts-0.020
 invitations-0.020
ils-0.020
 Sale-0.019
Token brightly
Token ID35254
Feature activation+7.15e-01

Pos logit contributions
cases+0.004
ils+0.004
 Sale+0.004
nuts+0.004
driver+0.004
Neg logit contributions
 sur-0.004
avis-0.004
 rely-0.004
 fro-0.004
please-0.004
Token.
Token ID13
Feature activation-1.03e+00

Pos logit contributions
 sur+0.025
avis+0.024
 rely+0.024
please+0.023
 listener+0.023
Neg logit contributions
cases-0.020
ils-0.019
nuts-0.019
 Sale-0.018
 invitations-0.018
Token Sally
Token ID25737
Feature activation-1.95e+00

Pos logit contributions
cases+0.031
 Sale+0.029
nuts+0.029
ils+0.028
 invitations+0.028
Neg logit contributions
 sur-0.033
avis-0.033
 rely-0.032
please-0.031
 listener-0.030
Token on
Token ID319
Feature activation-3.50e+00

Pos logit contributions
 day+0.042
 sur+0.042
avis+0.041
 morning+0.041
 listener+0.041
Neg logit contributions
nuts-0.035
cases-0.035
ils-0.032
 invitations-0.032
ades-0.030
Token,
Token ID11
Feature activation-9.13e-01

Pos logit contributions
cases+0.156
nuts+0.156
ils+0.150
driver+0.143
 invitations+0.142
Neg logit contributions
 day-0.059
 sur-0.058
 rely-0.056
 morning-0.056
avis-0.056
Token they
Token ID484
Feature activation-2.55e+00

Pos logit contributions
cases+0.034
ils+0.033
nuts+0.032
 Sale+0.031
 invitations+0.031
Neg logit contributions
 sur-0.024
avis-0.023
 rely-0.022
please-0.021
 listener-0.021
Token knew
Token ID2993
Feature activation-1.88e+00

Pos logit contributions
cases+0.103
ils+0.098
nuts+0.097
 Sale+0.095
 invitations+0.094
Neg logit contributions
 sur-0.059
avis-0.057
 rely-0.055
please-0.052
 listener-0.052
Token they
Token ID484
Feature activation-2.55e+00

Pos logit contributions
cases+0.045
ils+0.042
nuts+0.041
 Sale+0.039
 invitations+0.038
Neg logit contributions
 sur-0.075
avis-0.074
 rely-0.072
please-0.070
 listener-0.069
Token could
Token ID714
Feature activation+5.39e+00

Pos logit contributions
cases+0.120
nuts+0.115
 invitations+0.114
 Sale+0.114
ados+0.111
Neg logit contributions
 rely-0.040
 sur-0.039
avis-0.036
please-0.036
 listener-0.035
Token trust
Token ID3774
Feature activation+1.89e+00

Pos logit contributions
 sur+0.130
avis+0.128
 listener+0.117
 crit+0.114
 polar+0.114
Neg logit contributions
nuts-0.203
 Sale-0.198
cases-0.197
ils-0.195
 puzzles-0.187
Token each
Token ID1123
Feature activation+2.55e-01

Pos logit contributions
 sur+0.073
avis+0.073
 rely+0.071
please+0.068
 fro+0.068
Neg logit contributions
cases-0.047
nuts-0.042
ils-0.041
 Sale-0.040
 invitations-0.039
Token other
Token ID584
Feature activation+3.26e+00

Pos logit contributions
 sur+0.009
avis+0.009
 rely+0.008
please+0.008
 Arctic+0.008
Neg logit contributions
ils-0.007
cases-0.007
nuts-0.007
 invitations-0.007
 Sale-0.006
Token no
Token ID645
Feature activation-4.03e-01

Pos logit contributions
 sur+0.078
 rely+0.076
 listener+0.073
avis+0.071
 day+0.071
Neg logit contributions
cases-0.121
 Sale-0.119
nuts-0.117
ils-0.115
 invitations-0.114
Token matter
Token ID2300
Feature activation+2.67e+00

Pos logit contributions
 puzzles+0.014
ils+0.013
 all+0.013
 sandwiches+0.013
 invitations+0.013
Neg logit contributions
 Arctic-0.010
 sur-0.010
isc-0.010
 polar-0.010
avis-0.010
Token on
Token ID319
Feature activation-3.52e+00

Pos logit contributions
 sur+0.056
avis+0.056
 rely+0.055
 listener+0.053
please+0.053
Neg logit contributions
cases-0.029
ils-0.027
nuts-0.026
 invitations-0.025
 Sale-0.025
Token,
Token ID11
Feature activation-8.10e-01

Pos logit contributions
cases+0.128
ils+0.121
nuts+0.121
 Sale+0.117
 invitations+0.116
Neg logit contributions
avis-0.096
 sur-0.096
 rely-0.092
 polar-0.087
please-0.086
Token they
Token ID484
Feature activation-2.33e+00

Pos logit contributions
 all+0.023
es+0.021
ls+0.020
ils+0.020
nuts+0.020
Neg logit contributions
 sur-0.029
avis-0.028
 rely-0.025
 listener-0.025
raid-0.025
Token knew
Token ID2993
Feature activation-1.72e+00

Pos logit contributions
cases+0.091
ils+0.087
nuts+0.087
 Sale+0.084
 invitations+0.083
Neg logit contributions
 sur-0.056
avis-0.055
 rely-0.053
please-0.050
 listener-0.049
Token they
Token ID484
Feature activation-2.32e+00

Pos logit contributions
cases+0.034
ils+0.032
nuts+0.032
 Sale+0.030
 invitations+0.029
Neg logit contributions
 sur-0.073
avis-0.072
 rely-0.070
please-0.069
 listener-0.068
Token could
Token ID714
Feature activation+5.55e+00

Pos logit contributions
cases+0.110
nuts+0.104
 invitations+0.103
 Sale+0.103
ados+0.100
Neg logit contributions
 rely-0.037
 sur-0.036
 listener-0.034
avis-0.033
please-0.033
Token always
Token ID1464
Feature activation+4.75e+00

Pos logit contributions
 sur+0.131
avis+0.128
 listener+0.118
 crit+0.113
 Are+0.112
Neg logit contributions
nuts-0.212
 Sale-0.208
ils-0.205
cases-0.201
 puzzles-0.195
Token help
Token ID1037
Feature activation+2.21e+00

Pos logit contributions
avis+0.136
 sur+0.124
heart+0.119
 listener+0.116
ancer+0.113
Neg logit contributions
nuts-0.160
ils-0.159
 Sale-0.156
 puzzles-0.152
cases-0.151
Token each
Token ID1123
Feature activation+4.26e-01

Pos logit contributions
 sur+0.122
avis+0.115
 beacon+0.114
 pleasure+0.113
 listener+0.113
Neg logit contributions
ils-0.019
 puzzles-0.013
 all-0.012
nuts-0.011
es-0.011
Token other
Token ID584
Feature activation+3.30e+00

Pos logit contributions
 rely+0.015
 sur+0.015
avis+0.014
 pleasure+0.014
 Are+0.014
Neg logit contributions
ils-0.011
 invitations-0.011
cases-0.011
 puzzles-0.011
 Sale-0.010
Token when
Token ID618
Feature activation+4.78e+00

Pos logit contributions
 sur+0.094
avis+0.092
 rely+0.091
 listener+0.085
please+0.085
Neg logit contributions
cases-0.114
ils-0.109
nuts-0.108
 Sale-0.106
 invitations-0.103
Token take
Token ID1011
Feature activation-1.11e+00

Pos logit contributions
avis+0.083
 sur+0.082
 listener+0.074
please+0.073
 Kang+0.072
Neg logit contributions
cases-0.140
nuts-0.138
ils-0.137
 Sale-0.132
 invitations-0.130
Token in
Token ID287
Feature activation+1.85e-01

Pos logit contributions
cases+0.036
ils+0.035
nuts+0.034
 Sale+0.033
 invitations+0.032
Neg logit contributions
 sur-0.034
avis-0.034
 rely-0.032
please-0.031
 listener-0.031
Token oxygen
Token ID11863
Feature activation+1.42e+00

Pos logit contributions
 rely+0.005
 sur+0.005
avis+0.005
 pleasure+0.005
 polar+0.005
Neg logit contributions
cases-0.007
ils-0.006
nuts-0.006
 Sale-0.006
ports-0.006
Token for
Token ID329
Feature activation-3.27e+00

Pos logit contributions
 sur+0.039
 rely+0.038
avis+0.036
 fro+0.035
 listener+0.035
Neg logit contributions
cases-0.051
ils-0.048
 Sale-0.047
nuts-0.047
 invitations-0.045
Token all
Token ID477
Feature activation-7.84e-01

Pos logit contributions
cases+0.108
nuts+0.097
ils+0.096
driver+0.091
 invitations+0.091
Neg logit contributions
 rely-0.101
 sur-0.101
avis-0.101
please-0.095
 polar-0.094
Token the
Token ID262
Feature activation+3.90e-01

Pos logit contributions
cases+0.025
ils+0.025
nuts+0.024
 Sale+0.023
 invitations+0.023
Neg logit contributions
 sur-0.024
avis-0.024
 rely-0.023
please-0.022
 listener-0.022
Token animals
Token ID4695
Feature activation+2.12e+00

Pos logit contributions
avis+0.012
 sur+0.012
 Kang+0.010
please+0.010
 fro+0.010
Neg logit contributions
ils-0.013
 puzzles-0.012
nuts-0.012
cases-0.012
 Sale-0.012
Token to
Token ID284
Feature activation-1.09e-01

Pos logit contributions
 sur+0.064
 rely+0.062
avis+0.061
please+0.058
 listener+0.058
Neg logit contributions
cases-0.069
ils-0.064
nuts-0.064
 Sale-0.064
 invitations-0.062
Token breathe
Token ID18044
Feature activation-1.28e-01

Pos logit contributions
cases+0.004
ils+0.004
nuts+0.004
 Sale+0.004
 invitations+0.004
Neg logit contributions
 sur-0.002
avis-0.002
 rely-0.002
 listener-0.002
please-0.002
Token.
Token ID13
Feature activation-6.71e-02

Pos logit contributions
cases+0.005
 Sale+0.005
nuts+0.005
ils+0.004
driver+0.004
Neg logit contributions
 rely-0.003
 sur-0.003
 place-0.003
 cos-0.003
 bra-0.003
Token<|endoftext|>
Token ID50256
Feature activation-5.31e+00

Pos logit contributions
 Sale+0.002
cases+0.002
nuts+0.002
ils+0.002
es+0.002
Neg logit contributions
 sur-0.002
avis-0.002
please-0.002
 listener-0.002
 rely-0.002
Token mixed
Token ID7668
Feature activation-1.25e+00

Pos logit contributions
cases+0.007
ils+0.007
nuts+0.006
 Sale+0.006
 invitations+0.006
Neg logit contributions
 sur-0.004
avis-0.004
 rely-0.004
 listener-0.004
please-0.004
Token some
Token ID617
Feature activation-3.12e+00

Pos logit contributions
cases+0.040
 Sale+0.036
ils+0.036
nuts+0.035
driver+0.033
Neg logit contributions
 sur-0.039
avis-0.039
 fro-0.037
 rely-0.037
please-0.037
Token ingredients
Token ID9391
Feature activation+1.51e-01

Pos logit contributions
cases+0.113
 Sale+0.104
ils+0.103
nuts+0.102
driver+0.099
Neg logit contributions
 sur-0.084
avis-0.079
 rely-0.079
 crit-0.078
 fro-0.078
Token together
Token ID1978
Feature activation-2.65e+00

Pos logit contributions
 sur+0.004
 rely+0.004
avis+0.004
please+0.003
 day+0.003
Neg logit contributions
cases-0.005
 Sale-0.005
ils-0.005
ports-0.005
 invitations-0.005
Token and
Token ID290
Feature activation-8.01e-01

Pos logit contributions
cases+0.102
 Sale+0.095
ils+0.095
 invitations+0.091
nuts+0.091
Neg logit contributions
 sur-0.063
 rely-0.062
avis-0.059
please-0.057
 day-0.056
Token made
Token ID925
Feature activation+9.21e-01

Pos logit contributions
cases+0.029
ils+0.028
nuts+0.028
 Sale+0.027
 invitations+0.026
Neg logit contributions
 sur-0.021
avis-0.021
 rely-0.020
please-0.019
 listener-0.019
Token a
Token ID257
Feature activation-8.03e-01

Pos logit contributions
 sur+0.034
avis+0.034
 rely+0.033
please+0.032
 listener+0.032
Neg logit contributions
cases-0.024
ils-0.022
nuts-0.022
 Sale-0.021
 invitations-0.020
Token y
Token ID331
Feature activation-1.12e+00

Pos logit contributions
cases+0.026
ils+0.024
nuts+0.024
 Sale+0.023
 invitations+0.023
Neg logit contributions
 sur-0.025
avis-0.025
 rely-0.024
please-0.023
 listener-0.023
Tokenummy
Token ID13513
Feature activation-1.28e-01

Pos logit contributions
cases+0.037
ils+0.034
nuts+0.034
 Sale+0.033
 invitations+0.032
Neg logit contributions
 sur-0.034
avis-0.034
 rely-0.032
please-0.031
 listener-0.031
Token shrimp
Token ID30060
Feature activation-2.19e+00

Pos logit contributions
cases+0.004
ils+0.004
nuts+0.004
 Sale+0.004
driver+0.004
Neg logit contributions
 sur-0.004
avis-0.004
 rely-0.004
 fro-0.003
 polar-0.003
Token dish
Token ID9433
Feature activation-1.33e+00

Pos logit contributions
cases+0.084
 Sale+0.080
ports+0.077
ils+0.075
nuts+0.072
Neg logit contributions
 sur-0.052
 rely-0.051
 fro-0.049
 clim-0.049
 polar-0.047
Token You
Token ID921
Feature activation-2.02e+00

Pos logit contributions
cases+0.021
 Sale+0.020
driver+0.020
ils+0.019
nuts+0.019
Neg logit contributions
 sur-0.022
 rely-0.021
please-0.021
avis-0.021
 bas-0.021
Token have
Token ID423
Feature activation-1.54e+00

Pos logit contributions
cases+0.063
nuts+0.061
ils+0.060
 invitations+0.058
 Sale+0.057
Neg logit contributions
 sur-0.063
 rely-0.061
avis-0.060
 ask-0.059
 fro-0.058
Token to
Token ID284
Feature activation+4.77e-02

Pos logit contributions
cases+0.059
nuts+0.056
ils+0.054
fs+0.052
ados+0.051
Neg logit contributions
 sur-0.038
avis-0.038
 check-0.037
please-0.036
 place-0.036
Token stay
Token ID2652
Feature activation+3.27e+00

Pos logit contributions
 sur+0.001
avis+0.001
please+0.001
 rely+0.001
 listener+0.001
Neg logit contributions
cases-0.002
ils-0.002
nuts-0.002
 Sale-0.002
 invitations-0.002
Token hidden
Token ID7104
Feature activation-7.73e-01

Pos logit contributions
 sur+0.101
avis+0.097
 rely+0.097
 polar+0.092
 ask+0.091
Neg logit contributions
cases-0.109
nuts-0.104
ils-0.102
 Sale-0.100
driver-0.094
Token until
Token ID1566
Feature activation+1.43e+00

Pos logit contributions
cases+0.030
nuts+0.028
 Sale+0.028
ils+0.027
 invitations+0.027
Neg logit contributions
 sur-0.019
 place-0.018
avis-0.017
 clim-0.017
 polar-0.017
Token I
Token ID314
Feature activation+2.74e+00

Pos logit contributions
 sur+0.037
avis+0.036
 rely+0.036
 ask+0.034
 fro+0.034
Neg logit contributions
cases-0.054
ils-0.052
nuts-0.051
 invitations-0.048
driver-0.048
Token find
Token ID1064
Feature activation+1.16e+00

Pos logit contributions
avis+0.081
please+0.075
 sur+0.074
heart+0.074
 Kang+0.072
Neg logit contributions
cases-0.093
ils-0.088
 Sale-0.085
es-0.082
nuts-0.082
Token you
Token ID345
Feature activation-4.76e-01

Pos logit contributions
avis+0.032
 sur+0.032
 check+0.031
 polar+0.030
 rely+0.030
Neg logit contributions
cases-0.041
nuts-0.038
ils-0.037
 Sale-0.035
driver-0.035
Token!"
Token ID2474
Feature activation-1.84e+00

Pos logit contributions
cases+0.016
nuts+0.016
ils+0.016
 Sale+0.015
 invitations+0.015
Neg logit contributions
 rely-0.013
 sur-0.013
 fro-0.012
avis-0.012
 bas-0.012
Token He
Token ID679
Feature activation-8.68e-02

Pos logit contributions
cases+0.069
driver+0.062
 Sale+0.062
 invitations+0.061
es+0.060
Neg logit contributions
 rely-0.048
 sur-0.045
please-0.045
 bas-0.045
avis-0.045
Token and
Token ID290
Feature activation+3.48e-01

Pos logit contributions
cases+0.020
ils+0.018
nuts+0.018
 Sale+0.018
 invitations+0.017
Neg logit contributions
avis-0.020
 sur-0.020
 fro-0.019
 rely-0.019
 ask-0.019
Token smile
Token ID8212
Feature activation-1.61e+00

Pos logit contributions
 sur+0.006
 rely+0.006
avis+0.006
 fro+0.005
 ask+0.005
Neg logit contributions
cases-0.016
ils-0.015
nuts-0.015
 Sale-0.015
 invitations-0.014
Token.
Token ID13
Feature activation+1.07e+00

Pos logit contributions
 Sale+0.069
cases+0.069
nuts+0.068
ils+0.068
 invitations+0.065
Neg logit contributions
 sur-0.029
 listener-0.028
 ask-0.028
 rely-0.027
 day-0.027
Token Then
Token ID3244
Feature activation+4.84e+00

Pos logit contributions
 sur+0.037
avis+0.036
 rely+0.035
please+0.034
 listener+0.034
Neg logit contributions
cases-0.031
ils-0.029
nuts-0.029
 Sale-0.028
 invitations-0.028
Token they
Token ID484
Feature activation-1.96e+00

Pos logit contributions
 sur+0.161
avis+0.157
 rely+0.153
 ask+0.150
 fro+0.149
Neg logit contributions
cases-0.147
nuts-0.139
ils-0.137
 Sale-0.129
driver-0.129
Token start
Token ID923
Feature activation+8.84e-01

Pos logit contributions
cases+0.090
nuts+0.085
ados+0.084
ils+0.083
 Sale+0.083
Neg logit contributions
 sur-0.037
 ask-0.035
 rely-0.035
 check-0.034
avis-0.031
Token to
Token ID284
Feature activation+2.50e-01

Pos logit contributions
 sur+0.021
 check+0.020
avis+0.020
 day+0.020
 visit+0.020
Neg logit contributions
cases-0.035
ils-0.034
nuts-0.033
 Sale-0.031
ados-0.031
Token pick
Token ID2298
Feature activation+2.77e-01

Pos logit contributions
 sur+0.004
avis+0.004
 rely+0.004
 listener+0.004
please+0.004
Neg logit contributions
cases-0.011
ils-0.011
nuts-0.011
 Sale-0.011
 invitations-0.011
Token up
Token ID510
Feature activation+2.50e-01

Pos logit contributions
 sur+0.008
avis+0.008
 fro+0.007
 crit+0.007
 bra+0.007
Neg logit contributions
cases-0.010
 Sale-0.009
nuts-0.009
ils-0.009
driver-0.008
Token the
Token ID262
Feature activation+1.06e+00

Pos logit contributions
 fro+0.005
 boy+0.005
 L+0.005
 bra+0.005
 girl+0.005
Neg logit contributions
cases-0.010
 Sale-0.010
driver-0.010
nuts-0.010
ils-0.010
Token blocks
Token ID7021
Feature activation-2.90e+00

Pos logit contributions
 boy+0.039
 day+0.037
 girl+0.037
 sur+0.036
 rely+0.036
Neg logit contributions
ils-0.030
cases-0.028
 Sale-0.026
es-0.024
nuts-0.024
Token caught
Token ID4978
Feature activation-1.98e+00

Pos logit contributions
 sur+0.012
avis+0.012
 rely+0.012
 listener+0.011
please+0.011
Neg logit contributions
cases-0.014
ils-0.014
nuts-0.013
 Sale-0.013
 invitations-0.013
Token.
Token ID13
Feature activation-6.53e-03

Pos logit contributions
cases+0.075
ils+0.066
 Sale+0.065
nuts+0.065
 invitations+0.064
Neg logit contributions
 sur-0.054
avis-0.052
 rely-0.051
 polar-0.049
please-0.049
Token
Token ID220
Feature activation-2.22e-01

Pos logit contributions
 Sale+0.000
cases+0.000
driver+0.000
nuts+0.000
ifying+0.000
Neg logit contributions
 sur-0.000
avis-0.000
 bas-0.000
 listener-0.000
please-0.000
Token
Token ID198
Feature activation+8.16e-01

Pos logit contributions
cases+0.008
nuts+0.008
ils+0.008
 Sale+0.008
 invitations+0.007
Neg logit contributions
 sur-0.006
avis-0.006
 rely-0.005
 listener-0.005
please-0.005
Token
Token ID198
Feature activation+8.50e-01

Pos logit contributions
 sur+0.016
avis+0.016
 rely+0.015
 listener+0.014
please+0.014
Neg logit contributions
cases-0.035
ils-0.034
nuts-0.034
 Sale-0.033
 invitations-0.032
TokenThe
Token ID464
Feature activation+1.09e+00

Pos logit contributions
avis+0.022
 sur+0.021
 rely+0.020
please+0.019
 Arctic+0.018
Neg logit contributions
cases-0.033
ils-0.032
 invitations-0.031
nuts-0.031
ados-0.030
Token puppy
Token ID26188
Feature activation-2.32e-01

Pos logit contributions
 sur+0.025
avis+0.025
 rely+0.024
please+0.023
 listener+0.022
Neg logit contributions
cases-0.044
ils-0.042
nuts-0.042
 Sale-0.040
 invitations-0.040
Token's
Token ID338
Feature activation+1.96e+00

Pos logit contributions
cases+0.009
ils+0.009
nuts+0.009
 Sale+0.009
 invitations+0.009
Neg logit contributions
 sur-0.005
avis-0.005
 rely-0.005
please-0.005
 listener-0.005
Token owners
Token ID4393
Feature activation+5.46e-01

Pos logit contributions
 sur+0.057
avis+0.054
please+0.051
 rely+0.050
 listener+0.049
Neg logit contributions
cases-0.065
ils-0.064
nuts-0.064
 Sale-0.061
 invitations-0.061
Token laughed
Token ID13818
Feature activation-3.93e+00

Pos logit contributions
 sur+0.015
 rely+0.014
avis+0.014
please+0.013
 listener+0.013
Neg logit contributions
cases-0.019
ils-0.018
nuts-0.018
 Sale-0.018
 invitations-0.018
Token and
Token ID290
Feature activation-7.72e-01

Pos logit contributions
 Sale+0.164
ils+0.162
cases+0.161
nuts+0.154
ports+0.150
Neg logit contributions
 sur-0.076
 bra-0.076
heart-0.071
 rely-0.070
 pleasure-0.070
Token she
Token ID673
Feature activation-1.49e-02

Pos logit contributions
cases+0.059
ils+0.057
nuts+0.057
 Sale+0.055
 invitations+0.053
Neg logit contributions
avis-0.031
 sur-0.030
 fro-0.028
 rely-0.027
 listener-0.026
Token heard
Token ID2982
Feature activation-5.39e+00

Pos logit contributions
cases+0.001
ils+0.001
nuts+0.001
 Sale+0.000
 invitations+0.000
Neg logit contributions
 sur-0.000
avis-0.000
 rely-0.000
 listener-0.000
 Kang-0.000
Token a
Token ID257
Feature activation-1.10e+00

Pos logit contributions
cases+0.166
ils+0.162
nuts+0.161
 Sale+0.153
driver+0.148
Neg logit contributions
avis-0.176
 fro-0.164
 sur-0.164
 polar-0.162
 bra-0.161
Token loud
Token ID7812
Feature activation-2.88e+00

Pos logit contributions
cases+0.028
nuts+0.026
 Sale+0.025
ils+0.025
 invitations+0.025
Neg logit contributions
 rely-0.040
 sur-0.039
avis-0.039
 listener-0.039
 pleasure-0.038
Token bark
Token ID21405
Feature activation-5.56e-02

Pos logit contributions
cases+0.107
ils+0.102
nuts+0.102
 Sale+0.099
 invitations+0.097
Neg logit contributions
 sur-0.073
avis-0.071
 rely-0.070
 listener-0.066
please-0.066
Token from
Token ID422
Feature activation-2.01e+00

Pos logit contributions
cases+0.002
ils+0.002
 Sale+0.002
nuts+0.002
 invitations+0.002
Neg logit contributions
 sur-0.001
avis-0.001
 polar-0.001
 fro-0.001
 rely-0.001
Token her
Token ID607
Feature activation+4.32e-01

Pos logit contributions
cases+0.051
ils+0.051
nuts+0.049
 invitations+0.048
 puzzles+0.047
Neg logit contributions
avis-0.072
 sur-0.071
 rely-0.069
please-0.067
 listener-0.066
Token dog
Token ID3290
Feature activation-7.23e-01

Pos logit contributions
 sur+0.013
avis+0.012
 rely+0.012
please+0.012
 listener+0.012
Neg logit contributions
cases-0.014
nuts-0.014
ils-0.013
 sandwiches-0.013
 invitations-0.013
Token,
Token ID11
Feature activation-1.35e+00

Pos logit contributions
cases+0.023
ils+0.022
nuts+0.022
 Sale+0.021
 invitations+0.021
Neg logit contributions
 sur-0.023
avis-0.022
 rely-0.021
 listener-0.021
please-0.020
Token Max
Token ID5436
Feature activation-1.96e+00

Pos logit contributions
cases+0.045
nuts+0.043
es+0.040
 Sale+0.039
ils+0.039
Neg logit contributions
 sur-0.040
 listener-0.039
avis-0.038
 rely-0.036
 pleasure-0.035
Token.
Token ID13
Feature activation-5.11e-01

Pos logit contributions
cases+0.049
nuts+0.048
es+0.047
ils+0.047
driver+0.043
Neg logit contributions
 sur-0.073
avis-0.070
 listener-0.068
 pleasure-0.067
please-0.067
Token a
Token ID257
Feature activation-1.16e+00

Pos logit contributions
cases+0.030
nuts+0.028
ils+0.028
 Sale+0.026
 invitations+0.025
Neg logit contributions
 sur-0.039
avis-0.038
 rely-0.037
please-0.036
 fro-0.036
Token walk
Token ID2513
Feature activation+1.44e-01

Pos logit contributions
cases+0.046
ils+0.043
nuts+0.043
 invitations+0.042
 Sale+0.042
Neg logit contributions
 sur-0.028
avis-0.027
 rely-0.026
please-0.025
 listener-0.024
Token in
Token ID287
Feature activation-1.87e-01

Pos logit contributions
 sur+0.004
avis+0.004
 rely+0.004
 listener+0.004
please+0.004
Neg logit contributions
cases-0.005
ils-0.005
nuts-0.004
 Sale-0.004
 invitations-0.004
Token the
Token ID262
Feature activation+2.22e-01

Pos logit contributions
cases+0.004
 Sale+0.004
 puzzles+0.004
nuts+0.004
 invitations+0.004
Neg logit contributions
 sur-0.007
avis-0.007
please-0.007
 rely-0.007
 listener-0.007
Token grass
Token ID8701
Feature activation+4.07e+00

Pos logit contributions
 sur+0.005
 rely+0.005
avis+0.005
please+0.005
 listener+0.005
Neg logit contributions
cases-0.009
ils-0.008
 Sale-0.008
nuts-0.008
 invitations-0.008
Token,
Token ID11
Feature activation-1.52e+00

Pos logit contributions
avis+0.133
 sur+0.133
 rely+0.132
please+0.125
 listener+0.122
Neg logit contributions
cases-0.125
nuts-0.117
ils-0.110
es-0.107
 Sale-0.106
Token he
Token ID339
Feature activation-2.59e-01

Pos logit contributions
cases+0.061
ils+0.058
nuts+0.058
 Sale+0.056
 invitations+0.055
Neg logit contributions
 sur-0.035
avis-0.034
 rely-0.033
please-0.031
 listener-0.031
Token found
Token ID1043
Feature activation-1.65e+00

Pos logit contributions
cases+0.009
ils+0.008
es+0.008
nuts+0.008
ports+0.008
Neg logit contributions
 rely-0.008
 pleasure-0.008
avis-0.007
 sur-0.007
 listener-0.007
Token a
Token ID257
Feature activation-1.11e+00

Pos logit contributions
cases+0.036
ils+0.034
nuts+0.033
 Sale+0.031
 invitations+0.031
Neg logit contributions
 sur-0.068
avis-0.067
 rely-0.066
please-0.063
 listener-0.063
Token pen
Token ID3112
Feature activation-1.19e+00

Pos logit contributions
cases+0.031
nuts+0.030
 puzzles+0.029
 invitations+0.028
 sandwiches+0.028
Neg logit contributions
avis-0.037
 sur-0.035
 rely-0.035
please-0.034
 listener-0.033
Token and
Token ID290
Feature activation-9.04e-01

Pos logit contributions
cases+0.039
 Sale+0.038
ils+0.037
ports+0.037
 invitations+0.036
Neg logit contributions
 sur-0.034
 rely-0.033
 clim-0.033
 polar-0.032
avis-0.032
Token proud
Token ID6613
Feature activation-1.83e+00

Pos logit contributions
avis+0.032
 sur+0.031
 rely+0.029
please+0.029
 listener+0.028
Neg logit contributions
cases-0.066
ils-0.064
nuts-0.062
 invitations-0.061
 Sale-0.061
Token!
Token ID0
Feature activation-1.23e+00

Pos logit contributions
cases+0.047
ils+0.045
nuts+0.044
 invitations+0.042
es+0.042
Neg logit contributions
avis-0.068
 rely-0.067
 sur-0.065
please-0.065
 Kang-0.063
Token
Token ID198
Feature activation+7.01e-01

Pos logit contributions
cases+0.040
ils+0.038
nuts+0.037
 Sale+0.036
 invitations+0.035
Neg logit contributions
 sur-0.039
avis-0.038
 rely-0.036
 listener-0.035
please-0.035
Token
Token ID198
Feature activation+7.36e-01

Pos logit contributions
 sur+0.014
avis+0.014
 rely+0.013
please+0.013
 listener+0.012
Neg logit contributions
cases-0.030
ils-0.029
nuts-0.028
 Sale-0.028
 invitations-0.027
TokenWhen
Token ID2215
Feature activation+1.91e+00

Pos logit contributions
 sur+0.017
please+0.015
avis+0.015
 rely+0.015
Yes+0.015
Neg logit contributions
cases-0.031
ados-0.029
 invitations-0.029
nuts-0.028
ils-0.028
Token they
Token ID484
Feature activation-2.51e+00

Pos logit contributions
 sur+0.063
avis+0.058
 day+0.057
 polar+0.057
 fro+0.056
Neg logit contributions
cases-0.064
nuts-0.060
ils-0.059
driver-0.058
 invitations-0.053
Token reached
Token ID4251
Feature activation+1.70e+00

Pos logit contributions
cases+0.098
nuts+0.093
ils+0.090
driver+0.090
ados+0.089
Neg logit contributions
 sur-0.064
avis-0.058
 rely-0.056
 fro-0.055
 polar-0.055
Token the
Token ID262
Feature activation+6.47e-01

Pos logit contributions
 sur+0.049
avis+0.047
 rely+0.047
 polar+0.046
 listener+0.045
Neg logit contributions
cases-0.059
nuts-0.056
ils-0.056
 Sale-0.052
driver-0.051
Token entrance
Token ID10384
Feature activation+8.17e-01

Pos logit contributions
 sur+0.018
 place+0.017
 rely+0.016
avis+0.016
 listener+0.016
Neg logit contributions
cases-0.024
ils-0.023
nuts-0.022
driver-0.021
es-0.021
Token of
Token ID286
Feature activation+2.54e+00

Pos logit contributions
 sur+0.024
avis+0.023
 rely+0.023
 listener+0.022
please+0.021
Neg logit contributions
cases-0.028
ils-0.026
nuts-0.026
 Sale-0.025
 invitations-0.025
Token the
Token ID262
Feature activation+6.36e-01

Pos logit contributions
 sur+0.067
 polar+0.065
 place+0.065
 check+0.062
 rely+0.061
Neg logit contributions
cases-0.103
ils-0.094
nuts-0.091
driver-0.085
es-0.085
Token and
Token ID290
Feature activation-7.63e-01

Pos logit contributions
cases+0.073
 Sale+0.070
ils+0.070
nuts+0.068
 invitations+0.067
Neg logit contributions
 sur-0.061
 rely-0.059
avis-0.058
 listener-0.056
 bas-0.056
Token sang
Token ID25889
Feature activation-2.99e-01

Pos logit contributions
cases+0.026
ils+0.025
nuts+0.024
 Sale+0.023
 invitations+0.023
Neg logit contributions
 sur-0.023
avis-0.022
 rely-0.021
 listener-0.021
please-0.020
Token as
Token ID355
Feature activation+4.54e+00

Pos logit contributions
cases+0.010
nuts+0.010
ils+0.010
 Sale+0.010
driver+0.009
Neg logit contributions
 sur-0.008
avis-0.008
 fro-0.008
 rely-0.008
please-0.008
Token loud
Token ID7812
Feature activation-2.90e+00

Pos logit contributions
 sur+0.158
avis+0.156
 rely+0.151
 listener+0.146
please+0.145
Neg logit contributions
cases-0.128
ils-0.122
nuts-0.121
 Sale-0.117
 invitations-0.113
Token as
Token ID355
Feature activation+4.55e+00

Pos logit contributions
cases+0.103
ils+0.098
nuts+0.097
 Sale+0.094
 invitations+0.093
Neg logit contributions
 sur-0.080
avis-0.077
 rely-0.075
please-0.072
 polar-0.071
Token they
Token ID484
Feature activation-2.22e+00

Pos logit contributions
 sur+0.159
avis+0.157
 rely+0.153
please+0.148
 listener+0.147
Neg logit contributions
cases-0.127
ils-0.119
nuts-0.118
 Sale-0.113
 invitations-0.111
Token could
Token ID714
Feature activation+5.44e+00

Pos logit contributions
cases+0.084
ils+0.081
nuts+0.079
 Sale+0.076
 invitations+0.075
Neg logit contributions
 sur-0.056
avis-0.055
 rely-0.053
 listener-0.050
 Arctic-0.049
Token,
Token ID11
Feature activation-1.01e+00

Pos logit contributions
 sur+0.152
avis+0.149
 rely+0.148
please+0.139
 fro+0.137
Neg logit contributions
cases-0.190
ils-0.185
nuts-0.180
 Sale-0.176
 invitations-0.173
Token but
Token ID475
Feature activation+2.06e+00

Pos logit contributions
cases+0.037
nuts+0.035
ils+0.034
 invitations+0.033
 puzzles+0.033
Neg logit contributions
avis-0.027
 sur-0.027
 rely-0.025
 listener-0.025
please-0.024
Token Billy
Token ID15890
Feature activation-1.48e+00

Pos logit contributions
 sur+0.066
avis+0.065
 rely+0.063
please+0.060
 listener+0.060
Neg logit contributions
cases-0.065
ils-0.061
nuts-0.060
 Sale-0.058
 invitations-0.057
Token was
Token ID373
Feature activation+6.58e-01

Pos logit contributions
cases+0.053
ils+0.050
es+0.048
nuts+0.047
driver+0.044
Neg logit contributions
 sur-0.041
avis-0.040
 pleasure-0.039
 rely-0.038
 listener-0.037
Token faster
Token ID5443
Feature activation+1.04e+00

Pos logit contributions
cases+0.006
ils+0.006
nuts+0.006
 Sale+0.006
 invitations+0.006
Neg logit contributions
 sur-0.005
avis-0.005
 rely-0.004
please-0.004
 listener-0.004
Token.
Token ID13
Feature activation+1.02e+00

Pos logit contributions
 sur+0.030
avis+0.029
 rely+0.028
 listener+0.027
please+0.027
Neg logit contributions
cases-0.036
ils-0.034
nuts-0.033
 Sale-0.033
 invitations-0.032
Token The
Token ID383
Feature activation+7.90e-01

Pos logit contributions
 sur+0.027
avis+0.026
 rely+0.026
 listener+0.025
please+0.025
Neg logit contributions
cases-0.037
nuts-0.035
 Sale-0.035
ils-0.034
 invitations-0.033
Token dog
Token ID3290
Feature activation+9.32e-02

Pos logit contributions
 sur+0.020
avis+0.019
 rely+0.019
please+0.018
 listener+0.018
Neg logit contributions
cases-0.030
ils-0.029
nuts-0.028
 Sale-0.028
 invitations-0.027
Token ch
Token ID442
Feature activation+3.37e+00

Pos logit contributions
 sur+0.002
avis+0.002
 rely+0.002
 fro+0.002
please+0.002
Neg logit contributions
cases-0.004
nuts-0.003
ils-0.003
 invitations-0.003
 Sale-0.003
Tokenases
Token ID1386
Feature activation-1.61e+00

Pos logit contributions
 sur+0.117
avis+0.112
 rely+0.108
 fro+0.107
 listener+0.105
Neg logit contributions
cases-0.103
nuts-0.093
ils-0.091
 Sale-0.089
 invitations-0.089
Token them
Token ID606
Feature activation-5.01e-02

Pos logit contributions
cases+0.057
nuts+0.054
ils+0.053
 invitations+0.051
 Sale+0.051
Neg logit contributions
 sur-0.043
avis-0.042
 fro-0.041
 ask-0.040
 bra-0.040
Token and
Token ID290
Feature activation-2.43e-01

Pos logit contributions
cases+0.002
 Sale+0.002
ils+0.002
ports+0.002
nuts+0.002
Neg logit contributions
 sur-0.001
 rely-0.001
 fro-0.001
 clim-0.001
 bra-0.001
Token tries
Token ID8404
Feature activation+2.03e+00

Pos logit contributions
cases+0.009
ils+0.009
nuts+0.009
 Sale+0.009
 invitations+0.009
Neg logit contributions
 sur-0.006
avis-0.006
 rely-0.006
 fro-0.006
 polar-0.005
Token to
Token ID284
Feature activation+3.31e-01

Pos logit contributions
 sur+0.058
 rely+0.054
avis+0.054
 fro+0.053
 ask+0.052
Neg logit contributions
cases-0.072
ils-0.069
nuts-0.065
 Sale-0.064
 invitations-0.064
Token bite
Token ID13197
Feature activation-2.68e+00

Pos logit contributions
 sur+0.009
avis+0.009
 rely+0.008
 listener+0.008
please+0.008
Neg logit contributions
cases-0.012
ils-0.012
nuts-0.012
 Sale-0.011
 invitations-0.011
Token help
Token ID1037
Feature activation+2.08e+00

Pos logit contributions
cases+0.029
nuts+0.029
 invitations+0.029
ils+0.028
 Sale+0.028
Neg logit contributions
 sur-0.025
avis-0.025
please-0.022
 Kang-0.022
 Arctic-0.022
Token.
Token ID13
Feature activation+6.79e-02

Pos logit contributions
 sur+0.071
avis+0.069
 rely+0.066
please+0.065
 Arctic+0.064
Neg logit contributions
cases-0.060
nuts-0.058
ils-0.058
 invitations-0.054
 Sale-0.054
Token
Token ID198
Feature activation+9.52e-01

Pos logit contributions
 sur+0.002
avis+0.002
 polar+0.002
 rely+0.002
please+0.002
Neg logit contributions
cases-0.002
 Sale-0.002
nuts-0.002
es-0.002
 invitations-0.002
Token
Token ID198
Feature activation+9.84e-01

Pos logit contributions
 sur+0.023
avis+0.021
 listener+0.021
 rely+0.020
please+0.020
Neg logit contributions
cases-0.036
 Sale-0.035
ils-0.035
nuts-0.035
es-0.034
TokenThe
Token ID464
Feature activation+1.06e+00

Pos logit contributions
 sur+0.028
avis+0.027
 rely+0.026
please+0.026
 listener+0.024
Neg logit contributions
cases-0.036
ils-0.034
nuts-0.033
 invitations-0.032
 Sale-0.031
Token moral
Token ID6573
Feature activation-8.31e+00

Pos logit contributions
 sur+0.024
avis+0.023
 rely+0.022
please+0.022
 Kang+0.020
Neg logit contributions
cases-0.042
ils-0.042
 invitations-0.041
nuts-0.041
 puzzles-0.040
Token of
Token ID286
Feature activation+2.49e+00

Pos logit contributions
cases+0.265
ils+0.257
 Sale+0.254
nuts+0.248
 invitations+0.244
Neg logit contributions
 sur-0.251
avis-0.250
please-0.239
 rely-0.239
 polar-0.233
Token the
Token ID262
Feature activation+4.56e-01

Pos logit contributions
 sur+0.092
avis+0.089
 rely+0.087
please+0.085
 listener+0.085
Neg logit contributions
cases-0.064
ils-0.063
nuts-0.060
 Sale-0.059
 invitations-0.058
Token story
Token ID1621
Feature activation-6.38e+00

Pos logit contributions
 sur+0.015
avis+0.015
 rely+0.014
 listener+0.014
please+0.014
Neg logit contributions
cases-0.014
ils-0.013
nuts-0.013
 Sale-0.012
 invitations-0.012
Token is
Token ID318
Feature activation+8.48e-01

Pos logit contributions
cases+0.285
nuts+0.284
ils+0.269
ados+0.266
 sandwiches+0.265
Neg logit contributions
 day-0.111
 rely-0.108
 ask-0.103
 place-0.101
 sur-0.100
Token that
Token ID326
Feature activation+2.16e+00

Pos logit contributions
 sur+0.034
avis+0.033
 rely+0.033
 listener+0.032
please+0.031
Neg logit contributions
cases-0.020
ils-0.018
nuts-0.018
 Sale-0.017
 invitations-0.017
Token in
Token ID287
Feature activation+4.20e-01

Pos logit contributions
 Sale+0.000
ils+0.000
opes+0.000
cases+0.000
nuts+0.000
Neg logit contributions
 day-0.000
 place-0.000
 morning-0.000
 check-0.000
 night-0.000
Token different
Token ID1180
Feature activation-4.80e-01

Pos logit contributions
avis+0.011
 place+0.011
 rely+0.011
 check+0.010
 sur+0.010
Neg logit contributions
cases-0.016
ils-0.015
nuts-0.015
 Sale-0.014
driver-0.013
Token rooms
Token ID9519
Feature activation+3.98e-01

Pos logit contributions
cases+0.014
nuts+0.014
ils+0.013
 Sale+0.013
driver+0.012
Neg logit contributions
 sur-0.015
avis-0.015
 rely-0.015
 Arctic-0.014
 ask-0.014
Token and
Token ID290
Feature activation-2.97e-02

Pos logit contributions
 sur+0.008
avis+0.008
 rely+0.008
 ask+0.008
 place+0.008
Neg logit contributions
ils-0.015
nuts-0.015
 Sale-0.015
cases-0.015
ados-0.014
Token makes
Token ID1838
Feature activation+1.99e+00

Pos logit contributions
nuts+0.001
cases+0.001
ils+0.001
 Sale+0.001
ados+0.001
Neg logit contributions
 ask-0.001
 rely-0.001
 sur-0.001
 check-0.001
avis-0.001
Token stories
Token ID3923
Feature activation-7.13e+00

Pos logit contributions
avis+0.069
 sur+0.068
 rely+0.065
 fro+0.064
 polar+0.063
Neg logit contributions
cases-0.055
nuts-0.053
ils-0.052
 Sale-0.051
driver-0.048
Token for
Token ID329
Feature activation-3.09e+00

Pos logit contributions
ils+0.265
 Sale+0.259
nuts+0.255
cases+0.254
ports+0.251
Neg logit contributions
 rely-0.163
 sur-0.159
 bra-0.153
avis-0.151
 place-0.151
Token her
Token ID607
Feature activation+7.88e-01

Pos logit contributions
cases+0.107
nuts+0.099
ils+0.094
driver+0.094
 Sale+0.091
Neg logit contributions
avis-0.091
 sur-0.084
 rely-0.084
 fro-0.082
please-0.082
Token dolls
Token ID36062
Feature activation-6.95e-01

Pos logit contributions
 sur+0.025
avis+0.025
 rely+0.024
please+0.023
 listener+0.022
Neg logit contributions
cases-0.025
ils-0.023
nuts-0.023
 invitations-0.022
 puzzles-0.022
Token.
Token ID13
Feature activation+5.71e-01

Pos logit contributions
 Sale+0.026
cases+0.026
nuts+0.026
ils+0.026
ports+0.025
Neg logit contributions
 rely-0.016
 sur-0.015
 ask-0.015
avis-0.015
 place-0.014
Token
Token ID198
Feature activation+1.20e+00

Pos logit contributions
 sur+0.020
 rely+0.019
 matter+0.018
 bas+0.018
avis+0.018
Neg logit contributions
cases-0.016
 Sale-0.015
es-0.015
driver-0.015
 puzzles-0.014
Token favorite
Token ID4004
Feature activation+1.85e-01

Pos logit contributions
 sur+0.105
 Kang+0.098
 Arctic+0.097
 polar+0.097
heart+0.097
Neg logit contributions
 invitations-0.089
cases-0.088
 puzzles-0.087
 papers-0.083
ils-0.083
Token thing
Token ID1517
Feature activation+1.19e-01

Pos logit contributions
 sur+0.007
avis+0.007
 rely+0.007
 Kang+0.007
please+0.006
Neg logit contributions
ils-0.004
 puzzles-0.004
 invitations-0.004
 pages-0.004
cases-0.004
Token to
Token ID284
Feature activation-3.81e-01

Pos logit contributions
 sur+0.003
avis+0.003
 rely+0.003
please+0.003
 listener+0.003
Neg logit contributions
cases-0.004
ils-0.004
nuts-0.004
 Sale-0.004
 invitations-0.004
Token write
Token ID3551
Feature activation-1.35e+00

Pos logit contributions
cases+0.013
ils+0.012
 puzzles+0.012
 Sale+0.012
 invitations+0.012
Neg logit contributions
 sur-0.011
 Kang-0.010
avis-0.010
 listener-0.010
 Arctic-0.010
Token was
Token ID373
Feature activation+5.97e-01

Pos logit contributions
 Sale+0.042
cases+0.041
nuts+0.041
ils+0.038
ifying+0.038
Neg logit contributions
 rely-0.037
 bra-0.037
avis-0.036
 day-0.036
 ask-0.036
Token stories
Token ID3923
Feature activation-7.56e+00

Pos logit contributions
 sur+0.021
avis+0.021
 rely+0.020
please+0.019
 listener+0.019
Neg logit contributions
cases-0.017
ils-0.016
nuts-0.015
 Sale-0.015
 invitations-0.015
Token about
Token ID546
Feature activation-6.44e-01

Pos logit contributions
cases+0.249
ils+0.248
nuts+0.241
 Sale+0.239
ports+0.229
Neg logit contributions
 sur-0.211
 rely-0.208
avis-0.201
 listener-0.195
please-0.193
Token sailing
Token ID28499
Feature activation+2.75e+00

Pos logit contributions
cases+0.021
ils+0.019
nuts+0.019
 Sale+0.018
 invitations+0.018
Neg logit contributions
 sur-0.020
avis-0.020
 rely-0.019
please-0.019
 fro-0.018
Token.
Token ID13
Feature activation-1.05e+00

Pos logit contributions
 place+0.078
 day+0.078
 rely+0.077
 sur+0.076
 fro+0.076
Neg logit contributions
cases-0.092
nuts-0.083
 Sale-0.083
ils-0.079
driver-0.078
Token Every
Token ID3887
Feature activation+3.33e+00

Pos logit contributions
 Sale+0.024
 magazines+0.023
 puzzles+0.022
cases+0.021
ados+0.020
Neg logit contributions
 sur-0.045
 polar-0.040
 rely-0.039
 bas-0.039
rim-0.038
Token day
Token ID1110
Feature activation+7.53e-01

Pos logit contributions
 sur+0.058
please+0.057
avis+0.055
 rely+0.054
 crit+0.054
Neg logit contributions
ils-0.137
 Sale-0.136
 puzzles-0.136
 invitations-0.135
 pages-0.135
Token happy
Token ID3772
Feature activation-3.43e-01

Pos logit contributions
cases+0.003
ils+0.003
 invitations+0.003
es+0.003
nuts+0.003
Neg logit contributions
avis-0.003
 sur-0.002
 rely-0.002
 listener-0.002
 Kang-0.002
Token.
Token ID13
Feature activation+8.97e-01

Pos logit contributions
cases+0.012
ils+0.012
nuts+0.011
 Sale+0.011
 invitations+0.011
Neg logit contributions
 sur-0.010
avis-0.010
 rely-0.009
 listener-0.009
please-0.009
Token
Token ID198
Feature activation+1.38e+00

Pos logit contributions
 rely+0.043
 matter+0.042
heart+0.042
 bas+0.042
 sur+0.042
Neg logit contributions
cases-0.013
 Sale-0.011
ports-0.010
icks-0.010
driver-0.009
Token
Token ID198
Feature activation+1.38e+00

Pos logit contributions
 sur+0.037
avis+0.035
 listener+0.034
 rely+0.034
please+0.033
Neg logit contributions
cases-0.049
 Sale-0.047
ils-0.047
nuts-0.046
es-0.045
TokenThe
Token ID464
Feature activation+1.09e+00

Pos logit contributions
 sur+0.050
avis+0.049
 rely+0.048
please+0.046
 listener+0.046
Neg logit contributions
cases-0.037
ils-0.035
nuts-0.034
 Sale-0.033
 invitations-0.033
Token moral
Token ID6573
Feature activation-7.97e+00

Pos logit contributions
 sur+0.031
avis+0.030
 rely+0.029
 listener+0.028
 polar+0.028
Neg logit contributions
cases-0.039
ils-0.038
nuts-0.037
 Sale-0.036
 invitations-0.035
Token of
Token ID286
Feature activation+2.56e+00

Pos logit contributions
cases+0.270
ils+0.259
 Sale+0.255
nuts+0.253
 invitations+0.247
Neg logit contributions
avis-0.229
 sur-0.226
 rely-0.218
please-0.214
 fro-0.210
Token the
Token ID262
Feature activation+6.02e-01

Pos logit contributions
 sur+0.090
avis+0.088
 rely+0.086
please+0.083
 listener+0.083
Neg logit contributions
cases-0.070
ils-0.068
nuts-0.067
 Sale-0.065
 invitations-0.064
Token story
Token ID1621
Feature activation-6.16e+00

Pos logit contributions
 sur+0.018
avis+0.017
 rely+0.016
 polar+0.016
 listener+0.016
Neg logit contributions
cases-0.022
ils-0.020
nuts-0.020
driver-0.019
es-0.019
Token is
Token ID318
Feature activation+1.06e+00

Pos logit contributions
nuts+0.292
cases+0.288
ils+0.282
ados+0.270
 invitations+0.269
Neg logit contributions
 day-0.109
 sur-0.093
 place-0.093
 ask-0.090
 But-0.089
Token:
Token ID25
Feature activation-4.18e-01

Pos logit contributions
 sur+0.045
avis+0.044
 rely+0.044
 ask+0.044
 listener+0.042
Neg logit contributions
cases-0.025
ils-0.021
nuts-0.020
driver-0.019
 invitations-0.018
Token away
Token ID1497
Feature activation-3.77e+00

Pos logit contributions
cases+0.047
 Sale+0.045
ils+0.045
nuts+0.044
 invitations+0.044
Neg logit contributions
 sur-0.037
 rely-0.036
 fro-0.034
 polar-0.034
avis-0.034
Token.
Token ID13
Feature activation+2.30e-02

Pos logit contributions
cases+0.136
ils+0.129
nuts+0.127
 Sale+0.126
 invitations+0.124
Neg logit contributions
 sur-0.104
 rely-0.098
avis-0.096
 polar-0.093
 listener-0.092
Token
Token ID198
Feature activation+9.38e-01

Pos logit contributions
avis+0.001
 sur+0.001
 rely+0.001
 listener+0.001
 matter+0.001
Neg logit contributions
cases-0.001
 Sale-0.001
nuts-0.001
ils-0.001
ados-0.001
Token
Token ID198
Feature activation+1.01e+00

Pos logit contributions
avis+0.023
 sur+0.023
 listener+0.022
 rely+0.021
 matter+0.021
Neg logit contributions
cases-0.034
 Sale-0.033
nuts-0.033
ils-0.032
es-0.032
TokenThe
Token ID464
Feature activation+9.81e-01

Pos logit contributions
 sur+0.026
avis+0.024
 rely+0.024
please+0.024
 Kang+0.023
Neg logit contributions
cases-0.040
ils-0.038
 invitations-0.037
nuts-0.036
 Sale-0.035
Token moral
Token ID6573
Feature activation-8.34e+00

Pos logit contributions
 sur+0.025
avis+0.022
 rely+0.021
 polar+0.021
 fro+0.021
Neg logit contributions
cases-0.040
ils-0.038
nuts-0.038
 Sale-0.036
driver-0.036
Token of
Token ID286
Feature activation+2.41e+00

Pos logit contributions
cases+0.275
ils+0.264
 Sale+0.261
nuts+0.261
 invitations+0.248
Neg logit contributions
avis-0.245
 sur-0.243
 rely-0.235
please-0.231
 fro-0.224
Token the
Token ID262
Feature activation+3.76e-01

Pos logit contributions
 sur+0.088
avis+0.086
 rely+0.084
please+0.082
 listener+0.082
Neg logit contributions
cases-0.061
ils-0.061
nuts-0.059
 Sale-0.058
 invitations-0.056
Token story
Token ID1621
Feature activation-6.40e+00

Pos logit contributions
 sur+0.012
avis+0.011
 rely+0.011
 polar+0.010
 fro+0.010
Neg logit contributions
cases-0.013
nuts-0.012
ils-0.012
driver-0.011
 invitations-0.011
Token is
Token ID318
Feature activation+8.65e-01

Pos logit contributions
cases+0.291
nuts+0.290
 invitations+0.284
ils+0.281
ados+0.275
Neg logit contributions
 day-0.101
 sur-0.097
 rely-0.092
 place-0.091
 ask-0.091
Token that
Token ID326
Feature activation+2.04e+00

Pos logit contributions
 sur+0.033
 rely+0.032
avis+0.032
 listener+0.030
 ask+0.030
Neg logit contributions
cases-0.024
ils-0.021
nuts-0.021
 invitations-0.020
 Sale-0.019