TOP ACTIVATIONS
MAX = 9.593

looked through the ey ep iece and saw the tiny scales
looked through the ey ep iece . He saw the tiny
voice that said : " Warning ! It is not safe
. It says : " Warning : Do not touch .
sign that said : " Warning ! There is bad news
A voice said , " Warning , warning , bomb activated
she always wore her fan ci est dress to the museum
dressed up in her fan ci est clothes and asked her
to put on her fan ci est sunglasses .
the daughter found the fan ci est dress she had ever
a seat in the motor bike behind him . He waved
a man on a motor bike ! The goat
will buy you the fan ci est dress in the store
the brav est and boun ci est bunny in the whole
new home with the fan ci est king in the land
by saying , " Mar riage is when two people who
. " Mar riage is when two people love
cat said , " Mar riage is when two animals love
, be careful ! Sp icy food can make your tum
. " Sp icy is good , but not

MIN ACTIVATIONS
MIN = -7.071

The popcorn was sticky and gre asy . They used a
bigger and the trees looked gre ener . Finally
much or it will look gre asy ."
had grown to be even gre ener and brighter than before
she had blue eyes and blonde hair . One
much better ! They were gre ener and taller than before
it again , this time gre ated and louder . The
't take them all . Care fully , it picked out
They are too salty and gre asy . You should eat
cars of different colors and sizes . He likes to make
to follow the lights . Care fully she opened the door
, holding a doll with blonde hair . "
ie got an idea . Care fully , she placed the
trains of different colors and sizes . He makes a track
cars of different colors and sizes . He likes to make
years old and had beautiful blonde hair . One
saw a big tree . Care fully the tire opened the
different colors , shapes and sizes . Everyone was amazed at
her rain boots and rain coat so she wouldn 't get
A small girl with curly blonde hair was standing there .

INTERVAL 5.427 - 9.593
CONTAINS 0.196%

wanted to go see it again . L
should have given it back sooner . Can you forgive me
not want to break it again . They also gave some
, Fin was never lazy again and always had a new
wanted to ' spl ash ' around , but she wasn

INTERVAL 1.261 - 5.427
CONTAINS 16.451%

. Finally they reached a good spot to
and Sue went on the slide and had fun . They
ily was sad she couldn 't play outside , but she
and still couldn 't find one . He was getting very
had to be careful . The ball was very big and

INTERVAL -2.905 - 1.261
CONTAINS 79.922%

favorite hat everywhere she went . It was a big ,
, me too !" Ben says . " Can I hold
every day . <|endoftext|> Once there was a big dog .
saw this and came over to give her a hug .
you , conductor ." Anna said . " You

INTERVAL -7.071 - -2.905
CONTAINS 3.431%

the spot for a long time and never forgot this special
lift this box !" He bent down , put his hands
Max bark ed happily and w agged his tail as they
not think Ben was being nice . She had an idea
wish Max would be a good cat and not make a
Token looked
Token ID3114
Feature activation+3.86e-01

Pos logit contributions
wn+0.021
 aback+0.020
coat+0.017
 swe+0.017
 ripe+0.016
Neg logit contributions
glass-0.041
orse-0.040
 generation-0.037
work-0.036
 planet-0.036
Token through
Token ID832
Feature activation+1.02e+00

Pos logit contributions
orse+0.019
glass+0.018
 whale+0.018
work+0.018
 ship+0.017
Neg logit contributions
wn-0.022
coat-0.021
 aback-0.019
arl-0.018
ucker-0.017
Token the
Token ID262
Feature activation-7.10e-01

Pos logit contributions
glass+0.072
orse+0.071
work+0.064
 generation+0.064
bike+0.063
Neg logit contributions
 aback-0.041
wn-0.039
 swe-0.035
 like-0.035
 past-0.034
Token ey
Token ID1926
Feature activation-2.61e+00

Pos logit contributions
wn+0.039
 aback+0.038
coat+0.034
 swe+0.032
 like+0.031
Neg logit contributions
glass-0.040
orse-0.037
 planet-0.036
work-0.035
 whale-0.034
Tokenep
Token ID538
Feature activation+1.24e+00

Pos logit contributions
wn+0.124
 aback+0.118
coat+0.107
 swe+0.105
 like+0.102
Neg logit contributions
glass-0.161
orse-0.159
 generation-0.149
work-0.145
 planet-0.143
Tokeniece
Token ID8535
Feature activation+9.59e+00

Pos logit contributions
orse+0.073
glass+0.072
 generation+0.069
 planet+0.065
work+0.065
Neg logit contributions
wn-0.061
 aback-0.060
 like-0.056
coat-0.054
 past-0.053
Token and
Token ID290
Feature activation-1.08e-01

Pos logit contributions
glass+0.728
orse+0.720
 generation+0.673
work+0.662
bike+0.647
Neg logit contributions
wn-0.320
 aback-0.310
coat-0.271
 swe-0.257
 like-0.254
Token saw
Token ID2497
Feature activation+1.21e+00

Pos logit contributions
wn+0.004
 aback+0.004
coat+0.003
 swe+0.003
 like+0.003
Neg logit contributions
glass-0.008
orse-0.008
 generation-0.007
work-0.007
 planet-0.007
Token the
Token ID262
Feature activation-7.18e-01

Pos logit contributions
glass+0.084
orse+0.084
 generation+0.077
work+0.073
bike+0.073
Neg logit contributions
 aback-0.050
wn-0.046
 like-0.044
 sweet-0.044
 swe-0.043
Token tiny
Token ID7009
Feature activation-2.59e+00

Pos logit contributions
wn+0.035
 aback+0.035
coat+0.032
 swe+0.031
 like+0.030
Neg logit contributions
glass-0.044
orse-0.043
 generation-0.039
work-0.038
bike-0.037
Token scales
Token ID16252
Feature activation-1.08e+00

Pos logit contributions
 aback+0.120
wn+0.119
coat+0.109
 swe+0.108
 like+0.106
Neg logit contributions
glass-0.163
orse-0.159
 generation-0.145
work-0.140
bike-0.139
Token looked
Token ID3114
Feature activation+3.59e-01

Pos logit contributions
wn+0.002
 aback+0.002
 swe+0.001
coat+0.001
 like+0.001
Neg logit contributions
glass-0.004
orse-0.004
 generation-0.004
work-0.004
 planet-0.004
Token through
Token ID832
Feature activation+1.07e+00

Pos logit contributions
glass+0.019
orse+0.019
work+0.017
 whale+0.017
 ship+0.017
Neg logit contributions
wn-0.019
coat-0.018
 aback-0.018
 blonde-0.015
ucker-0.015
Token the
Token ID262
Feature activation-8.13e-01

Pos logit contributions
glass+0.068
orse+0.068
 generation+0.061
work+0.061
 planet+0.059
Neg logit contributions
wn-0.049
 aback-0.049
coat-0.043
 swe-0.043
 like-0.042
Token ey
Token ID1926
Feature activation-2.90e+00

Pos logit contributions
wn+0.045
 aback+0.044
coat+0.041
 like+0.036
 swe+0.036
Neg logit contributions
glass-0.043
 planet-0.040
orse-0.039
 whale-0.039
work-0.038
Tokenep
Token ID538
Feature activation+9.61e-01

Pos logit contributions
wn+0.141
 aback+0.136
coat+0.123
 swe+0.121
 like+0.117
Neg logit contributions
glass-0.176
orse-0.174
 generation-0.161
work-0.158
 planet-0.154
Tokeniece
Token ID8535
Feature activation+9.38e+00

Pos logit contributions
orse+0.062
glass+0.061
 generation+0.058
work+0.056
 planet+0.055
Neg logit contributions
wn-0.042
 aback-0.041
 like-0.037
coat-0.037
 swe-0.035
Token.
Token ID13
Feature activation-5.00e-01

Pos logit contributions
glass+0.702
orse+0.697
 generation+0.655
work+0.642
bike+0.631
Neg logit contributions
wn-0.317
 aback-0.300
coat-0.271
 like-0.257
 swe-0.255
Token He
Token ID679
Feature activation-5.79e-01

Pos logit contributions
wn+0.025
 aback+0.024
coat+0.022
 swe+0.022
 ripe+0.021
Neg logit contributions
glass-0.031
orse-0.030
 generation-0.027
work-0.026
 planet-0.026
Token saw
Token ID2497
Feature activation+1.24e+00

Pos logit contributions
wn+0.020
 aback+0.019
coat+0.017
 swe+0.016
 like+0.016
Neg logit contributions
glass-0.043
orse-0.043
 generation-0.040
work-0.040
 planet-0.039
Token the
Token ID262
Feature activation-7.99e-01

Pos logit contributions
glass+0.084
orse+0.084
 generation+0.078
work+0.075
bike+0.073
Neg logit contributions
 aback-0.052
wn-0.050
 like-0.046
 swe-0.046
 sweet-0.046
Token tiny
Token ID7009
Feature activation-2.54e+00

Pos logit contributions
wn+0.039
 aback+0.039
coat+0.034
 swe+0.034
 like+0.032
Neg logit contributions
glass-0.049
orse-0.048
 generation-0.044
work-0.044
 planet-0.043
Token voice
Token ID3809
Feature activation-8.95e-01

Pos logit contributions
wn+0.000
 aback+0.000
coat+0.000
 swe+0.000
 like+0.000
Neg logit contributions
glass-0.001
orse-0.001
 generation-0.000
work-0.000
 planet-0.000
Token that
Token ID326
Feature activation+6.49e-01

Pos logit contributions
 aback+0.031
wn+0.030
 like+0.027
 swe+0.026
coat+0.025
Neg logit contributions
glass-0.066
orse-0.065
 generation-0.062
 planet-0.059
work-0.059
Token said
Token ID531
Feature activation-1.06e+00

Pos logit contributions
glass+0.050
orse+0.049
 generation+0.046
work+0.045
 planet+0.044
Neg logit contributions
 aback-0.021
wn-0.020
coat-0.017
 swe-0.017
 like-0.016
Token:
Token ID25
Feature activation+1.77e+00

Pos logit contributions
 aback+0.038
wn+0.035
 like+0.032
 swe+0.031
coat+0.031
Neg logit contributions
glass-0.077
orse-0.076
 generation-0.074
work-0.070
 planet-0.069
Token "
Token ID366
Feature activation-1.52e+00

Pos logit contributions
glass+0.135
orse+0.127
 generation+0.124
work+0.120
 planet+0.119
Neg logit contributions
 aback-0.058
 like-0.052
 sweet-0.050
wn-0.049
 past-0.047
TokenWarning
Token ID20361
Feature activation+9.12e+00

Pos logit contributions
 aback+0.072
wn+0.072
coat+0.064
 swe+0.062
 like+0.060
Neg logit contributions
glass-0.096
orse-0.095
 generation-0.085
work-0.085
 planet-0.083
Token!
Token ID0
Feature activation+1.86e+00

Pos logit contributions
glass+0.622
orse+0.618
 generation+0.569
work+0.557
 planet+0.541
Neg logit contributions
 aback-0.380
wn-0.378
coat-0.329
 swe-0.322
 like-0.317
Token It
Token ID632
Feature activation+4.63e-01

Pos logit contributions
glass+0.113
orse+0.107
 generation+0.098
work+0.097
 planet+0.097
Neg logit contributions
 aback-0.095
wn-0.082
 like-0.076
coat-0.072
 sweet-0.071
Token is
Token ID318
Feature activation-1.12e-03

Pos logit contributions
glass+0.028
orse+0.027
 generation+0.025
work+0.025
 planet+0.024
Neg logit contributions
wn-0.023
 aback-0.023
coat-0.021
 swe-0.020
 ripe-0.019
Token not
Token ID407
Feature activation-2.84e-02

Pos logit contributions
wn+0.000
 aback+0.000
coat+0.000
 swe+0.000
 ripe+0.000
Neg logit contributions
glass-0.000
orse-0.000
 generation-0.000
work-0.000
 planet-0.000
Token safe
Token ID3338
Feature activation-1.20e+00

Pos logit contributions
 aback+0.001
wn+0.001
coat+0.001
ucker+0.001
 swe+0.001
Neg logit contributions
glass-0.002
orse-0.002
 generation-0.002
work-0.002
 planet-0.002
Token.
Token ID13
Feature activation-4.51e-01

Pos logit contributions
glass+0.014
orse+0.014
 generation+0.013
work+0.013
 planet+0.013
Neg logit contributions
wn-0.007
 aback-0.007
coat-0.006
 swe-0.006
 like-0.006
Token It
Token ID632
Feature activation+4.60e-01

Pos logit contributions
 aback+0.018
wn+0.016
 like+0.016
 sweet+0.014
 past+0.014
Neg logit contributions
orse-0.032
glass-0.031
bike-0.029
fer-0.029
ich-0.029
Token says
Token ID1139
Feature activation+7.98e-02

Pos logit contributions
glass+0.032
orse+0.031
 generation+0.029
work+0.029
 planet+0.028
Neg logit contributions
wn-0.019
 aback-0.018
coat-0.016
 swe-0.016
 like-0.015
Token:
Token ID25
Feature activation+1.78e+00

Pos logit contributions
 generation+0.005
orse+0.005
glass+0.005
bike+0.005
 harbour+0.005
Neg logit contributions
 like-0.003
 aback-0.003
 sweet-0.003
wn-0.003
 nice-0.003
Token "
Token ID366
Feature activation-1.55e+00

Pos logit contributions
glass+0.138
orse+0.134
 generation+0.132
bike+0.127
work+0.127
Neg logit contributions
 aback-0.046
 like-0.045
 sweet-0.045
 past-0.043
wn-0.042
TokenWarning
Token ID20361
Feature activation+9.03e+00

Pos logit contributions
 aback+0.069
wn+0.069
 like+0.062
 past+0.061
coat+0.056
Neg logit contributions
orse-0.099
glass-0.090
 planet-0.083
 crew-0.083
bike-0.083
Token:
Token ID25
Feature activation+1.78e+00

Pos logit contributions
glass+0.500
orse+0.491
 generation+0.443
work+0.435
 planet+0.421
Neg logit contributions
wn-0.494
 aback-0.489
coat-0.446
 swe-0.437
 like-0.421
Token Do
Token ID2141
Feature activation+1.19e+00

Pos logit contributions
orse+0.108
glass+0.106
 generation+0.099
fer+0.097
bike+0.097
Neg logit contributions
 aback-0.079
 sweet-0.078
wn-0.077
 like-0.077
 past-0.075
Token not
Token ID407
Feature activation-1.55e-02

Pos logit contributions
glass+0.083
orse+0.080
 generation+0.080
work+0.078
fer+0.077
Neg logit contributions
 fun-0.041
wn-0.041
 sweet-0.041
 like-0.041
 past-0.038
Token touch
Token ID3638
Feature activation+1.32e+00

Pos logit contributions
wn+0.001
 aback+0.001
 like+0.000
 swe+0.000
coat+0.000
Neg logit contributions
glass-0.001
orse-0.001
 generation-0.001
work-0.001
bike-0.001
Token.
Token ID13
Feature activation-4.47e-01

Pos logit contributions
orse+0.087
glass+0.086
 generation+0.078
work+0.075
fer+0.074
Neg logit contributions
 aback-0.058
wn-0.057
 like-0.052
 swe-0.051
 sweet-0.049
Token sign
Token ID1051
Feature activation-6.07e-02

Pos logit contributions
wn+0.008
 aback+0.008
coat+0.007
 swe+0.007
 like+0.006
Neg logit contributions
glass-0.016
orse-0.016
 generation-0.015
work-0.015
 planet-0.015
Token that
Token ID326
Feature activation+5.97e-01

Pos logit contributions
wn+0.003
 aback+0.003
coat+0.003
 like+0.003
 swe+0.002
Neg logit contributions
orse-0.004
glass-0.004
 generation-0.004
work-0.003
 planet-0.003
Token said
Token ID531
Feature activation-1.02e+00

Pos logit contributions
glass+0.049
orse+0.049
 generation+0.046
work+0.045
 planet+0.044
Neg logit contributions
 aback-0.016
wn-0.015
coat-0.013
 swe-0.012
 like-0.012
Token:
Token ID25
Feature activation+1.87e+00

Pos logit contributions
 aback+0.034
wn+0.032
 like+0.031
 sweet+0.030
coat+0.028
Neg logit contributions
orse-0.075
glass-0.074
 generation-0.073
work-0.068
bike-0.067
Token "
Token ID366
Feature activation-1.38e+00

Pos logit contributions
glass+0.149
orse+0.145
 generation+0.140
work+0.135
 planet+0.132
Neg logit contributions
 sweet-0.046
 aback-0.046
 past-0.045
 like-0.045
 nice-0.042
TokenWarning
Token ID20361
Feature activation+9.03e+00

Pos logit contributions
wn+0.068
 aback+0.067
coat+0.060
 swe+0.057
 like+0.057
Neg logit contributions
orse-0.084
glass-0.083
 generation-0.074
work-0.074
 planet-0.072
Token!
Token ID0
Feature activation+1.88e+00

Pos logit contributions
glass+0.589
orse+0.585
 generation+0.535
work+0.525
 planet+0.511
Neg logit contributions
wn-0.403
 aback-0.399
coat-0.354
 swe-0.345
 like-0.335
Token There
Token ID1318
Feature activation-3.75e-01

Pos logit contributions
orse+0.116
glass+0.113
field+0.100
 generation+0.098
fer+0.097
Neg logit contributions
 aback-0.090
wn-0.079
 like-0.074
 past-0.072
 sweet-0.072
Token is
Token ID318
Feature activation+1.49e-01

Pos logit contributions
 aback+0.014
wn+0.014
 like+0.012
 swe+0.012
 sweet+0.012
Neg logit contributions
orse-0.027
glass-0.027
 generation-0.024
fer-0.024
work-0.023
Token bad
Token ID2089
Feature activation-2.53e-01

Pos logit contributions
orse+0.011
glass+0.010
fer+0.010
 generation+0.010
work+0.009
Neg logit contributions
 aback-0.005
wn-0.005
 like-0.005
 past-0.005
 fun-0.005
Token news
Token ID1705
Feature activation+3.44e+00

Pos logit contributions
wn+0.014
 aback+0.014
coat+0.013
 swe+0.013
 ripe+0.012
Neg logit contributions
glass-0.014
orse-0.013
 generation-0.012
work-0.012
 planet-0.012
Token A
Token ID317
Feature activation+1.84e+00

Pos logit contributions
wn+0.017
 aback+0.017
coat+0.015
 swe+0.015
 ripe+0.014
Neg logit contributions
glass-0.021
orse-0.021
 generation-0.019
work-0.018
 planet-0.018
Token voice
Token ID3809
Feature activation-1.04e+00

Pos logit contributions
glass+0.121
orse+0.120
 generation+0.111
work+0.108
bike+0.105
Neg logit contributions
wn-0.079
 aback-0.078
coat-0.070
 swe-0.069
 like-0.066
Token said
Token ID531
Feature activation-1.09e+00

Pos logit contributions
 aback+0.036
wn+0.036
coat+0.031
 swe+0.030
 like+0.030
Neg logit contributions
glass-0.077
orse-0.077
 generation-0.072
work-0.070
 planet-0.068
Token,
Token ID11
Feature activation-4.19e-02

Pos logit contributions
 aback+0.029
 like+0.023
 sweet+0.022
wn+0.021
 swe+0.020
Neg logit contributions
 generation-0.091
orse-0.089
glass-0.089
bike-0.086
work-0.084
Token "
Token ID366
Feature activation-1.56e+00

Pos logit contributions
 aback+0.002
wn+0.002
 swe+0.001
coat+0.001
 like+0.001
Neg logit contributions
glass-0.003
orse-0.003
 generation-0.003
work-0.003
 planet-0.003
TokenWarning
Token ID20361
Feature activation+8.94e+00

Pos logit contributions
 aback+0.064
wn+0.060
 past+0.052
 like+0.052
 swe+0.049
Neg logit contributions
orse-0.112
glass-0.103
 planet-0.097
 generation-0.097
work-0.096
Token,
Token ID11
Feature activation-4.81e-02

Pos logit contributions
glass+0.591
orse+0.590
 generation+0.541
work+0.531
 planet+0.515
Neg logit contributions
wn-0.389
 aback-0.387
coat-0.339
 swe-0.332
 like-0.327
Token warning
Token ID6509
Feature activation+1.06e+00

Pos logit contributions
 sweet+0.002
 like+0.002
 aback+0.002
wn+0.002
 nice+0.002
Neg logit contributions
orse-0.003
glass-0.003
 generation-0.003
field-0.003
bike-0.003
Token,
Token ID11
Feature activation-4.66e-02

Pos logit contributions
glass+0.069
orse+0.066
work+0.063
 generation+0.062
 planet+0.061
Neg logit contributions
wn-0.047
 aback-0.045
coat-0.042
 swe-0.039
 ripe-0.038
Token bomb
Token ID5194
Feature activation+5.19e+00

Pos logit contributions
 aback+0.002
wn+0.002
 like+0.002
 sweet+0.002
 swe+0.002
Neg logit contributions
orse-0.003
glass-0.003
 generation-0.003
work-0.003
field-0.003
Token activated
Token ID13906
Feature activation-2.55e-01

Pos logit contributions
glass+0.319
orse+0.306
work+0.298
 planet+0.277
 generation+0.273
Neg logit contributions
wn-0.257
 aback-0.247
coat-0.226
 ripe-0.223
 swe-0.220
Token she
Token ID673
Feature activation-8.89e-01

Pos logit contributions
glass+0.006
orse+0.006
 generation+0.005
work+0.005
 planet+0.005
Neg logit contributions
wn-0.004
 aback-0.003
coat-0.003
 swe-0.003
 ripe-0.003
Token always
Token ID1464
Feature activation-3.67e+00

Pos logit contributions
 aback+0.030
wn+0.029
 like+0.026
coat+0.025
 swe+0.024
Neg logit contributions
glass-0.067
orse-0.066
 generation-0.061
work-0.060
 planet-0.059
Token wore
Token ID12408
Feature activation-1.14e+00

Pos logit contributions
wn+0.148
 aback+0.146
coat+0.127
 swe+0.124
 like+0.120
Neg logit contributions
glass-0.257
orse-0.253
 generation-0.233
work-0.229
 planet-0.225
Token her
Token ID607
Feature activation-1.53e+00

Pos logit contributions
wn+0.048
 aback+0.047
coat+0.046
ucker+0.042
inters+0.040
Neg logit contributions
glass-0.075
 generation-0.069
orse-0.069
 planet-0.065
work-0.064
Token fan
Token ID4336
Feature activation+1.15e-02

Pos logit contributions
wn+0.059
 aback+0.058
 swe+0.051
coat+0.051
 like+0.050
Neg logit contributions
glass-0.108
orse-0.108
 generation-0.099
work-0.097
bike-0.096
Tokenci
Token ID979
Feature activation+8.61e+00

Pos logit contributions
orse+0.001
glass+0.001
 generation+0.001
work+0.001
 planet+0.001
Neg logit contributions
wn-0.000
 aback-0.000
coat-0.000
 swe-0.000
 like-0.000
Tokenest
Token ID395
Feature activation-2.08e-02

Pos logit contributions
glass+0.600
orse+0.589
 generation+0.550
work+0.544
 planet+0.528
Neg logit contributions
wn-0.343
 aback-0.334
coat-0.297
 swe-0.287
 ripe-0.277
Token dress
Token ID6576
Feature activation+6.78e-01

Pos logit contributions
wn+0.001
 aback+0.001
coat+0.001
 swe+0.001
 ripe+0.001
Neg logit contributions
glass-0.001
orse-0.001
 generation-0.001
work-0.001
 planet-0.001
Token to
Token ID284
Feature activation-3.51e-01

Pos logit contributions
orse+0.049
glass+0.049
 generation+0.045
work+0.044
 planet+0.043
Neg logit contributions
wn-0.025
 aback-0.024
coat-0.023
 like-0.021
 swe-0.021
Token the
Token ID262
Feature activation-6.69e-01

Pos logit contributions
wn+0.014
 aback+0.013
 like+0.013
 swe+0.012
 sweet+0.012
Neg logit contributions
glass-0.024
orse-0.024
work-0.022
 generation-0.022
bike-0.022
Token museum
Token ID13257
Feature activation+3.01e+00

Pos logit contributions
 like+0.034
 fun+0.033
 past+0.033
wn+0.033
 sweet+0.032
Neg logit contributions
orse-0.040
glass-0.037
bike-0.033
work-0.031
fer-0.031
Token dressed
Token ID12049
Feature activation-4.27e-01

Pos logit contributions
wn+0.023
 aback+0.022
coat+0.020
 swe+0.019
 like+0.018
Neg logit contributions
glass-0.037
orse-0.037
 generation-0.034
work-0.034
 planet-0.033
Token up
Token ID510
Feature activation+7.47e-01

Pos logit contributions
wn+0.016
 aback+0.016
coat+0.014
 swe+0.013
 ripe+0.013
Neg logit contributions
glass-0.031
orse-0.030
 generation-0.028
work-0.028
 planet-0.027
Token in
Token ID287
Feature activation+8.00e-02

Pos logit contributions
glass+0.057
orse+0.057
 generation+0.052
work+0.051
 planet+0.050
Neg logit contributions
 aback-0.025
wn-0.025
coat-0.022
 swe-0.021
 like-0.021
Token her
Token ID607
Feature activation-1.64e+00

Pos logit contributions
glass+0.006
orse+0.006
 generation+0.005
work+0.005
 planet+0.005
Neg logit contributions
wn-0.003
 aback-0.003
coat-0.003
 swe-0.003
 like-0.003
Token fan
Token ID4336
Feature activation-1.04e-01

Pos logit contributions
wn+0.057
 swe+0.056
 aback+0.055
coat+0.054
 sweet+0.054
Neg logit contributions
orse-0.120
glass-0.119
 generation-0.110
work-0.106
bike-0.106
Tokenci
Token ID979
Feature activation+8.55e+00

Pos logit contributions
coat+0.003
wn+0.003
 aback+0.003
 swe+0.003
 like+0.003
Neg logit contributions
orse-0.008
glass-0.008
 generation-0.007
 Whale-0.007
fer-0.007
Tokenest
Token ID395
Feature activation-1.38e-01

Pos logit contributions
glass+0.645
orse+0.640
 generation+0.584
work+0.575
 planet+0.564
Neg logit contributions
wn-0.299
 aback-0.298
coat-0.257
 swe-0.255
 like-0.233
Token clothes
Token ID8242
Feature activation-3.22e+00

Pos logit contributions
wn+0.006
 aback+0.006
coat+0.005
 swe+0.005
 ripe+0.005
Neg logit contributions
glass-0.009
orse-0.009
 generation-0.008
work-0.008
 planet-0.008
Token and
Token ID290
Feature activation-3.20e-02

Pos logit contributions
wn+0.110
 aback+0.107
coat+0.098
 like+0.091
 swe+0.090
Neg logit contributions
glass-0.242
orse-0.240
 generation-0.224
work-0.216
 planet-0.215
Token asked
Token ID1965
Feature activation-3.44e+00

Pos logit contributions
wn+0.001
 aback+0.001
coat+0.001
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.002
orse-0.002
 generation-0.002
work-0.002
 planet-0.002
Token her
Token ID607
Feature activation-1.64e+00

Pos logit contributions
wn+0.135
 aback+0.128
coat+0.118
 swe+0.107
 ripe+0.102
Neg logit contributions
glass-0.244
orse-0.239
 generation-0.227
work-0.221
 planet-0.217
Token to
Token ID284
Feature activation-4.11e-01

Pos logit contributions
glass+0.122
orse+0.120
 generation+0.112
work+0.111
 planet+0.109
Neg logit contributions
wn-0.045
 aback-0.044
coat-0.037
 swe-0.035
 like-0.033
Token put
Token ID1234
Feature activation-2.39e+00

Pos logit contributions
wn+0.017
 aback+0.017
coat+0.015
 swe+0.015
 ripe+0.014
Neg logit contributions
glass-0.028
orse-0.027
 generation-0.025
work-0.025
 planet-0.025
Token on
Token ID319
Feature activation+1.03e+00

Pos logit contributions
wn+0.132
 aback+0.130
coat+0.126
 swe+0.116
inters+0.114
Neg logit contributions
glass-0.123
orse-0.115
work-0.107
 generation-0.107
 whale-0.095
Token her
Token ID607
Feature activation-1.57e+00

Pos logit contributions
glass+0.068
orse+0.067
 generation+0.063
work+0.061
 planet+0.059
Neg logit contributions
wn-0.044
 aback-0.043
coat-0.039
 swe-0.036
 like-0.034
Token fan
Token ID4336
Feature activation-3.34e-02

Pos logit contributions
 aback+0.063
wn+0.063
coat+0.053
 swe+0.049
 like+0.048
Neg logit contributions
glass-0.110
orse-0.109
work-0.098
 generation-0.098
 planet-0.096
Tokenci
Token ID979
Feature activation+8.55e+00

Pos logit contributions
wn+0.002
 aback+0.002
coat+0.002
 swe+0.002
 like+0.002
Neg logit contributions
orse-0.001
glass-0.001
 generation-0.001
 planet-0.001
work-0.001
Tokenest
Token ID395
Feature activation-1.33e-01

Pos logit contributions
glass+0.585
orse+0.574
 generation+0.533
work+0.527
 planet+0.511
Neg logit contributions
wn-0.352
 aback-0.347
coat-0.302
 swe-0.292
 ripe-0.290
Token sunglasses
Token ID36251
Feature activation-2.32e+00

Pos logit contributions
 aback+0.006
wn+0.006
coat+0.005
 swe+0.005
 ripe+0.005
Neg logit contributions
glass-0.009
orse-0.008
 generation-0.007
work-0.007
 planet-0.007
Token.
Token ID13
Feature activation+9.00e-03

Pos logit contributions
wn+0.092
 aback+0.091
coat+0.078
 swe+0.077
 ripe+0.072
Neg logit contributions
glass-0.164
orse-0.161
 generation-0.148
work-0.147
 planet-0.143
Token
Token ID198
Feature activation+4.68e-02

Pos logit contributions
orse+0.001
glass+0.000
 generation+0.000
work+0.000
 planet+0.000
Neg logit contributions
wn-0.000
 aback-0.000
 swe-0.000
coat-0.000
 ripe-0.000
Token
Token ID198
Feature activation+3.21e-02

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.003
work+0.003
 planet+0.002
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.002
 swe-0.002
 ripe-0.002
Token the
Token ID262
Feature activation-6.85e-01

Pos logit contributions
glass+0.010
orse+0.010
 generation+0.009
 planet+0.009
fer+0.009
Neg logit contributions
wn-0.003
 aback-0.003
 paid-0.003
coat-0.003
 past-0.002
Token daughter
Token ID4957
Feature activation+2.59e+00

Pos logit contributions
wn+0.036
 swe+0.036
 aback+0.035
coat+0.035
 sweet+0.035
Neg logit contributions
orse-0.040
glass-0.038
 generation-0.034
work-0.033
 Whale-0.032
Token found
Token ID1043
Feature activation-2.39e-01

Pos logit contributions
glass+0.185
orse+0.184
 generation+0.170
work+0.167
 planet+0.164
Neg logit contributions
wn-0.096
 aback-0.095
coat-0.087
 swe-0.084
 like-0.080
Token the
Token ID262
Feature activation-6.69e-01

Pos logit contributions
wn+0.011
 aback+0.011
coat+0.010
 swe+0.009
 like+0.009
Neg logit contributions
glass-0.016
orse-0.015
 generation-0.014
work-0.014
 planet-0.013
Token fan
Token ID4336
Feature activation+1.76e-02

Pos logit contributions
wn+0.027
 aback+0.026
coat+0.021
 swe+0.020
 ripe+0.020
Neg logit contributions
glass-0.047
orse-0.045
work-0.043
 generation-0.043
 planet-0.043
Tokenci
Token ID979
Feature activation+8.49e+00

Pos logit contributions
orse+0.001
glass+0.001
 generation+0.001
growth+0.001
 Whale+0.001
Neg logit contributions
coat-0.001
wn-0.001
 aback-0.001
 like-0.001
 swe-0.001
Tokenest
Token ID395
Feature activation-1.44e-01

Pos logit contributions
glass+0.633
orse+0.630
 generation+0.570
work+0.555
 planet+0.552
Neg logit contributions
wn-0.307
 aback-0.307
coat-0.269
 swe-0.260
 like-0.245
Token dress
Token ID6576
Feature activation+5.54e-01

Pos logit contributions
wn+0.006
 aback+0.006
coat+0.006
 swe+0.005
 like+0.005
Neg logit contributions
glass-0.010
orse-0.009
 generation-0.009
work-0.008
 planet-0.008
Token she
Token ID673
Feature activation-8.81e-01

Pos logit contributions
orse+0.040
glass+0.040
 generation+0.037
work+0.036
 planet+0.036
Neg logit contributions
wn-0.020
 aback-0.019
coat-0.018
 swe-0.017
 like-0.017
Token had
Token ID550
Feature activation+1.13e+00

Pos logit contributions
 aback+0.042
wn+0.041
coat+0.037
 swe+0.036
 ripe+0.034
Neg logit contributions
glass-0.056
orse-0.055
 generation-0.050
work-0.050
 planet-0.048
Token ever
Token ID1683
Feature activation+4.42e+00

Pos logit contributions
glass+0.065
orse+0.064
 generation+0.058
work+0.057
bike+0.056
Neg logit contributions
wn-0.058
 aback-0.057
coat-0.053
 like-0.052
 swe-0.050
Token a
Token ID257
Feature activation-1.88e-01

Pos logit contributions
wn+0.061
 aback+0.060
coat+0.058
 swe+0.054
 ripe+0.053
Neg logit contributions
glass-0.078
orse-0.074
 generation-0.070
work-0.068
 planet-0.065
Token seat
Token ID5852
Feature activation+1.71e+00

Pos logit contributions
wn+0.008
 aback+0.007
coat+0.006
 swe+0.006
 like+0.006
Neg logit contributions
glass-0.013
orse-0.013
work-0.012
 planet-0.012
 generation-0.012
Token in
Token ID287
Feature activation+2.13e-01

Pos logit contributions
orse+0.135
glass+0.131
 generation+0.126
 planet+0.120
 Whale+0.119
Neg logit contributions
wn-0.055
 aback-0.053
coat-0.050
 like-0.048
 past-0.044
Token the
Token ID262
Feature activation-6.03e-01

Pos logit contributions
orse+0.015
glass+0.015
 generation+0.013
work+0.013
ich+0.013
Neg logit contributions
 aback-0.009
wn-0.009
 like-0.008
 swe-0.008
coat-0.008
Token motor
Token ID5584
Feature activation+2.85e+00

Pos logit contributions
wn+0.037
 aback+0.036
coat+0.033
 swe+0.033
 like+0.032
Neg logit contributions
orse-0.030
glass-0.030
 generation-0.026
work-0.026
 planet-0.024
Tokenbike
Token ID32256
Feature activation+8.40e+00

Pos logit contributions
orse+0.232
glass+0.226
 generation+0.215
growth+0.209
 Whale+0.203
Neg logit contributions
 aback-0.075
coat-0.074
wn-0.072
 like-0.069
 past-0.066
Token behind
Token ID2157
Feature activation+2.81e+00

Pos logit contributions
orse+0.663
glass+0.655
 generation+0.605
work+0.590
 planet+0.588
Neg logit contributions
 aback-0.257
wn-0.254
coat-0.227
 like-0.220
 past-0.202
Token him
Token ID683
Feature activation+5.62e-01

Pos logit contributions
orse+0.221
glass+0.215
ich+0.195
 generation+0.195
fer+0.192
Neg logit contributions
 aback-0.094
wn-0.088
 like-0.081
 past-0.078
coat-0.074
Token.
Token ID13
Feature activation-5.42e-02

Pos logit contributions
glass+0.041
orse+0.041
 generation+0.037
work+0.037
 planet+0.036
Neg logit contributions
wn-0.021
 aback-0.021
coat-0.018
 swe-0.017
 like-0.016
Token He
Token ID679
Feature activation-2.81e-01

Pos logit contributions
wn+0.003
 aback+0.003
coat+0.003
 swe+0.003
 ripe+0.002
Neg logit contributions
glass-0.003
orse-0.003
 generation-0.003
work-0.003
bike-0.003
Token waved
Token ID26834
Feature activation-4.37e-01

Pos logit contributions
wn+0.011
 aback+0.011
coat+0.009
 swe+0.009
 like+0.009
Neg logit contributions
glass-0.020
orse-0.019
 generation-0.019
work-0.018
 planet-0.018
Token a
Token ID257
Feature activation-1.92e-01

Pos logit contributions
glass+0.010
orse+0.010
 generation+0.009
 planet+0.009
work+0.009
Neg logit contributions
wn-0.008
 aback-0.008
coat-0.008
ets-0.007
 swe-0.007
Token man
Token ID582
Feature activation-1.56e+00

Pos logit contributions
wn+0.006
 aback+0.006
coat+0.005
 swe+0.005
 sweet+0.005
Neg logit contributions
glass-0.015
orse-0.015
 generation-0.014
work-0.013
bike-0.013
Token on
Token ID319
Feature activation+9.71e-01

Pos logit contributions
 aback+0.058
 like+0.057
wn+0.056
coat+0.051
 past+0.049
Neg logit contributions
glass-0.108
orse-0.102
 generation-0.100
work-0.096
 planet-0.095
Token a
Token ID257
Feature activation-1.97e-01

Pos logit contributions
orse+0.074
glass+0.073
work+0.066
fer+0.065
 generation+0.065
Neg logit contributions
 aback-0.032
wn-0.031
 like-0.029
 past-0.028
 fun-0.027
Token motor
Token ID5584
Feature activation+2.83e+00

Pos logit contributions
wn+0.010
 aback+0.009
coat+0.009
 swe+0.008
 like+0.008
Neg logit contributions
glass-0.012
orse-0.012
 generation-0.011
work-0.011
 planet-0.010
Tokenbike
Token ID32256
Feature activation+8.37e+00

Pos logit contributions
glass+0.211
orse+0.211
 generation+0.197
growth+0.186
work+0.183
Neg logit contributions
 like-0.088
 aback-0.085
wn-0.085
coat-0.084
 past-0.079
Token!
Token ID0
Feature activation+1.80e+00

Pos logit contributions
glass+0.632
orse+0.632
 generation+0.582
work+0.563
 planet+0.556
Neg logit contributions
 aback-0.276
wn-0.269
 like-0.252
coat-0.242
 past-0.222
Token
Token ID198
Feature activation+7.02e-02

Pos logit contributions
glass+0.106
orse+0.104
 generation+0.094
work+0.093
 planet+0.090
Neg logit contributions
wn-0.092
 aback-0.092
coat-0.083
 swe-0.081
 like-0.078
Token
Token ID198
Feature activation+5.83e-02

Pos logit contributions
glass+0.004
orse+0.004
 generation+0.004
 planet+0.003
work+0.003
Neg logit contributions
wn-0.004
 aback-0.004
 swe-0.003
coat-0.003
 prett-0.003
TokenThe
Token ID464
Feature activation+3.98e+00

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.002
work+0.002
bike+0.002
Neg logit contributions
wn-0.004
 aback-0.004
coat-0.003
 swe-0.003
 ripe-0.003
Token goat
Token ID26917
Feature activation+2.24e+00

Pos logit contributions
glass+0.200
orse+0.197
 generation+0.176
work+0.172
 planet+0.166
Neg logit contributions
wn-0.240
 aback-0.238
coat-0.219
 swe-0.216
 like-0.209
Token will
Token ID481
Feature activation-2.15e+00

Pos logit contributions
wn+0.013
 aback+0.012
 swe+0.012
coat+0.011
 ripe+0.010
Neg logit contributions
orse-0.021
glass-0.020
 generation-0.019
work-0.019
 planet-0.018
Token buy
Token ID2822
Feature activation+1.36e+00

Pos logit contributions
wn+0.102
 aback+0.101
coat+0.088
 swe+0.087
 ripe+0.081
Neg logit contributions
glass-0.135
orse-0.134
 generation-0.125
work-0.123
 planet-0.120
Token you
Token ID345
Feature activation+1.55e+00

Pos logit contributions
glass+0.085
orse+0.084
 generation+0.076
work+0.075
 planet+0.073
Neg logit contributions
wn-0.064
 aback-0.064
 swe-0.058
coat-0.056
 like-0.055
Token the
Token ID262
Feature activation-6.68e-01

Pos logit contributions
glass+0.100
orse+0.098
 generation+0.090
work+0.088
 planet+0.086
Neg logit contributions
wn-0.071
 aback-0.070
coat-0.063
 swe-0.062
 like-0.060
Token fan
Token ID4336
Feature activation-1.30e-02

Pos logit contributions
 aback+0.031
wn+0.031
coat+0.025
inters+0.025
ucker+0.024
Neg logit contributions
glass-0.042
orse-0.039
 planet-0.039
work-0.038
 generation-0.038
Tokenci
Token ID979
Feature activation+8.35e+00

Pos logit contributions
 aback+0.000
wn+0.000
coat+0.000
 like+0.000
 swe+0.000
Neg logit contributions
glass-0.001
orse-0.001
 generation-0.001
 planet-0.001
bike-0.001
Tokenest
Token ID395
Feature activation-2.22e-01

Pos logit contributions
glass+0.598
orse+0.590
 generation+0.546
work+0.539
 planet+0.526
Neg logit contributions
wn-0.320
 aback-0.315
coat-0.275
 swe-0.267
 like-0.252
Token dress
Token ID6576
Feature activation+5.34e-01

Pos logit contributions
wn+0.009
 aback+0.009
coat+0.008
 swe+0.008
 ripe+0.007
Neg logit contributions
glass-0.015
orse-0.015
 generation-0.014
work-0.014
 planet-0.013
Token in
Token ID287
Feature activation+1.91e-01

Pos logit contributions
glass+0.038
orse+0.037
 generation+0.035
work+0.034
 planet+0.033
Neg logit contributions
wn-0.021
 aback-0.020
coat-0.018
 swe-0.017
 like-0.017
Token the
Token ID262
Feature activation-6.89e-01

Pos logit contributions
glass+0.014
orse+0.014
bike+0.013
work+0.013
 generation+0.012
Neg logit contributions
 aback-0.007
wn-0.007
 like-0.006
 swe-0.006
 sweet-0.006
Token store
Token ID3650
Feature activation-5.48e-01

Pos logit contributions
wn+0.037
 aback+0.036
 swe+0.033
coat+0.033
 like+0.032
Neg logit contributions
orse-0.040
glass-0.039
 generation-0.034
work-0.034
bike-0.033
Token the
Token ID262
Feature activation-7.31e-01

Pos logit contributions
orse+0.050
glass+0.050
 generation+0.046
 planet+0.045
work+0.045
Neg logit contributions
wn-0.041
 aback-0.039
coat-0.038
ets-0.034
 swe-0.034
Token brav
Token ID49025
Feature activation+2.59e+00

Pos logit contributions
wn+0.035
 aback+0.035
coat+0.030
 swe+0.029
 ripe+0.027
Neg logit contributions
glass-0.045
orse-0.044
 generation-0.041
work-0.041
 planet-0.041
Tokenest
Token ID395
Feature activation-3.09e-01

Pos logit contributions
orse+0.145
glass+0.143
 generation+0.142
work+0.136
 planet+0.130
Neg logit contributions
wn-0.124
 aback-0.117
coat-0.113
 swe-0.109
 ripe-0.108
Token and
Token ID290
Feature activation+2.18e-03

Pos logit contributions
wn+0.015
 aback+0.015
coat+0.013
 swe+0.013
 ripe+0.012
Neg logit contributions
glass-0.019
orse-0.019
 generation-0.017
work-0.017
 planet-0.017
Token boun
Token ID31283
Feature activation-2.38e+00

Pos logit contributions
glass+0.000
orse+0.000
 generation+0.000
work+0.000
 planet+0.000
Neg logit contributions
wn-0.000
 aback-0.000
coat-0.000
 swe-0.000
 like-0.000
Tokenci
Token ID979
Feature activation+8.29e+00

Pos logit contributions
wn+0.099
 aback+0.098
coat+0.086
 swe+0.083
 like+0.080
Neg logit contributions
glass-0.164
orse-0.161
 generation-0.148
work-0.145
 planet-0.142
Tokenest
Token ID395
Feature activation-2.96e-01

Pos logit contributions
glass+0.645
orse+0.632
 generation+0.582
work+0.568
 planet+0.559
Neg logit contributions
 aback-0.273
wn-0.272
coat-0.236
 swe-0.227
 like-0.219
Token bunny
Token ID44915
Feature activation+4.08e-01

Pos logit contributions
wn+0.014
 aback+0.014
coat+0.012
 swe+0.012
 ripe+0.012
Neg logit contributions
glass-0.019
orse-0.018
 generation-0.017
work-0.016
 planet-0.016
Token in
Token ID287
Feature activation+2.11e-01

Pos logit contributions
glass+0.029
orse+0.027
 generation+0.026
work+0.025
growth+0.025
Neg logit contributions
 aback-0.015
wn-0.015
 like-0.015
 sweet-0.013
coat-0.013
Token the
Token ID262
Feature activation-7.40e-01

Pos logit contributions
glass+0.016
orse+0.015
work+0.014
 generation+0.014
bike+0.014
Neg logit contributions
 aback-0.008
wn-0.007
 like-0.007
 past-0.006
 fun-0.006
Token whole
Token ID2187
Feature activation+2.84e+00

Pos logit contributions
wn+0.043
 aback+0.043
coat+0.039
 swe+0.038
 like+0.037
Neg logit contributions
glass-0.038
orse-0.038
 generation-0.034
work-0.033
 planet-0.033
Token new
Token ID649
Feature activation-6.00e-02

Pos logit contributions
wn+0.110
 swe+0.107
 aback+0.106
 fun+0.104
 sweet+0.103
Neg logit contributions
glass-0.077
orse-0.074
bike-0.066
work-0.064
 generation-0.064
Token home
Token ID1363
Feature activation+3.49e+00

Pos logit contributions
wn+0.004
 aback+0.004
 swe+0.004
coat+0.004
 like+0.004
Neg logit contributions
glass-0.002
orse-0.002
 generation-0.002
work-0.002
bike-0.002
Token with
Token ID351
Feature activation+4.60e-01

Pos logit contributions
orse+0.280
glass+0.280
 generation+0.261
work+0.255
bike+0.255
Neg logit contributions
 aback-0.092
wn-0.091
 like-0.084
coat-0.078
 swe-0.075
Token the
Token ID262
Feature activation-6.63e-01

Pos logit contributions
glass+0.033
orse+0.033
work+0.030
 generation+0.030
bike+0.029
Neg logit contributions
 aback-0.017
wn-0.017
 swe-0.015
coat-0.015
 like-0.014
Token fan
Token ID4336
Feature activation+1.53e-02

Pos logit contributions
wn+0.034
 aback+0.033
coat+0.030
 swe+0.029
 like+0.028
Neg logit contributions
glass-0.040
orse-0.039
 generation-0.035
work-0.035
 planet-0.034
Tokenci
Token ID979
Feature activation+8.18e+00

Pos logit contributions
glass+0.001
orse+0.001
 generation+0.001
 Whale+0.001
growth+0.001
Neg logit contributions
wn-0.000
 aback-0.000
 like-0.000
coat-0.000
 swe-0.000
Tokenest
Token ID395
Feature activation-3.08e-01

Pos logit contributions
glass+0.610
orse+0.601
 generation+0.552
work+0.542
 planet+0.533
Neg logit contributions
wn-0.292
 aback-0.290
coat-0.251
 swe-0.244
 like-0.232
Token king
Token ID5822
Feature activation+4.85e-01

Pos logit contributions
wn+0.015
 aback+0.015
coat+0.013
 swe+0.013
 ripe+0.012
Neg logit contributions
glass-0.019
orse-0.019
 generation-0.017
work-0.017
 planet-0.016
Token in
Token ID287
Feature activation+2.15e-01

Pos logit contributions
glass+0.031
orse+0.031
 generation+0.029
bike+0.028
work+0.028
Neg logit contributions
wn-0.021
 aback-0.020
coat-0.018
 like-0.018
 swe-0.018
Token the
Token ID262
Feature activation-6.71e-01

Pos logit contributions
glass+0.016
orse+0.016
bike+0.014
work+0.014
ich+0.014
Neg logit contributions
 aback-0.008
wn-0.007
 like-0.007
 past-0.006
 sweet-0.006
Token land
Token ID1956
Feature activation-1.13e-01

Pos logit contributions
wn+0.042
 aback+0.042
 swe+0.039
 sweet+0.038
coat+0.038
Neg logit contributions
glass-0.031
orse-0.031
work-0.026
bike-0.026
 generation-0.025
Token by
Token ID416
Feature activation-8.17e-01

Pos logit contributions
 aback+0.054
wn+0.054
coat+0.045
 swe+0.045
 like+0.043
Neg logit contributions
glass-0.131
orse-0.130
 generation-0.121
work-0.117
 planet-0.116
Token saying
Token ID2282
Feature activation-2.73e-01

Pos logit contributions
 sweet+0.036
wn+0.035
 aback+0.035
 like+0.035
 swe+0.034
Neg logit contributions
glass-0.052
orse-0.050
bike-0.045
 generation-0.044
fer-0.043
Token,
Token ID11
Feature activation-1.46e-02

Pos logit contributions
 aback+0.011
wn+0.011
 swe+0.010
 sweet+0.010
 like+0.010
Neg logit contributions
glass-0.018
orse-0.018
 generation-0.017
work-0.016
bike-0.016
Token "
Token ID366
Feature activation-1.36e+00

Pos logit contributions
wn+0.001
 aback+0.001
coat+0.001
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.001
orse-0.001
 generation-0.001
work-0.001
 planet-0.001
TokenMar
Token ID7676
Feature activation-1.06e+00

Pos logit contributions
wn+0.069
 aback+0.068
coat+0.063
 like+0.060
 swe+0.060
Neg logit contributions
orse-0.080
glass-0.080
 planet-0.069
 generation-0.069
work-0.068
Tokenriage
Token ID4087
Feature activation+8.14e+00

Pos logit contributions
wn+0.085
 aback+0.085
coat+0.079
 swe+0.077
 past+0.076
Neg logit contributions
orse-0.033
glass-0.032
 planet-0.023
 generation-0.022
bike-0.021
Token is
Token ID318
Feature activation+1.46e-01

Pos logit contributions
orse+0.575
glass+0.568
 generation+0.522
bike+0.504
work+0.499
Neg logit contributions
 aback-0.313
wn-0.308
coat-0.276
 like-0.271
 swe-0.267
Token when
Token ID618
Feature activation-5.36e-01

Pos logit contributions
glass+0.011
orse+0.010
 generation+0.010
work+0.009
 planet+0.009
Neg logit contributions
wn-0.006
 aback-0.006
coat-0.005
 swe-0.005
 like-0.005
Token two
Token ID734
Feature activation+9.21e-01

Pos logit contributions
 sweet+0.023
 like+0.023
 fun+0.022
 nice+0.021
 swe+0.020
Neg logit contributions
orse-0.035
glass-0.035
bike-0.032
work-0.031
fer-0.030
Token people
Token ID661
Feature activation-8.27e-01

Pos logit contributions
glass+0.053
orse+0.052
 generation+0.048
work+0.047
 planet+0.046
Neg logit contributions
wn-0.048
 aback-0.048
coat-0.043
 swe-0.042
 ripe-0.040
Token who
Token ID508
Feature activation+1.63e+00

Pos logit contributions
wn+0.034
 aback+0.034
coat+0.030
 swe+0.029
 ripe+0.027
Neg logit contributions
glass-0.057
orse-0.056
work-0.053
 generation-0.052
 planet-0.050
Token.
Token ID13
Feature activation-1.06e-01

Pos logit contributions
wn+0.042
 aback+0.039
coat+0.036
 swe+0.034
 ripe+0.033
Neg logit contributions
orse-0.064
glass-0.063
work-0.057
 generation-0.056
bike-0.055
Token
Token ID198
Feature activation-5.78e-02

Pos logit contributions
wn+0.005
 aback+0.005
coat+0.004
 swe+0.004
 like+0.004
Neg logit contributions
glass-0.007
orse-0.007
 generation-0.006
work-0.006
 planet-0.006
Token
Token ID198
Feature activation-7.00e-02

Pos logit contributions
 aback+0.003
wn+0.003
coat+0.002
 swe+0.002
 like+0.002
Neg logit contributions
glass-0.004
orse-0.004
 generation-0.003
work-0.003
 planet-0.003
Token"
Token ID1
Feature activation-6.32e-01

Pos logit contributions
 aback+0.003
wn+0.003
coat+0.003
 swe+0.003
 ripe+0.003
Neg logit contributions
glass-0.004
orse-0.004
 generation-0.004
work-0.004
 planet-0.004
TokenMar
Token ID7676
Feature activation-1.14e+00

Pos logit contributions
coat+0.025
 aback+0.025
wn+0.025
 like+0.024
 sweet+0.023
Neg logit contributions
orse-0.041
glass-0.041
 generation-0.038
 planet-0.037
work-0.037
Tokenriage
Token ID4087
Feature activation+8.13e+00

Pos logit contributions
 aback+0.068
wn+0.067
coat+0.062
 past+0.059
 like+0.059
Neg logit contributions
glass-0.059
orse-0.057
 generation-0.052
 planet-0.049
growth-0.045
Token is
Token ID318
Feature activation+6.03e-02

Pos logit contributions
glass+0.548
orse+0.545
 generation+0.501
work+0.486
bike+0.476
Neg logit contributions
 aback-0.341
wn-0.336
coat-0.300
 swe-0.294
 like-0.286
Token when
Token ID618
Feature activation-5.72e-01

Pos logit contributions
glass+0.004
orse+0.004
 generation+0.004
work+0.004
 planet+0.004
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.002
 like-0.002
 swe-0.002
Token two
Token ID734
Feature activation+8.59e-01

Pos logit contributions
 like+0.024
 sweet+0.024
 fun+0.023
 ripe+0.023
 aback+0.022
Neg logit contributions
glass-0.038
orse-0.037
work-0.034
bike-0.033
 generation-0.032
Token people
Token ID661
Feature activation-8.45e-01

Pos logit contributions
glass+0.052
orse+0.051
 generation+0.047
work+0.046
 planet+0.045
Neg logit contributions
wn-0.042
 aback-0.042
coat-0.038
 swe-0.037
 like-0.035
Token love
Token ID1842
Feature activation-2.00e-01

Pos logit contributions
wn+0.034
 aback+0.033
coat+0.030
 swe+0.029
 ripe+0.026
Neg logit contributions
glass-0.059
orse-0.058
work-0.054
 generation-0.054
 planet-0.052
Token cat
Token ID3797
Feature activation+7.07e-01

Pos logit contributions
glass+0.145
orse+0.145
 generation+0.135
work+0.132
fer+0.124
Neg logit contributions
 sweet-0.098
wn-0.096
 swe-0.095
coat-0.095
 aback-0.094
Token said
Token ID531
Feature activation-9.87e-01

Pos logit contributions
glass+0.053
orse+0.052
 generation+0.048
work+0.048
 planet+0.046
Neg logit contributions
wn-0.026
 aback-0.025
coat-0.021
 swe-0.020
 ripe-0.019
Token,
Token ID11
Feature activation+8.98e-02

Pos logit contributions
 aback+0.030
wn+0.028
coat+0.024
 swe+0.024
 like+0.023
Neg logit contributions
glass-0.079
orse-0.078
 generation-0.074
work-0.072
 planet-0.071
Token "
Token ID366
Feature activation-1.54e+00

Pos logit contributions
glass+0.007
orse+0.006
 generation+0.006
 planet+0.006
work+0.006
Neg logit contributions
 aback-0.003
wn-0.003
 like-0.003
 swe-0.003
 sweet-0.003
TokenMar
Token ID7676
Feature activation-1.28e+00

Pos logit contributions
wn+0.065
 aback+0.064
 like+0.063
coat+0.062
 prett+0.059
Neg logit contributions
glass-0.097
orse-0.095
 generation-0.085
 ship-0.085
 planet-0.084
Tokenriage
Token ID4087
Feature activation+8.07e+00

Pos logit contributions
 aback+0.075
wn+0.074
coat+0.069
 past+0.065
 swe+0.064
Neg logit contributions
glass-0.068
orse-0.068
 generation-0.060
 planet-0.058
work-0.054
Token is
Token ID318
Feature activation+1.41e-02

Pos logit contributions
orse+0.580
glass+0.574
 generation+0.533
bike+0.511
work+0.508
Neg logit contributions
 aback-0.298
wn-0.291
coat-0.263
 like-0.262
 swe-0.253
Token when
Token ID618
Feature activation-4.93e-01

Pos logit contributions
glass+0.001
orse+0.001
 generation+0.001
work+0.001
 planet+0.001
Neg logit contributions
wn-0.001
 aback-0.001
 like-0.000
coat-0.000
 swe-0.000
Token two
Token ID734
Feature activation+7.84e-01

Pos logit contributions
 like+0.020
 sweet+0.019
 fun+0.019
 nice+0.018
 past+0.018
Neg logit contributions
glass-0.034
orse-0.032
work-0.030
bike-0.029
 generation-0.028
Token animals
Token ID4695
Feature activation-1.93e+00

Pos logit contributions
glass+0.046
orse+0.045
 generation+0.041
work+0.040
 planet+0.039
Neg logit contributions
wn-0.040
 aback-0.040
coat-0.036
 swe-0.035
 like-0.034
Token love
Token ID1842
Feature activation-1.67e-01

Pos logit contributions
wn+0.079
 aback+0.079
coat+0.070
 swe+0.068
 ripe+0.063
Neg logit contributions
glass-0.132
orse-0.131
 generation-0.121
work-0.121
 planet-0.118
Token,
Token ID11
Feature activation+9.44e-02

Pos logit contributions
wn+0.002
 aback+0.002
 prett+0.002
 tast+0.001
 swe+0.001
Neg logit contributions
orse-0.002
glass-0.002
work-0.002
 planet-0.002
 generation-0.002
Token be
Token ID307
Feature activation+5.28e-01

Pos logit contributions
glass+0.005
orse+0.005
 generation+0.005
work+0.005
bike+0.005
Neg logit contributions
wn-0.005
 aback-0.005
 like-0.005
 swe-0.005
 sweet-0.005
Token careful
Token ID8161
Feature activation-2.34e+00

Pos logit contributions
glass+0.037
orse+0.035
 generation+0.033
ich+0.032
 crew+0.031
Neg logit contributions
 aback-0.019
coat-0.018
wn-0.017
ets-0.016
inters-0.016
Token!
Token ID0
Feature activation+1.91e+00

Pos logit contributions
wn+0.085
 aback+0.083
coat+0.077
 like+0.074
 sweet+0.071
Neg logit contributions
glass-0.172
orse-0.167
 generation-0.157
work-0.151
 planet-0.149
Token Sp
Token ID1338
Feature activation+1.82e+00

Pos logit contributions
glass+0.131
orse+0.118
fer+0.112
 generation+0.110
 planet+0.107
Neg logit contributions
 aback-0.081
wn-0.077
 like-0.075
 sweet-0.069
 swe-0.069
Tokenicy
Token ID4611
Feature activation+8.01e+00

Pos logit contributions
glass+0.112
orse+0.109
 generation+0.101
work+0.098
bike+0.097
Neg logit contributions
wn-0.089
 aback-0.084
coat-0.081
 swe-0.075
 like-0.075
Token food
Token ID2057
Feature activation-5.02e-01

Pos logit contributions
glass+0.515
orse+0.508
 generation+0.468
work+0.448
 planet+0.440
Neg logit contributions
wn-0.363
 aback-0.362
coat-0.322
 like-0.316
 swe-0.315
Token can
Token ID460
Feature activation-2.17e+00

Pos logit contributions
wn+0.022
 aback+0.021
coat+0.019
 swe+0.018
 ripe+0.017
Neg logit contributions
glass-0.033
orse-0.033
work-0.031
 planet-0.030
 generation-0.030
Token make
Token ID787
Feature activation-6.20e-01

Pos logit contributions
 aback+0.078
wn+0.078
 like+0.075
 swe+0.073
 sweet+0.072
Neg logit contributions
glass-0.158
orse-0.150
 generation-0.142
work-0.137
fer-0.136
Token your
Token ID534
Feature activation+7.66e-01

Pos logit contributions
wn+0.031
 aback+0.030
coat+0.027
 swe+0.026
 like+0.025
Neg logit contributions
glass-0.038
orse-0.037
 generation-0.034
work-0.033
 planet-0.033
Token tum
Token ID11814
Feature activation+1.37e+00

Pos logit contributions
glass+0.048
orse+0.047
 generation+0.042
work+0.042
bike+0.041
Neg logit contributions
wn-0.036
 aback-0.036
 swe-0.033
coat-0.032
 like-0.031
Token.
Token ID13
Feature activation-3.46e-01

Pos logit contributions
wn+0.047
 aback+0.047
coat+0.041
 swe+0.040
 like+0.039
Neg logit contributions
glass-0.102
orse-0.102
 generation-0.095
work-0.092
 planet-0.091
Token
Token ID198
Feature activation-2.09e-01

Pos logit contributions
wn+0.019
 aback+0.018
coat+0.016
 swe+0.015
 tast+0.015
Neg logit contributions
glass-0.020
orse-0.020
 generation-0.017
 planet-0.016
 Whale-0.016
Token
Token ID198
Feature activation-2.19e-01

Pos logit contributions
 aback+0.009
wn+0.009
coat+0.008
 swe+0.008
 like+0.008
Neg logit contributions
glass-0.014
orse-0.013
 generation-0.012
work-0.012
 planet-0.012
Token"
Token ID1
Feature activation-6.71e-01

Pos logit contributions
wn+0.010
 aback+0.009
coat+0.008
 swe+0.008
 like+0.008
Neg logit contributions
glass-0.015
orse-0.014
 generation-0.013
work-0.013
 planet-0.013
TokenSp
Token ID4561
Feature activation+2.37e+00

Pos logit contributions
wn+0.033
 aback+0.033
coat+0.030
 swe+0.029
 like+0.028
Neg logit contributions
glass-0.041
orse-0.041
 generation-0.037
work-0.036
 planet-0.035
Tokenicy
Token ID4611
Feature activation+8.00e+00

Pos logit contributions
glass+0.082
orse+0.079
 generation+0.067
work+0.064
 planet+0.062
Neg logit contributions
wn-0.180
 aback-0.178
coat-0.167
 swe-0.164
 like-0.161
Token is
Token ID318
Feature activation-1.42e-01

Pos logit contributions
glass+0.522
orse+0.513
 generation+0.472
work+0.462
 planet+0.452
Neg logit contributions
wn-0.357
 aback-0.355
coat-0.315
 swe-0.308
 like-0.296
Token good
Token ID922
Feature activation-2.95e+00

Pos logit contributions
wn+0.006
 aback+0.006
coat+0.006
inters+0.005
 swe+0.005
Neg logit contributions
orse-0.010
glass-0.009
work-0.009
 generation-0.008
 Whale-0.008
Token,
Token ID11
Feature activation+5.32e-02

Pos logit contributions
wn+0.141
 aback+0.129
coat+0.114
 swe+0.109
 ripe+0.107
Neg logit contributions
orse-0.189
glass-0.188
work-0.179
 Whale-0.168
bike-0.166
Token but
Token ID475
Feature activation-2.08e+00

Pos logit contributions
glass+0.004
orse+0.004
 generation+0.003
work+0.003
 planet+0.003
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.002
 swe-0.002
 like-0.002
Token not
Token ID407
Feature activation-1.03e-02

Pos logit contributions
 like+0.090
wn+0.088
 sweet+0.088
 aback+0.087
 swe+0.085
Neg logit contributions
glass-0.135
orse-0.127
 generation-0.120
bike-0.115
work-0.115
Token The
Token ID383
Feature activation+3.18e+00

Pos logit contributions
glass+0.142
orse+0.136
work+0.128
 generation+0.126
 planet+0.123
Neg logit contributions
 aback-0.075
wn-0.071
 like-0.066
coat-0.065
 swe-0.063
Token popcorn
Token ID38914
Feature activation+2.18e+00

Pos logit contributions
glass+0.193
orse+0.190
 generation+0.171
work+0.167
bike+0.160
Neg logit contributions
wn-0.151
 aback-0.148
 swe-0.142
coat-0.140
 sweet-0.138
Token was
Token ID373
Feature activation+4.67e-01

Pos logit contributions
glass+0.150
orse+0.148
 generation+0.139
work+0.133
 planet+0.130
Neg logit contributions
 aback-0.085
wn-0.085
coat-0.077
 swe-0.075
 like-0.074
Token sticky
Token ID23408
Feature activation-8.26e-01

Pos logit contributions
orse+0.032
glass+0.032
work+0.029
 generation+0.028
 whale+0.028
Neg logit contributions
 aback-0.020
wn-0.018
coat-0.016
 swe-0.014
ets-0.014
Token and
Token ID290
Feature activation+6.11e-01

Pos logit contributions
wn+0.038
 aback+0.037
coat+0.032
 swe+0.031
 prett+0.029
Neg logit contributions
glass-0.051
orse-0.050
work-0.048
 generation-0.045
bike-0.044
Token gre
Token ID10536
Feature activation-7.07e+00

Pos logit contributions
orse+0.044
glass+0.043
work+0.039
 generation+0.039
 planet+0.039
Neg logit contributions
 aback-0.024
wn-0.023
coat-0.020
 swe-0.017
ets-0.017
Tokenasy
Token ID4107
Feature activation+2.29e+00

Pos logit contributions
wn+0.350
 aback+0.347
 swe+0.307
coat+0.304
 fun+0.299
Neg logit contributions
glass-0.425
orse-0.416
work-0.390
 generation-0.371
fer-0.358
Token.
Token ID13
Feature activation-5.46e-02

Pos logit contributions
glass+0.150
orse+0.148
work+0.133
 generation+0.131
 planet+0.127
Neg logit contributions
wn-0.103
 aback-0.102
coat-0.087
 swe-0.081
 ripe-0.079
Token They
Token ID1119
Feature activation-4.89e-01

Pos logit contributions
 aback+0.002
wn+0.002
coat+0.002
 swe+0.002
 like+0.002
Neg logit contributions
glass-0.004
orse-0.004
work-0.003
 generation-0.003
 planet-0.003
Token used
Token ID973
Feature activation-1.72e+00

Pos logit contributions
wn+0.019
 aback+0.019
coat+0.017
 swe+0.016
 ripe+0.015
Neg logit contributions
glass-0.035
orse-0.034
 generation-0.031
work-0.031
 planet-0.030
Token a
Token ID257
Feature activation-3.08e-01

Pos logit contributions
wn+0.098
 aback+0.095
coat+0.093
ets+0.085
 swe+0.082
Neg logit contributions
glass-0.081
orse-0.081
work-0.075
 generation-0.071
 whale-0.068
Token bigger
Token ID5749
Feature activation-7.07e-01

Pos logit contributions
wn+0.123
coat+0.113
 aback+0.103
ucker+0.094
arl+0.093
Neg logit contributions
work-0.194
orse-0.193
glass-0.187
 generation-0.182
 planet-0.182
Token and
Token ID290
Feature activation+2.01e-01

Pos logit contributions
wn+0.031
 aback+0.029
coat+0.026
 swe+0.024
 ripe+0.023
Neg logit contributions
glass-0.047
orse-0.046
work-0.044
 generation-0.043
 planet-0.043
Token the
Token ID262
Feature activation-9.97e-01

Pos logit contributions
glass+0.013
orse+0.013
work+0.013
 generation+0.013
 planet+0.012
Neg logit contributions
wn-0.008
 aback-0.008
coat-0.007
inters-0.006
ets-0.006
Token trees
Token ID7150
Feature activation-2.10e+00

Pos logit contributions
wn+0.046
 aback+0.046
 swe+0.042
coat+0.041
 like+0.041
Neg logit contributions
glass-0.062
orse-0.062
 generation-0.054
work-0.053
bike-0.053
Token looked
Token ID3114
Feature activation+3.72e-01

Pos logit contributions
wn+0.092
 aback+0.091
coat+0.080
 swe+0.078
 ripe+0.074
Neg logit contributions
glass-0.138
orse-0.137
work-0.125
 generation-0.125
 planet-0.121
Token gre
Token ID10536
Feature activation-7.04e+00

Pos logit contributions
work+0.024
glass+0.024
orse+0.024
 generation+0.024
 planet+0.024
Neg logit contributions
wn-0.014
coat-0.014
 aback-0.013
ucker-0.011
arl-0.011
Tokenener
Token ID877
Feature activation-2.02e+00

Pos logit contributions
wn+0.334
 aback+0.329
coat+0.302
 swe+0.288
 fun+0.284
Neg logit contributions
glass-0.433
orse-0.415
work-0.411
 generation-0.395
fer-0.367
Token.
Token ID13
Feature activation-1.87e-02

Pos logit contributions
wn+0.073
 aback+0.072
coat+0.062
 swe+0.061
 like+0.060
Neg logit contributions
glass-0.148
orse-0.146
 generation-0.135
work-0.132
 planet-0.130
Token
Token ID198
Feature activation-8.37e-02

Pos logit contributions
 aback+0.001
wn+0.001
coat+0.001
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.001
orse-0.001
work-0.001
 generation-0.001
 planet-0.001
Token
Token ID198
Feature activation-1.13e-01

Pos logit contributions
 aback+0.004
wn+0.004
coat+0.003
 swe+0.003
 like+0.003
Neg logit contributions
glass-0.005
orse-0.005
work-0.005
 generation-0.005
 planet-0.005
TokenFinally
Token ID11158
Feature activation-7.08e-01

Pos logit contributions
wn+0.005
 aback+0.005
coat+0.004
 swe+0.004
 like+0.004
Neg logit contributions
glass-0.007
orse-0.007
work-0.007
 generation-0.006
bike-0.006
Token much
Token ID881
Feature activation-5.62e-01

Pos logit contributions
wn+0.103
 aback+0.101
coat+0.092
 swe+0.086
 tast+0.081
Neg logit contributions
glass-0.132
orse-0.131
 generation-0.121
work-0.117
 planet-0.116
Token or
Token ID393
Feature activation+8.07e-01

Pos logit contributions
wn+0.027
coat+0.025
 aback+0.025
ny+0.022
inters+0.022
Neg logit contributions
glass-0.028
orse-0.028
 movie-0.026
work-0.025
 planet-0.024
Token it
Token ID340
Feature activation+2.58e-02

Pos logit contributions
orse+0.050
glass+0.050
 generation+0.045
work+0.045
bike+0.044
Neg logit contributions
 aback-0.038
wn-0.037
coat-0.033
 swe-0.030
 ripe-0.029
Token will
Token ID481
Feature activation-2.21e+00

Pos logit contributions
glass+0.002
orse+0.002
work+0.002
 generation+0.001
 planet+0.001
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.001
 ripe-0.001
 swe-0.001
Token look
Token ID804
Feature activation-1.01e+00

Pos logit contributions
wn+0.109
 aback+0.108
coat+0.095
 swe+0.086
ets+0.083
Neg logit contributions
orse-0.132
glass-0.126
work-0.121
 generation-0.119
 planet-0.119
Token gre
Token ID10536
Feature activation-7.02e+00

Pos logit contributions
 aback+0.041
coat+0.039
wn+0.038
ere+0.032
arl+0.030
Neg logit contributions
orse-0.061
glass-0.059
work-0.057
 planet-0.056
 generation-0.056
Tokenasy
Token ID4107
Feature activation+2.30e+00

Pos logit contributions
 aback+0.356
wn+0.355
 fun+0.319
coat+0.307
 ripe+0.305
Neg logit contributions
glass-0.381
orse-0.356
work-0.354
fer-0.332
 generation-0.331
Token."
Token ID526
Feature activation+7.98e-01

Pos logit contributions
glass+0.136
orse+0.129
work+0.117
 generation+0.114
 planet+0.113
Neg logit contributions
wn-0.118
 aback-0.118
coat-0.101
 swe-0.090
 ripe-0.090
Token
Token ID220
Feature activation-1.93e+00

Pos logit contributions
orse+0.048
glass+0.047
 generation+0.042
work+0.041
bike+0.041
Neg logit contributions
wn-0.041
 aback-0.038
coat-0.035
 swe-0.035
 ripe-0.034
Token
Token ID198
Feature activation-4.75e-02

Pos logit contributions
wn+0.110
 aback+0.103
coat+0.097
 swe+0.096
 ripe+0.094
Neg logit contributions
orse-0.104
glass-0.102
 generation-0.089
work-0.087
bike-0.086
Token
Token ID198
Feature activation-6.87e-02

Pos logit contributions
wn+0.003
 aback+0.002
coat+0.002
 swe+0.002
 ripe+0.002
Neg logit contributions
orse-0.003
glass-0.003
 generation-0.002
bike-0.002
work-0.002
Token had
Token ID550
Feature activation+1.01e+00

Pos logit contributions
 aback+0.073
wn+0.073
coat+0.066
 swe+0.063
 like+0.061
Neg logit contributions
glass-0.114
orse-0.113
 generation-0.106
work-0.103
bike-0.100
Token grown
Token ID7334
Feature activation-2.19e+00

Pos logit contributions
glass+0.066
orse+0.065
 generation+0.060
work+0.059
 planet+0.058
Neg logit contributions
wn-0.044
 aback-0.043
coat-0.039
 swe-0.037
 ripe-0.034
Token to
Token ID284
Feature activation-3.35e-01

Pos logit contributions
wn+0.098
coat+0.094
arl+0.092
 aback+0.087
inters+0.080
Neg logit contributions
glass-0.128
work-0.124
 generation-0.117
orse-0.116
 planet-0.110
Token be
Token ID307
Feature activation+7.19e-01

Pos logit contributions
wn+0.013
 aback+0.013
 swe+0.011
 like+0.011
coat+0.011
Neg logit contributions
glass-0.024
orse-0.023
work-0.022
 generation-0.021
bike-0.021
Token even
Token ID772
Feature activation+2.04e+00

Pos logit contributions
glass+0.050
orse+0.050
 generation+0.045
work+0.044
 crew+0.042
Neg logit contributions
 aback-0.028
wn-0.027
coat-0.024
arl-0.022
 swe-0.022
Token gre
Token ID10536
Feature activation-7.01e+00

Pos logit contributions
orse+0.101
glass+0.099
 generation+0.093
work+0.092
 crew+0.086
Neg logit contributions
coat-0.106
wn-0.105
 aback-0.102
arl-0.094
inters-0.092
Tokenener
Token ID877
Feature activation-2.16e+00

Pos logit contributions
wn+0.300
 aback+0.295
coat+0.261
 swe+0.255
 ripe+0.242
Neg logit contributions
glass-0.474
orse-0.465
 generation-0.431
work-0.427
 planet-0.410
Token and
Token ID290
Feature activation+9.04e-02

Pos logit contributions
 aback+0.070
wn+0.070
coat+0.061
 like+0.059
 swe+0.058
Neg logit contributions
glass-0.164
orse-0.162
 generation-0.151
work-0.147
 planet-0.145
Token brighter
Token ID26083
Feature activation+5.80e-02

Pos logit contributions
glass+0.007
orse+0.007
 generation+0.006
work+0.006
 planet+0.006
Neg logit contributions
wn-0.003
 aback-0.003
coat-0.003
 swe-0.003
 like-0.003
Token than
Token ID621
Feature activation-3.94e+00

Pos logit contributions
glass+0.004
orse+0.004
 generation+0.004
work+0.004
 planet+0.004
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.002
 swe-0.002
 ripe-0.002
Token before
Token ID878
Feature activation+1.41e-01

Pos logit contributions
wn+0.201
 aback+0.199
coat+0.178
 swe+0.177
 like+0.175
Neg logit contributions
glass-0.233
orse-0.229
 generation-0.203
work-0.200
 planet-0.197
Token she
Token ID673
Feature activation-7.41e-01

Pos logit contributions
glass+0.042
orse+0.042
 generation+0.039
work+0.038
 planet+0.037
Neg logit contributions
wn-0.025
 aback-0.025
coat-0.022
 swe-0.021
 ripe-0.019
Token had
Token ID550
Feature activation+1.20e+00

Pos logit contributions
wn+0.035
 aback+0.034
coat+0.031
 swe+0.030
 ripe+0.028
Neg logit contributions
glass-0.047
orse-0.046
 generation-0.043
work-0.041
 planet-0.041
Token blue
Token ID4171
Feature activation-2.01e+00

Pos logit contributions
glass+0.076
orse+0.073
 generation+0.069
work+0.066
 planet+0.064
Neg logit contributions
wn-0.055
 aback-0.055
coat-0.050
 swe-0.048
 ripe-0.044
Token eyes
Token ID2951
Feature activation-2.62e+00

Pos logit contributions
wn+0.095
 aback+0.093
coat+0.080
 swe+0.078
 ripe+0.074
Neg logit contributions
glass-0.129
orse-0.126
work-0.117
 generation-0.115
 planet-0.114
Token and
Token ID290
Feature activation+1.02e-01

Pos logit contributions
wn+0.111
 aback+0.109
 swe+0.091
coat+0.089
 ripe+0.083
Neg logit contributions
glass-0.179
orse-0.171
work-0.164
 generation-0.159
bike-0.154
Token blonde
Token ID21541
Feature activation-7.00e+00

Pos logit contributions
glass+0.007
orse+0.007
 generation+0.006
work+0.006
 planet+0.006
Neg logit contributions
wn-0.004
 aback-0.004
coat-0.004
 swe-0.003
 like-0.003
Token hair
Token ID4190
Feature activation-2.97e+00

Pos logit contributions
 aback+0.299
wn+0.296
coat+0.235
 tast+0.229
 swe+0.223
Neg logit contributions
glass-0.480
orse-0.477
 generation-0.435
work-0.416
 planet-0.416
Token.
Token ID13
Feature activation-5.39e-03

Pos logit contributions
wn+0.110
 aback+0.108
coat+0.094
 swe+0.091
 like+0.086
Neg logit contributions
glass-0.217
orse-0.214
 generation-0.198
work-0.195
 planet-0.191
Token
Token ID198
Feature activation-1.50e-01

Pos logit contributions
 aback+0.000
wn+0.000
coat+0.000
 swe+0.000
 like+0.000
Neg logit contributions
glass-0.000
orse-0.000
work-0.000
 generation-0.000
 planet-0.000
Token
Token ID198
Feature activation-2.06e-01

Pos logit contributions
wn+0.007
 aback+0.007
coat+0.006
 swe+0.006
 like+0.006
Neg logit contributions
glass-0.010
orse-0.009
work-0.009
 generation-0.009
 planet-0.008
TokenOne
Token ID3198
Feature activation+2.01e+00

Pos logit contributions
wn+0.009
 aback+0.009
coat+0.008
 swe+0.007
 like+0.007
Neg logit contributions
glass-0.014
orse-0.013
work-0.013
 generation-0.012
 planet-0.012
Token much
Token ID881
Feature activation-5.23e-01

Pos logit contributions
wn+0.063
coat+0.058
 aback+0.058
ucker+0.045
arl+0.043
Neg logit contributions
glass-0.098
orse-0.098
work-0.096
 planet-0.093
 generation-0.092
Token better
Token ID1365
Feature activation-1.27e+00

Pos logit contributions
coat+0.032
wn+0.031
inters+0.026
ere+0.025
arl+0.024
Neg logit contributions
work-0.023
 whale-0.022
orse-0.021
 planet-0.021
 boat-0.020
Token!
Token ID0
Feature activation+2.02e+00

Pos logit contributions
wn+0.048
 aback+0.047
coat+0.041
 swe+0.039
 ripe+0.037
Neg logit contributions
glass-0.092
orse-0.091
 generation-0.084
work-0.083
 planet-0.081
Token They
Token ID1119
Feature activation+4.81e-02

Pos logit contributions
glass+0.137
orse+0.134
 generation+0.123
work+0.122
 planet+0.119
Neg logit contributions
 aback-0.087
wn-0.086
coat-0.076
 swe-0.074
 like-0.071
Token were
Token ID547
Feature activation+1.47e+00

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.003
work+0.003
bike+0.003
Neg logit contributions
wn-0.002
 aback-0.002
 swe-0.002
coat-0.002
 like-0.002
Token gre
Token ID10536
Feature activation-6.87e+00

Pos logit contributions
orse+0.095
glass+0.090
work+0.087
 whale+0.084
 planet+0.083
Neg logit contributions
wn-0.070
 aback-0.066
coat-0.063
inters-0.055
ets-0.053
Tokenener
Token ID877
Feature activation-1.67e+00

Pos logit contributions
wn+0.299
 aback+0.292
coat+0.260
 swe+0.257
 fun+0.243
Neg logit contributions
glass-0.456
orse-0.445
work-0.419
 generation-0.411
fer-0.389
Token and
Token ID290
Feature activation+1.53e-01

Pos logit contributions
wn+0.063
 aback+0.061
coat+0.053
 swe+0.051
 ripe+0.048
Neg logit contributions
glass-0.121
orse-0.120
 generation-0.110
work-0.110
 planet-0.107
Token taller
Token ID25242
Feature activation-6.89e-01

Pos logit contributions
glass+0.010
orse+0.010
 generation+0.009
work+0.009
 planet+0.009
Neg logit contributions
wn-0.007
 aback-0.007
 swe-0.006
coat-0.006
 like-0.006
Token than
Token ID621
Feature activation-3.65e+00

Pos logit contributions
wn+0.034
 aback+0.030
coat+0.028
 swe+0.025
inters+0.024
Neg logit contributions
glass-0.044
orse-0.044
work-0.041
 generation-0.038
 planet-0.038
Token before
Token ID878
Feature activation+4.93e-01

Pos logit contributions
wn+0.176
 aback+0.171
 like+0.157
 swe+0.155
coat+0.152
Neg logit contributions
glass-0.226
orse-0.222
 generation-0.197
 planet-0.192
bike-0.191
Token it
Token ID340
Feature activation+5.49e-02

Pos logit contributions
orse+0.263
glass+0.255
fer+0.252
 generation+0.250
field+0.250
Neg logit contributions
 aback-0.099
 sweet-0.090
 like-0.088
wn-0.087
 nice-0.086
Token again
Token ID757
Feature activation+5.48e+00

Pos logit contributions
glass+0.004
orse+0.004
 generation+0.004
work+0.004
 planet+0.004
Neg logit contributions
 aback-0.002
wn-0.002
coat-0.002
 swe-0.002
 like-0.002
Token,
Token ID11
Feature activation+5.02e-03

Pos logit contributions
glass+0.424
orse+0.416
 generation+0.390
work+0.384
 planet+0.375
Neg logit contributions
 aback-0.180
wn-0.179
coat-0.149
 swe-0.147
 like-0.146
Token this
Token ID428
Feature activation+2.01e+00

Pos logit contributions
glass+0.000
orse+0.000
 generation+0.000
work+0.000
 planet+0.000
Neg logit contributions
 aback-0.000
wn-0.000
 like-0.000
coat-0.000
 sweet-0.000
Token time
Token ID640
Feature activation-5.81e+00

Pos logit contributions
glass+0.169
orse+0.168
work+0.155
 generation+0.154
 planet+0.153
Neg logit contributions
wn-0.054
 aback-0.053
coat-0.042
 swe-0.041
 ripe-0.037
Token gre
Token ID10536
Feature activation-6.85e+00

Pos logit contributions
 like+0.225
 aback+0.223
wn+0.214
 past+0.206
 fun+0.195
Neg logit contributions
orse-0.381
glass-0.379
 generation-0.370
work-0.359
bike-0.354
Tokenated
Token ID515
Feature activation+5.56e+00

Pos logit contributions
wn+0.292
 aback+0.286
coat+0.254
 swe+0.251
 ripe+0.236
Neg logit contributions
glass-0.468
orse-0.458
 generation-0.422
work-0.420
 planet-0.402
Token and
Token ID290
Feature activation+4.71e-02

Pos logit contributions
glass+0.386
orse+0.372
 generation+0.346
work+0.345
 planet+0.338
Neg logit contributions
wn-0.229
 aback-0.223
coat-0.198
 swe-0.190
 ripe-0.180
Token louder
Token ID27089
Feature activation+5.91e-01

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.003
work+0.003
 planet+0.003
Neg logit contributions
 aback-0.002
wn-0.002
 swe-0.002
 sweet-0.002
 like-0.002
Token.
Token ID13
Feature activation-8.75e-02

Pos logit contributions
glass+0.047
orse+0.046
 generation+0.044
work+0.043
bike+0.042
Neg logit contributions
 aback-0.017
wn-0.016
 like-0.015
 sweet-0.014
 swe-0.014
Token The
Token ID383
Feature activation+2.43e+00

Pos logit contributions
wn+0.004
 aback+0.004
coat+0.004
 swe+0.004
 ripe+0.003
Neg logit contributions
glass-0.005
orse-0.005
 generation-0.005
work-0.005
 planet-0.005
Token't
Token ID470
Feature activation+1.74e+00

Pos logit contributions
glass+0.200
orse+0.185
 crew+0.176
 generation+0.174
fer+0.170
Neg logit contributions
 aback-0.088
 like-0.074
 time-0.073
 swe-0.073
wn-0.073
Token take
Token ID1011
Feature activation-4.71e-01

Pos logit contributions
glass+0.128
orse+0.120
work+0.114
ich+0.114
bike+0.113
Neg logit contributions
 like-0.065
wn-0.060
 aback-0.059
 sweet-0.051
 fun-0.051
Token them
Token ID606
Feature activation-2.40e+00

Pos logit contributions
wn+0.025
coat+0.025
 aback+0.024
 ripe+0.022
 swe+0.021
Neg logit contributions
glass-0.025
orse-0.025
 generation-0.024
 planet-0.023
work-0.022
Token all
Token ID477
Feature activation-1.70e+00

Pos logit contributions
wn+0.096
 aback+0.094
coat+0.083
 swe+0.080
 ripe+0.075
Neg logit contributions
glass-0.167
orse-0.164
 generation-0.152
work-0.151
 planet-0.148
Token.
Token ID13
Feature activation-8.64e-02

Pos logit contributions
wn+0.068
 aback+0.065
coat+0.059
 swe+0.056
 ripe+0.052
Neg logit contributions
orse-0.118
glass-0.118
work-0.109
 generation-0.109
 planet-0.106
Token Care
Token ID7276
Feature activation-6.85e+00

Pos logit contributions
wn+0.005
 aback+0.004
coat+0.004
 swe+0.004
 ripe+0.004
Neg logit contributions
glass-0.005
orse-0.005
 generation-0.005
work-0.004
 planet-0.004
Tokenfully
Token ID2759
Feature activation+3.47e+00

Pos logit contributions
 aback+0.314
wn+0.310
 like+0.280
coat+0.279
 swe+0.268
Neg logit contributions
glass-0.441
orse-0.431
 generation-0.395
work-0.378
 planet-0.374
Token,
Token ID11
Feature activation-3.76e-03

Pos logit contributions
glass+0.255
orse+0.251
 generation+0.233
work+0.230
 planet+0.225
Neg logit contributions
wn-0.127
 aback-0.126
coat-0.109
 swe-0.105
 like-0.100
Token it
Token ID340
Feature activation+1.34e-01

Pos logit contributions
 aback+0.000
 like+0.000
 past+0.000
 sweet+0.000
 ripe+0.000
Neg logit contributions
glass-0.000
orse-0.000
work-0.000
 generation-0.000
bike-0.000
Token picked
Token ID6497
Feature activation-1.49e+00

Pos logit contributions
glass+0.009
orse+0.009
 generation+0.009
work+0.008
 planet+0.008
Neg logit contributions
wn-0.005
 aback-0.005
coat-0.005
 swe-0.004
 ripe-0.004
Token out
Token ID503
Feature activation-1.60e+00

Pos logit contributions
wn+0.063
 aback+0.060
coat+0.055
 swe+0.052
 ripe+0.049
Neg logit contributions
glass-0.102
orse-0.099
 generation-0.093
work-0.092
 planet-0.089
Token They
Token ID1119
Feature activation-1.42e-01

Pos logit contributions
 aback+0.006
wn+0.006
 like+0.005
 swe+0.005
 sweet+0.005
Neg logit contributions
glass-0.007
orse-0.007
work-0.006
 generation-0.006
fer-0.006
Token are
Token ID389
Feature activation+1.14e+00

Pos logit contributions
 aback+0.007
wn+0.007
coat+0.007
 swe+0.006
 ripe+0.006
Neg logit contributions
orse-0.008
glass-0.008
work-0.007
 planet-0.007
 generation-0.007
Token too
Token ID1165
Feature activation-2.04e+00

Pos logit contributions
orse+0.072
glass+0.068
work+0.063
 planet+0.062
 whale+0.062
Neg logit contributions
 aback-0.057
wn-0.053
coat-0.052
 swe-0.041
ere-0.041
Token salty
Token ID36021
Feature activation+1.08e+00

Pos logit contributions
wn+0.087
 aback+0.085
coat+0.072
 swe+0.060
 prett+0.057
Neg logit contributions
orse-0.148
glass-0.136
 whale-0.128
 planet-0.127
 generation-0.124
Token and
Token ID290
Feature activation+1.74e-01

Pos logit contributions
glass+0.070
orse+0.068
work+0.063
 generation+0.062
 planet+0.061
Neg logit contributions
wn-0.049
 aback-0.048
coat-0.040
 swe-0.039
 prett-0.037
Token gre
Token ID10536
Feature activation-6.85e+00

Pos logit contributions
orse+0.013
glass+0.012
work+0.011
 whale+0.011
 generation+0.011
Neg logit contributions
 aback-0.007
wn-0.007
coat-0.006
ere-0.004
ets-0.004
Tokenasy
Token ID4107
Feature activation+2.55e+00

Pos logit contributions
wn+0.332
 aback+0.329
coat+0.294
 swe+0.289
 fun+0.280
Neg logit contributions
glass-0.420
orse-0.411
work-0.386
 generation-0.372
fer-0.355
Token.
Token ID13
Feature activation-1.25e-01

Pos logit contributions
glass+0.155
orse+0.153
work+0.134
 generation+0.130
 planet+0.129
Neg logit contributions
wn-0.129
 aback-0.129
coat-0.113
 swe-0.101
 prett-0.098
Token You
Token ID921
Feature activation+7.60e-01

Pos logit contributions
 aback+0.006
wn+0.006
 swe+0.006
 like+0.006
 sweet+0.005
Neg logit contributions
glass-0.008
orse-0.007
 generation-0.007
work-0.007
 planet-0.007
Token should
Token ID815
Feature activation+1.21e+00

Pos logit contributions
orse+0.050
glass+0.049
work+0.045
 planet+0.045
 generation+0.043
Neg logit contributions
 aback-0.034
wn-0.033
coat-0.030
 swe-0.029
 ripe-0.026
Token eat
Token ID4483
Feature activation+1.64e+00

Pos logit contributions
orse+0.071
glass+0.066
 generation+0.062
 planet+0.059
work+0.058
Neg logit contributions
 aback-0.066
wn-0.064
coat-0.056
 swe-0.054
 prett-0.051
Token cars
Token ID5006
Feature activation+7.11e-01

Pos logit contributions
glass+0.072
orse+0.064
 planet+0.058
 ship+0.055
 crew+0.053
Neg logit contributions
 aback-0.113
wn-0.109
coat-0.099
 tast-0.095
ucker-0.093
Token of
Token ID286
Feature activation+4.82e-01

Pos logit contributions
glass+0.048
orse+0.045
work+0.042
 planet+0.041
bike+0.040
Neg logit contributions
 aback-0.031
wn-0.031
 swe-0.029
 ripe-0.027
 tast-0.027
Token different
Token ID1180
Feature activation-1.08e+00

Pos logit contributions
glass+0.027
 planet+0.024
orse+0.024
 generation+0.023
 ship+0.022
Neg logit contributions
 aback-0.026
wn-0.026
coat-0.024
ucker-0.023
ited-0.022
Token colors
Token ID7577
Feature activation-5.08e+00

Pos logit contributions
 aback+0.047
wn+0.046
coat+0.041
 ripe+0.040
 tast+0.039
Neg logit contributions
glass-0.074
orse-0.067
work-0.066
 planet-0.066
 generation-0.063
Token and
Token ID290
Feature activation-2.78e-01

Pos logit contributions
wn+0.206
 aback+0.205
coat+0.174
 swe+0.171
 ripe+0.162
Neg logit contributions
glass-0.359
orse-0.350
work-0.328
 planet-0.316
 generation-0.315
Token sizes
Token ID10620
Feature activation-6.84e+00

Pos logit contributions
wn+0.011
 aback+0.011
coat+0.010
 swe+0.009
 ripe+0.009
Neg logit contributions
glass-0.020
orse-0.019
 generation-0.017
work-0.017
 planet-0.017
Token.
Token ID13
Feature activation-6.33e-01

Pos logit contributions
wn+0.299
 aback+0.298
coat+0.258
 swe+0.253
 ripe+0.239
Neg logit contributions
glass-0.463
orse-0.450
work-0.420
 planet-0.405
 generation-0.405
Token He
Token ID679
Feature activation-6.68e-01

Pos logit contributions
wn+0.026
 aback+0.026
coat+0.023
 swe+0.023
 like+0.022
Neg logit contributions
glass-0.043
orse-0.042
 generation-0.039
work-0.039
 planet-0.037
Token likes
Token ID7832
Feature activation-1.11e+00

Pos logit contributions
wn+0.027
 aback+0.027
coat+0.024
 swe+0.023
 ripe+0.022
Neg logit contributions
glass-0.046
orse-0.046
 generation-0.042
work-0.042
 planet-0.041
Token to
Token ID284
Feature activation-4.39e-01

Pos logit contributions
 aback+0.046
wn+0.046
coat+0.039
 swe+0.036
 ripe+0.035
Neg logit contributions
glass-0.078
orse-0.077
 generation-0.070
work-0.070
 planet-0.069
Token make
Token ID787
Feature activation-6.46e-01

Pos logit contributions
wn+0.019
 aback+0.019
coat+0.017
 swe+0.016
 ripe+0.015
Neg logit contributions
glass-0.029
orse-0.029
 generation-0.027
work-0.026
 planet-0.026
Token to
Token ID284
Feature activation-4.40e-01

Pos logit contributions
glass+0.121
orse+0.120
 generation+0.112
work+0.111
 planet+0.108
Neg logit contributions
 aback-0.040
wn-0.040
coat-0.032
 swe-0.032
 like-0.030
Token follow
Token ID1061
Feature activation-1.45e+00

Pos logit contributions
wn+0.017
 aback+0.017
coat+0.015
 like+0.015
 swe+0.015
Neg logit contributions
glass-0.031
orse-0.030
work-0.028
 generation-0.028
bike-0.027
Token the
Token ID262
Feature activation-6.00e-01

Pos logit contributions
 aback+0.047
wn+0.044
 like+0.039
 swe+0.038
 past+0.037
Neg logit contributions
glass-0.115
orse-0.114
work-0.104
 generation-0.103
bike-0.102
Token lights
Token ID7588
Feature activation-1.85e+00

Pos logit contributions
wn+0.030
 aback+0.030
coat+0.027
 swe+0.026
 ripe+0.025
Neg logit contributions
glass-0.037
orse-0.036
 generation-0.033
work-0.033
 planet-0.032
Token.
Token ID13
Feature activation-2.33e-02

Pos logit contributions
wn+0.059
 aback+0.056
 like+0.053
coat+0.051
 swe+0.047
Neg logit contributions
orse-0.144
glass-0.142
 generation-0.131
 Whale-0.127
bike-0.125
Token Care
Token ID7276
Feature activation-6.84e+00

Pos logit contributions
wn+0.001
 aback+0.001
coat+0.001
 swe+0.001
 ripe+0.001
Neg logit contributions
glass-0.002
 generation-0.001
orse-0.001
work-0.001
 planet-0.001
Tokenfully
Token ID2759
Feature activation+3.60e+00

Pos logit contributions
wn+0.359
 aback+0.354
coat+0.321
 swe+0.316
 ripe+0.303
Neg logit contributions
glass-0.394
orse-0.387
 generation-0.352
work-0.347
 planet-0.336
Token she
Token ID673
Feature activation-8.64e-01

Pos logit contributions
glass+0.262
orse+0.258
 generation+0.235
work+0.233
 planet+0.227
Neg logit contributions
wn-0.137
 aback-0.132
coat-0.118
 swe-0.114
 ripe-0.107
Token opened
Token ID4721
Feature activation-1.71e+00

Pos logit contributions
wn+0.032
 aback+0.032
coat+0.028
 swe+0.027
 like+0.026
Neg logit contributions
glass-0.063
orse-0.062
 generation-0.057
work-0.057
 planet-0.055
Token the
Token ID262
Feature activation-6.34e-01

Pos logit contributions
wn+0.086
 aback+0.081
coat+0.077
 swe+0.071
 tast+0.069
Neg logit contributions
glass-0.103
orse-0.099
 generation-0.095
work-0.093
 planet-0.092
Token door
Token ID3420
Feature activation+8.70e-01

Pos logit contributions
wn+0.036
 aback+0.035
coat+0.033
 swe+0.032
 ripe+0.031
Neg logit contributions
glass-0.035
orse-0.031
work-0.030
 planet-0.030
 generation-0.030
Token,
Token ID11
Feature activation+1.90e-02

Pos logit contributions
wn+0.046
 aback+0.044
coat+0.039
 swe+0.038
 ripe+0.036
Neg logit contributions
glass-0.086
orse-0.085
 generation-0.078
work-0.078
 planet-0.075
Token holding
Token ID4769
Feature activation+3.76e-01

Pos logit contributions
glass+0.001
orse+0.001
 generation+0.001
work+0.001
 planet+0.001
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.001
 swe-0.001
 like-0.001
Token a
Token ID257
Feature activation-4.48e-01

Pos logit contributions
glass+0.024
orse+0.023
 generation+0.022
work+0.021
 planet+0.021
Neg logit contributions
wn-0.017
 aback-0.016
coat-0.015
 swe-0.014
 ripe-0.014
Token doll
Token ID3654
Feature activation+2.44e-01

Pos logit contributions
 aback+0.020
wn+0.020
ets+0.017
inters+0.016
 tast+0.015
Neg logit contributions
glass-0.029
orse-0.028
 planet-0.026
 whale-0.025
 ship-0.024
Token with
Token ID351
Feature activation+4.13e-01

Pos logit contributions
orse+0.019
glass+0.018
 generation+0.017
 Whale+0.017
work+0.017
Neg logit contributions
wn-0.008
coat-0.007
 like-0.007
 aback-0.007
 past-0.006
Token blonde
Token ID21541
Feature activation-6.83e+00

Pos logit contributions
glass+0.028
orse+0.027
 generation+0.025
work+0.025
 planet+0.024
Neg logit contributions
wn-0.018
 aback-0.017
coat-0.016
 swe-0.015
 like-0.014
Token hair
Token ID4190
Feature activation-3.10e+00

Pos logit contributions
 aback+0.263
wn+0.260
coat+0.211
 tast+0.201
 swe+0.201
Neg logit contributions
glass-0.496
orse-0.490
 generation-0.441
 planet-0.429
work-0.426
Token.
Token ID13
Feature activation-1.35e-01

Pos logit contributions
wn+0.117
 aback+0.117
coat+0.098
 swe+0.096
 ripe+0.090
Neg logit contributions
glass-0.224
orse-0.221
 generation-0.203
work-0.202
 planet-0.196
Token
Token ID198
Feature activation-2.45e-01

Pos logit contributions
wn+0.007
 aback+0.007
coat+0.006
 swe+0.006
 ripe+0.006
Neg logit contributions
glass-0.008
orse-0.008
 generation-0.007
work-0.007
 planet-0.007
Token
Token ID198
Feature activation-2.70e-01

Pos logit contributions
 aback+0.010
wn+0.010
coat+0.009
 swe+0.009
 like+0.008
Neg logit contributions
glass-0.017
orse-0.016
work-0.015
 generation-0.015
bike-0.014
Token"
Token ID1
Feature activation-7.18e-01

Pos logit contributions
wn+0.012
 aback+0.012
coat+0.010
 swe+0.010
 like+0.010
Neg logit contributions
glass-0.018
orse-0.018
work-0.016
 generation-0.016
 planet-0.015
Tokenie
Token ID494
Feature activation-1.53e+00

Pos logit contributions
glass+0.036
orse+0.036
 generation+0.033
work+0.033
 planet+0.032
Neg logit contributions
wn-0.020
 aback-0.019
coat-0.017
 swe-0.017
 like-0.016
Token got
Token ID1392
Feature activation-4.25e-01

Pos logit contributions
 aback+0.061
wn+0.061
coat+0.054
 swe+0.053
 like+0.050
Neg logit contributions
glass-0.106
orse-0.104
 generation-0.097
work-0.095
 planet-0.093
Token an
Token ID281
Feature activation+4.67e-01

Pos logit contributions
wn+0.021
coat+0.020
ets+0.018
 aback+0.018
inters+0.016
Neg logit contributions
glass-0.025
orse-0.024
work-0.023
 generation-0.022
 whale-0.021
Token idea
Token ID2126
Feature activation-5.45e+00

Pos logit contributions
glass+0.038
orse+0.038
 generation+0.036
work+0.035
 planet+0.035
Neg logit contributions
wn-0.012
 aback-0.011
coat-0.011
 swe-0.009
 ripe-0.008
Token.
Token ID13
Feature activation-1.43e-01

Pos logit contributions
wn+0.240
 aback+0.223
coat+0.205
 swe+0.201
 prett+0.192
Neg logit contributions
glass-0.365
orse-0.359
work-0.326
 generation-0.320
 planet-0.312
Token Care
Token ID7276
Feature activation-6.82e+00

Pos logit contributions
wn+0.006
 aback+0.005
coat+0.005
 swe+0.005
 ripe+0.005
Neg logit contributions
glass-0.010
orse-0.009
 generation-0.009
work-0.009
bike-0.009
Tokenfully
Token ID2759
Feature activation+3.25e+00

Pos logit contributions
wn+0.354
 aback+0.344
coat+0.311
 swe+0.310
 ripe+0.300
Neg logit contributions
glass-0.393
orse-0.381
work-0.358
 generation-0.353
 planet-0.337
Token,
Token ID11
Feature activation-3.69e-02

Pos logit contributions
glass+0.230
orse+0.228
work+0.205
 generation+0.204
 planet+0.197
Neg logit contributions
wn-0.130
 aback-0.124
coat-0.113
 swe-0.107
 ripe-0.100
Token she
Token ID673
Feature activation-1.03e+00

Pos logit contributions
 aback+0.002
wn+0.002
coat+0.002
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.002
orse-0.002
 generation-0.002
work-0.002
 planet-0.002
Token placed
Token ID4624
Feature activation-2.08e+00

Pos logit contributions
wn+0.040
 aback+0.039
coat+0.033
 swe+0.032
 ripe+0.030
Neg logit contributions
orse-0.075
glass-0.074
work-0.067
 generation-0.066
 planet-0.065
Token the
Token ID262
Feature activation-8.53e-01

Pos logit contributions
wn+0.110
 aback+0.104
coat+0.100
 swe+0.093
ets+0.091
Neg logit contributions
glass-0.117
orse-0.112
work-0.107
 generation-0.106
 planet-0.099
Token trains
Token ID13404
Feature activation+2.21e+00

Pos logit contributions
glass+0.121
orse+0.117
 planet+0.108
 whale+0.104
 generation+0.104
Neg logit contributions
wn-0.103
 aback-0.102
coat-0.091
ucker-0.079
 tast-0.078
Token of
Token ID286
Feature activation+5.99e-01

Pos logit contributions
glass+0.160
orse+0.158
 generation+0.146
work+0.145
 planet+0.141
Neg logit contributions
wn-0.082
 aback-0.081
coat-0.070
 swe-0.069
 ripe-0.064
Token different
Token ID1180
Feature activation-8.58e-01

Pos logit contributions
glass+0.035
orse+0.033
 generation+0.030
 planet+0.030
work+0.030
Neg logit contributions
 aback-0.032
wn-0.031
coat-0.028
 swe-0.027
 ripe-0.026
Token colors
Token ID7577
Feature activation-4.90e+00

Pos logit contributions
 aback+0.032
wn+0.032
coat+0.027
 swe+0.026
 ripe+0.026
Neg logit contributions
glass-0.063
orse-0.061
work-0.058
 generation-0.057
 planet-0.056
Token and
Token ID290
Feature activation-5.82e-02

Pos logit contributions
wn+0.186
 aback+0.183
coat+0.156
 swe+0.153
 ripe+0.143
Neg logit contributions
glass-0.356
orse-0.350
work-0.325
 generation-0.320
 planet-0.314
Token sizes
Token ID10620
Feature activation-6.80e+00

Pos logit contributions
wn+0.002
 aback+0.002
coat+0.002
 swe+0.002
 ripe+0.002
Neg logit contributions
glass-0.004
orse-0.004
 generation-0.004
work-0.004
 planet-0.004
Token.
Token ID13
Feature activation-3.42e-01

Pos logit contributions
wn+0.269
 aback+0.266
coat+0.229
 swe+0.225
 ripe+0.209
Neg logit contributions
glass-0.483
orse-0.475
work-0.440
 generation-0.433
 planet-0.426
Token He
Token ID679
Feature activation-5.78e-01

Pos logit contributions
 aback+0.014
wn+0.014
coat+0.012
 swe+0.012
 like+0.012
Neg logit contributions
glass-0.024
orse-0.023
work-0.022
 generation-0.021
 planet-0.021
Token makes
Token ID1838
Feature activation-1.68e+00

Pos logit contributions
wn+0.023
 aback+0.023
coat+0.020
 swe+0.019
 like+0.018
Neg logit contributions
glass-0.041
orse-0.040
 generation-0.037
work-0.036
 planet-0.036
Token a
Token ID257
Feature activation-2.92e-01

Pos logit contributions
wn+0.076
 aback+0.075
coat+0.067
 swe+0.066
 like+0.064
Neg logit contributions
glass-0.109
orse-0.107
 generation-0.099
work-0.097
 planet-0.094
Token track
Token ID2610
Feature activation+1.27e+00

Pos logit contributions
wn+0.011
 aback+0.010
 swe+0.009
coat+0.009
 sweet+0.009
Neg logit contributions
glass-0.021
orse-0.021
 generation-0.020
work-0.019
bike-0.019
Token cars
Token ID5006
Feature activation+7.94e-01

Pos logit contributions
glass+0.075
orse+0.067
 planet+0.059
 ship+0.057
 crew+0.056
Neg logit contributions
 aback-0.117
wn-0.113
coat-0.104
 tast-0.099
 ju-0.098
Token of
Token ID286
Feature activation+5.26e-01

Pos logit contributions
glass+0.054
orse+0.051
work+0.048
 planet+0.046
bike+0.046
Neg logit contributions
 aback-0.035
wn-0.034
 swe-0.031
 ripe-0.029
coat-0.029
Token different
Token ID1180
Feature activation-1.04e+00

Pos logit contributions
glass+0.027
orse+0.024
 planet+0.023
 generation+0.023
work+0.022
Neg logit contributions
 aback-0.031
wn-0.031
coat-0.029
ucker-0.027
 tast-0.026
Token colors
Token ID7577
Feature activation-5.02e+00

Pos logit contributions
 aback+0.045
wn+0.045
coat+0.039
 ripe+0.039
 tast+0.038
Neg logit contributions
glass-0.071
orse-0.065
work-0.064
 planet-0.063
 generation-0.061
Token and
Token ID290
Feature activation-1.15e-01

Pos logit contributions
wn+0.195
 aback+0.194
coat+0.164
 swe+0.161
 ripe+0.152
Neg logit contributions
glass-0.361
orse-0.354
work-0.329
 generation-0.321
 planet-0.318
Token sizes
Token ID10620
Feature activation-6.80e+00

Pos logit contributions
 aback+0.004
wn+0.004
coat+0.004
 swe+0.003
 ripe+0.003
Neg logit contributions
glass-0.008
orse-0.008
 generation-0.007
work-0.007
 planet-0.007
Token.
Token ID13
Feature activation-4.17e-01

Pos logit contributions
 aback+0.292
wn+0.292
coat+0.252
 swe+0.247
 ripe+0.233
Neg logit contributions
glass-0.464
orse-0.453
work-0.422
 generation-0.407
 planet-0.407
Token He
Token ID679
Feature activation-6.16e-01

Pos logit contributions
wn+0.017
 aback+0.017
coat+0.015
 swe+0.015
 like+0.014
Neg logit contributions
glass-0.029
orse-0.028
 generation-0.026
work-0.026
 planet-0.025
Token likes
Token ID7832
Feature activation-1.08e+00

Pos logit contributions
wn+0.026
 aback+0.026
coat+0.023
 swe+0.021
 ripe+0.021
Neg logit contributions
glass-0.042
orse-0.041
 generation-0.038
work-0.038
 planet-0.037
Token to
Token ID284
Feature activation-4.35e-01

Pos logit contributions
 aback+0.046
wn+0.046
coat+0.040
 swe+0.037
 ripe+0.036
Neg logit contributions
glass-0.074
orse-0.073
 generation-0.067
work-0.066
 planet-0.065
Token make
Token ID787
Feature activation-6.53e-01

Pos logit contributions
 aback+0.019
wn+0.019
coat+0.016
 swe+0.016
 ripe+0.015
Neg logit contributions
glass-0.029
orse-0.029
 generation-0.026
work-0.026
 planet-0.025
Token years
Token ID812
Feature activation-1.77e+00

Pos logit contributions
orse+0.079
glass+0.076
 generation+0.063
 planet+0.062
 whale+0.060
Neg logit contributions
wn-0.138
 aback-0.129
coat-0.125
inters-0.114
arl-0.112
Token old
Token ID1468
Feature activation-2.60e+00

Pos logit contributions
wn+0.067
 aback+0.064
coat+0.055
 swe+0.048
 tast+0.046
Neg logit contributions
glass-0.127
orse-0.126
work-0.120
 generation-0.120
 planet-0.115
Token and
Token ID290
Feature activation+2.74e-01

Pos logit contributions
wn+0.120
 aback+0.105
coat+0.097
 swe+0.091
inters+0.087
Neg logit contributions
glass-0.173
orse-0.166
work-0.160
 generation-0.158
 planet-0.153
Token had
Token ID550
Feature activation+1.05e+00

Pos logit contributions
glass+0.019
orse+0.019
 generation+0.017
work+0.017
 planet+0.016
Neg logit contributions
wn-0.011
 aback-0.011
coat-0.010
 swe-0.010
 ripe-0.009
Token beautiful
Token ID4950
Feature activation-7.18e-01

Pos logit contributions
glass+0.062
orse+0.060
 generation+0.059
 planet+0.055
work+0.054
Neg logit contributions
wn-0.049
 aback-0.048
coat-0.044
arl-0.041
ucker-0.040
Token blonde
Token ID21541
Feature activation-6.78e+00

Pos logit contributions
wn+0.035
 aback+0.033
coat+0.029
arl+0.026
inters+0.026
Neg logit contributions
glass-0.044
orse-0.043
 planet-0.042
 generation-0.041
work-0.040
Token hair
Token ID4190
Feature activation-2.91e+00

Pos logit contributions
wn+0.266
 aback+0.263
coat+0.214
 swe+0.201
 tast+0.199
Neg logit contributions
glass-0.482
orse-0.475
 generation-0.444
work-0.429
 planet-0.426
Token.
Token ID13
Feature activation-1.01e-01

Pos logit contributions
wn+0.102
 aback+0.100
coat+0.089
 swe+0.085
 like+0.082
Neg logit contributions
glass-0.216
orse-0.214
 generation-0.199
work-0.194
 planet-0.191
Token
Token ID220
Feature activation-2.10e+00

Pos logit contributions
 aback+0.004
wn+0.004
coat+0.003
 swe+0.003
 like+0.003
Neg logit contributions
glass-0.007
orse-0.007
work-0.006
 generation-0.006
 planet-0.006
Token
Token ID198
Feature activation-1.64e-01

Pos logit contributions
wn+0.109
 aback+0.106
coat+0.097
 swe+0.096
 ripe+0.093
Neg logit contributions
glass-0.122
orse-0.120
 generation-0.109
work-0.106
 planet-0.103
TokenOne
Token ID3198
Feature activation+2.01e+00

Pos logit contributions
wn+0.008
 aback+0.008
coat+0.007
 swe+0.007
 like+0.007
Neg logit contributions
glass-0.010
orse-0.010
 generation-0.009
work-0.009
 planet-0.009
Token saw
Token ID2497
Feature activation+1.16e+00

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.003
work+0.003
 planet+0.003
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.001
 swe-0.001
 like-0.001
Token a
Token ID257
Feature activation-3.14e-01

Pos logit contributions
orse+0.086
glass+0.086
 generation+0.080
work+0.077
fer+0.076
Neg logit contributions
 aback-0.043
 sweet-0.038
wn-0.038
 like-0.038
 fun-0.037
Token big
Token ID1263
Feature activation-1.60e+00

Pos logit contributions
wn+0.012
 aback+0.011
coat+0.010
 swe+0.009
 like+0.009
Neg logit contributions
glass-0.023
orse-0.023
 generation-0.021
work-0.021
 planet-0.021
Token tree
Token ID5509
Feature activation+6.07e-01

Pos logit contributions
wn+0.075
 aback+0.074
coat+0.067
 swe+0.065
 like+0.063
Neg logit contributions
glass-0.101
orse-0.100
 generation-0.091
work-0.090
 planet-0.087
Token.
Token ID13
Feature activation-2.96e-02

Pos logit contributions
glass+0.045
orse+0.044
 generation+0.041
work+0.040
 planet+0.039
Neg logit contributions
wn-0.021
 aback-0.021
coat-0.019
 like-0.018
 swe-0.018
Token Care
Token ID7276
Feature activation-6.78e+00

Pos logit contributions
wn+0.002
 aback+0.002
coat+0.001
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.002
orse-0.002
 generation-0.002
work-0.002
 planet-0.001
Tokenfully
Token ID2759
Feature activation+3.38e+00

Pos logit contributions
 aback+0.295
wn+0.291
 like+0.264
coat+0.260
 swe+0.243
Neg logit contributions
glass-0.454
orse-0.447
 generation-0.407
work-0.389
bike-0.385
Token the
Token ID262
Feature activation-7.23e-01

Pos logit contributions
glass+0.245
orse+0.242
 generation+0.223
work+0.220
 planet+0.215
Neg logit contributions
wn-0.128
 aback-0.126
coat-0.110
 swe-0.106
 like-0.100
Token tire
Token ID15867
Feature activation+1.72e-01

Pos logit contributions
wn+0.040
 aback+0.039
ets+0.035
 swe+0.033
coat+0.033
Neg logit contributions
glass-0.039
bike-0.036
orse-0.036
 planet-0.034
 generation-0.033
Token opened
Token ID4721
Feature activation-1.76e+00

Pos logit contributions
glass+0.011
orse+0.011
 generation+0.010
work+0.010
 planet+0.010
Neg logit contributions
 aback-0.007
wn-0.007
coat-0.007
 swe-0.006
 like-0.006
Token the
Token ID262
Feature activation-7.07e-01

Pos logit contributions
wn+0.092
 aback+0.081
coat+0.080
 swe+0.073
ets+0.072
Neg logit contributions
glass-0.101
orse-0.100
 generation-0.095
work-0.094
 planet-0.093
Token different
Token ID1180
Feature activation-1.25e+00

Pos logit contributions
orse+0.095
glass+0.092
work+0.086
 generation+0.083
 planet+0.083
Neg logit contributions
 aback-0.076
wn-0.075
coat-0.066
inters-0.061
 swe-0.059
Token colors
Token ID7577
Feature activation-5.12e+00

Pos logit contributions
wn+0.064
 aback+0.056
coat+0.051
inters+0.046
arl+0.045
Neg logit contributions
work-0.076
 planet-0.074
glass-0.074
orse-0.071
 generation-0.070
Token,
Token ID11
Feature activation+1.83e-02

Pos logit contributions
wn+0.210
 aback+0.207
coat+0.173
 swe+0.170
 ripe+0.157
Neg logit contributions
glass-0.354
orse-0.353
work-0.333
 planet-0.320
 generation-0.318
Token shapes
Token ID15268
Feature activation-2.91e+00

Pos logit contributions
glass+0.001
orse+0.001
 generation+0.001
work+0.001
 planet+0.001
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.000
 swe-0.000
 ripe-0.000
Token and
Token ID290
Feature activation+8.34e-02

Pos logit contributions
wn+0.129
 aback+0.125
coat+0.107
 swe+0.104
inters+0.097
Neg logit contributions
glass-0.193
orse-0.192
work-0.183
 planet-0.175
 generation-0.169
Token sizes
Token ID10620
Feature activation-6.77e+00

Pos logit contributions
glass+0.006
orse+0.006
 generation+0.005
work+0.005
 planet+0.005
Neg logit contributions
wn-0.003
 aback-0.003
coat-0.003
 swe-0.002
 ripe-0.002
Token.
Token ID13
Feature activation-2.26e-01

Pos logit contributions
wn+0.303
 aback+0.301
coat+0.256
 swe+0.250
 ripe+0.233
Neg logit contributions
glass-0.445
orse-0.441
work-0.416
 planet-0.396
 generation-0.393
Token Everyone
Token ID11075
Feature activation+8.91e-01

Pos logit contributions
 aback+0.010
wn+0.010
coat+0.009
 like+0.009
 swe+0.009
Neg logit contributions
glass-0.015
orse-0.015
work-0.013
 generation-0.013
bike-0.013
Token was
Token ID373
Feature activation+5.13e-02

Pos logit contributions
glass+0.063
orse+0.061
 generation+0.058
work+0.055
 planet+0.054
Neg logit contributions
wn-0.034
 aback-0.034
 like-0.030
coat-0.030
 swe-0.030
Token amazed
Token ID24562
Feature activation+1.78e+00

Pos logit contributions
orse+0.003
glass+0.003
work+0.003
 generation+0.003
 planet+0.003
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.002
 swe-0.002
inters-0.002
Token at
Token ID379
Feature activation+2.42e+00

Pos logit contributions
work+0.109
glass+0.103
orse+0.102
 whale+0.100
 planet+0.099
Neg logit contributions
wn-0.094
coat-0.085
 aback-0.082
inters-0.070
 swe-0.068
Token her
Token ID607
Feature activation-1.77e+00

Pos logit contributions
glass+0.066
orse+0.064
 generation+0.060
work+0.059
 planet+0.058
Neg logit contributions
wn-0.041
 aback-0.040
coat-0.035
 swe-0.033
 ripe-0.032
Token rain
Token ID6290
Feature activation-3.58e-01

Pos logit contributions
wn+0.073
 aback+0.072
coat+0.063
 swe+0.061
 like+0.058
Neg logit contributions
glass-0.121
orse-0.120
 generation-0.110
work-0.108
 planet-0.106
Token boots
Token ID14412
Feature activation-3.70e+00

Pos logit contributions
 aback+0.008
wn+0.008
 ripe+0.007
 tast+0.006
 juicy+0.006
Neg logit contributions
glass-0.031
orse-0.029
work-0.028
bike-0.028
 generation-0.027
Token and
Token ID290
Feature activation+2.29e-01

Pos logit contributions
wn+0.139
 aback+0.134
 swe+0.113
coat+0.111
 ripe+0.109
Neg logit contributions
glass-0.269
orse-0.261
work-0.243
 generation-0.243
bike-0.236
Token rain
Token ID6290
Feature activation-3.53e-01

Pos logit contributions
glass+0.015
orse+0.015
 generation+0.013
work+0.013
 planet+0.013
Neg logit contributions
wn-0.010
 aback-0.010
coat-0.009
 swe-0.009
 like-0.008
Tokencoat
Token ID31434
Feature activation-6.77e+00

Pos logit contributions
 aback+0.010
wn+0.009
 ripe+0.008
 tast+0.007
 juicy+0.007
Neg logit contributions
glass-0.028
work-0.027
orse-0.026
bike-0.026
 generation-0.025
Token so
Token ID523
Feature activation-1.42e+00

Pos logit contributions
wn+0.269
 aback+0.263
 swe+0.218
 ripe+0.217
coat+0.211
Neg logit contributions
glass-0.475
orse-0.460
work-0.429
 generation-0.425
bike-0.423
Token she
Token ID673
Feature activation-1.07e+00

Pos logit contributions
wn+0.083
 aback+0.082
coat+0.073
 swe+0.070
 ripe+0.070
Neg logit contributions
orse-0.071
glass-0.071
 generation-0.063
work-0.062
 planet-0.061
Token wouldn
Token ID3636
Feature activation-2.26e+00

Pos logit contributions
wn+0.042
 aback+0.042
coat+0.036
 swe+0.035
 ripe+0.034
Neg logit contributions
glass-0.075
orse-0.075
 generation-0.068
work-0.068
 planet-0.066
Token't
Token ID470
Feature activation+1.75e+00

Pos logit contributions
 aback+0.116
wn+0.112
coat+0.110
 like+0.107
 swe+0.100
Neg logit contributions
glass-0.134
orse-0.126
 generation-0.120
work-0.117
 planet-0.112
Token get
Token ID651
Feature activation+2.62e-01

Pos logit contributions
orse+0.107
glass+0.103
 generation+0.093
work+0.092
 planet+0.092
Neg logit contributions
 aback-0.087
wn-0.086
 swe-0.073
 ripe-0.071
coat-0.071
Token A
Token ID317
Feature activation+1.73e+00

Pos logit contributions
glass+0.189
orse+0.186
 generation+0.169
work+0.166
 planet+0.162
Neg logit contributions
wn-0.147
 aback-0.146
coat-0.130
 swe-0.127
 like-0.123
Token small
Token ID1402
Feature activation-2.91e+00

Pos logit contributions
glass+0.113
orse+0.111
 generation+0.102
work+0.100
 planet+0.097
Neg logit contributions
 aback-0.076
wn-0.075
coat-0.067
 swe-0.066
 sweet-0.063
Token girl
Token ID2576
Feature activation-3.10e+00

Pos logit contributions
 aback+0.151
wn+0.149
coat+0.137
 swe+0.134
 like+0.131
Neg logit contributions
glass-0.168
orse-0.165
 generation-0.149
work-0.144
bike-0.139
Token with
Token ID351
Feature activation+3.60e-01

Pos logit contributions
wn+0.132
 aback+0.130
coat+0.115
 swe+0.112
 like+0.106
Neg logit contributions
glass-0.209
orse-0.206
 generation-0.190
work-0.187
 planet-0.182
Token curly
Token ID45731
Feature activation-3.52e+00

Pos logit contributions
glass+0.025
orse+0.024
 generation+0.022
work+0.022
 planet+0.021
Neg logit contributions
wn-0.015
 aback-0.015
coat-0.013
 swe-0.013
 like-0.012
Token blonde
Token ID21541
Feature activation-6.77e+00

Pos logit contributions
wn+0.127
 aback+0.125
coat+0.106
 swe+0.105
 tast+0.101
Neg logit contributions
glass-0.260
orse-0.260
 generation-0.236
work-0.228
 planet-0.226
Token hair
Token ID4190
Feature activation-3.13e+00

Pos logit contributions
wn+0.213
 aback+0.209
coat+0.176
 swe+0.170
 like+0.158
Neg logit contributions
glass-0.532
orse-0.525
 generation-0.489
work-0.482
 planet-0.473
Token was
Token ID373
Feature activation+2.00e-02

Pos logit contributions
wn+0.101
 aback+0.100
coat+0.090
 swe+0.084
 like+0.083
Neg logit contributions
glass-0.238
orse-0.236
 generation-0.224
work-0.215
 planet-0.214
Token standing
Token ID5055
Feature activation+1.05e+00

Pos logit contributions
glass+0.001
orse+0.001
 generation+0.001
work+0.001
 planet+0.001
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.001
 swe-0.001
 like-0.001
Token there
Token ID612
Feature activation-1.57e+00

Pos logit contributions
glass+0.079
orse+0.079
 generation+0.073
work+0.071
 planet+0.070
Neg logit contributions
wn-0.036
 aback-0.036
coat-0.031
 swe-0.030
 like-0.029
Token.
Token ID13
Feature activation-2.41e-01

Pos logit contributions
wn+0.062
 aback+0.061
coat+0.053
 swe+0.052
 ripe+0.048
Neg logit contributions
glass-0.111
orse-0.109
 generation-0.101
work-0.100
 planet-0.097
Token wanted
Token ID2227
Feature activation-8.59e-01

Pos logit contributions
wn+0.065
 aback+0.064
coat+0.056
 swe+0.055
 like+0.053
Neg logit contributions
glass-0.134
orse-0.131
 generation-0.123
work-0.122
 planet-0.118
Token to
Token ID284
Feature activation-3.43e-01

Pos logit contributions
 aback+0.022
wn+0.022
 swe+0.018
 like+0.017
coat+0.017
Neg logit contributions
orse-0.072
glass-0.072
 generation-0.066
work-0.066
bike-0.065
Token go
Token ID467
Feature activation-3.28e-01

Pos logit contributions
wn+0.013
 aback+0.013
coat+0.012
 like+0.012
 swe+0.012
Neg logit contributions
glass-0.024
orse-0.024
work-0.022
 generation-0.022
bike-0.021
Token see
Token ID766
Feature activation+4.57e-01

Pos logit contributions
wn+0.013
 aback+0.013
coat+0.012
 like+0.012
 swe+0.011
Neg logit contributions
glass-0.023
orse-0.023
 generation-0.021
work-0.021
 planet-0.020
Token it
Token ID340
Feature activation+2.31e-01

Pos logit contributions
glass+0.030
orse+0.030
 generation+0.027
work+0.027
bike+0.026
Neg logit contributions
 aback-0.020
wn-0.020
 swe-0.018
 like-0.018
coat-0.017
Token again
Token ID757
Feature activation+5.57e+00

Pos logit contributions
glass+0.014
orse+0.014
 generation+0.012
work+0.012
 planet+0.012
Neg logit contributions
wn-0.012
 aback-0.011
coat-0.010
 swe-0.010
 ripe-0.010
Token.
Token ID13
Feature activation-1.71e-01

Pos logit contributions
glass+0.403
orse+0.395
work+0.364
 generation+0.360
bike+0.354
Neg logit contributions
 aback-0.208
wn-0.205
 like-0.185
 swe-0.181
coat-0.180
Token
Token ID220
Feature activation-1.90e+00

Pos logit contributions
wn+0.010
 aback+0.008
coat+0.008
 ripe+0.008
 swe+0.008
Neg logit contributions
orse-0.009
glass-0.009
 generation-0.008
work-0.008
 planet-0.007
Token
Token ID198
Feature activation-7.15e-02

Pos logit contributions
wn+0.104
 aback+0.100
 swe+0.092
coat+0.092
 ripe+0.089
Neg logit contributions
glass-0.104
orse-0.104
 generation-0.092
work-0.088
bike-0.086
Token
Token ID198
Feature activation-8.19e-02

Pos logit contributions
wn+0.004
 aback+0.004
coat+0.003
 swe+0.003
 ripe+0.003
Neg logit contributions
glass-0.004
orse-0.004
 generation-0.004
 planet-0.003
work-0.003
TokenL
Token ID43
Feature activation-1.52e+00

Pos logit contributions
wn+0.003
 aback+0.003
coat+0.003
 swe+0.003
 like+0.003
Neg logit contributions
glass-0.006
orse-0.006
 generation-0.005
work-0.005
 planet-0.005
Token should
Token ID815
Feature activation+1.13e+00

Pos logit contributions
orse+0.018
glass+0.017
work+0.017
 generation+0.016
 planet+0.015
Neg logit contributions
wn-0.014
 aback-0.013
coat-0.012
 swe-0.012
 ripe-0.011
Token have
Token ID423
Feature activation+1.93e+00

Pos logit contributions
glass+0.073
orse+0.071
 generation+0.065
work+0.064
 planet+0.062
Neg logit contributions
wn-0.052
 aback-0.051
coat-0.046
 swe-0.046
 like-0.045
Token given
Token ID1813
Feature activation-2.06e+00

Pos logit contributions
glass+0.122
orse+0.120
 generation+0.110
work+0.109
bike+0.105
Neg logit contributions
wn-0.092
 aback-0.089
coat-0.082
 swe-0.075
 like-0.072
Token it
Token ID340
Feature activation-2.23e-01

Pos logit contributions
wn+0.104
 aback+0.099
coat+0.095
 swe+0.087
 ripe+0.084
Neg logit contributions
glass-0.123
orse-0.120
 generation-0.112
work-0.110
 planet-0.106
Token back
Token ID736
Feature activation+2.38e+00

Pos logit contributions
wn+0.009
 aback+0.009
coat+0.008
 swe+0.008
 like+0.007
Neg logit contributions
glass-0.016
orse-0.015
 generation-0.014
work-0.014
 planet-0.014
Token sooner
Token ID14556
Feature activation+5.56e+00

Pos logit contributions
glass+0.173
orse+0.171
 generation+0.158
work+0.156
 planet+0.153
Neg logit contributions
wn-0.090
 aback-0.088
coat-0.077
 swe-0.075
 like-0.070
Token.
Token ID13
Feature activation-3.28e-01

Pos logit contributions
glass+0.435
orse+0.426
 generation+0.401
work+0.396
bike+0.388
Neg logit contributions
 aback-0.163
wn-0.162
 like-0.154
 swe-0.148
 sweet-0.142
Token Can
Token ID1680
Feature activation+2.26e-01

Pos logit contributions
 aback+0.017
wn+0.016
 like+0.015
 sweet+0.014
 swe+0.014
Neg logit contributions
glass-0.020
orse-0.019
work-0.017
 generation-0.017
bike-0.017
Token you
Token ID345
Feature activation+1.34e+00

Pos logit contributions
glass+0.016
 generation+0.014
orse+0.014
growth+0.014
work+0.013
Neg logit contributions
 sweet-0.009
 like-0.009
 swe-0.008
 aback-0.008
 past-0.008
Token forgive
Token ID20927
Feature activation-4.14e-01

Pos logit contributions
glass+0.080
orse+0.080
work+0.071
 generation+0.071
 planet+0.070
Neg logit contributions
wn-0.068
 aback-0.066
coat-0.061
 swe-0.058
 ripe-0.056
Token me
Token ID502
Feature activation-4.50e+00

Pos logit contributions
wn+0.015
 aback+0.014
coat+0.012
 swe+0.011
 ripe+0.011
Neg logit contributions
glass-0.031
orse-0.031
 generation-0.029
work-0.028
 planet-0.028
Token not
Token ID407
Feature activation-1.10e-02

Pos logit contributions
glass+0.027
orse+0.026
 generation+0.024
work+0.023
 planet+0.022
Neg logit contributions
wn-0.023
 aback-0.022
coat-0.020
 swe-0.020
 like-0.020
Token want
Token ID765
Feature activation-7.84e-01

Pos logit contributions
 aback+0.000
wn+0.000
 swe+0.000
coat+0.000
 ripe+0.000
Neg logit contributions
orse-0.001
glass-0.001
 generation-0.001
 planet-0.001
work-0.001
Token to
Token ID284
Feature activation-3.44e-01

Pos logit contributions
wn+0.024
 aback+0.023
 swe+0.019
coat+0.019
 like+0.018
Neg logit contributions
glass-0.063
orse-0.062
 generation-0.058
work-0.057
 planet-0.056
Token break
Token ID2270
Feature activation+6.45e-01

Pos logit contributions
wn+0.015
 aback+0.015
coat+0.013
 swe+0.013
 like+0.012
Neg logit contributions
glass-0.023
orse-0.023
 generation-0.021
work-0.020
 planet-0.020
Token it
Token ID340
Feature activation+1.71e-01

Pos logit contributions
orse+0.042
glass+0.041
 generation+0.037
work+0.036
bike+0.035
Neg logit contributions
 aback-0.030
wn-0.029
 swe-0.027
 like-0.027
 sweet-0.026
Token again
Token ID757
Feature activation+5.61e+00

Pos logit contributions
glass+0.011
orse+0.011
 generation+0.010
work+0.010
 planet+0.009
Neg logit contributions
wn-0.008
 aback-0.008
coat-0.007
 swe-0.007
 ripe-0.007
Token.
Token ID13
Feature activation-2.49e-01

Pos logit contributions
glass+0.447
orse+0.441
 generation+0.409
work+0.408
bike+0.398
Neg logit contributions
 aback-0.164
wn-0.163
 like-0.144
 swe-0.141
coat-0.139
Token They
Token ID1119
Feature activation-2.36e-01

Pos logit contributions
 aback+0.009
wn+0.008
 like+0.008
 fun+0.008
 swe+0.008
Neg logit contributions
glass-0.018
orse-0.018
bike-0.017
work-0.017
fer-0.017
Token also
Token ID635
Feature activation+1.80e+00

Pos logit contributions
wn+0.010
 aback+0.010
coat+0.009
 swe+0.009
 ripe+0.008
Neg logit contributions
glass-0.016
orse-0.015
 generation-0.014
work-0.014
 planet-0.014
Token gave
Token ID2921
Feature activation-2.83e+00

Pos logit contributions
glass+0.120
orse+0.119
 generation+0.109
work+0.107
 planet+0.105
Neg logit contributions
 aback-0.076
wn-0.076
coat-0.065
 swe-0.063
 ripe-0.058
Token some
Token ID617
Feature activation+7.82e-01

Pos logit contributions
 aback+0.128
wn+0.122
 swe+0.121
 sweet+0.115
 like+0.114
Neg logit contributions
orse-0.187
glass-0.187
fer-0.162
 generation-0.162
bike-0.158
Token,
Token ID11
Feature activation+1.12e-01

Pos logit contributions
glass+0.077
orse+0.076
 generation+0.071
work+0.070
 planet+0.068
Neg logit contributions
wn-0.036
 aback-0.035
coat-0.030
 swe-0.029
 like-0.027
Token Fin
Token ID4463
Feature activation+4.40e-01

Pos logit contributions
glass+0.006
orse+0.006
 generation+0.005
work+0.005
 planet+0.005
Neg logit contributions
wn-0.006
 aback-0.006
coat-0.006
 swe-0.006
 like-0.005
Token was
Token ID373
Feature activation+2.51e-01

Pos logit contributions
glass+0.030
orse+0.029
 generation+0.027
work+0.026
 planet+0.026
Neg logit contributions
wn-0.019
 aback-0.018
coat-0.016
 swe-0.016
 ripe-0.015
Token never
Token ID1239
Feature activation+1.53e+00

Pos logit contributions
glass+0.016
orse+0.016
 generation+0.015
work+0.015
 planet+0.014
Neg logit contributions
wn-0.011
 aback-0.011
coat-0.010
 swe-0.010
 ripe-0.009
Token lazy
Token ID16931
Feature activation-1.40e-01

Pos logit contributions
orse+0.097
glass+0.084
work+0.084
 Whale+0.081
fer+0.080
Neg logit contributions
wn-0.070
coat-0.065
 aback-0.061
arl-0.058
 swe-0.056
Token again
Token ID757
Feature activation+5.55e+00

Pos logit contributions
wn+0.009
 aback+0.008
coat+0.008
 swe+0.007
 ripe+0.007
Neg logit contributions
glass-0.007
orse-0.007
work-0.006
 generation-0.006
 planet-0.006
Token and
Token ID290
Feature activation+1.42e-01

Pos logit contributions
glass+0.436
orse+0.422
 generation+0.398
work+0.395
 planet+0.385
Neg logit contributions
 aback-0.175
wn-0.169
 like-0.154
 swe-0.144
coat-0.139
Token always
Token ID1464
Feature activation-3.72e+00

Pos logit contributions
glass+0.010
orse+0.009
 generation+0.009
work+0.008
 planet+0.008
Neg logit contributions
 aback-0.006
wn-0.006
 like-0.005
 swe-0.005
 fun-0.005
Token had
Token ID550
Feature activation+1.17e+00

Pos logit contributions
wn+0.146
 aback+0.145
coat+0.126
 swe+0.123
 like+0.118
Neg logit contributions
glass-0.265
orse-0.260
 generation-0.241
work-0.237
 planet-0.232
Token a
Token ID257
Feature activation-1.62e-01

Pos logit contributions
orse+0.073
glass+0.071
 generation+0.067
 planet+0.064
work+0.062
Neg logit contributions
wn-0.052
coat-0.051
 aback-0.049
arl-0.044
ucker-0.041
Token new
Token ID649
Feature activation-7.79e-02

Pos logit contributions
wn+0.005
 aback+0.005
coat+0.004
 swe+0.004
 like+0.004
Neg logit contributions
orse-0.012
glass-0.012
 generation-0.012
 planet-0.011
work-0.011
Token wanted
Token ID2227
Feature activation-5.47e-01

Pos logit contributions
wn+0.039
 aback+0.038
coat+0.034
 swe+0.032
 ripe+0.031
Neg logit contributions
glass-0.066
orse-0.066
 generation-0.060
work-0.060
 planet-0.058
Token to
Token ID284
Feature activation+3.29e-03

Pos logit contributions
 aback+0.014
wn+0.014
 swe+0.012
coat+0.012
 like+0.012
Neg logit contributions
glass-0.045
orse-0.045
 generation-0.042
work-0.041
 planet-0.041
Token '
Token ID705
Feature activation-2.62e+00

Pos logit contributions
glass+0.000
orse+0.000
 generation+0.000
work+0.000
 planet+0.000
Neg logit contributions
wn-0.000
 aback-0.000
coat-0.000
 swe-0.000
 like-0.000
Tokenspl
Token ID22018
Feature activation+2.20e+00

Pos logit contributions
wn+0.132
 aback+0.130
 swe+0.118
coat+0.117
 sweet+0.110
Neg logit contributions
glass-0.154
orse-0.151
 generation-0.138
work-0.136
bike-0.132
Tokenash
Token ID1077
Feature activation+1.20e+00

Pos logit contributions
orse+0.134
glass+0.130
work+0.124
bike+0.117
 generation+0.115
Neg logit contributions
 aback-0.105
wn-0.099
 swe-0.094
 sweet-0.092
 ripe-0.091
Token'
Token ID6
Feature activation+5.97e+00

Pos logit contributions
glass+0.078
orse+0.076
 generation+0.071
 planet+0.067
work+0.066
Neg logit contributions
 aback-0.054
wn-0.054
coat-0.049
 like-0.048
 swe-0.047
Token around
Token ID1088
Feature activation-2.23e+00

Pos logit contributions
glass+0.429
orse+0.423
 generation+0.391
work+0.387
 planet+0.377
Neg logit contributions
wn-0.227
 aback-0.223
coat-0.194
 swe-0.188
 like-0.177
Token,
Token ID11
Feature activation+1.36e-01

Pos logit contributions
wn+0.089
 aback+0.087
coat+0.077
 swe+0.074
 ripe+0.070
Neg logit contributions
glass-0.156
orse-0.153
 generation-0.142
work-0.140
 planet-0.137
Token but
Token ID475
Feature activation-2.16e+00

Pos logit contributions
glass+0.010
orse+0.010
 generation+0.009
work+0.009
 planet+0.009
Neg logit contributions
wn-0.005
 aback-0.005
coat-0.004
 swe-0.004
 ripe-0.003
Token she
Token ID673
Feature activation-9.84e-01

Pos logit contributions
wn+0.104
 aback+0.104
coat+0.095
 swe+0.094
 like+0.092
Neg logit contributions
glass-0.133
orse-0.128
 generation-0.118
work-0.116
 planet-0.112
Token wasn
Token ID2492
Feature activation-1.26e+00

Pos logit contributions
wn+0.045
 aback+0.044
coat+0.039
 swe+0.038
 ripe+0.036
Neg logit contributions
glass-0.064
orse-0.063
 generation-0.057
work-0.057
 planet-0.055
Token.
Token ID13
Feature activation-1.64e-01

Pos logit contributions
glass+0.157
orse+0.149
 generation+0.143
growth+0.138
ich+0.137
Neg logit contributions
 aback-0.047
wn-0.042
 like-0.042
 swe-0.037
 sweet-0.037
Token
Token ID220
Feature activation-1.85e+00

Pos logit contributions
 aback+0.006
wn+0.005
 like+0.004
 swe+0.004
 sweet+0.004
Neg logit contributions
glass-0.013
orse-0.012
work-0.012
 generation-0.011
 planet-0.011
Token
Token ID198
Feature activation+1.40e-03

Pos logit contributions
wn+0.089
 aback+0.088
coat+0.079
 swe+0.078
 ripe+0.074
Neg logit contributions
glass-0.114
orse-0.113
 generation-0.103
work-0.101
 planet-0.098
Token
Token ID198
Feature activation-1.07e-02

Pos logit contributions
glass+0.000
orse+0.000
work+0.000
 generation+0.000
 planet+0.000
Neg logit contributions
 aback-0.000
wn-0.000
coat-0.000
 swe-0.000
 like-0.000
TokenFinally
Token ID11158
Feature activation-3.04e-01

Pos logit contributions
wn+0.001
 aback+0.001
coat+0.001
 like+0.001
 swe+0.001
Neg logit contributions
glass-0.001
orse-0.001
work-0.001
 generation-0.001
 planet-0.000
Token they
Token ID484
Feature activation+1.98e+00

Pos logit contributions
 aback+0.005
 like+0.004
 past+0.004
 sweet+0.004
wn+0.004
Neg logit contributions
glass-0.028
orse-0.027
 generation-0.027
work-0.026
bike-0.026
Token reached
Token ID4251
Feature activation-2.25e+00

Pos logit contributions
glass+0.141
orse+0.137
 generation+0.130
work+0.127
 planet+0.123
Neg logit contributions
wn-0.076
 aback-0.076
 like-0.064
coat-0.064
 swe-0.063
Token a
Token ID257
Feature activation-2.63e-01

Pos logit contributions
 aback+0.086
wn+0.083
 like+0.075
 swe+0.072
 ripe+0.070
Neg logit contributions
glass-0.165
orse-0.159
 generation-0.148
work-0.146
bike-0.144
Token good
Token ID922
Feature activation-2.98e+00

Pos logit contributions
wn+0.010
 aback+0.010
coat+0.009
 swe+0.009
 like+0.008
Neg logit contributions
glass-0.019
orse-0.018
 generation-0.017
work-0.017
bike-0.016
Token spot
Token ID4136
Feature activation-1.80e+00

Pos logit contributions
wn+0.163
 aback+0.161
coat+0.147
 swe+0.144
 ripe+0.137
Neg logit contributions
glass-0.165
orse-0.165
 generation-0.147
work-0.144
 planet-0.142
Token to
Token ID284
Feature activation-4.65e-01

Pos logit contributions
 aback+0.056
wn+0.055
coat+0.048
 like+0.046
 swe+0.043
Neg logit contributions
glass-0.141
orse-0.140
 generation-0.131
 planet-0.126
work-0.126
Token and
Token ID290
Feature activation+1.21e-02

Pos logit contributions
glass+0.095
orse+0.094
bike+0.086
work+0.083
 whale+0.080
Neg logit contributions
wn-0.073
 aback-0.067
 ripe-0.058
 swe-0.056
inters-0.056
Token Sue
Token ID26089
Feature activation-2.61e+00

Pos logit contributions
glass+0.001
orse+0.001
 generation+0.001
work+0.001
 planet+0.001
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.000
 swe-0.000
ets-0.000
Token went
Token ID1816
Feature activation-1.03e+00

Pos logit contributions
wn+0.121
 aback+0.119
 swe+0.099
coat+0.098
 ripe+0.097
Neg logit contributions
glass-0.165
orse-0.164
bike-0.146
work-0.146
 generation-0.141
Token on
Token ID319
Feature activation+1.07e+00

Pos logit contributions
wn+0.042
 aback+0.042
 swe+0.034
coat+0.034
 ripe+0.034
Neg logit contributions
glass-0.070
orse-0.067
work-0.063
bike-0.062
 generation-0.061
Token the
Token ID262
Feature activation-5.42e-01

Pos logit contributions
orse+0.089
glass+0.087
work+0.079
 generation+0.079
fer+0.077
Neg logit contributions
 aback-0.032
 fun-0.032
 past-0.030
 like-0.029
 sweet-0.028
Token slide
Token ID10649
Feature activation+3.33e+00

Pos logit contributions
 aback+0.033
wn+0.033
coat+0.029
 swe+0.028
 like+0.028
Neg logit contributions
glass-0.027
orse-0.027
 generation-0.023
work-0.023
bike-0.023
Token and
Token ID290
Feature activation+4.56e-02

Pos logit contributions
glass+0.252
orse+0.250
 generation+0.232
work+0.227
 planet+0.223
Neg logit contributions
wn-0.114
 aback-0.113
coat-0.100
 swe-0.095
 like-0.092
Token had
Token ID550
Feature activation+1.15e+00

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.003
work+0.003
 planet+0.003
Neg logit contributions
wn-0.002
 aback-0.002
coat-0.002
 swe-0.001
 like-0.001
Token fun
Token ID1257
Feature activation-4.06e+00

Pos logit contributions
glass+0.082
orse+0.080
 generation+0.074
work+0.073
 planet+0.072
Neg logit contributions
wn-0.044
 aback-0.043
coat-0.037
 swe-0.035
 ripe-0.033
Token.
Token ID13
Feature activation-2.64e-01

Pos logit contributions
wn+0.131
 aback+0.126
 like+0.116
coat+0.113
 swe+0.109
Neg logit contributions
glass-0.315
orse-0.305
 generation-0.292
work-0.286
 planet-0.277
Token They
Token ID1119
Feature activation-1.95e-01

Pos logit contributions
wn+0.012
 aback+0.012
coat+0.011
 swe+0.010
 like+0.010
Neg logit contributions
glass-0.017
orse-0.017
 generation-0.015
work-0.015
 planet-0.015
Tokenily
Token ID813
Feature activation-8.72e-02

Pos logit contributions
 aback+0.089
wn+0.087
coat+0.080
 swe+0.080
 ripe+0.077
Neg logit contributions
glass-0.074
orse-0.072
 generation-0.065
work-0.064
 planet-0.062
Token was
Token ID373
Feature activation+2.54e-01

Pos logit contributions
wn+0.004
 aback+0.004
coat+0.004
 swe+0.004
 ripe+0.003
Neg logit contributions
glass-0.005
orse-0.005
 generation-0.005
work-0.005
bike-0.005
Token sad
Token ID6507
Feature activation-3.69e-01

Pos logit contributions
orse+0.018
glass+0.018
 generation+0.016
work+0.016
bike+0.015
Neg logit contributions
wn-0.010
 aback-0.010
coat-0.009
 swe-0.008
ets-0.008
Token she
Token ID673
Feature activation-8.27e-01

Pos logit contributions
wn+0.017
 aback+0.016
coat+0.015
 swe+0.014
 ripe+0.014
Neg logit contributions
orse-0.023
glass-0.023
work-0.021
 generation-0.021
 planet-0.020
Token couldn
Token ID3521
Feature activation+2.70e+00

Pos logit contributions
wn+0.049
 aback+0.049
 swe+0.044
 ripe+0.043
coat+0.042
Neg logit contributions
glass-0.041
orse-0.040
work-0.036
 generation-0.036
bike-0.035
Token't
Token ID470
Feature activation+1.82e+00

Pos logit contributions
glass+0.210
orse+0.204
 crew+0.194
 generation+0.185
fer+0.180
Neg logit contributions
 aback-0.091
coat-0.077
 time-0.076
 like-0.075
wn-0.075
Token play
Token ID711
Feature activation+1.35e+00

Pos logit contributions
glass+0.133
orse+0.127
work+0.121
 generation+0.119
ich+0.119
Neg logit contributions
 like-0.067
wn-0.066
 aback-0.062
coat-0.058
 sweet-0.053
Token outside
Token ID2354
Feature activation+1.89e-01

Pos logit contributions
glass+0.092
orse+0.089
 generation+0.083
work+0.081
 planet+0.081
Neg logit contributions
wn-0.058
 aback-0.057
coat-0.049
 swe-0.048
 ripe-0.046
Token,
Token ID11
Feature activation+3.94e-02

Pos logit contributions
glass+0.014
orse+0.014
 generation+0.012
work+0.012
bike+0.012
Neg logit contributions
 aback-0.007
wn-0.007
 like-0.007
 swe-0.006
coat-0.006
Token but
Token ID475
Feature activation-2.09e+00

Pos logit contributions
glass+0.003
orse+0.003
 generation+0.003
work+0.003
 planet+0.003
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.001
 swe-0.001
 like-0.001
Token she
Token ID673
Feature activation-8.31e-01

Pos logit contributions
 aback+0.065
wn+0.063
 like+0.062
 swe+0.061
 sweet+0.059
Neg logit contributions
glass-0.167
orse-0.162
work-0.151
 generation-0.148
fer-0.146
Token and
Token ID290
Feature activation+8.81e-02

Pos logit contributions
wn+0.228
 aback+0.221
 swe+0.195
coat+0.194
 ripe+0.175
Neg logit contributions
glass-0.409
orse-0.404
 generation-0.370
work-0.364
 planet-0.355
Token still
Token ID991
Feature activation-1.18e+00

Pos logit contributions
glass+0.005
orse+0.005
 generation+0.005
work+0.005
 planet+0.005
Neg logit contributions
wn-0.004
 aback-0.004
coat-0.004
 swe-0.004
 like-0.004
Token couldn
Token ID3521
Feature activation+2.71e+00

Pos logit contributions
wn+0.052
 aback+0.051
coat+0.045
 like+0.045
 swe+0.043
Neg logit contributions
glass-0.077
orse-0.076
 generation-0.070
work-0.069
 planet-0.068
Token't
Token ID470
Feature activation+1.85e+00

Pos logit contributions
glass+0.184
orse+0.176
work+0.164
 crew+0.162
 whale+0.162
Neg logit contributions
 aback-0.118
wn-0.109
coat-0.102
 like-0.101
 time-0.095
Token find
Token ID1064
Feature activation+1.85e+00

Pos logit contributions
glass+0.122
orse+0.119
work+0.115
bike+0.111
ich+0.110
Neg logit contributions
 like-0.078
 aback-0.077
wn-0.076
coat-0.066
 idea-0.063
Token one
Token ID530
Feature activation+1.75e+00

Pos logit contributions
glass+0.106
orse+0.104
 generation+0.096
work+0.093
 planet+0.092
Neg logit contributions
wn-0.097
 aback-0.094
coat-0.089
 swe-0.082
 ripe-0.079
Token.
Token ID13
Feature activation-7.22e-02

Pos logit contributions
orse+0.104
glass+0.102
 generation+0.096
 planet+0.096
work+0.094
Neg logit contributions
wn-0.087
 aback-0.083
coat-0.076
inters-0.068
 swe-0.068
Token He
Token ID679
Feature activation-2.29e-01

Pos logit contributions
wn+0.003
 aback+0.003
coat+0.003
 swe+0.003
 ripe+0.003
Neg logit contributions
glass-0.004
orse-0.004
 generation-0.004
bike-0.004
work-0.004
Token was
Token ID373
Feature activation+2.24e-01

Pos logit contributions
wn+0.009
 aback+0.009
coat+0.008
 swe+0.008
 like+0.008
Neg logit contributions
glass-0.016
orse-0.016
 generation-0.014
work-0.014
 planet-0.014
Token getting
Token ID1972
Feature activation-6.76e-01

Pos logit contributions
glass+0.015
orse+0.015
 generation+0.014
work+0.013
 planet+0.013
Neg logit contributions
wn-0.010
 aback-0.009
coat-0.009
 swe-0.008
ets-0.008
Token very
Token ID845
Feature activation-3.55e-01

Pos logit contributions
wn+0.031
coat+0.030
 aback+0.028
ets+0.027
 swe+0.025
Neg logit contributions
glass-0.041
orse-0.040
 generation-0.037
work-0.036
 crew-0.035
Token had
Token ID550
Feature activation+9.34e-01

Pos logit contributions
wn+0.046
 aback+0.046
coat+0.040
 swe+0.039
 like+0.038
Neg logit contributions
glass-0.075
orse-0.074
 generation-0.068
work-0.067
 planet-0.065
Token to
Token ID284
Feature activation-4.08e-01

Pos logit contributions
glass+0.063
orse+0.061
 generation+0.058
work+0.055
 planet+0.054
Neg logit contributions
wn-0.038
 aback-0.037
coat-0.032
 swe-0.029
ets-0.029
Token be
Token ID307
Feature activation+5.42e-01

Pos logit contributions
wn+0.020
 aback+0.020
coat+0.018
 swe+0.017
 ripe+0.017
Neg logit contributions
glass-0.024
orse-0.024
 generation-0.022
 planet-0.021
work-0.021
Token careful
Token ID8161
Feature activation-2.51e+00

Pos logit contributions
glass+0.035
orse+0.032
 generation+0.030
 crew+0.029
ich+0.028
Neg logit contributions
 aback-0.022
wn-0.019
coat-0.019
arl-0.018
inters-0.018
Token.
Token ID13
Feature activation-2.72e-01

Pos logit contributions
wn+0.094
 aback+0.092
coat+0.082
 swe+0.079
 like+0.078
Neg logit contributions
glass-0.182
orse-0.180
 generation-0.166
work-0.163
 planet-0.161
Token The
Token ID383
Feature activation+2.29e+00

Pos logit contributions
wn+0.017
 aback+0.015
coat+0.015
 ripe+0.014
ets+0.014
Neg logit contributions
orse-0.013
glass-0.012
 generation-0.012
 Our-0.011
 planet-0.010
Token ball
Token ID2613
Feature activation+1.05e-01

Pos logit contributions
glass+0.159
orse+0.158
 generation+0.146
work+0.142
bike+0.136
Neg logit contributions
wn-0.091
 aback-0.088
 swe-0.085
coat-0.085
 sweet-0.084
Token was
Token ID373
Feature activation+1.78e-01

Pos logit contributions
glass+0.007
orse+0.007
work+0.006
 generation+0.006
 planet+0.006
Neg logit contributions
wn-0.005
 aback-0.005
coat-0.004
 swe-0.004
 ripe-0.004
Token very
Token ID845
Feature activation-3.51e-01

Pos logit contributions
orse+0.010
glass+0.009
work+0.008
ich+0.008
 nobody+0.008
Neg logit contributions
 aback-0.009
wn-0.008
ets-0.008
coat-0.007
ere-0.007
Token big
Token ID1263
Feature activation-1.68e+00

Pos logit contributions
 aback+0.016
wn+0.014
coat+0.012
ets+0.011
 swe+0.009
Neg logit contributions
orse-0.025
glass-0.022
 planet-0.021
 whale-0.021
 ship-0.020
Token and
Token ID290
Feature activation+9.45e-02

Pos logit contributions
wn+0.085
 aback+0.076
coat+0.070
 swe+0.066
 ripe+0.065
Neg logit contributions
glass-0.102
orse-0.096
 planet-0.091
work-0.091
 generation-0.089
Token favorite
Token ID4004
Feature activation+3.71e+00

Pos logit contributions
 aback+0.062
wn+0.060
coat+0.049
 swe+0.046
 ripe+0.045
Neg logit contributions
glass-0.114
orse-0.111
 generation-0.102
 planet-0.101
work-0.101
Token hat
Token ID6877
Feature activation-4.05e-01

Pos logit contributions
glass+0.235
orse+0.229
 generation+0.209
 planet+0.206
work+0.205
Neg logit contributions
 aback-0.177
wn-0.175
coat-0.151
 swe-0.143
 like-0.143
Token everywhere
Token ID8347
Feature activation-3.50e+00

Pos logit contributions
 aback+0.020
wn+0.019
 swe+0.016
 ripe+0.016
coat+0.016
Neg logit contributions
glass-0.026
orse-0.024
work-0.023
 generation-0.022
 planet-0.022
Token she
Token ID673
Feature activation-8.15e-01

Pos logit contributions
wn+0.117
 like+0.109
 sweet+0.107
coat+0.107
 aback+0.106
Neg logit contributions
glass-0.261
orse-0.258
 generation-0.242
work-0.241
 Whale-0.238
Token went
Token ID1816
Feature activation-1.01e+00

Pos logit contributions
 aback+0.036
wn+0.035
coat+0.031
 swe+0.030
 ripe+0.029
Neg logit contributions
glass-0.055
orse-0.053
 generation-0.049
work-0.048
 planet-0.047
Token.
Token ID13
Feature activation-9.94e-02

Pos logit contributions
 aback+0.046
wn+0.045
 ripe+0.039
coat+0.038
 swe+0.038
Neg logit contributions
glass-0.066
orse-0.062
 generation-0.060
work-0.059
 planet-0.058
Token It
Token ID632
Feature activation+5.13e-01

Pos logit contributions
wn+0.005
 aback+0.005
coat+0.004
 ripe+0.004
 tast+0.004
Neg logit contributions
orse-0.005
glass-0.005
 generation-0.005
 planet-0.005
work-0.005
Token was
Token ID373
Feature activation+2.72e-01

Pos logit contributions
glass+0.031
orse+0.031
 generation+0.028
work+0.028
 planet+0.027
Neg logit contributions
wn-0.025
 aback-0.025
coat-0.022
 swe-0.022
 ripe-0.021
Token a
Token ID257
Feature activation-1.27e-01

Pos logit contributions
orse+0.011
glass+0.011
 planet+0.011
 nobody+0.010
 generation+0.010
Neg logit contributions
 aback-0.018
wn-0.016
ets-0.015
ives-0.015
eal-0.014
Token big
Token ID1263
Feature activation-1.61e+00

Pos logit contributions
 aback+0.004
wn+0.004
coat+0.003
 swe+0.003
ets+0.003
Neg logit contributions
glass-0.010
orse-0.010
 generation-0.009
 planet-0.009
work-0.009
Token,
Token ID11
Feature activation+8.40e-02

Pos logit contributions
wn+0.064
 aback+0.063
coat+0.055
 swe+0.054
 like+0.051
Neg logit contributions
glass-0.114
orse-0.112
 generation-0.104
work-0.102
 planet-0.099
Token,
Token ID11
Feature activation+1.08e-01

Pos logit contributions
wn+0.121
 aback+0.106
coat+0.095
 swe+0.093
 prett+0.090
Neg logit contributions
orse-0.116
glass-0.112
 planet-0.105
work-0.103
 generation-0.100
Token me
Token ID502
Feature activation-4.22e+00

Pos logit contributions
orse+0.006
glass+0.006
 whale+0.005
 planet+0.005
 Whale+0.005
Neg logit contributions
wn-0.006
 aback-0.006
coat-0.005
inters-0.005
 swe-0.004
Token too
Token ID1165
Feature activation-2.09e+00

Pos logit contributions
wn+0.171
 aback+0.164
 swe+0.162
 gre+0.146
 ju+0.142
Neg logit contributions
orse-0.280
glass-0.269
work-0.268
 generation-0.239
 Whale-0.234
Token!"
Token ID2474
Feature activation+2.33e+00

Pos logit contributions
wn+0.097
 aback+0.092
coat+0.080
 swe+0.077
 prett+0.071
Neg logit contributions
orse-0.139
glass-0.136
 generation-0.126
 planet-0.125
work-0.123
Token Ben
Token ID3932
Feature activation-1.05e+00

Pos logit contributions
orse+0.131
glass+0.129
 generation+0.115
work+0.111
 planet+0.111
Neg logit contributions
wn-0.129
 aback-0.124
coat-0.115
 swe-0.110
 ripe-0.107
Token says
Token ID1139
Feature activation+1.53e-01

Pos logit contributions
wn+0.069
coat+0.060
 ripe+0.056
inters+0.056
arl+0.056
Neg logit contributions
orse-0.047
glass-0.046
work-0.042
 whale-0.041
 Whale-0.041
Token.
Token ID13
Feature activation-1.42e-01

Pos logit contributions
glass+0.011
orse+0.011
 generation+0.010
work+0.010
 planet+0.010
Neg logit contributions
wn-0.006
 aback-0.006
coat-0.005
 swe-0.005
 ripe-0.005
Token "
Token ID366
Feature activation-1.42e+00

Pos logit contributions
wn+0.008
 aback+0.008
coat+0.007
 swe+0.007
 ripe+0.007
Neg logit contributions
orse-0.008
glass-0.008
 Whale-0.006
work-0.006
 generation-0.006
TokenCan
Token ID6090
Feature activation+4.61e-01

Pos logit contributions
wn+0.058
 aback+0.056
 like+0.056
coat+0.052
 ripe+0.049
Neg logit contributions
orse-0.095
glass-0.092
work-0.089
 planet-0.085
 generation-0.085
Token I
Token ID314
Feature activation+4.63e-01

Pos logit contributions
glass+0.030
 generation+0.028
orse+0.027
 harbour+0.026
growth+0.026
Neg logit contributions
 sweet-0.021
 swe-0.021
 might-0.020
 like-0.020
 past-0.020
Token hold
Token ID1745
Feature activation-1.76e+00

Pos logit contributions
orse+0.027
glass+0.026
 planet+0.023
 generation+0.023
work+0.022
Neg logit contributions
wn-0.025
 aback-0.024
 swe-0.021
coat-0.021
 ripe-0.021
Token every
Token ID790
Feature activation-1.93e-01

Pos logit contributions
orse+0.015
glass+0.015
 generation+0.014
work+0.013
 planet+0.013
Neg logit contributions
wn-0.006
 aback-0.006
 like-0.006
coat-0.006
 swe-0.005
Token day
Token ID1110
Feature activation+1.12e+00

Pos logit contributions
wn+0.009
 aback+0.008
coat+0.008
 swe+0.008
 like+0.007
Neg logit contributions
glass-0.013
orse-0.012
work-0.011
 generation-0.011
bike-0.011
Token.
Token ID13
Feature activation+2.77e-03

Pos logit contributions
glass+0.087
orse+0.086
 generation+0.080
work+0.079
bike+0.078
Neg logit contributions
wn-0.035
 aback-0.034
 like-0.030
coat-0.030
 swe-0.028
Token<|endoftext|>
Token ID50256
Feature activation+2.55e-01

Pos logit contributions
glass+0.000
orse+0.000
work+0.000
 generation+0.000
fer+0.000
Neg logit contributions
 aback-0.000
wn-0.000
coat-0.000
 like-0.000
 swe-0.000
TokenOnce
Token ID7454
Feature activation-1.03e+00

Pos logit contributions
glass+0.016
orse+0.016
 generation+0.014
work+0.014
 planet+0.014
Neg logit contributions
wn-0.012
 aback-0.012
coat-0.011
 swe-0.010
 like-0.010
Token there
Token ID612
Feature activation-1.46e+00

Pos logit contributions
wn+0.033
 aback+0.032
coat+0.027
 swe+0.025
 ripe+0.023
Neg logit contributions
orse-0.081
glass-0.080
 generation-0.074
work-0.074
 planet-0.073
Token was
Token ID373
Feature activation+1.26e-01

Pos logit contributions
wn+0.077
 aback+0.075
coat+0.068
 swe+0.067
 ripe+0.064
Neg logit contributions
glass-0.084
orse-0.084
 generation-0.075
work-0.074
 planet-0.073
Token a
Token ID257
Feature activation-3.12e-01

Pos logit contributions
glass+0.009
orse+0.008
work+0.008
 generation+0.008
bike+0.008
Neg logit contributions
wn-0.005
 aback-0.005
coat-0.005
 like-0.004
 swe-0.004
Token big
Token ID1263
Feature activation-1.59e+00

Pos logit contributions
wn+0.008
 aback+0.008
coat+0.007
 swe+0.006
 like+0.006
Neg logit contributions
glass-0.026
orse-0.026
 generation-0.024
work-0.024
 planet-0.023
Token dog
Token ID3290
Feature activation+1.52e+00

Pos logit contributions
wn+0.070
 aback+0.069
coat+0.067
 swe+0.066
 sweet+0.065
Neg logit contributions
glass-0.103
orse-0.098
 generation-0.092
work-0.087
fer-0.084
Token.
Token ID13
Feature activation-4.41e-02

Pos logit contributions
glass+0.102
orse+0.095
 generation+0.093
work+0.090
fer+0.088
Neg logit contributions
 like-0.063
coat-0.058
 past-0.057
wn-0.057
 aback-0.056
Token saw
Token ID2497
Feature activation+1.20e+00

Pos logit contributions
orse+0.122
glass+0.122
work+0.107
 generation+0.104
 whale+0.103
Neg logit contributions
wn-0.100
 aback-0.092
coat-0.086
 swe-0.082
inters-0.081
Token this
Token ID428
Feature activation+2.03e+00

Pos logit contributions
glass+0.059
orse+0.059
 whale+0.055
work+0.055
 Whale+0.053
Neg logit contributions
wn-0.074
coat-0.067
 aback-0.061
inters-0.060
 tast-0.058
Token and
Token ID290
Feature activation+2.11e-02

Pos logit contributions
glass+0.145
orse+0.142
 generation+0.132
work+0.131
 planet+0.128
Neg logit contributions
wn-0.080
 aback-0.077
coat-0.067
 swe-0.065
 ripe-0.062
Token came
Token ID1625
Feature activation-3.01e+00

Pos logit contributions
orse+0.001
glass+0.001
work+0.001
 generation+0.001
bike+0.001
Neg logit contributions
wn-0.001
 aback-0.001
coat-0.001
 swe-0.001
ets-0.001
Token over
Token ID625
Feature activation-6.19e-01

Pos logit contributions
wn+0.135
 aback+0.123
coat+0.118
 swe+0.111
 tast+0.111
Neg logit contributions
glass-0.196
orse-0.190
 generation-0.177
work-0.175
 planet-0.167
Token to
Token ID284
Feature activation-4.00e-01

Pos logit contributions
wn+0.024
 aback+0.022
coat+0.020
 swe+0.019
 prett+0.018
Neg logit contributions
glass-0.045
orse-0.044
 generation-0.041
work-0.041
 planet-0.039
Token give
Token ID1577
Feature activation-1.18e+00

Pos logit contributions
wn+0.017
 aback+0.016
coat+0.014
 swe+0.014
 ripe+0.013
Neg logit contributions
glass-0.027
orse-0.027
 generation-0.025
work-0.025
 planet-0.024
Token her
Token ID607
Feature activation-1.63e+00

Pos logit contributions
wn+0.057
coat+0.056
arl+0.052
 aback+0.049
inters+0.049
Neg logit contributions
orse-0.063
glass-0.061
 generation-0.058
work-0.058
 whale-0.054
Token a
Token ID257
Feature activation-1.96e-01

Pos logit contributions
wn+0.075
 aback+0.074
coat+0.066
 swe+0.064
 like+0.062
Neg logit contributions
glass-0.105
orse-0.103
 generation-0.095
work-0.093
 planet-0.091
Token hug
Token ID16225
Feature activation-1.28e+00

Pos logit contributions
wn+0.008
 aback+0.007
ets+0.007
coat+0.007
inters+0.006
Neg logit contributions
glass-0.013
orse-0.012
work-0.012
 planet-0.012
 movie-0.012
Token.
Token ID13
Feature activation-9.68e-02

Pos logit contributions
wn+0.046
 aback+0.045
coat+0.039
 swe+0.038
 like+0.036
Neg logit contributions
glass-0.094
orse-0.094
 generation-0.087
work-0.085
 planet-0.083
Token you
Token ID345
Feature activation+1.43e+00

Pos logit contributions
glass+0.088
orse+0.084
 generation+0.082
bike+0.081
work+0.079
Neg logit contributions
 aback-0.051
 sweet-0.050
wn-0.049
 like-0.048
 past-0.047
Token,
Token ID11
Feature activation+6.39e-02

Pos logit contributions
glass+0.120
orse+0.117
 generation+0.113
bike+0.110
work+0.108
Neg logit contributions
 aback-0.037
wn-0.035
 sweet-0.033
 like-0.033
 swe-0.032
Token conductor
Token ID39206
Feature activation+4.64e-01

Pos logit contributions
glass+0.004
orse+0.004
 generation+0.004
bike+0.003
work+0.003
Neg logit contributions
 sweet-0.003
 swe-0.003
 like-0.003
 nice-0.003
 fun-0.003
Token."
Token ID526
Feature activation+6.00e-01

Pos logit contributions
glass+0.035
orse+0.034
 generation+0.033
 planet+0.031
work+0.031
Neg logit contributions
 aback-0.016
wn-0.016
coat-0.014
 like-0.013
 swe-0.013
Token Anna
Token ID11735
Feature activation-1.78e+00

Pos logit contributions
glass+0.044
work+0.041
 generation+0.041
orse+0.040
ich+0.039
Neg logit contributions
 aback-0.021
 like-0.020
 sweet-0.019
 past-0.019
 fun-0.019
Token said
Token ID531
Feature activation-1.08e+00

Pos logit contributions
 aback+0.043
wn+0.038
coat+0.036
 swe+0.035
 like+0.034
Neg logit contributions
glass-0.149
 generation-0.144
orse-0.143
work-0.138
 planet-0.135
Token.
Token ID13
Feature activation-3.22e-01

Pos logit contributions
 aback+0.035
wn+0.034
coat+0.030
 swe+0.029
 like+0.028
Neg logit contributions
glass-0.084
orse-0.082
 generation-0.078
work-0.076
 planet-0.075
Token
Token ID198
Feature activation-1.08e-01

Pos logit contributions
wn+0.016
 aback+0.015
coat+0.014
 swe+0.013
 ripe+0.013
Neg logit contributions
glass-0.020
orse-0.020
 generation-0.018
work-0.018
 planet-0.017
Token
Token ID198
Feature activation-1.30e-01

Pos logit contributions
 aback+0.004
wn+0.004
coat+0.004
 swe+0.004
 sweet+0.004
Neg logit contributions
glass-0.008
orse-0.007
work-0.007
 generation-0.007
bike-0.007
Token"
Token ID1
Feature activation-6.26e-01

Pos logit contributions
wn+0.004
 aback+0.004
coat+0.003
 swe+0.003
 sweet+0.003
Neg logit contributions
glass-0.010
orse-0.010
work-0.009
 generation-0.009
bike-0.009
TokenYou
Token ID1639
Feature activation-8.31e-01

Pos logit contributions
wn+0.028
 aback+0.027
coat+0.026
 like+0.025
 sweet+0.025
Neg logit contributions
glass-0.040
orse-0.039
 generation-0.036
work-0.035
 planet-0.034
Token the
Token ID262
Feature activation-6.65e-01

Pos logit contributions
glass+0.194
orse+0.186
work+0.173
fer+0.172
bike+0.171
Neg logit contributions
 aback-0.066
 fun-0.061
 like-0.060
 sweet-0.053
 nice-0.052
Token spot
Token ID4136
Feature activation-1.83e+00

Pos logit contributions
 aback+0.037
wn+0.037
 swe+0.035
 like+0.035
 sweet+0.034
Neg logit contributions
glass-0.036
orse-0.035
work-0.030
 generation-0.030
fer-0.028
Token for
Token ID329
Feature activation+7.55e-01

Pos logit contributions
 aback+0.060
wn+0.059
coat+0.052
 like+0.051
 swe+0.047
Neg logit contributions
glass-0.140
orse-0.138
 generation-0.128
work-0.126
bike-0.124
Token a
Token ID257
Feature activation-2.65e-01

Pos logit contributions
orse+0.056
work+0.056
glass+0.056
bike+0.055
ich+0.055
Neg logit contributions
 fun-0.025
wn-0.025
 like-0.024
 aback-0.023
 time-0.022
Token long
Token ID890
Feature activation+6.28e-02

Pos logit contributions
wn+0.007
 aback+0.007
coat+0.006
 swe+0.006
 like+0.006
Neg logit contributions
glass-0.022
orse-0.021
 generation-0.020
work-0.020
bike-0.019
Token time
Token ID640
Feature activation-5.89e+00

Pos logit contributions
orse+0.005
glass+0.005
 generation+0.005
 planet+0.005
work+0.005
Neg logit contributions
 aback-0.002
wn-0.002
 swe-0.001
 ripe-0.001
 tast-0.001
Token and
Token ID290
Feature activation+1.90e-02

Pos logit contributions
wn+0.183
 aback+0.182
 like+0.155
coat+0.154
 swe+0.144
Neg logit contributions
glass-0.458
orse-0.447
 generation-0.421
work-0.417
bike-0.409
Token never
Token ID1239
Feature activation+1.43e+00

Pos logit contributions
glass+0.001
orse+0.001
work+0.001
ich+0.001
 generation+0.001
Neg logit contributions
 aback-0.001
wn-0.001
 like-0.001
 fun-0.001
 sweet-0.001
Token forgot
Token ID16453
Feature activation-7.27e-01

Pos logit contributions
orse+0.097
glass+0.094
work+0.088
 generation+0.087
 planet+0.085
Neg logit contributions
wn-0.061
 aback-0.058
coat-0.053
 swe-0.050
 ripe-0.048
Token this
Token ID428
Feature activation+2.08e+00

Pos logit contributions
wn+0.033
 aback+0.032
coat+0.029
 swe+0.027
 ripe+0.026
Neg logit contributions
glass-0.047
orse-0.047
 generation-0.043
work-0.042
 planet-0.041
Token special
Token ID2041
Feature activation-4.66e-01

Pos logit contributions
glass+0.141
orse+0.138
 generation+0.127
work+0.125
 planet+0.122
Neg logit contributions
wn-0.089
 aback-0.088
coat-0.077
 swe-0.076
 like-0.073
Token lift
Token ID10303
Feature activation-2.16e-01

Pos logit contributions
wn+0.021
 aback+0.020
coat+0.018
 swe+0.017
 ripe+0.016
Neg logit contributions
glass-0.033
orse-0.032
 generation-0.030
work-0.029
 planet-0.029
Token this
Token ID428
Feature activation+2.08e+00

Pos logit contributions
wn+0.010
 aback+0.010
coat+0.009
 swe+0.009
 like+0.008
Neg logit contributions
glass-0.014
orse-0.013
 generation-0.012
work-0.012
 planet-0.012
Token box
Token ID3091
Feature activation-2.75e-01

Pos logit contributions
glass+0.120
 planet+0.114
 generation+0.109
orse+0.108
work+0.106
Neg logit contributions
 aback-0.107
wn-0.107
coat-0.093
ucker-0.088
inters-0.084
Token!"
Token ID2474
Feature activation+2.69e+00

Pos logit contributions
wn+0.011
 aback+0.011
coat+0.010
 swe+0.009
 ripe+0.009
Neg logit contributions
glass-0.019
orse-0.018
work-0.017
 generation-0.017
 planet-0.017
Token He
Token ID679
Feature activation-1.97e-01

Pos logit contributions
glass+0.162
orse+0.159
 generation+0.144
work+0.142
 planet+0.138
Neg logit contributions
wn-0.136
 aback-0.133
coat-0.121
 swe-0.119
 ripe-0.114
Token bent
Token ID17157
Feature activation-3.26e+00

Pos logit contributions
wn+0.007
 aback+0.007
coat+0.006
 swe+0.006
 ripe+0.006
Neg logit contributions
glass-0.015
orse-0.014
 generation-0.013
work-0.013
 planet-0.013
Token down
Token ID866
Feature activation-4.65e-02

Pos logit contributions
wn+0.123
 aback+0.114
coat+0.101
 swe+0.099
 tast+0.094
Neg logit contributions
glass-0.239
orse-0.229
work-0.214
bike-0.213
 generation-0.211
Token,
Token ID11
Feature activation-8.01e-03

Pos logit contributions
wn+0.002
 aback+0.002
coat+0.001
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.004
orse-0.003
 generation-0.003
work-0.003
 planet-0.003
Token put
Token ID1234
Feature activation-2.39e+00

Pos logit contributions
wn+0.000
 aback+0.000
coat+0.000
 swe+0.000
ets+0.000
Neg logit contributions
orse-0.001
glass-0.000
bike-0.000
 generation-0.000
work-0.000
Token his
Token ID465
Feature activation+5.08e-01

Pos logit contributions
wn+0.134
coat+0.131
 aback+0.130
 swe+0.120
ucker+0.117
Neg logit contributions
glass-0.125
orse-0.114
 generation-0.106
work-0.105
 planet-0.101
Token hands
Token ID2832
Feature activation-1.37e+00

Pos logit contributions
glass+0.031
orse+0.030
 generation+0.027
work+0.027
 planet+0.027
Neg logit contributions
 aback-0.026
wn-0.025
coat-0.023
 swe-0.022
 like-0.021
Token Max
Token ID5436
Feature activation-6.00e-01

Pos logit contributions
 aback+0.004
wn+0.004
coat+0.003
 like+0.003
 swe+0.003
Neg logit contributions
glass-0.008
orse-0.007
work-0.007
fer-0.007
 generation-0.007
Token bark
Token ID21405
Feature activation-4.55e-01

Pos logit contributions
wn+0.031
 aback+0.029
coat+0.026
 swe+0.026
 ripe+0.025
Neg logit contributions
orse-0.036
glass-0.035
 generation-0.031
bike-0.031
work-0.031
Tokened
Token ID276
Feature activation+3.50e+00

Pos logit contributions
wn+0.026
 aback+0.024
coat+0.023
 swe+0.023
 ripe+0.022
Neg logit contributions
glass-0.024
orse-0.023
work-0.021
 generation-0.021
bike-0.020
Token happily
Token ID18177
Feature activation-5.49e-01

Pos logit contributions
glass+0.220
orse+0.213
 generation+0.206
 planet+0.202
work+0.201
Neg logit contributions
wn-0.161
 aback-0.139
coat-0.139
inters-0.125
ucker-0.123
Token and
Token ID290
Feature activation-1.58e-02

Pos logit contributions
wn+0.017
 aback+0.017
coat+0.014
 swe+0.014
 like+0.013
Neg logit contributions
glass-0.043
orse-0.043
 generation-0.040
work-0.039
 planet-0.038
Token w
Token ID266
Feature activation-3.22e+00

Pos logit contributions
 aback+0.001
wn+0.001
coat+0.001
 swe+0.001
 like+0.001
Neg logit contributions
glass-0.001
orse-0.001
 generation-0.001
work-0.001
 planet-0.001
Tokenagged
Token ID14655
Feature activation-3.85e-01

Pos logit contributions
 aback+0.156
 swe+0.148
 ju+0.148
 ripe+0.147
wn+0.142
Neg logit contributions
orse-0.193
glass-0.191
 generation-0.157
work-0.156
 crew-0.150
Token his
Token ID465
Feature activation+3.05e-01

Pos logit contributions
wn+0.021
 aback+0.020
coat+0.019
 swe+0.018
 ripe+0.017
Neg logit contributions
glass-0.022
orse-0.021
 generation-0.019
work-0.019
 planet-0.018
Token tail
Token ID7894
Feature activation-5.62e-01

Pos logit contributions
glass+0.018
orse+0.018
 generation+0.016
work+0.016
 planet+0.016
Neg logit contributions
wn-0.015
 aback-0.015
coat-0.014
 swe-0.013
 ripe-0.013
Token as
Token ID355
Feature activation-4.54e+00

Pos logit contributions
wn+0.023
 aback+0.022
coat+0.019
 swe+0.019
 ripe+0.018
Neg logit contributions
glass-0.039
orse-0.039
 generation-0.036
work-0.035
 planet-0.035
Token they
Token ID484
Feature activation+1.82e+00

Pos logit contributions
wn+0.272
 aback+0.271
 swe+0.252
 like+0.250
coat+0.250
Neg logit contributions
glass-0.224
orse-0.215
 generation-0.192
work-0.190
bike-0.178
Token not
Token ID407
Feature activation-3.10e-01

Pos logit contributions
glass+0.018
orse+0.018
 generation+0.016
work+0.016
 planet+0.015
Neg logit contributions
wn-0.018
 aback-0.018
coat-0.017
 swe-0.016
 ripe-0.016
Token think
Token ID892
Feature activation+6.99e-01

Pos logit contributions
 aback+0.014
ucker+0.013
coat+0.013
 swe+0.013
 ju+0.012
Neg logit contributions
orse-0.019
glass-0.018
 generation-0.016
 ship-0.016
 planet-0.016
Token Ben
Token ID3932
Feature activation-1.18e+00

Pos logit contributions
glass+0.043
orse+0.042
 generation+0.038
work+0.038
 planet+0.036
Neg logit contributions
wn-0.034
 aback-0.034
coat-0.031
 swe-0.030
 like-0.029
Token was
Token ID373
Feature activation+1.31e-01

Pos logit contributions
wn+0.062
 aback+0.059
coat+0.054
 swe+0.053
 ripe+0.052
Neg logit contributions
glass-0.068
orse-0.066
work-0.060
 generation-0.060
bike-0.059
Token being
Token ID852
Feature activation+1.11e-01

Pos logit contributions
orse+0.007
glass+0.007
work+0.007
 whale+0.007
 Whale+0.007
Neg logit contributions
 aback-0.006
wn-0.006
coat-0.006
 swe-0.005
inters-0.005
Token nice
Token ID3621
Feature activation-3.63e+00

Pos logit contributions
glass+0.006
orse+0.005
work+0.005
 whale+0.005
fer+0.005
Neg logit contributions
coat-0.005
 aback-0.005
wn-0.005
 swe-0.004
ucker-0.004
Token.
Token ID13
Feature activation-5.23e-01

Pos logit contributions
wn+0.159
 aback+0.151
coat+0.134
 swe+0.129
 ripe+0.127
Neg logit contributions
glass-0.244
orse-0.233
work-0.220
 generation-0.215
 planet-0.211
Token She
Token ID1375
Feature activation-2.12e+00

Pos logit contributions
wn+0.015
 aback+0.015
coat+0.013
 swe+0.012
 like+0.011
Neg logit contributions
glass-0.042
orse-0.042
 generation-0.039
work-0.039
 planet-0.038
Token had
Token ID550
Feature activation+9.60e-01

Pos logit contributions
wn+0.102
 aback+0.100
coat+0.087
 swe+0.082
 ripe+0.079
Neg logit contributions
glass-0.133
orse-0.130
 generation-0.115
bike-0.115
 whale-0.113
Token an
Token ID281
Feature activation+4.70e-01

Pos logit contributions
glass+0.055
orse+0.053
 generation+0.049
work+0.047
bike+0.046
Neg logit contributions
wn-0.049
 aback-0.048
coat-0.044
 swe-0.042
ets-0.039
Token idea
Token ID2126
Feature activation-5.54e+00

Pos logit contributions
glass+0.041
orse+0.040
 generation+0.038
 planet+0.037
work+0.036
Neg logit contributions
 aback-0.010
wn-0.010
coat-0.008
 swe-0.007
 gre-0.006
Token wish
Token ID4601
Feature activation+3.02e+00

Pos logit contributions
wn+0.015
 aback+0.015
coat+0.013
 swe+0.013
 ripe+0.012
Neg logit contributions
glass-0.021
orse-0.020
 planet-0.019
 generation-0.019
work-0.018
Token Max
Token ID5436
Feature activation-3.31e-01

Pos logit contributions
glass+0.158
orse+0.155
 generation+0.154
bike+0.139
fer+0.137
Neg logit contributions
 sweet-0.163
 like-0.161
 swe-0.159
 fun-0.158
wn-0.156
Token would
Token ID561
Feature activation-8.56e-01

Pos logit contributions
wn+0.013
 aback+0.013
coat+0.012
 swe+0.011
 ripe+0.011
Neg logit contributions
glass-0.023
orse-0.023
 generation-0.021
work-0.021
 planet-0.020
Token be
Token ID307
Feature activation+5.63e-01

Pos logit contributions
wn+0.043
 aback+0.043
coat+0.038
 swe+0.037
 ripe+0.036
Neg logit contributions
glass-0.051
orse-0.051
 generation-0.046
work-0.045
 planet-0.044
Token a
Token ID257
Feature activation-1.50e-01

Pos logit contributions
glass+0.036
orse+0.036
work+0.032
 generation+0.031
 crew+0.031
Neg logit contributions
 aback-0.024
wn-0.024
coat-0.023
inters-0.020
ucker-0.019
Token good
Token ID922
Feature activation-2.96e+00

Pos logit contributions
wn+0.004
 aback+0.004
coat+0.003
 swe+0.003
 ripe+0.002
Neg logit contributions
glass-0.013
orse-0.013
 generation-0.012
work-0.012
 planet-0.012
Token cat
Token ID3797
Feature activation+9.02e-01

Pos logit contributions
wn+0.168
 aback+0.168
 swe+0.147
 ripe+0.147
coat+0.147
Neg logit contributions
glass-0.159
orse-0.156
 generation-0.140
work-0.139
 planet-0.138
Token and
Token ID290
Feature activation+1.50e-01

Pos logit contributions
glass+0.067
orse+0.066
 generation+0.061
work+0.060
 planet+0.059
Neg logit contributions
wn-0.032
 aback-0.032
coat-0.028
 like-0.027
 swe-0.027
Token not
Token ID407
Feature activation-1.31e-02

Pos logit contributions
glass+0.010
orse+0.010
 generation+0.009
work+0.009
 planet+0.009
Neg logit contributions
wn-0.006
 aback-0.006
 swe-0.006
coat-0.006
 like-0.006
Token make
Token ID787
Feature activation-6.30e-01

Pos logit contributions
 aback+0.001
wn+0.001
coat+0.001
 swe+0.001
 ripe+0.000
Neg logit contributions
orse-0.001
glass-0.001
work-0.001
 generation-0.001
 planet-0.001
Token a
Token ID257
Feature activation-1.37e-01

Pos logit contributions
wn+0.033
 aback+0.032
coat+0.030
 swe+0.029
 ripe+0.028
Neg logit contributions
glass-0.036
orse-0.036
 generation-0.033
work-0.032
 planet-0.031