TOP ACTIVATIONS
MAX = 0.557

!" Sara says , sob bing . Mom comes
," Lily said , sob bing . " Playing
a hard lesson . Clim bing the tree was a bad
saw Mia shaking and sob bing . " M
," Ben said , sob bing . " I just wanted
lazy , Mia . Clim bing is fun . I want
," Lily said , sob bing . " Don
my . Just keep scrub bing ," said his mom my
the water and started scrub bing the chair . It was
!" Lily said , sob bing . " Oh
They were shaking and sob bing . They nodded .
more water and kept scrub bing , but the dirt didn
," Tim said , sob bing . Sam felt
she saw a duck bob bing in the p uddle .
," Ben said , sob bing . " That
the bucket and started scrub bing . He worked hard and
saw the gray onion bob bing back and forth . She
all over . After scrub bing for a while , the
the bath robe and sob bing . Tom felt sorry for
with the brush and scrub bing with her hands until all

MIN ACTIVATIONS
MIN = -0.557

ground that he had never seen before . He bent down
giant symbol they had never seen before . It was yellow
beautiful thing she had ever seen . L ily
ches . He was never seen again .
pumpkin that had ever been seen . Everyday , the pumpkin
happy - he had never seen so much honey before .
He wished he had never seen the gun . He wished
They wished they had never seen the truck and the tree
amazed . He had never seen anything like it before .
fly . She had never seen anything like it before .
knowing that the angel had seen her and acknowledged her friendly
big and she had never seen one before .
rainbow . He had never seen anything so pretty .
He wished he had never seen the dog . He wished
wonderful animals he had never seen before . The boy was
somewhere special . He had seen animals that could reach the
down streets he had never seen before . Finally
discover toys she had not seen in a long time .
fire work they had ever seen . Tim and Sam were
unusual and she had never seen anything like it before .

INTERVAL 0.278 - 0.557
CONTAINS 0.752%

. He brave ly marched down to the beach expecting to
John stuck his head inside the hole and saw
, we 'll pick it up . That 's a very
girl , but she stayed still . He decided to approach
to her . She looked up and saw it perched atop

INTERVAL -0.000 - 0.278
CONTAINS 52.416%

we love bubbles . And we love the m op ."
go ?" Tom asked . " I don 't
't remember what happened . She searched high and
into a deep , dark hole . The old
to share and listen . They play with the bear on

INTERVAL -0.279 - -0.000
CONTAINS 44.879%

perfect . <|endoftext|> Once upon a time there was a jewel
is important to listen to the doctor 's advice . <|endoftext|>
happy . Bob was happy to have such a great Grand
I won â € ™ t pinch it , I just
a group of dolphins saw what had happened , and they

INTERVAL -0.557 - -0.279
CONTAINS 1.952%

his owner . The moral of the story is to always
Ad eline was so proud of herself . She knew she
huge room filled with lots of things . He wanted to
took off . Tom was very upset . He
. He remembered just in time and carefully climbed back up
Token!"
Token ID2474
Feature activation-6.62e-02

Pos logit contributions
bing+0.351
 cheek+0.298
 mant+0.293
 wrist+0.291
 up+0.289
Neg logit contributions
 seen-0.259
iles-0.244
elly-0.228
vious-0.224
grown-0.219
Token Sara
Token ID24799
Feature activation-1.62e-01

Pos logit contributions
 seen+0.114
iles+0.109
elly+0.101
emon+0.099
vious+0.099
Neg logit contributions
bing-0.149
 cheek-0.126
 up-0.125
 mant-0.123
 wrist-0.123
Token says
Token ID1139
Feature activation-9.02e-02

Pos logit contributions
 seen+0.254
iles+0.252
vious+0.231
elly+0.225
emon+0.219
Neg logit contributions
bing-0.383
 mant-0.331
 cheek-0.323
 wrist-0.320
 mantle-0.318
Token,
Token ID11
Feature activation-5.60e-02

Pos logit contributions
 seen+0.079
 of+0.065
iles+0.061
elly+0.059
angs+0.055
Neg logit contributions
bing-0.254
 mantle-0.242
 mant-0.242
idents-0.230
itating-0.229
Token sob
Token ID27355
Feature activation+1.47e-01

Pos logit contributions
 seen+0.093
aver+0.085
eh+0.077
 crocod+0.077
iles+0.075
Neg logit contributions
idents-0.109
bing-0.100
 regularly-0.098
 mantle-0.097
 mant-0.096
Tokenbing
Token ID4623
Feature activation+5.57e-01

Pos logit contributions
 mant+0.129
 mantle+0.117
itating+0.106
 regularly+0.106
 wrist+0.097
Neg logit contributions
iles-0.389
elly-0.381
 seen-0.372
ud-0.366
emy-0.356
Token.
Token ID13
Feature activation+1.37e-02

Pos logit contributions
bing+1.344
 mant+1.286
itating+1.231
 wrist+1.225
idents+1.206
Neg logit contributions
elly-0.662
 seen-0.657
iles-0.652
grown-0.625
vious-0.615
Token
Token ID198
Feature activation+1.38e-02

Pos logit contributions
bing+0.033
 forth+0.029
 mant+0.029
 cheek+0.028
 mantle+0.028
Neg logit contributions
 seen-0.020
elly-0.019
grown-0.018
iles-0.018
 happened-0.017
Token
Token ID198
Feature activation+1.37e-02

Pos logit contributions
bing+0.032
 cheek+0.027
 forth+0.027
 mant+0.027
 wrist+0.027
Neg logit contributions
 seen-0.022
iles-0.021
elly-0.020
grown-0.019
emon-0.019
TokenMom
Token ID29252
Feature activation+1.89e-01

Pos logit contributions
bing+0.028
 mant+0.025
 cheek+0.024
idents+0.024
 up+0.024
Neg logit contributions
 seen-0.023
iles-0.023
elly-0.023
grown-0.021
emon-0.021
Token comes
Token ID2058
Feature activation-2.29e-01

Pos logit contributions
bing+0.457
 up+0.400
 cheek+0.396
 head+0.391
 forth+0.384
Neg logit contributions
 seen-0.290
iles-0.273
elly-0.253
emon-0.250
vious-0.247
Token,"
Token ID553
Feature activation-3.73e-02

Pos logit contributions
bing+0.017
itating+0.014
 cheek+0.014
 mant+0.014
 mantle+0.014
Neg logit contributions
 seen-0.015
iles-0.014
emon-0.013
elly-0.013
grown-0.013
Token Lily
Token ID20037
Feature activation-8.52e-02

Pos logit contributions
 seen+0.053
iles+0.050
vous+0.047
elly+0.045
vious+0.045
Neg logit contributions
bing-0.094
 cheek-0.084
 up-0.084
 head-0.082
 wrist-0.081
Token said
Token ID531
Feature activation-9.09e-02

Pos logit contributions
 seen+0.126
iles+0.119
elly+0.110
vious+0.108
emon+0.108
Neg logit contributions
bing-0.211
 cheek-0.181
 mant-0.180
 up-0.179
 wrist-0.177
Token,
Token ID11
Feature activation-5.41e-02

Pos logit contributions
 seen+0.083
 of+0.064
elly+0.064
iles+0.063
grown+0.056
Neg logit contributions
bing-0.253
 mant-0.243
 mantle-0.242
itating-0.227
 edge-0.225
Token sob
Token ID27355
Feature activation+1.49e-01

Pos logit contributions
 seen+0.092
aver+0.085
 crocod+0.078
 cheese+0.077
grown+0.077
Neg logit contributions
idents-0.099
bing-0.094
 mantle-0.090
 mant-0.089
 regularly-0.089
Tokenbing
Token ID4623
Feature activation+5.54e-01

Pos logit contributions
 mant+0.097
 mantle+0.081
 regularly+0.073
itating+0.065
 wrist+0.060
Neg logit contributions
iles-0.424
elly-0.412
 seen-0.402
emy-0.401
ud-0.395
Token.
Token ID13
Feature activation+1.45e-02

Pos logit contributions
bing+1.295
 mant+1.190
itating+1.162
 wrist+1.146
 mantle+1.127
Neg logit contributions
 seen-0.748
iles-0.743
elly-0.721
vious-0.702
grown-0.685
Token
Token ID198
Feature activation+1.50e-02

Pos logit contributions
bing+0.033
 mant+0.028
 cheek+0.028
 forth+0.028
 up+0.027
Neg logit contributions
 seen-0.024
iles-0.022
elly-0.022
grown-0.021
emon-0.021
Token
Token ID198
Feature activation+1.49e-02

Pos logit contributions
bing+0.035
 mant+0.030
 forth+0.030
 cheek+0.030
 wrist+0.030
Neg logit contributions
 seen-0.023
iles-0.022
elly-0.021
grown-0.020
emon-0.020
Token"
Token ID1
Feature activation-1.16e-02

Pos logit contributions
bing+0.033
 mant+0.029
 cheek+0.028
 up+0.028
 mantle+0.028
Neg logit contributions
 seen-0.024
iles-0.023
elly-0.023
grown-0.021
emon-0.021
TokenPlaying
Token ID36530
Feature activation+9.96e-03

Pos logit contributions
 seen+0.023
iles+0.022
elly+0.021
emon+0.021
vious+0.021
Neg logit contributions
bing-0.023
 cheek-0.019
 up-0.019
 head-0.018
 mant-0.018
Token a
Token ID257
Feature activation-1.81e-01

Pos logit contributions
iles+0.090
 seen+0.088
elly+0.080
vous+0.080
emon+0.079
Neg logit contributions
bing-0.098
 up-0.090
 cheek-0.084
 wrist-0.081
 head-0.081
Token hard
Token ID1327
Feature activation+3.48e-02

Pos logit contributions
 seen+0.347
iles+0.339
elly+0.328
emon+0.316
vious+0.313
Neg logit contributions
bing-0.369
 forth-0.297
 wrist-0.295
 mant-0.295
 up-0.291
Token lesson
Token ID11483
Feature activation+8.14e-02

Pos logit contributions
bing+0.063
 cheek+0.051
 up+0.051
 head+0.050
 mant+0.049
Neg logit contributions
 seen-0.074
iles-0.072
elly-0.068
vious-0.067
emon-0.067
Token.
Token ID13
Feature activation+1.35e-02

Pos logit contributions
bing+0.141
 up+0.125
 cheek+0.113
 regularly+0.109
 head+0.109
Neg logit contributions
iles-0.173
 seen-0.171
vous-0.161
vious-0.159
emon-0.158
Token Clim
Token ID28573
Feature activation-5.18e-02

Pos logit contributions
bing+0.023
 cheek+0.019
 up+0.019
 wrist+0.018
 head+0.018
Neg logit contributions
 seen-0.030
iles-0.030
emon-0.027
vious-0.027
vous-0.027
Tokenbing
Token ID4623
Feature activation+5.53e-01

Pos logit contributions
iles+0.161
elly+0.155
ile+0.151
grown+0.150
aw+0.148
Neg logit contributions
 mant-0.026
 forehead-0.023
 doorstep-0.020
 mantle-0.019
 cheek-0.018
Token the
Token ID262
Feature activation-1.16e-02

Pos logit contributions
bing+1.178
 cheek+1.015
 mant+1.008
 wrist+1.006
itating+0.987
Neg logit contributions
iles-0.963
 seen-0.954
elly-0.883
emon-0.876
vious-0.872
Token tree
Token ID5509
Feature activation+1.59e-02

Pos logit contributions
 seen+0.022
iles+0.021
elly+0.020
vious+0.019
grown+0.019
Neg logit contributions
bing-0.024
 cheek-0.020
 up-0.020
 mant-0.020
itating-0.019
Token was
Token ID373
Feature activation+1.09e-01

Pos logit contributions
bing+0.033
 mant+0.028
itating+0.028
 cheek+0.027
 wrist+0.027
Neg logit contributions
 seen-0.030
iles-0.028
vious-0.026
emon-0.026
elly-0.025
Token a
Token ID257
Feature activation-1.81e-01

Pos logit contributions
bing+0.177
 cheek+0.141
 up+0.141
 head+0.136
 mant+0.135
Neg logit contributions
 seen-0.252
iles-0.246
elly-0.233
emon-0.230
vious-0.229
Token bad
Token ID2089
Feature activation-1.24e-01

Pos logit contributions
 seen+0.339
iles+0.329
elly+0.305
vious+0.301
emon+0.301
Neg logit contributions
bing-0.373
 cheek-0.318
 up-0.314
 head-0.309
 mant-0.306
Token saw
Token ID2497
Feature activation-1.98e-01

Pos logit contributions
bing+0.072
itating+0.062
 cheek+0.061
 wrist+0.060
 mant+0.060
Neg logit contributions
 seen-0.034
iles-0.026
grown-0.024
vious-0.023
 happened-0.022
Token Mia
Token ID32189
Feature activation-2.35e-01

Pos logit contributions
 seen+0.369
iles+0.355
elly+0.334
emon+0.329
vious+0.326
Neg logit contributions
bing-0.410
 up-0.366
 cheek-0.346
 head-0.341
 edge-0.337
Token shaking
Token ID17275
Feature activation+1.60e-01

Pos logit contributions
 seen+0.442
grown+0.369
iles+0.362
vious+0.348
 happened+0.347
Neg logit contributions
bing-0.473
idents-0.411
 mant-0.410
rier-0.398
itating-0.397
Token and
Token ID290
Feature activation+2.58e-02

Pos logit contributions
bing+0.335
 mant+0.288
itating+0.286
 cheek+0.284
 wrist+0.277
Neg logit contributions
 seen-0.292
iles-0.270
grown-0.260
elly-0.255
vious-0.251
Token sob
Token ID27355
Feature activation+1.44e-01

Pos logit contributions
bing+0.056
itating+0.049
 mant+0.047
 mantle+0.047
idents+0.046
Neg logit contributions
 seen-0.049
grown-0.040
iles-0.039
 happened-0.038
 crocod-0.036
Tokenbing
Token ID4623
Feature activation+5.52e-01

Pos logit contributions
 mant+0.156
 mantle+0.133
itating+0.128
 wrist+0.127
 cheek+0.125
Neg logit contributions
elly-0.349
 seen-0.344
iles-0.344
ud-0.329
angs-0.329
Token.
Token ID13
Feature activation+9.73e-03

Pos logit contributions
bing+1.297
 mant+1.178
 wrist+1.154
itating+1.142
 forth+1.124
Neg logit contributions
 seen-0.787
iles-0.774
elly-0.738
vious-0.725
grown-0.724
Token
Token ID198
Feature activation+8.61e-03

Pos logit contributions
bing+0.024
 up+0.021
 cheek+0.021
 head+0.020
 mant+0.020
Neg logit contributions
 seen-0.015
iles-0.014
elly-0.013
vious-0.012
emon-0.012
Token
Token ID198
Feature activation+8.28e-03

Pos logit contributions
bing+0.017
 up+0.015
 cheek+0.015
 head+0.014
 mant+0.014
Neg logit contributions
 seen-0.016
iles-0.016
elly-0.015
vious-0.015
emon-0.015
Token"
Token ID1
Feature activation-1.69e-02

Pos logit contributions
bing+0.017
 cheek+0.014
 up+0.014
 head+0.014
 mant+0.014
Neg logit contributions
 seen-0.015
iles-0.015
vious-0.014
elly-0.014
emon-0.014
TokenM
Token ID44
Feature activation-8.76e-02

Pos logit contributions
 seen+0.032
iles+0.031
elly+0.029
vious+0.029
emon+0.028
Neg logit contributions
bing-0.035
 cheek-0.029
 up-0.028
 head-0.028
 forth-0.028
Token,"
Token ID553
Feature activation-3.96e-02

Pos logit contributions
bing+0.378
 cheek+0.319
 mant+0.317
 up+0.316
 wrist+0.314
Neg logit contributions
 seen-0.288
iles-0.273
elly-0.253
vious-0.248
grown-0.248
Token Ben
Token ID3932
Feature activation-2.32e-01

Pos logit contributions
 seen+0.050
iles+0.048
elly+0.042
emon+0.042
vous+0.042
Neg logit contributions
bing-0.107
 cheek-0.095
 up-0.094
 head-0.093
 wrist-0.092
Token said
Token ID531
Feature activation-9.38e-02

Pos logit contributions
 seen+0.376
iles+0.366
elly+0.335
grown+0.330
emon+0.328
Neg logit contributions
bing-0.539
 cheek-0.467
 up-0.464
 head-0.458
 wrist-0.451
Token,
Token ID11
Feature activation-5.63e-02

Pos logit contributions
 seen+0.110
 of+0.092
elly+0.087
vious+0.086
grown+0.082
Neg logit contributions
bing-0.248
 mant-0.228
 mantle-0.224
itating-0.219
itate-0.215
Token sob
Token ID27355
Feature activation+1.45e-01

Pos logit contributions
 seen+0.097
aver+0.087
grown+0.083
 cheese+0.082
eh+0.081
Neg logit contributions
bing-0.104
idents-0.102
itating-0.094
 ourselves-0.092
 regularly-0.092
Tokenbing
Token ID4623
Feature activation+5.50e-01

Pos logit contributions
 mant+0.130
 mantle+0.114
idents+0.106
 wrist+0.106
itating+0.105
Neg logit contributions
elly-0.375
iles-0.374
 seen-0.358
aver-0.348
ud-0.348
Token.
Token ID13
Feature activation+1.29e-02

Pos logit contributions
bing+1.300
 mant+1.176
 wrist+1.157
itating+1.145
idents+1.124
Neg logit contributions
 seen-0.765
iles-0.737
vious-0.725
elly-0.724
grown-0.690
Token "
Token ID366
Feature activation+9.72e-02

Pos logit contributions
bing+0.026
 cheek+0.022
 up+0.022
 mant+0.021
 wrist+0.021
Neg logit contributions
 seen-0.024
iles-0.024
elly-0.022
grown-0.022
emon-0.022
TokenI
Token ID40
Feature activation+4.66e-02

Pos logit contributions
bing+0.201
 up+0.174
 forth+0.170
 mant+0.169
 head+0.168
Neg logit contributions
 seen-0.179
iles-0.171
elly-0.170
vious-0.165
vous-0.163
Token just
Token ID655
Feature activation+5.48e-02

Pos logit contributions
bing+0.084
 up+0.075
 cheek+0.073
 head+0.072
idents+0.072
Neg logit contributions
iles-0.094
 seen-0.088
elly-0.084
emon-0.084
grown-0.082
Token wanted
Token ID2227
Feature activation-1.20e-01

Pos logit contributions
bing+0.102
 up+0.100
 head+0.100
 cheek+0.095
 forth+0.095
Neg logit contributions
iles-0.103
elly-0.093
emon-0.091
 seen-0.086
br-0.083
Token lazy
Token ID16931
Feature activation-1.03e-01

Pos logit contributions
elly+0.247
 seen+0.238
iles+0.233
grown+0.232
vious+0.225
Neg logit contributions
bing-0.129
 up-0.123
 head-0.117
 cheek-0.111
 forth-0.103
Token,
Token ID11
Feature activation-5.94e-02

Pos logit contributions
 seen+0.218
iles+0.217
emon+0.210
vous+0.199
grown+0.198
Neg logit contributions
bing-0.181
 up-0.157
 head-0.151
 cheek-0.150
 forth-0.148
Token Mia
Token ID32189
Feature activation-2.41e-01

Pos logit contributions
 seen+0.084
iles+0.079
elly+0.072
grown+0.071
emon+0.070
Neg logit contributions
bing-0.150
 cheek-0.133
 up-0.132
 head-0.131
 wrist-0.129
Token.
Token ID13
Feature activation+1.06e-02

Pos logit contributions
 seen+0.499
iles+0.483
emon+0.459
elly+0.455
vious+0.451
Neg logit contributions
bing-0.450
 up-0.382
 head-0.376
 cheek-0.373
 forth-0.356
Token Clim
Token ID28573
Feature activation-5.49e-02

Pos logit contributions
bing+0.016
 cheek+0.014
 head+0.014
 wrist+0.013
 up+0.013
Neg logit contributions
 seen-0.026
iles-0.024
grown-0.023
emon-0.022
vious-0.022
Tokenbing
Token ID4623
Feature activation+5.50e-01

Pos logit contributions
iles+0.187
elly+0.177
 seen+0.175
grown+0.172
vious+0.170
Neg logit contributions
 mant-0.016
 cheek-0.012
bing-0.012
 forehead-0.012
 wrist-0.010
Token is
Token ID318
Feature activation+1.09e-02

Pos logit contributions
bing+1.155
 cheek+0.980
 wrist+0.958
 mant+0.957
 up+0.953
Neg logit contributions
 seen-0.994
iles-0.975
elly-0.901
vious-0.888
emon-0.885
Token fun
Token ID1257
Feature activation-1.42e-01

Pos logit contributions
bing+0.018
 up+0.015
 cheek+0.014
 head+0.014
 forth+0.014
Neg logit contributions
 seen-0.025
iles-0.024
elly-0.023
vious-0.023
emon-0.023
Token.
Token ID13
Feature activation+8.70e-03

Pos logit contributions
 seen+0.275
iles+0.266
elly+0.247
emon+0.246
vious+0.245
Neg logit contributions
bing-0.284
 up-0.238
 cheek-0.237
 head-0.231
 wrist-0.228
Token I
Token ID314
Feature activation+5.36e-03

Pos logit contributions
bing+0.015
 cheek+0.013
 head+0.012
 wrist+0.012
 mant+0.012
Neg logit contributions
 seen-0.019
iles-0.018
grown-0.017
emon-0.017
elly-0.017
Token want
Token ID765
Feature activation-1.28e-01

Pos logit contributions
bing+0.010
 up+0.009
 cheek+0.009
 head+0.009
idents+0.008
Neg logit contributions
 seen-0.010
iles-0.010
elly-0.010
emon-0.009
grown-0.009
Token,"
Token ID553
Feature activation-4.12e-02

Pos logit contributions
 seen+0.178
iles+0.172
elly+0.160
emon+0.157
vious+0.157
Neg logit contributions
bing-0.205
 cheek-0.172
 up-0.171
 head-0.168
 mant-0.167
Token Lily
Token ID20037
Feature activation-8.85e-02

Pos logit contributions
 seen+0.058
iles+0.053
elly+0.050
vious+0.049
emon+0.049
Neg logit contributions
bing-0.105
 mant-0.090
 cheek-0.090
itating-0.090
 up-0.089
Token said
Token ID531
Feature activation-9.26e-02

Pos logit contributions
 seen+0.105
vious+0.081
iles+0.078
aver+0.075
emon+0.074
Neg logit contributions
bing-0.250
 mant-0.220
itating-0.211
rier-0.208
 edge-0.208
Token,
Token ID11
Feature activation-5.72e-02

Pos logit contributions
 seen+0.096
 of+0.086
 happened+0.069
 Tom+0.064
grown+0.061
Neg logit contributions
bing-0.247
 edge-0.231
 mant-0.228
 mantle-0.227
idents-0.225
Token sob
Token ID27355
Feature activation+1.47e-01

Pos logit contributions
 seen+0.097
aver+0.084
eh+0.080
 cheese+0.079
grown+0.079
Neg logit contributions
bing-0.106
idents-0.104
 ourselves-0.101
itating-0.095
 forth-0.094
Tokenbing
Token ID4623
Feature activation+5.49e-01

Pos logit contributions
 mant+0.128
 regularly+0.112
itating+0.111
 mantle+0.111
 edge+0.104
Neg logit contributions
elly-0.351
 seen-0.350
ud-0.330
emy-0.329
iles-0.328
Token.
Token ID13
Feature activation+1.23e-02

Pos logit contributions
bing+1.293
itating+1.173
 mant+1.155
idents+1.149
 forth+1.130
Neg logit contributions
 seen-0.745
vious-0.666
elly-0.663
grown-0.654
iles-0.622
Token
Token ID198
Feature activation+9.32e-03

Pos logit contributions
bing+0.030
 forth+0.027
itating+0.025
 mant+0.025
 heads+0.025
Neg logit contributions
 seen-0.017
elly-0.016
 happened-0.015
grown-0.015
emon-0.015
Token
Token ID198
Feature activation+9.05e-03

Pos logit contributions
bing+0.024
 forth+0.021
 heads+0.020
 cheek+0.020
 head+0.020
Neg logit contributions
 seen-0.013
elly-0.011
iles-0.011
emon-0.011
grown-0.010
Token"
Token ID1
Feature activation-1.54e-02

Pos logit contributions
bing+0.021
rier+0.020
 forth+0.019
 mant+0.019
 up+0.019
Neg logit contributions
elly-0.011
 seen-0.011
emon-0.010
Tom-0.010
ha-0.010
TokenDon
Token ID3987
Feature activation+6.48e-02

Pos logit contributions
 seen+0.028
iles+0.026
elly+0.026
vious+0.025
emon+0.025
Neg logit contributions
bing-0.032
 up-0.028
 cheek-0.028
 head-0.027
 mant-0.027
Tokenmy
Token ID1820
Feature activation-4.81e-02

Pos logit contributions
 seen+0.303
iles+0.300
emon+0.290
elly+0.278
grown+0.278
Neg logit contributions
bing-0.267
 head-0.233
 cheek-0.230
 up-0.230
 forth-0.228
Token.
Token ID13
Feature activation+2.39e-02

Pos logit contributions
 seen+0.081
iles+0.075
elly+0.071
vious+0.069
grown+0.069
Neg logit contributions
bing-0.108
 mant-0.094
 mantle-0.092
itating-0.092
 wrist-0.092
Token Just
Token ID2329
Feature activation+1.48e-01

Pos logit contributions
bing+0.051
 up+0.043
 cheek+0.041
 forth+0.041
 mant+0.041
Neg logit contributions
 seen-0.042
iles-0.041
elly-0.040
emon-0.039
vious-0.038
Token keep
Token ID1394
Feature activation+1.41e-01

Pos logit contributions
bing+0.280
 cheek+0.227
itating+0.225
 up+0.224
 mant+0.223
Neg logit contributions
 seen-0.302
iles-0.288
emon-0.269
elly-0.267
vious-0.266
Token scrub
Token ID27268
Feature activation+1.67e-01

Pos logit contributions
bing+0.276
 cheek+0.259
 wrist+0.259
itating+0.258
 mant+0.256
Neg logit contributions
 seen-0.250
iles-0.230
grown-0.223
emon-0.201
 cheese-0.201
Tokenbing
Token ID4623
Feature activation+5.48e-01

Pos logit contributions
 doorstep+0.086
itating+0.082
 mant+0.074
 wrist+0.070
 mantle+0.067
Neg logit contributions
iles-0.475
 seen-0.448
aks-0.444
elly-0.443
aked-0.443
Token,"
Token ID553
Feature activation-3.52e-02

Pos logit contributions
bing+0.996
 wrist+0.969
itating+0.960
 cheek+0.924
etically+0.903
Neg logit contributions
iles-0.961
 seen-0.887
vious-0.863
elly-0.863
 cheese-0.862
Token said
Token ID531
Feature activation-9.36e-02

Pos logit contributions
 seen+0.043
aver+0.037
 cheese+0.034
iles+0.033
 happened+0.033
Neg logit contributions
bing-0.094
itating-0.087
 mant-0.084
 edge-0.082
 mantle-0.080
Token his
Token ID465
Feature activation-8.63e-02

Pos logit contributions
 seen+0.142
iles+0.121
emon+0.118
aver+0.118
grown+0.116
Neg logit contributions
bing-0.226
itating-0.195
 mant-0.191
idents-0.191
 mantle-0.190
Token mom
Token ID1995
Feature activation+2.57e-01

Pos logit contributions
iles+0.170
 seen+0.169
aver+0.161
 cheese+0.160
 crocod+0.159
Neg logit contributions
bing-0.156
itating-0.136
itate-0.117
 regularly-0.117
 mant-0.116
Tokenmy
Token ID1820
Feature activation-4.84e-02

Pos logit contributions
bing+0.641
 mant+0.580
 edge+0.568
 mantle+0.561
idents+0.560
Neg logit contributions
 seen-0.356
iles-0.295
vious-0.290
ah-0.287
vous-0.287
Token the
Token ID262
Feature activation-8.44e-03

Pos logit contributions
bing+0.157
 cheek+0.140
itating+0.137
 wrist+0.134
idents+0.134
Neg logit contributions
 seen-0.108
grown-0.097
iles-0.096
elly-0.094
emon-0.092
Token water
Token ID1660
Feature activation-5.56e-02

Pos logit contributions
 seen+0.014
elly+0.013
vious+0.012
iles+0.012
grown+0.012
Neg logit contributions
bing-0.019
itating-0.016
idents-0.016
etically-0.016
itate-0.016
Token and
Token ID290
Feature activation+2.81e-02

Pos logit contributions
 seen+0.091
iles+0.084
elly+0.082
emon+0.079
vious+0.078
Neg logit contributions
bing-0.123
 cheek-0.107
 wrist-0.106
 mant-0.106
itating-0.105
Token started
Token ID2067
Feature activation+9.94e-02

Pos logit contributions
bing+0.064
itating+0.056
 cheek+0.056
 mant+0.054
 wrist+0.054
Neg logit contributions
 seen-0.047
iles-0.041
emon-0.039
grown-0.039
vious-0.039
Token scrub
Token ID27268
Feature activation+1.70e-01

Pos logit contributions
bing+0.248
 cheek+0.220
itating+0.218
 wrist+0.216
 mant+0.215
Neg logit contributions
 seen-0.138
iles-0.122
grown-0.118
emon-0.118
elly-0.116
Tokenbing
Token ID4623
Feature activation+5.48e-01

Pos logit contributions
itating+0.209
 doorstep+0.193
 cheek+0.187
 mant+0.185
 mantle+0.179
Neg logit contributions
 seen-0.385
iles-0.365
ud-0.363
aked-0.356
grown-0.356
Token the
Token ID262
Feature activation-7.92e-03

Pos logit contributions
itating+1.169
bing+1.159
 cheek+1.139
 wrist+1.137
 doorstep+1.134
Neg logit contributions
 seen-0.703
iles-0.697
 cheese-0.697
grown-0.675
elly-0.665
Token chair
Token ID5118
Feature activation+1.22e-01

Pos logit contributions
 seen+0.014
elly+0.013
 cheese+0.013
grown+0.013
iles+0.013
Neg logit contributions
bing-0.015
itating-0.014
ishly-0.014
etically-0.013
itate-0.013
Token.
Token ID13
Feature activation+1.56e-02

Pos logit contributions
bing+0.284
 cheek+0.247
 mant+0.247
itating+0.247
 wrist+0.245
Neg logit contributions
 seen-0.183
emon-0.153
iles-0.153
elly-0.151
vious-0.149
Token It
Token ID632
Feature activation-2.37e-02

Pos logit contributions
bing+0.039
 forth+0.033
rier+0.033
itating+0.033
 cheek+0.032
Neg logit contributions
 Some-0.022
 seen-0.020
 cheese-0.019
elly-0.019
 Tom-0.018
Token was
Token ID373
Feature activation+1.07e-01

Pos logit contributions
 seen+0.039
iles+0.036
vious+0.034
vous+0.034
emon+0.034
Neg logit contributions
bing-0.053
 wrist-0.048
 cheek-0.047
 mant-0.046
itating-0.045
Token!"
Token ID2474
Feature activation-6.87e-02

Pos logit contributions
 seen+0.184
iles+0.177
elly+0.163
vious+0.161
emon+0.160
Neg logit contributions
bing-0.257
 cheek-0.219
 mant-0.216
 up-0.216
 wrist-0.215
Token Lily
Token ID20037
Feature activation-9.08e-02

Pos logit contributions
 seen+0.137
iles+0.129
elly+0.123
vious+0.121
emon+0.120
Neg logit contributions
bing-0.138
 cheek-0.113
 up-0.111
 mant-0.111
 wrist-0.109
Token said
Token ID531
Feature activation-9.71e-02

Pos logit contributions
 seen+0.140
iles+0.117
vious+0.113
elly+0.109
aver+0.109
Neg logit contributions
bing-0.224
 mant-0.197
 cheek-0.189
itating-0.189
 mantle-0.187
Token,
Token ID11
Feature activation-5.90e-02

Pos logit contributions
 seen+0.086
 of+0.074
aver+0.054
elly+0.053
grown+0.052
Neg logit contributions
bing-0.270
 mant-0.261
 mantle-0.260
idents-0.258
 edge-0.247
Token sob
Token ID27355
Feature activation+1.46e-01

Pos logit contributions
 seen+0.095
aver+0.087
eh+0.082
 crocod+0.078
grown+0.076
Neg logit contributions
idents-0.115
 ourselves-0.105
bing-0.104
 regularly-0.100
itating-0.097
Tokenbing
Token ID4623
Feature activation+5.47e-01

Pos logit contributions
 mant+0.097
 mantle+0.079
 regularly+0.072
idents+0.071
itating+0.064
Neg logit contributions
iles-0.395
 seen-0.394
elly-0.393
ud-0.385
emy-0.380
Token.
Token ID13
Feature activation+1.52e-02

Pos logit contributions
bing+1.284
 mant+1.196
itating+1.158
idents+1.157
 wrist+1.135
Neg logit contributions
 seen-0.731
iles-0.671
elly-0.671
vious-0.664
grown-0.662
Token
Token ID198
Feature activation+1.02e-02

Pos logit contributions
bing+0.038
 forth+0.034
 mant+0.033
 cheek+0.032
 mantle+0.032
Neg logit contributions
 seen-0.021
elly-0.020
grown-0.019
 happened-0.018
iles-0.018
Token
Token ID198
Feature activation+1.00e-02

Pos logit contributions
bing+0.025
 forth+0.022
 mant+0.022
 cheek+0.022
 head+0.022
Neg logit contributions
 seen-0.014
iles-0.013
elly-0.013
grown-0.012
emon-0.012
Token"
Token ID1
Feature activation-1.78e-02

Pos logit contributions
bing+0.024
 mant+0.022
 cheek+0.022
rier+0.021
idents+0.021
Neg logit contributions
 seen-0.014
elly-0.013
iles-0.013
grown-0.012
emon-0.012
TokenOh
Token ID5812
Feature activation-3.77e-02

Pos logit contributions
 seen+0.032
iles+0.031
elly+0.030
vious+0.029
emon+0.029
Neg logit contributions
bing-0.038
 up-0.033
 mant-0.032
 cheek-0.032
 head-0.031
Token They
Token ID1119
Feature activation+5.60e-02

Pos logit contributions
bing+0.012
 up+0.010
 cheek+0.010
 head+0.010
 wrist+0.009
Neg logit contributions
 seen-0.014
iles-0.014
vious-0.013
emon-0.013
elly-0.013
Token were
Token ID547
Feature activation-4.50e-02

Pos logit contributions
bing+0.112
 head+0.102
 up+0.099
 heads+0.098
 cheek+0.097
Neg logit contributions
iles-0.103
 seen-0.098
elly-0.098
emon-0.094
vous-0.091
Token shaking
Token ID17275
Feature activation+1.57e-01

Pos logit contributions
 seen+0.097
iles+0.090
grown+0.084
aver+0.084
elly+0.083
Neg logit contributions
bing-0.084
 cheek-0.069
idents-0.069
 wrist-0.068
 mant-0.068
Token and
Token ID290
Feature activation+2.18e-02

Pos logit contributions
bing+0.344
 cheek+0.299
 mant+0.297
itating+0.291
 wrist+0.290
Neg logit contributions
 seen-0.274
iles-0.254
grown-0.245
elly-0.238
vious-0.230
Token sob
Token ID27355
Feature activation+1.41e-01

Pos logit contributions
bing+0.042
 mant+0.035
itating+0.035
 mantle+0.035
 wrist+0.035
Neg logit contributions
 seen-0.045
iles-0.040
grown-0.039
vious-0.037
elly-0.036
Tokenbing
Token ID4623
Feature activation+5.47e-01

Pos logit contributions
 mant+0.111
 mantle+0.090
 wrist+0.085
 regularly+0.079
 cheek+0.078
Neg logit contributions
elly-0.392
iles-0.390
ud-0.372
angs-0.370
 seen-0.369
Token.
Token ID13
Feature activation+7.68e-03

Pos logit contributions
bing+1.279
 mant+1.145
 wrist+1.111
itating+1.104
 forth+1.094
Neg logit contributions
 seen-0.802
iles-0.790
elly-0.747
vious-0.732
grown-0.728
Token They
Token ID1119
Feature activation+5.72e-02

Pos logit contributions
bing+0.014
 up+0.012
 cheek+0.011
 head+0.011
 wrist+0.011
Neg logit contributions
 seen-0.016
iles-0.016
vious-0.015
elly-0.015
emon-0.015
Token nodded
Token ID14464
Feature activation+1.37e-02

Pos logit contributions
bing+0.124
 cheek+0.106
 up+0.106
 head+0.104
 wrist+0.103
Neg logit contributions
 seen-0.100
iles-0.098
elly-0.091
emon-0.090
vious-0.089
Token.
Token ID13
Feature activation+7.60e-03

Pos logit contributions
bing+0.033
 mant+0.030
 mantle+0.029
itating+0.029
idents+0.029
Neg logit contributions
 seen-0.021
grown-0.018
iles-0.018
elly-0.017
vious-0.017
Token
Token ID198
Feature activation+1.05e-02

Pos logit contributions
bing+0.014
 up+0.012
 cheek+0.012
 head+0.011
 wrist+0.011
Neg logit contributions
 seen-0.016
iles-0.015
vious-0.014
elly-0.014
emon-0.014
Token more
Token ID517
Feature activation-4.48e-02

Pos logit contributions
bing+0.382
 cheek+0.324
 up+0.322
 head+0.315
 mant+0.315
Neg logit contributions
 seen-0.306
iles-0.295
elly-0.273
emon-0.269
vious-0.268
Token water
Token ID1660
Feature activation-5.81e-02

Pos logit contributions
 seen+0.083
iles+0.081
elly+0.074
emon+0.074
vious+0.073
Neg logit contributions
bing-0.094
 cheek-0.079
 up-0.079
 head-0.077
 wrist-0.077
Token and
Token ID290
Feature activation+2.56e-02

Pos logit contributions
 seen+0.102
iles+0.095
elly+0.091
vious+0.089
grown+0.088
Neg logit contributions
bing-0.125
 cheek-0.107
 mant-0.106
 wrist-0.106
itating-0.105
Token kept
Token ID4030
Feature activation+2.75e-02

Pos logit contributions
bing+0.054
 cheek+0.046
 up+0.045
 mant+0.044
 wrist+0.044
Neg logit contributions
 seen-0.047
iles-0.045
elly-0.042
emon-0.041
vious-0.041
Token scrub
Token ID27268
Feature activation+1.67e-01

Pos logit contributions
bing+0.057
itating+0.057
idents+0.054
 cheek+0.053
ishly+0.053
Neg logit contributions
 seen-0.048
grown-0.042
 cheese-0.037
emon-0.037
iles-0.036
Tokenbing
Token ID4623
Feature activation+5.47e-01

Pos logit contributions
itating+0.154
 doorstep+0.124
 mant+0.122
 cheek+0.120
itate+0.120
Neg logit contributions
 seen-0.437
aked-0.408
iles-0.408
ud-0.406
grown-0.405
Token,
Token ID11
Feature activation-5.65e-02

Pos logit contributions
bing+1.156
itating+1.136
 wrist+1.087
 cheek+1.074
etically+1.067
Neg logit contributions
 seen-0.796
iles-0.760
 cheese-0.747
grown-0.729
vious-0.708
Token but
Token ID475
Feature activation+8.15e-02

Pos logit contributions
 seen+0.106
grown+0.087
vious+0.085
 cheese+0.085
emon+0.084
Neg logit contributions
bing-0.115
itating-0.102
idents-0.097
etically-0.097
 wrist-0.097
Token the
Token ID262
Feature activation-8.83e-03

Pos logit contributions
bing+0.172
 cheek+0.150
itating+0.147
 mant+0.147
 wrist+0.145
Neg logit contributions
 seen-0.152
iles-0.136
grown-0.129
emon-0.125
vious-0.124
Token dirt
Token ID13647
Feature activation-3.14e-02

Pos logit contributions
 seen+0.015
iles+0.014
vious+0.014
elly+0.013
grown+0.013
Neg logit contributions
bing-0.020
itating-0.016
etically-0.016
 cheek-0.015
idents-0.015
Token didn
Token ID1422
Feature activation+1.21e-01

Pos logit contributions
 seen+0.057
iles+0.054
elly+0.050
vious+0.050
emon+0.050
Neg logit contributions
bing-0.065
 cheek-0.056
 wrist-0.056
 mant-0.055
itating-0.055
Token,"
Token ID553
Feature activation-4.31e-02

Pos logit contributions
bing+0.126
idents+0.108
 mantle+0.105
 wrist+0.104
 mant+0.104
Neg logit contributions
 seen-0.108
iles-0.090
vious-0.087
elly-0.087
emon-0.085
Token Tim
Token ID5045
Feature activation-1.46e-01

Pos logit contributions
 seen+0.064
iles+0.064
elly+0.061
emon+0.058
aked+0.057
Neg logit contributions
bing-0.102
 up-0.092
 forth-0.091
 cheek-0.091
idents-0.089
Token said
Token ID531
Feature activation-9.69e-02

Pos logit contributions
iles+0.253
 seen+0.253
grown+0.234
elly+0.234
emon+0.228
Neg logit contributions
bing-0.317
 up-0.280
 cheek-0.279
 head-0.277
 heads-0.266
Token,
Token ID11
Feature activation-5.91e-02

Pos logit contributions
 seen+0.121
iles+0.100
elly+0.100
vious+0.099
grown+0.096
Neg logit contributions
bing-0.247
 mant-0.229
 mantle-0.224
 wrist-0.217
itating-0.217
Token sob
Token ID27355
Feature activation+1.42e-01

Pos logit contributions
 seen+0.110
aver+0.096
grown+0.095
vious+0.093
iles+0.093
Neg logit contributions
bing-0.112
idents-0.103
 mant-0.097
 mantle-0.096
itating-0.096
Tokenbing
Token ID4623
Feature activation+5.47e-01

Pos logit contributions
 mant+0.105
 mantle+0.092
 wrist+0.080
itating+0.080
idents+0.079
Neg logit contributions
iles-0.399
elly-0.392
 seen-0.379
emy-0.376
aver-0.372
Token.
Token ID13
Feature activation+8.45e-03

Pos logit contributions
bing+1.205
 mant+1.053
 wrist+1.040
itating+1.022
 cheek+1.011
Neg logit contributions
 seen-0.896
iles-0.870
elly-0.816
vious-0.813
emon-0.798
Token
Token ID198
Feature activation+9.90e-03

Pos logit contributions
bing+0.016
 up+0.013
 cheek+0.013
 head+0.013
 edge+0.012
Neg logit contributions
 seen-0.018
iles-0.017
vious-0.016
emon-0.016
elly-0.016
Token
Token ID198
Feature activation+9.79e-03

Pos logit contributions
bing+0.019
 up+0.016
 cheek+0.016
 head+0.015
 edge+0.015
Neg logit contributions
 seen-0.020
iles-0.019
vious-0.018
emon-0.018
elly-0.018
TokenSam
Token ID16305
Feature activation-2.50e-01

Pos logit contributions
bing+0.025
 up+0.021
 cheek+0.021
 forth+0.021
 head+0.020
Neg logit contributions
 seen-0.015
iles-0.014
vious-0.013
emon-0.013
elly-0.013
Token felt
Token ID2936
Feature activation+1.15e-01

Pos logit contributions
 seen+0.460
iles+0.433
vious+0.405
elly+0.401
emon+0.401
Neg logit contributions
bing-0.533
 cheek-0.442
 mant-0.438
 up-0.436
itating-0.430
Token she
Token ID673
Feature activation+1.40e-01

Pos logit contributions
 seen+0.093
iles+0.089
elly+0.082
vious+0.081
emon+0.081
Neg logit contributions
bing-0.129
 cheek-0.110
 up-0.108
 mant-0.107
 wrist-0.107
Token saw
Token ID2497
Feature activation-1.95e-01

Pos logit contributions
bing+0.340
 up+0.308
 head+0.303
 forth+0.298
 edge+0.291
Neg logit contributions
iles-0.202
 seen-0.192
elly-0.188
emon-0.186
vious-0.181
Token a
Token ID257
Feature activation-1.77e-01

Pos logit contributions
iles+0.310
 seen+0.299
elly+0.297
emon+0.281
vious+0.274
Neg logit contributions
bing-0.441
 up-0.421
 edge-0.379
 cheek-0.372
 head-0.369
Token duck
Token ID22045
Feature activation-1.24e-02

Pos logit contributions
 seen+0.310
iles+0.297
elly+0.276
emon+0.272
vious+0.271
Neg logit contributions
bing-0.387
 cheek-0.326
 up-0.325
 mant-0.318
 wrist-0.317
Token bob
Token ID29202
Feature activation+1.25e-01

Pos logit contributions
 seen+0.022
iles+0.021
elly+0.019
grown+0.019
emon+0.019
Neg logit contributions
bing-0.027
 mant-0.023
 cheek-0.023
 wrist-0.023
idents-0.023
Tokenbing
Token ID4623
Feature activation+5.47e-01

Pos logit contributions
 mant+0.129
 mantle+0.112
idents+0.108
 doorstep+0.108
 wrist+0.104
Neg logit contributions
 seen-0.313
ah-0.311
iles-0.306
aver-0.293
ile-0.290
Token in
Token ID287
Feature activation+7.07e-02

Pos logit contributions
bing+1.016
 wrist+0.872
 cheek+0.861
idents+0.859
 mant+0.856
Neg logit contributions
 seen-1.082
iles-1.056
elly-0.983
emon-0.979
vious-0.971
Token the
Token ID262
Feature activation-7.26e-03

Pos logit contributions
bing+0.131
 up+0.118
 edge+0.109
 cheek+0.109
 head+0.109
Neg logit contributions
iles-0.147
 seen-0.140
elly-0.138
vious-0.134
emon-0.134
Token p
Token ID279
Feature activation-4.61e-03

Pos logit contributions
 seen+0.011
iles+0.010
vious+0.010
elly+0.009
emon+0.009
Neg logit contributions
bing-0.018
idents-0.015
itating-0.015
 wrist-0.015
 regularly-0.015
Tokenuddle
Token ID24500
Feature activation+1.67e-01

Pos logit contributions
 seen+0.011
iles+0.011
elly+0.011
emon+0.010
vious+0.010
Neg logit contributions
bing-0.007
 cheek-0.005
 up-0.005
 head-0.005
 wrist-0.005
Token.
Token ID13
Feature activation+1.23e-02

Pos logit contributions
bing+0.354
 cheek+0.298
 mant+0.295
 wrist+0.293
 up+0.290
Neg logit contributions
 seen-0.302
iles-0.289
elly-0.273
vious-0.267
emon-0.265
Token,"
Token ID553
Feature activation-4.18e-02

Pos logit contributions
bing+0.362
 mant+0.312
itating+0.305
 cheek+0.305
 mantle+0.302
Neg logit contributions
 seen-0.231
iles-0.215
elly-0.199
vious-0.195
emon-0.191
Token Ben
Token ID3932
Feature activation-2.35e-01

Pos logit contributions
 seen+0.087
iles+0.084
elly+0.079
emon+0.078
vious+0.078
Neg logit contributions
bing-0.078
 up-0.066
 cheek-0.066
 head-0.063
 wrist-0.063
Token said
Token ID531
Feature activation-9.69e-02

Pos logit contributions
 seen+0.363
iles+0.347
elly+0.318
emon+0.312
vious+0.312
Neg logit contributions
bing-0.567
 cheek-0.488
 up-0.486
 mant-0.476
 head-0.476
Token,
Token ID11
Feature activation-5.89e-02

Pos logit contributions
 seen+0.131
vious+0.111
elly+0.111
grown+0.107
emon+0.106
Neg logit contributions
bing-0.244
 mant-0.219
 mantle-0.213
idents-0.212
 forth-0.211
Token sob
Token ID27355
Feature activation+1.43e-01

Pos logit contributions
 seen+0.105
aver+0.093
grown+0.090
vious+0.087
 ago+0.087
Neg logit contributions
bing-0.108
idents-0.106
 ourselves-0.095
itating-0.095
 mant-0.094
Tokenbing
Token ID4623
Feature activation+5.47e-01

Pos logit contributions
 mant+0.141
 mantle+0.126
idents+0.118
 wrist+0.114
itating+0.114
Neg logit contributions
elly-0.362
iles-0.360
 seen-0.346
angs-0.335
aver-0.334
Token.
Token ID13
Feature activation+1.14e-02

Pos logit contributions
bing+1.289
 mant+1.160
 wrist+1.139
itating+1.124
idents+1.114
Neg logit contributions
 seen-0.769
iles-0.742
vious-0.728
elly-0.724
grown-0.690
Token
Token ID198
Feature activation+9.71e-03

Pos logit contributions
bing+0.023
 cheek+0.020
 up+0.019
 mant+0.019
 head+0.019
Neg logit contributions
 seen-0.022
iles-0.021
elly-0.020
emon-0.019
vious-0.019
Token
Token ID198
Feature activation+9.93e-03

Pos logit contributions
bing+0.020
 cheek+0.017
 up+0.017
 head+0.017
 mant+0.017
Neg logit contributions
 seen-0.018
iles-0.017
elly-0.016
emon-0.016
vious-0.016
Token"
Token ID1
Feature activation-1.57e-02

Pos logit contributions
bing+0.021
 cheek+0.018
 up+0.018
 mant+0.017
 wrist+0.017
Neg logit contributions
 seen-0.018
iles-0.018
elly-0.017
emon-0.016
vious-0.016
TokenThat
Token ID2504
Feature activation+1.47e-01

Pos logit contributions
 seen+0.030
iles+0.029
elly+0.027
vious+0.027
emon+0.027
Neg logit contributions
bing-0.032
 cheek-0.027
 up-0.026
 mant-0.026
 head-0.026
Token the
Token ID262
Feature activation-8.44e-03

Pos logit contributions
bing+0.444
 cheek+0.382
 up+0.380
 head+0.372
 mant+0.372
Neg logit contributions
 seen-0.291
iles-0.279
elly-0.256
emon-0.251
vious-0.251
Token bucket
Token ID19236
Feature activation+5.86e-03

Pos logit contributions
 seen+0.015
elly+0.014
vious+0.014
grown+0.014
iles+0.014
Neg logit contributions
bing-0.017
itating-0.014
itate-0.014
 wrist-0.014
 cheek-0.013
Token and
Token ID290
Feature activation+2.78e-02

Pos logit contributions
bing+0.012
 cheek+0.010
 mant+0.010
 wrist+0.010
idents+0.010
Neg logit contributions
 seen-0.011
iles-0.010
elly-0.010
vious-0.010
grown-0.009
Token started
Token ID2067
Feature activation+9.99e-02

Pos logit contributions
bing+0.063
itating+0.056
 cheek+0.054
 mant+0.052
 wrist+0.051
Neg logit contributions
 seen-0.049
iles-0.040
grown-0.039
emon-0.039
vious-0.039
Token scrub
Token ID27268
Feature activation+1.69e-01

Pos logit contributions
bing+0.241
 cheek+0.218
itating+0.217
 mant+0.214
 mantle+0.213
Neg logit contributions
 seen-0.142
grown-0.125
emon-0.117
elly-0.116
iles-0.114
Tokenbing
Token ID4623
Feature activation+5.46e-01

Pos logit contributions
itating+0.184
 mant+0.161
 cheek+0.161
 doorstep+0.157
 wrist+0.151
Neg logit contributions
 seen-0.403
ud-0.388
aked-0.385
iles-0.384
grown-0.377
Token.
Token ID13
Feature activation+1.59e-02

Pos logit contributions
bing+1.140
itating+1.121
 wrist+1.117
 cheek+1.111
 mant+1.081
Neg logit contributions
 seen-0.757
iles-0.734
grown-0.723
 cheese-0.703
vious-0.699
Token He
Token ID679
Feature activation-7.94e-02

Pos logit contributions
bing+0.035
 cheek+0.030
 forth+0.029
 up+0.029
 head+0.029
Neg logit contributions
 seen-0.027
iles-0.026
elly-0.025
vious-0.024
grown-0.024
Token worked
Token ID3111
Feature activation-1.32e-01

Pos logit contributions
 seen+0.145
iles+0.133
vious+0.126
emon+0.125
elly+0.124
Neg logit contributions
bing-0.170
 cheek-0.147
 mant-0.143
itating-0.143
 wrist-0.140
Token hard
Token ID1327
Feature activation+3.95e-02

Pos logit contributions
 seen+0.214
iles+0.201
elly+0.192
grown+0.191
vious+0.187
Neg logit contributions
bing-0.284
 wrist-0.264
 cheek-0.262
 mant-0.258
 head-0.250
Token and
Token ID290
Feature activation+2.84e-02

Pos logit contributions
bing+0.084
 cheek+0.073
 wrist+0.072
 mant+0.072
 head+0.071
Neg logit contributions
 seen-0.069
iles-0.066
elly-0.063
vious-0.061
grown-0.061
Token saw
Token ID2497
Feature activation-1.97e-01

Pos logit contributions
bing+0.344
 up+0.306
 head+0.304
 forth+0.296
 cheek+0.295
Neg logit contributions
iles-0.192
 seen-0.184
elly-0.178
emon-0.177
vious-0.168
Token the
Token ID262
Feature activation-1.01e-02

Pos logit contributions
 seen+0.349
iles+0.348
elly+0.326
emon+0.317
vious+0.316
Neg logit contributions
bing-0.422
 up-0.374
 cheek-0.350
 head-0.347
 edge-0.347
Token gray
Token ID12768
Feature activation-1.03e-01

Pos logit contributions
 seen+0.015
iles+0.014
elly+0.013
vous+0.013
vious+0.013
Neg logit contributions
bing-0.024
 wrist-0.021
itating-0.020
itate-0.020
 cheek-0.020
Token onion
Token ID21670
Feature activation-3.99e-02

Pos logit contributions
 seen+0.174
iles+0.169
elly+0.157
vious+0.152
emon+0.152
Neg logit contributions
bing-0.229
 cheek-0.195
 up-0.193
 wrist-0.191
 mant-0.190
Token bob
Token ID29202
Feature activation+1.24e-01

Pos logit contributions
 seen+0.071
iles+0.067
grown+0.061
emon+0.061
elly+0.061
Neg logit contributions
bing-0.085
 cheek-0.073
 mant-0.072
idents-0.072
itating-0.071
Tokenbing
Token ID4623
Feature activation+5.45e-01

Pos logit contributions
 mant+0.101
 doorstep+0.081
idents+0.080
 mantle+0.077
 wrist+0.074
Neg logit contributions
ah-0.330
iles-0.316
aw-0.313
ile-0.311
 seen-0.310
Token back
Token ID736
Feature activation+2.79e-01

Pos logit contributions
bing+0.976
 wrist+0.829
itating+0.828
 cheek+0.825
idents+0.817
Neg logit contributions
 seen-1.095
iles-1.060
elly-0.994
vious-0.992
grown-0.985
Token and
Token ID290
Feature activation+2.59e-02

Pos logit contributions
bing+0.622
 mant+0.547
idents+0.540
 cheek+0.540
 mantle+0.529
Neg logit contributions
 seen-0.440
iles-0.407
elly-0.401
grown-0.387
 ago-0.365
Token forth
Token ID6071
Feature activation+4.45e-01

Pos logit contributions
bing+0.037
etically+0.032
 mantle+0.032
idents+0.032
 wrist+0.031
Neg logit contributions
 seen-0.061
grown-0.052
iles-0.050
 happened-0.050
 crocod-0.049
Token.
Token ID13
Feature activation+1.15e-02

Pos logit contributions
bing+1.013
 mant+0.916
idents+0.905
 mantle+0.890
itating+0.889
Neg logit contributions
 seen-0.709
grown-0.595
iles-0.590
emon-0.567
 happened-0.562
Token She
Token ID1375
Feature activation+7.26e-02

Pos logit contributions
bing+0.023
 cheek+0.019
 up+0.019
 head+0.019
 forth+0.019
Neg logit contributions
 seen-0.022
iles-0.021
elly-0.020
vious-0.020
emon-0.020
Token all
Token ID477
Feature activation+1.31e-01

Pos logit contributions
bing+0.136
itating+0.131
idents+0.128
etically+0.126
itate+0.124
Neg logit contributions
 seen-0.092
 cheese-0.088
 of-0.086
able-0.077
emon-0.077
Token over
Token ID625
Feature activation+1.60e-01

Pos logit contributions
bing+0.247
 wrist+0.204
 cheek+0.204
idents+0.199
 mant+0.198
Neg logit contributions
 seen-0.273
iles-0.258
elly-0.246
vious-0.242
grown-0.241
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
bing+0.336
 cheek+0.286
 up+0.278
 mant+0.277
 head+0.277
Neg logit contributions
 seen-0.296
iles-0.283
elly-0.265
grown-0.262
emon-0.258
Token After
Token ID2293
Feature activation+1.57e-01

Pos logit contributions
bing+0.022
 cheek+0.018
 up+0.018
 head+0.018
 mant+0.018
Neg logit contributions
 seen-0.021
iles-0.020
elly-0.019
emon-0.019
grown-0.019
Token scrub
Token ID27268
Feature activation+1.64e-01

Pos logit contributions
bing+0.332
 up+0.277
 cheek+0.273
 head+0.263
 mant+0.262
Neg logit contributions
 seen-0.290
iles-0.280
elly-0.270
vious-0.268
vous-0.263
Tokenbing
Token ID4623
Feature activation+5.45e-01

Pos logit contributions
itating+0.172
 mant+0.165
 wrist+0.164
 cheek+0.164
 mantle+0.164
Neg logit contributions
iles-0.394
 seen-0.390
aked-0.376
grown-0.372
emon-0.371
Token for
Token ID329
Feature activation-2.90e-02

Pos logit contributions
bing+1.230
 wrist+1.153
itating+1.138
 cheek+1.133
 mant+1.117
Neg logit contributions
iles-0.782
 seen-0.742
elly-0.695
emon-0.688
vious-0.685
Token a
Token ID257
Feature activation-1.84e-01

Pos logit contributions
 seen+0.052
iles+0.050
elly+0.047
vious+0.046
emon+0.046
Neg logit contributions
bing-0.063
 cheek-0.053
 up-0.053
 head-0.051
 mant-0.051
Token while
Token ID981
Feature activation-1.06e-03

Pos logit contributions
 seen+0.274
iles+0.255
elly+0.249
emon+0.237
grown+0.235
Neg logit contributions
bing-0.442
 up-0.376
 forth-0.370
 wrist-0.369
itating-0.366
Token,
Token ID11
Feature activation-5.70e-02

Pos logit contributions
 seen+0.002
iles+0.002
elly+0.002
emon+0.002
vious+0.002
Neg logit contributions
bing-0.002
 cheek-0.002
 up-0.002
 head-0.002
 mant-0.002
Token the
Token ID262
Feature activation-1.10e-02

Pos logit contributions
 seen+0.089
iles+0.079
grown+0.077
emon+0.074
elly+0.073
Neg logit contributions
bing-0.134
 cheek-0.118
itating-0.115
 mant-0.115
 wrist-0.114
Token the
Token ID262
Feature activation-1.35e-02

Pos logit contributions
bing+0.333
 cheek+0.288
itating+0.283
 mant+0.282
 wrist+0.275
Neg logit contributions
 seen-0.259
iles-0.243
grown-0.229
elly-0.222
vious-0.221
Token bath
Token ID7837
Feature activation-7.70e-02

Pos logit contributions
 seen+0.027
iles+0.024
grown+0.024
elly+0.024
vious+0.023
Neg logit contributions
bing-0.028
itating-0.023
 up-0.022
 forth-0.021
 wrist-0.021
Tokenrobe
Token ID25481
Feature activation+1.48e-01

Pos logit contributions
 seen+0.162
iles+0.155
elly+0.150
vious+0.143
grown+0.143
Neg logit contributions
bing-0.139
 cheek-0.113
 mant-0.113
itating-0.112
 wrist-0.111
Token and
Token ID290
Feature activation+2.35e-02

Pos logit contributions
bing+0.325
idents+0.270
 mant+0.268
 wrist+0.267
 cheek+0.267
Neg logit contributions
 seen-0.259
elly-0.226
vious-0.225
iles-0.221
grown-0.219
Token sob
Token ID27355
Feature activation+1.42e-01

Pos logit contributions
bing+0.049
itating+0.042
 cheek+0.041
 mant+0.041
 mantle+0.040
Neg logit contributions
 seen-0.045
iles-0.039
grown-0.038
emon-0.036
elly-0.036
Tokenbing
Token ID4623
Feature activation+5.45e-01

Pos logit contributions
 mant+0.135
 mantle+0.110
 cheek+0.107
 regularly+0.105
 wrist+0.103
Neg logit contributions
elly-0.373
ud-0.357
iles-0.354
 seen-0.352
grown-0.347
Token.
Token ID13
Feature activation+1.04e-02

Pos logit contributions
bing+1.170
 mant+0.993
 cheek+0.989
 wrist+0.983
itating+0.978
Neg logit contributions
 seen-0.949
iles-0.923
elly-0.862
vious-0.849
grown-0.846
Token Tom
Token ID4186
Feature activation-2.96e-01

Pos logit contributions
bing+0.020
 up+0.017
 cheek+0.017
 wrist+0.017
 head+0.017
Neg logit contributions
 seen-0.020
iles-0.020
vious-0.018
emon-0.018
elly-0.018
Token felt
Token ID2936
Feature activation+1.15e-01

Pos logit contributions
 seen+0.547
iles+0.512
vious+0.474
elly+0.471
grown+0.464
Neg logit contributions
bing-0.633
 cheek-0.532
 mant-0.529
 wrist-0.518
 up-0.517
Token sorry
Token ID7926
Feature activation+1.35e-01

Pos logit contributions
bing+0.231
 cheek+0.189
 wrist+0.189
 mant+0.187
 forth+0.186
Neg logit contributions
 seen-0.230
iles-0.208
vious-0.206
grown-0.199
elly-0.198
Token for
Token ID329
Feature activation-2.63e-02

Pos logit contributions
bing+0.352
 wrist+0.317
 mant+0.311
 wrists+0.298
itating+0.296
Neg logit contributions
 seen-0.178
iles-0.170
elly-0.170
vious-0.158
 of-0.155
Token with
Token ID351
Feature activation+1.74e-01

Pos logit contributions
bing+1.085
itating+1.067
 wrist+1.017
 cheek+1.011
 mant+0.996
Neg logit contributions
 seen-0.835
iles-0.785
 cheese-0.770
vious-0.765
grown-0.762
Token the
Token ID262
Feature activation-1.40e-02

Pos logit contributions
bing+0.361
 cheek+0.303
 up+0.299
 mant+0.294
 head+0.293
Neg logit contributions
 seen-0.326
iles-0.312
elly-0.290
vious-0.287
emon-0.287
Token brush
Token ID14093
Feature activation+6.32e-02

Pos logit contributions
 seen+0.023
iles+0.021
vious+0.020
elly+0.020
emon+0.020
Neg logit contributions
bing-0.032
itating-0.027
 up-0.026
 mant-0.026
etically-0.026
Token and
Token ID290
Feature activation+2.45e-02

Pos logit contributions
bing+0.140
 mant+0.122
idents+0.121
 cheek+0.120
itate+0.119
Neg logit contributions
 seen-0.102
elly-0.087
 of-0.087
vious-0.086
emon-0.083
Token scrub
Token ID27268
Feature activation+1.64e-01

Pos logit contributions
bing+0.053
itating+0.048
 cheek+0.046
idents+0.045
 edge+0.045
Neg logit contributions
 seen-0.044
grown-0.037
iles-0.037
emon-0.036
vious-0.035
Tokenbing
Token ID4623
Feature activation+5.45e-01

Pos logit contributions
itating+0.177
 mant+0.157
 cheek+0.156
 doorstep+0.152
 edge+0.147
Neg logit contributions
 seen-0.403
aked-0.389
ud-0.380
grown-0.374
iles-0.373
Token with
Token ID351
Feature activation+1.74e-01

Pos logit contributions
bing+1.088
itating+1.068
 wrist+1.012
 cheek+1.009
 mant+1.001
Neg logit contributions
 seen-0.832
iles-0.779
 of-0.768
 cheese-0.765
grown-0.762
Token her
Token ID607
Feature activation+9.62e-02

Pos logit contributions
bing+0.368
 cheek+0.310
 up+0.306
 mant+0.301
 head+0.300
Neg logit contributions
 seen-0.319
iles-0.304
elly-0.282
vious-0.280
emon-0.280
Token hands
Token ID2832
Feature activation+1.15e-01

Pos logit contributions
bing+0.188
itating+0.175
itate+0.160
etically+0.159
idents+0.150
Neg logit contributions
 seen-0.184
iles-0.163
 cheese-0.160
grown-0.155
elly-0.153
Token until
Token ID1566
Feature activation+2.30e-01

Pos logit contributions
bing+0.266
 mant+0.233
 cheek+0.228
idents+0.228
 wrist+0.228
Neg logit contributions
 seen-0.185
elly-0.156
grown-0.154
iles-0.154
vious-0.153
Token all
Token ID477
Feature activation+1.33e-01

Pos logit contributions
bing+0.467
 cheek+0.400
itating+0.397
 mant+0.388
 wrist+0.386
Neg logit contributions
 seen-0.449
grown-0.400
iles-0.394
emon-0.388
 happened-0.382
Token ground
Token ID2323
Feature activation+1.92e-01

Pos logit contributions
 seen+0.159
iles+0.156
elly+0.147
emon+0.146
vous+0.145
Neg logit contributions
bing-0.162
 up-0.145
 cheek-0.132
 head-0.132
 forth-0.131
Token that
Token ID326
Feature activation-7.44e-02

Pos logit contributions
bing+0.390
 cheek+0.323
 wrist+0.319
 mant+0.318
itating+0.312
Neg logit contributions
 seen-0.373
iles-0.346
elly-0.325
vious-0.323
grown-0.320
Token he
Token ID339
Feature activation-5.40e-02

Pos logit contributions
 seen+0.143
iles+0.138
elly+0.128
emon+0.127
vious+0.126
Neg logit contributions
bing-0.151
 cheek-0.126
 up-0.124
 wrist-0.122
 mant-0.122
Token had
Token ID550
Feature activation-6.36e-02

Pos logit contributions
 seen+0.101
iles+0.094
vious+0.087
elly+0.087
grown+0.086
Neg logit contributions
bing-0.116
 cheek-0.096
 up-0.094
 mant-0.094
itating-0.093
Token never
Token ID1239
Feature activation+1.40e-01

Pos logit contributions
iles+0.118
elly+0.112
emon+0.106
 seen+0.106
vious+0.104
Neg logit contributions
bing-0.126
 up-0.114
 head-0.111
 cheek-0.108
 mant-0.107
Token seen
Token ID1775
Feature activation-5.54e-01

Pos logit contributions
bing+0.377
 up+0.370
 forth+0.367
 head+0.366
 cheek+0.355
Neg logit contributions
iles-0.154
elly-0.134
ud-0.107
emon-0.104
angs-0.102
Token before
Token ID878
Feature activation+7.64e-02

Pos logit contributions
iles+1.280
 seen+1.267
elly+1.215
emon+1.196
vious+1.174
Neg logit contributions
bing-0.885
 up-0.749
 cheek-0.691
 mant-0.679
 head-0.676
Token.
Token ID13
Feature activation+9.62e-03

Pos logit contributions
bing+0.162
 cheek+0.138
 up+0.136
 mant+0.134
 wrist+0.134
Neg logit contributions
 seen-0.139
iles-0.133
elly-0.124
grown-0.122
vious-0.122
Token He
Token ID679
Feature activation-8.70e-02

Pos logit contributions
bing+0.019
 up+0.016
 cheek+0.016
 wrist+0.016
 mant+0.016
Neg logit contributions
 seen-0.018
iles-0.018
elly-0.016
vious-0.016
emon-0.016
Token bent
Token ID17157
Feature activation-1.70e-02

Pos logit contributions
 seen+0.166
iles+0.155
vious+0.144
emon+0.143
grown+0.143
Neg logit contributions
bing-0.183
 cheek-0.154
 up-0.148
 mant-0.148
itating-0.148
Token down
Token ID866
Feature activation+3.12e-01

Pos logit contributions
 seen+0.044
iles+0.040
grown+0.039
aked+0.038
ah+0.038
Neg logit contributions
bing-0.022
 mant-0.019
 cheek-0.018
 mantle-0.018
idents-0.018
Token giant
Token ID6175
Feature activation-2.38e-01

Pos logit contributions
 seen+0.065
iles+0.063
elly+0.060
emon+0.059
grown+0.059
Neg logit contributions
bing-0.055
 up-0.046
 cheek-0.045
 head-0.044
 wrist-0.043
Token symbol
Token ID6194
Feature activation-4.24e-03

Pos logit contributions
 seen+0.428
iles+0.420
elly+0.402
emon+0.392
vious+0.389
Neg logit contributions
bing-0.484
 up-0.424
 head-0.419
 cheek-0.418
 forth-0.403
Token they
Token ID484
Feature activation+1.45e-01

Pos logit contributions
 seen+0.007
iles+0.007
elly+0.007
emon+0.006
grown+0.006
Neg logit contributions
bing-0.009
 up-0.008
 cheek-0.008
 head-0.008
 forth-0.008
Token had
Token ID550
Feature activation-6.25e-02

Pos logit contributions
bing+0.321
 cheek+0.270
 up+0.262
 wrist+0.261
 mant+0.261
Neg logit contributions
 seen-0.262
iles-0.248
vious-0.227
elly-0.223
emon-0.223
Token never
Token ID1239
Feature activation+1.43e-01

Pos logit contributions
iles+0.106
elly+0.102
 seen+0.098
emon+0.096
vious+0.094
Neg logit contributions
bing-0.132
 up-0.120
 head-0.115
 cheek-0.112
 heads-0.111
Token seen
Token ID1775
Feature activation-5.50e-01

Pos logit contributions
bing+0.367
 up+0.357
 head+0.353
 forth+0.353
 cheek+0.340
Neg logit contributions
iles-0.161
elly-0.152
ud-0.127
emon-0.126
angs-0.118
Token before
Token ID878
Feature activation+7.47e-02

Pos logit contributions
 seen+1.083
iles+1.071
elly+1.023
emon+0.999
vious+0.989
Neg logit contributions
bing-1.056
 up-0.918
 cheek-0.867
 head-0.852
 mant-0.848
Token.
Token ID13
Feature activation+7.47e-03

Pos logit contributions
bing+0.165
 cheek+0.142
 wrist+0.139
 mant+0.138
 up+0.138
Neg logit contributions
 seen-0.130
iles-0.124
elly-0.114
grown-0.113
vious-0.112
Token It
Token ID632
Feature activation-3.34e-02

Pos logit contributions
bing+0.015
 up+0.012
 cheek+0.012
 wrist+0.012
 mant+0.012
Neg logit contributions
 seen-0.015
iles-0.014
elly-0.013
vious-0.013
emon-0.013
Token was
Token ID373
Feature activation+9.91e-02

Pos logit contributions
 seen+0.067
iles+0.063
vious+0.059
vous+0.059
emon+0.058
Neg logit contributions
bing-0.067
 cheek-0.056
 wrist-0.055
 mant-0.054
itating-0.053
Token yellow
Token ID7872
Feature activation-1.00e-01

Pos logit contributions
bing+0.208
 up+0.197
 head+0.182
 mant+0.178
 cheek+0.175
Neg logit contributions
iles-0.171
elly-0.166
 seen-0.163
vious-0.157
emon-0.154
Token beautiful
Token ID4950
Feature activation+2.77e-02

Pos logit contributions
bing+0.455
itating+0.393
 cheek+0.379
 wrist+0.378
 mant+0.371
Neg logit contributions
 seen-0.298
grown-0.243
iles-0.238
emon-0.237
vious-0.235
Token thing
Token ID1517
Feature activation-6.37e-02

Pos logit contributions
bing+0.057
 cheek+0.047
 up+0.047
 mant+0.046
itating+0.046
Neg logit contributions
 seen-0.052
iles-0.050
elly-0.047
emon-0.046
vious-0.046
Token she
Token ID673
Feature activation+1.20e-01

Pos logit contributions
 seen+0.115
iles+0.110
elly+0.102
emon+0.100
vious+0.100
Neg logit contributions
bing-0.137
 cheek-0.115
 mant-0.113
 up-0.112
 wrist-0.112
Token had
Token ID550
Feature activation-6.01e-02

Pos logit contributions
bing+0.287
 cheek+0.243
 mant+0.238
itating+0.238
 up+0.237
Neg logit contributions
 seen-0.199
iles-0.180
vious-0.165
grown-0.164
elly-0.163
Token ever
Token ID1683
Feature activation-1.29e-01

Pos logit contributions
iles+0.097
 seen+0.090
elly+0.090
emon+0.087
vious+0.086
Neg logit contributions
bing-0.134
 up-0.119
 head-0.115
 cheek-0.114
 wrist-0.113
Token seen
Token ID1775
Feature activation-5.50e-01

Pos logit contributions
iles+0.130
elly+0.097
emon+0.091
angs+0.085
aver+0.084
Neg logit contributions
bing-0.361
 up-0.346
 head-0.337
 forth-0.334
 cheek-0.329
Token.
Token ID13
Feature activation+6.08e-03

Pos logit contributions
 seen+1.106
iles+1.063
elly+0.994
vious+0.982
emon+0.981
Neg logit contributions
bing-1.059
 cheek-0.880
 up-0.864
 mant-0.850
 head-0.850
Token
Token ID198
Feature activation+7.51e-03

Pos logit contributions
bing+0.012
 cheek+0.010
 up+0.010
 wrist+0.009
 head+0.009
Neg logit contributions
 seen-0.012
iles-0.012
elly-0.011
emon-0.011
vious-0.011
Token
Token ID198
Feature activation+9.07e-03

Pos logit contributions
bing+0.014
 up+0.011
 cheek+0.011
 wrist+0.011
 edge+0.011
Neg logit contributions
 seen-0.015
iles-0.015
vious-0.014
elly-0.014
emon-0.014
TokenL
Token ID43
Feature activation-1.59e-01

Pos logit contributions
bing+0.017
 cheek+0.014
 up+0.014
 wrist+0.014
itating+0.014
Neg logit contributions
 seen-0.018
iles-0.018
elly-0.017
vious-0.016
emon-0.016
Tokenily
Token ID813
Feature activation-3.49e-01

Pos logit contributions
 seen+0.189
iles+0.174
 cheese+0.154
grown+0.150
 crocod+0.147
Neg logit contributions
bing-0.401
 up-0.364
 cheek-0.358
etically-0.358
 mantle-0.358
Tokenches
Token ID2052
Feature activation-6.82e-02

Pos logit contributions
bing+0.090
 up+0.080
 head+0.076
 cheek+0.076
 edge+0.074
Neg logit contributions
 seen-0.067
iles-0.066
elly-0.061
emon-0.060
grown-0.059
Token.
Token ID13
Feature activation-9.05e-04

Pos logit contributions
 seen+0.140
iles+0.132
vious+0.124
grown+0.122
elly+0.120
Neg logit contributions
bing-0.125
 wrist-0.106
 cheek-0.105
idents-0.104
 mant-0.103
Token He
Token ID679
Feature activation-9.35e-02

Pos logit contributions
 seen+0.002
iles+0.002
elly+0.002
emon+0.002
vious+0.002
Neg logit contributions
bing-0.002
 up-0.001
 cheek-0.001
 wrist-0.001
 head-0.001
Token was
Token ID373
Feature activation+9.44e-02

Pos logit contributions
 seen+0.175
iles+0.166
vious+0.153
elly+0.153
emon+0.152
Neg logit contributions
bing-0.197
 cheek-0.165
 mant-0.160
 up-0.160
 wrist-0.159
Token never
Token ID1239
Feature activation+1.47e-01

Pos logit contributions
bing+0.171
 cheek+0.140
 up+0.136
 wrist+0.135
 mant+0.135
Neg logit contributions
 seen-0.204
iles-0.195
elly-0.182
vious-0.182
emon-0.182
Token seen
Token ID1775
Feature activation-5.49e-01

Pos logit contributions
bing+0.370
 up+0.360
 forth+0.352
 head+0.348
 cheek+0.340
Neg logit contributions
iles-0.194
elly-0.169
emon-0.164
ud-0.143
ile-0.138
Token again
Token ID757
Feature activation+1.94e-01

Pos logit contributions
iles+1.139
 seen+1.135
elly+1.101
emon+1.085
aked+1.042
Neg logit contributions
bing-0.993
 up-0.914
 cheek-0.782
 edge-0.781
 head-0.775
Token.
Token ID13
Feature activation-6.77e-04

Pos logit contributions
bing+0.392
 up+0.327
 cheek+0.324
 head+0.317
 wrist+0.315
Neg logit contributions
 seen-0.370
iles-0.362
elly-0.338
emon-0.335
vious-0.331
Token
Token ID220
Feature activation+2.32e-01

Pos logit contributions
 seen+0.001
iles+0.001
emon+0.001
elly+0.001
vious+0.001
Neg logit contributions
bing-0.001
 up-0.001
 cheek-0.001
 wrist-0.001
 head-0.001
Token
Token ID198
Feature activation+7.27e-03

Pos logit contributions
bing+0.477
 cheek+0.399
 up+0.399
 head+0.388
 mant+0.387
Neg logit contributions
 seen-0.438
iles-0.423
elly-0.394
emon-0.389
vious-0.388
Token
Token ID198
Feature activation+8.29e-03

Pos logit contributions
bing+0.015
 cheek+0.012
 up+0.012
 head+0.012
 mant+0.012
Neg logit contributions
 seen-0.014
iles-0.013
elly-0.012
vious-0.012
emon-0.012
Token pumpkin
Token ID30089
Feature activation+9.24e-03

Pos logit contributions
bing+0.115
 cheek+0.096
 up+0.094
itating+0.094
 mant+0.094
Neg logit contributions
 seen-0.104
iles-0.099
elly-0.093
grown-0.091
emon-0.091
Token that
Token ID326
Feature activation-6.53e-02

Pos logit contributions
bing+0.021
 cheek+0.017
 mant+0.017
itating+0.017
 wrist+0.017
Neg logit contributions
 seen-0.016
iles-0.015
elly-0.014
emon-0.014
grown-0.014
Token had
Token ID550
Feature activation-5.53e-02

Pos logit contributions
 seen+0.125
iles+0.118
grown+0.108
emon+0.108
elly+0.108
Neg logit contributions
bing-0.136
 cheek-0.115
 wrist-0.111
itating-0.111
 mant-0.110
Token ever
Token ID1683
Feature activation-1.23e-01

Pos logit contributions
iles+0.092
elly+0.089
vious+0.084
emon+0.080
 seen+0.078
Neg logit contributions
bing-0.116
 up-0.107
 head-0.101
 mant-0.101
 mantle-0.100
Token been
Token ID587
Feature activation-2.22e-01

Pos logit contributions
iles+0.119
elly+0.104
emon+0.090
angs+0.084
aver+0.082
Neg logit contributions
bing-0.340
 up-0.323
 head-0.307
 mant-0.305
 regularly-0.303
Token seen
Token ID1775
Feature activation-5.49e-01

Pos logit contributions
iles+0.395
 seen+0.366
elly+0.366
vious+0.353
emon+0.349
Neg logit contributions
bing-0.478
 up-0.415
 cheek-0.396
 head-0.393
 mant-0.391
Token.
Token ID13
Feature activation+6.84e-03

Pos logit contributions
 seen+1.108
iles+1.058
elly+0.993
grown+0.982
vious+0.981
Neg logit contributions
bing-1.051
 cheek-0.875
 up-0.853
 wrist-0.846
 head-0.845
Token Everyday
Token ID48154
Feature activation-7.18e-02

Pos logit contributions
bing+0.013
 cheek+0.011
 up+0.011
 wrist+0.011
 mant+0.011
Neg logit contributions
 seen-0.014
iles-0.013
elly-0.012
vious-0.012
emon-0.012
Token,
Token ID11
Feature activation-6.68e-02

Pos logit contributions
 seen+0.129
iles+0.124
elly+0.115
emon+0.114
vious+0.113
Neg logit contributions
bing-0.154
 cheek-0.131
 up-0.129
 mant-0.127
 head-0.127
Token the
Token ID262
Feature activation-1.76e-02

Pos logit contributions
 seen+0.123
iles+0.118
grown+0.108
elly+0.108
emon+0.107
Neg logit contributions
bing-0.139
 cheek-0.119
 mant-0.115
 head-0.114
itating-0.114
Token pumpkin
Token ID30089
Feature activation+1.22e-02

Pos logit contributions
 seen+0.030
iles+0.029
emon+0.027
vious+0.026
elly+0.026
Neg logit contributions
bing-0.040
itating-0.033
 cheek-0.033
 up-0.033
 wrist-0.032
Token happy
Token ID3772
Feature activation+2.13e-01

Pos logit contributions
bing+0.104
 wrist+0.082
 mant+0.081
 cheek+0.081
 forth+0.080
Neg logit contributions
 seen-0.123
iles-0.114
elly-0.109
vious-0.109
grown-0.109
Token -
Token ID532
Feature activation-6.01e-02

Pos logit contributions
bing+0.461
 cheek+0.390
 up+0.388
 head+0.379
 mant+0.379
Neg logit contributions
 seen-0.380
iles-0.366
elly-0.339
emon-0.334
vious-0.334
Token he
Token ID339
Feature activation-4.79e-02

Pos logit contributions
 seen+0.107
iles+0.097
emon+0.092
grown+0.092
vious+0.090
Neg logit contributions
bing-0.133
 cheek-0.109
itating-0.108
 heads-0.108
idents-0.107
Token had
Token ID550
Feature activation-6.04e-02

Pos logit contributions
 seen+0.086
iles+0.078
vious+0.073
emon+0.072
grown+0.072
Neg logit contributions
bing-0.107
 cheek-0.089
itating-0.087
 mant-0.087
idents-0.087
Token never
Token ID1239
Feature activation+1.46e-01

Pos logit contributions
iles+0.104
elly+0.098
 seen+0.095
emon+0.093
vious+0.090
Neg logit contributions
bing-0.128
 up-0.118
 head-0.113
 cheek-0.110
 mant-0.110
Token seen
Token ID1775
Feature activation-5.49e-01

Pos logit contributions
bing+0.311
 forth+0.300
 up+0.298
 head+0.297
 cheek+0.285
Neg logit contributions
iles-0.242
elly-0.215
emon-0.195
angs-0.187
ud-0.187
Token so
Token ID523
Feature activation+6.00e-02

Pos logit contributions
iles+1.043
 seen+1.023
elly+0.993
emon+0.972
vious+0.938
Neg logit contributions
bing-1.089
 up-0.981
 cheek-0.894
 mant-0.888
 head-0.884
Token much
Token ID881
Feature activation-1.39e-01

Pos logit contributions
bing+0.121
 cheek+0.101
 up+0.101
 head+0.098
 mant+0.098
Neg logit contributions
 seen-0.115
iles-0.112
elly-0.104
emon-0.103
vious-0.102
Token honey
Token ID12498
Feature activation-1.06e-01

Pos logit contributions
iles+0.262
 seen+0.260
emon+0.244
elly+0.239
aver+0.233
Neg logit contributions
bing-0.281
 up-0.246
 cheek-0.233
 forth-0.228
 head-0.227
Token before
Token ID878
Feature activation+7.70e-02

Pos logit contributions
 seen+0.215
iles+0.206
elly+0.193
emon+0.192
vious+0.190
Neg logit contributions
bing-0.200
 cheek-0.167
 mant-0.162
 wrist-0.161
 up-0.161
Token.
Token ID13
Feature activation+3.04e-03

Pos logit contributions
bing+0.172
 cheek+0.148
 up+0.145
 wrist+0.144
 mant+0.144
Neg logit contributions
 seen-0.131
iles-0.125
elly-0.115
grown-0.114
vious-0.113
Token He
Token ID679
Feature activation-8.94e-02

Pos logit contributions
bing+0.007
 up+0.006
 cheek+0.006
 wrist+0.006
 mantle+0.006
Neg logit contributions
 seen-0.007
iles-0.007
emon-0.006
vious-0.006
elly-0.006
Token wished
Token ID16555
Feature activation-1.73e-03

Pos logit contributions
 seen+0.170
iles+0.165
elly+0.155
emon+0.152
vious+0.151
Neg logit contributions
bing-0.181
 up-0.152
 cheek-0.152
 head-0.149
 mant-0.147
Token he
Token ID339
Feature activation-4.20e-02

Pos logit contributions
 seen+0.003
iles+0.003
elly+0.003
vious+0.003
emon+0.003
Neg logit contributions
bing-0.004
 cheek-0.003
 up-0.003
 head-0.003
 mant-0.003
Token had
Token ID550
Feature activation-5.22e-02

Pos logit contributions
 seen+0.069
iles+0.063
vious+0.058
grown+0.058
emon+0.057
Neg logit contributions
bing-0.101
 cheek-0.086
 mant-0.084
 wrist-0.084
 up-0.083
Token never
Token ID1239
Feature activation+1.48e-01

Pos logit contributions
iles+0.100
elly+0.097
 seen+0.089
vious+0.088
emon+0.087
Neg logit contributions
bing-0.101
 up-0.093
 head-0.089
 cheek-0.088
 forth-0.087
Token seen
Token ID1775
Feature activation-5.48e-01

Pos logit contributions
 forth+0.299
 up+0.296
bing+0.288
 head+0.277
 cheek+0.265
Neg logit contributions
iles-0.265
elly-0.241
angs-0.218
ud-0.214
ile-0.212
Token the
Token ID262
Feature activation-2.01e-02

Pos logit contributions
iles+1.035
 seen+1.021
elly+1.006
emon+0.972
vious+0.943
Neg logit contributions
bing-1.082
 up-0.995
 mant-0.889
 cheek-0.887
 head-0.883
Token gun
Token ID2485
Feature activation-2.45e-02

Pos logit contributions
 seen+0.038
iles+0.037
elly+0.035
emon+0.034
vious+0.034
Neg logit contributions
bing-0.040
 up-0.035
 cheek-0.035
 head-0.034
 mant-0.034
Token.
Token ID13
Feature activation+1.35e-03

Pos logit contributions
 seen+0.055
iles+0.052
vious+0.050
elly+0.049
emon+0.049
Neg logit contributions
bing-0.042
 cheek-0.033
 wrist-0.033
 mant-0.033
 up-0.032
Token He
Token ID679
Feature activation-9.00e-02

Pos logit contributions
bing+0.003
 cheek+0.002
 up+0.002
 head+0.002
 mant+0.002
Neg logit contributions
 seen-0.003
iles-0.002
emon-0.002
elly-0.002
vious-0.002
Token wished
Token ID16555
Feature activation-4.46e-04

Pos logit contributions
 seen+0.168
iles+0.158
vious+0.147
elly+0.144
grown+0.144
Neg logit contributions
bing-0.194
 cheek-0.163
 up-0.158
 mant-0.158
 wrist-0.157
Token They
Token ID1119
Feature activation+5.00e-02

Pos logit contributions
bing+0.022
 cheek+0.018
 up+0.018
 head+0.018
 wrist+0.018
Neg logit contributions
 seen-0.019
iles-0.019
elly-0.017
emon-0.017
vious-0.017
Token wished
Token ID16555
Feature activation+1.22e-04

Pos logit contributions
bing+0.112
 cheek+0.094
 up+0.092
 mant+0.091
 wrist+0.091
Neg logit contributions
 seen-0.089
iles-0.084
vious-0.077
elly-0.076
grown-0.075
Token they
Token ID484
Feature activation+1.58e-01

Pos logit contributions
bing+0.000
 cheek+0.000
itating+0.000
 mant+0.000
 wrist+0.000
Neg logit contributions
 seen-0.000
iles-0.000
grown-0.000
emon-0.000
elly-0.000
Token had
Token ID550
Feature activation-5.40e-02

Pos logit contributions
bing+0.422
 cheek+0.359
 mant+0.353
 wrist+0.349
itating+0.344
Neg logit contributions
 seen-0.245
iles-0.194
grown-0.191
 happened-0.178
vious-0.177
Token never
Token ID1239
Feature activation+1.48e-01

Pos logit contributions
iles+0.105
elly+0.103
 seen+0.098
vious+0.097
emon+0.095
Neg logit contributions
bing-0.103
 up-0.092
 head-0.089
 cheek-0.089
 wrist-0.087
Token seen
Token ID1775
Feature activation-5.48e-01

Pos logit contributions
 forth+0.331
 up+0.325
 head+0.313
bing+0.302
 cheek+0.291
Neg logit contributions
iles-0.223
elly-0.222
ud-0.194
ocol-0.189
ubby-0.184
Token the
Token ID262
Feature activation-2.09e-02

Pos logit contributions
 seen+1.024
iles+1.016
elly+0.979
emon+0.951
vious+0.940
Neg logit contributions
bing-1.098
 up-0.981
 cheek-0.905
 head-0.895
 mant-0.892
Token truck
Token ID7779
Feature activation+2.54e-03

Pos logit contributions
 seen+0.038
iles+0.037
elly+0.034
emon+0.034
vious+0.034
Neg logit contributions
bing-0.044
 cheek-0.038
 up-0.038
 head-0.037
 mant-0.037
Token and
Token ID290
Feature activation+1.89e-02

Pos logit contributions
bing+0.005
 cheek+0.004
 wrist+0.004
 mant+0.004
 up+0.004
Neg logit contributions
 seen-0.005
iles-0.005
vious-0.005
elly-0.005
emon-0.005
Token the
Token ID262
Feature activation-1.90e-02

Pos logit contributions
bing+0.039
 cheek+0.033
 up+0.033
 head+0.032
 mant+0.032
Neg logit contributions
 seen-0.036
iles-0.034
elly-0.032
emon-0.031
vious-0.031
Token tree
Token ID5509
Feature activation+7.83e-03

Pos logit contributions
 seen+0.035
iles+0.034
elly+0.032
emon+0.031
vious+0.031
Neg logit contributions
bing-0.040
 cheek-0.033
 up-0.033
 head-0.033
 mant-0.032
Token amazed
Token ID24562
Feature activation+1.52e-01

Pos logit contributions
bing+0.136
 cheek+0.110
 up+0.107
 wrist+0.106
 mant+0.106
Neg logit contributions
 seen-0.168
iles-0.160
elly-0.150
vious-0.150
emon-0.150
Token.
Token ID13
Feature activation+3.14e-03

Pos logit contributions
bing+0.303
 up+0.266
 forth+0.245
 cheek+0.243
 head+0.241
Neg logit contributions
 seen-0.293
emon-0.286
iles-0.286
vous-0.265
elly-0.265
Token He
Token ID679
Feature activation-9.30e-02

Pos logit contributions
bing+0.006
 up+0.006
 cheek+0.005
 wrist+0.005
 head+0.005
Neg logit contributions
 seen-0.006
iles-0.006
elly-0.005
emon-0.005
vious-0.005
Token had
Token ID550
Feature activation-5.37e-02

Pos logit contributions
 seen+0.160
iles+0.150
vious+0.136
elly+0.136
emon+0.136
Neg logit contributions
bing-0.211
 cheek-0.180
 up-0.175
 mant-0.174
itating-0.173
Token never
Token ID1239
Feature activation+1.48e-01

Pos logit contributions
iles+0.113
elly+0.112
emon+0.104
 seen+0.102
vious+0.101
Neg logit contributions
bing-0.089
 up-0.081
 head-0.079
 mant-0.075
 mantle-0.075
Token seen
Token ID1775
Feature activation-5.48e-01

Pos logit contributions
bing+0.384
 head+0.377
 up+0.377
 forth+0.374
 wrist+0.362
Neg logit contributions
iles-0.167
elly-0.145
ud-0.123
emon-0.121
angs-0.114
Token anything
Token ID1997
Feature activation-5.42e-02

Pos logit contributions
iles+1.094
elly+1.061
 seen+1.061
emon+1.030
aked+0.991
Neg logit contributions
bing-1.017
 up-0.951
 edge-0.829
 mant-0.828
 head-0.823
Token like
Token ID588
Feature activation-2.25e-01

Pos logit contributions
 seen+0.092
iles+0.088
elly+0.081
vious+0.080
emon+0.080
Neg logit contributions
bing-0.122
 cheek-0.104
 up-0.103
 head-0.102
 mant-0.101
Token it
Token ID340
Feature activation+1.48e-01

Pos logit contributions
 seen+0.517
iles+0.509
elly+0.486
emon+0.478
vious+0.473
Neg logit contributions
bing-0.364
 up-0.303
 cheek-0.286
 mant-0.281
 head-0.279
Token before
Token ID878
Feature activation+7.95e-02

Pos logit contributions
bing+0.289
 cheek+0.240
 wrist+0.233
 mant+0.233
 up+0.232
Neg logit contributions
 seen-0.299
iles-0.285
elly-0.266
vious-0.266
emon-0.263
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
bing+0.169
 cheek+0.144
 up+0.141
 mant+0.140
 wrist+0.140
Neg logit contributions
 seen-0.144
iles-0.138
elly-0.128
grown-0.126
vious-0.125
Token fly
Token ID6129
Feature activation+1.02e-01

Pos logit contributions
emon+0.052
 seen+0.050
iles+0.050
aked+0.049
vous+0.048
Neg logit contributions
bing-0.037
 up-0.034
 forth-0.027
 edge-0.026
 tightly-0.026
Token.
Token ID13
Feature activation+1.27e-02

Pos logit contributions
bing+0.188
 cheek+0.153
 mant+0.150
 wrist+0.149
 up+0.148
Neg logit contributions
 seen-0.216
iles-0.207
elly-0.195
vious-0.194
emon-0.191
Token She
Token ID1375
Feature activation+6.95e-02

Pos logit contributions
bing+0.026
 cheek+0.022
 up+0.022
 head+0.021
 wrist+0.021
Neg logit contributions
 seen-0.024
iles-0.024
elly-0.022
emon-0.022
vious-0.022
Token had
Token ID550
Feature activation-4.61e-02

Pos logit contributions
bing+0.156
 cheek+0.133
 up+0.130
 mant+0.130
 wrist+0.128
Neg logit contributions
 seen-0.120
iles-0.114
vious-0.104
elly-0.104
emon-0.103
Token never
Token ID1239
Feature activation+1.53e-01

Pos logit contributions
iles+0.086
elly+0.082
emon+0.076
 seen+0.076
vious+0.075
Neg logit contributions
bing-0.090
 up-0.084
 head-0.080
 cheek-0.079
 mant-0.078
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
 forth+0.365
 up+0.360
bing+0.356
 head+0.355
 cheek+0.336
Neg logit contributions
iles-0.200
elly-0.186
ud-0.162
ocol-0.160
emon-0.153
Token anything
Token ID1997
Feature activation-5.56e-02

Pos logit contributions
iles+1.026
 seen+1.014
elly+0.978
emon+0.954
vous+0.916
Neg logit contributions
bing-1.098
 up-1.000
 cheek-0.908
 head-0.900
 mant-0.896
Token like
Token ID588
Feature activation-2.23e-01

Pos logit contributions
 seen+0.093
iles+0.089
elly+0.082
vious+0.081
emon+0.081
Neg logit contributions
bing-0.126
 cheek-0.108
 up-0.107
 head-0.105
 mant-0.105
Token it
Token ID340
Feature activation+1.49e-01

Pos logit contributions
 seen+0.501
iles+0.494
elly+0.467
emon+0.460
vious+0.455
Neg logit contributions
bing-0.371
 up-0.309
 cheek-0.295
 head-0.287
 mant-0.287
Token before
Token ID878
Feature activation+8.49e-02

Pos logit contributions
bing+0.282
 cheek+0.232
 up+0.227
 mant+0.225
 wrist+0.225
Neg logit contributions
 seen-0.309
iles-0.297
elly-0.278
vious-0.277
emon-0.275
Token.
Token ID13
Feature activation+1.49e-02

Pos logit contributions
bing+0.186
 cheek+0.161
 mant+0.158
 wrist+0.157
 head+0.155
Neg logit contributions
 seen-0.148
iles-0.138
grown-0.129
elly-0.129
vious-0.126
Token knowing
Token ID6970
Feature activation+7.63e-02

Pos logit contributions
bing+0.145
 cheek+0.121
itating+0.119
 mant+0.119
 up+0.119
Neg logit contributions
 seen-0.122
iles-0.113
elly-0.107
grown-0.105
vious-0.105
Token that
Token ID326
Feature activation-6.55e-02

Pos logit contributions
bing+0.160
 cheek+0.135
 up+0.134
 head+0.131
 mant+0.131
Neg logit contributions
 seen-0.141
iles-0.136
elly-0.126
emon-0.125
vious-0.124
Token the
Token ID262
Feature activation-1.80e-02

Pos logit contributions
 seen+0.131
iles+0.124
emon+0.116
grown+0.115
elly+0.114
Neg logit contributions
bing-0.130
 cheek-0.107
itating-0.105
 wrist-0.104
 mant-0.104
Token angel
Token ID18304
Feature activation+9.41e-02

Pos logit contributions
 seen+0.032
iles+0.031
elly+0.029
vious+0.028
emon+0.028
Neg logit contributions
bing-0.038
 cheek-0.033
 up-0.033
 head-0.032
 mant-0.032
Token had
Token ID550
Feature activation-5.66e-02

Pos logit contributions
bing+0.201
 cheek+0.167
 mant+0.166
 wrist+0.162
 up+0.162
Neg logit contributions
 seen-0.177
iles-0.164
vious-0.154
elly-0.152
emon-0.151
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
iles+0.091
elly+0.086
 seen+0.081
emon+0.081
vious+0.079
Neg logit contributions
bing-0.128
 up-0.116
 head-0.113
 cheek-0.111
 mant-0.109
Token her
Token ID607
Feature activation+9.07e-02

Pos logit contributions
 seen+1.032
iles+0.991
elly+0.920
emon+0.909
vious+0.908
Neg logit contributions
bing-1.125
 cheek-0.944
 up-0.930
 head-0.915
 mant-0.915
Token and
Token ID290
Feature activation+2.28e-02

Pos logit contributions
bing+0.187
itating+0.153
 cheek+0.150
 forth+0.149
idents+0.148
Neg logit contributions
 seen-0.174
iles-0.158
grown-0.147
elly-0.146
vious-0.144
Token acknowledged
Token ID10810
Feature activation-4.08e-02

Pos logit contributions
bing+0.050
itating+0.041
 cheek+0.041
 mant+0.040
 mantle+0.039
Neg logit contributions
 seen-0.044
iles-0.037
grown-0.036
vious-0.036
emon-0.035
Token her
Token ID607
Feature activation+9.06e-02

Pos logit contributions
 seen+0.077
iles+0.073
grown+0.067
elly+0.067
vious+0.066
Neg logit contributions
bing-0.084
 cheek-0.070
itating-0.070
 mant-0.070
 wrist-0.068
Token friendly
Token ID8030
Feature activation-9.65e-02

Pos logit contributions
bing+0.173
itating+0.142
idents+0.139
 forth+0.132
 mantle+0.132
Neg logit contributions
 seen-0.186
iles-0.155
 of-0.152
 cheese-0.151
grown-0.150
Token big
Token ID1263
Feature activation-1.02e-01

Pos logit contributions
bing+0.346
 cheek+0.293
 wrist+0.289
 mant+0.286
itating+0.281
Neg logit contributions
 seen-0.285
iles-0.265
elly-0.253
vious-0.251
grown-0.244
Token and
Token ID290
Feature activation+2.02e-02

Pos logit contributions
 seen+0.202
iles+0.199
emon+0.187
elly+0.187
vious+0.183
Neg logit contributions
bing-0.190
 head-0.168
 up-0.167
 cheek-0.166
 forth-0.161
Token she
Token ID673
Feature activation+1.32e-01

Pos logit contributions
bing+0.043
 cheek+0.036
 up+0.036
 mant+0.035
 wrist+0.035
Neg logit contributions
 seen-0.037
iles-0.035
elly-0.032
vious-0.032
emon-0.032
Token had
Token ID550
Feature activation-5.47e-02

Pos logit contributions
bing+0.302
 cheek+0.251
 mant+0.245
itating+0.245
 up+0.242
Neg logit contributions
 seen-0.231
iles-0.213
vious-0.197
emon-0.193
grown-0.193
Token never
Token ID1239
Feature activation+1.48e-01

Pos logit contributions
iles+0.092
elly+0.086
 seen+0.083
vious+0.081
emon+0.080
Neg logit contributions
bing-0.116
 up-0.107
 head-0.103
 cheek-0.102
 wrist-0.102
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
bing+0.381
 up+0.379
 head+0.373
 forth+0.372
 cheek+0.358
Neg logit contributions
iles-0.173
elly-0.152
ud-0.125
emon-0.124
angs-0.120
Token one
Token ID530
Feature activation-6.28e-02

Pos logit contributions
 seen+1.080
iles+1.063
elly+0.998
emon+0.983
vious+0.973
Neg logit contributions
bing-1.062
 up-0.902
 cheek-0.877
 head-0.855
 mant-0.854
Token before
Token ID878
Feature activation+8.29e-02

Pos logit contributions
 seen+0.134
iles+0.125
elly+0.119
vious+0.118
grown+0.117
Neg logit contributions
bing-0.117
 cheek-0.095
 mant-0.093
 heads-0.092
idents-0.092
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
bing+0.184
 cheek+0.160
 mant+0.156
 wrist+0.155
 head+0.155
Neg logit contributions
 seen-0.144
iles-0.133
grown-0.127
elly-0.126
emon-0.120
Token
Token ID220
Feature activation+2.38e-01

Pos logit contributions
bing+0.023
 cheek+0.019
 up+0.019
 head+0.019
 mant+0.019
Neg logit contributions
 seen-0.024
iles-0.023
elly-0.022
emon-0.021
vious-0.021
Token
Token ID198
Feature activation+1.07e-02

Pos logit contributions
bing+0.527
 cheek+0.448
 forth+0.442
 mant+0.438
 head+0.437
Neg logit contributions
 seen-0.414
iles-0.395
elly-0.372
emon-0.363
grown-0.362
Token rainbow
Token ID27223
Feature activation-5.19e-02

Pos logit contributions
 seen+0.107
iles+0.097
grown+0.092
emon+0.092
elly+0.090
Neg logit contributions
bing-0.113
 cheek-0.094
 mant-0.093
itating-0.092
 wrist-0.092
Token.
Token ID13
Feature activation+1.38e-02

Pos logit contributions
 seen+0.100
iles+0.096
elly+0.089
vious+0.088
emon+0.088
Neg logit contributions
bing-0.105
 cheek-0.088
 mant-0.086
 up-0.086
 wrist-0.085
Token He
Token ID679
Feature activation-8.51e-02

Pos logit contributions
bing+0.028
 cheek+0.023
 up+0.023
 head+0.023
 mant+0.023
Neg logit contributions
 seen-0.026
iles-0.025
elly-0.024
emon-0.023
vious-0.023
Token had
Token ID550
Feature activation-5.04e-02

Pos logit contributions
 seen+0.153
iles+0.145
vious+0.133
elly+0.132
emon+0.132
Neg logit contributions
bing-0.186
 cheek-0.157
 mant-0.152
 up-0.152
 wrist-0.151
Token never
Token ID1239
Feature activation+1.52e-01

Pos logit contributions
iles+0.098
elly+0.096
emon+0.089
 seen+0.086
vious+0.085
Neg logit contributions
bing-0.090
 up-0.086
 head-0.083
 mant-0.080
 heads-0.080
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
 forth+0.369
 up+0.362
 head+0.357
bing+0.355
 cheek+0.331
Neg logit contributions
iles-0.194
elly-0.179
ud-0.157
emon-0.144
ocol-0.142
Token anything
Token ID1997
Feature activation-5.32e-02

Pos logit contributions
iles+1.083
 seen+1.069
elly+1.044
emon+1.015
vous+0.981
Neg logit contributions
bing-1.032
 up-0.941
 cheek-0.841
 head-0.838
 mant-0.837
Token so
Token ID523
Feature activation+6.32e-02

Pos logit contributions
 seen+0.087
iles+0.085
elly+0.078
emon+0.077
vious+0.076
Neg logit contributions
bing-0.122
 up-0.105
 cheek-0.104
 head-0.102
 mant-0.102
Token pretty
Token ID2495
Feature activation+3.70e-02

Pos logit contributions
bing+0.128
 up+0.111
 cheek+0.108
 head+0.105
 mant+0.104
Neg logit contributions
 seen-0.118
iles-0.117
emon-0.108
elly-0.107
vious-0.104
Token.
Token ID13
Feature activation+6.16e-03

Pos logit contributions
bing+0.072
 cheek+0.059
 up+0.059
 head+0.057
 mant+0.057
Neg logit contributions
 seen-0.074
iles-0.072
elly-0.067
emon-0.066
vious-0.066
Token
Token ID220
Feature activation+2.39e-01

Pos logit contributions
bing+0.011
 up+0.010
 cheek+0.009
 wrist+0.009
 mantle+0.009
Neg logit contributions
 seen-0.013
iles-0.012
emon-0.011
elly-0.011
vious-0.011
Token He
Token ID679
Feature activation-9.21e-02

Pos logit contributions
bing+0.006
 up+0.005
 wrist+0.005
 edge+0.005
 head+0.005
Neg logit contributions
 seen-0.007
iles-0.007
vious-0.006
emon-0.006
vous-0.006
Token wished
Token ID16555
Feature activation+2.56e-03

Pos logit contributions
 seen+0.160
iles+0.160
elly+0.150
emon+0.148
grown+0.144
Neg logit contributions
bing-0.194
 up-0.170
 head-0.169
 cheek-0.168
 heads-0.163
Token he
Token ID339
Feature activation-3.95e-02

Pos logit contributions
bing+0.005
 up+0.005
 cheek+0.005
 head+0.004
 wrist+0.004
Neg logit contributions
 seen-0.005
iles-0.005
elly-0.004
vious-0.004
emon-0.004
Token had
Token ID550
Feature activation-5.23e-02

Pos logit contributions
 seen+0.065
iles+0.060
grown+0.055
elly+0.054
vious+0.054
Neg logit contributions
bing-0.094
 cheek-0.080
 mant-0.079
 up-0.078
 wrist-0.078
Token never
Token ID1239
Feature activation+1.53e-01

Pos logit contributions
iles+0.112
elly+0.110
vious+0.103
 seen+0.102
emon+0.101
Neg logit contributions
bing-0.090
 up-0.081
 head-0.079
 cheek-0.078
 wrist-0.076
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
 up+0.310
 forth+0.307
 head+0.293
bing+0.285
 cheek+0.271
Neg logit contributions
iles-0.267
elly-0.259
ud-0.235
ocol-0.233
ubby-0.226
Token the
Token ID262
Feature activation-2.21e-02

Pos logit contributions
iles+1.061
elly+1.047
 seen+1.034
emon+1.007
vious+0.974
Neg logit contributions
bing-1.039
 up-0.989
 head-0.855
 edge-0.851
 mant-0.849
Token dog
Token ID3290
Feature activation-1.50e-01

Pos logit contributions
 seen+0.042
iles+0.040
elly+0.038
emon+0.037
vious+0.037
Neg logit contributions
bing-0.045
 up-0.039
 cheek-0.039
 head-0.038
 mant-0.037
Token.
Token ID13
Feature activation+1.38e-03

Pos logit contributions
 seen+0.307
iles+0.282
vious+0.272
elly+0.265
emon+0.264
Neg logit contributions
bing-0.287
 mant-0.233
 cheek-0.232
 heads-0.231
idents-0.230
Token He
Token ID679
Feature activation-8.95e-02

Pos logit contributions
bing+0.002
 up+0.002
 head+0.002
 cheek+0.002
 wrist+0.002
Neg logit contributions
iles-0.003
 seen-0.003
emon-0.003
vious-0.003
elly-0.003
Token wished
Token ID16555
Feature activation+3.09e-03

Pos logit contributions
 seen+0.173
iles+0.164
elly+0.152
vious+0.152
emon+0.150
Neg logit contributions
bing-0.185
 cheek-0.154
 up-0.152
 mant-0.150
 wrist-0.149
Token wonderful
Token ID7932
Feature activation-2.22e-02

Pos logit contributions
bing+0.023
 cheek+0.019
 wrist+0.019
 mant+0.018
itating+0.018
Neg logit contributions
 seen-0.021
iles-0.019
grown-0.018
vious-0.018
elly-0.018
Token animals
Token ID4695
Feature activation-5.45e-02

Pos logit contributions
 seen+0.042
iles+0.041
elly+0.038
emon+0.038
vious+0.038
Neg logit contributions
bing-0.045
 cheek-0.038
 up-0.038
 head-0.037
 mant-0.037
Token he
Token ID339
Feature activation-4.85e-02

Pos logit contributions
 seen+0.101
iles+0.096
elly+0.089
emon+0.088
grown+0.088
Neg logit contributions
bing-0.114
 cheek-0.097
 up-0.095
 mant-0.094
 wrist-0.094
Token had
Token ID550
Feature activation-5.72e-02

Pos logit contributions
 seen+0.079
iles+0.077
elly+0.071
emon+0.070
vious+0.070
Neg logit contributions
bing-0.111
 cheek-0.095
 up-0.095
 head-0.093
 mant-0.092
Token never
Token ID1239
Feature activation+1.46e-01

Pos logit contributions
iles+0.089
elly+0.085
 seen+0.080
vious+0.080
emon+0.080
Neg logit contributions
bing-0.130
 up-0.116
 head-0.113
 cheek-0.110
 mant-0.110
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
bing+0.434
 up+0.418
 head+0.416
 forth+0.405
 cheek+0.402
Neg logit contributions
iles-0.111
elly-0.098
emon-0.078
ud-0.072
ha-0.063
Token before
Token ID878
Feature activation+7.35e-02

Pos logit contributions
 seen+1.272
iles+1.268
elly+1.204
emon+1.187
vious+1.175
Neg logit contributions
bing-0.860
 up-0.709
 cheek-0.671
 head-0.650
 mant-0.650
Token.
Token ID13
Feature activation+2.57e-03

Pos logit contributions
bing+0.156
 cheek+0.133
 up+0.130
 wrist+0.129
 mant+0.129
Neg logit contributions
 seen-0.135
iles-0.128
elly-0.120
grown-0.118
vious-0.117
Token The
Token ID383
Feature activation-4.35e-02

Pos logit contributions
bing+0.005
 up+0.004
 cheek+0.004
 mantle+0.004
 wrist+0.004
Neg logit contributions
 seen-0.005
iles-0.005
elly-0.004
vious-0.004
emon-0.004
Token boy
Token ID2933
Feature activation-2.01e-02

Pos logit contributions
 seen+0.080
iles+0.077
elly+0.071
emon+0.070
vious+0.070
Neg logit contributions
bing-0.092
 cheek-0.077
 up-0.077
 wrist-0.075
 mant-0.075
Token was
Token ID373
Feature activation+9.49e-02

Pos logit contributions
 seen+0.039
iles+0.036
grown+0.033
vious+0.033
emon+0.033
Neg logit contributions
bing-0.042
 cheek-0.035
idents-0.035
 mant-0.034
 wrist-0.034
Token somewhere
Token ID7382
Feature activation-6.20e-02

Pos logit contributions
bing+0.203
 cheek+0.168
 mant+0.167
 wrist+0.164
 mantle+0.163
Neg logit contributions
 seen-0.202
iles-0.185
grown-0.173
elly-0.173
vious-0.172
Token special
Token ID2041
Feature activation+7.50e-02

Pos logit contributions
 seen+0.116
iles+0.107
elly+0.100
grown+0.099
vious+0.098
Neg logit contributions
bing-0.133
 cheek-0.111
 wrist-0.109
 mant-0.108
 head-0.108
Token.
Token ID13
Feature activation+5.83e-03

Pos logit contributions
bing+0.153
 up+0.129
 cheek+0.129
 head+0.125
 wrist+0.125
Neg logit contributions
 seen-0.142
iles-0.138
elly-0.129
emon-0.127
vious-0.127
Token He
Token ID679
Feature activation-8.94e-02

Pos logit contributions
bing+0.011
 cheek+0.010
 up+0.010
 head+0.009
 wrist+0.009
Neg logit contributions
 seen-0.011
iles-0.011
elly-0.010
emon-0.010
vious-0.010
Token had
Token ID550
Feature activation-5.28e-02

Pos logit contributions
 seen+0.158
iles+0.156
elly+0.147
emon+0.143
vious+0.142
Neg logit contributions
bing-0.189
 up-0.163
 head-0.160
 cheek-0.160
 wrist-0.156
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
iles+0.099
elly+0.098
emon+0.088
vious+0.087
ile+0.085
Neg logit contributions
bing-0.099
 up-0.096
 head-0.090
 edge-0.088
 regularly-0.087
Token animals
Token ID4695
Feature activation-5.23e-02

Pos logit contributions
iles+0.948
 seen+0.942
elly+0.939
emon+0.903
vious+0.885
Neg logit contributions
bing-1.158
 up-1.062
 head-0.962
 mant-0.961
 edge-0.952
Token that
Token ID326
Feature activation-6.25e-02

Pos logit contributions
 seen+0.095
iles+0.090
elly+0.083
grown+0.082
vious+0.082
Neg logit contributions
bing-0.112
 cheek-0.096
 mant-0.093
 wrist-0.093
 up-0.093
Token could
Token ID714
Feature activation-2.86e-02

Pos logit contributions
 seen+0.111
iles+0.108
elly+0.101
emon+0.099
vious+0.099
Neg logit contributions
bing-0.134
 up-0.113
 cheek-0.113
 head-0.111
 mant-0.110
Token reach
Token ID3151
Feature activation+6.62e-02

Pos logit contributions
 seen+0.060
iles+0.058
emon+0.053
vious+0.052
elly+0.052
Neg logit contributions
bing-0.054
 cheek-0.044
 mant-0.043
 up-0.043
 forth-0.043
Token the
Token ID262
Feature activation-2.32e-02

Pos logit contributions
bing+0.127
 up+0.109
 cheek+0.103
 head+0.101
 mant+0.100
Neg logit contributions
 seen-0.132
iles-0.127
elly-0.123
vious-0.122
emon-0.121
Token down
Token ID866
Feature activation+3.08e-01

Pos logit contributions
bing+0.016
 cheek+0.014
 wrist+0.013
itating+0.013
 mant+0.013
Neg logit contributions
 seen-0.013
iles-0.012
grown-0.011
vious-0.011
emon-0.011
Token streets
Token ID6483
Feature activation+1.75e-01

Pos logit contributions
bing+0.594
 up+0.527
 head+0.493
 edge+0.488
 cheek+0.486
Neg logit contributions
iles-0.587
 seen-0.587
emon-0.555
elly-0.555
vious-0.541
Token he
Token ID339
Feature activation-4.91e-02

Pos logit contributions
bing+0.375
 cheek+0.320
 mant+0.313
 wrist+0.311
idents+0.310
Neg logit contributions
 seen-0.321
iles-0.294
grown-0.278
elly-0.274
vious-0.274
Token had
Token ID550
Feature activation-5.66e-02

Pos logit contributions
 seen+0.075
iles+0.071
elly+0.065
emon+0.064
vious+0.064
Neg logit contributions
bing-0.119
 cheek-0.103
 up-0.102
 mant-0.100
 head-0.100
Token never
Token ID1239
Feature activation+1.46e-01

Pos logit contributions
iles+0.101
elly+0.096
emon+0.092
 seen+0.091
vious+0.089
Neg logit contributions
bing-0.116
 up-0.106
 head-0.101
 cheek-0.098
 mant-0.098
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
bing+0.424
 up+0.410
 head+0.404
 forth+0.402
 cheek+0.390
Neg logit contributions
iles-0.132
elly-0.118
emon-0.097
ud-0.085
angs-0.082
Token before
Token ID878
Feature activation+7.78e-02

Pos logit contributions
iles+1.281
 seen+1.273
elly+1.213
emon+1.199
vious+1.172
Neg logit contributions
bing-0.851
 up-0.715
 cheek-0.662
 head-0.651
 mant-0.645
Token.
Token ID13
Feature activation+1.85e-03

Pos logit contributions
bing+0.162
 cheek+0.136
 up+0.135
 mant+0.132
 head+0.132
Neg logit contributions
 seen-0.145
iles-0.139
elly-0.130
vious-0.128
emon-0.128
Token
Token ID198
Feature activation+3.44e-03

Pos logit contributions
bing+0.004
 up+0.003
 mantle+0.003
 cheek+0.003
 wrist+0.003
Neg logit contributions
 seen-0.003
iles-0.003
emon-0.003
elly-0.003
vious-0.003
Token
Token ID198
Feature activation+4.73e-03

Pos logit contributions
bing+0.006
 up+0.005
 edge+0.005
itating+0.005
 wrist+0.005
Neg logit contributions
iles-0.007
 seen-0.007
vious-0.007
elly-0.007
emon-0.007
TokenFinally
Token ID11158
Feature activation+2.00e-01

Pos logit contributions
bing+0.009
 up+0.008
itating+0.008
 edge+0.008
 mantle+0.008
Neg logit contributions
 seen-0.009
iles-0.009
elly-0.008
vious-0.008
emon-0.008
Token discover
Token ID7073
Feature activation+9.27e-02

Pos logit contributions
 seen+0.220
iles+0.185
vious+0.180
vous+0.177
grown+0.177
Neg logit contributions
bing-0.217
itating-0.179
idents-0.173
 cheek-0.173
rier-0.164
Token toys
Token ID14958
Feature activation-5.18e-03

Pos logit contributions
bing+0.201
 cheek+0.173
 head+0.167
 mant+0.166
 wrist+0.166
Neg logit contributions
 seen-0.165
iles-0.158
grown-0.146
elly-0.145
vious-0.143
Token she
Token ID673
Feature activation+1.28e-01

Pos logit contributions
 seen+0.010
iles+0.009
elly+0.009
emon+0.009
vious+0.009
Neg logit contributions
bing-0.011
 up-0.009
 cheek-0.009
 head-0.009
 wrist-0.009
Token had
Token ID550
Feature activation-5.42e-02

Pos logit contributions
bing+0.290
 up+0.249
 cheek+0.247
 head+0.244
 wrist+0.241
Neg logit contributions
 seen-0.208
iles-0.205
elly-0.190
emon-0.187
vious-0.185
Token not
Token ID407
Feature activation+2.23e-01

Pos logit contributions
iles+0.092
elly+0.088
 seen+0.085
vious+0.084
emon+0.083
Neg logit contributions
bing-0.116
 up-0.103
 head-0.099
 mant-0.098
 mantle-0.097
Token seen
Token ID1775
Feature activation-5.47e-01

Pos logit contributions
bing+0.537
 up+0.502
 head+0.497
 cheek+0.484
 regularly+0.477
Neg logit contributions
iles-0.315
elly-0.290
emon-0.263
vious-0.259
ile-0.242
Token in
Token ID287
Feature activation+5.82e-02

Pos logit contributions
 seen+1.174
iles+1.152
elly+1.083
emon+1.070
vious+1.066
Neg logit contributions
bing-0.971
 up-0.801
 cheek-0.786
 head-0.766
 mant-0.762
Token a
Token ID257
Feature activation-1.92e-01

Pos logit contributions
bing+0.120
 up+0.105
 head+0.100
 edge+0.099
 wrist+0.099
Neg logit contributions
iles-0.107
 seen-0.105
emon-0.099
elly-0.099
vious-0.098
Token long
Token ID890
Feature activation-1.03e-01

Pos logit contributions
 seen+0.374
iles+0.350
elly+0.335
grown+0.334
vious+0.327
Neg logit contributions
bing-0.393
 cheek-0.319
 up-0.318
 wrist-0.310
 forth-0.309
Token time
Token ID640
Feature activation-3.15e-01

Pos logit contributions
iles+0.200
 seen+0.187
emon+0.183
vious+0.181
elly+0.178
Neg logit contributions
bing-0.208
 head-0.192
 up-0.180
 edge-0.180
 wrist-0.178
Token.
Token ID13
Feature activation+5.06e-03

Pos logit contributions
 seen+0.582
iles+0.546
elly+0.510
grown+0.508
emon+0.502
Neg logit contributions
bing-0.665
 cheek-0.565
 up-0.550
 mant-0.548
 wrist-0.546
Token fire
Token ID2046
Feature activation-5.33e-02

Pos logit contributions
bing+0.118
 cheek+0.098
 up+0.097
 mant+0.096
itating+0.095
Neg logit contributions
 seen-0.104
iles-0.100
elly-0.093
emon-0.092
vious-0.091
Tokenwork
Token ID1818
Feature activation+9.98e-02

Pos logit contributions
 seen+0.102
iles+0.098
elly+0.091
emon+0.089
grown+0.089
Neg logit contributions
bing-0.107
 wrist-0.089
 cheek-0.089
 mant-0.088
itating-0.088
Token they
Token ID484
Feature activation+1.56e-01

Pos logit contributions
bing+0.223
 cheek+0.189
 wrist+0.187
 mant+0.185
 up+0.184
Neg logit contributions
 seen-0.171
iles-0.162
elly-0.149
emon-0.147
vious-0.147
Token had
Token ID550
Feature activation-5.16e-02

Pos logit contributions
bing+0.370
 cheek+0.317
 up+0.313
 wrist+0.309
 mant+0.308
Neg logit contributions
 seen-0.255
iles-0.238
grown-0.214
vious-0.214
elly-0.214
Token ever
Token ID1683
Feature activation-1.23e-01

Pos logit contributions
iles+0.081
elly+0.077
 seen+0.076
vious+0.075
emon+0.074
Neg logit contributions
bing-0.117
 up-0.103
 head-0.099
 cheek-0.098
 mant-0.097
Token seen
Token ID1775
Feature activation-5.46e-01

Pos logit contributions
iles+0.094
elly+0.087
emon+0.068
angs+0.060
aver+0.059
Neg logit contributions
bing-0.361
 up-0.338
 forth-0.337
 head-0.333
 heads-0.330
Token.
Token ID13
Feature activation+6.03e-03

Pos logit contributions
 seen+1.014
iles+0.969
elly+0.896
grown+0.888
emon+0.884
Neg logit contributions
bing-1.137
 cheek-0.963
 up-0.937
 mant-0.934
 wrist-0.933
Token Tim
Token ID5045
Feature activation-1.55e-01

Pos logit contributions
bing+0.012
 up+0.010
 cheek+0.010
 wrist+0.009
 mant+0.009
Neg logit contributions
 seen-0.012
iles-0.012
elly-0.011
vious-0.011
emon-0.011
Token and
Token ID290
Feature activation+1.84e-02

Pos logit contributions
 seen+0.266
iles+0.252
elly+0.232
vious+0.231
grown+0.228
Neg logit contributions
bing-0.346
 cheek-0.296
 up-0.290
 mant-0.290
 wrist-0.288
Token Sam
Token ID3409
Feature activation-3.15e-01

Pos logit contributions
bing+0.048
 cheek+0.042
 up+0.042
 head+0.041
 wrist+0.041
Neg logit contributions
 seen-0.025
iles-0.024
elly-0.022
vious-0.021
emon-0.021
Token were
Token ID547
Feature activation-4.75e-02

Pos logit contributions
 seen+0.523
iles+0.469
vious+0.444
grown+0.424
vous+0.419
Neg logit contributions
bing-0.745
 cheek-0.639
 mant-0.620
 wrist-0.617
 up-0.616
Token unusual
Token ID8468
Feature activation-2.00e-02

Pos logit contributions
bing+0.186
 up+0.164
 head+0.156
 cheek+0.155
 mant+0.154
Neg logit contributions
iles-0.175
 seen-0.172
elly-0.165
vious-0.161
emon-0.158
Token and
Token ID290
Feature activation+1.50e-02

Pos logit contributions
 seen+0.037
iles+0.036
elly+0.033
emon+0.033
vious+0.033
Neg logit contributions
bing-0.042
 up-0.036
 cheek-0.035
 head-0.034
 mant-0.034
Token she
Token ID673
Feature activation+1.30e-01

Pos logit contributions
bing+0.033
itating+0.028
 edge+0.028
 cheek+0.027
 mant+0.027
Neg logit contributions
 seen-0.027
iles-0.023
grown-0.023
vous-0.022
vious-0.021
Token had
Token ID550
Feature activation-5.16e-02

Pos logit contributions
bing+0.310
 cheek+0.260
itating+0.257
 mant+0.256
 up+0.254
Neg logit contributions
 seen-0.220
iles-0.193
vious-0.181
vous-0.180
elly-0.178
Token never
Token ID1239
Feature activation+1.50e-01

Pos logit contributions
iles+0.113
 seen+0.108
elly+0.107
emon+0.103
vious+0.103
Neg logit contributions
bing-0.087
 up-0.074
 head-0.072
 cheek-0.071
 mant-0.071
Token seen
Token ID1775
Feature activation-5.46e-01

Pos logit contributions
 head+0.423
bing+0.422
 up+0.421
 forth+0.416
 cheek+0.409
Neg logit contributions
iles-0.147
elly-0.117
ud-0.095
emon-0.090
ile-0.088
Token anything
Token ID1997
Feature activation-5.71e-02

Pos logit contributions
 seen+1.163
iles+1.151
elly+1.082
emon+1.065
vious+1.056
Neg logit contributions
bing-0.977
 up-0.820
 cheek-0.791
 mant-0.771
 head-0.769
Token like
Token ID588
Feature activation-2.26e-01

Pos logit contributions
 seen+0.095
iles+0.089
elly+0.084
vious+0.083
emon+0.082
Neg logit contributions
bing-0.131
 cheek-0.112
 up-0.110
 head-0.109
 wrist-0.109
Token it
Token ID340
Feature activation+1.45e-01

Pos logit contributions
 seen+0.520
iles+0.506
elly+0.478
emon+0.472
vious+0.472
Neg logit contributions
bing-0.369
 cheek-0.294
 up-0.293
 head-0.283
 mant-0.282
Token before
Token ID878
Feature activation+7.80e-02

Pos logit contributions
bing+0.264
 cheek+0.216
 up+0.211
 mant+0.208
 wrist+0.208
Neg logit contributions
 seen-0.311
iles-0.298
elly-0.281
vious-0.279
emon-0.278
Token.
Token ID13
Feature activation+9.77e-03

Pos logit contributions
bing+0.176
 cheek+0.155
 mant+0.152
itating+0.150
 wrist+0.150
Neg logit contributions
 seen-0.130
iles-0.118
grown-0.115
elly-0.112
emon-0.107
Token.
Token ID13
Feature activation+9.20e-03

Pos logit contributions
bing+0.036
 cheek+0.031
 wrist+0.031
 mant+0.031
 up+0.031
Neg logit contributions
 seen-0.027
iles-0.026
elly-0.024
vious-0.023
grown-0.023
Token He
Token ID679
Feature activation-8.47e-02

Pos logit contributions
bing+0.018
 up+0.015
 cheek+0.015
 wrist+0.015
 mant+0.015
Neg logit contributions
 seen-0.018
iles-0.017
vious-0.016
elly-0.016
emon-0.016
Token brave
Token ID14802
Feature activation+4.65e-02

Pos logit contributions
 seen+0.159
iles+0.150
emon+0.138
vious+0.138
elly+0.138
Neg logit contributions
bing-0.178
 cheek-0.150
 up-0.146
 mant-0.146
 wrist-0.144
Tokenly
Token ID306
Feature activation-8.13e-02

Pos logit contributions
bing+0.077
 up+0.067
 head+0.067
 heads+0.067
 forth+0.065
Neg logit contributions
emon-0.103
iles-0.099
 seen-0.098
elly-0.098
vous-0.094
Token marched
Token ID23558
Feature activation+1.38e-02

Pos logit contributions
iles+0.144
 seen+0.141
elly+0.136
vious+0.130
emon+0.128
Neg logit contributions
bing-0.175
 up-0.150
 head-0.146
 cheek-0.144
 forth-0.143
Token down
Token ID866
Feature activation+3.11e-01

Pos logit contributions
bing+0.022
 cheek+0.019
 mant+0.019
 wrist+0.019
idents+0.019
Neg logit contributions
 seen-0.031
iles-0.028
grown-0.027
vious-0.027
aver-0.026
Token to
Token ID284
Feature activation-1.04e-01

Pos logit contributions
bing+0.608
 up+0.553
 edge+0.503
 head+0.499
 regularly+0.493
Neg logit contributions
iles-0.601
 seen-0.585
elly-0.567
emon-0.552
vious-0.548
Token the
Token ID262
Feature activation-1.58e-02

Pos logit contributions
 seen+0.187
iles+0.170
grown+0.157
emon+0.154
vious+0.153
Neg logit contributions
bing-0.234
 cheek-0.197
itating-0.193
idents-0.191
 mant-0.188
Token beach
Token ID10481
Feature activation+1.09e-01

Pos logit contributions
 seen+0.030
iles+0.028
grown+0.026
elly+0.026
vious+0.026
Neg logit contributions
bing-0.033
itating-0.026
idents-0.026
 cheek-0.026
 wrist-0.025
Token expecting
Token ID12451
Feature activation-4.46e-02

Pos logit contributions
bing+0.246
 cheek+0.213
 mant+0.213
itating+0.209
idents+0.207
Neg logit contributions
 seen-0.185
iles-0.162
grown-0.152
elly-0.147
vious-0.144
Token to
Token ID284
Feature activation-1.02e-01

Pos logit contributions
 seen+0.079
iles+0.079
elly+0.074
emon+0.072
vious+0.072
Neg logit contributions
bing-0.094
 up-0.082
 cheek-0.080
 head-0.079
 mant-0.077
Token
Token ID198
Feature activation+6.70e-03

Pos logit contributions
bing+0.532
 cheek+0.455
 forth+0.444
 head+0.442
 wrist+0.440
Neg logit contributions
 seen-0.421
iles-0.401
elly-0.374
emon-0.367
grown-0.365
Token
Token ID198
Feature activation+7.41e-03

Pos logit contributions
bing+0.015
 cheek+0.013
 head+0.012
 forth+0.012
 up+0.012
Neg logit contributions
 seen-0.012
iles-0.011
elly-0.010
grown-0.010
emon-0.010
TokenJohn
Token ID7554
Feature activation-1.28e-01

Pos logit contributions
bing+0.019
 cheek+0.016
 up+0.016
 head+0.016
 mant+0.016
Neg logit contributions
 seen-0.011
iles-0.010
elly-0.009
emon-0.009
vious-0.009
Token stuck
Token ID7819
Feature activation-8.76e-02

Pos logit contributions
 seen+0.254
iles+0.243
elly+0.227
vious+0.225
emon+0.224
Neg logit contributions
bing-0.252
 cheek-0.211
 up-0.206
 mant-0.204
 wrist-0.204
Token his
Token ID465
Feature activation-8.37e-02

Pos logit contributions
 seen+0.165
iles+0.150
grown+0.143
emon+0.138
vious+0.135
Neg logit contributions
bing-0.180
 cheek-0.155
 wrist-0.151
idents-0.151
itating-0.151
Token head
Token ID1182
Feature activation+4.36e-01

Pos logit contributions
 seen+0.168
iles+0.150
grown+0.142
emon+0.141
elly+0.141
Neg logit contributions
bing-0.162
itating-0.139
itate-0.128
idents-0.127
 forth-0.126
Token inside
Token ID2641
Feature activation+8.59e-02

Pos logit contributions
bing+0.936
 cheek+0.804
 mant+0.792
 wrist+0.785
idents+0.782
Neg logit contributions
 seen-0.787
iles-0.700
grown-0.672
vious-0.665
elly-0.664
Token the
Token ID262
Feature activation-9.39e-03

Pos logit contributions
bing+0.181
 up+0.155
 cheek+0.152
 head+0.149
 mant+0.148
Neg logit contributions
 seen-0.156
iles-0.155
elly-0.144
emon-0.143
vious-0.141
Token hole
Token ID7604
Feature activation+3.78e-02

Pos logit contributions
 seen+0.017
grown+0.015
elly+0.015
iles+0.015
vious+0.014
Neg logit contributions
bing-0.020
 cheek-0.017
itating-0.017
 wrist-0.017
 regularly-0.017
Token and
Token ID290
Feature activation+2.64e-02

Pos logit contributions
bing+0.081
 cheek+0.069
 up+0.068
 mant+0.067
 wrist+0.067
Neg logit contributions
 seen-0.068
iles-0.064
elly-0.060
vious-0.059
grown-0.059
Token saw
Token ID2497
Feature activation-1.98e-01

Pos logit contributions
bing+0.059
 cheek+0.051
itating+0.051
 wrist+0.050
 mant+0.049
Neg logit contributions
 seen-0.048
iles-0.039
grown-0.039
vious-0.037
 happened-0.036
Token,
Token ID11
Feature activation-6.09e-02

Pos logit contributions
 seen+0.049
iles+0.043
vious+0.040
elly+0.040
emon+0.039
Neg logit contributions
bing-0.090
 mant-0.079
 mantle-0.077
itating-0.076
 forth-0.075
Token we
Token ID356
Feature activation+2.17e-01

Pos logit contributions
 seen+0.112
iles+0.104
emon+0.098
vious+0.097
elly+0.097
Neg logit contributions
bing-0.130
 wrist-0.106
 cheek-0.105
itating-0.105
 head-0.103
Token'll
Token ID1183
Feature activation+1.28e-01

Pos logit contributions
bing+0.531
 cheek+0.448
 wrist+0.441
 mant+0.435
 up+0.428
Neg logit contributions
 seen-0.350
iles-0.324
vious-0.302
grown-0.294
elly-0.290
Token pick
Token ID2298
Feature activation+1.16e-01

Pos logit contributions
bing+0.300
 cheek+0.242
 forth+0.241
idents+0.237
itating+0.235
Neg logit contributions
 seen-0.223
iles-0.172
 of-0.168
emon-0.167
vious-0.161
Token it
Token ID340
Feature activation+1.54e-01

Pos logit contributions
bing+0.191
 cheek+0.155
 head+0.150
 mant+0.149
 wrist+0.148
Neg logit contributions
 seen-0.265
iles-0.255
elly-0.242
vious-0.240
grown-0.239
Token up
Token ID510
Feature activation+4.61e-01

Pos logit contributions
bing+0.141
 wrist+0.104
 cheek+0.103
itating+0.097
 mant+0.095
Neg logit contributions
 seen-0.442
iles-0.421
elly-0.415
emon-0.408
ha-0.400
Token.
Token ID13
Feature activation+9.80e-03

Pos logit contributions
bing+0.952
 cheek+0.794
 mant+0.781
 wrist+0.778
 head+0.777
Neg logit contributions
 seen-0.873
iles-0.825
elly-0.785
grown-0.768
vious-0.762
Token That
Token ID1320
Feature activation+9.32e-02

Pos logit contributions
bing+0.022
 forth+0.018
 cheek+0.018
 up+0.018
 heads+0.017
Neg logit contributions
 seen-0.016
elly-0.015
iles-0.015
vious-0.015
emon-0.015
Token's
Token ID338
Feature activation+1.72e-01

Pos logit contributions
bing+0.169
 cheek+0.131
 mant+0.128
itating+0.127
 mantle+0.126
Neg logit contributions
 seen-0.207
iles-0.192
vious-0.184
elly-0.182
emon-0.180
Token a
Token ID257
Feature activation-1.83e-01

Pos logit contributions
bing+0.390
 cheek+0.334
 wrist+0.327
 mant+0.320
itating+0.319
Neg logit contributions
 seen-0.295
iles-0.278
emon-0.255
grown-0.251
elly-0.250
Token very
Token ID845
Feature activation+1.99e-01

Pos logit contributions
 seen+0.322
iles+0.280
vous+0.280
grown+0.276
emon+0.274
Neg logit contributions
bing-0.417
itating-0.331
 up-0.325
 forth-0.323
 heads-0.320
Token girl
Token ID2576
Feature activation+5.61e-02

Pos logit contributions
 seen+0.026
iles+0.024
vious+0.023
elly+0.023
grown+0.023
Neg logit contributions
bing-0.026
 cheek-0.021
 mant-0.021
 up-0.021
 mantle-0.020
Token,
Token ID11
Feature activation-5.94e-02

Pos logit contributions
bing+0.133
idents+0.120
 mant+0.116
 cheek+0.115
itating+0.115
Neg logit contributions
 seen-0.086
grown-0.070
iles-0.070
 ago-0.068
vious-0.066
Token but
Token ID475
Feature activation+7.78e-02

Pos logit contributions
 seen+0.104
grown+0.085
iles+0.080
vious+0.079
aver+0.078
Neg logit contributions
bing-0.136
idents-0.117
 mant-0.113
 edge-0.111
 cheek-0.111
Token she
Token ID673
Feature activation+1.36e-01

Pos logit contributions
bing+0.179
 cheek+0.158
 wrist+0.157
 mant+0.155
idents+0.154
Neg logit contributions
 seen-0.130
grown-0.108
iles-0.108
vious-0.105
 happened-0.099
Token stayed
Token ID9658
Feature activation+8.10e-02

Pos logit contributions
bing+0.244
 cheek+0.200
 up+0.196
 mant+0.194
 wrist+0.192
Neg logit contributions
 seen-0.296
iles-0.286
elly-0.267
vious-0.265
emon-0.264
Token still
Token ID991
Feature activation+3.22e-01

Pos logit contributions
bing+0.158
idents+0.148
 mant+0.143
 wrist+0.140
 cheek+0.139
Neg logit contributions
 seen-0.169
grown-0.141
aver-0.126
ak-0.124
vious-0.118
Token.
Token ID13
Feature activation+9.11e-03

Pos logit contributions
bing+0.716
idents+0.647
 mant+0.630
 wrist+0.617
 mantle+0.613
Neg logit contributions
 seen-0.555
grown-0.458
 ago-0.438
iles-0.437
vious-0.411
Token He
Token ID679
Feature activation-8.47e-02

Pos logit contributions
bing+0.017
 cheek+0.013
 forth+0.013
 head+0.013
 heads+0.013
Neg logit contributions
 seen-0.020
iles-0.018
grown-0.018
elly-0.018
vious-0.017
Token decided
Token ID3066
Feature activation+1.76e-02

Pos logit contributions
 seen+0.158
iles+0.145
vious+0.134
grown+0.132
vous+0.130
Neg logit contributions
bing-0.183
 cheek-0.156
 mant-0.153
 wrist-0.151
itating-0.149
Token to
Token ID284
Feature activation-9.79e-02

Pos logit contributions
bing+0.049
 cheek+0.042
 mant+0.042
 wrist+0.042
itating+0.041
Neg logit contributions
 seen-0.023
grown-0.018
iles-0.017
vious-0.017
elly-0.016
Token approach
Token ID3164
Feature activation+1.05e-02

Pos logit contributions
 seen+0.184
iles+0.168
vious+0.150
emon+0.149
vous+0.147
Neg logit contributions
bing-0.211
 cheek-0.167
idents-0.166
itating-0.164
 forth-0.161
Token to
Token ID284
Feature activation-9.59e-02

Pos logit contributions
bing+0.215
 up+0.189
 cheek+0.181
 head+0.178
 edge+0.175
Neg logit contributions
 seen-0.183
iles-0.178
vious-0.165
emon-0.165
elly-0.164
Token her
Token ID607
Feature activation+1.00e-01

Pos logit contributions
 seen+0.186
iles+0.179
elly+0.166
emon+0.164
vious+0.164
Neg logit contributions
bing-0.194
 cheek-0.162
 up-0.160
 mant-0.157
 wrist-0.157
Token.
Token ID13
Feature activation+1.69e-02

Pos logit contributions
bing+0.229
 cheek+0.193
 mant+0.191
 forth+0.190
 mantle+0.188
Neg logit contributions
 seen-0.168
iles-0.157
elly-0.146
grown-0.140
emon-0.138
Token She
Token ID1375
Feature activation+7.24e-02

Pos logit contributions
bing+0.034
 cheek+0.029
 up+0.028
 mant+0.028
 wrist+0.028
Neg logit contributions
 seen-0.033
iles-0.031
elly-0.030
vious-0.029
emon-0.029
Token looked
Token ID3114
Feature activation+2.29e-02

Pos logit contributions
bing+0.154
 cheek+0.131
 mant+0.128
 up+0.127
 wrist+0.127
Neg logit contributions
 seen-0.133
iles-0.127
elly-0.116
emon-0.116
vious-0.115
Token up
Token ID510
Feature activation+4.66e-01

Pos logit contributions
bing+0.031
 wrist+0.025
 cheek+0.023
 mant+0.022
 mantle+0.021
Neg logit contributions
 seen-0.062
grown-0.056
iles-0.056
vious-0.055
elly-0.054
Token and
Token ID290
Feature activation+2.87e-02

Pos logit contributions
bing+1.032
 cheek+0.882
 mant+0.866
 wrist+0.861
 head+0.851
Neg logit contributions
 seen-0.813
iles-0.764
elly-0.721
grown-0.710
vious-0.704
Token saw
Token ID2497
Feature activation-1.97e-01

Pos logit contributions
bing+0.092
 wrist+0.084
 mant+0.084
 cheek+0.084
 mantle+0.082
Neg logit contributions
 seen-0.026
grown-0.015
iles-0.013
vious-0.012
 happened-0.010
Token it
Token ID340
Feature activation+1.59e-01

Pos logit contributions
 seen+0.343
iles+0.337
elly+0.319
emon+0.309
vious+0.305
Neg logit contributions
bing-0.429
 up-0.382
 cheek-0.361
 head-0.354
 edge-0.353
Token perched
Token ID49264
Feature activation-2.28e-02

Pos logit contributions
bing+0.316
 wrist+0.285
 mant+0.285
 cheek+0.281
idents+0.278
Neg logit contributions
 seen-0.296
iles-0.280
grown-0.252
emon-0.248
vious-0.248
Token atop
Token ID20156
Feature activation+1.02e-01

Pos logit contributions
 seen+0.044
iles+0.042
elly+0.039
grown+0.039
emon+0.039
Neg logit contributions
bing-0.046
 cheek-0.038
 mant-0.038
 wrist-0.038
idents-0.037
Token we
Token ID356
Feature activation+2.27e-01

Pos logit contributions
bing+0.013
itating+0.011
 mantle+0.010
etically+0.010
 edge+0.010
Neg logit contributions
 seen-0.010
emon-0.009
iles-0.009
 cheese-0.009
 of-0.008
Token love
Token ID1842
Feature activation-5.70e-02

Pos logit contributions
bing+0.500
 cheek+0.420
 up+0.418
 mant+0.410
 wrist+0.409
Neg logit contributions
 seen-0.402
iles-0.384
elly-0.355
vious-0.352
emon-0.351
Token bubbles
Token ID25037
Feature activation+2.33e-02

Pos logit contributions
 seen+0.097
iles+0.096
elly+0.087
emon+0.087
grown+0.086
Neg logit contributions
bing-0.120
itating-0.108
etically-0.105
 cheek-0.105
 wrist-0.103
Token.
Token ID13
Feature activation+1.19e-02

Pos logit contributions
bing+0.047
 cheek+0.039
 up+0.039
 mant+0.039
 wrist+0.038
Neg logit contributions
 seen-0.044
iles-0.043
elly-0.040
emon-0.039
vious-0.039
Token And
Token ID843
Feature activation+6.27e-03

Pos logit contributions
bing+0.023
 cheek+0.019
 up+0.019
 head+0.019
 mant+0.018
Neg logit contributions
 seen-0.024
iles-0.023
elly-0.021
emon-0.021
vious-0.021
Token we
Token ID356
Feature activation+2.26e-01

Pos logit contributions
bing+0.012
itating+0.011
 mantle+0.010
etically+0.010
 edge+0.010
Neg logit contributions
 seen-0.012
iles-0.011
emon-0.011
 cheese-0.010
 crocod-0.010
Token love
Token ID1842
Feature activation-5.70e-02

Pos logit contributions
bing+0.521
 cheek+0.437
 up+0.435
 mant+0.428
 wrist+0.428
Neg logit contributions
 seen-0.384
iles-0.364
elly-0.334
vious-0.333
emon-0.332
Token the
Token ID262
Feature activation-1.14e-02

Pos logit contributions
iles+0.099
 seen+0.099
elly+0.090
emon+0.089
grown+0.088
Neg logit contributions
bing-0.118
itating-0.105
 cheek-0.102
etically-0.102
 mant-0.101
Token m
Token ID285
Feature activation+1.19e-02

Pos logit contributions
 seen+0.020
iles+0.020
elly+0.018
emon+0.018
vious+0.018
Neg logit contributions
bing-0.024
 cheek-0.020
 up-0.020
 wrist-0.020
 mant-0.020
Tokenop
Token ID404
Feature activation-4.94e-02

Pos logit contributions
bing+0.029
 up+0.026
 mant+0.026
 cheek+0.025
 wrist+0.025
Neg logit contributions
iles-0.017
 seen-0.016
elly-0.016
emon-0.015
vious-0.015
Token."
Token ID526
Feature activation+4.95e-02

Pos logit contributions
iles+0.080
 seen+0.079
elly+0.075
vious+0.074
emon+0.071
Neg logit contributions
bing-0.109
 mant-0.095
itating-0.095
 wrist-0.093
etically-0.092
Token go
Token ID467
Feature activation+1.12e-01

Pos logit contributions
bing+0.281
 up+0.249
 head+0.242
 cheek+0.234
 forth+0.229
Neg logit contributions
 seen-0.356
elly-0.351
iles-0.351
grown-0.344
emon-0.336
Token?"
Token ID1701
Feature activation-1.59e-01

Pos logit contributions
bing+0.270
 up+0.242
 cheek+0.232
 forth+0.230
 head+0.230
Neg logit contributions
 seen-0.170
iles-0.165
elly-0.152
emon-0.150
grown-0.149
Token Tom
Token ID4186
Feature activation-2.98e-01

Pos logit contributions
iles+0.232
 seen+0.222
elly+0.208
emon+0.205
vous+0.202
Neg logit contributions
bing-0.381
 up-0.351
 head-0.337
 cheek-0.335
 edge-0.334
Token asked
Token ID1965
Feature activation-2.33e-02

Pos logit contributions
 seen+0.541
iles+0.528
elly+0.491
emon+0.490
grown+0.485
Neg logit contributions
bing-0.634
 up-0.534
 cheek-0.533
 head-0.528
 wrist-0.516
Token.
Token ID13
Feature activation+1.13e-02

Pos logit contributions
 seen+0.043
iles+0.041
elly+0.038
emon+0.038
vious+0.038
Neg logit contributions
bing-0.050
 cheek-0.042
 up-0.042
 head-0.041
 mant-0.041
Token
Token ID198
Feature activation+1.02e-02

Pos logit contributions
bing+0.022
 cheek+0.019
 up+0.018
 head+0.018
 wrist+0.018
Neg logit contributions
 seen-0.023
iles-0.022
elly-0.020
vious-0.020
emon-0.020
Token
Token ID198
Feature activation+1.02e-02

Pos logit contributions
bing+0.021
 cheek+0.018
 up+0.018
 head+0.017
 mant+0.017
Neg logit contributions
 seen-0.019
iles-0.018
elly-0.017
emon-0.017
vious-0.017
Token"
Token ID1
Feature activation-1.65e-02

Pos logit contributions
bing+0.021
 up+0.017
 cheek+0.017
 head+0.017
 wrist+0.017
Neg logit contributions
 seen-0.020
iles-0.019
elly-0.018
vious-0.018
emon-0.018
TokenI
Token ID40
Feature activation+4.63e-02

Pos logit contributions
 seen+0.036
iles+0.035
elly+0.033
emon+0.033
vious+0.033
Neg logit contributions
bing-0.029
 cheek-0.023
 up-0.023
 head-0.022
 mant-0.022
Token don
Token ID836
Feature activation+7.76e-02

Pos logit contributions
bing+0.077
 up+0.070
 head+0.068
idents+0.065
itating+0.063
Neg logit contributions
iles-0.097
elly-0.091
emon-0.091
 seen-0.090
vious-0.087
Token't
Token ID470
Feature activation+1.57e-01

Pos logit contributions
bing+0.156
 wrist+0.141
 cheek+0.139
 mant+0.139
 heads+0.139
Neg logit contributions
iles-0.129
 seen-0.126
 of-0.121
emon-0.120
vious-0.118
Token't
Token ID470
Feature activation+1.50e-01

Pos logit contributions
bing+0.377
 cheek+0.318
 mant+0.317
 wrist+0.317
 forth+0.313
Neg logit contributions
 seen-0.367
iles-0.355
vious-0.352
emon-0.347
vous-0.336
Token remember
Token ID3505
Feature activation+1.58e-01

Pos logit contributions
bing+0.327
itating+0.255
idents+0.252
 forth+0.250
 up+0.246
Neg logit contributions
 seen-0.310
iles-0.246
vous-0.242
emon-0.242
 of-0.240
Token what
Token ID644
Feature activation-5.28e-02

Pos logit contributions
bing+0.330
 cheek+0.278
 up+0.270
 mant+0.268
 head+0.268
Neg logit contributions
 seen-0.299
iles-0.282
grown-0.262
elly-0.262
emon-0.260
Token happened
Token ID3022
Feature activation-4.02e-01

Pos logit contributions
iles+0.111
 seen+0.110
elly+0.104
vious+0.103
emon+0.100
Neg logit contributions
bing-0.093
 up-0.079
 cheek-0.078
 wrist-0.076
 head-0.076
Token.
Token ID13
Feature activation+8.06e-03

Pos logit contributions
 seen+0.691
iles+0.606
grown+0.587
elly+0.581
vious+0.570
Neg logit contributions
bing-0.919
 mant-0.792
 cheek-0.788
 wrist-0.776
 head-0.772
Token
Token ID198
Feature activation+2.89e-03

Pos logit contributions
bing+0.018
 forth+0.015
 cheek+0.015
 head+0.015
 wrist+0.015
Neg logit contributions
 seen-0.014
elly-0.013
iles-0.013
grown-0.013
emon-0.012
Token
Token ID198
Feature activation+2.69e-03

Pos logit contributions
bing+0.006
 cheek+0.005
 head+0.005
 forth+0.005
 mant+0.005
Neg logit contributions
 seen-0.005
iles-0.005
elly-0.004
grown-0.004
emon-0.004
TokenShe
Token ID3347
Feature activation+1.48e-01

Pos logit contributions
bing+0.006
 cheek+0.005
 head+0.005
 wrist+0.005
 forth+0.005
Neg logit contributions
 seen-0.004
iles-0.004
elly-0.004
grown-0.004
vous-0.004
Token searched
Token ID16499
Feature activation-1.54e-06

Pos logit contributions
bing+0.302
 up+0.260
 head+0.256
 cheek+0.252
 forth+0.249
Neg logit contributions
 seen-0.274
iles-0.271
elly-0.256
vious-0.246
emon-0.246
Token high
Token ID1029
Feature activation+1.71e-02

Feature not active on prev token;
no predictions were affected.
Token and
Token ID290
Feature activation+2.25e-02

Pos logit contributions
bing+0.034
 cheek+0.030
 wrist+0.029
 mant+0.029
 head+0.028
Neg logit contributions
 seen-0.033
iles-0.031
elly-0.029
grown-0.029
vious-0.029
Token into
Token ID656
Feature activation+1.83e-01

Pos logit contributions
bing+0.010
 wrist+0.008
 mant+0.008
idents+0.008
 cheek+0.008
Neg logit contributions
 seen-0.012
iles-0.011
grown-0.011
elly-0.011
vious-0.010
Token a
Token ID257
Feature activation-1.81e-01

Pos logit contributions
bing+0.355
 up+0.320
 edge+0.304
 head+0.296
 wrist+0.290
Neg logit contributions
iles-0.359
elly-0.338
 seen-0.338
emon-0.336
vious-0.331
Token deep
Token ID2769
Feature activation-3.17e-02

Pos logit contributions
 seen+0.328
iles+0.311
elly+0.292
grown+0.288
vious+0.287
Neg logit contributions
bing-0.389
 cheek-0.324
 up-0.323
 mant-0.316
 wrist-0.316
Token,
Token ID11
Feature activation-5.91e-02

Pos logit contributions
iles+0.064
 seen+0.064
emon+0.061
vious+0.058
elly+0.058
Neg logit contributions
bing-0.061
 up-0.052
 head-0.051
 edge-0.050
 forth-0.049
Token dark
Token ID3223
Feature activation-1.36e-01

Pos logit contributions
 seen+0.107
grown+0.091
iles+0.090
vious+0.090
vous+0.087
Neg logit contributions
bing-0.131
itating-0.107
 wrist-0.106
idents-0.105
 regularly-0.103
Token hole
Token ID7604
Feature activation+3.60e-02

Pos logit contributions
iles+0.280
 seen+0.277
emon+0.268
vious+0.256
elly+0.248
Neg logit contributions
bing-0.257
 head-0.215
 up-0.213
 edge-0.210
 cheek-0.209
Token.
Token ID13
Feature activation+7.08e-03

Pos logit contributions
bing+0.065
 up+0.055
 head+0.051
 cheek+0.050
 edge+0.050
Neg logit contributions
 seen-0.077
iles-0.077
emon-0.073
elly-0.072
vious-0.072
Token
Token ID198
Feature activation+4.51e-03

Pos logit contributions
bing+0.014
 up+0.012
 cheek+0.012
 mant+0.012
 head+0.012
Neg logit contributions
 seen-0.014
iles-0.013
elly-0.012
vious-0.012
emon-0.012
Token
Token ID198
Feature activation+4.62e-03

Pos logit contributions
bing+0.009
 up+0.007
 cheek+0.007
 mant+0.007
 head+0.007
Neg logit contributions
 seen-0.009
iles-0.009
elly-0.008
vious-0.008
emon-0.008
TokenThe
Token ID464
Feature activation+7.08e-02

Pos logit contributions
bing+0.009
 up+0.007
 cheek+0.007
itating+0.007
 edge+0.007
Neg logit contributions
 seen-0.009
iles-0.009
elly-0.008
vious-0.008
grown-0.008
Token old
Token ID1468
Feature activation-3.05e-02

Pos logit contributions
bing+0.155
 cheek+0.130
 up+0.129
 wrist+0.127
 mant+0.127
Neg logit contributions
 seen-0.127
iles-0.122
elly-0.113
vious-0.112
emon-0.111
Token to
Token ID284
Feature activation-1.01e-01

Pos logit contributions
bing+0.722
itating+0.646
 mant+0.642
rily+0.609
 forth+0.606
Neg logit contributions
 seen-0.217
 happened-0.151
 enough-0.126
 Tom-0.122
grown-0.121
Token share
Token ID2648
Feature activation+1.29e-01

Pos logit contributions
 seen+0.206
iles+0.195
emon+0.179
aver+0.176
ah+0.171
Neg logit contributions
bing-0.199
itating-0.165
idents-0.155
 up-0.154
 mant-0.153
Token and
Token ID290
Feature activation+2.33e-02

Pos logit contributions
bing+0.294
itating+0.252
 cheek+0.250
 mant+0.247
 forth+0.242
Neg logit contributions
 seen-0.220
iles-0.203
grown-0.189
elly-0.187
emon-0.186
Token listen
Token ID6004
Feature activation+4.71e-02

Pos logit contributions
bing+0.048
itating+0.039
 mant+0.037
 up+0.037
 edge+0.036
Neg logit contributions
 seen-0.048
iles-0.042
emon-0.041
grown-0.039
vous-0.039
Token.
Token ID13
Feature activation+1.22e-02

Pos logit contributions
bing+0.132
 forth+0.120
 wrist+0.116
 mant+0.116
idents+0.115
Neg logit contributions
 seen-0.050
elly-0.042
grown-0.041
iles-0.040
 of-0.040
Token They
Token ID1119
Feature activation+5.38e-02

Pos logit contributions
bing+0.025
 cheek+0.021
 up+0.021
 forth+0.020
 head+0.020
Neg logit contributions
 seen-0.024
iles-0.023
elly-0.021
grown-0.021
emon-0.021
Token play
Token ID711
Feature activation+7.44e-02

Pos logit contributions
bing+0.110
 cheek+0.089
 up+0.087
 mant+0.086
itating+0.086
Neg logit contributions
 seen-0.106
iles-0.102
vious-0.095
emon-0.094
elly-0.093
Token with
Token ID351
Feature activation+1.71e-01

Pos logit contributions
bing+0.129
 cheek+0.105
itating+0.103
 mant+0.103
 forth+0.100
Neg logit contributions
 seen-0.164
iles-0.155
emon-0.149
vious-0.146
elly-0.145
Token the
Token ID262
Feature activation-2.05e-02

Pos logit contributions
bing+0.346
 up+0.294
 cheek+0.288
 head+0.283
 wrist+0.280
Neg logit contributions
 seen-0.330
iles-0.318
elly-0.301
emon-0.297
vious-0.297
Token bear
Token ID6842
Feature activation-1.22e-02

Pos logit contributions
 seen+0.040
iles+0.038
elly+0.036
emon+0.036
vious+0.035
Neg logit contributions
bing-0.042
 cheek-0.035
 up-0.035
 head-0.034
 mant-0.034
Token on
Token ID319
Feature activation+1.40e-02

Pos logit contributions
 seen+0.023
iles+0.022
vious+0.020
elly+0.020
emon+0.020
Neg logit contributions
bing-0.025
 mant-0.021
 cheek-0.021
itating-0.020
idents-0.020
Token perfect
Token ID2818
Feature activation+9.78e-02

Pos logit contributions
iles+0.086
 seen+0.086
emon+0.079
elly+0.079
vious+0.078
Neg logit contributions
bing-0.099
 up-0.085
 cheek-0.084
 head-0.082
 mant-0.081
Token.
Token ID13
Feature activation+1.12e-02

Pos logit contributions
bing+0.194
 up+0.164
 cheek+0.161
 head+0.157
 wrist+0.155
Neg logit contributions
 seen-0.188
iles-0.186
elly-0.173
emon-0.171
vious-0.171
Token<|endoftext|>
Token ID50256
Feature activation-1.10e-02

Pos logit contributions
bing+0.023
 cheek+0.019
 up+0.019
 forth+0.019
 head+0.018
Neg logit contributions
 seen-0.022
iles-0.021
elly-0.020
grown-0.019
emon-0.019
TokenOnce
Token ID7454
Feature activation-1.40e-01

Pos logit contributions
 seen+0.019
iles+0.019
vious+0.017
emon+0.017
elly+0.017
Neg logit contributions
bing-0.024
 up-0.021
 edge-0.020
 cheek-0.019
itating-0.019
Token upon
Token ID2402
Feature activation-1.18e-01

Pos logit contributions
 seen+0.221
iles+0.213
emon+0.196
elly+0.196
vious+0.194
Neg logit contributions
bing-0.325
 up-0.284
 cheek-0.280
 head-0.274
itating-0.270
Token a
Token ID257
Feature activation-1.85e-01

Pos logit contributions
iles+0.208
elly+0.193
 seen+0.182
emon+0.181
vious+0.180
Neg logit contributions
bing-0.245
 up-0.240
 cheek-0.221
 wrist-0.216
 mantle-0.213
Token time
Token ID640
Feature activation-3.09e-01

Pos logit contributions
iles+0.242
 seen+0.237
angs+0.229
vious+0.228
emon+0.224
Neg logit contributions
bing-0.444
 up-0.413
 head-0.408
 cheek-0.403
 regularly-0.394
Token there
Token ID612
Feature activation-1.36e-01

Pos logit contributions
 seen+0.600
iles+0.587
vious+0.559
elly+0.544
vous+0.543
Neg logit contributions
bing-0.617
 up-0.543
 cheek-0.502
itating-0.490
 edge-0.488
Token was
Token ID373
Feature activation+9.66e-02

Pos logit contributions
 seen+0.311
iles+0.305
elly+0.289
vious+0.286
emon+0.285
Neg logit contributions
bing-0.221
 up-0.179
 cheek-0.173
 head-0.170
 wrist-0.168
Token a
Token ID257
Feature activation-1.85e-01

Pos logit contributions
bing+0.195
 up+0.190
 herself+0.165
 mantle+0.157
 edge+0.155
Neg logit contributions
iles-0.178
 seen-0.163
angs-0.162
vious-0.159
elly-0.158
Token jewel
Token ID16927
Feature activation-3.75e-02

Pos logit contributions
 seen+0.291
iles+0.283
elly+0.265
grown+0.256
emon+0.255
Neg logit contributions
bing-0.433
 up-0.365
 cheek-0.363
 mant-0.361
 wrist-0.357
Token is
Token ID318
Feature activation+1.71e-02

Pos logit contributions
bing+0.318
 wrist+0.296
 cheek+0.285
 mant+0.276
 forehead+0.269
Neg logit contributions
iles-0.285
 seen-0.283
vious-0.279
 happened-0.271
elly-0.266
Token important
Token ID1593
Feature activation+4.69e-02

Pos logit contributions
bing+0.033
 cheek+0.027
 up+0.027
 wrist+0.026
 mant+0.026
Neg logit contributions
 seen-0.035
iles-0.034
elly-0.032
emon-0.031
vious-0.031
Token to
Token ID284
Feature activation-9.63e-02

Pos logit contributions
bing+0.111
 up+0.096
 cheek+0.095
 head+0.093
 wrist+0.092
Neg logit contributions
 seen-0.074
iles-0.072
elly-0.065
vious-0.065
emon-0.065
Token listen
Token ID6004
Feature activation+5.33e-02

Pos logit contributions
 seen+0.194
iles+0.187
emon+0.173
grown+0.168
elly+0.168
Neg logit contributions
bing-0.186
 cheek-0.153
itating-0.152
 wrist-0.150
 up-0.149
Token to
Token ID284
Feature activation-9.62e-02

Pos logit contributions
bing+0.137
 wrist+0.122
 cheek+0.121
 mant+0.119
 forth+0.118
Neg logit contributions
 seen-0.072
iles-0.065
elly-0.063
grown-0.061
vious-0.060
Token the
Token ID262
Feature activation-8.99e-03

Pos logit contributions
 seen+0.150
iles+0.150
grown+0.126
 cheese+0.122
elly+0.121
Neg logit contributions
bing-0.221
itating-0.200
 cheek-0.192
idents-0.190
 wrist-0.190
Token doctor
Token ID6253
Feature activation+2.16e-02

Pos logit contributions
 seen+0.017
iles+0.016
elly+0.015
vious+0.015
emon+0.015
Neg logit contributions
bing-0.019
 up-0.016
 cheek-0.015
 wrist-0.015
itating-0.015
Token's
Token ID338
Feature activation+1.77e-01

Pos logit contributions
bing+0.048
 mant+0.042
 cheek+0.041
itating+0.041
 mantle+0.040
Neg logit contributions
 seen-0.036
iles-0.034
vious-0.032
elly-0.031
emon-0.031
Token advice
Token ID5608
Feature activation+7.97e-02

Pos logit contributions
bing+0.330
itating+0.286
 mant+0.268
 wrist+0.267
 mantle+0.264
Neg logit contributions
 seen-0.361
iles-0.343
grown-0.322
elly-0.320
vous-0.316
Token.
Token ID13
Feature activation+1.28e-02

Pos logit contributions
bing+0.175
 cheek+0.148
 mant+0.146
 wrist+0.145
 up+0.144
Neg logit contributions
 seen-0.139
iles-0.133
elly-0.123
vious-0.121
emon-0.120
Token<|endoftext|>
Token ID50256
Feature activation-6.33e-03

Pos logit contributions
bing+0.026
 cheek+0.021
 up+0.021
 head+0.021
 mant+0.021
Neg logit contributions
 seen-0.025
iles-0.024
elly-0.022
emon-0.022
vious-0.022
Token happy
Token ID3772
Feature activation+2.32e-01

Pos logit contributions
bing+0.056
itating+0.045
 edge+0.045
 heads+0.044
 wrist+0.044
Neg logit contributions
 seen-0.057
grown-0.046
vous-0.046
vious-0.045
 happened-0.044
Token.
Token ID13
Feature activation+1.22e-02

Pos logit contributions
bing+0.512
 cheek+0.435
 mant+0.429
itating+0.425
 wrist+0.424
Neg logit contributions
 seen-0.408
iles-0.385
elly-0.360
grown-0.355
vious-0.349
Token Bob
Token ID5811
Feature activation-1.67e-01

Pos logit contributions
bing+0.025
 up+0.021
 cheek+0.021
 mantle+0.020
 mant+0.020
Neg logit contributions
 seen-0.023
iles-0.022
vious-0.021
emon-0.020
elly-0.020
Token was
Token ID373
Feature activation+1.07e-01

Pos logit contributions
 seen+0.304
iles+0.280
vious+0.266
vous+0.256
elly+0.255
Neg logit contributions
bing-0.359
 mant-0.302
 cheek-0.299
idents-0.299
itating-0.297
Token happy
Token ID3772
Feature activation+2.32e-01

Pos logit contributions
bing+0.202
idents+0.168
 heads+0.162
 cheek+0.161
 wrist+0.158
Neg logit contributions
 seen-0.241
vious-0.205
grown-0.203
aver-0.203
vous-0.202
Token to
Token ID284
Feature activation-9.48e-02

Pos logit contributions
bing+0.564
 mant+0.484
 cheek+0.481
itating+0.475
 wrist+0.472
Neg logit contributions
 seen-0.368
iles-0.330
grown-0.311
elly-0.311
vious-0.295
Token have
Token ID423
Feature activation+4.12e-03

Pos logit contributions
 seen+0.161
iles+0.139
emon+0.133
vious+0.132
vous+0.129
Neg logit contributions
bing-0.224
idents-0.188
itating-0.188
 up-0.186
 mant-0.184
Token such
Token ID884
Feature activation-1.31e-01

Pos logit contributions
bing+0.009
 up+0.008
 head+0.007
 forth+0.007
 regularly+0.007
Neg logit contributions
iles-0.007
elly-0.007
emon-0.006
vious-0.006
angs-0.006
Token a
Token ID257
Feature activation-1.77e-01

Pos logit contributions
iles+0.201
 seen+0.197
elly+0.189
emon+0.175
vious+0.175
Neg logit contributions
bing-0.309
 up-0.286
 head-0.269
 cheek-0.268
 mant-0.266
Token great
Token ID1049
Feature activation+2.94e-02

Pos logit contributions
 seen+0.322
iles+0.288
elly+0.286
grown+0.285
vous+0.284
Neg logit contributions
bing-0.397
 heads-0.312
itating-0.308
 forth-0.308
 cheek-0.307
Token Grand
Token ID5675
Feature activation+4.05e-02

Pos logit contributions
bing+0.059
 up+0.050
 cheek+0.050
 head+0.049
 wrist+0.049
Neg logit contributions
 seen-0.056
iles-0.055
elly-0.051
emon-0.050
vious-0.050
TokenI
Token ID40
Feature activation+4.04e-02

Pos logit contributions
bing+0.314
 cheek+0.261
 up+0.257
 wrist+0.257
itating+0.246
Neg logit contributions
iles-0.374
 seen-0.367
vious-0.337
emon-0.334
elly-0.331
Token won
Token ID1839
Feature activation-2.28e-02

Pos logit contributions
bing+0.066
 up+0.064
 head+0.061
 regularly+0.059
 cheek+0.059
Neg logit contributions
iles-0.082
 seen-0.076
grown-0.074
elly-0.074
emon-0.071
Tokenâ
Token ID22940
Feature activation+1.63e-01

Pos logit contributions
 seen+0.042
iles+0.040
emon+0.039
vious+0.039
elly+0.037
Neg logit contributions
bing-0.043
 mant-0.039
 heads-0.038
 cheek-0.038
 forth-0.038
Token€
Token ID26391
Feature activation+8.59e-02

Pos logit contributions
bing+0.281
 cheek+0.234
 up+0.228
 head+0.222
 wrist+0.222
Neg logit contributions
 seen-0.359
iles-0.352
grown-0.332
elly-0.328
emon-0.316
Tokenâ„¢
Token ID8151
Feature activation+2.38e-02

Pos logit contributions
bing+0.168
 wrist+0.140
 cheek+0.140
 up+0.139
 head+0.136
Neg logit contributions
iles-0.167
 seen-0.164
vious-0.151
elly-0.150
emon-0.150
Tokent
Token ID83
Feature activation-9.47e-02

Pos logit contributions
bing+0.049
 up+0.043
itating+0.042
 cheek+0.042
 edge+0.039
Neg logit contributions
 seen-0.043
iles-0.043
grown-0.038
emon-0.038
vious-0.038
Token pinch
Token ID25394
Feature activation-3.56e-02

Pos logit contributions
iles+0.176
 seen+0.175
elly+0.163
grown+0.163
emon+0.160
Neg logit contributions
bing-0.191
 cheek-0.167
 up-0.164
 head-0.162
 wrist-0.157
Token it
Token ID340
Feature activation+1.55e-01

Pos logit contributions
 seen+0.072
iles+0.069
elly+0.065
vious+0.064
emon+0.064
Neg logit contributions
bing-0.069
 cheek-0.057
 mant-0.056
 up-0.055
 head-0.055
Token,
Token ID11
Feature activation-6.12e-02

Pos logit contributions
bing+0.363
 mant+0.312
 wrist+0.307
 cheek+0.306
 mantle+0.304
Neg logit contributions
 seen-0.242
iles-0.225
vious-0.217
elly-0.214
emon-0.204
Token I
Token ID314
Feature activation+4.82e-03

Pos logit contributions
 seen+0.115
iles+0.104
vious+0.102
elly+0.101
emon+0.100
Neg logit contributions
bing-0.129
 mant-0.105
 cheek-0.102
 heads-0.102
itating-0.102
Token just
Token ID655
Feature activation+5.24e-02

Pos logit contributions
bing+0.009
 up+0.008
 head+0.008
 cheek+0.008
 wrist+0.008
Neg logit contributions
iles-0.009
 seen-0.009
elly-0.008
grown-0.008
emon-0.008
Token a
Token ID257
Feature activation-1.83e-01

Pos logit contributions
 seen+0.094
iles+0.084
grown+0.080
emon+0.077
elly+0.076
Neg logit contributions
bing-0.142
 cheek-0.122
 mant-0.120
 wrist-0.119
itating-0.118
Token group
Token ID1448
Feature activation-3.63e-02

Pos logit contributions
 seen+0.365
iles+0.359
elly+0.333
vious+0.333
emon+0.326
Neg logit contributions
bing-0.341
 cheek-0.313
 head-0.302
 up-0.294
 edge-0.291
Token of
Token ID286
Feature activation-3.69e-01

Pos logit contributions
 seen+0.029
iles+0.029
elly+0.025
emon+0.023
grown+0.022
Neg logit contributions
bing-0.111
 up-0.101
 head-0.098
 cheek-0.098
 forth-0.098
Token dolphins
Token ID36798
Feature activation-1.96e-01

Pos logit contributions
 seen+0.736
elly+0.718
emon+0.670
iles+0.669
vous+0.645
Neg logit contributions
bing-0.672
 up-0.619
 forth-0.579
 head-0.564
 heads-0.559
Token saw
Token ID2497
Feature activation-2.03e-01

Pos logit contributions
 seen+0.297
iles+0.291
elly+0.276
emon+0.268
vious+0.261
Neg logit contributions
bing-0.467
 up-0.407
 cheek-0.398
 head-0.396
 forth-0.393
Token what
Token ID644
Feature activation-5.56e-02

Pos logit contributions
 seen+0.434
iles+0.423
elly+0.406
vious+0.400
vous+0.400
Neg logit contributions
bing-0.364
 up-0.315
 cheek-0.284
 edge-0.277
 head-0.276
Token had
Token ID550
Feature activation-4.71e-02

Pos logit contributions
iles+0.103
elly+0.097
 seen+0.097
vious+0.093
aver+0.090
Neg logit contributions
bing-0.104
 up-0.100
 head-0.085
 cheek-0.085
 forth-0.085
Token happened
Token ID3022
Feature activation-4.03e-01

Pos logit contributions
iles+0.054
elly+0.052
emon+0.046
grown+0.042
vious+0.041
Neg logit contributions
bing-0.124
 up-0.122
 head-0.112
 cheek-0.110
itating-0.109
Token,
Token ID11
Feature activation-5.90e-02

Pos logit contributions
 seen+0.663
iles+0.565
grown+0.560
elly+0.544
 ago+0.521
Neg logit contributions
bing-0.929
 mant-0.835
 cheek-0.823
 wrist-0.811
 mantle-0.803
Token and
Token ID290
Feature activation+2.27e-02

Pos logit contributions
 seen+0.104
iles+0.093
grown+0.090
vious+0.088
elly+0.087
Neg logit contributions
bing-0.131
 wrist-0.112
 mant-0.110
 cheek-0.110
idents-0.107
Token they
Token ID484
Feature activation+1.63e-01

Pos logit contributions
bing+0.045
 cheek+0.038
 wrist+0.038
 mant+0.037
 head+0.036
Neg logit contributions
 seen-0.046
iles-0.042
grown-0.040
vious-0.040
emon-0.039
Token his
Token ID465
Feature activation-8.46e-02

Pos logit contributions
 seen+0.149
iles+0.139
grown+0.124
vous+0.119
vious+0.118
Neg logit contributions
bing-0.237
 cheek-0.207
itating-0.206
 wrist-0.205
idents-0.202
Token owner
Token ID4870
Feature activation+5.44e-02

Pos logit contributions
 seen+0.156
iles+0.154
elly+0.139
grown+0.139
vous+0.136
Neg logit contributions
bing-0.171
itating-0.145
 mant-0.140
 mantle-0.137
 up-0.135
Token.
Token ID13
Feature activation+1.50e-02

Pos logit contributions
bing+0.117
 cheek+0.100
 mant+0.100
idents+0.098
 wrist+0.097
Neg logit contributions
 seen-0.096
iles-0.089
vious-0.083
grown-0.083
elly-0.082
Token The
Token ID383
Feature activation-3.23e-02

Pos logit contributions
bing+0.032
 cheek+0.027
 up+0.026
 head+0.026
 forth+0.026
Neg logit contributions
 seen-0.028
iles-0.026
elly-0.025
grown-0.024
emon-0.024
Token moral
Token ID6573
Feature activation+8.24e-02

Pos logit contributions
 seen+0.072
iles+0.070
elly+0.066
emon+0.065
vious+0.065
Neg logit contributions
bing-0.056
 cheek-0.045
 up-0.045
 head-0.044
 mant-0.044
Token of
Token ID286
Feature activation-3.67e-01

Pos logit contributions
 up+0.203
bing+0.202
 regularly+0.188
 tightly+0.187
 cheek+0.185
Neg logit contributions
iles-0.092
 seen-0.087
angs-0.082
emon-0.073
eh-0.072
Token the
Token ID262
Feature activation-1.15e-02

Pos logit contributions
 seen+0.659
iles+0.634
elly+0.588
emon+0.581
vious+0.579
Neg logit contributions
bing-0.790
 cheek-0.669
 up-0.663
 head-0.650
 mant-0.649
Token story
Token ID1621
Feature activation-4.19e-02

Pos logit contributions
 seen+0.022
iles+0.021
elly+0.020
vious+0.019
grown+0.019
Neg logit contributions
bing-0.024
 cheek-0.020
 up-0.020
 head-0.020
 mant-0.020
Token is
Token ID318
Feature activation+1.42e-02

Pos logit contributions
 seen+0.065
iles+0.063
elly+0.061
vious+0.058
grown+0.057
Neg logit contributions
bing-0.101
 cheek-0.085
 mant-0.085
 wrist-0.084
 forth-0.084
Token to
Token ID284
Feature activation-9.74e-02

Pos logit contributions
bing+0.033
 cheek+0.028
 up+0.028
 head+0.027
 mant+0.027
Neg logit contributions
 seen-0.024
iles-0.023
elly-0.021
emon-0.021
vious-0.021
Token always
Token ID1464
Feature activation+3.62e-02

Pos logit contributions
 seen+0.169
iles+0.160
emon+0.143
vious+0.139
grown+0.137
Neg logit contributions
bing-0.215
 cheek-0.177
itating-0.177
idents-0.175
 mant-0.175
Token Ad
Token ID1215
Feature activation+1.16e-01

Pos logit contributions
bing+0.039
 forth+0.034
 cheek+0.032
 head+0.030
 forehead+0.030
Neg logit contributions
 seen-0.026
elly-0.026
grown-0.025
vious-0.024
 ago-0.024
Tokeneline
Token ID4470
Feature activation+4.64e-02

Pos logit contributions
bing+0.229
 cheek+0.191
 up+0.189
 mant+0.185
 head+0.185
Neg logit contributions
 seen-0.230
iles-0.218
elly-0.207
vious-0.205
emon-0.203
Token was
Token ID373
Feature activation+1.10e-01

Pos logit contributions
bing+0.100
 cheek+0.085
 mant+0.084
idents+0.083
itating+0.081
Neg logit contributions
 seen-0.084
iles-0.078
vious-0.074
emon-0.072
elly-0.070
Token so
Token ID523
Feature activation+7.83e-02

Pos logit contributions
bing+0.233
idents+0.198
 cheek+0.193
 heads+0.192
 mant+0.187
Neg logit contributions
 seen-0.224
vious-0.186
grown-0.180
vous-0.179
emon-0.177
Token proud
Token ID6613
Feature activation+2.99e-01

Pos logit contributions
bing+0.141
 forth+0.109
 mant+0.107
 heads+0.107
 cheek+0.106
Neg logit contributions
 seen-0.182
vious-0.169
vous-0.162
elly-0.160
grown-0.159
Token of
Token ID286
Feature activation-3.62e-01

Pos logit contributions
bing+0.759
 up+0.686
 cheek+0.655
 head+0.644
 forth+0.643
Neg logit contributions
iles-0.396
 seen-0.388
emon-0.360
vous-0.350
vious-0.341
Token herself
Token ID5223
Feature activation+3.60e-01

Pos logit contributions
 seen+0.666
grown+0.631
iles+0.598
 of+0.586
 ago+0.578
Neg logit contributions
bing-0.690
idents-0.635
 cheek-0.600
itating-0.593
 edge-0.589
Token.
Token ID13
Feature activation+1.70e-02

Pos logit contributions
bing+0.871
 mant+0.743
 cheek+0.738
 heads+0.737
itating+0.731
Neg logit contributions
 seen-0.550
iles-0.497
emon-0.472
elly-0.471
grown-0.464
Token She
Token ID1375
Feature activation+7.30e-02

Pos logit contributions
bing+0.043
 forth+0.038
 cheek+0.035
 forehead+0.034
 heads+0.034
Neg logit contributions
 seen-0.024
elly-0.024
grown-0.023
 ago-0.023
vious-0.021
Token knew
Token ID2993
Feature activation+1.03e-02

Pos logit contributions
bing+0.175
 cheek+0.145
idents+0.144
 mant+0.144
itating+0.144
Neg logit contributions
 seen-0.122
iles-0.104
vious-0.102
emon-0.097
ha-0.096
Token she
Token ID673
Feature activation+1.41e-01

Pos logit contributions
bing+0.025
 cheek+0.022
 heads+0.022
 mant+0.021
 head+0.021
Neg logit contributions
 seen-0.016
iles-0.014
grown-0.014
emon-0.013
elly-0.013
Token huge
Token ID3236
Feature activation-6.93e-02

Pos logit contributions
 seen+0.314
iles+0.301
elly+0.281
emon+0.275
grown+0.275
Neg logit contributions
bing-0.407
 cheek-0.342
 up-0.340
 mant-0.335
 wrist-0.333
Token room
Token ID2119
Feature activation+1.98e-01

Pos logit contributions
 seen+0.136
iles+0.130
elly+0.120
emon+0.120
vious+0.119
Neg logit contributions
bing-0.139
 up-0.121
 head-0.119
 cheek-0.118
 forth-0.116
Token filled
Token ID5901
Feature activation-1.47e-01

Pos logit contributions
bing+0.349
 up+0.288
 cheek+0.284
 head+0.278
 wrist+0.275
Neg logit contributions
 seen-0.427
iles-0.425
vious-0.393
elly-0.392
emon-0.390
Token with
Token ID351
Feature activation+1.71e-01

Pos logit contributions
 seen+0.336
iles+0.315
elly+0.304
grown+0.301
emon+0.299
Neg logit contributions
bing-0.239
 cheek-0.198
idents-0.189
 head-0.188
itating-0.188
Token lots
Token ID6041
Feature activation-1.21e-02

Pos logit contributions
bing+0.322
 up+0.287
 cheek+0.270
 head+0.269
 mant+0.267
Neg logit contributions
iles-0.336
 seen-0.332
emon-0.312
elly-0.311
vious-0.302
Token of
Token ID286
Feature activation-3.72e-01

Pos logit contributions
 seen+0.011
iles+0.011
emon+0.009
aver+0.008
vous+0.008
Neg logit contributions
bing-0.036
 up-0.033
 cheek-0.032
 mant-0.032
 wrist-0.031
Token things
Token ID1243
Feature activation+8.36e-02

Pos logit contributions
 seen+0.734
iles+0.715
emon+0.667
elly+0.663
vious+0.652
Neg logit contributions
bing-0.724
 up-0.624
 cheek-0.611
 mant-0.589
 head-0.589
Token.
Token ID13
Feature activation+1.13e-02

Pos logit contributions
bing+0.184
 cheek+0.156
 mant+0.154
 wrist+0.152
 head+0.152
Neg logit contributions
 seen-0.150
iles-0.139
elly-0.129
grown-0.128
vious-0.127
Token He
Token ID679
Feature activation-8.62e-02

Pos logit contributions
bing+0.027
 cheek+0.023
 up+0.023
 head+0.023
 mant+0.023
Neg logit contributions
 seen-0.018
iles-0.017
elly-0.015
emon-0.015
vious-0.015
Token wanted
Token ID2227
Feature activation-1.23e-01

Pos logit contributions
 seen+0.156
iles+0.150
elly+0.139
emon+0.137
vious+0.137
Neg logit contributions
bing-0.185
 cheek-0.156
 up-0.155
 head-0.152
 mant-0.151
Token to
Token ID284
Feature activation-9.94e-02

Pos logit contributions
iles+0.219
elly+0.212
 seen+0.209
vious+0.205
emon+0.198
Neg logit contributions
bing-0.252
 up-0.237
 edge-0.213
 head-0.213
 mant-0.210
Token took
Token ID1718
Feature activation-7.90e-02

Pos logit contributions
bing+0.057
 cheek+0.049
itating+0.049
 mant+0.048
 up+0.047
Neg logit contributions
 seen-0.040
iles-0.036
vious-0.033
emon-0.033
grown-0.033
Token off
Token ID572
Feature activation+2.71e-01

Pos logit contributions
 seen+0.188
iles+0.184
elly+0.175
emon+0.172
vious+0.171
Neg logit contributions
bing-0.124
 up-0.099
 cheek-0.097
 head-0.094
 mant-0.093
Token.
Token ID13
Feature activation+1.08e-02

Pos logit contributions
bing+0.539
 up+0.489
 cheek+0.451
 head+0.439
 wrist+0.438
Neg logit contributions
iles-0.534
 seen-0.504
elly-0.494
vious-0.480
emon-0.476
Token
Token ID198
Feature activation+5.12e-03

Pos logit contributions
bing+0.026
 up+0.023
 cheek+0.022
idents+0.022
 head+0.022
Neg logit contributions
 seen-0.017
iles-0.016
vious-0.014
elly-0.014
emon-0.014
Token
Token ID198
Feature activation+5.18e-03

Pos logit contributions
bing+0.009
 up+0.008
 cheek+0.008
itating+0.008
 edge+0.008
Neg logit contributions
 seen-0.011
iles-0.010
vious-0.010
emon-0.010
elly-0.010
TokenTom
Token ID13787
Feature activation-3.04e-01

Pos logit contributions
bing+0.013
 up+0.011
 cheek+0.011
 head+0.011
itating+0.011
Neg logit contributions
 seen-0.007
iles-0.007
elly-0.006
vious-0.006
emon-0.006
Token was
Token ID373
Feature activation+1.03e-01

Pos logit contributions
 seen+0.581
iles+0.542
vious+0.510
elly+0.500
emon+0.498
Neg logit contributions
bing-0.625
 cheek-0.533
 mant-0.525
 wrist-0.516
itating-0.507
Token very
Token ID845
Feature activation+2.01e-01

Pos logit contributions
bing+0.151
 cheek+0.116
 up+0.116
 head+0.112
 mant+0.111
Neg logit contributions
 seen-0.255
iles-0.249
elly-0.236
emon-0.234
vious-0.233
Token upset
Token ID9247
Feature activation+1.64e-01

Pos logit contributions
bing+0.324
 cheek+0.259
 up+0.258
 head+0.250
 mant+0.245
Neg logit contributions
 seen-0.471
iles-0.460
elly-0.433
emon-0.429
vious-0.427
Token.
Token ID13
Feature activation+1.18e-02

Pos logit contributions
bing+0.302
 up+0.256
 cheek+0.238
 head+0.234
 forth+0.234
Neg logit contributions
 seen-0.351
iles-0.343
emon-0.328
vous-0.317
vious-0.316
Token He
Token ID679
Feature activation-8.31e-02

Pos logit contributions
bing+0.027
 up+0.023
 cheek+0.023
 head+0.023
 mant+0.022
Neg logit contributions
 seen-0.020
iles-0.019
emon-0.017
elly-0.017
vious-0.017
Token.
Token ID13
Feature activation+1.35e-02

Pos logit contributions
bing+0.862
 up+0.778
 edge+0.695
 head+0.688
 regularly+0.684
Neg logit contributions
iles-0.949
 seen-0.910
emon-0.899
elly-0.870
vious-0.869
Token He
Token ID679
Feature activation-7.94e-02

Pos logit contributions
bing+0.027
 up+0.023
 cheek+0.023
 head+0.023
 wrist+0.022
Neg logit contributions
 seen-0.026
iles-0.026
emon-0.023
elly-0.023
grown-0.023
Token remembered
Token ID12086
Feature activation-7.10e-02

Pos logit contributions
 seen+0.151
iles+0.142
vious+0.130
emon+0.130
elly+0.129
Neg logit contributions
bing-0.167
 cheek-0.141
 mant-0.136
 wrist-0.135
itating-0.135
Token just
Token ID655
Feature activation+5.41e-02

Pos logit contributions
iles+0.134
 seen+0.132
elly+0.129
emon+0.125
vious+0.124
Neg logit contributions
bing-0.141
 up-0.125
 head-0.117
 cheek-0.116
 forth-0.115
Token in
Token ID287
Feature activation+6.93e-02

Pos logit contributions
bing+0.112
 up+0.100
 head+0.096
 cheek+0.096
 forth+0.094
Neg logit contributions
iles-0.099
elly-0.093
 seen-0.092
emon-0.089
vious-0.087
Token time
Token ID640
Feature activation-3.01e-01

Pos logit contributions
bing+0.130
 up+0.113
 head+0.112
 wrist+0.107
 edge+0.106
Neg logit contributions
iles-0.135
elly-0.131
vious-0.129
emon-0.128
 seen-0.126
Token and
Token ID290
Feature activation+2.90e-02

Pos logit contributions
 seen+0.538
iles+0.483
grown+0.459
elly+0.448
vious+0.439
Neg logit contributions
bing-0.651
 mant-0.556
 cheek-0.555
idents-0.554
itating-0.545
Token carefully
Token ID7773
Feature activation+3.02e-01

Pos logit contributions
bing+0.062
 cheek+0.053
 mant+0.051
itating+0.051
 wrist+0.051
Neg logit contributions
 seen-0.055
iles-0.049
vious-0.046
grown-0.046
emon-0.045
Token climbed
Token ID19952
Feature activation-1.91e-01

Pos logit contributions
bing+0.627
 up+0.545
 head+0.539
 cheek+0.523
 forth+0.516
Neg logit contributions
iles-0.540
 seen-0.518
elly-0.517
vous-0.488
vious-0.487
Token back
Token ID736
Feature activation+2.77e-01

Pos logit contributions
 seen+0.462
iles+0.446
grown+0.420
emon+0.414
vious+0.411
Neg logit contributions
bing-0.285
 cheek-0.237
 wrist-0.231
 mant-0.225
idents-0.224
Token up
Token ID510
Feature activation+4.65e-01

Pos logit contributions
bing+0.452
 cheek+0.371
 mant+0.357
idents+0.351
 mantle+0.347
Neg logit contributions
 seen-0.649
iles-0.619
elly-0.588
grown-0.585
vious-0.570