TOP ACTIVATIONS
MAX = 4.763

and looked around for something to cover her up .
the things that are important to you . And from that
She looked around for something to wrap the sun flower but
There are many other things to do at the festival .
see what I will do to you . You will be
see what I can do to fix them ?"
parents were going crazy trying to help him with the packing
explore the west and hoped to have the best time ever
, " I 'm trying to look for a place to
he had to do something to make them happy .
bad behavior and being kind can make a big difference .
to hurry to the lake to swim .
and kept hoping and struggling to believe they would . <|endoftext|>
the world and wondering who to join with . One day
, excited for what was to come . <|endoftext|> Once upon
he needed to do something to get the books he lf
out what they should do to become successful .
's fly this important helicopter to the moon ."
find new and special ways to wrap gifts . Everyone was
something special to be able to go on any journey .

MIN ACTIVATIONS
MIN = -4.600

her best friend , a little cat . One sunny day
a mom my and a little girl . The mom my
a mom my and a little girl . The mom my
. He was a curious little boy and loved going on
hold it up . A little girl named Lily loved to
, the elephant met a little boy . The boy asked
. A little boy was playing in the
mouse was a very curious little creature . His home was
scary there . A brave little boy went into the dark
her best friend , a little girl named Mia .
. His owner , a little girl named Lily , wanted
. She was a very curious girl who was always looking
<|endoftext|> Tim was an enthusiastic little boy . He loved to
One day a little girl came up to the
there was a depend able little girl named Sue . She
show his friend , a little girl named Lily . Lily
sunny day , a happy little girl named Lily came to
. Max was a crazy little boy who loved to play
there was a very polite little bear . He was always
. One day , a little girl came to the house

INTERVAL 2.422 - 4.763
CONTAINS 3.746%

to play with their toys in the garden . One day
her heart knowing that she would still get to celebrate her
, " Sl ipping means to try to get somewhere without
Okay , Lily , we can play somewhere else . But
wait to get started . " On your mark

INTERVAL 0.081 - 2.422
CONTAINS 41.717%

face . <|endoftext|> Once upon a time there was a little
Oh no , Ben , are you okay ?"
followed the voice . When she stepped out of her cave
, M andy . " Let 's go
couldn 't get in . From that day on , Lily

INTERVAL -2.259 - 0.081
CONTAINS 51.327%

, " Your reward is a big , y ummy cookie
She found another tw ig and pretended to fight with Tom
and Mia picked him up and put him inside her bin
not . It 's mine . Mom gave it to me
. He tried to grab the chain from them .

INTERVAL -4.600 - -2.259
CONTAINS 3.210%

<|endoftext|> Once there was a kind dog named Spot . He
he were well and free . He imagined that he could
One day , Sarah and her mom went shopping . When
that he started to weep . He felt so sorry for
her room . It was big and colorful , but she
Token and
Token ID290
Feature activation+3.81e-02

Pos logit contributions
sed+1.520
 girl+1.434
 Pole+1.394
 boy+1.389
ometer+1.368
Neg logit contributions
ck-1.568
aches-1.527
 yourself-1.476
irs-1.437
 avoid-1.434
Token looked
Token ID3114
Feature activation-1.48e-01

Pos logit contributions
ck+0.078
aches+0.076
irs+0.073
 yourself+0.072
 avoid+0.071
Neg logit contributions
sed-0.067
 girl-0.066
 boy-0.064
 Pole-0.063
ometer-0.060
Token around
Token ID1088
Feature activation-4.84e-01

Pos logit contributions
sed+0.285
 girl+0.278
 boy+0.272
 Pole+0.267
ometer+0.259
Neg logit contributions
ck-0.283
aches-0.274
irs-0.261
 yourself-0.258
 avoid-0.255
Token for
Token ID329
Feature activation+1.87e+00

Pos logit contributions
sed+1.014
 girl+0.985
 Pole+0.969
 boy+0.967
ometer+0.940
Neg logit contributions
ck-0.833
aches-0.799
irs-0.764
 yourself-0.760
 avoid-0.755
Token something
Token ID1223
Feature activation+3.18e-01

Pos logit contributions
ck+3.255
aches+3.175
 achieve+3.103
irs+3.042
 yourself+2.857
Neg logit contributions
 girl-3.983
sed-3.934
 boy-3.819
 eyes-3.711
 mouse-3.538
Token to
Token ID284
Feature activation+4.76e+00

Pos logit contributions
ck+0.564
aches+0.547
irs+0.525
 yourself+0.512
 avoid+0.501
Neg logit contributions
sed-0.648
 girl-0.641
 boy-0.625
 Pole-0.607
ometer-0.589
Token cover
Token ID3002
Feature activation+1.84e+00

Pos logit contributions
irs+11.969
lled+10.637
cks+10.543
ck+10.381
aches+10.293
Neg logit contributions
 boy-9.286
 girl-9.176
 tum-8.421
sed-8.110
 mouth-8.058
Token her
Token ID607
Feature activation+1.43e-01

Pos logit contributions
ck+2.749
aches+2.656
irs+2.539
 achieve+2.525
 avoid+2.378
Neg logit contributions
 girl-4.350
 boy-4.223
sed-4.217
ometer-3.966
 eyes-3.929
Token up
Token ID510
Feature activation+1.24e+00

Pos logit contributions
ck+0.361
aches+0.353
irs+0.342
 yourself+0.337
 avoid+0.333
Neg logit contributions
sed-0.184
 girl-0.180
 boy-0.173
 Pole-0.166
ometer-0.158
Token.
Token ID13
Feature activation-1.33e+00

Pos logit contributions
ck+2.417
aches+2.379
irs+2.305
 yourself+2.176
 achieve+2.167
Neg logit contributions
 girl-2.371
 boy-2.297
sed-2.251
 Pole-2.101
ometer-2.099
Token
Token ID220
Feature activation+2.16e+00

Pos logit contributions
sed+2.664
 girl+2.562
 Pole+2.519
 boy+2.489
ometer+2.367
Neg logit contributions
ck-2.331
 yourself-2.286
aches-2.279
irs-2.267
 avoid-2.227
Token the
Token ID262
Feature activation-3.55e-01

Pos logit contributions
aches+4.220
ck+4.108
irs+3.888
 achieve+3.817
irl+3.679
Neg logit contributions
 girl-5.395
 boy-5.305
sed-5.068
 mouse-4.965
ometer-4.800
Token things
Token ID1243
Feature activation+1.71e+00

Pos logit contributions
sed+0.659
 girl+0.632
 boy+0.616
 Pole+0.616
ometer+0.596
Neg logit contributions
ck-0.693
aches-0.680
 yourself-0.650
irs-0.634
 avoid-0.633
Token that
Token ID326
Feature activation+1.59e+00

Pos logit contributions
ck+3.493
aches+3.401
irs+3.309
 avoid+3.099
 yourself+3.076
Neg logit contributions
 girl-3.056
 boy-3.012
sed-2.884
 Pole-2.863
ometer-2.746
Token are
Token ID389
Feature activation+2.29e+00

Pos logit contributions
ck+3.114
aches+2.967
irs+2.916
 yourself+2.759
asting+2.696
Neg logit contributions
 girl-3.112
 boy-3.106
sed-3.028
 Pole-2.871
ometer-2.774
Token important
Token ID1593
Feature activation+2.51e-01

Pos logit contributions
ck+3.916
irs+3.908
aches+3.699
 achieve+3.489
e+3.474
Neg logit contributions
 girl-4.838
sed-4.827
 boy-4.762
 mouse-4.371
 Pole-4.338
Token to
Token ID284
Feature activation+4.69e+00

Pos logit contributions
ck+0.473
aches+0.460
irs+0.440
 yourself+0.429
 avoid+0.424
Neg logit contributions
sed-0.483
 girl-0.476
 boy-0.465
 Pole-0.452
ometer-0.437
Token you
Token ID345
Feature activation+1.99e+00

Pos logit contributions
ck+8.873
asting+8.548
irs+8.445
aches+8.314
irl+8.213
Neg logit contributions
 girl-10.789
 boy-10.646
 "-10.171
sed-10.113
 mouse-9.799
Token.
Token ID13
Feature activation-1.94e-02

Pos logit contributions
ck+4.329
irs+4.174
aches+4.172
irl+3.773
 yourself+3.752
Neg logit contributions
 girl-3.417
sed-3.369
 boy-3.331
ometer-3.063
 Pole-3.053
Token And
Token ID843
Feature activation+6.08e-01

Pos logit contributions
sed+0.038
 girl+0.037
 boy+0.036
 Pole+0.035
ometer+0.034
Neg logit contributions
ck-0.036
aches-0.035
irs-0.034
 yourself-0.033
 avoid-0.032
Token from
Token ID422
Feature activation+4.19e-01

Pos logit contributions
ck+1.194
aches+1.172
irs+1.125
 yourself+1.087
 avoid+1.065
Neg logit contributions
 girl-1.129
sed-1.117
 boy-1.104
 Pole-1.049
ometer-1.018
Token that
Token ID326
Feature activation+3.63e-01

Pos logit contributions
ck+0.709
aches+0.695
irs+0.667
 yourself+0.644
 achieve+0.633
Neg logit contributions
sed-0.886
 girl-0.875
 boy-0.856
 Pole-0.829
ometer-0.803
Token She
Token ID1375
Feature activation-1.47e-03

Pos logit contributions
sed+2.529
 girl+2.382
 Pole+2.335
 boy+2.315
ometer+2.207
Neg logit contributions
ck-2.281
irs-2.240
aches-2.229
 yourself-2.181
 avoid-2.130
Token looked
Token ID3114
Feature activation-9.33e-01

Pos logit contributions
sed+0.003
 girl+0.003
 boy+0.003
 Pole+0.003
ometer+0.003
Neg logit contributions
ck-0.003
aches-0.003
irs-0.003
 yourself-0.003
 avoid-0.002
Token around
Token ID1088
Feature activation-6.67e-01

Pos logit contributions
sed+1.937
 Pole+1.829
 girl+1.823
 boy+1.803
ometer+1.780
Neg logit contributions
ck-1.700
aches-1.623
irs-1.531
 avoid-1.506
 yourself-1.499
Token for
Token ID329
Feature activation+1.59e+00

Pos logit contributions
sed+1.172
 girl+1.116
 Pole+1.109
 boy+1.090
ometer+1.072
Neg logit contributions
ck-1.375
aches-1.325
irs-1.283
 avoid-1.277
 yourself-1.269
Token something
Token ID1223
Feature activation+1.95e-01

Pos logit contributions
ck+2.587
aches+2.536
irs+2.435
 achieve+2.386
 yourself+2.282
Neg logit contributions
 girl-3.549
 boy-3.437
sed-3.405
 eyes-3.221
 Pole-3.173
Token to
Token ID284
Feature activation+4.60e+00

Pos logit contributions
ck+0.357
aches+0.347
irs+0.331
 yourself+0.325
 avoid+0.319
Neg logit contributions
sed-0.385
 girl-0.380
 boy-0.371
 Pole-0.361
ometer-0.349
Token wrap
Token ID14441
Feature activation+7.41e-01

Pos logit contributions
irs+10.010
lled+9.176
ck+8.745
cks+8.622
aches+8.298
Neg logit contributions
 girl-9.906
 boy-9.322
 cheeks-8.999
sed-8.950
 mouth-8.799
Token the
Token ID262
Feature activation-8.74e-01

Pos logit contributions
ck+1.276
aches+1.226
irs+1.187
 achieve+1.136
 yourself+1.131
Neg logit contributions
 girl-1.543
sed-1.541
 boy-1.504
 Pole-1.441
ometer-1.424
Token sun
Token ID4252
Feature activation-4.80e-01

Pos logit contributions
sed+1.443
 Pole+1.355
 girl+1.352
 boy+1.327
ometer+1.280
Neg logit contributions
ck-1.862
aches-1.856
 yourself-1.781
 avoid-1.721
irs-1.715
Tokenflower
Token ID25547
Feature activation-2.51e-01

Pos logit contributions
sed+0.925
 girl+0.900
 boy+0.879
 Pole+0.863
ometer+0.819
Neg logit contributions
ck-0.905
aches-0.888
irs-0.831
 yourself-0.826
 avoid-0.812
Token but
Token ID475
Feature activation+5.26e-01

Pos logit contributions
sed+0.486
 girl+0.473
 boy+0.462
 Pole+0.455
ometer+0.434
Neg logit contributions
ck-0.468
aches-0.459
irs-0.430
 yourself-0.429
 avoid-0.421
Token There
Token ID1318
Feature activation+1.10e+00

Pos logit contributions
sed+1.366
 girl+1.355
 boy+1.306
 Pole+1.287
ometer+1.168
Neg logit contributions
ck-2.340
 yourself-2.313
aches-2.239
irs-2.222
 avoid-2.214
Token are
Token ID389
Feature activation+9.23e-01

Pos logit contributions
ck+2.085
aches+1.986
irs+1.937
 yourself+1.836
 avoid+1.811
Neg logit contributions
 girl-2.155
 boy-2.123
sed-2.095
 Pole-1.997
ometer-1.935
Token many
Token ID867
Feature activation+1.09e+00

Pos logit contributions
ck+1.559
aches+1.488
irs+1.468
 achieve+1.368
 yourself+1.355
Neg logit contributions
sed-1.979
 girl-1.957
 boy-1.926
 Pole-1.827
ometer-1.786
Token other
Token ID584
Feature activation+7.72e-01

Pos logit contributions
ck+2.328
aches+2.274
irs+2.232
 achieve+2.111
 yourself+2.109
Neg logit contributions
 girl-1.864
 boy-1.824
sed-1.757
 Pole-1.666
ometer-1.558
Token things
Token ID1243
Feature activation+1.24e+00

Pos logit contributions
ck+1.810
aches+1.768
irs+1.720
 yourself+1.642
 achieve+1.640
Neg logit contributions
 girl-1.150
 boy-1.122
sed-1.118
 Pole-1.047
ometer-0.973
Token to
Token ID284
Feature activation+4.59e+00

Pos logit contributions
ck+2.617
aches+2.570
irs+2.482
 yourself+2.358
 achieve+2.349
Neg logit contributions
sed-2.132
 girl-2.100
 boy-2.055
 Pole-1.986
ometer-1.890
Token do
Token ID466
Feature activation+3.04e+00

Pos logit contributions
irs+10.286
lled+9.614
asting+9.274
ck+9.229
aches+9.195
Neg logit contributions
 boy-9.633
 girl-9.280
 dog-8.325
sed-8.261
 Pole-8.234
Token at
Token ID379
Feature activation+4.05e+00

Pos logit contributions
ck+7.594
aches+7.419
irs+7.404
 ease+6.968
asting+6.598
Neg logit contributions
 girl-4.379
 boy-4.337
sed-4.300
 "-3.877
 Pole-3.859
Token the
Token ID262
Feature activation+5.51e-01

Pos logit contributions
ck+9.226
aches+8.910
irs+8.422
asting+7.916
ered+7.712
Neg logit contributions
 boy-8.602
 girl-8.225
sed-7.208
 Pole-7.105
 mouse-6.844
Token festival
Token ID10876
Feature activation+1.18e+00

Pos logit contributions
ck+1.150
aches+1.113
irs+1.084
 yourself+1.038
 avoid+1.028
Neg logit contributions
sed-0.959
 girl-0.958
 boy-0.937
 Pole-0.899
ometer-0.855
Token.
Token ID13
Feature activation-4.83e-01

Pos logit contributions
ck+2.451
aches+2.371
irs+2.321
 yourself+2.186
 avoid+2.170
Neg logit contributions
 girl-2.102
 boy-2.060
sed-2.037
 Pole-1.940
ometer-1.847
Token see
Token ID766
Feature activation+1.35e+00

Pos logit contributions
irs+4.918
ck+4.893
aches+4.729
lled+4.360
asting+4.319
Neg logit contributions
sed-6.404
 girl-6.285
 boy-6.254
 mouse-5.788
ometer-5.668
Token what
Token ID644
Feature activation+1.37e-01

Pos logit contributions
ck+2.290
aches+2.257
irs+2.125
 achieve+2.042
 avoid+2.004
Neg logit contributions
 girl-2.878
 boy-2.818
sed-2.810
 Pole-2.687
 mouse-2.570
Token I
Token ID314
Feature activation+2.45e+00

Pos logit contributions
ck+0.235
aches+0.228
irs+0.217
 yourself+0.212
 avoid+0.208
Neg logit contributions
sed-0.288
 girl-0.284
 boy-0.277
 Pole-0.271
ometer-0.263
Token will
Token ID481
Feature activation+3.23e+00

Pos logit contributions
aches+4.662
ck+4.626
irs+4.553
 yourself+4.312
 ourselves+4.171
Neg logit contributions
 boy-5.038
 girl-5.014
sed-4.973
 Pole-4.482
ometer-4.425
Token do
Token ID466
Feature activation+2.27e+00

Pos logit contributions
irs+6.133
ck+6.071
aches+6.069
lled+5.344
 ease+5.285
Neg logit contributions
sed-6.690
 girl-6.646
 boy-6.581
 "-6.092
 mouse-6.044
Token to
Token ID284
Feature activation+4.59e+00

Pos logit contributions
ck+5.329
aches+5.146
irs+5.119
 ease+4.825
 avoid+4.642
Neg logit contributions
 girl-3.651
 boy-3.581
sed-3.373
 Pole-3.177
 "-3.148
Token you
Token ID345
Feature activation+3.15e+00

Pos logit contributions
irs+9.868
lled+9.093
aches+8.925
ck+8.712
asting+8.555
Neg logit contributions
 boy-10.113
 girl-9.978
 cat-9.181
 mouse-9.119
sed-8.801
Token.
Token ID13
Feature activation-8.19e-01

Pos logit contributions
ck+6.958
irs+6.896
aches+6.833
lled+6.535
 ease+6.414
Neg logit contributions
 girl-5.483
 boy-5.459
 mouse-4.974
sed-4.878
 Pole-4.841
Token You
Token ID921
Feature activation+2.01e+00

Pos logit contributions
sed+1.177
 girl+1.159
 boy+1.118
 Pole+1.096
ometer+1.021
Neg logit contributions
ck-1.903
 yourself-1.867
aches-1.841
irs-1.819
 avoid-1.801
Token will
Token ID481
Feature activation+2.88e+00

Pos logit contributions
ck+4.326
aches+4.197
irs+4.180
 yourself+3.998
asting+3.872
Neg logit contributions
 girl-3.463
 boy-3.455
sed-3.425
 Pole-3.144
ometer-3.128
Token be
Token ID307
Feature activation+8.83e-01

Pos logit contributions
irs+4.985
ck+4.932
aches+4.738
lled+4.441
asting+4.330
Neg logit contributions
sed-6.450
 girl-6.367
 boy-6.294
 mouse-5.856
 Pole-5.680
Token see
Token ID766
Feature activation+1.80e+00

Pos logit contributions
irs+6.818
ck+6.597
lled+6.450
aches+6.338
asting+6.128
Neg logit contributions
 girl-7.849
 boy-7.819
sed-7.546
 Pole-7.237
 mouse-7.180
Token what
Token ID644
Feature activation+7.78e-01

Pos logit contributions
ck+2.727
aches+2.705
irs+2.496
 achieve+2.477
 avoid+2.417
Neg logit contributions
 girl-4.173
 boy-4.083
sed-3.990
 Pole-3.906
 mouse-3.753
Token I
Token ID314
Feature activation+2.34e+00

Pos logit contributions
ck+1.488
aches+1.426
irs+1.385
 yourself+1.329
 avoid+1.317
Neg logit contributions
 girl-1.496
sed-1.472
 boy-1.469
 Pole-1.392
ometer-1.346
Token can
Token ID460
Feature activation+3.56e+00

Pos logit contributions
ck+4.825
aches+4.779
irs+4.734
 yourself+4.585
 ourselves+4.370
Neg logit contributions
 boy-4.424
 girl-4.409
sed-4.217
 Pole-3.964
ometer-3.858
Token do
Token ID466
Feature activation+2.48e+00

Pos logit contributions
irs+6.631
ck+6.508
aches+6.292
lled+5.881
asting+5.838
Neg logit contributions
 boy-7.820
 girl-7.713
 Pole-7.215
sed-7.212
 mouse-6.970
Token to
Token ID284
Feature activation+4.58e+00

Pos logit contributions
ck+5.729
irs+5.490
aches+5.447
 ease+5.195
 avoid+5.064
Neg logit contributions
 girl-4.124
 boy-4.082
sed-3.776
 Pole-3.679
 "-3.496
Token fix
Token ID4259
Feature activation+2.12e+00

Pos logit contributions
irs+9.502
lled+8.640
ck+8.457
aches+8.339
cks+8.267
Neg logit contributions
 boy-10.599
 girl-10.339
 cat-9.271
 mouse-9.217
 eyes-8.906
Token them
Token ID606
Feature activation+2.43e+00

Pos logit contributions
ck+3.030
aches+3.012
 achieve+2.749
irs+2.725
asting+2.700
Neg logit contributions
 girl-5.212
 boy-5.150
sed-4.832
 Pole-4.789
 mouse-4.763
Token?"
Token ID1701
Feature activation+2.82e-01

Pos logit contributions
ck+4.477
irs+4.292
aches+4.188
 ease+3.936
 avoid+3.891
Neg logit contributions
 girl-5.054
 boy-5.053
 Pole-4.764
sed-4.584
ometer-4.529
Token
Token ID198
Feature activation+1.98e+00

Pos logit contributions
ck+0.520
aches+0.507
irs+0.483
 yourself+0.473
 avoid+0.465
Neg logit contributions
sed-0.556
 girl-0.553
 boy-0.540
 Pole-0.524
ometer-0.506
Token
Token ID198
Feature activation+8.49e-01

Pos logit contributions
aches+4.256
ck+4.205
irs+4.054
 happen+3.920
 yourself+3.914
Neg logit contributions
 girl-3.520
 boy-3.491
sed-3.175
 Pole-3.056
ometer-3.008
Token parents
Token ID3397
Feature activation+9.53e-01

Pos logit contributions
sed+2.050
 girl+1.997
 Pole+1.948
 boy+1.899
ometer+1.824
Neg logit contributions
ck-2.692
aches-2.651
 avoid-2.535
 yourself-2.513
irs-2.504
Token were
Token ID547
Feature activation+7.82e-01

Pos logit contributions
ck+1.921
aches+1.865
irs+1.799
 yourself+1.757
 avoid+1.724
Neg logit contributions
 girl-1.718
sed-1.695
 boy-1.669
 Pole-1.589
ometer-1.502
Token going
Token ID1016
Feature activation+2.76e-01

Pos logit contributions
ck+1.598
aches+1.553
irs+1.505
 yourself+1.450
 avoid+1.444
Neg logit contributions
 girl-1.384
sed-1.382
 boy-1.348
 Pole-1.283
ometer-1.223
Token crazy
Token ID7165
Feature activation-3.97e-01

Pos logit contributions
ck+0.579
aches+0.564
irs+0.541
 yourself+0.530
 avoid+0.523
Neg logit contributions
sed-0.472
 girl-0.467
 boy-0.453
 Pole-0.440
ometer-0.423
Token trying
Token ID2111
Feature activation+1.14e+00

Pos logit contributions
sed+0.653
 girl+0.622
 boy+0.601
 Pole+0.593
ometer+0.576
Neg logit contributions
ck-0.868
aches-0.846
irs-0.816
 yourself-0.813
 avoid-0.801
Token to
Token ID284
Feature activation+4.56e+00

Pos logit contributions
ck+2.622
aches+2.527
irs+2.454
 yourself+2.372
 avoid+2.345
Neg logit contributions
 girl-1.711
sed-1.700
 boy-1.667
 Pole-1.574
ometer-1.509
Token help
Token ID1037
Feature activation+1.70e+00

Pos logit contributions
irs+9.115
cks+8.388
ck+8.383
aches+8.122
irl+7.914
Neg logit contributions
 boy-10.853
 girl-10.623
 mouse-9.366
 cat-9.186
 man-9.103
Token him
Token ID683
Feature activation+1.63e+00

Pos logit contributions
ck+3.393
aches+3.296
irs+3.156
 yourself+2.994
 achieve+2.971
Neg logit contributions
 girl-3.154
 boy-3.108
sed-2.966
 Pole-2.909
 mouse-2.762
Token with
Token ID351
Feature activation+1.93e+00

Pos logit contributions
ck+3.462
irs+3.309
aches+3.287
 yourself+3.095
 avoid+3.015
Neg logit contributions
 girl-2.773
 boy-2.713
sed-2.676
 Pole-2.546
ometer-2.530
Token the
Token ID262
Feature activation-5.13e-01

Pos logit contributions
ck+3.181
aches+3.097
 achieve+2.853
irs+2.830
 yourself+2.759
Neg logit contributions
 girl-4.279
sed-4.233
 boy-4.225
 Pole-3.952
 eyes-3.938
Token packing
Token ID24157
Feature activation+7.92e-01

Pos logit contributions
sed+0.814
 girl+0.772
 boy+0.744
 Pole+0.742
ometer+0.716
Neg logit contributions
ck-1.132
aches-1.117
irs-1.067
 yourself-1.059
 avoid-1.053
Token explore
Token ID7301
Feature activation+1.11e+00

Pos logit contributions
ck+5.592
aches+5.268
irs+5.250
cks+4.733
 yourself+4.726
Neg logit contributions
 girl-8.271
 boy-8.074
sed-7.903
 "-7.674
 eyes-7.621
Token the
Token ID262
Feature activation+9.20e-02

Pos logit contributions
ck+2.061
aches+2.008
irs+1.890
 yourself+1.834
 achieve+1.802
Neg logit contributions
 girl-2.199
 boy-2.137
sed-2.117
 Pole-2.023
 mouse-1.914
Token west
Token ID7421
Feature activation+5.19e-01

Pos logit contributions
ck+0.184
aches+0.180
irs+0.172
 yourself+0.169
 avoid+0.167
Neg logit contributions
sed-0.167
 girl-0.164
 boy-0.159
 Pole-0.155
ometer-0.150
Token and
Token ID290
Feature activation+8.73e-01

Pos logit contributions
ck+1.122
aches+1.093
irs+1.050
 yourself+1.025
 avoid+1.009
Neg logit contributions
 girl-0.857
sed-0.851
 boy-0.831
 Pole-0.803
ometer-0.764
Token hoped
Token ID10719
Feature activation+7.73e-01

Pos logit contributions
ck+1.340
aches+1.295
irs+1.236
 yourself+1.179
 avoid+1.128
Neg logit contributions
 girl-2.011
sed-1.997
 boy-1.968
 Pole-1.885
ometer-1.816
Token to
Token ID284
Feature activation+4.54e+00

Pos logit contributions
ck+1.541
aches+1.496
irs+1.434
 yourself+1.396
 achieve+1.367
Neg logit contributions
 girl-1.409
sed-1.402
 boy-1.368
 Pole-1.314
ometer-1.261
Token have
Token ID423
Feature activation+2.09e+00

Pos logit contributions
irs+8.329
ck+8.250
cks+7.944
aches+7.874
irl+7.342
Neg logit contributions
 girl-11.617
 boy-11.545
 "-10.205
 woman-9.983
 child-9.918
Token the
Token ID262
Feature activation-2.14e-01

Pos logit contributions
ck+3.620
aches+3.483
irs+3.371
 achieve+3.276
 avoid+3.199
Neg logit contributions
 girl-4.576
 boy-4.481
sed-4.351
 eyes-4.024
 mouse-4.011
Token best
Token ID1266
Feature activation+3.93e-01

Pos logit contributions
sed+0.413
 girl+0.400
 boy+0.390
 Pole+0.388
ometer+0.376
Neg logit contributions
ck-0.401
aches-0.392
irs-0.370
 yourself-0.370
 avoid-0.365
Token time
Token ID640
Feature activation+5.85e-01

Pos logit contributions
ck+0.655
aches+0.633
irs+0.604
 yourself+0.588
 avoid+0.576
Neg logit contributions
sed-0.845
 girl-0.841
 boy-0.822
 Pole-0.797
ometer-0.772
Token ever
Token ID1683
Feature activation+6.82e-01

Pos logit contributions
ck+1.095
aches+1.067
irs+1.016
 yourself+0.995
 avoid+0.979
Neg logit contributions
sed-1.123
 girl-1.119
 boy-1.090
 Pole-1.058
ometer-1.015
Token,
Token ID11
Feature activation-2.26e+00

Pos logit contributions
sed+2.523
 girl+2.398
 Pole+2.363
 boy+2.326
ometer+2.190
Neg logit contributions
ck-3.557
aches-3.552
 yourself-3.308
ered-3.252
 achieve-3.225
Token "
Token ID366
Feature activation+9.29e-01

Pos logit contributions
sed+2.019
 girl+1.843
 boy+1.715
 Pole+1.616
ometer+1.560
Neg logit contributions
ck-6.660
aches-6.568
 yourself-6.341
 achieve-6.332
 avoid-6.286
TokenI
Token ID40
Feature activation+2.20e+00

Pos logit contributions
ck+1.737
aches+1.709
irs+1.632
 yourself+1.549
 avoid+1.541
Neg logit contributions
 girl-1.837
 boy-1.794
sed-1.793
 Pole-1.708
ometer-1.657
Token'm
Token ID1101
Feature activation+1.16e+00

Pos logit contributions
ck+3.885
aches+3.844
irs+3.800
 yourself+3.667
 ourselves+3.535
Neg logit contributions
 girl-4.758
 boy-4.662
sed-4.642
 Pole-4.281
ometer-4.237
Token trying
Token ID2111
Feature activation+1.08e+00

Pos logit contributions
ck+2.290
aches+2.243
irs+2.151
 avoid+2.075
 achieve+2.062
Neg logit contributions
sed-2.174
 girl-2.129
 boy-2.063
 Pole-2.002
ometer-1.901
Token to
Token ID284
Feature activation+4.54e+00

Pos logit contributions
ck+2.522
aches+2.447
irs+2.373
 achieve+2.259
 avoid+2.257
Neg logit contributions
sed-1.622
 girl-1.611
 boy-1.548
 Pole-1.470
ometer-1.435
Token look
Token ID804
Feature activation+5.91e-01

Pos logit contributions
irs+10.220
ck+9.188
cks+8.992
aches+8.944
asting+8.488
Neg logit contributions
 girl-10.111
 boy-9.623
 rabbit-9.223
 mouse-8.768
sed-8.709
Token for
Token ID329
Feature activation+2.45e+00

Pos logit contributions
ck+1.231
aches+1.204
irs+1.155
 yourself+1.121
 achieve+1.112
Neg logit contributions
 girl-1.014
sed-1.003
 boy-0.983
 Pole-0.944
ometer-0.905
Token a
Token ID257
Feature activation-9.66e-01

Pos logit contributions
ck+4.788
aches+4.691
 achieve+4.499
irs+4.406
asting+4.325
Neg logit contributions
 girl-4.864
 boy-4.686
sed-4.546
 eyes-4.448
 Pole-4.417
Token place
Token ID1295
Feature activation+8.84e-01

Pos logit contributions
sed+1.719
 girl+1.574
 Pole+1.553
 boy+1.537
ometer+1.492
Neg logit contributions
ck-1.967
aches-1.936
 avoid-1.842
 yourself-1.837
irs-1.824
Token to
Token ID284
Feature activation+4.01e+00

Pos logit contributions
ck+1.931
aches+1.865
irs+1.814
 avoid+1.728
 yourself+1.725
Neg logit contributions
sed-1.452
 girl-1.447
 boy-1.405
 Pole-1.352
ometer-1.313
Token he
Token ID339
Feature activation-6.67e-01

Pos logit contributions
ck+0.476
aches+0.463
irs+0.445
 yourself+0.435
 avoid+0.428
Neg logit contributions
sed-0.465
 girl-0.459
 boy-0.448
 Pole-0.435
ometer-0.420
Token had
Token ID550
Feature activation+8.80e-01

Pos logit contributions
sed+1.249
 girl+1.217
 Pole+1.188
 boy+1.175
ometer+1.120
Neg logit contributions
ck-1.301
aches-1.256
 avoid-1.170
 yourself-1.164
irs-1.164
Token to
Token ID284
Feature activation+3.33e+00

Pos logit contributions
ck+1.982
aches+1.910
irs+1.880
 yourself+1.800
 avoid+1.785
Neg logit contributions
sed-1.423
 girl-1.385
 boy-1.355
 Pole-1.270
ometer-1.227
Token do
Token ID466
Feature activation+2.08e+00

Pos logit contributions
irs+6.301
ck+5.667
cks+5.408
lled+5.229
aches+5.177
Neg logit contributions
 girl-7.416
 boy-7.174
sed-6.808
 mouse-6.450
 eyes-6.414
Token something
Token ID1223
Feature activation+5.40e-01

Pos logit contributions
ck+3.581
irs+3.351
aches+3.309
asting+2.991
 avoid+2.925
Neg logit contributions
 girl-4.458
 boy-4.381
sed-4.311
 eyes-4.091
 Pole-4.038
Token to
Token ID284
Feature activation+4.53e+00

Pos logit contributions
ck+1.024
aches+0.986
irs+0.960
 yourself+0.926
 avoid+0.910
Neg logit contributions
sed-1.030
 girl-1.025
 boy-1.000
 Pole-0.965
ometer-0.935
Token make
Token ID787
Feature activation+1.68e+00

Pos logit contributions
irs+10.163
lled+8.617
cks+8.563
ck+8.395
asting+7.717
Neg logit contributions
 girl-9.582
 boy-9.303
 mouse-8.573
sed-8.501
 cat-8.476
Token them
Token ID606
Feature activation+2.10e+00

Pos logit contributions
ck+2.587
aches+2.496
irs+2.348
 achieve+2.227
 avoid+2.216
Neg logit contributions
 girl-3.859
 boy-3.825
sed-3.713
 mouse-3.525
 Pole-3.524
Token happy
Token ID3772
Feature activation+5.08e-01

Pos logit contributions
ck+3.819
irs+3.746
aches+3.508
 yourself+3.289
 avoid+3.239
Neg logit contributions
 girl-4.377
 boy-4.241
sed-4.032
 Pole-3.796
ometer-3.784
Token.
Token ID13
Feature activation-1.31e+00

Pos logit contributions
ck+1.034
aches+1.002
irs+0.968
 yourself+0.934
 avoid+0.927
Neg logit contributions
sed-0.900
 girl-0.896
 boy-0.874
 Pole-0.840
ometer-0.810
Token
Token ID198
Feature activation+2.59e+00

Pos logit contributions
sed+2.569
 girl+2.442
 Pole+2.335
 boy+2.314
ometer+2.205
Neg logit contributions
ck-2.384
irs-2.324
 yourself-2.313
aches-2.297
 avoid-2.268
Token bad
Token ID2089
Feature activation-8.13e-01

Pos logit contributions
ck+2.191
aches+2.131
irs+1.972
 achieve+1.889
 avoid+1.865
Neg logit contributions
 girl-3.021
 boy-2.929
sed-2.855
 mouse-2.661
 Pole-2.645
Token behavior
Token ID4069
Feature activation+1.35e+00

Pos logit contributions
sed+1.635
 Pole+1.536
 girl+1.491
ometer+1.469
 boy+1.453
Neg logit contributions
ck-1.488
aches-1.479
 yourself-1.424
irs-1.374
 avoid-1.366
Token and
Token ID290
Feature activation+8.18e-01

Pos logit contributions
ck+3.099
aches+3.004
irs+2.892
 avoid+2.736
 yourself+2.736
Neg logit contributions
 girl-2.161
 boy-2.105
sed-2.090
 Pole-1.895
ometer-1.885
Token being
Token ID852
Feature activation+5.33e-01

Pos logit contributions
ck+1.437
aches+1.391
irs+1.341
 yourself+1.260
 avoid+1.250
Neg logit contributions
 girl-1.695
sed-1.694
 boy-1.656
 Pole-1.557
ometer-1.525
Token kind
Token ID1611
Feature activation-1.53e-01

Pos logit contributions
ck+1.113
aches+1.088
irs+1.059
 yourself+1.007
 achieve+1.006
Neg logit contributions
 girl-0.917
sed-0.912
 boy-0.889
 Pole-0.829
ometer-0.809
Token can
Token ID460
Feature activation+4.52e+00

Pos logit contributions
sed+0.269
 girl+0.262
 boy+0.254
 Pole+0.253
ometer+0.239
Neg logit contributions
ck-0.310
aches-0.302
irs-0.289
 yourself-0.288
 avoid-0.283
Token make
Token ID787
Feature activation+2.62e+00

Pos logit contributions
irs+9.359
cks+8.808
ck+8.766
e+8.546
lled+8.241
Neg logit contributions
 girl-9.958
 boy-9.849
 "-8.845
sed-8.719
 cat-8.672
Token a
Token ID257
Feature activation-6.99e-01

Pos logit contributions
ck+4.771
aches+4.641
irs+4.403
asting+4.286
 achieve+4.255
Neg logit contributions
 girl-5.748
 boy-5.577
sed-4.947
 mouse-4.840
 eyes-4.808
Token big
Token ID1263
Feature activation-1.96e+00

Pos logit contributions
sed+1.460
 Pole+1.372
 girl+1.367
 boy+1.327
ometer+1.312
Neg logit contributions
ck-1.226
aches-1.192
 avoid-1.137
 achieve-1.116
 yourself-1.111
Token difference
Token ID3580
Feature activation+9.11e-01

Pos logit contributions
sed+4.538
 Pole+4.402
ometer+4.302
elt+4.201
 girl+4.143
Neg logit contributions
ck-2.994
 avoid-2.815
aches-2.799
 achieve-2.680
 yourself-2.649
Token.
Token ID13
Feature activation-2.81e-02

Pos logit contributions
ck+1.799
aches+1.748
irs+1.692
 achieve+1.610
 avoid+1.603
Neg logit contributions
 girl-1.720
 boy-1.668
sed-1.667
 Pole-1.544
ometer-1.505
Token to
Token ID284
Feature activation+3.25e+00

Pos logit contributions
ck+1.714
aches+1.664
irs+1.621
 yourself+1.575
 achieve+1.572
Neg logit contributions
sed-1.195
 girl-1.179
 boy-1.149
 Pole-1.088
ometer-1.050
Token hurry
Token ID23290
Feature activation+1.26e+00

Pos logit contributions
ck+5.889
irs+5.714
lled+5.418
cks+5.307
aches+5.254
Neg logit contributions
 girl-7.146
 boy-7.076
sed-6.727
 mouse-6.461
 cat-6.452
Token to
Token ID284
Feature activation+4.30e+00

Pos logit contributions
ck+2.762
aches+2.662
irs+2.609
 achieve+2.469
 avoid+2.441
Neg logit contributions
 girl-2.134
 boy-2.090
sed-2.053
 Pole-1.964
ometer-1.898
Token the
Token ID262
Feature activation-4.38e-01

Pos logit contributions
ck+9.088
lled+8.653
aches+8.611
irs+8.469
cks+8.364
Neg logit contributions
 boy-9.028
 girl-8.991
 dog-8.169
 cart-8.106
 cat-7.990
Token lake
Token ID13546
Feature activation+7.62e-01

Pos logit contributions
sed+0.751
 girl+0.725
 boy+0.700
 Pole+0.693
ometer+0.675
Neg logit contributions
ck-0.903
aches-0.891
 yourself-0.851
irs-0.845
 avoid-0.837
Token to
Token ID284
Feature activation+4.52e+00

Pos logit contributions
ck+1.518
aches+1.458
irs+1.438
 yourself+1.371
 avoid+1.352
Neg logit contributions
 girl-1.390
sed-1.381
 boy-1.361
 Pole-1.297
ometer-1.257
Token swim
Token ID9422
Feature activation+8.65e-01

Pos logit contributions
irs+10.326
ck+10.240
lled+9.837
aches+9.441
cks+9.328
Neg logit contributions
 boy-9.505
 girl-9.353
 truck-8.400
 cart-8.382
 dog-8.312
Token.
Token ID13
Feature activation-1.13e+00

Pos logit contributions
ck+1.710
aches+1.668
irs+1.632
 yourself+1.542
 achieve+1.535
Neg logit contributions
 girl-1.608
sed-1.580
 boy-1.579
 Pole-1.490
ometer-1.450
Token
Token ID220
Feature activation+2.10e+00

Pos logit contributions
sed+2.250
 girl+2.141
 Pole+2.074
 boy+2.064
ometer+1.967
Neg logit contributions
ck-2.062
aches-1.993
irs-1.970
 yourself-1.930
 avoid-1.918
Token
Token ID198
Feature activation+2.28e+00

Pos logit contributions
ck+4.658
aches+4.488
 yourself+4.163
irs+4.151
 achieve+4.149
Neg logit contributions
 boy-3.531
 girl-3.521
sed-3.501
ometer-3.397
 Pole-3.323
Token
Token ID198
Feature activation+1.83e+00

Pos logit contributions
ck+5.074
aches+4.984
irs+4.733
 happen+4.729
 yourself+4.677
Neg logit contributions
 girl-3.917
 boy-3.872
sed-3.476
 Pole-3.425
ometer-3.381
Token and
Token ID290
Feature activation+5.22e-01

Pos logit contributions
ck+0.316
aches+0.307
irs+0.294
 yourself+0.288
 avoid+0.284
Neg logit contributions
sed-0.307
 girl-0.303
 boy-0.295
 Pole-0.287
ometer-0.277
Token kept
Token ID4030
Feature activation+1.32e+00

Pos logit contributions
ck+0.930
aches+0.904
irs+0.871
 yourself+0.842
 avoid+0.822
Neg logit contributions
 girl-1.055
sed-1.053
 boy-1.029
 Pole-0.991
ometer-0.958
Token hoping
Token ID7725
Feature activation+1.28e+00

Pos logit contributions
ck+2.344
aches+2.318
irs+2.197
 yourself+2.112
 avoid+2.110
Neg logit contributions
 girl-2.671
sed-2.642
 boy-2.594
 Pole-2.440
ometer-2.370
Token and
Token ID290
Feature activation+1.53e+00

Pos logit contributions
ck+2.751
aches+2.663
irs+2.613
 yourself+2.486
 achieve+2.432
Neg logit contributions
 girl-2.156
sed-2.087
 boy-2.079
 Pole-1.959
ometer-1.894
Token struggling
Token ID9648
Feature activation+1.22e+00

Pos logit contributions
ck+2.723
irs+2.644
aches+2.640
 yourself+2.422
asting+2.374
Neg logit contributions
 girl-3.187
sed-3.123
 boy-3.092
 Pole-2.887
ometer-2.798
Token to
Token ID284
Feature activation+4.52e+00

Pos logit contributions
ck+2.474
aches+2.425
irs+2.340
 achieve+2.207
 yourself+2.194
Neg logit contributions
 girl-2.203
sed-2.134
 boy-2.129
 Pole-1.966
ometer-1.950
Token believe
Token ID1975
Feature activation+6.78e-01

Pos logit contributions
irs+9.644
ck+8.586
cks+8.496
asting+8.441
aches+8.282
Neg logit contributions
 girl-9.959
 boy-9.489
sed-8.652
 cat-8.496
 mouse-8.468
Token they
Token ID484
Feature activation+1.67e+00

Pos logit contributions
ck+1.135
aches+1.100
irs+1.048
 yourself+1.000
 avoid+0.999
Neg logit contributions
 girl-1.454
sed-1.437
 boy-1.412
 Pole-1.358
ometer-1.321
Token would
Token ID561
Feature activation+3.57e+00

Pos logit contributions
ck+3.952
aches+3.921
irs+3.816
 yourself+3.740
 achieve+3.605
Neg logit contributions
 girl-2.545
 boy-2.448
sed-2.441
 Pole-2.194
ometer-2.082
Token.
Token ID13
Feature activation+3.47e-02

Pos logit contributions
ck+6.943
irs+6.939
aches+6.671
cks+6.198
asting+5.980
Neg logit contributions
 girl-7.397
sed-7.254
 boy-7.089
 "-6.636
 mouse-6.518
Token<|endoftext|>
Token ID50256
Feature activation+2.56e-01

Pos logit contributions
ck+0.063
aches+0.061
irs+0.058
 yourself+0.057
 avoid+0.056
Neg logit contributions
sed-0.069
 girl-0.068
 boy-0.067
 Pole-0.065
ometer-0.063
Token the
Token ID262
Feature activation-8.78e-01

Pos logit contributions
ck+2.269
aches+2.184
irs+2.074
 achieve+2.049
 avoid+1.954
Neg logit contributions
 girl-3.559
 boy-3.532
sed-3.529
 Pole-3.299
 eyes-3.237
Token world
Token ID995
Feature activation-9.38e-01

Pos logit contributions
sed+1.562
 Pole+1.458
 girl+1.434
 boy+1.411
ometer+1.402
Neg logit contributions
ck-1.774
aches-1.736
 yourself-1.705
irs-1.653
 avoid-1.642
Token and
Token ID290
Feature activation-3.37e-02

Pos logit contributions
sed+2.048
 girl+1.920
 boy+1.864
 Pole+1.850
ometer+1.804
Neg logit contributions
ck-1.608
aches-1.560
irs-1.466
 yourself-1.435
 achieve-1.424
Token wondering
Token ID11263
Feature activation+4.22e-01

Pos logit contributions
sed+0.069
 girl+0.068
 boy+0.067
 Pole+0.066
ometer+0.064
Neg logit contributions
ck-0.059
aches-0.057
irs-0.054
 yourself-0.053
 avoid-0.053
Token who
Token ID508
Feature activation+4.44e-01

Pos logit contributions
ck+0.590
aches+0.563
irs+0.534
 yourself+0.513
 avoid+0.504
Neg logit contributions
sed-1.013
 girl-1.009
 boy-0.989
 Pole-0.964
ometer-0.941
Token to
Token ID284
Feature activation+4.51e+00

Pos logit contributions
ck+0.793
aches+0.770
irs+0.737
 yourself+0.718
 avoid+0.706
Neg logit contributions
sed-0.896
 girl-0.895
 boy-0.874
 Pole-0.845
ometer-0.820
Token join
Token ID4654
Feature activation+1.86e+00

Pos logit contributions
irs+10.171
lled+8.855
ck+8.681
aches+8.607
inations+8.582
Neg logit contributions
 girl-9.977
 boy-9.937
sed-8.905
 cat-8.753
 dog-8.634
Token with
Token ID351
Feature activation+1.99e+00

Pos logit contributions
ck+3.334
aches+3.327
 achieve+3.090
 avoid+3.013
irs+3.007
Neg logit contributions
 girl-3.898
 boy-3.785
sed-3.658
 eyes-3.488
 mouse-3.462
Token.
Token ID13
Feature activation-1.88e+00

Pos logit contributions
ck+2.905
aches+2.822
 achieve+2.753
irs+2.631
 avoid+2.556
Neg logit contributions
 girl-4.637
 boy-4.566
sed-4.545
 eyes-4.312
 mouse-4.296
Token One
Token ID1881
Feature activation-1.12e+00

Pos logit contributions
sed+3.889
 girl+3.726
 boy+3.568
 Pole+3.476
ometer+3.448
Neg logit contributions
irs-3.224
 yourself-3.160
ck-3.153
aches-3.130
 ease-3.068
Token day
Token ID1110
Feature activation-2.15e+00

Pos logit contributions
sed+2.347
 Pole+2.187
 girl+2.111
ometer+2.064
 boy+2.034
Neg logit contributions
ck-2.042
aches-1.935
 avoid-1.851
asting-1.836
 yourself-1.772
Token,
Token ID11
Feature activation-7.41e-01

Pos logit contributions
sed+2.194
 girl+2.095
 Pole+2.087
 boy+2.048
ometer+1.958
Neg logit contributions
ck-1.953
aches-1.906
 avoid-1.768
irs-1.766
 yourself-1.752
Token excited
Token ID6568
Feature activation-1.66e+00

Pos logit contributions
sed+1.421
 girl+1.390
 Pole+1.363
 boy+1.354
ometer+1.332
Neg logit contributions
ck-1.401
aches-1.364
 avoid-1.289
 yourself-1.281
irs-1.267
Token for
Token ID329
Feature activation+1.29e+00

Pos logit contributions
sed+3.289
ometer+3.082
 Pole+3.060
 girl+3.060
 boy+3.014
Neg logit contributions
ck-3.117
aches-2.979
 yourself-2.936
 happen-2.853
irs-2.853
Token what
Token ID644
Feature activation-3.50e-01

Pos logit contributions
ck+1.919
aches+1.875
irs+1.779
 achieve+1.718
 yourself+1.702
Neg logit contributions
 girl-2.997
sed-2.995
 boy-2.904
 Pole-2.762
 mouse-2.735
Token was
Token ID373
Feature activation+1.48e+00

Pos logit contributions
sed+0.659
 girl+0.638
 Pole+0.629
 boy+0.621
ometer+0.598
Neg logit contributions
ck-0.677
aches-0.666
irs-0.625
 yourself-0.620
 avoid-0.615
Token to
Token ID284
Feature activation+4.50e+00

Pos logit contributions
ck+2.824
aches+2.707
irs+2.686
 avoid+2.596
 yourself+2.587
Neg logit contributions
 girl-2.832
sed-2.801
 boy-2.730
 Pole-2.678
ometer-2.504
Token come
Token ID1282
Feature activation+9.14e-01

Pos logit contributions
irs+10.827
ck+10.381
lled+10.213
asting+9.963
cks+9.541
Neg logit contributions
 girl-8.247
 boy-7.794
 "-7.511
 cat-7.354
sed-7.343
Token.
Token ID13
Feature activation-2.33e-01

Pos logit contributions
ck+2.120
aches+2.091
irs+2.008
 yourself+1.938
 avoid+1.915
Neg logit contributions
 girl-1.368
sed-1.351
 boy-1.313
 Pole-1.252
ometer-1.218
Token<|endoftext|>
Token ID50256
Feature activation+1.85e+00

Pos logit contributions
sed+0.443
 girl+0.434
 boy+0.423
 Pole+0.416
ometer+0.398
Neg logit contributions
ck-0.444
aches-0.431
irs-0.417
 yourself-0.406
 avoid-0.401
TokenOnce
Token ID7454
Feature activation+1.00e+00

Pos logit contributions
ck+4.169
aches+4.090
irs+3.811
 yourself+3.722
 achieve+3.664
Neg logit contributions
 boy-3.173
 girl-3.153
sed-2.978
 Pole-2.862
 mouse-2.725
Token upon
Token ID2402
Feature activation+2.51e+00

Pos logit contributions
ck+2.101
aches+2.039
 yourself+1.967
irs+1.952
 achieve+1.912
Neg logit contributions
 girl-1.742
sed-1.717
 boy-1.705
 Pole-1.606
ometer-1.557
Token he
Token ID339
Feature activation-7.50e-01

Pos logit contributions
ck+0.268
aches+0.260
irs+0.248
 yourself+0.243
 avoid+0.239
Neg logit contributions
sed-0.296
 girl-0.292
 boy-0.285
 Pole-0.279
ometer-0.269
Token needed
Token ID2622
Feature activation+1.28e+00

Pos logit contributions
sed+1.403
 girl+1.370
 Pole+1.339
 boy+1.319
ometer+1.256
Neg logit contributions
ck-1.472
aches-1.413
 avoid-1.322
 yourself-1.309
irs-1.307
Token to
Token ID284
Feature activation+3.19e+00

Pos logit contributions
ck+2.937
aches+2.818
irs+2.740
 achieve+2.656
 avoid+2.622
Neg logit contributions
sed-1.982
 girl-1.978
 boy-1.923
 Pole-1.821
ometer-1.760
Token do
Token ID466
Feature activation+2.17e+00

Pos logit contributions
irs+5.699
ck+5.364
aches+5.126
cks+4.900
lled+4.836
Neg logit contributions
 boy-6.990
 girl-6.953
sed-6.585
 Pole-6.334
 eyes-6.214
Token something
Token ID1223
Feature activation+1.01e+00

Pos logit contributions
ck+3.826
irs+3.595
aches+3.523
asting+3.188
 avoid+3.142
Neg logit contributions
 girl-4.535
 boy-4.475
sed-4.337
 Pole-4.179
 eyes-4.161
Token to
Token ID284
Feature activation+4.50e+00

Pos logit contributions
ck+2.026
aches+1.936
irs+1.924
 yourself+1.828
 avoid+1.788
Neg logit contributions
 girl-1.838
sed-1.801
 boy-1.795
 Pole-1.707
ometer-1.673
Token get
Token ID651
Feature activation+1.49e+00

Pos logit contributions
irs+9.455
lled+8.203
ck+8.177
cks+8.139
aches+7.954
Neg logit contributions
 boy-9.727
 girl-9.407
 truck-8.798
sed-8.751
 mouse-8.550
Token the
Token ID262
Feature activation-6.70e-01

Pos logit contributions
ck+2.889
aches+2.804
irs+2.758
 achieve+2.658
 yourself+2.594
Neg logit contributions
 girl-2.788
 boy-2.730
sed-2.707
 Pole-2.574
ometer-2.528
Token books
Token ID3835
Feature activation+1.69e+00

Pos logit contributions
sed+1.124
 girl+1.063
 boy+1.026
 Pole+1.022
ometer+0.993
Neg logit contributions
ck-1.423
aches-1.405
 yourself-1.334
irs-1.333
 avoid-1.326
Tokenhe
Token ID258
Feature activation+6.38e-02

Pos logit contributions
ck+3.862
aches+3.681
irs+3.650
 yourself+3.490
 avoid+3.407
Neg logit contributions
 girl-2.731
 boy-2.687
 Pole-2.552
sed-2.534
ometer-2.481
Tokenlf
Token ID1652
Feature activation+6.99e-01

Pos logit contributions
ck+0.120
aches+0.117
irs+0.111
 yourself+0.109
 avoid+0.108
Neg logit contributions
sed-0.124
 girl-0.121
 boy-0.118
 Pole-0.116
ometer-0.112
Token out
Token ID503
Feature activation+1.15e+00

Pos logit contributions
ck+2.019
aches+1.975
 achieve+1.857
irs+1.840
 yourself+1.807
Neg logit contributions
 girl-2.682
 boy-2.621
sed-2.588
 Pole-2.450
ometer-2.391
Token what
Token ID644
Feature activation-1.27e-01

Pos logit contributions
ck+1.879
aches+1.837
irs+1.744
 yourself+1.702
 achieve+1.670
Neg logit contributions
 girl-2.502
sed-2.438
 boy-2.434
 Pole-2.300
ometer-2.240
Token they
Token ID484
Feature activation+1.62e+00

Pos logit contributions
sed+0.235
 girl+0.229
 boy+0.221
 Pole+0.220
ometer+0.211
Neg logit contributions
ck-0.250
aches-0.244
irs-0.232
 yourself-0.228
 avoid-0.226
Token should
Token ID815
Feature activation+2.94e+00

Pos logit contributions
ck+3.939
aches+3.891
irs+3.818
 yourself+3.795
 achieve+3.644
Neg logit contributions
 girl-2.341
 boy-2.317
sed-2.240
 Pole-2.090
ometer-1.898
Token do
Token ID466
Feature activation+2.74e+00

Pos logit contributions
ck+5.098
irs+4.971
aches+4.645
asting+4.446
 yourself+4.436
Neg logit contributions
 boy-6.824
 girl-6.729
sed-6.504
 Pole-6.173
 "-5.932
Token to
Token ID284
Feature activation+4.50e+00

Pos logit contributions
ck+6.807
irs+6.535
aches+6.535
 ease+6.141
asting+5.983
Neg logit contributions
 girl-4.099
 boy-4.074
sed-3.763
 "-3.492
 Pole-3.483
Token become
Token ID1716
Feature activation+1.24e+00

Pos logit contributions
irs+9.360
ck+8.512
cks+8.118
lled+8.035
asting+7.570
Neg logit contributions
 boy-10.179
 girl-9.997
sed-8.818
 cat-8.747
 mouse-8.713
Token successful
Token ID4388
Feature activation-4.25e-02

Pos logit contributions
ck+2.108
aches+2.010
irs+1.980
 avoid+1.844
 achieve+1.840
Neg logit contributions
 girl-2.664
 boy-2.615
sed-2.577
 Pole-2.435
ometer-2.382
Token.
Token ID13
Feature activation-1.04e+00

Pos logit contributions
sed+0.083
 girl+0.081
 boy+0.079
 Pole+0.077
ometer+0.075
Neg logit contributions
ck-0.079
aches-0.077
irs-0.074
 yourself-0.073
 avoid-0.071
Token
Token ID198
Feature activation+2.61e+00

Pos logit contributions
sed+1.603
 girl+1.517
 boy+1.440
 Pole+1.439
ometer+1.357
Neg logit contributions
ck-2.350
irs-2.293
aches-2.264
 yourself-2.214
 avoid-2.205
Token
Token ID198
Feature activation+1.95e+00

Pos logit contributions
ck+5.835
aches+5.826
 yourself+5.600
 belong+5.481
 happen+5.448
Neg logit contributions
 boy-4.391
 girl-4.374
 Pole-3.900
sed-3.834
ometer-3.807
Token's
Token ID338
Feature activation+2.68e+00

Pos logit contributions
ck+2.967
aches+2.908
irs+2.812
 avoid+2.731
 achieve+2.730
Neg logit contributions
 girl-1.757
 boy-1.712
sed-1.612
 Pole-1.569
ometer-1.451
Token fly
Token ID6129
Feature activation+2.30e+00

Pos logit contributions
irs+4.672
ck+4.555
aches+4.304
asting+4.091
lled+3.950
Neg logit contributions
 girl-5.929
 boy-5.854
sed-5.666
 mouse-5.262
 eyes-5.176
Token this
Token ID428
Feature activation+8.13e-01

Pos logit contributions
ck+5.174
irs+5.045
aches+5.028
asting+4.628
 achieve+4.545
Neg logit contributions
 girl-3.876
 boy-3.791
sed-3.611
 Pole-3.414
ometer-3.334
Token important
Token ID1593
Feature activation+1.21e-01

Pos logit contributions
ck+1.827
aches+1.773
irs+1.739
 yourself+1.675
 avoid+1.661
Neg logit contributions
 girl-1.283
sed-1.260
 boy-1.244
 Pole-1.192
ometer-1.133
Token helicopter
Token ID15392
Feature activation+9.08e-01

Pos logit contributions
ck+0.241
aches+0.235
irs+0.225
 yourself+0.221
 avoid+0.218
Neg logit contributions
sed-0.221
 girl-0.217
 boy-0.212
 Pole-0.206
ometer-0.199
Token to
Token ID284
Feature activation+4.50e+00

Pos logit contributions
ck+1.506
aches+1.441
irs+1.413
 yourself+1.321
 avoid+1.305
Neg logit contributions
sed-1.962
 girl-1.961
 boy-1.921
 Pole-1.856
ometer-1.822
Token the
Token ID262
Feature activation+2.75e-01

Pos logit contributions
irs+9.553
ck+9.219
aches+9.101
asting+9.051
lled+8.912
Neg logit contributions
 girl-9.422
 boy-9.232
 mouse-8.743
sed-8.711
 "-8.489
Token moon
Token ID8824
Feature activation+7.31e-01

Pos logit contributions
ck+0.590
aches+0.576
irs+0.555
 yourself+0.542
 avoid+0.535
Neg logit contributions
sed-0.457
 girl-0.453
 boy-0.440
 Pole-0.425
ometer-0.406
Token."
Token ID526
Feature activation-1.33e+00

Pos logit contributions
ck+1.194
aches+1.151
irs+1.111
 yourself+1.053
 avoid+1.042
Neg logit contributions
 girl-1.590
sed-1.582
 boy-1.556
 Pole-1.512
ometer-1.467
Token
Token ID198
Feature activation+1.61e+00

Pos logit contributions
sed+2.351
 girl+2.223
 boy+2.084
ometer+2.068
 Pole+2.055
Neg logit contributions
ck-2.702
aches-2.648
 yourself-2.628
irs-2.621
 avoid-2.596
Token
Token ID198
Feature activation+9.35e-01

Pos logit contributions
ck+3.349
aches+3.297
irs+3.081
 yourself+3.080
 belong+2.986
Neg logit contributions
 boy-2.852
 girl-2.839
sed-2.729
 Pole-2.699
ometer-2.575
Token find
Token ID1064
Feature activation+9.15e-01

Pos logit contributions
ck+4.354
aches+4.344
irs+4.233
 yourself+3.917
cks+3.855
Neg logit contributions
 girl-5.369
 boy-5.129
sed-5.033
 mouse-4.640
 eyes-4.600
Token new
Token ID649
Feature activation-2.63e-01

Pos logit contributions
ck+1.509
aches+1.463
irs+1.363
 achieve+1.340
 yourself+1.330
Neg logit contributions
 girl-2.019
sed-1.959
 boy-1.957
 Pole-1.839
ometer-1.798
Token and
Token ID290
Feature activation+1.31e+00

Pos logit contributions
sed+0.527
 girl+0.508
 Pole+0.499
 boy+0.498
ometer+0.480
Neg logit contributions
ck-0.477
aches-0.465
irs-0.443
 yourself-0.440
 avoid-0.432
Token special
Token ID2041
Feature activation-4.05e-01

Pos logit contributions
ck+2.485
aches+2.404
irs+2.371
 yourself+2.233
 achieve+2.170
Neg logit contributions
 girl-2.589
 boy-2.511
sed-2.506
 Pole-2.302
ometer-2.253
Token ways
Token ID2842
Feature activation+1.53e+00

Pos logit contributions
sed+0.802
 girl+0.771
 Pole+0.760
 boy+0.757
ometer+0.728
Neg logit contributions
ck-0.744
aches-0.726
irs-0.692
 yourself-0.689
 avoid-0.677
Token to
Token ID284
Feature activation+4.49e+00

Pos logit contributions
ck+3.648
aches+3.563
irs+3.427
 yourself+3.320
 achieve+3.251
Neg logit contributions
 girl-2.246
sed-2.176
 boy-2.118
 Pole-1.993
ometer-1.951
Token wrap
Token ID14441
Feature activation+1.03e+00

Pos logit contributions
irs+9.091
cks+8.812
ck+8.790
aches+8.698
lled+8.503
Neg logit contributions
 girl-9.810
 boy-9.149
 "-8.593
sed-8.416
 eyes-8.322
Token gifts
Token ID13201
Feature activation+9.78e-01

Pos logit contributions
ck+1.519
aches+1.456
irs+1.363
 achieve+1.335
 avoid+1.310
Neg logit contributions
 girl-2.438
sed-2.381
 boy-2.368
 Pole-2.257
ometer-2.247
Token.
Token ID13
Feature activation-2.19e-01

Pos logit contributions
ck+2.098
aches+2.020
irs+1.953
 yourself+1.889
 avoid+1.883
Neg logit contributions
 girl-1.653
sed-1.625
 boy-1.592
 Pole-1.531
ometer-1.499
Token Everyone
Token ID11075
Feature activation-3.03e-01

Pos logit contributions
sed+0.385
 girl+0.375
 boy+0.365
 Pole+0.357
ometer+0.340
Neg logit contributions
ck-0.446
aches-0.435
irs-0.426
 yourself-0.415
 avoid-0.409
Token was
Token ID373
Feature activation+1.53e+00

Pos logit contributions
sed+0.551
 girl+0.527
 Pole+0.515
 boy+0.514
ometer+0.494
Neg logit contributions
ck-0.617
aches-0.596
irs-0.569
 yourself-0.560
 avoid-0.556
Token something
Token ID1223
Feature activation+4.73e-01

Pos logit contributions
ck+3.888
aches+3.723
irs+3.644
 achieve+3.520
 avoid+3.425
Neg logit contributions
 girl-2.428
sed-2.402
 boy-2.380
 Pole-2.212
 eyes-2.122
Token special
Token ID2041
Feature activation+7.28e-01

Pos logit contributions
ck+0.971
aches+0.940
irs+0.910
 yourself+0.884
 avoid+0.872
Neg logit contributions
sed-0.833
 girl-0.824
 boy-0.803
 Pole-0.776
ometer-0.750
Token to
Token ID284
Feature activation+3.79e+00

Pos logit contributions
ck+1.574
aches+1.519
irs+1.481
 yourself+1.425
 avoid+1.416
Neg logit contributions
sed-1.214
 girl-1.204
 boy-1.173
 Pole-1.121
ometer-1.088
Token be
Token ID307
Feature activation+1.87e+00

Pos logit contributions
ck+7.459
irs+7.220
lled+7.036
aches+6.747
asting+6.504
Neg logit contributions
 boy-8.181
 girl-8.170
 eyes-7.493
 mouse-7.312
 "-7.304
Token able
Token ID1498
Feature activation+1.06e+00

Pos logit contributions
ck+3.834
irs+3.748
aches+3.643
 avoid+3.415
 achieve+3.403
Neg logit contributions
 girl-3.442
 boy-3.348
sed-3.217
 Pole-3.085
ometer-3.001
Token to
Token ID284
Feature activation+4.49e+00

Pos logit contributions
ck+2.523
aches+2.429
irs+2.377
 achieve+2.285
 yourself+2.273
Neg logit contributions
sed-1.554
 girl-1.542
 boy-1.494
 Pole-1.417
ometer-1.374
Token go
Token ID467
Feature activation+2.58e+00

Pos logit contributions
irs+9.314
ck+9.198
cks+8.499
lled+8.270
aches+8.250
Neg logit contributions
 girl-9.527
 boy-9.491
 Pole-8.799
 cart-8.370
 eyes-8.321
Token on
Token ID319
Feature activation+2.40e+00

Pos logit contributions
ck+6.237
aches+6.031
irs+5.954
lled+5.478
 avoid+5.454
Neg logit contributions
 girl-4.076
 boy-3.951
 Pole-3.852
sed-3.463
 cart-3.330
Token any
Token ID597
Feature activation+1.12e+00

Pos logit contributions
ck+4.521
aches+4.327
irs+4.023
 avoid+3.790
asting+3.758
Neg logit contributions
 girl-4.945
 boy-4.921
 Pole-4.732
sed-4.680
 eyes-4.610
Token journey
Token ID7002
Feature activation+1.93e+00

Pos logit contributions
ck+2.036
aches+1.995
irs+1.905
 yourself+1.823
 avoid+1.793
Neg logit contributions
 girl-2.300
 boy-2.254
sed-2.167
 Pole-2.149
 mouse-1.982
Token.
Token ID13
Feature activation-1.87e-01

Pos logit contributions
ck+4.311
aches+4.116
irs+4.102
 avoid+3.785
 yourself+3.758
Neg logit contributions
 girl-3.202
 boy-3.179
sed-3.054
 Pole-3.039
ometer-2.866
Token her
Token ID607
Feature activation-1.50e+00

Pos logit contributions
ck+1.771
aches+1.657
irs+1.610
 achieve+1.588
 avoid+1.577
Neg logit contributions
 girl-2.528
sed-2.495
 boy-2.461
 Pole-2.317
 mouse-2.272
Token best
Token ID1266
Feature activation-1.09e+00

Pos logit contributions
 Pole+2.696
sed+2.662
 girl+2.600
 boy+2.527
ometer+2.522
Neg logit contributions
aches-2.915
ck-2.854
 yourself-2.670
 avoid-2.655
irs-2.619
Token friend
Token ID1545
Feature activation-1.70e+00

Pos logit contributions
sed+1.515
 Pole+1.448
 girl+1.417
 boy+1.365
ometer+1.304
Neg logit contributions
aches-2.597
ck-2.581
 avoid-2.431
irs-2.428
 yourself-2.398
Token,
Token ID11
Feature activation-4.65e-01

Pos logit contributions
sed+2.905
 Pole+2.717
 girl+2.643
ometer+2.638
 boy+2.539
Neg logit contributions
ck-3.664
aches-3.622
 yourself-3.514
e-3.426
irs-3.410
Token a
Token ID257
Feature activation-2.90e+00

Pos logit contributions
sed+1.124
 girl+1.097
 Pole+1.074
 boy+1.071
ometer+1.057
Neg logit contributions
ck-0.640
aches-0.620
irs-0.577
 yourself-0.571
 avoid-0.555
Token little
Token ID1310
Feature activation-4.60e+00

Pos logit contributions
sed+6.077
elt+5.688
ometer+5.684
 Pole+5.629
board+5.369
Neg logit contributions
aches-5.004
 avoid-4.688
 achieve-4.670
ck-4.599
 yourself-4.561
Token cat
Token ID3797
Feature activation-1.05e+00

Pos logit contributions
sed+7.240
 Pole+6.985
ometer+6.096
icol+5.805
board+5.648
Neg logit contributions
 know-11.338
 belong-10.983
 yourself-10.951
 enjoy-10.928
ck-10.781
Token.
Token ID13
Feature activation-1.27e+00

Pos logit contributions
sed+2.034
 girl+1.937
 Pole+1.914
 boy+1.881
ometer+1.830
Neg logit contributions
ck-2.010
aches-1.941
 yourself-1.821
irs-1.803
 avoid-1.800
Token One
Token ID1881
Feature activation-1.08e+00

Pos logit contributions
sed+2.541
 girl+2.399
 boy+2.308
 Pole+2.307
ometer+2.282
Neg logit contributions
ck-2.233
irs-2.213
aches-2.213
 yourself-2.128
 avoid-2.117
Token sunny
Token ID27737
Feature activation-1.27e+00

Pos logit contributions
sed+2.217
 Pole+2.080
 girl+1.996
ometer+1.958
 boy+1.941
Neg logit contributions
ck-1.963
aches-1.931
 avoid-1.796
asting-1.772
 yourself-1.718
Token day
Token ID1110
Feature activation-2.11e+00

Pos logit contributions
sed+2.686
 Pole+2.482
 girl+2.409
ometer+2.407
 boy+2.354
Neg logit contributions
aches-2.220
ck-2.172
 yourself-2.044
 avoid-2.012
asting-1.954
Token a
Token ID257
Feature activation-3.00e+00

Pos logit contributions
ck+0.729
aches+0.698
irs+0.661
 yourself+0.630
 avoid+0.626
Neg logit contributions
 girl-1.142
sed-1.141
 boy-1.120
 Pole-1.085
ometer-1.051
Token mom
Token ID1995
Feature activation-2.45e-01

Pos logit contributions
sed+7.292
elt+6.795
 Pole+6.371
ometer+6.370
 cheeks+6.221
Neg logit contributions
 avoid-4.588
 achieve-4.502
aches-4.500
ck-4.389
 yourself-4.193
Tokenmy
Token ID1820
Feature activation-1.79e+00

Pos logit contributions
sed+0.470
 girl+0.448
 boy+0.437
 Pole+0.432
ometer+0.421
Neg logit contributions
ck-0.475
aches-0.462
irs-0.439
 yourself-0.435
 avoid-0.426
Token and
Token ID290
Feature activation-5.98e-01

Pos logit contributions
sed+3.347
 girl+2.967
 Pole+2.967
ometer+2.919
 boy+2.853
Neg logit contributions
ck-3.655
aches-3.533
 yourself-3.339
asting-3.316
irs-3.275
Token a
Token ID257
Feature activation-3.10e+00

Pos logit contributions
sed+1.103
 girl+1.056
 Pole+1.044
 boy+1.026
ometer+1.017
Neg logit contributions
ck-1.174
aches-1.143
irs-1.091
 yourself-1.082
 avoid-1.051
Token little
Token ID1310
Feature activation-4.59e+00

Pos logit contributions
sed+6.614
elt+6.103
 Pole+6.093
ometer+5.883
board+5.798
Neg logit contributions
aches-5.396
 avoid-5.281
 achieve-5.093
ck-5.092
 know-5.049
Token girl
Token ID2576
Feature activation-9.07e-01

Pos logit contributions
sed+6.529
 Pole+6.026
board+5.107
ometer+4.932
elt+4.612
Neg logit contributions
 know-12.596
 enjoy-12.041
ck-12.031
 belong-11.843
 avoid-11.833
Token.
Token ID13
Feature activation-1.56e+00

Pos logit contributions
sed+1.827
 Pole+1.739
 girl+1.715
 boy+1.687
ometer+1.660
Neg logit contributions
ck-1.674
aches-1.606
 yourself-1.518
irs-1.511
 avoid-1.493
Token The
Token ID383
Feature activation-2.01e+00

Pos logit contributions
sed+3.617
 girl+3.373
ometer+3.304
 boy+3.276
 Pole+3.267
Neg logit contributions
irs-2.316
ck-2.280
aches-2.220
 yourself-2.184
 avoid-2.129
Token mom
Token ID1995
Feature activation-7.11e-01

Pos logit contributions
sed+3.239
 Pole+2.791
ometer+2.767
elt+2.753
hog+2.527
Neg logit contributions
aches-4.470
ck-4.454
 avoid-4.338
 yourself-4.217
asting-4.187
Tokenmy
Token ID1820
Feature activation-1.17e+00

Pos logit contributions
sed+1.606
 Pole+1.486
 girl+1.482
 boy+1.443
ometer+1.438
Neg logit contributions
ck-1.168
aches-1.130
 yourself-1.043
irs-1.038
 avoid-1.014
Token a
Token ID257
Feature activation-3.00e+00

Pos logit contributions
ck+0.729
aches+0.698
irs+0.661
 yourself+0.630
 avoid+0.626
Neg logit contributions
 girl-1.142
sed-1.141
 boy-1.120
 Pole-1.085
ometer-1.051
Token mom
Token ID1995
Feature activation-2.45e-01

Pos logit contributions
sed+7.292
elt+6.795
 Pole+6.371
ometer+6.370
 cheeks+6.221
Neg logit contributions
 avoid-4.588
 achieve-4.502
aches-4.500
ck-4.389
 yourself-4.193
Tokenmy
Token ID1820
Feature activation-1.79e+00

Pos logit contributions
sed+0.470
 girl+0.448
 boy+0.437
 Pole+0.432
ometer+0.421
Neg logit contributions
ck-0.475
aches-0.462
irs-0.439
 yourself-0.435
 avoid-0.426
Token and
Token ID290
Feature activation-5.98e-01

Pos logit contributions
sed+3.347
 girl+2.967
 Pole+2.967
ometer+2.919
 boy+2.853
Neg logit contributions
ck-3.655
aches-3.533
 yourself-3.339
asting-3.316
irs-3.275
Token a
Token ID257
Feature activation-3.10e+00

Pos logit contributions
sed+1.103
 girl+1.056
 Pole+1.044
 boy+1.026
ometer+1.017
Neg logit contributions
ck-1.174
aches-1.143
irs-1.091
 yourself-1.082
 avoid-1.051
Token little
Token ID1310
Feature activation-4.59e+00

Pos logit contributions
sed+6.614
elt+6.103
 Pole+6.093
ometer+5.883
board+5.798
Neg logit contributions
aches-5.396
 avoid-5.281
 achieve-5.093
ck-5.092
 know-5.049
Token girl
Token ID2576
Feature activation-9.07e-01

Pos logit contributions
sed+6.529
 Pole+6.026
board+5.107
ometer+4.932
elt+4.612
Neg logit contributions
 know-12.596
 enjoy-12.041
ck-12.031
 belong-11.843
 avoid-11.833
Token.
Token ID13
Feature activation-1.56e+00

Pos logit contributions
sed+1.827
 Pole+1.739
 girl+1.715
 boy+1.687
ometer+1.660
Neg logit contributions
ck-1.674
aches-1.606
 yourself-1.518
irs-1.511
 avoid-1.493
Token The
Token ID383
Feature activation-2.01e+00

Pos logit contributions
sed+3.617
 girl+3.373
ometer+3.304
 boy+3.276
 Pole+3.267
Neg logit contributions
irs-2.316
ck-2.280
aches-2.220
 yourself-2.184
 avoid-2.129
Token mom
Token ID1995
Feature activation-7.11e-01

Pos logit contributions
sed+3.239
 Pole+2.791
ometer+2.767
elt+2.753
hog+2.527
Neg logit contributions
aches-4.470
ck-4.454
 avoid-4.338
 yourself-4.217
asting-4.187
Tokenmy
Token ID1820
Feature activation-1.17e+00

Pos logit contributions
sed+1.606
 Pole+1.486
 girl+1.482
 boy+1.443
ometer+1.438
Neg logit contributions
ck-1.168
aches-1.130
 yourself-1.043
irs-1.038
 avoid-1.014
Token.
Token ID13
Feature activation-7.94e-01

Pos logit contributions
sed+0.838
 girl+0.822
 boy+0.798
 Pole+0.794
ometer+0.763
Neg logit contributions
ck-0.684
aches-0.656
irs-0.616
 yourself-0.613
 avoid-0.610
Token He
Token ID679
Feature activation-8.07e-01

Pos logit contributions
sed+1.082
 girl+1.046
 Pole+0.997
 boy+0.984
ometer+0.945
Neg logit contributions
ck-1.891
aches-1.841
 yourself-1.821
irs-1.821
 avoid-1.804
Token was
Token ID373
Feature activation-9.44e-01

Pos logit contributions
sed+1.509
 girl+1.471
 Pole+1.434
 boy+1.404
ometer+1.380
Neg logit contributions
ck-1.569
aches-1.528
 avoid-1.433
 yourself-1.424
 achieve-1.407
Token a
Token ID257
Feature activation-2.95e+00

Pos logit contributions
sed+2.058
 Pole+1.998
 girl+1.990
 boy+1.927
ometer+1.913
Neg logit contributions
ck-1.530
aches-1.479
 yourself-1.400
 avoid-1.380
 achieve-1.345
Token curious
Token ID11040
Feature activation-4.26e+00

Pos logit contributions
sed+6.862
 Pole+6.446
ometer+6.123
 eyes+6.119
board+6.113
Neg logit contributions
 avoid-4.555
ck-4.399
 achieve-4.152
aches-4.120
e-4.083
Token little
Token ID1310
Feature activation-4.59e+00

Pos logit contributions
sed+8.210
ometer+6.719
 Pole+6.410
elt+6.005
hog+5.990
Neg logit contributions
 know-9.461
 avoid-9.128
asting-9.127
 yourself-8.895
 appreciate-8.865
Token boy
Token ID2933
Feature activation-1.06e+00

Pos logit contributions
sed+8.257
 Pole+6.807
 cheeks+6.251
omm+6.035
icol+6.013
Neg logit contributions
 know-10.995
 avoid-10.936
 belong-10.830
 enjoy-10.565
 yourself-10.410
Token and
Token ID290
Feature activation-4.68e-01

Pos logit contributions
sed+2.230
 girl+2.111
 Pole+2.103
 boy+2.037
ometer+1.971
Neg logit contributions
ck-1.911
aches-1.796
 yourself-1.723
 avoid-1.721
e-1.713
Token loved
Token ID6151
Feature activation-3.76e-01

Pos logit contributions
sed+0.880
 girl+0.852
 Pole+0.836
 boy+0.824
ometer+0.800
Neg logit contributions
ck-0.906
aches-0.888
 yourself-0.829
 avoid-0.826
irs-0.819
Token going
Token ID1016
Feature activation+6.41e-01

Pos logit contributions
sed+0.696
 girl+0.677
 boy+0.653
 Pole+0.646
ometer+0.625
Neg logit contributions
ck-0.739
aches-0.723
irs-0.693
 yourself-0.684
 avoid-0.677
Token on
Token ID319
Feature activation+1.28e+00

Pos logit contributions
ck+1.332
aches+1.287
irs+1.240
 yourself+1.202
 avoid+1.186
Neg logit contributions
 girl-1.122
sed-1.113
 boy-1.092
 Pole-1.059
ometer-1.016
Token hold
Token ID1745
Feature activation+3.04e-01

Pos logit contributions
ck+4.587
irs+4.453
aches+4.380
 yourself+4.108
asting+4.102
Neg logit contributions
 girl-6.002
 boy-5.925
sed-5.712
 eyes-5.463
 mouse-5.388
Token it
Token ID340
Feature activation-4.49e-01

Pos logit contributions
ck+0.487
aches+0.471
irs+0.445
 yourself+0.435
 avoid+0.429
Neg logit contributions
sed-0.672
 girl-0.668
 boy-0.653
 Pole-0.636
ometer-0.621
Token up
Token ID510
Feature activation+4.85e-01

Pos logit contributions
sed+0.889
 girl+0.863
 boy+0.841
 Pole+0.832
ometer+0.786
Neg logit contributions
ck-0.831
aches-0.809
irs-0.764
 yourself-0.757
 avoid-0.745
Token.
Token ID13
Feature activation-1.58e+00

Pos logit contributions
ck+0.976
aches+0.953
irs+0.915
 yourself+0.890
 avoid+0.878
Neg logit contributions
sed-0.873
 girl-0.867
 boy-0.844
 Pole-0.809
ometer-0.790
Token A
Token ID317
Feature activation-3.32e+00

Pos logit contributions
sed+3.197
 girl+3.031
 boy+2.894
 Pole+2.860
ometer+2.829
Neg logit contributions
irs-2.773
aches-2.773
ck-2.772
 yourself-2.712
 avoid-2.596
Token little
Token ID1310
Feature activation-4.59e+00

Pos logit contributions
sed+6.589
elt+6.028
ometer+5.456
 Pole+5.381
hog+5.179
Neg logit contributions
 avoid-6.614
 achieve-6.604
aches-6.563
ck-6.525
asting-6.316
Token girl
Token ID2576
Feature activation-1.59e+00

Pos logit contributions
sed+7.226
 Pole+6.136
ometer+5.648
elt+5.616
 cheeks+5.241
Neg logit contributions
e-12.038
ck-11.890
 yourself-11.652
 belong-11.566
 know-11.504
Token named
Token ID3706
Feature activation+2.97e-03

Pos logit contributions
sed+3.125
 Pole+3.016
 girl+2.890
ometer+2.884
 boy+2.837
Neg logit contributions
ck-3.093
aches-2.883
e-2.820
 avoid-2.722
 yourself-2.708
Token Lily
Token ID20037
Feature activation-5.68e-01

Pos logit contributions
ck+0.007
aches+0.007
irs+0.007
 yourself+0.007
 avoid+0.007
Neg logit contributions
sed-0.004
 girl-0.004
 boy-0.004
 Pole-0.004
ometer-0.004
Token loved
Token ID6151
Feature activation-6.25e-01

Pos logit contributions
sed+0.964
 girl+0.921
 Pole+0.892
 boy+0.888
ometer+0.860
Neg logit contributions
ck-1.223
aches-1.184
 yourself-1.121
irs-1.118
 avoid-1.111
Token to
Token ID284
Feature activation+1.58e+00

Pos logit contributions
sed+1.271
 girl+1.225
 Pole+1.191
 boy+1.190
ometer+1.159
Neg logit contributions
ck-1.123
aches-1.105
irs-1.050
 yourself-1.029
 avoid-1.004
Token,
Token ID11
Feature activation-1.29e+00

Pos logit contributions
sed+3.721
 girl+3.482
 Pole+3.439
ometer+3.367
 boy+3.317
Neg logit contributions
ck-4.180
aches-4.162
 avoid-3.959
 yourself-3.924
e-3.885
Token the
Token ID262
Feature activation-1.83e+00

Pos logit contributions
sed+3.788
 Pole+3.580
ometer+3.557
 girl+3.533
 boy+3.429
Neg logit contributions
ck-1.202
aches-1.168
 avoid-1.065
 yourself-1.031
irs-0.956
Token elephant
Token ID20950
Feature activation-1.27e+00

Pos logit contributions
sed+1.984
ometer+1.873
 Pole+1.743
 girl+1.691
elt+1.594
Neg logit contributions
ck-4.977
aches-4.924
 yourself-4.744
 avoid-4.668
 ease-4.587
Token met
Token ID1138
Feature activation+2.07e-01

Pos logit contributions
sed+2.416
 girl+2.249
 Pole+2.218
ometer+2.154
 boy+2.120
Neg logit contributions
ck-2.557
aches-2.475
 avoid-2.284
 yourself-2.243
 achieve-2.242
Token a
Token ID257
Feature activation-2.81e+00

Pos logit contributions
ck+0.283
aches+0.272
irs+0.256
 yourself+0.248
 avoid+0.243
Neg logit contributions
sed-0.508
 girl-0.503
 boy-0.493
 Pole-0.482
ometer-0.470
Token little
Token ID1310
Feature activation-4.54e+00

Pos logit contributions
sed+5.768
elt+5.325
 Pole+5.208
ometer+5.116
umm+5.067
Neg logit contributions
ck-5.152
aches-5.146
 avoid-5.095
 achieve-4.968
 know-4.887
Token boy
Token ID2933
Feature activation-7.22e-01

Pos logit contributions
sed+8.302
 Pole+6.613
ometer+6.317
 cheeks+6.304
umm+6.180
Neg logit contributions
ck-11.462
 know-11.063
 belong-10.745
 avoid-10.633
 yourself-10.530
Token.
Token ID13
Feature activation-1.96e+00

Pos logit contributions
sed+1.340
 girl+1.270
 Pole+1.248
 boy+1.228
ometer+1.199
Neg logit contributions
ck-1.487
aches-1.408
irs-1.331
 yourself-1.329
 avoid-1.324
Token The
Token ID383
Feature activation-1.90e+00

Pos logit contributions
sed+3.869
 girl+3.486
ometer+3.316
 Pole+3.308
 boy+3.188
Neg logit contributions
ck-3.598
aches-3.565
 avoid-3.513
irs-3.510
 know-3.501
Token boy
Token ID2933
Feature activation-2.10e+00

Pos logit contributions
sed+2.172
ometer+1.960
 Pole+1.830
 girl+1.717
elt+1.711
Neg logit contributions
ck-5.138
aches-5.039
 avoid-4.825
 yourself-4.765
 know-4.703
Token asked
Token ID1965
Feature activation-2.17e-01

Pos logit contributions
sed+4.157
 Pole+4.088
umm+3.972
ometer+3.908
 girl+3.884
Neg logit contributions
ck-4.112
aches-3.862
 avoid-3.533
 know-3.505
e-3.449
Token.
Token ID13
Feature activation-1.25e+00

Pos logit contributions
sed+1.275
 girl+1.255
 boy+1.221
 Pole+1.206
ometer+1.160
Neg logit contributions
ck-1.033
aches-1.023
irs-0.956
 yourself-0.937
 avoid-0.910
Token
Token ID220
Feature activation+2.29e+00

Pos logit contributions
sed+2.512
 girl+2.364
 boy+2.271
 Pole+2.249
ometer+2.241
Neg logit contributions
ck-2.237
aches-2.196
irs-2.169
 yourself-2.106
 avoid-2.078
Token
Token ID198
Feature activation+2.29e+00

Pos logit contributions
ck+5.302
aches+5.102
 yourself+4.773
irs+4.738
 achieve+4.716
Neg logit contributions
 boy-3.801
 girl-3.767
sed-3.575
ometer-3.513
 "-3.442
Token
Token ID198
Feature activation+3.42e-01

Pos logit contributions
ck+5.274
aches+5.195
 yourself+4.914
irs+4.853
 belong+4.697
Neg logit contributions
 boy-3.791
 girl-3.779
 Pole-3.413
sed-3.287
 "-3.266
TokenA
Token ID32
Feature activation-3.37e+00

Pos logit contributions
ck+0.334
aches+0.315
irs+0.290
 yourself+0.275
 avoid+0.265
Neg logit contributions
sed-0.968
 girl-0.968
 boy-0.951
 Pole-0.932
ometer-0.908
Token little
Token ID1310
Feature activation-4.52e+00

Pos logit contributions
sed+6.723
elt+5.654
 eyes+5.420
ometer+5.411
 Pole+5.330
Neg logit contributions
ck-6.941
aches-6.820
 avoid-6.770
asting-6.750
 know-6.569
Token boy
Token ID2933
Feature activation-1.08e+00

Pos logit contributions
sed+6.092
 Pole+5.348
ometer+4.885
 cheeks+4.309
icol+4.264
Neg logit contributions
e-12.612
ck-12.544
 belong-12.538
 have-12.535
 know-12.472
Token was
Token ID373
Feature activation-9.87e-01

Pos logit contributions
sed+2.067
 Pole+1.935
 girl+1.918
ometer+1.900
 boy+1.852
Neg logit contributions
ck-2.195
aches-2.052
e-1.981
 avoid-1.964
 yourself-1.948
Token playing
Token ID2712
Feature activation+1.00e-01

Pos logit contributions
sed+1.942
 Pole+1.929
 girl+1.928
 boy+1.857
ometer+1.843
Neg logit contributions
ck-1.800
aches-1.772
 yourself-1.664
 avoid-1.603
 achieve-1.582
Token in
Token ID287
Feature activation+1.74e+00

Pos logit contributions
ck+0.188
aches+0.183
irs+0.175
 yourself+0.172
 avoid+0.169
Neg logit contributions
sed-0.194
 girl-0.191
 boy-0.186
 Pole-0.182
ometer-0.176
Token the
Token ID262
Feature activation-8.53e-01

Pos logit contributions
ck+2.807
aches+2.735
irs+2.587
 achieve+2.463
asting+2.366
Neg logit contributions
 boy-4.110
 girl-4.086
sed-3.839
 Pole-3.684
 mouse-3.668
Token mouse
Token ID10211
Feature activation-1.07e+00

Pos logit contributions
sed+4.326
ometer+3.759
 Pole+3.641
 girl+3.278
elt+3.267
Neg logit contributions
ck-6.213
 yourself-6.183
aches-6.034
asting-5.935
 avoid-5.930
Token was
Token ID373
Feature activation-1.68e-01

Pos logit contributions
sed+1.941
 girl+1.863
 Pole+1.819
 boy+1.803
ometer+1.791
Neg logit contributions
ck-2.162
aches-2.122
 yourself-1.972
 avoid-1.964
irs-1.950
Token a
Token ID257
Feature activation-2.98e+00

Pos logit contributions
sed+0.314
 girl+0.309
 boy+0.300
 Pole+0.297
ometer+0.285
Neg logit contributions
ck-0.326
aches-0.317
irs-0.300
 yourself-0.299
 avoid-0.293
Token very
Token ID845
Feature activation-1.47e+00

Pos logit contributions
sed+6.145
 Pole+5.558
elt+5.422
ometer+5.391
board+5.261
Neg logit contributions
ck-5.553
 avoid-5.518
 achieve-5.240
e-5.179
aches-5.147
Token curious
Token ID11040
Feature activation-4.05e+00

Pos logit contributions
sed+2.770
 Pole+2.606
 girl+2.564
 boy+2.508
ometer+2.498
Neg logit contributions
ck-2.912
aches-2.843
 avoid-2.657
 yourself-2.596
 achieve-2.565
Token little
Token ID1310
Feature activation-4.51e+00

Pos logit contributions
sed+8.396
ometer+6.913
 Pole+6.822
elt+6.302
ses+6.142
Neg logit contributions
 know-8.465
asting-8.269
 avoid-8.034
ck-8.029
 yourself-7.975
Token creature
Token ID7185
Feature activation-1.57e+00

Pos logit contributions
sed+8.028
 Pole+6.704
ometer+5.973
board+5.650
 cheeks+5.419
Neg logit contributions
 yourself-10.864
 avoid-10.773
ck-10.610
 know-10.572
 enjoy-10.403
Token.
Token ID13
Feature activation-1.39e+00

Pos logit contributions
sed+3.550
 girl+3.393
 Pole+3.388
 boy+3.327
ometer+3.256
Neg logit contributions
ck-2.525
aches-2.437
 yourself-2.219
 avoid-2.218
e-2.188
Token His
Token ID2399
Feature activation-2.23e+00

Pos logit contributions
sed+2.463
 girl+2.326
 Pole+2.274
 boy+2.210
ometer+2.151
Neg logit contributions
ck-2.791
 yourself-2.721
aches-2.720
irs-2.678
 avoid-2.677
Token home
Token ID1363
Feature activation-1.17e+00

Pos logit contributions
 Pole+3.969
 girl+3.911
 boy+3.773
sed+3.743
elt+3.508
Neg logit contributions
ck-4.339
aches-4.175
 yourself-4.079
 avoid-4.076
 know-4.033
Token was
Token ID373
Feature activation+2.70e-01

Pos logit contributions
sed+2.160
 girl+2.078
 Pole+2.026
 boy+1.983
ometer+1.908
Neg logit contributions
ck-2.278
aches-2.277
 avoid-2.092
 yourself-2.089
 achieve-2.051
Token scary
Token ID14343
Feature activation-1.89e+00

Pos logit contributions
sed+2.013
 girl+1.957
 Pole+1.943
 boy+1.882
ometer+1.796
Neg logit contributions
ck-2.436
aches-2.360
 yourself-2.195
irs-2.183
 avoid-2.166
Token there
Token ID612
Feature activation+7.02e-01

Pos logit contributions
sed+4.081
 girl+3.878
 Pole+3.811
 boy+3.705
ometer+3.698
Neg logit contributions
 yourself-3.120
ck-3.091
aches-2.979
 happen-2.868
 avoid-2.855
Token.
Token ID13
Feature activation-1.19e+00

Pos logit contributions
ck+1.475
aches+1.425
irs+1.388
 yourself+1.333
 avoid+1.318
Neg logit contributions
sed-1.213
 girl-1.199
 boy-1.173
 Pole-1.122
ometer-1.084
Token A
Token ID317
Feature activation-3.67e+00

Pos logit contributions
sed+2.070
 girl+1.933
 boy+1.834
 Pole+1.826
ometer+1.797
Neg logit contributions
ck-2.458
aches-2.447
 yourself-2.441
irs-2.399
 avoid-2.331
Token brave
Token ID14802
Feature activation-3.54e+00

Pos logit contributions
sed+7.264
elt+6.810
 Pole+6.569
umm+6.244
ometer+6.162
Neg logit contributions
aches-7.038
ck-6.957
 avoid-6.903
 know-6.745
 achieve-6.649
Token little
Token ID1310
Feature activation-4.49e+00

Pos logit contributions
sed+5.306
elt+4.955
 Pole+4.711
ometer+4.634
�+3.970
Neg logit contributions
ck-8.797
aches-8.299
e-8.240
 belong-8.024
 know-7.986
Token boy
Token ID2933
Feature activation-1.47e+00

Pos logit contributions
sed+6.707
 Pole+5.969
elt+5.263
umm+5.058
ometer+5.042
Neg logit contributions
 yourself-11.604
 belong-11.581
ck-11.549
 avoid-11.328
 know-11.282
Token went
Token ID1816
Feature activation-3.55e-01

Pos logit contributions
sed+2.847
 Pole+2.681
 girl+2.663
ometer+2.595
 boy+2.556
Neg logit contributions
ck-2.926
aches-2.754
e-2.590
 yourself-2.582
 avoid-2.579
Token into
Token ID656
Feature activation+1.19e+00

Pos logit contributions
sed+0.673
 girl+0.650
 boy+0.634
 Pole+0.622
ometer+0.605
Neg logit contributions
ck-0.689
aches-0.672
irs-0.638
 yourself-0.636
 avoid-0.627
Token the
Token ID262
Feature activation-2.46e+00

Pos logit contributions
ck+1.736
aches+1.718
irs+1.569
 yourself+1.490
 achieve+1.461
Neg logit contributions
 girl-2.905
 boy-2.880
sed-2.798
 Pole-2.685
 mouse-2.635
Token dark
Token ID3223
Feature activation-1.25e+00

Pos logit contributions
sed+4.461
 Pole+3.879
ometer+3.756
 girl+3.632
 boy+3.420
Neg logit contributions
 yourself-5.347
 avoid-5.096
 achieve-5.014
e-4.941
ck-4.941
Token her
Token ID607
Feature activation-1.35e+00

Pos logit contributions
ck+1.982
aches+1.868
irs+1.798
 achieve+1.779
 avoid+1.703
Neg logit contributions
 girl-3.111
sed-3.102
 boy-3.044
 mouse-2.840
 Pole-2.822
Token best
Token ID1266
Feature activation-1.04e+00

Pos logit contributions
 Pole+2.171
 girl+2.085
sed+2.069
 boy+2.001
ometer+1.910
Neg logit contributions
ck-2.855
aches-2.828
 avoid-2.726
 yourself-2.694
 ourselves-2.632
Token friend
Token ID1545
Feature activation-1.57e+00

Pos logit contributions
sed+1.444
 Pole+1.402
 girl+1.363
 boy+1.309
ometer+1.250
Neg logit contributions
ck-2.470
aches-2.460
 avoid-2.343
irs-2.320
 yourself-2.315
Token,
Token ID11
Feature activation-3.35e-01

Pos logit contributions
sed+2.575
 Pole+2.489
 girl+2.406
ometer+2.370
 boy+2.305
Neg logit contributions
ck-3.470
aches-3.394
 yourself-3.308
e-3.215
irs-3.183
Token a
Token ID257
Feature activation-3.06e+00

Pos logit contributions
sed+0.707
 girl+0.692
 Pole+0.674
 boy+0.673
ometer+0.657
Neg logit contributions
ck-0.563
aches-0.545
irs-0.514
 yourself-0.510
 avoid-0.504
Token little
Token ID1310
Feature activation-4.49e+00

Pos logit contributions
sed+6.165
 Pole+5.948
elt+5.712
ometer+5.690
board+5.669
Neg logit contributions
 avoid-5.374
aches-5.181
 achieve-5.000
 know-4.986
 yourself-4.977
Token girl
Token ID2576
Feature activation-1.06e+00

Pos logit contributions
 Pole+6.782
sed+6.467
ometer+5.575
board+5.275
icol+5.204
Neg logit contributions
 know-11.487
 avoid-11.305
 yourself-11.153
 belong-11.091
ck-11.057
Token named
Token ID3706
Feature activation-8.03e-01

Pos logit contributions
sed+2.009
 Pole+1.929
 girl+1.849
ometer+1.824
 boy+1.822
Neg logit contributions
ck-2.095
aches-1.977
 yourself-1.898
 avoid-1.887
e-1.880
Token Mia
Token ID32189
Feature activation+5.24e-01

Pos logit contributions
sed+1.234
 girl+1.187
 Pole+1.177
 boy+1.157
ometer+1.116
Neg logit contributions
ck-1.781
aches-1.744
 yourself-1.715
irs-1.700
 avoid-1.679
Token.
Token ID13
Feature activation-8.65e-01

Pos logit contributions
ck+0.998
aches+0.974
irs+0.940
 yourself+0.912
 avoid+0.893
Neg logit contributions
 girl-0.997
sed-0.994
 boy-0.970
 Pole-0.930
ometer-0.898
Token
Token ID198
Feature activation+2.07e+00

Pos logit contributions
sed+1.791
 girl+1.776
 Pole+1.723
 boy+1.718
ometer+1.654
Neg logit contributions
ck-1.456
aches-1.408
irs-1.366
 avoid-1.333
 yourself-1.322
Token.
Token ID13
Feature activation-1.28e+00

Pos logit contributions
ck+0.525
aches+0.511
irs+0.491
 yourself+0.479
 avoid+0.472
Neg logit contributions
sed-0.510
 girl-0.505
 boy-0.492
 Pole-0.476
ometer-0.461
Token His
Token ID2399
Feature activation-2.16e+00

Pos logit contributions
 girl+2.406
sed+2.396
 boy+2.230
 Pole+2.228
ometer+2.205
Neg logit contributions
ck-2.374
aches-2.370
 yourself-2.269
 avoid-2.250
irs-2.226
Token owner
Token ID4870
Feature activation-1.06e+00

Pos logit contributions
 Pole+3.430
 girl+3.266
 boy+3.056
sed+3.012
keeper+2.927
Neg logit contributions
ck-4.620
aches-4.604
 avoid-4.570
 yourself-4.496
 know-4.287
Token,
Token ID11
Feature activation-1.30e+00

Pos logit contributions
sed+2.000
 Pole+1.896
 girl+1.862
ometer+1.829
 boy+1.782
Neg logit contributions
ck-2.120
aches-2.066
 yourself-1.922
 avoid-1.890
irs-1.875
Token a
Token ID257
Feature activation-3.33e+00

Pos logit contributions
sed+2.729
 Pole+2.661
 girl+2.632
ometer+2.590
 boy+2.511
Neg logit contributions
ck-2.206
aches-2.155
 yourself-2.048
 avoid-1.994
irs-1.955
Token little
Token ID1310
Feature activation-4.48e+00

Pos logit contributions
sed+6.719
 Pole+6.265
umm+6.220
elt+6.172
ometer+6.009
Neg logit contributions
 avoid-5.909
aches-5.885
ck-5.754
 know-5.713
 achieve-5.606
Token girl
Token ID2576
Feature activation-1.69e+00

Pos logit contributions
sed+6.688
 Pole+6.164
icol+5.470
umm+5.418
 cheeks+5.136
Neg logit contributions
 know-11.668
ck-11.530
 yourself-11.346
 avoid-11.312
e-11.236
Token named
Token ID3706
Feature activation-1.26e+00

Pos logit contributions
sed+3.295
 Pole+3.194
 girl+3.015
ometer+3.003
 boy+2.946
Neg logit contributions
ck-3.312
e-3.162
aches-3.134
 yourself-2.997
 avoid-2.980
Token Lily
Token ID20037
Feature activation-1.05e+00

Pos logit contributions
sed+2.159
 Pole+1.998
 girl+1.997
ometer+1.945
 boy+1.942
Neg logit contributions
aches-2.615
ck-2.611
 yourself-2.592
irs-2.501
 avoid-2.477
Token,
Token ID11
Feature activation-5.46e-01

Pos logit contributions
sed+1.721
 girl+1.607
 Pole+1.607
 boy+1.547
ometer+1.516
Neg logit contributions
ck-2.332
aches-2.272
 yourself-2.173
 avoid-2.150
irs-2.129
Token wanted
Token ID2227
Feature activation+4.23e-01

Pos logit contributions
sed+0.919
 girl+0.893
 Pole+0.868
 boy+0.859
ometer+0.846
Neg logit contributions
ck-1.157
aches-1.125
 yourself-1.076
irs-1.068
 avoid-1.058
Token.
Token ID13
Feature activation-8.14e-01

Pos logit contributions
ck+2.505
aches+2.499
irs+2.492
 yourself+2.342
 achieve+2.338
Neg logit contributions
 girl-2.264
 boy-2.205
sed-2.147
 Pole-1.961
ometer-1.946
Token She
Token ID1375
Feature activation-2.91e-01

Pos logit contributions
sed+1.254
 girl+1.218
 Pole+1.213
 boy+1.174
ometer+1.146
Neg logit contributions
ck-1.823
aches-1.818
 yourself-1.711
 avoid-1.701
irs-1.694
Token was
Token ID373
Feature activation-5.79e-01

Pos logit contributions
sed+0.532
 girl+0.518
 Pole+0.502
 boy+0.502
ometer+0.483
Neg logit contributions
ck-0.582
aches-0.567
 yourself-0.534
irs-0.528
 avoid-0.528
Token a
Token ID257
Feature activation-2.73e+00

Pos logit contributions
sed+1.191
 girl+1.160
 Pole+1.156
 boy+1.133
ometer+1.105
Neg logit contributions
ck-1.003
aches-0.970
 yourself-0.912
irs-0.898
 avoid-0.898
Token very
Token ID845
Feature activation-1.64e+00

Pos logit contributions
sed+5.919
 Pole+5.742
board+5.372
ometer+5.349
 eyes+5.228
Neg logit contributions
ck-4.528
 avoid-4.476
aches-4.292
 yourself-4.108
e-4.095
Token curious
Token ID11040
Feature activation-4.47e+00

Pos logit contributions
sed+3.317
 Pole+3.175
 girl+3.067
ometer+3.033
 boy+3.020
Neg logit contributions
ck-2.948
aches-2.932
 avoid-2.695
 yourself-2.640
 belong-2.568
Token girl
Token ID2576
Feature activation-1.16e+00

Pos logit contributions
sed+8.414
 Pole+6.698
ometer+6.593
hog+5.745
ses+5.645
Neg logit contributions
 know-10.769
 yourself-10.124
 appreciate-10.083
 understand-10.025
 belong-10.009
Token who
Token ID508
Feature activation-6.94e-02

Pos logit contributions
sed+2.534
 Pole+2.432
 girl+2.343
 boy+2.301
ometer+2.248
Neg logit contributions
ck-1.982
aches-1.850
 yourself-1.781
 avoid-1.773
e-1.759
Token was
Token ID373
Feature activation-1.85e-01

Pos logit contributions
sed+0.130
 girl+0.126
 boy+0.123
 Pole+0.122
ometer+0.116
Neg logit contributions
ck-0.136
aches-0.132
irs-0.125
 yourself-0.124
 avoid-0.123
Token always
Token ID1464
Feature activation-2.25e-01

Pos logit contributions
sed+0.391
 girl+0.384
 boy+0.375
 Pole+0.372
ometer+0.358
Neg logit contributions
ck-0.313
aches-0.305
irs-0.285
 yourself-0.284
 avoid-0.279
Token looking
Token ID2045
Feature activation-1.20e+00

Pos logit contributions
sed+0.419
 girl+0.405
 boy+0.395
 Pole+0.394
ometer+0.379
Neg logit contributions
ck-0.438
aches-0.426
 yourself-0.408
irs-0.404
 avoid-0.398
Token<|endoftext|>
Token ID50256
Feature activation+2.00e-01

Pos logit contributions
sed+2.473
 girl+2.367
 boy+2.264
 Pole+2.252
ometer+2.181
Neg logit contributions
ck-2.136
irs-2.110
 yourself-2.076
aches-2.033
 avoid-2.018
TokenTim
Token ID14967
Feature activation-4.01e-01

Pos logit contributions
ck+0.425
aches+0.414
irs+0.398
 yourself+0.391
 avoid+0.386
Neg logit contributions
sed-0.335
 girl-0.331
 boy-0.321
 Pole-0.312
ometer-0.299
Token was
Token ID373
Feature activation-9.99e-02

Pos logit contributions
sed+0.820
 girl+0.792
 boy+0.771
 Pole+0.770
ometer+0.745
Neg logit contributions
ck-0.709
aches-0.686
 yourself-0.646
irs-0.646
 avoid-0.633
Token an
Token ID281
Feature activation-6.51e-01

Pos logit contributions
sed+0.191
 girl+0.188
 boy+0.183
 Pole+0.180
ometer+0.174
Neg logit contributions
ck-0.190
aches-0.184
irs-0.176
 yourself-0.174
 avoid-0.171
Token enthusiastic
Token ID21848
Feature activation-3.65e+00

Pos logit contributions
sed+1.153
 girl+1.079
 Pole+1.064
 boy+1.048
ometer+1.035
Neg logit contributions
ck-1.370
aches-1.316
 yourself-1.245
irs-1.241
 avoid-1.229
Token little
Token ID1310
Feature activation-4.47e+00

Pos logit contributions
sed+7.679
ometer+6.750
 Pole+6.636
elt+6.100
�+5.899
Neg logit contributions
asting-6.955
ck-6.951
 know-6.672
 avoid-6.589
 happen-6.546
Token boy
Token ID2933
Feature activation-1.37e+00

Pos logit contributions
sed+8.112
 Pole+6.433
 cheeks+5.800
ometer+5.791
board+5.684
Neg logit contributions
ck-10.436
 avoid-10.288
 enjoy-10.119
 belong-10.009
 ourselves-9.904
Token.
Token ID13
Feature activation-1.48e+00

Pos logit contributions
sed+3.216
 Pole+3.025
 girl+3.001
 boy+2.918
ometer+2.888
Neg logit contributions
ck-2.167
aches-1.978
 yourself-1.942
e-1.936
 avoid-1.893
Token He
Token ID679
Feature activation-2.92e-01

Pos logit contributions
sed+2.055
 girl+1.841
 Pole+1.718
ometer+1.669
 boy+1.657
Neg logit contributions
ck-3.612
aches-3.608
 yourself-3.547
 avoid-3.505
irs-3.493
Token loved
Token ID6151
Feature activation-3.42e-01

Pos logit contributions
sed+0.492
 girl+0.481
 boy+0.464
 Pole+0.460
ometer+0.442
Neg logit contributions
ck-0.620
aches-0.603
irs-0.572
 yourself-0.569
 avoid-0.565
Token to
Token ID284
Feature activation+1.66e+00

Pos logit contributions
sed+0.624
 girl+0.607
 boy+0.586
 Pole+0.579
ometer+0.562
Neg logit contributions
ck-0.679
aches-0.666
irs-0.636
 yourself-0.627
 avoid-0.620
Token
Token ID198
Feature activation+2.09e+00

Pos logit contributions
sed+1.597
 girl+1.520
 boy+1.456
 Pole+1.445
ometer+1.411
Neg logit contributions
ck-1.647
aches-1.625
irs-1.592
 yourself-1.586
 avoid-1.535
Token
Token ID198
Feature activation+4.42e-02

Pos logit contributions
ck+4.697
aches+4.618
irs+4.320
 yourself+4.302
 belong+4.119
Neg logit contributions
 boy-3.537
 girl-3.489
 Pole-3.215
sed-3.208
ometer-3.083
TokenOne
Token ID3198
Feature activation-1.57e+00

Pos logit contributions
ck+0.039
aches+0.037
irs+0.033
 yourself+0.032
 avoid+0.031
Neg logit contributions
sed-0.130
 girl-0.129
 boy-0.126
 Pole-0.125
ometer-0.122
Token day
Token ID1110
Feature activation-2.05e+00

Pos logit contributions
sed+3.451
 Pole+3.236
ometer+3.022
 girl+3.020
elt+2.905
Neg logit contributions
ck-2.723
aches-2.584
 avoid-2.490
asting-2.446
e-2.413
Token a
Token ID257
Feature activation-3.59e+00

Pos logit contributions
sed+3.786
 Pole+3.485
 girl+3.459
ometer+3.377
 boy+3.304
Neg logit contributions
aches-4.192
ck-4.192
 yourself-4.043
irs-3.919
e-3.914
Token little
Token ID1310
Feature activation-4.46e+00

Pos logit contributions
sed+7.367
elt+6.783
ometer+6.445
 Pole+6.443
 cheeks+6.128
Neg logit contributions
ck-6.782
 avoid-6.752
aches-6.673
 achieve-6.404
 know-6.325
Token girl
Token ID2576
Feature activation-1.59e+00

Pos logit contributions
sed+5.785
 Pole+4.665
ometer+4.034
 cheeks+4.012
icol+3.926
Neg logit contributions
ck-13.119
 yourself-12.734
 belong-12.688
 know-12.628
 avoid-12.623
Token came
Token ID1625
Feature activation+3.53e-01

Pos logit contributions
sed+3.080
 Pole+2.893
ometer+2.856
 girl+2.756
 boy+2.697
Neg logit contributions
ck-3.154
aches-2.931
e-2.850
 avoid-2.795
irs-2.794
Token up
Token ID510
Feature activation-4.80e-02

Pos logit contributions
ck+0.766
aches+0.750
irs+0.722
 yourself+0.704
 avoid+0.696
Neg logit contributions
sed-0.582
 girl-0.579
 boy-0.561
 Pole-0.537
ometer-0.520
Token to
Token ID284
Feature activation+2.21e+00

Pos logit contributions
sed+0.086
 girl+0.084
 boy+0.082
 Pole+0.081
ometer+0.077
Neg logit contributions
ck-0.097
aches-0.094
irs-0.090
 yourself-0.089
 avoid-0.088
Token the
Token ID262
Feature activation-1.09e+00

Pos logit contributions
ck+3.690
aches+3.656
irs+3.294
asting+3.291
 achieve+3.226
Neg logit contributions
 girl-5.350
 boy-5.194
sed-4.810
 mouse-4.687
 Pole-4.629
Token there
Token ID612
Feature activation+5.93e-01

Pos logit contributions
ck+2.238
aches+2.166
 yourself+2.088
irs+2.062
 achieve+2.024
Neg logit contributions
 girl-1.877
 boy-1.833
sed-1.832
 Pole-1.727
ometer-1.672
Token was
Token ID373
Feature activation+9.54e-01

Pos logit contributions
ck+1.324
aches+1.290
irs+1.251
 yourself+1.226
 achieve+1.204
Neg logit contributions
 girl-0.939
sed-0.933
 boy-0.918
 Pole-0.863
ometer-0.823
Token a
Token ID257
Feature activation-2.87e+00

Pos logit contributions
ck+1.437
aches+1.417
irs+1.312
 achieve+1.280
 yourself+1.272
Neg logit contributions
 girl-2.225
sed-2.187
 boy-2.176
 Pole-2.048
ometer-1.988
Token depend
Token ID4745
Feature activation+4.07e-01

Pos logit contributions
sed+5.388
elt+4.682
ometer+4.623
 Pole+4.618
 eyes+4.440
Neg logit contributions
ck-6.167
 avoid-6.023
aches-5.807
 achieve-5.627
asting-5.599
Tokenable
Token ID540
Feature activation-3.61e+00

Pos logit contributions
ck+0.772
aches+0.754
irs+0.728
 yourself+0.708
 achieve+0.697
Neg logit contributions
 girl-0.777
sed-0.771
 boy-0.758
 Pole-0.724
ometer-0.704
Token little
Token ID1310
Feature activation-4.45e+00

Pos logit contributions
sed+5.120
elt+4.789
 Pole+4.598
ometer+4.322
umm+4.129
Neg logit contributions
aches-8.870
ck-8.753
 avoid-8.388
asting-8.237
 yourself-8.203
Token girl
Token ID2576
Feature activation-1.29e+00

Pos logit contributions
sed+7.230
 Pole+6.163
ometer+5.402
elt+5.146
board+5.125
Neg logit contributions
ck-11.131
 avoid-10.733
 belong-10.678
 enjoy-10.599
 yourself-10.551
Token named
Token ID3706
Feature activation+3.09e-01

Pos logit contributions
sed+2.532
 Pole+2.450
 girl+2.341
 boy+2.306
ometer+2.286
Neg logit contributions
ck-2.500
aches-2.332
e-2.258
 avoid-2.229
 yourself-2.221
Token Sue
Token ID26089
Feature activation+1.06e+00

Pos logit contributions
ck+0.774
aches+0.757
irs+0.732
 yourself+0.719
 avoid+0.712
Neg logit contributions
sed-0.411
 girl-0.406
 boy-0.389
 Pole-0.372
ometer-0.354
Token.
Token ID13
Feature activation-4.29e-01

Pos logit contributions
ck+2.129
aches+2.103
irs+2.086
 yourself+1.977
 achieve+1.972
Neg logit contributions
 girl-1.958
 boy-1.899
sed-1.871
 Pole-1.720
ometer-1.699
Token She
Token ID1375
Feature activation-4.09e-01

Pos logit contributions
sed+0.621
 girl+0.601
 Pole+0.584
 boy+0.579
ometer+0.552
Neg logit contributions
ck-1.009
aches-1.000
irs-0.949
 yourself-0.947
 avoid-0.939
Token show
Token ID905
Feature activation+1.10e+00

Pos logit contributions
ck+5.993
aches+5.925
irs+5.828
lled+5.461
asting+5.364
Neg logit contributions
 boy-5.703
 girl-5.663
sed-5.590
 eyes-5.134
 mouth-5.083
Token his
Token ID465
Feature activation-4.57e-01

Pos logit contributions
ck+1.567
aches+1.534
irs+1.437
 achieve+1.426
 avoid+1.366
Neg logit contributions
 girl-2.641
sed-2.617
 boy-2.602
 Pole-2.452
ometer-2.423
Token friend
Token ID1545
Feature activation-1.00e+00

Pos logit contributions
sed+0.678
 girl+0.661
 Pole+0.651
 boy+0.638
ometer+0.601
Neg logit contributions
ck-1.059
aches-1.042
 yourself-0.982
irs-0.982
 avoid-0.975
Token,
Token ID11
Feature activation-4.50e-01

Pos logit contributions
sed+1.997
 girl+1.899
 Pole+1.865
 boy+1.816
ometer+1.777
Neg logit contributions
ck-1.896
aches-1.855
 yourself-1.737
irs-1.736
 avoid-1.692
Token a
Token ID257
Feature activation-2.43e+00

Pos logit contributions
sed+0.819
 girl+0.799
 Pole+0.775
 boy+0.770
ometer+0.748
Neg logit contributions
ck-0.889
aches-0.861
irs-0.824
 yourself-0.820
 avoid-0.812
Token little
Token ID1310
Feature activation-4.45e+00

Pos logit contributions
sed+5.271
 Pole+4.986
elt+4.719
ometer+4.708
 girl+4.581
Neg logit contributions
aches-4.038
 achieve-3.811
 know-3.777
ck-3.753
 avoid-3.709
Token girl
Token ID2576
Feature activation-1.07e+00

Pos logit contributions
 Pole+7.424
sed+7.402
ometer+6.312
umm+6.270
board+5.794
Neg logit contributions
 know-11.060
ck-10.531
 belong-10.251
 avoid-10.126
 yourself-10.097
Token named
Token ID3706
Feature activation-2.87e-01

Pos logit contributions
sed+2.055
 Pole+1.957
 girl+1.896
 boy+1.875
ometer+1.868
Neg logit contributions
ck-2.121
aches-1.994
 yourself-1.914
irs-1.913
 avoid-1.892
Token Lily
Token ID20037
Feature activation-5.40e-01

Pos logit contributions
sed+0.419
 girl+0.400
 Pole+0.388
 boy+0.388
ometer+0.371
Neg logit contributions
ck-0.672
aches-0.658
irs-0.639
 yourself-0.636
 avoid-0.622
Token.
Token ID13
Feature activation-2.37e+00

Pos logit contributions
sed+1.160
 girl+1.118
 boy+1.095
 Pole+1.092
ometer+1.060
Neg logit contributions
ck-0.916
aches-0.874
irs-0.834
 yourself-0.824
 avoid-0.810
Token Lily
Token ID20037
Feature activation-1.27e+00

Pos logit contributions
sed+3.997
 Pole+3.820
 girl+3.688
 boy+3.575
ometer+3.529
Neg logit contributions
 yourself-4.905
 know-4.866
irs-4.865
ck-4.800
 avoid-4.781
Token sunny
Token ID27737
Feature activation-1.18e+00

Pos logit contributions
sed+2.437
 Pole+2.267
 girl+2.162
ometer+2.143
 boy+2.102
Neg logit contributions
ck-2.034
aches-1.946
 avoid-1.863
asting-1.821
 yourself-1.783
Token day
Token ID1110
Feature activation-2.27e+00

Pos logit contributions
sed+2.482
 Pole+2.289
 girl+2.202
ometer+2.200
 boy+2.158
Neg logit contributions
ck-2.074
aches-2.065
 yourself-1.942
 avoid-1.916
asting-1.868
Token,
Token ID11
Feature activation-1.33e+00

Pos logit contributions
sed+4.299
 girl+4.043
 Pole+3.903
 boy+3.826
ometer+3.805
Neg logit contributions
ck-4.563
aches-4.444
 avoid-4.262
 yourself-4.245
e-4.244
Token a
Token ID257
Feature activation-3.28e+00

Pos logit contributions
sed+3.705
 girl+3.527
 Pole+3.515
ometer+3.485
 boy+3.391
Neg logit contributions
ck-1.378
aches-1.325
 avoid-1.234
 yourself-1.215
irs-1.119
Token happy
Token ID3772
Feature activation-3.29e+00

Pos logit contributions
sed+7.085
elt+6.361
 Pole+6.287
ometer+6.280
 cheeks+5.958
Neg logit contributions
 avoid-5.762
ck-5.625
aches-5.554
 achieve-5.522
asting-5.367
Token little
Token ID1310
Feature activation-4.44e+00

Pos logit contributions
sed+5.662
 Pole+5.144
ometer+5.073
elt+4.783
hog+4.491
Neg logit contributions
ck-7.294
aches-6.922
 avoid-6.841
 know-6.831
 happen-6.729
Token girl
Token ID2576
Feature activation-1.32e+00

Pos logit contributions
sed+6.820
 Pole+6.068
ometer+5.115
 cheeks+5.109
icol+5.102
Neg logit contributions
 belong-11.348
ck-11.108
 happen-11.019
 know-10.817
 avoid-10.747
Token named
Token ID3706
Feature activation-2.84e-01

Pos logit contributions
sed+2.461
 Pole+2.302
ometer+2.288
 girl+2.246
 boy+2.199
Neg logit contributions
ck-2.686
aches-2.482
irs-2.423
e-2.421
 yourself-2.385
Token Lily
Token ID20037
Feature activation-5.16e-01

Pos logit contributions
sed+0.431
 girl+0.412
 boy+0.400
 Pole+0.397
ometer+0.381
Neg logit contributions
ck-0.651
aches-0.640
irs-0.618
 yourself-0.615
 avoid-0.601
Token came
Token ID1625
Feature activation+6.98e-01

Pos logit contributions
sed+0.901
 girl+0.855
 Pole+0.829
 boy+0.828
ometer+0.809
Neg logit contributions
ck-1.083
aches-1.040
irs-0.998
 yourself-0.992
 avoid-0.979
Token to
Token ID284
Feature activation+2.04e+00

Pos logit contributions
ck+1.561
aches+1.541
irs+1.472
 yourself+1.421
 achieve+1.414
Neg logit contributions
 girl-1.128
sed-1.107
 boy-1.089
 Pole-1.004
ometer-0.984
Token.
Token ID13
Feature activation-2.37e+00

Pos logit contributions
sed+0.840
 girl+0.812
 boy+0.793
 Pole+0.786
ometer+0.762
Neg logit contributions
ck-0.700
aches-0.671
irs-0.644
 yourself-0.628
 avoid-0.624
Token Max
Token ID5436
Feature activation-1.35e+00

Pos logit contributions
sed+4.150
 girl+3.739
 Pole+3.626
ometer+3.593
 boy+3.475
Neg logit contributions
irs-4.777
 avoid-4.740
 yourself-4.663
 know-4.640
 achieve-4.635
Token was
Token ID373
Feature activation-2.61e-01

Pos logit contributions
sed+2.554
 girl+2.455
 Pole+2.414
ometer+2.368
 boy+2.343
Neg logit contributions
ck-2.581
aches-2.515
 avoid-2.365
 achieve-2.356
 yourself-2.323
Token a
Token ID257
Feature activation-2.68e+00

Pos logit contributions
sed+0.548
 girl+0.537
 boy+0.523
 Pole+0.523
ometer+0.505
Neg logit contributions
ck-0.449
aches-0.435
irs-0.409
 yourself-0.409
 avoid-0.399
Token crazy
Token ID7165
Feature activation-3.75e+00

Pos logit contributions
sed+6.214
 Pole+5.843
ometer+5.677
elt+5.649
 eyes+5.571
Neg logit contributions
 avoid-4.094
ck-4.040
 achieve-3.806
aches-3.686
 know-3.679
Token little
Token ID1310
Feature activation-4.44e+00

Pos logit contributions
sed+6.801
 Pole+6.776
ometer+6.314
elt+6.224
 cheeks+5.738
Neg logit contributions
ck-7.785
 avoid-7.613
 know-7.436
 happen-7.434
e-7.327
Token boy
Token ID2933
Feature activation-1.51e+00

Pos logit contributions
sed+10.041
 Pole+9.690
ometer+8.585
umm+8.513
 cheeks+8.426
Neg logit contributions
ck-8.171
 know-8.068
 avoid-8.057
 belong-7.875
 enjoy-7.747
Token who
Token ID508
Feature activation-3.33e-01

Pos logit contributions
sed+3.273
 Pole+3.094
 girl+3.056
 boy+2.950
ometer+2.911
Neg logit contributions
ck-2.703
e-2.492
aches-2.453
 avoid-2.428
 yourself-2.357
Token loved
Token ID6151
Feature activation-7.25e-01

Pos logit contributions
sed+0.565
 girl+0.548
 boy+0.527
 Pole+0.525
ometer+0.502
Neg logit contributions
ck-0.707
aches-0.687
irs-0.655
 yourself-0.646
 avoid-0.644
Token to
Token ID284
Feature activation+1.78e+00

Pos logit contributions
sed+1.454
 girl+1.413
 Pole+1.388
 boy+1.367
ometer+1.330
Neg logit contributions
ck-1.310
aches-1.279
irs-1.229
 yourself-1.218
 avoid-1.188
Token play
Token ID711
Feature activation+4.27e-01

Pos logit contributions
ck+2.933
irs+2.754
aches+2.724
 yourself+2.623
asting+2.558
Neg logit contributions
sed-3.919
 girl-3.916
 boy-3.856
 Pole-3.652
board-3.536
Token there
Token ID612
Feature activation+3.95e-01

Pos logit contributions
ck+2.245
aches+2.183
 yourself+2.111
irs+2.078
 achieve+2.033
Neg logit contributions
 girl-1.874
 boy-1.843
sed-1.841
 Pole-1.742
ometer-1.687
Token was
Token ID373
Feature activation+9.28e-01

Pos logit contributions
ck+0.860
aches+0.838
irs+0.809
 yourself+0.794
 avoid+0.778
Neg logit contributions
sed-0.642
 girl-0.639
 boy-0.623
 Pole-0.596
ometer-0.570
Token a
Token ID257
Feature activation-2.78e+00

Pos logit contributions
ck+1.391
aches+1.378
irs+1.276
 yourself+1.239
 achieve+1.238
Neg logit contributions
 girl-2.162
sed-2.131
 boy-2.121
 Pole-1.999
ometer-1.941
Token very
Token ID845
Feature activation-1.92e+00

Pos logit contributions
sed+4.977
elt+4.333
ometer+4.274
 Pole+4.245
 eyes+4.053
Neg logit contributions
ck-6.228
 avoid-6.008
aches-5.833
 achieve-5.638
asting-5.611
Token polite
Token ID23507
Feature activation-3.60e+00

Pos logit contributions
sed+3.996
 Pole+3.751
ometer+3.647
 girl+3.595
elt+3.528
Neg logit contributions
ck-3.489
aches-3.429
 avoid-3.158
irl-3.071
 yourself-3.044
Token little
Token ID1310
Feature activation-4.43e+00

Pos logit contributions
sed+5.436
ometer+4.440
 Pole+4.374
elt+4.360
 Sure+3.768
Neg logit contributions
aches-8.842
ck-8.669
asting-8.589
e-8.341
 yourself-8.303
Token bear
Token ID6842
Feature activation-8.46e-01

Pos logit contributions
sed+7.256
 Pole+6.242
elt+5.639
ometer+5.623
board+5.489
Neg logit contributions
ck-10.979
 belong-10.331
aches-10.315
 avoid-10.273
 ourselves-10.207
Token.
Token ID13
Feature activation-1.42e+00

Pos logit contributions
sed+1.622
 girl+1.590
 Pole+1.543
 boy+1.534
ometer+1.457
Neg logit contributions
ck-1.633
aches-1.563
 avoid-1.452
 yourself-1.437
irs-1.436
Token He
Token ID679
Feature activation-3.76e-02

Pos logit contributions
sed+2.050
 girl+1.922
 Pole+1.781
 boy+1.756
ometer+1.707
Neg logit contributions
aches-3.362
ck-3.350
 yourself-3.298
 avoid-3.230
 achieve-3.197
Token was
Token ID373
Feature activation-1.76e-01

Pos logit contributions
sed+0.065
 girl+0.064
 boy+0.062
 Pole+0.061
ometer+0.058
Neg logit contributions
ck-0.078
aches-0.076
irs-0.073
 yourself-0.072
 avoid-0.071
Token always
Token ID1464
Feature activation-5.85e-01

Pos logit contributions
sed+0.352
 girl+0.347
 boy+0.337
 Pole+0.334
ometer+0.322
Neg logit contributions
ck-0.318
aches-0.310
 yourself-0.289
irs-0.289
 avoid-0.284
Token.
Token ID13
Feature activation-9.34e-01

Pos logit contributions
sed+3.995
 girl+3.756
 boy+3.662
 Pole+3.523
ometer+3.491
Neg logit contributions
ck-3.311
aches-3.242
 yourself-3.100
irs-3.054
asting-2.968
Token One
Token ID1881
Feature activation-1.00e+00

Pos logit contributions
sed+1.850
 girl+1.743
 boy+1.663
 Pole+1.649
ometer+1.632
Neg logit contributions
ck-1.704
aches-1.690
irs-1.659
 yourself-1.641
 avoid-1.597
Token day
Token ID1110
Feature activation-1.98e+00

Pos logit contributions
sed+2.084
 Pole+1.933
 girl+1.873
ometer+1.827
 boy+1.809
Neg logit contributions
ck-1.812
aches-1.745
 avoid-1.668
asting-1.638
 yourself-1.603
Token,
Token ID11
Feature activation-1.34e+00

Pos logit contributions
sed+3.568
 girl+3.382
 Pole+3.344
 boy+3.226
ometer+3.183
Neg logit contributions
ck-4.064
aches-4.036
 yourself-3.849
 avoid-3.837
irs-3.798
Token a
Token ID257
Feature activation-3.72e+00

Pos logit contributions
sed+3.839
 Pole+3.647
ometer+3.596
 girl+3.591
 boy+3.482
Neg logit contributions
ck-1.327
aches-1.291
 avoid-1.192
 yourself-1.180
irs-1.079
Token little
Token ID1310
Feature activation-4.43e+00

Pos logit contributions
sed+8.355
elt+7.282
ometer+7.152
 Pole+7.102
 cheeks+6.840
Neg logit contributions
 avoid-6.594
 achieve-6.336
ck-6.293
aches-6.288
asting-6.269
Token girl
Token ID2576
Feature activation-1.59e+00

Pos logit contributions
sed+6.327
 Pole+4.804
 cheeks+4.473
ometer+4.240
icol+4.176
Neg logit contributions
ck-12.259
 yourself-12.108
 avoid-12.013
 know-11.952
 belong-11.951
Token came
Token ID1625
Feature activation+5.52e-01

Pos logit contributions
sed+3.130
 Pole+2.911
ometer+2.893
 girl+2.774
 boy+2.727
Neg logit contributions
ck-3.148
aches-2.892
e-2.872
irs-2.827
 avoid-2.796
Token to
Token ID284
Feature activation+2.44e+00

Pos logit contributions
ck+1.229
aches+1.210
irs+1.161
 yourself+1.125
 avoid+1.113
Neg logit contributions
 girl-0.886
sed-0.877
 boy-0.856
 Pole-0.805
ometer-0.784
Token the
Token ID262
Feature activation-1.22e+00

Pos logit contributions
ck+4.164
aches+4.121
irs+3.848
asting+3.722
cks+3.628
Neg logit contributions
 girl-5.799
 boy-5.669
sed-5.169
 mouse-5.103
 Pole-4.967
Token house
Token ID2156
Feature activation-6.51e-01

Pos logit contributions
sed+1.824
 girl+1.576
 Pole+1.568
ometer+1.527
 boy+1.509
Neg logit contributions
ck-2.805
aches-2.741
 yourself-2.710
 avoid-2.709
 achieve-2.648
Token to
Token ID284
Feature activation+1.49e+00

Pos logit contributions
sed+1.103
 girl+1.053
 Pole+1.026
 boy+1.011
ometer+0.982
Neg logit contributions
ck-1.185
aches-1.173
irs-1.111
 avoid-1.107
 yourself-1.105
Token play
Token ID711
Feature activation+1.54e-01

Pos logit contributions
ck+2.327
irs+2.192
aches+2.164
 yourself+2.072
 avoid+1.946
Neg logit contributions
 girl-3.435
 boy-3.380
sed-3.357
 Pole-3.170
ometer-3.038
Token with
Token ID351
Feature activation+2.46e-01

Pos logit contributions
ck+0.293
aches+0.285
irs+0.273
 yourself+0.267
 avoid+0.263
Neg logit contributions
sed-0.297
 girl-0.293
 boy-0.286
 Pole-0.278
ometer-0.269
Token their
Token ID511
Feature activation-1.78e+00

Pos logit contributions
ck+0.372
aches+0.359
irs+0.339
 yourself+0.331
 avoid+0.324
Neg logit contributions
sed-0.568
 girl-0.563
 boy-0.552
 Pole-0.538
ometer-0.523
Token toys
Token ID14958
Feature activation-2.32e-01

Pos logit contributions
 Pole+2.997
sed+2.983
 girl+2.861
ometer+2.825
elt+2.756
Neg logit contributions
ck-3.653
aches-3.602
 avoid-3.443
irs-3.352
 ourselves-3.332
Token in
Token ID287
Feature activation+2.85e+00

Pos logit contributions
sed+0.449
 girl+0.438
 boy+0.426
 Pole+0.421
ometer+0.406
Neg logit contributions
ck-0.436
aches-0.426
irs-0.401
 yourself-0.398
 avoid-0.392
Token the
Token ID262
Feature activation-1.07e+00

Pos logit contributions
aches+5.446
ck+5.386
irs+4.898
asting+4.803
 achieve+4.798
Neg logit contributions
 girl-6.454
 boy-6.390
 mouse-5.617
 eyes-5.548
 "-5.417
Token garden
Token ID11376
Feature activation-2.97e-01

Pos logit contributions
sed+1.903
 Pole+1.751
 girl+1.739
ometer+1.718
 boy+1.678
Neg logit contributions
aches-2.144
ck-2.128
 yourself-2.109
irs-2.049
 achieve-2.032
Token.
Token ID13
Feature activation-6.54e-01

Pos logit contributions
sed+0.615
 girl+0.598
 boy+0.584
 Pole+0.578
ometer+0.561
Neg logit contributions
ck-0.514
aches-0.504
irs-0.470
 yourself-0.470
 avoid-0.461
Token One
Token ID1881
Feature activation-7.97e-01

Pos logit contributions
sed+1.170
 girl+1.136
 boy+1.094
 Pole+1.077
ometer+1.048
Neg logit contributions
ck-1.316
aches-1.283
irs-1.241
 avoid-1.211
 yourself-1.207
Token day
Token ID1110
Feature activation-6.60e-01

Pos logit contributions
sed+1.597
 Pole+1.506
 girl+1.469
 boy+1.424
ometer+1.421
Neg logit contributions
ck-1.484
aches-1.468
 avoid-1.357
 yourself-1.343
asting-1.324
Token her
Token ID607
Feature activation-1.10e+00

Pos logit contributions
ck+3.062
aches+3.028
irs+2.729
asting+2.593
 achieve+2.589
Neg logit contributions
 girl-4.817
sed-4.601
 boy-4.588
 eyes-4.395
 mouse-4.241
Token heart
Token ID2612
Feature activation+3.10e-01

Pos logit contributions
sed+2.094
 Pole+2.010
 girl+2.004
 boy+1.977
ometer+1.943
Neg logit contributions
ck-2.046
aches-2.011
 yourself-1.935
 avoid-1.898
irs-1.890
Token knowing
Token ID6970
Feature activation+8.70e-01

Pos logit contributions
ck+0.640
aches+0.623
irs+0.601
 yourself+0.588
 avoid+0.580
Neg logit contributions
sed-0.536
 girl-0.531
 boy-0.515
 Pole-0.499
ometer-0.482
Token that
Token ID326
Feature activation+4.24e-01

Pos logit contributions
ck+1.604
aches+1.530
irs+1.478
 yourself+1.417
 avoid+1.413
Neg logit contributions
 girl-1.733
sed-1.712
 boy-1.677
 Pole-1.606
ometer-1.559
Token she
Token ID673
Feature activation+8.13e-01

Pos logit contributions
ck+0.783
aches+0.757
irs+0.727
 yourself+0.706
 avoid+0.696
Neg logit contributions
sed-0.827
 girl-0.826
 boy-0.805
 Pole-0.776
ometer-0.753
Token would
Token ID561
Feature activation+3.12e+00

Pos logit contributions
ck+1.824
aches+1.793
irs+1.750
 yourself+1.682
 achieve+1.653
Neg logit contributions
 girl-1.304
sed-1.280
 boy-1.261
 Pole-1.173
ometer-1.130
Token still
Token ID991
Feature activation+2.27e+00

Pos logit contributions
ck+5.233
irs+5.059
aches+4.901
cks+4.587
asting+4.272
Neg logit contributions
 girl-7.330
 boy-7.076
sed-6.951
 mouse-6.557
 Pole-6.505
Token get
Token ID651
Feature activation+1.53e+00

Pos logit contributions
ck+3.377
irs+3.332
aches+3.176
asting+2.812
 yourself+2.799
Neg logit contributions
 girl-5.545
 boy-5.403
sed-5.369
 Pole-4.932
 mouse-4.883
Token to
Token ID284
Feature activation+2.62e+00

Pos logit contributions
ck+2.909
aches+2.817
irs+2.752
 achieve+2.634
 avoid+2.604
Neg logit contributions
 girl-3.033
 boy-2.905
sed-2.840
 Pole-2.647
ometer-2.616
Token celebrate
Token ID10648
Feature activation+1.41e+00

Pos logit contributions
ck+4.830
aches+4.576
irs+4.546
asting+4.155
cks+4.150
Neg logit contributions
 girl-5.655
 boy-5.405
sed-5.181
 eyes-5.085
 mouse-4.888
Token her
Token ID607
Feature activation+2.33e-01

Pos logit contributions
ck+2.739
aches+2.605
irs+2.542
 avoid+2.387
 achieve+2.367
Neg logit contributions
 girl-2.793
 boy-2.687
sed-2.570
 Pole-2.463
ometer-2.375
Token,
Token ID11
Feature activation-1.77e+00

Pos logit contributions
sed+2.634
 Pole+2.469
 girl+2.426
 boy+2.359
ometer+2.337
Neg logit contributions
ck-3.195
aches-3.035
 yourself-2.911
 avoid-2.879
ered-2.830
Token "
Token ID366
Feature activation+1.43e+00

Pos logit contributions
sed+1.957
 Pole+1.832
 girl+1.815
ometer+1.776
 boy+1.751
Neg logit contributions
ck-4.864
aches-4.729
 yourself-4.622
 avoid-4.517
ered-4.406
TokenSl
Token ID11122
Feature activation+8.67e-01

Pos logit contributions
ck+2.844
aches+2.784
irs+2.753
 achieve+2.550
 yourself+2.543
Neg logit contributions
 girl-2.658
 boy-2.631
sed-2.557
 Pole-2.427
ometer-2.397
Tokenipping
Token ID4501
Feature activation+2.31e-01

Pos logit contributions
ck+1.483
aches+1.447
 yourself+1.354
irs+1.353
 avoid+1.337
Neg logit contributions
sed-1.783
 girl-1.752
 boy-1.718
 Pole-1.711
ometer-1.662
Token means
Token ID1724
Feature activation+4.99e-01

Pos logit contributions
ck+0.409
aches+0.397
irs+0.379
 yourself+0.369
 avoid+0.364
Neg logit contributions
sed-0.468
 girl-0.463
 boy-0.452
 Pole-0.441
ometer-0.427
Token to
Token ID284
Feature activation+3.98e+00

Pos logit contributions
ck+0.835
aches+0.804
irs+0.771
 yourself+0.739
 avoid+0.730
Neg logit contributions
sed-1.063
 girl-1.053
 boy-1.032
 Pole-1.005
ometer-0.975
Token try
Token ID1949
Feature activation+1.04e+00

Pos logit contributions
irs+8.041
aches+7.201
ck+7.132
cks+7.001
lled+6.800
Neg logit contributions
 boy-8.684
 girl-8.573
sed-7.949
 cat-7.324
 mouse-7.192
Token to
Token ID284
Feature activation+3.80e+00

Pos logit contributions
ck+2.184
aches+2.109
irs+2.048
 achieve+1.927
 avoid+1.923
Neg logit contributions
 girl-1.785
sed-1.780
 boy-1.745
 Pole-1.689
ometer-1.647
Token get
Token ID651
Feature activation+1.96e+00

Pos logit contributions
irs+7.805
ck+7.006
aches+6.896
cks+6.854
lled+6.407
Neg logit contributions
 boy-8.289
 girl-8.167
sed-7.217
 cart-6.955
 cat-6.943
Token somewhere
Token ID7382
Feature activation+1.62e+00

Pos logit contributions
ck+3.528
aches+3.453
irs+3.362
 achieve+3.269
e+3.065
Neg logit contributions
 girl-4.012
 boy-3.956
sed-3.865
 Pole-3.608
ometer-3.588
Token without
Token ID1231
Feature activation+1.96e+00

Pos logit contributions
ck+3.037
aches+2.949
irs+2.912
e+2.667
 achieve+2.648
Neg logit contributions
 girl-3.229
 boy-3.171
sed-3.110
 Pole-2.994
ometer-2.871
TokenOkay
Token ID16454
Feature activation-7.59e-01

Pos logit contributions
sed+0.479
 girl+0.463
 Pole+0.450
 boy+0.449
ometer+0.435
Neg logit contributions
ck-0.461
aches-0.444
 yourself-0.421
irs-0.420
 avoid-0.416
Token,
Token ID11
Feature activation-4.53e-01

Pos logit contributions
sed+1.134
 girl+1.055
 Pole+1.028
 boy+1.008
ometer+1.003
Neg logit contributions
ck-1.765
aches-1.728
 yourself-1.678
 avoid-1.637
irs-1.630
Token Lily
Token ID20037
Feature activation-9.14e-01

Pos logit contributions
sed+0.724
 girl+0.686
 Pole+0.670
 boy+0.662
ometer+0.652
Neg logit contributions
ck-1.015
aches-0.988
 yourself-0.948
irs-0.936
 avoid-0.933
Token,
Token ID11
Feature activation+3.43e-01

Pos logit contributions
sed+1.892
 girl+1.796
 Pole+1.795
 boy+1.732
ometer+1.701
Neg logit contributions
ck-1.597
aches-1.532
 yourself-1.458
 avoid-1.433
irs-1.391
Token we
Token ID356
Feature activation+1.20e+00

Pos logit contributions
ck+0.673
aches+0.656
irs+0.630
 yourself+0.612
 avoid+0.603
Neg logit contributions
sed-0.638
 girl-0.630
 boy-0.614
 Pole-0.596
ometer-0.573
Token can
Token ID460
Feature activation+2.58e+00

Pos logit contributions
ck+2.583
aches+2.482
irs+2.431
 yourself+2.361
 achieve+2.298
Neg logit contributions
 girl-2.049
sed-2.044
 boy-2.025
ometer-1.848
 Pole-1.840
Token play
Token ID711
Feature activation+9.64e-01

Pos logit contributions
ck+4.272
irs+4.241
aches+4.078
asting+3.977
lled+3.965
Neg logit contributions
 girl-5.729
sed-5.684
 boy-5.675
 Pole-5.249
ometer-5.229
Token somewhere
Token ID7382
Feature activation+1.34e+00

Pos logit contributions
ck+1.845
aches+1.810
irs+1.762
asting+1.642
 yourself+1.639
Neg logit contributions
 girl-1.856
 boy-1.816
sed-1.812
 Pole-1.678
 mouse-1.642
Token else
Token ID2073
Feature activation-6.50e-02

Pos logit contributions
ck+1.800
aches+1.796
irs+1.745
 yourself+1.587
 achieve+1.564
Neg logit contributions
 girl-3.333
 boy-3.278
sed-3.219
 Pole-3.102
ometer-2.988
Token.
Token ID13
Feature activation-1.61e+00

Pos logit contributions
sed+0.145
 girl+0.142
 boy+0.139
 Pole+0.137
ometer+0.133
Neg logit contributions
ck-0.103
aches-0.099
irs-0.093
 yourself-0.092
 avoid-0.090
Token But
Token ID887
Feature activation+7.27e-01

Pos logit contributions
 girl+2.171
sed+2.149
 Pole+2.103
 boy+2.046
ometer+1.875
Neg logit contributions
ck-3.774
 yourself-3.756
 avoid-3.704
 ourselves-3.639
aches-3.556
Token wait
Token ID4043
Feature activation+8.97e-01

Pos logit contributions
irs+5.506
 yourself+5.008
aches+4.985
ck+4.923
cks+4.852
Neg logit contributions
 girl-7.287
 boy-7.117
sed-6.883
 Pole-6.341
 mouse-6.262
Token to
Token ID284
Feature activation+3.47e+00

Pos logit contributions
ck+2.055
aches+2.004
irs+1.944
 yourself+1.899
 achieve+1.849
Neg logit contributions
sed-1.372
 girl-1.356
 boy-1.310
 Pole-1.252
ometer-1.184
Token get
Token ID651
Feature activation+1.53e+00

Pos logit contributions
ck+7.046
irs+6.918
aches+6.500
lled+6.390
 ourselves+6.382
Neg logit contributions
 girl-6.805
 boy-6.596
 eyes-5.911
sed-5.896
 cart-5.801
Token started
Token ID2067
Feature activation+4.93e-01

Pos logit contributions
ck+3.137
aches+3.035
irs+2.988
 achieve+2.911
 yourself+2.842
Neg logit contributions
 girl-2.767
 boy-2.716
sed-2.618
 Pole-2.487
ometer-2.408
Token.
Token ID13
Feature activation-4.21e-01

Pos logit contributions
ck+1.002
aches+0.975
irs+0.938
 yourself+0.914
 avoid+0.903
Neg logit contributions
sed-0.870
 girl-0.863
 boy-0.841
 Pole-0.812
ometer-0.781
Token
Token ID198
Feature activation+2.62e+00

Pos logit contributions
sed+0.707
 girl+0.687
 boy+0.661
 Pole+0.646
ometer+0.619
Neg logit contributions
ck-0.889
aches-0.876
irs-0.850
 yourself-0.836
 avoid-0.830
Token
Token ID198
Feature activation+1.90e+00

Pos logit contributions
ck+6.185
aches+6.067
 yourself+5.745
 happen+5.698
 belong+5.665
Neg logit contributions
 boy-4.184
 girl-4.136
 Pole-3.790
 "-3.744
ometer-3.736
Token"
Token ID1
Feature activation+6.42e-01

Pos logit contributions
ck+3.311
aches+3.167
irs+3.046
 yourself+2.956
 happen+2.904
Neg logit contributions
 girl-4.074
 boy-4.002
 Pole-3.738
sed-3.691
ometer-3.646
TokenOn
Token ID2202
Feature activation+2.45e+00

Pos logit contributions
ck+1.304
aches+1.277
irs+1.218
 yourself+1.184
 achieve+1.179
Neg logit contributions
 girl-1.162
sed-1.140
 boy-1.130
 Pole-1.077
ometer-1.025
Token your
Token ID534
Feature activation+1.13e-01

Pos logit contributions
ck+4.124
aches+3.827
irs+3.785
 avoid+3.473
 achieve+3.467
Neg logit contributions
 girl-5.488
 boy-5.480
sed-5.282
 Pole-5.071
 eyes-4.970
Token mark
Token ID1317
Feature activation+7.15e-01

Pos logit contributions
ck+0.223
aches+0.217
irs+0.208
 yourself+0.204
 avoid+0.201
Neg logit contributions
sed-0.207
 girl-0.204
 boy-0.198
 Pole-0.193
ometer-0.186
Token face
Token ID1986
Feature activation-5.99e-01

Pos logit contributions
sed+0.338
 girl+0.334
 boy+0.320
 Pole+0.312
ometer+0.295
Neg logit contributions
ck-0.716
aches-0.704
 yourself-0.673
irs-0.671
 avoid-0.664
Token.
Token ID13
Feature activation-1.24e+00

Pos logit contributions
sed+1.229
 girl+1.200
 boy+1.171
 Pole+1.154
ometer+1.116
Neg logit contributions
ck-1.058
aches-1.032
 yourself-0.968
irs-0.960
 avoid-0.939
Token<|endoftext|>
Token ID50256
Feature activation+4.34e-01

Pos logit contributions
sed+2.362
 girl+2.260
 boy+2.171
 Pole+2.141
ometer+2.074
Neg logit contributions
ck-2.342
irs-2.293
aches-2.280
 yourself-2.257
 achieve-2.171
TokenOnce
Token ID7454
Feature activation+5.93e-01

Pos logit contributions
ck+0.865
aches+0.840
irs+0.799
 yourself+0.781
 avoid+0.774
Neg logit contributions
 girl-0.789
sed-0.788
 boy-0.771
 Pole-0.745
ometer-0.715
Token upon
Token ID2402
Feature activation+2.32e+00

Pos logit contributions
ck+1.186
aches+1.152
irs+1.103
 yourself+1.091
 avoid+1.069
Neg logit contributions
 girl-1.076
sed-1.071
 boy-1.052
 Pole-1.008
ometer-0.972
Token a
Token ID257
Feature activation+1.82e-01

Pos logit contributions
irs+4.375
ck+4.369
aches+4.318
 achieve+3.963
 avoid+3.810
Neg logit contributions
 girl-4.990
 boy-4.948
sed-4.553
 mouse-4.355
 eyes-4.153
Token time
Token ID640
Feature activation-1.92e-01

Pos logit contributions
ck+0.239
aches+0.229
irs+0.215
 yourself+0.208
 avoid+0.203
Neg logit contributions
sed-0.455
 girl-0.450
 boy-0.441
 Pole-0.432
ometer-0.421
Token there
Token ID612
Feature activation+1.07e+00

Pos logit contributions
sed+0.255
 girl+0.249
 boy+0.239
 Pole+0.231
ometer+0.221
Neg logit contributions
ck-0.475
aches-0.466
irs-0.451
 yourself-0.446
 avoid-0.440
Token was
Token ID373
Feature activation+9.84e-01

Pos logit contributions
ck+2.527
aches+2.476
irs+2.411
 yourself+2.354
 achieve+2.304
Neg logit contributions
 girl-1.600
 boy-1.589
sed-1.550
 Pole-1.438
ometer-1.369
Token a
Token ID257
Feature activation-2.75e+00

Pos logit contributions
ck+1.479
aches+1.463
irs+1.369
 achieve+1.317
 yourself+1.311
Neg logit contributions
 girl-2.286
sed-2.278
 boy-2.237
 Pole-2.118
ometer-2.054
Token little
Token ID1310
Feature activation-3.52e+00

Pos logit contributions
sed+4.929
elt+4.403
ometer+4.363
 Pole+4.315
board+4.108
Neg logit contributions
ck-6.125
 avoid-5.724
aches-5.708
irl-5.472
 achieve-5.448
TokenOh
Token ID5812
Feature activation-6.14e-02

Pos logit contributions
ck+1.910
aches+1.898
irs+1.836
 achieve+1.732
 yourself+1.722
Neg logit contributions
 girl-1.833
 boy-1.789
sed-1.760
 Pole-1.661
ometer-1.593
Token no
Token ID645
Feature activation-4.62e-01

Pos logit contributions
sed+0.138
 girl+0.136
 boy+0.133
 Pole+0.131
ometer+0.127
Neg logit contributions
ck-0.096
aches-0.093
irs-0.087
 yourself-0.086
 avoid-0.084
Token,
Token ID11
Feature activation+8.36e-01

Pos logit contributions
sed+0.725
 girl+0.707
 boy+0.683
 Pole+0.679
ometer+0.652
Neg logit contributions
ck-1.039
aches-1.013
 yourself-0.973
irs-0.965
 avoid-0.957
Token Ben
Token ID3932
Feature activation-6.37e-01

Pos logit contributions
ck+1.687
aches+1.637
irs+1.597
 achieve+1.527
 yourself+1.514
Neg logit contributions
 girl-1.514
sed-1.513
 boy-1.482
 Pole-1.409
ometer-1.347
Token,
Token ID11
Feature activation+7.25e-01

Pos logit contributions
sed+1.145
 girl+1.096
 Pole+1.085
 boy+1.059
ometer+1.031
Neg logit contributions
ck-1.295
aches-1.255
 yourself-1.212
 avoid-1.179
irs-1.173
Token are
Token ID389
Feature activation+3.64e-01

Pos logit contributions
ck+1.225
aches+1.184
irs+1.157
 yourself+1.084
 achieve+1.083
Neg logit contributions
sed-1.546
 girl-1.543
 boy-1.516
 Pole-1.447
ometer-1.398
Token you
Token ID345
Feature activation+9.23e-01

Pos logit contributions
ck+0.379
aches+0.360
irs+0.336
 yourself+0.316
 avoid+0.311
Neg logit contributions
sed-1.013
 girl-1.001
 boy-0.984
 Pole-0.963
ometer-0.944
Token okay
Token ID8788
Feature activation-6.59e-02

Pos logit contributions
ck+1.626
aches+1.581
irs+1.547
 achieve+1.441
 avoid+1.441
Neg logit contributions
sed-1.926
 girl-1.915
 boy-1.868
 Pole-1.792
ometer-1.744
Token?"
Token ID1701
Feature activation-9.52e-01

Pos logit contributions
sed+0.178
 girl+0.175
 boy+0.172
 Pole+0.170
ometer+0.166
Neg logit contributions
ck-0.074
aches-0.071
irs-0.065
 yourself-0.064
 avoid-0.062
Token
Token ID198
Feature activation+1.95e+00

Pos logit contributions
sed+2.041
 girl+2.003
 boy+1.948
 Pole+1.946
ometer+1.893
Neg logit contributions
ck-1.562
aches-1.527
 yourself-1.502
 avoid-1.445
irs-1.444
Token
Token ID198
Feature activation+1.00e+00

Pos logit contributions
ck+4.158
aches+4.145
irs+3.889
 belong+3.809
asting+3.740
Neg logit contributions
 girl-3.305
 boy-3.289
sed-3.272
 Pole-3.108
ometer-2.970
Token followed
Token ID3940
Feature activation+9.79e-01

Pos logit contributions
sed+2.169
 girl+2.102
 Pole+2.077
 boy+2.055
ometer+2.013
Neg logit contributions
ck-2.266
aches-2.134
irs-1.990
 avoid-1.968
 yourself-1.958
Token the
Token ID262
Feature activation-1.54e+00

Pos logit contributions
ck+1.359
aches+1.347
irs+1.228
 yourself+1.213
 achieve+1.201
Neg logit contributions
 girl-2.375
sed-2.331
 boy-2.318
 Pole-2.230
ometer-2.177
Token voice
Token ID3809
Feature activation-4.75e-01

Pos logit contributions
sed+2.373
 Pole+2.174
ometer+2.152
 girl+2.125
 boy+2.058
Neg logit contributions
ck-3.502
aches-3.445
 yourself-3.416
 avoid-3.283
 happen-3.279
Token.
Token ID13
Feature activation-2.30e+00

Pos logit contributions
sed+0.933
 girl+0.893
 boy+0.875
 Pole+0.871
ometer+0.835
Neg logit contributions
ck-0.900
aches-0.861
irs-0.820
 avoid-0.808
 yourself-0.807
Token When
Token ID1649
Feature activation-8.43e-01

Pos logit contributions
sed+3.662
 girl+3.515
 boy+3.466
 Pole+3.258
 gir+3.029
Neg logit contributions
irs-5.130
 yourself-5.049
ck-4.906
 achieve-4.834
 avoid-4.830
Token she
Token ID673
Feature activation+4.53e-01

Pos logit contributions
sed+1.552
 girl+1.449
 boy+1.421
 Pole+1.408
ometer+1.377
Neg logit contributions
ck-1.690
aches-1.661
irs-1.612
 avoid-1.569
 yourself-1.567
Token stepped
Token ID10764
Feature activation-1.39e+00

Pos logit contributions
ck+1.014
aches+0.998
irs+0.959
 yourself+0.951
 avoid+0.931
Neg logit contributions
sed-0.711
 girl-0.709
 boy-0.686
 Pole-0.659
ometer-0.624
Token out
Token ID503
Feature activation-8.81e-01

Pos logit contributions
sed+3.196
 girl+2.987
 boy+2.971
 Pole+2.938
ometer+2.894
Neg logit contributions
ck-2.202
irs-2.094
aches-2.085
 yourself-2.022
 avoid-1.960
Token of
Token ID286
Feature activation+1.53e+00

Pos logit contributions
sed+2.073
 Pole+1.963
 girl+1.958
 boy+1.940
ometer+1.908
Neg logit contributions
ck-1.348
aches-1.270
irs-1.249
 avoid-1.184
 yourself-1.184
Token her
Token ID607
Feature activation-1.76e+00

Pos logit contributions
ck+2.348
aches+2.330
 achieve+2.160
 yourself+2.066
irs+2.047
Neg logit contributions
 girl-3.601
sed-3.557
 boy-3.502
 Pole-3.262
 mouse-3.242
Token cave
Token ID11527
Feature activation-1.70e+00

Pos logit contributions
sed+2.528
 girl+2.447
 Pole+2.414
 boy+2.369
ometer+2.317
Neg logit contributions
ck-3.996
aches-3.921
irs-3.861
 yourself-3.822
 avoid-3.796
Token,
Token ID11
Feature activation-1.57e-01

Pos logit contributions
sed+2.832
 girl+2.669
 Pole+2.630
 boy+2.587
ometer+2.548
Neg logit contributions
ck-2.505
aches-2.427
 yourself-2.333
irs-2.316
 avoid-2.293
Token M
Token ID337
Feature activation-6.63e-01

Pos logit contributions
sed+0.262
 girl+0.253
 boy+0.245
 Pole+0.241
ometer+0.233
Neg logit contributions
ck-0.337
aches-0.330
irs-0.317
 yourself-0.314
 avoid-0.309
Tokenandy
Token ID10757
Feature activation-1.08e+00

Pos logit contributions
sed+1.288
 girl+1.270
 boy+1.240
 Pole+1.187
ometer+1.181
Neg logit contributions
ck-1.242
aches-1.217
irs-1.174
 yourself-1.140
asting-1.105
Token.
Token ID13
Feature activation-1.34e+00

Pos logit contributions
sed+2.195
 girl+2.093
 boy+2.046
 Pole+2.031
ometer+2.001
Neg logit contributions
ck-1.970
aches-1.895
irs-1.768
 avoid-1.755
 yourself-1.753
Token
Token ID220
Feature activation+1.60e+00

Pos logit contributions
sed+2.851
 girl+2.662
 boy+2.583
 Pole+2.543
ometer+2.476
Neg logit contributions
irs-2.238
ck-2.230
aches-2.187
 yourself-2.158
 avoid-2.107
Token
Token ID198
Feature activation+1.51e+00

Pos logit contributions
ck+3.395
aches+3.240
irs+2.987
 avoid+2.956
 yourself+2.946
Neg logit contributions
 girl-2.840
 boy-2.822
sed-2.793
 Pole-2.711
ometer-2.694
Token
Token ID198
Feature activation+7.82e-01

Pos logit contributions
ck+3.115
aches+3.010
irs+2.798
 yourself+2.760
 avoid+2.732
Neg logit contributions
 girl-2.767
 boy-2.711
 Pole-2.605
sed-2.584
ometer-2.469
Token"
Token ID1
Feature activation+9.39e-02

Pos logit contributions
ck+1.310
aches+1.273
irs+1.202
 yourself+1.163
 avoid+1.147
Neg logit contributions
 girl-1.683
 boy-1.638
sed-1.630
 Pole-1.586
ometer-1.523
TokenLet
Token ID5756
Feature activation+1.33e+00

Pos logit contributions
ck+0.192
aches+0.187
irs+0.179
 yourself+0.176
 avoid+0.174
Neg logit contributions
sed-0.168
 girl-0.164
 boy-0.160
 Pole-0.156
ometer-0.150
Token's
Token ID338
Feature activation+2.14e+00

Pos logit contributions
ck+3.354
aches+3.294
irs+3.233
 achieve+3.117
 yourself+3.084
Neg logit contributions
 girl-1.851
 boy-1.788
sed-1.653
 Pole-1.602
ometer-1.469
Token go
Token ID467
Feature activation+1.60e+00

Pos logit contributions
ck+3.347
irs+3.267
aches+3.181
 yourself+2.936
lled+2.837
Neg logit contributions
 girl-4.913
 boy-4.874
sed-4.773
 Pole-4.513
 mouse-4.439
Token couldn
Token ID3521
Feature activation+2.76e+00

Pos logit contributions
ck+2.313
aches+2.210
irs+2.182
 yourself+2.120
 avoid+2.063
Neg logit contributions
 girl-1.878
sed-1.828
 boy-1.816
 Pole-1.707
ometer-1.659
Token't
Token ID470
Feature activation+2.88e+00

Pos logit contributions
ck+3.745
aches+3.552
irs+3.215
 yourself+3.040
asting+2.948
Neg logit contributions
 girl-7.212
 boy-7.152
sed-6.920
 Pole-6.867
ometer-6.862
Token get
Token ID651
Feature activation+1.26e+00

Pos logit contributions
irs+5.158
ck+4.774
 yourself+4.758
aches+4.724
cks+4.501
Neg logit contributions
 girl-6.320
 boy-6.285
sed-6.057
 mouse-5.760
 Pole-5.724
Token in
Token ID287
Feature activation+2.19e+00

Pos logit contributions
ck+2.430
aches+2.374
irs+2.324
 achieve+2.263
 yourself+2.234
Neg logit contributions
 girl-2.349
sed-2.288
 boy-2.279
 Pole-2.129
ometer-2.084
Token.
Token ID13
Feature activation-1.08e+00

Pos logit contributions
ck+4.299
aches+4.198
irs+3.901
 achieve+3.726
asting+3.644
Neg logit contributions
 girl-4.419
 boy-4.254
sed-4.165
 mouse-3.953
 eyes-3.879
Token From
Token ID3574
Feature activation+5.47e-01

Pos logit contributions
sed+2.276
 girl+2.163
 Pole+2.129
 boy+2.114
ometer+2.010
Neg logit contributions
ck-1.795
irs-1.749
aches-1.730
 yourself-1.721
 avoid-1.683
Token that
Token ID326
Feature activation-1.43e-01

Pos logit contributions
ck+1.000
aches+0.985
irs+0.951
 yourself+0.913
 achieve+0.906
Neg logit contributions
sed-1.080
 girl-1.076
 boy-1.049
 Pole-1.011
ometer-0.972
Token day
Token ID1110
Feature activation+8.77e-02

Pos logit contributions
sed+0.262
 girl+0.253
 boy+0.247
 Pole+0.245
ometer+0.235
Neg logit contributions
ck-0.286
aches-0.278
irs-0.265
 yourself-0.262
 avoid-0.260
Token on
Token ID319
Feature activation-7.22e-01

Pos logit contributions
ck+0.160
aches+0.156
irs+0.149
 yourself+0.146
 avoid+0.144
Neg logit contributions
sed-0.174
 girl-0.171
 boy-0.167
 Pole-0.163
ometer-0.157
Token,
Token ID11
Feature activation-9.31e-02

Pos logit contributions
sed+1.062
 girl+0.999
 boy+0.976
 Pole+0.975
ometer+0.926
Neg logit contributions
ck-1.698
aches-1.690
 yourself-1.624
irs-1.623
 avoid-1.596
Token Lily
Token ID20037
Feature activation-3.70e-02

Pos logit contributions
sed+0.137
 girl+0.131
 boy+0.127
 Pole+0.125
ometer+0.120
Neg logit contributions
ck-0.220
aches-0.215
irs-0.208
 yourself-0.204
 avoid-0.203
Token,
Token ID11
Feature activation-2.62e+00

Pos logit contributions
sed+3.914
 Pole+3.716
 girl+3.669
 boy+3.622
ometer+3.420
Neg logit contributions
aches-4.611
ck-4.572
 happen-4.263
 yourself-4.240
ered-4.199
Token "
Token ID366
Feature activation+3.30e-02

Pos logit contributions
sed+2.581
 girl+2.355
 boy+2.235
 Pole+2.195
ometer+2.182
Neg logit contributions
ck-7.412
aches-7.322
 yourself-7.057
 avoid-6.973
 achieve-6.958
TokenYour
Token ID7120
Feature activation-1.97e+00

Pos logit contributions
ck+0.056
aches+0.054
irs+0.051
 yourself+0.051
 avoid+0.050
Neg logit contributions
sed-0.071
 girl-0.069
 boy-0.068
 Pole-0.066
ometer-0.064
Token reward
Token ID6721
Feature activation+1.78e-01

Pos logit contributions
sed+4.141
 Pole+4.088
 girl+3.933
elt+3.836
 boy+3.815
Neg logit contributions
aches-3.259
ck-3.180
 yourself-3.167
 avoid-3.145
 happen-2.940
Token is
Token ID318
Feature activation+3.14e-01

Pos logit contributions
ck+0.361
aches+0.351
irs+0.337
 yourself+0.330
 avoid+0.326
Neg logit contributions
sed-0.318
 girl-0.313
 boy-0.304
 Pole-0.296
ometer-0.285
Token a
Token ID257
Feature activation-1.64e+00

Pos logit contributions
ck+0.543
aches+0.524
irs+0.502
 yourself+0.487
 avoid+0.481
Neg logit contributions
sed-0.656
 girl-0.650
 boy-0.635
 Pole-0.618
ometer-0.601
Token big
Token ID1263
Feature activation-2.32e+00

Pos logit contributions
sed+3.834
 Pole+3.657
 girl+3.492
elt+3.473
 boy+3.435
Neg logit contributions
aches-2.423
ck-2.383
 avoid-2.317
 yourself-2.290
 achieve-2.249
Token,
Token ID11
Feature activation-1.74e+00

Pos logit contributions
sed+4.870
 Pole+4.739
ometer+4.511
elt+4.497
 girl+4.292
Neg logit contributions
ck-3.886
aches-3.809
 avoid-3.714
 yourself-3.707
 happen-3.701
Token y
Token ID331
Feature activation-7.55e-01

Pos logit contributions
sed+3.587
 Pole+3.353
ometer+3.244
 girl+3.224
 boy+3.144
Neg logit contributions
ck-3.184
aches-3.061
 yourself-2.991
 avoid-2.876
asting-2.810
Tokenummy
Token ID13513
Feature activation-1.95e+00

Pos logit contributions
sed+0.803
 girl+0.725
 boy+0.692
 Pole+0.672
ometer+0.627
Neg logit contributions
ck-2.121
aches-2.082
irs-2.034
 yourself-2.003
 avoid-1.962
Token cookie
Token ID19751
Feature activation-2.58e-01

Pos logit contributions
sed+3.881
 Pole+3.611
ometer+3.459
 girl+3.357
elt+3.341
Neg logit contributions
aches-3.692
ck-3.638
 yourself-3.440
 happen-3.400
 avoid-3.333
Token She
Token ID1375
Feature activation-9.92e-01

Pos logit contributions
 girl+3.736
sed+3.696
 Pole+3.575
 boy+3.506
ometer+3.288
Neg logit contributions
ck-3.839
 yourself-3.796
irs-3.773
aches-3.743
 avoid-3.736
Token found
Token ID1043
Feature activation+6.28e-01

Pos logit contributions
sed+1.822
 girl+1.792
 Pole+1.770
 boy+1.724
ometer+1.714
Neg logit contributions
ck-1.982
aches-1.892
 avoid-1.759
 yourself-1.741
irs-1.727
Token another
Token ID1194
Feature activation-8.03e-01

Pos logit contributions
ck+0.896
aches+0.870
irs+0.818
 achieve+0.795
 yourself+0.788
Neg logit contributions
sed-1.504
 girl-1.502
 boy-1.474
 Pole-1.410
ometer-1.388
Token tw
Token ID665
Feature activation+2.05e+00

Pos logit contributions
sed+1.293
 girl+1.213
 Pole+1.210
 boy+1.163
ometer+1.152
Neg logit contributions
ck-1.769
aches-1.724
 avoid-1.633
 yourself-1.625
irs-1.603
Tokenig
Token ID328
Feature activation+2.36e-02

Pos logit contributions
irs+5.218
ck+5.198
aches+5.032
 achieve+4.969
 ourselves+4.965
Neg logit contributions
sed-2.590
 girl-2.477
 boy-2.374
ometer-2.320
 Pole-2.286
Token and
Token ID290
Feature activation-3.29e-01

Pos logit contributions
ck+0.048
aches+0.047
irs+0.045
 yourself+0.044
 avoid+0.044
Neg logit contributions
sed-0.042
 girl-0.041
 boy-0.040
 Pole-0.039
ometer-0.037
Token pretended
Token ID36766
Feature activation+1.25e-01

Pos logit contributions
sed+0.672
 girl+0.659
 boy+0.641
 Pole+0.640
ometer+0.619
Neg logit contributions
ck-0.585
aches-0.564
irs-0.527
 yourself-0.524
 avoid-0.520
Token to
Token ID284
Feature activation+2.78e+00

Pos logit contributions
ck+0.218
aches+0.212
irs+0.202
 yourself+0.197
 avoid+0.194
Neg logit contributions
sed-0.258
 girl-0.255
 boy-0.249
 Pole-0.243
ometer-0.235
Token fight
Token ID1907
Feature activation+4.18e-01

Pos logit contributions
irs+5.614
ck+5.114
aches+5.061
asting+5.019
 yourself+4.747
Neg logit contributions
 boy-5.680
sed-5.611
 girl-5.597
 gir-5.178
 Pole-5.096
Token with
Token ID351
Feature activation+1.62e+00

Pos logit contributions
ck+0.685
aches+0.666
irs+0.630
 yourself+0.609
 avoid+0.600
Neg logit contributions
sed-0.906
 girl-0.903
 boy-0.886
 Pole-0.855
ometer-0.833
Token Tom
Token ID4186
Feature activation-4.00e-01

Pos logit contributions
ck+2.511
aches+2.478
 achieve+2.341
irs+2.298
asting+2.262
Neg logit contributions
sed-3.697
 girl-3.650
 boy-3.606
 eyes-3.410
 Pole-3.375
Token and
Token ID290
Feature activation-1.49e+00

Pos logit contributions
sed+2.439
 girl+2.360
 boy+2.281
 Pole+2.271
ometer+2.166
Neg logit contributions
ck-2.951
aches-2.904
 achieve-2.730
 avoid-2.720
irs-2.715
Token Mia
Token ID32189
Feature activation-3.14e-01

Pos logit contributions
sed+2.951
 Pole+2.868
 girl+2.836
ometer+2.797
 boy+2.742
Neg logit contributions
ck-2.828
aches-2.711
 avoid-2.506
irs-2.458
 yourself-2.429
Token picked
Token ID6497
Feature activation-2.61e-01

Pos logit contributions
sed+0.527
 girl+0.517
 boy+0.502
 Pole+0.493
ometer+0.477
Neg logit contributions
ck-0.669
aches-0.651
irs-0.618
 yourself-0.610
 avoid-0.608
Token him
Token ID683
Feature activation+8.30e-01

Pos logit contributions
sed+0.619
 girl+0.604
 boy+0.591
 Pole+0.590
ometer+0.570
Neg logit contributions
ck-0.380
aches-0.364
irs-0.343
 yourself-0.335
 avoid-0.327
Token up
Token ID510
Feature activation+2.52e-01

Pos logit contributions
ck+1.572
aches+1.525
irs+1.497
 yourself+1.439
 achieve+1.432
Neg logit contributions
 girl-1.565
sed-1.545
 boy-1.527
 Pole-1.444
ometer-1.425
Token and
Token ID290
Feature activation-1.08e+00

Pos logit contributions
ck+0.484
aches+0.471
irs+0.452
 yourself+0.441
 avoid+0.435
Neg logit contributions
sed-0.479
 girl-0.473
 boy-0.461
 Pole-0.447
ometer-0.433
Token put
Token ID1234
Feature activation+8.61e-01

Pos logit contributions
sed+1.984
 girl+1.911
 Pole+1.896
 boy+1.854
ometer+1.843
Neg logit contributions
ck-2.179
aches-2.100
 avoid-1.952
irs-1.951
 yourself-1.936
Token him
Token ID683
Feature activation+5.63e-01

Pos logit contributions
ck+1.143
aches+1.111
irs+1.042
 achieve+1.039
 yourself+1.008
Neg logit contributions
 girl-2.131
sed-2.119
 boy-2.091
 Pole-1.996
ometer-1.959
Token inside
Token ID2641
Feature activation+8.85e-01

Pos logit contributions
ck+1.060
aches+1.028
irs+1.006
 yourself+0.969
 achieve+0.964
Neg logit contributions
sed-1.072
 girl-1.062
 boy-1.034
 Pole-0.991
ometer-0.968
Token her
Token ID607
Feature activation-7.94e-01

Pos logit contributions
ck+1.699
aches+1.641
irs+1.572
 yourself+1.529
 achieve+1.517
Neg logit contributions
 girl-1.688
sed-1.656
 boy-1.643
 Pole-1.555
ometer-1.518
Token bin
Token ID9874
Feature activation-1.33e+00

Pos logit contributions
sed+1.191
 girl+1.176
 boy+1.142
 Pole+1.127
ometer+1.067
Neg logit contributions
ck-1.799
aches-1.787
 yourself-1.699
 avoid-1.668
irs-1.659
Token not
Token ID407
Feature activation+6.08e-01

Pos logit contributions
ck+1.667
aches+1.616
irs+1.594
 achieve+1.546
 avoid+1.516
Neg logit contributions
sed-1.872
 girl-1.819
 boy-1.786
 Pole-1.697
ometer-1.686
Token.
Token ID13
Feature activation-1.61e+00

Pos logit contributions
ck+1.030
aches+0.996
irs+0.968
 avoid+0.918
 achieve+0.917
Neg logit contributions
sed-1.286
 girl-1.284
 boy-1.257
 Pole-1.198
ometer-1.172
Token It
Token ID632
Feature activation-1.86e-01

Pos logit contributions
 girl+2.689
sed+2.680
 boy+2.570
 Pole+2.538
ometer+2.244
Neg logit contributions
 yourself-3.382
ck-3.378
aches-3.274
irs-3.232
 ourselves-3.146
Token's
Token ID338
Feature activation+8.27e-01

Pos logit contributions
sed+0.305
 girl+0.298
 boy+0.288
 Pole+0.284
ometer+0.270
Neg logit contributions
ck-0.405
aches-0.396
irs-0.376
 yourself-0.375
 avoid-0.368
Token mine
Token ID6164
Feature activation+5.76e-01

Pos logit contributions
ck+1.405
aches+1.361
irs+1.335
 achieve+1.287
 avoid+1.269
Neg logit contributions
sed-1.743
 girl-1.709
 boy-1.677
 Pole-1.596
ometer-1.576
Token.
Token ID13
Feature activation-1.29e+00

Pos logit contributions
ck+0.974
aches+0.939
irs+0.905
 achieve+0.855
 yourself+0.855
Neg logit contributions
sed-1.227
 girl-1.213
 boy-1.189
 Pole-1.153
ometer-1.127
Token Mom
Token ID11254
Feature activation+1.11e-01

Pos logit contributions
sed+2.240
 girl+2.211
 boy+2.127
 Pole+2.119
ometer+1.931
Neg logit contributions
ck-2.652
 yourself-2.626
aches-2.545
irs-2.517
 avoid-2.435
Token gave
Token ID2921
Feature activation+1.38e+00

Pos logit contributions
ck+0.219
aches+0.213
irs+0.204
 yourself+0.201
 avoid+0.198
Neg logit contributions
sed-0.202
 girl-0.199
 boy-0.193
 Pole-0.188
ometer-0.182
Token it
Token ID340
Feature activation+1.90e-01

Pos logit contributions
ck+1.943
aches+1.913
 achieve+1.822
irs+1.781
 avoid+1.722
Neg logit contributions
 girl-3.383
 boy-3.313
sed-3.235
 Pole-3.066
ometer-3.039
Token to
Token ID284
Feature activation+2.47e+00

Pos logit contributions
ck+0.387
aches+0.377
irs+0.362
 yourself+0.355
 avoid+0.351
Neg logit contributions
sed-0.334
 girl-0.329
 boy-0.320
 Pole-0.311
ometer-0.300
Token me
Token ID502
Feature activation+5.34e-01

Pos logit contributions
ck+4.996
aches+4.943
irs+4.803
 achieve+4.650
lled+4.547
Neg logit contributions
 girl-4.663
 boy-4.551
sed-4.507
 Pole-4.059
 eyes-4.047
Token.
Token ID13
Feature activation-9.90e-01

Pos logit contributions
sed+0.125
 girl+0.123
 boy+0.119
 Pole+0.117
ometer+0.113
Neg logit contributions
ck-0.117
aches-0.113
irs-0.107
 yourself-0.106
 avoid-0.105
Token He
Token ID679
Feature activation-4.22e-01

Pos logit contributions
sed+1.338
 girl+1.285
 Pole+1.211
 boy+1.183
ometer+1.134
Neg logit contributions
ck-2.421
aches-2.359
 avoid-2.328
irs-2.327
 yourself-2.326
Token tried
Token ID3088
Feature activation+1.16e-01

Pos logit contributions
sed+0.752
 girl+0.749
 Pole+0.721
 boy+0.720
ometer+0.698
Neg logit contributions
ck-0.856
aches-0.825
irs-0.773
 avoid-0.770
 yourself-0.769
Token to
Token ID284
Feature activation+3.07e+00

Pos logit contributions
ck+0.247
aches+0.241
irs+0.232
 yourself+0.228
 avoid+0.225
Neg logit contributions
sed-0.195
 girl-0.192
 boy-0.186
 Pole-0.181
ometer-0.174
Token grab
Token ID5552
Feature activation+6.66e-01

Pos logit contributions
irs+5.759
aches+5.408
 yourself+5.083
asting+5.063
 achieve+5.048
Neg logit contributions
 girl-6.800
 boy-6.799
sed-6.136
 mouse-5.707
 dog-5.565
Token the
Token ID262
Feature activation+7.26e-02

Pos logit contributions
ck+0.864
aches+0.848
irs+0.790
 achieve+0.767
 yourself+0.757
Neg logit contributions
 girl-1.666
sed-1.650
 boy-1.640
 Pole-1.559
ometer-1.530
Token chain
Token ID6333
Feature activation-1.54e-02

Pos logit contributions
ck+0.163
aches+0.159
irs+0.153
 yourself+0.151
 avoid+0.149
Neg logit contributions
sed-0.114
 girl-0.112
 boy-0.108
 Pole-0.106
ometer-0.101
Token from
Token ID422
Feature activation+8.77e-01

Pos logit contributions
sed+0.028
 girl+0.027
 boy+0.026
 Pole+0.026
ometer+0.025
Neg logit contributions
ck-0.031
aches-0.030
irs-0.029
 yourself-0.028
 avoid-0.028
Token them
Token ID606
Feature activation-8.88e-01

Pos logit contributions
ck+1.035
aches+1.020
irs+0.951
 achieve+0.905
 yourself+0.901
Neg logit contributions
sed-2.311
 girl-2.309
 boy-2.281
 Pole-2.184
ometer-2.093
Token.
Token ID13
Feature activation-8.92e-01

Pos logit contributions
sed+1.927
 girl+1.883
 Pole+1.816
 boy+1.813
ometer+1.731
Neg logit contributions
ck-1.497
aches-1.434
 yourself-1.357
irs-1.352
 avoid-1.342
Token
Token ID198
Feature activation+2.63e+00

Pos logit contributions
sed+1.320
 girl+1.315
 boy+1.220
 Pole+1.209
ometer+1.131
Neg logit contributions
ck-2.047
aches-2.006
 avoid-1.955
irs-1.945
 yourself-1.936
Token<|endoftext|>
Token ID50256
Feature activation+6.67e-01

Pos logit contributions
sed+2.364
 girl+2.287
 Pole+2.271
 boy+2.232
ometer+2.159
Neg logit contributions
ck-1.644
irs-1.586
aches-1.565
 yourself-1.550
 avoid-1.496
TokenOnce
Token ID7454
Feature activation+1.19e+00

Pos logit contributions
ck+1.419
aches+1.380
irs+1.307
 yourself+1.277
 avoid+1.269
Neg logit contributions
 girl-1.146
sed-1.129
 boy-1.123
 Pole-1.070
ometer-1.020
Token there
Token ID612
Feature activation+6.16e-01

Pos logit contributions
ck+2.551
aches+2.451
 yourself+2.369
irs+2.338
 achieve+2.291
Neg logit contributions
 girl-2.050
 boy-2.015
sed-1.994
 Pole-1.895
ometer-1.822
Token was
Token ID373
Feature activation+1.09e+00

Pos logit contributions
ck+1.385
aches+1.347
irs+1.309
 yourself+1.284
 achieve+1.256
Neg logit contributions
 girl-0.965
sed-0.962
 boy-0.946
 Pole-0.889
ometer-0.851
Token a
Token ID257
Feature activation-2.56e+00

Pos logit contributions
ck+1.682
aches+1.650
irs+1.533
 achieve+1.486
 yourself+1.478
Neg logit contributions
 girl-2.507
 boy-2.459
sed-2.452
 Pole-2.303
 mouse-2.244
Token kind
Token ID1611
Feature activation-2.79e+00

Pos logit contributions
sed+5.108
elt+4.518
 Pole+4.455
ometer+4.409
 eyes+4.234
Neg logit contributions
ck-5.154
 avoid-4.978
aches-4.909
 achieve-4.729
irl-4.587
Token dog
Token ID3290
Feature activation-1.76e+00

Pos logit contributions
sed+4.838
 Pole+3.794
ometer+3.766
elt+3.622
 cheeks+3.488
Neg logit contributions
aches-6.393
ck-6.271
asting-6.014
 yourself-5.952
 know-5.917
Token named
Token ID3706
Feature activation-2.38e-01

Pos logit contributions
sed+3.686
 girl+3.496
 Pole+3.484
 boy+3.358
ometer+3.357
Neg logit contributions
ck-3.204
aches-3.037
e-2.862
 avoid-2.824
 yourself-2.803
Token Spot
Token ID15899
Feature activation+2.21e-01

Pos logit contributions
sed+0.351
 girl+0.344
 boy+0.330
 Pole+0.323
ometer+0.310
Neg logit contributions
ck-0.552
aches-0.538
irs-0.523
 yourself-0.517
 avoid-0.509
Token.
Token ID13
Feature activation-4.78e-01

Pos logit contributions
ck+0.408
aches+0.397
irs+0.380
 yourself+0.371
 avoid+0.365
Neg logit contributions
sed-0.435
 girl-0.430
 boy-0.420
 Pole-0.408
ometer-0.395
Token He
Token ID679
Feature activation-4.77e-01

Pos logit contributions
sed+0.727
 girl+0.720
 Pole+0.683
 boy+0.682
ometer+0.655
Neg logit contributions
ck-1.075
aches-1.051
irs-1.007
 yourself-1.007
 avoid-1.002
Token he
Token ID339
Feature activation+8.69e-01

Pos logit contributions
ck+0.508
aches+0.494
irs+0.476
 yourself+0.466
 avoid+0.459
Neg logit contributions
sed-0.418
 girl-0.413
 boy-0.402
 Pole-0.389
ometer-0.374
Token were
Token ID547
Feature activation+1.27e+00

Pos logit contributions
ck+1.922
aches+1.896
irs+1.865
 yourself+1.808
 achieve+1.769
Neg logit contributions
sed-1.394
 girl-1.361
 boy-1.337
 Pole-1.259
ometer-1.210
Token well
Token ID880
Feature activation+3.83e-01

Pos logit contributions
ck+2.513
aches+2.423
irs+2.418
 achieve+2.270
 yourself+2.264
Neg logit contributions
sed-2.293
 girl-2.286
 boy-2.218
 Pole-2.120
ometer-2.005
Token and
Token ID290
Feature activation+1.02e+00

Pos logit contributions
ck+0.781
aches+0.761
irs+0.733
 yourself+0.714
 avoid+0.706
Neg logit contributions
sed-0.675
 girl-0.669
 boy-0.650
 Pole-0.630
ometer-0.607
Token free
Token ID1479
Feature activation-6.61e-02

Pos logit contributions
ck+1.937
aches+1.873
irs+1.845
 yourself+1.740
 achieve+1.704
Neg logit contributions
sed-1.962
 girl-1.950
 boy-1.916
 Pole-1.817
ometer-1.753
Token.
Token ID13
Feature activation-2.81e+00

Pos logit contributions
sed+0.132
 girl+0.130
 boy+0.126
 Pole+0.124
ometer+0.120
Neg logit contributions
ck-0.120
aches-0.117
irs-0.111
 yourself-0.110
 avoid-0.108
Token He
Token ID679
Feature activation-4.42e-01

Pos logit contributions
sed+4.513
 girl+4.398
 Pole+4.056
 boy+3.945
 Girl+3.684
Neg logit contributions
 avoid-6.211
 yourself-6.193
aches-6.096
 achieve-6.035
ck-6.000
Token imagined
Token ID15758
Feature activation+1.19e+00

Pos logit contributions
sed+0.824
 girl+0.803
 Pole+0.784
 boy+0.775
ometer+0.751
Neg logit contributions
ck-0.873
aches-0.840
 avoid-0.781
irs-0.778
 yourself-0.774
Token that
Token ID326
Feature activation-1.15e-01

Pos logit contributions
ck+2.401
aches+2.317
irs+2.268
 yourself+2.174
 achieve+2.145
Neg logit contributions
sed-2.173
 girl-2.113
 boy-2.077
 Pole-2.031
ometer-1.867
Token he
Token ID339
Feature activation+9.75e-01

Pos logit contributions
sed+0.258
 girl+0.252
 boy+0.245
 Pole+0.243
ometer+0.236
Neg logit contributions
ck-0.180
aches-0.175
irs-0.164
 yourself-0.163
 avoid-0.160
Token could
Token ID714
Feature activation+2.80e+00

Pos logit contributions
ck+2.229
aches+2.195
irs+2.165
 yourself+2.092
 achieve+2.038
Neg logit contributions
sed-1.484
 girl-1.483
 boy-1.453
 Pole-1.349
ometer-1.301
Token One
Token ID1881
Feature activation-1.05e+00

Pos logit contributions
sed+0.250
 girl+0.244
 boy+0.236
 Pole+0.230
ometer+0.221
Neg logit contributions
ck-0.344
aches-0.337
irs-0.328
 yourself-0.322
 avoid-0.317
Token day
Token ID1110
Feature activation-1.60e+00

Pos logit contributions
sed+2.170
 Pole+2.019
 girl+1.938
ometer+1.909
 boy+1.890
Neg logit contributions
ck-1.905
aches-1.858
 avoid-1.744
asting-1.713
 yourself-1.709
Token,
Token ID11
Feature activation-4.87e-01

Pos logit contributions
sed+2.750
 Pole+2.533
 girl+2.492
 boy+2.445
ometer+2.413
Neg logit contributions
ck-3.433
aches-3.351
irs-3.335
 avoid-3.231
 yourself-3.204
Token Sarah
Token ID10490
Feature activation-9.74e-01

Pos logit contributions
sed+0.883
 girl+0.834
 Pole+0.822
 boy+0.814
ometer+0.792
Neg logit contributions
ck-0.976
aches-0.949
irs-0.922
 yourself-0.904
 avoid-0.899
Token and
Token ID290
Feature activation-8.19e-04

Pos logit contributions
sed+1.778
 girl+1.641
 Pole+1.612
ometer+1.583
 boy+1.580
Neg logit contributions
ck-1.992
aches-1.917
 avoid-1.810
 yourself-1.809
irs-1.788
Token her
Token ID607
Feature activation-2.59e+00

Pos logit contributions
sed+0.002
 girl+0.002
 boy+0.002
 Pole+0.002
ometer+0.002
Neg logit contributions
ck-0.001
aches-0.001
irs-0.001
 yourself-0.001
 avoid-0.001
Token mom
Token ID1995
Feature activation-5.11e-01

Pos logit contributions
 Pole+4.330
sed+4.125
ometer+3.916
 girl+3.774
 boy+3.684
Neg logit contributions
aches-5.465
ck-5.372
 yourself-5.162
 avoid-5.142
 ease-5.031
Token went
Token ID1816
Feature activation-6.40e-01

Pos logit contributions
sed+0.937
 girl+0.883
 Pole+0.863
 boy+0.856
ometer+0.837
Neg logit contributions
ck-1.031
aches-0.997
irs-0.956
 yourself-0.931
 avoid-0.926
Token shopping
Token ID9735
Feature activation+4.05e-01

Pos logit contributions
sed+1.190
 girl+1.148
 boy+1.122
 Pole+1.107
ometer+1.061
Neg logit contributions
ck-1.248
aches-1.216
irs-1.177
 yourself-1.147
 avoid-1.144
Token.
Token ID13
Feature activation-8.18e-01

Pos logit contributions
ck+0.854
aches+0.830
irs+0.799
 yourself+0.781
 avoid+0.770
Neg logit contributions
 girl-0.692
sed-0.692
 boy-0.673
 Pole-0.648
ometer-0.622
Token When
Token ID1649
Feature activation-1.59e-01

Pos logit contributions
sed+1.333
 girl+1.270
 boy+1.227
 Pole+1.218
ometer+1.168
Neg logit contributions
ck-1.745
irs-1.707
aches-1.681
 yourself-1.642
 avoid-1.641
Token that
Token ID326
Feature activation+1.14e-01

Pos logit contributions
sed+2.364
 girl+2.276
 Pole+2.199
 boy+2.187
ometer+2.103
Neg logit contributions
ck-2.057
aches-1.968
 yourself-1.922
 avoid-1.883
 achieve-1.843
Token he
Token ID339
Feature activation-9.61e-01

Pos logit contributions
ck+0.252
aches+0.246
irs+0.236
 yourself+0.233
 avoid+0.230
Neg logit contributions
sed-0.185
 girl-0.182
 boy-0.176
 Pole-0.171
ometer-0.164
Token started
Token ID2067
Feature activation-7.00e-02

Pos logit contributions
sed+1.793
 girl+1.761
 Pole+1.712
 boy+1.702
ometer+1.665
Neg logit contributions
ck-1.906
aches-1.804
 yourself-1.690
 avoid-1.679
irs-1.651
Token to
Token ID284
Feature activation+1.50e+00

Pos logit contributions
sed+0.118
 girl+0.116
 boy+0.113
 Pole+0.110
ometer+0.105
Neg logit contributions
ck-0.149
aches-0.145
irs-0.139
 yourself-0.137
 avoid-0.136
Token weep
Token ID49671
Feature activation-2.51e-01

Pos logit contributions
ck+2.780
aches+2.728
irs+2.714
 yourself+2.548
asting+2.497
Neg logit contributions
 girl-2.995
sed-2.958
 boy-2.926
 Pole-2.648
 mouse-2.619
Token.
Token ID13
Feature activation-2.94e+00

Pos logit contributions
sed+0.509
 girl+0.495
 boy+0.483
 Pole+0.480
ometer+0.458
Neg logit contributions
ck-0.450
aches-0.435
 yourself-0.414
irs-0.410
 avoid-0.405
Token He
Token ID679
Feature activation-6.05e-01

Pos logit contributions
sed+4.688
 girl+4.506
 Pole+4.231
 boy+4.123
suit+3.820
Neg logit contributions
 yourself-6.851
 achieve-6.458
 avoid-6.454
 know-6.411
aches-6.276
Token felt
Token ID2936
Feature activation+1.68e+00

Pos logit contributions
sed+1.120
 girl+1.102
 Pole+1.082
 boy+1.068
ometer+1.032
Neg logit contributions
ck-1.207
aches-1.143
 avoid-1.072
irs-1.061
 yourself-1.061
Token so
Token ID523
Feature activation+1.40e+00

Pos logit contributions
ck+4.032
aches+4.012
irs+3.962
 achieve+3.680
 yourself+3.666
Neg logit contributions
 girl-2.509
 boy-2.424
sed-2.392
 mouse-2.095
 Pole-2.079
Token sorry
Token ID7926
Feature activation-8.77e-01

Pos logit contributions
ck+3.543
aches+3.524
irs+3.471
 achieve+3.249
 yourself+3.244
Neg logit contributions
 girl-1.850
sed-1.803
 boy-1.780
 Pole-1.599
ometer-1.514
Token for
Token ID329
Feature activation+1.84e+00

Pos logit contributions
sed+1.667
 girl+1.606
 Pole+1.566
 boy+1.550
ometer+1.464
Neg logit contributions
ck-1.656
aches-1.606
 yourself-1.567
 avoid-1.538
 achieve-1.512
Token her
Token ID607
Feature activation-7.61e-01

Pos logit contributions
ck+3.235
aches+3.209
 achieve+2.966
irs+2.935
irl+2.815
Neg logit contributions
 girl-5.036
 boy-4.895
 mouse-4.447
sed-4.296
 Pole-4.274
Token room
Token ID2119
Feature activation-5.85e-01

Pos logit contributions
sed+0.983
 girl+0.926
 Pole+0.905
 boy+0.889
ometer+0.836
Neg logit contributions
ck-1.884
aches-1.850
 yourself-1.822
irs-1.785
 avoid-1.776
Token.
Token ID13
Feature activation-1.53e+00

Pos logit contributions
sed+1.237
 girl+1.219
 boy+1.187
 Pole+1.172
ometer+1.133
Neg logit contributions
ck-0.992
aches-0.972
 yourself-0.906
irs-0.891
 avoid-0.880
Token It
Token ID632
Feature activation-2.26e-01

Pos logit contributions
sed+2.535
 Pole+2.251
 girl+2.243
 boy+2.158
ometer+2.063
Neg logit contributions
ck-3.290
aches-3.282
 yourself-3.263
irs-3.217
 avoid-3.158
Token was
Token ID373
Feature activation-2.99e-01

Pos logit contributions
sed+0.391
 girl+0.381
 boy+0.368
 Pole+0.364
ometer+0.348
Neg logit contributions
ck-0.475
aches-0.461
irs-0.439
 yourself-0.435
 avoid-0.430
Token big
Token ID1263
Feature activation-2.56e+00

Pos logit contributions
sed+0.651
 girl+0.643
 boy+0.628
 Pole+0.623
ometer+0.599
Neg logit contributions
ck-0.486
aches-0.470
irs-0.441
 yourself-0.438
 avoid-0.428
Token and
Token ID290
Feature activation-7.28e-01

Pos logit contributions
sed+5.003
 Pole+4.529
ometer+4.523
 girl+4.340
 boy+4.184
Neg logit contributions
ck-4.948
 yourself-4.902
aches-4.617
 happen-4.608
 avoid-4.488
Token colorful
Token ID20239
Feature activation-2.75e+00

Pos logit contributions
sed+1.346
 girl+1.278
 Pole+1.262
 boy+1.236
ometer+1.201
Neg logit contributions
ck-1.441
aches-1.395
 yourself-1.323
 avoid-1.298
irs-1.295
Token,
Token ID11
Feature activation-8.74e-01

Pos logit contributions
sed+6.399
 Pole+5.920
 girl+5.831
 boy+5.676
ometer+5.648
Neg logit contributions
 yourself-4.355
ck-4.331
aches-4.248
 happen-4.229
 ourselves-4.002
Token but
Token ID475
Feature activation+5.50e-03

Pos logit contributions
sed+1.531
 girl+1.464
 Pole+1.447
 boy+1.414
ometer+1.392
Neg logit contributions
ck-1.809
aches-1.763
 yourself-1.714
 avoid-1.657
irs-1.633
Token she
Token ID673
Feature activation-6.81e-01

Pos logit contributions
ck+0.009
aches+0.009
irs+0.008
 yourself+0.008
 avoid+0.008
Neg logit contributions
sed-0.012
 girl-0.012
 boy-0.011
 Pole-0.011
ometer-0.011