TOP ACTIVATIONS
MAX = 3.615

solution was a seven - buck et array that stored the
against former President Barack Obama Barack Hussein Obama With low birth
MORE as a 68 - year - old woman 's cause
sup <|endoftext|> A 67 - year - old homeless woman known
former President Barack Obama Barack Hussein Obama With low birth rate
season . The 37 - year - old Hur tt just
Michael Flynn The 57 - year - old retired lieutenant general
said that the 26 - year - old Reed , who
Wilson . The 26 - year - old offensive tackle also
of beating an 88 - year - old World War II
Clinton Hillary Diane Rodham Clinton Sanders : ' I fully expect
a train - and - equ ip program to help K
, but the 31 - year - old left guard is
said the 82 - year - old grandmother .
do itself when correspondent Jesse Wat ters ventured down to New
The legendary 78 - year - old developer picked up
The 89 - year - old said :
be due to a military retirement fund , according to the
Clinton Hillary Diane Rodham Clinton Sanders : ' I fully expect
trees bearing a high - t ide mark of black oil

MIN ACTIVATIONS
MIN = -3.592

. At midnight , a cr on job or some other
Keller , with a doll cr ad led in her lap
F alling together in a cr um pled mess of mutual
group , including headset , cr anks and brakes . The
rave drugs come off their cr utch . There
f unt p ère , cr é ateur origin el du
way through a press of cr in oline and dra per
cast their ballots before the Nov . 8 election .
rat rice , A u cr é p usc ule de
, the police announced the fo iling of an Islamic Jihad
to Europe in the first half of 2010 , with the
large , running a daily cr on job would be comput
was prepared soon after N DA Government came to power .
paid for all those programmers cr anking out the overhaul ed
same measure is repeated ; etc . So this
41 - year - old cr ick eter added .
about : L oss of or curse on magical
" fab ric ation , fals ification or plagiar ism "
pay all its bills on Oct . 2 if Washington does
have changed . Ice has cr umbled off of the glacier

INTERVAL 1.814 - 3.615
CONTAINS 0.389%

type - 94 Jin - class and type - 96 ),
to exposure to air and water pollution , according to a
Peter Lur ie and Sidney Wolfe accuse the NIH , WHO
A three - year deal at $ 11 .
ific ... Flag Flag Reply 14 Recommend

INTERVAL 0.012 - 1.814
CONTAINS 48.960%

all of its most memorable moments . <|endoftext|> What do you
tell ( who was responsible ) because this was a situation
, a Chinatown native . Despite this , neighborhood
product . Atom Freeman : I would say D ines h
what we 've always done . As a result of that

INTERVAL -1.790 - 0.012
CONTAINS 50.348%

10 ." On Monday , the American Academy of
and Count ing from the air . With some
inv ader . In 480 BCE the Persian king Xer xes
he continues to investigate the intriguing and often terrifying intersection of
they have yet to reckon with A vel ine , an

INTERVAL -3.592 - -1.790
CONTAINS 0.304%

humanity s d ross . This will be only
groups financed include the National Council of La R aza and
: 200 W OC : MSI After burn er
Obama delivered . On October 8 , Obama was asked
at the irrational ex uber ance that once valued a bitcoin
Token solution
Token ID4610
Feature activation+5.24e-01

Pos logit contributions
ンジ+1.398
+1.318
 Valhalla+1.289
 pose+1.266
+1.256
Neg logit contributions
ours-2.247
oured-2.239
ateral-1.937
hap-1.919
aths-1.908
Token was
Token ID373
Feature activation+3.25e-01

Pos logit contributions
ンジ+1.183
+1.123
 Valhalla+1.099
 pose+1.081
+1.073
Neg logit contributions
ours-1.609
oured-1.602
ateral-1.371
hap-1.358
aths-1.349
Token a
Token ID257
Feature activation+6.75e-01

Pos logit contributions
ンジ+0.739
+0.702
 Valhalla+0.689
 pose+0.678
+0.672
Neg logit contributions
ours-0.989
oured-0.985
ateral-0.842
hap-0.833
aths-0.828
Token seven
Token ID3598
Feature activation+5.73e-01

Pos logit contributions
ンジ+1.414
+1.336
 Valhalla+1.314
 pose+1.290
+1.275
Neg logit contributions
ours-2.174
oured-2.166
ateral-1.870
hap-1.851
aths-1.841
Token-
Token ID12
Feature activation+9.37e-01

Pos logit contributions
ンジ+1.139
+1.073
 Valhalla+1.045
 pose+1.024
+1.019
Neg logit contributions
ours-1.914
oured-1.908
ateral-1.655
hap-1.639
aths-1.631
Tokenbuck
Token ID27041
Feature activation+3.62e+00

Pos logit contributions
ンジ+2.163
+2.041
 pose+2.008
 Valhalla+2.004
+1.975
Neg logit contributions
ours-2.781
oured-2.780
ateral-2.365
hap-2.334
yss-2.330
Tokenet
Token ID316
Feature activation+6.94e-01

Pos logit contributions
ンジ+8.105
 pose+7.904
 Valhalla+7.669
+7.539
 dream+7.520
Neg logit contributions
ours-10.059
oured-9.968
hap-8.480
yss-8.397
ateral-8.390
Token array
Token ID7177
Feature activation-4.04e-01

Pos logit contributions
ンジ+1.455
+1.375
 Valhalla+1.347
 pose+1.323
+1.312
Neg logit contributions
ours-2.237
oured-2.228
ateral-1.922
hap-1.905
aths-1.895
Token that
Token ID326
Feature activation+1.03e-01

Pos logit contributions
oured+1.164
ours+1.164
ateral+0.984
hap+0.967
aths+0.962
Neg logit contributions
ンジ-0.986
-0.931
 Valhalla-0.914
-0.894
 pose-0.888
Token stored
Token ID8574
Feature activation+9.29e-01

Pos logit contributions
ンジ+0.227
+0.215
 Valhalla+0.209
+0.205
 pose+0.204
Neg logit contributions
ours-0.322
oured-0.321
ateral-0.275
hap-0.272
aths-0.272
Token the
Token ID262
Feature activation+2.83e-01

Pos logit contributions
ンジ+1.944
+1.837
 Valhalla+1.819
 pose+1.789
+1.761
Neg logit contributions
ours-2.975
oured-2.966
ateral-2.558
hap-2.538
aths-2.514
Token against
Token ID1028
Feature activation+3.50e-01

Pos logit contributions
ンジ+0.105
+0.100
 Valhalla+0.097
+0.095
 pose+0.095
Neg logit contributions
ours-0.119
oured-0.119
ateral-0.100
hap-0.099
aths-0.099
Token former
Token ID1966
Feature activation+1.72e+00

Pos logit contributions
ンジ+0.761
+0.721
 Valhalla+0.704
 pose+0.691
+0.688
Neg logit contributions
ours-1.103
oured-1.099
ateral-0.945
hap-0.935
aths-0.930
Token President
Token ID1992
Feature activation-2.75e-02

Pos logit contributions
ンジ+3.968
 Valhalla+3.741
+3.721
 pose+3.706
+3.589
Neg logit contributions
oured-5.106
ours-5.105
ateral-4.380
hap-4.294
yss-4.288
Token Barack
Token ID8732
Feature activation+2.74e+00

Pos logit contributions
ours+0.092
oured+0.091
ateral+0.079
hap+0.078
aths+0.078
Neg logit contributions
ンジ-0.055
-0.052
 Valhalla-0.050
 pose-0.049
-0.049
Token Obama
Token ID2486
Feature activation+1.44e+00

Pos logit contributions
ンジ+4.930
 Valhalla+4.662
 pose+4.623
+4.553
+4.466
Neg logit contributions
oured-9.281
ours-9.242
yss-8.123
ateral-8.053
hap-8.051
Token Barack
Token ID8732
Feature activation+3.58e+00

Pos logit contributions
ンジ+3.346
 pose+3.185
 Valhalla+3.173
+3.166
+3.076
Neg logit contributions
ours-4.203
oured-4.182
ateral-3.565
hap-3.535
yss-3.492
Token Hussein
Token ID22983
Feature activation+3.44e+00

Pos logit contributions
ンジ+7.431
 pose+7.120
+6.964
 Valhalla+6.944
+6.897
Neg logit contributions
oured-11.144
ours-11.014
ateral-9.629
hap-9.621
yss-9.455
Token Obama
Token ID2486
Feature activation+2.34e+00

Pos logit contributions
 pose+5.415
ンジ+5.224
 Valhalla+4.991
 dream+4.584
+4.579
Neg logit contributions
ours-12.109
oured-12.076
yss-10.940
hap-10.794
aths-10.701
TokenWith
Token ID3152
Feature activation+9.34e-01

Pos logit contributions
ンジ+6.466
 pose+6.245
+6.159
 Valhalla+6.118
+6.092
Neg logit contributions
oured-5.626
ours-5.473
hap-4.659
ateral-4.612
yss-4.609
Token low
Token ID1877
Feature activation+6.19e-01

Pos logit contributions
ンジ+2.187
+2.062
 pose+2.044
 Valhalla+2.033
+2.013
Neg logit contributions
oured-2.784
ours-2.784
ateral-2.358
hap-2.294
aths-2.288
Token birth
Token ID4082
Feature activation+2.40e+00

Pos logit contributions
ンジ+1.422
+1.341
 Valhalla+1.331
+1.316
 pose+1.311
Neg logit contributions
oured-1.867
ours-1.867
ateral-1.591
hap-1.583
aths-1.561
Token MORE
Token ID12011
Feature activation-2.29e-01

Pos logit contributions
ンジ+2.655
+2.542
 Valhalla+2.503
 pose+2.476
+2.447
Neg logit contributions
ours-2.715
oured-2.703
ateral-2.262
hap-2.240
aths-2.233
Token as
Token ID355
Feature activation+7.91e-01

Pos logit contributions
ours+0.625
oured+0.621
ateral+0.520
hap+0.514
aths+0.510
Neg logit contributions
ンジ-0.604
-0.579
 Valhalla-0.556
-0.550
 pose-0.540
Token a
Token ID257
Feature activation+8.63e-01

Pos logit contributions
ンジ+1.740
+1.647
 Valhalla+1.636
 pose+1.619
+1.589
Neg logit contributions
ours-2.442
oured-2.435
ateral-2.090
hap-2.069
aths-2.054
Token 68
Token ID8257
Feature activation-1.71e-02

Pos logit contributions
ンジ+1.986
+1.887
 Valhalla+1.842
 pose+1.807
+1.804
Neg logit contributions
ours-2.612
oured-2.603
ateral-2.220
hap-2.197
aths-2.186
Token-
Token ID12
Feature activation+1.48e+00

Pos logit contributions
ours+0.052
oured+0.052
ateral+0.044
hap+0.044
aths+0.043
Neg logit contributions
ンジ-0.039
-0.037
 Valhalla-0.036
 pose-0.036
-0.036
Tokenyear
Token ID1941
Feature activation+3.45e+00

Pos logit contributions
ンジ+2.581
 pose+2.324
+2.306
 Valhalla+2.285
+2.278
Neg logit contributions
oured-5.183
ours-5.177
ateral-4.563
hap-4.509
yss-4.500
Token-
Token ID12
Feature activation+6.39e-01

Pos logit contributions
ンジ+6.337
 pose+6.054
 dream+5.807
 Valhalla+5.767
+5.701
Neg logit contributions
ours-11.016
oured-10.870
ateral-9.699
hap-9.664
yss-9.577
Tokenold
Token ID727
Feature activation+1.64e+00

Pos logit contributions
ンジ+1.817
+1.741
 Valhalla+1.713
 pose+1.691
+1.685
Neg logit contributions
ours-1.578
oured-1.573
ateral-1.293
hap-1.275
aths-1.265
Token woman
Token ID2415
Feature activation+3.92e-02

Pos logit contributions
ンジ+3.374
 Valhalla+3.181
 pose+3.167
+3.156
+3.033
Neg logit contributions
ours-5.254
oured-5.233
ateral-4.584
hap-4.496
aths-4.463
Token's
Token ID338
Feature activation-4.50e-01

Pos logit contributions
ンジ+0.093
+0.088
 Valhalla+0.086
+0.084
 pose+0.084
Neg logit contributions
ours-0.117
oured-0.116
ateral-0.098
hap-0.098
aths-0.097
Token cause
Token ID2728
Feature activation+3.84e-02

Pos logit contributions
oured+1.457
ours+1.452
ateral+1.243
hap+1.234
aths+1.231
Neg logit contributions
ンジ-0.956
-0.904
-0.848
 Valhalla-0.845
 pose-0.813
Tokensup
Token ID37330
Feature activation+1.26e+00

Pos logit contributions
ours+1.330
oured+1.325
ateral+1.127
hap+1.116
aths+1.101
Neg logit contributions
ンジ-1.131
-1.088
 Valhalla-1.061
 pose-1.030
-1.030
Token<|endoftext|>
Token ID50256
Feature activation+3.21e-01

Pos logit contributions
ンジ+3.017
+2.849
 pose+2.822
 Valhalla+2.808
+2.773
Neg logit contributions
oured-3.627
ours-3.609
ateral-3.075
hap-3.036
yss-3.012
TokenA
Token ID32
Feature activation+2.36e-01

Pos logit contributions
ンジ+0.849
+0.811
 Valhalla+0.795
 pose+0.785
+0.784
Neg logit contributions
ours-0.856
oured-0.854
ateral-0.714
hap-0.703
aths-0.699
Token 67
Token ID8275
Feature activation-2.56e-01

Pos logit contributions
ンジ+0.515
+0.488
 Valhalla+0.474
+0.466
 pose+0.464
Neg logit contributions
ours-0.745
oured-0.741
ateral-0.635
hap-0.632
aths-0.628
Token-
Token ID12
Feature activation+1.26e+00

Pos logit contributions
ours+0.763
oured+0.761
ateral+0.647
hap+0.641
aths+0.637
Neg logit contributions
ンジ-0.599
-0.570
 Valhalla-0.558
 pose-0.549
-0.546
Tokenyear
Token ID1941
Feature activation+3.45e+00

Pos logit contributions
ンジ+2.152
+1.947
 Valhalla+1.897
 pose+1.884
+1.877
Neg logit contributions
ours-4.494
oured-4.491
ateral-3.964
hap-3.892
yss-3.888
Token-
Token ID12
Feature activation+4.03e-01

Pos logit contributions
ンジ+6.449
 pose+6.049
+5.863
 Valhalla+5.846
 dream+5.786
Neg logit contributions
ours-10.942
oured-10.860
ateral-9.737
hap-9.466
yss-9.440
Tokenold
Token ID727
Feature activation+1.43e+00

Pos logit contributions
ンジ+1.171
+1.127
 Valhalla+1.105
 pose+1.091
+1.085
Neg logit contributions
ours-0.980
oured-0.974
ateral-0.794
hap-0.784
aths-0.779
Token homeless
Token ID10463
Feature activation+8.96e-01

Pos logit contributions
ンジ+2.698
+2.524
 Valhalla+2.520
 pose+2.473
+2.401
Neg logit contributions
ours-4.860
oured-4.850
ateral-4.255
hap-4.193
aths-4.172
Token woman
Token ID2415
Feature activation+2.14e-01

Pos logit contributions
ンジ+1.498
+1.394
 Valhalla+1.353
 pose+1.321
+1.310
Neg logit contributions
ours-3.271
oured-3.261
ateral-2.867
hap-2.841
aths-2.828
Token known
Token ID1900
Feature activation+4.40e-01

Pos logit contributions
ンジ+0.491
+0.465
 Valhalla+0.450
+0.443
 pose+0.439
Neg logit contributions
ours-0.656
oured-0.653
ateral-0.555
hap-0.550
aths-0.548
Token former
Token ID1966
Feature activation+1.72e+00

Pos logit contributions
ンジ+0.761
+0.721
 Valhalla+0.704
 pose+0.691
+0.688
Neg logit contributions
ours-1.103
oured-1.099
ateral-0.945
hap-0.935
aths-0.930
Token President
Token ID1992
Feature activation-2.75e-02

Pos logit contributions
ンジ+3.968
 Valhalla+3.741
+3.721
 pose+3.706
+3.589
Neg logit contributions
oured-5.106
ours-5.105
ateral-4.380
hap-4.294
yss-4.288
Token Barack
Token ID8732
Feature activation+2.74e+00

Pos logit contributions
ours+0.092
oured+0.091
ateral+0.079
hap+0.078
aths+0.078
Neg logit contributions
ンジ-0.055
-0.052
 Valhalla-0.050
 pose-0.049
-0.049
Token Obama
Token ID2486
Feature activation+1.44e+00

Pos logit contributions
ンジ+4.930
 Valhalla+4.662
 pose+4.623
+4.553
+4.466
Neg logit contributions
oured-9.281
ours-9.242
yss-8.123
ateral-8.053
hap-8.051
Token Barack
Token ID8732
Feature activation+3.58e+00

Pos logit contributions
ンジ+3.346
 pose+3.185
 Valhalla+3.173
+3.166
+3.076
Neg logit contributions
ours-4.203
oured-4.182
ateral-3.565
hap-3.535
yss-3.492
Token Hussein
Token ID22983
Feature activation+3.44e+00

Pos logit contributions
ンジ+7.431
 pose+7.120
+6.964
 Valhalla+6.944
+6.897
Neg logit contributions
oured-11.144
ours-11.014
ateral-9.629
hap-9.621
yss-9.455
Token Obama
Token ID2486
Feature activation+2.34e+00

Pos logit contributions
 pose+5.415
ンジ+5.224
 Valhalla+4.991
 dream+4.584
+4.579
Neg logit contributions
ours-12.109
oured-12.076
yss-10.940
hap-10.794
aths-10.701
TokenWith
Token ID3152
Feature activation+9.34e-01

Pos logit contributions
ンジ+6.466
 pose+6.245
+6.159
 Valhalla+6.118
+6.092
Neg logit contributions
oured-5.626
ours-5.473
hap-4.659
ateral-4.612
yss-4.609
Token low
Token ID1877
Feature activation+6.19e-01

Pos logit contributions
ンジ+2.187
+2.062
 pose+2.044
 Valhalla+2.033
+2.013
Neg logit contributions
oured-2.784
ours-2.784
ateral-2.358
hap-2.294
aths-2.288
Token birth
Token ID4082
Feature activation+2.40e+00

Pos logit contributions
ンジ+1.422
+1.341
 Valhalla+1.331
+1.316
 pose+1.311
Neg logit contributions
oured-1.867
ours-1.867
ateral-1.591
hap-1.583
aths-1.561
Token rate
Token ID2494
Feature activation+1.01e+00

Pos logit contributions
+3.640
 pose+3.529
ンジ+3.472
+3.449
 Valhalla+3.281
Neg logit contributions
oured-8.475
ours-8.247
yss-7.630
hap-7.466
apist-7.417
Token season
Token ID1622
Feature activation-4.23e-01

Pos logit contributions
ours+0.393
oured+0.393
ateral+0.344
hap+0.342
aths+0.339
Neg logit contributions
ンジ-0.169
-0.157
 Valhalla-0.148
-0.145
 pose-0.144
Token.
Token ID13
Feature activation+2.97e-01

Pos logit contributions
ours+1.124
oured+1.118
ateral+0.932
hap+0.920
aths+0.914
Neg logit contributions
ンジ-1.130
-1.081
 Valhalla-1.060
 pose-1.043
-1.041
Token The
Token ID383
Feature activation-3.42e-01

Pos logit contributions
ンジ+0.674
+0.639
 Valhalla+0.627
 pose+0.614
+0.611
Neg logit contributions
ours-0.908
oured-0.905
ateral-0.774
hap-0.765
aths-0.761
Token 37
Token ID5214
Feature activation-8.69e-02

Pos logit contributions
ours+1.145
oured+1.141
ateral+0.985
hap+0.979
aths+0.975
Neg logit contributions
ンジ-0.680
-0.643
 Valhalla-0.617
-0.613
 pose-0.599
Token-
Token ID12
Feature activation+1.08e+00

Pos logit contributions
ours+0.259
oured+0.259
ateral+0.220
hap+0.218
aths+0.216
Neg logit contributions
ンジ-0.204
-0.194
 Valhalla-0.189
 pose-0.186
-0.185
Tokenyear
Token ID1941
Feature activation+3.38e+00

Pos logit contributions
ンジ+1.945
+1.810
 Valhalla+1.766
 pose+1.730
+1.723
Neg logit contributions
ours-3.782
oured-3.777
ateral-3.301
hap-3.266
aths-3.249
Token-
Token ID12
Feature activation+5.02e-01

Pos logit contributions
ンジ+6.439
 Valhalla+6.211
 pose+6.125
 dream+5.908
+5.890
Neg logit contributions
ours-10.747
oured-10.676
ateral-9.285
yss-9.230
hap-9.207
Tokenold
Token ID727
Feature activation+1.49e+00

Pos logit contributions
ンジ+1.464
+1.409
 Valhalla+1.380
 pose+1.364
+1.356
Neg logit contributions
ours-1.220
oured-1.211
ateral-0.988
hap-0.975
aths-0.969
Token Hur
Token ID10077
Feature activation+1.75e+00

Pos logit contributions
ンジ+3.188
+2.993
 Valhalla+2.983
 pose+2.947
+2.861
Neg logit contributions
ours-4.672
oured-4.646
ateral-4.043
hap-3.972
yss-3.950
Tokentt
Token ID926
Feature activation+6.99e-01

Pos logit contributions
ンジ+4.096
+3.925
 Valhalla+3.854
 pose+3.822
+3.772
Neg logit contributions
ours-5.018
oured-5.003
ateral-4.256
hap-4.211
aths-4.155
Token just
Token ID655
Feature activation+1.14e+00

Pos logit contributions
ンジ+1.587
+1.507
 Valhalla+1.475
 pose+1.455
+1.440
Neg logit contributions
ours-2.124
oured-2.117
ateral-1.814
hap-1.795
aths-1.780
Token Michael
Token ID3899
Feature activation+1.26e+00

Pos logit contributions
ンジ+2.113
+1.995
 Valhalla+1.965
 pose+1.930
+1.911
Neg logit contributions
ours-3.111
oured-3.106
ateral-2.676
hap-2.643
aths-2.626
Token Flynn
Token ID14654
Feature activation+4.32e-01

Pos logit contributions
ンジ+2.555
+2.427
 Valhalla+2.366
 pose+2.335
+2.311
Neg logit contributions
ours-4.088
oured-4.078
ateral-3.516
hap-3.499
aths-3.474
Token The
Token ID383
Feature activation+2.65e-01

Pos logit contributions
ンジ+0.967
+0.916
 Valhalla+0.898
 pose+0.883
+0.877
Neg logit contributions
ours-1.333
oured-1.328
ateral-1.140
hap-1.127
aths-1.121
Token 57
Token ID7632
Feature activation-1.42e-01

Pos logit contributions
ンジ+0.601
+0.571
 Valhalla+0.557
 pose+0.546
+0.546
Neg logit contributions
ours-0.810
oured-0.808
ateral-0.689
hap-0.683
aths-0.679
Token-
Token ID12
Feature activation+1.24e+00

Pos logit contributions
ours+0.422
oured+0.421
ateral+0.359
hap+0.355
aths+0.353
Neg logit contributions
ンジ-0.332
-0.315
 Valhalla-0.309
 pose-0.304
-0.302
Tokenyear
Token ID1941
Feature activation+3.35e+00

Pos logit contributions
ンジ+2.128
+1.936
 Valhalla+1.890
 pose+1.880
+1.871
Neg logit contributions
oured-4.416
ours-4.395
ateral-3.883
hap-3.841
aths-3.820
Token-
Token ID12
Feature activation+3.78e-01

Pos logit contributions
ンジ+6.175
 pose+6.070
 Valhalla+6.001
 dream+5.814
+5.577
Neg logit contributions
ours-10.707
oured-10.651
ateral-9.379
yss-9.295
hap-9.292
Tokenold
Token ID727
Feature activation+1.68e+00

Pos logit contributions
ンジ+1.106
+1.066
 Valhalla+1.044
 pose+1.030
+1.025
Neg logit contributions
ours-0.917
oured-0.910
ateral-0.741
hap-0.731
aths-0.727
Token retired
Token ID9880
Feature activation+1.98e+00

Pos logit contributions
ンジ+3.124
 pose+2.917
 Valhalla+2.915
+2.866
+2.774
Neg logit contributions
ours-5.659
oured-5.634
ateral-5.028
hap-4.941
aths-4.887
Token lieutenant
Token ID30269
Feature activation+6.92e-01

Pos logit contributions
ンジ+4.435
 Valhalla+4.166
 pose+4.129
+4.098
+4.005
Neg logit contributions
oured-5.918
ours-5.917
ateral-5.173
hap-5.075
yss-4.985
Token general
Token ID2276
Feature activation+5.29e-01

Pos logit contributions
ンジ+1.540
+1.445
 Valhalla+1.434
 pose+1.421
+1.400
Neg logit contributions
ours-2.126
oured-2.118
ateral-1.823
hap-1.806
aths-1.789
Token said
Token ID531
Feature activation+6.02e-01

Pos logit contributions
ours+0.950
oured+0.945
ateral+0.799
hap+0.790
aths+0.789
Neg logit contributions
ンジ-0.766
-0.725
 Valhalla-0.719
-0.697
 pose-0.695
Token that
Token ID326
Feature activation+3.64e-01

Pos logit contributions
ンジ+1.281
+1.213
 Valhalla+1.187
 pose+1.167
+1.155
Neg logit contributions
ours-1.920
oured-1.914
ateral-1.648
hap-1.630
aths-1.622
Token the
Token ID262
Feature activation+3.73e-01

Pos logit contributions
ンジ+0.757
+0.716
 Valhalla+0.696
+0.682
 pose+0.680
Neg logit contributions
ours-1.184
oured-1.179
ateral-1.019
hap-1.010
aths-1.005
Token 26
Token ID2608
Feature activation-6.56e-01

Pos logit contributions
ンジ+0.652
+0.609
 Valhalla+0.591
 pose+0.576
+0.575
Neg logit contributions
ours-1.335
oured-1.331
ateral-1.166
hap-1.156
aths-1.151
Token-
Token ID12
Feature activation+1.17e+00

Pos logit contributions
ours+1.895
oured+1.884
ateral+1.584
aths+1.572
hap+1.572
Neg logit contributions
ンジ-1.594
-1.522
 Valhalla-1.494
-1.464
 pose-1.458
Tokenyear
Token ID1941
Feature activation+3.35e+00

Pos logit contributions
ンジ+2.053
+1.885
 Valhalla+1.845
 pose+1.822
+1.807
Neg logit contributions
oured-4.132
ours-4.131
ateral-3.618
hap-3.585
aths-3.562
Token-
Token ID12
Feature activation+4.43e-01

Pos logit contributions
ンジ+6.314
+5.884
 pose+5.739
 Valhalla+5.577
 dream+5.565
Neg logit contributions
ours-10.677
oured-10.645
yss-9.359
ateral-9.349
hap-9.340
Tokenold
Token ID727
Feature activation+1.51e+00

Pos logit contributions
ンジ+1.301
+1.254
 Valhalla+1.228
 pose+1.213
+1.204
Neg logit contributions
ours-1.070
oured-1.061
ateral-0.864
hap-0.851
aths-0.848
Token Reed
Token ID16105
Feature activation+2.57e-01

Pos logit contributions
ンジ+2.980
+2.800
 Valhalla+2.754
 pose+2.752
+2.635
Neg logit contributions
ours-4.996
oured-4.995
ateral-4.358
hap-4.306
yss-4.272
Token,
Token ID11
Feature activation+4.21e-01

Pos logit contributions
ンジ+0.572
+0.541
 Valhalla+0.530
+0.518
 pose+0.517
Neg logit contributions
ours-0.797
oured-0.794
ateral-0.680
hap-0.672
aths-0.670
Token who
Token ID508
Feature activation-8.80e-02

Pos logit contributions
ンジ+0.967
+0.918
 Valhalla+0.896
+0.879
 pose+0.877
Neg logit contributions
ours-1.283
oured-1.277
ateral-1.087
hap-1.075
aths-1.071
Token Wilson
Token ID8127
Feature activation+1.11e+00

Pos logit contributions
ours+0.514
oured+0.513
hap+0.439
ateral+0.438
aths+0.437
Neg logit contributions
ンジ-0.337
-0.317
 Valhalla-0.304
 pose-0.303
-0.301
Token.
Token ID13
Feature activation+3.78e-01

Pos logit contributions
ンジ+2.805
 Valhalla+2.691
+2.667
 pose+2.642
+2.596
Neg logit contributions
ours-3.067
oured-3.052
ateral-2.539
hap-2.522
aths-2.517
Token The
Token ID383
Feature activation-3.04e-01

Pos logit contributions
ンジ+0.861
+0.817
 Valhalla+0.799
 pose+0.786
+0.782
Neg logit contributions
ours-1.150
oured-1.146
ateral-0.980
hap-0.970
aths-0.964
Token 26
Token ID2608
Feature activation-6.05e-01

Pos logit contributions
ours+1.001
oured+0.996
ateral+0.856
hap+0.848
aths+0.847
Neg logit contributions
ンジ-0.625
-0.585
-0.565
 Valhalla-0.560
 pose-0.549
Token-
Token ID12
Feature activation+1.07e+00

Pos logit contributions
ours+1.757
oured+1.749
ateral+1.479
hap+1.471
aths+1.460
Neg logit contributions
ンジ-1.460
-1.391
 Valhalla-1.359
 pose-1.335
-1.333
Tokenyear
Token ID1941
Feature activation+3.33e+00

Pos logit contributions
ンジ+1.897
+1.761
 Valhalla+1.724
 pose+1.688
+1.681
Neg logit contributions
ours-3.770
oured-3.768
ateral-3.291
hap-3.260
aths-3.246
Token-
Token ID12
Feature activation+4.75e-01

Pos logit contributions
ンジ+6.331
 Valhalla+6.166
 pose+6.042
 dream+5.829
+5.827
Neg logit contributions
ours-10.547
oured-10.516
ateral-9.118
hap-9.109
yss-9.084
Tokenold
Token ID727
Feature activation+1.46e+00

Pos logit contributions
ンジ+1.385
+1.333
 Valhalla+1.306
 pose+1.290
+1.282
Neg logit contributions
ours-1.154
oured-1.146
ateral-0.935
hap-0.922
aths-0.916
Token offensive
Token ID5859
Feature activation-2.56e-02

Pos logit contributions
ンジ+2.959
+2.785
 Valhalla+2.776
 pose+2.742
+2.648
Neg logit contributions
ours-4.694
oured-4.692
ateral-4.087
hap-4.031
aths-3.992
Token tackle
Token ID9761
Feature activation+1.24e+00

Pos logit contributions
ours+0.103
oured+0.103
ateral+0.092
aths+0.091
hap+0.091
Neg logit contributions
ンジ-0.033
-0.030
 Valhalla-0.028
-0.028
 pose-0.027
Token also
Token ID635
Feature activation+6.78e-01

Pos logit contributions
ンジ+2.720
+2.584
 Valhalla+2.543
 pose+2.527
+2.462
Neg logit contributions
oured-3.820
ours-3.818
ateral-3.295
hap-3.262
aths-3.237
Token of
Token ID286
Feature activation-3.91e-01

Pos logit contributions
ours+2.480
oured+2.466
ateral+2.044
hap+2.015
aths+1.994
Neg logit contributions
ンジ-2.618
-2.512
 Valhalla-2.460
-2.413
 pose-2.407
Token beating
Token ID11226
Feature activation-3.58e-01

Pos logit contributions
ours+1.253
oured+1.245
ateral+1.077
hap+1.066
aths+1.049
Neg logit contributions
ンジ-0.837
-0.794
 Valhalla-0.751
-0.747
 pose-0.725
Token an
Token ID281
Feature activation+8.21e-01

Pos logit contributions
ours+1.046
oured+1.045
ateral+0.886
hap+0.876
aths+0.874
Neg logit contributions
ンジ-0.863
-0.827
-0.785
 Valhalla-0.777
 pose-0.769
Token 88
Token ID9193
Feature activation-4.58e-02

Pos logit contributions
ンジ+1.882
+1.774
 Valhalla+1.770
 pose+1.746
+1.715
Neg logit contributions
ours-2.469
oured-2.457
ateral-2.103
hap-2.081
aths-2.066
Token-
Token ID12
Feature activation+1.20e+00

Pos logit contributions
ours+0.131
oured+0.131
ateral+0.110
hap+0.109
aths+0.109
Neg logit contributions
ンジ-0.112
-0.108
 Valhalla-0.104
-0.103
 pose-0.103
Tokenyear
Token ID1941
Feature activation+3.29e+00

Pos logit contributions
ンジ+2.149
+1.989
 Valhalla+1.947
 pose+1.912
+1.897
Neg logit contributions
ours-4.231
oured-4.222
ateral-3.703
hap-3.662
aths-3.643
Token-
Token ID12
Feature activation+3.56e-01

Pos logit contributions
ンジ+6.335
 Valhalla+6.104
 pose+6.098
 dream+5.824
+5.816
Neg logit contributions
ours-10.388
oured-10.336
ateral-9.008
hap-8.914
yss-8.882
Tokenold
Token ID727
Feature activation+1.62e+00

Pos logit contributions
ンジ+1.044
+1.006
 Valhalla+0.985
 pose+0.972
+0.967
Neg logit contributions
ours-0.861
oured-0.855
ateral-0.695
hap-0.686
aths-0.682
Token World
Token ID2159
Feature activation+1.21e+00

Pos logit contributions
ンジ+3.153
 Valhalla+2.989
+2.971
 pose+2.941
+2.823
Neg logit contributions
ours-5.393
oured-5.381
ateral-4.703
hap-4.641
aths-4.612
Token War
Token ID1810
Feature activation+6.50e-03

Pos logit contributions
ンジ+1.410
+1.270
 Valhalla+1.220
 pose+1.173
+1.160
Neg logit contributions
ours-5.014
oured-5.000
ateral-4.468
hap-4.438
aths-4.419
Token II
Token ID2873
Feature activation-4.17e-01

Pos logit contributions
ンジ+0.016
+0.015
 Valhalla+0.014
+0.014
 pose+0.014
Neg logit contributions
oured-0.019
ours-0.019
ateral-0.016
hap-0.016
aths-0.016
Token Clinton
Token ID2605
Feature activation+3.06e-02

Pos logit contributions
ンジ+1.859
+1.750
 Valhalla+1.703
 pose+1.668
+1.659
Neg logit contributions
ours-3.213
oured-3.202
ateral-2.783
hap-2.757
aths-2.743
Token Hillary
Token ID5073
Feature activation+1.84e+00

Pos logit contributions
ンジ+0.067
+0.064
 Valhalla+0.062
+0.061
 pose+0.060
Neg logit contributions
ours-0.097
oured-0.096
ateral-0.082
hap-0.081
aths-0.081
Token Diane
Token ID26542
Feature activation+2.39e+00

Pos logit contributions
ンジ+3.828
+3.618
 Valhalla+3.508
 pose+3.399
+3.398
Neg logit contributions
ours-6.069
oured-6.047
ateral-5.217
hap-5.157
aths-5.126
Token Rodham
Token ID41848
Feature activation+1.59e+00

Pos logit contributions
ンジ+4.075
+3.804
 pose+3.756
 Valhalla+3.699
+3.646
Neg logit contributions
ours-8.491
oured-8.470
ateral-7.421
hap-7.420
yss-7.339
Token Clinton
Token ID2605
Feature activation+1.38e+00

Pos logit contributions
ンジ+2.858
 pose+2.629
 Valhalla+2.619
+2.613
+2.592
Neg logit contributions
oured-5.489
ours-5.468
ateral-4.794
aths-4.778
hap-4.762
TokenSanders
Token ID26747
Feature activation+3.28e+00

Pos logit contributions
ンジ+3.755
+3.543
 pose+3.520
 Valhalla+3.520
+3.517
Neg logit contributions
oured-3.473
ours-3.424
ateral-2.898
aths-2.841
hap-2.833
Token:
Token ID25
Feature activation+1.94e+00

Pos logit contributions
ンジ+5.486
 pose+5.484
 Valhalla+5.188
+5.121
 dream+5.020
Neg logit contributions
oured-10.945
ours-10.773
aths-9.737
ateral-9.579
uthor-9.406
Token '
Token ID705
Feature activation+3.80e-01

Pos logit contributions
ンジ+3.361
 Valhalla+3.178
+3.139
 pose+3.121
+3.104
Neg logit contributions
ours-6.793
oured-6.781
ateral-5.942
hap-5.938
aths-5.856
TokenI
Token ID40
Feature activation+9.65e-01

Pos logit contributions
ンジ+0.979
+0.936
 Valhalla+0.917
 pose+0.901
+0.899
Neg logit contributions
ours-1.050
oured-1.044
ateral-0.876
hap-0.866
aths-0.860
Token fully
Token ID3938
Feature activation+4.36e-02

Pos logit contributions
ンジ+1.574
 pose+1.434
+1.411
 Valhalla+1.405
 dream+1.361
Neg logit contributions
ours-3.513
oured-3.478
hap-3.095
ateral-3.095
yss-3.045
Token expect
Token ID1607
Feature activation+1.88e+00

Pos logit contributions
ンジ+0.094
+0.089
 Valhalla+0.087
+0.084
 pose+0.083
Neg logit contributions
oured-0.139
ours-0.139
ateral-0.120
hap-0.118
aths-0.117
Token a
Token ID257
Feature activation-6.84e-02

Pos logit contributions
ours+2.414
oured+2.408
ateral+2.040
hap+2.000
aths+1.985
Neg logit contributions
ンジ-2.105
-2.008
 Valhalla-1.929
-1.903
 pose-1.899
Token train
Token ID4512
Feature activation+1.02e+00

Pos logit contributions
ours+0.217
oured+0.216
ateral+0.186
hap+0.184
aths+0.183
Neg logit contributions
ンジ-0.148
-0.140
 Valhalla-0.135
-0.133
 pose-0.132
Token-
Token ID12
Feature activation-2.25e-01

Pos logit contributions
ンジ+2.588
+2.447
 Valhalla+2.421
 pose+2.415
+2.368
Neg logit contributions
ours-2.819
oured-2.798
ateral-2.352
hap-2.343
aths-2.336
Tokenand
Token ID392
Feature activation+8.69e-01

Pos logit contributions
ours+0.487
oured+0.484
ateral+0.385
hap+0.379
aths+0.376
Neg logit contributions
ンジ-0.712
-0.684
 Valhalla-0.675
 pose-0.667
-0.665
Token-
Token ID12
Feature activation+3.29e-01

Pos logit contributions
ンジ+2.088
+1.989
 Valhalla+1.938
 pose+1.903
+1.901
Neg logit contributions
ours-2.547
oured-2.537
ateral-2.156
hap-2.129
aths-2.117
Tokenequ
Token ID4853
Feature activation+3.24e+00

Pos logit contributions
ンジ+0.922
+0.883
 Valhalla+0.872
 pose+0.862
+0.857
Neg logit contributions
ours-0.820
oured-0.819
ateral-0.672
hap-0.667
aths-0.660
Tokenip
Token ID541
Feature activation+4.89e-01

Pos logit contributions
ンジ+6.864
 pose+6.635
 Valhalla+6.434
+6.386
+6.363
Neg logit contributions
oured-9.231
ours-9.145
yss-7.864
Ps-7.848
hap-7.732
Token program
Token ID1430
Feature activation+5.12e-01

Pos logit contributions
ンジ+1.073
+1.016
 Valhalla+0.997
 pose+0.982
+0.973
Neg logit contributions
ours-1.526
oured-1.520
ateral-1.304
hap-1.295
aths-1.285
Token to
Token ID284
Feature activation+3.67e-02

Pos logit contributions
ンジ+1.232
+1.174
 Valhalla+1.156
 pose+1.141
+1.128
Neg logit contributions
ours-1.479
oured-1.474
ateral-1.253
hap-1.242
aths-1.235
Token help
Token ID1037
Feature activation-6.97e-01

Pos logit contributions
ンジ+0.073
+0.069
 Valhalla+0.066
+0.065
 pose+0.064
Neg logit contributions
ours-0.123
oured-0.123
ateral-0.106
hap-0.105
aths-0.105
Token K
Token ID509
Feature activation-2.14e-01

Pos logit contributions
oured+2.219
ours+2.218
ateral+1.907
hap+1.877
aths+1.874
Neg logit contributions
ンジ-1.496
-1.421
 Valhalla-1.368
-1.342
 pose-1.312
Token,
Token ID11
Feature activation+2.11e-01

Pos logit contributions
ours+1.991
oured+1.969
ateral+1.668
hap+1.648
aths+1.637
Neg logit contributions
ンジ-1.796
-1.711
 Valhalla-1.679
-1.652
 pose-1.615
Token but
Token ID475
Feature activation+8.13e-01

Pos logit contributions
ンジ+0.536
+0.512
 Valhalla+0.500
 pose+0.491
+0.490
Neg logit contributions
ours-0.590
oured-0.586
ateral-0.494
hap-0.488
aths-0.486
Token the
Token ID262
Feature activation+4.80e-01

Pos logit contributions
ンジ+1.607
+1.513
 Valhalla+1.474
 pose+1.445
+1.437
Neg logit contributions
ours-2.722
oured-2.713
ateral-2.355
hap-2.332
aths-2.320
Token 31
Token ID3261
Feature activation-1.20e-01

Pos logit contributions
ンジ+1.023
+0.968
 Valhalla+0.941
 pose+0.924
+0.923
Neg logit contributions
ours-1.538
oured-1.533
ateral-1.319
hap-1.306
aths-1.301
Token-
Token ID12
Feature activation+8.99e-01

Pos logit contributions
ours+0.361
oured+0.360
ateral+0.307
hap+0.304
aths+0.302
Neg logit contributions
ンジ-0.280
-0.266
 Valhalla-0.260
 pose-0.256
-0.254
Tokenyear
Token ID1941
Feature activation+3.24e+00

Pos logit contributions
ンジ+1.625
+1.516
 Valhalla+1.478
 pose+1.448
+1.440
Neg logit contributions
ours-3.146
oured-3.142
ateral-2.746
hap-2.720
aths-2.706
Token-
Token ID12
Feature activation+5.55e-01

Pos logit contributions
ンジ+6.027
+5.585
 Valhalla+5.532
 pose+5.513
 dream+5.323
Neg logit contributions
ours-10.439
oured-10.375
ateral-9.116
hap-9.067
yss-9.017
Tokenold
Token ID727
Feature activation+1.52e+00

Pos logit contributions
ンジ+1.621
+1.559
 Valhalla+1.529
 pose+1.511
+1.501
Neg logit contributions
ours-1.346
oured-1.335
ateral-1.088
hap-1.072
aths-1.068
Token left
Token ID1364
Feature activation+6.09e-01

Pos logit contributions
ンジ+3.114
+2.906
 Valhalla+2.895
 pose+2.873
+2.772
Neg logit contributions
oured-4.898
ours-4.892
ateral-4.274
hap-4.224
yss-4.185
Token guard
Token ID4860
Feature activation+1.25e+00

Pos logit contributions
ンジ+1.431
+1.361
 Valhalla+1.332
 pose+1.311
+1.304
Neg logit contributions
ours-1.812
oured-1.805
ateral-1.536
hap-1.520
aths-1.511
Token is
Token ID318
Feature activation+8.36e-03

Pos logit contributions
ンジ+2.562
+2.411
 Valhalla+2.370
 pose+2.364
+2.304
Neg logit contributions
oured-4.016
ours-4.006
ateral-3.477
hap-3.451
aths-3.415
Token
Token ID251
Feature activation-2.72e-01

Pos logit contributions
ours+0.019
oured+0.019
ateral+0.016
hap+0.016
aths+0.016
Neg logit contributions
ンジ-0.014
-0.013
 Valhalla-0.013
 pose-0.013
-0.013
Token said
Token ID531
Feature activation+1.45e-01

Pos logit contributions
ours+0.815
oured+0.811
ateral+0.689
hap+0.683
aths+0.679
Neg logit contributions
ンジ-0.632
-0.605
 Valhalla-0.592
 pose-0.577
-0.576
Token the
Token ID262
Feature activation-2.66e-01

Pos logit contributions
ンジ+0.309
+0.292
 Valhalla+0.285
 pose+0.279
+0.278
Neg logit contributions
ours-0.465
oured-0.463
ateral-0.399
hap-0.395
aths-0.393
Token 82
Token ID9415
Feature activation-1.38e-01

Pos logit contributions
ours+0.819
oured+0.816
ateral+0.697
hap+0.693
aths+0.685
Neg logit contributions
ンジ-0.596
-0.570
 Valhalla-0.553
-0.541
 pose-0.538
Token-
Token ID12
Feature activation+6.89e-01

Pos logit contributions
ours+0.410
oured+0.409
ateral+0.348
hap+0.344
aths+0.342
Neg logit contributions
ンジ-0.327
-0.311
 Valhalla-0.304
 pose-0.299
-0.298
Tokenyear
Token ID1941
Feature activation+3.24e+00

Pos logit contributions
ンジ+1.293
+1.220
 Valhalla+1.183
 pose+1.156
+1.149
Neg logit contributions
ours-2.384
oured-2.373
ateral-2.067
hap-2.052
aths-2.040
Token-
Token ID12
Feature activation+1.14e-01

Pos logit contributions
ンジ+6.177
 pose+5.910
 Valhalla+5.896
+5.663
 dream+5.604
Neg logit contributions
ours-10.263
oured-10.214
ateral-8.917
hap-8.839
yss-8.837
Tokenold
Token ID727
Feature activation+1.38e+00

Pos logit contributions
ンジ+0.341
+0.329
 Valhalla+0.322
 pose+0.318
+0.316
Neg logit contributions
ours-0.272
oured-0.269
ateral-0.218
hap-0.215
aths-0.214
Token grandmother
Token ID18410
Feature activation+4.09e-01

Pos logit contributions
ンジ+2.918
+2.739
 Valhalla+2.736
 pose+2.713
+2.638
Neg logit contributions
ours-4.343
oured-4.332
ateral-3.780
aths-3.704
hap-3.682
Token.
Token ID13
Feature activation+8.83e-02

Pos logit contributions
ンジ+1.017
+0.970
 Valhalla+0.952
 pose+0.941
+0.933
Neg logit contributions
ours-1.158
oured-1.153
ateral-0.977
hap-0.965
aths-0.958
Token
Token ID564
Feature activation-6.15e-01

Pos logit contributions
ンジ+0.197
+0.187
 Valhalla+0.182
 pose+0.179
+0.178
Neg logit contributions
ours-0.273
oured-0.272
ateral-0.233
hap-0.230
aths-0.229
Tokendo
Token ID4598
Feature activation+1.24e+00

Pos logit contributions
ンジ+0.621
+0.593
 Valhalla+0.581
 pose+0.572
+0.570
Neg logit contributions
ours-0.690
oured-0.688
ateral-0.579
hap-0.572
aths-0.569
Token itself
Token ID2346
Feature activation+3.39e-01

Pos logit contributions
ンジ+3.079
 Valhalla+2.913
+2.895
 pose+2.864
+2.827
Neg logit contributions
ours-3.472
oured-3.458
ateral-2.927
hap-2.884
yss-2.876
Token when
Token ID618
Feature activation-1.47e-01

Pos logit contributions
ンジ+0.749
+0.712
 Valhalla+0.692
 pose+0.678
+0.677
Neg logit contributions
ours-1.058
oured-1.053
ateral-0.905
hap-0.896
aths-0.890
Token correspondent
Token ID21076
Feature activation-5.08e-01

Pos logit contributions
ours+0.448
oured+0.447
ateral+0.381
hap+0.379
aths+0.375
Neg logit contributions
ンジ-0.335
-0.320
 Valhalla-0.309
-0.304
 pose-0.301
Token Jesse
Token ID18033
Feature activation+2.17e+00

Pos logit contributions
ours+1.654
oured+1.643
hap+1.408
ateral+1.405
aths+1.393
Neg logit contributions
ンジ-1.063
-1.010
 Valhalla-0.957
-0.950
 pose-0.934
Token Wat
Token ID12242
Feature activation+3.20e+00

Pos logit contributions
ンジ+1.573
+1.446
 pose+1.322
 Valhalla+1.318
+1.231
Neg logit contributions
ours-9.810
oured-9.767
ateral-8.761
yss-8.718
hap-8.710
Tokenters
Token ID1010
Feature activation+6.53e-01

Pos logit contributions
ンジ+5.322
 Valhalla+5.008
terson+4.993
+4.959
 pose+4.875
Neg logit contributions
oured-10.678
ours-10.671
ateral-9.419
aths-9.215
uthor-9.211
Token ventured
Token ID44716
Feature activation+1.78e-01

Pos logit contributions
ンジ+1.517
+1.438
 Valhalla+1.408
 pose+1.394
+1.380
Neg logit contributions
ours-1.956
oured-1.947
ateral-1.663
hap-1.645
aths-1.634
Token down
Token ID866
Feature activation-5.07e-01

Pos logit contributions
ンジ+0.400
+0.381
 Valhalla+0.370
+0.363
 pose+0.362
Neg logit contributions
ours-0.548
oured-0.545
ateral-0.467
hap-0.463
aths-0.460
Token to
Token ID284
Feature activation+1.10e+00

Pos logit contributions
ours+1.510
oured+1.501
ateral+1.288
hap+1.276
aths+1.266
Neg logit contributions
ンジ-1.188
-1.139
 Valhalla-1.099
-1.085
 pose-1.075
Token New
Token ID968
Feature activation+1.32e+00

Pos logit contributions
ンジ+2.281
+2.145
 Valhalla+2.116
 pose+2.089
+2.052
Neg logit contributions
ours-3.569
oured-3.556
ateral-3.070
hap-3.041
aths-3.015
Token
Token ID198
Feature activation-1.58e-01

Pos logit contributions
ンジ+0.088
+0.084
 Valhalla+0.081
 pose+0.080
+0.079
Neg logit contributions
ours-0.103
oured-0.101
ateral-0.086
hap-0.085
aths-0.085
TokenThe
Token ID464
Feature activation-1.83e-01

Pos logit contributions
ours+0.433
oured+0.430
ateral+0.361
hap+0.358
aths+0.355
Neg logit contributions
ンジ-0.410
-0.392
 Valhalla-0.386
 pose-0.381
-0.376
Token legendary
Token ID13273
Feature activation+5.69e-01

Pos logit contributions
ours+0.591
oured+0.590
ateral+0.508
hap+0.507
aths+0.500
Neg logit contributions
ンジ-0.384
-0.360
 Valhalla-0.347
-0.344
 pose-0.341
Token 78
Token ID8699
Feature activation+1.08e-01

Pos logit contributions
ンジ+1.139
+1.072
 Valhalla+1.042
 pose+1.021
+1.018
Neg logit contributions
ours-1.891
oured-1.885
ateral-1.634
hap-1.620
aths-1.610
Token-
Token ID12
Feature activation+1.01e+00

Pos logit contributions
ンジ+0.259
+0.246
 Valhalla+0.241
 pose+0.237
+0.236
Neg logit contributions
ours-0.318
oured-0.317
ateral-0.269
hap-0.266
aths-0.265
Tokenyear
Token ID1941
Feature activation+3.13e+00

Pos logit contributions
ンジ+2.708
+2.588
 Valhalla+2.527
 pose+2.517
+2.502
Neg logit contributions
oured-2.631
ours-2.628
ateral-2.195
hap-2.148
yss-2.143
Token-
Token ID12
Feature activation+4.72e-01

Pos logit contributions
ンジ+5.896
+5.504
 pose+5.485
 Valhalla+5.391
 dream+5.288
Neg logit contributions
ours-10.141
oured-10.044
ateral-8.847
hap-8.703
yss-8.694
Tokenold
Token ID727
Feature activation+1.43e+00

Pos logit contributions
ンジ+1.368
+1.315
 Valhalla+1.291
 pose+1.274
+1.268
Neg logit contributions
ours-1.152
oured-1.145
ateral-0.935
hap-0.923
aths-0.917
Token developer
Token ID8517
Feature activation+1.86e-01

Pos logit contributions
ンジ+2.936
+2.777
 Valhalla+2.764
 pose+2.718
+2.656
Neg logit contributions
ours-4.621
oured-4.606
ateral-3.988
hap-3.925
aths-3.924
Token picked
Token ID6497
Feature activation+5.54e-01

Pos logit contributions
ンジ+0.437
+0.415
 Valhalla+0.406
 pose+0.398
+0.398
Neg logit contributions
ours-0.556
oured-0.554
ateral-0.471
hap-0.466
aths-0.463
Token up
Token ID510
Feature activation-6.63e-01

Pos logit contributions
ンジ+1.119
+1.054
 Valhalla+1.024
 pose+1.006
+1.002
Neg logit contributions
ours-1.832
oured-1.826
ateral-1.581
hap-1.568
aths-1.558
Token
Token ID198
Feature activation+4.35e-01

Pos logit contributions
ンジ+0.366
+0.348
 Valhalla+0.341
 pose+0.334
+0.333
Neg logit contributions
ours-0.503
oured-0.501
ateral-0.429
hap-0.426
aths-0.422
Token
Token ID198
Feature activation+1.89e-01

Pos logit contributions
ンジ+1.059
+1.014
 Valhalla+0.975
 pose+0.962
+0.954
Neg logit contributions
ours-1.273
oured-1.260
ateral-1.068
hap-1.059
aths-1.052
TokenThe
Token ID464
Feature activation-1.05e-01

Pos logit contributions
ンジ+0.478
+0.456
 Valhalla+0.447
 pose+0.441
+0.439
Neg logit contributions
ours-0.528
oured-0.526
ateral-0.443
hap-0.437
aths-0.434
Token 89
Token ID9919
Feature activation-4.98e-01

Pos logit contributions
ours+0.328
oured+0.328
ateral+0.280
hap+0.276
aths+0.276
Neg logit contributions
ンジ-0.233
-0.220
 Valhalla-0.212
-0.210
 pose-0.208
Token-
Token ID12
Feature activation+1.00e+00

Pos logit contributions
ours+1.533
oured+1.528
ateral+1.307
hap+1.294
aths+1.287
Neg logit contributions
ンジ-1.118
-1.060
 Valhalla-1.036
 pose-1.018
-1.013
Tokenyear
Token ID1941
Feature activation+3.13e+00

Pos logit contributions
ンジ+1.668
+1.548
 Valhalla+1.490
 pose+1.479
+1.470
Neg logit contributions
ours-3.603
oured-3.596
ateral-3.189
yss-3.155
hap-3.148
Token-
Token ID12
Feature activation+1.90e-01

Pos logit contributions
ンジ+5.958
 Valhalla+5.839
 pose+5.799
 dream+5.537
+5.501
Neg logit contributions
ours-9.840
oured-9.770
ateral-8.628
hap-8.560
yss-8.530
Tokenold
Token ID727
Feature activation+1.48e+00

Pos logit contributions
ンジ+0.562
+0.542
 Valhalla+0.530
 pose+0.524
+0.521
Neg logit contributions
ours-0.454
oured-0.450
ateral-0.364
hap-0.359
aths-0.357
Token said
Token ID531
Feature activation-1.15e-02

Pos logit contributions
ンジ+3.353
 Valhalla+3.206
+3.185
 pose+3.171
+3.072
Neg logit contributions
ours-4.388
oured-4.364
ateral-3.792
hap-3.748
yss-3.704
Token:
Token ID25
Feature activation+4.86e-01

Pos logit contributions
ours+0.037
oured+0.037
ateral+0.032
hap+0.031
aths+0.031
Neg logit contributions
ンジ-0.024
-0.023
 Valhalla-0.022
-0.022
 pose-0.022
Token
Token ID564
Feature activation+8.15e-01

Pos logit contributions
ンジ+0.715
+0.659
 Valhalla+0.634
 pose+0.621
+0.614
Neg logit contributions
ours-1.869
oured-1.862
ateral-1.653
hap-1.641
aths-1.631
Token be
Token ID307
Feature activation+5.19e-01

Pos logit contributions
ンジ+1.272
+1.192
 Valhalla+1.165
 pose+1.148
+1.130
Neg logit contributions
ours-2.413
oured-2.404
ateral-2.100
hap-2.080
aths-2.072
Token due
Token ID2233
Feature activation-2.43e-01

Pos logit contributions
ンジ+1.207
+1.143
 Valhalla+1.136
 pose+1.120
+1.106
Neg logit contributions
ours-1.541
oured-1.533
ateral-1.312
hap-1.299
aths-1.290
Token to
Token ID284
Feature activation+7.70e-02

Pos logit contributions
ours+0.739
oured+0.736
ateral+0.629
hap+0.622
aths+0.618
Neg logit contributions
ンジ-0.557
-0.529
 Valhalla-0.516
 pose-0.507
-0.505
Token a
Token ID257
Feature activation+6.59e-01

Pos logit contributions
ンジ+0.165
+0.156
 Valhalla+0.152
 pose+0.150
+0.149
Neg logit contributions
ours-0.245
oured-0.244
ateral-0.210
hap-0.208
aths-0.207
Token military
Token ID2422
Feature activation+6.40e-02

Pos logit contributions
ンジ+1.545
+1.469
 Valhalla+1.443
 pose+1.420
+1.409
Neg logit contributions
ours-1.959
oured-1.949
ateral-1.663
hap-1.644
aths-1.631
Token retirement
Token ID10737
Feature activation+3.12e+00

Pos logit contributions
ンジ+0.135
+0.127
 Valhalla+0.125
 pose+0.122
+0.122
Neg logit contributions
ours-0.206
oured-0.205
ateral-0.177
hap-0.175
aths-0.174
Token fund
Token ID1814
Feature activation+1.08e+00

Pos logit contributions
 pose+5.581
ンジ+5.455
 Valhalla+5.347
+5.339
 dream+5.335
Neg logit contributions
ours-10.238
oured-10.132
yss-8.866
hap-8.811
ateral-8.756
Token,
Token ID11
Feature activation+2.59e-01

Pos logit contributions
ンジ+2.537
+2.423
 Valhalla+2.416
 pose+2.405
+2.330
Neg logit contributions
ours-3.119
oured-3.098
ateral-2.651
hap-2.632
aths-2.611
Token according
Token ID1864
Feature activation-2.24e-01

Pos logit contributions
ンジ+0.594
+0.565
 Valhalla+0.552
 pose+0.543
+0.540
Neg logit contributions
ours-0.784
oured-0.781
ateral-0.667
hap-0.660
aths-0.656
Token to
Token ID284
Feature activation-1.82e-01

Pos logit contributions
ours+0.623
oured+0.615
ateral+0.521
hap+0.515
aths+0.513
Neg logit contributions
ンジ-0.575
-0.557
 Valhalla-0.529
-0.524
 pose-0.522
Token the
Token ID262
Feature activation-1.16e-01

Pos logit contributions
ours+0.547
oured+0.545
ateral+0.466
hap+0.462
aths+0.458
Neg logit contributions
ンジ-0.422
-0.402
 Valhalla-0.387
 pose-0.382
-0.381
Token Clinton
Token ID2605
Feature activation+3.49e-01

Pos logit contributions
ンジ+2.778
+2.593
 pose+2.559
 Valhalla+2.551
+2.498
Neg logit contributions
ours-5.221
oured-5.210
ateral-4.540
hap-4.495
aths-4.473
Token Hillary
Token ID5073
Feature activation+1.70e+00

Pos logit contributions
ンジ+0.739
+0.698
 Valhalla+0.678
+0.661
 pose+0.661
Neg logit contributions
ours-1.126
oured-1.122
ateral-0.967
hap-0.957
aths-0.953
Token Diane
Token ID26542
Feature activation+2.30e+00

Pos logit contributions
ンジ+3.794
+3.598
 Valhalla+3.448
+3.327
 pose+3.286
Neg logit contributions
ours-5.422
oured-5.408
ateral-4.597
hap-4.542
yss-4.523
Token Rodham
Token ID41848
Feature activation+1.23e+00

Pos logit contributions
ンジ+3.911
+3.640
 pose+3.623
 Valhalla+3.575
+3.512
Neg logit contributions
ours-8.151
oured-8.131
hap-7.149
ateral-7.128
aths-7.060
Token Clinton
Token ID2605
Feature activation+1.23e+00

Pos logit contributions
ンジ+2.306
+2.140
 Valhalla+2.116
 pose+2.103
+2.075
Neg logit contributions
ours-4.220
oured-4.219
ateral-3.675
hap-3.649
aths-3.649
TokenSanders
Token ID26747
Feature activation+3.10e+00

Pos logit contributions
ンジ+3.156
+2.969
 pose+2.956
 Valhalla+2.954
+2.932
Neg logit contributions
oured-3.287
ours-3.259
ateral-2.777
aths-2.740
hap-2.711
Token:
Token ID25
Feature activation+1.93e+00

Pos logit contributions
ンジ+4.513
 pose+4.403
 Valhalla+4.123
 dream+4.015
+3.973
Neg logit contributions
oured-11.259
ours-11.148
aths-10.118
ateral-9.939
hap-9.662
Token '
Token ID705
Feature activation+7.67e-01

Pos logit contributions
ンジ+3.316
 Valhalla+3.140
 pose+3.107
+3.085
+3.044
Neg logit contributions
ours-6.812
oured-6.794
ateral-5.975
hap-5.969
aths-5.887
TokenI
Token ID40
Feature activation+1.10e+00

Pos logit contributions
ンジ+1.868
+1.773
 Valhalla+1.753
 pose+1.739
+1.712
Neg logit contributions
oured-2.191
ours-2.177
ateral-1.857
hap-1.824
aths-1.815
Token fully
Token ID3938
Feature activation-9.04e-02

Pos logit contributions
ンジ+1.672
 pose+1.543
 Valhalla+1.483
 dream+1.481
+1.448
Neg logit contributions
ours-4.088
oured-4.028
hap-3.616
ateral-3.610
yss-3.575
Token expect
Token ID1607
Feature activation+1.77e+00

Pos logit contributions
oured+0.288
ours+0.287
ateral+0.248
hap+0.244
aths+0.242
Neg logit contributions
ンジ-0.196
-0.184
 Valhalla-0.181
-0.175
 pose-0.172
Token trees
Token ID7150
Feature activation+3.30e-01

Pos logit contributions
ours+0.938
oured+0.937
ateral+0.797
hap+0.790
aths+0.786
Neg logit contributions
ンジ-0.706
-0.668
 Valhalla-0.648
-0.639
 pose-0.630
Token bearing
Token ID14121
Feature activation-8.04e-01

Pos logit contributions
ンジ+0.745
+0.707
 Valhalla+0.692
 pose+0.681
+0.676
Neg logit contributions
ours-1.010
oured-1.007
ateral-0.861
hap-0.852
aths-0.848
Token a
Token ID257
Feature activation+3.83e-01

Pos logit contributions
oured+2.413
ours+2.402
ateral+2.060
hap+2.044
aths+2.026
Neg logit contributions
ンジ-1.852
-1.759
 Valhalla-1.693
-1.671
 pose-1.644
Token high
Token ID1029
Feature activation+7.55e-01

Pos logit contributions
ンジ+0.812
+0.768
 Valhalla+0.749
 pose+0.735
+0.732
Neg logit contributions
ours-1.228
oured-1.224
ateral-1.055
hap-1.045
aths-1.039
Token-
Token ID12
Feature activation+4.88e-01

Pos logit contributions
ンジ+1.645
+1.558
 Valhalla+1.528
 pose+1.504
+1.493
Neg logit contributions
ours-2.369
oured-2.359
ateral-2.025
hap-2.005
aths-1.995
Tokent
Token ID83
Feature activation+3.10e+00

Pos logit contributions
ンジ+1.306
+1.247
 Valhalla+1.226
 pose+1.213
+1.208
Neg logit contributions
ours-1.287
oured-1.284
ateral-1.069
hap-1.050
aths-1.043
Tokenide
Token ID485
Feature activation-5.46e-02

Pos logit contributions
ンジ+6.618
 Valhalla+6.390
+6.387
 pose+6.315
+6.241
Neg logit contributions
oured-8.394
ours-8.359
hap-7.327
ateral-7.293
yss-7.165
Token mark
Token ID1317
Feature activation+5.45e-01

Pos logit contributions
ours+0.176
oured+0.175
ateral+0.151
hap+0.150
aths+0.149
Neg logit contributions
ンジ-0.115
-0.109
 Valhalla-0.106
 pose-0.103
-0.103
Token of
Token ID286
Feature activation+1.99e-01

Pos logit contributions
ンジ+1.378
+1.314
 Valhalla+1.293
 pose+1.282
+1.266
Neg logit contributions
ours-1.514
oured-1.508
ateral-1.268
hap-1.252
aths-1.247
Token black
Token ID2042
Feature activation-6.92e-02

Pos logit contributions
ンジ+0.462
+0.439
 Valhalla+0.428
 pose+0.420
+0.420
Neg logit contributions
ours-0.597
oured-0.596
ateral-0.508
hap-0.503
aths-0.500
Token oil
Token ID3056
Feature activation-2.88e-01

Pos logit contributions
ours+0.220
oured+0.220
ateral+0.189
hap+0.187
aths+0.186
Neg logit contributions
ンジ-0.149
-0.140
 Valhalla-0.137
 pose-0.134
-0.134
Token.
Token ID13
Feature activation-4.08e-01

Pos logit contributions
ours+1.027
oured+1.022
ateral+0.850
hap+0.837
aths+0.834
Neg logit contributions
ンジ-1.059
-1.012
 Valhalla-0.990
-0.974
 pose-0.973
Token At
Token ID1629
Feature activation+1.09e-01

Pos logit contributions
ours+1.170
oured+1.159
ateral+0.990
aths+0.974
hap+0.966
Neg logit contributions
ンジ-1.003
-0.950
 pose-0.915
 Valhalla-0.911
-0.910
Token midnight
Token ID15896
Feature activation-1.43e-01

Pos logit contributions
ンジ+0.249
+0.236
 Valhalla+0.229
+0.225
 pose+0.225
Neg logit contributions
ours-0.333
oured-0.332
ateral-0.285
hap-0.281
aths-0.280
Token,
Token ID11
Feature activation-2.03e-01

Pos logit contributions
ours+0.414
oured+0.412
ateral+0.349
hap+0.345
aths+0.343
Neg logit contributions
ンジ-0.349
-0.333
 Valhalla-0.325
 pose-0.319
-0.319
Token a
Token ID257
Feature activation+5.00e-02

Pos logit contributions
ours+0.638
oured+0.635
ateral+0.546
aths+0.541
hap+0.541
Neg logit contributions
ンジ-0.449
-0.424
 Valhalla-0.406
-0.402
 pose-0.398
Token cr
Token ID1067
Feature activation-3.59e+00

Pos logit contributions
ンジ+0.109
+0.102
 Valhalla+0.098
+0.098
 pose+0.096
Neg logit contributions
ours-0.159
oured-0.158
ateral-0.136
hap-0.134
aths-0.134
Tokenon
Token ID261
Feature activation-4.04e-01

Pos logit contributions
oured+8.055
ours+8.021
ateral+6.569
hap+6.328
aths+6.214
Neg logit contributions
ンジ-10.985
-10.205
 Valhalla-9.706
-9.618
 pose-9.322
Token job
Token ID1693
Feature activation+3.75e-01

Pos logit contributions
ours+1.172
oured+1.172
ateral+0.984
hap+0.975
aths+0.972
Neg logit contributions
ンジ-0.979
-0.944
 Valhalla-0.918
-0.898
 pose-0.887
Token or
Token ID393
Feature activation+1.08e-01

Pos logit contributions
ンジ+0.788
+0.745
 Valhalla+0.727
 pose+0.713
+0.710
Neg logit contributions
ours-1.211
oured-1.207
ateral-1.041
hap-1.031
aths-1.026
Token some
Token ID617
Feature activation-5.15e-01

Pos logit contributions
ンジ+0.243
+0.230
 Valhalla+0.223
+0.219
 pose+0.219
Neg logit contributions
ours-0.332
oured-0.331
ateral-0.283
hap-0.281
aths-0.279
Token other
Token ID584
Feature activation-4.50e-01

Pos logit contributions
ours+1.575
oured+1.568
ateral+1.347
hap+1.322
aths+1.319
Neg logit contributions
ンジ-1.176
-1.113
 Valhalla-1.064
-1.058
 pose-1.041
Token Keller
Token ID35999
Feature activation-1.71e-01

Pos logit contributions
ンジ+2.055
+1.957
 Valhalla+1.920
 pose+1.901
+1.872
Neg logit contributions
ours-2.975
oured-2.959
ateral-2.553
hap-2.518
aths-2.506
Token,
Token ID11
Feature activation-3.36e-01

Pos logit contributions
ours+0.546
oured+0.544
ateral+0.468
hap+0.463
aths+0.460
Neg logit contributions
ンジ-0.367
-0.345
 Valhalla-0.335
-0.327
 pose-0.323
Token with
Token ID351
Feature activation-1.16e+00

Pos logit contributions
ours+1.033
oured+1.032
ateral+0.878
hap+0.868
aths+0.863
Neg logit contributions
ンジ-0.764
-0.719
 Valhalla-0.695
-0.682
 pose-0.670
Token a
Token ID257
Feature activation-1.13e-01

Pos logit contributions
oured+3.553
ours+3.544
hap+3.023
ateral+3.021
yss+2.966
Neg logit contributions
ンジ-2.624
-2.446
 Valhalla-2.367
-2.316
 pose-2.260
Token doll
Token ID3654
Feature activation-2.30e+00

Pos logit contributions
oured+0.369
ours+0.369
ateral+0.317
hap+0.317
aths+0.313
Neg logit contributions
ンジ-0.234
-0.219
 Valhalla-0.212
-0.208
 pose-0.201
Token cr
Token ID1067
Feature activation-3.56e+00

Pos logit contributions
oured+6.665
ours+6.628
hap+5.629
ateral+5.578
aths+5.479
Neg logit contributions
ンジ-5.441
-5.167
 Valhalla-4.931
-4.864
 pose-4.625
Tokenad
Token ID324
Feature activation-1.11e-01

Pos logit contributions
oured+6.433
ours+6.242
ateral+4.643
aths+4.532
hap+4.357
Neg logit contributions
ンジ-12.925
-12.082
 Valhalla-11.468
-11.385
 pose-10.920
Tokenled
Token ID992
Feature activation-8.50e-01

Pos logit contributions
ours+0.295
oured+0.295
ateral+0.245
hap+0.242
aths+0.240
Neg logit contributions
ンジ-0.299
-0.288
 Valhalla-0.281
-0.276
 pose-0.275
Token in
Token ID287
Feature activation+3.15e-01

Pos logit contributions
oured+2.484
ours+2.484
ateral+2.115
hap+2.113
aths+2.076
Neg logit contributions
ンジ-2.043
-1.960
 Valhalla-1.838
-1.834
 pose-1.778
Token her
Token ID607
Feature activation-2.38e-02

Pos logit contributions
ンジ+0.675
+0.636
 Valhalla+0.615
+0.603
 pose+0.602
Neg logit contributions
ours-1.008
oured-1.005
ateral-0.867
hap-0.859
aths-0.853
Token lap
Token ID14779
Feature activation-9.28e-01

Pos logit contributions
ours+0.080
oured+0.079
ateral+0.069
hap+0.069
aths+0.068
Neg logit contributions
ンジ-0.047
-0.045
 Valhalla-0.043
-0.042
 pose-0.041
TokenF
Token ID37
Feature activation-3.06e-01

Pos logit contributions
ンジ+0.087
+0.083
 Valhalla+0.082
 pose+0.080
+0.080
Neg logit contributions
ours-0.098
oured-0.098
ateral-0.083
hap-0.082
aths-0.081
Tokenalling
Token ID9221
Feature activation-5.92e-01

Pos logit contributions
ours+0.853
oured+0.845
ateral+0.709
hap+0.698
aths+0.697
Neg logit contributions
ンジ-0.790
-0.752
 Valhalla-0.735
 pose-0.722
-0.719
Token together
Token ID1978
Feature activation-3.48e-01

Pos logit contributions
ours+1.846
oured+1.837
ateral+1.584
hap+1.567
aths+1.555
Neg logit contributions
ンジ-1.307
-1.236
 Valhalla-1.191
 pose-1.172
-1.170
Token in
Token ID287
Feature activation+2.01e-01

Pos logit contributions
ours+1.050
oured+1.040
ateral+0.892
hap+0.883
aths+0.876
Neg logit contributions
ンジ-0.813
-0.769
 Valhalla-0.744
-0.729
 pose-0.726
Token a
Token ID257
Feature activation+4.43e-01

Pos logit contributions
ンジ+0.437
+0.413
 Valhalla+0.398
 pose+0.393
+0.392
Neg logit contributions
ours-0.634
oured-0.632
ateral-0.543
hap-0.538
aths-0.536
Token cr
Token ID1067
Feature activation-3.47e+00

Pos logit contributions
ンジ+0.918
+0.864
 Valhalla+0.827
+0.820
 pose+0.810
Neg logit contributions
ours-1.447
oured-1.443
ateral-1.244
hap-1.237
aths-1.232
Tokenum
Token ID388
Feature activation-1.69e-01

Pos logit contributions
oured+8.013
ours+7.924
ateral+6.492
onel+6.254
aths+6.216
Neg logit contributions
ンジ-10.753
-9.931
 Valhalla-9.431
-9.304
 pose-9.079
Tokenpled
Token ID10137
Feature activation-1.01e+00

Pos logit contributions
ours+0.546
oured+0.544
ateral+0.470
hap+0.465
aths+0.462
Neg logit contributions
ンジ-0.353
-0.334
 Valhalla-0.326
 pose-0.319
-0.318
Token mess
Token ID2085
Feature activation-1.59e+00

Pos logit contributions
oured+3.096
ours+3.076
hap+2.631
ateral+2.621
aths+2.578
Neg logit contributions
ンジ-2.304
-2.193
-2.047
 Valhalla-2.027
 pose-1.942
Token of
Token ID286
Feature activation-1.87e-01

Pos logit contributions
ours+4.590
oured+4.536
ateral+3.847
hap+3.811
aths+3.774
Neg logit contributions
ンジ-3.877
-3.714
 Valhalla-3.479
-3.477
 pose-3.385
Token mutual
Token ID13584
Feature activation+1.83e-01

Pos logit contributions
ours+0.586
oured+0.585
ateral+0.500
hap+0.496
aths+0.493
Neg logit contributions
ンジ-0.414
-0.390
 Valhalla-0.376
 pose-0.370
-0.370
Token group
Token ID1448
Feature activation-1.26e-01

Pos logit contributions
ンジ+3.301
+3.097
 Valhalla+3.092
 pose+3.052
+3.019
Neg logit contributions
ours-5.089
oured-5.048
hap-4.428
ateral-4.416
aths-4.374
Token,
Token ID11
Feature activation-1.99e-01

Pos logit contributions
ours+0.382
oured+0.381
ateral+0.323
hap+0.318
aths+0.316
Neg logit contributions
ンジ-0.294
-0.283
 Valhalla-0.273
-0.266
 pose-0.266
Token including
Token ID1390
Feature activation-1.86e-01

Pos logit contributions
ours+0.610
oured+0.608
ateral+0.519
hap+0.511
aths+0.510
Neg logit contributions
ンジ-0.454
-0.430
 Valhalla-0.416
 pose-0.410
-0.409
Token headset
Token ID23492
Feature activation+9.33e-01

Pos logit contributions
ours+0.571
oured+0.569
ateral+0.486
hap+0.479
aths+0.477
Neg logit contributions
ンジ-0.425
-0.402
 Valhalla-0.390
 pose-0.384
-0.383
Token,
Token ID11
Feature activation-1.74e-01

Pos logit contributions
ンジ+2.226
+2.118
 Valhalla+2.089
 pose+2.057
+2.035
Neg logit contributions
ours-2.725
oured-2.712
ateral-2.302
hap-2.286
aths-2.272
Token cr
Token ID1067
Feature activation-3.36e+00

Pos logit contributions
ours+0.562
oured+0.560
ateral+0.483
hap+0.476
aths+0.474
Neg logit contributions
ンジ-0.368
-0.348
 Valhalla-0.335
 pose-0.330
-0.330
Tokenanks
Token ID2283
Feature activation-1.96e-01

Pos logit contributions
ours+7.066
oured+6.908
ateral+5.765
aths+5.496
onel+5.207
Neg logit contributions
ンジ-10.720
-10.198
 Valhalla-9.618
-9.509
 pose-9.316
Token and
Token ID290
Feature activation+5.64e-01

Pos logit contributions
ours+0.622
oured+0.620
ateral+0.536
hap+0.535
aths+0.526
Neg logit contributions
ンジ-0.418
-0.400
 Valhalla-0.386
 pose-0.378
-0.376
Token brakes
Token ID23746
Feature activation+1.62e+00

Pos logit contributions
ンジ+1.205
+1.140
 Valhalla+1.122
 pose+1.100
+1.091
Neg logit contributions
ours-1.793
oured-1.786
ateral-1.538
hap-1.529
aths-1.518
Token.
Token ID13
Feature activation+2.82e-01

Pos logit contributions
ンジ+3.937
+3.791
 Valhalla+3.761
 pose+3.715
+3.644
Neg logit contributions
ours-4.488
oured-4.463
ateral-3.768
hap-3.751
aths-3.745
Token The
Token ID383
Feature activation-9.14e-02

Pos logit contributions
ンジ+0.659
+0.627
 Valhalla+0.613
 pose+0.603
+0.600
Neg logit contributions
ours-0.843
oured-0.840
ateral-0.715
hap-0.708
aths-0.704
Token rave
Token ID46242
Feature activation-3.61e-01

Pos logit contributions
ンジ+1.827
+1.724
 Valhalla+1.693
 pose+1.671
+1.652
Neg logit contributions
ours-2.595
oured-2.583
ateral-2.224
hap-2.203
aths-2.178
Token drugs
Token ID5010
Feature activation+6.55e-01

Pos logit contributions
ours+1.114
oured+1.110
ateral+0.945
aths+0.939
hap+0.935
Neg logit contributions
ンジ-0.802
-0.762
 Valhalla-0.743
-0.727
 pose-0.723
Token come
Token ID1282
Feature activation+6.52e-01

Pos logit contributions
ンジ+1.577
+1.497
 Valhalla+1.476
 pose+1.459
+1.441
Neg logit contributions
ours-1.902
oured-1.897
ateral-1.612
hap-1.594
aths-1.583
Token off
Token ID572
Feature activation-1.05e+00

Pos logit contributions
ンジ+1.412
+1.326
 Valhalla+1.311
 pose+1.292
+1.276
Neg logit contributions
ours-2.063
oured-2.050
ateral-1.769
hap-1.748
aths-1.737
Token their
Token ID511
Feature activation+7.96e-01

Pos logit contributions
ours+3.124
oured+3.111
ateral+2.636
hap+2.619
aths+2.600
Neg logit contributions
ンジ-2.423
-2.333
 Valhalla-2.219
-2.208
 pose-2.168
Token cr
Token ID1067
Feature activation-3.23e+00

Pos logit contributions
ンジ+1.733
+1.641
 Valhalla+1.601
 pose+1.571
+1.565
Neg logit contributions
ours-2.508
oured-2.499
ateral-2.147
hap-2.126
aths-2.115
Tokenutch
Token ID7140
Feature activation+4.32e-01

Pos logit contributions
ours+8.013
oured+7.813
aths+6.579
ateral+6.428
onel+6.341
Neg logit contributions
ンジ-9.290
-8.632
 Valhalla-8.074
-7.939
 pose-7.894
Token.
Token ID13
Feature activation+6.04e-02

Pos logit contributions
ンジ+1.118
+1.069
 Valhalla+1.049
 pose+1.034
+1.029
Neg logit contributions
ours-1.180
oured-1.176
ateral-0.986
hap-0.974
aths-0.967
Token
Token ID198
Feature activation+2.40e-01

Pos logit contributions
ンジ+0.147
+0.140
 Valhalla+0.137
 pose+0.136
+0.134
Neg logit contributions
ours-0.174
oured-0.173
ateral-0.146
hap-0.144
aths-0.144
Token
Token ID198
Feature activation+7.19e-02

Pos logit contributions
ンジ+0.588
+0.565
 Valhalla+0.543
 pose+0.535
+0.530
Neg logit contributions
ours-0.697
oured-0.688
ateral-0.583
hap-0.576
aths-0.574
TokenThere
Token ID1858
Feature activation-7.14e-01

Pos logit contributions
ンジ+0.186
+0.178
 Valhalla+0.175
 pose+0.172
+0.170
Neg logit contributions
ours-0.198
oured-0.197
ateral-0.165
hap-0.163
aths-0.162
Tokenf
Token ID69
Feature activation+7.43e-01

Pos logit contributions
ours+1.959
oured+1.928
ateral+1.590
hap+1.576
aths+1.562
Neg logit contributions
ンジ-2.245
-2.127
 Valhalla-2.088
-2.051
 pose-2.044
Tokenunt
Token ID2797
Feature activation+3.62e-01

Pos logit contributions
ンジ+2.103
+2.021
 Valhalla+1.987
 pose+1.966
+1.951
Neg logit contributions
oured-1.828
ours-1.827
ateral-1.503
hap-1.483
aths-1.473
Token p
Token ID279
Feature activation-8.41e-01

Pos logit contributions
ンジ+0.922
+0.883
 Valhalla+0.869
 pose+0.861
+0.852
Neg logit contributions
ours-0.990
oured-0.990
ateral-0.831
hap-0.826
aths-0.817
Tokenère
Token ID35979
Feature activation-5.57e-01

Pos logit contributions
ours+2.029
oured+2.001
ateral+1.619
aths+1.591
hap+1.586
Neg logit contributions
ンジ-2.498
-2.388
 Valhalla-2.349
-2.312
 pose-2.306
Token,
Token ID11
Feature activation+8.54e-02

Pos logit contributions
ours+1.670
oured+1.658
ateral+1.414
hap+1.395
aths+1.390
Neg logit contributions
ンジ-1.309
-1.238
 Valhalla-1.207
-1.188
 pose-1.177
Token cr
Token ID1067
Feature activation-3.22e+00

Pos logit contributions
ンジ+0.199
+0.189
 Valhalla+0.185
 pose+0.181
+0.181
Neg logit contributions
ours-0.256
oured-0.255
ateral-0.218
hap-0.215
aths-0.214
Tokené
Token ID2634
Feature activation-4.20e-02

Pos logit contributions
ours+6.991
oured+6.768
ateral+5.378
aths+5.283
hap+5.062
Neg logit contributions
ンジ-10.417
-9.900
 Valhalla-9.436
-9.136
 pose-8.885
Tokenateur
Token ID15093
Feature activation+2.18e-02

Pos logit contributions
ours+0.096
oured+0.095
ateral+0.077
hap+0.076
aths+0.075
Neg logit contributions
ンジ-0.128
-0.123
 Valhalla-0.122
-0.120
 pose-0.120
Token origin
Token ID8159
Feature activation+4.67e-01

Pos logit contributions
ンジ+0.051
+0.049
 Valhalla+0.048
 pose+0.047
+0.047
Neg logit contributions
ours-0.064
oured-0.064
ateral-0.055
hap-0.054
aths-0.054
Tokenel
Token ID417
Feature activation+7.75e-02

Pos logit contributions
ンジ+1.322
+1.267
 Valhalla+1.244
 pose+1.235
+1.224
Neg logit contributions
ours-1.151
oured-1.149
ateral-0.945
hap-0.939
aths-0.925
Token du
Token ID7043
Feature activation-5.16e-01

Pos logit contributions
ンジ+0.194
+0.185
 Valhalla+0.182
 pose+0.179
+0.178
Neg logit contributions
ours-0.218
oured-0.217
ateral-0.183
hap-0.181
aths-0.180
Token way
Token ID835
Feature activation-3.88e-01

Pos logit contributions
ンジ+0.474
+0.448
 Valhalla+0.435
+0.428
 pose+0.419
Neg logit contributions
ours-0.650
oured-0.645
ateral-0.552
hap-0.547
aths-0.545
Token through
Token ID832
Feature activation-2.89e-01

Pos logit contributions
ours+1.176
oured+1.162
ateral+0.995
hap+0.980
aths+0.977
Neg logit contributions
ンジ-0.908
-0.875
 Valhalla-0.827
-0.814
 pose-0.798
Token a
Token ID257
Feature activation+7.26e-01

Pos logit contributions
ours+0.860
oured+0.853
ateral+0.721
hap+0.716
aths+0.709
Neg logit contributions
ンジ-0.690
-0.658
 Valhalla-0.634
-0.624
 pose-0.617
Token press
Token ID1803
Feature activation-3.26e-01

Pos logit contributions
ンジ+1.358
+1.274
 Valhalla+1.239
 pose+1.211
+1.205
Neg logit contributions
ours-2.511
oured-2.502
ateral-2.182
hap-2.162
aths-2.151
Token of
Token ID286
Feature activation+3.56e-01

Pos logit contributions
ours+0.937
oured+0.933
ateral+0.783
hap+0.775
aths+0.768
Neg logit contributions
ンジ-0.803
-0.767
 Valhalla-0.754
-0.734
 pose-0.729
Token cr
Token ID1067
Feature activation-3.21e+00

Pos logit contributions
ンジ+0.766
+0.724
 Valhalla+0.707
 pose+0.693
+0.691
Neg logit contributions
ours-1.132
oured-1.127
ateral-0.970
hap-0.960
aths-0.955
Tokenin
Token ID259
Feature activation+6.01e-01

Pos logit contributions
ours+7.068
oured+7.020
aths+5.607
ateral+5.528
hap+5.326
Neg logit contributions
ンジ-10.347
-9.619
 Valhalla-9.257
-9.115
 pose-8.859
Tokenoline
Token ID14453
Feature activation-2.33e-01

Pos logit contributions
ンジ+1.385
+1.316
 Valhalla+1.287
 pose+1.265
+1.259
Neg logit contributions
ours-1.819
oured-1.813
ateral-1.547
hap-1.531
aths-1.522
Token and
Token ID290
Feature activation+7.53e-02

Pos logit contributions
ours+0.660
oured+0.656
ateral+0.553
hap+0.546
aths+0.544
Neg logit contributions
ンジ-0.583
-0.558
 Valhalla-0.547
-0.533
 pose-0.532
Token dra
Token ID28841
Feature activation+1.52e+00

Pos logit contributions
ンジ+0.157
+0.148
 Valhalla+0.144
+0.140
 pose+0.140
Neg logit contributions
ours-0.245
oured-0.244
ateral-0.211
hap-0.209
aths-0.208
Tokenper
Token ID525
Feature activation-1.54e-01

Pos logit contributions
ンジ+3.473
+3.271
 Valhalla+3.236
 pose+3.227
+3.156
Neg logit contributions
ours-4.497
oured-4.496
ateral-3.835
hap-3.780
aths-3.754
Token cast
Token ID3350
Feature activation-6.83e-01

Pos logit contributions
ours+0.195
oured+0.194
ateral+0.168
aths+0.167
hap+0.167
Neg logit contributions
ンジ-0.114
-0.108
 Valhalla-0.103
-0.100
 pose-0.097
Token their
Token ID511
Feature activation+7.31e-01

Pos logit contributions
ours+1.884
oured+1.881
ateral+1.578
aths+1.560
hap+1.545
Neg logit contributions
ンジ-1.758
-1.674
 Valhalla-1.626
-1.592
 pose-1.583
Token ballots
Token ID20103
Feature activation+4.09e-01

Pos logit contributions
ンジ+1.883
+1.799
 Valhalla+1.771
 pose+1.751
+1.736
Neg logit contributions
ours-2.002
oured-1.991
ateral-1.666
hap-1.654
aths-1.632
Token before
Token ID878
Feature activation-1.89e-01

Pos logit contributions
ンジ+0.934
+0.886
 Valhalla+0.869
 pose+0.856
+0.850
Neg logit contributions
ours-1.241
oured-1.238
ateral-1.057
hap-1.046
aths-1.039
Token the
Token ID262
Feature activation-7.64e-01

Pos logit contributions
ours+0.585
oured+0.583
ateral+0.498
hap+0.495
aths+0.493
Neg logit contributions
ンジ-0.426
-0.405
 Valhalla-0.390
-0.384
 pose-0.382
Token Nov
Token ID5267
Feature activation-3.21e+00

Pos logit contributions
ours+2.311
oured+2.302
aths+1.948
ateral+1.945
hap+1.938
Neg logit contributions
ンジ-1.777
-1.685
 Valhalla-1.601
-1.587
 pose-1.565
Token.
Token ID13
Feature activation-8.14e-01

Pos logit contributions
ours+6.260
oured+6.107
aths+4.716
ateral+4.529
onel+4.381
Neg logit contributions
ンジ-10.677
-10.224
-9.664
 Valhalla-9.627
 pose-9.533
Token 8
Token ID807
Feature activation+4.93e-01

Pos logit contributions
ours+2.027
oured+2.020
aths+1.639
ateral+1.639
hap+1.600
Neg logit contributions
ンジ-2.334
-2.254
 Valhalla-2.140
-2.128
 pose-2.100
Token election
Token ID3071
Feature activation-2.18e-01

Pos logit contributions
ンジ+1.376
+1.320
 Valhalla+1.308
 pose+1.296
+1.281
Neg logit contributions
ours-1.232
oured-1.228
ateral-1.016
hap-1.003
aths-0.986
Token.
Token ID13
Feature activation+2.68e-01

Pos logit contributions
ours+0.559
oured+0.555
ateral+0.458
hap+0.453
aths+0.453
Neg logit contributions
ンジ-0.608
-0.582
 Valhalla-0.565
-0.558
 pose-0.555
Token
Token ID198
Feature activation+3.46e-01

Pos logit contributions
ンジ+0.607
+0.576
 Valhalla+0.564
 pose+0.554
+0.552
Neg logit contributions
ours-0.819
oured-0.816
ateral-0.698
hap-0.691
aths-0.686
Tokenrat
Token ID10366
Feature activation+6.33e-01

Pos logit contributions
ours+0.375
oured+0.374
ateral+0.314
hap+0.310
aths+0.308
Neg logit contributions
ンジ-0.349
-0.329
 Valhalla-0.324
-0.318
 pose-0.317
Tokenrice
Token ID20970
Feature activation-1.10e+00

Pos logit contributions
ンジ+1.586
+1.509
 Valhalla+1.485
 pose+1.467
+1.455
Neg logit contributions
ours-1.769
oured-1.766
ateral-1.488
hap-1.479
aths-1.471
Token,
Token ID11
Feature activation+1.25e-01

Pos logit contributions
ours+3.249
oured+3.208
ateral+2.735
hap+2.695
aths+2.676
Neg logit contributions
ンジ-2.636
-2.508
 Valhalla-2.422
-2.384
 pose-2.364
TokenA
Token ID32
Feature activation-6.35e-02

Pos logit contributions
ンジ+0.288
+0.273
 Valhalla+0.267
 pose+0.262
+0.261
Neg logit contributions
ours-0.380
oured-0.378
ateral-0.323
hap-0.319
aths-0.318
Tokenu
Token ID84
Feature activation+3.11e-03

Pos logit contributions
ours+0.181
oured+0.180
ateral+0.152
hap+0.150
aths+0.149
Neg logit contributions
ンジ-0.158
-0.151
 Valhalla-0.147
-0.144
 pose-0.144
Token cr
Token ID1067
Feature activation-3.11e+00

Pos logit contributions
ンジ+0.007
+0.007
 Valhalla+0.007
+0.006
 pose+0.006
Neg logit contributions
ours-0.010
oured-0.009
ateral-0.008
hap-0.008
aths-0.008
Tokené
Token ID2634
Feature activation+2.38e-02

Pos logit contributions
ours+7.353
oured+7.130
ateral+5.695
aths+5.647
hap+5.443
Neg logit contributions
ンジ-9.329
-8.759
 Valhalla-8.354
-8.159
 pose-7.998
Tokenp
Token ID79
Feature activation+7.85e-01

Pos logit contributions
ンジ+0.066
+0.063
 Valhalla+0.062
 pose+0.061
+0.060
Neg logit contributions
ours-0.061
oured-0.061
ateral-0.051
hap-0.050
aths-0.050
Tokenusc
Token ID16241
Feature activation-5.49e-01

Pos logit contributions
ンジ+2.195
+2.113
 Valhalla+2.073
 pose+2.052
+2.036
Neg logit contributions
ours-1.959
oured-1.952
ateral-1.611
hap-1.596
aths-1.578
Tokenule
Token ID2261
Feature activation-2.75e-01

Pos logit contributions
ours+1.376
oured+1.355
ateral+1.118
hap+1.089
aths+1.088
Neg logit contributions
ンジ-1.574
-1.513
 Valhalla-1.485
-1.451
 pose-1.446
Token de
Token ID390
Feature activation-2.66e-01

Pos logit contributions
ours+0.799
oured+0.795
ateral+0.674
hap+0.666
aths+0.662
Neg logit contributions
ンジ-0.666
-0.635
 Valhalla-0.621
 pose-0.610
-0.608
Token,
Token ID11
Feature activation-1.82e-01

Pos logit contributions
ours+0.754
oured+0.751
ateral+0.636
hap+0.629
aths+0.625
Neg logit contributions
ンジ-0.640
-0.609
 Valhalla-0.597
 pose-0.588
-0.585
Token the
Token ID262
Feature activation-4.77e-01

Pos logit contributions
ours+0.561
oured+0.560
ateral+0.479
hap+0.474
aths+0.471
Neg logit contributions
ンジ-0.409
-0.391
 Valhalla-0.378
-0.371
 pose-0.369
Token police
Token ID1644
Feature activation-1.33e+00

Pos logit contributions
ours+1.481
oured+1.476
ateral+1.265
hap+1.249
aths+1.246
Neg logit contributions
ンジ-1.064
-1.013
 Valhalla-0.972
-0.959
 pose-0.946
Token announced
Token ID3414
Feature activation+5.29e-02

Pos logit contributions
oured+3.682
ours+3.674
ateral+3.055
aths+3.020
hap+3.005
Neg logit contributions
ンジ-3.430
-3.257
 Valhalla-3.114
-3.110
 pose-3.030
Token the
Token ID262
Feature activation-5.82e-01

Pos logit contributions
ンジ+0.121
+0.115
 Valhalla+0.113
 pose+0.111
+0.110
Neg logit contributions
ours-0.161
oured-0.160
ateral-0.137
hap-0.135
aths-0.135
Token fo
Token ID11511
Feature activation-3.10e+00

Pos logit contributions
ours+1.682
oured+1.680
ateral+1.415
hap+1.395
aths+1.394
Neg logit contributions
ンジ-1.430
-1.359
 Valhalla-1.311
-1.292
 pose-1.280
Tokeniling
Token ID4386
Feature activation-1.41e+00

Pos logit contributions
ours+4.995
oured+4.946
aths+3.683
onel+3.503
ateral+3.423
Neg logit contributions
ンジ-10.976
-10.700
 Valhalla-10.689
 pose-10.291
-10.279
Token of
Token ID286
Feature activation-3.23e-01

Pos logit contributions
ours+3.537
oured+3.513
ateral+2.908
hap+2.859
aths+2.838
Neg logit contributions
ンジ-3.943
-3.803
 Valhalla-3.705
-3.639
 pose-3.631
Token an
Token ID281
Feature activation+4.34e-02

Pos logit contributions
ours+0.983
oured+0.981
ateral+0.840
hap+0.832
aths+0.826
Neg logit contributions
ンジ-0.737
-0.704
 Valhalla-0.679
-0.664
 pose-0.662
Token Islamic
Token ID5533
Feature activation+2.50e-01

Pos logit contributions
ンジ+0.102
+0.097
 Valhalla+0.094
 pose+0.093
+0.092
Neg logit contributions
ours-0.129
oured-0.129
ateral-0.110
hap-0.109
aths-0.108
Token Jihad
Token ID35243
Feature activation+4.21e-01

Pos logit contributions
ンジ+0.556
+0.533
 Valhalla+0.515
+0.506
 pose+0.504
Neg logit contributions
ours-0.772
oured-0.770
ateral-0.659
hap-0.654
aths-0.649
Token to
Token ID284
Feature activation-1.35e-01

Pos logit contributions
ンジ+0.525
+0.499
 Valhalla+0.486
 pose+0.477
+0.476
Neg logit contributions
ours-0.703
oured-0.701
ateral-0.599
hap-0.592
aths-0.589
Token Europe
Token ID2031
Feature activation-5.00e-01

Pos logit contributions
ours+0.427
oured+0.426
ateral+0.366
hap+0.362
aths+0.361
Neg logit contributions
ンジ-0.291
-0.275
 Valhalla-0.267
-0.262
 pose-0.262
Token in
Token ID287
Feature activation+8.06e-01

Pos logit contributions
ours+1.532
oured+1.524
ateral+1.305
hap+1.287
aths+1.276
Neg logit contributions
ンジ-1.140
-1.087
 Valhalla-1.041
-1.026
 pose-1.012
Token the
Token ID262
Feature activation-1.98e-01

Pos logit contributions
ンジ+1.703
+1.607
 Valhalla+1.601
 pose+1.578
+1.544
Neg logit contributions
ours-2.562
oured-2.545
ateral-2.206
hap-2.181
aths-2.168
Token first
Token ID717
Feature activation+3.13e-01

Pos logit contributions
ours+0.627
oured+0.625
ateral+0.536
hap+0.531
aths+0.527
Neg logit contributions
ンジ-0.433
-0.411
 Valhalla-0.395
-0.388
 pose-0.384
Token half
Token ID2063
Feature activation-3.09e+00

Pos logit contributions
ンジ+0.280
+0.244
 Valhalla+0.228
 pose+0.216
+0.214
Neg logit contributions
ours-1.389
oured-1.385
ateral-1.246
hap-1.238
aths-1.233
Token of
Token ID286
Feature activation+6.45e-02

Pos logit contributions
oured+6.074
ours+6.070
ateral+4.600
onel+4.402
aths+4.383
Neg logit contributions
ンジ-9.731
-9.685
-8.954
 Valhalla-8.883
 pose-8.597
Token 2010
Token ID3050
Feature activation-6.40e-01

Pos logit contributions
ンジ+0.148
+0.141
 Valhalla+0.137
 pose+0.134
+0.134
Neg logit contributions
ours-0.196
oured-0.196
ateral-0.167
hap-0.165
aths-0.164
Token,
Token ID11
Feature activation-5.29e-01

Pos logit contributions
ours+1.794
oured+1.789
ateral+1.503
hap+1.484
aths+1.471
Neg logit contributions
ンジ-1.622
-1.547
 Valhalla-1.507
-1.479
 pose-1.467
Token with
Token ID351
Feature activation-1.18e+00

Pos logit contributions
ours+1.600
oured+1.598
ateral+1.357
hap+1.340
aths+1.334
Neg logit contributions
ンジ-1.226
-1.167
 Valhalla-1.126
-1.106
 pose-1.096
Token the
Token ID262
Feature activation-1.06e+00

Pos logit contributions
oured+3.481
ours+3.479
ateral+2.933
hap+2.879
aths+2.879
Neg logit contributions
ンジ-2.787
-2.645
 Valhalla-2.540
-2.513
 pose-2.482
Token large
Token ID1588
Feature activation-7.73e-01

Pos logit contributions
oured+2.039
ours+2.035
ateral+1.706
hap+1.686
aths+1.682
Neg logit contributions
ンジ-1.881
-1.807
 Valhalla-1.741
-1.732
 pose-1.716
Token,
Token ID11
Feature activation-3.31e-01

Pos logit contributions
ours+2.106
oured+2.097
ateral+1.764
hap+1.736
aths+1.729
Neg logit contributions
ンジ-2.002
-1.917
 Valhalla-1.861
-1.838
 pose-1.830
Token running
Token ID2491
Feature activation-4.61e-01

Pos logit contributions
ours+1.012
oured+1.008
ateral+0.863
aths+0.852
hap+0.850
Neg logit contributions
ンジ-0.755
-0.719
 Valhalla-0.683
-0.678
 pose-0.667
Token a
Token ID257
Feature activation+1.73e-01

Pos logit contributions
ours+1.383
oured+1.374
ateral+1.175
aths+1.160
hap+1.158
Neg logit contributions
ンジ-1.077
-1.029
 Valhalla-0.984
-0.977
 pose-0.967
Token daily
Token ID4445
Feature activation-8.83e-01

Pos logit contributions
ンジ+0.383
+0.364
 Valhalla+0.353
+0.347
 pose+0.346
Neg logit contributions
ours-0.538
oured-0.536
ateral-0.460
hap-0.455
aths-0.453
Token cr
Token ID1067
Feature activation-3.06e+00

Pos logit contributions
ours+2.764
oured+2.758
hap+2.347
ateral+2.347
aths+2.338
Neg logit contributions
ンジ-1.943
-1.845
 Valhalla-1.726
-1.726
 pose-1.671
Tokenon
Token ID261
Feature activation-6.24e-01

Pos logit contributions
ours+5.560
oured+5.339
ateral+4.140
hap+4.010
onel+3.986
Neg logit contributions
ンジ-10.915
-10.376
 Valhalla-10.033
-9.833
 pose-9.518
Token job
Token ID1693
Feature activation+4.24e-01

Pos logit contributions
ours+1.814
oured+1.808
ateral+1.532
hap+1.513
aths+1.506
Neg logit contributions
ンジ-1.512
-1.457
 Valhalla-1.411
-1.379
 pose-1.349
Token would
Token ID561
Feature activation+1.02e-01

Pos logit contributions
ンジ+0.864
+0.815
 Valhalla+0.793
 pose+0.775
+0.773
Neg logit contributions
ours-1.399
oured-1.393
ateral-1.207
hap-1.194
aths-1.188
Token be
Token ID307
Feature activation+2.21e-01

Pos logit contributions
ンジ+0.200
+0.189
 Valhalla+0.181
+0.177
 pose+0.172
Neg logit contributions
ours-0.345
oured-0.344
ateral-0.301
hap-0.297
aths-0.295
Token comput
Token ID2653
Feature activation+8.29e-01

Pos logit contributions
ンジ+0.500
+0.475
 Valhalla+0.462
 pose+0.453
+0.453
Neg logit contributions
ours-0.678
oured-0.675
ateral-0.578
hap-0.572
aths-0.568
Token was
Token ID373
Feature activation-3.28e-01

Pos logit contributions
ンジ+1.515
+1.435
 Valhalla+1.404
 pose+1.389
+1.368
Neg logit contributions
ours-2.325
oured-2.314
ateral-2.000
hap-1.976
aths-1.965
Token prepared
Token ID5597
Feature activation+7.45e-01

Pos logit contributions
oured+0.989
ours+0.989
ateral+0.839
hap+0.835
aths+0.828
Neg logit contributions
ンジ-0.756
-0.725
 Valhalla-0.699
-0.686
 pose-0.681
Token soon
Token ID2582
Feature activation-1.04e-01

Pos logit contributions
ンジ+1.620
+1.528
 Valhalla+1.516
 pose+1.506
+1.469
Neg logit contributions
ours-2.319
oured-2.309
ateral-1.977
aths-1.955
yss-1.946
Token after
Token ID706
Feature activation-1.88e-01

Pos logit contributions
ours+0.309
oured+0.308
ateral+0.262
hap+0.259
aths+0.258
Neg logit contributions
ンジ-0.243
-0.231
 Valhalla-0.226
 pose-0.222
-0.221
Token N
Token ID399
Feature activation-6.45e-01

Pos logit contributions
ours+0.582
oured+0.580
ateral+0.497
hap+0.491
aths+0.489
Neg logit contributions
ンジ-0.420
-0.398
 Valhalla-0.390
 pose-0.383
-0.381
TokenDA
Token ID5631
Feature activation-3.03e+00

Pos logit contributions
oured+1.273
ours+1.266
ateral+0.974
hap+0.964
aths+0.960
Neg logit contributions
ンジ-2.192
-2.122
 Valhalla-2.072
 pose-2.041
-2.038
Token Government
Token ID5070
Feature activation-1.46e+00

Pos logit contributions
oured+7.858
ours+7.698
hap+6.500
aths+6.347
ateral+6.258
Neg logit contributions
ンジ-7.943
-7.569
 Valhalla-6.969
-6.966
terson-6.744
Token came
Token ID1625
Feature activation-3.82e-01

Pos logit contributions
oured+4.347
ours+4.346
ateral+3.636
hap+3.633
aths+3.617
Neg logit contributions
ンジ-3.434
-3.248
 Valhalla-3.093
-3.071
 pose-2.978
Token to
Token ID284
Feature activation-3.04e-01

Pos logit contributions
ours+1.135
oured+1.130
ateral+0.963
hap+0.951
aths+0.948
Neg logit contributions
ンジ-0.901
-0.859
 Valhalla-0.837
 pose-0.818
-0.817
Token power
Token ID1176
Feature activation-6.35e-01

Pos logit contributions
ours+0.928
oured+0.924
ateral+0.789
hap+0.782
aths+0.777
Neg logit contributions
ンジ-0.695
-0.659
 Valhalla-0.641
 pose-0.629
-0.628
Token.
Token ID13
Feature activation-2.78e-01

Pos logit contributions
ours+1.784
oured+1.777
ateral+1.495
hap+1.481
aths+1.469
Neg logit contributions
ンジ-1.604
-1.531
 Valhalla-1.492
-1.465
 pose-1.460
Token paid
Token ID3432
Feature activation+5.94e-01

Pos logit contributions
ours+3.157
oured+3.148
ateral+2.668
hap+2.649
aths+2.628
Neg logit contributions
ンジ-2.356
-2.254
-2.066
 Valhalla-2.066
 pose-2.021
Token for
Token ID329
Feature activation+4.51e-01

Pos logit contributions
ンジ+1.275
+1.214
 Valhalla+1.167
+1.150
 pose+1.147
Neg logit contributions
ours-1.902
oured-1.894
ateral-1.625
hap-1.614
aths-1.606
Token all
Token ID477
Feature activation-5.32e-01

Pos logit contributions
ンジ+0.996
+0.945
 Valhalla+0.921
 pose+0.905
+0.901
Neg logit contributions
ours-1.406
oured-1.401
ateral-1.201
hap-1.189
aths-1.183
Token those
Token ID883
Feature activation+2.07e-01

Pos logit contributions
ours+1.548
oured+1.540
ateral+1.309
hap+1.287
aths+1.282
Neg logit contributions
ンジ-1.282
-1.232
 Valhalla-1.180
-1.173
 pose-1.171
Token programmers
Token ID24867
Feature activation+6.77e-01

Pos logit contributions
ンジ+0.441
+0.417
 Valhalla+0.402
+0.396
 pose+0.396
Neg logit contributions
ours-0.666
oured-0.664
ateral-0.571
hap-0.565
aths-0.563
Token cr
Token ID1067
Feature activation-3.02e+00

Pos logit contributions
ンジ+1.597
+1.517
 Valhalla+1.494
 pose+1.470
+1.457
Neg logit contributions
ours-2.004
oured-1.997
ateral-1.702
hap-1.681
aths-1.671
Tokenanking
Token ID15230
Feature activation-1.01e+00

Pos logit contributions
ours+6.641
oured+6.535
ateral+5.123
aths+5.099
hap+4.909
Neg logit contributions
ンジ-9.872
-9.168
 Valhalla-8.733
-8.660
 pose-8.573
Token out
Token ID503
Feature activation+8.78e-03

Pos logit contributions
ours+2.894
oured+2.883
hap+2.419
ateral+2.405
aths+2.396
Neg logit contributions
ンジ-2.435
-2.351
-2.245
 Valhalla-2.215
 pose-2.198
Token the
Token ID262
Feature activation+1.69e-01

Pos logit contributions
ンジ+0.019
+0.018
 Valhalla+0.018
+0.017
 pose+0.017
Neg logit contributions
ours-0.027
oured-0.027
ateral-0.023
hap-0.023
aths-0.023
Token overhaul
Token ID18708
Feature activation-2.59e-03

Pos logit contributions
ンジ+0.368
+0.350
 Valhalla+0.338
 pose+0.332
+0.332
Neg logit contributions
ours-0.535
oured-0.533
ateral-0.458
hap-0.453
aths-0.451
Tokened
Token ID276
Feature activation-9.13e-01

Pos logit contributions
ours+0.007
oured+0.007
ateral+0.006
hap+0.006
aths+0.006
Neg logit contributions
ンジ-0.007
-0.007
 Valhalla-0.007
 pose-0.006
-0.006
Token same
Token ID976
Feature activation+8.87e-01

Pos logit contributions
ours+0.906
oured+0.903
ateral+0.777
hap+0.769
aths+0.765
Neg logit contributions
ンジ-0.598
-0.563
 Valhalla-0.543
-0.534
 pose-0.527
Token measure
Token ID3953
Feature activation+4.19e-01

Pos logit contributions
ンジ+1.855
+1.758
 Valhalla+1.723
 pose+1.705
+1.680
Neg logit contributions
ours-2.847
oured-2.837
ateral-2.452
hap-2.425
aths-2.411
Token is
Token ID318
Feature activation-2.60e-01

Pos logit contributions
ンジ+0.907
+0.861
 Valhalla+0.839
 pose+0.826
+0.821
Neg logit contributions
ours-1.321
oured-1.317
ateral-1.133
hap-1.122
aths-1.115
Token repeated
Token ID5100
Feature activation+2.80e-01

Pos logit contributions
ours+0.763
oured+0.759
ateral+0.641
hap+0.637
aths+0.631
Neg logit contributions
ンジ-0.626
-0.594
 Valhalla-0.580
-0.568
 pose-0.565
Token;
Token ID26
Feature activation-1.90e-02

Pos logit contributions
ンジ+0.643
+0.611
 Valhalla+0.598
 pose+0.589
+0.585
Neg logit contributions
ours-0.846
oured-0.843
ateral-0.719
hap-0.712
aths-0.708
Token etc
Token ID3503
Feature activation-2.99e+00

Pos logit contributions
ours+0.054
oured+0.054
ateral+0.046
hap+0.045
aths+0.045
Neg logit contributions
ンジ-0.047
-0.045
 Valhalla-0.044
 pose-0.043
-0.043
Token.
Token ID13
Feature activation-2.90e-01

Pos logit contributions
ours+6.612
oured+6.522
ateral+5.299
aths+5.084
hap+5.079
Neg logit contributions
ンジ-8.880
-8.443
 Valhalla-8.262
-8.204
 pose-8.047
Token So
Token ID1406
Feature activation-8.82e-01

Pos logit contributions
ours+0.846
oured+0.842
ateral+0.715
hap+0.705
aths+0.702
Neg logit contributions
ンジ-0.700
-0.665
 Valhalla-0.651
 pose-0.639
-0.637
Token this
Token ID428
Feature activation+2.85e-01

Pos logit contributions
ours+2.690
oured+2.675
ateral+2.291
aths+2.270
hap+2.264
Neg logit contributions
ンジ-2.008
-1.899
 Valhalla-1.847
-1.805
 pose-1.780
Token
Token ID564
Feature activation+1.59e-01

Pos logit contributions
ンジ+0.604
+0.572
 Valhalla+0.555
+0.543
 pose+0.541
Neg logit contributions
ours-0.918
oured-0.915
ateral-0.789
hap-0.781
aths-0.776
Token
Token ID246
Feature activation+6.48e-01

Pos logit contributions
ンジ+0.372
+0.354
 Valhalla+0.346
 pose+0.341
+0.339
Neg logit contributions
ours-0.474
oured-0.472
ateral-0.402
hap-0.398
aths-0.395
Token 41
Token ID6073
Feature activation-6.47e-01

Pos logit contributions
ours+0.011
oured+0.011
ateral+0.010
hap+0.010
aths+0.009
Neg logit contributions
ンジ-0.009
-0.008
 Valhalla-0.008
-0.008
 pose-0.008
Token-
Token ID12
Feature activation+4.22e-01

Pos logit contributions
ours+1.803
oured+1.789
hap+1.493
ateral+1.488
aths+1.466
Neg logit contributions
ンジ-1.639
-1.590
 Valhalla-1.539
-1.510
 pose-1.501
Tokenyear
Token ID1941
Feature activation+2.91e+00

Pos logit contributions
ンジ+0.822
+0.783
 Valhalla+0.758
 pose+0.735
+0.732
Neg logit contributions
ours-1.440
oured-1.429
ateral-1.239
hap-1.236
aths-1.225
Token-
Token ID12
Feature activation+1.15e-01

Pos logit contributions
ンジ+5.723
 Valhalla+5.485
 pose+5.483
 dream+5.305
+5.233
Neg logit contributions
ours-9.178
oured-9.105
ateral-8.018
yss-7.916
hap-7.899
Tokenold
Token ID727
Feature activation+1.48e+00

Pos logit contributions
ンジ+0.344
+0.333
 Valhalla+0.325
 pose+0.321
+0.319
Neg logit contributions
ours-0.274
oured-0.271
ateral-0.218
hap-0.216
aths-0.215
Token cr
Token ID1067
Feature activation-2.98e+00

Pos logit contributions
ンジ+3.365
 Valhalla+3.161
+3.144
 pose+3.127
+3.040
Neg logit contributions
ours-4.404
oured-4.383
ateral-3.813
hap-3.705
yss-3.699
Tokenick
Token ID624
Feature activation-2.00e-02

Pos logit contributions
ours+6.140
oured+6.037
aths+4.636
hap+4.612
ateral+4.612
Neg logit contributions
ンジ-9.746
-9.184
 Valhalla-9.095
-8.669
 pose-8.574
Tokeneter
Token ID2357
Feature activation-4.69e-02

Pos logit contributions
ours+0.058
oured+0.058
ateral+0.049
hap+0.048
aths+0.048
Neg logit contributions
ンジ-0.049
-0.047
 Valhalla-0.046
-0.045
 pose-0.045
Token added
Token ID2087
Feature activation-2.55e-02

Pos logit contributions
ours+0.134
oured+0.133
ateral+0.112
hap+0.111
aths+0.110
Neg logit contributions
ンジ-0.116
-0.111
 Valhalla-0.108
-0.106
 pose-0.106
Token.
Token ID13
Feature activation+1.73e-01

Pos logit contributions
ours+0.066
oured+0.065
ateral+0.054
hap+0.053
aths+0.053
Neg logit contributions
ンジ-0.070
-0.067
 Valhalla-0.066
 pose-0.065
-0.065
Token
Token ID198
Feature activation+5.61e-01

Pos logit contributions
ンジ+0.375
+0.354
 Valhalla+0.347
 pose+0.341
+0.338
Neg logit contributions
ours-0.547
oured-0.544
ateral-0.468
hap-0.463
aths-0.460
Token about
Token ID546
Feature activation+1.00e+00

Pos logit contributions
ンジ+0.349
+0.331
 Valhalla+0.324
 pose+0.319
+0.317
Neg logit contributions
ours-0.462
oured-0.461
ateral-0.394
hap-0.389
aths-0.387
Token:
Token ID25
Feature activation+1.22e+00

Pos logit contributions
ンジ+2.216
+2.112
 Valhalla+2.097
 pose+2.062
+2.026
Neg logit contributions
ours-3.059
oured-3.050
ateral-2.597
hap-2.595
aths-2.570
Token
Token ID198
Feature activation+5.71e-01

Pos logit contributions
ンジ+2.596
+2.480
 Valhalla+2.446
 pose+2.411
+2.389
Neg logit contributions
ours-3.791
oured-3.787
ateral-3.263
hap-3.225
aths-3.193
Token
Token ID198
Feature activation+2.57e-01

Pos logit contributions
ンジ+1.354
+1.292
 Valhalla+1.250
 pose+1.232
+1.222
Neg logit contributions
ours-1.704
oured-1.691
ateral-1.440
hap-1.424
aths-1.417
TokenL
Token ID43
Feature activation-8.64e-01

Pos logit contributions
ンジ+0.696
+0.666
 Valhalla+0.653
 pose+0.645
+0.644
Neg logit contributions
ours-0.672
oured-0.670
ateral-0.557
hap-0.549
aths-0.545
Tokenoss
Token ID793
Feature activation-2.97e+00

Pos logit contributions
ours+2.069
oured+2.062
ateral+1.710
aths+1.651
hap+1.646
Neg logit contributions
ンジ-2.562
-2.454
 Valhalla-2.393
-2.361
 pose-2.356
Token of
Token ID286
Feature activation-7.30e-01

Pos logit contributions
ours+7.021
oured+6.893
ateral+5.758
hap+5.650
aths+5.593
Neg logit contributions
ンジ-8.512
-8.196
 Valhalla-7.705
-7.634
 pose-7.535
Token or
Token ID393
Feature activation-7.07e-02

Pos logit contributions
ours+2.246
oured+2.235
ateral+1.927
hap+1.903
aths+1.894
Neg logit contributions
ンジ-1.649
-1.567
 Valhalla-1.488
-1.474
 pose-1.462
Token curse
Token ID17328
Feature activation+7.59e-02

Pos logit contributions
ours+0.213
oured+0.212
ateral+0.181
hap+0.179
aths+0.178
Neg logit contributions
ンジ-0.164
-0.155
 Valhalla-0.152
 pose-0.149
-0.149
Token on
Token ID319
Feature activation-1.66e-01

Pos logit contributions
ンジ+0.182
+0.173
 Valhalla+0.170
 pose+0.168
+0.167
Neg logit contributions
ours-0.221
oured-0.221
ateral-0.187
hap-0.185
aths-0.184
Token magical
Token ID10883
Feature activation+3.92e-01

Pos logit contributions
ours+0.513
oured+0.512
ateral+0.438
hap+0.434
aths+0.432
Neg logit contributions
ンジ-0.372
-0.352
 Valhalla-0.342
 pose-0.336
-0.336
Token "
Token ID366
Feature activation+1.01e-01

Pos logit contributions
ンジ+1.360
+1.287
 Valhalla+1.265
 pose+1.250
+1.232
Neg logit contributions
ours-2.047
oured-2.040
ateral-1.756
hap-1.738
aths-1.729
Tokenfab
Token ID36434
Feature activation+8.56e-01

Pos logit contributions
ンジ+0.292
+0.280
 Valhalla+0.275
 pose+0.273
+0.271
Neg logit contributions
ours-0.245
oured-0.245
ateral-0.200
hap-0.197
aths-0.196
Tokenric
Token ID1173
Feature activation-1.62e-01

Pos logit contributions
ンジ+1.472
+1.368
 Valhalla+1.319
 pose+1.310
+1.296
Neg logit contributions
ours-3.047
oured-3.046
ateral-2.687
hap-2.675
aths-2.656
Tokenation
Token ID341
Feature activation-1.05e+00

Pos logit contributions
ours+0.418
oured+0.415
ateral+0.344
hap+0.338
aths+0.337
Neg logit contributions
ンジ-0.445
-0.427
 Valhalla-0.419
 pose-0.412
-0.410
Token,
Token ID11
Feature activation-2.35e-01

Pos logit contributions
ours+2.731
oured+2.710
ateral+2.266
hap+2.226
aths+2.196
Neg logit contributions
ンジ-2.854
-2.754
 Valhalla-2.661
-2.618
 pose-2.595
Token fals
Token ID27807
Feature activation-2.95e+00

Pos logit contributions
ours+0.713
oured+0.709
ateral+0.603
hap+0.600
aths+0.594
Neg logit contributions
ンジ-0.541
-0.515
 Valhalla-0.497
-0.487
 pose-0.475
Tokenification
Token ID2649
Feature activation-2.18e+00

Pos logit contributions
ours+8.190
oured+8.131
yss+6.839
ateral+6.799
aths+6.718
Neg logit contributions
ンジ-7.368
-6.882
 Valhalla-6.613
-6.318
 pose-6.174
Token or
Token ID393
Feature activation-6.40e-01

Pos logit contributions
ours+5.467
oured+5.399
ateral+4.531
hap+4.458
yss+4.401
Neg logit contributions
ンジ-5.981
-5.793
 Valhalla-5.567
-5.466
 pose-5.401
Token plagiar
Token ID42054
Feature activation-3.62e-01

Pos logit contributions
ours+1.972
oured+1.962
ateral+1.674
hap+1.658
aths+1.642
Neg logit contributions
ンジ-1.439
-1.373
 Valhalla-1.321
-1.303
 pose-1.250
Tokenism
Token ID1042
Feature activation-9.84e-01

Pos logit contributions
ours+1.150
oured+1.146
ateral+0.990
aths+0.973
hap+0.972
Neg logit contributions
ンジ-0.785
-0.746
 Valhalla-0.720
-0.701
 pose-0.701
Token"
Token ID1
Feature activation+2.14e-01

Pos logit contributions
ours+2.620
oured+2.588
ateral+2.182
hap+2.133
aths+2.115
Neg logit contributions
ンジ-2.644
-2.540
 Valhalla-2.441
-2.401
 pose-2.353
Token pay
Token ID1414
Feature activation+7.00e-01

Pos logit contributions
ンジ+0.220
+0.206
 Valhalla+0.199
+0.193
 pose+0.193
Neg logit contributions
ours-0.461
oured-0.460
ateral-0.403
hap-0.400
aths-0.398
Token all
Token ID477
Feature activation-3.26e-01

Pos logit contributions
ンジ+1.450
+1.364
 Valhalla+1.336
 pose+1.314
+1.303
Neg logit contributions
ours-2.273
oured-2.264
ateral-1.960
hap-1.938
aths-1.929
Token its
Token ID663
Feature activation+5.07e-03

Pos logit contributions
ours+0.976
oured+0.972
ateral+0.829
hap+0.819
aths+0.814
Neg logit contributions
ンジ-0.758
-0.722
 Valhalla-0.705
 pose-0.693
-0.690
Token bills
Token ID9024
Feature activation+9.46e-01

Pos logit contributions
ンジ+0.009
+0.009
 Valhalla+0.008
 pose+0.008
+0.008
Neg logit contributions
ours-0.018
oured-0.018
ateral-0.015
hap-0.015
aths-0.015
Token on
Token ID319
Feature activation-2.41e-01

Pos logit contributions
ンジ+2.145
+2.030
 Valhalla+2.017
 pose+2.005
+1.967
Neg logit contributions
oured-2.843
ours-2.840
ateral-2.421
hap-2.402
aths-2.394
Token Oct
Token ID2556
Feature activation-2.93e+00

Pos logit contributions
ours+0.824
oured+0.822
ateral+0.714
hap+0.709
aths+0.704
Neg logit contributions
ンジ-0.465
-0.439
 Valhalla-0.419
-0.410
 pose-0.409
Token.
Token ID13
Feature activation-3.93e-01

Pos logit contributions
ours+5.886
oured+5.744
ateral+4.367
aths+4.364
hap+4.239
Neg logit contributions
ンジ-9.593
-9.211
-8.724
 pose-8.677
 Valhalla-8.636
Token 2
Token ID362
Feature activation-2.25e-01

Pos logit contributions
ours+1.064
oured+1.061
ateral+0.878
aths+0.870
hap+0.869
Neg logit contributions
ンジ-1.036
-0.991
 Valhalla-0.956
 pose-0.949
-0.944
Token if
Token ID611
Feature activation-3.69e-01

Pos logit contributions
ours+0.621
oured+0.618
ateral+0.519
hap+0.513
aths+0.510
Neg logit contributions
ンジ-0.579
-0.553
 Valhalla-0.542
 pose-0.534
-0.531
Token Washington
Token ID2669
Feature activation-5.21e-01

Pos logit contributions
ours+1.135
oured+1.133
ateral+0.968
hap+0.959
aths+0.954
Neg logit contributions
ンジ-0.834
-0.794
 Valhalla-0.767
-0.752
 pose-0.747
Token does
Token ID857
Feature activation-6.34e-01

Pos logit contributions
ours+1.700
oured+1.696
ateral+1.457
aths+1.450
hap+1.437
Neg logit contributions
ンジ-1.081
-1.017
 Valhalla-0.993
-0.963
 pose-0.959
Token have
Token ID423
Feature activation-3.78e-01

Pos logit contributions
ンジ+0.814
+0.768
 Valhalla+0.749
 pose+0.734
+0.730
Neg logit contributions
ours-1.303
oured-1.298
ateral-1.123
hap-1.112
aths-1.106
Token changed
Token ID3421
Feature activation+5.52e-01

Pos logit contributions
ours+0.976
oured+0.972
ateral+0.804
hap+0.794
aths+0.788
Neg logit contributions
ンジ-1.040
-0.997
 Valhalla-0.977
 pose-0.962
-0.960
Token.
Token ID13
Feature activation-3.21e-01

Pos logit contributions
ンジ+1.243
+1.176
 pose+1.163
 Valhalla+1.162
+1.133
Neg logit contributions
ours-1.676
oured-1.673
ateral-1.424
hap-1.417
aths-1.409
Token Ice
Token ID6663
Feature activation+1.71e-01

Pos logit contributions
ours+0.931
oured+0.925
ateral+0.786
hap+0.773
aths+0.771
Neg logit contributions
ンジ-0.778
-0.739
 Valhalla-0.722
 pose-0.715
-0.710
Token has
Token ID468
Feature activation-1.85e-02

Pos logit contributions
ンジ+0.381
+0.361
 Valhalla+0.353
 pose+0.347
+0.345
Neg logit contributions
ours-0.528
oured-0.526
ateral-0.451
hap-0.446
aths-0.444
Token cr
Token ID1067
Feature activation-2.92e+00

Pos logit contributions
ours+0.050
oured+0.050
ateral+0.042
hap+0.041
aths+0.041
Neg logit contributions
ンジ-0.048
-0.046
 Valhalla-0.045
 pose-0.045
-0.044
Tokenumbled
Token ID11137
Feature activation-5.50e-01

Pos logit contributions
oured+5.943
ours+5.932
ateral+4.529
aths+4.454
onel+4.089
Neg logit contributions
ンジ-10.056
-9.360
-8.991
 Valhalla-8.917
 pose-8.785
Token off
Token ID572
Feature activation-6.31e-01

Pos logit contributions
ours+1.638
oured+1.627
ateral+1.390
hap+1.372
aths+1.365
Neg logit contributions
ンジ-1.297
-1.236
 Valhalla-1.199
-1.180
 pose-1.174
Token of
Token ID286
Feature activation-1.06e-01

Pos logit contributions
ours+1.891
oured+1.881
ateral+1.610
hap+1.586
aths+1.578
Neg logit contributions
ンジ-1.473
-1.405
 Valhalla-1.357
-1.339
 pose-1.332
Token the
Token ID262
Feature activation-3.50e-01

Pos logit contributions
ours+0.322
oured+0.321
ateral+0.275
hap+0.272
aths+0.270
Neg logit contributions
ンジ-0.243
-0.231
 Valhalla-0.225
 pose-0.221
-0.220
Token glacier
Token ID44539
Feature activation-4.79e-01

Pos logit contributions
ours+1.128
oured+1.124
ateral+0.971
hap+0.960
aths+0.958
Neg logit contributions
ンジ-0.737
-0.698
 Valhalla-0.670
-0.662
 pose-0.652
Token type
Token ID2099
Feature activation-3.08e-01

Pos logit contributions
ours+4.524
oured+4.487
ateral+3.823
aths+3.742
hap+3.741
Neg logit contributions
ンジ-3.679
-3.531
 Valhalla-3.405
-3.391
 pose-3.343
Token-
Token ID12
Feature activation-1.35e+00

Pos logit contributions
ours+0.913
oured+0.909
ateral+0.774
hap+0.766
aths+0.762
Neg logit contributions
ンジ-0.726
-0.690
 Valhalla-0.676
 pose-0.665
-0.661
Token94
Token ID5824
Feature activation-1.07e+00

Pos logit contributions
ours+3.239
oured+3.187
ateral+2.610
hap+2.561
aths+2.544
Neg logit contributions
ンジ-3.890
-3.786
 Valhalla-3.650
 pose-3.642
-3.562
Token Jin
Token ID17297
Feature activation-6.94e-01

Pos logit contributions
ours+3.237
oured+3.218
ateral+2.738
hap+2.719
aths+2.695
Neg logit contributions
ンジ-2.458
-2.354
 Valhalla-2.248
-2.243
 pose-2.230
Token-
Token ID12
Feature activation+3.68e-01

Pos logit contributions
ours+1.946
oured+1.937
hap+1.622
ateral+1.616
aths+1.592
Neg logit contributions
ンジ-1.740
-1.680
 Valhalla-1.639
-1.608
 pose-1.600
Tokenclass
Token ID4871
Feature activation+2.39e+00

Pos logit contributions
ンジ+0.550
+0.500
 Valhalla+0.485
 pose+0.474
+0.471
Neg logit contributions
ours-1.406
oured-1.405
ateral-1.244
hap-1.228
aths-1.227
Token and
Token ID290
Feature activation-1.84e-01

Pos logit contributions
ンジ+5.389
 pose+5.038
 Valhalla+5.001
+4.937
+4.785
Neg logit contributions
oured-6.914
ours-6.891
ateral-5.910
hap-5.896
aths-5.867
Token type
Token ID2099
Feature activation+9.80e-01

Pos logit contributions
ours+0.513
oured+0.511
ateral+0.430
hap+0.425
aths+0.422
Neg logit contributions
ンジ-0.466
-0.446
 Valhalla-0.435
 pose-0.428
-0.428
Token-
Token ID12
Feature activation-6.54e-01

Pos logit contributions
ンジ+2.323
+2.210
 Valhalla+2.164
 pose+2.129
+2.119
Neg logit contributions
ours-2.894
oured-2.883
ateral-2.451
hap-2.425
aths-2.410
Token96
Token ID4846
Feature activation-9.34e-01

Pos logit contributions
ours+1.768
oured+1.752
ateral+1.470
hap+1.450
aths+1.443
Neg logit contributions
ンジ-1.712
-1.650
 Valhalla-1.605
 pose-1.585
-1.567
Token),
Token ID828
Feature activation-3.93e-01

Pos logit contributions
ours+2.790
oured+2.780
ateral+2.367
hap+2.349
aths+2.323
Neg logit contributions
ンジ-2.165
-2.071
 Valhalla-1.991
-1.977
 pose-1.965
Token to
Token ID284
Feature activation-4.42e-01

Pos logit contributions
ours+0.446
oured+0.442
ateral+0.375
hap+0.369
aths+0.364
Neg logit contributions
ンジ-0.410
-0.398
 Valhalla-0.383
-0.374
 pose-0.368
Token exposure
Token ID7111
Feature activation+3.32e-01

Pos logit contributions
ours+1.375
oured+1.370
ateral+1.171
hap+1.161
aths+1.150
Neg logit contributions
ンジ-0.985
-0.940
 Valhalla-0.910
-0.889
 pose-0.873
Token to
Token ID284
Feature activation+4.50e-01

Pos logit contributions
ンジ+0.760
+0.717
 Valhalla+0.704
 pose+0.704
+0.694
Neg logit contributions
oured-1.003
ours-0.998
ateral-0.853
hap-0.844
yss-0.843
Token air
Token ID1633
Feature activation+3.80e-01

Pos logit contributions
ンジ+0.952
+0.897
 Valhalla+0.880
 pose+0.872
+0.858
Neg logit contributions
ours-1.443
oured-1.438
ateral-1.241
hap-1.227
aths-1.219
Token and
Token ID290
Feature activation+8.20e-01

Pos logit contributions
ンジ+0.729
+0.685
 Valhalla+0.667
 pose+0.655
+0.650
Neg logit contributions
ours-1.293
oured-1.289
ateral-1.122
hap-1.111
aths-1.106
Token water
Token ID1660
Feature activation+2.04e+00

Pos logit contributions
ンジ+1.422
+1.319
 Valhalla+1.291
 pose+1.287
+1.261
Neg logit contributions
ours-2.925
oured-2.911
ateral-2.563
hap-2.535
aths-2.515
Token pollution
Token ID12231
Feature activation-2.17e-02

Pos logit contributions
ンジ+4.483
 pose+4.291
+4.276
 Valhalla+4.142
+4.089
Neg logit contributions
oured-6.099
ours-6.097
ateral-5.298
yss-5.231
hap-5.207
Token,
Token ID11
Feature activation+3.85e-01

Pos logit contributions
ours+0.061
oured+0.061
ateral+0.051
hap+0.051
aths+0.050
Neg logit contributions
ンジ-0.055
-0.052
 Valhalla-0.051
 pose-0.050
-0.050
Token according
Token ID1864
Feature activation-5.76e-01

Pos logit contributions
ンジ+0.836
+0.791
 Valhalla+0.773
 pose+0.761
+0.756
Neg logit contributions
ours-1.214
oured-1.209
ateral-1.040
hap-1.029
aths-1.023
Token to
Token ID284
Feature activation-3.98e-01

Pos logit contributions
ours+1.551
oured+1.527
ateral+1.288
hap+1.279
aths+1.272
Neg logit contributions
ンジ-1.509
-1.473
 Valhalla-1.400
-1.379
 pose-1.369
Token a
Token ID257
Feature activation+2.82e-01

Pos logit contributions
ours+1.180
oured+1.175
ateral+1.000
hap+0.994
aths+0.982
Neg logit contributions
ンジ-0.942
-0.904
 Valhalla-0.874
-0.855
 pose-0.853
Token Peter
Token ID5613
Feature activation+1.48e-01

Pos logit contributions
ours+0.277
oured+0.276
ateral+0.237
hap+0.235
aths+0.234
Neg logit contributions
ンジ-0.188
-0.179
 Valhalla-0.172
-0.169
 pose-0.169
Token Lur
Token ID47851
Feature activation+1.36e-01

Pos logit contributions
ンジ+0.310
+0.291
 Valhalla+0.282
 pose+0.279
+0.278
Neg logit contributions
ours-0.483
oured-0.480
ateral-0.413
hap-0.410
aths-0.408
Tokenie
Token ID494
Feature activation+2.62e-01

Pos logit contributions
ンジ+0.332
+0.316
 Valhalla+0.309
 pose+0.303
+0.303
Neg logit contributions
ours-0.397
oured-0.395
ateral-0.334
hap-0.331
aths-0.328
Token and
Token ID290
Feature activation-2.27e-02

Pos logit contributions
ンジ+0.621
+0.591
 Valhalla+0.578
 pose+0.569
+0.566
Neg logit contributions
ours-0.772
oured-0.770
ateral-0.654
hap-0.647
aths-0.643
Token Sidney
Token ID35727
Feature activation+9.18e-01

Pos logit contributions
ours+0.076
oured+0.076
ateral+0.066
hap+0.065
aths+0.065
Neg logit contributions
ンジ-0.045
-0.043
 Valhalla-0.041
-0.040
 pose-0.040
Token Wolfe
Token ID37013
Feature activation+1.93e+00

Pos logit contributions
ンジ+1.848
+1.756
 Valhalla+1.724
 pose+1.680
+1.667
Neg logit contributions
ours-3.013
oured-3.008
ateral-2.609
hap-2.575
aths-2.561
Token accuse
Token ID24456
Feature activation-3.09e-01

Pos logit contributions
ンジ+3.941
 pose+3.850
 Valhalla+3.748
+3.746
+3.621
Neg logit contributions
oured-6.123
ours-6.116
ateral-5.281
hap-5.201
yss-5.128
Token the
Token ID262
Feature activation+2.87e-01

Pos logit contributions
ours+0.977
oured+0.973
ateral+0.834
hap+0.828
aths+0.824
Neg logit contributions
ンジ-0.672
-0.641
 Valhalla-0.615
-0.605
 pose-0.602
Token NIH
Token ID37483
Feature activation+1.75e-01

Pos logit contributions
ンジ+0.682
+0.649
 Valhalla+0.634
 pose+0.624
+0.622
Neg logit contributions
ours-0.849
oured-0.845
ateral-0.719
hap-0.711
aths-0.707
Token,
Token ID11
Feature activation+5.17e-01

Pos logit contributions
ンジ+0.462
+0.440
 Valhalla+0.434
 pose+0.430
+0.424
Neg logit contributions
ours-0.471
oured-0.469
ateral-0.392
hap-0.388
aths-0.384
Token WHO
Token ID19494
Feature activation-6.19e-01

Pos logit contributions
ンジ+1.162
+1.093
 Valhalla+1.086
 pose+1.068
+1.054
Neg logit contributions
ours-1.586
oured-1.581
ateral-1.357
hap-1.339
aths-1.334
Token
Token ID198
Feature activation+5.89e-01

Pos logit contributions
ンジ+0.984
+0.936
 Valhalla+0.912
 pose+0.902
+0.896
Neg logit contributions
ours-1.256
oured-1.249
ateral-1.066
hap-1.054
aths-1.048
Token
Token ID198
Feature activation+4.15e-01

Pos logit contributions
ンジ+1.386
+1.324
 Valhalla+1.284
 pose+1.266
+1.253
Neg logit contributions
ours-1.758
oured-1.746
ateral-1.487
hap-1.471
aths-1.463
TokenA
Token ID32
Feature activation+3.05e-01

Pos logit contributions
ンジ+1.072
+1.024
 Valhalla+1.004
 pose+0.989
+0.986
Neg logit contributions
ours-1.136
oured-1.132
ateral-0.949
hap-0.937
aths-0.931
Token three
Token ID1115
Feature activation+2.99e-01

Pos logit contributions
ンジ+0.666
+0.632
 Valhalla+0.615
+0.604
 pose+0.601
Neg logit contributions
ours-0.963
oured-0.959
ateral-0.823
hap-0.817
aths-0.811
Token-
Token ID12
Feature activation+4.86e-01

Pos logit contributions
ンジ+0.667
+0.633
 Valhalla+0.619
 pose+0.608
+0.605
Neg logit contributions
ours-0.924
oured-0.920
ateral-0.789
hap-0.780
aths-0.776
Tokenyear
Token ID1941
Feature activation+2.07e+00

Pos logit contributions
ンジ+0.918
+0.842
 Valhalla+0.827
 pose+0.819
+0.809
Neg logit contributions
oured-1.664
ours-1.663
ateral-1.444
hap-1.432
aths-1.430
Token deal
Token ID1730
Feature activation+5.16e-01

Pos logit contributions
ンジ+4.142
 Valhalla+3.835
+3.833
 pose+3.819
 dream+3.675
Neg logit contributions
ours-6.735
oured-6.665
ateral-5.874
hap-5.738
yss-5.684
Token at
Token ID379
Feature activation+7.18e-02

Pos logit contributions
ンジ+1.062
+1.005
 Valhalla+0.974
+0.954
 pose+0.949
Neg logit contributions
ours-1.697
oured-1.689
ateral-1.460
hap-1.447
aths-1.439
Token $
Token ID720
Feature activation+1.72e-01

Pos logit contributions
ンジ+0.152
+0.144
 Valhalla+0.139
+0.137
 pose+0.136
Neg logit contributions
ours-0.231
oured-0.230
ateral-0.199
hap-0.197
aths-0.195
Token11
Token ID1157
Feature activation+3.92e-01

Pos logit contributions
ンジ+0.522
+0.504
 Valhalla+0.494
 pose+0.489
+0.484
Neg logit contributions
ours-0.396
oured-0.392
ateral-0.316
hap-0.312
aths-0.309
Token.
Token ID13
Feature activation-3.20e-01

Pos logit contributions
ンジ+0.986
+0.931
 Valhalla+0.918
 pose+0.908
+0.895
Neg logit contributions
ours-1.103
oured-1.095
ateral-0.930
hap-0.915
aths-0.911
Tokenific
Token ID811
Feature activation-8.30e-02

Pos logit contributions
ンジ+2.482
+2.356
 Valhalla+2.295
+2.244
 pose+2.239
Neg logit contributions
ours-3.369
oured-3.351
ateral-2.875
hap-2.836
aths-2.826
Token...
Token ID986
Feature activation+2.18e-01

Pos logit contributions
ours+0.245
oured+0.244
ateral+0.207
hap+0.205
aths+0.204
Neg logit contributions
ンジ-0.197
-0.188
 Valhalla-0.184
 pose-0.181
-0.180
Token Flag
Token ID19762
Feature activation+1.72e+00

Pos logit contributions
ンジ+0.528
+0.502
 Valhalla+0.492
 pose+0.485
+0.482
Neg logit contributions
ours-0.633
oured-0.631
ateral-0.535
hap-0.529
aths-0.526
Token
Token ID198
Feature activation+2.79e-01

Pos logit contributions
ンジ+3.530
 pose+3.420
 Valhalla+3.304
+3.265
+3.219
Neg logit contributions
ours-5.401
oured-5.352
ateral-4.712
aths-4.676
hap-4.662
Token
Token ID198
Feature activation+1.47e-01

Pos logit contributions
ンジ+0.684
+0.659
 Valhalla+0.632
 pose+0.627
+0.614
Neg logit contributions
ours-0.813
oured-0.802
ateral-0.680
hap-0.676
aths-0.670
TokenFlag
Token ID34227
Feature activation+1.95e+00

Pos logit contributions
ンジ+0.321
+0.307
 Valhalla+0.298
 pose+0.293
+0.290
Neg logit contributions
ours-0.462
oured-0.459
ateral-0.394
hap-0.392
aths-0.388
Token Reply
Token ID14883
Feature activation+1.79e-01

Pos logit contributions
ンジ+5.078
 Valhalla+4.922
 pose+4.875
+4.806
+4.684
Neg logit contributions
ours-5.107
oured-5.078
ateral-4.291
aths-4.224
hap-4.210
Token
Token ID198
Feature activation-4.75e-02

Pos logit contributions
ンジ+0.419
+0.398
 Valhalla+0.387
+0.380
 pose+0.380
Neg logit contributions
ours-0.533
oured-0.530
ateral-0.454
hap-0.449
aths-0.446
Token
Token ID198
Feature activation-1.34e-01

Pos logit contributions
ours+0.133
oured+0.131
ateral+0.111
hap+0.110
aths+0.109
Neg logit contributions
ンジ-0.122
-0.118
 Valhalla-0.112
 pose-0.111
-0.109
Token14
Token ID1415
Feature activation+7.33e-01

Pos logit contributions
ours+0.314
oured+0.311
ateral+0.252
hap+0.251
aths+0.246
Neg logit contributions
ンジ-0.400
-0.387
 Valhalla-0.378
 pose-0.374
-0.370
Token Recommend
Token ID30361
Feature activation+2.01e+00

Pos logit contributions
ンジ+1.357
+1.267
 Valhalla+1.248
 pose+1.222
+1.207
Neg logit contributions
ours-2.549
oured-2.537
ateral-2.216
hap-2.190
aths-2.187
Token all
Token ID477
Feature activation+2.31e-01

Pos logit contributions
ンジ+0.334
+0.314
 Valhalla+0.297
+0.292
 pose+0.288
Neg logit contributions
ours-0.661
oured-0.659
ateral-0.574
hap-0.569
aths-0.568
Token of
Token ID286
Feature activation+5.16e-01

Pos logit contributions
ンジ+0.542
+0.515
 Valhalla+0.507
 pose+0.498
+0.495
Neg logit contributions
ours-0.689
oured-0.686
ateral-0.584
hap-0.579
aths-0.575
Token its
Token ID663
Feature activation+6.10e-01

Pos logit contributions
ンジ+1.146
+1.087
 Valhalla+1.069
 pose+1.048
+1.043
Neg logit contributions
ours-1.595
oured-1.590
ateral-1.362
hap-1.350
aths-1.342
Token most
Token ID749
Feature activation+5.49e-01

Pos logit contributions
ンジ+1.339
+1.269
 Valhalla+1.244
 pose+1.221
+1.213
Neg logit contributions
ours-1.908
oured-1.900
ateral-1.632
hap-1.616
aths-1.607
Token memorable
Token ID18078
Feature activation+1.16e+00

Pos logit contributions
ンジ+1.292
+1.230
 Valhalla+1.198
 pose+1.177
+1.174
Neg logit contributions
ours-1.636
oured-1.632
ateral-1.387
hap-1.373
aths-1.364
Token moments
Token ID7188
Feature activation+3.27e-01

Pos logit contributions
ンジ+2.169
+2.046
 Valhalla+2.031
 pose+1.988
+1.957
Neg logit contributions
ours-3.963
oured-3.947
ateral-3.454
hap-3.431
aths-3.399
Token.
Token ID13
Feature activation+5.18e-01

Pos logit contributions
ンジ+0.850
+0.812
 Valhalla+0.794
 pose+0.780
+0.779
Neg logit contributions
ours-0.897
oured-0.893
ateral-0.748
hap-0.738
aths-0.734
Token<|endoftext|>
Token ID50256
Feature activation+1.47e-01

Pos logit contributions
ンジ+1.224
+1.164
 Valhalla+1.140
 pose+1.121
+1.116
Neg logit contributions
ours-1.537
oured-1.531
ateral-1.303
hap-1.289
aths-1.281
TokenWhat
Token ID2061
Feature activation+3.94e-01

Pos logit contributions
ンジ+0.388
+0.371
 Valhalla+0.364
 pose+0.358
+0.357
Neg logit contributions
ours-0.393
oured-0.391
ateral-0.327
hap-0.322
aths-0.320
Token do
Token ID466
Feature activation-5.61e-01

Pos logit contributions
ンジ+0.863
+0.817
 Valhalla+0.794
 pose+0.778
+0.776
Neg logit contributions
ours-1.244
oured-1.238
ateral-1.064
hap-1.052
aths-1.047
Token you
Token ID345
Feature activation-8.41e-02

Pos logit contributions
ours+1.874
oured+1.868
ateral+1.614
hap+1.606
aths+1.598
Neg logit contributions
ンジ-1.120
-1.055
 Valhalla-1.003
 pose-0.988
-0.984
Token tell
Token ID1560
Feature activation-4.32e-01

Pos logit contributions
ンジ+1.938
 pose+1.838
+1.819
 Valhalla+1.793
+1.744
Neg logit contributions
ours-3.494
oured-3.482
ateral-3.023
hap-3.003
yss-2.981
Token (
Token ID357
Feature activation-1.55e+00

Pos logit contributions
ours+1.340
oured+1.334
ateral+1.139
hap+1.129
aths+1.123
Neg logit contributions
ンジ-0.971
-0.919
 Valhalla-0.888
-0.870
 pose-0.864
Tokenwho
Token ID8727
Feature activation-2.41e-01

Pos logit contributions
ours+3.840
oured+3.774
ateral+3.069
hap+3.056
aths+3.043
Neg logit contributions
ンジ-4.423
-4.299
 Valhalla-4.252
 pose-4.118
-4.073
Token was
Token ID373
Feature activation-7.39e-01

Pos logit contributions
ours+0.719
oured+0.716
ateral+0.609
hap+0.602
aths+0.600
Neg logit contributions
ンジ-0.567
-0.541
 Valhalla-0.530
 pose-0.518
-0.516
Token responsible
Token ID4497
Feature activation-8.27e-01

Pos logit contributions
ours+2.217
oured+2.208
ateral+1.871
hap+1.853
aths+1.850
Neg logit contributions
ンジ-1.720
-1.653
 Valhalla-1.615
-1.557
 pose-1.549
Token)
Token ID8
Feature activation+1.30e-01

Pos logit contributions
ours+2.430
oured+2.406
ateral+2.039
hap+2.012
aths+2.008
Neg logit contributions
ンジ-1.979
-1.894
 Valhalla-1.856
 pose-1.801
-1.797
Token because
Token ID780
Feature activation-6.18e-01

Pos logit contributions
ンジ+0.289
+0.274
 Valhalla+0.268
 pose+0.263
+0.262
Neg logit contributions
ours-0.403
oured-0.402
ateral-0.344
hap-0.341
aths-0.339
Token this
Token ID428
Feature activation+2.69e-01

Pos logit contributions
ours+1.812
oured+1.799
ateral+1.527
hap+1.507
aths+1.500
Neg logit contributions
ンジ-1.481
-1.419
 Valhalla-1.377
-1.345
 pose-1.338
Token was
Token ID373
Feature activation-9.09e-03

Pos logit contributions
ンジ+0.591
+0.560
 Valhalla+0.548
 pose+0.538
+0.535
Neg logit contributions
ours-0.839
oured-0.836
ateral-0.717
hap-0.710
aths-0.706
Token a
Token ID257
Feature activation+8.22e-01

Pos logit contributions
ours+0.027
oured+0.027
ateral+0.023
hap+0.023
aths+0.023
Neg logit contributions
ンジ-0.021
-0.020
 Valhalla-0.019
 pose-0.019
-0.019
Token situation
Token ID3074
Feature activation-8.91e-02

Pos logit contributions
ンジ+1.795
+1.696
 Valhalla+1.666
 pose+1.647
+1.623
Neg logit contributions
ours-2.577
oured-2.564
ateral-2.205
hap-2.182
aths-2.168
Token,
Token ID11
Feature activation+7.45e-01

Pos logit contributions
ンジ+2.581
+2.449
 Valhalla+2.431
 pose+2.373
+2.352
Neg logit contributions
ours-3.385
oured-3.379
ateral-2.892
hap-2.858
aths-2.833
Token a
Token ID257
Feature activation+8.58e-01

Pos logit contributions
ンジ+1.665
+1.573
 Valhalla+1.558
 pose+1.528
+1.517
Neg logit contributions
ours-2.285
oured-2.279
ateral-1.956
hap-1.939
aths-1.926
Token Chinatown
Token ID47043
Feature activation+1.41e-01

Pos logit contributions
ンジ+1.702
+1.602
 Valhalla+1.564
 pose+1.534
+1.522
Neg logit contributions
ours-2.867
oured-2.858
ateral-2.481
hap-2.457
aths-2.445
Token native
Token ID6868
Feature activation+9.32e-01

Pos logit contributions
ンジ+0.296
+0.281
 Valhalla+0.273
 pose+0.267
+0.266
Neg logit contributions
ours-0.457
oured-0.455
ateral-0.391
hap-0.386
aths-0.385
Token.
Token ID13
Feature activation+4.55e-01

Pos logit contributions
ンジ+2.196
+2.083
 Valhalla+2.050
 pose+2.023
+2.002
Neg logit contributions
ours-2.737
oured-2.728
ateral-2.339
hap-2.322
aths-2.297
Token
Token ID198
Feature activation+6.16e-01

Pos logit contributions
ンジ+1.043
+0.990
 Valhalla+0.969
 pose+0.952
+0.948
Neg logit contributions
ours-1.382
oured-1.377
ateral-1.176
hap-1.164
aths-1.157
Token
Token ID198
Feature activation+2.91e-01

Pos logit contributions
ンジ+1.465
+1.398
 Valhalla+1.352
 pose+1.332
+1.323
Neg logit contributions
ours-1.835
oured-1.821
ateral-1.551
hap-1.533
aths-1.525
TokenDespite
Token ID8332
Feature activation-2.41e-01

Pos logit contributions
ンジ+0.768
+0.735
 Valhalla+0.722
 pose+0.711
+0.707
Neg logit contributions
ours-0.785
oured-0.781
ateral-0.652
hap-0.644
aths-0.641
Token this
Token ID428
Feature activation+4.37e-01

Pos logit contributions
ours+0.746
oured+0.742
ateral+0.633
hap+0.632
aths+0.626
Neg logit contributions
ンジ-0.538
-0.511
 Valhalla-0.497
-0.489
 pose-0.488
Token,
Token ID11
Feature activation+2.75e-01

Pos logit contributions
ンジ+1.046
+0.996
 Valhalla+0.976
 pose+0.961
+0.955
Neg logit contributions
ours-1.283
oured-1.278
ateral-1.086
hap-1.073
aths-1.068
Token neighborhood
Token ID6232
Feature activation+3.50e-01

Pos logit contributions
ンジ+0.603
+0.573
 Valhalla+0.556
 pose+0.546
+0.545
Neg logit contributions
ours-0.863
oured-0.859
ateral-0.737
hap-0.731
aths-0.727
Token product
Token ID1720
Feature activation+3.96e-01

Pos logit contributions
ours+0.618
oured+0.615
ateral+0.530
hap+0.522
aths+0.521
Neg logit contributions
ンジ-0.406
-0.385
 Valhalla-0.370
-0.365
 pose-0.364
Token.
Token ID13
Feature activation-1.25e-01

Pos logit contributions
ンジ+0.973
+0.925
 Valhalla+0.915
 pose+0.899
+0.891
Neg logit contributions
ours-1.132
oured-1.128
ateral-0.954
hap-0.948
aths-0.940
Token Atom
Token ID33102
Feature activation+8.99e-01

Pos logit contributions
ours+0.371
oured+0.368
ateral+0.314
aths+0.308
hap+0.308
Neg logit contributions
ンジ-0.293
-0.280
 pose-0.270
 Valhalla-0.269
-0.267
Token Freeman
Token ID22394
Feature activation+9.10e-01

Pos logit contributions
ンジ+2.100
+1.989
 Valhalla+1.957
 pose+1.925
+1.921
Neg logit contributions
ours-2.669
oured-2.662
ateral-2.276
hap-2.247
aths-2.231
Token:
Token ID25
Feature activation+9.05e-01

Pos logit contributions
ンジ+2.065
+1.954
 Valhalla+1.924
 pose+1.897
+1.881
Neg logit contributions
ours-2.758
oured-2.756
ateral-2.356
hap-2.343
aths-2.314
Token I
Token ID314
Feature activation+2.60e-01

Pos logit contributions
ンジ+1.979
+1.874
 Valhalla+1.857
 pose+1.810
+1.798
Neg logit contributions
ours-2.822
oured-2.819
ateral-2.419
hap-2.403
aths-2.382
Token would
Token ID561
Feature activation+1.52e+00

Pos logit contributions
ンジ+0.528
+0.496
 Valhalla+0.483
+0.472
 pose+0.467
Neg logit contributions
ours-0.861
oured-0.859
ateral-0.742
hap-0.736
aths-0.732
Token say
Token ID910
Feature activation+1.58e+00

Pos logit contributions
ンジ+2.701
 pose+2.543
+2.533
 Valhalla+2.501
+2.405
Neg logit contributions
ours-5.299
oured-5.276
ateral-4.603
hap-4.555
aths-4.543
Token D
Token ID360
Feature activation-2.63e-01

Pos logit contributions
ンジ+3.269
 Valhalla+3.127
+3.099
 pose+3.062
+2.967
Neg logit contributions
oured-5.011
ours-4.996
ateral-4.306
hap-4.293
aths-4.252
Tokenines
Token ID1127
Feature activation+3.14e-01

Pos logit contributions
ours+0.634
oured+0.631
ateral+0.515
hap+0.509
aths+0.504
Neg logit contributions
ンジ-0.763
-0.733
 Valhalla-0.721
 pose-0.713
-0.709
Tokenh
Token ID71
Feature activation+1.49e+00

Pos logit contributions
ンジ+0.963
+0.923
 Valhalla+0.913
 pose+0.904
+0.899
Neg logit contributions
ours-0.707
oured-0.705
ateral-0.563
hap-0.554
aths-0.552
Token what
Token ID644
Feature activation-1.18e-01

Pos logit contributions
ours+0.285
oured+0.284
ateral+0.242
hap+0.240
aths+0.238
Neg logit contributions
ンジ-0.207
-0.196
 Valhalla-0.188
-0.185
 pose-0.184
Token we
Token ID356
Feature activation+2.91e-01

Pos logit contributions
ours+0.373
oured+0.372
ateral+0.318
hap+0.313
aths+0.312
Neg logit contributions
ンジ-0.256
-0.243
 Valhalla-0.234
-0.230
 pose-0.229
Token've
Token ID1053
Feature activation+1.19e-01

Pos logit contributions
ンジ+0.653
+0.618
 Valhalla+0.605
+0.591
 pose+0.590
Neg logit contributions
ours-0.902
oured-0.899
ateral-0.769
hap-0.760
aths-0.757
Token always
Token ID1464
Feature activation+1.10e+00

Pos logit contributions
ンジ+0.328
+0.315
 Valhalla+0.308
+0.303
 pose+0.303
Neg logit contributions
ours-0.306
oured-0.305
ateral-0.251
hap-0.248
aths-0.246
Token done
Token ID1760
Feature activation+7.00e-01

Pos logit contributions
ンジ+2.952
+2.820
 Valhalla+2.784
 pose+2.756
+2.724
Neg logit contributions
ours-2.862
oured-2.847
ateral-2.377
hap-2.348
aths-2.333
Token.
Token ID13
Feature activation+4.66e-01

Pos logit contributions
ンジ+1.792
+1.710
 Valhalla+1.683
 pose+1.658
+1.645
Neg logit contributions
ours-1.929
oured-1.923
ateral-1.613
hap-1.596
aths-1.587
Token As
Token ID1081
Feature activation-2.40e-01

Pos logit contributions
ンジ+1.027
+0.974
 Valhalla+0.949
 pose+0.936
+0.930
Neg logit contributions
ours-1.454
oured-1.448
ateral-1.242
hap-1.229
aths-1.223
Token a
Token ID257
Feature activation+3.52e-01

Pos logit contributions
ours+0.760
oured+0.757
ateral+0.652
hap+0.641
aths+0.640
Neg logit contributions
ンジ-0.524
-0.498
 Valhalla-0.477
-0.471
 pose-0.470
Token result
Token ID1255
Feature activation-8.72e-02

Pos logit contributions
ンジ+0.701
+0.662
 Valhalla+0.639
+0.629
 pose+0.625
Neg logit contributions
ours-1.175
oured-1.172
ateral-1.015
hap-1.001
aths-1.001
Token of
Token ID286
Feature activation-1.23e-02

Pos logit contributions
ours+0.244
oured+0.243
ateral+0.205
hap+0.202
aths+0.201
Neg logit contributions
ンジ-0.220
-0.210
 Valhalla-0.206
 pose-0.203
-0.202
Token that
Token ID326
Feature activation+2.15e-01

Pos logit contributions
ours+0.038
oured+0.038
ateral+0.033
hap+0.032
aths+0.032
Neg logit contributions
ンジ-0.027
-0.026
 Valhalla-0.025
 pose-0.025
-0.025
Token 10
Token ID838
Feature activation+3.28e-01

Pos logit contributions
ンジ+0.151
+0.143
 Valhalla+0.137
+0.134
 pose+0.132
Neg logit contributions
ours-0.254
oured-0.252
ateral-0.218
hap-0.217
aths-0.217
Token."
Token ID526
Feature activation-1.71e-01

Pos logit contributions
ンジ+0.694
+0.657
 Valhalla+0.641
 pose+0.629
+0.626
Neg logit contributions
ours-1.053
oured-1.049
ateral-0.905
hap-0.896
aths-0.891
Token
Token ID198
Feature activation+2.67e-02

Pos logit contributions
ours+0.517
oured+0.515
ateral+0.440
hap+0.435
aths+0.433
Neg logit contributions
ンジ-0.394
-0.374
 Valhalla-0.365
 pose-0.359
-0.358
Token
Token ID198
Feature activation-2.02e-01

Pos logit contributions
ンジ+0.067
+0.064
 Valhalla+0.061
 pose+0.060
+0.060
Neg logit contributions
ours-0.077
oured-0.076
ateral-0.064
hap-0.063
aths-0.063
TokenOn
Token ID2202
Feature activation-2.01e-01

Pos logit contributions
ours+0.554
oured+0.550
ateral+0.462
hap+0.459
aths+0.456
Neg logit contributions
ンジ-0.523
-0.502
 Valhalla-0.495
 pose-0.487
-0.479
Token Monday
Token ID3321
Feature activation-1.43e+00

Pos logit contributions
ours+0.625
oured+0.624
ateral+0.535
hap+0.531
aths+0.528
Neg logit contributions
ンジ-0.446
-0.425
 Valhalla-0.409
-0.404
 pose-0.402
Token,
Token ID11
Feature activation-4.27e-01

Pos logit contributions
ours+3.819
oured+3.816
ateral+3.184
hap+3.175
aths+3.158
Neg logit contributions
ンジ-3.735
-3.553
 Valhalla-3.425
-3.412
 pose-3.345
Token the
Token ID262
Feature activation-6.02e-01

Pos logit contributions
ours+1.325
oured+1.323
ateral+1.135
hap+1.130
aths+1.122
Neg logit contributions
ンジ-0.949
-0.906
 Valhalla-0.859
-0.858
 pose-0.838
Token American
Token ID1605
Feature activation+7.30e-01

Pos logit contributions
ours+1.829
oured+1.824
ateral+1.557
hap+1.556
aths+1.537
Neg logit contributions
ンジ-1.365
-1.307
-1.238
 Valhalla-1.235
 pose-1.216
Token Academy
Token ID8581
Feature activation+2.26e-02

Pos logit contributions
ンジ+1.395
+1.315
 Valhalla+1.269
 pose+1.246
+1.242
Neg logit contributions
ours-2.496
oured-2.490
ateral-2.167
hap-2.146
aths-2.135
Token of
Token ID286
Feature activation+1.39e-01

Pos logit contributions
ンジ+0.064
+0.062
 Valhalla+0.060
+0.059
 pose+0.059
Neg logit contributions
ours-0.057
oured-0.056
ateral-0.047
hap-0.046
aths-0.045
Token and
Token ID290
Feature activation+1.99e-01

Pos logit contributions
ンジ+1.368
+1.307
 Valhalla+1.293
 pose+1.276
+1.262
Neg logit contributions
ours-1.494
oured-1.486
hap-1.246
ateral-1.239
aths-1.226
Token Count
Token ID2764
Feature activation-9.39e-02

Pos logit contributions
ンジ+0.467
+0.445
 Valhalla+0.436
 pose+0.429
+0.426
Neg logit contributions
ours-0.594
oured-0.591
ateral-0.504
hap-0.498
aths-0.495
Tokening
Token ID278
Feature activation-2.84e-02

Pos logit contributions
ours+0.252
oured+0.250
ateral+0.209
hap+0.205
aths+0.205
Neg logit contributions
ンジ-0.250
-0.240
 Valhalla-0.234
 pose-0.231
-0.230
Token from
Token ID422
Feature activation+8.12e-01

Pos logit contributions
ours+0.088
oured+0.087
ateral+0.075
hap+0.074
aths+0.074
Neg logit contributions
ンジ-0.063
-0.061
 Valhalla-0.059
 pose-0.058
-0.058
Token the
Token ID262
Feature activation+2.38e-02

Pos logit contributions
ンジ+1.860
+1.757
 Valhalla+1.747
 pose+1.716
+1.700
Neg logit contributions
ours-2.438
oured-2.426
ateral-2.061
hap-2.045
aths-2.041
Token air
Token ID1633
Feature activation-5.96e-01

Pos logit contributions
ンジ+0.053
+0.051
 Valhalla+0.050
 pose+0.049
+0.048
Neg logit contributions
ours-0.073
oured-0.073
ateral-0.062
hap-0.062
aths-0.061
Token.
Token ID13
Feature activation-7.63e-02

Pos logit contributions
ours+1.697
oured+1.685
ateral+1.430
hap+1.408
aths+1.396
Neg logit contributions
ンジ-1.481
-1.415
 Valhalla-1.375
 pose-1.353
-1.353
Token
Token ID198
Feature activation+1.52e-01

Pos logit contributions
ours+0.224
oured+0.223
ateral+0.190
hap+0.187
aths+0.187
Neg logit contributions
ンジ-0.182
-0.174
 Valhalla-0.169
 pose-0.168
-0.166
Token
Token ID198
Feature activation-6.09e-02

Pos logit contributions
ンジ+0.379
+0.364
 Valhalla+0.348
 pose+0.343
+0.340
Neg logit contributions
ours-0.440
oured-0.435
ateral-0.368
hap-0.364
aths-0.362
TokenWith
Token ID3152
Feature activation-4.60e-01

Pos logit contributions
ours+0.167
oured+0.166
ateral+0.139
hap+0.138
aths+0.137
Neg logit contributions
ンジ-0.157
-0.151
 Valhalla-0.148
 pose-0.146
-0.144
Token some
Token ID617
Feature activation-1.25e+00

Pos logit contributions
ours+1.421
oured+1.418
ateral+1.218
hap+1.206
aths+1.198
Neg logit contributions
ンジ-1.030
-0.978
 Valhalla-0.947
 pose-0.931
-0.928
Token inv
Token ID800
Feature activation-6.53e-01

Pos logit contributions
ours+1.761
oured+1.757
hap+1.492
aths+1.487
ateral+1.484
Neg logit contributions
ンジ-1.229
-1.176
-1.098
 Valhalla-1.087
 pose-1.061
Tokenader
Token ID5067
Feature activation-8.35e-01

Pos logit contributions
ours+2.106
oured+2.100
ateral+1.808
hap+1.797
aths+1.783
Neg logit contributions
ンジ-1.391
-1.308
 Valhalla-1.265
-1.249
 pose-1.234
Token.
Token ID13
Feature activation-1.64e-02

Pos logit contributions
ours+2.327
oured+2.320
ateral+1.918
hap+1.899
aths+1.899
Neg logit contributions
ンジ-2.124
-2.038
 Valhalla-1.941
-1.936
 pose-1.910
Token In
Token ID554
Feature activation+1.95e-01

Pos logit contributions
ours+0.049
oured+0.049
ateral+0.042
hap+0.041
aths+0.041
Neg logit contributions
ンジ-0.038
-0.036
 Valhalla-0.035
 pose-0.035
-0.035
Token 480
Token ID23487
Feature activation-1.61e+00

Pos logit contributions
ンジ+0.436
+0.413
 Valhalla+0.401
 pose+0.395
+0.394
Neg logit contributions
ours-0.603
oured-0.601
ateral-0.515
hap-0.510
aths-0.507
Token BCE
Token ID37078
Feature activation-1.12e-01

Pos logit contributions
ours+4.711
oured+4.693
hap+3.923
ateral+3.918
aths+3.907
Neg logit contributions
ンジ-3.837
-3.661
-3.467
 Valhalla-3.465
 pose-3.430
Token the
Token ID262
Feature activation-6.29e-01

Pos logit contributions
ours+0.337
oured+0.335
ateral+0.285
hap+0.283
aths+0.281
Neg logit contributions
ンジ-0.261
-0.248
 Valhalla-0.242
 pose-0.238
-0.237
Token Persian
Token ID22035
Feature activation-4.84e-01

Pos logit contributions
ours+1.933
oured+1.925
ateral+1.630
aths+1.616
hap+1.613
Neg logit contributions
ンジ-1.432
-1.363
-1.292
 Valhalla-1.284
 pose-1.268
Token king
Token ID5822
Feature activation-5.74e-01

Pos logit contributions
ours+1.522
oured+1.517
ateral+1.287
aths+1.286
hap+1.280
Neg logit contributions
ンジ-1.059
-1.014
 Valhalla-0.967
-0.958
 pose-0.947
Token Xer
Token ID44264
Feature activation-7.01e-01

Pos logit contributions
ours+1.732
oured+1.727
ateral+1.462
hap+1.448
aths+1.446
Neg logit contributions
ンジ-1.334
-1.274
 Valhalla-1.230
 pose-1.211
-1.209
Tokenxes
Token ID48169
Feature activation-6.25e-01

Pos logit contributions
ours+1.663
oured+1.654
ateral+1.330
hap+1.324
aths+1.322
Neg logit contributions
ンジ-2.075
-2.023
 Valhalla-1.974
 pose-1.924
-1.918
Token he
Token ID339
Feature activation+8.02e-01

Pos logit contributions
ンジ+0.305
+0.287
 Valhalla+0.279
 pose+0.273
+0.272
Neg logit contributions
ours-0.549
oured-0.547
ateral-0.477
hap-0.472
aths-0.470
Token continues
Token ID4477
Feature activation+2.31e-01

Pos logit contributions
ンジ+1.638
+1.556
 Valhalla+1.501
 pose+1.491
+1.475
Neg logit contributions
oured-2.606
ours-2.603
ateral-2.260
hap-2.252
yss-2.213
Token to
Token ID284
Feature activation+5.57e-01

Pos logit contributions
ンジ+0.530
+0.504
 Valhalla+0.493
 pose+0.486
+0.481
Neg logit contributions
ours-0.699
oured-0.697
ateral-0.595
hap-0.589
aths-0.585
Token investigate
Token ID9161
Feature activation-1.04e-01

Pos logit contributions
ンジ+1.053
+0.988
 Valhalla+0.960
 pose+0.936
+0.936
Neg logit contributions
ours-1.918
oured-1.912
ateral-1.666
hap-1.651
aths-1.643
Token the
Token ID262
Feature activation+1.85e-01

Pos logit contributions
ours+0.307
oured+0.305
ateral+0.260
hap+0.257
aths+0.256
Neg logit contributions
ンジ-0.246
-0.235
 Valhalla-0.229
 pose-0.225
-0.225
Token intriguing
Token ID19827
Feature activation-5.34e-01

Pos logit contributions
ンジ+0.427
+0.406
 Valhalla+0.397
 pose+0.390
+0.388
Neg logit contributions
ours-0.558
oured-0.555
ateral-0.474
hap-0.469
aths-0.466
Token and
Token ID290
Feature activation-4.21e-01

Pos logit contributions
ours+1.539
oured+1.529
ateral+1.293
hap+1.279
aths+1.273
Neg logit contributions
ンジ-1.302
-1.252
 Valhalla-1.210
-1.191
 pose-1.183
Token often
Token ID1690
Feature activation+2.59e-01

Pos logit contributions
ours+1.288
oured+1.282
ateral+1.097
hap+1.087
aths+1.080
Neg logit contributions
ンジ-0.949
-0.917
 Valhalla-0.882
-0.864
 pose-0.854
Token terrifying
Token ID17623
Feature activation+3.94e-01

Pos logit contributions
ンジ+0.588
+0.561
 Valhalla+0.546
 pose+0.535
+0.535
Neg logit contributions
ours-0.788
oured-0.786
ateral-0.672
hap-0.665
aths-0.661
Token intersection
Token ID16246
Feature activation+5.82e-01

Pos logit contributions
ンジ+0.955
+0.904
 Valhalla+0.894
 pose+0.885
+0.874
Neg logit contributions
ours-1.139
oured-1.137
ateral-0.964
hap-0.955
aths-0.947
Token of
Token ID286
Feature activation+7.61e-01

Pos logit contributions
ンジ+1.443
+1.378
 pose+1.353
 Valhalla+1.344
+1.335
Neg logit contributions
oured-1.634
ours-1.626
ateral-1.380
hap-1.360
yss-1.359
Token they
Token ID484
Feature activation+9.62e-01

Pos logit contributions
ンジ+2.623
 Valhalla+2.490
+2.472
 pose+2.459
+2.391
Neg logit contributions
oured-3.989
ours-3.979
ateral-3.428
hap-3.414
aths-3.375
Token have
Token ID423
Feature activation+8.47e-02

Pos logit contributions
ンジ+1.974
+1.863
 Valhalla+1.836
 pose+1.833
+1.787
Neg logit contributions
ours-3.119
oured-3.109
ateral-2.691
hap-2.661
aths-2.646
Token yet
Token ID1865
Feature activation+3.18e-01

Pos logit contributions
ンジ+0.200
+0.190
 Valhalla+0.187
 pose+0.185
+0.183
Neg logit contributions
ours-0.250
oured-0.249
ateral-0.212
hap-0.210
aths-0.209
Token to
Token ID284
Feature activation+2.48e-01

Pos logit contributions
ンジ+0.759
+0.722
 Valhalla+0.707
 pose+0.695
+0.692
Neg logit contributions
ours-0.935
oured-0.931
ateral-0.791
hap-0.783
aths-0.778
Token reckon
Token ID43851
Feature activation-1.04e-01

Pos logit contributions
ンジ+0.490
+0.461
 Valhalla+0.446
+0.436
 pose+0.432
Neg logit contributions
ours-0.835
oured-0.832
ateral-0.723
hap-0.716
aths-0.714
Token with
Token ID351
Feature activation-4.16e-01

Pos logit contributions
ours+0.370
oured+0.369
ateral+0.323
hap+0.320
aths+0.319
Neg logit contributions
ンジ-0.186
-0.174
 Valhalla-0.169
 pose-0.165
-0.164
Token A
Token ID317
Feature activation-2.99e-01

Pos logit contributions
ours+1.253
oured+1.246
ateral+1.066
aths+1.047
hap+1.045
Neg logit contributions
ンジ-0.969
-0.926
 Valhalla-0.877
-0.874
 pose-0.868
Tokenvel
Token ID626
Feature activation-1.15e+00

Pos logit contributions
ours+0.831
oured+0.824
ateral+0.689
aths+0.681
hap+0.680
Neg logit contributions
ンジ-0.769
-0.733
 Valhalla-0.719
 pose-0.705
-0.703
Tokenine
Token ID500
Feature activation-1.06e-01

Pos logit contributions
ours+3.881
oured+3.851
ateral+3.372
aths+3.282
yss+3.268
Neg logit contributions
ンジ-2.330
-2.198
 Valhalla-2.084
-2.030
 pose-2.028
Token,
Token ID11
Feature activation-4.81e-01

Pos logit contributions
ours+0.303
oured+0.302
ateral+0.255
hap+0.252
aths+0.251
Neg logit contributions
ンジ-0.262
-0.250
 Valhalla-0.244
 pose-0.240
-0.239
Token an
Token ID281
Feature activation-1.47e-02

Pos logit contributions
ours+1.447
oured+1.443
ateral+1.220
aths+1.207
hap+1.201
Neg logit contributions
ンジ-1.127
-1.080
 Valhalla-1.023
-1.009
 pose-1.002
Token humanity
Token ID9265
Feature activation+1.74e-01

Pos logit contributions
ours+0.236
oured+0.234
ateral+0.200
aths+0.200
hap+0.199
Neg logit contributions
ンジ-0.167
-0.158
 Valhalla-0.149
-0.148
 pose-0.146
Token
Token ID447
Feature activation+1.73e-01

Pos logit contributions
ンジ+0.449
+0.429
 Valhalla+0.420
 pose+0.414
+0.412
Neg logit contributions
ours-0.479
oured-0.476
ateral-0.399
hap-0.395
aths-0.393
Token
Token ID247
Feature activation-2.78e-01

Pos logit contributions
ンジ+0.286
+0.267
 Valhalla+0.259
 pose+0.252
+0.251
Neg logit contributions
ours-0.633
oured-0.631
ateral-0.556
hap-0.551
aths-0.548
Tokens
Token ID82
Feature activation-8.84e-02

Pos logit contributions
ours+0.551
oured+0.545
ateral+0.423
hap+0.414
aths+0.412
Neg logit contributions
ンジ-0.932
-0.901
 Valhalla-0.885
 pose-0.874
-0.873
Token d
Token ID288
Feature activation-8.57e-01

Pos logit contributions
ours+0.266
oured+0.263
ateral+0.224
hap+0.222
aths+0.222
Neg logit contributions
ンジ-0.207
-0.197
 Valhalla-0.190
-0.187
 pose-0.187
Tokenross
Token ID1214
Feature activation-1.97e+00

Pos logit contributions
ours+2.073
oured+2.048
ateral+1.675
hap+1.647
aths+1.646
Neg logit contributions
ンジ-2.533
-2.421
 Valhalla-2.363
 pose-2.336
-2.320
Token.
Token ID13
Feature activation+1.42e-01

Pos logit contributions
ours+5.384
oured+5.273
aths+4.422
ateral+4.414
hap+4.377
Neg logit contributions
ンジ-5.055
-4.832
 Valhalla-4.657
-4.627
 pose-4.548
Token This
Token ID770
Feature activation-2.14e-01

Pos logit contributions
ンジ+0.335
+0.320
 Valhalla+0.309
 pose+0.307
+0.305
Neg logit contributions
ours-0.420
oured-0.418
ateral-0.355
hap-0.351
aths-0.351
Token will
Token ID481
Feature activation+3.66e-01

Pos logit contributions
ours+0.688
oured+0.684
ateral+0.589
hap+0.586
aths+0.583
Neg logit contributions
ンジ-0.452
-0.431
 Valhalla-0.409
-0.404
 pose-0.396
Token be
Token ID307
Feature activation+4.15e-01

Pos logit contributions
ンジ+0.737
+0.695
 Valhalla+0.673
+0.657
 pose+0.656
Neg logit contributions
ours-1.214
oured-1.209
ateral-1.049
hap-1.040
aths-1.033
Token only
Token ID691
Feature activation+8.21e-01

Pos logit contributions
ンジ+0.952
+0.904
 Valhalla+0.885
 pose+0.870
+0.865
Neg logit contributions
ours-1.259
oured-1.254
ateral-1.072
hap-1.060
aths-1.054
Token groups
Token ID2628
Feature activation+3.00e-01

Pos logit contributions
ンジ+3.028
 Valhalla+2.869
 pose+2.844
+2.843
+2.753
Neg logit contributions
oured-4.143
ours-4.141
ateral-3.566
hap-3.521
aths-3.499
Token financed
Token ID29768
Feature activation+2.57e-01

Pos logit contributions
ンジ+0.692
+0.657
 Valhalla+0.643
 pose+0.632
+0.629
Neg logit contributions
ours-0.907
oured-0.904
ateral-0.772
hap-0.763
aths-0.759
Token include
Token ID2291
Feature activation+5.16e-01

Pos logit contributions
ンジ+0.576
+0.544
 Valhalla+0.535
 pose+0.527
+0.523
Neg logit contributions
ours-0.792
oured-0.790
ateral-0.677
hap-0.669
aths-0.665
Token the
Token ID262
Feature activation-2.34e-01

Pos logit contributions
ンジ+1.141
+1.081
 Valhalla+1.058
 pose+1.039
+1.033
Neg logit contributions
ours-1.604
oured-1.599
ateral-1.372
hap-1.357
aths-1.350
Token National
Token ID2351
Feature activation+6.45e-01

Pos logit contributions
ours+0.722
oured+0.718
ateral+0.613
hap+0.609
aths+0.604
Neg logit contributions
ンジ-0.525
-0.502
 Valhalla-0.483
-0.476
 pose-0.475
Token Council
Token ID4281
Feature activation-1.96e+00

Pos logit contributions
ンジ+1.633
+1.557
 Valhalla+1.530
 pose+1.506
+1.499
Neg logit contributions
ours-1.802
oured-1.795
ateral-1.511
hap-1.492
aths-1.484
Token of
Token ID286
Feature activation-2.99e-01

Pos logit contributions
ours+5.312
oured+5.235
ateral+4.362
hap+4.338
yss+4.223
Neg logit contributions
ンジ-5.042
-4.917
 Valhalla-4.736
-4.663
 pose-4.565
Token La
Token ID4689
Feature activation-4.51e-01

Pos logit contributions
ours+0.789
oured+0.785
ateral+0.654
hap+0.646
aths+0.641
Neg logit contributions
ンジ-0.805
-0.775
 Valhalla-0.752
-0.743
 pose-0.742
Token R
Token ID371
Feature activation-6.90e-01

Pos logit contributions
ours+1.481
oured+1.461
aths+1.250
ateral+1.240
hap+1.233
Neg logit contributions
ンジ-0.937
-0.893
 Valhalla-0.857
-0.849
 pose-0.825
Tokenaza
Token ID7056
Feature activation-4.56e-01

Pos logit contributions
ours+2.069
oured+2.055
ateral+1.768
aths+1.728
hap+1.723
Neg logit contributions
ンジ-1.629
-1.570
 Valhalla-1.515
-1.480
 pose-1.474
Token and
Token ID290
Feature activation+1.81e-01

Pos logit contributions
ours+1.297
oured+1.290
ateral+1.088
hap+1.076
aths+1.070
Neg logit contributions
ンジ-1.134
-1.085
 Valhalla-1.059
 pose-1.040
-1.039
Token:
Token ID25
Feature activation+3.21e-01

Pos logit contributions
ours+2.650
oured+2.647
ateral+2.261
aths+2.224
hap+2.216
Neg logit contributions
ンジ-1.989
-1.880
 Valhalla-1.816
-1.789
 pose-1.776
Token 200
Token ID939
Feature activation+5.02e-01

Pos logit contributions
ンジ+0.851
+0.813
 Valhalla+0.800
 pose+0.787
+0.784
Neg logit contributions
ours-0.856
oured-0.852
ateral-0.712
hap-0.703
aths-0.699
TokenW
Token ID54
Feature activation+8.87e-01

Pos logit contributions
ンジ+1.351
+1.286
 Valhalla+1.269
 pose+1.254
+1.245
Neg logit contributions
ours-1.322
oured-1.314
ateral-1.100
hap-1.080
aths-1.078
Token
Token ID198
Feature activation-4.89e-01

Pos logit contributions
ンジ+2.107
+1.991
 Valhalla+1.965
 pose+1.931
+1.914
Neg logit contributions
ours-2.609
oured-2.595
ateral-2.213
hap-2.198
aths-2.174
Token
Token ID198
Feature activation-6.72e-01

Pos logit contributions
ours+1.337
oured+1.316
ateral+1.098
hap+1.085
aths+1.081
Neg logit contributions
ンジ-1.298
-1.252
 Valhalla-1.179
 pose-1.165
-1.155
TokenOC
Token ID4503
Feature activation-1.86e+00

Pos logit contributions
ours+1.765
oured+1.755
ateral+1.454
aths+1.432
hap+1.431
Neg logit contributions
ンジ-1.822
-1.748
 Valhalla-1.716
 pose-1.691
-1.667
Token:
Token ID25
Feature activation+1.73e-01

Pos logit contributions
oured+5.766
ours+5.712
ateral+4.926
Ps+4.848
yss+4.833
Neg logit contributions
ンジ-4.170
-3.982
-3.738
 Valhalla-3.735
 pose-3.646
Token MSI
Token ID30781
Feature activation+1.78e-01

Pos logit contributions
ンジ+0.391
+0.371
 Valhalla+0.363
 pose+0.357
+0.355
Neg logit contributions
ours-0.530
oured-0.527
ateral-0.452
hap-0.447
aths-0.445
Token After
Token ID2293
Feature activation-4.31e-01

Pos logit contributions
ンジ+0.399
+0.379
 Valhalla+0.370
 pose+0.364
+0.362
Neg logit contributions
ours-0.548
oured-0.546
ateral-0.467
hap-0.462
aths-0.460
Tokenburn
Token ID10899
Feature activation+1.26e-01

Pos logit contributions
ours+1.119
oured+1.111
ateral+0.923
aths+0.909
hap+0.908
Neg logit contributions
ンジ-1.185
-1.122
 Valhalla-1.100
 pose-1.084
-1.078
Tokener
Token ID263
Feature activation+3.27e-01

Pos logit contributions
ンジ+0.303
+0.289
 Valhalla+0.282
 pose+0.278
+0.277
Neg logit contributions
ours-0.369
oured-0.367
ateral-0.312
hap-0.308
aths-0.306
Token
Token ID198
Feature activation+3.20e-02

Pos logit contributions
ンジ+0.713
+0.682
 Valhalla+0.655
 pose+0.644
+0.640
Neg logit contributions
ours-0.852
oured-0.843
ateral-0.715
hap-0.708
aths-0.704
TokenObama
Token ID15948
Feature activation+7.50e-01

Pos logit contributions
ンジ+0.080
+0.077
 Valhalla+0.076
 pose+0.074
+0.073
Neg logit contributions
ours-0.090
oured-0.090
ateral-0.075
hap-0.075
aths-0.074
Token delivered
Token ID6793
Feature activation+2.87e-01

Pos logit contributions
ンジ+1.706
+1.613
 Valhalla+1.583
 pose+1.571
+1.551
Neg logit contributions
ours-2.275
oured-2.266
ateral-1.942
hap-1.920
aths-1.902
Token.
Token ID13
Feature activation-1.65e-01

Pos logit contributions
ンジ+0.631
+0.598
 Valhalla+0.583
 pose+0.573
+0.571
Neg logit contributions
ours-0.898
oured-0.895
ateral-0.768
hap-0.761
aths-0.756
Token On
Token ID1550
Feature activation-4.74e-01

Pos logit contributions
ours+0.492
oured+0.489
ateral+0.418
aths+0.413
hap+0.412
Neg logit contributions
ンジ-0.384
-0.366
 Valhalla-0.355
 pose-0.353
-0.348
Token October
Token ID3267
Feature activation-1.99e+00

Pos logit contributions
ours+1.512
oured+1.509
ateral+1.299
hap+1.293
aths+1.284
Neg logit contributions
ンジ-1.017
-0.973
 Valhalla-0.919
-0.913
 pose-0.900
Token 8
Token ID807
Feature activation+2.72e-01

Pos logit contributions
ours+4.817
oured+4.779
aths+3.824
ateral+3.776
hap+3.763
Neg logit contributions
ンジ-5.686
-5.475
-5.130
 pose-5.051
 Valhalla-4.983
Token,
Token ID11
Feature activation-1.48e-01

Pos logit contributions
ンジ+0.632
+0.600
 Valhalla+0.588
 pose+0.578
+0.575
Neg logit contributions
ours-0.816
oured-0.813
ateral-0.694
hap-0.686
aths-0.682
Token Obama
Token ID2486
Feature activation+5.26e-01

Pos logit contributions
ours+0.471
oured+0.470
ateral+0.405
hap+0.402
aths+0.401
Neg logit contributions
ンジ-0.315
-0.301
 Valhalla-0.289
-0.283
 pose-0.282
Token was
Token ID373
Feature activation-6.39e-01

Pos logit contributions
ンジ+1.233
+1.171
 Valhalla+1.149
 pose+1.135
+1.125
Neg logit contributions
ours-1.561
oured-1.555
ateral-1.327
hap-1.312
aths-1.301
Token asked
Token ID1965
Feature activation+5.30e-01

Pos logit contributions
ours+1.935
oured+1.934
ateral+1.637
hap+1.632
aths+1.628
Neg logit contributions
ンジ-1.467
-1.405
 Valhalla-1.355
-1.331
 pose-1.308
Token at
Token ID379
Feature activation+3.67e-01

Pos logit contributions
ours+0.051
oured+0.051
ateral+0.043
hap+0.043
aths+0.042
Neg logit contributions
ンジ-0.046
-0.044
 Valhalla-0.042
-0.042
 pose-0.042
Token the
Token ID262
Feature activation-3.08e-02

Pos logit contributions
ンジ+0.805
+0.764
 Valhalla+0.749
 pose+0.736
+0.730
Neg logit contributions
ours-1.149
oured-1.143
ateral-0.982
hap-0.971
aths-0.967
Token irrational
Token ID25086
Feature activation+1.19e-01

Pos logit contributions
ours+0.097
oured+0.097
ateral+0.083
hap+0.083
aths+0.082
Neg logit contributions
ンジ-0.067
-0.063
 Valhalla-0.061
-0.060
 pose-0.060
Token ex
Token ID409
Feature activation-2.23e-01

Pos logit contributions
ンジ+0.258
+0.245
 Valhalla+0.239
 pose+0.234
+0.233
Neg logit contributions
ours-0.375
oured-0.374
ateral-0.321
hap-0.318
aths-0.316
Tokenuber
Token ID18478
Feature activation+1.05e-01

Pos logit contributions
ours+0.638
oured+0.634
ateral+0.537
hap+0.531
aths+0.527
Neg logit contributions
ンジ-0.555
-0.527
 Valhalla-0.515
 pose-0.506
-0.504
Tokenance
Token ID590
Feature activation-1.83e+00

Pos logit contributions
ンジ+0.356
+0.344
 Valhalla+0.340
 pose+0.337
+0.335
Neg logit contributions
ours-0.200
oured-0.199
ateral-0.153
hap-0.150
aths-0.148
Token that
Token ID326
Feature activation+7.43e-02

Pos logit contributions
ours+4.772
oured+4.751
ateral+3.950
hap+3.854
aths+3.816
Neg logit contributions
ンジ-4.919
-4.776
-4.474
 Valhalla-4.452
 pose-4.335
Token once
Token ID1752
Feature activation-4.98e-02

Pos logit contributions
ンジ+0.167
+0.159
 Valhalla+0.154
+0.151
 pose+0.150
Neg logit contributions
ours-0.230
oured-0.229
ateral-0.196
hap-0.194
aths-0.193
Token valued
Token ID17560
Feature activation+1.33e+00

Pos logit contributions
ours+0.140
oured+0.140
ateral+0.117
hap+0.116
aths+0.116
Neg logit contributions
ンジ-0.126
-0.120
 Valhalla-0.117
-0.115
 pose-0.114
Token a
Token ID257
Feature activation+6.94e-01

Pos logit contributions
ンジ+2.744
 Valhalla+2.604
+2.590
 pose+2.559
+2.499
Neg logit contributions
ours-4.256
oured-4.234
ateral-3.684
hap-3.640
yss-3.603
Token bitcoin
Token ID8550
Feature activation+7.71e-01

Pos logit contributions
ンジ+1.420
+1.339
 Valhalla+1.305
 pose+1.278
+1.273
Neg logit contributions
ours-2.280
oured-2.273
ateral-1.965
hap-1.947
aths-1.937