TOP ACTIVATIONS
MAX = 13.863

} at least one blank character (< Space > or <
font 2 = New - Object System . Draw ing .
Super NES and Game Boy Advance . We hope
was to explain away this stain of irrational ity ( Det
- sl it experiment with Cloud s old Match
issue of Frank Miller 's comic mini - series Ron in
Grove , California . Hide Caption 15 of 15
/ 3 minutes until golden brown . These are lightly fried
figure in DC Entertainment 's comic book mythology . Part of
Pasadena , California . Hide Caption 11 of 15 Photos :
Irvine , California . Hide Caption 10 of 15 Photos :
2 . im read ( image Path ) image = im
aga , California . Hide Caption 13 of 15 Photos :
Pasadena , California . Hide Caption 12 of 15 Photos :
aggressive victims of bullying . Child Development , 68 , 6
extended universe , like the comic books , the games .
OB B ( Object Al igned B ounding Box
once again that the old address https :// . nice hash
sent in els who stand guard today along the front iers
b mp = new - object system . draw ing .

MIN ACTIVATIONS
MIN = -12.643

pass over the middle by Siem ian . Several plays later
co / ni WB qv fi W u Dan Mer
<|endoftext|> Haw aii Congress woman Tuls i G abb ard is
The two form - in - one heat sinks cool the
drops - re za - as lan - after - prof
The Two Towers [ edit ] Or cs sent
war with Germany [ edit ] From 2 February
Stone talked with WikiLeaks about DNC emails Trump urges North Korea
the feeling of a big yo - yo then this isn
prime - chall enge - 35 790 347 . html https
. And if it had all four it was a nickel
# Never Trump ers of all , Wisconsin conservative radio host
tubing , features a full Shim ano 600 tri - color
Follow @ Pro _ Cy cling Tweet
, Death , Where Is Thy Sting - a - ling
With a bare bod kin ? who would f
between Russian hackers and the DNC leak isn 't a '
, from £ 9 79 all - in clusive , departing
Media playback is unsupported on your device Media caption Prof Re
two basketball goals on opposite ends stand over an empty court

INTERVAL 7.236 - 13.863
CONTAINS 0.430%

see her GP , the child stated clearly that she had
- hour Turkey - backed offensive . AFP has more from
| int f d | + -------------------- --+ | [ address
fram work using the command server less or sl s .
15 | org . cl ojure / tools . reader |

INTERVAL 0.610 - 7.236
CONTAINS 39.529%

reported as abstract s . <|endoftext|> A group of Ottawa fitness
s or adject ives , representing qualities such as solid ity
Co - operation and Development ( O ECD ). In the
by Warriors TV s Laure nce Scott and radio
. Well , after that I did give it

INTERVAL -6.016 - 0.610
CONTAINS 58.651%

Kane had come inches from putting Ireland in front just two
. The Liberals are offering the same raises the Conservatives gave
into the meditation on relationships and life that we all know
a lot of other ways to use cannabis . Ed ibles
said . " They don 't appreciate the fear m ong

INTERVAL -12.643 - -6.016
CONTAINS 1.389%

I 'll make the large corporations pay ' fair share of
Other executions took place in Chong jin
[ 10 : 47 ] < met ros xual th ug
J m B z 5 s 8 General Flynn (@
was coerced into filing a police complaint . She stated they
Token}
Token ID92
Feature activation+3.03e+00

Pos logit contributions
 Soup+3.316
esville+2.926
 Uniform+2.906
efe+2.886
eers+2.881
Neg logit contributions
Fact-5.333
itability-5.212
amus-5.058
weights-4.998
 campaigner-4.808
Token at
Token ID379
Feature activation+9.07e+00

Pos logit contributions
 Soup+1.757
esville+1.562
 Uniform+1.519
eers+1.491
icho+1.481
Neg logit contributions
Fact-2.567
itability-2.500
amus-2.430
weights-2.426
quished-2.315
Token least
Token ID1551
Feature activation+2.70e+00

Pos logit contributions
 Soup+4.160
esville+3.666
 Uniform+3.536
eers+3.449
 Albion+3.364
Neg logit contributions
Fact-8.442
itability-8.249
ccording-8.107
weights-8.020
amus-7.945
Token one
Token ID530
Feature activation+7.06e-01

Pos logit contributions
 Soup+1.867
esville+1.688
 Uniform+1.652
eers+1.625
icho+1.622
Neg logit contributions
Fact-1.989
itability-1.909
amus-1.859
weights-1.858
quished-1.761
Token blank
Token ID9178
Feature activation+5.81e+00

Pos logit contributions
 Soup+0.480
esville+0.439
 Uniform+0.422
eers+0.420
icho+0.416
Neg logit contributions
Fact-0.528
itability-0.509
weights-0.495
amus-0.494
quished-0.469
Token character
Token ID2095
Feature activation+1.39e+01

Pos logit contributions
 Soup+3.128
esville+2.781
 Uniform+2.755
eers+2.658
icho+2.657
Neg logit contributions
Fact-5.013
itability-4.916
weights-4.845
amus-4.781
kson-4.514
Token (<
Token ID38155
Feature activation+9.98e-01

Pos logit contributions
 Soup+6.222
 Uniform+5.185
 Arche+5.093
esville+4.982
icho+4.935
Neg logit contributions
itability-11.966
Fact-11.959
amus-11.781
weights-11.490
kson-10.878
TokenSpace
Token ID14106
Feature activation+4.86e+00

Pos logit contributions
 Soup+0.521
esville+0.460
 Uniform+0.440
eers+0.436
icho+0.431
Neg logit contributions
Fact-0.903
itability-0.877
weights-0.858
amus-0.857
quished-0.821
Token>
Token ID29
Feature activation-2.53e+00

Pos logit contributions
 Soup+2.680
esville+2.363
 Uniform+2.287
eers+2.249
icho+2.217
Neg logit contributions
Fact-4.219
itability-4.152
weights-4.032
amus-4.017
ccording-3.901
Token or
Token ID393
Feature activation+4.83e+00

Pos logit contributions
Fact+1.838
itability+1.765
weights+1.721
amus+1.715
quished+1.629
Neg logit contributions
 Soup-1.766
esville-1.624
 Uniform-1.551
eers-1.549
icho-1.538
Token <
Token ID1279
Feature activation+1.53e-01

Pos logit contributions
 Soup+1.968
esville+1.645
 Uniform+1.583
eers+1.515
 Albion+1.502
Neg logit contributions
Fact-4.936
itability-4.822
weights-4.714
amus-4.710
quished-4.547
Tokenfont
Token ID10331
Feature activation+8.64e+00

Pos logit contributions
 Soup+1.994
esville+1.802
eers+1.736
 Uniform+1.729
icho+1.695
Neg logit contributions
Fact-2.746
itability-2.687
amus-2.593
weights-2.582
quished-2.491
Token2
Token ID17
Feature activation-1.11e+00

Pos logit contributions
 Soup+4.786
esville+4.155
 Uniform+4.095
ingu+3.909
efe+3.894
Neg logit contributions
itability-6.971
Fact-6.959
amus-6.744
 campaigner-6.533
weights-6.529
Token =
Token ID796
Feature activation+5.53e+00

Pos logit contributions
Fact+0.919
itability+0.871
weights+0.864
amus+0.858
anger+0.813
Neg logit contributions
 Soup-0.665
esville-0.605
 Uniform-0.577
icho-0.576
eers-0.574
Token New
Token ID968
Feature activation+3.82e+00

Pos logit contributions
 Soup+2.802
 Uniform+2.366
esville+2.303
 Arche+2.199
 Albion+2.194
Neg logit contributions
Fact-5.100
itability-5.069
weights-4.857
amus-4.824
 campaigner-4.646
Token-
Token ID12
Feature activation+4.01e+00

Pos logit contributions
 Soup+2.365
esville+2.127
 Uniform+2.057
eers+2.040
 Albion+2.013
Neg logit contributions
Fact-3.056
itability-2.994
amus-2.901
weights-2.894
ccording-2.792
TokenObject
Token ID10267
Feature activation+1.38e+01

Pos logit contributions
 Soup+2.403
esville+2.162
 Uniform+2.092
eers+2.074
 Albion+2.038
Neg logit contributions
itability-3.223
Fact-3.219
amus-3.134
weights-3.108
ccording-3.040
Token System
Token ID4482
Feature activation+2.62e+00

Pos logit contributions
 Soup+5.647
 Uniform+5.110
 Arche+4.259
esville+4.192
 Albion+4.102
Neg logit contributions
Fact-12.981
itability-12.938
amus-12.463
weights-12.436
ennes-11.913
Token.
Token ID13
Feature activation+1.14e+00

Pos logit contributions
 Soup+1.353
esville+1.195
 Uniform+1.139
eers+1.128
icho+1.118
Neg logit contributions
Fact-2.398
itability-2.322
weights-2.276
amus-2.270
quished-2.171
TokenDraw
Token ID25302
Feature activation+3.92e+00

Pos logit contributions
 Soup+0.733
esville+0.665
icho+0.638
 Uniform+0.637
eers+0.634
Neg logit contributions
Fact-0.914
itability-0.863
weights-0.854
amus-0.847
ysc-0.805
Tokening
Token ID278
Feature activation+5.14e+00

Pos logit contributions
 Soup+1.566
esville+1.354
 Uniform+1.250
eers+1.245
 Albion+1.211
Neg logit contributions
Fact-4.003
itability-3.918
weights-3.841
amus-3.832
quished-3.727
Token.
Token ID13
Feature activation+2.71e+00

Pos logit contributions
 Soup+2.441
esville+2.133
 Uniform+2.022
eers+2.005
 Albion+1.964
Neg logit contributions
Fact-4.805
itability-4.764
weights-4.624
amus-4.618
ccording-4.550
Token Super
Token ID3115
Feature activation+1.79e+00

Pos logit contributions
 Soup+1.526
esville+1.327
 Uniform+1.276
eers+1.257
 Albion+1.254
Neg logit contributions
Fact-2.635
itability-2.566
weights-2.514
amus-2.509
quished-2.410
Token NES
Token ID31925
Feature activation+1.48e-02

Pos logit contributions
 Soup+0.967
esville+0.897
efe+0.866
 Uniform+0.860
icho+0.857
Neg logit contributions
Fact-1.580
itability-1.493
amus-1.480
weights-1.477
anger-1.417
Token
Token ID8151
Feature activation-1.69e+00

Pos logit contributions
 Soup+0.008
esville+0.008
icho+0.007
eers+0.007
 Uniform+0.007
Neg logit contributions
Fact-0.013
itability-0.012
weights-0.012
amus-0.012
quished-0.012
Token and
Token ID290
Feature activation+1.57e+00

Pos logit contributions
Fact+1.272
itability+1.228
weights+1.197
amus+1.195
anger+1.147
Neg logit contributions
 Soup-1.123
esville-1.042
icho-0.999
efe-0.988
 Uniform-0.983
Token Game
Token ID3776
Feature activation+5.82e+00

Pos logit contributions
 Soup+1.096
esville+1.043
eers+0.980
 Uniform+0.980
icho+0.977
Neg logit contributions
Fact-1.143
itability-1.092
amus-1.075
weights-1.067
anger-1.018
Token Boy
Token ID6387
Feature activation+1.34e+01

Pos logit contributions
 Soup+3.425
esville+3.040
 Uniform+2.963
eers+2.895
 Albion+2.883
Neg logit contributions
Fact-4.845
itability-4.744
weights-4.595
amus-4.588
quished-4.443
Token
Token ID8151
Feature activation+4.54e+00

Pos logit contributions
 Soup+4.879
 Uniform+4.105
 Arche+3.815
 Enhance+3.587
 XIV+3.585
Neg logit contributions
Fact-13.256
itability-12.846
amus-12.663
weights-12.505
oiler-12.441
Token Advance
Token ID25170
Feature activation+8.02e+00

Pos logit contributions
 Soup+1.474
esville+1.195
 Uniform+1.104
eers+1.088
icho+1.062
Neg logit contributions
Fact-5.013
itability-4.887
weights-4.804
amus-4.801
quished-4.632
Token.
Token ID13
Feature activation+2.18e+00

Pos logit contributions
 Soup+3.605
esville+2.984
eers+2.915
 Uniform+2.882
 Albion+2.820
Neg logit contributions
Fact-7.797
itability-7.521
amus-7.496
weights-7.370
ysc-7.181
Token We
Token ID775
Feature activation+9.44e-01

Pos logit contributions
 Soup+1.831
esville+1.707
 Uniform+1.654
eers+1.652
icho+1.645
Neg logit contributions
Fact-1.280
itability-1.208
weights-1.178
amus-1.169
quished-1.090
Token hope
Token ID2911
Feature activation+7.09e-01

Pos logit contributions
 Soup+0.578
esville+0.522
 Uniform+0.500
eers+0.497
 Albion+0.494
Neg logit contributions
Fact-0.769
itability-0.744
weights-0.727
amus-0.724
quished-0.692
Token was
Token ID373
Feature activation-1.75e+00

Pos logit contributions
Fact+0.206
itability+0.198
weights+0.193
amus+0.193
quished+0.183
Neg logit contributions
 Soup-0.181
esville-0.165
 Uniform-0.159
eers-0.158
icho-0.157
Token to
Token ID284
Feature activation+1.46e+00

Pos logit contributions
Fact+1.232
itability+1.182
amus+1.155
weights+1.148
quished+1.082
Neg logit contributions
 Soup-1.263
esville-1.154
eers-1.114
 Uniform-1.113
icho-1.100
Token explain
Token ID4727
Feature activation+1.67e+00

Pos logit contributions
 Soup+1.061
esville+0.974
 Uniform+0.944
eers+0.938
icho+0.930
Neg logit contributions
Fact-1.019
itability-0.977
weights-0.950
amus-0.948
quished-0.899
Token away
Token ID1497
Feature activation+3.18e+00

Pos logit contributions
 Soup+1.182
esville+1.081
 Uniform+1.052
eers+1.041
 Albion+1.038
Neg logit contributions
Fact-1.190
itability-1.142
weights-1.118
amus-1.110
quished-1.058
Token this
Token ID428
Feature activation+1.05e+00

Pos logit contributions
 Soup+2.258
esville+2.068
 Uniform+2.020
 Albion+1.997
 Arche+1.984
Neg logit contributions
Fact-2.244
itability-2.173
weights-2.110
amus-2.109
quished-2.010
Token stain
Token ID18746
Feature activation+1.32e+01

Pos logit contributions
 Soup+0.741
esville+0.679
 Uniform+0.655
eers+0.653
icho+0.648
Neg logit contributions
Fact-0.754
itability-0.725
weights-0.704
amus-0.703
quished-0.663
Token of
Token ID286
Feature activation+3.88e+00

Pos logit contributions
 Soup+6.745
esville+6.255
 Albion+6.064
 Arche+6.047
ingu+5.952
Neg logit contributions
Fact-10.339
itability-10.162
weights-9.891
amus-9.625
quished-9.574
Token irrational
Token ID25086
Feature activation-2.47e+00

Pos logit contributions
 Soup+2.702
esville+2.435
 Uniform+2.420
 Albion+2.386
 Arche+2.372
Neg logit contributions
Fact-2.792
itability-2.692
weights-2.637
amus-2.620
quished-2.512
Tokenity
Token ID414
Feature activation+2.34e+00

Pos logit contributions
Fact+1.758
itability+1.710
amus+1.653
weights+1.649
anger+1.560
Neg logit contributions
 Soup-1.763
esville-1.615
 Uniform-1.555
eers-1.552
icho-1.545
Token (
Token ID357
Feature activation+9.79e-01

Pos logit contributions
 Soup+1.443
esville+1.306
 Uniform+1.251
eers+1.241
 Albion+1.239
Neg logit contributions
Fact-1.890
itability-1.826
weights-1.789
amus-1.776
quished-1.703
TokenDet
Token ID11242
Feature activation+4.08e+00

Pos logit contributions
 Soup+0.654
esville+0.596
 Uniform+0.576
eers+0.571
icho+0.567
Neg logit contributions
Fact-0.738
itability-0.715
weights-0.695
amus-0.694
quished-0.664
Token-
Token ID12
Feature activation+1.93e+00

Pos logit contributions
Fact+0.311
itability+0.297
weights+0.293
amus+0.289
anger+0.274
Neg logit contributions
 Soup-0.275
esville-0.251
 Uniform-0.241
eers-0.240
icho-0.239
Tokensl
Token ID6649
Feature activation+2.52e+00

Pos logit contributions
 Soup+1.485
esville+1.379
eers+1.331
 Uniform+1.323
icho+1.316
Neg logit contributions
Fact-1.267
itability-1.216
amus-1.180
weights-1.164
quished-1.111
Tokenit
Token ID270
Feature activation+1.64e+00

Pos logit contributions
 Soup+1.875
esville+1.721
icho+1.665
 Uniform+1.656
eers+1.651
Neg logit contributions
Fact-1.708
itability-1.626
weights-1.600
amus-1.586
quished-1.514
Token experiment
Token ID6306
Feature activation+4.55e+00

Pos logit contributions
 Soup+1.400
esville+1.301
 Uniform+1.266
eers+1.260
icho+1.252
Neg logit contributions
Fact-0.946
itability-0.901
weights-0.871
amus-0.867
quished-0.809
Token with
Token ID351
Feature activation+2.37e+00

Pos logit contributions
 Soup+2.523
esville+2.245
 Uniform+2.152
eers+2.146
 Albion+2.111
Neg logit contributions
Fact-3.965
itability-3.851
weights-3.753
amus-3.746
ccording-3.653
Token Cloud
Token ID10130
Feature activation+1.31e+01

Pos logit contributions
 Soup+1.049
esville+0.906
 Uniform+0.855
eers+0.846
icho+0.834
Neg logit contributions
Fact-2.341
itability-2.276
weights-2.232
amus-2.227
quished-2.143
Token
Token ID447
Feature activation+2.46e+00

Pos logit contributions
 Soup+6.220
 Uniform+5.293
esville+5.291
 Albion+5.200
 Arche+5.198
Neg logit contributions
itability-11.398
Fact-11.301
ccording-11.240
weights-10.916
amus-10.902
Token
Token ID247
Feature activation-1.57e+00

Pos logit contributions
 Soup+1.446
esville+1.299
 Uniform+1.246
eers+1.235
icho+1.224
Neg logit contributions
Fact-2.071
itability-2.002
weights-1.958
amus-1.952
quished-1.865
Tokens
Token ID82
Feature activation+9.79e-01

Pos logit contributions
Fact+1.305
itability+1.245
weights+1.242
amus+1.225
anger+1.178
Neg logit contributions
 Soup-0.944
esville-0.841
 Uniform-0.823
icho-0.810
 Albion-0.805
Token old
Token ID1468
Feature activation+3.25e+00

Pos logit contributions
 Soup+0.593
esville+0.533
 Uniform+0.510
eers+0.508
icho+0.508
Neg logit contributions
Fact-0.803
itability-0.774
weights-0.761
amus-0.757
 campaigner-0.720
Token Match
Token ID13225
Feature activation+5.82e+00

Pos logit contributions
 Soup+2.158
esville+1.941
 Uniform+1.896
eers+1.864
 Albion+1.856
Neg logit contributions
Fact-2.496
itability-2.415
weights-2.346
amus-2.339
quished-2.238
Token issue
Token ID2071
Feature activation+3.47e+00

Pos logit contributions
 Soup+0.884
esville+0.803
 Uniform+0.780
eers+0.773
 Albion+0.765
Neg logit contributions
Fact-0.963
itability-0.927
weights-0.903
amus-0.900
quished-0.855
Token of
Token ID286
Feature activation+2.42e+00

Pos logit contributions
 Soup+2.341
esville+2.067
 Uniform+2.046
eers+2.026
 Albion+1.999
Neg logit contributions
Fact-2.627
itability-2.537
weights-2.466
amus-2.446
quished-2.330
Token Frank
Token ID5278
Feature activation+3.13e+00

Pos logit contributions
 Soup+1.564
esville+1.392
 Uniform+1.373
eers+1.341
 Albion+1.340
Neg logit contributions
Fact-1.900
itability-1.836
amus-1.785
weights-1.784
quished-1.701
Token Miller
Token ID7920
Feature activation+4.21e+00

Pos logit contributions
 Soup+2.185
esville+1.949
 Uniform+1.936
icho+1.882
eers+1.878
Neg logit contributions
Fact-2.317
itability-2.229
amus-2.140
weights-2.130
quished-2.058
Token's
Token ID338
Feature activation+8.73e+00

Pos logit contributions
 Soup+2.309
esville+1.992
 Uniform+1.967
eers+1.925
 Albion+1.897
Neg logit contributions
Fact-3.676
itability-3.594
weights-3.507
amus-3.478
 campaigner-3.341
Token comic
Token ID9048
Feature activation+1.29e+01

Pos logit contributions
 Soup+5.734
 Uniform+5.097
 Albion+4.941
esville+4.938
 Legion+4.882
Neg logit contributions
Fact-6.499
itability-6.133
weights-6.082
amus-6.058
quished-5.764
Token mini
Token ID9927
Feature activation+5.70e+00

Pos logit contributions
 Soup+8.463
 Uniform+7.450
eers+7.317
 Legion+7.212
 Albion+7.193
Neg logit contributions
Fact-9.106
itability-8.912
amus-8.600
weights-8.494
ccording-8.231
Token-
Token ID12
Feature activation+4.35e+00

Pos logit contributions
 Soup+3.441
esville+3.031
eers+2.985
 Uniform+2.975
icho+2.947
Neg logit contributions
Fact-4.577
itability-4.466
amus-4.354
weights-4.303
ysc-4.150
Tokenseries
Token ID25076
Feature activation+6.79e+00

Pos logit contributions
 Soup+3.885
esville+3.606
 Uniform+3.519
eers+3.514
icho+3.496
Neg logit contributions
Fact-2.243
itability-2.142
amus-2.060
weights-2.024
 campaigner-1.924
Token Ron
Token ID6575
Feature activation-1.50e+00

Pos logit contributions
 Soup+4.113
 Uniform+3.610
esville+3.568
eers+3.499
 Albion+3.482
Neg logit contributions
Fact-5.491
itability-5.349
amus-5.185
weights-5.177
 campaigner-4.975
Tokenin
Token ID259
Feature activation+5.28e-01

Pos logit contributions
Fact+1.419
itability+1.376
amus+1.351
weights+1.347
anger+1.294
Neg logit contributions
 Soup-0.717
esville-0.630
eers-0.602
 Uniform-0.594
 Albion-0.584
Token Grove
Token ID24144
Feature activation+4.32e+00

Pos logit contributions
 Soup+2.286
esville+2.059
 Uniform+1.887
eers+1.870
 Albion+1.860
Neg logit contributions
Fact-4.654
itability-4.509
weights-4.404
amus-4.369
quished-4.256
Token,
Token ID11
Feature activation+1.47e+00

Pos logit contributions
 Soup+2.275
esville+2.036
 Uniform+1.910
icho+1.875
 Albion+1.867
Neg logit contributions
Fact-3.894
itability-3.798
weights-3.707
amus-3.674
ysc-3.536
Token California
Token ID3442
Feature activation-1.77e+00

Pos logit contributions
 Soup+0.862
esville+0.770
 Uniform+0.744
eers+0.735
 Albion+0.732
Neg logit contributions
Fact-1.238
itability-1.205
weights-1.173
amus-1.172
quished-1.125
Token.
Token ID13
Feature activation+9.40e-01

Pos logit contributions
Fact+1.575
itability+1.524
amus+1.493
weights+1.491
quished+1.429
Neg logit contributions
 Soup-0.945
esville-0.838
 Uniform-0.804
eers-0.796
icho-0.789
Token Hide
Token ID10415
Feature activation+7.89e+00

Pos logit contributions
 Soup+0.787
esville+0.724
 Uniform+0.713
eers+0.701
 Albion+0.697
Neg logit contributions
Fact-0.556
itability-0.539
weights-0.519
amus-0.518
quished-0.484
Token Caption
Token ID11260
Feature activation+1.29e+01

Pos logit contributions
 Soup+2.024
esville+1.416
 Uniform+1.414
 Albion+1.239
 Arche+1.222
Neg logit contributions
Fact-9.075
itability-9.057
ccording-8.990
weights-8.871
amus-8.759
Token 15
Token ID1315
Feature activation+4.03e-02

Pos logit contributions
 Soup+6.003
esville+5.141
 Arche+5.093
 Uniform+4.970
 XIV+4.833
Neg logit contributions
ccording-12.621
itability-11.848
ategory-11.527
Fact-11.465
quished-11.252
Token of
Token ID286
Feature activation+1.46e+00

Pos logit contributions
 Soup+0.030
esville+0.028
 Uniform+0.027
eers+0.027
 Albion+0.027
Neg logit contributions
Fact-0.027
itability-0.026
weights-0.025
amus-0.025
anger-0.024
Token 15
Token ID1315
Feature activation+2.75e+00

Pos logit contributions
 Soup+0.937
esville+0.852
 Uniform+0.817
eers+0.815
icho+0.809
Neg logit contributions
Fact-1.141
itability-1.097
weights-1.074
amus-1.069
ysc-1.014
Token
Token ID198
Feature activation+1.59e+00

Pos logit contributions
 Soup+1.907
esville+1.739
 Uniform+1.683
eers+1.669
icho+1.655
Neg logit contributions
Fact-2.025
itability-1.955
weights-1.901
amus-1.896
quished-1.800
Token
Token ID198
Feature activation+4.54e-02

Pos logit contributions
 Soup+1.128
esville+1.032
 Uniform+0.997
eers+0.992
icho+0.986
Neg logit contributions
Fact-1.148
itability-1.096
weights-1.070
amus-1.067
quished-1.005
Token/
Token ID14
Feature activation-1.71e+00

Pos logit contributions
Fact+1.156
itability+1.098
weights+1.078
amus+1.076
ysc+1.018
Neg logit contributions
 Soup-1.022
esville-0.933
 Uniform-0.897
icho-0.896
eers-0.896
Token3
Token ID18
Feature activation+2.97e+00

Pos logit contributions
Fact+1.150
itability+1.077
weights+1.058
amus+1.046
anger+0.989
Neg logit contributions
 Soup-1.307
esville-1.209
 Uniform-1.181
 Albion-1.166
eers-1.165
Token minutes
Token ID2431
Feature activation+2.23e+00

Pos logit contributions
 Soup+2.231
esville+2.039
 Uniform+1.984
eers+1.966
icho+1.955
Neg logit contributions
Fact-2.019
itability-1.938
weights-1.880
amus-1.874
quished-1.778
Token until
Token ID1566
Feature activation+6.02e+00

Pos logit contributions
 Soup+1.070
esville+0.944
 Uniform+0.889
eers+0.884
icho+0.877
Neg logit contributions
Fact-2.102
itability-2.044
weights-2.001
amus-2.001
ysc-1.917
Token golden
Token ID10861
Feature activation+6.93e+00

Pos logit contributions
 Soup+3.018
esville+2.514
 Uniform+2.465
icho+2.425
eers+2.418
Neg logit contributions
Fact-5.613
itability-5.414
amus-5.310
ccording-5.224
weights-5.221
Token brown
Token ID7586
Feature activation+1.27e+01

Pos logit contributions
 Soup+4.035
esville+3.545
 Uniform+3.507
 Albion+3.420
eers+3.410
Neg logit contributions
Fact-5.779
itability-5.571
ccording-5.398
amus-5.372
weights-5.335
Token.
Token ID13
Feature activation+3.41e+00

Pos logit contributions
 Soup+4.295
esville+3.598
eers+3.458
 Uniform+3.169
 Arche+3.142
Neg logit contributions
ccording-13.842
Fact-12.788
itability-12.604
quished-12.418
amus-12.253
Token These
Token ID2312
Feature activation+3.33e-01

Pos logit contributions
 Soup+2.531
esville+2.267
 Uniform+2.255
eers+2.197
 Albion+2.194
Neg logit contributions
Fact-2.340
itability-2.328
amus-2.221
weights-2.207
quished-2.091
Token are
Token ID389
Feature activation+4.26e+00

Pos logit contributions
 Soup+0.247
esville+0.227
 Uniform+0.220
eers+0.219
icho+0.217
Neg logit contributions
Fact-0.229
itability-0.219
weights-0.214
amus-0.213
ysc-0.200
Token lightly
Token ID15376
Feature activation-3.07e+00

Pos logit contributions
 Soup+2.832
esville+2.550
 Uniform+2.453
icho+2.453
 Albion+2.429
Neg logit contributions
Fact-3.231
itability-3.141
weights-3.019
amus-3.012
ccording-2.877
Token fried
Token ID23018
Feature activation+2.32e-01

Pos logit contributions
Fact+2.141
itability+2.043
weights+2.040
amus+1.960
quished+1.915
Neg logit contributions
 Soup-2.195
esville-2.055
 Uniform-1.993
eers-1.960
 Albion-1.945
Token figure
Token ID3785
Feature activation+1.99e+00

Pos logit contributions
 Soup+3.586
esville+3.267
 Uniform+3.223
eers+3.160
 Arche+3.131
Neg logit contributions
Fact-3.813
itability-3.591
weights-3.584
amus-3.526
ysc-3.421
Token in
Token ID287
Feature activation-1.06e+00

Pos logit contributions
 Soup+1.211
esville+1.094
 Uniform+1.049
eers+1.045
icho+1.032
Neg logit contributions
Fact-1.619
itability-1.564
weights-1.525
amus-1.522
quished-1.455
Token DC
Token ID6257
Feature activation+2.49e+00

Pos logit contributions
Fact+0.832
itability+0.806
weights+0.786
amus+0.781
anger+0.742
Neg logit contributions
 Soup-0.683
esville-0.622
 Uniform-0.594
eers-0.591
icho-0.588
Token Entertainment
Token ID11058
Feature activation+3.68e+00

Pos logit contributions
 Soup+1.217
esville+1.051
 Uniform+1.018
eers+0.994
icho+0.986
Neg logit contributions
Fact-2.346
itability-2.278
weights-2.231
amus-2.227
quished-2.142
Token's
Token ID338
Feature activation+3.80e+00

Pos logit contributions
 Soup+2.077
esville+1.846
 Uniform+1.787
eers+1.768
 Albion+1.734
Neg logit contributions
Fact-3.166
itability-3.074
weights-3.000
amus-2.991
quished-2.883
Token comic
Token ID9048
Feature activation+1.26e+01

Pos logit contributions
 Soup+2.615
esville+2.347
 Uniform+2.328
icho+2.268
 Albion+2.261
Neg logit contributions
Fact-2.818
itability-2.673
weights-2.629
amus-2.625
quished-2.510
Token book
Token ID1492
Feature activation+8.96e+00

Pos logit contributions
 Soup+7.266
eers+6.454
 Uniform+6.209
esville+6.197
icho+6.115
Neg logit contributions
Fact-9.780
itability-9.300
amus-9.116
ccording-8.816
weights-8.754
Token mythology
Token ID25871
Feature activation+1.09e+00

Pos logit contributions
 Soup+5.932
eers+5.262
 Uniform+5.246
esville+5.210
 Legion+5.156
Neg logit contributions
Fact-6.669
itability-6.252
amus-6.108
weights-6.056
ccording-5.909
Token.
Token ID13
Feature activation-8.68e-01

Pos logit contributions
 Soup+0.587
esville+0.521
 Uniform+0.498
eers+0.494
icho+0.489
Neg logit contributions
Fact-0.964
itability-0.934
weights-0.914
amus-0.912
quished-0.873
Token Part
Token ID2142
Feature activation+4.14e-01

Pos logit contributions
Fact+0.573
itability+0.542
weights+0.530
amus+0.528
quished+0.495
Neg logit contributions
 Soup-0.664
esville-0.618
eers-0.595
icho-0.592
 Uniform-0.592
Token of
Token ID286
Feature activation+1.58e+00

Pos logit contributions
 Soup+0.293
esville+0.268
 Uniform+0.259
eers+0.258
icho+0.256
Neg logit contributions
Fact-0.298
itability-0.286
weights-0.279
amus-0.278
quished-0.263
Token Pasadena
Token ID41010
Feature activation+5.20e-01

Pos logit contributions
 Soup+1.201
esville+1.077
 Uniform+1.044
 Albion+1.032
eers+1.022
Neg logit contributions
Fact-1.615
itability-1.568
weights-1.526
amus-1.520
quished-1.465
Token,
Token ID11
Feature activation+4.29e-01

Pos logit contributions
 Soup+0.293
esville+0.262
 Uniform+0.251
eers+0.249
icho+0.246
Neg logit contributions
Fact-0.450
itability-0.435
weights-0.426
amus-0.425
quished-0.406
Token California
Token ID3442
Feature activation-1.37e+00

Pos logit contributions
 Soup+0.259
esville+0.233
 Uniform+0.224
eers+0.223
icho+0.220
Neg logit contributions
Fact-0.354
itability-0.342
weights-0.334
amus-0.333
quished-0.318
Token.
Token ID13
Feature activation+5.16e-01

Pos logit contributions
Fact+1.224
itability+1.183
weights+1.159
amus+1.159
quished+1.111
Neg logit contributions
 Soup-0.728
esville-0.646
 Uniform-0.617
eers-0.611
icho-0.606
Token Hide
Token ID10415
Feature activation+8.32e+00

Pos logit contributions
 Soup+0.430
esville+0.396
 Uniform+0.388
eers+0.383
icho+0.381
Neg logit contributions
Fact-0.308
itability-0.297
weights-0.286
amus-0.286
quished-0.267
Token Caption
Token ID11260
Feature activation+1.24e+01

Pos logit contributions
 Soup+2.049
 Uniform+1.407
esville+1.403
 Albion+1.217
 Arche+1.199
Neg logit contributions
Fact-9.629
itability-9.612
ccording-9.593
weights-9.420
amus-9.295
Token 11
Token ID1367
Feature activation-3.76e+00

Pos logit contributions
 Soup+5.656
 Arche+4.809
 Uniform+4.690
 XIV+4.617
esville+4.469
Neg logit contributions
ccording-12.448
itability-11.362
amus-10.981
ategory-10.956
quished-10.932
Token of
Token ID286
Feature activation+1.12e+00

Pos logit contributions
Fact+2.283
weights+2.087
amus+2.072
itability+2.055
anger+1.940
Neg logit contributions
 Soup-3.073
esville-2.823
icho-2.764
 Uniform-2.750
ingu-2.746
Token 15
Token ID1315
Feature activation+2.23e+00

Pos logit contributions
 Soup+0.724
esville+0.660
 Uniform+0.633
eers+0.630
icho+0.627
Neg logit contributions
Fact-0.873
itability-0.838
weights-0.820
amus-0.818
anger-0.773
Token Photos
Token ID9434
Feature activation+7.75e+00

Pos logit contributions
 Soup+1.238
esville+1.099
 Uniform+1.056
eers+1.043
icho+1.029
Neg logit contributions
Fact-1.952
itability-1.903
weights-1.853
amus-1.851
quished-1.778
Token:
Token ID25
Feature activation+6.82e-01

Pos logit contributions
 Soup+3.644
esville+3.208
eers+3.071
 Uniform+3.064
 Arche+2.989
Neg logit contributions
ccording-7.365
itability-7.104
Fact-7.046
quished-6.818
amus-6.795
Token Irvine
Token ID37277
Feature activation-5.30e-01

Pos logit contributions
 Soup+0.400
esville+0.357
 Uniform+0.344
eers+0.340
 Albion+0.337
Neg logit contributions
Fact-0.613
itability-0.594
weights-0.582
amus-0.578
quished-0.554
Token,
Token ID11
Feature activation+2.26e+00

Pos logit contributions
Fact+0.450
itability+0.434
amus+0.426
weights+0.424
quished+0.407
Neg logit contributions
 Soup-0.305
esville-0.273
 Uniform-0.262
eers-0.261
icho-0.260
Token California
Token ID3442
Feature activation-1.27e+00

Pos logit contributions
 Soup+1.293
esville+1.159
 Uniform+1.122
eers+1.114
 Albion+1.104
Neg logit contributions
Fact-1.918
itability-1.867
weights-1.819
amus-1.818
ysc-1.733
Token.
Token ID13
Feature activation+2.52e-01

Pos logit contributions
Fact+1.139
itability+1.103
weights+1.080
amus+1.080
quished+1.037
Neg logit contributions
 Soup-0.667
esville-0.593
 Uniform-0.564
eers-0.560
icho-0.553
Token Hide
Token ID10415
Feature activation+8.66e+00

Pos logit contributions
 Soup+0.208
esville+0.191
 Uniform+0.187
eers+0.185
icho+0.184
Neg logit contributions
Fact-0.152
itability-0.147
weights-0.142
amus-0.141
quished-0.132
Token Caption
Token ID11260
Feature activation+1.23e+01

Pos logit contributions
 Soup+2.137
 Uniform+1.465
esville+1.463
 Albion+1.267
 Arche+1.249
Neg logit contributions
ccording-10.015
Fact-9.976
itability-9.976
weights-9.768
amus-9.631
Token 10
Token ID838
Feature activation-6.37e+00

Pos logit contributions
 Soup+5.458
 Uniform+4.329
 Arche+4.323
esville+4.134
 XIV+4.106
Neg logit contributions
ccording-12.676
itability-11.522
quished-11.398
ategory-11.385
amus-11.282
Token of
Token ID286
Feature activation+8.27e-01

Pos logit contributions
Fact+3.603
weights+3.229
amus+3.218
itability+3.135
anger+3.002
Neg logit contributions
 Soup-5.370
esville-4.934
ingu-4.898
icho-4.863
 Uniform-4.826
Token 15
Token ID1315
Feature activation+2.31e+00

Pos logit contributions
 Soup+0.537
esville+0.489
 Uniform+0.470
eers+0.467
icho+0.465
Neg logit contributions
Fact-0.643
itability-0.618
weights-0.604
amus-0.603
anger-0.570
Token Photos
Token ID9434
Feature activation+7.70e+00

Pos logit contributions
 Soup+1.284
esville+1.140
 Uniform+1.096
eers+1.081
icho+1.067
Neg logit contributions
Fact-2.023
itability-1.972
weights-1.921
amus-1.918
quished-1.842
Token:
Token ID25
Feature activation+5.28e-01

Pos logit contributions
 Soup+3.622
esville+3.187
eers+3.050
 Uniform+3.043
 Arche+2.969
Neg logit contributions
ccording-7.327
itability-7.059
Fact-6.999
quished-6.779
amus-6.757
Token2
Token ID17
Feature activation+3.12e+00

Pos logit contributions
 Soup+2.298
esville+2.083
 Uniform+2.019
eers+1.989
 Albion+1.983
Neg logit contributions
Fact-2.602
itability-2.547
weights-2.464
amus-2.461
ccording-2.396
Token.
Token ID764
Feature activation+9.98e-01

Pos logit contributions
 Soup+1.696
esville+1.507
 Uniform+1.441
eers+1.425
 Albion+1.408
Neg logit contributions
Fact-2.749
itability-2.676
weights-2.607
amus-2.605
quished-2.520
Token im
Token ID545
Feature activation-3.52e-01

Pos logit contributions
 Soup+0.743
esville+0.684
 Uniform+0.661
eers+0.659
icho+0.653
Neg logit contributions
Fact-0.682
itability-0.655
weights-0.637
amus-0.635
quished-0.598
Tokenread
Token ID961
Feature activation+4.38e+00

Pos logit contributions
Fact+0.163
itability+0.155
weights+0.149
amus+0.147
ysc+0.136
Neg logit contributions
 Soup-0.340
esville-0.318
 Uniform-0.311
eers-0.310
 Albion-0.309
Token (
Token ID357
Feature activation+1.78e+00

Pos logit contributions
 Soup+2.623
esville+2.353
 Uniform+2.285
 Albion+2.226
eers+2.226
Neg logit contributions
Fact-3.569
itability-3.493
amus-3.433
ccording-3.402
weights-3.377
Token image
Token ID2939
Feature activation+1.22e+01

Pos logit contributions
 Soup+1.237
esville+1.130
 Uniform+1.092
eers+1.086
icho+1.077
Neg logit contributions
Fact-1.308
itability-1.259
weights-1.226
amus-1.222
quished-1.157
TokenPath
Token ID15235
Feature activation+9.50e+00

Pos logit contributions
 Soup+3.951
esville+2.904
 Uniform+2.854
 Arche+2.636
 Albion+2.612
Neg logit contributions
ccording-13.292
itability-12.711
quished-12.590
weights-12.450
amus-12.348
Token )
Token ID1267
Feature activation+7.43e-01

Pos logit contributions
 Soup+4.091
esville+3.440
 Uniform+3.320
 Albion+3.256
eers+3.130
Neg logit contributions
ccording-9.549
itability-9.081
Fact-8.948
quished-8.825
amus-8.708
Token image
Token ID2939
Feature activation+1.17e+01

Pos logit contributions
 Soup+0.526
esville+0.483
eers+0.466
 Uniform+0.464
icho+0.460
Neg logit contributions
Fact-0.534
itability-0.513
weights-0.500
amus-0.499
ysc-0.471
Token =
Token ID796
Feature activation+6.63e+00

Pos logit contributions
 Soup+4.797
esville+3.796
 Uniform+3.753
 Arche+3.553
 Albion+3.542
Neg logit contributions
ccording-12.053
itability-11.229
quished-11.006
Fact-10.992
amus-10.948
Token im
Token ID545
Feature activation-3.59e-01

Pos logit contributions
 Soup+4.420
 Uniform+3.931
esville+3.915
icho+3.822
 Albion+3.793
Neg logit contributions
Fact-4.952
itability-4.772
weights-4.656
amus-4.635
ccording-4.629
Tokenaga
Token ID8126
Feature activation-1.18e+00

Pos logit contributions
 Soup+0.095
esville+0.088
 Uniform+0.086
eers+0.086
icho+0.085
Neg logit contributions
Fact-0.059
itability-0.056
amus-0.054
weights-0.054
quished-0.050
Token,
Token ID11
Feature activation-4.44e-03

Pos logit contributions
Fact+0.981
itability+0.938
amus+0.928
weights+0.921
quished+0.882
Neg logit contributions
 Soup-0.702
esville-0.625
 Uniform-0.609
eers-0.608
 Albion-0.597
Token California
Token ID3442
Feature activation-2.02e+00

Pos logit contributions
Fact+0.004
itability+0.004
weights+0.004
amus+0.004
quished+0.003
Neg logit contributions
 Soup-0.003
esville-0.002
 Uniform-0.002
eers-0.002
icho-0.002
Token.
Token ID13
Feature activation+8.09e-02

Pos logit contributions
Fact+1.785
itability+1.724
weights+1.688
amus+1.688
quished+1.621
Neg logit contributions
 Soup-1.094
esville-0.975
 Uniform-0.931
eers-0.925
icho-0.914
Token Hide
Token ID10415
Feature activation+8.30e+00

Pos logit contributions
 Soup+0.067
esville+0.062
 Uniform+0.061
eers+0.060
 Albion+0.060
Neg logit contributions
Fact-0.048
itability-0.047
amus-0.045
weights-0.045
quished-0.042
Token Caption
Token ID11260
Feature activation+1.22e+01

Pos logit contributions
 Soup+2.062
esville+1.441
 Uniform+1.436
 Albion+1.251
eers+1.240
Neg logit contributions
Fact-9.569
itability-9.566
ccording-9.542
weights-9.356
amus-9.242
Token 13
Token ID1511
Feature activation-1.42e+00

Pos logit contributions
 Soup+4.891
eers+4.174
 Arche+4.112
esville+4.048
 Uniform+3.907
Neg logit contributions
ccording-12.508
itability-12.006
weights-11.547
Fact-11.502
amus-11.467
Token of
Token ID286
Feature activation+1.35e+00

Pos logit contributions
Fact+0.930
itability+0.868
weights+0.859
amus+0.855
anger+0.800
Neg logit contributions
 Soup-1.102
esville-1.010
 Uniform-0.981
icho-0.976
eers-0.974
Token 15
Token ID1315
Feature activation+1.86e+00

Pos logit contributions
 Soup+0.868
esville+0.789
 Uniform+0.758
eers+0.755
icho+0.749
Neg logit contributions
Fact-1.058
itability-1.018
weights-0.995
amus-0.993
anger-0.938
Token Photos
Token ID9434
Feature activation+7.22e+00

Pos logit contributions
 Soup+1.044
esville+0.931
 Uniform+0.893
eers+0.884
icho+0.874
Neg logit contributions
Fact-1.616
itability-1.569
weights-1.531
amus-1.528
quished-1.464
Token:
Token ID25
Feature activation+7.00e-01

Pos logit contributions
 Soup+3.480
esville+3.070
 Uniform+2.934
eers+2.932
 Arche+2.858
Neg logit contributions
ccording-6.725
itability-6.565
Fact-6.532
amus-6.291
quished-6.286
Token Pasadena
Token ID41010
Feature activation-1.50e-02

Pos logit contributions
 Soup+0.358
esville+0.316
 Uniform+0.302
eers+0.298
 Albion+0.296
Neg logit contributions
Fact-0.627
itability-0.608
weights-0.595
amus-0.594
quished-0.570
Token,
Token ID11
Feature activation+2.50e+00

Pos logit contributions
Fact+0.013
itability+0.013
weights+0.012
amus+0.012
quished+0.012
Neg logit contributions
 Soup-0.008
esville-0.007
 Uniform-0.007
eers-0.007
icho-0.007
Token California
Token ID3442
Feature activation-9.08e-01

Pos logit contributions
 Soup+1.411
esville+1.265
 Uniform+1.215
eers+1.210
 Albion+1.207
Neg logit contributions
Fact-2.136
itability-2.084
amus-2.033
weights-2.018
quished-1.934
Token.
Token ID13
Feature activation+4.90e-02

Pos logit contributions
Fact+0.819
itability+0.794
weights+0.777
amus+0.776
quished+0.744
Neg logit contributions
 Soup-0.477
esville-0.423
 Uniform-0.403
eers-0.399
icho-0.395
Token Hide
Token ID10415
Feature activation+8.25e+00

Pos logit contributions
 Soup+0.041
esville+0.038
 Uniform+0.037
eers+0.037
icho+0.037
Neg logit contributions
Fact-0.029
itability-0.028
weights-0.027
amus-0.027
quished-0.025
Token Caption
Token ID11260
Feature activation+1.22e+01

Pos logit contributions
 Soup+2.062
 Uniform+1.422
esville+1.416
 Albion+1.242
 Arche+1.213
Neg logit contributions
Fact-9.546
itability-9.530
ccording-9.490
weights-9.327
amus-9.213
Token 12
Token ID1105
Feature activation-5.19e+00

Pos logit contributions
 Soup+5.197
 Uniform+4.286
 Arche+4.251
esville+4.224
 Albion+3.981
Neg logit contributions
ccording-12.643
itability-11.838
ategory-11.417
amus-11.304
quished-11.292
Token of
Token ID286
Feature activation+1.75e+00

Pos logit contributions
Fact+3.029
weights+2.753
amus+2.728
itability+2.697
anger+2.552
Neg logit contributions
 Soup-4.322
esville-3.976
ingu-3.911
icho-3.902
 Uniform-3.889
Token 15
Token ID1315
Feature activation+2.22e+00

Pos logit contributions
 Soup+1.115
esville+1.012
 Uniform+0.973
eers+0.967
icho+0.959
Neg logit contributions
Fact-1.387
itability-1.336
weights-1.305
amus-1.302
quished-1.234
Token Photos
Token ID9434
Feature activation+7.56e+00

Pos logit contributions
 Soup+1.233
esville+1.095
 Uniform+1.052
eers+1.039
icho+1.026
Neg logit contributions
Fact-1.941
itability-1.892
weights-1.842
amus-1.840
quished-1.767
Token:
Token ID25
Feature activation+7.00e-01

Pos logit contributions
 Soup+3.591
esville+3.161
eers+3.026
 Uniform+3.019
 Arche+2.948
Neg logit contributions
ccording-7.136
itability-6.913
Fact-6.863
quished-6.632
amus-6.618
Token aggressive
Token ID8361
Feature activation-5.05e+00

Pos logit contributions
Fact+1.854
itability+1.791
weights+1.744
amus+1.731
anger+1.629
Neg logit contributions
 Soup-1.897
esville-1.748
 Uniform-1.680
eers-1.670
icho-1.667
Token victims
Token ID4970
Feature activation+1.07e+00

Pos logit contributions
Fact+3.422
itability+3.303
weights+3.256
amus+3.209
 campaigner+3.043
Neg logit contributions
 Soup-3.675
esville-3.410
icho-3.272
 Uniform-3.258
efe-3.246
Token of
Token ID286
Feature activation-1.44e+00

Pos logit contributions
 Soup+0.650
esville+0.583
 Uniform+0.563
eers+0.559
icho+0.553
Neg logit contributions
Fact-0.874
itability-0.846
amus-0.824
weights-0.824
quished-0.789
Token bullying
Token ID20714
Feature activation-1.45e-01

Pos logit contributions
Fact+0.970
itability+0.930
weights+0.906
amus+0.899
quished+0.847
Neg logit contributions
 Soup-1.083
esville-0.997
 Uniform-0.961
eers-0.958
icho-0.956
Token.
Token ID764
Feature activation+3.46e+00

Pos logit contributions
Fact+0.122
itability+0.118
weights+0.115
amus+0.115
quished+0.110
Neg logit contributions
 Soup-0.085
esville-0.076
 Uniform-0.073
eers-0.073
icho-0.072
Token Child
Token ID5932
Feature activation+1.20e+01

Pos logit contributions
 Soup+1.935
esville+1.701
 Uniform+1.674
 Albion+1.631
eers+1.614
Neg logit contributions
Fact-2.988
itability-2.911
weights-2.864
amus-2.843
quished-2.754
Token Development
Token ID7712
Feature activation+3.29e+00

Pos logit contributions
 Soup+6.291
esville+5.626
 Arche+5.543
 Uniform+5.432
eers+5.364
Neg logit contributions
Fact-9.796
itability-9.326
quished-9.255
weights-9.226
amus-9.202
Token,
Token ID11
Feature activation+4.57e-01

Pos logit contributions
 Soup+1.875
esville+1.684
 Uniform+1.609
eers+1.590
icho+1.564
Neg logit contributions
Fact-2.799
itability-2.738
weights-2.667
amus-2.645
quished-2.555
Token 68
Token ID8257
Feature activation+3.24e+00

Pos logit contributions
 Soup+0.226
esville+0.200
 Uniform+0.189
eers+0.189
icho+0.187
Neg logit contributions
Fact-0.425
itability-0.411
weights-0.404
amus-0.403
quished-0.385
Token,
Token ID11
Feature activation+2.96e+00

Pos logit contributions
 Soup+1.756
esville+1.551
 Uniform+1.492
eers+1.468
 Albion+1.446
Neg logit contributions
Fact-2.860
itability-2.805
weights-2.728
amus-2.722
quished-2.626
Token 6
Token ID718
Feature activation-1.15e-01

Pos logit contributions
 Soup+1.562
esville+1.360
 Uniform+1.329
eers+1.290
 Albion+1.290
Neg logit contributions
Fact-2.675
itability-2.609
weights-2.541
amus-2.539
quished-2.447
Token extended
Token ID7083
Feature activation-2.04e+00

Pos logit contributions
 Soup+0.665
esville+0.599
 Uniform+0.582
eers+0.576
 Albion+0.571
Neg logit contributions
Fact-0.805
itability-0.775
weights-0.758
amus-0.755
quished-0.721
Token universe
Token ID6881
Feature activation+6.72e-01

Pos logit contributions
Fact+1.382
itability+1.333
weights+1.298
amus+1.283
anger+1.209
Neg logit contributions
 Soup-1.523
esville-1.411
icho-1.353
eers-1.352
 Uniform-1.347
Token,
Token ID11
Feature activation+1.31e+00

Pos logit contributions
 Soup+0.408
esville+0.367
 Uniform+0.353
eers+0.351
icho+0.347
Neg logit contributions
Fact-0.553
itability-0.534
weights-0.522
amus-0.521
quished-0.497
Token like
Token ID588
Feature activation+8.73e-01

Pos logit contributions
 Soup+0.920
esville+0.836
 Uniform+0.816
eers+0.807
 Albion+0.803
Neg logit contributions
Fact-0.946
itability-0.908
weights-0.885
amus-0.882
quished-0.839
Token the
Token ID262
Feature activation+8.88e-01

Pos logit contributions
 Soup+0.589
esville+0.538
 Uniform+0.517
eers+0.515
icho+0.512
Neg logit contributions
Fact-0.658
itability-0.635
weights-0.618
amus-0.616
quished-0.584
Token comic
Token ID9048
Feature activation+1.20e+01

Pos logit contributions
 Soup+0.596
esville+0.542
 Uniform+0.524
eers+0.520
icho+0.515
Neg logit contributions
Fact-0.673
itability-0.648
weights-0.633
amus-0.631
quished-0.599
Token books
Token ID3835
Feature activation+6.61e+00

Pos logit contributions
 Soup+6.425
eers+5.611
 Uniform+5.449
 Arche+5.353
 Albion+5.324
Neg logit contributions
Fact-9.886
itability-9.593
amus-9.451
weights-9.077
 campaigner-8.923
Token,
Token ID11
Feature activation+2.36e+00

Pos logit contributions
 Soup+3.729
esville+3.214
 Uniform+3.214
eers+3.167
 Albion+3.140
Neg logit contributions
Fact-5.590
itability-5.383
weights-5.264
amus-5.216
quished-5.066
Token the
Token ID262
Feature activation+2.89e+00

Pos logit contributions
 Soup+1.625
esville+1.465
 Uniform+1.441
 Albion+1.416
eers+1.416
Neg logit contributions
Fact-1.743
itability-1.671
weights-1.631
amus-1.626
quished-1.552
Token games
Token ID1830
Feature activation+1.01e+00

Pos logit contributions
 Soup+1.774
esville+1.574
 Uniform+1.549
eers+1.521
 Albion+1.516
Neg logit contributions
Fact-2.359
itability-2.267
weights-2.222
amus-2.220
quished-2.132
Token.
Token ID13
Feature activation+4.69e-01

Pos logit contributions
 Soup+0.595
esville+0.534
 Uniform+0.514
eers+0.510
icho+0.504
Neg logit contributions
Fact-0.841
itability-0.813
weights-0.795
amus-0.793
quished-0.757
Token
Token ID198
Feature activation-4.61e+00

Pos logit contributions
Fact+1.006
itability+0.963
weights+0.943
amus+0.940
ysc+0.886
Neg logit contributions
 Soup-0.970
esville-0.895
eers-0.858
icho-0.857
 Uniform-0.856
Token
Token ID198
Feature activation-5.39e+00

Pos logit contributions
Fact+2.860
weights+2.611
amus+2.556
itability+2.554
anger+2.387
Neg logit contributions
 Soup-3.706
esville-3.447
icho-3.354
ingu-3.324
 Uniform-3.304
TokenOB
Token ID9864
Feature activation+9.11e-01

Pos logit contributions
Fact+3.902
weights+3.541
itability+3.535
amus+3.491
anger+3.317
Neg logit contributions
 Soup-3.891
esville-3.536
 Uniform-3.471
 Albion-3.411
eers-3.399
TokenB
Token ID33
Feature activation+2.97e+00

Pos logit contributions
 Soup+0.615
esville+0.560
 Uniform+0.542
eers+0.536
 Albion+0.533
Neg logit contributions
Fact-0.684
itability-0.659
weights-0.644
amus-0.643
quished-0.611
Token (
Token ID357
Feature activation+1.60e+00

Pos logit contributions
 Soup+1.130
esville+0.927
 Uniform+0.890
eers+0.841
icho+0.839
Neg logit contributions
Fact-3.113
itability-3.050
weights-2.987
amus-2.980
quished-2.894
TokenObject
Token ID10267
Feature activation+1.19e+01

Pos logit contributions
 Soup+1.088
esville+0.992
 Uniform+0.961
eers+0.949
icho+0.945
Neg logit contributions
Fact-1.192
itability-1.153
amus-1.122
weights-1.120
quished-1.067
Token Al
Token ID978
Feature activation+1.15e-03

Pos logit contributions
 Soup+7.903
esville+6.997
 Uniform+6.989
 Arche+6.888
 Albion+6.630
Neg logit contributions
Fact-8.307
itability-8.198
amus-7.920
weights-7.842
quished-7.711
Tokenigned
Token ID3916
Feature activation+4.55e+00

Pos logit contributions
 Soup+0.001
esville+0.001
 Uniform+0.001
eers+0.001
icho+0.001
Neg logit contributions
Fact-0.001
itability-0.001
weights-0.001
amus-0.001
quished-0.001
Token B
Token ID347
Feature activation+5.58e-01

Pos logit contributions
 Soup+2.954
 Uniform+2.626
esville+2.607
 Albion+2.509
 Arche+2.498
Neg logit contributions
Fact-3.535
itability-3.433
amus-3.350
weights-3.342
quished-3.258
Tokenounding
Token ID9969
Feature activation+3.24e+00

Pos logit contributions
 Soup+0.457
esville+0.423
 Uniform+0.412
eers+0.409
icho+0.408
Neg logit contributions
Fact-0.339
itability-0.323
amus-0.314
weights-0.313
quished-0.293
Token Box
Token ID8315
Feature activation+9.01e+00

Pos logit contributions
 Soup+1.976
esville+1.746
 Uniform+1.731
eers+1.686
 Albion+1.674
Neg logit contributions
Fact-2.650
itability-2.583
amus-2.509
weights-2.507
quished-2.415
Token once
Token ID1752
Feature activation+3.20e+00

Pos logit contributions
Fact+2.891
itability+2.845
weights+2.730
amus+2.683
ysc+2.583
Neg logit contributions
 Soup-2.697
esville-2.494
eers-2.390
 Uniform-2.381
icho-2.371
Token again
Token ID757
Feature activation+2.35e+00

Pos logit contributions
 Soup+2.648
esville+2.461
 Uniform+2.389
 Albion+2.369
eers+2.360
Neg logit contributions
Fact-1.897
itability-1.820
amus-1.769
weights-1.763
ccording-1.691
Token that
Token ID326
Feature activation+6.29e+00

Pos logit contributions
 Soup+1.828
esville+1.682
 Uniform+1.637
eers+1.627
 Albion+1.616
Neg logit contributions
Fact-1.529
itability-1.465
weights-1.417
amus-1.413
quished-1.338
Token the
Token ID262
Feature activation+4.35e+00

Pos logit contributions
 Soup+3.890
esville+3.476
 Uniform+3.404
 Albion+3.339
eers+3.316
Neg logit contributions
Fact-5.035
itability-4.881
amus-4.730
weights-4.725
ccording-4.654
Token old
Token ID1468
Feature activation+7.27e+00

Pos logit contributions
 Soup+2.280
esville+2.027
 Uniform+1.949
eers+1.917
 Albion+1.900
Neg logit contributions
Fact-3.925
itability-3.810
amus-3.721
weights-3.712
quished-3.592
Token address
Token ID2209
Feature activation+1.19e+01

Pos logit contributions
 Soup+3.679
esville+3.242
 Uniform+3.193
 Albion+3.089
efe+3.086
Neg logit contributions
Fact-6.539
itability-6.282
weights-6.226
amus-6.187
quished-6.129
Token https
Token ID3740
Feature activation+6.89e+00

Pos logit contributions
 Soup+6.578
 Uniform+5.758
 Arche+5.535
esville+5.504
efe+5.462
Neg logit contributions
Fact-9.227
itability-9.096
weights-8.798
amus-8.793
quished-8.558
Token://
Token ID1378
Feature activation+1.06e+01

Pos logit contributions
 Soup+1.879
esville+1.488
 Uniform+1.361
eers+1.307
icho+1.260
Neg logit contributions
Fact-7.710
itability-7.590
weights-7.470
ccording-7.450
amus-7.409
Token.
Token ID764
Feature activation+2.49e+00

Pos logit contributions
 Soup+5.761
esville+5.023
 Uniform+4.924
eers+4.920
efe+4.768
Neg logit contributions
Fact-8.648
itability-8.439
amus-8.085
weights-8.071
quished-7.894
Tokennice
Token ID44460
Feature activation+1.12e+00

Pos logit contributions
 Soup+1.273
esville+1.118
eers+1.074
icho+1.071
 Uniform+1.066
Neg logit contributions
Fact-2.288
itability-2.204
weights-2.160
amus-2.147
quished-2.081
Tokenhash
Token ID17831
Feature activation+2.13e+00

Pos logit contributions
 Soup+0.985
esville+0.917
 Uniform+0.894
eers+0.889
icho+0.882
Neg logit contributions
Fact-0.624
itability-0.593
weights-0.573
amus-0.571
ysc-0.528
Token sent
Token ID1908
Feature activation+5.99e-01

Pos logit contributions
Fact+1.043
itability+1.000
weights+0.985
amus+0.976
quished+0.924
Neg logit contributions
 Soup-0.950
esville-0.870
 Uniform-0.836
eers-0.829
icho-0.829
Tokenin
Token ID259
Feature activation-1.09e+00

Pos logit contributions
 Soup+0.371
esville+0.335
 Uniform+0.322
eers+0.320
icho+0.317
Neg logit contributions
Fact-0.484
itability-0.468
weights-0.457
amus-0.455
quished-0.434
Tokenels
Token ID1424
Feature activation+6.23e+00

Pos logit contributions
Fact+0.568
itability+0.538
weights+0.523
amus+0.522
ysc+0.480
Neg logit contributions
 Soup-0.983
esville-0.914
 Uniform-0.891
eers-0.886
icho-0.882
Token who
Token ID508
Feature activation+6.72e+00

Pos logit contributions
 Soup+3.864
esville+3.435
 Uniform+3.363
 Albion+3.334
eers+3.310
Neg logit contributions
Fact-4.968
itability-4.798
amus-4.640
weights-4.620
quished-4.461
Token stand
Token ID1302
Feature activation+1.03e+01

Pos logit contributions
 Soup+4.702
esville+4.310
 Uniform+4.220
 Albion+4.175
eers+4.150
Neg logit contributions
Fact-4.776
itability-4.533
weights-4.393
amus-4.348
quished-4.145
Token guard
Token ID4860
Feature activation+1.19e+01

Pos logit contributions
 Soup+5.770
esville+5.169
 Uniform+5.139
 Albion+5.114
 Arche+5.013
Neg logit contributions
Fact-8.195
itability-8.062
amus-7.656
weights-7.510
ysc-7.440
Token today
Token ID1909
Feature activation+5.28e+00

Pos logit contributions
 Soup+6.174
esville+5.878
 Albion+5.612
 Uniform+5.576
 Arche+5.512
Neg logit contributions
Fact-9.714
itability-9.413
amus-9.012
weights-8.974
ccording-8.862
Token along
Token ID1863
Feature activation+4.03e+00

Pos logit contributions
 Soup+2.871
esville+2.596
 Uniform+2.471
 Albion+2.434
eers+2.426
Neg logit contributions
Fact-4.607
itability-4.475
amus-4.354
weights-4.332
quished-4.160
Token the
Token ID262
Feature activation-7.41e-02

Pos logit contributions
 Soup+2.577
esville+2.336
 Uniform+2.264
 Albion+2.236
 Arche+2.220
Neg logit contributions
Fact-3.165
itability-3.065
amus-2.982
weights-2.956
quished-2.800
Token front
Token ID2166
Feature activation+3.82e+00

Pos logit contributions
Fact+0.053
itability+0.051
weights+0.050
amus+0.050
quished+0.047
Neg logit contributions
 Soup-0.053
esville-0.048
 Uniform-0.047
eers-0.046
icho-0.046
Tokeniers
Token ID3183
Feature activation+5.88e-01

Pos logit contributions
 Soup+1.861
esville+1.640
 Uniform+1.566
eers+1.552
icho+1.533
Neg logit contributions
Fact-3.590
itability-3.486
amus-3.392
weights-3.374
quished-3.242
Tokenb
Token ID65
Feature activation-1.01e+00

Pos logit contributions
Fact+0.808
itability+0.775
weights+0.761
amus+0.756
ysc+0.718
Neg logit contributions
 Soup-0.693
esville-0.626
 Uniform-0.606
eers-0.602
icho-0.597
Tokenmp
Token ID3149
Feature activation+2.67e+00

Pos logit contributions
Fact+0.540
itability+0.511
weights+0.496
amus+0.493
ysc+0.456
Neg logit contributions
 Soup-0.905
esville-0.842
 Uniform-0.822
eers-0.820
 Albion-0.814
Token =
Token ID796
Feature activation+1.55e+00

Pos logit contributions
 Soup+1.438
esville+1.269
 Uniform+1.219
eers+1.195
icho+1.184
Neg logit contributions
Fact-2.349
itability-2.309
weights-2.249
amus-2.240
quished-2.150
Token new
Token ID649
Feature activation+3.84e+00

Pos logit contributions
 Soup+0.941
esville+0.838
 Uniform+0.816
 Albion+0.796
eers+0.794
Neg logit contributions
Fact-1.283
itability-1.249
weights-1.213
amus-1.212
quished-1.159
Token-
Token ID12
Feature activation+3.72e+00

Pos logit contributions
 Soup+2.386
esville+2.145
 Uniform+2.085
eers+2.046
 Albion+2.035
Neg logit contributions
Fact-3.086
itability-3.012
amus-2.923
weights-2.912
quished-2.800
Tokenobject
Token ID15252
Feature activation+1.19e+01

Pos logit contributions
 Soup+2.992
esville+2.811
 Uniform+2.710
eers+2.698
efe+2.690
Neg logit contributions
Fact-2.225
itability-2.177
amus-2.108
weights-2.068
ccording-2.012
Token system
Token ID1080
Feature activation+1.38e+00

Pos logit contributions
 Soup+7.037
 Uniform+6.630
esville+6.241
 Albion+6.139
 Arche+6.120
Neg logit contributions
Fact-8.820
itability-8.749
amus-8.464
weights-8.314
quished-8.039
Token.
Token ID13
Feature activation+4.39e-01

Pos logit contributions
 Soup+0.746
esville+0.658
 Uniform+0.634
icho+0.625
eers+0.622
Neg logit contributions
Fact-1.239
itability-1.189
weights-1.173
amus-1.167
ysc-1.111
Tokendraw
Token ID19334
Feature activation+2.28e+00

Pos logit contributions
 Soup+0.291
esville+0.266
 Uniform+0.256
eers+0.254
icho+0.252
Neg logit contributions
Fact-0.335
itability-0.324
amus-0.315
weights-0.314
quished-0.299
Tokening
Token ID278
Feature activation+3.39e+00

Pos logit contributions
 Soup+0.981
esville+0.840
 Uniform+0.795
eers+0.785
icho+0.775
Neg logit contributions
Fact-2.275
itability-2.210
weights-2.168
amus-2.164
ysc-2.078
Token.
Token ID13
Feature activation+1.53e+00

Pos logit contributions
 Soup+1.702
esville+1.502
 Uniform+1.425
eers+1.416
icho+1.393
Neg logit contributions
Fact-3.129
itability-3.045
weights-2.975
amus-2.970
quished-2.863
Token pass
Token ID1208
Feature activation+2.08e+00

Pos logit contributions
Fact+0.110
itability+0.106
weights+0.104
amus+0.104
quished+0.099
Neg logit contributions
 Soup-0.072
esville-0.064
 Uniform-0.061
eers-0.061
icho-0.060
Token over
Token ID625
Feature activation-1.16e+00

Pos logit contributions
 Soup+1.318
esville+1.185
 Uniform+1.154
 Albion+1.147
eers+1.137
Neg logit contributions
Fact-1.646
itability-1.578
amus-1.548
weights-1.538
quished-1.477
Token the
Token ID262
Feature activation-2.36e+00

Pos logit contributions
Fact+0.854
itability+0.822
weights+0.800
amus+0.798
quished+0.756
Neg logit contributions
 Soup-0.806
esville-0.736
 Uniform-0.711
eers-0.707
icho-0.701
Token middle
Token ID3504
Feature activation-3.12e+00

Pos logit contributions
Fact+1.465
itability+1.382
weights+1.363
amus+1.348
 campaigner+1.263
Neg logit contributions
 Soup-1.891
esville-1.764
icho-1.716
eers-1.704
 Uniform-1.692
Token by
Token ID416
Feature activation-4.89e+00

Pos logit contributions
Fact+2.284
itability+2.204
weights+2.175
amus+2.149
quished+2.028
Neg logit contributions
 Soup-2.144
esville-1.972
eers-1.882
 Uniform-1.878
icho-1.863
Token Siem
Token ID45196
Feature activation-1.26e+01

Pos logit contributions
Fact+3.297
itability+3.140
weights+3.108
amus+3.054
 campaigner+2.904
Neg logit contributions
 Soup-3.571
esville-3.311
eers-3.209
icho-3.196
 Uniform-3.173
Tokenian
Token ID666
Feature activation-2.35e+00

Pos logit contributions
Fact+6.189
itability+6.122
amus+6.076
weights+5.477
kson+5.389
Neg logit contributions
 Soup-10.559
esville-9.681
eers-9.572
WARE-9.383
 Uniform-9.286
Token.
Token ID13
Feature activation-1.23e+00

Pos logit contributions
Fact+1.708
itability+1.641
weights+1.610
amus+1.599
quished+1.507
Neg logit contributions
 Soup-1.640
esville-1.498
 Uniform-1.438
eers-1.437
icho-1.420
Token Several
Token ID12168
Feature activation-4.44e+00

Pos logit contributions
Fact+0.870
itability+0.830
weights+0.813
amus+0.808
quished+0.765
Neg logit contributions
 Soup-0.884
esville-0.816
eers-0.784
 Uniform-0.782
icho-0.781
Token plays
Token ID5341
Feature activation+5.54e-02

Pos logit contributions
Fact+2.901
itability+2.762
weights+2.757
amus+2.674
 campaigner+2.541
Neg logit contributions
 Soup-3.343
esville-3.102
icho-3.028
efe-2.986
 Uniform-2.963
Token later
Token ID1568
Feature activation-1.41e+00

Pos logit contributions
 Soup+0.040
esville+0.036
 Uniform+0.035
eers+0.035
icho+0.035
Neg logit contributions
Fact-0.040
itability-0.038
weights-0.037
amus-0.037
quished-0.035
Tokenco
Token ID1073
Feature activation-1.87e+00

Pos logit contributions
Fact+1.544
itability+1.423
weights+1.420
amus+1.378
anger+1.250
Neg logit contributions
 Soup-4.198
esville-3.927
 Uniform-3.919
eers-3.862
 XIV-3.816
Token/
Token ID14
Feature activation-4.50e+00

Pos logit contributions
Fact+1.406
itability+1.312
weights+1.310
amus+1.297
anger+1.225
Neg logit contributions
 Soup-1.298
esville-1.152
 Uniform-1.132
eers-1.118
icho-1.111
Tokenni
Token ID8461
Feature activation-7.66e+00

Pos logit contributions
Fact+3.163
itability+2.989
amus+2.928
weights+2.920
ysc+2.778
Neg logit contributions
 Soup-3.275
esville-2.981
 Uniform-2.923
 Albion-2.872
eers-2.861
TokenWB
Token ID45607
Feature activation-6.83e+00

Pos logit contributions
Fact+5.156
itability+4.772
amus+4.720
weights+4.667
ysc+4.481
Neg logit contributions
 Soup-5.665
esville-5.218
 Uniform-5.104
 Albion-5.008
eers-4.983
Tokenqv
Token ID44179
Feature activation-7.91e+00

Pos logit contributions
Fact+4.620
itability+4.310
amus+4.249
weights+4.233
anger+4.037
Neg logit contributions
 Soup-5.097
esville-4.606
 Uniform-4.570
 Albion-4.496
eers-4.444
Tokenfi
Token ID12463
Feature activation-1.25e+01

Pos logit contributions
Fact+5.321
itability+4.949
amus+4.886
weights+4.838
ysc+4.651
Neg logit contributions
 Soup-5.827
esville-5.328
 Uniform-5.256
 Albion-5.160
 Arche-5.104
TokenW
Token ID54
Feature activation-9.12e+00

Pos logit contributions
Fact+8.170
itability+7.455
amus+7.411
weights+7.319
anger+7.151
Neg logit contributions
 Soup-8.824
esville-8.051
 Uniform-8.002
 Albion-7.890
 Arche-7.770
Tokenu
Token ID84
Feature activation-9.56e+00

Pos logit contributions
Fact+6.114
itability+5.656
amus+5.621
weights+5.581
anger+5.356
Neg logit contributions
 Soup-6.622
esville-5.976
 Uniform-5.946
 Albion-5.852
 Arche-5.775
Token
Token ID851
Feature activation-2.17e+00

Pos logit contributions
Fact+6.137
itability+5.653
amus+5.494
weights+5.471
anger+5.229
Neg logit contributions
 Soup-7.057
esville-6.572
eers-6.421
icho-6.337
 Uniform-6.296
Token Dan
Token ID6035
Feature activation-2.78e+00

Pos logit contributions
Fact+1.461
itability+1.393
amus+1.353
weights+1.352
quished+1.270
Neg logit contributions
 Soup-1.635
esville-1.511
eers-1.462
 Uniform-1.457
icho-1.448
Token Mer
Token ID4638
Feature activation-4.79e+00

Pos logit contributions
Fact+2.060
itability+1.969
amus+1.943
weights+1.920
anger+1.819
Neg logit contributions
 Soup-1.887
esville-1.716
eers-1.684
icho-1.676
 Uniform-1.665
Token<|endoftext|>
Token ID50256
Feature activation+5.36e+00

Pos logit contributions
 Soup+0.735
esville+0.665
 Uniform+0.647
eers+0.637
icho+0.634
Neg logit contributions
Fact-0.785
itability-0.760
weights-0.740
amus-0.737
quished-0.698
TokenHaw
Token ID33055
Feature activation-1.22e+00

Pos logit contributions
 Soup+3.612
esville+3.340
eers+3.234
icho+3.170
 Uniform+3.158
Neg logit contributions
itability-3.747
Fact-3.696
amus-3.614
weights-3.576
 campaigner-3.467
Tokenaii
Token ID42648
Feature activation-3.12e+00

Pos logit contributions
Fact+0.560
itability+0.530
amus+0.506
weights+0.503
quished+0.460
Neg logit contributions
 Soup-1.180
esville-1.101
 Uniform-1.083
eers-1.075
 Albion-1.068
Token Congress
Token ID3162
Feature activation-5.56e+00

Pos logit contributions
Fact+2.341
itability+2.254
amus+2.202
weights+2.194
 campaigner+2.080
Neg logit contributions
 Soup-2.073
esville-1.915
eers-1.829
icho-1.814
 Uniform-1.811
Tokenwoman
Token ID8580
Feature activation-2.96e+00

Pos logit contributions
Fact+3.430
itability+3.231
weights+3.198
amus+3.165
 campaigner+2.949
Neg logit contributions
 Soup-4.413
esville-4.093
 Albion-3.962
icho-3.932
 Uniform-3.923
Token Tuls
Token ID33219
Feature activation-1.25e+01

Pos logit contributions
Fact+1.829
itability+1.732
weights+1.710
amus+1.689
ysc+1.573
Neg logit contributions
 Soup-2.373
esville-2.220
eers-2.136
 Uniform-2.123
icho-2.121
Tokeni
Token ID72
Feature activation-7.18e+00

Pos logit contributions
Fact+8.560
amus+8.246
itability+8.160
weights+7.842
anger+7.666
Neg logit contributions
eers-7.852
 Soup-7.766
esville-7.656
WARE-7.543
 Uniform-7.264
Token G
Token ID402
Feature activation-2.56e+00

Pos logit contributions
Fact+5.083
amus+4.745
itability+4.708
weights+4.686
 campaigner+4.486
Neg logit contributions
 Soup-4.896
esville-4.688
eers-4.640
efe-4.485
icho-4.475
Tokenabb
Token ID6485
Feature activation-4.39e+00

Pos logit contributions
Fact+1.664
itability+1.565
weights+1.543
amus+1.505
anger+1.464
Neg logit contributions
 Soup-2.011
esville-1.861
 Albion-1.825
eers-1.779
 Uniform-1.772
Tokenard
Token ID446
Feature activation-7.43e-01

Pos logit contributions
Fact+3.531
amus+3.355
itability+3.307
anger+3.257
weights+3.244
Neg logit contributions
 Soup-2.663
esville-2.392
eers-2.381
 Albion-2.362
 Uniform-2.296
Token is
Token ID318
Feature activation-1.10e+00

Pos logit contributions
Fact+0.558
itability+0.537
weights+0.524
amus+0.523
quished+0.496
Neg logit contributions
 Soup-0.503
esville-0.458
 Uniform-0.442
eers-0.439
icho-0.436
Token The
Token ID383
Feature activation-7.91e-01

Pos logit contributions
Fact+0.259
itability+0.248
weights+0.241
amus+0.240
quished+0.227
Neg logit contributions
 Soup-0.270
esville-0.248
 Uniform-0.239
eers-0.239
icho-0.237
Token two
Token ID734
Feature activation-1.23e+00

Pos logit contributions
Fact+0.581
itability+0.559
weights+0.544
amus+0.544
ysc+0.512
Neg logit contributions
 Soup-0.547
esville-0.501
eers-0.481
icho-0.480
 Uniform-0.478
Token form
Token ID1296
Feature activation-8.94e+00

Pos logit contributions
Fact+0.918
itability+0.881
weights+0.863
amus+0.860
ysc+0.812
Neg logit contributions
 Soup-0.830
esville-0.758
eers-0.727
 Uniform-0.726
icho-0.722
Token-
Token ID12
Feature activation-5.21e+00

Pos logit contributions
Fact+6.098
weights+5.792
itability+5.675
amus+5.597
anger+5.362
Neg logit contributions
 Soup-6.384
esville-5.755
 Uniform-5.596
efe-5.579
 Albion-5.518
Tokenin
Token ID259
Feature activation-5.29e+00

Pos logit contributions
Fact+4.927
weights+4.729
itability+4.692
amus+4.614
anger+4.493
Neg logit contributions
 Soup-2.521
esville-2.172
 Uniform-2.088
 Albion-2.040
icho-2.013
Token-
Token ID12
Feature activation-1.24e+01

Pos logit contributions
Fact+3.853
itability+3.685
weights+3.657
amus+3.590
anger+3.420
Neg logit contributions
 Soup-3.613
esville-3.335
 Uniform-3.171
efe-3.149
 Albion-3.146
Tokenone
Token ID505
Feature activation-2.44e-01

Pos logit contributions
Fact+8.673
weights+8.656
itability+8.307
anger+7.970
amus+7.947
Neg logit contributions
 Soup-7.906
esville-7.244
 Uniform-7.095
 Albion-7.067
 XIV-6.944
Token heat
Token ID4894
Feature activation-5.41e+00

Pos logit contributions
Fact+0.115
itability+0.108
weights+0.104
amus+0.103
quished+0.094
Neg logit contributions
 Soup-0.234
esville-0.219
 Uniform-0.214
eers-0.213
icho-0.212
Token sinks
Token ID38614
Feature activation-1.53e+00

Pos logit contributions
Fact+3.677
weights+3.554
itability+3.529
amus+3.441
anger+3.283
Neg logit contributions
 Soup-3.905
esville-3.591
icho-3.463
 Albion-3.452
 Uniform-3.448
Token cool
Token ID3608
Feature activation-2.96e+00

Pos logit contributions
Fact+1.097
itability+1.055
weights+1.034
amus+1.029
anger+0.967
Neg logit contributions
 Soup-1.087
esville-0.996
eers-0.955
 Uniform-0.955
icho-0.951
Token the
Token ID262
Feature activation-1.89e+00

Pos logit contributions
Fact+2.104
itability+2.024
amus+1.984
weights+1.982
quished+1.864
Neg logit contributions
 Soup-2.095
esville-1.915
 Uniform-1.844
icho-1.843
eers-1.837
Tokendrops
Token ID49253
Feature activation-8.75e+00

Pos logit contributions
Fact+4.314
itability+4.078
weights+4.072
amus+4.064
anger+3.853
Neg logit contributions
 Soup-4.171
esville-3.784
 Uniform-3.689
 Albion-3.668
 XIV-3.633
Token-
Token ID12
Feature activation-5.39e+00

Pos logit contributions
Fact+6.038
itability+5.759
weights+5.642
amus+5.598
anger+5.340
Neg logit contributions
 Soup-6.137
esville-5.642
eers-5.564
 Uniform-5.482
 Albion-5.472
Tokenre
Token ID260
Feature activation-4.53e+00

Pos logit contributions
Fact+3.921
itability+3.713
weights+3.684
amus+3.640
anger+3.469
Neg logit contributions
 Soup-3.807
esville-3.448
 Uniform-3.397
 Albion-3.345
eers-3.335
Tokenza
Token ID4496
Feature activation-4.64e+00

Pos logit contributions
Fact+3.752
itability+3.633
amus+3.601
weights+3.566
anger+3.414
Neg logit contributions
 Soup-2.707
 Uniform-2.364
 Albion-2.305
esville-2.292
eers-2.252
Token-
Token ID12
Feature activation-7.03e+00

Pos logit contributions
Fact+3.417
itability+3.270
weights+3.226
amus+3.216
quished+3.044
Neg logit contributions
 Soup-3.189
esville-2.902
 Uniform-2.803
 Albion-2.769
eers-2.760
Tokenas
Token ID292
Feature activation-1.24e+01

Pos logit contributions
Fact+4.336
amus+4.226
itability+4.196
weights+4.144
anger+3.951
Neg logit contributions
 Soup-5.574
 Uniform-5.049
esville-5.029
 Albion-4.985
 Arche-4.871
Tokenlan
Token ID9620
Feature activation-7.73e-01

Pos logit contributions
itability+4.154
amus+3.729
Fact+3.558
weights+3.448
anger+3.196
Neg logit contributions
 Soup-13.044
 Uniform-12.179
 XIV-12.101
 Gleaming-11.870
��極-11.816
Token-
Token ID12
Feature activation-4.85e+00

Pos logit contributions
Fact+0.603
itability+0.582
weights+0.568
amus+0.566
quished+0.538
Neg logit contributions
 Soup-0.501
esville-0.455
 Uniform-0.438
eers-0.435
icho-0.431
Tokenafter
Token ID8499
Feature activation-7.43e+00

Pos logit contributions
Fact+3.477
itability+3.309
weights+3.282
amus+3.251
anger+3.084
Neg logit contributions
 Soup-3.440
esville-3.142
 Uniform-3.064
 Albion-3.023
eers-3.018
Token-
Token ID12
Feature activation-7.07e+00

Pos logit contributions
Fact+5.171
itability+4.971
weights+4.907
amus+4.834
quished+4.577
Neg logit contributions
 Soup-5.226
esville-4.718
 Uniform-4.648
 Albion-4.604
eers-4.586
Tokenprof
Token ID5577
Feature activation-8.49e-01

Pos logit contributions
Fact+5.178
itability+4.886
weights+4.881
amus+4.849
anger+4.709
Neg logit contributions
 Soup-4.847
esville-4.335
 Uniform-4.320
 Albion-4.278
 Arche-4.231
TokenThe
Token ID464
Feature activation-9.79e-01

Pos logit contributions
Fact+2.970
amus+2.676
itability+2.676
weights+2.674
anger+2.544
Neg logit contributions
 Soup-2.929
esville-2.726
 Uniform-2.653
icho-2.607
eers-2.572
Token Two
Token ID4930
Feature activation+2.72e-01

Pos logit contributions
Fact+0.701
itability+0.672
weights+0.656
amus+0.655
quished+0.619
Neg logit contributions
 Soup-0.694
esville-0.640
 Uniform-0.616
eers-0.612
icho-0.610
Token Towers
Token ID29031
Feature activation-3.59e+00

Pos logit contributions
 Soup+0.202
esville+0.186
 Uniform+0.180
eers+0.179
icho+0.178
Neg logit contributions
Fact-0.187
itability-0.179
weights-0.174
amus-0.173
quished-0.164
Token [
Token ID685
Feature activation-5.10e+00

Pos logit contributions
Fact+2.430
itability+2.318
weights+2.300
amus+2.283
anger+2.170
Neg logit contributions
 Soup-2.594
esville-2.438
 Uniform-2.329
icho-2.311
eers-2.295
Token edit
Token ID4370
Feature activation-3.14e-01

Pos logit contributions
Fact+2.749
itability+2.477
weights+2.463
amus+2.446
ysc+2.261
Neg logit contributions
 Soup-4.497
esville-4.291
icho-4.128
eers-4.124
 Uniform-4.109
Token ]
Token ID2361
Feature activation-1.23e+01

Pos logit contributions
Fact+0.236
itability+0.222
amus+0.220
weights+0.220
anger+0.205
Neg logit contributions
 Soup-0.213
esville-0.196
icho-0.191
 Uniform-0.187
eers-0.186
Token
Token ID198
Feature activation-8.25e+00

Pos logit contributions
Fact+5.877
amus+4.953
weights+4.856
itability+4.557
anger+4.480
Neg logit contributions
 Soup-10.217
ingu-10.128
esville-10.038
icho-9.902
��極-9.701
Token
Token ID198
Feature activation-7.35e+00

Pos logit contributions
Fact+4.404
weights+3.833
amus+3.782
itability+3.679
anger+3.511
Neg logit contributions
 Soup-7.071
esville-6.643
ingu-6.578
icho-6.510
��極-6.411
TokenOr
Token ID5574
Feature activation-2.77e+00

Pos logit contributions
Fact+5.200
weights+4.572
amus+4.551
itability+4.521
anger+4.404
Neg logit contributions
 Soup-5.349
esville-4.981
 Uniform-4.916
icho-4.794
efe-4.673
Tokencs
Token ID6359
Feature activation+4.87e-01

Pos logit contributions
Fact+1.981
itability+1.898
amus+1.889
weights+1.855
anger+1.770
Neg logit contributions
 Soup-1.949
esville-1.797
eers-1.735
 Uniform-1.728
efe-1.693
Token sent
Token ID1908
Feature activation-2.49e+00

Pos logit contributions
 Soup+0.350
esville+0.320
 Uniform+0.311
eers+0.309
 Albion+0.306
Neg logit contributions
Fact-0.347
itability-0.333
weights-0.324
amus-0.323
quished-0.306
Token war
Token ID1175
Feature activation-2.81e+00

Pos logit contributions
 Soup+0.640
esville+0.583
 Uniform+0.570
 Albion+0.561
eers+0.560
Neg logit contributions
Fact-0.637
itability-0.614
amus-0.595
weights-0.594
quished-0.563
Token with
Token ID351
Feature activation-3.43e+00

Pos logit contributions
Fact+2.092
itability+2.012
weights+1.967
amus+1.953
quished+1.857
Neg logit contributions
 Soup-1.900
esville-1.745
 Uniform-1.664
eers-1.661
icho-1.651
Token Germany
Token ID4486
Feature activation-3.40e+00

Pos logit contributions
Fact+2.399
itability+2.305
weights+2.273
amus+2.234
 campaigner+2.113
Neg logit contributions
 Soup-2.438
esville-2.269
eers-2.169
icho-2.157
 Uniform-2.145
Token [
Token ID685
Feature activation-5.96e+00

Pos logit contributions
Fact+2.460
itability+2.342
weights+2.300
amus+2.299
ysc+2.171
Neg logit contributions
 Soup-2.371
esville-2.184
icho-2.093
eers-2.089
 Uniform-2.084
Token edit
Token ID4370
Feature activation-4.00e-01

Pos logit contributions
Fact+3.164
itability+2.800
weights+2.798
amus+2.767
anger+2.578
Neg logit contributions
 Soup-5.330
esville-5.005
eers-4.847
icho-4.846
 Uniform-4.826
Token ]
Token ID2361
Feature activation-1.22e+01

Pos logit contributions
Fact+0.304
itability+0.288
weights+0.284
amus+0.283
anger+0.266
Neg logit contributions
 Soup-0.268
esville-0.245
icho-0.237
 Uniform-0.234
eers-0.233
Token
Token ID198
Feature activation-1.09e+01

Pos logit contributions
Fact+6.085
amus+5.074
weights+5.068
itability+4.815
anger+4.601
Neg logit contributions
 Soup-10.062
ingu-9.781
esville-9.733
icho-9.686
��極-9.409
Token
Token ID198
Feature activation-9.96e+00

Pos logit contributions
Fact+5.309
weights+4.475
amus+4.374
itability+4.210
anger+4.101
Neg logit contributions
 Soup-9.475
��極-9.035
esville-8.929
ingu-8.916
icho-8.900
TokenFrom
Token ID4863
Feature activation-3.34e+00

Pos logit contributions
Fact+7.052
weights+6.022
itability+5.877
amus+5.823
anger+5.791
Neg logit contributions
 Soup-7.304
esville-6.830
 Uniform-6.571
icho-6.568
eers-6.386
Token 2
Token ID362
Feature activation-4.53e+00

Pos logit contributions
Fact+2.519
itability+2.406
weights+2.366
amus+2.352
anger+2.230
Neg logit contributions
 Soup-2.224
esville-2.033
eers-1.963
icho-1.949
 Uniform-1.927
Token February
Token ID3945
Feature activation-3.03e+00

Pos logit contributions
Fact+2.830
itability+2.669
weights+2.637
amus+2.619
ysc+2.446
Neg logit contributions
 Soup-3.564
esville-3.289
icho-3.202
eers-3.193
 Uniform-3.179
Token Stone
Token ID8026
Feature activation+5.23e+00

Pos logit contributions
Fact+2.674
itability+2.625
weights+2.535
amus+2.472
 campaigner+2.388
Neg logit contributions
 Soup-2.626
esville-2.479
icho-2.362
 Uniform-2.345
eers-2.342
Token talked
Token ID6619
Feature activation-3.31e+00

Pos logit contributions
 Soup+3.400
esville+3.022
efe+2.903
 Uniform+2.897
 Albion+2.894
Neg logit contributions
Fact-4.063
itability-3.927
amus-3.904
weights-3.829
quished-3.704
Token with
Token ID351
Feature activation-3.06e+00

Pos logit contributions
Fact+2.977
itability+2.902
weights+2.831
amus+2.803
 campaigner+2.688
Neg logit contributions
 Soup-1.718
esville-1.568
eers-1.480
 Uniform-1.466
icho-1.445
Token WikiLeaks
Token ID19766
Feature activation-8.53e+00

Pos logit contributions
Fact+2.297
itability+2.205
weights+2.158
amus+2.154
quished+2.047
Neg logit contributions
 Soup-2.086
esville-1.890
 Uniform-1.831
eers-1.814
icho-1.800
Token about
Token ID546
Feature activation-3.54e+00

Pos logit contributions
itability+6.643
Fact+6.541
 campaigner+6.409
weights+6.333
kson+6.097
Neg logit contributions
 Soup-5.128
esville-4.815
icho-4.554
ingu-4.426
 Arche-4.417
Token DNC
Token ID17824
Feature activation-1.22e+01

Pos logit contributions
Fact+2.199
itability+2.091
weights+2.032
amus+2.024
anger+1.893
Neg logit contributions
 Soup-2.871
esville-2.658
 Uniform-2.592
eers-2.563
 Albion-2.550
Token emails
Token ID7237
Feature activation-9.19e+00

Pos logit contributions
Fact+6.741
itability+6.567
weights+6.548
 campaigner+6.298
 hinges+5.875
Neg logit contributions
esville-9.621
icho-9.457
 Soup-9.394
 Arche-9.054
 Albion-9.002
Token Trump
Token ID1301
Feature activation-1.84e+00

Pos logit contributions
Fact+6.232
itability+6.195
weights+5.809
 campaigner+5.660
kson+5.641
Neg logit contributions
 Soup-6.426
esville-6.253
icho-5.912
 Uniform-5.776
 Arche-5.763
Token urges
Token ID28804
Feature activation-3.20e+00

Pos logit contributions
Fact+1.372
itability+1.323
weights+1.300
amus+1.277
ysc+1.223
Neg logit contributions
 Soup-1.256
esville-1.144
icho-1.102
 Uniform-1.092
eers-1.091
Token North
Token ID2258
Feature activation-4.86e-01

Pos logit contributions
Fact+2.401
itability+2.317
weights+2.247
amus+2.244
 campaigner+2.163
Neg logit contributions
 Soup-2.130
esville-1.958
eers-1.884
 Albion-1.862
 Uniform-1.858
Token Korea
Token ID4969
Feature activation-6.87e+00

Pos logit contributions
Fact+0.381
itability+0.373
weights+0.368
amus+0.362
anger+0.359
Neg logit contributions
 Soup-0.298
esville-0.280
eers-0.269
 Uniform-0.264
 Albion-0.255
Token the
Token ID262
Feature activation-2.66e+00

Pos logit contributions
Fact+2.121
itability+2.045
weights+2.008
amus+1.968
ysc+1.856
Neg logit contributions
 Soup-2.279
esville-2.109
 Uniform-2.030
eers-2.029
efe-1.999
Token feeling
Token ID4203
Feature activation-3.38e+00

Pos logit contributions
Fact+1.877
itability+1.792
weights+1.756
amus+1.739
ysc+1.638
Neg logit contributions
 Soup-1.898
esville-1.768
eers-1.703
 Uniform-1.691
icho-1.676
Token of
Token ID286
Feature activation-3.52e+00

Pos logit contributions
Fact+2.257
itability+2.158
weights+2.125
amus+2.090
anger+1.975
Neg logit contributions
 Soup-2.525
esville-2.316
 Uniform-2.262
eers-2.255
efe-2.225
Token a
Token ID257
Feature activation-2.47e+00

Pos logit contributions
Fact+2.341
itability+2.257
weights+2.216
amus+2.168
anger+2.042
Neg logit contributions
 Soup-2.643
esville-2.441
eers-2.365
 Uniform-2.355
efe-2.337
Token big
Token ID1263
Feature activation-8.36e+00

Pos logit contributions
Fact+1.638
itability+1.569
weights+1.533
amus+1.519
anger+1.426
Neg logit contributions
 Soup-1.872
esville-1.744
eers-1.679
 Uniform-1.676
 Albion-1.654
Token yo
Token ID27406
Feature activation-1.21e+01

Pos logit contributions
Fact+5.332
weights+5.115
itability+5.061
amus+4.936
anger+4.717
Neg logit contributions
 Soup-6.061
esville-5.805
eers-5.590
 Uniform-5.442
efe-5.419
Token-
Token ID12
Feature activation-3.01e+00

Pos logit contributions
Fact+7.485
itability+6.929
amus+6.884
weights+6.828
ysc+6.558
Neg logit contributions
 Soup-8.746
esville-8.251
eers-8.024
 Uniform-7.768
 Albion-7.765
Tokenyo
Token ID8226
Feature activation-4.37e+00

Pos logit contributions
Fact+1.722
itability+1.641
weights+1.607
amus+1.595
ysc+1.485
Neg logit contributions
 Soup-2.560
esville-2.387
eers-2.318
 Uniform-2.312
 Albion-2.295
Token then
Token ID788
Feature activation-5.90e+00

Pos logit contributions
Fact+3.035
weights+2.912
itability+2.889
amus+2.831
 campaigner+2.686
Neg logit contributions
 Soup-3.124
esville-2.845
 Uniform-2.745
eers-2.739
efe-2.736
Token this
Token ID428
Feature activation-2.55e+00

Pos logit contributions
Fact+3.881
itability+3.711
weights+3.708
amus+3.580
 campaigner+3.409
Neg logit contributions
 Soup-4.294
esville-4.027
efe-3.895
eers-3.865
 Uniform-3.847
Token isn
Token ID2125
Feature activation+2.76e+00

Pos logit contributions
Fact+1.704
itability+1.632
weights+1.603
amus+1.589
anger+1.489
Neg logit contributions
 Soup-1.905
esville-1.780
eers-1.713
 Uniform-1.712
efe-1.686
Tokenprime
Token ID35505
Feature activation-7.43e-02

Pos logit contributions
Fact+3.385
itability+3.341
weights+3.257
amus+3.214
anger+3.120
Neg logit contributions
 Soup-4.436
 Uniform-4.010
esville-3.904
 Albion-3.892
 Arche-3.817
Token-
Token ID12
Feature activation-2.10e+00

Pos logit contributions
Fact+0.057
itability+0.055
weights+0.054
amus+0.053
anger+0.050
Neg logit contributions
 Soup-0.050
esville-0.045
 Uniform-0.043
eers-0.043
 Albion-0.043
Tokenchall
Token ID36747
Feature activation-1.52e+00

Pos logit contributions
Fact+1.136
itability+1.076
weights+1.051
amus+1.036
anger+0.978
Neg logit contributions
 Soup-1.889
esville-1.738
 Uniform-1.727
 Albion-1.707
eers-1.687
Tokenenge
Token ID3540
Feature activation-1.23e+00

Pos logit contributions
Fact+0.970
itability+0.925
amus+0.900
weights+0.891
anger+0.855
Neg logit contributions
 Soup-1.200
esville-1.108
 Uniform-1.078
 Albion-1.068
eers-1.062
Token-
Token ID12
Feature activation-3.04e+00

Pos logit contributions
Fact+0.950
itability+0.917
weights+0.893
amus+0.893
quished+0.846
Neg logit contributions
 Soup-0.809
esville-0.736
 Uniform-0.710
eers-0.701
icho-0.697
Token35
Token ID2327
Feature activation-1.20e+01

Pos logit contributions
Fact+1.726
itability+1.638
amus+1.588
weights+1.585
ysc+1.473
Neg logit contributions
 Soup-2.602
esville-2.424
 Uniform-2.382
eers-2.345
icho-2.327
Token790
Token ID37750
Feature activation-1.01e+01

Pos logit contributions
Fact+6.887
amus+6.456
itability+6.395
weights+6.268
anger+6.053
Neg logit contributions
 Soup-9.311
 Uniform-8.378
esville-8.355
eers-8.194
 Albion-8.175
Token347
Token ID30995
Feature activation-9.88e+00

Pos logit contributions
Fact+5.919
itability+5.513
amus+5.498
weights+5.381
ysc+5.079
Neg logit contributions
 Soup-7.971
esville-7.223
 Uniform-7.162
 XIV-7.018
 Albion-6.963
Token.
Token ID13
Feature activation-9.13e+00

Pos logit contributions
Fact+7.002
itability+6.657
amus+6.644
weights+6.441
ysc+6.330
Neg logit contributions
 Soup-6.413
esville-5.850
 Uniform-5.623
icho-5.592
 Albion-5.585
Tokenhtml
Token ID6494
Feature activation-9.95e+00

Pos logit contributions
Fact+4.254
itability+3.993
weights+3.885
amus+3.876
anger+3.665
Neg logit contributions
 Soup-8.434
 Uniform-7.891
esville-7.785
 Arche-7.774
icho-7.732
Token https
Token ID3740
Feature activation-2.15e+00

Pos logit contributions
Fact+5.291
itability+4.568
weights+4.489
amus+4.343
anger+4.172
Neg logit contributions
 Soup-8.106
esville-7.904
icho-7.871
ingu-7.653
rations-7.519
Token.
Token ID13
Feature activation+1.09e+00

Pos logit contributions
Fact+0.910
itability+0.877
weights+0.865
amus+0.853
anger+0.806
Neg logit contributions
 Soup-0.825
esville-0.754
 Uniform-0.726
eers-0.720
icho-0.716
Token And
Token ID843
Feature activation-3.89e+00

Pos logit contributions
 Soup+0.867
esville+0.799
 Uniform+0.778
eers+0.772
icho+0.766
Neg logit contributions
Fact-0.685
itability-0.657
weights-0.636
amus-0.635
quished-0.597
Token if
Token ID611
Feature activation-1.63e-01

Pos logit contributions
Fact+2.508
itability+2.392
weights+2.360
amus+2.318
 campaigner+2.173
Neg logit contributions
 Soup-2.954
esville-2.761
 Uniform-2.682
icho-2.656
eers-2.650
Token it
Token ID340
Feature activation-5.37e+00

Pos logit contributions
Fact+0.113
itability+0.109
weights+0.106
amus+0.105
quished+0.099
Neg logit contributions
 Soup-0.119
esville-0.109
 Uniform-0.106
eers-0.105
icho-0.104
Token had
Token ID550
Feature activation-8.27e+00

Pos logit contributions
Fact+3.775
amus+3.652
weights+3.638
itability+3.629
ysc+3.404
Neg logit contributions
 Soup-3.739
esville-3.440
 Uniform-3.346
 Albion-3.261
eers-3.235
Token all
Token ID477
Feature activation-1.17e+01

Pos logit contributions
Fact+5.102
weights+5.006
amus+4.963
itability+4.911
 hinges+4.504
Neg logit contributions
 Soup-6.164
esville-5.690
 Uniform-5.561
efe-5.481
 Albion-5.456
Token four
Token ID1440
Feature activation-6.22e+00

Pos logit contributions
amus+6.913
Fact+6.839
weights+6.779
itability+6.727
 hinges+6.140
Neg logit contributions
 Soup-8.638
esville-7.851
 Uniform-7.829
efe-7.657
 Albion-7.541
Token it
Token ID340
Feature activation-5.12e+00

Pos logit contributions
Fact+3.093
itability+2.996
weights+2.953
amus+2.912
anger+2.609
Neg logit contributions
 Soup-5.612
esville-5.220
 Uniform-5.099
efe-5.051
eers-5.015
Token was
Token ID373
Feature activation-1.30e+00

Pos logit contributions
Fact+3.457
itability+3.334
amus+3.314
weights+3.305
anger+3.076
Neg logit contributions
 Soup-3.756
esville-3.468
 Uniform-3.388
 Albion-3.301
eers-3.293
Token a
Token ID257
Feature activation-2.33e+00

Pos logit contributions
Fact+0.931
itability+0.901
weights+0.883
amus+0.878
 campaigner+0.822
Neg logit contributions
 Soup-0.910
esville-0.841
 Uniform-0.809
icho-0.803
eers-0.801
Token nickel
Token ID30666
Feature activation+1.92e+00

Pos logit contributions
Fact+1.671
itability+1.612
weights+1.569
amus+1.564
 campaigner+1.481
Neg logit contributions
 Soup-1.635
esville-1.511
 Uniform-1.442
icho-1.433
eers-1.432
Token #
Token ID1303
Feature activation+5.88e-01

Pos logit contributions
Fact+2.972
weights+2.827
itability+2.783
amus+2.680
 campaigner+2.627
Neg logit contributions
 Soup-4.249
esville-3.942
icho-3.859
 Uniform-3.810
efe-3.777
TokenNever
Token ID12295
Feature activation-3.64e+00

Pos logit contributions
 Soup+0.232
esville+0.197
 Uniform+0.184
eers+0.182
icho+0.179
Neg logit contributions
Fact-0.605
itability-0.591
weights-0.580
amus-0.579
quished-0.558
TokenTrump
Token ID6170
Feature activation-2.66e+00

Pos logit contributions
Fact+2.471
itability+2.233
weights+2.228
amus+2.158
anger+2.058
Neg logit contributions
 Soup-2.877
esville-2.600
 Uniform-2.596
 Albion-2.557
eers-2.547
Tokeners
Token ID364
Feature activation-6.76e+00

Pos logit contributions
Fact+1.550
itability+1.470
weights+1.446
amus+1.426
 campaigner+1.341
Neg logit contributions
 Soup-2.251
esville-2.083
 Uniform-2.033
icho-2.020
 Albion-1.999
Token of
Token ID286
Feature activation-2.75e+00

Pos logit contributions
Fact+4.885
weights+4.589
itability+4.567
amus+4.565
 campaigner+4.370
Neg logit contributions
 Soup-4.589
esville-4.172
icho-4.082
eers-4.067
 Uniform-4.042
Token all
Token ID477
Feature activation-1.17e+01

Pos logit contributions
Fact+1.938
itability+1.864
weights+1.818
amus+1.799
anger+1.703
Neg logit contributions
 Soup-1.975
esville-1.817
 Uniform-1.755
eers-1.749
icho-1.740
Token,
Token ID11
Feature activation-2.87e+00

Pos logit contributions
Fact+6.097
weights+5.774
itability+5.667
amus+5.442
 campaigner+5.258
Neg logit contributions
 Soup-9.695
esville-8.875
 Uniform-8.862
eers-8.816
icho-8.757
Token Wisconsin
Token ID9279
Feature activation-4.14e+00

Pos logit contributions
Fact+2.005
itability+1.920
weights+1.863
amus+1.857
 campaigner+1.764
Neg logit contributions
 Soup-2.077
esville-1.919
eers-1.846
 Uniform-1.839
icho-1.838
Token conservative
Token ID5940
Feature activation-5.30e+00

Pos logit contributions
Fact+2.962
itability+2.838
amus+2.771
weights+2.750
 campaigner+2.648
Neg logit contributions
 Soup-2.913
esville-2.632
eers-2.553
 Uniform-2.543
 Albion-2.529
Token radio
Token ID5243
Feature activation-2.96e+00

Pos logit contributions
Fact+3.039
itability+2.865
weights+2.791
amus+2.779
 campaigner+2.711
Neg logit contributions
 Soup-4.428
esville-4.139
icho-4.031
eers-3.968
 Uniform-3.955
Token host
Token ID2583
Feature activation+1.10e+00

Pos logit contributions
Fact+1.788
itability+1.696
weights+1.637
amus+1.635
 campaigner+1.540
Neg logit contributions
 Soup-2.424
esville-2.262
 Uniform-2.192
eers-2.181
icho-2.176
Token tubing
Token ID43128
Feature activation-2.47e+00

Pos logit contributions
 Soup+0.958
esville+0.865
 Uniform+0.851
eers+0.843
 Albion+0.842
Neg logit contributions
Fact-1.043
itability-0.998
amus-0.970
weights-0.957
quished-0.930
Token,
Token ID11
Feature activation-1.03e+00

Pos logit contributions
Fact+1.880
itability+1.807
weights+1.791
amus+1.770
 campaigner+1.669
Neg logit contributions
 Soup-1.618
esville-1.485
icho-1.412
 Uniform-1.410
eers-1.406
Token features
Token ID3033
Feature activation-2.10e+00

Pos logit contributions
Fact+0.754
itability+0.726
amus+0.703
weights+0.702
quished+0.670
Neg logit contributions
 Soup-0.715
esville-0.652
 Uniform-0.633
eers-0.627
 Albion-0.622
Token a
Token ID257
Feature activation-8.99e-01

Pos logit contributions
Fact+1.373
itability+1.320
weights+1.288
amus+1.276
quished+1.194
Neg logit contributions
 Soup-1.616
esville-1.491
 Uniform-1.441
eers-1.436
icho-1.430
Token full
Token ID1336
Feature activation+5.01e-01

Pos logit contributions
Fact+0.633
itability+0.610
weights+0.594
amus+0.591
quished+0.558
Neg logit contributions
 Soup-0.648
esville-0.595
 Uniform-0.574
eers-0.572
icho-0.568
Token Shim
Token ID31698
Feature activation-1.17e+01

Pos logit contributions
 Soup+0.359
esville+0.329
 Uniform+0.319
eers+0.316
icho+0.314
Neg logit contributions
Fact-0.357
itability-0.342
amus-0.333
weights-0.332
quished-0.315
Tokenano
Token ID5733
Feature activation-3.60e+00

Pos logit contributions
Fact+6.802
weights+6.504
itability+6.378
amus+6.242
anger+6.136
Neg logit contributions
 Soup-8.795
esville-8.510
eers-8.040
 XIV-7.820
 Uniform-7.759
Token 600
Token ID10053
Feature activation-1.24e+00

Pos logit contributions
Fact+2.577
itability+2.478
weights+2.421
amus+2.395
anger+2.284
Neg logit contributions
 Soup-2.503
esville-2.321
eers-2.203
icho-2.187
 Uniform-2.185
Token tri
Token ID1333
Feature activation-3.99e+00

Pos logit contributions
Fact+0.875
itability+0.841
weights+0.819
amus+0.816
quished+0.771
Neg logit contributions
 Soup-0.894
esville-0.821
 Uniform-0.793
eers-0.789
icho-0.783
Token-
Token ID12
Feature activation+1.40e-01

Pos logit contributions
Fact+2.924
itability+2.819
weights+2.791
amus+2.761
anger+2.661
Neg logit contributions
 Soup-2.726
esville-2.501
eers-2.405
 Uniform-2.384
 Albion-2.357
Tokencolor
Token ID8043
Feature activation+5.58e+00

Pos logit contributions
 Soup+0.098
esville+0.090
 Uniform+0.087
eers+0.086
 Albion+0.086
Neg logit contributions
Fact-0.101
itability-0.098
amus-0.095
weights-0.094
quished-0.090
Token
Token ID198
Feature activation-9.04e-01

Pos logit contributions
Fact+0.078
itability+0.073
weights+0.072
amus+0.072
anger+0.067
Neg logit contributions
 Soup-0.082
esville-0.075
 Uniform-0.072
icho-0.072
eers-0.072
TokenFollow
Token ID7155
Feature activation-9.23e-01

Pos logit contributions
Fact+0.680
itability+0.648
weights+0.635
amus+0.633
quished+0.599
Neg logit contributions
 Soup-0.617
esville-0.562
 Uniform-0.544
eers-0.537
icho-0.534
Token @
Token ID2488
Feature activation-1.30e+00

Pos logit contributions
Fact+0.665
itability+0.638
weights+0.624
amus+0.623
ysc+0.588
Neg logit contributions
 Soup-0.648
esville-0.599
 Uniform-0.573
eers-0.572
icho-0.566
TokenPro
Token ID2964
Feature activation-5.09e-01

Pos logit contributions
Fact+0.934
itability+0.893
weights+0.872
amus+0.872
quished+0.823
Neg logit contributions
 Soup-0.927
esville-0.849
 Uniform-0.824
eers-0.814
icho-0.809
Token_
Token ID62
Feature activation-3.73e+00

Pos logit contributions
Fact+0.393
itability+0.377
weights+0.368
amus+0.368
quished+0.349
Neg logit contributions
 Soup-0.336
esville-0.304
 Uniform-0.294
eers-0.292
icho-0.289
TokenCy
Token ID20418
Feature activation-1.17e+01

Pos logit contributions
Fact+2.594
itability+2.390
weights+2.358
amus+2.336
ysc+2.266
Neg logit contributions
 Soup-2.768
esville-2.578
 Uniform-2.507
icho-2.458
 XIV-2.441
Tokencling
Token ID8493
Feature activation-8.47e+00

Pos logit contributions
Fact+2.279
amus+1.750
weights+1.595
itability+1.425
ysc+1.365
Neg logit contributions
 Soup-13.642
esville-12.891
 XIV-12.841
 Albion-12.731
ariat-12.719
Token
Token ID198
Feature activation-3.69e+00

Pos logit contributions
Fact+5.608
amus+5.118
weights+5.094
ysc+4.962
itability+4.923
Neg logit contributions
 Soup-6.242
esville-5.763
icho-5.500
efe-5.436
eers-5.396
Token
Token ID198
Feature activation-2.81e+00

Pos logit contributions
Fact+2.333
weights+2.101
itability+2.092
amus+2.085
anger+1.943
Neg logit contributions
 Soup-2.933
esville-2.724
icho-2.656
ingu-2.634
 Uniform-2.628
TokenTweet
Token ID47845
Feature activation+7.26e-01

Pos logit contributions
Fact+2.018
itability+1.856
amus+1.839
weights+1.834
anger+1.724
Neg logit contributions
 Soup-2.040
esville-1.863
 Uniform-1.823
eers-1.783
icho-1.779
Token
Token ID198
Feature activation-3.25e+00

Pos logit contributions
 Soup+0.481
esville+0.436
eers+0.420
 Uniform+0.420
 Albion+0.414
Neg logit contributions
Fact-0.555
itability-0.536
weights-0.525
amus-0.520
quished-0.495
Token,
Token ID11
Feature activation-2.54e+00

Pos logit contributions
Fact+2.897
itability+2.790
amus+2.734
weights+2.699
anger+2.580
Neg logit contributions
 Soup-2.496
esville-2.322
eers-2.212
 Uniform-2.205
icho-2.194
Token Death
Token ID5830
Feature activation-1.53e+00

Pos logit contributions
Fact+1.825
itability+1.754
amus+1.706
weights+1.702
quished+1.609
Neg logit contributions
 Soup-1.772
esville-1.646
 Uniform-1.577
eers-1.576
icho-1.572
Token,
Token ID11
Feature activation-7.36e-01

Pos logit contributions
Fact+1.166
itability+1.122
weights+1.103
amus+1.093
quished+1.042
Neg logit contributions
 Soup-1.005
esville-0.911
 Uniform-0.886
icho-0.879
eers-0.878
Token Where
Token ID6350
Feature activation-2.82e+00

Pos logit contributions
Fact+0.537
itability+0.517
weights+0.503
amus+0.502
quished+0.475
Neg logit contributions
 Soup-0.515
esville-0.470
 Uniform-0.454
eers-0.451
icho-0.448
Token Is
Token ID1148
Feature activation-3.11e+00

Pos logit contributions
Fact+2.025
itability+1.942
amus+1.915
weights+1.900
anger+1.790
Neg logit contributions
 Soup-1.943
esville-1.831
eers-1.746
icho-1.732
 Uniform-1.728
Token Thy
Token ID31468
Feature activation-1.17e+01

Pos logit contributions
Fact+2.190
itability+2.146
amus+2.068
weights+2.044
anger+1.935
Neg logit contributions
 Soup-2.189
esville-2.049
eers-1.967
 Uniform-1.937
icho-1.937
Token Sting
Token ID35785
Feature activation+9.68e-01

Pos logit contributions
itability+7.994
Fact+7.918
amus+7.735
weights+7.353
anger+7.210
Neg logit contributions
esville-6.905
eers-6.718
icho-6.512
 Soup-6.488
ingu-6.375
Token-
Token ID12
Feature activation+3.04e+00

Pos logit contributions
 Soup+0.749
esville+0.691
 Uniform+0.671
eers+0.666
icho+0.661
Neg logit contributions
Fact-0.636
itability-0.608
weights-0.590
amus-0.587
quished-0.553
Tokena
Token ID64
Feature activation-3.87e-01

Pos logit contributions
 Soup+2.112
esville+1.917
 Uniform+1.850
eers+1.843
icho+1.816
Neg logit contributions
Fact-2.214
itability-2.166
weights-2.100
amus-2.089
quished-1.982
Token-
Token ID12
Feature activation-4.26e+00

Pos logit contributions
Fact+0.300
itability+0.289
amus+0.281
weights+0.281
quished+0.267
Neg logit contributions
 Soup-0.253
esville-0.230
 Uniform-0.221
eers-0.220
icho-0.218
Tokenling
Token ID1359
Feature activation+1.09e+00

Pos logit contributions
Fact+3.170
itability+2.975
weights+2.896
amus+2.873
anger+2.749
Neg logit contributions
 Soup-2.888
esville-2.699
icho-2.614
 Uniform-2.599
eers-2.584
Token
Token ID198
Feature activation+3.04e+00

Pos logit contributions
 Soup+1.129
esville+1.028
 Uniform+0.999
eers+0.993
 Albion+0.981
Neg logit contributions
Fact-1.261
itability-1.209
weights-1.176
amus-1.169
quished-1.121
Token
Token ID198
Feature activation+1.35e+00

Pos logit contributions
 Soup+2.040
esville+1.856
 Uniform+1.795
eers+1.784
 Albion+1.763
Neg logit contributions
Fact-2.279
itability-2.218
weights-2.145
amus-2.145
ccording-2.066
TokenWith
Token ID3152
Feature activation+6.97e-01

Pos logit contributions
 Soup+0.969
esville+0.885
 Uniform+0.860
eers+0.855
 Albion+0.845
Neg logit contributions
Fact-0.945
itability-0.920
weights-0.891
amus-0.887
ysc-0.846
Token a
Token ID257
Feature activation+5.07e-01

Pos logit contributions
 Soup+0.501
esville+0.460
 Uniform+0.446
eers+0.444
 Albion+0.439
Neg logit contributions
Fact-0.495
itability-0.473
weights-0.460
amus-0.459
quished-0.436
Token bare
Token ID6247
Feature activation+1.34e+00

Pos logit contributions
 Soup+0.365
esville+0.334
 Uniform+0.325
eers+0.323
icho+0.320
Neg logit contributions
Fact-0.360
itability-0.345
weights-0.336
amus-0.334
quished-0.318
Token bod
Token ID16429
Feature activation-1.16e+01

Pos logit contributions
 Soup+0.980
esville+0.902
 Uniform+0.877
eers+0.871
 Albion+0.863
Neg logit contributions
Fact-0.928
itability-0.889
weights-0.862
amus-0.857
quished-0.818
Tokenkin
Token ID5116
Feature activation+4.62e-01

Pos logit contributions
itability+7.595
Fact+7.502
amus+7.386
weights+7.152
ysc+7.136
Neg logit contributions
 Soup-8.038
esville-7.367
eers-7.131
 Albion-7.098
 Uniform-6.966
Token?
Token ID30
Feature activation+2.11e+00

Pos logit contributions
 Soup+0.312
esville+0.284
 Uniform+0.275
eers+0.273
icho+0.270
Neg logit contributions
Fact-0.349
itability-0.336
weights-0.327
amus-0.326
quished-0.310
Token who
Token ID508
Feature activation+2.08e+00

Pos logit contributions
 Soup+1.470
esville+1.331
 Uniform+1.300
eers+1.279
 Albion+1.276
Neg logit contributions
Fact-1.543
itability-1.494
weights-1.447
amus-1.442
quished-1.378
Token would
Token ID561
Feature activation-2.85e+00

Pos logit contributions
 Soup+1.499
esville+1.379
 Uniform+1.331
eers+1.326
 Albion+1.313
Neg logit contributions
Fact-1.472
itability-1.410
weights-1.367
amus-1.362
quished-1.301
Token f
Token ID277
Feature activation+2.94e+00

Pos logit contributions
Fact+1.901
itability+1.844
amus+1.791
weights+1.789
 campaigner+1.683
Neg logit contributions
 Soup-2.140
esville-1.961
 Uniform-1.900
icho-1.882
eers-1.876
Token between
Token ID1022
Feature activation-4.03e+00

Pos logit contributions
Fact+0.901
itability+0.871
weights+0.853
amus+0.848
quished+0.810
Neg logit contributions
 Soup-0.632
esville-0.568
 Uniform-0.544
eers-0.540
icho-0.536
Token Russian
Token ID3394
Feature activation-2.28e+00

Pos logit contributions
Fact+2.671
itability+2.564
weights+2.505
amus+2.473
 campaigner+2.340
Neg logit contributions
 Soup-3.051
esville-2.818
 Uniform-2.725
icho-2.709
eers-2.705
Token hackers
Token ID17110
Feature activation-7.34e+00

Pos logit contributions
Fact+1.322
itability+1.254
weights+1.221
amus+1.204
 campaigner+1.121
Neg logit contributions
 Soup-1.933
esville-1.803
 Uniform-1.749
eers-1.733
icho-1.729
Token and
Token ID290
Feature activation-3.19e+00

Pos logit contributions
Fact+4.520
itability+4.339
weights+4.251
amus+4.165
 campaigner+3.985
Neg logit contributions
 Soup-5.725
esville-5.376
icho-5.159
 Uniform-5.142
eers-5.081
Token the
Token ID262
Feature activation-3.60e+00

Pos logit contributions
Fact+2.333
itability+2.246
weights+2.194
amus+2.179
quished+2.071
Neg logit contributions
 Soup-2.214
esville-2.023
 Uniform-1.953
eers-1.942
icho-1.932
Token DNC
Token ID17824
Feature activation-1.16e+01

Pos logit contributions
Fact+2.315
itability+2.223
weights+2.165
amus+2.138
quished+2.024
Neg logit contributions
 Soup-2.792
esville-2.594
icho-2.500
eers-2.496
 Uniform-2.495
Token leak
Token ID13044
Feature activation-4.23e+00

Pos logit contributions
Fact+6.491
itability+6.258
weights+6.044
 campaigner+5.816
 hinges+5.805
Neg logit contributions
 Soup-8.837
esville-8.521
icho-8.354
 Albion-8.180
 Uniform-7.963
Token isn
Token ID2125
Feature activation-2.11e+00

Pos logit contributions
Fact+2.966
itability+2.833
weights+2.797
amus+2.756
 campaigner+2.614
Neg logit contributions
 Soup-3.010
esville-2.787
icho-2.678
 Uniform-2.676
eers-2.659
Token't
Token ID470
Feature activation-7.55e+00

Pos logit contributions
Fact+2.079
itability+1.991
amus+1.970
weights+1.958
anger+1.875
Neg logit contributions
 Soup-0.949
esville-0.809
 Uniform-0.780
icho-0.769
eers-0.758
Token a
Token ID257
Feature activation-5.37e+00

Pos logit contributions
Fact+5.316
itability+4.945
weights+4.841
amus+4.789
 campaigner+4.516
Neg logit contributions
 Soup-5.256
esville-4.958
eers-4.685
icho-4.670
 Uniform-4.660
Token '
Token ID705
Feature activation-3.82e+00

Pos logit contributions
Fact+3.172
itability+2.983
amus+2.813
weights+2.810
 campaigner+2.658
Neg logit contributions
 Soup-4.453
esville-4.207
eers-4.023
icho-4.006
 Uniform-4.000
Token,
Token ID11
Feature activation-4.38e-01

Pos logit contributions
Fact+0.274
itability+0.263
weights+0.257
amus+0.257
quished+0.243
Neg logit contributions
 Soup-0.232
esville-0.211
 Uniform-0.203
eers-0.201
icho-0.200
Token from
Token ID422
Feature activation-1.38e+00

Pos logit contributions
Fact+0.322
itability+0.309
weights+0.302
amus+0.301
quished+0.285
Neg logit contributions
 Soup-0.304
esville-0.278
 Uniform-0.268
eers-0.267
icho-0.265
Token £
Token ID4248
Feature activation-4.44e-01

Pos logit contributions
Fact+0.990
itability+0.952
weights+0.930
amus+0.926
quished+0.873
Neg logit contributions
 Soup-0.972
esville-0.892
 Uniform-0.859
eers-0.854
icho-0.850
Token9
Token ID24
Feature activation-3.31e-01

Pos logit contributions
Fact+0.315
itability+0.303
weights+0.294
amus+0.294
quished+0.278
Neg logit contributions
 Soup-0.320
esville-0.293
 Uniform-0.284
eers-0.282
icho-0.280
Token79
Token ID3720
Feature activation-2.36e+00

Pos logit contributions
Fact+0.230
itability+0.220
weights+0.214
amus+0.214
quished+0.202
Neg logit contributions
 Soup-0.244
esville-0.224
 Uniform-0.217
eers-0.216
icho-0.214
Token all
Token ID477
Feature activation-1.15e+01

Pos logit contributions
Fact+1.721
itability+1.657
weights+1.622
amus+1.617
quished+1.526
Neg logit contributions
 Soup-1.647
esville-1.505
 Uniform-1.441
eers-1.435
icho-1.430
Token-
Token ID12
Feature activation-3.79e+00

Pos logit contributions
Fact+7.037
weights+6.806
itability+6.728
amus+6.554
 campaigner+6.377
Neg logit contributions
 Soup-8.361
esville-7.776
icho-7.511
 Uniform-7.351
efe-7.346
Tokenin
Token ID259
Feature activation-2.55e+00

Pos logit contributions
Fact+3.784
itability+3.665
weights+3.627
amus+3.589
anger+3.473
Neg logit contributions
 Soup-1.619
esville-1.380
 Uniform-1.298
icho-1.262
eers-1.258
Tokenclusive
Token ID5731
Feature activation-1.73e+00

Pos logit contributions
Fact+2.099
itability+2.049
amus+1.987
weights+1.986
quished+1.898
Neg logit contributions
 Soup-1.514
esville-1.359
 Uniform-1.298
eers-1.290
icho-1.290
Token,
Token ID11
Feature activation-2.16e+00

Pos logit contributions
Fact+1.126
itability+1.078
weights+1.054
amus+1.049
quished+0.986
Neg logit contributions
 Soup-1.339
esville-1.235
 Uniform-1.191
icho-1.186
eers-1.181
Token departing
Token ID37717
Feature activation-2.29e+00

Pos logit contributions
Fact+1.547
itability+1.483
weights+1.453
amus+1.442
quished+1.363
Neg logit contributions
 Soup-1.524
esville-1.403
 Uniform-1.343
eers-1.339
icho-1.336
TokenMedia
Token ID13152
Feature activation-1.79e+00

Pos logit contributions
Fact+1.440
itability+1.353
weights+1.322
amus+1.312
anger+1.242
Neg logit contributions
 Soup-1.449
esville-1.332
 Uniform-1.297
icho-1.276
eers-1.272
Token playback
Token ID16388
Feature activation-1.67e+00

Pos logit contributions
Fact+1.328
itability+1.270
amus+1.236
weights+1.235
 campaigner+1.171
Neg logit contributions
 Soup-1.232
esville-1.131
 Uniform-1.085
eers-1.075
icho-1.072
Token is
Token ID318
Feature activation+3.59e-01

Pos logit contributions
Fact+1.143
itability+1.091
amus+1.076
weights+1.069
ysc+1.000
Neg logit contributions
 Soup-1.236
esville-1.135
icho-1.098
 Uniform-1.094
eers-1.093
Token unsupported
Token ID24222
Feature activation-4.69e+00

Pos logit contributions
 Soup+0.239
esville+0.218
 Uniform+0.209
eers+0.209
icho+0.207
Neg logit contributions
Fact-0.274
itability-0.264
weights-0.257
amus-0.256
ysc-0.243
Token on
Token ID319
Feature activation-4.56e+00

Pos logit contributions
Fact+3.227
itability+2.988
weights+2.960
amus+2.942
ysc+2.781
Neg logit contributions
 Soup-3.413
esville-3.225
icho-3.148
ingu-3.078
 Uniform-3.033
Token your
Token ID534
Feature activation-1.15e+01

Pos logit contributions
Fact+3.047
itability+2.857
weights+2.838
amus+2.769
anger+2.662
Neg logit contributions
 Soup-3.385
icho-3.098
esville-3.095
 Uniform-3.054
efe-3.009
Token device
Token ID3335
Feature activation-6.56e+00

Pos logit contributions
Fact+4.224
itability+3.786
weights+3.778
amus+3.658
 campaigner+3.636
Neg logit contributions
 Soup-10.822
icho-10.618
esville-10.329
efe-10.137
 Arche-10.007
Token Media
Token ID6343
Feature activation-7.51e-01

Pos logit contributions
Fact+3.433
weights+3.070
amus+3.069
itability+3.039
kson+2.774
Neg logit contributions
 Soup-5.579
esville-5.466
icho-5.379
ingu-5.245
efe-5.167
Token caption
Token ID8305
Feature activation-2.50e+00

Pos logit contributions
Fact+0.525
itability+0.506
weights+0.490
amus+0.487
ysc+0.458
Neg logit contributions
 Soup-0.551
esville-0.508
 Uniform-0.491
icho-0.488
efe-0.483
Token Prof
Token ID4415
Feature activation-2.07e+00

Pos logit contributions
Fact+1.732
itability+1.643
weights+1.601
amus+1.594
quished+1.505
Neg logit contributions
 Soup-1.819
esville-1.696
eers-1.631
icho-1.626
 Uniform-1.617
Token Re
Token ID797
Feature activation-2.84e+00

Pos logit contributions
Fact+1.317
itability+1.271
weights+1.224
amus+1.218
quished+1.146
Neg logit contributions
 Soup-1.624
esville-1.505
 Uniform-1.461
eers-1.456
icho-1.446
Token two
Token ID734
Feature activation-3.09e+00

Pos logit contributions
Fact+2.192
itability+2.089
weights+2.062
amus+2.023
quished+1.910
Neg logit contributions
 Soup-2.540
esville-2.361
eers-2.268
 Uniform-2.258
icho-2.253
Token basketball
Token ID9669
Feature activation+6.46e-01

Pos logit contributions
Fact+2.114
itability+2.035
weights+2.031
amus+1.965
 campaigner+1.869
Neg logit contributions
 Soup-2.242
esville-2.085
icho-1.985
 Uniform-1.980
efe-1.975
Token goals
Token ID4661
Feature activation-2.57e+00

Pos logit contributions
 Soup+0.452
esville+0.413
 Uniform+0.400
eers+0.397
icho+0.394
Neg logit contributions
Fact-0.470
itability-0.453
weights-0.441
amus-0.439
quished-0.416
Token on
Token ID319
Feature activation-1.60e+00

Pos logit contributions
Fact+1.891
itability+1.820
weights+1.793
amus+1.781
ysc+1.689
Neg logit contributions
 Soup-1.761
esville-1.614
 Uniform-1.542
eers-1.535
icho-1.524
Token opposite
Token ID6697
Feature activation-4.90e+00

Pos logit contributions
Fact+1.076
itability+1.032
weights+1.008
amus+0.999
quished+0.943
Neg logit contributions
 Soup-1.208
esville-1.114
 Uniform-1.076
eers-1.072
icho-1.065
Token ends
Token ID5645
Feature activation-1.15e+01

Pos logit contributions
Fact+2.327
weights+2.213
itability+2.162
amus+2.070
 hinges+1.959
Neg logit contributions
 Soup-4.541
esville-4.277
efe-4.126
 Uniform-4.105
icho-4.095
Token stand
Token ID1302
Feature activation+1.47e+00

Pos logit contributions
Fact+7.439
weights+7.254
amus+7.229
itability+7.113
quished+6.966
Neg logit contributions
 Soup-7.694
esville-7.028
eers-6.840
 Uniform-6.748
icho-6.733
Token over
Token ID625
Feature activation-1.53e+00

Pos logit contributions
 Soup+0.937
esville+0.846
 Uniform+0.819
eers+0.813
icho+0.806
Neg logit contributions
Fact-1.162
itability-1.124
amus-1.089
weights-1.089
quished-1.036
Token an
Token ID281
Feature activation-2.44e+00

Pos logit contributions
Fact+1.016
itability+0.973
weights+0.949
amus+0.943
quished+0.887
Neg logit contributions
 Soup-1.173
esville-1.082
 Uniform-1.047
eers-1.043
icho-1.037
Token empty
Token ID6565
Feature activation-8.88e-01

Pos logit contributions
Fact+1.728
itability+1.661
weights+1.629
amus+1.604
quished+1.527
Neg logit contributions
 Soup-1.734
esville-1.587
eers-1.534
 Uniform-1.526
icho-1.525
Token court
Token ID2184
Feature activation-8.48e-01

Pos logit contributions
Fact+0.601
itability+0.577
weights+0.563
amus+0.560
quished+0.526
Neg logit contributions
 Soup-0.663
esville-0.612
 Uniform-0.590
eers-0.588
icho-0.586
Token see
Token ID766
Feature activation+5.66e+00

Pos logit contributions
 Soup+3.551
esville+3.154
 Uniform+3.133
 Albion+3.078
eers+2.996
Neg logit contributions
Fact-4.096
itability-3.880
weights-3.841
amus-3.818
quished-3.705
Token her
Token ID607
Feature activation+2.67e+00

Pos logit contributions
 Soup+4.229
 Uniform+3.727
esville+3.722
 Albion+3.705
 Arche+3.682
Neg logit contributions
Fact-3.875
itability-3.640
amus-3.588
weights-3.577
quished-3.468
Token GP
Token ID14714
Feature activation-5.44e-01

Pos logit contributions
 Soup+1.995
esville+1.825
 Uniform+1.771
eers+1.748
 Albion+1.743
Neg logit contributions
Fact-1.835
itability-1.736
weights-1.694
amus-1.690
quished-1.617
Token,
Token ID11
Feature activation+2.33e+00

Pos logit contributions
Fact+0.435
itability+0.421
weights+0.411
amus+0.410
quished+0.390
Neg logit contributions
 Soup-0.341
esville-0.308
 Uniform-0.297
eers-0.295
icho-0.292
Token the
Token ID262
Feature activation+2.93e+00

Pos logit contributions
 Soup+1.768
esville+1.602
 Uniform+1.574
 Albion+1.561
eers+1.546
Neg logit contributions
Fact-1.563
itability-1.500
amus-1.466
weights-1.459
quished-1.389
Token child
Token ID1200
Feature activation+8.35e+00

Pos logit contributions
 Soup+2.183
esville+1.990
 Uniform+1.942
eers+1.919
 Albion+1.918
Neg logit contributions
Fact-2.006
itability-1.914
amus-1.875
weights-1.863
quished-1.790
Token stated
Token ID5081
Feature activation-1.03e-01

Pos logit contributions
 Soup+4.947
esville+4.580
eers+4.411
 Uniform+4.312
 Arche+4.267
Neg logit contributions
Fact-6.618
itability-6.345
amus-6.116
weights-6.061
quished-6.002
Token clearly
Token ID4084
Feature activation+4.90e+00

Pos logit contributions
Fact+0.071
itability+0.068
weights+0.066
amus+0.066
quished+0.063
Neg logit contributions
 Soup-0.076
esville-0.070
 Uniform-0.068
eers-0.067
icho-0.067
Token that
Token ID326
Feature activation+1.28e+00

Pos logit contributions
 Soup+3.485
esville+3.136
 Albion+3.090
 Uniform+3.080
icho+3.048
Neg logit contributions
Fact-3.464
itability-3.311
amus-3.233
weights-3.221
quished-3.100
Token she
Token ID673
Feature activation+2.13e+00

Pos logit contributions
 Soup+0.993
esville+0.908
 Uniform+0.886
eers+0.879
 Albion+0.875
Neg logit contributions
Fact-0.843
itability-0.807
weights-0.784
amus-0.783
quished-0.742
Token had
Token ID550
Feature activation-2.42e+00

Pos logit contributions
 Soup+1.337
esville+1.210
 Uniform+1.164
eers+1.163
 Albion+1.148
Neg logit contributions
Fact-1.704
itability-1.642
weights-1.593
amus-1.592
quished-1.540
Token-
Token ID12
Feature activation+2.24e+00

Pos logit contributions
 Soup+2.466
esville+2.237
 Uniform+2.160
 Albion+2.094
icho+2.093
Neg logit contributions
Fact-3.534
itability-3.413
weights-3.320
amus-3.292
quished-3.198
Tokenhour
Token ID9769
Feature activation-1.63e+00

Pos logit contributions
 Soup+1.640
esville+1.511
 Uniform+1.451
icho+1.439
 Albion+1.435
Neg logit contributions
Fact-1.529
itability-1.477
amus-1.425
weights-1.422
quished-1.346
Token Turkey
Token ID7137
Feature activation-1.70e+00

Pos logit contributions
Fact+1.073
itability+1.032
weights+1.007
amus+1.000
quished+0.942
Neg logit contributions
 Soup-1.243
esville-1.151
eers-1.108
 Uniform-1.105
icho-1.096
Token-
Token ID12
Feature activation+4.58e+00

Pos logit contributions
Fact+1.263
itability+1.212
weights+1.185
amus+1.180
quished+1.122
Neg logit contributions
 Soup-1.154
esville-1.063
 Uniform-1.020
eers-1.019
icho-1.004
Tokenbacked
Token ID17078
Feature activation+3.97e+00

Pos logit contributions
 Soup+2.858
esville+2.577
icho+2.484
eers+2.473
 Uniform+2.453
Neg logit contributions
Fact-3.508
itability-3.477
weights-3.339
amus-3.328
quished-3.196
Token offensive
Token ID5859
Feature activation+8.04e+00

Pos logit contributions
 Soup+2.850
esville+2.556
 Uniform+2.554
 Albion+2.512
icho+2.505
Neg logit contributions
Fact-2.816
itability-2.707
amus-2.621
weights-2.610
quished-2.489
Token.
Token ID13
Feature activation+2.37e+00

Pos logit contributions
 Soup+4.910
esville+4.358
 Albion+4.312
icho+4.290
eers+4.253
Neg logit contributions
Fact-6.261
itability-6.048
amus-5.867
weights-5.842
quished-5.631
Token AFP
Token ID8916
Feature activation-1.34e+00

Pos logit contributions
 Soup+1.650
esville+1.497
 Uniform+1.465
 Albion+1.439
eers+1.438
Neg logit contributions
Fact-1.723
itability-1.682
weights-1.632
amus-1.627
ysc-1.537
Token has
Token ID468
Feature activation-5.94e+00

Pos logit contributions
Fact+0.964
itability+0.918
weights+0.896
amus+0.893
quished+0.846
Neg logit contributions
 Soup-0.955
esville-0.875
 Uniform-0.845
eers-0.840
icho-0.836
Token more
Token ID517
Feature activation-1.91e+00

Pos logit contributions
Fact+4.075
itability+3.839
weights+3.718
amus+3.709
 campaigner+3.490
Neg logit contributions
 Soup-4.338
esville-3.948
icho-3.855
eers-3.834
 Uniform-3.803
Token from
Token ID422
Feature activation-2.90e+00

Pos logit contributions
Fact+1.552
itability+1.489
weights+1.467
amus+1.446
quished+1.382
Neg logit contributions
 Soup-1.182
esville-1.071
icho-1.019
eers-1.017
 Uniform-1.017
Token |
Token ID930
Feature activation+4.40e+00

Pos logit contributions
 Soup+0.410
esville+0.375
eers+0.358
 Uniform+0.356
icho+0.355
Neg logit contributions
Fact-0.514
itability-0.490
amus-0.481
weights-0.481
anger-0.455
Token int
Token ID493
Feature activation+1.66e+00

Pos logit contributions
 Soup+2.227
 Uniform+1.914
esville+1.901
icho+1.818
eers+1.802
Neg logit contributions
Fact-4.017
itability-3.959
amus-3.840
weights-3.832
quished-3.656
Token f
Token ID277
Feature activation-2.42e+00

Pos logit contributions
 Soup+1.329
esville+1.231
 Uniform+1.199
eers+1.193
icho+1.183
Neg logit contributions
Fact-1.033
itability-0.990
amus-0.961
weights-0.956
quished-0.897
Tokend
Token ID67
Feature activation+1.89e+00

Pos logit contributions
Fact+1.295
itability+1.217
weights+1.206
amus+1.180
anger+1.097
Neg logit contributions
 Soup-2.174
esville-1.980
 Uniform-1.962
eers-1.953
 Albion-1.952
Token |
Token ID930
Feature activation+5.28e+00

Pos logit contributions
 Soup+1.169
esville+1.057
 Uniform+1.014
eers+1.010
icho+1.001
Neg logit contributions
Fact-1.529
itability-1.470
weights-1.439
amus-1.437
quished-1.361
Token +
Token ID1343
Feature activation+8.68e+00

Pos logit contributions
 Soup+2.238
esville+1.826
 Uniform+1.808
eers+1.725
icho+1.705
Neg logit contributions
Fact-5.253
itability-5.207
amus-5.049
weights-5.035
quished-4.811
Token--------------------
Token ID19351
Feature activation+7.04e+00

Pos logit contributions
 Soup+5.975
esville+5.441
eers+5.254
efe+5.232
icho+5.226
Neg logit contributions
Fact-5.941
itability-5.904
ccording-5.833
amus-5.735
weights-5.664
Token--+
Token ID44785
Feature activation+3.59e+00

Pos logit contributions
 Soup+1.993
esville+1.647
eers+1.462
 Uniform+1.447
icho+1.428
Neg logit contributions
Fact-7.871
itability-7.710
weights-7.596
amus-7.547
ccording-7.451
Token |
Token ID930
Feature activation+4.79e+00

Pos logit contributions
 Soup+2.140
esville+1.911
 Uniform+1.849
eers+1.822
icho+1.803
Neg logit contributions
Fact-2.987
itability-2.905
weights-2.832
amus-2.821
quished-2.707
Token [
Token ID685
Feature activation+2.80e+00

Pos logit contributions
 Soup+3.246
 Uniform+2.870
esville+2.865
eers+2.770
icho+2.757
Neg logit contributions
Fact-3.564
itability-3.517
weights-3.386
amus-3.368
quished-3.163
Tokenaddress
Token ID21975
Feature activation+7.68e-01

Pos logit contributions
 Soup+1.790
esville+1.631
 Uniform+1.585
eers+1.561
icho+1.544
Neg logit contributions
Fact-2.175
itability-2.109
weights-2.038
amus-2.034
quished-1.962
Token fram
Token ID5346
Feature activation-1.57e+00

Pos logit contributions
 Soup+1.256
esville+1.129
 Uniform+1.091
eers+1.085
icho+1.072
Neg logit contributions
Fact-1.610
itability-1.556
amus-1.516
weights-1.513
quished-1.454
Tokenwork
Token ID1818
Feature activation-8.51e-02

Pos logit contributions
Fact+1.201
itability+1.157
weights+1.129
amus+1.128
ysc+1.068
Neg logit contributions
 Soup-1.037
esville-0.940
 Uniform-0.909
eers-0.899
 Albion-0.895
Token using
Token ID1262
Feature activation-1.94e+00

Pos logit contributions
Fact+0.064
itability+0.061
weights+0.060
amus+0.060
quished+0.057
Neg logit contributions
 Soup-0.058
esville-0.053
 Uniform-0.051
eers-0.051
icho-0.050
Token the
Token ID262
Feature activation+2.65e-01

Pos logit contributions
Fact+1.414
itability+1.355
weights+1.332
amus+1.329
quished+1.256
Neg logit contributions
 Soup-1.336
esville-1.227
 Uniform-1.175
eers-1.171
icho-1.169
Token command
Token ID3141
Feature activation+6.37e+00

Pos logit contributions
 Soup+0.173
esville+0.157
 Uniform+0.151
eers+0.150
icho+0.149
Neg logit contributions
Fact-0.206
itability-0.199
weights-0.194
amus-0.193
quished-0.184
Token server
Token ID4382
Feature activation+7.25e+00

Pos logit contributions
 Soup+4.155
esville+3.697
eers+3.672
 Uniform+3.630
 XIV+3.516
Neg logit contributions
Fact-4.888
itability-4.756
amus-4.528
weights-4.525
quished-4.312
Tokenless
Token ID1203
Feature activation+2.48e+00

Pos logit contributions
 Soup+2.258
esville+1.896
eers+1.722
 Uniform+1.713
efe+1.604
Neg logit contributions
Fact-7.722
itability-7.571
amus-7.431
quished-7.417
weights-7.192
Token or
Token ID393
Feature activation+1.37e+00

Pos logit contributions
 Soup+1.668
esville+1.517
 Uniform+1.470
eers+1.456
icho+1.440
Neg logit contributions
Fact-1.852
itability-1.799
amus-1.748
weights-1.736
quished-1.673
Token sl
Token ID1017
Feature activation+3.10e+00

Pos logit contributions
 Soup+0.966
esville+0.881
 Uniform+0.854
eers+0.847
icho+0.839
Neg logit contributions
Fact-0.995
itability-0.957
weights-0.929
amus-0.926
quished-0.878
Tokens
Token ID82
Feature activation+2.82e-01

Pos logit contributions
 Soup+1.551
esville+1.338
eers+1.276
icho+1.260
 Uniform+1.252
Neg logit contributions
Fact-2.862
itability-2.785
amus-2.712
weights-2.699
quished-2.607
Token.
Token ID764
Feature activation-6.89e-01

Pos logit contributions
 Soup+0.200
esville+0.183
 Uniform+0.177
eers+0.176
icho+0.175
Neg logit contributions
Fact-0.203
itability-0.195
weights-0.190
amus-0.189
quished-0.180
Token 15
Token ID1315
Feature activation+2.00e+00

Pos logit contributions
 Soup+1.845
esville+1.664
 Uniform+1.601
eers+1.586
icho+1.573
Neg logit contributions
Fact-2.403
itability-2.330
weights-2.267
amus-2.264
quished-2.159
Token |
Token ID930
Feature activation+1.89e+00

Pos logit contributions
 Soup+1.218
esville+1.097
 Uniform+1.054
eers+1.047
icho+1.037
Neg logit contributions
Fact-1.643
itability-1.586
weights-1.550
amus-1.546
quished-1.474
Token org
Token ID8745
Feature activation+4.96e+00

Pos logit contributions
 Soup+1.358
esville+1.234
 Uniform+1.211
eers+1.194
icho+1.182
Neg logit contributions
Fact-1.348
itability-1.301
weights-1.259
amus-1.256
quished-1.184
Token.
Token ID764
Feature activation-1.30e+00

Pos logit contributions
 Soup+2.508
esville+2.226
 Uniform+2.114
eers+2.108
 Albion+2.070
Neg logit contributions
Fact-4.485
itability-4.399
ccording-4.300
weights-4.272
amus-4.252
Token cl
Token ID537
Feature activation+4.31e+00

Pos logit contributions
Fact+0.741
itability+0.706
weights+0.685
amus+0.677
 campaigner+0.632
Neg logit contributions
 Soup-1.105
esville-1.035
 Uniform-0.999
icho-0.994
eers-0.991
Tokenojure
Token ID32511
Feature activation+7.91e+00

Pos logit contributions
 Soup+1.652
esville+1.403
icho+1.318
eers+1.312
 Uniform+1.270
Neg logit contributions
Fact-4.468
itability-4.310
weights-4.250
amus-4.239
ccording-4.150
Token /
Token ID1220
Feature activation+2.70e-01

Pos logit contributions
 Soup+3.557
esville+3.088
 Uniform+2.977
eers+2.890
 Albion+2.858
Neg logit contributions
Fact-7.483
ccording-7.421
itability-7.335
weights-7.149
amus-7.110
Token tools
Token ID4899
Feature activation+4.62e+00

Pos logit contributions
 Soup+0.192
esville+0.176
 Uniform+0.170
eers+0.169
icho+0.168
Neg logit contributions
Fact-0.195
itability-0.187
weights-0.182
amus-0.181
quished-0.172
Token.
Token ID764
Feature activation-1.56e-01

Pos logit contributions
 Soup+2.416
esville+2.113
 Uniform+2.041
eers+2.020
icho+1.985
Neg logit contributions
Fact-4.158
itability-4.049
weights-3.945
amus-3.940
quished-3.824
Token reader
Token ID9173
Feature activation+6.19e-01

Pos logit contributions
Fact+0.111
itability+0.107
weights+0.104
amus+0.103
quished+0.098
Neg logit contributions
 Soup-0.112
esville-0.103
 Uniform-0.100
eers-0.099
icho-0.098
Token |
Token ID930
Feature activation+1.13e+00

Pos logit contributions
 Soup+0.375
esville+0.336
 Uniform+0.325
eers+0.321
icho+0.317
Neg logit contributions
Fact-0.509
itability-0.494
amus-0.482
weights-0.479
quished-0.460
Token reported
Token ID2098
Feature activation+3.49e-01

Pos logit contributions
 Soup+4.035
esville+3.649
 Uniform+3.588
 Albion+3.532
eers+3.477
Neg logit contributions
Fact-4.189
itability-4.066
amus-3.959
weights-3.921
quished-3.798
Token as
Token ID355
Feature activation-2.52e+00

Pos logit contributions
 Soup+0.230
esville+0.209
 Uniform+0.201
eers+0.200
icho+0.198
Neg logit contributions
Fact-0.270
itability-0.260
weights-0.253
amus-0.253
quished-0.240
Token abstract
Token ID12531
Feature activation+6.07e+00

Pos logit contributions
Fact+1.821
itability+1.752
weights+1.708
amus+1.685
quished+1.590
Neg logit contributions
 Soup-1.762
esville-1.635
eers-1.572
efe-1.554
 Uniform-1.554
Tokens
Token ID82
Feature activation+2.20e+00

Pos logit contributions
 Soup+3.525
esville+3.144
 Uniform+3.049
eers+3.009
 Albion+2.957
Neg logit contributions
Fact-5.038
itability-4.892
amus-4.786
weights-4.720
ysc-4.573
Token.
Token ID13
Feature activation-2.32e-01

Pos logit contributions
 Soup+1.397
esville+1.262
 Uniform+1.219
eers+1.203
icho+1.195
Neg logit contributions
Fact-1.745
itability-1.684
weights-1.644
amus-1.639
quished-1.569
Token<|endoftext|>
Token ID50256
Feature activation+5.18e+00

Pos logit contributions
Fact+0.159
itability+0.152
weights+0.148
amus+0.147
quished+0.138
Neg logit contributions
 Soup-0.173
esville-0.160
eers-0.154
 Uniform-0.154
icho-0.153
TokenA
Token ID32
Feature activation+3.85e+00

Pos logit contributions
 Soup+3.491
esville+3.227
eers+3.125
icho+3.064
 Uniform+3.055
Neg logit contributions
itability-3.630
Fact-3.583
amus-3.506
weights-3.462
 campaigner-3.362
Token group
Token ID1448
Feature activation+3.52e+00

Pos logit contributions
 Soup+2.799
esville+2.572
 Uniform+2.505
eers+2.458
icho+2.449
Neg logit contributions
Fact-2.685
itability-2.557
weights-2.518
amus-2.479
quished-2.367
Token of
Token ID286
Feature activation-1.98e+00

Pos logit contributions
 Soup+2.379
esville+2.165
 Uniform+2.101
eers+2.089
 Albion+2.051
Neg logit contributions
Fact-2.605
itability-2.535
amus-2.444
weights-2.443
quished-2.339
Token Ottawa
Token ID14074
Feature activation-4.24e+00

Pos logit contributions
Fact+1.391
itability+1.346
weights+1.319
amus+1.302
 campaigner+1.236
Neg logit contributions
 Soup-1.417
esville-1.308
 Uniform-1.252
icho-1.250
eers-1.249
Token fitness
Token ID13547
Feature activation-5.62e+00

Pos logit contributions
Fact+2.921
itability+2.802
weights+2.761
amus+2.739
 campaigner+2.601
Neg logit contributions
 Soup-3.028
esville-2.796
 Uniform-2.703
eers-2.684
efe-2.675
Tokens
Token ID82
Feature activation-6.89e-01

Pos logit contributions
 Soup+1.638
esville+1.439
 Uniform+1.375
eers+1.367
icho+1.341
Neg logit contributions
Fact-2.840
itability-2.777
amus-2.714
weights-2.666
quished-2.611
Token or
Token ID393
Feature activation-6.11e-01

Pos logit contributions
Fact+0.556
itability+0.536
weights+0.525
amus+0.523
quished+0.497
Neg logit contributions
 Soup-0.428
esville-0.387
 Uniform-0.372
eers-0.370
icho-0.366
Token adject
Token ID31129
Feature activation+1.35e+00

Pos logit contributions
Fact+0.428
itability+0.411
weights+0.400
amus+0.399
quished+0.377
Neg logit contributions
 Soup-0.445
esville-0.408
 Uniform-0.395
eers-0.393
icho-0.390
Tokenives
Token ID1083
Feature activation+3.62e+00

Pos logit contributions
 Soup+0.721
esville+0.640
 Uniform+0.613
eers+0.603
icho+0.601
Neg logit contributions
Fact-1.208
itability-1.168
weights-1.147
amus-1.143
ysc-1.089
Token,
Token ID11
Feature activation+2.71e+00

Pos logit contributions
 Soup+2.128
esville+1.899
 Uniform+1.831
eers+1.820
 Albion+1.813
Neg logit contributions
Fact-2.992
itability-2.940
amus-2.846
weights-2.826
quished-2.748
Token representing
Token ID10200
Feature activation+1.46e+00

Pos logit contributions
 Soup+1.906
esville+1.728
 Uniform+1.699
 Albion+1.664
icho+1.660
Neg logit contributions
Fact-1.957
itability-1.901
amus-1.850
weights-1.829
quished-1.762
Token qualities
Token ID14482
Feature activation-1.13e+00

Pos logit contributions
 Soup+1.067
esville+0.977
 Uniform+0.950
eers+0.943
icho+0.935
Neg logit contributions
Fact-1.016
itability-0.978
amus-0.950
weights-0.947
quished-0.901
Token such
Token ID884
Feature activation+1.21e+00

Pos logit contributions
Fact+0.843
itability+0.808
weights+0.789
amus+0.784
anger+0.742
Neg logit contributions
 Soup-0.771
esville-0.704
 Uniform-0.678
eers-0.676
icho-0.671
Token as
Token ID355
Feature activation+1.53e-01

Pos logit contributions
 Soup+0.863
esville+0.791
 Uniform+0.764
eers+0.760
icho+0.755
Neg logit contributions
Fact-0.866
itability-0.830
weights-0.810
amus-0.807
quished-0.760
Token solid
Token ID4735
Feature activation-2.16e+00

Pos logit contributions
 Soup+0.105
esville+0.096
 Uniform+0.093
eers+0.092
icho+0.091
Neg logit contributions
Fact-0.113
itability-0.109
weights-0.106
amus-0.105
quished-0.100
Tokenity
Token ID414
Feature activation-2.65e+00

Pos logit contributions
Fact+1.567
itability+1.520
weights+1.476
amus+1.466
ysc+1.384
Neg logit contributions
 Soup-1.501
esville-1.387
eers-1.333
 Uniform-1.330
icho-1.319
Token Co
Token ID1766
Feature activation+5.24e+00

Pos logit contributions
Fact+0.160
itability+0.152
weights+0.149
amus+0.148
 campaigner+0.138
Neg logit contributions
 Soup-0.171
esville-0.159
eers-0.154
icho-0.153
 Uniform-0.152
Token-
Token ID12
Feature activation+5.94e+00

Pos logit contributions
 Soup+2.933
esville+2.564
 Albion+2.513
 Uniform+2.508
efe+2.478
Neg logit contributions
itability-4.336
Fact-4.320
amus-4.251
weights-4.199
quished-4.149
Tokenoperation
Token ID27184
Feature activation+2.05e+00

Pos logit contributions
 Soup+2.971
esville+2.556
 Uniform+2.460
 Albion+2.449
efe+2.399
Neg logit contributions
itability-5.308
Fact-5.274
amus-5.203
weights-5.128
quished-5.076
Token and
Token ID290
Feature activation+3.78e+00

Pos logit contributions
 Soup+1.446
esville+1.321
 Uniform+1.281
eers+1.271
 Albion+1.255
Neg logit contributions
Fact-1.476
itability-1.441
weights-1.394
amus-1.387
quished-1.331
Token Development
Token ID7712
Feature activation+1.54e+00

Pos logit contributions
 Soup+2.146
esville+1.849
 Uniform+1.840
eers+1.767
 Albion+1.766
Neg logit contributions
Fact-3.242
itability-3.184
weights-3.080
amus-3.071
quished-2.993
Token (
Token ID357
Feature activation+4.54e+00

Pos logit contributions
 Soup+0.939
esville+0.843
 Uniform+0.821
eers+0.812
 Albion+0.801
Neg logit contributions
Fact-1.249
itability-1.215
weights-1.179
amus-1.178
quished-1.131
TokenO
Token ID46
Feature activation+2.67e+00

Pos logit contributions
 Soup+3.246
esville+2.988
eers+2.902
 Uniform+2.885
 Albion+2.849
Neg logit contributions
Fact-3.156
itability-3.069
amus-2.987
weights-2.953
quished-2.855
TokenECD
Token ID27295
Feature activation+1.36e+00

Pos logit contributions
 Soup+2.497
esville+2.344
 Uniform+2.281
eers+2.280
icho+2.260
Neg logit contributions
Fact-1.298
itability-1.243
weights-1.176
amus-1.170
quished-1.099
Token).
Token ID737
Feature activation+2.62e-01

Pos logit contributions
 Soup+0.945
esville+0.857
 Uniform+0.839
eers+0.831
icho+0.819
Neg logit contributions
Fact-1.003
itability-0.966
amus-0.938
weights-0.935
quished-0.889
Token In
Token ID554
Feature activation-1.18e-02

Pos logit contributions
 Soup+0.202
esville+0.187
 Uniform+0.180
eers+0.180
icho+0.179
Neg logit contributions
Fact-0.172
itability-0.164
weights-0.160
amus-0.159
quished-0.149
Token the
Token ID262
Feature activation-3.61e-01

Pos logit contributions
Fact+0.008
itability+0.008
weights+0.008
amus+0.008
quished+0.007
Neg logit contributions
 Soup-0.009
esville-0.008
 Uniform-0.008
eers-0.008
icho-0.008
Token by
Token ID416
Feature activation+4.54e-01

Pos logit contributions
 Soup+2.294
esville+2.101
 Uniform+2.014
eers+2.011
 Albion+1.990
Neg logit contributions
Fact-3.219
itability-3.119
weights-3.047
amus-3.018
quished-2.880
Token Warriors
Token ID12090
Feature activation+1.99e+00

Pos logit contributions
 Soup+0.319
esville+0.292
 Uniform+0.283
eers+0.280
 Albion+0.278
Neg logit contributions
Fact-0.330
itability-0.318
weights-0.309
amus-0.308
quished-0.293
Token TV
Token ID3195
Feature activation+2.88e+00

Pos logit contributions
 Soup+1.379
esville+1.257
 Uniform+1.223
eers+1.199
 Albion+1.188
Neg logit contributions
Fact-1.458
itability-1.415
weights-1.380
amus-1.370
quished-1.308
Token
Token ID447
Feature activation+4.70e-01

Pos logit contributions
 Soup+1.957
esville+1.775
 Uniform+1.738
eers+1.712
 Albion+1.683
Neg logit contributions
Fact-2.135
itability-2.083
weights-2.023
amus-2.014
ysc-1.926
Token
Token ID247
Feature activation-8.93e-02

Pos logit contributions
 Soup+0.298
esville+0.269
 Uniform+0.259
eers+0.258
icho+0.256
Neg logit contributions
Fact-0.375
itability-0.362
amus-0.352
weights-0.351
quished-0.334
Tokens
Token ID82
Feature activation+1.25e+00

Pos logit contributions
Fact+0.078
itability+0.075
weights+0.074
amus+0.074
quished+0.070
Neg logit contributions
 Soup-0.050
esville-0.044
 Uniform-0.042
eers-0.042
icho-0.042
Token Laure
Token ID13826
Feature activation+3.35e+00

Pos logit contributions
 Soup+0.962
esville+0.886
 Uniform+0.862
eers+0.853
 Albion+0.848
Neg logit contributions
Fact-0.821
itability-0.787
weights-0.765
amus-0.762
quished-0.720
Tokennce
Token ID1198
Feature activation+7.10e-01

Pos logit contributions
 Soup+2.362
esville+2.157
 Uniform+2.092
eers+2.066
 Albion+2.065
Neg logit contributions
Fact-2.421
itability-2.316
weights-2.262
amus-2.198
quished-2.136
Token Scott
Token ID4746
Feature activation+2.95e+00

Pos logit contributions
 Soup+0.507
esville+0.464
 Uniform+0.449
eers+0.446
icho+0.443
Neg logit contributions
Fact-0.508
itability-0.488
weights-0.475
amus-0.473
quished-0.449
Token and
Token ID290
Feature activation+6.17e-01

Pos logit contributions
 Soup+1.971
esville+1.789
 Uniform+1.732
eers+1.706
 Albion+1.697
Neg logit contributions
Fact-2.237
itability-2.175
weights-2.107
amus-2.090
quished-1.997
Token radio
Token ID5243
Feature activation+1.02e+00

Pos logit contributions
 Soup+0.491
esville+0.452
 Uniform+0.441
 Albion+0.436
eers+0.436
Neg logit contributions
Fact-0.392
itability-0.376
weights-0.364
amus-0.362
quished-0.342
Token.
Token ID13
Feature activation-7.80e-01

Pos logit contributions
Fact+1.303
itability+1.249
amus+1.232
weights+1.231
anger+1.166
Neg logit contributions
 Soup-0.974
esville-0.880
 Uniform-0.842
eers-0.833
 Albion-0.829
Token
Token ID447
Feature activation+1.63e+00

Pos logit contributions
Fact+0.530
itability+0.505
weights+0.492
amus+0.490
quished+0.461
Neg logit contributions
 Soup-0.583
esville-0.539
eers-0.519
 Uniform-0.519
icho-0.516
Token
Token ID251
Feature activation+1.01e+00

Pos logit contributions
 Soup+1.000
esville+0.901
 Uniform+0.867
eers+0.860
icho+0.852
Neg logit contributions
Fact-1.330
itability-1.285
weights-1.256
amus-1.252
quished-1.193
Token Well
Token ID3894
Feature activation+1.02e+00

Pos logit contributions
 Soup+0.752
esville+0.688
 Uniform+0.671
eers+0.664
icho+0.659
Neg logit contributions
Fact-0.684
itability-0.663
weights-0.642
amus-0.640
quished-0.606
Token,
Token ID11
Feature activation-3.33e-01

Pos logit contributions
 Soup+0.609
esville+0.546
 Uniform+0.524
eers+0.520
icho+0.515
Neg logit contributions
Fact-0.855
itability-0.829
weights-0.810
amus-0.807
quished-0.769
Token after
Token ID706
Feature activation+1.45e+00

Pos logit contributions
Fact+0.241
itability+0.232
weights+0.226
amus+0.225
quished+0.214
Neg logit contributions
 Soup-0.234
esville-0.214
 Uniform-0.207
eers-0.205
icho-0.204
Token that
Token ID326
Feature activation+8.29e-01

Pos logit contributions
 Soup+1.068
esville+0.978
 Uniform+0.943
eers+0.935
 Albion+0.935
Neg logit contributions
Fact-1.015
itability-0.974
amus-0.945
weights-0.944
quished-0.901
Token I
Token ID314
Feature activation+2.91e+00

Pos logit contributions
 Soup+0.554
esville+0.503
 Uniform+0.486
eers+0.482
icho+0.478
Neg logit contributions
Fact-0.630
itability-0.608
weights-0.593
amus-0.590
quished-0.562
Token did
Token ID750
Feature activation+1.10e-01

Pos logit contributions
 Soup+1.996
esville+1.819
 Uniform+1.756
eers+1.745
 Albion+1.733
Neg logit contributions
Fact-2.153
itability-2.078
weights-2.019
amus-1.993
quished-1.924
Token give
Token ID1577
Feature activation+1.64e+00

Pos logit contributions
 Soup+0.077
esville+0.071
 Uniform+0.069
eers+0.068
icho+0.068
Neg logit contributions
Fact-0.079
itability-0.076
amus-0.074
weights-0.074
ysc-0.069
Token it
Token ID340
Feature activation-4.31e+00

Pos logit contributions
 Soup+1.151
esville+1.047
 Uniform+1.011
eers+1.006
 Albion+0.997
Neg logit contributions
Fact-1.196
itability-1.154
weights-1.125
amus-1.114
quished-1.077
Token Kane
Token ID20055
Feature activation-1.47e+00

Pos logit contributions
Fact+2.444
itability+2.327
weights+2.306
amus+2.300
 campaigner+2.151
Neg logit contributions
 Soup-2.656
esville-2.480
eers-2.399
 Uniform-2.390
icho-2.380
Token had
Token ID550
Feature activation-4.54e+00

Pos logit contributions
Fact+1.084
itability+1.042
weights+1.018
amus+1.017
quished+0.962
Neg logit contributions
 Soup-1.009
esville-0.921
 Uniform-0.889
eers-0.883
icho-0.877
Token come
Token ID1282
Feature activation-1.49e+00

Pos logit contributions
Fact+2.986
itability+2.895
amus+2.846
weights+2.845
 campaigner+2.656
Neg logit contributions
 Soup-3.408
esville-3.135
 Uniform-3.049
icho-3.029
efe-3.016
Token inches
Token ID8331
Feature activation-1.16e-02

Pos logit contributions
Fact+1.125
itability+1.085
weights+1.075
amus+1.070
ysc+1.010
Neg logit contributions
 Soup-0.985
esville-0.902
 Uniform-0.875
eers-0.867
icho-0.864
Token from
Token ID422
Feature activation-2.08e+00

Pos logit contributions
Fact+0.009
itability+0.008
weights+0.008
amus+0.008
quished+0.008
Neg logit contributions
 Soup-0.008
esville-0.007
 Uniform-0.007
eers-0.007
icho-0.007
Token putting
Token ID5137
Feature activation-2.51e+00

Pos logit contributions
Fact+1.442
itability+1.396
weights+1.360
amus+1.357
 campaigner+1.279
Neg logit contributions
 Soup-1.513
esville-1.380
 Uniform-1.343
eers-1.334
icho-1.330
Token Ireland
Token ID7517
Feature activation-3.82e+00

Pos logit contributions
Fact+1.759
itability+1.690
weights+1.664
amus+1.651
anger+1.552
Neg logit contributions
 Soup-1.808
esville-1.662
 Uniform-1.610
icho-1.596
eers-1.590
Token in
Token ID287
Feature activation-1.52e+00

Pos logit contributions
Fact+2.549
itability+2.455
weights+2.444
amus+2.418
 campaigner+2.258
Neg logit contributions
 Soup-2.832
esville-2.556
 Uniform-2.548
icho-2.537
eers-2.510
Token front
Token ID2166
Feature activation+8.63e-03

Pos logit contributions
Fact+0.739
itability+0.699
amus+0.675
weights+0.666
anger+0.607
Neg logit contributions
 Soup-1.426
esville-1.330
 Uniform-1.297
eers-1.297
icho-1.293
Token just
Token ID655
Feature activation+3.51e-01

Pos logit contributions
 Soup+0.006
esville+0.005
 Uniform+0.005
eers+0.005
icho+0.005
Neg logit contributions
Fact-0.007
itability-0.006
weights-0.006
amus-0.006
quished-0.006
Token two
Token ID734
Feature activation-6.09e-02

Pos logit contributions
 Soup+0.251
esville+0.231
 Uniform+0.223
eers+0.222
icho+0.220
Neg logit contributions
Fact-0.250
itability-0.240
weights-0.234
amus-0.233
quished-0.220
Token.
Token ID13
Feature activation-3.65e+00

Pos logit contributions
Fact+3.017
itability+2.901
amus+2.848
weights+2.840
 campaigner+2.699
Neg logit contributions
 Soup-2.333
esville-2.060
icho-2.004
 Uniform-1.984
eers-1.982
Token The
Token ID383
Feature activation-6.37e-01

Pos logit contributions
Fact+2.474
itability+2.290
amus+2.279
weights+2.268
anger+2.107
Neg logit contributions
 Soup-2.680
esville-2.524
eers-2.425
icho-2.411
ingu-2.392
Token Liberals
Token ID20162
Feature activation-1.72e+00

Pos logit contributions
Fact+0.427
itability+0.408
weights+0.398
amus+0.396
 campaigner+0.370
Neg logit contributions
 Soup-0.481
esville-0.445
 Uniform-0.428
eers-0.428
icho-0.426
Token are
Token ID389
Feature activation+1.40e+00

Pos logit contributions
Fact+1.207
itability+1.149
weights+1.122
amus+1.121
 campaigner+1.057
Neg logit contributions
 Soup-1.253
esville-1.146
 Uniform-1.105
eers-1.103
icho-1.100
Token offering
Token ID6011
Feature activation-3.74e-02

Pos logit contributions
 Soup+1.029
esville+0.944
 Uniform+0.915
eers+0.909
icho+0.903
Neg logit contributions
Fact-0.967
itability-0.928
weights-0.903
amus-0.900
quished-0.849
Token the
Token ID262
Feature activation-3.82e-01

Pos logit contributions
Fact+0.027
itability+0.026
weights+0.026
amus+0.025
quished+0.024
Neg logit contributions
 Soup-0.026
esville-0.024
 Uniform-0.023
eers-0.023
icho-0.023
Token same
Token ID976
Feature activation+2.81e+00

Pos logit contributions
Fact+0.275
itability+0.264
weights+0.258
amus+0.257
quished+0.241
Neg logit contributions
 Soup-0.270
esville-0.248
 Uniform-0.238
eers-0.238
icho-0.236
Token raises
Token ID12073
Feature activation-4.46e-01

Pos logit contributions
 Soup+1.904
esville+1.731
 Uniform+1.674
eers+1.660
 Albion+1.645
Neg logit contributions
Fact-2.102
itability-2.025
amus-1.974
weights-1.973
quished-1.878
Token the
Token ID262
Feature activation-1.84e+00

Pos logit contributions
Fact+0.322
itability+0.309
weights+0.301
amus+0.300
quished+0.284
Neg logit contributions
 Soup-0.316
esville-0.289
 Uniform-0.279
eers-0.277
icho-0.275
Token Conservatives
Token ID17897
Feature activation+2.40e+00

Pos logit contributions
Fact+1.078
itability+1.022
weights+0.993
amus+0.984
 campaigner+0.914
Neg logit contributions
 Soup-1.551
esville-1.448
 Uniform-1.398
eers-1.395
icho-1.390
Token gave
Token ID2921
Feature activation+3.72e+00

Pos logit contributions
 Soup+1.667
esville+1.525
 Uniform+1.475
eers+1.469
 Albion+1.456
Neg logit contributions
Fact-1.757
itability-1.688
weights-1.640
amus-1.637
quished-1.563
Token into
Token ID656
Feature activation-2.57e+00

Pos logit contributions
Fact+0.197
itability+0.190
weights+0.186
amus+0.185
quished+0.176
Neg logit contributions
 Soup-0.166
esville-0.150
 Uniform-0.145
eers-0.144
icho-0.143
Token the
Token ID262
Feature activation-2.12e+00

Pos logit contributions
Fact+1.674
itability+1.605
weights+1.555
amus+1.552
quished+1.452
Neg logit contributions
 Soup-1.982
esville-1.838
 Uniform-1.777
eers-1.766
icho-1.763
Token meditation
Token ID16901
Feature activation-2.53e+00

Pos logit contributions
Fact+1.531
itability+1.489
amus+1.436
weights+1.434
anger+1.354
Neg logit contributions
 Soup-1.468
esville-1.357
 Uniform-1.297
icho-1.296
eers-1.292
Token on
Token ID319
Feature activation+9.44e-01

Pos logit contributions
Fact+1.888
itability+1.823
weights+1.783
amus+1.775
ysc+1.685
Neg logit contributions
 Soup-1.712
esville-1.565
 Uniform-1.504
eers-1.496
icho-1.488
Token relationships
Token ID6958
Feature activation-4.57e+00

Pos logit contributions
 Soup+0.677
esville+0.618
 Uniform+0.602
 Albion+0.595
eers+0.594
Neg logit contributions
Fact-0.675
itability-0.643
weights-0.629
amus-0.629
quished-0.598
Token and
Token ID290
Feature activation-2.22e+00

Pos logit contributions
Fact+3.469
itability+3.331
weights+3.238
amus+3.212
anger+3.070
Neg logit contributions
 Soup-3.014
esville-2.760
 Uniform-2.649
eers-2.649
icho-2.605
Token life
Token ID1204
Feature activation-2.87e+00

Pos logit contributions
Fact+1.528
itability+1.475
weights+1.437
amus+1.425
anger+1.347
Neg logit contributions
 Soup-1.620
esville-1.489
 Uniform-1.434
eers-1.433
icho-1.431
Token that
Token ID326
Feature activation+3.33e+00

Pos logit contributions
Fact+2.276
itability+2.183
weights+2.153
amus+2.117
anger+2.033
Neg logit contributions
 Soup-1.821
esville-1.635
 Uniform-1.572
eers-1.565
icho-1.561
Token we
Token ID356
Feature activation+2.86e+00

Pos logit contributions
 Soup+2.350
esville+2.120
 Uniform+2.063
eers+2.060
 Albion+2.044
Neg logit contributions
Fact-2.392
itability-2.304
weights-2.242
amus-2.233
quished-2.139
Token all
Token ID477
Feature activation-2.15e+00

Pos logit contributions
 Soup+2.068
esville+1.913
eers+1.850
 Uniform+1.841
 Albion+1.825
Neg logit contributions
Fact-2.006
itability-1.906
weights-1.866
amus-1.849
quished-1.770
Token know
Token ID760
Feature activation+1.38e+00

Pos logit contributions
Fact+1.391
itability+1.357
amus+1.309
weights+1.301
 campaigner+1.220
Neg logit contributions
 Soup-1.657
esville-1.519
 Uniform-1.479
icho-1.470
eers-1.467
Token a
Token ID257
Feature activation-3.57e-01

Pos logit contributions
 Soup+0.753
esville+0.691
 Uniform+0.669
eers+0.666
icho+0.659
Neg logit contributions
Fact-0.728
itability-0.699
weights-0.680
amus-0.678
quished-0.643
Token lot
Token ID1256
Feature activation-2.41e+00

Pos logit contributions
Fact+0.237
itability+0.228
amus+0.220
weights+0.220
 campaigner+0.206
Neg logit contributions
 Soup-0.272
esville-0.251
 Uniform-0.242
eers-0.241
icho-0.240
Token of
Token ID286
Feature activation-2.77e+00

Pos logit contributions
Fact+1.623
itability+1.555
weights+1.515
amus+1.513
 campaigner+1.415
Neg logit contributions
 Soup-1.808
esville-1.655
 Uniform-1.605
icho-1.599
eers-1.591
Token other
Token ID584
Feature activation-2.78e+00

Pos logit contributions
Fact+1.788
itability+1.729
weights+1.682
amus+1.667
 campaigner+1.572
Neg logit contributions
 Soup-2.134
esville-1.973
icho-1.905
 Uniform-1.896
efe-1.881
Token ways
Token ID2842
Feature activation-3.81e+00

Pos logit contributions
Fact+1.852
itability+1.752
weights+1.732
amus+1.698
 campaigner+1.606
Neg logit contributions
 Soup-2.114
esville-1.949
 Uniform-1.877
icho-1.868
eers-1.852
Token to
Token ID284
Feature activation-7.77e-01

Pos logit contributions
Fact+2.648
itability+2.526
amus+2.469
weights+2.460
ysc+2.308
Neg logit contributions
 Soup-2.773
esville-2.535
 Uniform-2.464
efe-2.442
eers-2.424
Token use
Token ID779
Feature activation-4.17e+00

Pos logit contributions
Fact+0.537
itability+0.516
weights+0.500
amus+0.500
ysc+0.468
Neg logit contributions
 Soup-0.573
esville-0.526
 Uniform-0.510
eers-0.507
icho-0.504
Token cannabis
Token ID9499
Feature activation-2.62e+00

Pos logit contributions
Fact+2.663
itability+2.551
amus+2.487
weights+2.477
anger+2.310
Neg logit contributions
 Soup-3.242
esville-2.969
 Uniform-2.883
icho-2.872
eers-2.862
Token.
Token ID13
Feature activation-6.07e-01

Pos logit contributions
Fact+2.159
itability+2.077
weights+2.040
amus+2.036
 campaigner+1.930
Neg logit contributions
 Soup-1.580
esville-1.403
 Uniform-1.360
eers-1.346
icho-1.344
Token Ed
Token ID1717
Feature activation+2.41e+00

Pos logit contributions
Fact+0.401
itability+0.379
weights+0.370
amus+0.369
ysc+0.344
Neg logit contributions
 Soup-0.464
esville-0.431
eers-0.415
icho-0.414
 Uniform-0.414
Tokenibles
Token ID18764
Feature activation+1.40e+00

Pos logit contributions
 Soup+1.348
esville+1.202
 Uniform+1.147
eers+1.136
icho+1.134
Neg logit contributions
Fact-2.094
itability-2.016
weights-1.979
amus-1.965
quished-1.898
Token said
Token ID531
Feature activation+5.54e-01

Pos logit contributions
Fact+2.940
itability+2.776
amus+2.750
weights+2.731
 campaigner+2.617
Neg logit contributions
 Soup-3.183
esville-2.918
 Uniform-2.857
icho-2.843
eers-2.802
Token.
Token ID13
Feature activation+4.01e+00

Pos logit contributions
 Soup+0.317
esville+0.283
 Uniform+0.272
eers+0.270
icho+0.267
Neg logit contributions
Fact-0.474
itability-0.459
weights-0.448
amus-0.447
quished-0.427
Token "
Token ID366
Feature activation+2.92e-01

Pos logit contributions
 Soup+2.324
esville+2.066
 Uniform+2.011
 Albion+1.974
icho+1.967
Neg logit contributions
Fact-3.347
itability-3.254
weights-3.193
amus-3.171
quished-3.013
TokenThey
Token ID2990
Feature activation-4.48e-01

Pos logit contributions
 Soup+0.212
esville+0.193
 Uniform+0.188
eers+0.187
 Albion+0.185
Neg logit contributions
Fact-0.205
itability-0.199
weights-0.193
amus-0.192
quished-0.183
Token don
Token ID836
Feature activation+1.11e+00

Pos logit contributions
Fact+0.302
itability+0.290
weights+0.282
amus+0.281
quished+0.264
Neg logit contributions
 Soup-0.337
esville-0.310
 Uniform-0.301
eers-0.299
icho-0.297
Token't
Token ID470
Feature activation-1.11e+00

Pos logit contributions
 Soup+0.443
esville+0.374
 Uniform+0.353
eers+0.346
icho+0.345
Neg logit contributions
Fact-1.146
itability-1.109
weights-1.091
amus-1.091
anger-1.045
Token appreciate
Token ID9144
Feature activation+7.81e-01

Pos logit contributions
Fact+0.715
itability+0.686
weights+0.663
amus+0.662
 campaigner+0.620
Neg logit contributions
 Soup-0.870
esville-0.803
 Uniform-0.778
eers-0.776
icho-0.772
Token the
Token ID262
Feature activation-3.92e-01

Pos logit contributions
 Soup+0.570
esville+0.521
 Uniform+0.506
eers+0.503
 Albion+0.499
Neg logit contributions
Fact-0.547
itability-0.525
weights-0.511
amus-0.510
quished-0.486
Token fear
Token ID3252
Feature activation-2.07e+00

Pos logit contributions
Fact+0.271
itability+0.260
weights+0.252
amus+0.251
 campaigner+0.236
Neg logit contributions
 Soup-0.289
esville-0.266
 Uniform-0.257
eers-0.256
icho-0.255
Token m
Token ID285
Feature activation-2.53e+00

Pos logit contributions
Fact+1.451
itability+1.392
weights+1.359
amus+1.353
 campaigner+1.272
Neg logit contributions
 Soup-1.505
esville-1.377
 Uniform-1.341
eers-1.327
icho-1.322
Tokenong
Token ID506
Feature activation-2.09e-01

Pos logit contributions
Fact+1.712
itability+1.654
amus+1.607
weights+1.601
anger+1.543
Neg logit contributions
 Soup-1.915
 Uniform-1.725
esville-1.709
 XIV-1.678
 Albion-1.667
Token I
Token ID314
Feature activation-4.24e+00

Pos logit contributions
Fact+2.233
itability+2.122
weights+2.030
amus+2.018
 campaigner+1.931
Neg logit contributions
 Soup-2.820
esville-2.642
eers-2.572
 Uniform-2.534
icho-2.527
Token'll
Token ID1183
Feature activation-1.90e+00

Pos logit contributions
Fact+3.572
itability+3.444
amus+3.408
weights+3.369
 campaigner+3.252
Neg logit contributions
 Soup-2.453
esville-2.217
eers-2.164
icho-2.122
 Uniform-2.096
Token make
Token ID787
Feature activation-5.89e+00

Pos logit contributions
Fact+0.670
itability+0.621
weights+0.587
amus+0.581
quished+0.516
Neg logit contributions
 Soup-2.038
esville-1.927
 Uniform-1.881
eers-1.873
icho-1.864
Token the
Token ID262
Feature activation-3.12e+00

Pos logit contributions
Fact+4.367
itability+4.164
weights+4.094
amus+4.032
quished+3.857
Neg logit contributions
 Soup-3.934
esville-3.692
eers-3.547
 Albion-3.446
 Arche-3.430
Token large
Token ID1588
Feature activation-4.86e+00

Pos logit contributions
Fact+2.084
itability+1.998
weights+1.929
 campaigner+1.872
amus+1.868
Neg logit contributions
 Soup-2.338
esville-2.189
eers-2.143
 Uniform-2.108
icho-2.090
Token corporations
Token ID10225
Feature activation-7.52e+00

Pos logit contributions
Fact+3.014
itability+2.936
weights+2.874
amus+2.671
 campaigner+2.641
Neg logit contributions
 Soup-3.711
esville-3.622
eers-3.609
 Uniform-3.486
 Albion-3.479
Token pay
Token ID1414
Feature activation-6.05e+00

Pos logit contributions
Fact+5.342
itability+5.215
weights+5.086
 campaigner+4.945
quished+4.856
Neg logit contributions
eers-4.876
 Soup-4.839
esville-4.762
icho-4.492
 Uniform-4.407
Token '
Token ID705
Feature activation-4.60e+00

Pos logit contributions
Fact+3.925
itability+3.680
weights+3.612
amus+3.506
anger+3.445
Neg logit contributions
 Soup-4.649
esville-4.405
icho-4.341
eers-4.182
 Albion-4.181
Tokenfair
Token ID22043
Feature activation-2.72e+00

Pos logit contributions
Fact+3.895
itability+3.739
weights+3.712
amus+3.657
anger+3.477
Neg logit contributions
 Soup-2.727
esville-2.409
 Albion-2.302
icho-2.296
 Uniform-2.267
Token share
Token ID2648
Feature activation-5.37e+00

Pos logit contributions
Fact+1.553
itability+1.471
weights+1.426
amus+1.400
anger+1.315
Neg logit contributions
 Soup-2.331
esville-2.176
icho-2.125
 Uniform-2.123
 Albion-2.106
Token of
Token ID286
Feature activation-4.40e+00

Pos logit contributions
Fact+3.403
itability+3.217
amus+3.135
weights+3.131
 campaigner+2.939
Neg logit contributions
 Soup-4.250
esville-3.882
 Uniform-3.835
eers-3.797
icho-3.796
Token
Token ID447
Feature activation+7.29e-01

Pos logit contributions
 Soup+1.383
esville+1.246
 Uniform+1.225
 Albion+1.195
eers+1.195
Neg logit contributions
Fact-1.446
itability-1.412
amus-1.370
weights-1.370
quished-1.293
Token
Token ID247
Feature activation-3.57e-01

Pos logit contributions
 Soup+0.492
esville+0.448
 Uniform+0.433
eers+0.430
icho+0.427
Neg logit contributions
Fact-0.547
itability-0.528
weights-0.516
amus-0.513
quished-0.487
Token
Token ID198
Feature activation+2.14e+00

Pos logit contributions
Fact+0.256
itability+0.245
weights+0.239
amus+0.238
quished+0.225
Neg logit contributions
 Soup-0.254
esville-0.233
 Uniform-0.225
eers-0.224
icho-0.222
Token
Token ID198
Feature activation+1.47e+00

Pos logit contributions
 Soup+1.484
esville+1.355
 Uniform+1.309
eers+1.302
icho+1.289
Neg logit contributions
Fact-1.574
itability-1.518
weights-1.477
amus-1.473
quished-1.401
TokenOther
Token ID6395
Feature activation-3.26e+00

Pos logit contributions
 Soup+1.049
esville+0.958
eers+0.924
 Uniform+0.923
icho+0.913
Neg logit contributions
Fact-1.035
itability-1.021
amus-0.986
weights-0.984
quished-0.934
Token executions
Token ID30632
Feature activation-6.54e+00

Pos logit contributions
Fact+2.184
itability+2.081
weights+2.067
amus+2.004
 campaigner+1.933
Neg logit contributions
 Soup-2.419
esville-2.264
eers-2.168
efe-2.157
 Uniform-2.148
Token took
Token ID1718
Feature activation-5.99e+00

Pos logit contributions
Fact+4.121
itability+3.920
weights+3.867
amus+3.783
 campaigner+3.716
Neg logit contributions
 Soup-5.025
esville-4.574
icho-4.462
efe-4.436
 Uniform-4.430
Token place
Token ID1295
Feature activation-2.94e+00

Pos logit contributions
Fact+2.109
weights+1.956
itability+1.925
amus+1.836
 campaigner+1.641
Neg logit contributions
 Soup-6.292
esville-5.932
icho-5.827
 Uniform-5.798
eers-5.785
Token in
Token ID287
Feature activation-1.16e-01

Pos logit contributions
Fact+2.200
itability+2.110
weights+2.069
amus+2.041
ysc+1.959
Neg logit contributions
 Soup-1.980
esville-1.787
 Uniform-1.729
eers-1.729
efe-1.705
Token Chong
Token ID46892
Feature activation+1.58e+00

Pos logit contributions
Fact+0.082
itability+0.079
weights+0.077
amus+0.077
quished+0.072
Neg logit contributions
 Soup-0.083
esville-0.076
 Uniform-0.074
eers-0.073
icho-0.073
Tokenjin
Token ID18594
Feature activation-8.41e-01

Pos logit contributions
 Soup+1.056
esville+0.957
 Uniform+0.924
eers+0.916
icho+0.915
Neg logit contributions
Fact-1.202
itability-1.156
weights-1.127
amus-1.124
quished-1.072
Token [
Token ID685
Feature activation-4.14e-01

Pos logit contributions
Fact+0.501
itability+0.481
weights+0.468
amus+0.467
quished+0.441
Neg logit contributions
 Soup-0.519
esville-0.476
 Uniform-0.460
eers-0.458
icho-0.454
Token10
Token ID940
Feature activation-2.09e+00

Pos logit contributions
Fact+0.284
itability+0.273
weights+0.265
amus+0.264
quished+0.250
Neg logit contributions
 Soup-0.307
esville-0.282
 Uniform-0.273
eers-0.271
icho-0.270
Token:
Token ID25
Feature activation-1.63e+00

Pos logit contributions
Fact+1.714
itability+1.633
weights+1.612
amus+1.609
anger+1.525
Neg logit contributions
 Soup-1.277
esville-1.141
 Uniform-1.104
eers-1.095
icho-1.089
Token47
Token ID2857
Feature activation-2.01e+00

Pos logit contributions
Fact+1.068
itability+0.996
weights+0.980
amus+0.977
anger+0.912
Neg logit contributions
 Soup-1.265
esville-1.172
 Uniform-1.134
eers-1.130
icho-1.125
Token]
Token ID60
Feature activation-5.17e+00

Pos logit contributions
Fact+1.571
itability+1.499
weights+1.477
amus+1.476
anger+1.394
Neg logit contributions
 Soup-1.289
esville-1.178
eers-1.128
 Uniform-1.128
icho-1.115
Token <
Token ID1279
Feature activation-7.67e+00

Pos logit contributions
Fact+5.041
itability+4.825
amus+4.759
weights+4.746
anger+4.572
Neg logit contributions
 Soup-2.317
esville-2.045
eers-1.918
 Uniform-1.912
icho-1.853
Tokenmet
Token ID4164
Feature activation-1.95e+00

Pos logit contributions
Fact+5.670
amus+5.380
itability+5.243
weights+5.184
anger+5.062
Neg logit contributions
 Soup-5.149
 Uniform-4.652
esville-4.647
 Albion-4.477
eers-4.449
Tokenros
Token ID4951
Feature activation-4.03e+00

Pos logit contributions
Fact+1.416
itability+1.366
amus+1.332
weights+1.326
ysc+1.258
Neg logit contributions
 Soup-1.377
esville-1.250
 Uniform-1.219
 Albion-1.196
eers-1.196
Tokenxual
Token ID5541
Feature activation-7.09e-01

Pos logit contributions
Fact+3.011
itability+2.887
amus+2.841
weights+2.813
ysc+2.685
Neg logit contributions
 Soup-2.746
esville-2.498
 Uniform-2.421
 Albion-2.387
eers-2.376
Tokenth
Token ID400
Feature activation-4.15e+00

Pos logit contributions
Fact+0.575
itability+0.555
weights+0.542
amus+0.541
quished+0.515
Neg logit contributions
 Soup-0.439
esville-0.396
 Uniform-0.381
eers-0.378
icho-0.375
Tokenug
Token ID1018
Feature activation-3.48e+00

Pos logit contributions
Fact+3.046
itability+2.918
amus+2.898
weights+2.845
ysc+2.744
Neg logit contributions
 Soup-2.863
esville-2.609
 Uniform-2.534
 Albion-2.507
eers-2.506
TokenJ
Token ID41
Feature activation-3.42e+00

Pos logit contributions
Fact+5.188
itability+4.870
amus+4.813
weights+4.783
ysc+4.588
Neg logit contributions
 Soup-5.157
esville-4.666
 Uniform-4.612
 Albion-4.525
eers-4.456
Tokenm
Token ID76
Feature activation-1.04e+01

Pos logit contributions
Fact+2.333
itability+2.211
amus+2.159
weights+2.157
ysc+2.038
Neg logit contributions
 Soup-2.561
esville-2.344
 Uniform-2.289
eers-2.258
 Albion-2.255
TokenB
Token ID33
Feature activation-7.56e+00

Pos logit contributions
Fact+6.897
itability+6.376
weights+6.270
amus+6.269
ysc+6.060
Neg logit contributions
 Soup-7.575
esville-6.896
 Uniform-6.836
 Albion-6.730
eers-6.631
Tokenz
Token ID89
Feature activation-6.75e+00

Pos logit contributions
Fact+5.043
itability+4.699
weights+4.622
amus+4.621
anger+4.426
Neg logit contributions
 Soup-5.666
esville-5.116
 Uniform-5.058
 Albion-5.005
eers-4.941
Token5
Token ID20
Feature activation-9.67e+00

Pos logit contributions
Fact+4.599
itability+4.292
amus+4.232
weights+4.225
ysc+4.024
Neg logit contributions
 Soup-4.982
esville-4.553
 Uniform-4.475
 Albion-4.397
eers-4.365
Tokens
Token ID82
Feature activation-8.55e+00

Pos logit contributions
Fact+6.836
itability+6.364
amus+6.344
weights+6.279
ysc+6.052
Neg logit contributions
 Soup-6.581
 Uniform-5.900
esville-5.898
 Albion-5.789
 Arche-5.704
Token8
Token ID23
Feature activation-7.81e+00

Pos logit contributions
Fact+5.707
itability+5.267
amus+5.217
weights+5.180
anger+4.948
Neg logit contributions
 Soup-6.303
esville-5.739
 Uniform-5.686
eers-5.599
 Albion-5.597
Token
Token ID851
Feature activation+5.30e-01

Pos logit contributions
Fact+5.271
itability+4.919
amus+4.855
weights+4.812
anger+4.549
Neg logit contributions
 Soup-5.614
esville-5.177
icho-5.032
eers-4.985
 Uniform-4.952
Token General
Token ID3611
Feature activation+3.70e+00

Pos logit contributions
 Soup+0.393
esville+0.359
 Uniform+0.352
eers+0.346
 Albion+0.344
Neg logit contributions
Fact-0.365
itability-0.352
weights-0.341
amus-0.339
quished-0.322
Token Flynn
Token ID14654
Feature activation-5.02e+00

Pos logit contributions
 Soup+2.933
esville+2.671
 Uniform+2.638
efe+2.565
eers+2.560
Neg logit contributions
Fact-2.291
itability-2.250
weights-2.167
amus-2.136
ysc-2.028
Token (@
Token ID4275
Feature activation+8.50e-01

Pos logit contributions
Fact+4.203
itability+3.998
amus+3.929
weights+3.913
anger+3.757
Neg logit contributions
 Soup-2.855
esville-2.627
eers-2.513
icho-2.489
 Albion-2.464
Token was
Token ID373
Feature activation+1.22e+00

Pos logit contributions
 Soup+1.355
esville+1.236
 Uniform+1.191
eers+1.186
 Albion+1.172
Neg logit contributions
Fact-1.570
itability-1.515
weights-1.474
amus-1.461
quished-1.411
Token coerced
Token ID48474
Feature activation-2.29e-01

Pos logit contributions
 Soup+0.815
esville+0.741
 Uniform+0.715
eers+0.710
icho+0.705
Neg logit contributions
Fact-0.930
itability-0.896
weights-0.874
amus-0.869
quished-0.828
Token into
Token ID656
Feature activation-4.37e+00

Pos logit contributions
Fact+0.193
itability+0.187
weights+0.182
amus+0.182
quished+0.174
Neg logit contributions
 Soup-0.134
esville-0.119
 Uniform-0.115
eers-0.114
icho-0.113
Token filing
Token ID12180
Feature activation-6.23e+00

Pos logit contributions
Fact+2.549
itability+2.420
amus+2.352
weights+2.308
 campaigner+2.218
Neg logit contributions
 Soup-3.589
esville-3.333
eers-3.279
icho-3.239
efe-3.238
Token a
Token ID257
Feature activation-1.67e+00

Pos logit contributions
Fact+4.115
itability+3.895
amus+3.769
weights+3.719
anger+3.578
Neg logit contributions
 Soup-4.657
esville-4.258
eers-4.191
icho-4.185
 Albion-4.110
Token police
Token ID1644
Feature activation-6.05e+00

Pos logit contributions
Fact+1.108
itability+1.060
amus+1.025
weights+1.020
anger+0.960
Neg logit contributions
 Soup-1.282
esville-1.182
eers-1.146
icho-1.137
 Uniform-1.136
Token complaint
Token ID8224
Feature activation-3.92e+00

Pos logit contributions
Fact+2.638
itability+2.411
amus+2.258
weights+2.222
anger+2.098
Neg logit contributions
 Soup-5.989
esville-5.626
eers-5.516
icho-5.471
 Albion-5.426
Token.
Token ID13
Feature activation-7.31e-01

Pos logit contributions
Fact+2.907
itability+2.794
amus+2.740
weights+2.708
 campaigner+2.576
Neg logit contributions
 Soup-2.642
esville-2.415
icho-2.327
eers-2.318
 Uniform-2.309
Token She
Token ID1375
Feature activation-1.19e+00

Pos logit contributions
Fact+0.516
itability+0.494
weights+0.481
amus+0.480
quished+0.453
Neg logit contributions
 Soup-0.528
esville-0.485
 Uniform-0.468
eers-0.467
icho-0.463
Token stated
Token ID5081
Feature activation-3.21e+00

Pos logit contributions
Fact+0.839
itability+0.805
amus+0.786
weights+0.785
anger+0.737
Neg logit contributions
 Soup-0.863
esville-0.788
 Uniform-0.764
eers-0.759
icho-0.757
Token they
Token ID484
Feature activation-3.51e-01

Pos logit contributions
Fact+2.069
itability+1.976
weights+1.912
amus+1.894
anger+1.796
Neg logit contributions
 Soup-2.485
esville-2.292
eers-2.220
 Uniform-2.214
icho-2.203