TOP ACTIVATIONS
MAX = 10.606

appearance with Treasury Secretary Steven Mn uchin , that the debt
. Blank fe in , Cohn s mentor and
designed stone mug costs 16 euros ($ 18 ) from the
of law , Fe ing old said in a
. Fe eling Hel pless
the tens of millions of dollars . Story continues
attention , and millions of dollars of outside spending , to
hat - tr ick and Fra izer Campbell 's opener ,
. S . economist Paul Krugman dubbed the revision from an
there have been reports that Bannon might be edged out ,
. President Hassan Rouhani has shown he is listening
of the 2008 film Kung Fu Panda , is an anthrop
graft last year as President Xi Jinping continued to wage a
White House chief strategist Stephen Bannon during the presidential campaign ),
with hundreds of thousands of dollars .
Section III : Lean Six Sigma What
by blocking national highway connecting Roh t ak to Delhi .
was the brig adier general Pier a G att es chi
. In St Petersburg , for example , D
as Treasury secretary . Blank fe in , Cohn

MIN ACTIVATIONS
MIN = -11.022

techniques to write to multiple dat acent ers in different cities
with multiple calls over the satellite phone , Sullivan is finally
Big gest at - bat of my life ,
media , saying an oft - qu oted blog post from
skin condition , an autos omal dominant palm opl ant ar
a junior to earn all - state honors .
drama parody sk it " Veter in arian 's Hospital ",
. Typically the most eas ter ly aim point was targeted
are here to protect the civilians and to free their city
au , who is partially disabled , told authorities that he
to the bottom of the container . When moisture starts to
is a play - by - play guy who is not
it is directed or ab et ted by states who are
uk Moor claw , and Col bie Sterling all of
of critical illness - related cort ic oster oid ins uff
find a specific pet - colour combination , by setting both
be for " active " registered voters , i . e
natural habitat . The fire ant epidemic arose when the killer
lives , including hundreds of civilians . The hypocrisy
Development of the F - 35 fighter jet will cost

INTERVAL 5.199 - 10.606
CONTAINS 0.421%

well founded in fairness and economics . But like most reforms
He has one year and $ 1 . 35 million left
get tes , basil and spinach , and named in her
it , said Ryan , attribut ing the foul
, you will get a $ 1 , 000 check from

INTERVAL -0.208 - 5.199
CONTAINS 55.987%

own John Mu ir hike . Bob In man had written
ama i , k ū pa a , al
> lol m 2 k [ 10 : 47 ] <
of the business ; having [ a big ] enough premium
y would pay off , and that it would come to

INTERVAL -5.615 - -0.208
CONTAINS 42.942%

L sound , something that Phil adel ph ians do more
using your new found Deep Conf idence . Let
About former meth dealer and escort Mike Jones , H
also need to authorize each other to use the credit and
of the new polymer notes are not raised . The Bra

INTERVAL -11.022 - -5.615
CONTAINS 0.650%

ite cleric in Iraq , who has been outspoken against the
2016 . REUTERS / Cl od agh Kil c oy ne
act worth $ 20 million that was passed to help repair
or wage employment , whether regular or casual , less than
2 . D ensity as ambient plus deviation . Fl uid
Token appearance
Token ID5585
Feature activation+2.10e+00

Pos logit contributions
 holiest+3.686
emi+3.668
 athlet+3.592
cean+3.580
 Lanka+3.579
Neg logit contributions
chrom-2.484
roit-2.315
proof-2.312
rans-2.219
eor-2.212
Token with
Token ID351
Feature activation+3.71e+00

Pos logit contributions
 holiest+1.373
emi+1.358
 athlet+1.326
 Lanka+1.308
cean+1.307
Neg logit contributions
chrom-1.735
proof-1.641
roit-1.640
bos-1.610
eor-1.592
Token Treasury
Token ID13419
Feature activation+5.26e+00

Pos logit contributions
 holiest+2.654
emi+2.631
 athlet+2.592
cean+2.559
 goodbye+2.551
Neg logit contributions
chrom-2.848
roit-2.678
proof-2.662
eor-2.605
rans-2.597
Token Secretary
Token ID4986
Feature activation+4.52e+00

Pos logit contributions
 holiest+2.140
emi+2.111
 athlet+2.057
 goodbye+2.001
cean+1.988
Neg logit contributions
chrom-5.671
roit-5.445
proof-5.440
eor-5.358
rans-5.323
Token Steven
Token ID8239
Feature activation+6.95e+00

Pos logit contributions
emi+2.537
 holiest+2.498
 athlet+2.438
olicy+2.384
cean+2.382
Neg logit contributions
chrom-4.157
proof-3.989
roit-3.988
bos-3.927
eor-3.859
Token Mn
Token ID36030
Feature activation+1.06e+01

Pos logit contributions
 holiest+4.011
emi+3.981
 athlet+3.958
 goodbye+3.930
olicy+3.870
Neg logit contributions
chrom-6.319
roit-5.991
proof-5.891
bos-5.828
rans-5.816
Tokenuchin
Token ID43416
Feature activation+6.47e+00

Pos logit contributions
 holiest+5.013
emi+4.942
 athlet+4.760
 goodbye+4.727
cean+4.693
Neg logit contributions
chrom-10.311
proof-9.800
roit-9.774
inant-9.688
bos-9.653
Token,
Token ID11
Feature activation+2.54e+00

Pos logit contributions
 holiest+3.867
emi+3.790
 athlet+3.743
 goodbye+3.727
cean+3.670
Neg logit contributions
chrom-5.698
roit-5.355
proof-5.340
eor-5.241
obyl-5.223
Token that
Token ID326
Feature activation-1.78e-01

Pos logit contributions
 holiest+1.542
emi+1.520
 athlet+1.503
cean+1.465
 goodbye+1.461
Neg logit contributions
chrom-2.240
proof-2.106
roit-2.094
rans-2.048
eor-2.048
Token the
Token ID262
Feature activation+2.62e+00

Pos logit contributions
chrom+0.140
proof+0.133
roit+0.131
bos+0.127
eor+0.127
Neg logit contributions
emi-0.123
 holiest-0.122
 athlet-0.120
olicy-0.119
cean-0.117
Token debt
Token ID5057
Feature activation+7.11e+00

Pos logit contributions
emi+1.575
olicy+1.562
 holiest+1.551
cean+1.539
 athlet+1.534
Neg logit contributions
chrom-2.321
proof-2.191
roit-2.156
rans-2.118
eor-2.106
Token.
Token ID13
Feature activation+2.79e+00

Pos logit contributions
 holiest+3.079
emi+2.997
 athlet+2.964
 goodbye+2.930
cean+2.917
Neg logit contributions
chrom-4.162
proof-3.920
roit-3.915
eor-3.838
rans-3.816
Token Blank
Token ID31990
Feature activation+6.52e+00

Pos logit contributions
 athlet+1.772
 holiest+1.744
emi+1.649
 goodbye+1.629
cean+1.621
Neg logit contributions
chrom-2.452
proof-2.352
roit-2.310
eor-2.287
rans-2.273
Tokenfe
Token ID5036
Feature activation+8.60e+00

Pos logit contributions
 holiest+4.803
 athlet+4.688
emi+4.618
 goodbye+4.604
olicy+4.577
Neg logit contributions
chrom-4.979
proof-4.629
roit-4.616
rans-4.555
obyl-4.505
Tokenin
Token ID259
Feature activation+5.02e+00

Pos logit contributions
emi+7.226
 holiest+7.151
 goodbye+7.015
 bargaining+6.943
 athlet+6.915
Neg logit contributions
chrom-5.396
proof-4.781
obyl-4.780
roit-4.757
rans-4.666
Token,
Token ID11
Feature activation+3.03e+00

Pos logit contributions
 holiest+3.619
 athlet+3.492
emi+3.465
 goodbye+3.465
olicy+3.461
Neg logit contributions
chrom-3.934
proof-3.677
roit-3.666
rans-3.580
eor-3.579
Token Cohn
Token ID45005
Feature activation+1.04e+01

Pos logit contributions
 holiest+2.205
emi+2.171
cean+2.147
 athlet+2.127
 Lanka+2.088
Neg logit contributions
chrom-2.320
proof-2.176
roit-2.167
bos-2.134
rans-2.115
Token
Token ID447
Feature activation+3.16e+00

Pos logit contributions
 holiest+5.804
emi+5.802
 goodbye+5.685
 athlet+5.608
cean+5.550
Neg logit contributions
chrom-9.373
roit-8.759
proof-8.758
obyl-8.592
bos-8.591
Token
Token ID247
Feature activation+3.83e+00

Pos logit contributions
 holiest+2.489
 athlet+2.444
emi+2.439
cean+2.373
olicy+2.372
Neg logit contributions
chrom-2.266
proof-2.113
roit-2.103
rans-2.033
inant-2.026
Tokens
Token ID82
Feature activation+5.49e+00

Pos logit contributions
 holiest+3.285
emi+3.218
 athlet+3.205
 goodbye+3.192
cean+3.151
Neg logit contributions
chrom-2.408
roit-2.208
proof-2.203
eor-2.145
bos-2.122
Token mentor
Token ID22387
Feature activation+5.46e+00

Pos logit contributions
 holiest+3.429
emi+3.364
 athlet+3.346
cean+3.283
olicy+3.276
Neg logit contributions
chrom-4.740
roit-4.544
proof-4.514
eor-4.396
bos-4.374
Token and
Token ID290
Feature activation+3.71e+00

Pos logit contributions
 holiest+3.506
emi+3.394
 athlet+3.355
 goodbye+3.316
cean+3.308
Neg logit contributions
chrom-4.657
proof-4.394
roit-4.383
eor-4.277
rans-4.255
Token designed
Token ID3562
Feature activation+9.52e-01

Pos logit contributions
emi+0.521
olicy+0.507
 Lanka+0.505
cean+0.502
assi+0.489
Neg logit contributions
chrom-0.577
proof-0.537
roit-0.529
bos-0.522
rans-0.518
Token stone
Token ID7815
Feature activation+1.29e+00

Pos logit contributions
emi+0.678
cean+0.673
 holiest+0.672
olicy+0.665
 athlet+0.655
Neg logit contributions
chrom-0.730
proof-0.695
roit-0.665
bos-0.661
eor-0.652
Token mug
Token ID25152
Feature activation+3.43e+00

Pos logit contributions
emi+0.899
 holiest+0.899
cean+0.882
olicy+0.881
 goodbye+0.873
Neg logit contributions
chrom-1.011
proof-0.965
roit-0.916
bos-0.912
eor-0.904
Token costs
Token ID3484
Feature activation+6.00e+00

Pos logit contributions
 holiest+2.505
emi+2.461
cean+2.413
olicy+2.406
 goodbye+2.400
Neg logit contributions
chrom-2.635
proof-2.463
roit-2.409
eor-2.373
bos-2.364
Token 16
Token ID1467
Feature activation+5.77e-01

Pos logit contributions
 holiest+3.915
emi+3.797
 athlet+3.763
cean+3.739
olicy+3.725
Neg logit contributions
chrom-5.055
proof-4.746
roit-4.716
rans-4.611
bos-4.589
Token euros
Token ID21138
Feature activation+9.53e+00

Pos logit contributions
 holiest+0.374
cean+0.360
emi+0.359
 goodbye+0.358
olicy+0.356
Neg logit contributions
chrom-0.484
roit-0.457
proof-0.456
rans-0.451
bos-0.445
Token ($
Token ID7198
Feature activation+5.81e+00

Pos logit contributions
 holiest+5.650
emi+5.589
 athlet+5.517
 goodbye+5.504
cean+5.399
Neg logit contributions
chrom-8.290
roit-7.791
proof-7.755
eor-7.657
obyl-7.639
Token18
Token ID1507
Feature activation+1.97e+00

Pos logit contributions
 holiest+3.952
emi+3.890
 goodbye+3.836
 athlet+3.836
cean+3.783
Neg logit contributions
chrom-4.620
roit-4.316
proof-4.295
eor-4.214
obyl-4.201
Token)
Token ID8
Feature activation+1.24e+00

Pos logit contributions
 holiest+1.310
emi+1.262
 athlet+1.249
 goodbye+1.244
olicy+1.223
Neg logit contributions
chrom-1.633
proof-1.533
roit-1.532
rans-1.505
bos-1.488
Token from
Token ID422
Feature activation+1.44e+00

Pos logit contributions
 holiest+0.819
 Lanka+0.782
emi+0.781
cean+0.773
 athlet+0.772
Neg logit contributions
chrom-1.038
proof-0.984
roit-0.970
rans-0.952
bos-0.939
Token the
Token ID262
Feature activation+1.51e+00

Pos logit contributions
 holiest+0.934
cean+0.920
emi+0.911
olicy+0.903
 athlet+0.898
Neg logit contributions
chrom-1.199
proof-1.132
roit-1.124
bos-1.103
rans-1.103
Token of
Token ID286
Feature activation+2.76e+00

Pos logit contributions
 holiest+1.470
emi+1.434
 athlet+1.410
 goodbye+1.404
cean+1.376
Neg logit contributions
chrom-3.151
roit-2.983
proof-2.977
eor-2.932
obyl-2.923
Token law
Token ID1099
Feature activation-8.61e-01

Pos logit contributions
 holiest+1.683
emi+1.651
cean+1.622
olicy+1.620
 athlet+1.620
Neg logit contributions
chrom-2.414
proof-2.323
roit-2.309
rans-2.232
inant-2.208
Token,
Token ID11
Feature activation+2.37e+00

Pos logit contributions
chrom+0.673
roit+0.641
proof+0.640
rans+0.618
bos+0.602
Neg logit contributions
 holiest-0.587
emi-0.581
 athlet-0.567
cean-0.567
olicy-0.563
Token
Token ID447
Feature activation+2.08e+00

Pos logit contributions
 holiest+1.637
emi+1.600
 athlet+1.573
cean+1.564
 goodbye+1.540
Neg logit contributions
chrom-1.914
proof-1.804
roit-1.795
rans-1.762
bos-1.726
Token
Token ID251
Feature activation+5.10e-03

Pos logit contributions
 holiest+1.272
 athlet+1.269
emi+1.231
olicy+1.221
cean+1.208
Neg logit contributions
chrom-1.813
roit-1.734
proof-1.724
eor-1.687
rans-1.684
Token Fe
Token ID5452
Feature activation+9.47e+00

Pos logit contributions
 holiest+0.003
 athlet+0.003
emi+0.003
 goodbye+0.003
cean+0.003
Neg logit contributions
chrom-0.004
proof-0.004
roit-0.004
rans-0.004
eor-0.004
Tokening
Token ID278
Feature activation+2.96e+00

Pos logit contributions
emi+7.829
 holiest+7.776
cean+7.624
 athlet+7.560
 goodbye+7.558
Neg logit contributions
chrom-5.936
roit-5.459
proof-5.430
obyl-5.303
eor-5.303
Tokenold
Token ID727
Feature activation+6.93e-03

Pos logit contributions
 holiest+2.618
 athlet+2.563
cean+2.524
 goodbye+2.516
 bargaining+2.514
Neg logit contributions
chrom-1.798
roit-1.712
proof-1.658
rans-1.652
bos-1.622
Token said
Token ID531
Feature activation+1.71e+00

Pos logit contributions
 holiest+0.005
 athlet+0.005
olicy+0.005
cean+0.005
 goodbye+0.005
Neg logit contributions
chrom-0.005
roit-0.005
proof-0.005
rans-0.004
eor-0.004
Token in
Token ID287
Feature activation+4.46e+00

Pos logit contributions
 holiest+1.116
emi+1.080
 athlet+1.076
cean+1.061
olicy+1.049
Neg logit contributions
chrom-1.436
roit-1.353
proof-1.340
bos-1.315
rans-1.310
Token a
Token ID257
Feature activation+3.19e+00

Pos logit contributions
 holiest+2.642
emi+2.588
 athlet+2.561
 goodbye+2.518
cean+2.503
Neg logit contributions
chrom-4.028
proof-3.809
roit-3.804
rans-3.705
bos-3.703
Token.
Token ID13
Feature activation+3.21e+00

Pos logit contributions
 holiest+0.369
 Lanka+0.353
olicy+0.351
emi+0.350
cean+0.348
Neg logit contributions
chrom-0.455
proof-0.433
roit-0.427
eor-0.418
rans-0.416
Token
Token ID447
Feature activation+3.15e+00

Pos logit contributions
 holiest+2.069
 athlet+2.027
emi+1.944
 goodbye+1.930
cean+1.915
Neg logit contributions
chrom-2.796
proof-2.641
roit-2.599
eor-2.574
bos-2.542
Token
Token ID251
Feature activation+8.50e-01

Pos logit contributions
 holiest+2.190
emi+2.159
 athlet+2.151
cean+2.102
 goodbye+2.100
Neg logit contributions
chrom-2.486
roit-2.324
proof-2.320
eor-2.269
rans-2.257
Token
Token ID198
Feature activation+2.94e+00

Pos logit contributions
 holiest+0.560
 athlet+0.557
emi+0.533
olicy+0.531
 goodbye+0.530
Neg logit contributions
chrom-0.715
proof-0.674
roit-0.668
eor-0.654
bos-0.648
Token
Token ID198
Feature activation+2.89e+00

Pos logit contributions
 holiest+1.954
 athlet+1.937
emi+1.876
olicy+1.854
 goodbye+1.839
Neg logit contributions
chrom-2.490
proof-2.336
roit-2.295
eor-2.257
bos-2.247
TokenFe
Token ID14304
Feature activation+9.36e+00

Pos logit contributions
 holiest+2.067
 athlet+2.066
 goodbye+1.992
emi+1.970
 Lanka+1.970
Neg logit contributions
chrom-2.336
proof-2.176
roit-2.129
bos-2.103
eor-2.090
Tokeneling
Token ID10809
Feature activation+6.90e+00

Pos logit contributions
 holiest+7.717
emi+7.695
 athlet+7.521
 goodbye+7.517
cean+7.511
Neg logit contributions
chrom-5.952
proof-5.451
roit-5.443
eor-5.300
obyl-5.288
Token Hel
Token ID5053
Feature activation+3.69e+00

Pos logit contributions
 holiest+4.188
emi+4.115
 athlet+4.043
cean+4.031
 goodbye+4.010
Neg logit contributions
chrom-6.118
proof-5.767
roit-5.746
rans-5.578
bos-5.572
Tokenpless
Token ID14570
Feature activation+3.76e+00

Pos logit contributions
 holiest+2.977
emi+2.909
 athlet+2.908
 goodbye+2.899
 bargaining+2.858
Neg logit contributions
chrom-2.522
proof-2.333
roit-2.333
bos-2.260
obyl-2.256
Token
Token ID198
Feature activation+2.57e+00

Pos logit contributions
 holiest+2.491
emi+2.400
 athlet+2.387
olicy+2.372
 goodbye+2.372
Neg logit contributions
chrom-3.108
proof-2.965
roit-2.934
obyl-2.874
bos-2.868
Token
Token ID198
Feature activation+2.62e+00

Pos logit contributions
 holiest+1.762
 athlet+1.753
emi+1.685
olicy+1.684
 bargaining+1.653
Neg logit contributions
chrom-2.125
proof-2.002
roit-1.949
eor-1.917
bos-1.914
Token the
Token ID262
Feature activation+3.05e+00

Pos logit contributions
 holiest+3.013
emi+2.951
 athlet+2.919
 goodbye+2.915
cean+2.857
Neg logit contributions
chrom-4.056
roit-3.802
proof-3.796
eor-3.704
rans-3.694
Token tens
Token ID11192
Feature activation+4.09e+00

Pos logit contributions
emi+1.924
olicy+1.899
 holiest+1.899
cean+1.868
 goodbye+1.866
Neg logit contributions
chrom-2.612
proof-2.450
roit-2.444
rans-2.394
eor-2.377
Token of
Token ID286
Feature activation+3.41e+00

Pos logit contributions
 holiest+2.295
 athlet+2.198
emi+2.193
 goodbye+2.137
cean+2.114
Neg logit contributions
chrom-3.896
proof-3.681
roit-3.639
rans-3.618
eor-3.578
Token millions
Token ID5242
Feature activation+6.19e+00

Pos logit contributions
 holiest+1.950
emi+1.935
cean+1.878
 athlet+1.874
 goodbye+1.863
Neg logit contributions
chrom-3.101
proof-2.948
rans-2.909
roit-2.907
bos-2.883
Token of
Token ID286
Feature activation+3.44e+00

Pos logit contributions
 holiest+3.277
emi+3.204
 athlet+3.156
 goodbye+3.144
cean+3.089
Neg logit contributions
chrom-5.878
roit-5.549
proof-5.535
eor-5.437
obyl-5.418
Token dollars
Token ID5054
Feature activation+9.33e+00

Pos logit contributions
 holiest+1.550
emi+1.492
cean+1.439
 goodbye+1.419
 athlet+1.406
Neg logit contributions
chrom-3.595
proof-3.442
roit-3.436
rans-3.412
bos-3.357
Token.
Token ID13
Feature activation+2.99e+00

Pos logit contributions
 holiest+5.375
emi+5.274
 athlet+5.228
 goodbye+5.191
cean+5.087
Neg logit contributions
chrom-8.334
roit-7.843
proof-7.823
obyl-7.714
eor-7.712
Token
Token ID198
Feature activation+2.86e+00

Pos logit contributions
 holiest+2.037
 athlet+2.017
emi+1.941
 goodbye+1.934
cean+1.907
Neg logit contributions
chrom-2.456
proof-2.334
roit-2.304
eor-2.260
bos-2.242
Token
Token ID198
Feature activation+2.82e+00

Pos logit contributions
 holiest+1.947
 athlet+1.935
emi+1.872
olicy+1.858
 bargaining+1.828
Neg logit contributions
chrom-2.372
proof-2.235
roit-2.183
eor-2.149
bos-2.144
TokenStory
Token ID11605
Feature activation+4.14e+00

Pos logit contributions
 holiest+1.914
 athlet+1.891
 goodbye+1.844
emi+1.823
 Lanka+1.815
Neg logit contributions
chrom-2.370
proof-2.235
roit-2.196
bos-2.162
eor-2.156
Token continues
Token ID4477
Feature activation+3.93e+00

Pos logit contributions
 holiest+2.466
emi+2.367
 athlet+2.348
 goodbye+2.320
olicy+2.269
Neg logit contributions
chrom-3.750
proof-3.537
roit-3.523
eor-3.464
rans-3.458
Token attention
Token ID3241
Feature activation+4.40e+00

Pos logit contributions
emi+2.135
 holiest+2.130
olicy+2.123
cean+2.094
 athlet+2.079
Neg logit contributions
chrom-2.381
proof-2.246
roit-2.208
bos-2.180
rans-2.161
Token,
Token ID11
Feature activation+2.99e+00

Pos logit contributions
 holiest+2.768
emi+2.716
olicy+2.662
 athlet+2.656
 goodbye+2.636
Neg logit contributions
chrom-3.776
roit-3.546
proof-3.532
rans-3.475
eor-3.452
Token and
Token ID290
Feature activation+3.81e+00

Pos logit contributions
emi+2.051
 holiest+2.046
cean+1.989
 goodbye+1.979
olicy+1.979
Neg logit contributions
chrom-2.381
proof-2.254
roit-2.220
rans-2.207
bos-2.194
Token millions
Token ID5242
Feature activation+5.94e+00

Pos logit contributions
 holiest+2.583
emi+2.583
cean+2.478
 athlet+2.474
olicy+2.462
Neg logit contributions
chrom-3.112
proof-2.953
roit-2.934
rans-2.880
bos-2.877
Token of
Token ID286
Feature activation+3.63e+00

Pos logit contributions
 holiest+3.579
emi+3.486
 athlet+3.431
cean+3.395
 goodbye+3.392
Neg logit contributions
chrom-5.295
roit-4.998
proof-4.997
rans-4.885
bos-4.867
Token dollars
Token ID5054
Feature activation+9.16e+00

Pos logit contributions
 holiest+2.416
emi+2.384
cean+2.342
olicy+2.296
 athlet+2.283
Neg logit contributions
chrom-3.041
proof-2.863
roit-2.863
rans-2.800
bos-2.793
Token of
Token ID286
Feature activation+3.83e+00

Pos logit contributions
 holiest+5.543
emi+5.447
 athlet+5.382
 goodbye+5.365
cean+5.273
Neg logit contributions
chrom-7.950
roit-7.462
proof-7.442
eor-7.309
obyl-7.279
Token outside
Token ID2354
Feature activation+2.89e+00

Pos logit contributions
 holiest+2.629
emi+2.625
cean+2.568
olicy+2.546
 athlet+2.507
Neg logit contributions
chrom-3.079
proof-2.894
roit-2.871
bos-2.834
rans-2.811
Token spending
Token ID4581
Feature activation+5.57e+00

Pos logit contributions
 holiest+1.765
emi+1.752
olicy+1.695
cean+1.685
 goodbye+1.680
Neg logit contributions
chrom-2.547
proof-2.413
roit-2.405
bos-2.356
rans-2.349
Token,
Token ID11
Feature activation+3.33e+00

Pos logit contributions
 holiest+3.529
emi+3.442
olicy+3.394
cean+3.383
 athlet+3.376
Neg logit contributions
chrom-4.782
proof-4.517
roit-4.510
bos-4.409
eor-4.405
Token to
Token ID284
Feature activation+1.85e+00

Pos logit contributions
 holiest+2.369
emi+2.346
cean+2.322
olicy+2.297
 athlet+2.278
Neg logit contributions
chrom-2.580
proof-2.440
roit-2.405
rans-2.386
bos-2.369
Token hat
Token ID6877
Feature activation+6.81e+00

Pos logit contributions
 holiest+3.167
emi+3.131
olicy+3.084
cean+3.078
 athlet+3.040
Neg logit contributions
chrom-4.180
roit-3.955
proof-3.951
bos-3.833
eor-3.817
Token-
Token ID12
Feature activation-7.53e-01

Pos logit contributions
 holiest+4.083
emi+3.959
 athlet+3.934
 goodbye+3.910
 Lanka+3.864
Neg logit contributions
chrom-6.028
roit-5.691
proof-5.685
eor-5.564
rans-5.529
Tokentr
Token ID2213
Feature activation+1.49e+00

Pos logit contributions
chrom+0.343
proof+0.313
roit+0.299
bos+0.289
rans+0.286
Neg logit contributions
 holiest-0.794
 Lanka-0.783
cean-0.771
emi-0.771
 athlet-0.768
Tokenick
Token ID624
Feature activation+5.60e+00

Pos logit contributions
 holiest+1.032
 athlet+0.999
 goodbye+0.985
emi+0.975
 Lanka+0.968
Neg logit contributions
chrom-1.210
roit-1.136
proof-1.128
rans-1.121
eor-1.108
Token and
Token ID290
Feature activation+3.45e+00

Pos logit contributions
 holiest+3.411
emi+3.333
 athlet+3.280
cean+3.256
 goodbye+3.227
Neg logit contributions
chrom-4.922
proof-4.677
roit-4.670
eor-4.526
bos-4.525
Token Fra
Token ID39313
Feature activation+9.10e+00

Pos logit contributions
 holiest+2.119
emi+2.050
cean+2.023
olicy+2.014
 athlet+1.987
Neg logit contributions
chrom-3.013
proof-2.892
roit-2.876
bos-2.787
eor-2.778
Tokenizer
Token ID7509
Feature activation+3.56e+00

Pos logit contributions
emi+8.493
 holiest+8.373
 athlet+8.265
 Lanka+8.242
cean+8.229
Neg logit contributions
chrom-4.791
roit-4.274
proof-4.224
obyl-4.181
inant-4.135
Token Campbell
Token ID14327
Feature activation+9.42e-01

Pos logit contributions
 holiest+2.467
emi+2.375
 athlet+2.362
olicy+2.346
cean+2.334
Neg logit contributions
chrom-2.837
proof-2.699
roit-2.668
bos-2.639
rans-2.590
Token's
Token ID338
Feature activation+4.43e+00

Pos logit contributions
 holiest+0.696
emi+0.668
 athlet+0.667
 goodbye+0.663
cean+0.661
Neg logit contributions
chrom-0.703
proof-0.659
roit-0.658
bos-0.639
eor-0.638
Token opener
Token ID21996
Feature activation+3.38e+00

Pos logit contributions
 holiest+2.839
emi+2.792
olicy+2.750
cean+2.750
 athlet+2.728
Neg logit contributions
chrom-3.719
roit-3.523
proof-3.519
bos-3.411
eor-3.402
Token,
Token ID11
Feature activation+2.67e+00

Pos logit contributions
 holiest+2.149
emi+2.108
cean+2.061
olicy+2.060
 athlet+2.059
Neg logit contributions
chrom-2.854
proof-2.740
roit-2.720
bos-2.644
rans-2.627
Token.
Token ID13
Feature activation+1.66e+00

Pos logit contributions
 holiest+1.759
 athlet+1.724
 goodbye+1.702
emi+1.699
 bargaining+1.694
Neg logit contributions
chrom-1.999
roit-1.868
proof-1.851
bos-1.848
rans-1.821
TokenS
Token ID50
Feature activation+5.68e+00

Pos logit contributions
 holiest+1.241
emi+1.220
 athlet+1.208
 goodbye+1.205
cean+1.196
Neg logit contributions
chrom-1.213
roit-1.128
proof-1.121
eor-1.098
obyl-1.096
Token.
Token ID13
Feature activation+1.41e+00

Pos logit contributions
 holiest+2.846
emi+2.800
 athlet+2.766
 goodbye+2.755
 bargaining+2.717
Neg logit contributions
chrom-5.382
proof-5.092
roit-5.048
obyl-5.024
bos-4.963
Token economist
Token ID17646
Feature activation+5.44e+00

Pos logit contributions
 holiest+0.834
emi+0.823
 athlet+0.804
 goodbye+0.790
cean+0.786
Neg logit contributions
chrom-1.256
proof-1.201
roit-1.196
eor-1.175
bos-1.169
Token Paul
Token ID3362
Feature activation+2.29e+00

Pos logit contributions
 holiest+3.262
emi+3.211
 athlet+3.118
cean+3.106
 goodbye+3.094
Neg logit contributions
chrom-4.869
roit-4.629
proof-4.624
eor-4.525
bos-4.508
Token Krugman
Token ID46311
Feature activation+9.02e+00

Pos logit contributions
 holiest+1.301
 athlet+1.274
 goodbye+1.264
emi+1.259
cean+1.228
Neg logit contributions
chrom-2.137
roit-2.040
proof-2.018
bos-1.988
eor-1.985
Token dubbed
Token ID17494
Feature activation+6.86e-02

Pos logit contributions
 holiest+6.582
emi+6.482
 athlet+6.417
 goodbye+6.400
cean+6.311
Neg logit contributions
chrom-6.720
roit-6.228
proof-6.211
eor-6.076
obyl-6.059
Token the
Token ID262
Feature activation+1.44e+00

Pos logit contributions
 holiest+0.046
emi+0.046
 athlet+0.045
cean+0.045
olicy+0.044
Neg logit contributions
chrom-0.055
proof-0.052
roit-0.052
bos-0.051
rans-0.050
Token revision
Token ID18440
Feature activation+3.16e+00

Pos logit contributions
emi+1.037
 holiest+1.027
cean+1.017
olicy+1.004
 athlet+1.002
Neg logit contributions
chrom-1.096
proof-1.033
roit-1.027
rans-0.999
bos-0.997
Token from
Token ID422
Feature activation+1.32e+00

Pos logit contributions
 holiest+2.102
emi+2.074
 athlet+2.024
 goodbye+2.001
cean+1.992
Neg logit contributions
chrom-2.632
proof-2.488
roit-2.448
bos-2.401
rans-2.397
Token an
Token ID281
Feature activation+2.66e+00

Pos logit contributions
 holiest+0.838
emi+0.837
 athlet+0.811
cean+0.810
assi+0.788
Neg logit contributions
chrom-1.113
proof-1.061
roit-1.052
bos-1.040
rans-1.036
Token there
Token ID612
Feature activation+1.96e+00

Pos logit contributions
emi+2.012
 holiest+1.992
 athlet+1.924
olicy+1.920
cean+1.919
Neg logit contributions
chrom-2.506
proof-2.366
roit-2.349
rans-2.308
bos-2.296
Token have
Token ID423
Feature activation+2.05e+00

Pos logit contributions
 holiest+1.482
emi+1.457
 athlet+1.427
olicy+1.424
 goodbye+1.421
Neg logit contributions
chrom-1.436
proof-1.339
roit-1.326
rans-1.308
eor-1.301
Token been
Token ID587
Feature activation+3.19e-01

Pos logit contributions
 holiest+1.418
emi+1.392
olicy+1.379
 athlet+1.367
cean+1.361
Neg logit contributions
chrom-1.611
proof-1.525
roit-1.520
rans-1.505
bos-1.491
Token reports
Token ID3136
Feature activation+3.82e+00

Pos logit contributions
emi+0.260
 holiest+0.259
olicy+0.256
cean+0.255
 athlet+0.253
Neg logit contributions
chrom-0.207
proof-0.196
roit-0.190
rans-0.187
eor-0.184
Token that
Token ID326
Feature activation-2.66e-01

Pos logit contributions
 holiest+2.278
emi+2.243
olicy+2.180
 athlet+2.179
cean+2.163
Neg logit contributions
chrom-3.404
proof-3.211
roit-3.210
rans-3.143
eor-3.106
Token Bannon
Token ID19688
Feature activation+8.98e+00

Pos logit contributions
chrom+0.214
proof+0.203
roit+0.200
rans+0.199
bos+0.195
Neg logit contributions
emi-0.178
 holiest-0.176
olicy-0.171
 athlet-0.170
cean-0.168
Token might
Token ID1244
Feature activation+2.78e+00

Pos logit contributions
 holiest+5.916
emi+5.810
 athlet+5.740
 goodbye+5.717
cean+5.643
Neg logit contributions
chrom-7.379
roit-6.901
proof-6.884
eor-6.740
obyl-6.714
Token be
Token ID307
Feature activation+9.36e-01

Pos logit contributions
 holiest+2.237
emi+2.179
olicy+2.156
 Lanka+2.148
 athlet+2.142
Neg logit contributions
chrom-1.906
roit-1.775
proof-1.765
rans-1.741
eor-1.724
Token edged
Token ID45871
Feature activation+5.37e+00

Pos logit contributions
 holiest+0.641
emi+0.640
olicy+0.630
cean+0.627
 Lanka+0.619
Neg logit contributions
chrom-0.734
roit-0.692
proof-0.690
rans-0.675
bos-0.659
Token out
Token ID503
Feature activation+2.46e+00

Pos logit contributions
 holiest+3.194
emi+3.158
cean+3.070
olicy+3.039
 athlet+3.038
Neg logit contributions
chrom-4.826
proof-4.556
roit-4.522
rans-4.443
bos-4.414
Token,
Token ID11
Feature activation+3.08e+00

Pos logit contributions
 holiest+1.497
emi+1.469
olicy+1.441
cean+1.404
 Lanka+1.391
Neg logit contributions
chrom-2.185
proof-2.045
roit-2.035
rans-2.015
bos-2.004
Token.
Token ID13
Feature activation+3.37e+00

Pos logit contributions
 holiest+1.536
olicy+1.510
emi+1.504
 athlet+1.475
cean+1.469
Neg logit contributions
chrom-1.864
roit-1.744
proof-1.739
rans-1.727
eor-1.718
Token
Token ID198
Feature activation+3.22e+00

Pos logit contributions
 athlet+2.156
 holiest+2.149
emi+2.078
 goodbye+2.074
olicy+2.052
Neg logit contributions
chrom-2.875
proof-2.722
roit-2.695
eor-2.653
rans-2.639
Token
Token ID198
Feature activation+3.08e+00

Pos logit contributions
 holiest+2.155
 athlet+2.147
emi+2.067
olicy+2.051
 goodbye+2.027
Neg logit contributions
chrom-2.712
proof-2.553
roit-2.500
eor-2.462
rans-2.450
TokenPresident
Token ID10364
Feature activation+4.48e+00

Pos logit contributions
 athlet+2.204
 holiest+2.193
 goodbye+2.132
emi+2.111
 bargaining+2.108
Neg logit contributions
chrom-2.457
proof-2.317
roit-2.266
eor-2.228
rans-2.223
Token Hassan
Token ID24485
Feature activation+8.26e+00

Pos logit contributions
 holiest+2.985
 athlet+2.908
 goodbye+2.881
emi+2.869
 bargaining+2.829
Neg logit contributions
chrom-3.712
roit-3.492
proof-3.481
bos-3.406
eor-3.397
Token Rouhani
Token ID34915
Feature activation+8.97e+00

Pos logit contributions
 holiest+3.499
emi+3.410
 athlet+3.340
 goodbye+3.321
cean+3.251
Neg logit contributions
chrom-8.700
roit-8.253
proof-8.233
eor-8.109
obyl-8.093
Token has
Token ID468
Feature activation+3.76e+00

Pos logit contributions
 holiest+6.232
emi+6.128
 athlet+6.061
 goodbye+6.042
cean+5.960
Neg logit contributions
chrom-7.032
roit-6.555
proof-6.534
eor-6.397
obyl-6.373
Token shown
Token ID3402
Feature activation+9.55e-01

Pos logit contributions
 holiest+2.857
emi+2.806
olicy+2.792
 goodbye+2.751
cean+2.747
Neg logit contributions
chrom-2.747
proof-2.573
roit-2.552
rans-2.489
eor-2.452
Token he
Token ID339
Feature activation+1.94e+00

Pos logit contributions
emi+0.543
 holiest+0.543
olicy+0.541
cean+0.531
assi+0.529
Neg logit contributions
chrom-0.865
proof-0.832
roit-0.814
rans-0.799
eor-0.778
Token is
Token ID318
Feature activation+3.88e+00

Pos logit contributions
 holiest+1.429
olicy+1.385
emi+1.384
 Lanka+1.373
 goodbye+1.370
Neg logit contributions
chrom-1.462
proof-1.375
roit-1.352
rans-1.348
eor-1.327
Token listening
Token ID8680
Feature activation+4.00e+00

Pos logit contributions
 holiest+2.474
emi+2.455
assi+2.412
olicy+2.406
cean+2.394
Neg logit contributions
chrom-3.298
proof-3.172
roit-3.145
rans-3.067
bos-3.001
Token of
Token ID286
Feature activation+7.35e-01

Pos logit contributions
chrom+1.181
proof+1.113
roit+1.094
inant+1.081
rans+1.078
Neg logit contributions
 holiest-0.806
emi-0.790
cean-0.779
olicy-0.775
 athlet-0.763
Token the
Token ID262
Feature activation+6.14e-01

Pos logit contributions
 holiest+0.460
 athlet+0.453
emi+0.448
olicy+0.442
 goodbye+0.436
Neg logit contributions
chrom-0.630
proof-0.592
roit-0.586
rans-0.579
eor-0.579
Token 2008
Token ID3648
Feature activation-1.04e-01

Pos logit contributions
 holiest+0.402
 athlet+0.398
olicy+0.396
emi+0.396
 goodbye+0.387
Neg logit contributions
chrom-0.504
proof-0.468
roit-0.465
rans-0.460
bos-0.456
Token film
Token ID2646
Feature activation-7.20e-01

Pos logit contributions
chrom+0.074
proof+0.070
roit+0.069
rans+0.067
bos+0.067
Neg logit contributions
 holiest-0.079
emi-0.078
olicy-0.077
 athlet-0.077
cean-0.076
Token Kung
Token ID44753
Feature activation+5.51e+00

Pos logit contributions
chrom+0.625
proof+0.594
roit+0.577
eor+0.569
rans+0.567
Neg logit contributions
 holiest-0.438
 athlet-0.431
olicy-0.423
emi-0.421
 goodbye-0.417
Token Fu
Token ID13333
Feature activation+8.91e+00

Pos logit contributions
 holiest+3.416
 athlet+3.306
emi+3.300
 goodbye+3.271
cean+3.247
Neg logit contributions
chrom-4.790
roit-4.518
proof-4.515
obyl-4.398
eor-4.395
Token Panda
Token ID41112
Feature activation+3.69e+00

Pos logit contributions
 holiest+3.050
emi+2.948
 athlet+2.882
 goodbye+2.858
cean+2.780
Neg logit contributions
chrom-10.109
roit-9.635
proof-9.611
eor-9.479
obyl-9.457
Token,
Token ID11
Feature activation+5.66e-01

Pos logit contributions
 holiest+2.408
 athlet+2.332
emi+2.314
 goodbye+2.311
cean+2.295
Neg logit contributions
chrom-3.074
proof-2.917
roit-2.885
bos-2.817
rans-2.814
Token is
Token ID318
Feature activation+2.00e+00

Pos logit contributions
 holiest+0.413
olicy+0.407
emi+0.405
cean+0.395
 goodbye+0.390
Neg logit contributions
chrom-0.426
proof-0.399
rans-0.387
bos-0.386
roit-0.386
Token an
Token ID281
Feature activation+1.97e+00

Pos logit contributions
 holiest+1.352
olicy+1.325
emi+1.325
 goodbye+1.296
cean+1.291
Neg logit contributions
chrom-1.615
proof-1.521
rans-1.484
roit-1.474
obyl-1.441
Token anthrop
Token ID17911
Feature activation+4.19e-02

Pos logit contributions
 holiest+1.300
emi+1.276
olicy+1.275
 goodbye+1.263
cean+1.242
Neg logit contributions
chrom-1.605
proof-1.514
roit-1.509
rans-1.492
eor-1.468
Token graft
Token ID38455
Feature activation+1.36e+00

Pos logit contributions
emi+0.454
 holiest+0.445
cean+0.435
olicy+0.435
 athlet+0.433
Neg logit contributions
chrom-0.523
proof-0.492
roit-0.488
bos-0.479
rans-0.477
Token last
Token ID938
Feature activation-1.58e-01

Pos logit contributions
 holiest+0.911
emi+0.902
 athlet+0.891
cean+0.888
 goodbye+0.880
Neg logit contributions
chrom-1.103
proof-1.047
roit-1.026
rans-1.002
eor-0.996
Token year
Token ID614
Feature activation+2.16e+00

Pos logit contributions
chrom+0.129
proof+0.122
roit+0.122
rans+0.119
bos+0.118
Neg logit contributions
olicy-0.100
cean-0.100
emi-0.099
 athlet-0.099
 holiest-0.099
Token as
Token ID355
Feature activation+1.08e+00

Pos logit contributions
 holiest+1.480
emi+1.433
 athlet+1.422
olicy+1.408
cean+1.408
Neg logit contributions
chrom-1.747
proof-1.629
roit-1.611
rans-1.604
bos-1.599
Token President
Token ID1992
Feature activation+2.08e+00

Pos logit contributions
emi+0.727
 holiest+0.724
cean+0.716
 athlet+0.708
olicy+0.693
Neg logit contributions
chrom-0.862
proof-0.825
roit-0.812
bos-0.804
rans-0.795
Token Xi
Token ID21313
Feature activation+8.85e+00

Pos logit contributions
 holiest+0.859
 athlet+0.824
emi+0.788
 goodbye+0.779
cean+0.779
Neg logit contributions
chrom-2.226
proof-2.153
bos-2.140
roit-2.139
eor-2.113
Token Jinping
Token ID39823
Feature activation+7.47e+00

Pos logit contributions
 holiest+2.926
emi+2.800
 athlet+2.724
 goodbye+2.695
cean+2.649
Neg logit contributions
chrom-10.247
roit-9.771
proof-9.763
eor-9.606
obyl-9.590
Token continued
Token ID3767
Feature activation+9.93e-01

Pos logit contributions
 holiest+5.429
emi+5.344
 athlet+5.239
 goodbye+5.231
cean+5.204
Neg logit contributions
chrom-5.699
proof-5.297
roit-5.297
eor-5.161
bos-5.143
Token to
Token ID284
Feature activation-8.50e-01

Pos logit contributions
emi+0.633
 holiest+0.628
assi+0.610
 athlet+0.605
cean+0.605
Neg logit contributions
chrom-0.831
proof-0.799
roit-0.785
eor-0.767
bos-0.761
Token wage
Token ID7699
Feature activation+6.12e+00

Pos logit contributions
chrom+0.579
proof+0.546
roit+0.540
eor+0.529
rans+0.508
Neg logit contributions
emi-0.672
 holiest-0.669
assi-0.648
olicy-0.646
cean-0.645
Token a
Token ID257
Feature activation+1.44e+00

Pos logit contributions
 holiest+3.713
emi+3.685
cean+3.576
 athlet+3.563
 goodbye+3.539
Neg logit contributions
chrom-5.449
proof-5.168
roit-5.108
eor-4.983
bos-4.976
Token White
Token ID2635
Feature activation+1.41e+00

Pos logit contributions
emi+0.012
 holiest+0.012
 athlet+0.012
cean+0.012
externalActionCode+0.012
Neg logit contributions
chrom-0.016
proof-0.015
roit-0.015
rans-0.015
bos-0.015
Token House
Token ID2097
Feature activation+4.55e+00

Pos logit contributions
 holiest+1.059
emi+1.008
 athlet+1.005
 bargaining+1.003
olicy+0.996
Neg logit contributions
chrom-1.051
proof-1.006
roit-0.976
eor-0.976
rans-0.975
Token chief
Token ID4039
Feature activation+4.29e+00

Pos logit contributions
 holiest+3.003
emi+2.979
 athlet+2.898
cean+2.860
 goodbye+2.855
Neg logit contributions
chrom-3.773
roit-3.555
proof-3.534
eor-3.457
rans-3.454
Token strategist
Token ID25651
Feature activation+5.61e+00

Pos logit contributions
 holiest+2.052
emi+2.042
 athlet+1.927
 goodbye+1.921
cean+1.912
Neg logit contributions
chrom-4.293
roit-4.131
proof-4.125
bos-4.018
rans-4.017
Token Stephen
Token ID7970
Feature activation+5.82e+00

Pos logit contributions
 holiest+2.498
emi+2.432
 athlet+2.374
 goodbye+2.331
cean+2.322
Neg logit contributions
chrom-5.848
proof-5.550
roit-5.550
eor-5.448
rans-5.430
Token Bannon
Token ID19688
Feature activation+8.84e+00

Pos logit contributions
 athlet+3.745
 holiest+3.716
emi+3.682
 goodbye+3.570
 bargaining+3.564
Neg logit contributions
chrom-5.021
roit-4.684
proof-4.683
rans-4.570
eor-4.549
Token during
Token ID1141
Feature activation+2.44e+00

Pos logit contributions
 holiest+5.605
emi+5.488
 athlet+5.424
 goodbye+5.391
cean+5.323
Neg logit contributions
chrom-7.497
roit-7.025
proof-7.006
eor-6.869
rans-6.841
Token the
Token ID262
Feature activation+2.39e+00

Pos logit contributions
emi+1.539
 holiest+1.538
 athlet+1.528
cean+1.505
olicy+1.473
Neg logit contributions
chrom-2.031
proof-1.945
roit-1.927
rans-1.913
eor-1.891
Token presidential
Token ID4787
Feature activation+3.84e+00

Pos logit contributions
emi+1.495
olicy+1.464
cean+1.463
 athlet+1.461
 holiest+1.443
Neg logit contributions
chrom-2.025
proof-1.945
roit-1.930
rans-1.904
bos-1.885
Token campaign
Token ID1923
Feature activation+2.90e+00

Pos logit contributions
 holiest+2.649
emi+2.617
 athlet+2.601
olicy+2.554
cean+2.539
Neg logit contributions
chrom-3.037
proof-2.847
roit-2.841
rans-2.805
bos-2.779
Token),
Token ID828
Feature activation+1.38e+00

Pos logit contributions
 holiest+2.060
emi+2.003
 athlet+1.989
 goodbye+1.960
cean+1.955
Neg logit contributions
chrom-2.273
proof-2.109
roit-2.103
rans-2.069
bos-2.057
Token with
Token ID351
Feature activation+4.02e+00

Pos logit contributions
 holiest+1.955
emi+1.917
cean+1.888
 goodbye+1.873
 athlet+1.869
Neg logit contributions
chrom-2.442
proof-2.320
roit-2.280
rans-2.243
bos-2.232
Token hundreds
Token ID5179
Feature activation+2.60e+00

Pos logit contributions
 holiest+2.525
emi+2.449
 athlet+2.433
cean+2.419
 goodbye+2.409
Neg logit contributions
chrom-3.497
proof-3.341
roit-3.270
rans-3.211
bos-3.198
Token of
Token ID286
Feature activation+3.15e+00

Pos logit contributions
 holiest+1.514
emi+1.440
 goodbye+1.429
 athlet+1.426
 Lanka+1.420
Neg logit contributions
chrom-2.385
proof-2.252
roit-2.231
rans-2.211
eor-2.194
Token thousands
Token ID4138
Feature activation+4.81e+00

Pos logit contributions
 holiest+2.019
emi+1.944
cean+1.912
 athlet+1.906
 goodbye+1.894
Neg logit contributions
chrom-2.706
proof-2.583
roit-2.533
bos-2.505
rans-2.504
Token of
Token ID286
Feature activation+3.10e+00

Pos logit contributions
 holiest+2.670
emi+2.564
 athlet+2.527
 goodbye+2.523
cean+2.506
Neg logit contributions
chrom-4.545
proof-4.295
roit-4.284
rans-4.223
eor-4.187
Token dollars
Token ID5054
Feature activation+8.82e+00

Pos logit contributions
 holiest+1.484
emi+1.383
cean+1.377
 athlet+1.346
 goodbye+1.332
Neg logit contributions
chrom-3.206
proof-3.082
roit-3.038
rans-3.014
bos-2.980
Token.
Token ID13
Feature activation+2.89e+00

Pos logit contributions
 holiest+5.285
emi+5.167
 athlet+5.099
 goodbye+5.082
cean+5.009
Neg logit contributions
chrom-7.785
roit-7.316
proof-7.302
eor-7.160
rans-7.131
Token
Token ID447
Feature activation+3.17e+00

Pos logit contributions
 holiest+1.864
 athlet+1.810
 goodbye+1.745
emi+1.744
cean+1.727
Neg logit contributions
chrom-2.540
proof-2.411
roit-2.359
eor-2.340
rans-2.315
Token
Token ID251
Feature activation+6.33e-01

Pos logit contributions
 holiest+2.423
emi+2.405
 athlet+2.381
cean+2.339
olicy+2.330
Neg logit contributions
chrom-2.294
roit-2.138
proof-2.132
rans-2.087
eor-2.075
Token
Token ID198
Feature activation+2.63e+00

Pos logit contributions
 holiest+0.402
 athlet+0.384
 goodbye+0.382
emi+0.382
cean+0.376
Neg logit contributions
chrom-0.550
proof-0.524
roit-0.514
eor-0.510
rans-0.503
Token
Token ID198
Feature activation+2.57e+00

Pos logit contributions
 athlet+1.887
 holiest+1.880
olicy+1.786
emi+1.773
 bargaining+1.754
Neg logit contributions
chrom-2.155
proof-2.023
roit-1.960
eor-1.931
bos-1.912
Token
Token ID198
Feature activation+2.48e+00

Pos logit contributions
 holiest+1.518
emi+1.463
 athlet+1.456
 goodbye+1.432
cean+1.428
Neg logit contributions
chrom-1.886
proof-1.775
roit-1.750
eor-1.719
bos-1.716
Token
Token ID198
Feature activation+2.51e+00

Pos logit contributions
 holiest+1.736
 athlet+1.735
olicy+1.661
emi+1.658
 bargaining+1.628
Neg logit contributions
chrom-2.042
proof-1.923
roit-1.862
eor-1.835
bos-1.829
TokenSection
Token ID16375
Feature activation+7.96e-01

Pos logit contributions
 holiest+1.866
 athlet+1.833
emi+1.812
 goodbye+1.803
 bargaining+1.778
Neg logit contributions
chrom-1.938
proof-1.808
roit-1.761
bos-1.736
eor-1.729
Token III
Token ID6711
Feature activation+4.83e+00

Pos logit contributions
 holiest+0.525
 athlet+0.504
emi+0.502
cean+0.502
 goodbye+0.494
Neg logit contributions
chrom-0.663
proof-0.632
roit-0.625
bos-0.611
eor-0.609
Token:
Token ID25
Feature activation+2.20e+00

Pos logit contributions
 holiest+3.309
emi+3.236
 athlet+3.199
 goodbye+3.194
cean+3.158
Neg logit contributions
chrom-3.890
proof-3.637
roit-3.627
eor-3.548
bos-3.538
Token Lean
Token ID45661
Feature activation+8.78e+00

Pos logit contributions
 holiest+1.470
 athlet+1.420
emi+1.418
cean+1.392
 goodbye+1.387
Neg logit contributions
chrom-1.838
proof-1.735
roit-1.692
bos-1.674
rans-1.667
Token Six
Token ID9699
Feature activation+1.79e+00

Pos logit contributions
 holiest+5.910
emi+5.821
 goodbye+5.746
 athlet+5.737
 bargaining+5.659
Neg logit contributions
chrom-6.964
roit-6.504
proof-6.443
obyl-6.358
eor-6.332
Token Sigma
Token ID31669
Feature activation+6.33e+00

Pos logit contributions
 holiest+1.132
emi+1.098
 athlet+1.095
 goodbye+1.090
cean+1.064
Neg logit contributions
chrom-1.538
proof-1.468
roit-1.427
eor-1.414
bos-1.402
Token
Token ID198
Feature activation+1.94e+00

Pos logit contributions
 holiest+4.288
emi+4.198
 athlet+4.137
 goodbye+4.122
cean+4.080
Neg logit contributions
chrom-5.158
proof-4.830
roit-4.795
eor-4.702
bos-4.683
Token
Token ID198
Feature activation+2.05e+00

Pos logit contributions
 athlet+1.381
 holiest+1.379
olicy+1.322
emi+1.318
 bargaining+1.291
Neg logit contributions
chrom-1.578
proof-1.490
roit-1.437
eor-1.415
rans-1.409
TokenWhat
Token ID2061
Feature activation+1.38e+00

Pos logit contributions
 holiest+1.495
 athlet+1.463
emi+1.458
 goodbye+1.441
 Lanka+1.425
Neg logit contributions
chrom-1.619
proof-1.518
roit-1.470
bos-1.446
rans-1.442
Token by
Token ID416
Feature activation+2.25e+00

Pos logit contributions
 holiest+2.671
 athlet+2.619
emi+2.614
 goodbye+2.589
cean+2.567
Neg logit contributions
chrom-2.963
proof-2.767
roit-2.754
rans-2.714
obyl-2.706
Token blocking
Token ID12013
Feature activation+1.37e+00

Pos logit contributions
 holiest+1.422
emi+1.412
 goodbye+1.392
cean+1.388
 athlet+1.387
Neg logit contributions
chrom-1.923
proof-1.794
roit-1.790
rans-1.763
eor-1.744
Token national
Token ID2260
Feature activation+1.53e+00

Pos logit contributions
emi+0.770
 holiest+0.760
 athlet+0.750
olicy+0.742
cean+0.738
Neg logit contributions
chrom-1.249
proof-1.187
roit-1.168
rans-1.162
bos-1.146
Token highway
Token ID12763
Feature activation+4.18e+00

Pos logit contributions
emi+1.163
 holiest+1.161
 athlet+1.141
olicy+1.138
 goodbye+1.121
Neg logit contributions
chrom-1.105
proof-1.024
roit-1.020
obyl-0.996
rans-0.995
Token connecting
Token ID14320
Feature activation+6.87e-01

Pos logit contributions
 holiest+2.723
emi+2.695
 athlet+2.661
cean+2.629
olicy+2.625
Neg logit contributions
chrom-3.484
roit-3.289
proof-3.283
rans-3.181
bos-3.181
Token Roh
Token ID32694
Feature activation+8.73e+00

Pos logit contributions
emi+0.413
 holiest+0.406
 athlet+0.400
olicy+0.400
cean+0.397
Neg logit contributions
chrom-0.602
proof-0.571
roit-0.564
eor-0.558
bos-0.554
Tokent
Token ID83
Feature activation+5.15e-01

Pos logit contributions
emi+8.307
 holiest+8.300
 Lanka+8.151
 goodbye+8.114
cean+8.106
Neg logit contributions
chrom-4.305
roit-3.865
proof-3.772
eor-3.690
obyl-3.634
Tokenak
Token ID461
Feature activation+4.26e+00

Pos logit contributions
 holiest+0.375
 athlet+0.368
 goodbye+0.359
 bargaining+0.358
 forgiveness+0.355
Neg logit contributions
chrom-0.393
roit-0.375
obyl-0.373
proof-0.372
rans-0.371
Token to
Token ID284
Feature activation+1.26e+00

Pos logit contributions
 holiest+2.669
emi+2.605
 athlet+2.588
 goodbye+2.578
cean+2.540
Neg logit contributions
chrom-3.665
proof-3.436
roit-3.436
rans-3.362
eor-3.361
Token Delhi
Token ID12517
Feature activation+6.00e+00

Pos logit contributions
emi+0.751
 holiest+0.749
 athlet+0.740
olicy+0.739
cean+0.730
Neg logit contributions
chrom-1.097
proof-1.043
roit-1.039
eor-1.025
bos-1.018
Token.
Token ID13
Feature activation+2.90e+00

Pos logit contributions
 holiest+3.649
emi+3.575
 athlet+3.538
 goodbye+3.519
cean+3.470
Neg logit contributions
chrom-5.231
roit-4.915
proof-4.901
eor-4.804
obyl-4.788
Token was
Token ID373
Feature activation+2.66e+00

Pos logit contributions
 holiest+2.200
emi+2.140
 athlet+2.125
 goodbye+2.123
olicy+2.118
Neg logit contributions
chrom-3.210
proof-3.044
roit-3.025
rans-2.966
bos-2.960
Token the
Token ID262
Feature activation+3.24e+00

Pos logit contributions
 holiest+1.805
emi+1.782
cean+1.781
 goodbye+1.761
olicy+1.738
Neg logit contributions
chrom-2.146
proof-2.020
roit-1.998
rans-1.941
bos-1.932
Token brig
Token ID24393
Feature activation+4.77e+00

Pos logit contributions
emi+2.098
 holiest+2.092
olicy+2.080
 goodbye+2.073
cean+2.072
Neg logit contributions
chrom-2.684
roit-2.531
proof-2.530
inant-2.454
bos-2.447
Tokenadier
Token ID38868
Feature activation+2.81e+00

Pos logit contributions
 holiest+2.908
emi+2.855
 athlet+2.824
 goodbye+2.812
 bargaining+2.744
Neg logit contributions
chrom-4.175
roit-3.923
proof-3.922
rans-3.856
eor-3.849
Token general
Token ID2276
Feature activation+1.16e+00

Pos logit contributions
 holiest+1.387
emi+1.347
 goodbye+1.334
olicy+1.317
 athlet+1.313
Neg logit contributions
chrom-2.792
proof-2.674
roit-2.665
eor-2.614
rans-2.612
Token Pier
Token ID13762
Feature activation+8.70e+00

Pos logit contributions
 holiest+0.758
emi+0.735
cean+0.730
olicy+0.726
 goodbye+0.722
Neg logit contributions
chrom-0.959
proof-0.911
roit-0.906
bos-0.882
eor-0.880
Tokena
Token ID64
Feature activation+3.27e+00

Pos logit contributions
 holiest+7.178
emi+7.141
 athlet+7.016
cean+6.997
 goodbye+6.992
Neg logit contributions
chrom-5.542
roit-5.064
proof-5.061
eor-4.926
obyl-4.921
Token G
Token ID402
Feature activation+4.09e+00

Pos logit contributions
 holiest+2.079
emi+2.007
 goodbye+2.000
 athlet+1.985
olicy+1.968
Neg logit contributions
chrom-2.811
roit-2.627
proof-2.625
eor-2.579
bos-2.566
Tokenatt
Token ID1078
Feature activation+3.85e+00

Pos logit contributions
 holiest+3.661
 goodbye+3.570
 athlet+3.550
 bargaining+3.525
emi+3.515
Neg logit contributions
chrom-2.494
roit-2.325
proof-2.292
bos-2.225
rans-2.217
Tokenes
Token ID274
Feature activation+2.84e+00

Pos logit contributions
 holiest+3.279
 goodbye+3.189
 athlet+3.188
 bargaining+3.146
emi+3.130
Neg logit contributions
chrom-2.489
roit-2.305
proof-2.260
rans-2.224
eor-2.212
Tokenchi
Token ID11072
Feature activation+4.11e+00

Pos logit contributions
 holiest+2.036
 athlet+1.954
 goodbye+1.953
emi+1.934
olicy+1.932
Neg logit contributions
chrom-2.227
roit-2.082
proof-2.068
bos-2.044
rans-2.017
Token.
Token ID13
Feature activation+2.76e+00

Pos logit contributions
 holiest+0.954
emi+0.920
 athlet+0.911
 goodbye+0.899
 Lanka+0.898
Neg logit contributions
chrom-1.195
roit-1.129
proof-1.119
eor-1.101
rans-1.093
Token
Token ID198
Feature activation+2.72e+00

Pos logit contributions
 holiest+1.769
 athlet+1.731
emi+1.710
 goodbye+1.686
cean+1.674
Neg logit contributions
chrom-2.351
proof-2.212
roit-2.208
eor-2.156
rans-2.144
Token
Token ID198
Feature activation+2.78e+00

Pos logit contributions
 holiest+1.853
 athlet+1.840
emi+1.776
olicy+1.771
 bargaining+1.738
Neg logit contributions
chrom-2.267
proof-2.135
roit-2.082
eor-2.053
bos-2.044
TokenIn
Token ID818
Feature activation+5.88e+00

Pos logit contributions
 holiest+2.007
 athlet+1.997
emi+1.933
 goodbye+1.931
 bargaining+1.926
Neg logit contributions
chrom-2.192
proof-2.054
roit-2.017
bos-1.971
rans-1.968
Token St
Token ID520
Feature activation+3.97e+00

Pos logit contributions
 holiest+3.537
emi+3.446
 athlet+3.426
 goodbye+3.386
cean+3.348
Neg logit contributions
chrom-5.232
proof-4.949
roit-4.947
eor-4.833
bos-4.816
Token Petersburg
Token ID27051
Feature activation+8.61e+00

Pos logit contributions
 holiest+2.574
 goodbye+2.495
 athlet+2.488
emi+2.481
 bargaining+2.457
Neg logit contributions
chrom-3.321
roit-3.189
proof-3.111
eor-3.081
rans-3.079
Token,
Token ID11
Feature activation+3.00e+00

Pos logit contributions
 holiest+5.034
emi+4.970
 goodbye+4.907
 athlet+4.857
 bargaining+4.765
Neg logit contributions
chrom-7.594
roit-7.097
proof-7.079
eor-6.979
obyl-6.961
Token for
Token ID329
Feature activation+3.47e+00

Pos logit contributions
 holiest+1.966
emi+1.932
 athlet+1.909
cean+1.899
olicy+1.877
Neg logit contributions
chrom-2.472
proof-2.342
roit-2.333
rans-2.300
bos-2.268
Token example
Token ID1672
Feature activation+9.01e-01

Pos logit contributions
 holiest+2.444
 athlet+2.409
emi+2.400
cean+2.369
olicy+2.365
Neg logit contributions
chrom-2.718
roit-2.545
proof-2.545
rans-2.484
eor-2.470
Token,
Token ID11
Feature activation+2.53e+00

Pos logit contributions
 holiest+0.519
emi+0.510
 goodbye+0.503
 athlet+0.501
cean+0.492
Neg logit contributions
chrom-0.811
roit-0.764
proof-0.760
eor-0.747
obyl-0.746
Token D
Token ID360
Feature activation+1.70e+00

Pos logit contributions
 holiest+1.662
emi+1.640
 athlet+1.621
cean+1.617
olicy+1.602
Neg logit contributions
chrom-2.080
proof-1.966
roit-1.945
rans-1.930
bos-1.902
Token as
Token ID355
Feature activation+3.59e+00

Pos logit contributions
 holiest+1.643
emi+1.579
 athlet+1.557
 goodbye+1.544
cean+1.543
Neg logit contributions
chrom-2.079
roit-1.958
proof-1.955
eor-1.911
rans-1.901
Token Treasury
Token ID13419
Feature activation+5.31e+00

Pos logit contributions
 holiest+2.209
emi+2.189
 athlet+2.139
 goodbye+2.102
cean+2.088
Neg logit contributions
chrom-3.128
proof-2.983
roit-2.972
bos-2.903
eor-2.881
Token secretary
Token ID7705
Feature activation+4.85e+00

Pos logit contributions
 holiest+2.176
emi+2.125
 athlet+2.083
 goodbye+2.024
cean+2.006
Neg logit contributions
chrom-5.732
proof-5.526
roit-5.523
eor-5.409
rans-5.397
Token.
Token ID13
Feature activation+2.79e+00

Pos logit contributions
 holiest+3.079
emi+2.997
 athlet+2.964
 goodbye+2.930
cean+2.917
Neg logit contributions
chrom-4.162
proof-3.920
roit-3.915
eor-3.838
rans-3.816
Token Blank
Token ID31990
Feature activation+6.52e+00

Pos logit contributions
 athlet+1.772
 holiest+1.744
emi+1.649
 goodbye+1.629
cean+1.621
Neg logit contributions
chrom-2.452
proof-2.352
roit-2.310
eor-2.287
rans-2.273
Tokenfe
Token ID5036
Feature activation+8.60e+00

Pos logit contributions
 holiest+4.803
 athlet+4.688
emi+4.618
 goodbye+4.604
olicy+4.577
Neg logit contributions
chrom-4.979
proof-4.629
roit-4.616
rans-4.555
obyl-4.505
Tokenin
Token ID259
Feature activation+5.02e+00

Pos logit contributions
emi+7.226
 holiest+7.151
 goodbye+7.015
 bargaining+6.943
 athlet+6.915
Neg logit contributions
chrom-5.396
proof-4.781
obyl-4.780
roit-4.757
rans-4.666
Token,
Token ID11
Feature activation+3.03e+00

Pos logit contributions
 holiest+3.619
 athlet+3.492
emi+3.465
 goodbye+3.465
olicy+3.461
Neg logit contributions
chrom-3.934
proof-3.677
roit-3.666
rans-3.580
eor-3.579
Token Cohn
Token ID45005
Feature activation+1.04e+01

Pos logit contributions
 holiest+2.205
emi+2.171
cean+2.147
 athlet+2.127
 Lanka+2.088
Neg logit contributions
chrom-2.320
proof-2.176
roit-2.167
bos-2.134
rans-2.115
Token
Token ID447
Feature activation+3.16e+00

Pos logit contributions
 holiest+5.804
emi+5.802
 goodbye+5.685
 athlet+5.608
cean+5.550
Neg logit contributions
chrom-9.373
roit-8.759
proof-8.758
obyl-8.592
bos-8.591
Token
Token ID247
Feature activation+3.83e+00

Pos logit contributions
 holiest+2.489
 athlet+2.444
emi+2.439
cean+2.373
olicy+2.372
Neg logit contributions
chrom-2.266
proof-2.113
roit-2.103
rans-2.033
inant-2.026
Token techniques
Token ID7605
Feature activation-5.87e+00

Pos logit contributions
chrom+5.732
proof+5.453
roit+5.306
rans+5.214
inant+5.113
Neg logit contributions
olicy-5.132
emi-5.015
 athlet-4.998
 holiest-4.993
QUI-4.930
Token to
Token ID284
Feature activation-6.45e+00

Pos logit contributions
chrom+4.186
proof+3.920
roit+3.891
rans+3.805
bos+3.786
Neg logit contributions
 holiest-3.985
olicy-3.968
cean-3.908
emi-3.872
assi-3.825
Token write
Token ID3551
Feature activation-4.52e+00

Pos logit contributions
chrom+3.918
proof+3.610
roit+3.417
rans+3.345
inant+3.322
Neg logit contributions
olicy-5.302
 holiest-5.251
emi-5.195
QUI-5.138
assi-5.077
Token to
Token ID284
Feature activation-5.94e+00

Pos logit contributions
chrom+3.601
proof+3.394
inant+3.149
bos+3.141
roit+3.140
Neg logit contributions
emi-2.958
cean-2.945
olicy-2.914
 holiest-2.889
QUI-2.885
Token multiple
Token ID3294
Feature activation-7.57e+00

Pos logit contributions
chrom+4.658
proof+4.337
bos+4.290
inant+4.129
roit+4.115
Neg logit contributions
 holiest-3.879
 athlet-3.800
emi-3.750
cean-3.740
assi-3.716
Token dat
Token ID4818
Feature activation-1.10e+01

Pos logit contributions
chrom+5.311
proof+4.978
bos+4.848
inant+4.665
rans+4.603
Neg logit contributions
emi-5.354
 athlet-5.243
 holiest-5.201
olicy-5.135
assi-5.130
Tokenacent
Token ID12643
Feature activation-8.09e+00

Pos logit contributions
chrom+4.953
rans+4.656
bos+4.508
eor+4.438
proof+4.416
Neg logit contributions
 holiest-10.919
QUI-10.393
 athlet-10.288
 forgiveness-10.110
CONCLUS-10.028
Tokeners
Token ID364
Feature activation-5.09e+00

Pos logit contributions
chrom+4.406
rans+4.365
ards+4.342
roit+4.302
eor+4.298
Neg logit contributions
 holiest-7.257
 athlet-7.186
 srfAttach-6.899
 Lanka-6.855
 forgiveness-6.813
Token in
Token ID287
Feature activation-2.00e+00

Pos logit contributions
chrom+3.491
proof+3.313
bos+3.191
inant+3.128
eor+3.118
Neg logit contributions
olicy-3.754
 holiest-3.739
emi-3.650
QUI-3.635
cean-3.579
Token different
Token ID1180
Feature activation-6.39e+00

Pos logit contributions
chrom+1.656
proof+1.534
roit+1.477
bos+1.469
rans+1.457
Neg logit contributions
 holiest-1.311
emi-1.262
 athlet-1.256
cean-1.245
olicy-1.241
Token cities
Token ID4736
Feature activation-3.09e+00

Pos logit contributions
chrom+4.205
proof+3.901
izations+3.748
bos+3.698
inant+3.683
Neg logit contributions
emi-5.013
olicy-4.853
 athlet-4.838
cean-4.794
assi-4.768
Token with
Token ID351
Feature activation-4.38e+00

Pos logit contributions
 holiest+1.439
 goodbye+1.409
emi+1.408
 athlet+1.404
 bargaining+1.382
Neg logit contributions
chrom-2.184
roit-2.046
proof-2.026
obyl-2.015
eor-2.011
Token multiple
Token ID3294
Feature activation-7.80e+00

Pos logit contributions
chrom+3.493
proof+3.318
roit+3.258
rans+3.201
bos+3.196
Neg logit contributions
olicy-2.775
emi-2.760
cean-2.719
 holiest-2.684
QUI-2.658
Token calls
Token ID3848
Feature activation-5.36e+00

Pos logit contributions
chrom+5.900
proof+5.520
roit+5.349
bos+5.319
eor+5.310
Neg logit contributions
emi-5.133
olicy-4.925
cean-4.844
QUI-4.787
 holiest-4.786
Token over
Token ID625
Feature activation-4.71e+00

Pos logit contributions
chrom+3.935
roit+3.746
proof+3.739
rans+3.658
bos+3.623
Neg logit contributions
 holiest-3.716
olicy-3.620
emi-3.565
 Lanka-3.489
 athlet-3.459
Token the
Token ID262
Feature activation-4.99e+00

Pos logit contributions
chrom+3.702
proof+3.452
rans+3.344
bos+3.317
roit+3.303
Neg logit contributions
 holiest-3.085
 athlet-3.041
olicy-3.029
emi-3.019
cean-2.983
Token satellite
Token ID11210
Feature activation-1.06e+01

Pos logit contributions
chrom+4.124
proof+3.847
bos+3.746
rans+3.738
roit+3.716
Neg logit contributions
olicy-3.007
cean-2.956
emi-2.944
 athlet-2.935
QUI-2.922
Token phone
Token ID3072
Feature activation-5.73e+00

Pos logit contributions
chrom+7.062
proof+6.661
rans+6.503
isation+6.280
roit+6.267
Neg logit contributions
olicy-7.586
 holiest-7.310
 athlet-7.285
QUI-7.272
emi-7.256
Token,
Token ID11
Feature activation-4.54e+00

Pos logit contributions
chrom+4.029
proof+3.715
roit+3.621
rans+3.607
bos+3.490
Neg logit contributions
olicy-4.223
 holiest-4.175
emi-4.144
 athlet-4.133
QUI-4.085
Token Sullivan
Token ID18501
Feature activation-2.68e+00

Pos logit contributions
chrom+3.441
proof+3.264
roit+3.190
rans+3.167
bos+3.132
Neg logit contributions
emi-3.018
olicy-2.997
 holiest-2.995
cean-2.945
QUI-2.940
Token is
Token ID318
Feature activation-2.82e+00

Pos logit contributions
chrom+1.736
proof+1.639
roit+1.598
rans+1.572
bos+1.568
Neg logit contributions
 holiest-2.184
olicy-2.129
 athlet-2.088
emi-2.076
 Lanka-2.045
Token finally
Token ID3443
Feature activation-4.43e+00

Pos logit contributions
chrom+2.137
proof+2.064
roit+2.030
rans+1.983
bos+1.939
Neg logit contributions
emi-1.898
 holiest-1.852
olicy-1.845
cean-1.839
assi-1.829
Token
Token ID250
Feature activation-4.43e+00

Pos logit contributions
chrom+4.302
eor+4.030
proof+3.863
roit+3.855
rans+3.808
Neg logit contributions
 holiest-4.787
 goodbye-4.644
 athlet-4.635
emi-4.622
cean-4.590
TokenBig
Token ID12804
Feature activation-4.96e+00

Pos logit contributions
chrom+3.637
proof+3.421
eor+3.343
rans+3.332
roit+3.330
Neg logit contributions
 holiest-2.761
 athlet-2.714
emi-2.684
 Lanka-2.650
 bargaining-2.639
Tokengest
Token ID3495
Feature activation-3.92e+00

Pos logit contributions
chrom+3.977
proof+3.873
eor+3.753
bos+3.717
roit+3.651
Neg logit contributions
 holiest-3.068
 athlet-2.999
olicy-2.976
emi-2.949
QUI-2.872
Token at
Token ID379
Feature activation-5.55e+00

Pos logit contributions
chrom+3.109
proof+3.004
eor+2.850
roit+2.850
rans+2.834
Neg logit contributions
emi-2.511
olicy-2.478
 holiest-2.414
cean-2.410
 athlet-2.362
Token-
Token ID12
Feature activation-9.18e+00

Pos logit contributions
chrom+4.531
roit+4.335
eor+4.258
rans+4.254
proof+4.252
Neg logit contributions
 holiest-3.369
olicy-3.351
 athlet-3.295
cean-3.235
 Lanka-3.198
Tokenbat
Token ID8664
Feature activation-1.03e+01

Pos logit contributions
chrom+5.591
proof+5.251
rans+5.151
bos+5.049
obyl+4.955
Neg logit contributions
 holiest-7.562
 athlet-7.290
cean-7.207
 goodbye-7.180
olicy-7.158
Token of
Token ID286
Feature activation-3.84e+00

Pos logit contributions
chrom+7.425
proof+7.144
roit+7.075
eor+6.951
inant+6.924
Neg logit contributions
olicy-6.434
 holiest-6.182
cean-6.166
jri-6.049
 Lanka-6.045
Token my
Token ID616
Feature activation-3.11e+00

Pos logit contributions
chrom+3.093
roit+3.008
proof+2.929
eor+2.907
bos+2.844
Neg logit contributions
olicy-2.443
 holiest-2.362
 athlet-2.339
cean-2.295
emi-2.261
Token life
Token ID1204
Feature activation-3.45e+00

Pos logit contributions
chrom+2.612
roit+2.489
eor+2.449
proof+2.416
rans+2.382
Neg logit contributions
olicy-1.873
 holiest-1.868
emi-1.860
 athlet-1.835
cean-1.824
Token,
Token ID11
Feature activation-3.63e+00

Pos logit contributions
chrom+2.574
roit+2.417
proof+2.396
eor+2.369
bos+2.357
Neg logit contributions
 holiest-2.425
olicy-2.392
cean-2.312
emi-2.284
 Lanka-2.264
Token
Token ID447
Feature activation-4.35e+00

Pos logit contributions
chrom+2.979
roit+2.799
proof+2.786
bos+2.730
rans+2.722
Neg logit contributions
olicy-2.199
 holiest-2.171
emi-2.138
 Lanka-2.124
cean-2.121
Token media
Token ID2056
Feature activation-7.77e+00

Pos logit contributions
 holiest+0.898
emi+0.872
 goodbye+0.868
 athlet+0.866
 bargaining+0.854
Neg logit contributions
chrom-1.250
roit-1.174
proof-1.166
eor-1.151
obyl-1.149
Token,
Token ID11
Feature activation-5.54e+00

Pos logit contributions
chrom+5.594
proof+5.348
roit+5.259
rans+5.093
eor+5.031
Neg logit contributions
olicy-5.247
cean-5.212
emi-5.138
 holiest-5.135
assi-5.061
Token saying
Token ID2282
Feature activation-6.37e+00

Pos logit contributions
chrom+4.210
proof+3.980
roit+3.954
rans+3.933
bos+3.864
Neg logit contributions
emi-3.613
cean-3.520
 holiest-3.518
olicy-3.494
assi-3.421
Token an
Token ID281
Feature activation-4.38e+00

Pos logit contributions
chrom+4.587
proof+4.427
roit+4.352
bos+4.313
rans+4.232
Neg logit contributions
emi-4.358
olicy-4.266
cean-4.230
 holiest-4.217
QUI-4.092
Token oft
Token ID28639
Feature activation-3.11e+00

Pos logit contributions
chrom+3.341
roit+3.268
proof+3.254
rans+3.161
eor+3.063
Neg logit contributions
emi-2.784
 holiest-2.775
olicy-2.745
cean-2.702
 Lanka-2.696
Token-
Token ID12
Feature activation-1.01e+01

Pos logit contributions
chrom+2.453
rans+2.277
roit+2.270
proof+2.236
bos+2.220
Neg logit contributions
 holiest-2.113
 Lanka-2.079
emi-2.078
olicy-2.025
 athlet-2.016
Tokenqu
Token ID421
Feature activation-8.75e+00

Pos logit contributions
chrom+5.836
rans+5.469
proof+5.264
roit+5.156
identified+5.145
Neg logit contributions
 holiest-9.071
 Lanka-8.647
 goodbye-8.487
 srfAttach-8.417
 athlet-8.407
Tokenoted
Token ID5191
Feature activation-5.34e+00

Pos logit contributions
roit+1.482
chrom+1.426
rans+1.345
bos+1.267
inant+1.188
Neg logit contributions
 bargaining-11.126
 holiest-11.040
 goodbye-10.982
 srfAttach-10.821
 forgiveness-10.708
Token blog
Token ID4130
Feature activation-5.92e+00

Pos logit contributions
chrom+4.032
proof+3.929
rans+3.757
roit+3.687
bos+3.630
Neg logit contributions
olicy-3.533
emi-3.509
cean-3.437
assi-3.383
 holiest-3.362
Token post
Token ID1281
Feature activation-7.15e+00

Pos logit contributions
chrom+3.731
proof+3.610
roit+3.496
eor+3.444
rans+3.425
Neg logit contributions
 holiest-4.766
 athlet-4.658
emi-4.643
 Lanka-4.526
 srfAttach-4.486
Token from
Token ID422
Feature activation-4.65e+00

Pos logit contributions
chrom+4.844
proof+4.755
roit+4.567
bos+4.410
ards+4.355
Neg logit contributions
olicy-5.120
emi-5.037
 srfAttach-4.988
 holiest-4.982
cean-4.866
Token skin
Token ID4168
Feature activation-7.29e+00

Pos logit contributions
 holiest+0.698
emi+0.680
 goodbye+0.677
 athlet+0.675
 bargaining+0.665
Neg logit contributions
chrom-0.982
roit-0.924
proof-0.917
eor-0.906
obyl-0.903
Token condition
Token ID4006
Feature activation-8.32e+00

Pos logit contributions
chrom+5.664
proof+5.271
roit+5.212
rans+5.025
obyl+4.990
Neg logit contributions
cean-4.702
 holiest-4.658
olicy-4.647
 athlet-4.627
 Lanka-4.618
Token,
Token ID11
Feature activation-5.51e+00

Pos logit contributions
chrom+5.838
proof+5.521
roit+5.296
rans+5.227
bos+5.210
Neg logit contributions
cean-5.757
olicy-5.714
assi-5.585
 holiest-5.546
 Lanka-5.476
Token an
Token ID281
Feature activation-4.50e+00

Pos logit contributions
chrom+4.219
proof+3.955
roit+3.894
bos+3.839
rans+3.815
Neg logit contributions
cean-3.611
olicy-3.589
emi-3.537
 holiest-3.532
assi-3.511
Token autos
Token ID44619
Feature activation-6.18e+00

Pos logit contributions
chrom+3.451
roit+3.323
proof+3.200
rans+3.166
eor+3.126
Neg logit contributions
 holiest-2.997
 goodbye-2.941
 srfAttach-2.919
 Lanka-2.912
 bargaining-2.887
Tokenomal
Token ID18048
Feature activation-1.01e+01

Pos logit contributions
chrom+3.311
rans+2.778
roit+2.727
inant+2.702
proof+2.681
Neg logit contributions
 bargaining-5.835
 goodbye-5.829
 holiest-5.807
 athlet-5.767
 Lanka-5.659
Token dominant
Token ID11410
Feature activation-6.24e+00

Pos logit contributions
chrom+7.089
inant+6.303
roit+6.139
proof+6.121
isation+5.824
Neg logit contributions
olicy-7.639
cean-7.367
emi-7.306
 holiest-7.174
assi-7.169
Token palm
Token ID18057
Feature activation-3.22e+00

Pos logit contributions
chrom+4.339
roit+3.778
proof+3.665
inant+3.657
rans+3.641
Neg logit contributions
olicy-4.948
emi-4.892
cean-4.876
 holiest-4.732
assi-4.675
Tokenopl
Token ID20106
Feature activation-3.30e+00

Pos logit contributions
chrom+2.438
roit+2.258
inant+2.251
proof+2.245
rans+2.203
Neg logit contributions
 holiest-2.301
olicy-2.222
cean-2.205
emi-2.204
 Lanka-2.182
Tokenant
Token ID415
Feature activation-8.29e+00

Pos logit contributions
chrom+1.545
proof+1.362
rans+1.352
roit+1.343
eor+1.303
Neg logit contributions
 holiest-3.425
 bargaining-3.338
 goodbye-3.303
 athlet-3.249
QUI-3.237
Tokenar
Token ID283
Feature activation-5.85e+00

Pos logit contributions
chrom+5.950
inant+5.473
rans+5.403
proof+5.370
roit+5.326
Neg logit contributions
 holiest-6.164
olicy-5.928
cean-5.794
 goodbye-5.785
 bargaining-5.732
Token a
Token ID257
Feature activation+1.14e+00

Pos logit contributions
chrom+2.362
proof+2.298
bos+2.244
rans+2.187
roit+2.182
Neg logit contributions
emi-2.194
cean-2.183
olicy-2.153
CONCLUS-2.120
assi-2.105
Token junior
Token ID13430
Feature activation-6.49e+00

Pos logit contributions
 holiest+0.698
emi+0.680
 goodbye+0.677
 athlet+0.675
 bargaining+0.665
Neg logit contributions
chrom-0.982
roit-0.924
proof-0.917
eor-0.906
obyl-0.903
Token to
Token ID284
Feature activation-7.00e+00

Pos logit contributions
chrom+4.831
roit+4.684
proof+4.549
rans+4.481
eor+4.458
Neg logit contributions
 holiest-4.153
olicy-4.146
emi-4.089
QUI-4.065
assi-3.955
Token earn
Token ID5160
Feature activation-3.06e+00

Pos logit contributions
chrom+5.213
roit+5.113
proof+5.001
rans+4.987
bos+4.854
Neg logit contributions
emi-4.333
 holiest-4.325
olicy-4.306
cean-4.177
QUI-4.152
Token all
Token ID477
Feature activation-7.16e+00

Pos logit contributions
chrom+2.510
proof+2.415
roit+2.339
bos+2.293
rans+2.254
Neg logit contributions
olicy-1.886
 holiest-1.873
cean-1.865
QUI-1.846
emi-1.842
Token-
Token ID12
Feature activation-1.00e+01

Pos logit contributions
chrom+5.409
proof+5.182
roit+5.030
eor+4.816
rans+4.814
Neg logit contributions
 holiest-4.517
emi-4.484
 srfAttach-4.473
olicy-4.429
QUI-4.392
Tokenstate
Token ID5219
Feature activation-8.54e+00

Pos logit contributions
chrom+5.633
proof+5.366
roit+5.091
rans+5.006
bos+4.827
Neg logit contributions
 holiest-8.575
olicy-8.205
 Lanka-8.171
 goodbye-8.167
 athlet-8.073
Token honors
Token ID25279
Feature activation-3.73e+00

Pos logit contributions
chrom+6.251
roit+6.085
proof+5.998
bos+5.932
inant+5.776
Neg logit contributions
olicy-5.472
emi-5.369
assi-5.297
 holiest-5.275
QUI-5.159
Token.
Token ID13
Feature activation-3.91e+00

Pos logit contributions
chrom+2.733
roit+2.650
proof+2.569
bos+2.506
inant+2.458
Neg logit contributions
 holiest-2.642
emi-2.618
cean-2.547
olicy-2.546
assi-2.512
Token
Token ID198
Feature activation-3.52e+00

Pos logit contributions
chrom+3.151
proof+2.941
roit+2.911
bos+2.872
rans+2.823
Neg logit contributions
 holiest-2.610
 athlet-2.534
olicy-2.479
emi-2.444
cean-2.440
Token
Token ID198
Feature activation-3.24e+00

Pos logit contributions
chrom+2.558
proof+2.425
bos+2.229
roit+2.227
rans+2.214
Neg logit contributions
 athlet-2.927
 holiest-2.845
 nomine-2.774
 referen-2.770
olicy-2.769
Token drama
Token ID10512
Feature activation-5.84e+00

Pos logit contributions
chrom+0.674
roit+0.633
proof+0.629
eor+0.620
obyl+0.619
Neg logit contributions
 holiest-0.479
emi-0.469
 athlet-0.465
 goodbye-0.464
 bargaining-0.457
Token parody
Token ID29695
Feature activation-6.89e+00

Pos logit contributions
chrom+4.256
proof+4.049
roit+4.014
rans+3.869
inant+3.844
Neg logit contributions
 holiest-3.909
emi-3.865
cean-3.842
olicy-3.814
 athlet-3.773
Token sk
Token ID1341
Feature activation-6.59e+00

Pos logit contributions
chrom+4.936
proof+4.636
roit+4.618
rans+4.381
inant+4.379
Neg logit contributions
emi-4.764
 holiest-4.646
cean-4.613
olicy-4.527
QUI-4.509
Tokenit
Token ID270
Feature activation-4.27e+00

Pos logit contributions
chrom+2.643
roit+2.530
bos+2.520
rans+2.498
eor+2.451
Neg logit contributions
 holiest-6.853
 bargaining-6.825
 athlet-6.732
QUI-6.676
 goodbye-6.651
Token "
Token ID366
Feature activation-4.53e+00

Pos logit contributions
chrom+3.204
roit+3.016
proof+3.006
rans+2.957
eor+2.892
Neg logit contributions
 holiest-2.934
cean-2.842
emi-2.830
 athlet-2.781
olicy-2.768
TokenVeter
Token ID45182
Feature activation-9.92e+00

Pos logit contributions
chrom+3.467
proof+3.295
bos+3.121
roit+3.118
eor+3.079
Neg logit contributions
 athlet-3.186
 holiest-3.175
emi-3.069
olicy-3.062
 Lanka-3.053
Tokenin
Token ID259
Feature activation-3.43e+00

Pos logit contributions
chrom+4.108
roit+3.941
inant+3.804
rans+3.732
proof+3.692
Neg logit contributions
 athlet-10.074
 holiest-9.884
 goodbye-9.681
 Lanka-9.581
olicy-9.448
Tokenarian
Token ID3699
Feature activation-6.00e+00

Pos logit contributions
chrom+1.920
proof+1.735
ards+1.694
obyl+1.683
roit+1.666
Neg logit contributions
 holiest-3.149
 forgiveness-3.057
 bargaining-3.052
 athlet-3.049
 goodbye-3.013
Token's
Token ID338
Feature activation-2.92e+00

Pos logit contributions
chrom+4.776
proof+4.636
roit+4.463
bos+4.434
eor+4.312
Neg logit contributions
 holiest-3.739
 athlet-3.594
olicy-3.589
emi-3.530
externalActionCode-3.504
Token Hospital
Token ID9256
Feature activation-6.72e+00

Pos logit contributions
chrom+2.510
proof+2.455
bos+2.368
roit+2.363
rans+2.278
Neg logit contributions
emi-1.649
cean-1.625
 athlet-1.607
olicy-1.603
 holiest-1.587
Token",
Token ID1600
Feature activation-3.71e+00

Pos logit contributions
chrom+5.015
roit+4.940
proof+4.936
bos+4.812
izations+4.808
Neg logit contributions
 holiest-4.311
 athlet-4.239
cean-4.138
olicy-4.086
CONCLUS-4.067
Token.
Token ID13
Feature activation-3.80e+00

Pos logit contributions
chrom+4.587
proof+4.270
bos+4.225
eor+4.200
roit+4.158
Neg logit contributions
olicy-4.629
 holiest-4.390
 Lanka-4.281
emi-4.264
QUI-4.240
Token Typically
Token ID27095
Feature activation-4.98e+00

Pos logit contributions
chrom+3.166
proof+2.940
eor+2.924
roit+2.841
bos+2.839
Neg logit contributions
 athlet-2.508
 holiest-2.438
olicy-2.430
 bargaining-2.346
 goodbye-2.307
Token the
Token ID262
Feature activation-3.53e+00

Pos logit contributions
chrom+3.686
bos+3.420
proof+3.392
rans+3.341
eor+3.334
Neg logit contributions
olicy-3.547
emi-3.462
 holiest-3.364
assi-3.337
 athlet-3.334
Token most
Token ID749
Feature activation-3.49e+00

Pos logit contributions
chrom+2.811
bos+2.552
proof+2.543
eor+2.540
roit+2.511
Neg logit contributions
olicy-2.399
emi-2.325
assi-2.271
 athlet-2.256
QUI-2.246
Token eas
Token ID1396
Feature activation-2.70e+00

Pos logit contributions
chrom+3.050
roit+2.861
proof+2.813
eor+2.776
bos+2.729
Neg logit contributions
olicy-2.123
emi-2.098
 srfAttach-1.971
assi-1.933
QUI-1.917
Tokenter
Token ID353
Feature activation-9.92e+00

Pos logit contributions
chrom+1.039
proof+0.926
roit+0.874
bos+0.858
rans+0.843
Neg logit contributions
 holiest-2.988
 goodbye-2.963
 bargaining-2.939
emi-2.934
 athlet-2.917
Tokenly
Token ID306
Feature activation-3.76e+00

Pos logit contributions
chrom+5.508
proof+5.442
bos+5.344
roit+5.107
rans+4.962
Neg logit contributions
emi-8.471
olicy-8.352
 holiest-8.039
 goodbye-8.033
 forgiveness-7.860
Token aim
Token ID4031
Feature activation-1.90e+00

Pos logit contributions
chrom+2.917
bos+2.675
proof+2.658
eor+2.626
roit+2.549
Neg logit contributions
olicy-2.656
emi-2.609
assi-2.498
QUI-2.485
 athlet-2.471
Token point
Token ID966
Feature activation-2.85e+00

Pos logit contributions
chrom+1.505
proof+1.425
roit+1.386
eor+1.367
bos+1.353
Neg logit contributions
 athlet-1.303
emi-1.302
 holiest-1.294
 goodbye-1.271
olicy-1.271
Token was
Token ID373
Feature activation-2.58e+00

Pos logit contributions
chrom+2.125
proof+1.974
bos+1.959
eor+1.924
roit+1.911
Neg logit contributions
olicy-2.076
emi-2.013
 Lanka-1.941
 holiest-1.935
QUI-1.929
Token targeted
Token ID7977
Feature activation-5.52e+00

Pos logit contributions
chrom+1.964
proof+1.835
roit+1.816
bos+1.803
eor+1.784
Neg logit contributions
olicy-1.785
emi-1.749
 holiest-1.735
QUI-1.722
assi-1.703
Token are
Token ID389
Feature activation-5.37e+00

Pos logit contributions
chrom+3.631
proof+3.305
roit+3.283
bos+3.253
rans+3.248
Neg logit contributions
 holiest-4.312
olicy-4.205
 srfAttach-4.190
emi-4.129
cean-4.104
Token here
Token ID994
Feature activation-7.06e+00

Pos logit contributions
chrom+4.051
proof+3.921
roit+3.808
rans+3.801
bos+3.793
Neg logit contributions
olicy-3.514
emi-3.451
assi-3.392
cean-3.392
 holiest-3.340
Token to
Token ID284
Feature activation-6.43e+00

Pos logit contributions
chrom+4.792
bos+4.650
proof+4.636
roit+4.602
inant+4.439
Neg logit contributions
olicy-5.050
 holiest-4.996
emi-4.890
assi-4.811
cean-4.774
Token protect
Token ID1805
Feature activation-7.07e+00

Pos logit contributions
chrom+3.540
proof+3.427
roit+3.304
rans+3.182
bos+3.077
Neg logit contributions
olicy-5.429
emi-5.400
 holiest-5.388
assi-5.237
cean-5.198
Token the
Token ID262
Feature activation-4.58e+00

Pos logit contributions
chrom+5.304
roit+5.082
proof+5.062
rans+4.933
bos+4.906
Neg logit contributions
olicy-4.528
emi-4.512
cean-4.401
assi-4.377
 holiest-4.254
Token civilians
Token ID10380
Feature activation-9.82e+00

Pos logit contributions
chrom+3.707
proof+3.431
rans+3.429
roit+3.426
bos+3.334
Neg logit contributions
emi-2.874
olicy-2.842
assi-2.811
cean-2.773
 versa-2.686
Token and
Token ID290
Feature activation-3.77e+00

Pos logit contributions
proof+6.312
chrom+6.297
rans+6.218
roit+5.949
ards+5.936
Neg logit contributions
olicy-6.899
 srfAttach-6.763
QUI-6.586
CONCLUS-6.559
emi-6.549
Token to
Token ID284
Feature activation-5.17e+00

Pos logit contributions
chrom+2.686
proof+2.598
roit+2.573
bos+2.491
rans+2.474
Neg logit contributions
emi-2.654
cean-2.605
 holiest-2.575
olicy-2.551
 athlet-2.506
Token free
Token ID1479
Feature activation-3.03e+00

Pos logit contributions
chrom+2.725
proof+2.617
roit+2.474
rans+2.402
bos+2.361
Neg logit contributions
olicy-4.576
emi-4.564
 holiest-4.501
cean-4.421
assi-4.414
Token their
Token ID511
Feature activation-3.87e+00

Pos logit contributions
chrom+2.356
proof+2.255
bos+2.196
rans+2.172
roit+2.171
Neg logit contributions
olicy-1.994
emi-1.983
cean-1.967
 holiest-1.932
 athlet-1.905
Token city
Token ID1748
Feature activation-4.84e+00

Pos logit contributions
chrom+2.801
proof+2.674
roit+2.566
rans+2.551
bos+2.509
Neg logit contributions
emi-2.782
cean-2.727
olicy-2.689
assi-2.622
 srfAttach-2.619
Tokenau
Token ID559
Feature activation-5.16e+00

Pos logit contributions
chrom+2.514
roit+2.348
proof+2.341
eor+2.293
obyl+2.284
Neg logit contributions
 holiest-2.109
emi-2.073
 athlet-2.049
 goodbye-2.043
cean-2.014
Token,
Token ID11
Feature activation-4.94e+00

Pos logit contributions
chrom+4.045
proof+3.779
roit+3.772
rans+3.724
obyl+3.708
Neg logit contributions
 holiest-3.409
 athlet-3.306
olicy-3.278
 goodbye-3.192
emi-3.155
Token who
Token ID508
Feature activation-7.82e+00

Pos logit contributions
chrom+4.021
proof+3.797
roit+3.794
rans+3.693
bos+3.688
Neg logit contributions
 holiest-2.977
emi-2.927
olicy-2.914
cean-2.885
 athlet-2.844
Token is
Token ID318
Feature activation-4.32e+00

Pos logit contributions
chrom+5.370
proof+5.034
bos+4.953
roit+4.932
rans+4.819
Neg logit contributions
olicy-5.630
 holiest-5.418
QUI-5.355
CONCLUS-5.338
emi-5.332
Token partially
Token ID12387
Feature activation-5.29e+00

Pos logit contributions
chrom+3.216
proof+3.103
roit+2.988
rans+2.972
bos+2.855
Neg logit contributions
emi-2.853
olicy-2.850
cean-2.803
 Lanka-2.765
 versa-2.756
Token disabled
Token ID10058
Feature activation-9.78e+00

Pos logit contributions
chrom+3.954
proof+3.868
rans+3.628
roit+3.522
ards+3.470
Neg logit contributions
cean-3.565
emi-3.539
olicy-3.530
 Lanka-3.443
 holiest-3.357
Token,
Token ID11
Feature activation-4.89e+00

Pos logit contributions
chrom+6.500
proof+6.196
roit+6.064
rans+5.887
bos+5.835
Neg logit contributions
olicy-6.841
emi-6.807
 holiest-6.776
assi-6.673
cean-6.541
Token told
Token ID1297
Feature activation-6.69e+00

Pos logit contributions
chrom+3.460
proof+3.250
bos+3.213
rans+3.172
roit+3.162
Neg logit contributions
olicy-3.567
cean-3.438
 holiest-3.419
emi-3.417
QUI-3.369
Token authorities
Token ID4773
Feature activation-5.28e+00

Pos logit contributions
chrom+5.050
roit+4.828
proof+4.746
bos+4.671
rans+4.662
Neg logit contributions
 holiest-4.313
olicy-4.205
emi-4.195
cean-4.118
 athlet-4.053
Token that
Token ID326
Feature activation-6.64e+00

Pos logit contributions
chrom+3.667
proof+3.462
bos+3.397
roit+3.358
rans+3.322
Neg logit contributions
 holiest-3.827
olicy-3.801
emi-3.793
cean-3.698
QUI-3.638
Token he
Token ID339
Feature activation-5.25e+00

Pos logit contributions
chrom+4.736
proof+4.525
bos+4.407
roit+4.397
rans+4.295
Neg logit contributions
olicy-4.527
 holiest-4.509
emi-4.500
cean-4.354
QUI-4.324
Token to
Token ID284
Feature activation-1.45e-01

Pos logit contributions
chrom+1.697
proof+1.566
bos+1.545
roit+1.528
inant+1.491
Neg logit contributions
 holiest-2.150
cean-2.134
emi-2.133
olicy-2.122
QUI-2.113
Token the
Token ID262
Feature activation-5.72e+00

Pos logit contributions
chrom+0.123
roit+0.116
proof+0.115
eor+0.113
obyl+0.113
Neg logit contributions
 holiest-0.091
emi-0.089
 goodbye-0.088
 athlet-0.088
 bargaining-0.086
Token bottom
Token ID4220
Feature activation-7.36e+00

Pos logit contributions
chrom+4.691
proof+4.359
roit+4.346
bos+4.312
rans+4.261
Neg logit contributions
olicy-3.403
emi-3.387
cean-3.351
assi-3.257
QUI-3.251
Token of
Token ID286
Feature activation-5.49e+00

Pos logit contributions
chrom+5.728
proof+5.349
roit+5.323
rans+5.296
eor+5.168
Neg logit contributions
emi-4.384
QUI-4.371
cean-4.371
 athlet-4.363
 holiest-4.340
Token the
Token ID262
Feature activation-5.71e+00

Pos logit contributions
chrom+4.199
proof+3.907
roit+3.803
eor+3.776
rans+3.772
Neg logit contributions
 athlet-3.511
olicy-3.499
 holiest-3.485
cean-3.460
emi-3.450
Token container
Token ID9290
Feature activation-9.78e+00

Pos logit contributions
chrom+4.108
proof+3.775
roit+3.773
bos+3.720
eor+3.656
Neg logit contributions
olicy-3.990
cean-3.860
emi-3.848
 athlet-3.804
QUI-3.767
Token.
Token ID13
Feature activation-5.28e+00

Pos logit contributions
chrom+6.674
proof+6.468
roit+6.135
rans+6.107
obyl+6.052
Neg logit contributions
 holiest-6.581
olicy-6.503
emi-6.420
QUI-6.345
cean-6.249
TokenWhen
Token ID2215
Feature activation-6.81e+00

Pos logit contributions
chrom+4.247
proof+4.003
roit+3.861
eor+3.837
obyl+3.818
Neg logit contributions
 holiest-3.391
 athlet-3.381
emi-3.270
 bargaining-3.253
cean-3.236
Token moisture
Token ID20160
Feature activation-6.90e+00

Pos logit contributions
chrom+4.995
proof+4.711
roit+4.587
rans+4.549
eor+4.543
Neg logit contributions
 holiest-4.631
olicy-4.558
emi-4.514
 athlet-4.475
cean-4.357
Token starts
Token ID4940
Feature activation-3.47e+00

Pos logit contributions
proof+4.941
chrom+4.824
rans+4.510
roit+4.462
ards+4.304
Neg logit contributions
olicy-4.802
 holiest-4.698
 Lanka-4.691
 athlet-4.660
emi-4.657
Token to
Token ID284
Feature activation-5.55e+00

Pos logit contributions
chrom+2.584
rans+2.497
proof+2.481
roit+2.476
bos+2.445
Neg logit contributions
olicy-2.423
emi-2.362
 holiest-2.351
 Lanka-2.334
cean-2.294
Token is
Token ID318
Feature activation+1.66e+00

Pos logit contributions
chrom+1.225
rans+1.171
proof+1.166
roit+1.160
bos+1.118
Neg logit contributions
 holiest-1.416
cean-1.367
olicy-1.366
 srfAttach-1.363
emi-1.361
Token a
Token ID257
Feature activation-5.26e+00

Pos logit contributions
 holiest+0.983
emi+0.959
 goodbye+0.956
 athlet+0.955
 bargaining+0.936
Neg logit contributions
chrom-1.466
roit-1.378
proof-1.367
eor-1.355
obyl-1.350
Token play
Token ID711
Feature activation-5.31e+00

Pos logit contributions
chrom+4.590
proof+4.284
roit+4.158
rans+4.067
inant+4.060
Neg logit contributions
olicy-3.059
emi-2.998
cean-2.887
 athlet-2.875
assi-2.867
Token-
Token ID12
Feature activation-1.02e+01

Pos logit contributions
chrom+4.068
proof+3.990
roit+3.878
bos+3.786
inant+3.712
Neg logit contributions
 holiest-3.542
 Lanka-3.485
olicy-3.377
cean-3.340
emi-3.336
Tokenby
Token ID1525
Feature activation-7.18e+00

Pos logit contributions
chrom+5.359
proof+5.209
bos+4.857
rans+4.801
roit+4.735
Neg logit contributions
 Lanka-9.213
 holiest-9.027
 bargaining-8.741
 srfAttach-8.627
cean-8.618
Token-
Token ID12
Feature activation-9.76e+00

Pos logit contributions
chrom+4.995
proof+4.889
roit+4.739
rans+4.493
eor+4.388
Neg logit contributions
olicy-5.395
 athlet-5.296
 holiest-5.262
retty-5.042
 Lanka-4.998
Tokenplay
Token ID1759
Feature activation-5.73e+00

Pos logit contributions
chrom+2.714
proof+2.478
rans+2.048
bos+1.902
roit+1.855
Neg logit contributions
 holiest-12.154
 athlet-11.935
 referen-11.675
olicy-11.570
 Lanka-11.450
Token guy
Token ID3516
Feature activation-5.83e+00

Pos logit contributions
chrom+4.150
roit+4.033
proof+3.942
inant+3.830
rans+3.802
Neg logit contributions
olicy-3.882
emi-3.859
 holiest-3.788
cean-3.723
QUI-3.719
Token who
Token ID508
Feature activation-8.23e+00

Pos logit contributions
chrom+4.016
proof+3.849
roit+3.774
bos+3.731
rans+3.716
Neg logit contributions
olicy-4.249
 holiest-4.217
emi-4.094
 srfAttach-4.072
 Lanka-4.037
Token is
Token ID318
Feature activation-3.37e+00

Pos logit contributions
chrom+5.400
proof+5.147
roit+5.058
bos+4.990
rans+4.929
Neg logit contributions
olicy-6.157
 srfAttach-5.960
 holiest-5.949
emi-5.889
QUI-5.745
Token not
Token ID407
Feature activation-6.00e+00

Pos logit contributions
chrom+2.598
proof+2.569
roit+2.525
rans+2.444
bos+2.408
Neg logit contributions
olicy-2.217
emi-2.156
 holiest-2.116
cean-2.114
 Lanka-2.085
Token it
Token ID340
Feature activation-5.44e+00

Pos logit contributions
chrom+3.631
proof+3.360
roit+3.333
bos+3.318
eor+3.295
Neg logit contributions
emi-3.089
olicy-3.071
 holiest-3.042
 athlet-3.024
cean-2.953
Token is
Token ID318
Feature activation-3.88e+00

Pos logit contributions
chrom+3.858
proof+3.618
bos+3.533
roit+3.488
rans+3.472
Neg logit contributions
olicy-4.104
emi-3.911
cean-3.813
 holiest-3.778
 Lanka-3.766
Token directed
Token ID7924
Feature activation-5.46e+00

Pos logit contributions
chrom+3.239
proof+3.054
bos+2.944
rans+2.921
roit+2.914
Neg logit contributions
emi-2.443
cean-2.393
olicy-2.369
 holiest-2.299
assi-2.266
Token or
Token ID393
Feature activation-4.51e+00

Pos logit contributions
chrom+4.325
proof+4.109
roit+4.104
bos+3.976
rans+3.954
Neg logit contributions
olicy-3.463
emi-3.425
assi-3.303
 holiest-3.299
cean-3.218
Token ab
Token ID450
Feature activation-7.03e+00

Pos logit contributions
chrom+3.090
proof+2.936
roit+2.874
bos+2.817
rans+2.770
Neg logit contributions
olicy-3.483
emi-3.357
 Lanka-3.315
 holiest-3.279
 srfAttach-3.259
Tokenet
Token ID316
Feature activation-9.72e+00

Pos logit contributions
chrom+4.438
roit+4.263
proof+4.102
bos+4.041
eor+4.015
Neg logit contributions
 holiest-5.841
 goodbye-5.746
 forgiveness-5.615
 athlet-5.607
 Lanka-5.571
Tokented
Token ID1513
Feature activation-8.28e+00

Pos logit contributions
roit+1.734
chrom+1.697
proof+1.671
rans+1.352
eor+1.273
Neg logit contributions
 holiest-12.112
 goodbye-11.936
emi-11.833
 athlet-11.676
 Lanka-11.631
Token by
Token ID416
Feature activation-4.16e+00

Pos logit contributions
chrom+5.648
roit+5.387
proof+5.366
rans+5.140
bos+5.088
Neg logit contributions
olicy-5.785
 holiest-5.777
emi-5.762
assi-5.605
cean-5.594
Token states
Token ID2585
Feature activation-4.96e+00

Pos logit contributions
chrom+3.405
proof+3.171
roit+3.164
bos+3.123
rans+3.097
Neg logit contributions
cean-2.556
olicy-2.550
emi-2.529
 holiest-2.528
assi-2.464
Token who
Token ID508
Feature activation-6.65e+00

Pos logit contributions
chrom+3.736
rans+3.581
roit+3.561
bos+3.511
proof+3.494
Neg logit contributions
 holiest-3.300
olicy-3.281
 Lanka-3.256
emi-3.233
cean-3.187
Token are
Token ID389
Feature activation-3.84e+00

Pos logit contributions
chrom+4.230
roit+3.957
proof+3.846
bos+3.794
rans+3.785
Neg logit contributions
 srfAttach-5.149
 holiest-5.139
olicy-5.067
emi-5.052
QUI-5.017
Tokenuk
Token ID2724
Feature activation-2.53e+00

Pos logit contributions
chrom+3.571
roit+3.391
proof+3.379
eor+3.356
bos+3.303
Neg logit contributions
 holiest-3.431
 goodbye-3.374
olicy-3.362
 athlet-3.329
 bargaining-3.297
Token Moor
Token ID31451
Feature activation-5.20e+00

Pos logit contributions
chrom+1.994
roit+1.889
proof+1.883
rans+1.870
bos+1.859
Neg logit contributions
 holiest-1.728
 athlet-1.706
 goodbye-1.657
emi-1.644
olicy-1.627
Tokenclaw
Token ID43143
Feature activation-3.35e+00

Pos logit contributions
chrom+3.693
proof+3.579
rans+3.489
bos+3.484
roit+3.447
Neg logit contributions
 holiest-3.870
 athlet-3.730
 goodbye-3.727
 bargaining-3.720
olicy-3.640
Token,
Token ID11
Feature activation-2.75e+00

Pos logit contributions
chrom+2.605
proof+2.462
roit+2.431
bos+2.403
eor+2.398
Neg logit contributions
 holiest-2.305
 goodbye-2.218
 athlet-2.202
olicy-2.184
emi-2.182
Token and
Token ID290
Feature activation-2.02e+00

Pos logit contributions
chrom+2.314
proof+2.175
roit+2.163
eor+2.149
bos+2.119
Neg logit contributions
 holiest-1.657
 athlet-1.640
olicy-1.632
emi-1.618
cean-1.618
Token Col
Token ID1623
Feature activation-9.70e+00

Pos logit contributions
chrom+1.684
proof+1.609
roit+1.584
eor+1.554
bos+1.549
Neg logit contributions
olicy-1.214
emi-1.212
cean-1.208
 holiest-1.203
 athlet-1.198
Tokenbie
Token ID12590
Feature activation-3.46e+00

Pos logit contributions
chrom+6.615
roit+6.310
bos+6.192
eor+6.086
inant+6.050
Neg logit contributions
 holiest-6.938
 goodbye-6.933
 athlet-6.820
olicy-6.759
 bargaining-6.720
Token Sterling
Token ID23647
Feature activation-3.51e+00

Pos logit contributions
chrom+2.986
proof+2.818
roit+2.783
eor+2.770
bos+2.767
Neg logit contributions
 holiest-2.061
 athlet-2.002
olicy-1.991
 goodbye-1.966
emi-1.958
Token
Token ID960
Feature activation-3.72e+00

Pos logit contributions
chrom+2.694
bos+2.524
proof+2.522
roit+2.507
rans+2.472
Neg logit contributions
 holiest-2.435
olicy-2.350
emi-2.288
 athlet-2.275
cean-2.275
Tokenall
Token ID439
Feature activation-5.79e+00

Pos logit contributions
chrom+2.248
proof+2.120
rans+1.949
eor+1.942
roit+1.940
Neg logit contributions
 holiest-3.228
 athlet-3.220
 Lanka-3.199
olicy-3.148
emi-3.106
Token of
Token ID286
Feature activation-3.02e+00

Pos logit contributions
chrom+4.545
proof+4.318
roit+4.299
rans+4.212
inant+4.207
Neg logit contributions
 holiest-3.590
cean-3.502
olicy-3.500
emi-3.445
 Lanka-3.396
Token of
Token ID286
Feature activation-4.25e+00

Pos logit contributions
chrom+3.859
proof+3.616
bos+3.598
roit+3.588
inant+3.552
Neg logit contributions
 holiest-2.470
emi-2.380
olicy-2.369
 athlet-2.366
 goodbye-2.354
Token critical
Token ID4688
Feature activation-4.85e+00

Pos logit contributions
chrom+3.402
proof+3.138
roit+3.112
bos+3.092
rans+3.028
Neg logit contributions
olicy-2.804
cean-2.796
emi-2.750
 holiest-2.736
assi-2.650
Token illness
Token ID8526
Feature activation-4.52e+00

Pos logit contributions
chrom+3.663
roit+3.412
proof+3.336
eor+3.238
rans+3.235
Neg logit contributions
 holiest-3.426
emi-3.362
olicy-3.357
cean-3.316
 athlet-3.260
Token-
Token ID12
Feature activation-8.86e+00

Pos logit contributions
chrom+3.254
proof+3.099
roit+3.011
bos+2.948
rans+2.911
Neg logit contributions
 holiest-3.252
olicy-3.233
cean-3.216
emi-3.129
 Lanka-3.112
Tokenrelated
Token ID5363
Feature activation-4.91e+00

Pos logit contributions
chrom+4.390
proof+4.110
linked+3.886
bos+3.742
rans+3.678
Neg logit contributions
 holiest-8.249
 Lanka-8.202
cean-8.055
 athlet-8.053
olicy-8.051
Token cort
Token ID12794
Feature activation-9.61e+00

Pos logit contributions
chrom+3.392
proof+3.114
roit+3.069
bos+2.987
izations+2.974
Neg logit contributions
cean-3.766
olicy-3.755
emi-3.712
 holiest-3.691
assi-3.625
Tokenic
Token ID291
Feature activation-4.44e+00

Pos logit contributions
chrom+2.555
roit+2.239
proof+2.105
rans+2.014
inant+1.872
Neg logit contributions
 holiest-11.299
 goodbye-11.211
 bargaining-11.208
 athlet-11.081
QUI-11.026
Tokenoster
Token ID6197
Feature activation-6.89e+00

Pos logit contributions
chrom+2.323
roit+2.148
obyl+2.038
proof+2.035
rans+2.028
Neg logit contributions
 holiest-4.208
 bargaining-4.141
 Lanka-4.137
 goodbye-4.092
 athlet-4.086
Tokenoid
Token ID1868
Feature activation-5.07e+00

Pos logit contributions
chrom+1.370
proof+1.284
roit+1.160
rans+1.139
bos+1.113
Neg logit contributions
 holiest-8.526
 athlet-8.514
 Lanka-8.485
emi-8.407
olicy-8.361
Token ins
Token ID1035
Feature activation-9.86e-02

Pos logit contributions
chrom+3.662
proof+3.516
roit+3.469
bos+3.316
rans+3.219
Neg logit contributions
olicy-3.774
emi-3.704
 holiest-3.658
cean-3.656
 athlet-3.560
Tokenuff
Token ID1648
Feature activation-5.02e+00

Pos logit contributions
chrom+0.067
proof+0.060
rans+0.059
roit+0.059
bos+0.059
Neg logit contributions
 holiest-0.082
 athlet-0.082
 bargaining-0.080
emi-0.080
 goodbye-0.080
Token find
Token ID1064
Feature activation-6.07e+00

Pos logit contributions
chrom+4.528
proof+4.167
roit+3.976
eor+3.927
rans+3.911
Neg logit contributions
 holiest-5.356
olicy-5.307
emi-5.277
cean-5.107
QUI-5.102
Token a
Token ID257
Feature activation-4.95e+00

Pos logit contributions
chrom+4.886
proof+4.613
bos+4.408
roit+4.365
rans+4.339
Neg logit contributions
olicy-3.864
emi-3.783
cean-3.737
assi-3.606
 holiest-3.594
Token specific
Token ID2176
Feature activation-9.23e+00

Pos logit contributions
chrom+4.138
proof+3.735
roit+3.624
bos+3.580
inant+3.573
Neg logit contributions
olicy-3.141
emi-3.096
cean-3.050
 athlet-3.005
 holiest-2.981
Token pet
Token ID4273
Feature activation-7.53e+00

Pos logit contributions
chrom+6.719
proof+6.210
inant+6.001
roit+5.852
bos+5.832
Neg logit contributions
olicy-6.130
emi-6.101
cean-6.026
QUI-5.822
assi-5.811
Token-
Token ID12
Feature activation-8.82e+00

Pos logit contributions
chrom+5.695
proof+5.382
roit+5.345
rans+5.256
eor+5.169
Neg logit contributions
 holiest-4.994
cean-4.920
emi-4.902
 athlet-4.845
CONCLUS-4.832
Tokencolour
Token ID49903
Feature activation-9.51e+00

Pos logit contributions
chrom+5.039
proof+4.937
bos+4.345
rans+4.337
roit+4.273
Neg logit contributions
 Lanka-7.687
 holiest-7.567
 athlet-7.474
emi-7.327
 forgiveness-7.271
Token combination
Token ID6087
Feature activation-5.65e+00

Pos logit contributions
chrom+7.097
proof+6.683
roit+6.236
inant+6.160
isation+6.107
Neg logit contributions
olicy-6.308
cean-6.167
 holiest-6.128
CONCLUS-6.083
emi-6.071
Token,
Token ID11
Feature activation-3.73e+00

Pos logit contributions
chrom+4.253
proof+3.971
roit+3.837
rans+3.735
inant+3.725
Neg logit contributions
olicy-3.886
cean-3.849
emi-3.834
 holiest-3.814
CONCLUS-3.758
Token by
Token ID416
Feature activation-3.04e+00

Pos logit contributions
chrom+2.847
proof+2.651
roit+2.602
bos+2.531
rans+2.531
Neg logit contributions
 holiest-2.556
olicy-2.547
emi-2.538
cean-2.497
 Lanka-2.471
Token setting
Token ID4634
Feature activation-3.86e+00

Pos logit contributions
chrom+2.232
proof+2.042
roit+2.016
bos+1.970
rans+1.956
Neg logit contributions
olicy-2.221
 holiest-2.160
emi-2.158
cean-2.137
 goodbye-2.109
Token both
Token ID1111
Feature activation-4.72e+00

Pos logit contributions
chrom+3.174
proof+2.984
roit+2.883
bos+2.833
rans+2.809
Neg logit contributions
 holiest-2.506
emi-2.435
cean-2.400
olicy-2.392
 athlet-2.381
Token be
Token ID307
Feature activation-6.41e-01

Pos logit contributions
chrom+0.935
proof+0.899
rans+0.882
roit+0.862
eor+0.840
Neg logit contributions
 holiest-0.856
emi-0.855
olicy-0.851
cean-0.836
assi-0.830
Token for
Token ID329
Feature activation-5.06e+00

Pos logit contributions
chrom+0.565
roit+0.531
proof+0.528
eor+0.521
obyl+0.519
Neg logit contributions
 holiest-0.383
emi-0.375
 athlet-0.371
 goodbye-0.370
 bargaining-0.363
Token "
Token ID366
Feature activation-5.46e+00

Pos logit contributions
chrom+4.156
proof+3.956
roit+3.922
rans+3.818
bos+3.812
Neg logit contributions
olicy-3.043
emi-3.014
 holiest-2.990
cean-2.968
assi-2.879
Tokenactive
Token ID5275
Feature activation-6.32e+00

Pos logit contributions
chrom+3.780
proof+3.627
bos+3.350
rans+3.343
roit+3.341
Neg logit contributions
 holiest-4.246
 athlet-4.150
 Lanka-4.126
olicy-4.108
 goodbye-4.093
Token"
Token ID1
Feature activation-6.59e+00

Pos logit contributions
chrom+5.181
proof+4.876
roit+4.833
bos+4.730
eor+4.639
Neg logit contributions
 holiest-3.948
emi-3.813
cean-3.777
olicy-3.755
ofi-3.687
Token registered
Token ID6823
Feature activation-9.45e+00

Pos logit contributions
chrom+4.976
proof+4.644
roit+4.561
bos+4.476
rans+4.379
Neg logit contributions
emi-4.407
olicy-4.376
 holiest-4.344
cean-4.330
assi-4.250
Token voters
Token ID4446
Feature activation-6.64e+00

Pos logit contributions
chrom+6.315
roit+5.758
proof+5.680
bos+5.604
rans+5.492
Neg logit contributions
 holiest-6.932
cean-6.775
emi-6.685
assi-6.672
olicy-6.671
Token,
Token ID11
Feature activation-4.82e+00

Pos logit contributions
chrom+4.517
proof+4.346
rans+4.231
roit+4.182
bos+4.149
Neg logit contributions
 holiest-4.842
olicy-4.788
emi-4.660
cean-4.647
 Lanka-4.577
Token i
Token ID1312
Feature activation-3.92e+00

Pos logit contributions
chrom+3.596
proof+3.449
rans+3.384
roit+3.365
bos+3.308
Neg logit contributions
cean-3.240
emi-3.208
 holiest-3.182
assi-3.114
olicy-3.108
Token.
Token ID13
Feature activation-4.92e+00

Pos logit contributions
chrom+2.941
proof+2.681
roit+2.635
rans+2.628
ards+2.606
Neg logit contributions
 holiest-2.806
 goodbye-2.752
 athlet-2.701
 Lanka-2.649
 bargaining-2.646
Tokene
Token ID68
Feature activation-3.63e+00

Pos logit contributions
chrom+1.104
proof+0.917
eor+0.837
roit+0.685
rans+0.652
Neg logit contributions
 holiest-6.332
 athlet-6.272
olicy-6.161
 Lanka-6.087
 bargaining-6.035
Token natural
Token ID3288
Feature activation-6.26e+00

Pos logit contributions
chrom+4.098
roit+3.805
proof+3.802
bos+3.727
rans+3.716
Neg logit contributions
emi-3.241
olicy-3.163
cean-3.117
assi-3.109
QUI-3.027
Token habitat
Token ID20018
Feature activation-7.63e+00

Pos logit contributions
chrom+4.704
proof+4.308
roit+4.236
isation+4.195
izations+4.153
Neg logit contributions
emi-4.425
assi-4.349
olicy-4.295
 holiest-4.180
 Lanka-4.101
Token.
Token ID13
Feature activation-4.56e+00

Pos logit contributions
chrom+5.597
proof+5.395
rans+5.275
roit+5.156
bos+5.119
Neg logit contributions
olicy-4.984
emi-4.821
 athlet-4.821
cean-4.817
assi-4.754
Token The
Token ID383
Feature activation-3.48e+00

Pos logit contributions
chrom+3.671
proof+3.475
roit+3.366
eor+3.351
bos+3.316
Neg logit contributions
 athlet-3.074
 holiest-3.000
olicy-2.905
 bargaining-2.851
emi-2.809
Token fire
Token ID2046
Feature activation-6.62e+00

Pos logit contributions
chrom+2.844
proof+2.674
roit+2.624
bos+2.582
rans+2.577
Neg logit contributions
emi-2.186
olicy-2.186
 athlet-2.148
 holiest-2.135
cean-2.130
Token ant
Token ID1885
Feature activation-9.45e+00

Pos logit contributions
proof+5.001
chrom+4.891
roit+4.504
eor+4.435
rans+4.429
Neg logit contributions
 athlet-4.523
 holiest-4.496
olicy-4.459
cean-4.417
 Lanka-4.415
Token epidemic
Token ID18195
Feature activation-4.34e+00

Pos logit contributions
proof+6.578
chrom+6.567
roit+6.187
rans+6.021
eor+6.001
Neg logit contributions
 holiest-6.524
emi-6.341
QUI-6.306
 srfAttach-6.273
 athlet-6.268
Token arose
Token ID21172
Feature activation-3.50e+00

Pos logit contributions
chrom+3.062
proof+2.944
bos+2.796
roit+2.789
inant+2.741
Neg logit contributions
olicy-3.233
 holiest-3.157
cean-3.142
emi-3.128
QUI-3.097
Token when
Token ID618
Feature activation-6.10e+00

Pos logit contributions
chrom+2.620
proof+2.439
roit+2.413
bos+2.406
rans+2.364
Neg logit contributions
emi-2.451
 holiest-2.435
olicy-2.420
assi-2.364
cean-2.346
Token the
Token ID262
Feature activation-3.02e+00

Pos logit contributions
chrom+4.591
proof+4.429
bos+4.280
eor+4.252
roit+4.244
Neg logit contributions
emi-3.987
olicy-3.984
 holiest-3.946
QUI-3.935
cean-3.840
Token killer
Token ID11868
Feature activation-3.20e+00

Pos logit contributions
chrom+2.437
proof+2.277
roit+2.241
eor+2.223
rans+2.217
Neg logit contributions
emi-1.976
olicy-1.972
cean-1.914
 holiest-1.904
 athlet-1.884
Token lives
Token ID3160
Feature activation-3.94e+00

Pos logit contributions
chrom+4.062
proof+3.795
roit+3.728
rans+3.720
bos+3.699
Neg logit contributions
emi-3.125
 holiest-3.122
cean-3.102
olicy-3.102
 athlet-3.007
Token,
Token ID11
Feature activation-5.26e+00

Pos logit contributions
chrom+2.931
proof+2.827
rans+2.776
bos+2.759
obyl+2.723
Neg logit contributions
olicy-2.739
 holiest-2.622
emi-2.621
cean-2.599
assi-2.592
Token including
Token ID1390
Feature activation-7.48e+00

Pos logit contributions
chrom+3.935
proof+3.764
rans+3.696
bos+3.671
roit+3.641
Neg logit contributions
cean-3.445
olicy-3.389
emi-3.348
 holiest-3.347
assi-3.322
Token hundreds
Token ID5179
Feature activation-5.38e+00

Pos logit contributions
chrom+5.324
proof+5.283
bos+5.001
roit+4.921
rans+4.905
Neg logit contributions
olicy-4.926
cean-4.840
emi-4.820
QUI-4.701
assi-4.671
Token of
Token ID286
Feature activation-4.67e+00

Pos logit contributions
chrom+4.303
proof+4.101
rans+4.044
bos+4.044
roit+3.992
Neg logit contributions
olicy-3.264
cean-3.241
 holiest-3.229
emi-3.209
 athlet-3.162
Token civilians
Token ID10380
Feature activation-9.43e+00

Pos logit contributions
chrom+3.498
proof+3.328
bos+3.294
roit+3.241
rans+3.218
Neg logit contributions
emi-3.150
olicy-3.109
cean-3.086
 holiest-3.043
 srfAttach-3.040
Token.
Token ID13
Feature activation-3.33e+00

Pos logit contributions
chrom+5.907
rans+5.827
proof+5.791
eor+5.726
roit+5.620
Neg logit contributions
olicy-6.912
 holiest-6.595
cean-6.567
 athlet-6.513
emi-6.464
Token
Token ID198
Feature activation-2.76e+00

Pos logit contributions
chrom+2.683
proof+2.562
roit+2.474
eor+2.470
bos+2.467
Neg logit contributions
 athlet-2.261
 holiest-2.177
olicy-2.111
 bargaining-2.082
cean-2.080
Token
Token ID198
Feature activation-2.46e+00

Pos logit contributions
chrom+2.020
proof+1.917
roit+1.768
bos+1.766
rans+1.757
Neg logit contributions
 athlet-2.285
 holiest-2.194
olicy-2.163
 referen-2.141
 nomine-2.134
TokenThe
Token ID464
Feature activation-2.09e+00

Pos logit contributions
chrom+1.902
proof+1.795
bos+1.720
roit+1.702
eor+1.677
Neg logit contributions
 athlet-1.857
 holiest-1.782
olicy-1.753
 goodbye-1.751
 bargaining-1.739
Token hypocrisy
Token ID29442
Feature activation-8.54e-01

Pos logit contributions
chrom+1.654
proof+1.565
roit+1.533
bos+1.533
rans+1.519
Neg logit contributions
olicy-1.361
emi-1.354
 athlet-1.334
cean-1.331
 goodbye-1.305
Token
Token ID198
Feature activation-4.74e+00

Pos logit contributions
chrom+3.527
proof+3.262
roit+2.994
bos+2.954
rans+2.935
Neg logit contributions
 athlet-4.203
 holiest-4.041
olicy-3.913
 nomine-3.834
 bargaining-3.821
TokenDevelopment
Token ID41206
Feature activation-5.68e+00

Pos logit contributions
chrom+3.752
proof+3.497
roit+3.293
bos+3.291
rans+3.255
Neg logit contributions
 holiest-3.337
 athlet-3.310
 Lanka-3.234
olicy-3.209
 goodbye-3.203
Token of
Token ID286
Feature activation-4.59e+00

Pos logit contributions
chrom+4.490
proof+4.161
eor+3.968
rans+3.965
roit+3.965
Neg logit contributions
 holiest-3.664
emi-3.600
 athlet-3.583
olicy-3.562
cean-3.509
Token the
Token ID262
Feature activation-4.89e+00

Pos logit contributions
chrom+3.734
proof+3.455
roit+3.374
bos+3.342
inant+3.303
Neg logit contributions
emi-2.882
cean-2.878
 holiest-2.873
olicy-2.848
 goodbye-2.812
Token F
Token ID376
Feature activation-2.69e+00

Pos logit contributions
chrom+3.866
proof+3.576
roit+3.528
bos+3.469
rans+3.460
Neg logit contributions
cean-3.117
olicy-3.091
emi-3.089
 holiest-3.008
 athlet-3.001
Token-
Token ID12
Feature activation-9.36e+00

Pos logit contributions
chrom+1.880
roit+1.759
proof+1.746
bos+1.714
obyl+1.712
Neg logit contributions
 holiest-2.071
 athlet-2.046
 goodbye-2.016
 bargaining-1.981
cean-1.980
Token35
Token ID2327
Feature activation-2.72e+00

Pos logit contributions
chrom+6.391
proof+5.912
bos+5.669
roit+5.361
linked+5.299
Neg logit contributions
 holiest-6.901
 athlet-6.812
cean-6.717
 goodbye-6.576
emi-6.568
Token fighter
Token ID10543
Feature activation-2.80e+00

Pos logit contributions
chrom+1.992
proof+1.922
roit+1.888
bos+1.858
eor+1.838
Neg logit contributions
 holiest-1.942
emi-1.893
cean-1.878
 athlet-1.876
olicy-1.864
Token jet
Token ID12644
Feature activation-4.26e+00

Pos logit contributions
chrom+1.925
proof+1.869
roit+1.814
bos+1.765
rans+1.753
Neg logit contributions
 holiest-2.144
olicy-2.095
emi-2.089
 athlet-2.089
 Lanka-2.070
Token will
Token ID481
Feature activation-5.50e+00

Pos logit contributions
chrom+3.092
proof+2.955
roit+2.866
bos+2.800
inant+2.771
Neg logit contributions
 holiest-3.074
olicy-3.009
cean-2.986
emi-2.977
QUI-2.959
Token cost
Token ID1575
Feature activation-1.15e+00

Pos logit contributions
chrom+3.153
roit+2.935
proof+2.875
bos+2.856
eor+2.839
Neg logit contributions
olicy-4.667
 holiest-4.662
QUI-4.630
assi-4.566
emi-4.489
Token well
Token ID880
Feature activation+2.72e+00

Pos logit contributions
 holiest+1.908
emi+1.883
cean+1.866
 athlet+1.834
olicy+1.832
Neg logit contributions
chrom-2.348
proof-2.235
roit-2.196
rans-2.168
bos-2.158
Token founded
Token ID9393
Feature activation+3.41e+00

Pos logit contributions
 holiest+1.912
emi+1.871
 athlet+1.848
cean+1.845
 goodbye+1.833
Neg logit contributions
chrom-2.138
proof-2.038
roit-2.014
bos-1.987
rans-1.942
Token in
Token ID287
Feature activation+4.27e+00

Pos logit contributions
 holiest+2.230
emi+2.166
 athlet+2.131
 goodbye+2.111
cean+2.098
Neg logit contributions
chrom-2.881
proof-2.718
roit-2.696
bos-2.649
rans-2.639
Token fairness
Token ID22692
Feature activation+4.22e+00

Pos logit contributions
 holiest+2.602
emi+2.577
 athlet+2.534
cean+2.514
 goodbye+2.480
Neg logit contributions
chrom-3.775
proof-3.586
roit-3.540
rans-3.488
eor-3.466
Token and
Token ID290
Feature activation+2.90e+00

Pos logit contributions
 holiest+2.704
 athlet+2.619
emi+2.617
 goodbye+2.585
cean+2.564
Neg logit contributions
chrom-3.605
proof-3.413
roit-3.361
bos-3.313
eor-3.301
Token economics
Token ID12446
Feature activation+7.23e+00

Pos logit contributions
emi+1.846
 holiest+1.823
cean+1.811
 athlet+1.757
olicy+1.749
Neg logit contributions
chrom-2.506
proof-2.400
roit-2.377
bos-2.279
rans-2.272
Token.
Token ID13
Feature activation+2.18e+00

Pos logit contributions
 holiest+4.536
emi+4.430
 athlet+4.388
 goodbye+4.351
cean+4.295
Neg logit contributions
chrom-6.246
proof-5.873
roit-5.862
eor-5.740
rans-5.711
Token But
Token ID887
Feature activation+3.12e+00

Pos logit contributions
 holiest+1.383
 athlet+1.378
emi+1.331
cean+1.299
 goodbye+1.288
Neg logit contributions
chrom-1.904
proof-1.824
roit-1.794
rans-1.759
eor-1.743
Token like
Token ID588
Feature activation+3.52e+00

Pos logit contributions
 holiest+2.126
emi+2.104
 athlet+2.047
cean+2.045
olicy+1.996
Neg logit contributions
chrom-2.543
proof-2.412
roit-2.389
rans-2.357
bos-2.339
Token most
Token ID749
Feature activation+2.75e+00

Pos logit contributions
emi+2.217
 holiest+2.183
 athlet+2.145
cean+2.141
olicy+2.069
Neg logit contributions
chrom-3.059
proof-2.905
roit-2.887
bos-2.821
rans-2.815
Token reforms
Token ID12506
Feature activation+5.58e+00

Pos logit contributions
emi+1.800
 holiest+1.788
cean+1.758
 athlet+1.735
olicy+1.711
Neg logit contributions
chrom-2.291
proof-2.180
roit-2.178
rans-2.126
bos-2.111
Token He
Token ID679
Feature activation+2.43e+00

Pos logit contributions
 holiest+1.213
 athlet+1.157
cean+1.137
 goodbye+1.134
 bargaining+1.133
Neg logit contributions
chrom-1.552
proof-1.462
roit-1.460
bos-1.426
rans-1.412
Token has
Token ID468
Feature activation+3.35e+00

Pos logit contributions
olicy+1.771
 holiest+1.770
QUI+1.710
emi+1.704
 Lanka+1.698
Neg logit contributions
chrom-1.875
proof-1.745
roit-1.732
rans-1.708
bos-1.701
Token one
Token ID530
Feature activation+2.39e+00

Pos logit contributions
 holiest+2.294
olicy+2.270
emi+2.251
cean+2.200
 goodbye+2.183
Neg logit contributions
chrom-2.707
proof-2.555
roit-2.522
rans-2.481
bos-2.463
Token year
Token ID614
Feature activation+3.97e+00

Pos logit contributions
 holiest+1.782
olicy+1.752
emi+1.733
cean+1.732
 goodbye+1.712
Neg logit contributions
chrom-1.754
proof-1.653
roit-1.652
rans-1.599
inant-1.589
Token and
Token ID290
Feature activation+2.68e+00

Pos logit contributions
 holiest+2.225
emi+2.152
 Lanka+2.100
cean+2.092
olicy+2.086
Neg logit contributions
chrom-3.695
roit-3.487
proof-3.480
bos-3.414
rans-3.398
Token $
Token ID720
Feature activation+5.57e+00

Pos logit contributions
 holiest+1.536
olicy+1.522
cean+1.515
emi+1.492
 Lanka+1.486
Neg logit contributions
chrom-2.444
proof-2.360
roit-2.330
bos-2.255
rans-2.254
Token1
Token ID16
Feature activation+2.79e+00

Pos logit contributions
 holiest+4.032
 athlet+3.956
emi+3.908
 goodbye+3.852
 bargaining+3.834
Neg logit contributions
chrom-4.355
proof-4.052
roit-4.018
eor-3.914
rans-3.914
Token.
Token ID13
Feature activation+1.66e+00

Pos logit contributions
 holiest+1.716
emi+1.671
 athlet+1.657
 goodbye+1.655
cean+1.635
Neg logit contributions
chrom-2.423
roit-2.273
proof-2.267
bos-2.221
rans-2.218
Token35
Token ID2327
Feature activation+3.72e+00

Pos logit contributions
 athlet+1.292
 holiest+1.286
 bargaining+1.227
olicy+1.220
emi+1.211
Neg logit contributions
chrom-1.239
proof-1.160
roit-1.121
bos-1.115
rans-1.115
Token million
Token ID1510
Feature activation+4.47e+00

Pos logit contributions
 holiest+2.268
emi+2.213
 athlet+2.193
 goodbye+2.183
cean+2.156
Neg logit contributions
chrom-3.254
proof-3.051
roit-3.047
bos-2.992
rans-2.980
Token left
Token ID1364
Feature activation+1.79e+00

Pos logit contributions
 holiest+2.247
emi+2.162
cean+2.085
 athlet+2.080
 goodbye+2.064
Neg logit contributions
chrom-4.419
roit-4.217
proof-4.201
bos-4.135
rans-4.128
Tokenget
Token ID1136
Feature activation-1.10e+00

Pos logit contributions
chrom+0.441
roit+0.397
proof+0.395
eor+0.390
bos+0.387
Neg logit contributions
 holiest-0.795
 athlet-0.771
 goodbye-0.763
emi-0.758
 bargaining-0.756
Tokentes
Token ID4879
Feature activation-1.13e+00

Pos logit contributions
chrom+0.452
roit+0.419
proof+0.398
rans+0.390
eor+0.388
Neg logit contributions
 holiest-1.194
 goodbye-1.164
 Lanka-1.151
 athlet-1.151
emi-1.146
Token,
Token ID11
Feature activation+2.02e-01

Pos logit contributions
chrom+0.900
roit+0.842
proof+0.838
bos+0.837
eor+0.824
Neg logit contributions
 holiest-0.773
olicy-0.756
 goodbye-0.739
emi-0.739
 athlet-0.737
Token basil
Token ID37792
Feature activation-1.52e+00

Pos logit contributions
 holiest+0.129
emi+0.128
olicy+0.126
 athlet+0.125
cean+0.124
Neg logit contributions
chrom-0.172
proof-0.160
roit-0.159
bos-0.158
rans-0.154
Token and
Token ID290
Feature activation+1.35e+00

Pos logit contributions
chrom+1.204
proof+1.147
roit+1.137
bos+1.123
obyl+1.113
Neg logit contributions
 holiest-1.022
olicy-1.005
emi-1.003
 athlet-0.987
cean-0.974
Token spinach
Token ID39129
Feature activation+6.07e+00

Pos logit contributions
emi+0.871
 holiest+0.869
olicy+0.852
cean+0.846
 athlet+0.843
Neg logit contributions
chrom-1.151
proof-1.072
roit-1.060
bos-1.051
obyl-1.017
Token,
Token ID11
Feature activation+1.28e+00

Pos logit contributions
 holiest+3.731
emi+3.661
 athlet+3.609
 goodbye+3.582
cean+3.542
Neg logit contributions
chrom-5.293
roit-4.963
proof-4.959
eor-4.846
bos-4.843
Token and
Token ID290
Feature activation+2.29e+00

Pos logit contributions
 holiest+0.882
emi+0.875
olicy+0.868
cean+0.862
 athlet+0.849
Neg logit contributions
chrom-1.012
proof-0.945
roit-0.937
bos-0.936
rans-0.908
Token named
Token ID3706
Feature activation-4.81e-01

Pos logit contributions
 holiest+1.836
emi+1.836
olicy+1.801
cean+1.784
 Lanka+1.766
Neg logit contributions
chrom-1.576
proof-1.452
roit-1.441
bos-1.422
rans-1.389
Token in
Token ID287
Feature activation+3.12e+00

Pos logit contributions
chrom+0.361
proof+0.342
roit+0.334
bos+0.331
eor+0.329
Neg logit contributions
 holiest-0.342
emi-0.341
 athlet-0.335
olicy-0.332
cean-0.332
Token her
Token ID607
Feature activation+1.44e+00

Pos logit contributions
 holiest+1.571
 athlet+1.569
olicy+1.565
emi+1.559
cean+1.499
Neg logit contributions
chrom-3.054
proof-2.874
roit-2.860
rans-2.844
bos-2.812
Token it
Token ID340
Feature activation+2.73e+00

Pos logit contributions
 holiest+1.024
emi+1.014
 athlet+1.011
cean+1.010
olicy+0.989
Neg logit contributions
chrom-1.077
proof-1.027
roit-1.005
eor-0.984
rans-0.980
Token,
Token ID11
Feature activation+2.83e+00

Pos logit contributions
 holiest+1.785
emi+1.680
cean+1.675
 Lanka+1.663
 athlet+1.660
Neg logit contributions
chrom-2.354
proof-2.201
roit-2.191
eor-2.143
rans-2.141
Token
Token ID447
Feature activation+2.76e+00

Pos logit contributions
 holiest+1.911
emi+1.833
cean+1.822
 Lanka+1.806
 athlet+1.804
Neg logit contributions
chrom-2.346
proof-2.201
roit-2.199
rans-2.159
bos-2.118
Token
Token ID251
Feature activation+7.52e-01

Pos logit contributions
 holiest+2.113
 athlet+2.100
emi+2.095
cean+2.053
olicy+2.049
Neg logit contributions
chrom-2.002
roit-1.889
proof-1.867
rans-1.836
eor-1.800
Token said
Token ID531
Feature activation+2.10e+00

Pos logit contributions
 holiest+0.536
 athlet+0.522
emi+0.519
cean+0.513
 goodbye+0.512
Neg logit contributions
chrom-0.590
proof-0.550
roit-0.547
rans-0.532
eor-0.527
Token Ryan
Token ID6047
Feature activation+7.09e+00

Pos logit contributions
 holiest+1.295
emi+1.263
 athlet+1.259
cean+1.245
olicy+1.221
Neg logit contributions
chrom-1.835
roit-1.742
proof-1.734
rans-1.682
bos-1.678
Token,
Token ID11
Feature activation+2.76e+00

Pos logit contributions
 holiest+4.245
emi+4.145
 athlet+4.107
 goodbye+4.086
cean+4.018
Neg logit contributions
chrom-6.270
roit-5.906
proof-5.885
eor-5.774
bos-5.749
Token attribut
Token ID24548
Feature activation+3.17e+00

Pos logit contributions
 holiest+2.017
emi+1.983
 athlet+1.960
cean+1.939
olicy+1.916
Neg logit contributions
chrom-2.114
proof-1.976
roit-1.973
rans-1.928
bos-1.906
Tokening
Token ID278
Feature activation+2.69e+00

Pos logit contributions
 holiest+2.998
 athlet+2.953
 goodbye+2.926
emi+2.894
olicy+2.882
Neg logit contributions
chrom-1.719
roit-1.598
proof-1.593
eor-1.550
inant-1.520
Token the
Token ID262
Feature activation+2.81e+00

Pos logit contributions
 holiest+1.714
 athlet+1.686
emi+1.663
cean+1.658
 goodbye+1.634
Neg logit contributions
chrom-2.269
proof-2.155
roit-2.146
eor-2.088
rans-2.077
Token foul
Token ID15626
Feature activation+4.05e+00

Pos logit contributions
 holiest+2.067
emi+2.054
cean+2.041
 athlet+2.023
 goodbye+2.022
Neg logit contributions
chrom-2.108
roit-1.951
proof-1.940
rans-1.888
eor-1.887
Token,
Token ID11
Feature activation+3.00e+00

Pos logit contributions
 holiest+2.685
emi+2.572
 athlet+2.544
 Lanka+2.534
cean+2.526
Neg logit contributions
chrom-3.395
roit-3.189
proof-3.178
eor-3.113
bos-3.108
Token you
Token ID345
Feature activation+2.32e+00

Pos logit contributions
 holiest+2.136
emi+2.109
cean+2.072
 athlet+2.052
 Lanka+2.050
Neg logit contributions
chrom-2.320
proof-2.182
roit-2.149
eor-2.126
rans-2.116
Token will
Token ID481
Feature activation+1.64e+00

Pos logit contributions
 holiest+1.915
olicy+1.889
emi+1.884
 Lanka+1.857
cean+1.854
Neg logit contributions
chrom-1.560
proof-1.442
roit-1.400
rans-1.393
bos-1.383
Token get
Token ID651
Feature activation+2.98e+00

Pos logit contributions
olicy+1.464
 holiest+1.462
emi+1.458
cean+1.432
 Lanka+1.428
Neg logit contributions
chrom-0.963
proof-0.895
rans-0.872
roit-0.859
bos-0.844
Token a
Token ID257
Feature activation+3.36e+00

Pos logit contributions
 holiest+1.765
emi+1.762
cean+1.736
olicy+1.717
 athlet+1.704
Neg logit contributions
chrom-2.634
proof-2.538
roit-2.444
bos-2.434
rans-2.419
Token $
Token ID720
Feature activation+6.40e+00

Pos logit contributions
emi+2.190
 holiest+2.186
olicy+2.142
 athlet+2.131
cean+2.129
Neg logit contributions
chrom-2.814
proof-2.667
roit-2.597
rans-2.554
bos-2.548
Token1
Token ID16
Feature activation+4.07e+00

Pos logit contributions
 holiest+4.443
emi+4.349
 athlet+4.331
 goodbye+4.274
 bargaining+4.220
Neg logit contributions
chrom-5.108
proof-4.761
roit-4.735
eor-4.625
rans-4.621
Token,
Token ID11
Feature activation+3.15e+00

Pos logit contributions
 holiest+2.321
emi+2.286
 goodbye+2.246
 athlet+2.236
 bargaining+2.198
Neg logit contributions
chrom-3.669
roit-3.463
proof-3.433
eor-3.399
obyl-3.380
Token000
Token ID830
Feature activation+4.61e+00

Pos logit contributions
 holiest+2.067
emi+2.016
 athlet+2.014
 goodbye+1.975
 Lanka+1.960
Neg logit contributions
chrom-2.644
proof-2.466
roit-2.442
rans-2.410
eor-2.398
Token check
Token ID2198
Feature activation+2.03e+00

Pos logit contributions
 holiest+4.114
emi+4.061
 athlet+3.996
cean+3.978
 Lanka+3.956
Neg logit contributions
chrom-2.762
proof-2.554
roit-2.520
eor-2.435
rans-2.434
Token from
Token ID422
Feature activation+2.75e+00

Pos logit contributions
 holiest+1.317
emi+1.251
olicy+1.242
 Lanka+1.239
cean+1.232
Neg logit contributions
chrom-1.733
proof-1.659
roit-1.633
eor-1.589
rans-1.587
Token own
Token ID898
Feature activation+1.22e+00

Pos logit contributions
 holiest+0.656
emi+0.654
cean+0.641
olicy+0.638
 athlet+0.631
Neg logit contributions
chrom-0.811
roit-0.766
proof-0.762
eor-0.758
rans-0.744
Token John
Token ID1757
Feature activation-3.03e-01

Pos logit contributions
emi+0.829
 holiest+0.812
cean+0.794
olicy+0.792
 Lanka+0.778
Neg logit contributions
chrom-0.988
proof-0.940
roit-0.932
eor-0.916
rans-0.908
Token Mu
Token ID8252
Feature activation+1.78e+00

Pos logit contributions
chrom+0.294
roit+0.278
proof+0.278
eor+0.273
obyl+0.272
Neg logit contributions
 holiest-0.154
emi-0.151
 athlet-0.149
 goodbye-0.148
cean-0.146
Tokenir
Token ID343
Feature activation-5.70e-01

Pos logit contributions
 holiest+1.854
 athlet+1.824
emi+1.795
 goodbye+1.793
olicy+1.782
Neg logit contributions
chrom-0.801
proof-0.726
roit-0.716
eor-0.697
rans-0.696
Token hike
Token ID18390
Feature activation+3.64e+00

Pos logit contributions
chrom+0.444
proof+0.423
rans+0.407
eor+0.405
roit+0.404
Neg logit contributions
emi-0.395
 holiest-0.393
cean-0.377
 athlet-0.376
olicy-0.375
Token.
Token ID13
Feature activation+1.11e+00

Pos logit contributions
 holiest+2.352
emi+2.316
 athlet+2.254
 goodbye+2.241
 bargaining+2.233
Neg logit contributions
chrom-3.071
proof-2.913
roit-2.864
eor-2.819
rans-2.807
Token Bob
Token ID5811
Feature activation-1.83e+00

Pos logit contributions
 athlet+0.745
 holiest+0.737
emi+0.715
olicy+0.709
 goodbye+0.708
Neg logit contributions
chrom-0.931
proof-0.873
roit-0.854
eor-0.844
bos-0.841
Token In
Token ID554
Feature activation+3.46e+00

Pos logit contributions
chrom+1.589
proof+1.508
roit+1.491
bos+1.481
eor+1.460
Neg logit contributions
 holiest-1.122
 athlet-1.077
 goodbye-1.065
 bargaining-1.054
emi-1.053
Tokenman
Token ID805
Feature activation-6.45e-01

Pos logit contributions
 holiest+2.841
 goodbye+2.797
 athlet+2.761
emi+2.715
 bargaining+2.709
Neg logit contributions
chrom-2.399
proof-2.189
roit-2.161
eor-2.141
rans-2.120
Token had
Token ID550
Feature activation+1.56e-01

Pos logit contributions
chrom+0.506
proof+0.481
roit+0.466
bos+0.463
rans+0.462
Neg logit contributions
 holiest-0.450
olicy-0.435
 Lanka-0.433
 athlet-0.432
 goodbye-0.432
Token written
Token ID3194
Feature activation+3.57e-01

Pos logit contributions
 holiest+0.112
olicy+0.111
emi+0.109
 Lanka+0.107
cean+0.106
Neg logit contributions
chrom-0.120
proof-0.113
rans-0.109
roit-0.109
bos-0.108
Tokenama
Token ID1689
Feature activation+1.36e+00

Pos logit contributions
 holiest+2.168
 athlet+2.093
 goodbye+2.073
emi+2.072
 bargaining+2.059
Neg logit contributions
chrom-2.600
roit-2.455
proof-2.436
eor-2.411
rans-2.403
Tokeni
Token ID72
Feature activation+1.98e+00

Pos logit contributions
 holiest+0.969
 athlet+0.928
olicy+0.922
 bargaining+0.917
 goodbye+0.916
Neg logit contributions
chrom-1.060
roit-0.992
proof-0.984
eor-0.981
rans-0.974
Token,
Token ID11
Feature activation+1.00e+00

Pos logit contributions
 holiest+1.277
 athlet+1.224
olicy+1.219
emi+1.207
 bargaining+1.207
Neg logit contributions
chrom-1.657
roit-1.570
proof-1.544
rans-1.536
eor-1.534
Token k
Token ID479
Feature activation+1.49e+00

Pos logit contributions
 holiest+0.649
emi+0.624
olicy+0.623
 athlet+0.622
cean+0.618
Neg logit contributions
chrom-0.835
roit-0.780
rans-0.776
eor-0.773
proof-0.772
Tokenū
Token ID20317
Feature activation+1.35e+00

Pos logit contributions
 holiest+1.272
 bargaining+1.230
 athlet+1.229
 goodbye+1.221
 forgiveness+1.202
Neg logit contributions
chrom-0.960
roit-0.899
eor-0.883
proof-0.882
rans-0.877
Tokenpa
Token ID8957
Feature activation+3.38e+00

Pos logit contributions
 holiest+1.032
 athlet+0.992
 bargaining+0.983
emi+0.982
 goodbye+0.981
Neg logit contributions
chrom-0.990
roit-0.920
proof-0.912
eor-0.909
rans-0.905
Token
Token ID134
Feature activation+4.48e+00

Pos logit contributions
 holiest+2.350
 athlet+2.258
emi+2.242
 goodbye+2.235
 bargaining+2.227
Neg logit contributions
chrom-2.697
roit-2.519
proof-2.504
eor-2.477
rans-2.464
Token
Token ID119
Feature activation-2.42e+00

Pos logit contributions
 holiest+3.888
 athlet+3.826
emi+3.805
 goodbye+3.776
cean+3.745
Neg logit contributions
chrom-2.800
proof-2.534
roit-2.526
eor-2.483
obyl-2.461
Tokena
Token ID64
Feature activation+8.24e-01

Pos logit contributions
chrom+1.308
eor+1.221
rans+1.195
roit+1.194
proof+1.185
Neg logit contributions
 holiest-2.288
 bargaining-2.194
 athlet-2.192
 goodbye-2.175
CONCLUS-2.145
Token,
Token ID11
Feature activation+8.98e-01

Pos logit contributions
 holiest+0.548
olicy+0.522
 athlet+0.521
 bargaining+0.516
 goodbye+0.516
Neg logit contributions
chrom-0.676
roit-0.637
eor-0.626
proof-0.626
rans-0.624
Token al
Token ID435
Feature activation+4.34e-01

Pos logit contributions
 holiest+0.586
emi+0.560
olicy+0.560
 athlet+0.559
cean+0.557
Neg logit contributions
chrom-0.746
roit-0.699
rans-0.691
proof-0.690
eor-0.690
Token>
Token ID29
Feature activation+2.99e-01

Pos logit contributions
chrom+2.964
proof+2.777
roit+2.710
bos+2.678
rans+2.624
Neg logit contributions
 holiest-3.681
 goodbye-3.488
 bargaining-3.477
 Lanka-3.441
 athlet-3.426
Token lol
Token ID19462
Feature activation+1.33e+00

Pos logit contributions
emi+0.197
 holiest+0.195
olicy+0.194
cean+0.192
QUI+0.189
Neg logit contributions
chrom-0.243
proof-0.224
roit-0.224
bos-0.222
eor-0.219
Token m
Token ID285
Feature activation+9.53e-02

Pos logit contributions
 holiest+0.945
emi+0.931
 athlet+0.912
 bargaining+0.910
olicy+0.910
Neg logit contributions
chrom-1.030
proof-0.949
roit-0.948
bos-0.929
eor-0.927
Token2
Token ID17
Feature activation-2.04e+00

Pos logit contributions
 holiest+0.059
emi+0.058
 athlet+0.057
 goodbye+0.057
cean+0.056
Neg logit contributions
chrom-0.082
roit-0.077
proof-0.077
eor-0.075
obyl-0.075
Tokenk
Token ID74
Feature activation-2.82e-01

Pos logit contributions
chrom+1.344
proof+1.249
roit+1.194
eor+1.192
bos+1.189
Neg logit contributions
 holiest-1.738
 athlet-1.714
olicy-1.669
emi-1.655
 bargaining-1.622
Token [
Token ID685
Feature activation+4.14e-01

Pos logit contributions
chrom+0.239
proof+0.223
roit+0.223
eor+0.218
bos+0.218
Neg logit contributions
 holiest-0.177
emi-0.175
 athlet-0.172
cean-0.170
olicy-0.169
Token10
Token ID940
Feature activation+9.10e-01

Pos logit contributions
 athlet+0.339
 holiest+0.338
olicy+0.326
 bargaining+0.322
emi+0.321
Neg logit contributions
chrom-0.294
proof-0.273
bos-0.261
roit-0.259
eor-0.259
Token:
Token ID25
Feature activation+5.72e-01

Pos logit contributions
 athlet+0.702
 holiest+0.700
olicy+0.680
emi+0.673
 bargaining+0.662
Neg logit contributions
chrom-0.674
proof-0.637
roit-0.608
eor-0.603
bos-0.601
Token47
Token ID2857
Feature activation-4.33e+00

Pos logit contributions
 athlet+0.474
 holiest+0.470
olicy+0.457
emi+0.451
 bargaining+0.449
Neg logit contributions
chrom-0.400
proof-0.364
eor-0.355
rans-0.348
roit-0.346
Token]
Token ID60
Feature activation-1.43e-01

Pos logit contributions
chrom+2.908
proof+2.718
rans+2.539
eor+2.514
roit+2.490
Neg logit contributions
 athlet-3.793
 holiest-3.702
 referen-3.670
 nomine-3.649
olicy-3.619
Token <
Token ID1279
Feature activation+1.41e+00

Pos logit contributions
chrom+0.127
proof+0.119
roit+0.119
eor+0.117
obyl+0.116
Neg logit contributions
 holiest-0.086
emi-0.084
 goodbye-0.083
 athlet-0.083
 bargaining-0.081
Token of
Token ID286
Feature activation+2.81e+00

Pos logit contributions
 holiest+1.707
emi+1.678
cean+1.672
olicy+1.633
 Lanka+1.630
Neg logit contributions
chrom-2.683
proof-2.593
roit-2.560
rans-2.504
bos-2.494
Token the
Token ID262
Feature activation+2.46e+00

Pos logit contributions
 holiest+1.860
emi+1.850
cean+1.818
 athlet+1.773
 Lanka+1.761
Neg logit contributions
chrom-2.332
proof-2.277
roit-2.174
bos-2.140
rans-2.131
Token business
Token ID1597
Feature activation+2.43e+00

Pos logit contributions
 holiest+1.747
emi+1.728
cean+1.725
olicy+1.689
 Lanka+1.658
Neg logit contributions
chrom-1.908
proof-1.861
roit-1.781
bos-1.742
rans-1.731
Token;
Token ID26
Feature activation+2.79e+00

Pos logit contributions
 holiest+1.652
cean+1.600
emi+1.597
 Lanka+1.581
olicy+1.564
Neg logit contributions
chrom-1.979
proof-1.918
roit-1.847
rans-1.817
eor-1.803
Token having
Token ID1719
Feature activation+2.05e+00

Pos logit contributions
 holiest+1.916
emi+1.915
 Lanka+1.832
olicy+1.827
cean+1.824
Neg logit contributions
chrom-2.235
proof-2.149
roit-2.103
rans-2.065
bos-2.047
Token [
Token ID685
Feature activation+3.14e+00

Pos logit contributions
 holiest+1.365
emi+1.351
cean+1.322
olicy+1.316
 athlet+1.298
Neg logit contributions
chrom-1.707
proof-1.644
bos-1.569
roit-1.558
rans-1.542
Tokena
Token ID64
Feature activation+2.85e+00

Pos logit contributions
 holiest+2.858
 athlet+2.778
emi+2.760
 goodbye+2.751
 Lanka+2.735
Neg logit contributions
chrom-1.917
proof-1.778
roit-1.702
rans-1.689
bos-1.678
Token big
Token ID1263
Feature activation+3.21e+00

Pos logit contributions
 holiest+1.798
emi+1.789
cean+1.737
olicy+1.714
 athlet+1.714
Neg logit contributions
chrom-2.476
proof-2.336
roit-2.315
rans-2.264
bos-2.242
Token]
Token ID60
Feature activation+1.76e+00

Pos logit contributions
 holiest+2.268
emi+2.226
 athlet+2.156
cean+2.156
 Lanka+2.121
Neg logit contributions
chrom-2.563
proof-2.432
roit-2.393
rans-2.340
eor-2.327
Token enough
Token ID1576
Feature activation+4.50e+00

Pos logit contributions
emi+1.221
 holiest+1.214
cean+1.185
assi+1.166
olicy+1.163
Neg logit contributions
chrom-1.402
proof-1.350
roit-1.303
bos-1.287
rans-1.274
Token premium
Token ID8683
Feature activation+4.16e+00

Pos logit contributions
 holiest+3.180
emi+3.178
cean+3.066
olicy+3.020
 athlet+3.010
Neg logit contributions
chrom-3.586
proof-3.418
roit-3.316
bos-3.262
rans-3.250
Tokeny
Token ID88
Feature activation+1.69e+00

Pos logit contributions
 holiest+0.200
olicy+0.196
emi+0.195
 athlet+0.193
cean+0.193
Neg logit contributions
chrom-0.100
proof-0.095
roit-0.090
eor-0.087
rans-0.087
Token would
Token ID561
Feature activation-1.51e-01

Pos logit contributions
emi+1.232
 holiest+1.213
 athlet+1.169
cean+1.162
 Lanka+1.162
Neg logit contributions
chrom-1.299
proof-1.244
roit-1.204
bos-1.189
rans-1.161
Token pay
Token ID1414
Feature activation+3.23e+00

Pos logit contributions
chrom+0.096
proof+0.089
roit+0.088
rans+0.087
eor+0.085
Neg logit contributions
 holiest-0.130
olicy-0.126
emi-0.125
assi-0.123
cean-0.123
Token off
Token ID572
Feature activation+6.92e-01

Pos logit contributions
 holiest+1.525
cean+1.472
emi+1.445
olicy+1.442
 athlet+1.426
Neg logit contributions
chrom-3.280
proof-3.177
roit-3.147
bos-3.079
rans-3.064
Token,
Token ID11
Feature activation+8.07e-01

Pos logit contributions
 holiest+0.492
emi+0.472
olicy+0.469
 Lanka+0.468
assi+0.461
Neg logit contributions
chrom-0.548
proof-0.517
roit-0.504
rans-0.500
bos-0.499
Token and
Token ID290
Feature activation+1.78e+00

Pos logit contributions
 holiest+0.547
emi+0.533
olicy+0.524
 Lanka+0.517
cean+0.513
Neg logit contributions
chrom-0.653
proof-0.621
roit-0.603
rans-0.603
bos-0.593
Token that
Token ID326
Feature activation-1.53e+00

Pos logit contributions
 holiest+1.223
emi+1.218
olicy+1.184
 athlet+1.172
cean+1.166
Neg logit contributions
chrom-1.417
proof-1.346
roit-1.324
rans-1.303
eor-1.294
Token it
Token ID340
Feature activation+9.34e-01

Pos logit contributions
chrom+1.141
proof+1.082
roit+1.042
rans+1.013
bos+1.006
Neg logit contributions
emi-1.113
 holiest-1.109
olicy-1.076
 athlet-1.076
 Lanka-1.064
Token would
Token ID561
Feature activation-3.98e-01

Pos logit contributions
 holiest+0.666
olicy+0.641
emi+0.640
 Lanka+0.637
 athlet+0.631
Neg logit contributions
chrom-0.736
proof-0.687
roit-0.685
obyl-0.670
eor-0.668
Token come
Token ID1282
Feature activation+1.68e+00

Pos logit contributions
chrom+0.252
proof+0.235
roit+0.233
rans+0.227
eor+0.227
Neg logit contributions
 holiest-0.340
olicy-0.331
emi-0.329
QUI-0.323
assi-0.322
Token to
Token ID284
Feature activation-6.67e-01

Pos logit contributions
emi+1.171
 holiest+1.154
cean+1.145
olicy+1.112
assi+1.098
Neg logit contributions
chrom-1.352
proof-1.286
bos-1.265
roit-1.242
rans-1.221
Token L
Token ID406
Feature activation+3.22e+00

Pos logit contributions
 holiest+2.048
emi+2.030
cean+1.981
olicy+1.977
 Lanka+1.969
Neg logit contributions
chrom-2.280
roit-2.176
proof-2.148
rans-2.113
eor-2.073
Token sound
Token ID2128
Feature activation+2.02e+00

Pos logit contributions
 holiest+2.213
emi+2.156
 athlet+2.123
 goodbye+2.109
cean+2.108
Neg logit contributions
chrom-2.610
roit-2.453
proof-2.436
rans-2.379
bos-2.367
Token,
Token ID11
Feature activation+1.25e+00

Pos logit contributions
 holiest+1.324
emi+1.278
 Lanka+1.278
 athlet+1.255
cean+1.254
Neg logit contributions
chrom-1.705
proof-1.618
roit-1.587
rans-1.558
inant-1.548
Token something
Token ID1223
Feature activation+3.43e+00

Pos logit contributions
 holiest+0.836
emi+0.818
olicy+0.809
 Lanka+0.789
cean+0.787
Neg logit contributions
chrom-1.036
proof-0.975
roit-0.970
rans-0.952
eor-0.944
Token that
Token ID326
Feature activation-1.46e+00

Pos logit contributions
 holiest+2.271
emi+2.253
olicy+2.244
cean+2.212
 bargaining+2.164
Neg logit contributions
chrom-2.857
proof-2.693
roit-2.665
bos-2.596
rans-2.592
Token Phil
Token ID4543
Feature activation-1.06e+00

Pos logit contributions
chrom+1.128
proof+1.050
roit+1.050
rans+1.019
bos+1.015
Neg logit contributions
olicy-1.037
emi-1.023
 holiest-1.021
cean-0.984
 goodbye-0.963
Tokenadel
Token ID6959
Feature activation-1.22e+00

Pos logit contributions
chrom+0.870
proof+0.805
bos+0.796
roit+0.788
inant+0.768
Neg logit contributions
 holiest-0.743
emi-0.716
 goodbye-0.709
olicy-0.707
 bargaining-0.698
Tokenph
Token ID746
Feature activation+3.23e-01

Pos logit contributions
chrom+0.532
proof+0.483
bos+0.481
roit+0.480
rans+0.450
Neg logit contributions
 holiest-1.302
 goodbye-1.260
 Lanka-1.248
 bargaining-1.242
 athlet-1.238
Tokenians
Token ID1547
Feature activation+1.86e+00

Pos logit contributions
 holiest+0.322
 goodbye+0.312
cean+0.311
emi+0.311
 athlet+0.309
Neg logit contributions
chrom-0.155
roit-0.150
proof-0.146
eor-0.142
rans-0.142
Token do
Token ID466
Feature activation+1.09e+00

Pos logit contributions
 holiest+1.420
emi+1.386
olicy+1.377
cean+1.363
 Lanka+1.358
Neg logit contributions
chrom-1.377
proof-1.276
roit-1.271
rans-1.254
eor-1.236
Token more
Token ID517
Feature activation+2.61e+00

Pos logit contributions
 holiest+0.692
emi+0.671
 Lanka+0.670
cean+0.662
olicy+0.661
Neg logit contributions
chrom-0.935
proof-0.874
roit-0.872
eor-0.864
rans-0.863
Token using
Token ID1262
Feature activation-6.97e-01

Pos logit contributions
olicy+0.007
 holiest+0.007
emi+0.007
cean+0.007
 Lanka+0.007
Neg logit contributions
chrom-0.009
proof-0.008
roit-0.008
rans-0.008
bos-0.008
Token your
Token ID534
Feature activation+5.58e-01

Pos logit contributions
chrom+0.581
proof+0.554
roit+0.545
bos+0.532
eor+0.519
Neg logit contributions
olicy-0.447
 holiest-0.446
cean-0.438
 goodbye-0.433
emi-0.431
Token new
Token ID649
Feature activation-8.62e-01

Pos logit contributions
 holiest+0.355
cean+0.355
emi+0.354
olicy+0.352
 goodbye+0.351
Neg logit contributions
chrom-0.467
roit-0.444
proof-0.441
bos-0.432
inant-0.421
Token found
Token ID1043
Feature activation-1.20e+00

Pos logit contributions
chrom+0.678
proof+0.645
roit+0.636
bos+0.621
inant+0.606
Neg logit contributions
 holiest-0.600
cean-0.591
emi-0.589
olicy-0.584
 goodbye-0.579
Token Deep
Token ID10766
Feature activation+7.25e-01

Pos logit contributions
chrom+1.018
roit+0.982
proof+0.977
bos+0.945
inant+0.926
Neg logit contributions
 holiest-0.737
emi-0.735
cean-0.735
 goodbye-0.727
olicy-0.726
Token Conf
Token ID7326
Feature activation-1.47e+00

Pos logit contributions
 holiest+0.492
 goodbye+0.483
 athlet+0.469
olicy+0.469
emi+0.464
Neg logit contributions
chrom-0.595
proof-0.566
roit-0.563
bos-0.553
rans-0.542
Tokenidence
Token ID1704
Feature activation-2.44e-01

Pos logit contributions
chrom+0.791
proof+0.718
roit+0.711
inant+0.690
rans+0.687
Neg logit contributions
 goodbye-1.416
 holiest-1.414
 athlet-1.385
 Lanka-1.364
 bargaining-1.337
Token.
Token ID13
Feature activation+6.00e-01

Pos logit contributions
chrom+0.191
proof+0.184
roit+0.182
bos+0.176
inant+0.175
Neg logit contributions
 holiest-0.170
olicy-0.163
 goodbye-0.163
emi-0.163
 Lanka-0.158
Token
Token ID198
Feature activation+1.05e+00

Pos logit contributions
 holiest+0.425
 athlet+0.418
olicy+0.414
emi+0.403
 goodbye+0.399
Neg logit contributions
chrom-0.480
proof-0.455
roit-0.446
eor-0.432
bos-0.430
Token
Token ID198
Feature activation+9.73e-01

Pos logit contributions
 athlet+0.751
 holiest+0.749
olicy+0.723
emi+0.713
 bargaining+0.702
Neg logit contributions
chrom-0.848
proof-0.802
roit-0.771
eor-0.758
bos-0.758
TokenLet
Token ID5756
Feature activation+1.16e+00

Pos logit contributions
 holiest+0.714
 athlet+0.709
olicy+0.694
 Lanka+0.689
 goodbye+0.682
Neg logit contributions
chrom-0.779
proof-0.733
roit-0.703
bos-0.696
eor-0.687
Token
Token ID198
Feature activation-1.08e+00

Pos logit contributions
chrom+0.784
proof+0.745
roit+0.702
rans+0.695
eor+0.693
Neg logit contributions
 athlet-0.776
 holiest-0.764
olicy-0.745
emi-0.724
 bargaining-0.717
TokenAbout
Token ID8585
Feature activation-4.17e-01

Pos logit contributions
chrom+0.831
proof+0.790
roit+0.749
rans+0.746
bos+0.738
Neg logit contributions
 athlet-0.809
 holiest-0.792
olicy-0.783
 Lanka-0.779
 bargaining-0.778
Token former
Token ID1966
Feature activation-2.42e+00

Pos logit contributions
chrom+0.340
proof+0.328
rans+0.320
roit+0.319
eor+0.310
Neg logit contributions
emi-0.267
cean-0.265
 holiest-0.263
olicy-0.262
 athlet-0.261
Token meth
Token ID11248
Feature activation-1.01e+00

Pos logit contributions
chrom+1.962
proof+1.881
roit+1.823
rans+1.812
bos+1.790
Neg logit contributions
emi-1.536
olicy-1.518
cean-1.490
 athlet-1.485
ofi-1.469
Token dealer
Token ID16456
Feature activation-1.18e+00

Pos logit contributions
chrom+0.834
proof+0.799
roit+0.770
inant+0.766
rans+0.765
Neg logit contributions
emi-0.650
 holiest-0.650
 athlet-0.629
cean-0.625
olicy-0.622
Token and
Token ID290
Feature activation-1.17e+00

Pos logit contributions
chrom+0.976
proof+0.939
roit+0.919
rans+0.904
eor+0.894
Neg logit contributions
emi-0.741
 holiest-0.735
cean-0.726
olicy-0.721
 athlet-0.716
Token escort
Token ID27675
Feature activation-1.40e+00

Pos logit contributions
chrom+0.933
proof+0.903
roit+0.897
rans+0.873
bos+0.860
Neg logit contributions
emi-0.758
olicy-0.750
cean-0.742
QUI-0.718
 holiest-0.716
Token Mike
Token ID4995
Feature activation-8.28e-01

Pos logit contributions
chrom+1.170
proof+1.121
roit+1.097
eor+1.071
rans+1.062
Neg logit contributions
emi-0.871
 holiest-0.861
cean-0.835
externalActionCode-0.826
 athlet-0.826
Token Jones
Token ID5437
Feature activation-1.68e+00

Pos logit contributions
chrom+0.711
proof+0.672
roit+0.663
bos+0.658
eor+0.653
Neg logit contributions
 holiest-0.514
 athlet-0.506
emi-0.498
 goodbye-0.492
 bargaining-0.488
Token,
Token ID11
Feature activation-1.41e+00

Pos logit contributions
chrom+1.286
proof+1.216
roit+1.196
obyl+1.170
rans+1.167
Neg logit contributions
 holiest-1.179
 athlet-1.157
olicy-1.136
emi-1.132
cean-1.125
Token H
Token ID367
Feature activation+4.71e-01

Pos logit contributions
chrom+1.087
proof+1.024
roit+1.015
rans+1.002
bos+0.974
Neg logit contributions
emi-0.957
 athlet-0.956
cean-0.956
 holiest-0.954
olicy-0.948
Token also
Token ID635
Feature activation-4.30e+00

Pos logit contributions
chrom+2.219
proof+2.056
rans+2.030
roit+2.021
eor+1.992
Neg logit contributions
 holiest-2.539
olicy-2.424
QUI-2.422
emi-2.401
 athlet-2.391
Token need
Token ID761
Feature activation-1.15e+00

Pos logit contributions
chrom+2.669
proof+2.459
rans+2.429
roit+2.396
eor+2.358
Neg logit contributions
 holiest-3.668
QUI-3.556
olicy-3.552
 srfAttach-3.508
emi-3.478
Token to
Token ID284
Feature activation-4.06e+00

Pos logit contributions
chrom+0.986
proof+0.943
roit+0.909
bos+0.897
rans+0.894
Neg logit contributions
 holiest-0.717
emi-0.690
 athlet-0.688
 goodbye-0.683
cean-0.679
Token authorize
Token ID29145
Feature activation-3.01e+00

Pos logit contributions
chrom+2.521
proof+2.334
rans+2.258
roit+2.221
bos+2.186
Neg logit contributions
 holiest-3.435
olicy-3.322
emi-3.322
cean-3.272
QUI-3.266
Token each
Token ID1123
Feature activation-3.54e+00

Pos logit contributions
chrom+2.355
proof+2.287
roit+2.190
bos+2.163
rans+2.138
Neg logit contributions
 holiest-2.028
emi-1.982
olicy-1.913
 athlet-1.909
cean-1.896
Token other
Token ID584
Feature activation-4.94e+00

Pos logit contributions
chrom+2.772
proof+2.599
roit+2.533
eor+2.507
inant+2.500
Neg logit contributions
 holiest-2.393
emi-2.291
assi-2.266
 athlet-2.223
 goodbye-2.198
Token to
Token ID284
Feature activation-4.30e+00

Pos logit contributions
chrom+3.749
proof+3.547
roit+3.412
rans+3.385
obyl+3.380
Neg logit contributions
 holiest-3.481
emi-3.287
 Lanka-3.239
cean-3.203
 athlet-3.170
Token use
Token ID779
Feature activation-3.18e+00

Pos logit contributions
chrom+2.515
proof+2.349
rans+2.262
roit+2.228
inant+2.179
Neg logit contributions
 holiest-3.680
olicy-3.668
emi-3.633
QUI-3.557
cean-3.525
Token the
Token ID262
Feature activation-2.40e+00

Pos logit contributions
chrom+2.488
proof+2.377
roit+2.234
rans+2.232
bos+2.227
Neg logit contributions
 holiest-2.100
emi-2.098
olicy-2.069
cean-2.062
assi-2.055
Token credit
Token ID3884
Feature activation-8.62e-01

Pos logit contributions
chrom+1.773
proof+1.670
roit+1.600
bos+1.594
eor+1.577
Neg logit contributions
emi-1.714
 holiest-1.712
olicy-1.699
 athlet-1.675
cean-1.672
Token and
Token ID290
Feature activation-2.19e+00

Pos logit contributions
chrom+0.602
proof+0.560
roit+0.548
bos+0.544
eor+0.525
Neg logit contributions
 holiest-0.675
emi-0.667
 athlet-0.659
olicy-0.652
 goodbye-0.651
Token of
Token ID286
Feature activation+8.71e-01

Pos logit contributions
chrom+1.684
proof+1.567
rans+1.540
roit+1.519
eor+1.511
Neg logit contributions
 holiest-1.255
olicy-1.242
emi-1.215
 Lanka-1.207
cean-1.206
Token the
Token ID262
Feature activation+2.77e-01

Pos logit contributions
 holiest+0.567
emi+0.556
cean+0.555
 athlet+0.551
olicy+0.549
Neg logit contributions
chrom-0.729
proof-0.688
roit-0.664
rans-0.662
eor-0.659
Token new
Token ID649
Feature activation-4.09e-01

Pos logit contributions
cean+0.188
olicy+0.188
emi+0.186
 holiest+0.185
 athlet+0.179
Neg logit contributions
chrom-0.223
proof-0.209
roit-0.202
rans-0.201
bos-0.200
Token polymer
Token ID30528
Feature activation-2.36e+00

Pos logit contributions
chrom+0.337
proof+0.321
roit+0.305
rans+0.305
bos+0.304
Neg logit contributions
 holiest-0.269
cean-0.269
emi-0.267
olicy-0.264
 athlet-0.263
Token notes
Token ID4710
Feature activation+1.79e+00

Pos logit contributions
chrom+1.624
proof+1.500
bos+1.392
rans+1.383
roit+1.376
Neg logit contributions
 holiest-1.902
emi-1.879
olicy-1.859
cean-1.848
 athlet-1.831
Token are
Token ID389
Feature activation-3.10e-01

Pos logit contributions
 holiest+1.242
emi+1.218
olicy+1.202
cean+1.201
 athlet+1.190
Neg logit contributions
chrom-1.444
proof-1.343
roit-1.315
rans-1.298
eor-1.292
Token not
Token ID407
Feature activation-1.40e+00

Pos logit contributions
chrom+0.230
proof+0.213
rans+0.209
roit+0.207
eor+0.201
Neg logit contributions
olicy-0.229
emi-0.227
 holiest-0.226
cean-0.222
 athlet-0.220
Token raised
Token ID4376
Feature activation+5.24e-01

Pos logit contributions
chrom+1.109
proof+1.034
rans+1.022
roit+0.987
ards+0.971
Neg logit contributions
 holiest-0.961
olicy-0.958
cean-0.949
emi-0.944
 athlet-0.942
Token.
Token ID13
Feature activation+6.09e-02

Pos logit contributions
 holiest+0.363
emi+0.360
olicy+0.359
cean+0.354
 Lanka+0.349
Neg logit contributions
chrom-0.418
proof-0.396
roit-0.382
rans-0.380
bos-0.371
Token The
Token ID383
Feature activation+1.22e+00

Pos logit contributions
 athlet+0.039
 holiest+0.039
 bargaining+0.037
emi+0.037
olicy+0.037
Neg logit contributions
chrom-0.053
proof-0.050
roit-0.048
eor-0.048
rans-0.048
Token Bra
Token ID9718
Feature activation-6.23e-02

Pos logit contributions
emi+0.810
olicy+0.804
cean+0.797
 holiest+0.797
 athlet+0.775
Neg logit contributions
chrom-1.008
proof-0.936
roit-0.922
bos-0.909
rans-0.905
Tokenite
Token ID578
Feature activation-1.87e+00

Pos logit contributions
chrom+0.913
proof+0.793
roit+0.764
rans+0.743
eor+0.721
Neg logit contributions
 goodbye-3.125
 holiest-3.085
 athlet-3.071
 srfAttach-3.060
olicy-3.041
Token cleric
Token ID33824
Feature activation-3.47e+00

Pos logit contributions
chrom+1.516
proof+1.473
roit+1.449
rans+1.409
eor+1.403
Neg logit contributions
olicy-1.178
cean-1.143
 goodbye-1.133
emi-1.126
 srfAttach-1.123
Token in
Token ID287
Feature activation-1.03e+00

Pos logit contributions
chrom+2.293
proof+2.174
rans+2.143
roit+2.119
linked+2.095
Neg logit contributions
olicy-2.680
emi-2.617
cean-2.592
 goodbye-2.586
 srfAttach-2.583
Token Iraq
Token ID3908
Feature activation-1.62e+00

Pos logit contributions
chrom+0.901
proof+0.876
roit+0.855
eor+0.847
rans+0.838
Neg logit contributions
olicy-0.591
emi-0.581
 athlet-0.580
cean-0.575
 goodbye-0.567
Token,
Token ID11
Feature activation-2.82e+00

Pos logit contributions
chrom+1.143
proof+1.082
roit+1.067
rans+1.056
eor+1.029
Neg logit contributions
cean-1.191
emi-1.182
olicy-1.180
 athlet-1.180
 holiest-1.165
Token who
Token ID508
Feature activation-5.99e+00

Pos logit contributions
chrom+1.935
proof+1.830
rans+1.820
roit+1.779
bos+1.777
Neg logit contributions
olicy-2.124
cean-2.091
 athlet-2.072
emi-2.067
 holiest-2.033
Token has
Token ID468
Feature activation-2.63e+00

Pos logit contributions
chrom+3.906
proof+3.739
bos+3.678
rans+3.605
roit+3.584
Neg logit contributions
olicy-4.560
emi-4.396
 srfAttach-4.299
cean-4.291
 goodbye-4.253
Token been
Token ID587
Feature activation-5.73e+00

Pos logit contributions
chrom+1.830
proof+1.762
rans+1.713
roit+1.712
bos+1.679
Neg logit contributions
olicy-1.954
emi-1.925
cean-1.886
 holiest-1.864
 goodbye-1.862
Token outspoken
Token ID27687
Feature activation-3.06e+00

Pos logit contributions
chrom+3.793
proof+3.651
linked+3.563
roit+3.533
rans+3.485
Neg logit contributions
emi-4.141
olicy-4.104
 srfAttach-4.078
cean-4.050
 holiest-4.002
Token against
Token ID1028
Feature activation-3.61e+00

Pos logit contributions
chrom+2.303
proof+2.232
roit+2.194
rans+2.121
bos+2.114
Neg logit contributions
olicy-2.112
emi-2.102
 holiest-2.094
cean-2.039
QUI-2.000
Token the
Token ID262
Feature activation-3.05e+00

Pos logit contributions
chrom+2.811
proof+2.692
roit+2.631
bos+2.610
rans+2.554
Neg logit contributions
emi-2.385
olicy-2.317
 athlet-2.272
 srfAttach-2.258
cean-2.230
Token 2016
Token ID1584
Feature activation-6.43e+00

Pos logit contributions
chrom+4.508
proof+4.213
roit+4.207
rans+4.131
bos+4.121
Neg logit contributions
olicy-2.548
 holiest-2.484
 athlet-2.483
cean-2.481
emi-2.453
Token.
Token ID13
Feature activation-4.35e+00

Pos logit contributions
chrom+4.565
roit+4.409
proof+4.278
eor+4.254
rans+4.198
Neg logit contributions
 holiest-4.468
 athlet-4.333
olicy-4.296
 bargaining-4.279
emi-4.218
Token REUTERS
Token ID15862
Feature activation-2.87e+00

Pos logit contributions
chrom+2.960
proof+2.759
roit+2.689
bos+2.644
eor+2.634
Neg logit contributions
 athlet-3.392
 holiest-3.324
 bargaining-3.277
olicy-3.271
 goodbye-3.265
Token/
Token ID14
Feature activation-4.43e+00

Pos logit contributions
chrom+1.966
proof+1.846
roit+1.768
rans+1.755
eor+1.740
Neg logit contributions
 athlet-2.382
 holiest-2.304
olicy-2.280
emi-2.228
 bargaining-2.200
TokenCl
Token ID2601
Feature activation-6.83e+00

Pos logit contributions
chrom+3.291
proof+3.183
eor+3.035
bos+2.993
roit+2.984
Neg logit contributions
 athlet-3.273
 holiest-3.165
 bargaining-3.141
emi-3.054
olicy-3.015
Tokenod
Token ID375
Feature activation-8.11e+00

Pos logit contributions
roit+3.317
chrom+3.316
obyl+3.292
rans+3.285
bos+3.246
Neg logit contributions
 holiest-6.457
 goodbye-6.406
 bargaining-6.368
 athlet-6.334
QUI-6.326
Tokenagh
Token ID10471
Feature activation-1.48e+00

Pos logit contributions
chrom+4.407
rans+4.161
proof+4.111
obyl+4.085
bos+4.079
Neg logit contributions
 goodbye-7.367
 athlet-7.300
 holiest-7.189
 versa-7.153
 forgiveness-7.088
Token Kil
Token ID19902
Feature activation-2.58e+00

Pos logit contributions
chrom+1.327
proof+1.294
bos+1.278
roit+1.268
rans+1.261
Neg logit contributions
 athlet-0.831
 holiest-0.831
 versa-0.819
 goodbye-0.819
emi-0.804
Tokenc
Token ID66
Feature activation-6.03e+00

Pos logit contributions
chrom+1.600
roit+1.502
proof+1.479
bos+1.470
rans+1.445
Neg logit contributions
 goodbye-2.262
 athlet-2.225
olicy-2.204
 holiest-2.203
emi-2.169
Tokenoy
Token ID726
Feature activation-5.52e+00

Pos logit contributions
proof+1.782
obyl+1.730
chrom+1.580
ards+1.579
oad+1.563
Neg logit contributions
externalActionCode-6.849
 versa-6.781
 holiest-6.737
 athlet-6.727
 bargaining-6.681
Tokenne
Token ID710
Feature activation-5.64e+00

Pos logit contributions
chrom+2.540
roit+2.301
eor+2.259
proof+2.177
rans+2.074
Neg logit contributions
 holiest-5.687
 bargaining-5.551
 athlet-5.537
QUI-5.537
olicy-5.498
Token act
Token ID719
Feature activation-4.84e+00

Pos logit contributions
 holiest+1.438
 goodbye+1.406
 athlet+1.402
emi+1.402
 bargaining+1.383
Neg logit contributions
chrom-2.144
roit-2.013
proof-1.988
obyl-1.982
eor-1.977
Token worth
Token ID2861
Feature activation-1.67e+00

Pos logit contributions
chrom+3.743
roit+3.580
proof+3.559
bos+3.476
rans+3.447
Neg logit contributions
 holiest-3.151
emi-3.078
cean-3.075
olicy-3.046
 Lanka-3.015
Token $
Token ID720
Feature activation-1.51e+00

Pos logit contributions
chrom+1.362
proof+1.353
bos+1.287
roit+1.282
rans+1.279
Neg logit contributions
emi-1.017
 holiest-1.016
olicy-1.002
cean-0.994
assi-0.991
Token20
Token ID1238
Feature activation-6.45e+00

Pos logit contributions
chrom+1.098
proof+1.019
roit+0.970
rans+0.970
bos+0.962
Neg logit contributions
 holiest-1.197
 athlet-1.194
olicy-1.135
 bargaining-1.134
emi-1.129
Token million
Token ID1510
Feature activation-2.73e+00

Pos logit contributions
chrom+4.906
proof+4.600
bos+4.599
rans+4.482
roit+4.476
Neg logit contributions
 holiest-4.140
 Lanka-4.055
emi-4.046
 goodbye-4.030
cean-4.026
Token that
Token ID326
Feature activation-7.12e+00

Pos logit contributions
chrom+2.077
proof+1.982
roit+1.980
bos+1.934
rans+1.923
Neg logit contributions
 holiest-1.880
emi-1.837
olicy-1.833
cean-1.827
 Lanka-1.783
Token was
Token ID373
Feature activation-4.99e+00

Pos logit contributions
chrom+5.073
proof+4.718
roit+4.703
bos+4.637
rans+4.518
Neg logit contributions
emi-4.948
olicy-4.873
 holiest-4.755
assi-4.755
cean-4.734
Token passed
Token ID3804
Feature activation-4.96e+00

Pos logit contributions
chrom+3.524
rans+3.260
roit+3.230
proof+3.208
bos+3.113
Neg logit contributions
emi-3.640
 holiest-3.614
cean-3.571
 srfAttach-3.523
olicy-3.519
Token to
Token ID284
Feature activation-5.73e+00

Pos logit contributions
chrom+3.575
proof+3.413
roit+3.353
bos+3.304
rans+3.245
Neg logit contributions
 holiest-3.477
emi-3.452
cean-3.390
assi-3.383
 Lanka-3.356
Token help
Token ID1037
Feature activation-2.95e+00

Pos logit contributions
chrom+4.036
roit+3.868
proof+3.842
rans+3.711
bos+3.684
Neg logit contributions
 holiest-4.051
cean-3.957
emi-3.951
assi-3.869
olicy-3.858
Token repair
Token ID9185
Feature activation-4.37e+00

Pos logit contributions
chrom+2.250
proof+2.161
roit+2.129
bos+2.087
rans+2.076
Neg logit contributions
emi-2.048
 holiest-2.028
cean-1.971
assi-1.953
olicy-1.937
Token or
Token ID393
Feature activation-2.53e+00

Pos logit contributions
chrom+2.389
roit+2.181
proof+2.171
rans+2.122
isation+2.114
Neg logit contributions
 holiest-2.462
emi-2.366
olicy-2.307
QUI-2.293
 goodbye-2.287
Token wage
Token ID7699
Feature activation+2.06e+00

Pos logit contributions
chrom+2.084
proof+1.924
roit+1.908
rans+1.898
eor+1.858
Neg logit contributions
 holiest-1.631
emi-1.585
 Lanka-1.549
olicy-1.548
 goodbye-1.538
Token employment
Token ID7184
Feature activation-3.67e+00

Pos logit contributions
 holiest+1.378
emi+1.349
 goodbye+1.317
 athlet+1.317
olicy+1.312
Neg logit contributions
chrom-1.727
proof-1.629
roit-1.583
rans-1.554
eor-1.551
Token,
Token ID11
Feature activation-3.18e+00

Pos logit contributions
chrom+2.678
proof+2.542
roit+2.469
bos+2.436
rans+2.424
Neg logit contributions
 holiest-2.642
cean-2.559
olicy-2.543
emi-2.539
CONCLUS-2.504
Token whether
Token ID1771
Feature activation-4.92e+00

Pos logit contributions
chrom+2.391
proof+2.221
roit+2.181
rans+2.176
bos+2.105
Neg logit contributions
 holiest-2.195
olicy-2.178
emi-2.134
cean-2.123
QUI-2.121
Token regular
Token ID3218
Feature activation-6.75e+00

Pos logit contributions
chrom+3.590
roit+3.309
proof+3.157
rans+3.156
bos+3.155
Neg logit contributions
olicy-3.510
 holiest-3.460
emi-3.447
 Lanka-3.394
 srfAttach-3.344
Token or
Token ID393
Feature activation-3.06e+00

Pos logit contributions
chrom+4.537
roit+4.357
isation+4.344
proof+4.285
bos+4.199
Neg logit contributions
 holiest-5.006
emi-4.784
olicy-4.780
 srfAttach-4.671
CONCLUS-4.662
Token casual
Token ID9341
Feature activation-5.28e+00

Pos logit contributions
chrom+2.437
roit+2.298
proof+2.220
bos+2.214
eor+2.183
Neg logit contributions
 holiest-1.989
olicy-1.983
 Lanka-1.936
emi-1.930
 goodbye-1.911
Token,
Token ID11
Feature activation-3.78e+00

Pos logit contributions
chrom+3.661
roit+3.427
proof+3.422
isation+3.373
rans+3.336
Neg logit contributions
 holiest-3.949
olicy-3.783
emi-3.765
 Lanka-3.646
 srfAttach-3.633
Token less
Token ID1342
Feature activation-3.15e+00

Pos logit contributions
chrom+2.868
proof+2.637
roit+2.625
rans+2.622
bos+2.555
Neg logit contributions
 holiest-2.582
olicy-2.543
emi-2.488
cean-2.457
 Lanka-2.454
Token than
Token ID621
Feature activation-2.09e+00

Pos logit contributions
chrom+2.101
roit+1.939
proof+1.913
rans+1.853
ards+1.819
Neg logit contributions
olicy-2.526
emi-2.521
 holiest-2.521
 goodbye-2.463
 Lanka-2.459
Token 2
Token ID362
Feature activation-2.20e+00

Pos logit contributions
chrom+0.520
proof+0.488
roit+0.487
bos+0.478
eor+0.476
Neg logit contributions
 holiest-0.353
emi-0.346
 athlet-0.340
 goodbye-0.337
olicy-0.335
Token.
Token ID13
Feature activation-1.72e+00

Pos logit contributions
chrom+1.660
proof+1.547
roit+1.522
eor+1.502
rans+1.499
Neg logit contributions
 holiest-1.632
emi-1.561
 athlet-1.547
 goodbye-1.535
 bargaining-1.530
TokenD
Token ID35
Feature activation-2.72e+00

Pos logit contributions
chrom+1.419
proof+1.328
roit+1.286
rans+1.281
bos+1.268
Neg logit contributions
 holiest-1.191
 athlet-1.166
 bargaining-1.116
olicy-1.107
emi-1.104
Tokenensity
Token ID6377
Feature activation-3.71e-01

Pos logit contributions
chrom+1.004
roit+0.931
inant+0.887
proof+0.876
rans+0.854
Neg logit contributions
 holiest-3.017
 athlet-2.993
emi-2.923
 goodbye-2.922
 bargaining-2.915
Token as
Token ID355
Feature activation-7.56e-03

Pos logit contributions
chrom+0.298
proof+0.278
roit+0.272
rans+0.270
obyl+0.268
Neg logit contributions
 holiest-0.257
 athlet-0.249
emi-0.247
olicy-0.245
 goodbye-0.241
Token ambient
Token ID25237
Feature activation-5.63e+00

Pos logit contributions
chrom+0.006
proof+0.006
roit+0.006
bos+0.006
rans+0.006
Neg logit contributions
 holiest-0.005
emi-0.005
 athlet-0.005
olicy-0.005
cean-0.005
Token plus
Token ID5556
Feature activation+1.25e-01

Pos logit contributions
chrom+4.094
proof+3.899
roit+3.787
rans+3.706
bos+3.679
Neg logit contributions
 holiest-4.146
emi-3.963
 athlet-3.913
olicy-3.904
 Lanka-3.809
Token deviation
Token ID28833
Feature activation-8.98e-01

Pos logit contributions
 holiest+0.102
emi+0.100
 athlet+0.099
 goodbye+0.097
olicy+0.097
Neg logit contributions
chrom-0.084
proof-0.077
roit-0.076
eor-0.074
obyl-0.074
Token.
Token ID13
Feature activation-9.41e-01

Pos logit contributions
chrom+0.708
proof+0.675
roit+0.660
rans+0.650
obyl+0.647
Neg logit contributions
 holiest-0.642
emi-0.605
 athlet-0.604
olicy-0.592
 goodbye-0.583
Token Fl
Token ID1610
Feature activation+1.54e+00

Pos logit contributions
chrom+0.763
proof+0.720
roit+0.706
rans+0.696
obyl+0.692
Neg logit contributions
 holiest-0.653
 athlet-0.635
emi-0.617
 bargaining-0.614
 goodbye-0.614
Tokenuid
Token ID27112
Feature activation-6.70e-01

Pos logit contributions
 holiest+1.490
 goodbye+1.463
 athlet+1.455
 bargaining+1.446
cean+1.437
Neg logit contributions
chrom-0.840
roit-0.766
proof-0.761
obyl-0.739
bos-0.732