TOP ACTIVATIONS
MAX = 11.573

government of President Dil ma Rousse ff and no one in
not clear why Mr . Mor neau located his investment holding
." Mr . Mor neau , who at times
party abandoned President Dil ma Rousse ff s governing
as racist or ridiculous . Gork a is reported to be
Mr . Mor neau , who at times appeared
in Ottawa , Mr . Mor neau also jointly owns four
clear why Mr . Mor neau located his investment holding company
. Mr . Mor neau 's Alberta - based
Mr . Mor neau 's Alberta - based investment
Canadians ," Mr . Mor neau replied . " We also
that are usually employed to mitigate high bandwidth Layer 3 &
Also , Mr . Mor neau says he is sole owner
class Canadians ," Mr . Mor neau replied . " We
raised . Mc Connell was une qu
Ottawa , Mr . Mor neau also jointly owns four numbered
deal to save the Def erred Action for Childhood Arri vals
. S . Ambassador to Haiti , once dubbed the U
millionaire owners and shareholders of Mor neau She pell will not
owners and shareholders of Mor neau She pell will not have

MIN ACTIVATIONS
MIN = -9.794

too many times . The drummer looked like Tim Y oh
guitar . Then , a singer with a bal acl ava
spot , implying that the restaurant was boiling a soup pot
to the bedroom from the kitchen , with the toilet being
sites , many of the scenes look familiar to today 's
a lot happens : the characters go to a lot of
the gratification spectrum . The characters carefully controlled inter
box ers . Every time my nose
So i F on ! There was a sudden
movement , shun po . I m
ifted and put on a basement of concrete . Ma ori
way , and a cerebral journalist played by Diane Ke aton
version of one of his characters , and just It
as Ted Falcon i , playing guitar . Then , a
I say it was an actor s clinic .
un ruly slave . The story was almost certainly Greek propaganda
Cour tyard , where the characters boast weird ly elong ated
on that set . The boys had a lot of fun
burdens of meaning passed from writer to reader . The one
one of them ; it s just a very

INTERVAL 6.231 - 11.573
CONTAINS 0.182%

" Special needs vouchers are a product of calculating
along with the commitment to NATO . May spoke
Removal Technology in the Twin Fro z r IV enhancing the
international efforts to combat climate change . B ilateral cooperation with
need to match each network change with a respective fine -

INTERVAL 0.889 - 6.231
CONTAINS 32.642%

IM AGES HAVE BE EN T AK EN DOWN UP
who had heard of investor and Democratic money man George Soros
. S . currency , and making a $ 9 billion
tenure , and characterized Hag el 's resignation as his own
cor ned beef and thr ice - cooked bacon tossed with

INTERVAL -4.452 - 0.889
CONTAINS 65.925%

Aviation Week s Lara Sel ig man reminded folks
, the region near the cal ving front shows a dramatic
was again Rob ben who wre aked havoc in the Mexico
in the jam on a media bus ) pic . twitter
's photos , click here : Follow @ Pro

INTERVAL -9.794 - -4.452
CONTAINS 1.250%

's origin as a musician at a piano bar , who
who showed none , is atop . Such are the wages
um op us . A mel odic and melancholy meditation on
Moore said in an interview last year ,
sensational black and white love letter to New York City may
Token government
Token ID1230
Feature activation+4.28e+00

Pos logit contributions
 confir+1.823
 destro+1.635
 consolidate+1.577
 rely+1.530
 consume+1.522
Neg logit contributions
club-2.078
cent-2.073
wives-2.050
horse-2.040
cart-1.998
Token of
Token ID286
Feature activation+1.79e-01

Pos logit contributions
 confir+4.484
 destro+4.059
 consolidate+3.982
 rely+3.887
 comply+3.840
Neg logit contributions
club-3.505
cent-3.474
wives-3.450
horse-3.401
masters-3.347
Token President
Token ID1992
Feature activation+2.66e+00

Pos logit contributions
 confir+0.194
 destro+0.178
 consolidate+0.172
 rely+0.168
 consume+0.168
Neg logit contributions
club-0.144
cent-0.144
wives-0.141
horse-0.141
cart-0.137
Token Dil
Token ID29328
Feature activation+7.56e+00

Pos logit contributions
 confir+2.803
 destro+2.555
 consolidate+2.508
 rely+2.456
 consume+2.427
Neg logit contributions
club-2.160
cent-2.142
wives-2.131
horse-2.106
masters-2.062
Tokenma
Token ID2611
Feature activation+4.78e+00

Pos logit contributions
 confir+8.354
 consolidate+7.935
 rely+7.787
 consume+7.770
 destro+7.740
Neg logit contributions
club-4.826
cent-4.739
horse-4.732
lines-4.727
wives-4.715
Token Rousse
Token ID42849
Feature activation+1.16e+01

Pos logit contributions
 confir+4.228
 destro+3.750
 consolidate+3.677
 rely+3.600
 comply+3.535
Neg logit contributions
club-4.731
cent-4.666
horse-4.602
wives-4.593
masters-4.527
Tokenff
Token ID487
Feature activation+5.84e+00

Pos logit contributions
 confir+8.356
 comply+8.042
 consolidate+8.022
 rely+7.915
 consume+7.772
Neg logit contributions
club-10.783
wives-10.734
lines-10.468
horse-10.375
cent-10.360
Token and
Token ID290
Feature activation+3.27e+00

Pos logit contributions
 confir+6.157
 destro+5.551
 consolidate+5.454
 rely+5.332
 comply+5.300
Neg logit contributions
club-4.641
cent-4.558
wives-4.531
horse-4.496
masters-4.415
Token no
Token ID645
Feature activation+3.52e+00

Pos logit contributions
 confir+3.146
 destro+2.833
 consolidate+2.760
 rely+2.678
 consume+2.655
Neg logit contributions
club-3.025
cent-3.014
horse-2.963
wives-2.955
cart-2.908
Token one
Token ID530
Feature activation+3.79e+00

Pos logit contributions
 confir+3.300
 destro+2.938
 consolidate+2.869
 rely+2.784
 consume+2.765
Neg logit contributions
club-3.317
cent-3.306
horse-3.252
wives-3.247
cart-3.185
Token in
Token ID287
Feature activation+9.96e-02

Pos logit contributions
 confir+3.539
 destro+3.128
 consolidate+3.063
 rely+2.971
 consume+2.962
Neg logit contributions
club-3.601
cent-3.593
wives-3.547
horse-3.524
cart-3.475
Token not
Token ID407
Feature activation+4.45e+00

Pos logit contributions
club+0.006
cent+0.006
wives+0.006
horse+0.006
lines+0.006
Neg logit contributions
 confir-0.006
 destro-0.006
 consolidate-0.006
 rely-0.006
 consume-0.005
Token clear
Token ID1598
Feature activation+4.37e+00

Pos logit contributions
 confir+3.749
 destro+3.284
 consolidate+3.170
 rely+3.136
 consume+3.091
Neg logit contributions
club-4.591
cent-4.583
wives-4.528
horse-4.475
cart-4.434
Token why
Token ID1521
Feature activation+7.15e+00

Pos logit contributions
 confir+3.884
 destro+3.456
 consolidate+3.360
 rely+3.280
 consume+3.267
Neg logit contributions
club-4.304
cent-4.296
wives-4.232
horse-4.200
cart-4.152
Token Mr
Token ID1770
Feature activation+2.69e+00

Pos logit contributions
 confir+7.347
 destro+6.591
 consolidate+6.539
 rely+6.430
 consume+6.423
Neg logit contributions
club-5.815
cent-5.715
wives-5.601
horse-5.586
cart-5.548
Token.
Token ID13
Feature activation-6.82e-01

Pos logit contributions
 confir+2.675
 consolidate+2.415
 destro+2.400
 rely+2.385
 consume+2.368
Neg logit contributions
club-2.257
cent-2.235
wives-2.209
horse-2.182
lines-2.148
Token Mor
Token ID3461
Feature activation+1.00e+01

Pos logit contributions
club+0.474
cent+0.473
wives+0.463
horse+0.460
lines+0.450
Neg logit contributions
 confir-0.813
 destro-0.750
 consolidate-0.728
 rely-0.713
 consume-0.712
Tokenneau
Token ID46533
Feature activation+9.29e+00

Pos logit contributions
 confir+7.461
 consolidate+6.907
 consume+6.773
 rely+6.680
 comply+6.664
Neg logit contributions
club-9.825
wives-9.710
cent-9.576
masters-9.528
lines-9.431
Token located
Token ID5140
Feature activation-4.82e-01

Pos logit contributions
 confir+8.108
 consolidate+7.164
 destro+7.097
 rely+7.085
 consume+6.900
Neg logit contributions
club-8.696
cent-8.636
wives-8.539
masters-8.401
laughs-8.399
Token his
Token ID465
Feature activation+1.15e+00

Pos logit contributions
club+0.370
cent+0.369
wives+0.361
horse+0.360
cart+0.352
Neg logit contributions
 confir-0.539
 destro-0.494
 consolidate-0.480
 rely-0.470
 consume-0.469
Token investment
Token ID4896
Feature activation+2.34e+00

Pos logit contributions
 confir+1.172
 destro+1.064
 consolidate+1.033
 rely+1.011
 consume+1.005
Neg logit contributions
club-0.988
cent-0.980
wives-0.967
horse-0.961
lines-0.944
Token holding
Token ID4769
Feature activation+1.13e+00

Pos logit contributions
 confir+2.488
 destro+2.260
 consolidate+2.224
 rely+2.194
 consume+2.156
Neg logit contributions
club-1.882
cent-1.874
wives-1.872
horse-1.815
laughs-1.802
Token."
Token ID526
Feature activation-6.73e-01

Pos logit contributions
 confir+3.497
 destro+3.121
 consolidate+3.053
 rely+2.981
 consume+2.974
Neg logit contributions
club-3.352
cent-3.328
wives-3.279
horse-3.266
cart-3.202
Token
Token ID198
Feature activation+1.62e+00

Pos logit contributions
club+0.490
cent+0.489
wives+0.479
horse+0.477
lines+0.467
Neg logit contributions
 confir-0.778
 destro-0.717
 consolidate-0.695
 rely-0.682
 consume-0.680
Token
Token ID198
Feature activation+1.40e+00

Pos logit contributions
 confir+1.759
 consolidate+1.600
 destro+1.597
 rely+1.580
 consume+1.570
Neg logit contributions
club-1.223
cent-1.220
wives-1.201
horse-1.190
lines-1.163
TokenMr
Token ID5246
Feature activation-9.06e-01

Pos logit contributions
 confir+1.624
 destro+1.486
 consolidate+1.476
 rely+1.456
 consume+1.451
Neg logit contributions
club-0.930
cent-0.924
wives-0.912
horse-0.904
lines-0.889
Token.
Token ID13
Feature activation-2.06e+00

Pos logit contributions
cent+0.715
club+0.713
wives+0.699
horse+0.697
lines+0.683
Neg logit contributions
 confir-1.008
 destro-0.927
 consolidate-0.887
 rely-0.865
 consume-0.864
Token Mor
Token ID3461
Feature activation+1.00e+01

Pos logit contributions
club+1.604
cent+1.604
wives+1.571
horse+1.564
lines+1.533
Neg logit contributions
 confir-2.280
 destro-2.092
 consolidate-2.021
 rely-1.978
 consume-1.974
Tokenneau
Token ID46533
Feature activation+9.81e+00

Pos logit contributions
 confir+7.360
 consolidate+6.767
 consume+6.668
 rely+6.577
 comply+6.569
Neg logit contributions
club-9.864
wives-9.741
cent-9.601
masters-9.535
lines-9.451
Token,
Token ID11
Feature activation+4.92e-01

Pos logit contributions
 confir+9.637
 destro+8.516
 consolidate+8.415
 rely+8.336
 consume+8.199
Neg logit contributions
club-8.082
cent-7.911
wives-7.905
masters-7.727
horse-7.725
Token who
Token ID508
Feature activation-5.81e-01

Pos logit contributions
 confir+0.569
 destro+0.518
 consolidate+0.509
 rely+0.502
 consume+0.498
Neg logit contributions
club-0.360
cent-0.351
wives-0.349
horse-0.348
cart-0.341
Token at
Token ID379
Feature activation-2.65e+00

Pos logit contributions
club+0.587
cent+0.587
wives+0.579
horse+0.576
lines+0.568
Neg logit contributions
 confir-0.509
 destro-0.456
 consolidate-0.436
 rely-0.423
 consume-0.422
Token times
Token ID1661
Feature activation+6.12e-01

Pos logit contributions
club+2.085
cent+2.082
wives+2.041
horse+2.032
lines+1.991
Neg logit contributions
 confir-2.912
 destro-2.668
 consolidate-2.582
 rely-2.527
 consume-2.520
Token party
Token ID2151
Feature activation+1.98e-01

Pos logit contributions
 confir+0.615
 destro+0.553
 consolidate+0.532
 rely+0.520
 consume+0.519
Neg logit contributions
club-0.623
cent-0.623
wives-0.613
horse-0.611
lines-0.600
Token abandoned
Token ID9958
Feature activation+1.54e+00

Pos logit contributions
 confir+0.189
 destro+0.170
 consolidate+0.164
 rely+0.160
 consume+0.159
Neg logit contributions
club-0.185
cent-0.185
wives-0.182
horse-0.181
cart-0.178
Token President
Token ID1992
Feature activation+1.34e+00

Pos logit contributions
 confir+1.548
 destro+1.399
 consolidate+1.362
 rely+1.329
 consume+1.321
Neg logit contributions
club-1.359
cent-1.353
wives-1.335
horse-1.322
cart-1.299
Token Dil
Token ID29328
Feature activation+4.06e+00

Pos logit contributions
 confir+1.422
 destro+1.290
 consolidate+1.261
 rely+1.239
 consume+1.226
Neg logit contributions
club-1.101
cent-1.094
wives-1.088
horse-1.073
masters-1.056
Tokenma
Token ID2611
Feature activation+2.70e+00

Pos logit contributions
 confir+4.687
 consolidate+4.434
 destro+4.387
 rely+4.345
 consume+4.300
Neg logit contributions
club-2.627
cent-2.568
horse-2.560
lines-2.536
masters-2.535
Token Rousse
Token ID42849
Feature activation+9.88e+00

Pos logit contributions
 confir+2.363
 destro+2.087
 consolidate+2.052
 rely+2.005
 consume+1.972
Neg logit contributions
club-2.740
cent-2.711
wives-2.666
horse-2.663
lines-2.632
Tokenff
Token ID487
Feature activation+3.66e+00

Pos logit contributions
 confir+7.857
 consolidate+7.285
 comply+7.178
 rely+7.135
 deprive+7.061
Neg logit contributions
club-9.313
wives-9.083
lines-8.940
horse-8.893
cent-8.867
Token
Token ID447
Feature activation+1.30e+00

Pos logit contributions
 confir+3.429
 destro+3.054
 consolidate+2.994
 rely+2.916
 consume+2.879
Neg logit contributions
club-3.437
cent-3.422
wives-3.378
horse-3.355
masters-3.298
Token
Token ID247
Feature activation+5.89e-01

Pos logit contributions
 confir+1.761
 destro+1.629
 consolidate+1.589
 rely+1.567
 consume+1.560
Neg logit contributions
club-0.679
cent-0.674
wives-0.661
horse-0.650
cart-0.638
Tokens
Token ID82
Feature activation-1.05e-01

Pos logit contributions
 confir+0.746
 destro+0.687
 consolidate+0.672
 rely+0.659
 consume+0.656
Neg logit contributions
club-0.365
cent-0.363
wives-0.356
horse-0.353
lines-0.342
Token governing
Token ID15030
Feature activation+2.50e+00

Pos logit contributions
club+0.109
cent+0.109
wives+0.108
horse+0.107
lines+0.106
Neg logit contributions
 confir-0.088
 destro-0.078
 consolidate-0.075
 rely-0.073
 consume-0.073
Token as
Token ID355
Feature activation+5.59e-01

Pos logit contributions
club+0.331
cent+0.330
wives+0.325
horse+0.323
lines+0.318
Neg logit contributions
 confir-0.342
 destro-0.309
 consolidate-0.299
 rely-0.292
 consume-0.291
Token racist
Token ID9811
Feature activation+4.83e-01

Pos logit contributions
 confir+0.483
 destro+0.430
 consolidate+0.415
 rely+0.404
 consume+0.402
Neg logit contributions
club-0.572
cent-0.569
wives-0.562
horse-0.560
lines-0.551
Token or
Token ID393
Feature activation+2.58e+00

Pos logit contributions
 confir+0.512
 destro+0.467
 consolidate+0.455
 rely+0.445
 consume+0.443
Neg logit contributions
club-0.397
cent-0.396
wives-0.389
horse-0.386
lines-0.378
Token ridiculous
Token ID11441
Feature activation+1.95e+00

Pos logit contributions
 confir+2.160
 destro+1.916
 consolidate+1.861
 rely+1.812
 consume+1.806
Neg logit contributions
club-2.697
cent-2.677
wives-2.658
horse-2.633
lines-2.610
Token.
Token ID13
Feature activation-4.67e-01

Pos logit contributions
 confir+2.017
 destro+1.830
 consolidate+1.791
 rely+1.762
 consume+1.749
Neg logit contributions
club-1.628
cent-1.621
wives-1.590
horse-1.586
cart-1.551
Token Gork
Token ID49882
Feature activation+9.82e+00

Pos logit contributions
cent+0.312
club+0.311
wives+0.305
horse+0.302
lines+0.294
Neg logit contributions
 confir-0.555
 destro-0.510
 consolidate-0.505
 rely-0.498
 consume-0.493
Tokena
Token ID64
Feature activation+1.48e+00

Pos logit contributions
 confir+10.801
 consolidate+10.218
 rely+10.199
 consume+10.121
 comply+9.991
Neg logit contributions
wives-5.838
club-5.779
cent-5.745
horse-5.497
lines-5.430
Token is
Token ID318
Feature activation-1.33e+00

Pos logit contributions
 confir+1.342
 destro+1.201
 consolidate+1.158
 rely+1.132
 consume+1.126
Neg logit contributions
club-1.444
cent-1.442
wives-1.420
horse-1.412
cart-1.387
Token reported
Token ID2098
Feature activation-1.37e+00

Pos logit contributions
club+1.171
cent+1.169
wives+1.152
horse+1.145
lines+1.123
Neg logit contributions
 confir-1.344
 destro-1.218
 consolidate-1.177
 rely-1.151
 consume-1.147
Token to
Token ID284
Feature activation+1.25e+00

Pos logit contributions
club+1.258
cent+1.257
wives+1.235
horse+1.231
lines+1.210
Neg logit contributions
 confir-1.320
 destro-1.194
 consolidate-1.148
 rely-1.119
 consume-1.117
Token be
Token ID307
Feature activation+1.52e+00

Pos logit contributions
 confir+0.726
 destro+0.605
 consolidate+0.574
 rely+0.553
 consume+0.548
Neg logit contributions
cent-1.642
club-1.640
wives-1.624
horse-1.618
cart-1.600
Token
Token ID198
Feature activation+1.62e+00

Pos logit contributions
club+0.490
cent+0.489
wives+0.479
horse+0.477
lines+0.467
Neg logit contributions
 confir-0.778
 destro-0.717
 consolidate-0.695
 rely-0.682
 consume-0.680
Token
Token ID198
Feature activation+1.40e+00

Pos logit contributions
 confir+1.759
 consolidate+1.600
 destro+1.597
 rely+1.580
 consume+1.570
Neg logit contributions
club-1.223
cent-1.220
wives-1.201
horse-1.190
lines-1.163
TokenMr
Token ID5246
Feature activation-9.06e-01

Pos logit contributions
 confir+1.624
 destro+1.486
 consolidate+1.476
 rely+1.456
 consume+1.451
Neg logit contributions
club-0.930
cent-0.924
wives-0.912
horse-0.904
lines-0.889
Token.
Token ID13
Feature activation-2.06e+00

Pos logit contributions
cent+0.715
club+0.713
wives+0.699
horse+0.697
lines+0.683
Neg logit contributions
 confir-1.008
 destro-0.927
 consolidate-0.887
 rely-0.865
 consume-0.864
Token Mor
Token ID3461
Feature activation+1.00e+01

Pos logit contributions
club+1.604
cent+1.604
wives+1.571
horse+1.564
lines+1.533
Neg logit contributions
 confir-2.280
 destro-2.092
 consolidate-2.021
 rely-1.978
 consume-1.974
Tokenneau
Token ID46533
Feature activation+9.81e+00

Pos logit contributions
 confir+7.360
 consolidate+6.767
 consume+6.668
 rely+6.577
 comply+6.569
Neg logit contributions
club-9.864
wives-9.741
cent-9.601
masters-9.535
lines-9.451
Token,
Token ID11
Feature activation+4.92e-01

Pos logit contributions
 confir+9.637
 destro+8.516
 consolidate+8.415
 rely+8.336
 consume+8.199
Neg logit contributions
club-8.082
cent-7.911
wives-7.905
masters-7.727
horse-7.725
Token who
Token ID508
Feature activation-5.81e-01

Pos logit contributions
 confir+0.569
 destro+0.518
 consolidate+0.509
 rely+0.502
 consume+0.498
Neg logit contributions
club-0.360
cent-0.351
wives-0.349
horse-0.348
cart-0.341
Token at
Token ID379
Feature activation-2.65e+00

Pos logit contributions
club+0.587
cent+0.587
wives+0.579
horse+0.576
lines+0.568
Neg logit contributions
 confir-0.509
 destro-0.456
 consolidate-0.436
 rely-0.423
 consume-0.422
Token times
Token ID1661
Feature activation+6.12e-01

Pos logit contributions
club+2.085
cent+2.082
wives+2.041
horse+2.032
lines+1.991
Neg logit contributions
 confir-2.912
 destro-2.668
 consolidate-2.582
 rely-2.527
 consume-2.520
Token appeared
Token ID4120
Feature activation-1.93e+00

Pos logit contributions
 confir+0.529
 destro+0.472
 consolidate+0.453
 rely+0.442
 consume+0.440
Neg logit contributions
club-0.626
cent-0.625
wives-0.615
horse-0.613
lines-0.603
Token in
Token ID287
Feature activation-5.40e-01

Pos logit contributions
club+0.611
cent+0.609
wives+0.598
horse+0.595
lines+0.584
Neg logit contributions
 confir-0.777
 destro-0.708
 consolidate-0.685
 rely-0.670
 consume-0.668
Token Ottawa
Token ID14074
Feature activation+2.16e+00

Pos logit contributions
club+0.424
cent+0.420
wives+0.414
horse+0.410
cart+0.404
Neg logit contributions
 confir-0.595
 destro-0.545
 consolidate-0.530
 rely-0.519
 consume-0.517
Token,
Token ID11
Feature activation+1.46e+00

Pos logit contributions
 confir+2.170
 destro+1.955
 consolidate+1.915
 rely+1.877
 consume+1.849
Neg logit contributions
club-1.898
cent-1.873
wives-1.848
horse-1.824
cart-1.812
Token Mr
Token ID1770
Feature activation+8.88e-02

Pos logit contributions
 confir+1.618
 destro+1.476
 consolidate+1.440
 rely+1.411
 consume+1.407
Neg logit contributions
club-1.136
cent-1.129
wives-1.106
horse-1.103
cart-1.082
Token.
Token ID13
Feature activation-1.72e+00

Pos logit contributions
 confir+0.097
 destro+0.089
 consolidate+0.086
 rely+0.084
 consume+0.083
Neg logit contributions
club-0.071
cent-0.071
wives-0.070
horse-0.069
lines-0.068
Token Mor
Token ID3461
Feature activation+9.41e+00

Pos logit contributions
cent+1.177
club+1.174
wives+1.149
horse+1.144
lines+1.115
Neg logit contributions
 confir-2.073
 destro-1.916
 consolidate-1.855
 rely-1.818
 consume-1.814
Tokenneau
Token ID46533
Feature activation+9.02e+00

Pos logit contributions
 confir+7.165
 consolidate+6.602
 consume+6.471
 rely+6.387
 comply+6.376
Neg logit contributions
club-9.190
wives-9.089
cent-8.966
masters-8.912
lines-8.816
Token also
Token ID635
Feature activation+2.90e+00

Pos logit contributions
 confir+8.312
 consolidate+7.496
 destro+7.414
 rely+7.334
 consume+7.215
Neg logit contributions
club-7.936
cent-7.803
wives-7.770
laughs-7.681
cart-7.624
Token jointly
Token ID26913
Feature activation+2.89e-01

Pos logit contributions
 confir+2.400
 destro+2.120
 consolidate+2.059
 rely+2.005
 consume+1.985
Neg logit contributions
club-3.065
cent-3.055
wives-3.008
horse-2.997
cart-2.951
Token owns
Token ID12216
Feature activation-1.02e+00

Pos logit contributions
 confir+0.247
 destro+0.220
 consolidate+0.211
 rely+0.205
 consume+0.204
Neg logit contributions
club-0.299
cent-0.298
wives-0.294
horse-0.293
lines-0.288
Token four
Token ID1440
Feature activation-1.53e+00

Pos logit contributions
club+0.770
cent+0.769
wives+0.753
horse+0.749
lines+0.732
Neg logit contributions
 confir-1.165
 destro-1.068
 consolidate-1.038
 rely-1.017
 consume-1.015
Token clear
Token ID1598
Feature activation+4.37e+00

Pos logit contributions
 confir+3.749
 destro+3.284
 consolidate+3.170
 rely+3.136
 consume+3.091
Neg logit contributions
club-4.591
cent-4.583
wives-4.528
horse-4.475
cart-4.434
Token why
Token ID1521
Feature activation+7.15e+00

Pos logit contributions
 confir+3.884
 destro+3.456
 consolidate+3.360
 rely+3.280
 consume+3.267
Neg logit contributions
club-4.304
cent-4.296
wives-4.232
horse-4.200
cart-4.152
Token Mr
Token ID1770
Feature activation+2.69e+00

Pos logit contributions
 confir+7.347
 destro+6.591
 consolidate+6.539
 rely+6.430
 consume+6.423
Neg logit contributions
club-5.815
cent-5.715
wives-5.601
horse-5.586
cart-5.548
Token.
Token ID13
Feature activation-6.82e-01

Pos logit contributions
 confir+2.675
 consolidate+2.415
 destro+2.400
 rely+2.385
 consume+2.368
Neg logit contributions
club-2.257
cent-2.235
wives-2.209
horse-2.182
lines-2.148
Token Mor
Token ID3461
Feature activation+1.00e+01

Pos logit contributions
club+0.474
cent+0.473
wives+0.463
horse+0.460
lines+0.450
Neg logit contributions
 confir-0.813
 destro-0.750
 consolidate-0.728
 rely-0.713
 consume-0.712
Tokenneau
Token ID46533
Feature activation+9.29e+00

Pos logit contributions
 confir+7.461
 consolidate+6.907
 consume+6.773
 rely+6.680
 comply+6.664
Neg logit contributions
club-9.825
wives-9.710
cent-9.576
masters-9.528
lines-9.431
Token located
Token ID5140
Feature activation-4.82e-01

Pos logit contributions
 confir+8.108
 consolidate+7.164
 destro+7.097
 rely+7.085
 consume+6.900
Neg logit contributions
club-8.696
cent-8.636
wives-8.539
masters-8.401
laughs-8.399
Token his
Token ID465
Feature activation+1.15e+00

Pos logit contributions
club+0.370
cent+0.369
wives+0.361
horse+0.360
cart+0.352
Neg logit contributions
 confir-0.539
 destro-0.494
 consolidate-0.480
 rely-0.470
 consume-0.469
Token investment
Token ID4896
Feature activation+2.34e+00

Pos logit contributions
 confir+1.172
 destro+1.064
 consolidate+1.033
 rely+1.011
 consume+1.005
Neg logit contributions
club-0.988
cent-0.980
wives-0.967
horse-0.961
lines-0.944
Token holding
Token ID4769
Feature activation+1.13e+00

Pos logit contributions
 confir+2.488
 destro+2.260
 consolidate+2.224
 rely+2.194
 consume+2.156
Neg logit contributions
club-1.882
cent-1.874
wives-1.872
horse-1.815
laughs-1.802
Token company
Token ID1664
Feature activation+6.56e-02

Pos logit contributions
 confir+1.287
 destro+1.177
 consolidate+1.160
 rely+1.140
 consume+1.127
Neg logit contributions
club-0.838
cent-0.835
wives-0.827
horse-0.812
cart-0.795
Token.
Token ID13
Feature activation-5.74e-01

Pos logit contributions
club+0.784
cent+0.783
wives+0.768
horse+0.765
lines+0.750
Neg logit contributions
 confir-1.041
 destro-0.952
 consolidate-0.919
 rely-0.898
 consume-0.897
Token
Token ID198
Feature activation+1.90e+00

Pos logit contributions
cent+0.404
club+0.404
wives+0.395
horse+0.394
lines+0.386
Neg logit contributions
 confir-0.671
 destro-0.617
 consolidate-0.604
 rely-0.593
 consume-0.591
Token
Token ID198
Feature activation+1.75e+00

Pos logit contributions
 confir+2.041
 consolidate+1.863
 destro+1.848
 rely+1.842
 consume+1.828
Neg logit contributions
club-1.442
cent-1.438
wives-1.417
horse-1.402
lines-1.371
TokenMr
Token ID5246
Feature activation+8.32e-01

Pos logit contributions
 confir+2.039
 destro+1.867
 consolidate+1.858
 rely+1.830
 consume+1.823
Neg logit contributions
club-1.146
cent-1.143
wives-1.124
horse-1.113
lines-1.095
Token.
Token ID13
Feature activation-1.57e+00

Pos logit contributions
 confir+0.880
 destro+0.800
 consolidate+0.783
 rely+0.768
 consume+0.765
Neg logit contributions
club-0.678
cent-0.675
wives-0.664
horse-0.659
lines-0.647
Token Mor
Token ID3461
Feature activation+9.27e+00

Pos logit contributions
cent+1.099
club+1.099
wives+1.074
horse+1.069
lines+1.044
Neg logit contributions
 confir-1.872
 destro-1.728
 consolidate-1.674
 rely-1.640
 consume-1.637
Tokenneau
Token ID46533
Feature activation+9.26e+00

Pos logit contributions
 confir+7.072
 consolidate+6.466
 consume+6.339
 rely+6.263
 comply+6.252
Neg logit contributions
club-9.082
wives-8.976
cent-8.866
masters-8.794
lines-8.709
Token's
Token ID338
Feature activation-3.90e-01

Pos logit contributions
 confir+8.568
 consolidate+7.459
 destro+7.456
 rely+7.354
 consume+7.168
Neg logit contributions
club-8.247
cent-8.122
wives-8.071
horse-7.935
masters-7.903
Token Alberta
Token ID15555
Feature activation+4.26e+00

Pos logit contributions
club+0.363
cent+0.362
wives+0.356
horse+0.355
lines+0.349
Neg logit contributions
 confir-0.373
 destro-0.337
 consolidate-0.325
 rely-0.317
 consume-0.316
Token-
Token ID12
Feature activation-1.41e+00

Pos logit contributions
 confir+4.372
 destro+3.910
 consolidate+3.867
 rely+3.791
 consume+3.720
Neg logit contributions
club-3.596
wives-3.542
cent-3.538
horse-3.507
lines-3.444
Tokenbased
Token ID3106
Feature activation+6.33e-01

Pos logit contributions
club+0.857
cent+0.854
wives+0.835
horse+0.828
lines+0.808
Neg logit contributions
 confir-1.805
 destro-1.673
 consolidate-1.630
 rely-1.600
 consume-1.596
Token
Token ID198
Feature activation+1.90e+00

Pos logit contributions
cent+0.404
club+0.404
wives+0.395
horse+0.394
lines+0.386
Neg logit contributions
 confir-0.671
 destro-0.617
 consolidate-0.604
 rely-0.593
 consume-0.591
Token
Token ID198
Feature activation+1.75e+00

Pos logit contributions
 confir+2.041
 consolidate+1.863
 destro+1.848
 rely+1.842
 consume+1.828
Neg logit contributions
club-1.442
cent-1.438
wives-1.417
horse-1.402
lines-1.371
TokenMr
Token ID5246
Feature activation+8.32e-01

Pos logit contributions
 confir+2.039
 destro+1.867
 consolidate+1.858
 rely+1.830
 consume+1.823
Neg logit contributions
club-1.146
cent-1.143
wives-1.124
horse-1.113
lines-1.095
Token.
Token ID13
Feature activation-1.57e+00

Pos logit contributions
 confir+0.880
 destro+0.800
 consolidate+0.783
 rely+0.768
 consume+0.765
Neg logit contributions
club-0.678
cent-0.675
wives-0.664
horse-0.659
lines-0.647
Token Mor
Token ID3461
Feature activation+9.27e+00

Pos logit contributions
cent+1.099
club+1.099
wives+1.074
horse+1.069
lines+1.044
Neg logit contributions
 confir-1.872
 destro-1.728
 consolidate-1.674
 rely-1.640
 consume-1.637
Tokenneau
Token ID46533
Feature activation+9.26e+00

Pos logit contributions
 confir+7.072
 consolidate+6.466
 consume+6.339
 rely+6.263
 comply+6.252
Neg logit contributions
club-9.082
wives-8.976
cent-8.866
masters-8.794
lines-8.709
Token's
Token ID338
Feature activation-3.90e-01

Pos logit contributions
 confir+8.568
 consolidate+7.459
 destro+7.456
 rely+7.354
 consume+7.168
Neg logit contributions
club-8.247
cent-8.122
wives-8.071
horse-7.935
masters-7.903
Token Alberta
Token ID15555
Feature activation+4.26e+00

Pos logit contributions
club+0.363
cent+0.362
wives+0.356
horse+0.355
lines+0.349
Neg logit contributions
 confir-0.373
 destro-0.337
 consolidate-0.325
 rely-0.317
 consume-0.316
Token-
Token ID12
Feature activation-1.41e+00

Pos logit contributions
 confir+4.372
 destro+3.910
 consolidate+3.867
 rely+3.791
 consume+3.720
Neg logit contributions
club-3.596
wives-3.542
cent-3.538
horse-3.507
lines-3.444
Tokenbased
Token ID3106
Feature activation+6.33e-01

Pos logit contributions
club+0.857
cent+0.854
wives+0.835
horse+0.828
lines+0.808
Neg logit contributions
 confir-1.805
 destro-1.673
 consolidate-1.630
 rely-1.600
 consume-1.596
Token investment
Token ID4896
Feature activation+2.85e+00

Pos logit contributions
 confir+0.641
 destro+0.580
 consolidate+0.562
 rely+0.549
 consume+0.546
Neg logit contributions
club-0.554
cent-0.552
wives-0.544
horse-0.541
lines-0.532
Token Canadians
Token ID14008
Feature activation+4.24e+00

Pos logit contributions
 confir+0.656
 destro+0.603
 consolidate+0.587
 rely+0.576
 consume+0.574
Neg logit contributions
club-0.437
cent-0.435
wives-0.426
horse-0.424
cart-0.415
Token,"
Token ID553
Feature activation-1.45e-01

Pos logit contributions
 confir+4.128
 destro+3.749
 consolidate+3.707
 rely+3.649
 consume+3.632
Neg logit contributions
club-3.770
cent-3.752
wives-3.699
horse-3.679
cart-3.623
Token Mr
Token ID1770
Feature activation-6.03e-01

Pos logit contributions
club+0.128
cent+0.128
wives+0.126
horse+0.125
lines+0.123
Neg logit contributions
 confir-0.145
 destro-0.131
 consolidate-0.127
 rely-0.124
 consume-0.124
Token.
Token ID13
Feature activation-1.40e+00

Pos logit contributions
cent+0.478
club+0.477
wives+0.467
horse+0.466
lines+0.457
Neg logit contributions
 confir-0.666
 destro-0.612
 consolidate-0.587
 rely-0.573
 consume-0.573
Token Mor
Token ID3461
Feature activation+9.15e+00

Pos logit contributions
cent+0.965
club+0.965
wives+0.942
horse+0.938
lines+0.916
Neg logit contributions
 confir-1.679
 destro-1.552
 consolidate-1.502
 rely-1.472
 consume-1.469
Tokenneau
Token ID46533
Feature activation+9.23e+00

Pos logit contributions
 confir+6.887
 consolidate+6.338
 consume+6.231
 rely+6.157
 comply+6.136
Neg logit contributions
club-9.016
wives-8.908
cent-8.765
masters-8.696
lines-8.636
Token replied
Token ID8712
Feature activation-1.71e+00

Pos logit contributions
 confir+7.976
 destro+6.969
 consolidate+6.944
 rely+6.825
 consume+6.726
Neg logit contributions
club-8.521
wives-8.411
cent-8.283
laughs-8.217
horse-8.148
Token.
Token ID13
Feature activation-7.40e-01

Pos logit contributions
club+1.401
cent+1.401
wives+1.373
horse+1.368
lines+1.342
Neg logit contributions
 confir-1.827
 destro-1.671
 consolidate-1.612
 rely-1.575
 consume-1.573
Token "
Token ID366
Feature activation+1.03e-02

Pos logit contributions
club+0.545
cent+0.544
wives+0.534
horse+0.531
lines+0.519
Neg logit contributions
 confir-0.847
 destro-0.779
 consolidate-0.757
 rely-0.743
 consume-0.740
TokenWe
Token ID1135
Feature activation+1.87e+00

Pos logit contributions
 confir+0.012
 destro+0.011
 consolidate+0.011
 rely+0.011
 consume+0.011
Neg logit contributions
club-0.007
cent-0.007
wives-0.007
horse-0.007
lines-0.006
Token also
Token ID635
Feature activation+3.44e+00

Pos logit contributions
 confir+1.408
 destro+1.237
 consolidate+1.171
 rely+1.131
 consume+1.128
Neg logit contributions
club-2.122
cent-2.121
wives-2.090
horse-2.084
lines-2.056
Token that
Token ID326
Feature activation+3.70e+00

Pos logit contributions
 confir+1.342
 destro+1.218
 consolidate+1.180
 rely+1.158
 consume+1.151
Neg logit contributions
club-1.188
cent-1.172
wives-1.165
horse-1.162
cart-1.136
Token are
Token ID389
Feature activation-2.77e-02

Pos logit contributions
 confir+2.551
 destro+2.216
 consolidate+2.129
 rely+2.074
 consume+2.053
Neg logit contributions
club-4.430
cent-4.398
wives-4.373
horse-4.354
cart-4.288
Token usually
Token ID3221
Feature activation+2.08e+00

Pos logit contributions
club+0.029
cent+0.029
wives+0.028
horse+0.028
cart+0.028
Neg logit contributions
 confir-0.024
 destro-0.021
 consolidate-0.020
 rely-0.020
 consume-0.020
Token employed
Token ID9322
Feature activation+1.16e-01

Pos logit contributions
 confir+1.788
 destro+1.611
 consolidate+1.541
 rely+1.512
 consume+1.500
Neg logit contributions
club-2.128
cent-2.111
wives-2.092
horse-2.083
cart-2.051
Token to
Token ID284
Feature activation+2.53e+00

Pos logit contributions
 confir+0.116
 destro+0.106
 consolidate+0.102
 rely+0.099
 consume+0.099
Neg logit contributions
club-0.102
cent-0.102
wives-0.100
horse-0.100
lines-0.098
Token mitigate
Token ID24237
Feature activation+9.22e+00

Pos logit contributions
 confir+1.600
 destro+1.369
 consolidate+1.307
 rely+1.250
 consume+1.248
Neg logit contributions
club-3.172
cent-3.163
wives-3.143
horse-3.125
cart-3.086
Token high
Token ID1029
Feature activation+2.80e+00

Pos logit contributions
 confir+8.989
 destro+8.273
 consolidate+8.035
 consume+8.010
 rely+7.979
Neg logit contributions
wives-7.620
club-7.566
horse-7.453
cent-7.354
laughs-7.313
Token bandwidth
Token ID19484
Feature activation+3.68e+00

Pos logit contributions
 confir+2.749
 destro+2.479
 consolidate+2.410
 rely+2.370
 consume+2.362
Neg logit contributions
club-2.516
wives-2.508
cent-2.495
horse-2.470
cart-2.438
Token Layer
Token ID34398
Feature activation+3.14e+00

Pos logit contributions
 confir+3.676
 destro+3.342
 consolidate+3.239
 consume+3.204
 rely+3.203
Neg logit contributions
wives-3.225
club-3.224
cent-3.178
horse-3.163
cart-3.101
Token 3
Token ID513
Feature activation-6.34e-01

Pos logit contributions
 confir+3.706
 destro+3.396
 consolidate+3.306
 consume+3.290
 rely+3.265
Neg logit contributions
cent-2.179
club-2.155
horse-2.137
wives-2.126
cart-2.061
Token &
Token ID1222
Feature activation-1.62e-01

Pos logit contributions
club+0.563
cent+0.562
wives+0.554
horse+0.551
cart+0.540
Neg logit contributions
 confir-0.635
 destro-0.576
 consolidate-0.555
 rely-0.543
 consume-0.541
TokenAlso
Token ID7583
Feature activation+1.67e+00

Pos logit contributions
 confir+1.775
 destro+1.627
 consolidate+1.618
 rely+1.593
 consume+1.588
Neg logit contributions
club-0.960
cent-0.954
wives-0.935
horse-0.929
lines-0.914
Token,
Token ID11
Feature activation+7.76e-01

Pos logit contributions
 confir+1.692
 destro+1.534
 consolidate+1.484
 rely+1.452
 consume+1.446
Neg logit contributions
cent-1.462
club-1.460
wives-1.431
horse-1.421
lines-1.398
Token Mr
Token ID1770
Feature activation-1.81e-01

Pos logit contributions
 confir+0.815
 destro+0.740
 consolidate+0.719
 rely+0.704
 consume+0.702
Neg logit contributions
club-0.651
cent-0.647
wives-0.635
horse-0.633
lines-0.620
Token.
Token ID13
Feature activation-1.97e+00

Pos logit contributions
club+0.144
cent+0.144
wives+0.141
horse+0.141
lines+0.138
Neg logit contributions
 confir-0.197
 destro-0.181
 consolidate-0.174
 rely-0.171
 consume-0.170
Token Mor
Token ID3461
Feature activation+8.74e+00

Pos logit contributions
cent+1.374
club+1.371
wives+1.342
horse+1.335
lines+1.304
Neg logit contributions
 confir-2.355
 destro-2.175
 consolidate-2.106
 rely-2.063
 consume-2.060
Tokenneau
Token ID46533
Feature activation+9.18e+00

Pos logit contributions
 confir+6.810
 consolidate+6.221
 consume+6.095
 rely+6.024
 comply+6.013
Neg logit contributions
club-8.550
wives-8.410
cent-8.361
masters-8.249
lines-8.182
Token says
Token ID1139
Feature activation-6.36e-01

Pos logit contributions
 confir+8.359
 consolidate+7.325
 destro+7.302
 rely+7.192
 consume+7.045
Neg logit contributions
club-8.329
cent-8.239
wives-8.108
masters-7.993
horse-7.980
Token he
Token ID339
Feature activation+1.71e+00

Pos logit contributions
cent+0.538
club+0.538
wives+0.528
horse+0.525
lines+0.516
Neg logit contributions
 confir-0.663
 destro-0.605
 consolidate-0.582
 rely-0.568
 consume-0.567
Token is
Token ID318
Feature activation-5.65e-01

Pos logit contributions
 confir+1.427
 destro+1.270
 consolidate+1.218
 rely+1.185
 consume+1.179
Neg logit contributions
club-1.792
cent-1.790
wives-1.762
horse-1.757
cart-1.730
Token sole
Token ID6195
Feature activation+2.00e+00

Pos logit contributions
club+0.533
cent+0.533
wives+0.524
horse+0.522
lines+0.513
Neg logit contributions
 confir-0.533
 destro-0.481
 consolidate-0.463
 rely-0.451
 consume-0.450
Token owner
Token ID4870
Feature activation-1.32e+00

Pos logit contributions
 confir+1.561
 destro+1.373
 consolidate+1.316
 rely+1.284
 consume+1.282
Neg logit contributions
club-2.208
cent-2.195
wives-2.168
horse-2.161
lines-2.141
Tokenclass
Token ID4871
Feature activation+5.80e-01

Pos logit contributions
club+0.498
cent+0.495
wives+0.486
horse+0.482
cart+0.473
Neg logit contributions
 confir-0.939
 destro-0.873
 consolidate-0.856
 rely-0.840
 consume-0.839
Token Canadians
Token ID14008
Feature activation+4.24e+00

Pos logit contributions
 confir+0.656
 destro+0.603
 consolidate+0.587
 rely+0.576
 consume+0.574
Neg logit contributions
club-0.437
cent-0.435
wives-0.426
horse-0.424
cart-0.415
Token,"
Token ID553
Feature activation-1.45e-01

Pos logit contributions
 confir+4.128
 destro+3.749
 consolidate+3.707
 rely+3.649
 consume+3.632
Neg logit contributions
club-3.770
cent-3.752
wives-3.699
horse-3.679
cart-3.623
Token Mr
Token ID1770
Feature activation-6.03e-01

Pos logit contributions
club+0.128
cent+0.128
wives+0.126
horse+0.125
lines+0.123
Neg logit contributions
 confir-0.145
 destro-0.131
 consolidate-0.127
 rely-0.124
 consume-0.124
Token.
Token ID13
Feature activation-1.40e+00

Pos logit contributions
cent+0.478
club+0.477
wives+0.467
horse+0.466
lines+0.457
Neg logit contributions
 confir-0.666
 destro-0.612
 consolidate-0.587
 rely-0.573
 consume-0.573
Token Mor
Token ID3461
Feature activation+9.15e+00

Pos logit contributions
cent+0.965
club+0.965
wives+0.942
horse+0.938
lines+0.916
Neg logit contributions
 confir-1.679
 destro-1.552
 consolidate-1.502
 rely-1.472
 consume-1.469
Tokenneau
Token ID46533
Feature activation+9.23e+00

Pos logit contributions
 confir+6.887
 consolidate+6.338
 consume+6.231
 rely+6.157
 comply+6.136
Neg logit contributions
club-9.016
wives-8.908
cent-8.765
masters-8.696
lines-8.636
Token replied
Token ID8712
Feature activation-1.71e+00

Pos logit contributions
 confir+7.976
 destro+6.969
 consolidate+6.944
 rely+6.825
 consume+6.726
Neg logit contributions
club-8.521
wives-8.411
cent-8.283
laughs-8.217
horse-8.148
Token.
Token ID13
Feature activation-7.40e-01

Pos logit contributions
club+1.401
cent+1.401
wives+1.373
horse+1.368
lines+1.342
Neg logit contributions
 confir-1.827
 destro-1.671
 consolidate-1.612
 rely-1.575
 consume-1.573
Token "
Token ID366
Feature activation+1.03e-02

Pos logit contributions
club+0.545
cent+0.544
wives+0.534
horse+0.531
lines+0.519
Neg logit contributions
 confir-0.847
 destro-0.779
 consolidate-0.757
 rely-0.743
 consume-0.740
TokenWe
Token ID1135
Feature activation+1.87e+00

Pos logit contributions
 confir+0.012
 destro+0.011
 consolidate+0.011
 rely+0.011
 consume+0.011
Neg logit contributions
club-0.007
cent-0.007
wives-0.007
horse-0.007
lines-0.006
Token raised
Token ID4376
Feature activation+2.51e+00

Pos logit contributions
 confir+1.853
 destro+1.637
 consolidate+1.585
 rely+1.553
 consume+1.542
Neg logit contributions
club-2.273
cent-2.269
wives-2.229
horse-2.216
cart-2.196
Token.
Token ID13
Feature activation-7.37e-02

Pos logit contributions
 confir+2.478
 destro+2.242
 consolidate+2.174
 rely+2.134
 consume+2.124
Neg logit contributions
club-2.252
cent-2.238
wives-2.197
horse-2.189
cart-2.148
Token
Token ID198
Feature activation+2.48e+00

Pos logit contributions
club+0.053
cent+0.053
wives+0.052
horse+0.051
lines+0.050
Neg logit contributions
 confir-0.084
 destro-0.077
 consolidate-0.076
 rely-0.075
 consume-0.074
Token
Token ID198
Feature activation+2.17e+00

Pos logit contributions
 confir+2.611
 consolidate+2.396
 rely+2.373
 consume+2.356
 destro+2.356
Neg logit contributions
club-1.905
cent-1.893
wives-1.874
horse-1.849
lines-1.807
TokenMc
Token ID9742
Feature activation+1.80e+00

Pos logit contributions
 confir+2.486
 destro+2.275
 consolidate+2.265
 rely+2.229
 consume+2.219
Neg logit contributions
club-1.486
cent-1.473
wives-1.465
horse-1.438
lines-1.408
TokenConnell
Token ID15559
Feature activation+9.08e+00

Pos logit contributions
 confir+1.606
 destro+1.427
 consolidate+1.410
 rely+1.385
 consume+1.367
Neg logit contributions
cent-1.740
wives-1.735
club-1.727
lines-1.692
horse-1.689
Token was
Token ID373
Feature activation-1.20e+00

Pos logit contributions
 confir+9.467
 destro+8.495
 consolidate+8.261
 rely+8.255
 comply+8.103
Neg logit contributions
club-7.057
wives-6.919
cent-6.916
horse-6.830
masters-6.708
Token
Token ID564
Feature activation-1.05e+00

Pos logit contributions
club+1.148
cent+1.147
wives+1.128
horse+1.124
lines+1.106
Neg logit contributions
 confir-1.123
 destro-1.013
 consolidate-0.972
 rely-0.946
 consume-0.944
Token
Token ID250
Feature activation+1.84e+00

Pos logit contributions
cent+0.668
club+0.667
wives+0.657
horse+0.649
cart+0.634
Neg logit contributions
 confir-1.318
 destro-1.211
 consolidate-1.181
 rely-1.158
 consume-1.157
Tokenune
Token ID1726
Feature activation+3.24e+00

Pos logit contributions
 confir+2.286
 destro+2.085
 consolidate+2.051
 rely+2.015
 comply+2.014
Neg logit contributions
club-1.118
wives-1.115
horse-1.099
cent-1.095
lines-1.063
Tokenqu
Token ID421
Feature activation+2.58e+00

Pos logit contributions
 confir+3.800
 destro+3.470
 consolidate+3.428
 rely+3.376
 consume+3.368
Neg logit contributions
club-2.230
wives-2.205
horse-2.132
lines-2.120
masters-2.097
Token Ottawa
Token ID14074
Feature activation+2.16e+00

Pos logit contributions
club+0.424
cent+0.420
wives+0.414
horse+0.410
cart+0.404
Neg logit contributions
 confir-0.595
 destro-0.545
 consolidate-0.530
 rely-0.519
 consume-0.517
Token,
Token ID11
Feature activation+1.46e+00

Pos logit contributions
 confir+2.170
 destro+1.955
 consolidate+1.915
 rely+1.877
 consume+1.849
Neg logit contributions
club-1.898
cent-1.873
wives-1.848
horse-1.824
cart-1.812
Token Mr
Token ID1770
Feature activation+8.88e-02

Pos logit contributions
 confir+1.618
 destro+1.476
 consolidate+1.440
 rely+1.411
 consume+1.407
Neg logit contributions
club-1.136
cent-1.129
wives-1.106
horse-1.103
cart-1.082
Token.
Token ID13
Feature activation-1.72e+00

Pos logit contributions
 confir+0.097
 destro+0.089
 consolidate+0.086
 rely+0.084
 consume+0.083
Neg logit contributions
club-0.071
cent-0.071
wives-0.070
horse-0.069
lines-0.068
Token Mor
Token ID3461
Feature activation+9.41e+00

Pos logit contributions
cent+1.177
club+1.174
wives+1.149
horse+1.144
lines+1.115
Neg logit contributions
 confir-2.073
 destro-1.916
 consolidate-1.855
 rely-1.818
 consume-1.814
Tokenneau
Token ID46533
Feature activation+9.02e+00

Pos logit contributions
 confir+7.165
 consolidate+6.602
 consume+6.471
 rely+6.387
 comply+6.376
Neg logit contributions
club-9.190
wives-9.089
cent-8.966
masters-8.912
lines-8.816
Token also
Token ID635
Feature activation+2.90e+00

Pos logit contributions
 confir+8.312
 consolidate+7.496
 destro+7.414
 rely+7.334
 consume+7.215
Neg logit contributions
club-7.936
cent-7.803
wives-7.770
laughs-7.681
cart-7.624
Token jointly
Token ID26913
Feature activation+2.89e-01

Pos logit contributions
 confir+2.400
 destro+2.120
 consolidate+2.059
 rely+2.005
 consume+1.985
Neg logit contributions
club-3.065
cent-3.055
wives-3.008
horse-2.997
cart-2.951
Token owns
Token ID12216
Feature activation-1.02e+00

Pos logit contributions
 confir+0.247
 destro+0.220
 consolidate+0.211
 rely+0.205
 consume+0.204
Neg logit contributions
club-0.299
cent-0.298
wives-0.294
horse-0.293
lines-0.288
Token four
Token ID1440
Feature activation-1.53e+00

Pos logit contributions
club+0.770
cent+0.769
wives+0.753
horse+0.749
lines+0.732
Neg logit contributions
 confir-1.165
 destro-1.068
 consolidate-1.038
 rely-1.017
 consume-1.015
Token numbered
Token ID25840
Feature activation-1.19e+00

Pos logit contributions
club+1.350
cent+1.349
wives+1.325
horse+1.320
lines+1.296
Neg logit contributions
 confir-1.532
 destro-1.392
 consolidate-1.340
 rely-1.308
 consume-1.305
Token deal
Token ID1730
Feature activation+1.54e+00

Pos logit contributions
 confir+2.408
 destro+2.118
 consolidate+2.046
 rely+1.998
 comply+1.993
Neg logit contributions
club-2.986
cent-2.955
wives-2.918
horse-2.913
lines-2.855
Token to
Token ID284
Feature activation+2.48e+00

Pos logit contributions
 confir+1.570
 destro+1.422
 consolidate+1.376
 rely+1.346
 consume+1.340
Neg logit contributions
club-1.347
cent-1.343
wives-1.318
horse-1.313
lines-1.290
Token save
Token ID3613
Feature activation+4.14e+00

Pos logit contributions
 confir+1.687
 destro+1.455
 consolidate+1.386
 rely+1.335
 consume+1.326
Neg logit contributions
club-3.001
cent-2.996
wives-2.960
horse-2.950
lines-2.912
Token the
Token ID262
Feature activation+1.98e+00

Pos logit contributions
 confir+4.261
 destro+3.822
 consolidate+3.780
 rely+3.747
 comply+3.711
Neg logit contributions
club-3.452
cent-3.434
horse-3.372
wives-3.356
cart-3.288
Token Def
Token ID2896
Feature activation+2.61e+00

Pos logit contributions
 confir+1.988
 destro+1.785
 consolidate+1.749
 rely+1.727
 comply+1.712
Neg logit contributions
club-1.730
cent-1.729
horse-1.706
wives-1.698
cart-1.663
Tokenerred
Token ID17436
Feature activation+9.01e+00

Pos logit contributions
 confir+1.972
 destro+1.708
 consolidate+1.655
 rely+1.625
 comply+1.606
Neg logit contributions
club-2.898
cent-2.880
wives-2.879
horse-2.850
lines-2.795
Token Action
Token ID7561
Feature activation+5.20e+00

Pos logit contributions
 confir+7.678
 consolidate+6.587
 rely+6.573
 destro+6.510
 comply+6.444
Neg logit contributions
wives-8.665
club-8.663
cent-8.581
horse-8.520
laughs-8.402
Token for
Token ID329
Feature activation+2.68e+00

Pos logit contributions
 confir+4.655
 destro+4.078
 consolidate+4.059
 rely+3.983
 consume+3.925
Neg logit contributions
club-5.010
cent-4.942
wives-4.940
horse-4.924
cart-4.884
Token Childhood
Token ID40992
Feature activation+1.74e+00

Pos logit contributions
 confir+2.945
 destro+2.655
 consolidate+2.593
 rely+2.570
 consume+2.554
Neg logit contributions
club-2.092
cent-2.063
horse-2.049
wives-2.025
cart-2.012
Token Arri
Token ID27350
Feature activation+3.56e+00

Pos logit contributions
 confir+2.670
 destro+2.487
 consolidate+2.448
 rely+2.425
 consume+2.411
Neg logit contributions
club-0.614
cent-0.611
wives-0.591
horse-0.590
cart-0.556
Tokenvals
Token ID12786
Feature activation+2.31e+00

Pos logit contributions
 confir+2.805
 consolidate+2.450
 destro+2.443
 rely+2.431
 consume+2.416
Neg logit contributions
club-3.644
horse-3.635
cent-3.615
wives-3.599
lines-3.543
Token.
Token ID13
Feature activation+1.19e+00

Pos logit contributions
 confir+1.950
 destro+1.765
 consolidate+1.746
 rely+1.719
 consume+1.711
Neg logit contributions
club-1.571
cent-1.563
wives-1.540
horse-1.529
lines-1.497
TokenS
Token ID50
Feature activation+3.07e+00

Pos logit contributions
 confir+1.513
 destro+1.397
 consolidate+1.389
 rely+1.377
 consume+1.368
Neg logit contributions
club-0.682
cent-0.677
wives-0.667
horse-0.659
lines-0.638
Token.
Token ID13
Feature activation+1.31e+00

Pos logit contributions
 confir+3.189
 destro+2.901
 consolidate+2.795
 rely+2.758
 consume+2.745
Neg logit contributions
club-2.584
cent-2.560
wives-2.533
horse-2.522
lines-2.484
Token Ambassador
Token ID20656
Feature activation+2.45e+00

Pos logit contributions
 confir+1.239
 destro+1.116
 consolidate+1.075
 rely+1.050
 consume+1.047
Neg logit contributions
club-1.234
cent-1.228
wives-1.214
horse-1.207
lines-1.188
Token to
Token ID284
Feature activation+1.33e+00

Pos logit contributions
 confir+2.219
 destro+1.993
 consolidate+1.930
 rely+1.880
 consume+1.860
Neg logit contributions
club-2.376
cent-2.363
wives-2.337
horse-2.328
cart-2.282
Token Haiti
Token ID25051
Feature activation+8.99e+00

Pos logit contributions
 confir+1.243
 destro+1.118
 consolidate+1.087
 rely+1.057
 consume+1.051
Neg logit contributions
club-1.261
cent-1.256
wives-1.240
horse-1.234
lines-1.210
Token,
Token ID11
Feature activation+2.67e+00

Pos logit contributions
 confir+8.245
 destro+7.209
 consolidate+7.146
 rely+7.049
 comply+6.953
Neg logit contributions
club-8.121
horse-7.975
cent-7.972
wives-7.837
haw-7.747
Token once
Token ID1752
Feature activation+1.41e+00

Pos logit contributions
 confir+2.473
 destro+2.193
 consolidate+2.123
 rely+2.090
 consume+2.057
Neg logit contributions
club-2.577
cent-2.556
wives-2.522
horse-2.511
cart-2.470
Token dubbed
Token ID17494
Feature activation-9.09e-01

Pos logit contributions
 confir+1.343
 destro+1.206
 consolidate+1.163
 rely+1.142
 consume+1.134
Neg logit contributions
club-1.321
cent-1.318
wives-1.304
horse-1.290
cart-1.271
Token the
Token ID262
Feature activation+5.24e-01

Pos logit contributions
club+0.739
cent+0.739
wives+0.724
horse+0.721
lines+0.707
Neg logit contributions
 confir-0.978
 destro-0.894
 consolidate-0.864
 rely-0.844
 consume-0.843
Token U
Token ID471
Feature activation+2.85e+00

Pos logit contributions
 confir+0.504
 destro+0.456
 consolidate+0.439
 rely+0.428
 consume+0.427
Neg logit contributions
club-0.485
cent-0.484
wives-0.477
horse-0.475
lines-0.466
Token millionaire
Token ID45172
Feature activation+6.75e-01

Pos logit contributions
club+0.269
cent+0.269
wives+0.265
horse+0.264
lines+0.259
Neg logit contributions
 confir-0.270
 destro-0.244
 consolidate-0.234
 rely-0.228
 consume-0.228
Token owners
Token ID4393
Feature activation-1.92e+00

Pos logit contributions
 confir+0.683
 destro+0.621
 consolidate+0.601
 rely+0.589
 consume+0.587
Neg logit contributions
club-0.589
cent-0.588
wives-0.577
horse-0.575
cart-0.566
Token and
Token ID290
Feature activation+1.09e+00

Pos logit contributions
cent+1.886
club+1.886
wives+1.858
horse+1.846
lines+1.819
Neg logit contributions
 confir-1.733
 destro-1.565
 consolidate-1.467
 rely-1.415
 consume-1.413
Token shareholders
Token ID19195
Feature activation+1.32e+00

Pos logit contributions
 confir+1.071
 destro+0.971
 consolidate+0.940
 rely+0.916
 consume+0.914
Neg logit contributions
club-0.992
cent-0.989
wives-0.972
horse-0.971
lines-0.953
Token of
Token ID286
Feature activation-1.56e+00

Pos logit contributions
 confir+1.264
 destro+1.141
 consolidate+1.104
 rely+1.079
 consume+1.076
Neg logit contributions
club-1.227
cent-1.225
wives-1.205
horse-1.201
cart-1.180
Token Mor
Token ID3461
Feature activation+8.94e+00

Pos logit contributions
cent+1.281
club+1.281
wives+1.256
horse+1.250
lines+1.226
Neg logit contributions
 confir-1.671
 destro-1.528
 consolidate-1.474
 rely-1.440
 consume-1.437
Tokenneau
Token ID46533
Feature activation+8.94e+00

Pos logit contributions
 confir+7.190
 consolidate+6.518
 consume+6.401
 rely+6.401
 comply+6.309
Neg logit contributions
club-8.412
wives-8.297
cent-8.264
laughs-8.157
lines-8.140
Token She
Token ID1375
Feature activation+2.30e+00

Pos logit contributions
 confir+10.537
 destro+9.625
 consolidate+9.585
 rely+9.467
 consume+9.339
Neg logit contributions
club-5.586
wives-5.508
cent-5.481
lines-5.423
horse-5.385
Tokenpell
Token ID23506
Feature activation+4.74e+00

Pos logit contributions
 confir+1.800
 destro+1.585
 consolidate+1.530
 rely+1.485
 consume+1.482
Neg logit contributions
club-2.501
cent-2.495
wives-2.464
horse-2.455
lines-2.423
Token will
Token ID481
Feature activation+2.37e+00

Pos logit contributions
 confir+3.873
 destro+3.417
 consolidate+3.371
 rely+3.319
 consume+3.281
Neg logit contributions
club-4.947
cent-4.944
horse-4.874
wives-4.857
cart-4.807
Token not
Token ID407
Feature activation+5.23e+00

Pos logit contributions
 confir+1.666
 destro+1.446
 consolidate+1.381
 rely+1.334
 consume+1.329
Neg logit contributions
club-2.794
cent-2.793
wives-2.758
horse-2.751
cart-2.712
Token owners
Token ID4393
Feature activation-1.92e+00

Pos logit contributions
 confir+0.683
 destro+0.621
 consolidate+0.601
 rely+0.589
 consume+0.587
Neg logit contributions
club-0.589
cent-0.588
wives-0.577
horse-0.575
cart-0.566
Token and
Token ID290
Feature activation+1.09e+00

Pos logit contributions
cent+1.886
club+1.886
wives+1.858
horse+1.846
lines+1.819
Neg logit contributions
 confir-1.733
 destro-1.565
 consolidate-1.467
 rely-1.415
 consume-1.413
Token shareholders
Token ID19195
Feature activation+1.32e+00

Pos logit contributions
 confir+1.071
 destro+0.971
 consolidate+0.940
 rely+0.916
 consume+0.914
Neg logit contributions
club-0.992
cent-0.989
wives-0.972
horse-0.971
lines-0.953
Token of
Token ID286
Feature activation-1.56e+00

Pos logit contributions
 confir+1.264
 destro+1.141
 consolidate+1.104
 rely+1.079
 consume+1.076
Neg logit contributions
club-1.227
cent-1.225
wives-1.205
horse-1.201
cart-1.180
Token Mor
Token ID3461
Feature activation+8.94e+00

Pos logit contributions
cent+1.281
club+1.281
wives+1.256
horse+1.250
lines+1.226
Neg logit contributions
 confir-1.671
 destro-1.528
 consolidate-1.474
 rely-1.440
 consume-1.437
Tokenneau
Token ID46533
Feature activation+8.94e+00

Pos logit contributions
 confir+7.190
 consolidate+6.518
 consume+6.401
 rely+6.401
 comply+6.309
Neg logit contributions
club-8.412
wives-8.297
cent-8.264
laughs-8.157
lines-8.140
Token She
Token ID1375
Feature activation+2.30e+00

Pos logit contributions
 confir+10.537
 destro+9.625
 consolidate+9.585
 rely+9.467
 consume+9.339
Neg logit contributions
club-5.586
wives-5.508
cent-5.481
lines-5.423
horse-5.385
Tokenpell
Token ID23506
Feature activation+4.74e+00

Pos logit contributions
 confir+1.800
 destro+1.585
 consolidate+1.530
 rely+1.485
 consume+1.482
Neg logit contributions
club-2.501
cent-2.495
wives-2.464
horse-2.455
lines-2.423
Token will
Token ID481
Feature activation+2.37e+00

Pos logit contributions
 confir+3.873
 destro+3.417
 consolidate+3.371
 rely+3.319
 consume+3.281
Neg logit contributions
club-4.947
cent-4.944
horse-4.874
wives-4.857
cart-4.807
Token not
Token ID407
Feature activation+5.23e+00

Pos logit contributions
 confir+1.666
 destro+1.446
 consolidate+1.381
 rely+1.334
 consume+1.329
Neg logit contributions
club-2.794
cent-2.793
wives-2.758
horse-2.751
cart-2.712
Token have
Token ID423
Feature activation+2.07e+00

Pos logit contributions
 confir+4.199
 destro+3.688
 consolidate+3.623
 rely+3.534
 consume+3.525
Neg logit contributions
club-5.598
cent-5.593
wives-5.514
horse-5.513
cart-5.431
Token too
Token ID1165
Feature activation+1.21e+00

Pos logit contributions
club+2.322
cent+2.322
wives+2.269
horse+2.257
lines+2.214
Neg logit contributions
 confir-3.669
 destro-3.377
 consolidate-3.266
 rely-3.196
 consume-3.193
Token many
Token ID867
Feature activation+3.73e-01

Pos logit contributions
 confir+1.188
 destro+1.083
 consolidate+1.045
 rely+1.023
 consume+1.018
Neg logit contributions
club-1.077
cent-1.067
wives-1.058
horse-1.053
cart-1.036
Token times
Token ID1661
Feature activation-1.34e+00

Pos logit contributions
 confir+0.430
 destro+0.397
 consolidate+0.385
 rely+0.380
 consume+0.378
Neg logit contributions
club-0.270
cent-0.269
wives-0.264
horse-0.264
cart-0.259
Token.
Token ID13
Feature activation-2.24e+00

Pos logit contributions
club+1.116
cent+1.114
wives+1.095
horse+1.089
lines+1.068
Neg logit contributions
 confir-1.406
 destro-1.285
 consolidate-1.242
 rely-1.216
 consume-1.213
Token The
Token ID383
Feature activation-3.45e+00

Pos logit contributions
cent+1.302
club+1.298
wives+1.273
horse+1.252
lines+1.220
Neg logit contributions
 confir-2.890
 destro-2.686
 consolidate-2.631
 rely-2.591
 consume-2.582
Token drummer
Token ID34269
Feature activation-9.79e+00

Pos logit contributions
club+2.567
cent+2.563
wives+2.509
horse+2.496
lines+2.449
Neg logit contributions
 confir-3.932
 destro-3.615
 consolidate-3.501
 rely-3.424
 consume-3.420
Token looked
Token ID3114
Feature activation-2.89e+00

Pos logit contributions
cent+8.864
club+8.706
horse+8.567
wives+8.552
lines+8.460
Neg logit contributions
 confir-9.054
 destro-8.138
 consolidate-7.830
 consume-7.544
 deprive-7.494
Token like
Token ID588
Feature activation+3.00e-01

Pos logit contributions
club+2.364
cent+2.362
wives+2.313
horse+2.308
lines+2.264
Neg logit contributions
 confir-3.087
 destro-2.817
 consolidate-2.721
 rely-2.658
 consume-2.652
Token Tim
Token ID5045
Feature activation-2.87e+00

Pos logit contributions
 confir+0.298
 destro+0.272
 consolidate+0.261
 rely+0.256
 consume+0.255
Neg logit contributions
club-0.266
cent-0.266
wives-0.262
horse-0.260
lines-0.256
Token Y
Token ID575
Feature activation-2.74e+00

Pos logit contributions
cent+1.955
club+1.949
horse+1.897
wives+1.893
lines+1.856
Neg logit contributions
 confir-3.469
 destro-3.193
 consolidate-3.091
 consume-3.030
 rely-3.030
Tokenoh
Token ID1219
Feature activation-2.22e-01

Pos logit contributions
cent+2.025
club+2.024
wives+1.978
horse+1.970
lines+1.929
Neg logit contributions
 confir-3.161
 destro-2.909
 consolidate-2.813
 rely-2.754
 consume-2.749
Token guitar
Token ID10047
Feature activation-6.35e+00

Pos logit contributions
cent+6.784
club+6.775
horse+6.652
wives+6.572
lines+6.518
Neg logit contributions
 confir-8.368
 destro-7.574
 consolidate-7.297
 consume-7.090
 rely-7.068
Token.
Token ID13
Feature activation-3.76e+00

Pos logit contributions
club+5.032
cent+5.022
wives+4.907
horse+4.904
lines+4.866
Neg logit contributions
 confir-6.871
 destro-6.263
 consolidate-6.052
 consume-5.905
 rely-5.877
Token Then
Token ID3244
Feature activation-1.86e-01

Pos logit contributions
club+2.187
cent+2.185
wives+2.128
horse+2.112
lines+2.055
Neg logit contributions
 confir-4.903
 destro-4.558
 consolidate-4.437
 rely-4.359
 consume-4.351
Token,
Token ID11
Feature activation-1.95e+00

Pos logit contributions
cent+0.174
club+0.174
wives+0.171
horse+0.170
lines+0.167
Neg logit contributions
 confir-0.177
 destro-0.160
 consolidate-0.154
 rely-0.150
 consume-0.150
Token a
Token ID257
Feature activation-5.24e+00

Pos logit contributions
club+1.802
cent+1.801
wives+1.771
horse+1.762
lines+1.733
Neg logit contributions
 confir-1.879
 destro-1.701
 consolidate-1.637
 rely-1.596
 consume-1.592
Token singer
Token ID14015
Feature activation-9.51e+00

Pos logit contributions
club+4.173
cent+4.164
horse+4.064
wives+4.049
cart+3.963
Neg logit contributions
 confir-5.659
 destro-5.159
 consolidate-5.001
 rely-4.884
 consume-4.880
Token with
Token ID351
Feature activation-3.74e+00

Pos logit contributions
cent+7.786
club+7.731
horse+7.605
wives+7.588
lines+7.448
Neg logit contributions
 confir-9.666
 destro-8.825
 consolidate-8.551
 consume-8.242
 rely-8.230
Token a
Token ID257
Feature activation-3.91e+00

Pos logit contributions
club+3.021
cent+3.013
wives+2.949
horse+2.945
lines+2.892
Neg logit contributions
 confir-4.015
 destro-3.675
 consolidate-3.547
 rely-3.464
 consume-3.459
Token bal
Token ID3652
Feature activation-2.95e+00

Pos logit contributions
club+3.404
cent+3.387
horse+3.325
wives+3.312
cart+3.255
Neg logit contributions
 confir-3.940
 destro-3.577
 consolidate-3.459
 rely-3.370
 consume-3.359
Tokenacl
Token ID37779
Feature activation-2.97e+00

Pos logit contributions
club+2.328
cent+2.325
horse+2.275
wives+2.263
cart+2.221
Neg logit contributions
 confir-3.260
 destro-2.970
 consolidate-2.861
 rely-2.804
 consume-2.801
Tokenava
Token ID4170
Feature activation-2.01e+00

Pos logit contributions
club+2.390
cent+2.378
horse+2.330
wives+2.310
cart+2.265
Neg logit contributions
 confir-3.240
 destro-2.963
 consolidate-2.858
 rely-2.791
 consume-2.776
Token spot
Token ID4136
Feature activation-3.41e+00

Pos logit contributions
cent+2.063
club+2.062
wives+2.023
horse+2.016
lines+1.982
Neg logit contributions
 confir-2.390
 destro-2.174
 consolidate-2.093
 rely-2.040
 consume-2.035
Token,
Token ID11
Feature activation-9.36e-01

Pos logit contributions
cent+2.783
club+2.778
wives+2.720
horse+2.711
lines+2.670
Neg logit contributions
 confir-3.644
 destro-3.329
 consolidate-3.215
 rely-3.135
 consume-3.125
Token implying
Token ID30253
Feature activation-3.25e+00

Pos logit contributions
club+0.861
cent+0.860
wives+0.845
horse+0.842
lines+0.827
Neg logit contributions
 confir-0.907
 destro-0.821
 consolidate-0.790
 rely-0.770
 consume-0.769
Token that
Token ID326
Feature activation+8.82e-01

Pos logit contributions
club+2.853
cent+2.847
wives+2.807
horse+2.797
lines+2.755
Neg logit contributions
 confir-3.233
 destro-2.935
 consolidate-2.828
 rely-2.733
 consume-2.732
Token the
Token ID262
Feature activation-2.13e+00

Pos logit contributions
 confir+0.828
 destro+0.747
 consolidate+0.718
 rely+0.700
 consume+0.698
Neg logit contributions
club-0.837
cent-0.836
wives-0.821
horse-0.819
lines-0.805
Token restaurant
Token ID7072
Feature activation-8.95e+00

Pos logit contributions
cent+1.963
club+1.960
wives+1.932
horse+1.923
lines+1.892
Neg logit contributions
 confir-2.045
 destro-1.850
 consolidate-1.780
 rely-1.730
 consume-1.725
Token was
Token ID373
Feature activation-4.35e+00

Pos logit contributions
club+8.477
wives+8.425
cent+8.385
horse+8.250
lines+8.197
Neg logit contributions
 confir-7.836
 destro-7.078
 consolidate-6.747
 consume-6.383
 rely-6.327
Token boiling
Token ID24372
Feature activation-2.61e+00

Pos logit contributions
cent+4.189
club+4.187
wives+4.116
horse+4.086
lines+4.038
Neg logit contributions
 confir-3.981
 destro-3.575
 consolidate-3.433
 rely-3.316
 consume-3.297
Token a
Token ID257
Feature activation-3.51e+00

Pos logit contributions
cent+2.169
club+2.168
wives+2.132
horse+2.120
lines+2.084
Neg logit contributions
 confir-2.747
 destro-2.507
 consolidate-2.417
 rely-2.354
 consume-2.347
Token soup
Token ID17141
Feature activation-4.27e+00

Pos logit contributions
cent+3.087
club+3.078
horse+3.025
wives+3.019
cart+2.961
Neg logit contributions
 confir-3.516
 destro-3.194
 consolidate-3.084
 rely-3.000
 consume-2.989
Token pot
Token ID1787
Feature activation-3.13e+00

Pos logit contributions
cent+3.790
club+3.766
horse+3.713
wives+3.705
cart+3.636
Neg logit contributions
 confir-4.269
 destro-3.877
 consolidate-3.727
 rely-3.619
 consume-3.616
Token to
Token ID284
Feature activation-2.11e-01

Pos logit contributions
club+5.153
cent+5.132
wives+5.020
horse+5.011
lines+4.950
Neg logit contributions
 confir-5.867
 destro-5.340
 consolidate-5.149
 rely-5.007
 consume-4.984
Token the
Token ID262
Feature activation-2.65e-01

Pos logit contributions
club+0.188
cent+0.188
wives+0.185
horse+0.184
lines+0.181
Neg logit contributions
 confir-0.210
 destro-0.190
 consolidate-0.184
 rely-0.179
 consume-0.179
Token bedroom
Token ID14043
Feature activation-4.37e+00

Pos logit contributions
club+0.232
cent+0.232
wives+0.228
horse+0.227
lines+0.223
Neg logit contributions
 confir-0.268
 destro-0.245
 consolidate-0.235
 rely-0.230
 consume-0.229
Token from
Token ID422
Feature activation-2.57e+00

Pos logit contributions
club+3.653
cent+3.648
wives+3.583
horse+3.557
lines+3.512
Neg logit contributions
 confir-4.543
 destro-4.158
 consolidate-4.001
 rely-3.898
 consume-3.895
Token the
Token ID262
Feature activation-5.84e-01

Pos logit contributions
club+2.077
cent+2.075
wives+2.035
horse+2.025
lines+1.987
Neg logit contributions
 confir-2.780
 destro-2.543
 consolidate-2.457
 rely-2.402
 consume-2.397
Token kitchen
Token ID9592
Feature activation-8.86e+00

Pos logit contributions
club+0.500
cent+0.500
wives+0.490
horse+0.489
lines+0.479
Neg logit contributions
 confir-0.602
 destro-0.549
 consolidate-0.528
 rely-0.517
 consume-0.516
Token,
Token ID11
Feature activation-1.63e+00

Pos logit contributions
club+7.417
cent+7.393
wives+7.321
horse+7.266
lines+7.199
Neg logit contributions
 confir-8.830
 destro-8.079
 consolidate-7.780
 consume-7.486
 rely-7.481
Token with
Token ID351
Feature activation-3.33e+00

Pos logit contributions
club+1.467
cent+1.466
wives+1.441
horse+1.435
lines+1.410
Neg logit contributions
 confir-1.603
 destro-1.454
 consolidate-1.400
 rely-1.365
 consume-1.362
Token the
Token ID262
Feature activation-1.30e+00

Pos logit contributions
club+2.659
cent+2.657
wives+2.605
horse+2.590
lines+2.548
Neg logit contributions
 confir-3.626
 destro-3.326
 consolidate-3.208
 rely-3.135
 consume-3.129
Token toilet
Token ID16146
Feature activation-2.44e+00

Pos logit contributions
club+1.138
cent+1.136
wives+1.115
horse+1.111
lines+1.090
Neg logit contributions
 confir-1.325
 destro-1.204
 consolidate-1.161
 rely-1.135
 consume-1.132
Token being
Token ID852
Feature activation-2.90e+00

Pos logit contributions
club+2.172
cent+2.171
wives+2.132
horse+2.123
lines+2.087
Neg logit contributions
 confir-2.435
 destro-2.210
 consolidate-2.128
 rely-2.076
 consume-2.071
Token sites
Token ID5043
Feature activation-4.74e+00

Pos logit contributions
cent+2.470
club+2.462
wives+2.418
horse+2.408
lines+2.376
Neg logit contributions
 confir-2.679
 destro-2.425
 consolidate-2.333
 rely-2.270
 consume-2.263
Token,
Token ID11
Feature activation+7.72e-01

Pos logit contributions
cent+3.995
club+3.959
wives+3.901
horse+3.892
lines+3.830
Neg logit contributions
 confir-4.921
 destro-4.496
 consolidate-4.305
 consume-4.197
 rely-4.187
Token many
Token ID867
Feature activation-1.58e+00

Pos logit contributions
 confir+0.797
 destro+0.726
 consolidate+0.705
 rely+0.692
 consume+0.688
Neg logit contributions
club-0.661
cent-0.654
wives-0.646
horse-0.642
cart-0.630
Token of
Token ID286
Feature activation-2.74e+00

Pos logit contributions
cent+1.413
club+1.411
wives+1.385
horse+1.381
lines+1.358
Neg logit contributions
 confir-1.577
 destro-1.432
 consolidate-1.376
 consume-1.339
 rely-1.339
Token the
Token ID262
Feature activation-1.30e+00

Pos logit contributions
club+2.058
cent+2.057
wives+2.013
horse+2.004
lines+1.962
Neg logit contributions
 confir-3.119
 destro-2.866
 consolidate-2.774
 rely-2.715
 consume-2.711
Token scenes
Token ID8188
Feature activation-8.57e+00

Pos logit contributions
club+1.146
cent+1.144
wives+1.124
horse+1.120
lines+1.099
Neg logit contributions
 confir-1.301
 destro-1.183
 consolidate-1.139
 rely-1.112
 consume-1.109
Token look
Token ID804
Feature activation-3.93e+00

Pos logit contributions
cent+9.103
club+8.876
lines+8.797
horse+8.783
wives+8.766
Neg logit contributions
 confir-6.651
 destro-5.940
 consolidate-5.438
 deprive-5.151
uyomi-5.146
Token familiar
Token ID5385
Feature activation-1.11e+00

Pos logit contributions
cent+3.623
club+3.611
horse+3.547
wives+3.542
lines+3.500
Neg logit contributions
 confir-3.775
 destro-3.420
 consolidate-3.271
 rely-3.180
 consume-3.173
Token to
Token ID284
Feature activation+1.66e-02

Pos logit contributions
club+0.942
cent+0.942
wives+0.924
horse+0.920
lines+0.903
Neg logit contributions
 confir-1.149
 destro-1.048
 consolidate-1.010
 rely-0.986
 consume-0.984
Token today
Token ID1909
Feature activation+1.26e+00

Pos logit contributions
 confir+0.016
 destro+0.015
 consolidate+0.014
 rely+0.014
 consume+0.014
Neg logit contributions
club-0.015
cent-0.015
wives-0.015
horse-0.015
lines-0.014
Token's
Token ID338
Feature activation-2.22e+00

Pos logit contributions
 confir+1.063
 destro+0.952
 consolidate+0.911
 rely+0.891
 consume+0.888
Neg logit contributions
club-1.294
cent-1.289
wives-1.273
horse-1.267
lines-1.249
Token a
Token ID257
Feature activation-3.56e+00

Pos logit contributions
 confir+2.197
 destro+1.995
 consolidate+1.940
 rely+1.908
 consume+1.899
Neg logit contributions
club-1.902
cent-1.892
wives-1.864
horse-1.850
cart-1.813
Token lot
Token ID1256
Feature activation-1.52e+00

Pos logit contributions
cent+2.886
club+2.879
horse+2.824
wives+2.820
lines+2.762
Neg logit contributions
 confir-3.816
 destro-3.489
 consolidate-3.365
 rely-3.284
 consume-3.279
Token happens
Token ID4325
Feature activation-4.25e-01

Pos logit contributions
cent+1.726
club+1.723
wives+1.703
horse+1.697
lines+1.677
Neg logit contributions
 confir-1.149
 destro-1.008
 consolidate-0.955
 consume-0.917
 rely-0.917
Token:
Token ID25
Feature activation-4.43e+00

Pos logit contributions
club+0.356
cent+0.355
wives+0.349
horse+0.347
lines+0.341
Neg logit contributions
 confir-0.445
 destro-0.406
 consolidate-0.393
 rely-0.385
 consume-0.384
Token the
Token ID262
Feature activation-2.06e+00

Pos logit contributions
club+3.745
cent+3.738
wives+3.671
horse+3.664
lines+3.607
Neg logit contributions
 confir-4.555
 destro-4.165
 consolidate-3.981
 rely-3.882
 consume-3.876
Token characters
Token ID3435
Feature activation-8.43e+00

Pos logit contributions
club+1.630
cent+1.629
wives+1.597
horse+1.590
lines+1.559
Neg logit contributions
 confir-2.258
 destro-2.069
 consolidate-1.999
 rely-1.955
 consume-1.952
Token go
Token ID467
Feature activation-3.85e+00

Pos logit contributions
cent+9.232
club+9.200
wives+9.093
lines+9.042
horse+8.998
Neg logit contributions
 confir-6.164
 destro-5.480
 consolidate-4.869
enei-4.661
 deprive-4.623
Token to
Token ID284
Feature activation+1.73e-01

Pos logit contributions
club+2.997
cent+2.988
wives+2.923
horse+2.915
lines+2.870
Neg logit contributions
 confir-4.252
 destro-3.904
 consolidate-3.753
 rely-3.656
 consume-3.653
Token a
Token ID257
Feature activation-2.46e+00

Pos logit contributions
 confir+0.178
 destro+0.162
 consolidate+0.157
 consume+0.153
 rely+0.153
Neg logit contributions
cent-0.148
club-0.148
wives-0.146
horse-0.145
cart-0.142
Token lot
Token ID1256
Feature activation+8.85e-01

Pos logit contributions
club+2.167
cent+2.162
wives+2.121
horse+2.114
lines+2.076
Neg logit contributions
 confir-2.483
 destro-2.255
 consolidate-2.172
 rely-2.121
 consume-2.115
Token of
Token ID286
Feature activation-2.64e+00

Pos logit contributions
 confir+1.029
 destro+0.950
 consolidate+0.922
 rely+0.905
 consume+0.903
Neg logit contributions
cent-0.632
club-0.629
wives-0.620
horse-0.619
lines-0.600
Token the
Token ID262
Feature activation-1.58e+00

Pos logit contributions
club+2.058
cent+2.056
wives+2.015
horse+2.006
lines+1.965
Neg logit contributions
 confir-2.881
 destro-2.641
 consolidate-2.553
 rely-2.498
 consume-2.493
Token gratification
Token ID47459
Feature activation-1.22e+00

Pos logit contributions
club+1.513
cent+1.512
wives+1.486
horse+1.481
lines+1.458
Neg logit contributions
 confir-1.472
 destro-1.327
 consolidate-1.273
 rely-1.239
 consume-1.236
Token spectrum
Token ID10958
Feature activation-1.37e+00

Pos logit contributions
club+1.257
cent+1.256
wives+1.237
horse+1.233
lines+1.214
Neg logit contributions
 confir-1.052
 destro-0.939
 consolidate-0.898
 rely-0.872
 consume-0.870
Token.
Token ID13
Feature activation-2.79e+00

Pos logit contributions
club+1.132
cent+1.130
wives+1.109
horse+1.104
lines+1.083
Neg logit contributions
 confir-1.452
 destro-1.327
 consolidate-1.281
 rely-1.253
 consume-1.251
Token The
Token ID383
Feature activation-4.30e+00

Pos logit contributions
club+1.564
cent+1.561
wives+1.518
horse+1.508
lines+1.463
Neg logit contributions
 confir-3.688
 destro-3.435
 consolidate-3.350
 rely-3.296
 consume-3.286
Token characters
Token ID3435
Feature activation-8.38e+00

Pos logit contributions
cent+3.860
club+3.849
wives+3.794
horse+3.770
lines+3.720
Neg logit contributions
 confir-4.211
 destro-3.827
 consolidate-3.673
 consume-3.575
 rely-3.566
Token
Token ID447
Feature activation-4.37e+00

Pos logit contributions
cent+8.988
club+8.848
wives+8.791
horse+8.675
lines+8.633
Neg logit contributions
 confir-6.418
 destro-5.688
 consolidate-5.274
 consume-4.992
 deprive-4.958
Token
Token ID247
Feature activation-1.44e+00

Pos logit contributions
club+2.393
cent+2.379
horse+2.306
wives+2.300
lines+2.236
Neg logit contributions
 confir-5.865
 destro-5.521
 consolidate-5.311
 rely-5.168
 consume-5.167
Token carefully
Token ID7773
Feature activation-1.54e+00

Pos logit contributions
club+0.832
cent+0.829
wives+0.808
horse+0.805
lines+0.780
Neg logit contributions
 confir-1.890
 destro-1.758
 consolidate-1.710
 rely-1.684
 consume-1.680
Token controlled
Token ID6856
Feature activation-2.39e+00

Pos logit contributions
cent+1.560
club+1.559
wives+1.533
horse+1.527
lines+1.505
Neg logit contributions
 confir-1.352
 destro-1.210
 consolidate-1.154
 rely-1.120
 consume-1.119
Token inter
Token ID987
Feature activation-1.40e+00

Pos logit contributions
club+2.063
cent+2.062
wives+2.025
horse+2.016
lines+1.982
Neg logit contributions
 confir-2.444
 destro-2.224
 consolidate-2.145
 rely-2.091
 consume-2.088
Token box
Token ID3091
Feature activation-5.62e+00

Pos logit contributions
club+6.333
cent+6.298
wives+6.234
horse+6.225
lines+6.119
Neg logit contributions
 confir-7.411
 destro-6.737
 consolidate-6.413
 rely-6.217
 consume-6.208
Tokeners
Token ID364
Feature activation-4.90e+00

Pos logit contributions
club+4.233
cent+4.200
horse+4.129
wives+4.120
cart+4.001
Neg logit contributions
 confir-6.329
 destro-5.835
 consolidate-5.606
 rely-5.501
 consume-5.465
Token.
Token ID13
Feature activation-4.73e+00

Pos logit contributions
cent+4.276
club+4.268
wives+4.185
horse+4.164
lines+4.099
Neg logit contributions
 confir-4.923
 destro-4.462
 consolidate-4.272
 rely-4.133
 consume-4.129
Token
Token ID198
Feature activation-2.57e+00

Pos logit contributions
club+2.954
cent+2.949
wives+2.874
horse+2.859
lines+2.786
Neg logit contributions
 confir-5.988
 destro-5.553
 consolidate-5.382
 rely-5.280
 consume-5.273
Token
Token ID198
Feature activation-2.02e+00

Pos logit contributions
club+1.842
cent+1.840
wives+1.796
horse+1.791
lines+1.752
Neg logit contributions
 confir-3.035
 destro-2.802
 consolidate-2.694
 rely-2.634
 consume-2.633
Token
Token ID447
Feature activation-8.32e+00

Pos logit contributions
club+1.318
cent+1.316
wives+1.286
horse+1.278
lines+1.247
Neg logit contributions
 confir-2.487
 destro-2.301
 consolidate-2.240
 rely-2.199
 consume-2.195
Token
Token ID250
Feature activation-2.23e+00

Pos logit contributions
club+3.508
cent+3.487
horse+3.347
wives+3.345
cart+3.218
Neg logit contributions
 confir-12.060
 destro-11.337
 consolidate-11.023
 rely-10.843
 consume-10.818
TokenEvery
Token ID6109
Feature activation-2.26e+00

Pos logit contributions
club+1.499
cent+1.494
wives+1.466
horse+1.457
lines+1.423
Neg logit contributions
 confir-2.700
 destro-2.500
 consolidate-2.422
 rely-2.380
 consume-2.377
Token time
Token ID640
Feature activation-4.60e-01

Pos logit contributions
club+1.933
cent+1.929
wives+1.901
horse+1.885
lines+1.857
Neg logit contributions
 confir-2.341
 destro-2.135
 consolidate-2.056
 rely-2.012
 consume-2.009
Token my
Token ID616
Feature activation-2.18e+00

Pos logit contributions
club+0.380
cent+0.379
wives+0.372
horse+0.370
lines+0.363
Neg logit contributions
 confir-0.489
 destro-0.446
 consolidate-0.431
 rely-0.422
 consume-0.421
Token nose
Token ID9686
Feature activation-3.87e+00

Pos logit contributions
club+1.868
cent+1.867
wives+1.831
horse+1.823
lines+1.790
Neg logit contributions
 confir-2.242
 destro-2.042
 consolidate-1.968
 rely-1.924
 consume-1.920
Token So
Token ID1406
Feature activation-2.71e+00

Pos logit contributions
club+4.382
cent+4.338
wives+4.214
horse+4.183
lines+4.129
Neg logit contributions
 confir-8.941
 destro-8.316
 consolidate-7.934
 consume-7.789
 rely-7.747
Tokeni
Token ID72
Feature activation-3.16e+00

Pos logit contributions
cent+2.475
club+2.471
wives+2.422
horse+2.414
lines+2.382
Neg logit contributions
 confir-2.629
 destro-2.388
 consolidate-2.272
 consume-2.209
 rely-2.205
Token F
Token ID376
Feature activation-3.98e+00

Pos logit contributions
club+2.487
cent+2.484
wives+2.435
horse+2.420
lines+2.373
Neg logit contributions
 confir-3.464
 destro-3.175
 consolidate-3.067
 rely-2.996
 consume-2.990
Tokenon
Token ID261
Feature activation-8.89e-01

Pos logit contributions
cent+3.040
club+3.033
wives+2.954
horse+2.946
lines+2.894
Neg logit contributions
 confir-4.476
 destro-4.124
 consolidate-3.950
 consume-3.857
 rely-3.855
Token!
Token ID0
Feature activation-2.89e+00

Pos logit contributions
club+0.706
cent+0.705
wives+0.693
horse+0.690
cart+0.675
Neg logit contributions
 confir-0.971
 destro-0.889
 consolidate-0.857
 rely-0.841
 consume-0.839
Token
Token ID447
Feature activation-8.32e+00

Pos logit contributions
club+1.846
cent+1.843
wives+1.797
horse+1.787
lines+1.744
Neg logit contributions
 confir-3.606
 destro-3.342
 consolidate-3.243
 rely-3.181
 consume-3.176
Token
Token ID251
Feature activation-2.23e+00

Pos logit contributions
club+3.480
cent+3.464
horse+3.338
wives+3.332
cart+3.196
Neg logit contributions
 confir-12.147
 destro-11.401
 consolidate-11.098
 rely-10.925
 consume-10.910
Token There
Token ID1318
Feature activation-8.45e-01

Pos logit contributions
cent+1.503
club+1.502
wives+1.468
horse+1.460
lines+1.422
Neg logit contributions
 confir-2.710
 destro-2.501
 consolidate-2.433
 rely-2.391
 consume-2.385
Token was
Token ID373
Feature activation-3.83e+00

Pos logit contributions
club+0.842
cent+0.841
wives+0.828
horse+0.825
lines+0.812
Neg logit contributions
 confir-0.754
 destro-0.677
 consolidate-0.648
 rely-0.629
 consume-0.628
Token a
Token ID257
Feature activation-3.94e+00

Pos logit contributions
club+3.206
cent+3.202
wives+3.135
horse+3.129
lines+3.071
Neg logit contributions
 confir-4.007
 destro-3.663
 consolidate-3.532
 rely-3.440
 consume-3.433
Token sudden
Token ID4802
Feature activation-1.98e+00

Pos logit contributions
club+3.545
cent+3.536
wives+3.468
horse+3.465
lines+3.397
Neg logit contributions
 confir-3.886
 destro-3.525
 consolidate-3.386
 rely-3.299
 consume-3.290
Token movement
Token ID3356
Feature activation-1.82e+00

Pos logit contributions
 confir+0.084
 destro+0.076
 consolidate+0.073
 rely+0.072
 consume+0.071
Neg logit contributions
cent-0.080
club-0.079
wives-0.079
horse-0.078
cart-0.076
Token,
Token ID11
Feature activation-5.31e-01

Pos logit contributions
club+1.522
cent+1.519
wives+1.493
horse+1.483
lines+1.456
Neg logit contributions
 confir-1.914
 destro-1.748
 consolidate-1.687
 rely-1.651
 consume-1.646
Token shun
Token ID32814
Feature activation-4.87e-01

Pos logit contributions
club+0.464
cent+0.462
wives+0.455
horse+0.452
cart+0.444
Neg logit contributions
 confir-0.540
 destro-0.492
 consolidate-0.474
 rely-0.465
 consume-0.463
Tokenpo
Token ID7501
Feature activation-6.32e-02

Pos logit contributions
club+0.408
cent+0.408
wives+0.403
horse+0.399
lines+0.390
Neg logit contributions
 confir-0.509
 destro-0.465
 consolidate-0.448
 rely-0.440
 consume-0.438
Token.
Token ID13
Feature activation-3.06e+00

Pos logit contributions
club+0.051
cent+0.051
wives+0.050
horse+0.050
cart+0.049
Neg logit contributions
 confir-0.068
 destro-0.063
 consolidate-0.060
 rely-0.059
 consume-0.059
Token
Token ID564
Feature activation-8.25e+00

Pos logit contributions
club+1.785
cent+1.783
wives+1.736
horse+1.724
lines+1.677
Neg logit contributions
 confir-3.983
 destro-3.702
 consolidate-3.601
 rely-3.539
 consume-3.532
Token
Token ID250
Feature activation-1.84e+00

Pos logit contributions
club+3.503
cent+3.469
wives+3.360
horse+3.348
lines+3.231
Neg logit contributions
 confir-11.852
 destro-11.160
 consolidate-10.820
 rely-10.652
 consume-10.624
TokenI
Token ID40
Feature activation-2.46e+00

Pos logit contributions
club+1.226
cent+1.223
wives+1.200
horse+1.191
lines+1.161
Neg logit contributions
 confir-2.240
 destro-2.071
 consolidate-2.013
 rely-1.979
 consume-1.974
Token
Token ID447
Feature activation-2.45e+00

Pos logit contributions
cent+2.144
club+2.143
horse+2.099
wives+2.092
cart+2.050
Neg logit contributions
 confir-2.483
 destro-2.254
 consolidate-2.172
 consume-2.109
 rely-2.106
Token
Token ID247
Feature activation-3.90e+00

Pos logit contributions
club+1.186
cent+1.180
horse+1.141
wives+1.140
lines+1.093
Neg logit contributions
 confir-3.448
 destro-3.237
 consolidate-3.128
 rely-3.074
 consume-3.073
Tokenm
Token ID76
Feature activation-2.40e+00

Pos logit contributions
club+2.890
cent+2.889
wives+2.827
horse+2.813
lines+2.753
Neg logit contributions
 confir-4.460
 destro-4.104
 consolidate-3.971
 rely-3.885
 consume-3.879
Tokenifted
Token ID21715
Feature activation+2.92e-01

Pos logit contributions
club+0.598
cent+0.594
wives+0.585
horse+0.582
lines+0.569
Neg logit contributions
 confir-1.021
 destro-0.943
 consolidate-0.916
 rely-0.899
 consume-0.898
Token and
Token ID290
Feature activation+8.72e-01

Pos logit contributions
 confir+0.301
 destro+0.274
 consolidate+0.266
 rely+0.261
 consume+0.259
Neg logit contributions
club-0.250
cent-0.248
wives-0.244
horse-0.243
lines-0.238
Token put
Token ID1234
Feature activation-3.08e+00

Pos logit contributions
 confir+0.797
 destro+0.718
 consolidate+0.690
 rely+0.671
 consume+0.669
Neg logit contributions
club-0.849
cent-0.846
wives-0.835
horse-0.831
lines-0.818
Token on
Token ID319
Feature activation-2.92e+00

Pos logit contributions
cent+2.421
club+2.419
wives+2.370
horse+2.359
lines+2.315
Neg logit contributions
 confir-3.391
 destro-3.109
 consolidate-2.997
 rely-2.929
 consume-2.927
Token a
Token ID257
Feature activation-2.33e+00

Pos logit contributions
cent+2.591
club+2.590
wives+2.545
horse+2.534
lines+2.494
Neg logit contributions
 confir-2.908
 destro-2.644
 consolidate-2.541
 rely-2.477
 consume-2.475
Token basement
Token ID17012
Feature activation-8.24e+00

Pos logit contributions
club+2.199
cent+2.197
wives+2.158
horse+2.153
lines+2.117
Neg logit contributions
 confir-2.187
 destro-1.972
 consolidate-1.893
 rely-1.844
 consume-1.841
Token of
Token ID286
Feature activation-3.34e+00

Pos logit contributions
cent+7.874
club+7.815
wives+7.693
horse+7.668
cart+7.613
Neg logit contributions
 confir-7.167
 destro-6.482
 consolidate-6.171
 consume-5.997
 comply-5.923
Token concrete
Token ID10017
Feature activation-3.96e+00

Pos logit contributions
cent+2.996
club+2.993
wives+2.943
horse+2.929
lines+2.887
Neg logit contributions
 confir-3.291
 destro-2.989
 consolidate-2.871
 rely-2.801
 consume-2.795
Token.
Token ID13
Feature activation-1.70e+00

Pos logit contributions
cent+3.334
club+3.327
wives+3.271
horse+3.249
lines+3.216
Neg logit contributions
 confir-4.104
 destro-3.749
 consolidate-3.601
 consume-3.517
 rely-3.516
Token Ma
Token ID6669
Feature activation-2.50e+00

Pos logit contributions
club+1.021
cent+1.018
wives+0.991
horse+0.985
cart+0.958
Neg logit contributions
 confir-2.176
 destro-2.017
 consolidate-1.972
 rely-1.944
 consume-1.932
Tokenori
Token ID10145
Feature activation+5.82e-01

Pos logit contributions
club+1.989
cent+1.988
wives+1.948
horse+1.940
lines+1.901
Neg logit contributions
 confir-2.726
 destro-2.497
 consolidate-2.413
 rely-2.360
 consume-2.355
Tokenway
Token ID1014
Feature activation-3.22e+00

Pos logit contributions
club+1.060
cent+1.057
wives+1.002
horse+0.990
lines+0.937
Neg logit contributions
 confir-5.444
 destro-5.128
 consolidate-5.013
 rely-4.941
 consume-4.934
Token,
Token ID11
Feature activation-1.49e+00

Pos logit contributions
cent+2.552
club+2.544
wives+2.499
horse+2.488
lines+2.439
Neg logit contributions
 confir-3.521
 destro-3.222
 consolidate-3.113
 rely-3.036
 consume-3.036
Token and
Token ID290
Feature activation-1.10e+00

Pos logit contributions
club+1.153
cent+1.148
wives+1.126
horse+1.122
lines+1.098
Neg logit contributions
 confir-1.670
 destro-1.533
 consolidate-1.483
 rely-1.455
 consume-1.451
Token a
Token ID257
Feature activation-4.57e+00

Pos logit contributions
cent+1.034
club+1.033
wives+1.017
horse+1.012
lines+0.996
Neg logit contributions
 confir-1.050
 destro-0.948
 consolidate-0.910
 rely-0.885
 consume-0.883
Token cerebral
Token ID31169
Feature activation-5.22e+00

Pos logit contributions
cent+4.227
club+4.220
wives+4.141
horse+4.130
lines+4.058
Neg logit contributions
 confir-4.311
 destro-3.909
 consolidate-3.754
 consume-3.625
 rely-3.625
Token journalist
Token ID10099
Feature activation-8.21e+00

Pos logit contributions
cent+4.537
club+4.504
horse+4.426
wives+4.419
cart+4.340
Neg logit contributions
 confir-5.219
 destro-4.757
 consolidate-4.592
 consume-4.452
 rely-4.419
Token played
Token ID2826
Feature activation-5.08e+00

Pos logit contributions
cent+6.931
club+6.854
horse+6.768
wives+6.764
lines+6.641
Neg logit contributions
 confir-8.252
 destro-7.500
 consolidate-7.250
 consume-6.954
 comply-6.942
Token by
Token ID416
Feature activation-2.67e+00

Pos logit contributions
cent+3.571
club+3.548
wives+3.476
horse+3.469
lines+3.385
Neg logit contributions
 confir-6.004
 destro-5.523
 consolidate-5.361
 consume-5.233
 rely-5.202
Token Diane
Token ID26542
Feature activation-4.21e+00

Pos logit contributions
club+1.622
cent+1.622
wives+1.579
horse+1.570
lines+1.531
Neg logit contributions
 confir-3.410
 destro-3.165
 consolidate-3.075
 rely-3.015
 consume-3.012
Token Ke
Token ID3873
Feature activation-4.06e+00

Pos logit contributions
cent+2.430
club+2.368
horse+2.338
wives+2.336
lines+2.262
Neg logit contributions
 confir-5.592
 destro-5.178
 consolidate-5.009
 consume-4.889
 rely-4.884
Tokenaton
Token ID13951
Feature activation-7.81e-01

Pos logit contributions
cent+2.907
club+2.821
wives+2.780
horse+2.743
cart+2.739
Neg logit contributions
 confir-4.896
 destro-4.516
 consolidate-4.316
 rely-4.205
 consume-4.188
Token version
Token ID2196
Feature activation-4.01e+00

Pos logit contributions
club+0.757
cent+0.755
wives+0.744
horse+0.739
lines+0.727
Neg logit contributions
 confir-0.827
 destro-0.751
 consolidate-0.723
 rely-0.706
 consume-0.704
Token of
Token ID286
Feature activation-2.53e+00

Pos logit contributions
cent+2.991
club+2.989
wives+2.914
horse+2.907
lines+2.850
Neg logit contributions
 confir-4.570
 destro-4.208
 consolidate-4.069
 consume-3.977
 rely-3.973
Token one
Token ID530
Feature activation+1.83e-01

Pos logit contributions
club+2.012
cent+2.007
wives+1.970
horse+1.959
lines+1.919
Neg logit contributions
 confir-2.765
 destro-2.533
 consolidate-2.448
 rely-2.397
 consume-2.390
Token of
Token ID286
Feature activation-2.64e+00

Pos logit contributions
 confir+0.210
 destro+0.194
 consolidate+0.187
 rely+0.185
 consume+0.183
Neg logit contributions
club-0.134
wives-0.132
cent-0.132
horse-0.129
masters-0.127
Token his
Token ID465
Feature activation-1.66e+00

Pos logit contributions
club+1.959
cent+1.957
wives+1.916
horse+1.906
lines+1.866
Neg logit contributions
 confir-3.033
 destro-2.790
 consolidate-2.701
 rely-2.645
 consume-2.640
Token characters
Token ID3435
Feature activation-8.21e+00

Pos logit contributions
club+1.231
cent+1.226
wives+1.203
horse+1.196
lines+1.166
Neg logit contributions
 confir-1.912
 destro-1.759
 consolidate-1.704
 rely-1.671
 consume-1.666
Token,
Token ID11
Feature activation-1.37e-01

Pos logit contributions
cent+6.844
club+6.772
wives+6.667
horse+6.625
lines+6.567
Neg logit contributions
 confir-8.345
 destro-7.636
 consolidate-7.306
 consume-7.091
 deprive-7.046
Token and
Token ID290
Feature activation-5.40e-02

Pos logit contributions
club+0.109
cent+0.108
wives+0.107
horse+0.106
lines+0.104
Neg logit contributions
 confir-0.149
 destro-0.137
 consolidate-0.133
 rely-0.130
 consume-0.130
Token just
Token ID655
Feature activation+1.28e-01

Pos logit contributions
club+0.047
cent+0.047
wives+0.046
horse+0.046
lines+0.045
Neg logit contributions
 confir-0.055
 destro-0.050
 consolidate-0.048
 rely-0.047
 consume-0.047
Token
Token ID1399
Feature activation-2.16e+00

Pos logit contributions
 confir+0.120
 destro+0.108
 consolidate+0.104
 rely+0.101
 consume+0.101
Neg logit contributions
club-0.121
cent-0.121
wives-0.119
horse-0.118
lines-0.117
Token It
Token ID632
Feature activation-6.47e-01

Pos logit contributions
club+1.768
cent+1.766
wives+1.733
horse+1.725
lines+1.691
Neg logit contributions
 confir-2.316
 destro-2.117
 consolidate-2.045
 rely-2.000
 consume-1.995
Token as
Token ID355
Feature activation-2.40e+00

Pos logit contributions
club+4.072
cent+4.060
wives+3.980
horse+3.972
lines+3.903
Neg logit contributions
 confir-4.373
 destro-3.951
 consolidate-3.793
 consume-3.695
 rely-3.693
Token Ted
Token ID11396
Feature activation-4.24e+00

Pos logit contributions
cent+2.164
club+2.163
wives+2.123
horse+2.117
lines+2.078
Neg logit contributions
 confir-2.363
 destro-2.144
 consolidate-2.063
 rely-2.009
 consume-2.002
Token Falcon
Token ID17621
Feature activation-3.81e+00

Pos logit contributions
cent+3.228
club+3.204
wives+3.154
horse+3.133
lines+3.076
Neg logit contributions
 confir-4.775
 destro-4.382
 consolidate-4.244
 consume-4.138
 rely-4.132
Tokeni
Token ID72
Feature activation-3.57e+00

Pos logit contributions
club+3.089
cent+3.088
wives+3.025
horse+3.023
lines+2.954
Neg logit contributions
 confir-4.092
 destro-3.742
 consolidate-3.610
 rely-3.525
 consume-3.521
Token,
Token ID11
Feature activation-8.41e-01

Pos logit contributions
cent+3.173
club+3.167
wives+3.113
horse+3.097
lines+3.040
Neg logit contributions
 confir-3.559
 destro-3.231
 consolidate-3.109
 rely-3.029
 consume-3.023
Token playing
Token ID2712
Feature activation-8.20e+00

Pos logit contributions
club+0.665
cent+0.662
wives+0.652
horse+0.649
lines+0.636
Neg logit contributions
 confir-0.921
 destro-0.846
 consolidate-0.816
 rely-0.800
 consume-0.799
Token guitar
Token ID10047
Feature activation-6.35e+00

Pos logit contributions
cent+6.784
club+6.775
horse+6.652
wives+6.572
lines+6.518
Neg logit contributions
 confir-8.368
 destro-7.574
 consolidate-7.297
 consume-7.090
 rely-7.068
Token.
Token ID13
Feature activation-3.76e+00

Pos logit contributions
club+5.032
cent+5.022
wives+4.907
horse+4.904
lines+4.866
Neg logit contributions
 confir-6.871
 destro-6.263
 consolidate-6.052
 consume-5.905
 rely-5.877
Token Then
Token ID3244
Feature activation-1.86e-01

Pos logit contributions
club+2.187
cent+2.185
wives+2.128
horse+2.112
lines+2.055
Neg logit contributions
 confir-4.903
 destro-4.558
 consolidate-4.437
 rely-4.359
 consume-4.351
Token,
Token ID11
Feature activation-1.95e+00

Pos logit contributions
cent+0.174
club+0.174
wives+0.171
horse+0.170
lines+0.167
Neg logit contributions
 confir-0.177
 destro-0.160
 consolidate-0.154
 rely-0.150
 consume-0.150
Token a
Token ID257
Feature activation-5.24e+00

Pos logit contributions
club+1.802
cent+1.801
wives+1.771
horse+1.762
lines+1.733
Neg logit contributions
 confir-1.879
 destro-1.701
 consolidate-1.637
 rely-1.596
 consume-1.592
Token I
Token ID314
Feature activation-9.49e-01

Pos logit contributions
 confir+2.602
 destro+2.366
 consolidate+2.284
 rely+2.243
 consume+2.242
Neg logit contributions
club-2.294
cent-2.270
horse-2.235
wives-2.232
cart-2.182
Token say
Token ID910
Feature activation-1.74e+00

Pos logit contributions
club+0.979
cent+0.979
wives+0.962
horse+0.959
lines+0.944
Neg logit contributions
 confir-0.814
 destro-0.727
 consolidate-0.694
 rely-0.673
 consume-0.671
Token it
Token ID340
Feature activation+1.95e+00

Pos logit contributions
club+1.542
cent+1.541
wives+1.514
horse+1.508
lines+1.482
Neg logit contributions
 confir-1.744
 destro-1.584
 consolidate-1.524
 rely-1.487
 consume-1.484
Token was
Token ID373
Feature activation-1.09e+00

Pos logit contributions
 confir+1.773
 destro+1.605
 consolidate+1.543
 rely+1.516
 consume+1.505
Neg logit contributions
club-1.876
cent-1.869
wives-1.847
horse-1.831
cart-1.800
Token an
Token ID281
Feature activation-2.04e+00

Pos logit contributions
club+0.965
cent+0.962
wives+0.948
horse+0.943
lines+0.925
Neg logit contributions
 confir-1.098
 destro-0.997
 consolidate-0.962
 rely-0.941
 consume-0.937
Token actor
Token ID8674
Feature activation-8.10e+00

Pos logit contributions
club+1.988
cent+1.988
wives+1.954
horse+1.947
lines+1.916
Neg logit contributions
 confir-1.860
 destro-1.674
 consolidate-1.603
 rely-1.560
 consume-1.557
Token
Token ID447
Feature activation-3.72e+00

Pos logit contributions
club+7.437
cent+7.434
horse+7.270
wives+7.269
lines+7.126
Neg logit contributions
 confir-7.487
 destro-6.784
 consolidate-6.496
 consume-6.294
 rely-6.283
Token
Token ID247
Feature activation-2.36e+00

Pos logit contributions
club+1.925
cent+1.917
wives+1.861
horse+1.855
lines+1.793
Neg logit contributions
 confir-5.090
 destro-4.751
 consolidate-4.630
 rely-4.548
 consume-4.542
Tokens
Token ID82
Feature activation-4.74e+00

Pos logit contributions
club+1.942
cent+1.937
wives+1.905
horse+1.895
lines+1.860
Neg logit contributions
 confir-2.522
 destro-2.304
 consolidate-2.225
 rely-2.180
 consume-2.173
Token clinic
Token ID15760
Feature activation-4.27e+00

Pos logit contributions
cent+4.520
club+4.505
wives+4.422
horse+4.395
lines+4.327
Neg logit contributions
 confir-4.424
 destro-4.002
 consolidate-3.836
 consume-3.729
 rely-3.725
Token.
Token ID13
Feature activation-2.44e+00

Pos logit contributions
cent+3.538
club+3.535
wives+3.465
horse+3.459
lines+3.386
Neg logit contributions
 confir-4.507
 destro-4.121
 consolidate-3.964
 consume-3.862
 rely-3.862
Token un
Token ID555
Feature activation-1.95e+00

Pos logit contributions
club+1.185
cent+1.184
wives+1.165
horse+1.160
lines+1.142
Neg logit contributions
 confir-1.090
 destro-0.979
 consolidate-0.939
 rely-0.914
 consume-0.911
Tokenruly
Token ID34715
Feature activation+1.52e+00

Pos logit contributions
club+1.387
cent+1.386
wives+1.359
horse+1.349
lines+1.322
Neg logit contributions
 confir-2.271
 destro-2.092
 consolidate-2.033
 rely-1.993
 consume-1.990
Token slave
Token ID11778
Feature activation-3.52e+00

Pos logit contributions
 confir+1.424
 destro+1.287
 consolidate+1.241
 consume+1.217
 rely+1.217
Neg logit contributions
club-1.440
cent-1.429
horse-1.397
wives-1.397
cart-1.376
Token.
Token ID13
Feature activation-1.25e+00

Pos logit contributions
club+2.827
cent+2.825
wives+2.786
horse+2.769
lines+2.711
Neg logit contributions
 confir-3.817
 destro-3.489
 consolidate-3.361
 rely-3.285
 consume-3.275
Token The
Token ID383
Feature activation-2.61e+00

Pos logit contributions
club+0.734
cent+0.730
wives+0.712
horse+0.700
lines+0.682
Neg logit contributions
 confir-1.601
 destro-1.492
 consolidate-1.466
 rely-1.447
 consume-1.435
Token story
Token ID1621
Feature activation-8.08e+00

Pos logit contributions
cent+2.381
club+2.377
wives+2.340
horse+2.332
lines+2.296
Neg logit contributions
 confir-2.543
 destro-2.300
 consolidate-2.214
 rely-2.157
 consume-2.153
Token was
Token ID373
Feature activation-3.20e+00

Pos logit contributions
cent+7.609
horse+7.423
club+7.413
lines+7.363
wives+7.338
Neg logit contributions
 confir-7.303
 destro-6.663
 consolidate-6.344
 consume-6.189
 comply-6.124
Token almost
Token ID2048
Feature activation-1.65e+00

Pos logit contributions
cent+3.044
club+3.017
horse+2.981
wives+2.976
lines+2.932
Neg logit contributions
 confir-2.964
 destro-2.682
 consolidate-2.567
 rely-2.490
 consume-2.489
Token certainly
Token ID3729
Feature activation+1.74e+00

Pos logit contributions
cent+1.631
club+1.621
wives+1.598
horse+1.597
lines+1.572
Neg logit contributions
 confir-1.485
 destro-1.338
 consolidate-1.283
 rely-1.238
 consume-1.237
Token Greek
Token ID8312
Feature activation-3.95e-01

Pos logit contributions
 confir+1.714
 destro+1.554
 consolidate+1.499
 rely+1.470
 consume+1.458
Neg logit contributions
club-1.568
cent-1.546
wives-1.535
horse-1.514
lines-1.493
Token propaganda
Token ID11613
Feature activation-1.42e+00

Pos logit contributions
club+0.340
cent+0.338
wives+0.333
horse+0.331
lines+0.325
Neg logit contributions
 confir-0.406
 destro-0.369
 consolidate-0.357
 rely-0.349
 consume-0.348
Token Cour
Token ID2734
Feature activation-6.88e+00

Pos logit contributions
cent+3.161
club+3.150
wives+3.060
horse+3.058
lines+2.986
Neg logit contributions
 confir-6.502
 destro-6.026
 consolidate-5.829
 rely-5.721
 consume-5.714
Tokentyard
Token ID30308
Feature activation-4.16e+00

Pos logit contributions
cent+3.286
club+3.235
wives+3.118
horse+3.085
lines+3.022
Neg logit contributions
 confir-9.686
 destro-9.072
 consolidate-8.832
 rely-8.674
 consume-8.667
Token,
Token ID11
Feature activation-4.31e-01

Pos logit contributions
club+3.229
cent+3.226
wives+3.161
horse+3.146
lines+3.082
Neg logit contributions
 confir-4.624
 destro-4.242
 consolidate-4.102
 rely-4.015
 consume-4.007
Token where
Token ID810
Feature activation-1.53e+00

Pos logit contributions
club+0.362
cent+0.359
wives+0.355
horse+0.352
cart+0.345
Neg logit contributions
 confir-0.451
 destro-0.412
 consolidate-0.399
 rely-0.392
 consume-0.390
Token the
Token ID262
Feature activation-3.26e+00

Pos logit contributions
club+1.289
cent+1.288
wives+1.264
horse+1.259
lines+1.235
Neg logit contributions
 confir-1.600
 destro-1.459
 consolidate-1.408
 rely-1.375
 consume-1.372
Token characters
Token ID3435
Feature activation-8.02e+00

Pos logit contributions
cent+2.995
club+2.989
wives+2.932
horse+2.927
lines+2.887
Neg logit contributions
 confir-3.151
 destro-2.853
 consolidate-2.744
 consume-2.666
 rely-2.665
Token boast
Token ID28020
Feature activation-5.02e+00

Pos logit contributions
cent+8.800
club+8.718
wives+8.626
horse+8.556
lines+8.499
Neg logit contributions
 confir-5.976
 destro-5.219
 consolidate-4.785
 deprive-4.520
 consume-4.517
Token weird
Token ID7650
Feature activation-3.37e+00

Pos logit contributions
club+4.404
cent+4.404
wives+4.314
horse+4.314
lines+4.248
Neg logit contributions
 confir-5.046
 destro-4.577
 consolidate-4.407
 consume-4.280
 rely-4.277
Tokenly
Token ID306
Feature activation-2.08e+00

Pos logit contributions
club+2.984
cent+2.973
horse+2.915
wives+2.913
lines+2.880
Neg logit contributions
 confir-3.374
 destro-3.061
 consolidate-2.940
 consume-2.867
 rely-2.864
Token elong
Token ID39696
Feature activation-2.97e+00

Pos logit contributions
club+1.943
cent+1.941
wives+1.903
horse+1.902
lines+1.869
Neg logit contributions
 confir-1.982
 destro-1.791
 consolidate-1.717
 rely-1.672
 consume-1.669
Tokenated
Token ID515
Feature activation-3.41e+00

Pos logit contributions
cent+2.315
club+2.308
wives+2.259
horse+2.246
lines+2.201
Neg logit contributions
 confir-3.301
 destro-3.024
 consolidate-2.918
 rely-2.856
 consume-2.850
Token on
Token ID319
Feature activation-3.09e+00

Pos logit contributions
club+1.848
cent+1.847
wives+1.813
horse+1.805
lines+1.772
Neg logit contributions
 confir-2.249
 destro-2.049
 consolidate-1.976
 rely-1.930
 consume-1.925
Token that
Token ID326
Feature activation+2.97e+00

Pos logit contributions
cent+2.572
club+2.571
wives+2.522
horse+2.511
lines+2.466
Neg logit contributions
 confir-3.252
 destro-2.968
 consolidate-2.865
 rely-2.796
 consume-2.790
Token set
Token ID900
Feature activation-3.22e+00

Pos logit contributions
 confir+3.103
 destro+2.837
 consolidate+2.708
 rely+2.696
 consume+2.691
Neg logit contributions
club-2.475
cent-2.445
wives-2.414
horse-2.357
cart-2.330
Token.
Token ID13
Feature activation-2.52e+00

Pos logit contributions
club+2.667
cent+2.664
wives+2.614
horse+2.602
lines+2.553
Neg logit contributions
 confir-3.404
 destro-3.109
 consolidate-3.000
 rely-2.933
 consume-2.927
Token The
Token ID383
Feature activation-4.51e+00

Pos logit contributions
club+1.449
cent+1.445
wives+1.407
horse+1.398
cart+1.358
Neg logit contributions
 confir-3.273
 destro-3.045
 consolidate-2.984
 rely-2.942
 consume-2.927
Token boys
Token ID6510
Feature activation-7.94e+00

Pos logit contributions
cent+3.990
club+3.982
wives+3.939
horse+3.903
lines+3.847
Neg logit contributions
 confir-4.493
 destro-4.076
 consolidate-3.934
 rely-3.822
 consume-3.817
Token had
Token ID550
Feature activation-3.41e+00

Pos logit contributions
club+7.595
cent+7.584
wives+7.533
horse+7.409
lines+7.293
Neg logit contributions
 confir-7.024
 destro-6.308
 consolidate-5.959
 deprive-5.683
 consume-5.650
Token a
Token ID257
Feature activation-2.99e+00

Pos logit contributions
club+3.047
cent+3.044
wives+2.993
horse+2.983
lines+2.936
Neg logit contributions
 confir-3.383
 destro-3.068
 consolidate-2.962
 rely-2.878
 consume-2.875
Token lot
Token ID1256
Feature activation-1.47e+00

Pos logit contributions
club+2.543
cent+2.540
wives+2.496
horse+2.490
lines+2.437
Neg logit contributions
 confir-3.092
 destro-2.817
 consolidate-2.719
 rely-2.650
 consume-2.648
Token of
Token ID286
Feature activation-2.50e+00

Pos logit contributions
club+1.068
cent+1.067
wives+1.043
horse+1.038
lines+1.016
Neg logit contributions
 confir-1.705
 destro-1.571
 consolidate-1.521
 rely-1.491
 consume-1.487
Token fun
Token ID1257
Feature activation-1.74e+00

Pos logit contributions
club+2.357
cent+2.355
wives+2.317
horse+2.309
lines+2.272
Neg logit contributions
 confir-2.368
 destro-2.138
 consolidate-2.054
 rely-1.997
 consume-1.995
Token burdens
Token ID27127
Feature activation-2.62e+00

Pos logit contributions
club+0.800
cent+0.798
wives+0.784
horse+0.782
cart+0.766
Neg logit contributions
 confir-0.891
 destro-0.808
 consolidate-0.780
 rely-0.762
 consume-0.760
Token of
Token ID286
Feature activation-1.98e+00

Pos logit contributions
cent+2.373
club+2.353
wives+2.326
horse+2.321
lines+2.273
Neg logit contributions
 confir-2.583
 destro-2.338
 consolidate-2.229
 rely-2.172
 consume-2.165
Token meaning
Token ID3616
Feature activation-1.93e+00

Pos logit contributions
club+1.765
cent+1.762
wives+1.732
horse+1.725
lines+1.694
Neg logit contributions
 confir-1.978
 destro-1.797
 consolidate-1.730
 rely-1.689
 consume-1.685
Token passed
Token ID3804
Feature activation-1.97e+00

Pos logit contributions
cent+2.012
club+2.004
wives+1.977
horse+1.974
lines+1.943
Neg logit contributions
 confir-1.633
 destro-1.454
 consolidate-1.377
 rely-1.333
 consume-1.327
Token from
Token ID422
Feature activation-2.43e+00

Pos logit contributions
cent+1.572
club+1.567
wives+1.541
horse+1.533
lines+1.506
Neg logit contributions
 confir-2.155
 destro-1.975
 consolidate-1.898
 rely-1.852
 consume-1.849
Token writer
Token ID6260
Feature activation-7.91e+00

Pos logit contributions
cent+2.101
club+2.099
wives+2.061
horse+2.053
lines+2.017
Neg logit contributions
 confir-2.482
 destro-2.259
 consolidate-2.175
 rely-2.122
 consume-2.118
Token to
Token ID284
Feature activation-2.87e-01

Pos logit contributions
club+6.775
cent+6.760
horse+6.645
wives+6.585
lines+6.502
Neg logit contributions
 confir-7.978
 destro-7.268
 consolidate-6.907
 consume-6.707
 rely-6.702
Token reader
Token ID9173
Feature activation-6.40e+00

Pos logit contributions
club+0.196
cent+0.195
wives+0.192
horse+0.189
lines+0.185
Neg logit contributions
 confir-0.347
 destro-0.321
 consolidate-0.311
 consume-0.306
 rely-0.306
Token.
Token ID13
Feature activation-1.22e+00

Pos logit contributions
cent+4.910
club+4.873
horse+4.802
cart+4.755
wives+4.741
Neg logit contributions
 confir-7.053
 destro-6.485
 consolidate-6.120
 rely-5.954
 comply-5.946
Token The
Token ID383
Feature activation-3.10e+00

Pos logit contributions
cent+0.700
club+0.699
wives+0.681
horse+0.676
lines+0.652
Neg logit contributions
 confir-1.579
 destro-1.468
 consolidate-1.445
 rely-1.425
 consume-1.416
Token one
Token ID530
Feature activation-2.31e+00

Pos logit contributions
cent+2.941
club+2.935
wives+2.886
horse+2.876
lines+2.841
Neg logit contributions
 confir-2.911
 destro-2.625
 consolidate-2.513
 rely-2.445
 consume-2.444
Token one
Token ID530
Feature activation-2.00e-01

Pos logit contributions
 confir+2.281
 destro+2.065
 consolidate+2.000
 rely+1.965
 consume+1.951
Neg logit contributions
club-1.926
cent-1.915
wives-1.892
horse-1.868
cart-1.842
Token of
Token ID286
Feature activation-2.75e+00

Pos logit contributions
club+0.148
cent+0.147
wives+0.144
horse+0.142
lines+0.141
Neg logit contributions
 confir-0.229
 destro-0.211
 consolidate-0.204
 rely-0.201
 consume-0.201
Token them
Token ID606
Feature activation+4.03e-01

Pos logit contributions
club+1.902
cent+1.902
wives+1.862
horse+1.851
lines+1.811
Neg logit contributions
 confir-3.290
 destro-3.038
 consolidate-2.943
 rely-2.882
 consume-2.878
Token;
Token ID26
Feature activation-1.41e+00

Pos logit contributions
 confir+0.429
 destro+0.391
 consolidate+0.378
 rely+0.371
 consume+0.370
Neg logit contributions
club-0.332
cent-0.331
wives-0.325
horse-0.323
cart-0.317
Token it
Token ID340
Feature activation+3.63e-02

Pos logit contributions
club+1.299
cent+1.298
wives+1.278
horse+1.272
lines+1.250
Neg logit contributions
 confir-1.358
 destro-1.228
 consolidate-1.178
 rely-1.147
 consume-1.145
Token
Token ID447
Feature activation-7.85e+00

Pos logit contributions
 confir+0.032
 destro+0.029
 consolidate+0.027
 rely+0.026
 consume+0.026
Neg logit contributions
club-0.037
cent-0.037
wives-0.036
horse-0.036
lines-0.035
Token
Token ID247
Feature activation-3.64e+00

Pos logit contributions
club+3.811
cent+3.776
wives+3.674
horse+3.663
lines+3.524
Neg logit contributions
 confir-10.858
 destro-10.195
 consolidate-9.879
 consume-9.719
 rely-9.708
Tokens
Token ID82
Feature activation-4.06e+00

Pos logit contributions
cent+2.633
club+2.632
wives+2.572
horse+2.559
lines+2.508
Neg logit contributions
 confir-4.234
 destro-3.898
 consolidate-3.782
 rely-3.699
 consume-3.693
Token just
Token ID655
Feature activation-3.43e-02

Pos logit contributions
cent+3.716
club+3.709
wives+3.652
horse+3.631
lines+3.577
Neg logit contributions
 confir-3.911
 destro-3.533
 consolidate-3.402
 rely-3.306
 consume-3.304
Token a
Token ID257
Feature activation-3.05e+00

Pos logit contributions
club+0.029
cent+0.029
wives+0.029
horse+0.028
cart+0.028
Neg logit contributions
 confir-0.036
 destro-0.032
 consolidate-0.031
 rely-0.031
 consume-0.031
Token very
Token ID845
Feature activation-1.89e+00

Pos logit contributions
club+2.835
cent+2.831
wives+2.785
horse+2.779
lines+2.730
Neg logit contributions
 confir-2.900
 destro-2.620
 consolidate-2.517
 rely-2.455
 consume-2.446
Token
Token ID198
Feature activation+2.55e+00

Pos logit contributions
club+0.097
cent+0.096
wives+0.095
horse+0.094
lines+0.093
Neg logit contributions
 confir-0.158
 destro-0.146
 consolidate-0.142
 rely-0.140
 consume-0.139
Token
Token ID198
Feature activation+2.31e+00

Pos logit contributions
 confir+2.679
 consolidate+2.460
 rely+2.433
 destro+2.419
 consume+2.416
Neg logit contributions
club-1.955
cent-1.945
wives-1.926
horse-1.903
lines-1.858
Token"
Token ID1
Feature activation-1.93e-02

Pos logit contributions
 confir+2.681
 destro+2.472
 consolidate+2.446
 rely+2.413
 comply+2.397
Neg logit contributions
club-1.533
cent-1.525
wives-1.519
horse-1.493
lines-1.470
TokenSpecial
Token ID13409
Feature activation+1.16e+00

Pos logit contributions
club+0.013
cent+0.013
wives+0.013
horse+0.012
lines+0.012
Neg logit contributions
 confir-0.023
 destro-0.021
 consolidate-0.021
 rely-0.021
 consume-0.020
Token needs
Token ID2476
Feature activation+4.43e+00

Pos logit contributions
 confir+0.755
 destro+0.651
 consolidate+0.618
 rely+0.602
 consume+0.592
Neg logit contributions
club-1.430
cent-1.423
wives-1.409
horse-1.406
cart-1.391
Token vouchers
Token ID41262
Feature activation+6.76e+00

Pos logit contributions
 confir+3.998
 destro+3.618
 consolidate+3.511
 rely+3.486
 consume+3.429
Neg logit contributions
club-4.276
cent-4.227
wives-4.201
horse-4.187
cart-4.133
Token are
Token ID389
Feature activation+1.60e+00

Pos logit contributions
 confir+5.032
 destro+4.459
 consolidate+4.440
 rely+4.372
 consume+4.335
Neg logit contributions
club-7.444
cent-7.360
horse-7.326
wives-7.314
laughs-7.247
Token a
Token ID257
Feature activation-2.38e-01

Pos logit contributions
 confir+1.501
 destro+1.357
 consolidate+1.320
 rely+1.290
 consume+1.287
Neg logit contributions
club-1.498
cent-1.486
wives-1.469
horse-1.462
cart-1.439
Token product
Token ID1720
Feature activation-1.11e+00

Pos logit contributions
club+0.243
cent+0.243
wives+0.239
horse+0.238
lines+0.235
Neg logit contributions
 confir-0.207
 destro-0.185
 consolidate-0.177
 rely-0.172
 consume-0.171
Token of
Token ID286
Feature activation-5.27e-01

Pos logit contributions
club+0.775
cent+0.774
wives+0.757
horse+0.752
lines+0.736
Neg logit contributions
 confir-1.327
 destro-1.225
 consolidate-1.188
 rely-1.165
 consume-1.163
Token calculating
Token ID26019
Feature activation+1.75e+00

Pos logit contributions
club+0.482
cent+0.481
wives+0.473
horse+0.471
lines+0.463
Neg logit contributions
 confir-0.514
 destro-0.466
 consolidate-0.448
 rely-0.437
 consume-0.436
Token along
Token ID1863
Feature activation-1.34e+00

Pos logit contributions
 confir+0.919
 destro+0.837
 consolidate+0.812
 rely+0.795
 consume+0.791
Neg logit contributions
club-0.756
cent-0.752
wives-0.739
horse-0.735
lines-0.721
Token with
Token ID351
Feature activation+4.41e-01

Pos logit contributions
cent+0.981
club+0.979
wives+0.958
horse+0.958
lines+0.944
Neg logit contributions
 confir-1.557
 destro-1.439
 consolidate-1.376
 rely-1.342
 consume-1.341
Token the
Token ID262
Feature activation+2.81e-01

Pos logit contributions
 confir+0.436
 destro+0.395
 consolidate+0.382
 rely+0.374
 consume+0.372
Neg logit contributions
club-0.396
cent-0.395
wives-0.389
horse-0.387
cart-0.380
Token commitment
Token ID7901
Feature activation+3.34e+00

Pos logit contributions
 confir+0.255
 destro+0.228
 consolidate+0.220
 rely+0.215
 consume+0.214
Neg logit contributions
club-0.275
cent-0.274
wives-0.270
horse-0.269
cart-0.265
Token to
Token ID284
Feature activation+1.33e+00

Pos logit contributions
 confir+3.254
 destro+2.926
 consolidate+2.853
 rely+2.800
 consume+2.776
Neg logit contributions
club-3.006
cent-2.955
wives-2.937
horse-2.912
cart-2.884
Token NATO
Token ID11332
Feature activation+7.00e+00

Pos logit contributions
 confir+1.153
 destro+1.031
 consolidate+0.988
 rely+0.960
 consume+0.957
Neg logit contributions
club-1.360
cent-1.358
wives-1.338
horse-1.333
lines-1.313
Token.
Token ID13
Feature activation-4.80e-01

Pos logit contributions
 confir+6.332
 destro+5.678
 consolidate+5.611
 rely+5.550
 consume+5.426
Neg logit contributions
cent-6.472
wives-6.436
club-6.433
laughs-6.262
horse-6.251
Token
Token ID198
Feature activation+2.81e+00

Pos logit contributions
club+0.351
cent+0.350
wives+0.344
horse+0.340
cart+0.333
Neg logit contributions
 confir-0.541
 destro-0.497
 consolidate-0.490
 rely-0.482
 consume-0.477
Token
Token ID198
Feature activation+2.52e+00

Pos logit contributions
 confir+2.877
 consolidate+2.662
 rely+2.640
 consume+2.615
 comply+2.602
Neg logit contributions
club-2.176
cent-2.158
wives-2.144
horse-2.113
lines-2.066
TokenMay
Token ID6747
Feature activation+3.62e+00

Pos logit contributions
 confir+2.847
 consolidate+2.624
 destro+2.595
 rely+2.589
 consume+2.563
Neg logit contributions
club-1.687
cent-1.665
wives-1.656
horse-1.630
lines-1.591
Token spoke
Token ID5158
Feature activation-1.98e+00

Pos logit contributions
 confir+3.247
 destro+2.870
 consolidate+2.807
 rely+2.754
 consume+2.711
Neg logit contributions
club-3.558
cent-3.546
wives-3.487
horse-3.477
cart-3.419
Token Removal
Token ID46209
Feature activation+1.81e+00

Pos logit contributions
 confir+0.458
 destro+0.419
 consolidate+0.405
 consume+0.395
 rely+0.394
Neg logit contributions
club-0.355
cent-0.353
wives-0.351
horse-0.346
cart-0.339
Token Technology
Token ID8987
Feature activation+2.09e+00

Pos logit contributions
 confir+1.960
 destro+1.797
 consolidate+1.750
 rely+1.707
 consume+1.707
Neg logit contributions
club-1.440
cent-1.434
wives-1.417
horse-1.403
cart-1.375
Token in
Token ID287
Feature activation-1.46e+00

Pos logit contributions
 confir+2.066
 destro+1.870
 consolidate+1.814
 rely+1.770
 consume+1.766
Neg logit contributions
club-1.875
cent-1.864
wives-1.846
horse-1.825
cart-1.798
Token the
Token ID262
Feature activation+6.51e-01

Pos logit contributions
club+1.198
cent+1.195
wives+1.175
horse+1.167
cart+1.146
Neg logit contributions
 confir-1.568
 destro-1.433
 consolidate-1.386
 rely-1.354
 consume-1.351
Token Twin
Token ID14968
Feature activation+2.86e+00

Pos logit contributions
 confir+0.657
 destro+0.598
 consolidate+0.577
 rely+0.562
 consume+0.562
Neg logit contributions
club-0.572
cent-0.572
wives-0.565
horse-0.559
cart-0.551
Token Fro
Token ID9734
Feature activation+6.90e+00

Pos logit contributions
 confir+3.602
 destro+3.313
 consolidate+3.271
 rely+3.228
 consume+3.224
Neg logit contributions
cent-1.713
club-1.699
wives-1.657
horse-1.640
cart-1.608
Tokenz
Token ID89
Feature activation+2.76e+00

Pos logit contributions
 confir+6.350
 consolidate+5.938
 consume+5.918
 rely+5.877
 comply+5.820
Neg logit contributions
club-5.626
wives-5.533
cent-5.523
horse-5.409
laughs-5.319
Tokenr
Token ID81
Feature activation+3.57e+00

Pos logit contributions
 confir+3.389
 destro+3.131
 consolidate+3.104
 consume+3.067
 rely+3.063
Neg logit contributions
cent-1.699
club-1.692
wives-1.660
horse-1.630
lines-1.580
Token IV
Token ID8363
Feature activation+3.25e+00

Pos logit contributions
 confir+3.502
 destro+3.168
 consolidate+3.102
 consume+3.024
 rely+3.016
Neg logit contributions
cent-3.187
club-3.178
wives-3.130
horse-3.074
cart-3.068
Token enhancing
Token ID27496
Feature activation+3.59e+00

Pos logit contributions
 confir+3.278
 destro+2.962
 consolidate+2.889
 consume+2.823
 rely+2.799
Neg logit contributions
cent-2.843
club-2.841
wives-2.814
horse-2.776
cart-2.726
Token the
Token ID262
Feature activation+1.09e+00

Pos logit contributions
 confir+3.214
 destro+2.905
 consolidate+2.813
 consume+2.734
 rely+2.721
Neg logit contributions
club-3.519
cent-3.502
wives-3.500
horse-3.430
cart-3.411
Token international
Token ID3230
Feature activation+2.37e+00

Pos logit contributions
 confir+0.626
 destro+0.559
 consolidate+0.536
 rely+0.521
 consume+0.519
Neg logit contributions
club-0.750
cent-0.748
wives-0.738
horse-0.735
cart-0.723
Token efforts
Token ID4040
Feature activation+1.46e+00

Pos logit contributions
 confir+1.637
 destro+1.419
 consolidate+1.375
 rely+1.327
 comply+1.320
Neg logit contributions
club-2.778
cent-2.768
wives-2.760
horse-2.734
cart-2.698
Token to
Token ID284
Feature activation+3.32e+00

Pos logit contributions
 confir+1.401
 destro+1.268
 consolidate+1.224
 rely+1.196
 consume+1.191
Neg logit contributions
club-1.348
cent-1.342
wives-1.324
horse-1.314
lines-1.295
Token combat
Token ID5249
Feature activation+1.57e+00

Pos logit contributions
 confir+2.209
 destro+1.908
 consolidate+1.831
 rely+1.763
 consume+1.753
Neg logit contributions
club-4.048
cent-4.042
wives-4.002
horse-3.985
cart-3.937
Token climate
Token ID4258
Feature activation+3.69e+00

Pos logit contributions
 confir+1.046
 destro+0.906
 consolidate+0.858
 rely+0.830
 consume+0.827
Neg logit contributions
club-1.918
cent-1.912
wives-1.895
horse-1.878
lines-1.857
Token change
Token ID1487
Feature activation+6.43e+00

Pos logit contributions
 confir+1.914
 destro+1.607
 consolidate+1.495
 rely+1.459
 consume+1.417
Neg logit contributions
wives-4.966
club-4.956
cent-4.936
horse-4.878
cart-4.819
Token.
Token ID13
Feature activation+2.10e-01

Pos logit contributions
 confir+6.175
 destro+5.670
 consolidate+5.513
 rely+5.462
 consume+5.388
Neg logit contributions
club-5.570
cent-5.531
wives-5.526
horse-5.427
masters-5.338
Token B
Token ID347
Feature activation-1.40e+00

Pos logit contributions
 confir+0.269
 destro+0.250
 consolidate+0.247
 rely+0.244
 consume+0.241
Neg logit contributions
club-0.121
cent-0.120
wives-0.119
horse-0.116
lines-0.113
Tokenilateral
Token ID14796
Feature activation+6.07e+00

Pos logit contributions
club+1.450
cent+1.435
wives+1.428
horse+1.417
lines+1.393
Neg logit contributions
 confir-1.201
 destro-1.067
 consolidate-1.028
 rely-0.998
 consume-0.995
Token cooperation
Token ID11113
Feature activation+2.56e+00

Pos logit contributions
 confir+5.022
 destro+4.438
 consolidate+4.389
 rely+4.296
 comply+4.250
Neg logit contributions
wives-6.115
club-6.046
cent-6.017
horse-5.954
haw-5.862
Token with
Token ID351
Feature activation-3.35e-01

Pos logit contributions
 confir+2.423
 destro+2.187
 consolidate+2.134
 rely+2.092
 consume+2.079
Neg logit contributions
club-2.387
cent-2.366
wives-2.337
horse-2.312
cart-2.280
Token need
Token ID761
Feature activation+6.95e+00

Pos logit contributions
 confir+3.530
 destro+3.054
 consolidate+2.980
 rely+2.948
 consume+2.899
Neg logit contributions
club-5.940
cent-5.919
wives-5.891
horse-5.872
cart-5.803
Token to
Token ID284
Feature activation+5.56e+00

Pos logit contributions
 confir+6.438
 destro+5.798
 consolidate+5.668
 rely+5.570
 consume+5.520
Neg logit contributions
club-6.374
wives-6.338
cent-6.293
horse-6.248
cart-6.145
Token match
Token ID2872
Feature activation+4.23e+00

Pos logit contributions
 confir+4.168
 destro+3.650
 consolidate+3.576
 rely+3.508
 consume+3.453
Neg logit contributions
club-6.256
cent-6.239
wives-6.230
horse-6.174
cart-6.110
Token each
Token ID1123
Feature activation+3.92e+00

Pos logit contributions
 confir+4.250
 destro+3.862
 consolidate+3.754
 consume+3.694
 rely+3.689
Neg logit contributions
club-3.624
wives-3.610
cent-3.588
horse-3.561
cart-3.489
Token network
Token ID3127
Feature activation+4.66e+00

Pos logit contributions
 confir+3.862
 destro+3.521
 consolidate+3.392
 consume+3.341
 rely+3.339
Neg logit contributions
club-3.467
wives-3.432
cent-3.401
horse-3.377
cart-3.329
Token change
Token ID1487
Feature activation+6.77e+00

Pos logit contributions
 confir+4.525
 destro+4.075
 consolidate+3.948
 rely+3.899
 consume+3.881
Neg logit contributions
club-4.220
wives-4.197
cent-4.137
horse-4.106
cart-4.043
Token with
Token ID351
Feature activation+1.65e+00

Pos logit contributions
 confir+6.649
 destro+5.992
 consolidate+5.853
 rely+5.813
 consume+5.785
Neg logit contributions
club-5.830
wives-5.763
horse-5.655
cent-5.639
laughs-5.590
Token a
Token ID257
Feature activation-2.73e-01

Pos logit contributions
 confir+1.715
 destro+1.560
 consolidate+1.514
 rely+1.484
 consume+1.481
Neg logit contributions
club-1.395
cent-1.388
wives-1.376
horse-1.364
cart-1.335
Token respective
Token ID11756
Feature activation+1.09e+00

Pos logit contributions
cent+0.263
club+0.263
wives+0.258
horse+0.257
lines+0.253
Neg logit contributions
 confir-0.253
 destro-0.228
 consolidate-0.219
 rely-0.213
 consume-0.212
Token fine
Token ID3734
Feature activation+5.95e+00

Pos logit contributions
 confir+1.027
 destro+0.926
 consolidate+0.891
 rely+0.869
 consume+0.867
Neg logit contributions
club-1.037
cent-1.033
wives-1.021
horse-1.016
cart-0.997
Token-
Token ID12
Feature activation-1.55e+00

Pos logit contributions
 confir+6.408
 destro+5.809
 consolidate+5.634
 consume+5.518
 rely+5.500
Neg logit contributions
cent-4.652
wives-4.627
club-4.614
horse-4.605
laughs-4.514
Token
Token ID198
Feature activation+1.48e+00

Pos logit contributions
 confir+1.735
 destro+1.579
 consolidate+1.578
 rely+1.555
 consume+1.548
Neg logit contributions
club-1.202
cent-1.199
wives-1.181
horse-1.169
lines-1.142
TokenIM
Token ID3955
Feature activation-1.19e+00

Pos logit contributions
 confir+1.732
 destro+1.596
 consolidate+1.571
 rely+1.558
 consume+1.552
Neg logit contributions
club-0.970
cent-0.965
wives-0.953
horse-0.940
lines-0.918
TokenAGES
Token ID25552
Feature activation-4.32e-01

Pos logit contributions
cent+0.897
club+0.891
wives+0.873
horse+0.871
lines+0.845
Neg logit contributions
 confir-1.352
 destro-1.236
 consolidate-1.211
 rely-1.189
 consume-1.181
Token HAVE
Token ID21515
Feature activation-5.98e-01

Pos logit contributions
club+0.333
cent+0.333
wives+0.328
horse+0.326
cart+0.318
Neg logit contributions
 confir-0.480
 destro-0.439
 consolidate-0.428
 rely-0.422
 consume-0.418
Token BE
Token ID9348
Feature activation-1.61e-01

Pos logit contributions
club+0.491
cent+0.490
wives+0.482
horse+0.480
cart+0.468
Neg logit contributions
 confir-0.638
 destro-0.582
 consolidate-0.562
 rely-0.551
 consume-0.549
TokenEN
Token ID1677
Feature activation+2.24e+00

Pos logit contributions
club+0.121
cent+0.121
wives+0.119
horse+0.118
lines+0.116
Neg logit contributions
 confir-0.182
 destro-0.167
 consolidate-0.162
 rely-0.159
 consume-0.159
Token T
Token ID309
Feature activation-6.81e-01

Pos logit contributions
 confir+2.579
 destro+2.381
 consolidate+2.308
 rely+2.268
 consume+2.259
Neg logit contributions
club-1.591
cent-1.590
wives-1.577
horse-1.565
lines-1.511
TokenAK
Token ID10206
Feature activation+3.03e+00

Pos logit contributions
club+0.349
cent+0.348
wives+0.339
horse+0.337
lines+0.326
Neg logit contributions
 confir-0.935
 destro-0.873
 consolidate-0.850
 rely-0.836
 consume-0.835
TokenEN
Token ID1677
Feature activation+2.86e+00

Pos logit contributions
 confir+3.203
 destro+2.884
 consolidate+2.875
 rely+2.842
 consume+2.834
Neg logit contributions
club-2.397
cent-2.369
wives-2.337
horse-2.310
lines-2.292
Token DOWN
Token ID30320
Feature activation+1.72e+00

Pos logit contributions
 confir+3.286
 destro+3.010
 consolidate+2.957
 rely+2.913
 consume+2.886
Neg logit contributions
cent-2.002
club-2.002
wives-1.991
horse-1.967
lines-1.900
Token UP
Token ID15958
Feature activation+1.17e+00

Pos logit contributions
 confir+1.805
 destro+1.648
 consolidate+1.603
 rely+1.577
 consume+1.568
Neg logit contributions
club-1.402
cent-1.395
wives-1.381
horse-1.362
lines-1.328
Token who
Token ID508
Feature activation+9.54e-01

Pos logit contributions
 confir+3.168
 destro+2.879
 consolidate+2.811
 rely+2.792
 consume+2.772
Neg logit contributions
club-2.035
cent-2.007
wives-2.003
horse-1.972
cart-1.955
Token had
Token ID550
Feature activation+7.65e-01

Pos logit contributions
 confir+0.847
 destro+0.760
 consolidate+0.727
 rely+0.706
 consume+0.704
Neg logit contributions
club-0.954
cent-0.954
wives-0.939
horse-0.936
lines-0.922
Token heard
Token ID2982
Feature activation+2.36e-01

Pos logit contributions
 confir+0.752
 destro+0.681
 consolidate+0.656
 rely+0.641
 consume+0.639
Neg logit contributions
club-0.692
cent-0.691
wives-0.679
horse-0.676
lines-0.664
Token of
Token ID286
Feature activation-7.94e-01

Pos logit contributions
 confir+0.271
 destro+0.249
 consolidate+0.241
 rely+0.239
 consume+0.237
Neg logit contributions
club-0.173
cent-0.172
wives-0.169
horse-0.168
cart-0.164
Token investor
Token ID15811
Feature activation+2.41e-02

Pos logit contributions
club+0.621
cent+0.620
wives+0.608
horse+0.605
lines+0.593
Neg logit contributions
 confir-0.878
 destro-0.805
 consolidate-0.778
 rely-0.761
 consume-0.760
Token and
Token ID290
Feature activation+1.69e+00

Pos logit contributions
 confir+0.030
 destro+0.028
 consolidate+0.027
 rely+0.027
 consume+0.027
Neg logit contributions
cent-0.015
club-0.015
wives-0.015
horse-0.015
lines-0.015
Token Democratic
Token ID4390
Feature activation+1.13e+00

Pos logit contributions
 confir+1.555
 destro+1.405
 consolidate+1.352
 rely+1.318
 consume+1.316
Neg logit contributions
club-1.626
cent-1.623
wives-1.606
horse-1.593
lines-1.570
Token money
Token ID1637
Feature activation+1.68e+00

Pos logit contributions
 confir+1.000
 destro+0.898
 consolidate+0.864
 rely+0.844
 consume+0.838
Neg logit contributions
club-1.114
cent-1.114
wives-1.104
horse-1.095
lines-1.080
Tokenman
Token ID805
Feature activation-1.47e+00

Pos logit contributions
 confir+1.962
 destro+1.828
 consolidate+1.775
 rely+1.744
 consume+1.731
Neg logit contributions
club-1.182
cent-1.180
wives-1.173
horse-1.134
lines-1.125
Token George
Token ID4502
Feature activation-2.05e-01

Pos logit contributions
club+0.866
cent+0.865
wives+0.840
horse+0.837
lines+0.814
Neg logit contributions
 confir-1.916
 destro-1.779
 consolidate-1.729
 rely-1.698
 consume-1.696
Token Soros
Token ID27900
Feature activation+4.77e+00

Pos logit contributions
club+0.223
cent+0.221
wives+0.220
horse+0.218
lines+0.215
Neg logit contributions
 confir-0.161
 destro-0.143
 consolidate-0.139
 rely-0.135
 consume-0.134
Token.
Token ID13
Feature activation+4.88e-01

Pos logit contributions
 confir+1.020
 destro+0.930
 consolidate+0.909
 rely+0.891
 consume+0.888
Neg logit contributions
club-0.784
cent-0.781
wives-0.769
horse-0.763
lines-0.748
TokenS
Token ID50
Feature activation+1.17e+00

Pos logit contributions
 confir+0.644
 destro+0.598
 consolidate+0.589
 rely+0.581
 consume+0.579
Neg logit contributions
club-0.267
cent-0.266
wives-0.260
horse-0.257
lines-0.249
Token.
Token ID13
Feature activation+4.56e-01

Pos logit contributions
 confir+1.251
 destro+1.139
 consolidate+1.108
 rely+1.088
 consume+1.086
Neg logit contributions
club-0.958
cent-0.952
wives-0.940
horse-0.932
lines-0.912
Token currency
Token ID7395
Feature activation+1.16e+00

Pos logit contributions
 confir+0.352
 destro+0.310
 consolidate+0.298
 rely+0.291
 consume+0.290
Neg logit contributions
club-0.507
cent-0.504
wives-0.502
horse-0.497
cart-0.491
Token,
Token ID11
Feature activation+1.64e+00

Pos logit contributions
 confir+1.097
 destro+0.985
 consolidate+0.958
 rely+0.942
 consume+0.933
Neg logit contributions
cent-1.077
club-1.075
wives-1.065
horse-1.055
cart-1.039
Token and
Token ID290
Feature activation+1.44e+00

Pos logit contributions
 confir+1.484
 destro+1.333
 consolidate+1.296
 rely+1.258
 consume+1.252
Neg logit contributions
club-1.611
cent-1.598
wives-1.585
horse-1.572
cart-1.552
Token making
Token ID1642
Feature activation-2.68e-01

Pos logit contributions
 confir+1.269
 destro+1.135
 consolidate+1.095
 rely+1.065
 consume+1.057
Neg logit contributions
club-1.458
cent-1.448
wives-1.435
horse-1.423
cart-1.404
Token a
Token ID257
Feature activation-3.07e-01

Pos logit contributions
club+0.263
cent+0.262
wives+0.258
horse+0.257
lines+0.253
Neg logit contributions
 confir-0.244
 destro-0.219
 consolidate-0.211
 rely-0.205
 consume-0.205
Token $
Token ID720
Feature activation+1.89e+00

Pos logit contributions
club+0.291
cent+0.291
wives+0.287
horse+0.285
lines+0.280
Neg logit contributions
 confir-0.288
 destro-0.260
 consolidate-0.250
 rely-0.243
 consume-0.242
Token9
Token ID24
Feature activation+1.00e+00

Pos logit contributions
 confir+2.316
 destro+2.138
 consolidate+2.122
 rely+2.094
 consume+2.084
Neg logit contributions
cent-1.142
club-1.142
wives-1.137
horse-1.111
lines-1.082
Token billion
Token ID2997
Feature activation+5.26e+00

Pos logit contributions
 confir+1.070
 destro+0.977
 consolidate+0.948
 rely+0.928
 consume+0.924
Neg logit contributions
club-0.822
cent-0.818
wives-0.809
horse-0.801
lines-0.790
Token tenure
Token ID17081
Feature activation+2.62e+00

Pos logit contributions
 confir+2.254
 destro+2.010
 consolidate+1.986
 rely+1.931
 consume+1.925
Neg logit contributions
club-2.217
cent-2.187
wives-2.177
horse-2.162
cart-2.125
Token,
Token ID11
Feature activation+1.83e+00

Pos logit contributions
 confir+2.617
 destro+2.371
 consolidate+2.327
 rely+2.276
 consume+2.271
Neg logit contributions
club-2.276
cent-2.265
wives-2.232
horse-2.196
cart-2.186
Token and
Token ID290
Feature activation+1.36e+00

Pos logit contributions
 confir+1.708
 destro+1.525
 consolidate+1.489
 rely+1.452
 consume+1.445
Neg logit contributions
club-1.760
cent-1.740
wives-1.720
horse-1.711
cart-1.678
Token characterized
Token ID16264
Feature activation-3.01e-01

Pos logit contributions
 confir+1.345
 destro+1.207
 consolidate+1.172
 rely+1.147
 consume+1.137
Neg logit contributions
club-1.233
cent-1.223
wives-1.211
horse-1.203
cart-1.179
Token Hag
Token ID21375
Feature activation+5.76e+00

Pos logit contributions
club+0.214
cent+0.213
wives+0.209
horse+0.207
cart+0.203
Neg logit contributions
 confir-0.354
 destro-0.326
 consolidate-0.317
 rely-0.311
 consume-0.311
Tokenel
Token ID417
Feature activation+4.51e+00

Pos logit contributions
 confir+6.597
 consolidate+6.221
 rely+6.133
 consume+6.085
 comply+6.074
Neg logit contributions
club-3.686
wives-3.631
cent-3.607
horse-3.535
laughs-3.436
Token's
Token ID338
Feature activation-9.23e-01

Pos logit contributions
 confir+3.845
 destro+3.412
 consolidate+3.358
 rely+3.268
 consume+3.227
Neg logit contributions
club-4.534
cent-4.492
wives-4.417
horse-4.406
laughs-4.358
Token resignation
Token ID18525
Feature activation+2.78e+00

Pos logit contributions
club+0.938
cent+0.936
wives+0.922
horse+0.919
lines+0.905
Neg logit contributions
 confir-0.804
 destro-0.719
 consolidate-0.689
 rely-0.670
 consume-0.668
Token as
Token ID355
Feature activation+2.09e+00

Pos logit contributions
 confir+2.517
 destro+2.240
 consolidate+2.200
 rely+2.145
 consume+2.133
Neg logit contributions
club-2.697
cent-2.687
wives-2.646
horse-2.630
cart-2.590
Token his
Token ID465
Feature activation+2.19e+00

Pos logit contributions
 confir+2.296
 destro+2.096
 consolidate+2.058
 rely+2.007
 consume+1.995
Neg logit contributions
club-1.645
cent-1.630
wives-1.605
horse-1.591
cart-1.568
Token own
Token ID898
Feature activation-6.99e-01

Pos logit contributions
 confir+2.283
 destro+2.074
 consolidate+2.035
 rely+1.982
 consume+1.973
Neg logit contributions
club-1.835
cent-1.819
wives-1.796
horse-1.788
cart-1.763
Token cor
Token ID1162
Feature activation-4.35e-01

Pos logit contributions
 confir+1.740
 destro+1.598
 consolidate+1.538
 rely+1.517
 consume+1.508
Neg logit contributions
club-1.419
cent-1.414
wives-1.395
horse-1.385
lines-1.373
Tokenned
Token ID2817
Feature activation-1.67e+00

Pos logit contributions
club+0.283
cent+0.282
wives+0.277
horse+0.277
cart+0.268
Neg logit contributions
 confir-0.533
 destro-0.494
 consolidate-0.479
 rely-0.473
 consume-0.471
Token beef
Token ID12023
Feature activation-1.23e+00

Pos logit contributions
club+1.512
cent+1.511
wives+1.484
horse+1.478
lines+1.452
Neg logit contributions
 confir-1.640
 destro-1.487
 consolidate-1.431
 rely-1.396
 consume-1.393
Token and
Token ID290
Feature activation+1.60e+00

Pos logit contributions
cent+1.041
club+1.040
wives+1.021
horse+1.015
lines+0.998
Neg logit contributions
 confir-1.273
 destro-1.162
 consolidate-1.120
 rely-1.097
 consume-1.095
Token thr
Token ID5636
Feature activation-7.38e-01

Pos logit contributions
 confir+1.657
 destro+1.513
 consolidate+1.463
 rely+1.433
 consume+1.429
Neg logit contributions
club-1.357
cent-1.349
wives-1.335
horse-1.311
lines-1.299
Tokenice
Token ID501
Feature activation+1.13e+00

Pos logit contributions
cent+0.536
club+0.536
wives+0.526
horse+0.524
lines+0.516
Neg logit contributions
 confir-0.842
 destro-0.774
 consolidate-0.759
 rely-0.746
 consume-0.738
Token-
Token ID12
Feature activation-1.95e+00

Pos logit contributions
 confir+1.240
 destro+1.139
 consolidate+1.104
 rely+1.094
 consume+1.091
Neg logit contributions
club-0.880
cent-0.869
wives-0.857
horse-0.843
lines-0.838
Tokencooked
Token ID46591
Feature activation+1.22e+00

Pos logit contributions
club+1.361
cent+1.359
wives+1.333
horse+1.321
lines+1.296
Neg logit contributions
 confir-2.299
 destro-2.124
 consolidate-2.060
 consume-2.020
 rely-2.020
Token bacon
Token ID21385
Feature activation-4.95e-01

Pos logit contributions
 confir+1.226
 destro+1.115
 consolidate+1.079
 rely+1.054
 consume+1.051
Neg logit contributions
club-1.069
cent-1.065
wives-1.047
horse-1.037
lines-1.021
Token tossed
Token ID22591
Feature activation-1.20e+00

Pos logit contributions
club+0.412
cent+0.411
wives+0.403
horse+0.401
lines+0.395
Neg logit contributions
 confir-0.523
 destro-0.477
 consolidate-0.461
 rely-0.452
 consume-0.451
Token with
Token ID351
Feature activation-1.86e+00

Pos logit contributions
club+0.957
cent+0.954
wives+0.937
horse+0.931
lines+0.913
Neg logit contributions
 confir-1.315
 destro-1.205
 consolidate-1.166
 rely-1.143
 consume-1.139
Token Aviation
Token ID25186
Feature activation-3.07e+00

Pos logit contributions
 confir+0.421
 destro+0.381
 consolidate+0.368
 rely+0.361
 consume+0.360
Neg logit contributions
club-0.383
cent-0.379
wives-0.375
horse-0.373
cart-0.366
Token Week
Token ID6119
Feature activation-8.06e-01

Pos logit contributions
club+2.063
cent+2.061
wives+2.012
horse+2.004
lines+1.957
Neg logit contributions
 confir-3.724
 destro-3.441
 consolidate-3.338
 rely-3.272
 consume-3.267
Token
Token ID447
Feature activation-1.81e+00

Pos logit contributions
club+0.717
cent+0.713
wives+0.702
horse+0.697
cart+0.686
Neg logit contributions
 confir-0.812
 destro-0.734
 consolidate-0.706
 rely-0.693
 consume-0.689
Token
Token ID247
Feature activation-9.78e-01

Pos logit contributions
club+1.083
cent+1.081
wives+1.056
horse+1.046
lines+1.019
Neg logit contributions
 confir-2.339
 destro-2.166
 consolidate-2.112
 rely-2.077
 consume-2.074
Tokens
Token ID82
Feature activation-2.12e+00

Pos logit contributions
club+0.522
cent+0.520
wives+0.511
horse+0.502
cart+0.484
Neg logit contributions
 confir-1.324
 destro-1.228
 consolidate-1.195
 rely-1.179
 consume-1.176
Token Lara
Token ID29849
Feature activation-1.93e+00

Pos logit contributions
club+1.285
cent+1.281
wives+1.250
horse+1.241
lines+1.208
Neg logit contributions
 confir-2.722
 destro-2.527
 consolidate-2.455
 rely-2.413
 consume-2.408
Token Sel
Token ID15300
Feature activation-4.30e-01

Pos logit contributions
club+1.299
cent+1.289
wives+1.264
horse+1.255
lines+1.225
Neg logit contributions
 confir-2.338
 destro-2.161
 consolidate-2.100
 rely-2.063
 consume-2.056
Tokenig
Token ID328
Feature activation-9.41e-01

Pos logit contributions
club+0.297
cent+0.294
wives+0.289
horse+0.286
lines+0.282
Neg logit contributions
 confir-0.508
 destro-0.470
 consolidate-0.458
 rely-0.449
 consume-0.448
Tokenman
Token ID805
Feature activation-2.29e+00

Pos logit contributions
club+0.628
cent+0.623
wives+0.612
horse+0.607
lines+0.593
Neg logit contributions
 confir-1.153
 destro-1.063
 consolidate-1.027
 rely-1.012
 consume-1.006
Token reminded
Token ID14516
Feature activation+1.84e+00

Pos logit contributions
club+1.983
cent+1.978
wives+1.945
horse+1.935
lines+1.898
Neg logit contributions
 confir-2.342
 destro-2.130
 consolidate-2.052
 rely-2.005
 consume-2.000
Token folks
Token ID7974
Feature activation-2.50e+00

Pos logit contributions
 confir+1.943
 destro+1.765
 consolidate+1.719
 rely+1.701
 consume+1.687
Neg logit contributions
club-1.509
cent-1.505
wives-1.484
horse-1.469
lines-1.438
Token,
Token ID11
Feature activation+3.34e-01

Pos logit contributions
cent+2.680
club+2.665
wives+2.621
horse+2.612
lines+2.568
Neg logit contributions
 confir-3.200
 destro-2.914
 consolidate-2.806
 rely-2.735
 consume-2.730
Token the
Token ID262
Feature activation-1.13e+00

Pos logit contributions
 confir+0.341
 destro+0.310
 consolidate+0.300
 rely+0.294
 consume+0.293
Neg logit contributions
club-0.289
cent-0.287
wives-0.283
horse-0.282
cart-0.276
Token region
Token ID3814
Feature activation-4.40e+00

Pos logit contributions
cent+1.074
club+1.073
wives+1.055
horse+1.051
lines+1.037
Neg logit contributions
 confir-1.062
 destro-0.959
 consolidate-0.919
 rely-0.895
 consume-0.893
Token near
Token ID1474
Feature activation-4.06e+00

Pos logit contributions
cent+3.793
club+3.752
horse+3.706
wives+3.702
lines+3.671
Neg logit contributions
 confir-4.477
 destro-4.082
 consolidate-3.916
 rely-3.791
 comply-3.788
Token the
Token ID262
Feature activation-4.89e-01

Pos logit contributions
cent+3.121
club+3.107
horse+3.045
wives+3.039
lines+2.995
Neg logit contributions
 confir-4.526
 destro-4.147
 consolidate-4.009
 rely-3.922
 consume-3.906
Token cal
Token ID2386
Feature activation-4.42e-01

Pos logit contributions
club+0.470
cent+0.469
wives+0.463
horse+0.460
lines+0.452
Neg logit contributions
 confir-0.453
 destro-0.409
 consolidate-0.392
 rely-0.382
 consume-0.382
Tokenving
Token ID1075
Feature activation+2.61e-02

Pos logit contributions
club+0.252
cent+0.251
wives+0.247
horse+0.243
lines+0.236
Neg logit contributions
 confir-0.579
 destro-0.539
 consolidate-0.527
 rely-0.517
 consume-0.516
Token front
Token ID2166
Feature activation-1.75e+00

Pos logit contributions
 confir+0.024
 destro+0.022
 consolidate+0.021
 rely+0.020
 consume+0.020
Neg logit contributions
club-0.025
cent-0.025
wives-0.025
horse-0.025
cart-0.024
Token shows
Token ID2523
Feature activation-3.58e+00

Pos logit contributions
club+1.713
cent+1.712
wives+1.682
horse+1.678
lines+1.658
Neg logit contributions
 confir-1.584
 destro-1.423
 consolidate-1.361
 rely-1.323
 consume-1.318
Token a
Token ID257
Feature activation-2.41e+00

Pos logit contributions
cent+3.071
club+3.068
horse+3.003
wives+3.001
lines+2.961
Neg logit contributions
 confir-3.662
 destro-3.334
 consolidate-3.203
 rely-3.128
 consume-3.114
Token dramatic
Token ID10092
Feature activation-3.35e+00

Pos logit contributions
cent+2.289
club+2.285
wives+2.241
horse+2.239
lines+2.211
Neg logit contributions
 confir-2.263
 destro-2.039
 consolidate-1.955
 rely-1.907
 consume-1.899
Token was
Token ID373
Feature activation-2.11e+00

Pos logit contributions
 confir+0.111
 destro+0.099
 consolidate+0.095
 rely+0.093
 consume+0.092
Neg logit contributions
club-0.125
cent-0.125
wives-0.123
horse-0.122
lines-0.120
Token again
Token ID757
Feature activation+1.53e+00

Pos logit contributions
club+1.812
cent+1.810
wives+1.774
horse+1.769
lines+1.736
Neg logit contributions
 confir-2.166
 destro-1.973
 consolidate-1.901
 rely-1.855
 consume-1.852
Token Rob
Token ID3851
Feature activation+9.41e-01

Pos logit contributions
 confir+1.766
 destro+1.630
 consolidate+1.582
 rely+1.565
 consume+1.545
Neg logit contributions
cent-1.108
club-1.102
wives-1.094
horse-1.075
lines-1.050
Tokenben
Token ID11722
Feature activation+3.46e+00

Pos logit contributions
 confir+1.162
 destro+1.075
 consolidate+1.041
 rely+1.021
 consume+1.020
Neg logit contributions
club-0.619
cent-0.618
wives-0.603
horse-0.601
lines-0.585
Token who
Token ID508
Feature activation-2.39e-02

Pos logit contributions
 confir+3.680
 destro+3.362
 consolidate+3.268
 rely+3.223
 consume+3.182
Neg logit contributions
club-2.763
wives-2.723
cent-2.711
horse-2.679
lines-2.639
Token wre
Token ID30154
Feature activation-3.21e+00

Pos logit contributions
club+0.025
cent+0.025
wives+0.024
horse+0.024
lines+0.024
Neg logit contributions
 confir-0.020
 destro-0.018
 consolidate-0.017
 rely-0.017
 consume-0.017
Tokenaked
Token ID4335
Feature activation-1.09e+00

Pos logit contributions
cent+2.385
club+2.384
horse+2.322
wives+2.321
lines+2.264
Neg logit contributions
 confir-3.690
 destro-3.378
 consolidate-3.270
 rely-3.199
 consume-3.197
Token havoc
Token ID29860
Feature activation-8.39e-01

Pos logit contributions
club+0.853
cent+0.850
wives+0.838
horse+0.829
lines+0.813
Neg logit contributions
 confir-1.201
 destro-1.103
 consolidate-1.067
 rely-1.044
 consume-1.041
Token in
Token ID287
Feature activation-2.09e+00

Pos logit contributions
club+0.701
cent+0.700
wives+0.689
horse+0.683
lines+0.671
Neg logit contributions
 confir-0.880
 destro-0.804
 consolidate-0.779
 rely-0.762
 consume-0.759
Token the
Token ID262
Feature activation-1.25e+00

Pos logit contributions
club+1.700
cent+1.692
wives+1.677
horse+1.659
cart+1.626
Neg logit contributions
 confir-2.252
 destro-2.066
 consolidate-2.001
 rely-1.956
 consume-1.947
Token Mexico
Token ID5828
Feature activation+2.18e+00

Pos logit contributions
club+1.066
cent+1.063
wives+1.046
horse+1.041
lines+1.021
Neg logit contributions
 confir-1.296
 destro-1.182
 consolidate-1.141
 rely-1.115
 consume-1.112
Token in
Token ID287
Feature activation-1.12e+00

Pos logit contributions
 confir+0.908
 destro+0.829
 consolidate+0.805
 rely+0.787
 consume+0.787
Neg logit contributions
club-0.714
cent-0.714
wives-0.702
horse-0.695
cart-0.682
Token the
Token ID262
Feature activation+2.07e-01

Pos logit contributions
club+0.885
cent+0.882
wives+0.872
horse+0.861
cart+0.844
Neg logit contributions
 confir-1.225
 destro-1.122
 consolidate-1.087
 rely-1.062
 consume-1.061
Token jam
Token ID16853
Feature activation-1.25e+00

Pos logit contributions
 confir+0.216
 destro+0.197
 consolidate+0.191
 rely+0.186
 consume+0.186
Neg logit contributions
club-0.174
cent-0.173
wives-0.171
horse-0.170
lines-0.166
Token on
Token ID319
Feature activation-1.70e+00

Pos logit contributions
club+1.009
cent+1.008
wives+0.989
horse+0.984
lines+0.964
Neg logit contributions
 confir-1.352
 destro-1.238
 consolidate-1.197
 rely-1.170
 consume-1.168
Token a
Token ID257
Feature activation-2.09e+00

Pos logit contributions
club+1.329
cent+1.327
wives+1.301
horse+1.294
lines+1.267
Neg logit contributions
 confir-1.875
 destro-1.718
 consolidate-1.662
 rely-1.626
 consume-1.623
Token media
Token ID2056
Feature activation-1.51e+00

Pos logit contributions
club+1.804
cent+1.803
wives+1.768
horse+1.764
lines+1.732
Neg logit contributions
 confir-2.138
 destro-1.945
 consolidate-1.876
 rely-1.832
 consume-1.828
Token bus
Token ID1323
Feature activation-2.64e+00

Pos logit contributions
club+1.334
cent+1.332
wives+1.313
horse+1.302
lines+1.280
Neg logit contributions
 confir-1.527
 destro-1.388
 consolidate-1.338
 rely-1.307
 consume-1.304
Token)
Token ID8
Feature activation+3.72e-01

Pos logit contributions
club+2.106
cent+2.104
wives+2.059
horse+2.056
lines+2.016
Neg logit contributions
 confir-2.880
 destro-2.636
 consolidate-2.545
 rely-2.488
 consume-2.483
Token pic
Token ID8301
Feature activation-3.34e+00

Pos logit contributions
 confir+0.419
 destro+0.385
 consolidate+0.373
 rely+0.366
 consume+0.365
Neg logit contributions
club-0.283
cent-0.283
wives-0.277
horse-0.276
cart-0.270
Token.
Token ID13
Feature activation-1.66e+00

Pos logit contributions
cent+2.538
club+2.523
horse+2.470
wives+2.457
lines+2.418
Neg logit contributions
 confir-3.932
 destro-3.650
 consolidate-3.413
 consume-3.307
 rely-3.302
Tokentwitter
Token ID6956
Feature activation-8.56e-01

Pos logit contributions
cent+1.096
club+1.095
wives+1.070
horse+1.062
lines+1.036
Neg logit contributions
 confir-2.019
 destro-1.865
 consolidate-1.822
 rely-1.790
 consume-1.787
Token's
Token ID338
Feature activation-2.35e+00

Pos logit contributions
 confir+0.670
 destro+0.603
 consolidate+0.579
 rely+0.566
 consume+0.562
Neg logit contributions
club-0.693
cent-0.691
wives-0.683
horse-0.677
lines-0.667
Token photos
Token ID5205
Feature activation-5.23e+00

Pos logit contributions
club+2.029
cent+2.022
wives+1.991
horse+1.979
lines+1.940
Neg logit contributions
 confir-2.416
 destro-2.200
 consolidate-2.121
 rely-2.074
 consume-2.067
Token,
Token ID11
Feature activation-7.12e-01

Pos logit contributions
cent+4.857
club+4.809
horse+4.731
wives+4.712
lines+4.668
Neg logit contributions
 confir-4.964
 destro-4.514
 consolidate-4.298
 consume-4.152
 rely-4.146
Token click
Token ID3904
Feature activation-2.98e+00

Pos logit contributions
club+0.719
cent+0.717
wives+0.708
horse+0.705
cart+0.692
Neg logit contributions
 confir-0.626
 destro-0.558
 consolidate-0.538
 rely-0.525
 consume-0.523
Token here
Token ID994
Feature activation-1.46e-01

Pos logit contributions
club+1.746
cent+1.744
wives+1.697
horse+1.687
lines+1.642
Neg logit contributions
 confir-3.888
 destro-3.614
 consolidate-3.513
 rely-3.450
 consume-3.444
Token:
Token ID25
Feature activation-3.46e+00

Pos logit contributions
club+0.115
cent+0.115
wives+0.113
horse+0.112
lines+0.110
Neg logit contributions
 confir-0.160
 destro-0.146
 consolidate-0.142
 rely-0.139
 consume-0.139
Token
Token ID198
Feature activation+1.14e+00

Pos logit contributions
club+2.377
cent+2.376
wives+2.321
horse+2.309
lines+2.257
Neg logit contributions
 confir-4.160
 destro-3.842
 consolidate-3.726
 rely-3.651
 consume-3.646
Token
Token ID198
Feature activation+1.10e+00

Pos logit contributions
 confir+1.263
 destro+1.152
 consolidate+1.142
 rely+1.124
 consume+1.119
Neg logit contributions
club-0.852
cent-0.850
wives-0.836
horse-0.829
lines-0.809
TokenFollow
Token ID7155
Feature activation-3.47e+00

Pos logit contributions
 confir+1.307
 destro+1.203
 consolidate+1.186
 rely+1.172
 consume+1.160
Neg logit contributions
club-0.724
cent-0.714
wives-0.712
horse-0.696
lines-0.679
Token @
Token ID2488
Feature activation-5.11e+00

Pos logit contributions
club+2.465
cent+2.461
wives+2.411
horse+2.394
lines+2.340
Neg logit contributions
 confir-4.088
 destro-3.770
 consolidate-3.652
 rely-3.581
 consume-3.573
TokenPro
Token ID2964
Feature activation-2.12e+00

Pos logit contributions
cent+3.167
club+3.157
wives+3.069
horse+3.069
lines+2.990
Neg logit contributions
 confir-6.489
 destro-6.018
 consolidate-5.846
 rely-5.734
 consume-5.733
Token's
Token ID338
Feature activation-2.95e+00

Pos logit contributions
 confir+2.873
 destro+2.621
 consolidate+2.526
 consume+2.479
 rely+2.478
Neg logit contributions
club-2.973
cent-2.940
wives-2.895
horse-2.885
lines-2.856
Token origin
Token ID8159
Feature activation-3.34e+00

Pos logit contributions
cent+2.748
club+2.748
wives+2.701
horse+2.690
lines+2.645
Neg logit contributions
 confir-2.821
 destro-2.547
 consolidate-2.447
 rely-2.383
 consume-2.378
Token as
Token ID355
Feature activation-2.55e+00

Pos logit contributions
club+2.369
cent+2.363
wives+2.311
horse+2.300
lines+2.250
Neg logit contributions
 confir-3.938
 destro-3.633
 consolidate-3.525
 rely-3.457
 consume-3.451
Token a
Token ID257
Feature activation-3.10e+00

Pos logit contributions
club+1.895
cent+1.892
wives+1.853
horse+1.844
lines+1.805
Neg logit contributions
 confir-2.921
 destro-2.687
 consolidate-2.601
 rely-2.549
 consume-2.544
Token musician
Token ID21623
Feature activation-5.90e+00

Pos logit contributions
club+2.715
cent+2.714
wives+2.664
horse+2.656
lines+2.605
Neg logit contributions
 confir-3.132
 destro-2.846
 consolidate-2.742
 rely-2.672
 consume-2.667
Token at
Token ID379
Feature activation-4.86e+00

Pos logit contributions
cent+4.747
club+4.734
horse+4.631
wives+4.629
lines+4.520
Neg logit contributions
 confir-6.338
 destro-5.805
 consolidate-5.577
 rely-5.416
 consume-5.416
Token a
Token ID257
Feature activation-2.64e+00

Pos logit contributions
club+3.630
cent+3.627
wives+3.547
horse+3.527
lines+3.459
Neg logit contributions
 confir-5.530
 destro-5.087
 consolidate-4.908
 rely-4.797
 consume-4.796
Token piano
Token ID19132
Feature activation-4.01e+00

Pos logit contributions
club+2.261
cent+2.257
wives+2.215
horse+2.207
lines+2.166
Neg logit contributions
 confir-2.723
 destro-2.480
 consolidate-2.389
 rely-2.334
 consume-2.329
Token bar
Token ID2318
Feature activation-3.78e+00

Pos logit contributions
club+3.362
cent+3.351
wives+3.285
horse+3.269
lines+3.210
Neg logit contributions
 confir-4.225
 destro-3.857
 consolidate-3.711
 rely-3.624
 consume-3.623
Token,
Token ID11
Feature activation-1.39e+00

Pos logit contributions
club+3.132
cent+3.127
wives+3.069
horse+3.051
lines+2.998
Neg logit contributions
 confir-4.005
 destro-3.655
 consolidate-3.525
 rely-3.444
 consume-3.437
Token who
Token ID508
Feature activation-2.98e+00

Pos logit contributions
club+1.212
cent+1.210
wives+1.189
horse+1.184
lines+1.162
Neg logit contributions
 confir-1.413
 destro-1.286
 consolidate-1.240
 rely-1.212
 consume-1.210
Token who
Token ID508
Feature activation-1.98e-01

Pos logit contributions
 confir+0.373
 destro+0.340
 consolidate+0.331
 rely+0.324
 consume+0.323
Neg logit contributions
club-0.291
cent-0.288
wives-0.285
horse-0.282
lines-0.277
Token showed
Token ID3751
Feature activation-9.98e-01

Pos logit contributions
club+0.202
cent+0.201
wives+0.198
horse+0.198
lines+0.195
Neg logit contributions
 confir-0.172
 destro-0.154
 consolidate-0.147
 rely-0.143
 consume-0.142
Token none
Token ID4844
Feature activation+1.98e+00

Pos logit contributions
club+1.001
cent+1.001
wives+0.991
horse+0.981
lines+0.967
Neg logit contributions
 confir-0.881
 destro-0.788
 consolidate-0.761
 rely-0.741
 consume-0.737
Token,
Token ID11
Feature activation+7.54e-01

Pos logit contributions
 confir+2.237
 destro+2.041
 consolidate+1.993
 rely+1.972
 consume+1.959
Neg logit contributions
club-1.469
cent-1.459
wives-1.443
horse-1.429
cart-1.405
Token is
Token ID318
Feature activation-2.41e+00

Pos logit contributions
 confir+0.648
 destro+0.579
 consolidate+0.556
 rely+0.541
 consume+0.538
Neg logit contributions
club-0.775
cent-0.772
wives-0.762
horse-0.758
lines-0.746
Token atop
Token ID20156
Feature activation-4.66e+00

Pos logit contributions
cent+2.329
club+2.318
wives+2.282
horse+2.274
lines+2.238
Neg logit contributions
 confir-2.220
 destro-1.998
 consolidate-1.904
 rely-1.853
 consume-1.851
Token.
Token ID13
Feature activation-2.26e+00

Pos logit contributions
cent+3.697
club+3.677
horse+3.616
wives+3.600
lines+3.551
Neg logit contributions
 confir-5.067
 destro-4.652
 consolidate-4.465
 rely-4.377
 consume-4.372
Token Such
Token ID8013
Feature activation-1.08e-01

Pos logit contributions
club+1.463
cent+1.460
wives+1.427
horse+1.415
lines+1.382
Neg logit contributions
 confir-2.794
 destro-2.587
 consolidate-2.518
 rely-2.472
 consume-2.465
Token are
Token ID389
Feature activation-2.72e+00

Pos logit contributions
club+0.104
cent+0.104
wives+0.102
horse+0.102
lines+0.100
Neg logit contributions
 confir-0.101
 destro-0.091
 consolidate-0.087
 rely-0.085
 consume-0.085
Token the
Token ID262
Feature activation-6.99e-01

Pos logit contributions
cent+2.117
club+2.115
horse+2.074
wives+2.069
lines+2.027
Neg logit contributions
 confir-3.004
 destro-2.757
 consolidate-2.656
 rely-2.598
 consume-2.596
Token wages
Token ID9400
Feature activation-2.08e+00

Pos logit contributions
club+0.723
cent+0.722
wives+0.711
horse+0.709
lines+0.698
Neg logit contributions
 confir-0.598
 destro-0.533
 consolidate-0.510
 rely-0.495
 consume-0.494
Tokenum
Token ID388
Feature activation-1.35e-01

Pos logit contributions
club+0.614
cent+0.614
wives+0.605
horse+0.602
lines+0.592
Neg logit contributions
 confir-0.788
 destro-0.721
 consolidate-0.702
 rely-0.685
 consume-0.682
Token op
Token ID1034
Feature activation-2.79e+00

Pos logit contributions
club+0.115
cent+0.114
wives+0.114
horse+0.113
cart+0.111
Neg logit contributions
 confir-0.138
 destro-0.126
 consolidate-0.122
 rely-0.120
 consume-0.119
Tokenus
Token ID385
Feature activation-2.21e+00

Pos logit contributions
cent+1.717
club+1.716
wives+1.660
horse+1.658
cart+1.612
Neg logit contributions
 confir-3.545
 destro-3.293
 consolidate-3.196
 rely-3.140
 consume-3.129
Token.
Token ID13
Feature activation-2.36e+00

Pos logit contributions
club+1.796
cent+1.794
wives+1.761
horse+1.752
lines+1.719
Neg logit contributions
 confir-2.367
 destro-2.164
 consolidate-2.091
 rely-2.045
 consume-2.041
Token A
Token ID317
Feature activation-3.79e+00

Pos logit contributions
club+1.287
cent+1.281
wives+1.250
horse+1.238
lines+1.203
Neg logit contributions
 confir-3.146
 destro-2.934
 consolidate-2.868
 rely-2.823
 consume-2.814
Token mel
Token ID7758
Feature activation-4.93e+00

Pos logit contributions
cent+3.371
club+3.369
wives+3.299
horse+3.290
lines+3.227
Neg logit contributions
 confir-3.777
 destro-3.427
 consolidate-3.300
 rely-3.218
 consume-3.210
Tokenodic
Token ID29512
Feature activation-3.55e+00

Pos logit contributions
cent+4.093
club+4.018
horse+3.943
wives+3.933
cart+3.863
Neg logit contributions
 confir-5.257
 destro-4.796
 consolidate-4.593
 consume-4.481
 rely-4.466
Token and
Token ID290
Feature activation-3.04e-02

Pos logit contributions
club+3.185
cent+3.182
horse+3.116
wives+3.112
lines+3.059
Neg logit contributions
 confir-3.505
 destro-3.176
 consolidate-3.048
 consume-2.967
 rely-2.965
Token melancholy
Token ID47886
Feature activation-1.90e+00

Pos logit contributions
club+0.028
cent+0.028
wives+0.028
horse+0.028
lines+0.027
Neg logit contributions
 confir-0.029
 destro-0.026
 consolidate-0.025
 rely-0.025
 consume-0.025
Token meditation
Token ID16901
Feature activation-2.46e+00

Pos logit contributions
club+1.655
cent+1.653
wives+1.620
horse+1.617
lines+1.586
Neg logit contributions
 confir-1.940
 destro-1.764
 consolidate-1.698
 rely-1.655
 consume-1.654
Token on
Token ID319
Feature activation-2.99e+00

Pos logit contributions
club+1.981
cent+1.979
wives+1.940
horse+1.932
lines+1.894
Neg logit contributions
 confir-2.656
 destro-2.431
 consolidate-2.348
 rely-2.296
 consume-2.291
Token
Token ID251
Feature activation-1.32e+00

Pos logit contributions
club+1.731
cent+1.730
horse+1.686
wives+1.681
lines+1.637
Neg logit contributions
 confir-3.914
 destro-3.657
 consolidate-3.538
 rely-3.460
 consume-3.458
Token Moore
Token ID8877
Feature activation+2.94e-01

Pos logit contributions
club+1.169
cent+1.166
wives+1.148
horse+1.143
lines+1.120
Neg logit contributions
 confir-1.315
 destro-1.195
 consolidate-1.152
 rely-1.125
 consume-1.122
Token said
Token ID531
Feature activation-2.39e+00

Pos logit contributions
 confir+0.282
 destro+0.255
 consolidate+0.245
 rely+0.240
 consume+0.238
Neg logit contributions
club-0.274
cent-0.273
wives-0.269
horse-0.268
cart-0.263
Token in
Token ID287
Feature activation-3.40e+00

Pos logit contributions
cent+1.969
club+1.969
wives+1.930
horse+1.921
lines+1.886
Neg logit contributions
 confir-2.544
 destro-2.324
 consolidate-2.244
 rely-2.191
 consume-2.188
Token an
Token ID281
Feature activation-2.33e+00

Pos logit contributions
club+2.679
cent+2.667
wives+2.624
horse+2.609
lines+2.549
Neg logit contributions
 confir-3.746
 destro-3.432
 consolidate-3.321
 rely-3.252
 consume-3.243
Token interview
Token ID2720
Feature activation-4.85e+00

Pos logit contributions
cent+1.780
club+1.778
wives+1.735
horse+1.728
lines+1.703
Neg logit contributions
 confir-2.615
 destro-2.409
 consolidate-2.326
 rely-2.270
 consume-2.269
Token last
Token ID938
Feature activation-1.10e+00

Pos logit contributions
cent+3.862
club+3.840
wives+3.766
horse+3.761
lines+3.696
Neg logit contributions
 confir-5.227
 destro-4.806
 consolidate-4.638
 rely-4.515
 consume-4.513
Token year
Token ID614
Feature activation-1.16e+00

Pos logit contributions
club+0.878
cent+0.877
wives+0.860
horse+0.855
lines+0.837
Neg logit contributions
 confir-1.205
 destro-1.103
 consolidate-1.068
 rely-1.046
 consume-1.043
Token,
Token ID11
Feature activation-5.89e-01

Pos logit contributions
club+0.955
cent+0.952
wives+0.937
horse+0.931
lines+0.913
Neg logit contributions
 confir-1.235
 destro-1.128
 consolidate-1.090
 rely-1.067
 consume-1.065
Token
Token ID564
Feature activation-2.91e+00

Pos logit contributions
club+0.487
cent+0.484
wives+0.477
horse+0.475
lines+0.464
Neg logit contributions
 confir-0.627
 destro-0.573
 consolidate-0.553
 rely-0.543
 consume-0.541
Token
Token ID250
Feature activation-1.39e-01

Pos logit contributions
club+1.796
cent+1.794
wives+1.743
horse+1.737
lines+1.699
Neg logit contributions
 confir-3.695
 destro-3.435
 consolidate-3.327
 rely-3.262
 consume-3.258
Token sensational
Token ID31826
Feature activation-3.65e+00

Pos logit contributions
cent+2.455
club+2.452
wives+2.398
horse+2.390
lines+2.341
Neg logit contributions
 confir-3.397
 destro-3.108
 consolidate-3.002
 rely-2.937
 consume-2.932
Token black
Token ID2042
Feature activation-3.62e+00

Pos logit contributions
cent+3.107
club+3.105
wives+3.038
horse+3.030
lines+2.972
Neg logit contributions
 confir-3.785
 destro-3.442
 consolidate-3.315
 rely-3.241
 consume-3.235
Token and
Token ID290
Feature activation+1.87e-01

Pos logit contributions
club+2.547
cent+2.545
wives+2.481
horse+2.479
lines+2.416
Neg logit contributions
 confir-4.277
 destro-3.946
 consolidate-3.819
 rely-3.742
 consume-3.733
Token white
Token ID2330
Feature activation-2.28e+00

Pos logit contributions
 confir+0.161
 destro+0.144
 consolidate+0.138
 rely+0.134
 consume+0.134
Neg logit contributions
club-0.191
cent-0.190
wives-0.188
horse-0.187
lines-0.184
Token love
Token ID1842
Feature activation-2.11e+00

Pos logit contributions
club+1.956
cent+1.955
wives+1.917
horse+1.911
lines+1.875
Neg logit contributions
 confir-2.341
 destro-2.133
 consolidate-2.054
 rely-2.004
 consume-2.000
Token letter
Token ID3850
Feature activation-4.61e+00

Pos logit contributions
club+1.513
cent+1.512
wives+1.483
horse+1.469
lines+1.440
Neg logit contributions
 confir-2.468
 destro-2.271
 consolidate-2.202
 rely-2.161
 consume-2.156
Token to
Token ID284
Feature activation-1.60e-01

Pos logit contributions
cent+4.220
club+4.206
wives+4.120
horse+4.105
lines+4.051
Neg logit contributions
 confir-4.450
 destro-4.026
 consolidate-3.874
 consume-3.761
 rely-3.761
Token New
Token ID968
Feature activation-2.60e+00

Pos logit contributions
club+0.126
cent+0.125
wives+0.123
horse+0.123
lines+0.120
Neg logit contributions
 confir-0.175
 destro-0.161
 consolidate-0.156
 rely-0.153
 consume-0.152
Token York
Token ID1971
Feature activation+4.59e-01

Pos logit contributions
club+1.634
cent+1.631
wives+1.590
horse+1.583
lines+1.540
Neg logit contributions
 confir-3.275
 destro-3.035
 consolidate-2.951
 rely-2.896
 consume-2.891
Token City
Token ID2254
Feature activation-1.90e+00

Pos logit contributions
 confir+0.456
 destro+0.415
 consolidate+0.399
 rely+0.390
 consume+0.389
Neg logit contributions
club-0.410
cent-0.410
wives-0.402
horse-0.402
lines-0.393
Token may
Token ID743
Feature activation+1.23e+00

Pos logit contributions
cent+1.889
club+1.888
wives+1.859
horse+1.851
lines+1.825
Neg logit contributions
 confir-1.698
 destro-1.523
 consolidate-1.458
 rely-1.417
 consume-1.413