TOP ACTIVATIONS
MAX = 1.266

> {{ phone . sn ipp et }} p >
whole . Image credit : Cop Richard , royalty
chant : Long live the king ! Long live
display . Sl ov akia s lone
", " -- images ", required = True , help ="
- and - a - half , in 2018 there
story " <|endoftext|> ( Phys . org ) -- An international
Google . Photo credit : Jeff Roberts
WASHINGTON , February 9 , 2014 In her recently published
whatever they are ! Long live disp ersion as such !
H err mann , ever the e got ist , seldom
super important , and the closer you are to the basket
came despite some strong head wind s for the GOP among
The work is planned to happen in two phases , according
to den uclear ize ahead of summit Venezuela 's Maduro says
mainstream big names like Neil G aiman or influential
. I also did two coats and it had to dry
matters WASHINGTON , February 9 , 2014 In
to protect polling stations ahead of the ' b anned '
and Thomas Hudson came in close to the center of the

MIN ACTIVATIONS
MIN = -1.527

Hide Caption 4 of 15 Photos : Carly Fiorina 's political
Hide Caption 5 of 15 Photos : Carly Fiorina 's political
comment on Facebook Share this comment on Twitter NY
comment on Facebook Share this comment on Twitter NY
mail this to a friend Print able version <|endoftext|> A university
Hide Caption 11 of 15 Photos : Carly Fiorina 's political
comment on Facebook Share this comment on Twitter NY
Hide Caption 6 of 15 Photos : Carly Fiorina 's political
Hide Caption 10 of 15 Photos : Carly Fiorina 's political
mail this to a friend Print able version <|endoftext|> Rob C
Hide Caption 9 of 15 Photos : Carly Fiorina 's political
comment on Facebook Share this comment on Twitter NY
Hide Caption 7 of 15 Photos : Carly Fiorina 's political
Recommend Share this comment on Facebook Share this comment
Hide Caption 8 of 15 Photos : Carly Fiorina 's political
Department of Justice were not returned . Gal ar
Hide Caption 12 of 15 Photos : Carly Fiorina 's political
Share this comment on Facebook Share this comment on Twitter
Share this comment on Facebook Share this comment on Twitter
Hide Caption 14 of 15 Photos : Carly Fiorina 's political

INTERVAL 0.568 - 1.266
CONTAINS 0.790%

declines to disclose his tax rate , stating in an ABC
. There had also been a 40 per cent
- be tack ler and then accelerated toward the end zone
2013 ( J U BA ) Military clashes have occurred
https :// twitter . com / om

INTERVAL -0.130 - 0.568
CONTAINS 72.841%

, The Chronicle 13 of 15 Photo : Le a Suzuki
made it to its final physical location and is now redundant
report , " he pointed to ' two African - Americans
you 've read the legal documents , maybe not . Here
78 - year - old developer picked up the former Federal

INTERVAL -0.829 - -0.130
CONTAINS 26.298%

months for the smaller Prot oc er at ops hatch lings
at around 30 . After I joined Pf izer Pharmaceutical s
chickens can do math ? They might not sc
his role in So ap .) Twenty - eight of the
? To Destiny , the developer 's next series of games

INTERVAL -1.527 - -0.829
CONTAINS 0.071%

Any reproduction , copying , or redistribution ( elect ronic or
2012 Canada 16 10 Posts # 14 On September 20 2013
the international cinema stage is simple : look pretty and play
of the game . <|endoftext|> Share . Two iconic horror titles
mail this to a friend Print able version <|endoftext|> A university
Token>
Token ID29
Feature activation+3.92e-01

Pos logit contributions
*)+1.370
allah+1.363
 coincide+1.343
 closer+1.308
 flee+1.290
Neg logit contributions
zos-2.627
ozo-2.573
sonian-2.517
yx-2.497
ints-2.252
Token{{
Token ID27007
Feature activation+5.14e-02

Pos logit contributions
*)+1.480
allah+1.462
 coincide+1.437
 flee+1.405
 closer+1.404
Neg logit contributions
ozo-1.929
zos-1.929
sonian-1.852
yx-1.831
ints-1.655
Token phone
Token ID3072
Feature activation+1.26e-01

Pos logit contributions
*)+0.203
allah+0.202
 coincide+0.199
 flee+0.195
 closer+0.193
Neg logit contributions
zos-0.246
ozo-0.246
sonian-0.235
yx-0.234
ints-0.210
Token.
Token ID13
Feature activation-5.27e-01

Pos logit contributions
*)+0.432
allah+0.426
 coincide+0.422
 closer+0.414
 flee+0.413
Neg logit contributions
zos-0.667
ozo-0.664
sonian-0.641
yx-0.635
ints-0.572
Tokensn
Token ID16184
Feature activation+3.02e-01

Pos logit contributions
zos+2.557
ozo+2.547
sonian+2.448
yx+2.439
ints+2.199
Neg logit contributions
*)-2.029
allah-2.023
 coincide-2.004
 flee-1.951
 bombard-1.923
Tokenipp
Token ID3974
Feature activation+1.27e+00

Pos logit contributions
*)+1.268
allah+1.268
 coincide+1.246
 flee+1.219
 closer+1.203
Neg logit contributions
zos-1.355
ozo-1.333
sonian-1.305
yx-1.271
ints-1.134
Tokenet
Token ID316
Feature activation+1.34e-01

Pos logit contributions
*)+5.119
allah+5.117
 coincide+4.912
 flee+4.861
 closer+4.845
Neg logit contributions
ozo-5.582
zos-5.455
sonian-5.317
yx-5.103
ints-4.626
Token }}
Token ID34949
Feature activation+4.09e-01

Pos logit contributions
*)+0.525
allah+0.524
 coincide+0.520
 flee+0.507
 closer+0.503
Neg logit contributions
zos-0.637
ozo-0.637
sonian-0.613
yx-0.607
ints-0.543
Token
Token ID3556
Feature activation+7.45e-01

Pos logit contributions
*)+1.585
 coincide+1.554
allah+1.553
 flee+1.537
 closer+1.533
Neg logit contributions
zos-1.969
ozo-1.954
sonian-1.892
yx-1.868
ints-1.674
Tokenp
Token ID79
Feature activation+1.52e-01

Pos logit contributions
*)+2.690
allah+2.683
 coincide+2.645
 flee+2.570
 closer+2.531
Neg logit contributions
zos-3.794
ozo-3.770
sonian-3.630
yx-3.614
ints-3.275
Token>
Token ID29
Feature activation+5.70e-04

Pos logit contributions
*)+0.416
allah+0.413
 coincide+0.405
>"+0.397
 closer+0.395
Neg logit contributions
zos-0.912
ozo-0.891
sonian-0.877
yx-0.866
ints-0.786
Token whole
Token ID2187
Feature activation+3.24e-01

Pos logit contributions
*)+0.056
allah+0.056
 coincide+0.055
 flee+0.054
 bombard+0.053
Neg logit contributions
zos-0.084
ozo-0.084
sonian-0.081
yx-0.081
ints-0.073
Token.
Token ID13
Feature activation+1.37e-01

Pos logit contributions
allah+1.148
*)+1.145
 coincide+1.133
 closer+1.097
 flee+1.094
Neg logit contributions
zos-1.679
ozo-1.665
sonian-1.616
yx-1.597
ints-1.453
Token
Token ID198
Feature activation-1.46e-01

Pos logit contributions
*)+0.539
allah+0.534
 coincide+0.529
 flee+0.514
 closer+0.507
Neg logit contributions
zos-0.660
ozo-0.656
sonian-0.632
yx-0.627
ints-0.566
Token
Token ID198
Feature activation+1.87e-02

Pos logit contributions
zos+0.616
ozo+0.612
yx+0.599
sonian+0.585
ints+0.518
Neg logit contributions
 murd-0.658
 coincide-0.650
 bombard-0.641
allah-0.640
 flee-0.633
TokenImage
Token ID5159
Feature activation+9.85e-02

Pos logit contributions
*)+0.066
allah+0.066
 coincide+0.065
 flee+0.063
 closer+0.062
Neg logit contributions
zos-0.097
ozo-0.096
sonian-0.093
yx-0.092
ints-0.084
Token credit
Token ID3884
Feature activation+1.23e+00

Pos logit contributions
*)+0.384
allah+0.382
 coincide+0.379
 flee+0.368
 bombard+0.363
Neg logit contributions
zos-0.476
ozo-0.474
sonian-0.456
yx-0.452
ints-0.408
Token:
Token ID25
Feature activation+6.73e-01

Pos logit contributions
*)+3.846
allah+3.838
 coincide+3.671
 flee+3.628
 murd+3.529
Neg logit contributions
zos-6.761
ozo-6.615
sonian-6.442
yx-6.300
ints-5.854
Token Cop
Token ID6955
Feature activation-3.80e-02

Pos logit contributions
allah+2.876
*)+2.873
 coincide+2.869
 flee+2.811
 murd+2.768
Neg logit contributions
zos-2.962
ozo-2.934
sonian-2.825
yx-2.770
ints-2.495
Token Richard
Token ID6219
Feature activation+2.33e-02

Pos logit contributions
zos+0.169
ozo+0.167
sonian+0.160
yx+0.159
ints+0.142
Neg logit contributions
allah-0.163
*)-0.162
 coincide-0.160
 flee-0.157
 murd-0.154
Token,
Token ID11
Feature activation+2.94e-01

Pos logit contributions
*)+0.103
allah+0.103
 coincide+0.100
 flee+0.099
 closer+0.098
Neg logit contributions
zos-0.101
ozo-0.100
sonian-0.096
yx-0.095
ints-0.085
Token royalty
Token ID29359
Feature activation+3.93e-01

Pos logit contributions
*)+1.198
allah+1.192
 coincide+1.183
 flee+1.162
 closer+1.141
Neg logit contributions
zos-1.367
ozo-1.357
sonian-1.306
yx-1.294
ints-1.166
Token chant
Token ID33541
Feature activation+2.26e-01

Pos logit contributions
zos+0.210
ozo+0.209
sonian+0.201
yx+0.201
ints+0.182
Neg logit contributions
*)-0.149
allah-0.149
 coincide-0.148
 flee-0.143
 bombard-0.141
Token:
Token ID25
Feature activation+1.75e-01

Pos logit contributions
*)+0.837
allah+0.836
 coincide+0.828
 flee+0.805
 closer+0.795
Neg logit contributions
zos-1.132
ozo-1.129
sonian-1.085
yx-1.078
ints-0.976
Token
Token ID564
Feature activation-3.57e-02

Pos logit contributions
*)+0.545
allah+0.544
 coincide+0.538
 flee+0.526
 bombard+0.513
Neg logit contributions
zos-0.985
ozo-0.984
sonian-0.950
yx-0.940
ints-0.866
Token
Token ID250
Feature activation+5.72e-02

Pos logit contributions
zos+0.158
ozo+0.156
yx+0.151
sonian+0.150
ints+0.134
Neg logit contributions
*)-0.153
allah-0.152
 coincide-0.151
 flee-0.147
 bombard-0.147
TokenLong
Token ID14617
Feature activation-3.30e-01

Pos logit contributions
*)+0.223
allah+0.222
 coincide+0.220
 flee+0.214
 bombard+0.211
Neg logit contributions
zos-0.275
ozo-0.274
sonian-0.263
yx-0.262
ints-0.236
Token live
Token ID2107
Feature activation+1.23e+00

Pos logit contributions
zos+1.876
ozo+1.864
yx+1.805
sonian+1.798
ints+1.651
Neg logit contributions
*)-0.996
allah-0.993
 coincide-0.984
 flee-0.941
 bombard-0.930
Token the
Token ID262
Feature activation+4.41e-01

Pos logit contributions
allah+4.569
 flee+4.499
*)+4.467
 coincide+4.466
 closer+4.423
Neg logit contributions
ozo-5.848
zos-5.825
sonian-5.749
yx-5.474
ints-5.047
Token king
Token ID5822
Feature activation+3.59e-01

Pos logit contributions
allah+1.805
*)+1.803
 coincide+1.788
 flee+1.786
 murd+1.750
Neg logit contributions
ozo-2.018
zos-2.015
sonian-1.952
yx-1.894
ints-1.716
Token!
Token ID0
Feature activation+4.13e-01

Pos logit contributions
allah+1.072
*)+1.056
 coincide+1.026
 flee+1.015
 murd+0.993
Neg logit contributions
zos-2.057
ozo-2.050
sonian-1.986
yx-1.969
ints-1.822
Token Long
Token ID5882
Feature activation+6.63e-02

Pos logit contributions
*)+1.572
allah+1.563
 coincide+1.534
 flee+1.510
 closer+1.482
Neg logit contributions
zos-2.045
ozo-2.034
sonian-1.965
yx-1.922
ints-1.749
Token live
Token ID2107
Feature activation+9.37e-01

Pos logit contributions
allah+0.179
*)+0.179
 coincide+0.177
 flee+0.175
 closer+0.168
Neg logit contributions
zos-0.400
ozo-0.399
sonian-0.389
yx-0.383
ints-0.356
Token display
Token ID3359
Feature activation-7.36e-02

Pos logit contributions
*)+1.106
allah+1.097
 coincide+1.091
 flee+1.065
 closer+1.052
Neg logit contributions
zos-1.589
ozo-1.579
sonian-1.515
yx-1.511
ints-1.362
Token.
Token ID13
Feature activation+5.26e-02

Pos logit contributions
zos+0.364
ozo+0.362
sonian+0.348
yx+0.347
ints+0.314
Neg logit contributions
*)-0.277
allah-0.276
 coincide-0.273
 flee-0.266
 bombard-0.262
Token
Token ID198
Feature activation-7.33e-02

Pos logit contributions
*)+0.200
allah+0.198
 coincide+0.195
 flee+0.190
 closer+0.189
Neg logit contributions
ozo-0.259
zos-0.259
yx-0.246
sonian-0.245
ints-0.222
Token
Token ID198
Feature activation-7.44e-02

Pos logit contributions
zos+0.315
ozo+0.314
yx+0.307
sonian+0.302
ints+0.270
Neg logit contributions
 murd-0.322
 coincide-0.320
allah-0.316
 bombard-0.315
 flee-0.311
TokenSl
Token ID11122
Feature activation+2.58e-01

Pos logit contributions
zos+0.370
ozo+0.368
yx+0.351
sonian+0.351
ints+0.314
Neg logit contributions
*)-0.284
allah-0.279
 coincide-0.272
 flee-0.265
 closer-0.263
Tokenov
Token ID709
Feature activation+1.17e+00

Pos logit contributions
allah+1.163
*)+1.158
 coincide+1.139
 flee+1.116
 closer+1.100
Neg logit contributions
zos-1.080
ozo-1.075
sonian-1.040
yx-1.020
ints-0.896
Tokenakia
Token ID21897
Feature activation+1.33e-01

Pos logit contributions
 closer+3.081
*)+3.078
allah+3.038
 flee+2.972
ş+2.963
Neg logit contributions
zos-6.730
ozo-6.710
sonian-6.579
yx-6.512
 Transformers-5.912
Token
Token ID447
Feature activation+1.71e-01

Pos logit contributions
*)+0.511
allah+0.510
 coincide+0.506
 flee+0.493
 bombard+0.486
Neg logit contributions
zos-0.651
ozo-0.648
sonian-0.623
yx-0.620
ints-0.558
Token
Token ID247
Feature activation+2.53e-01

Pos logit contributions
*)+0.890
allah+0.867
 coincide+0.862
 closer+0.846
 flee+0.843
Neg logit contributions
ozo-0.622
zos-0.612
sonian-0.573
yx-0.566
ints-0.484
Tokens
Token ID82
Feature activation+9.52e-02

Pos logit contributions
*)+1.021
allah+1.014
 coincide+1.000
 flee+0.977
 closer+0.969
Neg logit contributions
zos-1.198
ozo-1.189
sonian-1.135
yx-1.132
ints-1.010
Token lone
Token ID13497
Feature activation-1.44e-01

Pos logit contributions
*)+0.370
allah+0.367
 coincide+0.364
 flee+0.356
 closer+0.352
Neg logit contributions
zos-0.462
ozo-0.459
sonian-0.439
yx-0.439
ints-0.393
Token",
Token ID1600
Feature activation+4.51e-01

Pos logit contributions
 coincide+1.186
*)+1.180
allah+1.175
 flee+1.157
 bombard+1.149
Neg logit contributions
zos-1.994
ozo-1.987
sonian-1.929
yx-1.926
ints-1.756
Token "
Token ID366
Feature activation+2.44e-01

Pos logit contributions
allah+1.341
*)+1.335
 coincide+1.322
 flee+1.273
 bombard+1.265
Neg logit contributions
zos-2.589
ozo-2.573
yx-2.501
sonian-2.497
ints-2.291
Token--
Token ID438
Feature activation+4.94e-01

Pos logit contributions
allah+1.079
*)+1.079
 coincide+1.077
 flee+1.053
 bombard+1.046
Neg logit contributions
zos-1.039
ozo-1.034
yx-0.989
sonian-0.986
ints-0.879
Tokenimages
Token ID17566
Feature activation+4.12e-01

Pos logit contributions
*)+2.351
allah+2.344
 coincide+2.319
 flee+2.268
 closer+2.240
Neg logit contributions
zos-1.940
ozo-1.937
sonian-1.844
yx-1.831
ints-1.603
Token",
Token ID1600
Feature activation+3.58e-01

Pos logit contributions
allah+1.313
 coincide+1.312
*)+1.307
 flee+1.280
 bombard+1.265
Neg logit contributions
zos-2.272
ozo-2.267
sonian-2.192
yx-2.184
ints-2.001
Token required
Token ID2672
Feature activation+1.15e+00

Pos logit contributions
*)+1.101
 coincide+1.099
allah+1.066
 closer+1.064
 flee+1.061
Neg logit contributions
ozo-2.006
zos-1.992
sonian-1.960
yx-1.932
ints-1.742
Token=
Token ID28
Feature activation+6.59e-01

Pos logit contributions
*)+3.885
allah+3.835
 coincide+3.754
 flee+3.625
 closer+3.610
Neg logit contributions
zos-6.129
ozo-6.083
sonian-5.871
yx-5.824
ints-5.339
TokenTrue
Token ID17821
Feature activation+1.99e-01

Pos logit contributions
*)+2.028
allah+2.025
 coincide+2.007
 flee+1.942
 bombard+1.914
Neg logit contributions
zos-3.701
ozo-3.686
sonian-3.562
yx-3.558
ints-3.256
Token,
Token ID11
Feature activation-1.96e-01

Pos logit contributions
 coincide+0.834
allah+0.828
 murd+0.825
*)+0.818
 bombard+0.817
Neg logit contributions
zos-0.897
ozo-0.896
yx-0.868
sonian-0.860
ints-0.766
Token help
Token ID1037
Feature activation+1.31e-01

Pos logit contributions
zos+0.891
ozo+0.880
yx+0.863
sonian+0.863
ints+0.783
Neg logit contributions
*)-0.772
allah-0.766
 murd-0.751
 coincide-0.745
 bombard-0.741
Token="
Token ID2625
Feature activation+1.05e-01

Pos logit contributions
 bombard+0.429
 coincide+0.428
allah+0.428
 murd+0.425
 flee+0.421
Neg logit contributions
ozo-0.709
zos-0.708
yx-0.696
sonian-0.689
ints-0.625
Token-
Token ID12
Feature activation+9.01e-02

Pos logit contributions
*)+0.317
allah+0.317
 coincide+0.313
 flee+0.302
 closer+0.298
Neg logit contributions
zos-0.497
ozo-0.495
sonian-0.478
yx-0.476
ints-0.434
Tokenand
Token ID392
Feature activation+3.58e-01

Pos logit contributions
*)+0.244
allah+0.243
 coincide+0.239
 flee+0.230
 bombard+0.225
Neg logit contributions
zos-0.541
ozo-0.539
sonian-0.522
yx-0.520
ints-0.480
Token-
Token ID12
Feature activation+6.79e-01

Pos logit contributions
*)+1.393
allah+1.390
 coincide+1.381
 flee+1.341
 closer+1.331
Neg logit contributions
ozo-1.733
zos-1.731
sonian-1.670
yx-1.647
ints-1.482
Tokena
Token ID64
Feature activation+5.99e-01

Pos logit contributions
allah+2.430
*)+2.412
 coincide+2.320
 closer+2.291
 flee+2.249
Neg logit contributions
ozo-3.493
zos-3.479
sonian-3.372
yx-3.291
ints-2.960
Token-
Token ID12
Feature activation+7.94e-01

Pos logit contributions
*)+2.534
allah+2.530
 coincide+2.525
 flee+2.463
 bombard+2.441
Neg logit contributions
zos-2.668
ozo-2.661
sonian-2.553
yx-2.549
ints-2.277
Tokenhalf
Token ID13959
Feature activation+1.11e+00

Pos logit contributions
*)+2.824
allah+2.815
 coincide+2.783
 flee+2.701
 bombard+2.660
Neg logit contributions
zos-4.089
ozo-4.073
sonian-3.923
yx-3.909
ints-3.547
Token,
Token ID11
Feature activation+2.34e-01

Pos logit contributions
allah+3.670
*)+3.616
 coincide+3.555
 closer+3.494
 flee+3.446
Neg logit contributions
zos-5.825
ozo-5.745
sonian-5.680
yx-5.527
ints-5.109
Token in
Token ID287
Feature activation+7.23e-02

Pos logit contributions
*)+0.657
allah+0.655
 coincide+0.641
 flee+0.618
 closer+0.606
Neg logit contributions
zos-1.386
ozo-1.385
sonian-1.344
yx-1.335
ints-1.231
Token 2018
Token ID2864
Feature activation+5.47e-02

Pos logit contributions
*)+0.286
allah+0.285
 coincide+0.283
 flee+0.275
 bombard+0.271
Neg logit contributions
zos-0.343
ozo-0.342
sonian-0.329
yx-0.327
ints-0.294
Token there
Token ID612
Feature activation-1.37e-01

Pos logit contributions
*)+0.192
allah+0.191
 coincide+0.189
 flee+0.183
 closer+0.180
Neg logit contributions
zos-0.285
ozo-0.285
sonian-0.274
yx-0.273
ints-0.249
Token
Token ID447
Feature activation+2.92e-01

Pos logit contributions
zos+0.668
ozo+0.666
sonian+0.640
yx+0.638
ints+0.576
Neg logit contributions
*)-0.522
allah-0.521
 coincide-0.516
 flee-0.502
 bombard-0.495
Token story
Token ID1621
Feature activation+2.29e-01

Pos logit contributions
*)+0.166
allah+0.166
 coincide+0.164
 flee+0.162
 closer+0.160
Neg logit contributions
zos-0.167
ozo-0.166
yx-0.158
sonian-0.158
ints-0.140
Token"
Token ID1
Feature activation-3.17e-01

Pos logit contributions
*)+0.751
allah+0.749
 coincide+0.745
 flee+0.722
 closer+0.714
Neg logit contributions
zos-1.240
ozo-1.236
sonian-1.185
yx-1.184
ints-1.080
Token<|endoftext|>
Token ID50256
Feature activation+2.46e-01

Pos logit contributions
zos+1.580
ozo+1.574
sonian+1.515
yx+1.510
ints+1.365
Neg logit contributions
*)-1.174
allah-1.171
 coincide-1.159
 flee-1.127
 bombard-1.112
Token(
Token ID7
Feature activation+2.62e-02

Pos logit contributions
*)+0.968
allah+0.948
 coincide+0.918
 flee+0.892
 closer+0.888
Neg logit contributions
zos-1.196
ozo-1.189
sonian-1.141
yx-1.128
ints-1.019
TokenPhys
Token ID43215
Feature activation-2.47e-01

Pos logit contributions
*)+0.113
allah+0.110
 coincide+0.109
 flee+0.106
 closer+0.104
Neg logit contributions
zos-0.119
ozo-0.118
sonian-0.112
yx-0.112
ints-0.100
Token.
Token ID13
Feature activation+1.10e+00

Pos logit contributions
zos+1.289
ozo+1.284
sonian+1.235
yx+1.224
ints+1.120
Neg logit contributions
*)-0.863
allah-0.861
 coincide-0.845
 flee-0.820
 closer-0.812
Tokenorg
Token ID2398
Feature activation+4.63e-01

Pos logit contributions
*)+3.075
allah+3.012
 coincide+2.866
 flee+2.805
 closer+2.787
Neg logit contributions
ozo-6.525
zos-6.469
yx-6.286
sonian-6.220
ints-5.748
Token)
Token ID8
Feature activation+6.67e-01

Pos logit contributions
*)+1.380
allah+1.331
 coincide+1.321
 flee+1.274
 closer+1.259
Neg logit contributions
zos-2.735
ozo-2.682
yx-2.592
sonian-2.565
ints-2.375
Token --
Token ID1377
Feature activation+6.83e-02

Pos logit contributions
*)+2.522
 coincide+2.461
allah+2.442
 closer+2.411
 flee+2.378
Neg logit contributions
zos-3.342
ozo-3.292
sonian-3.146
yx-3.082
ints-2.828
Token An
Token ID1052
Feature activation-4.02e-02

Pos logit contributions
*)+0.281
allah+0.277
 coincide+0.276
 flee+0.267
 closer+0.267
Neg logit contributions
zos-0.317
ozo-0.314
sonian-0.302
yx-0.300
ints-0.268
Token international
Token ID3230
Feature activation+7.90e-02

Pos logit contributions
zos+0.186
ozo+0.185
sonian+0.177
yx+0.176
ints+0.158
Neg logit contributions
allah-0.164
*)-0.163
 coincide-0.162
 flee-0.158
 closer-0.156
Token Google
Token ID3012
Feature activation+1.35e-01

Pos logit contributions
zos+1.250
ozo+1.244
sonian+1.190
yx+1.188
ints+1.062
Neg logit contributions
*)-1.173
allah-1.171
 coincide-1.159
 flee-1.131
 bombard-1.115
Token.
Token ID13
Feature activation-3.00e-01

Pos logit contributions
*)+0.486
allah+0.485
 coincide+0.475
 flee+0.465
 bombard+0.460
Neg logit contributions
zos-0.686
ozo-0.683
sonian-0.660
yx-0.657
ints-0.596
Token
Token ID198
Feature activation-4.42e-01

Pos logit contributions
zos+1.285
ozo+1.281
yx+1.221
sonian+1.221
ints+1.080
Neg logit contributions
allah-1.320
*)-1.316
 coincide-1.314
 flee-1.284
 murd-1.270
Token
Token ID198
Feature activation-2.72e-01

Pos logit contributions
zos+1.761
ozo+1.759
yx+1.730
sonian+1.687
ints+1.477
Neg logit contributions
 murd-2.106
 coincide-2.053
 bombard-2.032
allah-2.004
 flee-1.994
TokenPhoto
Token ID6191
Feature activation-2.67e-01

Pos logit contributions
zos+1.320
ozo+1.316
sonian+1.264
yx+1.262
ints+1.138
Neg logit contributions
allah-1.036
 coincide-1.033
*)-1.031
 flee-1.007
 bombard-0.994
Token credit
Token ID3884
Feature activation+1.10e+00

Pos logit contributions
zos+1.287
ozo+1.279
sonian+1.228
yx+1.224
ints+1.103
Neg logit contributions
*)-1.039
allah-1.036
 coincide-1.026
 flee-1.000
 bombard-0.986
Token:
Token ID25
Feature activation+4.62e-01

Pos logit contributions
allah+3.743
*)+3.735
 coincide+3.648
 flee+3.596
 closer+3.489
Neg logit contributions
zos-5.695
ozo-5.547
sonian-5.357
yx-5.293
ints-4.924
Token Jeff
Token ID5502
Feature activation+2.46e-01

Pos logit contributions
allah+1.984
*)+1.972
 coincide+1.967
 flee+1.943
 closer+1.901
Neg logit contributions
zos-2.021
ozo-2.013
sonian-1.913
yx-1.903
ints-1.717
Token Roberts
Token ID10918
Feature activation-4.97e-02

Pos logit contributions
*)+1.102
allah+1.099
 coincide+1.075
 flee+1.063
 closer+1.048
Neg logit contributions
ozo-1.025
zos-1.024
sonian-0.975
yx-0.970
ints-0.857
Token
Token ID198
Feature activation+5.45e-02

Pos logit contributions
zos+0.239
ozo+0.238
sonian+0.229
yx+0.229
ints+0.206
Neg logit contributions
*)-0.194
allah-0.194
 coincide-0.190
 flee-0.187
 closer-0.183
Token
Token ID198
Feature activation-6.99e-02

Pos logit contributions
 murd+0.237
 coincide+0.236
allah+0.233
 bombard+0.231
 flee+0.230
Neg logit contributions
ozo-0.237
zos-0.237
yx-0.230
sonian-0.227
ints-0.202
TokenWASHINGTON
Token ID21793
Feature activation+3.08e-01

Pos logit contributions
*)+1.106
allah+1.084
 coincide+1.063
 flee+1.031
 closer+1.020
Neg logit contributions
zos-1.444
ozo-1.427
sonian-1.383
yx-1.367
ints-1.231
Token,
Token ID11
Feature activation-7.72e-02

Pos logit contributions
*)+1.160
allah+1.139
 coincide+1.134
 flee+1.101
 closer+1.090
Neg logit contributions
zos-1.540
ozo-1.521
yx-1.463
sonian-1.449
ints-1.316
Token February
Token ID3945
Feature activation+1.03e+00

Pos logit contributions
zos+0.291
ozo+0.290
sonian+0.276
yx+0.273
ints+0.239
Neg logit contributions
*)-0.380
allah-0.380
 coincide-0.376
 flee-0.369
 bombard-0.365
Token 9
Token ID860
Feature activation+8.27e-01

Pos logit contributions
*)+3.784
allah+3.724
 coincide+3.694
 flee+3.581
 closer+3.571
Neg logit contributions
zos-5.156
ozo-5.146
sonian-4.941
yx-4.881
ints-4.437
Token,
Token ID11
Feature activation+8.74e-01

Pos logit contributions
allah+2.793
*)+2.758
 coincide+2.744
 flee+2.573
 closer+2.527
Neg logit contributions
zos-4.429
ozo-4.336
yx-4.192
sonian-4.176
ints-3.812
Token 2014
Token ID1946
Feature activation+1.09e+00

Pos logit contributions
allah+2.666
*)+2.653
 coincide+2.643
 flee+2.503
 closer+2.495
Neg logit contributions
zos-4.953
ozo-4.907
sonian-4.745
yx-4.698
ints-4.311
Token
Token ID851
Feature activation+4.00e-01

Pos logit contributions
*)+3.598
allah+3.541
 coincide+3.463
 flee+3.273
    +3.235
Neg logit contributions
zos-5.797
ozo-5.752
sonian-5.631
yx-5.517
ints-5.034
Token In
Token ID554
Feature activation+1.34e-01

Pos logit contributions
*)+1.697
allah+1.671
 coincide+1.661
 flee+1.620
 closer+1.604
Neg logit contributions
zos-1.805
ozo-1.782
sonian-1.725
yx-1.702
ints-1.522
Token her
Token ID607
Feature activation+2.55e-01

Pos logit contributions
allah+0.541
*)+0.538
 coincide+0.536
 flee+0.522
 closer+0.517
Neg logit contributions
zos-0.627
ozo-0.619
sonian-0.606
yx-0.595
ints-0.532
Token recently
Token ID2904
Feature activation-1.06e-01

Pos logit contributions
allah+1.021
*)+1.020
 coincide+1.010
 flee+0.994
 closer+0.979
Neg logit contributions
zos-1.199
ozo-1.192
sonian-1.154
yx-1.138
ints-1.025
Token published
Token ID3199
Feature activation+5.47e-02

Pos logit contributions
zos+0.508
ozo+0.504
sonian+0.488
yx+0.483
ints+0.436
Neg logit contributions
*)-0.418
allah-0.418
 coincide-0.417
 flee-0.405
 closer-0.401
Tokenwhatever
Token ID39664
Feature activation+4.33e-01

Pos logit contributions
*)+0.506
allah+0.505
 coincide+0.499
 flee+0.487
 closer+0.480
Neg logit contributions
zos-0.638
ozo-0.634
sonian-0.611
yx-0.606
ints-0.548
Token they
Token ID484
Feature activation+6.75e-02

Pos logit contributions
*)+1.678
 coincide+1.676
allah+1.668
 flee+1.623
 closer+1.611
Neg logit contributions
zos-2.103
ozo-2.087
sonian-1.991
yx-1.974
ints-1.779
Token are
Token ID389
Feature activation-3.52e-01

Pos logit contributions
*)+0.246
allah+0.245
 coincide+0.243
 flee+0.237
 bombard+0.233
Neg logit contributions
zos-0.342
ozo-0.340
sonian-0.328
yx-0.326
ints-0.295
Token!
Token ID0
Feature activation-1.23e-02

Pos logit contributions
zos+1.700
ozo+1.694
sonian+1.627
yx+1.624
ints+1.464
Neg logit contributions
*)-1.350
allah-1.350
 coincide-1.337
 flee-1.301
 bombard-1.282
Token Long
Token ID5882
Feature activation-7.81e-02

Pos logit contributions
zos+0.059
ozo+0.059
sonian+0.057
yx+0.056
ints+0.051
Neg logit contributions
*)-0.047
allah-0.047
 coincide-0.047
 flee-0.046
 bombard-0.045
Token live
Token ID2107
Feature activation+1.07e+00

Pos logit contributions
zos+0.367
ozo+0.366
sonian+0.351
yx+0.349
ints+0.313
Neg logit contributions
*)-0.314
allah-0.312
 coincide-0.311
 flee-0.303
 bombard-0.299
Token disp
Token ID4596
Feature activation-3.31e-02

Pos logit contributions
*)+4.101
 coincide+4.095
allah+4.071
 flee+3.959
 closer+3.945
Neg logit contributions
ozo-5.089
zos-5.085
sonian-4.895
yx-4.781
ints-4.313
Tokenersion
Token ID6900
Feature activation-3.62e-01

Pos logit contributions
zos+0.203
ozo+0.202
sonian+0.196
yx+0.195
ints+0.180
Neg logit contributions
allah-0.085
*)-0.085
 coincide-0.083
 flee-0.080
 bombard-0.078
Token as
Token ID355
Feature activation+1.55e-02

Pos logit contributions
zos+1.967
ozo+1.961
sonian+1.892
yx+1.888
ints+1.724
Neg logit contributions
*)-1.177
allah-1.172
 coincide-1.160
 flee-1.124
 bombard-1.106
Token such
Token ID884
Feature activation+1.90e-02

Pos logit contributions
*)+0.062
allah+0.062
 coincide+0.062
 flee+0.060
 closer+0.059
Neg logit contributions
zos-0.073
ozo-0.072
sonian-0.069
yx-0.069
ints-0.062
Token!
Token ID0
Feature activation-1.97e-01

Pos logit contributions
*)+0.063
allah+0.063
 coincide+0.063
 flee+0.061
 bombard+0.060
Neg logit contributions
zos-0.103
ozo-0.102
sonian-0.099
yx-0.098
ints-0.089
TokenH
Token ID39
Feature activation-6.19e-02

Pos logit contributions
*)+0.779
allah+0.777
 coincide+0.768
 flee+0.749
 closer+0.740
Neg logit contributions
zos-0.911
ozo-0.910
sonian-0.872
yx-0.868
ints-0.781
Tokenerr
Token ID8056
Feature activation-2.40e-01

Pos logit contributions
zos+0.345
ozo+0.343
sonian+0.333
yx+0.331
ints+0.302
Neg logit contributions
*)-0.193
allah-0.193
 coincide-0.189
 flee-0.183
 bombard-0.180
Tokenmann
Token ID9038
Feature activation+6.52e-02

Pos logit contributions
ozo+0.971
zos+0.962
sonian+0.955
yx+0.917
ints+0.818
Neg logit contributions
 coincide-1.104
*)-1.092
allah-1.092
 flee-1.088
 bombard-1.077
Token,
Token ID11
Feature activation+2.89e-01

Pos logit contributions
*)+0.266
allah+0.266
 coincide+0.263
 flee+0.257
 closer+0.253
Neg logit contributions
zos-0.300
ozo-0.300
sonian-0.286
yx-0.286
ints-0.256
Token ever
Token ID1683
Feature activation+3.36e-01

Pos logit contributions
*)+1.125
allah+1.124
 coincide+1.111
 flee+1.085
 closer+1.076
Neg logit contributions
zos-1.379
ozo-1.375
yx-1.318
sonian-1.316
ints-1.188
Token the
Token ID262
Feature activation+1.07e+00

Pos logit contributions
allah+1.326
*)+1.313
 coincide+1.305
 closer+1.285
 flee+1.273
Neg logit contributions
zos-1.589
ozo-1.586
sonian-1.526
yx-1.520
ints-1.361
Token e
Token ID304
Feature activation+3.62e-01

Pos logit contributions
allah+4.082
*)+4.019
 closer+3.996
 coincide+3.992
 flee+3.971
Neg logit contributions
zos-5.075
ozo-4.989
sonian-4.836
yx-4.763
ints-4.296
Tokengot
Token ID23442
Feature activation+4.06e-03

Pos logit contributions
*)+1.576
allah+1.570
 coincide+1.539
 flee+1.504
 bombard+1.483
Neg logit contributions
ozo-1.558
zos-1.553
sonian-1.496
yx-1.477
ints-1.318
Tokenist
Token ID396
Feature activation-1.65e-01

Pos logit contributions
allah+0.019
*)+0.019
 coincide+0.019
 flee+0.018
 bombard+0.018
Neg logit contributions
zos-0.017
ozo-0.016
sonian-0.016
yx-0.015
ints-0.014
Token,
Token ID11
Feature activation+4.90e-01

Pos logit contributions
zos+0.815
ozo+0.808
sonian+0.778
yx+0.776
ints+0.703
Neg logit contributions
allah-0.620
*)-0.619
 coincide-0.610
 flee-0.596
 bombard-0.588
Token seldom
Token ID25129
Feature activation+3.78e-01

Pos logit contributions
allah+1.975
*)+1.958
 coincide+1.930
 flee+1.890
 closer+1.873
Neg logit contributions
ozo-2.264
zos-2.262
sonian-2.153
yx-2.141
ints-1.926
Token super
Token ID2208
Feature activation+1.37e-01

Pos logit contributions
*)+1.153
allah+1.150
 coincide+1.140
 flee+1.110
 closer+1.101
Neg logit contributions
zos-1.373
ozo-1.367
sonian-1.316
yx-1.305
ints-1.173
Token important
Token ID1593
Feature activation+4.65e-01

Pos logit contributions
allah+0.542
*)+0.542
 coincide+0.537
 flee+0.523
 closer+0.516
Neg logit contributions
zos-0.654
ozo-0.652
sonian-0.626
yx-0.623
ints-0.558
Token,
Token ID11
Feature activation+1.28e-01

Pos logit contributions
*)+1.680
allah+1.669
 coincide+1.667
 flee+1.625
 closer+1.625
Neg logit contributions
zos-2.365
ozo-2.355
sonian-2.266
yx-2.244
ints-2.018
Token and
Token ID290
Feature activation+2.58e-01

Pos logit contributions
*)+0.475
allah+0.474
 coincide+0.470
 flee+0.458
 closer+0.452
Neg logit contributions
zos-0.642
ozo-0.641
sonian-0.617
yx-0.613
ints-0.555
Token the
Token ID262
Feature activation+6.55e-02

Pos logit contributions
*)+1.006
allah+1.003
 coincide+0.998
 flee+0.971
 closer+0.959
Neg logit contributions
zos-1.241
ozo-1.239
sonian-1.192
yx-1.184
ints-1.064
Token closer
Token ID5699
Feature activation+1.06e+00

Pos logit contributions
*)+0.227
allah+0.226
 coincide+0.224
 flee+0.217
 bombard+0.214
Neg logit contributions
zos-0.343
ozo-0.342
sonian-0.329
yx-0.328
ints-0.299
Token you
Token ID345
Feature activation+3.84e-01

Pos logit contributions
 closer+3.922
 coincide+3.840
*)+3.799
 flee+3.755
allah+3.754
Neg logit contributions
zos-5.260
ozo-5.215
sonian-5.079
yx-4.981
ints-4.458
Token are
Token ID389
Feature activation+3.24e-02

Pos logit contributions
*)+1.323
allah+1.323
 coincide+1.291
 flee+1.252
 murd+1.247
Neg logit contributions
zos-2.020
ozo-2.005
yx-1.936
sonian-1.927
ints-1.760
Token to
Token ID284
Feature activation-1.11e-01

Pos logit contributions
*)+0.137
allah+0.137
 coincide+0.136
 flee+0.132
 closer+0.132
Neg logit contributions
zos-0.145
ozo-0.145
sonian-0.140
yx-0.138
ints-0.124
Token the
Token ID262
Feature activation+8.08e-02

Pos logit contributions
zos+0.509
ozo+0.509
sonian+0.489
yx+0.485
ints+0.433
Neg logit contributions
*)-0.454
allah-0.453
 coincide-0.450
 flee-0.440
 closer-0.434
Token basket
Token ID7988
Feature activation+8.65e-02

Pos logit contributions
*)+0.189
allah+0.188
 coincide+0.186
 flee+0.179
 closer+0.175
Neg logit contributions
zos-0.516
ozo-0.515
sonian-0.499
yx-0.497
ints-0.459
Token came
Token ID1625
Feature activation+2.20e-01

Pos logit contributions
*)+1.751
 coincide+1.745
allah+1.740
 flee+1.692
 closer+1.679
Neg logit contributions
zos-2.182
ozo-2.171
sonian-2.093
yx-2.070
ints-1.862
Token despite
Token ID3805
Feature activation+6.32e-01

Pos logit contributions
*)+0.823
allah+0.820
 coincide+0.820
 flee+0.794
 closer+0.790
Neg logit contributions
zos-1.091
ozo-1.089
sonian-1.051
yx-1.039
ints-0.939
Token some
Token ID617
Feature activation+3.50e-01

Pos logit contributions
allah+2.411
 coincide+2.411
*)+2.409
 closer+2.386
 flee+2.368
Neg logit contributions
ozo-3.065
zos-3.065
sonian-2.980
yx-2.869
ints-2.591
Token strong
Token ID1913
Feature activation+3.11e-01

Pos logit contributions
*)+1.349
 coincide+1.339
allah+1.338
 closer+1.318
 flee+1.317
Neg logit contributions
zos-1.705
ozo-1.694
sonian-1.628
yx-1.607
ints-1.444
Token head
Token ID1182
Feature activation+1.66e-01

Pos logit contributions
*)+1.132
allah+1.128
 coincide+1.121
 flee+1.092
 closer+1.084
Neg logit contributions
zos-1.580
ozo-1.566
sonian-1.517
yx-1.504
ints-1.356
Tokenwind
Token ID7972
Feature activation+1.05e+00

Pos logit contributions
*)+0.692
allah+0.686
 coincide+0.681
 flee+0.668
 closer+0.667
Neg logit contributions
zos-0.748
ozo-0.741
sonian-0.724
yx-0.711
ints-0.628
Tokens
Token ID82
Feature activation+6.16e-01

Pos logit contributions
*)+4.167
allah+4.140
 coincide+4.067
 closer+4.047
 bombard+4.033
Neg logit contributions
zos-4.910
ozo-4.900
sonian-4.717
yx-4.607
ints-4.020
Token for
Token ID329
Feature activation+2.16e-01

Pos logit contributions
*)+2.278
 coincide+2.264
allah+2.258
 closer+2.213
 flee+2.197
Neg logit contributions
zos-3.103
ozo-3.042
sonian-3.002
yx-2.917
ints-2.633
Token the
Token ID262
Feature activation+1.76e-01

Pos logit contributions
 coincide+0.828
*)+0.828
allah+0.824
 flee+0.812
 closer+0.807
Neg logit contributions
zos-1.045
ozo-1.043
sonian-1.015
yx-0.984
ints-0.887
Token GOP
Token ID6796
Feature activation+5.05e-02

Pos logit contributions
*)+0.614
allah+0.612
 coincide+0.608
 flee+0.594
 closer+0.586
Neg logit contributions
zos-0.916
ozo-0.914
sonian-0.883
yx-0.872
ints-0.793
Token among
Token ID1871
Feature activation+3.98e-01

Pos logit contributions
*)+0.199
allah+0.198
 coincide+0.197
 flee+0.193
 closer+0.190
Neg logit contributions
zos-0.241
ozo-0.240
sonian-0.231
yx-0.229
ints-0.206
TokenThe
Token ID464
Feature activation+9.80e-02

Pos logit contributions
*)+0.380
allah+0.374
 coincide+0.367
 flee+0.355
 closer+0.352
Neg logit contributions
zos-0.502
ozo-0.499
sonian-0.479
yx-0.476
ints-0.429
Token work
Token ID670
Feature activation+4.13e-01

Pos logit contributions
*)+0.393
allah+0.392
 coincide+0.389
 flee+0.380
 bombard+0.374
Neg logit contributions
zos-0.460
ozo-0.458
sonian-0.440
yx-0.437
ints-0.392
Token is
Token ID318
Feature activation+9.34e-02

Pos logit contributions
 coincide+1.545
allah+1.538
*)+1.537
 flee+1.486
 closer+1.482
Neg logit contributions
ozo-2.039
zos-2.032
yx-1.950
sonian-1.949
ints-1.763
Token planned
Token ID6027
Feature activation+3.10e-01

Pos logit contributions
*)+0.368
allah+0.367
 coincide+0.366
 flee+0.356
 closer+0.352
Neg logit contributions
zos-0.444
ozo-0.443
sonian-0.426
yx-0.423
ints-0.381
Token to
Token ID284
Feature activation+4.09e-01

Pos logit contributions
 coincide+1.216
*)+1.208
allah+1.199
 closer+1.185
 flee+1.171
Neg logit contributions
ozo-1.487
zos-1.477
sonian-1.427
yx-1.417
ints-1.268
Token happen
Token ID1645
Feature activation+1.05e+00

Pos logit contributions
allah+1.303
*)+1.302
 coincide+1.246
 murd+1.225
 flee+1.224
Neg logit contributions
zos-2.257
ozo-2.248
yx-2.172
sonian-2.160
ints-1.986
Token in
Token ID287
Feature activation+4.93e-01

Pos logit contributions
 coincide+4.011
*)+3.952
 closer+3.945
allah+3.912
 flee+3.805
Neg logit contributions
zos-5.101
ozo-5.077
sonian-4.914
yx-4.864
ints-4.371
Token two
Token ID734
Feature activation+4.17e-01

Pos logit contributions
 coincide+1.920
allah+1.914
*)+1.912
 flee+1.853
 closer+1.851
Neg logit contributions
zos-2.360
ozo-2.348
sonian-2.261
yx-2.239
ints-2.005
Token phases
Token ID21164
Feature activation+8.51e-01

Pos logit contributions
*)+1.098
 coincide+1.094
allah+1.090
 flee+1.045
 closer+1.029
Neg logit contributions
zos-2.527
ozo-2.523
sonian-2.442
yx-2.431
ints-2.241
Token,
Token ID11
Feature activation+3.57e-01

Pos logit contributions
 coincide+2.949
*)+2.910
 closer+2.903
allah+2.890
 flee+2.809
Neg logit contributions
zos-4.412
ozo-4.392
sonian-4.253
yx-4.214
ints-3.818
Token according
Token ID1864
Feature activation-1.81e-01

Pos logit contributions
 coincide+1.356
*)+1.355
allah+1.349
 flee+1.310
 closer+1.305
Neg logit contributions
zos-1.742
ozo-1.741
sonian-1.682
yx-1.661
ints-1.499
Token to
Token ID284
Feature activation-5.86e-02

Pos logit contributions
*)+0.750
allah+0.749
 coincide+0.743
 flee+0.725
 bombard+0.717
Neg logit contributions
zos-0.715
ozo-0.712
sonian-0.681
yx-0.678
ints-0.601
Token den
Token ID2853
Feature activation-2.29e-01

Pos logit contributions
ozo+0.312
zos+0.308
yx+0.301
sonian+0.295
ints+0.273
Neg logit contributions
allah-0.201
*)-0.198
 coincide-0.193
 flee-0.187
 murd-0.185
Tokenuclear
Token ID4016
Feature activation+3.21e-01

Pos logit contributions
ozo+1.124
zos+1.103
yx+1.070
sonian+1.061
ints+0.960
Neg logit contributions
*)-0.876
allah-0.873
 coincide-0.865
 murd-0.828
 flee-0.826
Tokenize
Token ID1096
Feature activation+1.40e-01

Pos logit contributions
allah+1.577
*)+1.575
 coincide+1.567
 flee+1.528
 bombard+1.513
Neg logit contributions
zos-1.210
ozo-1.207
yx-1.146
sonian-1.142
ints-1.001
Token ahead
Token ID4058
Feature activation+7.34e-01

Pos logit contributions
*)+0.558
allah+0.556
 coincide+0.551
 flee+0.537
 closer+0.529
Neg logit contributions
zos-0.662
ozo-0.658
sonian-0.632
yx-0.627
ints-0.564
Token of
Token ID286
Feature activation+1.05e+00

Pos logit contributions
*)+2.731
allah+2.717
 coincide+2.688
 flee+2.615
 closer+2.571
Neg logit contributions
zos-3.663
ozo-3.644
sonian-3.505
yx-3.491
ints-3.165
Token summit
Token ID14237
Feature activation+4.54e-01

Pos logit contributions
 coincide+4.022
allah+3.997
*)+3.975
 closer+3.853
 flee+3.843
Neg logit contributions
zos-5.162
ozo-5.019
yx-4.818
sonian-4.734
ints-4.285
Token Venezuela
Token ID16666
Feature activation-7.73e-02

Pos logit contributions
allah+1.819
*)+1.796
 coincide+1.771
 flee+1.742
 closer+1.717
Neg logit contributions
zos-2.175
ozo-2.125
sonian-2.059
yx-2.023
ints-1.824
Token's
Token ID338
Feature activation+1.20e-01

Pos logit contributions
zos+0.407
ozo+0.405
sonian+0.394
yx+0.390
ints+0.356
Neg logit contributions
*)-0.263
allah-0.260
 coincide-0.258
 flee-0.248
 murd-0.246
Token Maduro
Token ID33007
Feature activation+5.37e-02

Pos logit contributions
*)+0.505
allah+0.504
 coincide+0.499
 flee+0.487
 bombard+0.481
Neg logit contributions
zos-0.539
ozo-0.536
sonian-0.515
yx-0.512
ints-0.459
Token says
Token ID1139
Feature activation+1.85e-01

Pos logit contributions
*)+0.206
allah+0.206
 coincide+0.204
 flee+0.201
 closer+0.196
Neg logit contributions
zos-0.261
ozo-0.261
yx-0.250
sonian-0.249
ints-0.225
Token mainstream
Token ID8661
Feature activation-3.38e-02

Pos logit contributions
*)+1.787
allah+1.769
 coincide+1.755
 flee+1.722
 closer+1.707
Neg logit contributions
zos-1.977
ozo-1.966
sonian-1.895
yx-1.876
ints-1.680
Token big
Token ID1263
Feature activation+6.21e-01

Pos logit contributions
zos+0.166
ozo+0.165
sonian+0.159
yx+0.159
ints+0.143
Neg logit contributions
*)-0.128
allah-0.127
 coincide-0.126
 flee-0.123
 bombard-0.121
Token names
Token ID3891
Feature activation+2.15e-01

Pos logit contributions
*)+2.652
allah+2.645
 coincide+2.572
 flee+2.551
 bombard+2.506
Neg logit contributions
zos-2.713
ozo-2.703
sonian-2.608
yx-2.583
ints-2.321
Token like
Token ID588
Feature activation+1.79e-01

Pos logit contributions
*)+0.828
allah+0.822
 coincide+0.815
 flee+0.796
 bombard+0.784
Neg logit contributions
zos-1.044
ozo-1.037
sonian-1.002
yx-0.994
ints-0.898
Token Neil
Token ID15929
Feature activation+1.11e-01

Pos logit contributions
*)+0.811
allah+0.807
 coincide+0.799
 flee+0.786
 closer+0.774
Neg logit contributions
ozo-0.748
zos-0.748
sonian-0.712
yx-0.705
ints-0.630
Token G
Token ID402
Feature activation+1.04e+00

Pos logit contributions
*)+0.478
allah+0.477
 coincide+0.473
 flee+0.462
 bombard+0.456
Neg logit contributions
zos-0.485
ozo-0.483
sonian-0.462
yx-0.460
ints-0.410
Tokenaiman
Token ID47840
Feature activation-1.20e-01

Pos logit contributions
allah+4.776
*)+4.738
 coincide+4.549
 bombard+4.541
 closer+4.516
Neg logit contributions
zos-4.064
ozo-4.027
yx-3.909
sonian-3.908
ints-3.374
Token
Token ID960
Feature activation+2.71e-01

Pos logit contributions
zos+0.631
ozo+0.627
sonian+0.606
yx+0.603
ints+0.549
Neg logit contributions
*)-0.417
allah-0.416
 coincide-0.411
 flee-0.400
 bombard-0.393
Tokenor
Token ID273
Feature activation+1.94e-01

Pos logit contributions
*)+0.993
allah+0.989
 coincide+0.961
 flee+0.935
 closer+0.928
Neg logit contributions
zos-1.367
ozo-1.362
sonian-1.308
yx-1.305
ints-1.185
Token influential
Token ID14212
Feature activation+9.16e-02

Pos logit contributions
*)+0.833
allah+0.829
 coincide+0.828
 flee+0.807
 closer+0.799
Neg logit contributions
zos-0.851
ozo-0.849
sonian-0.811
yx-0.804
ints-0.725
Token
Token ID564
Feature activation+1.14e-01

Pos logit contributions
*)+0.351
allah+0.348
 coincide+0.346
 flee+0.337
 bombard+0.331
Neg logit contributions
zos-0.448
ozo-0.447
sonian-0.429
yx-0.427
ints-0.386
Token.
Token ID13
Feature activation+1.58e-01

Pos logit contributions
*)+2.001
 coincide+1.987
allah+1.982
 flee+1.930
 closer+1.909
Neg logit contributions
zos-2.694
ozo-2.687
sonian-2.605
yx-2.577
ints-2.327
Token I
Token ID314
Feature activation+1.48e-01

Pos logit contributions
*)+0.676
allah+0.669
 coincide+0.663
 flee+0.648
 closer+0.639
Neg logit contributions
zos-0.705
ozo-0.704
sonian-0.673
yx-0.668
ints-0.599
Token also
Token ID635
Feature activation+3.46e-01

Pos logit contributions
*)+0.612
allah+0.611
 coincide+0.602
 flee+0.587
 murd+0.582
Neg logit contributions
zos-0.672
ozo-0.666
sonian-0.640
yx-0.640
ints-0.572
Token did
Token ID750
Feature activation+4.29e-01

Pos logit contributions
*)+1.466
allah+1.462
 coincide+1.450
 flee+1.416
 bombard+1.398
Neg logit contributions
zos-1.545
ozo-1.538
sonian-1.473
yx-1.467
ints-1.310
Token two
Token ID734
Feature activation+3.93e-01

Pos logit contributions
*)+1.664
allah+1.657
 coincide+1.643
 flee+1.620
 bombard+1.593
Neg logit contributions
ozo-2.066
zos-2.060
sonian-1.986
yx-1.949
ints-1.766
Token coats
Token ID30720
Feature activation+1.04e+00

Pos logit contributions
*)+1.500
 coincide+1.487
allah+1.484
 flee+1.459
 closer+1.443
Neg logit contributions
ozo-1.917
zos-1.917
sonian-1.842
yx-1.821
ints-1.646
Token and
Token ID290
Feature activation+4.65e-01

Pos logit contributions
*)+3.595
 coincide+3.554
allah+3.521
 closer+3.500
 flee+3.456
Neg logit contributions
zos-5.348
ozo-5.311
sonian-5.141
yx-5.050
ints-4.617
Token it
Token ID340
Feature activation+3.42e-01

Pos logit contributions
*)+1.829
 coincide+1.825
allah+1.817
 flee+1.780
 closer+1.762
Neg logit contributions
ozo-2.217
zos-2.201
sonian-2.128
yx-2.084
ints-1.889
Token had
Token ID550
Feature activation+1.19e-01

Pos logit contributions
*)+1.290
allah+1.286
 coincide+1.277
 flee+1.237
 closer+1.223
Neg logit contributions
zos-1.684
ozo-1.682
sonian-1.611
yx-1.599
ints-1.452
Token to
Token ID284
Feature activation-1.06e-02

Pos logit contributions
*)+0.472
allah+0.471
 coincide+0.469
 flee+0.459
 closer+0.453
Neg logit contributions
zos-0.565
ozo-0.563
sonian-0.541
yx-0.536
ints-0.483
Token dry
Token ID5894
Feature activation+3.61e-01

Pos logit contributions
zos+0.056
ozo+0.056
sonian+0.054
yx+0.054
ints+0.049
Neg logit contributions
*)-0.037
allah-0.036
 coincide-0.036
 flee-0.035
 bombard-0.034
Token matters
Token ID6067
Feature activation+1.69e-01

Pos logit contributions
zos+0.345
ozo+0.343
sonian+0.329
yx+0.328
ints+0.294
Neg logit contributions
*)-0.306
allah-0.306
 coincide-0.304
 flee-0.296
 bombard-0.292
Token
Token ID198
Feature activation+3.89e-01

Pos logit contributions
*)+0.629
allah+0.626
 coincide+0.623
 flee+0.607
 closer+0.601
Neg logit contributions
zos-0.845
ozo-0.838
sonian-0.807
yx-0.804
ints-0.723
Token
Token ID198
Feature activation+2.90e-01

Pos logit contributions
*)+1.357
allah+1.334
 coincide+1.284
    +1.277
 closer+1.253
Neg logit contributions
zos-2.066
ozo-2.030
sonian-1.972
yx-1.951
ints-1.782
TokenWASHINGTON
Token ID21793
Feature activation+3.08e-01

Pos logit contributions
*)+1.106
allah+1.084
 coincide+1.063
 flee+1.031
 closer+1.020
Neg logit contributions
zos-1.444
ozo-1.427
sonian-1.383
yx-1.367
ints-1.231
Token,
Token ID11
Feature activation-7.72e-02

Pos logit contributions
*)+1.160
allah+1.139
 coincide+1.134
 flee+1.101
 closer+1.090
Neg logit contributions
zos-1.540
ozo-1.521
yx-1.463
sonian-1.449
ints-1.316
Token February
Token ID3945
Feature activation+1.03e+00

Pos logit contributions
zos+0.291
ozo+0.290
sonian+0.276
yx+0.273
ints+0.239
Neg logit contributions
*)-0.380
allah-0.380
 coincide-0.376
 flee-0.369
 bombard-0.365
Token 9
Token ID860
Feature activation+8.27e-01

Pos logit contributions
*)+3.784
allah+3.724
 coincide+3.694
 flee+3.581
 closer+3.571
Neg logit contributions
zos-5.156
ozo-5.146
sonian-4.941
yx-4.881
ints-4.437
Token,
Token ID11
Feature activation+8.74e-01

Pos logit contributions
allah+2.793
*)+2.758
 coincide+2.744
 flee+2.573
 closer+2.527
Neg logit contributions
zos-4.429
ozo-4.336
yx-4.192
sonian-4.176
ints-3.812
Token 2014
Token ID1946
Feature activation+1.09e+00

Pos logit contributions
allah+2.666
*)+2.653
 coincide+2.643
 flee+2.503
 closer+2.495
Neg logit contributions
zos-4.953
ozo-4.907
sonian-4.745
yx-4.698
ints-4.311
Token
Token ID851
Feature activation+4.00e-01

Pos logit contributions
*)+3.598
allah+3.541
 coincide+3.463
 flee+3.273
    +3.235
Neg logit contributions
zos-5.797
ozo-5.752
sonian-5.631
yx-5.517
ints-5.034
Token In
Token ID554
Feature activation+1.34e-01

Pos logit contributions
*)+1.697
allah+1.671
 coincide+1.661
 flee+1.620
 closer+1.604
Neg logit contributions
zos-1.805
ozo-1.782
sonian-1.725
yx-1.702
ints-1.522
Token to
Token ID284
Feature activation+3.35e-01

Pos logit contributions
*)+1.833
allah+1.825
 coincide+1.823
 flee+1.779
 closer+1.765
Neg logit contributions
zos-2.495
ozo-2.477
sonian-2.423
yx-2.380
ints-2.149
Token protect
Token ID1805
Feature activation+1.70e-01

Pos logit contributions
*)+1.093
 coincide+1.092
allah+1.090
 flee+1.057
 bombard+1.040
Neg logit contributions
zos-1.823
ozo-1.819
sonian-1.758
yx-1.743
ints-1.592
Token polling
Token ID13985
Feature activation+8.42e-02

Pos logit contributions
 coincide+0.659
*)+0.656
allah+0.655
 flee+0.643
 closer+0.634
Neg logit contributions
zos-0.822
ozo-0.813
sonian-0.791
yx-0.772
ints-0.696
Token stations
Token ID8985
Feature activation+3.32e-01

Pos logit contributions
*)+0.258
 coincide+0.256
allah+0.255
 flee+0.247
 closer+0.242
Neg logit contributions
zos-0.476
ozo-0.475
sonian-0.461
yx-0.458
ints-0.418
Token ahead
Token ID4058
Feature activation+4.02e-01

Pos logit contributions
*)+1.249
allah+1.244
 coincide+1.239
 flee+1.205
 closer+1.190
Neg logit contributions
zos-1.646
ozo-1.637
sonian-1.586
yx-1.568
ints-1.413
Token of
Token ID286
Feature activation+1.03e+00

Pos logit contributions
allah+1.611
*)+1.608
 coincide+1.598
 murd+1.580
 flee+1.562
Neg logit contributions
ozo-1.867
zos-1.863
yx-1.814
sonian-1.782
ints-1.607
Token the
Token ID262
Feature activation+3.66e-01

Pos logit contributions
 coincide+3.779
 closer+3.697
allah+3.657
 flee+3.635
*)+3.621
Neg logit contributions
ozo-5.165
zos-5.114
sonian-5.041
yx-4.806
ints-4.277
Token '
Token ID705
Feature activation-6.53e-02

Pos logit contributions
*)+1.317
allah+1.308
 coincide+1.303
 flee+1.270
 closer+1.255
Neg logit contributions
zos-1.870
ozo-1.869
sonian-1.805
yx-1.780
ints-1.609
Tokenb
Token ID65
Feature activation+6.43e-02

Pos logit contributions
zos+0.319
ozo+0.318
sonian+0.305
yx+0.304
ints+0.274
Neg logit contributions
*)-0.250
allah-0.249
 coincide-0.246
 flee-0.240
 bombard-0.236
Tokenanned
Token ID3577
Feature activation+4.21e-01

Pos logit contributions
*)+0.268
allah+0.267
 coincide+0.263
 flee+0.255
 closer+0.253
Neg logit contributions
zos-0.292
ozo-0.289
sonian-0.280
yx-0.279
ints-0.246
Token'
Token ID6
Feature activation-3.34e-02

Pos logit contributions
*)+1.570
 coincide+1.569
allah+1.564
 flee+1.529
 closer+1.508
Neg logit contributions
ozo-2.085
zos-2.066
sonian-2.004
yx-1.993
ints-1.765
Token and
Token ID290
Feature activation-3.90e-02

Pos logit contributions
*)+0.910
allah+0.909
 coincide+0.895
 flee+0.880
 closer+0.865
Neg logit contributions
zos-1.186
ozo-1.176
sonian-1.136
yx-1.122
ints-1.020
Token Thomas
Token ID5658
Feature activation-1.38e-01

Pos logit contributions
zos+0.181
ozo+0.180
sonian+0.173
yx+0.170
ints+0.154
Neg logit contributions
*)-0.158
allah-0.158
 coincide-0.157
 flee-0.155
 closer-0.152
Token Hudson
Token ID19995
Feature activation+1.85e-01

Pos logit contributions
zos+0.683
ozo+0.680
sonian+0.654
yx+0.652
ints+0.589
Neg logit contributions
*)-0.516
allah-0.514
 coincide-0.509
 flee-0.496
 bombard-0.488
Token came
Token ID1625
Feature activation+1.79e-01

Pos logit contributions
*)+0.661
allah+0.659
 coincide+0.647
 flee+0.635
 closer+0.624
Neg logit contributions
zos-0.955
ozo-0.949
sonian-0.912
yx-0.907
ints-0.825
Token in
Token ID287
Feature activation+1.49e-01

Pos logit contributions
*)+0.709
allah+0.707
 coincide+0.701
 flee+0.685
 closer+0.678
Neg logit contributions
zos-0.852
ozo-0.851
sonian-0.815
yx-0.813
ints-0.730
Token close
Token ID1969
Feature activation+1.02e+00

Pos logit contributions
*)+0.566
allah+0.564
 coincide+0.560
 flee+0.547
 closer+0.541
Neg logit contributions
zos-0.737
ozo-0.736
sonian-0.705
yx-0.702
ints-0.633
Token to
Token ID284
Feature activation-1.16e-01

Pos logit contributions
 coincide+3.753
*)+3.729
allah+3.706
 closer+3.702
 flee+3.637
Neg logit contributions
zos-5.202
ozo-5.135
yx-4.880
sonian-4.863
ints-4.406
Token the
Token ID262
Feature activation-7.23e-02

Pos logit contributions
zos+0.552
ozo+0.550
sonian+0.528
yx+0.526
ints+0.474
Neg logit contributions
*)-0.457
allah-0.456
 coincide-0.451
 flee-0.440
 bombard-0.434
Token center
Token ID3641
Feature activation-1.34e-01

Pos logit contributions
zos+0.350
ozo+0.348
sonian+0.334
yx+0.332
ints+0.300
Neg logit contributions
*)-0.281
allah-0.279
 coincide-0.277
 flee-0.272
 bombard-0.268
Token of
Token ID286
Feature activation-1.50e-01

Pos logit contributions
zos+0.672
ozo+0.667
sonian+0.642
yx+0.638
ints+0.580
Neg logit contributions
*)-0.493
allah-0.492
 coincide-0.488
 flee-0.476
 closer-0.468
Token the
Token ID262
Feature activation-5.00e-02

Pos logit contributions
zos+0.687
ozo+0.681
sonian+0.658
yx+0.648
ints+0.583
Neg logit contributions
*)-0.622
allah-0.622
 coincide-0.615
 flee-0.601
 closer-0.593
Token Hide
Token ID10415
Feature activation-5.23e-01

Pos logit contributions
zos+0.587
ozo+0.585
sonian+0.561
yx+0.558
ints+0.503
Neg logit contributions
*)-0.504
allah-0.499
 coincide-0.495
 flee-0.483
 closer-0.476
Token Caption
Token ID11260
Feature activation-8.67e-01

Pos logit contributions
zos+2.169
ozo+2.143
sonian+2.047
yx+2.038
ints+1.798
Neg logit contributions
*)-2.424
allah-2.419
 coincide-2.389
 flee-2.351
 closer-2.322
Token 4
Token ID604
Feature activation-2.77e-01

Pos logit contributions
zos+3.937
ozo+3.896
yx+3.838
sonian+3.765
 Correction+3.424
Neg logit contributions
allah-3.439
 coincide-3.413
*)-3.343
 murd-3.342
 flee-3.337
Token of
Token ID286
Feature activation-2.44e-01

Pos logit contributions
ozo+1.220
zos+1.218
yx+1.189
sonian+1.157
ints+1.029
Neg logit contributions
 murd-1.187
 coincide-1.177
allah-1.160
 flee-1.147
*)-1.145
Token 15
Token ID1315
Feature activation-4.37e-01

Pos logit contributions
zos+1.155
ozo+1.143
yx+1.113
sonian+1.096
ints+0.986
Neg logit contributions
allah-0.961
 coincide-0.957
*)-0.951
 flee-0.926
 murd-0.923
Token Photos
Token ID9434
Feature activation-1.53e+00

Pos logit contributions
zos+1.646
ozo+1.637
sonian+1.555
yx+1.548
ints+1.349
Neg logit contributions
*)-2.159
allah-2.153
 coincide-2.137
 flee-2.095
 bombard-2.071
Token:
Token ID25
Feature activation-4.41e-01

Pos logit contributions
zos+5.324
yx+5.323
ozo+5.291
sonian+4.962
 Transformers+4.496
Neg logit contributions
 murd-7.574
allah-7.209
 neighb-7.203
 accur-7.052
 bombard-7.040
Token Carly
Token ID35982
Feature activation-4.98e-01

Pos logit contributions
zos+1.832
ozo+1.824
sonian+1.741
yx+1.734
ints+1.533
Neg logit contributions
*)-2.006
allah-2.000
 coincide-1.985
 flee-1.941
 bombard-1.918
Token Fiorina
Token ID49028
Feature activation-5.93e-01

Pos logit contributions
zos+1.663
ozo+1.649
sonian+1.562
yx+1.552
ints+1.325
Neg logit contributions
*)-2.668
allah-2.656
 coincide-2.646
 flee-2.583
 bombard-2.568
Token's
Token ID338
Feature activation-3.61e-01

Pos logit contributions
zos+2.730
ozo+2.702
sonian+2.611
yx+2.592
ints+2.346
Neg logit contributions
*)-2.401
allah-2.378
 coincide-2.342
 flee-2.263
 bombard-2.256
Token political
Token ID1964
Feature activation-2.82e-01

Pos logit contributions
zos+1.904
ozo+1.896
sonian+1.825
yx+1.823
ints+1.660
Neg logit contributions
*)-1.240
allah-1.233
 coincide-1.220
 flee-1.178
 bombard-1.162
Token Hide
Token ID10415
Feature activation+3.57e-02

Pos logit contributions
*)+1.684
allah+1.657
 coincide+1.637
 flee+1.601
 closer+1.577
Neg logit contributions
zos-3.005
ozo-2.988
sonian-2.916
yx-2.867
ints-2.649
Token Caption
Token ID11260
Feature activation-5.76e-01

Pos logit contributions
 coincide+0.183
 murd+0.182
allah+0.181
*)+0.181
 bombard+0.180
Neg logit contributions
ozo-0.127
zos-0.124
sonian-0.120
yx-0.119
ints-0.103
Token 5
Token ID642
Feature activation+2.43e-02

Pos logit contributions
zos+2.787
ozo+2.725
yx+2.725
sonian+2.611
ints+2.453
Neg logit contributions
allah-2.207
 murd-2.205
 coincide-2.186
 flee-2.155
*)-2.125
Token of
Token ID286
Feature activation+1.01e-01

Pos logit contributions
 murd+0.103
 coincide+0.103
allah+0.101
 bombard+0.100
 flee+0.099
Neg logit contributions
ozo-0.109
zos-0.109
yx-0.106
sonian-0.104
ints-0.092
Token 15
Token ID1315
Feature activation-9.26e-02

Pos logit contributions
allah+0.435
 coincide+0.426
 murd+0.425
*)+0.421
 bombard+0.419
Neg logit contributions
zos-0.440
ozo-0.436
yx-0.435
sonian-0.422
ints-0.378
Token Photos
Token ID9434
Feature activation-1.46e+00

Pos logit contributions
zos+0.274
ozo+0.270
yx+0.268
sonian+0.253
ints+0.217
Neg logit contributions
 murd-0.536
 coincide-0.534
allah-0.526
 bombard-0.525
 flee-0.521
Token:
Token ID25
Feature activation-1.90e-01

Pos logit contributions
yx+5.147
ozo+4.976
zos+4.955
sonian+4.611
ints+4.080
Neg logit contributions
 neighb-7.745
 murd-7.701
 conduc-7.510
senal-7.503
 rall-7.438
Token Carly
Token ID35982
Feature activation-5.34e-01

Pos logit contributions
zos+0.684
ozo+0.680
yx+0.679
sonian+0.651
 Correction+0.578
Neg logit contributions
allah-0.943
 coincide-0.942
 murd-0.937
*)-0.927
 bombard-0.926
Token Fiorina
Token ID49028
Feature activation-5.16e-01

Pos logit contributions
zos+1.301
ozo+1.268
yx+1.226
sonian+1.189
ints+0.968
Neg logit contributions
 coincide-3.263
allah-3.259
 murd-3.249
*)-3.238
 bombard-3.228
Token's
Token ID338
Feature activation-2.67e-01

Pos logit contributions
zos+1.944
ozo+1.931
sonian+1.888
yx+1.860
ints+1.618
Neg logit contributions
allah-2.470
 murd-2.458
 coincide-2.436
*)-2.435
 bombard-2.386
Token political
Token ID1964
Feature activation-6.90e-02

Pos logit contributions
ozo+1.428
zos+1.422
sonian+1.409
yx+1.388
ints+1.269
Neg logit contributions
allah-0.859
*)-0.854
 coincide-0.820
 murd-0.813
 bombard-0.807
Token comment
Token ID2912
Feature activation-1.23e+00

Pos logit contributions
zos+1.897
yx+1.882
ozo+1.867
sonian+1.814
ints+1.596
Neg logit contributions
allah-2.296
 murd-2.284
*)-2.279
 coincide-2.278
 bombard-2.251
Token on
Token ID319
Feature activation-5.14e-01

Pos logit contributions
ozo+4.491
yx+4.485
zos+4.455
sonian+4.213
rade+3.727
Neg logit contributions
senal-6.143
 murd-6.092
 neighb-5.997
 unnecess-5.918
 conduc-5.798
Token Facebook
Token ID3203
Feature activation-6.77e-01

Pos logit contributions
ozo+1.717
zos+1.697
yx+1.693
sonian+1.641
ints+1.399
Neg logit contributions
 coincide-2.693
 murd-2.685
*)-2.654
allah-2.643
 flee-2.642
Token Share
Token ID8734
Feature activation-1.25e+00

Pos logit contributions
yx+2.577
ozo+2.496
zos+2.484
sonian+2.418
rade+2.171
Neg logit contributions
 murd-3.194
allah-3.152
 coincide-3.127
 bombard-3.115
*)-3.110
Token this
Token ID428
Feature activation-7.74e-01

Pos logit contributions
ozo+4.786
yx+4.733
zos+4.713
sonian+4.502
rade+4.040
Neg logit contributions
 murd-5.812
senal-5.781
 coincide-5.552
allah-5.527
 unnecess-5.508
Token comment
Token ID2912
Feature activation-1.42e+00

Pos logit contributions
yx+2.847
zos+2.805
ozo+2.796
sonian+2.735
ints+2.365
Neg logit contributions
 murd-3.626
allah-3.621
 coincide-3.575
*)-3.575
 bombard-3.549
Token on
Token ID319
Feature activation-7.30e-01

Pos logit contributions
yx+5.127
ozo+5.110
zos+5.052
sonian+4.758
rade+4.248
Neg logit contributions
senal-7.307
 murd-7.051
 neighb-6.963
 unnecess-6.900
 conduc-6.758
Token Twitter
Token ID3009
Feature activation-6.91e-01

Pos logit contributions
zos+2.325
ozo+2.291
yx+2.270
sonian+2.170
 Transformers+1.891
Neg logit contributions
 murd-3.919
allah-3.904
 coincide-3.887
*)-3.855
 bombard-3.813
Token
Token ID198
Feature activation-4.61e-01

Pos logit contributions
zos+3.422
ozo+3.400
sonian+3.264
yx+3.252
ints+2.940
Neg logit contributions
allah-2.613
*)-2.603
 coincide-2.582
 flee-2.519
 bombard-2.479
Token
Token ID198
Feature activation-5.07e-01

Pos logit contributions
zos+1.877
ozo+1.867
yx+1.843
sonian+1.788
ints+1.580
Neg logit contributions
 murd-2.134
 coincide-2.105
 bombard-2.079
allah-2.055
 flee-2.048
TokenNY
Token ID12805
Feature activation-4.51e-01

Pos logit contributions
zos+1.986
ozo+1.975
sonian+1.880
yx+1.871
ints+1.641
Neg logit contributions
*)-2.431
allah-2.427
 coincide-2.406
 flee-2.355
 bombard-2.328
Token comment
Token ID2912
Feature activation-1.32e+00

Pos logit contributions
zos+1.922
yx+1.919
ozo+1.896
sonian+1.843
ints+1.625
Neg logit contributions
allah-2.332
 murd-2.320
 coincide-2.314
*)-2.312
 bombard-2.290
Token on
Token ID319
Feature activation-5.96e-01

Pos logit contributions
yx+4.724
ozo+4.722
zos+4.687
sonian+4.425
rade+3.919
Neg logit contributions
senal-6.611
 murd-6.489
 neighb-6.411
 unnecess-6.335
 conduc-6.191
Token Facebook
Token ID3203
Feature activation-6.80e-01

Pos logit contributions
ozo+1.936
yx+1.924
zos+1.907
sonian+1.852
ints+1.572
Neg logit contributions
 coincide-3.166
 murd-3.159
*)-3.107
 flee-3.105
allah-3.098
Token Share
Token ID8734
Feature activation-1.24e+00

Pos logit contributions
yx+2.589
ozo+2.512
zos+2.498
sonian+2.429
ints+2.185
Neg logit contributions
 murd-3.202
allah-3.156
 coincide-3.143
 bombard-3.131
*)-3.122
Token this
Token ID428
Feature activation-7.44e-01

Pos logit contributions
ozo+4.739
yx+4.672
zos+4.668
sonian+4.450
rade+4.000
Neg logit contributions
 murd-5.707
senal-5.650
 coincide-5.469
allah-5.434
 unnecess-5.389
Token comment
Token ID2912
Feature activation-1.40e+00

Pos logit contributions
yx+2.764
zos+2.724
ozo+2.717
sonian+2.656
ints+2.300
Neg logit contributions
allah-3.459
 murd-3.459
 coincide-3.418
*)-3.417
 bombard-3.389
Token on
Token ID319
Feature activation-7.13e-01

Pos logit contributions
yx+5.083
ozo+5.076
zos+5.041
sonian+4.735
rade+4.223
Neg logit contributions
senal-7.057
 murd-6.836
 neighb-6.728
 unnecess-6.656
 conduc-6.525
Token Twitter
Token ID3009
Feature activation-6.42e-01

Pos logit contributions
zos+2.281
ozo+2.244
yx+2.226
sonian+2.124
 Transformers+1.854
Neg logit contributions
 murd-3.803
allah-3.800
 coincide-3.786
*)-3.760
 bombard-3.706
Token
Token ID198
Feature activation-4.51e-01

Pos logit contributions
zos+3.197
ozo+3.175
sonian+3.046
yx+3.037
ints+2.747
Neg logit contributions
allah-2.416
*)-2.404
 coincide-2.386
 flee-2.329
 murd-2.293
Token
Token ID198
Feature activation-5.13e-01

Pos logit contributions
zos+1.836
ozo+1.826
yx+1.802
sonian+1.750
ints+1.545
Neg logit contributions
 murd-2.088
 coincide-2.060
 bombard-2.035
allah-2.012
 flee-2.006
TokenNY
Token ID12805
Feature activation-4.88e-01

Pos logit contributions
zos+1.994
ozo+1.984
sonian+1.888
yx+1.880
ints+1.647
Neg logit contributions
*)-2.468
allah-2.462
 coincide-2.445
 flee-2.393
 bombard-2.367
Tokenmail
Token ID4529
Feature activation-5.91e-01

Pos logit contributions
ozo+0.819
zos+0.816
sonian+0.794
yx+0.777
ints+0.724
Neg logit contributions
*)-0.289
allah-0.276
 coincide-0.266
 bombard-0.260
 flee-0.256
Token this
Token ID428
Feature activation-2.98e-01

Pos logit contributions
zos+2.921
ozo+2.904
yx+2.797
sonian+2.777
ints+2.519
Neg logit contributions
*)-2.208
allah-2.187
 coincide-2.185
 flee-2.110
 murd-2.093
Token to
Token ID284
Feature activation-1.16e-01

Pos logit contributions
zos+1.237
ozo+1.234
sonian+1.186
yx+1.176
ints+1.041
Neg logit contributions
*)-1.349
allah-1.347
 coincide-1.335
 flee-1.304
 bombard-1.293
Token a
Token ID257
Feature activation+2.50e-02

Pos logit contributions
zos+0.507
ozo+0.506
sonian+0.484
yx+0.482
ints+0.430
Neg logit contributions
*)-0.504
allah-0.502
 coincide-0.497
 flee-0.486
 bombard-0.480
Token friend
Token ID1545
Feature activation-2.88e-01

Pos logit contributions
allah+0.065
*)+0.065
 coincide+0.063
 flee+0.061
 bombard+0.060
Neg logit contributions
ozo-0.152
zos-0.152
sonian-0.147
yx-0.147
ints-0.136
Token Print
Token ID12578
Feature activation-1.36e+00

Pos logit contributions
ozo+1.352
zos+1.348
yx+1.312
sonian+1.308
ints+1.183
Neg logit contributions
allah-1.136
 coincide-1.126
*)-1.123
 flee-1.075
 murd-1.069
Tokenable
Token ID540
Feature activation-6.30e-01

Pos logit contributions
zos+6.229
ozo+6.202
yx+6.024
sonian+5.827
 Correction+5.449
Neg logit contributions
 coincide-5.273
*)-5.203
 murd-5.088
allah-5.066
 bombard-5.061
Token version
Token ID2196
Feature activation-4.76e-01

Pos logit contributions
zos+2.269
ozo+2.237
yx+2.135
sonian+2.107
ints+1.847
Neg logit contributions
*)-3.191
allah-3.157
 coincide-3.156
 bombard-3.056
 murd-3.055
Token<|endoftext|>
Token ID50256
Feature activation+1.67e-01

Pos logit contributions
ozo+2.195
zos+2.186
yx+2.128
sonian+2.114
 Correction+1.923
Neg logit contributions
allah-1.891
*)-1.885
 coincide-1.853
 murd-1.836
 flee-1.803
TokenA
Token ID32
Feature activation+1.30e-01

Pos logit contributions
*)+0.655
allah+0.643
 coincide+0.623
 flee+0.606
 closer+0.602
Neg logit contributions
zos-0.813
ozo-0.808
sonian-0.777
yx-0.768
ints-0.693
Token university
Token ID6403
Feature activation+2.38e-01

Pos logit contributions
*)+0.512
allah+0.512
 coincide+0.505
 flee+0.497
 closer+0.492
Neg logit contributions
ozo-0.620
zos-0.619
sonian-0.589
yx-0.585
ints-0.528
Token Hide
Token ID10415
Feature activation-2.19e-01

Pos logit contributions
*)+0.998
allah+0.982
 coincide+0.971
 flee+0.949
 closer+0.934
Neg logit contributions
zos-1.579
ozo-1.571
sonian-1.522
yx-1.501
ints-1.382
Token Caption
Token ID11260
Feature activation-5.77e-01

Pos logit contributions
ozo+0.742
zos+0.728
sonian+0.706
yx+0.703
ints+0.597
Neg logit contributions
 murd-1.152
 coincide-1.152
allah-1.139
 bombard-1.136
*)-1.130
Token 11
Token ID1367
Feature activation-1.59e-01

Pos logit contributions
zos+1.974
ozo+1.965
yx+1.918
sonian+1.864
ints+1.637
Neg logit contributions
 murd-2.997
allah-2.957
 coincide-2.906
*)-2.897
 flee-2.896
Token of
Token ID286
Feature activation+1.76e-01

Pos logit contributions
ozo+0.682
zos+0.672
yx+0.669
sonian+0.651
ints+0.572
Neg logit contributions
 murd-0.705
 coincide-0.693
 bombard-0.682
allah-0.680
 flee-0.672
Token 15
Token ID1315
Feature activation+2.64e-02

Pos logit contributions
allah+0.744
 coincide+0.733
 murd+0.732
*)+0.724
 bombard+0.721
Neg logit contributions
zos-0.780
ozo-0.775
yx-0.771
sonian-0.752
ints-0.678
Token Photos
Token ID9434
Feature activation-1.36e+00

Pos logit contributions
 coincide+0.148
 murd+0.148
 bombard+0.146
allah+0.145
 flee+0.145
Neg logit contributions
zos-0.082
ozo-0.081
yx-0.080
sonian-0.076
ints-0.066
Token:
Token ID25
Feature activation-1.65e-01

Pos logit contributions
yx+4.852
ozo+4.767
zos+4.696
sonian+4.444
ints+3.898
Neg logit contributions
 murd-7.044
 neighb-7.030
senal-6.937
 conduc-6.825
 rall-6.785
Token Carly
Token ID35982
Feature activation-4.97e-01

Pos logit contributions
yx+0.579
zos+0.577
ozo+0.573
sonian+0.543
 Correction+0.487
Neg logit contributions
 murd-0.841
 coincide-0.839
 bombard-0.830
allah-0.829
*)-0.814
Token Fiorina
Token ID49028
Feature activation-4.26e-01

Pos logit contributions
zos+1.249
ozo+1.234
yx+1.197
sonian+1.146
ints+0.930
Neg logit contributions
 murd-3.031
 coincide-3.019
 bombard-3.015
allah-2.985
*)-2.957
Token's
Token ID338
Feature activation-2.50e-01

Pos logit contributions
zos+1.629
ozo+1.622
yx+1.579
sonian+1.572
ints+1.356
Neg logit contributions
 murd-2.066
allah-2.030
 coincide-2.030
 bombard-2.001
*)-1.981
Token political
Token ID1964
Feature activation-2.45e-02

Pos logit contributions
ozo+1.296
sonian+1.283
zos+1.283
yx+1.260
ints+1.140
Neg logit contributions
allah-0.829
*)-0.827
 murd-0.805
 coincide-0.793
 bombard-0.791
Token comment
Token ID2912
Feature activation-1.18e+00

Pos logit contributions
zos+1.573
yx+1.547
ozo+1.542
sonian+1.496
ints+1.323
Neg logit contributions
allah-1.878
 coincide-1.865
*)-1.862
 murd-1.856
 bombard-1.837
Token on
Token ID319
Feature activation-4.94e-01

Pos logit contributions
ozo+4.342
zos+4.327
yx+4.301
sonian+4.056
rade+3.596
Neg logit contributions
 murd-5.821
senal-5.791
 neighb-5.715
 unnecess-5.626
 bombard-5.522
Token Facebook
Token ID3203
Feature activation-6.19e-01

Pos logit contributions
ozo+1.673
zos+1.658
yx+1.633
sonian+1.588
ints+1.368
Neg logit contributions
 coincide-2.565
*)-2.553
 murd-2.547
allah-2.541
 flee-2.519
Token Share
Token ID8734
Feature activation-1.20e+00

Pos logit contributions
yx+2.293
ozo+2.223
zos+2.218
sonian+2.162
rade+1.946
Neg logit contributions
 murd-2.936
allah-2.919
*)-2.917
 coincide-2.891
 flee-2.867
Token this
Token ID428
Feature activation-7.18e-01

Pos logit contributions
ozo+4.682
zos+4.626
yx+4.585
sonian+4.384
rade+3.960
Neg logit contributions
 murd-5.526
senal-5.372
 coincide-5.323
allah-5.282
 flee-5.230
Token comment
Token ID2912
Feature activation-1.35e+00

Pos logit contributions
yx+2.649
zos+2.630
ozo+2.614
sonian+2.555
ints+2.212
Neg logit contributions
allah-3.372
 murd-3.362
 coincide-3.330
*)-3.325
 bombard-3.297
Token on
Token ID319
Feature activation-6.04e-01

Pos logit contributions
yx+4.923
ozo+4.915
zos+4.888
sonian+4.578
rade+4.102
Neg logit contributions
senal-6.798
 murd-6.650
 neighb-6.537
 unnecess-6.463
 conduc-6.346
Token Twitter
Token ID3009
Feature activation-5.89e-01

Pos logit contributions
zos+2.004
ozo+1.976
yx+1.936
sonian+1.871
 Transformers+1.617
Neg logit contributions
allah-3.186
 coincide-3.173
 murd-3.173
*)-3.159
 bombard-3.101
Token
Token ID198
Feature activation-4.06e-01

Pos logit contributions
zos+2.911
ozo+2.892
sonian+2.775
yx+2.765
ints+2.501
Neg logit contributions
allah-2.234
*)-2.224
 coincide-2.208
 flee-2.154
 bombard-2.120
Token
Token ID198
Feature activation-4.69e-01

Pos logit contributions
zos+1.675
ozo+1.667
yx+1.640
sonian+1.596
ints+1.414
Neg logit contributions
 murd-1.857
 coincide-1.838
 bombard-1.810
allah-1.796
 flee-1.787
TokenNY
Token ID12805
Feature activation-4.79e-01

Pos logit contributions
zos+1.852
ozo+1.842
sonian+1.754
yx+1.746
ints+1.533
Neg logit contributions
*)-2.232
allah-2.228
 coincide-2.208
 flee-2.160
 bombard-2.136
Token Hide
Token ID10415
Feature activation-1.91e-02

Pos logit contributions
*)+1.241
allah+1.219
 coincide+1.205
 flee+1.172
 closer+1.153
Neg logit contributions
zos-2.206
ozo-2.193
sonian-2.134
yx-2.101
ints-1.938
Token Caption
Token ID11260
Feature activation-4.69e-01

Pos logit contributions
ozo+0.067
zos+0.066
sonian+0.063
yx+0.063
ints+0.054
Neg logit contributions
 coincide-0.099
 murd-0.098
allah-0.098
*)-0.097
 bombard-0.097
Token 6
Token ID718
Feature activation+4.66e-02

Pos logit contributions
ozo+2.270
zos+2.257
yx+2.252
sonian+2.149
 Correction+1.959
Neg logit contributions
 murd-1.812
 coincide-1.758
 bombard-1.719
allah-1.710
 flee-1.697
Token of
Token ID286
Feature activation+1.73e-01

Pos logit contributions
 murd+0.197
 coincide+0.197
allah+0.194
 bombard+0.192
 flee+0.191
Neg logit contributions
ozo-0.209
zos-0.207
yx-0.204
sonian-0.200
ints-0.177
Token 15
Token ID1315
Feature activation-5.40e-02

Pos logit contributions
allah+0.737
 coincide+0.723
 murd+0.722
*)+0.716
 bombard+0.712
Neg logit contributions
zos-0.762
ozo-0.759
yx-0.753
sonian-0.735
ints-0.660
Token Photos
Token ID9434
Feature activation-1.34e+00

Pos logit contributions
zos+0.162
ozo+0.160
yx+0.158
sonian+0.149
ints+0.130
Neg logit contributions
 murd-0.311
 coincide-0.309
 bombard-0.305
allah-0.303
 flee-0.303
Token:
Token ID25
Feature activation-1.80e-01

Pos logit contributions
yx+4.789
ozo+4.695
zos+4.626
sonian+4.365
ints+3.842
Neg logit contributions
 neighb-6.984
 murd-6.983
senal-6.854
 conduc-6.772
 rall-6.729
Token Carly
Token ID35982
Feature activation-4.90e-01

Pos logit contributions
yx+0.632
zos+0.629
ozo+0.626
sonian+0.594
 Correction+0.531
Neg logit contributions
 murd-0.916
 coincide-0.913
allah-0.906
 bombard-0.905
*)-0.891
Token Fiorina
Token ID49028
Feature activation-5.56e-01

Pos logit contributions
zos+1.227
ozo+1.214
yx+1.174
sonian+1.123
ints+0.911
Neg logit contributions
 murd-2.991
 coincide-2.978
 bombard-2.972
allah-2.950
*)-2.921
Token's
Token ID338
Feature activation-2.69e-01

Pos logit contributions
zos+2.058
ozo+2.057
yx+2.004
sonian+2.003
ints+1.708
Neg logit contributions
 murd-2.743
allah-2.695
 coincide-2.687
 bombard-2.659
*)-2.612
Token political
Token ID1964
Feature activation-1.26e-02

Pos logit contributions
ozo+1.380
sonian+1.371
zos+1.364
yx+1.341
ints+1.213
Neg logit contributions
allah-0.904
*)-0.896
 murd-0.875
 bombard-0.865
 coincide-0.859
Token Hide
Token ID10415
Feature activation-1.48e-01

Pos logit contributions
*)+1.279
allah+1.249
 coincide+1.236
 flee+1.215
 closer+1.197
Neg logit contributions
zos-1.878
ozo-1.875
sonian-1.816
yx-1.787
ints-1.639
Token Caption
Token ID11260
Feature activation-4.95e-01

Pos logit contributions
ozo+0.511
zos+0.502
yx+0.486
sonian+0.485
ints+0.414
Neg logit contributions
 coincide-0.766
 murd-0.765
allah-0.759
 bombard-0.756
*)-0.754
Token 10
Token ID838
Feature activation+1.09e-01

Pos logit contributions
zos+2.064
ozo+2.022
yx+1.999
sonian+1.940
ints+1.762
Neg logit contributions
 murd-2.216
 coincide-2.200
 bombard-2.166
allah-2.159
*)-2.140
Token of
Token ID286
Feature activation+1.45e-01

Pos logit contributions
 murd+0.457
 coincide+0.456
allah+0.451
 bombard+0.446
*)+0.443
Neg logit contributions
ozo-0.490
zos-0.487
yx-0.477
sonian-0.469
ints-0.415
Token 15
Token ID1315
Feature activation-2.53e-02

Pos logit contributions
allah+0.616
 murd+0.608
 coincide+0.607
*)+0.600
 bombard+0.598
Neg logit contributions
zos-0.639
ozo-0.635
yx-0.632
sonian-0.616
ints-0.555
Token Photos
Token ID9434
Feature activation-1.34e+00

Pos logit contributions
zos+0.078
ozo+0.077
yx+0.076
sonian+0.072
ints+0.062
Neg logit contributions
 coincide-0.142
 murd-0.142
 bombard-0.140
 flee-0.139
allah-0.139
Token:
Token ID25
Feature activation-1.28e-01

Pos logit contributions
yx+4.814
ozo+4.726
zos+4.660
sonian+4.408
ints+3.874
Neg logit contributions
 murd-6.928
 neighb-6.908
senal-6.807
 conduc-6.693
 rall-6.660
Token Carly
Token ID35982
Feature activation-4.52e-01

Pos logit contributions
yx+0.453
zos+0.453
ozo+0.449
sonian+0.426
 Correction+0.380
Neg logit contributions
 murd-0.650
 coincide-0.650
allah-0.642
 bombard-0.642
*)-0.631
Token Fiorina
Token ID49028
Feature activation-4.61e-01

Pos logit contributions
zos+1.154
ozo+1.139
yx+1.105
sonian+1.059
ints+0.864
Neg logit contributions
 murd-2.745
 coincide-2.736
 bombard-2.731
allah-2.708
*)-2.683
Token's
Token ID338
Feature activation-2.38e-01

Pos logit contributions
zos+1.762
ozo+1.751
yx+1.709
sonian+1.702
ints+1.466
Neg logit contributions
 murd-2.249
allah-2.206
 coincide-2.205
 bombard-2.176
*)-2.149
Token political
Token ID1964
Feature activation+1.18e-02

Pos logit contributions
ozo+1.243
zos+1.231
sonian+1.229
yx+1.208
ints+1.093
Neg logit contributions
allah-0.785
*)-0.783
 murd-0.762
 coincide-0.751
 bombard-0.748
Tokenmail
Token ID4529
Feature activation-6.38e-01

Pos logit contributions
ozo+1.296
zos+1.294
sonian+1.251
yx+1.234
ints+1.146
Neg logit contributions
*)-0.481
allah-0.456
 coincide-0.446
 bombard-0.431
 flee-0.429
Token this
Token ID428
Feature activation-3.03e-01

Pos logit contributions
zos+3.126
ozo+3.116
yx+2.999
sonian+2.976
ints+2.699
Neg logit contributions
*)-2.404
 coincide-2.386
allah-2.380
 flee-2.295
 murd-2.283
Token to
Token ID284
Feature activation-1.65e-01

Pos logit contributions
ozo+1.229
zos+1.227
sonian+1.186
yx+1.172
ints+1.034
Neg logit contributions
*)-1.392
allah-1.390
 coincide-1.378
 flee-1.345
 murd-1.339
Token a
Token ID257
Feature activation+1.27e-02

Pos logit contributions
zos+0.705
ozo+0.705
sonian+0.674
yx+0.669
ints+0.598
Neg logit contributions
*)-0.729
allah-0.723
 coincide-0.715
 flee-0.701
 murd-0.695
Token friend
Token ID1545
Feature activation-3.32e-01

Pos logit contributions
allah+0.034
*)+0.033
 coincide+0.032
 flee+0.031
 bombard+0.031
Neg logit contributions
ozo-0.077
zos-0.077
sonian-0.075
yx-0.075
ints-0.069
Token Print
Token ID12578
Feature activation-1.33e+00

Pos logit contributions
ozo+1.503
zos+1.483
sonian+1.461
yx+1.460
ints+1.312
Neg logit contributions
 coincide-1.360
allah-1.356
*)-1.336
 flee-1.286
 murd-1.283
Tokenable
Token ID540
Feature activation-6.39e-01

Pos logit contributions
zos+6.137
ozo+6.106
yx+5.914
sonian+5.743
 Correction+5.343
Neg logit contributions
 coincide-5.146
*)-5.057
 murd-4.938
 bombard-4.923
allah-4.907
Token version
Token ID2196
Feature activation-5.08e-01

Pos logit contributions
zos+2.252
ozo+2.222
yx+2.122
sonian+2.081
ints+1.821
Neg logit contributions
*)-3.273
 coincide-3.248
allah-3.234
 murd-3.140
 bombard-3.139
Token<|endoftext|>
Token ID50256
Feature activation+1.83e-01

Pos logit contributions
ozo+2.363
zos+2.345
yx+2.281
sonian+2.271
 Correction+2.059
Neg logit contributions
*)-2.007
allah-2.005
 coincide-1.979
 murd-1.946
 flee-1.916
TokenRob
Token ID14350
Feature activation+1.92e-01

Pos logit contributions
*)+0.715
allah+0.702
 coincide+0.679
 flee+0.661
 closer+0.657
Neg logit contributions
zos-0.889
ozo-0.884
sonian-0.848
yx-0.839
ints-0.757
Token C
Token ID327
Feature activation-8.28e-03

Pos logit contributions
allah+0.850
*)+0.844
 coincide+0.832
 flee+0.819
 closer+0.805
Neg logit contributions
zos-0.816
ozo-0.804
yx-0.775
sonian-0.773
ints-0.689
Token Hide
Token ID10415
Feature activation-1.68e-01

Pos logit contributions
*)+0.754
allah+0.744
 coincide+0.737
 flee+0.722
 closer+0.711
Neg logit contributions
zos-1.028
ozo-1.025
sonian-0.992
yx-0.975
ints-0.889
Token Caption
Token ID11260
Feature activation-5.67e-01

Pos logit contributions
ozo+0.581
zos+0.571
yx+0.551
sonian+0.551
ints+0.470
Neg logit contributions
 coincide-0.874
 murd-0.872
allah-0.865
 bombard-0.861
*)-0.859
Token 9
Token ID860
Feature activation+3.71e-02

Pos logit contributions
zos+2.222
ozo+2.196
yx+2.173
sonian+2.111
ints+1.873
Neg logit contributions
 murd-2.639
allah-2.634
 coincide-2.617
 bombard-2.579
 flee-2.575
Token of
Token ID286
Feature activation+2.87e-01

Pos logit contributions
 murd+0.160
 coincide+0.158
allah+0.156
 bombard+0.155
 flee+0.153
Neg logit contributions
ozo-0.165
zos-0.163
yx-0.160
sonian-0.157
ints-0.138
Token 15
Token ID1315
Feature activation+3.22e-02

Pos logit contributions
allah+1.203
 coincide+1.187
 murd+1.183
*)+1.175
 bombard+1.166
Neg logit contributions
zos-1.283
ozo-1.277
yx-1.264
sonian-1.235
ints-1.113
Token Photos
Token ID9434
Feature activation-1.33e+00

Pos logit contributions
 coincide+0.181
 murd+0.181
 bombard+0.178
allah+0.177
 flee+0.177
Neg logit contributions
zos-0.100
ozo-0.099
yx-0.097
sonian-0.092
ints-0.080
Token:
Token ID25
Feature activation-1.05e-01

Pos logit contributions
yx+4.765
ozo+4.678
zos+4.612
sonian+4.365
ints+3.838
Neg logit contributions
 murd-6.902
 neighb-6.887
senal-6.777
 conduc-6.677
 rall-6.636
Token Carly
Token ID35982
Feature activation-4.81e-01

Pos logit contributions
zos+0.372
yx+0.371
ozo+0.370
sonian+0.350
 Correction+0.313
Neg logit contributions
 coincide-0.527
 murd-0.527
allah-0.523
 bombard-0.521
*)-0.514
Token Fiorina
Token ID49028
Feature activation-5.22e-01

Pos logit contributions
zos+1.209
ozo+1.196
yx+1.160
sonian+1.109
ints+0.906
Neg logit contributions
 murd-2.931
 coincide-2.919
 bombard-2.914
allah-2.891
*)-2.862
Token's
Token ID338
Feature activation-2.38e-01

Pos logit contributions
zos+1.958
ozo+1.949
yx+1.904
sonian+1.902
ints+1.634
Neg logit contributions
 murd-2.554
allah-2.509
 coincide-2.499
 bombard-2.467
*)-2.445
Token political
Token ID1964
Feature activation+2.17e-02

Pos logit contributions
ozo+1.240
zos+1.228
sonian+1.227
yx+1.206
ints+1.094
Neg logit contributions
allah-0.781
*)-0.780
 murd-0.757
 coincide-0.748
 bombard-0.746
Token comment
Token ID2912
Feature activation-1.09e+00

Pos logit contributions
zos+1.511
ozo+1.473
yx+1.458
sonian+1.422
ints+1.307
Neg logit contributions
allah-1.666
 coincide-1.649
*)-1.629
 flee-1.613
 bombard-1.606
Token on
Token ID319
Feature activation-4.86e-01

Pos logit contributions
zos+3.801
ozo+3.620
yx+3.342
sonian+3.179
rade+3.104
Neg logit contributions
*)-5.501
 murd-5.279
 flee-5.246
 coincide-5.194
allah-5.163
Token Facebook
Token ID3203
Feature activation-5.71e-01

Pos logit contributions
zos+1.866
ozo+1.837
yx+1.757
sonian+1.750
ints+1.545
Neg logit contributions
*)-2.368
allah-2.350
 coincide-2.320
 flee-2.286
 murd-2.256
Token Share
Token ID8734
Feature activation-1.24e+00

Pos logit contributions
zos+2.480
ozo+2.465
yx+2.370
sonian+2.369
ints+2.111
Neg logit contributions
*)-2.475
allah-2.419
 coincide-2.394
 flee-2.325
ş-2.314
Token this
Token ID428
Feature activation-7.02e-01

Pos logit contributions
zos+5.309
ozo+5.248
yx+5.043
sonian+4.953
ints+4.607
Neg logit contributions
*)-5.212
allah-5.182
 coincide-5.177
 murd-5.110
 flee-5.045
Token comment
Token ID2912
Feature activation-1.32e+00

Pos logit contributions
zos+2.695
ozo+2.654
yx+2.630
sonian+2.592
ints+2.284
Neg logit contributions
allah-3.247
*)-3.229
 murd-3.213
 coincide-3.209
 bombard-3.165
Token on
Token ID319
Feature activation-5.69e-01

Pos logit contributions
zos+4.895
ozo+4.842
yx+4.732
sonian+4.441
rade+3.979
Neg logit contributions
 murd-6.426
senal-6.216
 neighb-6.150
 unnecess-6.142
*)-6.043
Token Twitter
Token ID3009
Feature activation-5.03e-01

Pos logit contributions
zos+2.036
ozo+1.978
sonian+1.921
yx+1.906
ints+1.649
Neg logit contributions
*)-2.920
allah-2.864
 coincide-2.856
 murd-2.802
 bombard-2.800
Token
Token ID198
Feature activation-3.72e-01

Pos logit contributions
zos+2.346
ozo+2.339
sonian+2.251
yx+2.248
ints+2.018
Neg logit contributions
*)-2.012
allah-1.999
 coincide-1.979
 flee-1.927
 bombard-1.902
Token
Token ID198
Feature activation-4.41e-01

Pos logit contributions
zos+1.537
ozo+1.534
yx+1.503
sonian+1.475
ints+1.303
Neg logit contributions
 murd-1.690
 coincide-1.673
 bombard-1.646
allah-1.635
 flee-1.626
TokenNY
Token ID12805
Feature activation-4.32e-01

Pos logit contributions
zos+1.937
ozo+1.927
sonian+1.843
yx+1.834
ints+1.635
Neg logit contributions
*)-1.907
allah-1.904
 coincide-1.882
 flee-1.837
 bombard-1.815
Token Hide
Token ID10415
Feature activation-1.70e-01

Pos logit contributions
zos+0.903
ozo+0.899
sonian+0.868
yx+0.866
ints+0.791
Neg logit contributions
*)-0.530
allah-0.528
 coincide-0.522
 flee-0.506
 bombard-0.497
Token Caption
Token ID11260
Feature activation-5.88e-01

Pos logit contributions
ozo+0.583
zos+0.572
yx+0.554
sonian+0.553
ints+0.470
Neg logit contributions
 coincide-0.888
 murd-0.885
allah-0.878
*)-0.874
 bombard-0.871
Token 7
Token ID767
Feature activation-4.04e-02

Pos logit contributions
zos+2.154
ozo+2.108
sonian+2.066
yx+2.058
ints+1.775
Neg logit contributions
 murd-2.909
allah-2.847
 coincide-2.846
 flee-2.834
 bombard-2.831
Token of
Token ID286
Feature activation+1.87e-01

Pos logit contributions
ozo+0.178
zos+0.176
yx+0.174
sonian+0.170
ints+0.149
Neg logit contributions
 murd-0.175
 coincide-0.173
allah-0.171
 bombard-0.170
 flee-0.168
Token 15
Token ID1315
Feature activation+2.82e-03

Pos logit contributions
allah+0.797
 coincide+0.784
 murd+0.783
*)+0.772
 bombard+0.772
Neg logit contributions
zos-0.825
ozo-0.820
yx-0.815
sonian-0.794
ints-0.715
Token Photos
Token ID9434
Feature activation-1.32e+00

Pos logit contributions
 murd+0.016
 coincide+0.016
 bombard+0.016
allah+0.016
 flee+0.016
Neg logit contributions
zos-0.009
ozo-0.009
yx-0.008
sonian-0.008
ints-0.007
Token:
Token ID25
Feature activation-1.60e-01

Pos logit contributions
yx+4.722
ozo+4.624
zos+4.574
sonian+4.321
ints+3.794
Neg logit contributions
 murd-6.851
 neighb-6.844
senal-6.718
 conduc-6.636
 rall-6.583
Token Carly
Token ID35982
Feature activation-4.99e-01

Pos logit contributions
yx+0.562
zos+0.561
ozo+0.557
sonian+0.527
 Correction+0.473
Neg logit contributions
 murd-0.816
 coincide-0.814
allah-0.807
 bombard-0.806
*)-0.792
Token Fiorina
Token ID49028
Feature activation-5.62e-01

Pos logit contributions
zos+1.243
ozo+1.226
yx+1.192
sonian+1.135
ints+0.922
Neg logit contributions
 murd-3.052
 coincide-3.038
 bombard-3.035
allah-3.009
*)-2.977
Token's
Token ID338
Feature activation-2.77e-01

Pos logit contributions
zos+2.071
ozo+2.069
yx+2.023
sonian+2.020
ints+1.719
Neg logit contributions
 murd-2.779
allah-2.727
 coincide-2.722
 bombard-2.694
 flee-2.642
Token political
Token ID1964
Feature activation+3.46e-02

Pos logit contributions
ozo+1.418
sonian+1.410
zos+1.404
yx+1.381
ints+1.248
Neg logit contributions
allah-0.933
*)-0.927
 murd-0.904
 bombard-0.893
 coincide-0.889
Token
Token ID198
Feature activation-5.18e-01

Pos logit contributions
ozo+1.991
zos+1.980
yx+1.971
sonian+1.897
ints+1.696
Neg logit contributions
 coincide-2.142
 murd-2.141
 bombard-2.115
allah-2.107
*)-2.095
Token
Token ID198
Feature activation-4.12e-01

Pos logit contributions
zos+2.049
ozo+2.035
yx+2.025
sonian+1.969
ints+1.747
Neg logit contributions
 murd-2.448
 coincide-2.395
 bombard-2.381
allah-2.338
 flee-2.333
TokenRecommend
Token ID41248
Feature activation-6.01e-01

Pos logit contributions
zos+1.448
ozo+1.432
yx+1.404
sonian+1.366
ints+1.217
Neg logit contributions
 coincide-2.113
 bombard-2.094
 flee-2.089
allah-2.079
*)-2.050
Token Share
Token ID8734
Feature activation-1.08e+00

Pos logit contributions
ozo+2.337
zos+2.310
yx+2.307
sonian+2.234
ints+2.020
Neg logit contributions
*)-2.744
allah-2.725
 coincide-2.713
 bombard-2.684
 murd-2.653
Token this
Token ID428
Feature activation-5.05e-01

Pos logit contributions
zos+4.095
ozo+4.087
yx+4.075
sonian+3.878
rade+3.495
Neg logit contributions
 murd-5.017
senal-4.846
allah-4.842
 coincide-4.820
 flee-4.718
Token comment
Token ID2912
Feature activation-1.32e+00

Pos logit contributions
zos+1.922
yx+1.919
ozo+1.896
sonian+1.843
ints+1.625
Neg logit contributions
allah-2.332
 murd-2.320
 coincide-2.314
*)-2.312
 bombard-2.290
Token on
Token ID319
Feature activation-5.96e-01

Pos logit contributions
yx+4.724
ozo+4.722
zos+4.687
sonian+4.425
rade+3.919
Neg logit contributions
senal-6.611
 murd-6.489
 neighb-6.411
 unnecess-6.335
 conduc-6.191
Token Facebook
Token ID3203
Feature activation-6.80e-01

Pos logit contributions
ozo+1.936
yx+1.924
zos+1.907
sonian+1.852
ints+1.572
Neg logit contributions
 coincide-3.166
 murd-3.159
*)-3.107
 flee-3.105
allah-3.098
Token Share
Token ID8734
Feature activation-1.24e+00

Pos logit contributions
yx+2.589
ozo+2.512
zos+2.498
sonian+2.429
ints+2.185
Neg logit contributions
 murd-3.202
allah-3.156
 coincide-3.143
 bombard-3.131
*)-3.122
Token this
Token ID428
Feature activation-7.44e-01

Pos logit contributions
ozo+4.739
yx+4.672
zos+4.668
sonian+4.450
rade+4.000
Neg logit contributions
 murd-5.707
senal-5.650
 coincide-5.469
allah-5.434
 unnecess-5.389
Token comment
Token ID2912
Feature activation-1.40e+00

Pos logit contributions
yx+2.764
zos+2.724
ozo+2.717
sonian+2.656
ints+2.300
Neg logit contributions
allah-3.459
 murd-3.459
 coincide-3.418
*)-3.417
 bombard-3.389
Token Hide
Token ID10415
Feature activation-1.51e-02

Pos logit contributions
*)+1.064
allah+1.044
 coincide+1.032
 flee+1.004
 closer+0.986
Neg logit contributions
zos-1.904
ozo-1.897
sonian-1.841
yx-1.815
ints-1.679
Token Caption
Token ID11260
Feature activation-4.91e-01

Pos logit contributions
ozo+0.053
zos+0.052
sonian+0.050
yx+0.050
ints+0.043
Neg logit contributions
 coincide-0.078
 murd-0.077
allah-0.077
*)-0.077
 bombard-0.076
Token 8
Token ID807
Feature activation+1.08e-01

Pos logit contributions
zos+2.083
ozo+2.052
yx+2.048
sonian+1.995
 Correction+1.758
Neg logit contributions
 murd-2.189
 bombard-2.149
 coincide-2.138
allah-2.091
 flee-2.069
Token of
Token ID286
Feature activation+2.55e-01

Pos logit contributions
 murd+0.457
 coincide+0.454
allah+0.450
 bombard+0.445
 flee+0.442
Neg logit contributions
ozo-0.486
zos-0.480
yx-0.472
sonian-0.465
ints-0.411
Token 15
Token ID1315
Feature activation+7.50e-02

Pos logit contributions
allah+1.075
 coincide+1.058
 murd+1.056
*)+1.046
 bombard+1.041
Neg logit contributions
zos-1.131
ozo-1.124
yx-1.115
sonian-1.089
ints-0.981
Token Photos
Token ID9434
Feature activation-1.29e+00

Pos logit contributions
 coincide+0.420
 murd+0.419
 bombard+0.413
allah+0.411
 flee+0.411
Neg logit contributions
zos-0.235
ozo-0.233
yx-0.228
sonian-0.216
ints-0.189
Token:
Token ID25
Feature activation-1.01e-01

Pos logit contributions
yx+4.636
ozo+4.562
zos+4.506
sonian+4.259
ints+3.753
Neg logit contributions
 murd-6.641
 neighb-6.603
senal-6.476
 conduc-6.395
 rall-6.366
Token Carly
Token ID35982
Feature activation-4.77e-01

Pos logit contributions
yx+0.359
zos+0.359
ozo+0.357
sonian+0.339
 Correction+0.302
Neg logit contributions
 murd-0.514
 coincide-0.513
allah-0.508
 bombard-0.508
*)-0.500
Token Fiorina
Token ID49028
Feature activation-5.32e-01

Pos logit contributions
zos+1.200
ozo+1.184
yx+1.149
sonian+1.100
ints+0.895
Neg logit contributions
 murd-2.905
 coincide-2.895
 bombard-2.894
allah-2.867
*)-2.841
Token's
Token ID338
Feature activation-2.51e-01

Pos logit contributions
zos+1.981
ozo+1.977
yx+1.933
sonian+1.931
ints+1.649
Neg logit contributions
 murd-2.614
allah-2.569
 coincide-2.561
 bombard-2.534
*)-2.500
Token political
Token ID1964
Feature activation+7.63e-02

Pos logit contributions
ozo+1.299
sonian+1.289
zos+1.287
yx+1.264
ints+1.145
Neg logit contributions
allah-0.833
*)-0.830
 murd-0.809
 bombard-0.799
 coincide-0.797
Token Department
Token ID2732
Feature activation-3.17e-01

Pos logit contributions
zos+0.380
ozo+0.378
sonian+0.359
yx+0.357
ints+0.310
Neg logit contributions
*)-0.517
allah-0.516
 coincide-0.512
 flee-0.502
 bombard-0.497
Token of
Token ID286
Feature activation-2.58e-01

Pos logit contributions
ozo+1.379
zos+1.364
yx+1.327
sonian+1.318
ints+1.154
Neg logit contributions
 murd-1.367
 coincide-1.352
allah-1.347
*)-1.332
 bombard-1.331
Token Justice
Token ID4796
Feature activation-3.14e-01

Pos logit contributions
zos+1.079
ozo+1.077
sonian+1.032
yx+1.016
ints+0.900
Neg logit contributions
allah-1.164
*)-1.163
 coincide-1.150
 flee-1.129
 closer-1.114
Token were
Token ID547
Feature activation-5.99e-01

Pos logit contributions
zos+1.438
ozo+1.436
sonian+1.372
yx+1.371
ints+1.227
Neg logit contributions
*)-1.289
allah-1.282
 coincide-1.267
 flee-1.242
 murd-1.224
Token not
Token ID407
Feature activation-5.08e-01

Pos logit contributions
zos+2.966
ozo+2.940
yx+2.848
sonian+2.818
ints+2.545
Neg logit contributions
*)-2.225
allah-2.217
 coincide-2.201
 flee-2.137
 murd-2.113
Token returned
Token ID4504
Feature activation-1.29e+00

Pos logit contributions
zos+2.239
ozo+2.218
yx+2.158
sonian+2.108
ints+1.886
Neg logit contributions
*)-2.149
allah-2.136
 coincide-2.130
 flee-2.085
 murd-2.068
Token.
Token ID13
Feature activation-4.10e-01

Pos logit contributions
zos+5.795
ozo+5.778
yx+5.586
sonian+5.431
ints+4.891
Neg logit contributions
*)-5.122
 coincide-5.078
allah-5.038
 flee-4.966
 murd-4.962
Token
Token ID198
Feature activation-3.75e-01

Pos logit contributions
zos+1.987
ozo+1.978
sonian+1.904
yx+1.896
ints+1.712
Neg logit contributions
*)-1.585
allah-1.578
 coincide-1.560
 flee-1.521
 closer-1.495
Token
Token ID198
Feature activation-2.94e-01

Pos logit contributions
zos+1.538
ozo+1.538
yx+1.509
sonian+1.472
ints+1.306
Neg logit contributions
 murd-1.736
 coincide-1.699
 bombard-1.681
allah-1.661
 flee-1.655
TokenGal
Token ID26552
Feature activation-1.52e-01

Pos logit contributions
zos+1.391
ozo+1.385
sonian+1.330
yx+1.326
ints+1.191
Neg logit contributions
*)-1.176
allah-1.170
 coincide-1.156
 flee-1.126
 bombard-1.108
Tokenar
Token ID283
Feature activation+1.32e-01

Pos logit contributions
zos+0.586
ozo+0.577
yx+0.545
sonian+0.538
ints+0.473
Neg logit contributions
 coincide-0.755
 murd-0.748
*)-0.746
 bombard-0.734
allah-0.733
Token Hide
Token ID10415
Feature activation-2.06e-01

Pos logit contributions
*)+1.588
allah+1.549
 coincide+1.527
 flee+1.500
 closer+1.477
Neg logit contributions
zos-3.007
ozo-2.984
sonian-2.911
yx-2.860
ints-2.654
Token Caption
Token ID11260
Feature activation-4.69e-01

Pos logit contributions
ozo+0.695
zos+0.682
sonian+0.660
yx+0.660
ints+0.560
Neg logit contributions
 coincide-1.081
 murd-1.080
allah-1.067
 bombard-1.065
*)-1.060
Token 12
Token ID1105
Feature activation+4.94e-02

Pos logit contributions
zos+2.052
yx+1.988
ozo+1.983
sonian+1.927
 Transformers+1.731
Neg logit contributions
 murd-2.037
 coincide-2.004
 bombard-1.992
allah-1.959
*)-1.951
Token of
Token ID286
Feature activation+2.34e-01

Pos logit contributions
 murd+0.212
 coincide+0.210
allah+0.207
 bombard+0.206
 flee+0.204
Neg logit contributions
ozo-0.220
zos-0.217
yx-0.214
sonian-0.210
ints-0.185
Token 15
Token ID1315
Feature activation+7.84e-02

Pos logit contributions
allah+0.987
 coincide+0.974
 murd+0.971
*)+0.964
 bombard+0.956
Neg logit contributions
zos-1.041
ozo-1.035
yx-1.028
sonian-1.002
ints-0.905
Token Photos
Token ID9434
Feature activation-1.28e+00

Pos logit contributions
 coincide+0.437
 murd+0.436
 bombard+0.431
allah+0.429
 flee+0.428
Neg logit contributions
zos-0.247
ozo-0.244
yx-0.238
sonian-0.227
ints-0.199
Token:
Token ID25
Feature activation-1.36e-01

Pos logit contributions
yx+4.655
ozo+4.586
zos+4.522
sonian+4.279
ints+3.764
Neg logit contributions
 murd-6.614
 neighb-6.565
senal-6.474
 conduc-6.368
 rall-6.334
Token Carly
Token ID35982
Feature activation-4.82e-01

Pos logit contributions
zos+0.481
yx+0.481
ozo+0.478
sonian+0.453
 Correction+0.406
Neg logit contributions
 murd-0.692
 coincide-0.691
allah-0.683
 bombard-0.683
*)-0.671
Token Fiorina
Token ID49028
Feature activation-4.10e-01

Pos logit contributions
zos+1.214
ozo+1.199
yx+1.163
sonian+1.112
ints+0.905
Neg logit contributions
 murd-2.941
 coincide-2.928
 bombard-2.925
allah-2.896
*)-2.869
Token's
Token ID338
Feature activation-2.29e-01

Pos logit contributions
zos+1.576
ozo+1.565
yx+1.525
sonian+1.510
ints+1.313
Neg logit contributions
 murd-1.986
allah-1.957
 coincide-1.951
 bombard-1.922
*)-1.905
Token political
Token ID1964
Feature activation+1.41e-02

Pos logit contributions
ozo+1.194
zos+1.181
sonian+1.178
yx+1.160
ints+1.049
Neg logit contributions
allah-0.758
*)-0.756
 murd-0.736
 coincide-0.726
 bombard-0.722
Token Share
Token ID8734
Feature activation-1.05e+00

Pos logit contributions
ozo+2.350
yx+2.323
zos+2.318
sonian+2.248
ints+2.042
Neg logit contributions
*)-2.739
allah-2.726
 coincide-2.709
 bombard-2.680
 murd-2.659
Token this
Token ID428
Feature activation-4.97e-01

Pos logit contributions
ozo+4.047
zos+4.046
yx+4.020
sonian+3.840
rade+3.455
Neg logit contributions
 murd-4.858
allah-4.704
 coincide-4.666
senal-4.652
 flee-4.582
Token comment
Token ID2912
Feature activation-1.23e+00

Pos logit contributions
zos+1.897
yx+1.882
ozo+1.867
sonian+1.814
ints+1.596
Neg logit contributions
allah-2.296
 murd-2.284
*)-2.279
 coincide-2.278
 bombard-2.251
Token on
Token ID319
Feature activation-5.14e-01

Pos logit contributions
ozo+4.491
yx+4.485
zos+4.455
sonian+4.213
rade+3.727
Neg logit contributions
senal-6.143
 murd-6.092
 neighb-5.997
 unnecess-5.918
 conduc-5.798
Token Facebook
Token ID3203
Feature activation-6.77e-01

Pos logit contributions
ozo+1.717
zos+1.697
yx+1.693
sonian+1.641
ints+1.399
Neg logit contributions
 coincide-2.693
 murd-2.685
*)-2.654
allah-2.643
 flee-2.642
Token Share
Token ID8734
Feature activation-1.25e+00

Pos logit contributions
yx+2.577
ozo+2.496
zos+2.484
sonian+2.418
rade+2.171
Neg logit contributions
 murd-3.194
allah-3.152
 coincide-3.127
 bombard-3.115
*)-3.110
Token this
Token ID428
Feature activation-7.74e-01

Pos logit contributions
ozo+4.786
yx+4.733
zos+4.713
sonian+4.502
rade+4.040
Neg logit contributions
 murd-5.812
senal-5.781
 coincide-5.552
allah-5.527
 unnecess-5.508
Token comment
Token ID2912
Feature activation-1.42e+00

Pos logit contributions
yx+2.847
zos+2.805
ozo+2.796
sonian+2.735
ints+2.365
Neg logit contributions
 murd-3.626
allah-3.621
 coincide-3.575
*)-3.575
 bombard-3.549
Token on
Token ID319
Feature activation-7.30e-01

Pos logit contributions
yx+5.127
ozo+5.110
zos+5.052
sonian+4.758
rade+4.248
Neg logit contributions
senal-7.307
 murd-7.051
 neighb-6.963
 unnecess-6.900
 conduc-6.758
Token Twitter
Token ID3009
Feature activation-6.91e-01

Pos logit contributions
zos+2.325
ozo+2.291
yx+2.270
sonian+2.170
 Transformers+1.891
Neg logit contributions
 murd-3.919
allah-3.904
 coincide-3.887
*)-3.855
 bombard-3.813
Token
Token ID198
Feature activation-4.61e-01

Pos logit contributions
zos+3.422
ozo+3.400
sonian+3.264
yx+3.252
ints+2.940
Neg logit contributions
allah-2.613
*)-2.603
 coincide-2.582
 flee-2.519
 bombard-2.479
Token Share
Token ID8734
Feature activation-1.08e+00

Pos logit contributions
yx+3.019
zos+3.005
ozo+2.962
sonian+2.834
 Correction+2.664
Neg logit contributions
allah-3.944
 coincide-3.913
*)-3.904
 accur-3.847
 murd-3.838
Token this
Token ID428
Feature activation-3.69e-01

Pos logit contributions
zos+4.483
ozo+4.276
yx+4.202
sonian+4.060
ints+3.756
Neg logit contributions
*)-4.774
allah-4.729
 coincide-4.691
 murd-4.641
 flee-4.547
Token comment
Token ID2912
Feature activation-1.09e+00

Pos logit contributions
zos+1.511
ozo+1.473
yx+1.458
sonian+1.422
ints+1.307
Neg logit contributions
allah-1.666
 coincide-1.649
*)-1.629
 flee-1.613
 bombard-1.606
Token on
Token ID319
Feature activation-4.86e-01

Pos logit contributions
zos+3.801
ozo+3.620
yx+3.342
sonian+3.179
rade+3.104
Neg logit contributions
*)-5.501
 murd-5.279
 flee-5.246
 coincide-5.194
allah-5.163
Token Facebook
Token ID3203
Feature activation-5.71e-01

Pos logit contributions
zos+1.866
ozo+1.837
yx+1.757
sonian+1.750
ints+1.545
Neg logit contributions
*)-2.368
allah-2.350
 coincide-2.320
 flee-2.286
 murd-2.256
Token Share
Token ID8734
Feature activation-1.24e+00

Pos logit contributions
zos+2.480
ozo+2.465
yx+2.370
sonian+2.369
ints+2.111
Neg logit contributions
*)-2.475
allah-2.419
 coincide-2.394
 flee-2.325
ş-2.314
Token this
Token ID428
Feature activation-7.02e-01

Pos logit contributions
zos+5.309
ozo+5.248
yx+5.043
sonian+4.953
ints+4.607
Neg logit contributions
*)-5.212
allah-5.182
 coincide-5.177
 murd-5.110
 flee-5.045
Token comment
Token ID2912
Feature activation-1.32e+00

Pos logit contributions
zos+2.695
ozo+2.654
yx+2.630
sonian+2.592
ints+2.284
Neg logit contributions
allah-3.247
*)-3.229
 murd-3.213
 coincide-3.209
 bombard-3.165
Token on
Token ID319
Feature activation-5.69e-01

Pos logit contributions
zos+4.895
ozo+4.842
yx+4.732
sonian+4.441
rade+3.979
Neg logit contributions
 murd-6.426
senal-6.216
 neighb-6.150
 unnecess-6.142
*)-6.043
Token Twitter
Token ID3009
Feature activation-5.03e-01

Pos logit contributions
zos+2.036
ozo+1.978
sonian+1.921
yx+1.906
ints+1.649
Neg logit contributions
*)-2.920
allah-2.864
 coincide-2.856
 murd-2.802
 bombard-2.800
Token
Token ID198
Feature activation-3.72e-01

Pos logit contributions
zos+2.346
ozo+2.339
sonian+2.251
yx+2.248
ints+2.018
Neg logit contributions
*)-2.012
allah-1.999
 coincide-1.979
 flee-1.927
 bombard-1.902
Token Hide
Token ID10415
Feature activation-1.32e-01

Pos logit contributions
*)+0.976
allah+0.958
 coincide+0.951
 flee+0.929
 closer+0.913
Neg logit contributions
zos-1.617
ozo-1.612
sonian-1.566
yx-1.541
ints-1.419
Token Caption
Token ID11260
Feature activation-5.22e-01

Pos logit contributions
ozo+0.458
zos+0.452
yx+0.435
sonian+0.432
ints+0.371
Neg logit contributions
 coincide-0.687
 murd-0.684
allah-0.681
 bombard-0.677
*)-0.676
Token 14
Token ID1478
Feature activation+2.17e-01

Pos logit contributions
zos+2.124
ozo+2.100
yx+2.043
sonian+1.950
 Transformers+1.758
Neg logit contributions
allah-2.404
 coincide-2.393
 murd-2.382
 bombard-2.372
 flee-2.298
Token of
Token ID286
Feature activation+3.33e-01

Pos logit contributions
 coincide+0.903
 murd+0.903
allah+0.893
 bombard+0.885
 flee+0.878
Neg logit contributions
ozo-0.981
zos-0.973
yx-0.955
sonian-0.941
ints-0.834
Token 15
Token ID1315
Feature activation+1.18e-01

Pos logit contributions
allah+1.373
 coincide+1.356
 murd+1.343
*)+1.343
 bombard+1.328
Neg logit contributions
zos-1.517
ozo-1.509
yx-1.487
sonian-1.459
ints-1.318
Token Photos
Token ID9434
Feature activation-1.24e+00

Pos logit contributions
 coincide+0.655
 murd+0.650
allah+0.645
 bombard+0.644
 flee+0.641
Neg logit contributions
zos-0.378
ozo-0.372
yx-0.363
sonian-0.347
ints-0.303
Token:
Token ID25
Feature activation-9.31e-02

Pos logit contributions
yx+4.550
ozo+4.487
zos+4.422
sonian+4.191
ints+3.683
Neg logit contributions
 murd-6.362
 neighb-6.305
senal-6.195
 conduc-6.108
 rall-6.081
Token Carly
Token ID35982
Feature activation-4.01e-01

Pos logit contributions
zos+0.332
yx+0.331
ozo+0.330
sonian+0.312
 Correction+0.280
Neg logit contributions
 coincide-0.469
 murd-0.468
allah-0.465
 bombard-0.463
*)-0.457
Token Fiorina
Token ID49028
Feature activation-3.94e-01

Pos logit contributions
zos+1.031
ozo+1.023
yx+0.988
sonian+0.945
ints+0.775
Neg logit contributions
 murd-2.426
 coincide-2.418
 bombard-2.414
allah-2.393
*)-2.371
Token's
Token ID338
Feature activation-1.98e-01

Pos logit contributions
zos+1.510
ozo+1.505
yx+1.469
sonian+1.457
ints+1.259
Neg logit contributions
 murd-1.910
 coincide-1.877
allah-1.877
 bombard-1.854
*)-1.827
Token political
Token ID1964
Feature activation+4.33e-02

Pos logit contributions
ozo+1.036
zos+1.026
sonian+1.023
yx+1.007
ints+0.911
Neg logit contributions
allah-0.650
*)-0.648
 murd-0.628
 coincide-0.622
 bombard-0.617
Token declines
Token ID24459
Feature activation-6.61e-02

Pos logit contributions
*)+0.678
allah+0.676
 coincide+0.665
 flee+0.649
 closer+0.639
Neg logit contributions
zos-0.911
ozo-0.906
sonian-0.870
yx-0.869
ints-0.785
Token to
Token ID284
Feature activation+1.53e-01

Pos logit contributions
ozo+0.306
zos+0.306
sonian+0.294
yx+0.291
ints+0.259
Neg logit contributions
allah-0.270
*)-0.269
 coincide-0.268
 closer-0.262
 flee-0.262
Token disclose
Token ID15771
Feature activation+1.30e-01

Pos logit contributions
*)+0.618
allah+0.616
 coincide+0.610
 flee+0.595
 bombard+0.587
Neg logit contributions
zos-0.709
ozo-0.707
sonian-0.678
yx-0.676
ints-0.606
Token his
Token ID465
Feature activation+7.75e-02

Pos logit contributions
*)+0.510
allah+0.508
 coincide+0.504
 flee+0.492
 closer+0.485
Neg logit contributions
zos-0.620
ozo-0.616
sonian-0.595
yx-0.589
ints-0.530
Token tax
Token ID1687
Feature activation+1.59e-01

Pos logit contributions
*)+0.288
allah+0.287
 coincide+0.284
 flee+0.277
 bombard+0.273
Neg logit contributions
zos-0.387
ozo-0.386
sonian-0.372
yx-0.370
ints-0.334
Token rate
Token ID2494
Feature activation+7.35e-01

Pos logit contributions
*)+0.589
allah+0.587
 coincide+0.584
 flee+0.570
 closer+0.560
Neg logit contributions
zos-0.791
ozo-0.788
sonian-0.764
yx-0.755
ints-0.684
Token,
Token ID11
Feature activation+7.38e-02

Pos logit contributions
*)+2.645
allah+2.626
 coincide+2.604
 closer+2.537
 flee+2.534
Neg logit contributions
zos-3.729
ozo-3.695
sonian-3.637
yx-3.552
ints-3.228
Token stating
Token ID12316
Feature activation+1.08e-01

Pos logit contributions
*)+0.290
allah+0.289
 coincide+0.287
 flee+0.279
 bombard+0.275
Neg logit contributions
zos-0.352
ozo-0.351
sonian-0.337
yx-0.336
ints-0.302
Token in
Token ID287
Feature activation-1.34e-01

Pos logit contributions
*)+0.383
allah+0.383
 coincide+0.377
 flee+0.369
 closer+0.364
Neg logit contributions
zos-0.561
ozo-0.559
sonian-0.543
yx-0.537
ints-0.488
Token an
Token ID281
Feature activation-4.93e-01

Pos logit contributions
zos+0.634
ozo+0.631
sonian+0.613
yx+0.603
ints+0.541
Neg logit contributions
allah-0.533
*)-0.531
 coincide-0.528
 flee-0.517
 closer-0.509
Token ABC
Token ID9738
Feature activation-4.26e-01

Pos logit contributions
zos+2.583
ozo+2.581
sonian+2.474
yx+2.472
ints+2.253
Neg logit contributions
*)-1.686
allah-1.665
 coincide-1.659
 flee-1.601
 bombard-1.596
Token.
Token ID13
Feature activation-4.88e-02

Pos logit contributions
 coincide+0.502
allah+0.501
*)+0.498
 flee+0.485
 closer+0.484
Neg logit contributions
zos-0.717
ozo-0.712
sonian-0.690
yx-0.677
ints-0.619
Token
Token ID198
Feature activation-1.68e-01

Pos logit contributions
zos+0.261
ozo+0.260
sonian+0.252
yx+0.248
ints+0.225
Neg logit contributions
allah-0.166
*)-0.166
 coincide-0.162
 closer-0.158
 flee-0.157
Token
Token ID198
Feature activation-1.36e-01

Pos logit contributions
zos+0.709
ozo+0.707
yx+0.694
sonian+0.678
ints+0.602
Neg logit contributions
 murd-0.749
 coincide-0.747
 bombard-0.737
allah-0.729
 flee-0.726
TokenThere
Token ID1858
Feature activation-8.54e-02

Pos logit contributions
zos+0.701
ozo+0.698
sonian+0.672
yx+0.667
ints+0.604
Neg logit contributions
*)-0.495
allah-0.489
 coincide-0.478
 flee-0.463
 closer-0.459
Token had
Token ID550
Feature activation+4.04e-01

Pos logit contributions
zos+0.444
ozo+0.441
sonian+0.425
yx+0.422
ints+0.383
Neg logit contributions
allah-0.302
*)-0.301
 coincide-0.300
 flee-0.290
 bombard-0.285
Token also
Token ID635
Feature activation+6.09e-01

Pos logit contributions
 coincide+1.267
*)+1.251
allah+1.235
 closer+1.225
 flee+1.189
Neg logit contributions
zos-2.263
ozo-2.235
sonian-2.222
yx-2.123
ints-1.971
Token been
Token ID587
Feature activation+1.20e-01

Pos logit contributions
 coincide+1.755
 closer+1.681
allah+1.672
*)+1.670
 flee+1.631
Neg logit contributions
zos-3.560
sonian-3.550
ozo-3.536
yx-3.347
ints-3.100
Token a
Token ID257
Feature activation+1.45e-01

Pos logit contributions
*)+0.490
allah+0.489
 coincide+0.484
 flee+0.472
 bombard+0.466
Neg logit contributions
zos-0.556
ozo-0.553
sonian-0.531
yx-0.529
ints-0.474
Token 40
Token ID2319
Feature activation-4.01e-01

Pos logit contributions
*)+0.487
allah+0.486
 coincide+0.479
 flee+0.463
 bombard+0.456
Neg logit contributions
zos-0.776
ozo-0.772
yx-0.746
sonian-0.745
ints-0.679
Token per
Token ID583
Feature activation-6.16e-02

Pos logit contributions
zos+2.305
ozo+2.278
sonian+2.233
yx+2.196
ints+2.025
Neg logit contributions
*)-1.204
allah-1.203
 coincide-1.187
 flee-1.155
 closer-1.131
Token cent
Token ID1247
Feature activation+5.95e-02

Pos logit contributions
ozo+0.289
zos+0.289
yx+0.282
sonian+0.275
ints+0.250
Neg logit contributions
 murd-0.245
 coincide-0.244
*)-0.243
allah-0.242
 bombard-0.240
Token-
Token ID12
Feature activation+1.50e-01

Pos logit contributions
allah+0.019
 coincide+0.018
*)+0.018
 flee+0.018
 closer+0.018
Neg logit contributions
zos-0.024
ozo-0.023
sonian-0.023
yx-0.023
ints-0.021
Tokenbe
Token ID1350
Feature activation+1.87e-01

Pos logit contributions
 coincide+0.581
*)+0.580
 murd+0.576
allah+0.574
 bombard+0.568
Neg logit contributions
zos-0.724
ozo-0.724
yx-0.699
sonian-0.684
ints-0.624
Token tack
Token ID6331
Feature activation+3.19e-01

Pos logit contributions
*)+0.665
allah+0.664
 coincide+0.661
 flee+0.647
 closer+0.636
Neg logit contributions
zos-0.958
ozo-0.955
sonian-0.922
yx-0.914
ints-0.831
Tokenler
Token ID1754
Feature activation-1.55e-01

Pos logit contributions
allah+1.108
*)+1.098
 coincide+1.090
 flee+1.071
 closer+1.070
Neg logit contributions
zos-1.682
ozo-1.656
sonian-1.598
yx-1.579
ints-1.451
Token and
Token ID290
Feature activation+3.73e-01

Pos logit contributions
zos+0.751
ozo+0.749
sonian+0.720
yx+0.719
ints+0.648
Neg logit contributions
*)-0.593
allah-0.591
 coincide-0.585
 flee-0.568
 bombard-0.561
Token then
Token ID788
Feature activation+7.41e-01

Pos logit contributions
 coincide+1.405
*)+1.404
allah+1.404
 flee+1.376
 bombard+1.348
Neg logit contributions
zos-1.839
ozo-1.835
sonian-1.763
yx-1.745
ints-1.585
Token accelerated
Token ID23312
Feature activation-1.83e-01

Pos logit contributions
 coincide+2.820
*)+2.815
allah+2.810
 flee+2.787
 closer+2.753
Neg logit contributions
zos-3.593
ozo-3.574
sonian-3.455
yx-3.370
ints-3.097
Token toward
Token ID3812
Feature activation+3.51e-01

Pos logit contributions
zos+0.844
ozo+0.841
yx+0.807
sonian+0.806
ints+0.722
Neg logit contributions
*)-0.742
allah-0.739
 coincide-0.731
 flee-0.709
 bombard-0.702
Token the
Token ID262
Feature activation+3.00e-01

Pos logit contributions
allah+1.444
 coincide+1.438
*)+1.425
 flee+1.412
 closer+1.412
Neg logit contributions
zos-1.621
ozo-1.603
sonian-1.558
yx-1.521
ints-1.366
Token end
Token ID886
Feature activation+4.02e-01

Pos logit contributions
allah+1.103
*)+1.100
 coincide+1.097
 flee+1.085
 closer+1.069
Neg logit contributions
zos-1.517
ozo-1.503
sonian-1.450
yx-1.432
ints-1.303
Token zone
Token ID6516
Feature activation+2.29e-01

Pos logit contributions
allah+1.207
 closer+1.176
 flee+1.170
 coincide+1.170
*)+1.161
Neg logit contributions
zos-2.337
ozo-2.252
sonian-2.248
yx-2.164
ints-2.004
Token 2013
Token ID2211
Feature activation+5.64e-01

Pos logit contributions
*)+1.740
allah+1.735
 coincide+1.721
 flee+1.668
 closer+1.657
Neg logit contributions
zos-2.560
ozo-2.543
sonian-2.462
yx-2.430
ints-2.217
Token (
Token ID357
Feature activation+6.23e-01

Pos logit contributions
*)+2.058
allah+2.044
 coincide+2.035
 flee+1.967
 closer+1.953
Neg logit contributions
zos-2.857
ozo-2.834
sonian-2.733
yx-2.695
ints-2.438
TokenJ
Token ID41
Feature activation+1.05e-01

Pos logit contributions
*)+2.542
allah+2.530
 coincide+2.493
 flee+2.393
 closer+2.369
Neg logit contributions
zos-2.869
ozo-2.863
sonian-2.720
yx-2.706
ints-2.428
TokenU
Token ID52
Feature activation+1.59e-01

Pos logit contributions
*)+0.443
allah+0.442
 coincide+0.435
 flee+0.425
 closer+0.420
Neg logit contributions
zos-0.474
ozo-0.468
sonian-0.451
yx-0.448
ints-0.401
TokenBA
Token ID4339
Feature activation-4.27e-01

Pos logit contributions
*)+0.667
allah+0.666
 coincide+0.649
 flee+0.642
 closer+0.634
Neg logit contributions
zos-0.717
ozo-0.712
yx-0.685
sonian-0.680
ints-0.600
Token)
Token ID8
Feature activation+6.03e-01

Pos logit contributions
zos+2.159
ozo+2.150
sonian+2.070
yx+2.067
ints+1.870
Neg logit contributions
*)-1.570
allah-1.565
 coincide-1.544
 flee-1.502
 bombard-1.476
Token
Token ID784
Feature activation+2.44e-01

Pos logit contributions
allah+2.312
*)+2.278
 coincide+2.237
 flee+2.164
    +2.138
Neg logit contributions
zos-3.013
ozo-2.995
sonian-2.865
yx-2.792
ints-2.535
Token Military
Token ID12842
Feature activation-1.18e-01

Pos logit contributions
allah+1.021
*)+1.017
 coincide+1.008
 flee+0.979
 closer+0.970
Neg logit contributions
zos-1.112
ozo-1.105
sonian-1.059
yx-1.048
ints-0.936
Token clashes
Token ID21022
Feature activation+7.21e-02

Pos logit contributions
zos+0.636
ozo+0.633
sonian+0.611
yx+0.609
ints+0.554
Neg logit contributions
*)-0.395
allah-0.394
 coincide-0.390
 flee-0.379
 bombard-0.374
Token have
Token ID423
Feature activation+3.05e-01

Pos logit contributions
*)+0.289
allah+0.288
 coincide+0.288
 flee+0.279
 bombard+0.275
Neg logit contributions
zos-0.340
ozo-0.339
sonian-0.327
yx-0.324
ints-0.291
Token occurred
Token ID5091
Feature activation+5.28e-01

Pos logit contributions
 coincide+1.143
*)+1.139
allah+1.138
 flee+1.105
 closer+1.093
Neg logit contributions
ozo-1.505
zos-1.503
sonian-1.461
yx-1.434
ints-1.303
Token
Token ID447
Feature activation+1.70e-01

Pos logit contributions
*)+0.158
allah+0.157
 coincide+0.155
 flee+0.151
 bombard+0.148
Neg logit contributions
zos-0.266
ozo-0.265
sonian-0.256
yx-0.255
ints-0.233
Token
Token ID251
Feature activation+1.73e-01

Pos logit contributions
*)+0.670
allah+0.661
 coincide+0.653
 flee+0.637
 closer+0.625
Neg logit contributions
ozo-0.825
zos-0.821
sonian-0.783
yx-0.771
ints-0.696
Token
Token ID198
Feature activation+6.93e-02

Pos logit contributions
*)+0.660
allah+0.657
 coincide+0.647
 flee+0.630
 closer+0.620
Neg logit contributions
zos-0.851
ozo-0.847
sonian-0.816
yx-0.810
ints-0.733
Token
Token ID198
Feature activation+1.12e-01

Pos logit contributions
 coincide+0.298
 murd+0.297
allah+0.295
 bombard+0.293
 flee+0.291
Neg logit contributions
zos-0.304
ozo-0.304
yx-0.296
sonian-0.291
ints-0.258
Tokenhttps
Token ID5450
Feature activation+4.63e-01

Pos logit contributions
*)+0.421
allah+0.416
 coincide+0.408
 flee+0.396
 closer+0.391
Neg logit contributions
zos-0.557
ozo-0.555
sonian-0.534
yx-0.531
ints-0.480
Token://
Token ID1378
Feature activation+6.66e-01

Pos logit contributions
 coincide+1.999
allah+1.996
 murd+1.974
*)+1.974
 flee+1.970
Neg logit contributions
zos-2.028
ozo-2.010
sonian-1.939
yx-1.937
ints-1.711
Tokentwitter
Token ID6956
Feature activation+3.89e-01

Pos logit contributions
allah+2.633
*)+2.586
 coincide+2.507
 flee+2.427
ş+2.388
Neg logit contributions
zos-3.163
ozo-3.138
sonian-3.015
yx-2.966
ints-2.682
Token.
Token ID13
Feature activation+3.28e-01

Pos logit contributions
 coincide+1.523
*)+1.520
allah+1.517
 flee+1.486
 murd+1.482
Neg logit contributions
zos-1.852
ozo-1.852
sonian-1.784
yx-1.783
ints-1.594
Tokencom
Token ID785
Feature activation+2.85e-01

Pos logit contributions
 coincide+1.280
*)+1.264
allah+1.262
 flee+1.257
 bombard+1.251
Neg logit contributions
zos-1.577
ozo-1.574
sonian-1.520
yx-1.516
ints-1.357
Token/
Token ID14
Feature activation+9.65e-02

Pos logit contributions
allah+1.075
*)+1.060
 coincide+1.017
 murd+1.006
 closer+0.995
Neg logit contributions
zos-1.413
ozo-1.394
sonian-1.355
yx-1.340
ints-1.213
Tokenom
Token ID296
Feature activation-2.93e-02

Pos logit contributions
allah+0.412
*)+0.405
 coincide+0.397
 flee+0.388
 murd+0.383
Neg logit contributions
zos-0.431
ozo-0.430
sonian-0.412
yx-0.407
ints-0.365
Token,
Token ID11
Feature activation-3.63e-01

Pos logit contributions
zos+1.392
ozo+1.391
yx+1.357
sonian+1.339
ints+1.171
Neg logit contributions
 murd-1.514
 coincide-1.504
 bombard-1.486
allah-1.459
 flee-1.459
Token The
Token ID383
Feature activation+2.45e-01

Pos logit contributions
zos+1.281
ozo+1.281
yx+1.244
sonian+1.221
ints+1.039
Neg logit contributions
 coincide-1.861
 murd-1.857
 bombard-1.853
allah-1.850
*)-1.841
Token Chronicle
Token ID14160
Feature activation+1.79e-02

Pos logit contributions
allah+1.153
*)+1.149
 coincide+1.144
 flee+1.122
 bombard+1.121
Neg logit contributions
zos-0.979
ozo-0.973
sonian-0.937
yx-0.929
ints-0.808
Token 13
Token ID1511
Feature activation+1.88e-01

Pos logit contributions
allah+0.084
*)+0.083
 coincide+0.083
 flee+0.081
 bombard+0.080
Neg logit contributions
zos-0.072
ozo-0.072
yx-0.068
sonian-0.068
ints-0.060
Token of
Token ID286
Feature activation+4.94e-01

Pos logit contributions
allah+0.773
 murd+0.772
 coincide+0.772
*)+0.760
 bombard+0.756
Neg logit contributions
ozo-0.859
zos-0.847
yx-0.830
sonian-0.823
ints-0.732
Token 15
Token ID1315
Feature activation-9.39e-02

Pos logit contributions
allah+1.975
 coincide+1.952
*)+1.951
 murd+1.911
 flee+1.906
Neg logit contributions
zos-2.320
ozo-2.315
yx-2.250
sonian-2.235
ints-2.020
Token Photo
Token ID5555
Feature activation-4.77e-01

Pos logit contributions
zos+0.306
ozo+0.304
yx+0.289
sonian+0.286
ints+0.244
Neg logit contributions
allah-0.506
 coincide-0.506
*)-0.505
 murd-0.500
 bombard-0.498
Token:
Token ID25
Feature activation+4.17e-01

Pos logit contributions
ozo+1.990
zos+1.974
yx+1.955
sonian+1.906
ints+1.691
Neg logit contributions
 murd-2.155
 coincide-2.107
 bombard-2.089
allah-2.077
 flee-2.057
Token Le
Token ID1004
Feature activation-1.11e-03

Pos logit contributions
allah+1.458
*)+1.430
 coincide+1.418
 flee+1.407
 closer+1.389
Neg logit contributions
zos-2.180
ozo-2.148
sonian-2.082
yx-2.044
ints-1.875
Tokena
Token ID64
Feature activation-7.36e-02

Pos logit contributions
ozo+0.005
zos+0.005
yx+0.005
sonian+0.005
ints+0.004
Neg logit contributions
 coincide-0.005
 murd-0.005
 bombard-0.005
 flee-0.005
*)-0.005
Token Suzuki
Token ID35807
Feature activation-3.19e-01

Pos logit contributions
zos+0.201
ozo+0.198
yx+0.184
sonian+0.184
ints+0.149
Neg logit contributions
allah-0.441
 coincide-0.440
*)-0.439
 murd-0.432
 flee-0.431
Token made
Token ID925
Feature activation-2.58e-01

Pos logit contributions
*)+0.398
allah+0.397
 coincide+0.395
 flee+0.384
 closer+0.379
Neg logit contributions
zos-0.470
ozo-0.468
sonian-0.450
yx-0.447
ints-0.403
Token it
Token ID340
Feature activation+3.11e-01

Pos logit contributions
zos+1.169
ozo+1.162
sonian+1.114
yx+1.112
ints+0.996
Neg logit contributions
*)-1.067
allah-1.066
 coincide-1.055
 flee-1.027
 bombard-1.016
Token to
Token ID284
Feature activation+3.18e-01

Pos logit contributions
*)+1.241
allah+1.239
 coincide+1.229
 flee+1.200
 closer+1.197
Neg logit contributions
zos-1.469
ozo-1.464
sonian-1.403
yx-1.399
ints-1.258
Token its
Token ID663
Feature activation+1.44e-01

Pos logit contributions
allah+1.317
*)+1.317
 coincide+1.309
 flee+1.280
 closer+1.261
Neg logit contributions
ozo-1.439
zos-1.439
sonian-1.381
yx-1.368
ints-1.231
Token final
Token ID2457
Feature activation+2.23e-01

Pos logit contributions
*)+0.600
allah+0.599
 coincide+0.593
 flee+0.578
 bombard+0.571
Neg logit contributions
zos-0.652
ozo-0.649
sonian-0.622
yx-0.620
ints-0.554
Token physical
Token ID3518
Feature activation+2.41e-01

Pos logit contributions
*)+0.871
allah+0.869
 coincide+0.861
 flee+0.841
 closer+0.828
Neg logit contributions
zos-1.071
ozo-1.071
sonian-1.028
yx-1.022
ints-0.922
Token location
Token ID4067
Feature activation+3.01e-01

Pos logit contributions
*)+0.821
allah+0.820
 coincide+0.814
 flee+0.790
 bombard+0.776
Neg logit contributions
zos-1.273
ozo-1.269
sonian-1.222
yx-1.218
ints-1.110
Token and
Token ID290
Feature activation+6.53e-02

Pos logit contributions
*)+1.094
allah+1.093
 coincide+1.087
 flee+1.053
 closer+1.046
Neg logit contributions
zos-1.525
ozo-1.521
sonian-1.464
yx-1.454
ints-1.319
Token is
Token ID318
Feature activation+2.42e-01

Pos logit contributions
*)+0.262
allah+0.261
 coincide+0.259
 flee+0.253
 bombard+0.249
Neg logit contributions
zos-0.306
ozo-0.305
sonian-0.293
yx-0.291
ints-0.262
Token now
Token ID783
Feature activation+6.19e-02

Pos logit contributions
*)+0.901
allah+0.897
 coincide+0.889
 flee+0.866
 closer+0.852
Neg logit contributions
zos-1.203
ozo-1.198
sonian-1.153
yx-1.148
ints-1.040
Token redundant
Token ID30806
Feature activation+3.02e-01

Pos logit contributions
*)+0.237
allah+0.237
 coincide+0.234
 flee+0.228
 bombard+0.225
Neg logit contributions
zos-0.301
ozo-0.300
sonian-0.288
yx-0.287
ints-0.259
Token report
Token ID989
Feature activation-1.37e-01

Pos logit contributions
zos+1.423
ozo+1.413
sonian+1.361
yx+1.360
ints+1.230
Neg logit contributions
allah-1.057
*)-1.053
 coincide-1.039
 flee-1.010
 bombard-0.997
Token,
Token ID11
Feature activation+7.48e-03

Pos logit contributions
ozo+0.697
zos+0.697
sonian+0.669
yx+0.667
ints+0.605
Neg logit contributions
allah-0.500
*)-0.498
 coincide-0.492
 flee-0.479
 closer-0.474
Token "
Token ID366
Feature activation-1.54e-01

Pos logit contributions
allah+0.021
*)+0.021
 coincide+0.020
 closer+0.020
 flee+0.020
Neg logit contributions
ozo-0.045
zos-0.044
sonian-0.043
yx-0.043
ints-0.039
Tokenhe
Token ID258
Feature activation-4.13e-01

Pos logit contributions
zos+0.752
ozo+0.749
sonian+0.720
yx+0.718
ints+0.647
Neg logit contributions
*)-0.591
allah-0.590
 coincide-0.585
 flee-0.569
 bombard-0.561
Token pointed
Token ID6235
Feature activation-7.31e-02

Pos logit contributions
zos+1.792
ozo+1.781
sonian+1.706
yx+1.699
ints+1.513
Neg logit contributions
*)-1.798
allah-1.792
 coincide-1.780
 flee-1.738
 bombard-1.717
Token to
Token ID284
Feature activation+2.11e-01

Pos logit contributions
ozo+0.339
zos+0.339
sonian+0.326
yx+0.323
ints+0.290
Neg logit contributions
*)-0.301
allah-0.301
 coincide-0.296
 flee-0.290
 closer-0.290
Token '
Token ID705
Feature activation+1.47e-01

Pos logit contributions
*)+0.845
allah+0.843
 coincide+0.838
 flee+0.815
 closer+0.808
Neg logit contributions
zos-0.991
ozo-0.990
sonian-0.959
yx-0.943
ints-0.844
Tokentwo
Token ID11545
Feature activation-4.24e-02

Pos logit contributions
*)+0.567
allah+0.563
 coincide+0.549
 flee+0.532
 closer+0.530
Neg logit contributions
zos-0.717
ozo-0.717
sonian-0.691
yx-0.681
ints-0.613
Token African
Token ID5510
Feature activation-1.08e-03

Pos logit contributions
zos+0.209
ozo+0.209
sonian+0.201
yx+0.200
ints+0.181
Neg logit contributions
*)-0.160
allah-0.159
 coincide-0.158
 flee-0.153
 bombard-0.151
Token-
Token ID12
Feature activation+1.71e-01

Pos logit contributions
zos+0.005
ozo+0.005
sonian+0.005
yx+0.005
ints+0.004
Neg logit contributions
*)-0.004
allah-0.004
 coincide-0.004
 flee-0.004
 bombard-0.004
TokenAmericans
Token ID17636
Feature activation+7.16e-01

Pos logit contributions
*)+0.664
allah+0.661
 coincide+0.648
 flee+0.635
 closer+0.626
Neg logit contributions
ozo-0.824
zos-0.822
sonian-0.794
yx-0.787
ints-0.708
Token you
Token ID345
Feature activation+2.90e-02

Pos logit contributions
zos+1.031
ozo+1.027
sonian+0.983
yx+0.981
ints+0.877
Neg logit contributions
*)-0.944
allah-0.941
 coincide-0.933
 flee-0.910
 bombard-0.899
Token've
Token ID1053
Feature activation-2.10e-01

Pos logit contributions
allah+0.114
*)+0.114
 coincide+0.112
 flee+0.109
 bombard+0.108
Neg logit contributions
zos-0.138
ozo-0.138
yx-0.132
sonian-0.132
ints-0.119
Token read
Token ID1100
Feature activation-3.73e-02

Pos logit contributions
zos+0.979
ozo+0.974
yx+0.935
sonian+0.929
ints+0.834
Neg logit contributions
allah-0.841
*)-0.836
 coincide-0.824
 flee-0.804
 bombard-0.794
Token the
Token ID262
Feature activation+1.98e-02

Pos logit contributions
zos+0.180
ozo+0.178
sonian+0.172
yx+0.170
ints+0.154
Neg logit contributions
*)-0.146
allah-0.145
 coincide-0.144
 flee-0.140
 closer-0.139
Token legal
Token ID2742
Feature activation-1.92e-01

Pos logit contributions
*)+0.082
allah+0.082
 coincide+0.081
 flee+0.080
 closer+0.079
Neg logit contributions
zos-0.090
ozo-0.090
sonian-0.087
yx-0.086
ints-0.077
Token documents
Token ID4963
Feature activation+1.59e-01

Pos logit contributions
zos+0.957
ozo+0.952
sonian+0.917
yx+0.913
ints+0.826
Neg logit contributions
*)-0.713
allah-0.711
 coincide-0.705
 flee-0.686
 bombard-0.676
Token,
Token ID11
Feature activation-6.14e-02

Pos logit contributions
*)+0.612
allah+0.610
 coincide+0.605
 flee+0.589
 bombard+0.581
Neg logit contributions
zos-0.772
ozo-0.768
sonian-0.740
yx-0.736
ints-0.664
Token maybe
Token ID3863
Feature activation-1.62e-02

Pos logit contributions
zos+0.307
ozo+0.305
sonian+0.295
yx+0.293
ints+0.265
Neg logit contributions
*)-0.227
allah-0.227
 coincide-0.225
 flee-0.219
 bombard-0.216
Token not
Token ID407
Feature activation-3.01e-01

Pos logit contributions
zos+0.078
ozo+0.078
sonian+0.075
yx+0.075
ints+0.067
Neg logit contributions
*)-0.063
allah-0.063
 coincide-0.062
 flee-0.061
 closer-0.060
Token.
Token ID13
Feature activation-1.36e-01

Pos logit contributions
zos+1.516
ozo+1.509
sonian+1.457
yx+1.449
ints+1.311
Neg logit contributions
allah-1.108
*)-1.105
 coincide-1.095
 flee-1.064
 bombard-1.048
Token Here
Token ID3423
Feature activation-6.18e-02

Pos logit contributions
zos+0.612
ozo+0.610
yx+0.583
sonian+0.583
ints+0.519
Neg logit contributions
*)-0.568
allah-0.568
 coincide-0.565
 flee-0.551
 bombard-0.546
Token 78
Token ID8699
Feature activation+4.64e-02

Pos logit contributions
zos+0.192
ozo+0.191
sonian+0.183
yx+0.182
ints+0.163
Neg logit contributions
*)-0.173
allah-0.172
 coincide-0.171
 flee-0.167
 bombard-0.165
Token-
Token ID12
Feature activation-6.21e-02

Pos logit contributions
*)+0.180
allah+0.176
 coincide+0.174
 flee+0.172
 closer+0.168
Neg logit contributions
zos-0.228
ozo-0.223
sonian-0.214
yx-0.211
ints-0.195
Tokenyear
Token ID1941
Feature activation+1.11e-01

Pos logit contributions
zos+0.296
ozo+0.294
sonian+0.276
yx+0.276
ints+0.251
Neg logit contributions
*)-0.250
allah-0.247
 coincide-0.239
 flee-0.237
quart-0.237
Token-
Token ID12
Feature activation+2.18e-01

Pos logit contributions
*)+0.412
allah+0.409
 coincide+0.408
 flee+0.403
 bombard+0.396
Neg logit contributions
zos-0.553
ozo-0.542
sonian-0.522
yx-0.521
ints-0.481
Tokenold
Token ID727
Feature activation+2.57e-01

Pos logit contributions
 coincide+1.419
allah+1.406
*)+1.404
 flee+1.397
 murd+1.396
Neg logit contributions
ozo-0.486
zos-0.486
yx-0.453
sonian-0.447
ints-0.339
Token developer
Token ID8517
Feature activation-1.01e-01

Pos logit contributions
*)+1.030
allah+1.019
 coincide+1.016
 flee+1.000
 closer+0.986
Neg logit contributions
zos-1.211
ozo-1.200
sonian-1.147
yx-1.141
ints-1.040
Token picked
Token ID6497
Feature activation-2.82e-01

Pos logit contributions
zos+0.461
ozo+0.459
sonian+0.439
yx+0.438
ints+0.393
Neg logit contributions
*)-0.417
allah-0.415
 coincide-0.411
 flee-0.402
 bombard-0.396
Token up
Token ID510
Feature activation-5.52e-01

Pos logit contributions
zos+1.326
ozo+1.320
sonian+1.267
yx+1.262
ints+1.134
Neg logit contributions
*)-1.129
allah-1.126
 coincide-1.116
 flee-1.088
 bombard-1.073
Token the
Token ID262
Feature activation-1.12e-01

Pos logit contributions
zos+2.454
ozo+2.453
sonian+2.362
yx+2.359
ints+2.101
Neg logit contributions
*)-2.325
allah-2.315
 coincide-2.290
 flee-2.229
 bombard-2.208
Token former
Token ID1966
Feature activation-2.71e-02

Pos logit contributions
zos+0.549
ozo+0.547
sonian+0.525
yx+0.523
ints+0.472
Neg logit contributions
*)-0.426
allah-0.425
 coincide-0.422
 flee-0.412
 bombard-0.405
Token Federal
Token ID5618
Feature activation-1.97e-01

Pos logit contributions
zos+0.118
ozo+0.117
sonian+0.112
yx+0.111
ints+0.099
Neg logit contributions
*)-0.119
allah-0.118
 coincide-0.117
 flee-0.115
 bombard-0.114
Token months
Token ID1933
Feature activation-2.34e-01

Pos logit contributions
zos+0.182
ozo+0.181
sonian+0.175
yx+0.174
ints+0.160
Neg logit contributions
*)-0.098
 coincide-0.097
allah-0.097
 flee-0.094
 closer-0.093
Token for
Token ID329
Feature activation-2.56e-02

Pos logit contributions
zos+1.150
ozo+1.148
sonian+1.103
yx+1.100
ints+0.994
Neg logit contributions
*)-0.879
allah-0.878
 coincide-0.867
 flee-0.844
 bombard-0.833
Token the
Token ID262
Feature activation-7.70e-02

Pos logit contributions
zos+0.108
ozo+0.107
sonian+0.104
yx+0.101
ints+0.090
Neg logit contributions
*)-0.115
allah-0.115
 coincide-0.114
 flee-0.112
 closer-0.111
Token smaller
Token ID4833
Feature activation-1.54e-01

Pos logit contributions
zos+0.277
ozo+0.275
sonian+0.262
yx+0.259
ints+0.225
Neg logit contributions
*)-0.393
allah-0.392
 coincide-0.390
 flee-0.384
 closer-0.379
Token Prot
Token ID5038
Feature activation-1.27e-01

Pos logit contributions
zos+0.614
ozo+0.609
sonian+0.582
yx+0.578
ints+0.509
Neg logit contributions
*)-0.729
allah-0.727
 coincide-0.723
 flee-0.709
 bombard-0.700
Tokenoc
Token ID420
Feature activation-4.36e-01

Pos logit contributions
zos+0.532
ozo+0.532
sonian+0.505
yx+0.504
ints+0.446
Neg logit contributions
*)-0.572
allah-0.570
 coincide-0.567
 flee-0.554
 bombard-0.549
Tokener
Token ID263
Feature activation+9.20e-02

Pos logit contributions
ozo+1.952
zos+1.949
yx+1.870
sonian+1.818
ints+1.634
Neg logit contributions
 murd-1.907
 coincide-1.858
 bombard-1.829
*)-1.821
allah-1.813
Tokenat
Token ID265
Feature activation-3.96e-01

Pos logit contributions
*)+0.365
allah+0.364
 coincide+0.361
 flee+0.351
 bombard+0.347
Neg logit contributions
zos-0.435
ozo-0.433
sonian-0.416
yx-0.414
ints-0.372
Tokenops
Token ID2840
Feature activation-4.51e-01

Pos logit contributions
ozo+0.888
zos+0.866
yx+0.791
sonian+0.717
ints+0.582
Neg logit contributions
 murd-2.635
 coincide-2.627
*)-2.613
allah-2.583
 flee-2.580
Token hatch
Token ID25834
Feature activation-4.54e-01

Pos logit contributions
ozo+2.163
zos+2.161
sonian+2.069
yx+2.068
ints+1.858
Neg logit contributions
allah-1.754
*)-1.754
 coincide-1.733
 flee-1.682
 bombard-1.665
Tokenlings
Token ID17783
Feature activation-2.77e-01

Pos logit contributions
zos+2.132
ozo+2.123
sonian+2.036
yx+2.031
ints+1.823
Neg logit contributions
*)-1.822
allah-1.818
 coincide-1.800
 flee-1.753
 bombard-1.732
Token at
Token ID379
Feature activation+1.65e-01

Pos logit contributions
*)+1.259
 coincide+1.251
allah+1.250
 flee+1.222
 closer+1.213
Neg logit contributions
ozo-1.294
zos-1.286
sonian-1.237
yx-1.226
ints-1.088
Token around
Token ID1088
Feature activation+2.11e-01

Pos logit contributions
allah+0.665
*)+0.664
 coincide+0.660
 flee+0.643
 closer+0.635
Neg logit contributions
ozo-0.771
zos-0.771
sonian-0.739
yx-0.734
ints-0.657
Token 30
Token ID1542
Feature activation+1.03e-01

Pos logit contributions
*)+0.847
allah+0.845
 coincide+0.839
 flee+0.817
 bombard+0.806
Neg logit contributions
zos-0.986
ozo-0.983
sonian-0.944
yx-0.940
ints-0.843
Token.
Token ID13
Feature activation-2.10e-01

Pos logit contributions
*)+0.377
allah+0.376
 coincide+0.372
 flee+0.362
 bombard+0.356
Neg logit contributions
zos-0.523
ozo-0.521
sonian-0.503
yx-0.498
ints-0.452
Token After
Token ID2293
Feature activation-1.89e-01

Pos logit contributions
zos+0.928
ozo+0.920
yx+0.881
sonian+0.880
ints+0.784
Neg logit contributions
allah-0.901
*)-0.900
 coincide-0.893
 flee-0.873
 bombard-0.864
Token I
Token ID314
Feature activation-1.67e-01

Pos logit contributions
zos+0.873
ozo+0.865
yx+0.832
sonian+0.830
ints+0.749
Neg logit contributions
*)-0.766
allah-0.761
 coincide-0.754
 flee-0.734
 murd-0.727
Token joined
Token ID5399
Feature activation-2.46e-01

Pos logit contributions
zos+0.747
ozo+0.736
yx+0.710
sonian+0.705
ints+0.635
Neg logit contributions
allah-0.701
*)-0.698
 coincide-0.693
 flee-0.670
 murd-0.669
Token Pf
Token ID38477
Feature activation+1.27e-01

Pos logit contributions
zos+1.093
ozo+1.087
sonian+1.040
yx+1.038
ints+0.925
Neg logit contributions
*)-1.052
allah-1.049
 coincide-1.039
 flee-1.014
 bombard-1.002
Tokenizer
Token ID7509
Feature activation-4.78e-01

Pos logit contributions
allah+0.478
*)+0.475
 coincide+0.465
 flee+0.456
 closer+0.450
Neg logit contributions
zos-0.627
ozo-0.626
sonian-0.611
yx-0.607
ints-0.540
Token Pharmaceutical
Token ID37175
Feature activation-1.83e-01

Pos logit contributions
zos+2.256
ozo+2.246
sonian+2.151
yx+2.146
ints+1.930
Neg logit contributions
*)-1.901
allah-1.897
 coincide-1.881
 flee-1.830
 bombard-1.810
Tokens
Token ID82
Feature activation-2.94e-01

Pos logit contributions
zos+0.783
ozo+0.779
sonian+0.755
yx+0.743
ints+0.662
Neg logit contributions
*)-0.809
allah-0.805
 coincide-0.797
 flee-0.784
 closer-0.774
Token chickens
Token ID25972
Feature activation+2.43e-01

Pos logit contributions
zos+2.121
ozo+2.116
yx+2.040
sonian+2.034
ints+1.847
Neg logit contributions
*)-1.462
allah-1.454
 coincide-1.433
 flee-1.389
 bombard-1.370
Token can
Token ID460
Feature activation-5.45e-02

Pos logit contributions
*)+0.858
allah+0.853
 coincide+0.849
 flee+0.827
 bombard+0.812
Neg logit contributions
zos-1.259
ozo-1.251
sonian-1.209
yx-1.202
ints-1.091
Token do
Token ID466
Feature activation-8.44e-02

Pos logit contributions
ozo+0.284
zos+0.283
yx+0.273
sonian+0.272
ints+0.248
Neg logit contributions
allah-0.189
*)-0.188
 coincide-0.184
 flee-0.177
 murd-0.177
Token math
Token ID10688
Feature activation-4.54e-01

Pos logit contributions
zos+0.402
ozo+0.401
sonian+0.386
yx+0.384
ints+0.345
Neg logit contributions
*)-0.332
allah-0.331
 coincide-0.328
 flee-0.320
 bombard-0.316
Token?
Token ID30
Feature activation-5.53e-02

Pos logit contributions
zos+2.338
ozo+2.324
sonian+2.242
yx+2.240
ints+2.029
Neg logit contributions
*)-1.607
allah-1.605
 coincide-1.587
 flee-1.539
 bombard-1.512
Token They
Token ID1119
Feature activation-2.53e-01

Pos logit contributions
zos+0.244
ozo+0.243
sonian+0.232
yx+0.232
ints+0.207
Neg logit contributions
allah-0.236
*)-0.235
 coincide-0.234
 flee-0.228
 bombard-0.226
Token might
Token ID1244
Feature activation-3.89e-01

Pos logit contributions
zos+1.171
ozo+1.162
yx+1.118
sonian+1.106
ints+1.005
Neg logit contributions
allah-1.019
*)-1.015
 coincide-0.993
 murd-0.967
 flee-0.958
Token not
Token ID407
Feature activation-1.39e-01

Pos logit contributions
zos+1.849
ozo+1.846
yx+1.788
sonian+1.766
ints+1.609
Neg logit contributions
allah-1.491
*)-1.480
 coincide-1.445
 murd-1.411
 flee-1.393
Token
Token ID564
Feature activation+5.75e-02

Pos logit contributions
zos+0.691
ozo+0.690
yx+0.665
sonian+0.659
ints+0.601
Neg logit contributions
allah-0.513
*)-0.513
 coincide-0.502
 murd-0.488
 flee-0.484
Token
Token ID250
Feature activation+1.14e-01

Pos logit contributions
*)+0.270
allah+0.266
 coincide+0.265
 flee+0.258
 closer+0.255
Neg logit contributions
ozo-0.235
zos-0.235
sonian-0.225
yx-0.220
ints-0.193
Tokensc
Token ID1416
Feature activation+1.03e-01

Pos logit contributions
*)+0.463
allah+0.461
 coincide+0.452
 flee+0.442
 murd+0.434
Neg logit contributions
zos-0.538
ozo-0.536
sonian-0.519
yx-0.511
ints-0.456
Token his
Token ID465
Feature activation+8.24e-02

Pos logit contributions
zos+0.103
ozo+0.103
sonian+0.099
yx+0.098
ints+0.087
Neg logit contributions
*)-0.095
allah-0.094
 coincide-0.093
 flee-0.091
 closer-0.090
Token role
Token ID2597
Feature activation-3.03e-01

Pos logit contributions
*)+0.304
allah+0.302
 coincide+0.299
 flee+0.294
 closer+0.291
Neg logit contributions
zos-0.415
ozo-0.414
sonian-0.402
yx-0.395
ints-0.356
Token in
Token ID287
Feature activation-1.46e-01

Pos logit contributions
zos+1.479
ozo+1.468
sonian+1.447
yx+1.410
ints+1.264
Neg logit contributions
*)-1.188
allah-1.171
 coincide-1.159
 closer-1.147
 flee-1.128
Token So
Token ID1406
Feature activation-1.87e-02

Pos logit contributions
zos+0.645
ozo+0.641
sonian+0.618
yx+0.610
ints+0.545
Neg logit contributions
*)-0.625
allah-0.622
 coincide-0.615
 flee-0.602
 closer-0.596
Tokenap
Token ID499
Feature activation+6.94e-03

Pos logit contributions
zos+0.080
ozo+0.080
sonian+0.077
yx+0.076
ints+0.068
Neg logit contributions
allah-0.082
*)-0.082
 coincide-0.080
 flee-0.079
 murd-0.078
Token.)
Token ID2014
Feature activation-5.37e-01

Pos logit contributions
*)+0.033
allah+0.033
 coincide+0.033
 flee+0.032
 closer+0.032
Neg logit contributions
zos-0.027
ozo-0.027
sonian-0.026
yx-0.026
ints-0.022
Token Twenty
Token ID22381
Feature activation-2.76e-01

Pos logit contributions
zos+2.444
ozo+2.429
sonian+2.337
yx+2.328
ints+2.081
Neg logit contributions
allah-2.210
 coincide-2.194
*)-2.187
 flee-2.140
 bombard-2.124
Token-
Token ID12
Feature activation-4.19e-01

Pos logit contributions
zos+1.295
ozo+1.290
sonian+1.237
yx+1.232
ints+1.107
Neg logit contributions
*)-1.108
allah-1.101
 coincide-1.095
 flee-1.065
 bombard-1.048
Tokeneight
Token ID26022
Feature activation-4.18e-01

Pos logit contributions
zos+1.957
ozo+1.951
sonian+1.874
yx+1.870
ints+1.679
Neg logit contributions
*)-1.675
allah-1.674
 coincide-1.663
 flee-1.623
 bombard-1.608
Token of
Token ID286
Feature activation-2.59e-01

Pos logit contributions
zos+2.010
ozo+2.000
sonian+1.922
yx+1.914
ints+1.724
Neg logit contributions
*)-1.636
allah-1.627
 coincide-1.616
 flee-1.573
 bombard-1.549
Token the
Token ID262
Feature activation-3.91e-01

Pos logit contributions
zos+1.199
ozo+1.197
sonian+1.148
yx+1.141
ints+1.022
Neg logit contributions
*)-1.059
allah-1.056
 coincide-1.047
 flee-1.023
 closer-1.009
Token?
Token ID30
Feature activation-2.48e-01

Pos logit contributions
zos+0.199
ozo+0.198
sonian+0.191
yx+0.190
ints+0.173
Neg logit contributions
*)-0.135
allah-0.135
 coincide-0.133
 flee-0.129
 bombard-0.127
Token To
Token ID1675
Feature activation+1.72e-01

Pos logit contributions
zos+1.151
ozo+1.146
sonian+1.099
yx+1.095
ints+0.982
Neg logit contributions
*)-1.007
allah-1.004
 coincide-0.995
 flee-0.970
 bombard-0.957
Token Destiny
Token ID17886
Feature activation+2.13e-01

Pos logit contributions
allah+0.620
*)+0.619
 coincide+0.619
 flee+0.599
 bombard+0.590
Neg logit contributions
zos-0.878
ozo-0.876
sonian-0.847
yx-0.838
ints-0.760
Token,
Token ID11
Feature activation+1.84e-01

Pos logit contributions
*)+0.814
allah+0.812
 coincide+0.808
 flee+0.785
 closer+0.775
Neg logit contributions
zos-1.046
ozo-1.036
sonian-0.997
yx-0.991
ints-0.896
Token the
Token ID262
Feature activation+1.54e-02

Pos logit contributions
*)+0.839
 coincide+0.835
allah+0.834
 flee+0.816
 closer+0.814
Neg logit contributions
zos-0.768
ozo-0.757
sonian-0.729
yx-0.719
ints-0.645
Token developer
Token ID8517
Feature activation-2.05e-01

Pos logit contributions
*)+0.060
allah+0.060
 coincide+0.059
 flee+0.058
 closer+0.057
Neg logit contributions
zos-0.074
ozo-0.074
sonian-0.071
yx-0.070
ints-0.064
Token's
Token ID338
Feature activation-1.01e-01

Pos logit contributions
zos+0.910
ozo+0.906
sonian+0.868
yx+0.864
ints+0.770
Neg logit contributions
*)-0.870
allah-0.868
 coincide-0.861
 flee-0.840
 bombard-0.829
Token next
Token ID1306
Feature activation+1.54e-01

Pos logit contributions
zos+0.485
ozo+0.482
sonian+0.463
yx+0.460
ints+0.415
Neg logit contributions
*)-0.397
allah-0.396
 coincide-0.394
 flee-0.384
 bombard-0.378
Token series
Token ID2168
Feature activation+1.22e-01

Pos logit contributions
 coincide+0.626
*)+0.626
allah+0.625
 flee+0.610
 closer+0.604
Neg logit contributions
zos-0.716
ozo-0.713
sonian-0.685
yx-0.674
ints-0.612
Token of
Token ID286
Feature activation+6.13e-01

Pos logit contributions
*)+0.447
 coincide+0.446
allah+0.446
 flee+0.432
 closer+0.429
Neg logit contributions
zos-0.621
ozo-0.618
sonian-0.593
yx-0.589
ints-0.539
Token games
Token ID1830
Feature activation+1.34e-01

Pos logit contributions
 coincide+2.560
allah+2.544
*)+2.521
 flee+2.493
 closer+2.487
Neg logit contributions
zos-2.773
ozo-2.739
sonian-2.628
yx-2.550
ints-2.322
Token Any
Token ID4377
Feature activation+2.76e-02

Pos logit contributions
zos+0.446
ozo+0.443
sonian+0.427
yx+0.423
ints+0.383
Neg logit contributions
*)-0.356
allah-0.354
 coincide-0.349
 flee-0.340
 bombard-0.334
Token reproduction
Token ID20728
Feature activation-4.75e-02

Pos logit contributions
*)+0.105
allah+0.105
 coincide+0.104
 flee+0.101
 closer+0.101
Neg logit contributions
zos-0.136
ozo-0.135
sonian-0.129
yx-0.128
ints-0.116
Token,
Token ID11
Feature activation-7.12e-01

Pos logit contributions
zos+0.247
ozo+0.245
sonian+0.236
yx+0.236
ints+0.214
Neg logit contributions
*)-0.168
allah-0.168
 coincide-0.167
 flee-0.161
 closer-0.159
Token copying
Token ID23345
Feature activation-1.10e-01

Pos logit contributions
ozo+3.169
zos+3.168
yx+3.063
sonian+3.038
ints+2.713
Neg logit contributions
*)-2.953
allah-2.935
 coincide-2.913
 flee-2.839
 murd-2.810
Token,
Token ID11
Feature activation-9.96e-01

Pos logit contributions
zos+0.567
ozo+0.566
sonian+0.545
yx+0.543
ints+0.493
Neg logit contributions
*)-0.393
allah-0.392
 coincide-0.388
 flee-0.377
 closer-0.372
Token or
Token ID393
Feature activation-8.36e-01

Pos logit contributions
zos+4.978
ozo+4.969
yx+4.814
sonian+4.789
ints+4.338
Neg logit contributions
*)-3.586
allah-3.568
 coincide-3.543
 flee-3.439
 murd-3.397
Token redistribution
Token ID41425
Feature activation-3.89e-01

Pos logit contributions
zos+4.143
ozo+4.122
yx+3.983
sonian+3.964
ints+3.590
Neg logit contributions
*)-3.101
allah-3.084
 coincide-3.057
 flee-2.973
 murd-2.936
Token (
Token ID357
Feature activation-3.62e-01

Pos logit contributions
zos+1.960
ozo+1.954
sonian+1.880
yx+1.874
ints+1.698
Neg logit contributions
*)-1.415
allah-1.410
 coincide-1.397
 flee-1.359
 bombard-1.341
Tokenelect
Token ID9509
Feature activation+2.44e-01

Pos logit contributions
zos+1.872
ozo+1.866
sonian+1.798
yx+1.794
ints+1.627
Neg logit contributions
*)-1.271
allah-1.268
 coincide-1.259
 flee-1.223
 bombard-1.203
Tokenronic
Token ID4565
Feature activation+1.52e-01

Pos logit contributions
*)+1.144
allah+1.127
 coincide+1.106
 flee+1.093
 closer+1.092
Neg logit contributions
zos-0.992
ozo-0.976
sonian-0.941
yx-0.923
ints-0.820
Token or
Token ID393
Feature activation-7.01e-01

Pos logit contributions
*)+0.502
allah+0.490
 coincide+0.488
 flee+0.475
 bombard+0.470
Neg logit contributions
zos-0.829
ozo-0.826
sonian-0.792
yx-0.789
ints-0.725
Token 2012
Token ID2321
Feature activation-8.64e-03

Pos logit contributions
zos+0.966
ozo+0.963
yx+0.928
sonian+0.916
ints+0.823
Neg logit contributions
allah-0.975
*)-0.972
 coincide-0.971
 flee-0.948
 bombard-0.940
Token Canada
Token ID3340
Feature activation-1.74e-01

Pos logit contributions
zos+0.035
ozo+0.035
sonian+0.034
yx+0.033
ints+0.029
Neg logit contributions
*)-0.040
allah-0.040
 coincide-0.040
 flee-0.039
 closer-0.038
Token 16
Token ID1467
Feature activation+6.11e-02

Pos logit contributions
ozo+0.807
zos+0.807
yx+0.784
sonian+0.766
ints+0.694
Neg logit contributions
allah-0.697
*)-0.692
 coincide-0.692
 flee-0.676
 bombard-0.669
Token10
Token ID940
Feature activation+7.95e-02

Pos logit contributions
*)+0.216
allah+0.215
 coincide+0.212
 flee+0.205
 closer+0.203
Neg logit contributions
zos-0.318
ozo-0.316
sonian-0.305
yx-0.302
ints-0.276
Token Posts
Token ID12043
Feature activation-8.92e-01

Pos logit contributions
allah+0.400
 coincide+0.400
*)+0.396
 flee+0.394
 bombard+0.390
Neg logit contributions
zos-0.288
ozo-0.288
yx-0.278
sonian-0.272
ints-0.236
Token #
Token ID1303
Feature activation-1.21e+00

Pos logit contributions
zos+3.861
ozo+3.835
sonian+3.672
yx+3.647
ints+3.242
Neg logit contributions
*)-3.921
allah-3.916
 coincide-3.887
 flee-3.796
 bombard-3.749
Token14
Token ID1415
Feature activation-9.22e-01

Pos logit contributions
zos+5.212
yx+5.177
ozo+5.137
sonian+4.783
ints+4.538
Neg logit contributions
 murd-5.147
 neighb-5.093
 bombard-4.969
 coincide-4.937
 flee-4.914
Token On
Token ID1550
Feature activation-6.52e-01

Pos logit contributions
zos+3.930
ozo+3.908
yx+3.763
sonian+3.732
 Correction+3.315
Neg logit contributions
*)-3.940
allah-3.939
 coincide-3.862
 flee-3.775
 bombard-3.770
Token September
Token ID2693
Feature activation-4.90e-01

Pos logit contributions
ozo+2.559
yx+2.535
zos+2.486
sonian+2.340
 Transformers+2.140
Neg logit contributions
*)-3.041
allah-3.011
 bombard-2.944
 murd-2.936
 coincide-2.920
Token 20
Token ID1160
Feature activation-7.20e-01

Pos logit contributions
zos+2.567
ozo+2.539
sonian+2.441
yx+2.425
ints+2.207
Neg logit contributions
*)-1.753
allah-1.747
 coincide-1.719
 flee-1.691
 closer-1.661
Token 2013
Token ID2211
Feature activation-5.90e-01

Pos logit contributions
zos+2.869
yx+2.846
ozo+2.800
sonian+2.679
ints+2.387
Neg logit contributions
 murd-3.310
 bombard-3.285
allah-3.241
 neighb-3.169
*)-3.134
Token the
Token ID262
Feature activation-3.86e-01

Pos logit contributions
zos+1.482
ozo+1.475
sonian+1.416
yx+1.406
ints+1.265
Neg logit contributions
*)-1.293
allah-1.291
 coincide-1.278
 flee-1.249
 bombard-1.233
Token international
Token ID3230
Feature activation-1.97e-01

Pos logit contributions
zos+1.756
ozo+1.748
sonian+1.677
yx+1.674
ints+1.493
Neg logit contributions
*)-1.597
allah-1.593
 coincide-1.579
 flee-1.537
 bombard-1.516
Token cinema
Token ID22041
Feature activation-5.35e-01

Pos logit contributions
zos+0.875
ozo+0.871
sonian+0.835
yx+0.830
ints+0.742
Neg logit contributions
*)-0.840
allah-0.838
 coincide-0.831
 flee-0.812
 bombard-0.802
Token stage
Token ID3800
Feature activation-3.06e-01

Pos logit contributions
zos+2.378
ozo+2.374
yx+2.281
sonian+2.267
ints+2.006
Neg logit contributions
*)-2.260
allah-2.259
 coincide-2.223
 flee-2.164
 murd-2.142
Token is
Token ID318
Feature activation-4.86e-01

Pos logit contributions
zos+1.426
ozo+1.418
sonian+1.367
yx+1.359
ints+1.219
Neg logit contributions
*)-1.230
allah-1.225
 coincide-1.213
 flee-1.186
 bombard-1.168
Token simple
Token ID2829
Feature activation-9.73e-01

Pos logit contributions
zos+2.267
ozo+2.258
sonian+2.160
yx+2.150
ints+1.940
Neg logit contributions
allah-1.982
*)-1.978
 coincide-1.969
 flee-1.921
 closer-1.909
Token:
Token ID25
Feature activation-2.76e-01

Pos logit contributions
zos+4.579
ozo+4.564
sonian+4.411
yx+4.403
ints+3.921
Neg logit contributions
*)-3.745
allah-3.742
 coincide-3.699
 flee-3.587
 murd-3.546
Token look
Token ID804
Feature activation-4.29e-01

Pos logit contributions
zos+1.325
ozo+1.321
yx+1.271
sonian+1.266
ints+1.139
Neg logit contributions
*)-1.069
allah-1.066
 coincide-1.046
 flee-1.016
 murd-1.002
Token pretty
Token ID2495
Feature activation-1.90e-01

Pos logit contributions
zos+2.120
ozo+2.113
sonian+2.035
yx+2.022
ints+1.833
Neg logit contributions
*)-1.600
allah-1.596
 coincide-1.578
 flee-1.534
 bombard-1.514
Token and
Token ID290
Feature activation-2.07e-01

Pos logit contributions
zos+0.948
ozo+0.944
sonian+0.909
yx+0.906
ints+0.819
Neg logit contributions
*)-0.709
allah-0.707
 coincide-0.700
 flee-0.681
 bombard-0.671
Token play
Token ID711
Feature activation-2.14e-01

Pos logit contributions
zos+1.030
ozo+1.023
sonian+0.984
yx+0.984
ints+0.888
Neg logit contributions
*)-0.772
allah-0.770
 coincide-0.759
 flee-0.737
 bombard-0.726
Token of
Token ID286
Feature activation+1.25e-01

Pos logit contributions
*)+1.316
 coincide+1.314
allah+1.295
 closer+1.276
 flee+1.257
Neg logit contributions
zos-1.906
ozo-1.887
sonian-1.835
yx-1.806
ints-1.642
Token the
Token ID262
Feature activation+1.11e-01

Pos logit contributions
*)+0.543
 coincide+0.541
allah+0.541
 flee+0.528
 closer+0.522
Neg logit contributions
zos-0.543
ozo-0.536
sonian-0.520
yx-0.506
ints-0.455
Token game
Token ID983
Feature activation+2.52e-01

Pos logit contributions
*)+0.473
 coincide+0.471
allah+0.469
 flee+0.459
 closer+0.454
Neg logit contributions
zos-0.497
ozo-0.493
sonian-0.475
yx-0.466
ints-0.419
Token.
Token ID13
Feature activation-2.49e-01

Pos logit contributions
*)+0.906
 coincide+0.901
allah+0.896
 closer+0.871
 flee+0.868
Neg logit contributions
zos-1.284
ozo-1.279
sonian-1.237
yx-1.223
ints-1.112
Token<|endoftext|>
Token ID50256
Feature activation+2.06e-01

Pos logit contributions
zos+1.193
ozo+1.188
sonian+1.142
yx+1.136
ints+1.024
Neg logit contributions
*)-0.976
allah-0.971
 coincide-0.963
 flee-0.938
 bombard-0.923
TokenShare
Token ID11649
Feature activation-8.66e-01

Pos logit contributions
*)+0.805
allah+0.790
 coincide+0.765
 flee+0.744
 closer+0.740
Neg logit contributions
zos-1.000
ozo-0.995
sonian-0.956
yx-0.944
ints-0.853
Token.
Token ID13
Feature activation+2.93e-01

Pos logit contributions
zos+4.098
ozo+4.085
yx+3.926
sonian+3.914
ints+3.523
Neg logit contributions
*)-3.368
 coincide-3.351
allah-3.351
 flee-3.241
 bombard-3.191
Token Two
Token ID4930
Feature activation+8.52e-02

Pos logit contributions
*)+1.261
allah+1.252
 coincide+1.244
 flee+1.212
 closer+1.207
Neg logit contributions
zos-1.293
ozo-1.281
sonian-1.235
yx-1.218
ints-1.093
Token iconic
Token ID14133
Feature activation+2.10e-01

Pos logit contributions
*)+0.333
 coincide+0.333
allah+0.331
 flee+0.324
 closer+0.322
Neg logit contributions
zos-0.408
ozo-0.405
sonian-0.393
yx-0.387
ints-0.349
Token horror
Token ID9961
Feature activation-1.36e-01

Pos logit contributions
*)+0.903
 coincide+0.898
allah+0.897
 flee+0.884
 closer+0.876
Neg logit contributions
zos-0.919
ozo-0.912
sonian-0.889
yx-0.864
ints-0.775
Token titles
Token ID8714
Feature activation+1.64e-01

Pos logit contributions
zos+0.533
ozo+0.529
sonian+0.506
yx+0.501
ints+0.441
Neg logit contributions
*)-0.653
allah-0.651
 coincide-0.647
 flee-0.634
 bombard-0.629
Tokenmail
Token ID4529
Feature activation-5.91e-01

Pos logit contributions
ozo+0.819
zos+0.816
sonian+0.794
yx+0.777
ints+0.724
Neg logit contributions
*)-0.289
allah-0.276
 coincide-0.266
 bombard-0.260
 flee-0.256
Token this
Token ID428
Feature activation-2.98e-01

Pos logit contributions
zos+2.921
ozo+2.904
yx+2.797
sonian+2.777
ints+2.519
Neg logit contributions
*)-2.208
allah-2.187
 coincide-2.185
 flee-2.110
 murd-2.093
Token to
Token ID284
Feature activation-1.16e-01

Pos logit contributions
zos+1.237
ozo+1.234
sonian+1.186
yx+1.176
ints+1.041
Neg logit contributions
*)-1.349
allah-1.347
 coincide-1.335
 flee-1.304
 bombard-1.293
Token a
Token ID257
Feature activation+2.50e-02

Pos logit contributions
zos+0.507
ozo+0.506
sonian+0.484
yx+0.482
ints+0.430
Neg logit contributions
*)-0.504
allah-0.502
 coincide-0.497
 flee-0.486
 bombard-0.480
Token friend
Token ID1545
Feature activation-2.88e-01

Pos logit contributions
allah+0.065
*)+0.065
 coincide+0.063
 flee+0.061
 bombard+0.060
Neg logit contributions
ozo-0.152
zos-0.152
sonian-0.147
yx-0.147
ints-0.136
Token Print
Token ID12578
Feature activation-1.36e+00

Pos logit contributions
ozo+1.352
zos+1.348
yx+1.312
sonian+1.308
ints+1.183
Neg logit contributions
allah-1.136
 coincide-1.126
*)-1.123
 flee-1.075
 murd-1.069
Tokenable
Token ID540
Feature activation-6.30e-01

Pos logit contributions
zos+6.229
ozo+6.202
yx+6.024
sonian+5.827
 Correction+5.449
Neg logit contributions
 coincide-5.273
*)-5.203
 murd-5.088
allah-5.066
 bombard-5.061
Token version
Token ID2196
Feature activation-4.76e-01

Pos logit contributions
zos+2.269
ozo+2.237
yx+2.135
sonian+2.107
ints+1.847
Neg logit contributions
*)-3.191
allah-3.157
 coincide-3.156
 bombard-3.056
 murd-3.055
Token<|endoftext|>
Token ID50256
Feature activation+1.67e-01

Pos logit contributions
ozo+2.195
zos+2.186
yx+2.128
sonian+2.114
 Correction+1.923
Neg logit contributions
allah-1.891
*)-1.885
 coincide-1.853
 murd-1.836
 flee-1.803
TokenA
Token ID32
Feature activation+1.30e-01

Pos logit contributions
*)+0.655
allah+0.643
 coincide+0.623
 flee+0.606
 closer+0.602
Neg logit contributions
zos-0.813
ozo-0.808
sonian-0.777
yx-0.768
ints-0.693
Token university
Token ID6403
Feature activation+2.38e-01

Pos logit contributions
*)+0.512
allah+0.512
 coincide+0.505
 flee+0.497
 closer+0.492
Neg logit contributions
ozo-0.620
zos-0.619
sonian-0.589
yx-0.585
ints-0.528