TOP ACTIVATIONS
MAX = 4.353

Dolphins knew how to exploit mism atches in the condensed area
Commander ," with its two mism atched heroes , Jack Aub
seem way too early to handic ap the 2017 Oscars ,
accepted an offer from L ipp inc ott to write a
DE AD LIN ES LO OM Ra ising the
non - profit group , comm utes from her home in
H are wood is officially listed as an offensive tackle ,
H t / W t Comp . Rating 247 ESPN Rivals
factions at this gathering , i . e . an Old
Rec ru ited with pick No . 43
. Thomas , 39 , appeared in 53 games
9 d 970 b L isted below are links to we
The situation may rec ede or deterior ate and
rick Garland s nomination to the Supreme Court .
Painter also said Ross must rec use himself from all Russia
! <|endoftext|> Marc Dillon at M WC 2014 with Mak ia
Patri k Elias , Claude Lem ie ux , Scott Stevens
with particles , avoiding that numerical diffusion ; this particle -
in Sir sa , while comm ending the work
November 2012 , he was listed number one in the national

MIN ACTIVATIONS
MIN = -4.338

s army . Bang ura , who took up
) 358 af d 4 bd 02 de 3
Follow @ Jessica Dur ando on Twitter
ya Sabha member from Gujarat Shank ar Bh ai Veg ad
ial Name (' system . windows . forms ') | out
fair play . Ven ky s plans
. 104 - 106 ( Atlanta : Scholars Press , 1985
the Australian Capital Territory ( ACT ) and the Northern Territory
pattern of ownership of consumer dur ables among Indian households .
with the Winnipeg Jets / Atlanta Th ras hers franchise ,
W IS Q ARS ). Atlanta , GA : US Department
Africa after all . " Dar ren must be in Fab
, 1991 2009 . Atlanta , GA : US Department
what 's called a " d owntown allowance ."
O Korea as d Gu M ih
done by the University of Victoria s Environmental Law
nd ed . 2010 . Atlanta , GA : US Department
Disease Control and Prevention in Atlanta , Georgia , to halt
isolation from the architecture - pur ging mainstream of the world
major competitors in the Data Visual isation market , a position

INTERVAL 2.181 - 4.353
CONTAINS 0.351%

only Ravens All - Pro Marshal Y anda per Pro
confirmed . FR OM THE AR CH IVE :
are not arrested randomly in North Korea . If they are
phan Black BBC America Matthew Mc Con aug hey True Detective
factions at this gathering , i . e . an Old

INTERVAL 0.008 - 2.181
CONTAINS 48.319%

ura said . The position of the
of attack . Now notice the variety of ways
/ her self as someone he / she is not through
ous Vac ations - Who To Vote For - Jimmy Tide
to eliminate the other and with the screen door effect you

INTERVAL -2.165 - 0.008
CONTAINS 50.964%

destruct ible - Vehicles show four stages of damage , so
worth the Democratic votes . Georgia Rep . St acey Evans
of The ology of Em ory University , in Atlanta .
In my case , it s for a R
Latin as on Facebook ! Do you think that

INTERVAL -4.338 - -2.165
CONTAINS 0.365%

My sample of Got lands d rick a And
call for his " Travel Ban " on Twitter as an
the Sund arb ans in Bangladesh , told reporters .
weather , and ever - d wind ling resources .
bin / sh / s bin / iw config w l
Token Dolphins
Token ID21668
Feature activation+3.02e-01

Pos logit contributions
door+0.053
+0.052
▬▬+0.049
 pulp+0.047
taboola+0.046
Neg logit contributions
atched-0.063
apped-0.062
onduct-0.055
ursed-0.051
umbered-0.050
Token knew
Token ID2993
Feature activation-1.99e-01

Pos logit contributions
atched+0.746
apped+0.739
onduct+0.646
ursed+0.599
uked+0.596
Neg logit contributions
door-0.650
-0.639
▬▬-0.602
 pulp-0.577
taboola-0.566
Token how
Token ID703
Feature activation-9.06e-01

Pos logit contributions
door+0.402
+0.391
▬▬+0.366
 pulp+0.356
taboola+0.344
Neg logit contributions
atched-0.522
apped-0.514
onduct-0.454
umbered-0.423
uked-0.423
Token to
Token ID284
Feature activation-1.13e+00

Pos logit contributions
door+1.686
+1.654
▬▬+1.545
 pulp+1.471
taboola+1.436
Neg logit contributions
atched-2.498
apped-2.471
onduct-2.192
ursed-2.055
umbered-2.048
Token exploit
Token ID14561
Feature activation+1.30e+00

Pos logit contributions
door+1.979
+1.942
 pulp+1.787
▬▬+1.774
ners+1.690
Neg logit contributions
atched-3.209
apped-3.160
onduct-2.834
uked-2.661
ursed-2.659
Token mism
Token ID32691
Feature activation+4.35e+00

Pos logit contributions
atched+3.209
apped+3.196
onduct+2.780
ursed+2.581
umbered+2.570
Neg logit contributions
door-2.759
-2.709
▬▬-2.575
 pulp-2.439
taboola-2.420
Tokenatches
Token ID20981
Feature activation+1.39e+00

Pos logit contributions
atched+3.357
apped+2.947
onduct+1.304
ursed+0.772
uked+0.749
Neg logit contributions
-16.974
door-16.835
▬▬-16.667
taboola-16.004
 pulp-15.891
Token in
Token ID287
Feature activation+2.60e-01

Pos logit contributions
atched+3.435
apped+3.421
onduct+2.983
ursed+2.766
umbered+2.761
Neg logit contributions
door-2.937
-2.892
▬▬-2.734
 pulp-2.606
taboola-2.575
Token the
Token ID262
Feature activation+5.16e-01

Pos logit contributions
atched+0.680
apped+0.675
onduct+0.593
ursed+0.553
umbered+0.552
Neg logit contributions
door-0.521
-0.511
▬▬-0.481
 pulp-0.458
taboola-0.451
Token condensed
Token ID38784
Feature activation+1.00e+00

Pos logit contributions
atched+1.421
apped+1.412
onduct+1.249
ursed+1.170
umbered+1.167
Neg logit contributions
door-0.961
-0.941
▬▬-0.881
 pulp-0.836
taboola-0.821
Token area
Token ID1989
Feature activation-7.48e-01

Pos logit contributions
atched+2.923
apped+2.906
onduct+2.591
ursed+2.436
umbered+2.429
Neg logit contributions
door-1.696
-1.657
▬▬-1.542
 pulp-1.450
taboola-1.430
Token Commander
Token ID13353
Feature activation-1.41e+00

Pos logit contributions
atched+0.381
apped+0.377
onduct+0.332
ursed+0.310
umbered+0.309
Neg logit contributions
door-0.289
-0.284
▬▬-0.266
 pulp-0.256
taboola-0.249
Token,"
Token ID553
Feature activation+7.71e-01

Pos logit contributions
door+2.497
+2.422
▬▬+2.252
 pulp+2.188
ners+2.100
Neg logit contributions
atched-3.921
apped-3.889
onduct-3.460
ython-3.271
umbered-3.239
Token with
Token ID351
Feature activation+1.12e+00

Pos logit contributions
atched+1.968
apped+1.953
onduct+1.711
ursed+1.591
umbered+1.585
Neg logit contributions
door-1.592
-1.561
▬▬-1.470
 pulp-1.412
taboola-1.382
Token its
Token ID663
Feature activation+1.51e-01

Pos logit contributions
atched+2.954
apped+2.934
onduct+2.583
ursed+2.418
umbered+2.397
Neg logit contributions
door-2.185
-2.138
▬▬-2.019
taboola-1.886
 pulp-1.858
Token two
Token ID734
Feature activation-7.70e-02

Pos logit contributions
atched+0.411
apped+0.409
onduct+0.361
ursed+0.338
umbered+0.337
Neg logit contributions
door-0.285
-0.279
▬▬-0.261
 pulp-0.248
taboola-0.244
Token mism
Token ID32691
Feature activation+3.87e+00

Pos logit contributions
door+0.150
+0.147
▬▬+0.138
 pulp+0.132
taboola+0.129
Neg logit contributions
atched-0.206
apped-0.204
onduct-0.180
ursed-0.168
umbered-0.168
Tokenatched
Token ID14265
Feature activation+1.21e+00

Pos logit contributions
atched+4.903
apped+4.389
onduct+3.006
ixed+2.559
uked+2.551
Neg logit contributions
-13.323
door-13.180
▬▬-13.012
taboola-12.398
 pulp-12.253
Token heroes
Token ID10281
Feature activation+1.09e+00

Pos logit contributions
atched+3.122
apped+3.101
onduct+2.711
ursed+2.540
umbered+2.533
Neg logit contributions
door-2.422
-2.384
▬▬-2.247
taboola-2.112
 pulp-2.104
Token,
Token ID11
Feature activation+7.96e-02

Pos logit contributions
atched+2.740
apped+2.725
onduct+2.379
umbered+2.215
uked+2.201
Neg logit contributions
door-2.279
-2.238
▬▬-2.127
taboola-2.002
 pulp-1.996
Token Jack
Token ID3619
Feature activation-1.41e-01

Pos logit contributions
atched+0.191
apped+0.189
onduct+0.164
ursed+0.152
umbered+0.151
Neg logit contributions
door-0.177
-0.174
▬▬-0.164
 pulp-0.159
taboola-0.155
Token Aub
Token ID19723
Feature activation-6.31e-01

Pos logit contributions
door+0.284
+0.278
▬▬+0.260
 pulp+0.250
taboola+0.243
Neg logit contributions
atched-0.369
apped-0.366
onduct-0.322
ursed-0.300
umbered-0.299
Token seem
Token ID1283
Feature activation-7.16e-01

Pos logit contributions
atched+0.557
apped+0.553
onduct+0.483
ursed+0.446
uked+0.446
Neg logit contributions
door-0.477
-0.469
▬▬-0.441
 pulp-0.426
taboola-0.417
Token way
Token ID835
Feature activation-1.95e-01

Pos logit contributions
door+1.232
+1.201
▬▬+1.122
 pulp+1.073
ners+1.028
Neg logit contributions
atched-2.071
apped-2.060
onduct-1.840
ursed-1.724
uked-1.714
Token too
Token ID1165
Feature activation-1.66e-01

Pos logit contributions
door+0.356
+0.348
▬▬+0.324
 pulp+0.309
taboola+0.302
Neg logit contributions
atched-0.544
apped-0.540
onduct-0.480
ursed-0.449
umbered-0.447
Token early
Token ID1903
Feature activation+1.49e+00

Pos logit contributions
door+0.283
+0.277
▬▬+0.257
 pulp+0.243
taboola+0.238
Neg logit contributions
atched-0.481
apped-0.478
onduct-0.426
ursed-0.400
umbered-0.400
Token to
Token ID284
Feature activation-1.34e-01

Pos logit contributions
atched+3.732
apped+3.714
onduct+3.240
umbered+3.009
ursed+2.973
Neg logit contributions
door-3.084
-3.021
▬▬-2.883
 pulp-2.729
taboola-2.693
Token handic
Token ID29103
Feature activation+3.68e+00

Pos logit contributions
door+0.272
+0.266
▬▬+0.249
 pulp+0.244
taboola+0.236
Neg logit contributions
atched-0.346
apped-0.343
onduct-0.303
ursed-0.282
uked-0.281
Tokenap
Token ID499
Feature activation+1.28e+00

Pos logit contributions
apped+8.723
atched+8.310
onduct+7.127
umbered+6.478
uked+6.476
Neg logit contributions
-8.388
door-8.302
▬▬-8.176
taboola-7.494
 pulp-7.417
Token the
Token ID262
Feature activation+3.27e-01

Pos logit contributions
atched+3.236
apped+3.211
onduct+2.802
ursed+2.605
umbered+2.592
Neg logit contributions
door-2.673
-2.612
▬▬-2.470
 pulp-2.351
taboola-2.314
Token 2017
Token ID2177
Feature activation+3.79e-01

Pos logit contributions
atched+0.866
apped+0.861
onduct+0.759
ursed+0.708
umbered+0.706
Neg logit contributions
door-0.643
-0.631
▬▬-0.591
 pulp-0.566
taboola-0.554
Token Oscars
Token ID46485
Feature activation+1.14e+00

Pos logit contributions
atched+0.943
apped+0.936
onduct+0.817
ursed+0.759
umbered+0.756
Neg logit contributions
door-0.805
-0.791
▬▬-0.746
 pulp-0.715
taboola-0.703
Token,
Token ID11
Feature activation-9.27e-03

Pos logit contributions
atched+2.863
apped+2.842
onduct+2.485
umbered+2.307
ursed+2.299
Neg logit contributions
door-2.369
-2.326
▬▬-2.210
 pulp-2.096
taboola-2.073
Token accepted
Token ID6292
Feature activation+6.35e-01

Pos logit contributions
door+2.110
+2.097
 pulp+1.965
▬▬+1.931
ners+1.873
Neg logit contributions
atched-2.291
apped-2.233
onduct-1.969
ython-1.812
uked-1.805
Token an
Token ID281
Feature activation+4.87e-01

Pos logit contributions
atched+1.632
apped+1.622
onduct+1.422
ursed+1.323
umbered+1.320
Neg logit contributions
door-1.298
-1.273
▬▬-1.200
 pulp-1.141
taboola-1.126
Token offer
Token ID2897
Feature activation+9.46e-01

Pos logit contributions
atched+1.228
apped+1.219
onduct+1.065
ursed+0.990
umbered+0.988
Neg logit contributions
door-1.021
-1.001
▬▬-0.943
 pulp-0.905
taboola-0.888
Token from
Token ID422
Feature activation+1.84e-01

Pos logit contributions
atched+2.359
apped+2.343
onduct+2.044
ursed+1.900
umbered+1.894
Neg logit contributions
door-1.999
-1.967
▬▬-1.857
 pulp-1.770
taboola-1.746
Token L
Token ID406
Feature activation+1.17e+00

Pos logit contributions
atched+0.479
apped+0.475
onduct+0.418
ursed+0.389
umbered+0.388
Neg logit contributions
door-0.370
-0.363
▬▬-0.341
 pulp-0.329
taboola-0.320
Tokenipp
Token ID3974
Feature activation+3.66e+00

Pos logit contributions
atched+2.843
apped+2.830
onduct+2.457
umbered+2.277
ursed+2.276
Neg logit contributions
door-2.540
-2.495
▬▬-2.371
 pulp-2.254
taboola-2.232
Tokeninc
Token ID1939
Feature activation+2.19e+00

Pos logit contributions
apped+8.572
atched+8.485
onduct+7.524
ursed+6.900
ixed+6.880
Neg logit contributions
door-7.671
-7.652
▬▬-7.531
taboola-6.906
yrinth-6.675
Tokenott
Token ID1252
Feature activation-1.18e-01

Pos logit contributions
apped+3.994
atched+3.885
onduct+3.228
ursed+2.924
uked+2.886
Neg logit contributions
door-6.106
-6.056
▬▬-5.925
taboola-5.620
 pulp-5.534
Token to
Token ID284
Feature activation+2.24e-01

Pos logit contributions
door+0.242
+0.238
▬▬+0.222
 pulp+0.215
taboola+0.209
Neg logit contributions
atched-0.304
apped-0.301
onduct-0.263
ursed-0.246
uked-0.245
Token write
Token ID3551
Feature activation+7.26e-01

Pos logit contributions
atched+0.630
apped+0.625
onduct+0.554
uked+0.521
ursed+0.519
Neg logit contributions
door-0.401
-0.392
▬▬-0.361
 pulp-0.360
taboola-0.341
Token a
Token ID257
Feature activation+1.32e-01

Pos logit contributions
atched+1.858
apped+1.846
onduct+1.617
ursed+1.505
umbered+1.501
Neg logit contributions
door-1.491
-1.463
▬▬-1.379
 pulp-1.315
taboola-1.294
TokenDE
Token ID7206
Feature activation-1.04e+00

Pos logit contributions
door+0.367
+0.360
▬▬+0.339
 pulp+0.321
taboola+0.317
Neg logit contributions
atched-0.480
apped-0.476
onduct-0.419
ursed-0.391
umbered-0.389
TokenAD
Token ID2885
Feature activation-9.77e-01

Pos logit contributions
+2.263
door+2.251
▬▬+2.129
taboola+1.998
 pulp+1.981
Neg logit contributions
atched-2.504
apped-2.494
onduct-2.152
ursed-2.030
uked-1.989
TokenLIN
Token ID34509
Feature activation+7.14e-01

Pos logit contributions
door+2.197
+2.135
▬▬+2.039
 pulp+1.950
taboola+1.936
Neg logit contributions
atched-2.322
apped-2.300
onduct-1.989
ursed-1.857
umbered-1.849
TokenES
Token ID1546
Feature activation+1.88e+00

Pos logit contributions
atched+1.652
apped+1.637
onduct+1.416
ursed+1.302
umbered+1.298
Neg logit contributions
door-1.648
-1.622
▬▬-1.541
 pulp-1.475
taboola-1.452
Token LO
Token ID17579
Feature activation+1.82e+00

Pos logit contributions
atched+4.462
apped+4.440
onduct+3.853
umbered+3.558
ursed+3.551
Neg logit contributions
door-4.116
-4.042
▬▬-3.803
 pulp-3.677
taboola-3.586
TokenOM
Token ID2662
Feature activation+3.64e+00

Pos logit contributions
atched+4.106
apped+4.095
onduct+3.489
umbered+3.219
ursed+3.214
Neg logit contributions
door-4.291
-4.198
▬▬-3.988
 pulp-3.844
taboola-3.792
Token
Token ID198
Feature activation+2.97e-01

Pos logit contributions
apped+7.959
atched+7.958
onduct+6.734
ursed+6.233
umbered+6.166
Neg logit contributions
door-8.055
-7.991
▬▬-7.517
 pulp-7.353
taboola-7.220
Token
Token ID198
Feature activation+2.07e-01

Pos logit contributions
atched+0.821
apped+0.812
onduct+0.720
uked+0.678
umbered+0.678
Neg logit contributions
door-0.565
-0.540
▬▬-0.503
 pulp-0.488
taboola-0.478
TokenRa
Token ID21762
Feature activation+1.32e-01

Pos logit contributions
atched+0.542
apped+0.539
onduct+0.474
ursed+0.442
umbered+0.441
Neg logit contributions
door-0.412
-0.404
▬▬-0.380
 pulp-0.362
taboola-0.356
Tokenising
Token ID1710
Feature activation+1.19e+00

Pos logit contributions
atched+0.270
apped+0.267
onduct+0.227
ursed+0.206
umbered+0.205
Neg logit contributions
door-0.337
-0.332
▬▬-0.316
 pulp-0.305
taboola-0.301
Token the
Token ID262
Feature activation+5.58e-01

Pos logit contributions
atched+2.999
apped+2.987
onduct+2.592
umbered+2.418
ursed+2.413
Neg logit contributions
door-2.485
-2.418
▬▬-2.293
 pulp-2.171
taboola-2.136
Token non
Token ID1729
Feature activation-1.86e-01

Pos logit contributions
door+0.014
+0.014
▬▬+0.013
 pulp+0.013
taboola+0.012
Neg logit contributions
atched-0.022
apped-0.022
onduct-0.020
ursed-0.018
umbered-0.018
Token-
Token ID12
Feature activation+6.63e-01

Pos logit contributions
door+0.370
+0.365
▬▬+0.342
 pulp+0.322
taboola+0.317
Neg logit contributions
atched-0.490
apped-0.489
onduct-0.427
uked-0.397
umbered-0.397
Tokenprofit
Token ID9183
Feature activation+3.74e-01

Pos logit contributions
atched+1.846
apped+1.842
onduct+1.625
uked+1.520
umbered+1.508
Neg logit contributions
door-1.190
-1.183
▬▬-1.111
 pulp-1.047
taboola-1.020
Token group
Token ID1448
Feature activation-2.48e-01

Pos logit contributions
atched+1.015
apped+1.008
onduct+0.892
ursed+0.834
umbered+0.832
Neg logit contributions
door-0.713
-0.699
▬▬-0.653
 pulp-0.625
taboola-0.609
Token,
Token ID11
Feature activation-4.42e-01

Pos logit contributions
door+0.484
+0.475
▬▬+0.443
 pulp+0.427
taboola+0.414
Neg logit contributions
atched-0.660
apped-0.654
onduct-0.578
ursed-0.541
umbered-0.538
Token comm
Token ID725
Feature activation+3.63e+00

Pos logit contributions
door+0.999
+0.982
▬▬+0.923
 pulp+0.901
taboola+0.872
Neg logit contributions
atched-1.033
apped-1.024
onduct-0.886
uked-0.819
ursed-0.819
Tokenutes
Token ID1769
Feature activation+1.26e+00

Pos logit contributions
atched+7.096
apped+7.049
onduct+5.830
uked+5.429
ixed+5.349
Neg logit contributions
door-9.410
-9.225
▬▬-9.013
 pulp-8.521
taboola-8.383
Token from
Token ID422
Feature activation+3.15e-01

Pos logit contributions
atched+3.138
apped+3.120
onduct+2.722
ursed+2.537
umbered+2.518
Neg logit contributions
door-2.659
-2.630
▬▬-2.482
 pulp-2.356
taboola-2.333
Token her
Token ID607
Feature activation-2.14e-01

Pos logit contributions
atched+0.826
apped+0.820
onduct+0.720
ursed+0.672
umbered+0.670
Neg logit contributions
door-0.627
-0.616
▬▬-0.579
 pulp-0.550
taboola-0.543
Token home
Token ID1363
Feature activation+1.30e-01

Pos logit contributions
door+0.428
+0.418
▬▬+0.393
 pulp+0.378
taboola+0.368
Neg logit contributions
atched-0.559
apped-0.555
onduct-0.489
ursed-0.456
umbered-0.455
Token in
Token ID287
Feature activation-3.53e-01

Pos logit contributions
atched+0.314
apped+0.311
onduct+0.270
ursed+0.250
uked+0.249
Neg logit contributions
door-0.284
-0.280
▬▬-0.265
 pulp-0.252
taboola-0.250
TokenH
Token ID39
Feature activation-9.71e-02

Pos logit contributions
atched+0.410
apped+0.407
onduct+0.357
ursed+0.333
umbered+0.332
Neg logit contributions
door-0.316
-0.310
▬▬-0.291
 pulp-0.277
taboola-0.273
Tokenare
Token ID533
Feature activation-4.34e-01

Pos logit contributions
door+0.158
+0.147
▬▬+0.135
 pulp+0.131
ners+0.127
Neg logit contributions
apped-0.288
atched-0.285
onduct-0.256
umbered-0.242
ython-0.242
Tokenwood
Token ID3822
Feature activation+1.44e-01

Pos logit contributions
door+0.835
+0.778
 pulp+0.719
▬▬+0.709
ners+0.698
Neg logit contributions
atched-1.174
apped-1.164
onduct-1.022
uked-0.978
umbered-0.969
Token is
Token ID318
Feature activation+7.22e-01

Pos logit contributions
atched+0.343
apped+0.339
onduct+0.295
ursed+0.273
ython+0.273
Neg logit contributions
door-0.321
-0.317
▬▬-0.296
 pulp-0.288
taboola-0.280
Token officially
Token ID8720
Feature activation+1.50e+00

Pos logit contributions
atched+1.822
apped+1.807
onduct+1.581
ursed+1.469
umbered+1.465
Neg logit contributions
door-1.515
-1.487
▬▬-1.402
 pulp-1.342
taboola-1.316
Token listed
Token ID5610
Feature activation+3.62e+00

Pos logit contributions
atched+3.672
apped+3.666
onduct+3.179
ursed+2.952
umbered+2.952
Neg logit contributions
door-3.204
-3.140
▬▬-2.973
 pulp-2.834
taboola-2.815
Token as
Token ID355
Feature activation-4.97e-01

Pos logit contributions
atched+6.720
apped+6.698
onduct+5.525
umbered+5.179
ursed+5.158
Neg logit contributions
door-9.251
-9.210
▬▬-8.853
taboola-8.436
 pulp-8.405
Token an
Token ID281
Feature activation+4.79e-01

Pos logit contributions
door+0.955
+0.936
▬▬+0.878
 pulp+0.840
taboola+0.812
Neg logit contributions
atched-1.343
apped-1.326
onduct-1.178
ython-1.102
ursed-1.094
Token offensive
Token ID5859
Feature activation+2.18e-01

Pos logit contributions
atched+1.271
apped+1.261
onduct+1.111
ursed+1.036
umbered+1.034
Neg logit contributions
door-0.941
-0.923
▬▬-0.866
 pulp-0.826
taboola-0.811
Token tackle
Token ID9761
Feature activation+8.56e-01

Pos logit contributions
atched+0.542
apped+0.540
onduct+0.470
ursed+0.438
umbered+0.437
Neg logit contributions
door-0.463
-0.454
▬▬-0.430
 pulp-0.409
taboola-0.406
Token,
Token ID11
Feature activation+9.50e-02

Pos logit contributions
atched+2.097
apped+2.084
onduct+1.811
ursed+1.681
umbered+1.677
Neg logit contributions
door-1.849
-1.817
▬▬-1.719
 pulp-1.642
taboola-1.620
Token H
Token ID367
Feature activation+1.81e+00

Pos logit contributions
atched+2.518
apped+2.502
onduct+2.196
ursed+2.045
umbered+2.041
Neg logit contributions
door-1.936
-1.897
▬▬-1.787
 pulp-1.700
taboola-1.670
Tokent
Token ID83
Feature activation+1.86e-01

Pos logit contributions
atched+3.639
apped+3.560
onduct+2.993
uked+2.726
ursed+2.722
Neg logit contributions
door-4.660
-4.557
▬▬-4.436
 pulp-4.253
taboola-4.204
Token/
Token ID14
Feature activation+1.63e+00

Pos logit contributions
atched+0.482
apped+0.479
onduct+0.420
ursed+0.391
umbered+0.391
Neg logit contributions
door-0.375
-0.368
▬▬-0.347
 pulp-0.331
taboola-0.325
TokenW
Token ID54
Feature activation+5.60e-01

Pos logit contributions
atched+3.799
apped+3.748
onduct+3.249
uked+2.959
ursed+2.934
Neg logit contributions
-3.596
door-3.535
▬▬-3.402
 pulp-3.280
taboola-3.253
Tokent
Token ID83
Feature activation-3.75e-01

Pos logit contributions
apped+1.179
atched+1.168
onduct+1.017
ursed+0.933
umbered+0.930
Neg logit contributions
door-1.408
-1.382
▬▬-1.281
 pulp-1.257
taboola-1.222
Token Comp
Token ID3082
Feature activation+3.62e+00

Pos logit contributions
door+0.807
+0.793
▬▬+0.737
 pulp+0.711
taboola+0.696
Neg logit contributions
atched-0.922
apped-0.921
onduct-0.813
ursed-0.755
uked-0.749
Token.
Token ID13
Feature activation+9.82e-01

Pos logit contributions
atched+9.772
apped+9.491
onduct+8.385
aired+7.729
ython+7.716
Neg logit contributions
door-6.521
-6.154
▬▬-6.124
taboola-5.650
 pulp-5.611
Token Rating
Token ID12028
Feature activation+6.83e-01

Pos logit contributions
atched+2.384
apped+2.362
onduct+2.051
ursed+1.896
umbered+1.893
Neg logit contributions
door-2.149
-2.105
▬▬-1.997
 pulp-1.907
taboola-1.877
Token 247
Token ID30179
Feature activation+1.47e+00

Pos logit contributions
atched+1.760
apped+1.748
onduct+1.533
ursed+1.427
umbered+1.424
Neg logit contributions
door-1.391
-1.365
▬▬-1.285
 pulp-1.227
taboola-1.206
Token ESPN
Token ID10409
Feature activation+1.21e+00

Pos logit contributions
atched+3.680
apped+3.645
onduct+3.180
umbered+2.971
ursed+2.968
Neg logit contributions
door-3.066
-2.977
▬▬-2.808
 pulp-2.719
taboola-2.650
Token Rivals
Token ID42307
Feature activation+1.43e+00

Pos logit contributions
atched+3.189
apped+3.164
onduct+2.779
umbered+2.597
ursed+2.593
Neg logit contributions
door-2.399
-2.316
▬▬-2.213
 pulp-2.082
taboola-2.079
Token factions
Token ID18783
Feature activation+6.16e-01

Pos logit contributions
atched+1.699
apped+1.686
onduct+1.454
ursed+1.339
umbered+1.336
Neg logit contributions
door-1.708
-1.679
▬▬-1.593
 pulp-1.530
taboola-1.507
Token at
Token ID379
Feature activation-2.91e-01

Pos logit contributions
atched+1.553
apped+1.543
onduct+1.348
umbered+1.251
ursed+1.250
Neg logit contributions
door-1.285
-1.264
▬▬-1.195
 pulp-1.133
taboola-1.125
Token this
Token ID428
Feature activation-3.52e-01

Pos logit contributions
door+0.575
+0.564
▬▬+0.530
 pulp+0.504
taboola+0.496
Neg logit contributions
atched-0.768
apped-0.763
onduct-0.671
ursed-0.626
umbered-0.625
Token gathering
Token ID11228
Feature activation-1.50e-02

Pos logit contributions
door+0.724
+0.704
▬▬+0.663
 pulp+0.646
ners+0.616
Neg logit contributions
atched-0.904
apped-0.897
onduct-0.792
ursed-0.734
umbered-0.732
Token,
Token ID11
Feature activation-1.81e-01

Pos logit contributions
door+0.032
+0.031
▬▬+0.030
 pulp+0.028
taboola+0.028
Neg logit contributions
atched-0.037
apped-0.037
onduct-0.032
ursed-0.030
umbered-0.030
Token i
Token ID1312
Feature activation+3.59e+00

Pos logit contributions
door+0.357
+0.349
▬▬+0.327
 pulp+0.315
taboola+0.306
Neg logit contributions
atched-0.477
apped-0.473
onduct-0.416
ursed-0.389
umbered-0.388
Token.
Token ID13
Feature activation+1.32e+00

Pos logit contributions
apped+7.577
atched+7.534
onduct+6.508
ursed+5.935
uked+5.902
Neg logit contributions
door-8.347
-8.213
▬▬-8.039
taboola-7.466
yrinth-7.464
Tokene
Token ID68
Feature activation+1.28e+00

Pos logit contributions
atched+2.653
apped+2.631
onduct+2.211
ursed+2.014
umbered+2.010
Neg logit contributions
door-3.467
-3.410
▬▬-3.247
 pulp-3.143
taboola-3.099
Token.
Token ID13
Feature activation+8.59e-01

Pos logit contributions
atched+2.503
apped+2.473
onduct+2.114
umbered+1.895
uked+1.850
Neg logit contributions
door-3.344
-3.227
▬▬-3.162
taboola-3.002
 pulp-2.994
Token an
Token ID281
Feature activation+1.46e-01

Pos logit contributions
atched+2.241
apped+2.229
onduct+1.951
ursed+1.828
uked+1.818
Neg logit contributions
door-1.706
-1.669
▬▬-1.586
taboola-1.479
 pulp-1.474
Token Old
Token ID5706
Feature activation+1.20e-01

Pos logit contributions
atched+0.397
apped+0.394
onduct+0.348
ursed+0.326
umbered+0.325
Neg logit contributions
door-0.277
-0.271
▬▬-0.254
 pulp-0.242
taboola-0.237
Token
Token ID251
Feature activation-6.12e-02

Pos logit contributions
atched+0.674
apped+0.670
onduct+0.583
ursed+0.537
umbered+0.537
Neg logit contributions
door-0.611
-0.582
▬▬-0.565
 pulp-0.544
ners-0.531
Token
Token ID198
Feature activation-1.21e-01

Pos logit contributions
door+0.129
+0.127
▬▬+0.120
 pulp+0.116
taboola+0.113
Neg logit contributions
atched-0.152
apped-0.152
onduct-0.132
umbered-0.122
ursed-0.122
Token
Token ID198
Feature activation-1.24e-01

Pos logit contributions
door+0.224
+0.211
▬▬+0.195
 pulp+0.192
ners+0.188
Neg logit contributions
atched-0.340
apped-0.336
onduct-0.298
hovah-0.283
uked-0.283
TokenRec
Token ID6690
Feature activation+1.78e+00

Pos logit contributions
door+0.250
+0.245
▬▬+0.231
 pulp+0.220
taboola+0.216
Neg logit contributions
atched-0.321
apped-0.319
onduct-0.280
ursed-0.261
umbered-0.260
Tokenru
Token ID622
Feature activation+6.70e-02

Pos logit contributions
apped+5.469
atched+5.460
onduct+4.882
ursed+4.582
umbered+4.565
Neg logit contributions
door-2.699
-2.655
▬▬-2.512
 pulp-2.326
taboola-2.274
Tokenited
Token ID863
Feature activation+3.58e+00

Pos logit contributions
atched+0.163
apped+0.162
onduct+0.142
ursed+0.132
uked+0.131
Neg logit contributions
door-0.144
-0.141
▬▬-0.132
 pulp-0.127
taboola-0.126
Token with
Token ID351
Feature activation+2.79e-01

Pos logit contributions
apped+8.638
atched+8.547
onduct+7.404
ursed+6.849
umbered+6.840
Neg logit contributions
door-7.403
-7.217
▬▬-6.915
taboola-6.532
 pulp-6.515
Token pick
Token ID2298
Feature activation+1.90e+00

Pos logit contributions
atched+0.733
apped+0.727
onduct+0.642
ursed+0.599
umbered+0.597
Neg logit contributions
door-0.553
-0.544
▬▬-0.510
 pulp-0.487
taboola-0.475
Token No
Token ID1400
Feature activation+1.16e+00

Pos logit contributions
atched+4.704
apped+4.678
onduct+4.037
umbered+3.781
ursed+3.768
Neg logit contributions
door-4.012
-3.926
▬▬-3.760
 pulp-3.579
taboola-3.532
Token.
Token ID13
Feature activation+1.44e+00

Pos logit contributions
atched+2.779
apped+2.760
onduct+2.389
umbered+2.212
ursed+2.203
Neg logit contributions
door-2.542
-2.496
▬▬-2.382
 pulp-2.269
taboola-2.238
Token43
Token ID3559
Feature activation+1.01e+00

Pos logit contributions
atched+3.510
apped+3.487
onduct+3.027
ursed+2.801
umbered+2.800
Neg logit contributions
door-3.106
-3.047
▬▬-2.887
 pulp-2.764
taboola-2.728
Token.
Token ID13
Feature activation-2.36e-01

Pos logit contributions
door+0.671
+0.656
▬▬+0.616
 pulp+0.593
taboola+0.579
Neg logit contributions
atched-0.835
apped-0.826
onduct-0.722
ursed-0.677
uked-0.673
Token
Token ID198
Feature activation-1.05e-01

Pos logit contributions
door+0.460
+0.451
▬▬+0.422
 pulp+0.403
taboola+0.395
Neg logit contributions
atched-0.629
apped-0.625
onduct-0.551
ursed-0.517
umbered-0.514
Token
Token ID198
Feature activation-1.49e-01

Pos logit contributions
door+0.195
+0.183
▬▬+0.169
 pulp+0.167
ners+0.164
Neg logit contributions
atched-0.296
apped-0.292
onduct-0.260
hovah-0.248
umbered-0.246
TokenThomas
Token ID22405
Feature activation-8.22e-01

Pos logit contributions
door+0.287
+0.281
▬▬+0.264
 pulp+0.250
taboola+0.246
Neg logit contributions
atched-0.403
apped-0.400
onduct-0.354
ursed-0.331
umbered-0.330
Token,
Token ID11
Feature activation+6.16e-01

Pos logit contributions
door+1.757
+1.719
▬▬+1.588
 pulp+1.574
ners+1.537
Neg logit contributions
atched-2.020
apped-1.987
onduct-1.747
ython-1.627
ursed-1.624
Token 39
Token ID5014
Feature activation+3.54e+00

Pos logit contributions
atched+1.526
apped+1.513
onduct+1.321
ursed+1.225
umbered+1.218
Neg logit contributions
door-1.322
-1.300
▬▬-1.226
 pulp-1.177
taboola-1.150
Token,
Token ID11
Feature activation+1.82e-01

Pos logit contributions
atched+7.466
apped+7.306
onduct+6.195
umbered+5.680
ython+5.588
Neg logit contributions
door-8.079
-7.727
▬▬-7.623
 pulp-7.132
yrinth-7.124
Token appeared
Token ID4120
Feature activation+1.27e-01

Pos logit contributions
atched+0.418
apped+0.413
onduct+0.358
ursed+0.329
ython+0.328
Neg logit contributions
door-0.424
-0.417
▬▬-0.394
 pulp-0.381
taboola-0.373
Token in
Token ID287
Feature activation+6.18e-01

Pos logit contributions
atched+0.307
apped+0.304
onduct+0.265
ursed+0.245
umbered+0.244
Neg logit contributions
door-0.279
-0.275
▬▬-0.259
 pulp-0.249
taboola-0.244
Token 53
Token ID7192
Feature activation+2.25e+00

Pos logit contributions
atched+1.532
apped+1.522
onduct+1.328
ursed+1.233
umbered+1.232
Neg logit contributions
door-1.323
-1.298
▬▬-1.221
 pulp-1.180
taboola-1.147
Token games
Token ID1830
Feature activation+3.92e-01

Pos logit contributions
atched+4.867
apped+4.826
onduct+4.136
ursed+3.758
uked+3.738
Neg logit contributions
door-5.440
-5.308
▬▬-5.131
taboola-4.822
 pulp-4.801
Token9
Token ID24
Feature activation+1.14e+00

Pos logit contributions
door+1.218
+1.186
▬▬+1.115
 pulp+1.073
taboola+1.056
Neg logit contributions
atched-1.435
apped-1.426
onduct-1.247
ursed-1.159
umbered-1.155
Tokend
Token ID67
Feature activation-2.76e+00

Pos logit contributions
atched+2.767
apped+2.746
onduct+2.388
ursed+2.210
umbered+2.206
Neg logit contributions
door-2.466
-2.428
▬▬-2.300
 pulp-2.193
taboola-2.161
Token970
Token ID43587
Feature activation+8.79e-01

Pos logit contributions
door+5.477
+5.271
▬▬+4.953
 pulp+4.822
taboola+4.745
Neg logit contributions
atched-6.822
apped-6.727
onduct-5.961
ursed-5.560
uked-5.533
Tokenb
Token ID65
Feature activation-1.70e+00

Pos logit contributions
atched+2.144
apped+2.130
onduct+1.852
ursed+1.716
umbered+1.712
Neg logit contributions
door-1.905
-1.877
▬▬-1.775
 pulp-1.693
taboola-1.667
Token L
Token ID406
Feature activation+8.00e-01

Pos logit contributions
door+3.490
+3.365
▬▬+3.176
 pulp+3.078
taboola+3.023
Neg logit contributions
atched-4.211
apped-4.167
onduct-3.674
ursed-3.402
umbered-3.398
Tokenisted
Token ID6347
Feature activation+3.54e+00

Pos logit contributions
atched+1.947
apped+1.935
onduct+1.677
ursed+1.555
umbered+1.554
Neg logit contributions
door-1.745
-1.713
▬▬-1.620
 pulp-1.551
taboola-1.527
Token below
Token ID2174
Feature activation+4.40e-01

Pos logit contributions
atched+8.418
apped+8.301
onduct+7.233
ursed+6.766
umbered+6.685
Neg logit contributions
door-7.266
-7.251
▬▬-6.820
taboola-6.533
 pulp-6.459
Token are
Token ID389
Feature activation+4.00e-01

Pos logit contributions
atched+1.116
apped+1.108
onduct+0.970
ursed+0.902
umbered+0.900
Neg logit contributions
door-0.915
-0.897
▬▬-0.845
 pulp-0.809
taboola-0.794
Token links
Token ID6117
Feature activation+6.48e-02

Pos logit contributions
atched+1.041
apped+1.034
onduct+0.908
ursed+0.846
umbered+0.844
Neg logit contributions
door-0.804
-0.789
▬▬-0.742
 pulp-0.707
taboola-0.695
Token to
Token ID284
Feature activation+3.11e-02

Pos logit contributions
atched+0.173
apped+0.172
onduct+0.151
ursed+0.141
umbered+0.141
Neg logit contributions
door-0.126
-0.124
▬▬-0.116
 pulp-0.111
taboola-0.109
Token we
Token ID356
Feature activation+3.61e-01

Pos logit contributions
atched+0.082
apped+0.081
onduct+0.072
ursed+0.067
umbered+0.067
Neg logit contributions
door-0.062
-0.060
▬▬-0.057
 pulp-0.054
taboola-0.053
Token
Token ID447
Feature activation+8.31e-02

Pos logit contributions
door+0.323
+0.317
▬▬+0.298
 pulp+0.285
taboola+0.280
Neg logit contributions
atched-0.418
apped-0.416
onduct-0.365
ursed-0.340
umbered-0.339
Token
Token ID250
Feature activation-4.40e-02

Pos logit contributions
atched+0.287
apped+0.285
onduct+0.259
ursed+0.247
umbered+0.246
Neg logit contributions
-0.098
door-0.097
▬▬-0.085
 pulp-0.077
taboola-0.076
TokenThe
Token ID464
Feature activation-3.41e-01

Pos logit contributions
door+0.084
+0.082
▬▬+0.077
 pulp+0.073
taboola+0.072
Neg logit contributions
atched-0.119
apped-0.118
onduct-0.105
ursed-0.098
umbered-0.098
Token situation
Token ID3074
Feature activation+4.67e-01

Pos logit contributions
door+0.680
+0.666
▬▬+0.624
 pulp+0.601
taboola+0.584
Neg logit contributions
atched-0.895
apped-0.887
onduct-0.783
ursed-0.729
umbered-0.728
Token may
Token ID743
Feature activation+2.06e-01

Pos logit contributions
atched+1.098
apped+1.088
onduct+0.943
ursed+0.872
umbered+0.866
Neg logit contributions
door-1.059
-1.041
▬▬-0.983
 pulp-0.945
taboola-0.933
Token rec
Token ID664
Feature activation+3.53e+00

Pos logit contributions
atched+0.532
apped+0.527
onduct+0.463
ursed+0.431
uked+0.429
Neg logit contributions
door-0.421
-0.413
▬▬-0.387
 pulp-0.373
taboola-0.366
Tokenede
Token ID18654
Feature activation+5.35e-01

Pos logit contributions
apped+6.705
atched+6.533
onduct+5.759
ursed+5.231
umbered+4.963
Neg logit contributions
door-9.101
-9.018
▬▬-9.012
 pulp-8.402
ners-8.092
Token or
Token ID393
Feature activation-1.73e-01

Pos logit contributions
atched+1.371
apped+1.350
onduct+1.185
ursed+1.106
uked+1.104
Neg logit contributions
door-1.112
-1.086
▬▬-1.023
 pulp-0.981
taboola-0.964
Token deterior
Token ID17975
Feature activation+6.13e-01

Pos logit contributions
door+0.359
+0.357
▬▬+0.334
 pulp+0.319
taboola+0.317
Neg logit contributions
atched-0.437
apped-0.433
onduct-0.379
ursed-0.355
umbered-0.351
Tokenate
Token ID378
Feature activation+1.32e+00

Pos logit contributions
atched+2.020
apped+2.010
onduct+1.812
ursed+1.720
umbered+1.719
Neg logit contributions
door-0.813
-0.787
▬▬-0.708
 pulp-0.661
taboola-0.644
Token and
Token ID290
Feature activation+3.37e-01

Pos logit contributions
atched+3.285
apped+3.278
onduct+2.859
umbered+2.649
ursed+2.646
Neg logit contributions
door-2.772
-2.727
▬▬-2.576
 pulp-2.454
taboola-2.413
Tokenrick
Token ID5557
Feature activation+1.46e+00

Pos logit contributions
atched+3.148
apped+3.143
onduct+2.713
uked+2.511
umbered+2.508
Neg logit contributions
door-2.808
-2.791
▬▬-2.635
 pulp-2.524
taboola-2.524
Token Garland
Token ID35466
Feature activation+1.43e+00

Pos logit contributions
apped+3.168
atched+3.119
onduct+2.698
umbered+2.469
ursed+2.454
Neg logit contributions
door-3.521
-3.459
▬▬-3.309
taboola-3.169
 pulp-3.135
Token
Token ID447
Feature activation+7.18e-01

Pos logit contributions
atched+3.501
apped+3.494
onduct+3.040
umbered+2.814
ursed+2.808
Neg logit contributions
door-3.049
-3.005
▬▬-2.845
 pulp-2.705
taboola-2.682
Token
Token ID247
Feature activation+1.38e-01

Pos logit contributions
atched+2.170
apped+2.162
onduct+1.936
ursed+1.825
umbered+1.822
Neg logit contributions
door-1.130
-1.052
▬▬-1.018
 pulp-0.963
taboola-0.924
Tokens
Token ID82
Feature activation+2.52e-01

Pos logit contributions
atched+0.355
apped+0.352
onduct+0.309
ursed+0.288
umbered+0.287
Neg logit contributions
door-0.282
-0.277
▬▬-0.261
 pulp-0.249
taboola-0.245
Token nomination
Token ID11872
Feature activation+3.51e+00

Pos logit contributions
atched+0.587
apped+0.582
onduct+0.503
ursed+0.465
umbered+0.464
Neg logit contributions
door-0.579
-0.568
▬▬-0.538
 pulp-0.518
taboola-0.509
Token to
Token ID284
Feature activation+1.10e+00

Pos logit contributions
apped+7.962
atched+7.953
onduct+6.875
umbered+6.501
ursed+6.291
Neg logit contributions
door-7.465
-7.422
▬▬-7.036
 pulp-6.702
ners-6.628
Token the
Token ID262
Feature activation+9.82e-01

Pos logit contributions
atched+2.794
apped+2.776
onduct+2.433
ursed+2.255
uked+2.254
Neg logit contributions
door-2.265
-2.218
▬▬-2.092
 pulp-1.981
taboola-1.968
Token Supreme
Token ID5617
Feature activation+6.50e-01

Pos logit contributions
atched+2.312
apped+2.295
onduct+1.984
ursed+1.830
umbered+1.826
Neg logit contributions
door-2.214
-2.177
▬▬-2.062
 pulp-1.973
taboola-1.953
Token Court
Token ID3078
Feature activation+5.22e-01

Pos logit contributions
atched+1.706
apped+1.688
onduct+1.478
ursed+1.395
umbered+1.392
Neg logit contributions
door-1.323
-1.285
▬▬-1.200
 pulp-1.156
taboola-1.120
Token.
Token ID13
Feature activation+7.82e-02

Pos logit contributions
atched+1.314
apped+1.304
onduct+1.140
ursed+1.059
umbered+1.055
Neg logit contributions
door-1.100
-1.079
▬▬-1.018
 pulp-0.974
taboola-0.956
Token Painter
Token ID49102
Feature activation-2.50e-01

Pos logit contributions
door+0.160
+0.157
▬▬+0.147
 pulp+0.141
taboola+0.138
Neg logit contributions
atched-0.207
apped-0.205
onduct-0.181
ursed-0.168
umbered-0.168
Token also
Token ID635
Feature activation+1.44e-01

Pos logit contributions
door+0.524
+0.518
▬▬+0.482
 pulp+0.467
ners+0.455
Neg logit contributions
atched-0.627
apped-0.618
onduct-0.541
ursed-0.505
uked-0.501
Token said
Token ID531
Feature activation-6.93e-01

Pos logit contributions
atched+0.354
apped+0.350
onduct+0.305
ursed+0.284
umbered+0.282
Neg logit contributions
door-0.310
-0.304
▬▬-0.286
 pulp-0.277
taboola-0.270
Token Ross
Token ID9847
Feature activation-9.62e-01

Pos logit contributions
door+1.362
+1.329
▬▬+1.244
 pulp+1.205
ners+1.168
Neg logit contributions
atched-1.832
apped-1.808
onduct-1.586
umbered-1.500
ursed-1.494
Token must
Token ID1276
Feature activation+2.97e-01

Pos logit contributions
door+2.101
+2.092
▬▬+1.936
 pulp+1.868
taboola+1.846
Neg logit contributions
atched-2.312
apped-2.271
onduct-1.970
ursed-1.854
uked-1.844
Token rec
Token ID664
Feature activation+3.50e+00

Pos logit contributions
atched+0.767
apped+0.762
onduct+0.668
ursed+0.622
umbered+0.621
Neg logit contributions
door-0.606
-0.594
▬▬-0.559
 pulp-0.534
taboola-0.524
Tokenuse
Token ID1904
Feature activation+1.76e+00

Pos logit contributions
apped+9.927
atched+9.649
onduct+9.085
ursed+8.386
uked+8.143
Neg logit contributions
door-5.911
-5.738
▬▬-5.729
taboola-5.123
 pulp-5.037
Token himself
Token ID2241
Feature activation+1.92e-01

Pos logit contributions
atched+3.365
apped+3.353
onduct+2.823
ursed+2.540
uked+2.518
Neg logit contributions
door-4.702
-4.630
▬▬-4.435
 pulp-4.270
taboola-4.241
Token from
Token ID422
Feature activation+1.67e-01

Pos logit contributions
atched+0.451
apped+0.445
onduct+0.381
ursed+0.354
umbered+0.354
Neg logit contributions
door-0.440
-0.430
▬▬-0.406
 pulp-0.393
taboola-0.386
Token all
Token ID477
Feature activation-1.81e-01

Pos logit contributions
atched+0.417
apped+0.413
onduct+0.360
umbered+0.337
ursed+0.336
Neg logit contributions
door-0.353
-0.347
▬▬-0.327
 pulp-0.314
taboola-0.306
Token Russia
Token ID3284
Feature activation-4.96e-01

Pos logit contributions
door+0.415
+0.408
▬▬+0.385
 pulp+0.372
ners+0.362
Neg logit contributions
atched-0.421
apped-0.414
onduct-0.356
umbered-0.333
ursed-0.333
Token!
Token ID0
Feature activation+5.99e-02

Pos logit contributions
atched+1.702
apped+1.690
onduct+1.484
ursed+1.373
umbered+1.369
Neg logit contributions
door-1.364
-1.347
▬▬-1.274
 pulp-1.213
taboola-1.200
Token<|endoftext|>
Token ID50256
Feature activation-1.19e+00

Pos logit contributions
atched+0.154
apped+0.152
onduct+0.133
ursed+0.123
umbered+0.123
Neg logit contributions
door-0.123
-0.121
▬▬-0.114
 pulp-0.109
taboola-0.108
TokenMarc
Token ID22697
Feature activation-9.06e-01

Pos logit contributions
door+2.362
+2.287
▬▬+2.163
taboola+2.033
 pulp+2.011
Neg logit contributions
atched-3.158
apped-3.123
onduct-2.770
ursed-2.591
umbered-2.581
Token Dillon
Token ID40057
Feature activation+6.17e-01

Pos logit contributions
door+1.749
+1.722
▬▬+1.568
 pulp+1.532
ners+1.499
Neg logit contributions
atched-2.374
apped-2.362
onduct-2.080
uked-1.958
ursed-1.955
Token at
Token ID379
Feature activation+1.40e+00

Pos logit contributions
atched+1.523
apped+1.511
onduct+1.318
ursed+1.222
umbered+1.219
Neg logit contributions
door-1.326
-1.301
▬▬-1.229
 pulp-1.177
taboola-1.157
Token M
Token ID337
Feature activation+3.49e+00

Pos logit contributions
atched+3.663
apped+3.640
onduct+3.208
ursed+2.973
ython+2.963
Neg logit contributions
door-2.782
-2.737
▬▬-2.578
taboola-2.410
 pulp-2.409
TokenWC
Token ID27353
Feature activation+3.24e-01

Pos logit contributions
apped+8.550
atched+8.425
onduct+7.333
ursed+6.767
uked+6.759
Neg logit contributions
door-7.197
-7.161
▬▬-6.880
 pulp-6.427
taboola-6.316
Token 2014
Token ID1946
Feature activation+1.87e-01

Pos logit contributions
atched+0.827
apped+0.820
onduct+0.716
ursed+0.668
umbered+0.667
Neg logit contributions
door-0.672
-0.660
▬▬-0.620
 pulp-0.594
taboola-0.580
Token with
Token ID351
Feature activation+5.04e-01

Pos logit contributions
atched+0.470
apped+0.466
onduct+0.407
ursed+0.379
umbered+0.379
Neg logit contributions
door-0.395
-0.387
▬▬-0.365
 pulp-0.350
taboola-0.342
Token Mak
Token ID15841
Feature activation+1.50e-01

Pos logit contributions
atched+1.293
apped+1.284
onduct+1.126
ursed+1.047
umbered+1.044
Neg logit contributions
door-1.031
-1.012
▬▬-0.953
 pulp-0.908
taboola-0.895
Tokenia
Token ID544
Feature activation+1.75e+00

Pos logit contributions
atched+0.387
apped+0.385
onduct+0.337
ursed+0.314
umbered+0.314
Neg logit contributions
door-0.305
-0.299
▬▬-0.281
 pulp-0.269
taboola-0.264
Token Patri
Token ID9606
Feature activation+2.80e+00

Pos logit contributions
atched+2.436
apped+2.418
onduct+2.093
ursed+1.933
umbered+1.930
Neg logit contributions
door-2.285
-2.249
▬▬-2.131
 pulp-2.043
taboola-2.011
Tokenk
Token ID74
Feature activation-9.00e-01

Pos logit contributions
atched+5.573
apped+5.453
onduct+4.595
umbered+4.136
uked+4.129
Neg logit contributions
door-7.307
-7.026
▬▬-6.875
taboola-6.644
yrinth-6.628
Token Elias
Token ID41462
Feature activation+1.27e+00

Pos logit contributions
+1.770
door+1.767
▬▬+1.589
 pulp+1.549
 Gum+1.503
Neg logit contributions
atched-2.382
apped-2.333
onduct-2.086
uked-1.950
ursed-1.943
Token,
Token ID837
Feature activation+8.80e-01

Pos logit contributions
atched+3.162
apped+3.144
onduct+2.744
umbered+2.542
ursed+2.537
Neg logit contributions
door-2.684
-2.636
▬▬-2.488
 pulp-2.378
taboola-2.341
Token Claude
Token ID40559
Feature activation+7.60e-02

Pos logit contributions
atched+2.171
apped+2.157
onduct+1.880
ursed+1.741
umbered+1.740
Neg logit contributions
door-1.885
-1.849
▬▬-1.749
 pulp-1.671
taboola-1.647
Token Lem
Token ID20607
Feature activation+3.47e+00

Pos logit contributions
atched+0.172
apped+0.169
onduct+0.146
uked+0.135
ursed+0.134
Neg logit contributions
door-0.180
-0.178
▬▬-0.168
 pulp-0.164
ners-0.159
Tokenie
Token ID494
Feature activation+1.25e+00

Pos logit contributions
atched+8.433
apped+8.414
onduct+7.239
ursed+6.671
umbered+6.643
Neg logit contributions
-7.068
▬▬-7.004
door-6.969
taboola-6.909
 pulp-6.461
Tokenux
Token ID2821
Feature activation+2.53e-01

Pos logit contributions
atched+2.741
apped+2.727
onduct+2.335
ursed+2.132
uked+2.123
Neg logit contributions
door-3.023
-2.974
▬▬-2.843
 pulp-2.717
taboola-2.711
Token,
Token ID11
Feature activation+7.57e-01

Pos logit contributions
atched+0.624
apped+0.619
onduct+0.540
ursed+0.500
umbered+0.499
Neg logit contributions
door-0.543
-0.533
▬▬-0.503
 pulp-0.481
taboola-0.474
Token Scott
Token ID4746
Feature activation+9.83e-01

Pos logit contributions
atched+1.826
apped+1.814
onduct+1.576
umbered+1.456
ursed+1.455
Neg logit contributions
door-1.661
-1.631
▬▬-1.544
 pulp-1.476
taboola-1.459
Token Stevens
Token ID20019
Feature activation-3.96e-01

Pos logit contributions
apped+2.301
atched+2.298
onduct+1.970
umbered+1.803
ursed+1.801
Neg logit contributions
door-2.214
-2.171
▬▬-2.112
taboola-2.011
 pulp-1.983
Token with
Token ID351
Feature activation-2.24e-01

Pos logit contributions
atched+2.848
apped+2.844
onduct+2.498
ursed+2.323
umbered+2.315
Neg logit contributions
door-2.188
-2.141
▬▬-2.027
 pulp-1.913
taboola-1.900
Token particles
Token ID13166
Feature activation-1.26e-01

Pos logit contributions
door+0.434
+0.426
▬▬+0.400
 pulp+0.381
taboola+0.373
Neg logit contributions
atched-0.599
apped-0.595
onduct-0.524
ursed-0.490
umbered-0.489
Token,
Token ID11
Feature activation-3.61e-01

Pos logit contributions
door+0.251
+0.245
▬▬+0.230
 pulp+0.221
taboola+0.215
Neg logit contributions
atched-0.331
apped-0.327
onduct-0.288
ursed-0.269
uked-0.269
Token avoiding
Token ID14928
Feature activation-1.74e+00

Pos logit contributions
door+0.682
+0.669
▬▬+0.620
 pulp+0.606
taboola+0.581
Neg logit contributions
atched-0.980
apped-0.969
onduct-0.858
ursed-0.810
uked-0.809
Token that
Token ID326
Feature activation+3.22e-01

Pos logit contributions
door+3.363
+3.340
▬▬+3.094
 pulp+3.036
taboola+2.890
Neg logit contributions
atched-4.626
apped-4.532
onduct-4.040
umbered-3.767
uked-3.753
Token numerical
Token ID29052
Feature activation+3.41e+00

Pos logit contributions
atched+0.841
apped+0.835
onduct+0.734
ursed+0.684
umbered+0.682
Neg logit contributions
door-0.645
-0.633
▬▬-0.595
 pulp-0.567
taboola-0.558
Token diffusion
Token ID44258
Feature activation-4.83e-01

Pos logit contributions
apped+8.385
atched+8.183
onduct+7.179
ursed+6.788
umbered+6.670
Neg logit contributions
door-6.663
-6.453
▬▬-6.233
taboola-5.869
yrinth-5.745
Token;
Token ID26
Feature activation+4.98e-01

Pos logit contributions
door+0.960
+0.939
▬▬+0.879
 pulp+0.852
taboola+0.820
Neg logit contributions
atched-1.274
apped-1.256
onduct-1.107
uked-1.036
umbered-1.032
Token this
Token ID428
Feature activation-2.85e-02

Pos logit contributions
atched+1.297
apped+1.287
onduct+1.131
ursed+1.054
umbered+1.051
Neg logit contributions
door-1.001
-0.982
▬▬-0.923
 pulp-0.882
taboola-0.865
Token particle
Token ID18758
Feature activation-2.69e-01

Pos logit contributions
door+0.060
+0.059
▬▬+0.055
 pulp+0.054
taboola+0.052
Neg logit contributions
atched-0.072
apped-0.071
onduct-0.062
ursed-0.058
umbered-0.058
Token-
Token ID12
Feature activation+1.43e-01

Pos logit contributions
door+0.529
+0.519
▬▬+0.486
 pulp+0.465
taboola+0.454
Neg logit contributions
atched-0.718
apped-0.711
onduct-0.627
ursed-0.586
umbered-0.584
Token in
Token ID287
Feature activation+4.00e-01

Pos logit contributions
atched+1.931
apped+1.919
onduct+1.669
ursed+1.546
umbered+1.543
Neg logit contributions
door-1.715
-1.686
▬▬-1.594
 pulp-1.524
taboola-1.499
Token Sir
Token ID7361
Feature activation-2.06e-01

Pos logit contributions
atched+1.051
apped+1.043
onduct+0.918
ursed+0.855
umbered+0.853
Neg logit contributions
door-0.794
-0.780
▬▬-0.733
 pulp-0.696
taboola-0.687
Tokensa
Token ID11400
Feature activation-1.18e+00

Pos logit contributions
door+0.419
+0.408
▬▬+0.379
 pulp+0.366
taboola+0.359
Neg logit contributions
atched-0.534
apped-0.526
onduct-0.465
ursed-0.434
umbered-0.432
Token,
Token ID11
Feature activation+5.93e-01

Pos logit contributions
door+2.291
+2.224
▬▬+2.071
 pulp+2.017
taboola+1.935
Neg logit contributions
atched-3.152
apped-3.118
onduct-2.744
ursed-2.593
umbered-2.566
Token while
Token ID981
Feature activation+1.18e+00

Pos logit contributions
atched+1.588
apped+1.577
onduct+1.392
ursed+1.298
umbered+1.295
Neg logit contributions
door-1.145
-1.124
▬▬-1.054
 pulp-1.000
taboola-0.986
Token comm
Token ID725
Feature activation+3.40e+00

Pos logit contributions
atched+3.099
apped+3.081
onduct+2.713
ursed+2.522
umbered+2.519
Neg logit contributions
door-2.304
-2.265
▬▬-2.132
 pulp-2.015
taboola-2.000
Tokenending
Token ID1571
Feature activation-5.36e-01

Pos logit contributions
atched+8.066
apped+8.062
onduct+6.936
uked+6.496
ixed+6.408
Neg logit contributions
door-7.208
-7.033
▬▬-6.872
 pulp-6.498
taboola-6.278
Token the
Token ID262
Feature activation+3.08e-01

Pos logit contributions
door+1.050
+1.033
▬▬+0.966
 pulp+0.930
taboola+0.898
Neg logit contributions
atched-1.419
apped-1.410
onduct-1.239
ursed-1.159
umbered-1.159
Token
Token ID564
Feature activation-1.29e+00

Pos logit contributions
atched+0.827
apped+0.822
onduct+0.725
ursed+0.677
umbered+0.675
Neg logit contributions
door-0.595
-0.584
▬▬-0.548
 pulp-0.520
taboola-0.512
Token
Token ID246
Feature activation+2.71e-01

Pos logit contributions
+2.083
door+2.071
▬▬+1.869
 pulp+1.755
taboola+1.718
Neg logit contributions
atched-3.897
apped-3.866
onduct-3.460
umbered-3.265
ursed-3.262
Tokenwork
Token ID1818
Feature activation-5.63e-01

Pos logit contributions
atched+0.706
apped+0.702
onduct+0.617
ursed+0.574
umbered+0.573
Neg logit contributions
door-0.544
-0.533
▬▬-0.503
 pulp-0.481
taboola-0.471
Token November
Token ID3389
Feature activation-5.91e-01

Pos logit contributions
atched+2.471
apped+2.456
onduct+2.148
ursed+1.995
umbered+1.988
Neg logit contributions
door-2.026
-1.994
▬▬-1.883
 pulp-1.785
taboola-1.775
Token 2012
Token ID2321
Feature activation-4.79e-01

Pos logit contributions
door+1.174
+1.142
▬▬+1.057
 pulp+1.034
ners+1.007
Neg logit contributions
atched-1.560
apped-1.539
onduct-1.353
uked-1.272
umbered-1.271
Token,
Token ID11
Feature activation-5.81e-01

Pos logit contributions
door+0.949
+0.930
▬▬+0.870
 pulp+0.833
taboola+0.816
Neg logit contributions
atched-1.259
apped-1.251
onduct-1.101
ursed-1.029
umbered-1.025
Token he
Token ID339
Feature activation+1.64e-01

Pos logit contributions
door+1.172
+1.144
▬▬+1.069
 pulp+1.036
taboola+1.007
Neg logit contributions
atched-1.503
apped-1.493
onduct-1.311
ursed-1.221
umbered-1.221
Token was
Token ID373
Feature activation-1.76e-01

Pos logit contributions
atched+0.368
apped+0.364
onduct+0.313
uked+0.287
ursed+0.287
Neg logit contributions
door-0.389
-0.382
▬▬-0.361
 pulp-0.350
taboola-0.344
Token listed
Token ID5610
Feature activation+3.37e+00

Pos logit contributions
door+0.365
+0.357
▬▬+0.337
 pulp+0.324
taboola+0.315
Neg logit contributions
atched-0.447
apped-0.442
onduct-0.389
ursed-0.362
ython-0.361
Token number
Token ID1271
Feature activation+1.36e+00

Pos logit contributions
atched+6.647
apped+6.625
onduct+5.531
umbered+5.166
ursed+5.123
Neg logit contributions
door-8.389
-8.305
▬▬-8.032
 pulp-7.596
taboola-7.586
Token one
Token ID530
Feature activation+8.87e-03

Pos logit contributions
atched+3.250
apped+3.227
onduct+2.783
ursed+2.598
umbered+2.583
Neg logit contributions
door-2.979
-2.929
▬▬-2.787
 pulp-2.645
taboola-2.637
Token in
Token ID287
Feature activation+8.16e-02

Pos logit contributions
atched+0.022
apped+0.022
onduct+0.019
ursed+0.017
umbered+0.017
Neg logit contributions
door-0.019
-0.019
▬▬-0.018
 pulp-0.017
taboola-0.017
Token the
Token ID262
Feature activation+1.47e-01

Pos logit contributions
atched+0.213
apped+0.211
onduct+0.186
ursed+0.173
umbered+0.173
Neg logit contributions
door-0.164
-0.161
▬▬-0.151
 pulp-0.144
taboola-0.142
Token national
Token ID2260
Feature activation+1.89e+00

Pos logit contributions
atched+0.372
apped+0.370
onduct+0.324
ursed+0.301
umbered+0.300
Neg logit contributions
door-0.307
-0.301
▬▬-0.284
 pulp-0.272
taboola-0.267
Tokens
Token ID82
Feature activation-1.82e-01

Pos logit contributions
door+0.863
+0.846
▬▬+0.798
 pulp+0.763
taboola+0.756
Neg logit contributions
atched-0.954
apped-0.946
onduct-0.821
ursed-0.762
umbered-0.759
Token army
Token ID5428
Feature activation+2.02e-01

Pos logit contributions
door+0.353
+0.343
▬▬+0.322
 pulp+0.310
taboola+0.297
Neg logit contributions
atched-0.491
apped-0.486
onduct-0.430
umbered-0.402
ursed-0.402
Token.
Token ID13
Feature activation-4.46e-01

Pos logit contributions
atched+0.506
apped+0.502
onduct+0.438
ursed+0.408
umbered+0.406
Neg logit contributions
door-0.425
-0.418
▬▬-0.394
 pulp-0.377
taboola-0.370
Token
Token ID198
Feature activation-2.42e-01

Pos logit contributions
door+0.889
+0.871
▬▬+0.820
 pulp+0.781
taboola+0.767
Neg logit contributions
atched-1.169
apped-1.161
onduct-1.021
ursed-0.951
umbered-0.949
Token
Token ID198
Feature activation-3.54e-01

Pos logit contributions
door+0.450
+0.421
▬▬+0.393
 pulp+0.386
ners+0.377
Neg logit contributions
atched-0.682
apped-0.673
onduct-0.598
hovah-0.569
uked-0.566
TokenBang
Token ID43984
Feature activation-4.34e+00

Pos logit contributions
door+0.660
+0.646
▬▬+0.605
 pulp+0.573
taboola+0.564
Neg logit contributions
atched-0.974
apped-0.968
onduct-0.857
ursed-0.802
umbered-0.800
Tokenura
Token ID5330
Feature activation-1.29e+00

Pos logit contributions
door+6.499
+5.920
ible+5.707
ners+5.178
regate+5.119
Neg logit contributions
atched-12.355
apped-12.253
onduct-10.938
hovah-10.826
umbered-10.406
Token,
Token ID11
Feature activation+3.24e-01

Pos logit contributions
door+2.629
+2.604
▬▬+2.401
 pulp+2.330
taboola+2.282
Neg logit contributions
atched-3.269
apped-3.213
onduct-2.821
ursed-2.643
uked-2.631
Token who
Token ID508
Feature activation+1.17e+00

Pos logit contributions
atched+0.813
apped+0.807
onduct+0.705
ursed+0.655
umbered+0.653
Neg logit contributions
door-0.682
-0.670
▬▬-0.631
 pulp-0.603
taboola-0.593
Token took
Token ID1718
Feature activation+2.74e-01

Pos logit contributions
atched+2.642
apped+2.628
onduct+2.260
ursed+2.081
umbered+2.079
Neg logit contributions
door-2.726
-2.682
▬▬-2.551
 pulp-2.445
taboola-2.413
Token up
Token ID510
Feature activation+1.03e-01

Pos logit contributions
atched+0.749
apped+0.742
onduct+0.657
umbered+0.614
ursed+0.613
Neg logit contributions
door-0.518
-0.508
▬▬-0.474
 pulp-0.454
taboola-0.443
Token)
Token ID8
Feature activation-2.23e-01

Pos logit contributions
door+2.508
+2.462
▬▬+2.314
 pulp+2.220
taboola+2.179
Neg logit contributions
atched-2.988
apped-2.967
onduct-2.593
ursed-2.408
umbered-2.403
Token
Token ID198
Feature activation-1.37e-01

Pos logit contributions
door+0.474
+0.467
▬▬+0.443
 pulp+0.419
taboola+0.416
Neg logit contributions
atched-0.558
apped-0.552
onduct-0.482
ursed-0.447
umbered-0.446
Token
Token ID198
Feature activation-1.95e-01

Pos logit contributions
door+0.255
+0.239
▬▬+0.221
 pulp+0.218
ners+0.213
Neg logit contributions
atched-0.385
apped-0.382
onduct-0.339
hovah-0.322
uked-0.321
Token358
Token ID31128
Feature activation+1.13e+00

Pos logit contributions
door+0.419
+0.408
▬▬+0.384
 pulp+0.368
taboola+0.364
Neg logit contributions
atched-0.483
apped-0.480
onduct-0.419
ursed-0.389
umbered-0.388
Tokenaf
Token ID1878
Feature activation-1.19e+00

Pos logit contributions
atched+2.749
apped+2.730
onduct+2.372
ursed+2.195
umbered+2.192
Neg logit contributions
door-2.463
-2.433
▬▬-2.307
 pulp-2.201
taboola-2.163
Tokend
Token ID67
Feature activation-4.16e+00

Pos logit contributions
door+2.478
+2.373
▬▬+2.191
 pulp+2.139
taboola+2.120
Neg logit contributions
atched-2.946
apped-2.939
onduct-2.582
ursed-2.401
umbered-2.391
Token4
Token ID19
Feature activation-6.84e-01

Pos logit contributions
door+7.681
+7.090
ible+6.682
taboola+6.483
 pulp+6.461
Neg logit contributions
atched-9.901
apped-9.846
onduct-8.885
hovah-8.329
ursed-8.241
Tokenbd
Token ID17457
Feature activation-1.85e+00

Pos logit contributions
door+1.484
+1.433
▬▬+1.341
 pulp+1.305
taboola+1.281
Neg logit contributions
atched-1.663
apped-1.652
onduct-1.446
ursed-1.338
umbered-1.333
Token02
Token ID2999
Feature activation-1.46e+00

Pos logit contributions
door+3.748
+3.574
▬▬+3.295
 pulp+3.236
taboola+3.199
Neg logit contributions
atched-4.619
apped-4.584
onduct-4.071
ursed-3.774
uked-3.750
Tokende
Token ID2934
Feature activation-8.22e-01

Pos logit contributions
door+3.079
+2.939
▬▬+2.738
 pulp+2.688
taboola+2.637
Neg logit contributions
atched-3.576
apped-3.545
onduct-3.130
ursed-2.892
umbered-2.890
Token3
Token ID18
Feature activation+8.48e-01

Pos logit contributions
door+1.715
+1.656
▬▬+1.535
 pulp+1.497
taboola+1.477
Neg logit contributions
atched-2.048
apped-2.041
onduct-1.790
ursed-1.661
umbered-1.660
Token
Token ID198
Feature activation-1.20e-01

Pos logit contributions
door+0.477
+0.468
▬▬+0.438
 pulp+0.417
taboola+0.409
Neg logit contributions
atched-0.657
apped-0.652
onduct-0.575
ursed-0.537
umbered-0.536
Token
Token ID198
Feature activation-2.18e-01

Pos logit contributions
door+0.223
+0.210
▬▬+0.195
 pulp+0.191
ners+0.187
Neg logit contributions
atched-0.335
apped-0.331
onduct-0.294
uked-0.278
umbered-0.278
TokenFollow
Token ID7155
Feature activation-1.46e+00

Pos logit contributions
door+0.410
+0.401
▬▬+0.375
 pulp+0.355
taboola+0.350
Neg logit contributions
atched-0.600
apped-0.596
onduct-0.527
ursed-0.494
umbered-0.492
Token @
Token ID2488
Feature activation+1.47e-01

Pos logit contributions
door+2.943
+2.888
▬▬+2.714
 pulp+2.613
taboola+2.559
Neg logit contributions
atched-3.717
apped-3.682
onduct-3.243
ursed-3.018
uked-3.017
TokenJessica
Token ID45572
Feature activation-1.15e+00

Pos logit contributions
atched+0.371
apped+0.369
onduct+0.322
ursed+0.299
umbered+0.299
Neg logit contributions
door-0.305
-0.300
▬▬-0.282
 pulp-0.270
taboola-0.265
TokenDur
Token ID36927
Feature activation-4.04e+00

Pos logit contributions
door+2.400
+2.343
▬▬+2.187
 pulp+2.070
ners+2.032
Neg logit contributions
atched-2.930
apped-2.904
onduct-2.544
ursed-2.388
umbered-2.387
Tokenando
Token ID25440
Feature activation-1.29e-01

Pos logit contributions
door+8.541
+8.123
ible+7.557
ners+7.541
▬▬+7.500
Neg logit contributions
atched-8.966
apped-8.752
onduct-7.625
umbered-7.139
ursed-6.997
Token on
Token ID319
Feature activation+4.94e-01

Pos logit contributions
door+0.268
+0.263
▬▬+0.248
 pulp+0.237
taboola+0.233
Neg logit contributions
atched-0.329
apped-0.326
onduct-0.286
ursed-0.265
umbered-0.265
Token Twitter
Token ID3009
Feature activation+2.25e-02

Pos logit contributions
atched+1.360
apped+1.352
onduct+1.196
uked+1.115
ursed+1.113
Neg logit contributions
door-0.905
-0.899
▬▬-0.828
 pulp-0.791
taboola-0.779
Token
Token ID198
Feature activation-3.17e-01

Pos logit contributions
atched+0.056
apped+0.056
onduct+0.049
ursed+0.045
umbered+0.045
Neg logit contributions
door-0.048
-0.047
▬▬-0.044
 pulp-0.042
taboola-0.042
Token
Token ID198
Feature activation-3.34e-01

Pos logit contributions
door+0.586
+0.548
▬▬+0.510
 pulp+0.500
ners+0.494
Neg logit contributions
atched-0.889
apped-0.881
onduct-0.783
hovah-0.746
uked-0.744
Tokenya
Token ID3972
Feature activation+2.06e-01

Pos logit contributions
door+1.430
+1.396
▬▬+1.304
 pulp+1.272
ners+1.256
Neg logit contributions
atched-1.277
apped-1.253
onduct-1.083
ursed-0.993
umbered-0.992
Token Sabha
Token ID41941
Feature activation+2.59e-01

Pos logit contributions
atched+0.321
apped+0.318
onduct+0.253
ursed+0.222
umbered+0.221
Neg logit contributions
door-0.627
-0.619
▬▬-0.593
 pulp-0.579
ners-0.569
Token member
Token ID2888
Feature activation-2.85e-01

Pos logit contributions
atched+0.663
apped+0.657
onduct+0.575
ursed+0.536
umbered+0.535
Neg logit contributions
door-0.535
-0.523
▬▬-0.491
 pulp-0.472
taboola-0.462
Token from
Token ID422
Feature activation+8.50e-01

Pos logit contributions
door+0.581
+0.570
▬▬+0.533
 pulp+0.515
taboola+0.501
Neg logit contributions
atched-0.736
apped-0.729
onduct-0.638
ursed-0.596
umbered-0.596
Token Gujarat
Token ID30311
Feature activation-8.12e-01

Pos logit contributions
atched+2.252
apped+2.237
onduct+1.967
ursed+1.835
uked+1.832
Neg logit contributions
door-1.666
-1.636
▬▬-1.540
 pulp-1.455
taboola-1.443
Token Shank
Token ID41595
Feature activation-3.80e+00

Pos logit contributions
door+1.589
+1.545
▬▬+1.428
 pulp+1.399
taboola+1.347
Neg logit contributions
atched-2.158
apped-2.138
onduct-1.885
ursed-1.766
umbered-1.761
Tokenar
Token ID283
Feature activation-9.04e-01

Pos logit contributions
door+6.439
+6.223
ners+5.410
 pulp+5.293
ible+5.154
Neg logit contributions
atched-10.364
apped-10.298
onduct-9.129
umbered-8.752
ursed-8.688
Token Bh
Token ID16581
Feature activation-9.55e-01

Pos logit contributions
door+1.670
+1.645
▬▬+1.467
 pulp+1.457
ners+1.413
Neg logit contributions
atched-2.489
apped-2.433
onduct-2.185
ursed-2.064
umbered-2.051
Tokenai
Token ID1872
Feature activation-4.30e-01

Pos logit contributions
door+1.718
+1.645
 pulp+1.478
ners+1.459
▬▬+1.459
Neg logit contributions
atched-2.632
apped-2.594
onduct-2.301
umbered-2.212
ursed-2.174
Token Veg
Token ID8427
Feature activation-1.20e+00

Pos logit contributions
door+0.910
+0.893
▬▬+0.839
 pulp+0.807
taboola+0.787
Neg logit contributions
atched-1.072
apped-1.064
onduct-0.928
ursed-0.865
umbered-0.861
Tokenad
Token ID324
Feature activation-1.81e+00

Pos logit contributions
door+2.383
+2.332
▬▬+2.149
 pulp+2.109
taboola+2.037
Neg logit contributions
atched-3.128
apped-3.112
onduct-2.734
ursed-2.567
umbered-2.546
Tokenial
Token ID498
Feature activation+9.67e-01

Pos logit contributions
door+1.851
+1.697
ners+1.432
ible+1.399
▬▬+1.368
Neg logit contributions
atched-5.507
apped-5.458
onduct-4.986
hovah-4.900
ursed-4.757
TokenName
Token ID5376
Feature activation+3.88e-01

Pos logit contributions
atched+2.227
apped+2.217
onduct+1.901
umbered+1.758
ursed+1.746
Neg logit contributions
door-2.206
-2.181
▬▬-2.067
 pulp-1.989
taboola-1.953
Token('
Token ID10786
Feature activation+4.41e-01

Pos logit contributions
atched+0.802
apped+0.786
onduct+0.663
umbered+0.607
ursed+0.604
Neg logit contributions
door-0.988
-0.976
▬▬-0.936
taboola-0.899
 pulp-0.898
Tokensystem
Token ID10057
Feature activation-1.45e-01

Pos logit contributions
atched+1.266
apped+1.258
onduct+1.116
umbered+1.047
uked+1.046
Neg logit contributions
door-0.764
-0.751
▬▬-0.708
 pulp-0.666
taboola-0.640
Token.
Token ID13
Feature activation-5.89e-01

Pos logit contributions
door+0.293
+0.279
▬▬+0.259
 pulp+0.253
taboola+0.251
Neg logit contributions
atched-0.383
apped-0.377
onduct-0.331
ursed-0.313
umbered-0.313
Tokenwindows
Token ID28457
Feature activation-3.78e+00

Pos logit contributions
door+1.122
+1.091
▬▬+1.018
 pulp+0.969
taboola+0.959
Neg logit contributions
atched-1.598
apped-1.588
onduct-1.404
ursed-1.315
umbered-1.312
Token.
Token ID13
Feature activation-1.12e+00

Pos logit contributions
door+7.336
+6.984
▬▬+6.353
ners+6.310
 pulp+6.102
Neg logit contributions
atched-9.801
apped-9.502
onduct-8.317
ursed-8.043
uked-7.964
Tokenforms
Token ID23914
Feature activation-1.37e+00

Pos logit contributions
door+2.245
+2.177
▬▬+2.024
 pulp+1.936
taboola+1.934
Neg logit contributions
atched-2.931
apped-2.894
onduct-2.555
ursed-2.392
umbered-2.386
Token')
Token ID11537
Feature activation-2.26e-02

Pos logit contributions
door+2.875
+2.800
▬▬+2.623
 pulp+2.510
taboola+2.491
Neg logit contributions
atched-3.417
apped-3.388
onduct-2.935
ursed-2.771
uked-2.768
Token |
Token ID930
Feature activation-2.12e-01

Pos logit contributions
door+0.048
+0.047
▬▬+0.045
 pulp+0.043
taboola+0.042
Neg logit contributions
atched-0.056
apped-0.056
onduct-0.049
ursed-0.045
umbered-0.045
Token out
Token ID503
Feature activation-5.65e-01

Pos logit contributions
door+0.467
+0.459
▬▬+0.434
 pulp+0.417
taboola+0.410
Neg logit contributions
atched-0.511
apped-0.507
onduct-0.441
ursed-0.408
umbered-0.407
Token fair
Token ID3148
Feature activation-3.89e-01

Pos logit contributions
door+0.224
+0.219
▬▬+0.206
 pulp+0.199
ners+0.192
Neg logit contributions
atched-0.298
apped-0.295
onduct-0.261
ursed-0.243
umbered-0.242
Token play
Token ID711
Feature activation-3.37e-01

Pos logit contributions
door+0.766
+0.754
▬▬+0.707
 pulp+0.680
ners+0.664
Neg logit contributions
atched-1.031
apped-1.016
onduct-0.894
ursed-0.839
uked-0.836
Token.
Token ID13
Feature activation-9.22e-02

Pos logit contributions
door+0.713
+0.699
▬▬+0.655
 pulp+0.635
ners+0.622
Neg logit contributions
atched-0.849
apped-0.837
onduct-0.726
uked-0.679
ursed-0.678
Token
Token ID198
Feature activation+1.78e-01

Pos logit contributions
door+0.182
+0.178
▬▬+0.167
 pulp+0.159
taboola+0.156
Neg logit contributions
atched-0.245
apped-0.243
onduct-0.214
ursed-0.200
umbered-0.200
Token
Token ID198
Feature activation+5.11e-02

Pos logit contributions
atched+0.489
apped+0.484
onduct+0.429
uked+0.405
ursed+0.404
Neg logit contributions
door-0.340
-0.323
▬▬-0.301
 pulp-0.293
taboola-0.287
TokenVen
Token ID37522
Feature activation-3.77e+00

Pos logit contributions
atched+0.140
apped+0.139
onduct+0.123
ursed+0.115
umbered+0.115
Neg logit contributions
door-0.097
-0.094
▬▬-0.089
 pulp-0.083
taboola-0.083
Tokenky
Token ID2584
Feature activation-6.92e-01

Pos logit contributions
door+5.603
+5.025
ners+4.793
ible+4.673
gary+4.525
Neg logit contributions
atched-10.854
apped-10.618
onduct-9.945
hovah-9.548
ursed-9.268
Token
Token ID447
Feature activation+6.26e-02

Pos logit contributions
door+1.470
+1.448
▬▬+1.358
 pulp+1.304
taboola+1.278
Neg logit contributions
atched-1.722
apped-1.697
onduct-1.490
ursed-1.382
uked-1.376
Token
Token ID247
Feature activation-2.85e-01

Pos logit contributions
atched+0.127
apped+0.126
onduct+0.107
ursed+0.097
umbered+0.097
Neg logit contributions
door-0.162
-0.156
▬▬-0.152
 pulp-0.147
taboola-0.144
Tokens
Token ID82
Feature activation-2.81e-01

Pos logit contributions
door+0.652
+0.642
▬▬+0.607
 pulp+0.583
taboola+0.575
Neg logit contributions
atched-0.661
apped-0.655
onduct-0.567
ursed-0.522
umbered-0.521
Token plans
Token ID3352
Feature activation+5.85e-01

Pos logit contributions
door+0.578
+0.567
▬▬+0.534
 pulp+0.513
ners+0.500
Neg logit contributions
atched-0.720
apped-0.712
onduct-0.627
ursed-0.582
umbered-0.581
Token.
Token ID13
Feature activation+9.03e-01

Pos logit contributions
atched+0.657
apped+0.653
onduct+0.569
ursed+0.528
umbered+0.527
Neg logit contributions
door-0.567
-0.558
▬▬-0.526
 pulp-0.503
taboola-0.495
Token 104
Token ID14436
Feature activation+1.70e+00

Pos logit contributions
atched+2.231
apped+2.213
onduct+1.931
ursed+1.791
umbered+1.787
Neg logit contributions
door-1.933
-1.894
▬▬-1.792
 pulp-1.710
taboola-1.688
Token-
Token ID12
Feature activation-2.81e-01

Pos logit contributions
atched+4.186
apped+4.120
onduct+3.602
umbered+3.395
ursed+3.346
Neg logit contributions
door-3.560
-3.461
▬▬-3.318
taboola-3.100
 pulp-3.077
Token106
Token ID15801
Feature activation+1.06e-02

Pos logit contributions
door+0.615
+0.601
▬▬+0.565
 pulp+0.543
taboola+0.537
Neg logit contributions
atched-0.684
apped-0.680
onduct-0.593
ursed-0.549
umbered-0.547
Token (
Token ID357
Feature activation+7.22e-01

Pos logit contributions
atched+0.027
apped+0.027
onduct+0.024
ursed+0.022
umbered+0.022
Neg logit contributions
door-0.022
-0.021
▬▬-0.020
 pulp-0.019
taboola-0.019
TokenAtlanta
Token ID43482
Feature activation-3.75e+00

Pos logit contributions
atched+1.940
apped+1.927
onduct+1.697
ursed+1.587
umbered+1.585
Neg logit contributions
door-1.389
-1.363
▬▬-1.282
 pulp-1.217
taboola-1.195
Token:
Token ID25
Feature activation-1.56e-01

Pos logit contributions
door+6.665
+6.633
 pulp+6.130
▬▬+5.949
ners+5.820
Neg logit contributions
atched-9.853
apped-9.831
onduct-8.743
uked-8.242
umbered-8.136
Token Scholars
Token ID26381
Feature activation+1.72e-02

Pos logit contributions
door+0.302
+0.297
▬▬+0.278
 pulp+0.266
taboola+0.260
Neg logit contributions
atched-0.418
apped-0.415
onduct-0.366
ursed-0.342
uked-0.341
Token Press
Token ID4332
Feature activation-9.79e-01

Pos logit contributions
atched+0.045
apped+0.044
onduct+0.039
ursed+0.036
umbered+0.036
Neg logit contributions
door-0.035
-0.034
▬▬-0.032
 pulp-0.030
taboola-0.030
Token,
Token ID11
Feature activation-4.10e-01

Pos logit contributions
door+1.886
+1.842
▬▬+1.701
 pulp+1.644
ners+1.616
Neg logit contributions
atched-2.637
apped-2.619
onduct-2.314
uked-2.157
ursed-2.150
Token 1985
Token ID12863
Feature activation+4.97e-01

Pos logit contributions
door+0.730
+0.713
▬▬+0.661
 pulp+0.642
taboola+0.618
Neg logit contributions
atched-1.168
apped-1.160
onduct-1.032
uked-0.968
ursed-0.964
Token the
Token ID262
Feature activation+1.08e-01

Pos logit contributions
door+0.133
+0.131
▬▬+0.121
 pulp+0.117
ners+0.112
Neg logit contributions
atched-0.213
apped-0.212
onduct-0.188
umbered-0.176
ursed-0.176
Token Australian
Token ID6638
Feature activation+1.20e-01

Pos logit contributions
atched+0.303
apped+0.301
onduct+0.268
umbered+0.251
ursed+0.250
Neg logit contributions
door-0.194
-0.190
▬▬-0.177
 pulp-0.170
taboola-0.162
Token Capital
Token ID9747
Feature activation-2.99e+00

Pos logit contributions
atched+0.379
apped+0.377
onduct+0.340
ursed+0.321
umbered+0.321
Neg logit contributions
door-0.173
-0.169
▬▬-0.154
 pulp-0.145
taboola-0.140
Token Territory
Token ID25219
Feature activation-6.18e-01

Pos logit contributions
+3.977
door+3.950
 pulp+3.345
▬▬+3.252
 Gum+3.250
Neg logit contributions
atched-9.392
apped-9.179
onduct-8.434
uked-7.931
ython-7.907
Token (
Token ID357
Feature activation+2.37e-01

Pos logit contributions
door+1.246
+1.220
▬▬+1.155
 pulp+1.096
taboola+1.079
Neg logit contributions
atched-1.607
apped-1.596
onduct-1.402
ursed-1.307
umbered-1.304
TokenACT
Token ID10659
Feature activation-3.75e+00

Pos logit contributions
atched+0.564
apped+0.558
onduct+0.485
umbered+0.447
ursed+0.447
Neg logit contributions
door-0.532
-0.524
▬▬-0.495
 pulp-0.474
taboola-0.469
Token)
Token ID8
Feature activation-7.47e-01

Pos logit contributions
door+6.619
+6.553
▬▬+5.780
 pulp+5.771
ible+5.684
Neg logit contributions
atched-10.066
apped-9.691
onduct-8.592
uked-8.454
ython-8.168
Token and
Token ID290
Feature activation-6.02e-01

Pos logit contributions
door+1.518
+1.489
▬▬+1.403
 pulp+1.338
taboola+1.316
Neg logit contributions
atched-1.929
apped-1.916
onduct-1.681
ursed-1.565
umbered-1.562
Token the
Token ID262
Feature activation+1.75e-01

Pos logit contributions
door+1.087
+1.072
▬▬+0.989
 pulp+0.960
ners+0.924
Neg logit contributions
atched-1.678
apped-1.666
onduct-1.476
umbered-1.386
uked-1.382
Token Northern
Token ID8342
Feature activation+4.36e-02

Pos logit contributions
atched+0.523
apped+0.520
onduct+0.465
ursed+0.438
umbered+0.437
Neg logit contributions
door-0.285
-0.279
▬▬-0.258
 pulp-0.245
taboola-0.236
Token Territory
Token ID25219
Feature activation-1.57e-01

Pos logit contributions
atched+0.131
apped+0.129
onduct+0.117
uked+0.111
ursed+0.110
Neg logit contributions
door-0.070
-0.069
▬▬-0.061
 pulp-0.061
ners-0.058
Token pattern
Token ID3912
Feature activation+7.82e-02

Pos logit contributions
door+0.387
+0.380
▬▬+0.356
 pulp+0.339
taboola+0.332
Neg logit contributions
atched-0.533
apped-0.531
onduct-0.468
ursed-0.438
umbered-0.437
Token of
Token ID286
Feature activation-1.24e-01

Pos logit contributions
atched+0.213
apped+0.212
onduct+0.187
ursed+0.175
umbered+0.175
Neg logit contributions
door-0.148
-0.145
▬▬-0.136
 pulp-0.129
taboola-0.126
Token ownership
Token ID9238
Feature activation-6.99e-01

Pos logit contributions
door+0.241
+0.237
▬▬+0.222
 pulp+0.211
taboola+0.207
Neg logit contributions
atched-0.331
apped-0.329
onduct-0.290
umbered-0.271
ursed-0.271
Token of
Token ID286
Feature activation-4.49e-02

Pos logit contributions
door+1.334
+1.306
▬▬+1.230
 pulp+1.171
taboola+1.136
Neg logit contributions
atched-1.888
apped-1.862
onduct-1.657
ursed-1.548
uked-1.546
Token consumer
Token ID7172
Feature activation-8.56e-01

Pos logit contributions
door+0.088
+0.086
▬▬+0.081
 pulp+0.078
taboola+0.076
Neg logit contributions
atched-0.119
apped-0.118
onduct-0.104
ursed-0.097
umbered-0.097
Token dur
Token ID22365
Feature activation-3.72e+00

Pos logit contributions
door+2.027
+1.967
▬▬+1.885
 pulp+1.833
taboola+1.766
Neg logit contributions
atched-1.923
apped-1.902
onduct-1.632
umbered-1.523
ursed-1.517
Tokenables
Token ID2977
Feature activation+6.41e-01

Pos logit contributions
door+5.395
+4.787
ible+4.646
ners+4.386
 pulp+4.340
Neg logit contributions
atched-11.073
apped-10.922
onduct-9.791
hovah-9.461
ython-9.454
Token among
Token ID1871
Feature activation+4.90e-02

Pos logit contributions
atched+1.648
apped+1.637
onduct+1.435
ursed+1.335
umbered+1.332
Neg logit contributions
door-1.311
-1.287
▬▬-1.213
 pulp-1.156
taboola-1.137
Token Indian
Token ID3942
Feature activation+1.06e+00

Pos logit contributions
atched+0.136
apped+0.135
onduct+0.120
ursed+0.112
umbered+0.112
Neg logit contributions
door-0.090
-0.088
▬▬-0.082
 pulp-0.078
taboola-0.077
Token households
Token ID12503
Feature activation+1.88e-02

Pos logit contributions
atched+2.720
apped+2.712
onduct+2.371
ursed+2.211
umbered+2.197
Neg logit contributions
door-2.162
-2.122
▬▬-2.000
 pulp-1.895
taboola-1.882
Token.
Token ID13
Feature activation-6.08e-01

Pos logit contributions
atched+0.048
apped+0.047
onduct+0.041
ursed+0.039
uked+0.038
Neg logit contributions
door-0.039
-0.039
▬▬-0.036
 pulp-0.035
taboola-0.034
Token with
Token ID351
Feature activation-2.96e-01

Pos logit contributions
atched+0.775
apped+0.767
onduct+0.670
ursed+0.622
uked+0.620
Neg logit contributions
door-0.673
-0.660
▬▬-0.623
 pulp-0.598
taboola-0.588
Token the
Token ID262
Feature activation+2.04e-01

Pos logit contributions
door+0.574
+0.565
▬▬+0.528
 pulp+0.510
taboola+0.491
Neg logit contributions
atched-0.788
apped-0.785
onduct-0.690
ursed-0.647
umbered-0.645
Token Winnipeg
Token ID23434
Feature activation-1.20e+00

Pos logit contributions
atched+0.527
apped+0.523
onduct+0.459
ursed+0.428
umbered+0.426
Neg logit contributions
door-0.416
-0.408
▬▬-0.382
 pulp-0.367
taboola-0.358
Token Jets
Token ID14728
Feature activation+1.91e+00

Pos logit contributions
door+2.191
+2.137
▬▬+1.936
 pulp+1.906
 Gum+1.813
Neg logit contributions
atched-3.345
apped-3.313
onduct-2.943
ursed-2.767
umbered-2.751
Token/
Token ID14
Feature activation+3.06e-01

Pos logit contributions
atched+4.549
apped+4.528
onduct+3.973
umbered+3.632
ursed+3.615
Neg logit contributions
door-4.112
-4.028
▬▬-3.875
 pulp-3.649
taboola-3.611
TokenAtlanta
Token ID43482
Feature activation-3.66e+00

Pos logit contributions
atched+0.799
apped+0.794
onduct+0.698
ursed+0.650
umbered+0.649
Neg logit contributions
door-0.613
-0.601
▬▬-0.566
 pulp-0.539
taboola-0.530
Token Th
Token ID536
Feature activation-2.01e+00

Pos logit contributions
+4.994
door+4.871
▬▬+4.324
 pulp+4.289
 Gum+3.984
Neg logit contributions
atched-11.442
apped-11.197
onduct-10.240
uked-9.810
ython-9.774
Tokenras
Token ID8847
Feature activation-3.15e-01

Pos logit contributions
+3.722
door+3.700
ners+3.237
 pulp+3.219
▬▬+3.188
Neg logit contributions
atched-5.363
apped-5.347
onduct-4.642
ursed-4.474
umbered-4.466
Tokenhers
Token ID7084
Feature activation+1.13e+00

Pos logit contributions
door+0.715
+0.698
▬▬+0.657
 pulp+0.630
ners+0.628
Neg logit contributions
atched-0.741
apped-0.736
onduct-0.634
umbered-0.589
uked-0.585
Token franchise
Token ID8663
Feature activation+6.75e-01

Pos logit contributions
atched+2.823
apped+2.816
onduct+2.475
umbered+2.280
ursed+2.267
Neg logit contributions
door-2.361
-2.300
▬▬-2.203
taboola-2.070
 pulp-2.066
Token,
Token ID11
Feature activation-2.65e-01

Pos logit contributions
atched+1.689
apped+1.680
onduct+1.464
ursed+1.358
umbered+1.358
Neg logit contributions
door-1.422
-1.392
▬▬-1.321
 pulp-1.252
taboola-1.232
TokenW
Token ID54
Feature activation-4.46e-01

Pos logit contributions
atched+1.786
apped+1.772
onduct+1.542
umbered+1.426
ursed+1.426
Neg logit contributions
door-1.600
-1.569
▬▬-1.486
 pulp-1.426
taboola-1.403
TokenIS
Token ID1797
Feature activation-1.78e-02

Pos logit contributions
door+0.956
+0.947
▬▬+0.883
 pulp+0.844
taboola+0.833
Neg logit contributions
atched-1.083
apped-1.080
onduct-0.939
ursed-0.877
uked-0.876
TokenQ
Token ID48
Feature activation-8.56e-01

Pos logit contributions
door+0.038
+0.038
▬▬+0.036
 pulp+0.034
taboola+0.034
Neg logit contributions
atched-0.044
apped-0.043
onduct-0.038
ursed-0.035
umbered-0.035
TokenARS
Token ID27415
Feature activation+6.68e-02

Pos logit contributions
door+2.005
+1.996
▬▬+1.875
 pulp+1.811
ners+1.765
Neg logit contributions
atched-1.931
apped-1.909
onduct-1.623
uked-1.530
umbered-1.528
Token).
Token ID737
Feature activation+2.57e-01

Pos logit contributions
atched+0.166
apped+0.165
onduct+0.144
ursed+0.133
umbered+0.133
Neg logit contributions
door-0.142
-0.140
▬▬-0.132
 pulp-0.126
taboola-0.124
Token Atlanta
Token ID9371
Feature activation-3.65e+00

Pos logit contributions
atched+0.675
apped+0.670
onduct+0.589
ursed+0.550
umbered+0.549
Neg logit contributions
door-0.512
-0.502
▬▬-0.472
 pulp-0.450
taboola-0.443
Token,
Token ID11
Feature activation-4.01e-01

Pos logit contributions
door+6.585
+6.524
▬▬+5.932
 pulp+5.822
ners+5.594
Neg logit contributions
apped-9.799
atched-9.704
onduct-8.643
ursed-8.139
uked-8.137
Token GA
Token ID14545
Feature activation-9.93e-01

Pos logit contributions
door+0.626
+0.608
▬▬+0.560
 pulp+0.529
taboola+0.516
Neg logit contributions
atched-1.227
apped-1.221
onduct-1.096
ursed-1.033
umbered-1.033
Token:
Token ID25
Feature activation+1.52e-01

Pos logit contributions
door+2.258
+2.219
▬▬+2.103
 pulp+2.018
taboola+1.988
Neg logit contributions
atched-2.322
apped-2.305
onduct-1.992
ursed-1.839
umbered-1.834
Token US
Token ID1294
Feature activation+1.61e-02

Pos logit contributions
atched+0.406
apped+0.404
onduct+0.356
ursed+0.333
umbered+0.332
Neg logit contributions
door-0.293
-0.287
▬▬-0.270
 pulp-0.257
taboola-0.252
Token Department
Token ID2732
Feature activation-2.09e+00

Pos logit contributions
atched+0.046
apped+0.046
onduct+0.041
ursed+0.038
umbered+0.038
Neg logit contributions
door-0.028
-0.028
▬▬-0.026
 pulp-0.024
taboola-0.024
Token Africa
Token ID5478
Feature activation-4.24e-01

Pos logit contributions
door+0.139
+0.135
▬▬+0.124
 pulp+0.121
ners+0.118
Neg logit contributions
atched-0.200
apped-0.197
onduct-0.175
ursed-0.166
ython-0.163
Token after
Token ID706
Feature activation-1.02e+00

Pos logit contributions
door+0.865
+0.845
▬▬+0.787
 pulp+0.769
ners+0.744
Neg logit contributions
atched-1.094
apped-1.082
onduct-0.949
ursed-0.886
uked-0.884
Token all
Token ID477
Feature activation+4.31e-01

Pos logit contributions
door+1.931
+1.891
▬▬+1.762
 pulp+1.701
ners+1.645
Neg logit contributions
atched-2.770
apped-2.739
onduct-2.423
umbered-2.286
ursed-2.275
Token.
Token ID13
Feature activation+1.29e-01

Pos logit contributions
atched+1.057
apped+1.051
onduct+0.915
ursed+0.848
umbered+0.846
Neg logit contributions
door-0.925
-0.910
▬▬-0.860
 pulp-0.821
taboola-0.816
Token "
Token ID366
Feature activation-6.47e-02

Pos logit contributions
atched+0.330
apped+0.328
onduct+0.287
ursed+0.266
umbered+0.266
Neg logit contributions
door-0.266
-0.261
▬▬-0.247
 pulp-0.235
taboola-0.232
TokenDar
Token ID32708
Feature activation-3.65e+00

Pos logit contributions
door+0.137
+0.134
▬▬+0.127
 pulp+0.121
taboola+0.119
Neg logit contributions
atched-0.162
apped-0.161
onduct-0.140
ursed-0.131
umbered-0.130
Tokenren
Token ID918
Feature activation-1.02e+00

Pos logit contributions
door+4.605
+4.329
ners+4.000
▬▬+3.650
 pulp+3.587
Neg logit contributions
atched-11.887
apped-11.632
onduct-10.722
ython-10.167
umbered-10.075
Token must
Token ID1276
Feature activation+5.11e-01

Pos logit contributions
door+2.185
+2.155
▬▬+2.006
 pulp+1.947
ners+1.908
Neg logit contributions
atched-2.503
apped-2.460
onduct-2.179
umbered-2.005
ursed-2.005
Token be
Token ID307
Feature activation+8.82e-01

Pos logit contributions
atched+1.340
apped+1.332
onduct+1.170
ursed+1.090
umbered+1.088
Neg logit contributions
door-1.018
-0.999
▬▬-0.941
 pulp-0.894
taboola-0.881
Token in
Token ID287
Feature activation+6.24e-01

Pos logit contributions
atched+2.464
apped+2.454
onduct+2.170
ursed+2.033
umbered+2.030
Neg logit contributions
door-1.601
-1.563
▬▬-1.462
 pulp-1.383
taboola-1.367
Token Fab
Token ID14236
Feature activation-1.04e+00

Pos logit contributions
atched+1.650
apped+1.645
onduct+1.443
ursed+1.349
uked+1.343
Neg logit contributions
door-1.222
-1.198
▬▬-1.125
 pulp-1.065
taboola-1.056
Token,
Token ID11
Feature activation-4.61e-02

Pos logit contributions
atched+0.361
apped+0.359
onduct+0.314
ursed+0.292
umbered+0.292
Neg logit contributions
door-0.287
-0.282
▬▬-0.266
 pulp-0.253
taboola-0.249
Token 1991
Token ID10249
Feature activation+1.48e-01

Pos logit contributions
door+0.083
+0.081
▬▬+0.075
 pulp+0.072
taboola+0.070
Neg logit contributions
atched-0.130
apped-0.129
onduct-0.115
ursed-0.108
uked-0.107
Token
Token ID1906
Feature activation-9.87e-01

Pos logit contributions
atched+0.417
apped+0.414
onduct+0.368
umbered+0.344
ursed+0.344
Neg logit contributions
door-0.263
-0.258
▬▬-0.242
 pulp-0.228
taboola-0.223
Token2009
Token ID10531
Feature activation+6.15e-01

Pos logit contributions
door+1.888
+1.831
▬▬+1.713
taboola+1.614
 pulp+1.606
Neg logit contributions
atched-2.681
apped-2.666
onduct-2.347
uked-2.216
umbered-2.209
Token.
Token ID13
Feature activation-2.47e-01

Pos logit contributions
atched+1.500
apped+1.485
onduct+1.301
umbered+1.201
ursed+1.196
Neg logit contributions
door-1.331
-1.305
▬▬-1.244
 pulp-1.182
taboola-1.167
Token Atlanta
Token ID9371
Feature activation-3.63e+00

Pos logit contributions
door+0.475
+0.465
▬▬+0.435
 pulp+0.417
taboola+0.406
Neg logit contributions
atched-0.666
apped-0.663
onduct-0.584
ursed-0.546
uked-0.545
Token,
Token ID11
Feature activation-4.73e-01

Pos logit contributions
door+6.581
+6.478
▬▬+5.928
 pulp+5.801
taboola+5.590
Neg logit contributions
apped-9.730
atched-9.717
onduct-8.614
uked-8.115
ursed-8.096
Token GA
Token ID14545
Feature activation-1.10e+00

Pos logit contributions
door+0.624
+0.585
▬▬+0.537
 pulp+0.510
taboola+0.483
Neg logit contributions
atched-1.567
apped-1.562
onduct-1.414
ython-1.353
umbered-1.352
Token:
Token ID25
Feature activation-5.98e-02

Pos logit contributions
door+2.171
+2.072
 pulp+1.966
▬▬+1.959
taboola+1.872
Neg logit contributions
atched-2.876
apped-2.865
onduct-2.490
uked-2.395
hovah-2.387
Token US
Token ID1294
Feature activation+2.14e-02

Pos logit contributions
door+0.123
+0.121
▬▬+0.113
 pulp+0.108
taboola+0.107
Neg logit contributions
atched-0.153
apped-0.152
onduct-0.134
umbered-0.124
ursed-0.124
Token Department
Token ID2732
Feature activation-1.99e+00

Pos logit contributions
atched+0.062
apped+0.062
onduct+0.055
umbered+0.052
ursed+0.051
Neg logit contributions
door-0.037
-0.036
▬▬-0.034
 pulp-0.031
taboola-0.031
Token what
Token ID644
Feature activation+7.10e-02

Pos logit contributions
door+0.675
+0.656
▬▬+0.609
 pulp+0.584
taboola+0.564
Neg logit contributions
atched-1.113
apped-1.105
onduct-0.984
umbered-0.925
ursed-0.924
Token's
Token ID338
Feature activation+5.44e-01

Pos logit contributions
atched+0.193
apped+0.191
onduct+0.169
ursed+0.158
umbered+0.158
Neg logit contributions
door-0.135
-0.133
▬▬-0.124
 pulp-0.119
taboola-0.116
Token called
Token ID1444
Feature activation-1.32e+00

Pos logit contributions
atched+1.398
apped+1.391
onduct+1.217
ursed+1.135
umbered+1.132
Neg logit contributions
door-1.108
-1.085
▬▬-1.026
 pulp-0.975
taboola-0.962
Token a
Token ID257
Feature activation-1.04e+00

Pos logit contributions
door+2.494
+2.433
▬▬+2.266
 pulp+2.189
taboola+2.120
Neg logit contributions
atched-3.582
apped-3.534
onduct-3.143
ursed-2.946
umbered-2.937
Token "
Token ID366
Feature activation-5.12e-01

Pos logit contributions
door+1.866
+1.816
▬▬+1.682
 pulp+1.623
taboola+1.563
Neg logit contributions
atched-2.944
apped-2.907
onduct-2.605
ursed-2.438
umbered-2.435
Tokend
Token ID67
Feature activation-3.60e+00

Pos logit contributions
door+1.017
+0.993
▬▬+0.931
 pulp+0.888
taboola+0.875
Neg logit contributions
atched-1.350
apped-1.339
onduct-1.180
ursed-1.100
umbered-1.098
Tokenowntown
Token ID22748
Feature activation-1.57e+00

Pos logit contributions
door+7.047
+6.641
ners+6.087
ible+6.076
▬▬+6.045
Neg logit contributions
atched-8.793
apped-8.540
onduct-7.836
ython-7.357
ursed-7.235
Token allowance
Token ID24930
Feature activation-3.54e-01

Pos logit contributions
door+2.975
+2.873
▬▬+2.673
 pulp+2.607
taboola+2.503
Neg logit contributions
atched-4.267
apped-4.200
onduct-3.755
uked-3.507
ursed-3.499
Token."
Token ID526
Feature activation-4.90e-02

Pos logit contributions
door+0.614
+0.601
▬▬+0.560
 pulp+0.529
taboola+0.518
Neg logit contributions
atched-1.019
apped-1.013
onduct-0.901
ursed-0.846
umbered-0.845
Token
Token ID198
Feature activation-6.50e-02

Pos logit contributions
door+0.094
+0.092
▬▬+0.086
 pulp+0.082
taboola+0.081
Neg logit contributions
atched-0.132
apped-0.131
onduct-0.116
ursed-0.108
umbered-0.108
Token
Token ID198
Feature activation-2.44e-01

Pos logit contributions
door+0.120
+0.113
▬▬+0.105
 pulp+0.103
ners+0.101
Neg logit contributions
atched-0.183
apped-0.181
onduct-0.160
hovah-0.153
umbered-0.152
TokenO
Token ID46
Feature activation-1.84e-01

Pos logit contributions
door+1.507
+1.485
▬▬+1.385
 pulp+1.333
taboola+1.292
Neg logit contributions
atched-1.810
apped-1.790
onduct-1.564
ursed-1.462
umbered-1.461
Token Korea
Token ID4969
Feature activation+1.14e-01

Pos logit contributions
door+0.379
+0.372
▬▬+0.351
 pulp+0.334
taboola+0.326
Neg logit contributions
atched-0.470
apped-0.466
onduct-0.408
ursed-0.381
umbered-0.381
Token
Token ID198
Feature activation-4.71e-01

Pos logit contributions
atched+0.288
apped+0.286
onduct+0.250
ursed+0.233
umbered+0.232
Neg logit contributions
door-0.236
-0.231
▬▬-0.218
 pulp-0.209
taboola-0.205
Token
Token ID198
Feature activation-5.41e-01

Pos logit contributions
door+0.857
+0.797
▬▬+0.743
 pulp+0.725
ners+0.719
Neg logit contributions
atched-1.337
apped-1.329
onduct-1.175
hovah-1.137
umbered-1.124
Tokenas
Token ID292
Feature activation-2.09e-01

Pos logit contributions
door+1.084
+1.054
▬▬+0.999
 pulp+0.938
taboola+0.931
Neg logit contributions
atched-1.417
apped-1.407
onduct-1.236
umbered-1.159
ursed-1.158
Tokend
Token ID67
Feature activation-3.59e+00

Pos logit contributions
door+0.420
+0.412
▬▬+0.387
 pulp+0.369
taboola+0.362
Neg logit contributions
atched-0.544
apped-0.540
onduct-0.474
ursed-0.443
umbered-0.443
Token
Token ID198
Feature activation+5.68e-02

Pos logit contributions
door+7.763
+7.496
▬▬+6.921
ners+6.848
ible+6.791
Neg logit contributions
atched-7.938
apped-7.842
onduct-6.877
ython-6.522
umbered-6.515
Token
Token ID198
Feature activation+7.28e-02

Pos logit contributions
atched+0.158
apped+0.156
onduct+0.138
umbered+0.131
uked+0.131
Neg logit contributions
door-0.107
-0.102
▬▬-0.094
 pulp-0.092
ners-0.090
TokenGu
Token ID8205
Feature activation-7.46e-01

Pos logit contributions
atched+0.188
apped+0.186
onduct+0.163
ursed+0.152
umbered+0.152
Neg logit contributions
door-0.149
-0.145
▬▬-0.137
 pulp-0.130
taboola-0.129
TokenM
Token ID44
Feature activation+2.12e+00

Pos logit contributions
door+1.597
+1.560
▬▬+1.463
 pulp+1.408
taboola+1.391
Neg logit contributions
atched-1.827
apped-1.812
onduct-1.593
ursed-1.478
umbered-1.474
Tokenih
Token ID4449
Feature activation+2.01e+00

Pos logit contributions
atched+5.255
apped+5.154
onduct+4.441
uked+4.168
ixed+4.133
Neg logit contributions
door-4.549
-4.427
▬▬-4.314
 pulp-4.113
taboola-4.068
Token done
Token ID1760
Feature activation-1.33e-01

Pos logit contributions
atched+0.734
apped+0.729
onduct+0.637
ursed+0.591
umbered+0.590
Neg logit contributions
door-0.619
-0.607
▬▬-0.573
 pulp-0.548
taboola-0.539
Token by
Token ID416
Feature activation-7.23e-02

Pos logit contributions
door+0.288
+0.282
▬▬+0.265
 pulp+0.256
taboola+0.250
Neg logit contributions
atched-0.326
apped-0.323
onduct-0.282
ursed-0.261
umbered-0.260
Token the
Token ID262
Feature activation+2.04e-01

Pos logit contributions
door+0.141
+0.138
▬▬+0.130
 pulp+0.124
taboola+0.121
Neg logit contributions
atched-0.193
apped-0.191
onduct-0.168
ursed-0.157
umbered-0.157
Token University
Token ID2059
Feature activation-1.31e+00

Pos logit contributions
atched+0.566
apped+0.563
onduct+0.498
ursed+0.467
umbered+0.466
Neg logit contributions
door-0.376
-0.368
▬▬-0.344
 pulp-0.326
taboola-0.321
Token of
Token ID286
Feature activation-4.56e-01

Pos logit contributions
door+2.163
+2.105
▬▬+1.925
 pulp+1.859
ners+1.794
Neg logit contributions
atched-3.876
apped-3.839
onduct-3.431
ursed-3.243
umbered-3.231
Token Victoria
Token ID12313
Feature activation-3.57e+00

Pos logit contributions
door+0.725
+0.705
▬▬+0.648
 pulp+0.620
ners+0.589
Neg logit contributions
atched-1.382
apped-1.372
onduct-1.232
ursed-1.163
umbered-1.159
Token
Token ID447
Feature activation-1.00e-01

Pos logit contributions
door+6.395
+6.248
 pulp+5.800
▬▬+5.511
ners+5.397
Neg logit contributions
atched-9.600
apped-9.359
onduct-8.303
ursed-7.908
ython-7.678
Token
Token ID247
Feature activation-4.82e-01

Pos logit contributions
door+0.152
+0.146
▬▬+0.136
 pulp+0.128
taboola+0.124
Neg logit contributions
atched-0.311
apped-0.309
onduct-0.278
ursed-0.262
umbered-0.261
Tokens
Token ID82
Feature activation-1.10e-01

Pos logit contributions
door+0.998
+0.980
▬▬+0.921
 pulp+0.880
taboola+0.866
Neg logit contributions
atched-1.228
apped-1.219
onduct-1.066
ursed-0.993
umbered-0.989
Token Environmental
Token ID13272
Feature activation-2.76e-01

Pos logit contributions
door+0.203
+0.199
▬▬+0.186
 pulp+0.177
taboola+0.173
Neg logit contributions
atched-0.305
apped-0.303
onduct-0.268
ursed-0.251
umbered-0.250
Token Law
Token ID3854
Feature activation-1.02e+00

Pos logit contributions
door+0.506
+0.495
▬▬+0.456
 pulp+0.439
taboola+0.426
Neg logit contributions
atched-0.769
apped-0.762
onduct-0.673
ursed-0.636
umbered-0.631
Tokennd
Token ID358
Feature activation-1.76e+00

Pos logit contributions
atched+1.306
apped+1.296
onduct+1.114
ursed+1.025
umbered+1.024
Neg logit contributions
door-1.353
-1.330
▬▬-1.264
 pulp-1.212
taboola-1.196
Token ed
Token ID1225
Feature activation-2.41e-01

Pos logit contributions
door+3.277
+3.210
▬▬+2.938
 pulp+2.857
ners+2.810
Neg logit contributions
atched-4.806
apped-4.794
onduct-4.264
ursed-4.013
uked-3.964
Token.
Token ID13
Feature activation+1.84e-01

Pos logit contributions
door+0.550
+0.541
▬▬+0.511
 pulp+0.491
taboola+0.484
Neg logit contributions
atched-0.564
apped-0.560
onduct-0.484
ursed-0.447
uked-0.445
Token 2010
Token ID3050
Feature activation+9.79e-01

Pos logit contributions
atched+0.505
apped+0.502
onduct+0.444
ursed+0.415
umbered+0.414
Neg logit contributions
door-0.344
-0.337
▬▬-0.315
 pulp-0.300
taboola-0.294
Token.
Token ID13
Feature activation-1.15e-01

Pos logit contributions
atched+2.271
apped+2.214
onduct+1.934
umbered+1.805
ursed+1.753
Neg logit contributions
door-2.218
-2.166
▬▬-2.095
 pulp-1.981
taboola-1.968
Token Atlanta
Token ID9371
Feature activation-3.56e+00

Pos logit contributions
door+0.228
+0.223
▬▬+0.209
 pulp+0.200
taboola+0.196
Neg logit contributions
atched-0.304
apped-0.303
onduct-0.267
ursed-0.249
uked-0.248
Token,
Token ID11
Feature activation-3.79e-01

Pos logit contributions
door+6.705
+6.601
▬▬+6.072
 pulp+5.950
ners+5.766
Neg logit contributions
apped-9.297
atched-9.269
onduct-8.211
uked-7.697
ursed-7.680
Token GA
Token ID14545
Feature activation-1.07e+00

Pos logit contributions
door+0.586
+0.571
▬▬+0.527
 pulp+0.494
taboola+0.483
Neg logit contributions
atched-1.165
apped-1.158
onduct-1.038
ursed-0.980
umbered-0.978
Token:
Token ID25
Feature activation+1.88e-01

Pos logit contributions
door+2.447
+2.406
▬▬+2.283
 pulp+2.187
taboola+2.158
Neg logit contributions
atched-2.477
apped-2.455
onduct-2.122
ursed-1.955
umbered-1.951
Token US
Token ID1294
Feature activation-1.04e-01

Pos logit contributions
atched+0.493
apped+0.490
onduct+0.431
ursed+0.401
umbered+0.401
Neg logit contributions
door-0.374
-0.367
▬▬-0.345
 pulp-0.328
taboola-0.323
Token Department
Token ID2732
Feature activation-1.95e+00

Pos logit contributions
door+0.177
+0.173
▬▬+0.161
 pulp+0.151
taboola+0.149
Neg logit contributions
atched-0.303
apped-0.301
onduct-0.268
umbered-0.252
ursed-0.252
Token Disease
Token ID17344
Feature activation-2.00e+00

Pos logit contributions
door+3.437
+3.313
▬▬+3.076
 pulp+3.041
taboola+2.955
Neg logit contributions
atched-3.867
apped-3.774
onduct-3.346
ursed-3.121
uked-3.112
Token Control
Token ID6779
Feature activation-1.26e+00

Pos logit contributions
door+3.477
+3.369
▬▬+3.056
ners+3.045
 pulp+2.998
Neg logit contributions
atched-5.606
apped-5.535
onduct-4.820
ursed-4.724
uked-4.662
Token and
Token ID290
Feature activation-1.29e+00

Pos logit contributions
door+2.408
+2.361
▬▬+2.190
 pulp+2.100
taboola+2.051
Neg logit contributions
atched-3.389
apped-3.375
onduct-2.968
ursed-2.782
uked-2.778
Token Prevention
Token ID18313
Feature activation-1.12e+00

Pos logit contributions
door+2.186
+2.107
▬▬+1.900
 pulp+1.888
ners+1.823
Neg logit contributions
atched-3.712
apped-3.692
onduct-3.284
ursed-3.154
ython-3.095
Token in
Token ID287
Feature activation-7.69e-01

Pos logit contributions
door+2.082
+2.041
▬▬+1.889
 pulp+1.802
taboola+1.758
Neg logit contributions
atched-3.097
apped-3.084
onduct-2.717
ursed-2.559
uked-2.549
Token Atlanta
Token ID9371
Feature activation-3.56e+00

Pos logit contributions
door+1.415
+1.372
▬▬+1.269
 pulp+1.239
ners+1.192
Neg logit contributions
apped-2.132
atched-2.131
onduct-1.882
ython-1.760
umbered-1.759
Token,
Token ID11
Feature activation-6.55e-01

Pos logit contributions
+5.709
door+5.666
▬▬+5.112
 pulp+5.041
taboola+4.927
Neg logit contributions
atched-10.077
apped-9.964
onduct-8.842
uked-8.420
ursed-8.382
Token Georgia
Token ID7859
Feature activation-1.73e+00

Pos logit contributions
door+1.111
+1.080
▬▬+1.000
 pulp+0.963
taboola+0.936
Neg logit contributions
atched-1.903
apped-1.901
onduct-1.684
ython-1.599
ursed-1.591
Token,
Token ID11
Feature activation-5.67e-01

Pos logit contributions
door+3.305
+3.268
▬▬+3.053
 pulp+2.911
taboola+2.869
Neg logit contributions
atched-4.623
apped-4.591
onduct-4.030
uked-3.795
ursed-3.792
Token to
Token ID284
Feature activation+2.75e-01

Pos logit contributions
door+1.059
+1.039
▬▬+0.970
 pulp+0.927
taboola+0.906
Neg logit contributions
atched-1.554
apped-1.545
onduct-1.364
ursed-1.279
umbered-1.274
Token halt
Token ID17369
Feature activation-2.91e-01

Pos logit contributions
atched+0.739
apped+0.733
onduct+0.646
ursed+0.604
umbered+0.603
Neg logit contributions
door-0.531
-0.520
▬▬-0.487
 pulp-0.467
taboola-0.456
Token isolation
Token ID15133
Feature activation-3.85e-01

Pos logit contributions
door+0.293
+0.287
▬▬+0.268
 pulp+0.260
taboola+0.250
Neg logit contributions
atched-0.394
apped-0.391
onduct-0.346
uked-0.320
ursed-0.320
Token from
Token ID422
Feature activation-5.54e-01

Pos logit contributions
door+0.850
+0.835
▬▬+0.787
 pulp+0.761
taboola+0.745
Neg logit contributions
atched-0.930
apped-0.919
onduct-0.803
ursed-0.742
uked-0.738
Token the
Token ID262
Feature activation+1.02e-01

Pos logit contributions
door+1.049
+1.029
▬▬+0.964
 pulp+0.926
taboola+0.899
Neg logit contributions
atched-1.502
apped-1.496
onduct-1.326
ursed-1.235
umbered-1.232
Token architecture
Token ID10959
Feature activation-1.85e+00

Pos logit contributions
atched+0.279
apped+0.277
onduct+0.245
ursed+0.229
umbered+0.229
Neg logit contributions
door-0.191
-0.187
▬▬-0.175
 pulp-0.168
taboola-0.163
Token-
Token ID12
Feature activation-6.46e-01

Pos logit contributions
+3.414
door+3.410
▬▬+3.137
 pulp+3.046
taboola+2.942
Neg logit contributions
atched-5.018
apped-4.936
onduct-4.404
ursed-4.098
uked-4.096
Tokenpur
Token ID14225
Feature activation-3.56e+00

Pos logit contributions
door+1.348
+1.308
▬▬+1.229
 pulp+1.180
taboola+1.162
Neg logit contributions
atched-1.637
apped-1.624
onduct-1.425
ursed-1.326
umbered-1.323
Tokenging
Token ID2667
Feature activation-1.06e+00

Pos logit contributions
door+6.723
+6.319
 pulp+5.830
ners+5.804
▬▬+5.765
Neg logit contributions
atched-9.003
apped-8.982
onduct-7.957
ython-7.469
ursed-7.426
Token mainstream
Token ID8661
Feature activation+7.25e-01

Pos logit contributions
door+2.061
+2.033
▬▬+1.906
 pulp+1.869
taboola+1.777
Neg logit contributions
atched-2.831
apped-2.805
onduct-2.486
ython-2.301
umbered-2.300
Token of
Token ID286
Feature activation+5.72e-02

Pos logit contributions
atched+1.888
apped+1.876
onduct+1.647
ursed+1.537
umbered+1.536
Neg logit contributions
door-1.450
-1.416
▬▬-1.332
 pulp-1.266
taboola-1.250
Token the
Token ID262
Feature activation-5.85e-02

Pos logit contributions
atched+0.155
apped+0.153
onduct+0.135
ursed+0.127
umbered+0.126
Neg logit contributions
door-0.109
-0.107
▬▬-0.100
 pulp-0.095
taboola-0.094
Token world
Token ID995
Feature activation+4.90e-01

Pos logit contributions
door+0.105
+0.103
▬▬+0.096
 pulp+0.091
taboola+0.089
Neg logit contributions
atched-0.166
apped-0.164
onduct-0.146
ursed-0.137
umbered-0.137
Token major
Token ID1688
Feature activation+2.14e-01

Pos logit contributions
door+1.232
+1.209
▬▬+1.123
 pulp+1.093
ners+1.041
Neg logit contributions
atched-2.022
apped-2.000
onduct-1.785
uked-1.682
ursed-1.682
Token competitors
Token ID13861
Feature activation+7.18e-01

Pos logit contributions
atched+0.561
apped+0.556
onduct+0.490
ursed+0.459
uked+0.456
Neg logit contributions
door-0.426
-0.419
▬▬-0.391
 pulp-0.379
taboola-0.368
Token in
Token ID287
Feature activation-5.54e-01

Pos logit contributions
atched+1.780
apped+1.768
onduct+1.541
ursed+1.430
umbered+1.427
Neg logit contributions
door-1.533
-1.506
▬▬-1.422
 pulp-1.359
taboola-1.339
Token the
Token ID262
Feature activation-8.74e-02

Pos logit contributions
door+1.063
+1.041
▬▬+0.973
 pulp+0.941
taboola+0.913
Neg logit contributions
atched-1.493
apped-1.479
onduct-1.307
ursed-1.225
umbered-1.222
Token Data
Token ID6060
Feature activation-9.12e-01

Pos logit contributions
door+0.164
+0.160
▬▬+0.149
 pulp+0.145
taboola+0.140
Neg logit contributions
atched-0.239
apped-0.237
onduct-0.211
ursed-0.197
uked-0.197
Token Visual
Token ID15612
Feature activation-3.56e+00

Pos logit contributions
door+1.705
+1.665
▬▬+1.549
 pulp+1.491
taboola+1.447
Neg logit contributions
atched-2.505
apped-2.477
onduct-2.202
umbered-2.062
ursed-2.062
Tokenisation
Token ID5612
Feature activation-1.38e+00

Pos logit contributions
door+6.425
+6.253
▬▬+5.715
 pulp+5.690
ners+5.582
Neg logit contributions
atched-9.540
apped-9.232
onduct-8.307
ursed-8.105
uked-7.799
Token market
Token ID1910
Feature activation-7.27e-02

Pos logit contributions
door+2.604
+2.538
▬▬+2.343
 pulp+2.306
ners+2.197
Neg logit contributions
atched-3.739
apped-3.663
onduct-3.271
ursed-3.095
uked-3.076
Token,
Token ID11
Feature activation-7.54e-01

Pos logit contributions
door+0.150
+0.147
▬▬+0.138
 pulp+0.133
taboola+0.130
Neg logit contributions
atched-0.186
apped-0.184
onduct-0.161
ursed-0.150
uked-0.150
Token a
Token ID257
Feature activation-2.66e-02

Pos logit contributions
door+1.458
+1.416
▬▬+1.335
 pulp+1.294
taboola+1.254
Neg logit contributions
atched-2.008
apped-1.982
onduct-1.747
ursed-1.653
uked-1.649
Token position
Token ID2292
Feature activation+7.88e-01

Pos logit contributions
door+0.050
+0.049
▬▬+0.046
 pulp+0.044
taboola+0.042
Neg logit contributions
atched-0.073
apped-0.072
onduct-0.065
ursed-0.060
umbered-0.060
Token only
Token ID691
Feature activation+2.64e-01

Pos logit contributions
atched+0.462
apped+0.459
onduct+0.409
ursed+0.383
umbered+0.382
Neg logit contributions
door-0.286
-0.280
▬▬-0.261
 pulp-0.248
taboola-0.241
Token Ravens
Token ID20091
Feature activation-2.49e-01

Pos logit contributions
atched+0.661
apped+0.656
onduct+0.573
ursed+0.532
umbered+0.531
Neg logit contributions
door-0.560
-0.550
▬▬-0.518
 pulp-0.496
taboola-0.487
Token All
Token ID1439
Feature activation-1.53e-01

Pos logit contributions
door+0.509
+0.505
▬▬+0.471
 pulp+0.452
ners+0.442
Neg logit contributions
atched-0.641
apped-0.632
onduct-0.560
ursed-0.520
umbered-0.516
Token-
Token ID12
Feature activation+1.21e+00

Pos logit contributions
door+0.292
+0.286
▬▬+0.269
 pulp+0.255
taboola+0.251
Neg logit contributions
atched-0.411
apped-0.410
onduct-0.361
ursed-0.337
umbered-0.336
TokenPro
Token ID2964
Feature activation+5.85e-01

Pos logit contributions
apped+3.058
atched+3.047
onduct+2.628
ixed+2.475
ursed+2.468
Neg logit contributions
door-2.426
-2.408
▬▬-2.261
 pulp-2.193
taboola-2.129
Token Marshal
Token ID37899
Feature activation+2.21e+00

Pos logit contributions
atched+1.455
apped+1.448
onduct+1.262
ursed+1.170
umbered+1.170
Neg logit contributions
door-1.239
-1.213
▬▬-1.149
 pulp-1.097
taboola-1.084
Token Y
Token ID575
Feature activation+1.46e+00

Pos logit contributions
atched+6.209
apped+6.177
onduct+5.469
ursed+5.146
umbered+5.125
Neg logit contributions
door-3.980
-3.926
▬▬-3.661
 pulp-3.463
taboola-3.387
Tokenanda
Token ID5282
Feature activation+2.06e-01

Pos logit contributions
atched+2.843
apped+2.818
onduct+2.357
ursed+2.135
umbered+2.128
Neg logit contributions
door-3.923
-3.859
▬▬-3.681
 pulp-3.562
taboola-3.515
Token
Token ID851
Feature activation-1.15e-02

Pos logit contributions
atched+0.543
apped+0.539
onduct+0.474
ursed+0.443
umbered+0.441
Neg logit contributions
door-0.408
-0.401
▬▬-0.376
 pulp-0.358
taboola-0.353
Token per
Token ID583
Feature activation+3.10e-01

Pos logit contributions
door+0.024
+0.023
▬▬+0.022
 pulp+0.021
taboola+0.021
Neg logit contributions
atched-0.029
apped-0.029
onduct-0.025
ursed-0.024
umbered-0.024
Token Pro
Token ID1041
Feature activation+1.39e+00

Pos logit contributions
atched+0.809
apped+0.804
onduct+0.706
ursed+0.658
umbered+0.657
Neg logit contributions
door-0.623
-0.611
▬▬-0.575
 pulp-0.548
taboola-0.539
Token confirmed
Token ID4999
Feature activation+5.57e-01

Pos logit contributions
atched+1.558
apped+1.544
onduct+1.334
ursed+1.227
umbered+1.221
Neg logit contributions
door-1.582
-1.554
▬▬-1.472
 pulp-1.421
taboola-1.395
Token.
Token ID13
Feature activation+2.70e-02

Pos logit contributions
atched+1.353
apped+1.346
onduct+1.170
umbered+1.082
ursed+1.072
Neg logit contributions
door-1.208
-1.187
▬▬-1.138
 pulp-1.074
taboola-1.060
Token
Token ID198
Feature activation-2.26e-01

Pos logit contributions
atched+0.067
apped+0.066
onduct+0.057
umbered+0.053
uked+0.052
Neg logit contributions
door-0.058
-0.057
▬▬-0.054
 pulp-0.052
taboola-0.051
Token
Token ID198
Feature activation-2.71e-01

Pos logit contributions
door+0.420
+0.393
▬▬+0.364
 pulp+0.358
ners+0.351
Neg logit contributions
atched-0.635
apped-0.628
onduct-0.560
hovah-0.532
uked-0.530
TokenFR
Token ID10913
Feature activation+1.10e+00

Pos logit contributions
door+0.515
+0.503
▬▬+0.472
 pulp+0.446
taboola+0.441
Neg logit contributions
atched-0.738
apped-0.733
onduct-0.649
ursed-0.608
umbered-0.605
TokenOM
Token ID2662
Feature activation+2.89e+00

Pos logit contributions
atched+2.268
apped+2.253
onduct+1.903
ursed+1.733
umbered+1.731
Neg logit contributions
door-2.803
-2.756
▬▬-2.632
 pulp-2.536
taboola-2.500
Token THE
Token ID3336
Feature activation+7.07e-01

Pos logit contributions
atched+6.717
apped+6.628
onduct+5.779
ursed+5.317
umbered+5.244
Neg logit contributions
door-6.230
-6.186
▬▬-5.810
taboola-5.522
 pulp-5.474
Token AR
Token ID5923
Feature activation-7.30e-01

Pos logit contributions
atched+1.753
apped+1.740
onduct+1.519
ursed+1.410
umbered+1.408
Neg logit contributions
door-1.515
-1.484
▬▬-1.404
 pulp-1.343
taboola-1.320
TokenCH
Token ID3398
Feature activation+4.82e-01

Pos logit contributions
door+1.411
+1.396
▬▬+1.304
 pulp+1.237
taboola+1.231
Neg logit contributions
atched-1.947
apped-1.930
onduct-1.723
uked-1.613
ursed-1.594
TokenIVE
Token ID9306
Feature activation+9.04e-01

Pos logit contributions
atched+1.421
apped+1.410
onduct+1.259
ursed+1.185
umbered+1.182
Neg logit contributions
door-0.808
-0.791
▬▬-0.734
 pulp-0.693
taboola-0.675
Token:
Token ID25
Feature activation+1.01e+00

Pos logit contributions
atched+2.196
apped+2.178
onduct+1.892
ursed+1.750
umbered+1.746
Neg logit contributions
door-1.974
-1.944
▬▬-1.834
 pulp-1.758
taboola-1.727
Token are
Token ID389
Feature activation-4.64e-01

Pos logit contributions
door+0.747
+0.731
▬▬+0.682
 pulp+0.660
ners+0.640
Neg logit contributions
atched-1.045
apped-1.031
onduct-0.910
ursed-0.854
umbered-0.850
Token not
Token ID407
Feature activation-1.76e+00

Pos logit contributions
door+0.897
+0.876
▬▬+0.822
 pulp+0.794
ners+0.767
Neg logit contributions
atched-1.245
apped-1.231
onduct-1.091
ursed-1.018
ython-1.015
Token arrested
Token ID5169
Feature activation-5.14e-01

Pos logit contributions
door+3.274
+3.168
▬▬+2.954
 pulp+2.911
ners+2.814
Neg logit contributions
atched-4.781
apped-4.718
onduct-4.241
ython-3.964
ursed-3.929
Token randomly
Token ID15456
Feature activation+1.26e+00

Pos logit contributions
door+1.075
+1.033
▬▬+0.978
 pulp+0.954
ners+0.926
Neg logit contributions
atched-1.307
apped-1.283
onduct-1.132
uked-1.053
umbered-1.049
Token in
Token ID287
Feature activation+2.18e-01

Pos logit contributions
atched+3.111
apped+3.095
onduct+2.696
ursed+2.504
umbered+2.497
Neg logit contributions
door-2.675
-2.636
▬▬-2.489
 pulp-2.368
taboola-2.349
Token North
Token ID2258
Feature activation+2.39e+00

Pos logit contributions
atched+0.623
apped+0.619
onduct+0.551
ursed+0.517
umbered+0.516
Neg logit contributions
door-0.383
-0.374
▬▬-0.348
 pulp-0.331
taboola-0.323
Token Korea
Token ID4969
Feature activation+1.31e+00

Pos logit contributions
apped+5.545
atched+5.450
onduct+4.672
umbered+4.470
uked+4.418
Neg logit contributions
door-5.204
-5.112
▬▬-5.090
yrinth-4.657
 pulp-4.649
Token.
Token ID13
Feature activation-7.71e-02

Pos logit contributions
atched+3.222
apped+3.210
onduct+2.798
ursed+2.595
umbered+2.585
Neg logit contributions
door-2.792
-2.763
▬▬-2.612
 pulp-2.470
taboola-2.456
Token If
Token ID1002
Feature activation-5.25e-01

Pos logit contributions
door+0.149
+0.145
▬▬+0.136
 pulp+0.130
taboola+0.127
Neg logit contributions
atched-0.208
apped-0.207
onduct-0.182
ursed-0.171
umbered-0.171
Token they
Token ID484
Feature activation-6.01e-01

Pos logit contributions
door+1.087
+1.059
▬▬+0.988
 pulp+0.965
ners+0.935
Neg logit contributions
atched-1.334
apped-1.325
onduct-1.164
umbered-1.087
ursed-1.081
Token are
Token ID389
Feature activation-1.32e-01

Pos logit contributions
door+1.144
+1.114
▬▬+1.032
 pulp+1.015
taboola+0.978
Neg logit contributions
atched-1.614
apped-1.599
onduct-1.418
uked-1.320
umbered-1.318
Tokenphan
Token ID19080
Feature activation+1.02e+00

Pos logit contributions
atched+2.408
apped+2.397
onduct+2.129
ursed+2.007
umbered+2.007
Neg logit contributions
door-1.265
-1.237
▬▬-1.153
 pulp-1.079
taboola-1.065
Token Black
Token ID2619
Feature activation+2.22e-01

Pos logit contributions
atched+3.334
apped+3.313
onduct+3.018
ursed+2.860
umbered+2.850
Neg logit contributions
door-1.360
-1.335
▬▬-1.190
 pulp-1.127
ners-1.088
Token BBC
Token ID7823
Feature activation+1.44e+00

Pos logit contributions
atched+0.577
apped+0.573
onduct+0.503
ursed+0.469
umbered+0.468
Neg logit contributions
door-0.448
-0.440
▬▬-0.413
 pulp-0.396
taboola-0.388
Token America
Token ID2253
Feature activation-6.52e-01

Pos logit contributions
atched+3.769
apped+3.752
onduct+3.290
ursed+3.043
umbered+3.032
Neg logit contributions
door-2.843
-2.759
▬▬-2.632
taboola-2.482
 pulp-2.453
Token Matthew
Token ID9308
Feature activation+3.96e-01

Pos logit contributions
door+1.275
+1.252
▬▬+1.164
 pulp+1.128
ners+1.093
Neg logit contributions
atched-1.722
apped-1.713
onduct-1.512
ursed-1.415
uked-1.414
Token Mc
Token ID1982
Feature activation+2.95e+00

Pos logit contributions
atched+0.952
apped+0.945
onduct+0.822
ursed+0.765
uked+0.759
Neg logit contributions
door-0.875
-0.859
▬▬-0.807
 pulp-0.777
taboola-0.760
TokenCon
Token ID3103
Feature activation-1.43e+00

Pos logit contributions
atched+6.308
apped+6.257
onduct+5.422
aired+4.955
uked+4.943
Neg logit contributions
-6.844
door-6.706
▬▬-6.672
taboola-6.249
yrinth-6.216
Tokenaug
Token ID7493
Feature activation-9.97e-01

Pos logit contributions
door+1.724
+1.521
ners+1.337
taboola+1.276
gary+1.250
Neg logit contributions
atched-4.821
apped-4.760
onduct-4.261
ython-4.167
hovah-4.165
Tokenhey
Token ID20342
Feature activation+1.32e+00

Pos logit contributions
door+1.846
+1.695
taboola+1.548
▬▬+1.532
ners+1.513
Neg logit contributions
atched-2.796
apped-2.763
onduct-2.439
ursed-2.337
uked-2.328
Token True
Token ID6407
Feature activation-5.51e-01

Pos logit contributions
atched+3.557
apped+3.540
onduct+3.124
ursed+2.909
umbered+2.902
Neg logit contributions
door-2.479
-2.431
▬▬-2.293
 pulp-2.132
taboola-2.130
Token Detective
Token ID20220
Feature activation+5.96e-01

Pos logit contributions
door+1.450
+1.405
▬▬+1.318
 pulp+1.295
ners+1.289
Neg logit contributions
atched-1.105
apped-1.095
onduct-0.929
ursed-0.854
umbered-0.841
Token factions
Token ID18783
Feature activation+6.16e-01

Pos logit contributions
atched+1.699
apped+1.686
onduct+1.454
ursed+1.339
umbered+1.336
Neg logit contributions
door-1.708
-1.679
▬▬-1.593
 pulp-1.530
taboola-1.507
Token at
Token ID379
Feature activation-2.91e-01

Pos logit contributions
atched+1.553
apped+1.543
onduct+1.348
umbered+1.251
ursed+1.250
Neg logit contributions
door-1.285
-1.264
▬▬-1.195
 pulp-1.133
taboola-1.125
Token this
Token ID428
Feature activation-3.52e-01

Pos logit contributions
door+0.575
+0.564
▬▬+0.530
 pulp+0.504
taboola+0.496
Neg logit contributions
atched-0.768
apped-0.763
onduct-0.671
ursed-0.626
umbered-0.625
Token gathering
Token ID11228
Feature activation-1.50e-02

Pos logit contributions
door+0.724
+0.704
▬▬+0.663
 pulp+0.646
ners+0.616
Neg logit contributions
atched-0.904
apped-0.897
onduct-0.792
ursed-0.734
umbered-0.732
Token,
Token ID11
Feature activation-1.81e-01

Pos logit contributions
door+0.032
+0.031
▬▬+0.030
 pulp+0.028
taboola+0.028
Neg logit contributions
atched-0.037
apped-0.037
onduct-0.032
ursed-0.030
umbered-0.030
Token i
Token ID1312
Feature activation+3.59e+00

Pos logit contributions
door+0.357
+0.349
▬▬+0.327
 pulp+0.315
taboola+0.306
Neg logit contributions
atched-0.477
apped-0.473
onduct-0.416
ursed-0.389
umbered-0.388
Token.
Token ID13
Feature activation+1.32e+00

Pos logit contributions
apped+7.577
atched+7.534
onduct+6.508
ursed+5.935
uked+5.902
Neg logit contributions
door-8.347
-8.213
▬▬-8.039
taboola-7.466
yrinth-7.464
Tokene
Token ID68
Feature activation+1.28e+00

Pos logit contributions
atched+2.653
apped+2.631
onduct+2.211
ursed+2.014
umbered+2.010
Neg logit contributions
door-3.467
-3.410
▬▬-3.247
 pulp-3.143
taboola-3.099
Token.
Token ID13
Feature activation+8.59e-01

Pos logit contributions
atched+2.503
apped+2.473
onduct+2.114
umbered+1.895
uked+1.850
Neg logit contributions
door-3.344
-3.227
▬▬-3.162
taboola-3.002
 pulp-2.994
Token an
Token ID281
Feature activation+1.46e-01

Pos logit contributions
atched+2.241
apped+2.229
onduct+1.951
ursed+1.828
uked+1.818
Neg logit contributions
door-1.706
-1.669
▬▬-1.586
taboola-1.479
 pulp-1.474
Token Old
Token ID5706
Feature activation+1.20e-01

Pos logit contributions
atched+0.397
apped+0.394
onduct+0.348
ursed+0.326
umbered+0.325
Neg logit contributions
door-0.277
-0.271
▬▬-0.254
 pulp-0.242
taboola-0.237
Tokenura
Token ID5330
Feature activation-1.09e+00

Pos logit contributions
door+4.327
+3.998
ible+3.669
ners+3.549
taboola+3.508
Neg logit contributions
apped-7.526
atched-7.507
onduct-6.644
hovah-6.463
ython-6.318
Token said
Token ID531
Feature activation-1.58e-01

Pos logit contributions
door+2.241
+2.210
▬▬+2.061
 pulp+1.984
taboola+1.954
Neg logit contributions
atched-2.767
apped-2.736
onduct-2.395
ursed-2.241
uked-2.231
Token.
Token ID13
Feature activation-1.82e-01

Pos logit contributions
door+0.344
+0.338
▬▬+0.321
 pulp+0.307
taboola+0.302
Neg logit contributions
atched-0.383
apped-0.380
onduct-0.331
umbered-0.306
ursed-0.306
Token
Token ID198
Feature activation-2.14e-01

Pos logit contributions
door+0.383
+0.374
▬▬+0.356
 pulp+0.342
taboola+0.334
Neg logit contributions
atched-0.454
apped-0.450
onduct-0.393
ursed-0.365
umbered-0.364
Token
Token ID198
Feature activation-3.62e-01

Pos logit contributions
door+0.395
+0.370
▬▬+0.344
 pulp+0.338
taboola+0.331
Neg logit contributions
atched-0.606
apped-0.597
onduct-0.532
hovah-0.507
umbered-0.504
Token
Token ID447
Feature activation+1.06e-01

Pos logit contributions
door+0.690
+0.675
▬▬+0.634
 pulp+0.599
taboola+0.592
Neg logit contributions
atched-0.982
apped-0.975
onduct-0.862
ursed-0.805
umbered-0.804
Token
Token ID250
Feature activation+3.15e-02

Pos logit contributions
atched+0.363
apped+0.360
onduct+0.327
ursed+0.311
umbered+0.311
Neg logit contributions
door-0.127
-0.126
▬▬-0.111
 pulp-0.101
taboola-0.099
TokenThe
Token ID464
Feature activation-4.13e-01

Pos logit contributions
atched+0.086
apped+0.085
onduct+0.075
ursed+0.071
umbered+0.070
Neg logit contributions
door-0.060
-0.059
▬▬-0.055
 pulp-0.052
taboola-0.051
Token position
Token ID2292
Feature activation+7.54e-01

Pos logit contributions
door+0.803
+0.784
▬▬+0.732
 pulp+0.706
taboola+0.685
Neg logit contributions
atched-1.105
apped-1.096
onduct-0.968
ursed-0.904
umbered-0.902
Token of
Token ID286
Feature activation+3.44e-01

Pos logit contributions
atched+1.969
apped+1.957
onduct+1.718
ursed+1.601
umbered+1.599
Neg logit contributions
door-1.508
-1.478
▬▬-1.393
 pulp-1.327
taboola-1.305
Token the
Token ID262
Feature activation+9.60e-02

Pos logit contributions
atched+0.905
apped+0.899
onduct+0.791
ursed+0.737
umbered+0.736
Neg logit contributions
door-0.683
-0.669
▬▬-0.629
 pulp-0.599
taboola-0.589
Token of
Token ID286
Feature activation+5.26e-01

Pos logit contributions
door+1.780
+1.736
▬▬+1.609
 pulp+1.547
ners+1.496
Neg logit contributions
atched-2.945
apped-2.905
onduct-2.593
umbered-2.446
ursed-2.436
Token attack
Token ID1368
Feature activation-3.86e-01

Pos logit contributions
atched+1.606
apped+1.592
onduct+1.432
ursed+1.338
uked+1.337
Neg logit contributions
door-0.827
-0.799
▬▬-0.744
 pulp-0.685
taboola-0.685
Token.
Token ID13
Feature activation-2.52e-01

Pos logit contributions
door+0.801
+0.774
▬▬+0.730
 pulp+0.712
ners+0.687
Neg logit contributions
atched-0.987
apped-0.975
onduct-0.849
uked-0.796
umbered-0.791
Token
Token ID198
Feature activation-1.37e-01

Pos logit contributions
door+0.497
+0.486
▬▬+0.456
 pulp+0.436
taboola+0.428
Neg logit contributions
atched-0.668
apped-0.662
onduct-0.582
ursed-0.545
umbered-0.544
Token
Token ID198
Feature activation-1.38e-01

Pos logit contributions
door+0.254
+0.239
▬▬+0.220
 pulp+0.218
ners+0.214
Neg logit contributions
atched-0.387
apped-0.383
onduct-0.340
hovah-0.325
uked-0.322
TokenNow
Token ID3844
Feature activation+1.27e+00

Pos logit contributions
door+0.264
+0.257
▬▬+0.242
 pulp+0.228
taboola+0.226
Neg logit contributions
atched-0.374
apped-0.371
onduct-0.327
ursed-0.307
umbered-0.306
Token notice
Token ID4003
Feature activation+6.75e-01

Pos logit contributions
atched+3.210
apped+3.188
onduct+2.793
ursed+2.587
umbered+2.579
Neg logit contributions
door-2.633
-2.592
▬▬-2.445
 pulp-2.331
taboola-2.301
Token the
Token ID262
Feature activation+6.33e-02

Pos logit contributions
atched+1.784
apped+1.772
onduct+1.560
ursed+1.455
umbered+1.451
Neg logit contributions
door-1.329
-1.303
▬▬-1.225
 pulp-1.165
taboola-1.147
Token variety
Token ID4996
Feature activation+1.39e+00

Pos logit contributions
atched+0.170
apped+0.169
onduct+0.149
ursed+0.139
umbered+0.139
Neg logit contributions
door-0.122
-0.120
▬▬-0.112
 pulp-0.107
taboola-0.105
Token of
Token ID286
Feature activation+7.60e-02

Pos logit contributions
atched+3.697
apped+3.694
onduct+3.249
ursed+3.042
umbered+3.034
Neg logit contributions
door-2.655
-2.624
▬▬-2.480
 pulp-2.320
taboola-2.300
Token ways
Token ID2842
Feature activation-5.96e-01

Pos logit contributions
atched+0.193
apped+0.192
onduct+0.168
ursed+0.156
umbered+0.156
Neg logit contributions
door-0.158
-0.155
▬▬-0.146
 pulp-0.140
taboola-0.137
Token/
Token ID14
Feature activation+4.74e-01

Pos logit contributions
door+1.481
+1.450
▬▬+1.371
 pulp+1.317
taboola+1.297
Neg logit contributions
atched-1.579
apped-1.562
onduct-1.351
uked-1.260
ursed-1.253
Tokenher
Token ID372
Feature activation-3.95e-01

Pos logit contributions
atched+1.003
apped+0.996
onduct+0.838
umbered+0.774
ursed+0.771
Neg logit contributions
door-1.194
-1.169
▬▬-1.106
 pulp-1.062
taboola-1.050
Tokenself
Token ID944
Feature activation-1.61e-01

Pos logit contributions
door+0.731
+0.712
▬▬+0.667
 pulp+0.634
taboola+0.623
Neg logit contributions
atched-1.094
apped-1.086
onduct-0.960
ursed-0.901
umbered-0.900
Token as
Token ID355
Feature activation-3.97e-01

Pos logit contributions
door+0.378
+0.370
▬▬+0.351
 pulp+0.339
taboola+0.332
Neg logit contributions
atched-0.365
apped-0.361
onduct-0.310
ursed-0.286
uked-0.286
Token someone
Token ID2130
Feature activation+3.10e-01

Pos logit contributions
door+0.763
+0.751
▬▬+0.699
 pulp+0.671
ners+0.650
Neg logit contributions
atched-1.067
apped-1.056
onduct-0.935
uked-0.874
ursed-0.874
Token he
Token ID339
Feature activation+5.39e-01

Pos logit contributions
atched+0.769
apped+0.764
onduct+0.666
ursed+0.618
umbered+0.617
Neg logit contributions
door-0.659
-0.648
▬▬-0.611
 pulp-0.585
taboola-0.576
Token/
Token ID14
Feature activation+9.12e-01

Pos logit contributions
atched+1.392
apped+1.381
onduct+1.212
ursed+1.128
umbered+1.126
Neg logit contributions
door-1.097
-1.078
▬▬-1.010
 pulp-0.966
taboola-0.950
Tokenshe
Token ID7091
Feature activation-2.00e-01

Pos logit contributions
atched+2.376
apped+2.359
onduct+2.061
ursed+1.931
umbered+1.930
Neg logit contributions
door-1.850
-1.805
▬▬-1.683
 pulp-1.605
taboola-1.585
Token is
Token ID318
Feature activation+7.42e-01

Pos logit contributions
door+0.433
+0.426
▬▬+0.396
 pulp+0.384
ners+0.376
Neg logit contributions
atched-0.491
apped-0.485
onduct-0.423
ursed-0.393
uked-0.392
Token not
Token ID407
Feature activation-1.03e+00

Pos logit contributions
atched+1.900
apped+1.891
onduct+1.656
umbered+1.537
ursed+1.533
Neg logit contributions
door-1.515
-1.487
▬▬-1.407
 pulp-1.336
taboola-1.325
Token through
Token ID832
Feature activation+1.54e-01

Pos logit contributions
door+2.020
+1.983
▬▬+1.835
 pulp+1.771
ners+1.731
Neg logit contributions
atched-2.735
apped-2.695
onduct-2.380
ursed-2.251
ython-2.241
Tokenous
Token ID516
Feature activation+3.47e-01

Pos logit contributions
door+0.859
+0.833
▬▬+0.773
 pulp+0.739
taboola+0.724
Neg logit contributions
atched-1.268
apped-1.259
onduct-1.118
ursed-1.047
umbered-1.041
Token Vac
Token ID25709
Feature activation+8.46e-01

Pos logit contributions
atched+0.920
apped+0.914
onduct+0.805
ursed+0.751
umbered+0.749
Neg logit contributions
door-0.681
-0.667
▬▬-0.627
 pulp-0.597
taboola-0.587
Tokenations
Token ID602
Feature activation+6.84e-01

Pos logit contributions
atched+2.635
apped+2.617
onduct+2.374
umbered+2.222
ursed+2.211
Neg logit contributions
door-1.245
-1.228
▬▬-1.163
 pulp-1.068
taboola-1.043
Token-
Token ID12
Feature activation-1.54e-01

Pos logit contributions
atched+1.755
apped+1.752
onduct+1.546
uked+1.411
ursed+1.410
Neg logit contributions
door-1.378
-1.352
▬▬-1.285
 pulp-1.232
taboola-1.202
TokenWho
Token ID8241
Feature activation+2.26e-01

Pos logit contributions
door+0.311
+0.305
▬▬+0.287
 pulp+0.273
taboola+0.269
Neg logit contributions
atched-0.401
apped-0.398
onduct-0.350
ursed-0.326
umbered-0.325
Token To
Token ID1675
Feature activation+6.33e-01

Pos logit contributions
atched+0.573
apped+0.566
onduct+0.496
ursed+0.465
umbered+0.463
Neg logit contributions
door-0.470
-0.460
▬▬-0.431
 pulp-0.413
taboola-0.406
Token Vote
Token ID19175
Feature activation+1.54e+00

Pos logit contributions
atched+1.667
apped+1.654
onduct+1.453
ursed+1.370
umbered+1.365
Neg logit contributions
door-1.251
-1.216
▬▬-1.139
 pulp-1.092
taboola-1.069
Token For
Token ID1114
Feature activation+1.33e+00

Pos logit contributions
apped+3.936
atched+3.914
onduct+3.445
ursed+3.179
uked+3.170
Neg logit contributions
door-3.036
-3.030
▬▬-2.871
 pulp-2.746
taboola-2.721
Token-
Token ID12
Feature activation+3.50e-01

Pos logit contributions
atched+3.318
apped+3.311
onduct+2.886
umbered+2.671
ursed+2.652
Neg logit contributions
door-2.728
-2.673
▬▬-2.556
 pulp-2.431
taboola-2.412
TokenJimmy
Token ID40335
Feature activation+6.99e-01

Pos logit contributions
atched+0.910
apped+0.906
onduct+0.794
ursed+0.737
umbered+0.737
Neg logit contributions
door-0.699
-0.687
▬▬-0.646
 pulp-0.618
taboola-0.606
Token Tide
Token ID31105
Feature activation+9.27e-01

Pos logit contributions
atched+1.709
apped+1.698
onduct+1.476
ursed+1.365
umbered+1.364
Neg logit contributions
door-1.514
-1.487
▬▬-1.409
 pulp-1.346
taboola-1.326
Token to
Token ID284
Feature activation+1.27e-01

Pos logit contributions
atched+2.068
apped+2.058
onduct+1.801
ursed+1.683
umbered+1.681
Neg logit contributions
door-1.567
-1.536
▬▬-1.442
 pulp-1.368
taboola-1.350
Token eliminate
Token ID11005
Feature activation+2.00e-01

Pos logit contributions
atched+0.331
apped+0.329
onduct+0.289
ursed+0.269
umbered+0.269
Neg logit contributions
door-0.256
-0.251
▬▬-0.235
 pulp-0.226
taboola-0.221
Token the
Token ID262
Feature activation+1.94e-01

Pos logit contributions
atched+0.524
apped+0.520
onduct+0.457
ursed+0.426
umbered+0.425
Neg logit contributions
door-0.397
-0.389
▬▬-0.366
 pulp-0.349
taboola-0.343
Token other
Token ID584
Feature activation+9.50e-01

Pos logit contributions
atched+0.522
apped+0.518
onduct+0.456
ursed+0.426
umbered+0.424
Neg logit contributions
door-0.374
-0.367
▬▬-0.345
 pulp-0.327
taboola-0.323
Token and
Token ID290
Feature activation+4.67e-01

Pos logit contributions
atched+2.395
apped+2.379
onduct+2.079
ursed+1.933
umbered+1.927
Neg logit contributions
door-1.978
-1.944
▬▬-1.838
 pulp-1.746
taboola-1.726
Token with
Token ID351
Feature activation+8.74e-02

Pos logit contributions
atched+1.188
apped+1.178
onduct+1.031
umbered+0.961
ursed+0.960
Neg logit contributions
door-0.974
-0.956
▬▬-0.893
 pulp-0.867
ners-0.841
Token the
Token ID262
Feature activation+9.33e-02

Pos logit contributions
atched+0.234
apped+0.233
onduct+0.205
umbered+0.192
ursed+0.192
Neg logit contributions
door-0.170
-0.166
▬▬-0.155
 pulp-0.149
taboola-0.145
Token screen
Token ID3159
Feature activation+2.88e-01

Pos logit contributions
atched+0.253
apped+0.252
onduct+0.222
umbered+0.208
ursed+0.208
Neg logit contributions
door-0.178
-0.174
▬▬-0.162
 pulp-0.156
taboola-0.151
Token door
Token ID3420
Feature activation-3.09e-01

Pos logit contributions
atched+0.793
apped+0.787
onduct+0.698
ursed+0.653
umbered+0.653
Neg logit contributions
door-0.545
-0.528
▬▬-0.492
 pulp-0.470
taboola-0.460
Token effect
Token ID1245
Feature activation-5.88e-01

Pos logit contributions
door+0.611
+0.593
▬▬+0.550
 pulp+0.535
ners+0.517
Neg logit contributions
atched-0.823
apped-0.816
onduct-0.721
uked-0.673
umbered-0.673
Token you
Token ID345
Feature activation-1.49e+00

Pos logit contributions
door+1.193
+1.165
▬▬+1.069
 pulp+1.056
ners+1.025
Neg logit contributions
atched-1.525
apped-1.508
onduct-1.328
uked-1.242
umbered-1.236
Token destruct
Token ID15256
Feature activation-1.99e+00

Pos logit contributions
atched+1.520
apped+1.511
onduct+1.342
umbered+1.262
ursed+1.261
Neg logit contributions
door-0.921
-0.902
▬▬-0.839
 pulp-0.792
taboola-0.781
Tokenible
Token ID856
Feature activation+1.04e+00

Pos logit contributions
door+1.149
+0.986
ible+0.784
▬▬+0.761
 pulp+0.650
Neg logit contributions
atched-7.886
apped-7.824
onduct-7.325
ython-6.964
uked-6.929
Token -
Token ID532
Feature activation-3.94e-02

Pos logit contributions
atched+2.442
apped+2.438
onduct+2.102
ursed+1.949
umbered+1.936
Neg logit contributions
door-2.306
-2.288
▬▬-2.176
taboola-2.056
 pulp-2.052
Token Vehicles
Token ID31365
Feature activation+6.29e-01

Pos logit contributions
door+0.076
+0.074
▬▬+0.070
 pulp+0.067
taboola+0.065
Neg logit contributions
atched-0.106
apped-0.105
onduct-0.093
ursed-0.087
umbered-0.086
Token show
Token ID905
Feature activation-5.05e-01

Pos logit contributions
atched+1.617
apped+1.611
onduct+1.409
ursed+1.314
umbered+1.313
Neg logit contributions
door-1.274
-1.256
▬▬-1.184
 pulp-1.117
taboola-1.109
Token four
Token ID1440
Feature activation-3.81e-01

Pos logit contributions
door+1.021
+1.000
▬▬+0.937
 pulp+0.902
taboola+0.877
Neg logit contributions
atched-1.314
apped-1.298
onduct-1.142
ursed-1.062
umbered-1.061
Token stages
Token ID9539
Feature activation-1.27e+00

Pos logit contributions
door+0.725
+0.705
▬▬+0.658
 pulp+0.633
taboola+0.617
Neg logit contributions
atched-1.040
apped-1.028
onduct-0.913
ursed-0.851
uked-0.848
Token of
Token ID286
Feature activation-2.11e-01

Pos logit contributions
door+2.353
+2.253
▬▬+2.079
 pulp+2.030
ners+1.968
Neg logit contributions
atched-3.498
apped-3.468
onduct-3.070
ursed-2.888
umbered-2.885
Token damage
Token ID2465
Feature activation-7.67e-01

Pos logit contributions
door+0.397
+0.389
▬▬+0.364
 pulp+0.347
taboola+0.339
Neg logit contributions
atched-0.577
apped-0.573
onduct-0.506
ursed-0.474
umbered-0.473
Token,
Token ID11
Feature activation-3.50e-01

Pos logit contributions
door+1.529
+1.486
▬▬+1.383
 pulp+1.363
ners+1.322
Neg logit contributions
atched-2.013
apped-1.987
onduct-1.753
umbered-1.636
uked-1.634
Token so
Token ID523
Feature activation+4.03e-01

Pos logit contributions
door+0.680
+0.662
▬▬+0.619
 pulp+0.599
taboola+0.580
Neg logit contributions
atched-0.936
apped-0.928
onduct-0.820
ursed-0.766
umbered-0.766
Token worth
Token ID2861
Feature activation-5.00e-01

Pos logit contributions
door+0.591
+0.579
▬▬+0.541
 pulp+0.517
taboola+0.507
Neg logit contributions
atched-0.814
apped-0.807
onduct-0.715
ursed-0.668
umbered-0.664
Token the
Token ID262
Feature activation+3.88e-01

Pos logit contributions
door+0.925
+0.913
▬▬+0.839
 pulp+0.809
taboola+0.783
Neg logit contributions
atched-1.376
apped-1.370
onduct-1.214
ursed-1.144
umbered-1.141
Token Democratic
Token ID4390
Feature activation+9.98e-01

Pos logit contributions
atched+1.012
apped+1.005
onduct+0.882
ursed+0.822
umbered+0.820
Neg logit contributions
door-0.779
-0.763
▬▬-0.719
 pulp-0.684
taboola-0.674
Token votes
Token ID5690
Feature activation+3.06e-01

Pos logit contributions
atched+2.588
apped+2.581
onduct+2.254
umbered+2.104
ursed+2.100
Neg logit contributions
door-1.997
-1.947
▬▬-1.837
 pulp-1.739
taboola-1.724
Token.
Token ID13
Feature activation-8.99e-03

Pos logit contributions
atched+0.773
apped+0.768
onduct+0.672
ursed+0.625
umbered+0.623
Neg logit contributions
door-0.637
-0.626
▬▬-0.589
 pulp-0.563
taboola-0.554
Token Georgia
Token ID7859
Feature activation-9.41e-01

Pos logit contributions
door+0.018
+0.018
▬▬+0.017
 pulp+0.016
taboola+0.016
Neg logit contributions
atched-0.023
apped-0.023
onduct-0.020
ursed-0.019
umbered-0.019
Token Rep
Token ID1432
Feature activation+7.61e-01

Pos logit contributions
door+2.043
+2.019
▬▬+1.887
 pulp+1.808
taboola+1.779
Neg logit contributions
atched-2.281
apped-2.257
onduct-1.972
uked-1.836
umbered-1.825
Token.
Token ID13
Feature activation+2.31e-01

Pos logit contributions
atched+1.789
apped+1.770
onduct+1.534
umbered+1.400
uked+1.399
Neg logit contributions
door-1.736
-1.694
▬▬-1.637
 pulp-1.565
taboola-1.539
Token St
Token ID520
Feature activation-1.25e+00

Pos logit contributions
atched+0.573
apped+0.569
onduct+0.497
ursed+0.462
umbered+0.460
Neg logit contributions
door-0.494
-0.485
▬▬-0.458
 pulp-0.439
taboola-0.431
Tokenacey
Token ID25415
Feature activation+4.63e-01

Pos logit contributions
door+2.151
+2.067
▬▬+1.850
ners+1.804
 pulp+1.794
Neg logit contributions
atched-3.535
apped-3.491
onduct-3.156
ursed-2.992
ython-2.968
Token Evans
Token ID13922
Feature activation-1.07e+00

Pos logit contributions
atched+1.073
apped+1.065
onduct+0.919
ursed+0.847
umbered+0.845
Neg logit contributions
door-1.064
-1.045
▬▬-0.994
 pulp-0.951
taboola-0.940
Token of
Token ID286
Feature activation-2.56e-02

Pos logit contributions
atched+0.538
apped+0.535
onduct+0.471
ursed+0.440
umbered+0.439
Neg logit contributions
door-0.389
-0.382
▬▬-0.359
 pulp-0.341
taboola-0.335
Token The
Token ID383
Feature activation-3.60e-01

Pos logit contributions
door+0.046
+0.045
▬▬+0.042
 pulp+0.040
taboola+0.039
Neg logit contributions
atched-0.072
apped-0.072
onduct-0.064
ursed-0.060
umbered-0.060
Tokenology
Token ID1435
Feature activation+5.82e-01

Pos logit contributions
door+0.762
+0.742
▬▬+0.682
 pulp+0.673
ners+0.654
Neg logit contributions
atched-0.902
apped-0.887
onduct-0.781
umbered-0.735
ursed-0.726
Token of
Token ID286
Feature activation-2.32e-01

Pos logit contributions
atched+1.389
apped+1.379
onduct+1.195
ursed+1.104
umbered+1.103
Neg logit contributions
door-1.293
-1.272
▬▬-1.208
 pulp-1.154
taboola-1.138
Token Em
Token ID2295
Feature activation+4.80e-01

Pos logit contributions
door+0.407
+0.399
▬▬+0.369
 pulp+0.354
taboola+0.344
Neg logit contributions
atched-0.661
apped-0.657
onduct-0.585
ursed-0.549
umbered-0.548
Tokenory
Token ID652
Feature activation-2.21e-02

Pos logit contributions
atched+1.165
apped+1.157
onduct+1.005
ursed+0.930
umbered+0.928
Neg logit contributions
door-1.052
-1.034
▬▬-0.980
 pulp-0.936
taboola-0.923
Token University
Token ID2059
Feature activation-1.51e+00

Pos logit contributions
door+0.044
+0.043
▬▬+0.040
 pulp+0.038
taboola+0.038
Neg logit contributions
atched-0.058
apped-0.058
onduct-0.051
ursed-0.047
umbered-0.047
Token,
Token ID11
Feature activation-6.08e-01

Pos logit contributions
door+3.008
+2.978
▬▬+2.756
 pulp+2.671
ners+2.603
Neg logit contributions
atched-3.876
apped-3.830
onduct-3.379
ursed-3.161
ython-3.151
Token in
Token ID287
Feature activation+4.29e-01

Pos logit contributions
door+1.228
+1.220
▬▬+1.144
 pulp+1.097
taboola+1.071
Neg logit contributions
atched-1.559
apped-1.546
onduct-1.359
ursed-1.270
umbered-1.265
Token Atlanta
Token ID9371
Feature activation-3.29e+00

Pos logit contributions
atched+1.114
apped+1.105
onduct+0.968
ursed+0.904
uked+0.901
Neg logit contributions
door-0.865
-0.849
▬▬-0.800
 pulp-0.758
taboola-0.752
Token.
Token ID13
Feature activation-2.20e-01

Pos logit contributions
+6.250
door+6.236
▬▬+5.767
 pulp+5.516
taboola+5.468
Neg logit contributions
atched-8.519
apped-8.443
onduct-7.402
ursed-7.018
uked-6.936
Token In
Token ID554
Feature activation+4.73e-01

Pos logit contributions
door+0.177
+0.173
▬▬+0.163
 pulp+0.156
taboola+0.153
Neg logit contributions
atched-0.227
apped-0.225
onduct-0.198
ursed-0.185
umbered-0.184
Token my
Token ID616
Feature activation+4.09e-03

Pos logit contributions
atched+1.274
apped+1.266
onduct+1.117
ursed+1.044
umbered+1.042
Neg logit contributions
door-0.907
-0.889
▬▬-0.833
 pulp-0.793
taboola-0.779
Token case
Token ID1339
Feature activation-5.48e-01

Pos logit contributions
atched+0.013
apped+0.013
onduct+0.012
ursed+0.011
umbered+0.011
Neg logit contributions
door-0.006
-0.006
▬▬-0.005
 pulp-0.005
taboola-0.005
Token,
Token ID11
Feature activation+4.34e-01

Pos logit contributions
door+1.115
+1.096
▬▬+1.030
 pulp+0.983
taboola+0.966
Neg logit contributions
atched-1.411
apped-1.402
onduct-1.231
ursed-1.148
uked-1.147
Token it
Token ID340
Feature activation+5.65e-01

Pos logit contributions
atched+1.059
apped+1.052
onduct+0.915
ursed+0.847
umbered+0.845
Neg logit contributions
door-0.945
-0.928
▬▬-0.878
 pulp-0.840
taboola-0.827
Token
Token ID447
Feature activation-2.50e-01

Pos logit contributions
atched+1.348
apped+1.338
onduct+1.161
ursed+1.073
umbered+1.070
Neg logit contributions
door-1.260
-1.239
▬▬-1.172
 pulp-1.124
taboola-1.107
Token
Token ID247
Feature activation-3.25e-01

Pos logit contributions
+0.570
door+0.564
▬▬+0.525
 pulp+0.505
taboola+0.503
Neg logit contributions
atched-0.588
apped-0.582
onduct-0.500
umbered-0.464
ursed-0.462
Tokens
Token ID82
Feature activation-4.02e-01

Pos logit contributions
door+0.754
+0.742
▬▬+0.703
 pulp+0.675
taboola+0.667
Neg logit contributions
atched-0.744
apped-0.738
onduct-0.635
ursed-0.585
umbered-0.584
Token for
Token ID329
Feature activation-4.17e-01

Pos logit contributions
door+0.804
+0.788
▬▬+0.743
 pulp+0.708
taboola+0.693
Neg logit contributions
atched-1.049
apped-1.039
onduct-0.914
ursed-0.853
uked-0.852
Token a
Token ID257
Feature activation-1.29e-02

Pos logit contributions
door+0.839
+0.821
▬▬+0.771
 pulp+0.735
taboola+0.718
Neg logit contributions
atched-1.085
apped-1.077
onduct-0.947
ursed-0.887
uked-0.885
Token R
Token ID371
Feature activation+1.04e+00

Pos logit contributions
door+0.026
+0.025
▬▬+0.024
 pulp+0.023
taboola+0.022
Neg logit contributions
atched-0.033
apped-0.033
onduct-0.029
ursed-0.027
umbered-0.027
TokenLatin
Token ID49022
Feature activation-3.72e-01

Pos logit contributions
atched+0.723
apped+0.718
onduct+0.632
ursed+0.590
umbered+0.589
Neg logit contributions
door-0.537
-0.527
▬▬-0.495
 pulp-0.472
taboola-0.463
Tokenas
Token ID292
Feature activation+2.55e-01

Pos logit contributions
door+0.770
+0.750
▬▬+0.703
 pulp+0.677
taboola+0.664
Neg logit contributions
atched-0.950
apped-0.941
onduct-0.827
ursed-0.771
umbered-0.770
Token on
Token ID319
Feature activation+2.28e-01

Pos logit contributions
atched+0.636
apped+0.631
onduct+0.551
ursed+0.511
umbered+0.510
Neg logit contributions
door-0.539
-0.530
▬▬-0.501
 pulp-0.478
taboola-0.471
Token Facebook
Token ID3203
Feature activation-1.06e+00

Pos logit contributions
atched+0.634
apped+0.630
onduct+0.558
ursed+0.522
umbered+0.522
Neg logit contributions
door-0.416
-0.409
▬▬-0.382
 pulp-0.361
taboola-0.356
Token!
Token ID0
Feature activation-3.00e-02

Pos logit contributions
door+2.159
+2.113
▬▬+1.989
 pulp+1.915
taboola+1.881
Neg logit contributions
atched-2.700
apped-2.672
onduct-2.361
ursed-2.206
uked-2.205
Token
Token ID198
Feature activation-1.90e-01

Pos logit contributions
door+0.060
+0.059
▬▬+0.055
 pulp+0.053
taboola+0.052
Neg logit contributions
atched-0.079
apped-0.078
onduct-0.069
ursed-0.064
umbered-0.064
Token
Token ID198
Feature activation-2.04e-01

Pos logit contributions
door+0.355
+0.336
▬▬+0.313
 pulp+0.305
taboola+0.301
Neg logit contributions
atched-0.527
apped-0.522
onduct-0.465
uked-0.440
umbered-0.438
TokenDo
Token ID5211
Feature activation-1.07e+00

Pos logit contributions
door+0.398
+0.390
▬▬+0.366
 pulp+0.348
taboola+0.343
Neg logit contributions
atched-0.545
apped-0.542
onduct-0.478
ursed-0.446
umbered-0.445
Token you
Token ID345
Feature activation-1.18e+00

Pos logit contributions
door+2.294
+2.254
▬▬+2.137
 pulp+2.046
taboola+1.998
Neg logit contributions
apped-2.622
atched-2.621
onduct-2.304
umbered-2.131
ursed-2.120
Token think
Token ID892
Feature activation-1.81e+00

Pos logit contributions
door+2.018
+1.942
▬▬+1.800
 pulp+1.784
taboola+1.699
Neg logit contributions
atched-3.370
apped-3.353
onduct-3.016
umbered-2.836
uked-2.831
Token that
Token ID326
Feature activation-1.51e-01

Pos logit contributions
door+3.282
+3.143
▬▬+2.920
 pulp+2.879
ners+2.746
Neg logit contributions
atched-5.067
apped-5.015
onduct-4.507
umbered-4.233
uked-4.170
TokenMy
Token ID3666
Feature activation+2.80e-01

Pos logit contributions
door+0.030
+0.029
▬▬+0.028
 pulp+0.026
taboola+0.026
Neg logit contributions
atched-0.041
apped-0.040
onduct-0.036
ursed-0.033
umbered-0.033
Token sample
Token ID6291
Feature activation+9.82e-01

Pos logit contributions
atched+0.768
apped+0.763
onduct+0.675
ursed+0.631
umbered+0.630
Neg logit contributions
door-0.522
-0.511
▬▬-0.479
 pulp-0.454
taboola-0.447
Token of
Token ID286
Feature activation+6.08e-01

Pos logit contributions
atched+2.484
apped+2.469
onduct+2.154
ursed+2.008
uked+2.000
Neg logit contributions
door-2.040
-2.004
▬▬-1.891
 pulp-1.799
taboola-1.779
Token Got
Token ID11853
Feature activation+2.93e-01

Pos logit contributions
atched+1.637
apped+1.626
onduct+1.432
ursed+1.339
uked+1.334
Neg logit contributions
door-1.169
-1.146
▬▬-1.077
 pulp-1.020
taboola-1.007
Tokenlands
Token ID4447
Feature activation-1.22e-01

Pos logit contributions
atched+0.701
apped+0.691
onduct+0.601
umbered+0.555
ursed+0.551
Neg logit contributions
door-0.657
-0.640
▬▬-0.603
 pulp-0.585
ners-0.577
Tokend
Token ID67
Feature activation-2.89e+00

Pos logit contributions
door+0.299
+0.296
▬▬+0.283
 pulp+0.271
taboola+0.268
Neg logit contributions
atched-0.265
apped-0.263
onduct-0.223
ursed-0.204
umbered-0.204
Tokenrick
Token ID5557
Feature activation+2.57e-01

Pos logit contributions
door+4.044
+3.439
ners+3.242
ruck+3.205
taboola+3.202
Neg logit contributions
atched-8.860
apped-8.548
hovah-7.856
onduct-7.846
umbered-7.516
Tokena
Token ID64
Feature activation-6.90e-01

Pos logit contributions
atched+0.623
apped+0.615
onduct+0.538
umbered+0.500
ursed+0.498
Neg logit contributions
door-0.570
-0.549
▬▬-0.512
 pulp-0.501
ners-0.493
Token
Token ID198
Feature activation+1.02e-01

Pos logit contributions
door+1.462
+1.435
▬▬+1.340
 pulp+1.321
ners+1.273
Neg logit contributions
atched-1.716
apped-1.690
onduct-1.499
umbered-1.383
ursed-1.370
Token
Token ID198
Feature activation+1.49e-02

Pos logit contributions
atched+0.284
apped+0.282
onduct+0.250
umbered+0.236
uked+0.236
Neg logit contributions
door-0.193
-0.182
▬▬-0.169
 pulp-0.166
ners-0.162
TokenAnd
Token ID1870
Feature activation+1.51e+00

Pos logit contributions
atched+0.039
apped+0.039
onduct+0.035
ursed+0.032
umbered+0.032
Neg logit contributions
door-0.030
-0.029
▬▬-0.027
 pulp-0.026
taboola-0.025
Token call
Token ID869
Feature activation+3.59e-01

Pos logit contributions
door+0.058
+0.056
▬▬+0.053
 pulp+0.051
taboola+0.050
Neg logit contributions
atched-0.075
apped-0.075
onduct-0.066
ursed-0.061
umbered-0.061
Token for
Token ID329
Feature activation-4.49e-01

Pos logit contributions
apped+0.834
atched+0.834
onduct+0.712
ursed+0.662
uked+0.657
Neg logit contributions
door-0.814
-0.806
▬▬-0.767
 pulp-0.733
taboola-0.716
Token his
Token ID465
Feature activation-4.62e-01

Pos logit contributions
door+0.869
+0.854
▬▬+0.800
 pulp+0.766
ners+0.742
Neg logit contributions
atched-1.200
apped-1.194
onduct-1.053
umbered-0.987
ursed-0.980
Token "
Token ID366
Feature activation-1.70e-01

Pos logit contributions
door+0.929
+0.914
▬▬+0.853
 pulp+0.825
ners+0.797
Neg logit contributions
atched-1.198
apped-1.191
onduct-1.051
umbered-0.983
ursed-0.980
TokenTravel
Token ID33074
Feature activation-1.21e+00

Pos logit contributions
door+0.333
+0.326
▬▬+0.306
 pulp+0.290
taboola+0.285
Neg logit contributions
atched-0.451
apped-0.448
onduct-0.395
ursed-0.368
umbered-0.368
Token Ban
Token ID10274
Feature activation-2.39e+00

Pos logit contributions
door+2.186
+2.160
▬▬+1.985
 pulp+1.906
ners+1.862
Neg logit contributions
atched-3.377
apped-3.310
onduct-2.950
ursed-2.810
umbered-2.778
Token"
Token ID1
Feature activation+6.43e-01

Pos logit contributions
door+3.764
+3.584
▬▬+3.379
ners+3.190
taboola+3.186
Neg logit contributions
atched-7.154
apped-7.037
onduct-6.222
ursed-5.954
ython-5.930
Token on
Token ID319
Feature activation+1.18e-01

Pos logit contributions
atched+1.662
apped+1.653
onduct+1.452
ursed+1.351
umbered+1.348
Neg logit contributions
door-1.296
-1.275
▬▬-1.201
 pulp-1.141
taboola-1.125
Token Twitter
Token ID3009
Feature activation-4.44e-01

Pos logit contributions
atched+0.317
apped+0.314
onduct+0.278
umbered+0.259
ursed+0.258
Neg logit contributions
door-0.230
-0.225
▬▬-0.211
 pulp-0.201
taboola-0.197
Token as
Token ID355
Feature activation-5.05e-01

Pos logit contributions
door+0.890
+0.871
▬▬+0.816
 pulp+0.788
taboola+0.767
Neg logit contributions
atched-1.161
apped-1.145
onduct-1.010
ursed-0.944
umbered-0.941
Token an
Token ID281
Feature activation+9.09e-02

Pos logit contributions
door+1.009
+0.989
▬▬+0.921
 pulp+0.893
taboola+0.867
Neg logit contributions
atched-1.318
apped-1.303
onduct-1.152
umbered-1.074
ursed-1.071
Token the
Token ID262
Feature activation+5.23e-02

Pos logit contributions
atched+0.313
apped+0.311
onduct+0.274
ursed+0.257
umbered+0.256
Neg logit contributions
door-0.213
-0.209
▬▬-0.196
 pulp-0.185
taboola-0.183
Token Sund
Token ID3309
Feature activation-1.47e+00

Pos logit contributions
atched+0.144
apped+0.143
onduct+0.126
ursed+0.118
umbered+0.118
Neg logit contributions
door-0.098
-0.095
▬▬-0.089
 pulp-0.085
taboola-0.083
Tokenarb
Token ID38039
Feature activation-1.62e+00

Pos logit contributions
door+2.175
+1.983
 pulp+1.757
ible+1.754
taboola+1.743
Neg logit contributions
atched-4.488
apped-4.435
onduct-4.060
umbered-3.837
hovah-3.833
Tokenans
Token ID504
Feature activation-4.10e-01

Pos logit contributions
door+2.511
+2.359
ners+2.233
 pulp+2.092
ible+2.057
Neg logit contributions
atched-4.785
apped-4.666
onduct-4.251
hovah-4.143
umbered-4.039
Token in
Token ID287
Feature activation-3.95e-01

Pos logit contributions
door+0.753
+0.737
▬▬+0.685
 pulp+0.658
taboola+0.639
Neg logit contributions
atched-1.139
apped-1.131
onduct-1.003
ursed-0.940
umbered-0.937
Token Bangladesh
Token ID19483
Feature activation-2.28e+00

Pos logit contributions
door+0.713
+0.697
▬▬+0.650
 pulp+0.618
taboola+0.604
Neg logit contributions
atched-1.108
apped-1.102
onduct-0.978
ursed-0.916
umbered-0.916
Token,
Token ID11
Feature activation-5.56e-01

Pos logit contributions
door+4.288
+4.197
▬▬+3.884
 pulp+3.843
taboola+3.688
Neg logit contributions
atched-6.091
apped-6.018
onduct-5.268
ursed-4.959
umbered-4.949
Token told
Token ID1297
Feature activation+8.12e-01

Pos logit contributions
door+1.144
+1.123
▬▬+1.059
 pulp+1.015
taboola+0.995
Neg logit contributions
atched-1.416
apped-1.407
onduct-1.231
ursed-1.146
umbered-1.143
Token reporters
Token ID7638
Feature activation+9.83e-02

Pos logit contributions
atched+2.127
apped+2.114
onduct+1.859
umbered+1.735
ursed+1.731
Neg logit contributions
door-1.599
-1.573
▬▬-1.489
 pulp-1.388
taboola-1.386
Token.
Token ID13
Feature activation-5.09e-01

Pos logit contributions
atched+0.246
apped+0.244
onduct+0.213
ursed+0.198
umbered+0.197
Neg logit contributions
door-0.209
-0.204
▬▬-0.191
 pulp-0.185
taboola-0.180
Token
Token ID198
Feature activation-1.61e-02

Pos logit contributions
door+1.008
+0.992
▬▬+0.937
 pulp+0.889
taboola+0.869
Neg logit contributions
atched-1.337
apped-1.328
onduct-1.168
ursed-1.088
umbered-1.084
Token weather
Token ID6193
Feature activation+7.62e-01

Pos logit contributions
door+0.922
+0.905
▬▬+0.841
 pulp+0.810
taboola+0.783
Neg logit contributions
atched-1.364
apped-1.354
onduct-1.203
ursed-1.126
uked-1.122
Token,
Token ID11
Feature activation-4.98e-01

Pos logit contributions
atched+1.944
apped+1.931
onduct+1.689
ursed+1.569
umbered+1.569
Neg logit contributions
door-1.569
-1.539
▬▬-1.451
 pulp-1.385
taboola-1.363
Token and
Token ID290
Feature activation+7.06e-01

Pos logit contributions
door+0.963
+0.942
▬▬+0.883
 pulp+0.845
taboola+0.825
Neg logit contributions
atched-1.334
apped-1.324
onduct-1.169
ursed-1.093
uked-1.089
Token ever
Token ID1683
Feature activation-4.20e-01

Pos logit contributions
atched+1.883
apped+1.871
onduct+1.649
ursed+1.539
umbered+1.536
Neg logit contributions
door-1.371
-1.344
▬▬-1.263
 pulp-1.198
taboola-1.181
Token-
Token ID12
Feature activation+5.41e-01

Pos logit contributions
door+0.779
+0.761
▬▬+0.713
 pulp+0.674
taboola+0.662
Neg logit contributions
atched-1.161
apped-1.154
onduct-1.023
ursed-0.958
uked-0.955
Tokend
Token ID67
Feature activation-2.95e+00

Pos logit contributions
atched+1.415
apped+1.407
onduct+1.232
umbered+1.151
ursed+1.148
Neg logit contributions
door-1.063
-1.059
▬▬-0.997
 pulp-0.953
taboola-0.925
Tokenwind
Token ID7972
Feature activation-6.68e-01

Pos logit contributions
door+5.815
+5.297
ners+5.115
ible+5.081
 pulp+4.930
Neg logit contributions
atched-7.197
apped-7.047
onduct-6.455
ursed-5.982
ython-5.928
Tokenling
Token ID1359
Feature activation-1.05e-01

Pos logit contributions
door+0.789
+0.760
▬▬+0.673
 pulp+0.611
ners+0.604
Neg logit contributions
atched-2.301
apped-2.282
onduct-2.075
ursed-1.969
uked-1.969
Token resources
Token ID4133
Feature activation+1.17e+00

Pos logit contributions
door+0.206
+0.202
▬▬+0.189
 pulp+0.183
taboola+0.177
Neg logit contributions
atched-0.277
apped-0.274
onduct-0.243
ursed-0.227
uked-0.225
Token.
Token ID13
Feature activation-4.24e-01

Pos logit contributions
atched+2.920
apped+2.903
onduct+2.530
umbered+2.351
ursed+2.350
Neg logit contributions
door-2.464
-2.420
▬▬-2.289
 pulp-2.181
taboola-2.147
Token
Token ID198
Feature activation+4.59e-02

Pos logit contributions
door+0.814
+0.797
▬▬+0.753
 pulp+0.710
taboola+0.699
Neg logit contributions
atched-1.144
apped-1.135
onduct-1.005
ursed-0.940
uked-0.940
Tokenbin
Token ID8800
Feature activation-2.20e+00

Pos logit contributions
atched+2.551
apped+2.533
onduct+2.147
ursed+1.971
umbered+1.967
Neg logit contributions
door-2.972
-2.941
▬▬-2.794
 pulp-2.693
taboola-2.653
Token/
Token ID14
Feature activation+9.32e-01

Pos logit contributions
door+3.506
+3.249
▬▬+3.070
taboola+2.968
 pulp+2.958
Neg logit contributions
apped-6.567
atched-6.560
onduct-5.909
uked-5.548
hovah-5.480
Tokensh
Token ID1477
Feature activation-1.42e+00

Pos logit contributions
atched+2.077
apped+2.054
onduct+1.746
ursed+1.628
uked+1.616
Neg logit contributions
door-2.199
-2.179
▬▬-2.079
 pulp-1.990
taboola-1.954
Token /
Token ID1220
Feature activation+7.19e-01

Pos logit contributions
door+2.760
+2.680
▬▬+2.466
 pulp+2.438
ners+2.305
Neg logit contributions
atched-3.784
apped-3.745
onduct-3.317
ursed-3.118
uked-3.099
Tokens
Token ID82
Feature activation-7.05e-01

Pos logit contributions
atched+1.816
apped+1.805
onduct+1.577
ursed+1.466
umbered+1.463
Neg logit contributions
door-1.499
-1.473
▬▬-1.390
 pulp-1.326
taboola-1.305
Tokenbin
Token ID8800
Feature activation-2.75e+00

Pos logit contributions
door+1.649
+1.614
▬▬+1.522
 pulp+1.475
taboola+1.447
Neg logit contributions
atched-1.600
apped-1.585
onduct-1.377
ursed-1.266
uked-1.262
Token/
Token ID14
Feature activation+4.89e-01

Pos logit contributions
door+4.271
+4.123
▬▬+3.764
 pulp+3.704
taboola+3.672
Neg logit contributions
atched-8.011
apped-7.991
onduct-7.167
hovah-6.821
uked-6.812
Tokeniw
Token ID14246
Feature activation+2.34e+00

Pos logit contributions
atched+1.252
apped+1.244
onduct+1.089
ursed+1.014
umbered+1.012
Neg logit contributions
door-1.004
-0.986
▬▬-0.930
 pulp-0.887
taboola-0.872
Tokenconfig
Token ID11250
Feature activation+2.61e-02

Pos logit contributions
atched+5.375
apped+5.309
onduct+4.524
uked+4.168
ursed+4.167
Neg logit contributions
door-5.317
-5.291
▬▬-5.110
 pulp-4.805
taboola-4.773
Token w
Token ID266
Feature activation+1.72e-01

Pos logit contributions
atched+0.068
apped+0.067
onduct+0.059
ursed+0.055
uked+0.055
Neg logit contributions
door-0.053
-0.052
▬▬-0.048
 pulp-0.047
taboola-0.045
Tokenl
Token ID75
Feature activation+3.73e-01

Pos logit contributions
atched+0.427
apped+0.424
onduct+0.370
ursed+0.343
umbered+0.342
Neg logit contributions
door-0.369
-0.362
▬▬-0.342
 pulp-0.327
taboola-0.322