TOP ACTIVATIONS
MAX = 1.399

repeatedly criticized President Trump Donald John Trump House committee believes it
s " Ent ire classrooms of Somali children -
airport hold about 34 million gallons total , according to Gale
Im more n are made up of a mot ley crew
Democratic presidential nominee Hillary Clinton Hillary Diane Rodham Clinton Sanders :
that the traits / ent itle ments we normally associate with
us / wiki / trans cript s / sc aling bitcoin
elegant it seems out of place . At the menu
the book he learned that Georg es li è s
responded by signing the " Direct ive of Peoples ' Comm
History [ edit ] WE P was ratified
matchup between Donald Trump Donald John Trump House committee believes it
that make US generals sit up and listen . The stakes
History [ edit ] WE P was ratified as
Republican nominee Donald Trump Donald John Trump House committee believes it
us / wiki / trans cript s / sc aling bitcoin
158 ] See also [ edit ] <|endoftext|> I
on a media bus ) pic . twitter . com /
with Republican nominee Donald Trump Donald John Trump House committee believes
prison MORE and Hillary Clinton Hillary Diane Rodham Clinton Sanders :

MIN ACTIVATIONS
MIN = -1.470

conscience or acknowledgment of mis d eeds that destroyed the financial
, and he pointed his finger at me , and the
knee surgery , a broken finger , his manager and the
And don 't worry you will once again
HA V ANA ( Reuters ) - Latin American and Caribbean
tour , plenty of body parts were already set up ,
- level of intellectual rig or . But they surrounded themselves
( the proposal by Andy Stern in Ra ising the Floor
Frank iv sk . <|endoftext|> [ Ex clusive ] Best Buy
, I 'm not looking ahead , I 'm not looking
2015 FR ES NO , Calif . ( K
( Figure 2 A , [ Action Junior 3 , Inv
Y ip H ONG K ONG ( Reuters )
suffered a rupt ured Achilles tendon in Thursday 's game against
Media ite : [ Chuck Todd :]
ONG K ONG ( Reuters ) - One needs to look
re leasing factor ( CR F ) is a hormone secret
finish the season with 10 26 yards at 5 . 4
example with hope , never fear ' Mrs .
, 000 won ($ 8 52 . 54 ) smartphone .

INTERVAL 0.682 - 1.399
CONTAINS 0.526%

Ol af ?" I stood still . My mom
election approaches . For a few years , the economy was
I spent a long time looking for what used to be
The trade also reun ites Fel ton and Thomas with
go into the details of sound commercial banking , how it

INTERVAL -0.035 - 0.682
CONTAINS 55.447%

Center History Office , 1997 . B
ent access to this illegal content .
much as I do , but I think we can win
can use this on your opponent s high Level
have to learn to overtake . The race may only be

INTERVAL -0.753 - -0.035
CONTAINS 43.662%

my first real session with it was dec ided
that self - enc our agement , he managed to control
to schizophrenia , they disagree about exactly how it 's
common sense enough to get around our own space without man
for an equitable future they are the enemy .

INTERVAL -1.470 - -0.753
CONTAINS 0.366%

home - grown Pent ec ost al group , the new
career out of playing un hing ed lun atics . Maybe
agh Kil c oy ne A swath of secondary
." My eyes are still wide open . I wasn 't
who was hit by sh rapnel in World War II ;
Token repeatedly
Token ID7830
Feature activation-8.38e-02

Pos logit contributions
ibaba+0.447
emale+0.435
aimon+0.339
 mustache+0.334
 WRITE+0.334
Neg logit contributions
meg-0.317
star-0.306
-0.301
power-0.297
mania-0.296
Token criticized
Token ID12318
Feature activation-2.32e-01

Pos logit contributions
ibaba+0.362
emale+0.353
aimon+0.274
ijing+0.272
 mustache+0.271
Neg logit contributions
meg-0.249
star-0.239
-0.237
power-0.232
pres-0.232
Token President
Token ID1992
Feature activation-5.99e-02

Pos logit contributions
ibaba+0.978
emale+0.952
aimon+0.735
ijing+0.725
 WRITE+0.718
Neg logit contributions
meg-0.715
star-0.698
-0.683
pres-0.674
power-0.672
Token Trump
Token ID1301
Feature activation-3.70e-02

Pos logit contributions
ibaba+0.232
emale+0.225
ijing+0.167
aimon+0.167
 WRITE+0.164
Neg logit contributions
meg-0.207
star-0.201
-0.199
rovers-0.194
mania-0.194
Token Donald
Token ID3759
Feature activation+8.39e-01

Pos logit contributions
ibaba+0.174
emale+0.170
ijing+0.134
aimon+0.134
 mustache+0.133
Neg logit contributions
meg-0.097
star-0.092
-0.091
pres-0.089
power-0.088
Token John
Token ID1757
Feature activation+1.40e+00

Pos logit contributions
star+1.832
meg+1.830
mania+1.772
power+1.718
pres+1.676
Neg logit contributions
ibaba-4.305
emale-4.178
ijing-3.362
aimon-3.361
irtual-3.282
Token Trump
Token ID1301
Feature activation+8.01e-01

Pos logit contributions
meg+3.701
power+3.684
star+3.627
pres+3.568
rovers+3.513
Neg logit contributions
ibaba-6.499
emale-6.491
aimon-5.071
ijing-4.954
irtual-4.811
TokenHouse
Token ID18102
Feature activation+4.90e-01

Pos logit contributions
meg+1.913
star+1.871
power+1.789
pres+1.778
+1.769
Neg logit contributions
ibaba-3.964
emale-3.886
aimon-3.119
ijing-3.069
 WRITE-3.047
Token committee
Token ID5583
Feature activation+5.80e-02

Pos logit contributions
meg+1.353
star+1.341
power+1.311
mania+1.276
pres+1.265
Neg logit contributions
ibaba-2.252
emale-2.196
aimon-1.743
ijing-1.720
 WRITE-1.671
Token believes
Token ID5804
Feature activation-4.39e-01

Pos logit contributions
meg+0.086
star+0.083
+0.079
mania+0.078
pres+0.077
Neg logit contributions
ibaba-0.337
emale-0.331
aimon-0.275
ijing-0.272
 mustache-0.271
Token it
Token ID340
Feature activation+1.35e-01

Pos logit contributions
ibaba+1.815
emale+1.775
aimon+1.357
ijing+1.331
 WRITE+1.309
Neg logit contributions
meg-1.406
star-1.358
pres-1.315
-1.311
power-1.303
Tokens
Token ID82
Feature activation-7.66e-02

Pos logit contributions
meg+0.320
star+0.306
+0.295
mania+0.289
power+0.289
Neg logit contributions
ibaba-0.881
emale-0.861
aimon-0.706
ijing-0.699
 WRITE-0.693
Token"
Token ID1
Feature activation-1.38e-01

Pos logit contributions
ibaba+0.411
emale+0.401
aimon+0.330
ijing+0.327
 WRITE+0.324
Neg logit contributions
meg-0.149
star-0.142
-0.138
mania-0.135
power-0.134
Token
Token ID198
Feature activation-3.34e-01

Pos logit contributions
ibaba+0.682
emale+0.661
aimon+0.534
ijing+0.527
 WRITE+0.524
Neg logit contributions
meg-0.327
star-0.316
-0.307
mania-0.301
power-0.300
Token
Token ID198
Feature activation-4.73e-01

Pos logit contributions
ibaba+1.672
emale+1.624
aimon+1.325
 WRITE+1.306
ijing+1.302
Neg logit contributions
meg-0.751
star-0.724
-0.702
pres-0.688
rovers-0.687
TokenEnt
Token ID14539
Feature activation+1.35e-01

Pos logit contributions
ibaba+2.182
emale+2.129
aimon+1.707
ijing+1.691
 WRITE+1.668
Neg logit contributions
meg-1.176
star-1.143
-1.119
rovers-1.104
mania-1.092
Tokenire
Token ID557
Feature activation+1.35e+00

Pos logit contributions
meg+0.406
star+0.396
+0.394
mania+0.383
power+0.382
Neg logit contributions
ibaba-0.580
emale-0.563
aimon-0.439
ijing-0.428
 WRITE-0.419
Token classrooms
Token ID32579
Feature activation+2.85e-02

Pos logit contributions
meg+4.088
star+4.047
power+3.895
mania+3.892
pres+3.852
Neg logit contributions
ibaba-5.701
emale-5.536
aimon-4.320
 WRITE-4.218
ijing-4.183
Token of
Token ID286
Feature activation+1.88e-01

Pos logit contributions
meg+0.077
star+0.075
+0.074
pres+0.072
mania+0.072
Neg logit contributions
ibaba-0.130
emale-0.127
aimon-0.100
ijing-0.099
 WRITE-0.098
Token Somali
Token ID30886
Feature activation-5.72e-03

Pos logit contributions
meg+0.587
star+0.573
+0.560
mania+0.554
pres+0.553
Neg logit contributions
ibaba-0.791
emale-0.770
aimon-0.593
ijing-0.583
 WRITE-0.577
Token children
Token ID1751
Feature activation-2.37e-02

Pos logit contributions
ibaba+0.024
emale+0.023
aimon+0.017
ijing+0.017
 mustache+0.017
Neg logit contributions
meg-0.018
star-0.018
-0.017
rovers-0.017
power-0.017
Token -
Token ID532
Feature activation-8.47e-02

Pos logit contributions
ibaba+0.107
emale+0.105
aimon+0.083
ijing+0.082
 mustache+0.081
Neg logit contributions
meg-0.065
star-0.063
-0.062
pres-0.061
rovers-0.060
Token airport
Token ID9003
Feature activation+5.33e-02

Pos logit contributions
meg+0.844
star+0.825
+0.800
power+0.796
mania+0.796
Neg logit contributions
ibaba-1.132
emale-1.104
aimon-0.846
ijing-0.831
 WRITE-0.822
Token hold
Token ID1745
Feature activation+2.49e-01

Pos logit contributions
meg+0.152
star+0.149
+0.145
mania+0.143
power+0.143
Neg logit contributions
ibaba-0.237
emale-0.231
aimon-0.181
ijing-0.178
 WRITE-0.176
Token about
Token ID546
Feature activation+1.14e-01

Pos logit contributions
meg+0.696
star+0.677
+0.652
power+0.651
pres+0.651
Neg logit contributions
ibaba-1.130
emale-1.106
aimon-0.868
ijing-0.853
 WRITE-0.844
Token 34
Token ID4974
Feature activation+3.04e-01

Pos logit contributions
meg+0.361
star+0.353
power+0.342
pres+0.340
mania+0.340
Neg logit contributions
ibaba-0.472
emale-0.463
aimon-0.353
ijing-0.346
 WRITE-0.341
Token million
Token ID1510
Feature activation+6.24e-01

Pos logit contributions
meg+0.789
star+0.767
+0.740
power+0.735
pres+0.734
Neg logit contributions
ibaba-1.441
emale-1.408
aimon-1.119
ijing-1.102
 WRITE-1.092
Token gallons
Token ID23239
Feature activation+1.33e+00

Pos logit contributions
meg+1.716
star+1.667
+1.610
pres+1.601
power+1.600
Neg logit contributions
ibaba-2.856
emale-2.791
aimon-2.194
ijing-2.160
 WRITE-2.138
Token total
Token ID2472
Feature activation+7.16e-01

Pos logit contributions
meg+3.194
star+3.087
power+2.967
+2.954
pres+2.944
Neg logit contributions
ibaba-6.505
emale-6.407
aimon-5.132
ijing-5.027
 WRITE-4.997
Token,
Token ID11
Feature activation+4.45e-01

Pos logit contributions
meg+1.721
star+1.670
+1.604
power+1.594
pres+1.593
Neg logit contributions
ibaba-3.522
emale-3.446
aimon-2.765
ijing-2.725
 WRITE-2.701
Token according
Token ID1864
Feature activation+1.13e-01

Pos logit contributions
meg+1.202
star+1.170
+1.130
power+1.123
pres+1.122
Neg logit contributions
ibaba-2.061
emale-2.012
aimon-1.589
ijing-1.565
 WRITE-1.550
Token to
Token ID284
Feature activation+1.26e-01

Pos logit contributions
meg+0.265
star+0.258
+0.246
power+0.246
pres+0.245
Neg logit contributions
ibaba-0.564
emale-0.555
aimon-0.444
ijing-0.439
 WRITE-0.433
Token Gale
Token ID36483
Feature activation-3.11e-01

Pos logit contributions
meg+0.388
star+0.381
+0.371
pres+0.370
mania+0.368
Neg logit contributions
ibaba-0.536
emale-0.519
aimon-0.399
ijing-0.394
 WRITE-0.391
Token Im
Token ID1846
Feature activation+4.55e-01

Pos logit contributions
meg+0.831
star+0.813
+0.787
power+0.783
pres+0.783
Neg logit contributions
ibaba-1.156
emale-1.127
aimon-0.869
ijing-0.854
 WRITE-0.844
Tokenmore
Token ID3549
Feature activation-1.65e-01

Pos logit contributions
meg+1.258
star+1.226
+1.185
power+1.177
pres+1.177
Neg logit contributions
ibaba-2.073
emale-2.024
aimon-1.591
ijing-1.567
 WRITE-1.552
Tokenn
Token ID77
Feature activation-3.19e-01

Pos logit contributions
ibaba+0.769
emale+0.752
aimon+0.597
ijing+0.587
 WRITE+0.577
Neg logit contributions
meg-0.432
star-0.422
-0.409
pres-0.405
power-0.405
Token are
Token ID389
Feature activation+1.19e-01

Pos logit contributions
ibaba+1.479
emale+1.448
aimon+1.142
ijing+1.128
 WRITE+1.112
Neg logit contributions
meg-0.839
star-0.815
-0.795
mania-0.789
rovers-0.784
Token made
Token ID925
Feature activation+1.30e-01

Pos logit contributions
meg+0.332
star+0.324
+0.314
power+0.311
mania+0.311
Neg logit contributions
ibaba-0.537
emale-0.525
aimon-0.412
ijing-0.405
 WRITE-0.401
Token up
Token ID510
Feature activation+1.33e+00

Pos logit contributions
meg+0.340
star+0.329
+0.324
mania+0.318
power+0.317
Neg logit contributions
ibaba-0.607
emale-0.595
aimon-0.470
ijing-0.467
 WRITE-0.460
Token of
Token ID286
Feature activation+5.65e-01

Pos logit contributions
meg+2.937
star+2.821
+2.694
pres+2.675
power+2.674
Neg logit contributions
ibaba-6.774
emale-6.556
aimon-5.356
ijing-5.270
 WRITE-5.216
Token a
Token ID257
Feature activation+2.57e-02

Pos logit contributions
meg+1.656
star+1.617
+1.561
power+1.555
pres+1.552
Neg logit contributions
ibaba-2.486
emale-2.417
aimon-1.884
ijing-1.853
 WRITE-1.832
Token mot
Token ID2369
Feature activation+4.06e-01

Pos logit contributions
meg+0.073
star+0.071
+0.069
mania+0.069
power+0.068
Neg logit contributions
ibaba-0.115
emale-0.112
aimon-0.088
ijing-0.086
 mustache-0.086
Tokenley
Token ID1636
Feature activation+9.77e-02

Pos logit contributions
meg+1.109
star+1.075
mania+1.073
+1.068
power+1.051
Neg logit contributions
ibaba-1.858
emale-1.808
aimon-1.416
ijing-1.398
 mustache-1.378
Token crew
Token ID5462
Feature activation-5.28e-02

Pos logit contributions
meg+0.279
star+0.271
+0.263
mania+0.261
power+0.261
Neg logit contributions
ibaba-0.437
emale-0.427
aimon-0.334
ijing-0.328
 WRITE-0.325
TokenDemocratic
Token ID33939
Feature activation-4.48e-02

Pos logit contributions
meg+0.352
star+0.342
+0.334
mania+0.329
pres+0.328
Neg logit contributions
ibaba-0.681
emale-0.664
aimon-0.533
ijing-0.525
 WRITE-0.519
Token presidential
Token ID4787
Feature activation-4.72e-01

Pos logit contributions
ibaba+0.209
emale+0.205
aimon+0.162
ijing+0.162
 WRITE+0.161
Neg logit contributions
meg-0.117
star-0.112
-0.110
pres-0.108
power-0.108
Token nominee
Token ID10429
Feature activation-2.48e-01

Pos logit contributions
ibaba+1.827
emale+1.764
aimon+1.314
 margins+1.312
 WRITE+1.310
Neg logit contributions
meg-1.603
-1.547
star-1.544
rovers-1.504
mania-1.500
Token Hillary
Token ID5073
Feature activation+4.61e-01

Pos logit contributions
ibaba+0.866
emale+0.832
 WRITE+0.606
ijing+0.603
aimon+0.600
Neg logit contributions
meg-0.935
star-0.914
-0.908
rovers-0.893
mania-0.891
Token Clinton
Token ID2605
Feature activation-1.90e-01

Pos logit contributions
meg+1.758
star+1.726
+1.683
power+1.676
pres+1.676
Neg logit contributions
ibaba-1.619
emale-1.569
aimon-1.131
ijing-1.106
 WRITE-1.090
Token Hillary
Token ID5073
Feature activation+1.31e+00

Pos logit contributions
ibaba+0.840
emale+0.816
aimon+0.642
ijing+0.641
 WRITE+0.639
Neg logit contributions
meg-0.542
star-0.521
-0.515
power-0.504
rovers-0.504
Token Diane
Token ID26542
Feature activation+8.15e-01

Pos logit contributions
star+2.847
meg+2.836
mania+2.653
pres+2.630
power+2.582
Neg logit contributions
ibaba-6.596
emale-6.396
ijing-5.249
aimon-5.204
 mustache-5.140
Token Rodham
Token ID41848
Feature activation+8.26e-01

Pos logit contributions
star+2.599
meg+2.596
pres+2.514
mania+2.487
+2.469
Neg logit contributions
ibaba-3.340
emale-3.248
ijing-2.472
aimon-2.403
 mustache-2.386
Token Clinton
Token ID2605
Feature activation+6.67e-01

Pos logit contributions
meg+2.925
star+2.913
power+2.823
pres+2.808
mania+2.797
Neg logit contributions
ibaba-3.090
emale-3.057
aimon-2.222
ijing-2.205
 mustache-2.140
TokenSanders
Token ID26747
Feature activation+5.33e-01

Pos logit contributions
meg+1.512
star+1.466
+1.404
power+1.394
pres+1.394
Neg logit contributions
ibaba-3.376
emale-3.303
aimon-2.669
ijing-2.632
 WRITE-2.609
Token:
Token ID25
Feature activation+2.36e-02

Pos logit contributions
meg+1.305
star+1.272
mania+1.226
power+1.214
+1.213
Neg logit contributions
ibaba-2.602
emale-2.548
aimon-2.030
ijing-2.003
 WRITE-1.989
Token that
Token ID326
Feature activation+3.06e-01

Pos logit contributions
ibaba+0.055
emale+0.053
aimon+0.042
ijing+0.042
 WRITE+0.041
Neg logit contributions
meg-0.031
star-0.030
-0.030
power-0.029
mania-0.029
Token the
Token ID262
Feature activation+7.68e-02

Pos logit contributions
meg+0.747
star+0.726
power+0.694
pres+0.693
mania+0.693
Neg logit contributions
ibaba-1.497
emale-1.470
aimon-1.177
ijing-1.156
 WRITE-1.140
Token traits
Token ID12796
Feature activation+2.10e-01

Pos logit contributions
meg+0.247
star+0.242
+0.234
pres+0.234
power+0.233
Neg logit contributions
ibaba-0.316
emale-0.308
aimon-0.235
ijing-0.230
 WRITE-0.227
Token/
Token ID14
Feature activation+1.54e-01

Pos logit contributions
meg+0.547
star+0.533
+0.512
power+0.510
pres+0.510
Neg logit contributions
ibaba-0.992
emale-0.969
aimon-0.770
ijing-0.757
 WRITE-0.750
Tokenent
Token ID298
Feature activation+3.10e-01

Pos logit contributions
meg+0.402
star+0.391
+0.377
power+0.375
pres+0.374
Neg logit contributions
ibaba-0.729
emale-0.712
aimon-0.565
ijing-0.557
 WRITE-0.552
Tokenitle
Token ID2578
Feature activation+1.31e+00

Pos logit contributions
meg+0.877
star+0.856
+0.827
pres+0.823
power+0.823
Neg logit contributions
ibaba-1.392
emale-1.358
aimon-1.064
ijing-1.047
 WRITE-1.037
Tokenments
Token ID902
Feature activation+2.57e-01

Pos logit contributions
meg+1.653
star+1.528
power+1.456
pres+1.438
mania+1.399
Neg logit contributions
ibaba-7.882
emale-7.731
aimon-6.505
ijing-6.401
 WRITE-6.390
Token we
Token ID356
Feature activation-2.22e-01

Pos logit contributions
meg+0.673
star+0.655
+0.632
power+0.628
pres+0.628
Neg logit contributions
ibaba-1.212
emale-1.184
aimon-0.939
ijing-0.925
 WRITE-0.916
Token normally
Token ID7685
Feature activation-3.37e-01

Pos logit contributions
ibaba+0.981
emale+0.955
aimon+0.745
ijing+0.734
 WRITE+0.728
Neg logit contributions
meg-0.643
star-0.626
-0.609
power-0.603
pres-0.602
Token associate
Token ID11602
Feature activation-2.37e-01

Pos logit contributions
ibaba+1.361
emale+1.314
aimon+0.999
ijing+0.987
 WRITE+0.985
Neg logit contributions
meg-1.102
star-1.070
-1.058
power-1.039
pres-1.038
Token with
Token ID351
Feature activation+1.38e-02

Pos logit contributions
ibaba+1.061
emale+1.030
aimon+0.806
ijing+0.798
 WRITE+0.788
Neg logit contributions
meg-0.669
star-0.655
-0.634
pres-0.632
power-0.629
Tokenus
Token ID385
Feature activation-1.44e-01

Pos logit contributions
meg+0.534
star+0.518
+0.508
pres+0.494
mania+0.494
Neg logit contributions
ibaba-1.207
emale-1.175
aimon-0.955
ijing-0.942
 WRITE-0.929
Token/
Token ID14
Feature activation-4.97e-01

Pos logit contributions
ibaba+0.695
emale+0.679
aimon+0.542
ijing+0.537
 WRITE+0.528
Neg logit contributions
meg-0.353
star-0.345
-0.343
pres-0.332
mania-0.330
Tokenwiki
Token ID15466
Feature activation-3.84e-01

Pos logit contributions
ibaba+2.382
emale+2.327
aimon+1.861
ijing+1.840
 WRITE+1.821
Neg logit contributions
meg-1.215
star-1.187
-1.162
mania-1.143
pres-1.135
Token/
Token ID14
Feature activation+4.59e-01

Pos logit contributions
ibaba+1.783
emale+1.748
aimon+1.387
ijing+1.379
 WRITE+1.372
Neg logit contributions
meg-0.979
star-0.948
-0.935
mania-0.917
pres-0.916
Tokentrans
Token ID7645
Feature activation+7.06e-01

Pos logit contributions
meg+1.185
star+1.152
+1.107
power+1.104
pres+1.102
Neg logit contributions
ibaba-2.182
emale-2.132
aimon-1.695
ijing-1.670
 WRITE-1.653
Tokencript
Token ID6519
Feature activation+1.29e+00

Pos logit contributions
meg+2.390
star+2.344
power+2.282
pres+2.270
mania+2.258
Neg logit contributions
ibaba-2.791
emale-2.700
aimon-2.020
ijing-1.994
 WRITE-1.977
Tokens
Token ID82
Feature activation+2.32e-01

Pos logit contributions
meg+3.427
star+3.322
power+3.207
pres+3.180
mania+3.171
Neg logit contributions
ibaba-6.001
emale-5.839
aimon-4.674
ijing-4.556
 WRITE-4.495
Token/
Token ID14
Feature activation+3.04e-01

Pos logit contributions
meg+0.649
star+0.635
+0.616
pres+0.609
mania+0.609
Neg logit contributions
ibaba-1.043
emale-1.019
aimon-0.798
ijing-0.787
 WRITE-0.780
Tokensc
Token ID1416
Feature activation+3.23e-01

Pos logit contributions
meg+0.769
star+0.748
+0.722
mania+0.716
pres+0.715
Neg logit contributions
ibaba-1.456
emale-1.423
aimon-1.134
ijing-1.118
 WRITE-1.108
Tokenaling
Token ID4272
Feature activation+3.57e-01

Pos logit contributions
meg+0.907
star+0.885
+0.858
power+0.850
pres+0.850
Neg logit contributions
ibaba-1.460
emale-1.424
aimon-1.117
ijing-1.100
 WRITE-1.088
Tokenbitcoin
Token ID35395
Feature activation+4.26e-01

Pos logit contributions
meg+0.817
star+0.795
+0.767
mania+0.758
pres+0.756
Neg logit contributions
ibaba-1.792
emale-1.754
aimon-1.412
ijing-1.396
 WRITE-1.384
Token elegant
Token ID19992
Feature activation+4.38e-02

Pos logit contributions
meg+0.075
star+0.073
+0.071
power+0.070
mania+0.070
Neg logit contributions
ibaba-0.108
emale-0.106
aimon-0.082
ijing-0.081
 WRITE-0.080
Token it
Token ID340
Feature activation+4.86e-01

Pos logit contributions
meg+0.114
star+0.111
+0.107
pres+0.106
power+0.106
Neg logit contributions
ibaba-0.206
emale-0.201
aimon-0.160
ijing-0.158
 WRITE-0.156
Token seems
Token ID2331
Feature activation+3.37e-01

Pos logit contributions
meg+1.281
star+1.245
+1.203
power+1.194
pres+1.193
Neg logit contributions
ibaba-2.280
emale-2.226
aimon-1.765
ijing-1.740
 WRITE-1.723
Token out
Token ID503
Feature activation-1.45e-01

Pos logit contributions
meg+0.883
star+0.855
+0.827
mania+0.823
power+0.820
Neg logit contributions
ibaba-1.591
emale-1.551
aimon-1.232
ijing-1.220
 WRITE-1.206
Token of
Token ID286
Feature activation+6.66e-01

Pos logit contributions
ibaba+0.681
emale+0.668
aimon+0.529
ijing+0.523
 WRITE+0.517
Neg logit contributions
meg-0.373
star-0.361
-0.359
mania-0.350
pres-0.346
Token place
Token ID1295
Feature activation+1.25e+00

Pos logit contributions
meg+2.132
star+2.091
+2.019
pres+2.017
power+2.017
Neg logit contributions
ibaba-2.741
emale-2.676
aimon-2.041
ijing-2.002
 WRITE-1.975
Token.
Token ID13
Feature activation-2.43e-01

Pos logit contributions
meg+2.901
star+2.839
power+2.692
pres+2.689
+2.685
Neg logit contributions
ibaba-6.221
emale-6.113
aimon-4.873
ijing-4.796
 WRITE-4.734
Token At
Token ID1629
Feature activation-5.18e-02

Pos logit contributions
ibaba+1.119
emale+1.091
aimon+0.864
ijing+0.850
 WRITE+0.849
Neg logit contributions
meg-0.654
star-0.634
-0.615
power-0.610
pres-0.609
Token the
Token ID262
Feature activation-1.64e-01

Pos logit contributions
ibaba+0.229
emale+0.222
aimon+0.174
ijing+0.172
 WRITE+0.169
Neg logit contributions
meg-0.150
star-0.146
-0.143
power-0.141
pres-0.141
Token menu
Token ID6859
Feature activation-2.20e-01

Pos logit contributions
ibaba+0.664
emale+0.642
aimon+0.492
ijing+0.485
 WRITE+0.480
Neg logit contributions
meg-0.535
star-0.519
-0.514
mania-0.507
power-0.506
Token
Token ID447
Feature activation-1.12e-01

Pos logit contributions
ibaba+1.005
emale+0.971
aimon+0.771
ijing+0.759
 WRITE+0.749
Neg logit contributions
meg-0.606
star-0.582
-0.578
mania-0.567
pres-0.565
Token the
Token ID262
Feature activation-4.25e-02

Pos logit contributions
ibaba+1.948
emale+1.887
aimon+1.467
ijing+1.441
 WRITE+1.438
Neg logit contributions
meg-1.415
star-1.379
-1.357
mania-1.340
pres-1.328
Token book
Token ID1492
Feature activation+2.26e-01

Pos logit contributions
ibaba+0.179
emale+0.174
aimon+0.134
ijing+0.132
 WRITE+0.131
Neg logit contributions
meg-0.132
star-0.128
-0.126
mania-0.124
pres-0.124
Token he
Token ID339
Feature activation+1.79e-02

Pos logit contributions
meg+0.621
star+0.604
+0.587
mania+0.580
pres+0.579
Neg logit contributions
ibaba-1.036
emale-1.011
aimon-0.798
ijing-0.786
 WRITE-0.780
Token learned
Token ID4499
Feature activation-8.27e-02

Pos logit contributions
meg+0.047
star+0.045
+0.044
power+0.043
pres+0.043
Neg logit contributions
ibaba-0.084
emale-0.082
aimon-0.065
ijing-0.064
 WRITE-0.064
Token that
Token ID326
Feature activation+7.71e-02

Pos logit contributions
ibaba+0.376
emale+0.363
aimon+0.290
ijing+0.284
 WRITE+0.283
Neg logit contributions
meg-0.230
star-0.222
-0.216
mania-0.214
pres-0.214
Token Georg
Token ID6850
Feature activation+1.25e+00

Pos logit contributions
meg+0.221
star+0.215
+0.210
mania+0.207
pres+0.206
Neg logit contributions
ibaba-0.343
emale-0.334
aimon-0.262
ijing-0.257
 WRITE-0.257
Tokenes
Token ID274
Feature activation+4.51e-01

Pos logit contributions
meg+3.091
star+2.997
pres+2.874
mania+2.869
power+2.868
Neg logit contributions
ibaba-6.049
emale-5.918
aimon-4.731
ijing-4.664
 WRITE-4.622
Token
Token ID43579
Feature activation-3.21e-01

Pos logit contributions
meg+1.296
star+1.262
+1.228
power+1.214
mania+1.210
Neg logit contributions
ibaba-2.006
emale-1.960
aimon-1.530
ijing-1.509
 WRITE-1.496
Tokenli
Token ID4528
Feature activation+4.53e-01

Pos logit contributions
ibaba+1.383
emale+1.349
aimon+1.040
ijing+1.026
 WRITE+1.013
Neg logit contributions
meg-0.963
star-0.932
-0.923
mania-0.904
pres-0.901
Tokenè
Token ID14064
Feature activation+1.82e-01

Pos logit contributions
meg+1.107
star+1.052
+1.040
mania+1.030
rovers+1.007
Neg logit contributions
ibaba-2.206
emale-2.167
ijing-1.728
aimon-1.719
 WRITE-1.683
Tokens
Token ID82
Feature activation+2.54e-01

Pos logit contributions
meg+0.247
star+0.236
pres+0.216
power+0.214
mania+0.213
Neg logit contributions
ibaba-1.094
emale-1.074
aimon-0.901
ijing-0.889
 WRITE-0.883
Token responded
Token ID7082
Feature activation+3.18e-01

Pos logit contributions
meg+0.165
star+0.159
+0.156
power+0.153
pres+0.153
Neg logit contributions
ibaba-0.280
emale-0.273
aimon-0.216
ijing-0.213
 mustache-0.212
Token by
Token ID416
Feature activation-6.58e-01

Pos logit contributions
meg+0.808
star+0.782
+0.758
mania+0.751
pres+0.746
Neg logit contributions
ibaba-1.516
emale-1.484
aimon-1.188
ijing-1.171
 WRITE-1.155
Token signing
Token ID8415
Feature activation+3.77e-01

Pos logit contributions
ibaba+2.558
emale+2.475
aimon+1.895
 mustache+1.863
ijing+1.851
Neg logit contributions
meg-2.178
star-2.124
-2.107
rovers-2.076
mania-2.062
Token the
Token ID262
Feature activation+2.56e-01

Pos logit contributions
meg+1.137
star+1.109
+1.078
mania+1.069
power+1.068
Neg logit contributions
ibaba-1.626
emale-1.585
aimon-1.228
ijing-1.208
 WRITE-1.199
Token "
Token ID366
Feature activation+4.26e-01

Pos logit contributions
meg+0.772
star+0.754
+0.731
power+0.727
pres+0.727
Neg logit contributions
ibaba-1.101
emale-1.072
aimon-0.830
ijing-0.816
 WRITE-0.807
TokenDirect
Token ID13470
Feature activation+1.25e+00

Pos logit contributions
meg+1.044
star+1.012
+0.977
mania+0.969
power+0.967
Neg logit contributions
ibaba-2.069
emale-2.024
aimon-1.620
ijing-1.597
 WRITE-1.583
Tokenive
Token ID425
Feature activation+3.30e-01

Pos logit contributions
meg+3.719
star+3.673
power+3.582
pres+3.549
sim+3.482
Neg logit contributions
ibaba-5.366
emale-5.204
aimon-4.093
ijing-3.938
 WRITE-3.917
Token of
Token ID286
Feature activation+3.07e-01

Pos logit contributions
meg+0.837
star+0.812
+0.787
mania+0.777
power+0.775
Neg logit contributions
ibaba-1.580
emale-1.546
aimon-1.233
ijing-1.217
 WRITE-1.207
Token Peoples
Token ID41770
Feature activation-2.22e-01

Pos logit contributions
meg+0.918
star+0.896
+0.869
power+0.864
pres+0.863
Neg logit contributions
ibaba-1.332
emale-1.298
aimon-1.006
ijing-0.990
 WRITE-0.980
Token'
Token ID6
Feature activation-4.56e-01

Pos logit contributions
ibaba+1.056
emale+1.027
 WRITE+0.823
aimon+0.820
ijing+0.816
Neg logit contributions
meg-0.545
star-0.523
-0.523
rovers-0.504
mania-0.501
Token Comm
Token ID1520
Feature activation-2.73e-01

Pos logit contributions
ibaba+2.086
emale+2.036
 WRITE+1.609
aimon+1.596
ijing+1.588
Neg logit contributions
meg-1.213
-1.170
star-1.159
rovers-1.126
mania-1.125
Token
Token ID198
Feature activation+4.92e-01

Pos logit contributions
meg+0.928
star+0.901
power+0.855
mania+0.852
pres+0.851
Neg logit contributions
ibaba-2.374
emale-2.336
aimon-1.893
ijing-1.874
 WRITE-1.846
TokenHistory
Token ID18122
Feature activation+9.46e-03

Pos logit contributions
meg+1.195
star+1.161
+1.117
pres+1.106
mania+1.106
Neg logit contributions
ibaba-2.408
emale-2.354
aimon-1.888
ijing-1.861
 WRITE-1.845
Token [
Token ID685
Feature activation+2.72e-01

Pos logit contributions
meg+0.026
star+0.025
+0.025
pres+0.024
mania+0.024
Neg logit contributions
ibaba-0.043
emale-0.042
aimon-0.033
ijing-0.033
 WRITE-0.032
Token edit
Token ID4370
Feature activation+1.74e-02

Pos logit contributions
meg+0.715
star+0.699
power+0.671
pres+0.669
mania+0.667
Neg logit contributions
ibaba-1.285
emale-1.259
aimon-0.994
ijing-0.980
 WRITE-0.971
Token ]
Token ID2361
Feature activation+9.84e-01

Pos logit contributions
meg+0.053
star+0.051
+0.050
power+0.050
pres+0.050
Neg logit contributions
ibaba-0.075
emale-0.073
aimon-0.056
ijing-0.055
 WRITE-0.055
Token
Token ID198
Feature activation+1.23e+00

Pos logit contributions
meg+1.825
star+1.801
power+1.712
mania+1.707
pres+1.686
Neg logit contributions
ibaba-5.400
emale-5.366
aimon-4.334
ijing-4.311
 mustache-4.237
Token
Token ID198
Feature activation+1.20e+00

Pos logit contributions
meg+2.166
star+2.146
power+2.032
mania+2.014
pres+2.000
Neg logit contributions
emale-6.897
ibaba-6.878
ijing-5.533
aimon-5.522
ortunately-5.499
TokenWE
Token ID8845
Feature activation+1.81e-01

Pos logit contributions
meg+2.763
star+2.679
power+2.580
pres+2.572
+2.560
Neg logit contributions
ibaba-6.061
emale-5.920
aimon-4.757
ijing-4.695
 mustache-4.669
TokenP
Token ID47
Feature activation+1.22e-01

Pos logit contributions
meg+0.334
star+0.320
pres+0.305
power+0.305
+0.304
Neg logit contributions
ibaba-0.998
emale-0.981
aimon-0.804
ijing-0.793
 mustache-0.785
Token was
Token ID373
Feature activation+1.69e-01

Pos logit contributions
meg+0.300
star+0.291
+0.282
pres+0.277
power+0.277
Neg logit contributions
ibaba-0.589
emale-0.575
aimon-0.461
ijing-0.455
 WRITE-0.451
Token ratified
Token ID39067
Feature activation+1.02e-01

Pos logit contributions
meg+0.390
star+0.378
+0.364
power+0.360
pres+0.359
Neg logit contributions
ibaba-0.851
emale-0.832
aimon-0.672
ijing-0.663
 WRITE-0.657
Token matchup
Token ID18406
Feature activation+3.78e-02

Pos logit contributions
meg+0.666
star+0.652
+0.632
power+0.629
pres+0.629
Neg logit contributions
ibaba-0.851
emale-0.829
aimon-0.632
ijing-0.621
 WRITE-0.613
Token between
Token ID1022
Feature activation-3.10e-01

Pos logit contributions
meg+0.114
star+0.111
+0.110
power+0.107
rovers+0.106
Neg logit contributions
ibaba-0.162
emale-0.157
aimon-0.121
ijing-0.121
 WRITE-0.120
Token Donald
Token ID3759
Feature activation+1.70e-01

Pos logit contributions
ibaba+1.266
emale+1.235
aimon+0.938
 WRITE+0.930
ijing+0.928
Neg logit contributions
meg-0.989
star-0.960
-0.951
power-0.935
rovers-0.928
Token Trump
Token ID1301
Feature activation-2.11e-02

Pos logit contributions
meg+0.588
star+0.567
+0.566
pres+0.554
power+0.552
Neg logit contributions
ibaba-0.651
emale-0.636
aimon-0.472
ijing-0.468
 WRITE-0.464
Token Donald
Token ID3759
Feature activation+1.03e+00

Pos logit contributions
ibaba+0.097
emale+0.095
ijing+0.075
aimon+0.074
 mustache+0.074
Neg logit contributions
meg-0.056
star-0.054
-0.053
power-0.052
rovers-0.052
Token John
Token ID1757
Feature activation+1.22e+00

Pos logit contributions
meg+2.229
star+2.209
mania+2.138
power+2.055
sim+1.983
Neg logit contributions
ibaba-5.276
emale-5.116
ijing-4.149
aimon-4.148
irtual-4.069
Token Trump
Token ID1301
Feature activation+7.76e-01

Pos logit contributions
meg+3.216
power+3.178
star+3.171
pres+3.090
rovers+3.050
Neg logit contributions
emale-5.665
ibaba-5.652
aimon-4.461
ijing-4.346
irtual-4.204
TokenHouse
Token ID18102
Feature activation+5.45e-01

Pos logit contributions
meg+1.853
star+1.815
power+1.736
pres+1.726
+1.708
Neg logit contributions
ibaba-3.842
emale-3.771
aimon-3.032
ijing-2.982
 WRITE-2.948
Token committee
Token ID5583
Feature activation+1.21e-01

Pos logit contributions
meg+1.520
star+1.508
power+1.469
mania+1.434
pres+1.409
Neg logit contributions
ibaba-2.510
emale-2.421
aimon-1.953
ijing-1.920
irtual-1.841
Token believes
Token ID5804
Feature activation-2.75e-01

Pos logit contributions
meg+0.175
star+0.167
+0.158
mania+0.156
pres+0.154
Neg logit contributions
ibaba-0.706
emale-0.693
aimon-0.578
ijing-0.572
 WRITE-0.569
Token it
Token ID340
Feature activation+2.14e-01

Pos logit contributions
ibaba+1.128
emale+1.096
aimon+0.835
ijing+0.820
 WRITE+0.802
Neg logit contributions
meg-0.907
star-0.865
pres-0.838
rovers-0.832
power-0.832
Token that
Token ID326
Feature activation+3.11e-01

Pos logit contributions
meg+0.784
star+0.762
+0.738
pres+0.732
rovers+0.730
Neg logit contributions
ibaba-1.309
emale-1.273
aimon-1.004
ijing-0.995
 WRITE-0.989
Token make
Token ID787
Feature activation+3.67e-01

Pos logit contributions
meg+0.893
star+0.872
+0.842
power+0.839
pres+0.838
Neg logit contributions
ibaba-1.386
emale-1.351
aimon-1.056
ijing-1.038
 WRITE-1.026
Token US
Token ID1294
Feature activation+1.46e-01

Pos logit contributions
meg+1.078
star+1.052
+1.020
pres+1.012
power+1.012
Neg logit contributions
ibaba-1.606
emale-1.566
aimon-1.217
ijing-1.200
 WRITE-1.189
Token generals
Token ID26770
Feature activation+1.70e-01

Pos logit contributions
meg+0.445
star+0.435
+0.423
mania+0.418
power+0.417
Neg logit contributions
ibaba-0.620
emale-0.605
aimon-0.465
ijing-0.459
 WRITE-0.455
Token sit
Token ID1650
Feature activation+2.16e-01

Pos logit contributions
meg+0.479
star+0.463
+0.453
pres+0.447
power+0.446
Neg logit contributions
ibaba-0.764
emale-0.743
aimon-0.582
ijing-0.579
 mustache-0.575
Token up
Token ID510
Feature activation+1.20e+00

Pos logit contributions
meg+0.590
star+0.570
+0.559
pres+0.549
power+0.547
Neg logit contributions
ibaba-0.992
emale-0.967
aimon-0.761
ijing-0.754
 WRITE-0.744
Token and
Token ID290
Feature activation+2.32e-01

Pos logit contributions
meg+2.769
star+2.697
power+2.619
mania+2.575
+2.572
Neg logit contributions
ibaba-5.982
emale-5.896
aimon-4.736
ijing-4.666
 WRITE-4.574
Token listen
Token ID6004
Feature activation+3.31e-01

Pos logit contributions
meg+0.639
star+0.622
+0.603
mania+0.597
pres+0.597
Neg logit contributions
ibaba-1.060
emale-1.034
aimon-0.813
ijing-0.801
 WRITE-0.799
Token.
Token ID13
Feature activation+7.51e-02

Pos logit contributions
meg+0.768
star+0.742
+0.711
mania+0.706
pres+0.703
Neg logit contributions
ibaba-1.652
emale-1.612
aimon-1.304
ijing-1.293
 WRITE-1.283
Token The
Token ID383
Feature activation+2.02e-01

Pos logit contributions
meg+0.200
star+0.194
+0.188
power+0.186
pres+0.186
Neg logit contributions
ibaba-0.350
emale-0.342
aimon-0.270
ijing-0.266
 WRITE-0.265
Token stakes
Token ID21147
Feature activation-1.01e-01

Pos logit contributions
meg+0.616
star+0.602
+0.583
power+0.580
pres+0.580
Neg logit contributions
ibaba-0.863
emale-0.841
aimon-0.649
ijing-0.638
 WRITE-0.631
TokenHistory
Token ID18122
Feature activation+9.46e-03

Pos logit contributions
meg+1.195
star+1.161
+1.117
pres+1.106
mania+1.106
Neg logit contributions
ibaba-2.408
emale-2.354
aimon-1.888
ijing-1.861
 WRITE-1.845
Token [
Token ID685
Feature activation+2.72e-01

Pos logit contributions
meg+0.026
star+0.025
+0.025
pres+0.024
mania+0.024
Neg logit contributions
ibaba-0.043
emale-0.042
aimon-0.033
ijing-0.033
 WRITE-0.032
Token edit
Token ID4370
Feature activation+1.74e-02

Pos logit contributions
meg+0.715
star+0.699
power+0.671
pres+0.669
mania+0.667
Neg logit contributions
ibaba-1.285
emale-1.259
aimon-0.994
ijing-0.980
 WRITE-0.971
Token ]
Token ID2361
Feature activation+9.84e-01

Pos logit contributions
meg+0.053
star+0.051
+0.050
power+0.050
pres+0.050
Neg logit contributions
ibaba-0.075
emale-0.073
aimon-0.056
ijing-0.055
 WRITE-0.055
Token
Token ID198
Feature activation+1.23e+00

Pos logit contributions
meg+1.825
star+1.801
power+1.712
mania+1.707
pres+1.686
Neg logit contributions
ibaba-5.400
emale-5.366
aimon-4.334
ijing-4.311
 mustache-4.237
Token
Token ID198
Feature activation+1.20e+00

Pos logit contributions
meg+2.166
star+2.146
power+2.032
mania+2.014
pres+2.000
Neg logit contributions
emale-6.897
ibaba-6.878
ijing-5.533
aimon-5.522
ortunately-5.499
TokenWE
Token ID8845
Feature activation+1.81e-01

Pos logit contributions
meg+2.763
star+2.679
power+2.580
pres+2.572
+2.560
Neg logit contributions
ibaba-6.061
emale-5.920
aimon-4.757
ijing-4.695
 mustache-4.669
TokenP
Token ID47
Feature activation+1.22e-01

Pos logit contributions
meg+0.334
star+0.320
pres+0.305
power+0.305
+0.304
Neg logit contributions
ibaba-0.998
emale-0.981
aimon-0.804
ijing-0.793
 mustache-0.785
Token was
Token ID373
Feature activation+1.69e-01

Pos logit contributions
meg+0.300
star+0.291
+0.282
pres+0.277
power+0.277
Neg logit contributions
ibaba-0.589
emale-0.575
aimon-0.461
ijing-0.455
 WRITE-0.451
Token ratified
Token ID39067
Feature activation+1.02e-01

Pos logit contributions
meg+0.390
star+0.378
+0.364
power+0.360
pres+0.359
Neg logit contributions
ibaba-0.851
emale-0.832
aimon-0.672
ijing-0.663
 WRITE-0.657
Token as
Token ID355
Feature activation+1.57e-01

Pos logit contributions
meg+0.261
star+0.253
+0.245
power+0.241
pres+0.241
Neg logit contributions
ibaba-0.484
emale-0.475
aimon-0.379
ijing-0.374
 WRITE-0.371
Token Republican
Token ID3415
Feature activation+5.34e-02

Pos logit contributions
meg+0.326
star+0.319
+0.312
power+0.309
mania+0.308
Neg logit contributions
ibaba-0.440
emale-0.429
aimon-0.329
ijing-0.325
 WRITE-0.321
Token nominee
Token ID10429
Feature activation-7.67e-02

Pos logit contributions
meg+0.164
star+0.159
+0.158
power+0.154
pres+0.154
Neg logit contributions
ibaba-0.224
emale-0.219
aimon-0.168
ijing-0.167
 WRITE-0.167
Token Donald
Token ID3759
Feature activation-3.31e-02

Pos logit contributions
ibaba+0.309
emale+0.301
aimon+0.229
 WRITE+0.228
ijing+0.228
Neg logit contributions
meg-0.247
-0.240
star-0.239
rovers-0.232
power-0.232
Token Trump
Token ID1301
Feature activation-3.38e-04

Pos logit contributions
ibaba+0.135
emale+0.131
aimon+0.100
ijing+0.098
 WRITE+0.097
Neg logit contributions
meg-0.108
star-0.106
-0.103
power-0.102
pres-0.102
Token Donald
Token ID3759
Feature activation+1.19e+00

Pos logit contributions
ibaba+0.001
emale+0.001
aimon+0.001
ijing+0.001
 WRITE+0.001
Neg logit contributions
meg-0.001
star-0.001
-0.001
power-0.001
pres-0.001
Token John
Token ID1757
Feature activation+1.20e+00

Pos logit contributions
star+2.548
meg+2.536
mania+2.431
power+2.359
sim+2.263
Neg logit contributions
ibaba-6.152
emale-5.954
aimon-4.837
ijing-4.832
irtual-4.755
Token Trump
Token ID1301
Feature activation+9.80e-01

Pos logit contributions
meg+3.213
power+3.188
star+3.168
pres+3.096
rovers+3.056
Neg logit contributions
emale-5.561
ibaba-5.522
aimon-4.322
ijing-4.242
irtual-4.104
TokenHouse
Token ID18102
Feature activation+5.59e-01

Pos logit contributions
meg+2.330
star+2.286
power+2.192
pres+2.173
+2.146
Neg logit contributions
ibaba-4.858
emale-4.762
aimon-3.827
ijing-3.766
 WRITE-3.728
Token committee
Token ID5583
Feature activation+8.39e-02

Pos logit contributions
meg+1.555
star+1.549
power+1.514
mania+1.465
pres+1.450
Neg logit contributions
ibaba-2.574
emale-2.487
aimon-2.002
ijing-1.964
irtual-1.888
Token believes
Token ID5804
Feature activation-3.60e-01

Pos logit contributions
meg+0.122
star+0.116
+0.110
mania+0.108
pres+0.107
Neg logit contributions
ibaba-0.491
emale-0.483
aimon-0.402
ijing-0.398
 WRITE-0.396
Token it
Token ID340
Feature activation+1.58e-01

Pos logit contributions
ibaba+1.441
emale+1.406
aimon+1.064
ijing+1.042
 WRITE+1.018
Neg logit contributions
meg-1.217
star-1.168
pres-1.134
-1.134
power-1.127
Tokenus
Token ID385
Feature activation-3.15e-02

Pos logit contributions
meg+0.613
star+0.587
power+0.546
pres+0.545
+0.544
Neg logit contributions
ibaba-2.226
emale-2.189
aimon-1.814
ijing-1.794
 WRITE-1.784
Token/
Token ID14
Feature activation-3.30e-01

Pos logit contributions
ibaba+0.153
emale+0.150
aimon+0.119
ijing+0.118
 WRITE+0.117
Neg logit contributions
meg-0.077
star-0.075
-0.073
pres-0.072
mania-0.072
Tokenwiki
Token ID15466
Feature activation-2.70e-01

Pos logit contributions
ibaba+1.494
emale+1.452
aimon+1.149
ijing+1.132
 WRITE+1.123
Neg logit contributions
meg-0.893
star-0.871
-0.855
mania-0.843
rovers-0.840
Token/
Token ID14
Feature activation+5.37e-01

Pos logit contributions
ibaba+1.320
emale+1.285
aimon+1.037
ijing+1.021
 WRITE+1.016
Neg logit contributions
meg-0.642
star-0.620
-0.605
mania-0.593
rovers-0.592
Tokentrans
Token ID7645
Feature activation+5.32e-01

Pos logit contributions
meg+1.346
star+1.308
+1.262
mania+1.251
pres+1.250
Neg logit contributions
ibaba-2.589
emale-2.530
aimon-2.020
ijing-1.991
 WRITE-1.973
Tokencript
Token ID6519
Feature activation+1.20e+00

Pos logit contributions
meg+1.328
star+1.311
pres+1.261
power+1.253
sim+1.235
Neg logit contributions
ibaba-2.548
emale-2.515
aimon-1.990
 WRITE-1.964
ijing-1.960
Tokens
Token ID82
Feature activation+2.22e-01

Pos logit contributions
meg+1.234
star+1.184
pres+1.082
power+1.062
mania+1.035
Neg logit contributions
ibaba-7.514
emale-7.499
aimon-6.294
ijing-6.237
 mustache-6.185
Token/
Token ID14
Feature activation+4.32e-01

Pos logit contributions
meg+0.543
star+0.528
+0.510
power+0.505
mania+0.504
Neg logit contributions
ibaba-1.077
emale-1.054
aimon-0.843
ijing-0.833
 WRITE-0.828
Tokensc
Token ID1416
Feature activation+3.22e-01

Pos logit contributions
meg+0.921
star+0.889
+0.862
mania+0.852
pres+0.843
Neg logit contributions
ibaba-2.230
emale-2.184
aimon-1.776
ijing-1.751
 WRITE-1.736
Tokenaling
Token ID4272
Feature activation+3.40e-01

Pos logit contributions
meg+0.933
star+0.897
power+0.879
pres+0.869
mania+0.868
Neg logit contributions
ibaba-1.428
emale-1.401
aimon-1.084
 mustache-1.076
 WRITE-1.071
Tokenbitcoin
Token ID35395
Feature activation+5.37e-01

Pos logit contributions
meg+1.145
star+1.125
+1.108
pres+1.090
mania+1.088
Neg logit contributions
ibaba-1.341
emale-1.299
aimon-0.976
ijing-0.965
 WRITE-0.937
Token158
Token ID21273
Feature activation+4.36e-01

Pos logit contributions
meg+0.171
star+0.163
+0.159
mania+0.159
power+0.158
Neg logit contributions
ibaba-0.296
emale-0.293
aimon-0.229
ijing-0.227
 WRITE-0.225
Token]
Token ID60
Feature activation+5.63e-01

Pos logit contributions
meg+1.266
star+1.234
+1.194
mania+1.186
pres+1.185
Neg logit contributions
ibaba-1.919
emale-1.885
aimon-1.456
ijing-1.439
 WRITE-1.426
Token
Token ID198
Feature activation+6.06e-01

Pos logit contributions
meg+1.426
star+1.390
+1.334
mania+1.329
power+1.329
Neg logit contributions
ibaba-2.697
emale-2.634
aimon-2.099
ijing-2.070
 WRITE-2.045
Token
Token ID198
Feature activation+6.43e-01

Pos logit contributions
meg+1.205
star+1.177
power+1.115
mania+1.109
pres+1.107
Neg logit contributions
ibaba-3.261
emale-3.223
aimon-2.603
ijing-2.587
 mustache-2.537
TokenSee
Token ID6214
Feature activation+8.30e-01

Pos logit contributions
meg+1.476
star+1.430
+1.374
power+1.361
pres+1.361
Neg logit contributions
ibaba-3.234
emale-3.164
aimon-2.553
ijing-2.518
 WRITE-2.497
Token also
Token ID635
Feature activation+1.20e+00

Pos logit contributions
meg+1.910
star+1.867
power+1.771
mania+1.768
+1.759
Neg logit contributions
ibaba-4.155
emale-4.085
aimon-3.275
ijing-3.239
 WRITE-3.194
Token [
Token ID685
Feature activation+3.37e-01

Pos logit contributions
meg+3.221
star+3.150
mania+3.027
+3.022
power+3.013
Neg logit contributions
ibaba-5.502
emale-5.392
aimon-4.263
ijing-4.186
 WRITE-4.132
Token edit
Token ID4370
Feature activation+8.95e-02

Pos logit contributions
meg+0.875
star+0.860
power+0.821
pres+0.820
mania+0.817
Neg logit contributions
ibaba-1.601
emale-1.572
aimon-1.240
ijing-1.223
 WRITE-1.209
Token ]
Token ID2361
Feature activation+1.01e+00

Pos logit contributions
meg+0.273
star+0.266
+0.259
power+0.256
pres+0.256
Neg logit contributions
ibaba-0.382
emale-0.372
aimon-0.288
ijing-0.283
 WRITE-0.280
Token<|endoftext|>
Token ID50256
Feature activation+2.04e-01

Pos logit contributions
meg+2.338
star+2.270
+2.176
power+2.162
mania+2.160
Neg logit contributions
ibaba-5.024
emale-4.915
aimon-3.960
ijing-3.905
 WRITE-3.870
TokenI
Token ID40
Feature activation+2.95e-02

Pos logit contributions
meg+0.506
star+0.491
+0.480
mania+0.473
pres+0.471
Neg logit contributions
ibaba-0.976
emale-0.951
aimon-0.763
ijing-0.752
 WRITE-0.744
Token on
Token ID319
Feature activation+1.25e-01

Pos logit contributions
meg+0.812
star+0.793
+0.762
mania+0.761
power+0.760
Neg logit contributions
ibaba-1.294
emale-1.263
aimon-0.988
ijing-0.974
 WRITE-0.963
Token a
Token ID257
Feature activation+4.41e-02

Pos logit contributions
meg+0.394
star+0.385
+0.373
power+0.372
pres+0.371
Neg logit contributions
ibaba-0.522
emale-0.508
aimon-0.389
ijing-0.383
 WRITE-0.379
Token media
Token ID2056
Feature activation-2.59e-02

Pos logit contributions
meg+0.148
star+0.144
+0.141
mania+0.140
pres+0.140
Neg logit contributions
ibaba-0.175
emale-0.170
aimon-0.128
ijing-0.126
 WRITE-0.125
Token bus
Token ID1323
Feature activation+1.38e-01

Pos logit contributions
ibaba+0.103
emale+0.100
aimon+0.075
ijing+0.074
 WRITE+0.073
Neg logit contributions
meg-0.087
star-0.085
-0.083
power-0.082
mania-0.082
Token)
Token ID8
Feature activation+2.55e-01

Pos logit contributions
meg+0.409
star+0.399
+0.386
power+0.384
pres+0.384
Neg logit contributions
ibaba-0.601
emale-0.586
aimon-0.455
ijing-0.448
 WRITE-0.443
Token pic
Token ID8301
Feature activation+1.19e+00

Pos logit contributions
meg+0.623
star+0.603
+0.584
power+0.576
mania+0.576
Neg logit contributions
ibaba-1.245
emale-1.218
aimon-0.975
ijing-0.962
 WRITE-0.956
Token.
Token ID13
Feature activation+3.65e-01

Pos logit contributions
meg+1.711
star+1.704
mania+1.538
power+1.536
pres+1.529
Neg logit contributions
ibaba-7.112
emale-7.075
aimon-5.780
ijing-5.776
 mustache-5.660
Tokentwitter
Token ID6956
Feature activation+4.46e-01

Pos logit contributions
meg+0.689
star+0.664
+0.630
power+0.625
pres+0.624
Neg logit contributions
ibaba-1.988
emale-1.949
aimon-1.600
ijing-1.582
 WRITE-1.568
Token.
Token ID13
Feature activation+1.74e-01

Pos logit contributions
meg+0.736
star+0.711
power+0.662
mania+0.660
+0.660
Neg logit contributions
ibaba-2.542
emale-2.502
aimon-2.064
ijing-2.046
 WRITE-2.019
Tokencom
Token ID785
Feature activation-3.07e-01

Pos logit contributions
meg+0.356
star+0.344
+0.327
power+0.325
pres+0.325
Neg logit contributions
ibaba-0.918
emale-0.901
aimon-0.733
ijing-0.725
 WRITE-0.718
Token/
Token ID14
Feature activation-1.62e-01

Pos logit contributions
ibaba+1.509
emale+1.469
aimon+1.186
ijing+1.167
 WRITE+1.165
Neg logit contributions
meg-0.731
star-0.705
-0.683
pres-0.674
mania-0.674
Token with
Token ID351
Feature activation+1.05e-01

Pos logit contributions
ibaba+0.507
emale+0.493
aimon+0.389
ijing+0.387
 WRITE+0.382
Neg logit contributions
meg-0.307
star-0.297
-0.293
power-0.287
pres-0.285
Token Republican
Token ID3415
Feature activation+5.34e-02

Pos logit contributions
meg+0.326
star+0.319
+0.312
power+0.309
mania+0.308
Neg logit contributions
ibaba-0.440
emale-0.429
aimon-0.329
ijing-0.325
 WRITE-0.321
Token nominee
Token ID10429
Feature activation-7.67e-02

Pos logit contributions
meg+0.164
star+0.159
+0.158
power+0.154
pres+0.154
Neg logit contributions
ibaba-0.224
emale-0.219
aimon-0.168
ijing-0.167
 WRITE-0.167
Token Donald
Token ID3759
Feature activation-3.31e-02

Pos logit contributions
ibaba+0.309
emale+0.301
aimon+0.229
 WRITE+0.228
ijing+0.228
Neg logit contributions
meg-0.247
-0.240
star-0.239
rovers-0.232
power-0.232
Token Trump
Token ID1301
Feature activation-3.38e-04

Pos logit contributions
ibaba+0.135
emale+0.131
aimon+0.100
ijing+0.098
 WRITE+0.097
Neg logit contributions
meg-0.108
star-0.106
-0.103
power-0.102
pres-0.102
Token Donald
Token ID3759
Feature activation+1.19e+00

Pos logit contributions
ibaba+0.001
emale+0.001
aimon+0.001
ijing+0.001
 WRITE+0.001
Neg logit contributions
meg-0.001
star-0.001
-0.001
power-0.001
pres-0.001
Token John
Token ID1757
Feature activation+1.20e+00

Pos logit contributions
star+2.548
meg+2.536
mania+2.431
power+2.359
sim+2.263
Neg logit contributions
ibaba-6.152
emale-5.954
aimon-4.837
ijing-4.832
irtual-4.755
Token Trump
Token ID1301
Feature activation+9.80e-01

Pos logit contributions
meg+3.213
power+3.188
star+3.168
pres+3.096
rovers+3.056
Neg logit contributions
emale-5.561
ibaba-5.522
aimon-4.322
ijing-4.242
irtual-4.104
TokenHouse
Token ID18102
Feature activation+5.59e-01

Pos logit contributions
meg+2.330
star+2.286
power+2.192
pres+2.173
+2.146
Neg logit contributions
ibaba-4.858
emale-4.762
aimon-3.827
ijing-3.766
 WRITE-3.728
Token committee
Token ID5583
Feature activation+8.39e-02

Pos logit contributions
meg+1.555
star+1.549
power+1.514
mania+1.465
pres+1.450
Neg logit contributions
ibaba-2.574
emale-2.487
aimon-2.002
ijing-1.964
irtual-1.888
Token believes
Token ID5804
Feature activation-3.60e-01

Pos logit contributions
meg+0.122
star+0.116
+0.110
mania+0.108
pres+0.107
Neg logit contributions
ibaba-0.491
emale-0.483
aimon-0.402
ijing-0.398
 WRITE-0.396
Token prison
Token ID3770
Feature activation+5.08e-01

Pos logit contributions
meg+0.786
star+0.774
pres+0.747
power+0.746
mania+0.745
Neg logit contributions
ibaba-0.933
emale-0.913
aimon-0.688
ijing-0.668
 WRITE-0.664
Token MORE
Token ID12011
Feature activation+3.42e-01

Pos logit contributions
meg+1.562
star+1.519
power+1.477
pres+1.476
+1.469
Neg logit contributions
ibaba-2.171
emale-2.124
aimon-1.641
ijing-1.610
 mustache-1.572
Token and
Token ID290
Feature activation+3.49e-01

Pos logit contributions
meg+0.877
star+0.844
+0.821
pres+0.806
power+0.805
Neg logit contributions
ibaba-1.633
emale-1.593
aimon-1.261
ijing-1.255
 WRITE-1.244
Token Hillary
Token ID5073
Feature activation+3.92e-01

Pos logit contributions
meg+1.174
star+1.151
+1.117
pres+1.113
power+1.112
Neg logit contributions
ibaba-1.379
emale-1.341
aimon-1.010
ijing-0.990
 WRITE-0.977
Token Clinton
Token ID2605
Feature activation+1.88e-03

Pos logit contributions
meg+1.442
star+1.424
mania+1.385
pres+1.384
power+1.381
Neg logit contributions
ibaba-1.425
emale-1.392
aimon-1.017
ijing-0.987
 WRITE-0.969
Token Hillary
Token ID5073
Feature activation+1.19e+00

Pos logit contributions
meg+0.007
star+0.007
+0.007
power+0.007
pres+0.007
Neg logit contributions
ibaba-0.007
emale-0.006
aimon-0.005
 WRITE-0.005
ijing-0.005
Token Diane
Token ID26542
Feature activation+8.40e-01

Pos logit contributions
star+2.565
meg+2.554
mania+2.390
power+2.367
pres+2.347
Neg logit contributions
ibaba-6.088
emale-5.869
ijing-4.812
aimon-4.739
 mustache-4.700
Token Rodham
Token ID41848
Feature activation+8.00e-01

Pos logit contributions
star+2.662
meg+2.654
pres+2.579
mania+2.555
+2.547
Neg logit contributions
ibaba-3.473
emale-3.375
ijing-2.544
aimon-2.492
 mustache-2.468
Token Clinton
Token ID2605
Feature activation+7.09e-01

Pos logit contributions
meg+2.797
star+2.780
power+2.704
pres+2.691
mania+2.677
Neg logit contributions
ibaba-3.020
emale-3.012
aimon-2.176
ijing-2.176
 mustache-2.106
TokenSanders
Token ID26747
Feature activation+5.07e-01

Pos logit contributions
meg+1.548
star+1.503
+1.431
pres+1.426
power+1.425
Neg logit contributions
ibaba-3.646
emale-3.569
aimon-2.895
ijing-2.856
 WRITE-2.831
Token:
Token ID25
Feature activation-2.86e-03

Pos logit contributions
meg+1.234
star+1.201
mania+1.162
power+1.149
+1.142
Neg logit contributions
ibaba-2.482
emale-2.431
aimon-1.937
ijing-1.913
 WRITE-1.898
Token conscience
Token ID18346
Feature activation-4.62e-02

Pos logit contributions
meg+0.460
star+0.449
+0.439
mania+0.433
power+0.433
Neg logit contributions
ibaba-0.611
emale-0.591
aimon-0.457
ijing-0.447
 WRITE-0.445
Token or
Token ID393
Feature activation-6.89e-02

Pos logit contributions
ibaba+0.224
emale+0.216
aimon+0.177
 mustache+0.175
ijing+0.175
Neg logit contributions
meg-0.111
star-0.107
-0.106
rovers-0.103
mania-0.102
Token acknowledgment
Token ID48182
Feature activation-2.16e-01

Pos logit contributions
ibaba+0.285
emale+0.274
aimon+0.213
ijing+0.208
 mustache+0.207
Neg logit contributions
meg-0.219
star-0.213
-0.210
rovers-0.206
mania-0.206
Token of
Token ID286
Feature activation+1.61e-01

Pos logit contributions
ibaba+0.992
emale+0.949
aimon+0.767
ijing+0.759
 WRITE+0.754
Neg logit contributions
meg-0.585
star-0.565
-0.552
rovers-0.541
mania-0.540
Token mis
Token ID2984
Feature activation+4.80e-01

Pos logit contributions
meg+0.481
star+0.469
+0.458
mania+0.452
pres+0.451
Neg logit contributions
ibaba-0.699
emale-0.678
aimon-0.530
ijing-0.520
 WRITE-0.516
Tokend
Token ID67
Feature activation-1.47e+00

Pos logit contributions
meg+1.235
star+1.202
pres+1.154
+1.153
power+1.153
Neg logit contributions
ibaba-2.279
emale-2.227
aimon-1.768
ijing-1.746
 WRITE-1.730
Tokeneeds
Token ID39642
Feature activation-4.13e-01

Pos logit contributions
ibaba+3.956
emale+3.780
aimon+2.614
ijing+2.216
 WRITE+2.122
Neg logit contributions
meg-6.644
-6.606
star-6.574
pres-6.406
mania-6.373
Token that
Token ID326
Feature activation+1.92e-02

Pos logit contributions
ibaba+1.927
emale+1.852
aimon+1.493
ijing+1.492
 mustache+1.480
Neg logit contributions
meg-1.065
star-1.028
-1.017
rovers-0.996
mania-0.979
Token destroyed
Token ID6572
Feature activation-2.15e-01

Pos logit contributions
meg+0.050
star+0.049
+0.047
mania+0.047
pres+0.046
Neg logit contributions
ibaba-0.090
emale-0.088
aimon-0.070
ijing-0.069
 WRITE-0.069
Token the
Token ID262
Feature activation-1.02e-01

Pos logit contributions
ibaba+0.916
emale+0.888
aimon+0.686
ijing+0.675
 WRITE+0.675
Neg logit contributions
meg-0.653
star-0.637
-0.625
mania-0.616
pres-0.615
Token financial
Token ID3176
Feature activation+5.32e-02

Pos logit contributions
ibaba+0.412
emale+0.401
aimon+0.303
 mustache+0.299
ijing+0.299
Neg logit contributions
meg-0.330
star-0.321
-0.315
mania-0.312
pres-0.311
Token,
Token ID11
Feature activation+8.74e-02

Pos logit contributions
ibaba+0.170
emale+0.165
aimon+0.132
ijing+0.130
 WRITE+0.129
Neg logit contributions
meg-0.090
star-0.087
-0.085
mania-0.084
pres-0.084
Token and
Token ID290
Feature activation+1.31e-01

Pos logit contributions
meg+0.237
star+0.230
+0.224
mania+0.221
power+0.221
Neg logit contributions
ibaba-0.403
emale-0.391
aimon-0.310
ijing-0.305
 WRITE-0.303
Token he
Token ID339
Feature activation-6.29e-02

Pos logit contributions
meg+0.342
star+0.332
+0.321
power+0.318
pres+0.318
Neg logit contributions
ibaba-0.615
emale-0.599
aimon-0.477
ijing-0.469
 WRITE-0.466
Token pointed
Token ID6235
Feature activation+8.50e-02

Pos logit contributions
ibaba+0.297
emale+0.288
aimon+0.230
ijing+0.227
 WRITE+0.226
Neg logit contributions
meg-0.164
star-0.158
-0.154
power-0.152
pres-0.152
Token his
Token ID465
Feature activation-5.21e-01

Pos logit contributions
meg+0.233
star+0.226
+0.219
mania+0.218
power+0.217
Neg logit contributions
ibaba-0.387
emale-0.377
aimon-0.298
ijing-0.294
 WRITE-0.292
Token finger
Token ID7660
Feature activation-1.43e+00

Pos logit contributions
ibaba+1.919
emale+1.873
aimon+1.397
 mustache+1.382
ijing+1.355
Neg logit contributions
meg-1.843
star-1.802
-1.762
mania-1.761
rovers-1.754
Token at
Token ID379
Feature activation-1.61e-01

Pos logit contributions
ibaba+6.068
emale+5.851
 mustache+4.730
aimon+4.712
ijing+4.704
Neg logit contributions
meg-3.869
star-3.771
rovers-3.713
-3.707
mania-3.632
Token me
Token ID502
Feature activation+2.02e-01

Pos logit contributions
ibaba+0.688
emale+0.666
aimon+0.519
ijing+0.509
 WRITE+0.508
Neg logit contributions
meg-0.485
star-0.472
-0.462
mania-0.457
pres-0.456
Token,
Token ID11
Feature activation-1.28e-01

Pos logit contributions
meg+0.494
star+0.478
+0.463
mania+0.460
power+0.457
Neg logit contributions
ibaba-0.984
emale-0.962
aimon-0.772
ijing-0.762
 WRITE-0.754
Token and
Token ID290
Feature activation-1.02e-01

Pos logit contributions
ibaba+0.563
emale+0.547
aimon+0.428
ijing+0.421
 mustache+0.420
Neg logit contributions
meg-0.368
star-0.358
-0.349
mania-0.346
power-0.344
Token the
Token ID262
Feature activation-7.48e-02

Pos logit contributions
ibaba+0.465
emale+0.452
aimon+0.357
 WRITE+0.351
ijing+0.350
Neg logit contributions
meg-0.281
star-0.273
-0.266
mania-0.263
power-0.263
Token knee
Token ID10329
Feature activation-3.06e-01

Pos logit contributions
ibaba+0.095
emale+0.093
aimon+0.073
ijing+0.072
 WRITE+0.072
Neg logit contributions
meg-0.059
star-0.057
-0.056
power-0.055
mania-0.055
Token surgery
Token ID8185
Feature activation-6.84e-01

Pos logit contributions
ibaba+1.216
emale+1.169
aimon+0.889
 mustache+0.878
ijing+0.872
Neg logit contributions
meg-1.018
star-0.988
-0.976
pres-0.961
power-0.956
Token,
Token ID11
Feature activation-3.32e-02

Pos logit contributions
ibaba+2.997
emale+2.904
aimon+2.296
 mustache+2.291
 margins+2.285
Neg logit contributions
meg-1.886
-1.843
star-1.818
power-1.752
pres-1.750
Token a
Token ID257
Feature activation-9.23e-02

Pos logit contributions
ibaba+0.138
emale+0.134
aimon+0.103
 mustache+0.101
ijing+0.101
Neg logit contributions
meg-0.105
star-0.102
-0.100
pres-0.098
power-0.098
Token broken
Token ID5445
Feature activation-7.86e-03

Pos logit contributions
ibaba+0.375
emale+0.364
aimon+0.278
ijing+0.272
 WRITE+0.270
Neg logit contributions
meg-0.301
star-0.293
-0.287
power-0.284
pres-0.284
Token finger
Token ID7660
Feature activation-1.43e+00

Pos logit contributions
ibaba+0.027
emale+0.026
aimon+0.019
ijing+0.018
 WRITE+0.018
Neg logit contributions
meg-0.030
star-0.030
-0.029
power-0.029
mania-0.029
Token,
Token ID11
Feature activation-1.61e-02

Pos logit contributions
ibaba+6.024
emale+5.825
 mustache+4.690
aimon+4.666
 buckle+4.570
Neg logit contributions
meg-3.984
-3.849
star-3.838
rovers-3.730
pres-3.700
Token his
Token ID465
Feature activation-2.44e-01

Pos logit contributions
ibaba+0.067
emale+0.065
aimon+0.051
 mustache+0.050
ijing+0.049
Neg logit contributions
meg-0.050
star-0.049
-0.048
pres-0.047
power-0.047
Token manager
Token ID4706
Feature activation+3.10e-01

Pos logit contributions
ibaba+1.023
emale+1.000
aimon+0.773
 mustache+0.756
ijing+0.754
Neg logit contributions
meg-0.752
star-0.730
-0.716
power-0.711
pres-0.707
Token and
Token ID290
Feature activation+2.41e-01

Pos logit contributions
meg+0.846
star+0.822
+0.797
power+0.789
pres+0.789
Neg logit contributions
ibaba-1.427
emale-1.392
aimon-1.100
ijing-1.081
 WRITE-1.074
Token the
Token ID262
Feature activation+5.76e-02

Pos logit contributions
meg+0.716
star+0.699
+0.678
power+0.674
pres+0.673
Neg logit contributions
ibaba-1.050
emale-1.023
aimon-0.794
ijing-0.781
 WRITE-0.773
Token
Token ID198
Feature activation+1.23e-01

Pos logit contributions
meg+0.028
star+0.027
+0.026
power+0.026
mania+0.026
Neg logit contributions
ibaba-0.054
emale-0.053
aimon-0.042
ijing-0.042
 WRITE-0.041
Token
Token ID198
Feature activation+1.14e-01

Pos logit contributions
meg+0.265
star+0.256
+0.246
pres+0.243
power+0.243
Neg logit contributions
ibaba-0.631
emale-0.617
aimon-0.502
ijing-0.495
 WRITE-0.492
TokenAnd
Token ID1870
Feature activation-4.99e-01

Pos logit contributions
meg+0.271
star+0.264
+0.253
power+0.252
pres+0.251
Neg logit contributions
ibaba-0.564
emale-0.551
aimon-0.443
ijing-0.437
 WRITE-0.432
Token don
Token ID836
Feature activation-3.73e-02

Pos logit contributions
ibaba+2.325
emale+2.265
aimon+1.800
ijing+1.770
 WRITE+1.766
Neg logit contributions
meg-1.314
star-1.278
-1.254
mania-1.231
pres-1.229
Token't
Token ID470
Feature activation+3.53e-01

Pos logit contributions
ibaba+0.196
emale+0.192
aimon+0.157
ijing+0.155
 WRITE+0.154
Neg logit contributions
meg-0.077
star-0.074
-0.071
pres-0.070
power-0.070
Token worry
Token ID5490
Feature activation-1.38e+00

Pos logit contributions
meg+1.009
star+0.987
+0.951
power+0.949
mania+0.947
Neg logit contributions
ibaba-1.580
emale-1.541
aimon-1.203
ijing-1.185
 WRITE-1.168
Token
Token ID784
Feature activation-7.80e-02

Pos logit contributions
ibaba+5.827
emale+5.644
 WRITE+4.519
aimon+4.518
 mustache+4.451
Neg logit contributions
meg-3.856
rovers-3.736
star-3.730
-3.717
pres-3.662
Token you
Token ID345
Feature activation-8.68e-02

Pos logit contributions
ibaba+0.364
emale+0.356
aimon+0.282
ijing+0.277
 WRITE+0.275
Neg logit contributions
meg-0.207
star-0.202
-0.195
power-0.193
pres-0.193
Token will
Token ID481
Feature activation+2.08e-01

Pos logit contributions
ibaba+0.407
emale+0.398
aimon+0.315
ijing+0.311
 WRITE+0.308
Neg logit contributions
meg-0.228
star-0.222
-0.215
power-0.213
pres-0.213
Token once
Token ID1752
Feature activation+4.54e-01

Pos logit contributions
meg+0.567
star+0.554
power+0.533
pres+0.531
mania+0.530
Neg logit contributions
ibaba-0.958
emale-0.935
aimon-0.736
ijing-0.725
 WRITE-0.711
Token again
Token ID757
Feature activation+3.52e-01

Pos logit contributions
meg+1.120
star+1.088
power+1.042
+1.041
mania+1.039
Neg logit contributions
ibaba-2.209
emale-2.162
aimon-1.730
ijing-1.702
 WRITE-1.684
TokenHA
Token ID7801
Feature activation-1.76e-02

Pos logit contributions
meg+0.455
star+0.442
+0.432
mania+0.426
pres+0.424
Neg logit contributions
ibaba-0.881
emale-0.858
aimon-0.689
ijing-0.678
 WRITE-0.671
TokenV
Token ID53
Feature activation-5.29e-01

Pos logit contributions
ibaba+0.085
emale+0.083
aimon+0.066
ijing+0.066
 WRITE+0.065
Neg logit contributions
meg-0.043
star-0.042
-0.040
power-0.040
pres-0.040
TokenANA
Token ID31574
Feature activation-6.29e-01

Pos logit contributions
ibaba+1.740
emale+1.665
aimon+1.180
 WRITE+1.171
ijing+1.165
Neg logit contributions
meg-2.102
star-2.049
-2.038
pres-2.002
power-1.993
Token (
Token ID357
Feature activation-8.43e-01

Pos logit contributions
ibaba+2.856
emale+2.795
aimon+2.210
 WRITE+2.202
ijing+2.173
Neg logit contributions
meg-1.709
star-1.648
-1.601
rovers-1.585
power-1.578
TokenReuters
Token ID12637
Feature activation-3.68e-01

Pos logit contributions
ibaba+3.268
emale+3.160
 WRITE+2.417
aimon+2.412
ijing+2.387
Neg logit contributions
meg-2.723
star-2.650
-2.593
rovers-2.589
power-2.567
Token)
Token ID8
Feature activation-1.33e+00

Pos logit contributions
ibaba+1.504
emale+1.464
aimon+1.115
ijing+1.095
 WRITE+1.083
Neg logit contributions
meg-1.188
star-1.162
-1.129
power-1.123
pres-1.122
Token -
Token ID532
Feature activation-8.66e-01

Pos logit contributions
ibaba+4.955
emale+4.737
 WRITE+3.648
aimon+3.576
ijing+3.479
Neg logit contributions
meg-4.578
star-4.463
rovers-4.391
-4.386
pres-4.331
Token Latin
Token ID9133
Feature activation-7.85e-02

Pos logit contributions
ibaba+3.362
emale+3.237
 WRITE+2.453
aimon+2.400
ijing+2.394
Neg logit contributions
meg-2.881
star-2.811
-2.784
rovers-2.723
pres-2.715
Token American
Token ID1605
Feature activation+1.20e-01

Pos logit contributions
ibaba+0.296
emale+0.286
ijing+0.210
aimon+0.209
 mustache+0.204
Neg logit contributions
meg-0.277
star-0.269
-0.268
pres-0.265
rovers-0.263
Token and
Token ID290
Feature activation-2.30e-01

Pos logit contributions
meg+0.364
star+0.355
+0.345
pres+0.343
power+0.342
Neg logit contributions
ibaba-0.514
emale-0.501
aimon-0.387
ijing-0.381
 WRITE-0.377
Token Caribbean
Token ID18020
Feature activation+1.47e-01

Pos logit contributions
ibaba+0.927
emale+0.896
aimon+0.684
ijing+0.673
 mustache+0.667
Neg logit contributions
meg-0.747
star-0.731
-0.725
pres-0.714
mania-0.711
Token tour
Token ID4205
Feature activation+1.48e-01

Pos logit contributions
meg+0.103
star+0.100
+0.098
power+0.097
mania+0.097
Neg logit contributions
ibaba-0.170
emale-0.166
aimon-0.131
ijing-0.129
 WRITE-0.128
Token,
Token ID11
Feature activation+9.13e-02

Pos logit contributions
meg+0.387
star+0.376
+0.364
power+0.361
pres+0.360
Neg logit contributions
ibaba-0.699
emale-0.683
aimon-0.543
ijing-0.535
 WRITE-0.529
Token plenty
Token ID6088
Feature activation-7.74e-01

Pos logit contributions
meg+0.249
star+0.242
+0.236
mania+0.233
power+0.233
Neg logit contributions
ibaba-0.418
emale-0.408
aimon-0.321
ijing-0.316
 WRITE-0.315
Token of
Token ID286
Feature activation+5.08e-02

Pos logit contributions
ibaba+3.415
emale+3.360
aimon+2.607
 mustache+2.589
ijing+2.550
Neg logit contributions
meg-2.104
star-2.037
-2.021
pres-2.014
mania-2.005
Token body
Token ID1767
Feature activation+6.71e-02

Pos logit contributions
meg+0.147
star+0.143
+0.140
power+0.138
mania+0.138
Neg logit contributions
ibaba-0.224
emale-0.219
aimon-0.171
ijing-0.168
 WRITE-0.167
Token parts
Token ID3354
Feature activation-1.32e+00

Pos logit contributions
meg+0.191
star+0.186
+0.183
mania+0.181
rovers+0.179
Neg logit contributions
ibaba-0.296
emale-0.291
aimon-0.226
ijing-0.223
 mustache-0.221
Token were
Token ID547
Feature activation-1.54e-01

Pos logit contributions
ibaba+5.566
emale+5.487
 mustache+4.286
aimon+4.274
 buckle+4.189
Neg logit contributions
meg-3.652
-3.603
star-3.573
rovers-3.506
mania-3.456
Token already
Token ID1541
Feature activation-1.39e-01

Pos logit contributions
ibaba+0.695
emale+0.678
aimon+0.536
ijing+0.528
 mustache+0.526
Neg logit contributions
meg-0.425
star-0.413
-0.406
power-0.398
mania-0.398
Token set
Token ID900
Feature activation-5.29e-02

Pos logit contributions
ibaba+0.622
emale+0.605
aimon+0.475
ijing+0.468
 WRITE+0.464
Neg logit contributions
meg-0.396
star-0.385
-0.376
power-0.371
mania-0.370
Token up
Token ID510
Feature activation+5.74e-01

Pos logit contributions
ibaba+0.251
emale+0.244
aimon+0.194
ijing+0.192
 WRITE+0.190
Neg logit contributions
meg-0.136
star-0.132
-0.128
power-0.127
mania-0.127
Token,
Token ID11
Feature activation+1.15e-02

Pos logit contributions
meg+1.361
star+1.321
+1.263
pres+1.260
power+1.256
Neg logit contributions
ibaba-2.849
emale-2.788
aimon-2.238
ijing-2.206
 WRITE-2.186
Token-
Token ID12
Feature activation-2.02e-02

Pos logit contributions
ibaba+1.708
emale+1.637
aimon+1.192
ijing+1.149
 mustache+1.147
Neg logit contributions
meg-1.957
star-1.924
-1.889
pres-1.873
mania-1.870
Tokenlevel
Token ID5715
Feature activation-3.39e-01

Pos logit contributions
ibaba+0.097
emale+0.095
aimon+0.076
ijing+0.075
 WRITE+0.074
Neg logit contributions
meg-0.050
star-0.049
-0.048
mania-0.047
pres-0.047
Token of
Token ID286
Feature activation-2.93e-01

Pos logit contributions
ibaba+1.409
emale+1.361
aimon+1.057
 mustache+1.038
ijing+1.036
Neg logit contributions
meg-1.064
star-1.036
-1.011
rovers-0.997
pres-0.995
Token intellectual
Token ID9028
Feature activation-3.63e-02

Pos logit contributions
ibaba+1.246
emale+1.201
aimon+0.948
 mustache+0.927
ijing+0.927
Neg logit contributions
meg-0.889
star-0.863
-0.848
rovers-0.828
mania-0.827
Token rig
Token ID7805
Feature activation+2.01e-01

Pos logit contributions
ibaba+0.157
emale+0.151
aimon+0.118
ijing+0.116
 WRITE+0.115
Neg logit contributions
meg-0.109
star-0.106
-0.104
pres-0.102
mania-0.102
Tokenor
Token ID273
Feature activation-1.28e+00

Pos logit contributions
meg+0.583
star+0.568
+0.559
mania+0.557
pres+0.548
Neg logit contributions
ibaba-0.882
emale-0.854
aimon-0.670
ijing-0.662
 WRITE-0.645
Token.
Token ID13
Feature activation+1.12e-01

Pos logit contributions
ibaba+5.686
emale+5.394
aimon+4.391
 mustache+4.384
ijing+4.372
Neg logit contributions
meg-3.329
star-3.212
-3.191
rovers-3.127
mania-3.117
Token But
Token ID887
Feature activation-1.11e-01

Pos logit contributions
meg+0.310
star+0.302
+0.292
power+0.290
mania+0.290
Neg logit contributions
ibaba-0.508
emale-0.496
aimon-0.390
ijing-0.384
 WRITE-0.381
Token they
Token ID484
Feature activation-2.65e-02

Pos logit contributions
ibaba+0.503
emale+0.488
aimon+0.386
ijing+0.380
 WRITE+0.379
Neg logit contributions
meg-0.307
star-0.299
-0.290
mania-0.287
power-0.286
Token surrounded
Token ID11191
Feature activation+3.38e-02

Pos logit contributions
ibaba+0.122
emale+0.119
aimon+0.094
ijing+0.093
 WRITE+0.092
Neg logit contributions
meg-0.072
star-0.070
-0.068
mania-0.067
power-0.067
Token themselves
Token ID2405
Feature activation+5.52e-01

Pos logit contributions
meg+0.097
star+0.094
+0.091
mania+0.090
power+0.090
Neg logit contributions
ibaba-0.151
emale-0.146
aimon-0.115
ijing-0.113
 WRITE-0.112
Token (
Token ID357
Feature activation-4.78e-02

Pos logit contributions
ibaba+0.175
emale+0.170
aimon+0.134
ijing+0.132
 WRITE+0.131
Neg logit contributions
meg-0.103
star-0.101
-0.098
mania-0.097
power-0.097
Tokenthe
Token ID1169
Feature activation-1.16e-01

Pos logit contributions
ibaba+0.239
emale+0.233
aimon+0.188
ijing+0.186
 WRITE+0.184
Neg logit contributions
meg-0.109
star-0.106
-0.103
mania-0.101
power-0.101
Token proposal
Token ID6961
Feature activation-3.68e-01

Pos logit contributions
ibaba+0.501
emale+0.489
aimon+0.379
ijing+0.373
 WRITE+0.370
Neg logit contributions
meg-0.346
star-0.338
-0.328
mania-0.326
power-0.326
Token by
Token ID416
Feature activation-7.25e-02

Pos logit contributions
ibaba+1.700
emale+1.652
aimon+1.313
ijing+1.299
 WRITE+1.280
Neg logit contributions
meg-0.989
star-0.960
-0.931
power-0.925
mania-0.924
Token Andy
Token ID12382
Feature activation-2.40e-01

Pos logit contributions
ibaba+0.311
emale+0.303
aimon+0.234
ijing+0.231
 WRITE+0.228
Neg logit contributions
meg-0.220
star-0.215
-0.209
mania-0.207
pres-0.207
Token Stern
Token ID22990
Feature activation-1.28e+00

Pos logit contributions
ibaba+1.047
emale+1.013
aimon+0.799
ijing+0.786
 WRITE+0.780
Neg logit contributions
meg-0.706
star-0.676
-0.669
pres-0.668
power-0.662
Token in
Token ID287
Feature activation-8.60e-02

Pos logit contributions
ibaba+5.860
emale+5.750
aimon+4.606
ijing+4.580
 mustache+4.454
Neg logit contributions
meg-3.185
-3.044
star-3.003
mania-2.948
power-2.942
Token Ra
Token ID7567
Feature activation+1.84e-01

Pos logit contributions
ibaba+0.375
emale+0.366
aimon+0.285
ijing+0.281
 WRITE+0.278
Neg logit contributions
meg-0.254
star-0.247
-0.240
pres-0.238
mania-0.238
Tokenising
Token ID1710
Feature activation+5.70e-02

Pos logit contributions
meg+0.620
star+0.607
+0.593
mania+0.588
power+0.587
Neg logit contributions
ibaba-0.725
emale-0.705
aimon-0.531
ijing-0.521
 WRITE-0.514
Token the
Token ID262
Feature activation-3.88e-01

Pos logit contributions
meg+0.176
star+0.172
+0.167
power+0.166
pres+0.166
Neg logit contributions
ibaba-0.241
emale-0.235
aimon-0.181
ijing-0.178
 WRITE-0.176
Token Floor
Token ID22343
Feature activation-3.77e-01

Pos logit contributions
ibaba+1.536
emale+1.503
aimon+1.113
ijing+1.113
 WRITE+1.110
Neg logit contributions
meg-1.293
star-1.257
-1.230
mania-1.230
pres-1.224
TokenFrank
Token ID17439
Feature activation+2.48e-02

Pos logit contributions
meg+0.066
+0.064
star+0.064
pres+0.062
rovers+0.062
Neg logit contributions
ibaba-0.088
emale-0.087
ijing-0.066
aimon-0.066
 WRITE-0.065
Tokeniv
Token ID452
Feature activation-2.42e-02

Pos logit contributions
meg+0.044
star+0.042
+0.040
mania+0.040
power+0.039
Neg logit contributions
ibaba-0.138
emale-0.134
aimon-0.111
ijing-0.110
 WRITE-0.108
Tokensk
Token ID8135
Feature activation+2.82e-01

Pos logit contributions
ibaba+0.121
emale+0.114
ijing+0.095
aimon+0.095
irtual+0.093
Neg logit contributions
meg-0.054
-0.054
star-0.051
mania-0.051
power-0.050
Token.
Token ID13
Feature activation-2.70e-01

Pos logit contributions
meg+0.703
star+0.685
+0.661
mania+0.655
power+0.655
Neg logit contributions
ibaba-1.357
emale-1.324
aimon-1.057
ijing-1.046
 WRITE-1.033
Token<|endoftext|>
Token ID50256
Feature activation+2.05e-02

Pos logit contributions
ibaba+1.191
emale+1.160
aimon+0.903
 WRITE+0.896
ijing+0.889
Neg logit contributions
meg-0.773
star-0.753
-0.735
pres-0.723
power-0.720
Token[
Token ID58
Feature activation-1.27e+00

Pos logit contributions
meg+0.051
star+0.049
+0.048
mania+0.047
pres+0.047
Neg logit contributions
ibaba-0.098
emale-0.095
aimon-0.077
ijing-0.075
 WRITE-0.075
TokenEx
Token ID3109
Feature activation+5.18e-02

Pos logit contributions
ibaba+5.912
emale+5.692
aimon+4.566
ijing+4.551
 WRITE+4.544
Neg logit contributions
meg-3.131
-3.001
star-2.991
rovers-2.908
mania-2.900
Tokenclusive
Token ID5731
Feature activation-9.12e-03

Pos logit contributions
meg+0.144
star+0.140
+0.138
mania+0.135
power+0.135
Neg logit contributions
ibaba-0.233
emale-0.229
aimon-0.178
ijing-0.176
 WRITE-0.173
Token]
Token ID60
Feature activation+1.90e-01

Pos logit contributions
ibaba+0.040
emale+0.039
aimon+0.031
ijing+0.030
 WRITE+0.030
Neg logit contributions
meg-0.026
star-0.026
-0.025
mania-0.025
pres-0.025
Token Best
Token ID6705
Feature activation+1.32e-01

Pos logit contributions
meg+0.593
star+0.578
+0.564
pres+0.559
power+0.559
Neg logit contributions
ibaba-0.799
emale-0.778
aimon-0.597
ijing-0.588
 WRITE-0.584
Token Buy
Token ID11763
Feature activation-3.95e-01

Pos logit contributions
meg+0.413
star+0.403
+0.393
power+0.390
mania+0.390
Neg logit contributions
ibaba-0.553
emale-0.539
aimon-0.413
ijing-0.407
 WRITE-0.402
Token,
Token ID11
Feature activation-1.07e-01

Pos logit contributions
ibaba+0.207
emale+0.202
aimon+0.158
ijing+0.156
 WRITE+0.156
Neg logit contributions
meg-0.125
star-0.121
-0.118
mania-0.116
power-0.116
Token I
Token ID314
Feature activation-5.61e-03

Pos logit contributions
ibaba+0.496
emale+0.484
aimon+0.383
ijing+0.377
 WRITE+0.375
Neg logit contributions
meg-0.285
star-0.277
-0.269
power-0.265
mania-0.265
Token'm
Token ID1101
Feature activation+1.42e-02

Pos logit contributions
ibaba+0.026
emale+0.026
aimon+0.020
ijing+0.020
 WRITE+0.020
Neg logit contributions
meg-0.015
star-0.014
-0.014
pres-0.014
power-0.014
Token not
Token ID407
Feature activation-1.75e-01

Pos logit contributions
meg+0.041
star+0.040
+0.038
power+0.038
pres+0.038
Neg logit contributions
ibaba-0.064
emale-0.062
aimon-0.049
ijing-0.048
 WRITE-0.047
Token looking
Token ID2045
Feature activation+2.10e-01

Pos logit contributions
ibaba+0.754
emale+0.735
aimon+0.571
ijing+0.566
 WRITE+0.560
Neg logit contributions
meg-0.523
star-0.506
-0.496
power-0.488
pres-0.488
Token ahead
Token ID4058
Feature activation-1.27e+00

Pos logit contributions
meg+0.539
star+0.518
+0.511
mania+0.499
pres+0.499
Neg logit contributions
ibaba-0.996
emale-0.969
aimon-0.773
ijing-0.769
 WRITE-0.761
Token,
Token ID11
Feature activation-1.22e-01

Pos logit contributions
ibaba+5.596
emale+5.448
aimon+4.371
ijing+4.369
 WRITE+4.364
Neg logit contributions
meg-3.289
-3.179
star-3.137
rovers-3.093
power-3.022
Token I
Token ID314
Feature activation-1.01e-01

Pos logit contributions
ibaba+0.561
emale+0.548
aimon+0.431
ijing+0.425
 WRITE+0.423
Neg logit contributions
meg-0.332
star-0.323
-0.314
power-0.310
mania-0.310
Token'm
Token ID1101
Feature activation-1.36e-01

Pos logit contributions
ibaba+0.489
emale+0.478
aimon+0.383
ijing+0.379
 WRITE+0.378
Neg logit contributions
meg-0.250
star-0.241
-0.236
pres-0.231
power-0.230
Token not
Token ID407
Feature activation-1.28e-01

Pos logit contributions
ibaba+0.568
emale+0.551
aimon+0.426
ijing+0.420
 WRITE+0.415
Neg logit contributions
meg-0.427
star-0.414
-0.406
rovers-0.400
pres-0.400
Token looking
Token ID2045
Feature activation+1.96e-01

Pos logit contributions
ibaba+0.519
emale+0.504
aimon+0.385
ijing+0.379
 WRITE+0.376
Neg logit contributions
meg-0.416
star-0.404
-0.396
power-0.391
pres-0.391
Token 2015
Token ID1853
Feature activation-2.57e-02

Pos logit contributions
meg+0.101
star+0.098
+0.096
mania+0.095
pres+0.095
Neg logit contributions
ibaba-0.184
emale-0.179
aimon-0.143
ijing-0.141
 WRITE-0.140
Token
Token ID198
Feature activation-1.81e-01

Pos logit contributions
ibaba+0.123
emale+0.120
aimon+0.096
ijing+0.095
 WRITE+0.095
Neg logit contributions
meg-0.064
star-0.062
-0.061
mania-0.060
pres-0.060
Token
Token ID198
Feature activation-1.36e-01

Pos logit contributions
ibaba+0.871
emale+0.851
aimon+0.686
ijing+0.676
 WRITE+0.673
Neg logit contributions
meg-0.436
star-0.422
-0.412
pres-0.405
mania-0.405
TokenFR
Token ID10913
Feature activation-3.29e-01

Pos logit contributions
ibaba+0.651
emale+0.636
aimon+0.511
ijing+0.503
 WRITE+0.499
Neg logit contributions
meg-0.334
star-0.323
-0.317
mania-0.311
rovers-0.311
TokenES
Token ID1546
Feature activation-3.37e-01

Pos logit contributions
ibaba+1.416
emale+1.382
aimon+1.078
 WRITE+1.066
ijing+1.064
Neg logit contributions
meg-0.967
star-0.943
-0.923
power-0.905
rovers-0.902
TokenNO
Token ID15285
Feature activation-1.27e+00

Pos logit contributions
ibaba+1.541
emale+1.523
ijing+1.184
aimon+1.182
 WRITE+1.176
Neg logit contributions
meg-0.899
star-0.884
-0.853
power-0.849
rovers-0.844
Token,
Token ID11
Feature activation-2.05e-01

Pos logit contributions
emale+5.213
ibaba+5.192
 WRITE+4.099
aimon+4.045
ijing+3.980
Neg logit contributions
meg-3.699
star-3.580
-3.491
rovers-3.465
mania-3.446
Token Calif
Token ID3192
Feature activation-4.24e-01

Pos logit contributions
ibaba+0.799
emale+0.779
aimon+0.586
 WRITE+0.574
ijing+0.574
Neg logit contributions
meg-0.697
star-0.680
-0.665
power-0.660
mania-0.658
Token.
Token ID13
Feature activation-3.26e-01

Pos logit contributions
ibaba+2.210
emale+2.169
aimon+1.762
ijing+1.750
 WRITE+1.744
Neg logit contributions
meg-0.858
-0.817
star-0.816
mania-0.793
power-0.778
Token (
Token ID357
Feature activation-5.55e-01

Pos logit contributions
ibaba+1.428
emale+1.394
aimon+1.086
 WRITE+1.086
ijing+1.072
Neg logit contributions
meg-0.938
star-0.910
-0.888
mania-0.879
power-0.874
TokenK
Token ID42
Feature activation-3.88e-01

Pos logit contributions
ibaba+2.690
emale+2.632
aimon+2.139
 WRITE+2.124
ijing+2.095
Neg logit contributions
meg-1.279
star-1.235
-1.202
rovers-1.186
mania-1.186
Token (
Token ID357
Feature activation-3.94e-01

Pos logit contributions
meg+0.245
star+0.237
+0.229
power+0.226
pres+0.225
Neg logit contributions
ibaba-0.514
emale-0.504
aimon-0.405
ijing-0.400
 WRITE-0.397
TokenFigure
Token ID11337
Feature activation-4.55e-01

Pos logit contributions
ibaba+1.896
emale+1.861
aimon+1.486
ijing+1.465
 WRITE+1.449
Neg logit contributions
meg-0.958
star-0.929
-0.904
rovers-0.887
mania-0.886
Token 2
Token ID362
Feature activation-3.80e-01

Pos logit contributions
ibaba+1.753
emale+1.704
aimon+1.262
ijing+1.254
 WRITE+1.239
Neg logit contributions
meg-1.572
star-1.532
-1.500
power-1.484
pres-1.482
TokenA
Token ID32
Feature activation-4.92e-02

Pos logit contributions
ibaba+1.620
emale+1.586
aimon+1.211
ijing+1.204
 WRITE+1.191
Neg logit contributions
meg-1.153
star-1.125
-1.101
mania-1.087
power-1.085
Token,
Token ID11
Feature activation-1.40e-01

Pos logit contributions
ibaba+0.193
emale+0.188
aimon+0.140
ijing+0.138
 WRITE+0.137
Neg logit contributions
meg-0.167
star-0.163
-0.159
mania-0.158
power-0.158
Token [
Token ID685
Feature activation-1.26e+00

Pos logit contributions
ibaba+0.635
emale+0.620
aimon+0.486
ijing+0.480
 WRITE+0.475
Neg logit contributions
meg-0.388
star-0.377
-0.366
mania-0.363
power-0.363
TokenAction
Token ID12502
Feature activation-3.47e-02

Pos logit contributions
ibaba+5.830
emale+5.689
aimon+4.520
ijing+4.495
 WRITE+4.433
Neg logit contributions
meg-3.101
star-2.968
-2.967
rovers-2.930
mania-2.892
Token Junior
Token ID20000
Feature activation+2.50e-01

Pos logit contributions
ibaba+0.149
emale+0.145
aimon+0.111
ijing+0.110
 WRITE+0.109
Neg logit contributions
meg-0.106
star-0.103
-0.100
pres-0.099
power-0.099
Token 3
Token ID513
Feature activation-1.41e-01

Pos logit contributions
meg+0.787
star+0.771
+0.748
power+0.744
mania+0.744
Neg logit contributions
ibaba-1.045
emale-1.015
aimon-0.779
ijing-0.766
 WRITE-0.755
Token,
Token ID11
Feature activation-1.33e-01

Pos logit contributions
ibaba+0.637
emale+0.626
aimon+0.488
ijing+0.483
 WRITE+0.479
Neg logit contributions
meg-0.394
star-0.382
-0.371
mania-0.368
pres-0.367
Token Inv
Token ID10001
Feature activation-4.63e-01

Pos logit contributions
ibaba+0.551
emale+0.538
aimon+0.411
ijing+0.403
 WRITE+0.400
Neg logit contributions
meg-0.425
star-0.415
-0.403
mania-0.401
pres-0.401
Token Y
Token ID575
Feature activation+8.63e-02

Pos logit contributions
ibaba+0.004
emale+0.004
aimon+0.002
ijing+0.002
 WRITE+0.002
Neg logit contributions
meg-0.005
star-0.005
-0.005
power-0.005
pres-0.005
Tokenip
Token ID541
Feature activation+2.68e-01

Pos logit contributions
meg+0.252
star+0.246
+0.243
mania+0.237
pres+0.235
Neg logit contributions
ibaba-0.381
emale-0.370
aimon-0.288
ijing-0.287
 WRITE-0.280
Token
Token ID198
Feature activation-6.14e-01

Pos logit contributions
meg+0.742
star+0.724
+0.699
power+0.695
mania+0.694
Neg logit contributions
ibaba-1.220
emale-1.191
aimon-0.938
ijing-0.921
 WRITE-0.912
Token
Token ID198
Feature activation-4.21e-01

Pos logit contributions
ibaba+2.972
emale+2.849
aimon+2.339
 WRITE+2.325
ijing+2.283
Neg logit contributions
meg-1.450
star-1.391
-1.367
rovers-1.350
pres-1.332
TokenH
Token ID39
Feature activation-2.75e-01

Pos logit contributions
ibaba+1.833
emale+1.790
aimon+1.385
ijing+1.381
 WRITE+1.363
Neg logit contributions
meg-1.226
star-1.197
-1.162
pres-1.157
rovers-1.154
TokenONG
Token ID18494
Feature activation-1.24e+00

Pos logit contributions
ibaba+0.941
emale+0.906
aimon+0.645
ijing+0.642
 WRITE+0.618
Neg logit contributions
meg-1.071
star-1.057
-1.036
mania-1.027
pres-1.025
Token K
Token ID509
Feature activation-5.22e-01

Pos logit contributions
ibaba+4.783
emale+4.670
 WRITE+3.827
 PLA+3.756
ijing+3.668
Neg logit contributions
meg-3.975
rovers-3.811
star-3.791
power-3.642
pres-3.629
TokenONG
Token ID18494
Feature activation-9.23e-01

Pos logit contributions
ibaba+1.149
emale+1.119
aimon+0.632
ijing+0.627
 WRITE+0.597
Neg logit contributions
meg-2.635
star-2.587
rovers-2.540
-2.537
mania-2.529
Token (
Token ID357
Feature activation-7.45e-01

Pos logit contributions
ibaba+4.054
emale+3.999
ijing+3.113
 WRITE+3.108
aimon+3.064
Neg logit contributions
meg-2.573
star-2.497
rovers-2.416
pres-2.399
-2.391
TokenReuters
Token ID12637
Feature activation-3.18e-01

Pos logit contributions
ibaba+2.910
emale+2.792
ijing+2.129
 WRITE+2.119
aimon+2.108
Neg logit contributions
meg-2.446
star-2.400
-2.344
rovers-2.330
pres-2.320
Token)
Token ID8
Feature activation-1.22e+00

Pos logit contributions
ibaba+1.315
emale+1.280
aimon+0.975
ijing+0.959
 WRITE+0.945
Neg logit contributions
meg-1.020
star-0.999
-0.969
power-0.965
mania-0.964
Token suffered
Token ID6989
Feature activation-2.23e-01

Pos logit contributions
ibaba+1.958
emale+1.912
aimon+1.515
 mustache+1.494
 WRITE+1.491
Neg logit contributions
meg-1.171
star-1.131
-1.113
rovers-1.089
pres-1.088
Token a
Token ID257
Feature activation-3.10e-01

Pos logit contributions
ibaba+0.940
emale+0.907
aimon+0.701
 mustache+0.691
 WRITE+0.689
Neg logit contributions
meg-0.690
star-0.674
-0.661
pres-0.655
power-0.653
Token rupt
Token ID31580
Feature activation-6.92e-01

Pos logit contributions
ibaba+1.227
emale+1.183
aimon+0.896
 mustache+0.879
ijing+0.876
Neg logit contributions
meg-1.036
star-1.013
-0.996
pres-0.982
mania-0.979
Tokenured
Token ID1522
Feature activation-9.43e-02

Pos logit contributions
ibaba+2.320
emale+2.251
aimon+1.604
ijing+1.552
 WRITE+1.516
Neg logit contributions
meg-2.708
-2.680
star-2.637
mania-2.598
pres-2.567
Token Achilles
Token ID39561
Feature activation-6.54e-01

Pos logit contributions
ibaba+0.307
emale+0.295
aimon+0.206
ijing+0.201
 WRITE+0.200
Neg logit contributions
meg-0.384
star-0.377
-0.370
pres-0.368
mania-0.367
Token tendon
Token ID42205
Feature activation-1.23e+00

Pos logit contributions
ibaba+2.004
emale+1.948
aimon+1.343
 mustache+1.320
 WRITE+1.302
Neg logit contributions
meg-2.716
star-2.670
-2.636
pres-2.613
mania-2.592
Token in
Token ID287
Feature activation-1.25e-01

Pos logit contributions
ibaba+5.580
emale+5.430
aimon+4.368
 mustache+4.361
 WRITE+4.293
Neg logit contributions
meg-3.146
star-3.063
-3.019
rovers-2.950
pres-2.941
Token Thursday
Token ID3635
Feature activation-7.83e-02

Pos logit contributions
ibaba+0.566
emale+0.549
aimon+0.436
ijing+0.427
 WRITE+0.424
Neg logit contributions
meg-0.345
star-0.336
-0.328
pres-0.323
power-0.323
Token's
Token ID338
Feature activation-1.60e-01

Pos logit contributions
ibaba+0.395
emale+0.383
aimon+0.313
ijing+0.307
 mustache+0.307
Neg logit contributions
meg-0.178
-0.169
star-0.168
rovers-0.165
power-0.162
Token game
Token ID983
Feature activation-2.07e-01

Pos logit contributions
ibaba+0.641
emale+0.621
aimon+0.473
ijing+0.462
 WRITE+0.461
Neg logit contributions
meg-0.526
star-0.515
-0.504
pres-0.498
power-0.498
Token against
Token ID1028
Feature activation-3.59e-01

Pos logit contributions
ibaba+1.038
emale+1.015
aimon+0.819
ijing+0.809
 WRITE+0.807
Neg logit contributions
meg-0.470
star-0.453
-0.442
power-0.432
mania-0.430
Token Media
Token ID6343
Feature activation-2.10e-01

Pos logit contributions
ibaba+0.497
emale+0.486
aimon+0.372
ijing+0.368
 WRITE+0.366
Neg logit contributions
meg-0.364
star-0.354
-0.347
pres-0.343
mania-0.343
Tokenite
Token ID578
Feature activation+8.92e-02

Pos logit contributions
ibaba+1.036
emale+1.014
aimon+0.813
ijing+0.806
 WRITE+0.806
Neg logit contributions
meg-0.495
star-0.475
-0.466
pres-0.456
rovers-0.454
Token:
Token ID25
Feature activation-2.00e-01

Pos logit contributions
meg+0.220
star+0.213
+0.206
pres+0.204
mania+0.204
Neg logit contributions
ibaba-0.432
emale-0.422
aimon-0.337
ijing-0.335
 WRITE-0.331
Token
Token ID198
Feature activation-3.54e-01

Pos logit contributions
ibaba+0.975
emale+0.951
aimon+0.761
ijing+0.757
 WRITE+0.755
Neg logit contributions
meg-0.480
star-0.462
-0.447
mania-0.444
pres-0.440
Token
Token ID198
Feature activation-4.22e-01

Pos logit contributions
ibaba+1.728
emale+1.663
aimon+1.364
 WRITE+1.350
ijing+1.334
Neg logit contributions
meg-0.829
star-0.793
-0.780
rovers-0.767
pres-0.756
Token[
Token ID58
Feature activation-1.22e+00

Pos logit contributions
ibaba+2.023
emale+1.989
aimon+1.597
ijing+1.584
 WRITE+1.574
Neg logit contributions
meg-0.992
star-0.959
-0.934
rovers-0.921
mania-0.919
TokenChuck
Token ID44324
Feature activation-1.60e-01

Pos logit contributions
ibaba+5.565
emale+5.473
ijing+4.333
aimon+4.333
 WRITE+4.304
Neg logit contributions
meg-3.056
star-2.936
-2.912
rovers-2.831
mania-2.803
Token Todd
Token ID14377
Feature activation+3.65e-01

Pos logit contributions
ibaba+0.669
emale+0.656
aimon+0.501
ijing+0.499
 WRITE+0.497
Neg logit contributions
meg-0.488
-0.470
star-0.465
pres-0.463
rovers-0.454
Token:]
Token ID47715
Feature activation-3.94e-01

Pos logit contributions
meg+1.048
star+1.019
+0.991
pres+0.986
power+0.982
Neg logit contributions
ibaba-1.616
emale-1.588
aimon-1.234
ijing-1.220
 WRITE-1.211
Token
Token ID564
Feature activation+1.49e-01

Pos logit contributions
ibaba+1.713
emale+1.672
aimon+1.295
 WRITE+1.286
ijing+1.282
Neg logit contributions
meg-1.152
star-1.121
-1.093
mania-1.079
power-1.077
Token
Token ID250
Feature activation+1.70e-01

Pos logit contributions
meg+0.527
star+0.516
+0.504
mania+0.501
pres+0.500
Neg logit contributions
ibaba-0.561
emale-0.542
aimon-0.399
ijing-0.396
 WRITE-0.392
TokenONG
Token ID18494
Feature activation-1.24e+00

Pos logit contributions
ibaba+0.941
emale+0.906
aimon+0.645
ijing+0.642
 WRITE+0.618
Neg logit contributions
meg-1.071
star-1.057
-1.036
mania-1.027
pres-1.025
Token K
Token ID509
Feature activation-5.22e-01

Pos logit contributions
ibaba+4.783
emale+4.670
 WRITE+3.827
 PLA+3.756
ijing+3.668
Neg logit contributions
meg-3.975
rovers-3.811
star-3.791
power-3.642
pres-3.629
TokenONG
Token ID18494
Feature activation-9.23e-01

Pos logit contributions
ibaba+1.149
emale+1.119
aimon+0.632
ijing+0.627
 WRITE+0.597
Neg logit contributions
meg-2.635
star-2.587
rovers-2.540
-2.537
mania-2.529
Token (
Token ID357
Feature activation-7.45e-01

Pos logit contributions
ibaba+4.054
emale+3.999
ijing+3.113
 WRITE+3.108
aimon+3.064
Neg logit contributions
meg-2.573
star-2.497
rovers-2.416
pres-2.399
-2.391
TokenReuters
Token ID12637
Feature activation-3.18e-01

Pos logit contributions
ibaba+2.910
emale+2.792
ijing+2.129
 WRITE+2.119
aimon+2.108
Neg logit contributions
meg-2.446
star-2.400
-2.344
rovers-2.330
pres-2.320
Token)
Token ID8
Feature activation-1.22e+00

Pos logit contributions
ibaba+1.315
emale+1.280
aimon+0.975
ijing+0.959
 WRITE+0.945
Neg logit contributions
meg-1.020
star-0.999
-0.969
power-0.965
mania-0.964
Token -
Token ID532
Feature activation-7.36e-01

Pos logit contributions
ibaba+4.463
emale+4.297
 WRITE+3.331
aimon+3.220
ijing+3.157
Neg logit contributions
meg-4.238
star-4.116
rovers-4.063
-4.032
pres-4.009
Token One
Token ID1881
Feature activation-4.34e-01

Pos logit contributions
ibaba+2.902
emale+2.796
ijing+2.100
 WRITE+2.095
aimon+2.090
Neg logit contributions
meg-2.434
star-2.385
-2.337
pres-2.302
power-2.299
Token needs
Token ID2476
Feature activation-1.82e-01

Pos logit contributions
ibaba+2.015
emale+1.938
ijing+1.538
aimon+1.533
 mustache+1.508
Neg logit contributions
meg-1.151
star-1.126
-1.097
pres-1.092
rovers-1.089
Token to
Token ID284
Feature activation-1.61e-01

Pos logit contributions
ibaba+0.807
emale+0.781
aimon+0.609
ijing+0.606
 WRITE+0.598
Neg logit contributions
meg-0.527
star-0.512
-0.500
mania-0.498
power-0.492
Token look
Token ID804
Feature activation+3.98e-01

Pos logit contributions
ibaba+0.711
emale+0.690
aimon+0.540
ijing+0.534
 WRITE+0.531
Neg logit contributions
meg-0.463
star-0.451
-0.441
mania-0.435
power-0.434
Tokenre
Token ID260
Feature activation-5.41e-02

Pos logit contributions
ibaba+0.438
emale+0.431
aimon+0.346
ijing+0.340
 WRITE+0.336
Neg logit contributions
meg-0.220
-0.210
star-0.209
mania-0.206
rovers-0.204
Tokenleasing
Token ID48764
Feature activation+3.69e-02

Pos logit contributions
ibaba+0.207
emale+0.202
aimon+0.152
ijing+0.148
 WRITE+0.144
Neg logit contributions
meg-0.185
star-0.181
-0.180
rovers-0.178
mania-0.177
Token factor
Token ID5766
Feature activation+2.42e-01

Pos logit contributions
meg+0.148
star+0.144
+0.143
mania+0.142
pres+0.142
Neg logit contributions
ibaba-0.121
emale-0.117
aimon-0.082
ijing-0.080
 WRITE-0.079
Token (
Token ID357
Feature activation-2.91e-01

Pos logit contributions
meg+0.614
star+0.597
+0.580
power+0.575
pres+0.572
Neg logit contributions
ibaba-1.153
emale-1.123
aimon-0.892
ijing-0.884
 WRITE-0.878
TokenCR
Token ID9419
Feature activation-1.39e-01

Pos logit contributions
ibaba+1.287
emale+1.257
aimon+0.983
ijing+0.979
 WRITE+0.970
Neg logit contributions
meg-0.804
star-0.790
-0.778
pres-0.762
mania-0.762
TokenF
Token ID37
Feature activation-1.20e+00

Pos logit contributions
ibaba+0.656
emale+0.649
aimon+0.514
ijing+0.508
 WRITE+0.505
Neg logit contributions
meg-0.344
-0.339
star-0.332
rovers-0.327
power-0.323
Token)
Token ID8
Feature activation-5.56e-01

Pos logit contributions
ibaba+4.544
emale+4.446
aimon+3.306
ijing+3.284
 mustache+3.206
Neg logit contributions
-3.864
meg-3.820
rovers-3.758
power-3.730
star-3.706
Token is
Token ID318
Feature activation-9.26e-03

Pos logit contributions
ibaba+2.491
emale+2.396
aimon+1.887
ijing+1.874
 WRITE+1.864
Neg logit contributions
meg-1.518
star-1.483
-1.465
power-1.437
mania-1.433
Token a
Token ID257
Feature activation+8.01e-02

Pos logit contributions
ibaba+0.040
emale+0.039
aimon+0.030
ijing+0.030
 WRITE+0.029
Neg logit contributions
meg-0.028
star-0.027
-0.027
mania-0.026
power-0.026
Token hormone
Token ID17770
Feature activation+4.35e-01

Pos logit contributions
meg+0.253
star+0.247
+0.243
mania+0.240
pres+0.239
Neg logit contributions
ibaba-0.332
emale-0.323
aimon-0.247
ijing-0.243
 WRITE-0.241
Token secret
Token ID3200
Feature activation+5.63e-01

Pos logit contributions
meg+1.125
star+1.093
+1.059
power+1.050
mania+1.049
Neg logit contributions
ibaba-2.054
emale-2.004
aimon-1.594
ijing-1.573
 WRITE-1.558
Token finish
Token ID5461
Feature activation+1.76e-01

Pos logit contributions
ibaba+1.573
emale+1.520
aimon+1.169
 WRITE+1.153
 mustache+1.145
Neg logit contributions
meg-1.199
star-1.169
-1.154
pres-1.136
mania-1.132
Token the
Token ID262
Feature activation-9.57e-02

Pos logit contributions
meg+0.485
star+0.473
+0.457
power+0.454
pres+0.454
Neg logit contributions
ibaba-0.801
emale-0.782
aimon-0.615
ijing-0.606
 WRITE-0.600
Token season
Token ID1622
Feature activation-1.51e-01

Pos logit contributions
ibaba+0.460
emale+0.448
aimon+0.360
ijing+0.353
 WRITE+0.353
Neg logit contributions
meg-0.238
star-0.230
-0.225
pres-0.222
mania-0.222
Token with
Token ID351
Feature activation-2.37e-01

Pos logit contributions
ibaba+0.680
emale+0.663
aimon+0.521
ijing+0.513
 WRITE+0.509
Neg logit contributions
meg-0.423
star-0.411
-0.399
pres-0.396
power-0.395
Token 10
Token ID838
Feature activation-1.66e-01

Pos logit contributions
ibaba+0.934
emale+0.906
aimon+0.683
ijing+0.670
 WRITE+0.665
Neg logit contributions
meg-0.797
star-0.781
-0.763
pres-0.757
mania-0.755
Token26
Token ID2075
Feature activation-1.20e+00

Pos logit contributions
ibaba+0.710
emale+0.684
aimon+0.531
ijing+0.525
 WRITE+0.519
Neg logit contributions
meg-0.501
star-0.489
-0.477
pres-0.474
power-0.470
Token yards
Token ID5695
Feature activation-3.21e-01

Pos logit contributions
ibaba+4.667
emale+4.423
aimon+3.404
 mustache+3.390
 margins+3.331
Neg logit contributions
meg-3.927
star-3.817
-3.793
rovers-3.760
pres-3.686
Token at
Token ID379
Feature activation+1.78e-01

Pos logit contributions
ibaba+1.459
emale+1.405
aimon+1.125
 WRITE+1.105
 mustache+1.101
Neg logit contributions
meg-0.880
star-0.850
-0.830
rovers-0.821
pres-0.820
Token 5
Token ID642
Feature activation+9.93e-02

Pos logit contributions
meg+0.573
star+0.558
+0.546
pres+0.543
mania+0.541
Neg logit contributions
ibaba-0.726
emale-0.706
aimon-0.540
ijing-0.528
 WRITE-0.524
Token.
Token ID13
Feature activation-1.43e-01

Pos logit contributions
meg+0.200
star+0.192
+0.185
pres+0.183
power+0.183
Neg logit contributions
ibaba-0.526
emale-0.514
aimon-0.421
ijing-0.415
 WRITE-0.412
Token4
Token ID19
Feature activation-1.68e-01

Pos logit contributions
ibaba+0.655
emale+0.639
aimon+0.505
ijing+0.496
 WRITE+0.493
Neg logit contributions
meg-0.389
star-0.378
-0.367
pres-0.364
power-0.363
Token example
Token ID1672
Feature activation-4.97e-01

Pos logit contributions
emale+2.496
ibaba+2.488
aimon+1.868
 margins+1.835
 mustache+1.822
Neg logit contributions
meg-2.133
star-2.057
-2.045
mania-2.019
pres-2.015
Token with
Token ID351
Feature activation-3.81e-01

Pos logit contributions
ibaba+2.270
emale+2.190
aimon+1.750
ijing+1.748
 WRITE+1.727
Neg logit contributions
meg-1.310
star-1.256
-1.240
mania-1.221
rovers-1.220
Token hope
Token ID2911
Feature activation-5.20e-01

Pos logit contributions
ibaba+1.605
emale+1.569
aimon+1.214
ijing+1.202
 WRITE+1.195
Neg logit contributions
meg-1.155
star-1.116
-1.103
mania-1.083
rovers-1.081
Token,
Token ID11
Feature activation+7.37e-02

Pos logit contributions
ibaba+2.388
emale+2.294
aimon+1.843
ijing+1.824
 WRITE+1.815
Neg logit contributions
meg-1.374
star-1.316
rovers-1.291
-1.288
mania-1.272
Token never
Token ID1239
Feature activation-2.73e-02

Pos logit contributions
meg+0.226
star+0.219
+0.216
mania+0.212
pres+0.211
Neg logit contributions
ibaba-0.312
emale-0.303
aimon-0.235
ijing-0.230
 WRITE-0.229
Token fear
Token ID3252
Feature activation-1.20e+00

Pos logit contributions
ibaba+0.124
emale+0.121
aimon+0.095
ijing+0.094
 WRITE+0.094
Neg logit contributions
meg-0.075
star-0.073
-0.071
mania-0.070
pres-0.070
Token'
Token ID6
Feature activation-2.83e-03

Pos logit contributions
ibaba+5.192
emale+4.980
aimon+3.975
 mustache+3.892
 WRITE+3.889
Neg logit contributions
meg-3.330
-3.224
star-3.203
rovers-3.167
pres-3.143
Token
Token ID198
Feature activation+1.27e-01

Pos logit contributions
ibaba+0.014
emale+0.014
aimon+0.011
ijing+0.011
 WRITE+0.011
Neg logit contributions
meg-0.007
star-0.006
-0.006
mania-0.006
pres-0.006
Token
Token ID198
Feature activation-3.17e-03

Pos logit contributions
meg+0.275
star+0.265
+0.255
pres+0.252
mania+0.252
Neg logit contributions
ibaba-0.652
emale-0.637
aimon-0.519
ijing-0.511
 WRITE-0.508
TokenMrs
Token ID27034
Feature activation+2.03e-01

Pos logit contributions
ibaba+0.015
emale+0.015
aimon+0.012
ijing+0.012
 WRITE+0.012
Neg logit contributions
meg-0.008
star-0.007
-0.007
mania-0.007
rovers-0.007
Token.
Token ID13
Feature activation+9.79e-02

Pos logit contributions
meg+0.349
star+0.336
+0.315
power+0.314
mania+0.313
Neg logit contributions
ibaba-1.140
emale-1.119
aimon-0.924
ijing-0.913
 WRITE-0.904
Token,
Token ID11
Feature activation+3.06e-01

Pos logit contributions
ibaba+0.129
emale+0.126
aimon+0.100
ijing+0.099
 WRITE+0.098
Neg logit contributions
meg-0.066
star-0.064
-0.062
mania-0.061
pres-0.061
Token000
Token ID830
Feature activation+1.12e-01

Pos logit contributions
meg+0.879
star+0.858
+0.828
power+0.824
pres+0.822
Neg logit contributions
ibaba-1.361
emale-1.329
aimon-1.038
ijing-1.021
 WRITE-1.012
Token won
Token ID1839
Feature activation+6.43e-01

Pos logit contributions
meg+0.320
star+0.312
+0.301
power+0.300
pres+0.300
Neg logit contributions
ibaba-0.501
emale-0.488
aimon-0.383
ijing-0.375
 WRITE-0.372
Token ($
Token ID7198
Feature activation+1.98e-01

Pos logit contributions
meg+1.924
star+1.885
power+1.824
+1.816
pres+1.814
Neg logit contributions
ibaba-2.776
emale-2.729
aimon-2.108
ijing-2.052
 WRITE-2.033
Token8
Token ID23
Feature activation-1.99e-01

Pos logit contributions
meg+0.534
star+0.521
+0.503
power+0.499
pres+0.498
Neg logit contributions
ibaba-0.927
emale-0.904
aimon-0.716
ijing-0.705
 mustache-0.696
Token52
Token ID4309
Feature activation-1.19e+00

Pos logit contributions
ibaba+0.921
emale+0.899
aimon+0.710
ijing+0.700
 WRITE+0.693
Neg logit contributions
meg-0.533
star-0.520
-0.502
power-0.498
pres-0.498
Token.
Token ID13
Feature activation-1.63e-01

Pos logit contributions
ibaba+4.896
emale+4.697
ijing+3.608
aimon+3.582
 WRITE+3.542
Neg logit contributions
meg-3.657
star-3.626
mania-3.528
rovers-3.524
-3.514
Token54
Token ID4051
Feature activation-5.44e-01

Pos logit contributions
ibaba+0.747
emale+0.729
aimon+0.574
ijing+0.566
 WRITE+0.559
Neg logit contributions
meg-0.446
star-0.435
-0.420
power-0.417
pres-0.417
Token)
Token ID8
Feature activation+4.47e-01

Pos logit contributions
ibaba+2.136
emale+2.059
aimon+1.540
ijing+1.538
 WRITE+1.521
Neg logit contributions
meg-1.824
star-1.797
-1.752
rovers-1.749
mania-1.745
Token smartphone
Token ID11745
Feature activation-2.02e-01

Pos logit contributions
meg+1.383
star+1.355
power+1.311
pres+1.305
+1.302
Neg logit contributions
ibaba-1.891
emale-1.859
aimon-1.428
ijing-1.393
 WRITE-1.376
Token.
Token ID13
Feature activation+1.51e-02

Pos logit contributions
ibaba+0.992
emale+0.966
aimon+0.775
ijing+0.767
 WRITE+0.762
Neg logit contributions
meg-0.486
star-0.471
-0.456
pres-0.450
mania-0.449
Token Ol
Token ID6544
Feature activation-2.67e-01

Pos logit contributions
ibaba+0.379
emale+0.369
aimon+0.289
ijing+0.284
 WRITE+0.281
Neg logit contributions
meg-0.244
star-0.238
-0.230
power-0.229
pres-0.229
Tokenaf
Token ID1878
Feature activation-4.81e-01

Pos logit contributions
ibaba+1.402
emale+1.368
aimon+1.125
ijing+1.099
 WRITE+1.087
Neg logit contributions
meg-0.542
-0.530
star-0.525
power-0.509
mania-0.507
Token?"
Token ID1701
Feature activation+3.22e-01

Pos logit contributions
ibaba+2.552
emale+2.499
aimon+2.068
 mustache+2.025
ijing+2.023
Neg logit contributions
meg-0.933
star-0.875
-0.873
rovers-0.852
power-0.845
Token I
Token ID314
Feature activation+4.16e-01

Pos logit contributions
meg+0.842
star+0.819
+0.791
power+0.785
mania+0.784
Neg logit contributions
ibaba-1.518
emale-1.482
aimon-1.176
ijing-1.158
 WRITE-1.149
Token stood
Token ID6204
Feature activation+3.04e-01

Pos logit contributions
meg+1.141
star+1.115
+1.071
power+1.069
pres+1.068
Neg logit contributions
ibaba-1.906
emale-1.863
aimon-1.467
ijing-1.444
 WRITE-1.424
Token still
Token ID991
Feature activation+8.41e-01

Pos logit contributions
meg+0.806
star+0.782
+0.765
mania+0.752
pres+0.752
Neg logit contributions
ibaba-1.415
emale-1.381
aimon-1.094
ijing-1.079
 WRITE-1.073
Token.
Token ID13
Feature activation+2.11e-01

Pos logit contributions
meg+2.042
star+1.996
power+1.908
mania+1.904
pres+1.900
Neg logit contributions
ibaba-4.114
emale-4.031
aimon-3.225
ijing-3.178
 WRITE-3.133
Token
Token ID198
Feature activation+3.54e-01

Pos logit contributions
meg+0.565
star+0.551
+0.532
power+0.528
pres+0.528
Neg logit contributions
ibaba-0.978
emale-0.955
aimon-0.755
ijing-0.743
 WRITE-0.736
Token
Token ID198
Feature activation+2.96e-01

Pos logit contributions
meg+0.737
star+0.716
power+0.678
+0.677
mania+0.677
Neg logit contributions
ibaba-1.866
emale-1.834
aimon-1.487
ijing-1.472
 WRITE-1.452
TokenMy
Token ID3666
Feature activation-1.03e-01

Pos logit contributions
meg+0.568
star+0.544
+0.523
mania+0.512
rovers+0.510
Neg logit contributions
ibaba-1.594
emale-1.561
aimon-1.286
ijing-1.266
 WRITE-1.260
Token mom
Token ID1995
Feature activation+2.77e-01

Pos logit contributions
ibaba+0.409
emale+0.397
aimon+0.302
ijing+0.295
 mustache+0.293
Neg logit contributions
meg-0.340
star-0.332
-0.325
pres-0.322
mania-0.322
Token election
Token ID3071
Feature activation+5.59e-02

Pos logit contributions
ibaba+0.658
emale+0.638
aimon+0.492
ijing+0.487
 WRITE+0.483
Neg logit contributions
meg-0.472
star-0.460
-0.451
pres-0.445
mania-0.444
Token approaches
Token ID10581
Feature activation-8.59e-01

Pos logit contributions
meg+0.175
star+0.171
+0.166
power+0.165
mania+0.165
Neg logit contributions
ibaba-0.235
emale-0.229
aimon-0.175
ijing-0.172
 WRITE-0.170
Token.
Token ID13
Feature activation-3.50e-02

Pos logit contributions
ibaba+4.309
emale+4.195
aimon+3.416
ijing+3.391
 WRITE+3.381
Neg logit contributions
meg-1.869
star-1.821
-1.792
power-1.737
mania-1.737
Token For
Token ID1114
Feature activation+3.11e-01

Pos logit contributions
ibaba+0.161
emale+0.157
aimon+0.124
ijing+0.122
 WRITE+0.122
Neg logit contributions
meg-0.095
star-0.092
-0.089
power-0.088
pres-0.088
Token a
Token ID257
Feature activation+9.59e-01

Pos logit contributions
meg+0.901
star+0.880
+0.853
power+0.846
pres+0.846
Neg logit contributions
ibaba-1.375
emale-1.341
aimon-1.045
ijing-1.029
 WRITE-1.019
Token few
Token ID1178
Feature activation+8.97e-01

Pos logit contributions
meg+2.508
star+2.445
power+2.343
pres+2.338
mania+2.328
Neg logit contributions
ibaba-4.510
emale-4.410
aimon-3.514
ijing-3.442
 WRITE-3.413
Token years
Token ID812
Feature activation+8.06e-01

Pos logit contributions
meg+2.279
star+2.218
+2.123
pres+2.121
power+2.117
Neg logit contributions
ibaba-4.291
emale-4.190
aimon-3.350
ijing-3.277
 WRITE-3.259
Token,
Token ID11
Feature activation+2.56e-01

Pos logit contributions
meg+2.032
star+1.975
+1.906
power+1.889
mania+1.887
Neg logit contributions
ibaba-3.873
emale-3.781
aimon-3.019
ijing-2.979
 WRITE-2.950
Token the
Token ID262
Feature activation+1.05e-02

Pos logit contributions
meg+0.772
star+0.755
+0.733
power+0.727
mania+0.726
Neg logit contributions
ibaba-1.105
emale-1.076
aimon-0.833
ijing-0.821
 WRITE-0.813
Token economy
Token ID3773
Feature activation-2.67e-01

Pos logit contributions
meg+0.033
star+0.032
+0.031
power+0.031
mania+0.031
Neg logit contributions
ibaba-0.044
emale-0.043
aimon-0.033
ijing-0.032
 WRITE-0.032
Token was
Token ID373
Feature activation-2.96e-01

Pos logit contributions
ibaba+1.206
emale+1.169
aimon+0.924
ijing+0.920
 mustache+0.919
Neg logit contributions
meg-0.723
star-0.703
-0.695
rovers-0.679
pres-0.675
TokenI
Token ID40
Feature activation+1.61e-01

Pos logit contributions
ibaba+0.777
emale+0.761
aimon+0.608
ijing+0.599
 WRITE+0.595
Neg logit contributions
meg-0.409
star-0.395
-0.386
power-0.379
rovers-0.379
Token spent
Token ID3377
Feature activation+2.82e-01

Pos logit contributions
meg+0.400
star+0.388
+0.374
power+0.371
pres+0.371
Neg logit contributions
ibaba-0.779
emale-0.762
aimon-0.609
ijing-0.600
 WRITE-0.595
Token a
Token ID257
Feature activation+4.74e-01

Pos logit contributions
meg+0.777
star+0.758
+0.730
power+0.728
mania+0.727
Neg logit contributions
ibaba-1.292
emale-1.261
aimon-0.992
ijing-0.977
 WRITE-0.966
Token long
Token ID890
Feature activation+4.89e-01

Pos logit contributions
meg+1.254
star+1.220
+1.179
power+1.170
pres+1.169
Neg logit contributions
ibaba-2.219
emale-2.167
aimon-1.717
ijing-1.690
 WRITE-1.675
Token time
Token ID640
Feature activation+2.47e-01

Pos logit contributions
meg+1.048
star+1.013
+0.971
power+0.961
mania+0.961
Neg logit contributions
ibaba-2.531
emale-2.478
aimon-2.013
ijing-1.987
 WRITE-1.971
Token looking
Token ID2045
Feature activation+8.27e-01

Pos logit contributions
meg+0.738
star+0.720
+0.698
power+0.694
pres+0.693
Neg logit contributions
ibaba-1.075
emale-1.047
aimon-0.813
ijing-0.800
 WRITE-0.792
Token for
Token ID329
Feature activation-4.22e-01

Pos logit contributions
meg+2.020
star+1.956
+1.889
mania+1.868
pres+1.867
Neg logit contributions
ibaba-4.045
emale-3.949
aimon-3.164
ijing-3.128
 WRITE-3.099
Token what
Token ID644
Feature activation-8.82e-02

Pos logit contributions
ibaba+1.772
emale+1.721
aimon+1.320
ijing+1.308
 mustache+1.304
Neg logit contributions
meg-1.299
star-1.263
-1.250
pres-1.225
power-1.221
Token used
Token ID973
Feature activation-1.11e-01

Pos logit contributions
ibaba+0.403
emale+0.392
aimon+0.309
ijing+0.306
 mustache+0.303
Neg logit contributions
meg-0.243
star-0.235
-0.230
pres-0.226
power-0.225
Token to
Token ID284
Feature activation+1.46e-01

Pos logit contributions
ibaba+0.545
emale+0.533
aimon+0.428
ijing+0.421
 WRITE+0.418
Neg logit contributions
meg-0.270
star-0.262
-0.252
power-0.250
mania-0.250
Token be
Token ID307
Feature activation+2.64e-01

Pos logit contributions
meg+0.457
star+0.447
+0.442
mania+0.435
power+0.431
Neg logit contributions
ibaba-0.605
emale-0.592
aimon-0.452
ijing-0.445
 mustache-0.444
Token
Token ID198
Feature activation-2.56e-01

Pos logit contributions
ibaba+1.085
emale+1.049
aimon+0.860
 WRITE+0.848
ijing+0.841
Neg logit contributions
meg-0.498
star-0.478
-0.468
rovers-0.458
pres-0.455
TokenThe
Token ID464
Feature activation-1.77e-01

Pos logit contributions
ibaba+1.227
emale+1.197
aimon+0.972
 WRITE+0.946
ijing+0.945
Neg logit contributions
meg-0.614
star-0.590
-0.579
rovers-0.570
mania-0.566
Token trade
Token ID3292
Feature activation+6.03e-02

Pos logit contributions
ibaba+0.711
emale+0.689
aimon+0.525
ijing+0.513
 WRITE+0.512
Neg logit contributions
meg-0.580
star-0.565
-0.555
pres-0.549
mania-0.548
Token also
Token ID635
Feature activation+2.40e-01

Pos logit contributions
meg+0.166
star+0.161
+0.157
pres+0.155
power+0.154
Neg logit contributions
ibaba-0.276
emale-0.268
aimon-0.213
ijing-0.208
 WRITE-0.208
Token reun
Token ID15398
Feature activation+5.11e-01

Pos logit contributions
meg+0.669
star+0.653
+0.629
power+0.627
pres+0.627
Neg logit contributions
ibaba-1.090
emale-1.065
aimon-0.836
ijing-0.823
 WRITE-0.814
Tokenites
Token ID2737
Feature activation+7.09e-01

Pos logit contributions
meg+1.082
star+1.050
+0.994
power+0.994
pres+0.990
Neg logit contributions
ibaba-2.663
emale-2.607
aimon-2.120
ijing-2.093
 WRITE-2.077
Token Fel
Token ID13937
Feature activation+2.69e-01

Pos logit contributions
meg+2.136
star+2.097
power+2.016
mania+2.013
+2.012
Neg logit contributions
ibaba-3.052
emale-2.978
aimon-2.296
ijing-2.266
 WRITE-2.227
Tokenton
Token ID1122
Feature activation-2.28e-01

Pos logit contributions
meg+0.508
star+0.490
power+0.460
pres+0.460
mania+0.459
Neg logit contributions
ibaba-1.460
emale-1.432
aimon-1.174
ijing-1.162
 WRITE-1.153
Token and
Token ID290
Feature activation+1.46e-01

Pos logit contributions
ibaba+1.038
emale+1.009
aimon+0.801
ijing+0.781
 WRITE+0.773
Neg logit contributions
meg-0.624
star-0.602
-0.592
power-0.584
rovers-0.583
Token Thomas
Token ID5658
Feature activation-1.11e-01

Pos logit contributions
meg+0.434
star+0.421
+0.412
pres+0.407
power+0.406
Neg logit contributions
ibaba-0.633
emale-0.616
aimon-0.481
ijing-0.470
 WRITE-0.469
Token with
Token ID351
Feature activation+9.08e-02

Pos logit contributions
ibaba+0.520
emale+0.504
aimon+0.403
ijing+0.394
 WRITE+0.392
Neg logit contributions
meg-0.292
star-0.284
-0.275
pres-0.272
rovers-0.272
Token go
Token ID467
Feature activation+4.75e-02

Pos logit contributions
meg+0.635
star+0.618
+0.601
mania+0.592
power+0.591
Neg logit contributions
ibaba-1.170
emale-1.142
aimon-0.909
ijing-0.895
 WRITE-0.891
Token into
Token ID656
Feature activation+3.60e-01

Pos logit contributions
meg+0.122
star+0.119
+0.115
pres+0.114
mania+0.113
Neg logit contributions
ibaba-0.225
emale-0.219
aimon-0.175
ijing-0.173
 WRITE-0.171
Token the
Token ID262
Feature activation+3.70e-02

Pos logit contributions
meg+0.965
star+0.943
+0.912
pres+0.903
power+0.900
Neg logit contributions
ibaba-1.675
emale-1.631
aimon-1.293
ijing-1.273
 WRITE-1.260
Token details
Token ID3307
Feature activation+3.30e-01

Pos logit contributions
meg+0.110
star+0.108
+0.105
pres+0.104
mania+0.104
Neg logit contributions
ibaba-0.160
emale-0.156
aimon-0.121
ijing-0.119
 WRITE-0.118
Token of
Token ID286
Feature activation-1.06e-01

Pos logit contributions
meg+0.849
star+0.826
+0.794
mania+0.789
pres+0.789
Neg logit contributions
ibaba-1.565
emale-1.523
aimon-1.217
ijing-1.203
 WRITE-1.191
Token sound
Token ID2128
Feature activation+7.09e-01

Pos logit contributions
ibaba+0.457
emale+0.443
aimon+0.345
ijing+0.339
 WRITE+0.336
Neg logit contributions
meg-0.317
star-0.312
-0.306
pres-0.300
rovers-0.300
Token commercial
Token ID5068
Feature activation+1.15e-01

Pos logit contributions
meg+2.092
star+2.044
+1.976
power+1.969
mania+1.967
Neg logit contributions
ibaba-3.102
emale-3.026
aimon-2.352
ijing-2.310
 WRITE-2.291
Token banking
Token ID11754
Feature activation+2.25e-01

Pos logit contributions
meg+0.343
star+0.332
+0.329
pres+0.322
rovers+0.322
Neg logit contributions
ibaba-0.496
emale-0.478
aimon-0.371
ijing-0.368
 margins-0.364
Token,
Token ID11
Feature activation-1.95e-01

Pos logit contributions
meg+0.573
star+0.556
+0.542
pres+0.528
power+0.526
Neg logit contributions
ibaba-1.075
emale-1.045
aimon-0.836
ijing-0.829
 WRITE-0.817
Token how
Token ID703
Feature activation-3.55e-01

Pos logit contributions
ibaba+0.869
emale+0.843
aimon+0.659
ijing+0.653
 mustache+0.646
Neg logit contributions
meg-0.551
star-0.537
-0.528
mania-0.517
pres-0.516
Token it
Token ID340
Feature activation-2.12e-01

Pos logit contributions
ibaba+1.585
emale+1.525
aimon+1.201
ijing+1.196
 WRITE+1.188
Neg logit contributions
meg-0.995
star-0.960
-0.959
rovers-0.934
pres-0.928
Token Center
Token ID3337
Feature activation+1.48e-02

Pos logit contributions
meg+0.711
star+0.690
+0.677
mania+0.668
pres+0.667
Neg logit contributions
ibaba-0.901
emale-0.878
aimon-0.664
ijing-0.656
 WRITE-0.651
Token History
Token ID7443
Feature activation-3.19e-02

Pos logit contributions
meg+0.041
star+0.040
+0.039
pres+0.039
mania+0.038
Neg logit contributions
ibaba-0.066
emale-0.065
aimon-0.051
ijing-0.050
 WRITE-0.050
Token Office
Token ID4452
Feature activation+3.23e-01

Pos logit contributions
ibaba+0.140
emale+0.136
aimon+0.106
ijing+0.105
 WRITE+0.105
Neg logit contributions
meg-0.093
star-0.090
-0.088
mania-0.087
pres-0.086
Token,
Token ID11
Feature activation+3.43e-01

Pos logit contributions
meg+0.880
star+0.852
+0.829
pres+0.822
mania+0.822
Neg logit contributions
ibaba-1.483
emale-1.449
aimon-1.141
ijing-1.127
 WRITE-1.120
Token 1997
Token ID8309
Feature activation+4.85e-03

Pos logit contributions
meg+0.849
star+0.829
+0.794
power+0.790
pres+0.789
Neg logit contributions
ibaba-1.669
emale-1.630
aimon-1.303
ijing-1.284
 WRITE-1.268
Token.
Token ID13
Feature activation+1.56e-01

Pos logit contributions
meg+0.010
star+0.009
+0.009
mania+0.009
pres+0.009
Neg logit contributions
ibaba-0.026
emale-0.025
aimon-0.021
ijing-0.020
 WRITE-0.020
Token
Token ID198
Feature activation-2.58e-01

Pos logit contributions
meg+0.377
star+0.362
+0.351
mania+0.350
pres+0.348
Neg logit contributions
ibaba-0.769
emale-0.750
aimon-0.601
ijing-0.594
 WRITE-0.593
Token
Token ID198
Feature activation-3.19e-01

Pos logit contributions
ibaba+1.272
emale+1.228
aimon+1.006
 WRITE+0.994
ijing+0.985
Neg logit contributions
meg-0.595
star-0.571
-0.563
rovers-0.549
pres-0.545
Token
Token ID447
Feature activation-3.60e-01

Pos logit contributions
ibaba+1.479
emale+1.453
aimon+1.160
ijing+1.147
 WRITE+1.139
Neg logit contributions
meg-0.795
star-0.759
-0.750
mania-0.744
rovers-0.735
Token
Token ID250
Feature activation+3.57e-02

Pos logit contributions
ibaba+1.348
emale+1.289
ijing+0.952
aimon+0.950
 WRITE+0.934
Neg logit contributions
meg-1.287
star-1.257
-1.226
mania-1.225
pres-1.221
TokenB
Token ID33
Feature activation+2.68e-01

Pos logit contributions
meg+0.088
star+0.084
+0.082
mania+0.081
power+0.080
Neg logit contributions
ibaba-0.173
emale-0.169
aimon-0.135
ijing-0.134
 WRITE-0.132
Tokenent
Token ID298
Feature activation+1.64e-02

Pos logit contributions
meg+0.524
star+0.510
+0.491
mania+0.487
pres+0.486
Neg logit contributions
ibaba-1.053
emale-1.030
aimon-0.825
ijing-0.813
 WRITE-0.806
Token access
Token ID1895
Feature activation-1.89e-01

Pos logit contributions
meg+0.053
star+0.052
+0.050
power+0.050
pres+0.050
Neg logit contributions
ibaba-0.067
emale-0.066
aimon-0.050
ijing-0.049
 WRITE-0.049
Token to
Token ID284
Feature activation-2.61e-01

Pos logit contributions
ibaba+0.902
emale+0.869
aimon+0.701
ijing+0.694
 WRITE+0.680
Neg logit contributions
meg-0.474
star-0.468
-0.455
mania-0.447
power-0.442
Token this
Token ID428
Feature activation-2.14e-01

Pos logit contributions
ibaba+1.081
emale+1.046
aimon+0.802
ijing+0.798
 WRITE+0.794
Neg logit contributions
meg-0.810
star-0.796
-0.784
rovers-0.771
mania-0.768
Token illegal
Token ID5293
Feature activation+4.04e-01

Pos logit contributions
ibaba+0.840
emale+0.804
aimon+0.608
ijing+0.599
 WRITE+0.595
Neg logit contributions
meg-0.719
star-0.707
-0.694
mania-0.685
rovers-0.683
Token content
Token ID2695
Feature activation+6.14e-02

Pos logit contributions
meg+1.461
star+1.432
+1.389
power+1.389
pres+1.386
Neg logit contributions
ibaba-1.499
emale-1.463
aimon-1.076
ijing-1.053
 WRITE-1.035
Token.
Token ID13
Feature activation-9.32e-02

Pos logit contributions
meg+0.127
star+0.123
+0.118
mania+0.117
power+0.116
Neg logit contributions
ibaba-0.322
emale-0.312
aimon-0.257
ijing-0.255
 WRITE-0.251
Token
Token ID447
Feature activation+1.30e-01

Pos logit contributions
ibaba+0.439
emale+0.428
aimon+0.340
 WRITE+0.337
ijing+0.336
Neg logit contributions
meg-0.239
star-0.233
-0.225
mania-0.223
power-0.222
Token
Token ID251
Feature activation-1.53e-01

Pos logit contributions
meg+0.462
star+0.452
mania+0.440
+0.439
pres+0.439
Neg logit contributions
ibaba-0.496
emale-0.476
aimon-0.354
ijing-0.349
 WRITE-0.348
Token
Token ID198
Feature activation-2.33e-01

Pos logit contributions
ibaba+0.740
emale+0.719
aimon+0.576
 WRITE+0.573
ijing+0.571
Neg logit contributions
meg-0.375
star-0.364
-0.351
mania-0.349
rovers-0.347
Token
Token ID198
Feature activation-1.74e-01

Pos logit contributions
ibaba+1.151
emale+1.114
aimon+0.912
 WRITE+0.901
ijing+0.892
Neg logit contributions
meg-0.534
star-0.512
-0.503
rovers-0.492
pres-0.489
Token much
Token ID881
Feature activation-9.38e-02

Pos logit contributions
ibaba+0.203
emale+0.197
aimon+0.154
 mustache+0.151
ijing+0.151
Neg logit contributions
meg-0.138
star-0.135
-0.132
pres-0.131
power-0.130
Token as
Token ID355
Feature activation-2.76e-01

Pos logit contributions
ibaba+0.460
emale+0.447
aimon+0.362
ijing+0.356
 mustache+0.355
Neg logit contributions
meg-0.223
star-0.215
-0.211
pres-0.209
power-0.207
Token I
Token ID314
Feature activation-1.82e-01

Pos logit contributions
ibaba+1.312
emale+1.275
aimon+1.022
ijing+1.007
 WRITE+1.006
Neg logit contributions
meg-0.702
star-0.681
-0.665
pres-0.655
power-0.651
Token do
Token ID466
Feature activation-1.84e-01

Pos logit contributions
ibaba+0.718
emale+0.691
aimon+0.523
ijing+0.517
 WRITE+0.514
Neg logit contributions
meg-0.612
star-0.597
-0.588
pres-0.579
mania-0.575
Token,
Token ID11
Feature activation+1.74e-01

Pos logit contributions
ibaba+0.937
emale+0.914
aimon+0.742
ijing+0.734
 WRITE+0.730
Neg logit contributions
meg-0.404
star-0.390
-0.382
pres-0.378
power-0.373
Token but
Token ID475
Feature activation+6.70e-02

Pos logit contributions
meg+0.482
star+0.470
+0.455
power+0.451
mania+0.451
Neg logit contributions
ibaba-0.795
emale-0.775
aimon-0.610
ijing-0.600
 WRITE-0.595
Token I
Token ID314
Feature activation+9.76e-02

Pos logit contributions
meg+0.175
star+0.170
+0.164
pres+0.163
power+0.163
Neg logit contributions
ibaba-0.316
emale-0.309
aimon-0.245
ijing-0.241
 WRITE-0.239
Token think
Token ID892
Feature activation-2.19e-01

Pos logit contributions
meg+0.253
star+0.246
+0.239
pres+0.236
mania+0.236
Neg logit contributions
ibaba-0.461
emale-0.449
aimon-0.357
ijing-0.352
 WRITE-0.350
Token we
Token ID356
Feature activation-3.12e-02

Pos logit contributions
ibaba+1.028
emale+1.000
aimon+0.795
ijing+0.783
 mustache+0.776
Neg logit contributions
meg-0.577
star-0.559
-0.545
pres-0.540
mania-0.538
Token can
Token ID460
Feature activation-6.29e-02

Pos logit contributions
ibaba+0.144
emale+0.140
aimon+0.111
ijing+0.109
 WRITE+0.108
Neg logit contributions
meg-0.084
star-0.082
-0.080
pres-0.079
mania-0.079
Token win
Token ID1592
Feature activation+3.77e-01

Pos logit contributions
ibaba+0.263
emale+0.255
aimon+0.195
ijing+0.193
 WRITE+0.191
Neg logit contributions
meg-0.197
star-0.193
-0.188
pres-0.186
mania-0.186
Token can
Token ID460
Feature activation+1.63e-02

Pos logit contributions
meg+0.374
star+0.364
+0.351
power+0.349
pres+0.349
Neg logit contributions
ibaba-0.677
emale-0.662
aimon-0.525
ijing-0.518
 WRITE-0.513
Token use
Token ID779
Feature activation+9.70e-03

Pos logit contributions
meg+0.043
star+0.041
+0.040
mania+0.040
pres+0.040
Neg logit contributions
ibaba-0.077
emale-0.075
aimon-0.059
 WRITE-0.059
ijing-0.059
Token this
Token ID428
Feature activation+5.47e-02

Pos logit contributions
meg+0.026
star+0.025
+0.025
mania+0.024
pres+0.024
Neg logit contributions
ibaba-0.045
emale-0.043
aimon-0.034
 WRITE-0.034
ijing-0.034
Token on
Token ID319
Feature activation+1.56e-01

Pos logit contributions
meg+0.140
star+0.137
+0.133
mania+0.131
pres+0.130
Neg logit contributions
ibaba-0.259
emale-0.252
aimon-0.202
 WRITE-0.199
ijing-0.198
Token your
Token ID534
Feature activation+1.30e-01

Pos logit contributions
meg+0.483
star+0.471
+0.459
mania+0.455
pres+0.454
Neg logit contributions
ibaba-0.660
emale-0.642
aimon-0.494
ijing-0.486
 WRITE-0.483
Token opponent
Token ID6125
Feature activation+5.90e-01

Pos logit contributions
meg+0.469
star+0.459
+0.449
mania+0.445
power+0.445
Neg logit contributions
ibaba-0.479
emale-0.465
aimon-0.343
ijing-0.336
 WRITE-0.333
Token
Token ID447
Feature activation-2.00e-01

Pos logit contributions
meg+1.424
star+1.386
+1.327
power+1.324
pres+1.320
Neg logit contributions
ibaba-2.892
emale-2.831
aimon-2.270
ijing-2.236
 WRITE-2.216
Token
Token ID247
Feature activation-4.11e-02

Pos logit contributions
ibaba+0.878
emale+0.854
aimon+0.665
ijing+0.655
 WRITE+0.648
Neg logit contributions
meg-0.588
star-0.574
-0.556
mania-0.553
pres-0.552
Tokens
Token ID82
Feature activation-2.23e-02

Pos logit contributions
ibaba+0.240
emale+0.236
aimon+0.196
ijing+0.194
 WRITE+0.192
Neg logit contributions
meg-0.060
star-0.056
-0.055
mania-0.052
rovers-0.052
Token high
Token ID1029
Feature activation+6.84e-02

Pos logit contributions
ibaba+0.103
emale+0.101
aimon+0.080
ijing+0.078
 WRITE+0.078
Neg logit contributions
meg-0.059
star-0.058
-0.056
pres-0.055
mania-0.055
Token Level
Token ID5684
Feature activation+2.21e-02

Pos logit contributions
meg+0.230
star+0.225
+0.219
pres+0.218
mania+0.218
Neg logit contributions
ibaba-0.269
emale-0.262
aimon-0.198
ijing-0.194
 WRITE-0.193
Token have
Token ID423
Feature activation+3.19e-01

Pos logit contributions
meg+0.476
star+0.464
+0.447
power+0.447
mania+0.445
Neg logit contributions
ibaba-0.770
emale-0.752
aimon-0.589
ijing-0.580
 WRITE-0.573
Token to
Token ID284
Feature activation+2.26e-01

Pos logit contributions
meg+0.874
star+0.851
+0.824
pres+0.817
power+0.816
Neg logit contributions
ibaba-1.463
emale-1.426
aimon-1.125
ijing-1.108
 WRITE-1.097
Token learn
Token ID2193
Feature activation+2.25e-01

Pos logit contributions
meg+0.679
star+0.664
+0.641
power+0.641
pres+0.640
Neg logit contributions
ibaba-0.980
emale-0.956
aimon-0.740
ijing-0.727
 WRITE-0.718
Token to
Token ID284
Feature activation-3.99e-01

Pos logit contributions
meg+0.611
star+0.595
+0.575
power+0.572
pres+0.571
Neg logit contributions
ibaba-1.033
emale-1.010
aimon-0.796
ijing-0.783
 WRITE-0.776
Token overtake
Token ID48155
Feature activation+1.28e-01

Pos logit contributions
ibaba+1.680
emale+1.626
aimon+1.256
ijing+1.235
 WRITE+1.234
Neg logit contributions
meg-1.235
star-1.202
-1.182
mania-1.159
pres-1.159
Token.
Token ID13
Feature activation+2.95e-01

Pos logit contributions
meg+0.311
star+0.301
+0.293
pres+0.286
mania+0.286
Neg logit contributions
ibaba-0.626
emale-0.609
aimon-0.491
ijing-0.485
 WRITE-0.481
Token The
Token ID383
Feature activation+1.59e-01

Pos logit contributions
meg+0.810
star+0.797
mania+0.768
power+0.768
pres+0.761
Neg logit contributions
ibaba-1.359
emale-1.334
aimon-1.043
ijing-1.031
 mustache-1.010
Token race
Token ID3234
Feature activation-2.87e-01

Pos logit contributions
meg+0.491
star+0.481
power+0.465
mania+0.464
+0.463
Neg logit contributions
ibaba-0.671
emale-0.655
aimon-0.503
ijing-0.493
 WRITE-0.487
Token may
Token ID743
Feature activation+8.27e-02

Pos logit contributions
ibaba+1.216
emale+1.171
aimon+0.912
ijing+0.896
 WRITE+0.886
Neg logit contributions
meg-0.881
star-0.853
-0.847
rovers-0.826
pres-0.825
Token only
Token ID691
Feature activation-1.85e-01

Pos logit contributions
meg+0.239
star+0.232
+0.228
pres+0.224
mania+0.223
Neg logit contributions
ibaba-0.366
emale-0.356
aimon-0.279
ijing-0.274
 WRITE-0.273
Token be
Token ID307
Feature activation-1.76e-01

Pos logit contributions
ibaba+0.802
emale+0.782
aimon+0.608
ijing+0.597
 WRITE+0.592
Neg logit contributions
meg-0.548
star-0.535
-0.521
pres-0.515
mania-0.515
Token my
Token ID616
Feature activation-1.57e-01

Pos logit contributions
ibaba+1.644
emale+1.590
aimon+1.228
 WRITE+1.208
ijing+1.207
Neg logit contributions
meg-1.171
star-1.149
-1.131
pres-1.113
rovers-1.109
Token first
Token ID717
Feature activation+8.06e-02

Pos logit contributions
ibaba+0.711
emale+0.695
aimon+0.547
 WRITE+0.538
ijing+0.538
Neg logit contributions
meg-0.433
star-0.420
-0.412
power-0.408
pres-0.407
Token real
Token ID1103
Feature activation-1.29e-01

Pos logit contributions
meg+0.224
star+0.218
+0.213
pres+0.210
power+0.210
Neg logit contributions
ibaba-0.366
emale-0.357
aimon-0.280
ijing-0.276
 WRITE-0.275
Token session
Token ID6246
Feature activation-2.73e-01

Pos logit contributions
ibaba+0.563
emale+0.548
aimon+0.427
ijing+0.420
 WRITE+0.417
Neg logit contributions
meg-0.381
star-0.371
-0.362
pres-0.359
power-0.359
Token with
Token ID351
Feature activation-2.03e-01

Pos logit contributions
ibaba+1.272
emale+1.237
aimon+0.984
ijing+0.971
 WRITE+0.967
Neg logit contributions
meg-0.720
star-0.695
-0.684
pres-0.673
rovers-0.668
Token it
Token ID340
Feature activation-8.21e-02

Pos logit contributions
ibaba+0.902
emale+0.880
aimon+0.686
ijing+0.676
 WRITE+0.671
Neg logit contributions
meg-0.584
star-0.569
-0.552
pres-0.549
power-0.548
Token was
Token ID373
Feature activation-2.30e-01

Pos logit contributions
ibaba+0.391
emale+0.381
aimon+0.304
ijing+0.300
 WRITE+0.298
Neg logit contributions
meg-0.208
star-0.201
-0.197
pres-0.193
mania-0.193
Token
Token ID564
Feature activation-3.37e-01

Pos logit contributions
ibaba+1.033
emale+1.003
aimon+0.788
ijing+0.781
 WRITE+0.778
Neg logit contributions
meg-0.646
star-0.628
-0.609
mania-0.607
pres-0.606
Token
Token ID250
Feature activation-2.83e-01

Pos logit contributions
ibaba+1.262
emale+1.220
aimon+0.891
ijing+0.890
 WRITE+0.875
Neg logit contributions
meg-1.197
star-1.176
-1.144
mania-1.143
pres-1.140
Tokendec
Token ID12501
Feature activation-1.29e-01

Pos logit contributions
ibaba+1.386
emale+1.359
aimon+1.082
ijing+1.071
 WRITE+1.058
Neg logit contributions
meg-0.679
star-0.656
-0.641
mania-0.629
rovers-0.625
Tokenided
Token ID1384
Feature activation+2.87e-02

Pos logit contributions
ibaba+0.533
emale+0.519
aimon+0.397
ijing+0.390
 WRITE+0.386
Neg logit contributions
meg-0.408
star-0.399
-0.388
power-0.385
pres-0.385
Token that
Token ID326
Feature activation+1.01e-01

Pos logit contributions
meg+0.430
star+0.420
+0.406
power+0.403
pres+0.403
Neg logit contributions
ibaba-0.694
emale-0.678
aimon-0.531
ijing-0.523
 WRITE-0.518
Token self
Token ID2116
Feature activation+1.83e-01

Pos logit contributions
meg+0.281
star+0.274
+0.265
mania+0.263
power+0.263
Neg logit contributions
ibaba-0.461
emale-0.449
aimon-0.354
ijing-0.348
 WRITE-0.345
Token-
Token ID12
Feature activation+1.25e-02

Pos logit contributions
meg+0.393
star+0.366
+0.365
rovers+0.352
mania+0.351
Neg logit contributions
ibaba-0.945
emale-0.921
aimon-0.752
 mustache-0.748
ijing-0.742
Tokenenc
Token ID12685
Feature activation-4.13e-01

Pos logit contributions
meg+0.028
star+0.027
+0.027
mania+0.026
power+0.025
Neg logit contributions
ibaba-0.063
emale-0.062
aimon-0.050
ijing-0.049
 mustache-0.049
Tokenour
Token ID454
Feature activation-2.00e-01

Pos logit contributions
ibaba+1.788
emale+1.751
aimon+1.379
ijing+1.330
 WRITE+1.304
Neg logit contributions
meg-1.209
-1.183
star-1.163
pres-1.125
power-1.124
Tokenagement
Token ID5082
Feature activation-2.19e-01

Pos logit contributions
ibaba+0.871
emale+0.853
aimon+0.665
ijing+0.653
 mustache+0.644
Neg logit contributions
meg-0.579
star-0.560
-0.554
mania-0.546
pres-0.545
Token,
Token ID11
Feature activation-1.51e-01

Pos logit contributions
ibaba+0.983
emale+0.955
aimon+0.755
ijing+0.739
 WRITE+0.736
Neg logit contributions
meg-0.617
star-0.602
-0.583
mania-0.579
power-0.579
Token he
Token ID339
Feature activation-2.04e-03

Pos logit contributions
ibaba+0.679
emale+0.662
aimon+0.520
 WRITE+0.510
ijing+0.509
Neg logit contributions
meg-0.424
star-0.410
-0.400
mania-0.396
power-0.395
Token managed
Token ID5257
Feature activation+2.42e-01

Pos logit contributions
ibaba+0.009
emale+0.009
aimon+0.007
ijing+0.007
 WRITE+0.007
Neg logit contributions
meg-0.006
star-0.006
-0.006
power-0.006
pres-0.006
Token to
Token ID284
Feature activation+2.07e-01

Pos logit contributions
meg+0.603
star+0.586
mania+0.564
+0.564
power+0.561
Neg logit contributions
ibaba-1.171
emale-1.143
aimon-0.913
ijing-0.899
 WRITE-0.894
Token control
Token ID1630
Feature activation+2.74e-01

Pos logit contributions
meg+0.666
star+0.652
+0.630
pres+0.630
power+0.630
Neg logit contributions
ibaba-0.855
emale-0.834
aimon-0.636
ijing-0.625
 WRITE-0.612
Token to
Token ID284
Feature activation+1.43e-01

Pos logit contributions
meg+2.009
star+1.949
+1.878
power+1.864
mania+1.864
Neg logit contributions
ibaba-3.897
emale-3.806
aimon-3.051
ijing-3.005
 WRITE-2.971
Token schizophrenia
Token ID22794
Feature activation+2.55e-01

Pos logit contributions
meg+0.385
star+0.375
+0.364
power+0.359
pres+0.359
Neg logit contributions
ibaba-0.659
emale-0.643
aimon-0.511
ijing-0.502
 WRITE-0.498
Token,
Token ID11
Feature activation-1.69e-01

Pos logit contributions
meg+0.712
star+0.692
+0.676
pres+0.663
rovers+0.663
Neg logit contributions
ibaba-1.153
emale-1.119
aimon-0.884
ijing-0.874
 WRITE-0.863
Token they
Token ID484
Feature activation-3.34e-02

Pos logit contributions
ibaba+0.763
emale+0.741
aimon+0.588
 WRITE+0.576
ijing+0.576
Neg logit contributions
meg-0.469
star-0.455
-0.448
power-0.436
mania-0.435
Token disagree
Token ID12546
Feature activation+5.82e-02

Pos logit contributions
ibaba+0.151
emale+0.147
aimon+0.116
ijing+0.115
 WRITE+0.114
Neg logit contributions
meg-0.093
star-0.090
-0.089
mania-0.087
power-0.086
Token about
Token ID546
Feature activation-3.18e-01

Pos logit contributions
meg+0.157
star+0.153
+0.149
mania+0.147
power+0.145
Neg logit contributions
ibaba-0.265
emale-0.260
aimon-0.207
ijing-0.205
 mustache-0.203
Token exactly
Token ID3446
Feature activation+2.55e-01

Pos logit contributions
ibaba+1.310
emale+1.265
aimon+0.984
ijing+0.962
 mustache+0.957
Neg logit contributions
meg-1.006
star-0.985
-0.963
mania-0.949
power-0.946
Token how
Token ID703
Feature activation-7.65e-02

Pos logit contributions
meg+0.700
star+0.680
+0.659
mania+0.654
power+0.651
Neg logit contributions
ibaba-1.168
emale-1.134
aimon-0.897
ijing-0.884
 WRITE-0.878
Token
Token ID960
Feature activation+3.73e-01

Pos logit contributions
ibaba+0.348
emale+0.334
aimon+0.270
 mustache+0.266
ijing+0.266
Neg logit contributions
meg-0.208
star-0.201
-0.199
rovers-0.193
power-0.192
Tokenit
Token ID270
Feature activation-4.16e-02

Pos logit contributions
meg+1.002
star+0.974
+0.946
mania+0.936
power+0.933
Neg logit contributions
ibaba-1.726
emale-1.685
aimon-1.331
ijing-1.313
 WRITE-1.298
Token's
Token ID338
Feature activation+1.17e-01

Pos logit contributions
ibaba+0.195
emale+0.187
ijing+0.151
aimon+0.151
 mustache+0.148
Neg logit contributions
meg-0.109
star-0.104
-0.102
rovers-0.100
mania-0.099
Token common
Token ID2219
Feature activation-7.69e-02

Pos logit contributions
ibaba+0.821
emale+0.792
aimon+0.622
ijing+0.612
 WRITE+0.611
Neg logit contributions
meg-0.542
star-0.529
-0.515
pres-0.510
mania-0.509
Token sense
Token ID2565
Feature activation+1.11e-01

Pos logit contributions
ibaba+0.282
emale+0.273
aimon+0.200
ijing+0.196
 WRITE+0.195
Neg logit contributions
meg-0.281
star-0.275
-0.268
mania-0.267
pres-0.267
Token enough
Token ID1576
Feature activation-4.06e-01

Pos logit contributions
meg+0.262
star+0.255
+0.245
pres+0.243
mania+0.243
Neg logit contributions
ibaba-0.551
emale-0.537
aimon-0.434
ijing-0.428
 WRITE-0.426
Token to
Token ID284
Feature activation+3.20e-01

Pos logit contributions
ibaba+1.915
emale+1.841
aimon+1.494
 WRITE+1.478
ijing+1.472
Neg logit contributions
meg-1.035
star-0.999
-0.969
mania-0.967
pres-0.959
Token get
Token ID651
Feature activation-6.90e-02

Pos logit contributions
meg+0.929
star+0.907
power+0.876
pres+0.875
+0.873
Neg logit contributions
ibaba-1.415
emale-1.384
aimon-1.076
ijing-1.059
 WRITE-1.038
Token around
Token ID1088
Feature activation-2.03e-01

Pos logit contributions
ibaba+0.320
emale+0.311
aimon+0.247
ijing+0.243
 WRITE+0.242
Neg logit contributions
meg-0.185
star-0.180
-0.175
power-0.173
pres-0.173
Token our
Token ID674
Feature activation-2.05e-01

Pos logit contributions
ibaba+0.999
emale+0.970
aimon+0.791
ijing+0.777
 WRITE+0.773
Neg logit contributions
meg-0.480
star-0.467
-0.451
power-0.446
pres-0.445
Token own
Token ID898
Feature activation-8.26e-02

Pos logit contributions
ibaba+0.869
emale+0.846
aimon+0.660
ijing+0.648
 WRITE+0.646
Neg logit contributions
meg-0.620
star-0.606
-0.590
mania-0.586
power-0.583
Token space
Token ID2272
Feature activation+1.07e-01

Pos logit contributions
ibaba+0.342
emale+0.332
aimon+0.258
ijing+0.254
 WRITE+0.253
Neg logit contributions
meg-0.257
star-0.251
-0.246
mania-0.243
pres-0.242
Token without
Token ID1231
Feature activation-3.32e-02

Pos logit contributions
meg+0.237
star+0.230
+0.222
power+0.219
mania+0.218
Neg logit contributions
ibaba-0.548
emale-0.534
aimon-0.435
ijing-0.428
 WRITE-0.427
Token man
Token ID582
Feature activation-1.20e-01

Pos logit contributions
ibaba+0.138
emale+0.134
aimon+0.104
ijing+0.102
 WRITE+0.101
Neg logit contributions
meg-0.104
star-0.102
-0.099
pres-0.098
mania-0.098
Token for
Token ID329
Feature activation-2.40e-01

Pos logit contributions
meg+0.601
star+0.583
+0.562
mania+0.560
pres+0.557
Neg logit contributions
ibaba-1.358
emale-1.318
aimon-1.063
ijing-1.058
 WRITE-1.048
Token an
Token ID281
Feature activation+2.00e-02

Pos logit contributions
ibaba+0.958
emale+0.923
aimon+0.698
ijing+0.691
 WRITE+0.686
Neg logit contributions
meg-0.798
star-0.780
mania-0.760
-0.760
pres-0.754
Token equitable
Token ID37164
Feature activation+3.25e-02

Pos logit contributions
meg+0.074
star+0.073
+0.071
pres+0.071
power+0.070
Neg logit contributions
ibaba-0.072
emale-0.070
aimon-0.051
ijing-0.050
 WRITE-0.049
Token future
Token ID2003
Feature activation-1.97e-02

Pos logit contributions
meg+0.102
star+0.099
+0.096
power+0.096
pres+0.096
Neg logit contributions
ibaba-0.136
emale-0.133
aimon-0.102
ijing-0.100
 WRITE-0.099
Token
Token ID784
Feature activation+1.67e-01

Pos logit contributions
ibaba+0.102
emale+0.099
aimon+0.081
ijing+0.080
 WRITE+0.080
Neg logit contributions
meg-0.042
star-0.041
-0.039
mania-0.039
pres-0.038
Token they
Token ID484
Feature activation-3.54e-01

Pos logit contributions
meg+0.399
star+0.387
+0.372
mania+0.369
power+0.369
Neg logit contributions
ibaba-0.822
emale-0.802
aimon-0.644
ijing-0.636
 WRITE-0.630
Token are
Token ID389
Feature activation-2.40e-01

Pos logit contributions
ibaba+1.566
emale+1.514
ijing+1.174
aimon+1.165
 WRITE+1.159
Neg logit contributions
meg-1.024
star-0.997
-0.969
mania-0.959
rovers-0.953
Token the
Token ID262
Feature activation+7.28e-03

Pos logit contributions
ibaba+1.033
emale+1.001
aimon+0.775
ijing+0.768
 WRITE+0.755
Neg logit contributions
meg-0.725
star-0.706
-0.688
mania-0.683
pres-0.680
Token enemy
Token ID4472
Feature activation+1.97e-01

Pos logit contributions
meg+0.023
star+0.022
+0.021
power+0.021
pres+0.021
Neg logit contributions
ibaba-0.031
emale-0.030
aimon-0.023
ijing-0.023
 WRITE-0.022
Token.
Token ID13
Feature activation+6.86e-02

Pos logit contributions
meg+0.439
star+0.424
+0.410
mania+0.407
pres+0.405
Neg logit contributions
ibaba-1.003
emale-0.972
aimon-0.791
ijing-0.782
 WRITE-0.775
Token
Token ID198
Feature activation+1.61e-01

Pos logit contributions
meg+0.171
star+0.166
+0.160
mania+0.159
power+0.159
Neg logit contributions
ibaba-0.331
emale-0.323
aimon-0.258
ijing-0.254
 WRITE-0.253
Token home
Token ID1363
Feature activation-9.63e-02

Pos logit contributions
ibaba+1.905
emale+1.866
aimon+1.450
 mustache+1.429
ijing+1.427
Neg logit contributions
meg-1.286
star-1.254
-1.231
pres-1.210
mania-1.209
Token-
Token ID12
Feature activation-5.46e-01

Pos logit contributions
ibaba+0.472
emale+0.460
aimon+0.371
ijing+0.367
 WRITE+0.361
Neg logit contributions
meg-0.229
star-0.222
-0.216
mania-0.214
pres-0.212
Tokengrown
Token ID22377
Feature activation-3.09e-01

Pos logit contributions
ibaba+2.509
emale+2.449
aimon+1.956
ijing+1.919
 WRITE+1.881
Neg logit contributions
meg-1.409
star-1.368
-1.361
mania-1.339
rovers-1.316
Token Pent
Token ID9696
Feature activation+2.43e-02

Pos logit contributions
ibaba+1.311
emale+1.281
aimon+0.993
 mustache+0.976
ijing+0.974
Neg logit contributions
meg-0.930
star-0.909
-0.893
pres-0.880
mania-0.878
Tokenec
Token ID721
Feature activation-4.90e-01

Pos logit contributions
meg+0.073
star+0.071
+0.069
mania+0.069
power+0.069
Neg logit contributions
ibaba-0.104
emale-0.102
aimon-0.079
ijing-0.078
 WRITE-0.077
Tokenost
Token ID455
Feature activation-8.74e-01

Pos logit contributions
ibaba+2.642
emale+2.595
aimon+2.143
ijing+2.100
 WRITE+2.082
Neg logit contributions
meg-0.916
-0.877
star-0.861
mania-0.833
pres-0.826
Tokenal
Token ID282
Feature activation+1.66e-02

Pos logit contributions
ibaba+3.845
emale+3.807
aimon+3.039
ijing+2.956
 mustache+2.870
Neg logit contributions
meg-2.389
star-2.305
-2.291
pres-2.272
mania-2.254
Token group
Token ID1448
Feature activation+1.12e-01

Pos logit contributions
meg+0.054
star+0.053
+0.052
pres+0.051
mania+0.051
Neg logit contributions
ibaba-0.067
emale-0.066
aimon-0.050
ijing-0.049
 WRITE-0.048
Token,
Token ID11
Feature activation-1.54e-01

Pos logit contributions
meg+0.278
star+0.272
+0.263
mania+0.260
pres+0.260
Neg logit contributions
ibaba-0.538
emale-0.525
aimon-0.421
ijing-0.415
 WRITE-0.410
Token the
Token ID262
Feature activation-1.04e-01

Pos logit contributions
ibaba+0.661
emale+0.646
aimon+0.500
ijing+0.492
 WRITE+0.489
Neg logit contributions
meg-0.460
star-0.449
-0.436
power-0.434
mania-0.433
Token new
Token ID649
Feature activation-5.27e-01

Pos logit contributions
ibaba+0.444
emale+0.434
aimon+0.335
ijing+0.330
 WRITE+0.327
Neg logit contributions
meg-0.315
star-0.308
-0.300
power-0.297
mania-0.297
Token career
Token ID3451
Feature activation-3.27e-01

Pos logit contributions
meg+0.329
star+0.322
+0.312
pres+0.310
power+0.310
Neg logit contributions
ibaba-0.452
emale-0.441
aimon-0.339
ijing-0.334
 WRITE-0.329
Token out
Token ID503
Feature activation+1.73e-01

Pos logit contributions
ibaba+1.469
emale+1.427
aimon+1.129
 mustache+1.119
ijing+1.110
Neg logit contributions
meg-0.906
star-0.873
-0.864
pres-0.849
power-0.846
Token of
Token ID286
Feature activation+6.70e-01

Pos logit contributions
meg+0.420
star+0.408
+0.393
power+0.389
pres+0.389
Neg logit contributions
ibaba-0.844
emale-0.826
aimon-0.661
ijing-0.652
 WRITE-0.646
Token playing
Token ID2712
Feature activation-9.43e-02

Pos logit contributions
meg+2.053
star+2.010
pres+1.936
power+1.933
+1.932
Neg logit contributions
ibaba-2.869
emale-2.788
aimon-2.147
ijing-2.114
 WRITE-2.076
Token un
Token ID555
Feature activation+5.18e-02

Pos logit contributions
ibaba+0.403
emale+0.393
aimon+0.303
 WRITE+0.300
ijing+0.299
Neg logit contributions
meg-0.283
star-0.276
-0.271
mania-0.267
pres-0.267
Tokenhing
Token ID722
Feature activation-8.06e-01

Pos logit contributions
meg+0.125
star+0.120
+0.119
mania+0.116
power+0.115
Neg logit contributions
ibaba-0.252
emale-0.248
aimon-0.199
ijing-0.196
 WRITE-0.192
Tokened
Token ID276
Feature activation-3.45e-01

Pos logit contributions
ibaba+4.197
emale+4.111
aimon+3.340
ijing+3.312
 WRITE+3.281
Neg logit contributions
meg-1.674
star-1.613
-1.562
mania-1.542
power-1.529
Token lun
Token ID14678
Feature activation+6.88e-02

Pos logit contributions
ibaba+1.465
emale+1.435
aimon+1.109
 mustache+1.097
 WRITE+1.096
Neg logit contributions
meg-1.029
star-1.001
-0.984
mania-0.971
pres-0.968
Tokenatics
Token ID23372
Feature activation+1.90e-01

Pos logit contributions
meg+0.167
star+0.160
+0.159
mania+0.157
pres+0.155
Neg logit contributions
ibaba-0.335
emale-0.327
aimon-0.265
ijing-0.260
 WRITE-0.256
Token.
Token ID13
Feature activation-2.56e-01

Pos logit contributions
meg+0.464
star+0.451
+0.437
pres+0.432
mania+0.430
Neg logit contributions
ibaba-0.917
emale-0.895
aimon-0.723
ijing-0.713
 mustache-0.708
Token Maybe
Token ID6674
Feature activation-2.76e-01

Pos logit contributions
ibaba+1.149
emale+1.122
aimon+0.879
 WRITE+0.871
ijing+0.865
Neg logit contributions
meg-0.718
star-0.694
-0.678
power-0.670
mania-0.670
Tokenagh
Token ID10471
Feature activation+1.77e-01

Pos logit contributions
ibaba+1.204
emale+1.154
aimon+0.884
ijing+0.845
 WRITE+0.835
Neg logit contributions
meg-1.093
-1.070
pres-1.040
star-1.038
mania-1.030
Token Kil
Token ID19902
Feature activation-4.03e-01

Pos logit contributions
meg+0.484
star+0.469
+0.460
pres+0.455
power+0.453
Neg logit contributions
ibaba-0.803
emale-0.788
aimon-0.615
 WRITE-0.609
ijing-0.604
Tokenc
Token ID66
Feature activation-1.77e-02

Pos logit contributions
ibaba+1.898
emale+1.829
aimon+1.478
ijing+1.452
 WRITE+1.433
Neg logit contributions
meg-1.030
-1.010
star-0.991
rovers-0.965
mania-0.962
Tokenoy
Token ID726
Feature activation-5.29e-02

Pos logit contributions
ibaba+0.072
emale+0.070
aimon+0.054
ijing+0.053
 WRITE+0.052
Neg logit contributions
meg-0.058
star-0.056
-0.055
pres-0.054
mania-0.054
Tokenne
Token ID710
Feature activation+5.06e-02

Pos logit contributions
ibaba+0.270
emale+0.265
aimon+0.214
ijing+0.211
 WRITE+0.210
Neg logit contributions
meg-0.115
star-0.112
-0.108
mania-0.107
pres-0.107
Token
Token ID198
Feature activation-8.70e-01

Pos logit contributions
meg+0.121
star+0.118
+0.114
pres+0.112
mania+0.112
Neg logit contributions
ibaba-0.249
emale-0.243
aimon-0.195
ijing-0.193
 WRITE-0.192
Token
Token ID198
Feature activation-7.06e-01

Pos logit contributions
ibaba+4.043
emale+3.814
aimon+3.167
 WRITE+3.159
 margins+3.063
Neg logit contributions
meg-2.150
-2.064
star-2.054
rovers-2.030
pres-1.973
TokenA
Token ID32
Feature activation-2.26e-01

Pos logit contributions
ibaba+3.257
emale+3.172
aimon+2.558
ijing+2.536
 WRITE+2.514
Neg logit contributions
meg-1.712
star-1.661
-1.627
rovers-1.617
mania-1.591
Token swath
Token ID50176
Feature activation-6.77e-01

Pos logit contributions
ibaba+0.978
emale+0.948
aimon+0.732
ijing+0.726
 WRITE+0.716
Neg logit contributions
meg-0.674
star-0.657
-0.649
mania-0.640
pres-0.635
Token of
Token ID286
Feature activation+2.46e-02

Pos logit contributions
ibaba+3.014
emale+2.924
ijing+2.297
 mustache+2.285
aimon+2.277
Neg logit contributions
meg-1.847
-1.791
star-1.788
pres-1.746
sim-1.734
Token secondary
Token ID9233
Feature activation-4.53e-02

Pos logit contributions
meg+0.080
star+0.078
+0.076
pres+0.075
power+0.075
Neg logit contributions
ibaba-0.100
emale-0.097
aimon-0.074
ijing-0.073
 WRITE-0.072
Token."
Token ID526
Feature activation-8.29e-02

Pos logit contributions
ibaba+1.427
emale+1.397
aimon+1.153
 WRITE+1.132
ijing+1.126
Neg logit contributions
meg-0.572
-0.543
star-0.538
rovers-0.534
mania-0.528
Token My
Token ID2011
Feature activation-7.55e-02

Pos logit contributions
ibaba+0.390
emale+0.380
aimon+0.302
 WRITE+0.298
ijing+0.296
Neg logit contributions
meg-0.217
star-0.209
-0.205
mania-0.201
power-0.201
Token eyes
Token ID2951
Feature activation-3.84e-01

Pos logit contributions
ibaba+0.291
emale+0.282
aimon+0.212
ijing+0.207
 WRITE+0.206
Neg logit contributions
meg-0.260
star-0.255
-0.249
mania-0.247
power-0.247
Token are
Token ID389
Feature activation-5.24e-01

Pos logit contributions
ibaba+1.493
emale+1.445
aimon+1.105
 WRITE+1.099
 mustache+1.096
Neg logit contributions
meg-1.272
-1.234
star-1.227
pres-1.219
mania-1.207
Token still
Token ID991
Feature activation-4.08e-01

Pos logit contributions
ibaba+2.043
emale+1.980
aimon+1.514
 mustache+1.495
ijing+1.494
Neg logit contributions
meg-1.742
star-1.674
-1.662
pres-1.655
rovers-1.653
Token wide
Token ID3094
Feature activation-7.85e-01

Pos logit contributions
ibaba+1.487
emale+1.446
aimon+1.079
ijing+1.060
 mustache+1.058
Neg logit contributions
meg-1.462
-1.408
star-1.400
pres-1.396
mania-1.394
Token open
Token ID1280
Feature activation-2.75e-01

Pos logit contributions
ibaba+3.003
emale+2.934
 mustache+2.282
 WRITE+2.249
aimon+2.235
Neg logit contributions
meg-2.559
-2.507
rovers-2.465
mania-2.440
pres-2.429
Token.
Token ID13
Feature activation-6.67e-02

Pos logit contributions
ibaba+1.319
emale+1.278
aimon+1.035
 mustache+1.033
 WRITE+1.025
Neg logit contributions
meg-0.673
star-0.646
-0.643
rovers-0.631
mania-0.628
Token I
Token ID314
Feature activation+5.26e-02

Pos logit contributions
ibaba+0.306
emale+0.299
aimon+0.237
 WRITE+0.233
ijing+0.232
Neg logit contributions
meg-0.181
star-0.175
-0.171
mania-0.168
power-0.168
Token wasn
Token ID2492
Feature activation-1.88e-02

Pos logit contributions
meg+0.142
star+0.138
+0.134
pres+0.133
mania+0.133
Neg logit contributions
ibaba-0.242
emale-0.236
aimon-0.187
ijing-0.184
 WRITE-0.183
Token't
Token ID470
Feature activation-1.21e-01

Pos logit contributions
ibaba+0.098
emale+0.095
aimon+0.078
ijing+0.077
 WRITE+0.077
Neg logit contributions
meg-0.039
star-0.038
-0.036
pres-0.036
power-0.036
Token who
Token ID508
Feature activation-2.63e-01

Pos logit contributions
ibaba+2.318
emale+2.257
aimon+1.807
ijing+1.784
 mustache+1.779
Neg logit contributions
meg-1.331
star-1.280
-1.261
rovers-1.240
power-1.236
Token was
Token ID373
Feature activation-3.02e-01

Pos logit contributions
ibaba+1.185
emale+1.158
aimon+0.911
ijing+0.902
 mustache+0.902
Neg logit contributions
meg-0.728
star-0.706
-0.690
power-0.682
rovers-0.677
Token hit
Token ID2277
Feature activation-3.42e-01

Pos logit contributions
ibaba+1.296
emale+1.266
aimon+0.979
ijing+0.965
 WRITE+0.954
Neg logit contributions
meg-0.910
star-0.885
-0.864
power-0.854
mania-0.852
Token by
Token ID416
Feature activation-7.27e-02

Pos logit contributions
ibaba+1.582
emale+1.557
aimon+1.226
ijing+1.211
 WRITE+1.211
Neg logit contributions
meg-0.892
star-0.857
-0.851
rovers-0.831
pres-0.828
Token sh
Token ID427
Feature activation+8.72e-02

Pos logit contributions
ibaba+0.306
emale+0.297
aimon+0.230
ijing+0.225
 WRITE+0.224
Neg logit contributions
meg-0.225
star-0.220
-0.216
power-0.213
pres-0.212
Tokenrapnel
Token ID48766
Feature activation-9.30e-01

Pos logit contributions
meg+0.309
star+0.304
+0.299
pres+0.294
mania+0.293
Neg logit contributions
ibaba-0.329
emale-0.319
aimon-0.237
ijing-0.232
 WRITE-0.227
Token in
Token ID287
Feature activation+5.12e-02

Pos logit contributions
ibaba+4.226
emale+4.101
 mustache+3.381
aimon+3.349
 buckle+3.320
Neg logit contributions
meg-2.301
-2.241
star-2.222
rovers-2.136
pres-2.110
Token World
Token ID2159
Feature activation-2.06e-01

Pos logit contributions
meg+0.145
star+0.141
+0.137
power+0.136
pres+0.136
Neg logit contributions
ibaba-0.230
emale-0.223
aimon-0.175
ijing-0.173
 WRITE-0.171
Token War
Token ID1810
Feature activation-5.51e-01

Pos logit contributions
ibaba+1.233
emale+1.204
aimon+1.015
ijing+1.004
 WRITE+0.995
Neg logit contributions
meg-0.269
star-0.251
-0.241
pres-0.231
mania-0.230
Token II
Token ID2873
Feature activation-3.10e-01

Pos logit contributions
ibaba+2.765
emale+2.671
aimon+2.162
ijing+2.148
 WRITE+2.132
Neg logit contributions
meg-1.254
star-1.206
-1.193
pres-1.151
power-1.146
Token;
Token ID26
Feature activation+1.42e-01

Pos logit contributions
ibaba+1.518
emale+1.473
aimon+1.187
 mustache+1.183
ijing+1.176
Neg logit contributions
meg-0.739
star-0.706
-0.696
mania-0.678
power-0.675