TOP ACTIVATIONS
MAX = 1.826

Tony Shal h oub 2003 Larry David
Matt Le Bl anc 2003 Bernie Mac
2003 Larry David 2003 Matt Le Bl anc
2002 Bernie Mac 2002 Matthew Perry 2003 Tony
2003 Bernie Mac 2003 Eric McCorm ack
2003 Eric McCorm ack 2003 Ray Rom ano 2004
Kel sey Gram mer 2002 Matt Le Bl anc
Matt Le Bl anc 2002 Bernie Mac
Kel sey Gram mer 2004 Larry David
2002 Ray Rom ano 2002 Kel sey Gram mer
Draw ing . Color ] :: From Ar gb ($ FC
. Maybe their calls will go un he eded certainly
Draw ing . Color ] :: From Ar gb ($ FC
we move into hybrids and electric cars , gas tax revenues
Matt Le Bl anc 2004 John R itter
2003 Ray Rom ano 2004 Kel sey Gram mer
for oil and red for natural gas ." Jan Re es
2003 Eric McCorm ack 2003 Ray Rom
2004 Larry David 2004 Matt Le Bl anc
on the rifle . The Model of 1903 used tang ent

MIN ACTIVATIONS
MIN = -1.618

in white and is now Gand alf the White , for
, Ar ag orn , Gand alf , Gim li the
. There they unexpectedly meet Gand alf . Gand alf explains
, with the advice of Gand alf , chooses companions for
's The Hobbit , introducing Gand alf . AL
a horrible scene of a child being eaten alive by As
of 48 %). It remains to be seen whether the C
where it was forged . Fro do takes this task upon
midterm election . It remains to be seen whether this sort
for donations . It remains to be seen in the next
an historical condition of the human intellect to which the element
the C UP ) remains to be seen . Let us
called a Bal rog . Gand alf faces the Bal rog
Without a working star tracker , ion thrust ing
Helm 's Deep , while Gand alf dep arts to seek
to take the Ring from Fro do , but Fro do
are nine in number : Fro do , Sam , Merry
torn between his loyalty to Fro do and his desire for
County , South Carolina , commemor ates her birthplace .[ 61
, is strictly prohibited without the express written permission of Money

INTERVAL 0.965 - 1.826
CONTAINS 0.053%

baking and is an excellent b inder too . Ch ia
the game 's popularity has w aned since last summer ,
in advance for any assistance you may be able to give
2002 Ray Rom ano 2002 Kel sey Gram mer
the company s value derived mostly from Booker

INTERVAL 0.104 - 0.965
CONTAINS 33.547%

Co os County is not an isolated case . Across the
: These " lab yrinth ine worlds filled with secrets encourage
and their ability to solve complex problems such as climate change
regions ( i . e ., MR Z text ) against
. " I love Hels ing borg and Sweden , but

INTERVAL -0.757 - 0.104
CONTAINS 66.080%

it mostly just all happened in the same day .
a small amount of tooth from each fossil .
ai was a T atar ( not to be confused with
Myst ification is a part of it .
is less adrenaline - fuel ed , has a slightly slower

INTERVAL -1.618 - -0.757
CONTAINS 0.320%

meters per second ( 5 , 9 46 miles per hour
was introduced as an Excel add - in back in 2013
ensity ) * Get Volume () ; return f Mass ;
will be thr ongs of ad oring fans wanting his aut
gas fracking operations began , had 8 57 earthquakes last year
Token Tony
Token ID8832
Feature activation+7.27e-02

Pos logit contributions
arest+2.171
unia+2.074
asa+1.741
 Grammy+1.724
ounding+1.695
Neg logit contributions
velt-2.113
catentry-1.982
REDACTED-1.882
••-1.839
atical-1.831
Token Shal
Token ID47974
Feature activation+5.74e-03

Pos logit contributions
arest+0.707
unia+0.677
asa+0.570
ounding+0.553
 Grammy+0.549
Neg logit contributions
velt-0.654
catentry-0.602
REDACTED-0.587
••-0.571
atical-0.564
Tokenh
Token ID71
Feature activation+1.76e-01

Pos logit contributions
arest+0.065
unia+0.062
asa+0.053
ounding+0.052
 Dame+0.051
Neg logit contributions
velt-0.043
catentry-0.038
REDACTED-0.036
••-0.036
joy-0.035
Tokenoub
Token ID12944
Feature activation+9.11e-01

Pos logit contributions
arest+1.489
unia+1.420
asa+1.166
ounding+1.131
ine+1.090
Neg logit contributions
velt-1.803
catentry-1.707
REDACTED-1.672
••-1.621
 gamb-1.591
Token
Token ID198
Feature activation+4.13e-01

Pos logit contributions
arest+5.889
unia+5.265
ine+4.489
asa+4.340
 Grammy+4.311
Neg logit contributions
velt-9.884
catentry-9.804
REDACTED-9.305
••-9.076
atical-8.909
Token
Token ID198
Feature activation+1.83e+00

Pos logit contributions
arest+2.977
unia+2.781
asa+2.208
ine+2.172
 replay+2.065
Neg logit contributions
velt-4.665
catentry-4.472
REDACTED-4.196
atical-4.187
••-4.175
Token2003
Token ID16088
Feature activation+1.51e-01

Pos logit contributions
arest+8.081
unia+6.365
rel+6.072
ine+5.877
asa+5.389
Neg logit contributions
velt-19.494
 scapego-18.984
 gamb-18.334
 Typh-18.273
 swast-18.097
Token Larry
Token ID13633
Feature activation+2.94e-01

Pos logit contributions
arest+1.426
unia+1.363
asa+1.140
 Grammy+1.117
ounding+1.110
Neg logit contributions
velt-1.398
catentry-1.300
REDACTED-1.245
••-1.219
atical-1.207
Token David
Token ID3271
Feature activation+1.06e+00

Pos logit contributions
arest+2.607
unia+2.481
asa+2.109
 Grammy+2.087
ine+2.066
Neg logit contributions
velt-2.732
catentry-2.638
REDACTED-2.610
••-2.446
atical-2.420
Token
Token ID198
Feature activation+5.24e-01

Pos logit contributions
arest+6.628
unia+5.839
 Grammy+5.174
ine+4.905
asa+4.904
Neg logit contributions
velt-11.332
catentry-11.163
REDACTED-10.494
atical-10.295
••-10.221
Token
Token ID198
Feature activation+1.73e+00

Pos logit contributions
arest+3.567
unia+3.304
asa+2.613
ine+2.613
rel+2.451
Neg logit contributions
velt-6.046
catentry-5.839
atical-5.461
 gamb-5.434
REDACTED-5.424
Token Matt
Token ID4705
Feature activation+8.68e-02

Pos logit contributions
velt+0.805
catentry+0.736
REDACTED+0.722
••+0.707
atical+0.695
Neg logit contributions
arest-0.848
unia-0.813
asa-0.678
ounding-0.661
 Grammy-0.641
Token Le
Token ID1004
Feature activation+8.21e-01

Pos logit contributions
arest+0.681
unia+0.647
asa+0.516
ounding+0.499
 Grammy+0.488
Neg logit contributions
velt-0.948
catentry-0.890
REDACTED-0.870
••-0.848
atical-0.839
TokenBl
Token ID3629
Feature activation+3.14e-01

Pos logit contributions
arest+6.276
unia+5.850
 Grammy+4.589
ounding+4.531
asa+4.463
Neg logit contributions
velt-8.545
REDACTED-7.803
catentry-7.790
 Typh-7.589
••-7.573
Tokenanc
Token ID1192
Feature activation+1.12e+00

Pos logit contributions
arest+2.454
unia+2.276
asa+1.862
ounding+1.758
ine+1.741
Neg logit contributions
velt-3.418
catentry-3.315
REDACTED-3.209
••-3.105
 Typh-3.105
Token
Token ID198
Feature activation+4.37e-01

Pos logit contributions
arest+6.356
unia+5.532
asa+4.698
 Grammy+4.635
ine+4.531
Neg logit contributions
velt-12.343
catentry-12.022
atical-11.433
REDACTED-11.401
••-11.295
Token
Token ID198
Feature activation+1.78e+00

Pos logit contributions
arest+3.099
unia+2.897
asa+2.294
ine+2.269
 replay+2.146
Neg logit contributions
velt-4.960
catentry-4.756
atical-4.456
REDACTED-4.455
 gamb-4.444
Token2003
Token ID16088
Feature activation-1.67e-01

Pos logit contributions
arest+7.794
unia+6.051
ine+5.874
Depth+5.785
rel+5.743
Neg logit contributions
velt-19.580
 destro-19.437
 scapego-19.303
 gamb-18.678
 Typh-18.283
Token Bernie
Token ID10477
Feature activation+1.50e-01

Pos logit contributions
velt+1.481
catentry+1.344
REDACTED+1.328
••+1.303
 gamb+1.277
Neg logit contributions
arest-1.634
unia-1.571
asa-1.312
ounding-1.281
 replay-1.230
Token Mac
Token ID4100
Feature activation+1.04e+00

Pos logit contributions
arest+1.388
unia+1.333
asa+1.109
 Grammy+1.069
ounding+1.062
Neg logit contributions
velt-1.407
catentry-1.303
REDACTED-1.263
••-1.235
atical-1.225
Token
Token ID198
Feature activation+4.45e-01

Pos logit contributions
arest+6.626
unia+5.916
 Grammy+4.802
asa+4.764
ine+4.694
Neg logit contributions
velt-11.218
catentry-10.854
REDACTED-10.391
••-10.315
atical-10.277
Token
Token ID198
Feature activation+1.65e+00

Pos logit contributions
arest+3.148
unia+2.944
asa+2.330
ine+2.299
 replay+2.179
Neg logit contributions
velt-5.060
catentry-4.860
atical-4.552
REDACTED-4.551
 gamb-4.535
Token
Token ID198
Feature activation+1.83e+00

Pos logit contributions
arest+2.977
unia+2.781
asa+2.208
ine+2.172
 replay+2.065
Neg logit contributions
velt-4.665
catentry-4.472
REDACTED-4.196
atical-4.187
••-4.175
Token2003
Token ID16088
Feature activation+1.51e-01

Pos logit contributions
arest+8.081
unia+6.365
rel+6.072
ine+5.877
asa+5.389
Neg logit contributions
velt-19.494
 scapego-18.984
 gamb-18.334
 Typh-18.273
 swast-18.097
Token Larry
Token ID13633
Feature activation+2.94e-01

Pos logit contributions
arest+1.426
unia+1.363
asa+1.140
 Grammy+1.117
ounding+1.110
Neg logit contributions
velt-1.398
catentry-1.300
REDACTED-1.245
••-1.219
atical-1.207
Token David
Token ID3271
Feature activation+1.06e+00

Pos logit contributions
arest+2.607
unia+2.481
asa+2.109
 Grammy+2.087
ine+2.066
Neg logit contributions
velt-2.732
catentry-2.638
REDACTED-2.610
••-2.446
atical-2.420
Token
Token ID198
Feature activation+5.24e-01

Pos logit contributions
arest+6.628
unia+5.839
 Grammy+5.174
ine+4.905
asa+4.904
Neg logit contributions
velt-11.332
catentry-11.163
REDACTED-10.494
atical-10.295
••-10.221
Token
Token ID198
Feature activation+1.73e+00

Pos logit contributions
arest+3.567
unia+3.304
asa+2.613
ine+2.613
rel+2.451
Neg logit contributions
velt-6.046
catentry-5.839
atical-5.461
 gamb-5.434
REDACTED-5.424
Token2003
Token ID16088
Feature activation-8.84e-02

Pos logit contributions
arest+8.008
unia+6.374
rel+5.863
ine+5.817
Depth+5.397
Neg logit contributions
velt-18.671
 scapego-18.223
 gamb-17.756
 Typh-17.561
 Gamb-17.272
Token Matt
Token ID4705
Feature activation+8.68e-02

Pos logit contributions
velt+0.805
catentry+0.736
REDACTED+0.722
••+0.707
atical+0.695
Neg logit contributions
arest-0.848
unia-0.813
asa-0.678
ounding-0.661
 Grammy-0.641
Token Le
Token ID1004
Feature activation+8.21e-01

Pos logit contributions
arest+0.681
unia+0.647
asa+0.516
ounding+0.499
 Grammy+0.488
Neg logit contributions
velt-0.948
catentry-0.890
REDACTED-0.870
••-0.848
atical-0.839
TokenBl
Token ID3629
Feature activation+3.14e-01

Pos logit contributions
arest+6.276
unia+5.850
 Grammy+4.589
ounding+4.531
asa+4.463
Neg logit contributions
velt-8.545
REDACTED-7.803
catentry-7.790
 Typh-7.589
••-7.573
Tokenanc
Token ID1192
Feature activation+1.12e+00

Pos logit contributions
arest+2.454
unia+2.276
asa+1.862
ounding+1.758
ine+1.741
Neg logit contributions
velt-3.418
catentry-3.315
REDACTED-3.209
••-3.105
 Typh-3.105
Token
Token ID198
Feature activation+1.57e+00

Pos logit contributions
arest+2.933
unia+2.740
asa+2.186
ine+2.143
 replay+2.036
Neg logit contributions
velt-4.557
catentry-4.364
REDACTED-4.102
atical-4.094
 gamb-4.079
Token2002
Token ID16942
Feature activation-1.54e-01

Pos logit contributions
arest+8.685
unia+7.375
rel+7.146
Depth+6.774
ine+6.641
Neg logit contributions
velt-15.933
 destro-15.755
 gamb-15.700
 scapego-15.670
 capitals-15.217
Token Bernie
Token ID10477
Feature activation+4.02e-01

Pos logit contributions
velt+1.394
catentry+1.266
REDACTED+1.254
••+1.227
atical+1.204
Neg logit contributions
arest-1.484
unia-1.427
asa-1.186
ounding-1.158
 replay-1.111
Token Mac
Token ID4100
Feature activation+1.40e-01

Pos logit contributions
arest+3.762
unia+3.550
asa+3.038
 Grammy+2.932
ine+2.924
Neg logit contributions
velt-3.685
catentry-3.555
REDACTED-3.360
••-3.235
atical-3.225
Token
Token ID198
Feature activation+4.61e-01

Pos logit contributions
arest+1.298
unia+1.237
asa+1.032
ounding+1.003
 Grammy+0.977
Neg logit contributions
velt-1.325
catentry-1.231
REDACTED-1.195
••-1.172
atical-1.154
Token
Token ID198
Feature activation+1.65e+00

Pos logit contributions
arest+3.247
unia+3.026
asa+2.403
ine+2.371
rel+2.243
Neg logit contributions
velt-5.250
catentry-5.041
atical-4.730
REDACTED-4.717
 gamb-4.709
Token2002
Token ID16942
Feature activation+2.81e-02

Pos logit contributions
arest+9.579
unia+8.222
rel+7.892
Depth+7.479
ine+7.468
Neg logit contributions
velt-15.706
 scapego-15.534
 gamb-15.489
 swast-15.038
 capitals-15.024
Token Matthew
Token ID9308
Feature activation+4.41e-01

Pos logit contributions
arest+0.267
unia+0.255
asa+0.213
ounding+0.207
 Grammy+0.205
Neg logit contributions
velt-0.259
catentry-0.239
REDACTED-0.231
••-0.226
atical-0.224
Token Perry
Token ID14105
Feature activation+8.47e-01

Pos logit contributions
arest+4.351
unia+4.206
asa+3.553
 Grammy+3.481
ine+3.454
Neg logit contributions
velt-3.815
catentry-3.712
REDACTED-3.542
••-3.307
atical-3.288
Token 2003
Token ID5816
Feature activation+2.30e-01

Pos logit contributions
arest+5.561
unia+5.098
 Grammy+4.364
asa+4.278
ine+4.134
Neg logit contributions
velt-8.934
catentry-8.849
atical-8.432
REDACTED-8.378
••-8.299
Token Tony
Token ID8832
Feature activation+7.27e-02

Pos logit contributions
arest+2.171
unia+2.074
asa+1.741
 Grammy+1.724
ounding+1.695
Neg logit contributions
velt-2.113
catentry-1.982
REDACTED-1.882
••-1.839
atical-1.831
Token
Token ID198
Feature activation+1.78e+00

Pos logit contributions
arest+3.099
unia+2.897
asa+2.294
ine+2.269
 replay+2.146
Neg logit contributions
velt-4.960
catentry-4.756
atical-4.456
REDACTED-4.455
 gamb-4.444
Token2003
Token ID16088
Feature activation-1.67e-01

Pos logit contributions
arest+7.794
unia+6.051
ine+5.874
Depth+5.785
rel+5.743
Neg logit contributions
velt-19.580
 destro-19.437
 scapego-19.303
 gamb-18.678
 Typh-18.283
Token Bernie
Token ID10477
Feature activation+1.50e-01

Pos logit contributions
velt+1.481
catentry+1.344
REDACTED+1.328
••+1.303
 gamb+1.277
Neg logit contributions
arest-1.634
unia-1.571
asa-1.312
ounding-1.281
 replay-1.230
Token Mac
Token ID4100
Feature activation+1.04e+00

Pos logit contributions
arest+1.388
unia+1.333
asa+1.109
 Grammy+1.069
ounding+1.062
Neg logit contributions
velt-1.407
catentry-1.303
REDACTED-1.263
••-1.235
atical-1.225
Token
Token ID198
Feature activation+4.45e-01

Pos logit contributions
arest+6.626
unia+5.916
 Grammy+4.802
asa+4.764
ine+4.694
Neg logit contributions
velt-11.218
catentry-10.854
REDACTED-10.391
••-10.315
atical-10.277
Token
Token ID198
Feature activation+1.65e+00

Pos logit contributions
arest+3.148
unia+2.944
asa+2.330
ine+2.299
 replay+2.179
Neg logit contributions
velt-5.060
catentry-4.860
atical-4.552
REDACTED-4.551
 gamb-4.535
Token2003
Token ID16088
Feature activation-1.85e-01

Pos logit contributions
arest+7.618
unia+6.014
ine+5.590
Depth+5.569
rel+5.499
Neg logit contributions
 destro-18.462
velt-18.434
 scapego-18.176
 gamb-17.699
 Typh-17.279
Token Eric
Token ID7651
Feature activation+6.79e-02

Pos logit contributions
velt+1.625
catentry+1.470
REDACTED+1.453
••+1.427
joy+1.398
Neg logit contributions
arest-1.822
unia-1.753
asa-1.469
ounding-1.434
 replay-1.377
Token McCorm
Token ID42037
Feature activation-1.73e-01

Pos logit contributions
arest+0.744
unia+0.717
asa+0.618
ounding+0.599
 Grammy+0.595
Neg logit contributions
velt-0.524
catentry-0.476
REDACTED-0.460
••-0.447
atical-0.441
Tokenack
Token ID441
Feature activation+1.31e+00

Pos logit contributions
velt+1.204
catentry+1.048
atical+1.007
REDACTED+1.005
••+0.987
Neg logit contributions
arest-2.019
unia-1.951
asa-1.664
ounding-1.657
 Grammy-1.657
Token
Token ID198
Feature activation+4.29e-01

Pos logit contributions
arest+7.426
unia+6.561
 Grammy+5.644
ine+5.613
asa+5.508
Neg logit contributions
velt-13.919
catentry-13.331
REDACTED-12.689
••-12.592
joy-12.590
Token2003
Token ID16088
Feature activation-1.85e-01

Pos logit contributions
arest+7.618
unia+6.014
ine+5.590
Depth+5.569
rel+5.499
Neg logit contributions
 destro-18.462
velt-18.434
 scapego-18.176
 gamb-17.699
 Typh-17.279
Token Eric
Token ID7651
Feature activation+6.79e-02

Pos logit contributions
velt+1.625
catentry+1.470
REDACTED+1.453
••+1.427
joy+1.398
Neg logit contributions
arest-1.822
unia-1.753
asa-1.469
ounding-1.434
 replay-1.377
Token McCorm
Token ID42037
Feature activation-1.73e-01

Pos logit contributions
arest+0.744
unia+0.717
asa+0.618
ounding+0.599
 Grammy+0.595
Neg logit contributions
velt-0.524
catentry-0.476
REDACTED-0.460
••-0.447
atical-0.441
Tokenack
Token ID441
Feature activation+1.31e+00

Pos logit contributions
velt+1.204
catentry+1.048
atical+1.007
REDACTED+1.005
••+0.987
Neg logit contributions
arest-2.019
unia-1.951
asa-1.664
ounding-1.657
 Grammy-1.657
Token
Token ID198
Feature activation+4.29e-01

Pos logit contributions
arest+7.426
unia+6.561
 Grammy+5.644
ine+5.613
asa+5.508
Neg logit contributions
velt-13.919
catentry-13.331
REDACTED-12.689
••-12.592
joy-12.590
Token
Token ID198
Feature activation+1.61e+00

Pos logit contributions
arest+3.064
unia+2.869
asa+2.280
ine+2.241
 replay+2.129
Neg logit contributions
velt-4.866
catentry-4.666
REDACTED-4.374
atical-4.366
••-4.350
Token2003
Token ID16088
Feature activation-9.33e-02

Pos logit contributions
arest+7.665
unia+6.108
ine+5.715
rel+5.621
Depth+5.320
Neg logit contributions
velt-17.715
 scapego-17.179
 gamb-16.768
 Typh-16.555
 Gamb-16.263
Token Ray
Token ID7760
Feature activation-1.74e-01

Pos logit contributions
velt+0.834
catentry+0.760
REDACTED+0.746
••+0.731
atical+0.718
Neg logit contributions
arest-0.909
unia-0.873
asa-0.730
ounding-0.711
 Grammy-0.688
Token Rom
Token ID3570
Feature activation-4.72e-02

Pos logit contributions
velt+1.208
catentry+1.054
REDACTED+1.036
••+0.999
 gamb+0.992
Neg logit contributions
arest-2.030
unia-1.955
asa-1.701
ounding-1.673
 Grammy-1.615
Tokenano
Token ID5733
Feature activation+1.36e+00

Pos logit contributions
velt+0.281
catentry+0.244
REDACTED+0.235
••+0.227
atical+0.222
Neg logit contributions
arest-0.601
unia-0.583
asa-0.511
ounding-0.502
 Grammy-0.495
Token 2004
Token ID5472
Feature activation+5.61e-02

Pos logit contributions
arest+7.428
unia+6.268
 Grammy+5.540
ine+5.360
asa+5.282
Neg logit contributions
velt-14.881
catentry-14.014
••-13.475
REDACTED-13.369
atical-13.321
Token Kel
Token ID15150
Feature activation+2.19e-01

Pos logit contributions
velt+0.295
catentry+0.272
REDACTED+0.265
••+0.260
atical+0.256
Neg logit contributions
arest-0.294
unia-0.281
asa-0.234
ounding-0.227
 Grammy-0.223
Tokensey
Token ID4397
Feature activation+1.26e-01

Pos logit contributions
arest+1.978
unia+1.889
asa+1.569
ounding+1.494
ine+1.490
Neg logit contributions
velt-2.106
catentry-1.971
REDACTED-1.923
••-1.882
 gamb-1.849
Token Gram
Token ID20159
Feature activation+5.66e-02

Pos logit contributions
arest+0.931
unia+0.889
asa+0.692
 Grammy+0.668
ounding+0.664
Neg logit contributions
velt-1.421
catentry-1.335
REDACTED-1.305
••-1.276
atical-1.273
Tokenmer
Token ID647
Feature activation+8.63e-01

Pos logit contributions
arest+0.472
unia+0.449
asa+0.364
ounding+0.352
 Grammy+0.342
Neg logit contributions
velt-0.586
catentry-0.545
REDACTED-0.533
••-0.523
atical-0.516
Token
Token ID198
Feature activation+4.44e-01

Pos logit contributions
arest+5.085
unia+4.575
asa+3.726
ine+3.698
 Dame+3.620
Neg logit contributions
velt-9.980
catentry-9.791
atical-9.301
 destro-9.118
REDACTED-9.059
Token
Token ID198
Feature activation+1.58e+00

Pos logit contributions
arest+3.130
unia+2.923
asa+2.317
ine+2.295
 replay+2.165
Neg logit contributions
velt-5.065
catentry-4.858
atical-4.555
REDACTED-4.550
 gamb-4.535
Token2002
Token ID16942
Feature activation-4.09e-01

Pos logit contributions
arest+7.822
unia+6.547
rel+6.261
Depth+6.155
ine+5.792
Neg logit contributions
 destro-17.776
 gamb-16.936
velt-16.916
 scapego-16.725
 capitals-16.301
Token Matt
Token ID4705
Feature activation+7.23e-02

Pos logit contributions
velt+3.835
REDACTED+3.478
••+3.455
catentry+3.417
 Gamb+3.398
Neg logit contributions
arest-3.695
unia-3.574
asa-2.897
ounding-2.830
 replay-2.663
Token Le
Token ID1004
Feature activation+3.45e-01

Pos logit contributions
arest+0.713
unia+0.684
asa+0.576
ounding+0.560
 Grammy+0.549
Neg logit contributions
velt-0.642
catentry-0.596
REDACTED-0.579
••-0.562
atical-0.553
TokenBl
Token ID3629
Feature activation+5.27e-01

Pos logit contributions
arest+3.165
unia+3.004
asa+2.513
ounding+2.429
ine+2.409
Neg logit contributions
velt-3.223
catentry-3.043
REDACTED-2.957
••-2.865
 gamb-2.830
Tokenanc
Token ID1192
Feature activation+1.24e+00

Pos logit contributions
arest+3.831
unia+3.574
asa+2.909
ounding+2.782
ine+2.775
Neg logit contributions
velt-6.028
catentry-5.999
REDACTED-5.796
••-5.450
 Typh-5.436
Token Matt
Token ID4705
Feature activation+7.23e-02

Pos logit contributions
velt+3.835
REDACTED+3.478
••+3.455
catentry+3.417
 Gamb+3.398
Neg logit contributions
arest-3.695
unia-3.574
asa-2.897
ounding-2.830
 replay-2.663
Token Le
Token ID1004
Feature activation+3.45e-01

Pos logit contributions
arest+0.713
unia+0.684
asa+0.576
ounding+0.560
 Grammy+0.549
Neg logit contributions
velt-0.642
catentry-0.596
REDACTED-0.579
••-0.562
atical-0.553
TokenBl
Token ID3629
Feature activation+5.27e-01

Pos logit contributions
arest+3.165
unia+3.004
asa+2.513
ounding+2.429
ine+2.409
Neg logit contributions
velt-3.223
catentry-3.043
REDACTED-2.957
••-2.865
 gamb-2.830
Tokenanc
Token ID1192
Feature activation+1.24e+00

Pos logit contributions
arest+3.831
unia+3.574
asa+2.909
ounding+2.782
ine+2.775
Neg logit contributions
velt-6.028
catentry-5.999
REDACTED-5.796
••-5.450
 Typh-5.436
Token
Token ID198
Feature activation+4.05e-01

Pos logit contributions
arest+7.904
unia+6.824
asa+6.223
 Grammy+5.958
ine+5.859
Neg logit contributions
velt-12.129
catentry-11.946
atical-11.450
REDACTED-11.414
••-11.353
Token
Token ID198
Feature activation+1.57e+00

Pos logit contributions
arest+2.933
unia+2.740
asa+2.186
ine+2.143
 replay+2.036
Neg logit contributions
velt-4.557
catentry-4.364
REDACTED-4.102
atical-4.094
 gamb-4.079
Token2002
Token ID16942
Feature activation-1.54e-01

Pos logit contributions
arest+8.685
unia+7.375
rel+7.146
Depth+6.774
ine+6.641
Neg logit contributions
velt-15.933
 destro-15.755
 gamb-15.700
 scapego-15.670
 capitals-15.217
Token Bernie
Token ID10477
Feature activation+4.02e-01

Pos logit contributions
velt+1.394
catentry+1.266
REDACTED+1.254
••+1.227
atical+1.204
Neg logit contributions
arest-1.484
unia-1.427
asa-1.186
ounding-1.158
 replay-1.111
Token Mac
Token ID4100
Feature activation+1.40e-01

Pos logit contributions
arest+3.762
unia+3.550
asa+3.038
 Grammy+2.932
ine+2.924
Neg logit contributions
velt-3.685
catentry-3.555
REDACTED-3.360
••-3.235
atical-3.225
Token
Token ID198
Feature activation+4.61e-01

Pos logit contributions
arest+1.298
unia+1.237
asa+1.032
ounding+1.003
 Grammy+0.977
Neg logit contributions
velt-1.325
catentry-1.231
REDACTED-1.195
••-1.172
atical-1.154
Token
Token ID198
Feature activation+1.65e+00

Pos logit contributions
arest+3.247
unia+3.026
asa+2.403
ine+2.371
rel+2.243
Neg logit contributions
velt-5.250
catentry-5.041
atical-4.730
REDACTED-4.717
 gamb-4.709
Token Kel
Token ID15150
Feature activation+3.31e-01

Pos logit contributions
arest+0.529
unia+0.506
asa+0.422
ounding+0.411
 Grammy+0.408
Neg logit contributions
velt-0.519
catentry-0.480
REDACTED-0.466
••-0.454
atical-0.449
Tokensey
Token ID4397
Feature activation+1.74e-01

Pos logit contributions
arest+2.927
unia+2.794
asa+2.303
ine+2.203
ounding+2.178
Neg logit contributions
velt-3.237
catentry-3.050
REDACTED-2.976
••-2.913
 gamb-2.860
Token Gram
Token ID20159
Feature activation-2.04e-02

Pos logit contributions
arest+1.254
unia+1.203
asa+0.926
 Grammy+0.908
ounding+0.885
Neg logit contributions
velt-1.981
catentry-1.867
REDACTED-1.821
••-1.779
atical-1.779
Tokenmer
Token ID647
Feature activation+7.38e-01

Pos logit contributions
velt+0.207
catentry+0.192
REDACTED+0.188
••+0.184
atical+0.182
Neg logit contributions
arest-0.173
unia-0.165
asa-0.134
ounding-0.130
 Grammy-0.127
Token
Token ID198
Feature activation+2.92e-01

Pos logit contributions
arest+4.695
unia+4.307
asa+3.432
ine+3.414
 Dame+3.235
Neg logit contributions
velt-8.445
catentry-8.219
atical-7.778
REDACTED-7.674
 gamb-7.639
Token
Token ID198
Feature activation+1.47e+00

Pos logit contributions
arest+2.229
unia+2.103
asa+1.683
ine+1.626
 replay+1.576
Neg logit contributions
velt-3.213
catentry-3.049
REDACTED-2.900
••-2.866
atical-2.863
Token2004
Token ID15724
Feature activation-2.51e-01

Pos logit contributions
arest+5.591
unia+4.689
Depth+3.839
asa+3.528
ine+3.517
Neg logit contributions
velt-19.141
 destro-18.144
 scapego-17.594
 gamb-17.584
 Typh-16.800
Token Larry
Token ID13633
Feature activation+5.35e-02

Pos logit contributions
velt+2.244
catentry+2.026
REDACTED+2.024
••+1.998
joy+1.956
Neg logit contributions
arest-2.403
unia-2.315
asa-1.928
ounding-1.871
 replay-1.789
Token David
Token ID3271
Feature activation+6.48e-01

Pos logit contributions
arest+0.445
unia+0.424
asa+0.343
ounding+0.332
 Grammy+0.323
Neg logit contributions
velt-0.556
catentry-0.516
REDACTED-0.505
••-0.496
atical-0.489
Token
Token ID198
Feature activation+4.99e-01

Pos logit contributions
arest+4.287
unia+3.990
asa+3.182
ine+3.141
 replay+2.993
Neg logit contributions
velt-7.382
catentry-7.149
atical-6.768
REDACTED-6.660
••-6.611
Token
Token ID198
Feature activation+1.31e+00

Pos logit contributions
arest+3.431
unia+3.184
ine+2.516
asa+2.506
rel+2.351
Neg logit contributions
velt-5.738
catentry-5.525
atical-5.171
 gamb-5.163
REDACTED-5.138
Token 2002
Token ID6244
Feature activation+2.78e-01

Pos logit contributions
arest+5.534
unia+4.777
 Grammy+4.152
asa+4.079
ine+3.854
Neg logit contributions
velt-9.213
catentry-8.993
atical-8.544
••-8.466
REDACTED-8.434
Token Ray
Token ID7760
Feature activation+8.03e-02

Pos logit contributions
arest+2.498
unia+2.380
 Grammy+2.014
asa+1.993
 Dame+1.943
Neg logit contributions
velt-2.647
catentry-2.505
REDACTED-2.365
atical-2.312
••-2.300
Token Rom
Token ID3570
Feature activation+1.25e-01

Pos logit contributions
arest+0.890
unia+0.860
asa+0.737
ounding+0.720
 Grammy+0.711
Neg logit contributions
velt-0.609
catentry-0.553
REDACTED-0.534
••-0.521
atical-0.511
Tokenano
Token ID5733
Feature activation+1.19e+00

Pos logit contributions
arest+1.554
unia+1.499
asa+1.314
ounding+1.282
ine+1.259
Neg logit contributions
velt-0.792
catentry-0.718
REDACTED-0.683
••-0.660
 gamb-0.647
Token
Token ID198
Feature activation+2.24e-01

Pos logit contributions
arest+6.395
unia+5.563
ine+4.605
asa+4.540
 Dame+4.461
Neg logit contributions
velt-13.406
catentry-12.967
••-12.619
atical-12.581
 destro-12.483
Token
Token ID198
Feature activation+1.46e+00

Pos logit contributions
arest+1.766
unia+1.670
asa+1.345
ine+1.285
ounding+1.272
Neg logit contributions
velt-2.406
catentry-2.268
REDACTED-2.174
••-2.146
atical-2.135
Token2002
Token ID16942
Feature activation-3.15e-02

Pos logit contributions
arest+8.582
unia+7.304
rel+6.567
asa+6.374
Depth+6.141
Neg logit contributions
velt-14.541
 gamb-14.193
 scapego-13.903
 Gamb-13.762
 Typh-13.739
Token Kel
Token ID15150
Feature activation+2.19e-01

Pos logit contributions
velt+0.295
catentry+0.272
REDACTED+0.265
••+0.260
atical+0.256
Neg logit contributions
arest-0.294
unia-0.281
asa-0.234
ounding-0.227
 Grammy-0.223
Tokensey
Token ID4397
Feature activation+1.26e-01

Pos logit contributions
arest+1.978
unia+1.889
asa+1.569
ounding+1.494
ine+1.490
Neg logit contributions
velt-2.106
catentry-1.971
REDACTED-1.923
••-1.882
 gamb-1.849
Token Gram
Token ID20159
Feature activation+5.66e-02

Pos logit contributions
arest+0.931
unia+0.889
asa+0.692
 Grammy+0.668
ounding+0.664
Neg logit contributions
velt-1.421
catentry-1.335
REDACTED-1.305
••-1.276
atical-1.273
Tokenmer
Token ID647
Feature activation+8.63e-01

Pos logit contributions
arest+0.472
unia+0.449
asa+0.364
ounding+0.352
 Grammy+0.342
Neg logit contributions
velt-0.586
catentry-0.545
REDACTED-0.533
••-0.523
atical-0.516
TokenDraw
Token ID25302
Feature activation+2.38e-01

Pos logit contributions
arest+3.450
unia+3.167
ine+2.464
Depth+2.451
asa+2.415
Neg logit contributions
velt-6.500
catentry-6.115
 gamb-5.980
atical-5.881
REDACTED-5.850
Tokening
Token ID278
Feature activation+4.94e-01

Pos logit contributions
arest+1.560
unia+1.453
asa+1.107
ounding+1.069
ine+1.041
Neg logit contributions
velt-2.879
catentry-2.728
REDACTED-2.672
••-2.629
 gamb-2.608
Token.
Token ID13
Feature activation+7.76e-01

Pos logit contributions
arest+3.506
unia+3.261
asa+2.610
ine+2.577
rel+2.409
Neg logit contributions
velt-5.596
catentry-5.352
 gamb-5.029
atical-5.017
REDACTED-4.970
TokenColor
Token ID10258
Feature activation-1.81e-01

Pos logit contributions
arest+5.179
unia+4.787
Depth+4.019
ine+3.947
asa+3.844
Neg logit contributions
velt-8.752
catentry-7.993
 gamb-7.975
 scapego-7.810
 Typh-7.697
Token]
Token ID60
Feature activation+4.21e-01

Pos logit contributions
velt+1.110
catentry+0.956
REDACTED+0.952
••+0.913
joy+0.880
Neg logit contributions
arest-2.272
unia-2.213
asa-1.913
ounding-1.899
 Grammy-1.866
Token::
Token ID3712
Feature activation+1.46e+00

Pos logit contributions
arest+2.828
unia+2.611
asa+2.035
ine+1.930
ounding+1.844
Neg logit contributions
velt-4.964
catentry-4.734
 gamb-4.566
atical-4.503
REDACTED-4.469
TokenFrom
Token ID4863
Feature activation+7.09e-01

Pos logit contributions
arest+5.979
Depth+5.185
Draft+4.358
Gas+4.352
ine+4.206
Neg logit contributions
 destro-18.467
velt-18.091
 gamb-17.707
ittens-17.222
 lapt-17.063
TokenAr
Token ID3163
Feature activation+5.10e-01

Pos logit contributions
arest+7.240
unia+6.752
asa+6.055
Depth+5.917
ine+5.872
Neg logit contributions
velt-5.716
catentry-5.198
 gamb-5.002
 scapego-4.863
atical-4.783
Tokengb
Token ID22296
Feature activation+3.89e-01

Pos logit contributions
arest+4.066
unia+3.931
asa+3.233
ounding+3.012
ine+2.957
Neg logit contributions
velt-5.294
catentry-5.081
REDACTED-4.911
••-4.877
 Typh-4.817
Token($
Token ID16763
Feature activation+2.18e-01

Pos logit contributions
arest+3.286
unia+3.165
asa+2.556
ounding+2.513
ine+2.465
Neg logit contributions
velt-3.880
catentry-3.603
 gamb-3.493
REDACTED-3.465
atical-3.452
TokenFC
Token ID4851
Feature activation+3.73e-02

Pos logit contributions
arest+2.388
unia+2.277
asa+1.966
ounding+1.929
ine+1.889
Neg logit contributions
velt-1.683
catentry-1.525
REDACTED-1.431
••-1.430
 gamb-1.425
Token.
Token ID13
Feature activation-3.16e-01

Pos logit contributions
arest+1.729
unia+1.661
asa+1.347
ounding+1.293
ine+1.272
Neg logit contributions
velt-2.071
catentry-1.954
REDACTED-1.891
••-1.846
joy-1.814
Token Maybe
Token ID6674
Feature activation-1.19e-01

Pos logit contributions
velt+2.676
••+2.388
catentry+2.365
REDACTED+2.353
 Gamb+2.340
Neg logit contributions
arest-3.182
unia-3.085
ounding-2.585
asa-2.583
 replay-2.415
Token their
Token ID511
Feature activation-2.26e-01

Pos logit contributions
velt+1.070
catentry+0.971
REDACTED+0.952
••+0.935
 gamb+0.923
Neg logit contributions
arest-1.153
unia-1.106
asa-0.925
ounding-0.906
 Grammy-0.878
Token calls
Token ID3848
Feature activation+2.37e-01

Pos logit contributions
velt+1.996
catentry+1.808
REDACTED+1.765
 gamb+1.755
••+1.742
Neg logit contributions
arest-2.207
unia-2.130
asa-1.797
ounding-1.750
 Grammy-1.681
Token will
Token ID481
Feature activation+1.18e-01

Pos logit contributions
arest+2.187
unia+2.069
asa+1.708
ounding+1.688
 replay+1.635
Neg logit contributions
velt-2.249
catentry-2.111
REDACTED-2.052
••-2.000
 Typh-1.942
Token go
Token ID467
Feature activation+1.42e+00

Pos logit contributions
arest+1.117
unia+1.069
asa+0.883
ounding+0.870
 replay+0.852
Neg logit contributions
velt-1.096
catentry-1.012
REDACTED-0.987
••-0.963
atical-0.945
Token un
Token ID555
Feature activation+1.54e-01

Pos logit contributions
arest+10.372
unia+9.362
asa+8.337
 Grammy+8.033
 ov+7.955
Neg logit contributions
velt-12.900
catentry-12.332
REDACTED-11.992
••-11.495
-11.130
Tokenhe
Token ID258
Feature activation+7.65e-02

Pos logit contributions
arest+1.461
unia+1.376
asa+1.149
ounding+1.135
ine+1.103
Neg logit contributions
velt-1.435
catentry-1.331
REDACTED-1.288
••-1.265
atical-1.237
Tokeneded
Token ID15395
Feature activation-7.40e-02

Pos logit contributions
arest+0.621
unia+0.589
asa+0.472
ounding+0.458
 Grammy+0.444
Neg logit contributions
velt-0.811
catentry-0.757
REDACTED-0.741
••-0.726
atical-0.716
Token
Token ID851
Feature activation-2.60e-01

Pos logit contributions
velt+0.717
catentry+0.661
REDACTED+0.646
••+0.635
atical+0.625
Neg logit contributions
arest-0.667
unia-0.638
asa-0.526
ounding-0.512
 Grammy-0.498
Token certainly
Token ID3729
Feature activation+5.21e-03

Pos logit contributions
velt+2.217
catentry+1.983
REDACTED+1.949
••+1.936
 gamb+1.916
Neg logit contributions
arest-2.614
unia-2.506
asa-2.120
ounding-2.090
 Grammy-2.000
TokenDraw
Token ID25302
Feature activation+2.56e-01

Pos logit contributions
arest+3.865
unia+3.474
Depth+2.641
ine+2.615
ounding+2.561
Neg logit contributions
velt-8.064
catentry-7.638
 gamb-7.436
REDACTED-7.370
 Typh-7.320
Tokening
Token ID278
Feature activation+6.93e-01

Pos logit contributions
arest+1.677
unia+1.564
asa+1.189
ounding+1.151
ine+1.118
Neg logit contributions
velt-3.095
catentry-2.930
REDACTED-2.871
••-2.823
 gamb-2.802
Token.
Token ID13
Feature activation+6.30e-01

Pos logit contributions
arest+4.561
unia+4.208
ine+3.343
asa+3.329
Depth+3.229
Neg logit contributions
velt-7.928
catentry-7.588
atical-7.250
 gamb-7.205
 destro-7.078
TokenColor
Token ID10258
Feature activation-6.17e-02

Pos logit contributions
arest+4.810
unia+4.441
asa+3.620
ine+3.598
Depth+3.563
Neg logit contributions
velt-6.633
catentry-6.155
 gamb-5.968
REDACTED-5.866
 Typh-5.831
Token]
Token ID60
Feature activation+3.88e-01

Pos logit contributions
velt+0.422
catentry+0.379
REDACTED+0.363
••+0.352
atical+0.346
Neg logit contributions
arest-0.732
unia-0.706
asa-0.615
ounding-0.603
 Grammy-0.592
Token::
Token ID3712
Feature activation+1.40e+00

Pos logit contributions
arest+2.709
unia+2.545
asa+1.976
ounding+1.871
ine+1.867
Neg logit contributions
velt-4.467
catentry-4.239
REDACTED-4.054
 gamb-4.042
atical-4.034
TokenFrom
Token ID4863
Feature activation+9.99e-01

Pos logit contributions
arest+9.816
unia+8.113
Depth+7.931
ine+7.742
asa+7.557
Neg logit contributions
velt-13.074
catentry-12.408
 gamb-12.147
joy-11.583
 Gamb-11.530
TokenAr
Token ID3163
Feature activation+3.48e-01

Pos logit contributions
arest+9.237
unia+8.471
asa+7.655
Depth+7.555
ine+7.437
Neg logit contributions
velt-8.301
catentry-7.683
 gamb-7.353
 scapego-7.105
 Typh-7.084
Tokengb
Token ID22296
Feature activation+5.92e-01

Pos logit contributions
arest+2.967
unia+2.851
asa+2.346
ounding+2.240
ine+2.183
Neg logit contributions
velt-3.503
catentry-3.304
REDACTED-3.206
••-3.175
 Typh-3.092
Token($
Token ID16763
Feature activation+2.30e-01

Pos logit contributions
arest+4.717
unia+4.528
ounding+3.648
asa+3.606
ine+3.583
Neg logit contributions
velt-5.969
catentry-5.709
REDACTED-5.403
 gamb-5.342
••-5.303
TokenFC
Token ID4851
Feature activation-1.01e-01

Pos logit contributions
arest+1.698
unia+1.573
asa+1.251
ounding+1.212
ine+1.162
Neg logit contributions
velt-2.582
catentry-2.431
REDACTED-2.339
••-2.337
 gamb-2.329
Token we
Token ID356
Feature activation-3.46e-01

Pos logit contributions
arest+1.403
unia+1.337
asa+1.092
ounding+1.044
ine+1.039
Neg logit contributions
velt-1.664
catentry-1.539
REDACTED-1.513
••-1.468
atical-1.453
Token move
Token ID1445
Feature activation+2.56e-02

Pos logit contributions
velt+3.353
catentry+3.084
 gamb+2.978
atical+2.936
••+2.927
Neg logit contributions
arest-3.062
unia-2.903
asa-2.385
ounding-2.359
 Grammy-2.278
Token into
Token ID656
Feature activation+2.98e-01

Pos logit contributions
arest+0.235
unia+0.224
asa+0.186
ounding+0.180
 Grammy+0.177
Neg logit contributions
velt-0.244
catentry-0.225
REDACTED-0.221
••-0.214
atical-0.212
Token hybrids
Token ID45456
Feature activation+3.76e-01

Pos logit contributions
arest+2.674
unia+2.546
asa+2.122
 Grammy+2.048
ounding+2.029
Neg logit contributions
velt-2.826
catentry-2.602
REDACTED-2.599
••-2.474
atical-2.460
Token and
Token ID290
Feature activation+5.84e-01

Pos logit contributions
arest+3.179
unia+3.051
asa+2.486
ounding+2.419
ine+2.405
Neg logit contributions
velt-3.705
REDACTED-3.436
catentry-3.417
atical-3.248
••-3.233
Token electric
Token ID5186
Feature activation+1.38e+00

Pos logit contributions
arest+5.234
unia+5.079
asa+4.331
ine+4.235
 replay+4.149
Neg logit contributions
velt-5.218
REDACTED-4.869
catentry-4.774
••-4.443
atical-4.420
Token cars
Token ID5006
Feature activation+1.67e-01

Pos logit contributions
unia+9.171
arest+9.037
ine+7.946
asa+7.172
 autos+7.050
Neg logit contributions
velt-12.872
REDACTED-12.527
catentry-12.142
atical-11.525
zai-11.458
Token,
Token ID11
Feature activation-1.20e-02

Pos logit contributions
arest+1.455
unia+1.393
asa+1.143
ounding+1.111
 Grammy+1.088
Neg logit contributions
velt-1.633
catentry-1.510
REDACTED-1.502
••-1.434
atical-1.431
Token gas
Token ID3623
Feature activation+5.93e-01

Pos logit contributions
velt+0.113
catentry+0.104
REDACTED+0.102
••+0.100
atical+0.098
Neg logit contributions
arest-0.111
unia-0.106
asa-0.088
ounding-0.085
 Grammy-0.083
Token tax
Token ID1687
Feature activation+1.74e-01

Pos logit contributions
arest+5.095
unia+4.909
ine+4.000
asa+3.934
ounding+3.912
Neg logit contributions
velt-5.555
REDACTED-5.345
catentry-5.282
••-5.155
atical-4.895
Token revenues
Token ID13089
Feature activation-1.38e-04

Pos logit contributions
arest+1.679
unia+1.610
asa+1.348
ounding+1.312
 replay+1.286
Neg logit contributions
velt-1.567
catentry-1.432
REDACTED-1.419
••-1.362
joy-1.342
Token Matt
Token ID4705
Feature activation+2.59e-02

Pos logit contributions
velt+1.923
catentry+1.729
REDACTED+1.719
••+1.695
joy+1.666
Neg logit contributions
arest-2.183
unia-2.102
asa-1.768
ounding-1.719
 replay-1.653
Token Le
Token ID1004
Feature activation+9.56e-01

Pos logit contributions
arest+0.211
unia+0.201
asa+0.162
ounding+0.157
 Grammy+0.152
Neg logit contributions
velt-0.274
catentry-0.254
REDACTED-0.249
••-0.245
atical-0.241
TokenBl
Token ID3629
Feature activation+4.83e-03

Pos logit contributions
arest+6.480
unia+6.043
Depth+4.706
ine+4.546
rel+4.542
Neg logit contributions
velt-10.587
 Typh-9.529
catentry-9.450
atical-9.361
 scapego-9.315
Tokenanc
Token ID1192
Feature activation+9.97e-01

Pos logit contributions
arest+0.041
unia+0.039
asa+0.032
ounding+0.031
 Grammy+0.030
Neg logit contributions
velt-0.050
catentry-0.046
REDACTED-0.045
••-0.044
 gamb-0.043
Token
Token ID198
Feature activation+4.37e-01

Pos logit contributions
arest+5.639
unia+5.096
asa+4.199
ine+3.936
 replay+3.775
Neg logit contributions
velt-11.603
catentry-11.194
atical-10.783
 destro-10.718
••-10.574
Token
Token ID198
Feature activation+1.37e+00

Pos logit contributions
arest+3.098
unia+2.890
asa+2.286
ine+2.266
 replay+2.138
Neg logit contributions
velt-4.974
catentry-4.770
atical-4.471
REDACTED-4.467
 gamb-4.461
Token2004
Token ID15724
Feature activation-2.30e-01

Pos logit contributions
arest+6.126
unia+5.240
Depth+4.315
asa+4.018
ine+3.985
Neg logit contributions
velt-17.366
 destro-17.001
 gamb-16.213
 scapego-16.079
 Gamb-15.399
Token John
Token ID1757
Feature activation-1.38e-01

Pos logit contributions
velt+2.059
catentry+1.862
REDACTED+1.851
••+1.827
joy+1.789
Neg logit contributions
arest-2.215
unia-2.134
asa-1.779
ounding-1.732
 replay-1.662
Token R
Token ID371
Feature activation-1.85e-01

Pos logit contributions
velt+1.036
catentry+0.933
REDACTED+0.904
••+0.881
atical+0.864
Neg logit contributions
arest-1.539
unia-1.485
asa-1.276
ounding-1.248
 Grammy-1.225
Tokenitter
Token ID1967
Feature activation+7.66e-01

Pos logit contributions
velt+1.492
catentry+1.318
REDACTED+1.298
••+1.258
 gamb+1.248
Neg logit contributions
arest-1.941
unia-1.887
asa-1.593
 Grammy-1.564
 Dame-1.546
Token
Token ID198
Feature activation+3.34e-01

Pos logit contributions
arest+5.070
unia+4.651
asa+3.784
 Grammy+3.731
ounding+3.485
Neg logit contributions
velt-8.564
catentry-8.096
REDACTED-7.881
joy-7.764
atical-7.717
Token
Token ID198
Feature activation+4.29e-01

Pos logit contributions
arest+7.426
unia+6.561
 Grammy+5.644
ine+5.613
asa+5.508
Neg logit contributions
velt-13.919
catentry-13.331
REDACTED-12.689
••-12.592
joy-12.590
Token
Token ID198
Feature activation+1.61e+00

Pos logit contributions
arest+3.064
unia+2.869
asa+2.280
ine+2.241
 replay+2.129
Neg logit contributions
velt-4.866
catentry-4.666
REDACTED-4.374
atical-4.366
••-4.350
Token2003
Token ID16088
Feature activation-9.33e-02

Pos logit contributions
arest+7.665
unia+6.108
ine+5.715
rel+5.621
Depth+5.320
Neg logit contributions
velt-17.715
 scapego-17.179
 gamb-16.768
 Typh-16.555
 Gamb-16.263
Token Ray
Token ID7760
Feature activation-1.74e-01

Pos logit contributions
velt+0.834
catentry+0.760
REDACTED+0.746
••+0.731
atical+0.718
Neg logit contributions
arest-0.909
unia-0.873
asa-0.730
ounding-0.711
 Grammy-0.688
Token Rom
Token ID3570
Feature activation-4.72e-02

Pos logit contributions
velt+1.208
catentry+1.054
REDACTED+1.036
••+0.999
 gamb+0.992
Neg logit contributions
arest-2.030
unia-1.955
asa-1.701
ounding-1.673
 Grammy-1.615
Tokenano
Token ID5733
Feature activation+1.36e+00

Pos logit contributions
velt+0.281
catentry+0.244
REDACTED+0.235
••+0.227
atical+0.222
Neg logit contributions
arest-0.601
unia-0.583
asa-0.511
ounding-0.502
 Grammy-0.495
Token 2004
Token ID5472
Feature activation+5.61e-02

Pos logit contributions
arest+7.428
unia+6.268
 Grammy+5.540
ine+5.360
asa+5.282
Neg logit contributions
velt-14.881
catentry-14.014
••-13.475
REDACTED-13.369
atical-13.321
Token Kel
Token ID15150
Feature activation+3.31e-01

Pos logit contributions
arest+0.529
unia+0.506
asa+0.422
ounding+0.411
 Grammy+0.408
Neg logit contributions
velt-0.519
catentry-0.480
REDACTED-0.466
••-0.454
atical-0.449
Tokensey
Token ID4397
Feature activation+1.74e-01

Pos logit contributions
arest+2.927
unia+2.794
asa+2.303
ine+2.203
ounding+2.178
Neg logit contributions
velt-3.237
catentry-3.050
REDACTED-2.976
••-2.913
 gamb-2.860
Token Gram
Token ID20159
Feature activation-2.04e-02

Pos logit contributions
arest+1.254
unia+1.203
asa+0.926
 Grammy+0.908
ounding+0.885
Neg logit contributions
velt-1.981
catentry-1.867
REDACTED-1.821
••-1.779
atical-1.779
Tokenmer
Token ID647
Feature activation+7.38e-01

Pos logit contributions
velt+0.207
catentry+0.192
REDACTED+0.188
••+0.184
atical+0.182
Neg logit contributions
arest-0.173
unia-0.165
asa-0.134
ounding-0.130
 Grammy-0.127
Token for
Token ID329
Feature activation+7.07e-02

Pos logit contributions
arest+2.320
unia+2.205
asa+1.850
ounding+1.806
ine+1.765
Neg logit contributions
velt-2.255
catentry-2.134
REDACTED-2.071
••-2.026
atical-1.986
Token oil
Token ID3056
Feature activation+3.30e-01

Pos logit contributions
arest+0.660
unia+0.631
asa+0.527
ounding+0.510
 Grammy+0.506
Neg logit contributions
velt-0.660
catentry-0.609
REDACTED-0.594
••-0.578
atical-0.572
Token and
Token ID290
Feature activation+1.42e-01

Pos logit contributions
arest+2.794
unia+2.711
asa+2.182
ounding+2.143
ine+2.120
Neg logit contributions
velt-3.238
catentry-3.062
REDACTED-2.998
••-2.911
atical-2.859
Token red
Token ID2266
Feature activation-1.63e-01

Pos logit contributions
arest+1.433
unia+1.378
asa+1.166
 Grammy+1.131
ounding+1.131
Neg logit contributions
velt-1.216
catentry-1.124
REDACTED-1.084
••-1.050
atical-1.036
Token for
Token ID329
Feature activation+5.73e-01

Pos logit contributions
velt+1.452
catentry+1.324
REDACTED+1.292
••+1.265
atical+1.247
Neg logit contributions
arest-1.597
unia-1.533
asa-1.286
ounding-1.251
 Grammy-1.224
Token natural
Token ID3288
Feature activation+1.33e+00

Pos logit contributions
arest+3.169
unia+3.004
 replay+2.118
 Grammy+2.096
asa+2.085
Neg logit contributions
velt-7.295
catentry-6.953
REDACTED-6.873
••-6.562
joy-6.539
Token gas
Token ID3623
Feature activation+5.22e-01

Pos logit contributions
unia+3.373
arest+2.950
 gas+2.118
 replay+1.777
ine+1.681
Neg logit contributions
catentry-18.856
velt-18.427
REDACTED-17.988
riber-17.346
ItemTracker-17.140
Token."
Token ID526
Feature activation+5.26e-02

Pos logit contributions
arest+4.218
unia+4.067
ine+3.160
asa+3.146
ounding+3.131
Neg logit contributions
velt-5.225
catentry-5.016
REDACTED-4.951
••-4.706
atical-4.638
Token Jan
Token ID2365
Feature activation+4.06e-01

Pos logit contributions
arest+0.464
unia+0.444
asa+0.366
ounding+0.353
 Grammy+0.351
Neg logit contributions
velt-0.517
catentry-0.481
REDACTED-0.468
atical-0.453
••-0.452
Token Re
Token ID797
Feature activation+6.80e-02

Pos logit contributions
arest+3.870
unia+3.719
asa+3.169
ine+3.101
 Grammy+3.011
Neg logit contributions
velt-3.557
catentry-3.467
REDACTED-3.248
atical-3.123
••-3.119
Tokenes
Token ID274
Feature activation-1.74e-01

Pos logit contributions
arest+0.540
unia+0.513
asa+0.410
ounding+0.396
 Grammy+0.384
Neg logit contributions
velt-0.732
catentry-0.682
REDACTED-0.668
••-0.656
atical-0.647
Token
Token ID198
Feature activation+4.45e-01

Pos logit contributions
arest+6.626
unia+5.916
 Grammy+4.802
asa+4.764
ine+4.694
Neg logit contributions
velt-11.218
catentry-10.854
REDACTED-10.391
••-10.315
atical-10.277
Token
Token ID198
Feature activation+1.65e+00

Pos logit contributions
arest+3.148
unia+2.944
asa+2.330
ine+2.299
 replay+2.179
Neg logit contributions
velt-5.060
catentry-4.860
atical-4.552
REDACTED-4.551
 gamb-4.535
Token2003
Token ID16088
Feature activation-1.85e-01

Pos logit contributions
arest+7.618
unia+6.014
ine+5.590
Depth+5.569
rel+5.499
Neg logit contributions
 destro-18.462
velt-18.434
 scapego-18.176
 gamb-17.699
 Typh-17.279
Token Eric
Token ID7651
Feature activation+6.79e-02

Pos logit contributions
velt+1.625
catentry+1.470
REDACTED+1.453
••+1.427
joy+1.398
Neg logit contributions
arest-1.822
unia-1.753
asa-1.469
ounding-1.434
 replay-1.377
Token McCorm
Token ID42037
Feature activation-1.73e-01

Pos logit contributions
arest+0.744
unia+0.717
asa+0.618
ounding+0.599
 Grammy+0.595
Neg logit contributions
velt-0.524
catentry-0.476
REDACTED-0.460
••-0.447
atical-0.441
Tokenack
Token ID441
Feature activation+1.31e+00

Pos logit contributions
velt+1.204
catentry+1.048
atical+1.007
REDACTED+1.005
••+0.987
Neg logit contributions
arest-2.019
unia-1.951
asa-1.664
ounding-1.657
 Grammy-1.657
Token
Token ID198
Feature activation+4.29e-01

Pos logit contributions
arest+7.426
unia+6.561
 Grammy+5.644
ine+5.613
asa+5.508
Neg logit contributions
velt-13.919
catentry-13.331
REDACTED-12.689
••-12.592
joy-12.590
Token
Token ID198
Feature activation+1.61e+00

Pos logit contributions
arest+3.064
unia+2.869
asa+2.280
ine+2.241
 replay+2.129
Neg logit contributions
velt-4.866
catentry-4.666
REDACTED-4.374
atical-4.366
••-4.350
Token2003
Token ID16088
Feature activation-9.33e-02

Pos logit contributions
arest+7.665
unia+6.108
ine+5.715
rel+5.621
Depth+5.320
Neg logit contributions
velt-17.715
 scapego-17.179
 gamb-16.768
 Typh-16.555
 Gamb-16.263
Token Ray
Token ID7760
Feature activation-1.74e-01

Pos logit contributions
velt+0.834
catentry+0.760
REDACTED+0.746
••+0.731
atical+0.718
Neg logit contributions
arest-0.909
unia-0.873
asa-0.730
ounding-0.711
 Grammy-0.688
Token Rom
Token ID3570
Feature activation-4.72e-02

Pos logit contributions
velt+1.208
catentry+1.054
REDACTED+1.036
••+0.999
 gamb+0.992
Neg logit contributions
arest-2.030
unia-1.955
asa-1.701
ounding-1.673
 Grammy-1.615
Token
Token ID198
Feature activation+1.47e+00

Pos logit contributions
arest+2.229
unia+2.103
asa+1.683
ine+1.626
 replay+1.576
Neg logit contributions
velt-3.213
catentry-3.049
REDACTED-2.900
••-2.866
atical-2.863
Token2004
Token ID15724
Feature activation-2.51e-01

Pos logit contributions
arest+5.591
unia+4.689
Depth+3.839
asa+3.528
ine+3.517
Neg logit contributions
velt-19.141
 destro-18.144
 scapego-17.594
 gamb-17.584
 Typh-16.800
Token Larry
Token ID13633
Feature activation+5.35e-02

Pos logit contributions
velt+2.244
catentry+2.026
REDACTED+2.024
••+1.998
joy+1.956
Neg logit contributions
arest-2.403
unia-2.315
asa-1.928
ounding-1.871
 replay-1.789
Token David
Token ID3271
Feature activation+6.48e-01

Pos logit contributions
arest+0.445
unia+0.424
asa+0.343
ounding+0.332
 Grammy+0.323
Neg logit contributions
velt-0.556
catentry-0.516
REDACTED-0.505
••-0.496
atical-0.489
Token
Token ID198
Feature activation+4.99e-01

Pos logit contributions
arest+4.287
unia+3.990
asa+3.182
ine+3.141
 replay+2.993
Neg logit contributions
velt-7.382
catentry-7.149
atical-6.768
REDACTED-6.660
••-6.611
Token
Token ID198
Feature activation+1.31e+00

Pos logit contributions
arest+3.431
unia+3.184
ine+2.516
asa+2.506
rel+2.351
Neg logit contributions
velt-5.738
catentry-5.525
atical-5.171
 gamb-5.163
REDACTED-5.138
Token2004
Token ID15724
Feature activation-2.21e-01

Pos logit contributions
arest+5.434
unia+4.529
Depth+3.615
asa+3.460
ine+3.244
Neg logit contributions
velt-17.310
 destro-16.872
 gamb-16.142
 scapego-15.999
 Typh-15.391
Token Matt
Token ID4705
Feature activation+2.59e-02

Pos logit contributions
velt+1.923
catentry+1.729
REDACTED+1.719
••+1.695
joy+1.666
Neg logit contributions
arest-2.183
unia-2.102
asa-1.768
ounding-1.719
 replay-1.653
Token Le
Token ID1004
Feature activation+9.56e-01

Pos logit contributions
arest+0.211
unia+0.201
asa+0.162
ounding+0.157
 Grammy+0.152
Neg logit contributions
velt-0.274
catentry-0.254
REDACTED-0.249
••-0.245
atical-0.241
TokenBl
Token ID3629
Feature activation+4.83e-03

Pos logit contributions
arest+6.480
unia+6.043
Depth+4.706
ine+4.546
rel+4.542
Neg logit contributions
velt-10.587
 Typh-9.529
catentry-9.450
atical-9.361
 scapego-9.315
Tokenanc
Token ID1192
Feature activation+9.97e-01

Pos logit contributions
arest+0.041
unia+0.039
asa+0.032
ounding+0.031
 Grammy+0.030
Neg logit contributions
velt-0.050
catentry-0.046
REDACTED-0.045
••-0.044
 gamb-0.043
Token on
Token ID319
Feature activation+2.48e-02

Pos logit contributions
arest+0.973
unia+0.925
asa+0.767
ounding+0.744
 Grammy+0.725
Neg logit contributions
velt-1.086
catentry-1.009
REDACTED-0.979
••-0.959
atical-0.944
Token the
Token ID262
Feature activation-6.27e-02

Pos logit contributions
arest+0.228
unia+0.217
asa+0.181
ounding+0.176
 Grammy+0.172
Neg logit contributions
velt-0.235
catentry-0.217
REDACTED-0.211
••-0.206
atical-0.204
Token rifle
Token ID11575
Feature activation-2.29e-02

Pos logit contributions
velt+0.533
catentry+0.487
REDACTED+0.474
••+0.463
 gamb+0.456
Neg logit contributions
arest-0.639
unia-0.615
asa-0.518
ounding-0.506
 Grammy-0.495
Token.
Token ID13
Feature activation-1.52e-01

Pos logit contributions
velt+0.220
catentry+0.203
REDACTED+0.198
••+0.194
atical+0.191
Neg logit contributions
arest-0.208
unia-0.199
asa-0.165
ounding-0.160
 Grammy-0.156
Token The
Token ID383
Feature activation-1.20e-01

Pos logit contributions
velt+1.285
catentry+1.176
REDACTED+1.141
••+1.121
 gamb+1.098
Neg logit contributions
arest-1.549
unia-1.494
asa-1.254
ounding-1.231
 Grammy-1.193
Token Model
Token ID9104
Feature activation+1.30e+00

Pos logit contributions
velt+1.063
catentry+0.973
REDACTED+0.949
••+0.926
 gamb+0.917
Neg logit contributions
arest-1.178
unia-1.138
asa-0.947
ounding-0.925
 Grammy-0.904
Token of
Token ID286
Feature activation+5.79e-01

Pos logit contributions
arest+9.079
unia+8.052
asa+7.753
ine+7.661
rel+7.080
Neg logit contributions
velt-11.839
catentry-11.340
NetMessage-10.761
REDACTED-10.703
joy-10.459
Token 1903
Token ID41625
Feature activation+3.41e-01

Pos logit contributions
arest+4.108
unia+3.642
asa+3.103
ounding+2.919
ine+2.914
Neg logit contributions
velt-6.315
catentry-5.982
REDACTED-5.855
atical-5.726
joy-5.693
Token used
Token ID973
Feature activation+7.65e-02

Pos logit contributions
arest+3.033
unia+2.885
asa+2.468
ine+2.338
ounding+2.335
Neg logit contributions
velt-3.270
catentry-3.060
REDACTED-2.932
••-2.846
atical-2.840
Token tang
Token ID13875
Feature activation-3.86e-01

Pos logit contributions
arest+0.682
unia+0.648
asa+0.538
ounding+0.522
 Grammy+0.511
Neg logit contributions
velt-0.746
catentry-0.694
REDACTED-0.674
••-0.659
atical-0.651
Tokenent
Token ID298
Feature activation+4.22e-01

Pos logit contributions
velt+3.452
catentry+3.201
REDACTED+3.102
••+3.091
 gamb+3.084
Neg logit contributions
arest-3.618
unia-3.487
asa-2.819
 Dame-2.752
 Grammy-2.747
Token in
Token ID287
Feature activation-2.73e-01

Pos logit contributions
velt+3.324
catentry+2.957
REDACTED+2.935
 gamb+2.900
••+2.872
Neg logit contributions
arest-3.689
unia-3.558
asa-2.969
ounding-2.931
 Grammy-2.840
Token white
Token ID2330
Feature activation-3.01e-01

Pos logit contributions
velt+2.431
catentry+2.217
REDACTED+2.169
••+2.136
 gamb+2.130
Neg logit contributions
arest-2.649
unia-2.551
asa-2.133
ounding-2.089
 Grammy-2.019
Token and
Token ID290
Feature activation-2.22e-01

Pos logit contributions
velt+2.794
catentry+2.522
REDACTED+2.473
••+2.456
 gamb+2.450
Neg logit contributions
arest-2.821
unia-2.700
asa-2.243
ounding-2.202
 Grammy-2.135
Token is
Token ID318
Feature activation-1.32e-01

Pos logit contributions
velt+1.916
catentry+1.724
REDACTED+1.702
••+1.675
 gamb+1.673
Neg logit contributions
arest-2.211
unia-2.128
asa-1.787
ounding-1.769
 Grammy-1.700
Token now
Token ID783
Feature activation+1.30e-01

Pos logit contributions
velt+1.155
catentry+1.056
REDACTED+1.034
••+1.011
 gamb+1.004
Neg logit contributions
arest-1.297
unia-1.247
asa-1.044
ounding-1.025
 Grammy-0.999
Token Gand
Token ID17727
Feature activation-1.62e+00

Pos logit contributions
arest+1.233
unia+1.177
asa+0.992
ounding+0.946
ine+0.937
Neg logit contributions
velt-1.221
catentry-1.136
REDACTED-1.087
••-1.066
atical-1.044
Tokenalf
Token ID1604
Feature activation-7.20e-01

Pos logit contributions
velt+7.630
joy+6.355
atical+6.201
nah+6.093
andal+5.995
Neg logit contributions
arest-15.933
unia-15.589
 Grammy-15.377
ailability-15.115
 replay-14.798
Token the
Token ID262
Feature activation-4.27e-01

Pos logit contributions
velt+5.886
 gamb+5.134
REDACTED+5.131
joy+5.115
••+5.112
Neg logit contributions
arest-6.972
unia-6.671
ounding-5.684
asa-5.588
 Grammy-5.562
Token White
Token ID2635
Feature activation-5.43e-01

Pos logit contributions
velt+3.430
catentry+3.113
 gamb+3.082
 Gamb+3.054
••+3.036
Neg logit contributions
arest-4.367
unia-4.230
asa-3.564
ounding-3.544
 Grammy-3.397
Token,
Token ID11
Feature activation-5.29e-01

Pos logit contributions
velt+4.851
joy+4.292
••+4.285
catentry+4.254
REDACTED+4.241
Neg logit contributions
arest-5.132
unia-4.940
asa-4.136
ounding-4.071
 Grammy-3.925
Token for
Token ID329
Feature activation-2.19e-01

Pos logit contributions
velt+4.461
 gamb+3.964
catentry+3.956
REDACTED+3.931
 Gamb+3.915
Neg logit contributions
arest-5.142
unia-4.987
ounding-4.196
asa-4.174
 Grammy-3.974
Token,
Token ID11
Feature activation-7.33e-01

Pos logit contributions
velt+4.187
REDACTED+3.712
catentry+3.710
joy+3.662
 gamb+3.630
Neg logit contributions
arest-4.299
unia-4.067
asa-3.354
ounding-3.353
 replay-3.282
Token Ar
Token ID943
Feature activation-6.57e-01

Pos logit contributions
velt+6.319
 Gamb+5.838
 Typh+5.777
joy+5.735
 gamb+5.698
Neg logit contributions
arest-6.751
unia-6.248
ounding-5.465
asa-5.259
Depth-5.014
Tokenag
Token ID363
Feature activation-3.23e-01

Pos logit contributions
velt+6.237
joy+5.639
nah+5.585
atical+5.563
zai+5.427
Neg logit contributions
arest-5.440
unia-4.867
 equival-4.226
 Grammy-4.213
 replay-4.165
Tokenorn
Token ID1211
Feature activation-3.70e-01

Pos logit contributions
velt+2.934
REDACTED+2.601
catentry+2.588
atical+2.576
nah+2.566
Neg logit contributions
arest-2.971
unia-2.892
ounding-2.357
 Grammy-2.347
asa-2.329
Token,
Token ID11
Feature activation-7.68e-01

Pos logit contributions
velt+3.440
catentry+3.113
REDACTED+3.069
••+3.026
 gamb+3.002
Neg logit contributions
arest-3.434
unia-3.278
asa-2.705
ounding-2.656
 replay-2.589
Token Gand
Token ID17727
Feature activation-1.50e+00

Pos logit contributions
velt+6.438
 Gamb+5.991
 Typh+5.934
joy+5.900
 gamb+5.828
Neg logit contributions
arest-7.141
unia-6.608
ounding-5.813
asa-5.578
Depth-5.360
Tokenalf
Token ID1604
Feature activation-6.20e-01

Pos logit contributions
velt+5.586
joy+4.117
atical+3.995
 gamb+3.898
nah+3.882
Neg logit contributions
arest-17.170
unia-16.211
 Grammy-15.949
ailability-15.858
 ov-15.682
Token,
Token ID11
Feature activation-8.80e-01

Pos logit contributions
velt+5.407
REDACTED+4.774
 gamb+4.745
catentry+4.744
••+4.724
Neg logit contributions
arest-5.899
unia-5.580
asa-4.665
ounding-4.636
 replay-4.526
Token Gim
Token ID41123
Feature activation-1.22e+00

Pos logit contributions
velt+7.160
 Gamb+6.744
 Typh+6.646
joy+6.596
 gamb+6.570
Neg logit contributions
arest-8.098
unia-7.472
ounding-6.700
asa-6.363
Depth-6.184
Tokenli
Token ID4528
Feature activation-5.93e-01

Pos logit contributions
velt+4.262
nah+3.870
 Gamb+3.817
••+3.653
joy+3.485
Neg logit contributions
arest-15.222
unia-14.893
ounding-14.286
Depth-13.845
xton-13.836
Token the
Token ID262
Feature activation-4.26e-01

Pos logit contributions
velt+5.223
catentry+4.624
joy+4.581
REDACTED+4.572
 gamb+4.561
Neg logit contributions
arest-5.584
unia-5.335
asa-4.415
ounding-4.397
 replay-4.327
Token.
Token ID13
Feature activation-5.10e-01

Pos logit contributions
velt+0.551
catentry+0.503
REDACTED+0.490
••+0.479
atical+0.471
Neg logit contributions
arest-0.649
unia-0.624
asa-0.527
ounding-0.514
 Grammy-0.503
Token There
Token ID1318
Feature activation-3.20e-02

Pos logit contributions
velt+4.143
 Gamb+3.668
••+3.612
joy+3.585
catentry+3.567
Neg logit contributions
arest-5.209
unia-4.974
ounding-4.255
asa-4.186
uple-4.015
Token they
Token ID484
Feature activation-9.21e-01

Pos logit contributions
velt+0.314
catentry+0.292
REDACTED+0.284
••+0.278
atical+0.275
Neg logit contributions
arest-0.284
unia-0.272
asa-0.224
ounding-0.217
 Grammy-0.212
Token unexpectedly
Token ID25884
Feature activation-1.21e+00

Pos logit contributions
velt+6.804
 gamb+6.329
atical+6.007
 scapego+5.837
 Typh+5.822
Neg logit contributions
arest-9.260
unia-8.739
asa-7.621
ounding-7.610
Depth-7.468
Token meet
Token ID1826
Feature activation-2.93e-01

Pos logit contributions
velt+8.142
 gamb+7.799
 Typh+7.244
 scapego+7.205
atical+7.201
Neg logit contributions
arest-11.868
unia-11.253
asa-9.900
ounding-9.803
Depth-9.692
Token Gand
Token ID17727
Feature activation-1.50e+00

Pos logit contributions
velt+2.609
catentry+2.367
REDACTED+2.336
 gamb+2.287
••+2.279
Neg logit contributions
arest-2.829
unia-2.684
asa-2.248
ounding-2.225
 Grammy-2.138
Tokenalf
Token ID1604
Feature activation-3.06e-01

Pos logit contributions
velt+5.709
joy+4.512
atical+4.047
nah+4.034
 gamb+3.972
Neg logit contributions
arest-16.813
unia-16.162
 Grammy-15.963
ailability-15.673
 ov-15.272
Token.
Token ID13
Feature activation-4.54e-01

Pos logit contributions
velt+2.798
catentry+2.504
REDACTED+2.479
••+2.442
 gamb+2.440
Neg logit contributions
arest-2.881
unia-2.757
asa-2.293
ounding-2.255
 Grammy-2.207
Token Gand
Token ID17727
Feature activation-1.18e+00

Pos logit contributions
velt+3.694
 Gamb+3.260
••+3.243
joy+3.217
catentry+3.201
Neg logit contributions
arest-4.634
unia-4.467
ounding-3.788
asa-3.754
uple-3.591
Tokenalf
Token ID1604
Feature activation-4.04e-01

Pos logit contributions
velt+5.265
joy+4.029
 gamb+4.012
atical+3.793
bard+3.712
Neg logit contributions
arest-13.655
unia-13.619
 Grammy-12.767
Û-12.529
ailability-12.514
Token explains
Token ID6688
Feature activation-2.99e-01

Pos logit contributions
velt+3.294
 gamb+2.917
REDACTED+2.871
catentry+2.847
joy+2.803
Neg logit contributions
arest-4.124
unia-3.994
ounding-3.378
asa-3.364
 Grammy-3.256
Token,
Token ID11
Feature activation-5.08e-01

Pos logit contributions
velt+2.926
REDACTED+2.577
catentry+2.576
 gamb+2.537
••+2.503
Neg logit contributions
arest-3.492
unia-3.353
asa-2.813
ounding-2.804
 Grammy-2.710
Token with
Token ID351
Feature activation-3.16e-01

Pos logit contributions
velt+4.161
 gamb+3.724
catentry+3.706
joy+3.681
 Gamb+3.677
Neg logit contributions
arest-5.063
unia-4.852
asa-4.088
ounding-4.081
 Grammy-3.893
Token the
Token ID262
Feature activation-3.26e-01

Pos logit contributions
velt+2.706
catentry+2.450
REDACTED+2.418
••+2.354
joy+2.353
Neg logit contributions
arest-3.161
unia-3.029
asa-2.555
ounding-2.491
 Grammy-2.424
Token advice
Token ID5608
Feature activation-1.26e-01

Pos logit contributions
velt+3.151
catentry+2.921
REDACTED+2.861
 gamb+2.802
••+2.796
Neg logit contributions
arest-2.886
unia-2.771
asa-2.276
ounding-2.202
 Grammy-2.117
Token of
Token ID286
Feature activation-6.55e-01

Pos logit contributions
velt+1.261
catentry+1.169
REDACTED+1.141
••+1.120
atical+1.104
Neg logit contributions
arest-1.093
unia-1.044
asa-0.853
ounding-0.828
 Grammy-0.808
Token Gand
Token ID17727
Feature activation-1.49e+00

Pos logit contributions
velt+5.379
 Gamb+4.913
REDACTED+4.861
joy+4.859
catentry+4.839
Neg logit contributions
arest-6.371
unia-6.025
ounding-5.071
asa-5.049
Depth-4.804
Tokenalf
Token ID1604
Feature activation-6.39e-01

Pos logit contributions
velt+5.291
nah+3.849
joy+3.691
 gamb+3.565
atical+3.229
Neg logit contributions
arest-17.446
unia-16.755
 Grammy-16.310
ailability-16.252
 Dame-15.862
Token,
Token ID11
Feature activation-3.56e-01

Pos logit contributions
velt+5.582
REDACTED+4.897
 gamb+4.882
catentry+4.864
••+4.824
Neg logit contributions
arest-6.124
unia-5.832
ounding-4.856
asa-4.843
 Grammy-4.681
Token chooses
Token ID19769
Feature activation-7.04e-02

Pos logit contributions
velt+2.908
REDACTED+2.550
catentry+2.547
 gamb+2.523
••+2.466
Neg logit contributions
arest-3.703
unia-3.561
ounding-3.016
asa-3.012
 Grammy-2.875
Token companions
Token ID19429
Feature activation-1.67e-01

Pos logit contributions
velt+0.653
catentry+0.603
REDACTED+0.586
••+0.573
atical+0.565
Neg logit contributions
arest-0.664
unia-0.637
asa-0.531
ounding-0.515
 Grammy-0.504
Token for
Token ID329
Feature activation-1.96e-03

Pos logit contributions
velt+1.518
catentry+1.377
REDACTED+1.353
••+1.325
 gamb+1.312
Neg logit contributions
arest-1.611
unia-1.543
asa-1.283
ounding-1.254
 Grammy-1.225
Token's
Token ID338
Feature activation-3.99e-01

Pos logit contributions
velt+4.669
REDACTED+3.999
catentry+3.982
 gamb+3.956
joy+3.951
Neg logit contributions
arest-6.238
unia-6.013
asa-5.072
ounding-4.994
 replay-4.921
Token The
Token ID383
Feature activation-1.01e+00

Pos logit contributions
velt+3.473
catentry+3.138
REDACTED+3.074
 gamb+3.072
••+3.037
Neg logit contributions
arest-3.914
unia-3.707
asa-3.103
ounding-3.052
 replay-2.918
Token Hobbit
Token ID43877
Feature activation-9.02e-01

Pos logit contributions
velt+5.895
 Gamb+5.843
joy+5.808
 gamb+5.531
••+5.496
Neg logit contributions
arest-10.847
unia-10.003
ounding-9.248
uple-8.559
xton-8.553
Token,
Token ID11
Feature activation-3.06e-01

Pos logit contributions
velt+7.239
 gamb+6.273
atical+6.237
••+6.210
joy+6.073
Neg logit contributions
arest-8.483
unia-8.277
asa-7.054
ounding-6.888
 Dame-6.722
Token introducing
Token ID16118
Feature activation-5.16e-01

Pos logit contributions
velt+2.763
catentry+2.511
REDACTED+2.455
••+2.428
 gamb+2.409
Neg logit contributions
arest-2.938
unia-2.810
asa-2.336
ounding-2.286
 Grammy-2.206
Token Gand
Token ID17727
Feature activation-1.46e+00

Pos logit contributions
velt+4.471
catentry+3.996
 gamb+3.953
REDACTED+3.952
 Typh+3.922
Neg logit contributions
arest-5.015
unia-4.765
asa-3.985
ounding-3.938
 Grammy-3.780
Tokenalf
Token ID1604
Feature activation-8.10e-01

Pos logit contributions
velt+6.042
joy+5.090
atical+4.760
rez+4.653
 gamb+4.603
Neg logit contributions
arest-15.897
unia-15.510
 Grammy-15.031
 Dame-14.617
Depth-14.538
Token.
Token ID13
Feature activation-3.02e-01

Pos logit contributions
velt+6.509
 gamb+5.750
••+5.702
joy+5.669
 Typh+5.617
Neg logit contributions
arest-7.831
unia-7.487
ounding-6.390
 Grammy-6.235
asa-6.178
Token
Token ID198
Feature activation-2.07e-01

Pos logit contributions
velt+2.987
catentry+2.748
REDACTED+2.694
••+2.658
 gamb+2.604
Neg logit contributions
arest-2.648
unia-2.529
asa-2.064
ounding-2.012
 Grammy-1.949
Token
Token ID198
Feature activation-5.74e-01

Pos logit contributions
velt+1.908
REDACTED+1.744
catentry+1.720
••+1.698
joy+1.672
Neg logit contributions
arest-1.920
unia-1.855
asa-1.522
ounding-1.522
 Grammy-1.460
TokenAL
Token ID1847
Feature activation-6.96e-01

Pos logit contributions
velt+4.832
REDACTED+4.590
••+4.502
catentry+4.430
joy+4.293
Neg logit contributions
arest-5.588
unia-5.448
ounding-4.459
asa-4.379
 replay-4.349
Token a
Token ID257
Feature activation+3.91e-01

Pos logit contributions
velt+2.089
REDACTED+1.933
catentry+1.919
••+1.872
 gamb+1.854
Neg logit contributions
arest-2.192
unia-2.133
ounding-1.738
asa-1.724
 Grammy-1.687
Token horrible
Token ID12361
Feature activation+7.74e-01

Pos logit contributions
arest+2.652
unia+2.437
asa+1.917
 replay+1.745
 Grammy+1.736
Neg logit contributions
velt-4.700
catentry-4.387
REDACTED-4.303
••-4.216
joy-4.132
Token scene
Token ID3715
Feature activation-7.69e-02

Pos logit contributions
arest+5.457
unia+4.985
asa+4.149
ine+4.112
 replay+4.086
Neg logit contributions
velt-8.334
catentry-7.966
••-7.788
-7.601
REDACTED-7.521
Token of
Token ID286
Feature activation+5.37e-02

Pos logit contributions
velt+0.737
catentry+0.678
REDACTED+0.672
••+0.657
atical+0.645
Neg logit contributions
arest-0.695
unia-0.666
asa-0.547
ounding-0.539
 Grammy-0.522
Token a
Token ID257
Feature activation-3.13e-01

Pos logit contributions
arest+0.464
unia+0.442
asa+0.361
ounding+0.350
 Grammy+0.341
Neg logit contributions
velt-0.541
catentry-0.501
REDACTED-0.489
••-0.480
atical-0.474
Token child
Token ID1200
Feature activation-1.38e+00

Pos logit contributions
velt+2.003
catentry+1.796
 gamb+1.780
REDACTED+1.760
••+1.723
Neg logit contributions
arest-3.746
unia-3.608
asa-3.124
ounding-3.089
 Grammy-3.009
Token being
Token ID852
Feature activation-2.27e-01

Pos logit contributions
velt+1.735
 gamb+1.632
REDACTED+1.512
 bal+1.421
nah+1.171
Neg logit contributions
arest-19.852
unia-19.552
ailability-18.555
Û-17.873
��-17.601
Token eaten
Token ID17065
Feature activation-4.78e-01

Pos logit contributions
velt+1.462
catentry+1.277
REDACTED+1.276
 gamb+1.242
••+1.234
Neg logit contributions
arest-2.747
unia-2.671
asa-2.292
ounding-2.291
 Grammy-2.233
Token alive
Token ID6776
Feature activation-8.26e-01

Pos logit contributions
velt+3.218
REDACTED+2.812
catentry+2.637
joy+2.608
 gamb+2.605
Neg logit contributions
arest-5.584
unia-5.510
asa-4.693
 Grammy-4.645
ounding-4.634
Token by
Token ID416
Feature activation-1.79e-01

Pos logit contributions
velt+5.396
REDACTED+4.795
••+4.707
joy+4.598
catentry+4.536
Neg logit contributions
arest-9.170
unia-8.963
ailability-8.007
asa-7.654
ounding-7.631
Token As
Token ID1081
Feature activation-1.30e-01

Pos logit contributions
velt+1.692
catentry+1.559
REDACTED+1.523
••+1.492
atical+1.470
Neg logit contributions
arest-1.660
unia-1.588
asa-1.315
ounding-1.281
 Grammy-1.250
Token of
Token ID286
Feature activation-7.83e-02

Pos logit contributions
velt+1.030
catentry+0.947
REDACTED+0.922
••+0.910
atical+0.898
Neg logit contributions
arest-1.012
unia-0.968
asa-0.802
ounding-0.779
 Grammy-0.761
Token 48
Token ID4764
Feature activation+4.46e-01

Pos logit contributions
velt+0.678
catentry+0.618
REDACTED+0.603
••+0.594
atical+0.584
Neg logit contributions
arest-0.785
unia-0.754
asa-0.634
ounding-0.620
 Grammy-0.604
Token%).
Token ID18823
Feature activation-1.50e-01

Pos logit contributions
arest+3.877
unia+3.570
asa+3.018
ounding+2.932
ine+2.875
Neg logit contributions
velt-4.386
catentry-4.184
REDACTED-3.967
atical-3.818
••-3.804
Token It
Token ID632
Feature activation+2.66e-02

Pos logit contributions
velt+1.305
catentry+1.194
REDACTED+1.166
••+1.152
 gamb+1.119
Neg logit contributions
arest-1.489
unia-1.433
asa-1.196
ounding-1.180
 Grammy-1.134
Token remains
Token ID3793
Feature activation+1.92e-01

Pos logit contributions
arest+0.247
unia+0.236
asa+0.196
ounding+0.190
 Grammy+0.186
Neg logit contributions
velt-0.252
catentry-0.233
REDACTED-0.226
••-0.222
atical-0.218
Token to
Token ID284
Feature activation-1.38e+00

Pos logit contributions
arest+1.742
unia+1.643
asa+1.371
ounding+1.319
ine+1.303
Neg logit contributions
velt-1.868
catentry-1.723
REDACTED-1.668
••-1.619
joy-1.604
Token be
Token ID307
Feature activation-5.86e-01

Pos logit contributions
velt+7.762
REDACTED+7.139
catentry+7.106
atical+6.812
 gamb+6.663
Neg logit contributions
arest-13.983
unia-13.908
asa-12.501
ounding-11.784
ailability-11.676
Token seen
Token ID1775
Feature activation-2.81e-01

Pos logit contributions
velt+3.631
 gamb+3.182
catentry+3.163
REDACTED+3.078
atical+3.063
Neg logit contributions
arest-6.939
unia-6.814
asa-5.876
ounding-5.867
 Grammy-5.693
Token whether
Token ID1771
Feature activation-1.86e-01

Pos logit contributions
velt+2.273
catentry+2.043
REDACTED+2.000
••+1.974
 gamb+1.944
Neg logit contributions
arest-2.957
unia-2.860
asa-2.416
ounding-2.368
 Grammy-2.313
Token the
Token ID262
Feature activation-1.71e-01

Pos logit contributions
velt+1.644
catentry+1.504
REDACTED+1.467
••+1.447
 gamb+1.422
Neg logit contributions
arest-1.819
unia-1.745
asa-1.455
ounding-1.432
 Grammy-1.391
Token C
Token ID327
Feature activation-9.76e-02

Pos logit contributions
velt+1.499
catentry+1.369
REDACTED+1.339
••+1.308
 gamb+1.300
Neg logit contributions
arest-1.681
unia-1.620
asa-1.355
ounding-1.328
 Grammy-1.290
Token where
Token ID810
Feature activation-3.12e-01

Pos logit contributions
velt+0.025
catentry+0.023
REDACTED+0.023
••+0.022
atical+0.022
Neg logit contributions
arest-0.026
unia-0.025
asa-0.021
ounding-0.020
 Grammy-0.020
Token it
Token ID340
Feature activation-4.93e-01

Pos logit contributions
velt+2.602
catentry+2.332
REDACTED+2.297
••+2.243
 gamb+2.233
Neg logit contributions
arest-3.206
unia-3.052
asa-2.592
ounding-2.560
 Grammy-2.480
Token was
Token ID373
Feature activation-6.07e-01

Pos logit contributions
velt+4.417
REDACTED+3.957
 gamb+3.951
catentry+3.847
atical+3.835
Neg logit contributions
arest-4.655
unia-4.478
asa-3.743
ounding-3.697
 Grammy-3.555
Token forged
Token ID24918
Feature activation-4.16e-01

Pos logit contributions
velt+5.037
 gamb+4.610
REDACTED+4.585
joy+4.394
 Typh+4.373
Neg logit contributions
arest-5.929
unia-5.741
ounding-4.794
asa-4.790
 Grammy-4.574
Token.
Token ID13
Feature activation-6.27e-01

Pos logit contributions
velt+3.644
REDACTED+3.264
catentry+3.225
••+3.192
 gamb+3.182
Neg logit contributions
arest-4.041
unia-3.876
asa-3.249
ounding-3.181
 Grammy-3.095
Token Fro
Token ID9734
Feature activation-1.38e+00

Pos logit contributions
velt+5.057
 Typh+4.630
 Gamb+4.547
••+4.524
REDACTED+4.516
Neg logit contributions
arest-6.337
unia-6.022
ounding-5.127
asa-5.068
uple-4.866
Tokendo
Token ID4598
Feature activation-7.84e-01

Pos logit contributions
velt+4.742
joy+3.921
nah+3.782
 gamb+3.587
 bal+3.316
Neg logit contributions
arest-17.895
unia-16.882
ounding-15.650
awei-15.368
Depth-15.339
Token takes
Token ID2753
Feature activation-1.26e-02

Pos logit contributions
velt+6.289
 gamb+5.660
REDACTED+5.431
••+5.363
 Gamb+5.360
Neg logit contributions
arest-7.827
unia-7.543
ounding-6.443
asa-6.269
Depth-6.199
Token this
Token ID428
Feature activation-2.43e-03

Pos logit contributions
velt+0.113
catentry+0.104
REDACTED+0.101
••+0.099
atical+0.098
Neg logit contributions
arest-0.122
unia-0.118
asa-0.098
ounding-0.096
 Grammy-0.094
Token task
Token ID4876
Feature activation+8.99e-02

Pos logit contributions
velt+0.023
catentry+0.021
REDACTED+0.021
••+0.020
atical+0.020
Neg logit contributions
arest-0.022
unia-0.021
asa-0.018
ounding-0.017
 Grammy-0.017
Token upon
Token ID2402
Feature activation-4.48e-01

Pos logit contributions
arest+0.837
unia+0.802
asa+0.669
ounding+0.650
 Grammy+0.634
Neg logit contributions
velt-0.844
catentry-0.779
REDACTED-0.757
••-0.744
atical-0.728
Token midterm
Token ID44066
Feature activation-3.28e-01

Pos logit contributions
arest+2.043
unia+1.940
asa+1.599
 Grammy+1.581
ounding+1.579
Neg logit contributions
velt-2.093
catentry-1.942
REDACTED-1.923
atical-1.815
••-1.815
Token election
Token ID3071
Feature activation-1.63e-01

Pos logit contributions
velt+3.002
catentry+2.763
••+2.693
 gamb+2.676
REDACTED+2.666
Neg logit contributions
arest-3.036
unia-2.921
asa-2.445
ounding-2.367
 Dame-2.310
Token.
Token ID13
Feature activation-2.79e-01

Pos logit contributions
velt+1.515
catentry+1.393
REDACTED+1.356
••+1.343
 gamb+1.322
Neg logit contributions
arest-1.521
unia-1.465
asa-1.216
ounding-1.183
 Grammy-1.153
Token It
Token ID632
Feature activation+5.11e-02

Pos logit contributions
velt+2.380
catentry+2.151
••+2.129
REDACTED+2.118
 Gamb+2.073
Neg logit contributions
arest-2.796
unia-2.703
asa-2.273
ounding-2.215
 replay-2.126
Token remains
Token ID3793
Feature activation+1.32e-01

Pos logit contributions
arest+0.478
unia+0.455
asa+0.380
ounding+0.369
 Grammy+0.361
Neg logit contributions
velt-0.479
catentry-0.444
REDACTED-0.432
••-0.420
atical-0.415
Token to
Token ID284
Feature activation-1.38e+00

Pos logit contributions
arest+1.183
unia+1.123
asa+0.931
ounding+0.905
 Grammy+0.899
Neg logit contributions
velt-1.280
catentry-1.181
REDACTED-1.147
••-1.106
atical-1.105
Token be
Token ID307
Feature activation-6.61e-01

Pos logit contributions
velt+7.836
catentry+6.964
REDACTED+6.870
atical+6.843
 gamb+6.790
Neg logit contributions
arest-14.135
unia-13.745
asa-12.393
ounding-11.674
ailability-11.580
Token seen
Token ID1775
Feature activation-8.22e-02

Pos logit contributions
velt+3.954
 gamb+3.486
catentry+3.351
••+3.294
atical+3.293
Neg logit contributions
arest-7.964
unia-7.824
asa-6.697
ounding-6.688
 Grammy-6.441
Token whether
Token ID1771
Feature activation-4.49e-01

Pos logit contributions
velt+0.706
catentry+0.646
REDACTED+0.628
••+0.612
atical+0.604
Neg logit contributions
arest-0.831
unia-0.798
asa-0.675
ounding-0.658
 Grammy-0.645
Token this
Token ID428
Feature activation-1.57e-01

Pos logit contributions
velt+3.784
••+3.399
catentry+3.378
REDACTED+3.354
 gamb+3.316
Neg logit contributions
arest-4.505
unia-4.313
asa-3.639
ounding-3.537
 replay-3.371
Token sort
Token ID3297
Feature activation-2.71e-01

Pos logit contributions
velt+1.467
catentry+1.344
REDACTED+1.312
••+1.301
 gamb+1.285
Neg logit contributions
arest-1.471
unia-1.407
asa-1.162
ounding-1.132
 Grammy-1.094
Token for
Token ID329
Feature activation+3.11e-01

Pos logit contributions
arest+4.806
unia+4.534
asa+3.784
ine+3.725
 replay+3.678
Neg logit contributions
velt-5.766
catentry-5.510
REDACTED-5.292
atical-5.040
joy-4.971
Token donations
Token ID10976
Feature activation+8.74e-02

Pos logit contributions
arest+2.809
unia+2.666
asa+2.229
 Grammy+2.178
ounding+2.173
Neg logit contributions
velt-2.959
catentry-2.752
REDACTED-2.694
atical-2.574
••-2.570
Token.
Token ID13
Feature activation-8.39e-02

Pos logit contributions
arest+0.794
unia+0.758
asa+0.629
ounding+0.612
 Grammy+0.600
Neg logit contributions
velt-0.840
catentry-0.782
REDACTED-0.759
••-0.739
atical-0.729
Token It
Token ID632
Feature activation+3.78e-01

Pos logit contributions
velt+0.705
catentry+0.638
REDACTED+0.624
••+0.619
 gamb+0.600
Neg logit contributions
arest-0.863
unia-0.833
asa-0.701
ounding-0.689
 replay-0.663
Token remains
Token ID3793
Feature activation+6.60e-02

Pos logit contributions
arest+3.410
unia+3.233
asa+2.743
ounding+2.601
ine+2.582
Neg logit contributions
velt-3.561
catentry-3.388
REDACTED-3.247
atical-3.141
••-3.109
Token to
Token ID284
Feature activation-1.37e+00

Pos logit contributions
arest+0.617
unia+0.588
asa+0.491
ounding+0.477
 Grammy+0.469
Neg logit contributions
velt-0.617
catentry-0.569
REDACTED-0.554
••-0.539
atical-0.535
Token be
Token ID307
Feature activation-8.03e-01

Pos logit contributions
velt+7.908
REDACTED+7.089
catentry+6.966
atical+6.657
nah+6.607
Neg logit contributions
arest-14.106
unia-13.683
asa-12.569
ounding-11.888
 Grammy-11.522
Token seen
Token ID1775
Feature activation-1.91e-01

Pos logit contributions
velt+4.050
 gamb+3.680
atical+3.378
catentry+3.223
nah+3.200
Neg logit contributions
arest-10.127
unia-9.985
ounding-8.768
asa-8.649
 Grammy-8.253
Token in
Token ID287
Feature activation-6.58e-01

Pos logit contributions
velt+1.626
catentry+1.471
REDACTED+1.438
••+1.417
 gamb+1.393
Neg logit contributions
arest-1.941
unia-1.869
asa-1.572
ounding-1.535
 Grammy-1.498
Token the
Token ID262
Feature activation-7.89e-01

Pos logit contributions
velt+5.190
 gamb+4.569
••+4.548
catentry+4.464
 Gamb+4.406
Neg logit contributions
unia-6.618
arest-6.566
asa-5.518
ounding-5.220
 Dame-5.128
Token next
Token ID1306
Feature activation-9.76e-02

Pos logit contributions
velt+6.471
 gamb+5.979
••+5.888
 Gamb+5.881
catentry+5.810
Neg logit contributions
arest-7.405
unia-7.389
asa-6.200
ounding-5.919
 Dame-5.625
Token an
Token ID281
Feature activation-4.52e-01

Pos logit contributions
arest+1.038
unia+0.988
asa+0.816
ounding+0.790
 Grammy+0.778
Neg logit contributions
velt-1.108
catentry-1.028
REDACTED-1.006
••-0.970
atical-0.958
Token historical
Token ID6754
Feature activation-2.60e-01

Pos logit contributions
velt+4.012
catentry+3.667
atical+3.601
REDACTED+3.570
••+3.529
Neg logit contributions
arest-4.269
unia-4.024
asa-3.335
ounding-3.296
 Grammy-3.266
Token condition
Token ID4006
Feature activation-5.29e-02

Pos logit contributions
velt+2.450
catentry+2.229
REDACTED+2.183
••+2.167
atical+2.143
Neg logit contributions
arest-2.383
unia-2.288
asa-1.890
ounding-1.842
 Grammy-1.771
Token of
Token ID286
Feature activation-2.21e-01

Pos logit contributions
velt+0.485
catentry+0.445
REDACTED+0.435
••+0.426
atical+0.419
Neg logit contributions
arest-0.505
unia-0.484
asa-0.403
ounding-0.393
 Grammy-0.384
Token the
Token ID262
Feature activation-5.55e-01

Pos logit contributions
velt+2.011
catentry+1.826
REDACTED+1.780
••+1.769
atical+1.751
Neg logit contributions
arest-2.093
unia-2.008
asa-1.674
ounding-1.637
 Grammy-1.590
Token human
Token ID1692
Feature activation-1.36e+00

Pos logit contributions
velt+4.797
 gamb+4.236
catentry+4.218
••+4.211
 Typh+4.190
Neg logit contributions
arest-5.247
unia-5.111
asa-4.261
ounding-4.194
 Grammy-4.053
Token intellect
Token ID7654
Feature activation-7.06e-02

Pos logit contributions
velt+9.186
 gamb+8.337
••+8.111
 scapego+8.072
nah+7.827
Neg logit contributions
arest-12.847
unia-12.534
ounding-10.410
asa-10.397
 Grammy-10.043
Token to
Token ID284
Feature activation-2.88e-01

Pos logit contributions
velt+0.639
catentry+0.585
REDACTED+0.572
••+0.561
atical+0.553
Neg logit contributions
arest-0.681
unia-0.652
asa-0.546
ounding-0.532
 Grammy-0.521
Token which
Token ID543
Feature activation-8.50e-02

Pos logit contributions
velt+2.577
catentry+2.339
REDACTED+2.283
••+2.264
atical+2.247
Neg logit contributions
arest-2.770
unia-2.644
asa-2.231
ounding-2.187
 Grammy-2.115
Token the
Token ID262
Feature activation-4.14e-01

Pos logit contributions
velt+0.789
catentry+0.725
REDACTED+0.707
••+0.693
atical+0.683
Neg logit contributions
arest-0.801
unia-0.767
asa-0.638
ounding-0.622
 Grammy-0.607
Token element
Token ID5002
Feature activation+2.03e-01

Pos logit contributions
velt+3.631
catentry+3.248
 gamb+3.204
••+3.189
REDACTED+3.168
Neg logit contributions
arest-3.966
unia-3.856
asa-3.205
ounding-3.144
 Grammy-3.027
Token the
Token ID262
Feature activation+1.70e-02

Pos logit contributions
velt+2.149
catentry+1.977
REDACTED+1.915
••+1.890
atical+1.859
Neg logit contributions
arest-2.461
unia-2.371
asa-1.991
ounding-1.957
 Grammy-1.897
Token C
Token ID327
Feature activation+9.88e-02

Pos logit contributions
arest+0.174
unia+0.167
asa+0.141
ounding+0.138
 Grammy+0.135
Neg logit contributions
velt-0.144
catentry-0.131
REDACTED-0.128
••-0.125
atical-0.123
TokenUP
Token ID8577
Feature activation+2.24e-01

Pos logit contributions
arest+1.205
unia+1.167
asa+1.022
ounding+0.997
ine+0.979
Neg logit contributions
velt-0.646
catentry-0.572
REDACTED-0.548
••-0.530
 gamb-0.523
Token)
Token ID8
Feature activation-3.72e-01

Pos logit contributions
arest+1.937
unia+1.839
asa+1.525
ounding+1.469
ine+1.437
Neg logit contributions
velt-2.243
catentry-2.076
REDACTED-2.032
••-1.964
joy-1.945
Token remains
Token ID3793
Feature activation+1.85e-01

Pos logit contributions
velt+3.303
catentry+3.022
REDACTED+2.934
 gamb+2.920
••+2.912
Neg logit contributions
arest-3.545
unia-3.428
asa-2.846
ounding-2.794
 Grammy-2.752
Token to
Token ID284
Feature activation-1.35e+00

Pos logit contributions
arest+1.672
unia+1.588
asa+1.311
ounding+1.278
 Grammy+1.255
Neg logit contributions
velt-1.803
catentry-1.659
REDACTED-1.603
••-1.566
joy-1.544
Token be
Token ID307
Feature activation-6.46e-01

Pos logit contributions
velt+7.321
REDACTED+6.691
atical+6.669
 gamb+6.381
catentry+6.364
Neg logit contributions
arest-13.773
unia-13.529
asa-12.639
ailability-11.923
ounding-11.719
Token seen
Token ID1775
Feature activation-4.30e-01

Pos logit contributions
velt+3.978
 gamb+3.687
atical+3.522
REDACTED+3.447
catentry+3.433
Neg logit contributions
arest-7.548
unia-7.416
ounding-6.407
asa-6.361
 Grammy-6.112
Token.
Token ID13
Feature activation-4.41e-01

Pos logit contributions
velt+3.834
catentry+3.459
REDACTED+3.457
••+3.427
 gamb+3.394
Neg logit contributions
arest-4.052
unia-3.923
asa-3.211
ounding-3.157
 Grammy-3.091
Token Let
Token ID3914
Feature activation-2.28e-01

Pos logit contributions
velt+3.514
••+3.184
catentry+3.183
 Gamb+3.146
 Typh+3.126
Neg logit contributions
arest-4.545
unia-4.428
ounding-3.723
asa-3.682
uple-3.463
Token us
Token ID514
Feature activation-4.34e-01

Pos logit contributions
velt+2.272
catentry+2.102
REDACTED+2.063
••+2.032
atical+1.997
Neg logit contributions
arest-1.963
unia-1.874
asa-1.527
ounding-1.490
 Grammy-1.457
Token called
Token ID1444
Feature activation-5.99e-01

Pos logit contributions
velt+2.329
catentry+2.127
REDACTED+2.077
••+2.039
 gamb+2.037
Neg logit contributions
arest-2.640
unia-2.539
asa-2.116
ounding-2.087
 Grammy-2.038
Token a
Token ID257
Feature activation-6.82e-01

Pos logit contributions
velt+5.175
 Typh+4.766
 Gamb+4.714
 gamb+4.639
catentry+4.636
Neg logit contributions
arest-5.589
unia-5.280
ounding-4.480
asa-4.383
 Grammy-4.158
Token Bal
Token ID8528
Feature activation-7.19e-01

Pos logit contributions
velt+5.980
 Typh+5.454
 gamb+5.447
catentry+5.421
 Gamb+5.415
Neg logit contributions
arest-6.063
unia-5.842
ounding-4.856
asa-4.776
 Grammy-4.641
Tokenrog
Token ID3828
Feature activation-5.05e-01

Pos logit contributions
velt+7.054
joy+6.336
nah+6.331
 gamb+6.217
catentry+6.200
Neg logit contributions
arest-5.654
unia-5.399
 Grammy-4.369
 replay-4.336
asa-4.301
Token.
Token ID13
Feature activation-5.51e-01

Pos logit contributions
velt+4.496
catentry+4.030
REDACTED+4.018
 gamb+3.987
atical+3.968
Neg logit contributions
arest-4.705
unia-4.540
asa-3.709
ounding-3.684
 Grammy-3.633
Token Gand
Token ID17727
Feature activation-1.33e+00

Pos logit contributions
velt+4.299
 Gamb+3.903
 Typh+3.862
••+3.857
REDACTED+3.762
Neg logit contributions
arest-5.714
unia-5.461
ounding-4.628
asa-4.596
uple-4.432
Tokenalf
Token ID1604
Feature activation-8.67e-01

Pos logit contributions
velt+5.749
 gamb+4.194
joy+4.146
atical+4.034
nah+3.901
Neg logit contributions
arest-15.290
unia-15.074
ailability-14.537
 Grammy-14.493
Û-13.939
Token faces
Token ID6698
Feature activation-6.26e-01

Pos logit contributions
velt+6.662
 gamb+6.111
REDACTED+5.694
 Typh+5.659
 Gamb+5.589
Neg logit contributions
arest-8.622
unia-8.366
ounding-7.176
Depth-7.109
asa-6.999
Token the
Token ID262
Feature activation-4.23e-01

Pos logit contributions
velt+4.970
REDACTED+4.469
 gamb+4.407
 Gamb+4.355
joy+4.319
Neg logit contributions
arest-6.452
unia-6.107
asa-5.150
ounding-5.106
 Grammy-4.918
Token Bal
Token ID8528
Feature activation-7.75e-01

Pos logit contributions
velt+3.511
catentry+3.192
 gamb+3.171
REDACTED+3.143
 Gamb+3.122
Neg logit contributions
arest-4.209
unia-4.108
asa-3.405
ounding-3.360
 Grammy-3.267
Tokenrog
Token ID3828
Feature activation-4.56e-01

Pos logit contributions
velt+3.061
joy+2.257
nah+2.235
catentry+2.143
atical+2.051
Neg logit contributions
arest-10.497
unia-10.176
 Grammy-9.220
 Dame-9.150
asa-9.121
Token
Token ID198
Feature activation+1.29e-01

Pos logit contributions
velt+1.141
catentry+1.051
REDACTED+1.028
••+1.009
atical+0.992
Neg logit contributions
arest-1.094
unia-1.047
asa-0.862
ounding-0.842
 Grammy-0.815
Token
Token ID198
Feature activation-1.04e-02

Pos logit contributions
arest+1.067
unia+1.014
asa+0.823
ounding+0.790
ine+0.775
Neg logit contributions
velt-1.347
catentry-1.257
REDACTED-1.218
••-1.199
atical-1.187
TokenWithout
Token ID16249
Feature activation-2.29e-01

Pos logit contributions
velt+0.096
catentry+0.088
REDACTED+0.086
••+0.084
atical+0.083
Neg logit contributions
arest-0.099
unia-0.094
asa-0.079
ounding-0.077
 Grammy-0.075
Token a
Token ID257
Feature activation-6.59e-01

Pos logit contributions
velt+2.080
catentry+1.901
REDACTED+1.864
••+1.819
joy+1.799
Neg logit contributions
arest-2.198
unia-2.114
asa-1.759
ounding-1.710
 Grammy-1.656
Token working
Token ID1762
Feature activation-6.51e-01

Pos logit contributions
velt+5.728
catentry+5.096
REDACTED+5.094
joy+5.039
 gamb+5.024
Neg logit contributions
arest-6.186
unia-6.123
asa-4.915
ounding-4.864
 Dame-4.661
Token star
Token ID3491
Feature activation-1.33e+00

Pos logit contributions
velt+5.423
catentry+4.815
 gamb+4.793
REDACTED+4.786
joy+4.784
Neg logit contributions
arest-6.331
unia-6.233
asa-5.051
ounding-4.977
uple-4.768
Token tracker
Token ID30013
Feature activation-4.72e-01

Pos logit contributions
velt+6.038
catentry+5.288
atical+4.921
REDACTED+4.785
 gamb+4.715
Neg logit contributions
arest-15.799
unia-15.661
 Dame-13.639
ounding-13.232
asa-13.225
Token,
Token ID11
Feature activation-4.65e-02

Pos logit contributions
velt+4.179
catentry+3.838
REDACTED+3.836
••+3.714
joy+3.672
Neg logit contributions
arest-4.545
unia-4.334
asa-3.555
ounding-3.498
 Dame-3.366
Token ion
Token ID22088
Feature activation+4.71e-01

Pos logit contributions
velt+0.431
catentry+0.396
REDACTED+0.387
••+0.379
atical+0.373
Neg logit contributions
arest-0.438
unia-0.420
asa-0.349
ounding-0.340
 Grammy-0.332
Token thrust
Token ID14613
Feature activation-2.62e-01

Pos logit contributions
arest+4.259
unia+4.055
asa+3.439
ine+3.321
 replay+3.312
Neg logit contributions
velt-4.400
catentry-4.135
REDACTED-4.030
••-3.896
nah-3.797
Tokening
Token ID278
Feature activation-1.85e-01

Pos logit contributions
velt+2.311
catentry+2.087
REDACTED+2.047
••+2.033
atical+2.018
Neg logit contributions
arest-2.557
unia-2.411
asa-2.042
ounding-1.978
 Grammy-1.951
Token Helm
Token ID28351
Feature activation-3.86e-01

Pos logit contributions
velt+2.822
catentry+2.551
REDACTED+2.523
••+2.503
joy+2.485
Neg logit contributions
arest-2.752
unia-2.621
asa-2.169
ounding-2.144
 Grammy-2.050
Token's
Token ID338
Feature activation-5.44e-01

Pos logit contributions
velt+3.437
joy+3.032
catentry+2.983
••+2.948
atical+2.944
Neg logit contributions
arest-3.579
unia-3.464
asa-2.899
ounding-2.874
 replay-2.824
Token Deep
Token ID10766
Feature activation-5.08e-02

Pos logit contributions
velt+5.558
joy+5.296
catentry+5.186
 Gamb+5.092
 gamb+5.055
Neg logit contributions
arest-4.403
unia-4.302
uple-3.202
ounding-3.154
iping-3.138
Token,
Token ID11
Feature activation-1.59e-01

Pos logit contributions
velt+0.493
catentry+0.456
REDACTED+0.445
••+0.436
atical+0.430
Neg logit contributions
arest-0.456
unia-0.436
asa-0.359
ounding-0.349
 Grammy-0.341
Token while
Token ID981
Feature activation-2.51e-01

Pos logit contributions
velt+1.378
catentry+1.252
REDACTED+1.225
••+1.203
 gamb+1.192
Neg logit contributions
arest-1.579
unia-1.520
asa-1.276
ounding-1.251
 Grammy-1.219
Token Gand
Token ID17727
Feature activation-1.33e+00

Pos logit contributions
velt+1.954
catentry+1.735
REDACTED+1.711
••+1.671
 gamb+1.669
Neg logit contributions
arest-2.713
unia-2.608
asa-2.222
ounding-2.197
 Grammy-2.131
Tokenalf
Token ID1604
Feature activation-5.08e-01

Pos logit contributions
velt+5.685
joy+4.291
 gamb+4.005
nah+3.900
atical+3.892
Neg logit contributions
arest-15.216
unia-15.210
 Grammy-14.371
ailability-14.315
Û-13.927
Token dep
Token ID1207
Feature activation-1.05e+00

Pos logit contributions
velt+4.197
 gamb+3.737
REDACTED+3.623
••+3.606
catentry+3.584
Neg logit contributions
arest-5.163
unia-4.997
ounding-4.172
asa-4.153
Depth-4.029
Tokenarts
Token ID5889
Feature activation-3.38e-01

Pos logit contributions
velt+4.889
rikes+4.231
joy+4.206
atical+4.104
rez+4.102
Neg logit contributions
unia-12.201
arest-11.769
 Grammy-10.874
 replay-10.655
 Dame-10.605
Token to
Token ID284
Feature activation-1.27e-02

Pos logit contributions
velt+2.978
REDACTED+2.668
catentry+2.643
joy+2.627
••+2.619
Neg logit contributions
arest-3.270
unia-3.136
asa-2.622
ounding-2.577
 Grammy-2.507
Token seek
Token ID5380
Feature activation-4.20e-01

Pos logit contributions
velt+0.118
catentry+0.109
REDACTED+0.106
••+0.104
atical+0.102
Neg logit contributions
arest-0.121
unia-0.116
asa-0.096
ounding-0.093
 Grammy-0.091
Token to
Token ID284
Feature activation-2.36e-01

Pos logit contributions
velt+2.351
catentry+2.150
REDACTED+2.101
••+2.058
 gamb+2.033
Neg logit contributions
arest-2.471
unia-2.370
asa-1.981
ounding-1.926
 Grammy-1.888
Token take
Token ID1011
Feature activation+2.15e-01

Pos logit contributions
velt+2.094
catentry+1.899
REDACTED+1.851
 gamb+1.822
••+1.813
Neg logit contributions
arest-2.293
unia-2.201
asa-1.852
ounding-1.811
 Grammy-1.765
Token the
Token ID262
Feature activation-2.71e-01

Pos logit contributions
arest+2.103
unia+2.049
asa+1.716
ounding+1.645
 Dame+1.635
Neg logit contributions
velt-1.896
catentry-1.792
REDACTED-1.726
••-1.650
atical-1.638
Token Ring
Token ID12569
Feature activation-6.39e-01

Pos logit contributions
velt+2.144
catentry+1.938
REDACTED+1.879
 gamb+1.872
••+1.860
Neg logit contributions
arest-2.861
unia-2.764
asa-2.350
ounding-2.295
 Grammy-2.234
Token from
Token ID422
Feature activation-5.96e-01

Pos logit contributions
velt+5.285
 gamb+4.623
joy+4.583
REDACTED+4.579
 Gamb+4.540
Neg logit contributions
arest-6.246
unia-6.055
ounding-5.004
asa-4.974
 Grammy-4.850
Token Fro
Token ID9734
Feature activation-1.32e+00

Pos logit contributions
velt+4.475
 gamb+3.981
 Gamb+3.970
joy+3.942
REDACTED+3.895
Neg logit contributions
arest-6.318
unia-5.947
ounding-5.167
asa-5.115
 Grammy-4.847
Tokendo
Token ID4598
Feature activation-6.07e-01

Pos logit contributions
velt+5.043
joy+3.907
nah+3.719
cery+3.418
worn+3.215
Neg logit contributions
arest-16.590
unia-15.838
ailability-15.760
 Grammy-15.327
ounding-14.986
Token,
Token ID11
Feature activation-5.75e-01

Pos logit contributions
velt+5.209
 gamb+4.544
••+4.515
joy+4.473
REDACTED+4.451
Neg logit contributions
arest-5.916
unia-5.678
ounding-4.766
asa-4.707
 Grammy-4.577
Token but
Token ID475
Feature activation-5.68e-01

Pos logit contributions
velt+4.294
 gamb+3.810
 Gamb+3.726
joy+3.658
catentry+3.646
Neg logit contributions
arest-6.131
unia-5.894
ounding-5.104
asa-5.087
uple-4.833
Token Fro
Token ID9734
Feature activation-1.11e+00

Pos logit contributions
velt+4.567
 gamb+4.124
 Gamb+4.041
 Typh+3.973
joy+3.963
Neg logit contributions
arest-5.813
unia-5.445
ounding-4.733
asa-4.667
uple-4.439
Tokendo
Token ID4598
Feature activation-6.12e-01

Pos logit contributions
velt+5.582
joy+4.315
nah+4.029
atical+3.752
cery+3.736
Neg logit contributions
arest-13.552
unia-12.882
ailability-12.714
 Grammy-12.414
ounding-12.158
Token are
Token ID389
Feature activation-1.58e-01

Pos logit contributions
velt+4.726
REDACTED+4.193
catentry+4.127
 gamb+4.124
••+4.080
Neg logit contributions
arest-5.341
unia-5.135
asa-4.285
ounding-4.265
 Grammy-4.116
Token nine
Token ID5193
Feature activation-9.04e-02

Pos logit contributions
velt+1.384
catentry+1.258
REDACTED+1.240
••+1.207
 gamb+1.199
Neg logit contributions
arest-1.570
unia-1.503
asa-1.264
ounding-1.236
 Grammy-1.206
Token in
Token ID287
Feature activation+2.96e-01

Pos logit contributions
velt+0.829
catentry+0.762
REDACTED+0.744
••+0.729
atical+0.719
Neg logit contributions
arest-0.859
unia-0.823
asa-0.686
ounding-0.669
 Grammy-0.653
Token number
Token ID1271
Feature activation-9.02e-03

Pos logit contributions
arest+2.697
unia+2.574
asa+2.163
ine+2.102
ounding+2.099
Neg logit contributions
velt-2.778
catentry-2.573
REDACTED-2.516
••-2.436
atical-2.367
Token:
Token ID25
Feature activation-3.96e-01

Pos logit contributions
velt+0.088
catentry+0.082
REDACTED+0.080
••+0.078
atical+0.077
Neg logit contributions
arest-0.080
unia-0.077
asa-0.063
ounding-0.061
 Grammy-0.060
Token Fro
Token ID9734
Feature activation-1.32e+00

Pos logit contributions
velt+3.438
catentry+3.086
REDACTED+3.084
••+3.048
joy+3.038
Neg logit contributions
arest-3.858
unia-3.647
asa-3.061
ounding-3.059
 Grammy-2.902
Tokendo
Token ID4598
Feature activation-5.96e-01

Pos logit contributions
velt+5.628
joy+4.421
nah+4.338
 gamb+4.033
cery+4.016
Neg logit contributions
arest-15.757
unia-14.967
 Grammy-14.069
ailability-14.068
ounding-13.899
Token,
Token ID11
Feature activation-8.54e-01

Pos logit contributions
velt+5.092
 gamb+4.520
••+4.491
 Gamb+4.487
REDACTED+4.469
Neg logit contributions
arest-5.766
unia-5.484
ounding-4.650
asa-4.556
 replay-4.452
Token Sam
Token ID3409
Feature activation-6.83e-01

Pos logit contributions
velt+7.026
 Gamb+6.651
 gamb+6.508
 Typh+6.463
joy+6.451
Neg logit contributions
arest-7.834
unia-7.346
ounding-6.477
asa-6.121
Depth-5.984
Token,
Token ID11
Feature activation-6.83e-01

Pos logit contributions
velt+5.250
 Gamb+4.857
••+4.854
nah+4.811
 gamb+4.735
Neg logit contributions
arest-6.762
unia-6.225
ounding-5.646
 replay-5.232
Depth-5.228
Token Merry
Token ID39001
Feature activation-8.26e-01

Pos logit contributions
velt+5.938
 Gamb+5.479
 Typh+5.347
joy+5.330
 gamb+5.303
Neg logit contributions
arest-6.321
unia-5.891
ounding-5.142
asa-4.906
Depth-4.693
Token torn
Token ID12445
Feature activation-2.75e-01

Pos logit contributions
velt+0.149
catentry+0.138
REDACTED+0.134
••+0.131
atical+0.129
Neg logit contributions
arest-0.149
unia-0.143
asa-0.119
ounding-0.115
 Grammy-0.113
Token between
Token ID1022
Feature activation-2.36e-01

Pos logit contributions
velt+2.025
catentry+1.767
REDACTED+1.743
••+1.721
atical+1.718
Neg logit contributions
arest-3.077
unia-2.973
asa-2.532
ounding-2.505
 Grammy-2.458
Token his
Token ID465
Feature activation-8.94e-02

Pos logit contributions
velt+2.058
catentry+1.865
REDACTED+1.829
••+1.797
 gamb+1.781
Neg logit contributions
arest-2.340
unia-2.250
asa-1.886
ounding-1.852
 Grammy-1.804
Token loyalty
Token ID15843
Feature activation+5.56e-02

Pos logit contributions
velt+0.821
catentry+0.754
REDACTED+0.736
••+0.721
atical+0.711
Neg logit contributions
arest-0.850
unia-0.815
asa-0.680
ounding-0.662
 Grammy-0.646
Token to
Token ID284
Feature activation-3.82e-01

Pos logit contributions
arest+0.499
unia+0.479
asa+0.394
ounding+0.382
ine+0.374
Neg logit contributions
velt-0.539
catentry-0.505
REDACTED-0.485
••-0.478
joy-0.466
Token Fro
Token ID9734
Feature activation-1.31e+00

Pos logit contributions
velt+3.071
catentry+2.738
REDACTED+2.680
 gamb+2.656
••+2.645
Neg logit contributions
arest-3.979
unia-3.820
asa-3.263
ounding-3.233
 Grammy-3.111
Tokendo
Token ID4598
Feature activation-4.01e-01

Pos logit contributions
velt+5.113
joy+4.067
nah+3.793
cery+3.478
 gamb+3.270
Neg logit contributions
arest-16.511
unia-15.667
ailability-14.896
ounding-14.875
 Grammy-14.817
Token and
Token ID290
Feature activation-2.49e-01

Pos logit contributions
velt+3.537
catentry+3.153
REDACTED+3.111
••+3.088
 gamb+3.062
Neg logit contributions
arest-3.918
unia-3.767
asa-3.148
ounding-3.088
 Grammy-3.005
Token his
Token ID465
Feature activation+2.03e-01

Pos logit contributions
velt+2.040
catentry+1.818
REDACTED+1.795
••+1.757
 gamb+1.755
Neg logit contributions
arest-2.600
unia-2.499
asa-2.123
ounding-2.092
 Grammy-2.036
Token desire
Token ID6227
Feature activation-1.21e-01

Pos logit contributions
arest+1.973
unia+1.887
asa+1.571
ounding+1.529
 Grammy+1.524
Neg logit contributions
velt-1.820
catentry-1.687
REDACTED-1.636
••-1.598
joy-1.544
Token for
Token ID329
Feature activation-2.38e-01

Pos logit contributions
velt+1.129
catentry+1.042
REDACTED+1.011
••+0.991
atical+0.975
Neg logit contributions
arest-1.140
unia-1.093
asa-0.908
ounding-0.884
 Grammy-0.862
Token County
Token ID3418
Feature activation-1.34e-01

Pos logit contributions
arest+2.413
unia+2.296
asa+1.862
ine+1.758
 Grammy+1.758
Neg logit contributions
velt-3.447
catentry-3.293
REDACTED-3.176
••-3.094
atical-3.081
Token,
Token ID11
Feature activation-1.32e-01

Pos logit contributions
velt+1.296
catentry+1.191
REDACTED+1.167
••+1.148
 gamb+1.128
Neg logit contributions
arest-1.214
unia-1.162
asa-0.954
ounding-0.930
 Grammy-0.904
Token South
Token ID2520
Feature activation+3.86e-02

Pos logit contributions
velt+1.289
catentry+1.184
REDACTED+1.162
••+1.146
 gamb+1.129
Neg logit contributions
arest-1.178
unia-1.127
asa-0.920
ounding-0.901
 Grammy-0.869
Token Carolina
Token ID5913
Feature activation-1.10e-01

Pos logit contributions
arest+0.354
unia+0.339
asa+0.280
ounding+0.273
 Grammy+0.269
Neg logit contributions
velt-0.369
catentry-0.341
REDACTED-0.333
••-0.324
atical-0.320
Token,
Token ID11
Feature activation-3.88e-02

Pos logit contributions
velt+1.061
catentry+0.979
REDACTED+0.956
••+0.938
atical+0.924
Neg logit contributions
arest-0.998
unia-0.955
asa-0.787
ounding-0.766
 Grammy-0.747
Token commemor
Token ID28723
Feature activation-1.29e+00

Pos logit contributions
velt+0.352
catentry+0.324
REDACTED+0.315
••+0.308
atical+0.304
Neg logit contributions
arest-0.373
unia-0.358
asa-0.300
ounding-0.292
 Grammy-0.286
Tokenates
Token ID689
Feature activation-1.25e-02

Pos logit contributions
atical+5.215
velt+5.178
joy+4.660
icates+4.496
herent+4.275
Neg logit contributions
unia-15.660
arest-14.734
 Grammy-13.249
Û-13.078
 replay-12.796
Token her
Token ID607
Feature activation+1.61e-01

Pos logit contributions
velt+0.116
catentry+0.107
REDACTED+0.104
••+0.102
atical+0.101
Neg logit contributions
arest-0.118
unia-0.113
asa-0.094
ounding-0.092
 Grammy-0.090
Token birthplace
Token ID48145
Feature activation+3.47e-01

Pos logit contributions
arest+1.444
unia+1.382
asa+1.147
ounding+1.110
 Grammy+1.108
Neg logit contributions
velt-1.564
catentry-1.468
REDACTED-1.422
••-1.365
atical-1.353
Token.[
Token ID3693
Feature activation-1.02e-01

Pos logit contributions
arest+2.889
unia+2.779
asa+2.291
 Grammy+2.218
ounding+2.209
Neg logit contributions
velt-3.512
catentry-3.369
REDACTED-3.175
••-3.056
 gamb-3.041
Token61
Token ID5333
Feature activation-5.51e-01

Pos logit contributions
velt+0.716
catentry+0.639
REDACTED+0.617
••+0.600
atical+0.588
Neg logit contributions
arest-1.199
unia-1.158
asa-1.003
ounding-0.984
 Grammy-0.966
Token,
Token ID11
Feature activation-8.44e-01

Pos logit contributions
velt+5.199
catentry+4.692
REDACTED+4.686
••+4.555
atical+4.505
Neg logit contributions
arest-5.872
unia-5.600
asa-4.602
ounding-4.521
 Grammy-4.456
Token is
Token ID318
Feature activation-1.21e+00

Pos logit contributions
velt+6.517
REDACTED+5.856
catentry+5.828
 gamb+5.675
atical+5.657
Neg logit contributions
arest-8.432
unia-8.113
asa-6.758
ounding-6.722
Depth-6.581
Token strictly
Token ID14084
Feature activation-5.82e-01

Pos logit contributions
velt+7.945
REDACTED+7.322
riber+6.925
••+6.812
 gamb+6.803
Neg logit contributions
arest-12.369
unia-12.175
asa-10.138
ounding-10.063
Depth-9.919
Token prohibited
Token ID12244
Feature activation-8.47e-01

Pos logit contributions
velt+3.270
REDACTED+2.944
catentry+2.856
••+2.833
atical+2.753
Neg logit contributions
arest-7.261
unia-7.045
asa-6.054
ounding-6.008
Depth-5.900
Token without
Token ID1231
Feature activation-1.00e+00

Pos logit contributions
velt+6.510
••+5.917
REDACTED+5.913
joy+5.705
atical+5.685
Neg logit contributions
arest-8.475
unia-8.245
ounding-6.808
asa-6.767
 replay-6.607
Token the
Token ID262
Feature activation-1.29e+00

Pos logit contributions
velt+7.133
atical+6.340
REDACTED+6.182
catentry+6.152
••+6.121
Neg logit contributions
arest-10.155
unia-9.982
asa-8.633
uple-8.114
ounding-8.069
Token express
Token ID4911
Feature activation-1.25e+00

Pos logit contributions
velt+7.415
atical+6.678
catentry+6.277
riber+6.234
 Gamb+6.166
Neg logit contributions
unia-13.474
arest-13.468
asa-11.617
Depth-11.196
rel-11.120
Token written
Token ID3194
Feature activation-1.05e+00

Pos logit contributions
velt+5.852
atical+5.093
catentry+5.032
REDACTED+4.852
••+4.639
Neg logit contributions
unia-14.750
arest-14.691
asa-12.681
Depth-12.475
ounding-12.288
Token permission
Token ID7170
Feature activation-5.46e-01

Pos logit contributions
velt+6.667
REDACTED+5.967
atical+5.959
catentry+5.879
••+5.565
Neg logit contributions
arest-11.427
unia-11.217
asa-9.638
ounding-9.488
Depth-9.280
Token of
Token ID286
Feature activation-4.49e-01

Pos logit contributions
velt+4.855
REDACTED+4.418
catentry+4.411
••+4.346
atical+4.301
Neg logit contributions
arest-5.085
unia-4.876
asa-4.069
ounding-3.938
 replay-3.889
Token Money
Token ID12911
Feature activation-4.95e-01

Pos logit contributions
velt+3.712
catentry+3.339
REDACTED+3.285
••+3.266
 Gamb+3.219
Neg logit contributions
arest-4.552
unia-4.436
asa-3.726
ounding-3.620
 replay-3.499
Token baking
Token ID16871
Feature activation+1.85e-01

Pos logit contributions
arest+4.010
unia+3.769
asa+3.073
ounding+3.065
 replay+2.993
Neg logit contributions
velt-4.528
catentry-4.143
REDACTED-4.109
••-3.989
atical-3.882
Token and
Token ID290
Feature activation+3.32e-01

Pos logit contributions
arest+1.643
unia+1.577
asa+1.277
ounding+1.253
ine+1.226
Neg logit contributions
velt-1.817
catentry-1.687
REDACTED-1.656
••-1.593
atical-1.563
Token is
Token ID318
Feature activation-1.48e-01

Pos logit contributions
arest+2.916
unia+2.755
asa+2.280
ounding+2.235
 replay+2.220
Neg logit contributions
velt-3.216
REDACTED-2.986
catentry-2.942
••-2.844
atical-2.808
Token an
Token ID281
Feature activation-8.27e-02

Pos logit contributions
velt+1.358
catentry+1.248
REDACTED+1.214
••+1.193
 gamb+1.181
Neg logit contributions
arest-1.410
unia-1.358
asa-1.126
ounding-1.100
 Grammy-1.068
Token excellent
Token ID6275
Feature activation+1.32e-01

Pos logit contributions
velt+0.764
catentry+0.705
REDACTED+0.683
••+0.668
atical+0.664
Neg logit contributions
arest-0.784
unia-0.750
asa-0.619
ounding-0.605
 Grammy-0.591
Token b
Token ID275
Feature activation+1.04e+00

Pos logit contributions
arest+1.208
unia+1.158
asa+0.966
ounding+0.942
 replay+0.931
Neg logit contributions
velt-1.249
catentry-1.153
REDACTED-1.134
••-1.094
atical-1.079
Tokeninder
Token ID5540
Feature activation+2.98e-01

Pos logit contributions
arest+8.272
unia+7.157
ounding+6.513
ine+6.348
asa+6.296
Neg logit contributions
velt-10.293
REDACTED-9.756
catentry-9.664
 Gamb-8.942
••-8.831
Token too
Token ID1165
Feature activation+1.01e-01

Pos logit contributions
arest+2.549
unia+2.426
asa+2.004
ounding+1.962
ine+1.919
Neg logit contributions
velt-2.972
catentry-2.773
REDACTED-2.692
••-2.606
atical-2.556
Token.
Token ID13
Feature activation+3.16e-02

Pos logit contributions
arest+0.849
unia+0.807
asa+0.659
ounding+0.638
 Grammy+0.620
Neg logit contributions
velt-1.037
catentry-0.964
REDACTED-0.945
••-0.916
atical-0.908
Token Ch
Token ID609
Feature activation-2.33e-01

Pos logit contributions
arest+0.281
unia+0.268
asa+0.221
ounding+0.214
 Grammy+0.210
Neg logit contributions
velt-0.311
catentry-0.288
REDACTED-0.281
••-0.274
atical-0.271
Tokenia
Token ID544
Feature activation+2.02e-01

Pos logit contributions
velt+2.051
catentry+1.845
REDACTED+1.803
••+1.786
atical+1.765
Neg logit contributions
arest-2.271
unia-2.193
asa-1.831
ounding-1.783
 Grammy-1.766
Token the
Token ID262
Feature activation+3.82e-01

Pos logit contributions
velt+2.089
catentry+1.893
REDACTED+1.854
••+1.817
 gamb+1.801
Neg logit contributions
arest-2.321
unia-2.228
asa-1.872
ounding-1.829
 Grammy-1.768
Token game
Token ID983
Feature activation+1.62e-01

Pos logit contributions
arest+3.143
unia+2.948
 replay+2.387
asa+2.382
 Grammy+2.369
Neg logit contributions
velt-3.906
catentry-3.700
REDACTED-3.593
••-3.489
atical-3.430
Token's
Token ID338
Feature activation+1.79e-01

Pos logit contributions
arest+1.440
unia+1.382
asa+1.138
ounding+1.095
 Grammy+1.090
Neg logit contributions
velt-1.566
catentry-1.470
REDACTED-1.428
••-1.389
atical-1.364
Token popularity
Token ID11533
Feature activation+1.87e-01

Pos logit contributions
arest+1.642
unia+1.569
asa+1.295
 replay+1.272
ounding+1.262
Neg logit contributions
velt-1.692
catentry-1.576
REDACTED-1.541
••-1.479
atical-1.457
Token has
Token ID468
Feature activation+7.20e-01

Pos logit contributions
arest+1.620
unia+1.549
asa+1.261
ounding+1.233
 Grammy+1.219
Neg logit contributions
velt-1.863
catentry-1.756
REDACTED-1.710
••-1.655
joy-1.616
Token w
Token ID266
Feature activation+1.05e+00

Pos logit contributions
arest+6.375
unia+6.320
asa+5.137
 Grammy+5.115
ounding+5.107
Neg logit contributions
velt-6.620
catentry-6.376
REDACTED-5.971
••-5.617
joy-5.525
Tokenaned
Token ID22739
Feature activation+3.84e-03

Pos logit contributions
arest+6.682
unia+5.785
ounding+4.685
ine+4.637
asa+4.450
Neg logit contributions
velt-12.325
REDACTED-12.112
catentry-12.031
 Gamb-11.037
joy-10.938
Token since
Token ID1201
Feature activation+1.04e-01

Pos logit contributions
arest+0.034
unia+0.033
asa+0.027
ounding+0.026
 Grammy+0.026
Neg logit contributions
velt-0.037
catentry-0.035
REDACTED-0.034
••-0.033
atical-0.033
Token last
Token ID938
Feature activation-2.22e-01

Pos logit contributions
arest+0.897
unia+0.860
asa+0.702
ounding+0.682
 Grammy+0.679
Neg logit contributions
velt-1.026
catentry-0.960
REDACTED-0.938
••-0.912
atical-0.903
Token summer
Token ID3931
Feature activation+8.14e-02

Pos logit contributions
velt+2.055
catentry+1.894
REDACTED+1.839
••+1.826
 gamb+1.808
Neg logit contributions
arest-2.082
unia-1.996
asa-1.652
ounding-1.616
 Grammy-1.551
Token,
Token ID11
Feature activation-4.27e-01

Pos logit contributions
arest+0.699
unia+0.663
asa+0.547
ounding+0.530
 Grammy+0.524
Neg logit contributions
velt-0.820
catentry-0.768
REDACTED-0.747
••-0.721
atical-0.717
Token in
Token ID287
Feature activation-5.51e-01

Pos logit contributions
velt+3.360
REDACTED+3.033
catentry+3.024
••+2.903
 gamb+2.900
Neg logit contributions
arest-3.719
unia-3.632
asa-3.020
ounding-2.904
 Grammy-2.881
Token advance
Token ID5963
Feature activation-4.99e-01

Pos logit contributions
velt+2.570
REDACTED+2.253
catentry+2.081
joy+2.063
atical+1.994
Neg logit contributions
unia-7.268
arest-7.251
asa-6.354
 Grammy-6.210
 Dame-6.150
Token for
Token ID329
Feature activation+1.23e-01

Pos logit contributions
velt+4.259
REDACTED+3.821
catentry+3.780
••+3.670
 gamb+3.650
Neg logit contributions
arest-4.878
unia-4.780
asa-4.038
 Grammy-3.838
ounding-3.804
Token any
Token ID597
Feature activation+1.56e-01

Pos logit contributions
arest+1.254
unia+1.196
asa+1.019
ounding+1.005
 replay+0.979
Neg logit contributions
velt-1.053
catentry-0.973
REDACTED-0.929
••-0.922
atical-0.900
Token assistance
Token ID6829
Feature activation+1.73e-01

Pos logit contributions
arest+1.338
unia+1.273
asa+1.047
ounding+1.024
 replay+0.998
Neg logit contributions
velt-1.572
catentry-1.460
REDACTED-1.423
••-1.403
atical-1.378
Token you
Token ID345
Feature activation+1.05e+00

Pos logit contributions
arest+1.573
unia+1.498
asa+1.246
ounding+1.221
ine+1.192
Neg logit contributions
velt-1.647
catentry-1.540
REDACTED-1.484
••-1.453
atical-1.439
Token may
Token ID743
Feature activation+8.32e-01

Pos logit contributions
arest+6.775
unia+6.494
ine+5.148
 replay+5.125
rel+5.110
Neg logit contributions
catentry-11.204
velt-11.191
REDACTED-10.820
ONSORED-10.543
••-10.521
Token be
Token ID307
Feature activation+1.04e+00

Pos logit contributions
arest+6.091
unia+5.764
ine+4.798
 replay+4.530
ounding+4.492
Neg logit contributions
velt-8.754
catentry-8.461
••-8.206
REDACTED-8.082
ONSORED-7.951
Token able
Token ID1498
Feature activation+4.75e-01

Pos logit contributions
arest+7.206
unia+6.614
ine+5.435
asa+5.419
ounding+5.309
Neg logit contributions
velt-11.138
catentry-10.743
REDACTED-10.459
••-10.314
-9.941
Token to
Token ID284
Feature activation+7.38e-01

Pos logit contributions
arest+4.024
unia+3.823
asa+3.130
ounding+3.062
ine+3.057
Neg logit contributions
velt-4.651
catentry-4.449
••-4.260
REDACTED-4.251
atical-4.115
Token give
Token ID1577
Feature activation+3.25e-01

Pos logit contributions
arest+5.553
unia+5.326
ine+4.273
asa+4.248
 replay+4.162
Neg logit contributions
velt-7.728
catentry-7.367
REDACTED-7.176
••-7.034
joy-6.896
Token 2002
Token ID6244
Feature activation+2.78e-01

Pos logit contributions
arest+5.534
unia+4.777
 Grammy+4.152
asa+4.079
ine+3.854
Neg logit contributions
velt-9.213
catentry-8.993
atical-8.544
••-8.466
REDACTED-8.434
Token Ray
Token ID7760
Feature activation+8.03e-02

Pos logit contributions
arest+2.498
unia+2.380
 Grammy+2.014
asa+1.993
 Dame+1.943
Neg logit contributions
velt-2.647
catentry-2.505
REDACTED-2.365
atical-2.312
••-2.300
Token Rom
Token ID3570
Feature activation+1.25e-01

Pos logit contributions
arest+0.890
unia+0.860
asa+0.737
ounding+0.720
 Grammy+0.711
Neg logit contributions
velt-0.609
catentry-0.553
REDACTED-0.534
••-0.521
atical-0.511
Tokenano
Token ID5733
Feature activation+1.19e+00

Pos logit contributions
arest+1.554
unia+1.499
asa+1.314
ounding+1.282
ine+1.259
Neg logit contributions
velt-0.792
catentry-0.718
REDACTED-0.683
••-0.660
 gamb-0.647
Token
Token ID198
Feature activation+2.24e-01

Pos logit contributions
arest+6.395
unia+5.563
ine+4.605
asa+4.540
 Dame+4.461
Neg logit contributions
velt-13.406
catentry-12.967
••-12.619
atical-12.581
 destro-12.483
Token
Token ID198
Feature activation+1.46e+00

Pos logit contributions
arest+1.766
unia+1.670
asa+1.345
ine+1.285
ounding+1.272
Neg logit contributions
velt-2.406
catentry-2.268
REDACTED-2.174
••-2.146
atical-2.135
Token2002
Token ID16942
Feature activation-3.15e-02

Pos logit contributions
arest+8.582
unia+7.304
rel+6.567
asa+6.374
Depth+6.141
Neg logit contributions
velt-14.541
 gamb-14.193
 scapego-13.903
 Gamb-13.762
 Typh-13.739
Token Kel
Token ID15150
Feature activation+2.19e-01

Pos logit contributions
velt+0.295
catentry+0.272
REDACTED+0.265
••+0.260
atical+0.256
Neg logit contributions
arest-0.294
unia-0.281
asa-0.234
ounding-0.227
 Grammy-0.223
Tokensey
Token ID4397
Feature activation+1.26e-01

Pos logit contributions
arest+1.978
unia+1.889
asa+1.569
ounding+1.494
ine+1.490
Neg logit contributions
velt-2.106
catentry-1.971
REDACTED-1.923
••-1.882
 gamb-1.849
Token Gram
Token ID20159
Feature activation+5.66e-02

Pos logit contributions
arest+0.931
unia+0.889
asa+0.692
 Grammy+0.668
ounding+0.664
Neg logit contributions
velt-1.421
catentry-1.335
REDACTED-1.305
••-1.276
atical-1.273
Tokenmer
Token ID647
Feature activation+8.63e-01

Pos logit contributions
arest+0.472
unia+0.449
asa+0.364
ounding+0.352
 Grammy+0.342
Neg logit contributions
velt-0.586
catentry-0.545
REDACTED-0.533
••-0.523
atical-0.516
Token the
Token ID262
Feature activation+6.54e-02

Pos logit contributions
arest+0.014
unia+0.014
asa+0.011
ounding+0.011
 Grammy+0.011
Neg logit contributions
velt-0.013
catentry-0.012
REDACTED-0.012
••-0.011
atical-0.011
Token company
Token ID1664
Feature activation+2.97e-01

Pos logit contributions
arest+0.603
unia+0.577
asa+0.478
ounding+0.465
 Grammy+0.459
Neg logit contributions
velt-0.618
catentry-0.573
REDACTED-0.558
••-0.544
atical-0.536
Token
Token ID447
Feature activation-1.56e-02

Pos logit contributions
arest+2.628
unia+2.537
asa+2.115
 Grammy+2.015
ine+2.013
Neg logit contributions
velt-2.836
catentry-2.718
REDACTED-2.579
••-2.494
joy-2.471
Token
Token ID247
Feature activation+1.23e-01

Pos logit contributions
velt+0.163
catentry+0.151
REDACTED+0.148
••+0.145
atical+0.143
Neg logit contributions
arest-0.130
unia-0.124
asa-0.100
ounding-0.097
 Grammy-0.094
Tokens
Token ID82
Feature activation+7.33e-03

Pos logit contributions
arest+1.009
unia+0.959
asa+0.779
ounding+0.745
ine+0.731
Neg logit contributions
velt-1.297
catentry-1.209
REDACTED-1.175
••-1.146
atical-1.139
Token value
Token ID1988
Feature activation+9.83e-01

Pos logit contributions
arest+0.070
unia+0.067
asa+0.056
ounding+0.054
 Grammy+0.053
Neg logit contributions
velt-0.067
catentry-0.062
REDACTED-0.061
••-0.059
atical-0.058
Token derived
Token ID10944
Feature activation+1.88e-01

Pos logit contributions
arest+8.179
unia+8.084
asa+6.717
ine+6.481
 replay+6.422
Neg logit contributions
velt-8.547
catentry-8.524
REDACTED-7.988
joy-7.439
 Typh-7.419
Token mostly
Token ID4632
Feature activation+4.42e-01

Pos logit contributions
arest+1.542
unia+1.488
asa+1.192
ounding+1.160
ine+1.150
Neg logit contributions
velt-1.957
catentry-1.839
REDACTED-1.777
••-1.718
joy-1.699
Token from
Token ID422
Feature activation-2.88e-02

Pos logit contributions
arest+3.330
unia+3.230
asa+2.622
ine+2.500
ounding+2.478
Neg logit contributions
velt-4.706
catentry-4.488
REDACTED-4.267
••-4.150
 Typh-4.134
Token Booker
Token ID31004
Feature activation+2.66e-02

Pos logit contributions
velt+0.269
catentry+0.248
REDACTED+0.241
••+0.236
atical+0.233
Neg logit contributions
arest-0.271
unia-0.259
asa-0.215
ounding-0.210
 Grammy-0.206
Token
Token ID447
Feature activation+1.36e-01

Pos logit contributions
arest+0.262
unia+0.251
asa+0.211
ounding+0.206
 Grammy+0.201
Neg logit contributions
velt-0.234
catentry-0.217
REDACTED-0.210
••-0.204
atical-0.202
TokenCo
Token ID7222
Feature activation+3.92e-02

Pos logit contributions
arest+0.507
unia+0.485
asa+0.405
ounding+0.393
 Grammy+0.384
Neg logit contributions
velt-0.496
catentry-0.456
REDACTED-0.442
••-0.430
 gamb-0.430
Tokenos
Token ID418
Feature activation+2.07e-01

Pos logit contributions
arest+0.360
unia+0.345
asa+0.286
ounding+0.278
 Grammy+0.272
Neg logit contributions
velt-0.372
catentry-0.342
REDACTED-0.334
••-0.327
atical-0.323
Token County
Token ID3418
Feature activation-1.58e-01

Pos logit contributions
arest+1.548
unia+1.474
asa+1.170
ounding+1.125
ine+1.116
Neg logit contributions
velt-2.272
catentry-2.161
REDACTED-2.089
atical-2.039
••-2.038
Token is
Token ID318
Feature activation-6.81e-02

Pos logit contributions
velt+1.449
catentry+1.324
REDACTED+1.298
••+1.279
 gamb+1.259
Neg logit contributions
arest-1.494
unia-1.432
asa-1.189
ounding-1.163
 Grammy-1.131
Token not
Token ID407
Feature activation+4.34e-01

Pos logit contributions
velt+0.629
catentry+0.577
REDACTED+0.564
••+0.554
 gamb+0.545
Neg logit contributions
arest-0.643
unia-0.616
asa-0.512
ounding-0.500
 Grammy-0.486
Token an
Token ID281
Feature activation+1.18e-01

Pos logit contributions
arest+3.744
unia+3.615
asa+2.981
 Grammy+2.902
ounding+2.881
Neg logit contributions
velt-4.337
catentry-4.018
REDACTED-3.862
••-3.698
atical-3.696
Token isolated
Token ID11557
Feature activation-1.60e-01

Pos logit contributions
arest+1.072
unia+1.036
asa+0.862
ounding+0.839
 Grammy+0.824
Neg logit contributions
velt-1.130
catentry-1.033
REDACTED-1.015
••-0.986
atical-0.967
Token case
Token ID1339
Feature activation+2.01e-01

Pos logit contributions
velt+1.316
catentry+1.193
REDACTED+1.162
••+1.154
 gamb+1.142
Neg logit contributions
arest-1.666
unia-1.599
asa-1.356
ounding-1.321
 Grammy-1.276
Token.
Token ID13
Feature activation+6.61e-02

Pos logit contributions
arest+1.662
unia+1.584
asa+1.293
ounding+1.242
ine+1.219
Neg logit contributions
velt-2.082
catentry-1.969
REDACTED-1.880
••-1.814
atical-1.802
Token Across
Token ID27497
Feature activation+1.73e-01

Pos logit contributions
arest+0.633
unia+0.606
asa+0.510
ounding+0.493
 Grammy+0.493
Neg logit contributions
velt-0.601
catentry-0.556
REDACTED-0.539
••-0.520
atical-0.520
Token the
Token ID262
Feature activation+4.97e-02

Pos logit contributions
arest+1.546
unia+1.458
asa+1.200
 Grammy+1.177
ounding+1.172
Neg logit contributions
velt-1.679
catentry-1.587
REDACTED-1.540
••-1.475
atical-1.459
Token:
Token ID25
Feature activation+1.36e-01

Pos logit contributions
velt+2.076
catentry+1.910
REDACTED+1.870
••+1.839
atical+1.815
Neg logit contributions
arest-1.947
unia-1.871
asa-1.542
ounding-1.495
 Grammy-1.456
Token These
Token ID2312
Feature activation+1.16e-01

Pos logit contributions
arest+1.297
unia+1.241
asa+1.042
ounding+1.014
 replay+1.008
Neg logit contributions
velt-1.240
catentry-1.141
REDACTED-1.112
••-1.086
atical-1.069
Token "
Token ID366
Feature activation+2.20e-01

Pos logit contributions
arest+1.075
unia+1.028
asa+0.855
ounding+0.830
 replay+0.816
Neg logit contributions
velt-1.094
catentry-1.008
REDACTED-0.982
••-0.961
atical-0.943
Tokenlab
Token ID23912
Feature activation-1.99e-01

Pos logit contributions
arest+2.229
unia+2.124
asa+1.812
ounding+1.758
ine+1.754
Neg logit contributions
velt-1.880
catentry-1.712
REDACTED-1.653
••-1.625
atical-1.603
Tokenyrinth
Token ID21324
Feature activation+4.37e-01

Pos logit contributions
velt+1.781
catentry+1.631
REDACTED+1.592
atical+1.560
••+1.552
Neg logit contributions
arest-1.895
unia-1.832
asa-1.520
ounding-1.467
 Grammy-1.461
Tokenine
Token ID500
Feature activation+2.82e-01

Pos logit contributions
arest+3.614
unia+3.428
ine+2.876
asa+2.822
 replay+2.738
Neg logit contributions
velt-4.468
catentry-4.218
REDACTED-4.104
••-3.983
joy-3.904
Token worlds
Token ID11621
Feature activation+1.78e-01

Pos logit contributions
arest+2.566
unia+2.462
asa+2.041
 replay+1.990
ounding+1.978
Neg logit contributions
velt-2.670
catentry-2.495
REDACTED-2.416
••-2.347
atical-2.323
Token filled
Token ID5901
Feature activation+7.47e-02

Pos logit contributions
arest+1.599
unia+1.526
asa+1.253
ounding+1.230
 replay+1.196
Neg logit contributions
velt-1.733
catentry-1.626
REDACTED-1.589
••-1.540
atical-1.502
Token with
Token ID351
Feature activation+8.02e-02

Pos logit contributions
arest+0.622
unia+0.596
asa+0.483
ounding+0.476
ine+0.460
Neg logit contributions
velt-0.771
catentry-0.720
REDACTED-0.704
••-0.685
atical-0.671
Token secrets
Token ID13141
Feature activation+2.87e-01

Pos logit contributions
arest+0.753
unia+0.721
asa+0.600
ounding+0.585
 replay+0.575
Neg logit contributions
velt-0.745
catentry-0.689
REDACTED-0.669
••-0.655
atical-0.644
Token encourage
Token ID7898
Feature activation+4.55e-01

Pos logit contributions
arest+2.514
unia+2.388
asa+1.951
ounding+1.913
 replay+1.898
Neg logit contributions
velt-2.842
catentry-2.663
REDACTED-2.593
••-2.523
joy-2.450
Token and
Token ID290
Feature activation+8.56e-02

Pos logit contributions
velt+1.034
catentry+0.943
REDACTED+0.920
••+0.905
 gamb+0.893
Neg logit contributions
arest-1.122
unia-1.075
asa-0.898
ounding-0.878
 Grammy-0.857
Token their
Token ID511
Feature activation-2.80e-02

Pos logit contributions
arest+0.794
unia+0.759
asa+0.630
ounding+0.613
 Grammy+0.603
Neg logit contributions
velt-0.806
catentry-0.745
REDACTED-0.728
••-0.710
atical-0.699
Token ability
Token ID2694
Feature activation-4.21e-02

Pos logit contributions
velt+0.257
catentry+0.236
REDACTED+0.230
••+0.225
atical+0.222
Neg logit contributions
arest-0.267
unia-0.255
asa-0.213
ounding-0.208
 Grammy-0.202
Token to
Token ID284
Feature activation-7.45e-02

Pos logit contributions
velt+0.396
catentry+0.366
REDACTED+0.358
••+0.349
atical+0.344
Neg logit contributions
arest-0.389
unia-0.373
asa-0.308
ounding-0.301
 Grammy-0.293
Token solve
Token ID8494
Feature activation+4.36e-01

Pos logit contributions
velt+0.710
catentry+0.652
REDACTED+0.636
••+0.625
 gamb+0.619
Neg logit contributions
arest-0.683
unia-0.652
asa-0.540
ounding-0.527
 Grammy-0.512
Token complex
Token ID3716
Feature activation+4.16e-01

Pos logit contributions
arest+3.593
unia+3.446
asa+2.729
 Grammy+2.686
ounding+2.677
Neg logit contributions
velt-4.450
catentry-4.209
REDACTED-4.123
••-3.969
joy-3.899
Token problems
Token ID2761
Feature activation-1.06e-01

Pos logit contributions
arest+3.128
unia+2.994
asa+2.312
 Grammy+2.272
 replay+2.258
Neg logit contributions
velt-4.534
catentry-4.259
REDACTED-4.210
••-4.036
joy-4.025
Token such
Token ID884
Feature activation+2.62e-02

Pos logit contributions
velt+1.003
catentry+0.922
REDACTED+0.902
••+0.885
atical+0.875
Neg logit contributions
arest-0.975
unia-0.933
asa-0.772
ounding-0.753
 Grammy-0.735
Token as
Token ID355
Feature activation+1.05e-01

Pos logit contributions
arest+0.244
unia+0.233
asa+0.194
ounding+0.189
 Grammy+0.184
Neg logit contributions
velt-0.245
catentry-0.226
REDACTED-0.221
••-0.216
atical-0.213
Token climate
Token ID4258
Feature activation+4.11e-01

Pos logit contributions
arest+0.922
unia+0.882
asa+0.720
ounding+0.700
 Grammy+0.689
Neg logit contributions
velt-1.032
catentry-0.960
REDACTED-0.940
••-0.912
atical-0.902
Token change
Token ID1487
Feature activation+3.47e-01

Pos logit contributions
unia+2.964
arest+2.953
asa+2.189
ounding+2.184
ine+2.183
Neg logit contributions
velt-4.493
catentry-4.327
REDACTED-4.261
joy-4.024
••-4.023
Token regions
Token ID7652
Feature activation+1.29e-01

Pos logit contributions
arest+3.407
unia+3.096
asa+2.485
ine+2.418
ounding+2.415
Neg logit contributions
velt-4.828
catentry-4.515
REDACTED-4.412
••-4.384
joy-4.318
Token (
Token ID357
Feature activation+2.26e-01

Pos logit contributions
arest+1.174
unia+1.117
asa+0.930
ounding+0.904
ine+0.886
Neg logit contributions
velt-1.231
catentry-1.146
REDACTED-1.106
••-1.085
atical-1.064
Tokeni
Token ID72
Feature activation+6.37e-01

Pos logit contributions
arest+2.283
unia+2.167
asa+1.858
ounding+1.802
ine+1.773
Neg logit contributions
velt-1.944
catentry-1.769
REDACTED-1.714
••-1.683
 gamb-1.652
Token.
Token ID13
Feature activation-9.31e-02

Pos logit contributions
arest+4.582
unia+4.278
asa+3.403
ine+3.292
ounding+3.174
Neg logit contributions
velt-7.090
catentry-6.844
••-6.413
 gamb-6.336
REDACTED-6.316
Tokene
Token ID68
Feature activation+6.46e-01

Pos logit contributions
velt+0.731
catentry+0.662
REDACTED+0.650
••+0.630
joy+0.619
Neg logit contributions
arest-1.002
unia-0.967
asa-0.827
ounding-0.811
 Grammy-0.800
Token.,
Token ID1539
Feature activation+3.82e-01

Pos logit contributions
arest+4.521
unia+4.064
asa+3.182
ounding+3.079
ine+3.022
Neg logit contributions
velt-7.331
catentry-7.059
REDACTED-6.712
••-6.537
atical-6.473
Token MR
Token ID17242
Feature activation-2.84e-01

Pos logit contributions
arest+3.482
unia+3.244
asa+2.725
ounding+2.634
ine+2.622
Neg logit contributions
velt-3.607
catentry-3.346
REDACTED-3.251
••-3.191
joy-3.127
TokenZ
Token ID57
Feature activation+5.53e-01

Pos logit contributions
velt+2.468
catentry+2.240
REDACTED+2.196
••+2.157
atical+2.121
Neg logit contributions
arest-2.799
unia-2.713
asa-2.261
ounding-2.188
 Grammy-2.168
Token text
Token ID2420
Feature activation+4.58e-01

Pos logit contributions
arest+4.754
unia+4.407
asa+3.684
ounding+3.611
ine+3.585
Neg logit contributions
velt-5.412
catentry-5.054
REDACTED-4.823
••-4.782
joy-4.709
Token)
Token ID8
Feature activation-7.03e-03

Pos logit contributions
arest+3.985
unia+3.750
asa+3.120
ounding+3.058
ine+3.017
Neg logit contributions
velt-4.477
catentry-4.199
REDACTED-4.040
••-4.000
atical-3.887
Token against
Token ID1028
Feature activation+5.96e-01

Pos logit contributions
velt+0.064
catentry+0.059
REDACTED+0.057
••+0.056
atical+0.055
Neg logit contributions
arest-0.068
unia-0.065
asa-0.054
ounding-0.053
 Grammy-0.052
Token.
Token ID13
Feature activation-2.42e-01

Pos logit contributions
arest+0.948
unia+0.905
asa+0.740
ounding+0.715
 Grammy+0.701
Neg logit contributions
velt-1.156
catentry-1.086
REDACTED-1.045
••-1.024
atical-1.014
Token "
Token ID366
Feature activation-1.14e-01

Pos logit contributions
velt+2.311
catentry+2.128
REDACTED+2.082
••+2.048
atical+2.008
Neg logit contributions
arest-2.206
unia-2.113
asa-1.742
ounding-1.693
 Grammy-1.648
TokenI
Token ID40
Feature activation-2.56e-01

Pos logit contributions
velt+0.963
catentry+0.878
REDACTED+0.867
••+0.852
joy+0.826
Neg logit contributions
arest-1.162
unia-1.124
asa-0.941
ounding-0.925
 Grammy-0.897
Token love
Token ID1842
Feature activation-1.88e-01

Pos logit contributions
velt+2.241
catentry+1.974
 gamb+1.973
REDACTED+1.944
atical+1.934
Neg logit contributions
arest-2.515
unia-2.424
asa-2.004
ounding-1.999
 Grammy-1.934
Token Hels
Token ID33889
Feature activation-1.41e-01

Pos logit contributions
velt+1.721
catentry+1.566
REDACTED+1.538
••+1.505
 gamb+1.488
Neg logit contributions
arest-1.797
unia-1.731
asa-1.435
ounding-1.403
 Grammy-1.365
Tokening
Token ID278
Feature activation+2.39e-01

Pos logit contributions
velt+1.570
catentry+1.414
REDACTED+1.400
••+1.391
 gamb+1.379
Neg logit contributions
arest-1.056
unia-1.010
asa-0.787
ounding-0.758
 Grammy-0.748
Tokenborg
Token ID23297
Feature activation-6.10e-02

Pos logit contributions
arest+3.065
unia+2.959
asa+2.622
ounding+2.558
 Grammy+2.531
Neg logit contributions
velt-1.413
catentry-1.263
REDACTED-1.185
••-1.142
atical-1.120
Token and
Token ID290
Feature activation+7.05e-02

Pos logit contributions
velt+0.570
catentry+0.520
REDACTED+0.510
••+0.500
 gamb+0.494
Neg logit contributions
arest-0.570
unia-0.547
asa-0.454
ounding-0.442
 Grammy-0.431
Token Sweden
Token ID10710
Feature activation-2.64e-01

Pos logit contributions
arest+0.648
unia+0.619
asa+0.517
ounding+0.499
 Grammy+0.492
Neg logit contributions
velt-0.668
catentry-0.620
REDACTED-0.604
••-0.591
atical-0.580
Token,
Token ID11
Feature activation+9.51e-02

Pos logit contributions
velt+2.403
catentry+2.159
REDACTED+2.125
 gamb+2.077
joy+2.073
Neg logit contributions
arest-2.518
unia-2.436
asa-2.004
ounding-1.959
 Grammy-1.915
Token but
Token ID475
Feature activation-5.70e-02

Pos logit contributions
arest+0.896
unia+0.854
asa+0.714
ounding+0.693
 Grammy+0.685
Neg logit contributions
velt-0.880
catentry-0.817
REDACTED-0.794
••-0.775
atical-0.760
Token it
Token ID340
Feature activation+1.07e-01

Pos logit contributions
velt+1.040
catentry+0.953
REDACTED+0.932
••+0.911
atical+0.900
Neg logit contributions
arest-1.124
unia-1.076
asa-0.903
ounding-0.880
 Grammy-0.856
Token mostly
Token ID4632
Feature activation-6.35e-02

Pos logit contributions
arest+0.988
unia+0.951
asa+0.790
ounding+0.763
 replay+0.751
Neg logit contributions
velt-1.000
catentry-0.944
REDACTED-0.906
••-0.889
joy-0.865
Token just
Token ID655
Feature activation-5.67e-02

Pos logit contributions
velt+0.550
catentry+0.500
REDACTED+0.489
••+0.478
atical+0.471
Neg logit contributions
arest-0.638
unia-0.612
asa-0.516
ounding-0.504
 Grammy-0.493
Token all
Token ID477
Feature activation-6.10e-02

Pos logit contributions
velt+0.497
catentry+0.455
REDACTED+0.443
••+0.433
atical+0.427
Neg logit contributions
arest-0.563
unia-0.541
asa-0.455
ounding-0.444
 Grammy-0.434
Token happened
Token ID3022
Feature activation-1.32e-01

Pos logit contributions
velt+0.500
catentry+0.451
REDACTED+0.442
••+0.431
 gamb+0.426
Neg logit contributions
arest-0.641
unia-0.616
asa-0.524
ounding-0.511
 Grammy-0.501
Token in
Token ID287
Feature activation+5.67e-03

Pos logit contributions
velt+1.223
catentry+1.118
REDACTED+1.099
••+1.072
atical+1.063
Neg logit contributions
arest-1.242
unia-1.183
asa-0.984
ounding-0.961
 Grammy-0.934
Token the
Token ID262
Feature activation-4.14e-01

Pos logit contributions
arest+0.054
unia+0.052
asa+0.043
ounding+0.042
 Grammy+0.041
Neg logit contributions
velt-0.052
catentry-0.048
REDACTED-0.047
••-0.046
atical-0.045
Token same
Token ID976
Feature activation+2.90e-01

Pos logit contributions
velt+3.580
catentry+3.295
REDACTED+3.249
 gamb+3.190
••+3.165
Neg logit contributions
arest-4.013
unia-3.852
asa-3.220
ounding-3.132
 Grammy-3.053
Token day
Token ID1110
Feature activation+1.46e-01

Pos logit contributions
arest+2.512
unia+2.420
asa+2.004
 replay+1.951
 Grammy+1.930
Neg logit contributions
velt-2.858
catentry-2.660
REDACTED-2.541
••-2.536
atical-2.476
Token.
Token ID13
Feature activation-1.07e-01

Pos logit contributions
arest+1.253
unia+1.197
asa+0.977
ounding+0.952
 Grammy+0.929
Neg logit contributions
velt-1.469
catentry-1.371
REDACTED-1.325
••-1.311
atical-1.281
Token
Token ID198
Feature activation-2.81e-01

Pos logit contributions
velt+0.988
catentry+0.907
REDACTED+0.890
••+0.876
atical+0.853
Neg logit contributions
arest-1.020
unia-0.977
asa-0.814
ounding-0.794
 Grammy-0.765
Token a
Token ID257
Feature activation-5.36e-01

Pos logit contributions
velt+2.746
catentry+2.464
REDACTED+2.449
••+2.394
 gamb+2.387
Neg logit contributions
arest-2.968
unia-2.824
asa-2.368
ounding-2.318
 Grammy-2.243
Token small
Token ID1402
Feature activation-2.19e-01

Pos logit contributions
velt+4.606
 gamb+4.060
REDACTED+4.050
catentry+4.043
••+3.962
Neg logit contributions
arest-5.270
unia-5.002
asa-4.238
ounding-4.141
 Grammy-3.963
Token amount
Token ID2033
Feature activation-5.07e-01

Pos logit contributions
velt+1.907
catentry+1.727
REDACTED+1.693
 gamb+1.653
••+1.653
Neg logit contributions
arest-2.188
unia-2.088
asa-1.759
ounding-1.719
 Grammy-1.672
Token of
Token ID286
Feature activation-3.08e-01

Pos logit contributions
velt+4.524
REDACTED+4.032
catentry+4.004
atical+3.994
 gamb+3.982
Neg logit contributions
arest-4.845
unia-4.558
asa-3.843
ounding-3.754
 Grammy-3.647
Token tooth
Token ID16162
Feature activation-4.80e-01

Pos logit contributions
velt+2.837
catentry+2.571
REDACTED+2.547
 gamb+2.483
••+2.481
Neg logit contributions
arest-2.909
unia-2.761
asa-2.307
ounding-2.253
 Grammy-2.172
Token from
Token ID422
Feature activation-1.59e-01

Pos logit contributions
velt+4.256
catentry+3.896
joy+3.780
••+3.737
atical+3.737
Neg logit contributions
arest-4.562
unia-4.309
asa-3.602
ounding-3.540
 Dame-3.470
Token each
Token ID1123
Feature activation-8.50e-02

Pos logit contributions
velt+1.482
catentry+1.358
REDACTED+1.331
••+1.303
atical+1.285
Neg logit contributions
arest-1.497
unia-1.430
asa-1.189
ounding-1.159
 Grammy-1.130
Token fossil
Token ID12584
Feature activation-3.58e-01

Pos logit contributions
velt+0.794
catentry+0.729
REDACTED+0.714
••+0.698
 gamb+0.689
Neg logit contributions
arest-0.795
unia-0.760
asa-0.631
ounding-0.615
 Grammy-0.600
Token.
Token ID13
Feature activation-2.00e-02

Pos logit contributions
velt+3.184
catentry+2.837
REDACTED+2.833
 gamb+2.818
••+2.774
Neg logit contributions
arest-3.461
unia-3.294
asa-2.764
ounding-2.691
 Dame-2.615
Token
Token ID198
Feature activation-6.63e-02

Pos logit contributions
velt+0.196
catentry+0.181
REDACTED+0.177
••+0.172
atical+0.171
Neg logit contributions
arest-0.178
unia-0.170
asa-0.140
ounding-0.136
 Grammy-0.133
Token
Token ID198
Feature activation-1.20e-01

Pos logit contributions
velt+0.645
catentry+0.592
REDACTED+0.587
••+0.574
joy+0.564
Neg logit contributions
arest-0.591
unia-0.566
asa-0.463
ounding-0.455
 Grammy-0.442
Tokenai
Token ID1872
Feature activation+6.58e-02

Pos logit contributions
arest+2.568
unia+2.476
asa+2.085
ine+1.989
ounding+1.975
Neg logit contributions
velt-2.548
catentry-2.386
REDACTED-2.333
••-2.261
atical-2.209
Token was
Token ID373
Feature activation+2.59e-01

Pos logit contributions
arest+0.612
unia+0.586
asa+0.490
ounding+0.471
ine+0.461
Neg logit contributions
velt-0.618
catentry-0.575
REDACTED-0.558
••-0.545
atical-0.538
Token a
Token ID257
Feature activation+1.28e-01

Pos logit contributions
arest+2.373
unia+2.265
asa+1.914
ounding+1.816
ine+1.810
Neg logit contributions
velt-2.466
catentry-2.285
REDACTED-2.204
••-2.151
atical-2.106
Token T
Token ID309
Feature activation+8.19e-02

Pos logit contributions
arest+1.197
unia+1.140
asa+0.959
ounding+0.925
 Grammy+0.915
Neg logit contributions
velt-1.199
catentry-1.104
REDACTED-1.076
••-1.049
atical-1.029
Tokenatar
Token ID9459
Feature activation-2.24e-01

Pos logit contributions
arest+0.788
unia+0.753
asa+0.636
ounding+0.617
ine+0.599
Neg logit contributions
velt-0.741
catentry-0.691
REDACTED-0.670
••-0.652
 gamb-0.642
Token (
Token ID357
Feature activation-2.40e-02

Pos logit contributions
velt+2.087
catentry+1.924
REDACTED+1.881
••+1.850
 gamb+1.846
Neg logit contributions
arest-2.089
unia-1.998
asa-1.636
ounding-1.607
 Grammy-1.565
Tokennot
Token ID1662
Feature activation+4.97e-01

Pos logit contributions
velt+0.216
catentry+0.199
REDACTED+0.194
••+0.190
atical+0.187
Neg logit contributions
arest-0.232
unia-0.223
asa-0.186
ounding-0.182
 Grammy-0.178
Token to
Token ID284
Feature activation+1.88e-01

Pos logit contributions
arest+4.546
unia+4.271
asa+3.687
ine+3.578
ounding+3.503
Neg logit contributions
velt-4.713
catentry-4.329
REDACTED-4.138
••-4.024
joy-3.956
Token be
Token ID307
Feature activation+4.86e-01

Pos logit contributions
arest+1.566
unia+1.489
ounding+1.192
asa+1.192
 Grammy+1.137
Neg logit contributions
velt-1.966
catentry-1.832
REDACTED-1.769
••-1.732
atical-1.715
Token confused
Token ID10416
Feature activation-7.35e-02

Pos logit contributions
arest+2.561
unia+2.324
asa+1.644
ounding+1.561
ine+1.515
Neg logit contributions
velt-6.495
catentry-6.178
REDACTED-5.935
••-5.872
joy-5.754
Token with
Token ID351
Feature activation-1.31e-01

Pos logit contributions
velt+0.736
catentry+0.683
REDACTED+0.665
••+0.651
atical+0.642
Neg logit contributions
arest-0.641
unia-0.611
asa-0.502
ounding-0.487
 Grammy-0.474
Token Myst
Token ID32347
Feature activation+3.01e-01

Pos logit contributions
velt+1.110
catentry+1.012
REDACTED+0.989
••+0.979
 gamb+0.946
Neg logit contributions
arest-1.295
unia-1.244
asa-1.048
ounding-1.032
 Grammy-0.995
Tokenification
Token ID2649
Feature activation+5.89e-01

Pos logit contributions
arest+2.525
unia+2.410
asa+1.947
ounding+1.897
ine+1.836
Neg logit contributions
velt-3.117
catentry-2.903
REDACTED-2.861
••-2.766
 gamb-2.722
Token is
Token ID318
Feature activation-2.75e-02

Pos logit contributions
arest+5.073
unia+4.909
asa+4.010
ounding+3.884
ine+3.840
Neg logit contributions
velt-5.606
catentry-5.388
REDACTED-5.125
••-4.955
joy-4.871
Token a
Token ID257
Feature activation+2.87e-01

Pos logit contributions
velt+0.252
catentry+0.232
REDACTED+0.226
••+0.221
atical+0.218
Neg logit contributions
arest-0.262
unia-0.251
asa-0.209
ounding-0.204
 Grammy-0.199
Token part
Token ID636
Feature activation-2.20e-01

Pos logit contributions
arest+2.731
unia+2.601
asa+2.155
ounding+2.116
 replay+2.086
Neg logit contributions
velt-2.612
catentry-2.421
REDACTED-2.372
••-2.265
atical-2.242
Token of
Token ID286
Feature activation-1.67e-01

Pos logit contributions
velt+2.178
catentry+2.015
REDACTED+1.970
••+1.938
atical+1.914
Neg logit contributions
arest-1.934
unia-1.846
asa-1.512
ounding-1.468
 Grammy-1.433
Token it
Token ID340
Feature activation+2.69e-01

Pos logit contributions
velt+1.538
catentry+1.404
REDACTED+1.369
••+1.346
 gamb+1.326
Neg logit contributions
arest-1.588
unia-1.525
asa-1.276
ounding-1.241
 Grammy-1.209
Token.
Token ID13
Feature activation-7.43e-02

Pos logit contributions
arest+2.277
unia+2.168
asa+1.772
ounding+1.720
ine+1.695
Neg logit contributions
velt-2.718
catentry-2.542
REDACTED-2.466
••-2.384
joy-2.341
Token
Token ID447
Feature activation+1.04e-01

Pos logit contributions
velt+0.653
catentry+0.596
REDACTED+0.581
••+0.573
atical+0.559
Neg logit contributions
arest-0.736
unia-0.708
asa-0.595
ounding-0.582
 Grammy-0.565
Token
Token ID251
Feature activation+1.32e-01

Pos logit contributions
arest+0.723
unia+0.686
asa+0.527
ounding+0.502
 Grammy+0.495
Neg logit contributions
velt-1.216
catentry-1.133
REDACTED-1.116
••-1.091
 gamb-1.081
Token
Token ID198
Feature activation+5.34e-02

Pos logit contributions
arest+1.195
unia+1.139
asa+0.944
ounding+0.912
 Grammy+0.906
Neg logit contributions
velt-1.257
catentry-1.166
REDACTED-1.136
atical-1.100
••-1.092
Token is
Token ID318
Feature activation-1.36e-02

Pos logit contributions
arest+0.181
unia+0.174
asa+0.145
ounding+0.140
 Grammy+0.137
Neg logit contributions
velt-0.179
catentry-0.166
REDACTED-0.161
••-0.158
atical-0.155
Token less
Token ID1342
Feature activation+1.02e-01

Pos logit contributions
velt+0.127
catentry+0.117
REDACTED+0.114
••+0.111
atical+0.109
Neg logit contributions
arest-0.127
unia-0.122
asa-0.101
ounding-0.098
 Grammy-0.096
Token adrenaline
Token ID37607
Feature activation+1.33e-01

Pos logit contributions
arest+0.927
unia+0.888
asa+0.739
ounding+0.716
 replay+0.706
Neg logit contributions
velt-0.981
catentry-0.903
REDACTED-0.882
••-0.864
atical-0.846
Token-
Token ID12
Feature activation+3.26e-01

Pos logit contributions
arest+1.293
unia+1.241
asa+1.045
ounding+1.030
 replay+1.004
Neg logit contributions
velt-1.188
catentry-1.092
REDACTED-1.066
••-1.044
atical-1.016
Tokenfuel
Token ID25802
Feature activation+6.04e-02

Pos logit contributions
arest+2.910
unia+2.784
asa+2.324
ounding+2.313
ine+2.279
Neg logit contributions
velt-3.093
catentry-2.908
REDACTED-2.819
••-2.772
atical-2.676
Tokened
Token ID276
Feature activation-1.76e-01

Pos logit contributions
arest+0.546
unia+0.524
asa+0.432
ounding+0.425
ine+0.410
Neg logit contributions
velt-0.581
catentry-0.537
REDACTED-0.526
••-0.516
atical-0.503
Token,
Token ID11
Feature activation-1.51e-01

Pos logit contributions
velt+1.656
catentry+1.524
REDACTED+1.484
••+1.455
 gamb+1.446
Neg logit contributions
arest-1.633
unia-1.556
asa-1.290
ounding-1.250
 Grammy-1.226
Token has
Token ID468
Feature activation+1.72e-01

Pos logit contributions
velt+1.280
catentry+1.163
REDACTED+1.135
••+1.104
 gamb+1.096
Neg logit contributions
arest-1.545
unia-1.488
asa-1.256
ounding-1.226
 Grammy-1.197
Token a
Token ID257
Feature activation+6.26e-03

Pos logit contributions
arest+1.508
unia+1.451
asa+1.192
 replay+1.146
ounding+1.143
Neg logit contributions
velt-1.690
catentry-1.568
REDACTED-1.512
••-1.487
atical-1.456
Token slightly
Token ID4622
Feature activation+1.57e-02

Pos logit contributions
arest+0.059
unia+0.057
asa+0.048
ounding+0.046
 Grammy+0.045
Neg logit contributions
velt-0.058
catentry-0.053
REDACTED-0.052
••-0.051
atical-0.050
Token slower
Token ID13611
Feature activation-9.35e-02

Pos logit contributions
arest+0.148
unia+0.143
asa+0.119
ounding+0.116
 Grammy+0.114
Neg logit contributions
velt-0.144
catentry-0.133
REDACTED-0.129
••-0.127
atical-0.124
Token meters
Token ID10700
Feature activation-2.79e-01

Pos logit contributions
velt+4.051
catentry+3.684
REDACTED+3.679
••+3.641
atical+3.571
Neg logit contributions
arest-3.591
unia-3.453
asa-2.785
ounding-2.741
 replay-2.629
Token per
Token ID583
Feature activation-2.06e-03

Pos logit contributions
velt+3.301
catentry+3.050
REDACTED+3.023
••+2.978
 gamb+2.950
Neg logit contributions
arest-1.899
unia-1.786
asa-1.355
ounding-1.311
 Grammy-1.250
Token second
Token ID1218
Feature activation-4.15e-01

Pos logit contributions
velt+0.019
catentry+0.017
REDACTED+0.017
••+0.016
atical+0.016
Neg logit contributions
arest-0.020
unia-0.019
asa-0.016
ounding-0.015
 Grammy-0.015
Token (
Token ID357
Feature activation-3.27e-02

Pos logit contributions
velt+3.794
catentry+3.426
REDACTED+3.385
••+3.326
atical+3.277
Neg logit contributions
arest-3.898
unia-3.752
asa-3.112
ounding-3.027
 Grammy-2.952
Token5
Token ID20
Feature activation+1.20e-01

Pos logit contributions
velt+0.303
catentry+0.279
REDACTED+0.273
••+0.267
atical+0.263
Neg logit contributions
arest-0.309
unia-0.296
asa-0.246
ounding-0.240
 Grammy-0.235
Token,
Token ID11
Feature activation-7.96e-01

Pos logit contributions
arest+0.985
unia+0.931
asa+0.764
ounding+0.742
 Grammy+0.731
Neg logit contributions
velt-1.244
catentry-1.155
REDACTED-1.119
atical-1.095
••-1.090
Token9
Token ID24
Feature activation+3.10e-01

Pos logit contributions
velt+5.374
REDACTED+4.870
catentry+4.778
••+4.775
joy+4.654
Neg logit contributions
arest-8.546
unia-8.382
 Grammy-7.058
asa-7.037
ounding-6.999
Token46
Token ID3510
Feature activation+3.48e-01

Pos logit contributions
arest+2.863
unia+2.746
asa+2.311
ine+2.237
ounding+2.220
Neg logit contributions
velt-2.902
catentry-2.692
REDACTED-2.558
atical-2.506
 gamb-2.495
Token miles
Token ID4608
Feature activation-3.75e-01

Pos logit contributions
arest+2.799
unia+2.627
asa+2.203
ounding+2.087
 Grammy+2.080
Neg logit contributions
velt-3.620
catentry-3.442
REDACTED-3.307
nah-3.190
••-3.172
Token per
Token ID583
Feature activation-7.51e-03

Pos logit contributions
velt+4.374
catentry+4.034
REDACTED+3.993
••+3.967
 gamb+3.911
Neg logit contributions
arest-2.612
unia-2.449
asa-1.856
ounding-1.797
 Grammy-1.731
Token hour
Token ID1711
Feature activation-2.28e-01

Pos logit contributions
velt+0.072
catentry+0.067
REDACTED+0.065
••+0.063
atical+0.062
Neg logit contributions
arest-0.068
unia-0.066
asa-0.054
ounding-0.053
 Grammy-0.052
Token was
Token ID373
Feature activation+2.72e-02

Pos logit contributions
velt+0.539
catentry+0.495
REDACTED+0.483
••+0.472
atical+0.465
Neg logit contributions
arest-0.572
unia-0.549
asa-0.459
ounding-0.447
 Grammy-0.437
Token introduced
Token ID5495
Feature activation-9.88e-02

Pos logit contributions
arest+0.271
unia+0.261
asa+0.220
ounding+0.214
 Grammy+0.211
Neg logit contributions
velt-0.237
catentry-0.217
REDACTED-0.211
••-0.206
atical-0.201
Token as
Token ID355
Feature activation-7.04e-02

Pos logit contributions
velt+0.916
catentry+0.838
REDACTED+0.819
••+0.806
atical+0.794
Neg logit contributions
arest-0.931
unia-0.891
asa-0.741
ounding-0.723
 Grammy-0.703
Token an
Token ID281
Feature activation-1.05e-01

Pos logit contributions
velt+0.681
catentry+0.630
REDACTED+0.615
••+0.603
atical+0.596
Neg logit contributions
arest-0.635
unia-0.606
asa-0.499
ounding-0.486
 Grammy-0.473
Token Excel
Token ID24134
Feature activation-2.14e-01

Pos logit contributions
velt+0.966
catentry+0.892
REDACTED+0.867
••+0.850
atical+0.843
Neg logit contributions
arest-0.998
unia-0.950
asa-0.787
ounding-0.768
 Grammy-0.751
Token add
Token ID751
Feature activation-8.89e-01

Pos logit contributions
velt+1.898
catentry+1.744
REDACTED+1.707
••+1.698
 gamb+1.665
Neg logit contributions
arest-2.068
unia-1.979
asa-1.645
ounding-1.594
 Grammy-1.578
Token-
Token ID12
Feature activation-7.21e-01

Pos logit contributions
velt+5.653
joy+5.218
catentry+5.124
••+5.101
atical+4.900
Neg logit contributions
arest-9.841
unia-9.262
asa-7.844
 Grammy-7.765
 Dame-7.672
Tokenin
Token ID259
Feature activation-4.31e-01

Pos logit contributions
velt+6.889
joy+6.396
REDACTED+6.342
••+6.342
catentry+6.264
Neg logit contributions
arest-5.966
unia-5.558
 Grammy-4.512
asa-4.455
 Dame-4.394
Token back
Token ID736
Feature activation+6.01e-03

Pos logit contributions
velt+3.653
catentry+3.311
••+3.252
REDACTED+3.245
 gamb+3.206
Neg logit contributions
arest-4.314
unia-4.051
asa-3.424
ounding-3.329
 Grammy-3.237
Token in
Token ID287
Feature activation-7.57e-02

Pos logit contributions
arest+0.056
unia+0.054
asa+0.045
ounding+0.043
 Grammy+0.042
Neg logit contributions
velt-0.056
catentry-0.052
REDACTED-0.050
••-0.049
atical-0.049
Token 2013
Token ID2211
Feature activation-2.00e-01

Pos logit contributions
velt+0.737
catentry+0.681
REDACTED+0.665
••+0.653
atical+0.644
Neg logit contributions
arest-0.678
unia-0.648
asa-0.534
ounding-0.519
 Grammy-0.505
Tokenensity
Token ID6377
Feature activation+4.20e-01

Pos logit contributions
arest+0.881
unia+0.830
asa+0.632
ounding+0.607
 Grammy+0.582
Neg logit contributions
velt-1.559
catentry-1.464
REDACTED-1.435
••-1.414
 gamb-1.396
Token )
Token ID1267
Feature activation-4.21e-01

Pos logit contributions
arest+2.867
unia+2.706
asa+2.101
ounding+2.021
ine+2.013
Neg logit contributions
velt-4.917
catentry-4.568
REDACTED-4.460
••-4.339
nah-4.296
Token *
Token ID1635
Feature activation-2.61e-01

Pos logit contributions
velt+3.613
catentry+3.329
••+3.189
REDACTED+3.184
atical+3.175
Neg logit contributions
arest-4.105
unia-3.960
asa-3.305
ounding-3.226
 Grammy-3.173
Token Get
Token ID3497
Feature activation-9.04e-02

Pos logit contributions
velt+2.341
catentry+2.168
REDACTED+2.122
••+2.079
 gamb+2.060
Neg logit contributions
arest-2.477
unia-2.384
asa-1.993
ounding-1.928
 Grammy-1.892
TokenVolume
Token ID31715
Feature activation-8.53e-02

Pos logit contributions
velt+0.936
catentry+0.869
REDACTED+0.853
••+0.836
atical+0.826
Neg logit contributions
arest-0.751
unia-0.716
asa-0.580
ounding-0.561
 Grammy-0.548
Token()
Token ID3419
Feature activation-8.42e-01

Pos logit contributions
velt+0.860
catentry+0.796
REDACTED+0.780
••+0.767
atical+0.757
Neg logit contributions
arest-0.734
unia-0.698
asa-0.571
ounding-0.550
 Grammy-0.540
Token ;
Token ID2162
Feature activation-4.54e-02

Pos logit contributions
velt+6.592
catentry+5.988
 gamb+5.784
REDACTED+5.736
••+5.730
Neg logit contributions
arest-8.129
unia-7.950
asa-6.670
 Grammy-6.443
 Dame-6.421
Token return
Token ID1441
Feature activation-2.12e-01

Pos logit contributions
velt+0.430
catentry+0.396
REDACTED+0.386
••+0.378
atical+0.373
Neg logit contributions
arest-0.420
unia-0.402
asa-0.332
ounding-0.324
 Grammy-0.317
Token f
Token ID277
Feature activation-1.52e-01

Pos logit contributions
velt+1.980
catentry+1.833
REDACTED+1.783
••+1.753
 gamb+1.739
Neg logit contributions
arest-1.940
unia-1.870
asa-1.555
ounding-1.500
 Grammy-1.461
TokenMass
Token ID20273
Feature activation-2.78e-01

Pos logit contributions
velt+1.570
catentry+1.456
REDACTED+1.428
••+1.403
atical+1.385
Neg logit contributions
arest-1.264
unia-1.206
asa-0.975
ounding-0.943
 Grammy-0.922
Token ;
Token ID2162
Feature activation-6.99e-01

Pos logit contributions
velt+2.496
catentry+2.266
REDACTED+2.232
••+2.207
atical+2.187
Neg logit contributions
arest-2.649
unia-2.559
asa-2.121
ounding-2.064
 Grammy-2.033
Token will
Token ID481
Feature activation-4.11e-02

Pos logit contributions
velt+1.904
catentry+1.729
REDACTED+1.698
••+1.669
 gamb+1.646
Neg logit contributions
arest-2.105
unia-2.020
asa-1.683
ounding-1.655
 Grammy-1.613
Token be
Token ID307
Feature activation-6.74e-02

Pos logit contributions
velt+0.402
catentry+0.372
REDACTED+0.363
••+0.355
atical+0.350
Neg logit contributions
arest-0.367
unia-0.351
asa-0.288
ounding-0.280
 Grammy-0.273
Token thr
Token ID5636
Feature activation-6.14e-01

Pos logit contributions
velt+0.635
catentry+0.584
REDACTED+0.571
••+0.559
atical+0.551
Neg logit contributions
arest-0.625
unia-0.598
asa-0.496
ounding-0.483
 Grammy-0.471
Tokenongs
Token ID28079
Feature activation-3.25e-01

Pos logit contributions
velt+4.795
atical+4.144
catentry+4.090
joy+4.057
••+4.025
Neg logit contributions
arest-6.216
unia-6.079
 Grammy-5.112
asa-5.072
 Dame-5.028
Token of
Token ID286
Feature activation-2.29e-01

Pos logit contributions
velt+2.989
REDACTED+2.690
catentry+2.685
••+2.652
 gamb+2.649
Neg logit contributions
arest-3.047
unia-2.901
asa-2.394
ounding-2.322
 Grammy-2.286
Token ad
Token ID512
Feature activation-7.94e-01

Pos logit contributions
velt+2.013
catentry+1.820
REDACTED+1.797
••+1.767
 gamb+1.746
Neg logit contributions
arest-2.270
unia-2.169
asa-1.817
ounding-1.779
 Grammy-1.726
Tokenoring
Token ID3255
Feature activation-2.44e-01

Pos logit contributions
velt+5.859
atical+5.190
 gamb+5.152
nah+5.137
joy+5.120
Neg logit contributions
arest-7.903
unia-7.620
 Dame-6.487
 Grammy-6.439
asa-6.414
Token fans
Token ID3296
Feature activation+7.00e-02

Pos logit contributions
velt+2.147
catentry+1.948
REDACTED+1.911
••+1.889
 gamb+1.874
Neg logit contributions
arest-2.412
unia-2.307
asa-1.936
ounding-1.893
 Grammy-1.828
Token wanting
Token ID10291
Feature activation-3.02e-01

Pos logit contributions
arest+0.675
unia+0.647
asa+0.542
ounding+0.529
 Grammy+0.516
Neg logit contributions
velt-0.633
catentry-0.584
REDACTED-0.567
••-0.554
atical-0.544
Token his
Token ID465
Feature activation+3.88e-01

Pos logit contributions
velt+2.599
catentry+2.345
••+2.296
REDACTED+2.293
 gamb+2.256
Neg logit contributions
arest-3.024
unia-2.879
asa-2.434
ounding-2.378
 Grammy-2.309
Token aut
Token ID1960
Feature activation+3.93e-01

Pos logit contributions
arest+3.590
unia+3.428
asa+2.839
 replay+2.838
 Grammy+2.778
Neg logit contributions
velt-3.613
catentry-3.393
REDACTED-3.220
••-3.080
joy-3.071
Token gas
Token ID3623
Feature activation+5.10e-01

Pos logit contributions
velt+0.667
catentry+0.614
REDACTED+0.600
••+0.588
atical+0.579
Neg logit contributions
arest-0.654
unia-0.626
asa-0.519
ounding-0.505
 Grammy-0.493
Token fracking
Token ID25441
Feature activation-2.22e-01

Pos logit contributions
arest+4.252
unia+4.127
asa+3.336
ounding+3.297
ine+3.273
Neg logit contributions
velt-4.975
catentry-4.707
REDACTED-4.691
••-4.532
 Gamb-4.426
Token operations
Token ID4560
Feature activation-5.30e-01

Pos logit contributions
velt+1.946
catentry+1.747
REDACTED+1.717
••+1.701
atical+1.677
Neg logit contributions
arest-2.213
unia-2.114
asa-1.775
ounding-1.740
 Grammy-1.696
Token began
Token ID2540
Feature activation-6.02e-01

Pos logit contributions
velt+4.164
••+3.627
 gamb+3.606
REDACTED+3.598
atical+3.583
Neg logit contributions
arest-5.625
unia-5.331
asa-4.497
ounding-4.425
 replay-4.326
Token,
Token ID11
Feature activation-4.43e-01

Pos logit contributions
velt+4.982
••+4.422
REDACTED+4.352
catentry+4.284
 gamb+4.275
Neg logit contributions
arest-6.139
unia-5.825
asa-4.881
ounding-4.731
 replay-4.636
Token had
Token ID550
Feature activation-7.74e-01

Pos logit contributions
velt+3.764
catentry+3.417
REDACTED+3.369
 gamb+3.351
••+3.327
Neg logit contributions
arest-4.356
unia-4.175
asa-3.476
ounding-3.464
 Grammy-3.327
Token 8
Token ID807
Feature activation-1.08e-01

Pos logit contributions
velt+5.879
 gamb+5.555
••+5.405
 Gamb+5.380
 Typh+5.329
Neg logit contributions
arest-7.832
unia-7.376
asa-6.259
ounding-6.257
xton-5.862
Token57
Token ID3553
Feature activation-8.24e-02

Pos logit contributions
velt+0.966
catentry+0.889
REDACTED+0.864
••+0.846
atical+0.833
Neg logit contributions
arest-1.053
unia-1.012
asa-0.846
ounding-0.823
 Grammy-0.803
Token earthquakes
Token ID29781
Feature activation-1.34e-01

Pos logit contributions
velt+0.658
catentry+0.599
REDACTED+0.581
••+0.567
atical+0.557
Neg logit contributions
arest-0.881
unia-0.848
asa-0.723
ounding-0.707
 Grammy-0.693
Token last
Token ID938
Feature activation-6.97e-02

Pos logit contributions
velt+1.202
catentry+1.101
REDACTED+1.072
••+1.052
 gamb+1.042
Neg logit contributions
arest-1.311
unia-1.256
asa-1.048
ounding-1.023
 Grammy-0.997
Token year
Token ID614
Feature activation+6.96e-02

Pos logit contributions
velt+0.658
catentry+0.607
REDACTED+0.592
••+0.581
atical+0.572
Neg logit contributions
arest-0.644
unia-0.616
asa-0.511
ounding-0.497
 Grammy-0.485