TOP ACTIVATIONS
MAX = 1.017

While [ Key Ex ists Q [ step New
2 - 1 / 2 "] [ EC O " A
ijk a an Z ee "] [ Date " 2016 .
time [ 10 : 47 ] < met ros xual th
- July 6 , 2010 Subject : R iff
/ u I ba H Ex TT j B orm
- February 9 , 2010 Subject : R ipe
2016 . 01 . 29 "] [ Round " 11 "]
uk / in coming / article 35 79 23 12 .
35 790 347 . html https :// www . bel fast
- January 15 , 2010 Subject : Que er
- December 5 , 2009 Subject : Que ers
to have a none - too - sub tle dig at
https :// www . eff . org / d ee
v 2 . morph ology Ex ( gray , c v
but maybe Kris Let ang also lacks character ? I
presidential nominee Hillary Clinton Hillary Diane Rodham Clinton Sanders : '
Tata Steel Chess Chall engers "] [ Site " W ijk
k [ 10 : 47 ] < qu ij ii >
against 41 , 88 11 tanks against 35 12 ), with

MIN ACTIVATIONS
MIN = -1.066

SEC OND PART RE QU IR ES A BIT MORE EFF
AND STEP H EN M atur in first meet in April
. Materials and Methods Experimental subjects Three female r
" What does it mean to an intro vert online
mates , V Ps See Gallery Donald Trump passes on '
Q . What does this mean in terms of data collection
based simulations . Remember from Part 6 that when
man : What does that mean , operators in the industry
Morris ville ( Note : edited for length .)
the firing line during Commons Question Period Wednesday as Conservative MPs
What does this mean ? A lot of messages
Posts # 16 Glad to hear everything worked out great .
it to ' lead by example with hope , never fear
T akeru Kob ay ashi s
of the statements . For example , the sax ophon ist
is a silly , innocuous example ( who cares if a
he aping tablespoons of peanut butter : Bl end
Amazon itself . Remember , though , that thanks
images " ] ) : # load the image , resize
- protected expressions . For example , in 2010 the ACLU

INTERVAL 0.496 - 1.017
CONTAINS 0.331%

me one easy se gue into the metal portion of this
at the off - season additions Marc Ber ge vin has
more . Also , as before , the grid automatically sizes
January 15 , 2010 Subject : Que er magnet
assured her that if she filled out the police report ,

INTERVAL -0.025 - 0.496
CONTAINS 56.986%

ville has already broken up that trio . L add played
their message through . The har bing ers of
. Participant Anna B erc owitz noted four primary factions at
part of his appeal . I also recognize that his campaign
accepted a two - year deal at the minimum from the

INTERVAL -0.545 - -0.025
CONTAINS 42.498%

story , using flashbacks to do so not unlike Patty
promised to build a stronger relationship with ab orig inals .
hen ando ah ) speaks on the floor of the House
prescriptions are dangerous . A run - up in prescribing from
to some reporters that he was in New York City ,

INTERVAL -1.066 - -0.545
CONTAINS 0.185%

obsolete . This does not mean that older teachers who remain
in Philly mirror ing certain attributes of black speech .
" for short . Just remember , it 's enough "
. clone (), pieces : self . pieces . clone (),
by President Barack Obama , CNN confirmed from several sources .
Token
Token ID198
Feature activation+4.22e-02

Pos logit contributions
+0.415
Page+0.414
Bloom+0.409
mination+0.398
��+0.389
Neg logit contributions
 BDS-0.421
 NEO-0.373
 positively-0.324
 withd-0.310
 negatively-0.307
Token
Token ID198
Feature activation+1.74e-02

Pos logit contributions
Page+0.126
+0.125
Bloom+0.124
mination+0.120
Range+0.117
Neg logit contributions
 BDS-0.122
 NEO-0.108
 withd-0.094
 positively-0.094
 negatively-0.089
TokenWhile
Token ID3633
Feature activation+1.26e-01

Pos logit contributions
+0.041
Page+0.041
Bloom+0.040
mination+0.039
��+0.038
Neg logit contributions
 BDS-0.061
 NEO-0.055
 positively-0.049
 withd-0.047
 negatively-0.047
Token[
Token ID58
Feature activation+7.77e-02

Pos logit contributions
Page+0.357
+0.357
Bloom+0.352
mination+0.342
��+0.337
Neg logit contributions
 BDS-0.382
 NEO-0.340
 positively-0.295
 withd-0.284
ppelin-0.282
Token Key
Token ID7383
Feature activation-3.41e-01

Pos logit contributions
+0.185
Page+0.183
Bloom+0.181
mination+0.175
��+0.172
Neg logit contributions
 BDS-0.273
 NEO-0.247
 positively-0.219
 withd-0.211
 negatively-0.210
TokenEx
Token ID3109
Feature activation+1.02e+00

Pos logit contributions
 BDS+1.224
 NEO+1.109
 positively+0.988
 withd+0.950
 negatively+0.948
Neg logit contributions
-0.778
Bloom-0.758
Page-0.752
mination-0.734
��-0.727
Tokenists
Token ID1023
Feature activation+3.34e-01

Pos logit contributions
Page+2.447
+2.445
Bloom+2.377
mination+2.318
Range+2.258
Neg logit contributions
 BDS-3.513
 NEO-3.194
 positively-2.838
 withd-2.738
 negatively-2.715
TokenQ
Token ID48
Feature activation+2.85e-01

Pos logit contributions
+0.826
Page+0.822
Bloom+0.811
mination+0.785
��+0.770
Neg logit contributions
 BDS-1.139
 NEO-1.027
 positively-0.912
 withd-0.877
 negatively-0.871
Token[
Token ID58
Feature activation+1.11e-01

Pos logit contributions
+0.683
Page+0.680
Bloom+0.669
mination+0.648
��+0.635
Neg logit contributions
 BDS-0.995
 NEO-0.900
 positively-0.801
 withd-0.771
 negatively-0.765
Tokenstep
Token ID9662
Feature activation+2.71e-01

Pos logit contributions
+0.277
Page+0.275
Bloom+0.271
mination+0.263
��+0.258
Neg logit contributions
 BDS-0.374
 NEO-0.337
 positively-0.299
 withd-0.287
 negatively-0.285
TokenNew
Token ID3791
Feature activation+1.39e-01

Pos logit contributions
+0.694
Page+0.690
Bloom+0.681
mination+0.660
��+0.648
Neg logit contributions
 BDS-0.903
 NEO-0.812
 positively-0.718
 withd-0.689
 negatively-0.684
Token2
Token ID17
Feature activation-1.37e-01

Pos logit contributions
+0.192
Page+0.189
Bloom+0.187
mination+0.181
��+0.179
Neg logit contributions
 BDS-0.277
 NEO-0.250
 positively-0.222
 withd-0.212
 negatively-0.212
Token-
Token ID12
Feature activation-1.00e-01

Pos logit contributions
 BDS+0.461
 NEO+0.415
 positively+0.367
 negatively+0.349
 withd+0.348
Neg logit contributions
-0.352
Bloom-0.344
Page-0.344
mination-0.330
��-0.327
Token1
Token ID16
Feature activation-3.23e-02

Pos logit contributions
 BDS+0.351
 NEO+0.319
 positively+0.283
 withd+0.270
 negatively+0.270
Neg logit contributions
-0.239
Page-0.235
Bloom-0.234
mination-0.225
��-0.223
Token/
Token ID14
Feature activation+2.71e-02

Pos logit contributions
 BDS+0.095
 NEO+0.084
 positively+0.073
 withd+0.069
 negatively+0.069
Neg logit contributions
-0.095
Page-0.095
Bloom-0.094
mination-0.091
��-0.090
Token2
Token ID17
Feature activation-2.02e-01

Pos logit contributions
+0.066
Page+0.065
Bloom+0.064
mination+0.062
��+0.061
Neg logit contributions
 BDS-0.094
 NEO-0.085
 positively-0.075
 withd-0.072
 negatively-0.072
Token"]
Token ID8973
Feature activation+1.01e+00

Pos logit contributions
 BDS+0.548
 NEO+0.481
 positively+0.409
 withd+0.384
 negatively+0.383
Neg logit contributions
-0.644
Page-0.636
Bloom-0.634
mination-0.615
��-0.610
Token [
Token ID685
Feature activation+1.85e-01

Pos logit contributions
Page+2.193
+2.177
Bloom+2.139
mination+2.064
Range+1.990
Neg logit contributions
 BDS-3.714
 NEO-3.338
 withd-3.096
 positively-3.084
 negatively-2.969
TokenEC
Token ID2943
Feature activation+4.58e-01

Pos logit contributions
Page+0.347
+0.346
Bloom+0.337
mination+0.319
Range+0.310
Neg logit contributions
 BDS-0.746
 NEO-0.681
 positively-0.624
 withd-0.608
 negatively-0.603
TokenO
Token ID46
Feature activation+1.89e-01

Pos logit contributions
+1.115
Page+1.108
Bloom+1.091
mination+1.060
��+1.035
Neg logit contributions
 BDS-1.584
 NEO-1.428
 positively-1.268
 withd-1.226
 negatively-1.213
Token "
Token ID366
Feature activation+1.25e-02

Pos logit contributions
+0.545
Page+0.541
Bloom+0.534
mination+0.521
��+0.511
Neg logit contributions
 BDS-0.567
 NEO-0.505
 positively-0.440
 withd-0.422
 negatively-0.416
TokenA
Token ID32
Feature activation-2.43e-01

Pos logit contributions
Page+0.031
+0.031
Bloom+0.031
mination+0.030
��+0.029
Neg logit contributions
 BDS-0.042
 NEO-0.038
 positively-0.034
 withd-0.033
 negatively-0.032
Tokenijk
Token ID45961
Feature activation+1.46e-01

Pos logit contributions
+0.318
Page+0.316
Bloom+0.312
mination+0.303
��+0.298
Neg logit contributions
 BDS-0.407
 NEO-0.366
 positively-0.323
 withd-0.310
 negatively-0.308
Token a
Token ID257
Feature activation-1.76e-01

Pos logit contributions
+0.392
Page+0.391
Bloom+0.388
mination+0.378
��+0.367
Neg logit contributions
 BDS-0.460
 NEO-0.413
 positively-0.367
 withd-0.348
 negatively-0.345
Tokenan
Token ID272
Feature activation+4.10e-03

Pos logit contributions
 BDS+0.556
 NEO+0.497
 positively+0.436
 withd+0.417
 negatively+0.414
Neg logit contributions
-0.479
Page-0.474
Bloom-0.470
mination-0.456
��-0.449
Token Z
Token ID1168
Feature activation-2.00e-02

Pos logit contributions
Page+0.014
+0.014
Bloom+0.014
��+0.014
mination+0.013
Neg logit contributions
 BDS-0.010
 NEO-0.009
 positively-0.007
 withd-0.007
ppelin-0.007
Tokenee
Token ID1453
Feature activation+1.27e-01

Pos logit contributions
 BDS+0.080
 NEO+0.073
 positively+0.066
 withd+0.064
 negatively+0.064
Neg logit contributions
Page-0.037
-0.037
Bloom-0.036
mination-0.035
��-0.034
Token"]
Token ID8973
Feature activation+1.01e+00

Pos logit contributions
+0.359
Page+0.358
Bloom+0.355
mination+0.345
��+0.339
Neg logit contributions
 BDS-0.387
 NEO-0.344
 positively-0.302
 withd-0.290
 negatively-0.286
Token [
Token ID685
Feature activation+2.02e-02

Pos logit contributions
Page+2.073
+2.019
Bloom+2.000
mination+1.934
Range+1.874
Neg logit contributions
 BDS-3.803
 NEO-3.453
 withd-3.231
 positively-3.213
 negatively-3.100
TokenDate
Token ID10430
Feature activation+3.91e-01

Pos logit contributions
Page+0.018
+0.018
Bloom+0.017
mination+0.015
Range+0.014
Neg logit contributions
 BDS-0.101
 NEO-0.094
 positively-0.087
 withd-0.086
 negatively-0.085
Token "
Token ID366
Feature activation+7.12e-02

Pos logit contributions
+1.127
Page+1.124
Bloom+1.110
mination+1.075
��+1.060
Neg logit contributions
 BDS-1.173
 NEO-1.040
 positively-0.906
 withd-0.872
 negatively-0.857
Token2016
Token ID5304
Feature activation-2.65e-01

Pos logit contributions
+0.160
Page+0.159
Bloom+0.157
mination+0.151
��+0.148
Neg logit contributions
 BDS-0.259
 NEO-0.235
 positively-0.210
 withd-0.203
 negatively-0.202
Token.
Token ID13
Feature activation-2.99e-01

Pos logit contributions
 BDS+0.857
 NEO+0.769
 positively+0.677
 withd+0.651
 negatively+0.644
Neg logit contributions
-0.702
Page-0.697
Bloom-0.689
mination-0.669
��-0.657
Token time
Token ID640
Feature activation+1.38e-01

Pos logit contributions
+0.380
Page+0.376
Bloom+0.373
mination+0.364
��+0.356
Neg logit contributions
 BDS-0.372
 NEO-0.330
 positively-0.285
 withd-0.273
ppelin-0.270
Token [
Token ID685
Feature activation+1.97e-01

Pos logit contributions
+0.354
Page+0.351
Bloom+0.346
mination+0.336
��+0.329
Neg logit contributions
 BDS-0.459
 NEO-0.413
 positively-0.365
 withd-0.351
 negatively-0.348
Token10
Token ID940
Feature activation+1.33e-01

Pos logit contributions
+0.448
Page+0.443
Bloom+0.438
mination+0.423
��+0.415
Neg logit contributions
 BDS-0.712
 NEO-0.647
 positively-0.578
 withd-0.557
 negatively-0.554
Token:
Token ID25
Feature activation+1.70e-01

Pos logit contributions
+0.376
Page+0.373
Bloom+0.367
mination+0.358
��+0.352
Neg logit contributions
 BDS-0.407
 NEO-0.362
 positively-0.317
 withd-0.304
 negatively-0.300
Token47
Token ID2857
Feature activation+6.21e-01

Pos logit contributions
Page+0.390
+0.388
Bloom+0.382
mination+0.367
Beast+0.355
Neg logit contributions
 BDS-0.611
 NEO-0.554
 positively-0.498
 withd-0.488
 negatively-0.477
Token]
Token ID60
Feature activation+1.01e+00

Pos logit contributions
+1.616
Page+1.613
Bloom+1.593
mination+1.540
��+1.494
Neg logit contributions
 BDS-2.005
 NEO-1.805
 positively-1.594
 withd-1.552
 negatively-1.522
Token <
Token ID1279
Feature activation+8.34e-01

Pos logit contributions
+1.866
Page+1.839
Bloom+1.806
mination+1.737
��+1.678
Neg logit contributions
 BDS-4.024
 NEO-3.683
 positively-3.341
ppelin-3.251
 withd-3.251
Tokenmet
Token ID4164
Feature activation+3.69e-01

Pos logit contributions
Page+2.052
+2.047
Bloom+2.021
mination+1.943
Beast+1.898
Neg logit contributions
 BDS-2.854
 NEO-2.568
 positively-2.288
 withd-2.213
 negatively-2.194
Tokenros
Token ID4951
Feature activation+2.43e-01

Pos logit contributions
+0.864
Page+0.855
Bloom+0.846
mination+0.818
��+0.801
Neg logit contributions
 BDS-1.307
 NEO-1.183
 positively-1.054
 withd-1.015
 negatively-1.009
Tokenxual
Token ID5541
Feature activation+1.41e-01

Pos logit contributions
+0.545
Page+0.540
Bloom+0.533
mination+0.514
��+0.503
Neg logit contributions
 BDS-0.888
 NEO-0.806
 positively-0.720
 withd-0.695
 negatively-0.691
Tokenth
Token ID400
Feature activation+4.48e-02

Pos logit contributions
+0.247
Page+0.244
Bloom+0.241
mination+0.230
��+0.225
Neg logit contributions
 BDS-0.579
 NEO-0.532
 positively-0.484
 withd-0.468
 negatively-0.467
Token -
Token ID532
Feature activation+6.07e-01

Pos logit contributions
+0.173
Page+0.170
Bloom+0.169
mination+0.164
��+0.161
Neg logit contributions
 BDS-0.232
 NEO-0.209
 positively-0.186
 withd-0.178
 negatively-0.177
Token July
Token ID2901
Feature activation+5.57e-01

Pos logit contributions
Page+1.888
+1.882
Bloom+1.860
mination+1.802
Beast+1.757
Neg logit contributions
 BDS-1.650
 NEO-1.451
 positively-1.251
 withd-1.230
ppelin-1.191
Token 6
Token ID718
Feature activation+3.03e-01

Pos logit contributions
Page+1.427
+1.411
Bloom+1.396
mination+1.345
Beast+1.308
Neg logit contributions
 BDS-1.822
 NEO-1.645
 withd-1.478
 positively-1.473
 negatively-1.407
Token,
Token ID11
Feature activation+2.70e-01

Pos logit contributions
Page+0.811
+0.802
Bloom+0.793
mination+0.766
Range+0.744
Neg logit contributions
 BDS-0.973
 NEO-0.878
 withd-0.786
 positively-0.771
 negatively-0.738
Token 2010
Token ID3050
Feature activation+6.83e-01

Pos logit contributions
+0.804
Page+0.798
Bloom+0.791
mination+0.770
��+0.759
Neg logit contributions
 BDS-0.788
 NEO-0.697
 positively-0.603
 withd-0.574
 negatively-0.570
Token
Token ID198
Feature activation+9.68e-01

Pos logit contributions
Page+1.973
+1.925
Bloom+1.917
mination+1.855
Beast+1.804
Neg logit contributions
 BDS-2.033
 NEO-1.822
 withd-1.690
 positively-1.603
 negatively-1.533
Token
Token ID198
Feature activation+7.43e-01

Pos logit contributions
Page+2.667
Bloom+2.565
+2.556
mination+2.462
Range+2.428
Neg logit contributions
 BDS-2.974
 NEO-2.714
 withd-2.656
 positively-2.385
 compr-2.313
TokenSubject
Token ID19776
Feature activation+2.99e-01

Pos logit contributions
Page+1.291
+1.240
Bloom+1.238
mination+1.169
Reviewed+1.140
Neg logit contributions
 BDS-3.130
 NEO-2.897
 withd-2.666
 positively-2.626
 negatively-2.555
Token:
Token ID25
Feature activation-3.16e-02

Pos logit contributions
Page+0.806
+0.787
Bloom+0.776
mination+0.752
Range+0.733
Neg logit contributions
 BDS-0.962
 NEO-0.869
 withd-0.794
 positively-0.759
 negatively-0.730
Token R
Token ID371
Feature activation-2.99e-01

Pos logit contributions
 BDS+0.093
 NEO+0.083
 positively+0.072
 withd+0.069
ppelin+0.068
Neg logit contributions
-0.092
Page-0.092
Bloom-0.091
mination-0.088
��-0.086
Tokeniff
Token ID733
Feature activation+1.45e-01

Pos logit contributions
 BDS+0.998
 NEO+0.897
 positively+0.794
 withd+0.760
 negatively+0.756
Neg logit contributions
-0.764
Page-0.754
Bloom-0.748
mination-0.725
��-0.717
Token/
Token ID14
Feature activation+3.46e-02

Pos logit contributions
 BDS+1.005
 NEO+0.891
 positively+0.772
 withd+0.739
 negatively+0.730
Neg logit contributions
-1.009
Page-1.001
Bloom-0.992
mination-0.965
��-0.948
Tokenu
Token ID84
Feature activation+1.71e-01

Pos logit contributions
+0.084
Page+0.084
Bloom+0.083
mination+0.080
��+0.078
Neg logit contributions
 BDS-0.119
 NEO-0.108
 positively-0.096
 withd-0.092
 negatively-0.091
TokenI
Token ID40
Feature activation+1.55e-01

Pos logit contributions
+0.424
Page+0.423
Bloom+0.418
mination+0.403
��+0.395
Neg logit contributions
 BDS-0.579
 NEO-0.523
 positively-0.463
 withd-0.447
 negatively-0.443
Tokenba
Token ID7012
Feature activation+3.75e-01

Pos logit contributions
+0.369
Page+0.367
Bloom+0.362
mination+0.350
��+0.342
Neg logit contributions
 BDS-0.547
 NEO-0.495
 positively-0.440
 withd-0.425
 negatively-0.422
TokenH
Token ID39
Feature activation+5.81e-01

Pos logit contributions
Page+0.932
+0.930
Bloom+0.919
mination+0.884
��+0.864
Neg logit contributions
 BDS-1.275
 NEO-1.150
 positively-1.019
 withd-0.989
 negatively-0.976
TokenEx
Token ID3109
Feature activation+9.65e-01

Pos logit contributions
Page+1.390
+1.377
Bloom+1.369
mination+1.310
Beast+1.273
Neg logit contributions
 BDS-2.030
 NEO-1.839
 positively-1.635
 withd-1.593
 negatively-1.570
TokenTT
Token ID15751
Feature activation+4.93e-01

Pos logit contributions
Page+2.361
Bloom+2.318
+2.314
mination+2.214
Beast+2.157
Neg logit contributions
 BDS-3.310
 NEO-2.997
 positively-2.657
 withd-2.613
 negatively-2.558
Tokenj
Token ID73
Feature activation+1.52e-01

Pos logit contributions
Page+1.185
+1.178
Bloom+1.162
mination+1.120
��+1.085
Neg logit contributions
 BDS-1.715
 NEO-1.551
 positively-1.378
 withd-1.345
 negatively-1.321
Token
Token ID851
Feature activation+4.40e-01

Pos logit contributions
+0.387
Page+0.385
Bloom+0.379
mination+0.367
��+0.360
Neg logit contributions
 BDS-0.507
 NEO-0.456
 positively-0.403
 withd-0.390
 negatively-0.385
Token B
Token ID347
Feature activation+5.34e-03

Pos logit contributions
+1.329
Page+1.320
Bloom+1.300
mination+1.264
��+1.247
Neg logit contributions
 BDS-1.249
 NEO-1.102
 positively-0.952
 withd-0.910
 negatively-0.902
Tokenorm
Token ID579
Feature activation-1.30e-01

Pos logit contributions
+0.014
Page+0.014
Bloom+0.013
mination+0.013
��+0.013
Neg logit contributions
 BDS-0.018
 NEO-0.016
 positively-0.014
 withd-0.014
 negatively-0.013
Token -
Token ID532
Feature activation+5.28e-01

Pos logit contributions
+0.179
Page+0.176
Bloom+0.173
mination+0.165
��+0.161
Neg logit contributions
 BDS-0.451
 NEO-0.415
 positively-0.378
 withd-0.366
 negatively-0.365
Token February
Token ID3945
Feature activation+4.55e-01

Pos logit contributions
Page+1.587
+1.580
Bloom+1.564
mination+1.513
Beast+1.476
Neg logit contributions
 BDS-1.498
 NEO-1.324
 positively-1.147
 withd-1.124
ppelin-1.093
Token 9
Token ID860
Feature activation+2.39e-01

Pos logit contributions
Page+1.186
+1.180
Bloom+1.163
mination+1.125
Beast+1.090
Neg logit contributions
 BDS-1.474
 NEO-1.326
 positively-1.182
 withd-1.170
 negatively-1.128
Token,
Token ID11
Feature activation+2.67e-01

Pos logit contributions
Page+0.643
+0.639
Bloom+0.631
mination+0.610
Range+0.591
Neg logit contributions
 BDS-0.760
 NEO-0.685
 withd-0.607
 positively-0.602
 negatively-0.575
Token 2010
Token ID3050
Feature activation+6.12e-01

Pos logit contributions
+0.815
Page+0.807
Bloom+0.803
mination+0.780
��+0.774
Neg logit contributions
 BDS-0.757
 NEO-0.667
 positively-0.575
 withd-0.543
 negatively-0.542
Token
Token ID198
Feature activation+9.27e-01

Pos logit contributions
Page+1.788
+1.750
Bloom+1.741
mination+1.683
Beast+1.643
Neg logit contributions
 BDS-1.809
 NEO-1.616
 withd-1.481
 positively-1.421
 negatively-1.355
Token
Token ID198
Feature activation+7.15e-01

Pos logit contributions
Page+2.573
Bloom+2.477
+2.474
mination+2.383
Range+2.350
Neg logit contributions
 BDS-2.838
 NEO-2.581
 withd-2.509
 positively-2.266
 compr-2.181
TokenSubject
Token ID19776
Feature activation+2.88e-01

Pos logit contributions
Page+1.272
+1.233
Bloom+1.228
mination+1.160
Beast+1.123
Neg logit contributions
 BDS-2.975
 NEO-2.746
 withd-2.505
 positively-2.487
 negatively-2.410
Token:
Token ID25
Feature activation-2.96e-02

Pos logit contributions
Page+0.779
+0.764
Bloom+0.752
mination+0.729
Range+0.710
Neg logit contributions
 BDS-0.924
 NEO-0.832
 withd-0.756
 positively-0.728
 negatively-0.699
Token R
Token ID371
Feature activation-3.65e-01

Pos logit contributions
 BDS+0.087
 NEO+0.077
 positively+0.067
 withd+0.064
ppelin+0.063
Neg logit contributions
-0.086
Page-0.086
Bloom-0.085
mination-0.083
��-0.081
Tokenipe
Token ID3757
Feature activation+5.55e-02

Pos logit contributions
 BDS+1.192
 NEO+1.069
 positively+0.945
 withd+0.901
 negatively+0.899
Neg logit contributions
-0.958
Page-0.941
Bloom-0.934
mination-0.907
��-0.896
Token2016
Token ID5304
Feature activation-2.65e-01

Pos logit contributions
+0.160
Page+0.159
Bloom+0.157
mination+0.151
��+0.148
Neg logit contributions
 BDS-0.259
 NEO-0.235
 positively-0.210
 withd-0.203
 negatively-0.202
Token.
Token ID13
Feature activation-2.99e-01

Pos logit contributions
 BDS+0.857
 NEO+0.769
 positively+0.677
 withd+0.651
 negatively+0.644
Neg logit contributions
-0.702
Page-0.697
Bloom-0.689
mination-0.669
��-0.657
Token01
Token ID486
Feature activation-2.00e-01

Pos logit contributions
 BDS+0.982
 NEO+0.883
 positively+0.773
ppelin+0.737
 negatively+0.735
Neg logit contributions
-0.779
Page-0.765
Bloom-0.759
mination-0.741
��-0.739
Token.
Token ID13
Feature activation-3.28e-01

Pos logit contributions
 BDS+0.656
 NEO+0.590
 positively+0.522
 withd+0.510
 negatively+0.498
Neg logit contributions
-0.520
Page-0.520
Bloom-0.512
mination-0.495
��-0.482
Token29
Token ID1959
Feature activation-2.34e-01

Pos logit contributions
 BDS+1.141
 NEO+1.034
 positively+0.909
ppelin+0.874
 negatively+0.866
Neg logit contributions
-0.786
Page-0.766
Bloom-0.762
��-0.753
mination-0.744
Token"]
Token ID8973
Feature activation+9.25e-01

Pos logit contributions
 BDS+0.618
 NEO+0.541
 positively+0.457
 withd+0.430
 negatively+0.428
Neg logit contributions
-0.758
Page-0.752
Bloom-0.746
mination-0.728
��-0.720
Token [
Token ID685
Feature activation-7.26e-02

Pos logit contributions
Page+2.009
+1.973
Bloom+1.949
mination+1.896
Range+1.828
Neg logit contributions
 BDS-3.417
 NEO-3.060
 positively-2.827
 withd-2.813
 negatively-2.707
TokenRound
Token ID22685
Feature activation+7.19e-01

Pos logit contributions
 BDS+0.307
 NEO+0.281
 positively+0.259
 withd+0.254
ppelin+0.251
Neg logit contributions
Page-0.122
-0.120
Bloom-0.117
Range-0.111
mination-0.110
Token "
Token ID366
Feature activation+1.98e-01

Pos logit contributions
+1.980
Page+1.979
Bloom+1.952
mination+1.891
��+1.861
Neg logit contributions
 BDS-2.232
 NEO-1.989
 positively-1.743
 withd-1.707
 negatively-1.655
Token11
Token ID1157
Feature activation-2.72e-01

Pos logit contributions
+0.454
Page+0.451
Bloom+0.445
mination+0.428
��+0.418
Neg logit contributions
 BDS-0.712
 NEO-0.646
 positively-0.579
 withd-0.560
 negatively-0.555
Token"]
Token ID8973
Feature activation+8.14e-01

Pos logit contributions
 BDS+0.811
 NEO+0.720
 positively+0.624
 withd+0.594
 negatively+0.591
Neg logit contributions
-0.792
Page-0.784
Bloom-0.778
mination-0.756
��-0.746
Tokenuk
Token ID2724
Feature activation+2.51e-02

Pos logit contributions
 BDS+0.120
 NEO+0.107
 withd+0.093
 positively+0.091
 negatively+0.087
Neg logit contributions
Page-0.130
-0.129
Bloom-0.127
mination-0.123
Range-0.120
Token/
Token ID14
Feature activation+3.12e-01

Pos logit contributions
+0.073
Page+0.073
Bloom+0.072
mination+0.070
��+0.068
Neg logit contributions
 BDS-0.075
 NEO-0.066
 positively-0.058
 withd-0.056
 negatively-0.055
Tokenin
Token ID259
Feature activation+9.00e-02

Pos logit contributions
+0.756
Page+0.755
Bloom+0.740
mination+0.721
��+0.700
Neg logit contributions
 BDS-1.078
 NEO-0.975
 positively-0.861
 withd-0.831
ppelin-0.826
Tokencoming
Token ID4976
Feature activation+1.41e-01

Pos logit contributions
Page+0.174
+0.169
Bloom+0.169
mination+0.165
jury+0.158
Neg logit contributions
 BDS-0.355
 NEO-0.329
 positively-0.294
 withd-0.291
 negatively-0.284
Token/
Token ID14
Feature activation+2.25e-01

Pos logit contributions
+0.414
Page+0.411
Bloom+0.406
mination+0.395
��+0.389
Neg logit contributions
 BDS-0.415
 NEO-0.368
 positively-0.319
 withd-0.305
 negatively-0.302
Tokenarticle
Token ID20205
Feature activation+9.13e-01

Pos logit contributions
Page+0.420
+0.417
Bloom+0.405
mination+0.395
��+0.375
Neg logit contributions
 BDS-0.905
 NEO-0.831
 positively-0.742
 withd-0.724
ppelin-0.722
Token35
Token ID2327
Feature activation+2.42e-01

Pos logit contributions
Page+1.587
+1.570
Bloom+1.533
mination+1.448
��+1.382
Neg logit contributions
 BDS-3.781
 NEO-3.494
 positively-3.173
 withd-3.108
 negatively-3.062
Token79
Token ID3720
Feature activation+1.40e-01

Pos logit contributions
+0.570
Page+0.564
Bloom+0.557
mination+0.539
��+0.528
Neg logit contributions
 BDS-0.854
 NEO-0.773
 positively-0.689
 withd-0.665
 negatively-0.660
Token23
Token ID1954
Feature activation+2.18e-01

Pos logit contributions
+0.341
Page+0.340
Bloom+0.335
mination+0.323
��+0.315
Neg logit contributions
 BDS-0.484
 NEO-0.438
 positively-0.390
 withd-0.377
 negatively-0.373
Token12
Token ID1065
Feature activation+2.66e-01

Pos logit contributions
+0.501
Page+0.498
Bloom+0.490
mination+0.473
��+0.458
Neg logit contributions
 BDS-0.784
 NEO-0.713
 positively-0.637
 withd-0.618
 negatively-0.611
Token.
Token ID13
Feature activation+2.09e-02

Pos logit contributions
Page+0.682
+0.680
Bloom+0.666
mination+0.646
Beast+0.626
Neg logit contributions
 BDS-0.879
 NEO-0.793
 positively-0.701
 withd-0.694
 negatively-0.669
Token35
Token ID2327
Feature activation+2.59e-01

Pos logit contributions
 BDS+0.283
 NEO+0.257
 positively+0.229
 withd+0.221
 negatively+0.219
Neg logit contributions
-0.185
Page-0.183
Bloom-0.181
mination-0.175
��-0.172
Token790
Token ID37750
Feature activation+1.38e-01

Pos logit contributions
+0.604
Page+0.601
Bloom+0.591
mination+0.571
��+0.555
Neg logit contributions
 BDS-0.919
 NEO-0.834
 positively-0.743
 withd-0.720
 negatively-0.712
Token347
Token ID30995
Feature activation+3.68e-01

Pos logit contributions
+0.317
Page+0.316
Bloom+0.310
mination+0.300
��+0.291
Neg logit contributions
 BDS-0.495
 NEO-0.449
 positively-0.401
 withd-0.389
 negatively-0.385
Token.
Token ID13
Feature activation+5.09e-01

Pos logit contributions
Page+0.933
+0.930
Bloom+0.904
mination+0.884
��+0.853
Neg logit contributions
 BDS-1.229
 NEO-1.109
 positively-0.979
 withd-0.958
 negatively-0.934
Tokenhtml
Token ID6494
Feature activation+1.05e-01

Pos logit contributions
Page+0.835
+0.797
Bloom+0.760
mination+0.738
aspx+0.722
Neg logit contributions
 BDS-2.186
 NEO-2.022
 withd-1.831
 positively-1.820
 negatively-1.779
Token https
Token ID3740
Feature activation+9.13e-01

Pos logit contributions
+0.236
Page+0.235
Bloom+0.229
mination+0.221
��+0.214
Neg logit contributions
 BDS-0.380
 NEO-0.346
 positively-0.312
 withd-0.304
 negatively-0.299
Token://
Token ID1378
Feature activation+3.90e-01

Pos logit contributions
Page+2.051
+2.020
Bloom+1.991
mination+1.850
Range+1.837
Neg logit contributions
 BDS-3.324
 NEO-3.046
 withd-2.930
 positively-2.720
 negatively-2.642
Tokenwww
Token ID2503
Feature activation+8.34e-02

Pos logit contributions
+0.803
Page+0.799
Bloom+0.785
mination+0.756
��+0.728
Neg logit contributions
 BDS-1.493
 NEO-1.367
 positively-1.223
 withd-1.197
 negatively-1.180
Token.
Token ID13
Feature activation+7.26e-02

Pos logit contributions
+0.215
Page+0.215
Bloom+0.211
mination+0.204
Beast+0.197
Neg logit contributions
 BDS-0.275
 NEO-0.248
 positively-0.220
 withd-0.217
 negatively-0.210
Tokenbel
Token ID6667
Feature activation+2.85e-01

Pos logit contributions
+0.226
Page+0.224
Bloom+0.222
��+0.216
mination+0.215
Neg logit contributions
 BDS-0.200
 NEO-0.176
 positively-0.152
ppelin-0.144
 negatively-0.142
Tokenfast
Token ID7217
Feature activation-1.52e-02

Pos logit contributions
Page+0.674
+0.672
Bloom+0.658
mination+0.640
Beast+0.613
Neg logit contributions
 BDS-1.006
 NEO-0.918
 positively-0.805
 withd-0.800
 negatively-0.775
Token -
Token ID532
Feature activation+6.53e-01

Pos logit contributions
+0.261
Page+0.257
Bloom+0.253
mination+0.242
��+0.236
Neg logit contributions
 BDS-0.625
 NEO-0.575
 positively-0.523
 withd-0.506
 negatively-0.504
Token January
Token ID3269
Feature activation+4.61e-01

Pos logit contributions
Page+2.020
+2.009
Bloom+1.993
mination+1.925
Beast+1.881
Neg logit contributions
 BDS-1.784
 NEO-1.571
 positively-1.356
 withd-1.340
ppelin-1.300
Token 15
Token ID1315
Feature activation+4.10e-01

Pos logit contributions
Page+1.190
+1.183
Bloom+1.166
mination+1.125
Beast+1.093
Neg logit contributions
 BDS-1.506
 NEO-1.357
 positively-1.212
 withd-1.204
 negatively-1.158
Token,
Token ID11
Feature activation+2.94e-01

Pos logit contributions
Page+1.089
+1.074
Bloom+1.065
mination+1.024
Range+0.998
Neg logit contributions
 BDS-1.321
 NEO-1.196
 withd-1.085
 positively-1.054
 negatively-1.011
Token 2010
Token ID3050
Feature activation+6.65e-01

Pos logit contributions
+0.889
Page+0.882
Bloom+0.874
mination+0.852
��+0.839
Neg logit contributions
 BDS-0.841
 NEO-0.743
 positively-0.640
 withd-0.609
 negatively-0.604
Token
Token ID198
Feature activation+9.02e-01

Pos logit contributions
Page+1.934
+1.890
Bloom+1.881
mination+1.816
Beast+1.775
Neg logit contributions
 BDS-1.968
 NEO-1.760
 withd-1.622
 positively-1.551
 negatively-1.480
Token
Token ID198
Feature activation+6.83e-01

Pos logit contributions
Page+2.516
Bloom+2.424
+2.423
mination+2.332
Range+2.300
Neg logit contributions
 BDS-2.755
 NEO-2.503
 withd-2.419
 positively-2.197
 negatively-2.112
TokenSubject
Token ID19776
Feature activation+3.25e-01

Pos logit contributions
Page+1.222
+1.190
Bloom+1.183
mination+1.115
Beast+1.080
Neg logit contributions
 BDS-2.830
 NEO-2.609
 withd-2.376
 positively-2.363
 negatively-2.288
Token:
Token ID25
Feature activation+1.13e-02

Pos logit contributions
Page+0.874
+0.855
Bloom+0.843
mination+0.817
Range+0.796
Neg logit contributions
 BDS-1.045
 NEO-0.942
 withd-0.861
 positively-0.825
 negatively-0.793
Token Que
Token ID4670
Feature activation-1.63e-01

Pos logit contributions
+0.033
Page+0.033
Bloom+0.032
mination+0.031
��+0.031
Neg logit contributions
 BDS-0.033
 NEO-0.030
 positively-0.025
 withd-0.025
ppelin-0.024
Tokener
Token ID263
Feature activation-1.57e-01

Pos logit contributions
 BDS+0.622
 NEO+0.566
 positively+0.511
 negatively+0.490
 withd+0.490
Neg logit contributions
-0.341
Page-0.332
Bloom-0.330
mination-0.316
��-0.311
Token -
Token ID532
Feature activation+6.58e-01

Pos logit contributions
+0.356
Page+0.351
Bloom+0.345
mination+0.329
��+0.320
Neg logit contributions
 BDS-0.880
 NEO-0.810
 positively-0.737
 withd-0.714
 negatively-0.711
Token December
Token ID3426
Feature activation+5.15e-01

Pos logit contributions
Page+2.044
+2.041
Bloom+2.010
mination+1.944
Beast+1.899
Neg logit contributions
 BDS-1.777
 NEO-1.560
 withd-1.350
 positively-1.347
ppelin-1.296
Token 5
Token ID642
Feature activation+5.20e-01

Pos logit contributions
Page+1.315
+1.308
Bloom+1.288
mination+1.243
Beast+1.206
Neg logit contributions
 BDS-1.688
 NEO-1.525
 positively-1.362
 withd-1.360
 negatively-1.302
Token,
Token ID11
Feature activation+3.25e-01

Pos logit contributions
Page+1.375
+1.356
Bloom+1.341
mination+1.297
Range+1.259
Neg logit contributions
 BDS-1.676
 NEO-1.517
 withd-1.368
 positively-1.335
 negatively-1.279
Token 2009
Token ID3717
Feature activation+6.59e-01

Pos logit contributions
+0.985
Page+0.981
Bloom+0.968
mination+0.946
��+0.926
Neg logit contributions
 BDS-0.923
 NEO-0.816
 positively-0.702
 withd-0.673
ppelin-0.663
Token
Token ID198
Feature activation+8.98e-01

Pos logit contributions
Page+1.922
+1.884
Bloom+1.873
mination+1.812
Beast+1.770
Neg logit contributions
 BDS-1.943
 NEO-1.732
 withd-1.585
 positively-1.519
 negatively-1.448
Token
Token ID198
Feature activation+6.89e-01

Pos logit contributions
Page+2.503
+2.413
Bloom+2.411
mination+2.322
Range+2.286
Neg logit contributions
 BDS-2.744
 NEO-2.494
 withd-2.412
 positively-2.188
 negatively-2.103
TokenSubject
Token ID19776
Feature activation+2.37e-01

Pos logit contributions
Page+1.246
+1.220
Bloom+1.207
mination+1.141
Beast+1.103
Neg logit contributions
 BDS-2.833
 NEO-2.610
 withd-2.375
 positively-2.364
 negatively-2.286
Token:
Token ID25
Feature activation+2.25e-02

Pos logit contributions
Page+0.644
+0.633
Bloom+0.624
mination+0.605
Range+0.588
Neg logit contributions
 BDS-0.756
 NEO-0.680
 withd-0.611
 positively-0.593
 negatively-0.569
Token Que
Token ID4670
Feature activation-3.00e-01

Pos logit contributions
+0.066
Page+0.066
Bloom+0.065
mination+0.063
��+0.062
Neg logit contributions
 BDS-0.066
 NEO-0.059
 positively-0.051
 withd-0.049
ppelin-0.048
Tokeners
Token ID364
Feature activation-8.48e-02

Pos logit contributions
 BDS+1.158
 NEO+1.054
 positively+0.955
 negatively+0.917
ppelin+0.916
Neg logit contributions
-0.616
Bloom-0.600
Page-0.597
mination-0.573
��-0.559
Token to
Token ID284
Feature activation-1.06e-01

Pos logit contributions
 BDS+0.715
 NEO+0.642
 positively+0.567
 withd+0.543
 negatively+0.540
Neg logit contributions
-0.569
Page-0.563
Bloom-0.558
mination-0.541
��-0.532
Token have
Token ID423
Feature activation-8.79e-02

Pos logit contributions
 BDS+0.303
 NEO+0.269
 positively+0.228
 withd+0.220
ppelin+0.219
Neg logit contributions
-0.316
Page-0.314
Bloom-0.310
mination-0.302
��-0.297
Token a
Token ID257
Feature activation-1.07e-01

Pos logit contributions
 BDS+0.271
 NEO+0.242
 positively+0.211
 withd+0.202
 negatively+0.200
Neg logit contributions
-0.246
Page-0.244
Bloom-0.241
mination-0.234
��-0.231
Token none
Token ID4844
Feature activation-3.51e-01

Pos logit contributions
 BDS+0.307
 NEO+0.271
 positively+0.234
 withd+0.223
 negatively+0.221
Neg logit contributions
-0.321
Page-0.319
Bloom-0.316
mination-0.308
��-0.303
Token-
Token ID12
Feature activation+3.53e-01

Pos logit contributions
 BDS+1.191
 NEO+1.074
 positively+0.958
 negatively+0.916
 withd+0.913
Neg logit contributions
-0.874
Page-0.862
Bloom-0.859
mination-0.827
��-0.822
Tokentoo
Token ID18820
Feature activation+8.97e-01

Pos logit contributions
+0.903
Page+0.898
Bloom+0.889
mination+0.861
��+0.843
Neg logit contributions
 BDS-1.176
 NEO-1.058
 positively-0.933
 withd-0.898
 negatively-0.891
Token-
Token ID12
Feature activation+4.88e-02

Pos logit contributions
+2.130
Bloom+2.106
Page+2.104
mination+2.009
��+1.954
Neg logit contributions
 BDS-3.090
 NEO-2.808
 positively-2.463
 withd-2.426
ppelin-2.411
Tokensub
Token ID7266
Feature activation+5.60e-01

Pos logit contributions
+0.134
Page+0.133
Bloom+0.131
mination+0.128
��+0.127
Neg logit contributions
 BDS-0.153
 NEO-0.136
 positively-0.120
 withd-0.114
 negatively-0.114
Tokentle
Token ID7100
Feature activation-1.04e-01

Pos logit contributions
+1.035
Page+1.021
Bloom+1.006
mination+0.966
��+0.937
Neg logit contributions
 BDS-2.259
 NEO-2.072
 positively-1.871
 withd-1.819
 negatively-1.805
Token dig
Token ID3100
Feature activation-4.34e-01

Pos logit contributions
 BDS+0.300
 NEO+0.265
 positively+0.229
 withd+0.218
 negatively+0.216
Neg logit contributions
-0.314
Page-0.311
Bloom-0.308
mination-0.300
��-0.296
Token at
Token ID379
Feature activation-7.14e-02

Pos logit contributions
 BDS+1.341
 NEO+1.193
 positively+1.049
 withd+0.996
 negatively+0.995
Neg logit contributions
-1.213
Page-1.204
Bloom-1.194
mination-1.154
��-1.148
Token
Token ID198
Feature activation+2.10e-01

Pos logit contributions
Page+1.075
+1.059
Bloom+1.050
mination+1.013
Range+0.989
Neg logit contributions
 BDS-1.084
 NEO-0.974
 withd-0.881
 positively-0.846
 negatively-0.807
Tokenhttps
Token ID5450
Feature activation+4.82e-01

Pos logit contributions
+0.447
Page+0.439
Bloom+0.434
mination+0.422
��+0.415
Neg logit contributions
 BDS-0.786
 NEO-0.715
 positively-0.641
 withd-0.616
 negatively-0.615
Token://
Token ID1378
Feature activation+4.43e-01

Pos logit contributions
Page+1.149
+1.136
Bloom+1.116
mination+1.060
Range+1.038
Neg logit contributions
 BDS-1.705
 NEO-1.546
 withd-1.410
 positively-1.382
 negatively-1.329
Tokenwww
Token ID2503
Feature activation+2.20e-01

Pos logit contributions
+0.947
Page+0.939
Bloom+0.926
mination+0.892
��+0.868
Neg logit contributions
 BDS-1.660
 NEO-1.514
 positively-1.357
 withd-1.315
 negatively-1.305
Token.
Token ID13
Feature activation+2.52e-01

Pos logit contributions
Page+0.563
+0.561
Bloom+0.552
mination+0.533
Beast+0.515
Neg logit contributions
 BDS-0.730
 NEO-0.659
 positively-0.585
 withd-0.583
 negatively-0.559
Tokeneff
Token ID14822
Feature activation+8.93e-01

Pos logit contributions
+0.589
Page+0.582
Bloom+0.575
mination+0.555
��+0.548
Neg logit contributions
 BDS-0.895
 NEO-0.810
 positively-0.723
 withd-0.694
 negatively-0.692
Token.
Token ID13
Feature activation+3.12e-03

Pos logit contributions
Page+2.177
+2.135
Bloom+2.122
mination+2.056
mile+1.984
Neg logit contributions
 BDS-3.041
 NEO-2.766
 withd-2.534
 positively-2.467
 negatively-2.372
Tokenorg
Token ID2398
Feature activation-1.03e-02

Pos logit contributions
+0.007
Page+0.007
Bloom+0.007
mination+0.007
��+0.006
Neg logit contributions
 BDS-0.011
 NEO-0.010
 positively-0.009
 withd-0.009
 negatively-0.009
Token/
Token ID14
Feature activation+2.33e-01

Pos logit contributions
 BDS+0.031
 NEO+0.027
 positively+0.024
 withd+0.023
 negatively+0.023
Neg logit contributions
-0.030
Page-0.030
Bloom-0.030
mination-0.029
��-0.028
Tokend
Token ID67
Feature activation-6.63e-02

Pos logit contributions
+0.536
Page+0.529
Bloom+0.524
mination+0.507
��+0.498
Neg logit contributions
 BDS-0.834
 NEO-0.756
 positively-0.674
 withd-0.649
 negatively-0.645
Tokenee
Token ID1453
Feature activation+4.27e-01

Pos logit contributions
 BDS+0.262
 NEO+0.240
 positively+0.217
 withd+0.210
 negatively+0.209
Neg logit contributions
-0.128
Page-0.127
Bloom-0.125
mination-0.120
��-0.117
Tokenv
Token ID85
Feature activation+3.16e-02

Pos logit contributions
 BDS+0.249
 NEO+0.228
 positively+0.208
ppelin+0.201
 negatively+0.201
Neg logit contributions
-0.105
Page-0.102
Bloom-0.101
mination-0.096
��-0.096
Token2
Token ID17
Feature activation+4.39e-02

Pos logit contributions
Page+0.073
+0.073
Bloom+0.072
mination+0.069
Range+0.067
Neg logit contributions
 BDS-0.113
 NEO-0.103
 positively-0.092
 withd-0.091
 negatively-0.088
Token.
Token ID13
Feature activation+3.88e-02

Pos logit contributions
Page+0.114
+0.114
Bloom+0.112
mination+0.108
Range+0.105
Neg logit contributions
 BDS-0.144
 NEO-0.130
 positively-0.115
 withd-0.113
 negatively-0.110
Tokenmorph
Token ID24503
Feature activation+2.00e-01

Pos logit contributions
+0.099
Page+0.098
Bloom+0.097
mination+0.094
��+0.092
Neg logit contributions
 BDS-0.130
 NEO-0.117
 positively-0.103
 withd-0.099
 negatively-0.098
Tokenology
Token ID1435
Feature activation+9.53e-02

Pos logit contributions
+0.503
Page+0.499
Bloom+0.494
mination+0.477
��+0.469
Neg logit contributions
 BDS-0.675
 NEO-0.609
 positively-0.539
 withd-0.518
 negatively-0.514
TokenEx
Token ID3109
Feature activation+8.88e-01

Pos logit contributions
+0.234
Page+0.231
Bloom+0.229
mination+0.221
��+0.218
Neg logit contributions
 BDS-0.328
 NEO-0.296
 positively-0.262
 withd-0.252
 negatively-0.251
Token(
Token ID7
Feature activation-8.51e-02

Pos logit contributions
+1.935
Page+1.918
Bloom+1.897
mination+1.833
Range+1.766
Neg logit contributions
 BDS-3.277
 NEO-3.002
 positively-2.688
 withd-2.603
 negatively-2.578
Tokengray
Token ID44605
Feature activation+2.77e-01

Pos logit contributions
 BDS+0.293
 NEO+0.265
 positively+0.235
 withd+0.226
 negatively+0.224
Neg logit contributions
-0.208
Page-0.206
Bloom-0.204
mination-0.197
��-0.194
Token,
Token ID11
Feature activation+7.72e-02

Pos logit contributions
+0.745
Page+0.739
Bloom+0.731
mination+0.710
��+0.699
Neg logit contributions
 BDS-0.886
 NEO-0.794
 positively-0.697
 withd-0.668
 negatively-0.663
Token c
Token ID269
Feature activation-1.08e-02

Pos logit contributions
+0.220
Page+0.218
Bloom+0.216
mination+0.210
��+0.207
Neg logit contributions
 BDS-0.235
 NEO-0.209
 positively-0.182
 withd-0.174
 negatively-0.172
Tokenv
Token ID85
Feature activation+1.43e-01

Pos logit contributions
 BDS+0.045
 NEO+0.041
 positively+0.037
ppelin+0.036
 negatively+0.036
Neg logit contributions
-0.019
Page-0.018
Bloom-0.018
mination-0.017
��-0.017
Token
Token ID784
Feature activation+6.64e-02

Pos logit contributions
+0.210
Page+0.209
Bloom+0.206
mination+0.201
��+0.197
Neg logit contributions
 BDS-0.247
 NEO-0.221
 positively-0.193
 withd-0.186
ppelin-0.184
Token but
Token ID475
Feature activation+2.22e-01

Pos logit contributions
+0.189
Page+0.187
Bloom+0.185
mination+0.180
��+0.177
Neg logit contributions
 BDS-0.202
 NEO-0.180
 positively-0.156
 withd-0.150
ppelin-0.149
Token maybe
Token ID3863
Feature activation+1.58e-01

Pos logit contributions
Page+0.583
+0.582
Bloom+0.573
mination+0.551
��+0.544
Neg logit contributions
 BDS-0.713
 NEO-0.639
 positively-0.557
 withd-0.543
ppelin-0.538
Token Kris
Token ID27748
Feature activation-1.12e-02

Pos logit contributions
+0.403
Page+0.401
Bloom+0.395
mination+0.380
��+0.374
Neg logit contributions
 BDS-0.525
 NEO-0.472
 positively-0.415
 withd-0.402
ppelin-0.400
Token Let
Token ID3914
Feature activation+4.73e-02

Pos logit contributions
 BDS+0.032
 NEO+0.028
 positively+0.024
 withd+0.023
 negatively+0.023
Neg logit contributions
-0.034
Page-0.034
Bloom-0.034
mination-0.033
��-0.032
Tokenang
Token ID648
Feature activation+8.87e-01

Pos logit contributions
+0.086
Page+0.083
Bloom+0.082
��+0.079
mination+0.079
Neg logit contributions
 BDS-0.193
 NEO-0.177
 positively-0.161
ppelin-0.155
odic-0.155
Token also
Token ID635
Feature activation+1.38e-04

Pos logit contributions
Page+2.083
+2.078
Bloom+2.016
mination+1.967
Beast+1.901
Neg logit contributions
 BDS-3.093
 NEO-2.793
 positively-2.419
ppelin-2.392
 withd-2.389
Token lacks
Token ID16523
Feature activation-2.48e-01

Pos logit contributions
+0.000
Page+0.000
Bloom+0.000
mination+0.000
��+0.000
Neg logit contributions
 BDS-0.000
 NEO-0.000
 positively-0.000
 withd-0.000
ppelin-0.000
Token character
Token ID2095
Feature activation-3.70e-02

Pos logit contributions
 BDS+0.684
 NEO+0.601
 positively+0.514
 withd+0.489
 negatively+0.484
Neg logit contributions
-0.775
Page-0.769
Bloom-0.762
mination-0.742
��-0.732
Token?
Token ID30
Feature activation-1.27e-01

Pos logit contributions
 BDS+0.119
 NEO+0.106
 positively+0.093
 withd+0.090
ppelin+0.089
Neg logit contributions
-0.099
Page-0.098
Bloom-0.097
mination-0.094
��-0.092
Token I
Token ID314
Feature activation-5.28e-03

Pos logit contributions
 BDS+0.375
 NEO+0.333
 positively+0.290
 withd+0.279
ppelin+0.275
Neg logit contributions
-0.370
Page-0.368
Bloom-0.363
mination-0.353
��-0.346
Token presidential
Token ID4787
Feature activation+3.78e-01

Pos logit contributions
 BDS+0.298
 NEO+0.269
 positively+0.241
 withd+0.233
ppelin+0.232
Neg logit contributions
-0.194
Page-0.193
Bloom-0.190
mination-0.183
��-0.179
Token nominee
Token ID10429
Feature activation+3.63e-01

Pos logit contributions
Page+1.002
+0.989
Bloom+0.975
mination+0.949
��+0.931
Neg logit contributions
 BDS-1.207
 NEO-1.076
 positively-0.950
ppelin-0.923
 withd-0.918
Token Hillary
Token ID5073
Feature activation+1.59e-01

Pos logit contributions
Page+0.813
+0.802
Bloom+0.789
mination+0.775
Reviewed+0.738
Neg logit contributions
 BDS-1.291
 NEO-1.174
 positively-1.051
ppelin-1.031
 withd-1.029
Token Clinton
Token ID2605
Feature activation+2.02e-01

Pos logit contributions
+0.356
Page+0.355
Bloom+0.349
mination+0.337
��+0.327
Neg logit contributions
 BDS-0.574
 NEO-0.523
 positively-0.465
 withd-0.454
ppelin-0.448
Token Hillary
Token ID5073
Feature activation+1.19e-01

Pos logit contributions
Page+0.486
+0.484
Bloom+0.476
mination+0.461
��+0.449
Neg logit contributions
 BDS-0.697
 NEO-0.630
 positively-0.555
ppelin-0.538
 withd-0.537
Token Diane
Token ID26542
Feature activation+8.85e-01

Pos logit contributions
+0.295
Page+0.291
Bloom+0.289
mination+0.280
��+0.277
Neg logit contributions
 BDS-0.405
 NEO-0.366
 positively-0.325
 negatively-0.310
ppelin-0.309
Token Rodham
Token ID41848
Feature activation+6.48e-01

Pos logit contributions
Page+0.602
Bloom+0.582
+0.575
mination+0.549
Range+0.449
Neg logit contributions
 BDS-4.522
 NEO-4.285
 positively-3.956
 withd-3.956
ppelin-3.879
Token Clinton
Token ID2605
Feature activation+4.73e-02

Pos logit contributions
Page+1.447
+1.445
Bloom+1.414
mination+1.370
��+1.319
Neg logit contributions
 BDS-2.353
 NEO-2.143
 positively-1.914
 withd-1.883
ppelin-1.834
TokenSanders
Token ID26747
Feature activation+4.91e-01

Pos logit contributions
+0.100
Page+0.099
Bloom+0.098
mination+0.094
��+0.092
Neg logit contributions
 BDS-0.179
 NEO-0.163
 positively-0.146
 withd-0.142
 negatively-0.141
Token:
Token ID25
Feature activation+2.71e-03

Pos logit contributions
Page+1.245
Bloom+1.223
+1.219
Charge+1.142
Journal+1.134
Neg logit contributions
 BDS-1.638
 NEO-1.464
 positively-1.283
 withd-1.282
ppelin-1.268
Token '
Token ID705
Feature activation-4.61e-02

Pos logit contributions
+0.008
Page+0.008
Bloom+0.008
mination+0.008
��+0.008
Neg logit contributions
 BDS-0.007
 NEO-0.007
 positively-0.006
 withd-0.006
 negatively-0.005
Token Tata
Token ID49963
Feature activation+2.44e-01

Pos logit contributions
 BDS+0.170
 NEO+0.151
 positively+0.131
 withd+0.124
 negatively+0.124
Neg logit contributions
-0.165
Page-0.163
Bloom-0.162
mination-0.158
��-0.155
Token Steel
Token ID7851
Feature activation+4.86e-01

Pos logit contributions
+0.689
Page+0.686
Bloom+0.678
mination+0.660
��+0.651
Neg logit contributions
 BDS-0.743
 NEO-0.661
 positively-0.581
 withd-0.559
 negatively-0.552
Token Chess
Token ID25774
Feature activation+5.07e-01

Pos logit contributions
+1.331
Page+1.323
Bloom+1.304
mination+1.272
��+1.251
Neg logit contributions
 BDS-1.520
 NEO-1.357
 positively-1.195
 withd-1.153
 negatively-1.138
Token Chall
Token ID10772
Feature activation+2.56e-01

Pos logit contributions
+1.262
Page+1.259
Bloom+1.236
mination+1.201
��+1.176
Neg logit contributions
 BDS-1.713
 NEO-1.542
 positively-1.374
 withd-1.332
 negatively-1.315
Tokenengers
Token ID9302
Feature activation+7.48e-01

Pos logit contributions
Page+0.519
+0.515
Bloom+0.506
mination+0.488
��+0.469
Neg logit contributions
 BDS-0.987
 NEO-0.905
 positively-0.813
 withd-0.795
 negatively-0.784
Token"]
Token ID8973
Feature activation+8.81e-01

Pos logit contributions
Page+1.864
+1.841
Bloom+1.814
mination+1.752
��+1.721
Neg logit contributions
 BDS-2.484
 NEO-2.235
 positively-2.018
 withd-1.972
 negatively-1.930
Token [
Token ID685
Feature activation+7.64e-02

Pos logit contributions
Page+1.938
+1.916
Bloom+1.876
mination+1.840
Range+1.757
Neg logit contributions
 BDS-3.208
 NEO-2.894
 positively-2.634
 withd-2.595
ppelin-2.520
TokenSite
Token ID29123
Feature activation-5.27e-03

Pos logit contributions
Page+0.154
+0.153
Bloom+0.149
mination+0.143
��+0.140
Neg logit contributions
 BDS-0.297
 NEO-0.271
 positively-0.245
 withd-0.238
 negatively-0.235
Token "
Token ID366
Feature activation+1.47e-01

Pos logit contributions
 BDS+0.015
 NEO+0.014
 positively+0.012
 negatively+0.011
 withd+0.011
Neg logit contributions
-0.016
Page-0.015
Bloom-0.015
mination-0.015
��-0.015
TokenW
Token ID54
Feature activation+1.23e-01

Pos logit contributions
+0.353
Page+0.351
Bloom+0.347
mination+0.334
��+0.328
Neg logit contributions
 BDS-0.511
 NEO-0.462
 positively-0.412
 withd-0.397
 negatively-0.394
Tokenijk
Token ID45961
Feature activation+1.46e-01

Pos logit contributions
+0.318
Page+0.316
Bloom+0.312
mination+0.303
��+0.298
Neg logit contributions
 BDS-0.407
 NEO-0.366
 positively-0.323
 withd-0.310
 negatively-0.308
Tokenk
Token ID74
Feature activation-9.34e-02

Pos logit contributions
 BDS+0.792
 NEO+0.714
 positively+0.631
 withd+0.605
 negatively+0.602
Neg logit contributions
-0.603
Page-0.595
Bloom-0.591
mination-0.572
��-0.563
Token [
Token ID685
Feature activation+1.84e-01

Pos logit contributions
 BDS+0.310
 NEO+0.279
 positively+0.247
 withd+0.237
 negatively+0.235
Neg logit contributions
-0.240
Page-0.237
Bloom-0.235
mination-0.228
��-0.225
Token10
Token ID940
Feature activation+3.25e-02

Pos logit contributions
Page+0.464
+0.460
Bloom+0.455
mination+0.438
Beast+0.426
Neg logit contributions
 BDS-0.624
 NEO-0.563
 positively-0.498
 withd-0.487
 negatively-0.476
Token:
Token ID25
Feature activation+4.10e-02

Pos logit contributions
Page+0.090
+0.090
Bloom+0.088
mination+0.086
Range+0.083
Neg logit contributions
 BDS-0.101
 NEO-0.090
 withd-0.079
 positively-0.079
 negatively-0.075
Token47
Token ID2857
Feature activation+3.76e-01

Pos logit contributions
+0.110
Page+0.110
Bloom+0.108
mination+0.105
��+0.103
Neg logit contributions
 BDS-0.131
 NEO-0.117
 positively-0.103
 withd-0.099
 negatively-0.098
Token]
Token ID60
Feature activation+8.76e-01

Pos logit contributions
Page+0.823
+0.808
Bloom+0.800
mination+0.761
Range+0.737
Neg logit contributions
 BDS-1.389
 NEO-1.262
 withd-1.166
 positively-1.142
 negatively-1.099
Token <
Token ID1279
Feature activation+7.34e-01

Pos logit contributions
+1.591
Page+1.579
Bloom+1.547
mination+1.489
��+1.445
Neg logit contributions
 BDS-3.526
 NEO-3.233
 positively-2.932
 withd-2.850
ppelin-2.841
Tokenqu
Token ID421
Feature activation+4.28e-01

Pos logit contributions
Page+1.821
+1.810
Bloom+1.794
mination+1.731
Beast+1.689
Neg logit contributions
 BDS-2.498
 NEO-2.250
 positively-1.994
 withd-1.938
 negatively-1.916
Tokenij
Token ID2926
Feature activation+1.54e-01

Pos logit contributions
Page+1.100
+1.093
Bloom+1.081
mination+1.051
��+1.023
Neg logit contributions
 BDS-1.422
 NEO-1.278
 positively-1.124
 withd-1.087
 negatively-1.076
Tokenii
Token ID4178
Feature activation+1.07e-01

Pos logit contributions
+0.383
Page+0.382
Bloom+0.376
mination+0.369
��+0.355
Neg logit contributions
 BDS-0.526
 NEO-0.472
 positively-0.418
 withd-0.406
 negatively-0.400
Token>
Token ID29
Feature activation-5.15e-03

Pos logit contributions
+0.207
Page+0.204
Bloom+0.201
mination+0.193
��+0.189
Neg logit contributions
 BDS-0.424
 NEO-0.388
 positively-0.351
 withd-0.339
 negatively-0.337
Token against
Token ID1028
Feature activation-3.03e-01

Pos logit contributions
+0.623
Page+0.619
Bloom+0.611
mination+0.594
��+0.583
Neg logit contributions
 BDS-0.763
 NEO-0.684
 positively-0.600
 withd-0.577
 negatively-0.572
Token 41
Token ID6073
Feature activation+1.95e-01

Pos logit contributions
 BDS+0.952
 NEO+0.850
 positively+0.745
 withd+0.712
 negatively+0.708
Neg logit contributions
-0.833
Page-0.825
Bloom-0.817
mination-0.795
��-0.782
Token,
Token ID11
Feature activation+6.72e-02

Pos logit contributions
+0.540
Page+0.537
Bloom+0.529
mination+0.516
��+0.506
Neg logit contributions
 BDS-0.606
 NEO-0.541
 positively-0.471
 withd-0.453
ppelin-0.448
Token 88
Token ID9193
Feature activation+1.69e-01

Pos logit contributions
+0.177
Page+0.176
Bloom+0.174
mination+0.169
��+0.166
Neg logit contributions
 BDS-0.218
 NEO-0.196
 positively-0.172
 withd-0.165
 negatively-0.164
Token11
Token ID1157
Feature activation+7.76e-02

Pos logit contributions
+0.466
Page+0.464
Bloom+0.457
mination+0.446
��+0.438
Neg logit contributions
 BDS-0.528
 NEO-0.472
 positively-0.410
 withd-0.395
ppelin-0.390
Token tanks
Token ID11657
Feature activation+8.70e-01

Pos logit contributions
+0.214
Page+0.212
Bloom+0.210
mination+0.204
��+0.200
Neg logit contributions
 BDS-0.243
 NEO-0.217
 positively-0.190
 withd-0.182
 negatively-0.181
Token against
Token ID1028
Feature activation-1.52e-01

Pos logit contributions
Page+2.198
+2.175
Bloom+2.148
mination+2.090
��+2.037
Neg logit contributions
 BDS-2.875
 NEO-2.603
 positively-2.247
 withd-2.196
ppelin-2.159
Token 35
Token ID3439
Feature activation+1.42e-01

Pos logit contributions
 BDS+0.486
 NEO+0.435
 positively+0.382
 withd+0.367
 negatively+0.363
Neg logit contributions
-0.410
Page-0.407
Bloom-0.402
mination-0.390
��-0.383
Token12
Token ID1065
Feature activation+8.60e-02

Pos logit contributions
+0.375
Page+0.372
Bloom+0.368
mination+0.357
��+0.351
Neg logit contributions
 BDS-0.458
 NEO-0.411
 positively-0.360
 withd-0.346
 negatively-0.343
Token),
Token ID828
Feature activation+1.82e-01

Pos logit contributions
+0.244
Page+0.242
Bloom+0.240
mination+0.233
��+0.230
Neg logit contributions
 BDS-0.261
 NEO-0.232
 positively-0.201
 withd-0.193
 negatively-0.191
Token with
Token ID351
Feature activation+8.12e-02

Pos logit contributions
Page+0.519
+0.518
Bloom+0.511
mination+0.497
��+0.488
Neg logit contributions
 BDS-0.547
 NEO-0.486
 positively-0.420
 withd-0.403
ppelin-0.400
Token SEC
Token ID10729
Feature activation-8.27e-02

Pos logit contributions
 BDS+0.129
 NEO+0.114
 positively+0.098
 withd+0.094
 negatively+0.093
Neg logit contributions
-0.137
Page-0.136
Bloom-0.135
mination-0.132
��-0.130
TokenOND
Token ID18672
Feature activation-3.58e-01

Pos logit contributions
 BDS+0.313
 NEO+0.286
 positively+0.255
 withd+0.246
 negatively+0.245
Neg logit contributions
-0.174
Page-0.172
Bloom-0.170
mination-0.163
��-0.161
Token PART
Token ID16652
Feature activation-4.19e-01

Pos logit contributions
 BDS+1.035
 NEO+0.916
 positively+0.789
 withd+0.750
 negatively+0.745
Neg logit contributions
-1.076
Page-1.067
Bloom-1.059
mination-1.031
��-1.016
Token RE
Token ID4526
Feature activation-2.69e-02

Pos logit contributions
 BDS+1.002
 NEO+0.862
 positively+0.715
 withd+0.669
 negatively+0.663
Neg logit contributions
-1.464
Page-1.453
Bloom-1.443
mination-1.411
��-1.394
TokenQU
Token ID10917
Feature activation-4.01e-01

Pos logit contributions
 BDS+0.094
 NEO+0.085
 positively+0.076
 withd+0.074
 negatively+0.073
Neg logit contributions
-0.064
Page-0.063
Bloom-0.063
mination-0.060
��-0.059
TokenIR
Token ID4663
Feature activation-1.02e+00

Pos logit contributions
 BDS+1.484
 NEO+1.352
 positively+1.212
 withd+1.164
ppelin+1.159
Neg logit contributions
-0.884
Bloom-0.861
Page-0.851
mination-0.818
��-0.808
TokenES
Token ID1546
Feature activation-2.94e-01

Pos logit contributions
 BDS+2.629
 NEO+2.286
 positively+1.922
 negatively+1.785
 TED+1.768
Neg logit contributions
-3.387
Page-3.359
Bloom-3.338
mination-3.251
��-3.237
Token A
Token ID317
Feature activation-5.55e-01

Pos logit contributions
 BDS+0.868
 NEO+0.770
 positively+0.667
 withd+0.636
 negatively+0.631
Neg logit contributions
-0.865
Page-0.858
Bloom-0.850
mination-0.828
��-0.815
Token BIT
Token ID36992
Feature activation-8.31e-02

Pos logit contributions
 BDS+1.536
 NEO+1.351
 positively+1.150
 withd+1.077
 negatively+1.076
Neg logit contributions
-1.738
Page-1.717
Bloom-1.709
mination-1.668
��-1.646
Token MORE
Token ID12011
Feature activation-4.00e-01

Pos logit contributions
 BDS+0.235
 NEO+0.208
 positively+0.180
 withd+0.173
ppelin+0.170
Neg logit contributions
-0.252
Page-0.250
Bloom-0.248
mination-0.241
��-0.237
Token EFF
Token ID33659
Feature activation+1.35e-01

Pos logit contributions
 BDS+1.095
 NEO+0.961
 positively+0.821
 withd+0.779
 negatively+0.772
Neg logit contributions
-1.260
Page-1.251
Bloom-1.240
mination-1.210
��-1.193
Token AND
Token ID5357
Feature activation-5.62e-02

Pos logit contributions
+0.107
Page+0.107
Bloom+0.105
mination+0.102
��+0.100
Neg logit contributions
 BDS-0.116
 NEO-0.104
 positively-0.090
 withd-0.087
 negatively-0.086
Token STEP
Token ID49154
Feature activation-2.34e-01

Pos logit contributions
 BDS+0.158
 NEO+0.140
 positively+0.121
 withd+0.116
 negatively+0.115
Neg logit contributions
-0.171
Page-0.170
Bloom-0.168
mination-0.164
��-0.161
TokenH
Token ID39
Feature activation-3.45e-02

Pos logit contributions
 BDS+0.704
 NEO+0.626
 positively+0.545
 withd+0.520
 negatively+0.516
Neg logit contributions
-0.673
Page-0.667
Bloom-0.661
mination-0.643
��-0.633
TokenEN
Token ID1677
Feature activation-1.88e-01

Pos logit contributions
 BDS+0.116
 NEO+0.105
 positively+0.094
 withd+0.092
ppelin+0.090
Neg logit contributions
-0.086
Page-0.085
Bloom-0.085
mination-0.081
Beast-0.079
Token M
Token ID337
Feature activation-4.29e-01

Pos logit contributions
 BDS+0.540
 NEO+0.477
 positively+0.413
 withd+0.394
 negatively+0.390
Neg logit contributions
-0.563
Page-0.559
Bloom-0.554
mination-0.540
��-0.531
Tokenatur
Token ID2541
Feature activation-9.37e-01

Pos logit contributions
 BDS+1.488
 NEO+1.345
 positively+1.190
 negatively+1.138
 withd+1.136
Neg logit contributions
-1.042
Page-1.024
Bloom-1.019
mination-0.983
��-0.973
Tokenin
Token ID259
Feature activation-1.82e-01

Pos logit contributions
 BDS+2.906
 NEO+2.594
 positively+2.303
ppelin+2.180
 negatively+2.179
Neg logit contributions
-2.579
Bloom-2.518
Page-2.513
��-2.461
mination-2.432
Token first
Token ID717
Feature activation-6.37e-02

Pos logit contributions
 BDS+0.570
 NEO+0.511
 positively+0.445
 withd+0.428
 negatively+0.425
Neg logit contributions
-0.494
Page-0.494
Bloom-0.487
mination-0.472
��-0.464
Token meet
Token ID1826
Feature activation-1.74e-02

Pos logit contributions
 BDS+0.188
 NEO+0.168
 positively+0.144
 withd+0.138
ppelin+0.137
Neg logit contributions
Page-0.185
-0.184
Bloom-0.182
mination-0.177
��-0.174
Token in
Token ID287
Feature activation-4.15e-02

Pos logit contributions
 BDS+0.053
 NEO+0.047
 positively+0.041
 withd+0.039
 negatively+0.039
Neg logit contributions
-0.049
Page-0.049
Bloom-0.048
mination-0.047
��-0.046
Token April
Token ID3035
Feature activation+1.06e-01

Pos logit contributions
 BDS+0.123
 NEO+0.110
 positively+0.095
 withd+0.092
 negatively+0.090
Neg logit contributions
-0.120
Page-0.120
Bloom-0.119
mination-0.115
��-0.113
Token.
Token ID13
Feature activation+1.17e-01

Pos logit contributions
 BDS+0.164
 NEO+0.146
 positively+0.127
ppelin+0.122
 withd+0.122
Neg logit contributions
Page-0.150
-0.150
Bloom-0.147
mination-0.144
��-0.141
Token
Token ID198
Feature activation+9.50e-02

Pos logit contributions
Page+0.335
+0.335
Bloom+0.329
mination+0.321
Range+0.312
Neg logit contributions
 BDS-0.351
 NEO-0.312
 positively-0.274
 withd-0.265
ppelin-0.261
Token
Token ID198
Feature activation+9.42e-02

Pos logit contributions
Page+0.287
+0.286
Bloom+0.283
mination+0.274
Beast+0.267
Neg logit contributions
 BDS-0.271
 NEO-0.241
 positively-0.209
 withd-0.208
 negatively-0.198
TokenMaterials
Token ID41657
Feature activation-6.41e-01

Pos logit contributions
Page+0.231
+0.230
Bloom+0.228
mination+0.219
��+0.214
Neg logit contributions
 BDS-0.324
 NEO-0.293
 positively-0.260
 withd-0.252
 negatively-0.249
Token and
Token ID290
Feature activation-4.74e-02

Pos logit contributions
 BDS+1.951
 NEO+1.736
 positively+1.511
 withd+1.441
 negatively+1.433
Neg logit contributions
-1.826
Page-1.808
Bloom-1.793
mination-1.744
��-1.719
Token Methods
Token ID25458
Feature activation-9.32e-01

Pos logit contributions
 BDS+0.165
 NEO+0.151
 positively+0.134
 withd+0.132
ppelin+0.131
Neg logit contributions
Page-0.111
-0.110
Bloom-0.108
mination-0.105
��-0.103
Token Experimental
Token ID32286
Feature activation-5.97e-01

Pos logit contributions
 BDS+2.861
 NEO+2.546
 positively+2.187
 negatively+2.075
 withd+2.066
Neg logit contributions
-2.667
Page-2.610
Bloom-2.607
mination-2.527
��-2.523
Token subjects
Token ID7481
Feature activation-4.94e-01

Pos logit contributions
 BDS+1.912
 NEO+1.714
 positively+1.508
 withd+1.447
 negatively+1.435
Neg logit contributions
-1.595
Page-1.588
Bloom-1.565
mination-1.523
��-1.495
Token Three
Token ID7683
Feature activation-2.31e-02

Pos logit contributions
 BDS+1.492
 NEO+1.326
 positively+1.157
 withd+1.107
ppelin+1.101
Neg logit contributions
-1.406
Page-1.404
Bloom-1.385
mination-1.347
��-1.324
Token female
Token ID4048
Feature activation+4.38e-02

Pos logit contributions
 BDS+0.066
 NEO+0.058
 positively+0.050
 withd+0.048
ppelin+0.047
Neg logit contributions
-0.070
Page-0.069
Bloom-0.069
mination-0.067
��-0.066
Token r
Token ID374
Feature activation-2.18e-01

Pos logit contributions
+0.128
Page+0.127
Bloom+0.125
mination+0.122
��+0.120
Neg logit contributions
 BDS-0.128
 NEO-0.113
 positively-0.098
ppelin-0.095
 withd-0.094
Token
Token ID198
Feature activation+6.21e-02

Pos logit contributions
Page+0.198
+0.196
Bloom+0.195
mination+0.188
Range+0.183
Neg logit contributions
 BDS-0.194
 NEO-0.173
 withd-0.154
 positively-0.150
 negatively-0.143
Token"
Token ID1
Feature activation+1.19e-01

Pos logit contributions
+0.148
Page+0.147
Bloom+0.145
mination+0.140
��+0.137
Neg logit contributions
 BDS-0.218
 NEO-0.197
 positively-0.176
 withd-0.170
 negatively-0.168
TokenWhat
Token ID2061
Feature activation+5.48e-02

Pos logit contributions
Page+0.284
+0.283
Bloom+0.280
mination+0.269
��+0.262
Neg logit contributions
 BDS-0.416
 NEO-0.376
 positively-0.336
 withd-0.326
 negatively-0.322
Token does
Token ID857
Feature activation-3.75e-03

Pos logit contributions
Page+0.143
+0.142
Bloom+0.140
mination+0.136
��+0.132
Neg logit contributions
 BDS-0.177
 NEO-0.159
 positively-0.140
 withd-0.137
ppelin-0.136
Token it
Token ID340
Feature activation-3.18e-01

Pos logit contributions
 BDS+0.011
 NEO+0.010
 positively+0.009
 withd+0.009
ppelin+0.008
Neg logit contributions
-0.011
Page-0.011
Bloom-0.010
mination-0.010
��-0.010
Token mean
Token ID1612
Feature activation-9.27e-01

Pos logit contributions
 BDS+0.977
 NEO+0.871
 positively+0.758
 withd+0.728
ppelin+0.721
Neg logit contributions
-0.893
Page-0.887
Bloom-0.877
mination-0.853
��-0.838
Token to
Token ID284
Feature activation-5.43e-02

Pos logit contributions
 BDS+3.001
 NEO+2.696
 positively+2.383
 negatively+2.270
 withd+2.233
Neg logit contributions
-2.459
Bloom-2.409
Page-2.400
mination-2.344
��-2.321
Token an
Token ID281
Feature activation-2.04e-01

Pos logit contributions
 BDS+0.148
 NEO+0.130
 positively+0.110
 withd+0.107
ppelin+0.105
Neg logit contributions
-0.170
Page-0.169
Bloom-0.167
mination-0.162
��-0.160
Token intro
Token ID18951
Feature activation+1.85e-01

Pos logit contributions
 BDS+0.569
 NEO+0.500
 positively+0.430
 withd+0.410
 negatively+0.405
Neg logit contributions
-0.632
Page-0.627
Bloom-0.622
mination-0.606
��-0.597
Tokenvert
Token ID1851
Feature activation-8.65e-02

Pos logit contributions
+0.469
Page+0.465
Bloom+0.460
mination+0.447
��+0.436
Neg logit contributions
 BDS-0.618
 NEO-0.556
 positively-0.492
 withd-0.474
 negatively-0.470
Token online
Token ID2691
Feature activation-1.28e-01

Pos logit contributions
 BDS+0.284
 NEO+0.255
 positively+0.224
 withd+0.217
ppelin+0.214
Neg logit contributions
-0.224
Page-0.223
Bloom-0.219
mination-0.213
��-0.208
Token mates
Token ID27973
Feature activation-1.95e-01

Pos logit contributions
 BDS+0.669
 NEO+0.602
 positively+0.530
 withd+0.508
 negatively+0.506
Neg logit contributions
-0.531
Page-0.525
Bloom-0.521
mination-0.505
��-0.499
Token,
Token ID11
Feature activation-1.12e-01

Pos logit contributions
 BDS+0.488
 NEO+0.423
 positively+0.355
 withd+0.335
 negatively+0.331
Neg logit contributions
-0.659
Page-0.654
Bloom-0.649
mination-0.634
��-0.626
Token V
Token ID569
Feature activation-3.07e-01

Pos logit contributions
 BDS+0.347
 NEO+0.309
 positively+0.270
 withd+0.260
ppelin+0.257
Neg logit contributions
-0.309
Page-0.306
Bloom-0.303
mination-0.294
��-0.289
TokenPs
Token ID12016
Feature activation-6.24e-02

Pos logit contributions
 BDS+1.012
 NEO+0.909
 positively+0.802
 withd+0.769
 negatively+0.764
Neg logit contributions
-0.794
Page-0.788
Bloom-0.778
mination-0.757
��-0.742
Token See
Token ID4091
Feature activation-6.92e-02

Pos logit contributions
 BDS+0.197
 NEO+0.176
 positively+0.155
 withd+0.149
 negatively+0.147
Neg logit contributions
-0.170
Page-0.169
Bloom-0.167
mination-0.162
��-0.159
Token Gallery
Token ID12917
Feature activation-9.15e-01

Pos logit contributions
 BDS+0.240
 NEO+0.217
 positively+0.193
 withd+0.188
ppelin+0.185
Neg logit contributions
-0.166
Page-0.165
Bloom-0.163
mination-0.158
��-0.154
Token Donald
Token ID3759
Feature activation-5.08e-01

Pos logit contributions
 BDS+2.798
 NEO+2.498
 positively+2.152
 negatively+2.045
 TED+2.034
Neg logit contributions
-2.586
Bloom-2.574
Page-2.548
mination-2.502
��-2.467
Token Trump
Token ID1301
Feature activation-2.10e-01

Pos logit contributions
 BDS+1.740
 NEO+1.567
 positively+1.387
 negatively+1.325
ppelin+1.315
Neg logit contributions
-1.254
Page-1.236
Bloom-1.231
mination-1.197
��-1.190
Token passes
Token ID8318
Feature activation-1.57e-01

Pos logit contributions
 BDS+0.755
 NEO+0.685
 positively+0.612
 negatively+0.586
 withd+0.586
Neg logit contributions
-0.484
Page-0.478
Bloom-0.474
mination-0.458
��-0.450
Token on
Token ID319
Feature activation-1.25e-01

Pos logit contributions
 BDS+0.449
 NEO+0.397
 positively+0.342
 withd+0.326
 negatively+0.323
Neg logit contributions
-0.473
Page-0.469
Bloom-0.465
mination-0.453
��-0.447
Token '
Token ID705
Feature activation-2.14e-01

Pos logit contributions
 BDS+0.375
 NEO+0.334
 positively+0.290
 withd+0.277
 negatively+0.275
Neg logit contributions
-0.359
Page-0.356
Bloom-0.353
mination-0.343
��-0.338
TokenQ
Token ID48
Feature activation-2.41e-01

Pos logit contributions
 BDS+0.401
 NEO+0.361
 positively+0.320
 withd+0.306
 negatively+0.305
Neg logit contributions
-0.300
Page-0.296
Bloom-0.293
mination-0.285
��-0.280
Token.
Token ID13
Feature activation-2.56e-01

Pos logit contributions
 BDS+0.739
 NEO+0.658
 positively+0.574
 withd+0.548
 negatively+0.544
Neg logit contributions
-0.680
Page-0.674
Bloom-0.668
mination-0.650
��-0.640
Token What
Token ID1867
Feature activation-7.59e-02

Pos logit contributions
 BDS+0.749
 NEO+0.663
 positively+0.574
 withd+0.547
 negatively+0.543
Neg logit contributions
-0.759
Page-0.753
Bloom-0.746
mination-0.726
��-0.716
Token does
Token ID857
Feature activation-4.78e-03

Pos logit contributions
 BDS+0.250
 NEO+0.224
 positively+0.197
 withd+0.191
ppelin+0.189
Neg logit contributions
-0.197
Page-0.195
Bloom-0.193
mination-0.186
��-0.182
Token this
Token ID428
Feature activation-2.76e-01

Pos logit contributions
 BDS+0.016
 NEO+0.014
 positively+0.012
 withd+0.012
ppelin+0.012
Neg logit contributions
-0.013
Page-0.013
Bloom-0.012
mination-0.012
��-0.012
Token mean
Token ID1612
Feature activation-9.14e-01

Pos logit contributions
 BDS+0.849
 NEO+0.757
 positively+0.661
 withd+0.631
 negatively+0.627
Neg logit contributions
-0.776
Page-0.769
Bloom-0.762
mination-0.741
��-0.730
Token in
Token ID287
Feature activation-1.32e-01

Pos logit contributions
 BDS+2.890
 NEO+2.588
 positively+2.291
 negatively+2.186
 withd+2.143
Neg logit contributions
-2.472
Page-2.449
Bloom-2.446
mination-2.385
��-2.356
Token terms
Token ID2846
Feature activation+1.80e-01

Pos logit contributions
 BDS+0.415
 NEO+0.371
 positively+0.325
 withd+0.313
ppelin+0.309
Neg logit contributions
-0.362
Page-0.359
Bloom-0.355
mination-0.345
��-0.338
Token of
Token ID286
Feature activation+1.07e-01

Pos logit contributions
Page+0.473
+0.472
Bloom+0.464
mination+0.451
Range+0.435
Neg logit contributions
 BDS-0.580
 NEO-0.521
 positively-0.455
 withd-0.450
ppelin-0.437
Token data
Token ID1366
Feature activation-1.20e-02

Pos logit contributions
+0.306
Page+0.305
Bloom+0.301
mination+0.293
��+0.286
Neg logit contributions
 BDS-0.321
 NEO-0.285
 positively-0.247
 withd-0.239
ppelin-0.236
Token collection
Token ID4947
Feature activation-2.15e-01

Pos logit contributions
 BDS+0.036
 NEO+0.032
 positively+0.027
 withd+0.026
 negatively+0.026
Neg logit contributions
-0.035
Page-0.034
Bloom-0.034
mination-0.033
��-0.033
Tokenbased
Token ID3106
Feature activation+9.71e-03

Pos logit contributions
+0.094
Page+0.093
Bloom+0.092
mination+0.089
��+0.088
Neg logit contributions
 BDS-0.127
 NEO-0.115
 positively-0.102
 withd-0.097
 negatively-0.097
Token simulations
Token ID27785
Feature activation+1.61e-01

Pos logit contributions
+0.029
Page+0.029
Bloom+0.029
mination+0.028
��+0.028
Neg logit contributions
 BDS-0.028
 NEO-0.025
 positively-0.021
 withd-0.020
ppelin-0.020
Token.
Token ID13
Feature activation-8.71e-02

Pos logit contributions
Page+0.426
+0.421
Bloom+0.416
mination+0.406
��+0.393
Neg logit contributions
 BDS-0.522
 NEO-0.468
 positively-0.407
 withd-0.392
ppelin-0.391
Token
Token ID198
Feature activation-5.44e-02

Pos logit contributions
 BDS+0.250
 NEO+0.221
 positively+0.192
 withd+0.183
 negatively+0.181
Neg logit contributions
-0.262
Page-0.260
Bloom-0.258
mination-0.251
��-0.246
Token
Token ID198
Feature activation-1.39e-02

Pos logit contributions
 BDS+0.157
 NEO+0.140
 withd+0.122
 positively+0.121
 negatively+0.115
Neg logit contributions
Page-0.163
-0.162
Bloom-0.161
mination-0.155
Range-0.151
TokenRemember
Token ID16676
Feature activation-9.04e-01

Pos logit contributions
 BDS+0.048
 NEO+0.043
 positively+0.038
 withd+0.037
 negatively+0.037
Neg logit contributions
-0.034
Page-0.033
Bloom-0.033
mination-0.032
��-0.032
Token from
Token ID422
Feature activation+2.41e-02

Pos logit contributions
 BDS+2.653
 NEO+2.352
 positively+2.047
 negatively+1.932
 withd+1.915
Neg logit contributions
-2.677
Page-2.601
Bloom-2.597
mination-2.539
��-2.507
Token Part
Token ID2142
Feature activation-2.91e-01

Pos logit contributions
+0.073
Page+0.072
Bloom+0.072
mination+0.070
��+0.069
Neg logit contributions
 BDS-0.069
 NEO-0.061
 positively-0.052
 withd-0.050
ppelin-0.049
Token 6
Token ID718
Feature activation+2.64e-01

Pos logit contributions
 BDS+0.934
 NEO+0.836
 positively+0.733
 withd+0.701
 negatively+0.698
Neg logit contributions
-0.782
Page-0.774
Bloom-0.767
mination-0.745
��-0.734
Token that
Token ID326
Feature activation-1.23e-01

Pos logit contributions
Page+0.737
+0.737
Bloom+0.726
mination+0.707
��+0.695
Neg logit contributions
 BDS-0.815
 NEO-0.727
 positively-0.633
 withd-0.609
 negatively-0.601
Token when
Token ID618
Feature activation-2.25e-01

Pos logit contributions
 BDS+0.380
 NEO+0.339
 positively+0.296
 withd+0.283
 negatively+0.281
Neg logit contributions
-0.346
Page-0.343
Bloom-0.340
mination-0.330
��-0.325
Tokenman
Token ID805
Feature activation-1.50e-02

Pos logit contributions
+0.842
Page+0.839
Bloom+0.827
mination+0.809
��+0.784
Neg logit contributions
 BDS-0.983
 NEO-0.880
 positively-0.773
 withd-0.742
 negatively-0.736
Token:
Token ID25
Feature activation+1.00e-01

Pos logit contributions
 BDS+0.046
 NEO+0.041
 positively+0.036
 withd+0.034
 negatively+0.034
Neg logit contributions
-0.043
Page-0.042
Bloom-0.042
mination-0.041
��-0.040
Token What
Token ID1867
Feature activation-1.77e-01

Pos logit contributions
+0.294
Page+0.293
Bloom+0.289
mination+0.280
��+0.272
Neg logit contributions
 BDS-0.294
 NEO-0.261
 positively-0.228
 withd-0.221
 negatively-0.217
Token does
Token ID857
Feature activation-1.05e-01

Pos logit contributions
 BDS+0.575
 NEO+0.518
 positively+0.453
 withd+0.437
ppelin+0.433
Neg logit contributions
-0.465
Page-0.460
Bloom-0.455
mination-0.441
��-0.432
Token that
Token ID326
Feature activation-3.68e-01

Pos logit contributions
 BDS+0.331
 NEO+0.297
 positively+0.260
 withd+0.250
ppelin+0.247
Neg logit contributions
-0.285
Page-0.282
Bloom-0.279
mination-0.270
��-0.265
Token mean
Token ID1612
Feature activation-9.01e-01

Pos logit contributions
 BDS+1.137
 NEO+1.012
 positively+0.887
 withd+0.844
 negatively+0.843
Neg logit contributions
-1.030
Page-1.023
Bloom-1.014
mination-0.986
��-0.971
Token,
Token ID11
Feature activation-1.78e-01

Pos logit contributions
 BDS+2.841
 NEO+2.525
 positively+2.263
 negatively+2.149
 withd+2.114
Neg logit contributions
-2.435
Page-2.418
Bloom-2.414
mination-2.350
��-2.320
Token operators
Token ID12879
Feature activation+5.03e-02

Pos logit contributions
 BDS+0.555
 NEO+0.496
 positively+0.433
 withd+0.415
 negatively+0.412
Neg logit contributions
-0.493
Page-0.489
Bloom-0.484
mination-0.471
��-0.463
Token in
Token ID287
Feature activation-5.54e-02

Pos logit contributions
+0.133
Page+0.132
Bloom+0.130
mination+0.126
��+0.124
Neg logit contributions
 BDS-0.163
 NEO-0.146
 positively-0.129
 withd-0.123
 negatively-0.122
Token the
Token ID262
Feature activation-9.54e-02

Pos logit contributions
 BDS+0.169
 NEO+0.151
 positively+0.131
 withd+0.127
ppelin+0.126
Neg logit contributions
-0.155
Page-0.155
Bloom-0.154
mination-0.148
��-0.145
Token industry
Token ID2831
Feature activation-2.03e-02

Pos logit contributions
 BDS+0.253
 NEO+0.223
 positively+0.191
 withd+0.183
ppelin+0.181
Neg logit contributions
-0.301
Page-0.301
Bloom-0.299
mination-0.289
��-0.286
Token Morris
Token ID14433
Feature activation-2.06e-01

Pos logit contributions
+0.320
Page+0.318
Bloom+0.315
mination+0.305
��+0.299
Neg logit contributions
 BDS-0.351
 NEO-0.313
 positively-0.274
 withd-0.262
ppelin-0.261
Tokenville
Token ID4244
Feature activation+6.29e-02

Pos logit contributions
 BDS+0.630
 NEO+0.559
 positively+0.486
 withd+0.463
 negatively+0.462
Neg logit contributions
-0.587
Page-0.579
Bloom-0.574
mination-0.561
��-0.553
Token
Token ID198
Feature activation+9.04e-03

Pos logit contributions
+0.174
Page+0.174
Bloom+0.172
mination+0.167
��+0.164
Neg logit contributions
 BDS-0.195
 NEO-0.174
 positively-0.152
 withd-0.146
 negatively-0.145
Token
Token ID198
Feature activation-3.87e-02

Pos logit contributions
Page+0.027
+0.027
Bloom+0.027
mination+0.026
Beast+0.025
Neg logit contributions
 BDS-0.026
 NEO-0.023
 positively-0.020
 withd-0.020
 negatively-0.019
Token(
Token ID7
Feature activation-2.66e-01

Pos logit contributions
 BDS+0.140
 NEO+0.127
 positively+0.114
 withd+0.109
 negatively+0.109
Neg logit contributions
-0.088
Page-0.086
Bloom-0.085
mination-0.083
��-0.081
TokenNote
Token ID6425
Feature activation-8.81e-01

Pos logit contributions
 BDS+0.976
 NEO+0.885
 positively+0.791
 withd+0.763
 negatively+0.759
Neg logit contributions
-0.585
Page-0.573
Bloom-0.567
mination-0.553
��-0.548
Token:
Token ID25
Feature activation-6.03e-02

Pos logit contributions
 BDS+2.597
 NEO+2.294
 positively+1.995
 negatively+1.898
 withd+1.871
Neg logit contributions
-2.584
Bloom-2.537
Page-2.523
mination-2.479
��-2.459
Token edited
Token ID13012
Feature activation+3.21e-01

Pos logit contributions
 BDS+0.183
 NEO+0.163
 positively+0.142
 withd+0.136
 negatively+0.135
Neg logit contributions
-0.172
Page-0.170
Bloom-0.169
mination-0.164
��-0.161
Token for
Token ID329
Feature activation+2.77e-02

Pos logit contributions
+0.836
Page+0.834
Bloom+0.819
mination+0.797
��+0.777
Neg logit contributions
 BDS-1.047
 NEO-0.943
 positively-0.827
 withd-0.798
ppelin-0.797
Token length
Token ID4129
Feature activation-1.73e-01

Pos logit contributions
+0.076
Page+0.076
Bloom+0.075
mination+0.073
��+0.072
Neg logit contributions
 BDS-0.087
 NEO-0.077
 positively-0.068
 withd-0.065
 negatively-0.064
Token.)
Token ID2014
Feature activation+1.10e-01

Pos logit contributions
 BDS+0.553
 NEO+0.494
 positively+0.435
 withd+0.415
 negatively+0.415
Neg logit contributions
-0.467
Page-0.460
Bloom-0.458
mination-0.444
��-0.438
Token the
Token ID262
Feature activation-1.63e-01

Pos logit contributions
 BDS+0.418
 NEO+0.371
 positively+0.322
 withd+0.307
 negatively+0.305
Neg logit contributions
-0.407
Page-0.404
Bloom-0.400
mination-0.390
��-0.384
Token firing
Token ID9645
Feature activation-4.43e-02

Pos logit contributions
 BDS+0.462
 NEO+0.409
 positively+0.350
 withd+0.336
ppelin+0.332
Neg logit contributions
-0.497
Page-0.495
Bloom-0.488
mination-0.475
��-0.467
Token line
Token ID1627
Feature activation+5.75e-03

Pos logit contributions
 BDS+0.130
 NEO+0.115
 positively+0.099
 withd+0.095
 negatively+0.094
Neg logit contributions
-0.131
Page-0.130
Bloom-0.128
mination-0.125
��-0.123
Token during
Token ID1141
Feature activation-1.32e-01

Pos logit contributions
Page+0.016
+0.016
Bloom+0.016
mination+0.016
��+0.015
Neg logit contributions
 BDS-0.017
 NEO-0.016
 positively-0.013
 withd-0.013
ppelin-0.013
Token Commons
Token ID13815
Feature activation-1.26e-02

Pos logit contributions
 BDS+0.415
 NEO+0.373
 positively+0.325
 withd+0.315
ppelin+0.310
Neg logit contributions
-0.361
Page-0.359
Bloom-0.355
mination-0.344
��-0.338
Token Question
Token ID18233
Feature activation-8.71e-01

Pos logit contributions
 BDS+0.038
 NEO+0.034
 positively+0.029
 withd+0.028
 negatively+0.028
Neg logit contributions
-0.036
Page-0.036
Bloom-0.036
mination-0.035
��-0.034
Token Period
Token ID18581
Feature activation-5.59e-02

Pos logit contributions
 BDS+2.973
 NEO+2.672
 positively+2.362
 negatively+2.259
ppelin+2.250
Neg logit contributions
-2.160
Page-2.128
Bloom-2.127
mination-2.044
��-2.032
Token Wednesday
Token ID3583
Feature activation-2.81e-01

Pos logit contributions
 BDS+0.167
 NEO+0.149
 positively+0.128
 withd+0.123
ppelin+0.122
Neg logit contributions
-0.161
Page-0.161
Bloom-0.159
mination-0.155
��-0.151
Token as
Token ID355
Feature activation-1.03e-01

Pos logit contributions
 BDS+0.849
 NEO+0.757
 positively+0.653
 withd+0.628
ppelin+0.623
Neg logit contributions
-0.801
Page-0.798
Bloom-0.789
mination-0.767
��-0.754
Token Conservative
Token ID11132
Feature activation-3.73e-01

Pos logit contributions
 BDS+0.342
 NEO+0.309
 positively+0.272
 withd+0.262
ppelin+0.260
Neg logit contributions
-0.262
Page-0.260
Bloom-0.258
mination-0.249
��-0.245
Token MPs
Token ID14952
Feature activation-1.57e-01

Pos logit contributions
 BDS+1.254
 NEO+1.129
 positively+0.999
 withd+0.959
 negatively+0.954
Neg logit contributions
-0.941
Page-0.931
Bloom-0.922
mination-0.894
��-0.879
Token
Token ID198
Feature activation+3.54e-02

Pos logit contributions
Page+0.051
+0.051
Bloom+0.050
mination+0.048
Beast+0.047
Neg logit contributions
 BDS-0.053
 NEO-0.047
 positively-0.041
 withd-0.040
 negatively-0.039
Token
Token ID198
Feature activation+5.31e-02

Pos logit contributions
Page+0.106
+0.105
Bloom+0.104
mination+0.101
Range+0.098
Neg logit contributions
 BDS-0.102
 NEO-0.091
 withd-0.080
 positively-0.079
 negatively-0.075
TokenWhat
Token ID2061
Feature activation-2.33e-02

Pos logit contributions
Page+0.130
+0.130
Bloom+0.128
mination+0.123
��+0.120
Neg logit contributions
 BDS-0.183
 NEO-0.165
 positively-0.147
 withd-0.143
 negatively-0.141
Token does
Token ID857
Feature activation+2.45e-02

Pos logit contributions
 BDS+0.079
 NEO+0.071
 positively+0.062
 withd+0.061
ppelin+0.060
Neg logit contributions
-0.058
Page-0.058
Bloom-0.057
mination-0.055
��-0.053
Token this
Token ID428
Feature activation-2.90e-01

Pos logit contributions
+0.065
Page+0.064
Bloom+0.064
mination+0.061
��+0.060
Neg logit contributions
 BDS-0.079
 NEO-0.071
 positively-0.063
 withd-0.061
ppelin-0.060
Token mean
Token ID1612
Feature activation-8.52e-01

Pos logit contributions
 BDS+0.894
 NEO+0.797
 positively+0.697
 withd+0.665
 negatively+0.661
Neg logit contributions
-0.811
Page-0.804
Bloom-0.797
mination-0.775
��-0.763
Token?
Token ID30
Feature activation+4.16e-02

Pos logit contributions
 BDS+2.703
 NEO+2.419
 positively+2.147
 negatively+2.048
 withd+2.013
Neg logit contributions
-2.296
Page-2.272
Bloom-2.263
mination-2.204
��-2.177
Token A
Token ID317
Feature activation-3.45e-02

Pos logit contributions
Page+0.120
+0.120
Bloom+0.118
mination+0.114
Range+0.111
Neg logit contributions
 BDS-0.124
 NEO-0.109
 positively-0.097
 withd-0.094
ppelin-0.092
Token lot
Token ID1256
Feature activation+3.07e-01

Pos logit contributions
 BDS+0.099
 NEO+0.087
 positively+0.075
 withd+0.072
ppelin+0.071
Neg logit contributions
-0.104
Page-0.103
Bloom-0.102
mination-0.100
��-0.098
Token of
Token ID286
Feature activation-1.59e-02

Pos logit contributions
+0.797
Page+0.793
Bloom+0.781
mination+0.757
��+0.742
Neg logit contributions
 BDS-1.007
 NEO-0.909
 positively-0.792
 withd-0.771
ppelin-0.767
Token messages
Token ID6218
Feature activation-1.37e-01

Pos logit contributions
 BDS+0.048
 NEO+0.043
 positively+0.037
 withd+0.036
ppelin+0.036
Neg logit contributions
-0.045
Page-0.045
Bloom-0.045
mination-0.043
��-0.042
Token Posts
Token ID12043
Feature activation+1.73e-01

Pos logit contributions
 BDS+0.300
 NEO+0.270
 positively+0.237
 withd+0.227
 negatively+0.226
Neg logit contributions
-0.239
Page-0.236
Bloom-0.234
mination-0.227
��-0.224
Token #
Token ID1303
Feature activation-2.19e-01

Pos logit contributions
+0.428
Page+0.426
Bloom+0.418
mination+0.404
��+0.398
Neg logit contributions
 BDS-0.590
 NEO-0.532
 positively-0.471
 withd-0.455
 negatively-0.451
Token16
Token ID1433
Feature activation-2.87e-01

Pos logit contributions
 BDS+0.790
 NEO+0.717
 positively+0.640
 withd+0.614
 negatively+0.613
Neg logit contributions
-0.501
Page-0.495
Bloom-0.489
mination-0.473
��-0.468
Token Glad
Token ID28442
Feature activation-2.50e-01

Pos logit contributions
 BDS+0.784
 NEO+0.687
 positively+0.587
 withd+0.559
 negatively+0.553
Neg logit contributions
-0.905
Page-0.899
Bloom-0.891
mination-0.869
��-0.855
Token to
Token ID284
Feature activation-1.84e-01

Pos logit contributions
 BDS+0.813
 NEO+0.730
 positively+0.641
 withd+0.614
 negatively+0.610
Neg logit contributions
-0.663
Page-0.656
Bloom-0.650
mination-0.631
��-0.621
Token hear
Token ID3285
Feature activation-8.48e-01

Pos logit contributions
 BDS+0.486
 NEO+0.424
 positively+0.356
 withd+0.343
ppelin+0.339
Neg logit contributions
-0.594
Page-0.591
Bloom-0.583
mination-0.568
��-0.559
Token everything
Token ID2279
Feature activation-8.12e-02

Pos logit contributions
 BDS+2.673
 NEO+2.408
 positively+2.107
 negatively+1.992
 withd+1.990
Neg logit contributions
-2.323
Page-2.296
Bloom-2.278
mination-2.224
��-2.202
Token worked
Token ID3111
Feature activation-1.33e-01

Pos logit contributions
 BDS+0.275
 NEO+0.248
 positively+0.219
 withd+0.211
 negatively+0.210
Neg logit contributions
-0.203
Page-0.201
Bloom-0.199
mination-0.193
��-0.189
Token out
Token ID503
Feature activation+3.07e-03

Pos logit contributions
 BDS+0.422
 NEO+0.378
 positively+0.331
 withd+0.317
 negatively+0.315
Neg logit contributions
-0.363
Page-0.359
Bloom-0.356
mination-0.346
��-0.341
Token great
Token ID1049
Feature activation+3.50e-01

Pos logit contributions
+0.008
Page+0.008
Bloom+0.008
mination+0.008
��+0.008
Neg logit contributions
 BDS-0.010
 NEO-0.009
 positively-0.008
 withd-0.007
 negatively-0.007
Token.
Token ID13
Feature activation-1.77e-01

Pos logit contributions
+0.921
Page+0.918
Bloom+0.904
mination+0.874
��+0.857
Neg logit contributions
 BDS-1.130
 NEO-1.011
 positively-0.887
ppelin-0.854
 withd-0.854
Token it
Token ID340
Feature activation-1.90e-01

Pos logit contributions
Page+0.439
+0.438
Bloom+0.430
mination+0.414
Beast+0.399
Neg logit contributions
 BDS-0.631
 NEO-0.580
 withd-0.517
 positively-0.504
ppelin-0.487
Token to
Token ID284
Feature activation-1.65e-01

Pos logit contributions
 BDS+0.661
 NEO+0.598
 positively+0.528
 withd+0.517
ppelin+0.508
Neg logit contributions
-0.458
Page-0.456
Bloom-0.449
mination-0.434
��-0.422
Token '
Token ID705
Feature activation-2.08e-01

Pos logit contributions
 BDS+0.395
 NEO+0.341
 positively+0.281
 withd+0.266
ppelin+0.261
Neg logit contributions
-0.576
Page-0.573
Bloom-0.569
mination-0.556
��-0.548
Tokenlead
Token ID28230
Feature activation-3.57e-01

Pos logit contributions
 BDS+0.571
 NEO+0.502
 positively+0.429
 withd+0.407
 negatively+0.404
Neg logit contributions
-0.653
Page-0.648
Bloom-0.642
mination-0.627
��-0.618
Token by
Token ID416
Feature activation-1.47e-01

Pos logit contributions
 BDS+1.207
 NEO+1.086
 positively+0.965
 negatively+0.920
 withd+0.917
Neg logit contributions
-0.902
Page-0.891
Bloom-0.884
mination-0.857
��-0.844
Token example
Token ID1672
Feature activation-8.46e-01

Pos logit contributions
 BDS+0.451
 NEO+0.402
 positively+0.351
 withd+0.336
 negatively+0.333
Neg logit contributions
-0.412
Page-0.409
Bloom-0.405
mination-0.394
��-0.387
Token with
Token ID351
Feature activation-2.48e-01

Pos logit contributions
 BDS+2.431
 NEO+2.136
 positively+1.898
 withd+1.798
 negatively+1.782
Neg logit contributions
-2.539
Page-2.506
Bloom-2.491
mination-2.443
��-2.404
Token hope
Token ID2911
Feature activation-2.06e-01

Pos logit contributions
 BDS+0.699
 NEO+0.617
 positively+0.529
 withd+0.508
 negatively+0.500
Neg logit contributions
-0.756
Page-0.752
Bloom-0.745
mination-0.726
��-0.713
Token,
Token ID11
Feature activation-2.11e-01

Pos logit contributions
 BDS+0.632
 NEO+0.563
 positively+0.492
 withd+0.467
 negatively+0.466
Neg logit contributions
-0.584
Page-0.577
Bloom-0.573
mination-0.558
��-0.549
Token never
Token ID1239
Feature activation-7.90e-02

Pos logit contributions
 BDS+0.679
 NEO+0.609
 positively+0.535
 withd+0.514
 negatively+0.509
Neg logit contributions
-0.560
Page-0.556
Bloom-0.550
mination-0.534
��-0.525
Token fear
Token ID3252
Feature activation+2.08e-02

Pos logit contributions
 BDS+0.155
 NEO+0.129
 positively+0.101
 withd+0.093
 negatively+0.091
Neg logit contributions
-0.309
Page-0.308
Bloom-0.306
mination-0.300
��-0.296
Token
Token ID198
Feature activation-2.73e-02

Pos logit contributions
 BDS+0.124
 NEO+0.110
 positively+0.095
 withd+0.094
 negatively+0.090
Neg logit contributions
Page-0.132
-0.131
Bloom-0.130
mination-0.126
Beast-0.122
Token
Token ID447
Feature activation-1.08e-01

Pos logit contributions
 BDS+0.095
 NEO+0.086
 positively+0.077
 withd+0.074
 negatively+0.073
Neg logit contributions
-0.065
Page-0.065
Bloom-0.064
mination-0.062
��-0.061
Token
Token ID250
Feature activation+2.44e-02

Pos logit contributions
 BDS+0.405
 NEO+0.368
 positively+0.331
 negatively+0.317
ppelin+0.316
Neg logit contributions
-0.229
Page-0.225
Bloom-0.223
mination-0.215
��-0.209
TokenT
Token ID51
Feature activation+1.41e-01

Pos logit contributions
+0.062
Page+0.062
Bloom+0.061
mination+0.059
��+0.058
Neg logit contributions
 BDS-0.081
 NEO-0.073
 positively-0.065
 withd-0.062
 negatively-0.062
Tokenakeru
Token ID43246
Feature activation+1.76e-01

Pos logit contributions
+0.362
Page+0.358
Bloom+0.355
mination+0.344
��+0.339
Neg logit contributions
 BDS-0.465
 NEO-0.418
 positively-0.369
 withd-0.354
 negatively-0.352
Token Kob
Token ID23403
Feature activation-8.44e-01

Pos logit contributions
+0.499
Page+0.495
Bloom+0.490
mination+0.476
��+0.468
Neg logit contributions
 BDS-0.537
 NEO-0.478
 positively-0.417
 withd-0.398
 negatively-0.395
Tokenay
Token ID323
Feature activation-2.67e-02

Pos logit contributions
 BDS+2.559
 NEO+2.313
 positively+2.010
 negatively+1.911
ppelin+1.898
Neg logit contributions
-2.351
Bloom-2.321
Page-2.305
��-2.249
mination-2.227
Tokenashi
Token ID12144
Feature activation-9.48e-02

Pos logit contributions
 BDS+0.082
 NEO+0.073
 positively+0.064
 withd+0.061
 negatively+0.061
Neg logit contributions
-0.075
Page-0.074
Bloom-0.074
mination-0.072
��-0.070
Token
Token ID447
Feature activation-2.30e-01

Pos logit contributions
 BDS+0.328
 NEO+0.296
 positively+0.263
 withd+0.253
 negatively+0.251
Neg logit contributions
-0.231
Page-0.228
Bloom-0.226
mination-0.219
��-0.215
Token
Token ID247
Feature activation+1.12e-02

Pos logit contributions
 BDS+1.041
 NEO+0.956
 positively+0.884
 withd+0.858
 negatively+0.856
Neg logit contributions
-0.315
Page-0.305
Bloom-0.298
mination-0.283
Range-0.262
Tokens
Token ID82
Feature activation-9.20e-02

Pos logit contributions
+0.028
Page+0.028
Bloom+0.028
mination+0.027
��+0.026
Neg logit contributions
 BDS-0.037
 NEO-0.034
 positively-0.030
 withd-0.029
 negatively-0.028
Token of
Token ID286
Feature activation+1.31e-01

Pos logit contributions
Page+0.696
+0.694
Bloom+0.679
mination+0.659
��+0.649
Neg logit contributions
 BDS-0.826
 NEO-0.740
 positively-0.645
 withd-0.625
ppelin-0.625
Token the
Token ID262
Feature activation+1.03e-01

Pos logit contributions
+0.361
Page+0.359
Bloom+0.353
mination+0.343
��+0.339
Neg logit contributions
 BDS-0.401
 NEO-0.362
 positively-0.314
 withd-0.305
ppelin-0.302
Token statements
Token ID6299
Feature activation+2.54e-01

Pos logit contributions
+0.274
Page+0.271
Bloom+0.267
mination+0.259
��+0.256
Neg logit contributions
 BDS-0.330
 NEO-0.298
 positively-0.260
 withd-0.252
ppelin-0.251
Token.
Token ID13
Feature activation-4.82e-02

Pos logit contributions
Page+0.669
+0.664
Bloom+0.651
mination+0.630
��+0.622
Neg logit contributions
 BDS-0.818
 NEO-0.734
 positively-0.637
ppelin-0.621
 withd-0.619
Token For
Token ID1114
Feature activation+5.48e-03

Pos logit contributions
 BDS+0.144
 NEO+0.129
 positively+0.113
 withd+0.109
ppelin+0.107
Neg logit contributions
-0.138
Page-0.137
Bloom-0.135
mination-0.132
��-0.128
Token example
Token ID1672
Feature activation-8.43e-01

Pos logit contributions
+0.015
Page+0.015
Bloom+0.015
mination+0.014
��+0.014
Neg logit contributions
 BDS-0.017
 NEO-0.015
 positively-0.013
 withd-0.013
ppelin-0.013
Token,
Token ID11
Feature activation-2.67e-01

Pos logit contributions
 BDS+2.518
 NEO+2.226
 positively+1.940
 withd+1.843
 negatively+1.839
Neg logit contributions
-2.436
Page-2.408
Bloom-2.398
mination-2.343
��-2.292
Token the
Token ID262
Feature activation-1.69e-01

Pos logit contributions
 BDS+0.810
 NEO+0.722
 positively+0.628
 withd+0.601
 negatively+0.595
Neg logit contributions
-0.759
Page-0.754
Bloom-0.745
mination-0.725
��-0.714
Token sax
Token ID46909
Feature activation-1.42e-01

Pos logit contributions
 BDS+0.488
 NEO+0.433
 positively+0.374
 withd+0.357
ppelin+0.354
Neg logit contributions
-0.501
Page-0.498
Bloom-0.493
mination-0.481
��-0.474
Tokenophon
Token ID48982
Feature activation+1.88e-01

Pos logit contributions
 BDS+0.443
 NEO+0.396
 positively+0.347
 withd+0.331
 negatively+0.329
Neg logit contributions
-0.395
Page-0.390
Bloom-0.388
mination-0.376
��-0.370
Tokenist
Token ID396
Feature activation-9.58e-02

Pos logit contributions
+0.419
Page+0.417
Bloom+0.409
mination+0.396
��+0.389
Neg logit contributions
 BDS-0.684
 NEO-0.622
 positively-0.555
 withd-0.537
 negatively-0.533
Token is
Token ID318
Feature activation-3.67e-02

Pos logit contributions
+0.427
Page+0.425
Bloom+0.420
mination+0.407
��+0.398
Neg logit contributions
 BDS-0.525
 NEO-0.472
 positively-0.413
 withd-0.400
ppelin-0.397
Token a
Token ID257
Feature activation-3.63e-02

Pos logit contributions
 BDS+0.111
 NEO+0.099
 positively+0.086
 withd+0.083
ppelin+0.082
Neg logit contributions
-0.104
Page-0.103
Bloom-0.102
mination-0.099
��-0.098
Token silly
Token ID14397
Feature activation-9.70e-02

Pos logit contributions
 BDS+0.106
 NEO+0.094
 positively+0.081
 withd+0.077
 negatively+0.076
Neg logit contributions
-0.108
Page-0.107
Bloom-0.106
mination-0.103
��-0.101
Token,
Token ID11
Feature activation-1.08e-01

Pos logit contributions
 BDS+0.291
 NEO+0.259
 positively+0.225
 withd+0.216
 negatively+0.213
Neg logit contributions
-0.279
Page-0.277
Bloom-0.274
mination-0.266
��-0.262
Token innocuous
Token ID45427
Feature activation+2.56e-02

Pos logit contributions
 BDS+0.329
 NEO+0.293
 positively+0.255
 withd+0.244
 negatively+0.242
Neg logit contributions
-0.305
Page-0.302
Bloom-0.299
mination-0.291
��-0.287
Token example
Token ID1672
Feature activation-8.30e-01

Pos logit contributions
+0.073
Page+0.073
Bloom+0.072
mination+0.070
��+0.069
Neg logit contributions
 BDS-0.077
 NEO-0.069
 positively-0.060
 withd-0.058
ppelin-0.057
Token (
Token ID357
Feature activation-1.75e-01

Pos logit contributions
 BDS+2.563
 NEO+2.281
 positively+2.001
 withd+1.904
 negatively+1.901
Neg logit contributions
-2.327
Page-2.291
Bloom-2.284
mination-2.216
��-2.200
Tokenwho
Token ID8727
Feature activation-4.39e-02

Pos logit contributions
 BDS+0.595
 NEO+0.536
 positively+0.475
 withd+0.456
 negatively+0.453
Neg logit contributions
-0.435
Page-0.430
Bloom-0.425
mination-0.413
��-0.407
Token cares
Token ID16609
Feature activation-3.33e-01

Pos logit contributions
 BDS+0.154
 NEO+0.140
 positively+0.124
 withd+0.120
 negatively+0.119
Neg logit contributions
-0.104
Page-0.103
Bloom-0.102
mination-0.098
��-0.097
Token if
Token ID611
Feature activation+1.06e-01

Pos logit contributions
 BDS+1.097
 NEO+0.984
 positively+0.869
 withd+0.831
 negatively+0.829
Neg logit contributions
-0.865
Page-0.856
Bloom-0.849
mination-0.825
��-0.812
Token a
Token ID257
Feature activation+1.48e-02

Pos logit contributions
+0.272
Page+0.271
Bloom+0.268
mination+0.259
��+0.254
Neg logit contributions
 BDS-0.347
 NEO-0.311
 positively-0.275
 withd-0.265
ppelin-0.263
Token he
Token ID339
Feature activation+1.93e-01

Pos logit contributions
+0.076
Page+0.075
Bloom+0.074
mination+0.072
��+0.071
Neg logit contributions
 BDS-0.107
 NEO-0.097
 positively-0.086
 withd-0.083
 negatively-0.082
Tokenaping
Token ID9269
Feature activation-1.26e-01

Pos logit contributions
+0.506
Page+0.499
Bloom+0.496
mination+0.478
��+0.474
Neg logit contributions
 BDS-0.631
 NEO-0.567
 positively-0.501
 withd-0.477
 negatively-0.476
Token tablespoons
Token ID30064
Feature activation-4.12e-01

Pos logit contributions
 BDS+0.431
 NEO+0.386
 positively+0.344
 withd+0.328
 negatively+0.328
Neg logit contributions
-0.314
Page-0.310
Bloom-0.308
mination-0.299
��-0.295
Token of
Token ID286
Feature activation+5.36e-02

Pos logit contributions
 BDS+1.278
 NEO+1.137
 positively+0.999
 withd+0.948
 negatively+0.947
Neg logit contributions
-1.148
Page-1.138
Bloom-1.126
mination-1.095
��-1.082
Token peanut
Token ID26636
Feature activation-2.59e-01

Pos logit contributions
+0.168
Page+0.167
Bloom+0.165
mination+0.161
��+0.158
Neg logit contributions
 BDS-0.147
 NEO-0.130
 positively-0.110
 withd-0.105
ppelin-0.105
Token butter
Token ID9215
Feature activation-8.28e-01

Pos logit contributions
 BDS+0.654
 NEO+0.569
 positively+0.477
 withd+0.451
 negatively+0.447
Neg logit contributions
-0.870
Page-0.863
Bloom-0.857
mination-0.837
��-0.825
Token:
Token ID25
Feature activation+7.09e-02

Pos logit contributions
 BDS+2.575
 NEO+2.277
 positively+2.009
 withd+1.905
 negatively+1.899
Neg logit contributions
-2.298
Page-2.275
Bloom-2.252
mination-2.194
��-2.173
Token
Token ID198
Feature activation-1.89e-02

Pos logit contributions
+0.221
Page+0.219
Bloom+0.217
mination+0.212
��+0.209
Neg logit contributions
 BDS-0.197
 NEO-0.173
 positively-0.148
 withd-0.141
 negatively-0.140
Token
Token ID198
Feature activation-5.92e-03

Pos logit contributions
 BDS+0.054
 NEO+0.048
 positively+0.041
 withd+0.041
 negatively+0.039
Neg logit contributions
Page-0.057
-0.057
Bloom-0.057
mination-0.055
Beast-0.053
TokenBl
Token ID3629
Feature activation+6.25e-01

Pos logit contributions
 BDS+0.021
 NEO+0.019
 positively+0.017
 withd+0.016
 negatively+0.016
Neg logit contributions
-0.014
Page-0.014
Bloom-0.014
mination-0.013
��-0.013
Tokenend
Token ID437
Feature activation+2.36e-03

Pos logit contributions
Page+1.620
+1.609
Bloom+1.592
mination+1.559
Range+1.500
Neg logit contributions
 BDS-2.063
 NEO-1.861
 positively-1.617
 withd-1.574
 negatively-1.552
Token Amazon
Token ID6186
Feature activation-2.14e-02

Pos logit contributions
+0.628
Page+0.623
Bloom+0.617
mination+0.597
��+0.582
Neg logit contributions
 BDS-0.694
 NEO-0.619
 positively-0.539
 withd-0.520
ppelin-0.517
Token itself
Token ID2346
Feature activation+1.46e-01

Pos logit contributions
 BDS+0.069
 NEO+0.062
 positively+0.054
 withd+0.052
 negatively+0.052
Neg logit contributions
-0.057
Page-0.057
Bloom-0.056
mination-0.054
��-0.053
Token.
Token ID13
Feature activation-2.68e-03

Pos logit contributions
+0.385
Page+0.383
Bloom+0.377
mination+0.366
��+0.358
Neg logit contributions
 BDS-0.475
 NEO-0.427
 positively-0.373
 withd-0.359
ppelin-0.357
Token
Token ID198
Feature activation-6.86e-02

Pos logit contributions
 BDS+0.008
 NEO+0.007
 positively+0.006
 withd+0.006
 negatively+0.006
Neg logit contributions
-0.008
Page-0.008
Bloom-0.008
mination-0.007
��-0.007
Token
Token ID198
Feature activation-3.33e-02

Pos logit contributions
 BDS+0.197
 NEO+0.175
 withd+0.152
 positively+0.151
 negatively+0.143
Neg logit contributions
Page-0.207
-0.206
Bloom-0.203
mination-0.197
Range-0.192
TokenRemember
Token ID16676
Feature activation-8.23e-01

Pos logit contributions
 BDS+0.117
 NEO+0.105
 positively+0.094
 withd+0.090
 negatively+0.090
Neg logit contributions
-0.079
Page-0.078
Bloom-0.077
mination-0.075
��-0.074
Token,
Token ID11
Feature activation-1.27e-01

Pos logit contributions
 BDS+2.458
 NEO+2.182
 positively+1.901
 negatively+1.797
 withd+1.793
Neg logit contributions
-2.372
Page-2.337
Bloom-2.329
mination-2.278
��-2.268
Token though
Token ID996
Feature activation-2.09e-02

Pos logit contributions
 BDS+0.408
 NEO+0.366
 positively+0.321
 withd+0.308
 negatively+0.306
Neg logit contributions
-0.338
Page-0.335
Bloom-0.332
mination-0.322
��-0.317
Token,
Token ID11
Feature activation+4.72e-02

Pos logit contributions
 BDS+0.064
 NEO+0.057
 positively+0.050
 withd+0.048
 negatively+0.047
Neg logit contributions
-0.059
Page-0.058
Bloom-0.058
mination-0.056
��-0.055
Token that
Token ID326
Feature activation-5.11e-02

Pos logit contributions
+0.131
Page+0.130
Bloom+0.129
mination+0.125
��+0.123
Neg logit contributions
 BDS-0.147
 NEO-0.131
 positively-0.115
 withd-0.110
 negatively-0.109
Token thanks
Token ID5176
Feature activation+1.15e-01

Pos logit contributions
 BDS+0.162
 NEO+0.145
 positively+0.127
 withd+0.122
 negatively+0.121
Neg logit contributions
-0.139
Page-0.138
Bloom-0.136
mination-0.132
��-0.130
Tokenimages
Token ID17566
Feature activation-1.30e-01

Pos logit contributions
 BDS+0.301
 NEO+0.274
 positively+0.245
 withd+0.237
 negatively+0.235
Neg logit contributions
-0.182
Page-0.180
Bloom-0.178
mination-0.172
��-0.167
Token"
Token ID1
Feature activation+9.96e-02

Pos logit contributions
 BDS+0.418
 NEO+0.376
 withd+0.330
 positively+0.329
 negatively+0.313
Neg logit contributions
Page-0.350
-0.348
Bloom-0.341
mination-0.332
Range-0.321
Token ]
Token ID2361
Feature activation+3.63e-01

Pos logit contributions
+0.215
Page+0.213
Bloom+0.209
mination+0.202
��+0.196
Neg logit contributions
 BDS-0.371
 NEO-0.339
 positively-0.304
 withd-0.296
 negatively-0.291
Token )
Token ID1267
Feature activation+7.65e-02

Pos logit contributions
+0.741
Page+0.735
Bloom+0.722
mination+0.692
��+0.677
Neg logit contributions
 BDS-1.391
 NEO-1.273
 positively-1.147
 withd-1.113
 negatively-1.101
Token :
Token ID1058
Feature activation-3.40e-01

Pos logit contributions
+0.187
Page+0.186
Bloom+0.184
mination+0.178
��+0.174
Neg logit contributions
 BDS-0.262
 NEO-0.237
 positively-0.210
 withd-0.203
ppelin-0.201
Token #
Token ID1303
Feature activation-8.21e-01

Pos logit contributions
 BDS+1.139
 NEO+1.023
 positively+0.904
 negatively+0.863
 withd+0.861
Neg logit contributions
-0.874
Page-0.856
Bloom-0.849
mination-0.825
��-0.814
Token load
Token ID3440
Feature activation-3.48e-01

Pos logit contributions
 BDS+1.368
 NEO+1.094
 positively+0.816
 negatively+0.709
 withd+0.707
Neg logit contributions
-3.480
Page-3.441
Bloom-3.416
mination-3.374
��-3.334
Token the
Token ID262
Feature activation-1.88e-01

Pos logit contributions
 BDS+1.047
 NEO+0.931
 positively+0.809
 negatively+0.766
 withd+0.764
Neg logit contributions
-1.001
Page-0.991
Bloom-0.983
mination-0.958
��-0.947
Token image
Token ID2939
Feature activation-1.91e-01

Pos logit contributions
 BDS+0.542
 NEO+0.478
 positively+0.412
 withd+0.407
ppelin+0.395
Neg logit contributions
Page-0.557
-0.557
Bloom-0.549
mination-0.532
��-0.520
Token,
Token ID11
Feature activation-1.42e-01

Pos logit contributions
 BDS+0.594
 NEO+0.531
 positively+0.463
 withd+0.447
 negatively+0.441
Neg logit contributions
-0.530
Page-0.527
Bloom-0.521
mination-0.506
��-0.495
Token resize
Token ID47558
Feature activation-8.63e-02

Pos logit contributions
 BDS+0.488
 NEO+0.441
 positively+0.390
 withd+0.378
ppelin+0.373
Neg logit contributions
-0.344
Page-0.342
Bloom-0.338
mination-0.327
��-0.319
Token-
Token ID12
Feature activation-2.24e-02

Pos logit contributions
 BDS+0.637
 NEO+0.565
 positively+0.491
 withd+0.467
 negatively+0.466
Neg logit contributions
-0.618
Page-0.612
Bloom-0.608
mination-0.592
��-0.583
Tokenprotected
Token ID24326
Feature activation-1.18e-01

Pos logit contributions
 BDS+0.078
 NEO+0.070
 positively+0.062
 withd+0.060
 negatively+0.060
Neg logit contributions
-0.054
Page-0.054
Bloom-0.053
mination-0.051
��-0.051
Token expressions
Token ID14700
Feature activation-2.70e-01

Pos logit contributions
 BDS+0.335
 NEO+0.296
 positively+0.254
 withd+0.243
 negatively+0.240
Neg logit contributions
-0.357
Page-0.354
Bloom-0.350
mination-0.342
��-0.336
Token.
Token ID13
Feature activation-1.07e-01

Pos logit contributions
 BDS+0.855
 NEO+0.767
 positively+0.669
 withd+0.643
ppelin+0.637
Neg logit contributions
-0.733
Page-0.729
Bloom-0.719
mination-0.697
��-0.686
Token For
Token ID1114
Feature activation-1.20e-01

Pos logit contributions
 BDS+0.314
 NEO+0.278
 positively+0.241
 withd+0.230
 negatively+0.228
Neg logit contributions
-0.314
Page-0.312
Bloom-0.309
mination-0.301
��-0.297
Token example
Token ID1672
Feature activation-8.20e-01

Pos logit contributions
 BDS+0.378
 NEO+0.338
 positively+0.296
 withd+0.284
 negatively+0.282
Neg logit contributions
-0.328
Page-0.325
Bloom-0.322
mination-0.312
��-0.307
Token,
Token ID11
Feature activation-2.47e-01

Pos logit contributions
 BDS+2.422
 NEO+2.143
 positively+1.870
 withd+1.781
 negatively+1.773
Neg logit contributions
-2.372
Page-2.344
Bloom-2.336
mination-2.291
��-2.243
Token in
Token ID287
Feature activation-2.71e-01

Pos logit contributions
 BDS+0.743
 NEO+0.661
 positively+0.574
 withd+0.548
 negatively+0.544
Neg logit contributions
-0.710
Page-0.704
Bloom-0.697
mination-0.679
��-0.668
Token 2010
Token ID3050
Feature activation-3.89e-02

Pos logit contributions
 BDS+0.786
 NEO+0.696
 positively+0.601
 withd+0.575
 negatively+0.568
Neg logit contributions
-0.808
Page-0.804
Bloom-0.796
mination-0.774
��-0.762
Token the
Token ID262
Feature activation+1.96e-02

Pos logit contributions
 BDS+0.121
 NEO+0.108
 positively+0.094
 withd+0.090
 negatively+0.090
Neg logit contributions
-0.108
Page-0.107
Bloom-0.106
mination-0.103
��-0.101
Token ACLU
Token ID24381
Feature activation-5.52e-01

Pos logit contributions
+0.057
Page+0.057
Bloom+0.056
mination+0.054
��+0.054
Neg logit contributions
 BDS-0.058
 NEO-0.052
 positively-0.045
 withd-0.043
 negatively-0.042
Token me
Token ID502
Feature activation+1.17e-01

Pos logit contributions
+0.306
Page+0.304
Bloom+0.301
mination+0.292
��+0.287
Neg logit contributions
 BDS-0.349
 NEO-0.311
 positively-0.272
 withd-0.261
 negatively-0.259
Token one
Token ID530
Feature activation+2.40e-02

Pos logit contributions
+0.345
Page+0.341
Bloom+0.338
mination+0.328
��+0.323
Neg logit contributions
 BDS-0.342
 NEO-0.303
 positively-0.261
 withd-0.251
 negatively-0.248
Token easy
Token ID2562
Feature activation-1.22e-02

Pos logit contributions
+0.071
Page+0.071
Bloom+0.070
mination+0.068
��+0.067
Neg logit contributions
 BDS-0.070
 NEO-0.062
 positively-0.053
 withd-0.051
ppelin-0.051
Token se
Token ID384
Feature activation+2.26e-01

Pos logit contributions
 BDS+0.035
 NEO+0.031
 positively+0.027
 withd+0.026
 negatively+0.026
Neg logit contributions
-0.036
Page-0.036
Bloom-0.036
mination-0.035
��-0.034
Tokengue
Token ID18701
Feature activation+3.88e-01

Pos logit contributions
+0.590
Page+0.587
Bloom+0.580
mination+0.570
��+0.548
Neg logit contributions
 BDS-0.738
 NEO-0.662
 positively-0.583
 withd-0.561
 negatively-0.556
Token into
Token ID656
Feature activation+5.66e-01

Pos logit contributions
Page+1.022
+1.015
Bloom+0.995
mination+0.977
��+0.942
Neg logit contributions
 BDS-1.253
 NEO-1.127
 positively-0.979
 withd-0.953
ppelin-0.943
Token the
Token ID262
Feature activation+2.38e-01

Pos logit contributions
+1.614
Page+1.608
Bloom+1.595
mination+1.531
��+1.510
Neg logit contributions
 BDS-1.688
 NEO-1.494
 positively-1.291
 withd-1.259
ppelin-1.254
Token metal
Token ID6147
Feature activation+5.53e-01

Pos logit contributions
+0.683
Page+0.683
Bloom+0.677
mination+0.651
��+0.647
Neg logit contributions
 BDS-0.707
 NEO-0.629
 positively-0.542
 withd-0.525
ppelin-0.524
Token portion
Token ID6903
Feature activation+1.87e-01

Pos logit contributions
+1.538
Page+1.528
Bloom+1.513
mination+1.464
��+1.445
Neg logit contributions
 BDS-1.698
 NEO-1.521
 positively-1.316
 withd-1.271
ppelin-1.256
Token of
Token ID286
Feature activation-3.94e-02

Pos logit contributions
+0.494
Page+0.492
Bloom+0.485
mination+0.471
��+0.462
Neg logit contributions
 BDS-0.602
 NEO-0.541
 positively-0.472
 withd-0.457
ppelin-0.453
Token this
Token ID428
Feature activation+4.90e-02

Pos logit contributions
 BDS+0.121
 NEO+0.108
 positively+0.094
 withd+0.090
 negatively+0.089
Neg logit contributions
-0.111
Page-0.110
Bloom-0.109
mination-0.106
��-0.104
Token at
Token ID379
Feature activation-1.73e-03

Pos logit contributions
 BDS+1.314
 NEO+1.172
 positively+1.024
 withd+0.978
 negatively+0.971
Neg logit contributions
-1.198
Page-1.187
Bloom-1.176
mination-1.144
��-1.127
Token the
Token ID262
Feature activation+1.67e-02

Pos logit contributions
 BDS+0.005
 NEO+0.005
 positively+0.004
 withd+0.004
ppelin+0.004
Neg logit contributions
-0.005
Page-0.005
Bloom-0.005
mination-0.005
��-0.005
Token off
Token ID572
Feature activation-1.84e-01

Pos logit contributions
+0.049
Page+0.049
Bloom+0.049
mination+0.047
��+0.047
Neg logit contributions
 BDS-0.048
 NEO-0.043
 positively-0.037
 withd-0.036
ppelin-0.035
Token-
Token ID12
Feature activation+1.42e-01

Pos logit contributions
 BDS+0.610
 NEO+0.549
 positively+0.485
 withd+0.466
 negatively+0.463
Neg logit contributions
-0.470
Page-0.465
Bloom-0.461
mination-0.447
��-0.439
Tokenseason
Token ID6230
Feature activation+2.99e-02

Pos logit contributions
+0.327
Page+0.324
Bloom+0.320
mination+0.309
��+0.303
Neg logit contributions
 BDS-0.508
 NEO-0.461
 positively-0.411
 withd-0.396
 negatively-0.394
Token additions
Token ID19885
Feature activation+5.20e-01

Pos logit contributions
+0.079
Page+0.078
Bloom+0.078
mination+0.075
��+0.074
Neg logit contributions
 BDS-0.096
 NEO-0.086
 positively-0.076
 withd-0.073
ppelin-0.073
Token Marc
Token ID13067
Feature activation+2.94e-02

Pos logit contributions
Page+1.362
+1.351
Bloom+1.331
��+1.259
mination+1.256
Neg logit contributions
 BDS-1.675
 NEO-1.485
 positively-1.306
ppelin-1.272
 withd-1.271
Token Ber
Token ID4312
Feature activation+7.29e-02

Pos logit contributions
+0.106
Page+0.106
Bloom+0.105
mination+0.102
��+0.102
Neg logit contributions
 BDS-0.066
 NEO-0.056
 positively-0.045
 withd-0.043
ppelin-0.043
Tokenge
Token ID469
Feature activation+1.40e-01

Pos logit contributions
Page+0.179
+0.179
Bloom+0.176
mination+0.171
Beast+0.164
Neg logit contributions
 BDS-0.250
 NEO-0.227
 positively-0.200
 withd-0.197
 negatively-0.191
Tokenvin
Token ID7114
Feature activation+2.80e-01

Pos logit contributions
+0.424
Page+0.422
Bloom+0.418
mination+0.407
��+0.398
Neg logit contributions
 BDS-0.402
 NEO-0.356
 positively-0.305
 withd-0.294
 negatively-0.288
Token has
Token ID468
Feature activation-5.69e-02

Pos logit contributions
Page+0.715
+0.714
Bloom+0.695
��+0.668
mination+0.664
Neg logit contributions
 BDS-0.926
 NEO-0.826
 positively-0.716
ppelin-0.704
 withd-0.698
Token more
Token ID517
Feature activation+2.34e-01

Pos logit contributions
+0.370
Page+0.367
Bloom+0.363
mination+0.353
��+0.348
Neg logit contributions
 BDS-0.451
 NEO-0.405
 positively-0.356
 withd-0.340
 negatively-0.339
Token.
Token ID13
Feature activation+1.34e-01

Pos logit contributions
+0.607
Page+0.602
Bloom+0.596
mination+0.578
��+0.568
Neg logit contributions
 BDS-0.769
 NEO-0.691
 positively-0.608
 withd-0.585
 negatively-0.580
Token Also
Token ID4418
Feature activation+1.97e-01

Pos logit contributions
+0.381
Page+0.379
Bloom+0.374
mination+0.365
��+0.357
Neg logit contributions
 BDS-0.409
 NEO-0.364
 positively-0.318
 withd-0.306
 negatively-0.302
Token,
Token ID11
Feature activation+1.68e-01

Pos logit contributions
+0.556
Page+0.553
Bloom+0.545
mination+0.531
��+0.523
Neg logit contributions
 BDS-0.604
 NEO-0.538
 positively-0.467
 withd-0.450
ppelin-0.444
Token as
Token ID355
Feature activation+1.09e-01

Pos logit contributions
+0.450
Page+0.448
Bloom+0.442
mination+0.430
��+0.422
Neg logit contributions
 BDS-0.538
 NEO-0.482
 positively-0.422
 withd-0.406
ppelin-0.402
Token before
Token ID878
Feature activation+6.45e-01

Pos logit contributions
+0.310
Page+0.308
Bloom+0.305
mination+0.296
��+0.292
Neg logit contributions
 BDS-0.333
 NEO-0.296
 positively-0.258
 withd-0.247
 negatively-0.245
Token,
Token ID11
Feature activation+2.44e-01

Pos logit contributions
Page+1.733
+1.728
Bloom+1.702
mination+1.649
��+1.632
Neg logit contributions
 BDS-2.043
 NEO-1.840
 positively-1.600
 withd-1.547
ppelin-1.525
Token the
Token ID262
Feature activation+9.57e-02

Pos logit contributions
+0.657
Page+0.653
Bloom+0.645
mination+0.627
��+0.615
Neg logit contributions
 BDS-0.773
 NEO-0.692
 positively-0.606
 withd-0.583
ppelin-0.578
Token grid
Token ID10706
Feature activation-1.99e-01

Pos logit contributions
+0.273
Page+0.271
Bloom+0.268
mination+0.261
��+0.257
Neg logit contributions
 BDS-0.290
 NEO-0.258
 positively-0.225
 withd-0.215
ppelin-0.213
Token automatically
Token ID6338
Feature activation+8.12e-02

Pos logit contributions
 BDS+0.645
 NEO+0.579
 positively+0.513
 negatively+0.488
 withd+0.487
Neg logit contributions
-0.529
Page-0.520
Bloom-0.516
mination-0.498
��-0.496
Token sizes
Token ID10620
Feature activation+1.60e-01

Pos logit contributions
+0.206
Page+0.205
Bloom+0.203
mination+0.197
��+0.191
Neg logit contributions
 BDS-0.270
 NEO-0.243
 positively-0.213
ppelin-0.206
 withd-0.205
Token January
Token ID3269
Feature activation+4.61e-01

Pos logit contributions
Page+2.020
+2.009
Bloom+1.993
mination+1.925
Beast+1.881
Neg logit contributions
 BDS-1.784
 NEO-1.571
 positively-1.356
 withd-1.340
ppelin-1.300
Token 15
Token ID1315
Feature activation+4.10e-01

Pos logit contributions
Page+1.190
+1.183
Bloom+1.166
mination+1.125
Beast+1.093
Neg logit contributions
 BDS-1.506
 NEO-1.357
 positively-1.212
 withd-1.204
 negatively-1.158
Token,
Token ID11
Feature activation+2.94e-01

Pos logit contributions
Page+1.089
+1.074
Bloom+1.065
mination+1.024
Range+0.998
Neg logit contributions
 BDS-1.321
 NEO-1.196
 withd-1.085
 positively-1.054
 negatively-1.011
Token 2010
Token ID3050
Feature activation+6.65e-01

Pos logit contributions
+0.889
Page+0.882
Bloom+0.874
mination+0.852
��+0.839
Neg logit contributions
 BDS-0.841
 NEO-0.743
 positively-0.640
 withd-0.609
 negatively-0.604
Token
Token ID198
Feature activation+9.02e-01

Pos logit contributions
Page+1.934
+1.890
Bloom+1.881
mination+1.816
Beast+1.775
Neg logit contributions
 BDS-1.968
 NEO-1.760
 withd-1.622
 positively-1.551
 negatively-1.480
Token
Token ID198
Feature activation+6.83e-01

Pos logit contributions
Page+2.516
Bloom+2.424
+2.423
mination+2.332
Range+2.300
Neg logit contributions
 BDS-2.755
 NEO-2.503
 withd-2.419
 positively-2.197
 negatively-2.112
TokenSubject
Token ID19776
Feature activation+3.25e-01

Pos logit contributions
Page+1.222
+1.190
Bloom+1.183
mination+1.115
Beast+1.080
Neg logit contributions
 BDS-2.830
 NEO-2.609
 withd-2.376
 positively-2.363
 negatively-2.288
Token:
Token ID25
Feature activation+1.13e-02

Pos logit contributions
Page+0.874
+0.855
Bloom+0.843
mination+0.817
Range+0.796
Neg logit contributions
 BDS-1.045
 NEO-0.942
 withd-0.861
 positively-0.825
 negatively-0.793
Token Que
Token ID4670
Feature activation-1.63e-01

Pos logit contributions
+0.033
Page+0.033
Bloom+0.032
mination+0.031
��+0.031
Neg logit contributions
 BDS-0.033
 NEO-0.030
 positively-0.025
 withd-0.025
ppelin-0.024
Tokener
Token ID263
Feature activation-1.57e-01

Pos logit contributions
 BDS+0.622
 NEO+0.566
 positively+0.511
 negatively+0.490
 withd+0.490
Neg logit contributions
-0.341
Page-0.332
Bloom-0.330
mination-0.316
��-0.311
Token magnet
Token ID19972
Feature activation-3.66e-02

Pos logit contributions
 BDS+0.476
 NEO+0.424
 positively+0.369
 withd+0.353
 negatively+0.350
Neg logit contributions
-0.448
Page-0.444
Bloom-0.440
mination-0.428
��-0.421
Token assured
Token ID13933
Feature activation+1.57e-01

Pos logit contributions
 BDS+0.138
 NEO+0.125
 positively+0.108
 withd+0.105
ppelin+0.104
Neg logit contributions
Page-0.112
-0.112
Bloom-0.110
mination-0.107
��-0.105
Token her
Token ID607
Feature activation+9.90e-02

Pos logit contributions
+0.391
Page+0.388
Bloom+0.384
mination+0.371
��+0.364
Neg logit contributions
 BDS-0.534
 NEO-0.482
 positively-0.427
 withd-0.411
 negatively-0.407
Token that
Token ID326
Feature activation+1.38e-01

Pos logit contributions
+0.256
Page+0.254
Bloom+0.251
mination+0.243
��+0.239
Neg logit contributions
 BDS-0.326
 NEO-0.293
 positively-0.258
 withd-0.249
ppelin-0.246
Token if
Token ID611
Feature activation+2.02e-01

Pos logit contributions
+0.351
Page+0.349
Bloom+0.344
mination+0.333
��+0.328
Neg logit contributions
 BDS-0.462
 NEO-0.416
 positively-0.367
 withd-0.354
ppelin-0.351
Token she
Token ID673
Feature activation+8.72e-02

Pos logit contributions
+0.468
Page+0.464
Bloom+0.458
mination+0.441
��+0.434
Neg logit contributions
 BDS-0.720
 NEO-0.653
 positively-0.582
 withd-0.563
ppelin-0.558
Token filled
Token ID5901
Feature activation+5.29e-01

Pos logit contributions
+0.216
Page+0.216
Bloom+0.212
mination+0.206
��+0.201
Neg logit contributions
 BDS-0.294
 NEO-0.266
 positively-0.233
 withd-0.226
ppelin-0.225
Token out
Token ID503
Feature activation+2.82e-01

Pos logit contributions
+1.376
Page+1.367
Bloom+1.350
mination+1.299
��+1.270
Neg logit contributions
 BDS-1.719
 NEO-1.550
 positively-1.361
 withd-1.322
ppelin-1.304
Token the
Token ID262
Feature activation+1.62e-01

Pos logit contributions
+0.771
Page+0.765
Bloom+0.760
mination+0.732
��+0.721
Neg logit contributions
 BDS-0.880
 NEO-0.788
 positively-0.687
 withd-0.667
ppelin-0.664
Token police
Token ID1644
Feature activation-9.83e-02

Pos logit contributions
+0.473
Page+0.470
Bloom+0.466
mination+0.453
��+0.445
Neg logit contributions
 BDS-0.474
 NEO-0.422
 positively-0.365
 withd-0.350
ppelin-0.349
Token report
Token ID989
Feature activation+2.11e-02

Pos logit contributions
 BDS+0.268
 NEO+0.236
 positively+0.201
 withd+0.191
ppelin+0.190
Neg logit contributions
-0.309
Page-0.307
Bloom-0.304
mination-0.297
��-0.292
Token,
Token ID11
Feature activation+4.13e-02

Pos logit contributions
+0.057
Page+0.057
Bloom+0.056
mination+0.054
��+0.053
Neg logit contributions
 BDS-0.067
 NEO-0.060
 positively-0.052
 withd-0.050
ppelin-0.050
Tokenville
Token ID4244
Feature activation-9.77e-02

Pos logit contributions
 BDS+1.041
 NEO+0.952
 positively+0.839
 withd+0.823
 negatively+0.801
Neg logit contributions
Page-0.759
-0.751
Bloom-0.739
mination-0.720
��-0.700
Token has
Token ID468
Feature activation-3.35e-02

Pos logit contributions
 BDS+0.363
 NEO+0.331
 positively+0.296
 withd+0.286
 negatively+0.285
Neg logit contributions
-0.212
Page-0.210
Bloom-0.207
mination-0.200
��-0.196
Token already
Token ID1541
Feature activation-1.28e-01

Pos logit contributions
 BDS+0.113
 NEO+0.102
 positively+0.090
 withd+0.087
 negatively+0.086
Neg logit contributions
-0.084
Page-0.083
Bloom-0.083
mination-0.080
��-0.079
Token broken
Token ID5445
Feature activation+2.76e-01

Pos logit contributions
 BDS+0.445
 NEO+0.403
 positively+0.355
 withd+0.344
ppelin+0.342
Neg logit contributions
-0.308
Page-0.306
Bloom-0.300
mination-0.292
��-0.287
Token up
Token ID510
Feature activation-4.49e-01

Pos logit contributions
+0.755
Page+0.751
Bloom+0.741
mination+0.720
��+0.704
Neg logit contributions
 BDS-0.862
 NEO-0.774
 positively-0.667
ppelin-0.646
 withd-0.643
Token that
Token ID326
Feature activation+3.87e-02

Pos logit contributions
 BDS+1.353
 NEO+1.204
 positively+1.045
 withd+0.999
 negatively+0.991
Neg logit contributions
-1.288
Page-1.277
Bloom-1.264
mination-1.230
��-1.212
Token trio
Token ID19886
Feature activation-3.55e-02

Pos logit contributions
+0.119
Page+0.118
Bloom+0.117
mination+0.114
��+0.113
Neg logit contributions
 BDS-0.108
 NEO-0.095
 positively-0.081
 withd-0.077
ppelin-0.077
Token.
Token ID13
Feature activation+3.42e-03

Pos logit contributions
 BDS+0.113
 NEO+0.101
 positively+0.088
 withd+0.085
ppelin+0.084
Neg logit contributions
-0.096
Page-0.095
Bloom-0.094
mination-0.091
��-0.090
Token L
Token ID406
Feature activation-3.03e-01

Pos logit contributions
+0.010
Page+0.010
Bloom+0.010
mination+0.010
��+0.010
Neg logit contributions
 BDS-0.010
 NEO-0.009
 positively-0.008
 withd-0.007
 negatively-0.007
Tokenadd
Token ID2860
Feature activation-1.49e-01

Pos logit contributions
 BDS+1.021
 NEO+0.919
 positively+0.814
 withd+0.781
 negatively+0.777
Neg logit contributions
-0.761
Page-0.753
Bloom-0.745
mination-0.722
��-0.711
Token played
Token ID2826
Feature activation-2.15e-01

Pos logit contributions
 BDS+0.516
 NEO+0.466
 positively+0.413
 withd+0.398
 negatively+0.395
Neg logit contributions
-0.361
Page-0.358
Bloom-0.354
mination-0.343
��-0.336
Token their
Token ID511
Feature activation-2.27e-01

Pos logit contributions
Page+0.059
+0.058
Bloom+0.058
mination+0.056
��+0.054
Neg logit contributions
 BDS-0.068
 NEO-0.060
 positively-0.052
 withd-0.052
ppelin-0.051
Token message
Token ID3275
Feature activation-1.35e-01

Pos logit contributions
 BDS+0.662
 NEO+0.586
 positively+0.508
 withd+0.481
 negatively+0.479
Neg logit contributions
-0.675
Page-0.668
Bloom-0.663
mination-0.646
��-0.637
Token through
Token ID832
Feature activation-5.77e-02

Pos logit contributions
 BDS+0.396
 NEO+0.351
 positively+0.305
 withd+0.290
 negatively+0.288
Neg logit contributions
-0.397
Page-0.393
Bloom-0.390
mination-0.380
��-0.374
Token.
Token ID764
Feature activation+9.35e-02

Pos logit contributions
 BDS+0.187
 NEO+0.168
 positively+0.147
 withd+0.143
ppelin+0.141
Neg logit contributions
-0.152
Page-0.151
Bloom-0.149
mination-0.144
��-0.141
Token
Token ID198
Feature activation+6.80e-02

Pos logit contributions
+0.258
Page+0.257
Bloom+0.254
mination+0.246
��+0.240
Neg logit contributions
 BDS-0.292
 NEO-0.260
 positively-0.229
 withd-0.221
 negatively-0.217
Token
Token ID198
Feature activation+7.35e-02

Pos logit contributions
Page+0.203
+0.201
Bloom+0.199
mination+0.192
Range+0.187
Neg logit contributions
 BDS-0.197
 NEO-0.176
 withd-0.156
 positively-0.153
 negatively-0.145
TokenThe
Token ID464
Feature activation-2.71e-02

Pos logit contributions
Page+0.180
+0.180
Bloom+0.178
mination+0.170
��+0.166
Neg logit contributions
 BDS-0.253
 NEO-0.228
 positively-0.203
 withd-0.197
 negatively-0.194
Token har
Token ID3971
Feature activation-2.38e-03

Pos logit contributions
 BDS+0.079
 NEO+0.070
 positively+0.060
 withd+0.058
ppelin+0.058
Neg logit contributions
Page-0.080
-0.080
Bloom-0.079
mination-0.077
��-0.076
Tokenbing
Token ID4623
Feature activation+9.57e-02

Pos logit contributions
 BDS+0.008
 NEO+0.007
 positively+0.006
 withd+0.006
 negatively+0.006
Neg logit contributions
-0.006
Page-0.006
Bloom-0.006
mination-0.006
��-0.006
Tokeners
Token ID364
Feature activation+1.06e-01

Pos logit contributions
+0.185
Page+0.183
Bloom+0.180
mination+0.173
��+0.169
Neg logit contributions
 BDS-0.378
 NEO-0.346
 positively-0.313
 withd-0.303
 negatively-0.301
Token of
Token ID286
Feature activation+6.94e-02

Pos logit contributions
+0.286
Page+0.286
Bloom+0.282
mination+0.273
��+0.266
Neg logit contributions
 BDS-0.337
 NEO-0.301
 positively-0.263
 withd-0.255
ppelin-0.252
Token.
Token ID13
Feature activation+1.71e-01

Pos logit contributions
 BDS+0.878
 NEO+0.782
 positively+0.683
 withd+0.652
 negatively+0.648
Neg logit contributions
-0.796
Page-0.788
Bloom-0.782
mination-0.760
��-0.750
Token Participant
Token ID29880
Feature activation-1.11e-01

Pos logit contributions
+0.491
Page+0.488
Bloom+0.483
mination+0.467
Beast+0.455
Neg logit contributions
 BDS-0.511
 NEO-0.457
 positively-0.402
 withd-0.385
 negatively-0.381
Token Anna
Token ID11735
Feature activation-1.84e-02

Pos logit contributions
 BDS+0.343
 NEO+0.306
 positively+0.267
 withd+0.256
 negatively+0.254
Neg logit contributions
-0.310
Page-0.308
Bloom-0.305
mination-0.296
��-0.291
Token B
Token ID347
Feature activation+9.81e-02

Pos logit contributions
 BDS+0.055
 NEO+0.049
 positively+0.042
 withd+0.040
 negatively+0.040
Neg logit contributions
-0.054
Page-0.053
Bloom-0.053
mination-0.051
��-0.051
Tokenerc
Token ID2798
Feature activation-7.13e-02

Pos logit contributions
+0.260
Page+0.259
Bloom+0.256
mination+0.248
��+0.244
Neg logit contributions
 BDS-0.317
 NEO-0.285
 positively-0.250
 withd-0.241
 negatively-0.238
Tokenowitz
Token ID20951
Feature activation+1.59e-01

Pos logit contributions
 BDS+0.204
 NEO+0.180
 positively+0.155
 withd+0.148
 negatively+0.147
Neg logit contributions
-0.216
Page-0.214
Bloom-0.212
mination-0.207
��-0.203
Token noted
Token ID4367
Feature activation-3.60e-02

Pos logit contributions
+0.423
Page+0.420
Bloom+0.416
mination+0.403
��+0.396
Neg logit contributions
 BDS-0.510
 NEO-0.457
 positively-0.400
 withd-0.385
 negatively-0.381
Token four
Token ID1440
Feature activation-6.75e-02

Pos logit contributions
 BDS+0.110
 NEO+0.098
 positively+0.085
 withd+0.082
ppelin+0.081
Neg logit contributions
-0.101
Page-0.101
Bloom-0.100
mination-0.097
��-0.095
Token primary
Token ID4165
Feature activation-1.49e-01

Pos logit contributions
 BDS+0.198
 NEO+0.175
 positively+0.152
 withd+0.144
 negatively+0.143
Neg logit contributions
-0.200
Page-0.198
Bloom-0.196
mination-0.191
��-0.188
Token factions
Token ID18783
Feature activation-2.29e-01

Pos logit contributions
 BDS+0.446
 NEO+0.396
 positively+0.344
 withd+0.329
 negatively+0.326
Neg logit contributions
-0.431
Page-0.428
Bloom-0.424
mination-0.412
��-0.405
Token at
Token ID379
Feature activation+2.76e-02

Pos logit contributions
 BDS+0.695
 NEO+0.622
 positively+0.536
 withd+0.516
ppelin+0.512
Neg logit contributions
-0.649
Page-0.648
Bloom-0.641
mination-0.622
��-0.611
Token part
Token ID636
Feature activation+1.25e-01

Pos logit contributions
+0.042
Page+0.042
Bloom+0.042
mination+0.040
��+0.040
Neg logit contributions
 BDS-0.036
 NEO-0.031
 positively-0.027
 withd-0.025
 negatively-0.025
Token of
Token ID286
Feature activation-8.00e-02

Pos logit contributions
+0.335
Page+0.332
Bloom+0.329
mination+0.319
��+0.314
Neg logit contributions
 BDS-0.401
 NEO-0.359
 positively-0.314
 withd-0.303
ppelin-0.301
Token his
Token ID465
Feature activation-9.03e-02

Pos logit contributions
 BDS+0.248
 NEO+0.221
 positively+0.193
 withd+0.184
 negatively+0.183
Neg logit contributions
-0.223
Page-0.221
Bloom-0.219
mination-0.213
��-0.210
Token appeal
Token ID5198
Feature activation+3.52e-01

Pos logit contributions
 BDS+0.249
 NEO+0.219
 positively+0.188
 withd+0.177
 negatively+0.176
Neg logit contributions
-0.283
Page-0.281
Bloom-0.279
mination-0.272
��-0.268
Token.
Token ID13
Feature activation+8.03e-02

Pos logit contributions
+0.916
Page+0.910
Bloom+0.894
mination+0.869
��+0.854
Neg logit contributions
 BDS-1.156
 NEO-1.040
 positively-0.907
 withd-0.880
ppelin-0.880
Token I
Token ID314
Feature activation+1.12e-01

Pos logit contributions
+0.231
Page+0.230
Bloom+0.227
mination+0.221
��+0.216
Neg logit contributions
 BDS-0.241
 NEO-0.214
 positively-0.188
 withd-0.180
 negatively-0.178
Token also
Token ID635
Feature activation+6.72e-02

Pos logit contributions
+0.308
Page+0.306
Bloom+0.302
mination+0.293
��+0.289
Neg logit contributions
 BDS-0.353
 NEO-0.316
 positively-0.273
 withd-0.264
ppelin-0.262
Token recognize
Token ID7564
Feature activation+1.08e-01

Pos logit contributions
+0.205
Page+0.205
Bloom+0.202
mination+0.196
��+0.194
Neg logit contributions
 BDS-0.189
 NEO-0.167
 positively-0.140
 withd-0.136
ppelin-0.135
Token that
Token ID326
Feature activation+1.44e-01

Pos logit contributions
+0.294
Page+0.292
Bloom+0.289
mination+0.279
��+0.275
Neg logit contributions
 BDS-0.339
 NEO-0.304
 positively-0.265
 withd-0.255
ppelin-0.252
Token his
Token ID465
Feature activation-1.27e-02

Pos logit contributions
+0.385
Page+0.383
Bloom+0.379
mination+0.367
��+0.361
Neg logit contributions
 BDS-0.462
 NEO-0.414
 positively-0.363
 withd-0.349
ppelin-0.346
Token campaign
Token ID1923
Feature activation+1.86e-01

Pos logit contributions
 BDS+0.038
 NEO+0.034
 positively+0.030
 withd+0.028
 negatively+0.028
Neg logit contributions
-0.037
Page-0.036
Bloom-0.036
mination-0.035
��-0.034
Token accepted
Token ID6292
Feature activation-4.01e-01

Pos logit contributions
 BDS+0.811
 NEO+0.733
 positively+0.651
 withd+0.626
 negatively+0.622
Neg logit contributions
-0.568
Page-0.563
Bloom-0.556
mination-0.539
��-0.529
Token a
Token ID257
Feature activation+2.18e-01

Pos logit contributions
 BDS+1.223
 NEO+1.088
 positively+0.949
 withd+0.906
 negatively+0.900
Neg logit contributions
-1.139
Page-1.129
Bloom-1.118
mination-1.088
��-1.072
Token two
Token ID734
Feature activation-6.29e-02

Pos logit contributions
+0.646
Page+0.642
Bloom+0.636
mination+0.618
��+0.611
Neg logit contributions
 BDS-0.631
 NEO-0.560
 positively-0.480
 withd-0.460
ppelin-0.460
Token-
Token ID12
Feature activation+7.10e-02

Pos logit contributions
 BDS+0.211
 NEO+0.189
 positively+0.168
 withd+0.161
 negatively+0.160
Neg logit contributions
-0.160
Page-0.158
Bloom-0.157
mination-0.152
��-0.149
Tokenyear
Token ID1941
Feature activation+3.12e-01

Pos logit contributions
+0.140
Page+0.137
Bloom+0.136
mination+0.130
��+0.128
Neg logit contributions
 BDS-0.279
 NEO-0.255
 positively-0.231
 withd-0.223
 negatively-0.221
Token deal
Token ID1730
Feature activation+1.77e-01

Pos logit contributions
+0.858
Page+0.853
Bloom+0.844
mination+0.820
��+0.807
Neg logit contributions
 BDS-0.969
 NEO-0.868
 positively-0.758
 withd-0.726
ppelin-0.724
Token at
Token ID379
Feature activation+5.86e-02

Pos logit contributions
+0.492
Page+0.490
Bloom+0.484
mination+0.471
��+0.463
Neg logit contributions
 BDS-0.549
 NEO-0.491
 positively-0.427
 withd-0.409
ppelin-0.407
Token the
Token ID262
Feature activation+1.57e-02

Pos logit contributions
+0.162
Page+0.160
Bloom+0.159
mination+0.154
��+0.152
Neg logit contributions
 BDS-0.183
 NEO-0.164
 positively-0.143
 withd-0.137
ppelin-0.136
Token minimum
Token ID5288
Feature activation+3.45e-01

Pos logit contributions
+0.046
Page+0.045
Bloom+0.045
mination+0.044
��+0.043
Neg logit contributions
 BDS-0.047
 NEO-0.042
 positively-0.036
 withd-0.035
ppelin-0.034
Token from
Token ID422
Feature activation+9.89e-02

Pos logit contributions
+0.945
Page+0.940
Bloom+0.929
mination+0.904
��+0.887
Neg logit contributions
 BDS-1.074
 NEO-0.966
 positively-0.840
ppelin-0.807
 withd-0.805
Token the
Token ID262
Feature activation-7.65e-02

Pos logit contributions
+0.266
Page+0.264
Bloom+0.261
mination+0.253
��+0.249
Neg logit contributions
 BDS-0.315
 NEO-0.282
 positively-0.248
 withd-0.238
 negatively-0.236
Token story
Token ID1621
Feature activation-1.12e-01

Pos logit contributions
+0.306
Page+0.305
Bloom+0.300
mination+0.293
��+0.288
Neg logit contributions
 BDS-0.333
 NEO-0.297
 positively-0.258
 withd-0.248
ppelin-0.246
Token,
Token ID11
Feature activation-1.09e-02

Pos logit contributions
 BDS+0.353
 NEO+0.315
 positively+0.276
 withd+0.264
 negatively+0.262
Neg logit contributions
-0.309
Page-0.306
Bloom-0.303
mination-0.294
��-0.289
Token using
Token ID1262
Feature activation+1.78e-01

Pos logit contributions
 BDS+0.033
 NEO+0.030
 positively+0.026
 withd+0.025
ppelin+0.024
Neg logit contributions
-0.031
Page-0.031
Bloom-0.030
mination-0.029
��-0.029
Token flashbacks
Token ID49555
Feature activation+3.14e-01

Pos logit contributions
+0.499
Page+0.499
Bloom+0.493
mination+0.476
��+0.467
Neg logit contributions
 BDS-0.541
 NEO-0.486
 positively-0.419
 withd-0.408
ppelin-0.406
Token to
Token ID284
Feature activation-1.33e-01

Pos logit contributions
Page+0.795
+0.792
Bloom+0.779
mination+0.750
��+0.731
Neg logit contributions
 BDS-1.040
 NEO-0.942
 positively-0.817
 withd-0.797
ppelin-0.797
Token do
Token ID466
Feature activation-1.73e-01

Pos logit contributions
 BDS+0.374
 NEO+0.331
 positively+0.280
 withd+0.271
ppelin+0.269
Neg logit contributions
-0.405
Page-0.403
Bloom-0.398
mination-0.386
��-0.379
Token so
Token ID523
Feature activation-2.31e-01

Pos logit contributions
 BDS+0.496
 NEO+0.439
 positively+0.378
 withd+0.360
 negatively+0.357
Neg logit contributions
-0.521
Page-0.517
Bloom-0.513
mination-0.499
��-0.492
Token
Token ID784
Feature activation-1.26e-01

Pos logit contributions
 BDS+0.744
 NEO+0.667
 positively+0.585
 withd+0.561
 negatively+0.557
Neg logit contributions
-0.617
Page-0.612
Bloom-0.605
mination-0.587
��-0.578
Token not
Token ID407
Feature activation+1.05e-01

Pos logit contributions
 BDS+0.391
 NEO+0.349
 positively+0.305
 withd+0.292
 negatively+0.289
Neg logit contributions
-0.353
Page-0.350
Bloom-0.346
mination-0.336
��-0.331
Token unlike
Token ID5023
Feature activation+9.70e-03

Pos logit contributions
+0.277
Page+0.275
Bloom+0.272
mination+0.264
��+0.259
Neg logit contributions
 BDS-0.339
 NEO-0.305
 positively-0.268
 withd-0.257
ppelin-0.255
Token Patty
Token ID42417
Feature activation+3.96e-01

Pos logit contributions
+0.027
Page+0.027
Bloom+0.027
mination+0.026
��+0.026
Neg logit contributions
 BDS-0.029
 NEO-0.026
 positively-0.023
 withd-0.022
ppelin-0.022
Token promised
Token ID8072
Feature activation-2.75e-01

Pos logit contributions
 BDS+0.501
 NEO+0.448
 positively+0.396
 withd+0.378
 negatively+0.377
Neg logit contributions
-0.416
Page-0.412
Bloom-0.409
mination-0.397
��-0.390
Token to
Token ID284
Feature activation-2.63e-02

Pos logit contributions
 BDS+0.902
 NEO+0.809
 positively+0.716
 negatively+0.682
 withd+0.682
Neg logit contributions
-0.719
Page-0.710
Bloom-0.703
mination-0.684
��-0.672
Token build
Token ID1382
Feature activation-1.39e-02

Pos logit contributions
 BDS+0.075
 NEO+0.066
 positively+0.056
 withd+0.054
ppelin+0.054
Neg logit contributions
-0.079
Page-0.079
Bloom-0.078
mination-0.075
��-0.075
Token a
Token ID257
Feature activation-2.11e-01

Pos logit contributions
 BDS+0.042
 NEO+0.037
 positively+0.032
 withd+0.031
ppelin+0.031
Neg logit contributions
Page-0.040
-0.039
Bloom-0.039
mination-0.038
��-0.037
Token stronger
Token ID7387
Feature activation-4.33e-01

Pos logit contributions
 BDS+0.559
 NEO+0.489
 positively+0.415
 withd+0.393
 negatively+0.389
Neg logit contributions
-0.682
Page-0.677
Bloom-0.672
mination-0.655
��-0.647
Token relationship
Token ID2776
Feature activation-2.62e-02

Pos logit contributions
 BDS+1.237
 NEO+1.099
 positively+0.945
 withd+0.910
ppelin+0.894
Neg logit contributions
-1.295
Page-1.292
Bloom-1.278
mination-1.243
��-1.223
Token with
Token ID351
Feature activation-1.33e-01

Pos logit contributions
 BDS+0.080
 NEO+0.071
 positively+0.062
 withd+0.059
 negatively+0.059
Neg logit contributions
-0.074
Page-0.074
Bloom-0.073
mination-0.071
��-0.070
Token ab
Token ID450
Feature activation-1.39e-01

Pos logit contributions
 BDS+0.420
 NEO+0.377
 positively+0.329
 withd+0.319
ppelin+0.315
Neg logit contributions
-0.357
Page-0.357
Bloom-0.352
mination-0.342
��-0.335
Tokenorig
Token ID11612
Feature activation-1.96e-01

Pos logit contributions
 BDS+0.462
 NEO+0.416
 positively+0.367
 withd+0.356
 negatively+0.350
Neg logit contributions
-0.353
Page-0.352
Bloom-0.347
mination-0.338
��-0.329
Tokeninals
Token ID6897
Feature activation-7.93e-02

Pos logit contributions
 BDS+0.684
 NEO+0.617
 positively+0.547
 withd+0.531
 negatively+0.523
Neg logit contributions
Page-0.471
-0.471
Bloom-0.463
mination-0.453
��-0.437
Token.
Token ID13
Feature activation-5.65e-02

Pos logit contributions
 BDS+0.252
 NEO+0.226
 positively+0.197
 withd+0.189
ppelin+0.188
Neg logit contributions
-0.214
Page-0.213
Bloom-0.210
mination-0.204
��-0.200
Tokenhen
Token ID831
Feature activation-9.76e-02

Pos logit contributions
+0.395
Page+0.393
Bloom+0.388
mination+0.377
��+0.366
Neg logit contributions
 BDS-0.502
 NEO-0.451
 positively-0.397
 withd-0.384
 negatively-0.379
Tokenando
Token ID25440
Feature activation-1.47e-01

Pos logit contributions
 BDS+0.293
 NEO+0.260
 positively+0.226
 withd+0.214
 negatively+0.214
Neg logit contributions
-0.282
Page-0.279
Bloom-0.277
mination-0.269
��-0.267
Tokenah
Token ID993
Feature activation-3.31e-02

Pos logit contributions
 BDS+0.576
 NEO+0.526
 positively+0.476
ppelin+0.457
 negatively+0.457
Neg logit contributions
-0.293
Page-0.286
Bloom-0.284
mination-0.273
��-0.271
Token)
Token ID8
Feature activation-1.01e-01

Pos logit contributions
 BDS+0.110
 NEO+0.099
 positively+0.088
 withd+0.084
 negatively+0.084
Neg logit contributions
-0.085
Page-0.084
Bloom-0.083
mination-0.080
��-0.079
Token speaks
Token ID9209
Feature activation-2.26e-01

Pos logit contributions
 BDS+0.345
 NEO+0.311
 positively+0.275
 withd+0.265
ppelin+0.263
Neg logit contributions
-0.249
Page-0.248
Bloom-0.244
mination-0.237
��-0.232
Token on
Token ID319
Feature activation-6.12e-02

Pos logit contributions
 BDS+0.684
 NEO+0.609
 positively+0.528
 withd+0.506
ppelin+0.501
Neg logit contributions
-0.646
Page-0.641
Bloom-0.635
mination-0.617
��-0.608
Token the
Token ID262
Feature activation-8.29e-02

Pos logit contributions
 BDS+0.204
 NEO+0.185
 positively+0.163
 withd+0.158
ppelin+0.156
Neg logit contributions
Page-0.154
-0.153
Bloom-0.152
mination-0.145
��-0.144
Token floor
Token ID4314
Feature activation-3.96e-02

Pos logit contributions
 BDS+0.268
 NEO+0.242
 positively+0.212
 withd+0.205
ppelin+0.203
Neg logit contributions
Page-0.217
-0.216
Bloom-0.213
mination-0.206
��-0.204
Token of
Token ID286
Feature activation-1.72e-01

Pos logit contributions
 BDS+0.125
 NEO+0.112
 positively+0.097
 withd+0.094
ppelin+0.093
Neg logit contributions
-0.107
Page-0.107
Bloom-0.106
mination-0.103
��-0.101
Token the
Token ID262
Feature activation-9.58e-02

Pos logit contributions
 BDS+0.530
 NEO+0.474
 positively+0.412
 withd+0.397
ppelin+0.392
Neg logit contributions
-0.481
Page-0.480
Bloom-0.475
mination-0.460
��-0.454
Token House
Token ID2097
Feature activation+1.51e-01

Pos logit contributions
 BDS+0.310
 NEO+0.279
 positively+0.245
 withd+0.235
 negatively+0.233
Neg logit contributions
-0.253
Page-0.251
Bloom-0.248
mination-0.241
��-0.237
Token prescriptions
Token ID34092
Feature activation-9.23e-02

Pos logit contributions
 BDS+1.542
 NEO+1.361
 positively+1.183
 negatively+1.120
 withd+1.111
Neg logit contributions
-1.539
Page-1.522
Bloom-1.511
mination-1.474
��-1.465
Token are
Token ID389
Feature activation+3.33e-02

Pos logit contributions
 BDS+0.287
 NEO+0.257
 positively+0.222
 withd+0.214
ppelin+0.212
Neg logit contributions
Page-0.254
-0.254
Bloom-0.250
mination-0.243
��-0.238
Token dangerous
Token ID4923
Feature activation-1.87e-01

Pos logit contributions
+0.097
Page+0.097
Bloom+0.096
mination+0.093
��+0.092
Neg logit contributions
 BDS-0.098
 NEO-0.087
 positively-0.075
 withd-0.073
ppelin-0.072
Token.
Token ID13
Feature activation-6.09e-02

Pos logit contributions
 BDS+0.592
 NEO+0.530
 positively+0.464
 withd+0.445
ppelin+0.441
Neg logit contributions
-0.510
Page-0.505
Bloom-0.500
mination-0.486
��-0.478
Token A
Token ID317
Feature activation-1.46e-01

Pos logit contributions
 BDS+0.186
 NEO+0.165
 positively+0.144
 withd+0.138
 negatively+0.137
Neg logit contributions
-0.173
Page-0.171
Bloom-0.170
mination-0.165
��-0.162
Token run
Token ID1057
Feature activation-2.78e-01

Pos logit contributions
 BDS+0.408
 NEO+0.359
 positively+0.308
 withd+0.293
 negatively+0.290
Neg logit contributions
-0.453
Page-0.449
Bloom-0.445
mination-0.434
��-0.428
Token-
Token ID12
Feature activation+2.80e-02

Pos logit contributions
 BDS+0.928
 NEO+0.835
 positively+0.738
 withd+0.708
 negatively+0.704
Neg logit contributions
-0.710
Page-0.703
Bloom-0.696
mination-0.675
��-0.664
Tokenup
Token ID929
Feature activation-3.25e-01

Pos logit contributions
Page+0.069
+0.069
Bloom+0.068
mination+0.066
��+0.064
Neg logit contributions
 BDS-0.096
 NEO-0.086
 positively-0.076
 withd-0.073
 negatively-0.073
Token in
Token ID287
Feature activation-3.23e-02

Pos logit contributions
 BDS+0.987
 NEO+0.878
 positively+0.767
 withd+0.729
 negatively+0.728
Neg logit contributions
-0.929
Page-0.919
Bloom-0.912
mination-0.889
��-0.878
Token prescribing
Token ID39973
Feature activation+3.85e-02

Pos logit contributions
 BDS+0.083
 NEO+0.072
 positively+0.060
 withd+0.058
ppelin+0.057
Neg logit contributions
-0.107
Page-0.106
Bloom-0.105
mination-0.102
��-0.101
Token from
Token ID422
Feature activation-8.76e-02

Pos logit contributions
+0.098
Page+0.098
Bloom+0.097
mination+0.094
��+0.091
Neg logit contributions
 BDS-0.127
 NEO-0.114
 positively-0.100
 withd-0.097
ppelin-0.097
Token to
Token ID284
Feature activation-2.15e-01

Pos logit contributions
 BDS+0.177
 NEO+0.158
 positively+0.138
 withd+0.133
ppelin+0.131
Neg logit contributions
-0.156
Page-0.155
Bloom-0.153
mination-0.148
��-0.146
Token some
Token ID617
Feature activation+8.50e-02

Pos logit contributions
 BDS+0.686
 NEO+0.614
 positively+0.540
 withd+0.516
 negatively+0.513
Neg logit contributions
-0.578
Page-0.573
Bloom-0.568
mination-0.554
��-0.544
Token reporters
Token ID7638
Feature activation-2.16e-01

Pos logit contributions
+0.231
Page+0.229
Bloom+0.227
mination+0.221
��+0.217
Neg logit contributions
 BDS-0.269
 NEO-0.241
 positively-0.211
 withd-0.202
 negatively-0.201
Token that
Token ID326
Feature activation-1.17e-01

Pos logit contributions
 BDS+0.665
 NEO+0.593
 positively+0.517
 withd+0.495
 negatively+0.490
Neg logit contributions
-0.606
Page-0.602
Bloom-0.595
mination-0.579
��-0.570
Token he
Token ID339
Feature activation-4.53e-02

Pos logit contributions
 BDS+0.391
 NEO+0.352
 positively+0.311
 withd+0.298
 negatively+0.297
Neg logit contributions
-0.298
Page-0.296
Bloom-0.293
mination-0.284
��-0.279
Token was
Token ID373
Feature activation-1.19e-01

Pos logit contributions
 BDS+0.158
 NEO+0.143
 positively+0.126
 withd+0.122
ppelin+0.121
Neg logit contributions
-0.109
Page-0.108
Bloom-0.106
mination-0.103
��-0.101
Token in
Token ID287
Feature activation-6.40e-02

Pos logit contributions
 BDS+0.348
 NEO+0.308
 positively+0.265
 withd+0.254
ppelin+0.252
Neg logit contributions
-0.354
Page-0.352
Bloom-0.348
mination-0.338
��-0.334
Token New
Token ID968
Feature activation+7.93e-02

Pos logit contributions
 BDS+0.198
 NEO+0.177
 positively+0.154
 withd+0.148
ppelin+0.147
Neg logit contributions
-0.178
Page-0.177
Bloom-0.175
mination-0.169
��-0.167
Token York
Token ID1971
Feature activation-1.07e-01

Pos logit contributions
+0.193
Page+0.191
Bloom+0.189
mination+0.182
��+0.179
Neg logit contributions
 BDS-0.274
 NEO-0.248
 positively-0.220
 withd-0.212
 negatively-0.211
Token City
Token ID2254
Feature activation-2.14e-03

Pos logit contributions
 BDS+0.352
 NEO+0.317
 positively+0.279
 withd+0.267
 negatively+0.266
Neg logit contributions
-0.279
Page-0.277
Bloom-0.273
mination-0.265
��-0.261
Token,
Token ID11
Feature activation-1.17e-01

Pos logit contributions
 BDS+0.007
 NEO+0.006
 positively+0.005
 withd+0.005
ppelin+0.005
Neg logit contributions
-0.006
Page-0.006
Bloom-0.006
mination-0.005
��-0.005
Token obsolete
Token ID26533
Feature activation-2.35e-01

Pos logit contributions
+0.090
Page+0.090
Bloom+0.089
mination+0.086
��+0.085
Neg logit contributions
 BDS-0.078
 NEO-0.069
 positively-0.057
 withd-0.057
ppelin-0.056
Token.
Token ID13
Feature activation-1.37e-01

Pos logit contributions
 BDS+0.759
 NEO+0.680
 positively+0.597
 withd+0.572
ppelin+0.568
Neg logit contributions
-0.626
Page-0.621
Bloom-0.614
mination-0.595
��-0.586
Token This
Token ID770
Feature activation-2.39e-01

Pos logit contributions
 BDS+0.407
 NEO+0.361
 positively+0.312
 withd+0.297
 negatively+0.296
Neg logit contributions
-0.399
Page-0.395
Bloom-0.392
mination-0.382
��-0.377
Token does
Token ID857
Feature activation-2.14e-01

Pos logit contributions
 BDS+0.765
 NEO+0.685
 positively+0.601
 withd+0.578
ppelin+0.572
Neg logit contributions
-0.640
Page-0.636
Bloom-0.629
mination-0.611
��-0.600
Token not
Token ID407
Feature activation-2.01e-01

Pos logit contributions
 BDS+0.615
 NEO+0.544
 positively+0.465
 withd+0.447
ppelin+0.442
Neg logit contributions
-0.643
Page-0.638
Bloom-0.633
mination-0.617
��-0.607
Token mean
Token ID1612
Feature activation-6.77e-01

Pos logit contributions
 BDS+0.621
 NEO+0.557
 positively+0.473
 withd+0.466
ppelin+0.463
Neg logit contributions
-0.550
Page-0.550
Bloom-0.544
mination-0.530
��-0.517
Token that
Token ID326
Feature activation-1.81e-01

Pos logit contributions
 BDS+2.033
 NEO+1.810
 positively+1.583
 negatively+1.502
 withd+1.484
Neg logit contributions
-1.965
Page-1.940
Bloom-1.919
mination-1.874
��-1.857
Token older
Token ID4697
Feature activation+2.67e-01

Pos logit contributions
 BDS+0.564
 NEO+0.504
 positively+0.441
 withd+0.421
 negatively+0.418
Neg logit contributions
-0.502
Page-0.498
Bloom-0.493
mination-0.480
��-0.472
Token teachers
Token ID7799
Feature activation+1.09e-01

Pos logit contributions
+0.679
Page+0.675
Bloom+0.667
mination+0.647
��+0.632
Neg logit contributions
 BDS-0.885
 NEO-0.796
 positively-0.701
 withd-0.679
ppelin-0.673
Token who
Token ID508
Feature activation-1.10e-01

Pos logit contributions
Page+0.283
+0.282
Bloom+0.278
mination+0.271
��+0.262
Neg logit contributions
 BDS-0.355
 NEO-0.320
 positively-0.275
 withd-0.269
ppelin-0.267
Token remain
Token ID3520
Feature activation-1.09e-01

Pos logit contributions
 BDS+0.337
 NEO+0.300
 positively+0.258
 withd+0.249
ppelin+0.247
Neg logit contributions
-0.310
Page-0.309
Bloom-0.305
mination-0.297
��-0.292
Token in
Token ID287
Feature activation-2.01e-02

Pos logit contributions
 BDS+0.185
 NEO+0.165
 positively+0.143
 withd+0.137
ppelin+0.137
Neg logit contributions
-0.164
Page-0.163
Bloom-0.161
mination-0.156
��-0.154
Token Philly
Token ID30029
Feature activation-2.15e-01

Pos logit contributions
 BDS+0.060
 NEO+0.053
 positively+0.046
 withd+0.044
ppelin+0.044
Neg logit contributions
-0.058
Page-0.058
Bloom-0.058
mination-0.056
��-0.055
Token mirror
Token ID10162
Feature activation+1.82e-01

Pos logit contributions
 BDS+0.667
 NEO+0.595
 positively+0.520
 withd+0.497
 negatively+0.493
Neg logit contributions
-0.598
Page-0.593
Bloom-0.587
mination-0.570
��-0.561
Tokening
Token ID278
Feature activation-1.97e-03

Pos logit contributions
+0.483
Page+0.480
Bloom+0.474
mination+0.459
��+0.451
Neg logit contributions
 BDS-0.586
 NEO-0.525
 positively-0.458
 withd-0.445
ppelin-0.440
Token certain
Token ID1728
Feature activation-1.17e-01

Pos logit contributions
 BDS+0.006
 NEO+0.005
 positively+0.005
 withd+0.004
ppelin+0.004
Neg logit contributions
-0.006
Page-0.006
Bloom-0.006
mination-0.005
��-0.005
Token attributes
Token ID12608
Feature activation-5.93e-01

Pos logit contributions
 BDS+0.354
 NEO+0.314
 positively+0.274
 withd+0.262
ppelin+0.260
Neg logit contributions
-0.334
Page-0.331
Bloom-0.328
mination-0.319
��-0.313
Token of
Token ID286
Feature activation-1.30e-01

Pos logit contributions
 BDS+1.836
 NEO+1.640
 positively+1.435
 withd+1.366
 negatively+1.362
Neg logit contributions
-1.658
Page-1.641
Bloom-1.627
mination-1.585
��-1.562
Token black
Token ID2042
Feature activation+1.81e-01

Pos logit contributions
 BDS+0.393
 NEO+0.350
 positively+0.304
 withd+0.294
ppelin+0.290
Neg logit contributions
-0.368
Page-0.366
Bloom-0.362
mination-0.351
��-0.345
Token speech
Token ID4046
Feature activation+9.02e-02

Pos logit contributions
+0.506
Page+0.505
Bloom+0.496
mination+0.483
��+0.473
Neg logit contributions
 BDS-0.556
 NEO-0.495
 positively-0.430
 withd-0.418
ppelin-0.412
Token.
Token ID13
Feature activation-1.89e-02

Pos logit contributions
+0.247
Page+0.245
Bloom+0.242
mination+0.235
��+0.232
Neg logit contributions
 BDS-0.284
 NEO-0.253
 positively-0.221
 withd-0.213
ppelin-0.211
Token
Token ID198
Feature activation-3.25e-02

Pos logit contributions
 BDS+0.057
 NEO+0.051
 positively+0.044
 withd+0.042
 negatively+0.042
Neg logit contributions
-0.054
Page-0.054
Bloom-0.053
mination-0.052
��-0.051
Token"
Token ID1
Feature activation+7.79e-03

Pos logit contributions
+0.035
Page+0.035
Bloom+0.035
mination+0.034
��+0.033
Neg logit contributions
 BDS-0.047
 NEO-0.043
 positively-0.038
 withd-0.036
 negatively-0.036
Token for
Token ID329
Feature activation-6.76e-02

Pos logit contributions
+0.021
Page+0.021
Bloom+0.021
mination+0.020
��+0.020
Neg logit contributions
 BDS-0.025
 NEO-0.022
 positively-0.019
 withd-0.019
ppelin-0.018
Token short
Token ID1790
Feature activation+4.10e-02

Pos logit contributions
 BDS+0.191
 NEO+0.168
 positively+0.144
 withd+0.138
ppelin+0.137
Neg logit contributions
-0.206
Page-0.205
Bloom-0.203
mination-0.198
��-0.195
Token.
Token ID13
Feature activation-1.20e-01

Pos logit contributions
+0.109
Page+0.108
Bloom+0.106
mination+0.103
��+0.101
Neg logit contributions
 BDS-0.132
 NEO-0.119
 positively-0.104
 withd-0.100
ppelin-0.099
Token Just
Token ID2329
Feature activation-1.69e-01

Pos logit contributions
 BDS+0.355
 NEO+0.315
 positively+0.273
 withd+0.262
 negatively+0.259
Neg logit contributions
-0.350
Page-0.349
Bloom-0.345
mination-0.335
��-0.328
Token remember
Token ID3505
Feature activation-7.12e-01

Pos logit contributions
 BDS+0.513
 NEO+0.456
 positively+0.395
 withd+0.379
ppelin+0.376
Neg logit contributions
-0.482
Page-0.478
Bloom-0.472
mination-0.460
��-0.451
Token,
Token ID11
Feature activation-8.13e-02

Pos logit contributions
 BDS+2.143
 NEO+1.919
 positively+1.676
 negatively+1.578
 withd+1.568
Neg logit contributions
-2.040
Page-1.996
Bloom-1.990
mination-1.957
��-1.941
Token it
Token ID340
Feature activation-7.05e-02

Pos logit contributions
 BDS+0.242
 NEO+0.215
 positively+0.187
 withd+0.178
 negatively+0.177
Neg logit contributions
-0.237
Page-0.234
Bloom-0.233
mination-0.226
��-0.224
Token's
Token ID338
Feature activation+4.48e-02

Pos logit contributions
 BDS+0.251
 NEO+0.228
 positively+0.203
 withd+0.196
ppelin+0.194
Neg logit contributions
-0.163
Page-0.162
Bloom-0.160
mination-0.155
��-0.151
Token enough
Token ID1576
Feature activation-1.18e-01

Pos logit contributions
+0.125
Page+0.124
Bloom+0.123
mination+0.120
��+0.117
Neg logit contributions
 BDS-0.138
 NEO-0.123
 positively-0.107
 withd-0.103
ppelin-0.102
Token "
Token ID366
Feature activation+9.20e-02

Pos logit contributions
 BDS+0.394
 NEO+0.355
 positively+0.314
 withd+0.300
 negatively+0.299
Neg logit contributions
-0.303
Page-0.299
Bloom-0.297
mination-0.288
��-0.284
Token.
Token ID764
Feature activation+2.30e-01

Pos logit contributions
 BDS+1.883
 NEO+1.708
 positively+1.517
 withd+1.456
 negatively+1.447
Neg logit contributions
-1.384
Page-1.350
Bloom-1.345
mination-1.300
��-1.287
Tokenclone
Token ID21018
Feature activation+7.66e-03

Pos logit contributions
+0.603
Page+0.596
Bloom+0.591
mination+0.573
��+0.564
Neg logit contributions
 BDS-0.751
 NEO-0.674
 positively-0.594
 withd-0.569
 negatively-0.565
Token (),
Token ID29994
Feature activation+2.82e-01

Pos logit contributions
+0.018
Page+0.018
Bloom+0.018
mination+0.017
��+0.017
Neg logit contributions
 BDS-0.027
 NEO-0.025
 positively-0.022
 withd-0.021
 negatively-0.021
Token pieces
Token ID5207
Feature activation+3.01e-02

Pos logit contributions
+0.892
Page+0.885
Bloom+0.878
mination+0.856
��+0.845
Neg logit contributions
 BDS-0.770
 NEO-0.676
 positively-0.578
 withd-0.548
 negatively-0.543
Token :
Token ID1058
Feature activation-1.24e-01

Pos logit contributions
+0.076
Page+0.075
Bloom+0.074
mination+0.072
��+0.071
Neg logit contributions
 BDS-0.101
 NEO-0.091
 positively-0.081
 withd-0.077
 negatively-0.077
Token self
Token ID2116
Feature activation-6.84e-01

Pos logit contributions
 BDS+0.385
 NEO+0.345
 positively+0.301
 withd+0.286
 negatively+0.285
Neg logit contributions
-0.350
Page-0.344
Bloom-0.343
mination-0.334
��-0.329
Token.
Token ID764
Feature activation+1.64e-01

Pos logit contributions
 BDS+2.318
 NEO+2.100
 positively+1.870
 withd+1.791
 negatively+1.783
Neg logit contributions
-1.705
Page-1.665
Bloom-1.655
mination-1.609
��-1.580
Tokenpieces
Token ID34154
Feature activation-7.30e-02

Pos logit contributions
+0.403
Page+0.399
Bloom+0.395
mination+0.382
��+0.375
Neg logit contributions
 BDS-0.561
 NEO-0.506
 positively-0.449
 withd-0.432
 negatively-0.429
Token.
Token ID764
Feature activation+2.77e-02

Pos logit contributions
 BDS+0.243
 NEO+0.219
 positively+0.194
 withd+0.186
 negatively+0.185
Neg logit contributions
-0.186
Page-0.184
Bloom-0.182
mination-0.177
��-0.174
Tokenclone
Token ID21018
Feature activation+3.43e-02

Pos logit contributions
+0.085
Page+0.085
Bloom+0.084
mination+0.082
��+0.081
Neg logit contributions
 BDS-0.078
 NEO-0.069
 positively-0.059
 withd-0.056
 negatively-0.056
Token (),
Token ID29994
Feature activation+1.90e-01

Pos logit contributions
+0.086
Page+0.085
Bloom+0.084
mination+0.081
��+0.080
Neg logit contributions
 BDS-0.116
 NEO-0.105
 positively-0.093
 withd-0.089
 negatively-0.089
Token by
Token ID416
Feature activation-1.79e-01

Pos logit contributions
 BDS+1.066
 NEO+0.949
 positively+0.830
 withd+0.791
 negatively+0.787
Neg logit contributions
-0.979
Page-0.970
Bloom-0.962
mination-0.935
��-0.922
Token President
Token ID1992
Feature activation-2.16e-01

Pos logit contributions
 BDS+0.630
 NEO+0.571
 positively+0.508
 withd+0.491
ppelin+0.486
Neg logit contributions
-0.423
Page-0.420
Bloom-0.415
mination-0.400
��-0.393
Token Barack
Token ID8732
Feature activation-1.01e-01

Pos logit contributions
 BDS+0.719
 NEO+0.648
 positively+0.572
 withd+0.550
 negatively+0.546
Neg logit contributions
-0.550
Page-0.546
Bloom-0.540
mination-0.523
��-0.514
Token Obama
Token ID2486
Feature activation-1.92e-01

Pos logit contributions
 BDS+0.372
 NEO+0.340
 positively+0.305
 withd+0.301
ppelin+0.293
Neg logit contributions
-0.219
Page-0.218
Bloom-0.214
mination-0.205
��-0.198
Token,
Token ID11
Feature activation-2.90e-01

Pos logit contributions
 BDS+0.563
 NEO+0.499
 positively+0.431
 withd+0.412
 negatively+0.408
Neg logit contributions
-0.564
Page-0.561
Bloom-0.555
mination-0.540
��-0.532
Token CNN
Token ID8100
Feature activation-5.78e-01

Pos logit contributions
 BDS+0.910
 NEO+0.814
 positively+0.712
 withd+0.682
 negatively+0.677
Neg logit contributions
-0.794
Page-0.787
Bloom-0.780
mination-0.757
��-0.745
Token confirmed
Token ID4999
Feature activation+3.65e-02

Pos logit contributions
 BDS+2.124
 NEO+1.933
 positively+1.745
 withd+1.679
 negatively+1.670
Neg logit contributions
-1.274
Page-1.239
Bloom-1.237
mination-1.194
��-1.179
Token from
Token ID422
Feature activation-1.46e-01

Pos logit contributions
+0.103
Page+0.103
Bloom+0.102
mination+0.098
��+0.097
Neg logit contributions
 BDS-0.111
 NEO-0.099
 positively-0.085
 withd-0.083
ppelin-0.082
Token several
Token ID1811
Feature activation-2.25e-01

Pos logit contributions
 BDS+0.462
 NEO+0.415
 positively+0.362
 withd+0.352
ppelin+0.346
Neg logit contributions
-0.393
Page-0.391
Bloom-0.386
mination-0.375
��-0.367
Token sources
Token ID4237
Feature activation-3.16e-01

Pos logit contributions
 BDS+0.694
 NEO+0.619
 positively+0.540
 withd+0.519
ppelin+0.513
Neg logit contributions
-0.629
Page-0.625
Bloom-0.619
mination-0.602
��-0.591
Token.
Token ID13
Feature activation-1.25e-01

Pos logit contributions
 BDS+0.988
 NEO+0.886
 positively+0.763
 withd+0.742
ppelin+0.736
Neg logit contributions
Page-0.864
-0.864
Bloom-0.854
mination-0.830
��-0.808