TOP ACTIVATIONS
MAX = 2.991

, Mikhail Al "] [ Result " 1 / 2 -
re i Profile Blog Joined October 2002 United States
www . bel fast te legraph . co . uk /
K ada ver BB Profile Blog Joined June 2009 Germany 247
word by Patrick J . Buchanan , and critical essays by
Dark horse Profile Blog Joined December 2011 United States
Big Fan Profile Blog Joined December 2010 TL AD
J aaaa asper Profile Blog Joined April 2012 United States
Obama s 2008 convention speech and went unp un
www . bel fast te legraph . co . uk /
TB one - Profile Blog Joined November 2010 United States
g its amb ry / status / 9 175 669 15
Flag Reply 55 Recommend Recommend Share this
"] [ Site " W ijk a an Z ee "]
Share this comment on Facebook Share this comment on Twitter
Share this comment on Facebook Share this comment on Twitter
Share this comment on Facebook Share this comment on Twitter
by corporate America and the Wall Street cabal is their complete
Master of Dal K Profile Blog Joined June 2012 Canada 16
iser K ier an Profile Blog Joined August 2011 United States

MIN ACTIVATIONS
MIN = -2.766

Board mess seriously Article Continued Below
as scrap . Article Continued Below The
't afford an entire orchestra due to budget restrictions , but
12 more , with another due out this year . But
blocks as param eters ?
last train , which was due to arrive at the mountain
Final official results are due to be announced on 14
blocks as param eters ? is passing
, with a second print due in stores on the same
with [ CO 2 ] due to fertil ization of the
that will not be possible due to real life obligations .
el is a former chief executive officer of D n ip
t even do basic due diligence in selecting good people
The Senate committee is due to report next year .
a wholly benevolent foundation , due only to the fact the
bin and $ 8 million due to Carlos Quentin . But
is kered composer of four sym ph onies , has been
so few have it , due to the custom apps HTC
1989 , and was retired due to the hull being worn
Good rich All - Ter rain KO 2 tires specifically designed

INTERVAL 1.552 - 2.991
CONTAINS 0.116%

not forget …. MY HA IR . My glorious man e
l ort ot ft w > loving this meme [ 10
. See L ings , p . 232 n . 1
. 2001 - present Congress debates legislation on human
1986 ). The Ol we us B ully / Vict im

INTERVAL 0.113 - 1.552
CONTAINS 33.307%

the content . Some of the content includes the original publisher
aren 't yet clear , but Den elle Dixon , Mozilla
him special ; that 's what makes Dick Dick . Even
We ve been told to make life
argued that the Liberals are going after wealthy people who try

INTERVAL -1.326 - 0.113
CONTAINS 66.198%

asylum and design it to be filled of traps designed by
and crime . I don
to do that without jeopard izing the launch . But you
2013 , K arel de G uch t dismissed as
system , who confirmed that our hypothesis was correct and that

INTERVAL -2.766 - -1.326
CONTAINS 0.378%

today . This is partly due to the excellent conversion ,
[ 0 ], $ BG [ 1 ], $ BG [
long time and this sub wo ofer is worth every penny
John McCain . Newsletter Sign Up Continue reading the
. <|endoftext|> From John Brian King s book ,
Token,
Token ID11
Feature activation+2.97e-01

Pos logit contributions
 Commando+0.317
earch+0.306
senal+0.299
bt+0.252
ould+0.252
Neg logit contributions
 tem-0.277
 persona-0.273
 mig-0.270
 negatives-0.258
][/-0.250
Token Mikhail
Token ID42040
Feature activation-2.75e-01

Pos logit contributions
 tem+0.214
 persona+0.211
 mig+0.209
 negatives+0.200
][/+0.194
Neg logit contributions
 Commando-0.246
earch-0.236
senal-0.230
bt-0.195
ould-0.194
Token Al
Token ID978
Feature activation+1.95e-01

Pos logit contributions
 Commando+0.358
earch+0.350
senal+0.343
bt+0.313
ould+0.312
Neg logit contributions
 tem-0.067
 persona-0.065
 mig-0.063
 negatives-0.055
 ethnicity-0.048
Token"]
Token ID8973
Feature activation+1.74e+00

Pos logit contributions
 tem+0.140
 persona+0.137
 mig+0.136
 negatives+0.130
][/+0.126
Neg logit contributions
 Commando-0.163
earch-0.158
senal-0.153
bt-0.131
ould-0.130
Token [
Token ID685
Feature activation-5.92e-01

Pos logit contributions
 tem+1.036
 persona+1.016
 mig+1.004
 negatives+0.949
][/+0.916
Neg logit contributions
 Commando-1.660
earch-1.611
senal-1.593
bt-1.364
ould-1.362
TokenResult
Token ID23004
Feature activation+2.99e+00

Pos logit contributions
 Commando+0.511
earch+0.492
senal+0.476
bt+0.412
ould+0.409
Neg logit contributions
 tem-0.404
 persona-0.402
 mig-0.394
 negatives-0.378
][/-0.363
Token "
Token ID366
Feature activation-6.40e-02

Pos logit contributions
 tem+2.226
 persona+2.192
 mig+2.168
 negatives+2.075
][/+2.015
Neg logit contributions
 Commando-2.412
earch-2.326
senal-2.279
bt-1.904
ould-1.898
Token1
Token ID16
Feature activation+2.05e-01

Pos logit contributions
 Commando+0.043
earch+0.041
senal+0.040
bt+0.032
ould+0.032
Neg logit contributions
 tem-0.056
 persona-0.056
 mig-0.055
 negatives-0.053
][/-0.052
Token/
Token ID14
Feature activation-1.87e-01

Pos logit contributions
 tem+0.162
 persona+0.161
 mig+0.160
 negatives+0.153
][/+0.148
Neg logit contributions
 Commando-0.154
earch-0.147
senal-0.143
bt-0.120
ould-0.119
Token2
Token ID17
Feature activation-3.60e-02

Pos logit contributions
 Commando+0.142
earch+0.137
senal+0.129
bt+0.115
ould+0.112
Neg logit contributions
 persona-0.146
 tem-0.145
 mig-0.143
 negatives-0.138
 ethnicity-0.133
Token-
Token ID12
Feature activation+1.21e-02

Pos logit contributions
 Commando+0.027
earch+0.026
senal+0.025
bt+0.021
ould+0.021
Neg logit contributions
 tem-0.028
 persona-0.028
 mig-0.028
 negatives-0.027
][/-0.026
Token
Token ID198
Feature activation-3.86e-01

Pos logit contributions
 tem+0.726
 persona+0.715
 mig+0.706
 negatives+0.671
][/+0.650
Neg logit contributions
 Commando-0.963
earch-0.932
senal-0.913
bt-0.778
ould-0.776
Token
Token ID198
Feature activation-2.95e-01

Pos logit contributions
 Commando+0.287
earch+0.274
senal+0.250
bt+0.227
ould+0.221
Neg logit contributions
 tem-0.308
 persona-0.307
 mig-0.302
 negatives-0.292
 ethnicity-0.281
Tokenre
Token ID260
Feature activation-9.59e-02

Pos logit contributions
 Commando+0.232
earch+0.225
senal+0.218
bt+0.185
ould+0.183
Neg logit contributions
 tem-0.224
 persona-0.222
 mig-0.219
 negatives-0.210
][/-0.203
Tokeni
Token ID72
Feature activation+7.68e-01

Pos logit contributions
 Commando+0.075
earch+0.073
senal+0.071
bt+0.060
ould+0.060
Neg logit contributions
 tem-0.073
 persona-0.072
 mig-0.071
 negatives-0.068
][/-0.066
Token Profile
Token ID13118
Feature activation+4.21e-01

Pos logit contributions
 tem+0.556
 persona+0.548
 mig+0.542
 negatives+0.518
][/+0.502
Neg logit contributions
 Commando-0.634
earch-0.612
senal-0.598
bt-0.505
ould-0.503
Token Blog
Token ID14001
Feature activation+2.74e+00

Pos logit contributions
 tem+0.310
 persona+0.306
 mig+0.302
 negatives+0.289
][/+0.281
Neg logit contributions
 Commando-0.343
earch-0.331
senal-0.323
bt-0.272
ould-0.271
Token Joined
Token ID16709
Feature activation+7.36e-01

Pos logit contributions
 tem+2.235
 persona+2.201
 mig+2.180
 negatives+2.081
][/+2.054
Neg logit contributions
 Commando-2.003
earch-1.942
senal-1.925
bt-1.541
ould-1.534
Token October
Token ID3267
Feature activation+5.50e-01

Pos logit contributions
 tem+0.641
 persona+0.633
 mig+0.628
 negatives+0.604
][/+0.589
Neg logit contributions
 Commando-0.499
earch-0.479
senal-0.467
bt-0.375
ould-0.374
Token 2002
Token ID6244
Feature activation+6.38e-02

Pos logit contributions
 tem+0.448
 persona+0.441
 mig+0.437
 negatives+0.418
][/+0.409
Neg logit contributions
 Commando-0.403
earch-0.391
senal-0.384
ould-0.312
bt-0.312
Token United
Token ID1578
Feature activation-3.03e-02

Pos logit contributions
 tem+0.053
 persona+0.052
 mig+0.052
 negatives+0.050
][/+0.048
Neg logit contributions
 Commando-0.046
earch-0.044
senal-0.043
bt-0.035
ould-0.035
Token States
Token ID1829
Feature activation-4.65e-01

Pos logit contributions
 Commando+0.026
earch+0.025
senal+0.025
bt+0.021
ould+0.021
Neg logit contributions
 tem-0.021
 persona-0.020
 mig-0.020
 negatives-0.019
][/-0.019
Tokenwww
Token ID2503
Feature activation+1.03e-01

Pos logit contributions
 tem+0.141
 persona+0.139
 mig+0.138
 negatives+0.131
][/+0.127
Neg logit contributions
 Commando-0.166
earch-0.161
senal-0.156
bt-0.134
ould-0.133
Token.
Token ID13
Feature activation+4.21e-02

Pos logit contributions
 tem+0.077
 persona+0.076
 mig+0.075
 negatives+0.072
 ethnicity+0.070
Neg logit contributions
 Commando-0.082
earch-0.080
senal-0.072
bt-0.067
ould-0.065
Tokenbel
Token ID6667
Feature activation-7.53e-01

Pos logit contributions
 tem+0.031
 persona+0.030
 mig+0.030
 negatives+0.029
][/+0.028
Neg logit contributions
 Commando-0.034
earch-0.033
senal-0.032
bt-0.028
ould-0.027
Tokenfast
Token ID7217
Feature activation+1.26e+00

Pos logit contributions
 Commando+0.591
earch+0.576
senal+0.539
bt+0.480
ould+0.473
Neg logit contributions
 persona-0.563
 tem-0.561
 mig-0.560
 negatives-0.537
 ethnicity-0.516
Tokente
Token ID660
Feature activation+6.97e-01

Pos logit contributions
 tem+1.077
 persona+1.072
 mig+1.063
 negatives+1.029
][/+1.004
Neg logit contributions
 Commando-0.858
earch-0.821
senal-0.796
bt-0.654
ould-0.647
Tokenlegraph
Token ID16606
Feature activation+2.72e+00

Pos logit contributions
 tem+0.578
 persona+0.572
 mig+0.567
 negatives+0.544
][/+0.530
Neg logit contributions
 Commando-0.501
earch-0.480
senal-0.466
bt-0.384
ould-0.382
Token.
Token ID13
Feature activation+2.55e-01

Pos logit contributions
 tem+1.812
 persona+1.752
 mig+1.749
 negatives+1.652
][/+1.635
Neg logit contributions
 Commando-2.413
senal-2.392
earch-2.325
ould-1.927
bt-1.912
Tokenco
Token ID1073
Feature activation+4.94e-01

Pos logit contributions
 tem+0.207
 persona+0.203
 mig+0.201
 negatives+0.193
][/+0.189
Neg logit contributions
 Commando-0.189
earch-0.181
senal-0.180
ould-0.144
bt-0.144
Token.
Token ID13
Feature activation+3.91e-01

Pos logit contributions
 tem+0.345
 persona+0.340
 mig+0.336
 negatives+0.320
][/+0.311
Neg logit contributions
 Commando-0.421
earch-0.406
senal-0.401
bt-0.337
ould-0.336
Tokenuk
Token ID2724
Feature activation-6.31e-01

Pos logit contributions
 tem+0.277
 persona+0.271
 mig+0.268
 negatives+0.254
][/+0.250
Neg logit contributions
 Commando-0.332
senal-0.325
earch-0.319
ould-0.260
bt-0.257
Token/
Token ID14
Feature activation+1.76e-01

Pos logit contributions
 Commando+0.497
earch+0.479
senal+0.463
bt+0.392
ould+0.389
Neg logit contributions
 tem-0.480
 persona-0.475
 mig-0.470
 negatives-0.450
][/-0.435
TokenK
Token ID42
Feature activation+6.78e-02

Pos logit contributions
 Commando+0.077
earch+0.075
senal+0.073
bt+0.062
ould+0.062
Neg logit contributions
 tem-0.069
 persona-0.068
 mig-0.067
 negatives-0.064
][/-0.062
Tokenada
Token ID4763
Feature activation+2.61e-01

Pos logit contributions
 tem+0.051
 persona+0.050
 mig+0.049
 negatives+0.047
][/+0.046
Neg logit contributions
 Commando-0.054
earch-0.053
senal-0.051
bt-0.043
ould-0.043
Tokenver
Token ID332
Feature activation-4.70e-01

Pos logit contributions
 tem+0.188
 persona+0.186
 mig+0.185
 negatives+0.177
][/+0.171
Neg logit contributions
 Commando-0.216
earch-0.208
senal-0.202
bt-0.172
ould-0.171
TokenBB
Token ID15199
Feature activation+8.81e-01

Pos logit contributions
 Commando+0.506
earch+0.490
senal+0.481
bt+0.429
ould+0.426
Neg logit contributions
 tem-0.219
 persona-0.216
 mig-0.213
 negatives-0.198
][/-0.188
Token Profile
Token ID13118
Feature activation+3.37e-01

Pos logit contributions
 tem+0.434
 persona+0.426
 mig+0.419
 negatives+0.391
][/+0.373
Neg logit contributions
 Commando-0.931
earch-0.905
senal-0.889
bt-0.782
ould-0.780
Token Blog
Token ID14001
Feature activation+2.61e+00

Pos logit contributions
 tem+0.224
 persona+0.221
 mig+0.219
 negatives+0.208
][/+0.201
Neg logit contributions
 Commando-0.298
earch-0.287
senal-0.281
bt-0.240
ould-0.239
Token Joined
Token ID16709
Feature activation+9.83e-01

Pos logit contributions
 tem+2.121
 persona+2.088
 mig+2.067
 negatives+1.971
][/+1.943
Neg logit contributions
 Commando-1.912
earch-1.862
senal-1.852
ould-1.475
bt-1.470
Token June
Token ID2795
Feature activation+3.77e-01

Pos logit contributions
 tem+0.847
 persona+0.834
 mig+0.828
 negatives+0.793
][/+0.776
Neg logit contributions
 Commando-0.675
earch-0.655
senal-0.644
ould-0.514
bt-0.511
Token 2009
Token ID3717
Feature activation-4.28e-01

Pos logit contributions
 tem+0.302
 persona+0.299
 mig+0.296
 negatives+0.284
][/+0.276
Neg logit contributions
 Commando-0.282
earch-0.271
senal-0.265
bt-0.218
ould-0.217
Token Germany
Token ID4486
Feature activation+5.27e-02

Pos logit contributions
 Commando+0.290
earch+0.277
senal+0.269
bt+0.218
ould+0.216
Neg logit contributions
 tem-0.373
 persona-0.369
 mig-0.366
 negatives-0.353
][/-0.344
Token 247
Token ID30179
Feature activation-4.98e-01

Pos logit contributions
 tem+0.038
 persona+0.038
 mig+0.037
 negatives+0.036
][/+0.034
Neg logit contributions
 Commando-0.044
earch-0.042
senal-0.041
bt-0.035
ould-0.034
Tokenword
Token ID4775
Feature activation+4.19e-01

Pos logit contributions
 Commando+2.335
earch+2.285
senal+2.231
bt+2.006
ould+1.994
Neg logit contributions
 tem-0.775
 persona-0.757
 mig-0.751
 negatives-0.687
][/-0.644
Token by
Token ID416
Feature activation+1.01e-01

Pos logit contributions
 tem+0.260
 persona+0.256
 mig+0.253
 negatives+0.239
][/+0.231
Neg logit contributions
 Commando-0.390
earch-0.377
senal-0.371
bt-0.318
ould-0.318
Token Patrick
Token ID9925
Feature activation-4.55e-01

Pos logit contributions
 tem+0.067
 persona+0.065
 mig+0.065
 negatives+0.060
 ethnicity+0.059
Neg logit contributions
 Commando-0.091
earch-0.088
senal-0.087
bt-0.073
ould-0.073
Token J
Token ID449
Feature activation-2.80e-01

Pos logit contributions
 Commando+0.387
earch+0.372
senal+0.363
bt+0.310
ould+0.309
Neg logit contributions
 tem-0.317
 persona-0.314
 mig-0.311
 negatives-0.297
][/-0.286
Token.
Token ID13
Feature activation-9.01e-02

Pos logit contributions
 Commando+0.230
earch+0.223
senal+0.214
ould+0.183
bt+0.183
Neg logit contributions
 tem-0.206
 persona-0.202
 mig-0.200
 negatives-0.195
][/-0.187
Token Buchanan
Token ID36358
Feature activation+2.61e+00

Pos logit contributions
 Commando+0.066
earch+0.062
senal+0.061
bt+0.051
ould+0.050
Neg logit contributions
 tem-0.073
 persona-0.072
 mig-0.072
 negatives-0.070
 disadvantage-0.068
Token,
Token ID11
Feature activation+1.81e-01

Pos logit contributions
 tem+1.691
 mig+1.631
 persona+1.621
 negatives+1.493
 habits+1.431
Neg logit contributions
 Commando-2.372
earch-2.363
senal-2.291
bt-1.963
sbm-1.950
Token and
Token ID290
Feature activation-3.84e-02

Pos logit contributions
 persona+0.134
 mig+0.134
 tem+0.134
 negatives+0.128
][/+0.124
Neg logit contributions
 Commando-0.145
earch-0.140
senal-0.135
bt-0.115
ould-0.113
Token critical
Token ID4688
Feature activation-5.67e-02

Pos logit contributions
 Commando+0.040
earch+0.039
senal+0.038
bt+0.034
ould+0.033
Neg logit contributions
 tem-0.019
 persona-0.019
 mig-0.019
 negatives-0.018
][/-0.017
Token essays
Token ID27126
Feature activation-3.67e-01

Pos logit contributions
 Commando+0.066
earch+0.064
senal+0.063
bt+0.056
ould+0.056
Neg logit contributions
 tem-0.022
 persona-0.021
 mig-0.021
 negatives-0.019
][/-0.018
Token by
Token ID416
Feature activation+9.61e-02

Pos logit contributions
 Commando+0.365
senal+0.350
earch+0.349
ould+0.293
sbm+0.292
Neg logit contributions
 tem-0.217
 persona-0.210
 mig-0.203
 negatives-0.190
 ethnicity-0.184
Token
Token ID198
Feature activation-4.45e-01

Pos logit contributions
 Commando+0.292
earch+0.282
senal+0.276
bt+0.234
ould+0.233
Neg logit contributions
 tem-0.237
 persona-0.233
 mig-0.231
 negatives-0.220
][/-0.213
Token
Token ID198
Feature activation-2.76e-01

Pos logit contributions
 Commando+0.334
earch+0.319
senal+0.294
bt+0.263
ould+0.258
Neg logit contributions
 tem-0.353
 persona-0.351
 mig-0.346
 negatives-0.335
 ethnicity-0.322
TokenDark
Token ID17367
Feature activation+2.93e-01

Pos logit contributions
 Commando+0.218
earch+0.211
senal+0.205
bt+0.174
ould+0.172
Neg logit contributions
 tem-0.209
 persona-0.207
 mig-0.204
 negatives-0.196
][/-0.190
Tokenhorse
Token ID30527
Feature activation-9.81e-02

Pos logit contributions
 tem+0.222
 persona+0.219
 mig+0.217
 negatives+0.208
][/+0.202
Neg logit contributions
 Commando-0.233
earch-0.224
senal-0.219
bt-0.184
ould-0.182
Token Profile
Token ID13118
Feature activation+1.60e-01

Pos logit contributions
 Commando+0.100
earch+0.097
senal+0.095
bt+0.083
ould+0.083
Neg logit contributions
 tem-0.052
 persona-0.051
 mig-0.051
 negatives-0.048
][/-0.045
Token Blog
Token ID14001
Feature activation+2.58e+00

Pos logit contributions
 tem+0.111
 persona+0.110
 mig+0.109
 negatives+0.104
][/+0.100
Neg logit contributions
 Commando-0.137
earch-0.132
senal-0.129
bt-0.110
ould-0.109
Token Joined
Token ID16709
Feature activation+9.39e-01

Pos logit contributions
 tem+2.104
 persona+2.072
 mig+2.052
 negatives+1.957
][/+1.929
Neg logit contributions
 Commando-1.889
earch-1.836
senal-1.829
ould-1.453
bt-1.450
Token December
Token ID3426
Feature activation+2.73e-02

Pos logit contributions
 tem+0.812
 persona+0.800
 mig+0.794
 negatives+0.762
][/+0.745
Neg logit contributions
 Commando-0.643
earch-0.622
senal-0.610
ould-0.487
bt-0.486
Token 2011
Token ID2813
Feature activation-3.86e-01

Pos logit contributions
 tem+0.022
 persona+0.022
 mig+0.022
 negatives+0.021
][/+0.020
Neg logit contributions
 Commando-0.020
earch-0.019
senal-0.019
bt-0.015
ould-0.015
Token United
Token ID1578
Feature activation-5.39e-02

Pos logit contributions
 Commando+0.278
earch+0.266
senal+0.259
bt+0.213
ould+0.211
Neg logit contributions
 tem-0.320
 persona-0.316
 mig-0.313
 negatives-0.301
][/-0.293
Token States
Token ID1829
Feature activation-5.85e-01

Pos logit contributions
 Commando+0.048
earch+0.046
senal+0.045
bt+0.038
ould+0.038
Neg logit contributions
 tem-0.036
 persona-0.035
 mig-0.035
 negatives-0.033
][/-0.032
Token
Token ID198
Feature activation+1.43e-01

Pos logit contributions
 tem+0.226
 persona+0.222
 mig+0.220
 negatives+0.210
][/+0.204
Neg logit contributions
 Commando-0.266
earch-0.258
senal-0.253
bt-0.213
ould-0.212
Token
Token ID198
Feature activation+1.49e-01

Pos logit contributions
 tem+0.113
 persona+0.112
 mig+0.111
 negatives+0.107
 ethnicity+0.103
Neg logit contributions
 Commando-0.109
earch-0.104
senal-0.097
bt-0.086
ould-0.084
TokenBig
Token ID12804
Feature activation-9.01e-01

Pos logit contributions
 tem+0.110
 persona+0.109
 mig+0.108
 negatives+0.103
][/+0.100
Neg logit contributions
 Commando-0.121
earch-0.117
senal-0.114
bt-0.096
ould-0.096
TokenFan
Token ID22480
Feature activation-8.47e-01

Pos logit contributions
 Commando+0.792
earch+0.761
senal+0.747
bt+0.642
ould+0.636
Neg logit contributions
 tem-0.600
 persona-0.594
 mig-0.584
 negatives-0.554
][/-0.540
Token Profile
Token ID13118
Feature activation+2.86e-01

Pos logit contributions
 Commando+0.945
earch+0.917
senal+0.900
bt+0.805
ould+0.800
Neg logit contributions
 tem-0.365
 persona-0.359
 mig-0.352
 negatives-0.329
][/-0.308
Token Blog
Token ID14001
Feature activation+2.57e+00

Pos logit contributions
 tem+0.197
 persona+0.194
 mig+0.192
 negatives+0.184
][/+0.178
Neg logit contributions
 Commando-0.246
earch-0.238
senal-0.232
bt-0.198
ould-0.197
Token Joined
Token ID16709
Feature activation+9.28e-01

Pos logit contributions
 tem+2.088
 persona+2.057
 mig+2.034
 negatives+1.939
][/+1.912
Neg logit contributions
 Commando-1.886
earch-1.838
senal-1.831
ould-1.457
bt-1.449
Token December
Token ID3426
Feature activation+2.36e-01

Pos logit contributions
 tem+0.800
 persona+0.788
 mig+0.783
 negatives+0.750
][/+0.734
Neg logit contributions
 Commando-0.637
earch-0.617
senal-0.606
ould-0.484
bt-0.482
Token 2010
Token ID3050
Feature activation-4.36e-01

Pos logit contributions
 tem+0.191
 persona+0.188
 mig+0.187
 negatives+0.179
][/+0.174
Neg logit contributions
 Commando-0.175
earch-0.169
senal-0.166
bt-0.135
ould-0.135
Token TL
Token ID24811
Feature activation+9.08e-01

Pos logit contributions
 Commando+0.307
earch+0.293
senal+0.285
bt+0.233
ould+0.231
Neg logit contributions
 tem-0.369
 persona-0.365
 mig-0.362
 negatives-0.348
][/-0.339
TokenAD
Token ID2885
Feature activation+7.99e-01

Pos logit contributions
 tem+0.631
 persona+0.621
 mig+0.617
 negatives+0.584
][/+0.570
Neg logit contributions
 Commando-0.774
earch-0.751
senal-0.738
bt-0.622
ould-0.620
Token
Token ID198
Feature activation-1.33e-01

Pos logit contributions
 Commando+0.235
earch+0.225
senal+0.207
bt+0.185
ould+0.181
Neg logit contributions
 tem-0.249
 persona-0.248
 mig-0.244
 negatives-0.236
 ethnicity-0.227
TokenJ
Token ID41
Feature activation-1.44e-01

Pos logit contributions
 Commando+0.105
earch+0.101
senal+0.098
bt+0.083
ould+0.082
Neg logit contributions
 tem-0.101
 persona-0.099
 mig-0.098
 negatives-0.094
][/-0.091
Tokenaaaa
Token ID24794
Feature activation+4.24e-01

Pos logit contributions
 Commando+0.113
earch+0.110
senal+0.106
bt+0.090
ould+0.090
Neg logit contributions
 tem-0.110
 persona-0.109
 mig-0.108
 negatives-0.103
][/-0.100
Tokenasper
Token ID32981
Feature activation-3.67e-01

Pos logit contributions
 tem+0.339
 persona+0.334
 mig+0.331
 negatives+0.318
][/+0.309
Neg logit contributions
 Commando-0.319
earch-0.308
senal-0.298
bt-0.248
ould-0.248
Token Profile
Token ID13118
Feature activation+3.83e-01

Pos logit contributions
 Commando+0.469
earch+0.456
senal+0.448
bt+0.408
ould+0.406
Neg logit contributions
 tem-0.100
 persona-0.097
 mig-0.094
 negatives-0.084
 ethnicity-0.074
Token Blog
Token ID14001
Feature activation+2.57e+00

Pos logit contributions
 tem+0.279
 persona+0.275
 mig+0.272
 negatives+0.260
][/+0.252
Neg logit contributions
 Commando-0.315
earch-0.303
senal-0.296
bt-0.250
ould-0.249
Token Joined
Token ID16709
Feature activation+9.93e-01

Pos logit contributions
 tem+2.111
 persona+2.079
 mig+2.061
 negatives+1.968
][/+1.937
Neg logit contributions
 Commando-1.868
earch-1.809
senal-1.792
bt-1.435
ould-1.431
Token April
Token ID3035
Feature activation+1.35e-02

Pos logit contributions
 tem+0.859
 persona+0.847
 mig+0.841
 negatives+0.806
][/+0.788
Neg logit contributions
 Commando-0.679
earch-0.656
senal-0.643
ould-0.514
bt-0.513
Token 2012
Token ID2321
Feature activation-3.20e-01

Pos logit contributions
 tem+0.011
 persona+0.011
 mig+0.011
 negatives+0.010
][/+0.010
Neg logit contributions
 Commando-0.010
earch-0.010
senal-0.009
bt-0.008
ould-0.008
Token United
Token ID1578
Feature activation-8.66e-02

Pos logit contributions
 Commando+0.226
earch+0.216
senal+0.210
bt+0.171
ould+0.170
Neg logit contributions
 tem-0.271
 persona-0.268
 mig-0.265
 negatives-0.255
][/-0.249
Token States
Token ID1829
Feature activation-5.60e-01

Pos logit contributions
 Commando+0.076
earch+0.073
senal+0.072
bt+0.061
ould+0.061
Neg logit contributions
 tem-0.058
 persona-0.058
 mig-0.057
 negatives-0.054
][/-0.052
Token Obama
Token ID2486
Feature activation+2.78e-01

Pos logit contributions
 Commando+0.523
earch+0.504
senal+0.493
bt+0.424
ould+0.422
Neg logit contributions
 tem-0.377
 persona-0.371
 mig-0.367
 negatives-0.348
][/-0.338
Token
Token ID447
Feature activation+1.14e-01

Pos logit contributions
 tem+0.171
 persona+0.168
 mig+0.167
 negatives+0.158
][/+0.152
Neg logit contributions
 Commando-0.259
earch-0.250
senal-0.245
bt-0.212
ould-0.211
Token
Token ID247
Feature activation+1.58e-01

Pos logit contributions
 tem+0.094
 persona+0.093
 mig+0.092
 negatives+0.089
][/+0.086
Neg logit contributions
 Commando-0.082
earch-0.079
senal-0.077
ould-0.063
bt-0.063
Tokens
Token ID82
Feature activation+2.74e-01

Pos logit contributions
 tem+0.107
 persona+0.106
 mig+0.105
 negatives+0.100
][/+0.096
Neg logit contributions
 Commando-0.137
earch-0.132
senal-0.130
bt-0.111
ould-0.110
Token 2008
Token ID3648
Feature activation-2.07e-01

Pos logit contributions
 tem+0.187
 persona+0.184
 mig+0.183
 negatives+0.174
][/+0.169
Neg logit contributions
 Commando-0.238
earch-0.228
senal-0.222
bt-0.191
ould-0.190
Token convention
Token ID9831
Feature activation+2.54e+00

Pos logit contributions
 Commando+0.192
earch+0.184
senal+0.179
bt+0.157
ould+0.155
Neg logit contributions
 tem-0.129
 mig-0.126
 persona-0.126
 negatives-0.119
][/-0.115
Token speech
Token ID4046
Feature activation-8.70e-01

Pos logit contributions
 tem+1.593
 persona+1.570
 mig+1.543
 negatives+1.464
][/+1.410
Neg logit contributions
 Commando-2.341
earch-2.271
senal-2.233
bt-1.911
ould-1.907
Token and
Token ID290
Feature activation+1.20e-01

Pos logit contributions
 Commando+0.727
earch+0.702
senal+0.680
bt+0.582
ould+0.579
Neg logit contributions
 tem-0.613
 persona-0.605
 mig-0.603
 negatives-0.574
][/-0.559
Token went
Token ID1816
Feature activation+6.23e-01

Pos logit contributions
 tem+0.076
 persona+0.075
 mig+0.074
 negatives+0.070
][/+0.068
Neg logit contributions
 Commando-0.109
earch-0.105
senal-0.103
bt-0.089
ould-0.088
Token unp
Token ID8593
Feature activation+5.02e-01

Pos logit contributions
 tem+0.411
 persona+0.405
 mig+0.400
 negatives+0.380
][/+0.368
Neg logit contributions
 Commando-0.555
earch-0.537
senal-0.526
bt-0.449
ould-0.448
Tokenun
Token ID403
Feature activation-1.40e-01

Pos logit contributions
 tem+0.424
 persona+0.419
 mig+0.416
 negatives+0.399
][/+0.389
Neg logit contributions
 Commando-0.353
earch-0.339
senal-0.329
bt-0.269
ould-0.268
Tokenwww
Token ID2503
Feature activation+1.94e-01

Pos logit contributions
 tem+0.199
 persona+0.196
 mig+0.194
 negatives+0.185
][/+0.179
Neg logit contributions
 Commando-0.238
earch-0.230
senal-0.224
bt-0.190
ould-0.190
Token.
Token ID13
Feature activation+1.29e-01

Pos logit contributions
 tem+0.140
 persona+0.140
 mig+0.137
 negatives+0.132
 ethnicity+0.127
Neg logit contributions
 Commando-0.159
earch-0.154
senal-0.144
bt-0.128
ould-0.127
Tokenbel
Token ID6667
Feature activation-6.17e-01

Pos logit contributions
 tem+0.093
 persona+0.091
 mig+0.090
 negatives+0.086
][/+0.084
Neg logit contributions
 Commando-0.108
earch-0.104
senal-0.102
bt-0.085
ould-0.085
Tokenfast
Token ID7217
Feature activation+1.18e+00

Pos logit contributions
 Commando+0.498
earch+0.484
senal+0.456
bt+0.405
ould+0.398
Neg logit contributions
 tem-0.452
 persona-0.451
 mig-0.444
 negatives-0.426
 ethnicity-0.410
Tokente
Token ID660
Feature activation+8.18e-01

Pos logit contributions
 tem+1.004
 persona+0.999
 mig+0.987
 negatives+0.954
][/+0.924
Neg logit contributions
 Commando-0.808
earch-0.781
senal-0.743
bt-0.623
ould-0.617
Tokenlegraph
Token ID16606
Feature activation+2.53e+00

Pos logit contributions
 tem+0.683
 persona+0.678
 mig+0.670
 negatives+0.644
][/+0.625
Neg logit contributions
 Commando-0.581
earch-0.560
senal-0.536
bt-0.449
ould-0.446
Token.
Token ID13
Feature activation+3.68e-01

Pos logit contributions
 tem+1.705
 persona+1.656
 mig+1.650
 negatives+1.561
][/+1.542
Neg logit contributions
 Commando-2.223
senal-2.193
earch-2.142
ould-1.778
bt-1.766
Tokenco
Token ID1073
Feature activation+4.90e-01

Pos logit contributions
 tem+0.301
 persona+0.296
 mig+0.293
 negatives+0.281
][/+0.275
Neg logit contributions
 Commando-0.271
earch-0.260
senal-0.258
bt-0.206
ould-0.206
Token.
Token ID13
Feature activation+5.09e-01

Pos logit contributions
 tem+0.348
 persona+0.344
 mig+0.340
 negatives+0.324
][/+0.313
Neg logit contributions
 Commando-0.411
earch-0.397
senal-0.384
bt-0.330
ould-0.328
Tokenuk
Token ID2724
Feature activation-5.77e-01

Pos logit contributions
 tem+0.373
 persona+0.367
 mig+0.364
 negatives+0.346
][/+0.340
Neg logit contributions
 Commando-0.418
senal-0.402
earch-0.401
ould-0.327
bt-0.327
Token/
Token ID14
Feature activation+3.30e-01

Pos logit contributions
 Commando+0.448
earch+0.432
senal+0.410
bt+0.355
ould+0.350
Neg logit contributions
 tem-0.444
 persona-0.440
 mig-0.435
 negatives-0.418
 ethnicity-0.402
Token
Token ID198
Feature activation-3.61e-01

Pos logit contributions
 Commando+0.353
earch+0.338
senal+0.304
bt+0.280
ould+0.272
Neg logit contributions
 tem-0.384
 persona-0.383
 mig-0.378
 negatives-0.366
 ethnicity-0.352
TokenTB
Token ID22737
Feature activation-1.02e+00

Pos logit contributions
 Commando+0.285
earch+0.275
senal+0.267
bt+0.227
ould+0.224
Neg logit contributions
 tem-0.273
 persona-0.270
 mig-0.267
 negatives-0.256
][/-0.248
Tokenone
Token ID505
Feature activation-7.80e-01

Pos logit contributions
 Commando+0.851
earch+0.821
senal+0.802
bt+0.684
ould+0.682
Neg logit contributions
 tem-0.721
 persona-0.713
 mig-0.703
 negatives-0.675
][/-0.650
Token-
Token ID12
Feature activation+8.40e-02

Pos logit contributions
 Commando+0.782
earch+0.754
senal+0.738
bt+0.653
ould+0.646
Neg logit contributions
 tem-0.428
 persona-0.421
 mig-0.414
 negatives-0.392
][/-0.373
Token Profile
Token ID13118
Feature activation+4.09e-01

Pos logit contributions
 tem+0.058
 persona+0.057
 mig+0.057
 negatives+0.054
][/+0.052
Neg logit contributions
 Commando-0.072
earch-0.070
senal-0.068
bt-0.058
ould-0.058
Token Blog
Token ID14001
Feature activation+2.52e+00

Pos logit contributions
 tem+0.287
 persona+0.283
 mig+0.280
 negatives+0.267
][/+0.259
Neg logit contributions
 Commando-0.347
earch-0.335
senal-0.327
bt-0.278
ould-0.277
Token Joined
Token ID16709
Feature activation+7.58e-01

Pos logit contributions
 tem+2.049
 persona+2.018
 mig+1.995
 negatives+1.901
][/+1.876
Neg logit contributions
 Commando-1.853
earch-1.803
senal-1.797
ould-1.430
bt-1.424
Token November
Token ID3389
Feature activation+1.53e-01

Pos logit contributions
 tem+0.655
 persona+0.645
 mig+0.640
 negatives+0.614
][/+0.600
Neg logit contributions
 Commando-0.520
earch-0.503
senal-0.493
ould-0.394
bt-0.393
Token 2010
Token ID3050
Feature activation-4.88e-01

Pos logit contributions
 tem+0.126
 persona+0.124
 mig+0.123
 negatives+0.118
][/+0.115
Neg logit contributions
 Commando-0.111
earch-0.107
senal-0.105
bt-0.085
ould-0.085
Token United
Token ID1578
Feature activation+2.31e-02

Pos logit contributions
 Commando+0.343
earch+0.328
senal+0.320
bt+0.261
ould+0.259
Neg logit contributions
 tem-0.412
 persona-0.408
 mig-0.404
 negatives-0.388
][/-0.379
Token States
Token ID1829
Feature activation-4.53e-01

Pos logit contributions
 tem+0.015
 persona+0.015
 mig+0.015
 negatives+0.014
][/+0.014
Neg logit contributions
 Commando-0.020
earch-0.020
senal-0.019
bt-0.016
ould-0.016
Tokeng
Token ID70
Feature activation-4.47e-01

Pos logit contributions
 Commando+0.378
earch+0.365
senal+0.354
bt+0.296
ould+0.295
Neg logit contributions
 tem-0.380
 persona-0.376
 mig-0.372
 negatives-0.357
][/-0.346
Tokenits
Token ID896
Feature activation-4.29e-01

Pos logit contributions
 Commando+0.350
earch+0.339
senal+0.330
bt+0.277
ould+0.275
Neg logit contributions
 tem-0.341
 persona-0.337
 mig-0.333
 negatives-0.319
][/-0.309
Tokenamb
Token ID4131
Feature activation+1.41e+00

Pos logit contributions
 Commando+0.336
earch+0.328
senal+0.315
bt+0.268
ould+0.266
Neg logit contributions
 tem-0.326
 persona-0.323
 mig-0.319
 negatives-0.306
][/-0.297
Tokenry
Token ID563
Feature activation-6.55e-01

Pos logit contributions
 tem+1.062
 persona+1.047
 mig+1.039
 negatives+0.988
][/+0.970
Neg logit contributions
 Commando-1.125
earch-1.067
senal-1.064
bt-0.868
ould-0.865
Token/
Token ID14
Feature activation+1.87e-01

Pos logit contributions
 Commando+0.517
earch+0.501
senal+0.485
bt+0.411
ould+0.408
Neg logit contributions
 tem-0.494
 persona-0.488
 mig-0.483
 negatives-0.464
][/-0.449
Tokenstatus
Token ID13376
Feature activation+2.49e+00

Pos logit contributions
 tem+0.124
 persona+0.123
 mig+0.122
 negatives+0.116
][/+0.111
Neg logit contributions
 Commando-0.164
earch-0.158
senal-0.154
bt-0.134
ould-0.132
Token/
Token ID14
Feature activation+2.37e-01

Pos logit contributions
 tem+1.834
 persona+1.795
 mig+1.782
 negatives+1.691
][/+1.666
Neg logit contributions
 Commando-2.040
senal-1.977
earch-1.955
ould-1.604
bt-1.594
Token9
Token ID24
Feature activation+4.36e-01

Pos logit contributions
 tem+0.169
 persona+0.167
 mig+0.165
 negatives+0.157
][/+0.152
Neg logit contributions
 Commando-0.199
earch-0.192
senal-0.187
bt-0.159
ould-0.158
Token175
Token ID17430
Feature activation+3.67e-01

Pos logit contributions
 tem+0.302
 persona+0.298
 mig+0.295
 negatives+0.281
][/+0.272
Neg logit contributions
 Commando-0.373
earch-0.361
senal-0.353
bt-0.300
ould-0.299
Token669
Token ID36657
Feature activation+7.63e-01

Pos logit contributions
 tem+0.261
 persona+0.258
 mig+0.255
 negatives+0.243
][/+0.236
Neg logit contributions
 Commando-0.307
earch-0.296
senal-0.289
bt-0.245
ould-0.244
Token15
Token ID1314
Feature activation+2.07e-01

Pos logit contributions
 tem+0.543
 persona+0.535
 mig+0.530
 negatives+0.505
][/+0.491
Neg logit contributions
 Commando-0.639
earch-0.616
senal-0.605
bt-0.509
ould-0.508
TokenFlag
Token ID34227
Feature activation-2.42e-01

Pos logit contributions
 tem+0.101
 persona+0.099
 mig+0.099
 negatives+0.094
][/+0.092
Neg logit contributions
 Commando-0.091
earch-0.087
senal-0.085
bt-0.070
ould-0.070
Token Reply
Token ID14883
Feature activation+4.94e-01

Pos logit contributions
 Commando+0.202
earch+0.195
senal+0.190
bt+0.161
ould+0.160
Neg logit contributions
 tem-0.173
 persona-0.171
 mig-0.169
 negatives-0.161
][/-0.156
Token
Token ID198
Feature activation+6.12e-03

Pos logit contributions
 tem+0.375
 persona+0.370
 mig+0.367
 negatives+0.351
][/+0.341
Neg logit contributions
 Commando-0.390
earch-0.375
senal-0.368
bt-0.306
ould-0.305
Token
Token ID198
Feature activation+6.92e-02

Pos logit contributions
 tem+0.005
 persona+0.005
 mig+0.005
 negatives+0.004
][/+0.004
Neg logit contributions
 Commando-0.005
earch-0.005
senal-0.004
bt-0.004
ould-0.004
Token55
Token ID2816
Feature activation+1.70e-03

Pos logit contributions
 tem+0.050
 persona+0.050
 mig+0.049
 negatives+0.047
][/+0.045
Neg logit contributions
 Commando-0.057
earch-0.055
senal-0.054
bt-0.045
ould-0.045
Token Recommend
Token ID30361
Feature activation+2.46e+00

Pos logit contributions
 tem+0.001
 persona+0.001
 mig+0.001
 negatives+0.001
][/+0.001
Neg logit contributions
 Commando-0.002
earch-0.002
senal-0.002
bt-0.002
ould-0.002
Token
Token ID198
Feature activation+1.43e-01

Pos logit contributions
 tem+1.784
 persona+1.748
 mig+1.737
 negatives+1.653
][/+1.625
Neg logit contributions
 Commando-2.019
senal-2.014
earch-1.963
ould-1.596
bt-1.589
Token
Token ID198
Feature activation+1.77e-01

Pos logit contributions
 tem+0.112
 persona+0.111
 mig+0.110
 negatives+0.105
 ethnicity+0.102
Neg logit contributions
 Commando-0.109
earch-0.104
senal-0.098
bt-0.086
ould-0.084
TokenRecommend
Token ID41248
Feature activation+1.76e+00

Pos logit contributions
 tem+0.060
 persona+0.059
 mig+0.058
 negatives+0.052
][/+0.048
Neg logit contributions
 Commando-0.214
earch-0.209
senal-0.204
bt-0.185
ould-0.184
Token Share
Token ID8734
Feature activation+2.35e+00

Pos logit contributions
 tem+1.352
 persona+1.333
 mig+1.314
 negatives+1.256
][/+1.229
Neg logit contributions
 Commando-1.372
earch-1.348
senal-1.337
bt-1.081
ould-1.071
Token this
Token ID428
Feature activation+1.28e-01

Pos logit contributions
 tem+1.803
 persona+1.773
 mig+1.754
 negatives+1.683
][/+1.633
Neg logit contributions
 Commando-1.841
senal-1.800
earch-1.794
ould-1.442
bt-1.435
Token"]
Token ID8973
Feature activation+2.90e-01

Pos logit contributions
 tem+0.246
 persona+0.243
 mig+0.239
 negatives+0.226
][/+0.217
Neg logit contributions
 Commando-0.415
earch-0.402
senal-0.394
bt-0.343
ould-0.342
Token [
Token ID685
Feature activation-9.32e-01

Pos logit contributions
 tem+0.183
 persona+0.182
 mig+0.179
 negatives+0.170
][/+0.163
Neg logit contributions
 Commando-0.264
earch-0.255
senal-0.245
bt-0.215
ould-0.213
TokenSite
Token ID29123
Feature activation+5.14e-01

Pos logit contributions
 Commando+0.559
earch+0.526
senal+0.482
bt+0.411
ould+0.401
Neg logit contributions
 persona-0.871
 mig-0.864
 tem-0.858
 negatives-0.851
 ethnicity-0.819
Token "
Token ID366
Feature activation-4.72e-01

Pos logit contributions
 persona+0.389
 tem+0.389
 mig+0.384
 negatives+0.370
][/+0.357
Neg logit contributions
 Commando-0.401
earch-0.386
senal-0.370
bt-0.315
ould-0.312
TokenW
Token ID54
Feature activation-3.12e-01

Pos logit contributions
 Commando+0.372
earch+0.358
senal+0.348
bt+0.296
ould+0.293
Neg logit contributions
 tem-0.356
 persona-0.355
 mig-0.350
 negatives-0.337
 ethnicity-0.323
Tokenijk
Token ID45961
Feature activation+2.46e+00

Pos logit contributions
 Commando+0.253
earch+0.244
senal+0.236
bt+0.202
ould+0.201
Neg logit contributions
 tem-0.231
 persona-0.228
 mig-0.225
 negatives-0.217
][/-0.208
Token a
Token ID257
Feature activation+1.75e-01

Pos logit contributions
 tem+1.524
 persona+1.496
 mig+1.478
 negatives+1.396
][/+1.349
Neg logit contributions
 Commando-2.284
earch-2.212
senal-2.174
bt-1.870
ould-1.864
Tokenan
Token ID272
Feature activation-4.67e-01

Pos logit contributions
 tem+0.126
 persona+0.124
 mig+0.123
 negatives+0.118
][/+0.114
Neg logit contributions
 Commando-0.144
earch-0.139
senal-0.134
bt-0.115
ould-0.115
Token Z
Token ID1168
Feature activation+1.13e-01

Pos logit contributions
 Commando+0.389
earch+0.371
senal+0.358
bt+0.308
ould+0.306
Neg logit contributions
 tem-0.332
 persona-0.329
 mig-0.326
 negatives-0.313
][/-0.301
Tokenee
Token ID1453
Feature activation-1.19e-01

Pos logit contributions
 tem+0.083
 persona+0.081
 mig+0.081
 negatives+0.076
 ethnicity+0.074
Neg logit contributions
 Commando-0.093
earch-0.090
senal-0.085
ould-0.075
bt-0.074
Token"]
Token ID8973
Feature activation+1.79e+00

Pos logit contributions
 Commando+0.103
earch+0.099
senal+0.097
bt+0.083
ould+0.082
Neg logit contributions
 tem-0.081
 persona-0.080
 mig-0.079
 negatives-0.075
][/-0.073
Token Share
Token ID8734
Feature activation+2.35e+00

Pos logit contributions
 tem+1.439
 persona+1.418
 mig+1.399
 negatives+1.339
][/+1.313
Neg logit contributions
 Commando-1.428
earch-1.397
senal-1.384
bt-1.118
ould-1.109
Token this
Token ID428
Feature activation+1.97e-01

Pos logit contributions
 tem+1.829
 persona+1.801
 mig+1.783
 negatives+1.707
][/+1.665
Neg logit contributions
 Commando-1.820
earch-1.761
senal-1.761
ould-1.417
bt-1.415
Token comment
Token ID2912
Feature activation-9.68e-02

Pos logit contributions
 tem+0.072
 persona+0.071
 mig+0.071
 negatives+0.064
][/+0.059
Neg logit contributions
 Commando-0.229
earch-0.223
senal-0.209
bt-0.197
ould-0.195
Token on
Token ID319
Feature activation-1.78e-01

Pos logit contributions
 Commando+0.079
earch+0.076
senal+0.071
bt+0.064
ould+0.063
Neg logit contributions
 tem-0.069
 persona-0.069
 mig-0.068
 negatives-0.066
 ethnicity-0.063
Token Facebook
Token ID3203
Feature activation+1.26e-01

Pos logit contributions
 Commando+0.157
earch+0.151
senal+0.143
bt+0.127
ould+0.126
Neg logit contributions
 tem-0.119
 persona-0.117
 mig-0.116
 negatives-0.110
][/-0.106
Token Share
Token ID8734
Feature activation+2.46e+00

Pos logit contributions
 tem+0.104
 persona+0.103
 mig+0.102
 negatives+0.098
][/+0.095
Neg logit contributions
 Commando-0.091
earch-0.087
senal-0.083
bt-0.070
ould-0.070
Token this
Token ID428
Feature activation+1.10e-01

Pos logit contributions
 tem+1.929
 persona+1.902
 mig+1.883
 negatives+1.805
][/+1.760
Neg logit contributions
 Commando-1.881
earch-1.812
senal-1.799
bt-1.461
ould-1.458
Token comment
Token ID2912
Feature activation-1.77e-01

Pos logit contributions
 tem+0.040
 persona+0.040
 mig+0.039
 negatives+0.036
][/+0.033
Neg logit contributions
 Commando-0.127
earch-0.124
senal-0.116
bt-0.110
ould-0.109
Token on
Token ID319
Feature activation-2.48e-01

Pos logit contributions
 Commando+0.145
earch+0.139
senal+0.128
bt+0.117
ould+0.115
Neg logit contributions
 tem-0.128
 persona-0.127
 mig-0.126
 negatives-0.121
 ethnicity-0.115
Token Twitter
Token ID3009
Feature activation-4.83e-01

Pos logit contributions
 Commando+0.216
earch+0.207
senal+0.198
bt+0.174
ould+0.172
Neg logit contributions
 tem-0.168
 persona-0.166
 mig-0.164
 negatives-0.158
 ethnicity-0.151
Token
Token ID198
Feature activation-1.40e-01

Pos logit contributions
 Commando+0.380
earch+0.366
senal+0.357
bt+0.298
ould+0.297
Neg logit contributions
 tem-0.368
 persona-0.363
 mig-0.359
 negatives-0.344
][/-0.334
Token Share
Token ID8734
Feature activation+2.35e+00

Pos logit contributions
 tem+1.352
 persona+1.333
 mig+1.314
 negatives+1.256
][/+1.229
Neg logit contributions
 Commando-1.372
earch-1.348
senal-1.337
bt-1.081
ould-1.071
Token this
Token ID428
Feature activation+1.28e-01

Pos logit contributions
 tem+1.803
 persona+1.773
 mig+1.754
 negatives+1.683
][/+1.633
Neg logit contributions
 Commando-1.841
senal-1.800
earch-1.794
ould-1.442
bt-1.435
Token comment
Token ID2912
Feature activation-1.39e-01

Pos logit contributions
 tem+0.047
 persona+0.046
 mig+0.046
 negatives+0.041
][/+0.039
Neg logit contributions
 Commando-0.150
earch-0.146
senal-0.137
bt-0.129
ould-0.128
Token on
Token ID319
Feature activation-3.00e-01

Pos logit contributions
 Commando+0.114
earch+0.109
senal+0.102
bt+0.092
ould+0.091
Neg logit contributions
 tem-0.099
 persona-0.099
 mig-0.098
 negatives-0.094
 ethnicity-0.090
Token Facebook
Token ID3203
Feature activation-1.47e-01

Pos logit contributions
 Commando+0.264
earch+0.253
senal+0.241
bt+0.214
ould+0.212
Neg logit contributions
 tem-0.201
 persona-0.198
 mig-0.196
 negatives-0.188
][/-0.180
Token Share
Token ID8734
Feature activation+2.46e+00

Pos logit contributions
 Commando+0.105
earch+0.100
senal+0.095
bt+0.080
ould+0.080
Neg logit contributions
 tem-0.123
 persona-0.122
 mig-0.120
 negatives-0.116
][/-0.112
Token this
Token ID428
Feature activation+7.11e-03

Pos logit contributions
 tem+1.922
 persona+1.894
 mig+1.876
 negatives+1.798
][/+1.753
Neg logit contributions
 Commando-1.887
earch-1.820
senal-1.812
bt-1.467
ould-1.464
Token comment
Token ID2912
Feature activation-2.73e-01

Pos logit contributions
 tem+0.003
 persona+0.003
 mig+0.003
 negatives+0.002
][/+0.002
Neg logit contributions
 Commando-0.008
earch-0.008
senal-0.007
bt-0.007
ould-0.007
Token on
Token ID319
Feature activation-3.83e-01

Pos logit contributions
 Commando+0.223
earch+0.215
senal+0.196
bt+0.180
ould+0.177
Neg logit contributions
 tem-0.196
 persona-0.196
 mig-0.194
 negatives-0.186
 ethnicity-0.178
Token Twitter
Token ID3009
Feature activation-7.06e-01

Pos logit contributions
 Commando+0.332
earch+0.319
senal+0.303
bt+0.268
ould+0.263
Neg logit contributions
 tem-0.259
 persona-0.257
 mig-0.254
 negatives-0.244
 ethnicity-0.233
Token
Token ID198
Feature activation-1.94e-01

Pos logit contributions
 Commando+0.555
earch+0.534
senal+0.521
bt+0.436
ould+0.434
Neg logit contributions
 tem-0.538
 persona-0.531
 mig-0.525
 negatives-0.504
][/-0.489
Token Share
Token ID8734
Feature activation+2.29e+00

Pos logit contributions
 tem+1.340
 persona+1.322
 mig+1.302
 negatives+1.245
][/+1.214
Neg logit contributions
 Commando-1.345
earch-1.316
senal-1.304
bt-1.057
ould-1.047
Token this
Token ID428
Feature activation+2.26e-01

Pos logit contributions
 tem+1.578
 persona+1.544
 mig+1.508
 negatives+1.488
 ethnicity+1.416
Neg logit contributions
earch-1.983
 Commando-1.958
senal-1.953
sbm-1.619
govtrack-1.583
Token comment
Token ID2912
Feature activation-2.04e-01

Pos logit contributions
 tem+0.089
 persona+0.087
 mig+0.087
 negatives+0.077
][/+0.075
Neg logit contributions
 Commando-0.260
earch-0.252
senal-0.241
ould-0.222
bt-0.222
Token on
Token ID319
Feature activation-1.67e-01

Pos logit contributions
 Commando+0.173
earch+0.165
senal+0.161
ould+0.138
bt+0.138
Neg logit contributions
 tem-0.144
 persona-0.142
 mig-0.141
 negatives-0.134
][/-0.130
Token Facebook
Token ID3203
Feature activation-1.05e-01

Pos logit contributions
 Commando+0.147
earch+0.141
senal+0.138
ould+0.119
bt+0.119
Neg logit contributions
 tem-0.111
 persona-0.110
 mig-0.109
 negatives-0.104
][/-0.101
Token Share
Token ID8734
Feature activation+2.45e+00

Pos logit contributions
 Commando+0.075
earch+0.070
senal+0.069
bt+0.057
ould+0.057
Neg logit contributions
 tem-0.087
 persona-0.086
 mig-0.086
 negatives-0.083
][/-0.081
Token this
Token ID428
Feature activation+1.15e-01

Pos logit contributions
 tem+1.789
 persona+1.755
 mig+1.736
 negatives+1.677
][/+1.626
Neg logit contributions
 Commando-1.985
earch-1.964
senal-1.955
bt-1.590
ould-1.570
Token comment
Token ID2912
Feature activation-2.81e-01

Pos logit contributions
 tem+0.043
 persona+0.042
 mig+0.042
 negatives+0.038
][/+0.036
Neg logit contributions
 Commando-0.133
earch-0.129
senal-0.122
bt-0.114
ould-0.114
Token on
Token ID319
Feature activation-2.41e-01

Pos logit contributions
 Commando+0.229
earch+0.221
senal+0.206
bt+0.185
ould+0.184
Neg logit contributions
 tem-0.202
 persona-0.201
 mig-0.199
 negatives-0.191
][/-0.182
Token Twitter
Token ID3009
Feature activation-5.95e-01

Pos logit contributions
 Commando+0.209
earch+0.201
senal+0.194
bt+0.169
ould+0.167
Neg logit contributions
 tem-0.162
 persona-0.160
 mig-0.160
 negatives-0.153
][/-0.148
Token
Token ID198
Feature activation-2.75e-01

Pos logit contributions
 Commando+0.472
earch+0.453
senal+0.442
bt+0.371
ould+0.369
Neg logit contributions
 tem-0.450
 persona-0.444
 mig-0.439
 negatives-0.422
][/-0.409
Token by
Token ID416
Feature activation+6.51e-01

Pos logit contributions
 tem+0.513
 persona+0.505
 mig+0.500
 negatives+0.476
][/+0.461
Neg logit contributions
 Commando-0.638
earch-0.616
senal-0.601
bt-0.513
ould-0.510
Token corporate
Token ID6355
Feature activation+5.92e-02

Pos logit contributions
 tem+0.428
 persona+0.421
 mig+0.416
 negatives+0.396
][/+0.383
Neg logit contributions
 Commando-0.580
earch-0.560
senal-0.549
bt-0.470
ould-0.468
Token America
Token ID2253
Feature activation-4.37e-01

Pos logit contributions
 tem+0.038
 persona+0.037
 mig+0.037
 negatives+0.035
][/+0.034
Neg logit contributions
 Commando-0.054
earch-0.052
senal-0.051
bt-0.044
ould-0.044
Token and
Token ID290
Feature activation+2.74e-01

Pos logit contributions
 Commando+0.372
earch+0.359
senal+0.349
bt+0.298
ould+0.297
Neg logit contributions
 tem-0.304
 persona-0.300
 mig-0.297
 negatives-0.283
][/-0.275
Token the
Token ID262
Feature activation+4.92e-01

Pos logit contributions
 tem+0.171
 persona+0.168
 mig+0.166
 negatives+0.157
][/+0.152
Neg logit contributions
 Commando-0.253
earch-0.245
senal-0.239
bt-0.207
ould-0.205
Token Wall
Token ID5007
Feature activation+2.45e+00

Pos logit contributions
 tem+0.314
 persona+0.309
 mig+0.305
 negatives+0.289
][/+0.279
Neg logit contributions
 Commando-0.449
earch-0.436
senal-0.428
bt-0.366
ould-0.366
Token Street
Token ID3530
Feature activation-1.48e-02

Pos logit contributions
 tem+2.395
 persona+2.369
 mig+2.351
 negatives+2.265
][/+2.216
Neg logit contributions
 Commando-1.393
earch-1.336
senal-1.310
ould-0.989
bt-0.974
Token cabal
Token ID47548
Feature activation+1.04e-01

Pos logit contributions
 Commando+0.014
earch+0.013
senal+0.013
bt+0.011
ould+0.011
Neg logit contributions
 tem-0.009
 persona-0.009
 mig-0.009
 negatives-0.009
][/-0.008
Token is
Token ID318
Feature activation+1.76e-01

Pos logit contributions
 tem+0.072
 persona+0.071
 mig+0.070
 negatives+0.067
][/+0.064
Neg logit contributions
 Commando-0.090
earch-0.087
senal-0.085
bt-0.072
ould-0.072
Token their
Token ID511
Feature activation+8.76e-02

Pos logit contributions
 tem+0.126
 persona+0.124
 mig+0.123
 negatives+0.117
][/+0.114
Neg logit contributions
 Commando-0.147
earch-0.142
senal-0.138
bt-0.117
ould-0.117
Token complete
Token ID1844
Feature activation-4.32e-01

Pos logit contributions
 tem+0.051
 persona+0.050
 mig+0.049
 negatives+0.047
][/+0.045
Neg logit contributions
 Commando-0.085
earch-0.082
senal-0.080
bt-0.069
ould-0.069
TokenMaster
Token ID18254
Feature activation-7.22e-01

Pos logit contributions
 Commando+0.183
earch+0.177
senal+0.171
bt+0.145
ould+0.144
Neg logit contributions
 tem-0.175
 persona-0.174
 mig-0.172
 negatives-0.165
][/-0.159
Token of
Token ID286
Feature activation+7.33e-02

Pos logit contributions
 Commando+0.600
earch+0.574
senal+0.561
bt+0.480
ould+0.473
Neg logit contributions
 tem-0.518
 persona-0.513
 mig-0.507
 negatives-0.486
][/-0.470
Token Dal
Token ID12348
Feature activation+1.69e-01

Pos logit contributions
 tem+0.056
 persona+0.056
 mig+0.055
 negatives+0.053
][/+0.051
Neg logit contributions
 Commando-0.057
earch-0.054
senal-0.053
bt-0.045
ould-0.044
TokenK
Token ID42
Feature activation-1.14e-01

Pos logit contributions
 tem+0.129
 persona+0.128
 mig+0.127
 negatives+0.122
][/+0.118
Neg logit contributions
 Commando-0.131
earch-0.128
senal-0.123
bt-0.104
ould-0.103
Token Profile
Token ID13118
Feature activation+2.56e-01

Pos logit contributions
 Commando+0.091
earch+0.088
senal+0.085
bt+0.072
ould+0.072
Neg logit contributions
 tem-0.085
 persona-0.084
 mig-0.083
 negatives-0.080
][/-0.077
Token Blog
Token ID14001
Feature activation+2.43e+00

Pos logit contributions
 tem+0.168
 persona+0.166
 mig+0.164
 negatives+0.156
][/+0.151
Neg logit contributions
 Commando-0.229
earch-0.221
senal-0.216
bt-0.185
ould-0.184
Token Joined
Token ID16709
Feature activation+8.01e-01

Pos logit contributions
 tem+1.991
 persona+1.962
 mig+1.942
 negatives+1.857
][/+1.825
Neg logit contributions
 Commando-1.770
earch-1.718
senal-1.699
ould-1.360
bt-1.360
Token June
Token ID2795
Feature activation+4.16e-01

Pos logit contributions
 tem+0.695
 persona+0.686
 mig+0.680
 negatives+0.653
][/+0.638
Neg logit contributions
 Commando-0.546
earch-0.526
senal-0.514
bt-0.412
ould-0.411
Token 2012
Token ID2321
Feature activation-1.97e-01

Pos logit contributions
 tem+0.332
 persona+0.328
 mig+0.325
 negatives+0.311
][/+0.303
Neg logit contributions
 Commando-0.312
earch-0.300
senal-0.293
bt-0.242
ould-0.241
Token Canada
Token ID3340
Feature activation+7.02e-01

Pos logit contributions
 Commando+0.140
earch+0.134
senal+0.130
bt+0.107
ould+0.106
Neg logit contributions
 tem-0.165
 persona-0.163
 mig-0.162
 negatives-0.155
][/-0.151
Token 16
Token ID1467
Feature activation-5.42e-01

Pos logit contributions
 tem+0.504
 persona+0.497
 mig+0.492
 negatives+0.469
][/+0.455
Neg logit contributions
 Commando-0.584
earch-0.564
senal-0.553
bt-0.465
ould-0.464
Tokeniser
Token ID5847
Feature activation+9.70e-01

Pos logit contributions
 tem+0.333
 persona+0.328
 mig+0.324
 negatives+0.313
][/+0.301
Neg logit contributions
 Commando-0.354
earch-0.343
senal-0.331
bt-0.282
ould-0.280
TokenK
Token ID42
Feature activation-3.03e-02

Pos logit contributions
 tem+0.682
 persona+0.673
 mig+0.666
 negatives+0.636
][/+0.615
Neg logit contributions
 Commando-0.822
earch-0.790
senal-0.774
bt-0.659
ould-0.654
Tokenier
Token ID959
Feature activation+9.68e-01

Pos logit contributions
 Commando+0.026
earch+0.026
senal+0.025
bt+0.021
ould+0.021
Neg logit contributions
 tem-0.021
 persona-0.020
 mig-0.020
 negatives-0.019
][/-0.018
Tokenan
Token ID272
Feature activation+5.26e-02

Pos logit contributions
 tem+0.672
 persona+0.664
 mig+0.656
 negatives+0.626
][/+0.605
Neg logit contributions
 Commando-0.827
earch-0.797
senal-0.781
bt-0.666
ould-0.662
Token Profile
Token ID13118
Feature activation+2.43e-01

Pos logit contributions
 tem+0.013
 persona+0.012
 mig+0.012
 negatives+0.010
][/+0.009
Neg logit contributions
 Commando-0.069
earch-0.067
senal-0.066
bt-0.060
ould-0.060
Token Blog
Token ID14001
Feature activation+2.42e+00

Pos logit contributions
 tem+0.158
 persona+0.155
 mig+0.153
 negatives+0.146
][/+0.141
Neg logit contributions
 Commando-0.219
earch-0.212
senal-0.207
bt-0.178
ould-0.177
Token Joined
Token ID16709
Feature activation+7.77e-01

Pos logit contributions
 tem+1.971
 persona+1.942
 mig+1.922
 negatives+1.832
][/+1.807
Neg logit contributions
 Commando-1.769
earch-1.722
senal-1.709
ould-1.364
bt-1.360
Token August
Token ID2932
Feature activation+8.21e-01

Pos logit contributions
 tem+0.670
 persona+0.660
 mig+0.656
 negatives+0.628
][/+0.614
Neg logit contributions
 Commando-0.533
earch-0.516
senal-0.507
ould-0.405
bt-0.403
Token 2011
Token ID2813
Feature activation-2.91e-01

Pos logit contributions
 tem+0.661
 persona+0.651
 mig+0.645
 negatives+0.617
][/+0.603
Neg logit contributions
 Commando-0.610
earch-0.590
senal-0.580
ould-0.473
bt-0.473
Token United
Token ID1578
Feature activation-3.38e-02

Pos logit contributions
 Commando+0.209
earch+0.200
senal+0.195
bt+0.160
ould+0.159
Neg logit contributions
 tem-0.242
 persona-0.239
 mig-0.237
 negatives-0.228
][/-0.222
Token States
Token ID1829
Feature activation-4.89e-01

Pos logit contributions
 Commando+0.030
earch+0.029
senal+0.028
bt+0.024
ould+0.024
Neg logit contributions
 tem-0.023
 persona-0.022
 mig-0.022
 negatives-0.021
][/-0.020
Token Board
Token ID5926
Feature activation-8.88e-01

Pos logit contributions
 Commando+0.261
earch+0.248
senal+0.240
bt+0.192
ould+0.191
Neg logit contributions
 tem-0.368
 persona-0.364
 mig-0.362
 negatives-0.348
][/-0.339
Token mess
Token ID2085
Feature activation-3.62e-01

Pos logit contributions
 Commando+0.774
earch+0.746
senal+0.724
bt+0.623
ould+0.620
Neg logit contributions
 tem-0.597
 mig-0.587
 persona-0.587
 negatives-0.553
][/-0.542
Token seriously
Token ID6411
Feature activation-3.26e-01

Pos logit contributions
 Commando+0.339
earch+0.330
senal+0.319
bt+0.278
ould+0.277
Neg logit contributions
 tem-0.220
 persona-0.215
 mig-0.215
 negatives-0.203
][/-0.197
Token
Token ID198
Feature activation-3.55e-01

Pos logit contributions
 Commando+0.271
earch+0.262
senal+0.255
bt+0.216
ould+0.216
Neg logit contributions
 tem-0.231
 persona-0.230
 mig-0.227
 negatives-0.216
][/-0.210
Token
Token ID198
Feature activation-9.93e-02

Pos logit contributions
 Commando+0.265
earch+0.254
senal+0.233
bt+0.209
ould+0.205
Neg logit contributions
 tem-0.282
 persona-0.280
 mig-0.277
 negatives-0.267
 ethnicity-0.257
TokenArticle
Token ID14906
Feature activation-2.77e+00

Pos logit contributions
 Commando+0.076
earch+0.074
senal+0.071
bt+0.061
ould+0.060
Neg logit contributions
 tem-0.077
 persona-0.076
 mig-0.075
 negatives-0.072
 ethnicity-0.070
Token Continued
Token ID22168
Feature activation-7.77e-01

Pos logit contributions
 Commando+1.853
earch+1.761
senal+1.575
ould+1.425
bt+1.340
Neg logit contributions
 tem-2.350
 mig-2.345
 persona-2.337
][/-2.219
 negatives-2.214
Token Below
Token ID10383
Feature activation-2.60e-01

Pos logit contributions
 Commando+0.647
earch+0.619
senal+0.580
bt+0.516
ould+0.514
Neg logit contributions
 tem-0.558
 persona-0.552
 mig-0.542
 negatives-0.517
 ethnicity-0.499
Token
Token ID198
Feature activation+1.11e-02

Pos logit contributions
 Commando+0.198
earch+0.189
senal+0.178
bt+0.155
ould+0.153
Neg logit contributions
 tem-0.205
 persona-0.204
 mig-0.201
 negatives-0.194
 ethnicity-0.186
Token
Token ID198
Feature activation+1.09e-01

Pos logit contributions
 tem+0.009
 persona+0.009
 mig+0.009
 negatives+0.008
 ethnicity+0.008
Neg logit contributions
 Commando-0.008
earch-0.008
senal-0.008
bt-0.007
ould-0.007
Token
Token ID447
Feature activation+2.63e-03

Pos logit contributions
 tem+0.076
 persona+0.075
 mig+0.074
 negatives+0.071
 ethnicity+0.068
Neg logit contributions
 Commando-0.092
earch-0.090
senal-0.086
bt-0.075
ould-0.074
Token as
Token ID355
Feature activation-2.05e-01

Pos logit contributions
 tem+0.110
 persona+0.109
 mig+0.107
 negatives+0.102
][/+0.098
Neg logit contributions
 Commando-0.144
earch-0.140
senal-0.137
bt-0.116
ould-0.116
Token scrap
Token ID15881
Feature activation-9.63e-01

Pos logit contributions
 Commando+0.198
earch+0.192
senal+0.188
bt+0.164
ould+0.162
Neg logit contributions
 tem-0.118
 persona-0.116
 mig-0.115
 negatives-0.108
][/-0.105
Token.
Token ID13
Feature activation-1.40e-01

Pos logit contributions
 Commando+0.824
earch+0.794
senal+0.771
bt+0.668
ould+0.659
Neg logit contributions
 tem-0.658
 persona-0.648
 mig-0.645
 negatives-0.607
][/-0.595
Token
Token ID198
Feature activation-1.38e-01

Pos logit contributions
 Commando+0.115
earch+0.111
senal+0.108
bt+0.091
ould+0.091
Neg logit contributions
 tem-0.102
 persona-0.101
 mig-0.100
 negatives-0.095
][/-0.093
Token
Token ID198
Feature activation+8.98e-03

Pos logit contributions
 Commando+0.103
earch+0.098
senal+0.090
bt+0.081
ould+0.079
Neg logit contributions
 tem-0.110
 persona-0.109
 mig-0.108
 negatives-0.104
 ethnicity-0.100
TokenArticle
Token ID14906
Feature activation-2.69e+00

Pos logit contributions
 tem+0.006
 persona+0.006
 mig+0.006
 negatives+0.006
 ethnicity+0.006
Neg logit contributions
 Commando-0.008
earch-0.007
senal-0.007
bt-0.006
ould-0.006
Token Continued
Token ID22168
Feature activation-9.10e-01

Pos logit contributions
 Commando+1.864
earch+1.773
senal+1.661
ould+1.440
bt+1.393
Neg logit contributions
 persona-2.214
 tem-2.205
 mig-2.200
 negatives-2.104
][/-2.052
Token Below
Token ID10383
Feature activation-4.11e-01

Pos logit contributions
 Commando+0.751
earch+0.716
senal+0.664
bt+0.596
ould+0.594
Neg logit contributions
 tem-0.659
 persona-0.652
 mig-0.641
 negatives-0.612
 ethnicity-0.592
Token
Token ID198
Feature activation-2.07e-01

Pos logit contributions
 Commando+0.311
earch+0.298
senal+0.279
bt+0.244
ould+0.241
Neg logit contributions
 tem-0.325
 persona-0.323
 mig-0.318
 negatives-0.306
 ethnicity-0.295
Token
Token ID198
Feature activation-1.09e-01

Pos logit contributions
 Commando+0.156
earch+0.149
senal+0.138
bt+0.123
ould+0.120
Neg logit contributions
 tem-0.164
 persona-0.164
 mig-0.161
 negatives-0.156
 ethnicity-0.150
TokenThe
Token ID464
Feature activation-1.80e-01

Pos logit contributions
 Commando+0.095
earch+0.092
senal+0.088
bt+0.077
ould+0.076
Neg logit contributions
 tem-0.074
 persona-0.074
 mig-0.073
 negatives-0.069
 ethnicity-0.067
Token't
Token ID470
Feature activation-3.04e-01

Pos logit contributions
 tem+0.014
 persona+0.014
 mig+0.014
 negatives+0.013
 ethnicity+0.013
Neg logit contributions
 Commando-0.022
earch-0.021
senal-0.020
bt-0.018
ould-0.018
Token afford
Token ID5368
Feature activation-8.61e-01

Pos logit contributions
 Commando+0.263
earch+0.254
senal+0.246
bt+0.213
ould+0.212
Neg logit contributions
 tem-0.206
 persona-0.204
 mig-0.201
 negatives-0.192
][/-0.187
Token an
Token ID281
Feature activation-4.47e-01

Pos logit contributions
 Commando+0.760
earch+0.734
senal+0.719
bt+0.614
ould+0.612
Neg logit contributions
 tem-0.573
 persona-0.565
 mig-0.558
 negatives-0.530
][/-0.514
Token entire
Token ID2104
Feature activation+2.98e-01

Pos logit contributions
 Commando+0.408
earch+0.395
senal+0.387
bt+0.333
ould+0.332
Neg logit contributions
 tem-0.284
 persona-0.279
 mig-0.276
 negatives-0.261
][/-0.252
Token orchestra
Token ID40095
Feature activation-6.99e-01

Pos logit contributions
 tem+0.199
 persona+0.196
 mig+0.193
 negatives+0.184
][/+0.178
Neg logit contributions
 Commando-0.263
earch-0.254
senal-0.250
bt-0.212
ould-0.211
Token due
Token ID2233
Feature activation-2.65e+00

Pos logit contributions
 Commando+0.601
earch+0.580
senal+0.567
bt+0.483
ould+0.481
Neg logit contributions
 tem-0.482
 persona-0.474
 mig-0.470
 negatives-0.448
][/-0.433
Token to
Token ID284
Feature activation-1.43e-01

Pos logit contributions
 Commando+2.219
earch+2.141
senal+2.115
bt+1.780
ould+1.773
Neg logit contributions
 tem-1.868
 persona-1.841
 mig-1.821
 negatives-1.739
][/-1.686
Token budget
Token ID4466
Feature activation-7.44e-01

Pos logit contributions
 Commando+0.130
earch+0.126
senal+0.124
bt+0.106
ould+0.106
Neg logit contributions
 tem-0.092
 persona-0.090
 mig-0.089
 negatives-0.084
][/-0.082
Token restrictions
Token ID8733
Feature activation-1.67e-01

Pos logit contributions
 Commando+0.785
earch+0.760
senal+0.746
bt+0.659
ould+0.656
Neg logit contributions
 tem-0.366
 persona-0.360
 mig-0.352
 negatives-0.327
][/-0.316
Token,
Token ID11
Feature activation-2.23e-01

Pos logit contributions
 Commando+0.141
earch+0.136
senal+0.133
bt+0.113
ould+0.113
Neg logit contributions
 tem-0.118
 persona-0.116
 mig-0.115
 negatives-0.109
][/-0.106
Token but
Token ID475
Feature activation-5.32e-02

Pos logit contributions
 Commando+0.201
earch+0.194
senal+0.190
bt+0.163
ould+0.163
Neg logit contributions
 tem-0.145
 persona-0.143
 mig-0.141
 negatives-0.134
][/-0.129
Token 12
Token ID1105
Feature activation-5.22e-01

Pos logit contributions
 tem+0.491
 persona+0.483
 mig+0.477
 negatives+0.452
][/+0.431
Neg logit contributions
 Commando-0.765
earch-0.745
senal-0.735
bt-0.631
ould-0.630
Token more
Token ID517
Feature activation-1.91e-01

Pos logit contributions
 Commando+0.483
earch+0.468
senal+0.459
bt+0.395
ould+0.394
Neg logit contributions
 tem-0.326
 persona-0.321
 mig-0.317
 negatives-0.300
][/-0.290
Token,
Token ID11
Feature activation-1.26e-01

Pos logit contributions
 Commando+0.160
earch+0.154
senal+0.151
bt+0.128
ould+0.127
Neg logit contributions
 tem-0.135
 persona-0.133
 mig-0.132
 negatives-0.125
][/-0.122
Token with
Token ID351
Feature activation-2.50e-01

Pos logit contributions
 Commando+0.113
earch+0.109
senal+0.107
bt+0.092
ould+0.091
Neg logit contributions
 tem-0.082
 persona-0.081
 mig-0.080
 negatives-0.076
][/-0.073
Token another
Token ID1194
Feature activation-7.34e-01

Pos logit contributions
 Commando+0.229
earch+0.221
senal+0.217
bt+0.186
ould+0.186
Neg logit contributions
 tem-0.158
 persona-0.156
 mig-0.154
 negatives-0.146
][/-0.141
Token due
Token ID2233
Feature activation-2.63e+00

Pos logit contributions
 Commando+0.647
earch+0.624
senal+0.611
bt+0.522
ould+0.520
Neg logit contributions
 tem-0.491
 persona-0.483
 mig-0.478
 negatives-0.455
][/-0.440
Token out
Token ID503
Feature activation-5.88e-01

Pos logit contributions
 Commando+2.203
earch+2.113
senal+2.049
bt+1.754
ould+1.752
Neg logit contributions
 tem-1.832
 persona-1.811
 mig-1.797
 negatives-1.710
][/-1.662
Token this
Token ID428
Feature activation-1.66e-01

Pos logit contributions
 Commando+0.499
earch+0.480
senal+0.467
bt+0.399
ould+0.397
Neg logit contributions
 tem-0.411
 persona-0.405
 mig-0.402
 negatives-0.383
][/-0.372
Token year
Token ID614
Feature activation+3.39e-01

Pos logit contributions
 Commando+0.162
earch+0.157
senal+0.154
bt+0.134
ould+0.134
Neg logit contributions
 tem-0.094
 persona-0.092
 mig-0.091
 negatives-0.086
][/-0.083
Token.
Token ID13
Feature activation-2.02e-01

Pos logit contributions
 tem+0.237
 persona+0.234
 mig+0.232
 negatives+0.221
][/+0.214
Neg logit contributions
 Commando-0.288
earch-0.278
senal-0.272
bt-0.230
ould-0.230
Token But
Token ID887
Feature activation+8.72e-02

Pos logit contributions
 Commando+0.166
earch+0.160
senal+0.158
bt+0.132
ould+0.132
Neg logit contributions
 tem-0.146
 persona-0.144
 mig-0.143
 negatives-0.136
][/-0.132
Token blocks
Token ID7021
Feature activation-2.52e-01

Pos logit contributions
 tem+0.031
 persona+0.031
 mig+0.030
 negatives+0.029
][/+0.028
Neg logit contributions
 Commando-0.043
earch-0.042
senal-0.041
bt-0.035
ould-0.035
Token as
Token ID355
Feature activation-6.29e-02

Pos logit contributions
 Commando+0.215
earch+0.208
senal+0.200
bt+0.173
ould+0.171
Neg logit contributions
 tem-0.174
 persona-0.172
 mig-0.170
 negatives-0.162
][/-0.158
Token
Token ID564
Feature activation-1.47e-01

Pos logit contributions
 Commando+0.065
earch+0.063
senal+0.060
bt+0.054
ould+0.054
Neg logit contributions
 tem-0.032
 persona-0.031
 mig-0.031
 negatives-0.029
][/-0.028
Token
Token ID246
Feature activation+1.95e-01

Pos logit contributions
 Commando+0.109
earch+0.105
senal+0.102
bt+0.084
ould+0.084
Neg logit contributions
 tem-0.119
 persona-0.118
 mig-0.116
 negatives-0.112
][/-0.109
Tokenparam
Token ID17143
Feature activation-1.47e+00

Pos logit contributions
 tem+0.142
 persona+0.140
 mig+0.139
 negatives+0.132
][/+0.128
Neg logit contributions
 Commando-0.161
earch-0.155
senal-0.151
bt-0.128
ould-0.127
Tokeneters
Token ID7307
Feature activation-2.63e+00

Pos logit contributions
 Commando+1.343
earch+1.305
senal+1.264
bt+1.096
ould+1.094
Neg logit contributions
 tem-0.926
 persona-0.911
 mig-0.899
 negatives-0.852
][/-0.820
Token
Token ID447
Feature activation-1.00e-01

Pos logit contributions
 Commando+2.196
earch+2.127
senal+2.020
bt+1.751
ould+1.726
Neg logit contributions
 tem-1.845
 persona-1.830
 mig-1.812
 negatives-1.724
 ethnicity-1.661
Token
Token ID247
Feature activation+2.36e-02

Pos logit contributions
 Commando+0.079
earch+0.076
senal+0.074
bt+0.062
ould+0.062
Neg logit contributions
 tem-0.077
 persona-0.076
 mig-0.075
 negatives-0.072
][/-0.070
Token?
Token ID30
Feature activation-2.67e-01

Pos logit contributions
 tem+0.017
 persona+0.017
 mig+0.016
 negatives+0.016
][/+0.015
Neg logit contributions
 Commando-0.020
earch-0.019
senal-0.019
bt-0.016
ould-0.016
Token
Token ID198
Feature activation+6.11e-02

Pos logit contributions
 Commando+0.213
earch+0.205
senal+0.200
bt+0.168
ould+0.167
Neg logit contributions
 tem-0.200
 persona-0.198
 mig-0.196
 negatives-0.187
][/-0.182
Token
Token ID198
Feature activation+1.91e-01

Pos logit contributions
 tem+0.048
 persona+0.048
 mig+0.047
 negatives+0.046
 ethnicity+0.044
Neg logit contributions
 Commando-0.046
earch-0.044
senal-0.041
bt-0.036
ould-0.036
Token last
Token ID938
Feature activation-2.93e-01

Pos logit contributions
 Commando+0.145
earch+0.141
senal+0.138
bt+0.119
ould+0.118
Neg logit contributions
 tem-0.097
 persona-0.095
 mig-0.094
 negatives-0.089
][/-0.086
Token train
Token ID4512
Feature activation-3.04e-01

Pos logit contributions
 Commando+0.283
earch+0.274
senal+0.268
bt+0.233
ould+0.232
Neg logit contributions
 tem-0.171
 persona-0.168
 mig-0.165
 negatives-0.156
][/-0.150
Token,
Token ID11
Feature activation-2.32e-01

Pos logit contributions
 Commando+0.257
earch+0.247
senal+0.240
bt+0.205
ould+0.204
Neg logit contributions
 tem-0.214
 persona-0.210
 mig-0.208
 negatives-0.198
][/-0.192
Token which
Token ID543
Feature activation-2.45e-01

Pos logit contributions
 Commando+0.207
earch+0.200
senal+0.197
bt+0.168
ould+0.167
Neg logit contributions
 tem-0.152
 persona-0.150
 mig-0.148
 negatives-0.140
][/-0.136
Token was
Token ID373
Feature activation-2.68e-01

Pos logit contributions
 Commando+0.212
earch+0.204
senal+0.199
bt+0.170
ould+0.169
Neg logit contributions
 tem-0.167
 persona-0.165
 mig-0.163
 negatives-0.155
][/-0.151
Token due
Token ID2233
Feature activation-2.63e+00

Pos logit contributions
 Commando+0.243
earch+0.234
senal+0.228
bt+0.197
ould+0.196
Neg logit contributions
 tem-0.173
 persona-0.170
 mig-0.168
 negatives-0.160
][/-0.155
Token to
Token ID284
Feature activation-2.68e-01

Pos logit contributions
 Commando+2.162
earch+2.058
senal+1.992
bt+1.707
ould+1.701
Neg logit contributions
 tem-1.869
 persona-1.849
 mig-1.829
 negatives-1.753
][/-1.703
Token arrive
Token ID9240
Feature activation-7.88e-01

Pos logit contributions
 Commando+0.246
earch+0.236
senal+0.230
bt+0.200
ould+0.199
Neg logit contributions
 tem-0.170
 persona-0.167
 mig-0.165
 negatives-0.157
][/-0.152
Token at
Token ID379
Feature activation-4.36e-02

Pos logit contributions
 Commando+0.690
earch+0.660
senal+0.639
bt+0.554
ould+0.552
Neg logit contributions
 tem-0.528
 persona-0.521
 mig-0.515
 negatives-0.492
][/-0.478
Token the
Token ID262
Feature activation-3.71e-01

Pos logit contributions
 Commando+0.038
earch+0.037
senal+0.036
bt+0.031
ould+0.030
Neg logit contributions
 tem-0.030
 persona-0.029
 mig-0.029
 negatives-0.028
][/-0.027
Token mountain
Token ID8598
Feature activation-3.12e-01

Pos logit contributions
 Commando+0.320
earch+0.308
senal+0.301
bt+0.257
ould+0.255
Neg logit contributions
 tem-0.255
 persona-0.251
 mig-0.249
 negatives-0.237
][/-0.230
Token
Token ID198
Feature activation+1.38e-01

Pos logit contributions
 tem+0.083
 persona+0.083
 mig+0.082
 negatives+0.079
 ethnicity+0.076
Neg logit contributions
 Commando-0.080
earch-0.076
senal-0.071
bt-0.063
ould-0.062
TokenFinal
Token ID19006
Feature activation+3.47e-01

Pos logit contributions
 tem+0.096
 persona+0.095
 mig+0.093
 negatives+0.089
][/+0.086
Neg logit contributions
 Commando-0.117
earch-0.113
senal-0.110
bt-0.094
ould-0.094
Token official
Token ID1743
Feature activation-8.93e-02

Pos logit contributions
 tem+0.189
 persona+0.186
 mig+0.184
 negatives+0.172
][/+0.165
Neg logit contributions
 Commando-0.347
earch-0.336
senal-0.330
bt-0.289
ould-0.287
Token results
Token ID2482
Feature activation-4.60e-01

Pos logit contributions
 Commando+0.096
earch+0.093
senal+0.091
bt+0.082
ould+0.081
Neg logit contributions
 tem-0.041
 persona-0.040
 mig-0.040
 negatives-0.037
][/-0.035
Token are
Token ID389
Feature activation-3.67e-01

Pos logit contributions
 Commando+0.394
earch+0.379
senal+0.368
bt+0.317
ould+0.315
Neg logit contributions
 tem-0.317
 persona-0.313
 mig-0.309
 negatives-0.294
][/-0.286
Token due
Token ID2233
Feature activation-2.60e+00

Pos logit contributions
 Commando+0.354
earch+0.341
senal+0.332
bt+0.292
ould+0.290
Neg logit contributions
 tem-0.213
 persona-0.209
 mig-0.207
 negatives-0.195
][/-0.189
Token to
Token ID284
Feature activation-1.32e-01

Pos logit contributions
 Commando+2.209
earch+2.090
senal+2.022
bt+1.758
ould+1.755
Neg logit contributions
 tem-1.784
 persona-1.769
 mig-1.745
 negatives-1.665
][/-1.618
Token be
Token ID307
Feature activation-1.75e-01

Pos logit contributions
 Commando+0.104
earch+0.099
senal+0.096
bt+0.081
ould+0.081
Neg logit contributions
 tem-0.101
 persona-0.100
 mig-0.099
 negatives-0.094
][/-0.092
Token announced
Token ID3414
Feature activation+1.30e-01

Pos logit contributions
 Commando+0.149
earch+0.144
senal+0.139
bt+0.120
ould+0.119
Neg logit contributions
 tem-0.122
 persona-0.120
 mig-0.119
 negatives-0.113
][/-0.110
Token on
Token ID319
Feature activation-1.63e-01

Pos logit contributions
 tem+0.085
 persona+0.083
 mig+0.082
 negatives+0.078
][/+0.076
Neg logit contributions
 Commando-0.116
earch-0.112
senal-0.110
bt-0.094
ould-0.094
Token 14
Token ID1478
Feature activation-3.21e-01

Pos logit contributions
 Commando+0.136
earch+0.131
senal+0.127
bt+0.109
ould+0.108
Neg logit contributions
 tem-0.116
 persona-0.115
 mig-0.113
 negatives-0.108
][/-0.105
Token blocks
Token ID7021
Feature activation-2.89e-01

Pos logit contributions
 tem+0.013
 persona+0.012
 mig+0.012
 negatives+0.012
][/+0.011
Neg logit contributions
 Commando-0.016
earch-0.015
senal-0.014
bt-0.013
ould-0.012
Token as
Token ID355
Feature activation-9.05e-02

Pos logit contributions
 Commando+0.229
earch+0.222
senal+0.210
bt+0.182
ould+0.179
Neg logit contributions
 tem-0.217
 persona-0.215
 mig-0.212
 negatives-0.204
][/-0.198
Token
Token ID564
Feature activation-1.23e-01

Pos logit contributions
 Commando+0.074
earch+0.071
senal+0.070
bt+0.059
ould+0.058
Neg logit contributions
 tem-0.066
 persona-0.065
 mig-0.065
 negatives-0.062
][/-0.060
Token
Token ID246
Feature activation+1.84e-01

Pos logit contributions
 Commando+0.095
earch+0.091
senal+0.086
bt+0.074
ould+0.074
Neg logit contributions
 tem-0.096
 persona-0.094
 mig-0.094
 negatives-0.090
][/-0.087
Tokenparam
Token ID17143
Feature activation-1.41e+00

Pos logit contributions
 tem+0.164
 persona+0.162
 mig+0.161
 negatives+0.155
][/+0.151
Neg logit contributions
 Commando-0.121
earch-0.116
senal-0.112
bt-0.091
ould-0.090
Tokeneters
Token ID7307
Feature activation-2.56e+00

Pos logit contributions
 Commando+1.268
earch+1.247
senal+1.143
bt+1.052
ould+1.041
Neg logit contributions
 persona-0.885
 tem-0.882
 mig-0.863
 negatives-0.831
 ethnicity-0.802
Token
Token ID447
Feature activation-1.16e-01

Pos logit contributions
 Commando+2.310
earch+2.268
senal+2.038
bt+1.951
ould+1.907
Neg logit contributions
 tem-1.587
 persona-1.583
 mig-1.547
 negatives-1.504
 disadvantage-1.421
Token
Token ID247
Feature activation+1.69e-02

Pos logit contributions
 Commando+0.082
earch+0.079
senal+0.076
bt+0.063
ould+0.063
Neg logit contributions
 tem-0.096
 persona-0.095
 mig-0.094
 negatives-0.091
][/-0.088
Token?
Token ID30
Feature activation-2.47e-01

Pos logit contributions
 tem+0.013
 persona+0.013
 mig+0.013
 negatives+0.012
 ethnicity+0.012
Neg logit contributions
 Commando-0.013
earch-0.012
senal-0.011
bt-0.010
ould-0.010
Token is
Token ID318
Feature activation+2.53e-02

Pos logit contributions
 Commando+0.215
earch+0.208
senal+0.201
bt+0.174
ould+0.173
Neg logit contributions
 tem-0.166
 persona-0.165
 mig-0.163
 negatives-0.154
][/-0.150
Token passing
Token ID6427
Feature activation-7.71e-01

Pos logit contributions
 tem+0.019
 persona+0.019
 mig+0.018
 negatives+0.018
][/+0.017
Neg logit contributions
 Commando-0.020
earch-0.020
senal-0.019
bt-0.016
ould-0.016
Token,
Token ID11
Feature activation-1.09e-01

Pos logit contributions
 tem+0.308
 persona+0.304
 mig+0.301
 negatives+0.287
][/+0.278
Neg logit contributions
 Commando-0.369
earch-0.356
senal-0.348
bt-0.295
ould-0.294
Token with
Token ID351
Feature activation-1.00e-01

Pos logit contributions
 Commando+0.095
earch+0.091
senal+0.089
bt+0.076
ould+0.076
Neg logit contributions
 tem-0.074
 persona-0.073
 mig-0.072
 negatives-0.068
][/-0.066
Token a
Token ID257
Feature activation-2.37e-02

Pos logit contributions
 Commando+0.092
earch+0.088
senal+0.086
bt+0.075
ould+0.074
Neg logit contributions
 tem-0.063
 persona-0.063
 mig-0.062
 negatives-0.058
][/-0.057
Token second
Token ID1218
Feature activation+9.41e-01

Pos logit contributions
 Commando+0.022
earch+0.021
senal+0.021
bt+0.018
ould+0.018
Neg logit contributions
 tem-0.015
 persona-0.015
 mig-0.014
 negatives-0.014
][/-0.013
Token print
Token ID3601
Feature activation+5.16e-01

Pos logit contributions
 tem+0.559
 persona+0.550
 mig+0.541
 negatives+0.512
][/+0.491
Neg logit contributions
 Commando-0.900
earch-0.876
senal-0.865
bt-0.742
ould-0.742
Token due
Token ID2233
Feature activation-2.54e+00

Pos logit contributions
 tem+0.348
 persona+0.343
 mig+0.339
 negatives+0.323
][/+0.312
Neg logit contributions
 Commando-0.451
earch-0.436
senal-0.428
bt-0.363
ould-0.363
Token in
Token ID287
Feature activation-1.85e-01

Pos logit contributions
 Commando+2.137
earch+2.036
senal+1.947
bt+1.685
ould+1.682
Neg logit contributions
 tem-1.766
 persona-1.746
 mig-1.735
 negatives-1.653
][/-1.608
Token stores
Token ID7000
Feature activation+7.63e-01

Pos logit contributions
 Commando+0.143
earch+0.137
senal+0.132
bt+0.111
ould+0.110
Neg logit contributions
 tem-0.143
 persona-0.141
 mig-0.140
 negatives-0.134
][/-0.130
Token on
Token ID319
Feature activation-5.19e-02

Pos logit contributions
 tem+0.520
 persona+0.512
 mig+0.505
 negatives+0.481
][/+0.465
Neg logit contributions
 Commando-0.662
earch-0.641
senal-0.631
bt-0.535
ould-0.532
Token the
Token ID262
Feature activation-1.53e-01

Pos logit contributions
 Commando+0.038
earch+0.036
senal+0.035
bt+0.029
ould+0.028
Neg logit contributions
 tem-0.043
 persona-0.042
 mig-0.042
 negatives-0.040
][/-0.039
Token same
Token ID976
Feature activation-1.75e-02

Pos logit contributions
 Commando+0.144
earch+0.139
senal+0.135
bt+0.118
ould+0.117
Neg logit contributions
 tem-0.092
 persona-0.091
 mig-0.090
 negatives-0.085
][/-0.082
Token with
Token ID351
Feature activation-1.78e-01

Pos logit contributions
 tem+0.049
 persona+0.048
 mig+0.048
 negatives+0.045
][/+0.044
Neg logit contributions
 Commando-0.070
earch-0.067
senal-0.065
bt-0.057
ould-0.057
Token [
Token ID685
Feature activation-4.13e-01

Pos logit contributions
 Commando+0.172
earch+0.167
senal+0.163
ould+0.142
bt+0.142
Neg logit contributions
 tem-0.101
 persona-0.101
 mig-0.099
 negatives-0.093
][/-0.091
TokenCO
Token ID8220
Feature activation-9.63e-01

Pos logit contributions
 Commando+0.283
earch+0.271
senal+0.264
bt+0.215
ould+0.214
Neg logit contributions
 tem-0.355
 persona-0.352
 mig-0.348
 negatives-0.336
][/-0.327
Token 2
Token ID362
Feature activation-1.73e-01

Pos logit contributions
 Commando+0.778
earch+0.744
senal+0.708
bt+0.621
ould+0.619
Neg logit contributions
 persona-0.697
 tem-0.695
 mig-0.688
 negatives-0.650
 ethnicity-0.627
Token ]
Token ID2361
Feature activation-3.14e-01

Pos logit contributions
 Commando+0.152
earch+0.146
senal+0.143
ould+0.122
bt+0.122
Neg logit contributions
 tem-0.117
 persona-0.115
 mig-0.114
 negatives-0.108
][/-0.104
Token due
Token ID2233
Feature activation-2.54e+00

Pos logit contributions
 Commando+0.269
earch+0.260
senal+0.254
bt+0.217
ould+0.216
Neg logit contributions
 tem-0.217
 persona-0.214
 mig-0.212
 negatives-0.202
][/-0.196
Token to
Token ID284
Feature activation-1.30e-01

Pos logit contributions
 Commando+2.072
earch+2.011
senal+1.980
ould+1.667
bt+1.664
Neg logit contributions
 tem-1.802
 persona-1.794
 mig-1.770
 negatives-1.697
][/-1.651
Token fertil
Token ID19078
Feature activation-5.39e-02

Pos logit contributions
 Commando+0.121
earch+0.117
senal+0.114
ould+0.099
bt+0.098
Neg logit contributions
 tem-0.079
 persona-0.078
 mig-0.077
 negatives-0.073
][/-0.071
Tokenization
Token ID1634
Feature activation-3.19e-01

Pos logit contributions
 Commando+0.045
earch+0.043
senal+0.042
ould+0.036
bt+0.036
Neg logit contributions
 tem-0.039
 persona-0.038
 mig-0.038
 negatives-0.036
][/-0.035
Token of
Token ID286
Feature activation-1.37e-01

Pos logit contributions
 Commando+0.272
earch+0.262
senal+0.254
bt+0.218
ould+0.218
Neg logit contributions
 tem-0.220
 persona-0.219
 mig-0.216
 negatives-0.205
][/-0.200
Token the
Token ID262
Feature activation-3.26e-02

Pos logit contributions
 Commando+0.125
earch+0.121
senal+0.117
ould+0.102
bt+0.101
Neg logit contributions
 tem-0.087
 persona-0.086
 mig-0.085
 negatives-0.080
][/-0.078
Token that
Token ID326
Feature activation-1.82e-01

Pos logit contributions
 tem+0.088
 persona+0.087
 mig+0.086
 negatives+0.082
][/+0.080
Neg logit contributions
 Commando-0.099
earch-0.096
senal-0.094
bt-0.079
ould-0.079
Token will
Token ID481
Feature activation-1.02e-01

Pos logit contributions
 Commando+0.156
earch+0.150
senal+0.147
bt+0.125
ould+0.124
Neg logit contributions
 tem-0.126
 persona-0.125
 mig-0.123
 negatives-0.117
][/-0.114
Token not
Token ID407
Feature activation-1.63e-02

Pos logit contributions
 Commando+0.090
earch+0.087
senal+0.084
bt+0.074
ould+0.073
Neg logit contributions
 tem-0.067
 persona-0.066
 mig-0.065
 negatives-0.062
][/-0.060
Token be
Token ID307
Feature activation-2.31e-01

Pos logit contributions
 Commando+0.013
earch+0.013
senal+0.012
bt+0.010
ould+0.010
Neg logit contributions
 tem-0.012
 persona-0.012
 mig-0.012
 negatives-0.011
][/-0.011
Token possible
Token ID1744
Feature activation-3.49e-01

Pos logit contributions
 Commando+0.211
earch+0.204
senal+0.200
bt+0.172
ould+0.171
Neg logit contributions
 tem-0.146
 persona-0.144
 mig-0.142
 negatives-0.135
][/-0.131
Token due
Token ID2233
Feature activation-2.52e+00

Pos logit contributions
 Commando+0.304
earch+0.294
senal+0.287
bt+0.245
ould+0.244
Neg logit contributions
 tem-0.237
 persona-0.233
 mig-0.231
 negatives-0.220
][/-0.213
Token to
Token ID284
Feature activation-2.09e-01

Pos logit contributions
 Commando+2.034
earch+1.952
senal+1.923
bt+1.623
ould+1.596
Neg logit contributions
 tem-1.826
 persona-1.803
 mig-1.790
 negatives-1.715
][/-1.665
Token real
Token ID1103
Feature activation-3.57e-01

Pos logit contributions
 Commando+0.190
earch+0.183
senal+0.179
bt+0.154
ould+0.153
Neg logit contributions
 tem-0.133
 persona-0.132
 mig-0.130
 negatives-0.123
][/-0.120
Token life
Token ID1204
Feature activation-7.14e-02

Pos logit contributions
 Commando+0.339
earch+0.328
senal+0.322
bt+0.278
ould+0.277
Neg logit contributions
 tem-0.213
 persona-0.210
 mig-0.207
 negatives-0.195
][/-0.189
Token obligations
Token ID13675
Feature activation+3.67e-01

Pos logit contributions
 Commando+0.072
earch+0.070
senal+0.068
bt+0.060
ould+0.060
Neg logit contributions
 tem-0.038
 persona-0.037
 mig-0.037
 negatives-0.034
][/-0.033
Token.
Token ID13
Feature activation-2.13e-02

Pos logit contributions
 tem+0.259
 persona+0.256
 mig+0.253
 negatives+0.241
][/+0.234
Neg logit contributions
 Commando-0.309
earch-0.298
senal-0.291
bt-0.247
ould-0.246
Tokenel
Token ID417
Feature activation-1.32e-01

Pos logit contributions
 Commando+0.104
earch+0.100
senal+0.094
bt+0.078
ould+0.077
Neg logit contributions
 tem-0.150
 persona-0.149
 mig-0.148
 negatives-0.143
][/-0.138
Token is
Token ID318
Feature activation+1.16e-01

Pos logit contributions
 Commando+0.113
earch+0.109
senal+0.106
bt+0.091
ould+0.090
Neg logit contributions
 tem-0.092
 persona-0.090
 mig-0.089
 negatives-0.085
][/-0.083
Token a
Token ID257
Feature activation+1.20e-01

Pos logit contributions
 tem+0.072
 persona+0.070
 mig+0.070
 negatives+0.067
][/+0.065
Neg logit contributions
 Commando-0.108
earch-0.104
senal-0.101
bt-0.088
ould-0.087
Token former
Token ID1966
Feature activation+7.35e-01

Pos logit contributions
 tem+0.078
 persona+0.077
 mig+0.076
 negatives+0.072
][/+0.070
Neg logit contributions
 Commando-0.108
earch-0.104
senal-0.101
bt-0.087
ould-0.086
Token chief
Token ID4039
Feature activation-1.05e-01

Pos logit contributions
 tem+0.483
 persona+0.476
 mig+0.471
 negatives+0.448
][/+0.434
Neg logit contributions
 Commando-0.656
earch-0.632
senal-0.617
bt-0.529
ould-0.526
Token executive
Token ID4640
Feature activation-2.52e+00

Pos logit contributions
 Commando+0.098
earch+0.094
senal+0.092
bt+0.079
ould+0.079
Neg logit contributions
 tem-0.064
 persona-0.063
 mig-0.062
 negatives-0.059
][/-0.057
Token officer
Token ID3818
Feature activation-5.79e-01

Pos logit contributions
 Commando+2.157
earch+2.069
senal+1.992
bt+1.710
ould+1.685
Neg logit contributions
 tem-1.717
 persona-1.694
 mig-1.683
 negatives-1.625
][/-1.582
Token of
Token ID286
Feature activation-3.91e-01

Pos logit contributions
 Commando+0.506
earch+0.488
senal+0.476
bt+0.407
ould+0.404
Neg logit contributions
 tem-0.391
 persona-0.385
 mig-0.382
 negatives-0.364
][/-0.353
Token D
Token ID360
Feature activation-2.32e-01

Pos logit contributions
 Commando+0.336
earch+0.323
senal+0.314
bt+0.267
ould+0.265
Neg logit contributions
 tem-0.270
 persona-0.266
 mig-0.263
 negatives-0.251
][/-0.243
Tokenn
Token ID77
Feature activation-1.38e-01

Pos logit contributions
 Commando+0.181
earch+0.177
senal+0.170
bt+0.145
ould+0.144
Neg logit contributions
 tem-0.177
 persona-0.175
 mig-0.173
 negatives-0.166
][/-0.160
Tokenip
Token ID541
Feature activation-2.15e-01

Pos logit contributions
 Commando+0.103
earch+0.099
senal+0.096
bt+0.080
ould+0.079
Neg logit contributions
 tem-0.111
 persona-0.110
 mig-0.109
 negatives-0.104
][/-0.101
Token
Token ID247
Feature activation-1.40e-01

Pos logit contributions
 Commando+0.046
earch+0.045
senal+0.043
ould+0.036
bt+0.036
Neg logit contributions
 tem-0.053
 persona-0.053
 mig-0.052
 negatives-0.050
][/-0.049
Tokent
Token ID83
Feature activation-3.80e-02

Pos logit contributions
 Commando+0.099
earch+0.095
senal+0.092
bt+0.076
ould+0.075
Neg logit contributions
 tem-0.117
 persona-0.116
 mig-0.115
 negatives-0.110
][/-0.107
Token even
Token ID772
Feature activation-1.80e-01

Pos logit contributions
 Commando+0.034
earch+0.032
senal+0.032
bt+0.027
ould+0.027
Neg logit contributions
 tem-0.025
 persona-0.025
 mig-0.025
 negatives-0.023
][/-0.023
Token do
Token ID466
Feature activation+3.52e-02

Pos logit contributions
 Commando+0.154
earch+0.149
senal+0.145
bt+0.123
ould+0.123
Neg logit contributions
 tem-0.125
 persona-0.123
 mig-0.122
 negatives-0.116
][/-0.112
Token basic
Token ID4096
Feature activation+5.38e-01

Pos logit contributions
 tem+0.024
 persona+0.023
 mig+0.023
 negatives+0.022
][/+0.021
Neg logit contributions
 Commando-0.031
earch-0.030
senal-0.029
bt-0.025
ould-0.025
Token due
Token ID2233
Feature activation-2.52e+00

Pos logit contributions
 tem+0.298
 persona+0.293
 mig+0.289
 negatives+0.272
][/+0.260
Neg logit contributions
 Commando-0.535
earch-0.519
senal-0.511
bt-0.444
ould-0.443
Token diligence
Token ID34136
Feature activation+1.77e-01

Pos logit contributions
 Commando+2.484
earch+2.436
senal+2.366
bt+2.089
ould+2.069
Neg logit contributions
 tem-1.382
 persona-1.355
 mig-1.346
 negatives-1.262
][/-1.221
Token in
Token ID287
Feature activation-1.99e-01

Pos logit contributions
 tem+0.122
 persona+0.120
 mig+0.119
 negatives+0.113
][/+0.110
Neg logit contributions
 Commando-0.152
earch-0.147
senal-0.142
bt-0.122
ould-0.121
Token selecting
Token ID17246
Feature activation-1.66e-01

Pos logit contributions
 Commando+0.177
earch+0.172
senal+0.167
bt+0.144
ould+0.143
Neg logit contributions
 tem-0.130
 persona-0.128
 mig-0.127
 negatives-0.120
][/-0.116
Token good
Token ID922
Feature activation-2.92e-01

Pos logit contributions
 Commando+0.154
earch+0.149
senal+0.145
ould+0.126
bt+0.126
Neg logit contributions
 tem-0.102
 persona-0.101
 mig-0.100
 negatives-0.094
][/-0.091
Token people
Token ID661
Feature activation+2.02e-01

Pos logit contributions
 Commando+0.289
earch+0.280
senal+0.270
ould+0.240
bt+0.238
Neg logit contributions
 tem-0.161
 persona-0.159
 mig-0.158
 negatives-0.148
][/-0.145
Token
Token ID198
Feature activation-3.69e-01

Pos logit contributions
 Commando+0.386
earch+0.368
senal+0.330
bt+0.304
ould+0.297
Neg logit contributions
 tem-0.421
 persona-0.419
 mig-0.413
 negatives-0.399
 ethnicity-0.385
TokenThe
Token ID464
Feature activation-3.21e-01

Pos logit contributions
 Commando+0.325
earch+0.317
senal+0.304
bt+0.268
ould+0.264
Neg logit contributions
 tem-0.245
 persona-0.242
 mig-0.238
 negatives-0.228
][/-0.218
Token Senate
Token ID3845
Feature activation-6.21e-01

Pos logit contributions
 Commando+0.285
earch+0.275
senal+0.267
bt+0.230
ould+0.228
Neg logit contributions
 tem-0.212
 persona-0.209
 mig-0.207
 negatives-0.196
][/-0.191
Token committee
Token ID5583
Feature activation-5.37e-01

Pos logit contributions
 Commando+0.522
earch+0.502
senal+0.490
bt+0.416
ould+0.415
Neg logit contributions
 tem-0.441
 persona-0.434
 mig-0.430
 negatives-0.410
][/-0.398
Token is
Token ID318
Feature activation-1.06e-01

Pos logit contributions
 Commando+0.473
earch+0.457
senal+0.440
bt+0.383
ould+0.380
Neg logit contributions
 tem-0.357
 persona-0.350
 mig-0.348
 negatives-0.330
][/-0.322
Token due
Token ID2233
Feature activation-2.52e+00

Pos logit contributions
 Commando+0.099
earch+0.096
senal+0.094
bt+0.081
ould+0.081
Neg logit contributions
 tem-0.066
 persona-0.065
 mig-0.064
 negatives-0.060
][/-0.058
Token to
Token ID284
Feature activation-2.07e-01

Pos logit contributions
 Commando+2.062
earch+1.983
senal+1.922
bt+1.637
ould+1.634
Neg logit contributions
 tem-1.814
 persona-1.792
 mig-1.778
 negatives-1.704
][/-1.648
Token report
Token ID989
Feature activation+2.08e-01

Pos logit contributions
 Commando+0.193
earch+0.187
senal+0.179
bt+0.158
ould+0.157
Neg logit contributions
 tem-0.127
 persona-0.125
 mig-0.124
 negatives-0.117
][/-0.114
Token next
Token ID1306
Feature activation+2.32e-02

Pos logit contributions
 tem+0.134
 persona+0.132
 mig+0.130
 negatives+0.123
][/+0.119
Neg logit contributions
 Commando-0.188
earch-0.182
senal-0.178
bt-0.153
ould-0.152
Token year
Token ID614
Feature activation+1.95e-01

Pos logit contributions
 tem+0.015
 persona+0.015
 mig+0.014
 negatives+0.014
][/+0.013
Neg logit contributions
 Commando-0.021
earch-0.020
senal-0.020
bt-0.017
ould-0.017
Token.
Token ID13
Feature activation-4.18e-01

Pos logit contributions
 tem+0.133
 persona+0.131
 mig+0.129
 negatives+0.123
][/+0.119
Neg logit contributions
 Commando-0.170
earch-0.164
senal-0.161
bt-0.137
ould-0.136
Token a
Token ID257
Feature activation+6.24e-02

Pos logit contributions
 tem+0.075
 persona+0.074
 mig+0.073
 negatives+0.070
][/+0.068
Neg logit contributions
 Commando-0.104
earch-0.100
senal-0.097
bt-0.084
ould-0.084
Token wholly
Token ID18174
Feature activation-9.04e-02

Pos logit contributions
 tem+0.038
 persona+0.038
 mig+0.037
 negatives+0.035
][/+0.034
Neg logit contributions
 Commando-0.058
earch-0.056
senal-0.055
bt-0.048
ould-0.047
Token benevolent
Token ID45599
Feature activation-1.03e+00

Pos logit contributions
 Commando+0.086
earch+0.083
senal+0.081
bt+0.070
ould+0.070
Neg logit contributions
 tem-0.054
 persona-0.053
 mig-0.053
 negatives-0.050
][/-0.048
Token foundation
Token ID8489
Feature activation-4.66e-01

Pos logit contributions
 Commando+0.938
earch+0.897
senal+0.873
bt+0.757
ould+0.754
Neg logit contributions
 tem-0.660
 mig-0.646
 persona-0.645
 negatives-0.615
][/-0.594
Token,
Token ID11
Feature activation-2.37e-01

Pos logit contributions
 Commando+0.397
earch+0.383
senal+0.373
bt+0.318
ould+0.317
Neg logit contributions
 tem-0.324
 persona-0.318
 mig-0.316
 negatives-0.301
][/-0.292
Token due
Token ID2233
Feature activation-2.51e+00

Pos logit contributions
 Commando+0.209
earch+0.201
senal+0.196
bt+0.168
ould+0.168
Neg logit contributions
 tem-0.158
 persona-0.156
 mig-0.154
 negatives-0.147
][/-0.142
Token only
Token ID691
Feature activation-1.45e-01

Pos logit contributions
 Commando+2.129
earch+2.053
senal+2.003
bt+1.722
ould+1.703
Neg logit contributions
 tem-1.705
 persona-1.685
 mig-1.684
 negatives-1.604
][/-1.555
Token to
Token ID284
Feature activation+4.93e-03

Pos logit contributions
 Commando+0.124
earch+0.120
senal+0.117
bt+0.100
ould+0.099
Neg logit contributions
 tem-0.100
 persona-0.099
 mig-0.098
 negatives-0.093
][/-0.090
Token the
Token ID262
Feature activation-1.03e-02

Pos logit contributions
 tem+0.003
 persona+0.003
 mig+0.003
 negatives+0.003
][/+0.003
Neg logit contributions
 Commando-0.005
earch-0.004
senal-0.004
bt-0.004
ould-0.004
Token fact
Token ID1109
Feature activation-4.36e-01

Pos logit contributions
 Commando+0.010
earch+0.009
senal+0.009
bt+0.008
ould+0.008
Neg logit contributions
 tem-0.006
 persona-0.006
 mig-0.006
 negatives-0.006
][/-0.005
Token the
Token ID262
Feature activation+2.38e-02

Pos logit contributions
 Commando+0.325
earch+0.312
senal+0.303
bt+0.251
ould+0.250
Neg logit contributions
 tem-0.349
 persona-0.345
 mig-0.343
 negatives-0.328
][/-0.320
Tokenbin
Token ID8800
Feature activation-3.80e-01

Pos logit contributions
 Commando+0.032
earch+0.030
senal+0.027
bt+0.019
ould+0.018
Neg logit contributions
 tem-0.095
 persona-0.095
 mig-0.094
 negatives-0.092
 ethnicity-0.089
Token and
Token ID290
Feature activation-2.30e-01

Pos logit contributions
 Commando+0.320
earch+0.308
senal+0.301
ould+0.256
bt+0.256
Neg logit contributions
 tem-0.267
 persona-0.264
 mig-0.261
 negatives-0.250
][/-0.242
Token $
Token ID720
Feature activation-1.99e-01

Pos logit contributions
 Commando+0.203
earch+0.196
senal+0.192
bt+0.164
ould+0.164
Neg logit contributions
 tem-0.152
 persona-0.150
 mig-0.148
 negatives-0.141
][/-0.137
Token8
Token ID23
Feature activation-7.61e-01

Pos logit contributions
 Commando+0.161
earch+0.155
senal+0.148
bt+0.129
ould+0.127
Neg logit contributions
 tem-0.147
 persona-0.146
 mig-0.144
 negatives-0.138
 ethnicity-0.133
Token million
Token ID1510
Feature activation-3.89e-01

Pos logit contributions
 Commando+0.653
earch+0.631
senal+0.617
bt+0.528
ould+0.523
Neg logit contributions
 tem-0.522
 persona-0.514
 mig-0.508
 negatives-0.485
][/-0.470
Token due
Token ID2233
Feature activation-2.50e+00

Pos logit contributions
 Commando+0.334
earch+0.321
senal+0.312
bt+0.268
ould+0.266
Neg logit contributions
 tem-0.267
 persona-0.264
 mig-0.261
 negatives-0.248
][/-0.241
Token to
Token ID284
Feature activation-1.64e-01

Pos logit contributions
 Commando+2.034
earch+1.937
senal+1.897
bt+1.618
ould+1.596
Neg logit contributions
 tem-1.799
 persona-1.780
 mig-1.763
 negatives-1.701
][/-1.642
Token Carlos
Token ID17409
Feature activation+3.41e-01

Pos logit contributions
 Commando+0.136
earch+0.131
senal+0.128
bt+0.108
ould+0.108
Neg logit contributions
 tem-0.117
 persona-0.116
 mig-0.114
 negatives-0.109
][/-0.106
Token Quentin
Token ID42447
Feature activation+1.89e-01

Pos logit contributions
 tem+0.230
 persona+0.226
 mig+0.224
 negatives+0.213
][/+0.206
Neg logit contributions
 Commando-0.299
earch-0.289
senal-0.283
bt-0.241
ould-0.240
Token.
Token ID13
Feature activation-2.08e-01

Pos logit contributions
 tem+0.132
 persona+0.131
 mig+0.129
 negatives+0.123
][/+0.120
Neg logit contributions
 Commando-0.160
earch-0.155
senal-0.151
bt-0.128
ould-0.128
Token But
Token ID887
Feature activation+2.42e-02

Pos logit contributions
 Commando+0.175
earch+0.169
senal+0.165
bt+0.140
ould+0.139
Neg logit contributions
 tem-0.146
 persona-0.144
 mig-0.143
 negatives-0.136
][/-0.132
Tokenis
Token ID271
Feature activation+7.16e-01

Pos logit contributions
 tem+0.357
 persona+0.352
 mig+0.350
 negatives+0.334
][/+0.324
Neg logit contributions
 Commando-0.380
earch-0.368
senal-0.355
ould-0.305
bt-0.302
Tokenkered
Token ID28970
Feature activation-3.59e-01

Pos logit contributions
 tem+0.436
 persona+0.429
 mig+0.423
 negatives+0.399
][/+0.386
Neg logit contributions
 Commando-0.674
earch-0.652
senal-0.641
bt-0.551
ould-0.550
Token composer
Token ID26777
Feature activation-5.61e-01

Pos logit contributions
 Commando+0.320
earch+0.307
senal+0.300
bt+0.258
ould+0.257
Neg logit contributions
 tem-0.236
 persona-0.231
 mig-0.230
 negatives-0.219
][/-0.213
Token of
Token ID286
Feature activation-2.03e-01

Pos logit contributions
 Commando+0.474
earch+0.457
senal+0.447
bt+0.379
ould+0.377
Neg logit contributions
 tem-0.395
 persona-0.389
 mig-0.385
 negatives-0.368
][/-0.357
Token four
Token ID1440
Feature activation-1.96e-01

Pos logit contributions
 Commando+0.181
earch+0.174
senal+0.171
bt+0.146
ould+0.146
Neg logit contributions
 tem-0.133
 persona-0.131
 mig-0.130
 negatives-0.123
][/-0.119
Token sym
Token ID5659
Feature activation-2.50e+00

Pos logit contributions
 Commando+0.176
earch+0.170
senal+0.167
bt+0.143
ould+0.142
Neg logit contributions
 tem-0.128
 persona-0.126
 mig-0.125
 negatives-0.118
][/-0.114
Tokenph
Token ID746
Feature activation-4.85e-01

Pos logit contributions
 Commando+1.476
earch+1.431
senal+1.372
bt+1.108
ould+1.088
Neg logit contributions
 tem-2.343
 persona-2.338
 mig-2.305
 negatives-2.237
][/-2.187
Tokenonies
Token ID17300
Feature activation-1.29e+00

Pos logit contributions
 Commando+0.423
earch+0.416
senal+0.398
ould+0.349
bt+0.347
Neg logit contributions
 tem-0.319
 persona-0.318
 mig-0.314
 negatives-0.299
 ethnicity-0.288
Token,
Token ID11
Feature activation-3.60e-01

Pos logit contributions
 Commando+1.095
earch+1.055
senal+1.025
bt+0.873
ould+0.868
Neg logit contributions
 tem-0.902
 persona-0.890
 mig-0.882
 negatives-0.842
][/-0.819
Token has
Token ID468
Feature activation-3.67e-01

Pos logit contributions
 Commando+0.319
earch+0.308
senal+0.301
bt+0.258
ould+0.256
Neg logit contributions
 tem-0.239
 persona-0.235
 mig-0.232
 negatives-0.221
][/-0.214
Token been
Token ID587
Feature activation-2.20e-01

Pos logit contributions
 Commando+0.345
earch+0.333
senal+0.321
bt+0.282
ould+0.280
Neg logit contributions
 tem-0.220
 persona-0.217
 mig-0.215
 negatives-0.204
][/-0.199
Token so
Token ID523
Feature activation-1.31e-01

Pos logit contributions
 tem+0.443
 persona+0.438
 mig+0.432
 negatives+0.410
][/+0.399
Neg logit contributions
 Commando-0.576
earch-0.556
senal-0.541
bt-0.462
ould-0.461
Token few
Token ID1178
Feature activation+5.12e-01

Pos logit contributions
 Commando+0.121
earch+0.117
senal+0.115
bt+0.099
ould+0.098
Neg logit contributions
 tem-0.082
 persona-0.081
 mig-0.080
 negatives-0.075
][/-0.073
Token have
Token ID423
Feature activation-1.26e-01

Pos logit contributions
 tem+0.298
 persona+0.293
 mig+0.289
 negatives+0.272
][/+0.262
Neg logit contributions
 Commando-0.496
earch-0.481
senal-0.471
bt-0.409
ould-0.408
Token it
Token ID340
Feature activation+4.67e-02

Pos logit contributions
 Commando+0.118
earch+0.114
senal+0.111
bt+0.096
ould+0.096
Neg logit contributions
 tem-0.077
 persona-0.076
 mig-0.074
 negatives-0.070
][/-0.068
Token,
Token ID11
Feature activation-9.77e-02

Pos logit contributions
 tem+0.031
 persona+0.030
 mig+0.030
 negatives+0.028
][/+0.028
Neg logit contributions
 Commando-0.042
earch-0.040
senal-0.039
bt-0.034
ould-0.033
Token due
Token ID2233
Feature activation-2.50e+00

Pos logit contributions
 Commando+0.087
earch+0.084
senal+0.081
bt+0.070
ould+0.070
Neg logit contributions
 tem-0.064
 persona-0.064
 mig-0.063
 negatives-0.060
][/-0.058
Token to
Token ID284
Feature activation-4.68e-02

Pos logit contributions
 Commando+2.093
earch+2.015
senal+1.997
bt+1.683
ould+1.658
Neg logit contributions
 tem-1.734
 persona-1.715
 mig-1.699
 negatives-1.620
][/-1.573
Token the
Token ID262
Feature activation-6.29e-02

Pos logit contributions
 Commando+0.044
earch+0.042
senal+0.041
bt+0.036
ould+0.035
Neg logit contributions
 tem-0.028
 persona-0.028
 mig-0.028
 negatives-0.026
][/-0.025
Token custom
Token ID2183
Feature activation+1.02e+00

Pos logit contributions
 Commando+0.059
earch+0.057
senal+0.056
bt+0.049
ould+0.048
Neg logit contributions
 tem-0.038
 persona-0.037
 mig-0.037
 negatives-0.035
][/-0.034
Token apps
Token ID6725
Feature activation+3.75e-01

Pos logit contributions
 tem+0.646
 persona+0.636
 mig+0.628
 negatives+0.595
][/+0.575
Neg logit contributions
 Commando-0.939
earch-0.909
senal-0.892
bt-0.766
ould-0.764
Token HTC
Token ID22063
Feature activation+5.04e-01

Pos logit contributions
 tem+0.255
 persona+0.252
 mig+0.250
 negatives+0.238
][/+0.231
Neg logit contributions
 Commando-0.324
earch-0.314
senal-0.305
bt-0.260
ould-0.259
Token 1989
Token ID11104
Feature activation+2.37e-01

Pos logit contributions
 tem+0.190
 persona+0.187
 mig+0.186
 negatives+0.177
][/+0.171
Neg logit contributions
 Commando-0.260
earch-0.250
senal-0.243
ould-0.210
bt-0.210
Token,
Token ID11
Feature activation+1.58e-01

Pos logit contributions
 tem+0.167
 persona+0.165
 mig+0.163
 negatives+0.156
][/+0.151
Neg logit contributions
 Commando-0.200
earch-0.193
senal-0.189
bt-0.160
ould-0.159
Token and
Token ID290
Feature activation+8.19e-02

Pos logit contributions
 tem+0.108
 persona+0.106
 mig+0.105
 negatives+0.100
][/+0.097
Neg logit contributions
 Commando-0.137
earch-0.132
senal-0.129
bt-0.110
ould-0.109
Token was
Token ID373
Feature activation+1.65e-01

Pos logit contributions
 tem+0.054
 persona+0.053
 mig+0.053
 negatives+0.050
][/+0.049
Neg logit contributions
 Commando-0.073
earch-0.070
senal-0.068
bt-0.059
ould-0.058
Token retired
Token ID9880
Feature activation-3.86e-02

Pos logit contributions
 tem+0.104
 persona+0.102
 mig+0.101
 negatives+0.096
][/+0.093
Neg logit contributions
 Commando-0.152
earch-0.147
senal-0.143
bt-0.124
ould-0.123
Token due
Token ID2233
Feature activation-2.49e+00

Pos logit contributions
 Commando+0.033
earch+0.032
senal+0.031
bt+0.026
ould+0.026
Neg logit contributions
 tem-0.027
 persona-0.027
 mig-0.026
 negatives-0.025
][/-0.024
Token to
Token ID284
Feature activation-1.77e-02

Pos logit contributions
 Commando+1.936
earch+1.856
senal+1.757
ould+1.518
bt+1.513
Neg logit contributions
 tem-1.850
 persona-1.828
 mig-1.813
 negatives-1.729
][/-1.678
Token the
Token ID262
Feature activation+5.25e-02

Pos logit contributions
 Commando+0.017
earch+0.016
senal+0.016
bt+0.014
ould+0.014
Neg logit contributions
 tem-0.011
 persona-0.010
 mig-0.010
 negatives-0.010
][/-0.009
Token hull
Token ID23644
Feature activation+3.85e-01

Pos logit contributions
 tem+0.032
 persona+0.032
 mig+0.031
 negatives+0.029
][/+0.029
Neg logit contributions
 Commando-0.049
earch-0.047
senal-0.046
bt-0.040
ould-0.040
Token being
Token ID852
Feature activation-5.17e-02

Pos logit contributions
 tem+0.223
 persona+0.220
 mig+0.217
 negatives+0.204
][/+0.198
Neg logit contributions
 Commando-0.372
earch-0.359
senal-0.351
bt-0.306
ould-0.306
Token worn
Token ID12666
Feature activation+5.90e-01

Pos logit contributions
 Commando+0.050
earch+0.048
senal+0.047
bt+0.041
ould+0.041
Neg logit contributions
 tem-0.030
 persona-0.029
 mig-0.029
 negatives-0.027
][/-0.026
TokenGood
Token ID10248
Feature activation-2.71e-01

Pos logit contributions
 Commando+0.280
earch+0.269
senal+0.263
bt+0.218
ould+0.217
Neg logit contributions
 tem-0.294
 persona-0.290
 mig-0.287
 negatives-0.275
][/-0.268
Tokenrich
Token ID7527
Feature activation+2.93e-02

Pos logit contributions
 Commando+0.219
earch+0.212
senal+0.210
ould+0.173
bt+0.172
Neg logit contributions
 tem-0.202
 persona-0.199
 mig-0.195
 negatives-0.185
][/-0.181
Token All
Token ID1439
Feature activation+1.66e-01

Pos logit contributions
 tem+0.020
 persona+0.020
 mig+0.020
 negatives+0.019
][/+0.018
Neg logit contributions
 Commando-0.025
earch-0.024
senal-0.024
bt-0.020
ould-0.020
Token-
Token ID12
Feature activation-1.24e-01

Pos logit contributions
 tem+0.131
 persona+0.130
 mig+0.129
 negatives+0.123
][/+0.120
Neg logit contributions
 Commando-0.126
earch-0.121
senal-0.117
bt-0.098
ould-0.097
TokenTer
Token ID15156
Feature activation-1.14e-01

Pos logit contributions
 Commando+0.093
earch+0.090
senal+0.087
bt+0.072
ould+0.071
Neg logit contributions
 tem-0.099
 persona-0.098
 mig-0.098
 negatives-0.094
][/-0.091
Tokenrain
Token ID3201
Feature activation-2.47e+00

Pos logit contributions
 Commando+0.065
earch+0.062
senal+0.059
bt+0.046
ould+0.045
Neg logit contributions
 tem-0.112
 persona-0.111
 mig-0.110
 negatives-0.106
][/-0.104
Token KO
Token ID21054
Feature activation-4.34e-01

Pos logit contributions
 Commando+2.121
earch+2.040
senal+1.939
bt+1.672
ould+1.663
Neg logit contributions
 tem-1.696
 persona-1.687
 mig-1.676
 negatives-1.600
][/-1.552
Token2
Token ID17
Feature activation-2.06e-01

Pos logit contributions
 Commando+0.359
earch+0.347
senal+0.338
bt+0.285
ould+0.284
Neg logit contributions
 tem-0.313
 persona-0.309
 mig-0.306
 negatives-0.292
][/-0.283
Token tires
Token ID17920
Feature activation-9.27e-01

Pos logit contributions
 Commando+0.180
earch+0.174
senal+0.170
bt+0.145
ould+0.144
Neg logit contributions
 tem-0.138
 persona-0.136
 mig-0.135
 negatives-0.128
][/-0.124
Token specifically
Token ID5734
Feature activation-9.59e-02

Pos logit contributions
 Commando+0.773
earch+0.745
senal+0.722
bt+0.612
ould+0.610
Neg logit contributions
 tem-0.662
 persona-0.655
 mig-0.649
 negatives-0.620
][/-0.601
Token designed
Token ID3562
Feature activation-5.54e-01

Pos logit contributions
 Commando+0.083
earch+0.080
senal+0.077
bt+0.066
ould+0.066
Neg logit contributions
 tem-0.066
 persona-0.065
 mig-0.064
 negatives-0.061
][/-0.059
Token not
Token ID407
Feature activation+1.35e-01

Pos logit contributions
 tem+0.106
 persona+0.104
 mig+0.103
 negatives+0.097
][/+0.094
Neg logit contributions
 Commando-0.155
earch-0.150
senal-0.146
bt-0.127
ould-0.126
Token forget
Token ID6044
Feature activation-2.64e-01

Pos logit contributions
 tem+0.083
 persona+0.082
 mig+0.081
 negatives+0.077
][/+0.074
Neg logit contributions
 Commando-0.125
earch-0.121
senal-0.119
bt-0.103
ould-0.102
Token….
Token ID11580
Feature activation-5.69e-01

Pos logit contributions
 Commando+0.223
earch+0.216
senal+0.211
bt+0.179
ould+0.178
Neg logit contributions
 tem-0.185
 persona-0.182
 mig-0.180
 negatives-0.172
][/-0.167
Token MY
Token ID17615
Feature activation+4.30e-02

Pos logit contributions
 Commando+0.479
earch+0.462
senal+0.451
bt+0.383
ould+0.381
Neg logit contributions
 tem-0.402
 persona-0.397
 mig-0.392
 negatives-0.374
][/-0.363
Token HA
Token ID14558
Feature activation+3.96e-02

Pos logit contributions
 tem+0.029
 persona+0.029
 mig+0.029
 negatives+0.027
][/+0.026
Neg logit contributions
 Commando-0.037
earch-0.036
senal-0.035
bt-0.030
ould-0.030
TokenIR
Token ID4663
Feature activation+1.67e+00

Pos logit contributions
 tem+0.033
 persona+0.033
 mig+0.032
 negatives+0.031
][/+0.030
Neg logit contributions
 Commando-0.028
earch-0.027
senal-0.026
bt-0.022
ould-0.021
Token.
Token ID13
Feature activation+3.70e-01

Pos logit contributions
 tem+1.156
 persona+1.139
 mig+1.126
 negatives+1.070
][/+1.042
Neg logit contributions
 Commando-1.435
earch-1.390
senal-1.365
bt-1.152
ould-1.148
Token My
Token ID2011
Feature activation-8.34e-02

Pos logit contributions
 tem+0.263
 persona+0.259
 mig+0.257
 negatives+0.245
][/+0.237
Neg logit contributions
 Commando-0.310
earch-0.299
senal-0.293
bt-0.247
ould-0.246
Token glorious
Token ID21140
Feature activation-2.98e-01

Pos logit contributions
 Commando+0.079
earch+0.076
senal+0.075
bt+0.064
ould+0.064
Neg logit contributions
 tem-0.051
 persona-0.050
 mig-0.049
 negatives-0.046
][/-0.045
Token man
Token ID582
Feature activation+5.95e-01

Pos logit contributions
 Commando+0.280
earch+0.271
senal+0.265
bt+0.229
ould+0.228
Neg logit contributions
 tem-0.181
 persona-0.178
 mig-0.176
 negatives-0.167
][/-0.161
Tokene
Token ID68
Feature activation+3.80e-01

Pos logit contributions
 tem+0.441
 persona+0.437
 mig+0.429
 negatives+0.411
][/+0.397
Neg logit contributions
 Commando-0.481
earch-0.464
senal-0.456
ould-0.379
bt-0.379
Tokenl
Token ID75
Feature activation+4.79e-03

Pos logit contributions
 tem+0.463
 persona+0.457
 mig+0.452
 negatives+0.434
][/+0.420
Neg logit contributions
 Commando-0.459
earch-0.444
senal-0.430
bt-0.361
ould-0.359
Tokenort
Token ID419
Feature activation+2.32e-01

Pos logit contributions
 tem+0.004
 persona+0.004
 mig+0.003
 negatives+0.003
 ethnicity+0.003
Neg logit contributions
 Commando-0.004
earch-0.004
senal-0.004
bt-0.003
ould-0.003
Tokenot
Token ID313
Feature activation+7.14e-01

Pos logit contributions
 tem+0.154
 persona+0.152
 mig+0.149
 negatives+0.144
 ethnicity+0.137
Neg logit contributions
 Commando-0.205
earch-0.201
senal-0.194
bt-0.169
ould-0.167
Tokenft
Token ID701
Feature activation+1.36e+00

Pos logit contributions
 tem+0.479
 persona+0.472
 mig+0.465
 negatives+0.445
 ethnicity+0.426
Neg logit contributions
 Commando-0.626
earch-0.611
senal-0.587
bt-0.513
ould-0.509
Tokenw
Token ID86
Feature activation-3.57e-02

Pos logit contributions
 tem+0.764
 persona+0.752
 mig+0.740
 negatives+0.700
][/+0.668
Neg logit contributions
 Commando-1.333
earch-1.298
senal-1.267
bt-1.111
ould-1.105
Token>
Token ID29
Feature activation+1.69e+00

Pos logit contributions
 Commando+0.040
earch+0.039
senal+0.038
bt+0.035
ould+0.034
Neg logit contributions
 tem-0.015
 persona-0.015
 mig-0.014
 negatives-0.013
 ethnicity-0.012
Token loving
Token ID14442
Feature activation-7.34e-01

Pos logit contributions
 tem+1.098
 persona+1.079
 mig+1.068
 negatives+1.011
][/+0.977
Neg logit contributions
 Commando-1.519
earch-1.471
senal-1.446
bt-1.234
ould-1.230
Token this
Token ID428
Feature activation+3.56e-01

Pos logit contributions
 Commando+0.659
earch+0.635
senal+0.621
bt+0.536
ould+0.531
Neg logit contributions
 tem-0.476
 persona-0.470
 mig-0.462
 negatives-0.442
][/-0.428
Token meme
Token ID25336
Feature activation-1.95e-01

Pos logit contributions
 tem+0.224
 persona+0.220
 mig+0.218
 negatives+0.207
][/+0.199
Neg logit contributions
 Commando-0.327
earch-0.316
senal-0.310
bt-0.267
ould-0.266
Token [
Token ID685
Feature activation-2.10e-01

Pos logit contributions
 Commando+0.179
earch+0.174
senal+0.169
bt+0.147
ould+0.146
Neg logit contributions
 tem-0.122
 persona-0.120
 mig-0.118
 negatives-0.113
][/-0.109
Token10
Token ID940
Feature activation-2.49e-01

Pos logit contributions
 Commando+0.122
earch+0.117
senal+0.107
bt+0.091
ould+0.089
Neg logit contributions
 tem-0.201
 persona-0.200
 mig-0.198
 negatives-0.192
 ethnicity-0.186
Token.
Token ID13
Feature activation+4.13e-01

Pos logit contributions
 tem+0.079
 persona+0.078
 mig+0.077
 negatives+0.073
][/+0.071
Neg logit contributions
 Commando-0.103
earch-0.100
senal-0.097
bt-0.083
ould-0.083
Token See
Token ID4091
Feature activation+4.08e-01

Pos logit contributions
 tem+0.296
 persona+0.292
 mig+0.289
 negatives+0.276
][/+0.268
Neg logit contributions
 Commando-0.343
earch-0.331
senal-0.324
bt-0.273
ould-0.272
Token L
Token ID406
Feature activation+3.18e-01

Pos logit contributions
 tem+0.225
 persona+0.221
 mig+0.218
 negatives+0.205
][/+0.198
Neg logit contributions
 Commando-0.406
earch-0.394
senal-0.385
bt-0.337
ould-0.336
Tokenings
Token ID654
Feature activation-7.10e-02

Pos logit contributions
 tem+0.162
 persona+0.160
 mig+0.157
 negatives+0.148
][/+0.140
Neg logit contributions
 Commando-0.329
earch-0.322
senal-0.313
ould-0.278
bt-0.278
Token,
Token ID11
Feature activation+5.70e-01

Pos logit contributions
 Commando+0.061
earch+0.059
senal+0.058
bt+0.050
ould+0.049
Neg logit contributions
 tem-0.048
 persona-0.048
 mig-0.047
 negatives-0.045
][/-0.044
Token p
Token ID279
Feature activation+1.56e+00

Pos logit contributions
 tem+0.403
 persona+0.397
 mig+0.393
 negatives+0.375
][/+0.364
Neg logit contributions
 Commando-0.480
earch-0.462
senal-0.451
bt-0.383
ould-0.381
Token.
Token ID13
Feature activation+4.03e-01

Pos logit contributions
 tem+1.063
 persona+1.044
 mig+1.035
 negatives+0.981
][/+0.951
Neg logit contributions
 Commando-1.369
earch-1.307
senal-1.289
bt-1.080
ould-1.075
Token 232
Token ID31773
Feature activation+1.38e-01

Pos logit contributions
 tem+0.307
 persona+0.304
 mig+0.300
 negatives+0.288
][/+0.279
Neg logit contributions
 Commando-0.317
earch-0.305
senal-0.298
bt-0.249
ould-0.248
Token n
Token ID299
Feature activation+1.78e+00

Pos logit contributions
 tem+0.096
 persona+0.094
 mig+0.093
 negatives+0.089
][/+0.086
Neg logit contributions
 Commando-0.118
earch-0.114
senal-0.111
bt-0.095
ould-0.094
Token.
Token ID13
Feature activation+4.01e-01

Pos logit contributions
 tem+1.222
 persona+1.194
 mig+1.188
 negatives+1.127
][/+1.089
Neg logit contributions
 Commando-1.542
earch-1.476
senal-1.459
bt-1.223
ould-1.223
Token1
Token ID16
Feature activation+6.94e-01

Pos logit contributions
 tem+0.281
 persona+0.277
 mig+0.274
 negatives+0.261
][/+0.253
Neg logit contributions
 Commando-0.340
earch-0.328
senal-0.321
bt-0.272
ould-0.271
Token.
Token ID13
Feature activation+2.78e-01

Pos logit contributions
 tem+0.703
 persona+0.692
 mig+0.681
 negatives+0.652
 ethnicity+0.625
Neg logit contributions
 Commando-0.906
earch-0.877
senal-0.861
ould-0.731
bt-0.727
Token
Token ID198
Feature activation+1.20e-01

Pos logit contributions
 tem+0.201
 persona+0.198
 mig+0.196
 negatives+0.187
][/+0.182
Neg logit contributions
 Commando-0.229
earch-0.221
senal-0.216
bt-0.182
ould-0.181
Token
Token ID198
Feature activation+2.47e-01

Pos logit contributions
 tem+0.095
 persona+0.094
 mig+0.093
 negatives+0.089
 ethnicity+0.086
Neg logit contributions
 Commando-0.090
earch-0.087
senal-0.080
bt-0.071
ould-0.070
Token2001
Token ID14585
Feature activation+1.76e-01

Pos logit contributions
 tem+0.198
 persona+0.196
 mig+0.194
 negatives+0.186
][/+0.181
Neg logit contributions
 Commando-0.183
earch-0.177
senal-0.171
bt-0.143
ould-0.142
Token-
Token ID12
Feature activation+3.76e-01

Pos logit contributions
 tem+0.128
 persona+0.127
 mig+0.125
 negatives+0.120
][/+0.116
Neg logit contributions
 Commando-0.145
earch-0.140
senal-0.137
bt-0.115
ould-0.115
Tokenpresent
Token ID25579
Feature activation+1.56e+00

Pos logit contributions
 tem+0.282
 persona+0.279
 mig+0.276
 negatives+0.265
][/+0.256
Neg logit contributions
 Commando-0.300
earch-0.289
senal-0.281
bt-0.237
ould-0.236
Token Congress
Token ID3162
Feature activation+6.57e-01

Pos logit contributions
 tem+1.146
 persona+1.126
 mig+1.117
 negatives+1.063
][/+1.031
Neg logit contributions
 Commando-1.273
earch-1.231
senal-1.212
bt-1.014
ould-1.011
Token debates
Token ID15389
Feature activation+1.74e-01

Pos logit contributions
 tem+0.378
 persona+0.371
 mig+0.366
 negatives+0.345
][/+0.332
Neg logit contributions
 Commando-0.640
earch-0.621
senal-0.609
bt-0.529
ould-0.527
Token legislation
Token ID5520
Feature activation-3.30e-02

Pos logit contributions
 tem+0.109
 persona+0.107
 mig+0.106
 negatives+0.100
][/+0.097
Neg logit contributions
 Commando-0.160
earch-0.155
senal-0.152
bt-0.131
ould-0.130
Token on
Token ID319
Feature activation+2.44e-01

Pos logit contributions
 Commando+0.027
earch+0.026
senal+0.026
bt+0.022
ould+0.022
Neg logit contributions
 tem-0.024
 persona-0.023
 mig-0.023
 negatives-0.022
][/-0.021
Token human
Token ID1692
Feature activation-2.66e-01

Pos logit contributions
 tem+0.151
 persona+0.148
 mig+0.146
 negatives+0.139
][/+0.134
Neg logit contributions
 Commando-0.227
earch-0.219
senal-0.215
bt-0.185
ould-0.184
Token 1986
Token ID12113
Feature activation+6.75e-01

Pos logit contributions
 tem+0.660
 persona+0.651
 mig+0.645
 negatives+0.617
][/+0.601
Neg logit contributions
 Commando-0.626
earch-0.602
senal-0.590
bt-0.485
ould-0.484
Token ).
Token ID6739
Feature activation+9.45e-01

Pos logit contributions
 tem+0.361
 persona+0.355
 mig+0.350
 negatives+0.328
][/+0.313
Neg logit contributions
 Commando-0.685
earch-0.666
senal-0.654
bt-0.571
ould-0.569
Token The
Token ID383
Feature activation+4.18e-01

Pos logit contributions
 tem+0.696
 persona+0.687
 mig+0.679
 negatives+0.650
][/+0.630
Neg logit contributions
 Commando-0.771
earch-0.740
senal-0.722
bt-0.609
ould-0.606
Token Ol
Token ID6544
Feature activation-3.90e-01

Pos logit contributions
 tem+0.234
 persona+0.229
 mig+0.227
 negatives+0.212
][/+0.206
Neg logit contributions
 Commando-0.414
earch-0.400
senal-0.392
bt-0.342
ould-0.341
Tokenwe
Token ID732
Feature activation+5.85e-01

Pos logit contributions
 Commando+0.284
earch+0.272
senal+0.266
bt+0.218
ould+0.217
Neg logit contributions
 tem-0.321
 persona-0.317
 mig-0.314
 negatives-0.301
][/-0.294
Tokenus
Token ID385
Feature activation+1.73e+00

Pos logit contributions
 tem+0.478
 persona+0.472
 mig+0.468
 negatives+0.449
][/+0.437
Neg logit contributions
 Commando-0.429
earch-0.412
senal-0.401
bt-0.330
ould-0.328
Token B
Token ID347
Feature activation-2.62e-01

Pos logit contributions
 tem+1.044
 persona+1.025
 mig+1.014
 negatives+0.957
][/+0.925
Neg logit contributions
 Commando-1.634
earch-1.582
senal-1.551
bt-1.340
ould-1.336
Tokenully
Token ID2132
Feature activation+3.31e-01

Pos logit contributions
 Commando+0.207
earch+0.200
senal+0.194
ould+0.165
bt+0.164
Neg logit contributions
 tem-0.198
 persona-0.195
 mig-0.194
 negatives-0.186
][/-0.181
Token/
Token ID14
Feature activation-8.46e-03

Pos logit contributions
 tem+0.214
 mig+0.209
 persona+0.209
 negatives+0.198
][/+0.191
Neg logit contributions
 Commando-0.300
earch-0.288
senal-0.281
bt-0.243
ould-0.241
TokenVict
Token ID21944
Feature activation+7.24e-02

Pos logit contributions
 Commando+0.006
earch+0.006
senal+0.006
bt+0.005
ould+0.005
Neg logit contributions
 tem-0.007
 persona-0.007
 mig-0.007
 negatives-0.006
][/-0.006
Tokenim
Token ID320
Feature activation+1.73e+00

Pos logit contributions
 tem+0.045
 persona+0.044
 mig+0.044
 negatives+0.042
][/+0.040
Neg logit contributions
 Commando-0.067
earch-0.065
senal-0.063
bt-0.055
ould-0.055
Token the
Token ID262
Feature activation+2.37e-01

Pos logit contributions
 tem+0.052
 mig+0.051
 persona+0.051
 negatives+0.048
][/+0.046
Neg logit contributions
 Commando-0.079
earch-0.076
senal-0.075
bt-0.065
ould-0.065
Token content
Token ID2695
Feature activation+2.49e-02

Pos logit contributions
 tem+0.129
 persona+0.127
 mig+0.126
 negatives+0.117
][/+0.113
Neg logit contributions
 Commando-0.238
earch-0.231
senal-0.226
bt-0.198
ould-0.197
Token.
Token ID13
Feature activation+2.09e-01

Pos logit contributions
 tem+0.017
 persona+0.017
 mig+0.017
 negatives+0.016
][/+0.016
Neg logit contributions
 Commando-0.021
earch-0.020
senal-0.019
bt-0.017
ould-0.017
Token Some
Token ID2773
Feature activation+3.55e-01

Pos logit contributions
 tem+0.151
 persona+0.148
 mig+0.147
 negatives+0.140
][/+0.137
Neg logit contributions
 Commando-0.173
earch-0.167
senal-0.164
bt-0.137
ould-0.137
Token of
Token ID286
Feature activation+3.46e-01

Pos logit contributions
 tem+0.197
 persona+0.194
 mig+0.191
 negatives+0.179
][/+0.173
Neg logit contributions
 Commando-0.352
earch-0.341
senal-0.334
bt-0.292
ould-0.291
Token the
Token ID262
Feature activation+3.23e-01

Pos logit contributions
 tem+0.226
 persona+0.222
 mig+0.220
 negatives+0.208
][/+0.202
Neg logit contributions
 Commando-0.310
earch-0.300
senal-0.294
bt-0.251
ould-0.251
Token content
Token ID2695
Feature activation+3.88e-02

Pos logit contributions
 tem+0.160
 persona+0.157
 mig+0.156
 negatives+0.144
][/+0.139
Neg logit contributions
 Commando-0.339
earch-0.328
senal-0.322
bt-0.284
ould-0.282
Token includes
Token ID3407
Feature activation+2.43e-01

Pos logit contributions
 tem+0.025
 persona+0.025
 mig+0.025
 negatives+0.023
][/+0.022
Neg logit contributions
 Commando-0.035
earch-0.034
senal-0.032
ould-0.028
bt-0.028
Token the
Token ID262
Feature activation+3.26e-01

Pos logit contributions
 tem+0.127
 persona+0.125
 mig+0.123
 negatives+0.115
][/+0.111
Neg logit contributions
 Commando-0.249
earch-0.241
senal-0.236
bt-0.207
ould-0.206
Token original
Token ID2656
Feature activation+2.08e-02

Pos logit contributions
 tem+0.170
 persona+0.167
 mig+0.165
 negatives+0.154
][/+0.148
Neg logit contributions
 Commando-0.336
earch-0.325
senal-0.319
bt-0.280
ould-0.278
Token publisher
Token ID9991
Feature activation+6.25e-01

Pos logit contributions
 tem+0.012
 persona+0.011
 mig+0.011
 negatives+0.011
][/+0.010
Neg logit contributions
 Commando-0.021
earch-0.020
senal-0.020
bt-0.017
ould-0.017
Token aren
Token ID3588
Feature activation+2.19e-01

Pos logit contributions
 tem+0.863
 persona+0.850
 mig+0.840
 negatives+0.799
][/+0.773
Neg logit contributions
 Commando-1.127
earch-1.090
senal-1.077
bt-0.914
ould-0.911
Token't
Token ID470
Feature activation-1.16e-01

Pos logit contributions
 tem+0.133
 persona+0.133
 mig+0.130
 negatives+0.124
 ethnicity+0.118
Neg logit contributions
 Commando-0.204
earch-0.197
senal-0.182
bt-0.171
ould-0.168
Token yet
Token ID1865
Feature activation+8.43e-01

Pos logit contributions
 Commando+0.114
earch+0.110
senal+0.108
bt+0.094
ould+0.094
Neg logit contributions
 tem-0.066
 persona-0.065
 mig-0.064
 negatives-0.060
][/-0.058
Token clear
Token ID1598
Feature activation+3.15e-01

Pos logit contributions
 tem+0.453
 persona+0.445
 mig+0.438
 negatives+0.411
][/+0.394
Neg logit contributions
 Commando-0.854
earch-0.828
senal-0.815
bt-0.711
ould-0.709
Token,
Token ID11
Feature activation+1.44e-01

Pos logit contributions
 tem+0.225
 persona+0.222
 mig+0.220
 negatives+0.210
][/+0.205
Neg logit contributions
 Commando-0.261
earch-0.252
senal-0.244
bt-0.208
ould-0.207
Token but
Token ID475
Feature activation+1.95e-01

Pos logit contributions
 tem+0.087
 persona+0.085
 mig+0.084
 negatives+0.080
][/+0.077
Neg logit contributions
 Commando-0.136
earch-0.131
senal-0.128
bt-0.111
ould-0.111
Token Den
Token ID5601
Feature activation-2.05e-01

Pos logit contributions
 tem+0.119
 persona+0.117
 mig+0.115
 negatives+0.109
][/+0.106
Neg logit contributions
 Commando-0.184
earch-0.178
senal-0.174
bt-0.150
ould-0.149
Tokenelle
Token ID13485
Feature activation-3.68e-01

Pos logit contributions
 Commando+0.163
earch+0.157
senal+0.153
bt+0.129
ould+0.127
Neg logit contributions
 tem-0.155
 persona-0.153
 mig-0.151
 negatives-0.145
][/-0.141
Token Dixon
Token ID31178
Feature activation+8.43e-01

Pos logit contributions
 Commando+0.299
earch+0.288
senal+0.280
bt+0.236
ould+0.235
Neg logit contributions
 tem-0.272
 persona-0.268
 mig-0.266
 negatives-0.254
][/-0.247
Token,
Token ID11
Feature activation+5.28e-02

Pos logit contributions
 tem+0.600
 persona+0.591
 mig+0.585
 negatives+0.558
][/+0.541
Neg logit contributions
 Commando-0.706
earch-0.681
senal-0.668
bt-0.563
ould-0.562
Token Mozilla
Token ID29258
Feature activation+9.95e-01

Pos logit contributions
 tem+0.030
 persona+0.029
 mig+0.029
 negatives+0.027
][/+0.026
Neg logit contributions
 Commando-0.052
earch-0.050
senal-0.049
bt-0.043
ould-0.042
Token him
Token ID683
Feature activation+4.96e-02

Pos logit contributions
 tem+0.064
 persona+0.063
 mig+0.063
 negatives+0.060
][/+0.058
Neg logit contributions
 Commando-0.072
earch-0.069
senal-0.067
bt-0.057
ould-0.056
Token special
Token ID2041
Feature activation-2.08e-01

Pos logit contributions
 tem+0.032
 persona+0.032
 mig+0.031
 negatives+0.030
][/+0.029
Neg logit contributions
 Commando-0.045
earch-0.043
senal-0.042
bt-0.036
ould-0.036
Token;
Token ID26
Feature activation+7.45e-01

Pos logit contributions
 Commando+0.179
earch+0.172
senal+0.166
bt+0.143
ould+0.142
Neg logit contributions
 tem-0.142
 persona-0.140
 mig-0.139
 negatives-0.132
][/-0.129
Token that
Token ID326
Feature activation+2.76e-01

Pos logit contributions
 tem+0.512
 persona+0.504
 mig+0.499
 negatives+0.474
][/+0.460
Neg logit contributions
 Commando-0.643
earch-0.620
senal-0.606
bt-0.516
ould-0.514
Token's
Token ID338
Feature activation+1.16e-01

Pos logit contributions
 tem+0.183
 persona+0.180
 mig+0.178
 negatives+0.169
][/+0.164
Neg logit contributions
 Commando-0.244
earch-0.235
senal-0.230
bt-0.197
ould-0.196
Token what
Token ID644
Feature activation+5.76e-01

Pos logit contributions
 tem+0.074
 persona+0.072
 mig+0.072
 negatives+0.068
][/+0.066
Neg logit contributions
 Commando-0.106
earch-0.102
senal-0.100
bt-0.086
ould-0.086
Token makes
Token ID1838
Feature activation-1.30e-02

Pos logit contributions
 tem+0.318
 persona+0.312
 mig+0.308
 negatives+0.289
][/+0.278
Neg logit contributions
 Commando-0.574
earch-0.556
senal-0.546
bt-0.476
ould-0.475
Token Dick
Token ID11740
Feature activation-5.15e-01

Pos logit contributions
 Commando+0.010
earch+0.010
senal+0.009
bt+0.008
ould+0.008
Neg logit contributions
 tem-0.010
 persona-0.010
 mig-0.010
 negatives-0.010
][/-0.009
Token Dick
Token ID11740
Feature activation-4.46e-01

Pos logit contributions
 Commando+0.476
earch+0.458
senal+0.449
bt+0.385
ould+0.384
Neg logit contributions
 tem-0.322
 persona-0.315
 mig-0.314
 negatives-0.297
][/-0.287
Token.
Token ID13
Feature activation+7.45e-02

Pos logit contributions
 Commando+0.398
earch+0.383
senal+0.375
bt+0.320
ould+0.319
Neg logit contributions
 tem-0.295
 persona-0.288
 mig-0.287
 negatives-0.272
][/-0.264
Token Even
Token ID3412
Feature activation+1.25e-01

Pos logit contributions
 tem+0.052
 persona+0.051
 mig+0.051
 negatives+0.049
][/+0.047
Neg logit contributions
 Commando-0.063
earch-0.061
senal-0.060
bt-0.051
ould-0.050
Token
Token ID564
Feature activation+1.26e-01

Pos logit contributions
 tem+0.091
 persona+0.090
 mig+0.089
 negatives+0.084
][/+0.082
Neg logit contributions
 Commando-0.109
earch-0.105
senal-0.103
bt-0.087
ould-0.087
Token
Token ID250
Feature activation+1.85e-02

Pos logit contributions
 tem+0.106
 persona+0.105
 mig+0.104
 negatives+0.100
][/+0.098
Neg logit contributions
 Commando-0.089
earch-0.085
senal-0.083
bt-0.067
ould-0.067
TokenWe
Token ID1135
Feature activation+4.42e-01

Pos logit contributions
 tem+0.013
 persona+0.013
 mig+0.013
 negatives+0.012
][/+0.012
Neg logit contributions
 Commando-0.015
earch-0.015
senal-0.014
bt-0.012
ould-0.012
Token
Token ID447
Feature activation+4.61e-01

Pos logit contributions
 tem+0.297
 persona+0.292
 mig+0.289
 negatives+0.275
][/+0.266
Neg logit contributions
 Commando-0.388
earch-0.375
senal-0.367
bt-0.313
ould-0.312
Token
Token ID247
Feature activation+2.66e-01

Pos logit contributions
 tem+0.422
 persona+0.417
 mig+0.414
 negatives+0.400
][/+0.390
Neg logit contributions
 Commando-0.292
earch-0.279
senal-0.270
bt-0.215
ould-0.215
Tokenve
Token ID303
Feature activation+8.29e-01

Pos logit contributions
 tem+0.195
 persona+0.192
 mig+0.190
 negatives+0.181
][/+0.176
Neg logit contributions
 Commando-0.217
earch-0.210
senal-0.205
bt-0.172
ould-0.171
Token been
Token ID587
Feature activation+4.35e-01

Pos logit contributions
 tem+0.489
 persona+0.480
 mig+0.474
 negatives+0.447
][/+0.431
Neg logit contributions
 Commando-0.795
earch-0.771
senal-0.757
bt-0.655
ould-0.653
Token told
Token ID1297
Feature activation-2.04e-01

Pos logit contributions
 tem+0.285
 persona+0.281
 mig+0.277
 negatives+0.263
][/+0.254
Neg logit contributions
 Commando-0.390
earch-0.378
senal-0.371
bt-0.316
ould-0.316
Token to
Token ID284
Feature activation-1.45e-01

Pos logit contributions
 Commando+0.177
earch+0.171
senal+0.167
bt+0.142
ould+0.142
Neg logit contributions
 tem-0.139
 persona-0.137
 mig-0.135
 negatives-0.129
][/-0.125
Token make
Token ID787
Feature activation-1.06e-02

Pos logit contributions
 Commando+0.135
earch+0.130
senal+0.126
bt+0.110
ould+0.109
Neg logit contributions
 tem-0.090
 persona-0.089
 mig-0.088
 negatives-0.084
][/-0.081
Token life
Token ID1204
Feature activation+4.32e-01

Pos logit contributions
 Commando+0.010
earch+0.009
senal+0.009
bt+0.008
ould+0.008
Neg logit contributions
 tem-0.007
 persona-0.007
 mig-0.007
 negatives-0.006
][/-0.006
Token argued
Token ID7189
Feature activation+3.29e-01

Pos logit contributions
 tem+0.046
 persona+0.045
 mig+0.045
 negatives+0.042
][/+0.041
Neg logit contributions
 Commando-0.062
earch-0.060
senal-0.058
bt-0.050
ould-0.050
Token that
Token ID326
Feature activation-2.02e-01

Pos logit contributions
 tem+0.236
 persona+0.233
 mig+0.231
 negatives+0.220
][/+0.214
Neg logit contributions
 Commando-0.273
earch-0.263
senal-0.257
bt-0.217
ould-0.217
Token the
Token ID262
Feature activation-2.96e-02

Pos logit contributions
 Commando+0.180
earch+0.174
senal+0.169
bt+0.146
ould+0.145
Neg logit contributions
 tem-0.132
 persona-0.130
 mig-0.129
 negatives-0.122
][/-0.119
Token Liberals
Token ID20162
Feature activation+9.81e-01

Pos logit contributions
 Commando+0.028
earch+0.027
senal+0.026
bt+0.023
ould+0.023
Neg logit contributions
 tem-0.018
 persona-0.017
 mig-0.017
 negatives-0.016
][/-0.016
Token are
Token ID389
Feature activation+1.46e-01

Pos logit contributions
 tem+0.691
 persona+0.683
 mig+0.672
 negatives+0.641
][/+0.621
Neg logit contributions
 Commando-0.834
earch-0.804
senal-0.798
bt-0.666
ould-0.662
Token going
Token ID1016
Feature activation+3.03e-01

Pos logit contributions
 tem+0.096
 persona+0.095
 mig+0.094
 negatives+0.089
][/+0.087
Neg logit contributions
 Commando-0.129
earch-0.125
senal-0.122
bt-0.105
ould-0.104
Token after
Token ID706
Feature activation+2.15e-02

Pos logit contributions
 tem+0.186
 persona+0.183
 mig+0.181
 negatives+0.171
][/+0.165
Neg logit contributions
 Commando-0.283
earch-0.274
senal-0.269
bt-0.231
ould-0.231
Token wealthy
Token ID11574
Feature activation-1.09e-01

Pos logit contributions
 tem+0.014
 persona+0.014
 mig+0.014
 negatives+0.013
][/+0.013
Neg logit contributions
 Commando-0.019
earch-0.018
senal-0.018
bt-0.015
ould-0.015
Token people
Token ID661
Feature activation+7.87e-02

Pos logit contributions
 Commando+0.098
earch+0.094
senal+0.090
bt+0.079
ould+0.079
Neg logit contributions
 tem-0.071
 persona-0.069
 mig-0.069
 negatives-0.065
][/-0.064
Token who
Token ID508
Feature activation+1.15e-01

Pos logit contributions
 tem+0.056
 persona+0.055
 mig+0.055
 negatives+0.052
][/+0.050
Neg logit contributions
 Commando-0.066
earch-0.064
senal-0.062
bt-0.053
ould-0.053
Token try
Token ID1949
Feature activation-7.58e-02

Pos logit contributions
 tem+0.076
 persona+0.075
 mig+0.074
 negatives+0.071
][/+0.068
Neg logit contributions
 Commando-0.101
earch-0.098
senal-0.095
bt-0.082
ould-0.082
Token asylum
Token ID13142
Feature activation-6.08e-01

Pos logit contributions
 Commando+0.393
earch+0.380
senal+0.373
bt+0.323
ould+0.323
Neg logit contributions
 tem-0.241
 persona-0.237
 mig-0.234
 negatives-0.221
][/-0.212
Token and
Token ID290
Feature activation-1.78e-01

Pos logit contributions
 Commando+0.508
earch+0.490
senal+0.476
bt+0.405
ould+0.403
Neg logit contributions
 tem-0.431
 persona-0.425
 mig-0.423
 negatives-0.404
][/-0.390
Token design
Token ID1486
Feature activation-4.39e-01

Pos logit contributions
 Commando+0.172
earch+0.166
senal+0.163
bt+0.142
ould+0.142
Neg logit contributions
 tem-0.102
 persona-0.100
 mig-0.099
 negatives-0.094
][/-0.090
Token it
Token ID340
Feature activation-1.02e-01

Pos logit contributions
 Commando+0.402
earch+0.387
senal+0.379
bt+0.327
ould+0.327
Neg logit contributions
 tem-0.276
 persona-0.272
 mig-0.270
 negatives-0.255
][/-0.248
Token to
Token ID284
Feature activation-1.08e-01

Pos logit contributions
 Commando+0.090
earch+0.086
senal+0.084
bt+0.072
ould+0.072
Neg logit contributions
 tem-0.067
 persona-0.066
 mig-0.066
 negatives-0.063
][/-0.061
Token be
Token ID307
Feature activation-1.39e-01

Pos logit contributions
 Commando+0.100
earch+0.096
senal+0.093
bt+0.081
ould+0.081
Neg logit contributions
 tem-0.068
 persona-0.067
 mig-0.066
 negatives-0.063
][/-0.061
Token filled
Token ID5901
Feature activation-2.66e-01

Pos logit contributions
 Commando+0.128
earch+0.124
senal+0.122
bt+0.105
ould+0.105
Neg logit contributions
 tem-0.087
 persona-0.085
 mig-0.085
 negatives-0.080
][/-0.077
Token of
Token ID286
Feature activation+4.22e-02

Pos logit contributions
 Commando+0.199
earch+0.191
senal+0.186
bt+0.154
ould+0.154
Neg logit contributions
 tem-0.212
 persona-0.210
 mig-0.208
 negatives-0.200
][/-0.195
Token traps
Token ID20348
Feature activation-2.77e-01

Pos logit contributions
 tem+0.028
 persona+0.027
 mig+0.027
 negatives+0.026
][/+0.025
Neg logit contributions
 Commando-0.038
earch-0.036
senal-0.036
bt-0.030
ould-0.030
Token designed
Token ID3562
Feature activation-5.16e-01

Pos logit contributions
 Commando+0.232
earch+0.224
senal+0.218
bt+0.185
ould+0.185
Neg logit contributions
 tem-0.197
 persona-0.194
 mig-0.193
 negatives-0.184
][/-0.178
Token by
Token ID416
Feature activation+2.06e-01

Pos logit contributions
 Commando+0.447
earch+0.431
senal+0.420
bt+0.359
ould+0.358
Neg logit contributions
 tem-0.351
 persona-0.346
 mig-0.342
 negatives-0.326
][/-0.316
Token and
Token ID290
Feature activation+2.60e-02

Pos logit contributions
 tem+0.229
 persona+0.225
 mig+0.223
 negatives+0.213
][/+0.207
Neg logit contributions
 Commando-0.265
earch-0.256
senal-0.250
bt-0.211
ould-0.210
Token crime
Token ID4065
Feature activation-3.25e-01

Pos logit contributions
 tem+0.016
 persona+0.016
 mig+0.016
 negatives+0.015
][/+0.015
Neg logit contributions
 Commando-0.024
earch-0.023
senal-0.023
bt-0.020
ould-0.019
Token.
Token ID13
Feature activation+8.56e-02

Pos logit contributions
 Commando+0.274
earch+0.264
senal+0.258
bt+0.220
ould+0.218
Neg logit contributions
 tem-0.228
 persona-0.226
 mig-0.223
 negatives-0.213
][/-0.208
Token
Token ID198
Feature activation-6.11e-02

Pos logit contributions
 tem+0.064
 persona+0.063
 mig+0.062
 negatives+0.059
][/+0.058
Neg logit contributions
 Commando-0.069
earch-0.066
senal-0.065
bt-0.054
ould-0.054
Token
Token ID198
Feature activation+7.27e-02

Pos logit contributions
 Commando+0.046
earch+0.044
senal+0.040
bt+0.036
ould+0.036
Neg logit contributions
 tem-0.048
 persona-0.048
 mig-0.048
 negatives-0.046
 ethnicity-0.044
Token
Token ID447
Feature activation-1.19e-01

Pos logit contributions
 tem+0.047
 persona+0.047
 mig+0.046
 negatives+0.044
 ethnicity+0.042
Neg logit contributions
 Commando-0.065
earch-0.063
senal-0.061
bt-0.054
ould-0.053
Token
Token ID250
Feature activation-2.27e-02

Pos logit contributions
 Commando+0.066
earch+0.062
senal+0.059
bt+0.046
ould+0.046
Neg logit contributions
 tem-0.118
 persona-0.116
 mig-0.115
 negatives-0.112
][/-0.109
TokenI
Token ID40
Feature activation+1.69e-01

Pos logit contributions
 Commando+0.020
earch+0.019
senal+0.018
bt+0.016
ould+0.016
Neg logit contributions
 tem-0.015
 persona-0.015
 mig-0.015
 negatives-0.014
][/-0.014
Token don
Token ID836
Feature activation+4.34e-01

Pos logit contributions
 tem+0.110
 persona+0.109
 mig+0.107
 negatives+0.102
][/+0.099
Neg logit contributions
 Commando-0.151
earch-0.146
senal-0.143
bt-0.123
ould-0.122
Token
Token ID447
Feature activation+2.24e-01

Pos logit contributions
 tem+0.257
 persona+0.256
 mig+0.250
 negatives+0.239
 ethnicity+0.227
Neg logit contributions
 Commando-0.413
earch-0.399
senal-0.378
bt-0.345
ould-0.341
Token
Token ID247
Feature activation+1.80e-02

Pos logit contributions
 tem+0.216
 persona+0.213
 mig+0.211
 negatives+0.205
][/+0.200
Neg logit contributions
 Commando-0.130
earch-0.124
senal-0.117
bt-0.095
ould-0.094
Token to
Token ID284
Feature activation-2.94e-01

Pos logit contributions
 Commando+0.222
earch+0.215
senal+0.210
bt+0.179
ould+0.178
Neg logit contributions
 tem-0.175
 persona-0.172
 mig-0.170
 negatives-0.162
][/-0.157
Token do
Token ID466
Feature activation+2.65e-03

Pos logit contributions
 Commando+0.263
earch+0.251
senal+0.241
bt+0.213
ould+0.212
Neg logit contributions
 tem-0.192
 persona-0.190
 mig-0.187
 negatives-0.179
][/-0.173
Token that
Token ID326
Feature activation-9.63e-02

Pos logit contributions
 tem+0.002
 persona+0.002
 mig+0.002
 negatives+0.002
][/+0.002
Neg logit contributions
 Commando-0.002
earch-0.002
senal-0.002
bt-0.002
ould-0.002
Token without
Token ID1231
Feature activation-4.31e-01

Pos logit contributions
 Commando+0.082
earch+0.079
senal+0.077
bt+0.066
ould+0.066
Neg logit contributions
 tem-0.066
 persona-0.066
 mig-0.065
 negatives-0.062
][/-0.060
Token jeopard
Token ID22451
Feature activation-1.44e-01

Pos logit contributions
 Commando+0.364
earch+0.349
senal+0.339
bt+0.290
ould+0.289
Neg logit contributions
 tem-0.301
 persona-0.298
 mig-0.295
 negatives-0.281
][/-0.273
Tokenizing
Token ID2890
Feature activation-4.60e-01

Pos logit contributions
 Commando+0.117
earch+0.113
senal+0.110
bt+0.093
ould+0.093
Neg logit contributions
 tem-0.105
 persona-0.104
 mig-0.102
 negatives-0.098
][/-0.095
Token the
Token ID262
Feature activation-1.63e-01

Pos logit contributions
 Commando+0.410
earch+0.393
senal+0.382
bt+0.331
ould+0.329
Neg logit contributions
 tem-0.300
 persona-0.296
 mig-0.294
 negatives-0.279
][/-0.271
Token launch
Token ID4219
Feature activation-6.67e-01

Pos logit contributions
 Commando+0.155
earch+0.148
senal+0.145
bt+0.127
ould+0.126
Neg logit contributions
 tem-0.098
 persona-0.096
 mig-0.096
 negatives-0.090
][/-0.088
Token.
Token ID13
Feature activation-8.35e-02

Pos logit contributions
 Commando+0.584
earch+0.560
senal+0.543
bt+0.474
ould+0.465
Neg logit contributions
 tem-0.444
 persona-0.434
 mig-0.434
 negatives-0.413
][/-0.401
Token But
Token ID887
Feature activation+5.52e-02

Pos logit contributions
 Commando+0.072
earch+0.069
senal+0.068
bt+0.058
ould+0.058
Neg logit contributions
 tem-0.057
 persona-0.057
 mig-0.056
 negatives-0.053
][/-0.051
Token you
Token ID345
Feature activation-2.29e-01

Pos logit contributions
 tem+0.036
 persona+0.035
 mig+0.035
 negatives+0.033
][/+0.032
Neg logit contributions
 Commando-0.050
earch-0.048
senal-0.047
bt-0.040
ould-0.040
Token 2013
Token ID2211
Feature activation+1.70e-02

Pos logit contributions
 Commando+0.237
earch+0.228
senal+0.223
bt+0.188
ould+0.188
Neg logit contributions
 tem-0.208
 persona-0.205
 mig-0.203
 negatives-0.194
][/-0.188
Token,
Token ID11
Feature activation+1.64e-01

Pos logit contributions
 tem+0.012
 persona+0.012
 mig+0.012
 negatives+0.011
][/+0.011
Neg logit contributions
 Commando-0.014
earch-0.014
senal-0.013
bt-0.011
ould-0.011
Token K
Token ID509
Feature activation-1.98e-01

Pos logit contributions
 tem+0.109
 persona+0.108
 mig+0.106
 negatives+0.101
][/+0.098
Neg logit contributions
 Commando-0.145
earch-0.140
senal-0.137
bt-0.117
ould-0.116
Tokenarel
Token ID20318
Feature activation-6.74e-03

Pos logit contributions
 Commando+0.157
earch+0.153
senal+0.148
bt+0.125
ould+0.125
Neg logit contributions
 tem-0.149
 persona-0.147
 mig-0.145
 negatives-0.140
][/-0.134
Token de
Token ID390
Feature activation-4.15e-01

Pos logit contributions
 Commando+0.006
earch+0.005
senal+0.005
bt+0.004
ould+0.004
Neg logit contributions
 tem-0.005
 persona-0.005
 mig-0.005
 negatives-0.005
][/-0.004
Token G
Token ID402
Feature activation-2.53e-01

Pos logit contributions
 Commando+0.330
earch+0.307
senal+0.298
bt+0.254
ould+0.251
Neg logit contributions
 tem-0.317
 persona-0.315
 mig-0.307
 negatives-0.300
][/-0.291
Tokenuch
Token ID794
Feature activation-1.69e-01

Pos logit contributions
 Commando+0.196
earch+0.192
senal+0.182
ould+0.161
bt+0.158
Neg logit contributions
 tem-0.194
 persona-0.192
 mig-0.192
 negatives-0.182
 ethnicity-0.175
Tokent
Token ID83
Feature activation+6.34e-02

Pos logit contributions
 Commando+0.119
earch+0.114
senal+0.111
bt+0.091
ould+0.091
Neg logit contributions
 tem-0.143
 persona-0.141
 mig-0.139
 negatives-0.134
][/-0.131
Token dismissed
Token ID11126
Feature activation+4.83e-01

Pos logit contributions
 tem+0.044
 persona+0.043
 mig+0.043
 negatives+0.041
][/+0.039
Neg logit contributions
 Commando-0.054
earch-0.052
senal-0.051
bt-0.044
ould-0.043
Token as
Token ID355
Feature activation+8.14e-02

Pos logit contributions
 tem+0.303
 persona+0.299
 mig+0.295
 negatives+0.279
][/+0.270
Neg logit contributions
 Commando-0.445
earch-0.431
senal-0.423
bt-0.364
ould-0.362
Token
Token ID564
Feature activation+4.59e-04

Pos logit contributions
 tem+0.054
 persona+0.053
 mig+0.052
 negatives+0.050
][/+0.048
Neg logit contributions
 Commando-0.072
earch-0.070
senal-0.068
bt-0.059
ould-0.058
Token system
Token ID1080
Feature activation-1.62e-01

Pos logit contributions
 tem+0.098
 persona+0.096
 mig+0.095
 negatives+0.090
][/+0.087
Neg logit contributions
 Commando-0.162
earch-0.157
senal-0.154
bt-0.133
ould-0.133
Token,
Token ID11
Feature activation-2.05e-01

Pos logit contributions
 Commando+0.139
earch+0.135
senal+0.132
bt+0.112
ould+0.112
Neg logit contributions
 tem-0.111
 persona-0.109
 mig-0.108
 negatives-0.103
][/-0.100
Token who
Token ID508
Feature activation+2.21e-02

Pos logit contributions
 Commando+0.166
earch+0.160
senal+0.156
bt+0.132
ould+0.131
Neg logit contributions
 tem-0.152
 persona-0.150
 mig-0.148
 negatives-0.142
][/-0.138
Token confirmed
Token ID4999
Feature activation-4.67e-01

Pos logit contributions
 tem+0.014
 persona+0.013
 mig+0.013
 negatives+0.013
][/+0.012
Neg logit contributions
 Commando-0.020
earch-0.020
senal-0.019
bt-0.017
ould-0.017
Token that
Token ID326
Feature activation-1.97e-01

Pos logit contributions
 Commando+0.379
earch+0.365
senal+0.355
bt+0.300
ould+0.298
Neg logit contributions
 tem-0.344
 persona-0.340
 mig-0.336
 negatives-0.321
][/-0.313
Token our
Token ID674
Feature activation-3.84e-02

Pos logit contributions
 Commando+0.175
earch+0.169
senal+0.164
bt+0.141
ould+0.140
Neg logit contributions
 tem-0.130
 persona-0.129
 mig-0.127
 negatives-0.121
][/-0.118
Token hypothesis
Token ID14078
Feature activation+9.72e-02

Pos logit contributions
 Commando+0.040
earch+0.038
senal+0.037
bt+0.033
ould+0.033
Neg logit contributions
 tem-0.019
 persona-0.019
 mig-0.019
 negatives-0.018
][/-0.017
Token was
Token ID373
Feature activation-4.60e-02

Pos logit contributions
 tem+0.068
 persona+0.067
 mig+0.066
 negatives+0.063
][/+0.061
Neg logit contributions
 Commando-0.082
earch-0.080
senal-0.077
bt-0.066
ould-0.066
Token correct
Token ID3376
Feature activation-6.34e-02

Pos logit contributions
 Commando+0.041
earch+0.040
senal+0.039
bt+0.034
ould+0.034
Neg logit contributions
 tem-0.030
 persona-0.029
 mig-0.029
 negatives-0.027
][/-0.027
Token and
Token ID290
Feature activation-2.17e-02

Pos logit contributions
 Commando+0.055
earch+0.053
senal+0.052
bt+0.044
ould+0.044
Neg logit contributions
 tem-0.043
 persona-0.042
 mig-0.042
 negatives-0.040
][/-0.039
Token that
Token ID326
Feature activation+5.40e-02

Pos logit contributions
 Commando+0.020
earch+0.019
senal+0.018
bt+0.016
ould+0.016
Neg logit contributions
 tem-0.014
 persona-0.014
 mig-0.013
 negatives-0.013
][/-0.012
Token today
Token ID1909
Feature activation+7.20e-02

Pos logit contributions
 tem+0.622
 persona+0.612
 mig+0.606
 negatives+0.577
][/+0.558
Neg logit contributions
 Commando-0.789
earch-0.763
senal-0.747
bt-0.635
ould-0.633
Token.
Token ID13
Feature activation-6.56e-02

Pos logit contributions
 tem+0.051
 persona+0.050
 mig+0.050
 negatives+0.047
][/+0.046
Neg logit contributions
 Commando-0.061
earch-0.059
senal-0.057
bt-0.049
ould-0.048
Token This
Token ID770
Feature activation-9.54e-04

Pos logit contributions
 Commando+0.055
earch+0.054
senal+0.053
ould+0.044
bt+0.044
Neg logit contributions
 tem-0.046
 persona-0.046
 mig-0.045
 negatives-0.043
][/-0.042
Token is
Token ID318
Feature activation+8.28e-02

Pos logit contributions
 Commando+0.001
earch+0.001
senal+0.001
bt+0.001
ould+0.001
Neg logit contributions
 tem-0.001
 persona-0.001
 mig-0.001
 negatives-0.001
][/-0.001
Token partly
Token ID11476
Feature activation-4.72e-01

Pos logit contributions
 tem+0.050
 persona+0.049
 mig+0.048
 negatives+0.046
][/+0.044
Neg logit contributions
 Commando-0.079
earch-0.076
senal-0.074
bt-0.064
ould-0.064
Token due
Token ID2233
Feature activation-2.38e+00

Pos logit contributions
 Commando+0.409
earch+0.394
senal+0.383
bt+0.329
ould+0.326
Neg logit contributions
 tem-0.321
 persona-0.316
 mig-0.313
 negatives-0.297
][/-0.290
Token to
Token ID284
Feature activation-6.72e-03

Pos logit contributions
 Commando+1.956
earch+1.874
senal+1.853
ould+1.552
bt+1.550
Neg logit contributions
 tem-1.702
 persona-1.687
 mig-1.672
 negatives-1.591
][/-1.558
Token the
Token ID262
Feature activation-6.74e-03

Pos logit contributions
 Commando+0.006
earch+0.006
senal+0.006
bt+0.005
ould+0.005
Neg logit contributions
 tem-0.004
 persona-0.004
 mig-0.004
 negatives-0.004
][/-0.004
Token excellent
Token ID6275
Feature activation+7.52e-01

Pos logit contributions
 Commando+0.006
earch+0.006
senal+0.006
bt+0.005
ould+0.005
Neg logit contributions
 tem-0.004
 persona-0.004
 mig-0.004
 negatives-0.004
][/-0.004
Token conversion
Token ID11315
Feature activation+5.66e-01

Pos logit contributions
 tem+0.443
 persona+0.436
 mig+0.430
 negatives+0.406
][/+0.390
Neg logit contributions
 Commando-0.722
earch-0.701
senal-0.689
bt-0.597
ould-0.594
Token,
Token ID11
Feature activation+5.56e-02

Pos logit contributions
 tem+0.339
 persona+0.335
 mig+0.330
 negatives+0.311
][/+0.304
Neg logit contributions
 Commando-0.537
earch-0.518
senal-0.506
bt-0.438
ould-0.438
Token[
Token ID58
Feature activation-1.32e+00

Pos logit contributions
 tem+0.354
 persona+0.348
 mig+0.345
 negatives+0.329
][/+0.320
Neg logit contributions
 Commando-0.403
earch-0.389
senal-0.382
bt-0.320
ould-0.319
Token0
Token ID15
Feature activation-1.35e-01

Pos logit contributions
 Commando+0.882
earch+0.843
senal+0.814
bt+0.665
ould+0.656
Neg logit contributions
 tem-1.157
 persona-1.147
 mig-1.138
 negatives-1.093
][/-1.060
Token],
Token ID4357
Feature activation+2.87e-01

Pos logit contributions
 Commando+0.134
earch+0.130
senal+0.127
bt+0.111
ould+0.111
Neg logit contributions
 tem-0.075
 persona-0.074
 mig-0.073
 negatives-0.069
][/-0.065
Token$
Token ID3
Feature activation-4.65e-01

Pos logit contributions
 tem+0.220
 persona+0.217
 mig+0.215
 negatives+0.205
][/+0.200
Neg logit contributions
 Commando-0.225
earch-0.216
senal-0.212
bt-0.176
ould-0.176
TokenBG
Token ID40469
Feature activation+4.70e-01

Pos logit contributions
 Commando+0.358
earch+0.345
senal+0.336
bt+0.282
ould+0.279
Neg logit contributions
 tem-0.362
 persona-0.358
 mig-0.355
 negatives-0.339
][/-0.329
Token[
Token ID58
Feature activation-1.45e+00

Pos logit contributions
 tem+0.326
 persona+0.319
 mig+0.316
 negatives+0.302
][/+0.295
Neg logit contributions
 Commando-0.403
earch-0.388
senal-0.383
bt-0.321
ould-0.320
Token1
Token ID16
Feature activation+1.57e-01

Pos logit contributions
 Commando+1.056
earch+1.019
senal+0.978
bt+0.825
ould+0.815
Neg logit contributions
 tem-1.190
 persona-1.179
 mig-1.166
 negatives-1.113
 ethnicity-1.080
Token],
Token ID4357
Feature activation+1.84e-01

Pos logit contributions
 tem+0.079
 persona+0.078
 mig+0.077
 negatives+0.072
][/+0.068
Neg logit contributions
 Commando-0.164
earch-0.159
senal-0.157
bt-0.137
ould-0.137
Token$
Token ID3
Feature activation-5.94e-01

Pos logit contributions
 tem+0.136
 persona+0.134
 mig+0.133
 negatives+0.127
][/+0.123
Neg logit contributions
 Commando-0.148
earch-0.143
senal-0.140
bt-0.117
ould-0.117
TokenBG
Token ID40469
Feature activation+4.62e-01

Pos logit contributions
 Commando+0.424
earch+0.408
senal+0.396
bt+0.328
ould+0.324
Neg logit contributions
 tem-0.496
 persona-0.489
 mig-0.485
 negatives-0.466
][/-0.452
Token[
Token ID58
Feature activation-1.51e+00

Pos logit contributions
 tem+0.324
 persona+0.317
 mig+0.314
 negatives+0.299
][/+0.294
Neg logit contributions
 Commando-0.392
earch-0.378
senal-0.377
bt-0.312
ould-0.312
Token long
Token ID890
Feature activation+5.98e-02

Pos logit contributions
 Commando+0.001
earch+0.001
senal+0.001
bt+0.001
ould+0.001
Neg logit contributions
 tem-0.001
 persona-0.001
 mig-0.001
 negatives-0.001
][/-0.001
Token time
Token ID640
Feature activation+9.80e-02

Pos logit contributions
 tem+0.027
 persona+0.026
 mig+0.026
 negatives+0.024
][/+0.023
Neg logit contributions
 Commando-0.066
earch-0.064
senal-0.063
bt-0.056
ould-0.056
Token and
Token ID290
Feature activation-6.18e-02

Pos logit contributions
 tem+0.068
 persona+0.067
 mig+0.066
 negatives+0.063
][/+0.061
Neg logit contributions
 Commando-0.084
earch-0.081
senal-0.080
bt-0.068
ould-0.067
Token this
Token ID428
Feature activation+3.00e-01

Pos logit contributions
 Commando+0.053
earch+0.051
senal+0.050
bt+0.042
ould+0.042
Neg logit contributions
 tem-0.043
 persona-0.042
 mig-0.042
 negatives-0.040
][/-0.039
Token sub
Token ID850
Feature activation-2.54e-02

Pos logit contributions
 tem+0.195
 persona+0.192
 mig+0.190
 negatives+0.180
][/+0.174
Neg logit contributions
 Commando-0.269
earch-0.260
senal-0.255
bt-0.218
ould-0.218
Tokenwo
Token ID21638
Feature activation-2.07e+00

Pos logit contributions
 Commando+0.022
earch+0.021
senal+0.021
bt+0.018
ould+0.017
Neg logit contributions
 tem-0.018
 persona-0.017
 mig-0.017
 negatives-0.016
][/-0.016
Tokenofer
Token ID30288
Feature activation-1.09e+00

Pos logit contributions
 Commando+0.893
earch+0.864
senal+0.748
ould+0.578
bt+0.567
Neg logit contributions
 tem-2.273
 persona-2.267
 mig-2.258
 negatives-2.201
 ethnicity-2.148
Token is
Token ID318
Feature activation-5.70e-02

Pos logit contributions
 Commando+0.960
earch+0.922
senal+0.895
bt+0.772
ould+0.766
Neg logit contributions
 tem-0.723
 persona-0.717
 mig-0.708
 negatives-0.675
][/-0.656
Token worth
Token ID2861
Feature activation-1.47e-01

Pos logit contributions
 Commando+0.053
earch+0.051
senal+0.050
bt+0.043
ould+0.043
Neg logit contributions
 tem-0.036
 persona-0.035
 mig-0.035
 negatives-0.033
][/-0.032
Token every
Token ID790
Feature activation+1.76e-01

Pos logit contributions
 Commando+0.125
earch+0.121
senal+0.118
bt+0.100
ould+0.100
Neg logit contributions
 tem-0.103
 persona-0.102
 mig-0.101
 negatives-0.096
][/-0.093
Token penny
Token ID26369
Feature activation-6.42e-01

Pos logit contributions
 tem+0.155
 persona+0.153
 mig+0.152
 negatives+0.146
][/+0.142
Neg logit contributions
 Commando-0.119
earch-0.114
senal-0.111
bt-0.089
ould-0.088
Token John
Token ID1757
Feature activation+5.15e-01

Pos logit contributions
 Commando+0.028
earch+0.027
senal+0.026
bt+0.022
ould+0.022
Neg logit contributions
 tem-0.024
 persona-0.024
 mig-0.024
 negatives-0.022
][/-0.022
Token McCain
Token ID14264
Feature activation-1.06e+00

Pos logit contributions
 tem+0.424
 persona+0.420
 mig+0.415
 negatives+0.399
][/+0.389
Neg logit contributions
 Commando-0.375
earch-0.358
senal-0.346
bt-0.287
ould-0.286
Token.
Token ID13
Feature activation-1.31e-02

Pos logit contributions
 Commando+0.888
earch+0.855
senal+0.833
bt+0.708
ould+0.708
Neg logit contributions
 tem-0.753
 persona-0.742
 mig-0.737
 negatives-0.703
][/-0.686
Token
Token ID198
Feature activation+2.40e-01

Pos logit contributions
 Commando+0.011
earch+0.011
senal+0.010
bt+0.009
ould+0.009
Neg logit contributions
 tem-0.009
 persona-0.009
 mig-0.009
 negatives-0.009
][/-0.008
Token
Token ID198
Feature activation+3.14e-01

Pos logit contributions
 tem+0.189
 persona+0.188
 mig+0.185
 negatives+0.178
 ethnicity+0.172
Neg logit contributions
 Commando-0.181
earch-0.174
senal-0.162
bt-0.143
ould-0.141
TokenNewsletter
Token ID33031
Feature activation-1.61e+00

Pos logit contributions
 tem+0.230
 persona+0.226
 mig+0.224
 negatives+0.215
][/+0.207
Neg logit contributions
 Commando-0.257
earch-0.248
senal-0.239
bt-0.206
ould-0.204
Token Sign
Token ID5865
Feature activation+3.46e-01

Pos logit contributions
 Commando+1.484
earch+1.431
senal+1.338
bt+1.206
ould+1.195
Neg logit contributions
 tem-0.986
 persona-0.974
 mig-0.966
 negatives-0.924
][/-0.886
Token Up
Token ID3205
Feature activation+6.93e-01

Pos logit contributions
 tem+0.276
 persona+0.273
 mig+0.270
 negatives+0.259
][/+0.252
Neg logit contributions
 Commando-0.260
earch-0.249
senal-0.242
bt-0.201
ould-0.201
Token Continue
Token ID10054
Feature activation+8.63e-01

Pos logit contributions
 tem+0.477
 persona+0.470
 mig+0.465
 negatives+0.444
][/+0.429
Neg logit contributions
 Commando-0.597
earch-0.576
senal-0.561
bt-0.479
ould-0.477
Token reading
Token ID3555
Feature activation+2.02e-01

Pos logit contributions
 tem+0.589
 persona+0.580
 mig+0.574
 negatives+0.546
][/+0.529
Neg logit contributions
 Commando-0.747
earch-0.722
senal-0.709
bt-0.602
ould-0.600
Token the
Token ID262
Feature activation+6.01e-02

Pos logit contributions
 tem+0.129
 persona+0.127
 mig+0.125
 negatives+0.119
][/+0.115
Neg logit contributions
 Commando-0.184
earch-0.178
senal-0.175
bt-0.150
ould-0.149
Token.
Token ID13
Feature activation-1.39e-01

Pos logit contributions
 Commando+0.303
earch+0.293
senal+0.287
bt+0.244
ould+0.244
Neg logit contributions
 tem-0.229
 persona-0.226
 mig-0.223
 negatives-0.212
][/-0.205
Token<|endoftext|>
Token ID50256
Feature activation+1.12e-01

Pos logit contributions
 Commando+0.113
earch+0.109
senal+0.106
bt+0.089
ould+0.089
Neg logit contributions
 tem-0.102
 persona-0.101
 mig-0.100
 negatives-0.096
][/-0.093
TokenFrom
Token ID4863
Feature activation-8.23e-01

Pos logit contributions
 tem+0.081
 persona+0.080
 mig+0.079
 negatives+0.076
 ethnicity+0.073
Neg logit contributions
 Commando-0.093
earch-0.090
senal-0.087
bt-0.075
ould-0.074
Token John
Token ID1757
Feature activation-2.63e-01

Pos logit contributions
 Commando+0.696
earch+0.664
senal+0.645
bt+0.549
ould+0.547
Neg logit contributions
 tem-0.579
 persona-0.574
 mig-0.567
 negatives-0.543
][/-0.528
Token Brian
Token ID8403
Feature activation-1.22e-01

Pos logit contributions
 Commando+0.211
earch+0.203
senal+0.197
bt+0.167
ould+0.166
Neg logit contributions
 tem-0.196
 persona-0.195
 mig-0.192
 negatives-0.184
][/-0.178
Token King
Token ID2677
Feature activation-1.74e+00

Pos logit contributions
 Commando+0.100
earch+0.096
senal+0.094
bt+0.079
ould+0.079
Neg logit contributions
 tem-0.089
 persona-0.088
 mig-0.087
 negatives-0.083
][/-0.081
Token
Token ID447
Feature activation-4.11e-01

Pos logit contributions
 Commando+1.461
earch+1.409
senal+1.370
bt+1.165
ould+1.161
Neg logit contributions
 tem-1.223
 persona-1.208
 mig-1.197
 negatives-1.143
][/-1.111
Token
Token ID247
Feature activation-2.02e-01

Pos logit contributions
 Commando+0.347
earch+0.335
senal+0.328
bt+0.278
ould+0.277
Neg logit contributions
 tem-0.289
 persona-0.285
 mig-0.282
 negatives-0.269
][/-0.260
Tokens
Token ID82
Feature activation-5.36e-02

Pos logit contributions
 Commando+0.172
earch+0.166
senal+0.162
bt+0.138
ould+0.137
Neg logit contributions
 tem-0.142
 persona-0.139
 mig-0.138
 negatives-0.132
][/-0.127
Token book
Token ID1492
Feature activation-5.71e-01

Pos logit contributions
 Commando+0.047
earch+0.045
senal+0.044
bt+0.038
ould+0.038
Neg logit contributions
 tem-0.036
 persona-0.036
 mig-0.035
 negatives-0.033
][/-0.032
Token,
Token ID11
Feature activation-4.18e-01

Pos logit contributions
 Commando+0.484
earch+0.466
senal+0.455
bt+0.387
ould+0.385
Neg logit contributions
 tem-0.401
 persona-0.396
 mig-0.392
 negatives-0.373
][/-0.362