TOP ACTIVATIONS
MAX = 0.877

self . half _ m oves , depth : self .
in a skillet over medium - high heat . Place f
ussian , then apply the black hat # morph ological operator
= c v 2 . Ga ussian Bl ur ( gray
Beaut ifully Paid Hom age To Lost Leader O der us
= im utils . res ize ( image , height =
film E scape from Al cat raz
3 x 3 Ga ussian , then apply the black hat
- 31 . [ 24 ] Nah man ides
- to - 5 Working Women Education Fund : ... $
6 } < -> { 1 , 6 }, { 1
who say that making middle and high school ers start classes
the commitment to advancing civil and human rights these are
AIR s Health and Social Development program develops knowledge
v 2 . Ga ussian Bl ur ( gray , (
n pm install - g server less You
image , height = 600 ) gray = c v 2
parse . Arg ument Parser () ap . add _ argument
. J ag d ish Sh ind e Senior
, one might ask , why go that one small

MIN ACTIVATIONS
MIN = -0.973

43 ( S 4 ) DOI : 10 . 10 86
and policy Americans for Prosper ity : Dist orting
Facebook St umble Upon What are these
' 2016 primary Sanders : ' Damn right ' I 'll
Trump House committee believes it has evidence Trump requested putting ally
d [ CO 2 ] / dt = f
2011 6 2 2012 7 0 2013 3 0 2014 8
the argument parse and parse the arguments ap = arg parse
Trump House committee believes it has evidence Trump requested putting ally
s . t . I TA 11 Matte o Trent in
two quantum states during the pseud og ap , one of
Facebook St umble Upon What are these
Rodham Clinton Sanders : ' I fully expect ' fair treatment
[ index State [ Keys [ step New ][ [ i
14 Recommend Recommend Share this comment on
required = True , help =" path to images directory ")
' 2016 primary Sanders : ' Damn right ' I 'll
2009 10 2 2010 12 2 2011 6 2 2012 7
former President Barack Obama Barack Hussein Obama With low birth rate
. The player must know choose ) Permanent Voices in Your

INTERVAL 0.414 - 0.877
CONTAINS 0.821%

trials identifying and enroll ing patients with C IR CI ).
enrollment as part of the 2011 - 13 state budget ,
" P ulse period " , Float ) = 1 //
itself will be subject to sh aria in its dealings on
had lost when he murdered Yoshi . S aki refuses and

INTERVAL -0.048 - 0.414
CONTAINS 61.093%

: how to teach yourself to code Apple
Facebook Twitter Pinterest Liverpool s Adam L all ana
sound tracks for seasons 9 through 27 ( with the exception
do with the requests of other autonomous communities .
said . We were picking some of our grapes

INTERVAL -0.510 - -0.048
CONTAINS 37.884%

, it reads . Audio evidence
' 44 ' on sleeve at basketball game MORE and Hillary
don t short change the race . They criticized
( 31 ) Save a turret from redemption (
said the board 's ruling went the right way . He

INTERVAL -0.973 - -0.510
CONTAINS 0.202%

Warning - Message " F ailed to $ ($ _. Exception
thought he 'd be a hockey dad when his kids were
connection ( OS I ) layers . Second , attacks from
' 2016 primary Sanders : ' Damn right ' I 'll
T ( t )- T 0 ) The year
Token self
Token ID2116
Feature activation+3.02e-01

Pos logit contributions
 axes+1.594
ochond+1.515
 acceler+1.434
OTOS+1.385
opsis+1.384
Neg logit contributions
util-2.968
onen-2.931
idium-2.894
hler-2.816
cone-2.787
Token.
Token ID764
Feature activation-4.07e-01

Pos logit contributions
util+3.177
onen+3.001
hler+2.879
idium+2.854
unia+2.715
Neg logit contributions
escription-3.037
ochond-2.856
 axes-2.751
 pollut-2.733
OTOS-2.689
Tokenhalf
Token ID13959
Feature activation+1.19e-01

Pos logit contributions
 axes+3.812
ochond+3.707
OTOS+3.490
 acceler+3.467
escription+3.431
Neg logit contributions
util-4.100
onen-4.055
idium-4.038
hler-3.901
illance-3.840
Token_
Token ID62
Feature activation+3.38e-02

Pos logit contributions
util+1.260
onen+1.230
idium+1.217
hler+1.179
cone+1.146
Neg logit contributions
 axes-1.119
ochond-1.112
 acceler-1.052
OTOS-1.052
opsis-1.044
Tokenm
Token ID76
Feature activation+2.71e-01

Pos logit contributions
util+0.368
onen+0.361
idium+0.357
hler+0.347
cone+0.339
Neg logit contributions
 axes-0.302
ochond-0.299
 acceler-0.281
opsis-0.280
OTOS-0.278
Tokenoves
Token ID5241
Feature activation+8.77e-01

Pos logit contributions
util+2.330
onen+2.168
idium+2.044
hler+2.022
unia+1.963
Neg logit contributions
escription-3.111
 pollut-3.098
 axes-3.056
ochond-2.986
 physic-2.932
Token,
Token ID837
Feature activation-7.74e-02

Pos logit contributions
util+8.854
onen+8.277
idium+8.062
hler+8.026
ilogy+7.626
Neg logit contributions
ochond-7.598
 pollut-7.477
 axes-7.424
OTOS-7.374
opsis-7.323
Token depth
Token ID6795
Feature activation+2.13e-01

Pos logit contributions
 axes+0.648
ochond+0.640
OTOS+0.608
 acceler+0.606
 pollut+0.604
Neg logit contributions
util-0.892
onen-0.873
idium-0.859
hler-0.835
cone-0.818
Token :
Token ID1058
Feature activation-1.40e-01

Pos logit contributions
util+2.232
onen+2.158
hler+2.056
idium+2.025
cone+1.989
Neg logit contributions
escription-2.031
 axes-1.977
 pollut-1.976
ochond-1.975
opsis-1.873
Token self
Token ID2116
Feature activation+2.91e-01

Pos logit contributions
 axes+0.948
ochond+0.917
 acceler+0.859
opsis+0.841
OTOS+0.839
Neg logit contributions
util-1.850
onen-1.826
idium-1.803
hler-1.757
cone-1.730
Token.
Token ID764
Feature activation-3.57e-01

Pos logit contributions
util+3.106
onen+2.929
hler+2.806
idium+2.801
cone+2.661
Neg logit contributions
escription-2.855
ochond-2.726
 axes-2.643
 pollut-2.590
OTOS-2.566
Token in
Token ID287
Feature activation+2.45e-01

Pos logit contributions
util+3.602
onen+3.217
hler+3.108
cone+3.104
idium+3.051
Neg logit contributions
escription-5.220
ochond-4.765
 axes-4.692
OTOS-4.624
 pollut-4.598
Token a
Token ID257
Feature activation+5.72e-01

Pos logit contributions
util+2.183
onen+2.025
idium+1.936
cone+1.935
hler+1.909
Neg logit contributions
escription-2.918
ochond-2.705
 axes-2.608
OTOS-2.579
opsis-2.575
Token skillet
Token ID41306
Feature activation+3.79e-01

Pos logit contributions
util+4.314
onen+4.095
cone+4.022
hler+3.928
idium+3.605
Neg logit contributions
escription-7.368
ochond-6.789
 axes-6.662
OTOS-6.662
opsis-6.563
Token over
Token ID625
Feature activation+2.65e-01

Pos logit contributions
util+3.131
onen+3.058
hler+2.898
cone+2.757
idium+2.730
Neg logit contributions
escription-4.579
ochond-4.298
 axes-4.224
OTOS-4.203
opsis-4.134
Token medium
Token ID7090
Feature activation+6.90e-01

Pos logit contributions
util+2.242
onen+2.123
idium+2.027
hler+1.948
cone+1.931
Neg logit contributions
escription-3.197
ochond-3.094
OTOS-3.016
opsis-2.984
 axes-2.968
Token-
Token ID12
Feature activation+8.64e-01

Pos logit contributions
util+4.546
onen+4.043
hler+3.732
cone+3.721
idium+3.719
Neg logit contributions
escription-9.675
ochond-8.525
opsis-8.470
OTOS-8.451
 axes-8.168
Tokenhigh
Token ID8929
Feature activation+3.58e-01

Pos logit contributions
util+5.057
onen+4.956
hey+4.666
cone+4.287
hler+4.269
Neg logit contributions
escription-12.454
OTOS-11.693
 axes-11.651
ochond-11.380
opsis-11.249
Token heat
Token ID4894
Feature activation+4.36e-01

Pos logit contributions
util+3.469
onen+3.307
hler+3.191
cone+3.140
idium+3.091
Neg logit contributions
escription-3.827
ochond-3.570
OTOS-3.508
opsis-3.487
 axes-3.456
Token.
Token ID13
Feature activation+9.85e-02

Pos logit contributions
util+3.500
onen+3.183
hler+3.065
cone+3.011
idium+2.990
Neg logit contributions
escription-5.661
ochond-5.056
 axes-5.055
 pollut-5.026
 subsequ-4.922
Token Place
Token ID8474
Feature activation+6.41e-01

Pos logit contributions
util+0.783
onen+0.731
hler+0.682
idium+0.677
cone+0.664
Neg logit contributions
escription-1.208
 axes-1.206
ochond-1.194
 pollut-1.167
opsis-1.142
Token f
Token ID277
Feature activation+2.30e-01

Pos logit contributions
util+5.159
onen+4.839
idium+4.683
hler+4.587
cone+4.561
Neg logit contributions
escription-7.713
OTOS-6.912
ochond-6.896
opsis-6.739
 axes-6.554
Tokenussian
Token ID31562
Feature activation+5.09e-01

Pos logit contributions
util+2.414
onen+2.252
ussen+1.933
uristic+1.866
hler+1.836
Neg logit contributions
escription-8.224
OTOS-7.876
 physic-7.859
ochond-7.829
 pollut-7.822
Token,
Token ID11
Feature activation+8.18e-01

Pos logit contributions
util+4.020
onen+3.913
idium+3.662
hler+3.620
cone+3.434
Neg logit contributions
escription-6.160
ochond-5.854
OTOS-5.722
opsis-5.506
 axes-5.468
Token then
Token ID788
Feature activation+1.86e-01

Pos logit contributions
util+3.965
onen+3.359
hler+3.182
kov+3.134
unia+2.934
Neg logit contributions
escription-12.870
ochond-11.481
OTOS-11.261
opsis-11.213
phies-10.994
Token apply
Token ID4174
Feature activation+3.66e-01

Pos logit contributions
util+1.900
onen+1.815
hler+1.740
idium+1.716
cone+1.691
Neg logit contributions
escription-1.848
ochond-1.778
 axes-1.753
OTOS-1.711
opsis-1.686
Token the
Token ID262
Feature activation+5.55e-01

Pos logit contributions
util+3.285
onen+3.044
hler+2.950
idium+2.907
cone+2.801
Neg logit contributions
escription-4.246
ochond-3.922
OTOS-3.783
opsis-3.757
 axes-3.742
Token black
Token ID2042
Feature activation+8.64e-01

Pos logit contributions
util+5.485
onen+5.443
idium+5.216
hler+5.103
uristic+5.031
Neg logit contributions
escription-5.507
ochond-5.302
OTOS-4.981
opsis-4.920
phies-4.697
Tokenhat
Token ID5183
Feature activation+5.54e-01

Pos logit contributions
util+7.570
hler+7.153
onen+7.023
cone+6.908
hey+6.848
Neg logit contributions
escription-8.762
OTOS-8.700
ochond-8.443
opsis-8.320
 acceler-8.306
Token #
Token ID1303
Feature activation+4.86e-01

Pos logit contributions
util+3.532
onen+3.322
idium+3.175
hler+3.043
cone+2.993
Neg logit contributions
ochond-7.338
 axes-7.260
opsis-7.080
OTOS-7.053
 acceler-7.018
Token morph
Token ID17488
Feature activation-4.63e-02

Pos logit contributions
util+4.533
onen+4.407
hler+4.189
idium+4.070
cone+3.960
Neg logit contributions
escription-4.948
ochond-4.832
OTOS-4.739
 axes-4.602
opsis-4.546
Tokenological
Token ID2770
Feature activation+4.17e-01

Pos logit contributions
 axes+0.598
ochond+0.586
OTOS+0.576
 pollut+0.574
escription+0.572
Neg logit contributions
util-0.329
onen-0.321
idium-0.308
hler-0.293
cone-0.280
Token operator
Token ID10088
Feature activation+2.33e-01

Pos logit contributions
util+4.996
onen+4.898
hler+4.667
idium+4.530
cone+4.520
Neg logit contributions
escription-3.475
ochond-3.137
OTOS-3.036
 axes-2.977
opsis-2.911
Token =
Token ID796
Feature activation+2.65e-01

Pos logit contributions
util+6.252
onen+5.973
cone+5.708
hler+5.696
idium+5.639
Neg logit contributions
ochond-6.258
 axes-6.102
opsis-6.043
OTOS-5.990
 acceler-5.977
Token c
Token ID269
Feature activation+2.33e-01

Pos logit contributions
util+2.405
onen+2.292
idium+2.235
hler+2.160
cone+2.157
Neg logit contributions
ochond-2.874
 axes-2.793
OTOS-2.742
opsis-2.740
escription-2.730
Tokenv
Token ID85
Feature activation+4.24e-01

Pos logit contributions
util+1.876
onen+1.649
hler+1.565
cone+1.554
idium+1.552
Neg logit contributions
escription-2.949
 axes-2.901
OTOS-2.885
 pollut-2.834
ochond-2.815
Token2
Token ID17
Feature activation+5.42e-01

Pos logit contributions
util+3.121
onen+2.612
hler+2.547
idium+2.542
cone+2.488
Neg logit contributions
escription-5.844
 pollut-5.462
 axes-5.396
ochond-5.355
OTOS-5.293
Token.
Token ID13
Feature activation+4.30e-01

Pos logit contributions
util+4.321
onen+3.717
hler+3.547
cone+3.510
pn+3.417
Neg logit contributions
escription-7.343
 dilig-6.539
 subsequ-6.495
 pollut-6.479
 axes-6.451
TokenGa
Token ID35389
Feature activation+8.53e-01

Pos logit contributions
util+4.094
onen+3.877
idium+3.718
cone+3.658
hler+3.623
Neg logit contributions
ochond-4.427
 axes-4.380
 pollut-4.291
OTOS-4.262
opsis-4.240
Tokenussian
Token ID31562
Feature activation+4.68e-01

Pos logit contributions
util+4.201
onen+3.791
unia+3.350
ussen+3.191
hler+3.069
Neg logit contributions
 axes-12.030
ochond-11.996
 pollut-11.891
 acceler-11.834
 DAC-11.823
TokenBl
Token ID3629
Feature activation+4.44e-01

Pos logit contributions
util+4.461
onen+4.301
idium+4.142
hler+4.062
cone+4.045
Neg logit contributions
ochond-4.778
 axes-4.669
opsis-4.538
 pollut-4.536
escription-4.497
Tokenur
Token ID333
Feature activation+4.77e-01

Pos logit contributions
util+3.501
onen+3.282
idium+3.067
hler+3.000
ussen+2.929
Neg logit contributions
 axes-5.405
ochond-5.318
OTOS-5.270
 acceler-5.234
escription-5.175
Token(
Token ID7
Feature activation+1.74e-01

Pos logit contributions
util+4.728
onen+4.536
idium+4.440
hler+4.354
cone+4.281
Neg logit contributions
ochond-4.685
 axes-4.647
 pollut-4.456
opsis-4.440
OTOS-4.436
Tokengray
Token ID44605
Feature activation+4.74e-01

Pos logit contributions
util+1.534
onen+1.479
idium+1.437
hler+1.388
cone+1.373
Neg logit contributions
ochond-1.940
 axes-1.926
OTOS-1.859
opsis-1.854
 pollut-1.848
Token Beaut
Token ID13711
Feature activation-1.56e-01

Pos logit contributions
util+1.694
onen+1.655
idium+1.609
hler+1.566
cone+1.531
Neg logit contributions
ochond-1.475
 axes-1.473
 pollut-1.393
OTOS-1.390
opsis-1.382
Tokenifully
Token ID17049
Feature activation+1.07e-01

Pos logit contributions
 axes+1.880
ochond+1.875
opsis+1.809
OTOS+1.809
 pollut+1.797
Neg logit contributions
util-1.206
onen-1.166
idium-1.140
hler-1.098
cone-1.066
Token Paid
Token ID47355
Feature activation+2.57e-01

Pos logit contributions
util+1.087
onen+1.055
idium+1.023
hler+0.997
cone+0.980
Neg logit contributions
 axes-1.030
ochond-1.025
OTOS-0.977
 pollut-0.973
opsis-0.968
Token Hom
Token ID8074
Feature activation+3.67e-03

Pos logit contributions
util+2.490
onen+2.415
idium+2.328
hler+2.289
cone+2.215
Neg logit contributions
ochond-2.569
 axes-2.556
 pollut-2.423
OTOS-2.405
opsis-2.391
Tokenage
Token ID496
Feature activation+3.10e-01

Pos logit contributions
util+0.031
onen+0.030
idium+0.029
hler+0.028
cone+0.027
Neg logit contributions
 axes-0.042
ochond-0.042
OTOS-0.040
opsis-0.040
 pollut-0.040
Token To
Token ID1675
Feature activation+8.51e-01

Pos logit contributions
util+3.275
onen+3.162
idium+3.078
hler+3.027
cone+2.930
Neg logit contributions
 axes-2.840
ochond-2.834
OTOS-2.684
opsis-2.663
 pollut-2.662
Token Lost
Token ID9164
Feature activation+3.25e-01

Pos logit contributions
util+7.485
onen+7.481
idium+6.929
hler+6.629
ilogy+6.433
Neg logit contributions
 axes-8.297
ochond-8.232
 acceler-7.910
escription-7.852
 pollut-7.704
Token Leader
Token ID10540
Feature activation+3.22e-01

Pos logit contributions
util+3.246
onen+3.176
idium+3.010
hler+2.972
ilogy+2.883
Neg logit contributions
 axes-3.128
ochond-3.123
 pollut-2.980
 acceler-2.974
OTOS-2.960
Token O
Token ID440
Feature activation+3.48e-01

Pos logit contributions
util+3.100
onen+3.030
idium+2.929
hler+2.831
cone+2.779
Neg logit contributions
 axes-3.231
ochond-3.213
 pollut-3.053
OTOS-3.048
 acceler-3.043
Tokender
Token ID1082
Feature activation-2.27e-01

Pos logit contributions
util+2.243
onen+2.195
idium+2.117
hler+1.991
cone+1.977
Neg logit contributions
 axes-4.660
escription-4.613
ochond-4.610
 acceler-4.509
 pollut-4.477
Tokenus
Token ID385
Feature activation+8.11e-02

Pos logit contributions
 axes+2.241
ochond+2.234
OTOS+2.132
 pollut+2.130
 acceler+2.122
Neg logit contributions
util-2.301
onen-2.234
idium-2.188
hler-2.116
cone-2.065
Token =
Token ID796
Feature activation+3.38e-01

Pos logit contributions
util+2.911
onen+2.800
idium+2.735
hler+2.678
cone+2.628
Neg logit contributions
ochond-2.641
 axes-2.616
 pollut-2.523
opsis-2.500
OTOS-2.489
Token im
Token ID545
Feature activation+2.22e-01

Pos logit contributions
util+3.569
onen+3.377
idium+3.298
cone+3.249
hler+3.219
Neg logit contributions
ochond-3.147
escription-2.986
OTOS-2.978
 axes-2.960
opsis-2.946
Tokenutils
Token ID26791
Feature activation+3.27e-01

Pos logit contributions
util+0.775
onen+0.606
idium+0.578
hler+0.497
ilogy+0.447
Neg logit contributions
 axes-3.734
ochond-3.683
escription-3.671
 pollut-3.663
 acceler-3.639
Token.
Token ID13
Feature activation+2.19e-01

Pos logit contributions
util+3.044
onen+2.685
hler+2.617
idium+2.588
cone+2.479
Neg logit contributions
ochond-3.575
 axes-3.569
escription-3.567
 pollut-3.560
 acceler-3.432
Tokenres
Token ID411
Feature activation+1.21e-01

Pos logit contributions
util+2.032
onen+1.929
idium+1.862
hler+1.793
cone+1.779
Neg logit contributions
ochond-2.358
 axes-2.356
 pollut-2.290
OTOS-2.267
 acceler-2.259
Tokenize
Token ID1096
Feature activation+8.48e-01

Pos logit contributions
util+1.506
onen+1.451
idium+1.425
hler+1.391
cone+1.374
Neg logit contributions
ochond-0.912
 axes-0.908
OTOS-0.868
 pollut-0.867
 acceler-0.856
Token(
Token ID7
Feature activation+3.78e-01

Pos logit contributions
util+7.483
onen+6.854
cone+6.616
idium+6.521
irtual+6.394
Neg logit contributions
ochond-8.695
 axes-8.610
 pollut-8.598
 acceler-8.547
escription-8.369
Tokenimage
Token ID9060
Feature activation+4.68e-01

Pos logit contributions
util+2.862
onen+2.638
idium+2.542
cone+2.501
hler+2.443
Neg logit contributions
ochond-4.674
 axes-4.603
OTOS-4.518
 pollut-4.513
 acceler-4.493
Token,
Token ID11
Feature activation+1.80e-01

Pos logit contributions
util+5.358
onen+5.090
idium+4.998
hler+4.937
cone+4.876
Neg logit contributions
ochond-3.896
 axes-3.751
 pollut-3.678
opsis-3.633
 acceler-3.621
Token height
Token ID6001
Feature activation+1.56e-01

Pos logit contributions
util+1.959
onen+1.886
idium+1.851
hler+1.799
cone+1.791
Neg logit contributions
ochond-1.640
 axes-1.571
OTOS-1.544
opsis-1.538
escription-1.517
Token=
Token ID28
Feature activation+2.00e-01

Pos logit contributions
util+1.837
onen+1.781
idium+1.752
hler+1.711
cone+1.683
Neg logit contributions
ochond-1.261
 axes-1.249
opsis-1.189
 pollut-1.188
OTOS-1.187
Token film
Token ID2646
Feature activation+6.21e-01

Pos logit contributions
util+3.698
onen+3.618
idium+3.544
cone+3.413
hler+3.408
Neg logit contributions
ochond-3.069
 axes-3.033
OTOS-2.914
opsis-2.894
 pollut-2.860
Token
Token ID564
Feature activation+1.35e-01

Pos logit contributions
util+5.793
onen+5.602
idium+5.489
cone+5.357
ilogy+5.326
Neg logit contributions
ochond-6.151
 axes-6.078
OTOS-5.715
opsis-5.707
 pollut-5.666
Token
Token ID250
Feature activation+3.98e-01

Pos logit contributions
util+0.967
onen+0.932
idium+0.907
hler+0.878
cone+0.850
Neg logit contributions
ochond-1.728
 axes-1.711
escription-1.667
 pollut-1.658
opsis-1.643
TokenE
Token ID36
Feature activation+5.45e-02

Pos logit contributions
util+3.172
onen+3.041
idium+2.939
cone+2.829
hler+2.810
Neg logit contributions
ochond-4.622
 axes-4.608
 pollut-4.521
 acceler-4.449
opsis-4.428
Tokenscape
Token ID6794
Feature activation+4.31e-01

Pos logit contributions
util+0.501
onen+0.479
idium+0.468
hler+0.456
cone+0.453
Neg logit contributions
 axes-0.595
ochond-0.581
escription-0.564
 pollut-0.564
 acceler-0.563
Token from
Token ID422
Feature activation+8.33e-01

Pos logit contributions
util+4.487
onen+4.345
idium+4.295
hler+4.156
cone+4.075
Neg logit contributions
 axes-3.984
ochond-3.970
 pollut-3.773
opsis-3.760
OTOS-3.758
Token Al
Token ID978
Feature activation+3.03e-01

Pos logit contributions
util+5.318
onen+5.070
idium+4.934
hler+4.627
cone+4.578
Neg logit contributions
 axes-10.336
ochond-10.312
 pollut-9.979
opsis-9.979
escription-9.859
Tokencat
Token ID9246
Feature activation+1.05e-01

Pos logit contributions
util+2.191
onen+2.032
cone+1.895
idium+1.889
hler+1.830
Neg logit contributions
escription-4.076
 axes-3.975
OTOS-3.840
 acceler-3.838
 pollut-3.824
Tokenraz
Token ID3247
Feature activation+3.57e-01

Pos logit contributions
util+1.041
onen+1.010
idium+0.967
hler+0.945
cone+0.938
Neg logit contributions
escription-1.099
 axes-1.072
ochond-1.067
OTOS-1.028
 pollut-1.027
Token
Token ID447
Feature activation+1.16e-01

Pos logit contributions
util+3.375
onen+3.286
idium+3.212
hler+3.102
cone+3.046
Neg logit contributions
 axes-3.630
ochond-3.627
OTOS-3.460
 pollut-3.422
opsis-3.408
Token
Token ID251
Feature activation+1.19e-01

Pos logit contributions
util+1.121
onen+1.092
idium+1.069
hler+1.033
cone+1.010
Neg logit contributions
 axes-1.177
ochond-1.173
OTOS-1.121
opsis-1.119
 pollut-1.117
Token 3
Token ID513
Feature activation+1.87e-01

Pos logit contributions
util+3.427
onen+3.416
hler+3.172
idium+3.169
cone+3.165
Neg logit contributions
escription-3.941
ochond-3.834
OTOS-3.639
opsis-3.629
 axes-3.517
Tokenx
Token ID87
Feature activation-3.86e-02

Pos logit contributions
util+1.802
onen+1.733
idium+1.678
cone+1.630
hler+1.607
Neg logit contributions
escription-2.035
ochond-1.913
 axes-1.885
 pollut-1.884
OTOS-1.857
Token3
Token ID18
Feature activation+4.60e-01

Pos logit contributions
escription+0.414
ochond+0.410
 axes+0.407
 pollut+0.395
opsis+0.390
Neg logit contributions
util-0.362
onen-0.349
idium-0.335
hler-0.334
cone-0.325
Token Ga
Token ID12822
Feature activation+5.29e-01

Pos logit contributions
util+2.888
onen+2.717
hler+2.599
cone+2.596
idium+2.566
Neg logit contributions
escription-6.314
ochond-6.095
OTOS-5.858
opsis-5.811
 pollut-5.745
Tokenussian
Token ID31562
Feature activation+5.09e-01

Pos logit contributions
util+2.414
onen+2.252
ussen+1.933
uristic+1.866
hler+1.836
Neg logit contributions
escription-8.224
OTOS-7.876
 physic-7.859
ochond-7.829
 pollut-7.822
Token,
Token ID11
Feature activation+8.18e-01

Pos logit contributions
util+4.020
onen+3.913
idium+3.662
hler+3.620
cone+3.434
Neg logit contributions
escription-6.160
ochond-5.854
OTOS-5.722
opsis-5.506
 axes-5.468
Token then
Token ID788
Feature activation+1.86e-01

Pos logit contributions
util+3.965
onen+3.359
hler+3.182
kov+3.134
unia+2.934
Neg logit contributions
escription-12.870
ochond-11.481
OTOS-11.261
opsis-11.213
phies-10.994
Token apply
Token ID4174
Feature activation+3.66e-01

Pos logit contributions
util+1.900
onen+1.815
hler+1.740
idium+1.716
cone+1.691
Neg logit contributions
escription-1.848
ochond-1.778
 axes-1.753
OTOS-1.711
opsis-1.686
Token the
Token ID262
Feature activation+5.55e-01

Pos logit contributions
util+3.285
onen+3.044
hler+2.950
idium+2.907
cone+2.801
Neg logit contributions
escription-4.246
ochond-3.922
OTOS-3.783
opsis-3.757
 axes-3.742
Token black
Token ID2042
Feature activation+8.64e-01

Pos logit contributions
util+5.485
onen+5.443
idium+5.216
hler+5.103
uristic+5.031
Neg logit contributions
escription-5.507
ochond-5.302
OTOS-4.981
opsis-4.920
phies-4.697
Tokenhat
Token ID5183
Feature activation+5.54e-01

Pos logit contributions
util+7.570
hler+7.153
onen+7.023
cone+6.908
hey+6.848
Neg logit contributions
escription-8.762
OTOS-8.700
ochond-8.443
opsis-8.320
 acceler-8.306
Token-
Token ID12
Feature activation+2.89e-01

Pos logit contributions
util+1.061
onen+1.034
idium+1.013
hler+0.984
cone+0.960
Neg logit contributions
 axes-0.991
ochond-0.989
OTOS-0.943
opsis-0.939
 pollut-0.939
Token31
Token ID3132
Feature activation+1.77e-04

Pos logit contributions
util+2.519
onen+2.380
idium+2.299
hler+2.276
cone+2.174
Neg logit contributions
ochond-3.205
 axes-3.196
 pollut-3.120
OTOS-3.074
opsis-3.049
Token.
Token ID13
Feature activation+1.21e-01

Pos logit contributions
util+0.002
onen+0.002
idium+0.002
hler+0.002
cone+0.002
Neg logit contributions
 axes-0.002
ochond-0.002
OTOS-0.002
opsis-0.002
 acceler-0.002
Token
Token ID198
Feature activation+1.97e-01

Pos logit contributions
util+1.251
onen+1.219
idium+1.194
hler+1.161
cone+1.126
Neg logit contributions
ochond-1.151
 axes-1.148
 pollut-1.094
opsis-1.088
OTOS-1.088
Token
Token ID198
Feature activation+4.02e-01

Pos logit contributions
util+1.822
onen+1.707
idium+1.617
hler+1.615
cone+1.562
Neg logit contributions
escription-2.296
ochond-2.141
 axes-2.141
 pollut-2.092
opsis-2.039
Token[
Token ID58
Feature activation+8.11e-01

Pos logit contributions
util+3.956
onen+3.845
idium+3.764
hler+3.638
cone+3.560
Neg logit contributions
 axes-4.031
ochond-4.008
 acceler-3.817
 pollut-3.817
opsis-3.816
Token24
Token ID1731
Feature activation+3.68e-01

Pos logit contributions
util+3.589
onen+3.392
idium+2.902
hey+2.801
cone+2.776
Neg logit contributions
escription-12.964
 dilig-11.922
 subsequ-11.902
 axes-11.698
 pollut-11.617
Token]
Token ID60
Feature activation+1.47e-02

Pos logit contributions
util+3.713
onen+3.492
hler+3.435
idium+3.318
kov+3.218
Neg logit contributions
escription-4.022
 axes-3.616
ochond-3.582
 pollut-3.556
 subsequ-3.532
Token Nah
Token ID35201
Feature activation+6.22e-02

Pos logit contributions
util+0.151
onen+0.147
idium+0.144
hler+0.139
cone+0.136
Neg logit contributions
 axes-0.141
ochond-0.141
OTOS-0.134
 pollut-0.133
opsis-0.133
Tokenman
Token ID805
Feature activation-6.91e-02

Pos logit contributions
util+0.590
onen+0.578
idium+0.566
hler+0.543
cone+0.530
Neg logit contributions
 axes-0.646
ochond-0.639
OTOS-0.618
 pollut-0.616
opsis-0.615
Tokenides
Token ID1460
Feature activation-2.43e-01

Pos logit contributions
 axes+0.639
ochond+0.637
OTOS+0.605
opsis+0.604
 pollut+0.602
Neg logit contributions
util-0.734
onen-0.716
idium-0.702
hler-0.681
cone-0.667
Token-
Token ID12
Feature activation+2.94e-01

Pos logit contributions
 axes+0.757
ochond+0.744
OTOS+0.712
 pollut+0.708
 acceler+0.707
Neg logit contributions
util-0.930
onen-0.912
idium-0.890
hler-0.869
cone-0.846
Tokento
Token ID1462
Feature activation+1.72e-01

Pos logit contributions
util+3.201
onen+3.128
idium+3.069
hler+2.979
cone+2.917
Neg logit contributions
 axes-2.629
ochond-2.620
OTOS-2.488
opsis-2.480
 pollut-2.475
Token-
Token ID12
Feature activation-1.49e-01

Pos logit contributions
util+1.578
onen+1.522
idium+1.430
hler+1.430
cone+1.368
Neg logit contributions
escription-1.898
 axes-1.878
ochond-1.856
 pollut-1.810
opsis-1.768
Token5
Token ID20
Feature activation+4.06e-01

Pos logit contributions
 axes+1.682
ochond+1.677
OTOS+1.608
opsis+1.607
 pollut+1.605
Neg logit contributions
util-1.275
onen-1.237
idium-1.205
hler-1.161
cone-1.130
Token Working
Token ID14594
Feature activation+1.80e-01

Pos logit contributions
util+3.549
onen+3.425
idium+3.351
hler+3.305
cone+3.149
Neg logit contributions
 axes-4.383
ochond-4.376
opsis-4.206
OTOS-4.200
 acceler-4.156
Token Women
Token ID6926
Feature activation+8.11e-01

Pos logit contributions
util+1.703
onen+1.653
idium+1.616
hler+1.568
cone+1.522
Neg logit contributions
 axes-1.860
ochond-1.854
 acceler-1.769
OTOS-1.768
 pollut-1.766
Token Education
Token ID7868
Feature activation+3.54e-01

Pos logit contributions
util+6.258
hler+5.982
onen+5.929
idium+5.887
cone+5.545
Neg logit contributions
ochond-9.021
 axes-8.959
opsis-8.631
OTOS-8.587
escription-8.583
Token Fund
Token ID7557
Feature activation+2.83e-01

Pos logit contributions
util+4.060
onen+3.972
idium+3.885
hler+3.800
cone+3.704
Neg logit contributions
 axes-2.958
ochond-2.935
opsis-2.782
OTOS-2.778
 pollut-2.767
Token:
Token ID25
Feature activation+4.29e-01

Pos logit contributions
util+2.922
onen+2.849
idium+2.778
hler+2.709
cone+2.632
Neg logit contributions
ochond-2.662
 axes-2.652
OTOS-2.523
 pollut-2.513
opsis-2.505
Token...
Token ID2644
Feature activation+1.41e-01

Pos logit contributions
util+4.229
onen+4.085
idium+4.024
hler+3.883
cone+3.769
Neg logit contributions
ochond-4.209
 axes-4.099
OTOS-3.916
escription-3.915
 pollut-3.915
Token $
Token ID720
Feature activation+2.75e-01

Pos logit contributions
util+1.381
onen+1.337
idium+1.308
hler+1.269
cone+1.238
Neg logit contributions
ochond-1.418
 axes-1.406
OTOS-1.346
opsis-1.343
 pollut-1.340
Token 6
Token ID718
Feature activation+2.99e-01

Pos logit contributions
util+3.438
onen+3.315
idium+3.215
hler+3.165
cone+3.073
Neg logit contributions
ochond-3.895
 axes-3.824
 pollut-3.688
opsis-3.682
OTOS-3.664
Token}
Token ID92
Feature activation+1.72e-01

Pos logit contributions
util+2.470
onen+2.397
idium+2.335
hler+2.247
cone+2.184
Neg logit contributions
 axes-3.452
ochond-3.444
OTOS-3.308
opsis-3.301
 pollut-3.298
Token <
Token ID1279
Feature activation-3.42e-01

Pos logit contributions
util+1.554
onen+1.509
idium+1.466
hler+1.425
cone+1.380
Neg logit contributions
ochond-1.858
 axes-1.841
OTOS-1.769
opsis-1.768
 pollut-1.763
Token->
Token ID3784
Feature activation-5.03e-02

Pos logit contributions
 axes+2.352
ochond+2.342
OTOS+2.254
opsis+2.236
escription+2.198
Neg logit contributions
onen-4.189
util-4.180
idium-4.124
ilogy-3.951
ussen-3.943
Token {
Token ID1391
Feature activation+1.59e-01

Pos logit contributions
 axes+0.624
ochond+0.605
 acceler+0.591
 pollut+0.590
opsis+0.588
Neg logit contributions
util-0.382
onen-0.374
idium-0.366
hler-0.349
cone-0.337
Token1
Token ID16
Feature activation+8.04e-01

Pos logit contributions
util+1.365
onen+1.331
idium+1.301
hler+1.248
cone+1.216
Neg logit contributions
 axes-1.789
ochond-1.770
OTOS-1.707
opsis-1.704
 acceler-1.695
Token,
Token ID11
Feature activation+3.69e-01

Pos logit contributions
util+5.213
onen+4.824
hler+4.481
kov+4.307
hey+4.304
Neg logit contributions
escription-11.496
 dilig-10.500
 subsequ-10.144
 pollut-10.005
ochond-9.755
Token 6
Token ID718
Feature activation+3.82e-01

Pos logit contributions
util+3.630
onen+3.524
idium+3.435
hler+3.352
cone+3.260
Neg logit contributions
ochond-3.674
 axes-3.657
OTOS-3.494
opsis-3.491
 pollut-3.484
Token},
Token ID5512
Feature activation-1.84e-02

Pos logit contributions
util+3.202
onen+3.110
idium+3.024
hler+2.924
cone+2.839
Neg logit contributions
 axes-4.353
ochond-4.347
OTOS-4.174
opsis-4.163
 pollut-4.162
Token {
Token ID1391
Feature activation+1.64e-01

Pos logit contributions
 axes+0.232
ochond+0.226
 pollut+0.222
 acceler+0.221
opsis+0.220
Neg logit contributions
util-0.135
onen-0.132
idium-0.131
hler-0.123
cone-0.120
Token1
Token ID16
Feature activation+7.06e-01

Pos logit contributions
util+1.419
onen+1.385
idium+1.357
hler+1.300
cone+1.269
Neg logit contributions
 axes-1.836
ochond-1.815
opsis-1.751
OTOS-1.750
 acceler-1.740
Token who
Token ID508
Feature activation-3.25e-01

Pos logit contributions
 axes+2.634
ochond+2.515
 acceler+2.467
 pollut+2.412
OTOS+2.387
Neg logit contributions
util-3.832
onen-3.774
idium-3.721
hler-3.580
ilogy-3.550
Token say
Token ID910
Feature activation-2.54e-01

Pos logit contributions
 axes+2.833
ochond+2.762
 acceler+2.645
 pollut+2.633
opsis+2.630
Neg logit contributions
util-3.583
onen-3.502
idium-3.462
hler-3.314
avorite-3.279
Token that
Token ID326
Feature activation-3.36e-01

Pos logit contributions
 axes+2.096
ochond+2.044
 acceler+1.995
 pollut+1.956
OTOS+1.930
Neg logit contributions
util-2.941
onen-2.875
idium-2.853
hler-2.741
ilogy-2.713
Token making
Token ID1642
Feature activation-1.49e-01

Pos logit contributions
 axes+2.858
ochond+2.752
 acceler+2.735
 pollut+2.662
opsis+2.622
Neg logit contributions
util-3.802
onen-3.719
idium-3.686
hler-3.535
irtual-3.487
Token middle
Token ID3504
Feature activation-3.29e-01

Pos logit contributions
 axes+1.307
ochond+1.289
 acceler+1.233
 pollut+1.225
OTOS+1.221
Neg logit contributions
util-1.652
onen-1.615
idium-1.589
hler-1.537
cone-1.510
Token and
Token ID290
Feature activation+7.80e-01

Pos logit contributions
 axes+2.528
ochond+2.482
 acceler+2.368
 pollut+2.338
opsis+2.302
Neg logit contributions
util-3.954
onen-3.912
idium-3.884
hler-3.726
avorite-3.688
Token high
Token ID1029
Feature activation+6.47e-02

Pos logit contributions
util+5.805
onen+5.794
idium+5.669
hler+5.637
ussen+5.421
Neg logit contributions
escription-8.415
ochond-8.369
OTOS-8.091
 axes-7.980
opsis-7.961
Token school
Token ID1524
Feature activation-1.27e-01

Pos logit contributions
util+0.729
onen+0.710
idium+0.690
hler+0.678
cone+0.663
Neg logit contributions
ochond-0.547
 axes-0.546
opsis-0.528
OTOS-0.526
escription-0.514
Tokeners
Token ID364
Feature activation-2.61e-01

Pos logit contributions
 axes+1.139
ochond+1.133
OTOS+1.075
 acceler+1.073
 pollut+1.073
Neg logit contributions
util-1.380
onen-1.348
idium-1.325
hler-1.284
cone-1.258
Token start
Token ID923
Feature activation-9.94e-02

Pos logit contributions
 axes+2.251
ochond+2.200
 acceler+2.137
 pollut+2.107
opsis+2.075
Neg logit contributions
util-2.924
onen-2.870
idium-2.845
hler-2.721
ilogy-2.686
Token classes
Token ID6097
Feature activation-3.47e-01

Pos logit contributions
 axes+0.876
ochond+0.875
OTOS+0.831
opsis+0.829
 pollut+0.827
Neg logit contributions
util-1.097
onen-1.072
idium-1.049
hler-1.020
cone-1.000
Token the
Token ID262
Feature activation-9.01e-02

Pos logit contributions
 axes+1.052
ochond+1.046
OTOS+0.990
 pollut+0.988
opsis+0.988
Neg logit contributions
util-1.379
onen-1.349
idium-1.326
hler-1.286
cone-1.263
Token commitment
Token ID7901
Feature activation+2.00e-02

Pos logit contributions
 axes+0.758
ochond+0.754
OTOS+0.713
opsis+0.712
 pollut+0.711
Neg logit contributions
util-1.031
onen-1.009
idium-0.991
hler-0.963
cone-0.945
Token to
Token ID284
Feature activation+2.16e-02

Pos logit contributions
util+0.229
onen+0.224
idium+0.220
hler+0.214
cone+0.210
Neg logit contributions
 axes-0.169
ochond-0.168
OTOS-0.159
opsis-0.159
 pollut-0.158
Token advancing
Token ID19988
Feature activation-2.96e-01

Pos logit contributions
util+0.236
onen+0.231
idium+0.225
hler+0.220
cone+0.215
Neg logit contributions
ochond-0.192
 axes-0.191
opsis-0.182
OTOS-0.182
escription-0.180
Token civil
Token ID3026
Feature activation-8.00e-02

Pos logit contributions
 axes+2.459
ochond+2.398
 acceler+2.310
 pollut+2.290
OTOS+2.247
Neg logit contributions
util-3.423
onen-3.364
idium-3.319
hler-3.193
cone-3.148
Token and
Token ID290
Feature activation+7.75e-01

Pos logit contributions
ochond+0.724
 axes+0.724
OTOS+0.688
opsis+0.685
 pollut+0.682
Neg logit contributions
util-0.864
onen-0.843
idium-0.826
hler-0.803
cone-0.786
Token human
Token ID1692
Feature activation+1.95e-01

Pos logit contributions
util+6.578
onen+6.365
hler+6.028
idium+5.867
irtual+5.687
Neg logit contributions
escription-7.834
ochond-7.753
opsis-7.582
 axes-7.518
OTOS-7.337
Token rights
Token ID2489
Feature activation+2.48e-01

Pos logit contributions
util+1.728
onen+1.628
idium+1.609
hler+1.572
cone+1.505
Neg logit contributions
ochond-2.137
escription-2.062
 axes-2.054
opsis-2.047
OTOS-2.033
Token
Token ID960
Feature activation-2.10e-03

Pos logit contributions
util+2.584
onen+2.493
idium+2.419
hler+2.367
cone+2.331
Neg logit contributions
ochond-2.277
 axes-2.243
escription-2.222
OTOS-2.221
opsis-2.159
Tokenthese
Token ID27218
Feature activation+9.45e-02

Pos logit contributions
 axes+0.023
ochond+0.023
opsis+0.022
OTOS+0.022
 pollut+0.022
Neg logit contributions
util-0.019
onen-0.019
idium-0.018
hler-0.017
cone-0.017
Token are
Token ID389
Feature activation+1.81e-01

Pos logit contributions
util+1.000
onen+0.974
idium+0.947
hler+0.926
cone+0.909
Neg logit contributions
ochond-0.878
 axes-0.851
OTOS-0.834
opsis-0.832
escription-0.828
TokenAIR
Token ID42149
Feature activation-2.52e-01

Pos logit contributions
 axes+0.001
ochond+0.001
opsis+0.001
 pollut+0.001
OTOS+0.001
Neg logit contributions
util-0.001
onen-0.001
idium-0.001
hler-0.001
cone-0.001
Token
Token ID447
Feature activation-3.13e-01

Pos logit contributions
 axes+2.085
ochond+2.034
 acceler+1.980
 pollut+1.942
OTOS+1.933
Neg logit contributions
util-2.862
onen-2.835
idium-2.802
hler-2.710
cone-2.672
Token
Token ID247
Feature activation-3.93e-01

Pos logit contributions
 axes+2.984
ochond+2.939
opsis+2.827
OTOS+2.822
 acceler+2.817
Neg logit contributions
util-3.173
onen-3.111
idium-3.043
ilogy-2.898
hler-2.897
Tokens
Token ID82
Feature activation-2.39e-01

Pos logit contributions
 axes+3.286
ochond+3.193
 acceler+3.131
opsis+3.045
 pollut+3.039
Neg logit contributions
onen-4.302
util-4.295
idium-4.256
hler-4.104
avorite-4.095
Token Health
Token ID3893
Feature activation+2.66e-02

Pos logit contributions
 axes+2.079
ochond+2.020
 acceler+1.953
 pollut+1.952
opsis+1.923
Neg logit contributions
util-2.666
onen-2.613
idium-2.580
hler-2.495
cone-2.446
Token and
Token ID290
Feature activation+7.73e-01

Pos logit contributions
util+0.272
onen+0.266
idium+0.259
hler+0.252
cone+0.246
Neg logit contributions
 axes-0.256
ochond-0.255
opsis-0.243
OTOS-0.243
 pollut-0.242
Token Social
Token ID5483
Feature activation+3.06e-01

Pos logit contributions
util+6.273
onen+6.027
idium+5.611
hler+5.585
cone+5.421
Neg logit contributions
ochond-8.517
 axes-8.459
escription-8.275
opsis-8.243
OTOS-8.147
Token Development
Token ID7712
Feature activation-5.82e-03

Pos logit contributions
util+2.922
onen+2.815
idium+2.736
hler+2.710
illance+2.568
Neg logit contributions
ochond-3.112
 axes-3.111
opsis-3.016
OTOS-3.005
escription-2.961
Token program
Token ID1430
Feature activation+7.57e-02

Pos logit contributions
 axes+0.053
ochond+0.053
OTOS+0.050
 acceler+0.050
opsis+0.050
Neg logit contributions
util-0.062
onen-0.061
idium-0.060
hler-0.058
cone-0.057
Token develops
Token ID21126
Feature activation-6.67e-02

Pos logit contributions
util+0.871
onen+0.848
idium+0.830
hler+0.808
cone+0.791
Neg logit contributions
ochond-0.634
 axes-0.630
OTOS-0.602
opsis-0.597
 pollut-0.594
Token knowledge
Token ID3725
Feature activation-1.17e-01

Pos logit contributions
 axes+0.582
ochond+0.576
 pollut+0.547
 acceler+0.547
OTOS+0.545
Neg logit contributions
util-0.745
onen-0.728
idium-0.717
hler-0.695
cone-0.681
Tokenv
Token ID85
Feature activation+4.99e-01

Pos logit contributions
util+0.607
onen+0.556
idium+0.531
hler+0.523
cone+0.517
Neg logit contributions
escription-0.851
 axes-0.848
OTOS-0.830
ochond-0.830
 pollut-0.823
Token2
Token ID17
Feature activation+4.93e-01

Pos logit contributions
util+3.430
onen+2.844
idium+2.735
hler+2.733
cone+2.651
Neg logit contributions
escription-7.324
 pollut-6.648
 dilig-6.547
 subsequ-6.541
 axes-6.505
Token.
Token ID764
Feature activation+4.86e-01

Pos logit contributions
util+4.700
onen+4.230
hler+4.110
idium+3.984
cone+3.906
Neg logit contributions
escription-5.899
ochond-5.011
 pollut-4.976
 dilig-4.958
 subsequ-4.945
Token Ga
Token ID12822
Feature activation+7.40e-01

Pos logit contributions
util+4.889
onen+4.709
idium+4.571
hler+4.522
cone+4.478
Neg logit contributions
ochond-4.590
OTOS-4.454
opsis-4.389
escription-4.360
 axes-4.310
Tokenussian
Token ID31562
Feature activation+6.13e-01

Pos logit contributions
util+3.348
onen+3.048
ussen+2.505
uristic+2.491
hler+2.406
Neg logit contributions
escription-11.197
OTOS-11.055
 DAC-10.980
 physic-10.883
 pollut-10.849
TokenBl
Token ID3629
Feature activation+7.71e-01

Pos logit contributions
util+4.167
onen+3.903
hler+3.569
idium+3.505
kov+3.360
Neg logit contributions
escription-8.257
ochond-7.703
opsis-7.635
 pollut-7.610
 subsequ-7.544
Tokenur
Token ID333
Feature activation+4.94e-01

Pos logit contributions
util+1.770
uristic+1.176
onen+1.129
unia+0.741
ussen+0.495
Neg logit contributions
escription-14.059
OTOS-13.484
 physic-13.415
 axes-13.401
 dilig-13.304
Token (
Token ID357
Feature activation+3.80e-01

Pos logit contributions
util+3.496
onen+3.218
hler+2.978
idium+2.949
cone+2.806
Neg logit contributions
escription-6.883
 pollut-6.195
 axes-6.078
ochond-6.058
 subsequ-6.045
Token gray
Token ID12768
Feature activation+4.52e-01

Pos logit contributions
util+2.958
onen+2.827
idium+2.764
hler+2.661
cone+2.611
Neg logit contributions
escription-4.652
ochond-4.433
OTOS-4.432
opsis-4.377
phies-4.121
Token,
Token ID837
Feature activation+1.20e-02

Pos logit contributions
util+4.359
onen+4.085
idium+3.860
hler+3.799
kov+3.701
Neg logit contributions
escription-5.089
ochond-4.547
 subsequ-4.538
 pollut-4.502
OTOS-4.445
Token (
Token ID357
Feature activation+3.08e-01

Pos logit contributions
util+0.109
onen+0.105
idium+0.101
hler+0.098
cone+0.098
Neg logit contributions
ochond-0.131
escription-0.129
 axes-0.128
OTOS-0.126
opsis-0.126
Token
Token ID198
Feature activation+3.42e-01

Pos logit contributions
util+2.310
onen+2.149
hler+2.031
idium+2.026
cone+1.981
Neg logit contributions
escription-3.030
 axes-2.771
ochond-2.758
 pollut-2.724
OTOS-2.644
Tokenn
Token ID77
Feature activation+6.25e-01

Pos logit contributions
util+3.206
onen+3.031
idium+2.911
hler+2.849
cone+2.763
Neg logit contributions
 axes-3.701
ochond-3.591
 acceler-3.565
escription-3.533
 pollut-3.527
Tokenpm
Token ID4426
Feature activation+3.47e-01

Pos logit contributions
util+2.752
onen+2.515
hler+2.003
unia+1.838
idium+1.833
Neg logit contributions
OTOS-9.835
 axes-9.675
escription-9.668
 DAC-9.333
 physic-9.327
Token install
Token ID2721
Feature activation+5.50e-01

Pos logit contributions
util+3.026
onen+2.867
hler+2.603
idium+2.597
unia+2.546
Neg logit contributions
ochond-3.930
 axes-3.837
OTOS-3.768
escription-3.759
opsis-3.673
Token -
Token ID532
Feature activation+4.29e-01

Pos logit contributions
util+6.091
onen+5.901
idium+5.662
hler+5.496
ussen+5.418
Neg logit contributions
ochond-4.712
OTOS-4.524
 axes-4.499
escription-4.439
opsis-4.391
Tokeng
Token ID70
Feature activation+7.65e-01

Pos logit contributions
util+3.618
onen+3.142
hler+3.014
idium+2.962
cone+2.952
Neg logit contributions
escription-4.997
ochond-4.974
 pollut-4.775
 axes-4.766
 subsequ-4.718
Token server
Token ID4382
Feature activation+4.44e-02

Pos logit contributions
util+8.299
onen+7.975
idium+7.536
hler+7.455
ussen+7.213
Neg logit contributions
ochond-6.410
OTOS-6.106
 axes-6.081
escription-5.989
opsis-5.930
Tokenless
Token ID1203
Feature activation+4.13e-01

Pos logit contributions
util+0.461
onen+0.437
idium+0.425
hler+0.417
cone+0.408
Neg logit contributions
ochond-0.440
 axes-0.440
escription-0.424
OTOS-0.423
 pollut-0.417
Token
Token ID198
Feature activation-6.39e-02

Pos logit contributions
util+4.328
onen+4.162
idium+3.979
hler+3.941
cone+3.815
Neg logit contributions
ochond-3.893
 axes-3.872
OTOS-3.776
 pollut-3.640
opsis-3.634
Token
Token ID198
Feature activation-3.80e-01

Pos logit contributions
ochond+0.619
 axes+0.618
escription+0.611
 pollut+0.593
OTOS+0.587
Neg logit contributions
util-0.662
onen-0.639
idium-0.619
hler-0.606
cone-0.593
TokenYou
Token ID1639
Feature activation-1.42e-01

Pos logit contributions
 axes+3.890
ochond+3.872
opsis+3.747
OTOS+3.708
 acceler+3.656
Neg logit contributions
util-3.422
idium-3.375
onen-3.370
hler-3.228
avorite-3.219
Tokenimage
Token ID9060
Feature activation+4.68e-01

Pos logit contributions
util+2.862
onen+2.638
idium+2.542
cone+2.501
hler+2.443
Neg logit contributions
ochond-4.674
 axes-4.603
OTOS-4.518
 pollut-4.513
 acceler-4.493
Token,
Token ID11
Feature activation+1.80e-01

Pos logit contributions
util+5.358
onen+5.090
idium+4.998
hler+4.937
cone+4.876
Neg logit contributions
ochond-3.896
 axes-3.751
 pollut-3.678
opsis-3.633
 acceler-3.621
Token height
Token ID6001
Feature activation+1.56e-01

Pos logit contributions
util+1.959
onen+1.886
idium+1.851
hler+1.799
cone+1.791
Neg logit contributions
ochond-1.640
 axes-1.571
OTOS-1.544
opsis-1.538
escription-1.517
Token=
Token ID28
Feature activation+2.00e-01

Pos logit contributions
util+1.837
onen+1.781
idium+1.752
hler+1.711
cone+1.683
Neg logit contributions
ochond-1.261
 axes-1.249
opsis-1.189
 pollut-1.188
OTOS-1.187
Token600
Token ID8054
Feature activation+5.73e-01

Pos logit contributions
util+1.847
onen+1.742
idium+1.707
hler+1.658
cone+1.652
Neg logit contributions
ochond-2.164
 axes-2.118
opsis-2.069
 pollut-2.060
OTOS-2.059
Token)
Token ID8
Feature activation+7.62e-01

Pos logit contributions
util+5.249
onen+4.972
idium+4.946
hler+4.778
cone+4.713
Neg logit contributions
ochond-5.958
 pollut-5.617
opsis-5.581
OTOS-5.577
 axes-5.554
Token gray
Token ID12768
Feature activation+3.51e-01

Pos logit contributions
util+7.684
onen+7.004
cone+6.901
idium+6.756
hler+6.721
Neg logit contributions
ochond-6.919
escription-6.547
OTOS-6.514
opsis-6.291
 pollut-6.125
Token =
Token ID796
Feature activation+2.45e-01

Pos logit contributions
util+3.592
onen+3.459
idium+3.349
hler+3.283
cone+3.268
Neg logit contributions
ochond-3.345
 axes-3.316
OTOS-3.198
 acceler-3.198
 pollut-3.195
Token c
Token ID269
Feature activation+2.44e-01

Pos logit contributions
util+2.484
onen+2.372
idium+2.328
hler+2.255
cone+2.252
Neg logit contributions
ochond-2.407
 axes-2.314
OTOS-2.283
opsis-2.274
escription-2.267
Tokenv
Token ID85
Feature activation+4.22e-01

Pos logit contributions
util+1.968
onen+1.702
hler+1.629
idium+1.622
cone+1.613
Neg logit contributions
escription-3.073
OTOS-3.046
 axes-3.042
 pollut-2.974
ochond-2.973
Token2
Token ID17
Feature activation+4.66e-01

Pos logit contributions
util+3.138
onen+2.600
hler+2.541
idium+2.536
cone+2.471
Neg logit contributions
escription-5.793
 pollut-5.438
 axes-5.359
ochond-5.351
 subsequ-5.281
Tokenparse
Token ID29572
Feature activation+4.39e-01

Pos logit contributions
util+2.553
onen+2.342
idium+2.231
hler+2.204
cone+2.183
Neg logit contributions
 axes-3.641
ochond-3.635
 pollut-3.575
escription-3.575
OTOS-3.560
Token.
Token ID13
Feature activation+1.79e-01

Pos logit contributions
util+4.555
onen+4.282
idium+4.146
hler+4.121
cone+3.960
Neg logit contributions
ochond-4.191
 axes-4.100
OTOS-3.990
 pollut-3.969
escription-3.928
TokenArg
Token ID28100
Feature activation+5.73e-02

Pos logit contributions
util+1.782
onen+1.713
idium+1.661
hler+1.614
cone+1.570
Neg logit contributions
ochond-1.801
 axes-1.783
 pollut-1.731
OTOS-1.718
opsis-1.706
Tokenument
Token ID1713
Feature activation-3.45e-02

Pos logit contributions
util+0.625
onen+0.613
idium+0.601
hler+0.584
cone+0.571
Neg logit contributions
 axes-0.510
ochond-0.507
opsis-0.481
OTOS-0.481
 pollut-0.478
TokenParser
Token ID46677
Feature activation+4.94e-01

Pos logit contributions
 axes+0.287
ochond+0.287
opsis+0.268
OTOS+0.268
 acceler+0.264
Neg logit contributions
util-0.396
onen-0.392
idium-0.384
hler-0.373
cone-0.366
Token()
Token ID3419
Feature activation+7.60e-01

Pos logit contributions
util+5.108
onen+4.805
hler+4.720
idium+4.719
cone+4.487
Neg logit contributions
ochond-4.593
 axes-4.505
 pollut-4.390
escription-4.359
OTOS-4.347
Token ap
Token ID2471
Feature activation+4.58e-01

Pos logit contributions
util+8.331
onen+7.663
hler+7.625
cone+7.594
idium+7.472
Neg logit contributions
ochond-6.075
escription-5.864
OTOS-5.779
opsis-5.451
 pollut-5.387
Token.
Token ID13
Feature activation+8.76e-02

Pos logit contributions
util+4.454
onen+4.123
idium+4.026
hler+4.003
cone+3.873
Neg logit contributions
ochond-4.606
 axes-4.574
 pollut-4.492
OTOS-4.431
escription-4.381
Tokenadd
Token ID2860
Feature activation+2.96e-01

Pos logit contributions
util+0.819
onen+0.788
idium+0.763
hler+0.741
cone+0.727
Neg logit contributions
ochond-0.931
 axes-0.928
OTOS-0.897
 pollut-0.893
 acceler-0.891
Token_
Token ID62
Feature activation+2.82e-02

Pos logit contributions
util+2.893
onen+2.761
idium+2.635
hler+2.596
cone+2.581
Neg logit contributions
ochond-3.017
 axes-2.981
 pollut-2.926
 acceler-2.904
OTOS-2.880
Tokenargument
Token ID49140
Feature activation+7.39e-02

Pos logit contributions
util+0.313
onen+0.302
idium+0.295
hler+0.288
cone+0.284
Neg logit contributions
ochond-0.250
 axes-0.250
OTOS-0.239
 acceler-0.237
 pollut-0.237
Token.
Token ID13
Feature activation-4.80e-02

Pos logit contributions
util+1.531
onen+1.489
idium+1.463
hler+1.424
cone+1.386
Neg logit contributions
ochond-1.309
 axes-1.306
OTOS-1.247
opsis-1.236
 pollut-1.230
Token
Token ID198
Feature activation+1.21e-01

Pos logit contributions
 axes+0.449
ochond+0.447
OTOS+0.426
opsis+0.425
 pollut+0.424
Neg logit contributions
util-0.503
onen-0.491
idium-0.482
hler-0.467
cone-0.457
Token
Token ID198
Feature activation+1.52e-01

Pos logit contributions
util+1.138
onen+1.077
idium+1.028
hler+1.019
cone+0.989
Neg logit contributions
escription-1.348
ochond-1.286
 axes-1.279
 pollut-1.246
opsis-1.221
TokenJ
Token ID41
Feature activation+3.74e-01

Pos logit contributions
util+1.363
onen+1.311
idium+1.273
hler+1.223
cone+1.204
Neg logit contributions
 axes-1.671
ochond-1.660
 acceler-1.597
escription-1.587
 pollut-1.584
Tokenag
Token ID363
Feature activation-2.19e-01

Pos logit contributions
util+4.046
onen+3.926
idium+3.827
hler+3.700
cone+3.595
Neg logit contributions
 axes-3.351
ochond-3.345
 acceler-3.197
OTOS-3.171
 pollut-3.168
Tokend
Token ID67
Feature activation+7.57e-01

Pos logit contributions
 axes+2.268
ochond+2.246
opsis+2.167
 pollut+2.126
 acceler+2.123
Neg logit contributions
util-2.015
onen-1.990
idium-1.941
hler-1.902
cone-1.829
Tokenish
Token ID680
Feature activation+1.26e-01

Pos logit contributions
util+7.109
onen+6.817
idium+6.512
unia+6.453
hler+6.389
Neg logit contributions
 axes-7.240
ochond-7.195
 pollut-7.118
OTOS-7.025
 acceler-6.994
Token Sh
Token ID911
Feature activation+1.45e-01

Pos logit contributions
util+1.273
onen+1.241
idium+1.209
hler+1.170
cone+1.146
Neg logit contributions
 axes-1.226
ochond-1.225
 acceler-1.164
OTOS-1.163
 pollut-1.158
Tokenind
Token ID521
Feature activation-8.46e-03

Pos logit contributions
util+1.499
onen+1.470
idium+1.433
hler+1.384
cone+1.351
Neg logit contributions
 axes-1.394
ochond-1.358
 acceler-1.323
 pollut-1.319
OTOS-1.311
Tokene
Token ID68
Feature activation+9.33e-02

Pos logit contributions
 axes+0.080
ochond+0.080
OTOS+0.076
opsis+0.076
 pollut+0.076
Neg logit contributions
util-0.088
onen-0.086
idium-0.084
hler-0.082
cone-0.080
Token Senior
Token ID14017
Feature activation-1.47e-02

Pos logit contributions
util+0.973
onen+0.953
idium+0.930
hler+0.906
cone+0.882
Neg logit contributions
 axes-0.879
ochond-0.876
OTOS-0.834
opsis-0.830
 acceler-0.828
Token,
Token ID11
Feature activation+1.63e-03

Pos logit contributions
util+2.589
onen+2.530
idium+2.466
hler+2.368
cone+2.352
Neg logit contributions
ochond-2.344
 axes-2.300
OTOS-2.251
escription-2.227
opsis-2.215
Token one
Token ID530
Feature activation+1.66e-01

Pos logit contributions
util+0.018
onen+0.017
idium+0.017
hler+0.017
cone+0.016
Neg logit contributions
ochond-0.014
 axes-0.014
OTOS-0.014
opsis-0.014
 pollut-0.014
Token might
Token ID1244
Feature activation+2.15e-01

Pos logit contributions
util+1.822
onen+1.772
idium+1.729
hler+1.684
cone+1.682
Neg logit contributions
ochond-1.471
 axes-1.419
escription-1.411
OTOS-1.404
opsis-1.393
Token ask
Token ID1265
Feature activation+1.28e-01

Pos logit contributions
util+2.279
onen+2.193
idium+2.147
hler+2.113
cone+2.056
Neg logit contributions
ochond-1.991
OTOS-1.970
 axes-1.967
escription-1.901
opsis-1.882
Token,
Token ID11
Feature activation+9.20e-02

Pos logit contributions
util+1.350
onen+1.314
idium+1.289
hler+1.246
cone+1.223
Neg logit contributions
ochond-1.190
 axes-1.187
OTOS-1.131
opsis-1.129
 pollut-1.124
Token why
Token ID1521
Feature activation+7.54e-01

Pos logit contributions
util+0.942
onen+0.916
idium+0.896
hler+0.865
cone+0.850
Neg logit contributions
ochond-0.886
 axes-0.879
OTOS-0.844
opsis-0.840
escription-0.834
Token go
Token ID467
Feature activation+7.32e-02

Pos logit contributions
util+7.367
onen+7.008
idium+6.795
cone+6.644
hler+6.384
Neg logit contributions
escription-6.902
ochond-6.866
OTOS-6.804
opsis-6.655
 axes-6.571
Token that
Token ID326
Feature activation+9.93e-02

Pos logit contributions
util+0.797
onen+0.780
idium+0.764
hler+0.741
cone+0.729
Neg logit contributions
ochond-0.653
 axes-0.649
OTOS-0.621
opsis-0.620
 pollut-0.616
Token one
Token ID530
Feature activation-1.12e-01

Pos logit contributions
util+1.222
onen+1.198
idium+1.173
hler+1.148
cone+1.134
Neg logit contributions
ochond-0.743
 axes-0.736
opsis-0.702
OTOS-0.697
 pollut-0.690
Token small
Token ID1402
Feature activation+2.18e-01

Pos logit contributions
 axes+0.878
ochond+0.867
OTOS+0.817
 acceler+0.816
 pollut+0.816
Neg logit contributions
util-1.348
onen-1.317
idium-1.300
hler-1.261
cone-1.234
Token
Token ID198
Feature activation+2.64e-01

Pos logit contributions
util+2.359
onen+2.317
idium+2.238
cone+2.192
hler+2.189
Neg logit contributions
ochond-1.936
 axes-1.893
opsis-1.852
OTOS-1.833
escription-1.804
Token 43
Token ID5946
Feature activation-4.25e-01

Pos logit contributions
 axes+1.675
ochond+1.654
 acceler+1.586
opsis+1.579
 pollut+1.577
Neg logit contributions
util-1.829
onen-1.791
idium-1.759
hler-1.700
cone-1.673
Token (
Token ID357
Feature activation+4.64e-03

Pos logit contributions
 axes+3.170
ochond+3.027
opsis+3.000
 acceler+2.920
escription+2.904
Neg logit contributions
util-4.965
onen-4.901
idium-4.828
hler-4.757
cone-4.749
TokenS
Token ID50
Feature activation-9.64e-02

Pos logit contributions
util+0.042
onen+0.041
idium+0.040
hler+0.039
cone+0.038
Neg logit contributions
 axes-0.050
ochond-0.050
OTOS-0.047
opsis-0.047
 pollut-0.047
Token4
Token ID19
Feature activation-2.52e-01

Pos logit contributions
 axes+1.013
ochond+1.011
OTOS+0.967
opsis+0.965
 pollut+0.962
Neg logit contributions
util-0.898
onen-0.875
idium-0.854
hler-0.825
cone-0.807
Token)
Token ID8
Feature activation-4.50e-01

Pos logit contributions
 axes+2.318
ochond+2.279
opsis+2.172
 acceler+2.165
 pollut+2.156
Neg logit contributions
util-2.678
onen-2.620
idium-2.551
hler-2.449
ilogy-2.431
Token DOI
Token ID40722
Feature activation-9.73e-01

Pos logit contributions
 axes+4.118
ochond+4.011
 acceler+3.907
OTOS+3.837
opsis+3.837
Neg logit contributions
util-4.570
onen-4.484
idium-4.359
cone-4.267
hler-4.222
Token:
Token ID25
Feature activation-4.09e-01

Pos logit contributions
 axes+5.850
 acceler+5.711
OTOS+5.586
ochond+5.546
 pollut+5.537
Neg logit contributions
util-11.592
irtual-11.261
idium-11.261
onen-11.206
avorite-11.143
Token 10
Token ID838
Feature activation-6.46e-01

Pos logit contributions
 axes+3.046
ochond+2.947
 acceler+2.850
 pollut+2.847
OTOS+2.831
Neg logit contributions
util-4.920
onen-4.779
hler-4.775
idium-4.742
cone-4.649
Token.
Token ID13
Feature activation-9.54e-02

Pos logit contributions
 axes+5.030
ochond+5.009
OTOS+4.745
 acceler+4.734
opsis+4.710
Neg logit contributions
onen-7.314
util-7.306
avorite-7.221
idium-7.211
irtual-6.996
Token10
Token ID940
Feature activation-3.07e-01

Pos logit contributions
 axes+1.103
ochond+1.098
opsis+1.055
OTOS+1.055
 acceler+1.052
Neg logit contributions
util-0.784
onen-0.765
idium-0.748
hler-0.715
cone-0.700
Token86
Token ID4521
Feature activation-9.93e-02

Pos logit contributions
 axes+3.055
ochond+3.032
 acceler+2.903
 pollut+2.869
opsis+2.863
Neg logit contributions
util-3.022
onen-2.987
idium-2.894
hler-2.797
avorite-2.750
Token and
Token ID290
Feature activation+5.83e-02

Pos logit contributions
 axes+0.478
ochond+0.475
OTOS+0.451
 pollut+0.451
 acceler+0.450
Neg logit contributions
util-0.581
onen-0.568
idium-0.557
hler-0.540
cone-0.530
Token policy
Token ID2450
Feature activation+2.99e-02

Pos logit contributions
util+0.673
onen+0.657
idium+0.645
hler+0.627
cone+0.615
Neg logit contributions
ochond-0.484
 axes-0.477
OTOS-0.456
opsis-0.453
escription-0.449
Token
Token ID198
Feature activation-1.56e-01

Pos logit contributions
util+0.321
onen+0.314
idium+0.307
hler+0.299
cone+0.292
Neg logit contributions
ochond-0.272
 axes-0.271
OTOS-0.258
opsis-0.257
 pollut-0.256
Token
Token ID198
Feature activation-6.61e-02

Pos logit contributions
 axes+1.452
ochond+1.449
escription+1.383
 pollut+1.377
opsis+1.374
Neg logit contributions
util-1.663
onen-1.618
idium-1.580
hler-1.538
cone-1.506
TokenAmericans
Token ID17636
Feature activation-1.60e-01

Pos logit contributions
 axes+0.698
ochond+0.695
OTOS+0.669
opsis+0.665
 pollut+0.663
Neg logit contributions
util-0.611
onen-0.596
idium-0.583
hler-0.562
cone-0.548
Token for
Token ID329
Feature activation-9.38e-01

Pos logit contributions
 axes+1.416
ochond+1.402
 pollut+1.333
 acceler+1.331
OTOS+1.330
Neg logit contributions
util-1.763
onen-1.729
idium-1.690
hler-1.640
cone-1.609
Token Prosper
Token ID41342
Feature activation-3.87e-01

Pos logit contributions
 pollut+5.076
 axes+5.052
ochond+4.893
 DAC+4.765
 govern+4.598
Neg logit contributions
onen-12.534
avorite-12.457
ilogy-12.165
util-11.854
idium-11.812
Tokenity
Token ID414
Feature activation-2.46e-01

Pos logit contributions
 axes+3.999
ochond+3.969
OTOS+3.821
 pollut+3.813
 acceler+3.808
Neg logit contributions
util-3.618
onen-3.579
idium-3.462
hler-3.347
avorite-3.304
Token:
Token ID25
Feature activation-1.01e-01

Pos logit contributions
 axes+2.122
ochond+2.053
 acceler+2.013
 pollut+1.987
OTOS+1.967
Neg logit contributions
util-2.769
onen-2.720
idium-2.650
hler-2.573
cone-2.544
Token Dist
Token ID4307
Feature activation-3.79e-01

Pos logit contributions
 axes+0.950
ochond+0.942
OTOS+0.900
 pollut+0.897
 acceler+0.896
Neg logit contributions
util-1.054
onen-1.030
idium-1.009
hler-0.977
cone-0.957
Tokenorting
Token ID24707
Feature activation-8.39e-02

Pos logit contributions
ochond+4.457
 axes+4.407
opsis+4.284
 pollut+4.268
escription+4.163
Neg logit contributions
util-2.864
onen-2.845
idium-2.779
hler-2.678
cone-2.655
Token
Token ID198
Feature activation-3.37e-01

Pos logit contributions
 axes+3.386
ochond+3.348
OTOS+3.202
 acceler+3.187
opsis+3.165
Neg logit contributions
util-4.435
onen-4.357
idium-4.282
hler-4.152
cone-4.078
Token
Token ID198
Feature activation-4.42e-01

Pos logit contributions
 axes+2.835
ochond+2.823
OTOS+2.685
opsis+2.668
 acceler+2.663
Neg logit contributions
util-3.772
onen-3.715
idium-3.680
avorite-3.547
hler-3.546
TokenFacebook
Token ID12025
Feature activation-6.06e-01

Pos logit contributions
 axes+2.676
ochond+2.536
 acceler+2.446
opsis+2.367
OTOS+2.331
Neg logit contributions
idium-5.796
util-5.778
irtual-5.751
onen-5.666
hler-5.536
Token
Token ID198
Feature activation-1.74e-01

Pos logit contributions
 axes+4.428
ochond+4.397
OTOS+4.279
 acceler+4.221
opsis+4.191
Neg logit contributions
util-7.242
idium-7.137
avorite-7.053
onen-7.017
irtual-6.866
Token
Token ID198
Feature activation-3.60e-01

Pos logit contributions
 axes+1.614
ochond+1.611
escription+1.538
 pollut+1.532
OTOS+1.526
Neg logit contributions
util-1.845
onen-1.798
idium-1.753
hler-1.708
cone-1.671
TokenSt
Token ID1273
Feature activation-8.94e-01

Pos logit contributions
 axes+3.488
ochond+3.428
opsis+3.299
OTOS+3.298
 acceler+3.255
Neg logit contributions
util-3.583
idium-3.480
onen-3.446
hler-3.320
irtual-3.264
Tokenumble
Token ID10344
Feature activation-4.16e-01

Pos logit contributions
 axes+3.154
ait+3.026
yn+2.905
ochond+2.877
opsis+2.847
Neg logit contributions
��-13.331
util-13.057
hler-13.035
cone-13.015
avorite-12.918
TokenUpon
Token ID23792
Feature activation-3.34e-01

Pos logit contributions
 axes+3.139
ochond+3.043
opsis+2.998
 acceler+2.977
OTOS+2.911
Neg logit contributions
util-4.981
onen-4.887
hler-4.765
idium-4.760
avorite-4.665
Token What
Token ID1867
Feature activation-4.07e-01

Pos logit contributions
 axes+2.664
ochond+2.569
 acceler+2.485
 DAC+2.456
opsis+2.455
Neg logit contributions
util-3.977
onen-3.844
idium-3.813
cone-3.740
hler-3.726
Token are
Token ID389
Feature activation-1.90e-01

Pos logit contributions
 axes+2.963
opsis+2.844
OTOS+2.771
 acceler+2.709
 pollut+2.705
Neg logit contributions
util-4.956
onen-4.947
idium-4.861
hler-4.842
ussen-4.697
Token these
Token ID777
Feature activation-4.03e-01

Pos logit contributions
 axes+1.444
ochond+1.425
 acceler+1.363
 pollut+1.343
OTOS+1.342
Neg logit contributions
util-2.339
idium-2.284
onen-2.269
hler-2.202
cone-2.177
Token'
Token ID6
Feature activation-1.44e-02

Pos logit contributions
util+1.062
idium+1.037
hler+0.982
cone+0.976
onen+0.971
Neg logit contributions
ochond-1.293
OTOS-1.264
 axes-1.234
escription-1.219
opsis-1.198
Token 2016
Token ID1584
Feature activation-2.67e-01

Pos logit contributions
 axes+0.117
ochond+0.110
 pollut+0.110
 acceler+0.109
OTOS+0.104
Neg logit contributions
onen-0.170
util-0.169
idium-0.166
avorite-0.160
hler-0.158
Token primary
Token ID4165
Feature activation-3.57e-01

Pos logit contributions
 axes+2.123
 pollut+2.047
 acceler+2.015
 pickup+1.952
opsis+1.940
Neg logit contributions
util-3.198
onen-3.093
idium-3.041
avorite-2.998
hler-2.977
Token Sanders
Token ID5831
Feature activation-1.08e-01

Pos logit contributions
 axes+3.379
 acceler+3.200
opsis+3.123
 pollut+3.114
ochond+3.054
Neg logit contributions
util-3.775
onen-3.701
idium-3.566
irtual-3.485
hler-3.426
Token:
Token ID25
Feature activation-3.75e-01

Pos logit contributions
 axes+0.843
ochond+0.799
 acceler+0.784
 pollut+0.778
OTOS+0.763
Neg logit contributions
util-1.323
onen-1.295
idium-1.266
hler-1.245
avorite-1.221
Token '
Token ID705
Feature activation-8.41e-01

Pos logit contributions
 axes+2.164
 acceler+2.096
opsis+1.964
ochond+1.945
 pollut+1.919
Neg logit contributions
util-5.182
onen-5.069
cone-4.979
avorite-4.973
idium-4.940
TokenDamn
Token ID43343
Feature activation-6.50e-02

Pos logit contributions
 axes+6.427
 acceler+5.624
yn+5.490
ait+5.397
ochond+5.397
Neg logit contributions
avorite-10.239
util-9.628
onen-9.516
unia-9.467
idium-9.199
Token right
Token ID826
Feature activation-1.23e-01

Pos logit contributions
 axes+0.570
ochond+0.562
 pollut+0.527
 acceler+0.527
opsis+0.525
Neg logit contributions
util-0.728
onen-0.711
idium-0.702
hler-0.683
cone-0.672
Token'
Token ID6
Feature activation-2.71e-01

Pos logit contributions
 axes+1.037
ochond+1.022
 acceler+0.973
 pollut+0.972
opsis+0.968
Neg logit contributions
util-1.411
onen-1.365
idium-1.352
hler-1.308
ilogy-1.287
Token I
Token ID314
Feature activation-4.98e-01

Pos logit contributions
 axes+1.992
 acceler+1.985
OTOS+1.940
ochond+1.927
 pollut+1.851
Neg logit contributions
util-3.319
onen-3.288
idium-3.232
avorite-3.202
cone-3.120
Token'll
Token ID1183
Feature activation-3.94e-01

Pos logit contributions
 axes+3.971
ochond+3.765
 acceler+3.621
 pollut+3.591
 govern+3.547
Neg logit contributions
onen-5.944
util-5.680
irtual-5.613
idium-5.589
perties-5.482
Token Trump
Token ID1301
Feature activation-4.58e-01

Pos logit contributions
 axes+3.952
ochond+3.768
OTOS+3.686
 acceler+3.617
opsis+3.517
Neg logit contributions
util-9.402
idium-9.295
onen-9.190
avorite-9.173
ilogy-9.031
TokenHouse
Token ID18102
Feature activation-5.03e-01

Pos logit contributions
 axes+4.210
ochond+4.175
OTOS+4.092
opsis+3.943
 acceler+3.935
Neg logit contributions
onen-4.600
util-4.587
avorite-4.441
idium-4.399
ilogy-4.358
Token committee
Token ID5583
Feature activation-3.72e-01

Pos logit contributions
 axes+2.979
ochond+2.768
 acceler+2.708
OTOS+2.683
opsis+2.658
Neg logit contributions
util-6.915
onen-6.762
illance-6.598
irtual-6.525
hler-6.464
Token believes
Token ID5804
Feature activation-2.63e-01

Pos logit contributions
 axes+3.289
 acceler+3.273
ochond+3.183
OTOS+3.129
opsis+3.075
Neg logit contributions
util-4.052
onen-3.951
hler-3.836
idium-3.803
ilogy-3.731
Token it
Token ID340
Feature activation-7.11e-01

Pos logit contributions
 axes+2.270
ochond+2.231
 acceler+2.201
OTOS+2.116
opsis+2.111
Neg logit contributions
util-2.957
onen-2.859
idium-2.812
hler-2.784
ilogy-2.737
Token has
Token ID468
Feature activation-8.35e-01

Pos logit contributions
 acceler+4.119
 axes+3.848
 smokes+3.665
ochond+3.422
 drives+3.348
Neg logit contributions
onen-10.024
avorite-9.812
hler-9.700
ilogy-9.616
ussen-9.442
Token evidence
Token ID2370
Feature activation-3.40e-01

Pos logit contributions
 axes+4.370
 acceler+4.114
ochond+4.084
 pollut+3.839
 fung+3.791
Neg logit contributions
hler-11.931
onen-11.693
cone-11.359
ilogy-11.258
perties-11.250
Token Trump
Token ID1301
Feature activation-1.83e-01

Pos logit contributions
 axes+2.709
ochond+2.672
 acceler+2.574
 pollut+2.535
OTOS+2.500
Neg logit contributions
util-4.041
onen-3.973
idium-3.874
hler-3.800
irtual-3.712
Token requested
Token ID9167
Feature activation-2.21e-01

Pos logit contributions
 axes+1.525
ochond+1.437
 acceler+1.417
 pollut+1.405
OTOS+1.381
Neg logit contributions
util-2.132
onen-2.092
idium-2.026
irtual-1.991
avorite-1.985
Token putting
Token ID5137
Feature activation-3.15e-01

Pos logit contributions
 axes+1.736
 acceler+1.664
OTOS+1.598
ochond+1.593
 pollut+1.577
Neg logit contributions
util-2.638
onen-2.636
idium-2.633
hler-2.564
unia-2.496
Token ally
Token ID12525
Feature activation-2.91e-01

Pos logit contributions
 axes+2.627
ochond+2.511
OTOS+2.502
 acceler+2.466
 pollut+2.421
Neg logit contributions
util-3.651
onen-3.581
idium-3.483
ilogy-3.354
hler-3.330
Token
Token ID198
Feature activation-8.66e-02

Pos logit contributions
 axes+1.562
ochond+1.535
 pollut+1.459
opsis+1.457
 acceler+1.451
Neg logit contributions
util-2.055
onen-2.016
idium-1.969
hler-1.923
avorite-1.878
Token
Token ID198
Feature activation-3.30e-01

Pos logit contributions
 axes+0.840
ochond+0.840
escription+0.829
 pollut+0.802
opsis+0.797
Neg logit contributions
util-0.894
onen-0.863
idium-0.838
hler-0.818
cone-0.803
Tokend
Token ID67
Feature activation+3.54e-02

Pos logit contributions
 axes+3.373
ochond+3.360
opsis+3.211
 pollut+3.208
OTOS+3.208
Neg logit contributions
util-3.036
onen-3.008
idium-2.927
hler-2.842
avorite-2.796
Token[
Token ID58
Feature activation-4.06e-01

Pos logit contributions
util+0.353
onen+0.342
idium+0.334
hler+0.323
cone+0.316
Neg logit contributions
 axes-0.350
ochond-0.350
OTOS-0.335
 pollut-0.334
opsis-0.333
TokenCO
Token ID8220
Feature activation+4.36e-02

Pos logit contributions
 axes+4.030
ochond+3.979
 pollut+3.826
opsis+3.817
OTOS+3.793
Neg logit contributions
util-3.834
onen-3.783
idium-3.668
hler-3.565
avorite-3.550
Token 2
Token ID362
Feature activation-8.32e-01

Pos logit contributions
util+0.439
onen+0.428
idium+0.419
hler+0.406
cone+0.397
Neg logit contributions
 axes-0.426
ochond-0.425
OTOS-0.405
opsis-0.404
 pollut-0.403
Token ]
Token ID2361
Feature activation-5.33e-01

Pos logit contributions
 axes+6.510
 pollut+6.389
ochond+6.337
 acceler+6.293
 DAC+5.955
Neg logit contributions
onen-8.888
util-8.626
avorite-8.440
ilogy-8.327
idium-8.296
Token/
Token ID14
Feature activation-4.08e-01

Pos logit contributions
 axes+4.988
ochond+4.923
 pollut+4.723
 acceler+4.711
OTOS+4.672
Neg logit contributions
onen-5.242
util-5.240
idium-5.027
hler-4.923
avorite-4.897
Tokendt
Token ID28664
Feature activation-4.43e-01

Pos logit contributions
 axes+3.989
ochond+3.965
 pollut+3.811
opsis+3.783
OTOS+3.772
Neg logit contributions
util-3.864
onen-3.853
idium-3.757
hler-3.598
avorite-3.585
Token =
Token ID796
Feature activation-2.39e-01

Pos logit contributions
 axes+4.061
ochond+3.988
 pollut+3.847
 acceler+3.834
opsis+3.813
Neg logit contributions
util-4.526
onen-4.520
idium-4.420
hler-4.294
avorite-4.191
Token f
Token ID277
Feature activation-5.37e-02

Pos logit contributions
 axes+2.243
ochond+2.208
 pollut+2.123
 acceler+2.113
OTOS+2.094
Neg logit contributions
util-2.491
onen-2.444
idium-2.381
hler-2.320
avorite-2.262
Token 2011
Token ID2813
Feature activation-2.18e-02

Pos logit contributions
 axes+5.137
ochond+4.618
 acceler+4.453
opsis+4.432
 DAC+4.392
Neg logit contributions
util-9.560
idium-9.354
onen-9.234
ilogy-8.992
irtual-8.923
Token 6
Token ID718
Feature activation+1.48e-01

Pos logit contributions
ochond+0.210
 axes+0.210
escription+0.200
opsis+0.200
OTOS+0.199
Neg logit contributions
util-0.223
onen-0.217
idium-0.212
hler-0.206
cone-0.201
Token 2
Token ID362
Feature activation-5.20e-01

Pos logit contributions
util+1.546
onen+1.506
idium+1.463
hler+1.427
cone+1.400
Neg logit contributions
ochond-1.395
 axes-1.385
opsis-1.323
escription-1.319
 pollut-1.319
Token 2012
Token ID2321
Feature activation+5.94e-02

Pos logit contributions
 axes+3.617
ochond+3.357
 pollut+3.161
opsis+3.123
 acceler+3.077
Neg logit contributions
onen-6.526
util-6.490
idium-6.307
hler-6.230
cone-6.038
Token 7
Token ID767
Feature activation+9.12e-03

Pos logit contributions
util+0.604
onen+0.586
idium+0.572
hler+0.556
cone+0.543
Neg logit contributions
ochond-0.577
 axes-0.576
escription-0.554
opsis-0.549
 pollut-0.547
Token 0
Token ID657
Feature activation-8.27e-01

Pos logit contributions
util+0.099
onen+0.096
idium+0.095
hler+0.092
cone+0.090
Neg logit contributions
 axes-0.082
ochond-0.082
OTOS-0.078
opsis-0.078
 pollut-0.077
Token 2013
Token ID2211
Feature activation-5.80e-02

Pos logit contributions
ochond+5.011
 axes+5.011
 pollut+4.943
escription+4.923
OTOS+4.786
Neg logit contributions
util-9.904
idium-9.814
onen-9.542
unia-9.344
cone-9.336
Token 3
Token ID513
Feature activation+9.21e-02

Pos logit contributions
 axes+0.541
ochond+0.540
OTOS+0.513
opsis+0.512
 pollut+0.511
Neg logit contributions
util-0.612
onen-0.597
idium-0.584
hler-0.567
cone-0.555
Token 0
Token ID657
Feature activation-5.32e-01

Pos logit contributions
util+0.976
onen+0.949
idium+0.926
hler+0.903
cone+0.885
Neg logit contributions
ochond-0.855
 axes-0.852
opsis-0.810
OTOS-0.808
 pollut-0.808
Token 2014
Token ID1946
Feature activation-1.57e-02

Pos logit contributions
 axes+4.659
ochond+4.617
OTOS+4.533
 acceler+4.421
 pollut+4.414
Neg logit contributions
util-5.568
onen-5.302
hler-5.211
idium-5.198
illance-5.014
Token 8
Token ID807
Feature activation-1.05e-03

Pos logit contributions
 axes+0.146
ochond+0.146
opsis+0.139
OTOS+0.139
 pollut+0.138
Neg logit contributions
util-0.165
onen-0.161
idium-0.157
hler-0.153
cone-0.150
Token the
Token ID262
Feature activation-3.43e-01

Pos logit contributions
 axes+3.718
ochond+3.676
 acceler+3.475
opsis+3.463
OTOS+3.442
Neg logit contributions
util-5.522
onen-5.434
idium-5.389
hler-5.210
avorite-5.181
Token argument
Token ID4578
Feature activation-3.93e-01

Pos logit contributions
 axes+2.367
ochond+2.344
opsis+2.186
 pollut+2.183
 acceler+2.180
Neg logit contributions
util-4.431
onen-4.348
idium-4.283
hler-4.174
cone-4.109
Token parse
Token ID21136
Feature activation-3.82e-01

Pos logit contributions
 axes+2.743
ochond+2.617
opsis+2.505
 pollut+2.490
 acceler+2.489
Neg logit contributions
util-5.007
onen-5.002
idium-4.928
hler-4.792
avorite-4.731
Token and
Token ID290
Feature activation-5.56e-01

Pos logit contributions
 axes+3.123
ochond+3.098
OTOS+2.921
 acceler+2.921
opsis+2.916
Neg logit contributions
util-4.428
onen-4.344
idium-4.298
hler-4.152
avorite-4.094
Token parse
Token ID21136
Feature activation-3.48e-01

Pos logit contributions
 axes+3.628
ochond+3.433
opsis+3.376
 acceler+3.302
 pollut+3.283
Neg logit contributions
util-7.243
onen-7.206
idium-7.169
hler-6.937
avorite-6.888
Token the
Token ID262
Feature activation-8.24e-01

Pos logit contributions
 axes+2.918
ochond+2.901
OTOS+2.741
opsis+2.736
 acceler+2.730
Neg logit contributions
util-3.977
onen-3.898
idium-3.829
hler-3.718
cone-3.649
Token arguments
Token ID7159
Feature activation-2.33e-01

Pos logit contributions
 axes+5.031
 acceler+4.273
ochond+4.263
 offsets+4.164
OTOS+4.064
Neg logit contributions
onen-11.193
util-11.162
idium-10.964
avorite-10.791
irtual-10.755
Token ap
Token ID2471
Feature activation+2.87e-01

Pos logit contributions
 axes+1.150
ochond+1.035
 pollut+0.997
 acceler+0.989
opsis+0.955
Neg logit contributions
onen-3.492
util-3.486
idium-3.479
hler-3.357
irtual-3.285
Token =
Token ID796
Feature activation-7.71e-02

Pos logit contributions
util+2.689
onen+2.492
hler+2.446
idium+2.415
kov+2.297
Neg logit contributions
escription-3.209
 pollut-3.022
 axes-3.014
ochond-2.983
 acceler-2.892
Token arg
Token ID1822
Feature activation-2.05e-01

Pos logit contributions
ochond+0.589
 axes+0.582
OTOS+0.559
escription+0.558
opsis+0.550
Neg logit contributions
util-0.945
onen-0.920
idium-0.904
hler-0.884
cone-0.869
Tokenparse
Token ID29572
Feature activation+1.23e-01

Pos logit contributions
 axes+2.232
ochond+2.221
OTOS+2.135
 pollut+2.129
opsis+2.124
Neg logit contributions
util-1.848
onen-1.790
idium-1.747
hler-1.687
cone-1.646
Token Trump
Token ID1301
Feature activation-4.66e-01

Pos logit contributions
 axes+3.872
ochond+3.704
OTOS+3.689
 acceler+3.506
opsis+3.484
Neg logit contributions
util-9.158
onen-9.017
idium-9.010
avorite-8.969
hler-8.835
TokenHouse
Token ID18102
Feature activation-5.22e-01

Pos logit contributions
 axes+4.196
ochond+4.164
OTOS+4.106
opsis+3.927
 acceler+3.872
Neg logit contributions
onen-4.726
util-4.682
avorite-4.610
ilogy-4.526
idium-4.488
Token committee
Token ID5583
Feature activation-2.83e-01

Pos logit contributions
 axes+3.038
 acceler+2.750
ochond+2.746
OTOS+2.683
 pollut+2.613
Neg logit contributions
util-7.263
onen-7.083
illance-6.925
irtual-6.869
avorite-6.855
Token believes
Token ID5804
Feature activation-2.56e-01

Pos logit contributions
 axes+2.594
 acceler+2.540
ochond+2.505
OTOS+2.462
opsis+2.421
Neg logit contributions
util-3.023
onen-2.957
hler-2.834
idium-2.824
cone-2.773
Token it
Token ID340
Feature activation-7.10e-01

Pos logit contributions
 axes+2.263
ochond+2.230
 acceler+2.169
OTOS+2.116
 pollut+2.105
Neg logit contributions
util-2.838
onen-2.758
idium-2.711
hler-2.669
ilogy-2.613
Token has
Token ID468
Feature activation-8.22e-01

Pos logit contributions
 acceler+3.735
 axes+3.589
 smokes+3.234
ochond+3.173
 chops+2.932
Neg logit contributions
avorite-10.252
onen-10.194
hler-10.087
ilogy-9.738
irtual-9.733
Token evidence
Token ID2370
Feature activation-3.58e-01

Pos logit contributions
 axes+4.241
ochond+4.051
 acceler+3.999
 pollut+3.876
 fung+3.830
Neg logit contributions
hler-11.726
onen-11.474
cone-11.218
util-10.966
perties-10.927
Token Trump
Token ID1301
Feature activation-2.34e-01

Pos logit contributions
 axes+2.752
ochond+2.715
 acceler+2.607
 pollut+2.558
OTOS+2.536
Neg logit contributions
util-4.370
onen-4.280
idium-4.206
hler-4.092
irtual-4.040
Token requested
Token ID9167
Feature activation-2.47e-01

Pos logit contributions
 axes+1.903
ochond+1.770
 pollut+1.741
opsis+1.727
 acceler+1.723
Neg logit contributions
util-2.787
onen-2.729
irtual-2.624
idium-2.606
cone-2.579
Token putting
Token ID5137
Feature activation-3.13e-01

Pos logit contributions
 axes+1.740
 acceler+1.662
 pollut+1.562
ochond+1.561
OTOS+1.549
Neg logit contributions
idium-3.155
onen-3.153
util-3.147
hler-3.070
unia-3.004
Token ally
Token ID12525
Feature activation-2.82e-01

Pos logit contributions
 axes+2.678
ochond+2.582
OTOS+2.545
 acceler+2.504
 pollut+2.473
Neg logit contributions
util-3.558
onen-3.473
idium-3.394
hler-3.255
ilogy-3.247
Token s
Token ID264
Feature activation-1.60e-01

Pos logit contributions
 axes+4.238
OTOS+4.159
 acceler+4.110
 pollut+3.944
ochond+3.914
Neg logit contributions
idium-7.700
util-7.661
onen-7.647
ilogy-7.613
cone-7.568
Token.
Token ID13
Feature activation-3.57e-01

Pos logit contributions
 axes+1.527
ochond+1.522
 pollut+1.448
OTOS+1.447
escription+1.444
Neg logit contributions
util-1.670
onen-1.622
idium-1.587
hler-1.543
cone-1.508
Tokent
Token ID83
Feature activation-3.34e-01

Pos logit contributions
ochond+3.226
 axes+3.203
opsis+3.055
OTOS+3.038
 acceler+3.029
Neg logit contributions
util-3.744
onen-3.707
idium-3.667
avorite-3.548
hler-3.506
Token.
Token ID13
Feature activation-3.11e-01

Pos logit contributions
 axes+2.870
ochond+2.868
OTOS+2.739
opsis+2.726
 acceler+2.701
Neg logit contributions
util-3.666
onen-3.622
idium-3.585
avorite-3.501
hler-3.443
Token I
Token ID314
Feature activation+2.63e-02

Pos logit contributions
 axes+2.425
ochond+2.365
OTOS+2.332
 acceler+2.318
 pollut+2.296
Neg logit contributions
util-3.612
onen-3.595
idium-3.530
avorite-3.472
hler-3.429
TokenTA
Token ID5603
Feature activation-8.03e-01

Pos logit contributions
util+0.244
onen+0.231
idium+0.228
hler+0.222
cone+0.217
Neg logit contributions
ochond-0.283
 axes-0.283
escription-0.278
 pollut-0.273
opsis-0.267
Token 11
Token ID1367
Feature activation+1.20e-01

Pos logit contributions
 axes+4.769
OTOS+4.218
opsis+4.084
 pollut+4.038
 acceler+3.972
Neg logit contributions
idium-10.480
util-9.940
onen-9.833
ilogy-9.757
unia-9.607
Token Matte
Token ID38789
Feature activation-1.79e-01

Pos logit contributions
util+1.211
onen+1.181
idium+1.154
hler+1.119
cone+1.091
Neg logit contributions
ochond-1.170
 axes-1.168
 pollut-1.111
opsis-1.109
OTOS-1.106
Tokeno
Token ID78
Feature activation-2.09e-01

Pos logit contributions
 axes+1.823
ochond+1.819
OTOS+1.736
opsis+1.732
 pollut+1.731
Neg logit contributions
util-1.732
onen-1.687
idium-1.652
hler-1.597
cone-1.559
Token Trent
Token ID24269
Feature activation-5.76e-02

Pos logit contributions
 axes+1.583
ochond+1.554
opsis+1.485
 pollut+1.482
 acceler+1.475
Neg logit contributions
util-2.545
onen-2.496
idium-2.462
hler-2.392
cone-2.364
Tokenin
Token ID259
Feature activation-8.64e-02

Pos logit contributions
 axes+0.380
ochond+0.366
OTOS+0.357
 pollut+0.347
opsis+0.338
Neg logit contributions
util-0.782
onen-0.775
idium-0.742
ussen-0.729
cone-0.728
Token two
Token ID734
Feature activation-1.24e-01

Pos logit contributions
util+0.682
onen+0.667
idium+0.654
hler+0.632
cone+0.621
Neg logit contributions
ochond-0.599
 axes-0.595
OTOS-0.570
opsis-0.569
 pollut-0.564
Token quantum
Token ID14821
Feature activation+7.19e-02

Pos logit contributions
 axes+1.107
ochond+1.083
 acceler+1.039
 pollut+1.033
OTOS+1.024
Neg logit contributions
util-1.363
onen-1.330
idium-1.307
hler-1.267
cone-1.238
Token states
Token ID2585
Feature activation+7.13e-02

Pos logit contributions
util+0.794
onen+0.775
idium+0.760
hler+0.738
cone+0.724
Neg logit contributions
ochond-0.633
 axes-0.620
opsis-0.600
OTOS-0.597
escription-0.592
Token during
Token ID1141
Feature activation-6.38e-02

Pos logit contributions
util+0.742
onen+0.724
idium+0.709
hler+0.688
cone+0.675
Neg logit contributions
ochond-0.673
 axes-0.667
OTOS-0.642
opsis-0.640
 pollut-0.635
Token the
Token ID262
Feature activation+2.27e-02

Pos logit contributions
 axes+0.543
ochond+0.539
OTOS+0.510
 pollut+0.509
opsis+0.509
Neg logit contributions
util-0.725
onen-0.709
idium-0.697
hler-0.678
cone-0.664
Token pseud
Token ID25038
Feature activation-7.97e-01

Pos logit contributions
util+0.251
onen+0.245
idium+0.241
hler+0.233
cone+0.229
Neg logit contributions
ochond-0.199
 axes-0.198
OTOS-0.188
opsis-0.188
 pollut-0.186
Tokenog
Token ID519
Feature activation-1.48e-01

Pos logit contributions
ochond+5.490
opsis+4.918
 axes+4.774
OTOS+4.599
 pollut+4.572
Neg logit contributions
util-9.845
idium-9.599
onen-9.571
avorite-9.527
hler-9.321
Tokenap
Token ID499
Feature activation-1.43e-01

Pos logit contributions
 axes+1.754
ochond+1.728
OTOS+1.687
escription+1.680
 pollut+1.674
Neg logit contributions
util-1.215
onen-1.179
idium-1.126
hler-1.089
cone-1.075
Token,
Token ID11
Feature activation+8.92e-02

Pos logit contributions
 axes+1.286
ochond+1.274
 acceler+1.214
 pollut+1.209
OTOS+1.202
Neg logit contributions
util-1.554
onen-1.518
idium-1.490
hler-1.449
cone-1.410
Token one
Token ID530
Feature activation+5.81e-02

Pos logit contributions
util+0.930
onen+0.907
idium+0.886
hler+0.862
cone+0.846
Neg logit contributions
ochond-0.840
 axes-0.830
opsis-0.801
OTOS-0.800
escription-0.792
Token of
Token ID286
Feature activation-2.43e-02

Pos logit contributions
util+0.625
onen+0.612
idium+0.600
hler+0.580
cone+0.572
Neg logit contributions
ochond-0.525
 axes-0.521
OTOS-0.499
opsis-0.498
escription-0.493
Token
Token ID198
Feature activation-2.87e-01

Pos logit contributions
 axes+3.213
ochond+3.185
OTOS+3.042
 acceler+3.020
opsis+3.018
Neg logit contributions
util-4.086
onen-4.008
idium-3.932
hler-3.818
cone-3.748
Token
Token ID198
Feature activation-3.78e-01

Pos logit contributions
 axes+2.496
ochond+2.486
OTOS+2.363
opsis+2.352
 acceler+2.345
Neg logit contributions
util-3.170
onen-3.110
idium-3.067
hler-2.965
avorite-2.916
TokenFacebook
Token ID12025
Feature activation-5.77e-01

Pos logit contributions
 axes+2.466
ochond+2.347
 acceler+2.279
opsis+2.192
OTOS+2.184
Neg logit contributions
util-4.839
idium-4.815
onen-4.732
irtual-4.719
hler-4.616
Token
Token ID198
Feature activation-9.76e-02

Pos logit contributions
 axes+4.285
ochond+4.282
OTOS+4.179
 acceler+4.080
opsis+4.073
Neg logit contributions
util-6.829
idium-6.738
onen-6.652
avorite-6.643
hler-6.468
Token
Token ID198
Feature activation-3.40e-01

Pos logit contributions
 axes+0.947
ochond+0.946
escription+0.933
 pollut+0.907
opsis+0.896
Neg logit contributions
util-1.007
onen-0.975
idium-0.943
hler-0.924
cone-0.903
TokenSt
Token ID1273
Feature activation-7.96e-01

Pos logit contributions
 axes+3.312
ochond+3.264
OTOS+3.140
opsis+3.128
 acceler+3.102
Neg logit contributions
util-3.359
idium-3.259
onen-3.233
hler-3.117
irtual-3.065
Tokenumble
Token ID10344
Feature activation-3.52e-01

Pos logit contributions
 axes+3.162
ochond+2.981
ait+2.897
opsis+2.870
yn+2.851
Neg logit contributions
��-11.548
util-11.538
hler-11.450
onen-11.425
cone-11.425
TokenUpon
Token ID23792
Feature activation-2.66e-01

Pos logit contributions
 axes+2.839
ochond+2.788
opsis+2.696
 acceler+2.679
OTOS+2.657
Neg logit contributions
util-4.083
onen-3.998
idium-3.906
hler-3.869
avorite-3.757
Token What
Token ID1867
Feature activation-3.99e-01

Pos logit contributions
 axes+2.263
ochond+2.220
 acceler+2.135
opsis+2.134
 pollut+2.120
Neg logit contributions
util-3.020
onen-2.923
idium-2.905
hler-2.830
cone-2.793
Token are
Token ID389
Feature activation-1.39e-01

Pos logit contributions
 axes+2.973
opsis+2.846
OTOS+2.798
ochond+2.756
 pollut+2.748
Neg logit contributions
util-4.806
onen-4.769
idium-4.710
hler-4.662
cone-4.545
Token these
Token ID777
Feature activation-3.51e-01

Pos logit contributions
 axes+1.106
ochond+1.100
 acceler+1.036
OTOS+1.035
opsis+1.035
Neg logit contributions
util-1.663
onen-1.623
idium-1.608
hler-1.562
cone-1.536
Token Rodham
Token ID41848
Feature activation-3.81e-01

Pos logit contributions
ochond+1.457
OTOS+1.376
 axes+1.337
 acceler+1.268
 govern+1.230
Neg logit contributions
idium-7.581
util-7.563
onen-7.491
avorite-7.457
irtual-7.222
Token Clinton
Token ID2605
Feature activation-3.41e-02

Pos logit contributions
 axes+2.758
ochond+2.756
OTOS+2.589
 acceler+2.577
 pollut+2.558
Neg logit contributions
util-4.678
onen-4.607
idium-4.546
hler-4.431
avorite-4.411
TokenSanders
Token ID26747
Feature activation-1.60e-01

Pos logit contributions
 axes+0.312
ochond+0.309
OTOS+0.295
 acceler+0.293
opsis+0.292
Neg logit contributions
util-0.362
onen-0.356
idium-0.348
hler-0.339
avorite-0.334
Token:
Token ID25
Feature activation-2.08e-01

Pos logit contributions
 axes+1.338
ochond+1.307
 acceler+1.282
 pollut+1.249
OTOS+1.241
Neg logit contributions
util-1.854
onen-1.802
idium-1.771
hler-1.734
avorite-1.697
Token '
Token ID705
Feature activation-7.43e-01

Pos logit contributions
 axes+1.324
 acceler+1.274
ochond+1.222
OTOS+1.211
opsis+1.182
Neg logit contributions
util-2.806
onen-2.723
idium-2.712
hler-2.660
avorite-2.657
TokenI
Token ID40
Feature activation-7.96e-01

Pos logit contributions
 axes+5.175
ochond+4.740
 acceler+4.596
bec+4.126
 offsets+4.095
Neg logit contributions
avorite-9.375
util-9.269
unia-9.118
onen-9.094
idium-9.085
Token fully
Token ID3938
Feature activation-4.38e-01

Pos logit contributions
 axes+5.595
ochond+4.950
 govern+4.937
 acceler+4.538
 pollut+4.510
Neg logit contributions
idium-9.931
onen-9.806
irtual-9.800
util-9.750
cone-9.610
Token expect
Token ID1607
Feature activation-2.80e-01

Pos logit contributions
 axes+3.479
ochond+3.304
 acceler+3.235
 pollut+3.130
OTOS+3.068
Neg logit contributions
util-5.143
idium-5.110
onen-5.102
hler-4.878
ilogy-4.827
Token'
Token ID6
Feature activation-2.02e-02

Pos logit contributions
 axes+2.307
 pollut+2.095
 acceler+2.083
ochond+2.078
opsis+2.065
Neg logit contributions
util-3.308
onen-3.201
illance-3.139
idium-3.131
irtual-3.046
Token fair
Token ID3148
Feature activation-1.86e-01

Pos logit contributions
ochond+0.152
 axes+0.150
escription+0.145
opsis+0.145
 pollut+0.141
Neg logit contributions
util-0.251
onen-0.247
idium-0.240
hler-0.238
cone-0.232
Token treatment
Token ID3513
Feature activation-2.30e-01

Pos logit contributions
 axes+1.468
ochond+1.436
OTOS+1.366
opsis+1.363
 acceler+1.361
Neg logit contributions
util-2.242
onen-2.193
idium-2.172
hler-2.101
cone-2.073
Token[
Token ID58
Feature activation-8.61e-03

Pos logit contributions
 axes+1.793
ochond+1.774
opsis+1.707
OTOS+1.703
 acceler+1.688
Neg logit contributions
util-1.836
onen-1.813
idium-1.772
hler-1.716
cone-1.673
Token index
Token ID6376
Feature activation-2.91e-01

Pos logit contributions
 axes+0.091
ochond+0.091
OTOS+0.087
opsis+0.087
 pollut+0.087
Neg logit contributions
util-0.080
onen-0.078
idium-0.076
hler-0.073
cone-0.071
TokenState
Token ID9012
Feature activation+1.13e-01

Pos logit contributions
 axes+3.068
ochond+3.067
OTOS+2.949
opsis+2.926
 acceler+2.882
Neg logit contributions
util-2.598
onen-2.576
idium-2.535
hler-2.460
avorite-2.370
Token[
Token ID58
Feature activation-2.67e-01

Pos logit contributions
util+1.279
onen+1.249
idium+1.225
hler+1.192
cone+1.168
Neg logit contributions
ochond-0.968
 axes-0.968
 pollut-0.914
opsis-0.913
OTOS-0.912
TokenKeys
Token ID40729
Feature activation+2.26e-01

Pos logit contributions
 axes+2.806
ochond+2.777
OTOS+2.659
 acceler+2.656
opsis+2.643
Neg logit contributions
util-2.426
onen-2.356
idium-2.333
hler-2.260
avorite-2.224
Token[
Token ID58
Feature activation-7.89e-01

Pos logit contributions
util+2.192
onen+2.124
idium+2.068
hler+2.003
cone+1.966
Neg logit contributions
ochond-2.281
 axes-2.274
 pollut-2.181
OTOS-2.173
opsis-2.172
Tokenstep
Token ID9662
Feature activation+6.55e-02

Pos logit contributions
 axes+6.364
ochond+6.338
escription+6.062
 acceler+5.941
opsis+5.916
Neg logit contributions
idium-7.626
util-7.573
avorite-7.530
onen-7.427
hler-7.309
TokenNew
Token ID3791
Feature activation-5.32e-02

Pos logit contributions
util+0.530
onen+0.515
idium+0.503
hler+0.482
cone+0.468
Neg logit contributions
 axes-0.768
ochond-0.767
OTOS-0.738
opsis-0.735
 pollut-0.732
Token][
Token ID7131
Feature activation-4.04e-01

Pos logit contributions
 axes+0.472
ochond+0.469
opsis+0.443
OTOS+0.442
 acceler+0.441
Neg logit contributions
util-0.581
onen-0.569
idium-0.562
hler-0.543
cone-0.532
Token[
Token ID58
Feature activation-4.54e-01

Pos logit contributions
 axes+4.060
ochond+3.963
 acceler+3.817
OTOS+3.775
escription+3.765
Neg logit contributions
util-3.733
idium-3.633
onen-3.628
avorite-3.511
hler-3.500
Tokeni
Token ID72
Feature activation-1.95e-01

Pos logit contributions
 axes+4.623
ochond+4.524
 acceler+4.329
opsis+4.319
OTOS+4.305
Neg logit contributions
util-4.120
onen-4.009
idium-3.985
avorite-3.873
hler-3.872
Token
Token ID198
Feature activation-4.86e-01

Pos logit contributions
 axes+4.506
OTOS+4.319
 DAC+4.176
 acceler+4.146
ochond+4.091
Neg logit contributions
idium-8.495
onen-8.472
util-8.462
irtual-8.233
ussen-8.124
Token
Token ID198
Feature activation-6.96e-01

Pos logit contributions
 axes+3.620
ochond+3.605
OTOS+3.458
opsis+3.399
 acceler+3.390
Neg logit contributions
avorite-5.732
onen-5.703
util-5.698
idium-5.697
irtual-5.591
Token14
Token ID1415
Feature activation-3.74e-01

Pos logit contributions
OTOS+6.082
 axes+6.042
ochond+5.949
opsis+5.789
 acceler+5.551
Neg logit contributions
onen-6.841
idium-6.753
util-6.676
irtual-6.511
ilogy-6.434
Token Recommend
Token ID30361
Feature activation-4.73e-01

Pos logit contributions
 axes+3.133
ochond+2.985
OTOS+2.969
 pollut+2.939
 acceler+2.887
Neg logit contributions
util-4.171
idium-4.131
onen-4.101
hler-3.892
ilogy-3.887
Token
Token ID198
Feature activation-4.50e-01

Pos logit contributions
 axes+3.718
ochond+3.651
OTOS+3.503
opsis+3.481
 acceler+3.427
Neg logit contributions
util-5.480
onen-5.367
idium-5.350
avorite-5.204
irtual-5.145
Token
Token ID198
Feature activation-7.84e-01

Pos logit contributions
 axes+3.520
ochond+3.502
OTOS+3.344
opsis+3.299
 acceler+3.288
Neg logit contributions
util-5.188
onen-5.161
idium-5.146
avorite-5.107
irtual-4.971
TokenRecommend
Token ID41248
Feature activation-5.62e-01

Pos logit contributions
OTOS+6.036
 axes+5.703
ochond+5.421
opsis+5.413
 DAC+5.332
Neg logit contributions
onen-8.866
idium-8.863
util-8.588
irtual-8.448
ussen-8.396
Token Share
Token ID8734
Feature activation-4.36e-01

Pos logit contributions
 axes+3.162
opsis+2.871
ochond+2.764
 DAC+2.711
OTOS+2.688
Neg logit contributions
util-7.861
idium-7.713
onen-7.612
irtual-7.484
hler-7.353
Token this
Token ID428
Feature activation-6.22e-01

Pos logit contributions
 axes+3.265
ochond+3.196
opsis+3.085
 acceler+3.021
OTOS+2.998
Neg logit contributions
util-5.350
onen-5.280
idium-5.209
hler-5.038
cone-4.969
Token comment
Token ID2912
Feature activation-1.87e-01

Pos logit contributions
 axes+3.341
 acceler+2.977
OTOS+2.970
 pollut+2.914
 DAC+2.899
Neg logit contributions
onen-8.598
idium-8.594
util-8.523
avorite-8.372
hler-8.169
Token on
Token ID319
Feature activation-3.43e-01

Pos logit contributions
 axes+1.576
ochond+1.531
 acceler+1.492
OTOS+1.485
 pollut+1.479
Neg logit contributions
util-2.115
idium-2.073
onen-2.065
hler-1.993
avorite-1.951
Token required
Token ID2672
Feature activation-2.45e-02

Pos logit contributions
 axes+0.210
ochond+0.200
 pollut+0.195
 acceler+0.194
OTOS+0.190
Neg logit contributions
util-0.269
onen-0.267
idium-0.261
hler-0.253
irtual-0.249
Token=
Token ID28
Feature activation+2.81e-02

Pos logit contributions
ochond+0.211
 axes+0.211
escription+0.209
 pollut+0.203
 acceler+0.200
Neg logit contributions
util-0.281
onen-0.270
idium-0.263
hler-0.260
cone-0.254
TokenTrue
Token ID17821
Feature activation+6.09e-01

Pos logit contributions
util+0.209
onen+0.199
idium+0.190
hler+0.183
cone+0.178
Neg logit contributions
ochond-0.357
escription-0.356
 axes-0.353
 pollut-0.346
opsis-0.342
Token,
Token ID11
Feature activation-2.62e-02

Pos logit contributions
util+4.328
onen+3.921
hler+3.680
hey+3.558
kov+3.528
Neg logit contributions
escription-8.189
 dilig-7.571
 subsequ-7.453
ochond-7.432
 pollut-7.309
Token help
Token ID1037
Feature activation-2.49e-02

Pos logit contributions
ochond+0.272
 axes+0.263
escription+0.258
OTOS+0.258
opsis+0.255
Neg logit contributions
util-0.259
onen-0.248
hler-0.235
idium-0.235
cone-0.225
Token="
Token ID2625
Feature activation-7.84e-01

Pos logit contributions
ochond+0.195
 axes+0.195
escription+0.190
 pollut+0.187
 acceler+0.183
Neg logit contributions
util-0.306
onen-0.296
idium-0.289
hler-0.283
cone-0.276
Tokenpath
Token ID6978
Feature activation-5.19e-01

Pos logit contributions
 axes+5.681
ochond+5.575
opsis+5.154
 pollut+5.111
OTOS+5.099
Neg logit contributions
avorite-9.042
idium-8.947
onen-8.716
util-8.710
ilogy-8.552
Token to
Token ID284
Feature activation-3.12e-01

Pos logit contributions
 axes+3.859
ochond+3.822
 acceler+3.588
OTOS+3.579
opsis+3.572
Neg logit contributions
util-6.292
onen-6.210
idium-6.181
avorite-6.021
hler-5.993
Token images
Token ID4263
Feature activation-2.05e-02

Pos logit contributions
 axes+2.427
ochond+2.380
OTOS+2.264
 acceler+2.249
 pollut+2.243
Neg logit contributions
util-3.762
onen-3.697
idium-3.635
hler-3.545
cone-3.483
Token directory
Token ID8619
Feature activation-1.13e-01

Pos logit contributions
ochond+0.196
 axes+0.189
escription+0.188
OTOS+0.185
opsis+0.184
Neg logit contributions
util-0.215
onen-0.206
idium-0.202
hler-0.195
cone-0.195
Token")
Token ID4943
Feature activation-2.31e-01

Pos logit contributions
ochond+1.100
 axes+1.097
 pollut+1.050
OTOS+1.050
escription+1.049
Neg logit contributions
util-1.158
onen-1.121
idium-1.096
hler-1.068
cone-1.036
Token'
Token ID6
Feature activation+8.11e-02

Pos logit contributions
util+1.393
idium+1.351
hler+1.290
cone+1.278
onen+1.277
Neg logit contributions
ochond-1.618
OTOS-1.574
 axes-1.559
escription-1.525
opsis-1.508
Token 2016
Token ID1584
Feature activation-2.82e-01

Pos logit contributions
util+0.922
onen+0.913
idium+0.899
avorite+0.855
hler+0.850
Neg logit contributions
 axes-0.701
ochond-0.661
 pollut-0.657
 acceler-0.656
OTOS-0.635
Token primary
Token ID4165
Feature activation-3.73e-01

Pos logit contributions
 axes+2.653
 pollut+2.553
 acceler+2.534
opsis+2.475
 pickup+2.463
Neg logit contributions
util-2.970
onen-2.890
idium-2.862
avorite-2.751
irtual-2.750
Token Sanders
Token ID5831
Feature activation-1.44e-01

Pos logit contributions
 axes+3.381
 acceler+3.184
opsis+3.136
 DAC+3.097
 pollut+3.092
Neg logit contributions
util-4.100
onen-4.013
idium-3.883
irtual-3.857
avorite-3.725
Token:
Token ID25
Feature activation-3.37e-01

Pos logit contributions
 axes+1.090
 acceler+1.017
ochond+1.007
 pollut+0.992
OTOS+0.965
Neg logit contributions
util-1.810
onen-1.767
idium-1.717
hler-1.708
avorite-1.685
Token '
Token ID705
Feature activation-7.83e-01

Pos logit contributions
 axes+1.958
 acceler+1.878
ochond+1.769
opsis+1.765
 pollut+1.746
Neg logit contributions
util-4.671
onen-4.573
avorite-4.474
idium-4.458
cone-4.449
TokenDamn
Token ID43343
Feature activation-6.52e-02

Pos logit contributions
 axes+6.171
 acceler+5.521
ochond+5.335
yn+5.287
ait+5.181
Neg logit contributions
avorite-9.402
util-8.987
onen-8.789
unia-8.725
idium-8.552
Token right
Token ID826
Feature activation-7.17e-02

Pos logit contributions
 axes+0.581
ochond+0.575
 pollut+0.540
 acceler+0.540
opsis+0.539
Neg logit contributions
util-0.720
onen-0.703
idium-0.693
hler-0.674
cone-0.662
Token'
Token ID6
Feature activation-2.02e-01

Pos logit contributions
 axes+0.609
ochond+0.604
 pollut+0.572
 acceler+0.571
opsis+0.571
Neg logit contributions
util-0.817
onen-0.795
idium-0.784
hler-0.760
cone-0.747
Token I
Token ID314
Feature activation-4.94e-01

Pos logit contributions
 axes+1.489
 acceler+1.473
OTOS+1.463
ochond+1.455
 pollut+1.397
Neg logit contributions
util-2.480
onen-2.429
idium-2.412
avorite-2.371
cone-2.334
Token'll
Token ID1183
Feature activation-3.96e-01

Pos logit contributions
 axes+3.977
ochond+3.745
 acceler+3.638
 pollut+3.578
 govern+3.519
Neg logit contributions
onen-5.890
util-5.643
idium-5.562
irtual-5.557
perties-5.440
Token 2009
Token ID3717
Feature activation+5.52e-02

Pos logit contributions
 axes+2.974
ochond+2.962
OTOS+2.873
 pollut+2.863
 acceler+2.837
Neg logit contributions
util-4.160
idium-4.053
onen-4.008
cone-3.847
avorite-3.825
Token 10
Token ID838
Feature activation+2.84e-02

Pos logit contributions
util+0.554
onen+0.536
idium+0.522
hler+0.508
cone+0.496
Neg logit contributions
ochond-0.547
 axes-0.543
escription-0.527
opsis-0.520
OTOS-0.518
Token 2
Token ID362
Feature activation-3.41e-01

Pos logit contributions
util+0.303
onen+0.295
idium+0.289
hler+0.281
cone+0.275
Neg logit contributions
 axes-0.261
ochond-0.260
opsis-0.247
OTOS-0.247
 pollut-0.246
Token 2010
Token ID3050
Feature activation+1.46e-01

Pos logit contributions
 axes+2.736
ochond+2.631
 pollut+2.535
 acceler+2.521
opsis+2.481
Neg logit contributions
util-3.984
onen-3.879
idium-3.814
hler-3.772
ussen-3.674
Token 12
Token ID1105
Feature activation+7.09e-02

Pos logit contributions
util+1.429
onen+1.374
idium+1.337
hler+1.301
cone+1.268
Neg logit contributions
ochond-1.487
 axes-1.473
escription-1.453
opsis-1.415
 pollut-1.413
Token 2
Token ID362
Feature activation-7.81e-01

Pos logit contributions
util+0.746
onen+0.727
idium+0.709
hler+0.689
cone+0.676
Neg logit contributions
ochond-0.663
 axes-0.661
opsis-0.629
OTOS-0.628
escription-0.627
Token 2011
Token ID2813
Feature activation-2.18e-02

Pos logit contributions
 axes+5.137
ochond+4.618
 acceler+4.453
opsis+4.432
 DAC+4.392
Neg logit contributions
util-9.560
idium-9.354
onen-9.234
ilogy-8.992
irtual-8.923
Token 6
Token ID718
Feature activation+1.48e-01

Pos logit contributions
ochond+0.210
 axes+0.210
escription+0.200
opsis+0.200
OTOS+0.199
Neg logit contributions
util-0.223
onen-0.217
idium-0.212
hler-0.206
cone-0.201
Token 2
Token ID362
Feature activation-5.20e-01

Pos logit contributions
util+1.546
onen+1.506
idium+1.463
hler+1.427
cone+1.400
Neg logit contributions
ochond-1.395
 axes-1.385
opsis-1.323
escription-1.319
 pollut-1.319
Token 2012
Token ID2321
Feature activation+5.94e-02

Pos logit contributions
 axes+3.617
ochond+3.357
 pollut+3.161
opsis+3.123
 acceler+3.077
Neg logit contributions
onen-6.526
util-6.490
idium-6.307
hler-6.230
cone-6.038
Token 7
Token ID767
Feature activation+9.12e-03

Pos logit contributions
util+0.604
onen+0.586
idium+0.572
hler+0.556
cone+0.543
Neg logit contributions
ochond-0.577
 axes-0.576
escription-0.554
opsis-0.549
 pollut-0.547
Token former
Token ID1966
Feature activation+9.21e-02

Pos logit contributions
 axes+1.281
ochond+1.265
OTOS+1.205
 pollut+1.204
 acceler+1.202
Neg logit contributions
util-1.530
onen-1.496
idium-1.473
hler-1.423
cone-1.399
Token President
Token ID1992
Feature activation+1.99e-02

Pos logit contributions
util+1.051
onen+1.028
idium+1.005
hler+0.984
cone+0.960
Neg logit contributions
ochond-0.774
 axes-0.756
opsis-0.738
escription-0.724
OTOS-0.724
Token Barack
Token ID8732
Feature activation-1.15e-01

Pos logit contributions
util+0.209
onen+0.203
idium+0.198
hler+0.193
cone+0.188
Neg logit contributions
ochond-0.187
 axes-0.185
opsis-0.178
 pollut-0.178
OTOS-0.177
Token Obama
Token ID2486
Feature activation+3.02e-02

Pos logit contributions
ochond+0.962
 axes+0.951
escription+0.914
 pollut+0.904
opsis+0.904
Neg logit contributions
util-1.331
onen-1.289
idium-1.262
hler-1.233
cone-1.206
Token Barack
Token ID8732
Feature activation-5.19e-01

Pos logit contributions
util+0.328
onen+0.320
idium+0.315
hler+0.305
cone+0.299
Neg logit contributions
 axes-0.273
ochond-0.271
OTOS-0.258
opsis-0.257
 pollut-0.257
Token Hussein
Token ID22983
Feature activation-7.80e-01

Pos logit contributions
 axes+4.511
ochond+4.349
OTOS+4.206
opsis+4.187
 acceler+4.130
Neg logit contributions
onen-5.739
idium-5.603
util-5.579
irtual-5.467
avorite-5.425
Token Obama
Token ID2486
Feature activation-4.47e-01

Pos logit contributions
 axes+3.922
ochond+3.775
OTOS+3.695
 acceler+3.562
 pollut+3.262
Neg logit contributions
onen-10.890
util-10.821
idium-10.781
avorite-10.609
hler-10.562
TokenWith
Token ID3152
Feature activation-3.31e-01

Pos logit contributions
 axes+4.689
ochond+4.619
OTOS+4.521
 acceler+4.432
 pollut+4.329
Neg logit contributions
onen-3.981
util-3.939
hler-3.818
idium-3.809
avorite-3.731
Token low
Token ID1877
Feature activation-4.05e-01

Pos logit contributions
 axes+2.847
ochond+2.778
OTOS+2.702
 pollut+2.685
 acceler+2.683
Neg logit contributions
util-3.713
onen-3.694
idium-3.616
hler-3.492
irtual-3.423
Token birth
Token ID4082
Feature activation-1.00e-01

Pos logit contributions
 axes+3.382
ochond+3.357
 acceler+3.179
OTOS+3.153
 pollut+3.149
Neg logit contributions
util-4.686
onen-4.591
idium-4.514
hler-4.375
avorite-4.287
Token rate
Token ID2494
Feature activation-1.32e-01

Pos logit contributions
ochond+1.022
 axes+1.019
OTOS+0.982
opsis+0.976
 pollut+0.972
Neg logit contributions
util-0.967
onen-0.938
idium-0.921
hler-0.887
cone-0.878
Token.
Token ID13
Feature activation-3.57e-01

Pos logit contributions
 axes+1.230
ochond+1.224
escription+1.186
 pollut+1.174
OTOS+1.162
Neg logit contributions
util-1.284
onen-1.245
idium-1.212
hler-1.183
cone-1.151
Token The
Token ID383
Feature activation-2.06e-01

Pos logit contributions
ochond+3.017
 axes+3.011
OTOS+2.876
opsis+2.860
 acceler+2.836
Neg logit contributions
util-3.967
onen-3.918
idium-3.880
hler-3.731
avorite-3.730
Token player
Token ID2137
Feature activation-4.44e-01

Pos logit contributions
 axes+1.231
ochond+1.227
OTOS+1.151
escription+1.143
opsis+1.137
Neg logit contributions
util-2.867
onen-2.813
idium-2.754
hler-2.703
cone-2.659
Token must
Token ID1276
Feature activation-3.27e-01

Pos logit contributions
 axes+3.734
ochond+3.677
 acceler+3.554
 pollut+3.448
opsis+3.428
Neg logit contributions
util-5.013
onen-4.920
idium-4.903
avorite-4.759
hler-4.685
Token know
Token ID760
Feature activation-4.87e-01

Pos logit contributions
 axes+2.180
ochond+2.178
 govern+1.975
OTOS+1.947
 acceler+1.945
Neg logit contributions
onen-4.139
util-4.105
idium-4.079
cone-3.991
ilogy-3.913
Token choose
Token ID3853
Feature activation-7.72e-01

Pos logit contributions
 axes+4.662
ochond+4.603
 pollut+4.387
 acceler+4.382
opsis+4.318
Neg logit contributions
util-4.714
onen-4.701
idium-4.649
hler-4.489
cone-4.390
Token )
Token ID1267
Feature activation-2.44e-01

Pos logit contributions
 axes+6.657
ochond+6.225
 acceler+6.050
 pollut+5.956
opsis+5.816
Neg logit contributions
onen-7.980
util-7.825
idium-7.739
hler-7.521
cone-7.343
Token Permanent
Token ID36814
Feature activation-4.00e-01

Pos logit contributions
 axes+2.356
ochond+2.304
OTOS+2.213
 acceler+2.212
 pollut+2.190
Neg logit contributions
util-2.454
onen-2.436
idium-2.370
hler-2.294
cone-2.271
Token Voices
Token ID40440
Feature activation-1.50e-01

Pos logit contributions
 axes+3.670
ochond+3.483
 acceler+3.467
 pollut+3.432
OTOS+3.333
Neg logit contributions
util-4.229
onen-4.194
idium-4.023
hler-3.943
ilogy-3.942
Token in
Token ID287
Feature activation-3.93e-02

Pos logit contributions
 axes+1.289
ochond+1.268
 acceler+1.215
 pollut+1.213
opsis+1.206
Neg logit contributions
util-1.685
onen-1.652
idium-1.631
hler-1.574
ilogy-1.544
Token Your
Token ID3406
Feature activation+3.91e-02

Pos logit contributions
 axes+0.345
ochond+0.342
OTOS+0.325
 pollut+0.324
opsis+0.324
Neg logit contributions
util-0.436
onen-0.426
idium-0.418
hler-0.406
cone-0.398
Token trials
Token ID9867
Feature activation-6.88e-03

Pos logit contributions
util+2.134
onen+2.077
idium+2.021
hler+1.965
cone+1.937
Neg logit contributions
ochond-1.981
 axes-1.963
OTOS-1.890
opsis-1.883
escription-1.877
Token identifying
Token ID13720
Feature activation-9.52e-03

Pos logit contributions
 axes+0.063
ochond+0.063
OTOS+0.060
opsis+0.059
 pollut+0.059
Neg logit contributions
util-0.074
onen-0.072
idium-0.071
hler-0.069
cone-0.067
Token and
Token ID290
Feature activation+4.83e-01

Pos logit contributions
 axes+0.089
ochond+0.089
 acceler+0.084
 pollut+0.084
OTOS+0.084
Neg logit contributions
util-0.100
onen-0.098
idium-0.096
hler-0.093
cone-0.091
Token enroll
Token ID14627
Feature activation+4.34e-02

Pos logit contributions
util+4.536
onen+4.297
idium+4.151
hler+4.132
illance+4.055
Neg logit contributions
ochond-4.934
 axes-4.823
OTOS-4.757
opsis-4.744
escription-4.628
Tokening
Token ID278
Feature activation+8.59e-02

Pos logit contributions
util+0.572
onen+0.560
idium+0.551
hler+0.539
cone+0.529
Neg logit contributions
 axes-0.291
ochond-0.288
OTOS-0.271
opsis-0.270
 pollut-0.267
Token patients
Token ID3871
Feature activation+4.40e-01

Pos logit contributions
util+0.917
onen+0.895
idium+0.877
hler+0.852
cone+0.835
Neg logit contributions
ochond-0.787
 axes-0.787
OTOS-0.748
opsis-0.745
 pollut-0.742
Token with
Token ID351
Feature activation+3.51e-01

Pos logit contributions
util+4.366
onen+4.212
idium+4.082
hler+4.001
cone+3.960
Neg logit contributions
 axes-4.230
ochond-4.214
OTOS-4.132
escription-4.014
 pollut-3.963
Token C
Token ID327
Feature activation+1.22e-02

Pos logit contributions
util+3.470
onen+3.360
idium+3.300
hler+3.190
cone+3.158
Neg logit contributions
ochond-3.422
 axes-3.389
escription-3.276
OTOS-3.260
opsis-3.248
TokenIR
Token ID4663
Feature activation-1.93e-01

Pos logit contributions
util+0.159
onen+0.155
idium+0.153
hler+0.149
cone+0.146
Neg logit contributions
 axes-0.084
ochond-0.083
 pollut-0.078
opsis-0.077
 acceler-0.077
TokenCI
Token ID25690
Feature activation-2.38e-02

Pos logit contributions
 axes+2.279
ochond+2.272
OTOS+2.185
opsis+2.181
 acceler+2.174
Neg logit contributions
util-1.547
onen-1.505
idium-1.456
hler-1.403
cone-1.357
Token).
Token ID737
Feature activation-5.66e-02

Pos logit contributions
 axes+0.228
ochond+0.227
OTOS+0.216
opsis+0.216
 pollut+0.216
Neg logit contributions
util-0.245
onen-0.239
idium-0.234
hler-0.227
cone-0.222
Token enrollment
Token ID20753
Feature activation+2.94e-01

Pos logit contributions
util+1.697
onen+1.647
idium+1.610
hler+1.577
cone+1.545
Neg logit contributions
ochond-1.383
 axes-1.353
OTOS-1.315
escription-1.303
opsis-1.299
Token as
Token ID355
Feature activation+1.16e-01

Pos logit contributions
util+2.964
onen+2.882
idium+2.769
cone+2.726
hler+2.722
Neg logit contributions
ochond-2.818
escription-2.723
 axes-2.703
OTOS-2.688
opsis-2.656
Token part
Token ID636
Feature activation+2.36e-01

Pos logit contributions
util+1.236
onen+1.200
idium+1.182
hler+1.150
cone+1.128
Neg logit contributions
ochond-1.067
 axes-1.063
OTOS-1.017
opsis-1.014
 pollut-1.003
Token of
Token ID286
Feature activation+2.86e-01

Pos logit contributions
util+2.258
onen+2.157
hler+2.053
idium+2.024
cone+2.002
Neg logit contributions
escription-2.599
 axes-2.415
ochond-2.386
 pollut-2.328
opsis-2.314
Token the
Token ID262
Feature activation+1.69e-01

Pos logit contributions
util+3.101
onen+2.988
idium+2.914
hler+2.872
cone+2.807
Neg logit contributions
ochond-2.580
 axes-2.542
opsis-2.452
OTOS-2.442
escription-2.413
Token 2011
Token ID2813
Feature activation+4.50e-01

Pos logit contributions
util+1.835
onen+1.785
idium+1.734
hler+1.699
cone+1.657
Neg logit contributions
ochond-1.531
 axes-1.494
OTOS-1.455
opsis-1.445
escription-1.440
Token-
Token ID12
Feature activation-1.42e-01

Pos logit contributions
util+4.624
onen+4.379
idium+4.294
hler+4.167
cone+4.120
Neg logit contributions
ochond-4.291
 axes-4.078
escription-4.063
OTOS-4.060
opsis-3.999
Token13
Token ID1485
Feature activation+2.61e-01

Pos logit contributions
 axes+1.696
ochond+1.687
OTOS+1.626
 pollut+1.624
opsis+1.623
Neg logit contributions
util-1.097
onen-1.079
idium-1.055
hler-1.011
ilogy-0.969
Token state
Token ID1181
Feature activation+1.05e-01

Pos logit contributions
util+2.663
onen+2.556
idium+2.492
hler+2.424
cone+2.385
Neg logit contributions
ochond-2.552
 axes-2.472
escription-2.417
OTOS-2.411
opsis-2.402
Token budget
Token ID4466
Feature activation+2.25e-01

Pos logit contributions
util+0.976
onen+0.936
idium+0.913
hler+0.891
cone+0.874
Neg logit contributions
ochond-1.121
 axes-1.108
OTOS-1.070
opsis-1.070
escription-1.066
Token,
Token ID11
Feature activation+9.02e-02

Pos logit contributions
util+2.270
onen+2.183
idium+2.114
hler+2.058
cone+2.056
Neg logit contributions
ochond-2.194
 axes-2.125
escription-2.087
opsis-2.057
OTOS-2.044
Token "
Token ID366
Feature activation+1.70e-01

Pos logit contributions
util+2.487
onen+2.405
idium+2.318
hler+2.225
ilogy+2.144
Neg logit contributions
ochond-3.056
 axes-3.000
escription-2.944
OTOS-2.912
opsis-2.911
TokenP
Token ID47
Feature activation-1.45e-01

Pos logit contributions
util+0.926
onen+0.845
idium+0.773
cone+0.735
hler+0.731
Neg logit contributions
escription-2.555
 axes-2.509
ochond-2.503
 pollut-2.449
 acceler-2.406
Tokenulse
Token ID9615
Feature activation-3.22e-01

Pos logit contributions
 axes+1.574
ochond+1.569
OTOS+1.503
opsis+1.501
 pollut+1.499
Neg logit contributions
util-1.307
onen-1.270
idium-1.241
hler-1.196
cone-1.167
Token period
Token ID2278
Feature activation+7.57e-02

Pos logit contributions
 axes+2.453
ochond+2.426
 pollut+2.290
 acceler+2.280
opsis+2.273
Neg logit contributions
util-3.936
onen-3.863
idium-3.805
hler-3.704
cone-3.625
Token"
Token ID1
Feature activation+4.24e-01

Pos logit contributions
util+0.712
onen+0.684
idium+0.659
hler+0.643
cone+0.622
Neg logit contributions
escription-0.811
 axes-0.803
ochond-0.794
 pollut-0.772
OTOS-0.770
Token,
Token ID837
Feature activation+4.79e-01

Pos logit contributions
util+4.228
onen+3.936
idium+3.759
hler+3.729
kov+3.603
Neg logit contributions
escription-4.589
ochond-4.160
 pollut-4.110
 axes-4.099
 subsequ-3.995
Token Float
Token ID48436
Feature activation+3.06e-01

Pos logit contributions
util+4.129
onen+3.849
idium+3.631
hler+3.620
cone+3.545
Neg logit contributions
escription-5.481
ochond-5.238
 axes-5.056
opsis-5.048
OTOS-4.987
Token )
Token ID1267
Feature activation+6.25e-02

Pos logit contributions
util+3.671
onen+3.500
hler+3.440
idium+3.405
cone+3.248
Neg logit contributions
escription-2.689
 axes-2.393
ochond-2.387
 pollut-2.376
 subsequ-2.265
Token =
Token ID796
Feature activation+8.87e-02

Pos logit contributions
util+0.633
onen+0.602
idium+0.583
hler+0.579
cone+0.554
Neg logit contributions
escription-0.634
 axes-0.620
ochond-0.617
 pollut-0.601
opsis-0.584
Token 1
Token ID352
Feature activation+4.67e-01

Pos logit contributions
util+0.821
onen+0.797
idium+0.771
hler+0.744
cone+0.721
Neg logit contributions
escription-0.961
ochond-0.958
 axes-0.930
OTOS-0.908
 pollut-0.904
Token //
Token ID3373
Feature activation-1.88e-02

Pos logit contributions
util+2.979
onen+2.548
hler+2.511
idium+2.353
uristic+2.294
Neg logit contributions
escription-6.670
ochond-6.283
 axes-6.216
 pollut-6.131
phies-6.055
Token itself
Token ID2346
Feature activation-1.26e-01

Pos logit contributions
util+0.788
onen+0.769
idium+0.754
hler+0.731
cone+0.717
Neg logit contributions
ochond-0.646
 axes-0.644
OTOS-0.616
opsis-0.614
 pollut-0.610
Token will
Token ID481
Feature activation+6.54e-03

Pos logit contributions
 axes+1.105
ochond+1.092
 acceler+1.040
opsis+1.034
 pollut+1.034
Neg logit contributions
util-1.399
onen-1.369
idium-1.347
hler-1.305
cone-1.283
Token be
Token ID307
Feature activation+1.57e-01

Pos logit contributions
util+0.077
onen+0.075
idium+0.074
hler+0.072
cone+0.070
Neg logit contributions
ochond-0.053
 axes-0.053
OTOS-0.050
opsis-0.050
 pollut-0.050
Token subject
Token ID2426
Feature activation+1.21e-01

Pos logit contributions
util+1.676
onen+1.621
idium+1.578
hler+1.542
cone+1.500
Neg logit contributions
ochond-1.441
 axes-1.425
opsis-1.382
OTOS-1.373
 pollut-1.352
Token to
Token ID284
Feature activation+2.21e-01

Pos logit contributions
util+1.434
onen+1.410
idium+1.384
hler+1.350
cone+1.334
Neg logit contributions
ochond-0.961
 axes-0.955
OTOS-0.899
 pollut-0.895
 acceler-0.894
Token sh
Token ID427
Feature activation+4.20e-01

Pos logit contributions
util+2.370
onen+2.293
idium+2.242
hler+2.182
cone+2.110
Neg logit contributions
ochond-2.014
 axes-1.997
opsis-1.932
OTOS-1.905
escription-1.882
Tokenaria
Token ID10312
Feature activation+1.76e-01

Pos logit contributions
util+3.222
onen+3.150
idium+3.079
unia+2.742
hler+2.724
Neg logit contributions
 axes-5.119
ochond-5.027
OTOS-5.010
 pollut-4.973
 acceler-4.967
Token in
Token ID287
Feature activation+1.26e-01

Pos logit contributions
util+1.945
onen+1.893
idium+1.857
hler+1.799
cone+1.756
Neg logit contributions
ochond-1.548
 axes-1.518
opsis-1.470
OTOS-1.459
escription-1.431
Token its
Token ID663
Feature activation+1.98e-01

Pos logit contributions
util+1.384
onen+1.343
idium+1.313
hler+1.275
cone+1.249
Neg logit contributions
ochond-1.131
 axes-1.120
OTOS-1.076
opsis-1.071
 pollut-1.060
Token dealings
Token ID29043
Feature activation+5.53e-03

Pos logit contributions
util+2.311
onen+2.266
idium+2.213
hler+2.126
cone+2.101
Neg logit contributions
ochond-1.603
 axes-1.563
OTOS-1.533
opsis-1.513
escription-1.512
Token on
Token ID319
Feature activation+8.82e-02

Pos logit contributions
util+0.063
onen+0.061
idium+0.061
hler+0.059
cone+0.058
Neg logit contributions
 axes-0.047
ochond-0.047
opsis-0.044
 acceler-0.044
OTOS-0.044
Token had
Token ID550
Feature activation+4.05e-01

Pos logit contributions
util+4.273
onen+4.130
idium+4.019
hler+3.843
cone+3.785
Neg logit contributions
ochond-4.291
escription-4.240
OTOS-4.227
 pollut-4.079
 axes-4.065
Token lost
Token ID2626
Feature activation+2.59e-01

Pos logit contributions
util+3.762
onen+3.585
idium+3.470
hler+3.321
cone+3.269
Neg logit contributions
ochond-4.277
escription-4.168
OTOS-4.166
 axes-4.068
opsis-4.068
Token when
Token ID618
Feature activation+3.08e-02

Pos logit contributions
util+2.557
onen+2.483
idium+2.383
hler+2.311
cone+2.309
Neg logit contributions
ochond-2.572
 axes-2.494
OTOS-2.452
 pollut-2.440
opsis-2.421
Token he
Token ID339
Feature activation+3.24e-01

Pos logit contributions
util+0.329
onen+0.322
idium+0.315
hler+0.306
cone+0.300
Neg logit contributions
ochond-0.282
 axes-0.282
OTOS-0.268
opsis-0.267
 pollut-0.267
Token murdered
Token ID12864
Feature activation+1.19e-01

Pos logit contributions
util+3.132
onen+3.021
idium+2.936
hler+2.854
cone+2.790
Neg logit contributions
ochond-3.240
OTOS-3.187
escription-3.134
 axes-3.105
 pollut-3.096
Token Yoshi
Token ID38936
Feature activation+4.36e-01

Pos logit contributions
util+1.321
onen+1.291
idium+1.261
hler+1.226
cone+1.201
Neg logit contributions
ochond-1.029
 axes-1.026
OTOS-0.976
opsis-0.972
 pollut-0.969
Token.
Token ID13
Feature activation-7.97e-04

Pos logit contributions
util+4.141
onen+4.086
idium+3.892
hler+3.856
cone+3.752
Neg logit contributions
ochond-4.385
 axes-4.297
 pollut-4.180
OTOS-4.103
escription-4.099
Token S
Token ID311
Feature activation+1.57e-01

Pos logit contributions
 axes+0.008
ochond+0.008
 pollut+0.007
OTOS+0.007
opsis+0.007
Neg logit contributions
util-0.008
onen-0.008
idium-0.008
hler-0.008
cone-0.007
Tokenaki
Token ID8182
Feature activation-1.49e-01

Pos logit contributions
util+1.297
onen+1.225
idium+1.204
hler+1.143
cone+1.133
Neg logit contributions
 axes-1.885
escription-1.847
 pollut-1.841
ochond-1.821
 acceler-1.789
Token refuses
Token ID17567
Feature activation-4.86e-02

Pos logit contributions
 axes+1.250
ochond+1.223
 acceler+1.173
opsis+1.160
 pollut+1.156
Neg logit contributions
util-1.714
onen-1.677
idium-1.651
hler-1.602
cone-1.575
Token and
Token ID290
Feature activation-1.98e-01

Pos logit contributions
 axes+0.427
ochond+0.421
OTOS+0.400
 acceler+0.400
opsis+0.399
Neg logit contributions
util-0.539
onen-0.527
idium-0.518
hler-0.503
cone-0.492
Token:
Token ID25
Feature activation+4.00e-01

Pos logit contributions
util+3.090
onen+3.009
idium+2.920
hler+2.865
cone+2.775
Neg logit contributions
ochond-2.929
 axes-2.904
 pollut-2.799
opsis-2.787
OTOS-2.771
Token how
Token ID703
Feature activation+1.87e-01

Pos logit contributions
util+3.923
onen+3.772
idium+3.655
hler+3.593
cone+3.478
Neg logit contributions
ochond-3.962
 axes-3.842
OTOS-3.740
 pollut-3.703
opsis-3.701
Token to
Token ID284
Feature activation+1.97e-01

Pos logit contributions
util+2.012
onen+1.965
idium+1.909
hler+1.869
cone+1.816
Neg logit contributions
ochond-1.707
 axes-1.679
OTOS-1.623
opsis-1.607
 pollut-1.605
Token teach
Token ID4545
Feature activation+8.41e-02

Pos logit contributions
util+2.268
onen+2.206
idium+2.144
hler+2.110
cone+2.044
Neg logit contributions
ochond-1.661
 axes-1.639
OTOS-1.587
escription-1.574
opsis-1.555
Token yourself
Token ID3511
Feature activation+1.92e-01

Pos logit contributions
util+0.926
onen+0.905
idium+0.883
hler+0.862
cone+0.839
Neg logit contributions
ochond-0.746
 axes-0.737
OTOS-0.707
opsis-0.704
 pollut-0.702
Token to
Token ID284
Feature activation+1.10e-01

Pos logit contributions
util+2.095
onen+2.048
idium+1.985
hler+1.943
cone+1.881
Neg logit contributions
ochond-1.710
 axes-1.686
OTOS-1.627
opsis-1.617
 pollut-1.606
Token code
Token ID2438
Feature activation+8.42e-03

Pos logit contributions
util+1.430
onen+1.402
idium+1.364
hler+1.343
cone+1.312
Neg logit contributions
ochond-0.763
 axes-0.760
OTOS-0.715
opsis-0.707
escription-0.707
Token
Token ID198
Feature activation-1.31e-02

Pos logit contributions
util+0.092
onen+0.090
idium+0.088
hler+0.085
cone+0.084
Neg logit contributions
 axes-0.076
ochond-0.075
OTOS-0.071
opsis-0.071
 pollut-0.071
Token
Token ID198
Feature activation+2.90e-02

Pos logit contributions
escription+0.131
 axes+0.131
ochond+0.131
 pollut+0.126
opsis+0.124
Neg logit contributions
util-0.133
onen-0.127
idium-0.123
hler-0.121
cone-0.118
TokenApple
Token ID16108
Feature activation-4.61e-03

Pos logit contributions
util+0.263
onen+0.255
idium+0.248
hler+0.240
cone+0.234
Neg logit contributions
 axes-0.314
ochond-0.313
 pollut-0.299
OTOS-0.299
opsis-0.299
Token
Token ID447
Feature activation+9.69e-02

Pos logit contributions
ochond+0.043
 axes+0.043
OTOS+0.041
 pollut+0.041
opsis+0.041
Neg logit contributions
util-0.048
onen-0.047
idium-0.046
hler-0.045
cone-0.044
TokenFacebook
Token ID12025
Feature activation+8.58e-02

Pos logit contributions
util+0.183
onen+0.177
idium+0.172
hler+0.166
cone+0.162
Neg logit contributions
 axes-0.218
ochond-0.216
 pollut-0.208
opsis-0.207
 acceler-0.207
Token Twitter
Token ID3009
Feature activation-2.52e-01

Pos logit contributions
util+0.888
onen+0.868
idium+0.845
hler+0.820
cone+0.801
Neg logit contributions
 axes-0.820
ochond-0.815
opsis-0.774
 pollut-0.774
OTOS-0.773
Token Pinterest
Token ID17334
Feature activation+1.26e-01

Pos logit contributions
 axes+2.276
ochond+2.267
OTOS+2.152
opsis+2.148
 pollut+2.144
Neg logit contributions
util-2.731
onen-2.668
idium-2.617
hler-2.540
cone-2.488
Token Liverpool
Token ID11761
Feature activation-2.87e-02

Pos logit contributions
util+1.286
onen+1.251
idium+1.213
hler+1.183
cone+1.150
Neg logit contributions
 axes-1.226
ochond-1.225
 pollut-1.168
opsis-1.166
OTOS-1.157
Token
Token ID447
Feature activation+9.50e-02

Pos logit contributions
 axes+0.255
ochond+0.254
OTOS+0.241
opsis+0.240
 pollut+0.240
Neg logit contributions
util-0.315
onen-0.308
idium-0.302
hler-0.293
cone-0.287
Token
Token ID247
Feature activation+1.92e-02

Pos logit contributions
util+0.751
onen+0.725
idium+0.700
hler+0.692
cone+0.660
Neg logit contributions
ochond-1.140
 axes-1.134
 pollut-1.099
escription-1.098
OTOS-1.096
Tokens
Token ID82
Feature activation+9.95e-02

Pos logit contributions
util+0.204
onen+0.199
idium+0.194
hler+0.188
cone+0.184
Neg logit contributions
 axes-0.178
ochond-0.178
OTOS-0.169
 pollut-0.169
opsis-0.169
Token Adam
Token ID7244
Feature activation-1.07e-01

Pos logit contributions
util+1.048
onen+1.016
idium+0.994
hler+0.964
cone+0.939
Neg logit contributions
ochond-0.929
 axes-0.921
opsis-0.887
OTOS-0.887
 pollut-0.878
Token L
Token ID406
Feature activation+1.61e-01

Pos logit contributions
 axes+0.924
ochond+0.922
OTOS+0.872
opsis+0.872
 pollut+0.868
Neg logit contributions
util-1.190
onen-1.162
idium-1.143
hler-1.107
cone-1.091
Tokenall
Token ID439
Feature activation+1.33e-01

Pos logit contributions
util+1.274
onen+1.224
idium+1.196
hler+1.156
ussen+1.112
Neg logit contributions
 axes-1.931
ochond-1.909
OTOS-1.879
 pollut-1.864
opsis-1.860
Tokenana
Token ID2271
Feature activation-4.27e-02

Pos logit contributions
util+0.961
onen+0.937
idium+0.871
hler+0.836
cone+0.806
Neg logit contributions
 axes-1.700
ochond-1.665
OTOS-1.665
 pollut-1.639
escription-1.630
Token sound
Token ID2128
Feature activation-1.13e-02

Pos logit contributions
util+1.186
onen+1.157
idium+1.132
hler+1.099
cone+1.074
Neg logit contributions
ochond-1.053
 axes-1.051
OTOS-1.000
opsis-0.996
 pollut-0.994
Tokentracks
Token ID46074
Feature activation+1.44e-01

Pos logit contributions
 axes+0.079
ochond+0.079
OTOS+0.074
opsis+0.074
 pollut+0.073
Neg logit contributions
util-0.145
onen-0.142
idium-0.139
hler-0.136
cone-0.134
Token for
Token ID329
Feature activation+2.08e-02

Pos logit contributions
util+1.597
onen+1.557
idium+1.526
hler+1.488
cone+1.455
Neg logit contributions
ochond-1.269
 axes-1.266
OTOS-1.206
 pollut-1.194
opsis-1.192
Token seasons
Token ID7028
Feature activation-1.64e-01

Pos logit contributions
util+0.218
onen+0.213
idium+0.209
hler+0.202
cone+0.198
Neg logit contributions
 axes-0.196
ochond-0.195
OTOS-0.186
opsis-0.185
 pollut-0.185
Token 9
Token ID860
Feature activation+2.38e-01

Pos logit contributions
 axes+1.477
ochond+1.451
 acceler+1.387
opsis+1.385
 pollut+1.378
Neg logit contributions
util-1.787
onen-1.745
idium-1.714
hler-1.661
cone-1.629
Token through
Token ID832
Feature activation+9.82e-02

Pos logit contributions
util+2.486
onen+2.413
idium+2.378
hler+2.321
cone+2.272
Neg logit contributions
ochond-2.201
 axes-2.199
OTOS-2.097
 pollut-2.088
opsis-2.081
Token 27
Token ID2681
Feature activation+8.48e-02

Pos logit contributions
util+0.993
onen+0.960
idium+0.934
hler+0.908
cone+0.892
Neg logit contributions
ochond-0.963
 axes-0.954
escription-0.919
OTOS-0.916
opsis-0.913
Token (
Token ID357
Feature activation+4.38e-02

Pos logit contributions
util+0.903
onen+0.880
idium+0.865
hler+0.841
cone+0.823
Neg logit contributions
 axes-0.777
ochond-0.776
OTOS-0.737
 pollut-0.736
opsis-0.732
Tokenwith
Token ID4480
Feature activation-1.78e-01

Pos logit contributions
util+0.410
onen+0.397
idium+0.387
hler+0.375
cone+0.368
Neg logit contributions
 axes-0.462
ochond-0.460
 pollut-0.441
opsis-0.441
OTOS-0.441
Token the
Token ID262
Feature activation-7.83e-02

Pos logit contributions
 axes+1.618
ochond+1.574
 acceler+1.512
 pollut+1.510
opsis+1.505
Neg logit contributions
util-1.919
onen-1.899
idium-1.859
hler-1.776
cone-1.768
Token exception
Token ID6631
Feature activation-3.04e-01

Pos logit contributions
 axes+0.719
ochond+0.706
 pollut+0.674
 acceler+0.673
opsis+0.672
Neg logit contributions
util-0.838
onen-0.822
idium-0.808
hler-0.778
cone-0.766
Token do
Token ID466
Feature activation+2.07e-01

Pos logit contributions
util+3.504
onen+3.461
idium+3.373
hler+3.305
cone+3.238
Neg logit contributions
ochond-2.362
 axes-2.341
OTOS-2.216
opsis-2.214
escription-2.189
Token with
Token ID351
Feature activation+1.50e-01

Pos logit contributions
util+1.965
onen+1.936
idium+1.838
hler+1.808
cone+1.763
Neg logit contributions
escription-2.124
ochond-2.063
 axes-2.042
OTOS-2.004
opsis-1.976
Token the
Token ID262
Feature activation+1.25e-01

Pos logit contributions
util+1.607
onen+1.563
idium+1.537
hler+1.489
cone+1.457
Neg logit contributions
ochond-1.365
 axes-1.351
OTOS-1.295
opsis-1.290
 pollut-1.280
Token requests
Token ID7007
Feature activation+3.59e-02

Pos logit contributions
util+1.326
onen+1.291
idium+1.267
hler+1.229
cone+1.204
Neg logit contributions
ochond-1.148
 axes-1.136
OTOS-1.090
opsis-1.088
 pollut-1.075
Token of
Token ID286
Feature activation+1.41e-01

Pos logit contributions
util+0.410
onen+0.402
idium+0.393
hler+0.383
cone+0.375
Neg logit contributions
 axes-0.305
ochond-0.302
opsis-0.285
 acceler-0.285
 pollut-0.285
Token other
Token ID584
Feature activation+1.83e-01

Pos logit contributions
util+1.449
onen+1.408
idium+1.383
hler+1.338
cone+1.310
Neg logit contributions
ochond-1.345
 axes-1.340
OTOS-1.284
opsis-1.280
 pollut-1.270
Token autonomous
Token ID18284
Feature activation+1.77e-01

Pos logit contributions
util+1.971
onen+1.921
idium+1.882
hler+1.827
cone+1.798
Neg logit contributions
ochond-1.650
 axes-1.628
OTOS-1.578
opsis-1.567
escription-1.554
Token communities
Token ID5348
Feature activation-4.19e-02

Pos logit contributions
util+1.939
onen+1.900
idium+1.852
hler+1.794
cone+1.764
Neg logit contributions
ochond-1.576
 axes-1.551
OTOS-1.503
opsis-1.484
 pollut-1.479
Token.
Token ID13
Feature activation-2.23e-01

Pos logit contributions
 axes+0.383
ochond+0.380
opsis+0.361
OTOS+0.361
 acceler+0.361
Neg logit contributions
util-0.449
onen-0.438
idium-0.431
hler-0.417
cone-0.409
Token
Token ID447
Feature activation-1.07e-02

Pos logit contributions
 axes+2.282
ochond+2.281
OTOS+2.195
opsis+2.175
 acceler+2.161
Neg logit contributions
util-2.079
onen-2.041
idium-2.030
avorite-1.958
hler-1.933
Token
Token ID251
Feature activation-1.21e-01

Pos logit contributions
 axes+0.089
ochond+0.088
opsis+0.084
OTOS+0.083
 acceler+0.082
Neg logit contributions
util-0.124
onen-0.120
idium-0.119
irtual-0.114
hler-0.113
Token said
Token ID531
Feature activation-2.29e-01

Pos logit contributions
ochond+1.698
 axes+1.679
opsis+1.599
 acceler+1.597
 pollut+1.584
Neg logit contributions
util-1.595
onen-1.561
idium-1.548
avorite-1.465
illance-1.449
Token.
Token ID13
Feature activation-1.05e-01

Pos logit contributions
ochond+1.862
 axes+1.822
opsis+1.726
 acceler+1.712
 pollut+1.697
Neg logit contributions
util-2.641
onen-2.610
idium-2.570
hler-2.478
irtual-2.459
Token
Token ID564
Feature activation-8.62e-02

Pos logit contributions
 axes+0.975
ochond+0.973
OTOS+0.926
opsis+0.924
 acceler+0.919
Neg logit contributions
util-1.092
onen-1.074
idium-1.058
hler-1.019
cone-1.000
Token
Token ID250
Feature activation-2.23e-02

Pos logit contributions
ochond+1.045
 axes+1.043
 pollut+1.004
escription+1.001
OTOS+1.000
Neg logit contributions
util-0.673
onen-0.651
idium-0.630
hler-0.609
cone-0.588
TokenWe
Token ID1135
Feature activation-2.11e-01

Pos logit contributions
 axes+0.235
ochond+0.234
OTOS+0.224
opsis+0.224
 pollut+0.224
Neg logit contributions
util-0.207
onen-0.201
idium-0.197
hler-0.190
cone-0.186
Token were
Token ID547
Feature activation+1.97e-02

Pos logit contributions
 axes+1.927
ochond+1.923
opsis+1.823
 pollut+1.817
 acceler+1.812
Neg logit contributions
util-2.221
onen-2.191
idium-2.158
hler-2.062
cone-2.048
Token picking
Token ID10868
Feature activation-1.82e-01

Pos logit contributions
util+0.209
onen+0.204
idium+0.200
hler+0.194
cone+0.190
Neg logit contributions
 axes-0.182
ochond-0.182
OTOS-0.173
opsis-0.172
 pollut-0.172
Token some
Token ID617
Feature activation-1.75e-01

Pos logit contributions
 axes+1.647
ochond+1.618
 acceler+1.540
opsis+1.537
 pollut+1.533
Neg logit contributions
util-1.973
onen-1.933
idium-1.904
hler-1.850
cone-1.800
Token of
Token ID286
Feature activation+1.09e-03

Pos logit contributions
 axes+1.529
ochond+1.497
 acceler+1.428
opsis+1.427
 pollut+1.426
Neg logit contributions
util-1.949
onen-1.905
idium-1.873
hler-1.831
ilogy-1.775
Token our
Token ID674
Feature activation-1.68e-01

Pos logit contributions
util+0.012
onen+0.012
idium+0.012
hler+0.011
cone+0.011
Neg logit contributions
 axes-0.010
ochond-0.009
opsis-0.009
 pollut-0.009
 acceler-0.009
Token grapes
Token ID36656
Feature activation+1.79e-01

Pos logit contributions
 axes+1.549
ochond+1.504
 acceler+1.449
 pollut+1.445
opsis+1.437
Neg logit contributions
util-1.806
onen-1.769
idium-1.737
hler-1.697
cone-1.646
Token
Token ID251
Feature activation+2.39e-03

Pos logit contributions
 axes+0.729
ochond+0.727
OTOS+0.686
opsis+0.685
 pollut+0.684
Neg logit contributions
util-1.015
onen-0.993
idium-0.976
hler-0.949
cone-0.931
Token,
Token ID11
Feature activation-1.84e-01

Pos logit contributions
util+0.026
onen+0.026
idium+0.025
hler+0.024
cone+0.024
Neg logit contributions
 axes-0.021
ochond-0.021
opsis-0.020
OTOS-0.020
 acceler-0.020
Token it
Token ID340
Feature activation-1.53e-01

Pos logit contributions
 axes+1.483
ochond+1.452
 acceler+1.390
 pollut+1.376
OTOS+1.371
Neg logit contributions
util-2.166
onen-2.137
idium-2.085
hler-2.039
cone-2.010
Token reads
Token ID9743
Feature activation+1.98e-01

Pos logit contributions
 axes+1.029
ochond+1.007
 acceler+0.959
opsis+0.943
OTOS+0.937
Neg logit contributions
util-2.002
onen-1.967
idium-1.931
cone-1.882
hler-1.881
Token.
Token ID13
Feature activation-2.24e-01

Pos logit contributions
util+2.171
onen+2.126
idium+2.080
hler+2.022
cone+1.987
Neg logit contributions
 axes-1.785
ochond-1.773
opsis-1.688
OTOS-1.684
 pollut-1.677
Token
Token ID198
Feature activation-1.16e-01

Pos logit contributions
 axes+1.751
ochond+1.708
OTOS+1.686
 acceler+1.625
 pollut+1.621
Neg logit contributions
util-2.609
onen-2.603
idium-2.560
hler-2.496
cone-2.466
Token
Token ID198
Feature activation-4.24e-02

Pos logit contributions
ochond+1.109
 axes+1.109
escription+1.084
 pollut+1.058
opsis+1.051
Neg logit contributions
util-1.203
onen-1.163
idium-1.130
hler-1.106
cone-1.079
TokenAudio
Token ID21206
Feature activation+4.06e-02

Pos logit contributions
 axes+0.461
ochond+0.460
OTOS+0.445
opsis+0.440
 pollut+0.438
Neg logit contributions
util-0.375
onen-0.367
idium-0.359
hler-0.346
cone-0.337
Token evidence
Token ID2370
Feature activation-7.56e-02

Pos logit contributions
util+0.466
onen+0.455
idium+0.448
hler+0.435
cone+0.427
Neg logit contributions
 axes-0.340
ochond-0.340
OTOS-0.320
opsis-0.318
 pollut-0.318
Token
Token ID198
Feature activation-3.27e-02

Pos logit contributions
 axes+0.657
ochond+0.654
OTOS+0.618
opsis+0.618
 pollut+0.617
Neg logit contributions
util-0.844
onen-0.828
idium-0.811
hler-0.789
cone-0.773
Token
Token ID198
Feature activation+5.85e-02

Pos logit contributions
ochond+0.324
 axes+0.324
escription+0.323
 pollut+0.311
opsis+0.307
Neg logit contributions
util-0.331
onen-0.318
idium-0.308
hler-0.301
cone-0.295
Token '
Token ID705
Feature activation-2.15e-01

Pos logit contributions
 axes+1.958
ochond+1.890
 acceler+1.836
 pollut+1.834
OTOS+1.810
Neg logit contributions
util-2.643
onen-2.611
idium-2.544
hler-2.479
cone-2.430
Token44
Token ID2598
Feature activation-1.58e-01

Pos logit contributions
 axes+2.372
ochond+2.361
OTOS+2.271
 pollut+2.264
 acceler+2.255
Neg logit contributions
util-1.881
idium-1.811
onen-1.804
hler-1.688
ussen-1.673
Token'
Token ID6
Feature activation+3.43e-02

Pos logit contributions
 axes+1.511
ochond+1.462
 pollut+1.434
 acceler+1.418
OTOS+1.413
Neg logit contributions
util-1.637
onen-1.631
idium-1.583
hler-1.505
cone-1.475
Token on
Token ID319
Feature activation-5.01e-01

Pos logit contributions
util+0.353
onen+0.344
idium+0.337
hler+0.327
cone+0.320
Neg logit contributions
 axes-0.328
ochond-0.327
OTOS-0.311
opsis-0.311
 pollut-0.310
Token sleeve
Token ID23179
Feature activation-3.08e-01

Pos logit contributions
 axes+3.959
ochond+3.742
opsis+3.606
OTOS+3.603
 acceler+3.590
Neg logit contributions
onen-6.012
util-5.977
idium-5.730
hler-5.598
irtual-5.574
Token at
Token ID379
Feature activation-2.24e-01

Pos logit contributions
 axes+2.587
ochond+2.547
OTOS+2.524
opsis+2.466
 acceler+2.465
Neg logit contributions
util-3.464
idium-3.388
onen-3.377
cone-3.250
hler-3.234
Token basketball
Token ID9669
Feature activation-3.73e-01

Pos logit contributions
 axes+2.253
ochond+2.240
OTOS+2.138
 pollut+2.135
opsis+2.134
Neg logit contributions
util-2.200
onen-2.141
idium-2.099
hler-2.032
cone-1.981
Token game
Token ID983
Feature activation-2.89e-01

Pos logit contributions
 axes+3.418
ochond+3.315
 pollut+3.207
 acceler+3.191
OTOS+3.153
Neg logit contributions
util-4.020
onen-3.902
idium-3.867
hler-3.774
avorite-3.677
Token MORE
Token ID12011
Feature activation-2.81e-01

Pos logit contributions
 axes+2.302
ochond+2.271
OTOS+2.228
 acceler+2.200
 pollut+2.182
Neg logit contributions
util-3.401
idium-3.301
onen-3.257
irtual-3.173
avorite-3.143
Token and
Token ID290
Feature activation-3.11e-01

Pos logit contributions
 axes+2.404
ochond+2.344
OTOS+2.253
 acceler+2.251
 pollut+2.243
Neg logit contributions
util-3.174
onen-3.115
idium-3.069
hler-2.928
cone-2.927
Token Hillary
Token ID5073
Feature activation-2.01e-01

Pos logit contributions
 axes+2.748
ochond+2.666
 pollut+2.563
 acceler+2.562
OTOS+2.540
Neg logit contributions
util-3.428
onen-3.393
idium-3.326
hler-3.187
cone-3.146
Token don
Token ID836
Feature activation+1.86e-01

Pos logit contributions
util+0.205
onen+0.200
idium+0.196
hler+0.190
cone+0.186
Neg logit contributions
 axes-0.173
ochond-0.173
OTOS-0.164
opsis-0.164
 pollut-0.163
Token
Token ID447
Feature activation-1.60e-01

Pos logit contributions
util+1.550
onen+1.504
hler+1.399
idium+1.395
cone+1.317
Neg logit contributions
escription-2.200
 axes-2.149
ochond-2.137
 pollut-2.093
OTOS-2.069
Token
Token ID247
Feature activation-7.22e-02

Pos logit contributions
 axes+1.665
ochond+1.653
OTOS+1.584
opsis+1.581
 acceler+1.579
Neg logit contributions
util-1.505
onen-1.468
idium-1.432
hler-1.376
cone-1.350
Tokent
Token ID83
Feature activation+8.73e-02

Pos logit contributions
 axes+0.684
ochond+0.683
opsis+0.648
OTOS+0.646
 pollut+0.645
Neg logit contributions
util-0.746
onen-0.728
idium-0.713
hler-0.690
cone-0.676
Token short
Token ID1790
Feature activation-1.04e-01

Pos logit contributions
util+0.982
onen+0.955
idium+0.939
hler+0.916
cone+0.893
Neg logit contributions
ochond-0.748
 axes-0.746
OTOS-0.722
escription-0.708
opsis-0.704
Tokenchange
Token ID3803
Feature activation-7.44e-02

Pos logit contributions
 axes+0.983
ochond+0.979
 pollut+0.929
opsis+0.929
OTOS+0.928
Neg logit contributions
util-1.085
onen-1.060
idium-1.043
hler-1.010
cone-0.987
Token the
Token ID262
Feature activation+2.02e-01

Pos logit contributions
 axes+0.623
ochond+0.620
OTOS+0.586
opsis+0.584
 pollut+0.584
Neg logit contributions
util-0.854
onen-0.835
idium-0.821
hler-0.797
cone-0.782
Token race
Token ID3234
Feature activation+1.02e-01

Pos logit contributions
util+2.249
onen+2.225
idium+2.179
hler+2.124
cone+2.057
Neg logit contributions
ochond-1.707
 axes-1.673
opsis-1.636
OTOS-1.632
escription-1.629
Token.
Token ID13
Feature activation-2.19e-02

Pos logit contributions
util+1.077
onen+1.055
idium+1.030
hler+1.001
cone+0.978
Neg logit contributions
ochond-0.946
 axes-0.941
OTOS-0.904
 pollut-0.895
opsis-0.893
Token They
Token ID1119
Feature activation+3.26e-02

Pos logit contributions
 axes+0.203
ochond+0.202
OTOS+0.192
opsis+0.191
 pollut+0.191
Neg logit contributions
util-0.233
onen-0.228
idium-0.223
hler-0.216
cone-0.212
Token criticized
Token ID12318
Feature activation+1.56e-01

Pos logit contributions
util+0.350
onen+0.342
idium+0.335
hler+0.326
cone+0.318
Neg logit contributions
 axes-0.296
ochond-0.295
OTOS-0.282
opsis-0.280
 pollut-0.280
Token (
Token ID357
Feature activation+4.04e-01

Pos logit contributions
 axes+0.057
ochond+0.057
OTOS+0.054
opsis+0.054
 pollut+0.054
Neg logit contributions
util-0.062
onen-0.060
idium-0.059
hler-0.057
cone-0.056
Token31
Token ID3132
Feature activation+3.76e-01

Pos logit contributions
util+3.533
onen+3.340
idium+3.275
hler+3.181
cone+3.118
Neg logit contributions
ochond-4.445
 axes-4.406
 pollut-4.311
opsis-4.260
 acceler-4.230
Token)
Token ID8
Feature activation-5.20e-03

Pos logit contributions
util+4.071
onen+3.959
idium+3.899
hler+3.788
cone+3.700
Neg logit contributions
ochond-3.365
 axes-3.354
opsis-3.188
OTOS-3.187
 pollut-3.179
Token
Token ID198
Feature activation+3.50e-01

Pos logit contributions
 axes+0.046
ochond+0.045
OTOS+0.043
 acceler+0.043
opsis+0.043
Neg logit contributions
util-0.058
onen-0.057
idium-0.055
hler-0.054
cone-0.053
Token
Token ID198
Feature activation+1.00e-02

Pos logit contributions
util+2.980
onen+2.725
hler+2.583
idium+2.539
cone+2.517
Neg logit contributions
escription-4.501
 axes-3.996
ochond-3.996
 dilig-3.971
 pollut-3.953
TokenSave
Token ID16928
Feature activation-1.52e-01

Pos logit contributions
util+0.076
onen+0.074
idium+0.072
hler+0.069
cone+0.067
Neg logit contributions
 axes-0.123
ochond-0.123
opsis-0.118
 pollut-0.118
OTOS-0.118
Token a
Token ID257
Feature activation-5.78e-03

Pos logit contributions
 axes+1.403
ochond+1.394
OTOS+1.326
 pollut+1.322
 acceler+1.321
Neg logit contributions
util-1.619
onen-1.583
idium-1.549
hler-1.512
cone-1.475
Token turret
Token ID28101
Feature activation-1.45e-01

Pos logit contributions
 axes+0.052
ochond+0.052
OTOS+0.049
opsis+0.049
 pollut+0.049
Neg logit contributions
util-0.063
onen-0.061
idium-0.060
hler-0.058
cone-0.057
Token from
Token ID422
Feature activation-8.28e-02

Pos logit contributions
 axes+1.273
ochond+1.262
opsis+1.199
OTOS+1.199
 acceler+1.196
Neg logit contributions
util-1.598
onen-1.560
idium-1.533
hler-1.491
cone-1.455
Token redemption
Token ID26831
Feature activation-1.25e-01

Pos logit contributions
 axes+0.710
ochond+0.704
OTOS+0.667
 pollut+0.666
 acceler+0.666
Neg logit contributions
util-0.935
onen-0.915
idium-0.898
hler-0.874
cone-0.855
Token (
Token ID357
Feature activation+4.04e-01

Pos logit contributions
 axes+1.158
ochond+1.145
OTOS+1.091
opsis+1.091
 acceler+1.090
Neg logit contributions
util-1.322
onen-1.290
idium-1.263
hler-1.232
cone-1.200
Token said
Token ID531
Feature activation+1.41e-01

Pos logit contributions
util+0.888
onen+0.864
idium+0.846
hler+0.818
cone+0.798
Neg logit contributions
ochond-0.966
 axes-0.959
OTOS-0.922
escription-0.917
opsis-0.917
Token the
Token ID262
Feature activation+1.58e-01

Pos logit contributions
util+1.516
onen+1.475
idium+1.450
hler+1.413
cone+1.381
Neg logit contributions
ochond-1.285
 axes-1.279
opsis-1.216
OTOS-1.213
 pollut-1.209
Token board
Token ID3096
Feature activation-6.93e-02

Pos logit contributions
util+1.689
onen+1.645
idium+1.617
hler+1.574
cone+1.541
Neg logit contributions
ochond-1.443
 axes-1.418
opsis-1.369
OTOS-1.364
escription-1.356
Token's
Token ID338
Feature activation+1.16e-01

Pos logit contributions
 axes+0.629
ochond+0.627
OTOS+0.596
opsis+0.594
 pollut+0.593
Neg logit contributions
util-0.747
onen-0.729
idium-0.715
hler-0.694
cone-0.680
Token ruling
Token ID6922
Feature activation-1.29e-01

Pos logit contributions
util+1.202
onen+1.168
idium+1.150
hler+1.113
cone+1.086
Neg logit contributions
ochond-1.109
 axes-1.094
OTOS-1.056
opsis-1.050
 pollut-1.040
Token went
Token ID1816
Feature activation-6.43e-02

Pos logit contributions
 axes+1.137
ochond+1.126
 acceler+1.069
opsis+1.067
 pollut+1.067
Neg logit contributions
util-1.431
onen-1.398
idium-1.373
hler-1.332
cone-1.304
Token the
Token ID262
Feature activation-1.14e-01

Pos logit contributions
 axes+0.520
ochond+0.518
OTOS+0.489
opsis+0.488
 pollut+0.487
Neg logit contributions
util-0.756
onen-0.740
idium-0.727
hler-0.708
cone-0.694
Token right
Token ID826
Feature activation-1.27e-02

Pos logit contributions
 axes+1.056
ochond+1.035
 pollut+0.992
 acceler+0.990
OTOS+0.989
Neg logit contributions
util-1.214
onen-1.180
idium-1.162
hler-1.125
cone-1.097
Token way
Token ID835
Feature activation+1.61e-01

Pos logit contributions
ochond+0.115
 axes+0.114
OTOS+0.109
opsis+0.109
 pollut+0.108
Neg logit contributions
util-0.136
onen-0.134
idium-0.131
hler-0.127
cone-0.125
Token.
Token ID13
Feature activation-1.79e-02

Pos logit contributions
util+1.621
onen+1.582
idium+1.542
hler+1.500
cone+1.485
Neg logit contributions
ochond-1.561
 axes-1.547
OTOS-1.485
 pollut-1.475
opsis-1.471
Token He
Token ID679
Feature activation-1.21e-02

Pos logit contributions
 axes+0.149
ochond+0.149
OTOS+0.144
opsis+0.142
 acceler+0.140
Neg logit contributions
util-0.202
onen-0.198
idium-0.196
hler-0.189
avorite-0.187
TokenWarning
Token ID20361
Feature activation+3.15e-02

Pos logit contributions
ochond+0.752
 axes+0.748
OTOS+0.727
opsis+0.724
 pollut+0.720
Neg logit contributions
util-0.368
onen-0.361
idium-0.350
hler-0.332
cone-0.324
Token -
Token ID532
Feature activation+9.56e-02

Pos logit contributions
onen+0.324
util+0.322
idium+0.314
hler+0.308
cone+0.301
Neg logit contributions
ochond-0.300
 axes-0.298
 pollut-0.283
OTOS-0.283
opsis-0.282
TokenMessage
Token ID12837
Feature activation-1.96e-01

Pos logit contributions
util+0.885
onen+0.831
idium+0.814
hler+0.788
cone+0.757
Neg logit contributions
 axes-1.055
ochond-1.030
escription-1.017
 pollut-1.013
 acceler-1.011
Token "
Token ID366
Feature activation-2.18e-01

Pos logit contributions
ochond+1.674
 axes+1.645
escription+1.568
OTOS+1.560
opsis+1.557
Neg logit contributions
onen-2.151
util-2.123
idium-2.085
hler-2.046
cone-2.031
TokenF
Token ID37
Feature activation-2.07e-01

Pos logit contributions
 axes+2.376
ochond+2.326
OTOS+2.260
opsis+2.235
 acceler+2.228
Neg logit contributions
util-1.873
onen-1.863
idium-1.845
hler-1.765
cone-1.714
Tokenailed
Token ID6255
Feature activation-6.49e-01

Pos logit contributions
 axes+2.912
ochond+2.906
OTOS+2.812
 pollut+2.808
 acceler+2.804
Neg logit contributions
util-1.207
onen-1.151
idium-1.108
hler-1.046
cone-0.999
Token to
Token ID284
Feature activation-5.30e-01

Pos logit contributions
 axes+3.727
escription+3.412
 acceler+3.368
ochond+3.306
opsis+3.154
Neg logit contributions
onen-8.581
idium-8.554
ussen-8.472
util-8.113
hler-8.026
Token $
Token ID720
Feature activation-3.29e-01

Pos logit contributions
 axes+4.003
 acceler+3.620
 pollut+3.597
ochond+3.576
opsis+3.567
Neg logit contributions
idium-6.341
onen-6.223
util-6.186
ussen-6.007
hler-6.000
Token($
Token ID16763
Feature activation-4.03e-01

Pos logit contributions
 axes+2.919
ochond+2.896
opsis+2.807
OTOS+2.795
escription+2.742
Neg logit contributions
idium-3.406
util-3.405
onen-3.400
hler-3.224
ussen-3.193
Token_.
Token ID44807
Feature activation-2.01e-01

Pos logit contributions
 axes+4.903
ochond+4.873
opsis+4.784
OTOS+4.753
escription+4.687
Neg logit contributions
idium-2.775
onen-2.742
util-2.653
hler-2.510
ussen-2.498
TokenException
Token ID16922
Feature activation-1.73e-01

Pos logit contributions
 axes+2.120
ochond+2.090
OTOS+2.030
opsis+2.015
 acceler+1.986
Neg logit contributions
util-1.827
onen-1.802
idium-1.776
hler-1.708
ilogy-1.665
Token thought
Token ID1807
Feature activation-7.29e-02

Pos logit contributions
util+0.912
onen+0.878
idium+0.858
hler+0.845
cone+0.821
Neg logit contributions
ochond-0.868
 axes-0.847
OTOS-0.842
escription-0.830
opsis-0.822
Token he
Token ID339
Feature activation-6.39e-02

Pos logit contributions
 axes+0.642
ochond+0.638
OTOS+0.604
 pollut+0.602
opsis+0.602
Neg logit contributions
util-0.807
onen-0.789
idium-0.774
hler-0.752
cone-0.736
Token'd
Token ID1549
Feature activation-1.49e-01

Pos logit contributions
 axes+0.499
ochond+0.496
OTOS+0.466
opsis+0.466
 pollut+0.465
Neg logit contributions
util-0.771
onen-0.755
idium-0.742
hler-0.722
cone-0.709
Token be
Token ID307
Feature activation+1.02e-01

Pos logit contributions
 axes+1.115
ochond+1.105
 acceler+1.035
opsis+1.035
 pollut+1.034
Neg logit contributions
util-1.850
onen-1.816
idium-1.790
hler-1.736
cone-1.711
Token a
Token ID257
Feature activation+1.41e-01

Pos logit contributions
util+1.103
onen+1.077
idium+1.046
hler+1.033
cone+1.009
Neg logit contributions
ochond-0.924
 axes-0.909
OTOS-0.882
opsis-0.879
escription-0.869
Token hockey
Token ID12217
Feature activation-6.21e-01

Pos logit contributions
util+1.461
onen+1.439
idium+1.392
hler+1.368
cone+1.329
Neg logit contributions
ochond-1.322
 axes-1.313
OTOS-1.271
opsis-1.264
escription-1.242
Token dad
Token ID9955
Feature activation-2.07e-01

Pos logit contributions
 axes+4.249
ochond+4.108
 acceler+3.952
 pollut+3.928
OTOS+3.602
Neg logit contributions
util-7.772
idium-7.616
onen-7.616
irtual-7.315
ilogy-7.304
Token when
Token ID618
Feature activation-2.86e-01

Pos logit contributions
 axes+1.837
ochond+1.819
 acceler+1.729
opsis+1.728
OTOS+1.722
Neg logit contributions
util-2.261
onen-2.208
idium-2.174
hler-2.105
cone-2.067
Token his
Token ID465
Feature activation-5.75e-02

Pos logit contributions
 axes+2.330
ochond+2.237
 acceler+2.154
OTOS+2.131
 pollut+2.110
Neg logit contributions
util-3.367
onen-3.288
idium-3.220
hler-3.125
ilogy-3.093
Token kids
Token ID3988
Feature activation-1.16e-02

Pos logit contributions
 axes+0.528
ochond+0.526
OTOS+0.499
opsis+0.498
 pollut+0.498
Neg logit contributions
util-0.614
onen-0.599
idium-0.587
hler-0.570
cone-0.558
Token were
Token ID547
Feature activation+8.72e-02

Pos logit contributions
 axes+0.106
ochond+0.106
OTOS+0.101
opsis+0.100
 pollut+0.100
Neg logit contributions
util-0.125
onen-0.122
idium-0.120
hler-0.116
cone-0.114
Tokenconnection
Token ID38659
Feature activation+1.56e-02

Pos logit contributions
 axes+3.049
ochond+2.922
opsis+2.777
 pollut+2.700
 acceler+2.695
Neg logit contributions
util-6.422
onen-6.337
idium-6.269
avorite-6.195
hler-6.157
Token (
Token ID357
Feature activation-2.67e-01

Pos logit contributions
util+0.162
onen+0.158
idium+0.155
hler+0.150
cone+0.147
Neg logit contributions
 axes-0.149
ochond-0.148
OTOS-0.141
opsis-0.140
 pollut-0.140
TokenOS
Token ID2640
Feature activation-4.36e-02

Pos logit contributions
ochond+2.521
 axes+2.515
OTOS+2.457
opsis+2.395
 acceler+2.371
Neg logit contributions
util-2.713
idium-2.636
onen-2.630
hler-2.562
ilogy-2.521
TokenI
Token ID40
Feature activation-3.88e-01

Pos logit contributions
 axes+0.470
ochond+0.468
OTOS+0.448
opsis+0.447
 pollut+0.447
Neg logit contributions
util-0.397
onen-0.386
idium-0.377
hler-0.364
cone-0.355
Token)
Token ID8
Feature activation-1.15e-01

Pos logit contributions
 axes+3.248
ochond+3.076
 acceler+2.999
OTOS+2.932
 pollut+2.901
Neg logit contributions
util-4.382
onen-4.333
idium-4.275
avorite-4.204
hler-4.117
Token layers
Token ID11685
Feature activation-5.48e-01

Pos logit contributions
 axes+1.018
ochond+0.995
 acceler+0.949
 pollut+0.945
OTOS+0.939
Neg logit contributions
util-1.288
onen-1.261
idium-1.237
hler-1.203
cone-1.177
Token.
Token ID13
Feature activation-2.45e-04

Pos logit contributions
 axes+4.473
ochond+4.293
 acceler+4.212
opsis+4.054
 pollut+4.023
Neg logit contributions
util-6.115
onen-6.035
idium-5.967
ilogy-5.782
hler-5.731
Token Second
Token ID5498
Feature activation+1.18e-01

Pos logit contributions
 axes+0.002
ochond+0.002
OTOS+0.002
opsis+0.002
 pollut+0.002
Neg logit contributions
util-0.003
onen-0.003
idium-0.002
hler-0.002
cone-0.002
Token,
Token ID11
Feature activation+2.18e-01

Pos logit contributions
util+1.294
onen+1.264
idium+1.238
hler+1.206
cone+1.174
Neg logit contributions
 axes-1.068
ochond-1.063
OTOS-1.005
opsis-1.005
 acceler-1.003
Token attacks
Token ID3434
Feature activation+1.23e-01

Pos logit contributions
util+2.311
onen+2.256
idium+2.200
hler+2.129
cone+2.098
Neg logit contributions
ochond-1.983
 axes-1.962
escription-1.896
opsis-1.889
OTOS-1.872
Token from
Token ID422
Feature activation+6.90e-02

Pos logit contributions
util+1.349
onen+1.314
idium+1.284
hler+1.254
cone+1.234
Neg logit contributions
ochond-1.086
 axes-1.085
OTOS-1.046
opsis-1.031
 pollut-1.025
Token'
Token ID6
Feature activation+1.78e-01

Pos logit contributions
util+2.093
idium+2.072
hler+1.947
cone+1.937
ussen+1.929
Neg logit contributions
ochond-2.821
OTOS-2.793
escription-2.694
 axes-2.660
opsis-2.577
Token 2016
Token ID1584
Feature activation-1.14e-01

Pos logit contributions
util+1.949
onen+1.904
idium+1.868
hler+1.814
cone+1.778
Neg logit contributions
 axes-1.584
ochond-1.579
OTOS-1.498
opsis-1.494
 pollut-1.491
Token primary
Token ID4165
Feature activation-2.45e-01

Pos logit contributions
 axes+0.878
 pollut+0.846
 acceler+0.836
ochond+0.813
opsis+0.808
Neg logit contributions
util-1.409
onen-1.368
idium-1.342
hler-1.319
avorite-1.306
Token Sanders
Token ID5831
Feature activation-1.81e-02

Pos logit contributions
 axes+2.347
 acceler+2.210
opsis+2.176
 pollut+2.163
ochond+2.159
Neg logit contributions
util-2.586
onen-2.533
idium-2.429
irtual-2.363
hler-2.338
Token:
Token ID25
Feature activation-1.43e-01

Pos logit contributions
 axes+0.143
ochond+0.137
 acceler+0.134
 pollut+0.133
OTOS+0.131
Neg logit contributions
util-0.218
onen-0.214
idium-0.209
hler-0.206
cone-0.201
Token '
Token ID705
Feature activation-6.06e-01

Pos logit contributions
 axes+0.948
 acceler+0.911
ochond+0.896
 pollut+0.873
opsis+0.871
Neg logit contributions
util-1.904
onen-1.866
idium-1.818
cone-1.798
avorite-1.787
TokenDamn
Token ID43343
Feature activation+4.11e-03

Pos logit contributions
 axes+5.190
ochond+4.640
 acceler+4.589
OTOS+4.382
ait+4.355
Neg logit contributions
avorite-7.026
util-6.871
onen-6.643
unia-6.457
idium-6.345
Token right
Token ID826
Feature activation+6.18e-02

Pos logit contributions
util+0.045
onen+0.044
idium+0.044
hler+0.042
cone+0.042
Neg logit contributions
 axes-0.037
ochond-0.036
 pollut-0.034
opsis-0.034
 acceler-0.034
Token'
Token ID6
Feature activation-1.04e-01

Pos logit contributions
util+0.689
onen+0.676
idium+0.661
hler+0.644
cone+0.630
Neg logit contributions
 axes-0.535
ochond-0.534
OTOS-0.507
opsis-0.504
 pollut-0.502
Token I
Token ID314
Feature activation-3.62e-01

Pos logit contributions
 axes+0.814
ochond+0.799
 acceler+0.787
OTOS+0.777
 pollut+0.766
Neg logit contributions
util-1.250
onen-1.224
idium-1.202
avorite-1.171
cone-1.165
Token'll
Token ID1183
Feature activation-2.54e-01

Pos logit contributions
 axes+3.087
ochond+2.986
 acceler+2.858
 pollut+2.834
opsis+2.774
Neg logit contributions
onen-4.149
util-4.013
idium-3.933
irtual-3.846
hler-3.813
TokenT
Token ID51
Feature activation-2.71e-02

Pos logit contributions
 axes+2.690
ochond+2.682
OTOS+2.567
escription+2.558
 acceler+2.548
Neg logit contributions
util-2.264
onen-2.249
idium-2.191
hler-2.113
ussen-2.102
Token(
Token ID7
Feature activation-2.31e-01

Pos logit contributions
 axes+0.231
ochond+0.228
 pollut+0.216
OTOS+0.216
 acceler+0.216
Neg logit contributions
util-0.303
onen-0.301
idium-0.297
hler-0.288
cone-0.280
Tokent
Token ID83
Feature activation+8.82e-02

Pos logit contributions
 axes+2.253
ochond+2.243
OTOS+2.142
opsis+2.135
 pollut+2.133
Neg logit contributions
util-2.347
onen-2.283
idium-2.237
hler-2.168
cone-2.121
Token)-
Token ID13219
Feature activation-1.96e-01

Pos logit contributions
util+0.945
onen+0.934
idium+0.918
hler+0.887
cone+0.869
Neg logit contributions
ochond-0.804
 axes-0.803
opsis-0.755
OTOS-0.749
escription-0.748
Token T
Token ID309
Feature activation-2.88e-01

Pos logit contributions
 axes+1.525
ochond+1.521
OTOS+1.457
 acceler+1.435
 pollut+1.435
Neg logit contributions
util-2.295
onen-2.281
idium-2.226
hler-2.168
cone-2.148
Token 0
Token ID657
Feature activation-7.07e-01

Pos logit contributions
 axes+2.003
ochond+1.911
 acceler+1.849
OTOS+1.838
 pollut+1.835
Neg logit contributions
util-3.603
onen-3.580
idium-3.503
hler-3.444
cone-3.386
Token )
Token ID1267
Feature activation-5.19e-01

Pos logit contributions
 axes+2.650
ochond+2.411
 acceler+2.224
 pollut+2.109
OTOS+2.049
Neg logit contributions
onen-10.717
util-10.278
idium-10.169
cone-9.966
illance-9.943
Token
Token ID198
Feature activation-2.60e-01

Pos logit contributions
 axes+4.391
ochond+4.233
 acceler+4.148
 pollut+4.090
opsis+3.980
Neg logit contributions
onen-5.676
util-5.653
idium-5.516
hler-5.340
avorite-5.265
Token
Token ID198
Feature activation-4.25e-01

Pos logit contributions
 axes+2.312
ochond+2.303
OTOS+2.186
opsis+2.181
 pollut+2.174
Neg logit contributions
util-2.836
onen-2.773
idium-2.725
hler-2.642
cone-2.589
TokenThe
Token ID464
Feature activation-2.03e-01

Pos logit contributions
 axes+3.904
ochond+3.883
OTOS+3.700
 pollut+3.685
opsis+3.677
Neg logit contributions
util-4.240
onen-4.228
idium-4.158
avorite-4.021
hler-4.008
Token year
Token ID614
Feature activation-3.72e-01

Pos logit contributions
 axes+1.780
ochond+1.752
 acceler+1.677
 pollut+1.677
opsis+1.652
Neg logit contributions
util-2.269
onen-2.207
idium-2.175
hler-2.111
cone-2.070