TOP ACTIVATIONS
MAX = 5.236

encia ) could be re ph r ased as
e , k omo i hol a ka make make i
because they belonged to a culture within the Atlanta Hawks where
goes down to the cu ph old ers . Open your
, a k uko i hol a o
had on the web a decade ago , and now it
s eu ph oric effects are
to be more of a socio - economic construct than a
says she has been a travel and food writer for 22
high heat . Place f rit ters in the skillet (
When Trudeau chose St é ph ane Dion as minister of
mirrors for primates , k ong toys and pr imate c
W ether s Mysteries , comic book short stories , prose
: " Supporters are the life blood of any football club
thinks of when you mention Psych o , and with good
W ether s Mysteries , comic book short stories , prose
MORE as a 68 - year - old woman 's cause
was forced to write a letter explaining her actions in school
call it a ph o ,
this query to the cl ojure . files table SELECT *

MIN ACTIVATIONS
MIN = -5.444

F ( Low Carb High Fat ) diet improves the body
3 . THE BIG SH ORT People inside
Emirates and Palestine in Group A . Here , the most
with no strings attached . The hope is that
, Arsenal , Liverpool and Tottenham Hots pur stating the
Arsenal , Liverpool and Tottenham Hots pur stating the existing
conscious mind and your Inner Ass hole that you are actually
s work on the Space Launch System heavy - lift rocket
knowing exactly what the future holds . It was cruel ,
Earth far more than expected . This shows that the
sermon entitled ' Rem aining Aw ake through a Great Revolution
, but as you can see from the photos , M
the chaos ( sp oiler alert ) that Black inadvertently sets
F TC Compl aint Assistant to report persons
competed alongside Chelsea and Tottenham Hots pur for the Premier League
( contin ued ). Extreme Un ction . Holy Orders .
industrial output rose more than expected in September , boosted by
handling of Mam ad ou Sak ho drugs test Read more
first Thermal Disc our agement Beam test ( 2 )
third Grav itational L ensing Accuracy Testing Challenge ( GRE AT

INTERVAL 2.566 - 5.236
CONTAINS 0.691%

warp drive website for your business and engage your customers .
child . But eight - year - olds who aspire to
1800 . Learn about the history of batteries and find out
. dollars from rem itt ances , tourism and pal ad
with Bernie Sanders Newt Gingrich Former Speaker of the House Newt

INTERVAL -0.104 - 2.566
CONTAINS 53.520%

s obviously not impossible that this could happen again .
talk about Supreme to someone who isn t interested
and the DOJ claimed , flat out , that Kam al
a leg amput ation who was volunteering at the store at
1 and 2 were listed on Y AL SA 's "

INTERVAL -2.774 - -0.104
CONTAINS 45.438%

. A great example of this is Robot Entertainment : in
silence . It s enough to trick me into
when the initial space is nearly cub oid , for example
. S BI Wire x Asia will deliver a
technical fault that can be corrected . " The

INTERVAL -5.444 - -2.774
CONTAINS 0.352%

remove . MS I exclusive Un locked Digital Power is composed
click on Vide opl ayer Input Stream . Make sure that
competed alongside Chelsea and Tottenham Hots pur for the Premier League
both Shirt Arm and Training Wear partner . The
as well as the previously announced nominees ) is below .
Tokenencia
Token ID29634
Feature activation-1.36e+00

Pos logit contributions
IELD+3.377
hips+3.251
 Desire+3.079
ymm+3.061
 suppose+2.862
Neg logit contributions
umn-3.849
cient-3.661
anwhile-3.540
isode-3.470
arch-3.428
Token)
Token ID8
Feature activation-1.45e-01

Pos logit contributions
IELD+2.429
hips+2.318
 Desire+2.201
ymm+2.194
 suppose+2.052
Neg logit contributions
umn-2.784
cient-2.651
anwhile-2.529
isode-2.496
arch-2.491
Token could
Token ID714
Feature activation+9.91e-01

Pos logit contributions
IELD+0.268
hips+0.254
 Desire+0.242
ymm+0.242
 suppose+0.225
Neg logit contributions
umn-0.291
cient-0.273
anwhile-0.263
arch-0.259
isode-0.258
Token be
Token ID307
Feature activation-4.83e-01

Pos logit contributions
umn+1.923
cient+1.820
anwhile+1.744
isode+1.709
arch+1.707
Neg logit contributions
IELD-1.873
hips-1.737
 Desire-1.702
ymm-1.674
 Darling-1.548
Token re
Token ID302
Feature activation+1.21e+00

Pos logit contributions
IELD+0.856
hips+0.808
 Desire+0.774
ymm+0.758
 suppose+0.705
Neg logit contributions
umn-1.001
cient-0.954
anwhile-0.911
arch-0.899
isode-0.891
Tokenph
Token ID746
Feature activation+5.24e+00

Pos logit contributions
umn+1.766
cient+1.633
arch+1.538
anwhile+1.490
isode+1.475
Neg logit contributions
IELD-2.905
hips-2.744
 Desire-2.675
ymm-2.608
 suppose-2.494
Tokenr
Token ID81
Feature activation+1.54e+00

Pos logit contributions
umn+5.982
cient+5.568
arch+5.398
isode+5.155
iously+4.961
Neg logit contributions
IELD-13.108
hips-12.531
WARE-12.148
 Desire-11.947
 Darling-11.548
Tokenased
Token ID839
Feature activation+1.53e+00

Pos logit contributions
umn+1.163
cient+0.957
arch+0.868
isode+0.787
ophone+0.702
Neg logit contributions
IELD-4.903
hips-4.733
 Desire-4.578
ymm-4.507
WARE-4.362
Token as
Token ID355
Feature activation-3.58e-01

Pos logit contributions
umn+3.173
cient+2.998
arch+2.894
anwhile+2.842
isode+2.755
Neg logit contributions
IELD-2.678
hips-2.472
ymm-2.372
 Desire-2.335
 suppose-2.162
Token
Token ID564
Feature activation+5.28e-01

Pos logit contributions
IELD+0.585
hips+0.551
 Desire+0.518
ymm+0.517
 suppose+0.482
Neg logit contributions
umn-0.790
cient-0.750
anwhile-0.723
arch-0.712
isode-0.711
Token
Token ID246
Feature activation+1.28e+00

Pos logit contributions
umn+1.033
cient+0.991
arch+0.925
anwhile+0.923
isode+0.905
Neg logit contributions
IELD-0.981
hips-0.936
 Desire-0.898
ymm-0.886
 suppose-0.845
Tokene
Token ID68
Feature activation+1.86e+00

Pos logit contributions
umn+2.009
cient+1.846
arch+1.719
isode+1.681
ophone+1.670
Neg logit contributions
IELD-3.294
hips-3.099
 Desire-3.073
ymm-2.993
 suppose-2.913
Token,
Token ID11
Feature activation+4.56e-01

Pos logit contributions
umn+3.913
cient+3.687
anwhile+3.534
arch+3.483
isode+3.482
Neg logit contributions
IELD-3.191
hips-2.908
 Desire-2.876
ymm-2.825
 suppose-2.576
Token k
Token ID479
Feature activation+1.61e-01

Pos logit contributions
umn+0.974
cient+0.923
anwhile+0.877
isode+0.865
arch+0.864
Neg logit contributions
IELD-0.778
hips-0.721
 Desire-0.699
ymm-0.692
 suppose-0.635
Tokenomo
Token ID17902
Feature activation+3.69e+00

Pos logit contributions
umn+0.277
cient+0.258
anwhile+0.243
arch+0.242
isode+0.240
Neg logit contributions
IELD-0.348
hips-0.328
 Desire-0.322
ymm-0.312
 suppose-0.300
Token i
Token ID1312
Feature activation+3.06e-01

Pos logit contributions
umn+7.294
cient+6.872
arch+6.498
isode+6.409
ophone+6.366
Neg logit contributions
IELD-6.325
 Desire-5.592
ymm-5.554
hips-5.546
ENTION-5.063
Tokenhol
Token ID3937
Feature activation+4.80e+00

Pos logit contributions
umn+0.614
cient+0.579
anwhile+0.549
arch+0.548
isode+0.544
Neg logit contributions
IELD-0.565
hips-0.527
 Desire-0.515
ymm-0.506
 suppose-0.471
Tokena
Token ID64
Feature activation+1.06e+00

Pos logit contributions
umn+7.374
cient+6.397
arch+6.200
ophone+6.090
isode+5.950
Neg logit contributions
IELD-10.298
 Desire-9.213
hips-9.211
ymm-9.029
WARE-8.885
Token ka
Token ID38387
Feature activation+1.45e+00

Pos logit contributions
umn+2.168
cient+2.056
anwhile+1.938
arch+1.925
isode+1.924
Neg logit contributions
IELD-1.887
hips-1.716
ymm-1.685
 Desire-1.683
 suppose-1.536
Token make
Token ID787
Feature activation+6.92e-01

Pos logit contributions
umn+3.041
cient+2.898
isode+2.719
anwhile+2.711
arch+2.701
Neg logit contributions
IELD-2.468
hips-2.265
ymm-2.200
 Desire-2.182
 suppose-2.001
Tokenmake
Token ID15883
Feature activation+8.78e-01

Pos logit contributions
umn+1.374
cient+1.308
isode+1.260
anwhile+1.259
arch+1.238
Neg logit contributions
IELD-1.270
hips-1.195
 Desire-1.143
ymm-1.135
 suppose-1.054
Token i
Token ID1312
Feature activation-1.61e-01

Pos logit contributions
umn+1.881
cient+1.787
anwhile+1.688
arch+1.683
isode+1.680
Neg logit contributions
IELD-1.478
hips-1.361
 Desire-1.315
ymm-1.310
 suppose-1.197
Token because
Token ID780
Feature activation+2.26e+00

Pos logit contributions
umn+0.081
cient+0.079
anwhile+0.074
arch+0.074
isode+0.073
Neg logit contributions
IELD-0.055
hips-0.050
ymm-0.047
 Desire-0.047
 suppose-0.043
Token they
Token ID484
Feature activation+3.65e-01

Pos logit contributions
umn+3.938
cient+3.665
anwhile+3.539
arch+3.451
iously+3.361
Neg logit contributions
IELD-4.467
hips-4.195
ymm-4.031
 Desire-3.960
GV-3.745
Token belonged
Token ID19611
Feature activation+5.90e-01

Pos logit contributions
umn+0.723
cient+0.691
anwhile+0.654
isode+0.646
arch+0.644
Neg logit contributions
IELD-0.676
hips-0.631
 Desire-0.612
ymm-0.607
 suppose-0.559
Token to
Token ID284
Feature activation+9.20e-01

Pos logit contributions
umn+1.143
cient+1.096
arch+1.018
anwhile+1.001
isode+0.999
Neg logit contributions
IELD-1.129
hips-1.034
 Desire-1.004
ymm-0.998
 suppose-0.931
Token a
Token ID257
Feature activation+1.78e-01

Pos logit contributions
umn+2.028
cient+1.922
anwhile+1.858
arch+1.838
isode+1.793
Neg logit contributions
IELD-1.444
hips-1.366
ymm-1.292
 Desire-1.255
 suppose-1.185
Token culture
Token ID3968
Feature activation+4.74e+00

Pos logit contributions
umn+0.359
cient+0.343
anwhile+0.324
arch+0.324
isode+0.318
Neg logit contributions
IELD-0.319
hips-0.304
ymm-0.287
 Desire-0.283
 suppose-0.263
Token within
Token ID1626
Feature activation+1.28e+00

Pos logit contributions
umn+8.036
cient+7.924
arch+7.228
ophone+7.199
anwhile+7.054
Neg logit contributions
IELD-9.099
ymm-7.760
 Desire-7.554
hips-7.524
ELF-7.123
Token the
Token ID262
Feature activation+7.92e-01

Pos logit contributions
umn+2.688
cient+2.512
arch+2.402
anwhile+2.400
ophone+2.376
Neg logit contributions
IELD-2.179
hips-2.013
ymm-1.951
 Desire-1.874
 suppose-1.825
Token Atlanta
Token ID9371
Feature activation-2.07e+00

Pos logit contributions
umn+1.608
cient+1.517
arch+1.445
anwhile+1.438
ophone+1.402
Neg logit contributions
IELD-1.407
hips-1.336
ymm-1.277
 Desire-1.246
 suppose-1.192
Token Hawks
Token ID23445
Feature activation-3.46e+00

Pos logit contributions
IELD+3.370
hips+3.211
 Desire+3.040
ymm+3.008
 suppose+2.802
Neg logit contributions
umn-4.541
cient-4.341
anwhile-4.177
isode-4.120
arch-4.097
Token where
Token ID810
Feature activation+9.69e-01

Pos logit contributions
IELD+5.463
hips+5.344
 Desire+4.989
ymm+4.900
 suppose+4.601
Neg logit contributions
umn-7.524
cient-7.196
anwhile-7.014
isode-6.892
arch-6.747
Token goes
Token ID2925
Feature activation+7.21e-01

Pos logit contributions
umn+1.997
cient+1.937
anwhile+1.857
isode+1.819
arch+1.815
Neg logit contributions
IELD-1.501
hips-1.390
 Desire-1.338
ymm-1.318
 Darling-1.175
Token down
Token ID866
Feature activation+5.15e-01

Pos logit contributions
umn+1.457
cient+1.390
anwhile+1.338
arch+1.307
isode+1.300
Neg logit contributions
IELD-1.309
hips-1.214
 Desire-1.193
ymm-1.150
 suppose-1.065
Token to
Token ID284
Feature activation-2.00e-03

Pos logit contributions
umn+1.082
cient+1.037
anwhile+0.992
isode+0.982
arch+0.978
Neg logit contributions
IELD-0.877
hips-0.824
 Desire-0.803
ymm-0.782
 suppose-0.726
Token the
Token ID262
Feature activation+4.09e-01

Pos logit contributions
IELD+0.003
hips+0.003
 Desire+0.003
ymm+0.003
 suppose+0.002
Neg logit contributions
umn-0.004
cient-0.004
anwhile-0.004
arch-0.004
isode-0.004
Token cu
Token ID18912
Feature activation+1.02e+00

Pos logit contributions
umn+0.842
cient+0.807
anwhile+0.769
isode+0.760
arch+0.757
Neg logit contributions
IELD-0.724
hips-0.679
 Desire-0.644
ymm-0.644
 suppose-0.599
Tokenph
Token ID746
Feature activation+4.68e+00

Pos logit contributions
umn+1.471
cient+1.370
arch+1.271
anwhile+1.259
isode+1.251
Neg logit contributions
IELD-2.468
hips-2.354
 Desire-2.278
ymm-2.226
 suppose-2.112
Tokenold
Token ID727
Feature activation+1.37e+00

Pos logit contributions
umn+8.310
cient+7.922
ophone+7.746
arch+7.557
isode+7.457
Neg logit contributions
IELD-8.472
 Desire-7.902
hips-7.795
GV-7.374
ymm-7.281
Tokeners
Token ID364
Feature activation+1.21e+00

Pos logit contributions
umn+2.404
cient+2.273
arch+2.145
isode+2.117
anwhile+2.108
Neg logit contributions
IELD-2.799
hips-2.650
 Desire-2.559
ymm-2.527
 Darling-2.364
Token.
Token ID13
Feature activation-9.66e-02

Pos logit contributions
umn+2.381
cient+2.322
isode+2.187
arch+2.180
anwhile+2.171
Neg logit contributions
IELD-2.186
 Desire-1.954
hips-1.930
ymm-1.916
 suppose-1.764
TokenOpen
Token ID11505
Feature activation+3.71e-01

Pos logit contributions
IELD+0.147
hips+0.139
 Desire+0.131
ymm+0.131
 suppose+0.120
Neg logit contributions
umn-0.220
cient-0.211
anwhile-0.202
isode-0.200
arch-0.200
Token your
Token ID534
Feature activation+6.14e-02

Pos logit contributions
umn+0.823
cient+0.775
anwhile+0.753
arch+0.753
isode+0.751
Neg logit contributions
IELD-0.593
hips-0.551
ymm-0.521
 Desire-0.518
 suppose-0.483
Token,
Token ID11
Feature activation+5.04e-01

Pos logit contributions
umn+0.394
cient+0.372
anwhile+0.353
arch+0.352
isode+0.350
Neg logit contributions
IELD-0.336
hips-0.310
 Desire-0.301
ymm-0.299
 suppose-0.278
Token a
Token ID257
Feature activation+1.18e-01

Pos logit contributions
umn+1.065
cient+1.008
anwhile+0.958
isode+0.945
arch+0.945
Neg logit contributions
IELD-0.868
hips-0.805
 Desire-0.779
ymm-0.776
 suppose-0.712
Token k
Token ID479
Feature activation+2.33e-01

Pos logit contributions
umn+0.246
cient+0.234
anwhile+0.221
isode+0.220
arch+0.219
Neg logit contributions
IELD-0.209
hips-0.195
 Desire-0.188
ymm-0.186
 suppose-0.173
Tokenuko
Token ID29794
Feature activation+1.57e+00

Pos logit contributions
umn+0.398
cient+0.372
arch+0.348
anwhile+0.346
isode+0.344
Neg logit contributions
IELD-0.507
hips-0.477
 Desire-0.468
ymm-0.453
 suppose-0.438
Token i
Token ID1312
Feature activation+5.52e-01

Pos logit contributions
umn+3.248
cient+3.074
arch+2.888
anwhile+2.879
isode+2.867
Neg logit contributions
IELD-2.764
hips-2.526
 Desire-2.487
ymm-2.454
 suppose-2.251
Tokenhol
Token ID3937
Feature activation+4.63e+00

Pos logit contributions
umn+1.104
cient+1.041
anwhile+0.988
arch+0.984
isode+0.976
Neg logit contributions
IELD-1.014
hips-0.945
 Desire-0.922
ymm-0.908
 suppose-0.838
Tokena
Token ID64
Feature activation+8.52e-01

Pos logit contributions
umn+8.316
cient+7.476
ophone+7.133
arch+7.099
isode+6.901
Neg logit contributions
IELD-8.819
 Desire-7.838
hips-7.827
ymm-7.657
WARE-7.401
Token
Token ID220
Feature activation+8.26e-01

Pos logit contributions
umn+1.798
cient+1.698
anwhile+1.604
arch+1.598
isode+1.594
Neg logit contributions
IELD-1.482
hips-1.360
ymm-1.324
 Desire-1.324
 suppose-1.209
Token
Token ID134
Feature activation+6.95e-01

Pos logit contributions
umn+0.897
cient+0.798
arch+0.710
isode+0.693
ophone+0.690
Neg logit contributions
IELD-2.265
hips-2.184
 Desire-2.152
ymm-2.119
 suppose-2.037
Token
Token ID119
Feature activation+1.33e+00

Pos logit contributions
umn+1.534
cient+1.457
arch+1.374
isode+1.365
anwhile+1.361
Neg logit contributions
IELD-1.141
hips-1.077
 Desire-1.013
ymm-0.997
 suppose-0.933
Tokeno
Token ID78
Feature activation+1.69e+00

Pos logit contributions
umn+2.136
cient+1.970
arch+1.848
isode+1.803
ophone+1.799
Neg logit contributions
IELD-3.006
hips-2.822
 Desire-2.814
ymm-2.716
 suppose-2.645
Token had
Token ID550
Feature activation-1.29e-02

Pos logit contributions
umn+1.696
cient+1.629
anwhile+1.534
arch+1.531
isode+1.509
Neg logit contributions
IELD-1.513
hips-1.442
ymm-1.370
 Desire-1.357
 suppose-1.255
Token on
Token ID319
Feature activation-9.49e-02

Pos logit contributions
IELD+0.022
hips+0.020
 Desire+0.019
ymm+0.019
 suppose+0.017
Neg logit contributions
umn-0.028
cient-0.027
arch-0.025
anwhile-0.025
isode-0.025
Token the
Token ID262
Feature activation+9.12e-01

Pos logit contributions
IELD+0.156
hips+0.148
ymm+0.140
 Desire+0.138
 suppose+0.127
Neg logit contributions
umn-0.207
cient-0.198
anwhile-0.188
arch-0.187
ophone-0.184
Token web
Token ID3992
Feature activation+2.09e+00

Pos logit contributions
umn+1.701
cient+1.628
arch+1.531
anwhile+1.524
ophone+1.480
Neg logit contributions
IELD-1.746
hips-1.693
ymm-1.616
 Desire-1.564
 suppose-1.499
Token a
Token ID257
Feature activation-1.70e-01

Pos logit contributions
umn+4.059
cient+3.923
arch+3.811
ophone+3.738
isode+3.710
Neg logit contributions
IELD-3.706
hips-3.557
ymm-3.353
 Desire-3.344
 suppose-3.075
Token decade
Token ID5707
Feature activation+4.58e+00

Pos logit contributions
IELD+0.358
hips+0.344
 Desire+0.330
ymm+0.328
 suppose+0.309
Neg logit contributions
umn-0.291
cient-0.277
arch-0.258
anwhile-0.257
isode-0.256
Token ago
Token ID2084
Feature activation+1.62e+00

Pos logit contributions
umn+4.719
cient+4.288
arch+3.927
iously+3.730
isode+3.536
Neg logit contributions
IELD-11.829
ymm-10.802
 Desire-10.708
hips-10.586
GV-10.236
Token,
Token ID11
Feature activation-3.31e-02

Pos logit contributions
umn+3.106
cient+3.034
arch+2.810
ophone+2.781
isode+2.757
Neg logit contributions
IELD-3.069
hips-2.707
 Desire-2.659
ymm-2.658
 suppose-2.427
Token and
Token ID290
Feature activation+1.09e+00

Pos logit contributions
IELD+0.056
hips+0.053
ymm+0.050
 Desire+0.050
 suppose+0.046
Neg logit contributions
umn-0.071
cient-0.067
anwhile-0.065
arch-0.063
isode-0.063
Token now
Token ID783
Feature activation-1.06e+00

Pos logit contributions
umn+2.205
cient+2.112
anwhile+2.035
arch+1.991
ophone+1.967
Neg logit contributions
IELD-1.929
hips-1.774
ymm-1.684
 Desire-1.678
ELF-1.555
Token it
Token ID340
Feature activation+1.68e-01

Pos logit contributions
IELD+1.831
hips+1.735
 Desire+1.648
ymm+1.638
 suppose+1.517
Neg logit contributions
umn-2.222
cient-2.125
anwhile-2.040
arch-2.003
isode-1.999
Token
Token ID247
Feature activation-3.84e-01

Pos logit contributions
umn+1.579
cient+1.497
arch+1.396
anwhile+1.352
isode+1.342
Neg logit contributions
IELD-1.442
hips-1.363
 Desire-1.336
ymm-1.322
 suppose-1.241
Tokens
Token ID82
Feature activation-9.17e-01

Pos logit contributions
IELD+0.751
hips+0.704
 Desire+0.684
ymm+0.676
 suppose+0.638
Neg logit contributions
umn-0.728
cient-0.692
anwhile-0.656
arch-0.645
isode-0.643
Token
Token ID564
Feature activation-8.96e-02

Pos logit contributions
IELD+1.508
hips+1.422
ymm+1.347
 Desire+1.344
 suppose+1.206
Neg logit contributions
umn-2.011
cient-1.935
anwhile-1.844
arch-1.827
isode-1.807
Token
Token ID250
Feature activation+8.88e-02

Pos logit contributions
IELD+0.154
hips+0.148
ymm+0.141
 Desire+0.141
 suppose+0.130
Neg logit contributions
umn-0.188
cient-0.180
arch-0.169
anwhile-0.166
isode-0.166
Tokeneu
Token ID12496
Feature activation+1.99e+00

Pos logit contributions
umn+0.157
cient+0.151
anwhile+0.140
arch+0.140
isode+0.137
Neg logit contributions
IELD-0.183
hips-0.174
 Desire-0.169
ymm-0.166
 suppose-0.157
Tokenph
Token ID746
Feature activation+4.56e+00

Pos logit contributions
umn+3.221
cient+3.064
arch+2.859
anwhile+2.854
isode+2.778
Neg logit contributions
IELD-4.404
hips-4.231
 Desire-3.998
ymm-3.873
 suppose-3.655
Tokenoric
Token ID8146
Feature activation+1.01e+00

Pos logit contributions
umn+5.274
cient+4.858
arch+4.732
isode+4.421
ophone+4.356
Neg logit contributions
IELD-11.349
hips-10.600
 Desire-10.511
ymm-10.046
GV-9.908
Token
Token ID447
Feature activation-2.36e-01

Pos logit contributions
umn+2.040
cient+1.962
anwhile+1.845
arch+1.831
isode+1.806
Neg logit contributions
IELD-1.796
hips-1.719
ymm-1.641
 Desire-1.601
 suppose-1.466
Token
Token ID251
Feature activation-3.68e-01

Pos logit contributions
IELD+0.417
hips+0.396
ymm+0.380
 Desire+0.379
 suppose+0.353
Neg logit contributions
umn-0.489
cient-0.465
arch-0.437
anwhile-0.437
isode-0.430
Token effects
Token ID3048
Feature activation-9.23e-01

Pos logit contributions
IELD+0.655
hips+0.616
 Desire+0.588
ymm+0.585
 suppose+0.540
Neg logit contributions
umn-0.758
cient-0.727
anwhile-0.694
arch-0.682
isode-0.681
Token are
Token ID389
Feature activation-2.85e-02

Pos logit contributions
IELD+1.609
hips+1.511
 Desire+1.455
ymm+1.444
 suppose+1.324
Neg logit contributions
umn-1.929
cient-1.856
anwhile-1.762
isode-1.733
arch-1.731
Token to
Token ID284
Feature activation-1.23e+00

Pos logit contributions
umn+0.154
cient+0.151
arch+0.142
anwhile+0.139
isode+0.138
Neg logit contributions
IELD-0.142
hips-0.132
 Desire-0.129
ymm-0.127
 suppose-0.116
Token be
Token ID307
Feature activation-1.51e+00

Pos logit contributions
IELD+2.192
hips+2.092
 Desire+1.995
ymm+1.973
 suppose+1.846
Neg logit contributions
umn-2.516
cient-2.402
anwhile-2.297
isode-2.266
arch-2.251
Token more
Token ID517
Feature activation-4.72e-02

Pos logit contributions
IELD+2.398
hips+2.269
 Desire+2.156
ymm+2.123
 suppose+1.965
Neg logit contributions
umn-3.390
cient-3.260
anwhile-3.121
isode-3.083
arch-3.076
Token of
Token ID286
Feature activation-1.39e-01

Pos logit contributions
IELD+0.081
hips+0.075
 Desire+0.073
ymm+0.072
 suppose+0.066
Neg logit contributions
umn-0.098
cient-0.097
arch-0.091
anwhile-0.089
ophone-0.088
Token a
Token ID257
Feature activation-1.26e+00

Pos logit contributions
IELD+0.175
hips+0.164
ymm+0.152
 Desire+0.151
 suppose+0.135
Neg logit contributions
umn-0.352
cient-0.343
anwhile-0.325
arch-0.325
isode-0.324
Token socio
Token ID21790
Feature activation+4.55e+00

Pos logit contributions
IELD+2.337
hips+2.236
 Desire+2.126
ymm+2.117
 suppose+1.974
Neg logit contributions
umn-2.480
cient-2.369
anwhile-2.254
isode-2.227
arch-2.219
Token-
Token ID12
Feature activation+4.55e-01

Pos logit contributions
cient+5.942
umn+5.852
arch+5.617
isode+5.065
ocide+4.936
Neg logit contributions
IELD-10.835
hips-10.194
 Desire-9.880
ymm-9.653
 suppose-9.334
Tokeneconomic
Token ID17079
Feature activation+2.43e+00

Pos logit contributions
umn+0.407
cient+0.380
arch+0.329
isode+0.313
anwhile+0.312
Neg logit contributions
IELD-1.347
hips-1.288
 Desire-1.257
ymm-1.242
 suppose-1.206
Token construct
Token ID5678
Feature activation+1.13e-01

Pos logit contributions
umn+4.332
cient+4.267
arch+3.919
isode+3.869
ophone+3.860
Neg logit contributions
IELD-4.685
hips-4.322
ymm-4.195
 Desire-4.070
GV-3.867
Token than
Token ID621
Feature activation+7.07e-01

Pos logit contributions
umn+0.236
cient+0.228
arch+0.213
isode+0.213
anwhile+0.212
Neg logit contributions
IELD-0.197
hips-0.182
 Desire-0.177
ymm-0.175
 suppose-0.160
Token a
Token ID257
Feature activation-1.33e+00

Pos logit contributions
umn+1.658
cient+1.622
arch+1.537
isode+1.512
anwhile+1.510
Neg logit contributions
IELD-1.016
hips-0.917
 Desire-0.867
ymm-0.861
 Hold-0.767
Token says
Token ID1139
Feature activation+9.03e-01

Pos logit contributions
umn+1.017
cient+0.957
anwhile+0.910
arch+0.897
isode+0.887
Neg logit contributions
IELD-1.132
hips-1.056
 Desire-1.026
ymm-1.017
 suppose-0.952
Token she
Token ID673
Feature activation+3.65e-01

Pos logit contributions
umn+1.637
cient+1.543
anwhile+1.448
isode+1.443
arch+1.435
Neg logit contributions
IELD-1.826
hips-1.706
ymm-1.655
 Desire-1.628
 suppose-1.544
Token has
Token ID468
Feature activation+9.83e-01

Pos logit contributions
umn+0.692
cient+0.656
anwhile+0.622
arch+0.615
isode+0.608
Neg logit contributions
IELD-0.710
hips-0.666
 Desire-0.644
ymm-0.641
 suppose-0.596
Token been
Token ID587
Feature activation+6.26e-01

Pos logit contributions
umn+1.839
cient+1.748
isode+1.641
arch+1.640
anwhile+1.616
Neg logit contributions
IELD-1.918
hips-1.753
ymm-1.711
 Desire-1.704
 suppose-1.583
Token a
Token ID257
Feature activation+7.02e-01

Pos logit contributions
umn+1.256
cient+1.186
arch+1.132
isode+1.125
anwhile+1.113
Neg logit contributions
IELD-1.160
hips-1.059
ymm-1.018
 Desire-1.018
 suppose-0.933
Token travel
Token ID3067
Feature activation+4.40e+00

Pos logit contributions
umn+1.327
cient+1.246
isode+1.184
anwhile+1.181
arch+1.178
Neg logit contributions
IELD-1.362
hips-1.275
ymm-1.231
 Desire-1.197
 suppose-1.145
Token and
Token ID290
Feature activation+1.49e+00

Pos logit contributions
umn+5.057
cient+4.694
isode+4.272
arch+4.223
ophone+4.122
Neg logit contributions
IELD-11.001
hips-10.466
ymm-10.280
 Desire-9.954
ENTION-9.630
Token food
Token ID2057
Feature activation+1.16e+00

Pos logit contributions
umn+2.803
cient+2.646
isode+2.526
anwhile+2.512
arch+2.508
Neg logit contributions
IELD-2.893
hips-2.686
ymm-2.605
 Desire-2.549
 suppose-2.369
Token writer
Token ID6260
Feature activation+9.45e-01

Pos logit contributions
umn+1.569
cient+1.464
isode+1.340
arch+1.330
anwhile+1.329
Neg logit contributions
IELD-2.855
hips-2.743
ymm-2.670
 Desire-2.646
 suppose-2.505
Token for
Token ID329
Feature activation+1.37e+00

Pos logit contributions
umn+2.063
cient+1.950
arch+1.872
isode+1.861
anwhile+1.837
Neg logit contributions
IELD-1.592
hips-1.396
ymm-1.376
 Desire-1.359
 suppose-1.256
Token 22
Token ID2534
Feature activation+8.36e-02

Pos logit contributions
umn+2.882
cient+2.714
arch+2.566
isode+2.566
anwhile+2.543
Neg logit contributions
IELD-2.314
hips-2.123
ymm-2.073
 Desire-2.013
 suppose-1.920
Tokenhigh
Token ID8929
Feature activation-1.53e+00

Pos logit contributions
IELD+1.635
hips+1.581
 Desire+1.459
ymm+1.442
 suppose+1.333
Neg logit contributions
umn-2.891
cient-2.770
anwhile-2.721
isode-2.672
arch-2.623
Token heat
Token ID4894
Feature activation-1.15e+00

Pos logit contributions
IELD+2.780
hips+2.701
 Desire+2.558
ymm+2.526
 suppose+2.417
Neg logit contributions
umn-3.020
cient-2.878
anwhile-2.799
isode-2.739
arch-2.684
Token.
Token ID13
Feature activation+2.12e-01

Pos logit contributions
IELD+1.767
hips+1.693
 Desire+1.596
ymm+1.566
 suppose+1.462
Neg logit contributions
umn-2.619
cient-2.510
anwhile-2.447
isode-2.400
arch-2.368
Token Place
Token ID8474
Feature activation-4.77e-01

Pos logit contributions
umn+0.643
cient+0.617
arch+0.599
isode+0.594
ophone+0.593
Neg logit contributions
IELD-0.166
hips-0.143
ymm-0.131
 Desire-0.118
 suppose-0.095
Token f
Token ID277
Feature activation+6.30e-01

Pos logit contributions
IELD+0.880
hips+0.833
 Desire+0.796
ymm+0.793
 suppose+0.736
Neg logit contributions
umn-0.945
cient-0.902
anwhile-0.851
isode-0.845
arch-0.844
Tokenrit
Token ID799
Feature activation+4.40e+00

Pos logit contributions
umn+0.921
cient+0.862
anwhile+0.792
isode+0.786
arch+0.785
Neg logit contributions
IELD-1.491
hips-1.429
 Desire-1.395
ymm-1.376
 suppose-1.313
Tokenters
Token ID1010
Feature activation+1.52e+00

Pos logit contributions
umn+6.527
cient+5.933
arch+5.495
iously+5.424
ophone+5.093
Neg logit contributions
IELD-10.060
 Desire-9.213
ymm-8.925
hips-8.885
 glim-8.727
Token in
Token ID287
Feature activation+9.87e-01

Pos logit contributions
umn+3.181
cient+3.069
arch+2.910
iously+2.823
ophone+2.817
Neg logit contributions
IELD-2.647
hips-2.376
 Desire-2.318
ymm-2.304
 glim-2.099
Token the
Token ID262
Feature activation+1.88e+00

Pos logit contributions
umn+2.244
cient+2.159
arch+2.054
isode+2.013
ophone+2.012
Neg logit contributions
IELD-1.508
hips-1.403
 Desire-1.322
ymm-1.321
 suppose-1.197
Token skillet
Token ID41306
Feature activation-2.81e-01

Pos logit contributions
umn+3.227
cient+3.059
ophone+2.915
arch+2.870
anwhile+2.864
Neg logit contributions
IELD-3.873
hips-3.637
ymm-3.517
 Desire-3.510
 suppose-3.241
Token (
Token ID357
Feature activation+7.20e-01

Pos logit contributions
IELD+0.470
hips+0.445
 Desire+0.420
ymm+0.417
 suppose+0.388
Neg logit contributions
umn-0.607
cient-0.579
anwhile-0.552
isode-0.547
arch-0.546
Token When
Token ID1649
Feature activation+2.53e-01

Pos logit contributions
umn+0.419
cient+0.397
arch+0.379
anwhile+0.377
isode+0.372
Neg logit contributions
IELD-0.300
hips-0.278
ymm-0.262
 Desire-0.257
 suppose-0.245
Token Trudeau
Token ID17524
Feature activation-1.82e+00

Pos logit contributions
umn+0.532
cient+0.507
anwhile+0.479
arch+0.478
isode+0.466
Neg logit contributions
IELD-0.442
hips-0.411
 Desire-0.389
ymm-0.383
 suppose-0.352
Token chose
Token ID7690
Feature activation-1.87e-01

Pos logit contributions
IELD+3.357
hips+3.242
 Desire+3.071
ymm+3.042
 suppose+2.882
Neg logit contributions
umn-3.553
cient-3.397
anwhile-3.258
isode-3.225
arch-3.166
Token St
Token ID520
Feature activation-9.41e-01

Pos logit contributions
IELD+0.317
hips+0.293
ymm+0.279
 Desire+0.277
 suppose+0.259
Neg logit contributions
umn-0.403
cient-0.382
anwhile-0.364
arch-0.364
isode-0.356
Tokené
Token ID2634
Feature activation+1.84e+00

Pos logit contributions
IELD+2.085
hips+1.980
ymm+1.898
 Desire+1.891
 suppose+1.797
Neg logit contributions
umn-1.557
cient-1.455
arch-1.373
anwhile-1.365
isode-1.344
Tokenph
Token ID746
Feature activation+4.33e+00

Pos logit contributions
umn+2.143
cient+1.919
ophone+1.769
arch+1.752
isode+1.663
Neg logit contributions
IELD-5.036
hips-4.769
ymm-4.533
 Desire-4.530
 suppose-4.351
Tokenane
Token ID1531
Feature activation+4.52e-01

Pos logit contributions
umn+5.659
arch+5.326
cient+4.986
ocide+4.966
ophone+4.933
Neg logit contributions
IELD-10.435
WARE-9.671
hips-9.513
GV-9.125
 Desire-9.076
Token Dion
Token ID29628
Feature activation-7.41e-01

Pos logit contributions
umn+0.889
cient+0.830
anwhile+0.791
arch+0.787
isode+0.783
Neg logit contributions
IELD-0.844
hips-0.806
ymm-0.755
 Desire-0.744
 suppose-0.700
Token as
Token ID355
Feature activation+4.33e-01

Pos logit contributions
IELD+1.289
hips+1.224
 Desire+1.164
ymm+1.144
 suppose+1.073
Neg logit contributions
umn-1.552
cient-1.484
anwhile-1.411
arch-1.410
isode-1.386
Token minister
Token ID5342
Feature activation+7.49e-01

Pos logit contributions
umn+0.904
cient+0.842
arch+0.804
anwhile+0.800
isode+0.790
Neg logit contributions
IELD-0.761
hips-0.711
ymm-0.676
 Desire-0.670
 suppose-0.631
Token of
Token ID286
Feature activation+7.97e-02

Pos logit contributions
umn+1.503
cient+1.431
arch+1.353
isode+1.341
anwhile+1.335
Neg logit contributions
IELD-1.363
hips-1.235
 Desire-1.222
ymm-1.201
 suppose-1.126
Token mirrors
Token ID22353
Feature activation+7.35e-01

Pos logit contributions
IELD+0.955
hips+0.901
ymm+0.846
 Desire+0.839
 suppose+0.777
Neg logit contributions
umn-1.240
cient-1.196
anwhile-1.141
arch-1.127
isode-1.121
Token for
Token ID329
Feature activation-1.70e-01

Pos logit contributions
umn+1.470
cient+1.434
arch+1.354
anwhile+1.328
isode+1.318
Neg logit contributions
IELD-1.326
hips-1.211
 Desire-1.184
ymm-1.184
 suppose-1.074
Token primates
Token ID41316
Feature activation+1.37e+00

Pos logit contributions
IELD+0.280
hips+0.269
 Desire+0.250
ymm+0.248
 suppose+0.229
Neg logit contributions
umn-0.369
cient-0.355
anwhile-0.339
arch-0.336
isode-0.332
Token,
Token ID11
Feature activation+5.11e-01

Pos logit contributions
umn+2.716
cient+2.649
arch+2.528
ophone+2.434
anwhile+2.419
Neg logit contributions
IELD-2.495
hips-2.245
ymm-2.238
 Desire-2.190
 Darling-2.023
Token k
Token ID479
Feature activation-9.02e-01

Pos logit contributions
umn+1.059
cient+1.024
anwhile+0.971
arch+0.968
isode+0.954
Neg logit contributions
IELD-0.882
hips-0.822
 Desire-0.788
ymm-0.774
 suppose-0.705
Tokenong
Token ID506
Feature activation+4.31e+00

Pos logit contributions
IELD+1.876
hips+1.785
 Desire+1.728
ymm+1.678
 suppose+1.604
Neg logit contributions
umn-1.614
cient-1.532
arch-1.432
anwhile-1.422
isode-1.410
Token toys
Token ID14958
Feature activation+6.18e-01

Pos logit contributions
umn+7.576
arch+7.206
cient+7.176
ophone+6.737
anwhile+6.666
Neg logit contributions
IELD-8.141
hips-7.422
 Desire-7.097
ymm-6.927
ELF-6.789
Token and
Token ID290
Feature activation+8.21e-01

Pos logit contributions
umn+1.188
cient+1.151
arch+1.084
anwhile+1.076
ophone+1.062
Neg logit contributions
IELD-1.155
hips-1.055
 Desire-1.042
ymm-1.033
 suppose-0.962
Token pr
Token ID778
Feature activation+5.28e-01

Pos logit contributions
umn+1.658
cient+1.609
anwhile+1.526
arch+1.511
ophone+1.496
Neg logit contributions
IELD-1.454
hips-1.344
ymm-1.289
 Desire-1.281
 suppose-1.165
Tokenimate
Token ID1920
Feature activation+9.18e-02

Pos logit contributions
umn+1.061
cient+1.021
arch+0.960
ophone+0.933
isode+0.932
Neg logit contributions
IELD-0.984
hips-0.905
 Desire-0.887
ymm-0.861
WARE-0.811
Tokenc
Token ID66
Feature activation+1.65e+00

Pos logit contributions
umn+0.180
cient+0.172
arch+0.163
anwhile+0.162
isode+0.161
Neg logit contributions
IELD-0.170
hips-0.162
ymm-0.154
 Desire-0.153
 suppose-0.142
Token W
Token ID370
Feature activation-1.94e+00

Pos logit contributions
umn+1.574
cient+1.478
arch+1.404
isode+1.384
anwhile+1.379
Neg logit contributions
IELD-1.564
hips-1.546
 Desire-1.394
ymm-1.383
 suppose-1.336
Tokenether
Token ID6750
Feature activation+1.68e+00

Pos logit contributions
IELD+3.880
hips+3.746
 Desire+3.573
ymm+3.544
 suppose+3.347
Neg logit contributions
umn-3.485
cient-3.305
anwhile-3.171
isode-3.100
arch-3.059
Tokens
Token ID82
Feature activation-3.09e-01

Pos logit contributions
umn+3.105
cient+2.977
isode+2.872
anwhile+2.837
arch+2.815
Neg logit contributions
IELD-3.157
hips-3.002
ymm-2.860
 Desire-2.849
 suppose-2.711
Token Mysteries
Token ID28298
Feature activation+3.57e-02

Pos logit contributions
IELD+0.549
hips+0.506
ymm+0.487
 Desire+0.484
 suppose+0.463
Neg logit contributions
umn-0.638
cient-0.616
isode-0.582
anwhile-0.582
arch-0.580
Token,
Token ID11
Feature activation-5.12e-01

Pos logit contributions
umn+0.073
cient+0.071
isode+0.067
arch+0.066
anwhile+0.066
Neg logit contributions
IELD-0.063
hips-0.059
ymm-0.057
 Desire-0.056
 suppose-0.053
Token comic
Token ID9048
Feature activation+4.30e+00

Pos logit contributions
IELD+0.795
hips+0.747
ymm+0.697
 Desire+0.692
 suppose+0.647
Neg logit contributions
umn-1.171
cient-1.122
anwhile-1.079
isode-1.070
arch-1.058
Token book
Token ID1492
Feature activation+1.96e+00

Pos logit contributions
umn+7.479
cient+7.077
isode+6.959
arch+6.709
iously+6.554
Neg logit contributions
IELD-8.186
hips-7.707
ymm-7.353
 Desire-7.142
ENTION-6.803
Token short
Token ID1790
Feature activation+1.26e-01

Pos logit contributions
umn+3.736
cient+3.534
isode+3.404
anwhile+3.303
arch+3.300
Neg logit contributions
IELD-3.703
hips-3.486
ymm-3.332
 Desire-3.307
 suppose-3.115
Token stories
Token ID3923
Feature activation+9.34e-01

Pos logit contributions
umn+0.240
cient+0.227
isode+0.218
arch+0.212
anwhile+0.212
Neg logit contributions
IELD-0.244
hips-0.232
 Desire-0.220
ymm-0.219
 suppose-0.207
Token,
Token ID11
Feature activation-4.03e-01

Pos logit contributions
umn+1.831
cient+1.758
isode+1.646
arch+1.638
anwhile+1.629
Neg logit contributions
IELD-1.743
hips-1.596
ymm-1.550
 Desire-1.531
 suppose-1.458
Token prose
Token ID27149
Feature activation+8.61e-01

Pos logit contributions
IELD+0.730
hips+0.691
ymm+0.656
 Desire+0.653
 suppose+0.617
Neg logit contributions
umn-0.818
cient-0.781
anwhile-0.743
isode-0.738
arch-0.729
Token:
Token ID25
Feature activation+3.20e-01

Pos logit contributions
umn+3.006
cient+2.921
anwhile+2.751
arch+2.729
isode+2.703
Neg logit contributions
IELD-2.410
hips-2.219
ymm-2.134
 Desire-2.099
 suppose-1.936
Token "
Token ID366
Feature activation-4.32e-01

Pos logit contributions
umn+0.647
cient+0.617
anwhile+0.587
isode+0.582
arch+0.579
Neg logit contributions
IELD-0.577
hips-0.545
 Desire-0.522
ymm-0.517
 suppose-0.488
TokenSupporters
Token ID49422
Feature activation-1.09e+00

Pos logit contributions
IELD+0.703
hips+0.664
 Desire+0.638
ymm+0.627
 suppose+0.594
Neg logit contributions
umn-0.939
cient-0.906
arch-0.859
isode-0.854
anwhile-0.840
Token are
Token ID389
Feature activation+2.86e-01

Pos logit contributions
IELD+1.899
hips+1.762
ymm+1.694
 Desire+1.692
 suppose+1.570
Neg logit contributions
umn-2.293
cient-2.202
anwhile-2.078
arch-2.071
isode-2.055
Token the
Token ID262
Feature activation+1.11e+00

Pos logit contributions
umn+0.603
cient+0.588
arch+0.562
anwhile+0.547
isode+0.543
Neg logit contributions
IELD-0.490
hips-0.439
ymm-0.426
 Desire-0.422
ELF-0.377
Token life
Token ID1204
Feature activation+4.28e+00

Pos logit contributions
umn+2.138
cient+2.069
arch+1.952
anwhile+1.894
isode+1.874
Neg logit contributions
IELD-2.087
hips-1.999
ymm-1.884
 Desire-1.841
 suppose-1.758
Tokenblood
Token ID18041
Feature activation+1.64e+00

Pos logit contributions
cient+4.404
umn+4.360
arch+3.669
isode+3.398
iously+3.200
Neg logit contributions
IELD-11.181
ymm-10.517
hips-10.393
 Desire-10.376
ELF-10.175
Token of
Token ID286
Feature activation+2.69e-01

Pos logit contributions
umn+3.156
cient+2.993
isode+2.839
anwhile+2.827
arch+2.815
Neg logit contributions
IELD-3.047
hips-2.851
 Desire-2.808
ymm-2.794
 suppose-2.600
Token any
Token ID597
Feature activation+6.52e-01

Pos logit contributions
umn+0.628
cient+0.600
arch+0.571
anwhile+0.569
isode+0.566
Neg logit contributions
IELD-0.400
hips-0.370
ymm-0.351
 Desire-0.346
 suppose-0.319
Token football
Token ID4346
Feature activation+2.32e+00

Pos logit contributions
umn+1.254
cient+1.198
arch+1.130
isode+1.109
ophone+1.100
Neg logit contributions
IELD-1.231
hips-1.153
ymm-1.109
 Desire-1.089
 suppose-1.013
Token club
Token ID3430
Feature activation+1.31e-01

Pos logit contributions
umn+2.968
cient+2.908
isode+2.556
ophone+2.554
arch+2.536
Neg logit contributions
IELD-5.696
hips-5.295
ymm-5.185
 Desire-5.174
ELF-4.866
Token thinks
Token ID6834
Feature activation-4.87e-01

Pos logit contributions
IELD+2.160
hips+2.086
 Desire+1.983
ymm+1.957
 suppose+1.853
Neg logit contributions
umn-2.356
cient-2.245
anwhile-2.156
isode-2.120
arch-2.096
Token of
Token ID286
Feature activation-8.20e-01

Pos logit contributions
IELD+0.864
hips+0.825
 Desire+0.784
ymm+0.776
 suppose+0.729
Neg logit contributions
umn-0.999
cient-0.952
anwhile-0.910
isode-0.900
arch-0.900
Token when
Token ID618
Feature activation+1.12e+00

Pos logit contributions
IELD+1.208
hips+1.137
 Desire+1.068
ymm+1.055
 suppose+0.974
Neg logit contributions
umn-1.933
cient-1.860
anwhile-1.791
arch-1.777
isode-1.773
Token you
Token ID345
Feature activation+3.01e-01

Pos logit contributions
umn+2.144
cient+2.037
anwhile+1.949
arch+1.939
isode+1.905
Neg logit contributions
IELD-2.120
hips-2.002
ymm-1.918
 Desire-1.883
 suppose-1.760
Token mention
Token ID3068
Feature activation-3.96e-01

Pos logit contributions
umn+0.616
cient+0.587
anwhile+0.564
isode+0.555
arch+0.553
Neg logit contributions
IELD-0.538
hips-0.502
 Desire-0.482
ymm-0.481
 Darling-0.435
TokenPsych
Token ID31923
Feature activation+4.28e+00

Pos logit contributions
IELD+0.670
hips+0.634
 Desire+0.599
ymm+0.595
 suppose+0.558
Neg logit contributions
umn-0.847
cient-0.811
anwhile-0.775
arch-0.770
isode-0.760
Tokeno
Token ID78
Feature activation+2.07e+00

Pos logit contributions
umn+4.453
cient+3.682
arch+3.660
ophone+3.564
ocide+3.499
Neg logit contributions
IELD-11.444
hips-10.818
 Desire-10.651
ymm-10.441
 glim-10.261
Token,
Token ID11
Feature activation-3.81e-01

Pos logit contributions
umn+4.021
cient+3.845
arch+3.640
anwhile+3.633
isode+3.609
Neg logit contributions
IELD-3.692
hips-3.535
ymm-3.351
 Desire-3.344
 suppose-3.065
Token and
Token ID290
Feature activation+6.73e-01

Pos logit contributions
IELD+0.636
hips+0.606
 Desire+0.575
ymm+0.569
 suppose+0.526
Neg logit contributions
umn-0.817
cient-0.783
anwhile-0.751
arch-0.744
isode-0.738
Token with
Token ID351
Feature activation-1.88e+00

Pos logit contributions
umn+1.422
cient+1.365
anwhile+1.314
arch+1.299
isode+1.280
Neg logit contributions
IELD-1.135
hips-1.073
ymm-1.010
 Desire-1.004
 suppose-0.920
Token good
Token ID922
Feature activation-1.28e+00

Pos logit contributions
IELD+2.994
hips+2.850
 Desire+2.695
ymm+2.663
 suppose+2.478
Neg logit contributions
umn-4.179
cient-4.003
anwhile-3.853
isode-3.802
arch-3.776
Token W
Token ID370
Feature activation-1.80e+00

Pos logit contributions
umn+2.649
cient+2.532
ophone+2.372
arch+2.370
isode+2.327
Neg logit contributions
IELD-2.111
hips-2.012
ymm-1.903
 Desire-1.846
 suppose-1.783
Tokenether
Token ID6750
Feature activation+1.76e+00

Pos logit contributions
IELD+3.009
hips+2.961
ymm+2.737
 Desire+2.669
 suppose+2.534
Neg logit contributions
umn-3.768
anwhile-3.626
cient-3.571
isode-3.441
arch-3.353
Tokens
Token ID82
Feature activation-1.31e-01

Pos logit contributions
umn+3.122
cient+2.914
arch+2.779
iously+2.662
isode+2.635
Neg logit contributions
IELD-3.602
hips-3.301
 Desire-3.285
ymm-3.218
 glim-3.062
Token Mysteries
Token ID28298
Feature activation+3.91e-01

Pos logit contributions
IELD+0.237
hips+0.222
 Desire+0.213
ymm+0.211
 suppose+0.200
Neg logit contributions
umn-0.267
cient-0.253
arch-0.238
isode-0.237
anwhile-0.236
Token,
Token ID11
Feature activation-3.17e-01

Pos logit contributions
umn+0.781
cient+0.747
arch+0.702
isode+0.696
anwhile+0.690
Neg logit contributions
IELD-0.710
hips-0.668
 Desire-0.641
ymm-0.638
 suppose-0.600
Token comic
Token ID9048
Feature activation+4.27e+00

Pos logit contributions
IELD+0.577
hips+0.549
 Desire+0.523
ymm+0.519
 suppose+0.486
Neg logit contributions
umn-0.635
cient-0.608
anwhile-0.575
isode-0.573
arch-0.566
Token book
Token ID1492
Feature activation+2.12e+00

Pos logit contributions
umn+4.593
cient+4.413
arch+3.924
isode+3.705
ophone+3.650
Neg logit contributions
IELD-11.348
hips-10.548
ymm-10.317
 Desire-10.245
WARE-9.802
Token short
Token ID1790
Feature activation+4.37e-01

Pos logit contributions
umn+5.029
cient+4.822
isode+4.583
arch+4.535
ophone+4.468
Neg logit contributions
IELD-3.057
hips-2.806
 Desire-2.622
ymm-2.610
 suppose-2.309
Token stories
Token ID3923
Feature activation+1.19e+00

Pos logit contributions
umn+0.823
cient+0.782
arch+0.724
isode+0.723
anwhile+0.712
Neg logit contributions
IELD-0.859
hips-0.803
 Desire-0.779
ymm-0.768
 suppose-0.716
Token,
Token ID11
Feature activation-6.94e-02

Pos logit contributions
umn+2.252
cient+2.160
arch+2.024
ophone+1.968
isode+1.959
Neg logit contributions
IELD-2.284
hips-2.111
 Desire-2.052
ymm-2.043
 suppose-1.919
Token prose
Token ID27149
Feature activation+1.70e+00

Pos logit contributions
IELD+0.113
hips+0.107
 Desire+0.101
ymm+0.101
 suppose+0.093
Neg logit contributions
umn-0.152
cient-0.146
anwhile-0.139
isode-0.138
arch-0.137
Token MORE
Token ID12011
Feature activation-9.09e-01

Pos logit contributions
umn+1.339
cient+1.276
anwhile+1.220
arch+1.213
isode+1.195
Neg logit contributions
IELD-0.712
hips-0.668
 Desire-0.610
ymm-0.596
 suppose-0.557
Token as
Token ID355
Feature activation+9.97e-01

Pos logit contributions
IELD+1.542
hips+1.443
 Desire+1.379
ymm+1.362
 suppose+1.252
Neg logit contributions
umn-1.949
cient-1.869
anwhile-1.797
arch-1.778
isode-1.755
Token a
Token ID257
Feature activation+8.98e-01

Pos logit contributions
umn+2.172
cient+2.097
anwhile+2.015
arch+1.985
ophone+1.932
Neg logit contributions
IELD-1.603
hips-1.494
 Desire-1.428
ymm-1.366
GV-1.267
Token 68
Token ID8257
Feature activation+9.42e-01

Pos logit contributions
umn+1.946
cient+1.844
arch+1.761
anwhile+1.759
isode+1.715
Neg logit contributions
IELD-1.456
hips-1.406
ymm-1.292
 Desire-1.281
WARE-1.158
Token-
Token ID12
Feature activation+1.04e+00

Pos logit contributions
umn+1.392
cient+1.291
anwhile+1.199
arch+1.178
isode+1.169
Neg logit contributions
IELD-2.194
hips-2.132
 Desire-2.061
ymm-2.041
 suppose-1.931
Tokenyear
Token ID1941
Feature activation+4.26e+00

Pos logit contributions
umn+1.096
cient+0.959
arch+0.872
isode+0.852
anwhile+0.849
Neg logit contributions
IELD-2.915
hips-2.800
 Desire-2.740
ymm-2.683
 suppose-2.570
Token-
Token ID12
Feature activation+1.06e-01

Pos logit contributions
umn+5.365
cient+4.993
arch+4.623
anwhile+4.285
isode+4.275
Neg logit contributions
IELD-10.087
hips-9.890
ymm-9.464
 Desire-9.290
GV-8.837
Tokenold
Token ID727
Feature activation+1.37e+00

Pos logit contributions
umn+0.160
cient+0.151
anwhile+0.144
isode+0.140
arch+0.137
Neg logit contributions
IELD-0.244
hips-0.237
 Desire-0.228
ymm-0.226
 suppose-0.216
Token woman
Token ID2415
Feature activation+4.57e-01

Pos logit contributions
umn+2.652
cient+2.511
arch+2.404
anwhile+2.340
ophone+2.322
Neg logit contributions
IELD-2.568
hips-2.415
ymm-2.267
 Desire-2.209
ELF-2.066
Token's
Token ID338
Feature activation-5.08e-01

Pos logit contributions
umn+0.911
cient+0.870
arch+0.826
anwhile+0.819
isode+0.811
Neg logit contributions
IELD-0.842
hips-0.775
ymm-0.739
 Desire-0.738
 suppose-0.676
Token cause
Token ID2728
Feature activation+7.41e-01

Pos logit contributions
IELD+0.908
hips+0.861
ymm+0.813
 Desire+0.805
 suppose+0.739
Neg logit contributions
umn-1.033
cient-0.993
anwhile-0.937
arch-0.936
isode-0.922
Token was
Token ID373
Feature activation+8.32e-01

Pos logit contributions
IELD+1.567
hips+1.491
 Desire+1.418
ymm+1.412
 suppose+1.318
Neg logit contributions
umn-1.712
cient-1.633
anwhile-1.562
arch-1.537
isode-1.531
Token forced
Token ID4137
Feature activation+2.17e-01

Pos logit contributions
umn+1.567
cient+1.499
arch+1.431
anwhile+1.409
ophone+1.368
Neg logit contributions
IELD-1.608
hips-1.487
ymm-1.440
 Desire-1.437
 Darling-1.292
Token to
Token ID284
Feature activation+1.20e+00

Pos logit contributions
umn+0.431
cient+0.410
anwhile+0.391
arch+0.388
isode+0.382
Neg logit contributions
IELD-0.405
hips-0.377
 Desire-0.364
ymm-0.363
 suppose-0.334
Token write
Token ID3551
Feature activation+2.17e+00

Pos logit contributions
umn+2.230
cient+2.119
arch+2.019
isode+1.978
anwhile+1.972
Neg logit contributions
IELD-2.336
hips-2.163
ymm-2.132
 Desire-2.108
 Darling-1.939
Token a
Token ID257
Feature activation+6.38e-01

Pos logit contributions
umn+4.744
cient+4.616
arch+4.353
iously+4.352
anwhile+4.309
Neg logit contributions
IELD-3.445
hips-3.022
ymm-2.947
 Desire-2.805
WARE-2.621
Token letter
Token ID3850
Feature activation+4.25e+00

Pos logit contributions
umn+1.152
cient+1.097
arch+1.036
anwhile+1.035
isode+1.017
Neg logit contributions
IELD-1.284
hips-1.218
ymm-1.164
 Desire-1.126
 suppose-1.063
Token explaining
Token ID11170
Feature activation+1.53e+00

Pos logit contributions
umn+7.435
cient+7.198
arch+6.784
iously+6.566
anwhile+6.559
Neg logit contributions
IELD-7.972
hips-7.101
ymm-7.097
 Desire-6.772
WARE-6.326
Token her
Token ID607
Feature activation-6.30e-01

Pos logit contributions
umn+3.202
cient+3.035
arch+2.893
anwhile+2.869
iously+2.818
Neg logit contributions
IELD-2.609
hips-2.444
ymm-2.331
 Desire-2.265
 suppose-2.110
Token actions
Token ID4028
Feature activation+2.30e-01

Pos logit contributions
IELD+1.190
hips+1.137
ymm+1.082
 Desire+1.074
 suppose+1.005
Neg logit contributions
umn-1.221
cient-1.167
anwhile-1.101
arch-1.098
isode-1.072
Token in
Token ID287
Feature activation+2.15e-01

Pos logit contributions
umn+0.479
cient+0.464
anwhile+0.440
arch+0.434
isode+0.428
Neg logit contributions
IELD-0.406
hips-0.375
ymm-0.361
 Desire-0.358
 suppose-0.331
Token school
Token ID1524
Feature activation+6.26e-02

Pos logit contributions
umn+0.453
cient+0.432
anwhile+0.417
arch+0.415
isode+0.405
Neg logit contributions
IELD-0.370
hips-0.351
ymm-0.333
 Desire-0.320
 suppose-0.304
Token call
Token ID869
Feature activation+2.19e-01

Pos logit contributions
IELD+0.306
hips+0.289
 Desire+0.277
ymm+0.275
 suppose+0.253
Neg logit contributions
umn-0.365
cient-0.349
anwhile-0.334
arch-0.330
isode-0.328
Token it
Token ID340
Feature activation+4.92e-01

Pos logit contributions
umn+0.456
cient+0.436
anwhile+0.413
arch+0.412
isode+0.405
Neg logit contributions
IELD-0.385
hips-0.361
ymm-0.344
 Desire-0.337
 suppose-0.316
Token a
Token ID257
Feature activation-5.25e-01

Pos logit contributions
umn+1.108
cient+1.073
arch+1.018
anwhile+1.014
ophone+1.001
Neg logit contributions
IELD-0.783
hips-0.702
ymm-0.679
 Desire-0.654
 suppose-0.599
Token
Token ID564
Feature activation+1.41e-01

Pos logit contributions
IELD+0.958
hips+0.907
ymm+0.863
 Desire+0.860
 suppose+0.801
Neg logit contributions
umn-1.051
cient-1.008
anwhile-0.957
isode-0.948
arch-0.945
Token
Token ID246
Feature activation+8.45e-01

Pos logit contributions
umn+0.291
cient+0.277
arch+0.259
anwhile+0.256
isode+0.256
Neg logit contributions
IELD-0.248
hips-0.237
 Desire-0.225
ymm-0.225
 suppose-0.209
Tokenph
Token ID746
Feature activation+4.23e+00

Pos logit contributions
umn+1.486
cient+1.425
arch+1.327
anwhile+1.315
isode+1.308
Neg logit contributions
IELD-1.744
hips-1.633
 Desire-1.591
ymm-1.577
 suppose-1.494
Tokeno
Token ID78
Feature activation+2.21e+00

Pos logit contributions
umn+4.859
cient+4.442
ophone+4.163
arch+4.153
ocide+4.142
Neg logit contributions
IELD-10.690
hips-10.190
 Desire-9.940
ymm-9.648
 Darling-9.368
Token,
Token ID11
Feature activation-4.15e-01

Pos logit contributions
umn+4.266
cient+4.088
isode+3.881
arch+3.848
ophone+3.838
Neg logit contributions
IELD-4.100
hips-3.791
ymm-3.722
 Desire-3.613
 suppose-3.323
Token
Token ID447
Feature activation-1.21e-01

Pos logit contributions
IELD+0.832
hips+0.794
 Desire+0.763
ymm+0.763
 suppose+0.708
Neg logit contributions
umn-0.756
cient-0.719
anwhile-0.677
arch-0.670
isode-0.666
Token
Token ID247
Feature activation-5.71e-02

Pos logit contributions
IELD+0.213
hips+0.201
 Desire+0.194
ymm+0.192
 suppose+0.179
Neg logit contributions
umn-0.252
cient-0.240
anwhile-0.225
arch-0.224
isode-0.222
Token
Token ID447
Feature activation-5.00e-01

Pos logit contributions
IELD+0.097
hips+0.091
ymm+0.086
 Desire+0.085
 suppose+0.079
Neg logit contributions
umn-0.122
cient-0.117
anwhile-0.112
arch-0.110
isode-0.110
Token this
Token ID428
Feature activation+1.14e+00

Pos logit contributions
IELD+0.429
hips+0.399
ymm+0.379
 Desire+0.378
 suppose+0.337
Neg logit contributions
umn-0.633
cient-0.607
anwhile-0.579
arch-0.575
isode-0.572
Token query
Token ID12405
Feature activation-5.78e-01

Pos logit contributions
umn+1.784
cient+1.695
arch+1.582
anwhile+1.549
isode+1.545
Neg logit contributions
IELD-2.553
hips-2.415
ymm-2.339
 Desire-2.301
WARE-2.120
Token to
Token ID284
Feature activation+9.23e-02

Pos logit contributions
IELD+0.975
hips+0.920
 Desire+0.880
ymm+0.878
 suppose+0.796
Neg logit contributions
umn-1.239
cient-1.190
anwhile-1.134
arch-1.126
isode-1.117
Token the
Token ID262
Feature activation+5.90e-01

Pos logit contributions
umn+0.177
cient+0.168
anwhile+0.160
arch+0.159
isode+0.158
Neg logit contributions
IELD-0.175
hips-0.166
 Desire-0.160
ymm-0.160
 suppose-0.146
Token cl
Token ID537
Feature activation+1.61e+00

Pos logit contributions
umn+1.118
cient+1.055
arch+1.004
anwhile+1.003
isode+0.992
Neg logit contributions
IELD-1.142
hips-1.073
ymm-1.038
 Desire-1.033
 suppose-0.963
Tokenojure
Token ID32511
Feature activation+4.21e+00

Pos logit contributions
umn+2.384
cient+2.215
arch+2.078
anwhile+2.046
ophone+2.043
Neg logit contributions
IELD-3.865
hips-3.600
 Desire-3.534
ymm-3.422
 suppose-3.261
Token.
Token ID13
Feature activation+4.27e-01

Pos logit contributions
umn+7.541
cient+7.210
arch+7.117
anwhile+6.879
isode+6.626
Neg logit contributions
IELD-7.768
hips-7.397
 Desire-7.185
ymm-7.048
 suppose-6.431
Tokenfiles
Token ID16624
Feature activation+5.26e-01

Pos logit contributions
umn+0.485
cient+0.443
arch+0.408
anwhile+0.388
isode+0.383
Neg logit contributions
IELD-1.162
hips-1.116
 Desire-1.092
ymm-1.077
 suppose-1.037
Token table
Token ID3084
Feature activation+7.71e-01

Pos logit contributions
umn+1.033
cient+0.989
arch+0.931
anwhile+0.918
isode+0.908
Neg logit contributions
IELD-0.990
hips-0.910
 Desire-0.903
ymm-0.885
 suppose-0.825
Token SELECT
Token ID33493
Feature activation-1.45e-01

Pos logit contributions
umn+1.540
cient+1.469
arch+1.408
anwhile+1.381
isode+1.371
Neg logit contributions
IELD-1.386
hips-1.308
ymm-1.278
 Desire-1.261
 suppose-1.136
Token *
Token ID1635
Feature activation-1.66e-01

Pos logit contributions
IELD+0.275
hips+0.260
ymm+0.252
 Desire+0.250
 suppose+0.233
Neg logit contributions
umn-0.281
cient-0.265
arch-0.252
anwhile-0.251
isode-0.249
TokenF
Token ID37
Feature activation-2.39e-02

Pos logit contributions
umn+2.648
cient+2.467
arch+2.404
isode+2.343
anwhile+2.317
Neg logit contributions
IELD-3.009
hips-2.961
ymm-2.860
 Desire-2.810
WARE-2.537
Token (
Token ID357
Feature activation-1.63e-01

Pos logit contributions
IELD+0.048
hips+0.046
ymm+0.044
 Desire+0.044
 suppose+0.041
Neg logit contributions
umn-0.043
cient-0.041
anwhile-0.039
isode-0.039
arch-0.038
TokenLow
Token ID20535
Feature activation-3.75e-01

Pos logit contributions
IELD+0.296
hips+0.276
 Desire+0.267
ymm+0.264
 suppose+0.249
Neg logit contributions
umn-0.331
cient-0.316
arch-0.300
anwhile-0.298
isode-0.298
Token Carb
Token ID20453
Feature activation-2.83e+00

Pos logit contributions
IELD+0.748
hips+0.708
ymm+0.686
 Desire+0.668
 suppose+0.638
Neg logit contributions
umn-0.684
cient-0.650
anwhile-0.619
arch-0.615
isode-0.613
Token High
Token ID3334
Feature activation-3.34e+00

Pos logit contributions
IELD+6.391
hips+6.211
 Desire+5.984
ymm+5.902
 suppose+5.637
Neg logit contributions
umn-4.358
cient-4.097
anwhile-3.857
isode-3.787
arch-3.730
Token Fat
Token ID12301
Feature activation-5.44e+00

Pos logit contributions
IELD+5.625
hips+5.493
 Desire+5.260
ymm+5.009
 suppose+4.755
Neg logit contributions
umn-6.952
cient-6.671
anwhile-6.409
isode-6.288
ophone-6.150
Token)
Token ID8
Feature activation-8.01e-01

Pos logit contributions
hips+8.494
IELD+8.460
 Desire+7.981
ymm+7.394
 suppose+7.139
Neg logit contributions
umn-11.521
cient-11.120
anwhile-10.566
isode-10.345
ophone-10.219
Token diet
Token ID5496
Feature activation+7.29e-01

Pos logit contributions
IELD+1.561
hips+1.482
ymm+1.420
 Desire+1.408
 suppose+1.309
Neg logit contributions
umn-1.502
cient-1.423
anwhile-1.359
isode-1.355
arch-1.345
Token improves
Token ID19575
Feature activation+8.50e-01

Pos logit contributions
umn+1.448
cient+1.401
anwhile+1.321
arch+1.316
isode+1.307
Neg logit contributions
IELD-1.330
hips-1.206
 Desire-1.184
ymm-1.179
 suppose-1.077
Token the
Token ID262
Feature activation+9.02e-01

Pos logit contributions
umn+1.657
cient+1.576
arch+1.529
anwhile+1.493
isode+1.485
Neg logit contributions
IELD-1.546
hips-1.436
ymm-1.416
 Desire-1.354
WARE-1.279
Token body
Token ID1767
Feature activation-1.35e-01

Pos logit contributions
umn+1.713
cient+1.627
arch+1.558
anwhile+1.535
isode+1.503
Neg logit contributions
IELD-1.705
hips-1.599
ymm-1.550
 Desire-1.499
 suppose-1.421
Token
Token ID198
Feature activation+1.15e+00

Pos logit contributions
umn+3.423
cient+3.288
arch+3.128
ophone+3.013
iously+2.995
Neg logit contributions
IELD-2.715
hips-2.498
ymm-2.416
 Desire-2.400
 glim-2.274
Token3
Token ID18
Feature activation+6.42e-01

Pos logit contributions
umn+2.381
cient+2.286
arch+2.173
isode+2.113
ophone+2.070
Neg logit contributions
IELD-1.939
hips-1.837
 Desire-1.812
ymm-1.747
 suppose-1.667
Token.
Token ID13
Feature activation+1.72e-01

Pos logit contributions
umn+1.352
cient+1.307
anwhile+1.239
arch+1.227
isode+1.224
Neg logit contributions
IELD-1.073
hips-1.024
 Desire-0.969
ymm-0.953
 suppose-0.908
Token THE
Token ID3336
Feature activation+4.82e-01

Pos logit contributions
umn+0.385
cient+0.371
arch+0.351
anwhile+0.349
isode+0.347
Neg logit contributions
IELD-0.266
hips-0.253
ymm-0.236
 Desire-0.233
 suppose-0.223
Token BIG
Token ID26746
Feature activation-1.52e+00

Pos logit contributions
umn+1.057
cient+1.022
anwhile+0.977
arch+0.970
isode+0.952
Neg logit contributions
IELD-0.751
hips-0.723
ymm-0.675
 Desire-0.673
 suppose-0.635
Token SH
Token ID6006
Feature activation-4.84e+00

Pos logit contributions
IELD+2.350
hips+2.238
 Desire+2.113
ymm+2.088
 suppose+1.937
Neg logit contributions
umn-3.432
cient-3.290
anwhile-3.170
isode-3.131
arch-3.113
TokenORT
Token ID9863
Feature activation-8.86e-02

Pos logit contributions
IELD+7.444
hips+6.516
 Desire+6.049
ymm+5.859
ELF+5.717
Neg logit contributions
umn-11.155
anwhile-10.832
cient-10.744
isode-10.270
ophone-10.133
Token
Token ID198
Feature activation+1.18e-01

Pos logit contributions
IELD+0.135
hips+0.133
 Desire+0.127
ymm+0.127
 suppose+0.117
Neg logit contributions
umn-0.194
cient-0.186
arch-0.179
anwhile-0.177
isode-0.177
Token
Token ID198
Feature activation+2.22e-01

Pos logit contributions
umn+0.271
cient+0.261
arch+0.247
isode+0.245
anwhile+0.244
Neg logit contributions
IELD-0.178
hips-0.167
 Desire-0.158
ymm-0.157
 suppose-0.144
TokenPeople
Token ID8061
Feature activation+1.03e+00

Pos logit contributions
umn+0.466
cient+0.447
arch+0.424
isode+0.417
anwhile+0.408
Neg logit contributions
IELD-0.377
hips-0.359
 Desire-0.349
ymm-0.338
 suppose-0.321
Token inside
Token ID2641
Feature activation-3.29e-01

Pos logit contributions
umn+2.082
cient+2.006
anwhile+1.885
arch+1.880
isode+1.838
Neg logit contributions
IELD-1.862
hips-1.691
ymm-1.622
 Desire-1.613
WARE-1.483
Token Emirates
Token ID24880
Feature activation-2.11e+00

Pos logit contributions
IELD+3.063
hips+2.893
 Desire+2.718
ymm+2.682
 suppose+2.467
Neg logit contributions
umn-5.191
cient-4.987
anwhile-4.815
isode-4.757
arch-4.729
Token and
Token ID290
Feature activation+9.40e-01

Pos logit contributions
IELD+3.696
hips+3.535
 Desire+3.362
ymm+3.325
 suppose+3.116
Neg logit contributions
umn-4.370
cient-4.172
anwhile-4.006
isode-3.948
arch-3.916
Token Palestine
Token ID17263
Feature activation-7.52e-01

Pos logit contributions
umn+1.724
cient+1.644
anwhile+1.578
arch+1.563
isode+1.541
Neg logit contributions
IELD-1.821
hips-1.709
ymm-1.662
 Desire-1.598
 suppose-1.522
Token in
Token ID287
Feature activation+3.10e-01

Pos logit contributions
IELD+1.273
hips+1.204
 Desire+1.142
ymm+1.140
 suppose+1.053
Neg logit contributions
umn-1.601
cient-1.535
anwhile-1.466
arch-1.451
isode-1.450
Token Group
Token ID4912
Feature activation-1.01e+00

Pos logit contributions
umn+0.626
cient+0.596
arch+0.564
anwhile+0.563
isode+0.560
Neg logit contributions
IELD-0.557
hips-0.529
ymm-0.497
 Desire-0.490
 suppose-0.462
Token A
Token ID317
Feature activation-4.74e+00

Pos logit contributions
IELD+1.568
hips+1.464
ymm+1.387
 Desire+1.368
 suppose+1.262
Neg logit contributions
umn-2.328
cient-2.204
anwhile-2.129
isode-2.128
arch-2.126
Token.
Token ID13
Feature activation+2.69e-01

Pos logit contributions
IELD+6.864
hips+6.734
 Desire+6.189
ymm+6.182
 suppose+5.796
Neg logit contributions
umn-10.396
cient-10.047
anwhile-9.675
isode-9.523
arch-9.326
Token Here
Token ID3423
Feature activation-8.18e-01

Pos logit contributions
umn+0.602
cient+0.576
arch+0.550
anwhile+0.550
ophone+0.541
Neg logit contributions
IELD-0.416
hips-0.387
ymm-0.370
 Desire-0.361
 suppose-0.336
Token,
Token ID11
Feature activation-1.30e+00

Pos logit contributions
IELD+1.460
hips+1.393
 Desire+1.323
ymm+1.308
 suppose+1.219
Neg logit contributions
umn-1.667
cient-1.584
anwhile-1.532
arch-1.501
isode-1.497
Token the
Token ID262
Feature activation-7.17e-02

Pos logit contributions
IELD+2.275
hips+2.152
 Desire+2.044
ymm+2.031
 suppose+1.874
Neg logit contributions
umn-2.684
cient-2.563
anwhile-2.478
arch-2.426
isode-2.418
Token most
Token ID749
Feature activation-1.98e-01

Pos logit contributions
IELD+0.131
hips+0.125
ymm+0.118
 Desire+0.118
 suppose+0.109
Neg logit contributions
umn-0.142
cient-0.136
anwhile-0.130
arch-0.128
isode-0.126
Token
Token ID126
Feature activation+5.56e-01

Pos logit contributions
IELD+0.420
hips+0.391
 Desire+0.374
ymm+0.372
 suppose+0.346
Neg logit contributions
umn-0.475
cient-0.454
anwhile-0.432
arch-0.429
isode-0.426
Token
Token ID245
Feature activation+1.65e+00

Pos logit contributions
umn+1.038
cient+0.982
arch+0.914
anwhile+0.912
isode+0.904
Neg logit contributions
IELD-1.094
hips-1.029
 Desire-1.001
ymm-0.999
 suppose-0.938
Tokenwith
Token ID4480
Feature activation-4.17e-01

Pos logit contributions
umn+2.980
cient+2.827
arch+2.733
anwhile+2.650
isode+2.613
Neg logit contributions
IELD-3.280
hips-2.977
 Desire-2.946
ymm-2.900
 suppose-2.716
Token no
Token ID645
Feature activation-8.16e-02

Pos logit contributions
IELD+0.641
hips+0.595
 Desire+0.564
ymm+0.563
 suppose+0.523
Neg logit contributions
umn-0.958
cient-0.924
anwhile-0.881
arch-0.878
isode-0.871
Token strings
Token ID13042
Feature activation-5.41e-01

Pos logit contributions
IELD+0.148
hips+0.140
ymm+0.134
 Desire+0.132
 suppose+0.122
Neg logit contributions
umn-0.163
cient-0.156
anwhile-0.149
arch-0.148
isode-0.148
Token attached
Token ID7223
Feature activation-4.73e+00

Pos logit contributions
IELD+1.080
hips+1.023
 Desire+0.991
ymm+0.982
 suppose+0.916
Neg logit contributions
umn-0.989
cient-0.945
anwhile-0.888
arch-0.877
isode-0.873
Token.
Token ID13
Feature activation-3.79e-02

Pos logit contributions
IELD+6.422
hips+6.399
 Desire+5.792
ymm+5.615
 suppose+5.390
Neg logit contributions
umn-10.692
cient-10.214
anwhile-10.035
isode-9.953
arch-9.618
Token The
Token ID383
Feature activation-5.27e-01

Pos logit contributions
IELD+0.058
hips+0.055
ymm+0.052
 Desire+0.051
 suppose+0.048
Neg logit contributions
umn-0.086
cient-0.082
anwhile-0.078
arch-0.078
isode-0.077
Token hope
Token ID2911
Feature activation-4.38e-01

Pos logit contributions
IELD+1.007
hips+0.965
ymm+0.914
 Desire+0.913
 suppose+0.857
Neg logit contributions
umn-1.010
cient-0.960
anwhile-0.918
arch-0.902
isode-0.899
Token is
Token ID318
Feature activation-1.27e+00

Pos logit contributions
IELD+0.782
hips+0.746
 Desire+0.711
ymm+0.702
 suppose+0.653
Neg logit contributions
umn-0.890
cient-0.857
anwhile-0.819
arch-0.804
isode-0.798
Token that
Token ID326
Feature activation+2.11e-01

Pos logit contributions
IELD+2.174
hips+2.070
 Desire+1.965
ymm+1.945
 suppose+1.809
Neg logit contributions
umn-2.701
cient-2.581
anwhile-2.480
isode-2.436
arch-2.427
Token,
Token ID11
Feature activation+1.19e-01

Pos logit contributions
IELD+3.738
hips+3.552
 Desire+3.389
ymm+3.364
 suppose+3.138
Neg logit contributions
umn-4.444
cient-4.236
anwhile-4.068
isode-4.006
arch-3.992
Token Arsenal
Token ID13837
Feature activation-3.07e+00

Pos logit contributions
umn+0.265
cient+0.256
anwhile+0.246
arch+0.245
ophone+0.239
Neg logit contributions
IELD-0.186
hips-0.175
ymm-0.168
 Desire-0.161
 suppose-0.151
Token,
Token ID11
Feature activation+1.38e-01

Pos logit contributions
IELD+5.185
hips+4.961
 Desire+4.716
ymm+4.637
 suppose+4.344
Neg logit contributions
umn-6.464
cient-6.188
anwhile-5.967
isode-5.881
arch-5.805
Token Liverpool
Token ID11761
Feature activation-1.56e+00

Pos logit contributions
umn+0.305
cient+0.293
anwhile+0.282
arch+0.280
ophone+0.274
Neg logit contributions
IELD-0.219
hips-0.207
ymm-0.198
 Desire-0.190
 suppose-0.179
Token and
Token ID290
Feature activation+1.65e+00

Pos logit contributions
IELD+2.780
hips+2.651
 Desire+2.525
ymm+2.512
 suppose+2.357
Neg logit contributions
umn-3.182
cient-3.033
anwhile-2.909
isode-2.864
arch-2.849
Token Tottenham
Token ID24272
Feature activation-4.54e+00

Pos logit contributions
umn+3.508
cient+3.353
anwhile+3.258
arch+3.236
ophone+3.144
Neg logit contributions
IELD-2.641
hips-2.480
ymm-2.449
 Desire-2.259
WARE-2.149
Token Hots
Token ID41239
Feature activation-4.54e+00

Pos logit contributions
IELD+6.859
hips+6.799
 Desire+6.272
ymm+6.078
 suppose+5.692
Neg logit contributions
umn-9.959
cient-9.526
anwhile-9.203
isode-9.172
arch-8.883
Tokenpur
Token ID14225
Feature activation-2.36e+00

Pos logit contributions
hips+7.683
IELD+7.399
 Desire+6.880
ymm+6.709
 suppose+6.407
Neg logit contributions
umn-9.241
anwhile-9.149
cient-8.786
isode-8.597
theless-8.465
Token
Token ID851
Feature activation+5.72e-01

Pos logit contributions
IELD+3.992
hips+3.813
 Desire+3.619
ymm+3.578
 suppose+3.345
Neg logit contributions
umn-5.015
cient-4.794
anwhile-4.608
isode-4.544
arch-4.507
Token stating
Token ID12316
Feature activation+2.86e-01

Pos logit contributions
umn+1.180
cient+1.122
arch+1.070
anwhile+1.067
isode+1.042
Neg logit contributions
IELD-1.011
hips-0.921
ymm-0.894
 Desire-0.890
 suppose-0.818
Token the
Token ID262
Feature activation+8.85e-01

Pos logit contributions
umn+0.580
cient+0.556
arch+0.526
anwhile+0.524
isode+0.510
Neg logit contributions
IELD-0.502
hips-0.476
ymm-0.458
 Desire-0.445
 suppose-0.417
Token Arsenal
Token ID13837
Feature activation-3.07e+00

Pos logit contributions
umn+0.265
cient+0.256
anwhile+0.246
arch+0.245
ophone+0.239
Neg logit contributions
IELD-0.186
hips-0.175
ymm-0.168
 Desire-0.161
 suppose-0.151
Token,
Token ID11
Feature activation+1.38e-01

Pos logit contributions
IELD+5.185
hips+4.961
 Desire+4.716
ymm+4.637
 suppose+4.344
Neg logit contributions
umn-6.464
cient-6.188
anwhile-5.967
isode-5.881
arch-5.805
Token Liverpool
Token ID11761
Feature activation-1.56e+00

Pos logit contributions
umn+0.305
cient+0.293
anwhile+0.282
arch+0.280
ophone+0.274
Neg logit contributions
IELD-0.219
hips-0.207
ymm-0.198
 Desire-0.190
 suppose-0.179
Token and
Token ID290
Feature activation+1.65e+00

Pos logit contributions
IELD+2.780
hips+2.651
 Desire+2.525
ymm+2.512
 suppose+2.357
Neg logit contributions
umn-3.182
cient-3.033
anwhile-2.909
isode-2.864
arch-2.849
Token Tottenham
Token ID24272
Feature activation-4.54e+00

Pos logit contributions
umn+3.508
cient+3.353
anwhile+3.258
arch+3.236
ophone+3.144
Neg logit contributions
IELD-2.641
hips-2.480
ymm-2.449
 Desire-2.259
WARE-2.149
Token Hots
Token ID41239
Feature activation-4.54e+00

Pos logit contributions
IELD+6.859
hips+6.799
 Desire+6.272
ymm+6.078
 suppose+5.692
Neg logit contributions
umn-9.959
cient-9.526
anwhile-9.203
isode-9.172
arch-8.883
Tokenpur
Token ID14225
Feature activation-2.36e+00

Pos logit contributions
hips+7.683
IELD+7.399
 Desire+6.880
ymm+6.709
 suppose+6.407
Neg logit contributions
umn-9.241
anwhile-9.149
cient-8.786
isode-8.597
theless-8.465
Token
Token ID851
Feature activation+5.72e-01

Pos logit contributions
IELD+3.992
hips+3.813
 Desire+3.619
ymm+3.578
 suppose+3.345
Neg logit contributions
umn-5.015
cient-4.794
anwhile-4.608
isode-4.544
arch-4.507
Token stating
Token ID12316
Feature activation+2.86e-01

Pos logit contributions
umn+1.180
cient+1.122
arch+1.070
anwhile+1.067
isode+1.042
Neg logit contributions
IELD-1.011
hips-0.921
ymm-0.894
 Desire-0.890
 suppose-0.818
Token the
Token ID262
Feature activation+8.85e-01

Pos logit contributions
umn+0.580
cient+0.556
arch+0.526
anwhile+0.524
isode+0.510
Neg logit contributions
IELD-0.502
hips-0.476
ymm-0.458
 Desire-0.445
 suppose-0.417
Token existing
Token ID4683
Feature activation+3.55e-01

Pos logit contributions
umn+1.638
cient+1.584
arch+1.489
anwhile+1.462
ophone+1.426
Neg logit contributions
IELD-1.691
hips-1.645
ymm-1.550
 Desire-1.524
 suppose-1.442
Token conscious
Token ID6921
Feature activation+1.66e-01

Pos logit contributions
IELD+0.240
hips+0.224
ymm+0.213
 Desire+0.212
 suppose+0.193
Neg logit contributions
umn-0.305
cient-0.293
arch-0.277
anwhile-0.277
isode-0.276
Token mind
Token ID2000
Feature activation+1.39e-01

Pos logit contributions
umn+0.315
cient+0.303
arch+0.284
isode+0.281
anwhile+0.280
Neg logit contributions
IELD-0.321
hips-0.299
ymm-0.291
 Desire-0.282
 suppose-0.265
Token and
Token ID290
Feature activation+1.11e+00

Pos logit contributions
umn+0.266
cient+0.256
arch+0.240
anwhile+0.239
isode+0.237
Neg logit contributions
IELD-0.266
hips-0.250
ymm-0.242
 Desire-0.238
 suppose-0.221
Token your
Token ID534
Feature activation-1.07e+00

Pos logit contributions
umn+2.178
cient+2.145
arch+2.019
isode+1.981
ophone+1.953
Neg logit contributions
IELD-1.998
hips-1.854
ymm-1.780
 Desire-1.737
 Darling-1.605
Token Inner
Token ID24877
Feature activation-6.67e-01

Pos logit contributions
IELD+2.023
hips+1.931
 Desire+1.836
ymm+1.832
 suppose+1.711
Neg logit contributions
umn-2.056
cient-1.972
anwhile-1.860
isode-1.847
arch-1.843
Token Ass
Token ID2195
Feature activation-4.48e+00

Pos logit contributions
IELD+0.988
hips+0.926
ymm+0.855
 Desire+0.827
 suppose+0.790
Neg logit contributions
umn-1.584
cient-1.525
arch-1.455
isode-1.440
ophone-1.423
Tokenhole
Token ID13207
Feature activation-1.46e+00

Pos logit contributions
IELD+7.897
hips+7.801
ymm+7.470
 Desire+7.434
 suppose+6.779
Neg logit contributions
umn-8.513
cient-8.050
anwhile-7.987
isode-7.743
arch-7.501
Token that
Token ID326
Feature activation+7.49e-01

Pos logit contributions
IELD+2.496
hips+2.362
ymm+2.241
 Desire+2.207
 suppose+2.062
Neg logit contributions
umn-3.122
cient-2.987
arch-2.821
anwhile-2.820
isode-2.814
Token you
Token ID345
Feature activation-3.11e-01

Pos logit contributions
umn+1.504
cient+1.417
anwhile+1.358
arch+1.342
isode+1.333
Neg logit contributions
IELD-1.367
hips-1.279
ymm-1.212
 Desire-1.196
GV-1.097
Token are
Token ID389
Feature activation+2.75e-01

Pos logit contributions
IELD+0.567
hips+0.531
ymm+0.506
 Desire+0.504
 suppose+0.462
Neg logit contributions
umn-0.630
cient-0.598
anwhile-0.567
isode-0.558
arch-0.558
Token actually
Token ID1682
Feature activation-1.93e+00

Pos logit contributions
umn+0.600
cient+0.576
arch+0.546
anwhile+0.544
isode+0.539
Neg logit contributions
IELD-0.453
hips-0.413
ymm-0.395
 Desire-0.393
GV-0.351
Tokens
Token ID82
Feature activation-8.21e-01

Pos logit contributions
umn+0.038
cient+0.035
anwhile+0.034
arch+0.033
isode+0.033
Neg logit contributions
IELD-0.039
hips-0.037
 Desire-0.036
ymm-0.035
 suppose-0.034
Token work
Token ID670
Feature activation+6.95e-01

Pos logit contributions
IELD+1.430
hips+1.337
 Desire+1.264
ymm+1.259
 suppose+1.192
Neg logit contributions
umn-1.733
cient-1.662
anwhile-1.572
arch-1.563
isode-1.547
Token on
Token ID319
Feature activation+3.14e-01

Pos logit contributions
umn+1.468
cient+1.436
arch+1.344
anwhile+1.339
iously+1.319
Neg logit contributions
IELD-1.204
hips-1.071
 Desire-1.065
ymm-1.035
 suppose-0.958
Token the
Token ID262
Feature activation+1.10e+00

Pos logit contributions
umn+0.682
cient+0.651
arch+0.617
anwhile+0.615
isode+0.611
Neg logit contributions
IELD-0.524
hips-0.485
 Desire-0.462
ymm-0.458
 suppose-0.427
Token Space
Token ID4687
Feature activation+1.29e+00

Pos logit contributions
umn+2.394
cient+2.283
arch+2.178
isode+2.135
anwhile+2.134
Neg logit contributions
IELD-1.803
hips-1.642
ymm-1.573
 Desire-1.545
 suppose-1.525
Token Launch
Token ID21225
Feature activation-4.44e+00

Pos logit contributions
umn+3.170
cient+2.992
arch+2.952
anwhile+2.876
ophone+2.855
Neg logit contributions
IELD-1.744
hips-1.471
ymm-1.427
 Desire-1.399
 suppose-1.366
Token System
Token ID4482
Feature activation-9.10e-01

Pos logit contributions
IELD+3.623
hips+3.469
 Desire+3.023
ymm+2.850
 Hold+2.408
Neg logit contributions
umn-13.103
cient-12.688
isode-12.327
anwhile-12.311
arch-12.052
Token heavy
Token ID4334
Feature activation-8.16e-01

Pos logit contributions
IELD+1.651
hips+1.538
 Desire+1.480
ymm+1.469
 suppose+1.379
Neg logit contributions
umn-1.848
cient-1.753
anwhile-1.683
arch-1.662
isode-1.659
Token-
Token ID12
Feature activation-1.19e+00

Pos logit contributions
IELD+1.631
hips+1.538
 Desire+1.492
ymm+1.482
 suppose+1.393
Neg logit contributions
umn-1.498
cient-1.416
isode-1.335
arch-1.335
anwhile-1.331
Tokenlift
Token ID26282
Feature activation-1.27e+00

Pos logit contributions
IELD+2.283
hips+2.111
 Desire+2.094
ymm+2.031
 suppose+1.926
Neg logit contributions
umn-2.285
cient-2.157
arch-2.076
anwhile-2.047
isode-2.046
Token rocket
Token ID10701
Feature activation-1.49e+00

Pos logit contributions
IELD+2.099
hips+1.947
 Desire+1.869
ymm+1.854
 suppose+1.713
Neg logit contributions
umn-2.787
cient-2.659
anwhile-2.553
arch-2.526
isode-2.515
Token knowing
Token ID6970
Feature activation-2.86e-01

Pos logit contributions
IELD+1.300
hips+1.231
ymm+1.153
 Desire+1.147
 suppose+1.051
Neg logit contributions
umn-1.791
cient-1.724
anwhile-1.639
arch-1.633
isode-1.606
Token exactly
Token ID3446
Feature activation-4.79e-01

Pos logit contributions
IELD+0.491
hips+0.457
ymm+0.441
 Desire+0.433
GV+0.400
Neg logit contributions
umn-0.601
cient-0.578
anwhile-0.548
arch-0.545
isode-0.533
Token what
Token ID644
Feature activation+6.32e-01

Pos logit contributions
IELD+0.746
hips+0.687
ymm+0.661
 Desire+0.651
GV+0.595
Neg logit contributions
umn-1.087
cient-1.038
anwhile-0.989
arch-0.982
isode-0.978
Token the
Token ID262
Feature activation-6.07e-01

Pos logit contributions
umn+1.231
cient+1.168
arch+1.102
anwhile+1.097
isode+1.078
Neg logit contributions
IELD-1.168
hips-1.126
ymm-1.058
 Desire-1.051
GV-0.984
Token future
Token ID2003
Feature activation+7.88e-01

Pos logit contributions
IELD+1.125
hips+1.096
ymm+1.030
 Desire+1.029
 suppose+0.957
Neg logit contributions
umn-1.191
cient-1.139
anwhile-1.076
arch-1.068
isode-1.056
Token holds
Token ID6622
Feature activation-4.35e+00

Pos logit contributions
umn+1.507
cient+1.439
arch+1.354
isode+1.325
iously+1.312
Neg logit contributions
IELD-1.496
hips-1.417
ymm-1.351
 Desire-1.312
GV-1.226
Token.
Token ID13
Feature activation-6.02e-01

Pos logit contributions
IELD+6.243
hips+6.142
 Desire+5.679
ymm+5.512
 suppose+5.294
Neg logit contributions
umn-9.929
cient-9.408
anwhile-9.326
isode-9.117
arch-8.937
Token It
Token ID632
Feature activation+7.44e-02

Pos logit contributions
IELD+0.888
hips+0.839
ymm+0.794
 Desire+0.758
 suppose+0.727
Neg logit contributions
umn-1.408
cient-1.360
arch-1.282
isode-1.277
anwhile-1.273
Token was
Token ID373
Feature activation-3.10e-01

Pos logit contributions
umn+0.145
cient+0.139
arch+0.130
anwhile+0.129
isode+0.128
Neg logit contributions
IELD-0.141
hips-0.132
ymm-0.127
 Desire-0.127
WARE-0.114
Token cruel
Token ID12177
Feature activation+6.08e-01

Pos logit contributions
IELD+0.500
hips+0.466
ymm+0.442
 Desire+0.438
 suppose+0.391
Neg logit contributions
umn-0.685
cient-0.671
arch-0.630
anwhile-0.625
isode-0.617
Token,
Token ID11
Feature activation-1.05e+00

Pos logit contributions
umn+1.176
cient+1.145
arch+1.063
isode+1.046
iously+1.046
Neg logit contributions
IELD-1.138
hips-1.053
ymm-1.033
 Desire-1.011
GV-0.924
Token Earth
Token ID3668
Feature activation-1.12e+00

Pos logit contributions
IELD+0.138
hips+0.132
ymm+0.124
 Desire+0.123
WARE+0.115
Neg logit contributions
umn-0.147
cient-0.139
arch-0.131
anwhile-0.130
isode-0.129
Token
Token ID784
Feature activation+1.64e-01

Pos logit contributions
IELD+1.974
hips+1.858
 Desire+1.780
ymm+1.758
 suppose+1.626
Neg logit contributions
umn-2.328
cient-2.242
arch-2.107
anwhile-2.105
isode-2.103
Token far
Token ID1290
Feature activation-4.08e-01

Pos logit contributions
umn+0.351
cient+0.336
anwhile+0.314
arch+0.313
isode+0.312
Neg logit contributions
IELD-0.282
hips-0.260
 Desire-0.249
ymm-0.245
 suppose-0.224
Token more
Token ID517
Feature activation-1.38e+00

Pos logit contributions
IELD+0.699
hips+0.644
 Desire+0.619
ymm+0.612
 suppose+0.568
Neg logit contributions
umn-0.872
cient-0.840
isode-0.790
anwhile-0.788
arch-0.785
Token than
Token ID621
Feature activation+2.45e-01

Pos logit contributions
IELD+2.239
hips+2.087
 Desire+1.993
ymm+1.961
 suppose+1.807
Neg logit contributions
umn-3.047
cient-2.958
arch-2.784
anwhile-2.783
isode-2.745
Token expected
Token ID2938
Feature activation-4.32e+00

Pos logit contributions
umn+0.523
cient+0.503
arch+0.471
anwhile+0.471
isode+0.462
Neg logit contributions
IELD-0.419
hips-0.385
ymm-0.361
 Desire-0.359
WARE-0.329
Token.
Token ID13
Feature activation-3.36e-01

Pos logit contributions
IELD+6.060
hips+5.899
 Desire+5.524
ymm+5.399
 suppose+5.183
Neg logit contributions
umn-9.913
cient-9.415
anwhile-9.307
isode-9.199
ophone-8.925
Token This
Token ID770
Feature activation+2.77e-01

Pos logit contributions
IELD+0.521
hips+0.489
ymm+0.457
 Desire+0.452
 suppose+0.426
Neg logit contributions
umn-0.766
cient-0.735
arch-0.694
anwhile-0.691
isode-0.688
Token shows
Token ID2523
Feature activation-1.46e+00

Pos logit contributions
umn+0.528
cient+0.511
arch+0.475
anwhile+0.472
isode+0.467
Neg logit contributions
IELD-0.532
hips-0.496
 Desire-0.475
ymm-0.464
 suppose-0.432
Token that
Token ID326
Feature activation+2.35e-01

Pos logit contributions
IELD+2.426
hips+2.289
 Desire+2.171
ymm+2.132
 suppose+1.982
Neg logit contributions
umn-3.193
cient-3.057
anwhile-2.912
arch-2.883
isode-2.869
Token the
Token ID262
Feature activation+2.87e-01

Pos logit contributions
umn+0.513
cient+0.493
arch+0.470
anwhile+0.468
isode+0.458
Neg logit contributions
IELD-0.391
hips-0.357
 Desire-0.337
ymm-0.330
WARE-0.302
Token sermon
Token ID39362
Feature activation+2.54e+00

Pos logit contributions
umn+0.785
cient+0.749
arch+0.709
anwhile+0.701
isode+0.689
Neg logit contributions
IELD-0.807
hips-0.766
ymm-0.742
 Desire-0.718
 suppose-0.670
Token entitled
Token ID9080
Feature activation-8.17e-01

Pos logit contributions
umn+4.966
cient+4.841
arch+4.534
ophone+4.374
iously+4.369
Neg logit contributions
IELD-4.655
ymm-4.113
hips-4.028
 Desire-3.916
GV-3.702
Token '
Token ID705
Feature activation+5.65e-01

Pos logit contributions
IELD+1.373
hips+1.287
ymm+1.218
 Desire+1.201
 suppose+1.133
Neg logit contributions
umn-1.773
cient-1.699
anwhile-1.618
isode-1.609
arch-1.608
TokenRem
Token ID8413
Feature activation+3.76e-01

Pos logit contributions
umn+1.194
cient+1.147
arch+1.108
isode+1.080
anwhile+1.056
Neg logit contributions
IELD-0.955
hips-0.895
 Desire-0.854
ymm-0.852
 suppose-0.796
Tokenaining
Token ID1397
Feature activation-5.39e-02

Pos logit contributions
umn+0.770
cient+0.741
arch+0.710
ophone+0.688
isode+0.688
Neg logit contributions
IELD-0.681
hips-0.632
 Desire-0.601
ymm-0.598
 suppose-0.566
Token Aw
Token ID5851
Feature activation-4.31e+00

Pos logit contributions
IELD+0.087
hips+0.079
ymm+0.077
 Desire+0.074
 suppose+0.069
Neg logit contributions
umn-0.118
cient-0.116
anwhile-0.110
arch-0.109
isode-0.107
Tokenake
Token ID539
Feature activation-1.31e+00

Pos logit contributions
IELD+6.409
hips+6.329
 Desire+5.915
ymm+5.814
 suppose+5.317
Neg logit contributions
umn-9.554
cient-9.051
anwhile-8.892
isode-8.731
arch-8.443
Token through
Token ID832
Feature activation+4.81e-01

Pos logit contributions
IELD+2.200
hips+2.085
ymm+1.972
 Desire+1.965
 suppose+1.831
Neg logit contributions
umn-2.800
cient-2.698
anwhile-2.581
isode-2.546
arch-2.538
Token a
Token ID257
Feature activation-1.11e+00

Pos logit contributions
umn+1.098
cient+1.057
arch+1.018
anwhile+0.998
isode+0.987
Neg logit contributions
IELD-0.745
hips-0.675
ymm-0.639
 Desire-0.611
 suppose-0.601
Token Great
Token ID3878
Feature activation-1.84e+00

Pos logit contributions
IELD+1.775
hips+1.673
ymm+1.568
 Desire+1.544
 suppose+1.456
Neg logit contributions
umn-2.481
cient-2.373
anwhile-2.274
arch-2.258
isode-2.245
Token Revolution
Token ID9303
Feature activation-6.34e-01

Pos logit contributions
IELD+2.930
hips+2.800
 Desire+2.653
ymm+2.609
 suppose+2.430
Neg logit contributions
umn-4.078
cient-3.908
anwhile-3.765
isode-3.713
arch-3.679
Token,
Token ID11
Feature activation-6.83e-01

Pos logit contributions
umn+0.429
cient+0.420
arch+0.392
isode+0.390
anwhile+0.390
Neg logit contributions
IELD-0.378
hips-0.349
ymm-0.335
 Desire-0.332
 suppose-0.306
Token but
Token ID475
Feature activation+7.19e-01

Pos logit contributions
IELD+1.169
hips+1.105
ymm+1.047
 Desire+1.047
 suppose+0.947
Neg logit contributions
umn-1.457
cient-1.390
anwhile-1.322
arch-1.305
isode-1.303
Token as
Token ID355
Feature activation-7.99e-01

Pos logit contributions
umn+1.503
cient+1.436
anwhile+1.375
arch+1.356
isode+1.351
Neg logit contributions
IELD-1.234
hips-1.156
ymm-1.109
 Desire-1.081
ELF-0.970
Token you
Token ID345
Feature activation-9.41e-01

Pos logit contributions
IELD+1.398
hips+1.326
 Desire+1.253
ymm+1.228
 suppose+1.153
Neg logit contributions
umn-1.675
cient-1.594
anwhile-1.515
isode-1.509
arch-1.502
Token can
Token ID460
Feature activation-1.24e+00

Pos logit contributions
IELD+1.731
hips+1.654
 Desire+1.580
ymm+1.563
 suppose+1.465
Neg logit contributions
umn-1.864
cient-1.776
anwhile-1.701
isode-1.672
arch-1.663
Token see
Token ID766
Feature activation-4.25e+00

Pos logit contributions
IELD+2.186
hips+2.082
 Desire+1.983
ymm+1.964
 suppose+1.827
Neg logit contributions
umn-2.546
cient-2.429
anwhile-2.331
isode-2.291
arch-2.282
Token from
Token ID422
Feature activation-9.95e-01

Pos logit contributions
IELD+6.058
hips+5.948
 Desire+5.579
ymm+5.464
 suppose+5.239
Neg logit contributions
umn-9.434
cient-9.121
anwhile-8.908
isode-8.729
arch-8.634
Token the
Token ID262
Feature activation-5.86e-02

Pos logit contributions
IELD+1.421
hips+1.340
 Desire+1.246
ymm+1.241
 suppose+1.125
Neg logit contributions
umn-2.403
cient-2.295
anwhile-2.220
isode-2.181
arch-2.177
Token photos
Token ID5205
Feature activation+6.83e-01

Pos logit contributions
IELD+0.107
hips+0.101
 Desire+0.096
ymm+0.095
 suppose+0.089
Neg logit contributions
umn-0.119
cient-0.112
anwhile-0.107
arch-0.105
isode-0.105
Token,
Token ID11
Feature activation-7.83e-01

Pos logit contributions
umn+1.380
cient+1.315
anwhile+1.235
arch+1.231
isode+1.228
Neg logit contributions
IELD-1.240
hips-1.131
 Desire-1.109
ymm-1.102
 suppose-1.010
Token M
Token ID337
Feature activation-2.94e-01

Pos logit contributions
IELD+1.287
hips+1.219
ymm+1.151
 Desire+1.145
 suppose+1.041
Neg logit contributions
umn-1.718
cient-1.637
anwhile-1.580
isode-1.553
arch-1.545
Token the
Token ID262
Feature activation-6.10e-01

Pos logit contributions
IELD+3.656
hips+3.502
 Desire+3.332
ymm+3.302
 suppose+3.069
Neg logit contributions
umn-4.938
cient-4.726
anwhile-4.576
isode-4.505
arch-4.444
Token chaos
Token ID11918
Feature activation+5.11e-01

Pos logit contributions
IELD+1.167
hips+1.119
 Desire+1.066
ymm+1.060
 suppose+0.998
Neg logit contributions
umn-1.164
cient-1.107
anwhile-1.056
isode-1.039
arch-1.035
Token (
Token ID357
Feature activation-9.46e-01

Pos logit contributions
umn+1.011
cient+0.967
anwhile+0.912
isode+0.903
arch+0.903
Neg logit contributions
IELD-0.947
hips-0.886
ymm-0.856
 Desire-0.856
 suppose-0.787
Tokensp
Token ID2777
Feature activation+1.41e-01

Pos logit contributions
IELD+1.982
hips+1.896
 Desire+1.828
ymm+1.806
 suppose+1.715
Neg logit contributions
umn-1.644
cient-1.554
anwhile-1.467
arch-1.453
isode-1.447
Tokenoiler
Token ID20837
Feature activation-1.09e+00

Pos logit contributions
umn+0.197
cient+0.185
anwhile+0.170
arch+0.170
isode+0.169
Neg logit contributions
IELD-0.343
hips-0.326
 Desire-0.321
ymm-0.315
 suppose-0.303
Token alert
Token ID7995
Feature activation-4.24e+00

Pos logit contributions
IELD+1.946
hips+1.860
 Desire+1.775
ymm+1.760
 suppose+1.628
Neg logit contributions
umn-2.239
cient-2.129
anwhile-2.038
arch-2.017
isode-2.010
Token)
Token ID8
Feature activation-1.27e+00

Pos logit contributions
IELD+6.221
hips+5.972
 Desire+5.609
ymm+5.510
 suppose+5.264
Neg logit contributions
umn-9.454
cient-9.071
anwhile-8.821
isode-8.615
ophone-8.540
Token that
Token ID326
Feature activation-7.81e-01

Pos logit contributions
IELD+2.317
hips+2.213
 Desire+2.112
ymm+2.090
 suppose+1.957
Neg logit contributions
umn-2.558
cient-2.436
anwhile-2.333
isode-2.300
arch-2.286
Token Black
Token ID2619
Feature activation-1.35e+00

Pos logit contributions
IELD+1.419
hips+1.357
 Desire+1.289
ymm+1.278
 suppose+1.196
Neg logit contributions
umn-1.571
cient-1.494
anwhile-1.428
isode-1.409
arch-1.403
Token inadvertently
Token ID29243
Feature activation-2.12e-01

Pos logit contributions
IELD+2.572
hips+2.468
 Desire+2.358
ymm+2.335
 suppose+2.201
Neg logit contributions
umn-2.582
cient-2.455
anwhile-2.348
isode-2.312
arch-2.292
Token sets
Token ID5621
Feature activation-7.45e-01

Pos logit contributions
IELD+0.412
hips+0.388
 Desire+0.375
ymm+0.372
 suppose+0.347
Neg logit contributions
umn-0.404
cient-0.382
anwhile-0.363
arch-0.359
isode-0.358
Token
Token ID250
Feature activation+7.52e-01

Pos logit contributions
umn+0.912
cient+0.875
arch+0.824
isode+0.812
anwhile+0.804
Neg logit contributions
IELD-0.723
hips-0.687
 Desire-0.656
ymm-0.647
 suppose-0.604
TokenF
Token ID37
Feature activation-2.69e-01

Pos logit contributions
umn+1.583
cient+1.521
anwhile+1.436
arch+1.436
isode+1.432
Neg logit contributions
IELD-1.245
hips-1.209
 Desire-1.160
ymm-1.143
 suppose-1.077
TokenTC
Token ID4825
Feature activation+5.64e-01

Pos logit contributions
IELD+0.482
hips+0.465
 Desire+0.453
ymm+0.446
 suppose+0.412
Neg logit contributions
umn-0.536
cient-0.519
arch-0.490
isode-0.488
anwhile-0.483
Token Compl
Token ID20011
Feature activation-1.40e+00

Pos logit contributions
umn+1.270
cient+1.207
anwhile+1.152
isode+1.145
arch+1.144
Neg logit contributions
IELD-0.880
hips-0.838
ymm-0.787
 Desire-0.775
 suppose-0.727
Tokenaint
Token ID2913
Feature activation-4.82e-01

Pos logit contributions
IELD+2.796
hips+2.679
 Desire+2.570
ymm+2.554
 suppose+2.400
Neg logit contributions
umn-2.564
cient-2.429
anwhile-2.289
isode-2.275
arch-2.256
Token Assistant
Token ID15286
Feature activation-4.23e+00

Pos logit contributions
IELD+0.711
hips+0.673
ymm+0.639
 Desire+0.628
 suppose+0.579
Neg logit contributions
umn-1.135
cient-1.084
anwhile-1.040
arch-1.033
isode-1.032
Token
Token ID447
Feature activation-4.56e-01

Pos logit contributions
IELD+7.912
hips+7.678
 Desire+7.240
ymm+7.033
 suppose+6.768
Neg logit contributions
umn-7.801
cient-7.508
anwhile-7.253
isode-7.048
arch-6.914
Token
Token ID251
Feature activation+2.81e-01

Pos logit contributions
IELD+0.833
hips+0.791
 Desire+0.759
ymm+0.751
 suppose+0.706
Neg logit contributions
umn-0.916
cient-0.871
anwhile-0.821
arch-0.820
isode-0.816
Token to
Token ID284
Feature activation+1.50e-01

Pos logit contributions
umn+0.465
cient+0.435
anwhile+0.408
isode+0.406
arch+0.402
Neg logit contributions
IELD-0.609
hips-0.578
ymm-0.567
 Desire-0.562
 suppose-0.526
Token report
Token ID989
Feature activation+3.74e-01

Pos logit contributions
umn+0.270
cient+0.254
anwhile+0.242
arch+0.238
isode+0.237
Neg logit contributions
IELD-0.300
hips-0.289
ymm-0.278
 Desire-0.275
 suppose-0.256
Token persons
Token ID6506
Feature activation+1.93e+00

Pos logit contributions
umn+0.795
cient+0.772
isode+0.724
anwhile+0.724
arch+0.720
Neg logit contributions
IELD-0.633
hips-0.584
ymm-0.568
 Desire-0.561
 suppose-0.507
Token competed
Token ID32440
Feature activation+6.07e-01

Pos logit contributions
IELD+1.341
hips+1.275
 Desire+1.197
ymm+1.197
 suppose+1.117
Neg logit contributions
umn-1.522
cient-1.464
isode-1.381
anwhile-1.373
arch-1.372
Token alongside
Token ID7848
Feature activation-2.90e-01

Pos logit contributions
umn+1.257
cient+1.243
arch+1.180
iously+1.146
anwhile+1.140
Neg logit contributions
IELD-1.037
hips-0.916
 Desire-0.909
ymm-0.891
 suppose-0.820
Token Chelsea
Token ID12147
Feature activation-1.23e+00

Pos logit contributions
IELD+0.465
hips+0.440
ymm+0.420
 Desire+0.404
 suppose+0.383
Neg logit contributions
umn-0.638
cient-0.612
arch-0.588
anwhile-0.587
isode-0.574
Token and
Token ID290
Feature activation+2.05e+00

Pos logit contributions
IELD+2.058
hips+1.944
 Desire+1.851
ymm+1.839
 suppose+1.695
Neg logit contributions
umn-2.635
cient-2.521
anwhile-2.407
arch-2.403
isode-2.387
Token Tottenham
Token ID24272
Feature activation-4.00e+00

Pos logit contributions
umn+4.212
cient+4.066
arch+3.934
anwhile+3.906
ophone+3.793
Neg logit contributions
IELD-3.296
hips-3.144
ymm-3.050
 Desire-2.838
WARE-2.717
Token Hots
Token ID41239
Feature activation-4.22e+00

Pos logit contributions
IELD+5.986
hips+5.845
 Desire+5.414
ymm+5.325
 suppose+4.994
Neg logit contributions
umn-9.083
cient-8.642
anwhile-8.415
isode-8.272
arch-8.079
Tokenpur
Token ID14225
Feature activation-1.81e+00

Pos logit contributions
hips+7.312
IELD+7.113
 Desire+6.560
ymm+6.463
 suppose+6.123
Neg logit contributions
umn-8.526
anwhile-8.310
cient-8.039
isode-7.824
theless-7.654
Token for
Token ID329
Feature activation+5.97e-01

Pos logit contributions
IELD+2.975
hips+2.802
 Desire+2.669
ymm+2.643
 suppose+2.451
Neg logit contributions
umn-3.953
cient-3.798
anwhile-3.633
isode-3.597
arch-3.591
Token the
Token ID262
Feature activation+1.30e+00

Pos logit contributions
umn+1.294
cient+1.252
isode+1.174
arch+1.172
ophone+1.161
Neg logit contributions
IELD-0.961
hips-0.903
ymm-0.861
 Desire-0.812
WARE-0.776
Token Premier
Token ID9952
Feature activation+1.27e-01

Pos logit contributions
umn+2.579
cient+2.498
arch+2.325
isode+2.284
anwhile+2.248
Neg logit contributions
IELD-2.299
hips-2.190
ymm-2.093
 Desire-2.038
 suppose-1.928
Token League
Token ID4041
Feature activation-1.95e-01

Pos logit contributions
umn+0.331
cient+0.322
arch+0.308
isode+0.306
anwhile+0.304
Neg logit contributions
IELD-0.149
hips-0.138
 Desire-0.126
ymm-0.126
 suppose-0.112
Token (
Token ID357
Feature activation+1.18e-02

Pos logit contributions
umn+2.044
cient+1.961
anwhile+1.835
isode+1.820
arch+1.806
Neg logit contributions
IELD-1.786
hips-1.673
ymm-1.635
 Desire-1.551
 suppose-1.482
Tokencontin
Token ID18487
Feature activation+6.41e-01

Pos logit contributions
umn+0.022
cient+0.021
arch+0.020
isode+0.020
anwhile+0.020
Neg logit contributions
IELD-0.023
hips-0.022
 Desire-0.021
ymm-0.021
 suppose-0.020
Tokenued
Token ID1739
Feature activation-5.36e-01

Pos logit contributions
umn+1.027
cient+0.965
arch+0.883
isode+0.880
anwhile+0.876
Neg logit contributions
IELD-1.453
hips-1.379
 Desire-1.351
ymm-1.332
 suppose-1.267
Token).
Token ID737
Feature activation-3.88e-01

Pos logit contributions
IELD+0.881
hips+0.832
ymm+0.793
 Desire+0.783
 suppose+0.724
Neg logit contributions
umn-1.184
cient-1.125
anwhile-1.076
arch-1.070
isode-1.065
Token Extreme
Token ID18111
Feature activation-1.85e+00

Pos logit contributions
IELD+0.571
hips+0.532
ymm+0.510
 Desire+0.475
 suppose+0.459
Neg logit contributions
umn-0.912
cient-0.876
arch-0.830
isode-0.828
anwhile-0.825
Token Un
Token ID791
Feature activation-4.21e+00

Pos logit contributions
IELD+3.109
hips+2.959
ymm+2.788
 Desire+2.782
 suppose+2.584
Neg logit contributions
umn-3.989
cient-3.809
anwhile-3.628
isode-3.602
arch-3.582
Tokenction
Token ID596
Feature activation+8.65e-01

Pos logit contributions
hips+7.772
IELD+7.762
 Desire+7.370
ymm+7.233
 suppose+6.761
Neg logit contributions
umn-7.629
anwhile-7.392
cient-7.200
isode-6.873
ophone-6.710
Token.
Token ID13
Feature activation-5.45e-01

Pos logit contributions
umn+1.752
cient+1.666
arch+1.561
anwhile+1.544
isode+1.536
Neg logit contributions
IELD-1.550
hips-1.411
ymm-1.404
 Desire-1.338
 suppose-1.291
Token Holy
Token ID7439
Feature activation+8.12e-01

Pos logit contributions
IELD+0.806
hips+0.755
ymm+0.715
 Desire+0.676
 suppose+0.651
Neg logit contributions
umn-1.286
cient-1.230
arch-1.169
isode-1.165
anwhile-1.162
Token Orders
Token ID30689
Feature activation-1.22e+00

Pos logit contributions
umn+1.767
cient+1.683
arch+1.603
isode+1.590
ocide+1.580
Neg logit contributions
IELD-1.318
hips-1.219
ymm-1.174
 Desire-1.103
 suppose-1.079
Token.
Token ID13
Feature activation-3.55e-01

Pos logit contributions
IELD+2.022
hips+1.893
ymm+1.808
 Desire+1.802
 suppose+1.674
Neg logit contributions
umn-2.642
cient-2.536
anwhile-2.418
isode-2.381
arch-2.375
Token industrial
Token ID7593
Feature activation+2.03e+00

Pos logit contributions
umn+1.010
cient+0.976
anwhile+0.935
arch+0.924
isode+0.908
Neg logit contributions
IELD-0.808
hips-0.754
 Desire-0.726
ymm-0.723
ENTION-0.655
Token output
Token ID5072
Feature activation-1.43e+00

Pos logit contributions
umn+3.186
cient+3.080
anwhile+2.990
arch+2.867
iously+2.864
Neg logit contributions
IELD-4.332
hips-4.120
ymm-3.987
 Desire-3.908
ELF-3.742
Token rose
Token ID8278
Feature activation-2.66e-01

Pos logit contributions
IELD+2.706
hips+2.560
 Desire+2.464
ymm+2.431
 suppose+2.266
Neg logit contributions
umn-2.765
cient-2.640
anwhile-2.525
isode-2.478
arch-2.476
Token more
Token ID517
Feature activation+2.96e-01

Pos logit contributions
IELD+0.447
hips+0.411
 Desire+0.398
ymm+0.392
 suppose+0.354
Neg logit contributions
umn-0.568
cient-0.547
anwhile-0.527
arch-0.520
isode-0.514
Token than
Token ID621
Feature activation+1.53e+00

Pos logit contributions
umn+0.584
cient+0.564
anwhile+0.542
arch+0.534
isode+0.523
Neg logit contributions
IELD-0.540
hips-0.495
 Desire-0.491
ymm-0.481
 suppose-0.439
Token expected
Token ID2938
Feature activation-4.20e+00

Pos logit contributions
umn+3.117
cient+3.033
arch+2.832
anwhile+2.820
isode+2.785
Neg logit contributions
IELD-2.561
hips-2.381
ymm-2.253
 Desire-2.239
GV-2.076
Token in
Token ID287
Feature activation+5.74e-03

Pos logit contributions
IELD+6.002
hips+5.898
 Desire+5.433
ymm+5.404
 suppose+5.109
Neg logit contributions
umn-9.641
cient-9.128
anwhile-8.867
isode-8.852
arch-8.650
Token September
Token ID2693
Feature activation+1.13e+00

Pos logit contributions
umn+0.013
cient+0.012
arch+0.011
anwhile+0.011
isode+0.011
Neg logit contributions
IELD-0.009
hips-0.009
 Desire-0.008
ymm-0.008
 suppose-0.008
Token,
Token ID11
Feature activation-2.20e-01

Pos logit contributions
umn+2.180
cient+2.060
anwhile+1.970
arch+1.939
isode+1.912
Neg logit contributions
IELD-2.095
hips-1.949
 Desire-1.905
ymm-1.901
ELF-1.737
Token boosted
Token ID29657
Feature activation-5.51e-01

Pos logit contributions
IELD+0.376
hips+0.358
 Desire+0.340
ymm+0.334
 suppose+0.311
Neg logit contributions
umn-0.464
cient-0.444
anwhile-0.422
arch-0.418
isode-0.414
Token by
Token ID416
Feature activation-6.55e-01

Pos logit contributions
IELD+0.905
hips+0.855
 Desire+0.812
ymm+0.798
 suppose+0.737
Neg logit contributions
umn-1.198
cient-1.154
anwhile-1.117
arch-1.087
isode-1.080
Token handling
Token ID9041
Feature activation-1.03e+00

Pos logit contributions
umn+0.470
cient+0.449
anwhile+0.425
arch+0.419
ophone+0.416
Neg logit contributions
IELD-0.435
hips-0.419
ymm-0.398
 Desire-0.392
 suppose-0.367
Token of
Token ID286
Feature activation-9.91e-01

Pos logit contributions
IELD+1.867
hips+1.774
ymm+1.677
 Desire+1.676
 suppose+1.562
Neg logit contributions
umn-2.086
cient-1.988
anwhile-1.898
arch-1.858
isode-1.855
Token Mam
Token ID29926
Feature activation-1.19e+00

Pos logit contributions
IELD+1.789
hips+1.722
ymm+1.617
 Desire+1.614
 suppose+1.505
Neg logit contributions
umn-2.004
cient-1.919
anwhile-1.819
arch-1.793
isode-1.780
Tokenad
Token ID324
Feature activation+6.59e-01

Pos logit contributions
IELD+2.790
hips+2.661
 Desire+2.558
ymm+2.545
 suppose+2.438
Neg logit contributions
umn-1.779
cient-1.681
anwhile-1.554
arch-1.553
isode-1.535
Tokenou
Token ID280
Feature activation+4.17e-01

Pos logit contributions
umn+1.106
cient+1.050
arch+0.981
isode+0.967
ophone+0.952
Neg logit contributions
IELD-1.455
hips-1.344
ymm-1.298
 Desire-1.256
 suppose-1.229
Token Sak
Token ID13231
Feature activation-4.20e+00

Pos logit contributions
umn+0.853
cient+0.812
arch+0.791
anwhile+0.783
isode+0.772
Neg logit contributions
IELD-0.710
hips-0.661
ymm-0.661
 Desire-0.615
GV-0.583
Tokenho
Token ID8873
Feature activation+3.06e-01

Pos logit contributions
hips+5.653
IELD+5.543
 Desire+5.221
ymm+5.082
 suppose+4.536
Neg logit contributions
umn-9.660
cient-9.194
anwhile-9.144
theless-8.910
isode-8.834
Token drugs
Token ID5010
Feature activation+2.01e+00

Pos logit contributions
umn+0.625
cient+0.597
anwhile+0.576
arch+0.564
ophone+0.557
Neg logit contributions
IELD-0.537
hips-0.518
ymm-0.486
 Desire-0.483
 suppose-0.445
Token test
Token ID1332
Feature activation-1.09e+00

Pos logit contributions
umn+4.064
cient+3.884
anwhile+3.753
ophone+3.651
arch+3.639
Neg logit contributions
IELD-3.448
hips-3.217
ymm-3.074
 Desire-2.991
 suppose-2.753
Token Read
Token ID4149
Feature activation+6.31e-01

Pos logit contributions
IELD+1.687
hips+1.594
 Desire+1.497
ymm+1.493
 suppose+1.374
Neg logit contributions
umn-2.474
cient-2.376
anwhile-2.296
isode-2.251
arch-2.246
Token more
Token ID517
Feature activation+9.09e-01

Pos logit contributions
umn+1.280
cient+1.219
arch+1.153
isode+1.143
anwhile+1.141
Neg logit contributions
IELD-1.137
hips-1.066
 Desire-1.021
ymm-1.004
 suppose-0.937
Token first
Token ID717
Feature activation+9.16e-01

Pos logit contributions
umn+2.725
cient+2.633
anwhile+2.466
arch+2.458
isode+2.440
Neg logit contributions
IELD-2.365
hips-2.259
ymm-2.123
 Desire-2.069
 suppose-1.959
Token Thermal
Token ID41590
Feature activation-4.64e-01

Pos logit contributions
umn+1.971
cient+1.903
anwhile+1.789
arch+1.782
isode+1.781
Neg logit contributions
IELD-1.504
hips-1.426
ymm-1.333
 Desire-1.309
 suppose-1.229
Token Disc
Token ID8444
Feature activation-4.69e-01

Pos logit contributions
IELD+0.678
hips+0.652
ymm+0.605
 Desire+0.593
 suppose+0.553
Neg logit contributions
umn-1.094
cient-1.046
anwhile-1.012
isode-1.004
arch-1.000
Tokenour
Token ID454
Feature activation-7.87e-02

Pos logit contributions
IELD+0.827
hips+0.796
 Desire+0.752
ymm+0.747
 suppose+0.699
Neg logit contributions
umn-0.967
cient-0.916
anwhile-0.878
isode-0.870
arch-0.868
Tokenagement
Token ID5082
Feature activation-9.38e-01

Pos logit contributions
IELD+0.124
hips+0.116
ymm+0.107
 Desire+0.106
 suppose+0.101
Neg logit contributions
umn-0.181
cient-0.172
arch-0.163
isode-0.162
anwhile-0.162
Token Beam
Token ID25855
Feature activation-4.20e+00

Pos logit contributions
IELD+1.479
hips+1.402
ymm+1.319
 Desire+1.303
 suppose+1.211
Neg logit contributions
umn-2.107
cient-2.018
anwhile-1.925
isode-1.920
arch-1.908
Token test
Token ID1332
Feature activation-1.46e+00

Pos logit contributions
IELD+7.306
hips+7.045
 Desire+6.767
ymm+6.570
 suppose+6.235
Neg logit contributions
umn-8.344
cient-7.930
anwhile-7.688
isode-7.378
arch-7.369
Token (
Token ID357
Feature activation-3.73e-01

Pos logit contributions
IELD+2.285
hips+2.186
 Desire+2.057
ymm+2.025
 suppose+1.893
Neg logit contributions
umn-3.270
cient-3.130
anwhile-3.022
isode-2.973
arch-2.952
Token2
Token ID17
Feature activation-1.15e-01

Pos logit contributions
IELD+0.710
hips+0.679
 Desire+0.650
ymm+0.644
 suppose+0.608
Neg logit contributions
umn-0.714
cient-0.679
anwhile-0.644
isode-0.638
arch-0.635
Token)
Token ID8
Feature activation+9.63e-02

Pos logit contributions
IELD+0.172
hips+0.162
 Desire+0.153
ymm+0.151
 suppose+0.139
Neg logit contributions
umn-0.267
cient-0.257
arch-0.243
anwhile-0.243
isode-0.243
Token
Token ID198
Feature activation+3.58e-01

Pos logit contributions
umn+0.224
cient+0.215
anwhile+0.207
isode+0.204
arch+0.203
Neg logit contributions
IELD-0.143
hips-0.136
 Desire-0.128
ymm-0.127
 suppose-0.117
Token third
Token ID2368
Feature activation+1.89e+00

Pos logit contributions
umn+2.273
cient+2.200
arch+2.050
anwhile+2.017
isode+2.013
Neg logit contributions
IELD-2.037
hips-1.914
ymm-1.813
 Desire-1.791
 suppose-1.728
Token Grav
Token ID32599
Feature activation-6.00e-01

Pos logit contributions
umn+3.634
cient+3.499
arch+3.261
isode+3.204
ophone+3.164
Neg logit contributions
IELD-3.549
hips-3.283
 Desire-3.162
ymm-3.113
 suppose-2.913
Tokenitational
Token ID22181
Feature activation+1.62e-01

Pos logit contributions
IELD+1.323
hips+1.269
 Desire+1.231
ymm+1.208
 suppose+1.163
Neg logit contributions
umn-0.967
cient-0.923
arch-0.857
isode-0.851
anwhile-0.843
Token L
Token ID406
Feature activation-3.02e+00

Pos logit contributions
umn+0.323
cient+0.307
arch+0.297
ophone+0.284
isode+0.282
Neg logit contributions
IELD-0.298
hips-0.280
ymm-0.262
 Desire-0.260
 suppose-0.252
Tokenensing
Token ID26426
Feature activation+3.48e-01

Pos logit contributions
IELD+5.384
hips+5.167
ymm+4.891
 Desire+4.838
 suppose+4.571
Neg logit contributions
umn-5.965
cient-5.704
anwhile-5.612
isode-5.449
arch-5.309
Token Accuracy
Token ID33222
Feature activation-4.17e+00

Pos logit contributions
umn+0.699
cient+0.667
arch+0.634
isode+0.619
ophone+0.617
Neg logit contributions
IELD-0.626
hips-0.583
ymm-0.562
 Desire-0.552
 suppose-0.528
Token Testing
Token ID23983
Feature activation-3.54e+00

Pos logit contributions
IELD+6.257
hips+6.051
 Desire+5.768
ymm+5.605
 suppose+5.193
Neg logit contributions
umn-9.233
cient-8.875
anwhile-8.784
isode-8.559
arch-8.329
Token Challenge
Token ID13879
Feature activation-2.47e+00

Pos logit contributions
IELD+5.527
hips+5.372
 Desire+5.070
ymm+4.919
 suppose+4.610
Neg logit contributions
umn-7.685
cient-7.393
anwhile-7.281
isode-7.029
arch-6.947
Token (
Token ID357
Feature activation-9.06e-01

Pos logit contributions
IELD+3.976
hips+3.832
 Desire+3.620
ymm+3.563
 suppose+3.346
Neg logit contributions
umn-5.361
cient-5.123
anwhile-5.005
isode-4.887
arch-4.830
TokenGRE
Token ID28934
Feature activation+1.49e+00

Pos logit contributions
IELD+1.355
hips+1.285
 Desire+1.221
ymm+1.197
 suppose+1.109
Neg logit contributions
umn-2.104
cient-2.019
anwhile-1.933
isode-1.920
arch-1.910
TokenAT
Token ID1404
Feature activation-1.42e+00

Pos logit contributions
umn+2.830
cient+2.763
arch+2.651
isode+2.610
anwhile+2.609
Neg logit contributions
hips-2.621
IELD-2.601
 Desire-2.476
ymm-2.472
 suppose-2.290
Token warp
Token ID25825
Feature activation+4.80e-01

Pos logit contributions
IELD+0.591
hips+0.552
ymm+0.538
 Desire+0.536
 suppose+0.498
Neg logit contributions
umn-0.637
cient-0.612
anwhile-0.577
isode-0.576
arch-0.575
Token drive
Token ID3708
Feature activation-3.71e-01

Pos logit contributions
umn+0.903
cient+0.876
arch+0.821
anwhile+0.818
isode+0.817
Neg logit contributions
IELD-0.919
hips-0.873
 Desire-0.839
ymm-0.837
 suppose-0.773
Token website
Token ID3052
Feature activation+1.66e+00

Pos logit contributions
IELD+0.685
hips+0.643
 Desire+0.620
ymm+0.619
 suppose+0.572
Neg logit contributions
umn-0.731
cient-0.701
anwhile-0.665
isode-0.661
arch-0.659
Token for
Token ID329
Feature activation+2.82e-03

Pos logit contributions
umn+3.429
cient+3.286
isode+3.112
arch+3.107
anwhile+3.079
Neg logit contributions
IELD-2.791
hips-2.562
ymm-2.522
 Desire-2.476
 suppose-2.260
Token your
Token ID534
Feature activation+1.19e-01

Pos logit contributions
umn+0.006
cient+0.006
anwhile+0.005
arch+0.005
isode+0.005
Neg logit contributions
IELD-0.005
hips-0.004
ymm-0.004
 Desire-0.004
 suppose-0.004
Token business
Token ID1597
Feature activation+2.63e+00

Pos logit contributions
umn+0.236
cient+0.225
anwhile+0.212
isode+0.211
arch+0.211
Neg logit contributions
IELD-0.216
hips-0.205
ymm-0.196
 Desire-0.195
 suppose-0.183
Token and
Token ID290
Feature activation+1.23e+00

Pos logit contributions
umn+5.157
cient+4.997
arch+4.716
isode+4.653
anwhile+4.564
Neg logit contributions
IELD-4.610
hips-4.084
ymm-4.080
 Desire-4.017
 suppose-3.736
Token engage
Token ID8209
Feature activation+2.70e-01

Pos logit contributions
umn+2.412
cient+2.313
arch+2.176
anwhile+2.172
isode+2.168
Neg logit contributions
IELD-2.233
hips-2.060
ymm-2.030
 Desire-1.991
 suppose-1.831
Token your
Token ID534
Feature activation+4.48e-02

Pos logit contributions
umn+0.569
cient+0.557
arch+0.519
anwhile+0.518
isode+0.516
Neg logit contributions
IELD-0.457
hips-0.421
ymm-0.408
 Desire-0.406
 suppose-0.383
Token customers
Token ID4297
Feature activation-1.21e-02

Pos logit contributions
umn+0.096
cient+0.093
anwhile+0.087
arch+0.087
isode+0.087
Neg logit contributions
IELD-0.075
hips-0.070
ymm-0.067
 Desire-0.067
 suppose-0.062
Token.
Token ID13
Feature activation+6.83e-02

Pos logit contributions
IELD+0.020
hips+0.018
ymm+0.018
 Desire+0.018
 suppose+0.017
Neg logit contributions
umn-0.026
cient-0.026
anwhile-0.024
arch-0.024
isode-0.024
Token child
Token ID1200
Feature activation+1.76e+00

Pos logit contributions
IELD+1.150
hips+1.093
ymm+1.033
 Desire+1.025
 suppose+0.950
Neg logit contributions
umn-1.331
cient-1.272
anwhile-1.206
isode-1.197
arch-1.193
Token.
Token ID13
Feature activation+2.93e-01

Pos logit contributions
umn+3.306
cient+3.173
arch+2.958
isode+2.913
ophone+2.896
Neg logit contributions
IELD-3.376
hips-3.091
 Desire-2.988
ymm-2.965
ELF-2.785
Token But
Token ID887
Feature activation-3.45e-01

Pos logit contributions
umn+0.651
cient+0.627
arch+0.592
isode+0.583
ophone+0.582
Neg logit contributions
IELD-0.455
hips-0.424
ymm-0.403
 Desire-0.384
 suppose-0.372
Token eight
Token ID3624
Feature activation-1.03e+00

Pos logit contributions
IELD+0.569
hips+0.525
ymm+0.494
 Desire+0.492
ELF+0.443
Neg logit contributions
umn-0.753
cient-0.724
anwhile-0.690
arch-0.682
isode-0.673
Token-
Token ID12
Feature activation+7.31e-01

Pos logit contributions
IELD+2.195
hips+2.099
 Desire+2.037
ymm+2.004
 suppose+1.898
Neg logit contributions
umn-1.761
cient-1.667
anwhile-1.558
isode-1.536
arch-1.532
Tokenyear
Token ID1941
Feature activation+2.84e+00

Pos logit contributions
umn+0.752
cient+0.678
arch+0.593
anwhile+0.588
isode+0.585
Neg logit contributions
IELD-2.050
hips-1.964
 Desire-1.945
ymm-1.899
 suppose-1.839
Token-
Token ID12
Feature activation-4.51e-01

Pos logit contributions
umn+3.777
cient+3.528
arch+3.139
anwhile+3.119
isode+3.063
Neg logit contributions
IELD-6.748
hips-6.482
 Desire-6.381
ymm-6.320
 suppose-5.897
Tokenolds
Token ID10119
Feature activation+6.73e-02

Pos logit contributions
IELD+0.916
hips+0.890
 Desire+0.847
ymm+0.841
 suppose+0.797
Neg logit contributions
umn-0.797
cient-0.755
anwhile-0.731
isode-0.712
arch-0.698
Token who
Token ID508
Feature activation+8.97e-01

Pos logit contributions
umn+0.136
cient+0.131
arch+0.123
anwhile+0.123
isode+0.121
Neg logit contributions
IELD-0.122
hips-0.110
ymm-0.108
 Desire-0.107
ENTION-0.098
Token aspire
Token ID42926
Feature activation-3.70e-01

Pos logit contributions
umn+1.774
cient+1.688
anwhile+1.596
arch+1.581
isode+1.565
Neg logit contributions
IELD-1.675
hips-1.530
ymm-1.480
 Desire-1.472
WARE-1.361
Token to
Token ID284
Feature activation-7.45e-01

Pos logit contributions
IELD+0.676
hips+0.632
 Desire+0.610
ymm+0.608
 suppose+0.567
Neg logit contributions
umn-0.739
cient-0.711
anwhile-0.669
arch-0.665
isode-0.664
Token 1800
Token ID21431
Feature activation+1.12e+00

Pos logit contributions
umn+0.793
cient+0.776
arch+0.730
anwhile+0.725
isode+0.718
Neg logit contributions
IELD-0.591
hips-0.555
 Desire-0.518
ymm-0.518
 suppose-0.472
Token.
Token ID13
Feature activation-5.49e-01

Pos logit contributions
umn+2.214
cient+2.114
arch+2.017
anwhile+1.979
isode+1.960
Neg logit contributions
IELD-2.049
hips-1.934
 Desire-1.865
ymm-1.839
 suppose-1.682
Token Learn
Token ID14365
Feature activation-1.01e-02

Pos logit contributions
IELD+0.819
hips+0.768
ymm+0.720
 Desire+0.713
 suppose+0.662
Neg logit contributions
umn-1.281
cient-1.239
anwhile-1.173
arch-1.173
isode-1.166
Token about
Token ID546
Feature activation+4.98e-01

Pos logit contributions
IELD+0.017
hips+0.016
ymm+0.015
 Desire+0.015
 suppose+0.014
Neg logit contributions
umn-0.021
cient-0.021
anwhile-0.019
arch-0.019
isode-0.019
Token the
Token ID262
Feature activation+6.37e-01

Pos logit contributions
umn+1.116
cient+1.093
arch+1.029
anwhile+1.020
isode+0.999
Neg logit contributions
IELD-0.785
hips-0.726
ymm-0.688
 Desire-0.679
 suppose-0.621
Token history
Token ID2106
Feature activation+3.15e+00

Pos logit contributions
umn+1.111
cient+1.073
arch+0.994
anwhile+0.991
isode+0.955
Neg logit contributions
IELD-1.325
hips-1.261
ymm-1.199
 Desire-1.191
 suppose-1.132
Token of
Token ID286
Feature activation+1.94e-01

Pos logit contributions
umn+5.665
cient+5.459
anwhile+5.042
arch+5.021
isode+4.942
Neg logit contributions
IELD-6.053
 Desire-5.535
ymm-5.520
hips-5.489
 suppose-5.029
Token batteries
Token ID13591
Feature activation-3.20e-01

Pos logit contributions
umn+0.442
cient+0.433
arch+0.409
anwhile+0.408
isode+0.399
Neg logit contributions
IELD-0.301
hips-0.279
 Desire-0.259
ymm-0.259
 suppose-0.235
Token and
Token ID290
Feature activation+7.33e-01

Pos logit contributions
IELD+0.549
hips+0.502
 Desire+0.486
ymm+0.485
 suppose+0.442
Neg logit contributions
umn-0.686
cient-0.662
anwhile-0.629
arch-0.620
isode-0.615
Token find
Token ID1064
Feature activation-7.08e-01

Pos logit contributions
umn+1.557
cient+1.512
anwhile+1.439
arch+1.421
isode+1.405
Neg logit contributions
IELD-1.232
hips-1.150
 Desire-1.091
ymm-1.091
GV-0.971
Token out
Token ID503
Feature activation-3.51e-01

Pos logit contributions
IELD+1.275
hips+1.206
ymm+1.150
 Desire+1.145
 suppose+1.067
Neg logit contributions
umn-1.437
cient-1.394
arch-1.300
anwhile-1.295
isode-1.288
Token.
Token ID13
Feature activation+3.36e-02

Pos logit contributions
IELD+2.072
hips+1.972
 Desire+1.846
ymm+1.820
 suppose+1.682
Neg logit contributions
umn-3.291
cient-3.160
anwhile-3.051
isode-3.024
arch-2.984
Token dollars
Token ID5054
Feature activation+1.36e-01

Pos logit contributions
umn+0.068
cient+0.065
anwhile+0.062
arch+0.061
isode+0.060
Neg logit contributions
IELD-0.060
hips-0.057
ymm-0.054
 Desire-0.054
 suppose-0.050
Token from
Token ID422
Feature activation-1.33e+00

Pos logit contributions
umn+0.271
cient+0.259
anwhile+0.247
arch+0.243
isode+0.241
Neg logit contributions
IELD-0.251
hips-0.232
 Desire-0.227
ymm-0.224
 suppose-0.208
Token rem
Token ID816
Feature activation+1.62e+00

Pos logit contributions
IELD+2.320
hips+2.203
 Desire+2.099
ymm+2.085
 suppose+1.947
Neg logit contributions
umn-2.746
cient-2.629
anwhile-2.520
isode-2.475
arch-2.471
Tokenitt
Token ID715
Feature activation+1.48e+00

Pos logit contributions
umn+3.264
cient+3.179
arch+3.048
anwhile+2.971
ophone+2.970
Neg logit contributions
IELD-2.800
 Desire-2.585
hips-2.547
ymm-2.538
 suppose-2.343
Tokenances
Token ID1817
Feature activation+2.71e+00

Pos logit contributions
umn+3.423
cient+3.363
anwhile+3.258
arch+3.201
ophone+3.101
Neg logit contributions
IELD-2.143
 Desire-1.934
ymm-1.934
hips-1.900
 suppose-1.712
Token,
Token ID11
Feature activation-8.48e-01

Pos logit contributions
umn+4.934
cient+4.745
anwhile+4.547
arch+4.482
isode+4.369
Neg logit contributions
IELD-5.158
ymm-4.634
 Desire-4.598
hips-4.494
ENTION-4.207
Token tourism
Token ID19277
Feature activation+1.80e+00

Pos logit contributions
IELD+1.564
hips+1.476
 Desire+1.422
ymm+1.406
 suppose+1.310
Neg logit contributions
umn-1.677
cient-1.605
anwhile-1.541
isode-1.505
arch-1.502
Token and
Token ID290
Feature activation+1.17e+00

Pos logit contributions
umn+3.279
cient+3.122
arch+2.933
anwhile+2.930
isode+2.894
Neg logit contributions
IELD-3.465
ymm-3.216
 Desire-3.173
hips-3.132
 suppose-2.907
Token pal
Token ID6340
Feature activation-4.26e-01

Pos logit contributions
umn+2.121
cient+2.061
anwhile+1.948
arch+1.911
ophone+1.883
Neg logit contributions
IELD-2.276
hips-2.092
 Desire-2.052
ymm-2.049
 suppose-1.899
Tokenad
Token ID324
Feature activation+8.36e-01

Pos logit contributions
IELD+0.982
hips+0.934
 Desire+0.909
ymm+0.901
 suppose+0.865
Neg logit contributions
umn-0.657
cient-0.618
arch-0.569
isode-0.564
anwhile-0.563
Token with
Token ID351
Feature activation-4.72e-01

Pos logit contributions
IELD+0.699
hips+0.659
 Desire+0.635
ymm+0.631
 suppose+0.582
Neg logit contributions
umn-0.673
cient-0.635
anwhile-0.605
isode-0.601
arch-0.597
Token Bernie
Token ID10477
Feature activation-6.06e-01

Pos logit contributions
IELD+0.859
hips+0.816
ymm+0.765
 Desire+0.759
 suppose+0.709
Neg logit contributions
umn-0.953
cient-0.911
anwhile-0.861
arch-0.860
isode-0.844
Token Sanders
Token ID5831
Feature activation-2.95e+00

Pos logit contributions
IELD+1.097
hips+1.053
ymm+0.988
 Desire+0.985
 suppose+0.913
Neg logit contributions
umn-1.219
cient-1.164
anwhile-1.112
arch-1.105
isode-1.101
Token Newt
Token ID40893
Feature activation+2.02e+00

Pos logit contributions
IELD+4.530
hips+4.355
 Desire+4.140
ymm+4.033
 suppose+3.771
Neg logit contributions
umn-6.616
cient-6.352
anwhile-6.144
isode-6.033
arch-5.965
Token Gingrich
Token ID29855
Feature activation-2.30e-01

Pos logit contributions
umn+4.064
cient+3.802
arch+3.762
anwhile+3.622
isode+3.610
Neg logit contributions
IELD-3.561
hips-3.507
ymm-3.234
 Desire-3.139
WARE-2.888
Token Former
Token ID14466
Feature activation+2.94e+00

Pos logit contributions
IELD+0.399
hips+0.374
ymm+0.355
 Desire+0.352
 suppose+0.326
Neg logit contributions
umn-0.484
cient-0.461
anwhile-0.439
arch-0.436
isode-0.435
Token Speaker
Token ID14931
Feature activation-1.48e+00

Pos logit contributions
umn+6.388
cient+5.836
arch+5.784
anwhile+5.595
isode+5.554
Neg logit contributions
IELD-4.412
hips-4.135
ymm-4.079
 Desire-3.672
GV-3.523
Token of
Token ID286
Feature activation+1.19e-01

Pos logit contributions
IELD+2.555
hips+2.453
 Desire+2.329
ymm+2.297
 suppose+2.155
Neg logit contributions
umn-3.087
cient-2.953
anwhile-2.842
isode-2.793
arch-2.766
Token the
Token ID262
Feature activation+4.61e-01

Pos logit contributions
umn+0.270
cient+0.256
arch+0.247
isode+0.244
anwhile+0.244
Neg logit contributions
IELD-0.178
hips-0.172
ymm-0.164
 Desire-0.159
 suppose-0.149
Token House
Token ID2097
Feature activation-2.26e+00

Pos logit contributions
umn+1.063
cient+1.009
arch+0.977
anwhile+0.951
isode+0.949
Neg logit contributions
IELD-0.700
hips-0.659
ymm-0.634
 Desire-0.607
 suppose-0.562
Token Newt
Token ID40893
Feature activation+2.16e+00

Pos logit contributions
IELD+2.912
hips+2.748
 Desire+2.564
ymm+2.516
 suppose+2.299
Neg logit contributions
umn-5.687
cient-5.480
anwhile-5.305
isode-5.240
arch-5.201
Token
Token ID247
Feature activation-6.69e-01

Pos logit contributions
umn+0.265
cient+0.252
arch+0.234
anwhile+0.232
isode+0.229
Neg logit contributions
IELD-0.266
hips-0.251
 Desire-0.243
ymm-0.239
 suppose-0.226
Tokens
Token ID82
Feature activation-1.10e+00

Pos logit contributions
IELD+1.339
hips+1.265
 Desire+1.227
ymm+1.212
 suppose+1.144
Neg logit contributions
umn-1.231
cient-1.170
anwhile-1.106
arch-1.088
isode-1.083
Token obviously
Token ID6189
Feature activation-1.43e+00

Pos logit contributions
IELD+1.832
hips+1.729
 Desire+1.650
ymm+1.626
 suppose+1.488
Neg logit contributions
umn-2.403
cient-2.301
anwhile-2.201
arch-2.180
isode-2.160
Token not
Token ID407
Feature activation+2.65e-01

Pos logit contributions
IELD+2.320
hips+2.202
 Desire+2.089
ymm+2.057
 suppose+1.897
Neg logit contributions
umn-3.163
cient-3.032
anwhile-2.905
arch-2.868
isode-2.859
Token impossible
Token ID5340
Feature activation-8.49e-01

Pos logit contributions
umn+0.607
cient+0.586
arch+0.559
anwhile+0.553
ophone+0.544
Neg logit contributions
IELD-0.404
hips-0.376
 Desire-0.357
ymm-0.350
WARE-0.309
Token that
Token ID326
Feature activation+5.80e-02

Pos logit contributions
IELD+1.427
hips+1.352
 Desire+1.288
ymm+1.273
 suppose+1.169
Neg logit contributions
umn-1.824
cient-1.738
anwhile-1.675
arch-1.644
isode-1.641
Token this
Token ID428
Feature activation+7.67e-01

Pos logit contributions
umn+0.113
cient+0.107
anwhile+0.103
arch+0.101
isode+0.100
Neg logit contributions
IELD-0.108
hips-0.102
ymm-0.096
 Desire-0.096
 suppose-0.089
Token could
Token ID714
Feature activation-3.19e-01

Pos logit contributions
umn+1.401
cient+1.342
anwhile+1.264
arch+1.258
isode+1.235
Neg logit contributions
IELD-1.508
hips-1.428
 Desire-1.354
ymm-1.333
ELF-1.243
Token happen
Token ID1645
Feature activation-4.40e-01

Pos logit contributions
IELD+0.630
hips+0.600
 Desire+0.579
ymm+0.569
 suppose+0.533
Neg logit contributions
umn-0.593
cient-0.560
anwhile-0.534
isode-0.526
arch-0.524
Token again
Token ID757
Feature activation+4.20e-01

Pos logit contributions
IELD+0.757
hips+0.701
 Desire+0.684
ymm+0.663
 suppose+0.616
Neg logit contributions
umn-0.932
cient-0.892
anwhile-0.858
arch-0.845
isode-0.839
Token.
Token ID13
Feature activation-2.27e-01

Pos logit contributions
umn+0.868
cient+0.831
anwhile+0.791
arch+0.790
isode+0.776
Neg logit contributions
IELD-0.742
hips-0.678
 Desire-0.668
ymm-0.651
 suppose-0.593
Token talk
Token ID1561
Feature activation+9.64e-01

Pos logit contributions
IELD+0.325
hips+0.307
 Desire+0.292
ymm+0.291
 suppose+0.268
Neg logit contributions
umn-0.371
cient-0.353
anwhile-0.337
isode-0.334
arch-0.333
Token about
Token ID546
Feature activation+3.83e-01

Pos logit contributions
umn+1.833
cient+1.760
isode+1.651
arch+1.648
anwhile+1.644
Neg logit contributions
IELD-1.847
hips-1.694
ymm-1.647
 Desire-1.636
 suppose-1.527
Token Supreme
Token ID5617
Feature activation+5.33e-01

Pos logit contributions
umn+0.809
cient+0.773
arch+0.736
anwhile+0.729
isode+0.725
Neg logit contributions
IELD-0.656
hips-0.608
ymm-0.576
 Desire-0.559
 suppose-0.529
Token to
Token ID284
Feature activation+4.93e-02

Pos logit contributions
umn+1.048
cient+1.012
arch+0.968
ophone+0.953
anwhile+0.944
Neg logit contributions
IELD-0.940
hips-0.875
ymm-0.850
 Desire-0.844
 suppose-0.809
Token someone
Token ID2130
Feature activation+4.70e-01

Pos logit contributions
umn+0.107
cient+0.102
anwhile+0.097
isode+0.097
arch+0.097
Neg logit contributions
IELD-0.081
hips-0.076
ymm-0.072
 Desire-0.071
 suppose-0.065
Token who
Token ID508
Feature activation+1.72e+00

Pos logit contributions
umn+0.917
cient+0.898
arch+0.827
ophone+0.822
isode+0.818
Neg logit contributions
IELD-0.882
hips-0.802
ymm-0.787
 Desire-0.760
WARE-0.699
Token isn
Token ID2125
Feature activation+5.83e-01

Pos logit contributions
umn+3.177
cient+3.023
arch+2.860
anwhile+2.845
ophone+2.773
Neg logit contributions
IELD-3.368
hips-3.072
ymm-3.008
 Desire-2.935
WARE-2.773
Token
Token ID447
Feature activation+3.63e-01

Pos logit contributions
umn+1.066
cient+1.019
arch+0.949
isode+0.934
ophone+0.923
Neg logit contributions
IELD-1.157
hips-1.093
 Desire-1.060
ymm-1.052
 suppose-0.991
Token
Token ID247
Feature activation-5.02e-01

Pos logit contributions
umn+0.692
cient+0.661
arch+0.606
anwhile+0.598
isode+0.593
Neg logit contributions
IELD-0.704
hips-0.662
 Desire-0.645
ymm-0.636
 suppose-0.598
Tokent
Token ID83
Feature activation-5.17e-01

Pos logit contributions
IELD+1.068
hips+1.008
 Desire+0.986
ymm+0.972
 suppose+0.920
Neg logit contributions
umn-0.862
cient-0.819
anwhile-0.762
arch-0.752
isode-0.745
Token interested
Token ID4609
Feature activation-9.83e-01

Pos logit contributions
IELD+0.838
hips+0.779
ymm+0.738
 Desire+0.731
 suppose+0.658
Neg logit contributions
umn-1.155
cient-1.117
anwhile-1.051
arch-1.051
isode-1.042
Token and
Token ID290
Feature activation+1.07e+00

Pos logit contributions
umn+0.340
cient+0.323
anwhile+0.313
arch+0.305
isode+0.301
Neg logit contributions
IELD-0.279
hips-0.261
ymm-0.249
 Desire-0.246
 suppose-0.227
Token the
Token ID262
Feature activation+2.42e-01

Pos logit contributions
umn+2.171
cient+2.072
anwhile+2.015
arch+1.946
isode+1.941
Neg logit contributions
IELD-1.893
hips-1.797
ymm-1.697
 Desire-1.660
 suppose-1.525
Token DOJ
Token ID27055
Feature activation+4.54e-01

Pos logit contributions
umn+0.461
cient+0.435
anwhile+0.419
arch+0.410
isode+0.405
Neg logit contributions
IELD-0.462
hips-0.444
ymm-0.422
 Desire-0.418
 suppose-0.393
Token claimed
Token ID4752
Feature activation-1.84e-01

Pos logit contributions
umn+0.858
cient+0.813
anwhile+0.782
arch+0.772
isode+0.753
Neg logit contributions
IELD-0.864
hips-0.826
ymm-0.792
 Desire-0.782
 suppose-0.721
Token,
Token ID11
Feature activation-6.30e-01

Pos logit contributions
IELD+0.331
hips+0.314
ymm+0.297
 Desire+0.295
 suppose+0.276
Neg logit contributions
umn-0.372
cient-0.354
anwhile-0.338
isode-0.332
arch-0.331
Token flat
Token ID6228
Feature activation+1.75e-01

Pos logit contributions
IELD+1.083
hips+1.023
 Desire+0.971
ymm+0.962
 suppose+0.892
Neg logit contributions
umn-1.331
cient-1.273
anwhile-1.221
arch-1.198
isode-1.196
Token out
Token ID503
Feature activation-1.73e+00

Pos logit contributions
umn+0.266
cient+0.252
iously+0.235
isode+0.233
arch+0.232
Neg logit contributions
IELD-0.401
hips-0.390
ymm-0.374
 Desire-0.371
 suppose-0.352
Token,
Token ID11
Feature activation-1.28e+00

Pos logit contributions
IELD+2.876
hips+2.765
 Desire+2.612
ymm+2.578
 suppose+2.420
Neg logit contributions
umn-3.696
cient-3.536
anwhile-3.416
isode-3.363
arch-3.326
Token that
Token ID326
Feature activation+1.15e+00

Pos logit contributions
IELD+2.155
hips+2.055
 Desire+1.949
ymm+1.928
 suppose+1.800
Neg logit contributions
umn-2.745
cient-2.623
anwhile-2.521
isode-2.486
arch-2.468
Token Kam
Token ID12670
Feature activation-1.26e+00

Pos logit contributions
umn+2.257
cient+2.123
anwhile+2.028
iously+2.012
arch+1.992
Neg logit contributions
IELD-2.120
hips-1.998
ymm-1.875
 Desire-1.823
GV-1.705
Tokenal
Token ID282
Feature activation+9.93e-01

Pos logit contributions
IELD+2.591
hips+2.542
ymm+2.427
 Desire+2.424
 suppose+2.289
Neg logit contributions
umn-2.107
cient-2.007
anwhile-1.988
isode-1.919
ophone-1.840
Token a
Token ID257
Feature activation-1.10e+00

Pos logit contributions
IELD+0.670
hips+0.635
ymm+0.596
 Desire+0.580
 suppose+0.553
Neg logit contributions
umn-0.885
cient-0.850
anwhile-0.807
arch-0.806
isode-0.805
Token leg
Token ID1232
Feature activation+8.40e-01

Pos logit contributions
IELD+1.926
hips+1.843
ymm+1.741
 Desire+1.726
 suppose+1.614
Neg logit contributions
umn-2.274
cient-2.164
anwhile-2.063
isode-2.057
arch-2.042
Token amput
Token ID40338
Feature activation-5.70e-01

Pos logit contributions
umn+1.519
cient+1.427
isode+1.365
arch+1.355
ophone+1.334
Neg logit contributions
IELD-1.682
hips-1.628
ymm-1.524
 Desire-1.520
 suppose-1.438
Tokenation
Token ID341
Feature activation+3.35e-01

Pos logit contributions
IELD+1.471
hips+1.415
 Desire+1.373
ymm+1.357
 suppose+1.307
Neg logit contributions
umn-0.722
cient-0.663
anwhile-0.602
isode-0.601
arch-0.598
Token who
Token ID508
Feature activation+1.68e+00

Pos logit contributions
umn+0.679
cient+0.653
arch+0.616
anwhile+0.615
isode+0.614
Neg logit contributions
IELD-0.601
hips-0.558
ymm-0.536
 Desire-0.526
 suppose-0.490
Token was
Token ID373
Feature activation-9.85e-02

Pos logit contributions
umn+3.060
cient+2.902
anwhile+2.760
arch+2.753
isode+2.654
Neg logit contributions
IELD-3.280
hips-3.084
ymm-2.995
 Desire-2.879
 suppose-2.699
Token volunteering
Token ID41434
Feature activation-8.72e-01

Pos logit contributions
IELD+0.177
hips+0.165
ymm+0.158
 Desire+0.157
 suppose+0.141
Neg logit contributions
umn-0.201
cient-0.193
anwhile-0.182
arch-0.182
isode-0.178
Token at
Token ID379
Feature activation-1.22e-01

Pos logit contributions
IELD+1.370
hips+1.289
ymm+1.213
 Desire+1.212
 suppose+1.118
Neg logit contributions
umn-1.973
cient-1.899
anwhile-1.811
arch-1.794
isode-1.784
Token the
Token ID262
Feature activation+8.08e-01

Pos logit contributions
IELD+0.170
hips+0.161
ymm+0.149
 Desire+0.144
 suppose+0.137
Neg logit contributions
umn-0.297
cient-0.285
anwhile-0.274
isode-0.272
arch-0.270
Token store
Token ID3650
Feature activation+1.27e-01

Pos logit contributions
umn+1.436
cient+1.363
anwhile+1.290
arch+1.272
isode+1.255
Neg logit contributions
IELD-1.637
hips-1.578
ymm-1.480
 Desire-1.447
 suppose-1.411
Token at
Token ID379
Feature activation+1.08e-01

Pos logit contributions
umn+0.270
cient+0.259
anwhile+0.246
arch+0.245
isode+0.243
Neg logit contributions
IELD-0.218
hips-0.201
ymm-0.193
 Desire-0.192
 suppose-0.177
Token 1
Token ID352
Feature activation-1.43e+00

Pos logit contributions
umn+2.443
cient+2.333
arch+2.238
isode+2.224
anwhile+2.222
Neg logit contributions
IELD-1.777
hips-1.640
ymm-1.621
 Desire-1.549
 suppose-1.430
Token and
Token ID290
Feature activation+8.38e-01

Pos logit contributions
IELD+3.077
hips+2.971
 Desire+2.848
ymm+2.832
 suppose+2.686
Neg logit contributions
umn-2.402
cient-2.266
anwhile-2.143
isode-2.117
arch-2.098
Token 2
Token ID362
Feature activation-1.85e+00

Pos logit contributions
umn+1.860
cient+1.734
arch+1.667
isode+1.656
anwhile+1.642
Neg logit contributions
IELD-1.376
hips-1.289
ymm-1.224
 Desire-1.187
 suppose-1.108
Token were
Token ID547
Feature activation-1.36e-01

Pos logit contributions
IELD+3.202
hips+3.053
 Desire+2.898
ymm+2.877
 suppose+2.685
Neg logit contributions
umn-3.878
cient-3.700
anwhile-3.551
isode-3.507
arch-3.479
Token listed
Token ID5610
Feature activation-5.25e-01

Pos logit contributions
IELD+0.270
hips+0.252
 Desire+0.243
ymm+0.240
 suppose+0.221
Neg logit contributions
umn-0.254
cient-0.242
anwhile-0.227
arch-0.227
isode-0.224
Token on
Token ID319
Feature activation+3.55e-01

Pos logit contributions
IELD+0.878
hips+0.798
 Desire+0.769
ymm+0.768
 suppose+0.704
Neg logit contributions
umn-1.159
cient-1.109
arch-1.060
anwhile-1.059
isode-1.031
Token Y
Token ID575
Feature activation-1.19e+00

Pos logit contributions
umn+0.800
cient+0.759
arch+0.733
anwhile+0.728
isode+0.725
Neg logit contributions
IELD-0.558
hips-0.527
ymm-0.508
 Desire-0.489
 suppose-0.464
TokenAL
Token ID1847
Feature activation+5.69e-01

Pos logit contributions
IELD+2.320
hips+2.250
 Desire+2.175
ymm+2.133
 suppose+2.036
Neg logit contributions
umn-2.236
cient-2.111
anwhile-1.992
isode-1.992
arch-1.991
TokenSA
Token ID4090
Feature activation-5.30e-01

Pos logit contributions
umn+1.139
cient+1.083
anwhile+1.039
arch+1.036
isode+1.035
Neg logit contributions
IELD-0.988
hips-0.967
 Desire-0.934
ymm-0.920
 suppose-0.845
Token's
Token ID338
Feature activation-1.16e+00

Pos logit contributions
IELD+0.961
hips+0.922
ymm+0.879
 Desire+0.875
 suppose+0.822
Neg logit contributions
umn-1.062
cient-1.007
anwhile-0.963
arch-0.962
isode-0.944
Token "
Token ID366
Feature activation-2.63e-01

Pos logit contributions
IELD+2.337
hips+2.243
 Desire+2.143
ymm+2.139
 suppose+2.020
Neg logit contributions
umn-2.086
cient-1.971
anwhile-1.875
isode-1.841
arch-1.839
Token.
Token ID13
Feature activation+3.08e-02

Pos logit contributions
umn+0.185
cient+0.177
arch+0.167
isode+0.167
anwhile+0.166
Neg logit contributions
IELD-0.156
hips-0.143
 Desire-0.137
ymm-0.137
 suppose-0.126
Token A
Token ID317
Feature activation-2.67e+00

Pos logit contributions
umn+0.071
cient+0.068
anwhile+0.065
arch+0.065
isode+0.065
Neg logit contributions
IELD-0.046
hips-0.043
ymm-0.041
 Desire-0.040
 suppose-0.037
Token great
Token ID1049
Feature activation-5.00e-01

Pos logit contributions
IELD+4.369
hips+4.175
 Desire+3.967
ymm+3.898
 suppose+3.660
Neg logit contributions
umn-5.772
cient-5.511
anwhile-5.327
isode-5.232
arch-5.182
Token example
Token ID1672
Feature activation-7.65e-01

Pos logit contributions
IELD+0.947
hips+0.904
 Desire+0.861
ymm+0.860
 suppose+0.801
Neg logit contributions
umn-0.964
cient-0.916
anwhile-0.870
isode-0.864
arch-0.860
Token of
Token ID286
Feature activation-1.26e+00

Pos logit contributions
IELD+1.417
hips+1.338
 Desire+1.278
ymm+1.271
 suppose+1.170
Neg logit contributions
umn-1.522
cient-1.460
anwhile-1.382
isode-1.371
arch-1.369
Token this
Token ID428
Feature activation-2.88e-01

Pos logit contributions
IELD+1.792
hips+1.693
 Desire+1.578
ymm+1.572
 suppose+1.428
Neg logit contributions
umn-3.019
cient-2.892
anwhile-2.783
isode-2.757
arch-2.750
Token is
Token ID318
Feature activation-3.41e-01

Pos logit contributions
IELD+0.532
hips+0.502
ymm+0.476
 Desire+0.475
 suppose+0.440
Neg logit contributions
umn-0.570
cient-0.549
anwhile-0.517
isode-0.516
arch-0.514
Token Robot
Token ID16071
Feature activation+1.13e-01

Pos logit contributions
IELD+0.527
hips+0.491
ymm+0.467
 Desire+0.457
 suppose+0.416
Neg logit contributions
umn-0.777
cient-0.741
isode-0.711
anwhile-0.710
arch-0.709
Token Entertainment
Token ID11058
Feature activation-1.63e-01

Pos logit contributions
umn+0.246
cient+0.235
arch+0.226
isode+0.222
anwhile+0.218
Neg logit contributions
IELD-0.187
hips-0.176
ymm-0.165
 Desire-0.162
 suppose-0.156
Token:
Token ID25
Feature activation+6.19e-03

Pos logit contributions
IELD+0.306
hips+0.288
ymm+0.278
 Desire+0.273
 suppose+0.257
Neg logit contributions
umn-0.316
cient-0.297
arch-0.282
anwhile-0.280
isode-0.277
Token in
Token ID287
Feature activation+5.07e-02

Pos logit contributions
umn+0.014
cient+0.013
anwhile+0.013
isode+0.013
arch+0.012
Neg logit contributions
IELD-0.010
hips-0.009
ymm-0.009
 Desire-0.009
 suppose-0.008
Token silence
Token ID9550
Feature activation-2.51e-01

Pos logit contributions
IELD+0.788
hips+0.737
ymm+0.690
 Desire+0.684
 suppose+0.640
Neg logit contributions
umn-0.934
cient-0.892
anwhile-0.841
arch-0.839
isode-0.838
Token.
Token ID13
Feature activation-3.12e-01

Pos logit contributions
IELD+0.450
hips+0.414
 Desire+0.401
ymm+0.399
 suppose+0.365
Neg logit contributions
umn-0.514
cient-0.496
anwhile-0.466
isode-0.464
arch-0.464
Token It
Token ID632
Feature activation-9.18e-02

Pos logit contributions
IELD+0.470
hips+0.443
ymm+0.415
 Desire+0.403
 suppose+0.382
Neg logit contributions
umn-0.727
cient-0.693
isode-0.659
anwhile-0.659
arch-0.658
Token
Token ID447
Feature activation-3.60e-01

Pos logit contributions
IELD+0.175
hips+0.165
 Desire+0.159
ymm+0.157
 suppose+0.146
Neg logit contributions
umn-0.178
cient-0.169
anwhile-0.161
arch-0.159
isode-0.158
Token
Token ID247
Feature activation-7.71e-01

Pos logit contributions
IELD+0.580
hips+0.550
 Desire+0.526
ymm+0.518
 suppose+0.482
Neg logit contributions
umn-0.797
cient-0.763
anwhile-0.723
arch-0.719
isode-0.717
Tokens
Token ID82
Feature activation-8.45e-01

Pos logit contributions
IELD+1.449
hips+1.377
 Desire+1.321
ymm+1.308
 suppose+1.231
Neg logit contributions
umn-1.509
cient-1.435
anwhile-1.370
isode-1.348
arch-1.341
Token enough
Token ID1576
Feature activation-2.51e-02

Pos logit contributions
IELD+1.357
hips+1.263
 Desire+1.192
ymm+1.180
 suppose+1.077
Neg logit contributions
umn-1.902
cient-1.826
anwhile-1.737
arch-1.725
isode-1.721
Token to
Token ID284
Feature activation-1.13e+00

Pos logit contributions
IELD+0.047
hips+0.044
 Desire+0.042
ymm+0.041
 suppose+0.038
Neg logit contributions
umn-0.050
cient-0.048
arch-0.045
isode-0.044
anwhile-0.044
Token trick
Token ID6908
Feature activation+8.30e-01

Pos logit contributions
IELD+2.099
hips+2.008
 Desire+1.917
ymm+1.900
 suppose+1.779
Neg logit contributions
umn-2.214
cient-2.109
anwhile-2.015
isode-1.988
arch-1.973
Token me
Token ID502
Feature activation-6.73e-01

Pos logit contributions
umn+1.833
cient+1.752
arch+1.658
ophone+1.648
anwhile+1.632
Neg logit contributions
IELD-1.337
hips-1.266
ymm-1.178
 Desire-1.163
 suppose-1.067
Token into
Token ID656
Feature activation-9.86e-02

Pos logit contributions
IELD+1.189
hips+1.109
 Desire+1.068
ymm+1.050
 suppose+0.959
Neg logit contributions
umn-1.406
cient-1.354
arch-1.275
anwhile-1.270
isode-1.262
Token when
Token ID618
Feature activation+7.64e-01

Pos logit contributions
umn+0.245
cient+0.235
arch+0.222
anwhile+0.221
isode+0.219
Neg logit contributions
IELD-0.231
hips-0.218
 Desire-0.211
ymm-0.205
 Darling-0.189
Token the
Token ID262
Feature activation-2.03e-01

Pos logit contributions
umn+1.665
cient+1.577
arch+1.513
anwhile+1.497
ophone+1.471
Neg logit contributions
IELD-1.250
hips-1.148
 Desire-1.118
ymm-1.078
ELF-0.978
Token initial
Token ID4238
Feature activation-2.39e+00

Pos logit contributions
IELD+0.350
hips+0.330
 Desire+0.313
ymm+0.310
 suppose+0.285
Neg logit contributions
umn-0.433
cient-0.411
anwhile-0.390
arch-0.390
isode-0.384
Token space
Token ID2272
Feature activation-3.84e-01

Pos logit contributions
IELD+3.990
hips+3.822
 Desire+3.629
ymm+3.587
 suppose+3.364
Neg logit contributions
umn-5.070
cient-4.851
anwhile-4.679
isode-4.606
arch-4.550
Token is
Token ID318
Feature activation-7.86e-01

Pos logit contributions
IELD+0.689
hips+0.644
 Desire+0.619
ymm+0.610
 suppose+0.569
Neg logit contributions
umn-0.786
cient-0.751
anwhile-0.710
arch-0.710
isode-0.708
Token nearly
Token ID3016
Feature activation-8.90e-01

Pos logit contributions
IELD+1.275
hips+1.193
 Desire+1.137
ymm+1.109
 suppose+1.007
Neg logit contributions
umn-1.753
cient-1.686
anwhile-1.590
arch-1.589
isode-1.576
Token cub
Token ID13617
Feature activation-3.36e-01

Pos logit contributions
IELD+1.423
hips+1.341
 Desire+1.278
ymm+1.253
 suppose+1.145
Neg logit contributions
umn-1.987
cient-1.919
isode-1.814
anwhile-1.810
arch-1.805
Tokenoid
Token ID1868
Feature activation+3.18e-01

Pos logit contributions
IELD+0.833
hips+0.799
 Desire+0.776
ymm+0.762
 suppose+0.731
Neg logit contributions
umn-0.468
cient-0.432
isode-0.395
anwhile-0.393
arch-0.391
Token,
Token ID11
Feature activation-5.58e-01

Pos logit contributions
umn+0.640
cient+0.618
arch+0.577
ophone+0.568
isode+0.566
Neg logit contributions
IELD-0.584
hips-0.537
 Desire-0.522
ymm-0.510
 suppose-0.468
Token for
Token ID329
Feature activation-9.43e-02

Pos logit contributions
IELD+0.962
hips+0.913
 Desire+0.877
ymm+0.855
 suppose+0.780
Neg logit contributions
umn-1.179
cient-1.120
anwhile-1.063
arch-1.058
isode-1.045
Token example
Token ID1672
Feature activation-6.66e-01

Pos logit contributions
IELD+0.172
hips+0.164
 Desire+0.156
ymm+0.153
 suppose+0.141
Neg logit contributions
umn-0.189
cient-0.180
arch-0.170
anwhile-0.169
isode-0.165
Token.
Token ID13
Feature activation-2.91e-01

Pos logit contributions
umn+4.300
cient+4.180
arch+3.958
anwhile+3.862
ophone+3.782
Neg logit contributions
IELD-4.252
hips-3.811
 Desire-3.800
ymm-3.771
ELF-3.482
Token
Token ID198
Feature activation-1.58e-01

Pos logit contributions
IELD+0.447
hips+0.422
 Desire+0.394
ymm+0.392
 suppose+0.371
Neg logit contributions
umn-0.663
cient-0.635
anwhile-0.612
arch-0.606
isode-0.602
Token
Token ID198
Feature activation-1.26e-01

Pos logit contributions
IELD+0.235
hips+0.221
 Desire+0.209
ymm+0.207
 suppose+0.190
Neg logit contributions
umn-0.370
cient-0.355
arch-0.337
anwhile-0.337
isode-0.336
TokenS
Token ID50
Feature activation-1.90e+00

Pos logit contributions
IELD+0.229
hips+0.219
 Desire+0.213
ymm+0.207
 suppose+0.199
Neg logit contributions
umn-0.249
cient-0.239
arch-0.226
isode-0.224
anwhile-0.219
TokenBI
Token ID3483
Feature activation+1.85e+00

Pos logit contributions
IELD+2.134
hips+1.993
 Desire+1.832
ymm+1.802
 suppose+1.613
Neg logit contributions
umn-5.095
cient-4.917
anwhile-4.772
isode-4.720
arch-4.686
Token Wire
Token ID14712
Feature activation-5.56e-01

Pos logit contributions
umn+3.787
cient+3.602
anwhile+3.489
arch+3.434
ophone+3.370
Neg logit contributions
IELD-3.077
hips-2.959
 Desire-2.761
ymm-2.744
 suppose-2.595
Tokenx
Token ID87
Feature activation+6.54e-01

Pos logit contributions
IELD+1.157
hips+1.133
 Desire+1.078
ymm+1.076
 suppose+1.017
Neg logit contributions
umn-0.944
cient-0.896
anwhile-0.882
isode-0.847
arch-0.822
Token Asia
Token ID7229
Feature activation-7.54e-02

Pos logit contributions
umn+1.116
cient+1.049
anwhile+0.993
arch+0.991
ophone+0.960
Neg logit contributions
IELD-1.376
hips-1.299
 Desire-1.254
ymm-1.252
 suppose-1.177
Token will
Token ID481
Feature activation+6.64e-01

Pos logit contributions
IELD+0.141
hips+0.133
 Desire+0.127
ymm+0.127
 suppose+0.118
Neg logit contributions
umn-0.149
cient-0.140
anwhile-0.133
arch-0.132
isode-0.131
Token deliver
Token ID5203
Feature activation-1.02e+00

Pos logit contributions
umn+1.269
cient+1.200
anwhile+1.146
arch+1.132
isode+1.116
Neg logit contributions
IELD-1.278
hips-1.182
 Desire-1.148
ymm-1.141
 Darling-1.049
Token a
Token ID257
Feature activation-3.32e-01

Pos logit contributions
IELD+1.715
hips+1.610
 Desire+1.534
ymm+1.525
 suppose+1.419
Neg logit contributions
umn-2.177
cient-2.093
anwhile-2.000
arch-1.979
isode-1.962
Token technical
Token ID6276
Feature activation+3.63e-01

Pos logit contributions
IELD+2.492
hips+2.393
 Desire+2.267
ymm+2.255
 suppose+2.103
Neg logit contributions
umn-2.618
cient-2.507
anwhile-2.382
isode-2.348
arch-2.344
Token fault
Token ID8046
Feature activation-6.34e-01

Pos logit contributions
umn+0.648
cient+0.621
arch+0.577
isode+0.575
ophone+0.574
Neg logit contributions
IELD-0.733
hips-0.703
ymm-0.662
 Desire-0.659
 Darling-0.615
Token that
Token ID326
Feature activation+8.30e-01

Pos logit contributions
IELD+1.092
hips+1.024
 Desire+0.977
ymm+0.971
 suppose+0.894
Neg logit contributions
umn-1.328
cient-1.290
anwhile-1.219
isode-1.210
arch-1.209
Token can
Token ID460
Feature activation-8.56e-02

Pos logit contributions
umn+1.584
cient+1.517
anwhile+1.429
arch+1.423
isode+1.395
Neg logit contributions
IELD-1.553
hips-1.471
 Desire-1.411
ymm-1.395
 suppose-1.280
Token be
Token ID307
Feature activation-1.54e+00

Pos logit contributions
IELD+0.161
hips+0.152
 Desire+0.148
ymm+0.146
 suppose+0.134
Neg logit contributions
umn-0.166
cient-0.158
anwhile-0.151
isode-0.149
arch-0.147
Token corrected
Token ID19267
Feature activation-1.31e+00

Pos logit contributions
IELD+2.582
hips+2.460
 Desire+2.340
ymm+2.310
 suppose+2.152
Neg logit contributions
umn-3.289
cient-3.149
anwhile-3.021
isode-2.979
arch-2.962
Token.
Token ID13
Feature activation-4.54e-01

Pos logit contributions
IELD+2.154
hips+2.026
 Desire+1.934
ymm+1.912
 suppose+1.767
Neg logit contributions
umn-2.864
cient-2.754
anwhile-2.639
arch-2.596
isode-2.594
Token
Token ID198
Feature activation-1.63e-02

Pos logit contributions
IELD+0.728
hips+0.688
ymm+0.649
 Desire+0.642
 suppose+0.605
Neg logit contributions
umn-1.000
cient-0.964
anwhile-0.915
arch-0.911
isode-0.900
Token
Token ID198
Feature activation+1.21e-01

Pos logit contributions
IELD+0.024
hips+0.023
 Desire+0.022
ymm+0.021
 suppose+0.020
Neg logit contributions
umn-0.038
cient-0.037
arch-0.035
anwhile-0.034
isode-0.034
Token"
Token ID1
Feature activation-4.86e-01

Pos logit contributions
umn+0.223
cient+0.213
arch+0.200
isode+0.196
ophone+0.194
Neg logit contributions
IELD-0.238
hips-0.228
 Desire-0.222
ymm-0.217
 suppose-0.208
TokenThe
Token ID464
Feature activation-4.74e-01

Pos logit contributions
IELD+0.771
hips+0.730
 Desire+0.705
ymm+0.695
 suppose+0.643
Neg logit contributions
umn-1.070
cient-1.033
arch-0.984
isode-0.970
anwhile-0.958
Token remove
Token ID4781
Feature activation-1.56e+00

Pos logit contributions
umn+1.596
cient+1.518
anwhile+1.449
arch+1.432
isode+1.401
Neg logit contributions
IELD-1.686
hips-1.575
 Desire-1.520
ymm-1.510
 Darling-1.392
Token.
Token ID13
Feature activation-4.13e-01

Pos logit contributions
IELD+2.456
hips+2.321
 Desire+2.199
ymm+2.178
 suppose+2.025
Neg logit contributions
umn-3.510
cient-3.368
anwhile-3.239
isode-3.196
arch-3.193
TokenMS
Token ID5653
Feature activation+1.97e+00

Pos logit contributions
IELD+0.638
hips+0.602
 Desire+0.563
ymm+0.559
 suppose+0.528
Neg logit contributions
umn-0.941
cient-0.904
anwhile-0.866
arch-0.860
isode-0.856
TokenI
Token ID40
Feature activation+2.57e-01

Pos logit contributions
umn+3.462
cient+3.353
arch+3.225
iously+3.209
ophone+3.167
Neg logit contributions
IELD-3.753
hips-3.503
 Desire-3.441
ymm-3.406
 Darling-3.187
Token exclusive
Token ID8568
Feature activation-1.80e+00

Pos logit contributions
umn+0.553
cient+0.529
anwhile+0.514
arch+0.506
isode+0.497
Neg logit contributions
IELD-0.427
hips-0.406
ymm-0.378
 Desire-0.378
 suppose-0.351
Token Un
Token ID791
Feature activation-3.25e+00

Pos logit contributions
IELD+2.963
hips+2.824
 Desire+2.675
ymm+2.645
 suppose+2.468
Neg logit contributions
umn-3.909
cient-3.741
anwhile-3.598
isode-3.549
arch-3.524
Tokenlocked
Token ID24162
Feature activation-1.84e+00

Pos logit contributions
IELD+3.680
hips+3.457
ymm+3.179
 Desire+3.163
 suppose+2.762
Neg logit contributions
umn-8.490
cient-8.195
anwhile-8.099
isode-7.840
ophone-7.707
Token Digital
Token ID10231
Feature activation-3.57e-01

Pos logit contributions
IELD+3.303
hips+3.162
 Desire+3.008
ymm+2.979
 suppose+2.801
Neg logit contributions
umn-3.736
cient-3.566
anwhile-3.417
isode-3.369
arch-3.346
TokenPower
Token ID13434
Feature activation-9.64e-01

Pos logit contributions
IELD+0.446
hips+0.418
 Desire+0.389
ymm+0.385
 suppose+0.352
Neg logit contributions
umn-0.914
cient-0.883
anwhile-0.851
arch-0.844
isode-0.844
Token is
Token ID318
Feature activation+6.31e-01

Pos logit contributions
IELD+1.611
hips+1.524
 Desire+1.449
ymm+1.433
 suppose+1.339
Neg logit contributions
umn-2.076
cient-1.996
anwhile-1.912
arch-1.889
isode-1.883
Token composed
Token ID13160
Feature activation+1.25e+00

Pos logit contributions
umn+1.391
cient+1.327
arch+1.274
anwhile+1.269
isode+1.250
Neg logit contributions
IELD-1.021
hips-0.943
 Desire-0.899
ymm-0.883
 suppose-0.797
Token click
Token ID3904
Feature activation+3.63e-01

Pos logit contributions
umn+1.588
cient+1.494
anwhile+1.438
arch+1.409
isode+1.402
Neg logit contributions
IELD-1.523
hips-1.421
ymm-1.376
 Desire-1.360
GV-1.255
Token on
Token ID319
Feature activation-9.28e-02

Pos logit contributions
umn+0.786
cient+0.748
anwhile+0.727
arch+0.717
isode+0.712
Neg logit contributions
IELD-0.594
hips-0.550
ymm-0.534
 Desire-0.530
 suppose-0.483
Token Vide
Token ID42551
Feature activation+1.27e+00

Pos logit contributions
IELD+0.145
hips+0.135
ymm+0.130
 Desire+0.128
 suppose+0.117
Neg logit contributions
umn-0.209
cient-0.199
anwhile-0.194
arch-0.191
isode-0.191
Tokenopl
Token ID20106
Feature activation+2.32e+00

Pos logit contributions
umn+1.669
cient+1.511
ophone+1.448
arch+1.443
isode+1.371
Neg logit contributions
IELD-3.131
hips-3.034
 Desire-2.970
ymm-2.944
 suppose-2.815
Tokenayer
Token ID2794
Feature activation+1.08e+00

Pos logit contributions
umn+4.039
cient+3.676
isode+3.543
arch+3.522
ophone+3.452
Neg logit contributions
IELD-4.914
hips-4.506
 Desire-4.394
ymm-4.304
 rounded-4.168
Token Input
Token ID23412
Feature activation-3.04e+00

Pos logit contributions
umn+2.255
cient+2.137
arch+2.041
anwhile+2.022
isode+2.010
Neg logit contributions
IELD-1.815
hips-1.707
ymm-1.654
 Desire-1.612
 suppose-1.507
TokenStream
Token ID12124
Feature activation-1.38e+00

Pos logit contributions
IELD+4.990
hips+4.763
 Desire+4.529
ymm+4.452
 suppose+4.164
Neg logit contributions
umn-6.548
cient-6.294
anwhile-6.066
isode-5.966
arch-5.895
Token.
Token ID13
Feature activation-7.31e-01

Pos logit contributions
IELD+2.474
hips+2.364
 Desire+2.258
ymm+2.243
 suppose+2.100
Neg logit contributions
umn-2.800
cient-2.658
anwhile-2.549
arch-2.516
isode-2.516
Token Make
Token ID6889
Feature activation-3.43e-01

Pos logit contributions
IELD+1.122
hips+1.053
 Desire+0.992
ymm+0.989
 suppose+0.918
Neg logit contributions
umn-1.686
cient-1.611
anwhile-1.544
isode-1.531
arch-1.524
Token sure
Token ID1654
Feature activation-2.38e-01

Pos logit contributions
IELD+0.312
hips+0.288
ymm+0.254
 Desire+0.250
 suppose+0.207
Neg logit contributions
umn-1.000
cient-0.965
anwhile-0.936
isode-0.929
arch-0.926
Token that
Token ID326
Feature activation-1.31e-01

Pos logit contributions
IELD+0.408
hips+0.388
ymm+0.370
 Desire+0.365
 suppose+0.339
Neg logit contributions
umn-0.503
cient-0.479
anwhile-0.460
isode-0.455
arch-0.452
Token competed
Token ID32440
Feature activation+6.07e-01

Pos logit contributions
IELD+1.341
hips+1.275
 Desire+1.197
ymm+1.197
 suppose+1.117
Neg logit contributions
umn-1.522
cient-1.464
isode-1.381
anwhile-1.373
arch-1.372
Token alongside
Token ID7848
Feature activation-2.90e-01

Pos logit contributions
umn+1.257
cient+1.243
arch+1.180
iously+1.146
anwhile+1.140
Neg logit contributions
IELD-1.037
hips-0.916
 Desire-0.909
ymm-0.891
 suppose-0.820
Token Chelsea
Token ID12147
Feature activation-1.23e+00

Pos logit contributions
IELD+0.465
hips+0.440
ymm+0.420
 Desire+0.404
 suppose+0.383
Neg logit contributions
umn-0.638
cient-0.612
arch-0.588
anwhile-0.587
isode-0.574
Token and
Token ID290
Feature activation+2.05e+00

Pos logit contributions
IELD+2.058
hips+1.944
 Desire+1.851
ymm+1.839
 suppose+1.695
Neg logit contributions
umn-2.635
cient-2.521
anwhile-2.407
arch-2.403
isode-2.387
Token Tottenham
Token ID24272
Feature activation-4.00e+00

Pos logit contributions
umn+4.212
cient+4.066
arch+3.934
anwhile+3.906
ophone+3.793
Neg logit contributions
IELD-3.296
hips-3.144
ymm-3.050
 Desire-2.838
WARE-2.717
Token Hots
Token ID41239
Feature activation-4.22e+00

Pos logit contributions
IELD+5.986
hips+5.845
 Desire+5.414
ymm+5.325
 suppose+4.994
Neg logit contributions
umn-9.083
cient-8.642
anwhile-8.415
isode-8.272
arch-8.079
Tokenpur
Token ID14225
Feature activation-1.81e+00

Pos logit contributions
hips+7.312
IELD+7.113
 Desire+6.560
ymm+6.463
 suppose+6.123
Neg logit contributions
umn-8.526
anwhile-8.310
cient-8.039
isode-7.824
theless-7.654
Token for
Token ID329
Feature activation+5.97e-01

Pos logit contributions
IELD+2.975
hips+2.802
 Desire+2.669
ymm+2.643
 suppose+2.451
Neg logit contributions
umn-3.953
cient-3.798
anwhile-3.633
isode-3.597
arch-3.591
Token the
Token ID262
Feature activation+1.30e+00

Pos logit contributions
umn+1.294
cient+1.252
isode+1.174
arch+1.172
ophone+1.161
Neg logit contributions
IELD-0.961
hips-0.903
ymm-0.861
 Desire-0.812
WARE-0.776
Token Premier
Token ID9952
Feature activation+1.27e-01

Pos logit contributions
umn+2.579
cient+2.498
arch+2.325
isode+2.284
anwhile+2.248
Neg logit contributions
IELD-2.299
hips-2.190
ymm-2.093
 Desire-2.038
 suppose-1.928
Token League
Token ID4041
Feature activation-1.95e-01

Pos logit contributions
umn+0.331
cient+0.322
arch+0.308
isode+0.306
anwhile+0.304
Neg logit contributions
IELD-0.149
hips-0.138
 Desire-0.126
ymm-0.126
 suppose-0.112
Token both
Token ID1111
Feature activation+3.29e-01

Pos logit contributions
umn+1.141
cient+1.093
anwhile+1.040
arch+1.040
isode+1.030
Neg logit contributions
IELD-0.980
hips-0.921
 Desire-0.883
ymm-0.859
 suppose-0.821
Token Shirt
Token ID45180
Feature activation-1.88e+00

Pos logit contributions
umn+0.733
cient+0.705
arch+0.669
anwhile+0.665
isode+0.659
Neg logit contributions
IELD-0.515
hips-0.482
 Desire-0.459
ymm-0.453
 suppose-0.420
Token Arm
Token ID7057
Feature activation-3.77e+00

Pos logit contributions
IELD+3.492
hips+3.344
 Desire+3.188
ymm+3.159
 suppose+2.984
Neg logit contributions
umn-3.698
cient-3.515
anwhile-3.350
isode-3.319
arch-3.294
Token and
Token ID290
Feature activation-1.62e-01

Pos logit contributions
IELD+7.435
hips+7.146
 Desire+6.849
ymm+6.833
 suppose+6.400
Neg logit contributions
umn-6.733
cient-6.433
anwhile-6.181
isode-6.032
arch-5.921
Token Training
Token ID13614
Feature activation-1.70e+00

Pos logit contributions
IELD+0.238
hips+0.223
ymm+0.209
 Desire+0.207
 suppose+0.193
Neg logit contributions
umn-0.378
cient-0.362
anwhile-0.347
arch-0.344
isode-0.343
Token Wear
Token ID27539
Feature activation-3.14e+00

Pos logit contributions
IELD+2.195
hips+2.073
 Desire+1.922
ymm+1.893
 suppose+1.732
Neg logit contributions
umn-4.296
cient-4.127
anwhile-3.991
isode-3.952
arch-3.934
Token partner
Token ID5212
Feature activation-1.04e+00

Pos logit contributions
IELD+5.480
hips+5.301
 Desire+5.019
ymm+4.950
 suppose+4.659
Neg logit contributions
umn-6.349
cient-6.083
anwhile-5.847
isode-5.779
arch-5.675
Token.
Token ID13
Feature activation-4.13e-01

Pos logit contributions
IELD+1.690
hips+1.586
 Desire+1.514
ymm+1.504
 suppose+1.406
Neg logit contributions
umn-2.282
cient-2.199
anwhile-2.095
arch-2.074
isode-2.068
Token
Token ID198
Feature activation-2.37e-01

Pos logit contributions
IELD+0.647
hips+0.609
ymm+0.575
 Desire+0.568
 suppose+0.545
Neg logit contributions
umn-0.927
cient-0.891
arch-0.842
isode-0.841
anwhile-0.836
Token
Token ID198
Feature activation-2.00e-01

Pos logit contributions
IELD+0.351
hips+0.330
 Desire+0.311
ymm+0.309
 suppose+0.284
Neg logit contributions
umn-0.555
cient-0.534
anwhile-0.506
arch-0.506
isode-0.505
TokenThe
Token ID464
Feature activation-3.10e-01

Pos logit contributions
IELD+0.402
hips+0.387
 Desire+0.371
ymm+0.367
 suppose+0.348
Neg logit contributions
umn-0.362
cient-0.343
anwhile-0.326
isode-0.322
arch-0.319
Tokenas
Token ID292
Feature activation-5.56e-01

Pos logit contributions
IELD+0.164
hips+0.155
 Desire+0.152
ymm+0.149
 suppose+0.142
Neg logit contributions
umn-0.138
cient-0.131
arch-0.124
isode-0.122
anwhile-0.121
Token well
Token ID880
Feature activation-1.29e+00

Pos logit contributions
IELD+0.966
hips+0.910
 Desire+0.874
ymm+0.854
 suppose+0.803
Neg logit contributions
umn-1.172
cient-1.120
anwhile-1.064
isode-1.054
arch-1.051
Token as
Token ID355
Feature activation-1.43e+00

Pos logit contributions
IELD+1.811
hips+1.724
 Desire+1.622
ymm+1.588
 suppose+1.475
Neg logit contributions
umn-3.102
cient-2.983
anwhile-2.918
isode-2.857
arch-2.822
Token the
Token ID262
Feature activation-1.90e-01

Pos logit contributions
IELD+2.345
hips+2.227
 Desire+2.107
ymm+2.090
 suppose+1.943
Neg logit contributions
umn-3.142
cient-3.004
anwhile-2.888
isode-2.852
arch-2.830
Token previously
Token ID4271
Feature activation-4.93e-01

Pos logit contributions
IELD+0.338
hips+0.320
ymm+0.305
 Desire+0.303
 suppose+0.284
Neg logit contributions
umn-0.392
cient-0.373
anwhile-0.356
isode-0.352
arch-0.350
Token announced
Token ID3414
Feature activation-3.64e+00

Pos logit contributions
IELD+0.951
hips+0.906
 Desire+0.871
ymm+0.859
 suppose+0.806
Neg logit contributions
umn-0.938
cient-0.894
anwhile-0.845
isode-0.836
arch-0.832
Token nominees
Token ID26456
Feature activation-2.16e+00

Pos logit contributions
IELD+5.869
hips+5.767
 Desire+5.513
ymm+5.213
 suppose+4.936
Neg logit contributions
umn-7.547
cient-7.240
anwhile-7.113
isode-6.907
arch-6.870
Token)
Token ID8
Feature activation-4.24e-02

Pos logit contributions
IELD+3.263
hips+3.150
 Desire+2.944
ymm+2.892
 suppose+2.698
Neg logit contributions
umn-4.939
cient-4.735
anwhile-4.581
isode-4.508
arch-4.460
Token is
Token ID318
Feature activation+5.58e-01

Pos logit contributions
IELD+0.082
hips+0.076
 Desire+0.074
ymm+0.073
 suppose+0.068
Neg logit contributions
umn-0.081
cient-0.077
anwhile-0.074
arch-0.073
isode-0.072
Token below
Token ID2174
Feature activation+1.15e-01

Pos logit contributions
umn+1.136
cient+1.091
arch+1.044
anwhile+1.037
isode+1.010
Neg logit contributions
IELD-1.003
hips-0.914
 Desire-0.913
ymm-0.880
 Darling-0.788
Token.
Token ID13
Feature activation-8.42e-01

Pos logit contributions
umn+0.238
cient+0.228
arch+0.217
isode+0.215
anwhile+0.215
Neg logit contributions
IELD-0.202
hips-0.189
 Desire-0.183
ymm-0.183
 suppose-0.169