TOP ACTIVATIONS
MAX = 5.862

the galaxy , and the Milky Way is at least 100
While the First Amendment right to free expression is
2017 - feature - summary /# dr ill through
of its own in the Milky Way . The sun is
volatile east in November after Cong ol ese army troops fled
with Blake Wheeler and Mark Sche ife le has been strong
research and development ( R & D ) tot alled 2
aba osh an Mining Co Ltd , past the ram sh
school , run by the Cong regation of the Religious of
one portable package . Both Flat pak and Snap also claim
There was a sudden wind , and she was nowhere
libraries and packaging formats . Flat pak is here to change
Way . The Milky Way , too , moves
any Linux distribution with the Flat pak software installed should be
alo Bills : G Richie Inc ogn ito Contract
I am against the First Amendment . My tweets reflected my
jets made by Lockheed Martin Corp besides two 35 m patrol
has been lifted and the sun of their future comes shining
South Korea invested in R & D in 2014 out stri
turn : self . clone (), pieces : self . pieces

MIN ACTIVATIONS
MIN = -6.352

United Supporters Trust ( N UST ), N UFC Fans United
, it looks well n igh insol uble , given
of the aims of N UST is to develop formal representation
section on www . n uf c . co . uk
frequent Dinosaur Jr . collaborator Maur a Jasper , an essay
Stone hurst estate in W alth am and Som erville 's
first Ghostbusters , the ne igh bor hood in New York
, Great Depression <|endoftext|> W ALTH AM , Mass . --
extra for other rewards on top of a copy of the
O rec as filling the top three positions in the first
- third benefit cost on top of the salary for a
Stone hurst estate in W alth am and Som erville 's
and 360 Racing occupied the top two positions in the L
p ales in com ­ par ­ is ­ on to
would probably be on your top 100 list too , the
engine , that reaches a top speed of 345 km /
and put it on the top . How will
chased by Satan . On top of that , I got
be made to run on top of it soon , but
s largest owners of top class office buildings . Trump

INTERVAL 2.808 - 5.862
CONTAINS 0.424%

said Friday that being first lady ' has been the greatest
. It was a different story in the U . S
born Farmer on " Fr ances Farmer Will Have Her Revenge
over charged by between 10 % and 30 %.
Nations said last week that Congo had suspended a dozen senior

INTERVAL -0.245 - 2.808
CONTAINS 62.973%

would be like , Oh , they are killing
who looks like he d been sent by central
it first came about when L . A . attorney Andrew
that some prisoners were tortured . Photo
. I mean , that

INTERVAL -3.299 - -0.245
CONTAINS 36.454%

acqu it themselves well -- just not with as much composure
, back with a hopefully improved 100 % on ESA 2014
the CG PR OG RAM / END CG section ! ):
s steam electric eff luent guidelines regulating wastewater streams
K .: You almost want to say to the person you

INTERVAL -6.352 - -3.299
CONTAINS 0.149%

the bleak solitude of Super Metroid , and alongside Mario 64
more jun iper infusion on top as needed .
different . " Both of them love it and
email and never miss our top stories Free Sign Up
, but Obama actually seems pleased : He announced early on
Token the
Token ID262
Feature activation+2.15e-01

Pos logit contributions
 condem+0.034
 newsp+0.028
terms+0.028
 constitu+0.027
cffff+0.027
Neg logit contributions
arching-0.018
toe-0.016
adia-0.016
hang-0.015
estine-0.015
Token galaxy
Token ID16161
Feature activation+7.12e-01

Pos logit contributions
arching+0.139
toe+0.125
adia+0.124
hang+0.117
mat+0.117
Neg logit contributions
 condem-0.228
 newsp-0.185
terms-0.184
 constitu-0.180
cffff-0.177
Token,
Token ID11
Feature activation-2.28e-02

Pos logit contributions
arching+0.393
toe+0.350
adia+0.346
hang+0.322
estine+0.321
Neg logit contributions
 condem-0.821
 newsp-0.680
terms-0.676
 constitu-0.666
cffff-0.648
Token and
Token ID290
Feature activation-2.10e-01

Pos logit contributions
 condem+0.024
 newsp+0.020
terms+0.020
 constitu+0.019
cffff+0.019
Neg logit contributions
arching-0.015
toe-0.013
adia-0.013
hang-0.012
mat-0.012
Token the
Token ID262
Feature activation+1.55e-01

Pos logit contributions
 condem+0.221
 newsp+0.180
terms+0.179
 constitu+0.176
cffff+0.171
Neg logit contributions
arching-0.136
toe-0.123
adia-0.122
hang-0.115
mat-0.115
Token Milky
Token ID34822
Feature activation+5.86e+00

Pos logit contributions
arching+0.112
toe+0.102
adia+0.102
hang+0.097
estine+0.096
Neg logit contributions
 condem-0.152
 newsp-0.122
terms-0.121
 constitu-0.119
cffff-0.115
Token Way
Token ID6378
Feature activation-7.50e-01

Pos logit contributions
arching+3.699
toe+3.471
adia+3.385
zel+3.232
hang+3.206
Neg logit contributions
 condem-6.367
 newsp-5.061
 constitu-4.984
terms-4.878
cffff-4.838
Token is
Token ID318
Feature activation-8.08e-01

Pos logit contributions
 condem+0.788
 newsp+0.640
terms+0.635
 constitu+0.625
cffff+0.607
Neg logit contributions
arching-0.490
toe-0.444
adia-0.441
hang-0.416
estine-0.414
Token at
Token ID379
Feature activation-1.35e-01

Pos logit contributions
 condem+0.740
 newsp+0.581
terms+0.576
 constitu+0.565
cffff+0.544
Neg logit contributions
arching-0.637
toe-0.587
adia-0.584
hang-0.557
estine-0.555
Token least
Token ID1551
Feature activation+2.11e+00

Pos logit contributions
 condem+0.133
 newsp+0.106
terms+0.105
 constitu+0.103
cffff+0.100
Neg logit contributions
arching-0.097
toe-0.089
adia-0.088
hang-0.084
mat-0.084
Token 100
Token ID1802
Feature activation+1.29e-01

Pos logit contributions
arching+1.352
adia+1.211
toe+1.205
mat+1.140
hang+1.137
Neg logit contributions
 condem-2.242
 newsp-1.821
terms-1.802
 constitu-1.786
cffff-1.724
Token
Token ID198
Feature activation+3.37e-01

Pos logit contributions
arching+0.018
toe+0.016
adia+0.016
hang+0.015
estine+0.015
Neg logit contributions
 condem-0.033
 newsp-0.027
terms-0.027
 constitu-0.027
cffff-0.026
Token
Token ID198
Feature activation+1.32e-01

Pos logit contributions
arching+0.190
toe+0.169
adia+0.167
hang+0.156
estine+0.156
Neg logit contributions
 condem-0.382
terms-0.316
 newsp-0.316
 constitu-0.310
cffff-0.301
TokenWhile
Token ID3633
Feature activation+4.65e-01

Pos logit contributions
arching+0.096
toe+0.088
adia+0.088
hang+0.083
estine+0.083
Neg logit contributions
 condem-0.128
 newsp-0.102
terms-0.102
 constitu-0.100
cffff-0.096
Token the
Token ID262
Feature activation-2.14e-01

Pos logit contributions
arching+0.307
toe+0.278
adia+0.276
hang+0.261
mat+0.260
Neg logit contributions
 condem-0.484
 newsp-0.393
terms-0.391
 constitu-0.384
cffff-0.374
Token First
Token ID3274
Feature activation+1.51e+00

Pos logit contributions
 condem+0.208
 newsp+0.166
terms+0.164
 constitu+0.162
cffff+0.156
Neg logit contributions
arching-0.157
toe-0.144
adia-0.142
hang-0.136
estine-0.135
Token Amendment
Token ID8441
Feature activation+5.49e+00

Pos logit contributions
arching+0.910
toe+0.823
adia+0.814
hang+0.763
estine+0.759
Neg logit contributions
 condem-1.655
 newsp-1.355
terms-1.344
 constitu-1.330
cffff-1.293
Token right
Token ID826
Feature activation+1.57e+00

Pos logit contributions
arching+3.540
adia+3.249
toe+3.239
mat+3.120
estine+3.078
Neg logit contributions
 condem-5.447
terms-4.445
 newsp-4.441
cffff-4.420
 constitu-4.359
Token to
Token ID284
Feature activation+4.50e-01

Pos logit contributions
arching+1.008
toe+0.914
adia+0.908
estine+0.860
mat+0.859
Neg logit contributions
 condem-1.640
 newsp-1.342
terms-1.329
 constitu-1.317
cffff-1.290
Token free
Token ID1479
Feature activation+4.18e-01

Pos logit contributions
arching+0.369
toe+0.341
adia+0.339
hang+0.325
estine+0.324
Neg logit contributions
 condem-0.396
 newsp-0.307
terms-0.306
 constitu-0.299
cffff-0.290
Token expression
Token ID5408
Feature activation+2.78e+00

Pos logit contributions
arching+0.280
toe+0.255
adia+0.252
hang+0.238
estine+0.237
Neg logit contributions
 condem-0.432
 newsp-0.349
terms-0.347
 constitu-0.341
cffff-0.330
Token is
Token ID318
Feature activation-4.65e-01

Pos logit contributions
arching+1.777
adia+1.616
toe+1.613
mat+1.533
estine+1.527
Neg logit contributions
 condem-2.916
 newsp-2.383
terms-2.361
 constitu-2.319
cffff-2.291
Token2017
Token ID5539
Feature activation+1.57e+00

Pos logit contributions
arching+0.849
toe+0.770
adia+0.763
hang+0.722
mat+0.720
Neg logit contributions
 condem-1.349
 newsp-1.095
terms-1.078
 constitu-1.071
cffff-1.035
Token-
Token ID12
Feature activation+1.08e+00

Pos logit contributions
arching+0.969
toe+0.873
adia+0.863
hang+0.814
mat+0.809
Neg logit contributions
 condem-1.711
 newsp-1.400
terms-1.387
 constitu-1.371
cffff-1.330
Tokenfeature
Token ID30053
Feature activation+6.28e-02

Pos logit contributions
arching+0.712
toe+0.647
adia+0.638
hang+0.606
mat+0.604
Neg logit contributions
 condem-1.131
 newsp-0.919
terms-0.905
 constitu-0.899
cffff-0.869
Token-
Token ID12
Feature activation+1.43e+00

Pos logit contributions
arching+0.044
toe+0.040
adia+0.039
estine+0.037
hang+0.037
Neg logit contributions
 condem-0.063
terms-0.051
 newsp-0.051
 constitu-0.050
cffff-0.048
Tokensummary
Token ID49736
Feature activation+1.45e+00

Pos logit contributions
arching+0.967
toe+0.879
adia+0.868
hang+0.826
mat+0.825
Neg logit contributions
 condem-1.476
 newsp-1.195
terms-1.173
 constitu-1.168
cffff-1.131
Token/#
Token ID31113
Feature activation+5.38e+00

Pos logit contributions
arching+0.875
toe+0.787
adia+0.778
hang+0.732
mat+0.728
Neg logit contributions
 condem-1.588
 newsp-1.301
terms-1.291
 constitu-1.275
cffff-1.238
Tokendr
Token ID7109
Feature activation+6.01e-01

Pos logit contributions
arching+3.515
toe+3.231
adia+3.153
mat+3.057
hang+3.020
Neg logit contributions
 condem-5.559
 newsp-4.517
 constitu-4.458
terms-4.353
cffff-4.352
Tokenill
Token ID359
Feature activation+4.82e+00

Pos logit contributions
arching+0.470
toe+0.434
adia+0.430
hang+0.411
mat+0.410
Neg logit contributions
 condem-0.554
 newsp-0.435
terms-0.431
 constitu-0.424
cffff-0.408
Tokenthrough
Token ID9579
Feature activation+1.60e+00

Pos logit contributions
arching+3.390
toe+3.120
adia+3.029
mat+2.933
hang+2.903
Neg logit contributions
 condem-4.788
 newsp-3.910
 constitu-3.800
terms-3.681
cffff-3.613
Token
Token ID198
Feature activation+1.87e+00

Pos logit contributions
arching+0.982
toe+0.883
adia+0.873
hang+0.822
mat+0.818
Neg logit contributions
 condem-1.745
 newsp-1.428
terms-1.416
 constitu-1.399
cffff-1.357
Token
Token ID198
Feature activation+1.16e+00

Pos logit contributions
arching+0.982
toe+0.881
adia+0.862
mat+0.808
hang+0.808
Neg logit contributions
 condem-2.234
 newsp-1.856
terms-1.814
 constitu-1.812
cffff-1.765
Token of
Token ID286
Feature activation+6.15e-02

Pos logit contributions
 condem+0.354
 newsp+0.283
terms+0.281
 constitu+0.276
cffff+0.267
Neg logit contributions
arching-0.255
toe-0.234
adia-0.231
hang-0.220
estine-0.219
Token its
Token ID663
Feature activation+4.76e-01

Pos logit contributions
arching+0.047
toe+0.043
adia+0.043
hang+0.041
mat+0.041
Neg logit contributions
 condem-0.058
 newsp-0.046
terms-0.045
 constitu-0.045
cffff-0.043
Token own
Token ID898
Feature activation-1.54e+00

Pos logit contributions
arching+0.251
toe+0.221
adia+0.219
hang+0.204
mat+0.202
Neg logit contributions
 condem-0.560
 newsp-0.466
terms-0.464
 constitu-0.456
cffff-0.446
Token in
Token ID287
Feature activation-7.66e-02

Pos logit contributions
 condem+1.727
 newsp+1.422
terms+1.415
 constitu+1.391
cffff+1.344
Neg logit contributions
arching-0.888
toe-0.800
adia-0.786
hang-0.742
estine-0.738
Token the
Token ID262
Feature activation-2.83e-01

Pos logit contributions
 condem+0.082
 newsp+0.067
terms+0.066
 constitu+0.066
cffff+0.064
Neg logit contributions
arching-0.049
toe-0.044
adia-0.043
hang-0.041
mat-0.041
Token Milky
Token ID34822
Feature activation+5.36e+00

Pos logit contributions
 condem+0.298
 newsp+0.242
terms+0.240
 constitu+0.237
cffff+0.231
Neg logit contributions
arching-0.184
toe-0.166
adia-0.165
hang-0.155
mat-0.155
Token Way
Token ID6378
Feature activation-1.36e+00

Pos logit contributions
arching+3.403
toe+3.205
adia+3.120
mat+3.010
zel+3.000
Neg logit contributions
 condem-5.811
 newsp-4.621
 constitu-4.519
terms-4.428
cffff-4.403
Token.
Token ID13
Feature activation-5.90e-01

Pos logit contributions
 condem+1.361
 newsp+1.093
terms+1.086
 constitu+1.067
cffff+1.028
Neg logit contributions
arching-0.955
toe-0.872
adia-0.862
hang-0.820
estine-0.816
Token The
Token ID383
Feature activation-6.71e-01

Pos logit contributions
 condem+0.607
 newsp+0.491
terms+0.489
 constitu+0.480
cffff+0.464
Neg logit contributions
arching-0.398
toe-0.361
adia-0.357
hang-0.339
estine-0.337
Token sun
Token ID4252
Feature activation+3.31e+00

Pos logit contributions
 condem+0.691
 newsp+0.558
terms+0.554
 constitu+0.545
cffff+0.527
Neg logit contributions
arching-0.452
toe-0.412
adia-0.408
hang-0.386
estine-0.385
Token is
Token ID318
Feature activation-6.45e-01

Pos logit contributions
arching+2.161
toe+1.957
adia+1.951
mat+1.873
hang+1.836
Neg logit contributions
 condem-3.468
 newsp-2.826
terms-2.779
 constitu-2.763
cffff-2.685
Token volatile
Token ID22750
Feature activation+1.13e+00

Pos logit contributions
arching+0.030
toe+0.027
adia+0.027
hang+0.026
estine+0.026
Neg logit contributions
 condem-0.041
 newsp-0.033
terms-0.032
 constitu-0.032
cffff-0.031
Token east
Token ID7627
Feature activation+2.52e+00

Pos logit contributions
arching+0.805
toe+0.740
adia+0.731
mat+0.696
hang+0.696
Neg logit contributions
 condem-1.112
terms-0.887
 newsp-0.886
 constitu-0.861
cffff-0.843
Token in
Token ID287
Feature activation+7.92e-01

Pos logit contributions
arching+1.594
toe+1.425
adia+1.413
mat+1.340
hang+1.329
Neg logit contributions
 condem-2.718
 newsp-2.215
terms-2.199
 constitu-2.174
cffff-2.127
Token November
Token ID3389
Feature activation-8.26e-01

Pos logit contributions
arching+0.539
toe+0.490
adia+0.485
hang+0.460
mat+0.459
Neg logit contributions
 condem-0.810
 newsp-0.653
terms-0.648
 constitu-0.639
cffff-0.620
Token after
Token ID706
Feature activation+1.67e+00

Pos logit contributions
 condem+0.899
terms+0.736
 newsp+0.735
 constitu+0.719
cffff+0.700
Neg logit contributions
arching-0.507
toe-0.455
adia-0.448
hang-0.423
inion-0.422
Token Cong
Token ID2908
Feature activation+5.34e+00

Pos logit contributions
arching+1.142
toe+1.040
adia+1.031
hang+0.975
mat+0.973
Neg logit contributions
 condem-1.704
 newsp-1.374
terms-1.367
 constitu-1.342
cffff-1.303
Tokenol
Token ID349
Feature activation+9.48e-01

Pos logit contributions
arching+3.022
adia+2.799
toe+2.729
mat+2.592
estine+2.537
Neg logit contributions
 condem-6.025
 newsp-4.941
terms-4.891
 constitu-4.825
cffff-4.730
Tokenese
Token ID2771
Feature activation-8.10e-01

Pos logit contributions
arching+0.669
toe+0.611
adia+0.606
hang+0.575
estine+0.573
Neg logit contributions
 condem-0.946
 newsp-0.758
terms-0.753
 constitu-0.740
cffff-0.715
Token army
Token ID5428
Feature activation+2.55e+00

Pos logit contributions
 condem+0.760
 newsp+0.602
terms+0.591
 constitu+0.587
cffff+0.554
Neg logit contributions
arching-0.624
toe-0.570
adia-0.567
hang-0.542
estine-0.538
Token troops
Token ID6553
Feature activation+1.44e+00

Pos logit contributions
arching+1.869
toe+1.723
adia+1.710
mat+1.622
hang+1.618
Neg logit contributions
 condem-2.461
 newsp-1.956
terms-1.952
 constitu-1.903
cffff-1.866
Token fled
Token ID11468
Feature activation+1.97e+00

Pos logit contributions
arching+1.005
toe+0.918
adia+0.911
hang+0.863
estine+0.860
Neg logit contributions
 condem-1.448
 newsp-1.164
terms-1.156
 constitu-1.135
cffff-1.100
Token with
Token ID351
Feature activation-3.69e-01

Pos logit contributions
arching+0.565
toe+0.513
adia+0.508
hang+0.483
mat+0.481
Neg logit contributions
 condem-0.880
 newsp-0.711
terms-0.706
 constitu-0.696
cffff-0.672
Token Blake
Token ID13307
Feature activation-1.16e+00

Pos logit contributions
 condem+0.377
 newsp+0.304
terms+0.302
 constitu+0.297
cffff+0.288
Neg logit contributions
arching-0.251
toe-0.229
adia-0.227
hang-0.215
estine-0.214
Token Wheeler
Token ID27512
Feature activation+1.99e+00

Pos logit contributions
 condem+1.304
 newsp+1.070
terms+1.067
 constitu+1.047
cffff+1.016
Neg logit contributions
arching-0.676
toe-0.600
adia-0.593
estine-0.558
hang-0.556
Token and
Token ID290
Feature activation+7.14e-01

Pos logit contributions
arching+1.205
toe+1.085
adia+1.073
hang+1.013
mat+1.008
Neg logit contributions
 condem-2.190
 newsp-1.798
terms-1.781
 constitu-1.756
cffff-1.702
Token Mark
Token ID2940
Feature activation+1.05e+00

Pos logit contributions
arching+0.504
toe+0.462
adia+0.457
hang+0.435
mat+0.434
Neg logit contributions
 condem-0.711
 newsp-0.569
terms-0.565
 constitu-0.555
cffff-0.539
Token Sche
Token ID10011
Feature activation+5.23e+00

Pos logit contributions
arching+0.481
toe+0.420
adia+0.415
hang+0.384
estine+0.383
Neg logit contributions
 condem-1.308
 newsp-1.095
terms-1.089
 constitu-1.080
cffff-1.054
Tokenife
Token ID901
Feature activation-1.41e+00

Pos logit contributions
arching+3.605
toe+3.337
adia+3.336
mat+3.188
estine+3.165
Neg logit contributions
 condem-5.352
 newsp-4.273
terms-4.162
 constitu-4.158
cffff-4.023
Tokenle
Token ID293
Feature activation+1.45e-01

Pos logit contributions
 condem+1.547
terms+1.285
 newsp+1.267
 constitu+1.248
cffff+1.220
Neg logit contributions
arching-0.828
toe-0.736
adia-0.733
estine-0.683
inion-0.679
Token has
Token ID468
Feature activation-3.20e-01

Pos logit contributions
arching+0.090
toe+0.081
adia+0.080
hang+0.075
estine+0.075
Neg logit contributions
 condem-0.157
 newsp-0.128
terms-0.127
 constitu-0.125
cffff-0.122
Token been
Token ID587
Feature activation+7.74e-02

Pos logit contributions
 condem+0.323
 newsp+0.259
terms+0.257
 constitu+0.254
cffff+0.245
Neg logit contributions
arching-0.223
toe-0.203
adia-0.202
hang-0.191
estine-0.191
Token strong
Token ID1913
Feature activation+7.36e-01

Pos logit contributions
arching+0.054
toe+0.049
adia+0.049
hang+0.046
estine+0.046
Neg logit contributions
 condem-0.078
 newsp-0.063
terms-0.062
 constitu-0.061
cffff-0.059
Token research
Token ID2267
Feature activation+8.97e-01

Pos logit contributions
arching+0.080
toe+0.073
adia+0.072
hang+0.068
estine+0.068
Neg logit contributions
 condem-0.134
 newsp-0.109
terms-0.108
 constitu-0.106
cffff-0.103
Token and
Token ID290
Feature activation-3.43e-01

Pos logit contributions
arching+0.579
toe+0.519
adia+0.519
mat+0.494
hang+0.490
Neg logit contributions
 condem-0.950
 newsp-0.769
terms-0.765
 constitu-0.755
cffff-0.736
Token development
Token ID2478
Feature activation-5.15e-01

Pos logit contributions
 condem+0.338
 newsp+0.271
terms+0.268
 constitu+0.264
cffff+0.255
Neg logit contributions
arching-0.245
toe-0.225
adia-0.222
hang-0.212
estine-0.211
Token (
Token ID357
Feature activation-1.82e-01

Pos logit contributions
 condem+0.524
 newsp+0.422
terms+0.419
 constitu+0.412
cffff+0.399
Neg logit contributions
arching-0.354
toe-0.323
adia-0.320
hang-0.303
estine-0.302
TokenR
Token ID49
Feature activation+1.08e+00

Pos logit contributions
 condem+0.165
 newsp+0.129
terms+0.128
 constitu+0.126
cffff+0.121
Neg logit contributions
arching-0.146
toe-0.134
adia-0.133
hang-0.127
mat-0.127
Token&
Token ID5
Feature activation+5.21e+00

Pos logit contributions
arching+0.389
toe+0.320
adia+0.311
estine+0.283
hang+0.281
Neg logit contributions
 condem-1.447
 newsp-1.232
terms-1.229
 constitu-1.211
cffff-1.185
TokenD
Token ID35
Feature activation+5.82e-01

Pos logit contributions
arching+3.425
toe+3.232
adia+3.187
hang+3.022
mat+3.008
Neg logit contributions
 condem-5.484
 newsp-4.427
terms-4.229
 constitu-4.216
 confir-4.111
Token)
Token ID8
Feature activation+3.77e-01

Pos logit contributions
arching+0.351
toe+0.315
adia+0.312
hang+0.293
estine+0.292
Neg logit contributions
 condem-0.640
 newsp-0.526
terms-0.523
 constitu-0.515
cffff-0.499
Token tot
Token ID2006
Feature activation+6.66e-01

Pos logit contributions
arching+0.268
toe+0.244
adia+0.243
hang+0.231
mat+0.231
Neg logit contributions
 condem-0.374
 newsp-0.299
terms-0.297
 constitu-0.292
cffff-0.283
Tokenalled
Token ID4262
Feature activation-6.64e-01

Pos logit contributions
arching+0.482
toe+0.437
adia+0.434
mat+0.414
hang+0.413
Neg logit contributions
 condem-0.666
 newsp-0.531
terms-0.525
 constitu-0.519
cffff-0.498
Token 2
Token ID362
Feature activation+3.29e-01

Pos logit contributions
 condem+0.635
 newsp+0.504
terms+0.500
 constitu+0.491
cffff+0.473
Neg logit contributions
arching-0.496
toe-0.456
adia-0.452
hang-0.430
estine-0.429
Tokenaba
Token ID15498
Feature activation+8.57e-01

Pos logit contributions
arching+0.141
toe+0.127
adia+0.125
estine+0.117
mat+0.116
Neg logit contributions
 condem-0.296
 newsp-0.246
terms-0.244
 constitu-0.239
cffff-0.233
Tokenosh
Token ID3768
Feature activation-9.95e-01

Pos logit contributions
arching+0.515
toe+0.463
adia+0.457
hang+0.429
estine+0.428
Neg logit contributions
 condem-0.945
 newsp-0.775
terms-0.770
 constitu-0.759
cffff-0.736
Tokenan
Token ID272
Feature activation+5.76e-01

Pos logit contributions
 condem+1.128
 newsp+0.933
terms+0.925
 constitu+0.912
cffff+0.876
Neg logit contributions
arching-0.565
toe-0.504
adia-0.493
estine-0.463
zel-0.456
Token Mining
Token ID29269
Feature activation+1.70e+00

Pos logit contributions
arching+0.355
toe+0.319
adia+0.316
hang+0.297
estine+0.296
Neg logit contributions
 condem-0.627
 newsp-0.513
terms-0.510
 constitu-0.502
cffff-0.486
Token Co
Token ID1766
Feature activation+7.54e-01

Pos logit contributions
arching+1.027
toe+0.923
adia+0.914
hang+0.859
estine+0.855
Neg logit contributions
 condem-1.863
 newsp-1.528
terms-1.518
 constitu-1.495
cffff-1.451
Token Ltd
Token ID12052
Feature activation+5.20e+00

Pos logit contributions
arching+0.427
toe+0.380
adia+0.377
hang+0.352
estine+0.350
Neg logit contributions
 condem-0.858
 newsp-0.709
terms-0.705
 constitu-0.693
cffff-0.674
Token,
Token ID11
Feature activation+1.19e+00

Pos logit contributions
arching+2.741
toe+2.465
adia+2.444
mat+2.289
hang+2.256
Neg logit contributions
 condem-5.969
 newsp-4.982
terms-4.927
 constitu-4.923
cffff-4.764
Token past
Token ID1613
Feature activation-5.19e-01

Pos logit contributions
arching+0.768
toe+0.696
adia+0.689
hang+0.654
mat+0.649
Neg logit contributions
 condem-1.263
 newsp-1.029
terms-1.019
 constitu-1.007
cffff-0.975
Token the
Token ID262
Feature activation+6.90e-01

Pos logit contributions
 condem+0.556
 newsp+0.454
terms+0.450
 constitu+0.442
cffff+0.429
Neg logit contributions
arching-0.327
toe-0.297
adia-0.292
estine-0.276
inion-0.274
Token ram
Token ID15770
Feature activation-7.18e-01

Pos logit contributions
arching+0.483
toe+0.440
adia+0.436
hang+0.417
mat+0.413
Neg logit contributions
 condem-0.693
 newsp-0.556
terms-0.554
 constitu-0.544
cffff-0.527
Tokensh
Token ID1477
Feature activation-1.19e+00

Pos logit contributions
 condem+0.834
terms+0.693
 newsp+0.693
 constitu+0.681
cffff+0.661
Neg logit contributions
arching-0.382
toe-0.339
adia-0.333
estine-0.312
hang-0.308
Token school
Token ID1524
Feature activation-1.34e+00

Pos logit contributions
arching+0.154
toe+0.142
adia+0.141
hang+0.135
estine+0.135
Neg logit contributions
 condem-0.174
 newsp-0.136
terms-0.135
 constitu-0.132
cffff-0.127
Token,
Token ID11
Feature activation-2.68e-01

Pos logit contributions
 condem+1.382
 newsp+1.110
terms+1.100
 constitu+1.084
cffff+1.032
Neg logit contributions
arching-0.914
toe-0.830
adia-0.821
hang-0.781
estine-0.774
Token run
Token ID1057
Feature activation-2.24e+00

Pos logit contributions
 condem+0.301
 newsp+0.248
terms+0.246
 constitu+0.243
cffff+0.236
Neg logit contributions
arching-0.156
toe-0.139
adia-0.138
hang-0.129
estine-0.128
Token by
Token ID416
Feature activation-1.67e+00

Pos logit contributions
 condem+2.037
 newsp+1.585
terms+1.584
 constitu+1.545
cffff+1.499
Neg logit contributions
arching-1.768
toe-1.624
adia-1.609
hang-1.545
estine-1.544
Token the
Token ID262
Feature activation+2.98e-01

Pos logit contributions
 condem+1.731
 newsp+1.400
terms+1.384
 constitu+1.370
cffff+1.310
Neg logit contributions
arching-1.109
toe-1.010
adia-0.990
estine-0.943
hang-0.937
Token Cong
Token ID2908
Feature activation+5.14e+00

Pos logit contributions
arching+0.209
toe+0.191
adia+0.189
hang+0.179
estine+0.179
Neg logit contributions
 condem-0.299
 newsp-0.240
terms-0.238
 constitu-0.234
cffff-0.226
Tokenregation
Token ID43068
Feature activation+1.97e+00

Pos logit contributions
arching+2.862
adia+2.569
toe+2.534
hang+2.369
mat+2.368
Neg logit contributions
 condem-5.873
 newsp-4.887
terms-4.817
 constitu-4.769
cffff-4.605
Token of
Token ID286
Feature activation-2.69e-01

Pos logit contributions
arching+1.444
adia+1.331
toe+1.327
mat+1.258
hang+1.253
Neg logit contributions
 condem-1.899
 newsp-1.520
terms-1.503
 constitu-1.471
cffff-1.435
Token the
Token ID262
Feature activation+7.73e-02

Pos logit contributions
 condem+0.274
 newsp+0.221
terms+0.221
 constitu+0.217
cffff+0.209
Neg logit contributions
arching-0.183
toe-0.167
adia-0.165
hang-0.156
estine-0.156
Token Religious
Token ID23945
Feature activation+9.31e-01

Pos logit contributions
arching+0.055
toe+0.050
adia+0.050
hang+0.047
estine+0.047
Neg logit contributions
 condem-0.077
 newsp-0.062
terms-0.061
 constitu-0.060
cffff-0.058
Token of
Token ID286
Feature activation-7.73e-01

Pos logit contributions
arching+0.626
toe+0.568
adia+0.563
hang+0.533
estine+0.531
Neg logit contributions
 condem-0.960
 newsp-0.776
terms-0.772
 constitu-0.759
cffff-0.734
Token one
Token ID530
Feature activation+2.88e-01

Pos logit contributions
arching+0.048
toe+0.044
adia+0.044
hang+0.041
mat+0.041
Neg logit contributions
 condem-0.068
 newsp-0.054
terms-0.054
 constitu-0.053
cffff-0.052
Token portable
Token ID17726
Feature activation+6.95e-01

Pos logit contributions
arching+0.198
toe+0.179
adia+0.177
hang+0.169
estine+0.168
Neg logit contributions
 condem-0.293
 newsp-0.234
terms-0.234
 constitu-0.230
cffff-0.226
Token package
Token ID5301
Feature activation-8.35e-01

Pos logit contributions
arching+0.485
toe+0.441
adia+0.439
estine+0.417
hang+0.417
Neg logit contributions
 condem-0.697
terms-0.556
 newsp-0.554
 constitu-0.549
cffff-0.533
Token.
Token ID13
Feature activation+3.92e-01

Pos logit contributions
 condem+0.902
 newsp+0.737
terms+0.733
 constitu+0.722
cffff+0.700
Neg logit contributions
arching-0.521
toe-0.470
adia-0.466
hang-0.438
estine-0.437
Token Both
Token ID5747
Feature activation-2.88e+00

Pos logit contributions
arching+0.246
toe+0.223
adia+0.220
hang+0.208
mat+0.207
Neg logit contributions
 condem-0.422
 newsp-0.344
terms-0.341
 constitu-0.337
cffff-0.327
Token Flat
Token ID21939
Feature activation+5.14e+00

Pos logit contributions
 condem+2.659
 newsp+2.106
terms+2.079
 constitu+2.014
cffff+1.924
Neg logit contributions
arching-2.251
toe-2.069
adia-2.052
inion-1.959
hang-1.957
Tokenpak
Token ID41091
Feature activation+2.03e+00

Pos logit contributions
arching+1.979
toe+1.684
adia+1.661
mat+1.531
hang+1.498
Neg logit contributions
 condem-6.829
 newsp-5.768
 constitu-5.728
terms-5.618
cffff-5.558
Token and
Token ID290
Feature activation+2.99e-01

Pos logit contributions
arching+1.086
toe+0.965
adia+0.964
mat+0.893
hang+0.890
Neg logit contributions
 condem-2.354
 newsp-1.944
terms-1.932
 constitu-1.928
cffff-1.882
Token Snap
Token ID16026
Feature activation+1.56e+00

Pos logit contributions
arching+0.198
toe+0.179
adia+0.178
hang+0.168
estine+0.167
Neg logit contributions
 condem-0.312
 newsp-0.253
terms-0.251
 constitu-0.247
cffff-0.239
Token also
Token ID635
Feature activation-8.08e-01

Pos logit contributions
arching+1.050
toe+0.961
adia+0.953
hang+0.906
estine+0.904
Neg logit contributions
 condem-1.594
 newsp-1.285
terms-1.276
 constitu-1.267
cffff-1.228
Token claim
Token ID1624
Feature activation-2.01e-01

Pos logit contributions
 condem+0.808
 newsp+0.646
terms+0.643
 constitu+0.629
cffff+0.607
Neg logit contributions
arching-0.573
toe-0.519
adia-0.516
hang-0.488
estine-0.487
Token
Token ID251
Feature activation+1.65e+00

Pos logit contributions
arching+0.773
toe+0.710
adia+0.706
hang+0.664
estine+0.659
Neg logit contributions
 condem-1.490
 newsp-1.213
terms-1.212
 constitu-1.194
cffff-1.186
Token There
Token ID1318
Feature activation+4.90e-01

Pos logit contributions
arching+1.048
toe+0.953
adia+0.944
hang+0.893
mat+0.888
Neg logit contributions
 condem-1.761
 newsp-1.433
terms-1.420
 constitu-1.403
cffff-1.362
Token was
Token ID373
Feature activation+1.14e+00

Pos logit contributions
arching+0.309
toe+0.279
adia+0.276
hang+0.260
estine+0.260
Neg logit contributions
 condem-0.526
 newsp-0.430
terms-0.427
 constitu-0.420
cffff-0.407
Token a
Token ID257
Feature activation+7.65e-01

Pos logit contributions
arching+0.747
toe+0.678
adia+0.672
hang+0.635
mat+0.632
Neg logit contributions
 condem-1.198
 newsp-0.971
terms-0.966
 constitu-0.951
cffff-0.921
Token sudden
Token ID4802
Feature activation+9.83e-01

Pos logit contributions
arching+0.548
toe+0.502
adia+0.498
hang+0.475
mat+0.472
Neg logit contributions
 condem-0.755
 newsp-0.600
terms-0.600
 constitu-0.589
cffff-0.570
Token wind
Token ID2344
Feature activation+5.11e+00

Pos logit contributions
arching+0.700
toe+0.640
adia+0.634
hang+0.604
estine+0.600
Neg logit contributions
 condem-0.973
 newsp-0.777
terms-0.775
 constitu-0.760
cffff-0.737
Token,
Token ID11
Feature activation+1.30e+00

Pos logit contributions
arching+3.215
toe+2.903
adia+2.849
hang+2.757
mat+2.726
Neg logit contributions
 condem-5.441
 newsp-4.391
terms-4.389
 constitu-4.345
cffff-4.261
Token and
Token ID290
Feature activation+4.06e-01

Pos logit contributions
arching+0.776
toe+0.695
adia+0.691
hang+0.652
mat+0.645
Neg logit contributions
 condem-1.444
 newsp-1.185
terms-1.179
 constitu-1.162
cffff-1.130
Token she
Token ID673
Feature activation+3.97e-01

Pos logit contributions
arching+0.263
toe+0.239
adia+0.237
hang+0.224
mat+0.223
Neg logit contributions
 condem-0.427
 newsp-0.346
terms-0.345
 constitu-0.339
cffff-0.329
Token was
Token ID373
Feature activation+1.05e+00

Pos logit contributions
arching+0.276
toe+0.251
adia+0.249
hang+0.236
estine+0.235
Neg logit contributions
 condem-0.402
 newsp-0.323
terms-0.321
 constitu-0.315
cffff-0.305
Token nowhere
Token ID12062
Feature activation+5.46e-01

Pos logit contributions
arching+0.742
toe+0.679
adia+0.672
hang+0.641
mat+0.638
Neg logit contributions
 condem-1.045
 newsp-0.839
terms-0.834
 constitu-0.819
cffff-0.792
Token libraries
Token ID12782
Feature activation-1.51e+00

Pos logit contributions
 condem+0.242
 newsp+0.195
terms+0.194
 constitu+0.190
cffff+0.184
Neg logit contributions
arching-0.161
toe-0.147
adia-0.146
hang-0.138
estine-0.138
Token and
Token ID290
Feature activation-1.62e-01

Pos logit contributions
 condem+1.680
 newsp+1.382
terms+1.373
 constitu+1.351
cffff+1.306
Neg logit contributions
arching-0.888
toe-0.798
adia-0.787
estine-0.740
hang-0.738
Token packaging
Token ID16846
Feature activation-6.76e-01

Pos logit contributions
 condem+0.149
 newsp+0.117
terms+0.116
 constitu+0.114
cffff+0.111
Neg logit contributions
arching-0.126
toe-0.116
adia-0.115
hang-0.110
mat-0.110
Token formats
Token ID17519
Feature activation+1.30e+00

Pos logit contributions
 condem+0.694
 newsp+0.560
terms+0.556
 constitu+0.547
cffff+0.529
Neg logit contributions
arching-0.458
toe-0.417
adia-0.413
hang-0.391
estine-0.390
Token.
Token ID13
Feature activation+3.48e-01

Pos logit contributions
arching+0.755
toe+0.676
adia+0.669
hang+0.628
estine+0.624
Neg logit contributions
 condem-1.454
 newsp-1.197
terms-1.190
 constitu-1.174
cffff-1.141
Token Flat
Token ID21939
Feature activation+5.07e+00

Pos logit contributions
arching+0.235
toe+0.214
adia+0.212
hang+0.201
estine+0.200
Neg logit contributions
 condem-0.358
 newsp-0.289
terms-0.287
 constitu-0.283
cffff-0.274
Tokenpak
Token ID41091
Feature activation+2.12e+00

Pos logit contributions
arching+2.024
toe+1.695
adia+1.685
mat+1.532
hang+1.529
Neg logit contributions
 condem-6.623
 newsp-5.603
 constitu-5.549
terms-5.514
cffff-5.408
Token is
Token ID318
Feature activation-1.69e-01

Pos logit contributions
arching+1.379
toe+1.253
adia+1.246
mat+1.179
estine+1.173
Neg logit contributions
 condem-2.215
 newsp-1.797
terms-1.788
 constitu-1.771
cffff-1.731
Token here
Token ID994
Feature activation+2.35e+00

Pos logit contributions
 condem+0.178
 newsp+0.144
terms+0.143
 constitu+0.141
cffff+0.136
Neg logit contributions
arching-0.111
toe-0.101
adia-0.100
hang-0.094
estine-0.094
Token to
Token ID284
Feature activation+1.20e+00

Pos logit contributions
arching+1.498
toe+1.359
adia+1.343
mat+1.262
hang+1.262
Neg logit contributions
 condem-2.499
 newsp-2.034
terms-2.024
 constitu-1.998
cffff-1.933
Token change
Token ID1487
Feature activation+6.95e-01

Pos logit contributions
arching+0.776
toe+0.703
adia+0.696
hang+0.658
estine+0.655
Neg logit contributions
 condem-1.268
 newsp-1.031
terms-1.025
 constitu-1.009
cffff-0.978
Token Way
Token ID6378
Feature activation-1.05e+00

Pos logit contributions
arching+3.154
toe+2.957
adia+2.880
zel+2.755
mat+2.748
Neg logit contributions
 condem-5.250
 newsp-4.164
 constitu-4.085
terms-3.999
cffff-3.961
Token.
Token ID13
Feature activation-3.01e-01

Pos logit contributions
 condem+1.221
 newsp+1.015
terms+1.009
 constitu+0.994
cffff+0.966
Neg logit contributions
arching-0.562
toe-0.499
adia-0.492
hang-0.460
estine-0.456
Token
Token ID198
Feature activation+5.35e-02

Pos logit contributions
 condem+0.336
 newsp+0.276
terms+0.275
 constitu+0.271
cffff+0.263
Neg logit contributions
arching-0.177
toe-0.159
adia-0.157
hang-0.147
estine-0.147
Token
Token ID198
Feature activation-1.81e-01

Pos logit contributions
arching+0.030
toe+0.027
adia+0.027
hang+0.025
estine+0.025
Neg logit contributions
 condem-0.060
terms-0.050
 newsp-0.050
 constitu-0.049
cffff-0.048
TokenThe
Token ID464
Feature activation-7.60e-01

Pos logit contributions
 condem+0.177
 newsp+0.141
terms+0.140
 constitu+0.138
cffff+0.133
Neg logit contributions
arching-0.130
toe-0.119
adia-0.118
hang-0.113
estine-0.112
Token Milky
Token ID34822
Feature activation+5.06e+00

Pos logit contributions
 condem+0.770
 newsp+0.620
terms+0.616
 constitu+0.605
cffff+0.585
Neg logit contributions
arching-0.524
toe-0.478
adia-0.474
hang-0.449
estine-0.448
Token Way
Token ID6378
Feature activation-6.07e-01

Pos logit contributions
arching+3.289
toe+3.056
adia+2.995
hang+2.843
mat+2.842
Neg logit contributions
 condem-5.377
 newsp-4.279
 constitu-4.212
terms-4.158
cffff-4.113
Token,
Token ID11
Feature activation-3.51e-01

Pos logit contributions
 condem+0.637
 newsp+0.517
terms+0.513
 constitu+0.505
cffff+0.490
Neg logit contributions
arching-0.397
toe-0.360
adia-0.356
hang-0.337
estine-0.335
Token too
Token ID1165
Feature activation+5.15e-01

Pos logit contributions
 condem+0.374
 newsp+0.305
terms+0.302
 constitu+0.298
cffff+0.289
Neg logit contributions
arching-0.225
toe-0.203
adia-0.202
hang-0.190
mat-0.190
Token,
Token ID11
Feature activation-1.01e+00

Pos logit contributions
arching+0.284
toe+0.253
adia+0.249
hang+0.232
estine+0.231
Neg logit contributions
 condem-0.593
 newsp-0.490
terms-0.490
 constitu-0.483
cffff-0.468
Token moves
Token ID6100
Feature activation-7.44e-01

Pos logit contributions
 condem+1.042
 newsp+0.842
terms+0.836
 constitu+0.823
cffff+0.796
Neg logit contributions
arching-0.682
toe-0.620
adia-0.614
hang-0.581
estine-0.579
Token any
Token ID597
Feature activation-7.53e-02

Pos logit contributions
 condem+0.827
 newsp+0.672
terms+0.668
 constitu+0.656
cffff+0.636
Neg logit contributions
arching-0.509
toe-0.462
adia-0.457
hang-0.432
estine-0.430
Token Linux
Token ID7020
Feature activation-1.17e+00

Pos logit contributions
 condem+0.068
 newsp+0.053
terms+0.053
 constitu+0.052
cffff+0.050
Neg logit contributions
arching-0.060
toe-0.055
adia-0.055
hang-0.053
mat-0.052
Token distribution
Token ID6082
Feature activation-7.17e-01

Pos logit contributions
 condem+0.959
 newsp+0.726
terms+0.720
 constitu+0.705
cffff+0.676
Neg logit contributions
arching-1.037
toe-0.965
adia-0.959
hang-0.921
estine-0.918
Token with
Token ID351
Feature activation-1.03e-01

Pos logit contributions
 condem+0.785
 newsp+0.642
terms+0.638
 constitu+0.628
cffff+0.608
Neg logit contributions
arching-0.437
toe-0.394
adia-0.390
hang-0.367
estine-0.364
Token the
Token ID262
Feature activation+9.48e-02

Pos logit contributions
 condem+0.109
 newsp+0.089
terms+0.088
 constitu+0.087
cffff+0.084
Neg logit contributions
arching-0.066
toe-0.060
adia-0.060
hang-0.056
estine-0.056
Token Flat
Token ID21939
Feature activation+5.02e+00

Pos logit contributions
arching+0.064
toe+0.058
adia+0.058
hang+0.055
estine+0.055
Neg logit contributions
 condem-0.097
 newsp-0.079
terms-0.078
 constitu-0.077
cffff-0.074
Tokenpak
Token ID41091
Feature activation+2.04e+00

Pos logit contributions
arching+2.019
toe+1.710
adia+1.683
mat+1.531
hang+1.530
Neg logit contributions
 condem-6.529
 newsp-5.516
 constitu-5.459
terms-5.446
cffff-5.319
Token software
Token ID3788
Feature activation-1.47e-02

Pos logit contributions
arching+1.353
toe+1.224
adia+1.212
hang+1.147
mat+1.147
Neg logit contributions
 condem-2.115
 newsp-1.706
terms-1.700
 constitu-1.685
cffff-1.647
Token installed
Token ID6589
Feature activation-2.08e+00

Pos logit contributions
 condem+0.016
 newsp+0.014
terms+0.013
 constitu+0.013
cffff+0.013
Neg logit contributions
arching-0.009
toe-0.008
adia-0.008
hang-0.007
estine-0.007
Token should
Token ID815
Feature activation-6.79e-01

Pos logit contributions
 condem+2.307
 newsp+1.887
terms+1.886
 constitu+1.839
cffff+1.782
Neg logit contributions
arching-1.237
toe-1.111
adia-1.097
hang-1.032
estine-1.025
Token be
Token ID307
Feature activation+2.12e-01

Pos logit contributions
 condem+0.659
 newsp+0.523
terms+0.522
 constitu+0.508
cffff+0.492
Neg logit contributions
arching-0.498
toe-0.453
adia-0.450
hang-0.426
inion-0.425
Tokenalo
Token ID7335
Feature activation+2.19e+00

Pos logit contributions
 condem+0.764
 newsp+0.591
terms+0.581
 constitu+0.575
cffff+0.551
Neg logit contributions
arching-0.733
toe-0.678
adia-0.675
hang-0.647
estine-0.644
Token Bills
Token ID20209
Feature activation+5.98e-01

Pos logit contributions
arching+1.729
toe+1.597
adia+1.577
hang+1.522
mat+1.509
Neg logit contributions
 condem-1.996
 newsp-1.572
terms-1.559
 constitu-1.522
cffff-1.489
Token:
Token ID25
Feature activation+7.05e-01

Pos logit contributions
arching+0.352
toe+0.317
adia+0.313
hang+0.294
mat+0.293
Neg logit contributions
 condem-0.664
 newsp-0.547
terms-0.543
 constitu-0.535
cffff-0.523
Token G
Token ID402
Feature activation+2.31e-01

Pos logit contributions
arching+0.497
toe+0.458
adia+0.452
hang+0.430
mat+0.429
Neg logit contributions
 condem-0.703
 newsp-0.563
terms-0.557
 constitu-0.548
cffff-0.534
Token Richie
Token ID49209
Feature activation-1.72e-01

Pos logit contributions
arching+0.157
toe+0.144
adia+0.142
mat+0.134
hang+0.134
Neg logit contributions
 condem-0.237
 newsp-0.191
terms-0.190
 constitu-0.186
cffff-0.182
Token Inc
Token ID3457
Feature activation+4.99e+00

Pos logit contributions
 condem+0.158
 newsp+0.124
terms+0.124
 constitu+0.121
cffff+0.116
Neg logit contributions
arching-0.135
toe-0.124
adia-0.123
hang-0.117
estine-0.117
Tokenogn
Token ID2360
Feature activation-2.17e+00

Pos logit contributions
arching+2.471
toe+2.255
adia+2.225
zel+2.100
mat+2.028
Neg logit contributions
 condem-6.035
 newsp-5.085
 constitu-4.978
terms-4.968
cffff-4.786
Tokenito
Token ID10094
Feature activation+1.90e-01

Pos logit contributions
 condem+2.656
 newsp+2.226
terms+2.217
 constitu+2.188
cffff+2.132
Neg logit contributions
arching-1.047
toe-0.902
adia-0.883
hang-0.822
mat-0.812
Token
Token ID198
Feature activation+5.81e-01

Pos logit contributions
arching+0.116
toe+0.104
adia+0.103
hang+0.097
estine+0.096
Neg logit contributions
 condem-0.208
 newsp-0.170
terms-0.169
 constitu-0.167
cffff-0.162
Token
Token ID198
Feature activation+3.81e-01

Pos logit contributions
arching+0.317
toe+0.283
adia+0.279
hang+0.260
mat+0.259
Neg logit contributions
 condem-0.674
 newsp-0.559
terms-0.553
 constitu-0.547
cffff-0.532
TokenContract
Token ID45845
Feature activation+6.12e-01

Pos logit contributions
arching+0.281
toe+0.260
adia+0.254
hang+0.245
mat+0.245
Neg logit contributions
 condem-0.373
 newsp-0.296
terms-0.289
 constitu-0.288
cffff-0.278
Token I
Token ID314
Feature activation+1.27e+00

Pos logit contributions
 condem+0.069
 newsp+0.056
terms+0.055
 constitu+0.054
cffff+0.053
Neg logit contributions
arching-0.047
toe-0.043
adia-0.043
hang-0.041
estine-0.040
Token am
Token ID716
Feature activation+2.41e+00

Pos logit contributions
arching+0.818
toe+0.747
adia+0.739
hang+0.699
estine+0.697
Neg logit contributions
 condem-1.333
 newsp-1.089
terms-1.081
 constitu-1.063
cffff-1.034
Token against
Token ID1028
Feature activation+2.33e+00

Pos logit contributions
arching+1.580
adia+1.437
toe+1.435
mat+1.377
hang+1.364
Neg logit contributions
 condem-2.452
terms-2.024
 newsp-2.008
 constitu-1.960
cffff-1.937
Token the
Token ID262
Feature activation+1.16e+00

Pos logit contributions
arching+1.568
toe+1.427
adia+1.416
hang+1.343
mat+1.336
Neg logit contributions
 condem-2.374
 newsp-1.926
terms-1.917
 constitu-1.878
cffff-1.835
Token First
Token ID3274
Feature activation+1.61e+00

Pos logit contributions
arching+0.811
toe+0.740
adia+0.734
hang+0.700
estine+0.697
Neg logit contributions
 condem-1.147
 newsp-0.926
terms-0.919
 constitu-0.901
cffff-0.886
Token Amendment
Token ID8441
Feature activation+4.99e+00

Pos logit contributions
arching+0.966
toe+0.880
adia+0.868
hang+0.817
mat+0.810
Neg logit contributions
 condem-1.780
 newsp-1.455
terms-1.442
 constitu-1.428
cffff-1.387
Token.
Token ID13
Feature activation+5.58e-01

Pos logit contributions
arching+2.604
toe+2.320
adia+2.296
mat+2.135
estine+2.133
Neg logit contributions
 condem-5.791
 newsp-4.825
terms-4.798
 constitu-4.723
cffff-4.642
Token My
Token ID2011
Feature activation-4.09e-01

Pos logit contributions
arching+0.334
toe+0.303
adia+0.299
hang+0.281
mat+0.280
Neg logit contributions
 condem-0.616
 newsp-0.506
terms-0.501
 constitu-0.494
cffff-0.482
Token tweets
Token ID12665
Feature activation+4.11e-01

Pos logit contributions
 condem+0.410
 newsp+0.328
terms+0.326
 constitu+0.321
cffff+0.309
Neg logit contributions
arching-0.288
toe-0.263
adia-0.261
hang-0.247
estine-0.247
Token reflected
Token ID12548
Feature activation+1.99e-01

Pos logit contributions
arching+0.289
toe+0.264
adia+0.262
hang+0.249
mat+0.248
Neg logit contributions
 condem-0.409
 newsp-0.329
terms-0.326
 constitu-0.322
cffff-0.313
Token my
Token ID616
Feature activation+1.64e-01

Pos logit contributions
arching+0.126
toe+0.114
adia+0.112
hang+0.106
estine+0.106
Neg logit contributions
 condem-0.214
 newsp-0.174
terms-0.173
 constitu-0.170
cffff-0.165
Token jets
Token ID20792
Feature activation+9.19e-01

Pos logit contributions
 condem+0.859
 newsp+0.692
terms+0.689
 constitu+0.678
cffff+0.657
Neg logit contributions
arching-0.574
toe-0.522
adia-0.518
hang-0.491
mat-0.489
Token made
Token ID925
Feature activation+4.78e-01

Pos logit contributions
arching+0.531
toe+0.475
adia+0.471
hang+0.441
estine+0.438
Neg logit contributions
 condem-1.034
 newsp-0.853
terms-0.847
 constitu-0.835
cffff-0.811
Token by
Token ID416
Feature activation-4.63e-01

Pos logit contributions
arching+0.342
toe+0.313
adia+0.312
hang+0.296
mat+0.295
Neg logit contributions
 condem-0.470
 newsp-0.376
terms-0.374
 constitu-0.368
cffff-0.356
Token Lockheed
Token ID34238
Feature activation+1.70e+00

Pos logit contributions
 condem+0.469
 newsp+0.377
terms+0.374
 constitu+0.369
cffff+0.355
Neg logit contributions
arching-0.319
toe-0.293
adia-0.288
estine-0.274
hang-0.273
Token Martin
Token ID5780
Feature activation-4.84e-01

Pos logit contributions
arching+1.468
toe+1.367
adia+1.357
mat+1.301
hang+1.299
Neg logit contributions
 condem-1.424
 newsp-1.089
terms-1.077
 constitu-1.062
cffff-1.021
Token Corp
Token ID11421
Feature activation+4.98e+00

Pos logit contributions
 condem+0.538
 newsp+0.442
terms+0.439
 constitu+0.431
cffff+0.418
Neg logit contributions
arching-0.288
toe-0.258
adia-0.256
hang-0.240
estine-0.239
Token besides
Token ID13769
Feature activation+1.02e+00

Pos logit contributions
arching+2.766
adia+2.458
toe+2.444
mat+2.309
inion+2.259
Neg logit contributions
 condem-5.615
 newsp-4.656
 constitu-4.620
terms-4.575
cffff-4.459
Token two
Token ID734
Feature activation+4.79e-01

Pos logit contributions
arching+0.675
toe+0.611
adia+0.607
hang+0.574
mat+0.572
Neg logit contributions
 condem-1.053
 newsp-0.854
terms-0.848
 constitu-0.835
cffff-0.811
Token 35
Token ID3439
Feature activation+7.92e-01

Pos logit contributions
arching+0.310
toe+0.279
adia+0.277
hang+0.262
mat+0.261
Neg logit contributions
 condem-0.505
 newsp-0.412
terms-0.410
 constitu-0.404
cffff-0.394
Token m
Token ID285
Feature activation+4.14e-01

Pos logit contributions
arching+0.508
toe+0.462
adia+0.456
hang+0.432
mat+0.429
Neg logit contributions
 condem-0.838
 newsp-0.683
terms-0.679
 constitu-0.670
cffff-0.649
Token patrol
Token ID13969
Feature activation+8.58e-01

Pos logit contributions
arching+0.274
toe+0.248
adia+0.247
mat+0.233
hang+0.233
Neg logit contributions
 condem-0.433
 newsp-0.354
terms-0.349
 constitu-0.349
cffff-0.333
Token has
Token ID468
Feature activation-9.92e-01

Pos logit contributions
 condem+0.263
 newsp+0.215
terms+0.213
 constitu+0.210
cffff+0.203
Neg logit contributions
arching-0.153
toe-0.138
adia-0.136
hang-0.128
estine-0.128
Token been
Token ID587
Feature activation-2.60e-01

Pos logit contributions
 condem+1.001
 newsp+0.803
terms+0.795
 constitu+0.784
cffff+0.753
Neg logit contributions
arching-0.689
toe-0.631
adia-0.625
hang-0.593
estine-0.590
Token lifted
Token ID13663
Feature activation+2.16e-02

Pos logit contributions
 condem+0.266
 newsp+0.214
terms+0.213
 constitu+0.209
cffff+0.202
Neg logit contributions
arching-0.178
toe-0.162
adia-0.161
hang-0.152
estine-0.152
Token and
Token ID290
Feature activation+7.59e-01

Pos logit contributions
arching+0.014
toe+0.012
adia+0.012
hang+0.012
estine+0.012
Neg logit contributions
 condem-0.023
 newsp-0.019
terms-0.019
 constitu-0.018
cffff-0.018
Token the
Token ID262
Feature activation+6.82e-01

Pos logit contributions
arching+0.510
toe+0.463
adia+0.459
hang+0.435
estine+0.434
Neg logit contributions
 condem-0.781
 newsp-0.633
terms-0.630
 constitu-0.618
cffff-0.601
Token sun
Token ID4252
Feature activation+4.95e+00

Pos logit contributions
arching+0.491
toe+0.448
adia+0.445
hang+0.424
mat+0.422
Neg logit contributions
 condem-0.670
 newsp-0.534
terms-0.532
 constitu-0.522
cffff-0.506
Token of
Token ID286
Feature activation+3.29e-02

Pos logit contributions
arching+3.197
toe+2.867
adia+2.855
mat+2.750
hang+2.697
Neg logit contributions
 condem-5.197
 newsp-4.270
terms-4.169
 constitu-4.146
cffff-4.037
Token their
Token ID511
Feature activation+2.95e-01

Pos logit contributions
arching+0.021
toe+0.019
adia+0.018
hang+0.017
estine+0.017
Neg logit contributions
 condem-0.035
 newsp-0.029
terms-0.029
 constitu-0.028
cffff-0.027
Token future
Token ID2003
Feature activation+2.23e+00

Pos logit contributions
arching+0.211
toe+0.192
adia+0.191
hang+0.181
estine+0.181
Neg logit contributions
 condem-0.292
 newsp-0.234
terms-0.232
 constitu-0.228
cffff-0.221
Token comes
Token ID2058
Feature activation+3.46e-02

Pos logit contributions
arching+1.549
toe+1.403
adia+1.399
hang+1.329
mat+1.326
Neg logit contributions
 condem-2.249
 newsp-1.811
terms-1.798
 constitu-1.767
cffff-1.718
Token shining
Token ID22751
Feature activation-1.10e+00

Pos logit contributions
arching+0.024
toe+0.022
adia+0.021
hang+0.020
estine+0.020
Neg logit contributions
 condem-0.035
 newsp-0.028
terms-0.028
 constitu-0.028
cffff-0.027
Token South
Token ID2520
Feature activation+9.65e-01

Pos logit contributions
 condem+0.060
 newsp+0.049
terms+0.048
 constitu+0.048
cffff+0.046
Neg logit contributions
arching-0.038
toe-0.035
adia-0.034
hang-0.032
estine-0.032
Token Korea
Token ID4969
Feature activation+9.94e-01

Pos logit contributions
arching+0.689
toe+0.632
adia+0.624
hang+0.594
estine+0.591
Neg logit contributions
 condem-0.956
 newsp-0.765
terms-0.759
 constitu-0.746
cffff-0.718
Token invested
Token ID13356
Feature activation+5.47e-01

Pos logit contributions
arching+0.693
toe+0.634
adia+0.631
hang+0.596
mat+0.596
Neg logit contributions
 condem-0.995
 newsp-0.798
terms-0.794
 constitu-0.782
cffff-0.761
Token in
Token ID287
Feature activation-4.28e-01

Pos logit contributions
arching+0.317
toe+0.283
adia+0.280
hang+0.262
estine+0.261
Neg logit contributions
 condem-0.616
 newsp-0.508
terms-0.505
 constitu-0.497
cffff-0.483
Token R
Token ID371
Feature activation-2.64e-01

Pos logit contributions
 condem+0.463
 newsp+0.380
terms+0.376
 constitu+0.371
cffff+0.358
Neg logit contributions
arching-0.265
toe-0.241
adia-0.237
hang-0.223
estine-0.223
Token&
Token ID5
Feature activation+4.93e+00

Pos logit contributions
 condem+0.356
 newsp+0.305
terms+0.304
 constitu+0.300
cffff+0.293
Neg logit contributions
arching-0.092
toe-0.076
adia-0.074
estine-0.066
hang-0.066
TokenD
Token ID35
Feature activation+2.38e-01

Pos logit contributions
arching+3.179
toe+3.027
adia+2.973
mat+2.883
hang+2.848
Neg logit contributions
 condem-5.374
 newsp-4.305
 constitu-4.095
 confir-4.055
 destro-3.997
Token in
Token ID287
Feature activation-2.11e-01

Pos logit contributions
arching+0.156
toe+0.141
adia+0.140
hang+0.132
estine+0.132
Neg logit contributions
 condem-0.249
 newsp-0.202
terms-0.201
 constitu-0.198
cffff-0.192
Token 2014
Token ID1946
Feature activation+3.12e-01

Pos logit contributions
 condem+0.209
 newsp+0.167
terms+0.165
 constitu+0.163
cffff+0.158
Neg logit contributions
arching-0.151
toe-0.138
adia-0.136
hang-0.130
mat-0.129
Token out
Token ID503
Feature activation-4.89e-01

Pos logit contributions
arching+0.212
toe+0.193
adia+0.191
hang+0.181
mat+0.180
Neg logit contributions
 condem-0.320
 newsp-0.259
terms-0.257
 constitu-0.253
cffff-0.245
Tokenstri
Token ID33565
Feature activation+1.57e+00

Pos logit contributions
 condem+0.493
 newsp+0.396
terms+0.391
 constitu+0.387
cffff+0.374
Neg logit contributions
arching-0.341
toe-0.312
adia-0.309
hang-0.295
mat-0.295
Token turn
Token ID1210
Feature activation-7.81e-01

Pos logit contributions
arching+0.985
toe+0.896
adia+0.891
hang+0.852
mat+0.850
Neg logit contributions
 condem-1.497
 newsp-1.201
terms-1.197
 constitu-1.177
cffff-1.151
Token :
Token ID1058
Feature activation+1.25e+00

Pos logit contributions
 condem+0.880
 newsp+0.727
terms+0.726
 constitu+0.711
cffff+0.687
Neg logit contributions
arching-0.447
toe-0.400
adia-0.393
hang-0.367
estine-0.365
Token self
Token ID2116
Feature activation+5.00e-01

Pos logit contributions
arching+0.804
toe+0.728
adia+0.721
hang+0.681
estine+0.678
Neg logit contributions
 condem-1.326
 newsp-1.077
terms-1.071
 constitu-1.054
cffff-1.022
Token.
Token ID764
Feature activation+1.29e+00

Pos logit contributions
arching+0.312
toe+0.282
adia+0.279
hang+0.262
estine+0.260
Neg logit contributions
 condem-0.539
 newsp-0.440
terms-0.439
 constitu-0.431
cffff-0.418
Tokenclone
Token ID21018
Feature activation+4.77e-01

Pos logit contributions
arching+0.903
toe+0.825
adia+0.816
hang+0.778
mat+0.775
Neg logit contributions
 condem-1.289
 newsp-1.033
terms-1.027
 constitu-1.010
cffff-0.979
Token (),
Token ID29994
Feature activation+4.90e+00

Pos logit contributions
arching+0.256
toe+0.226
adia+0.224
hang+0.208
estine+0.207
Neg logit contributions
 condem-0.558
 newsp-0.464
terms-0.461
 constitu-0.454
cffff-0.442
Token pieces
Token ID5207
Feature activation-1.93e+00

Pos logit contributions
arching+3.437
toe+3.162
adia+3.122
mat+3.009
hang+3.008
Neg logit contributions
 condem-4.771
terms-3.788
 newsp-3.753
 constitu-3.721
cffff-3.707
Token :
Token ID1058
Feature activation+6.50e-01

Pos logit contributions
 condem+2.280
 newsp+1.907
terms+1.896
 constitu+1.871
cffff+1.809
Neg logit contributions
arching-0.986
toe-0.861
adia-0.854
hang-0.787
zel-0.781
Token self
Token ID2116
Feature activation+3.93e-01

Pos logit contributions
arching+0.383
toe+0.343
adia+0.339
hang+0.319
estine+0.317
Neg logit contributions
 condem-0.725
 newsp-0.596
terms-0.592
 constitu-0.584
cffff-0.567
Token.
Token ID764
Feature activation+1.11e+00

Pos logit contributions
arching+0.261
toe+0.238
adia+0.235
hang+0.222
estine+0.220
Neg logit contributions
 condem-0.409
 newsp-0.331
terms-0.330
 constitu-0.324
cffff-0.314
Tokenpieces
Token ID34154
Feature activation-1.44e+00

Pos logit contributions
arching+0.769
toe+0.703
adia+0.693
hang+0.662
mat+0.660
Neg logit contributions
 condem-1.126
 newsp-0.906
terms-0.895
 constitu-0.886
cffff-0.858
Token United
Token ID1578
Feature activation-2.25e-01

Pos logit contributions
 condem+1.761
 newsp+1.409
terms+1.397
 constitu+1.370
cffff+1.320
Neg logit contributions
arching-1.299
toe-1.196
adia-1.177
hang-1.119
estine-1.119
Token Supporters
Token ID39995
Feature activation-8.20e-01

Pos logit contributions
 condem+0.238
 newsp+0.193
terms+0.192
 constitu+0.189
cffff+0.183
Neg logit contributions
arching-0.145
toe-0.131
adia-0.131
hang-0.123
estine-0.123
Token Trust
Token ID9870
Feature activation+2.85e-01

Pos logit contributions
 condem+0.969
 newsp+0.809
terms+0.809
 constitu+0.792
cffff+0.768
Neg logit contributions
arching-0.424
toe-0.373
adia-0.364
hang-0.338
estine-0.337
Token (
Token ID357
Feature activation-3.57e-01

Pos logit contributions
arching+0.163
toe+0.146
adia+0.144
hang+0.135
estine+0.134
Neg logit contributions
 condem-0.322
 newsp-0.266
terms-0.264
 constitu-0.261
cffff-0.253
TokenN
Token ID45
Feature activation-4.78e-01

Pos logit contributions
 condem+0.347
 newsp+0.276
terms+0.275
 constitu+0.269
cffff+0.260
Neg logit contributions
arching-0.261
toe-0.239
adia-0.237
hang-0.225
estine-0.225
TokenUST
Token ID7759
Feature activation-6.35e+00

Pos logit contributions
 condem+0.407
 newsp+0.313
terms+0.312
 constitu+0.304
cffff+0.289
Neg logit contributions
arching-0.405
toe-0.375
adia-0.371
hang-0.358
estine-0.356
Token),
Token ID828
Feature activation+1.21e+00

Pos logit contributions
 condem+6.369
 newsp+5.107
terms+5.104
 constitu+5.036
cffff+4.845
Neg logit contributions
arching-4.239
toe-3.874
adia-3.753
estine-3.590
mat-3.571
Token N
Token ID399
Feature activation-1.45e+00

Pos logit contributions
arching+0.867
toe+0.792
adia+0.789
hang+0.749
mat+0.745
Neg logit contributions
 condem-1.198
 newsp-0.957
terms-0.951
 constitu-0.936
cffff-0.906
TokenUFC
Token ID44534
Feature activation-4.40e+00

Pos logit contributions
 condem+1.291
 newsp+1.006
terms+1.004
 constitu+0.977
cffff+0.942
Neg logit contributions
arching-1.176
toe-1.090
adia-1.071
hang-1.032
estine-1.025
Token Fans
Token ID24964
Feature activation-4.41e-01

Pos logit contributions
 condem+4.638
 newsp+3.803
terms+3.801
 constitu+3.694
cffff+3.562
Neg logit contributions
arching-2.758
toe-2.531
adia-2.421
hang-2.308
zel-2.300
Token United
Token ID1578
Feature activation-4.92e-01

Pos logit contributions
 condem+0.472
 newsp+0.385
terms+0.384
 constitu+0.376
cffff+0.365
Neg logit contributions
arching-0.278
toe-0.251
adia-0.247
hang-0.232
estine-0.232
Token,
Token ID11
Feature activation-8.10e-01

Pos logit contributions
arching+0.284
toe+0.253
adia+0.250
hang+0.232
estine+0.231
Neg logit contributions
 condem-0.587
 newsp-0.487
terms-0.483
 constitu-0.477
cffff-0.462
Token it
Token ID340
Feature activation-3.55e-01

Pos logit contributions
 condem+0.850
 newsp+0.690
terms+0.685
 constitu+0.674
cffff+0.652
Neg logit contributions
arching-0.530
toe-0.481
adia-0.476
hang-0.449
estine-0.448
Token looks
Token ID3073
Feature activation-1.26e+00

Pos logit contributions
 condem+0.368
 newsp+0.298
terms+0.296
 constitu+0.291
cffff+0.281
Neg logit contributions
arching-0.236
toe-0.214
adia-0.212
hang-0.200
estine-0.200
Token well
Token ID880
Feature activation-2.72e+00

Pos logit contributions
 condem+1.277
 newsp+1.027
terms+1.015
 constitu+1.004
cffff+0.952
Neg logit contributions
arching-0.882
toe-0.796
adia-0.790
inion-0.751
estine-0.748
Token n
Token ID299
Feature activation-1.30e+00

Pos logit contributions
 condem+2.757
 newsp+2.217
terms+2.188
 constitu+2.154
cffff+2.041
Neg logit contributions
arching-1.868
adia-1.690
toe-1.680
estine-1.587
hang-1.584
Tokenigh
Token ID394
Feature activation-6.33e+00

Pos logit contributions
 condem+1.086
 newsp+0.829
terms+0.827
 constitu+0.805
cffff+0.771
Neg logit contributions
arching-1.119
toe-1.040
adia-1.032
hang-0.991
estine-0.983
Token insol
Token ID35831
Feature activation-1.64e+00

Pos logit contributions
 condem+6.239
 newsp+5.071
terms+4.950
 constitu+4.918
 recip+4.659
Neg logit contributions
arching-4.289
adia-4.020
toe-3.950
hang-3.721
inion-3.696
Tokenuble
Token ID26664
Feature activation-4.85e-01

Pos logit contributions
 condem+2.046
terms+1.744
 newsp+1.712
 constitu+1.692
cffff+1.668
Neg logit contributions
arching-0.705
toe-0.606
adia-0.598
estine-0.532
inion-0.531
Token,
Token ID11
Feature activation-3.61e-01

Pos logit contributions
 condem+0.547
 newsp+0.453
terms+0.449
 constitu+0.441
cffff+0.428
Neg logit contributions
arching-0.278
toe-0.250
adia-0.246
estine-0.231
hang-0.230
Token given
Token ID1813
Feature activation-4.29e-01

Pos logit contributions
 condem+0.386
 newsp+0.314
terms+0.312
 constitu+0.307
cffff+0.297
Neg logit contributions
arching-0.229
toe-0.207
adia-0.205
hang-0.193
estine-0.193
Token
Token ID198
Feature activation-4.59e-01

Pos logit contributions
 condem+0.457
 newsp+0.373
terms+0.370
 constitu+0.364
cffff+0.352
Neg logit contributions
arching-0.272
toe-0.248
adia-0.245
estine-0.229
hang-0.229
Token of
Token ID286
Feature activation+5.36e-01

Pos logit contributions
arching+0.728
toe+0.666
adia+0.662
hang+0.632
estine+0.627
Neg logit contributions
 condem-0.930
 newsp-0.738
terms-0.733
 constitu-0.719
cffff-0.697
Token the
Token ID262
Feature activation+1.78e-01

Pos logit contributions
arching+0.322
toe+0.291
adia+0.287
hang+0.270
estine+0.269
Neg logit contributions
 condem-0.589
 newsp-0.484
terms-0.481
 constitu-0.473
cffff-0.458
Token aims
Token ID12031
Feature activation-2.96e-01

Pos logit contributions
arching+0.118
toe+0.107
adia+0.106
hang+0.100
estine+0.100
Neg logit contributions
 condem-0.185
 newsp-0.150
terms-0.149
 constitu-0.147
cffff-0.142
Token of
Token ID286
Feature activation-2.82e-01

Pos logit contributions
 condem+0.277
 newsp+0.219
terms+0.217
 constitu+0.213
cffff+0.205
Neg logit contributions
arching-0.226
toe-0.209
adia-0.207
hang-0.197
estine-0.197
Token N
Token ID399
Feature activation-1.47e+00

Pos logit contributions
 condem+0.308
 newsp+0.252
terms+0.251
 constitu+0.247
cffff+0.239
Neg logit contributions
arching-0.172
toe-0.156
adia-0.153
hang-0.144
estine-0.144
TokenUST
Token ID7759
Feature activation-5.94e+00

Pos logit contributions
 condem+1.095
terms+0.804
 newsp+0.804
 constitu+0.779
cffff+0.739
Neg logit contributions
arching-1.404
toe-1.317
adia-1.297
hang-1.259
mat-1.253
Token is
Token ID318
Feature activation-8.55e-01

Pos logit contributions
 condem+6.002
 newsp+4.932
terms+4.898
 constitu+4.767
cffff+4.578
Neg logit contributions
arching-3.836
toe-3.630
adia-3.361
estine-3.308
hang-3.283
Token to
Token ID284
Feature activation+3.96e-01

Pos logit contributions
 condem+0.878
 newsp+0.710
terms+0.706
 constitu+0.695
cffff+0.670
Neg logit contributions
arching-0.577
toe-0.527
adia-0.517
hang-0.492
mat-0.487
Token develop
Token ID1205
Feature activation+5.85e-01

Pos logit contributions
arching+0.279
toe+0.255
adia+0.253
hang+0.240
estine+0.239
Neg logit contributions
 condem-0.395
 newsp-0.317
terms-0.315
 constitu-0.310
cffff-0.299
Token formal
Token ID8766
Feature activation+2.25e+00

Pos logit contributions
arching+0.373
toe+0.337
adia+0.334
hang+0.315
estine+0.314
Neg logit contributions
 condem-0.624
 newsp-0.508
terms-0.505
 constitu-0.497
cffff-0.481
Token representation
Token ID10552
Feature activation+1.89e+00

Pos logit contributions
arching+1.478
adia+1.348
toe+1.339
hang+1.262
mat+1.259
Neg logit contributions
 condem-2.350
 newsp-1.895
terms-1.885
 constitu-1.856
cffff-1.809
Token section
Token ID2665
Feature activation-1.42e+00

Pos logit contributions
 condem+0.345
 newsp+0.293
terms+0.293
 constitu+0.287
cffff+0.280
Neg logit contributions
arching-0.118
toe-0.103
adia-0.100
estine-0.091
mat-0.089
Token on
Token ID319
Feature activation+9.68e-01

Pos logit contributions
 condem+1.536
 newsp+1.261
terms+1.249
 constitu+1.224
cffff+1.182
Neg logit contributions
arching-0.884
toe-0.800
adia-0.785
estine-0.741
hang-0.741
Token www
Token ID7324
Feature activation+1.13e-01

Pos logit contributions
arching+0.590
toe+0.529
adia+0.525
hang+0.495
mat+0.491
Neg logit contributions
 condem-1.061
 newsp-0.867
terms-0.862
 constitu-0.850
cffff-0.826
Token.
Token ID13
Feature activation-4.14e-01

Pos logit contributions
arching+0.056
toe+0.049
adia+0.048
hang+0.044
estine+0.044
Neg logit contributions
 condem-0.137
 newsp-0.114
terms-0.114
 constitu-0.112
cffff-0.109
Tokenn
Token ID77
Feature activation-4.67e-01

Pos logit contributions
 condem+0.436
terms+0.355
 newsp+0.354
 constitu+0.346
cffff+0.337
Neg logit contributions
arching-0.267
toe-0.242
adia-0.238
estine-0.226
hang-0.225
Tokenuf
Token ID3046
Feature activation-5.90e+00

Pos logit contributions
 condem+0.483
terms+0.395
 newsp+0.387
 constitu+0.379
cffff+0.372
Neg logit contributions
arching-0.307
toe-0.278
adia-0.270
hang-0.259
estine-0.258
Tokenc
Token ID66
Feature activation-7.98e-01

Pos logit contributions
 condem+6.110
terms+5.096
 newsp+4.958
 constitu+4.900
cffff+4.801
Neg logit contributions
arching-3.629
toe-3.364
adia-3.197
estine-3.081
hang-3.047
Token.
Token ID13
Feature activation-5.77e-01

Pos logit contributions
 condem+0.913
terms+0.762
 newsp+0.753
 constitu+0.738
cffff+0.722
Neg logit contributions
arching-0.436
toe-0.389
adia-0.379
estine-0.358
hang-0.357
Tokenco
Token ID1073
Feature activation-1.04e-01

Pos logit contributions
 condem+0.539
terms+0.427
 newsp+0.423
 constitu+0.410
cffff+0.400
Neg logit contributions
arching-0.443
toe-0.405
adia-0.402
estine-0.384
hang-0.384
Token.
Token ID13
Feature activation-3.06e-02

Pos logit contributions
 condem+0.122
terms+0.101
 newsp+0.101
 constitu+0.099
cffff+0.096
Neg logit contributions
arching-0.054
toe-0.048
adia-0.047
hang-0.044
estine-0.044
Tokenuk
Token ID2724
Feature activation-5.17e-01

Pos logit contributions
 condem+0.033
terms+0.027
 newsp+0.027
 constitu+0.026
cffff+0.025
Neg logit contributions
arching-0.019
toe-0.017
adia-0.017
hang-0.016
estine-0.016
Token frequent
Token ID10792
Feature activation-1.56e+00

Pos logit contributions
 condem+0.108
 newsp+0.087
terms+0.087
 constitu+0.085
cffff+0.082
Neg logit contributions
arching-0.070
toe-0.064
adia-0.063
hang-0.060
estine-0.059
Token Dinosaur
Token ID47808
Feature activation-1.13e+00

Pos logit contributions
 condem+1.468
 newsp+1.160
terms+1.151
 constitu+1.130
cffff+1.089
Neg logit contributions
arching-1.185
toe-1.090
adia-1.081
hang-1.031
estine-1.028
Token Jr
Token ID7504
Feature activation+1.71e+00

Pos logit contributions
 condem+1.130
 newsp+0.908
terms+0.898
 constitu+0.882
cffff+0.851
Neg logit contributions
arching-0.802
toe-0.731
adia-0.723
hang-0.686
estine-0.685
Token.
Token ID13
Feature activation-9.27e-01

Pos logit contributions
arching+0.928
toe+0.826
adia+0.826
hang+0.769
mat+0.763
Neg logit contributions
 condem-1.965
terms-1.627
 newsp-1.621
 constitu-1.607
cffff-1.562
Token collaborator
Token ID50160
Feature activation-1.90e+00

Pos logit contributions
 condem+0.840
 newsp+0.654
terms+0.652
 constitu+0.640
cffff+0.616
Neg logit contributions
arching-0.738
toe-0.682
adia-0.678
hang-0.649
mat-0.645
Token Maur
Token ID18867
Feature activation-5.73e+00

Pos logit contributions
 condem+1.879
 newsp+1.510
terms+1.501
 constitu+1.470
cffff+1.411
Neg logit contributions
arching-1.358
toe-1.230
adia-1.219
hang-1.159
estine-1.158
Tokena
Token ID64
Feature activation+1.65e-01

Pos logit contributions
 condem+5.204
terms+4.196
 newsp+4.187
 constitu+4.051
cffff+3.893
Neg logit contributions
arching-4.250
toe-3.924
adia-3.772
hang-3.667
estine-3.627
Token Jasper
Token ID42992
Feature activation+1.93e+00

Pos logit contributions
arching+0.116
toe+0.106
adia+0.105
hang+0.100
estine+0.100
Neg logit contributions
 condem-0.164
 newsp-0.132
terms-0.131
 constitu-0.129
cffff-0.124
Token,
Token ID11
Feature activation+5.42e-01

Pos logit contributions
arching+1.095
toe+0.991
adia+0.981
mat+0.915
hang+0.914
Neg logit contributions
 condem-2.181
 newsp-1.799
terms-1.790
 constitu-1.778
cffff-1.718
Token an
Token ID281
Feature activation+3.96e-01

Pos logit contributions
arching+0.363
toe+0.330
adia+0.327
hang+0.310
mat+0.308
Neg logit contributions
 condem-0.560
 newsp-0.453
terms-0.450
 constitu-0.443
cffff-0.429
Token essay
Token ID14268
Feature activation-2.26e-01

Pos logit contributions
arching+0.264
toe+0.240
adia+0.237
hang+0.224
estine+0.223
Neg logit contributions
 condem-0.411
 newsp-0.333
terms-0.330
 constitu-0.325
cffff-0.315
Token Stone
Token ID8026
Feature activation+5.17e-01

Pos logit contributions
 condem+0.337
 newsp+0.268
terms+0.266
 constitu+0.261
cffff+0.252
Neg logit contributions
arching-0.254
toe-0.233
adia-0.231
hang-0.220
estine-0.219
Tokenhurst
Token ID33500
Feature activation-9.71e-01

Pos logit contributions
arching+0.371
toe+0.339
adia+0.336
hang+0.319
estine+0.319
Neg logit contributions
 condem-0.508
 newsp-0.406
terms-0.404
 constitu-0.396
cffff-0.383
Token estate
Token ID7964
Feature activation+2.23e-01

Pos logit contributions
 condem+0.941
 newsp+0.750
terms+0.744
 constitu+0.730
cffff+0.703
Neg logit contributions
arching-0.713
toe-0.653
adia-0.646
hang-0.615
estine-0.613
Token in
Token ID287
Feature activation-1.80e-01

Pos logit contributions
arching+0.135
toe+0.122
adia+0.120
hang+0.113
estine+0.113
Neg logit contributions
 condem-0.244
 newsp-0.200
terms-0.199
 constitu-0.196
cffff-0.190
Token W
Token ID370
Feature activation-4.43e-01

Pos logit contributions
 condem+0.170
 newsp+0.135
terms+0.134
 constitu+0.131
cffff+0.127
Neg logit contributions
arching-0.136
toe-0.125
adia-0.124
hang-0.118
mat-0.118
Tokenalth
Token ID1094
Feature activation-5.60e+00

Pos logit contributions
 condem+0.326
 newsp+0.239
terms+0.236
 constitu+0.230
cffff+0.218
Neg logit contributions
arching-0.428
toe-0.402
adia-0.399
hang-0.384
estine-0.384
Tokenam
Token ID321
Feature activation+7.11e-01

Pos logit contributions
 condem+5.722
 newsp+4.640
terms+4.587
 constitu+4.501
cffff+4.321
Neg logit contributions
arching-3.673
toe-3.363
adia-3.287
mat-3.121
estine-3.112
Token and
Token ID290
Feature activation-5.23e-01

Pos logit contributions
arching+0.427
toe+0.384
adia+0.380
hang+0.357
estine+0.356
Neg logit contributions
 condem-0.784
 newsp-0.644
terms-0.639
 constitu-0.631
cffff-0.613
Token Som
Token ID9995
Feature activation-1.43e+00

Pos logit contributions
 condem+0.542
 newsp+0.439
terms+0.435
 constitu+0.429
cffff+0.415
Neg logit contributions
arching-0.349
toe-0.318
adia-0.314
hang-0.297
estine-0.296
Tokenerville
Token ID33487
Feature activation-3.22e+00

Pos logit contributions
 condem+1.072
terms+0.800
 newsp+0.790
 constitu+0.759
cffff+0.730
Neg logit contributions
arching-1.337
toe-1.245
adia-1.229
hang-1.183
estine-1.179
Token's
Token ID338
Feature activation-2.44e-01

Pos logit contributions
 condem+3.518
 newsp+2.898
terms+2.871
 constitu+2.820
cffff+2.724
Neg logit contributions
arching-1.941
toe-1.766
adia-1.712
hang-1.619
estine-1.614
Token first
Token ID717
Feature activation-4.40e-01

Pos logit contributions
arching+0.151
toe+0.138
adia+0.136
hang+0.130
mat+0.129
Neg logit contributions
 condem-0.226
 newsp-0.181
terms-0.180
 constitu-0.177
cffff-0.172
Token Ghostbusters
Token ID50037
Feature activation-9.13e-02

Pos logit contributions
 condem+0.371
 newsp+0.286
terms+0.284
 constitu+0.277
cffff+0.265
Neg logit contributions
arching-0.378
toe-0.350
adia-0.348
hang-0.333
estine-0.333
Token,
Token ID11
Feature activation+2.20e-01

Pos logit contributions
 condem+0.107
 newsp+0.088
terms+0.088
 constitu+0.087
cffff+0.084
Neg logit contributions
arching-0.049
toe-0.044
adia-0.043
hang-0.040
estine-0.040
Token the
Token ID262
Feature activation-2.05e-01

Pos logit contributions
arching+0.129
toe+0.116
adia+0.115
hang+0.107
estine+0.107
Neg logit contributions
 condem-0.246
 newsp-0.202
terms-0.201
 constitu-0.198
cffff-0.192
Tokenne
Token ID710
Feature activation-2.81e-01

Pos logit contributions
 condem+0.187
 newsp+0.148
terms+0.146
 constitu+0.143
cffff+0.137
Neg logit contributions
arching-0.162
toe-0.148
adia-0.148
estine-0.140
hang-0.140
Tokenigh
Token ID394
Feature activation-5.58e+00

Pos logit contributions
 condem+0.244
 newsp+0.189
terms+0.187
 constitu+0.184
cffff+0.176
Neg logit contributions
arching-0.233
toe-0.216
adia-0.214
estine-0.204
hang-0.204
Tokenbor
Token ID2865
Feature activation-1.56e+00

Pos logit contributions
 condem+6.322
 newsp+5.235
terms+5.231
 constitu+5.158
cffff+5.006
Neg logit contributions
arching-2.979
toe-2.660
adia-2.624
estine-2.450
hang-2.422
Tokenhood
Token ID2894
Feature activation-6.40e-01

Pos logit contributions
 condem+1.593
 newsp+1.286
terms+1.273
 constitu+1.253
cffff+1.209
Neg logit contributions
arching-1.055
toe-0.962
adia-0.951
mat-0.900
estine-0.899
Token in
Token ID287
Feature activation+2.99e-02

Pos logit contributions
 condem+0.696
 newsp+0.568
terms+0.562
 constitu+0.559
cffff+0.535
Neg logit contributions
arching-0.392
toe-0.355
adia-0.351
estine-0.326
hang-0.325
Token New
Token ID968
Feature activation-1.42e-01

Pos logit contributions
arching+0.024
toe+0.022
adia+0.022
hang+0.021
estine+0.021
Neg logit contributions
 condem-0.027
 newsp-0.021
terms-0.021
 constitu-0.020
cffff-0.020
Token York
Token ID1971
Feature activation+1.32e+00

Pos logit contributions
 condem+0.149
 newsp+0.120
terms+0.119
 constitu+0.118
cffff+0.114
Neg logit contributions
arching-0.094
toe-0.086
adia-0.085
hang-0.080
estine-0.080
Token,
Token ID11
Feature activation+1.03e+00

Pos logit contributions
arching+1.164
toe+1.056
adia+1.052
hang+0.990
mat+0.987
Neg logit contributions
 condem-1.870
 newsp-1.506
terms-1.495
 constitu-1.478
cffff-1.438
Token Great
Token ID3878
Feature activation+1.14e+00

Pos logit contributions
arching+0.788
toe+0.722
adia+0.718
hang+0.689
mat+0.686
Neg logit contributions
 condem-0.960
 newsp-0.751
terms-0.749
 constitu-0.736
cffff-0.717
Token Depression
Token ID22483
Feature activation+3.47e-01

Pos logit contributions
arching+1.104
toe+1.033
adia+1.030
hang+0.993
mat+0.988
Neg logit contributions
 condem-0.836
 newsp-0.608
terms-0.600
 constitu-0.589
cffff-0.562
Token<|endoftext|>
Token ID50256
Feature activation-6.89e-01

Pos logit contributions
arching+0.195
toe+0.174
adia+0.172
hang+0.161
mat+0.160
Neg logit contributions
 condem-0.396
 newsp-0.327
terms-0.325
 constitu-0.321
cffff-0.312
TokenW
Token ID54
Feature activation+6.07e-01

Pos logit contributions
 condem+0.670
 newsp+0.533
terms+0.532
 constitu+0.521
cffff+0.503
Neg logit contributions
arching-0.503
toe-0.460
adia-0.457
hang-0.434
estine-0.433
TokenALTH
Token ID40818
Feature activation-5.57e+00

Pos logit contributions
arching+0.433
toe+0.396
adia+0.393
hang+0.373
mat+0.371
Neg logit contributions
 condem-0.602
 newsp-0.482
terms-0.479
 constitu-0.470
cffff-0.455
TokenAM
Token ID2390
Feature activation+8.71e-01

Pos logit contributions
 condem+4.816
 newsp+3.774
terms+3.696
 constitu+3.672
cffff+3.474
Neg logit contributions
arching-4.541
toe-4.198
adia-4.164
estine-3.975
mat-3.973
Token,
Token ID11
Feature activation-4.63e-01

Pos logit contributions
arching+0.487
toe+0.436
adia+0.431
hang+0.403
mat+0.401
Neg logit contributions
 condem-0.996
 newsp-0.823
terms-0.817
 constitu-0.805
cffff-0.785
Token Mass
Token ID5674
Feature activation-1.63e+00

Pos logit contributions
 condem+0.299
 newsp+0.207
terms+0.204
 constitu+0.198
cffff+0.186
Neg logit contributions
arching-0.491
toe-0.463
adia-0.460
hang-0.445
estine-0.444
Token.
Token ID13
Feature activation-1.23e+00

Pos logit contributions
 condem+1.887
 newsp+1.568
terms+1.559
 constitu+1.550
cffff+1.502
Neg logit contributions
arching-0.877
toe-0.767
adia-0.750
hang-0.706
estine-0.705
Token --
Token ID1377
Feature activation-1.52e+00

Pos logit contributions
 condem+1.339
 newsp+1.097
terms+1.093
 constitu+1.077
cffff+1.039
Neg logit contributions
arching-0.756
toe-0.674
adia-0.669
hang-0.629
estine-0.627
Token extra
Token ID3131
Feature activation-1.73e+00

Pos logit contributions
 condem+1.953
 newsp+1.558
terms+1.548
 constitu+1.517
cffff+1.465
Neg logit contributions
arching-1.452
toe-1.330
adia-1.319
hang-1.253
estine-1.250
Token for
Token ID329
Feature activation-6.89e-01

Pos logit contributions
 condem+1.674
 newsp+1.334
terms+1.326
 constitu+1.293
cffff+1.249
Neg logit contributions
arching-1.275
toe-1.169
adia-1.160
hang-1.105
estine-1.101
Token other
Token ID584
Feature activation+3.91e-01

Pos logit contributions
 condem+0.740
 newsp+0.605
terms+0.602
 constitu+0.590
cffff+0.572
Neg logit contributions
arching-0.433
toe-0.391
adia-0.386
hang-0.363
estine-0.363
Token rewards
Token ID11530
Feature activation-6.92e-01

Pos logit contributions
arching+0.316
toe+0.292
adia+0.291
hang+0.279
mat+0.278
Neg logit contributions
 condem-0.349
 newsp-0.271
terms-0.268
 constitu-0.266
cffff-0.256
Token on
Token ID319
Feature activation-2.24e-02

Pos logit contributions
 condem+0.765
 newsp+0.628
terms+0.625
 constitu+0.614
cffff+0.596
Neg logit contributions
arching-0.415
toe-0.373
adia-0.368
hang-0.346
estine-0.344
Token top
Token ID1353
Feature activation-5.50e+00

Pos logit contributions
 condem+0.024
 newsp+0.019
terms+0.019
 constitu+0.019
cffff+0.018
Neg logit contributions
arching-0.015
toe-0.013
adia-0.013
hang-0.012
estine-0.012
Token of
Token ID286
Feature activation+5.07e-01

Pos logit contributions
 condem+4.660
terms+3.674
 newsp+3.612
 constitu+3.496
cffff+3.385
Neg logit contributions
arching-4.510
toe-4.138
adia-4.091
inion-3.940
estine-3.922
Token a
Token ID257
Feature activation-1.71e-01

Pos logit contributions
arching+0.318
toe+0.287
adia+0.284
hang+0.268
estine+0.266
Neg logit contributions
 condem-0.546
 newsp-0.446
terms-0.443
 constitu-0.437
cffff-0.423
Token copy
Token ID4866
Feature activation+8.31e-02

Pos logit contributions
 condem+0.157
 newsp+0.123
terms+0.123
 constitu+0.120
cffff+0.116
Neg logit contributions
arching-0.134
toe-0.123
adia-0.122
hang-0.117
mat-0.116
Token of
Token ID286
Feature activation+2.75e-01

Pos logit contributions
arching+0.062
toe+0.057
adia+0.057
mat+0.054
hang+0.054
Neg logit contributions
 condem-0.079
 newsp-0.063
terms-0.062
 constitu-0.061
cffff-0.059
Token the
Token ID262
Feature activation+1.84e-01

Pos logit contributions
arching+0.172
toe+0.156
adia+0.154
hang+0.145
mat+0.145
Neg logit contributions
 condem-0.296
 newsp-0.242
terms-0.240
 constitu-0.237
cffff-0.230
Token O
Token ID440
Feature activation-6.18e-01

Pos logit contributions
 condem+0.753
 newsp+0.611
terms+0.606
 constitu+0.597
cffff+0.577
Neg logit contributions
arching-0.474
toe-0.430
adia-0.426
hang-0.402
estine-0.402
Tokenrec
Token ID8344
Feature activation+9.32e-01

Pos logit contributions
 condem+0.662
 newsp+0.538
terms+0.536
 constitu+0.527
cffff+0.510
Neg logit contributions
arching-0.391
toe-0.351
adia-0.347
hang-0.330
estine-0.328
Tokenas
Token ID292
Feature activation+2.47e-01

Pos logit contributions
arching+0.577
toe+0.520
adia+0.514
hang+0.487
mat+0.484
Neg logit contributions
 condem-1.013
 newsp-0.828
terms-0.823
 constitu-0.812
cffff-0.792
Token filling
Token ID12591
Feature activation-5.43e-01

Pos logit contributions
arching+0.169
toe+0.154
adia+0.153
hang+0.145
mat+0.144
Neg logit contributions
 condem-0.252
 newsp-0.203
terms-0.202
 constitu-0.198
cffff-0.193
Token the
Token ID262
Feature activation-2.37e-01

Pos logit contributions
 condem+0.594
 newsp+0.487
terms+0.484
 constitu+0.477
cffff+0.461
Neg logit contributions
arching-0.331
toe-0.298
adia-0.295
hang-0.277
estine-0.277
Token top
Token ID1353
Feature activation-5.44e+00

Pos logit contributions
 condem+0.226
 newsp+0.179
terms+0.178
 constitu+0.175
cffff+0.169
Neg logit contributions
arching-0.177
toe-0.163
adia-0.161
hang-0.154
estine-0.153
Token three
Token ID1115
Feature activation+1.49e-01

Pos logit contributions
 condem+5.153
 newsp+4.117
terms+4.113
 constitu+4.006
cffff+3.776
Neg logit contributions
arching-3.975
toe-3.648
adia-3.594
estine-3.434
hang-3.417
Token positions
Token ID6116
Feature activation-4.18e-01

Pos logit contributions
arching+0.099
toe+0.090
adia+0.089
hang+0.084
mat+0.084
Neg logit contributions
 condem-0.155
 newsp-0.125
terms-0.125
 constitu-0.123
cffff-0.119
Token in
Token ID287
Feature activation-7.77e-02

Pos logit contributions
 condem+0.460
 newsp+0.377
terms+0.375
 constitu+0.369
cffff+0.357
Neg logit contributions
arching-0.252
toe-0.227
adia-0.224
hang-0.211
estine-0.210
Token the
Token ID262
Feature activation-4.31e-01

Pos logit contributions
 condem+0.082
 newsp+0.066
terms+0.066
 constitu+0.065
cffff+0.063
Neg logit contributions
arching-0.051
toe-0.046
adia-0.046
hang-0.043
mat-0.043
Token first
Token ID717
Feature activation-1.31e+00

Pos logit contributions
 condem+0.427
 newsp+0.341
terms+0.339
 constitu+0.333
cffff+0.323
Neg logit contributions
arching-0.307
toe-0.281
adia-0.278
hang-0.264
mat-0.263
Token-
Token ID12
Feature activation-2.04e-02

Pos logit contributions
arching+0.142
toe+0.130
adia+0.129
hang+0.123
mat+0.122
Neg logit contributions
 condem-0.195
 newsp-0.156
terms-0.155
 constitu-0.152
cffff-0.147
Tokenthird
Token ID17089
Feature activation-5.86e-01

Pos logit contributions
 condem+0.022
 newsp+0.018
terms+0.018
 constitu+0.017
cffff+0.017
Neg logit contributions
arching-0.013
toe-0.012
adia-0.012
hang-0.011
estine-0.011
Token benefit
Token ID4414
Feature activation-4.88e-01

Pos logit contributions
 condem+0.577
 newsp+0.461
terms+0.457
 constitu+0.450
cffff+0.434
Neg logit contributions
arching-0.422
toe-0.386
adia-0.383
hang-0.364
estine-0.363
Token cost
Token ID1575
Feature activation-2.55e-01

Pos logit contributions
 condem+0.513
 newsp+0.416
terms+0.414
 constitu+0.407
cffff+0.394
Neg logit contributions
arching-0.319
toe-0.289
adia-0.286
hang-0.270
estine-0.269
Token on
Token ID319
Feature activation+5.72e-02

Pos logit contributions
 condem+0.272
 newsp+0.222
terms+0.220
 constitu+0.216
cffff+0.210
Neg logit contributions
arching-0.163
toe-0.147
adia-0.146
hang-0.138
estine-0.138
Token top
Token ID1353
Feature activation-5.43e+00

Pos logit contributions
arching+0.038
toe+0.035
adia+0.034
hang+0.033
mat+0.032
Neg logit contributions
 condem-0.059
 newsp-0.048
terms-0.048
 constitu-0.047
cffff-0.045
Token of
Token ID286
Feature activation+3.15e-01

Pos logit contributions
 condem+4.655
terms+3.683
 newsp+3.645
 constitu+3.569
cffff+3.385
Neg logit contributions
arching-4.348
adia-3.999
toe-3.998
estine-3.852
inion-3.813
Token the
Token ID262
Feature activation-3.14e-02

Pos logit contributions
arching+0.204
toe+0.185
adia+0.183
hang+0.173
mat+0.172
Neg logit contributions
 condem-0.333
 newsp-0.270
terms-0.269
 constitu-0.265
cffff-0.257
Token salary
Token ID9588
Feature activation-7.84e-01

Pos logit contributions
 condem+0.030
 newsp+0.024
terms+0.024
 constitu+0.023
cffff+0.023
Neg logit contributions
arching-0.023
toe-0.021
adia-0.021
hang-0.020
mat-0.020
Token for
Token ID329
Feature activation+4.24e-02

Pos logit contributions
 condem+0.815
 newsp+0.661
terms+0.656
 constitu+0.645
cffff+0.623
Neg logit contributions
arching-0.521
toe-0.474
adia-0.469
hang-0.443
estine-0.442
Token a
Token ID257
Feature activation-1.77e-01

Pos logit contributions
arching+0.027
toe+0.024
adia+0.024
hang+0.023
mat+0.022
Neg logit contributions
 condem-0.046
 newsp-0.037
terms-0.037
 constitu-0.036
cffff-0.035
Token Stone
Token ID8026
Feature activation+6.01e-01

Pos logit contributions
 condem+0.750
 newsp+0.636
terms+0.634
 constitu+0.625
cffff+0.607
Neg logit contributions
arching-0.250
toe-0.212
adia-0.210
hang-0.189
mat-0.187
Tokenhurst
Token ID33500
Feature activation-1.06e+00

Pos logit contributions
arching+0.288
toe+0.254
adia+0.249
hang+0.232
mat+0.229
Neg logit contributions
 condem-0.745
 newsp-0.627
terms-0.613
 constitu-0.613
cffff-0.591
Token estate
Token ID7964
Feature activation+5.85e-02

Pos logit contributions
 condem+0.736
 newsp+0.527
terms+0.521
 constitu+0.507
cffff+0.479
Neg logit contributions
arching-1.062
toe-0.998
adia-0.992
hang-0.958
estine-0.955
Token in
Token ID287
Feature activation-4.02e-01

Pos logit contributions
arching+0.033
toe+0.029
adia+0.029
hang+0.027
estine+0.027
Neg logit contributions
 condem-0.067
 newsp-0.055
terms-0.055
 constitu-0.054
cffff-0.053
Token W
Token ID370
Feature activation-5.05e-01

Pos logit contributions
 condem+0.432
 newsp+0.354
terms+0.352
 constitu+0.346
cffff+0.333
Neg logit contributions
arching-0.251
toe-0.227
adia-0.225
hang-0.210
estine-0.209
Tokenalth
Token ID1094
Feature activation-5.36e+00

Pos logit contributions
 condem+0.311
 newsp+0.209
terms+0.206
 constitu+0.200
cffff+0.186
Neg logit contributions
arching-0.554
toe-0.522
adia-0.520
hang-0.503
mat-0.502
Tokenam
Token ID321
Feature activation+6.57e-01

Pos logit contributions
 condem+4.988
terms+4.225
 newsp+4.024
 constitu+4.012
cffff+3.894
Neg logit contributions
arching-3.685
toe-3.329
adia-3.239
estine-3.119
zel-3.082
Token and
Token ID290
Feature activation-7.11e-01

Pos logit contributions
arching+0.348
toe+0.309
adia+0.306
hang+0.285
mat+0.284
Neg logit contributions
 condem-0.778
 newsp-0.646
terms-0.636
 constitu-0.632
cffff-0.614
Token Som
Token ID9995
Feature activation-1.36e+00

Pos logit contributions
 condem+0.654
 newsp+0.513
terms+0.508
 constitu+0.499
cffff+0.481
Neg logit contributions
arching-0.558
toe-0.515
adia-0.511
hang-0.488
estine-0.487
Tokenerville
Token ID33487
Feature activation-3.21e+00

Pos logit contributions
 condem+1.027
 newsp+0.757
terms+0.754
 constitu+0.732
cffff+0.700
Neg logit contributions
arching-1.291
toe-1.208
adia-1.198
hang-1.155
estine-1.150
Token's
Token ID338
Feature activation-4.32e-01

Pos logit contributions
 condem+3.283
terms+2.789
 newsp+2.693
 constitu+2.656
soType+2.573
Neg logit contributions
arching-2.033
toe-1.792
adia-1.789
hang-1.690
estine-1.682
Token and
Token ID290
Feature activation+2.44e-01

Pos logit contributions
 condem+1.336
 newsp+1.081
terms+1.073
 constitu+1.056
cffff+1.019
Neg logit contributions
arching-0.860
toe-0.782
adia-0.773
hang-0.732
estine-0.730
Token 360
Token ID11470
Feature activation-7.21e-01

Pos logit contributions
arching+0.165
toe+0.150
adia+0.148
hang+0.140
mat+0.140
Neg logit contributions
 condem-0.252
 newsp-0.203
terms-0.202
 constitu-0.199
cffff-0.193
Token Racing
Token ID14954
Feature activation-7.97e-01

Pos logit contributions
 condem+0.728
 newsp+0.585
terms+0.581
 constitu+0.572
cffff+0.553
Neg logit contributions
arching-0.501
toe-0.457
adia-0.452
hang-0.429
estine-0.428
Token occupied
Token ID12030
Feature activation+7.16e-01

Pos logit contributions
 condem+0.803
 newsp+0.645
terms+0.641
 constitu+0.630
cffff+0.609
Neg logit contributions
arching-0.555
toe-0.506
adia-0.501
hang-0.476
estine-0.474
Token the
Token ID262
Feature activation-5.91e-01

Pos logit contributions
arching+0.533
toe+0.488
adia+0.485
hang+0.461
estine+0.460
Neg logit contributions
 condem-0.688
 newsp-0.545
terms-0.542
 constitu-0.532
cffff-0.515
Token top
Token ID1353
Feature activation-5.31e+00

Pos logit contributions
 condem+0.568
 newsp+0.451
terms+0.447
 constitu+0.440
cffff+0.423
Neg logit contributions
arching-0.439
toe-0.403
adia-0.399
hang-0.380
estine-0.379
Token two
Token ID734
Feature activation+9.57e-02

Pos logit contributions
 condem+4.924
 newsp+3.915
terms+3.910
 constitu+3.832
 recip+3.557
Neg logit contributions
arching-3.994
toe-3.687
adia-3.610
estine-3.465
hang-3.459
Token positions
Token ID6116
Feature activation-2.16e-01

Pos logit contributions
arching+0.071
toe+0.065
adia+0.065
hang+0.061
mat+0.061
Neg logit contributions
 condem-0.092
 newsp-0.073
terms-0.073
 constitu-0.071
cffff-0.069
Token in
Token ID287
Feature activation-1.23e-01

Pos logit contributions
 condem+0.234
 newsp+0.191
terms+0.190
 constitu+0.187
cffff+0.181
Neg logit contributions
arching-0.134
toe-0.121
adia-0.119
hang-0.113
estine-0.112
Token the
Token ID262
Feature activation-2.92e-01

Pos logit contributions
 condem+0.132
 newsp+0.108
terms+0.107
 constitu+0.105
cffff+0.102
Neg logit contributions
arching-0.077
toe-0.069
adia-0.069
hang-0.065
mat-0.064
Token L
Token ID406
Feature activation-8.47e-01

Pos logit contributions
 condem+0.270
 newsp+0.212
terms+0.211
 constitu+0.207
cffff+0.200
Neg logit contributions
arching-0.227
toe-0.209
adia-0.207
hang-0.198
mat-0.197
Token p
Token ID279
Feature activation+8.02e-02

Pos logit contributions
arching+0.225
toe+0.205
adia+0.203
hang+0.192
mat+0.191
Neg logit contributions
 condem-0.365
 newsp-0.297
terms-0.295
 constitu-0.290
cffff-0.282
Tokenales
Token ID2040
Feature activation-1.44e+00

Pos logit contributions
arching+0.052
toe+0.047
adia+0.046
hang+0.044
mat+0.044
Neg logit contributions
 condem-0.085
 newsp-0.069
terms-0.069
 constitu-0.068
cffff-0.065
Token in
Token ID287
Feature activation-2.20e-01

Pos logit contributions
 condem+1.588
 newsp+1.302
terms+1.294
 constitu+1.274
cffff+1.236
Neg logit contributions
arching-0.871
toe-0.781
adia-0.773
hang-0.727
estine-0.724
Token com
Token ID401
Feature activation-8.49e-02

Pos logit contributions
 condem+0.218
 newsp+0.175
terms+0.174
 constitu+0.170
cffff+0.166
Neg logit contributions
arching-0.156
toe-0.143
adia-0.141
mat-0.135
hang-0.135
Token­
Token ID3907
Feature activation-4.18e+00

Pos logit contributions
 condem+0.074
 newsp+0.057
terms+0.056
 constitu+0.055
cffff+0.053
Neg logit contributions
arching-0.071
toe-0.066
adia-0.065
mat-0.063
hang-0.063
Tokenpar
Token ID1845
Feature activation-5.25e+00

Pos logit contributions
 condem+4.530
terms+3.773
 newsp+3.652
 constitu+3.645
cffff+3.547
Neg logit contributions
arching-2.473
adia-2.198
toe-2.181
hang-2.039
zel-2.030
Token­
Token ID3907
Feature activation-3.57e+00

Pos logit contributions
 condem+4.126
terms+3.172
 newsp+3.076
 constitu+3.033
cffff+2.892
Neg logit contributions
arching-4.703
toe-4.351
adia-4.263
estine-4.147
hang-4.088
Tokenis
Token ID271
Feature activation-1.03e+00

Pos logit contributions
 condem+3.675
terms+3.105
 constitu+2.937
 newsp+2.904
cffff+2.881
Neg logit contributions
arching-2.289
adia-1.958
toe-1.943
estine-1.848
 Reese-1.848
Token­
Token ID3907
Feature activation-3.15e+00

Pos logit contributions
 condem+0.870
 newsp+0.665
terms+0.663
 constitu+0.648
cffff+0.620
Neg logit contributions
arching-0.883
toe-0.818
adia-0.812
estine-0.778
hang-0.777
Tokenon
Token ID261
Feature activation-8.97e-01

Pos logit contributions
 condem+3.043
terms+2.521
 newsp+2.389
 constitu+2.377
cffff+2.298
Neg logit contributions
arching-2.272
adia-1.993
toe-1.982
estine-1.940
inion-1.888
Token to
Token ID284
Feature activation-2.69e-01

Pos logit contributions
 condem+0.938
 newsp+0.760
terms+0.755
 constitu+0.743
cffff+0.722
Neg logit contributions
arching-0.589
toe-0.535
adia-0.530
mat-0.501
hang-0.501
Token would
Token ID561
Feature activation-6.92e-01

Pos logit contributions
 condem+0.620
 newsp+0.504
terms+0.502
 constitu+0.493
cffff+0.479
Neg logit contributions
arching-0.377
toe-0.340
adia-0.337
estine-0.317
hang-0.316
Token probably
Token ID2192
Feature activation+1.03e+00

Pos logit contributions
 condem+0.670
 newsp+0.531
terms+0.530
 constitu+0.521
cffff+0.502
Neg logit contributions
arching-0.509
toe-0.467
adia-0.463
estine-0.440
hang-0.438
Token be
Token ID307
Feature activation-5.97e-01

Pos logit contributions
arching+0.738
toe+0.676
adia+0.670
hang+0.640
mat+0.637
Neg logit contributions
 condem-1.007
 newsp-0.808
terms-0.800
 constitu-0.786
cffff-0.760
Token on
Token ID319
Feature activation+1.27e-01

Pos logit contributions
 condem+0.632
 newsp+0.514
terms+0.510
 constitu+0.502
cffff+0.486
Neg logit contributions
arching-0.386
toe-0.349
adia-0.345
hang-0.326
estine-0.325
Token your
Token ID534
Feature activation-3.97e-01

Pos logit contributions
arching+0.076
toe+0.069
adia+0.068
hang+0.064
mat+0.064
Neg logit contributions
 condem-0.140
 newsp-0.114
terms-0.114
 constitu-0.112
cffff-0.109
Token top
Token ID1353
Feature activation-5.09e+00

Pos logit contributions
 condem+0.419
 newsp+0.340
terms+0.339
 constitu+0.333
cffff+0.323
Neg logit contributions
arching-0.257
toe-0.233
adia-0.230
hang-0.218
mat-0.217
Token 100
Token ID1802
Feature activation+8.36e-01

Pos logit contributions
 condem+5.031
 newsp+4.027
terms+4.017
 constitu+3.962
cffff+3.801
Neg logit contributions
arching-3.574
toe-3.260
adia-3.233
hang-3.054
mat-3.053
Token list
Token ID1351
Feature activation+1.99e+00

Pos logit contributions
arching+0.519
toe+0.468
adia+0.466
hang+0.437
mat+0.436
Neg logit contributions
 condem-0.904
 newsp-0.736
terms-0.733
 constitu-0.721
cffff-0.700
Token too
Token ID1165
Feature activation+1.01e+00

Pos logit contributions
arching+1.305
adia+1.191
toe+1.190
mat+1.126
hang+1.121
Neg logit contributions
 condem-2.076
 newsp-1.670
terms-1.652
 constitu-1.631
cffff-1.579
Token,
Token ID11
Feature activation-3.61e-01

Pos logit contributions
arching+0.584
toe+0.522
adia+0.517
hang+0.484
estine+0.482
Neg logit contributions
 condem-1.140
 newsp-0.939
terms-0.933
 constitu-0.920
cffff-0.893
Token the
Token ID262
Feature activation-6.77e-01

Pos logit contributions
 condem+0.397
 newsp+0.326
terms+0.324
 constitu+0.319
cffff+0.309
Neg logit contributions
arching-0.217
toe-0.195
adia-0.193
hang-0.181
estine-0.181
Token engine
Token ID3113
Feature activation-1.39e+00

Pos logit contributions
 condem+1.894
 newsp+1.539
terms+1.521
 constitu+1.500
cffff+1.445
Neg logit contributions
arching-1.097
toe-0.985
adia-0.975
estine-0.925
hang-0.921
Token,
Token ID11
Feature activation+2.12e-01

Pos logit contributions
 condem+1.548
 newsp+1.266
terms+1.257
 constitu+1.238
cffff+1.194
Neg logit contributions
arching-0.821
toe-0.733
adia-0.725
estine-0.683
hang-0.681
Token that
Token ID326
Feature activation+4.49e-01

Pos logit contributions
arching+0.137
toe+0.124
adia+0.123
hang+0.116
mat+0.115
Neg logit contributions
 condem-0.224
 newsp-0.183
terms-0.181
 constitu-0.179
cffff-0.173
Token reaches
Token ID12229
Feature activation-1.93e+00

Pos logit contributions
arching+0.291
toe+0.264
adia+0.261
hang+0.247
mat+0.246
Neg logit contributions
 condem-0.474
 newsp-0.385
terms-0.383
 constitu-0.377
cffff-0.366
Token a
Token ID257
Feature activation-9.06e-01

Pos logit contributions
 condem+1.921
 newsp+1.540
terms+1.525
 constitu+1.505
cffff+1.442
Neg logit contributions
arching-1.365
toe-1.243
adia-1.235
estine-1.176
mat-1.168
Token top
Token ID1353
Feature activation-5.05e+00

Pos logit contributions
 condem+0.871
 newsp+0.692
terms+0.686
 constitu+0.675
cffff+0.649
Neg logit contributions
arching-0.672
toe-0.616
adia-0.613
estine-0.582
hang-0.581
Token speed
Token ID2866
Feature activation-9.44e-01

Pos logit contributions
 condem+4.427
terms+3.409
 newsp+3.403
 constitu+3.359
cffff+3.161
Neg logit contributions
arching-4.089
adia-3.796
toe-3.765
estine-3.639
mat-3.577
Token of
Token ID286
Feature activation+3.82e-01

Pos logit contributions
 condem+0.869
 newsp+0.679
terms+0.676
 constitu+0.663
cffff+0.638
Neg logit contributions
arching-0.740
toe-0.680
adia-0.675
estine-0.645
hang-0.643
Token 345
Token ID39937
Feature activation-1.68e+00

Pos logit contributions
arching+0.285
toe+0.262
adia+0.259
hang+0.247
mat+0.246
Neg logit contributions
 condem-0.366
 newsp-0.290
terms-0.289
 constitu-0.283
cffff-0.274
Tokenkm
Token ID13276
Feature activation-6.91e-01

Pos logit contributions
 condem+1.408
 newsp+1.075
terms+1.067
 constitu+1.040
cffff+0.999
Neg logit contributions
arching-1.437
adia-1.324
toe-1.322
estine-1.289
hang-1.271
Token/
Token ID14
Feature activation-1.91e+00

Pos logit contributions
 condem+0.745
 newsp+0.608
terms+0.604
 constitu+0.595
cffff+0.577
Neg logit contributions
arching-0.433
toe-0.391
adia-0.387
hang-0.365
estine-0.363
Token and
Token ID290
Feature activation+1.06e+00

Pos logit contributions
arching+0.488
toe+0.440
adia+0.436
hang+0.410
mat+0.408
Neg logit contributions
 condem-0.860
 newsp-0.706
terms-0.700
 constitu-0.691
cffff-0.670
Token put
Token ID1234
Feature activation+1.28e-01

Pos logit contributions
arching+0.769
toe+0.710
adia+0.702
hang+0.672
mat+0.665
Neg logit contributions
 condem-1.029
 newsp-0.822
terms-0.820
 constitu-0.803
cffff-0.779
Token it
Token ID340
Feature activation+2.56e-01

Pos logit contributions
arching+0.074
toe+0.066
adia+0.066
hang+0.061
estine+0.061
Neg logit contributions
 condem-0.144
 newsp-0.119
terms-0.118
 constitu-0.116
cffff-0.113
Token on
Token ID319
Feature activation+5.71e-01

Pos logit contributions
arching+0.161
toe+0.146
adia+0.144
hang+0.136
estine+0.136
Neg logit contributions
 condem-0.275
 newsp-0.225
terms-0.223
 constitu-0.220
cffff-0.213
Token the
Token ID262
Feature activation-2.46e-01

Pos logit contributions
arching+0.340
toe+0.305
adia+0.301
hang+0.285
estine+0.282
Neg logit contributions
 condem-0.632
 newsp-0.518
terms-0.518
 constitu-0.509
cffff-0.495
Token top
Token ID1353
Feature activation-5.05e+00

Pos logit contributions
 condem+0.255
 newsp+0.206
terms+0.205
 constitu+0.202
cffff+0.196
Neg logit contributions
arching-0.164
toe-0.150
adia-0.148
hang-0.140
estine-0.140
Token.
Token ID13
Feature activation+7.38e-02

Pos logit contributions
 condem+4.694
 newsp+3.728
terms+3.694
 constitu+3.607
cffff+3.474
Neg logit contributions
arching-3.791
toe-3.460
adia-3.440
hang-3.271
estine-3.266
Token
Token ID198
Feature activation+1.09e-01

Pos logit contributions
arching+0.045
toe+0.041
adia+0.040
hang+0.038
mat+0.038
Neg logit contributions
 condem-0.081
 newsp-0.066
terms-0.065
 constitu-0.065
cffff-0.063
Token
Token ID198
Feature activation-1.52e-01

Pos logit contributions
arching+0.062
toe+0.055
adia+0.054
hang+0.051
estine+0.051
Neg logit contributions
 condem-0.123
terms-0.102
 newsp-0.101
 constitu-0.100
cffff-0.097
TokenHow
Token ID2437
Feature activation-1.02e-01

Pos logit contributions
 condem+0.148
 newsp+0.118
terms+0.116
 constitu+0.115
cffff+0.111
Neg logit contributions
arching-0.111
toe-0.102
adia-0.101
hang-0.096
mat-0.096
Token will
Token ID481
Feature activation-5.56e-01

Pos logit contributions
 condem+0.111
 newsp+0.091
terms+0.090
 constitu+0.089
cffff+0.086
Neg logit contributions
arching-0.062
toe-0.056
adia-0.056
hang-0.052
mat-0.052
Token chased
Token ID26172
Feature activation+1.52e-01

Pos logit contributions
arching+0.682
toe+0.628
adia+0.620
hang+0.591
mat+0.590
Neg logit contributions
 condem-0.903
 newsp-0.719
terms-0.716
 constitu-0.704
cffff-0.681
Token by
Token ID416
Feature activation-4.01e-02

Pos logit contributions
arching+0.104
toe+0.095
adia+0.094
hang+0.089
mat+0.088
Neg logit contributions
 condem-0.155
 newsp-0.125
terms-0.124
 constitu-0.122
cffff-0.118
Token Satan
Token ID14918
Feature activation+9.29e-01

Pos logit contributions
 condem+0.041
 newsp+0.033
terms+0.033
 constitu+0.032
cffff+0.031
Neg logit contributions
arching-0.027
toe-0.025
adia-0.024
hang-0.023
mat-0.023
Token.
Token ID13
Feature activation+3.06e-01

Pos logit contributions
arching+0.540
toe+0.484
adia+0.481
hang+0.448
mat+0.447
Neg logit contributions
 condem-1.042
 newsp-0.858
terms-0.853
 constitu-0.842
cffff-0.817
Token On
Token ID1550
Feature activation-1.75e-01

Pos logit contributions
arching+0.199
toe+0.182
adia+0.180
hang+0.170
mat+0.170
Neg logit contributions
 condem-0.324
 newsp-0.262
terms-0.259
 constitu-0.256
cffff-0.250
Token top
Token ID1353
Feature activation-5.05e+00

Pos logit contributions
 condem+0.192
 newsp+0.157
terms+0.156
 constitu+0.154
cffff+0.149
Neg logit contributions
arching-0.107
toe-0.097
adia-0.095
hang-0.090
mat-0.089
Token of
Token ID286
Feature activation+4.46e-01

Pos logit contributions
 condem+4.329
terms+3.413
 newsp+3.370
 constitu+3.299
cffff+3.147
Neg logit contributions
arching-4.146
toe-3.769
adia-3.748
estine-3.602
inion-3.598
Token that
Token ID326
Feature activation+8.18e-01

Pos logit contributions
arching+0.285
toe+0.259
adia+0.256
hang+0.242
mat+0.241
Neg logit contributions
 condem-0.474
 newsp-0.386
terms-0.383
 constitu-0.377
cffff-0.366
Token,
Token ID11
Feature activation-3.31e-01

Pos logit contributions
arching+0.450
toe+0.400
adia+0.395
hang+0.369
estine+0.367
Neg logit contributions
 condem-0.944
 newsp-0.783
terms-0.778
 constitu-0.767
cffff-0.745
Token I
Token ID314
Feature activation-2.62e-01

Pos logit contributions
 condem+0.350
 newsp+0.284
terms+0.282
 constitu+0.278
cffff+0.269
Neg logit contributions
arching-0.214
toe-0.194
adia-0.192
hang-0.182
mat-0.181
Token got
Token ID1392
Feature activation-3.63e-02

Pos logit contributions
 condem+0.274
 newsp+0.222
terms+0.221
 constitu+0.216
cffff+0.210
Neg logit contributions
arching-0.172
toe-0.156
adia-0.154
estine-0.146
hang-0.145
Token be
Token ID307
Feature activation-1.15e-02

Pos logit contributions
 condem+0.292
 newsp+0.233
terms+0.232
 constitu+0.227
cffff+0.220
Neg logit contributions
arching-0.216
toe-0.198
adia-0.196
hang-0.186
estine-0.186
Token made
Token ID925
Feature activation+4.21e-01

Pos logit contributions
 condem+0.012
 newsp+0.010
terms+0.010
 constitu+0.010
cffff+0.009
Neg logit contributions
arching-0.007
toe-0.007
adia-0.007
hang-0.006
estine-0.006
Token to
Token ID284
Feature activation+1.74e-01

Pos logit contributions
arching+0.287
toe+0.261
adia+0.259
hang+0.245
mat+0.244
Neg logit contributions
 condem-0.432
 newsp-0.348
terms-0.346
 constitu-0.341
cffff-0.329
Token run
Token ID1057
Feature activation-1.59e+00

Pos logit contributions
arching+0.120
toe+0.111
adia+0.109
hang+0.104
mat+0.104
Neg logit contributions
 condem-0.176
 newsp-0.140
terms-0.139
 constitu-0.139
cffff-0.134
Token on
Token ID319
Feature activation-8.11e-01

Pos logit contributions
 condem+1.633
 newsp+1.324
terms+1.314
 constitu+1.288
cffff+1.249
Neg logit contributions
arching-1.067
toe-0.972
adia-0.960
hang-0.909
estine-0.908
Token top
Token ID1353
Feature activation-5.01e+00

Pos logit contributions
 condem+0.832
 newsp+0.672
terms+0.667
 constitu+0.656
cffff+0.635
Neg logit contributions
arching-0.550
toe-0.500
adia-0.495
hang-0.469
estine-0.467
Token of
Token ID286
Feature activation+4.87e-01

Pos logit contributions
 condem+4.334
terms+3.405
 newsp+3.378
 constitu+3.284
cffff+3.177
Neg logit contributions
arching-4.036
toe-3.747
adia-3.683
estine-3.576
inion-3.559
Token it
Token ID340
Feature activation+4.19e-01

Pos logit contributions
arching+0.318
toe+0.288
adia+0.286
hang+0.270
estine+0.268
Neg logit contributions
 condem-0.513
 newsp-0.416
terms-0.414
 constitu-0.408
cffff-0.394
Token soon
Token ID2582
Feature activation+9.87e-03

Pos logit contributions
arching+0.237
toe+0.211
adia+0.208
hang+0.195
estine+0.194
Neg logit contributions
 condem-0.476
 newsp-0.394
terms-0.391
 constitu-0.385
cffff-0.375
Token,
Token ID11
Feature activation+3.02e-02

Pos logit contributions
arching+0.006
toe+0.005
adia+0.005
hang+0.005
estine+0.005
Neg logit contributions
 condem-0.011
 newsp-0.009
terms-0.009
 constitu-0.009
cffff-0.009
Token but
Token ID475
Feature activation+5.99e-02

Pos logit contributions
arching+0.019
toe+0.017
adia+0.017
hang+0.016
estine+0.016
Neg logit contributions
 condem-0.033
 newsp-0.027
terms-0.027
 constitu-0.026
cffff-0.025
Token
Token ID247
Feature activation-9.27e-01

Pos logit contributions
 condem+0.804
 newsp+0.620
terms+0.619
 constitu+0.602
cffff+0.578
Neg logit contributions
arching-0.812
toe-0.751
adia-0.739
estine-0.711
mat-0.710
Tokens
Token ID82
Feature activation-5.02e-01

Pos logit contributions
 condem+0.854
terms+0.677
 newsp+0.672
 constitu+0.658
cffff+0.633
Neg logit contributions
arching-0.719
toe-0.664
adia-0.658
hang-0.624
estine-0.623
Token largest
Token ID4387
Feature activation+1.39e-01

Pos logit contributions
 condem+0.516
 newsp+0.417
terms+0.414
 constitu+0.407
cffff+0.393
Neg logit contributions
arching-0.340
toe-0.309
adia-0.306
hang-0.290
estine-0.289
Token owners
Token ID4393
Feature activation-1.51e+00

Pos logit contributions
arching+0.105
toe+0.097
adia+0.096
hang+0.092
mat+0.091
Neg logit contributions
 condem-0.130
 newsp-0.103
terms-0.103
 constitu-0.101
cffff-0.097
Token of
Token ID286
Feature activation-7.89e-01

Pos logit contributions
 condem+1.607
 newsp+1.306
terms+1.303
 constitu+1.276
cffff+1.241
Neg logit contributions
arching-0.976
toe-0.878
adia-0.868
hang-0.820
estine-0.819
Token top
Token ID1353
Feature activation-5.01e+00

Pos logit contributions
 condem+0.771
 newsp+0.616
terms+0.612
 constitu+0.601
cffff+0.581
Neg logit contributions
arching-0.573
toe-0.525
adia-0.521
hang-0.495
mat-0.493
Token class
Token ID1398
Feature activation+1.82e+00

Pos logit contributions
 condem+4.754
 newsp+3.747
terms+3.742
 constitu+3.654
cffff+3.501
Neg logit contributions
arching-3.768
toe-3.429
adia-3.359
estine-3.274
mat-3.223
Token office
Token ID2607
Feature activation+1.58e+00

Pos logit contributions
arching+1.310
adia+1.223
toe+1.209
hang+1.156
mat+1.147
Neg logit contributions
 condem-1.743
terms-1.394
 newsp-1.389
 constitu-1.360
cffff-1.332
Token buildings
Token ID6832
Feature activation-6.16e-01

Pos logit contributions
arching+1.138
adia+1.055
toe+1.048
mat+1.003
hang+0.993
Neg logit contributions
 condem-1.519
 newsp-1.214
terms-1.209
 constitu-1.193
cffff-1.160
Token.
Token ID13
Feature activation+3.24e-01

Pos logit contributions
 condem+0.715
 newsp+0.592
terms+0.590
 constitu+0.580
cffff+0.565
Neg logit contributions
arching-0.336
toe-0.298
adia-0.293
hang-0.274
estine-0.273
Token Trump
Token ID1301
Feature activation-1.24e+00

Pos logit contributions
arching+0.212
toe+0.195
adia+0.192
hang+0.182
mat+0.181
Neg logit contributions
 condem-0.340
 newsp-0.275
terms-0.272
 constitu-0.269
cffff-0.262
Token said
Token ID531
Feature activation+1.14e-01

Pos logit contributions
 condem+2.880
 newsp+2.308
terms+2.273
 constitu+2.236
cffff+2.154
Neg logit contributions
arching-1.972
toe-1.785
adia-1.758
hang-1.676
estine-1.668
Token Friday
Token ID3217
Feature activation+1.84e-02

Pos logit contributions
arching+0.077
toe+0.071
adia+0.070
hang+0.066
mat+0.066
Neg logit contributions
 condem-0.116
 newsp-0.094
terms-0.093
 constitu-0.092
cffff-0.089
Token that
Token ID326
Feature activation+3.11e-01

Pos logit contributions
arching+0.012
toe+0.011
adia+0.011
hang+0.010
mat+0.010
Neg logit contributions
 condem-0.019
 newsp-0.016
terms-0.015
 constitu-0.015
cffff-0.015
Token being
Token ID852
Feature activation+8.66e-01

Pos logit contributions
arching+0.233
toe+0.215
adia+0.214
hang+0.203
estine+0.203
Neg logit contributions
 condem-0.294
 newsp-0.233
terms-0.231
 constitu-0.227
cffff-0.220
Token first
Token ID717
Feature activation-1.73e-01

Pos logit contributions
arching+0.599
toe+0.549
adia+0.544
mat+0.520
estine+0.518
Neg logit contributions
 condem-0.863
 newsp-0.699
terms-0.694
 constitu-0.686
cffff-0.663
Token lady
Token ID10846
Feature activation+2.82e+00

Pos logit contributions
 condem+0.196
 newsp+0.161
terms+0.160
 constitu+0.158
cffff+0.153
Neg logit contributions
arching-0.099
toe-0.089
adia-0.088
hang-0.082
estine-0.082
Token '
Token ID705
Feature activation-6.31e-01

Pos logit contributions
arching+1.709
toe+1.581
adia+1.581
mat+1.486
estine+1.470
Neg logit contributions
 condem-2.998
 newsp-2.480
terms-2.441
 constitu-2.429
cffff-2.372
Tokenhas
Token ID10134
Feature activation-6.97e-01

Pos logit contributions
 condem+0.651
 newsp+0.525
terms+0.524
 constitu+0.514
cffff+0.498
Neg logit contributions
arching-0.423
toe-0.384
adia-0.381
hang-0.360
estine-0.360
Token been
Token ID587
Feature activation+2.14e-01

Pos logit contributions
 condem+0.706
 newsp+0.566
terms+0.563
 constitu+0.554
cffff+0.534
Neg logit contributions
arching-0.483
toe-0.440
adia-0.435
hang-0.414
estine-0.412
Token the
Token ID262
Feature activation+4.52e-01

Pos logit contributions
arching+0.139
toe+0.126
adia+0.125
hang+0.118
estine+0.117
Neg logit contributions
 condem-0.226
 newsp-0.183
terms-0.182
 constitu-0.179
cffff-0.173
Token greatest
Token ID6000
Feature activation+8.17e-01

Pos logit contributions
arching+0.271
toe+0.243
adia+0.241
hang+0.226
estine+0.225
Neg logit contributions
 condem-0.500
 newsp-0.410
terms-0.408
 constitu-0.401
cffff-0.389
Token.
Token ID13
Feature activation+2.56e-01

Pos logit contributions
arching+0.875
toe+0.782
adia+0.773
hang+0.724
estine+0.720
Neg logit contributions
 condem-1.725
 newsp-1.423
terms-1.415
 constitu-1.394
cffff-1.354
Token It
Token ID632
Feature activation-1.20e-01

Pos logit contributions
arching+0.158
toe+0.143
adia+0.142
hang+0.133
mat+0.133
Neg logit contributions
 condem-0.278
 newsp-0.227
terms-0.225
 constitu-0.222
cffff-0.216
Token was
Token ID373
Feature activation+1.19e-01

Pos logit contributions
 condem+0.124
 newsp+0.100
terms+0.100
 constitu+0.098
cffff+0.095
Neg logit contributions
arching-0.080
toe-0.073
adia-0.072
hang-0.068
estine-0.068
Token a
Token ID257
Feature activation+1.71e-01

Pos logit contributions
arching+0.076
toe+0.069
adia+0.068
hang+0.065
estine+0.064
Neg logit contributions
 condem-0.126
 newsp-0.102
terms-0.101
 constitu-0.100
cffff-0.097
Token different
Token ID1180
Feature activation+2.03e+00

Pos logit contributions
arching+0.125
toe+0.114
adia+0.113
hang+0.108
estine+0.107
Neg logit contributions
 condem-0.167
 newsp-0.133
terms-0.132
 constitu-0.130
cffff-0.126
Token story
Token ID1621
Feature activation+3.24e+00

Pos logit contributions
arching+1.263
toe+1.144
adia+1.133
hang+1.081
estine+1.068
Neg logit contributions
 condem-2.184
 newsp-1.780
terms-1.773
 constitu-1.746
cffff-1.717
Token in
Token ID287
Feature activation+1.99e-01

Pos logit contributions
arching+1.988
adia+1.784
toe+1.781
mat+1.675
hang+1.673
Neg logit contributions
 condem-3.517
 newsp-2.875
terms-2.847
 constitu-2.827
cffff-2.739
Token the
Token ID262
Feature activation-1.71e-01

Pos logit contributions
arching+0.130
toe+0.118
adia+0.117
hang+0.111
mat+0.110
Neg logit contributions
 condem-0.209
 newsp-0.169
terms-0.167
 constitu-0.165
cffff-0.162
Token U
Token ID471
Feature activation+5.71e-01

Pos logit contributions
 condem+0.167
 newsp+0.132
terms+0.131
 constitu+0.129
cffff+0.126
Neg logit contributions
arching-0.126
toe-0.115
adia-0.115
hang-0.109
mat-0.108
Token.
Token ID13
Feature activation-6.27e-01

Pos logit contributions
arching+0.277
toe+0.243
adia+0.240
hang+0.221
mat+0.220
Neg logit contributions
 condem-0.699
 newsp-0.586
terms-0.579
 constitu-0.574
cffff-0.560
TokenS
Token ID50
Feature activation+9.21e-01

Pos logit contributions
 condem+0.541
 newsp+0.417
terms+0.415
 constitu+0.406
cffff+0.388
Neg logit contributions
arching-0.527
toe-0.490
adia-0.485
hang-0.465
estine-0.463
Tokenborn
Token ID6286
Feature activation-8.94e-01

Pos logit contributions
 condem+1.473
 newsp+1.215
terms+1.210
 constitu+1.187
cffff+1.154
Neg logit contributions
arching-0.751
toe-0.671
adia-0.666
hang-0.620
estine-0.620
Token Farmer
Token ID37755
Feature activation-1.44e+00

Pos logit contributions
 condem+0.916
 newsp+0.742
terms+0.732
 constitu+0.722
cffff+0.691
Neg logit contributions
arching-0.613
toe-0.555
adia-0.549
hang-0.520
estine-0.518
Token on
Token ID319
Feature activation+1.32e-01

Pos logit contributions
 condem+1.451
 newsp+1.169
terms+1.162
 constitu+1.139
cffff+1.098
Neg logit contributions
arching-1.009
toe-0.914
adia-0.906
hang-0.862
estine-0.856
Token "
Token ID366
Feature activation-2.69e-02

Pos logit contributions
arching+0.092
toe+0.084
adia+0.083
hang+0.079
estine+0.078
Neg logit contributions
 condem-0.134
 newsp-0.108
terms-0.107
 constitu-0.105
cffff-0.102
TokenFr
Token ID6732
Feature activation-3.39e-01

Pos logit contributions
 condem+0.026
 newsp+0.021
terms+0.021
 constitu+0.020
cffff+0.020
Neg logit contributions
arching-0.020
toe-0.018
adia-0.018
hang-0.017
estine-0.017
Tokenances
Token ID1817
Feature activation+3.25e+00

Pos logit contributions
 condem+0.374
terms+0.308
 newsp+0.308
 constitu+0.302
cffff+0.294
Neg logit contributions
arching-0.195
toe-0.178
adia-0.174
hang-0.164
mat-0.162
Token Farmer
Token ID37755
Feature activation-1.08e+00

Pos logit contributions
arching+2.214
toe+2.021
adia+2.004
hang+1.908
mat+1.893
Neg logit contributions
 condem-3.335
 newsp-2.676
terms-2.652
 constitu-2.617
cffff-2.550
Token Will
Token ID2561
Feature activation+1.24e+00

Pos logit contributions
 condem+1.163
 newsp+0.953
terms+0.950
 constitu+0.931
cffff+0.900
Neg logit contributions
arching-0.670
toe-0.602
adia-0.597
hang-0.563
estine-0.557
Token Have
Token ID8192
Feature activation+4.19e-01

Pos logit contributions
arching+0.852
toe+0.779
adia+0.772
hang+0.735
estine+0.731
Neg logit contributions
 condem-1.263
 newsp-1.019
terms-1.008
 constitu-0.996
cffff-0.966
Token Her
Token ID2332
Feature activation+4.43e-01

Pos logit contributions
arching+0.270
toe+0.243
adia+0.242
hang+0.228
estine+0.227
Neg logit contributions
 condem-0.443
 newsp-0.361
terms-0.359
 constitu-0.353
cffff-0.341
Token Revenge
Token ID24366
Feature activation+3.07e-01

Pos logit contributions
arching+0.296
toe+0.268
adia+0.266
hang+0.251
estine+0.250
Neg logit contributions
 condem-0.459
 newsp-0.371
terms-0.370
 constitu-0.363
cffff-0.350
Token over
Token ID625
Feature activation-1.73e-01

Pos logit contributions
 condem+0.686
 newsp+0.544
terms+0.540
 constitu+0.531
cffff+0.512
Neg logit contributions
arching-0.538
toe-0.495
adia-0.490
hang-0.467
estine-0.465
Tokencharged
Token ID17200
Feature activation+1.04e+00

Pos logit contributions
 condem+0.163
 newsp+0.128
terms+0.128
 constitu+0.125
cffff+0.121
Neg logit contributions
arching-0.133
toe-0.122
adia-0.121
estine-0.115
hang-0.115
Token by
Token ID416
Feature activation-4.91e-01

Pos logit contributions
arching+0.787
toe+0.723
adia+0.715
hang+0.683
mat+0.680
Neg logit contributions
 condem-0.987
 newsp-0.783
terms-0.774
 constitu-0.761
cffff-0.733
Token between
Token ID1022
Feature activation-1.47e-01

Pos logit contributions
 condem+0.471
 newsp+0.373
terms+0.371
 constitu+0.364
cffff+0.351
Neg logit contributions
arching-0.367
toe-0.337
adia-0.334
hang-0.318
estine-0.317
Token 10
Token ID838
Feature activation+1.28e+00

Pos logit contributions
 condem+0.129
 newsp+0.100
terms+0.099
 constitu+0.097
cffff+0.094
Neg logit contributions
arching-0.122
toe-0.113
adia-0.112
hang-0.107
estine-0.107
Token%
Token ID4
Feature activation+3.17e+00

Pos logit contributions
arching+0.765
toe+0.689
adia+0.680
hang+0.636
mat+0.635
Neg logit contributions
 condem-1.420
 newsp-1.165
terms-1.157
 constitu-1.141
cffff-1.108
Token and
Token ID290
Feature activation+6.41e-01

Pos logit contributions
arching+1.732
toe+1.548
adia+1.527
hang+1.422
mat+1.420
Neg logit contributions
 condem-3.652
 newsp-3.029
terms-2.986
 constitu-2.964
cffff-2.887
Token 30
Token ID1542
Feature activation+6.40e-02

Pos logit contributions
arching+0.426
toe+0.387
adia+0.382
hang+0.364
mat+0.361
Neg logit contributions
 condem-0.669
 newsp-0.541
terms-0.535
 constitu-0.528
cffff-0.513
Token%.
Token ID7225
Feature activation+3.34e+00

Pos logit contributions
arching+0.034
toe+0.030
adia+0.030
hang+0.028
estine+0.028
Neg logit contributions
 condem-0.075
 newsp-0.062
terms-0.062
 constitu-0.061
cffff-0.059
Token
Token ID198
Feature activation-2.27e-01

Pos logit contributions
arching+1.913
toe+1.776
adia+1.727
mat+1.636
hang+1.619
Neg logit contributions
 condem-3.750
 newsp-3.091
terms-3.024
 constitu-3.002
cffff-2.965
Token
Token ID198
Feature activation-4.28e-01

Pos logit contributions
 condem+0.257
terms+0.213
 newsp+0.212
 constitu+0.208
cffff+0.202
Neg logit contributions
arching-0.128
toe-0.114
adia-0.113
hang-0.105
estine-0.105
Token Nations
Token ID7973
Feature activation+8.44e-01

Pos logit contributions
arching+2.339
toe+2.114
adia+2.108
estine+1.990
hang+1.985
Neg logit contributions
 condem-3.593
 newsp-2.900
terms-2.888
 constitu-2.834
cffff-2.757
Token said
Token ID531
Feature activation+2.85e-01

Pos logit contributions
arching+0.568
toe+0.517
adia+0.511
hang+0.484
estine+0.482
Neg logit contributions
 condem-0.872
 newsp-0.704
terms-0.699
 constitu-0.687
cffff-0.664
Token last
Token ID938
Feature activation-1.10e+00

Pos logit contributions
arching+0.183
toe+0.166
adia+0.164
hang+0.155
estine+0.154
Neg logit contributions
 condem-0.304
 newsp-0.247
terms-0.245
 constitu-0.241
cffff-0.233
Token week
Token ID1285
Feature activation+1.24e+00

Pos logit contributions
 condem+0.933
 newsp+0.717
terms+0.711
 constitu+0.693
cffff+0.658
Neg logit contributions
arching-0.936
toe-0.869
adia-0.861
estine-0.823
hang-0.822
Token that
Token ID326
Feature activation+4.80e-01

Pos logit contributions
arching+0.791
toe+0.715
adia+0.708
hang+0.668
estine+0.665
Neg logit contributions
 condem-1.329
 newsp-1.083
terms-1.076
 constitu-1.060
cffff-1.025
Token Congo
Token ID25935
Feature activation+2.82e+00

Pos logit contributions
arching+0.320
toe+0.290
adia+0.287
hang+0.272
estine+0.271
Neg logit contributions
 condem-0.499
 newsp-0.404
terms-0.400
 constitu-0.395
cffff-0.379
Token had
Token ID550
Feature activation+1.36e+00

Pos logit contributions
arching+1.838
toe+1.665
adia+1.654
mat+1.563
estine+1.557
Neg logit contributions
 condem-2.948
 newsp-2.397
terms-2.383
 constitu-2.338
cffff-2.287
Token suspended
Token ID9951
Feature activation-4.75e-01

Pos logit contributions
arching+0.892
toe+0.818
adia+0.806
mat+0.764
hang+0.761
Neg logit contributions
 condem-1.408
 newsp-1.149
terms-1.142
 constitu-1.117
cffff-1.100
Token a
Token ID257
Feature activation+8.79e-01

Pos logit contributions
 condem+0.496
 newsp+0.402
terms+0.397
 constitu+0.393
cffff+0.373
Neg logit contributions
arching-0.313
toe-0.285
adia-0.282
hang-0.268
estine-0.266
Token dozen
Token ID8667
Feature activation-9.65e-02

Pos logit contributions
arching+0.659
toe+0.605
adia+0.600
hang+0.571
estine+0.570
Neg logit contributions
 condem-0.840
 newsp-0.666
terms-0.660
 constitu-0.649
cffff-0.625
Token senior
Token ID4664
Feature activation-4.68e-01

Pos logit contributions
 condem+0.092
 newsp+0.073
terms+0.072
 constitu+0.071
cffff+0.068
Neg logit contributions
arching-0.073
toe-0.067
adia-0.066
hang-0.063
estine-0.063
Token would
Token ID561
Feature activation-6.51e-01

Pos logit contributions
arching+0.117
toe+0.107
adia+0.106
hang+0.100
estine+0.100
Neg logit contributions
 condem-0.170
 newsp-0.137
terms-0.136
 constitu-0.134
cffff-0.129
Token be
Token ID307
Feature activation-2.57e-01

Pos logit contributions
 condem+0.653
 newsp+0.523
terms+0.519
 constitu+0.509
cffff+0.492
Neg logit contributions
arching-0.459
toe-0.417
adia-0.414
estine-0.391
hang-0.389
Token like
Token ID588
Feature activation-2.89e-01

Pos logit contributions
 condem+0.270
 newsp+0.219
terms+0.217
 constitu+0.214
cffff+0.207
Neg logit contributions
arching-0.168
toe-0.153
adia-0.151
hang-0.143
estine-0.142
Token,
Token ID11
Feature activation-1.88e-02

Pos logit contributions
 condem+0.324
 newsp+0.267
terms+0.265
 constitu+0.262
cffff+0.254
Neg logit contributions
arching-0.168
toe-0.150
adia-0.148
estine-0.138
hang-0.138
Token
Token ID564
Feature activation-8.82e-02

Pos logit contributions
 condem+0.019
 newsp+0.015
terms+0.015
 constitu+0.015
cffff+0.015
Neg logit contributions
arching-0.013
toe-0.012
adia-0.012
hang-0.011
estine-0.011
Token
Token ID250
Feature activation+1.36e-01

Pos logit contributions
 condem+0.091
 newsp+0.074
terms+0.073
 constitu+0.072
cffff+0.070
Neg logit contributions
arching-0.059
toe-0.054
adia-0.053
hang-0.051
estine-0.050
TokenOh
Token ID5812
Feature activation+2.95e-01

Pos logit contributions
arching+0.096
toe+0.088
adia+0.087
hang+0.083
mat+0.083
Neg logit contributions
 condem-0.136
 newsp-0.109
terms-0.108
 constitu-0.106
cffff-0.102
Token,
Token ID11
Feature activation+2.49e-01

Pos logit contributions
arching+0.182
toe+0.164
adia+0.162
hang+0.152
estine+0.152
Neg logit contributions
 condem-0.321
 newsp-0.262
terms-0.261
 constitu-0.257
cffff-0.249
Token they
Token ID484
Feature activation-1.66e-01

Pos logit contributions
arching+0.164
toe+0.148
adia+0.147
hang+0.140
mat+0.138
Neg logit contributions
 condem-0.260
 newsp-0.211
terms-0.209
 constitu-0.206
cffff-0.200
Token are
Token ID389
Feature activation+6.11e-01

Pos logit contributions
 condem+0.172
 newsp+0.139
terms+0.138
 constitu+0.136
cffff+0.131
Neg logit contributions
arching-0.111
toe-0.101
adia-0.100
estine-0.095
hang-0.094
Token killing
Token ID5170
Feature activation-1.43e+00

Pos logit contributions
arching+0.410
toe+0.372
adia+0.369
hang+0.349
estine+0.348
Neg logit contributions
 condem-0.632
 newsp-0.511
terms-0.508
 constitu-0.499
cffff-0.484
Token who
Token ID508
Feature activation+9.80e-02

Pos logit contributions
arching+0.147
toe+0.133
adia+0.132
hang+0.125
mat+0.125
Neg logit contributions
 condem-0.239
 newsp-0.194
terms-0.193
 constitu-0.190
cffff-0.184
Token looks
Token ID3073
Feature activation-1.27e+00

Pos logit contributions
arching+0.067
toe+0.061
adia+0.060
hang+0.057
estine+0.057
Neg logit contributions
 condem-0.100
 newsp-0.081
terms-0.080
 constitu-0.079
cffff-0.077
Token like
Token ID588
Feature activation-2.71e+00

Pos logit contributions
 condem+1.316
 newsp+1.063
terms+1.052
 constitu+1.039
cffff+0.998
Neg logit contributions
arching-0.856
toe-0.774
adia-0.763
hang-0.721
estine-0.719
Token he
Token ID339
Feature activation+8.95e-02

Pos logit contributions
 condem+2.752
 newsp+2.232
terms+2.198
 constitu+2.167
cffff+2.094
Neg logit contributions
arching-1.859
toe-1.682
adia-1.653
inion-1.571
hang-1.561
Token
Token ID447
Feature activation-3.77e-01

Pos logit contributions
arching+0.059
toe+0.053
adia+0.053
hang+0.050
estine+0.049
Neg logit contributions
 condem-0.094
 newsp-0.076
terms-0.076
 constitu-0.074
cffff-0.072
Token
Token ID247
Feature activation-5.91e-02

Pos logit contributions
 condem+0.336
 newsp+0.262
terms+0.260
 constitu+0.254
cffff+0.245
Neg logit contributions
arching-0.307
toe-0.283
adia-0.281
hang-0.268
estine-0.268
Tokend
Token ID67
Feature activation+1.08e+00

Pos logit contributions
 condem+0.055
terms+0.044
 newsp+0.043
 constitu+0.042
cffff+0.041
Neg logit contributions
arching-0.045
toe-0.042
adia-0.041
hang-0.039
estine-0.039
Token been
Token ID587
Feature activation+1.21e-01

Pos logit contributions
arching+0.729
toe+0.663
adia+0.656
hang+0.622
estine+0.619
Neg logit contributions
 condem-1.111
 newsp-0.898
terms-0.892
 constitu-0.878
cffff-0.849
Token sent
Token ID1908
Feature activation+7.96e-01

Pos logit contributions
arching+0.085
toe+0.078
adia+0.077
hang+0.073
estine+0.073
Neg logit contributions
 condem-0.121
 newsp-0.097
terms-0.096
 constitu-0.095
cffff-0.092
Token by
Token ID416
Feature activation-1.51e-01

Pos logit contributions
arching+0.527
toe+0.479
adia+0.473
hang+0.447
mat+0.446
Neg logit contributions
 condem-0.828
 newsp-0.672
terms-0.667
 constitu-0.658
cffff-0.635
Token central
Token ID4318
Feature activation+1.23e+00

Pos logit contributions
 condem+0.152
 newsp+0.122
terms+0.121
 constitu+0.119
cffff+0.115
Neg logit contributions
arching-0.106
toe-0.097
adia-0.096
hang-0.091
estine-0.091
Token it
Token ID340
Feature activation-3.12e-02

Pos logit contributions
 condem+0.988
 newsp+0.806
terms+0.797
 constitu+0.787
cffff+0.762
Neg logit contributions
arching-0.575
toe-0.518
adia-0.512
estine-0.481
hang-0.480
Token first
Token ID717
Feature activation+3.57e-02

Pos logit contributions
 condem+0.033
 newsp+0.027
terms+0.027
 constitu+0.026
cffff+0.025
Neg logit contributions
arching-0.020
toe-0.018
adia-0.018
hang-0.017
estine-0.017
Token came
Token ID1625
Feature activation+1.74e+00

Pos logit contributions
arching+0.024
toe+0.022
adia+0.022
hang+0.020
estine+0.020
Neg logit contributions
 condem-0.037
 newsp-0.030
terms-0.030
 constitu-0.029
cffff-0.028
Token about
Token ID546
Feature activation+6.20e-01

Pos logit contributions
arching+1.137
toe+1.032
adia+1.025
hang+0.968
mat+0.964
Neg logit contributions
 condem-1.834
 newsp-1.492
terms-1.477
 constitu-1.457
cffff-1.414
Token when
Token ID618
Feature activation+1.08e+00

Pos logit contributions
arching+0.415
toe+0.377
adia+0.373
hang+0.353
estine+0.352
Neg logit contributions
 condem-0.642
 newsp-0.519
terms-0.517
 constitu-0.507
cffff-0.492
Token L
Token ID406
Feature activation+9.61e-03

Pos logit contributions
arching+0.737
toe+0.671
adia+0.665
hang+0.630
estine+0.628
Neg logit contributions
 condem-1.109
 newsp-0.894
terms-0.888
 constitu-0.874
cffff-0.845
Token.
Token ID13
Feature activation-3.48e-01

Pos logit contributions
arching+0.005
toe+0.005
adia+0.004
hang+0.004
estine+0.004
Neg logit contributions
 condem-0.011
 newsp-0.009
terms-0.009
 constitu-0.009
cffff-0.009
TokenA
Token ID32
Feature activation+5.12e-01

Pos logit contributions
 condem+0.347
 newsp+0.277
terms+0.276
 constitu+0.270
cffff+0.262
Neg logit contributions
arching-0.247
toe-0.226
adia-0.224
hang-0.213
mat-0.212
Token.
Token ID13
Feature activation-7.24e-01

Pos logit contributions
arching+0.282
toe+0.254
adia+0.245
hang+0.231
estine+0.230
Neg logit contributions
 condem-0.591
 newsp-0.491
terms-0.483
 constitu-0.480
cffff-0.463
Token attorney
Token ID6136
Feature activation+1.08e+00

Pos logit contributions
 condem+0.696
 newsp+0.554
terms+0.547
 constitu+0.540
cffff+0.517
Neg logit contributions
arching-0.539
toe-0.496
adia-0.489
hang-0.465
estine-0.465
Token Andrew
Token ID6858
Feature activation+1.13e+00

Pos logit contributions
arching+0.763
toe+0.698
adia+0.693
hang+0.657
mat+0.656
Neg logit contributions
 condem-1.080
 newsp-0.864
terms-0.859
 constitu-0.845
cffff-0.820
Token that
Token ID326
Feature activation-5.90e-01

Pos logit contributions
arching+0.155
toe+0.141
adia+0.139
hang+0.131
estine+0.131
Neg logit contributions
 condem-0.245
 newsp-0.198
terms-0.196
 constitu-0.194
cffff-0.187
Token some
Token ID617
Feature activation+1.07e+00

Pos logit contributions
 condem+0.620
 newsp+0.502
terms+0.496
 constitu+0.491
cffff+0.469
Neg logit contributions
arching-0.387
toe-0.351
adia-0.347
hang-0.326
estine-0.326
Token prisoners
Token ID10577
Feature activation-1.15e+00

Pos logit contributions
arching+0.803
toe+0.737
adia+0.732
hang+0.698
estine+0.695
Neg logit contributions
 condem-1.027
 newsp-0.815
terms-0.810
 constitu-0.794
cffff-0.769
Token were
Token ID547
Feature activation+1.15e+00

Pos logit contributions
 condem+1.132
terms+0.894
 newsp+0.893
 constitu+0.876
cffff+0.828
Neg logit contributions
arching-0.839
toe-0.766
adia-0.756
hang-0.717
estine-0.716
Token tortured
Token ID22485
Feature activation+6.09e-01

Pos logit contributions
arching+0.844
toe+0.778
adia+0.773
hang+0.736
mat+0.732
Neg logit contributions
 condem-1.107
 newsp-0.884
terms-0.881
 constitu-0.862
cffff-0.842
Token.
Token ID13
Feature activation+8.38e-03

Pos logit contributions
arching+0.354
toe+0.316
adia+0.312
hang+0.293
estine+0.291
Neg logit contributions
 condem-0.684
 newsp-0.564
terms-0.561
 constitu-0.552
cffff-0.537
Token
Token ID198
Feature activation+3.58e-01

Pos logit contributions
arching+0.005
toe+0.004
adia+0.004
hang+0.004
estine+0.004
Neg logit contributions
 condem-0.009
terms-0.008
 newsp-0.008
 constitu-0.008
cffff-0.007
Token
Token ID198
Feature activation+1.33e-01

Pos logit contributions
arching+0.203
toe+0.180
adia+0.179
hang+0.166
estine+0.166
Neg logit contributions
 condem-0.404
terms-0.335
 newsp-0.334
 constitu-0.328
cffff-0.318
TokenPhoto
Token ID6191
Feature activation-9.48e-01

Pos logit contributions
arching+0.097
toe+0.088
adia+0.088
hang+0.083
estine+0.083
Neg logit contributions
 condem-0.130
 newsp-0.104
terms-0.103
 constitu-0.101
cffff-0.097
Token
Token ID198
Feature activation-7.08e-01

Pos logit contributions
 condem+1.066
 newsp+0.876
terms+0.875
 constitu+0.861
cffff+0.830
Neg logit contributions
arching-0.547
toe-0.489
adia-0.486
hang-0.452
estine-0.448
Token
Token ID198
Feature activation-6.34e-01

Pos logit contributions
 condem+0.771
terms+0.647
 newsp+0.635
 constitu+0.627
cffff+0.607
Neg logit contributions
arching-0.419
toe-0.370
adia-0.369
estine-0.344
hang-0.343
Token.
Token ID13
Feature activation+4.93e-01

Pos logit contributions
arching+1.554
toe+1.409
adia+1.379
mat+1.302
hang+1.296
Neg logit contributions
 condem-3.092
 newsp-2.547
terms-2.529
 constitu-2.486
cffff-2.432
Token
Token ID447
Feature activation+2.55e-01

Pos logit contributions
arching+0.300
toe+0.270
adia+0.268
hang+0.252
estine+0.251
Neg logit contributions
 condem-0.539
 newsp-0.442
terms-0.439
 constitu-0.432
cffff-0.419
Token
Token ID251
Feature activation+8.34e-01

Pos logit contributions
arching+0.179
toe+0.163
adia+0.161
hang+0.153
estine+0.153
Neg logit contributions
 condem-0.256
 newsp-0.206
terms-0.204
 constitu-0.201
cffff-0.194
Token
Token ID198
Feature activation+3.73e-01

Pos logit contributions
arching+0.499
toe+0.449
adia+0.444
hang+0.417
estine+0.415
Neg logit contributions
 condem-0.922
 newsp-0.756
terms-0.751
 constitu-0.740
cffff-0.719
Token
Token ID198
Feature activation+1.86e-01

Pos logit contributions
arching+0.209
toe+0.185
adia+0.183
hang+0.171
estine+0.170
Neg logit contributions
 condem-0.426
 newsp-0.353
terms-0.352
 constitu-0.346
cffff-0.336
Token
Token ID447
Feature activation+2.88e-01

Pos logit contributions
arching+0.129
toe+0.117
adia+0.117
hang+0.110
estine+0.110
Neg logit contributions
 condem-0.187
 newsp-0.150
terms-0.150
 constitu-0.147
cffff-0.142
Token
Token ID250
Feature activation+5.19e-01

Pos logit contributions
arching+0.197
toe+0.179
adia+0.177
hang+0.168
estine+0.167
Neg logit contributions
 condem-0.294
 newsp-0.237
terms-0.235
 constitu-0.231
cffff-0.223
TokenI
Token ID40
Feature activation+1.38e+00

Pos logit contributions
arching+0.371
toe+0.339
adia+0.336
hang+0.320
estine+0.318
Neg logit contributions
 condem-0.514
 newsp-0.412
terms-0.408
 constitu-0.402
cffff-0.388
Token mean
Token ID1612
Feature activation+1.77e+00

Pos logit contributions
arching+0.863
toe+0.779
adia+0.771
hang+0.727
estine+0.724
Neg logit contributions
 condem-1.482
 newsp-1.210
terms-1.202
 constitu-1.184
cffff-1.148
Token,
Token ID11
Feature activation-4.10e-01

Pos logit contributions
arching+0.991
toe+0.887
adia+0.877
hang+0.824
mat+0.818
Neg logit contributions
 condem-2.026
 newsp-1.673
terms-1.661
 constitu-1.638
cffff-1.596
Token that
Token ID326
Feature activation-3.99e-01

Pos logit contributions
 condem+0.429
 newsp+0.348
terms+0.346
 constitu+0.340
cffff+0.330
Neg logit contributions
arching-0.270
toe-0.244
adia-0.242
hang-0.229
estine-0.228
Token acqu
Token ID4078
Feature activation-7.74e-01

Pos logit contributions
 condem+0.396
 newsp+0.314
terms+0.313
 constitu+0.306
cffff+0.297
Neg logit contributions
arching-0.298
toe-0.271
adia-0.270
hang-0.256
estine-0.255
Tokenit
Token ID270
Feature activation-1.89e-01

Pos logit contributions
 condem+0.934
terms+0.788
 newsp+0.779
 constitu+0.769
cffff+0.759
Neg logit contributions
arching-0.373
toe-0.322
adia-0.317
hang-0.293
estine-0.290
Token themselves
Token ID2405
Feature activation+5.39e-01

Pos logit contributions
 condem+0.198
 newsp+0.161
terms+0.160
 constitu+0.157
cffff+0.152
Neg logit contributions
arching-0.124
toe-0.112
adia-0.111
hang-0.105
estine-0.105
Token well
Token ID880
Feature activation-1.02e+00

Pos logit contributions
arching+0.380
toe+0.347
adia+0.344
hang+0.327
estine+0.326
Neg logit contributions
 condem-0.538
 newsp-0.431
terms-0.428
 constitu-0.421
cffff-0.407
Token --
Token ID1377
Feature activation-1.20e+00

Pos logit contributions
 condem+1.083
 newsp+0.881
terms+0.876
 constitu+0.862
cffff+0.835
Neg logit contributions
arching-0.657
toe-0.593
adia-0.588
hang-0.555
estine-0.553
Token just
Token ID655
Feature activation-1.56e+00

Pos logit contributions
 condem+1.289
 newsp+1.052
terms+1.045
 constitu+1.029
cffff+0.997
Neg logit contributions
arching-0.761
toe-0.688
adia-0.681
hang-0.642
estine-0.639
Token not
Token ID407
Feature activation-1.17e+00

Pos logit contributions
 condem+1.591
 newsp+1.283
terms+1.275
 constitu+1.253
cffff+1.209
Neg logit contributions
arching-1.063
toe-0.966
adia-0.958
hang-0.908
estine-0.905
Token with
Token ID351
Feature activation-4.09e-02

Pos logit contributions
 condem+1.205
 newsp+0.973
terms+0.966
 constitu+0.951
cffff+0.920
Neg logit contributions
arching-0.793
toe-0.720
adia-0.714
hang-0.676
estine-0.673
Token as
Token ID355
Feature activation+2.44e-01

Pos logit contributions
 condem+0.042
 newsp+0.034
terms+0.034
 constitu+0.034
cffff+0.032
Neg logit contributions
arching-0.027
toe-0.025
adia-0.025
hang-0.023
estine-0.023
Token much
Token ID881
Feature activation-8.77e-01

Pos logit contributions
arching+0.169
toe+0.154
adia+0.152
hang+0.144
mat+0.144
Neg logit contributions
 condem-0.247
 newsp-0.199
terms-0.198
 constitu-0.194
cffff-0.188
Token composure
Token ID49994
Feature activation-7.50e-01

Pos logit contributions
 condem+0.860
 newsp+0.687
terms+0.680
 constitu+0.671
cffff+0.645
Neg logit contributions
arching-0.634
toe-0.580
adia-0.576
hang-0.547
estine-0.547
Token,
Token ID11
Feature activation-9.14e-02

Pos logit contributions
arching+0.235
toe+0.211
adia+0.209
hang+0.196
mat+0.195
Neg logit contributions
 condem-0.445
 newsp-0.366
terms-0.364
 constitu-0.359
cffff-0.349
Token back
Token ID736
Feature activation-2.66e-01

Pos logit contributions
 condem+0.095
 newsp+0.077
terms+0.077
 constitu+0.076
cffff+0.073
Neg logit contributions
arching-0.060
toe-0.055
adia-0.054
hang-0.051
estine-0.051
Token with
Token ID351
Feature activation+5.25e-01

Pos logit contributions
 condem+0.278
 newsp+0.225
terms+0.224
 constitu+0.219
cffff+0.212
Neg logit contributions
arching-0.175
toe-0.159
adia-0.157
hang-0.148
estine-0.148
Token a
Token ID257
Feature activation+3.31e-03

Pos logit contributions
arching+0.351
toe+0.319
adia+0.316
hang+0.299
estine+0.298
Neg logit contributions
 condem-0.544
 newsp-0.440
terms-0.437
 constitu-0.430
cffff-0.416
Token hopefully
Token ID11481
Feature activation+2.44e-01

Pos logit contributions
arching+0.002
toe+0.002
adia+0.002
hang+0.002
estine+0.002
Neg logit contributions
 condem-0.003
 newsp-0.003
terms-0.003
 constitu-0.003
cffff-0.002
Token improved
Token ID6596
Feature activation-1.42e+00

Pos logit contributions
arching+0.176
toe+0.161
adia+0.160
hang+0.152
estine+0.152
Neg logit contributions
 condem-0.240
 newsp-0.192
terms-0.190
 constitu-0.187
cffff-0.181
Token 100
Token ID1802
Feature activation+1.20e-01

Pos logit contributions
 condem+1.394
 newsp+1.116
terms+1.096
 constitu+1.087
cffff+1.035
Neg logit contributions
arching-1.028
toe-0.942
adia-0.934
hang-0.883
estine-0.883
Token%
Token ID4
Feature activation+1.78e+00

Pos logit contributions
arching+0.062
toe+0.055
adia+0.054
hang+0.050
estine+0.050
Neg logit contributions
 condem-0.142
 newsp-0.119
terms-0.118
 constitu-0.116
cffff-0.113
Token on
Token ID319
Feature activation-6.32e-02

Pos logit contributions
arching+1.239
toe+1.129
adia+1.121
hang+1.067
mat+1.064
Neg logit contributions
 condem-1.787
 newsp-1.429
terms-1.428
 constitu-1.401
cffff-1.364
Token ESA
Token ID42212
Feature activation-1.75e+00

Pos logit contributions
 condem+0.064
 newsp+0.052
terms+0.052
 constitu+0.051
cffff+0.049
Neg logit contributions
arching-0.043
toe-0.039
adia-0.039
hang-0.037
estine-0.037
Token2014
Token ID4967
Feature activation+7.76e-01

Pos logit contributions
 condem+1.999
terms+1.654
 newsp+1.652
 constitu+1.615
cffff+1.570
Neg logit contributions
arching-0.979
toe-0.872
adia-0.862
hang-0.806
estine-0.803
Token the
Token ID262
Feature activation-2.69e-01

Pos logit contributions
arching+0.129
toe+0.116
adia+0.114
hang+0.108
mat+0.108
Neg logit contributions
 condem-0.225
 newsp-0.184
terms-0.183
 constitu-0.180
cffff-0.175
Token CG
Token ID29925
Feature activation-3.44e-01

Pos logit contributions
 condem+0.269
 newsp+0.215
terms+0.214
 constitu+0.210
cffff+0.204
Neg logit contributions
arching-0.189
toe-0.173
adia-0.171
hang-0.163
mat-0.163
TokenPR
Token ID4805
Feature activation-1.79e+00

Pos logit contributions
 condem+0.345
 newsp+0.278
terms+0.276
 constitu+0.271
cffff+0.263
Neg logit contributions
arching-0.241
toe-0.219
adia-0.217
mat-0.206
hang-0.206
TokenOG
Token ID7730
Feature activation-2.66e+00

Pos logit contributions
 condem+1.675
 newsp+1.321
terms+1.311
 constitu+1.286
cffff+1.239
Neg logit contributions
arching-1.373
toe-1.265
adia-1.253
hang-1.197
estine-1.192
TokenRAM
Token ID24115
Feature activation-1.05e+00

Pos logit contributions
 condem+2.439
 newsp+1.918
terms+1.910
 constitu+1.857
cffff+1.793
Neg logit contributions
arching-2.080
toe-1.942
adia-1.907
hang-1.822
estine-1.817
Token/
Token ID14
Feature activation-7.04e-01

Pos logit contributions
 condem+1.091
 newsp+0.882
terms+0.877
 constitu+0.863
cffff+0.835
Neg logit contributions
arching-0.701
toe-0.637
adia-0.631
hang-0.597
estine-0.595
Token END
Token ID23578
Feature activation-1.63e+00

Pos logit contributions
 condem+0.706
 newsp+0.567
terms+0.563
 constitu+0.554
cffff+0.535
Neg logit contributions
arching-0.493
toe-0.450
adia-0.446
hang-0.423
mat-0.422
TokenCG
Token ID39816
Feature activation+1.68e-01

Pos logit contributions
 condem+1.581
 newsp+1.259
terms+1.251
 constitu+1.228
cffff+1.185
Neg logit contributions
arching-1.191
toe-1.092
adia-1.083
hang-1.031
estine-1.027
Token section
Token ID2665
Feature activation-2.82e-01

Pos logit contributions
arching+0.118
toe+0.107
adia+0.106
mat+0.101
hang+0.101
Neg logit contributions
 condem-0.169
 newsp-0.136
terms-0.135
 constitu-0.133
cffff-0.129
Token!
Token ID0
Feature activation+5.83e-01

Pos logit contributions
 condem+0.292
 newsp+0.236
terms+0.235
 constitu+0.231
cffff+0.224
Neg logit contributions
arching-0.189
toe-0.172
adia-0.171
hang-0.161
mat-0.161
Token):
Token ID2599
Feature activation+1.22e+00

Pos logit contributions
arching+0.365
toe+0.328
adia+0.325
mat+0.307
hang+0.307
Neg logit contributions
 condem-0.630
 newsp-0.512
terms-0.509
 constitu-0.504
cffff-0.487
Token
Token ID247
Feature activation-1.88e-01

Pos logit contributions
 condem+0.258
 newsp+0.200
terms+0.199
 constitu+0.194
cffff+0.187
Neg logit contributions
arching-0.250
toe-0.231
adia-0.229
hang-0.219
estine-0.219
Tokens
Token ID82
Feature activation+7.47e-03

Pos logit contributions
 condem+0.178
 newsp+0.141
terms+0.140
 constitu+0.138
cffff+0.133
Neg logit contributions
arching-0.142
toe-0.131
adia-0.130
hang-0.124
estine-0.123
Token steam
Token ID13324
Feature activation+5.07e-02

Pos logit contributions
arching+0.005
toe+0.005
adia+0.005
hang+0.005
estine+0.004
Neg logit contributions
 condem-0.007
 newsp-0.006
terms-0.006
 constitu-0.006
cffff-0.006
Token electric
Token ID5186
Feature activation+7.18e-01

Pos logit contributions
arching+0.035
toe+0.032
adia+0.031
hang+0.030
estine+0.029
Neg logit contributions
 condem-0.052
 newsp-0.042
terms-0.041
 constitu-0.041
cffff-0.039
Token eff
Token ID914
Feature activation-1.33e+00

Pos logit contributions
arching+0.494
toe+0.449
adia+0.446
hang+0.422
mat+0.421
Neg logit contributions
 condem-0.729
 newsp-0.588
terms-0.584
 constitu-0.575
cffff-0.558
Tokenluent
Token ID28216
Feature activation-2.97e-01

Pos logit contributions
 condem+1.281
 newsp+1.019
terms+1.014
 constitu+0.990
cffff+0.956
Neg logit contributions
arching-0.985
toe-0.903
adia-0.894
hang-0.855
estine-0.850
Token guidelines
Token ID9949
Feature activation+1.14e+00

Pos logit contributions
 condem+0.282
 newsp+0.223
terms+0.221
 constitu+0.217
cffff+0.210
Neg logit contributions
arching-0.224
toe-0.206
adia-0.205
hang-0.195
estine-0.194
Token regulating
Token ID26379
Feature activation+7.15e-01

Pos logit contributions
arching+0.625
toe+0.555
adia+0.549
hang+0.512
estine+0.510
Neg logit contributions
 condem-1.313
 newsp-1.088
terms-1.081
 constitu-1.066
cffff-1.037
Token wastewater
Token ID42073
Feature activation-1.55e+00

Pos logit contributions
arching+0.525
toe+0.482
adia+0.478
hang+0.455
estine+0.453
Neg logit contributions
 condem-0.692
 newsp-0.550
terms-0.547
 constitu-0.538
cffff-0.520
Token streams
Token ID15190
Feature activation+6.60e-01

Pos logit contributions
 condem+1.488
 newsp+1.180
terms+1.171
 constitu+1.151
cffff+1.106
Neg logit contributions
arching-1.158
toe-1.062
adia-1.054
hang-1.005
estine-1.000
Token
Token ID960
Feature activation-1.85e+00

Pos logit contributions
arching+0.367
toe+0.326
adia+0.323
hang+0.301
estine+0.300
Neg logit contributions
 condem-0.757
 newsp-0.627
terms-0.623
 constitu-0.614
cffff-0.597
TokenK
Token ID42
Feature activation+7.66e-01

Pos logit contributions
 condem+0.306
 newsp+0.248
terms+0.248
 constitu+0.243
cffff+0.234
Neg logit contributions
arching-0.199
toe-0.180
adia-0.179
hang-0.169
estine-0.169
Token.:
Token ID11207
Feature activation+5.91e-01

Pos logit contributions
arching+0.336
toe+0.291
adia+0.287
hang+0.261
estine+0.260
Neg logit contributions
 condem-0.972
 newsp-0.819
terms-0.811
 constitu-0.804
cffff-0.783
Token You
Token ID921
Feature activation-5.34e-01

Pos logit contributions
arching+0.390
toe+0.356
adia+0.353
hang+0.334
mat+0.333
Neg logit contributions
 condem-0.617
 newsp-0.499
terms-0.495
 constitu-0.488
cffff-0.474
Token almost
Token ID2048
Feature activation+1.18e+00

Pos logit contributions
 condem+0.549
 newsp+0.443
terms+0.440
 constitu+0.433
cffff+0.418
Neg logit contributions
arching-0.362
toe-0.329
adia-0.325
estine-0.308
hang-0.306
Token want
Token ID765
Feature activation+1.81e-02

Pos logit contributions
arching+0.788
toe+0.717
adia+0.710
hang+0.674
mat+0.672
Neg logit contributions
 condem-1.211
 newsp-0.982
terms-0.974
 constitu-0.959
cffff-0.929
Token to
Token ID284
Feature activation-8.96e-01

Pos logit contributions
arching+0.012
toe+0.011
adia+0.011
hang+0.010
estine+0.010
Neg logit contributions
 condem-0.019
 newsp-0.015
terms-0.015
 constitu-0.015
cffff-0.015
Token say
Token ID910
Feature activation-1.43e-01

Pos logit contributions
 condem+0.867
 newsp+0.686
terms+0.681
 constitu+0.668
cffff+0.642
Neg logit contributions
arching-0.664
toe-0.607
adia-0.601
estine-0.573
inion-0.570
Token to
Token ID284
Feature activation-1.19e+00

Pos logit contributions
 condem+0.150
 newsp+0.122
terms+0.121
 constitu+0.119
cffff+0.115
Neg logit contributions
arching-0.093
toe-0.085
adia-0.084
hang-0.079
estine-0.079
Token the
Token ID262
Feature activation-6.52e-02

Pos logit contributions
 condem+1.260
 newsp+1.024
terms+1.017
 constitu+1.001
cffff+0.970
Neg logit contributions
arching-0.768
toe-0.696
adia-0.686
estine-0.650
hang-0.648
Token person
Token ID1048
Feature activation-8.31e-01

Pos logit contributions
 condem+0.065
 newsp+0.052
terms+0.052
 constitu+0.051
cffff+0.049
Neg logit contributions
arching-0.046
toe-0.042
adia-0.042
hang-0.040
estine-0.040
Token you
Token ID345
Feature activation-7.68e-01

Pos logit contributions
 condem+0.874
 newsp+0.710
terms+0.706
 constitu+0.694
cffff+0.672
Neg logit contributions
arching-0.541
toe-0.490
adia-0.485
hang-0.459
estine-0.458
Token the
Token ID262
Feature activation+2.72e-01

Pos logit contributions
 condem+0.352
 newsp+0.286
terms+0.284
 constitu+0.279
cffff+0.270
Neg logit contributions
arching-0.221
toe-0.200
adia-0.198
hang-0.187
estine-0.187
Token bleak
Token ID30942
Feature activation+1.99e+00

Pos logit contributions
arching+0.198
toe+0.181
adia+0.180
hang+0.171
estine+0.170
Neg logit contributions
 condem-0.266
 newsp-0.212
terms-0.210
 constitu-0.207
cffff-0.200
Token solitude
Token ID49917
Feature activation+4.86e-02

Pos logit contributions
arching+1.318
toe+1.196
adia+1.189
hang+1.121
estine+1.120
Neg logit contributions
 condem-2.062
 newsp-1.661
terms-1.652
 constitu-1.631
cffff-1.590
Token of
Token ID286
Feature activation-3.39e-01

Pos logit contributions
arching+0.034
toe+0.031
adia+0.030
hang+0.029
estine+0.029
Neg logit contributions
 condem-0.049
 newsp-0.040
terms-0.039
 constitu-0.039
cffff-0.037
Token Super
Token ID3115
Feature activation-2.13e+00

Pos logit contributions
 condem+0.357
 newsp+0.290
terms+0.288
 constitu+0.283
cffff+0.274
Neg logit contributions
arching-0.221
toe-0.200
adia-0.198
hang-0.187
estine-0.186
Token Metroid
Token ID45430
Feature activation-3.46e+00

Pos logit contributions
 condem+2.137
terms+1.726
 newsp+1.713
 constitu+1.664
cffff+1.617
Neg logit contributions
arching-1.494
toe-1.350
adia-1.331
inion-1.258
hang-1.256
Token,
Token ID11
Feature activation+3.38e-01

Pos logit contributions
 condem+3.797
 newsp+3.170
terms+3.156
 constitu+3.063
cffff+2.968
Neg logit contributions
arching-2.038
toe-1.801
adia-1.754
inion-1.664
hang-1.663
Token and
Token ID290
Feature activation+3.81e-02

Pos logit contributions
arching+0.197
toe+0.176
adia+0.174
hang+0.163
estine+0.163
Neg logit contributions
 condem-0.379
 newsp-0.312
terms-0.310
 constitu-0.306
cffff-0.297
Token alongside
Token ID7848
Feature activation+3.13e-01

Pos logit contributions
arching+0.024
toe+0.022
adia+0.021
hang+0.020
estine+0.020
Neg logit contributions
 condem-0.041
 newsp-0.034
terms-0.033
 constitu-0.033
cffff-0.032
Token Mario
Token ID10682
Feature activation-1.26e+00

Pos logit contributions
arching+0.194
toe+0.175
adia+0.173
hang+0.163
estine+0.162
Neg logit contributions
 condem-0.339
 newsp-0.277
terms-0.275
 constitu-0.271
cffff-0.263
Token 64
Token ID5598
Feature activation-1.85e+00

Pos logit contributions
 condem+1.362
 newsp+1.117
terms+1.111
 constitu+1.088
cffff+1.057
Neg logit contributions
arching-0.776
toe-0.699
adia-0.691
estine-0.651
hang-0.650
Token more
Token ID517
Feature activation-4.18e-01

Pos logit contributions
 condem+1.163
 newsp+0.943
terms+0.934
 constitu+0.920
cffff+0.886
Neg logit contributions
arching-0.724
toe-0.657
adia-0.654
hang-0.613
estine-0.613
Token jun
Token ID10891
Feature activation-2.02e+00

Pos logit contributions
 condem+0.379
 newsp+0.296
terms+0.294
 constitu+0.288
cffff+0.277
Neg logit contributions
arching-0.334
toe-0.308
adia-0.306
hang-0.292
estine-0.291
Tokeniper
Token ID9346
Feature activation+1.09e+00

Pos logit contributions
 condem+2.005
terms+1.631
 newsp+1.604
 constitu+1.577
cffff+1.539
Neg logit contributions
arching-1.400
toe-1.267
adia-1.264
estine-1.200
hang-1.199
Token infusion
Token ID35547
Feature activation-9.00e-01

Pos logit contributions
arching+0.732
toe+0.666
adia+0.659
hang+0.625
mat+0.621
Neg logit contributions
 condem-1.118
 newsp-0.903
terms-0.897
 constitu-0.882
cffff-0.855
Token on
Token ID319
Feature activation+5.99e-01

Pos logit contributions
 condem+0.942
 newsp+0.765
terms+0.760
 constitu+0.746
cffff+0.721
Neg logit contributions
arching-0.591
toe-0.537
adia-0.532
hang-0.502
estine-0.501
Token top
Token ID1353
Feature activation-3.32e+00

Pos logit contributions
arching+0.397
toe+0.360
adia+0.353
hang+0.339
mat+0.335
Neg logit contributions
 condem-0.626
 newsp-0.505
terms-0.502
 constitu-0.497
cffff-0.481
Token as
Token ID355
Feature activation+5.37e-01

Pos logit contributions
 condem+3.263
terms+2.626
 newsp+2.622
 constitu+2.553
cffff+2.472
Neg logit contributions
arching-2.335
toe-2.127
adia-2.113
estine-2.010
hang-2.009
Token needed
Token ID2622
Feature activation+1.03e+00

Pos logit contributions
arching+0.393
toe+0.360
adia+0.357
hang+0.340
mat+0.339
Neg logit contributions
 condem-0.523
 newsp-0.416
terms-0.414
 constitu-0.407
cffff-0.393
Token.
Token ID13
Feature activation+2.62e-02

Pos logit contributions
arching+0.544
toe+0.481
adia+0.475
hang+0.441
mat+0.440
Neg logit contributions
 condem-1.208
 newsp-1.005
terms-0.997
 constitu-0.985
cffff-0.957
Token
Token ID198
Feature activation+2.50e-01

Pos logit contributions
arching+0.017
toe+0.016
adia+0.015
hang+0.015
estine+0.015
Neg logit contributions
 condem-0.028
 newsp-0.022
terms-0.022
 constitu-0.022
cffff-0.021
Token
Token ID198
Feature activation-7.57e-02

Pos logit contributions
arching+0.139
toe+0.123
adia+0.122
hang+0.114
estine+0.113
Neg logit contributions
 condem-0.287
 newsp-0.237
terms-0.236
 constitu-0.233
cffff-0.226
Token different
Token ID1180
Feature activation+1.16e+00

Pos logit contributions
 condem+0.950
 newsp+0.771
terms+0.766
 constitu+0.753
cffff+0.729
Neg logit contributions
arching-0.597
toe-0.542
adia-0.536
hang-0.507
estine-0.505
Token.
Token ID13
Feature activation+4.34e-02

Pos logit contributions
arching+0.624
toe+0.553
adia+0.547
hang+0.509
mat+0.507
Neg logit contributions
 condem-1.355
 newsp-1.124
terms-1.117
 constitu-1.103
cffff-1.073
Token
Token ID198
Feature activation-4.36e-01

Pos logit contributions
arching+0.027
toe+0.025
adia+0.024
hang+0.023
mat+0.023
Neg logit contributions
 condem-0.047
 newsp-0.038
terms-0.038
 constitu-0.037
cffff-0.036
Token
Token ID198
Feature activation-6.70e-01

Pos logit contributions
 condem+0.486
terms+0.405
 newsp+0.401
 constitu+0.394
cffff+0.383
Neg logit contributions
arching-0.251
toe-0.222
adia-0.221
hang-0.205
estine-0.205
Token"
Token ID1
Feature activation-1.03e+00

Pos logit contributions
 condem+0.766
terms+0.640
 newsp+0.638
 constitu+0.627
cffff+0.609
Neg logit contributions
arching-0.365
toe-0.322
adia-0.321
estine-0.296
hang-0.295
TokenBoth
Token ID10265
Feature activation-3.50e+00

Pos logit contributions
 condem+0.997
 newsp+0.796
terms+0.794
 constitu+0.776
cffff+0.748
Neg logit contributions
arching-0.753
toe-0.687
adia-0.683
hang-0.648
estine-0.647
Token of
Token ID286
Feature activation-9.63e-01

Pos logit contributions
 condem+3.553
 newsp+2.876
terms+2.852
 constitu+2.797
cffff+2.656
Neg logit contributions
arching-2.387
toe-2.173
adia-2.138
hang-2.049
estine-2.029
Token them
Token ID606
Feature activation-1.01e+00

Pos logit contributions
 condem+1.061
 newsp+0.877
terms+0.870
 constitu+0.857
cffff+0.830
Neg logit contributions
arching-0.572
toe-0.515
adia-0.508
hang-0.477
inion-0.476
Token love
Token ID1842
Feature activation+2.93e-01

Pos logit contributions
 condem+1.030
 newsp+0.828
terms+0.825
 constitu+0.809
cffff+0.777
Neg logit contributions
arching-0.693
toe-0.626
adia-0.621
estine-0.587
hang-0.586
Token it
Token ID340
Feature activation-4.31e-01

Pos logit contributions
arching+0.187
toe+0.169
adia+0.168
hang+0.158
estine+0.158
Neg logit contributions
 condem-0.312
 newsp-0.254
terms-0.252
 constitu-0.249
cffff-0.241
Token and
Token ID290
Feature activation-1.72e-01

Pos logit contributions
 condem+0.484
 newsp+0.400
terms+0.398
 constitu+0.391
cffff+0.380
Neg logit contributions
arching-0.250
toe-0.222
adia-0.220
hang-0.207
inion-0.206
Token email
Token ID3053
Feature activation+1.39e+00

Pos logit contributions
 condem+0.468
terms+0.385
 newsp+0.379
 constitu+0.375
cffff+0.362
Neg logit contributions
arching-0.297
toe-0.270
adia-0.269
estine-0.256
hang-0.254
Token and
Token ID290
Feature activation+6.38e-01

Pos logit contributions
arching+0.791
toe+0.704
adia+0.695
hang+0.655
estine+0.651
Neg logit contributions
 condem-1.580
 newsp-1.305
terms-1.302
 constitu-1.279
cffff-1.239
Token never
Token ID1239
Feature activation+1.58e+00

Pos logit contributions
arching+0.437
toe+0.402
adia+0.398
hang+0.380
zel+0.372
Neg logit contributions
 condem-0.637
terms-0.518
 newsp-0.506
 constitu-0.500
cffff-0.477
Token miss
Token ID2051
Feature activation-4.23e-01

Pos logit contributions
arching+0.952
toe+0.856
adia+0.847
hang+0.795
estine+0.791
Neg logit contributions
 condem-1.742
 newsp-1.429
terms-1.421
 constitu-1.399
cffff-1.357
Token our
Token ID674
Feature activation-7.31e-01

Pos logit contributions
 condem+0.470
terms+0.388
 newsp+0.388
 constitu+0.379
cffff+0.370
Neg logit contributions
arching-0.247
toe-0.221
adia-0.217
hang-0.206
mat-0.204
Token top
Token ID1353
Feature activation-4.18e+00

Pos logit contributions
 condem+0.778
 newsp+0.627
terms+0.621
 constitu+0.611
cffff+0.599
Neg logit contributions
arching-0.479
toe-0.432
adia-0.430
hang-0.407
inion-0.403
Token stories
Token ID3923
Feature activation+2.97e+00

Pos logit contributions
 condem+3.574
terms+2.820
 newsp+2.750
 constitu+2.689
cffff+2.518
Neg logit contributions
arching-3.508
toe-3.226
adia-3.185
hang-3.079
leon-3.039
Token Free
Token ID3232
Feature activation-2.79e-01

Pos logit contributions
arching+1.697
toe+1.548
adia+1.471
inion+1.469
mat+1.432
Neg logit contributions
 condem-3.321
 newsp-2.822
 constitu-2.719
terms-2.687
cffff-2.637
Token Sign
Token ID5865
Feature activation-4.77e-01

Pos logit contributions
 condem+0.317
 newsp+0.262
terms+0.260
 constitu+0.257
cffff+0.250
Neg logit contributions
arching-0.158
toe-0.142
adia-0.140
hang-0.131
estine-0.130
Token Up
Token ID3205
Feature activation+2.37e-01

Pos logit contributions
 condem+0.480
terms+0.389
 newsp+0.387
 constitu+0.375
cffff+0.365
Neg logit contributions
arching-0.327
toe-0.297
adia-0.295
hang-0.282
mat-0.279
Token
Token ID198
Feature activation-2.11e-01

Pos logit contributions
arching+0.129
toe+0.115
adia+0.114
estine+0.106
hang+0.106
Neg logit contributions
 condem-0.276
 newsp-0.229
terms-0.227
 constitu-0.224
cffff-0.218
Token,
Token ID11
Feature activation-3.43e-01

Pos logit contributions
arching+0.715
toe+0.641
adia+0.633
hang+0.594
estine+0.592
Neg logit contributions
 condem-1.347
 newsp-1.108
terms-1.102
 constitu-1.084
cffff-1.052
Token but
Token ID475
Feature activation-5.22e-01

Pos logit contributions
 condem+0.364
 newsp+0.296
terms+0.294
 constitu+0.290
cffff+0.281
Neg logit contributions
arching-0.220
toe-0.199
adia-0.197
hang-0.186
estine-0.185
Token Obama
Token ID2486
Feature activation-1.50e+00

Pos logit contributions
 condem+0.535
 newsp+0.432
terms+0.429
 constitu+0.422
cffff+0.409
Neg logit contributions
arching-0.354
toe-0.323
adia-0.320
hang-0.303
estine-0.302
Token actually
Token ID1682
Feature activation+5.45e-01

Pos logit contributions
 condem+1.546
 newsp+1.247
terms+1.239
 constitu+1.218
cffff+1.175
Neg logit contributions
arching-1.009
toe-0.915
adia-0.905
hang-0.858
estine-0.853
Token seems
Token ID2331
Feature activation-1.78e+00

Pos logit contributions
arching+0.368
toe+0.338
adia+0.335
mat+0.318
hang+0.318
Neg logit contributions
 condem-0.555
 newsp-0.450
terms-0.446
 constitu-0.439
cffff-0.428
Token pleased
Token ID10607
Feature activation-3.39e+00

Pos logit contributions
 condem+1.803
 newsp+1.446
terms+1.432
 constitu+1.415
cffff+1.348
Neg logit contributions
arching-1.249
toe-1.128
adia-1.123
hang-1.059
estine-1.057
Token:
Token ID25
Feature activation-8.83e-01

Pos logit contributions
 condem+3.313
 newsp+2.648
terms+2.628
 constitu+2.572
cffff+2.475
Neg logit contributions
arching-2.459
toe-2.235
adia-2.234
hang-2.114
estine-2.098
Token He
Token ID679
Feature activation-1.85e-01

Pos logit contributions
 condem+0.924
 newsp+0.749
terms+0.744
 constitu+0.732
cffff+0.709
Neg logit contributions
arching-0.582
toe-0.528
adia-0.523
hang-0.494
estine-0.492
Token announced
Token ID3414
Feature activation+3.30e-01

Pos logit contributions
 condem+0.190
 newsp+0.153
terms+0.152
 constitu+0.150
cffff+0.145
Neg logit contributions
arching-0.125
toe-0.114
adia-0.113
hang-0.107
estine-0.107
Token early
Token ID1903
Feature activation-2.07e-02

Pos logit contributions
arching+0.211
toe+0.191
adia+0.189
hang+0.178
mat+0.178
Neg logit contributions
 condem-0.350
 newsp-0.285
terms-0.283
 constitu-0.279
cffff-0.271
Token on
Token ID319
Feature activation+8.81e-01

Pos logit contributions
 condem+0.022
 newsp+0.017
terms+0.017
 constitu+0.017
cffff+0.017
Neg logit contributions
arching-0.014
toe-0.012
adia-0.012
hang-0.012
estine-0.012