TOP ACTIVATIONS
MAX = 679.152

I â m sorry Jane . I
That â s a special necklace ,
That â s right , Emma .
That â s the gas from the
It â s almost just as beautiful
It â s a magic book !
her mom my watched in amaz ement as the machine flew
Let â s explore and have some
Let â s have a stretching race
Let â s play a joke .
Let â s not think about the
Let â s go find something else
Please don â t eat me , â
Who â s there ? â
I â m here to give you
It â s a books he lf
It â s a surprise ! â
You â ll see . â
Where â s my jelly ? â
't know the rh in oc eros was wild . Can

MIN ACTIVATIONS
MIN = -268.287

ball . They are clever . They know how to share
a spaceship !" Ben said . " We can fly to
learned that courage is important . They learned that soccer is
but they were more careful . <|endoftext|> Anna and Ben were
go outside now ?" Anna and Ben asked .
. They are happy friends . <|endoftext|> L ily and Tom
Can we go outside now ?" Anna and Ben asked .
. They ate and ate until their bell ies were full
. They smiled and laughed . They had a great adventure
the office . He looked at the computer and saw that
that John was still there . Mark smiled , because he
with you ?" Anna says . Ben opens the
play outside again . He hoped they could be happy and
, duck !" Mia says . The duck smiles
. She gets a box . She gets some paper .
have fun . <|endoftext|> Ben likes to play with his toy
He gave her the ball . " I 'm sorry ,
bad . It hurt me ." Anna said .
or toys , or books ," Max said .
followed them to the office . He looked at the computer

INTERVAL 442.292 - 679.152
CONTAINS 0.021%

Where â s my jelly ? â
Let â s not think about the
children put on their bath ro bes and ran around in
Let â s explore and have some
Let â s jump over it ,

INTERVAL 205.432 - 442.292
CONTAINS 11.437%

a time there was a blue symbol . The symbol was
thought she must have done something wrong and started to cry
She had chosen a good toy and a good price .
her if she made it herself . L ila said no
baby bear . He was very small and afraid . He

INTERVAL -31.428 - 205.432
CONTAINS 42.215%

said goodbye to Andy and thanked him again for the lime
wanted to make more money so he decided to raise some
trying until her toes touched the ground . Suddenly
the iron . " Let 's take it to
thanked Benny for saving him . Benny felt amazing knowing he

INTERVAL -268.287 - -31.428
CONTAINS 46.327%

This watch is mine ! You can 't have it !"
to take the steak away . But John was too proud
boy . They look for their moms . They
he opened the envelope , he noticed that the poem was
feeling troubled ." Edward was so happy .
Token
Token ID129
Feature activation+5.50e+02

Pos logit contributions
 White+0.268
 faded+0.267
Ancient+0.262
 wicked+0.257
chy+0.256
Neg logit contributions
iu-0.311
pieces-0.277
ap-0.256
 interests-0.252
n-0.248
Token
Token ID241
Feature activation+4.50e+02

Pos logit contributions
iu+0.747
pieces+0.733
 interests+0.671
nels+0.583
 occasions+0.582
Neg logit contributions
chy-0.714
 White-0.705
 whispered-0.684
 meant-0.676
 faded-0.669
TokenI
Token ID40
Feature activation+3.76e+02

Pos logit contributions
iu+0.898
pieces+0.826
ric+0.757
 interests+0.750
nels+0.730
Neg logit contributions
 faded-0.342
chy-0.337
 White-0.328
 meant-0.323
 whispered-0.314
Tokenâ
Token ID22940
Feature activation+4.22e+02

Pos logit contributions
iu+0.456
pieces+0.419
 interests+0.403
ric+0.381
 Amen+0.353
Neg logit contributions
 White-0.541
 faded-0.518
chy-0.510
Ancient-0.492
Dead-0.487
Token
Token ID26391
Feature activation+4.29e+02

Pos logit contributions
iu+0.520
pieces+0.454
ric+0.393
 interests+0.385
nels+0.363
Neg logit contributions
 faded-0.647
chy-0.641
 White-0.631
 meant-0.625
 whispered-0.620
Token
Token ID8151
Feature activation+4.90e+02

Pos logit contributions
iu+0.694
pieces+0.630
ric+0.558
 interests+0.555
nels+0.533
Neg logit contributions
 faded-0.482
 White-0.478
chy-0.475
 whispered-0.462
 meant-0.461
Tokenm
Token ID76
Feature activation+3.82e+02

Pos logit contributions
ric+0.710
pieces+0.686
iu+0.646
 interests+0.601
n+0.552
Neg logit contributions
 faded-0.590
 White-0.561
Dead-0.520
 whispered-0.515
 wicked-0.509
Token sorry
Token ID7926
Feature activation+3.86e+02

Pos logit contributions
iu+0.599
ric+0.565
pieces+0.536
 interests+0.484
 Majesty+0.471
Neg logit contributions
 faded-0.456
chy-0.412
Dead-0.411
Ancient-0.406
mail-0.403
Token Jane
Token ID12091
Feature activation+3.74e+02

Pos logit contributions
iu+0.658
pieces+0.595
 interests+0.549
ric+0.545
nels+0.523
Neg logit contributions
 faded-0.402
chy-0.387
 White-0.382
 whispered-0.372
 meant-0.366
Token.
Token ID13
Feature activation+3.09e+02

Pos logit contributions
iu+0.507
pieces+0.477
 interests+0.431
ric+0.427
 Amen+0.393
Neg logit contributions
Ancient-0.464
Dead-0.462
 Blood-0.438
 White-0.420
Factory-0.419
Token I
Token ID314
Feature activation+3.13e+02

Pos logit contributions
iu+0.416
pieces+0.373
ric+0.349
 interests+0.328
 Amen+0.326
Neg logit contributions
chy-0.446
 faded-0.442
 meant-0.401
 spilled-0.400
 whispered-0.398
Token
Token ID129
Feature activation+5.50e+02

Pos logit contributions
 faded+0.205
chy+0.201
 White+0.197
 whispered+0.193
 meant+0.192
Neg logit contributions
iu-0.404
pieces-0.367
ric-0.337
 interests-0.332
nels-0.322
Token
Token ID241
Feature activation+4.50e+02

Pos logit contributions
iu+0.747
pieces+0.733
 interests+0.671
nels+0.583
 occasions+0.582
Neg logit contributions
chy-0.714
 White-0.705
 whispered-0.684
 meant-0.676
 faded-0.669
TokenThat
Token ID2504
Feature activation+3.64e+02

Pos logit contributions
iu+0.898
pieces+0.826
ric+0.757
 interests+0.750
nels+0.730
Neg logit contributions
 faded-0.342
chy-0.337
 White-0.328
 meant-0.323
 whispered-0.314
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
iu+0.537
pieces+0.492
ric+0.457
 interests+0.448
nels+0.425
Neg logit contributions
 faded-0.447
chy-0.431
 White-0.422
 whispered-0.416
Dead-0.400
Token
Token ID26391
Feature activation+4.25e+02

Pos logit contributions
iu+0.471
pieces+0.401
ric+0.346
 interests+0.333
nels+0.312
Neg logit contributions
 faded-0.693
chy-0.677
 White-0.673
 whispered-0.660
 meant-0.655
Token
Token ID8151
Feature activation+4.87e+02

Pos logit contributions
iu+0.683
pieces+0.615
ric+0.557
 interests+0.547
nels+0.524
Neg logit contributions
 faded-0.492
chy-0.486
 White-0.473
 meant-0.468
 whispered-0.463
Tokens
Token ID82
Feature activation+3.94e+02

Pos logit contributions
ric+0.695
pieces+0.670
iu+0.615
 interests+0.580
n+0.547
Neg logit contributions
 faded-0.612
 White-0.595
 wicked-0.565
 whispered-0.552
Dead-0.547
Token a
Token ID257
Feature activation+3.73e+02

Pos logit contributions
iu+0.573
pieces+0.518
ric+0.487
 interests+0.465
nels+0.444
Neg logit contributions
 faded-0.510
chy-0.488
 White-0.483
 whispered-0.478
 meant-0.462
Token special
Token ID2041
Feature activation+3.79e+02

Pos logit contributions
iu+0.596
pieces+0.560
 interests+0.502
ric+0.496
nels+0.483
Neg logit contributions
 faded-0.390
 White-0.385
 meant-0.377
chy-0.376
 whispered-0.372
Token necklace
Token ID38987
Feature activation+3.65e+02

Pos logit contributions
iu+0.544
pieces+0.484
 interests+0.459
 occasions+0.441
ric+0.441
Neg logit contributions
chy-0.471
 faded-0.470
 White-0.467
 whispered-0.446
 meant-0.444
Token,
Token ID11
Feature activation+3.05e+02

Pos logit contributions
iu+0.470
pieces+0.446
 interests+0.398
nels+0.380
 occasions+0.374
Neg logit contributions
Dead-0.468
 faded-0.460
 White-0.456
Ancient-0.453
 Blood-0.440
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenThat
Token ID2504
Feature activation+3.71e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.20e+02

Pos logit contributions
iu+0.554
pieces+0.511
ric+0.470
 interests+0.461
nels+0.437
Neg logit contributions
 faded-0.444
chy-0.425
 White-0.424
 whispered-0.418
Dead-0.398
Token
Token ID26391
Feature activation+4.26e+02

Pos logit contributions
iu+0.455
pieces+0.382
ric+0.332
 interests+0.312
nels+0.294
Neg logit contributions
 faded-0.718
chy-0.693
 White-0.692
 whispered-0.684
 meant-0.673
Token
Token ID8151
Feature activation+4.86e+02

Pos logit contributions
iu+0.660
pieces+0.592
ric+0.541
 interests+0.524
nels+0.502
Neg logit contributions
 faded-0.519
chy-0.511
 meant-0.493
 White-0.492
 whispered-0.488
Tokens
Token ID82
Feature activation+3.93e+02

Pos logit contributions
ric+0.703
pieces+0.679
iu+0.626
 interests+0.581
 disl+0.561
Neg logit contributions
 faded-0.600
 White-0.581
 whispered-0.555
 wicked-0.549
Dead-0.537
Token right
Token ID826
Feature activation+3.73e+02

Pos logit contributions
iu+0.573
pieces+0.519
ric+0.489
 interests+0.465
nels+0.444
Neg logit contributions
 faded-0.505
chy-0.483
 White-0.477
 whispered-0.477
 meant-0.460
Token,
Token ID11
Feature activation+3.28e+02

Pos logit contributions
iu+0.524
ric+0.463
pieces+0.440
nels+0.438
 interests+0.434
Neg logit contributions
 faded-0.465
chy-0.456
 White-0.447
 whispered-0.435
 spoke-0.424
Token Emma
Token ID18966
Feature activation+3.53e+02

Pos logit contributions
iu+0.458
pieces+0.418
ric+0.366
 interests+0.364
nels+0.349
Neg logit contributions
 faded-0.431
chy-0.430
 White-0.415
 whispered-0.404
 meant-0.400
Token.
Token ID13
Feature activation+2.95e+02

Pos logit contributions
iu+0.416
pieces+0.395
ric+0.381
 interests+0.370
 Majesty+0.365
Neg logit contributions
Dead-0.461
 faded-0.455
 Blood-0.442
Ancient-0.437
mail-0.422
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenThat
Token ID2504
Feature activation+3.71e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.20e+02

Pos logit contributions
iu+0.554
pieces+0.511
ric+0.470
 interests+0.461
nels+0.437
Neg logit contributions
 faded-0.444
chy-0.425
 White-0.424
 whispered-0.418
Dead-0.398
Token
Token ID26391
Feature activation+4.26e+02

Pos logit contributions
iu+0.455
pieces+0.382
ric+0.332
 interests+0.312
nels+0.294
Neg logit contributions
 faded-0.718
chy-0.693
 White-0.692
 whispered-0.684
 meant-0.673
Token
Token ID8151
Feature activation+4.86e+02

Pos logit contributions
iu+0.660
pieces+0.592
ric+0.541
 interests+0.524
nels+0.502
Neg logit contributions
 faded-0.519
chy-0.511
 meant-0.493
 White-0.492
 whispered-0.488
Tokens
Token ID82
Feature activation+3.93e+02

Pos logit contributions
ric+0.703
pieces+0.679
iu+0.626
 interests+0.581
 disl+0.561
Neg logit contributions
 faded-0.600
 White-0.581
 whispered-0.555
 wicked-0.549
Dead-0.537
Token the
Token ID262
Feature activation+3.80e+02

Pos logit contributions
iu+0.573
pieces+0.519
ric+0.489
 interests+0.465
nels+0.444
Neg logit contributions
 faded-0.505
chy-0.483
 White-0.477
 whispered-0.477
 meant-0.460
Token gas
Token ID3623
Feature activation+3.77e+02

Pos logit contributions
iu+0.611
pieces+0.555
ric+0.520
 interests+0.499
nels+0.471
Neg logit contributions
 faded-0.432
chy-0.410
 White-0.405
 whispered-0.404
 meant-0.398
Token from
Token ID422
Feature activation+3.32e+02

Pos logit contributions
iu+0.543
ric+0.514
pieces+0.476
 interests+0.458
nels+0.456
Neg logit contributions
 faded-0.449
Dead-0.441
 White-0.426
chy-0.422
Ancient-0.421
Token the
Token ID262
Feature activation+3.33e+02

Pos logit contributions
iu+0.426
pieces+0.376
ric+0.343
 interests+0.330
 occasions+0.310
Neg logit contributions
chy-0.470
 faded-0.459
 White-0.439
Ancient-0.433
 meant-0.432
Token
Token ID129
Feature activation+5.50e+02

Pos logit contributions
 faded+0.304
chy+0.300
 White+0.290
 meant+0.289
 whispered+0.283
Neg logit contributions
iu-0.364
pieces-0.326
ric-0.299
 interests-0.290
nels-0.276
Token
Token ID241
Feature activation+4.50e+02

Pos logit contributions
iu+0.747
pieces+0.733
 interests+0.671
nels+0.583
 occasions+0.582
Neg logit contributions
chy-0.714
 White-0.705
 whispered-0.684
 meant-0.676
 faded-0.669
TokenIt
Token ID1026
Feature activation+3.74e+02

Pos logit contributions
iu+0.898
pieces+0.826
ric+0.757
 interests+0.750
nels+0.730
Neg logit contributions
 faded-0.342
chy-0.337
 White-0.328
 meant-0.323
 whispered-0.314
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
iu+0.431
pieces+0.430
ric+0.389
 interests+0.379
nels+0.342
Neg logit contributions
 faded-0.539
 White-0.513
Dead-0.501
Ancient-0.493
 whispered-0.487
Token
Token ID26391
Feature activation+4.25e+02

Pos logit contributions
iu+0.518
pieces+0.449
ric+0.395
 interests+0.379
nels+0.358
Neg logit contributions
 faded-0.646
chy-0.635
 White-0.624
 whispered-0.614
 meant-0.614
Token
Token ID8151
Feature activation+4.86e+02

Pos logit contributions
iu+0.661
pieces+0.592
ric+0.538
 interests+0.525
nels+0.503
Neg logit contributions
 faded-0.514
chy-0.507
 White-0.493
 meant-0.490
 whispered-0.483
Tokens
Token ID82
Feature activation+3.94e+02

Pos logit contributions
ric+0.712
pieces+0.667
iu+0.606
 interests+0.564
n+0.552
Neg logit contributions
 faded-0.601
 White-0.575
 wicked-0.555
Dead-0.546
 whispered-0.535
Token almost
Token ID2048
Feature activation+3.81e+02

Pos logit contributions
iu+0.533
ric+0.495
pieces+0.489
 interests+0.455
nels+0.420
Neg logit contributions
 faded-0.509
 White-0.484
 whispered-0.482
Dead-0.480
Ancient-0.479
Token just
Token ID655
Feature activation+3.60e+02

Pos logit contributions
iu+0.497
pieces+0.439
ric+0.421
 interests+0.407
 disl+0.394
Neg logit contributions
 White-0.518
Ancient-0.503
Dead-0.487
Doctor-0.472
mail-0.468
Token as
Token ID355
Feature activation+3.26e+02

Pos logit contributions
iu+0.488
pieces+0.447
ric+0.416
 interests+0.410
 disl+0.373
Neg logit contributions
 White-0.470
Ancient-0.458
 faded-0.442
Dead-0.441
chy-0.440
Token beautiful
Token ID4950
Feature activation+3.41e+02

Pos logit contributions
iu+0.443
pieces+0.410
 interests+0.385
ric+0.360
 disl+0.347
Neg logit contributions
chy-0.378
 depress-0.366
 faded-0.366
Dead-0.362
Ancient-0.355
Token
Token ID129
Feature activation+5.50e+02

Pos logit contributions
 faded+0.186
chy+0.179
 White+0.177
 meant+0.174
 whispered+0.173
Neg logit contributions
iu-0.434
pieces-0.398
ric-0.371
 interests-0.362
nels-0.351
Token
Token ID241
Feature activation+4.50e+02

Pos logit contributions
iu+0.747
pieces+0.733
 interests+0.671
nels+0.583
 occasions+0.582
Neg logit contributions
chy-0.714
 White-0.705
 whispered-0.684
 meant-0.676
 faded-0.669
TokenIt
Token ID1026
Feature activation+3.74e+02

Pos logit contributions
iu+0.898
pieces+0.826
ric+0.757
 interests+0.750
nels+0.730
Neg logit contributions
 faded-0.342
chy-0.337
 White-0.328
 meant-0.323
 whispered-0.314
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
iu+0.431
pieces+0.430
ric+0.389
 interests+0.379
nels+0.342
Neg logit contributions
 faded-0.539
 White-0.513
Dead-0.501
Ancient-0.493
 whispered-0.487
Token
Token ID26391
Feature activation+4.25e+02

Pos logit contributions
iu+0.518
pieces+0.449
ric+0.395
 interests+0.379
nels+0.358
Neg logit contributions
 faded-0.646
chy-0.635
 White-0.624
 whispered-0.614
 meant-0.614
Token
Token ID8151
Feature activation+4.86e+02

Pos logit contributions
iu+0.661
pieces+0.592
ric+0.538
 interests+0.525
nels+0.503
Neg logit contributions
 faded-0.514
chy-0.507
 White-0.493
 meant-0.490
 whispered-0.483
Tokens
Token ID82
Feature activation+3.94e+02

Pos logit contributions
ric+0.712
pieces+0.667
iu+0.606
 interests+0.564
n+0.552
Neg logit contributions
 faded-0.601
 White-0.575
 wicked-0.555
Dead-0.546
 whispered-0.535
Token a
Token ID257
Feature activation+3.72e+02

Pos logit contributions
iu+0.533
ric+0.495
pieces+0.489
 interests+0.455
nels+0.420
Neg logit contributions
 faded-0.509
 White-0.484
 whispered-0.482
Dead-0.480
Ancient-0.479
Token magic
Token ID5536
Feature activation+3.75e+02

Pos logit contributions
iu+0.581
pieces+0.549
ric+0.493
 interests+0.489
nels+0.470
Neg logit contributions
 meant-0.384
 faded-0.384
 White-0.380
chy-0.372
 whispered-0.371
Token book
Token ID1492
Feature activation+3.61e+02

Pos logit contributions
iu+0.529
pieces+0.492
 interests+0.458
 occasions+0.458
ric+0.437
Neg logit contributions
 White-0.446
Dead-0.432
chy-0.428
Ancient-0.426
 faded-0.425
Token!
Token ID0
Feature activation+2.97e+02

Pos logit contributions
pieces+0.358
iu+0.321
 occasions+0.305
 interests+0.302
n+0.281
Neg logit contributions
Dead-0.522
 White-0.511
Pr-0.511
Spider-0.504
 Blood-0.500
Token her
Token ID607
Feature activation+4.16e+02

Pos logit contributions
chy+0.199
 White+0.198
 faded+0.197
 whispered+0.180
 meant+0.179
Neg logit contributions
iu-0.416
pieces-0.394
 interests-0.359
ric-0.345
nels-0.330
Token mom
Token ID1995
Feature activation+3.73e+02

Pos logit contributions
iu+0.637
pieces+0.575
ric+0.518
 interests+0.515
nels+0.476
Neg logit contributions
 faded-0.502
chy-0.490
 White-0.488
 meant-0.482
 whispered-0.472
Tokenmy
Token ID1820
Feature activation+3.93e+02

Pos logit contributions
iu+0.560
pieces+0.553
ric+0.502
 interests+0.463
 Amen+0.448
Neg logit contributions
Dead-0.415
 faded-0.410
Ancient-0.409
chy-0.399
 Blood-0.394
Token watched
Token ID7342
Feature activation+3.15e+02

Pos logit contributions
iu+0.527
pieces+0.522
ric+0.483
 spend+0.453
 Amen+0.450
Neg logit contributions
Dead-0.474
Ancient-0.466
 Blood-0.451
 faded-0.441
Spider-0.427
Token in
Token ID287
Feature activation+3.02e+02

Pos logit contributions
iu+0.467
pieces+0.429
 interests+0.357
ric+0.355
nels+0.340
Neg logit contributions
 faded-0.403
chy-0.399
 White-0.383
 meant-0.382
 blocked-0.381
Token amaz
Token ID40642
Feature activation+4.75e+02

Pos logit contributions
iu+0.467
pieces+0.430
 interests+0.386
ric+0.379
 occasions+0.359
Neg logit contributions
chy-0.366
 faded-0.351
 White-0.345
 meant-0.336
 whispered-0.335
Tokenement
Token ID972
Feature activation+3.38e+02

Pos logit contributions
iu+0.986
ric+0.924
pieces+0.865
 interests+0.830
nels+0.818
Neg logit contributions
 answered-0.340
 faded-0.323
 meant-0.301
 shone-0.272
 said-0.265
Token as
Token ID355
Feature activation+2.98e+02

Pos logit contributions
iu+0.582
pieces+0.540
ric+0.483
 interests+0.478
nels+0.457
Neg logit contributions
 faded-0.344
chy-0.337
 White-0.331
 meant-0.321
 whispered-0.318
Token the
Token ID262
Feature activation+3.02e+02

Pos logit contributions
iu+0.468
pieces+0.456
 interests+0.393
ric+0.367
 occasions+0.362
Neg logit contributions
chy-0.319
 faded-0.307
 White-0.300
Dead-0.298
 whispered-0.293
Token machine
Token ID4572
Feature activation+3.05e+02

Pos logit contributions
iu+0.476
pieces+0.434
 interests+0.377
ric+0.366
nels+0.356
Neg logit contributions
chy-0.354
 faded-0.348
 White-0.340
 meant-0.335
 whispered-0.332
Token flew
Token ID13112
Feature activation+2.81e+02

Pos logit contributions
pieces+0.413
iu+0.359
n+0.344
 occasions+0.316
 wait+0.313
Neg logit contributions
 Blood-0.400
Dead-0.372
Spider-0.372
Pr-0.362
Ancient-0.360
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenLet
Token ID5756
Feature activation+3.59e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.609
ric+0.521
pieces+0.511
 Amen+0.469
 interests+0.461
Neg logit contributions
 meant-0.426
 whispered-0.402
 faded-0.395
chy-0.386
 spoke-0.378
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.457
pieces+0.393
ric+0.335
 interests+0.329
nels+0.307
Neg logit contributions
 faded-0.649
chy-0.647
 White-0.636
 meant-0.632
 whispered-0.626
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.679
pieces+0.610
 interests+0.543
ric+0.541
nels+0.519
Neg logit contributions
 White-0.450
 faded-0.449
chy-0.445
 meant-0.433
 whispered-0.431
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
pieces+0.744
iu+0.728
ric+0.721
 interests+0.642
n+0.593
Neg logit contributions
 faded-0.517
 White-0.488
chy-0.469
 whispered-0.458
Dead-0.450
Token explore
Token ID7301
Feature activation+3.61e+02

Pos logit contributions
iu+0.591
pieces+0.501
ric+0.440
 interests+0.430
 Amen+0.419
Neg logit contributions
 meant-0.487
 faded-0.478
chy-0.476
 whispered-0.471
 White-0.470
Token and
Token ID290
Feature activation+3.16e+02

Pos logit contributions
iu+0.537
pieces+0.507
 interests+0.451
ric+0.448
nels+0.436
Neg logit contributions
 faded-0.436
chy-0.434
 White-0.412
 meant-0.404
 whispered-0.396
Token have
Token ID423
Feature activation+3.10e+02

Pos logit contributions
iu+0.554
pieces+0.466
ric+0.452
 interests+0.429
 Amen+0.419
Neg logit contributions
 whispered-0.349
 faded-0.339
 meant-0.338
 White-0.334
chy-0.332
Token some
Token ID617
Feature activation+3.03e+02

Pos logit contributions
iu+0.509
pieces+0.467
 interests+0.422
ric+0.415
nels+0.397
Neg logit contributions
chy-0.340
 faded-0.340
 White-0.332
 meant-0.321
 whispered-0.316
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenLet
Token ID5756
Feature activation+3.59e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.609
ric+0.521
pieces+0.511
 Amen+0.469
 interests+0.461
Neg logit contributions
 meant-0.426
 whispered-0.402
 faded-0.395
chy-0.386
 spoke-0.378
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.457
pieces+0.393
ric+0.335
 interests+0.329
nels+0.307
Neg logit contributions
 faded-0.649
chy-0.647
 White-0.636
 meant-0.632
 whispered-0.626
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.679
pieces+0.610
 interests+0.543
ric+0.541
nels+0.519
Neg logit contributions
 White-0.450
 faded-0.449
chy-0.445
 meant-0.433
 whispered-0.431
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
pieces+0.744
iu+0.728
ric+0.721
 interests+0.642
n+0.593
Neg logit contributions
 faded-0.517
 White-0.488
chy-0.469
 whispered-0.458
Dead-0.450
Token have
Token ID423
Feature activation+3.54e+02

Pos logit contributions
iu+0.591
pieces+0.501
ric+0.440
 interests+0.430
 Amen+0.419
Neg logit contributions
 meant-0.487
 faded-0.478
chy-0.476
 whispered-0.471
 White-0.470
Token a
Token ID257
Feature activation+3.45e+02

Pos logit contributions
iu+0.573
pieces+0.523
 interests+0.471
ric+0.464
nels+0.444
Neg logit contributions
chy-0.398
 faded-0.396
 White-0.386
 meant-0.375
 whispered-0.370
Token stretching
Token ID20880
Feature activation+3.45e+02

Pos logit contributions
iu+0.524
pieces+0.493
 interests+0.436
ric+0.424
nels+0.413
Neg logit contributions
 faded-0.396
 White-0.395
 meant-0.391
chy-0.387
 whispered-0.375
Token race
Token ID3234
Feature activation+3.22e+02

Pos logit contributions
iu+0.564
pieces+0.511
ric+0.468
 interests+0.460
nels+0.441
Neg logit contributions
 faded-0.381
chy-0.376
 White-0.371
 meant-0.362
 whispered-0.359
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenLet
Token ID5756
Feature activation+3.59e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.609
ric+0.521
pieces+0.511
 Amen+0.469
 interests+0.461
Neg logit contributions
 meant-0.426
 whispered-0.402
 faded-0.395
chy-0.386
 spoke-0.378
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.457
pieces+0.393
ric+0.335
 interests+0.329
nels+0.307
Neg logit contributions
 faded-0.649
chy-0.647
 White-0.636
 meant-0.632
 whispered-0.626
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.679
pieces+0.610
 interests+0.543
ric+0.541
nels+0.519
Neg logit contributions
 White-0.450
 faded-0.449
chy-0.445
 meant-0.433
 whispered-0.431
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
pieces+0.744
iu+0.728
ric+0.721
 interests+0.642
n+0.593
Neg logit contributions
 faded-0.517
 White-0.488
chy-0.469
 whispered-0.458
Dead-0.450
Token play
Token ID711
Feature activation+3.63e+02

Pos logit contributions
iu+0.591
pieces+0.501
ric+0.440
 interests+0.430
 Amen+0.419
Neg logit contributions
 meant-0.487
 faded-0.478
chy-0.476
 whispered-0.471
 White-0.470
Token a
Token ID257
Feature activation+3.44e+02

Pos logit contributions
iu+0.554
pieces+0.523
 interests+0.455
ric+0.451
nels+0.450
Neg logit contributions
 faded-0.439
chy-0.430
 White-0.416
 meant-0.411
 whispered-0.400
Token joke
Token ID9707
Feature activation+3.43e+02

Pos logit contributions
iu+0.549
pieces+0.512
ric+0.451
 interests+0.446
nels+0.437
Neg logit contributions
 faded-0.385
 meant-0.380
 White-0.376
chy-0.369
 whispered-0.363
Token.
Token ID13
Feature activation+2.86e+02

Pos logit contributions
pieces+0.384
iu+0.366
nels+0.323
 interests+0.317
 occasions+0.315
Neg logit contributions
 faded-0.516
 White-0.491
Dead-0.487
chy-0.474
Ancient-0.473
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenLet
Token ID5756
Feature activation+3.59e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.609
ric+0.521
pieces+0.511
 Amen+0.469
 interests+0.461
Neg logit contributions
 meant-0.426
 whispered-0.402
 faded-0.395
chy-0.386
 spoke-0.378
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.457
pieces+0.393
ric+0.335
 interests+0.329
nels+0.307
Neg logit contributions
 faded-0.649
chy-0.647
 White-0.636
 meant-0.632
 whispered-0.626
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.679
pieces+0.610
 interests+0.543
ric+0.541
nels+0.519
Neg logit contributions
 White-0.450
 faded-0.449
chy-0.445
 meant-0.433
 whispered-0.431
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
pieces+0.744
iu+0.728
ric+0.721
 interests+0.642
n+0.593
Neg logit contributions
 faded-0.517
 White-0.488
chy-0.469
 whispered-0.458
Dead-0.450
Token not
Token ID407
Feature activation+3.66e+02

Pos logit contributions
iu+0.591
pieces+0.501
ric+0.440
 interests+0.430
 Amen+0.419
Neg logit contributions
 meant-0.487
 faded-0.478
chy-0.476
 whispered-0.471
 White-0.470
Token think
Token ID892
Feature activation+3.44e+02

Pos logit contributions
iu+0.540
pieces+0.476
ric+0.426
 interests+0.413
nels+0.402
Neg logit contributions
 faded-0.475
chy-0.470
 meant-0.461
 White-0.458
 whispered-0.453
Token about
Token ID546
Feature activation+3.17e+02

Pos logit contributions
iu+0.449
pieces+0.428
 interests+0.368
ric+0.357
nels+0.346
Neg logit contributions
 faded-0.473
chy-0.468
 White-0.453
Dead-0.445
 whispered-0.441
Token the
Token ID262
Feature activation+3.17e+02

Pos logit contributions
iu+0.441
pieces+0.398
ric+0.357
 interests+0.351
nels+0.330
Neg logit contributions
 faded-0.430
chy-0.428
 White-0.410
 meant-0.404
 whispered-0.403
Token
Token ID129
Feature activation+5.50e+02

Pos logit contributions
Ancient+0.420
 faded+0.415
Dead+0.412
 whispered+0.405
 smoked+0.404
Neg logit contributions
iu-0.182
 Amen-0.157
 interests-0.139
 Wed-0.130
n-0.120
Token
Token ID241
Feature activation+4.50e+02

Pos logit contributions
iu+0.747
pieces+0.733
 interests+0.671
nels+0.583
 occasions+0.582
Neg logit contributions
chy-0.714
 White-0.705
 whispered-0.684
 meant-0.676
 faded-0.669
TokenLet
Token ID5756
Feature activation+3.48e+02

Pos logit contributions
iu+0.898
pieces+0.826
ric+0.757
 interests+0.750
nels+0.730
Neg logit contributions
 faded-0.342
chy-0.337
 White-0.328
 meant-0.323
 whispered-0.314
Tokenâ
Token ID22940
Feature activation+3.98e+02

Pos logit contributions
iu+0.589
pieces+0.506
ric+0.496
 interests+0.456
 Amen+0.451
Neg logit contributions
 meant-0.403
 whispered-0.385
 faded-0.381
chy-0.376
 White-0.362
Token
Token ID26391
Feature activation+4.10e+02

Pos logit contributions
iu+0.483
pieces+0.420
 interests+0.362
ric+0.361
nels+0.338
Neg logit contributions
chy-0.604
 faded-0.599
 meant-0.597
 White-0.591
 whispered-0.586
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.729
pieces+0.662
 interests+0.592
ric+0.585
 occasions+0.574
Neg logit contributions
 White-0.395
 faded-0.383
 whispered-0.377
chy-0.377
 meant-0.375
Tokens
Token ID82
Feature activation+3.82e+02

Pos logit contributions
pieces+0.732
iu+0.719
ric+0.713
 interests+0.629
n+0.591
Neg logit contributions
 faded-0.531
 White-0.496
chy-0.483
 whispered-0.460
Dead-0.460
Token go
Token ID467
Feature activation+3.50e+02

Pos logit contributions
iu+0.606
pieces+0.523
ric+0.463
 interests+0.453
nels+0.435
Neg logit contributions
 meant-0.466
chy-0.459
 faded-0.459
 whispered-0.454
 White-0.448
Token find
Token ID1064
Feature activation+3.32e+02

Pos logit contributions
iu+0.591
pieces+0.531
ric+0.484
 interests+0.477
nels+0.453
Neg logit contributions
 faded-0.376
chy-0.373
 White-0.366
 meant-0.360
 whispered-0.359
Token something
Token ID1223
Feature activation+3.23e+02

Pos logit contributions
iu+0.514
pieces+0.467
ric+0.413
 interests+0.413
nels+0.396
Neg logit contributions
 faded-0.401
chy-0.397
 White-0.381
 meant-0.376
 whispered-0.372
Token else
Token ID2073
Feature activation+3.45e+02

Pos logit contributions
iu+0.423
pieces+0.412
 interests+0.376
nels+0.351
ric+0.349
Neg logit contributions
 White-0.425
 faded-0.403
Dead-0.386
Ancient-0.376
Doctor-0.374
Token
Token ID241
Feature activation+4.50e+02

Pos logit contributions
iu+0.747
pieces+0.733
 interests+0.671
nels+0.583
 occasions+0.582
Neg logit contributions
chy-0.714
 White-0.705
 whispered-0.684
 meant-0.676
 faded-0.669
TokenPlease
Token ID5492
Feature activation+3.86e+02

Pos logit contributions
iu+0.898
pieces+0.826
ric+0.757
 interests+0.750
nels+0.730
Neg logit contributions
 faded-0.342
chy-0.337
 White-0.328
 meant-0.323
 whispered-0.314
Token don
Token ID836
Feature activation+3.74e+02

Pos logit contributions
iu+0.494
pieces+0.439
ric+0.409
 interests+0.407
nels+0.369
Neg logit contributions
 faded-0.556
chy-0.532
Dead-0.518
 White-0.517
 whispered-0.500
Tokenâ
Token ID22940
Feature activation+4.06e+02

Pos logit contributions
iu+0.786
pieces+0.723
 interests+0.684
ric+0.676
 Majesty+0.656
Neg logit contributions
 meant-0.264
chy-0.257
 White-0.256
 faded-0.249
 whispered-0.229
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.507
pieces+0.448
ric+0.389
 interests+0.383
nels+0.364
Neg logit contributions
chy-0.612
 faded-0.605
 meant-0.601
 White-0.591
 whispered-0.587
Token
Token ID8151
Feature activation+4.71e+02

Pos logit contributions
iu+0.690
pieces+0.625
ric+0.567
 interests+0.558
nels+0.536
Neg logit contributions
 faded-0.446
chy-0.440
 White-0.431
 meant-0.424
 whispered-0.420
Tokent
Token ID83
Feature activation+3.75e+02

Pos logit contributions
pieces+0.709
iu+0.706
ric+0.682
 interests+0.618
earned+0.592
Neg logit contributions
 faded-0.528
 White-0.499
 whispered-0.481
Dead-0.470
 shone-0.465
Token eat
Token ID4483
Feature activation+3.62e+02

Pos logit contributions
iu+0.484
pieces+0.439
 interests+0.399
ric+0.398
 disl+0.355
Neg logit contributions
 faded-0.531
 White-0.524
chy-0.512
 whispered-0.498
Dead-0.492
Token me
Token ID502
Feature activation+3.60e+02

Pos logit contributions
iu+0.439
pieces+0.392
ric+0.351
 interests+0.341
nels+0.327
Neg logit contributions
 faded-0.550
chy-0.528
 White-0.520
 whispered-0.515
 spoke-0.505
Token,
Token ID11
Feature activation+3.10e+02

Pos logit contributions
iu+0.452
pieces+0.428
 interests+0.400
ric+0.386
earned+0.345
Neg logit contributions
 faded-0.509
 White-0.495
Dead-0.490
Ancient-0.481
chy-0.459
Tokenâ
Token ID22940
Feature activation+3.53e+02

Pos logit contributions
iu+0.445
pieces+0.408
 interests+0.357
ric+0.355
nels+0.340
Neg logit contributions
 faded-0.398
chy-0.391
 White-0.381
 whispered-0.369
 meant-0.366
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenWho
Token ID8241
Feature activation+3.64e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+3.99e+02

Pos logit contributions
iu+0.567
pieces+0.532
 interests+0.487
ric+0.466
nels+0.440
Neg logit contributions
 faded-0.414
chy-0.409
 White-0.406
 whispered-0.375
Dead-0.365
Token
Token ID26391
Feature activation+4.10e+02

Pos logit contributions
iu+0.481
pieces+0.417
ric+0.366
 interests+0.348
nels+0.330
Neg logit contributions
 faded-0.628
chy-0.617
 White-0.609
 meant-0.597
 whispered-0.597
Token
Token ID8151
Feature activation+4.71e+02

Pos logit contributions
iu+0.660
pieces+0.596
ric+0.545
 interests+0.530
nels+0.508
Neg logit contributions
 faded-0.471
chy-0.466
 White-0.445
 meant-0.444
 whispered-0.439
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
ric+0.742
pieces+0.723
iu+0.670
 interests+0.611
n+0.584
Neg logit contributions
 faded-0.526
 White-0.489
 whispered-0.455
 wicked-0.453
chy-0.452
Token there
Token ID612
Feature activation+3.51e+02

Pos logit contributions
iu+0.511
pieces+0.468
ric+0.430
 interests+0.405
nels+0.387
Neg logit contributions
 faded-0.523
chy-0.515
 White-0.502
 whispered-0.489
 meant-0.476
Token?
Token ID30
Feature activation+3.12e+02

Pos logit contributions
ric+0.334
pieces+0.317
iu+0.295
 interests+0.259
nels+0.244
Neg logit contributions
 faded-0.582
Dead-0.537
chy-0.534
 White-0.529
 whispered-0.519
Tokenâ
Token ID22940
Feature activation+3.45e+02

Pos logit contributions
iu+0.398
pieces+0.391
ric+0.352
 interests+0.318
 Amen+0.313
Neg logit contributions
 faded-0.455
chy-0.454
 whispered-0.399
 meant-0.393
 sore-0.391
Token
Token ID26391
Feature activation+3.54e+02

Pos logit contributions
iu+0.318
pieces+0.237
ric+0.235
 disl+0.195
nels+0.186
Neg logit contributions
 faded-0.667
 White-0.620
 led-0.613
chy-0.611
 blocked-0.606
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenI
Token ID40
Feature activation+3.60e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.465
pieces+0.422
 interests+0.395
ric+0.382
 Amen+0.354
Neg logit contributions
 White-0.493
 faded-0.488
chy-0.483
 whispered-0.452
Dead-0.452
Token
Token ID26391
Feature activation+4.08e+02

Pos logit contributions
iu+0.424
pieces+0.357
ric+0.305
 interests+0.287
nels+0.268
Neg logit contributions
 faded-0.700
chy-0.682
 White-0.674
 meant-0.660
 whispered-0.659
Token
Token ID8151
Feature activation+4.67e+02

Pos logit contributions
iu+0.662
pieces+0.599
ric+0.534
 interests+0.530
nels+0.507
Neg logit contributions
 faded-0.458
 White-0.456
chy-0.450
 whispered-0.439
 meant-0.437
Tokenm
Token ID76
Feature activation+3.67e+02

Pos logit contributions
ric+0.670
pieces+0.660
iu+0.626
 interests+0.582
earned+0.536
Neg logit contributions
 faded-0.556
 White-0.528
Dead-0.499
 whispered-0.494
chy-0.483
Token here
Token ID994
Feature activation+3.64e+02

Pos logit contributions
iu+0.558
ric+0.524
pieces+0.504
 interests+0.448
 Majesty+0.438
Neg logit contributions
 faded-0.453
Dead-0.416
chy-0.415
 whispered-0.405
mail-0.399
Token to
Token ID284
Feature activation+3.31e+02

Pos logit contributions
iu+0.428
pieces+0.406
ric+0.374
 interests+0.373
nels+0.339
Neg logit contributions
 faded-0.519
chy-0.497
Dead-0.486
 White-0.479
 whispered-0.475
Token give
Token ID1577
Feature activation+3.10e+02

Pos logit contributions
iu+0.506
pieces+0.432
ric+0.387
 interests+0.380
nels+0.372
Neg logit contributions
 White-0.422
 faded-0.421
chy-0.417
 whispered-0.415
 meant-0.412
Token you
Token ID345
Feature activation+3.10e+02

Pos logit contributions
iu+0.436
pieces+0.389
ric+0.340
 interests+0.340
nels+0.321
Neg logit contributions
 faded-0.423
chy-0.419
 White-0.403
 meant-0.399
 whispered-0.393
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenIt
Token ID1026
Feature activation+3.58e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+3.98e+02

Pos logit contributions
iu+0.483
pieces+0.453
ric+0.417
 interests+0.401
nels+0.382
Neg logit contributions
 faded-0.473
 White-0.447
chy-0.445
 whispered-0.435
Dead-0.432
Token
Token ID26391
Feature activation+4.04e+02

Pos logit contributions
iu+0.443
pieces+0.375
ric+0.328
 interests+0.308
nels+0.292
Neg logit contributions
 faded-0.665
chy-0.650
 White-0.639
 whispered-0.633
 meant-0.630
Token
Token ID8151
Feature activation+4.63e+02

Pos logit contributions
iu+0.652
pieces+0.589
ric+0.531
 interests+0.522
nels+0.501
Neg logit contributions
 faded-0.461
chy-0.455
 White-0.446
 meant-0.439
 whispered-0.435
Tokens
Token ID82
Feature activation+3.76e+02

Pos logit contributions
ric+0.675
pieces+0.646
iu+0.598
 interests+0.557
n+0.527
Neg logit contributions
 faded-0.559
 White-0.537
Dead-0.513
 whispered-0.509
 wicked-0.500
Token a
Token ID257
Feature activation+3.57e+02

Pos logit contributions
iu+0.530
pieces+0.481
ric+0.472
 interests+0.443
nels+0.415
Neg logit contributions
 faded-0.479
chy-0.454
 whispered-0.453
 White-0.452
Dead-0.442
Token books
Token ID3835
Feature activation+3.62e+02

Pos logit contributions
iu+0.557
pieces+0.527
 interests+0.469
ric+0.468
nels+0.451
Neg logit contributions
 faded-0.376
 meant-0.373
 White-0.371
chy-0.366
 whispered-0.360
Tokenhe
Token ID258
Feature activation+3.76e+02

Pos logit contributions
pieces+0.377
iu+0.361
ric+0.337
ap+0.302
nels+0.300
Neg logit contributions
 White-0.519
Pr-0.477
 faded-0.475
 Hen-0.475
Dead-0.473
Tokenlf
Token ID1652
Feature activation+3.88e+02

Pos logit contributions
iu+0.553
pieces+0.496
ric+0.435
 interests+0.435
nels+0.417
Neg logit contributions
 faded-0.481
chy-0.477
 White-0.469
 meant-0.452
 whispered-0.451
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenIt
Token ID1026
Feature activation+3.58e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+3.98e+02

Pos logit contributions
iu+0.483
pieces+0.453
ric+0.417
 interests+0.401
nels+0.382
Neg logit contributions
 faded-0.473
 White-0.447
chy-0.445
 whispered-0.435
Dead-0.432
Token
Token ID26391
Feature activation+4.04e+02

Pos logit contributions
iu+0.443
pieces+0.375
ric+0.328
 interests+0.308
nels+0.292
Neg logit contributions
 faded-0.665
chy-0.650
 White-0.639
 whispered-0.633
 meant-0.630
Token
Token ID8151
Feature activation+4.63e+02

Pos logit contributions
iu+0.652
pieces+0.589
ric+0.531
 interests+0.522
nels+0.501
Neg logit contributions
 faded-0.461
chy-0.455
 White-0.446
 meant-0.439
 whispered-0.435
Tokens
Token ID82
Feature activation+3.76e+02

Pos logit contributions
ric+0.675
pieces+0.646
iu+0.598
 interests+0.557
n+0.527
Neg logit contributions
 faded-0.559
 White-0.537
Dead-0.513
 whispered-0.509
 wicked-0.500
Token a
Token ID257
Feature activation+3.57e+02

Pos logit contributions
iu+0.530
pieces+0.481
ric+0.472
 interests+0.443
nels+0.415
Neg logit contributions
 faded-0.479
chy-0.454
 whispered-0.453
 White-0.452
Dead-0.442
Token surprise
Token ID5975
Feature activation+3.52e+02

Pos logit contributions
iu+0.557
pieces+0.527
 interests+0.469
ric+0.468
nels+0.451
Neg logit contributions
 faded-0.376
 meant-0.373
 White-0.371
chy-0.366
 whispered-0.360
Token!
Token ID0
Feature activation+2.96e+02

Pos logit contributions
iu+0.429
pieces+0.401
 occasions+0.351
 interests+0.332
nels+0.329
Neg logit contributions
chy-0.504
 White-0.496
 faded-0.495
Dead-0.481
Ancient-0.477
Tokenâ
Token ID22940
Feature activation+3.38e+02

Pos logit contributions
iu+0.436
pieces+0.394
ric+0.348
 interests+0.343
nels+0.327
Neg logit contributions
chy-0.384
 faded-0.382
 meant-0.357
 White-0.355
 whispered-0.354
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenYou
Token ID1639
Feature activation+3.47e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+3.92e+02

Pos logit contributions
iu+0.486
pieces+0.440
ric+0.395
 interests+0.386
nels+0.363
Neg logit contributions
 faded-0.462
chy-0.454
 White-0.441
 whispered-0.431
 meant-0.430
Token
Token ID26391
Feature activation+4.00e+02

Pos logit contributions
iu+0.349
pieces+0.277
ric+0.256
 interests+0.216
 disl+0.204
Neg logit contributions
 faded-0.757
chy-0.714
 White-0.706
 blocked-0.701
Dead-0.700
Token
Token ID8151
Feature activation+4.59e+02

Pos logit contributions
iu+0.622
pieces+0.559
ric+0.506
 interests+0.494
nels+0.475
Neg logit contributions
 faded-0.482
chy-0.476
 White-0.460
 meant-0.459
 whispered-0.453
Tokenll
Token ID297
Feature activation+4.01e+02

Pos logit contributions
ric+0.658
pieces+0.631
iu+0.576
 interests+0.549
 disl+0.497
Neg logit contributions
 faded-0.584
 White-0.535
Dead-0.531
 whispered-0.525
 wicked-0.518
Token see
Token ID766
Feature activation+3.45e+02

Pos logit contributions
iu+0.543
pieces+0.473
 interests+0.405
ric+0.402
nels+0.388
Neg logit contributions
chy-0.566
 faded-0.563
 meant-0.552
 White-0.550
 whispered-0.542
Token.
Token ID13
Feature activation+2.99e+02

Pos logit contributions
iu+0.424
pieces+0.395
 interests+0.345
ric+0.340
nels+0.319
Neg logit contributions
 faded-0.515
chy-0.508
 White-0.484
 whispered-0.468
 meant-0.467
Tokenâ
Token ID22940
Feature activation+3.43e+02

Pos logit contributions
iu+0.435
pieces+0.393
ric+0.351
 interests+0.342
nels+0.326
Neg logit contributions
 faded-0.395
chy-0.395
 whispered-0.368
 meant-0.368
 White-0.363
Token
Token ID26391
Feature activation+3.51e+02

Pos logit contributions
iu+0.332
pieces+0.267
ric+0.233
nels+0.211
 interests+0.209
Neg logit contributions
 faded-0.633
chy-0.606
 White-0.602
 meant-0.586
 whispered-0.585
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenWhere
Token ID8496
Feature activation+3.58e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+3.91e+02

Pos logit contributions
iu+0.608
pieces+0.537
ric+0.491
 interests+0.467
nels+0.467
Neg logit contributions
 meant-0.397
chy-0.393
 faded-0.392
 whispered-0.380
 White-0.376
Token
Token ID26391
Feature activation+3.99e+02

Pos logit contributions
iu+0.446
pieces+0.385
ric+0.329
 interests+0.322
nels+0.300
Neg logit contributions
 faded-0.633
chy-0.628
 White-0.618
 meant-0.613
 whispered-0.609
Token
Token ID8151
Feature activation+4.58e+02

Pos logit contributions
iu+0.739
pieces+0.676
ric+0.619
 interests+0.611
nels+0.590
Neg logit contributions
 faded-0.362
chy-0.355
 White-0.348
 meant-0.340
 whispered-0.336
Tokens
Token ID82
Feature activation+3.71e+02

Pos logit contributions
ric+0.690
pieces+0.686
iu+0.641
 interests+0.579
n+0.562
Neg logit contributions
 faded-0.516
 White-0.497
chy-0.449
 whispered-0.449
Dead-0.444
Token my
Token ID616
Feature activation+3.67e+02

Pos logit contributions
iu+0.545
pieces+0.493
ric+0.441
 interests+0.432
nels+0.413
Neg logit contributions
 faded-0.472
chy-0.466
 White-0.457
 whispered-0.444
 meant-0.442
Token jelly
Token ID27644
Feature activation+3.69e+02

Pos logit contributions
iu+0.508
pieces+0.449
ric+0.412
 interests+0.411
 Majesty+0.379
Neg logit contributions
 faded-0.492
chy-0.488
 meant-0.475
 White-0.471
 whispered-0.467
Token?
Token ID30
Feature activation+3.05e+02

Pos logit contributions
iu+0.428
pieces+0.385
ric+0.348
 interests+0.338
nels+0.332
Neg logit contributions
 White-0.551
 faded-0.537
Dead-0.512
Ancient-0.510
 whispered-0.505
Tokenâ
Token ID22940
Feature activation+3.38e+02

Pos logit contributions
iu+0.378
pieces+0.371
ric+0.333
 Amen+0.305
 interests+0.291
Neg logit contributions
chy-0.461
 faded-0.455
 spilled-0.404
 whispered-0.400
 smelled-0.396
Token't
Token ID470
Feature activation+3.58e+02

Pos logit contributions
chy+0.310
 faded+0.293
 White+0.291
 whispered+0.279
 meant+0.275
Neg logit contributions
iu-0.315
pieces-0.283
 disl-0.237
 interests-0.236
nels-0.228
Token know
Token ID760
Feature activation+3.19e+02

Pos logit contributions
iu+0.555
pieces+0.493
ric+0.454
nels+0.429
 interests+0.419
Neg logit contributions
 faded-0.451
chy-0.446
 meant-0.445
 whispered-0.438
 White-0.419
Token the
Token ID262
Feature activation+3.29e+02

Pos logit contributions
iu+0.364
pieces+0.328
 interests+0.281
ric+0.276
nels+0.253
Neg logit contributions
 faded-0.502
chy-0.499
 White-0.483
 whispered-0.474
 meant-0.472
Token rh
Token ID9529
Feature activation+3.95e+02

Pos logit contributions
iu+0.546
pieces+0.496
ric+0.458
 interests+0.452
nels+0.420
Neg logit contributions
 faded-0.348
 White-0.332
chy-0.329
 whispered-0.327
 meant-0.318
Tokenin
Token ID259
Feature activation+3.84e+02

Pos logit contributions
iu+0.755
pieces+0.683
ric+0.626
 interests+0.624
nels+0.613
Neg logit contributions
 faded-0.336
 White-0.320
chy-0.319
 meant-0.316
 whispered-0.304
Tokenoc
Token ID420
Feature activation+4.58e+02

Pos logit contributions
iu+0.329
pieces+0.265
ric+0.220
 interests+0.210
earned+0.194
Neg logit contributions
 faded-0.712
chy-0.697
 Blood-0.678
 White-0.669
 whispered-0.662
Tokeneros
Token ID27498
Feature activation+3.62e+02

Pos logit contributions
iu+0.597
pieces+0.561
 interests+0.500
nels+0.464
ric+0.463
Neg logit contributions
 White-0.620
 faded-0.605
 meant-0.590
chy-0.583
 Green-0.565
Token was
Token ID373
Feature activation+3.16e+02

Pos logit contributions
pieces+0.481
iu+0.466
n+0.403
 interests+0.398
 ye+0.395
Neg logit contributions
 Blood-0.433
 White-0.413
Ancient-0.409
Dead-0.402
 faded-0.402
Token wild
Token ID4295
Feature activation+3.25e+02

Pos logit contributions
iu+0.441
pieces+0.409
ric+0.389
 interests+0.373
 wre+0.361
Neg logit contributions
Ancient-0.369
 White-0.360
Dead-0.337
 whispered-0.329
Factory-0.329
Token.
Token ID13
Feature activation+2.74e+02

Pos logit contributions
iu+0.587
pieces+0.536
ric+0.494
 interests+0.489
nels+0.471
Neg logit contributions
 faded-0.301
chy-0.292
 White-0.288
 whispered-0.281
 meant-0.274
Token Can
Token ID1680
Feature activation+2.83e+02

Pos logit contributions
iu+0.377
pieces+0.348
ric+0.331
 Amen+0.323
 disl+0.301
Neg logit contributions
chy-0.385
 faded-0.368
 spilled-0.339
 shone-0.339
 melted-0.331
Token ball
Token ID2613
Feature activation-2.32e+02

Pos logit contributions
chy+0.210
 faded+0.201
 White+0.189
 whispered+0.188
 dark+0.185
Neg logit contributions
iu-0.469
pieces-0.418
ric-0.397
 interests-0.383
 Amen-0.369
Token.
Token ID13
Feature activation-2.50e+02

Pos logit contributions
 faded+0.262
 White+0.259
 meant+0.248
 whispered+0.248
 spoke+0.241
Neg logit contributions
iu-0.363
pieces-0.336
ric-0.298
 interests-0.292
nels-0.281
Token They
Token ID1119
Feature activation-2.44e+02

Pos logit contributions
 faded+0.245
chy+0.238
 meant+0.230
 whispered+0.227
 White+0.222
Neg logit contributions
iu-0.443
pieces-0.409
ric-0.377
 interests-0.377
nels-0.352
Token are
Token ID389
Feature activation-2.47e+02

Pos logit contributions
 faded+0.349
chy+0.348
 White+0.340
 meant+0.336
 whispered+0.334
Neg logit contributions
iu-0.326
pieces-0.286
ric-0.251
 interests-0.247
nels-0.234
Token clever
Token ID14169
Feature activation-2.44e+02

Pos logit contributions
 faded+0.239
chy+0.224
 White+0.222
 whispered+0.215
 meant+0.212
Neg logit contributions
iu-0.442
pieces-0.409
ric-0.373
 interests-0.369
nels-0.354
Token.
Token ID13
Feature activation-2.58e+02

Pos logit contributions
 faded+0.199
chy+0.196
 meant+0.189
 White+0.188
 whispered+0.184
Neg logit contributions
iu-0.477
pieces-0.438
ric-0.407
 interests-0.394
nels-0.385
Token They
Token ID1119
Feature activation-2.52e+02

Pos logit contributions
 faded+0.212
chy+0.208
 meant+0.195
 whispered+0.193
 White+0.191
Neg logit contributions
iu-0.499
pieces-0.462
ric-0.429
 interests-0.428
nels-0.405
Token know
Token ID760
Feature activation-2.59e+02

Pos logit contributions
 faded+0.368
chy+0.361
 White+0.358
 meant+0.353
 whispered+0.350
Neg logit contributions
iu-0.325
pieces-0.288
ric-0.252
 interests-0.247
nels-0.232
Token how
Token ID703
Feature activation-2.60e+02

Pos logit contributions
 faded+0.427
chy+0.426
 White+0.418
 meant+0.411
 whispered+0.409
Neg logit contributions
iu-0.286
pieces-0.249
 interests-0.208
ric-0.208
nels-0.193
Token to
Token ID284
Feature activation-2.60e+02

Pos logit contributions
chy+0.178
 hard+0.177
 shy+0.176
 dark+0.172
 faded+0.171
Neg logit contributions
iu-0.577
pieces-0.505
ric-0.485
 Amen-0.475
ode-0.457
Token share
Token ID2648
Feature activation-2.64e+02

Pos logit contributions
chy+0.332
 faded+0.301
 meant+0.298
 whispered+0.296
 White+0.291
Neg logit contributions
iu-0.428
pieces-0.374
ric-0.328
 interests-0.323
 Amen-0.321
Token a
Token ID257
Feature activation-2.38e+02

Pos logit contributions
 faded+0.365
chy+0.359
 White+0.351
 meant+0.347
 whispered+0.343
Neg logit contributions
iu-0.291
pieces-0.258
ric-0.224
 interests-0.220
nels-0.204
Token spaceship
Token ID40663
Feature activation-2.25e+02

Pos logit contributions
 meant+0.304
chy+0.299
 faded+0.296
 whispered+0.285
 White+0.284
Neg logit contributions
iu-0.337
pieces-0.314
 interests-0.279
 Majesty-0.269
 Amen-0.263
Token!"
Token ID2474
Feature activation-2.49e+02

Pos logit contributions
 faded+0.273
chy+0.269
 meant+0.266
 White+0.263
 whispered+0.263
Neg logit contributions
iu-0.352
pieces-0.311
ric-0.284
 interests-0.275
nels-0.263
Token Ben
Token ID3932
Feature activation-2.35e+02

Pos logit contributions
 faded+0.193
chy+0.191
 White+0.182
 meant+0.179
 whispered+0.172
Neg logit contributions
iu-0.494
pieces-0.457
ric-0.415
 interests-0.414
nels-0.401
Token said
Token ID531
Feature activation-2.47e+02

Pos logit contributions
chy+0.256
 faded+0.244
 dying+0.239
 rotting+0.236
Poor+0.235
Neg logit contributions
iu-0.384
pieces-0.332
 interests-0.308
n-0.298
 Amen-0.283
Token.
Token ID13
Feature activation-2.57e+02

Pos logit contributions
 faded+0.237
chy+0.232
 White+0.231
 meant+0.221
Dead+0.217
Neg logit contributions
iu-0.445
pieces-0.403
 interests-0.359
 Amen-0.341
nels-0.339
Token "
Token ID366
Feature activation-2.52e+02

Pos logit contributions
chy+0.362
 faded+0.325
 shone+0.308
 meant+0.305
 wil+0.305
Neg logit contributions
iu-0.344
pieces-0.291
 interests-0.267
 Amen-0.266
nels-0.249
TokenWe
Token ID1135
Feature activation-2.52e+02

Pos logit contributions
 faded+0.286
chy+0.281
 meant+0.266
 whispered+0.265
 White+0.260
Neg logit contributions
iu-0.417
pieces-0.385
 interests-0.349
ric-0.345
nels-0.328
Token can
Token ID460
Feature activation-2.55e+02

Pos logit contributions
 faded+0.295
chy+0.293
 White+0.285
 meant+0.284
 whispered+0.280
Neg logit contributions
iu-0.402
pieces-0.362
ric-0.325
 interests-0.319
nels-0.307
Token fly
Token ID6129
Feature activation-2.54e+02

Pos logit contributions
chy+0.351
 White+0.350
 faded+0.349
 whispered+0.345
 meant+0.339
Neg logit contributions
iu-0.359
pieces-0.312
 interests-0.270
ric-0.270
nels-0.254
Token to
Token ID284
Feature activation-2.60e+02

Pos logit contributions
 faded+0.274
chy+0.270
 White+0.266
 meant+0.260
 whispered+0.258
Neg logit contributions
iu-0.432
pieces-0.387
ric-0.350
 interests-0.346
nels-0.330
Token learned
Token ID4499
Feature activation-2.43e+02

Pos logit contributions
 White+0.262
chy+0.256
 faded+0.251
Dead+0.249
 Blood+0.245
Neg logit contributions
iu-0.341
pieces-0.316
 interests-0.304
ric-0.291
 disl-0.278
Token that
Token ID326
Feature activation-2.44e+02

Pos logit contributions
 faded+0.292
chy+0.285
 meant+0.284
 White+0.282
 whispered+0.278
Neg logit contributions
iu-0.391
pieces-0.352
ric-0.314
 interests-0.298
nels-0.290
Token courage
Token ID11917
Feature activation-2.39e+02

Pos logit contributions
chy+0.355
 faded+0.348
 White+0.343
 whispered+0.336
Ancient+0.323
Neg logit contributions
iu-0.301
pieces-0.267
 interests-0.263
ric-0.230
nels-0.216
Token is
Token ID318
Feature activation-2.45e+02

Pos logit contributions
 White+0.309
chy+0.306
Dead+0.300
 faded+0.297
Ancient+0.290
Neg logit contributions
iu-0.307
 interests-0.294
pieces-0.294
ric-0.275
nels-0.256
Token important
Token ID1593
Feature activation-2.43e+02

Pos logit contributions
chy+0.315
 White+0.311
 faded+0.308
Dead+0.296
 whispered+0.290
Neg logit contributions
iu-0.331
pieces-0.306
 interests-0.299
nels-0.261
ric-0.260
Token.
Token ID13
Feature activation-2.56e+02

Pos logit contributions
 faded+0.205
chy+0.201
 White+0.195
 meant+0.189
 whispered+0.188
Neg logit contributions
iu-0.467
pieces-0.429
ric-0.393
 interests-0.392
nels-0.377
Token They
Token ID1119
Feature activation-2.50e+02

Pos logit contributions
 faded+0.231
chy+0.231
 whispered+0.213
 White+0.211
 meant+0.211
Neg logit contributions
iu-0.476
pieces-0.433
ric-0.406
 interests-0.404
nels-0.384
Token learned
Token ID4499
Feature activation-2.59e+02

Pos logit contributions
 White+0.276
chy+0.269
Dead+0.263
 faded+0.262
 Blood+0.259
Neg logit contributions
iu-0.371
pieces-0.342
 interests-0.328
ric-0.313
 disl-0.299
Token that
Token ID326
Feature activation-2.59e+02

Pos logit contributions
 faded+0.321
chy+0.314
 meant+0.311
 White+0.310
 whispered+0.305
Neg logit contributions
iu-0.404
pieces-0.362
ric-0.323
 interests-0.309
nels-0.300
Token soccer
Token ID11783
Feature activation-2.47e+02

Pos logit contributions
chy+0.377
 faded+0.367
 White+0.363
 whispered+0.356
Ancient+0.346
Neg logit contributions
iu-0.318
 interests-0.284
pieces-0.282
ric-0.241
 disl-0.228
Token is
Token ID318
Feature activation-2.60e+02

Pos logit contributions
 White+0.341
chy+0.325
Dead+0.324
 faded+0.319
 whispered+0.311
Neg logit contributions
iu-0.305
pieces-0.294
 interests-0.294
nels-0.260
ric-0.255
Token but
Token ID475
Feature activation-2.42e+02

Pos logit contributions
 White+0.238
chy+0.237
 faded+0.234
 meant+0.228
 whispered+0.223
Neg logit contributions
iu-0.411
pieces-0.372
 interests-0.354
ric-0.332
 disl-0.317
Token they
Token ID484
Feature activation-2.36e+02

Pos logit contributions
chy+0.257
 faded+0.257
 White+0.251
 whispered+0.242
 meant+0.242
Neg logit contributions
iu-0.407
pieces-0.369
ric-0.335
 interests-0.333
nels-0.316
Token were
Token ID547
Feature activation-2.43e+02

Pos logit contributions
 White+0.257
 faded+0.243
chy+0.238
Ancient+0.236
 BR+0.234
Neg logit contributions
iu-0.364
pieces-0.325
 interests-0.312
ric-0.308
 disl-0.283
Token more
Token ID517
Feature activation-2.44e+02

Pos logit contributions
 faded+0.238
 White+0.233
chy+0.231
 meant+0.219
 whispered+0.219
Neg logit contributions
iu-0.421
pieces-0.386
 interests-0.352
ric-0.349
nels-0.330
Token careful
Token ID8161
Feature activation-2.41e+02

Pos logit contributions
 White+0.289
chy+0.286
 faded+0.281
 meant+0.261
 whispered+0.255
Neg logit contributions
iu-0.383
pieces-0.338
 interests-0.322
 disl-0.293
nels-0.291
Token.
Token ID13
Feature activation-2.56e+02

Pos logit contributions
 meant+0.123
chy+0.121
 faded+0.108
 hard+0.099
 shy+0.097
Neg logit contributions
iu-0.545
pieces-0.526
icol-0.487
ric-0.484
 Majesty-0.475
Token<|endoftext|>
Token ID50256
Feature activation-2.52e+02

Pos logit contributions
chy+0.244
 faded+0.239
 whispered+0.222
 meant+0.220
 blocked+0.214
Neg logit contributions
iu-0.462
pieces-0.418
 interests-0.397
ric-0.392
nels-0.369
TokenAnna
Token ID31160
Feature activation-2.47e+02

Pos logit contributions
chy+0.231
 faded+0.225
 whispered+0.209
 sore+0.192
Dead+0.191
Neg logit contributions
iu-0.442
pieces-0.403
ric-0.393
 interests-0.375
 Amen-0.366
Token and
Token ID290
Feature activation-2.59e+02

Pos logit contributions
Pr+0.319
 darker+0.318
Dead+0.315
 Blood+0.309
 White+0.300
Neg logit contributions
iu-0.299
 interests-0.275
 disl-0.239
icol-0.235
 occasions-0.222
Token Ben
Token ID3932
Feature activation-2.48e+02

Pos logit contributions
 whispered+0.161
 said+0.153
 meant+0.151
 answered+0.141
 faded+0.140
Neg logit contributions
iu-0.594
pieces-0.567
ric-0.551
 occasions-0.523
nels-0.522
Token were
Token ID547
Feature activation-2.59e+02

Pos logit contributions
 White+0.282
Dead+0.270
chy+0.270
 Blood+0.265
 faded+0.254
Neg logit contributions
iu-0.369
 interests-0.321
pieces-0.312
 disl-0.294
 spend-0.292
Token go
Token ID467
Feature activation-2.43e+02

Pos logit contributions
 faded+0.268
 whispered+0.261
chy+0.260
 meant+0.257
 smelled+0.253
Neg logit contributions
iu-0.411
pieces-0.359
ric-0.321
nels-0.314
 interests-0.312
Token outside
Token ID2354
Feature activation-2.40e+02

Pos logit contributions
 faded+0.221
 whispered+0.219
 dark+0.209
chy+0.208
 White+0.200
Neg logit contributions
iu-0.473
pieces-0.407
ric-0.383
 interests-0.378
 disl-0.374
Token now
Token ID783
Feature activation-2.44e+02

Pos logit contributions
chy+0.272
 faded+0.270
 White+0.262
 meant+0.258
 whispered+0.252
Neg logit contributions
iu-0.382
pieces-0.351
ric-0.314
 interests-0.312
nels-0.294
Token?"
Token ID1701
Feature activation-2.55e+02

Pos logit contributions
 faded+0.293
chy+0.285
 White+0.283
 meant+0.280
 whispered+0.278
Neg logit contributions
iu-0.385
pieces-0.348
ric-0.314
 interests-0.306
nels-0.290
Token Anna
Token ID11735
Feature activation-2.44e+02

Pos logit contributions
chy+0.275
 faded+0.258
 meant+0.245
 White+0.242
 blocked+0.235
Neg logit contributions
iu-0.427
pieces-0.390
 interests-0.353
 Amen-0.338
ric-0.338
Token and
Token ID290
Feature activation-2.55e+02

Pos logit contributions
chy+0.285
 faded+0.265
 flowed+0.247
 shone+0.244
 blocked+0.232
Neg logit contributions
iu-0.392
 interests-0.320
pieces-0.320
 Amen-0.319
nels-0.272
Token Ben
Token ID3932
Feature activation-2.44e+02

Pos logit contributions
 faded+0.198
chy+0.195
 White+0.185
 meant+0.184
 whispered+0.179
Neg logit contributions
iu-0.510
pieces-0.468
ric-0.430
 interests-0.426
nels-0.411
Token asked
Token ID1965
Feature activation-2.58e+02

Pos logit contributions
chy+0.295
 faded+0.273
Dead+0.253
Ancient+0.249
 rotten+0.248
Neg logit contributions
iu-0.379
pieces-0.323
 Amen-0.306
 interests-0.300
n-0.273
Token.
Token ID13
Feature activation-2.66e+02

Pos logit contributions
 faded+0.278
chy+0.271
 meant+0.256
Dead+0.254
 blocked+0.254
Neg logit contributions
iu-0.437
pieces-0.393
 interests-0.363
ric-0.339
nels-0.330
Token
Token ID198
Feature activation-2.65e+02

Pos logit contributions
chy+0.390
 faded+0.343
 meant+0.318
 shone+0.311
 melted+0.310
Neg logit contributions
iu-0.348
 interests-0.307
 Amen-0.301
pieces-0.288
nels-0.271
Token
Token ID198
Feature activation-2.67e+02

Pos logit contributions
chy+0.194
 faded+0.194
 meant+0.180
 White+0.177
 whispered+0.171
Neg logit contributions
iu-0.540
pieces-0.497
ric-0.461
 interests-0.458
nels-0.444
Token.
Token ID13
Feature activation-2.45e+02

Pos logit contributions
 White+0.230
chy+0.226
 faded+0.225
 meant+0.211
 whispered+0.209
Neg logit contributions
iu-0.394
pieces-0.379
 interests-0.342
ric-0.324
nels-0.315
Token They
Token ID1119
Feature activation-2.39e+02

Pos logit contributions
chy+0.327
 faded+0.317
 whispered+0.297
 blocked+0.297
 shone+0.295
Neg logit contributions
iu-0.349
 interests-0.318
pieces-0.307
ric-0.290
 disl-0.275
Token are
Token ID389
Feature activation-2.42e+02

Pos logit contributions
 faded+0.347
chy+0.343
 White+0.338
 meant+0.334
 whispered+0.331
Neg logit contributions
iu-0.312
pieces-0.275
ric-0.241
 interests-0.236
nels-0.223
Token happy
Token ID3772
Feature activation-2.42e+02

Pos logit contributions
 faded+0.218
chy+0.214
 White+0.209
 meant+0.203
 whispered+0.201
Neg logit contributions
iu-0.451
pieces-0.415
ric-0.378
 interests-0.375
nels-0.360
Token friends
Token ID2460
Feature activation-2.37e+02

Pos logit contributions
chy+0.211
 faded+0.209
 White+0.201
 meant+0.198
 whispered+0.194
Neg logit contributions
iu-0.461
pieces-0.421
ric-0.386
 interests-0.386
nels-0.370
Token.
Token ID13
Feature activation-2.55e+02

Pos logit contributions
chy+0.224
 faded+0.222
 White+0.215
 whispered+0.208
 meant+0.207
Neg logit contributions
iu-0.424
pieces-0.391
 interests-0.353
ric-0.347
 Amen-0.337
Token<|endoftext|>
Token ID50256
Feature activation-2.51e+02

Pos logit contributions
chy+0.243
 faded+0.240
 whispered+0.221
 meant+0.216
 blocked+0.214
Neg logit contributions
iu-0.460
pieces-0.414
 interests-0.403
ric-0.391
nels-0.369
TokenL
Token ID43
Feature activation-2.36e+02

Pos logit contributions
chy+0.176
 faded+0.175
 whispered+0.158
 meant+0.147
 White+0.145
Neg logit contributions
iu-0.502
pieces-0.465
ric-0.448
 interests-0.435
nels-0.423
Tokenily
Token ID813
Feature activation-2.46e+02

Pos logit contributions
 whispered+0.169
 meant+0.164
 faded+0.156
 answered+0.155
chy+0.154
Neg logit contributions
iu-0.482
pieces-0.433
ric-0.418
 Amen-0.408
 disl-0.405
Token and
Token ID290
Feature activation-2.58e+02

Pos logit contributions
chy+0.327
 White+0.324
 faded+0.324
Dead+0.316
 blocked+0.307
Neg logit contributions
iu-0.313
 interests-0.272
pieces-0.261
ric-0.248
 disl-0.242
Token Tom
Token ID4186
Feature activation-2.47e+02

Pos logit contributions
 said+0.164
 meant+0.163
 whispered+0.152
 She+0.148
 she+0.146
Neg logit contributions
iu-0.609
pieces-0.579
ric-0.564
immers-0.553
 occasions-0.538
TokenCan
Token ID6090
Feature activation-2.40e+02

Pos logit contributions
chy+0.316
 weak+0.269
 nervous+0.266
 tired+0.262
 bored+0.262
Neg logit contributions
iu-0.351
pieces-0.295
 interests-0.285
 Amen-0.284
 occasions-0.272
Token we
Token ID356
Feature activation-2.37e+02

Pos logit contributions
 White+0.251
 faded+0.249
 broke+0.245
 blocked+0.238
 fell+0.237
Neg logit contributions
iu-0.431
pieces-0.399
ric-0.377
 impression-0.367
 Majesty-0.352
Token go
Token ID467
Feature activation-2.43e+02

Pos logit contributions
 faded+0.268
 whispered+0.261
chy+0.260
 meant+0.257
 smelled+0.253
Neg logit contributions
iu-0.411
pieces-0.359
ric-0.321
nels-0.314
 interests-0.312
Token outside
Token ID2354
Feature activation-2.40e+02

Pos logit contributions
 faded+0.221
 whispered+0.219
 dark+0.209
chy+0.208
 White+0.200
Neg logit contributions
iu-0.473
pieces-0.407
ric-0.383
 interests-0.378
 disl-0.374
Token now
Token ID783
Feature activation-2.44e+02

Pos logit contributions
chy+0.272
 faded+0.270
 White+0.262
 meant+0.258
 whispered+0.252
Neg logit contributions
iu-0.382
pieces-0.351
ric-0.314
 interests-0.312
nels-0.294
Token?"
Token ID1701
Feature activation-2.55e+02

Pos logit contributions
 faded+0.293
chy+0.285
 White+0.283
 meant+0.280
 whispered+0.278
Neg logit contributions
iu-0.385
pieces-0.348
ric-0.314
 interests-0.306
nels-0.290
Token Anna
Token ID11735
Feature activation-2.44e+02

Pos logit contributions
chy+0.275
 faded+0.258
 meant+0.245
 White+0.242
 blocked+0.235
Neg logit contributions
iu-0.427
pieces-0.390
 interests-0.353
 Amen-0.338
ric-0.338
Token and
Token ID290
Feature activation-2.55e+02

Pos logit contributions
chy+0.285
 faded+0.265
 flowed+0.247
 shone+0.244
 blocked+0.232
Neg logit contributions
iu-0.392
 interests-0.320
pieces-0.320
 Amen-0.319
nels-0.272
Token Ben
Token ID3932
Feature activation-2.44e+02

Pos logit contributions
 faded+0.198
chy+0.195
 White+0.185
 meant+0.184
 whispered+0.179
Neg logit contributions
iu-0.510
pieces-0.468
ric-0.430
 interests-0.426
nels-0.411
Token asked
Token ID1965
Feature activation-2.58e+02

Pos logit contributions
chy+0.295
 faded+0.273
Dead+0.253
Ancient+0.249
 rotten+0.248
Neg logit contributions
iu-0.379
pieces-0.323
 Amen-0.306
 interests-0.300
n-0.273
Token.
Token ID13
Feature activation-2.66e+02

Pos logit contributions
 faded+0.278
chy+0.271
 meant+0.256
Dead+0.254
 blocked+0.254
Neg logit contributions
iu-0.437
pieces-0.393
 interests-0.363
ric-0.339
nels-0.330
Token.
Token ID13
Feature activation-2.41e+02

Pos logit contributions
 faded+0.188
 White+0.183
chy+0.181
 meant+0.173
 whispered+0.171
Neg logit contributions
iu-0.406
pieces-0.373
ric-0.339
 interests-0.334
nels-0.324
Token They
Token ID1119
Feature activation-2.34e+02

Pos logit contributions
 faded+0.206
chy+0.202
 meant+0.192
 White+0.191
 whispered+0.188
Neg logit contributions
iu-0.459
pieces-0.420
ric-0.384
 interests-0.382
nels-0.371
Token ate
Token ID15063
Feature activation-2.40e+02

Pos logit contributions
 White+0.294
chy+0.294
Dead+0.291
Ancient+0.277
 faded+0.277
Neg logit contributions
iu-0.325
pieces-0.287
 spend-0.250
 interests-0.239
ric-0.235
Token and
Token ID290
Feature activation-2.45e+02

Pos logit contributions
 faded+0.320
chy+0.318
 White+0.316
 whispered+0.305
Dead+0.304
Neg logit contributions
iu-0.335
pieces-0.294
ric-0.254
 interests-0.249
nels-0.241
Token ate
Token ID15063
Feature activation-2.46e+02

Pos logit contributions
chy+0.222
 White+0.219
 faded+0.215
Ancient+0.213
Dead+0.208
Neg logit contributions
iu-0.445
pieces-0.393
 interests-0.359
ric-0.351
 disl-0.336
Token until
Token ID1566
Feature activation-2.54e+02

Pos logit contributions
 White+0.338
 faded+0.337
chy+0.335
Dead+0.331
 whispered+0.321
Neg logit contributions
iu-0.332
pieces-0.286
 interests-0.241
ric-0.233
nels-0.231
Token their
Token ID511
Feature activation-2.43e+02

Pos logit contributions
chy+0.353
 White+0.344
 meant+0.342
 whispered+0.336
Dead+0.335
Neg logit contributions
iu-0.344
pieces-0.290
nels-0.238
 interests-0.236
 Amen-0.230
Token bell
Token ID8966
Feature activation-2.41e+02

Pos logit contributions
 White+0.374
 meant+0.373
chy+0.372
 faded+0.370
 whispered+0.367
Neg logit contributions
iu-0.295
pieces-0.249
 interests-0.220
ric-0.212
nels-0.200
Tokenies
Token ID444
Feature activation-2.39e+02

Pos logit contributions
 wicked+0.273
Dead+0.269
Ancient+0.268
Old+0.265
Today+0.260
Neg logit contributions
iu-0.340
pieces-0.304
ric-0.296
n-0.267
nels-0.262
Token were
Token ID547
Feature activation-2.56e+02

Pos logit contributions
Ancient+0.410
 Blood+0.407
Dead+0.399
Pr+0.396
Spider+0.395
Neg logit contributions
n-0.205
 spend-0.174
iu-0.170
 go-0.165
 how-0.160
Token full
Token ID1336
Feature activation-2.56e+02

Pos logit contributions
Ancient+0.366
Dead+0.365
 White+0.365
 Monster+0.333
 Did+0.332
Neg logit contributions
iu-0.336
pieces-0.261
ric-0.251
 interests-0.248
allion-0.240
Token.
Token ID13
Feature activation-2.41e+02

Pos logit contributions
chy+0.260
 faded+0.253
 meant+0.244
 blocked+0.239
Dead+0.238
Neg logit contributions
iu-0.366
pieces-0.333
 interests-0.301
ric-0.289
nels-0.281
Token They
Token ID1119
Feature activation-2.35e+02

Pos logit contributions
chy+0.372
 meant+0.327
 faded+0.324
 blocked+0.324
 spoke+0.319
Neg logit contributions
iu-0.293
 interests-0.256
 Amen-0.245
pieces-0.236
nels-0.230
Token smiled
Token ID13541
Feature activation-2.38e+02

Pos logit contributions
 Blood+0.300
 White+0.281
Ancient+0.277
 Salt+0.275
 late+0.272
Neg logit contributions
iu-0.298
 interests-0.258
 spend-0.241
pieces-0.241
 wait-0.239
Token and
Token ID290
Feature activation-2.44e+02

Pos logit contributions
chy+0.254
 White+0.245
 faded+0.239
 blocked+0.238
Dead+0.236
Neg logit contributions
iu-0.377
pieces-0.335
 interests-0.324
ric-0.300
 Amen-0.299
Token laughed
Token ID13818
Feature activation-2.42e+02

Pos logit contributions
chy+0.239
Ancient+0.211
 White+0.209
Dead+0.206
 Blood+0.194
Neg logit contributions
iu-0.427
pieces-0.361
 interests-0.352
icol-0.326
nels-0.326
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
chy+0.303
Dead+0.300
 White+0.297
 blocked+0.296
 faded+0.291
Neg logit contributions
iu-0.328
pieces-0.301
 interests-0.273
 Amen-0.263
ap-0.251
Token They
Token ID1119
Feature activation-2.48e+02

Pos logit contributions
chy+0.360
 faded+0.320
 meant+0.317
 blocked+0.316
 whispered+0.316
Neg logit contributions
iu-0.337
 interests-0.302
pieces-0.284
 Amen-0.276
nels-0.269
Token had
Token ID550
Feature activation-2.54e+02

Pos logit contributions
 Blood+0.312
 White+0.298
Ancient+0.292
 Salt+0.289
Factory+0.285
Neg logit contributions
iu-0.308
 interests-0.275
pieces-0.254
 believe-0.252
 spend-0.249
Token a
Token ID257
Feature activation-2.53e+02

Pos logit contributions
 White+0.370
chy+0.366
Ancient+0.358
Dead+0.358
oled+0.335
Neg logit contributions
iu-0.298
 interests-0.277
pieces-0.250
 disl-0.231
 wre-0.211
Token great
Token ID1049
Feature activation-2.53e+02

Pos logit contributions
 meant+0.315
chy+0.305
 White+0.299
 whispered+0.293
 faded+0.292
Neg logit contributions
iu-0.358
pieces-0.327
 interests-0.315
 Amen-0.293
ric-0.292
Token adventure
Token ID8855
Feature activation-2.52e+02

Pos logit contributions
 faded+0.230
chy+0.218
 White+0.208
 meant+0.205
 whispered+0.202
Neg logit contributions
iu-0.477
pieces-0.448
ric-0.412
nels-0.386
 interests-0.378
Token the
Token ID262
Feature activation-2.41e+02

Pos logit contributions
 White+0.176
 dark+0.167
 whispered+0.159
 shy+0.159
 faded+0.156
Neg logit contributions
iu-0.505
pieces-0.484
ric-0.431
nels-0.427
 interests-0.427
Token office
Token ID2607
Feature activation-2.29e+02

Pos logit contributions
 faded+0.282
chy+0.275
 White+0.270
 meant+0.268
 whispered+0.263
Neg logit contributions
iu-0.388
pieces-0.351
ric-0.312
 interests-0.312
nels-0.299
Token.
Token ID13
Feature activation-2.53e+02

Pos logit contributions
chy+0.240
 White+0.240
 faded+0.239
 meant+0.223
Dead+0.218
Neg logit contributions
iu-0.391
pieces-0.353
 interests-0.311
nels-0.299
ric-0.297
Token He
Token ID679
Feature activation-2.43e+02

Pos logit contributions
chy+0.202
 faded+0.197
 White+0.184
 meant+0.180
 whispered+0.178
Neg logit contributions
iu-0.497
pieces-0.452
 interests-0.422
ric-0.419
nels-0.407
Token looked
Token ID3114
Feature activation-2.49e+02

Pos logit contributions
chy+0.228
 White+0.221
 conditioning+0.191
 faded+0.191
 depress+0.186
Neg logit contributions
iu-0.434
pieces-0.366
 interests-0.363
nels-0.340
 wre-0.333
Token at
Token ID379
Feature activation-2.54e+02

Pos logit contributions
 faded+0.337
 White+0.328
chy+0.327
 meant+0.308
Dead+0.302
Neg logit contributions
iu-0.341
pieces-0.318
 interests-0.274
nels-0.260
ric-0.259
Token the
Token ID262
Feature activation-2.52e+02

Pos logit contributions
 White+0.255
 faded+0.253
 meant+0.246
chy+0.243
 whispered+0.238
Neg logit contributions
iu-0.462
pieces-0.418
ric-0.370
 interests-0.370
 Amen-0.361
Token computer
Token ID3644
Feature activation-2.39e+02

Pos logit contributions
 faded+0.317
 White+0.300
 whispered+0.298
chy+0.295
 meant+0.289
Neg logit contributions
iu-0.376
pieces-0.348
 interests-0.307
ric-0.298
nels-0.288
Token and
Token ID290
Feature activation-2.58e+02

Pos logit contributions
Pr+0.384
 White+0.368
Spider+0.352
 Hen+0.348
Dead+0.347
Neg logit contributions
iu-0.239
pieces-0.234
ap-0.195
 wre-0.191
 how-0.186
Token saw
Token ID2497
Feature activation-2.62e+02

Pos logit contributions
chy+0.208
 White+0.186
 faded+0.185
 meant+0.174
 whispered+0.158
Neg logit contributions
iu-0.510
pieces-0.471
 interests-0.436
ric-0.431
nels-0.416
Token that
Token ID326
Feature activation-2.62e+02

Pos logit contributions
chy+0.472
 faded+0.435
 White+0.433
 whispered+0.423
 spoke+0.411
Neg logit contributions
iu-0.248
pieces-0.232
ric-0.175
nels-0.173
 wre-0.171
Token that
Token ID326
Feature activation-2.39e+02

Pos logit contributions
chy+0.326
 faded+0.320
 White+0.301
 whispered+0.296
 meant+0.294
Neg logit contributions
iu-0.332
pieces-0.306
 interests-0.263
ric-0.261
nels-0.245
Token John
Token ID1757
Feature activation-2.26e+02

Pos logit contributions
chy+0.292
 faded+0.284
 White+0.262
 whispered+0.261
 depress+0.257
Neg logit contributions
iu-0.362
pieces-0.329
 interests-0.299
ric-0.285
nels-0.274
Token was
Token ID373
Feature activation-2.41e+02

Pos logit contributions
 Blood+0.332
Ancient+0.328
Pr+0.316
Dead+0.316
Factory+0.309
Neg logit contributions
pieces-0.226
ric-0.224
iu-0.204
 how-0.197
n-0.189
Token still
Token ID991
Feature activation-2.39e+02

Pos logit contributions
chy+0.252
Ancient+0.251
 White+0.250
Dead+0.246
 faded+0.245
Neg logit contributions
iu-0.378
pieces-0.341
ric-0.326
 interests-0.309
 Amen-0.302
Token there
Token ID612
Feature activation-2.42e+02

Pos logit contributions
chy+0.273
Dead+0.259
Ancient+0.259
 White+0.253
 faded+0.253
Neg logit contributions
iu-0.360
pieces-0.324
ric-0.313
 interests-0.306
 impression-0.279
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
Spider+0.436
 Blood+0.432
Dead+0.431
 smoked+0.422
Ancient+0.420
Neg logit contributions
pieces-0.166
 interests-0.163
 how-0.162
 activities-0.139
ric-0.135
Token Mark
Token ID2940
Feature activation-2.37e+02

Pos logit contributions
chy+0.289
 faded+0.277
 whispered+0.254
 spilled+0.254
 shone+0.244
Neg logit contributions
iu-0.425
pieces-0.371
 interests-0.355
ric-0.346
 Amen-0.335
Token smiled
Token ID13541
Feature activation-2.49e+02

Pos logit contributions
 Blood+0.343
Ancient+0.328
chy+0.314
jar+0.311
Dead+0.310
Neg logit contributions
iu-0.273
pieces-0.248
n-0.226
 how-0.220
 interests-0.219
Token,
Token ID11
Feature activation-2.59e+02

Pos logit contributions
chy+0.309
 blocked+0.288
 White+0.285
Dead+0.283
 spilled+0.282
Neg logit contributions
iu-0.344
 interests-0.293
pieces-0.287
ric-0.264
 Amen-0.264
Token because
Token ID780
Feature activation-2.61e+02

Pos logit contributions
chy+0.264
 White+0.243
 faded+0.240
 meant+0.237
Ancient+0.227
Neg logit contributions
iu-0.454
pieces-0.393
 interests-0.376
ric-0.366
nels-0.343
Token he
Token ID339
Feature activation-2.55e+02

Pos logit contributions
 faded+0.181
 meant+0.174
 White+0.172
chy+0.171
 whispered+0.164
Neg logit contributions
iu-0.547
pieces-0.503
ric-0.474
 interests-0.461
nels-0.449
Token with
Token ID351
Feature activation-2.35e+02

Pos logit contributions
 faded+0.255
chy+0.251
 White+0.243
 meant+0.238
 whispered+0.233
Neg logit contributions
iu-0.384
pieces-0.355
 interests-0.315
ric-0.311
nels-0.306
Token you
Token ID345
Feature activation-2.30e+02

Pos logit contributions
 faded+0.391
 White+0.389
chy+0.389
 whispered+0.382
 meant+0.381
Neg logit contributions
iu-0.259
pieces-0.214
ric-0.188
 interests-0.179
nels-0.166
Token?"
Token ID1701
Feature activation-2.46e+02

Pos logit contributions
 faded+0.467
chy+0.449
 White+0.446
Pr+0.443
 depress+0.436
Neg logit contributions
pieces-0.118
 interests-0.100
iu-0.090
 Amen-0.082
ap-0.081
Token Anna
Token ID11735
Feature activation-2.34e+02

Pos logit contributions
 faded+0.174
chy+0.165
 whispered+0.165
 White+0.165
 meant+0.162
Neg logit contributions
iu-0.515
pieces-0.476
ric-0.454
 interests-0.432
nels-0.424
Token says
Token ID1139
Feature activation-2.42e+02

Pos logit contributions
 faded+0.319
chy+0.307
 blocked+0.288
 White+0.285
 shone+0.282
Neg logit contributions
iu-0.319
pieces-0.275
 interests-0.255
 Amen-0.249
ric-0.232
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
 faded+0.218
chy+0.212
 White+0.208
 meant+0.204
 whispered+0.195
Neg logit contributions
iu-0.454
pieces-0.413
 interests-0.374
ric-0.366
nels-0.355
Token
Token ID198
Feature activation-2.53e+02

Pos logit contributions
chy+0.295
 faded+0.283
 blocked+0.258
 meant+0.255
 shone+0.253
Neg logit contributions
iu-0.393
pieces-0.342
 interests-0.330
 Amen-0.314
ext-0.304
Token
Token ID198
Feature activation-2.54e+02

Pos logit contributions
 faded+0.201
chy+0.200
 meant+0.186
 White+0.182
 whispered+0.176
Neg logit contributions
iu-0.494
pieces-0.453
 interests-0.421
ric-0.418
nels-0.406
TokenBen
Token ID11696
Feature activation-2.40e+02

Pos logit contributions
 faded+0.153
 whispered+0.149
 White+0.149
chy+0.145
 meant+0.135
Neg logit contributions
iu-0.560
pieces-0.513
ric-0.496
 interests-0.461
nels-0.443
Token opens
Token ID9808
Feature activation-2.55e+02

Pos logit contributions
 faded+0.349
chy+0.341
 White+0.341
 meant+0.327
Dead+0.324
Neg logit contributions
iu-0.301
pieces-0.255
 interests-0.237
ric-0.232
nels-0.224
Token the
Token ID262
Feature activation-2.54e+02

Pos logit contributions
 White+0.194
 faded+0.190
 whispered+0.181
chy+0.177
 meant+0.177
Neg logit contributions
iu-0.518
pieces-0.488
ric-0.449
 interests-0.444
nels-0.417
Token play
Token ID711
Feature activation-2.34e+02

Pos logit contributions
 hurt+0.232
 faded+0.226
 whispered+0.225
chy+0.223
 sick+0.220
Neg logit contributions
iu-0.439
pieces-0.403
ocol-0.367
icol-0.347
nels-0.345
Token outside
Token ID2354
Feature activation-2.33e+02

Pos logit contributions
 faded+0.257
chy+0.254
 White+0.248
 meant+0.243
 whispered+0.240
Neg logit contributions
iu-0.388
pieces-0.356
 interests-0.317
ric-0.317
nels-0.304
Token again
Token ID757
Feature activation-2.34e+02

Pos logit contributions
 White+0.301
 faded+0.301
 whispered+0.301
Dead+0.300
chy+0.300
Neg logit contributions
iu-0.282
pieces-0.281
 interests-0.236
 Amen-0.223
 occasions-0.221
Token.
Token ID13
Feature activation-2.49e+02

Pos logit contributions
chy+0.276
 faded+0.267
 White+0.256
 meant+0.251
 blocked+0.249
Neg logit contributions
iu-0.348
pieces-0.324
 interests-0.298
 occasions-0.276
nels-0.274
Token He
Token ID679
Feature activation-2.40e+02

Pos logit contributions
 White+0.119
 faded+0.112
chy+0.109
 meant+0.108
 whispered+0.098
Neg logit contributions
iu-0.584
pieces-0.561
ric-0.506
nels-0.489
 interests-0.489
Token hoped
Token ID10719
Feature activation-2.54e+02

Pos logit contributions
 White+0.260
 Blood+0.254
Ancient+0.247
chy+0.241
Dead+0.240
Neg logit contributions
iu-0.366
 interests-0.314
pieces-0.299
ric-0.291
 Majesty-0.280
Token they
Token ID484
Feature activation-2.45e+02

Pos logit contributions
 faded+0.242
chy+0.238
 White+0.232
 meant+0.228
 whispered+0.225
Neg logit contributions
iu-0.462
pieces-0.422
ric-0.387
 interests-0.380
nels-0.368
Token could
Token ID714
Feature activation-2.51e+02

Pos logit contributions
 White+0.264
 faded+0.255
chy+0.250
Dead+0.243
 whispered+0.239
Neg logit contributions
iu-0.380
pieces-0.358
 interests-0.340
ric-0.330
 Amen-0.301
Token be
Token ID307
Feature activation-2.52e+02

Pos logit contributions
chy+0.258
 faded+0.255
 hurt+0.254
 whispered+0.253
 sick+0.245
Neg logit contributions
iu-0.465
pieces-0.423
ocol-0.381
nels-0.360
earned-0.360
Token happy
Token ID3772
Feature activation-2.52e+02

Pos logit contributions
 faded+0.317
chy+0.309
 meant+0.302
 whispered+0.302
 White+0.299
Neg logit contributions
iu-0.393
pieces-0.343
ric-0.318
 interests-0.296
nels-0.292
Token and
Token ID290
Feature activation-2.57e+02

Pos logit contributions
chy+0.298
 White+0.291
 faded+0.289
 whispered+0.278
 meant+0.275
Neg logit contributions
iu-0.382
pieces-0.341
 interests-0.325
ric-0.304
nels-0.289
Token,
Token ID11
Feature activation-2.41e+02

Pos logit contributions
 White+0.163
 faded+0.159
 whispered+0.159
 shy+0.155
 meant+0.155
Neg logit contributions
iu-0.463
pieces-0.448
ric-0.424
 interests-0.404
 disl-0.390
Token duck
Token ID22045
Feature activation-2.24e+02

Pos logit contributions
 faded+0.278
 meant+0.271
 whispered+0.267
 White+0.265
 shy+0.262
Neg logit contributions
iu-0.391
pieces-0.360
ric-0.328
 interests-0.317
nels-0.301
Token!"
Token ID2474
Feature activation-2.47e+02

Pos logit contributions
 faded+0.312
chy+0.298
 White+0.297
 meant+0.291
Dead+0.288
Neg logit contributions
iu-0.304
pieces-0.269
ric-0.248
 interests-0.236
nels-0.227
Token Mia
Token ID32189
Feature activation-2.34e+02

Pos logit contributions
 faded+0.221
 White+0.210
 whispered+0.209
 meant+0.208
chy+0.207
Neg logit contributions
iu-0.470
pieces-0.429
ric-0.405
 interests-0.390
nels-0.379
Token says
Token ID1139
Feature activation-2.43e+02

Pos logit contributions
 faded+0.226
chy+0.222
 White+0.214
 meant+0.209
 whispered+0.201
Neg logit contributions
iu-0.418
pieces-0.383
ric-0.350
 interests-0.345
nels-0.330
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
 faded+0.151
chy+0.147
 whispered+0.141
 White+0.139
 meant+0.131
Neg logit contributions
iu-0.515
pieces-0.483
ric-0.464
 interests-0.447
nels-0.439
Token
Token ID198
Feature activation-2.53e+02

Pos logit contributions
chy+0.382
 faded+0.344
 meant+0.341
 shone+0.323
 contained+0.317
Neg logit contributions
iu-0.323
 Amen-0.263
ext-0.254
 interests-0.254
pieces-0.247
Token
Token ID198
Feature activation-2.55e+02

Pos logit contributions
 faded+0.135
 White+0.132
chy+0.130
 whispered+0.125
 meant+0.123
Neg logit contributions
iu-0.564
pieces-0.528
ric-0.484
 interests-0.479
nels-0.466
TokenThe
Token ID464
Feature activation-2.53e+02

Pos logit contributions
 White+0.234
 faded+0.231
 whispered+0.230
chy+0.216
 blocked+0.215
Neg logit contributions
iu-0.471
pieces-0.437
ric-0.396
 interests-0.385
 impression-0.365
Token duck
Token ID22045
Feature activation-2.40e+02

Pos logit contributions
chy+0.338
 faded+0.333
 White+0.330
 meant+0.321
 whispered+0.318
Neg logit contributions
iu-0.360
pieces-0.317
ric-0.283
 interests-0.282
nels-0.270
Token smiles
Token ID21845
Feature activation-2.52e+02

Pos logit contributions
 White+0.307
 meant+0.297
chy+0.289
Dead+0.287
Ancient+0.287
Neg logit contributions
iu-0.334
ric-0.292
pieces-0.291
 interests-0.274
nels-0.264
Token.
Token ID13
Feature activation-2.44e+02

Pos logit contributions
 White+0.320
chy+0.307
Dead+0.297
Spider+0.296
Ancient+0.294
Neg logit contributions
iu-0.297
 interests-0.236
pieces-0.235
 occasions-0.206
nels-0.196
Token She
Token ID1375
Feature activation-2.35e+02

Pos logit contributions
chy+0.291
 faded+0.274
 blocked+0.247
 meant+0.245
 melted+0.239
Neg logit contributions
iu-0.371
 interests-0.331
pieces-0.330
ric-0.306
nels-0.306
Token gets
Token ID3011
Feature activation-2.41e+02

Pos logit contributions
 whispered+0.223
 faded+0.222
chy+0.220
 led+0.207
 meant+0.205
Neg logit contributions
iu-0.438
pieces-0.403
ric-0.361
 interests-0.354
nels-0.342
Token a
Token ID257
Feature activation-2.42e+02

Pos logit contributions
 faded+0.221
chy+0.217
 White+0.216
 whispered+0.213
 meant+0.213
Neg logit contributions
iu-0.445
pieces-0.406
ric-0.377
 interests-0.370
nels-0.358
Token box
Token ID3091
Feature activation-2.32e+02

Pos logit contributions
 meant+0.316
chy+0.306
 White+0.306
 faded+0.295
 whispered+0.284
Neg logit contributions
iu-0.328
pieces-0.299
ric-0.271
 interests-0.265
nels-0.254
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
 White+0.234
 faded+0.230
chy+0.229
 meant+0.221
 whispered+0.216
Neg logit contributions
iu-0.402
pieces-0.364
 interests-0.326
ric-0.322
nels-0.312
Token She
Token ID1375
Feature activation-2.45e+02

Pos logit contributions
chy+0.331
 faded+0.302
 blocked+0.285
 meant+0.275
 melted+0.269
Neg logit contributions
iu-0.362
pieces-0.324
 interests-0.316
 Amen-0.288
nels-0.285
Token gets
Token ID3011
Feature activation-2.51e+02

Pos logit contributions
 whispered+0.237
 faded+0.235
chy+0.226
 led+0.224
 meant+0.217
Neg logit contributions
iu-0.449
pieces-0.421
ric-0.384
 interests-0.368
nels-0.357
Token some
Token ID617
Feature activation-2.50e+02

Pos logit contributions
chy+0.280
 faded+0.278
 White+0.266
 meant+0.259
 whispered+0.254
Neg logit contributions
iu-0.420
pieces-0.380
ric-0.335
 interests-0.330
nels-0.319
Token paper
Token ID3348
Feature activation-2.43e+02

Pos logit contributions
chy+0.333
 White+0.332
 faded+0.329
Dead+0.311
 meant+0.306
Neg logit contributions
iu-0.339
pieces-0.310
 interests-0.265
 occasions-0.254
ocol-0.252
Token.
Token ID13
Feature activation-2.64e+02

Pos logit contributions
 faded+0.215
chy+0.215
 White+0.211
 meant+0.205
 whispered+0.199
Neg logit contributions
iu-0.449
pieces-0.410
ric-0.375
 interests-0.369
nels-0.361
Token have
Token ID423
Feature activation-2.41e+02

Pos logit contributions
 faded+0.375
 White+0.371
chy+0.360
 meant+0.359
 whispered+0.355
Neg logit contributions
iu-0.265
pieces-0.238
ric-0.203
 interests-0.198
nels-0.178
Token fun
Token ID1257
Feature activation-2.38e+02

Pos logit contributions
chy+0.328
 White+0.327
 faded+0.323
 meant+0.308
Dead+0.305
Neg logit contributions
iu-0.326
pieces-0.300
 interests-0.276
ric-0.257
nels-0.239
Token.
Token ID13
Feature activation-2.50e+02

Pos logit contributions
 faded+0.132
 meant+0.120
chy+0.120
 White+0.120
 whispered+0.118
Neg logit contributions
iu-0.524
pieces-0.483
ric-0.460
 interests-0.441
earned-0.436
Token<|endoftext|>
Token ID50256
Feature activation-2.47e+02

Pos logit contributions
chy+0.273
 faded+0.265
 whispered+0.252
 blocked+0.243
 meant+0.243
Neg logit contributions
iu-0.416
pieces-0.372
 interests-0.364
ric-0.345
nels-0.331
TokenBen
Token ID11696
Feature activation-2.40e+02

Pos logit contributions
chy+0.155
 faded+0.155
 whispered+0.138
 White+0.130
 meant+0.130
Neg logit contributions
iu-0.515
pieces-0.476
ric-0.456
 interests-0.443
nels-0.434
Token likes
Token ID7832
Feature activation-2.54e+02

Pos logit contributions
 White+0.340
 faded+0.338
chy+0.330
Dead+0.326
 blocked+0.312
Neg logit contributions
iu-0.283
 interests-0.258
 disl-0.233
nels-0.229
ric-0.222
Token to
Token ID284
Feature activation-2.55e+02

Pos logit contributions
 faded+0.286
chy+0.282
 White+0.277
 meant+0.273
 whispered+0.270
Neg logit contributions
iu-0.415
pieces-0.375
ric-0.339
 interests-0.333
nels-0.320
Token play
Token ID711
Feature activation-2.56e+02

Pos logit contributions
 whispered+0.259
 answered+0.256
chy+0.255
 faded+0.249
 shy+0.248
Neg logit contributions
iu-0.444
pieces-0.383
ric-0.371
nels-0.338
 Amen-0.338
Token with
Token ID351
Feature activation-2.59e+02

Pos logit contributions
chy+0.213
 faded+0.212
 White+0.207
 whispered+0.205
 meant+0.202
Neg logit contributions
iu-0.497
pieces-0.450
ric-0.429
 interests-0.405
 Amen-0.386
Token his
Token ID465
Feature activation-2.52e+02

Pos logit contributions
 faded+0.254
chy+0.250
 White+0.247
 meant+0.242
 whispered+0.238
Neg logit contributions
iu-0.465
pieces-0.426
ric-0.388
 interests-0.379
nels-0.367
Token toy
Token ID13373
Feature activation-2.50e+02

Pos logit contributions
 faded+0.279
chy+0.277
 White+0.274
 meant+0.266
 whispered+0.264
Neg logit contributions
iu-0.418
pieces-0.378
ric-0.342
 interests-0.332
nels-0.319
Token He
Token ID679
Feature activation-2.33e+02

Pos logit contributions
chy+0.311
 faded+0.283
 spoke+0.282
 blocked+0.276
 whispered+0.275
Neg logit contributions
iu-0.330
 disl-0.280
 Majesty-0.278
 interests-0.276
ric-0.266
Token gave
Token ID2921
Feature activation-2.44e+02

Pos logit contributions
chy+0.261
 White+0.253
 Blood+0.233
 conditioning+0.231
Factory+0.228
Neg logit contributions
iu-0.336
 interests-0.293
pieces-0.267
nels-0.261
 wre-0.260
Token her
Token ID607
Feature activation-2.38e+02

Pos logit contributions
 whispered+0.156
 White+0.153
 meant+0.152
 faded+0.148
 shy+0.142
Neg logit contributions
iu-0.520
pieces-0.499
ric-0.466
 disl-0.457
 interests-0.456
Token the
Token ID262
Feature activation-2.42e+02

Pos logit contributions
 faded+0.199
chy+0.198
 White+0.194
 meant+0.189
 whispered+0.188
Neg logit contributions
iu-0.466
pieces-0.418
ric-0.388
 interests-0.373
nels-0.368
Token ball
Token ID2613
Feature activation-2.34e+02

Pos logit contributions
 faded+0.281
chy+0.277
 White+0.277
 whispered+0.269
 meant+0.267
Neg logit contributions
iu-0.375
pieces-0.343
 interests-0.316
ric-0.308
nels-0.295
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
 White+0.370
Dead+0.369
Ancient+0.348
 whispered+0.338
 faded+0.338
Neg logit contributions
iu-0.236
 interests-0.200
pieces-0.200
earned-0.181
nels-0.166
Token "
Token ID366
Feature activation-2.49e+02

Pos logit contributions
chy+0.359
 spoke+0.324
 blocked+0.324
 faded+0.316
 whispered+0.314
Neg logit contributions
iu-0.325
 disl-0.277
 Majesty-0.271
 interests-0.269
icol-0.249
TokenI
Token ID40
Feature activation-2.50e+02

Pos logit contributions
chy+0.296
 faded+0.263
 White+0.263
 meant+0.258
 dark+0.250
Neg logit contributions
iu-0.396
pieces-0.348
 Amen-0.334
 interests-0.332
 disl-0.321
Token'm
Token ID1101
Feature activation-2.50e+02

Pos logit contributions
 White+0.371
 faded+0.347
Dead+0.343
chy+0.340
Ancient+0.337
Neg logit contributions
iu-0.287
 interests-0.253
 disl-0.241
pieces-0.239
ric-0.223
Token sorry
Token ID7926
Feature activation-2.50e+02

Pos logit contributions
 faded+0.288
chy+0.284
 White+0.276
 meant+0.272
 whispered+0.268
Neg logit contributions
iu-0.405
pieces-0.365
ric-0.330
 interests-0.321
nels-0.309
Token,
Token ID11
Feature activation-2.63e+02

Pos logit contributions
 shy+0.121
chy+0.117
 White+0.116
 dark+0.109
 meant+0.109
Neg logit contributions
iu-0.593
pieces-0.569
ric-0.528
 interests-0.502
earned-0.495
Token bad
Token ID2089
Feature activation-2.30e+02

Pos logit contributions
 faded+0.271
chy+0.263
 White+0.260
 whispered+0.253
 meant+0.248
Neg logit contributions
iu-0.362
pieces-0.321
ric-0.301
 interests-0.293
nels-0.276
Token.
Token ID13
Feature activation-2.42e+02

Pos logit contributions
 faded+0.246
chy+0.242
 White+0.242
 whispered+0.227
 meant+0.224
Neg logit contributions
iu-0.386
pieces-0.344
 interests-0.321
ric-0.306
nels-0.298
Token It
Token ID632
Feature activation-2.34e+02

Pos logit contributions
 White+0.261
 meant+0.242
 faded+0.235
 shy+0.233
chy+0.228
Neg logit contributions
iu-0.424
pieces-0.406
ric-0.361
nels-0.342
 interests-0.334
Token hurt
Token ID5938
Feature activation-2.39e+02

Pos logit contributions
 faded+0.191
 White+0.187
chy+0.177
 whispered+0.173
Ancient+0.167
Neg logit contributions
iu-0.440
pieces-0.395
ric-0.372
 interests-0.364
nels-0.355
Token me
Token ID502
Feature activation-2.35e+02

Pos logit contributions
 faded+0.419
chy+0.409
Dead+0.397
 whispered+0.396
 White+0.395
Neg logit contributions
iu-0.228
pieces-0.183
 interests-0.176
ric-0.165
 occasions-0.149
Token."
Token ID526
Feature activation-2.54e+02

Pos logit contributions
 faded+0.326
Dead+0.312
 shone+0.307
 White+0.304
 spoke+0.302
Neg logit contributions
iu-0.316
pieces-0.274
ric-0.262
 interests-0.253
 Amen-0.234
Token Anna
Token ID11735
Feature activation-2.40e+02

Pos logit contributions
chy+0.304
 meant+0.282
 faded+0.280
 shone+0.277
 blocked+0.273
Neg logit contributions
iu-0.378
pieces-0.325
 Amen-0.307
 interests-0.298
 wre-0.277
Token said
Token ID531
Feature activation-2.52e+02

Pos logit contributions
chy+0.273
Today+0.239
 conditioning+0.233
 single+0.229
Ancient+0.228
Neg logit contributions
iu-0.400
pieces-0.295
 interests-0.287
 Amen-0.283
 wre-0.277
Token.
Token ID13
Feature activation-2.60e+02

Pos logit contributions
 faded+0.166
chy+0.163
 White+0.154
 whispered+0.153
 meant+0.150
Neg logit contributions
iu-0.532
pieces-0.500
ric-0.470
 interests-0.455
nels-0.445
Token
Token ID198
Feature activation-2.60e+02

Pos logit contributions
chy+0.389
 faded+0.359
 shone+0.345
 melted+0.341
 blocked+0.338
Neg logit contributions
iu-0.286
 interests-0.244
 Amen-0.242
pieces-0.228
ext-0.215
Token
Token ID198
Feature activation-2.61e+02

Pos logit contributions
 faded+0.225
chy+0.221
 meant+0.211
 White+0.198
 whispered+0.196
Neg logit contributions
iu-0.487
pieces-0.439
ric-0.418
 interests-0.413
nels-0.398
Token or
Token ID393
Feature activation-2.38e+02

Pos logit contributions
 faded+0.250
chy+0.248
 White+0.239
 meant+0.231
 whispered+0.230
Neg logit contributions
iu-0.424
pieces-0.389
 interests-0.350
ric-0.346
nels-0.334
Token toys
Token ID14958
Feature activation-2.35e+02

Pos logit contributions
 faded+0.263
chy+0.258
 White+0.253
 meant+0.251
 whispered+0.248
Neg logit contributions
iu-0.395
pieces-0.354
ric-0.325
 interests-0.315
nels-0.308
Token,
Token ID11
Feature activation-2.49e+02

Pos logit contributions
 faded+0.336
chy+0.330
 White+0.326
Dead+0.308
 whispered+0.308
Neg logit contributions
iu-0.297
pieces-0.276
 interests-0.246
ric-0.232
 occasions-0.227
Token or
Token ID393
Feature activation-2.43e+02

Pos logit contributions
 faded+0.282
chy+0.278
 White+0.271
 whispered+0.262
 meant+0.260
Neg logit contributions
iu-0.403
pieces-0.371
 interests-0.336
ric-0.333
nels-0.315
Token books
Token ID3835
Feature activation-2.34e+02

Pos logit contributions
 faded+0.264
chy+0.258
 meant+0.257
 White+0.254
 whispered+0.251
Neg logit contributions
iu-0.411
pieces-0.364
ric-0.337
nels-0.323
 interests-0.316
Token,"
Token ID553
Feature activation-2.54e+02

Pos logit contributions
 faded+0.343
 White+0.340
chy+0.335
 shone+0.328
 spoke+0.323
Neg logit contributions
iu-0.259
pieces-0.244
 interests-0.205
ric-0.198
 Amen-0.193
Token Max
Token ID5436
Feature activation-2.40e+02

Pos logit contributions
 faded+0.332
chy+0.331
 meant+0.304
 blocked+0.293
 shone+0.291
Neg logit contributions
iu-0.356
pieces-0.333
 interests-0.292
nels-0.275
ric-0.273
Token said
Token ID531
Feature activation-2.51e+02

Pos logit contributions
 faded+0.230
 dying+0.223
chy+0.214
 rotting+0.210
 sore+0.206
Neg logit contributions
iu-0.409
 interests-0.362
pieces-0.362
 disl-0.337
n-0.331
Token.
Token ID13
Feature activation-2.62e+02

Pos logit contributions
 faded+0.285
 White+0.282
chy+0.275
Dead+0.272
 meant+0.265
Neg logit contributions
iu-0.406
pieces-0.353
 interests-0.315
 Amen-0.303
ap-0.290
Token
Token ID198
Feature activation-2.61e+02

Pos logit contributions
chy+0.400
 wil+0.372
 faded+0.368
 shone+0.363
 spilled+0.358
Neg logit contributions
iu-0.265
 Amen-0.247
 interests-0.240
pieces-0.235
nels-0.212
Token
Token ID198
Feature activation-2.62e+02

Pos logit contributions
 faded+0.177
chy+0.174
 White+0.165
 meant+0.163
 whispered+0.159
Neg logit contributions
iu-0.543
pieces-0.502
ric-0.466
 interests-0.461
nels-0.447
Token followed
Token ID3940
Feature activation-2.44e+02

Pos logit contributions
 faded+0.161
chy+0.157
 White+0.152
 meant+0.148
 whispered+0.145
Neg logit contributions
iu-0.496
pieces-0.459
ric-0.425
 interests-0.420
nels-0.408
Token them
Token ID606
Feature activation-2.29e+02

Pos logit contributions
 White+0.264
 faded+0.260
 whispered+0.260
 meant+0.251
 shy+0.245
Neg logit contributions
iu-0.398
ric-0.378
pieces-0.361
 Majesty-0.333
 interests-0.330
Token to
Token ID284
Feature activation-2.42e+02

Pos logit contributions
 faded+0.233
chy+0.221
 White+0.218
 meant+0.210
 whispered+0.206
Neg logit contributions
iu-0.404
pieces-0.372
 interests-0.334
ric-0.330
nels-0.320
Token the
Token ID262
Feature activation-2.41e+02

Pos logit contributions
 White+0.176
 dark+0.167
 whispered+0.159
 shy+0.159
 faded+0.156
Neg logit contributions
iu-0.505
pieces-0.484
ric-0.431
nels-0.427
 interests-0.427
Token office
Token ID2607
Feature activation-2.29e+02

Pos logit contributions
 faded+0.282
chy+0.275
 White+0.270
 meant+0.268
 whispered+0.263
Neg logit contributions
iu-0.388
pieces-0.351
ric-0.312
 interests-0.312
nels-0.299
Token.
Token ID13
Feature activation-2.53e+02

Pos logit contributions
chy+0.240
 White+0.240
 faded+0.239
 meant+0.223
Dead+0.218
Neg logit contributions
iu-0.391
pieces-0.353
 interests-0.311
nels-0.299
ric-0.297
Token He
Token ID679
Feature activation-2.43e+02

Pos logit contributions
chy+0.202
 faded+0.197
 White+0.184
 meant+0.180
 whispered+0.178
Neg logit contributions
iu-0.497
pieces-0.452
 interests-0.422
ric-0.419
nels-0.407
Token looked
Token ID3114
Feature activation-2.49e+02

Pos logit contributions
chy+0.228
 White+0.221
 conditioning+0.191
 faded+0.191
 depress+0.186
Neg logit contributions
iu-0.434
pieces-0.366
 interests-0.363
nels-0.340
 wre-0.333
Token at
Token ID379
Feature activation-2.54e+02

Pos logit contributions
 faded+0.337
 White+0.328
chy+0.327
 meant+0.308
Dead+0.302
Neg logit contributions
iu-0.341
pieces-0.318
 interests-0.274
nels-0.260
ric-0.259
Token the
Token ID262
Feature activation-2.52e+02

Pos logit contributions
 White+0.255
 faded+0.253
 meant+0.246
chy+0.243
 whispered+0.238
Neg logit contributions
iu-0.462
pieces-0.418
ric-0.370
 interests-0.370
 Amen-0.361
Token computer
Token ID3644
Feature activation-2.39e+02

Pos logit contributions
 faded+0.317
 White+0.300
 whispered+0.298
chy+0.295
 meant+0.289
Neg logit contributions
iu-0.376
pieces-0.348
 interests-0.307
ric-0.298
nels-0.288
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenWhere
Token ID8496
Feature activation+3.58e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+3.91e+02

Pos logit contributions
iu+0.608
pieces+0.537
ric+0.491
 interests+0.467
nels+0.467
Neg logit contributions
 meant-0.397
chy-0.393
 faded-0.392
 whispered-0.380
 White-0.376
Token
Token ID26391
Feature activation+3.99e+02

Pos logit contributions
iu+0.446
pieces+0.385
ric+0.329
 interests+0.322
nels+0.300
Neg logit contributions
 faded-0.633
chy-0.628
 White-0.618
 meant-0.613
 whispered-0.609
Token
Token ID8151
Feature activation+4.58e+02

Pos logit contributions
iu+0.739
pieces+0.676
ric+0.619
 interests+0.611
nels+0.590
Neg logit contributions
 faded-0.362
chy-0.355
 White-0.348
 meant-0.340
 whispered-0.336
Tokens
Token ID82
Feature activation+3.71e+02

Pos logit contributions
ric+0.690
pieces+0.686
iu+0.641
 interests+0.579
n+0.562
Neg logit contributions
 faded-0.516
 White-0.497
chy-0.449
 whispered-0.449
Dead-0.444
Token my
Token ID616
Feature activation+3.67e+02

Pos logit contributions
iu+0.545
pieces+0.493
ric+0.441
 interests+0.432
nels+0.413
Neg logit contributions
 faded-0.472
chy-0.466
 White-0.457
 whispered-0.444
 meant-0.442
Token jelly
Token ID27644
Feature activation+3.69e+02

Pos logit contributions
iu+0.508
pieces+0.449
ric+0.412
 interests+0.411
 Majesty+0.379
Neg logit contributions
 faded-0.492
chy-0.488
 meant-0.475
 White-0.471
 whispered-0.467
Token?
Token ID30
Feature activation+3.05e+02

Pos logit contributions
iu+0.428
pieces+0.385
ric+0.348
 interests+0.338
nels+0.332
Neg logit contributions
 White-0.551
 faded-0.537
Dead-0.512
Ancient-0.510
 whispered-0.505
Tokenâ
Token ID22940
Feature activation+3.38e+02

Pos logit contributions
iu+0.378
pieces+0.371
ric+0.333
 Amen+0.305
 interests+0.291
Neg logit contributions
chy-0.461
 faded-0.455
 spilled-0.404
 whispered-0.400
 smelled-0.396
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenLet
Token ID5756
Feature activation+3.59e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.609
ric+0.521
pieces+0.511
 Amen+0.469
 interests+0.461
Neg logit contributions
 meant-0.426
 whispered-0.402
 faded-0.395
chy-0.386
 spoke-0.378
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.457
pieces+0.393
ric+0.335
 interests+0.329
nels+0.307
Neg logit contributions
 faded-0.649
chy-0.647
 White-0.636
 meant-0.632
 whispered-0.626
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.679
pieces+0.610
 interests+0.543
ric+0.541
nels+0.519
Neg logit contributions
 White-0.450
 faded-0.449
chy-0.445
 meant-0.433
 whispered-0.431
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
pieces+0.744
iu+0.728
ric+0.721
 interests+0.642
n+0.593
Neg logit contributions
 faded-0.517
 White-0.488
chy-0.469
 whispered-0.458
Dead-0.450
Token not
Token ID407
Feature activation+3.66e+02

Pos logit contributions
iu+0.591
pieces+0.501
ric+0.440
 interests+0.430
 Amen+0.419
Neg logit contributions
 meant-0.487
 faded-0.478
chy-0.476
 whispered-0.471
 White-0.470
Token think
Token ID892
Feature activation+3.44e+02

Pos logit contributions
iu+0.540
pieces+0.476
ric+0.426
 interests+0.413
nels+0.402
Neg logit contributions
 faded-0.475
chy-0.470
 meant-0.461
 White-0.458
 whispered-0.453
Token about
Token ID546
Feature activation+3.17e+02

Pos logit contributions
iu+0.449
pieces+0.428
 interests+0.368
ric+0.357
nels+0.346
Neg logit contributions
 faded-0.473
chy-0.468
 White-0.453
Dead-0.445
 whispered-0.441
Token the
Token ID262
Feature activation+3.17e+02

Pos logit contributions
iu+0.441
pieces+0.398
ric+0.357
 interests+0.351
nels+0.330
Neg logit contributions
 faded-0.430
chy-0.428
 White-0.410
 meant-0.404
 whispered-0.403
Token children
Token ID1751
Feature activation+3.44e+02

Pos logit contributions
Dead+0.275
Ancient+0.269
 Blood+0.269
chy+0.264
 White+0.247
Neg logit contributions
iu-0.362
pieces-0.313
 interests-0.301
icol-0.290
ric-0.289
Token put
Token ID1234
Feature activation+2.85e+02

Pos logit contributions
iu+0.554
pieces+0.546
ric+0.482
 interests+0.480
ap+0.427
Neg logit contributions
 White-0.355
chy-0.354
 faded-0.353
Dead-0.341
Ancient-0.341
Token on
Token ID319
Feature activation+3.11e+02

Pos logit contributions
iu+0.419
pieces+0.403
 interests+0.344
ric+0.337
nels+0.321
Neg logit contributions
 faded-0.347
chy-0.346
Dead-0.337
 White-0.333
Ancient-0.319
Token their
Token ID511
Feature activation+3.35e+02

Pos logit contributions
iu+0.458
pieces+0.434
 interests+0.375
ric+0.369
nels+0.350
Neg logit contributions
 faded-0.394
chy-0.391
 meant-0.375
 whispered-0.365
 White-0.364
Token bath
Token ID7837
Feature activation+3.26e+02

Pos logit contributions
iu+0.426
pieces+0.382
 interests+0.333
ric+0.322
nels+0.304
Neg logit contributions
 faded-0.492
 meant-0.488
 White-0.479
chy-0.477
Dead-0.465
Tokenro
Token ID305
Feature activation+4.45e+02

Pos logit contributions
iu+0.521
 interests+0.474
pieces+0.467
 occasions+0.451
ric+0.448
Neg logit contributions
Dead-0.348
 faded-0.339
 White-0.328
 meant-0.327
 blocked-0.323
Tokenbes
Token ID12636
Feature activation+3.82e+02

Pos logit contributions
iu+0.722
pieces+0.674
ric+0.585
 interests+0.576
nels+0.555
Neg logit contributions
chy-0.502
 faded-0.500
 whispered-0.482
 White-0.479
 meant-0.468
Token and
Token ID290
Feature activation+2.93e+02

Pos logit contributions
pieces+0.316
iu+0.289
 interests+0.280
!'+0.266
 err+0.265
Neg logit contributions
 Blood-0.629
Dead-0.628
Spider-0.616
Ancient-0.588
 faded-0.587
Token ran
Token ID4966
Feature activation+2.81e+02

Pos logit contributions
iu+0.468
pieces+0.439
 interests+0.399
ric+0.371
nels+0.369
Neg logit contributions
chy-0.298
Dead-0.280
Ancient-0.280
 White-0.280
 Blood-0.278
Token around
Token ID1088
Feature activation+2.80e+02

Pos logit contributions
iu+0.346
pieces+0.315
 interests+0.273
nels+0.259
ric+0.254
Neg logit contributions
 faded-0.402
 meant-0.399
chy-0.397
Dead-0.396
 White-0.394
Token in
Token ID287
Feature activation+2.71e+02

Pos logit contributions
pieces+0.269
iu+0.268
 occasions+0.229
 activities+0.218
 interests+0.216
Neg logit contributions
 meant-0.440
chy-0.439
Dead-0.436
 faded-0.436
 blocked-0.418
Token
Token ID129
Feature activation+5.14e+02

Pos logit contributions
iu+0.997
pieces+0.940
 interests+0.829
nels+0.821
ric+0.807
Neg logit contributions
chy-0.518
 faded-0.506
 White-0.500
 whispered-0.492
 meant-0.472
Token
Token ID241
Feature activation+4.42e+02

Pos logit contributions
iu+0.684
pieces+0.612
 interests+0.530
ric+0.528
nels+0.496
Neg logit contributions
chy-0.725
 faded-0.722
 White-0.711
 meant-0.700
 whispered-0.696
TokenLet
Token ID5756
Feature activation+3.59e+02

Pos logit contributions
iu+0.824
pieces+0.754
ric+0.705
 interests+0.688
nels+0.654
Neg logit contributions
 faded-0.398
chy-0.385
 White-0.373
 whispered-0.367
 meant-0.361
Tokenâ
Token ID22940
Feature activation+4.02e+02

Pos logit contributions
iu+0.609
ric+0.521
pieces+0.511
 Amen+0.469
 interests+0.461
Neg logit contributions
 meant-0.426
 whispered-0.402
 faded-0.395
chy-0.386
 spoke-0.378
Token
Token ID26391
Feature activation+4.12e+02

Pos logit contributions
iu+0.457
pieces+0.393
ric+0.335
 interests+0.329
nels+0.307
Neg logit contributions
 faded-0.649
chy-0.647
 White-0.636
 meant-0.632
 whispered-0.626
Token
Token ID8151
Feature activation+4.73e+02

Pos logit contributions
iu+0.679
pieces+0.610
 interests+0.543
ric+0.541
nels+0.519
Neg logit contributions
 White-0.450
 faded-0.449
chy-0.445
 meant-0.433
 whispered-0.431
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
pieces+0.744
iu+0.728
ric+0.721
 interests+0.642
n+0.593
Neg logit contributions
 faded-0.517
 White-0.488
chy-0.469
 whispered-0.458
Dead-0.450
Token explore
Token ID7301
Feature activation+3.61e+02

Pos logit contributions
iu+0.591
pieces+0.501
ric+0.440
 interests+0.430
 Amen+0.419
Neg logit contributions
 meant-0.487
 faded-0.478
chy-0.476
 whispered-0.471
 White-0.470
Token and
Token ID290
Feature activation+3.16e+02

Pos logit contributions
iu+0.537
pieces+0.507
 interests+0.451
ric+0.448
nels+0.436
Neg logit contributions
 faded-0.436
chy-0.434
 White-0.412
 meant-0.404
 whispered-0.396
Token have
Token ID423
Feature activation+3.10e+02

Pos logit contributions
iu+0.554
pieces+0.466
ric+0.452
 interests+0.429
 Amen+0.419
Neg logit contributions
 whispered-0.349
 faded-0.339
 meant-0.338
 White-0.334
chy-0.332
Token some
Token ID617
Feature activation+3.03e+02

Pos logit contributions
iu+0.509
pieces+0.467
 interests+0.422
ric+0.415
nels+0.397
Neg logit contributions
chy-0.340
 faded-0.340
 White-0.332
 meant-0.321
 whispered-0.316
Token
Token ID129
Feature activation+4.37e+02

Pos logit contributions
iu+0.854
pieces+0.786
ric+0.724
 interests+0.716
nels+0.693
Neg logit contributions
 faded-0.333
chy-0.326
 White-0.317
 meant-0.309
 whispered-0.305
Token
Token ID241
Feature activation+4.01e+02

Pos logit contributions
iu+0.545
pieces+0.456
ric+0.420
 disl+0.386
 interests+0.385
Neg logit contributions
 faded-0.682
chy-0.658
 meant-0.649
 White-0.643
 whispered-0.632
TokenLet
Token ID5756
Feature activation+3.41e+02

Pos logit contributions
iu+0.786
pieces+0.723
ric+0.660
 interests+0.653
nels+0.638
Neg logit contributions
 faded-0.318
chy-0.314
 White-0.306
 meant-0.301
 whispered-0.293
Tokenâ
Token ID22940
Feature activation+3.84e+02

Pos logit contributions
iu+0.647
ric+0.579
pieces+0.549
 Amen+0.521
ocol+0.507
Neg logit contributions
 meant-0.355
 whispered-0.334
 said-0.323
 answered-0.311
 faded-0.306
Token
Token ID26391
Feature activation+3.93e+02

Pos logit contributions
iu+0.432
pieces+0.371
ric+0.316
 interests+0.309
nels+0.289
Neg logit contributions
 faded-0.626
chy-0.622
 White-0.613
 meant-0.608
 whispered-0.603
Token
Token ID8151
Feature activation+4.53e+02

Pos logit contributions
iu+0.684
pieces+0.616
 interests+0.553
ric+0.549
 occasions+0.528
Neg logit contributions
 White-0.399
 faded-0.387
 whispered-0.378
chy-0.376
 meant-0.374
Tokens
Token ID82
Feature activation+3.66e+02

Pos logit contributions
pieces+0.725
iu+0.723
ric+0.697
 interests+0.635
earned+0.583
Neg logit contributions
 faded-0.478
 White-0.453
chy-0.448
 whispered-0.423
Dead-0.418
Token jump
Token ID4391
Feature activation+3.41e+02

Pos logit contributions
iu+0.584
pieces+0.484
ric+0.427
 Amen+0.415
 interests+0.414
Neg logit contributions
 meant-0.466
 whispered-0.453
 White-0.443
 faded-0.441
chy-0.438
Token over
Token ID625
Feature activation+3.25e+02

Pos logit contributions
iu+0.513
pieces+0.482
ric+0.432
nels+0.429
 interests+0.420
Neg logit contributions
 faded-0.396
chy-0.377
 White-0.377
 meant-0.367
 whispered-0.362
Token it
Token ID340
Feature activation+3.19e+02

Pos logit contributions
iu+0.473
pieces+0.442
ric+0.387
 interests+0.387
nels+0.375
Neg logit contributions
chy-0.420
 faded-0.407
 White-0.384
 meant-0.384
 whispered-0.374
Token,
Token ID11
Feature activation+2.83e+02

Pos logit contributions
pieces+0.312
iu+0.302
 interests+0.271
nels+0.269
ric+0.262
Neg logit contributions
 faded-0.464
Dead-0.454
 Blood-0.451
 White-0.450
 choking-0.445
Token a
Token ID257
Feature activation+3.76e+02

Pos logit contributions
Spider+0.399
Doctor+0.384
Pr+0.369
 White+0.367
Ancient+0.366
Neg logit contributions
 how-0.205
pieces-0.203
 interests-0.163
ric-0.138
 what-0.137
Token time
Token ID640
Feature activation+3.33e+02

Pos logit contributions
iu+0.613
pieces+0.566
 interests+0.516
ric+0.508
 Amen+0.489
Neg logit contributions
 White-0.385
 faded-0.385
 meant-0.379
chy-0.376
 whispered-0.368
Token there
Token ID612
Feature activation+3.15e+02

Pos logit contributions
iu+0.520
pieces+0.481
 interests+0.429
ric+0.419
nels+0.410
Neg logit contributions
chy-0.392
 White-0.377
 faded-0.373
 meant-0.344
Dead-0.340
Token was
Token ID373
Feature activation+3.01e+02

Pos logit contributions
ap+0.284
pieces+0.274
n+0.266
ric+0.263
iu+0.258
Neg logit contributions
chy-0.491
 sore-0.487
 swollen-0.472
 White-0.464
 stiff-0.461
Token a
Token ID257
Feature activation+3.00e+02

Pos logit contributions
iu+0.414
pieces+0.390
 interests+0.344
ric+0.329
earned+0.304
Neg logit contributions
 faded-0.377
chy-0.374
 whispered-0.363
 White-0.362
 meant-0.347
Token blue
Token ID4171
Feature activation+3.12e+02

Pos logit contributions
iu+0.449
pieces+0.407
 Amen+0.360
 Majesty+0.356
nels+0.354
Neg logit contributions
 meant-0.354
chy-0.349
 faded-0.344
 whispered-0.325
 depress-0.314
Token symbol
Token ID6194
Feature activation+3.10e+02

Pos logit contributions
iu+0.522
pieces+0.473
ric+0.436
 interests+0.421
nels+0.407
Neg logit contributions
chy-0.332
 faded-0.331
 meant-0.321
 White-0.316
 whispered-0.314
Token.
Token ID13
Feature activation+2.47e+02

Pos logit contributions
iu+0.433
pieces+0.403
ric+0.328
 interests+0.327
nels+0.310
Neg logit contributions
 White-0.391
chy-0.377
 faded-0.374
Ancient-0.370
 meant-0.355
Token The
Token ID383
Feature activation+2.63e+02

Pos logit contributions
iu+0.328
pieces+0.317
 Amen+0.261
 interests+0.257
ric+0.250
Neg logit contributions
chy-0.371
 faded-0.354
 spilled-0.327
 sore-0.316
 blocked-0.309
Token symbol
Token ID6194
Feature activation+2.75e+02

Pos logit contributions
iu+0.391
pieces+0.347
ric+0.305
 interests+0.304
nels+0.282
Neg logit contributions
chy-0.341
 faded-0.335
 White-0.328
 meant-0.322
 whispered-0.316
Token was
Token ID373
Feature activation+2.52e+02

Pos logit contributions
iu+0.445
pieces+0.423
 interests+0.377
nels+0.336
 Amen+0.328
Neg logit contributions
 White-0.291
chy-0.280
 depress-0.260
Ancient-0.256
 Hen-0.253
Token thought
Token ID1807
Feature activation+2.72e+02

Pos logit contributions
iu+0.376
 disl+0.348
 interests+0.340
pieces+0.338
 Majesty+0.323
Neg logit contributions
 White-0.379
Ancient-0.358
jar-0.353
Today-0.347
 Blood-0.347
Token she
Token ID673
Feature activation+2.68e+02

Pos logit contributions
iu+0.311
pieces+0.308
 interests+0.246
 occasions+0.222
nels+0.219
Neg logit contributions
Dead-0.399
Ancient-0.398
 White-0.393
 faded-0.390
 blocked-0.378
Token must
Token ID1276
Feature activation+2.60e+02

Pos logit contributions
iu+0.371
pieces+0.362
 interests+0.337
ric+0.317
nels+0.302
Neg logit contributions
 White-0.294
 faded-0.293
 late-0.286
Dead-0.283
Ancient-0.279
Token have
Token ID423
Feature activation+2.52e+02

Pos logit contributions
iu+0.389
pieces+0.350
ric+0.315
 interests+0.284
nels+0.280
Neg logit contributions
 meant-0.347
chy-0.338
 faded-0.337
 whispered-0.335
 answered-0.322
Token done
Token ID1760
Feature activation+2.51e+02

Pos logit contributions
iu+0.307
pieces+0.298
 interests+0.288
 occasions+0.255
n+0.238
Neg logit contributions
 White-0.359
chy-0.356
Ancient-0.331
 Hen-0.330
 depress-0.327
Token something
Token ID1223
Feature activation+2.44e+02

Pos logit contributions
iu+0.329
pieces+0.313
 interests+0.260
ric+0.258
 disl+0.251
Neg logit contributions
 faded-0.359
chy-0.352
 White-0.342
 whispered-0.331
 meant-0.326
Token wrong
Token ID2642
Feature activation+2.51e+02

Pos logit contributions
pieces+0.305
iu+0.272
 interests+0.266
 disl+0.255
 occasions+0.232
Neg logit contributions
 White-0.356
 faded-0.336
Dead-0.326
 Blood-0.325
mail-0.324
Token and
Token ID290
Feature activation+2.27e+02

Pos logit contributions
iu+0.396
pieces+0.367
 interests+0.352
ric+0.323
 occasions+0.321
Neg logit contributions
 faded-0.281
 White-0.274
chy-0.268
Dead-0.259
 blocked-0.253
Token started
Token ID2067
Feature activation+2.26e+02

Pos logit contributions
iu+0.468
pieces+0.430
ric+0.401
 interests+0.394
nels+0.383
Neg logit contributions
 faded-0.163
chy-0.159
 White-0.152
 meant-0.152
 whispered-0.149
Token to
Token ID284
Feature activation+2.20e+02

Pos logit contributions
iu+0.332
pieces+0.305
ric+0.276
 interests+0.271
 disl+0.253
Neg logit contributions
 faded-0.281
chy-0.278
 White-0.274
 meant-0.267
 whispered-0.262
Token cry
Token ID3960
Feature activation+2.28e+02

Pos logit contributions
iu+0.324
pieces+0.288
ric+0.255
 interests+0.251
nels+0.241
Neg logit contributions
 faded-0.285
chy-0.284
 White-0.277
 meant-0.275
 whispered-0.273
Token She
Token ID1375
Feature activation+2.81e+02

Pos logit contributions
iu+0.439
pieces+0.410
 Amen+0.375
ric+0.374
 interests+0.369
Neg logit contributions
chy-0.351
 faded-0.339
 meant-0.307
 spilled-0.301
 blocked-0.298
Token had
Token ID550
Feature activation+2.72e+02

Pos logit contributions
iu+0.461
pieces+0.422
 interests+0.405
ric+0.389
 disl+0.377
Neg logit contributions
 White-0.270
chy-0.259
 faded-0.247
Ancient-0.247
 Blood-0.245
Token chosen
Token ID7147
Feature activation+2.90e+02

Pos logit contributions
 interests+0.301
iu+0.297
pieces+0.295
 disl+0.250
 occasions+0.244
Neg logit contributions
 White-0.376
Ancient-0.360
chy-0.352
 depress-0.351
 faded-0.345
Token a
Token ID257
Feature activation+2.75e+02

Pos logit contributions
pieces+0.322
iu+0.309
 interests+0.284
ric+0.256
 how+0.254
Neg logit contributions
chy-0.425
 faded-0.424
Dead-0.410
 depress-0.401
Ancient-0.391
Token good
Token ID922
Feature activation+2.82e+02

Pos logit contributions
iu+0.353
pieces+0.345
 interests+0.336
ric+0.300
 Majesty+0.299
Neg logit contributions
 meant-0.339
chy-0.329
 White-0.324
 faded-0.324
 whispered-0.322
Token toy
Token ID13373
Feature activation+2.89e+02

Pos logit contributions
iu+0.436
pieces+0.392
ric+0.352
 interests+0.345
nels+0.331
Neg logit contributions
 faded-0.343
chy-0.339
 White-0.332
 meant-0.328
 whispered-0.325
Token and
Token ID290
Feature activation+2.59e+02

Pos logit contributions
iu+0.343
pieces+0.341
 occasions+0.264
nels+0.256
 interests+0.255
Neg logit contributions
 faded-0.397
Dead-0.394
 Blood-0.382
chy-0.380
 smelled-0.379
Token a
Token ID257
Feature activation+2.66e+02

Pos logit contributions
iu+0.471
pieces+0.436
 interests+0.399
ric+0.390
nels+0.378
Neg logit contributions
 faded-0.230
chy-0.228
 White-0.226
 meant-0.211
 whispered-0.211
Token good
Token ID922
Feature activation+2.73e+02

Pos logit contributions
iu+0.370
pieces+0.359
 interests+0.344
 Majesty+0.316
 Amen+0.314
Neg logit contributions
 meant-0.307
 whispered-0.289
chy-0.288
 White-0.287
Dead-0.279
Token price
Token ID2756
Feature activation+2.80e+02

Pos logit contributions
iu+0.451
pieces+0.404
 interests+0.368
ric+0.364
nels+0.347
Neg logit contributions
 faded-0.299
chy-0.296
 White-0.292
 meant-0.286
 whispered-0.285
Token.
Token ID13
Feature activation+2.25e+02

Pos logit contributions
pieces+0.372
iu+0.369
 interests+0.307
 occasions+0.305
nels+0.274
Neg logit contributions
Dead-0.370
chy-0.363
 White-0.362
 faded-0.353
 Blood-0.340
Token her
Token ID607
Feature activation+2.46e+02

Pos logit contributions
iu+0.269
pieces+0.268
 interests+0.222
 occasions+0.213
ocol+0.191
Neg logit contributions
chy-0.370
Dead-0.370
 faded-0.368
 blocked-0.358
 shone-0.346
Token if
Token ID611
Feature activation+2.34e+02

Pos logit contributions
pieces+0.324
iu+0.323
 interests+0.288
ric+0.262
 Majesty+0.249
Neg logit contributions
 faded-0.335
chy-0.334
 blocked-0.312
Dead-0.309
 White-0.309
Token she
Token ID673
Feature activation+2.40e+02

Pos logit contributions
iu+0.428
pieces+0.391
 interests+0.358
nels+0.342
ric+0.341
Neg logit contributions
chy-0.210
 faded-0.198
 White-0.190
 meant-0.183
 whispered-0.180
Token made
Token ID925
Feature activation+2.26e+02

Pos logit contributions
iu+0.330
 interests+0.305
pieces+0.293
earned+0.266
 spend+0.255
Neg logit contributions
 White-0.280
 Blood-0.268
 darker-0.265
 dark-0.264
 depress-0.260
Token it
Token ID340
Feature activation+2.34e+02

Pos logit contributions
iu+0.343
pieces+0.303
ric+0.276
 interests+0.272
nels+0.258
Neg logit contributions
 faded-0.280
chy-0.277
 White-0.276
 meant-0.272
 whispered-0.269
Token herself
Token ID5223
Feature activation+2.39e+02

Pos logit contributions
iu+0.336
pieces+0.330
 interests+0.282
ric+0.280
nels+0.260
Neg logit contributions
 faded-0.288
 White-0.286
Dead-0.274
 whispered-0.269
chy-0.268
Token.
Token ID13
Feature activation+1.96e+02

Pos logit contributions
iu+0.351
pieces+0.350
ric+0.301
 interests+0.301
 disl+0.277
Neg logit contributions
 White-0.288
 faded-0.281
chy-0.266
Dead-0.265
 whispered-0.260
Token L
Token ID406
Feature activation+2.32e+02

Pos logit contributions
iu+0.281
pieces+0.268
 interests+0.229
ric+0.220
 disl+0.219
Neg logit contributions
chy-0.270
 faded-0.256
 blocked-0.234
 meant-0.229
 spilled-0.229
Tokenila
Token ID10102
Feature activation+2.20e+02

Pos logit contributions
iu+0.398
pieces+0.364
ric+0.335
 interests+0.321
nels+0.306
Neg logit contributions
 faded-0.250
chy-0.245
 White-0.242
 meant-0.230
 whispered-0.225
Token said
Token ID531
Feature activation+2.07e+02

Pos logit contributions
iu+0.312
pieces+0.293
 interests+0.282
n+0.252
 how+0.241
Neg logit contributions
 White-0.261
 dark-0.255
 faded-0.252
 darker-0.252
chy-0.251
Token no
Token ID645
Feature activation+2.07e+02

Pos logit contributions
iu+0.224
pieces+0.207
 how+0.172
 interests+0.162
 Majesty+0.157
Neg logit contributions
 White-0.330
chy-0.327
Ancient-0.311
Factory-0.310
lab-0.308
Token baby
Token ID5156
Feature activation+3.10e+02

Pos logit contributions
iu+0.449
pieces+0.407
 Amen+0.360
 Majesty+0.356
nels+0.354
Neg logit contributions
 meant-0.354
chy-0.349
 faded-0.344
 whispered-0.325
 depress-0.314
Token bear
Token ID6842
Feature activation+2.98e+02

Pos logit contributions
iu+0.460
pieces+0.421
 Amen+0.376
ric+0.372
 interests+0.345
Neg logit contributions
 faded-0.361
chy-0.349
Ancient-0.338
 Blood-0.337
Dead-0.331
Token.
Token ID13
Feature activation+2.44e+02

Pos logit contributions
iu+0.335
pieces+0.315
ric+0.305
 Amen+0.258
 interests+0.254
Neg logit contributions
 faded-0.438
Dead-0.417
Ancient-0.416
 Blood-0.412
chy-0.405
Token He
Token ID679
Feature activation+2.62e+02

Pos logit contributions
iu+0.362
pieces+0.334
 Amen+0.308
ric+0.307
 interests+0.291
Neg logit contributions
 faded-0.308
chy-0.306
 spilled-0.275
 meant-0.269
 shone-0.265
Token was
Token ID373
Feature activation+2.52e+02

Pos logit contributions
iu+0.291
 interests+0.268
ric+0.260
 Amen+0.252
n+0.251
Neg logit contributions
Ancient-0.340
sheet-0.335
 conditioning-0.319
 depress-0.319
jar-0.310
Token very
Token ID845
Feature activation+2.53e+02

Pos logit contributions
iu+0.398
pieces+0.362
ric+0.347
 interests+0.318
 Amen+0.318
Neg logit contributions
 faded-0.280
chy-0.260
Dead-0.256
Ancient-0.254
 White-0.248
Token small
Token ID1402
Feature activation+2.67e+02

Pos logit contributions
iu+0.425
pieces+0.371
 disl+0.364
 interests+0.358
 Amen+0.355
Neg logit contributions
 faded-0.233
 White-0.224
Dead-0.214
mail-0.213
 led-0.211
Token and
Token ID290
Feature activation+2.42e+02

Pos logit contributions
iu+0.430
pieces+0.389
 interests+0.366
ric+0.352
nels+0.333
Neg logit contributions
 faded-0.279
chy-0.270
Dead-0.261
 White-0.261
 whispered-0.252
Token afraid
Token ID7787
Feature activation+2.55e+02

Pos logit contributions
iu+0.384
pieces+0.354
 Amen+0.343
ric+0.337
 disl+0.329
Neg logit contributions
 faded-0.223
Ancient-0.217
 depress-0.214
 White-0.208
Factory-0.205
Token.
Token ID13
Feature activation+2.20e+02

Pos logit contributions
iu+0.347
pieces+0.327
ric+0.293
 interests+0.292
nels+0.265
Neg logit contributions
 faded-0.314
chy-0.311
 whispered-0.307
Ancient-0.303
Dead-0.298
Token He
Token ID679
Feature activation+2.35e+02

Pos logit contributions
iu+0.314
pieces+0.284
ric+0.267
 interests+0.254
 Amen+0.252
Neg logit contributions
 faded-0.304
chy-0.293
 spilled-0.270
 shone-0.268
 blew-0.261
Token said
Token ID531
Feature activation+1.31e+02

Pos logit contributions
iu+0.226
pieces+0.197
 interests+0.190
ric+0.176
n+0.172
Neg logit contributions
 White-0.129
Ancient-0.129
chy-0.126
Dead-0.126
 Blood-0.124
Token goodbye
Token ID24829
Feature activation+1.29e+02

Pos logit contributions
iu+0.108
pieces+0.090
 interests+0.071
 Majesty+0.068
ap+0.061
Neg logit contributions
chy-0.244
 White-0.242
Ancient-0.235
 meant-0.234
Dead-0.233
Token to
Token ID284
Feature activation+1.23e+02

Pos logit contributions
iu+0.170
 interests+0.132
pieces+0.131
icol+0.120
ric+0.119
Neg logit contributions
chy-0.177
 meant-0.177
 faded-0.176
Dead-0.174
 blocked-0.171
Token Andy
Token ID12382
Feature activation+1.40e+02

Pos logit contributions
iu+0.277
pieces+0.249
 interests+0.240
nels+0.237
ric+0.233
Neg logit contributions
 White-0.068
 meant-0.067
 whispered-0.067
chy-0.062
 faded-0.061
Token and
Token ID290
Feature activation+1.15e+02

Pos logit contributions
iu+0.172
pieces+0.167
ric+0.139
 interests+0.136
 spend+0.134
Neg logit contributions
Dead-0.207
 Blood-0.201
Ancient-0.194
 faded-0.188
 blocked-0.188
Token thanked
Token ID26280
Feature activation+1.14e+02

Pos logit contributions
iu+0.229
pieces+0.211
ric+0.195
 interests+0.192
nels+0.186
Neg logit contributions
 faded-0.088
chy-0.086
 White-0.084
 meant-0.082
 whispered-0.080
Token him
Token ID683
Feature activation+1.21e+02

Pos logit contributions
iu+0.179
pieces+0.163
ric+0.148
 interests+0.147
nels+0.139
Neg logit contributions
 White-0.137
 faded-0.133
 meant-0.132
 whispered-0.130
chy-0.130
Token again
Token ID757
Feature activation+1.16e+02

Pos logit contributions
pieces+0.133
iu+0.132
 interests+0.101
ric+0.095
nels+0.091
Neg logit contributions
Dead-0.202
 faded-0.202
 blocked-0.197
 smoked-0.194
 Blood-0.190
Token for
Token ID329
Feature activation+1.05e+02

Pos logit contributions
iu+0.163
pieces+0.152
 interests+0.133
ric+0.127
 occasions+0.122
Neg logit contributions
chy-0.150
 blocked-0.141
 faded-0.139
 meant-0.138
Dead-0.138
Token the
Token ID262
Feature activation+1.08e+02

Pos logit contributions
iu+0.150
pieces+0.132
ric+0.115
 Amen+0.112
 interests+0.112
Neg logit contributions
chy-0.143
 faded-0.133
Ancient-0.131
 meant-0.131
 White-0.130
Token lime
Token ID28738
Feature activation+1.16e+02

Pos logit contributions
iu+0.147
pieces+0.125
 interests+0.116
 impression+0.102
 occasions+0.100
Neg logit contributions
 White-0.150
 faded-0.147
 whispered-0.145
 meant-0.145
Dead-0.145
Token wanted
Token ID2227
Feature activation+1.88e+01

Pos logit contributions
iu+0.031
 spend+0.029
 spent+0.025
n+0.024
 interests+0.024
Neg logit contributions
chy-0.045
 bleeding-0.045
 swollen-0.045
ching-0.045
lessness-0.045
Token to
Token ID284
Feature activation+1.78e+01

Pos logit contributions
iu+0.018
pieces+0.016
 interests+0.015
 wre+0.014
 occasions+0.014
Neg logit contributions
chy-0.033
 faded-0.031
opy-0.029
Dead-0.029
 startled-0.029
Token make
Token ID787
Feature activation+1.01e+01

Pos logit contributions
iu+0.028
pieces+0.025
ric+0.024
nels+0.022
 interests+0.022
Neg logit contributions
 meant-0.021
 whispered-0.021
 faded-0.021
 White-0.020
 answered-0.020
Token more
Token ID517
Feature activation+1.42e+01

Pos logit contributions
iu+0.018
pieces+0.016
ric+0.015
 interests+0.015
nels+0.014
Neg logit contributions
 faded-0.010
 White-0.010
 meant-0.010
chy-0.010
 whispered-0.010
Token money
Token ID1637
Feature activation+2.38e+01

Pos logit contributions
iu+0.020
pieces+0.018
 interests+0.016
ric+0.015
nels+0.014
Neg logit contributions
 faded-0.019
chy-0.019
 White-0.019
 whispered-0.018
 meant-0.018
Token so
Token ID523
Feature activation+6.23e+00

Pos logit contributions
iu+0.037
pieces+0.034
 interests+0.031
ric+0.029
nels+0.029
Neg logit contributions
 faded-0.027
Dead-0.027
 White-0.027
chy-0.027
 whispered-0.025
Token he
Token ID339
Feature activation+1.14e+01

Pos logit contributions
iu+0.012
pieces+0.011
 interests+0.010
ric+0.009
nels+0.009
Neg logit contributions
chy-0.006
 faded-0.006
 White-0.005
 whispered-0.005
 meant-0.005
Token decided
Token ID3066
Feature activation+1.59e+00

Pos logit contributions
iu+0.017
pieces+0.014
n+0.013
 interests+0.013
 spend+0.013
Neg logit contributions
 White-0.013
 swallowing-0.013
Dead-0.013
 faded-0.013
 bleeding-0.012
Token to
Token ID284
Feature activation-8.22e-02

Pos logit contributions
iu+0.002
pieces+0.002
 interests+0.002
ric+0.002
nels+0.002
Neg logit contributions
 faded-0.002
chy-0.002
 White-0.002
 meant-0.002
 whispered-0.002
Token raise
Token ID5298
Feature activation-7.92e-01

Pos logit contributions
 meant+0.000
 White+0.000
 faded+0.000
 whispered+0.000
chy+0.000
Neg logit contributions
iu-0.000
pieces-0.000
ric-0.000
 interests-0.000
nels-0.000
Token some
Token ID617
Feature activation-8.41e-01

Pos logit contributions
 faded+0.001
chy+0.001
 White+0.001
 meant+0.001
 whispered+0.001
Neg logit contributions
iu-0.001
pieces-0.001
ric-0.001
 interests-0.001
nels-0.001
Token trying
Token ID2111
Feature activation+1.04e+02

Pos logit contributions
iu+0.145
pieces+0.131
ric+0.122
 interests+0.118
 accuracy+0.109
Neg logit contributions
 White-0.128
 faded-0.121
Ancient-0.120
chy-0.119
 meant-0.118
Token until
Token ID1566
Feature activation+9.04e+01

Pos logit contributions
iu+0.132
pieces+0.128
 interests+0.117
ric+0.115
nels+0.106
Neg logit contributions
 faded-0.138
 White-0.134
 meant-0.131
Dead-0.131
 whispered-0.129
Token her
Token ID607
Feature activation+1.02e+02

Pos logit contributions
iu+0.106
pieces+0.102
 Wed+0.088
n+0.085
 spend+0.079
Neg logit contributions
chy-0.124
Ancient-0.119
umpy-0.118
 meant-0.118
Old-0.117
Token toes
Token ID23932
Feature activation+1.19e+02

Pos logit contributions
iu+0.127
pieces+0.106
 interests+0.102
ric+0.095
nels+0.089
Neg logit contributions
 faded-0.151
chy-0.146
 meant-0.145
 White-0.144
 whispered-0.144
Token touched
Token ID12615
Feature activation+9.64e+01

Pos logit contributions
ric+0.138
iu+0.136
n+0.131
 interests+0.128
pieces+0.121
Neg logit contributions
 White-0.155
Ancient-0.147
Doctor-0.146
 Green-0.145
Old-0.142
Token the
Token ID262
Feature activation+9.10e+01

Pos logit contributions
iu+0.102
pieces+0.100
ric+0.091
 interests+0.078
nels+0.077
Neg logit contributions
Dead-0.144
 whispered-0.139
 faded-0.138
 spoke-0.137
Ancient-0.137
Token ground
Token ID2323
Feature activation+1.03e+02

Pos logit contributions
iu+0.117
pieces+0.109
ric+0.100
 interests+0.093
 Majesty+0.093
Neg logit contributions
 faded-0.119
 spoke-0.116
 whispered-0.116
 meant-0.114
 answered-0.113
Token.
Token ID13
Feature activation+7.08e+01

Pos logit contributions
pieces+0.080
iu+0.067
 how+0.067
 as+0.063
ap+0.063
Neg logit contributions
Dead-0.176
 White-0.167
Ancient-0.166
Old-0.165
Doctor-0.163
Token
Token ID198
Feature activation+7.44e+01

Pos logit contributions
iu+0.084
pieces+0.081
ric+0.078
 interests+0.069
n+0.067
Neg logit contributions
 faded-0.103
chy-0.103
 wicked-0.097
 blocked-0.096
 smelled-0.094
Token
Token ID198
Feature activation+7.33e+01

Pos logit contributions
iu+0.098
pieces+0.093
ric+0.087
 interests+0.085
ext+0.082
Neg logit contributions
 faded-0.103
chy-0.100
 meant-0.092
 blocked-0.090
 shone-0.090
TokenSuddenly
Token ID38582
Feature activation+7.85e+01

Pos logit contributions
iu+0.107
pieces+0.104
 interests+0.091
nels+0.090
ric+0.089
Neg logit contributions
 faded-0.091
chy-0.091
 meant-0.089
 whispered-0.084
 White-0.082
Token the
Token ID262
Feature activation+8.30e+00

Pos logit contributions
iu+0.007
ric+0.006
pieces+0.006
 how+0.005
 interests+0.005
Neg logit contributions
 faded-0.017
 spoke-0.017
chy-0.017
Dead-0.017
 blocked-0.017
Token iron
Token ID6953
Feature activation+1.82e+01

Pos logit contributions
iu+0.012
pieces+0.011
ric+0.010
 interests+0.010
nels+0.010
Neg logit contributions
 faded-0.011
chy-0.010
 White-0.010
 whispered-0.010
 meant-0.010
Token.
Token ID13
Feature activation-9.50e+00

Pos logit contributions
iu+0.026
pieces+0.024
ric+0.021
nels+0.019
ode+0.019
Neg logit contributions
 faded-0.022
Dead-0.022
 White-0.022
 Blood-0.022
 blocked-0.021
Token
Token ID198
Feature activation-7.55e+00

Pos logit contributions
chy+0.015
 faded+0.014
 meant+0.013
 melted+0.013
 shone+0.013
Neg logit contributions
iu-0.010
ric-0.008
 interests-0.008
ext-0.007
ap-0.007
Token
Token ID198
Feature activation-9.20e+00

Pos logit contributions
 faded+0.006
chy+0.006
 White+0.005
 meant+0.005
 whispered+0.005
Neg logit contributions
iu-0.015
pieces-0.014
ric-0.013
 interests-0.013
nels-0.012
Token"
Token ID1
Feature activation-6.51e+00

Pos logit contributions
 whispered+0.006
 White+0.006
 faded+0.006
chy+0.006
Dead+0.006
Neg logit contributions
iu-0.019
pieces-0.018
ric-0.017
 interests-0.016
 impression-0.016
TokenLet
Token ID5756
Feature activation-8.65e+00

Pos logit contributions
chy+0.009
 sleeping+0.008
 White+0.008
 answered+0.008
 meant+0.008
Neg logit contributions
iu-0.008
nels-0.007
 disl-0.007
 interests-0.007
pieces-0.006
Token's
Token ID338
Feature activation-9.34e+00

Pos logit contributions
 said+0.005
 meant+0.005
 hurt+0.005
 whispered+0.004
 led+0.004
Neg logit contributions
iu-0.021
ric-0.020
pieces-0.020
 interests-0.019
ocol-0.019
Token take
Token ID1011
Feature activation-1.60e+01

Pos logit contributions
 hurt+0.009
 broke+0.009
 closed+0.008
 whispered+0.008
 turned+0.008
Neg logit contributions
iu-0.018
pieces-0.018
ric-0.015
 interests-0.015
 occasions-0.015
Token it
Token ID340
Feature activation-8.69e+00

Pos logit contributions
 White+0.015
 whispered+0.015
 faded+0.015
 dark+0.014
 meant+0.014
Neg logit contributions
iu-0.029
pieces-0.027
ric-0.025
 interests-0.024
 Amen-0.023
Token to
Token ID284
Feature activation-1.91e+01

Pos logit contributions
 faded+0.012
 White+0.012
Dead+0.012
chy+0.011
 meant+0.011
Neg logit contributions
iu-0.011
pieces-0.010
 interests-0.009
nels-0.008
ric-0.008
Token thanked
Token ID26280
Feature activation+4.09e+01

Pos logit contributions
iu+0.078
pieces+0.068
 disl+0.064
nels+0.063
 interests+0.062
Neg logit contributions
chy-0.039
Ancient-0.037
 White-0.037
 faded-0.035
Dead-0.035
Token Benny
Token ID44275
Feature activation+5.44e+01

Pos logit contributions
iu+0.071
pieces+0.065
ric+0.059
 interests+0.058
nels+0.056
Neg logit contributions
 faded-0.042
chy-0.042
 White-0.041
 meant-0.040
 whispered-0.040
Token for
Token ID329
Feature activation+3.68e+01

Pos logit contributions
iu+0.083
pieces+0.076
 interests+0.067
ric+0.067
nels+0.063
Neg logit contributions
 faded-0.067
chy-0.064
 meant-0.063
 White-0.063
 whispered-0.062
Token saving
Token ID8914
Feature activation+3.69e+01

Pos logit contributions
iu+0.056
pieces+0.049
ric+0.044
 interests+0.043
 Amen+0.043
Neg logit contributions
chy-0.046
 faded-0.045
 White-0.043
 meant-0.042
 whispered-0.042
Token him
Token ID683
Feature activation+4.01e+01

Pos logit contributions
iu+0.055
pieces+0.049
ric+0.044
 interests+0.043
nels+0.041
Neg logit contributions
 faded-0.048
chy-0.047
 White-0.047
 meant-0.046
 whispered-0.045
Token.
Token ID13
Feature activation+1.44e+01

Pos logit contributions
iu+0.054
pieces+0.052
 interests+0.046
 impression+0.045
ric+0.044
Neg logit contributions
Dead-0.053
 faded-0.051
 Blood-0.051
 shone-0.050
Ancient-0.049
Token Benny
Token ID44275
Feature activation+3.78e+01

Pos logit contributions
iu+0.026
pieces+0.023
ric+0.022
 interests+0.021
nels+0.021
Neg logit contributions
 faded-0.014
chy-0.014
 meant-0.013
 White-0.013
 whispered-0.013
Token felt
Token ID2936
Feature activation+2.01e+01

Pos logit contributions
iu+0.060
pieces+0.051
 interests+0.048
ric+0.047
nels+0.046
Neg logit contributions
Ancient-0.039
chy-0.039
Dead-0.038
 White-0.037
 Blood-0.037
Token amazing
Token ID4998
Feature activation+2.50e+01

Pos logit contributions
iu+0.039
pieces+0.036
ric+0.033
 interests+0.033
nels+0.032
Neg logit contributions
 faded-0.017
chy-0.016
 White-0.016
 meant-0.015
 whispered-0.015
Token knowing
Token ID6970
Feature activation+1.59e+01

Pos logit contributions
iu+0.049
pieces+0.047
ric+0.043
 interests+0.041
nels+0.040
Neg logit contributions
 faded-0.021
chy-0.020
 White-0.019
 meant-0.019
 whispered-0.019
Token he
Token ID339
Feature activation+1.81e+01

Pos logit contributions
iu+0.030
pieces+0.028
ric+0.026
 interests+0.025
nels+0.024
Neg logit contributions
 faded-0.015
 White-0.014
 meant-0.014
chy-0.014
 whispered-0.014
TokenThis
Token ID1212
Feature activation-7.39e+01

Pos logit contributions
chy+0.082
 faded+0.077
 White+0.074
 whispered+0.071
 meant+0.070
Neg logit contributions
iu-0.126
pieces-0.112
 interests-0.099
ric-0.099
 disl-0.094
Token watch
Token ID2342
Feature activation-7.74e+01

Pos logit contributions
chy+0.103
 faded+0.103
 White+0.098
 whispered+0.095
 blocked+0.094
Neg logit contributions
iu-0.093
pieces-0.085
ric-0.082
 interests-0.075
 Majesty-0.071
Token is
Token ID318
Feature activation-7.81e+01

Pos logit contributions
 faded+0.108
Dead+0.108
 White+0.106
 whispered+0.105
Ancient+0.105
Neg logit contributions
pieces-0.096
iu-0.090
ric-0.084
 interests-0.077
ap-0.072
Token mine
Token ID6164
Feature activation-6.97e+01

Pos logit contributions
Ancient+0.105
 White+0.100
 faded+0.099
 Lover+0.099
Dead+0.099
Neg logit contributions
iu-0.094
pieces-0.090
ric-0.081
 interests-0.078
 disl-0.075
Token!
Token ID0
Feature activation-9.23e+01

Pos logit contributions
 faded+0.099
 White+0.095
 whispered+0.094
chy+0.094
Dead+0.093
Neg logit contributions
iu-0.090
pieces-0.079
ric-0.073
 interests-0.069
nels-0.068
Token You
Token ID921
Feature activation-8.44e+01

Pos logit contributions
 faded+0.127
chy+0.125
 White+0.122
 meant+0.121
 whispered+0.120
Neg logit contributions
iu-0.129
pieces-0.114
ric-0.101
 interests-0.099
nels-0.094
Token can
Token ID460
Feature activation-9.03e+01

Pos logit contributions
 faded+0.112
 White+0.110
chy+0.108
Dead+0.106
Ancient+0.103
Neg logit contributions
iu-0.111
pieces-0.102
ric-0.093
 interests-0.091
nels-0.084
Token't
Token ID470
Feature activation-8.77e+01

Pos logit contributions
 faded+0.094
chy+0.093
 White+0.090
 meant+0.089
 whispered+0.088
Neg logit contributions
iu-0.157
pieces-0.142
ric-0.129
 interests-0.128
nels-0.123
Token have
Token ID423
Feature activation-9.60e+01

Pos logit contributions
 faded+0.124
chy+0.123
 White+0.121
 meant+0.120
 whispered+0.119
Neg logit contributions
iu-0.119
pieces-0.104
ric-0.093
 interests-0.090
nels-0.086
Token it
Token ID340
Feature activation-8.97e+01

Pos logit contributions
chy+0.129
 faded+0.128
 White+0.126
 meant+0.123
 whispered+0.122
Neg logit contributions
iu-0.136
pieces-0.121
ric-0.106
 interests-0.106
nels-0.099
Token!"
Token ID2474
Feature activation-1.06e+02

Pos logit contributions
 White+0.158
Dead+0.157
Ancient+0.153
Spider+0.149
 faded+0.149
Neg logit contributions
pieces-0.074
iu-0.073
 yours-0.067
 interests-0.062
 mine-0.056
Token to
Token ID284
Feature activation-1.86e+02

Pos logit contributions
chy+0.289
 faded+0.287
 White+0.267
 shone+0.265
Ancient+0.265
Neg logit contributions
iu-0.197
pieces-0.180
 interests-0.178
ric-0.167
 wre-0.144
Token take
Token ID1011
Feature activation-1.90e+02

Pos logit contributions
 whispered+0.194
 meant+0.182
 hurt+0.178
 scared+0.177
 shy+0.175
Neg logit contributions
iu-0.340
pieces-0.319
ric-0.312
 Majesty-0.273
icol-0.270
Token the
Token ID262
Feature activation-1.88e+02

Pos logit contributions
 faded+0.186
 White+0.185
 meant+0.182
chy+0.178
 whispered+0.177
Neg logit contributions
iu-0.336
pieces-0.304
ric-0.290
 interests-0.277
nels-0.268
Token steak
Token ID26320
Feature activation-1.76e+02

Pos logit contributions
 faded+0.221
chy+0.219
 meant+0.213
 White+0.212
 whispered+0.208
Neg logit contributions
iu-0.301
pieces-0.273
ric-0.250
 interests-0.239
nels-0.233
Token away
Token ID1497
Feature activation-1.88e+02

Pos logit contributions
Dead+0.287
 White+0.276
 Blood+0.269
Ancient+0.269
 faded+0.267
Neg logit contributions
iu-0.186
pieces-0.170
 interests-0.141
 impression-0.134
nels-0.131
Token.
Token ID13
Feature activation-2.06e+02

Pos logit contributions
Dead+0.262
 Blood+0.249
 faded+0.244
chy+0.243
Spider+0.237
Neg logit contributions
iu-0.241
pieces-0.221
 wre-0.181
ric-0.180
 interests-0.180
Token But
Token ID887
Feature activation-2.05e+02

Pos logit contributions
chy+0.274
 faded+0.271
 White+0.244
 shone+0.239
 whispered+0.239
Neg logit contributions
iu-0.309
pieces-0.267
 interests-0.243
ric-0.236
nels-0.230
Token John
Token ID1757
Feature activation-1.90e+02

Pos logit contributions
chy+0.284
 faded+0.281
 White+0.273
 whispered+0.265
 meant+0.264
Neg logit contributions
iu-0.284
pieces-0.250
ric-0.217
 interests-0.216
nels-0.205
Token was
Token ID373
Feature activation-2.05e+02

Pos logit contributions
Ancient+0.254
Dead+0.247
 Blood+0.244
jar+0.244
Factory+0.238
Neg logit contributions
iu-0.224
n-0.195
pieces-0.194
ric-0.177
 interests-0.175
Token too
Token ID1165
Feature activation-2.02e+02

Pos logit contributions
 White+0.224
chy+0.223
 faded+0.222
 meant+0.202
 whispered+0.201
Neg logit contributions
iu-0.332
pieces-0.301
 interests-0.267
ric-0.266
 Amen-0.250
Token proud
Token ID6613
Feature activation-2.07e+02

Pos logit contributions
 shone+0.207
 Lover+0.197
chy+0.190
 faded+0.186
 contained+0.183
Neg logit contributions
iu-0.319
 disl-0.266
 interests-0.264
 wre-0.258
 gol-0.243
Token boy
Token ID2933
Feature activation-5.86e+01

Pos logit contributions
 faded+0.070
chy+0.068
 White+0.067
 meant+0.066
 whispered+0.066
Neg logit contributions
iu-0.108
pieces-0.098
ric-0.089
 interests-0.087
nels-0.084
Token.
Token ID13
Feature activation-8.52e+01

Pos logit contributions
 faded+0.073
 White+0.072
chy+0.069
 shone+0.069
Dead+0.069
Neg logit contributions
iu-0.087
pieces-0.083
 interests-0.072
ric-0.070
nels-0.067
Token
Token ID198
Feature activation-8.41e+01

Pos logit contributions
chy+0.140
 faded+0.132
 shone+0.130
 blocked+0.127
 whispered+0.124
Neg logit contributions
iu-0.087
 interests-0.076
pieces-0.073
nels-0.069
 disl-0.068
Token
Token ID198
Feature activation-8.55e+01

Pos logit contributions
 faded+0.069
chy+0.068
 meant+0.063
 White+0.062
 whispered+0.061
Neg logit contributions
iu-0.163
pieces-0.149
ric-0.140
 interests-0.138
nels-0.133
TokenThey
Token ID2990
Feature activation-7.91e+01

Pos logit contributions
chy+0.106
 faded+0.106
 meant+0.103
 White+0.100
 whispered+0.097
Neg logit contributions
iu-0.130
pieces-0.118
 interests-0.105
nels-0.105
ric-0.102
Token look
Token ID804
Feature activation-8.08e+01

Pos logit contributions
chy+0.094
 whispered+0.093
 scared+0.089
 low+0.089
 shy+0.086
Neg logit contributions
iu-0.137
pieces-0.124
 occasions-0.114
 interests-0.112
ric-0.108
Token for
Token ID329
Feature activation-8.89e+01

Pos logit contributions
 faded+0.088
chy+0.086
 White+0.084
 meant+0.083
 whispered+0.082
Neg logit contributions
iu-0.137
pieces-0.124
ric-0.112
 interests-0.110
nels-0.106
Token their
Token ID511
Feature activation-8.14e+01

Pos logit contributions
 faded+0.087
 White+0.085
chy+0.084
 meant+0.082
 whispered+0.082
Neg logit contributions
iu-0.160
pieces-0.145
ric-0.135
 interests-0.130
nels-0.126
Token moms
Token ID34015
Feature activation-7.59e+01

Pos logit contributions
 faded+0.134
chy+0.134
 White+0.134
 meant+0.132
 whispered+0.130
Neg logit contributions
iu-0.090
pieces-0.076
 interests-0.066
ric-0.066
nels-0.060
Token.
Token ID13
Feature activation-1.02e+02

Pos logit contributions
 faded+0.082
chy+0.079
 White+0.077
Dead+0.075
 meant+0.073
Neg logit contributions
iu-0.125
pieces-0.116
 interests-0.107
ric-0.103
nels-0.098
Token They
Token ID1119
Feature activation-9.51e+01

Pos logit contributions
chy+0.161
 shone+0.151
 faded+0.151
 healed+0.148
 whispered+0.147
Neg logit contributions
iu-0.106
 interests-0.098
pieces-0.088
nels-0.082
ap-0.078
Token he
Token ID339
Feature activation-2.08e+02

Pos logit contributions
 faded+0.177
chy+0.175
 White+0.168
 meant+0.165
 whispered+0.163
Neg logit contributions
iu-0.420
pieces-0.385
ric-0.355
 interests-0.350
nels-0.339
Token opened
Token ID4721
Feature activation-2.17e+02

Pos logit contributions
chy+0.178
 White+0.176
 faded+0.165
 meant+0.155
 whispered+0.154
Neg logit contributions
iu-0.390
pieces-0.353
 interests-0.322
ric-0.321
nels-0.306
Token the
Token ID262
Feature activation-2.16e+02

Pos logit contributions
 faded+0.201
 White+0.200
chy+0.195
 whispered+0.191
 meant+0.190
Neg logit contributions
iu-0.402
pieces-0.367
ric-0.336
 interests-0.331
nels-0.315
Token envelope
Token ID22878
Feature activation-2.07e+02

Pos logit contributions
 faded+0.207
chy+0.203
 White+0.198
 meant+0.196
 whispered+0.195
Neg logit contributions
iu-0.394
pieces-0.356
ric-0.328
 interests-0.321
nels-0.311
Token,
Token ID11
Feature activation-2.27e+02

Pos logit contributions
chy+0.351
 faded+0.326
Dead+0.321
 sore+0.317
Pr+0.315
Neg logit contributions
iu-0.169
pieces-0.157
 ye-0.134
nels-0.129
ap-0.126
Token he
Token ID339
Feature activation-2.19e+02

Pos logit contributions
 faded+0.163
 White+0.159
chy+0.157
 meant+0.157
 whispered+0.151
Neg logit contributions
iu-0.469
pieces-0.435
ric-0.405
 interests-0.396
nels-0.383
Token noticed
Token ID6810
Feature activation-2.30e+02

Pos logit contributions
 White+0.245
chy+0.244
 faded+0.218
 conditioning+0.216
 meant+0.216
Neg logit contributions
iu-0.332
pieces-0.291
 interests-0.270
 Majesty-0.252
ric-0.251
Token that
Token ID326
Feature activation-2.30e+02

Pos logit contributions
chy+0.332
 faded+0.327
 White+0.311
 meant+0.304
 whispered+0.303
Neg logit contributions
iu-0.302
pieces-0.268
ric-0.235
 interests-0.231
nels-0.221
Token the
Token ID262
Feature activation-2.30e+02

Pos logit contributions
 faded+0.318
chy+0.317
 White+0.299
 whispered+0.293
Ancient+0.287
Neg logit contributions
iu-0.301
pieces-0.274
 interests-0.243
ric-0.228
 Amen-0.222
Token poem
Token ID21247
Feature activation-2.23e+02

Pos logit contributions
 faded+0.240
chy+0.236
 White+0.232
 meant+0.227
 whispered+0.224
Neg logit contributions
iu-0.393
pieces-0.358
ric-0.323
 interests-0.321
nels-0.307
Token was
Token ID373
Feature activation-2.34e+02

Pos logit contributions
Pr+0.358
Ancient+0.335
 White+0.334
Dead+0.323
Spider+0.319
Neg logit contributions
iu-0.210
pieces-0.188
n-0.186
 occasions-0.181
 activities-0.166
Token feeling
Token ID4203
Feature activation-1.96e+02

Pos logit contributions
 White+0.244
Dead+0.241
Ancient+0.240
Doctor+0.223
 Green+0.220
Neg logit contributions
iu-0.275
pieces-0.241
ric-0.219
 Majesty-0.213
 interests-0.211
Token troubled
Token ID17840
Feature activation-1.97e+02

Pos logit contributions
 White+0.194
Dead+0.181
Ancient+0.179
 Lover+0.174
Spider+0.168
Neg logit contributions
iu-0.320
pieces-0.304
 interests-0.277
ric-0.271
 disl-0.266
Token."
Token ID526
Feature activation-2.12e+02

Pos logit contributions
 faded+0.287
Dead+0.277
 spoke+0.275
 White+0.275
 whispered+0.273
Neg logit contributions
iu-0.232
pieces-0.199
 interests-0.193
ric-0.184
nels-0.169
Token
Token ID220
Feature activation-1.96e+02

Pos logit contributions
chy+0.228
 faded+0.225
 meant+0.214
 White+0.214
 whispered+0.202
Neg logit contributions
iu-0.361
pieces-0.329
 interests-0.294
ric-0.292
nels-0.279
Token
Token ID198
Feature activation-2.14e+02

Pos logit contributions
 faded+0.167
chy+0.160
 White+0.149
 meant+0.148
 whispered+0.144
Neg logit contributions
iu-0.377
pieces-0.346
ric-0.322
 interests-0.318
nels-0.303
Token
Token ID198
Feature activation-2.16e+02

Pos logit contributions
chy+0.191
 faded+0.189
 meant+0.182
 White+0.179
 blocked+0.167
Neg logit contributions
iu-0.399
pieces-0.365
 interests-0.335
ric-0.331
nels-0.318
TokenEdward
Token ID43982
Feature activation-1.94e+02

Pos logit contributions
 faded+0.263
chy+0.259
 White+0.254
 meant+0.250
 whispered+0.249
Neg logit contributions
iu-0.335
pieces-0.300
ric-0.271
 interests-0.265
nels-0.254
Token was
Token ID373
Feature activation-2.14e+02

Pos logit contributions
chy+0.216
 White+0.205
Ancient+0.200
 dark+0.199
Dead+0.194
Neg logit contributions
iu-0.316
 interests-0.253
pieces-0.247
nels-0.235
n-0.229
Token so
Token ID523
Feature activation-2.18e+02

Pos logit contributions
chy+0.189
 faded+0.187
 White+0.186
 meant+0.170
Ancient+0.169
Neg logit contributions
iu-0.388
pieces-0.355
ric-0.332
 interests-0.322
 Amen-0.315
Token happy
Token ID3772
Feature activation-2.15e+02

Pos logit contributions
 White+0.250
chy+0.229
 Lover+0.222
Factory+0.221
 Green+0.219
Neg logit contributions
iu-0.322
 Amen-0.290
 interests-0.276
 disl-0.276
pieces-0.267
Token.
Token ID13
Feature activation-2.28e+02

Pos logit contributions
 faded+0.181
chy+0.172
 meant+0.172
 White+0.170
 whispered+0.170
Neg logit contributions
iu-0.417
pieces-0.389
ric-0.357
 interests-0.346
nels-0.337