TOP ACTIVATIONS
MAX = 693.147

I â m sorry Jane . I
That â s a special necklace ,
It â s a magic book !
It â s almost just as beautiful
That â s right , Emma .
That â s the gas from the
Let â s go find something else
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
Who â s there ? â
Please don â t eat me , â
I â m here to give you
It â s a surprise ! â
It â s a books he lf
her mom my watched in amaz ement as the machine flew
You â ll see . â
Where â s my jelly ? â
Why don â t you come with me

MIN ACTIVATIONS
MIN = -265.320

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
or toys , or books ," Max said .
but they were more careful . <|endoftext|> Anna and Ben were
have fun . <|endoftext|> Ben likes to play with his toy
OK ," Tom and Lily say . " But we were
go outside now ?" Anna and Ben asked .
Can we go outside now ?" Anna and Ben asked .
. They are clever . They know how to share and
and happy and good things ," Sam says .
. They are happy friends . <|endoftext|> L ily and Tom
swim with you , duck !" And they sw am together
bad . It hurt me ." Anna said .
. They smiled and laughed . They had a great adventure
He gave her the ball . " I 'm sorry ,
it is ! Follow me !" The squirrel led
forgot about lunch . They loved grapes . And they loved
. Let 's play again ." They take their boats and
, duck !" Mia says . The duck smiles

INTERVAL 453.530 - 693.147
CONTAINS 0.029%

It â s a magic book !
That â s right , Emma .
his head out of his bur row and saw something new
of other people â s property . We should
I don â t think so . It

INTERVAL 213.913 - 453.530
CONTAINS 10.492%

put some band - aid s on them . Then you
ang rier and Lily felt even sad der . From that
. It was her friend , Tommy . " Hi Lily
in the circle . The mouse stopped and the
ummy food . One day , her mom made some toast

INTERVAL -25.703 - 213.913
CONTAINS 41.616%

end . <|endoftext|> Once upon a time , there was a
brought home a newspaper . Lily saw it and said ,
Tom dreamed that a big monster took all his toys and
. Tim my loved planes and always dreamed of flying one
slowly . Tim learned that asking for help is good .

INTERVAL -265.320 - -25.703
CONTAINS 47.863%

even though it r ained . I learned that sometimes we
surprised that he stopped roaring and his anger suddenly vanished .
for animals and bugs . They had a big book with
friend . <|endoftext|> L ily liked to play with clay .
One little girl , named Linda , came up to
Token
Token ID129
Feature activation+5.76e+02

Pos logit contributions
 Jones+0.253
 White+0.233
 faded+0.230
Ancient+0.227
 blank+0.222
Neg logit contributions
n-0.366
ered-0.328
 pause-0.327
 hey-0.318
 spend-0.313
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenI
Token ID40
Feature activation+3.93e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Tokenâ
Token ID22940
Feature activation+4.40e+02

Pos logit contributions
n+0.467
 disl+0.459
pieces+0.451
 interests+0.433
 Majesty+0.432
Neg logit contributions
 needed-0.614
 whispered-0.605
 meant-0.586
 Jones-0.562
 said-0.545
Token
Token ID26391
Feature activation+4.47e+02

Pos logit contributions
 disl+0.439
pieces+0.434
n+0.427
 Majesty+0.425
 interests+0.418
Neg logit contributions
 needed-0.760
 whispered-0.753
 meant-0.739
 Jones-0.701
 said-0.691
Token
Token ID8151
Feature activation+5.09e+02

Pos logit contributions
pieces+0.628
 disl+0.613
 victory+0.606
 occasions+0.597
iu+0.594
Neg logit contributions
 needed-0.606
 whispered-0.587
 Jones-0.565
 said-0.550
 meant-0.549
Tokenm
Token ID76
Feature activation+3.93e+02

Pos logit contributions
n+0.863
ric+0.807
pieces+0.777
 disl+0.777
ieved+0.769
Neg logit contributions
 whispered-0.555
 needed-0.511
 faded-0.509
 Jones-0.505
 White-0.489
Token sorry
Token ID7926
Feature activation+3.96e+02

Pos logit contributions
n+0.585
 disl+0.570
pieces+0.550
 Majesty+0.548
ric+0.545
Neg logit contributions
 needed-0.499
 whispered-0.494
 meant-0.467
 Jones-0.452
 faded-0.442
Token Jane
Token ID12091
Feature activation+3.86e+02

Pos logit contributions
n+0.640
 disl+0.633
pieces+0.626
 interests+0.606
 Majesty+0.605
Neg logit contributions
 needed-0.451
 whispered-0.441
 meant-0.423
 Jones-0.398
 said-0.382
Token.
Token ID13
Feature activation+3.13e+02

Pos logit contributions
n+0.607
 disl+0.539
pieces+0.534
 Majesty+0.532
 spend+0.528
Neg logit contributions
Fuel-0.422
Ancient-0.414
 whispered-0.412
 blank-0.406
 aisle-0.404
Token I
Token ID314
Feature activation+3.19e+02

Pos logit contributions
n+0.438
 disl+0.430
pieces+0.410
 Majesty+0.406
 interests+0.398
Neg logit contributions
 needed-0.431
 whispered-0.422
 meant-0.407
 faded-0.384
chy-0.379
Token
Token ID129
Feature activation+5.76e+02

Pos logit contributions
 needed+0.236
 whispered+0.230
 meant+0.220
 Jones+0.206
 said+0.196
Neg logit contributions
n-0.377
 disl-0.372
pieces-0.368
 interests-0.358
 Majesty-0.357
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenThat
Token ID2504
Feature activation+3.77e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.544
 disl+0.537
pieces+0.530
 Majesty+0.512
 interests+0.512
Neg logit contributions
 needed-0.495
 whispered-0.485
 meant-0.469
 Jones-0.443
 said-0.429
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.392
 disl+0.382
pieces+0.373
 interests+0.353
 Majesty+0.351
Neg logit contributions
 needed-0.809
 whispered-0.799
 meant-0.778
 Jones-0.751
 said-0.732
Token
Token ID8151
Feature activation+5.06e+02

Pos logit contributions
pieces+0.612
 disl+0.599
n+0.588
 victory+0.583
 interests+0.583
Neg logit contributions
 needed-0.605
 whispered-0.579
 meant-0.558
 Jones-0.550
 said-0.534
Tokens
Token ID82
Feature activation+4.05e+02

Pos logit contributions
n+0.838
ric+0.776
 disl+0.761
pieces+0.746
ieved+0.727
Neg logit contributions
 whispered-0.602
 Jones-0.557
 faded-0.544
 White-0.535
 needed-0.530
Token a
Token ID257
Feature activation+3.82e+02

Pos logit contributions
pieces+0.532
n+0.528
 disl+0.525
iu+0.517
 arrest+0.516
Neg logit contributions
 needed-0.583
 whispered-0.560
 meant-0.557
 said-0.513
 Jones-0.512
Token special
Token ID2041
Feature activation+3.86e+02

Pos logit contributions
n+0.562
 disl+0.556
pieces+0.547
 Majesty+0.532
 interests+0.530
Neg logit contributions
 needed-0.481
 whispered-0.476
 meant-0.458
 Jones-0.435
 said-0.414
Token necklace
Token ID38987
Feature activation+3.75e+02

Pos logit contributions
n+0.509
 disl+0.500
pieces+0.490
 interests+0.477
 Majesty+0.474
Neg logit contributions
 needed-0.549
 whispered-0.541
 meant-0.522
 Jones-0.498
 said-0.480
Token,
Token ID11
Feature activation+3.09e+02

Pos logit contributions
n+0.524
pieces+0.459
 disl+0.458
 interests+0.450
nels+0.440
Neg logit contributions
 needed-0.484
 whispered-0.481
 Jones-0.461
 bored-0.458
 White-0.452
Token
Token ID129
Feature activation+5.76e+02

Pos logit contributions
 needed+0.194
 whispered+0.187
 meant+0.179
 Jones+0.165
chy+0.156
Neg logit contributions
n-0.411
pieces-0.407
 disl-0.405
 interests-0.396
iu-0.395
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenIt
Token ID1026
Feature activation+3.91e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.552
 disl+0.533
pieces+0.520
 interests+0.506
 Majesty+0.502
Neg logit contributions
 whispered-0.516
 needed-0.508
 Jones-0.488
 meant-0.476
 faded-0.465
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.439
 disl+0.435
pieces+0.427
 Majesty+0.408
 interests+0.407
Neg logit contributions
 needed-0.758
 whispered-0.747
 meant-0.728
 Jones-0.699
 said-0.682
Token
Token ID8151
Feature activation+5.06e+02

Pos logit contributions
pieces+0.596
 disl+0.585
n+0.573
iu+0.567
 victory+0.566
Neg logit contributions
 needed-0.619
 whispered-0.599
 meant-0.572
 Jones-0.566
 said-0.550
Tokens
Token ID82
Feature activation+4.04e+02

Pos logit contributions
n+0.881
ric+0.828
 disl+0.783
pieces+0.782
ieved+0.765
Neg logit contributions
 whispered-0.549
 Jones-0.514
 faded-0.496
 needed-0.490
 blank-0.480
Token a
Token ID257
Feature activation+3.81e+02

Pos logit contributions
n+0.531
 disl+0.519
pieces+0.505
 interests+0.491
 Majesty+0.489
Neg logit contributions
 needed-0.580
 whispered-0.577
 meant-0.551
 Jones-0.536
 faded-0.511
Token magic
Token ID5536
Feature activation+3.82e+02

Pos logit contributions
n+0.557
 disl+0.553
pieces+0.544
 victory+0.534
 Majesty+0.534
Neg logit contributions
 needed-0.471
 whispered-0.471
 meant-0.456
 Jones-0.430
 faded-0.405
Token book
Token ID1492
Feature activation+3.67e+02

Pos logit contributions
n+0.523
 disl+0.506
 Majesty+0.497
pieces+0.494
 interests+0.491
Neg logit contributions
 needed-0.509
 whispered-0.508
 meant-0.487
 Jones-0.468
 bored-0.450
Token!
Token ID0
Feature activation+3.00e+02

Pos logit contributions
n+0.469
pieces+0.386
 victory+0.381
 occasions+0.378
 disl+0.371
Neg logit contributions
 Jones-0.513
 whispered-0.512
 needed-0.498
Fuel-0.494
 bored-0.490
Token
Token ID129
Feature activation+5.76e+02

Pos logit contributions
 needed+0.370
 whispered+0.364
 meant+0.355
 Jones+0.342
 faded+0.329
Neg logit contributions
n-0.291
 disl-0.276
pieces-0.269
 interests-0.261
iu-0.257
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenIt
Token ID1026
Feature activation+3.91e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.552
 disl+0.533
pieces+0.520
 interests+0.506
 Majesty+0.502
Neg logit contributions
 whispered-0.516
 needed-0.508
 Jones-0.488
 meant-0.476
 faded-0.465
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.439
 disl+0.435
pieces+0.427
 Majesty+0.408
 interests+0.407
Neg logit contributions
 needed-0.758
 whispered-0.747
 meant-0.728
 Jones-0.699
 said-0.682
Token
Token ID8151
Feature activation+5.06e+02

Pos logit contributions
pieces+0.596
 disl+0.585
n+0.573
iu+0.567
 victory+0.566
Neg logit contributions
 needed-0.619
 whispered-0.599
 meant-0.572
 Jones-0.566
 said-0.550
Tokens
Token ID82
Feature activation+4.04e+02

Pos logit contributions
n+0.881
ric+0.828
 disl+0.783
pieces+0.782
ieved+0.765
Neg logit contributions
 whispered-0.549
 Jones-0.514
 faded-0.496
 needed-0.490
 blank-0.480
Token almost
Token ID2048
Feature activation+3.87e+02

Pos logit contributions
n+0.531
 disl+0.519
pieces+0.505
 interests+0.491
 Majesty+0.489
Neg logit contributions
 needed-0.580
 whispered-0.577
 meant-0.551
 Jones-0.536
 faded-0.511
Token just
Token ID655
Feature activation+3.68e+02

Pos logit contributions
n+0.490
 disl+0.478
pieces+0.443
 Majesty+0.440
 interests+0.439
Neg logit contributions
 needed-0.558
 whispered-0.555
 Jones-0.544
 meant-0.535
 White-0.515
Token as
Token ID355
Feature activation+3.33e+02

Pos logit contributions
n+0.475
 disl+0.462
pieces+0.446
 interests+0.437
 Majesty+0.433
Neg logit contributions
 needed-0.525
 whispered-0.521
 meant-0.503
 Jones-0.502
 White-0.473
Token beautiful
Token ID4950
Feature activation+3.48e+02

Pos logit contributions
n+0.467
 disl+0.439
 victory+0.437
 interests+0.419
pieces+0.418
Neg logit contributions
 whispered-0.430
 needed-0.415
 Jones-0.414
 meant-0.407
 spoke-0.383
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenThat
Token ID2504
Feature activation+3.82e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.557
 disl+0.550
pieces+0.544
iu+0.526
 Majesty+0.525
Neg logit contributions
 needed-0.495
 whispered-0.484
 meant-0.469
 Jones-0.441
 said-0.428
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.391
 disl+0.379
pieces+0.368
 interests+0.347
iu+0.346
Neg logit contributions
 needed-0.813
 whispered-0.804
 meant-0.780
 Jones-0.757
 said-0.735
Token
Token ID8151
Feature activation+5.05e+02

Pos logit contributions
pieces+0.585
 disl+0.579
n+0.577
 interests+0.562
iu+0.560
Neg logit contributions
 needed-0.624
 whispered-0.603
 meant-0.582
 Jones-0.568
 said-0.550
Tokens
Token ID82
Feature activation+4.03e+02

Pos logit contributions
n+0.826
ric+0.768
 disl+0.761
pieces+0.741
ieved+0.711
Neg logit contributions
 whispered-0.614
 Jones-0.564
 faded-0.543
 needed-0.537
 White-0.532
Token right
Token ID826
Feature activation+3.82e+02

Pos logit contributions
pieces+0.536
 disl+0.527
n+0.525
iu+0.523
 arrest+0.519
Neg logit contributions
 needed-0.579
 meant-0.554
 whispered-0.550
 said-0.506
 Jones-0.501
Token,
Token ID11
Feature activation+3.32e+02

Pos logit contributions
n+0.558
 disl+0.546
 Majesty+0.519
 interests+0.516
pieces+0.514
Neg logit contributions
 needed-0.490
 whispered-0.486
 meant-0.454
 Jones-0.438
 faded-0.426
Token Emma
Token ID18966
Feature activation+3.64e+02

Pos logit contributions
n+0.436
 disl+0.431
pieces+0.426
iu+0.411
 interests+0.411
Neg logit contributions
 needed-0.475
 whispered-0.466
 meant-0.452
 Jones-0.428
 said-0.419
Token.
Token ID13
Feature activation+2.98e+02

Pos logit contributions
n+0.505
 disl+0.472
 Majesty+0.464
 spend+0.436
 interests+0.434
Neg logit contributions
 whispered-0.448
 needed-0.441
 faded-0.440
 dark-0.436
Fuel-0.429
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenThat
Token ID2504
Feature activation+3.82e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.557
 disl+0.550
pieces+0.544
iu+0.526
 Majesty+0.525
Neg logit contributions
 needed-0.495
 whispered-0.484
 meant-0.469
 Jones-0.441
 said-0.428
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.391
 disl+0.379
pieces+0.368
 interests+0.347
iu+0.346
Neg logit contributions
 needed-0.813
 whispered-0.804
 meant-0.780
 Jones-0.757
 said-0.735
Token
Token ID8151
Feature activation+5.05e+02

Pos logit contributions
pieces+0.585
 disl+0.579
n+0.577
 interests+0.562
iu+0.560
Neg logit contributions
 needed-0.624
 whispered-0.603
 meant-0.582
 Jones-0.568
 said-0.550
Tokens
Token ID82
Feature activation+4.03e+02

Pos logit contributions
n+0.826
ric+0.768
 disl+0.761
pieces+0.741
ieved+0.711
Neg logit contributions
 whispered-0.614
 Jones-0.564
 faded-0.543
 needed-0.537
 White-0.532
Token the
Token ID262
Feature activation+3.89e+02

Pos logit contributions
pieces+0.536
 disl+0.527
n+0.525
iu+0.523
 arrest+0.519
Neg logit contributions
 needed-0.579
 meant-0.554
 whispered-0.550
 said-0.506
 Jones-0.501
Token gas
Token ID3623
Feature activation+3.89e+02

Pos logit contributions
n+0.569
 disl+0.559
pieces+0.550
 Majesty+0.534
 interests+0.533
Neg logit contributions
 needed-0.499
 whispered-0.493
 meant-0.472
 Jones-0.451
 faded-0.430
Token from
Token ID422
Feature activation+3.39e+02

Pos logit contributions
n+0.565
 disl+0.536
ric+0.530
 interests+0.517
nels+0.516
Neg logit contributions
 whispered-0.499
 needed-0.478
 meant-0.463
 Jones-0.451
 faded-0.440
Token the
Token ID262
Feature activation+3.40e+02

Pos logit contributions
n+0.415
 disl+0.403
pieces+0.392
 arrest+0.383
 Majesty+0.379
Neg logit contributions
 needed-0.517
 whispered-0.510
 meant-0.494
 Jones-0.478
 faded-0.457
Token
Token ID129
Feature activation+5.76e+02

Pos logit contributions
 whispered+0.431
Ancient+0.400
 meant+0.390
 faded+0.387
Fuel+0.386
Neg logit contributions
n-0.238
 victory-0.218
 job-0.212
 blast-0.202
 spend-0.197
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenLet
Token ID5756
Feature activation+3.69e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Tokenâ
Token ID22940
Feature activation+4.17e+02

Pos logit contributions
ric+0.670
iu+0.662
 Majesty+0.659
 disl+0.650
latable+0.648
Neg logit contributions
 said-0.389
 meant-0.388
 needed-0.386
 whispered-0.378
 spoke-0.314
Token
Token ID26391
Feature activation+4.28e+02

Pos logit contributions
 Majesty+0.403
 disl+0.390
pieces+0.388
 interests+0.386
 arrest+0.376
Neg logit contributions
 needed-0.727
 whispered-0.721
 meant-0.715
 said-0.670
 Jones-0.665
Token
Token ID8151
Feature activation+4.93e+02

Pos logit contributions
pieces+0.671
 occasions+0.666
 disl+0.660
 victory+0.657
iu+0.654
Neg logit contributions
 needed-0.505
 whispered-0.479
 said-0.463
 Jones-0.461
 meant-0.446
Tokens
Token ID82
Feature activation+3.92e+02

Pos logit contributions
n+0.904
pieces+0.833
ric+0.825
 disl+0.813
ieved+0.794
Neg logit contributions
 whispered-0.483
 needed-0.455
 faded-0.435
 Jones-0.430
 meant-0.426
Token go
Token ID467
Feature activation+3.58e+02

Pos logit contributions
iu+0.539
latable+0.513
pieces+0.506
 disl+0.500
 arrest+0.494
Neg logit contributions
 needed-0.596
 said-0.577
 whispered-0.573
 meant-0.573
 Jones-0.495
Token find
Token ID1064
Feature activation+3.41e+02

Pos logit contributions
 disl+0.587
iu+0.573
 Majesty+0.559
 interests+0.558
pieces+0.557
Neg logit contributions
 whispered-0.429
 needed-0.420
 said-0.400
 meant-0.393
 Jones-0.382
Token something
Token ID1223
Feature activation+3.33e+02

Pos logit contributions
 disl+0.475
n+0.468
pieces+0.463
iu+0.461
 Majesty+0.453
Neg logit contributions
 needed-0.466
 whispered-0.465
 meant-0.449
 said-0.428
 Jones-0.417
Token else
Token ID2073
Feature activation+3.53e+02

Pos logit contributions
n+0.439
 disl+0.421
 victory+0.402
pieces+0.397
 interests+0.394
Neg logit contributions
 Jones-0.472
 whispered-0.465
 needed-0.450
 White-0.443
 faded-0.425
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenLet
Token ID5756
Feature activation+3.77e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
ric+0.681
iu+0.662
latable+0.645
 Majesty+0.640
 disl+0.627
Neg logit contributions
 meant-0.427
 said-0.425
 needed-0.412
 whispered-0.410
 spoke-0.348
Token
Token ID26391
Feature activation+4.29e+02

Pos logit contributions
 Majesty+0.383
pieces+0.381
 disl+0.377
 interests+0.376
 arrest+0.367
Neg logit contributions
 needed-0.742
 whispered-0.735
 meant-0.725
 said-0.682
 Jones-0.681
Token
Token ID8151
Feature activation+4.92e+02

Pos logit contributions
pieces+0.609
 disl+0.599
iu+0.596
 occasions+0.595
 victory+0.591
Neg logit contributions
 needed-0.565
 whispered-0.537
 Jones-0.518
 said-0.511
 meant-0.508
Tokens
Token ID82
Feature activation+3.91e+02

Pos logit contributions
n+0.892
pieces+0.831
ric+0.819
 disl+0.815
 interests+0.789
Neg logit contributions
 whispered-0.491
 needed-0.461
 Jones-0.434
 meant-0.432
 faded-0.432
Token explore
Token ID7301
Feature activation+3.69e+02

Pos logit contributions
iu+0.521
latable+0.479
pieces+0.473
 disl+0.465
earned+0.457
Neg logit contributions
 needed-0.616
 meant-0.599
 said-0.594
 whispered-0.590
 Jones-0.520
Token and
Token ID290
Feature activation+3.20e+02

Pos logit contributions
n+0.528
 disl+0.525
pieces+0.515
iu+0.504
 Majesty+0.502
Neg logit contributions
 needed-0.486
 whispered-0.479
 meant-0.461
 Jones-0.436
 said-0.427
Token have
Token ID423
Feature activation+3.15e+02

Pos logit contributions
iu+0.476
 disl+0.476
 Majesty+0.456
latable+0.455
ric+0.453
Neg logit contributions
 needed-0.441
 whispered-0.436
 said-0.418
 meant-0.401
 spoke-0.371
Token some
Token ID617
Feature activation+3.08e+02

Pos logit contributions
 disl+0.466
n+0.461
pieces+0.459
iu+0.458
ric+0.448
Neg logit contributions
 needed-0.404
 whispered-0.398
 meant-0.380
 said-0.359
 dark-0.346
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenLet
Token ID5756
Feature activation+3.77e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
ric+0.681
iu+0.662
latable+0.645
 Majesty+0.640
 disl+0.627
Neg logit contributions
 meant-0.427
 said-0.425
 needed-0.412
 whispered-0.410
 spoke-0.348
Token
Token ID26391
Feature activation+4.29e+02

Pos logit contributions
 Majesty+0.383
pieces+0.381
 disl+0.377
 interests+0.376
 arrest+0.367
Neg logit contributions
 needed-0.742
 whispered-0.735
 meant-0.725
 said-0.682
 Jones-0.681
Token
Token ID8151
Feature activation+4.92e+02

Pos logit contributions
pieces+0.609
 disl+0.599
iu+0.596
 occasions+0.595
 victory+0.591
Neg logit contributions
 needed-0.565
 whispered-0.537
 Jones-0.518
 said-0.511
 meant-0.508
Tokens
Token ID82
Feature activation+3.91e+02

Pos logit contributions
n+0.892
pieces+0.831
ric+0.819
 disl+0.815
 interests+0.789
Neg logit contributions
 whispered-0.491
 needed-0.461
 Jones-0.434
 meant-0.432
 faded-0.432
Token have
Token ID423
Feature activation+3.60e+02

Pos logit contributions
iu+0.521
latable+0.479
pieces+0.473
 disl+0.465
earned+0.457
Neg logit contributions
 needed-0.616
 meant-0.599
 said-0.594
 whispered-0.590
 Jones-0.520
Token a
Token ID257
Feature activation+3.53e+02

Pos logit contributions
pieces+0.539
iu+0.538
 disl+0.535
ric+0.532
n+0.531
Neg logit contributions
 needed-0.452
 whispered-0.445
 meant-0.424
 said-0.409
 dark-0.391
Token stretching
Token ID20880
Feature activation+3.54e+02

Pos logit contributions
n+0.483
 disl+0.477
pieces+0.468
iu+0.457
 arrest+0.453
Neg logit contributions
 needed-0.491
 whispered-0.481
 meant-0.461
 Jones-0.439
 said-0.431
Token race
Token ID3234
Feature activation+3.32e+02

Pos logit contributions
 disl+0.528
pieces+0.519
n+0.518
iu+0.504
 Majesty+0.501
Neg logit contributions
 needed-0.456
 whispered-0.443
 meant-0.429
 said-0.400
 Jones-0.399
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenLet
Token ID5756
Feature activation+3.77e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
ric+0.681
iu+0.662
latable+0.645
 Majesty+0.640
 disl+0.627
Neg logit contributions
 meant-0.427
 said-0.425
 needed-0.412
 whispered-0.410
 spoke-0.348
Token
Token ID26391
Feature activation+4.29e+02

Pos logit contributions
 Majesty+0.383
pieces+0.381
 disl+0.377
 interests+0.376
 arrest+0.367
Neg logit contributions
 needed-0.742
 whispered-0.735
 meant-0.725
 said-0.682
 Jones-0.681
Token
Token ID8151
Feature activation+4.92e+02

Pos logit contributions
pieces+0.609
 disl+0.599
iu+0.596
 occasions+0.595
 victory+0.591
Neg logit contributions
 needed-0.565
 whispered-0.537
 Jones-0.518
 said-0.511
 meant-0.508
Tokens
Token ID82
Feature activation+3.91e+02

Pos logit contributions
n+0.892
pieces+0.831
ric+0.819
 disl+0.815
 interests+0.789
Neg logit contributions
 whispered-0.491
 needed-0.461
 Jones-0.434
 meant-0.432
 faded-0.432
Token play
Token ID711
Feature activation+3.71e+02

Pos logit contributions
iu+0.521
latable+0.479
pieces+0.473
 disl+0.465
earned+0.457
Neg logit contributions
 needed-0.616
 meant-0.599
 said-0.594
 whispered-0.590
 Jones-0.520
Token a
Token ID257
Feature activation+3.52e+02

Pos logit contributions
 disl+0.558
iu+0.545
n+0.539
pieces+0.538
 Majesty+0.531
Neg logit contributions
 whispered-0.457
 needed-0.454
 meant-0.432
 said-0.418
 Jones-0.403
Token joke
Token ID9707
Feature activation+3.52e+02

Pos logit contributions
n+0.493
 disl+0.493
pieces+0.478
iu+0.470
 interests+0.468
Neg logit contributions
 needed-0.472
 whispered-0.463
 meant-0.439
 Jones-0.420
 said-0.414
Token.
Token ID13
Feature activation+2.88e+02

Pos logit contributions
n+0.454
pieces+0.397
 disl+0.388
nels+0.379
 victory+0.377
Neg logit contributions
 needed-0.538
 whispered-0.510
 faded-0.496
 Jones-0.489
 White-0.479
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenLet
Token ID5756
Feature activation+3.77e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.18e+02

Pos logit contributions
ric+0.681
iu+0.662
latable+0.645
 Majesty+0.640
 disl+0.627
Neg logit contributions
 meant-0.427
 said-0.425
 needed-0.412
 whispered-0.410
 spoke-0.348
Token
Token ID26391
Feature activation+4.29e+02

Pos logit contributions
 Majesty+0.383
pieces+0.381
 disl+0.377
 interests+0.376
 arrest+0.367
Neg logit contributions
 needed-0.742
 whispered-0.735
 meant-0.725
 said-0.682
 Jones-0.681
Token
Token ID8151
Feature activation+4.92e+02

Pos logit contributions
pieces+0.609
 disl+0.599
iu+0.596
 occasions+0.595
 victory+0.591
Neg logit contributions
 needed-0.565
 whispered-0.537
 Jones-0.518
 said-0.511
 meant-0.508
Tokens
Token ID82
Feature activation+3.91e+02

Pos logit contributions
n+0.892
pieces+0.831
ric+0.819
 disl+0.815
 interests+0.789
Neg logit contributions
 whispered-0.491
 needed-0.461
 Jones-0.434
 meant-0.432
 faded-0.432
Token not
Token ID407
Feature activation+3.72e+02

Pos logit contributions
iu+0.521
latable+0.479
pieces+0.473
 disl+0.465
earned+0.457
Neg logit contributions
 needed-0.616
 meant-0.599
 said-0.594
 whispered-0.590
 Jones-0.520
Token think
Token ID892
Feature activation+3.53e+02

Pos logit contributions
iu+0.470
latable+0.461
 Majesty+0.459
 disl+0.457
pieces+0.455
Neg logit contributions
 needed-0.578
 meant-0.559
 whispered-0.551
 said-0.511
 dark-0.492
Token about
Token ID546
Feature activation+3.23e+02

Pos logit contributions
n+0.465
 disl+0.443
pieces+0.442
 interests+0.424
 victory+0.416
Neg logit contributions
 needed-0.510
 whispered-0.504
 meant-0.481
 Jones-0.471
 faded-0.450
Token the
Token ID262
Feature activation+3.24e+02

Pos logit contributions
pieces+0.393
iu+0.391
n+0.390
 disl+0.388
 Majesty+0.375
Neg logit contributions
 needed-0.482
 whispered-0.477
 meant-0.466
 Jones-0.440
 said-0.438
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenWho
Token ID8241
Feature activation+3.80e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.15e+02

Pos logit contributions
n+0.566
 disl+0.565
pieces+0.556
iu+0.545
 Majesty+0.540
Neg logit contributions
 needed-0.480
 whispered-0.467
 meant-0.455
 Jones-0.416
 said-0.416
Token
Token ID26391
Feature activation+4.27e+02

Pos logit contributions
n+0.388
 disl+0.385
pieces+0.381
 interests+0.364
 Majesty+0.363
Neg logit contributions
 needed-0.749
 whispered-0.738
 meant-0.722
 Jones-0.692
 said-0.677
Token
Token ID8151
Feature activation+4.90e+02

Pos logit contributions
n+0.603
 disl+0.602
pieces+0.599
 interests+0.579
iu+0.578
Neg logit contributions
 needed-0.565
 whispered-0.550
 meant-0.529
 Jones-0.510
 said-0.492
Tokens
Token ID82
Feature activation+3.91e+02

Pos logit contributions
n+0.870
ric+0.821
pieces+0.799
 disl+0.792
ieved+0.758
Neg logit contributions
 whispered-0.502
 faded-0.455
 needed-0.452
 Jones-0.447
 White-0.429
Token there
Token ID612
Feature activation+3.59e+02

Pos logit contributions
n+0.466
 disl+0.464
pieces+0.460
iu+0.448
 interests+0.442
Neg logit contributions
 needed-0.611
 whispered-0.595
 meant-0.585
 Jones-0.553
 said-0.545
Token?
Token ID30
Feature activation+3.15e+02

Pos logit contributions
n+0.378
 disl+0.327
ric+0.326
pieces+0.316
 pause+0.312
Neg logit contributions
 whispered-0.610
 needed-0.594
 faded-0.579
 Jones-0.560
 meant-0.556
Tokenâ
Token ID22940
Feature activation+3.54e+02

Pos logit contributions
n+0.435
pieces+0.404
 disl+0.401
 Majesty+0.379
ric+0.378
Neg logit contributions
 needed-0.442
 whispered-0.441
 meant-0.421
 faded-0.415
chy-0.408
Token
Token ID26391
Feature activation+3.64e+02

Pos logit contributions
n+0.302
 disl+0.269
ric+0.226
iu+0.225
ieved+0.223
Neg logit contributions
 needed-0.717
 whispered-0.697
 meant-0.675
 faded-0.674
 Jones-0.663
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenPlease
Token ID5492
Feature activation+4.03e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Token don
Token ID836
Feature activation+3.88e+02

Pos logit contributions
n+0.510
 disl+0.509
pieces+0.504
iu+0.484
 Majesty+0.482
Neg logit contributions
 needed-0.597
 whispered-0.585
 meant-0.567
 Jones-0.541
 said-0.529
Tokenâ
Token ID22940
Feature activation+4.22e+02

Pos logit contributions
 Majesty+0.628
 disl+0.605
 interests+0.595
pieces+0.578
iu+0.564
Neg logit contributions
 needed-0.503
 meant-0.501
 whispered-0.470
 said-0.458
 Jones-0.427
Token
Token ID26391
Feature activation+4.28e+02

Pos logit contributions
 disl+0.420
 Majesty+0.418
pieces+0.418
 interests+0.401
 arrest+0.397
Neg logit contributions
 needed-0.733
 whispered-0.721
 meant-0.719
 said-0.670
 Jones-0.665
Token
Token ID8151
Feature activation+4.89e+02

Pos logit contributions
pieces+0.628
 disl+0.622
n+0.608
iu+0.603
 interests+0.600
Neg logit contributions
 needed-0.549
 whispered-0.529
 meant-0.508
 Jones-0.500
 said-0.488
Tokent
Token ID83
Feature activation+3.87e+02

Pos logit contributions
n+0.812
pieces+0.741
ric+0.728
 disl+0.722
ieved+0.709
Neg logit contributions
 whispered-0.568
 needed-0.525
 faded-0.502
 Jones-0.496
 meant-0.496
Token eat
Token ID4483
Feature activation+3.70e+02

Pos logit contributions
n+0.446
 disl+0.441
pieces+0.437
 Majesty+0.421
iu+0.419
Neg logit contributions
 needed-0.621
 whispered-0.608
 meant-0.595
 Jones-0.562
 said-0.553
Token me
Token ID502
Feature activation+3.69e+02

Pos logit contributions
n+0.412
 disl+0.406
pieces+0.399
 interests+0.381
 Majesty+0.381
Neg logit contributions
 needed-0.604
 whispered-0.595
 meant-0.578
 Jones-0.554
 said-0.539
Token,
Token ID11
Feature activation+3.14e+02

Pos logit contributions
n+0.488
 disl+0.459
pieces+0.451
 interests+0.448
 Majesty+0.438
Neg logit contributions
 needed-0.518
 whispered-0.515
 meant-0.498
 Jones-0.486
 faded-0.479
Tokenâ
Token ID22940
Feature activation+3.62e+02

Pos logit contributions
n+0.436
 disl+0.432
pieces+0.425
 Majesty+0.411
 interests+0.410
Neg logit contributions
 needed-0.425
 whispered-0.417
 meant-0.403
 Jones-0.382
 said-0.371
Token
Token ID129
Feature activation+4.59e+02

Pos logit contributions
pieces+0.838
 disl+0.838
n+0.822
 interests+0.811
iu+0.806
Neg logit contributions
 needed-0.412
 whispered-0.404
 meant-0.367
 Jones-0.361
 said-0.353
Token
Token ID241
Feature activation+4.22e+02

Pos logit contributions
n+0.495
 disl+0.481
pieces+0.454
 Majesty+0.442
iu+0.438
Neg logit contributions
 needed-0.790
 whispered-0.771
 meant-0.753
 Jones-0.723
 faded-0.704
TokenI
Token ID40
Feature activation+3.74e+02

Pos logit contributions
pieces+0.731
 disl+0.728
n+0.721
 Majesty+0.721
 arrest+0.710
Neg logit contributions
 needed-0.423
 whispered-0.400
 meant-0.394
 Jones-0.357
 said-0.350
Tokenâ
Token ID22940
Feature activation+4.17e+02

Pos logit contributions
 disl+0.442
n+0.441
pieces+0.440
iu+0.424
 Majesty+0.420
Neg logit contributions
 needed-0.588
 whispered-0.578
 meant-0.562
 Jones-0.528
 said-0.525
Token
Token ID26391
Feature activation+4.24e+02

Pos logit contributions
n+0.369
 disl+0.355
pieces+0.345
iu+0.325
 interests+0.324
Neg logit contributions
 needed-0.788
 whispered-0.776
 meant-0.756
 Jones-0.731
 faded-0.713
Token
Token ID8151
Feature activation+4.85e+02

Pos logit contributions
pieces+0.585
 disl+0.580
 victory+0.561
 interests+0.560
n+0.559
Neg logit contributions
 needed-0.578
 whispered-0.565
 Jones-0.541
 meant-0.529
 said-0.524
Tokenm
Token ID76
Feature activation+3.77e+02

Pos logit contributions
n+0.817
ric+0.760
pieces+0.742
 disl+0.740
ieved+0.722
Neg logit contributions
 whispered-0.535
 needed-0.494
 faded-0.483
 Jones-0.481
 White-0.465
Token here
Token ID994
Feature activation+3.73e+02

Pos logit contributions
n+0.547
 disl+0.531
pieces+0.515
 Majesty+0.511
ric+0.505
Neg logit contributions
 needed-0.496
 whispered-0.491
 meant-0.466
 Jones-0.447
 faded-0.438
Token to
Token ID284
Feature activation+3.38e+02

Pos logit contributions
n+0.478
 disl+0.457
pieces+0.442
 interests+0.435
 victory+0.435
Neg logit contributions
 needed-0.545
 whispered-0.541
 meant-0.514
 Jones-0.495
 faded-0.489
Token give
Token ID1577
Feature activation+3.18e+02

Pos logit contributions
iu+0.441
nels+0.410
pieces+0.409
 disl+0.401
 interests+0.397
Neg logit contributions
 whispered-0.517
 said-0.507
 needed-0.500
 meant-0.486
 dark-0.459
Token you
Token ID345
Feature activation+3.20e+02

Pos logit contributions
 disl+0.389
pieces+0.381
n+0.379
 Majesty+0.369
iu+0.368
Neg logit contributions
 needed-0.485
 whispered-0.484
 meant-0.465
 said-0.447
 Jones-0.444
Token
Token ID129
Feature activation+4.59e+02

Pos logit contributions
pieces+0.838
 disl+0.838
n+0.822
 interests+0.811
iu+0.806
Neg logit contributions
 needed-0.412
 whispered-0.404
 meant-0.367
 Jones-0.361
 said-0.353
Token
Token ID241
Feature activation+4.22e+02

Pos logit contributions
n+0.495
 disl+0.481
pieces+0.454
 Majesty+0.442
iu+0.438
Neg logit contributions
 needed-0.790
 whispered-0.771
 meant-0.753
 Jones-0.723
 faded-0.704
TokenIt
Token ID1026
Feature activation+3.71e+02

Pos logit contributions
pieces+0.731
 disl+0.728
n+0.721
 Majesty+0.721
 arrest+0.710
Neg logit contributions
 needed-0.423
 whispered-0.400
 meant-0.394
 Jones-0.357
 said-0.350
Tokenâ
Token ID22940
Feature activation+4.13e+02

Pos logit contributions
n+0.539
 disl+0.531
pieces+0.524
 interests+0.507
 Majesty+0.506
Neg logit contributions
 needed-0.483
 whispered-0.474
 meant-0.456
 Jones-0.434
 said-0.417
Token
Token ID26391
Feature activation+4.20e+02

Pos logit contributions
n+0.389
 disl+0.373
pieces+0.363
iu+0.343
 interests+0.343
Neg logit contributions
 needed-0.758
 whispered-0.748
 meant-0.726
 Jones-0.702
 faded-0.683
Token
Token ID8151
Feature activation+4.81e+02

Pos logit contributions
pieces+0.588
 disl+0.580
n+0.565
 interests+0.562
iu+0.562
Neg logit contributions
 needed-0.561
 whispered-0.546
 Jones-0.517
 meant-0.516
 said-0.500
Tokens
Token ID82
Feature activation+3.85e+02

Pos logit contributions
n+0.822
ric+0.769
 disl+0.741
pieces+0.738
ieved+0.710
Neg logit contributions
 whispered-0.538
 Jones-0.496
 needed-0.488
 faded-0.477
 White-0.463
Token a
Token ID257
Feature activation+3.66e+02

Pos logit contributions
n+0.508
 disl+0.504
pieces+0.499
iu+0.480
 Majesty+0.478
Neg logit contributions
 needed-0.551
 whispered-0.538
 meant-0.524
 Jones-0.494
 said-0.482
Token surprise
Token ID5975
Feature activation+3.60e+02

Pos logit contributions
n+0.524
 disl+0.519
pieces+0.511
 Majesty+0.500
 victory+0.499
Neg logit contributions
 needed-0.467
 whispered-0.463
 meant-0.449
 Jones-0.426
 faded-0.401
Token!
Token ID0
Feature activation+2.99e+02

Pos logit contributions
n+0.437
pieces+0.394
 disl+0.384
 occasions+0.379
 blast+0.372
Neg logit contributions
 needed-0.552
 whispered-0.543
 Jones-0.515
 meant-0.511
 bored-0.504
Tokenâ
Token ID22940
Feature activation+3.46e+02

Pos logit contributions
n+0.410
 disl+0.407
pieces+0.400
 interests+0.386
 Majesty+0.386
Neg logit contributions
 needed-0.410
 whispered-0.403
 meant-0.389
 Jones-0.372
 said-0.359
Token
Token ID129
Feature activation+4.59e+02

Pos logit contributions
pieces+0.838
 disl+0.838
n+0.822
 interests+0.811
iu+0.806
Neg logit contributions
 needed-0.412
 whispered-0.404
 meant-0.367
 Jones-0.361
 said-0.353
Token
Token ID241
Feature activation+4.22e+02

Pos logit contributions
n+0.495
 disl+0.481
pieces+0.454
 Majesty+0.442
iu+0.438
Neg logit contributions
 needed-0.790
 whispered-0.771
 meant-0.753
 Jones-0.723
 faded-0.704
TokenIt
Token ID1026
Feature activation+3.71e+02

Pos logit contributions
pieces+0.731
 disl+0.728
n+0.721
 Majesty+0.721
 arrest+0.710
Neg logit contributions
 needed-0.423
 whispered-0.400
 meant-0.394
 Jones-0.357
 said-0.350
Tokenâ
Token ID22940
Feature activation+4.13e+02

Pos logit contributions
n+0.539
 disl+0.531
pieces+0.524
 interests+0.507
 Majesty+0.506
Neg logit contributions
 needed-0.483
 whispered-0.474
 meant-0.456
 Jones-0.434
 said-0.417
Token
Token ID26391
Feature activation+4.20e+02

Pos logit contributions
n+0.389
 disl+0.373
pieces+0.363
iu+0.343
 interests+0.343
Neg logit contributions
 needed-0.758
 whispered-0.748
 meant-0.726
 Jones-0.702
 faded-0.683
Token
Token ID8151
Feature activation+4.81e+02

Pos logit contributions
pieces+0.588
 disl+0.580
n+0.565
 interests+0.562
iu+0.562
Neg logit contributions
 needed-0.561
 whispered-0.546
 Jones-0.517
 meant-0.516
 said-0.500
Tokens
Token ID82
Feature activation+3.85e+02

Pos logit contributions
n+0.822
ric+0.769
 disl+0.741
pieces+0.738
ieved+0.710
Neg logit contributions
 whispered-0.538
 Jones-0.496
 needed-0.488
 faded-0.477
 White-0.463
Token a
Token ID257
Feature activation+3.66e+02

Pos logit contributions
n+0.508
 disl+0.504
pieces+0.499
iu+0.480
 Majesty+0.478
Neg logit contributions
 needed-0.551
 whispered-0.538
 meant-0.524
 Jones-0.494
 said-0.482
Token books
Token ID3835
Feature activation+3.70e+02

Pos logit contributions
n+0.524
 disl+0.519
pieces+0.511
 Majesty+0.500
 victory+0.499
Neg logit contributions
 needed-0.467
 whispered-0.463
 meant-0.449
 Jones-0.426
 faded-0.401
Tokenhe
Token ID258
Feature activation+3.85e+02

Pos logit contributions
n+0.464
 blast+0.415
 victory+0.408
 spend+0.401
pieces+0.391
Neg logit contributions
 Jones-0.533
 whispered-0.528
 needed-0.520
 White-0.508
 spoke-0.482
Tokenlf
Token ID1652
Feature activation+3.98e+02

Pos logit contributions
n+0.534
 disl+0.529
pieces+0.522
 interests+0.504
 Majesty+0.502
Neg logit contributions
 needed-0.522
 whispered-0.512
 meant-0.494
 Jones-0.469
 said-0.456
Token her
Token ID607
Feature activation+4.38e+02

Pos logit contributions
 needed+0.231
 whispered+0.229
 Jones+0.215
 meant+0.215
 White+0.194
Neg logit contributions
n-0.388
 disl-0.385
pieces-0.383
 interests-0.374
 Majesty-0.364
Token mom
Token ID1995
Feature activation+3.92e+02

Pos logit contributions
n+0.607
 disl+0.598
pieces+0.589
 interests+0.570
 Majesty+0.569
Neg logit contributions
 needed-0.595
 whispered-0.585
 meant-0.565
 Jones-0.537
 said-0.518
Tokenmy
Token ID1820
Feature activation+4.11e+02

Pos logit contributions
n+0.640
pieces+0.609
 spend+0.597
 hey+0.575
ric+0.575
Neg logit contributions
 whispered-0.394
 needed-0.386
 meant-0.381
 Jones-0.381
 faded-0.373
Token watched
Token ID7342
Feature activation+3.29e+02

Pos logit contributions
n+0.643
 spend+0.615
pieces+0.579
 Majesty+0.566
 disl+0.558
Neg logit contributions
Ancient-0.430
 dark-0.426
 whispered-0.415
Fuel-0.413
 faded-0.408
Token in
Token ID287
Feature activation+3.14e+02

Pos logit contributions
n+0.447
pieces+0.434
 disl+0.429
iu+0.410
 interests+0.406
Neg logit contributions
 needed-0.463
 whispered-0.455
 meant-0.439
 Jones-0.415
 faded-0.403
Token amaz
Token ID40642
Feature activation+4.79e+02

Pos logit contributions
n+0.441
 disl+0.436
pieces+0.430
 interests+0.418
 arrest+0.415
Neg logit contributions
 needed-0.420
 whispered-0.412
 meant-0.397
 Jones-0.376
 spoke-0.361
Tokenement
Token ID972
Feature activation+3.53e+02

Pos logit contributions
 blast+0.979
 arrest+0.974
ric+0.962
 disl+0.954
n+0.941
Neg logit contributions
 said-0.362
 meant-0.345
 whispered-0.331
 needed-0.317
 answered-0.308
Token as
Token ID355
Feature activation+3.05e+02

Pos logit contributions
n+0.575
pieces+0.563
 disl+0.562
 interests+0.541
 Majesty+0.538
Neg logit contributions
 needed-0.393
 whispered-0.384
 meant-0.366
 Jones-0.345
 faded-0.329
Token the
Token ID262
Feature activation+3.10e+02

Pos logit contributions
n+0.477
pieces+0.470
 disl+0.468
 victory+0.455
 interests+0.447
Neg logit contributions
 needed-0.350
 whispered-0.346
 meant-0.327
 Jones-0.318
 spoke-0.294
Token machine
Token ID4572
Feature activation+3.16e+02

Pos logit contributions
n+0.433
 disl+0.425
pieces+0.421
 interests+0.405
 Majesty+0.405
Neg logit contributions
 needed-0.420
 whispered-0.413
 meant-0.398
 Jones-0.380
 said-0.363
Token flew
Token ID13112
Feature activation+2.88e+02

Pos logit contributions
n+0.507
pieces+0.440
 spend+0.421
 join+0.417
ered+0.416
Neg logit contributions
 Jones-0.363
Fuel-0.352
Ancient-0.340
Pr-0.336
mail-0.336
Token
Token ID129
Feature activation+4.59e+02

Pos logit contributions
pieces+0.838
 disl+0.838
n+0.822
 interests+0.811
iu+0.806
Neg logit contributions
 needed-0.412
 whispered-0.404
 meant-0.367
 Jones-0.361
 said-0.353
Token
Token ID241
Feature activation+4.22e+02

Pos logit contributions
n+0.495
 disl+0.481
pieces+0.454
 Majesty+0.442
iu+0.438
Neg logit contributions
 needed-0.790
 whispered-0.771
 meant-0.753
 Jones-0.723
 faded-0.704
TokenYou
Token ID1639
Feature activation+3.64e+02

Pos logit contributions
pieces+0.731
 disl+0.728
n+0.721
 Majesty+0.721
 arrest+0.710
Neg logit contributions
 needed-0.423
 whispered-0.400
 meant-0.394
 Jones-0.357
 said-0.350
Tokenâ
Token ID22940
Feature activation+4.07e+02

Pos logit contributions
 disl+0.470
pieces+0.461
n+0.458
iu+0.454
 Majesty+0.443
Neg logit contributions
 needed-0.547
 whispered-0.532
 meant-0.523
 said-0.487
 Jones-0.486
Token
Token ID26391
Feature activation+4.16e+02

Pos logit contributions
n+0.340
 disl+0.298
pieces+0.265
ric+0.254
iu+0.252
Neg logit contributions
 needed-0.829
 whispered-0.816
 meant-0.777
 faded-0.774
 Jones-0.773
Token
Token ID8151
Feature activation+4.78e+02

Pos logit contributions
pieces+0.548
 disl+0.547
n+0.540
 interests+0.525
iu+0.524
Neg logit contributions
 needed-0.586
 whispered-0.571
 meant-0.548
 Jones-0.540
 said-0.522
Tokenll
Token ID297
Feature activation+4.13e+02

Pos logit contributions
n+0.804
ric+0.763
 disl+0.733
pieces+0.733
 interests+0.704
Neg logit contributions
 whispered-0.541
 faded-0.489
 needed-0.487
 Jones-0.476
 dark-0.461
Token see
Token ID766
Feature activation+3.54e+02

Pos logit contributions
iu+0.447
 disl+0.435
pieces+0.434
 arrest+0.419
 Majesty+0.417
Neg logit contributions
 needed-0.698
 meant-0.683
 whispered-0.676
 said-0.645
chy-0.611
Token.
Token ID13
Feature activation+3.02e+02

Pos logit contributions
n+0.393
 disl+0.387
pieces+0.380
iu+0.365
 interests+0.363
Neg logit contributions
 needed-0.579
 whispered-0.571
 meant-0.555
 Jones-0.531
 said-0.520
Tokenâ
Token ID22940
Feature activation+3.51e+02

Pos logit contributions
n+0.411
 disl+0.405
pieces+0.399
 interests+0.385
 Majesty+0.384
Neg logit contributions
 needed-0.418
 whispered-0.411
 meant-0.397
 Jones-0.377
 said-0.365
Token
Token ID26391
Feature activation+3.60e+02

Pos logit contributions
n+0.294
 disl+0.271
pieces+0.249
iu+0.237
nels+0.233
Neg logit contributions
 needed-0.699
 whispered-0.684
 meant-0.665
 Jones-0.652
 faded-0.643
Token
Token ID129
Feature activation+4.59e+02

Pos logit contributions
pieces+0.838
 disl+0.838
n+0.822
 interests+0.811
iu+0.806
Neg logit contributions
 needed-0.412
 whispered-0.404
 meant-0.367
 Jones-0.361
 said-0.353
Token
Token ID241
Feature activation+4.22e+02

Pos logit contributions
n+0.495
 disl+0.481
pieces+0.454
 Majesty+0.442
iu+0.438
Neg logit contributions
 needed-0.790
 whispered-0.771
 meant-0.753
 Jones-0.723
 faded-0.704
TokenWhere
Token ID8496
Feature activation+3.72e+02

Pos logit contributions
pieces+0.731
 disl+0.728
n+0.721
 Majesty+0.721
 arrest+0.710
Neg logit contributions
 needed-0.423
 whispered-0.400
 meant-0.394
 Jones-0.357
 said-0.350
Tokenâ
Token ID22940
Feature activation+4.07e+02

Pos logit contributions
iu+0.560
 arrest+0.552
 disl+0.546
pieces+0.544
 Majesty+0.538
Neg logit contributions
 needed-0.503
 meant-0.493
 whispered-0.478
 said-0.457
 smelled-0.410
Token
Token ID26391
Feature activation+4.16e+02

Pos logit contributions
pieces+0.363
 disl+0.361
n+0.353
 interests+0.349
 Majesty+0.348
Neg logit contributions
 needed-0.745
 whispered-0.737
 meant-0.722
 Jones-0.688
 said-0.681
Token
Token ID8151
Feature activation+4.77e+02

Pos logit contributions
 disl+0.689
pieces+0.686
n+0.673
 interests+0.667
iu+0.666
Neg logit contributions
 needed-0.454
 whispered-0.440
 meant-0.415
 Jones-0.407
 said-0.391
Tokens
Token ID82
Feature activation+3.81e+02

Pos logit contributions
n+0.814
ric+0.745
pieces+0.734
 disl+0.723
ieved+0.705
Neg logit contributions
 whispered-0.523
 Jones-0.482
 needed-0.479
 faded-0.477
 White-0.465
Token my
Token ID616
Feature activation+3.78e+02

Pos logit contributions
pieces+0.500
iu+0.496
 disl+0.495
n+0.489
 interests+0.482
Neg logit contributions
 needed-0.562
 meant-0.541
 whispered-0.540
 said-0.505
 Jones-0.482
Token jelly
Token ID27644
Feature activation+3.83e+02

Pos logit contributions
n+0.465
 disl+0.456
pieces+0.448
 Majesty+0.435
 interests+0.433
Neg logit contributions
 needed-0.574
 whispered-0.565
 meant-0.548
 Jones-0.523
 said-0.506
Token?
Token ID30
Feature activation+3.08e+02

Pos logit contributions
n+0.422
 disl+0.403
pieces+0.388
 Majesty+0.380
 interests+0.376
Neg logit contributions
 needed-0.622
 whispered-0.620
 meant-0.593
 Jones-0.587
 White-0.565
Tokenâ
Token ID22940
Feature activation+3.45e+02

Pos logit contributions
n+0.418
pieces+0.380
 disl+0.380
ric+0.356
 Majesty+0.355
Neg logit contributions
 needed-0.451
 whispered-0.444
 meant-0.421
 faded-0.418
chy-0.417
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenWhy
Token ID5195
Feature activation+3.81e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Token don
Token ID836
Feature activation+3.73e+02

Pos logit contributions
pieces+0.461
n+0.459
 disl+0.453
iu+0.449
 arrest+0.442
Neg logit contributions
 needed-0.586
 whispered-0.572
 meant-0.569
 Jones-0.530
 said-0.529
Tokenâ
Token ID22940
Feature activation+4.06e+02

Pos logit contributions
 Majesty+0.666
 interests+0.630
 disl+0.605
 majesty+0.593
iu+0.592
Neg logit contributions
 meant-0.451
 needed-0.446
 said-0.403
 whispered-0.402
 White-0.371
Token
Token ID26391
Feature activation+4.14e+02

Pos logit contributions
 disl+0.389
pieces+0.389
 Majesty+0.382
n+0.378
 interests+0.375
Neg logit contributions
 needed-0.719
 whispered-0.709
 meant-0.701
 Jones-0.657
 said-0.656
Token
Token ID8151
Feature activation+4.75e+02

Pos logit contributions
 disl+0.576
pieces+0.576
n+0.574
iu+0.555
 interests+0.555
Neg logit contributions
 needed-0.555
 whispered-0.540
 meant-0.522
 Jones-0.502
 said-0.489
Tokent
Token ID83
Feature activation+3.77e+02

Pos logit contributions
n+0.767
pieces+0.683
ric+0.676
ieved+0.662
 disl+0.661
Neg logit contributions
 whispered-0.579
 Jones-0.529
 faded-0.524
 needed-0.523
 blank-0.520
Token you
Token ID345
Feature activation+3.68e+02

Pos logit contributions
n+0.492
 disl+0.485
pieces+0.480
 Majesty+0.463
iu+0.462
Neg logit contributions
 needed-0.547
 whispered-0.536
 meant-0.521
 Jones-0.493
 said-0.481
Token come
Token ID1282
Feature activation+3.37e+02

Pos logit contributions
n+0.433
 pause+0.383
 spend+0.383
 disl+0.371
 interests+0.370
Neg logit contributions
 whispered-0.554
 Jones-0.553
 needed-0.545
 aisle-0.545
 White-0.529
Token with
Token ID351
Feature activation+3.17e+02

Pos logit contributions
n+0.477
 disl+0.475
pieces+0.466
iu+0.453
 Majesty+0.449
Neg logit contributions
 needed-0.450
 whispered-0.445
 meant-0.428
 Jones-0.410
 said-0.394
Token me
Token ID502
Feature activation+3.27e+02

Pos logit contributions
n+0.342
 disl+0.334
pieces+0.325
 interests+0.314
 Majesty+0.310
Neg logit contributions
 needed-0.532
 whispered-0.523
 meant-0.508
 Jones-0.492
 faded-0.474
Token ball
Token ID2613
Feature activation-2.32e+02

Pos logit contributions
 needed+0.280
 whispered+0.277
 meant+0.255
 dark+0.248
 said+0.248
Neg logit contributions
n-0.381
 disl-0.381
iu-0.365
 Majesty-0.364
pieces-0.364
Token.
Token ID13
Feature activation-2.48e+02

Pos logit contributions
 whispered+0.286
 needed+0.281
 Jones+0.273
 meant+0.272
 spoke+0.256
Neg logit contributions
n-0.352
 victory-0.344
pieces-0.334
ap-0.328
 blast-0.326
Token They
Token ID1119
Feature activation-2.45e+02

Pos logit contributions
 needed+0.263
 whispered+0.259
 meant+0.247
 Jones+0.226
 spoke+0.220
Neg logit contributions
n-0.420
 disl-0.410
pieces-0.409
 interests-0.404
 Majesty-0.394
Token are
Token ID389
Feature activation-2.47e+02

Pos logit contributions
 needed+0.444
 whispered+0.435
 meant+0.420
 said+0.403
 Jones+0.401
Neg logit contributions
 disl-0.234
n-0.233
pieces-0.228
iu-0.221
 Majesty-0.220
Token clever
Token ID14169
Feature activation-2.44e+02

Pos logit contributions
 needed+0.264
 whispered+0.259
 meant+0.249
 Jones+0.227
 said+0.223
Neg logit contributions
n-0.417
 disl-0.415
pieces-0.408
iu-0.397
 interests-0.396
Token.
Token ID13
Feature activation-2.55e+02

Pos logit contributions
 needed+0.250
 whispered+0.244
 meant+0.231
 Jones+0.219
 faded+0.206
Neg logit contributions
n-0.418
 disl-0.412
pieces-0.409
 interests-0.401
 Majesty-0.398
Token They
Token ID1119
Feature activation-2.52e+02

Pos logit contributions
 needed+0.257
 whispered+0.252
 meant+0.240
 Jones+0.219
 faded+0.213
Neg logit contributions
n-0.447
 disl-0.437
pieces-0.435
 interests-0.431
 Majesty-0.418
Token know
Token ID760
Feature activation-2.56e+02

Pos logit contributions
 needed+0.440
 whispered+0.431
 meant+0.416
 said+0.400
 Jones+0.390
Neg logit contributions
n-0.258
 disl-0.256
pieces-0.255
iu-0.250
 Majesty-0.245
Token how
Token ID703
Feature activation-2.58e+02

Pos logit contributions
 needed+0.476
 whispered+0.470
 meant+0.459
 said+0.436
 Jones+0.435
Neg logit contributions
n-0.234
 disl-0.228
pieces-0.222
iu-0.215
 arrest-0.212
Token to
Token ID284
Feature activation-2.58e+02

Pos logit contributions
 needed+0.253
 dark+0.204
 sleepy+0.201
 bored+0.200
 whispered+0.195
Neg logit contributions
iu-0.514
 Majesty-0.499
han-0.495
n-0.493
pieces-0.488
Token share
Token ID2648
Feature activation-2.62e+02

Pos logit contributions
 needed+0.352
 whispered+0.343
chy+0.341
 said+0.328
 meant+0.328
Neg logit contributions
iu-0.383
pieces-0.369
 disl-0.363
 Majesty-0.363
 interests-0.338
Token a
Token ID257
Feature activation-2.37e+02

Pos logit contributions
 needed+0.409
 whispered+0.404
 meant+0.392
 Jones+0.377
 said+0.370
Neg logit contributions
n-0.236
 disl-0.236
pieces-0.229
iu-0.219
 Majesty-0.218
Token spaceship
Token ID40663
Feature activation-2.29e+02

Pos logit contributions
 needed+0.349
 whispered+0.345
 meant+0.341
 Jones+0.317
 spoke+0.306
Neg logit contributions
n-0.297
 victory-0.290
pieces-0.290
 Majesty-0.290
 blast-0.289
Token!"
Token ID2474
Feature activation-2.47e+02

Pos logit contributions
 needed+0.331
 whispered+0.330
 meant+0.321
 said+0.307
 Jones+0.300
Neg logit contributions
 disl-0.304
n-0.294
pieces-0.291
iu-0.289
 interests-0.281
Token Ben
Token ID3932
Feature activation-2.36e+02

Pos logit contributions
 needed+0.233
 meant+0.224
 whispered+0.215
 Jones+0.202
chy+0.201
Neg logit contributions
n-0.435
pieces-0.424
 disl-0.421
 arrest-0.416
 blast-0.412
Token said
Token ID531
Feature activation-2.47e+02

Pos logit contributions
lessness+0.227
chy+0.226
oky+0.224
 faded+0.216
 dark+0.214
Neg logit contributions
n-0.431
ered-0.357
 interests-0.344
iu-0.343
 how-0.339
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
 needed+0.261
 whispered+0.251
 meant+0.246
 Jones+0.227
 faded+0.221
Neg logit contributions
n-0.412
pieces-0.399
 disl-0.397
iu-0.390
 victory-0.385
Token "
Token ID366
Feature activation-2.50e+02

Pos logit contributions
 needed+0.345
chy+0.342
 whispered+0.334
 meant+0.334
 faded+0.310
Neg logit contributions
n-0.327
 join-0.309
 blast-0.300
 interests-0.286
 our-0.283
TokenWe
Token ID1135
Feature activation-2.54e+02

Pos logit contributions
 needed+0.346
 whispered+0.343
 meant+0.329
 Jones+0.312
 faded+0.307
Neg logit contributions
n-0.358
pieces-0.340
 disl-0.336
 interests-0.334
 arrest-0.323
Token can
Token ID460
Feature activation-2.53e+02

Pos logit contributions
 needed+0.360
 whispered+0.346
 meant+0.337
 said+0.312
 Jones+0.309
Neg logit contributions
 disl-0.347
pieces-0.344
n-0.344
iu-0.337
 Majesty-0.330
Token fly
Token ID6129
Feature activation-2.54e+02

Pos logit contributions
 whispered+0.426
 needed+0.419
 said+0.399
 meant+0.398
 Jones+0.380
Neg logit contributions
 disl-0.290
pieces-0.274
iu-0.274
earned-0.264
 interests-0.262
Token to
Token ID284
Feature activation-2.57e+02

Pos logit contributions
 needed+0.325
 whispered+0.321
 meant+0.307
 Jones+0.293
 dark+0.288
Neg logit contributions
 disl-0.375
n-0.373
iu-0.366
pieces-0.363
 Majesty-0.356
Token learned
Token ID4499
Feature activation-2.41e+02

Pos logit contributions
 Jones+0.302
 aisle+0.282
 whispered+0.277
 White+0.269
 needed+0.267
Neg logit contributions
n-0.332
 disl-0.325
 victory-0.313
 interests-0.310
 spend-0.309
Token that
Token ID326
Feature activation-2.43e+02

Pos logit contributions
 needed+0.334
 whispered+0.330
 meant+0.326
 said+0.309
 Jones+0.290
Neg logit contributions
pieces-0.341
n-0.340
 Majesty-0.333
iu-0.331
 disl-0.320
Token courage
Token ID11917
Feature activation-2.41e+02

Pos logit contributions
 whispered+0.407
 needed+0.399
 Jones+0.387
 meant+0.377
 aisle+0.367
Neg logit contributions
 victory-0.254
n-0.252
 interests-0.252
 disl-0.250
pieces-0.232
Token is
Token ID318
Feature activation-2.44e+02

Pos logit contributions
 whispered+0.330
 needed+0.321
 Jones+0.315
 meant+0.301
 White+0.296
Neg logit contributions
n-0.321
 disl-0.315
 interests-0.313
pieces-0.300
 victory-0.298
Token important
Token ID1593
Feature activation-2.43e+02

Pos logit contributions
 needed+0.370
 whispered+0.366
 meant+0.348
 Jones+0.347
 faded+0.327
Neg logit contributions
n-0.300
 disl-0.294
pieces-0.286
 interests-0.282
 victory-0.277
Token.
Token ID13
Feature activation-2.54e+02

Pos logit contributions
 needed+0.230
 whispered+0.226
 meant+0.212
 Jones+0.199
 said+0.187
Neg logit contributions
n-0.438
 disl-0.433
pieces-0.426
 interests-0.419
iu-0.414
Token They
Token ID1119
Feature activation-2.50e+02

Pos logit contributions
 whispered+0.272
 needed+0.272
 meant+0.250
 spoke+0.232
 faded+0.232
Neg logit contributions
 disl-0.430
n-0.421
 interests-0.413
pieces-0.401
ric-0.396
Token learned
Token ID4499
Feature activation-2.56e+02

Pos logit contributions
 Jones+0.318
 aisle+0.300
 whispered+0.293
 White+0.286
 needed+0.285
Neg logit contributions
n-0.355
 disl-0.348
 victory-0.335
 spend-0.333
 interests-0.332
Token that
Token ID326
Feature activation-2.58e+02

Pos logit contributions
 needed+0.364
 whispered+0.360
 meant+0.355
 said+0.336
 Jones+0.321
Neg logit contributions
n-0.352
pieces-0.351
 Majesty-0.342
iu-0.340
 disl-0.333
Token soccer
Token ID11783
Feature activation-2.50e+02

Pos logit contributions
 whispered+0.433
 needed+0.422
 Jones+0.413
 meant+0.397
 aisle+0.396
Neg logit contributions
 victory-0.276
 interests-0.271
n-0.267
 disl-0.264
pieces-0.244
Token is
Token ID318
Feature activation-2.60e+02

Pos logit contributions
 whispered+0.372
 Jones+0.363
 needed+0.352
 White+0.344
 meant+0.339
Neg logit contributions
 interests-0.301
n-0.300
 disl-0.298
 victory-0.298
pieces-0.282
Token or
Token ID393
Feature activation-2.38e+02

Pos logit contributions
 needed+0.281
 whispered+0.275
 meant+0.263
 Jones+0.249
 said+0.237
Neg logit contributions
n-0.381
 disl-0.375
pieces-0.371
 interests-0.360
 Majesty-0.359
Token toys
Token ID14958
Feature activation-2.35e+02

Pos logit contributions
 needed+0.329
 whispered+0.317
 meant+0.308
 said+0.287
 Jones+0.281
Neg logit contributions
n-0.335
 disl-0.330
pieces-0.323
iu-0.320
nels-0.316
Token,
Token ID11
Feature activation-2.45e+02

Pos logit contributions
 needed+0.360
 whispered+0.357
 meant+0.338
 Jones+0.334
 faded+0.327
Neg logit contributions
n-0.289
 disl-0.269
pieces-0.268
 interests-0.263
 arrest-0.261
Token or
Token ID393
Feature activation-2.43e+02

Pos logit contributions
 needed+0.304
 whispered+0.298
 meant+0.287
 Jones+0.271
 said+0.262
Neg logit contributions
n-0.368
 disl-0.364
pieces-0.359
 Majesty-0.348
 interests-0.347
Token books
Token ID3835
Feature activation-2.37e+02

Pos logit contributions
 needed+0.329
 whispered+0.311
 meant+0.304
 said+0.286
 dark+0.274
Neg logit contributions
n-0.354
 disl-0.349
iu-0.344
pieces-0.342
nels-0.341
Token,"
Token ID553
Feature activation-2.53e+02

Pos logit contributions
 needed+0.383
 whispered+0.377
 Jones+0.357
 spoke+0.353
 White+0.346
Neg logit contributions
n-0.271
 victory-0.241
pieces-0.234
 our-0.234
 blast-0.231
Token Max
Token ID5436
Feature activation-2.41e+02

Pos logit contributions
 needed+0.385
 meant+0.372
 faded+0.359
 whispered+0.358
chy+0.355
Neg logit contributions
n-0.300
 disl-0.296
 hey-0.289
 victory-0.288
pieces-0.285
Token said
Token ID531
Feature activation-2.51e+02

Pos logit contributions
lessness+0.195
 needed+0.187
 faded+0.185
 aisle+0.180
 bored+0.177
Neg logit contributions
n-0.481
 disl-0.437
 interests-0.419
ieved-0.401
 how-0.396
Token.
Token ID13
Feature activation-2.59e+02

Pos logit contributions
 needed+0.305
 meant+0.284
 whispered+0.282
 aisle+0.275
 Jones+0.268
Neg logit contributions
n-0.377
pieces-0.357
iu-0.356
 victory-0.354
 disl-0.352
Token
Token ID198
Feature activation-2.56e+02

Pos logit contributions
 needed+0.386
 whispered+0.376
chy+0.362
 meant+0.354
 spoke+0.351
Neg logit contributions
 blast-0.281
n-0.277
 interests-0.277
 victory-0.276
 join-0.271
Token
Token ID198
Feature activation-2.57e+02

Pos logit contributions
 needed+0.238
 whispered+0.231
 meant+0.220
 Jones+0.203
 faded+0.191
Neg logit contributions
n-0.468
 disl-0.461
pieces-0.455
 interests-0.446
iu-0.442
Token but
Token ID475
Feature activation-2.39e+02

Pos logit contributions
 needed+0.270
 whispered+0.264
 meant+0.255
 Jones+0.247
 spoke+0.235
Neg logit contributions
 disl-0.371
n-0.366
 interests-0.365
 arrest-0.361
 victory-0.359
Token they
Token ID484
Feature activation-2.37e+02

Pos logit contributions
 needed+0.298
 whispered+0.297
 meant+0.280
 Jones+0.277
chy+0.257
Neg logit contributions
n-0.354
 disl-0.349
pieces-0.338
 interests-0.336
 victory-0.332
Token were
Token ID547
Feature activation-2.41e+02

Pos logit contributions
 Jones+0.282
 whispered+0.263
 White+0.257
 aisle+0.253
 meant+0.251
Neg logit contributions
n-0.363
 disl-0.340
 interests-0.327
 spend-0.326
 hey-0.325
Token more
Token ID517
Feature activation-2.40e+02

Pos logit contributions
 needed+0.271
 whispered+0.265
 meant+0.254
 said+0.230
 Jones+0.229
Neg logit contributions
n-0.398
 disl-0.395
pieces-0.391
iu-0.379
 Majesty-0.378
Token careful
Token ID8161
Feature activation-2.42e+02

Pos logit contributions
 needed+0.361
 Jones+0.350
 whispered+0.347
 meant+0.339
 White+0.330
Neg logit contributions
 disl-0.298
 interests-0.285
n-0.284
 Majesty-0.272
pieces-0.271
Token.
Token ID13
Feature activation-2.53e+02

Pos logit contributions
 needed+0.188
 meant+0.175
 whispered+0.163
 said+0.147
 Jones+0.134
Neg logit contributions
 Majesty-0.500
 disl-0.498
pieces-0.494
 arrest-0.484
latable-0.481
Token<|endoftext|>
Token ID50256
Feature activation-2.53e+02

Pos logit contributions
 needed+0.291
 whispered+0.291
 meant+0.271
 spoke+0.259
chy+0.257
Neg logit contributions
 disl-0.401
 interests-0.396
n-0.395
 blast-0.375
pieces-0.375
TokenAnna
Token ID31160
Feature activation-2.47e+02

Pos logit contributions
 whispered+0.280
 needed+0.263
 meant+0.244
chy+0.242
 faded+0.239
Neg logit contributions
n-0.427
 blast-0.390
 victory-0.380
ric-0.376
 interests-0.373
Token and
Token ID290
Feature activation-2.57e+02

Pos logit contributions
Pr+0.309
lessness+0.303
 crooked+0.297
 darker+0.297
Fuel+0.296
Neg logit contributions
n-0.337
 arrest-0.323
 disl-0.312
 interests-0.303
 blast-0.301
Token Ben
Token ID3932
Feature activation-2.49e+02

Pos logit contributions
 needed+0.235
 whispered+0.227
 said+0.223
 meant+0.206
 Jones+0.187
Neg logit contributions
pieces-0.519
ric-0.511
n-0.511
 Majesty-0.510
 occasions-0.508
Token were
Token ID547
Feature activation-2.58e+02

Pos logit contributions
 Jones+0.262
 White+0.261
 whispered+0.254
chy+0.241
 blank+0.240
Neg logit contributions
 spend-0.380
n-0.377
 disl-0.374
 arrest-0.372
 blast-0.363
Token have
Token ID423
Feature activation-2.38e+02

Pos logit contributions
 needed+0.438
 whispered+0.433
 meant+0.423
 Jones+0.412
 faded+0.396
Neg logit contributions
n-0.210
 disl-0.205
pieces-0.200
 interests-0.188
 Majesty-0.187
Token fun
Token ID1257
Feature activation-2.36e+02

Pos logit contributions
 needed+0.371
 whispered+0.366
 meant+0.354
 Jones+0.347
 faded+0.327
Neg logit contributions
n-0.284
 disl-0.279
pieces-0.274
 interests-0.267
 Majesty-0.262
Token.
Token ID13
Feature activation-2.47e+02

Pos logit contributions
 needed+0.148
 whispered+0.140
 meant+0.128
 said+0.122
 dark+0.104
Neg logit contributions
 disl-0.504
n-0.496
 Majesty-0.495
earned-0.494
pieces-0.489
Token<|endoftext|>
Token ID50256
Feature activation-2.49e+02

Pos logit contributions
 needed+0.269
 whispered+0.269
 meant+0.248
 Jones+0.236
 spoke+0.230
Neg logit contributions
 disl-0.410
n-0.409
 interests-0.404
 blast-0.390
pieces-0.388
TokenBen
Token ID11696
Feature activation-2.42e+02

Pos logit contributions
 whispered+0.219
 needed+0.214
 meant+0.192
 faded+0.179
chy+0.177
Neg logit contributions
n-0.468
 blast-0.439
 victory-0.434
 disl-0.433
 arrest-0.431
Token likes
Token ID7832
Feature activation-2.53e+02

Pos logit contributions
 whispered+0.362
 needed+0.350
 White+0.346
 Jones+0.345
 faded+0.340
Neg logit contributions
 disl-0.286
n-0.272
 interests-0.269
 our-0.269
 arrest-0.263
Token to
Token ID284
Feature activation-2.52e+02

Pos logit contributions
 needed+0.355
 whispered+0.348
 meant+0.337
 Jones+0.320
 said+0.311
Neg logit contributions
n-0.341
 disl-0.336
pieces-0.332
 Majesty-0.320
 interests-0.319
Token play
Token ID711
Feature activation-2.55e+02

Pos logit contributions
 needed+0.339
 whispered+0.335
 said+0.328
 meant+0.304
 spoke+0.292
Neg logit contributions
ieved-0.369
 hey-0.369
iu-0.352
 Majesty-0.348
ric-0.346
Token with
Token ID351
Feature activation-2.57e+02

Pos logit contributions
 needed+0.245
 whispered+0.244
 said+0.222
 meant+0.220
 dark+0.211
Neg logit contributions
 disl-0.461
ric-0.454
 Majesty-0.453
pieces-0.447
iu-0.447
Token his
Token ID465
Feature activation-2.53e+02

Pos logit contributions
 needed+0.312
 whispered+0.302
 meant+0.292
 Jones+0.272
 said+0.270
Neg logit contributions
n-0.403
pieces-0.397
 disl-0.390
 Majesty-0.387
iu-0.380
Token toy
Token ID13373
Feature activation-2.53e+02

Pos logit contributions
 needed+0.350
 whispered+0.342
 meant+0.329
 Jones+0.316
 said+0.307
Neg logit contributions
n-0.356
 disl-0.345
pieces-0.343
 Majesty-0.331
iu-0.330
Token OK
Token ID7477
Feature activation-2.38e+02

Pos logit contributions
 needed+0.321
 whispered+0.321
 meant+0.309
 bored+0.303
 said+0.299
Neg logit contributions
 disl-0.356
earned-0.354
 occasions-0.353
pieces-0.350
 interests-0.348
Token,"
Token ID553
Feature activation-2.51e+02

Pos logit contributions
 needed+0.205
 Jones+0.198
 whispered+0.195
 meant+0.191
 said+0.188
Neg logit contributions
 disl-0.472
pieces-0.471
n-0.453
earned-0.453
latable-0.449
Token Tom
Token ID4186
Feature activation-2.40e+02

Pos logit contributions
 needed+0.316
 whispered+0.300
 meant+0.296
 Jones+0.284
chy+0.274
Neg logit contributions
n-0.377
 disl-0.375
pieces-0.367
 interests-0.355
 arrest-0.354
Token and
Token ID290
Feature activation-2.50e+02

Pos logit contributions
 needed+0.308
 faded+0.275
 shone+0.269
chy+0.256
 smelled+0.255
Neg logit contributions
n-0.360
 interests-0.342
 disl-0.334
 our-0.318
 pause-0.316
Token Lily
Token ID20037
Feature activation-2.44e+02

Pos logit contributions
 needed+0.240
 whispered+0.235
 meant+0.217
 said+0.205
 Jones+0.196
Neg logit contributions
n-0.460
pieces-0.458
 disl-0.452
ric-0.445
 Majesty-0.443
Token say
Token ID910
Feature activation-2.53e+02

Pos logit contributions
 needed+0.403
 whispered+0.374
 meant+0.374
 faded+0.364
 Jones+0.363
Neg logit contributions
n-0.278
 disl-0.260
pieces-0.253
 interests-0.247
 hey-0.243
Token.
Token ID13
Feature activation-2.59e+02

Pos logit contributions
 needed+0.219
 whispered+0.213
 meant+0.201
 Jones+0.185
 said+0.176
Neg logit contributions
n-0.477
 disl-0.473
pieces-0.468
 Majesty-0.456
 interests-0.455
Token "
Token ID366
Feature activation-2.56e+02

Pos logit contributions
 needed+0.341
chy+0.331
 whispered+0.330
 meant+0.320
 spoke+0.300
Neg logit contributions
 interests-0.356
 blast-0.345
 disl-0.341
 our-0.340
n-0.332
TokenBut
Token ID1537
Feature activation-2.60e+02

Pos logit contributions
 needed+0.405
 whispered+0.400
 meant+0.387
 Jones+0.369
 said+0.354
Neg logit contributions
n-0.316
 disl-0.307
 interests-0.293
pieces-0.286
 hey-0.283
Token we
Token ID356
Feature activation-2.59e+02

Pos logit contributions
 needed+0.314
 said+0.307
 meant+0.284
 whispered+0.270
 smelled+0.257
Neg logit contributions
pieces-0.487
earned-0.485
 arrest-0.476
latable-0.462
 impression-0.455
Token were
Token ID547
Feature activation-2.63e+02

Pos logit contributions
 needed+0.322
 whispered+0.301
 meant+0.297
 said+0.276
 Jones+0.252
Neg logit contributions
pieces-0.432
 Majesty-0.419
iu-0.414
 disl-0.414
n-0.412
Token go
Token ID467
Feature activation-2.40e+02

Pos logit contributions
 needed+0.326
 said+0.322
 whispered+0.303
 smelled+0.283
 meant+0.281
Neg logit contributions
 Majesty-0.399
heit-0.387
iu-0.384
 disl-0.380
pieces-0.369
Token outside
Token ID2354
Feature activation-2.39e+02

Pos logit contributions
 whispered+0.252
 needed+0.244
 said+0.243
 dark+0.225
 bored+0.209
Neg logit contributions
 disl-0.475
iu-0.448
 Majesty-0.438
 accuracy-0.425
heit-0.424
Token now
Token ID783
Feature activation-2.42e+02

Pos logit contributions
 needed+0.293
 whispered+0.288
 meant+0.276
 Jones+0.263
 said+0.255
Neg logit contributions
n-0.364
 disl-0.362
pieces-0.355
iu-0.345
 interests-0.344
Token?"
Token ID1701
Feature activation-2.53e+02

Pos logit contributions
 whispered+0.317
 needed+0.311
 meant+0.302
 said+0.299
 Jones+0.298
Neg logit contributions
pieces-0.367
 disl-0.363
n-0.349
ric-0.349
 interests-0.348
Token Anna
Token ID11735
Feature activation-2.43e+02

Pos logit contributions
 needed+0.328
 meant+0.312
 whispered+0.311
chy+0.304
 Jones+0.291
Neg logit contributions
n-0.342
 blast-0.337
 disl-0.336
pieces-0.333
 our-0.331
Token and
Token ID290
Feature activation-2.53e+02

Pos logit contributions
chy+0.330
 flowed+0.310
 needed+0.308
 shone+0.303
 faded+0.301
Neg logit contributions
n-0.329
 interests-0.288
iu-0.276
han-0.264
 Amen-0.253
Token Ben
Token ID3932
Feature activation-2.45e+02

Pos logit contributions
 needed+0.233
 meant+0.217
 whispered+0.216
 Jones+0.188
 faded+0.184
Neg logit contributions
n-0.450
pieces-0.445
 disl-0.443
 arrest-0.440
iu-0.436
Token asked
Token ID1965
Feature activation-2.58e+02

Pos logit contributions
chy+0.313
oky+0.288
 faded+0.286
 Lover+0.286
 needed+0.282
Neg logit contributions
n-0.369
 pause-0.328
 spend-0.323
 interests-0.292
iu-0.290
Token.
Token ID13
Feature activation-2.63e+02

Pos logit contributions
 needed+0.322
 whispered+0.307
 meant+0.307
 faded+0.288
 spoke+0.279
Neg logit contributions
n-0.375
 interests-0.368
pieces-0.363
ap-0.362
 victory-0.360
Token
Token ID198
Feature activation-2.60e+02

Pos logit contributions
 needed+0.402
chy+0.399
 whispered+0.380
 meant+0.376
 faded+0.355
Neg logit contributions
n-0.301
 interests-0.298
 blast-0.294
 join-0.281
 our-0.280
Token
Token ID198
Feature activation-2.61e+02

Pos logit contributions
 needed+0.245
 whispered+0.236
 meant+0.228
 Jones+0.208
 faded+0.197
Neg logit contributions
n-0.475
 disl-0.467
pieces-0.460
 interests-0.451
 Majesty-0.448
TokenCan
Token ID6090
Feature activation-2.40e+02

Pos logit contributions
 needed+0.345
 whispered+0.313
 Jones+0.310
 bored+0.310
chy+0.310
Neg logit contributions
n-0.319
 blast-0.307
 interests-0.290
 hey-0.288
 occasions-0.282
Token we
Token ID356
Feature activation-2.37e+02

Pos logit contributions
 needed+0.292
 said+0.266
 whispered+0.242
 she+0.239
 could+0.237
Neg logit contributions
 Majesty-0.442
 impression-0.429
pieces-0.425
 disl-0.422
latable-0.420
Token go
Token ID467
Feature activation-2.40e+02

Pos logit contributions
 needed+0.326
 said+0.322
 whispered+0.303
 smelled+0.283
 meant+0.281
Neg logit contributions
 Majesty-0.399
heit-0.387
iu-0.384
 disl-0.380
pieces-0.369
Token outside
Token ID2354
Feature activation-2.39e+02

Pos logit contributions
 whispered+0.252
 needed+0.244
 said+0.243
 dark+0.225
 bored+0.209
Neg logit contributions
 disl-0.475
iu-0.448
 Majesty-0.438
 accuracy-0.425
heit-0.424
Token now
Token ID783
Feature activation-2.42e+02

Pos logit contributions
 needed+0.293
 whispered+0.288
 meant+0.276
 Jones+0.263
 said+0.255
Neg logit contributions
n-0.364
 disl-0.362
pieces-0.355
iu-0.345
 interests-0.344
Token?"
Token ID1701
Feature activation-2.53e+02

Pos logit contributions
 whispered+0.317
 needed+0.311
 meant+0.302
 said+0.299
 Jones+0.298
Neg logit contributions
pieces-0.367
 disl-0.363
n-0.349
ric-0.349
 interests-0.348
Token Anna
Token ID11735
Feature activation-2.43e+02

Pos logit contributions
 needed+0.328
 meant+0.312
 whispered+0.311
chy+0.304
 Jones+0.291
Neg logit contributions
n-0.342
 blast-0.337
 disl-0.336
pieces-0.333
 our-0.331
Token and
Token ID290
Feature activation-2.53e+02

Pos logit contributions
chy+0.330
 flowed+0.310
 needed+0.308
 shone+0.303
 faded+0.301
Neg logit contributions
n-0.329
 interests-0.288
iu-0.276
han-0.264
 Amen-0.253
Token Ben
Token ID3932
Feature activation-2.45e+02

Pos logit contributions
 needed+0.233
 meant+0.217
 whispered+0.216
 Jones+0.188
 faded+0.184
Neg logit contributions
n-0.450
pieces-0.445
 disl-0.443
 arrest-0.440
iu-0.436
Token asked
Token ID1965
Feature activation-2.58e+02

Pos logit contributions
chy+0.313
oky+0.288
 faded+0.286
 Lover+0.286
 needed+0.282
Neg logit contributions
n-0.369
 pause-0.328
 spend-0.323
 interests-0.292
iu-0.290
Token.
Token ID13
Feature activation-2.63e+02

Pos logit contributions
 needed+0.322
 whispered+0.307
 meant+0.307
 faded+0.288
 spoke+0.279
Neg logit contributions
n-0.375
 interests-0.368
pieces-0.363
ap-0.362
 victory-0.360
Token.
Token ID13
Feature activation-2.48e+02

Pos logit contributions
 whispered+0.286
 needed+0.281
 Jones+0.273
 meant+0.272
 spoke+0.256
Neg logit contributions
n-0.352
 victory-0.344
pieces-0.334
ap-0.328
 blast-0.326
Token They
Token ID1119
Feature activation-2.45e+02

Pos logit contributions
 needed+0.263
 whispered+0.259
 meant+0.247
 Jones+0.226
 spoke+0.220
Neg logit contributions
n-0.420
 disl-0.410
pieces-0.409
 interests-0.404
 Majesty-0.394
Token are
Token ID389
Feature activation-2.47e+02

Pos logit contributions
 needed+0.444
 whispered+0.435
 meant+0.420
 said+0.403
 Jones+0.401
Neg logit contributions
 disl-0.234
n-0.233
pieces-0.228
iu-0.221
 Majesty-0.220
Token clever
Token ID14169
Feature activation-2.44e+02

Pos logit contributions
 needed+0.264
 whispered+0.259
 meant+0.249
 Jones+0.227
 said+0.223
Neg logit contributions
n-0.417
 disl-0.415
pieces-0.408
iu-0.397
 interests-0.396
Token.
Token ID13
Feature activation-2.55e+02

Pos logit contributions
 needed+0.250
 whispered+0.244
 meant+0.231
 Jones+0.219
 faded+0.206
Neg logit contributions
n-0.418
 disl-0.412
pieces-0.409
 interests-0.401
 Majesty-0.398
Token They
Token ID1119
Feature activation-2.52e+02

Pos logit contributions
 needed+0.257
 whispered+0.252
 meant+0.240
 Jones+0.219
 faded+0.213
Neg logit contributions
n-0.447
 disl-0.437
pieces-0.435
 interests-0.431
 Majesty-0.418
Token know
Token ID760
Feature activation-2.56e+02

Pos logit contributions
 needed+0.440
 whispered+0.431
 meant+0.416
 said+0.400
 Jones+0.390
Neg logit contributions
n-0.258
 disl-0.256
pieces-0.255
iu-0.250
 Majesty-0.245
Token how
Token ID703
Feature activation-2.58e+02

Pos logit contributions
 needed+0.476
 whispered+0.470
 meant+0.459
 said+0.436
 Jones+0.435
Neg logit contributions
n-0.234
 disl-0.228
pieces-0.222
iu-0.215
 arrest-0.212
Token to
Token ID284
Feature activation-2.58e+02

Pos logit contributions
 needed+0.253
 dark+0.204
 sleepy+0.201
 bored+0.200
 whispered+0.195
Neg logit contributions
iu-0.514
 Majesty-0.499
han-0.495
n-0.493
pieces-0.488
Token share
Token ID2648
Feature activation-2.62e+02

Pos logit contributions
 needed+0.352
 whispered+0.343
chy+0.341
 said+0.328
 meant+0.328
Neg logit contributions
iu-0.383
pieces-0.369
 disl-0.363
 Majesty-0.363
 interests-0.338
Token and
Token ID290
Feature activation-2.64e+02

Pos logit contributions
 whispered+0.256
 needed+0.249
 meant+0.244
 said+0.227
 Jones+0.218
Neg logit contributions
n-0.464
 Majesty-0.456
 disl-0.454
iu-0.454
pieces-0.454
Token and
Token ID290
Feature activation-2.39e+02

Pos logit contributions
 needed+0.210
 whispered+0.201
 meant+0.196
 Jones+0.180
 said+0.178
Neg logit contributions
n-0.438
 disl-0.434
pieces-0.434
iu-0.426
ric-0.424
Token happy
Token ID3772
Feature activation-2.38e+02

Pos logit contributions
 needed+0.260
 whispered+0.252
 meant+0.241
 said+0.231
 dark+0.206
Neg logit contributions
n-0.429
ric-0.413
iu-0.410
pieces-0.408
 arrest-0.397
Token and
Token ID290
Feature activation-2.41e+02

Pos logit contributions
 needed+0.260
 whispered+0.252
 meant+0.243
 said+0.231
 Jones+0.227
Neg logit contributions
n-0.405
pieces-0.405
 disl-0.395
iu-0.385
ric-0.384
Token good
Token ID922
Feature activation-2.38e+02

Pos logit contributions
 needed+0.299
 whispered+0.293
 meant+0.281
 said+0.263
 Jones+0.256
Neg logit contributions
n-0.377
pieces-0.367
iu-0.359
 disl-0.359
ric-0.357
Token things
Token ID1243
Feature activation-2.39e+02

Pos logit contributions
 needed+0.278
 whispered+0.273
 meant+0.260
 Jones+0.246
 faded+0.235
Neg logit contributions
n-0.378
 disl-0.374
pieces-0.367
 interests-0.359
 Majesty-0.358
Token,"
Token ID553
Feature activation-2.52e+02

Pos logit contributions
 needed+0.325
 whispered+0.320
 Jones+0.308
 White+0.297
 meant+0.292
Neg logit contributions
n-0.320
 disl-0.314
 interests-0.299
 victory-0.298
 our-0.292
Token Sam
Token ID3409
Feature activation-2.40e+02

Pos logit contributions
 needed+0.368
 whispered+0.353
 meant+0.352
chy+0.341
 faded+0.331
Neg logit contributions
 disl-0.328
pieces-0.321
n-0.320
 arrest-0.311
 victory-0.310
Token says
Token ID1139
Feature activation-2.50e+02

Pos logit contributions
 needed+0.278
 faded+0.249
 shone+0.234
chy+0.232
 smelled+0.222
Neg logit contributions
n-0.383
 disl-0.373
 interests-0.368
pieces-0.354
 Majesty-0.350
Token.
Token ID13
Feature activation-2.58e+02

Pos logit contributions
 needed+0.240
 whispered+0.232
 meant+0.225
 Jones+0.205
 faded+0.194
Neg logit contributions
n-0.448
pieces-0.438
 disl-0.438
iu-0.428
 interests-0.426
Token
Token ID198
Feature activation-2.55e+02

Pos logit contributions
chy+0.428
 needed+0.420
 meant+0.401
 whispered+0.396
 dusty+0.376
Neg logit contributions
n-0.249
 interests-0.249
 blast-0.248
 disl-0.244
 join-0.244
Token
Token ID198
Feature activation-2.56e+02

Pos logit contributions
 needed+0.231
 whispered+0.224
 meant+0.212
 Jones+0.195
 said+0.183
Neg logit contributions
n-0.472
 disl-0.467
pieces-0.461
 interests-0.450
 Majesty-0.448
Token.
Token ID13
Feature activation-2.42e+02

Pos logit contributions
 needed+0.247
 whispered+0.240
 meant+0.229
 Jones+0.226
 White+0.208
Neg logit contributions
n-0.388
pieces-0.387
 disl-0.378
 interests-0.376
 victory-0.374
Token They
Token ID1119
Feature activation-2.39e+02

Pos logit contributions
 needed+0.309
 whispered+0.302
 meant+0.283
chy+0.269
 spoke+0.268
Neg logit contributions
 disl-0.370
n-0.369
 interests-0.366
 blast-0.350
 victory-0.340
Token are
Token ID389
Feature activation-2.41e+02

Pos logit contributions
 needed+0.423
 whispered+0.417
 meant+0.404
 said+0.385
 Jones+0.382
Neg logit contributions
n-0.234
 disl-0.232
pieces-0.226
iu-0.221
 Majesty-0.217
Token happy
Token ID3772
Feature activation-2.42e+02

Pos logit contributions
 needed+0.265
 whispered+0.259
 meant+0.251
 said+0.231
 bored+0.226
Neg logit contributions
n-0.408
 disl-0.402
pieces-0.390
iu-0.389
 arrest-0.388
Token friends
Token ID2460
Feature activation-2.38e+02

Pos logit contributions
 needed+0.233
 whispered+0.228
 meant+0.217
 Jones+0.201
 said+0.188
Neg logit contributions
n-0.433
 disl-0.430
pieces-0.424
 interests-0.415
 Majesty-0.413
Token.
Token ID13
Feature activation-2.52e+02

Pos logit contributions
 needed+0.248
 whispered+0.245
 meant+0.231
 Jones+0.214
 spoke+0.206
Neg logit contributions
n-0.400
pieces-0.391
 disl-0.388
 victory-0.384
 interests-0.381
Token<|endoftext|>
Token ID50256
Feature activation-2.53e+02

Pos logit contributions
 needed+0.274
 whispered+0.272
 meant+0.253
 Jones+0.235
 faded+0.232
Neg logit contributions
 disl-0.419
n-0.418
 interests-0.412
pieces-0.395
 blast-0.395
TokenL
Token ID43
Feature activation-2.40e+02

Pos logit contributions
 whispered+0.235
 needed+0.229
 meant+0.207
 faded+0.195
chy+0.193
Neg logit contributions
n-0.463
 blast-0.435
 victory-0.432
 interests-0.428
 disl-0.427
Tokenily
Token ID813
Feature activation-2.45e+02

Pos logit contributions
 whispered+0.256
 needed+0.255
 meant+0.241
 said+0.228
 Jones+0.223
Neg logit contributions
 disl-0.415
n-0.394
 Majesty-0.391
pieces-0.391
 interests-0.385
Token and
Token ID290
Feature activation-2.56e+02

Pos logit contributions
 whispered+0.354
 needed+0.347
 meant+0.339
 Jones+0.335
 White+0.329
Neg logit contributions
 disl-0.291
n-0.290
 arrest-0.287
 interests-0.278
 Majesty-0.277
Token Tom
Token ID4186
Feature activation-2.48e+02

Pos logit contributions
 needed+0.259
 said+0.257
 whispered+0.237
 meant+0.233
 She+0.210
Neg logit contributions
 Majesty-0.535
immers-0.530
pieces-0.522
ric-0.520
 occasions-0.519
Token swim
Token ID9422
Feature activation-2.33e+02

Pos logit contributions
 needed+0.277
 whispered+0.264
 dark+0.261
 said+0.260
 sleepy+0.255
Neg logit contributions
pieces-0.386
iu-0.377
 interests-0.377
han-0.371
nels-0.361
Token with
Token ID351
Feature activation-2.38e+02

Pos logit contributions
 needed+0.338
 whispered+0.330
 meant+0.313
 Jones+0.302
 said+0.301
Neg logit contributions
 disl-0.310
n-0.304
pieces-0.303
iu-0.295
 Majesty-0.290
Token you
Token ID345
Feature activation-2.35e+02

Pos logit contributions
 needed+0.375
 whispered+0.366
 said+0.354
 dark+0.342
 meant+0.338
Neg logit contributions
 disl-0.281
iu-0.280
pieces-0.278
ric-0.277
 interests-0.272
Token,
Token ID11
Feature activation-2.44e+02

Pos logit contributions
 needed+0.408
 meant+0.406
 whispered+0.404
 faded+0.381
 spoke+0.378
Neg logit contributions
n-0.234
 Majesty-0.222
pieces-0.216
 victory-0.216
 interests-0.207
Token duck
Token ID22045
Feature activation-2.34e+02

Pos logit contributions
 needed+0.302
 whispered+0.296
 meant+0.284
 Jones+0.269
 faded+0.257
Neg logit contributions
n-0.372
 disl-0.366
pieces-0.361
 Majesty-0.350
 interests-0.349
Token!"
Token ID2474
Feature activation-2.52e+02

Pos logit contributions
 needed+0.380
 whispered+0.367
 meant+0.355
 Jones+0.342
 said+0.340
Neg logit contributions
n-0.268
 disl-0.268
pieces-0.265
iu-0.260
 interests-0.255
Token And
Token ID843
Feature activation-2.53e+02

Pos logit contributions
 meant+0.330
chy+0.328
 needed+0.323
 faded+0.312
 dusty+0.301
Neg logit contributions
n-0.385
 disl-0.331
iu-0.316
pieces-0.308
 arrest-0.307
Token they
Token ID484
Feature activation-2.50e+02

Pos logit contributions
 needed+0.336
 meant+0.320
 whispered+0.317
 Jones+0.311
chy+0.303
Neg logit contributions
n-0.362
pieces-0.347
 victory-0.329
 disl-0.326
 interests-0.325
Token sw
Token ID1509
Feature activation-2.38e+02

Pos logit contributions
 needed+0.342
 meant+0.342
 Jones+0.340
 whispered+0.337
 aisle+0.316
Neg logit contributions
n-0.334
 blast-0.304
 spend-0.301
 victory-0.298
pieces-0.295
Tokenam
Token ID321
Feature activation-2.51e+02

Pos logit contributions
 needed+0.291
 whispered+0.289
 said+0.272
 meant+0.254
ed+0.253
Neg logit contributions
 disl-0.357
pieces-0.350
 hey-0.337
 blast-0.333
 risk-0.325
Token together
Token ID1978
Feature activation-2.54e+02

Pos logit contributions
 needed+0.272
 whispered+0.271
 said+0.262
 dark+0.240
 meant+0.240
Neg logit contributions
 disl-0.465
ric-0.439
 Majesty-0.431
pieces-0.425
 hey-0.415
Token bad
Token ID2089
Feature activation-2.30e+02

Pos logit contributions
 needed+0.332
 whispered+0.326
 meant+0.317
 Jones+0.299
 said+0.292
Neg logit contributions
n-0.308
 disl-0.304
pieces-0.301
 Majesty-0.289
iu-0.288
Token.
Token ID13
Feature activation-2.41e+02

Pos logit contributions
 needed+0.269
 whispered+0.269
 meant+0.252
 Jones+0.250
 White+0.236
Neg logit contributions
n-0.354
 disl-0.345
 interests-0.339
pieces-0.333
 arrest-0.330
Token It
Token ID632
Feature activation-2.35e+02

Pos logit contributions
 needed+0.281
 meant+0.271
 whispered+0.269
 Jones+0.260
 White+0.253
Neg logit contributions
pieces-0.395
n-0.375
 victory-0.367
 Majesty-0.366
nels-0.365
Token hurt
Token ID5938
Feature activation-2.38e+02

Pos logit contributions
 needed+0.224
 whispered+0.222
 meant+0.208
 Jones+0.204
 faded+0.185
Neg logit contributions
n-0.419
 disl-0.411
pieces-0.401
 arrest-0.393
 interests-0.392
Token me
Token ID502
Feature activation-2.35e+02

Pos logit contributions
 needed+0.457
 whispered+0.454
 meant+0.439
 Jones+0.430
 spoke+0.421
Neg logit contributions
n-0.195
 disl-0.181
 interests-0.178
 arrest-0.172
 victory-0.170
Token."
Token ID526
Feature activation-2.52e+02

Pos logit contributions
 needed+0.333
 whispered+0.323
 spoke+0.319
 Jones+0.315
 faded+0.314
Neg logit contributions
n-0.318
 interests-0.274
 our-0.274
 victory-0.272
pieces-0.270
Token Anna
Token ID11735
Feature activation-2.40e+02

Pos logit contributions
 meant+0.345
 needed+0.344
 whispered+0.343
chy+0.320
 Jones+0.312
Neg logit contributions
 blast-0.316
 victory-0.312
 disl-0.306
n-0.302
 our-0.300
Token said
Token ID531
Feature activation-2.51e+02

Pos logit contributions
chy+0.248
 aisle+0.233
Today+0.232
 blank+0.230
books+0.229
Neg logit contributions
n-0.401
iu-0.354
 hey-0.347
 arrest-0.341
 punish-0.337
Token.
Token ID13
Feature activation-2.57e+02

Pos logit contributions
 needed+0.229
 whispered+0.222
 meant+0.211
 Jones+0.194
 said+0.182
Neg logit contributions
n-0.462
 disl-0.455
pieces-0.451
 interests-0.440
iu-0.440
Token
Token ID198
Feature activation-2.54e+02

Pos logit contributions
 needed+0.390
chy+0.373
 meant+0.364
 whispered+0.355
 shone+0.348
Neg logit contributions
 blast-0.274
 hey-0.270
 join-0.267
n-0.264
 interests-0.259
Token
Token ID198
Feature activation-2.56e+02

Pos logit contributions
 needed+0.285
 meant+0.268
 whispered+0.268
 Jones+0.241
 faded+0.239
Neg logit contributions
n-0.421
 disl-0.416
 arrest-0.398
 interests-0.397
pieces-0.391
Token.
Token ID13
Feature activation-2.39e+02

Pos logit contributions
 needed+0.264
 whispered+0.254
 meant+0.252
 spoke+0.229
 Jones+0.227
Neg logit contributions
n-0.362
pieces-0.346
 disl-0.344
 victory-0.343
 interests-0.341
Token They
Token ID1119
Feature activation-2.35e+02

Pos logit contributions
 needed+0.334
 whispered+0.318
 meant+0.314
chy+0.309
 spoke+0.294
Neg logit contributions
 disl-0.321
 blast-0.316
 our-0.314
 interests-0.313
n-0.306
Token smiled
Token ID13541
Feature activation-2.38e+02

Pos logit contributions
 Jones+0.295
elly+0.279
oky+0.278
 aisle+0.274
Fuel+0.271
Neg logit contributions
 blast-0.331
 spend-0.327
n-0.325
 stay-0.310
 pause-0.309
Token and
Token ID290
Feature activation-2.42e+02

Pos logit contributions
 needed+0.310
 aisle+0.301
 whispered+0.297
 meant+0.283
 spoke+0.281
Neg logit contributions
 victory-0.326
 interests-0.321
 our-0.311
ap-0.305
 Majesty-0.303
Token laughed
Token ID13818
Feature activation-2.44e+02

Pos logit contributions
 Jones+0.237
 aisle+0.237
 needed+0.237
chy+0.229
 meant+0.223
Neg logit contributions
 blast-0.382
n-0.375
 disl-0.368
 victory-0.364
 interests-0.362
Token.
Token ID13
Feature activation-2.52e+02

Pos logit contributions
 needed+0.344
 whispered+0.327
 aisle+0.324
 meant+0.315
 Jones+0.312
Neg logit contributions
 victory-0.329
ap-0.302
 interests-0.301
pieces-0.300
 our-0.299
Token They
Token ID1119
Feature activation-2.48e+02

Pos logit contributions
 needed+0.331
 whispered+0.323
 meant+0.309
chy+0.301
 spoke+0.291
Neg logit contributions
 disl-0.356
 interests-0.355
 blast-0.345
n-0.343
 our-0.337
Token had
Token ID550
Feature activation-2.51e+02

Pos logit contributions
 Jones+0.309
 aisle+0.297
Fuel+0.285
oky+0.283
elly+0.273
Neg logit contributions
n-0.349
 blast-0.346
 spend-0.345
 pause-0.324
 stay-0.323
Token a
Token ID257
Feature activation-2.52e+02

Pos logit contributions
 Jones+0.405
 aisle+0.403
 needed+0.391
 White+0.376
 meant+0.371
Neg logit contributions
n-0.280
 disl-0.274
 interests-0.270
 victory-0.268
 blast-0.266
Token great
Token ID1049
Feature activation-2.50e+02

Pos logit contributions
 needed+0.384
 whispered+0.379
 meant+0.372
 Jones+0.353
 said+0.336
Neg logit contributions
n-0.302
 disl-0.302
 victory-0.293
pieces-0.293
 blast-0.291
Token adventure
Token ID8855
Feature activation-2.53e+02

Pos logit contributions
 needed+0.274
 whispered+0.252
 meant+0.241
 dark+0.236
 Jones+0.229
Neg logit contributions
pieces-0.434
 disl-0.421
n-0.421
ric-0.411
ieved-0.405
Token He
Token ID679
Feature activation-2.33e+02

Pos logit contributions
 needed+0.279
 whispered+0.267
 meant+0.253
 spoke+0.247
chy+0.242
Neg logit contributions
 disl-0.389
 Majesty-0.375
 yeah-0.353
n-0.349
 interests-0.346
Token gave
Token ID2921
Feature activation-2.41e+02

Pos logit contributions
 Jones+0.261
chy+0.225
 White+0.225
 meant+0.223
 aisle+0.221
Neg logit contributions
n-0.365
 disl-0.342
 arrest-0.341
 interests-0.337
 blast-0.328
Token her
Token ID607
Feature activation-2.37e+02

Pos logit contributions
 whispered+0.240
 needed+0.232
 said+0.222
 meant+0.222
 Jones+0.204
Neg logit contributions
 disl-0.472
pieces-0.444
 blast-0.434
 arrest-0.432
 interests-0.429
Token the
Token ID262
Feature activation-2.39e+02

Pos logit contributions
 whispered+0.228
 needed+0.227
 meant+0.214
 said+0.205
 dark+0.197
Neg logit contributions
n-0.430
iu-0.426
 Majesty-0.424
 hey-0.411
 occasions-0.410
Token ball
Token ID2613
Feature activation-2.34e+02

Pos logit contributions
 needed+0.324
 whispered+0.317
 meant+0.307
 Jones+0.291
 said+0.282
Neg logit contributions
n-0.337
 disl-0.332
pieces-0.327
 Majesty-0.316
iu-0.316
Token.
Token ID13
Feature activation-2.51e+02

Pos logit contributions
 Jones+0.395
 whispered+0.390
 needed+0.373
 meant+0.373
 White+0.370
Neg logit contributions
 victory-0.238
n-0.220
 respect-0.217
 our-0.217
 interests-0.212
Token "
Token ID366
Feature activation-2.48e+02

Pos logit contributions
 needed+0.311
 whispered+0.294
 meant+0.279
 spoke+0.271
chy+0.269
Neg logit contributions
 disl-0.400
 Majesty-0.383
n-0.372
 yeah-0.370
 interests-0.356
TokenI
Token ID40
Feature activation-2.51e+02

Pos logit contributions
 needed+0.333
 Jones+0.316
 whispered+0.315
 meant+0.308
chy+0.293
Neg logit contributions
n-0.367
 disl-0.361
 interests-0.332
 arrest-0.330
 hey-0.328
Token'm
Token ID1101
Feature activation-2.49e+02

Pos logit contributions
 needed+0.417
 whispered+0.416
 meant+0.402
 Jones+0.401
 White+0.381
Neg logit contributions
n-0.270
 disl-0.266
 interests-0.245
pieces-0.240
 victory-0.236
Token sorry
Token ID7926
Feature activation-2.51e+02

Pos logit contributions
 whispered+0.331
 needed+0.327
 meant+0.316
 said+0.307
 bored+0.298
Neg logit contributions
 interests-0.355
pieces-0.354
 disl-0.354
 occasions-0.352
n-0.351
Token,
Token ID11
Feature activation-2.58e+02

Pos logit contributions
 needed+0.199
 whispered+0.184
 meant+0.181
 said+0.179
 Jones+0.169
Neg logit contributions
pieces-0.516
n-0.505
 disl-0.505
 arrest-0.499
iu-0.490
Token it
Token ID340
Feature activation-2.29e+02

Pos logit contributions
 needed+0.387
 whispered+0.375
 meant+0.366
 Jones+0.348
 said+0.345
Neg logit contributions
n-0.259
 disl-0.254
pieces-0.252
iu-0.245
 Majesty-0.242
Token is
Token ID318
Feature activation-2.35e+02

Pos logit contributions
 Jones+0.291
 whispered+0.276
 White+0.264
 aisle+0.263
Fuel+0.259
Neg logit contributions
n-0.344
 disl-0.316
 arrest-0.305
 spend-0.300
 Majesty-0.297
Token!
Token ID0
Feature activation-2.43e+02

Pos logit contributions
 whispered+0.334
 Jones+0.322
 aisle+0.320
 White+0.316
Fuel+0.314
Neg logit contributions
n-0.302
 victory-0.280
 Majesty-0.277
 disl-0.275
 our-0.274
Token Follow
Token ID7281
Feature activation-2.42e+02

Pos logit contributions
 needed+0.333
 whispered+0.321
 meant+0.318
 Jones+0.306
 White+0.296
Neg logit contributions
pieces-0.333
n-0.332
 disl-0.328
iu-0.324
 arrest-0.322
Token me
Token ID502
Feature activation-2.37e+02

Pos logit contributions
 whispered+0.271
 Jones+0.253
 said+0.253
 T+0.243
 needed+0.238
Neg logit contributions
pieces-0.404
iu-0.396
 disl-0.396
ric-0.387
latable-0.380
Token!"
Token ID2474
Feature activation-2.51e+02

Pos logit contributions
 needed+0.308
 whispered+0.307
 meant+0.289
 Jones+0.280
 said+0.271
Neg logit contributions
 disl-0.353
pieces-0.340
iu-0.339
n-0.333
 interests-0.322
Token
Token ID198
Feature activation-2.50e+02

Pos logit contributions
 needed+0.376
 whispered+0.371
 meant+0.367
 faded+0.362
chy+0.360
Neg logit contributions
n-0.319
 disl-0.300
pieces-0.299
 arrest-0.281
 blast-0.271
Token
Token ID198
Feature activation-2.51e+02

Pos logit contributions
 needed+0.219
 whispered+0.213
 meant+0.200
 Jones+0.183
 said+0.178
Neg logit contributions
n-0.469
 disl-0.467
pieces-0.462
 Majesty-0.454
iu-0.451
TokenThe
Token ID464
Feature activation-2.52e+02

Pos logit contributions
 needed+0.354
 whispered+0.348
 meant+0.335
 Jones+0.316
 said+0.312
Neg logit contributions
n-0.339
 disl-0.335
pieces-0.330
 Majesty-0.323
iu-0.320
Token squirrel
Token ID33039
Feature activation-2.46e+02

Pos logit contributions
 needed+0.276
 whispered+0.271
 meant+0.258
 Jones+0.244
 said+0.230
Neg logit contributions
n-0.420
 disl-0.414
pieces-0.408
 interests-0.398
 Majesty-0.397
Token led
Token ID2957
Feature activation-2.59e+02

Pos logit contributions
 Jones+0.304
 bored+0.284
 meant+0.269
 needed+0.269
 dark+0.266
Neg logit contributions
n-0.406
 arrest-0.336
 victory-0.332
 pause-0.330
ered-0.329
Token forgot
Token ID16453
Feature activation-2.38e+02

Pos logit contributions
 Jones+0.327
 aisle+0.315
oky+0.303
Fuel+0.297
elly+0.294
Neg logit contributions
n-0.291
 spend-0.285
 enjoy-0.284
 share-0.276
 blast-0.272
Token about
Token ID546
Feature activation-2.39e+02

Pos logit contributions
 needed+0.409
 whispered+0.404
 meant+0.391
 Jones+0.381
 faded+0.367
Neg logit contributions
n-0.246
 disl-0.242
pieces-0.233
 interests-0.225
 Majesty-0.221
Token lunch
Token ID9965
Feature activation-2.35e+02

Pos logit contributions
 needed+0.317
 whispered+0.311
 meant+0.300
 Jones+0.285
 said+0.275
Neg logit contributions
n-0.343
 disl-0.338
pieces-0.334
 interests-0.323
 Majesty-0.322
Token.
Token ID13
Feature activation-2.50e+02

Pos logit contributions
 White+0.372
Dead+0.362
 whispered+0.362
Ancient+0.358
 Jones+0.355
Neg logit contributions
 victory-0.245
n-0.245
 occasions-0.228
ap-0.226
 interests-0.225
Token They
Token ID1119
Feature activation-2.46e+02

Pos logit contributions
 whispered+0.298
 needed+0.295
 meant+0.277
 spoke+0.265
chy+0.258
Neg logit contributions
 disl-0.395
n-0.376
 interests-0.374
 blast-0.361
nels-0.351
Token loved
Token ID6151
Feature activation-2.51e+02

Pos logit contributions
 Jones+0.353
oky+0.327
elly+0.323
Fuel+0.323
Ancient+0.323
Neg logit contributions
 spend-0.310
n-0.306
 enjoy-0.304
 share-0.291
 wait-0.282
Token grapes
Token ID36656
Feature activation-2.42e+02

Pos logit contributions
 needed+0.367
 whispered+0.360
 meant+0.349
 Jones+0.333
 said+0.323
Neg logit contributions
n-0.327
 disl-0.322
pieces-0.318
 Majesty-0.305
 interests-0.305
Token.
Token ID13
Feature activation-2.60e+02

Pos logit contributions
 needed+0.317
 whispered+0.304
 Jones+0.300
 meant+0.293
 White+0.290
Neg logit contributions
n-0.334
 victory-0.321
pieces-0.308
 interests-0.305
 disl-0.302
Token And
Token ID843
Feature activation-2.60e+02

Pos logit contributions
 needed+0.276
 whispered+0.274
 meant+0.255
 Jones+0.237
 spoke+0.235
Neg logit contributions
 disl-0.440
n-0.432
 interests-0.423
pieces-0.413
 blast-0.408
Token they
Token ID484
Feature activation-2.57e+02

Pos logit contributions
 needed+0.266
 whispered+0.256
 meant+0.247
 said+0.228
 Jones+0.223
Neg logit contributions
n-0.458
 disl-0.457
pieces-0.454
 arrest-0.444
 Majesty-0.442
Token loved
Token ID6151
Feature activation-2.63e+02

Pos logit contributions
 Jones+0.348
 aisle+0.334
Fuel+0.321
Ancient+0.311
oky+0.310
Neg logit contributions
n-0.347
 disl-0.326
 spend-0.323
 enjoy-0.311
 stay-0.308
Token.
Token ID13
Feature activation-2.40e+02

Pos logit contributions
 needed+0.211
 whispered+0.207
 meant+0.194
 Jones+0.181
 said+0.178
Neg logit contributions
n-0.421
 disl-0.420
pieces-0.418
iu-0.408
 Majesty-0.406
Token Let
Token ID3914
Feature activation-2.40e+02

Pos logit contributions
 needed+0.236
 White+0.233
 whispered+0.227
 meant+0.222
 Jones+0.221
Neg logit contributions
aches-0.420
pieces-0.407
ered-0.392
n-0.390
 arrest-0.388
Token's
Token ID338
Feature activation-2.35e+02

Pos logit contributions
 said+0.182
 meant+0.172
 whispered+0.170
 needed+0.170
 dark+0.131
Neg logit contributions
ric-0.528
 disl-0.517
pieces-0.502
iu-0.502
 Majesty-0.498
Token play
Token ID711
Feature activation-2.39e+02

Pos logit contributions
 said+0.174
 went+0.141
 had+0.141
 made+0.136
 wide+0.135
Neg logit contributions
earned-0.560
ocol-0.553
pieces-0.503
beh-0.503
iu-0.498
Token again
Token ID757
Feature activation-2.40e+02

Pos logit contributions
 whispered+0.262
 needed+0.253
 said+0.249
 meant+0.236
 dark+0.232
Neg logit contributions
ric-0.401
iu-0.393
 disl-0.391
earned-0.386
n-0.379
Token."
Token ID526
Feature activation-2.51e+02

Pos logit contributions
 needed+0.269
 whispered+0.265
 meant+0.251
 Jones+0.237
 said+0.234
Neg logit contributions
n-0.392
 disl-0.391
pieces-0.388
iu-0.377
 Majesty-0.375
Token They
Token ID1119
Feature activation-2.48e+02

Pos logit contributions
 needed+0.278
 meant+0.270
 whispered+0.269
 Jones+0.239
chy+0.237
Neg logit contributions
 disl-0.408
n-0.408
 interests-0.400
pieces-0.393
iu-0.386
Token take
Token ID1011
Feature activation-2.52e+02

Pos logit contributions
 whispered+0.368
 said+0.361
 needed+0.356
chy+0.329
 smelled+0.310
Neg logit contributions
 disl-0.371
 occasions-0.354
earned-0.349
 Majesty-0.344
iu-0.341
Token their
Token ID511
Feature activation-2.48e+02

Pos logit contributions
 said+0.307
 needed+0.305
 whispered+0.300
 dark+0.298
 meant+0.291
Neg logit contributions
 disl-0.436
ric-0.418
 Majesty-0.407
n-0.372
 witness-0.368
Token boats
Token ID16659
Feature activation-2.47e+02

Pos logit contributions
 needed+0.301
 whispered+0.286
 said+0.268
 meant+0.268
 dark+0.262
Neg logit contributions
 Majesty-0.403
 disl-0.390
n-0.383
ric-0.377
pieces-0.375
Token and
Token ID290
Feature activation-2.56e+02

Pos logit contributions
 whispered+0.161
 needed+0.161
 said+0.140
 meant+0.138
 Jones+0.123
Neg logit contributions
 Majesty-0.526
 disl-0.524
n-0.518
pieces-0.511
iu-0.502
Token,
Token ID11
Feature activation-2.36e+02

Pos logit contributions
 whispered+0.217
 needed+0.210
 Jones+0.210
 said+0.207
 meant+0.199
Neg logit contributions
 disl-0.429
pieces-0.419
ric-0.408
n-0.404
 interests-0.401
Token duck
Token ID22045
Feature activation-2.26e+02

Pos logit contributions
 whispered+0.277
 said+0.270
 meant+0.269
 needed+0.265
 dark+0.235
Neg logit contributions
 disl-0.386
pieces-0.385
ric-0.370
 occasions-0.369
earned-0.369
Token!"
Token ID2474
Feature activation-2.45e+02

Pos logit contributions
 needed+0.356
 whispered+0.351
 meant+0.341
 Jones+0.327
 faded+0.317
Neg logit contributions
n-0.267
 disl-0.261
pieces-0.255
 Majesty-0.246
 interests-0.245
Token Mia
Token ID32189
Feature activation-2.34e+02

Pos logit contributions
 needed+0.298
 whispered+0.294
 meant+0.280
 said+0.265
 Jones+0.260
Neg logit contributions
n-0.383
 disl-0.378
pieces-0.373
 Majesty-0.365
 interests-0.363
Token says
Token ID1139
Feature activation-2.43e+02

Pos logit contributions
 needed+0.215
 whispered+0.193
 meant+0.190
 Jones+0.177
 faded+0.170
Neg logit contributions
n-0.433
 disl-0.425
pieces-0.422
 interests-0.412
 Majesty-0.409
Token.
Token ID13
Feature activation-2.51e+02

Pos logit contributions
 needed+0.208
 whispered+0.204
 meant+0.190
 Jones+0.177
 said+0.167
Neg logit contributions
n-0.461
 disl-0.459
pieces-0.452
 Majesty-0.442
 interests-0.441
Token
Token ID198
Feature activation-2.49e+02

Pos logit contributions
 needed+0.380
 meant+0.369
chy+0.360
 whispered+0.351
 faded+0.330
Neg logit contributions
n-0.298
 arrest-0.293
ext-0.283
 blast-0.282
 join-0.280
Token
Token ID198
Feature activation-2.50e+02

Pos logit contributions
 needed+0.208
 whispered+0.204
 meant+0.190
 Jones+0.175
 said+0.169
Neg logit contributions
n-0.475
 disl-0.472
pieces-0.470
 Majesty-0.457
iu-0.455
TokenThe
Token ID464
Feature activation-2.51e+02

Pos logit contributions
 whispered+0.302
 needed+0.290
 said+0.276
 meant+0.272
 Jones+0.260
Neg logit contributions
pieces-0.389
n-0.387
 disl-0.382
 Majesty-0.376
iu-0.370
Token duck
Token ID22045
Feature activation-2.42e+02

Pos logit contributions
 needed+0.366
 whispered+0.360
 meant+0.348
 Jones+0.334
 said+0.321
Neg logit contributions
n-0.328
 disl-0.322
pieces-0.317
 interests-0.306
 Majesty-0.306
Token smiles
Token ID21845
Feature activation-2.53e+02

Pos logit contributions
 needed+0.323
 meant+0.312
 whispered+0.309
 Jones+0.309
 White+0.281
Neg logit contributions
n-0.339
 disl-0.331
 Majesty-0.323
pieces-0.318
 interests-0.317
Token
Token ID129
Feature activation+5.76e+02

Pos logit contributions
 needed+0.194
 whispered+0.187
 meant+0.179
 Jones+0.165
chy+0.156
Neg logit contributions
n-0.411
pieces-0.407
 disl-0.405
 interests-0.396
iu-0.395
Token
Token ID241
Feature activation+4.76e+02

Pos logit contributions
 interests+0.659
pieces+0.652
 victory+0.601
 occasions+0.582
nels+0.572
Neg logit contributions
 whispered-0.925
 needed-0.890
 said-0.883
 meant-0.881
 Jones-0.831
TokenIt
Token ID1026
Feature activation+3.91e+02

Pos logit contributions
 Majesty+0.849
 disl+0.848
pieces+0.847
n+0.838
 arrest+0.827
Neg logit contributions
 needed-0.454
 whispered-0.428
 meant-0.428
 Jones-0.380
 said-0.375
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.552
 disl+0.533
pieces+0.520
 interests+0.506
 Majesty+0.502
Neg logit contributions
 whispered-0.516
 needed-0.508
 Jones-0.488
 meant-0.476
 faded-0.465
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.439
 disl+0.435
pieces+0.427
 Majesty+0.408
 interests+0.407
Neg logit contributions
 needed-0.758
 whispered-0.747
 meant-0.728
 Jones-0.699
 said-0.682
Token
Token ID8151
Feature activation+5.06e+02

Pos logit contributions
pieces+0.596
 disl+0.585
n+0.573
iu+0.567
 victory+0.566
Neg logit contributions
 needed-0.619
 whispered-0.599
 meant-0.572
 Jones-0.566
 said-0.550
Tokens
Token ID82
Feature activation+4.04e+02

Pos logit contributions
n+0.881
ric+0.828
 disl+0.783
pieces+0.782
ieved+0.765
Neg logit contributions
 whispered-0.549
 Jones-0.514
 faded-0.496
 needed-0.490
 blank-0.480
Token a
Token ID257
Feature activation+3.81e+02

Pos logit contributions
n+0.531
 disl+0.519
pieces+0.505
 interests+0.491
 Majesty+0.489
Neg logit contributions
 needed-0.580
 whispered-0.577
 meant-0.551
 Jones-0.536
 faded-0.511
Token magic
Token ID5536
Feature activation+3.82e+02

Pos logit contributions
n+0.557
 disl+0.553
pieces+0.544
 victory+0.534
 Majesty+0.534
Neg logit contributions
 needed-0.471
 whispered-0.471
 meant-0.456
 Jones-0.430
 faded-0.405
Token book
Token ID1492
Feature activation+3.67e+02

Pos logit contributions
n+0.523
 disl+0.506
 Majesty+0.497
pieces+0.494
 interests+0.491
Neg logit contributions
 needed-0.509
 whispered-0.508
 meant-0.487
 Jones-0.468
 bored-0.450
Token!
Token ID0
Feature activation+3.00e+02

Pos logit contributions
n+0.469
pieces+0.386
 victory+0.381
 occasions+0.378
 disl+0.371
Neg logit contributions
 Jones-0.513
 whispered-0.512
 needed-0.498
Fuel-0.494
 bored-0.490
Token
Token ID129
Feature activation+5.37e+02

Pos logit contributions
pieces+0.954
 disl+0.915
nels+0.911
latable+0.901
 interests+0.896
Neg logit contributions
 needed-0.668
 whispered-0.652
 said-0.621
 Jones-0.595
 meant-0.584
Token
Token ID241
Feature activation+4.65e+02

Pos logit contributions
pieces+0.575
 interests+0.555
n+0.549
 disl+0.542
 victory+0.525
Neg logit contributions
 whispered-0.880
 needed-0.878
 meant-0.849
 Jones-0.817
 said-0.811
TokenThat
Token ID2504
Feature activation+3.82e+02

Pos logit contributions
n+0.785
 disl+0.783
pieces+0.780
 Majesty+0.764
iu+0.756
Neg logit contributions
 needed-0.485
 whispered-0.466
 meant-0.453
 Jones-0.417
 said-0.404
Tokenâ
Token ID22940
Feature activation+4.36e+02

Pos logit contributions
n+0.557
 disl+0.550
pieces+0.544
iu+0.526
 Majesty+0.525
Neg logit contributions
 needed-0.495
 whispered-0.484
 meant-0.469
 Jones-0.441
 said-0.428
Token
Token ID26391
Feature activation+4.43e+02

Pos logit contributions
n+0.391
 disl+0.379
pieces+0.368
 interests+0.347
iu+0.346
Neg logit contributions
 needed-0.813
 whispered-0.804
 meant-0.780
 Jones-0.757
 said-0.735
Token
Token ID8151
Feature activation+5.05e+02

Pos logit contributions
pieces+0.585
 disl+0.579
n+0.577
 interests+0.562
iu+0.560
Neg logit contributions
 needed-0.624
 whispered-0.603
 meant-0.582
 Jones-0.568
 said-0.550
Tokens
Token ID82
Feature activation+4.03e+02

Pos logit contributions
n+0.826
ric+0.768
 disl+0.761
pieces+0.741
ieved+0.711
Neg logit contributions
 whispered-0.614
 Jones-0.564
 faded-0.543
 needed-0.537
 White-0.532
Token right
Token ID826
Feature activation+3.82e+02

Pos logit contributions
pieces+0.536
 disl+0.527
n+0.525
iu+0.523
 arrest+0.519
Neg logit contributions
 needed-0.579
 meant-0.554
 whispered-0.550
 said-0.506
 Jones-0.501
Token,
Token ID11
Feature activation+3.32e+02

Pos logit contributions
n+0.558
 disl+0.546
 Majesty+0.519
 interests+0.516
pieces+0.514
Neg logit contributions
 needed-0.490
 whispered-0.486
 meant-0.454
 Jones-0.438
 faded-0.426
Token Emma
Token ID18966
Feature activation+3.64e+02

Pos logit contributions
n+0.436
 disl+0.431
pieces+0.426
iu+0.411
 interests+0.411
Neg logit contributions
 needed-0.475
 whispered-0.466
 meant-0.452
 Jones-0.428
 said-0.419
Token.
Token ID13
Feature activation+2.98e+02

Pos logit contributions
n+0.505
 disl+0.472
 Majesty+0.464
 spend+0.436
 interests+0.434
Neg logit contributions
 whispered-0.448
 needed-0.441
 faded-0.440
 dark-0.436
Fuel-0.429
Token his
Token ID465
Feature activation+4.53e+02

Pos logit contributions
 needed+0.337
 whispered+0.337
 meant+0.324
 Jones+0.310
 said+0.303
Neg logit contributions
 disl-0.266
n-0.261
pieces-0.259
iu-0.253
latable-0.248
Token head
Token ID1182
Feature activation+3.98e+02

Pos logit contributions
n+0.618
 disl+0.610
pieces+0.601
iu+0.577
 interests+0.576
Neg logit contributions
 needed-0.629
 whispered-0.617
 meant-0.594
 Jones-0.567
 said-0.550
Token out
Token ID503
Feature activation+3.62e+02

Pos logit contributions
n+0.552
 spend+0.491
 interests+0.491
 pause+0.484
pieces+0.483
Neg logit contributions
Ancient-0.477
 Jones-0.472
 whispered-0.472
Fuel-0.468
 needed-0.467
Token of
Token ID286
Feature activation+3.48e+02

Pos logit contributions
n+0.456
pieces+0.450
 victory+0.450
 interests+0.444
 arrest+0.441
Neg logit contributions
 needed-0.481
 whispered-0.480
 meant-0.468
 Jones-0.461
 White-0.448
Token his
Token ID465
Feature activation+3.63e+02

Pos logit contributions
n+0.513
 disl+0.507
pieces+0.503
 Majesty+0.484
iu+0.483
Neg logit contributions
 needed-0.448
 whispered-0.441
 meant-0.423
 Jones-0.402
 said-0.392
Token bur
Token ID4356
Feature activation+4.65e+02

Pos logit contributions
n+0.492
 disl+0.485
pieces+0.480
 interests+0.465
 Majesty+0.464
Neg logit contributions
 needed-0.501
 whispered-0.493
 meant-0.478
 Jones-0.454
 said-0.436
Tokenrow
Token ID808
Feature activation+3.72e+02

Pos logit contributions
 disl+0.958
 blast+0.920
pieces+0.917
 victory+0.897
 hey+0.888
Neg logit contributions
 needed-0.421
 whispered-0.408
 meant-0.367
 Jones-0.335
chy-0.328
Token and
Token ID290
Feature activation+3.09e+02

Pos logit contributions
n+0.582
 interests+0.526
pieces+0.524
 Majesty+0.523
 disl+0.517
Neg logit contributions
 whispered-0.421
 needed-0.409
 meant-0.400
 Jones-0.374
 White-0.370
Token saw
Token ID2497
Feature activation+3.06e+02

Pos logit contributions
n+0.520
 disl+0.511
pieces+0.499
 Majesty+0.495
 interests+0.492
Neg logit contributions
 needed-0.311
 whispered-0.302
 meant-0.298
 Jones-0.283
 White-0.263
Token something
Token ID1223
Feature activation+3.03e+02

Pos logit contributions
n+0.373
 disl+0.363
pieces+0.353
 Majesty+0.342
 victory+0.341
Neg logit contributions
 needed-0.460
 whispered-0.454
 meant-0.435
 Jones-0.430
 faded-0.412
Token new
Token ID649
Feature activation+3.06e+02

Pos logit contributions
n+0.378
 victory+0.366
 disl+0.338
pieces+0.336
 exercise+0.331
Neg logit contributions
 Jones-0.453
 whispered-0.416
 White-0.415
 needed-0.394
Factory-0.390
Token of
Token ID286
Feature activation+4.07e+02

Pos logit contributions
 needed+0.356
 whispered+0.354
 meant+0.338
chy+0.334
 faded+0.326
Neg logit contributions
n-0.273
 victory-0.270
 arrest-0.269
 disl-0.261
 pause-0.260
Token other
Token ID584
Feature activation+3.99e+02

Pos logit contributions
n+0.582
 disl+0.575
pieces+0.570
iu+0.557
nels+0.548
Neg logit contributions
 needed-0.546
 whispered-0.525
 meant-0.510
 said-0.474
 Jones-0.474
Token people
Token ID661
Feature activation+3.48e+02

Pos logit contributions
pieces+0.467
n+0.466
 disl+0.456
 interests+0.452
 arrest+0.452
Neg logit contributions
 whispered-0.606
 needed-0.591
 meant-0.578
 Jones-0.563
 faded-0.543
Tokenâ
Token ID22940
Feature activation+3.98e+02

Pos logit contributions
n+0.539
pieces+0.521
 disl+0.520
 interests+0.507
 arrest+0.497
Neg logit contributions
 whispered-0.391
 needed-0.381
 meant-0.377
 Jones-0.376
 White-0.360
Token
Token ID26391
Feature activation+4.08e+02

Pos logit contributions
n+0.465
 disl+0.438
pieces+0.424
iu+0.409
 interests+0.404
Neg logit contributions
 needed-0.649
 whispered-0.642
 meant-0.615
 Jones-0.597
 faded-0.587
Token
Token ID8151
Feature activation+4.70e+02

Pos logit contributions
 disl+0.652
n+0.650
pieces+0.649
 interests+0.627
iu+0.626
Neg logit contributions
 needed-0.463
 whispered-0.450
 meant-0.425
 Jones-0.414
 said-0.397
Tokens
Token ID82
Feature activation+3.83e+02

Pos logit contributions
n+0.802
ric+0.735
pieces+0.722
 disl+0.722
ieved+0.697
Neg logit contributions
 whispered-0.511
 needed-0.476
 Jones-0.472
 meant-0.465
 faded-0.463
Token property
Token ID3119
Feature activation+4.12e+02

Pos logit contributions
n+0.544
 disl+0.533
pieces+0.516
 Majesty+0.515
 interests+0.509
Neg logit contributions
 whispered-0.501
 needed-0.493
 meant-0.474
 Jones-0.457
 faded-0.439
Token.
Token ID13
Feature activation+3.09e+02

Pos logit contributions
n+0.632
 disl+0.599
pieces+0.581
 interests+0.574
 victory+0.565
Neg logit contributions
 whispered-0.490
 needed-0.474
 Jones-0.462
 faded-0.432
 White-0.428
Token We
Token ID775
Feature activation+3.09e+02

Pos logit contributions
n+0.485
 disl+0.466
pieces+0.449
 Majesty+0.439
ric+0.437
Neg logit contributions
 needed-0.380
 whispered-0.375
 meant-0.358
 faded-0.339
chy-0.337
Token should
Token ID815
Feature activation+2.92e+02

Pos logit contributions
n+0.385
 disl+0.382
pieces+0.375
iu+0.362
 Majesty+0.362
Neg logit contributions
 needed-0.467
 whispered-0.457
 meant-0.444
 Jones-0.421
 said-0.412
Token
Token ID241
Feature activation+4.22e+02

Pos logit contributions
n+0.495
 disl+0.481
pieces+0.454
 Majesty+0.442
iu+0.438
Neg logit contributions
 needed-0.790
 whispered-0.771
 meant-0.753
 Jones-0.723
 faded-0.704
TokenI
Token ID40
Feature activation+3.74e+02

Pos logit contributions
pieces+0.731
 disl+0.728
n+0.721
 Majesty+0.721
 arrest+0.710
Neg logit contributions
 needed-0.423
 whispered-0.400
 meant-0.394
 Jones-0.357
 said-0.350
Token don
Token ID836
Feature activation+3.72e+02

Pos logit contributions
 disl+0.442
n+0.441
pieces+0.440
iu+0.424
 Majesty+0.420
Neg logit contributions
 needed-0.588
 whispered-0.578
 meant-0.562
 Jones-0.528
 said-0.525
Tokenâ
Token ID22940
Feature activation+4.05e+02

Pos logit contributions
 Majesty+0.484
 disl+0.479
 interests+0.465
pieces+0.464
iu+0.454
Neg logit contributions
 needed-0.554
 meant-0.541
 whispered-0.537
 said-0.499
 Jones-0.492
Token
Token ID26391
Feature activation+4.10e+02

Pos logit contributions
n+0.336
 disl+0.320
pieces+0.310
iu+0.290
 interests+0.289
Neg logit contributions
 needed-0.790
 whispered-0.778
 meant-0.757
 Jones-0.736
 faded-0.717
Token
Token ID8151
Feature activation+4.71e+02

Pos logit contributions
pieces+0.507
 disl+0.505
n+0.487
iu+0.485
 interests+0.484
Neg logit contributions
 needed-0.616
 whispered-0.601
 Jones-0.576
 meant-0.574
 said-0.559
Tokent
Token ID83
Feature activation+3.73e+02

Pos logit contributions
n+0.749
ric+0.676
pieces+0.674
 disl+0.657
ieved+0.654
Neg logit contributions
 whispered-0.578
 needed-0.529
 faded-0.517
 Jones-0.513
 White-0.497
Token think
Token ID892
Feature activation+3.55e+02

Pos logit contributions
 disl+0.416
pieces+0.414
n+0.414
 Majesty+0.402
iu+0.400
Neg logit contributions
 needed-0.607
 whispered-0.593
 meant-0.585
 Jones-0.546
 said-0.543
Token so
Token ID523
Feature activation+3.32e+02

Pos logit contributions
n+0.445
 disl+0.428
pieces+0.424
 interests+0.406
 Majesty+0.401
Neg logit contributions
 needed-0.536
 whispered-0.528
 meant-0.507
 Jones-0.493
 faded-0.473
Token.
Token ID13
Feature activation+2.94e+02

Pos logit contributions
pieces+0.429
 disl+0.428
n+0.426
iu+0.421
 interests+0.410
Neg logit contributions
 needed-0.487
 whispered-0.468
 meant-0.460
 said-0.432
 Jones-0.427
Token It
Token ID632
Feature activation+3.08e+02

Pos logit contributions
n+0.428
 disl+0.418
pieces+0.414
 Majesty+0.400
 interests+0.399
Neg logit contributions
 needed-0.382
 whispered-0.375
 meant-0.360
 Jones-0.337
 faded-0.331
Token put
Token ID1234
Feature activation+2.11e+02

Pos logit contributions
latable+0.331
iu+0.327
ric+0.323
ext+0.319
pieces+0.309
Neg logit contributions
 said-0.326
 needed-0.311
 whispered-0.305
 meant-0.272
 sleepy-0.260
Token some
Token ID617
Feature activation+2.14e+02

Pos logit contributions
 disl+0.235
n+0.225
pieces+0.222
 Majesty+0.218
 interests+0.217
Neg logit contributions
 needed-0.349
 whispered-0.345
 meant-0.335
 Jones-0.322
 said-0.321
Token band
Token ID4097
Feature activation+2.27e+02

Pos logit contributions
 disl+0.294
n+0.293
pieces+0.286
 Majesty+0.282
iu+0.279
Neg logit contributions
 needed-0.298
 whispered-0.291
 meant-0.281
 said-0.264
 Jones-0.258
Token-
Token ID12
Feature activation+2.68e+02

Pos logit contributions
latable+0.380
 hey+0.370
 Majesty+0.369
pieces+0.364
 victory+0.363
Neg logit contributions
 needed-0.251
 whispered-0.244
 meant-0.227
 said-0.223
chy-0.197
Tokenaid
Token ID1698
Feature activation+2.39e+02

Pos logit contributions
latable+0.392
 Majesty+0.392
 blast+0.374
 disl+0.372
nels+0.364
Neg logit contributions
 needed-0.354
 whispered-0.340
 meant-0.327
 said-0.317
 spoke-0.305
Tokens
Token ID82
Feature activation+2.19e+02

Pos logit contributions
n+0.443
pieces+0.422
 disl+0.417
 interests+0.414
iu+0.406
Neg logit contributions
 needed-0.214
 whispered-0.210
 meant-0.194
 Jones-0.191
 faded-0.181
Token on
Token ID319
Feature activation+2.09e+02

Pos logit contributions
n+0.324
 interests+0.299
 disl+0.293
 arrest+0.286
 victory+0.285
Neg logit contributions
 whispered-0.275
 needed-0.272
 Jones-0.267
 meant-0.266
 spoke-0.253
Token them
Token ID606
Feature activation+2.23e+02

Pos logit contributions
n+0.188
 arrest+0.172
pieces+0.169
 disl+0.169
 our+0.163
Neg logit contributions
 needed-0.400
 whispered-0.392
 meant-0.380
 spoke-0.367
 faded-0.362
Token.
Token ID13
Feature activation+1.82e+02

Pos logit contributions
n+0.278
 our+0.234
 interests+0.234
pieces+0.224
 stay+0.216
Neg logit contributions
 needed-0.344
 spoke-0.334
 Jones-0.333
 faded-0.333
 whispered-0.328
Token Then
Token ID3244
Feature activation+1.94e+02

Pos logit contributions
n+0.298
pieces+0.296
 disl+0.293
iu+0.288
 arrest+0.287
Neg logit contributions
 whispered-0.195
 needed-0.193
 Jones-0.188
 meant-0.186
 White-0.179
Token you
Token ID345
Feature activation+1.99e+02

Pos logit contributions
 disl+0.267
pieces+0.264
latable+0.259
ric+0.256
iu+0.253
Neg logit contributions
 needed-0.275
 whispered-0.271
 meant-0.258
 said-0.257
 Jones-0.250
Token ang
Token ID3550
Feature activation+3.31e+02

Pos logit contributions
n+0.502
 disl+0.501
pieces+0.496
 interests+0.477
 Majesty+0.474
Neg logit contributions
 needed-0.422
 whispered-0.412
 meant-0.399
 Jones-0.397
 White-0.367
Tokenrier
Token ID5277
Feature activation+4.02e+02

Pos logit contributions
 disl+0.442
n+0.434
pieces+0.428
 interests+0.422
 Majesty+0.410
Neg logit contributions
 needed-0.480
 whispered-0.467
 meant-0.460
 Jones-0.440
 said-0.426
Token and
Token ID290
Feature activation+2.83e+02

Pos logit contributions
n+0.632
 disl+0.608
 interests+0.580
iu+0.576
 hey+0.571
Neg logit contributions
 whispered-0.444
 needed-0.441
 White-0.409
 Jones-0.399
 dark-0.398
Token Lily
Token ID20037
Feature activation+3.07e+02

Pos logit contributions
n+0.531
 disl+0.525
pieces+0.511
 hey+0.501
iu+0.497
Neg logit contributions
 needed-0.235
 whispered-0.228
 Jones-0.215
 meant-0.215
 White-0.189
Token felt
Token ID2936
Feature activation+2.82e+02

Pos logit contributions
n+0.551
 hey+0.472
pieces+0.471
 spend+0.468
ieved+0.467
Neg logit contributions
 Jones-0.289
 dark-0.285
Ancient-0.282
 blank-0.278
 aisle-0.277
Token even
Token ID772
Feature activation+2.92e+02

Pos logit contributions
n+0.521
 disl+0.516
pieces+0.509
 interests+0.495
 Majesty+0.495
Neg logit contributions
 needed-0.253
 whispered-0.247
 meant-0.233
 Jones-0.219
 said-0.203
Token sad
Token ID6507
Feature activation+2.84e+02

Pos logit contributions
n+0.547
 disl+0.538
 arrest+0.535
pieces+0.532
 Majesty+0.529
Neg logit contributions
 needed-0.278
 whispered-0.272
 meant-0.256
 said-0.238
 spoke-0.218
Tokender
Token ID1082
Feature activation+3.91e+02

Pos logit contributions
 disl+0.439
n+0.438
 interests+0.418
pieces+0.416
 victory+0.411
Neg logit contributions
 whispered-0.328
 needed-0.323
 Jones-0.301
 meant-0.298
 spoke-0.287
Token.
Token ID13
Feature activation+2.48e+02

Pos logit contributions
 disl+0.606
n+0.590
 interests+0.550
pieces+0.547
 arrest+0.525
Neg logit contributions
 Jones-0.451
 whispered-0.441
 White-0.428
 needed-0.422
 aisle-0.398
Token From
Token ID3574
Feature activation+2.51e+02

Pos logit contributions
 disl+0.370
n+0.369
 Majesty+0.325
ric+0.323
 interests+0.317
Neg logit contributions
 faded-0.317
 whispered-0.315
 meant-0.307
 needed-0.305
 shone-0.297
Token that
Token ID326
Feature activation+2.51e+02

Pos logit contributions
n+0.251
 disl+0.246
pieces+0.228
nels+0.222
iu+0.221
Neg logit contributions
 needed-0.439
 whispered-0.427
 meant-0.420
 Jones-0.413
 faded-0.397
Token.
Token ID13
Feature activation+3.01e+02

Pos logit contributions
n+0.587
 disl+0.586
pieces+0.571
iu+0.558
 interests+0.553
Neg logit contributions
 needed-0.426
 whispered-0.420
 meant-0.392
 Jones-0.378
 said-0.369
Token It
Token ID632
Feature activation+3.14e+02

Pos logit contributions
n+0.484
 disl+0.466
pieces+0.446
 Majesty+0.443
 interests+0.438
Neg logit contributions
 needed-0.344
 meant-0.333
 whispered-0.328
chy-0.326
 faded-0.314
Token was
Token ID373
Feature activation+2.91e+02

Pos logit contributions
n+0.519
 disl+0.495
 blast+0.471
pieces+0.470
 interests+0.465
Neg logit contributions
 Jones-0.337
 needed-0.315
 whispered-0.311
 meant-0.290
 White-0.287
Token her
Token ID607
Feature activation+3.00e+02

Pos logit contributions
n+0.357
 victory+0.350
 blast+0.336
 disl+0.332
 our+0.330
Neg logit contributions
 needed-0.415
 whispered-0.407
 Jones-0.405
Fuel-0.378
 aisle-0.377
Token friend
Token ID1545
Feature activation+2.96e+02

Pos logit contributions
n+0.353
pieces+0.343
 Majesty+0.340
 disl+0.339
 victory+0.338
Neg logit contributions
 needed-0.462
 whispered-0.457
 meant-0.447
 Jones-0.424
 spoke-0.416
Token,
Token ID11
Feature activation+2.64e+02

Pos logit contributions
n+0.348
pieces+0.329
 spend+0.306
 our+0.301
 join+0.301
Neg logit contributions
Fuel-0.431
 needed-0.421
Gas-0.415
Ancient-0.415
Dead-0.413
Token Tommy
Token ID19919
Feature activation+2.93e+02

Pos logit contributions
n+0.328
pieces+0.309
 victory+0.302
 interests+0.301
 our+0.299
Neg logit contributions
 needed-0.375
 whispered-0.353
 meant-0.343
 faded-0.336
chy-0.330
Token.
Token ID13
Feature activation+2.43e+02

Pos logit contributions
n+0.387
 disl+0.335
 our+0.330
 spend+0.321
 arrest+0.316
Neg logit contributions
 faded-0.402
Ancient-0.390
Dead-0.384
 dark-0.382
 needed-0.382
Token "
Token ID366
Feature activation+2.52e+02

Pos logit contributions
n+0.390
 disl+0.389
 interests+0.366
pieces+0.366
 Majesty+0.362
Neg logit contributions
 needed-0.279
 whispered-0.267
 meant-0.254
chy-0.245
 spoke-0.234
TokenHi
Token ID17250
Feature activation+2.58e+02

Pos logit contributions
n+0.443
 disl+0.440
pieces+0.438
 Majesty+0.429
iu+0.424
Neg logit contributions
 needed-0.246
 whispered-0.243
 meant-0.231
 Jones-0.215
 said-0.204
Token Lily
Token ID20037
Feature activation+2.63e+02

Pos logit contributions
pieces+0.448
 disl+0.447
ric+0.437
n+0.436
iu+0.432
Neg logit contributions
 needed-0.268
 whispered-0.266
 meant-0.259
 said-0.248
 Jones-0.246
Token in
Token ID287
Feature activation+3.53e+02

Pos logit contributions
n+0.675
 arrest+0.642
 disl+0.638
pieces+0.638
 interests+0.629
Neg logit contributions
 whispered-0.404
 needed-0.398
 Jones-0.395
 meant-0.388
 aisle-0.378
Token the
Token ID262
Feature activation+3.70e+02

Pos logit contributions
n+0.522
 disl+0.515
pieces+0.509
 Majesty+0.491
 interests+0.491
Neg logit contributions
 needed-0.451
 whispered-0.442
 meant-0.426
 Jones-0.404
 said-0.389
Token circle
Token ID9197
Feature activation+3.54e+02

Pos logit contributions
n+0.559
 disl+0.554
pieces+0.550
 Majesty+0.538
 interests+0.531
Neg logit contributions
 needed-0.449
 whispered-0.446
 meant-0.431
 Jones-0.402
 said-0.386
Token.
Token ID13
Feature activation+2.70e+02

Pos logit contributions
n+0.543
pieces+0.525
 interests+0.518
 disl+0.501
 Majesty+0.495
Neg logit contributions
 needed-0.404
 meant-0.397
 whispered-0.396
 Jones-0.394
 White-0.380
Token
Token ID198
Feature activation+2.74e+02

Pos logit contributions
n+0.423
 disl+0.417
pieces+0.410
 Majesty+0.392
 interests+0.381
Neg logit contributions
 needed-0.312
 meant-0.302
 whispered-0.301
chy-0.291
 faded-0.291
Token
Token ID198
Feature activation+2.71e+02

Pos logit contributions
n+0.479
 disl+0.463
pieces+0.458
 interests+0.449
 Majesty+0.439
Neg logit contributions
 needed-0.283
 whispered-0.273
 meant-0.266
 Jones-0.246
 faded-0.239
TokenThe
Token ID464
Feature activation+2.87e+02

Pos logit contributions
n+0.393
 disl+0.386
pieces+0.380
 interests+0.367
 Majesty+0.365
Neg logit contributions
 needed-0.359
 whispered-0.350
 meant-0.341
 Jones-0.323
 faded-0.309
Token mouse
Token ID10211
Feature activation+3.02e+02

Pos logit contributions
 Majesty+0.443
 victory+0.443
pieces+0.436
n+0.426
 disl+0.423
Neg logit contributions
 whispered-0.334
 meant-0.329
 needed-0.321
 spoke-0.295
 Jones-0.290
Token stopped
Token ID5025
Feature activation+2.69e+02

Pos logit contributions
n+0.482
 Majesty+0.441
 disl+0.436
 blast+0.434
ered+0.433
Neg logit contributions
 Jones-0.341
Fuel-0.310
 meant-0.301
 aisle-0.301
 White-0.301
Token and
Token ID290
Feature activation+2.49e+02

Pos logit contributions
 disl+0.372
ric+0.362
n+0.360
 victory+0.349
 arrest+0.347
Neg logit contributions
 needed-0.364
 whispered-0.344
 meant-0.338
 aisle-0.336
 faded-0.313
Token the
Token ID262
Feature activation+2.64e+02

Pos logit contributions
n+0.400
 disl+0.393
 hey+0.382
 arrest+0.377
 spend+0.374
Neg logit contributions
 Jones-0.241
 meant-0.236
 aisle-0.229
 needed-0.226
elly-0.225
Tokenummy
Token ID13513
Feature activation+2.85e+02

Pos logit contributions
n+0.475
pieces+0.471
 disl+0.471
 Majesty+0.456
 interests+0.451
Neg logit contributions
 needed-0.319
 whispered-0.303
 meant-0.298
 Jones-0.285
 said-0.274
Token food
Token ID2057
Feature activation+2.75e+02

Pos logit contributions
n+0.356
 disl+0.352
pieces+0.346
 Majesty+0.335
 interests+0.332
Neg logit contributions
 needed-0.429
 whispered-0.421
 meant-0.409
 Jones-0.389
 said-0.381
Token.
Token ID13
Feature activation+2.29e+02

Pos logit contributions
n+0.347
 occasions+0.307
 exercise+0.307
 victory+0.306
 our+0.302
Neg logit contributions
 needed-0.393
 whispered-0.389
 faded-0.386
 spoke-0.370
 White-0.364
Token One
Token ID1881
Feature activation+2.33e+02

Pos logit contributions
n+0.347
 disl+0.337
pieces+0.335
 interests+0.325
 Majesty+0.323
Neg logit contributions
 needed-0.286
 whispered-0.283
 meant-0.273
 faded-0.253
 Jones-0.251
Token day
Token ID1110
Feature activation+2.41e+02

Pos logit contributions
n+0.284
 exercise+0.259
earned+0.258
 pause+0.247
 spend+0.244
Neg logit contributions
 whispered-0.360
 Jones-0.349
 faded-0.326
 spoke-0.322
 White-0.321
Token,
Token ID11
Feature activation+2.23e+02

Pos logit contributions
n+0.359
 disl+0.342
pieces+0.337
 arrest+0.330
 victory+0.328
Neg logit contributions
 needed-0.305
 whispered-0.302
 meant-0.294
 Jones-0.275
 faded-0.265
Token her
Token ID607
Feature activation+2.40e+02

Pos logit contributions
 disl+0.460
n+0.457
pieces+0.453
 Majesty+0.449
 interests+0.446
Neg logit contributions
 needed-0.161
 whispered-0.155
 meant-0.141
 said-0.128
 Jones-0.128
Token mom
Token ID1995
Feature activation+2.39e+02

Pos logit contributions
n+0.230
 Majesty+0.229
 interests+0.227
 victory+0.221
 arrest+0.217
Neg logit contributions
 whispered-0.400
 needed-0.385
 meant-0.383
 spoke-0.373
 faded-0.368
Token made
Token ID925
Feature activation+2.18e+02

Pos logit contributions
n+0.381
 spend+0.346
 pause+0.317
 interests+0.314
 hey+0.307
Neg logit contributions
 faded-0.246
Ancient-0.245
 blank-0.244
 Jones-0.240
 crooked-0.240
Token some
Token ID617
Feature activation+2.22e+02

Pos logit contributions
n+0.290
pieces+0.264
 disl+0.261
iu+0.248
 victory+0.248
Neg logit contributions
 needed-0.326
 whispered-0.318
 faded-0.300
 Jones-0.298
 meant-0.298
Token toast
Token ID27805
Feature activation+2.43e+02

Pos logit contributions
 victory+0.238
 exercise+0.236
pieces+0.236
n+0.227
 disl+0.221
Neg logit contributions
 whispered-0.349
 Jones-0.346
 needed-0.343
 faded-0.338
 spoke-0.327
Token end
Token ID886
Feature activation+1.55e+02

Pos logit contributions
n+0.176
 disl+0.170
pieces+0.168
iu+0.163
 Majesty+0.163
Neg logit contributions
 whispered-0.243
 needed-0.241
 meant-0.237
 Jones-0.229
 faded-0.215
Token.
Token ID13
Feature activation+1.28e+02

Pos logit contributions
n+0.306
 disl+0.304
pieces+0.303
 Majesty+0.300
 arrest+0.296
Neg logit contributions
 needed-0.124
 whispered-0.119
 meant-0.115
 Jones-0.102
 said-0.099
Token<|endoftext|>
Token ID50256
Feature activation+1.29e+02

Pos logit contributions
n+0.200
 disl+0.191
pieces+0.186
 blast+0.184
 interests+0.183
Neg logit contributions
 whispered-0.153
 needed-0.151
 meant-0.143
chy-0.135
 spoke-0.134
TokenOnce
Token ID7454
Feature activation+1.27e+02

Pos logit contributions
n+0.185
 blast+0.165
 arrest+0.153
 victory+0.152
ric+0.149
Neg logit contributions
 whispered-0.179
 needed-0.174
 faded-0.164
 meant-0.163
chy-0.163
Token upon
Token ID2402
Feature activation+1.56e+02

Pos logit contributions
n+0.193
 disl+0.184
pieces+0.184
 interests+0.183
 victory+0.178
Neg logit contributions
 whispered-0.151
 needed-0.147
 meant-0.144
 Jones-0.137
 faded-0.131
Token a
Token ID257
Feature activation+1.36e+02

Pos logit contributions
 Majesty+0.338
pieces+0.335
nels+0.328
 disl+0.328
han+0.325
Neg logit contributions
 needed-0.120
 dark-0.115
 whispered-0.113
 meant-0.108
 said-0.105
Token time
Token ID640
Feature activation+1.35e+02

Pos logit contributions
pieces+0.237
 disl+0.232
 Majesty+0.222
ric+0.219
nels+0.215
Neg logit contributions
 needed-0.158
 whispered-0.148
 dark-0.144
 said-0.142
chy-0.139
Token,
Token ID11
Feature activation+1.22e+02

Pos logit contributions
pieces+0.271
 disl+0.262
 Majesty+0.257
n+0.254
iu+0.252
Neg logit contributions
 needed-0.113
 whispered-0.104
 meant-0.100
 said-0.096
 Jones-0.095
Token there
Token ID612
Feature activation+1.27e+02

Pos logit contributions
n+0.197
 disl+0.195
pieces+0.193
 Majesty+0.187
 interests+0.186
Neg logit contributions
 needed-0.140
 whispered-0.136
 meant-0.131
 Jones-0.123
 said-0.118
Token was
Token ID373
Feature activation+1.23e+02

Pos logit contributions
n+0.200
 disl+0.180
 blast+0.177
 victory+0.177
 interests+0.176
Neg logit contributions
 whispered-0.154
 needed-0.142
 Jones-0.140
 meant-0.132
 faded-0.130
Token a
Token ID257
Feature activation+1.23e+02

Pos logit contributions
 disl+0.210
n+0.206
 Majesty+0.205
pieces+0.205
nels+0.200
Neg logit contributions
 needed-0.135
 whispered-0.129
 meant-0.127
 dark-0.120
 said-0.117
Token brought
Token ID3181
Feature activation+1.73e+02

Pos logit contributions
n+0.252
 spend+0.217
 interests+0.201
 our+0.193
 pause+0.191
Neg logit contributions
 faded-0.253
 Jones-0.251
 wicked-0.249
Ancient-0.248
 crooked-0.247
Token home
Token ID1363
Feature activation+1.79e+02

Pos logit contributions
n+0.178
 our+0.171
 victory+0.147
 your+0.145
 occasions+0.143
Neg logit contributions
 faded-0.295
 needed-0.292
 whispered-0.289
chy-0.286
 spoke-0.281
Token a
Token ID257
Feature activation+1.67e+02

Pos logit contributions
n+0.244
 disl+0.239
pieces+0.236
 interests+0.231
iu+0.226
Neg logit contributions
 needed-0.245
 whispered-0.241
 meant-0.233
 Jones-0.223
 faded-0.215
Token newspaper
Token ID7533
Feature activation+1.84e+02

Pos logit contributions
 victory+0.211
n+0.204
 blast+0.203
 pause+0.201
pieces+0.201
Neg logit contributions
 needed-0.235
 meant-0.235
 whispered-0.233
 Jones-0.216
 faded-0.209
Token.
Token ID13
Feature activation+1.38e+02

Pos logit contributions
n+0.196
 victory+0.176
 spend+0.175
han+0.174
pieces+0.171
Neg logit contributions
 meant-0.275
 whispered-0.275
 needed-0.273
 White-0.263
 faded-0.260
Token Lily
Token ID20037
Feature activation+1.64e+02

Pos logit contributions
n+0.194
 disl+0.185
 blast+0.178
pieces+0.178
 spend+0.175
Neg logit contributions
 whispered-0.182
 needed-0.181
 meant-0.175
 Jones-0.167
 faded-0.165
Token saw
Token ID2497
Feature activation+1.46e+02

Pos logit contributions
n+0.261
 disl+0.233
 interests+0.227
 arrest+0.225
 spend+0.222
Neg logit contributions
 Jones-0.197
 faded-0.179
lessness-0.177
 White-0.173
 dark-0.172
Token it
Token ID340
Feature activation+1.50e+02

Pos logit contributions
n+0.137
 disl+0.137
 how+0.132
 arrest+0.131
 our+0.128
Neg logit contributions
 needed-0.249
 Jones-0.244
 whispered-0.243
 faded-0.243
chy-0.243
Token and
Token ID290
Feature activation+1.32e+02

Pos logit contributions
n+0.135
 pause+0.121
 how+0.120
 go+0.114
 exercise+0.111
Neg logit contributions
 Jones-0.310
lessness-0.308
irk-0.303
 Did-0.274
Gas-0.270
Token said
Token ID531
Feature activation+1.33e+02

Pos logit contributions
n+0.240
 disl+0.235
 spend+0.219
 Majesty+0.218
 how+0.216
Neg logit contributions
 Jones-0.118
chy-0.113
oky-0.101
 meant-0.100
 faded-0.099
Token,
Token ID11
Feature activation+1.23e+02

Pos logit contributions
n+0.209
 disl+0.207
pieces+0.202
 Majesty+0.197
 interests+0.196
Neg logit contributions
 needed-0.156
 whispered-0.154
 meant-0.148
 Jones-0.139
 faded-0.132
Token Tom
Token ID4186
Feature activation+2.62e+01

Pos logit contributions
n+0.004
 disl+0.004
 interests+0.004
 our+0.004
 blast+0.004
Neg logit contributions
 needed-0.004
 whispered-0.004
chy-0.004
 shone-0.004
 meant-0.004
Token dreamed
Token ID27947
Feature activation+8.22e+00

Pos logit contributions
n+0.031
 arrest+0.028
 blast+0.025
 interests+0.023
 pause+0.023
Neg logit contributions
Pr-0.039
Ancient-0.038
Gas-0.038
 pods-0.037
 Jones-0.037
Token that
Token ID326
Feature activation+3.49e+00

Pos logit contributions
n+0.005
 victory+0.005
 hey+0.004
pieces+0.004
 how+0.004
Neg logit contributions
 Jones-0.017
 whispered-0.017
Fuel-0.016
 White-0.016
 needed-0.016
Token a
Token ID257
Feature activation+5.99e+00

Pos logit contributions
 our+0.004
 exercise+0.004
 victory+0.004
 how+0.003
n+0.003
Neg logit contributions
opy-0.006
oky-0.005
Ancient-0.005
Factory-0.005
 whispered-0.005
Token big
Token ID1263
Feature activation+1.34e+01

Pos logit contributions
 victory+0.008
 blast+0.008
 Majesty+0.007
 disl+0.007
n+0.007
Neg logit contributions
 whispered-0.009
 meant-0.009
 needed-0.008
 Jones-0.008
Fuel-0.008
Token monster
Token ID9234
Feature activation+2.03e+01

Pos logit contributions
 victory+0.018
 blast+0.018
 disl+0.017
 interests+0.017
n+0.017
Neg logit contributions
 whispered-0.018
 needed-0.018
 meant-0.018
 Jones-0.017
 spoke-0.016
Token took
Token ID1718
Feature activation-8.30e-01

Pos logit contributions
n+0.031
 disl+0.029
 arrest+0.028
 blast+0.028
 victory+0.027
Neg logit contributions
 Jones-0.024
 whispered-0.023
Fuel-0.022
Ancient-0.022
 White-0.022
Token all
Token ID477
Feature activation+1.43e+00

Pos logit contributions
 needed+0.002
 whispered+0.001
 meant+0.001
 Jones+0.001
 spoke+0.001
Neg logit contributions
n-0.001
 victory-0.001
 our-0.001
pieces-0.001
 blast-0.001
Token his
Token ID465
Feature activation+2.77e+00

Pos logit contributions
n+0.002
 disl+0.002
pieces+0.002
 interests+0.002
 arrest+0.002
Neg logit contributions
 needed-0.002
 whispered-0.002
 meant-0.002
 Jones-0.002
 faded-0.002
Token toys
Token ID14958
Feature activation+4.95e+00

Pos logit contributions
 interests+0.003
n+0.003
pieces+0.003
 victory+0.003
 disl+0.003
Neg logit contributions
 whispered-0.004
 Jones-0.004
 needed-0.004
 meant-0.004
 faded-0.004
Token and
Token ID290
Feature activation-1.08e+01

Pos logit contributions
 victory+0.006
n+0.006
 our+0.005
 arrest+0.005
 stay+0.005
Neg logit contributions
 White-0.007
Ancient-0.007
Pr-0.007
 whispered-0.007
Dead-0.007
Token.
Token ID13
Feature activation+1.58e+02

Pos logit contributions
n+0.377
 disl+0.372
pieces+0.366
 interests+0.360
iu+0.358
Neg logit contributions
 needed-0.187
 whispered-0.183
 meant-0.173
 Jones-0.161
 faded-0.152
Token Tim
Token ID5045
Feature activation+1.83e+02

Pos logit contributions
n+0.132
 disl+0.131
latable+0.128
ric+0.117
 spend+0.115
Neg logit contributions
 whispered-0.313
 needed-0.295
 faded-0.289
 spoke-0.284
 meant-0.280
Tokenmy
Token ID1820
Feature activation+1.94e+02

Pos logit contributions
n+0.226
 interests+0.211
 disl+0.207
ric+0.207
latable+0.187
Neg logit contributions
 whispered-0.266
 faded-0.263
 Jones-0.257
 needed-0.252
 spoke-0.250
Token loved
Token ID6151
Feature activation+1.63e+02

Pos logit contributions
n+0.289
 disl+0.269
 interests+0.263
 blast+0.248
ric+0.248
Neg logit contributions
 whispered-0.236
 Jones-0.227
 faded-0.225
 meant-0.218
 needed-0.212
Token planes
Token ID13016
Feature activation+1.79e+02

Pos logit contributions
n+0.236
pieces+0.236
 disl+0.230
 Majesty+0.226
iu+0.223
Neg logit contributions
 needed-0.215
 whispered-0.208
 meant-0.201
 Jones-0.190
 said-0.188
Token and
Token ID290
Feature activation+1.47e+02

Pos logit contributions
 disl+0.330
n+0.329
pieces+0.324
 Majesty+0.320
iu+0.317
Neg logit contributions
 needed-0.167
 whispered-0.161
 meant-0.155
 Jones-0.140
 said-0.137
Token always
Token ID1464
Feature activation+1.51e+02

Pos logit contributions
 hey+0.313
ieved+0.312
 disl+0.309
pieces+0.306
iu+0.303
Neg logit contributions
 needed-0.129
 whispered-0.108
 meant-0.104
 said-0.101
 spoke-0.085
Token dreamed
Token ID27947
Feature activation+1.43e+02

Pos logit contributions
n+0.271
 disl+0.264
 spend+0.260
 interests+0.260
 blast+0.250
Neg logit contributions
 Jones-0.122
 whispered-0.121
 aisle-0.119
 needed-0.118
 meant-0.114
Token of
Token ID286
Feature activation+1.48e+02

Pos logit contributions
n+0.144
 victory+0.140
pieces+0.135
 disl+0.133
 interests+0.132
Neg logit contributions
 whispered-0.249
 Jones-0.243
 needed-0.243
 meant-0.238
 White-0.223
Token flying
Token ID7348
Feature activation+1.47e+02

Pos logit contributions
n+0.199
 disl+0.197
pieces+0.192
 interests+0.187
 victory+0.187
Neg logit contributions
 needed-0.208
 whispered-0.204
 meant-0.196
 Jones-0.189
 faded-0.181
Token one
Token ID530
Feature activation+1.38e+02

Pos logit contributions
n+0.208
 disl+0.203
pieces+0.200
 interests+0.198
 victory+0.198
Neg logit contributions
 needed-0.191
 whispered-0.191
 meant-0.181
 Jones-0.178
 faded-0.166
Token slowly
Token ID6364
Feature activation+9.50e+00

Pos logit contributions
n+0.024
pieces+0.022
 victory+0.022
iu+0.021
ap+0.021
Neg logit contributions
 whispered-0.028
 needed-0.027
 meant-0.027
 Jones-0.027
 spoke-0.026
Token.
Token ID13
Feature activation-1.03e+01

Pos logit contributions
n+0.013
 victory+0.012
pieces+0.011
 join+0.011
 blast+0.011
Neg logit contributions
 Jones-0.013
 whispered-0.013
 needed-0.013
 meant-0.012
 spoke-0.012
Token Tim
Token ID5045
Feature activation+1.09e+01

Pos logit contributions
 whispered+0.018
 faded+0.018
 needed+0.018
chy+0.017
 meant+0.017
Neg logit contributions
 disl-0.011
n-0.010
 arrest-0.009
 hey-0.008
 yeah-0.008
Token learned
Token ID4499
Feature activation-1.00e+01

Pos logit contributions
n+0.017
 interests+0.014
 blast+0.014
 arrest+0.014
 disl+0.014
Neg logit contributions
 Jones-0.014
 dark-0.013
 meant-0.012
 White-0.012
 bored-0.012
Token that
Token ID326
Feature activation-1.07e+01

Pos logit contributions
 whispered+0.010
 meant+0.010
 said+0.010
 needed+0.010
 dark+0.009
Neg logit contributions
pieces-0.019
 Majesty-0.018
 blast-0.018
iu-0.018
ieved-0.017
Token asking
Token ID4737
Feature activation-1.33e+01

Pos logit contributions
 whispered+0.017
 needed+0.016
 Jones+0.015
 meant+0.015
 faded+0.015
Neg logit contributions
n-0.013
 disl-0.012
 victory-0.012
 interests-0.012
iu-0.011
Token for
Token ID329
Feature activation-1.52e+01

Pos logit contributions
 whispered+0.022
 faded+0.022
 needed+0.021
 spoke+0.021
 aisle+0.021
Neg logit contributions
 interests-0.014
n-0.014
 our-0.014
 respect-0.014
 victory-0.014
Token help
Token ID1037
Feature activation-1.82e+01

Pos logit contributions
 needed+0.024
 whispered+0.023
 meant+0.022
 said+0.021
 Jones+0.021
Neg logit contributions
pieces-0.019
n-0.018
 disl-0.018
iu-0.018
 Majesty-0.017
Token is
Token ID318
Feature activation-2.12e+01

Pos logit contributions
 needed+0.021
 whispered+0.020
 meant+0.019
 Jones+0.018
 faded+0.017
Neg logit contributions
n-0.030
 disl-0.029
pieces-0.029
 interests-0.028
 Majesty-0.028
Token good
Token ID922
Feature activation-1.85e+01

Pos logit contributions
 whispered+0.033
 needed+0.033
 Jones+0.032
 meant+0.032
 faded+0.030
Neg logit contributions
n-0.025
 disl-0.024
 interests-0.023
 victory-0.023
pieces-0.022
Token.
Token ID13
Feature activation-3.72e+01

Pos logit contributions
 needed+0.024
 meant+0.024
 whispered+0.023
 said+0.021
 Jones+0.021
Neg logit contributions
pieces-0.029
n-0.028
 disl-0.028
 Majesty-0.027
ric-0.027
Token even
Token ID772
Feature activation-1.02e+02

Pos logit contributions
 needed+0.090
 whispered+0.086
 said+0.084
 sleepy+0.079
 sleeping+0.079
Neg logit contributions
earned-0.201
pieces-0.193
 disl-0.190
han-0.179
ieved-0.177
Token though
Token ID996
Feature activation-1.06e+02

Pos logit contributions
 needed+0.108
 whispered+0.106
 meant+0.103
 bored+0.098
 dark+0.097
Neg logit contributions
ric-0.189
n-0.187
 Majesty-0.186
pieces-0.185
 arrest-0.185
Token it
Token ID340
Feature activation-1.06e+02

Pos logit contributions
 needed+0.163
 meant+0.150
 whispered+0.147
 dark+0.141
 said+0.140
Neg logit contributions
 disl-0.140
pieces-0.138
iu-0.136
n-0.135
 Majesty-0.134
Token r
Token ID374
Feature activation-1.04e+02

Pos logit contributions
 Jones+0.122
 whispered+0.120
 needed+0.112
 meant+0.107
Fuel+0.106
Neg logit contributions
n-0.170
 disl-0.155
 interests-0.150
 arrest-0.148
 blast-0.147
Tokenained
Token ID1328
Feature activation-1.08e+02

Pos logit contributions
 needed+0.092
 meant+0.087
 whispered+0.085
 Jones+0.079
 bored+0.073
Neg logit contributions
pieces-0.196
 disl-0.194
n-0.191
 arrest-0.187
 Majesty-0.186
Token.
Token ID13
Feature activation-1.26e+02

Pos logit contributions
 whispered+0.157
 needed+0.156
 meant+0.149
 Jones+0.143
 spoke+0.138
Neg logit contributions
n-0.142
 disl-0.136
pieces-0.134
 victory-0.130
 interests-0.130
Token I
Token ID314
Feature activation-1.20e+02

Pos logit contributions
 White+0.124
 needed+0.121
 meant+0.116
 Jones+0.113
 T+0.107
Neg logit contributions
aches-0.242
pieces-0.237
 disl-0.233
ered-0.229
immers-0.228
Token learned
Token ID4499
Feature activation-1.30e+02

Pos logit contributions
 needed+0.182
 whispered+0.177
 meant+0.172
 said+0.161
 Jones+0.161
Neg logit contributions
pieces-0.151
n-0.150
 disl-0.149
iu-0.144
 Majesty-0.144
Token that
Token ID326
Feature activation-1.30e+02

Pos logit contributions
 meant+0.150
 said+0.143
 whispered+0.142
 needed+0.140
 dark+0.133
Neg logit contributions
pieces-0.235
iu-0.229
 Majesty-0.225
 arrest-0.218
ocol-0.217
Token sometimes
Token ID3360
Feature activation-1.30e+02

Pos logit contributions
 needed+0.178
 whispered+0.176
 meant+0.168
 Jones+0.162
 said+0.155
Neg logit contributions
n-0.181
 disl-0.177
pieces-0.174
 interests-0.169
 Majesty-0.168
Token we
Token ID356
Feature activation-1.29e+02

Pos logit contributions
 needed+0.184
 meant+0.174
 whispered+0.172
 dark+0.159
 said+0.158
Neg logit contributions
pieces-0.184
 disl-0.183
n-0.180
iu-0.179
ric-0.175
Token surprised
Token ID6655
Feature activation-2.67e+01

Pos logit contributions
 Jones+0.044
 aisle+0.043
 needed+0.041
 pods+0.040
 whispered+0.039
Neg logit contributions
 disl-0.037
n-0.035
 victory-0.034
ieved-0.032
 our-0.032
Token that
Token ID326
Feature activation-3.42e+01

Pos logit contributions
 needed+0.029
 whispered+0.028
 meant+0.028
 said+0.028
 Jones+0.025
Neg logit contributions
pieces-0.047
 Majesty-0.045
 disl-0.045
ric-0.044
 blast-0.044
Token he
Token ID339
Feature activation-2.94e+01

Pos logit contributions
 whispered+0.034
 needed+0.034
 meant+0.032
 Jones+0.030
chy+0.029
Neg logit contributions
n-0.059
 disl-0.057
 victory-0.057
pieces-0.056
 Majesty-0.055
Token stopped
Token ID5025
Feature activation-3.56e+01

Pos logit contributions
 Jones+0.028
 needed+0.028
 whispered+0.027
 meant+0.027
 aisle+0.026
Neg logit contributions
n-0.051
 disl-0.048
pieces-0.047
 Majesty-0.047
 arrest-0.046
Token roaring
Token ID43170
Feature activation-2.98e+01

Pos logit contributions
 needed+0.043
 whispered+0.042
 meant+0.040
 Jones+0.038
 said+0.036
Neg logit contributions
n-0.056
 disl-0.055
pieces-0.054
 Majesty-0.053
iu-0.053
Token and
Token ID290
Feature activation-4.72e+01

Pos logit contributions
 needed+0.046
Fuel+0.043
 whispered+0.042
Ancient+0.041
 meant+0.040
Neg logit contributions
 victory-0.036
n-0.034
 pause-0.031
pieces-0.031
 how-0.031
Token his
Token ID465
Feature activation-3.87e+01

Pos logit contributions
 needed+0.044
 whispered+0.043
 meant+0.042
 Jones+0.039
 faded+0.036
Neg logit contributions
n-0.085
 disl-0.083
pieces-0.081
 Majesty-0.080
 arrest-0.080
Token anger
Token ID8993
Feature activation-3.78e+01

Pos logit contributions
 whispered+0.058
 meant+0.057
 needed+0.057
 Jones+0.053
 spoke+0.051
Neg logit contributions
 victory-0.049
n-0.047
 Majesty-0.047
 disl-0.047
 interests-0.046
Token suddenly
Token ID6451
Feature activation-4.48e+01

Pos logit contributions
 Jones+0.047
 aisle+0.044
Ancient+0.043
Fuel+0.043
 whispered+0.043
Neg logit contributions
n-0.057
 victory-0.053
 pause-0.052
 blast-0.052
 arrest-0.051
Token vanished
Token ID23717
Feature activation-4.55e+01

Pos logit contributions
 aisle+0.045
Fuel+0.043
 Jones+0.043
Ancient+0.043
 White+0.041
Neg logit contributions
 Majesty-0.071
n-0.070
 disl-0.070
 arrest-0.069
 victory-0.068
Token.
Token ID13
Feature activation-6.68e+01

Pos logit contributions
 whispered+0.062
 needed+0.062
 meant+0.058
 Jones+0.055
 said+0.054
Neg logit contributions
n-0.060
 victory-0.059
 disl-0.058
pieces-0.058
 Majesty-0.058
Token for
Token ID329
Feature activation-1.71e+02

Pos logit contributions
 needed+0.197
 whispered+0.193
 meant+0.186
 Jones+0.176
 said+0.167
Neg logit contributions
n-0.257
 disl-0.254
pieces-0.252
 interests-0.245
 Majesty-0.243
Token animals
Token ID4695
Feature activation-1.67e+02

Pos logit contributions
 whispered+0.207
 needed+0.201
 meant+0.199
 dark+0.197
 said+0.190
Neg logit contributions
n-0.274
iu-0.258
ieved-0.255
 disl-0.255
 Majesty-0.255
Token and
Token ID290
Feature activation-1.77e+02

Pos logit contributions
 needed+0.193
 whispered+0.181
 meant+0.176
 faded+0.171
 Jones+0.171
Neg logit contributions
n-0.265
pieces-0.255
 interests-0.248
 victory-0.248
 disl-0.247
Token bugs
Token ID11316
Feature activation-1.69e+02

Pos logit contributions
 needed+0.223
 whispered+0.208
 said+0.201
 dark+0.194
 meant+0.190
Neg logit contributions
ieved-0.337
han-0.310
iu-0.310
ext-0.300
immers-0.296
Token.
Token ID13
Feature activation-1.87e+02

Pos logit contributions
 needed+0.194
 meant+0.181
 whispered+0.179
 Jones+0.175
 White+0.164
Neg logit contributions
n-0.275
pieces-0.263
 disl-0.255
 arrest-0.253
 interests-0.249
Token They
Token ID1119
Feature activation-1.82e+02

Pos logit contributions
 needed+0.165
 whispered+0.157
 meant+0.150
 Jones+0.138
 said+0.133
Neg logit contributions
n-0.351
 disl-0.350
pieces-0.346
 Majesty-0.341
 arrest-0.339
Token had
Token ID550
Feature activation-1.86e+02

Pos logit contributions
elly+0.254
oky+0.250
 Jones+0.235
 late+0.231
 wicked+0.225
Neg logit contributions
 spend-0.255
n-0.231
 pause-0.225
 explore-0.219
 stay-0.217
Token a
Token ID257
Feature activation-1.87e+02

Pos logit contributions
 needed+0.247
 whispered+0.243
 meant+0.234
 Jones+0.224
 said+0.213
Neg logit contributions
n-0.266
 disl-0.261
pieces-0.257
 interests-0.250
 Majesty-0.248
Token big
Token ID1263
Feature activation-1.80e+02

Pos logit contributions
 needed+0.240
 whispered+0.236
 meant+0.229
 Jones+0.216
 said+0.207
Neg logit contributions
n-0.272
 disl-0.268
pieces-0.264
 interests-0.258
 Majesty-0.258
Token book
Token ID1492
Feature activation-1.80e+02

Pos logit contributions
 needed+0.235
 whispered+0.229
 meant+0.220
 Jones+0.210
 said+0.203
Neg logit contributions
n-0.265
 disl-0.262
pieces-0.258
iu-0.249
 interests-0.246
Token with
Token ID351
Feature activation-1.90e+02

Pos logit contributions
 needed+0.306
 whispered+0.297
 Jones+0.290
 meant+0.288
 White+0.284
Neg logit contributions
n-0.179
 interests-0.167
 victory-0.165
pieces-0.165
 blast-0.162
Token friend
Token ID1545
Feature activation-1.92e+01

Pos logit contributions
 needed+0.026
 whispered+0.025
 meant+0.024
 Jones+0.023
 said+0.023
Neg logit contributions
n-0.022
pieces-0.022
 disl-0.022
iu-0.021
 Majesty-0.020
Token.
Token ID13
Feature activation-3.95e+01

Pos logit contributions
 needed+0.024
 whispered+0.023
 meant+0.022
 Jones+0.022
 faded+0.021
Neg logit contributions
n-0.029
pieces-0.028
 interests-0.027
 arrest-0.026
 disl-0.026
Token<|endoftext|>
Token ID50256
Feature activation-4.04e+01

Pos logit contributions
 needed+0.039
 whispered+0.038
 meant+0.036
 Jones+0.034
 faded+0.032
Neg logit contributions
n-0.069
 disl-0.069
pieces-0.067
 interests-0.066
 arrest-0.065
TokenL
Token ID43
Feature activation-1.82e+01

Pos logit contributions
 whispered+0.041
 needed+0.041
 meant+0.038
 faded+0.036
chy+0.035
Neg logit contributions
n-0.070
 blast-0.065
 victory-0.063
 arrest-0.063
ric-0.063
Tokenily
Token ID813
Feature activation-2.40e+01

Pos logit contributions
 needed+0.020
 whispered+0.020
 meant+0.019
 said+0.018
 Jones+0.017
Neg logit contributions
 disl-0.032
n-0.030
 Majesty-0.030
pieces-0.030
 interests-0.029
Token liked
Token ID8288
Feature activation-4.02e+01

Pos logit contributions
Pr+0.030
lessness+0.030
 Jones+0.029
Fuel+0.029
 White+0.029
Neg logit contributions
 arrest-0.032
n-0.031
 disl-0.030
 interests-0.029
 blast-0.028
Token to
Token ID284
Feature activation-4.15e+01

Pos logit contributions
 needed+0.056
 whispered+0.054
 meant+0.053
 Jones+0.050
 said+0.049
Neg logit contributions
n-0.055
 disl-0.054
pieces-0.054
 Majesty-0.052
 interests-0.052
Token play
Token ID711
Feature activation-4.23e+01

Pos logit contributions
 needed+0.063
 whispered+0.062
 said+0.059
 meant+0.057
 spoke+0.056
Neg logit contributions
ieved-0.053
 hey-0.052
iu-0.052
 Majesty-0.051
ric-0.050
Token with
Token ID351
Feature activation-4.66e+01

Pos logit contributions
 whispered+0.047
 needed+0.045
 said+0.043
 meant+0.042
 Jones+0.040
Neg logit contributions
 disl-0.070
 Majesty-0.069
ric-0.068
n-0.068
iu-0.068
Token clay
Token ID21558
Feature activation-3.49e+01

Pos logit contributions
 needed+0.059
 whispered+0.057
 meant+0.055
 Jones+0.052
 faded+0.050
Neg logit contributions
 disl-0.070
n-0.070
pieces-0.068
 arrest-0.067
 interests-0.066
Token.
Token ID13
Feature activation-6.15e+01

Pos logit contributions
 whispered+0.042
 needed+0.041
 meant+0.039
 Jones+0.038
 faded+0.038
Neg logit contributions
pieces-0.050
n-0.050
 disl-0.049
 victory-0.049
 arrest-0.049
Token
Token ID198
Feature activation-1.14e+02

Pos logit contributions
 needed+0.137
 whispered+0.131
 meant+0.131
 faded+0.125
 Jones+0.120
Neg logit contributions
n-0.187
 blast-0.163
 interests-0.163
 disl-0.163
pieces-0.162
TokenOne
Token ID3198
Feature activation-1.15e+02

Pos logit contributions
 needed+0.201
 whispered+0.198
 meant+0.195
 Jones+0.191
 faded+0.182
Neg logit contributions
n-0.125
 disl-0.112
pieces-0.110
 blast-0.108
 arrest-0.108
Token little
Token ID1310
Feature activation-1.06e+02

Pos logit contributions
 whispered+0.149
 needed+0.143
 Jones+0.142
 meant+0.138
 spoke+0.129
Neg logit contributions
n-0.167
 victory-0.163
 blast-0.159
 interests-0.156
pieces-0.156
Token girl
Token ID2576
Feature activation-1.02e+02

Pos logit contributions
 needed+0.109
 whispered+0.108
 meant+0.103
chy+0.097
 Jones+0.096
Neg logit contributions
n-0.174
pieces-0.174
 victory-0.172
 disl-0.172
 interests-0.171
Token,
Token ID11
Feature activation-1.20e+02

Pos logit contributions
chy+0.120
 Jones+0.117
 blank+0.115
 meant+0.114
 faded+0.112
Neg logit contributions
n-0.155
pieces-0.139
 arrest-0.138
 spend-0.138
 blast-0.137
Token named
Token ID3706
Feature activation-1.16e+02

Pos logit contributions
chy+0.226
 faded+0.208
 needed+0.201
 meant+0.200
 crooked+0.197
Neg logit contributions
n-0.112
 hey-0.109
 join-0.106
 how-0.105
 our-0.102
Token Linda
Token ID21776
Feature activation-1.08e+02

Pos logit contributions
 needed+0.119
 meant+0.116
 whispered+0.111
 said+0.111
 Jones+0.110
Neg logit contributions
iu-0.201
 interests-0.200
 blast-0.200
ric-0.198
earned-0.197
Token,
Token ID11
Feature activation-1.25e+02

Pos logit contributions
 meant+0.141
 needed+0.137
 faded+0.137
 dark+0.136
 whispered+0.135
Neg logit contributions
n-0.155
pieces-0.136
 disl-0.136
 arrest-0.136
 spend-0.132
Token came
Token ID1625
Feature activation-1.24e+02

Pos logit contributions
chy+0.218
 crooked+0.209
 dusty+0.201
 stiff+0.200
oky+0.197
Neg logit contributions
 how-0.141
 join-0.128
 spend-0.127
n-0.125
 hey-0.117
Token up
Token ID510
Feature activation-1.20e+02

Pos logit contributions
 needed+0.166
 whispered+0.162
 meant+0.156
 Jones+0.151
 said+0.144
Neg logit contributions
 disl-0.179
n-0.176
pieces-0.170
iu-0.167
 Majesty-0.166
Token to
Token ID284
Feature activation-1.28e+02

Pos logit contributions
 whispered+0.210
 needed+0.207
 meant+0.203
chy+0.194
 Lover+0.193
Neg logit contributions
 victory-0.119
 join-0.107
ap-0.103
 arrest-0.102
n-0.100