TOP ACTIVATIONS
MAX = 0.804

I â m sorry Jane . I
It â s almost just as beautiful
It â s a magic book !
That â s a special necklace ,
Let â s go find something else
That â s right , Emma .
That â s the gas from the
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
't know the rh in oc eros was wild . Can
Why don â t you come with me
of other people â s property . We should
Where â s my jelly ? â
It â s a surprise ! â
It â s a books he lf

MIN ACTIVATIONS
MIN = -0.286

said goodbye . She walked safely home , feeling happy that
. This time , Rita successfully measured the cushion very carefully
bird flew away quickly and safely . From
always put her notebook away safely when she had finished using
" Help ! Help !" Sam said . He
. Jane nodded , and thus began the game ! Everyone
him how to remove it safely . His
and opened the box . Cheap flew out and circled around
it was time to go home . Tom and Anna invited
how to fold his wings properly so he could fly .
different body and a different life . It can act in
Soon they created a delicious treat . Billy was so proud
glad to get back home safely ! <|endoftext|> Once upon a
you , Ben . You saved me ," Sam says .
back . He had to leave the wheat behind , even
Sam if he understood the number rules . Sam nodded his
the pilot for flying them safely back home and everyone hugged
gift of running and jumping freely . <|endoftext|> Once upon a
They washed the fruits and cut them . They put them
the mineral and play together nicely . The moral

INTERVAL 0.532 - 0.804
CONTAINS 0.257%

garden and admire his sun fl owers . They also see
't know the rh in oc eros was wild . Can
It â s a magic book !
Let â s play a joke
Who â s there ? â

INTERVAL 0.259 - 0.532
CONTAINS 9.722%

Liam and asked him to help her think of a creative
He dropped the dolls and tried to run away . But
." Tom my played with Mitt ens for
had never seen one like it before . But Turtle was
The bus driver smiled back as he opened the door

INTERVAL -0.013 - 0.259
CONTAINS 36.095%

The bug felt better and said thank you .
He decided to make a special pie . So he got
hive . She waved her hands and the bees started to
were amazed and they all stopped to watch the bird for
." The mouse followed Benny as he led the

INTERVAL -0.286 - -0.013
CONTAINS 53.925%

to play with Tim my 's toy cars instead . They
toy . Finally , they found it ! The little girl
ears felt deaf . B enny decided to
bee 's answer . He thanked her and went back home
His mum said , " No , not today , Tom
Token
Token ID129
Feature activation+7.19e-01

Pos logit contributions
 faded+0.226
 Jones+0.223
 blank+0.215
 dark+0.211
 White+0.209
Neg logit contributions
n-0.341
 hey-0.320
 pause-0.319
 spend-0.309
ap-0.305
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenI
Token ID40
Feature activation+5.26e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Tokenâ
Token ID22940
Feature activation+5.39e-01

Pos logit contributions
 disl+0.515
n+0.503
 hey+0.496
pieces+0.491
ric+0.475
Neg logit contributions
 whispered-0.755
 needed-0.746
 meant-0.720
 faded-0.675
 said-0.667
Token
Token ID26391
Feature activation+5.81e-01

Pos logit contributions
 disl+0.463
pieces+0.449
 hey+0.441
 arrest+0.429
 Majesty+0.428
Neg logit contributions
 whispered-0.848
 needed-0.837
 meant-0.823
 said-0.770
 spoke-0.758
Token
Token ID8151
Feature activation+6.08e-01

Pos logit contributions
 hey+0.692
pieces+0.691
 disl+0.683
latable+0.663
 victory+0.654
Neg logit contributions
 needed-0.719
 whispered-0.716
 said-0.661
 meant-0.661
 spoke-0.624
Tokenm
Token ID76
Feature activation+4.59e-01

Pos logit contributions
n+0.873
ric+0.860
 disl+0.815
pieces+0.805
ieved+0.796
Neg logit contributions
 whispered-0.612
 faded-0.552
 needed-0.548
 meant-0.517
 dark-0.511
Token sorry
Token ID7926
Feature activation+4.58e-01

Pos logit contributions
 disl+0.586
n+0.579
ric+0.573
 hey+0.566
pieces+0.556
Neg logit contributions
 whispered-0.529
 needed-0.519
 meant-0.491
 faded-0.469
 spoke-0.459
Token Jane
Token ID12091
Feature activation+4.34e-01

Pos logit contributions
 disl+0.652
pieces+0.637
 hey+0.637
n+0.636
ric+0.620
Neg logit contributions
 whispered-0.462
 needed-0.457
 meant-0.435
 said-0.392
 faded-0.388
Token.
Token ID13
Feature activation+4.02e-01

Pos logit contributions
n+0.581
 spend+0.537
 disl+0.531
pieces+0.518
 hey+0.516
Neg logit contributions
 whispered-0.422
 blank-0.410
 dark-0.403
 aisle-0.395
Fuel-0.395
Token I
Token ID314
Feature activation+3.72e-01

Pos logit contributions
 disl+0.459
n+0.447
 hey+0.440
pieces+0.435
ric+0.428
Neg logit contributions
 whispered-0.521
 needed-0.517
 meant-0.496
 faded-0.465
 spoke-0.456
Token
Token ID129
Feature activation+7.19e-01

Pos logit contributions
 whispered+0.293
 needed+0.291
 meant+0.282
 faded+0.262
 spoke+0.259
Neg logit contributions
 disl-0.212
n-0.212
pieces-0.203
 hey-0.202
ric-0.200
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenIt
Token ID1026
Feature activation+5.06e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Tokenâ
Token ID22940
Feature activation+5.35e-01

Pos logit contributions
n+0.607
 disl+0.606
pieces+0.579
ric+0.569
 hey+0.565
Neg logit contributions
 whispered-0.608
 needed-0.575
 faded-0.546
 meant-0.543
 spoke-0.524
Token
Token ID26391
Feature activation+5.77e-01

Pos logit contributions
 disl+0.458
pieces+0.440
 hey+0.439
n+0.431
ric+0.417
Neg logit contributions
 whispered-0.841
 needed-0.836
 meant-0.811
 said-0.760
 faded-0.752
Token
Token ID8151
Feature activation+6.05e-01

Pos logit contributions
 hey+0.654
pieces+0.650
 disl+0.648
iu+0.618
latable+0.617
Neg logit contributions
 needed-0.740
 whispered-0.736
 meant-0.692
 said-0.667
 spoke-0.643
Tokens
Token ID82
Feature activation+4.66e-01

Pos logit contributions
n+0.895
ric+0.885
 disl+0.826
pieces+0.815
ieved+0.796
Neg logit contributions
 whispered-0.605
 faded-0.537
 needed-0.525
 blank-0.511
 meant-0.500
Token almost
Token ID2048
Feature activation+4.26e-01

Pos logit contributions
 disl+0.526
n+0.517
 hey+0.501
pieces+0.500
ric+0.499
Neg logit contributions
 whispered-0.608
 needed-0.592
 meant-0.568
 faded-0.534
 spoke-0.531
Token just
Token ID655
Feature activation+4.26e-01

Pos logit contributions
 disl+0.463
n+0.454
 spend+0.431
 hey+0.427
 blast+0.422
Neg logit contributions
 whispered-0.555
 needed-0.543
 meant-0.527
 Jones-0.494
 spoke-0.490
Token as
Token ID355
Feature activation+4.12e-01

Pos logit contributions
 disl+0.468
n+0.462
 hey+0.445
pieces+0.441
 spend+0.438
Neg logit contributions
 whispered-0.550
 needed-0.538
 meant-0.521
 faded-0.485
 spoke-0.483
Token beautiful
Token ID4950
Feature activation+4.01e-01

Pos logit contributions
n+0.487
 disl+0.475
 victory+0.460
 hey+0.456
 spend+0.453
Neg logit contributions
 whispered-0.486
 needed-0.454
 meant-0.452
 spoke-0.429
 faded-0.423
Token
Token ID129
Feature activation+7.19e-01

Pos logit contributions
 needed+0.193
 whispered+0.193
 meant+0.183
chy+0.162
 faded+0.159
Neg logit contributions
 hey-0.375
pieces-0.369
 disl-0.367
n-0.362
iu-0.357
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenIt
Token ID1026
Feature activation+5.06e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Tokenâ
Token ID22940
Feature activation+5.35e-01

Pos logit contributions
n+0.607
 disl+0.606
pieces+0.579
ric+0.569
 hey+0.565
Neg logit contributions
 whispered-0.608
 needed-0.575
 faded-0.546
 meant-0.543
 spoke-0.524
Token
Token ID26391
Feature activation+5.77e-01

Pos logit contributions
 disl+0.458
pieces+0.440
 hey+0.439
n+0.431
ric+0.417
Neg logit contributions
 whispered-0.841
 needed-0.836
 meant-0.811
 said-0.760
 faded-0.752
Token
Token ID8151
Feature activation+6.05e-01

Pos logit contributions
 hey+0.654
pieces+0.650
 disl+0.648
iu+0.618
latable+0.617
Neg logit contributions
 needed-0.740
 whispered-0.736
 meant-0.692
 said-0.667
 spoke-0.643
Tokens
Token ID82
Feature activation+4.66e-01

Pos logit contributions
n+0.895
ric+0.885
 disl+0.826
pieces+0.815
ieved+0.796
Neg logit contributions
 whispered-0.605
 faded-0.537
 needed-0.525
 blank-0.511
 meant-0.500
Token a
Token ID257
Feature activation+4.44e-01

Pos logit contributions
 disl+0.526
n+0.517
 hey+0.501
pieces+0.500
ric+0.499
Neg logit contributions
 whispered-0.608
 needed-0.592
 meant-0.568
 faded-0.534
 spoke-0.531
Token magic
Token ID5536
Feature activation+4.06e-01

Pos logit contributions
 disl+0.572
 blast+0.555
 hey+0.554
pieces+0.554
n+0.552
Neg logit contributions
 whispered-0.496
 needed-0.479
 meant-0.473
 spoke-0.421
 faded-0.419
Token book
Token ID1492
Feature activation+3.91e-01

Pos logit contributions
 disl+0.479
n+0.477
 blast+0.460
pieces+0.460
 Majesty+0.458
Neg logit contributions
 whispered-0.485
 needed-0.470
 meant-0.456
 spoke-0.418
 bored-0.412
Token!
Token ID0
Feature activation+3.88e-01

Pos logit contributions
n+0.420
 join+0.354
pieces+0.353
 blast+0.348
 disl+0.346
Neg logit contributions
 whispered-0.498
 needed-0.471
 bored-0.454
 Jones-0.451
 faded-0.450
Token
Token ID129
Feature activation+7.19e-01

Pos logit contributions
 whispered+0.238
 needed+0.233
 meant+0.219
 said+0.198
 spoke+0.196
Neg logit contributions
 disl-0.355
 hey-0.350
n-0.349
pieces-0.345
ric-0.339
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenThat
Token ID2504
Feature activation+4.92e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Tokenâ
Token ID22940
Feature activation+5.33e-01

Pos logit contributions
 disl+0.619
n+0.602
pieces+0.595
 hey+0.594
ric+0.582
Neg logit contributions
 whispered-0.576
 needed-0.567
 meant-0.538
 faded-0.499
 spoke-0.493
Token
Token ID26391
Feature activation+5.75e-01

Pos logit contributions
 disl+0.395
 hey+0.377
pieces+0.377
n+0.376
ric+0.357
Neg logit contributions
 whispered-0.899
 needed-0.894
 meant-0.867
 said-0.816
 faded-0.814
Token
Token ID8151
Feature activation+6.03e-01

Pos logit contributions
 hey+0.671
pieces+0.665
 disl+0.658
latable+0.632
iu+0.631
Neg logit contributions
 needed-0.723
 whispered-0.710
 meant-0.675
 said-0.648
 spoke-0.620
Tokens
Token ID82
Feature activation+4.65e-01

Pos logit contributions
n+0.843
ric+0.823
 disl+0.794
pieces+0.767
ieved+0.748
Neg logit contributions
 whispered-0.661
 faded-0.590
 needed-0.568
 blank-0.554
 dark-0.553
Token a
Token ID257
Feature activation+4.42e-01

Pos logit contributions
 disl+0.532
pieces+0.528
 hey+0.526
n+0.512
iu+0.511
Neg logit contributions
 needed-0.593
 whispered-0.588
 meant-0.572
 said-0.527
 faded-0.509
Token special
Token ID2041
Feature activation+4.19e-01

Pos logit contributions
 disl+0.572
n+0.555
pieces+0.554
 hey+0.554
 blast+0.543
Neg logit contributions
 whispered-0.497
 needed-0.486
 meant-0.471
 faded-0.425
 spoke-0.423
Token necklace
Token ID38987
Feature activation+4.17e-01

Pos logit contributions
 disl+0.486
n+0.475
pieces+0.465
 hey+0.462
ric+0.453
Neg logit contributions
 whispered-0.527
 needed-0.518
 meant-0.500
 faded-0.460
 spoke-0.457
Token,
Token ID11
Feature activation+3.79e-01

Pos logit contributions
n+0.490
 disl+0.441
pieces+0.435
nels+0.420
 interests+0.419
Neg logit contributions
 whispered-0.489
 needed-0.478
 faded-0.457
 bored-0.447
 dark-0.444
Token
Token ID129
Feature activation+7.19e-01

Pos logit contributions
 whispered+0.401
 meant+0.359
 faded+0.357
 spoke+0.354
Ancient+0.347
Neg logit contributions
n-0.215
 victory-0.200
 spend-0.198
 blast-0.196
 job-0.192
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenLet
Token ID5756
Feature activation+5.04e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Tokenâ
Token ID22940
Feature activation+5.28e-01

Pos logit contributions
ric+0.803
 disl+0.774
iu+0.774
latable+0.773
 hey+0.771
Neg logit contributions
 whispered-0.481
 meant-0.479
 needed-0.474
 said-0.473
 spoke-0.396
Token
Token ID26391
Feature activation+5.71e-01

Pos logit contributions
 disl+0.429
 Majesty+0.428
 hey+0.419
pieces+0.418
 interests+0.408
Neg logit contributions
 whispered-0.829
 needed-0.817
 meant-0.814
 said-0.763
 spoke-0.745
Token
Token ID8151
Feature activation+5.99e-01

Pos logit contributions
 hey+0.777
pieces+0.755
 disl+0.750
latable+0.745
iu+0.734
Neg logit contributions
 needed-0.624
 whispered-0.611
 said-0.582
 meant-0.561
 spoke-0.528
Tokens
Token ID82
Feature activation+4.59e-01

Pos logit contributions
n+0.934
ric+0.896
pieces+0.881
 disl+0.872
 hey+0.847
Neg logit contributions
 whispered-0.542
 needed-0.495
 faded-0.478
 meant-0.472
 dark-0.440
Token go
Token ID467
Feature activation+4.15e-01

Pos logit contributions
iu+0.539
latable+0.527
 disl+0.511
pieces+0.508
 arrest+0.495
Neg logit contributions
 needed-0.623
 whispered-0.615
 said-0.605
 meant-0.604
 spoke-0.526
Token find
Token ID1064
Feature activation+3.97e-01

Pos logit contributions
 disl+0.604
iu+0.582
 hey+0.580
pieces+0.566
ric+0.564
Neg logit contributions
 whispered-0.449
 needed-0.427
 said-0.409
 meant-0.405
 spoke-0.359
Token something
Token ID1223
Feature activation+4.00e-01

Pos logit contributions
 disl+0.475
ric+0.458
pieces+0.458
 hey+0.458
n+0.452
Neg logit contributions
 whispered-0.499
 needed-0.489
 meant-0.474
 said-0.448
 spoke-0.422
Token else
Token ID2073
Feature activation+3.82e-01

Pos logit contributions
 disl+0.451
n+0.451
 victory+0.418
pieces+0.415
 our+0.414
Neg logit contributions
 whispered-0.501
 needed-0.467
 Jones-0.466
 faded-0.455
 White-0.443
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenThat
Token ID2504
Feature activation+4.83e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.22e-01

Pos logit contributions
 disl+0.613
n+0.597
pieces+0.592
 hey+0.591
ric+0.577
Neg logit contributions
 whispered-0.559
 needed-0.550
 meant-0.523
 faded-0.482
 said-0.479
Token
Token ID26391
Feature activation+5.65e-01

Pos logit contributions
 disl+0.381
n+0.365
 hey+0.362
pieces+0.361
ric+0.343
Neg logit contributions
 whispered-0.887
 needed-0.881
 meant-0.854
 faded-0.805
 said-0.804
Token
Token ID8151
Feature activation+5.95e-01

Pos logit contributions
 hey+0.628
 disl+0.624
pieces+0.623
iu+0.594
latable+0.593
Neg logit contributions
 needed-0.735
 whispered-0.728
 meant-0.693
 said-0.657
 spoke-0.639
Tokens
Token ID82
Feature activation+4.57e-01

Pos logit contributions
n+0.820
ric+0.804
 disl+0.784
pieces+0.753
ieved+0.723
Neg logit contributions
 whispered-0.668
 faded-0.584
 needed-0.572
 blank-0.550
 spoke-0.545
Token right
Token ID826
Feature activation+4.23e-01

Pos logit contributions
 disl+0.527
pieces+0.525
 hey+0.521
iu+0.510
 arrest+0.506
Neg logit contributions
 needed-0.583
 whispered-0.572
 meant-0.563
 said-0.514
 faded-0.496
Token,
Token ID11
Feature activation+4.00e-01

Pos logit contributions
 disl+0.528
n+0.518
 hey+0.505
ric+0.500
nels+0.490
Neg logit contributions
 whispered-0.499
 needed-0.487
 meant-0.456
 spoke-0.434
 faded-0.433
Token Emma
Token ID18966
Feature activation+4.11e-01

Pos logit contributions
 disl+0.443
 hey+0.430
pieces+0.430
n+0.428
ric+0.416
Neg logit contributions
 whispered-0.526
 needed-0.522
 meant-0.502
 said-0.465
 faded-0.462
Token.
Token ID13
Feature activation+3.81e-01

Pos logit contributions
n+0.484
 disl+0.471
 spend+0.449
 Majesty+0.446
 hey+0.437
Neg logit contributions
 whispered-0.459
 faded-0.446
 needed-0.438
 dark-0.435
 blank-0.428
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenThat
Token ID2504
Feature activation+4.83e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.22e-01

Pos logit contributions
 disl+0.613
n+0.597
pieces+0.592
 hey+0.591
ric+0.577
Neg logit contributions
 whispered-0.559
 needed-0.550
 meant-0.523
 faded-0.482
 said-0.479
Token
Token ID26391
Feature activation+5.65e-01

Pos logit contributions
 disl+0.381
n+0.365
 hey+0.362
pieces+0.361
ric+0.343
Neg logit contributions
 whispered-0.887
 needed-0.881
 meant-0.854
 faded-0.805
 said-0.804
Token
Token ID8151
Feature activation+5.95e-01

Pos logit contributions
 hey+0.628
 disl+0.624
pieces+0.623
iu+0.594
latable+0.593
Neg logit contributions
 needed-0.735
 whispered-0.728
 meant-0.693
 said-0.657
 spoke-0.639
Tokens
Token ID82
Feature activation+4.57e-01

Pos logit contributions
n+0.820
ric+0.804
 disl+0.784
pieces+0.753
ieved+0.723
Neg logit contributions
 whispered-0.668
 faded-0.584
 needed-0.572
 blank-0.550
 spoke-0.545
Token the
Token ID262
Feature activation+4.41e-01

Pos logit contributions
 disl+0.527
pieces+0.525
 hey+0.521
iu+0.510
 arrest+0.506
Neg logit contributions
 needed-0.583
 whispered-0.572
 meant-0.563
 said-0.514
 faded-0.496
Token gas
Token ID3623
Feature activation+4.05e-01

Pos logit contributions
 disl+0.564
n+0.552
 hey+0.549
pieces+0.547
ric+0.536
Neg logit contributions
 whispered-0.506
 needed-0.497
 meant-0.476
 faded-0.436
 spoke-0.432
Token from
Token ID422
Feature activation+4.04e-01

Pos logit contributions
ric+0.501
n+0.501
 disl+0.491
nels+0.470
 hey+0.470
Neg logit contributions
 whispered-0.472
 needed-0.435
 meant-0.428
 faded-0.411
 spoke-0.406
Token the
Token ID262
Feature activation+3.89e-01

Pos logit contributions
 disl+0.418
n+0.412
 hey+0.410
pieces+0.396
 arrest+0.391
Neg logit contributions
 whispered-0.556
 needed-0.550
 meant-0.531
 spoke-0.495
 faded-0.494
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenLet
Token ID5756
Feature activation+4.94e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.17e-01

Pos logit contributions
ric+0.777
iu+0.737
latable+0.733
 hey+0.715
 disl+0.712
Neg logit contributions
 meant-0.511
 whispered-0.505
 said-0.503
 needed-0.492
 spoke-0.425
Token
Token ID26391
Feature activation+5.61e-01

Pos logit contributions
 disl+0.397
 hey+0.396
pieces+0.395
 Majesty+0.389
 arrest+0.381
Neg logit contributions
 whispered-0.830
 needed-0.820
 meant-0.811
 said-0.763
 spoke-0.749
Token
Token ID8151
Feature activation+5.91e-01

Pos logit contributions
 hey+0.690
pieces+0.671
 disl+0.668
latable+0.664
iu+0.657
Neg logit contributions
 needed-0.681
 whispered-0.666
 said-0.626
 meant-0.621
 spoke-0.587
Tokens
Token ID82
Feature activation+4.52e-01

Pos logit contributions
n+0.908
ric+0.876
pieces+0.867
 disl+0.862
 hey+0.836
Neg logit contributions
 whispered-0.545
 needed-0.498
 meant-0.475
 faded-0.471
 dark-0.433
Token explore
Token ID7301
Feature activation+4.09e-01

Pos logit contributions
iu+0.512
latable+0.486
 disl+0.469
pieces+0.467
ext+0.463
Neg logit contributions
 needed-0.637
 whispered-0.627
 meant-0.625
 said-0.617
 spoke-0.544
Token and
Token ID290
Feature activation+3.81e-01

Pos logit contributions
 disl+0.503
 hey+0.489
n+0.488
pieces+0.488
ric+0.473
Neg logit contributions
 whispered-0.489
 needed-0.484
 meant-0.464
 said-0.426
 faded-0.423
Token have
Token ID423
Feature activation+3.71e-01

Pos logit contributions
 disl+0.487
iu+0.477
ric+0.471
 hey+0.470
latable+0.468
Neg logit contributions
 whispered-0.481
 needed-0.474
 said-0.449
 meant-0.440
 spoke-0.409
Token some
Token ID617
Feature activation+3.53e-01

Pos logit contributions
 disl+0.468
 hey+0.465
pieces+0.455
ric+0.452
n+0.449
Neg logit contributions
 whispered-0.439
 needed-0.434
 meant-0.414
 said-0.386
 faded-0.375
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenLet
Token ID5756
Feature activation+4.94e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.17e-01

Pos logit contributions
ric+0.777
iu+0.737
latable+0.733
 hey+0.715
 disl+0.712
Neg logit contributions
 meant-0.511
 whispered-0.505
 said-0.503
 needed-0.492
 spoke-0.425
Token
Token ID26391
Feature activation+5.61e-01

Pos logit contributions
 disl+0.397
 hey+0.396
pieces+0.395
 Majesty+0.389
 arrest+0.381
Neg logit contributions
 whispered-0.830
 needed-0.820
 meant-0.811
 said-0.763
 spoke-0.749
Token
Token ID8151
Feature activation+5.91e-01

Pos logit contributions
 hey+0.690
pieces+0.671
 disl+0.668
latable+0.664
iu+0.657
Neg logit contributions
 needed-0.681
 whispered-0.666
 said-0.626
 meant-0.621
 spoke-0.587
Tokens
Token ID82
Feature activation+4.52e-01

Pos logit contributions
n+0.908
ric+0.876
pieces+0.867
 disl+0.862
 hey+0.836
Neg logit contributions
 whispered-0.545
 needed-0.498
 meant-0.475
 faded-0.471
 dark-0.433
Token have
Token ID423
Feature activation+4.23e-01

Pos logit contributions
iu+0.512
latable+0.486
 disl+0.469
pieces+0.467
ext+0.463
Neg logit contributions
 needed-0.637
 whispered-0.627
 meant-0.625
 said-0.617
 spoke-0.544
Token a
Token ID257
Feature activation+4.13e-01

Pos logit contributions
 hey+0.546
 disl+0.540
ric+0.539
pieces+0.534
iu+0.529
Neg logit contributions
 whispered-0.490
 needed-0.484
 meant-0.461
 said-0.439
 faded-0.416
Token stretching
Token ID20880
Feature activation+3.98e-01

Pos logit contributions
 disl+0.497
 hey+0.483
n+0.482
pieces+0.481
ric+0.469
Neg logit contributions
 whispered-0.505
 needed-0.501
 meant-0.478
 said-0.443
 faded-0.440
Token race
Token ID3234
Feature activation+3.79e-01

Pos logit contributions
 disl+0.524
 hey+0.514
pieces+0.508
n+0.498
iu+0.493
Neg logit contributions
 whispered-0.452
 needed-0.450
 meant-0.429
 said-0.397
 faded-0.390
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenLet
Token ID5756
Feature activation+4.94e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.17e-01

Pos logit contributions
ric+0.777
iu+0.737
latable+0.733
 hey+0.715
 disl+0.712
Neg logit contributions
 meant-0.511
 whispered-0.505
 said-0.503
 needed-0.492
 spoke-0.425
Token
Token ID26391
Feature activation+5.61e-01

Pos logit contributions
 disl+0.397
 hey+0.396
pieces+0.395
 Majesty+0.389
 arrest+0.381
Neg logit contributions
 whispered-0.830
 needed-0.820
 meant-0.811
 said-0.763
 spoke-0.749
Token
Token ID8151
Feature activation+5.91e-01

Pos logit contributions
 hey+0.690
pieces+0.671
 disl+0.668
latable+0.664
iu+0.657
Neg logit contributions
 needed-0.681
 whispered-0.666
 said-0.626
 meant-0.621
 spoke-0.587
Tokens
Token ID82
Feature activation+4.52e-01

Pos logit contributions
n+0.908
ric+0.876
pieces+0.867
 disl+0.862
 hey+0.836
Neg logit contributions
 whispered-0.545
 needed-0.498
 meant-0.475
 faded-0.471
 dark-0.433
Token play
Token ID711
Feature activation+4.18e-01

Pos logit contributions
iu+0.512
latable+0.486
 disl+0.469
pieces+0.467
ext+0.463
Neg logit contributions
 needed-0.637
 whispered-0.627
 meant-0.625
 said-0.617
 spoke-0.544
Token a
Token ID257
Feature activation+4.10e-01

Pos logit contributions
 disl+0.554
 hey+0.534
iu+0.532
ric+0.531
pieces+0.528
Neg logit contributions
 whispered-0.470
 needed-0.454
 meant-0.437
 said-0.419
 spoke-0.389
Token joke
Token ID9707
Feature activation+3.92e-01

Pos logit contributions
 disl+0.511
 hey+0.494
n+0.493
pieces+0.491
ric+0.481
Neg logit contributions
 whispered-0.487
 needed-0.483
 meant-0.457
 said-0.426
 faded-0.421
Token.
Token ID13
Feature activation+3.77e-01

Pos logit contributions
n+0.429
pieces+0.385
 disl+0.383
nels+0.370
 join+0.369
Neg logit contributions
 needed-0.530
 whispered-0.515
 faded-0.496
 spoke-0.461
 dusty-0.460
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenLet
Token ID5756
Feature activation+4.94e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.17e-01

Pos logit contributions
ric+0.777
iu+0.737
latable+0.733
 hey+0.715
 disl+0.712
Neg logit contributions
 meant-0.511
 whispered-0.505
 said-0.503
 needed-0.492
 spoke-0.425
Token
Token ID26391
Feature activation+5.61e-01

Pos logit contributions
 disl+0.397
 hey+0.396
pieces+0.395
 Majesty+0.389
 arrest+0.381
Neg logit contributions
 whispered-0.830
 needed-0.820
 meant-0.811
 said-0.763
 spoke-0.749
Token
Token ID8151
Feature activation+5.91e-01

Pos logit contributions
 hey+0.690
pieces+0.671
 disl+0.668
latable+0.664
iu+0.657
Neg logit contributions
 needed-0.681
 whispered-0.666
 said-0.626
 meant-0.621
 spoke-0.587
Tokens
Token ID82
Feature activation+4.52e-01

Pos logit contributions
n+0.908
ric+0.876
pieces+0.867
 disl+0.862
 hey+0.836
Neg logit contributions
 whispered-0.545
 needed-0.498
 meant-0.475
 faded-0.471
 dark-0.433
Token not
Token ID407
Feature activation+4.42e-01

Pos logit contributions
iu+0.512
latable+0.486
 disl+0.469
pieces+0.467
ext+0.463
Neg logit contributions
 needed-0.637
 whispered-0.627
 meant-0.625
 said-0.617
 spoke-0.544
Token think
Token ID892
Feature activation+4.10e-01

Pos logit contributions
latable+0.472
 hey+0.471
iu+0.469
 disl+0.468
pieces+0.457
Neg logit contributions
 needed-0.616
 whispered-0.605
 meant-0.602
 said-0.551
 spoke-0.528
Token about
Token ID546
Feature activation+3.83e-01

Pos logit contributions
n+0.454
 disl+0.448
pieces+0.441
 hey+0.429
ric+0.416
Neg logit contributions
 whispered-0.538
 needed-0.530
 meant-0.505
 faded-0.477
 spoke-0.472
Token the
Token ID262
Feature activation+3.78e-01

Pos logit contributions
 hey+0.404
 disl+0.403
pieces+0.399
iu+0.394
n+0.387
Neg logit contributions
 whispered-0.520
 needed-0.513
 meant-0.500
 said-0.468
 spoke-0.449
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenWho
Token ID8241
Feature activation+4.95e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.09e-01

Pos logit contributions
 disl+0.640
n+0.624
 hey+0.624
pieces+0.623
ric+0.603
Neg logit contributions
 whispered-0.557
 needed-0.550
 meant-0.525
 faded-0.479
 said-0.478
Token
Token ID26391
Feature activation+5.56e-01

Pos logit contributions
 disl+0.401
pieces+0.389
 hey+0.387
n+0.376
 arrest+0.366
Neg logit contributions
 whispered-0.830
 needed-0.825
 meant-0.805
 said-0.755
 spoke-0.747
Token
Token ID8151
Feature activation+5.86e-01

Pos logit contributions
 disl+0.661
 hey+0.650
pieces+0.649
n+0.632
iu+0.625
Neg logit contributions
 whispered-0.684
 needed-0.683
 meant-0.649
 said-0.606
 spoke-0.594
Tokens
Token ID82
Feature activation+4.50e-01

Pos logit contributions
n+0.881
ric+0.873
 disl+0.832
pieces+0.828
 hey+0.791
Neg logit contributions
 whispered-0.559
 faded-0.498
 needed-0.489
 blank-0.458
 meant-0.456
Token there
Token ID612
Feature activation+4.29e-01

Pos logit contributions
 disl+0.458
pieces+0.445
 hey+0.441
n+0.438
iu+0.430
Neg logit contributions
 whispered-0.635
 needed-0.635
 meant-0.614
 said-0.572
 spoke-0.560
Token?
Token ID30
Feature activation+4.02e-01

Pos logit contributions
n+0.366
ric+0.344
 hey+0.332
 disl+0.332
 spend+0.331
Neg logit contributions
 whispered-0.673
 needed-0.639
 faded-0.633
 meant-0.605
 spoke-0.603
Tokenâ
Token ID22940
Feature activation+4.09e-01

Pos logit contributions
n+0.435
 disl+0.424
pieces+0.417
 hey+0.414
ric+0.406
Neg logit contributions
 whispered-0.542
 needed-0.531
 meant-0.512
 faded-0.499
chy-0.480
Token
Token ID26391
Feature activation+4.57e-01

Pos logit contributions
n+0.271
 disl+0.261
ric+0.227
 hey+0.218
iu+0.209
Neg logit contributions
 needed-0.749
 whispered-0.742
 meant-0.711
 faded-0.710
 smelled-0.683
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenPlease
Token ID5492
Feature activation+5.15e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Token don
Token ID836
Feature activation+4.96e-01

Pos logit contributions
 disl+0.570
pieces+0.554
 hey+0.550
n+0.547
ric+0.532
Neg logit contributions
 whispered-0.683
 needed-0.678
 meant-0.652
 said-0.606
 faded-0.597
Tokenâ
Token ID22940
Feature activation+5.07e-01

Pos logit contributions
 Majesty+0.675
 disl+0.662
 interests+0.633
 hey+0.629
pieces+0.619
Neg logit contributions
 meant-0.609
 needed-0.600
 whispered-0.577
 said-0.559
 spoke-0.504
Token
Token ID26391
Feature activation+5.51e-01

Pos logit contributions
 disl+0.435
pieces+0.425
 Majesty+0.416
 hey+0.411
 blast+0.404
Neg logit contributions
 whispered-0.795
 needed-0.791
 meant-0.785
 said-0.732
 spoke-0.707
Token
Token ID8151
Feature activation+5.80e-01

Pos logit contributions
 disl+0.680
pieces+0.678
 hey+0.677
latable+0.654
iu+0.649
Neg logit contributions
 needed-0.654
 whispered-0.649
 meant-0.614
 said-0.593
 spoke-0.561
Tokent
Token ID83
Feature activation+4.54e-01

Pos logit contributions
n+0.808
ric+0.765
pieces+0.757
 disl+0.749
 hey+0.732
Neg logit contributions
 whispered-0.626
 needed-0.566
 faded-0.546
 meant-0.542
 shone-0.520
Token eat
Token ID4483
Feature activation+4.01e-01

Pos logit contributions
 disl+0.449
 hey+0.436
pieces+0.434
n+0.432
ric+0.417
Neg logit contributions
 whispered-0.655
 needed-0.651
 meant-0.630
 said-0.586
 faded-0.582
Token me
Token ID502
Feature activation+3.98e-01

Pos logit contributions
 disl+0.385
n+0.373
 hey+0.370
pieces+0.370
ric+0.356
Neg logit contributions
 whispered-0.587
 needed-0.582
 meant-0.561
 faded-0.524
 said-0.522
Token,
Token ID11
Feature activation+3.79e-01

Pos logit contributions
n+0.440
 disl+0.429
pieces+0.413
 spend+0.406
 interests+0.405
Neg logit contributions
 whispered-0.510
 needed-0.499
 meant-0.485
 faded-0.474
 spoke-0.460
Tokenâ
Token ID22940
Feature activation+4.12e-01

Pos logit contributions
 disl+0.442
 hey+0.430
pieces+0.429
n+0.429
ric+0.415
Neg logit contributions
 whispered-0.476
 needed-0.472
 meant-0.453
 said-0.418
 faded-0.415
Token
Token ID129
Feature activation+5.89e-01

Pos logit contributions
 disl+0.983
 hey+0.975
pieces+0.972
latable+0.940
iu+0.932
Neg logit contributions
 whispered-0.583
 needed-0.572
 meant-0.523
 said-0.507
 spoke-0.473
Token
Token ID241
Feature activation+5.68e-01

Pos logit contributions
 disl+0.527
n+0.511
 hey+0.495
pieces+0.493
ric+0.479
Neg logit contributions
 whispered-0.909
 needed-0.908
 meant-0.876
 faded-0.821
 said-0.814
TokenI
Token ID40
Feature activation+4.87e-01

Pos logit contributions
 disl+0.831
pieces+0.828
 hey+0.822
 Majesty+0.807
 arrest+0.805
Neg logit contributions
 needed-0.544
 whispered-0.529
 meant-0.516
 said-0.463
 faded-0.433
Tokenâ
Token ID22940
Feature activation+5.05e-01

Pos logit contributions
 disl+0.484
pieces+0.470
 hey+0.469
n+0.462
iu+0.451
Neg logit contributions
 whispered-0.704
 needed-0.699
 meant-0.676
 said-0.631
 faded-0.623
Token
Token ID26391
Feature activation+5.50e-01

Pos logit contributions
 disl+0.364
n+0.348
 hey+0.346
pieces+0.345
ric+0.327
Neg logit contributions
 whispered-0.865
 needed-0.860
 meant-0.833
 faded-0.785
 said-0.785
Token
Token ID8151
Feature activation+5.79e-01

Pos logit contributions
 hey+0.649
 disl+0.646
pieces+0.641
latable+0.618
iu+0.611
Neg logit contributions
 whispered-0.686
 needed-0.684
 meant-0.633
 said-0.630
 spoke-0.599
Tokenm
Token ID76
Feature activation+4.37e-01

Pos logit contributions
n+0.823
ric+0.806
 disl+0.773
pieces+0.764
ieved+0.744
Neg logit contributions
 whispered-0.590
 needed-0.531
 faded-0.525
 meant-0.502
 dark-0.481
Token here
Token ID994
Feature activation+4.13e-01

Pos logit contributions
 disl+0.541
n+0.535
ric+0.526
 hey+0.523
pieces+0.517
Neg logit contributions
 whispered-0.521
 needed-0.511
 meant-0.486
 faded-0.460
 spoke-0.453
Token to
Token ID284
Feature activation+3.92e-01

Pos logit contributions
 disl+0.443
n+0.443
 hey+0.430
pieces+0.420
 spend+0.415
Neg logit contributions
 whispered-0.548
 needed-0.538
 meant-0.513
 faded-0.491
 spoke-0.485
Token give
Token ID1577
Feature activation+3.69e-01

Pos logit contributions
iu+0.432
nels+0.404
pieces+0.403
 disl+0.402
ric+0.392
Neg logit contributions
 whispered-0.554
 said-0.532
 needed-0.525
 meant-0.516
 dark-0.481
Token you
Token ID345
Feature activation+3.68e-01

Pos logit contributions
 disl+0.387
 hey+0.374
pieces+0.373
ric+0.364
n+0.364
Neg logit contributions
 whispered-0.518
 needed-0.509
 meant-0.493
 said-0.467
 spoke-0.451
Token't
Token ID470
Feature activation+5.83e-01

Pos logit contributions
 whispered+0.278
 needed+0.271
 meant+0.261
 faded+0.240
chy+0.237
Neg logit contributions
 disl-0.281
n-0.276
 hey-0.274
pieces-0.266
 yeah-0.260
Token know
Token ID760
Feature activation+4.73e-01

Pos logit contributions
latable+0.673
ric+0.657
nels+0.648
pieces+0.645
iu+0.633
Neg logit contributions
 needed-0.816
 whispered-0.812
 meant-0.786
 said-0.755
 spoke-0.707
Token the
Token ID262
Feature activation+4.73e-01

Pos logit contributions
 disl+0.410
n+0.395
 hey+0.394
pieces+0.394
ric+0.376
Neg logit contributions
 whispered-0.736
 needed-0.730
 meant-0.707
 faded-0.661
 said-0.661
Token rh
Token ID9529
Feature activation+5.33e-01

Pos logit contributions
 disl+0.617
n+0.600
 hey+0.599
pieces+0.597
ric+0.588
Neg logit contributions
 whispered-0.520
 needed-0.502
 meant-0.480
 faded-0.445
 spoke-0.445
Tokenin
Token ID259
Feature activation+4.54e-01

Pos logit contributions
 disl+0.802
n+0.771
pieces+0.763
 hey+0.762
latable+0.750
Neg logit contributions
 needed-0.511
 whispered-0.504
 meant-0.476
 said-0.429
 faded-0.422
Tokenoc
Token ID420
Feature activation+5.77e-01

Pos logit contributions
n+0.381
 disl+0.336
 victory+0.315
pieces+0.308
earned+0.302
Neg logit contributions
 whispered-0.729
 faded-0.682
 needed-0.661
 blank-0.660
 meant-0.660
Tokeneros
Token ID27498
Feature activation+4.32e-01

Pos logit contributions
 blast+0.565
 disl+0.564
n+0.552
pieces+0.551
latable+0.542
Neg logit contributions
 needed-0.821
 whispered-0.799
 meant-0.781
 said-0.729
 Jones-0.721
Token was
Token ID373
Feature activation+3.84e-01

Pos logit contributions
n+0.589
 disl+0.530
 spend+0.522
pieces+0.519
 arrest+0.519
Neg logit contributions
 blank-0.433
 whispered-0.428
 faded-0.394
 dark-0.394
 aisle-0.390
Token wild
Token ID4295
Feature activation+3.79e-01

Pos logit contributions
n+0.456
 victory+0.447
 disl+0.447
ric+0.432
 arrest+0.431
Neg logit contributions
 whispered-0.444
 needed-0.410
 meant-0.394
 Jones-0.382
 aisle-0.380
Token.
Token ID13
Feature activation+3.80e-01

Pos logit contributions
 disl+0.544
n+0.527
 hey+0.523
pieces+0.520
 arrest+0.513
Neg logit contributions
 whispered-0.374
 needed-0.359
 meant-0.333
 faded-0.312
 spoke-0.310
Token Can
Token ID1680
Feature activation+3.68e-01

Pos logit contributions
 disl+0.449
n+0.447
 hey+0.419
ric+0.417
pieces+0.412
Neg logit contributions
 needed-0.460
 whispered-0.460
 meant-0.446
 faded-0.436
chy-0.434
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenWhy
Token ID5195
Feature activation+5.28e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Token don
Token ID836
Feature activation+4.91e-01

Pos logit contributions
 disl+0.541
pieces+0.533
 hey+0.530
n+0.522
iu+0.515
Neg logit contributions
 whispered-0.739
 needed-0.736
 meant-0.717
 said-0.666
 faded-0.648
Tokenâ
Token ID22940
Feature activation+5.01e-01

Pos logit contributions
 Majesty+0.733
 interests+0.689
 disl+0.673
 hey+0.664
iu+0.651
Neg logit contributions
 meant-0.572
 needed-0.557
 whispered-0.518
 said-0.514
 spoke-0.477
Token
Token ID26391
Feature activation+5.47e-01

Pos logit contributions
 disl+0.418
pieces+0.410
 hey+0.399
 Majesty+0.392
 arrest+0.387
Neg logit contributions
 whispered-0.798
 needed-0.791
 meant-0.781
 said-0.730
 spoke-0.713
Token
Token ID8151
Feature activation+5.76e-01

Pos logit contributions
 disl+0.641
 hey+0.634
pieces+0.633
iu+0.609
n+0.608
Neg logit contributions
 needed-0.682
 whispered-0.681
 meant-0.650
 said-0.612
 spoke-0.594
Tokent
Token ID83
Feature activation+4.51e-01

Pos logit contributions
n+0.777
ric+0.724
pieces+0.710
 disl+0.697
ieved+0.687
Neg logit contributions
 whispered-0.653
 faded-0.583
 needed-0.577
 blank-0.569
 dark-0.562
Token you
Token ID345
Feature activation+4.24e-01

Pos logit contributions
 disl+0.506
 hey+0.492
pieces+0.491
n+0.490
ric+0.474
Neg logit contributions
 whispered-0.592
 needed-0.587
 meant-0.565
 said-0.522
 faded-0.519
Token come
Token ID1282
Feature activation+3.93e-01

Pos logit contributions
n+0.412
 spend+0.396
 pause+0.387
 join+0.374
 hey+0.369
Neg logit contributions
 whispered-0.587
 needed-0.563
 aisle-0.554
 faded-0.553
 meant-0.546
Token with
Token ID351
Feature activation+3.71e-01

Pos logit contributions
 disl+0.492
 hey+0.477
pieces+0.476
n+0.475
iu+0.462
Neg logit contributions
 whispered-0.467
 needed-0.460
 meant-0.442
 said-0.406
 faded-0.402
Token me
Token ID502
Feature activation+3.61e-01

Pos logit contributions
 disl+0.341
n+0.331
 hey+0.324
pieces+0.322
 blast+0.309
Neg logit contributions
 whispered-0.556
 needed-0.554
 meant-0.533
 faded-0.503
 spoke-0.498
Token of
Token ID286
Feature activation+5.88e-01

Pos logit contributions
 whispered+0.334
 needed+0.328
 meant+0.314
chy+0.309
 faded+0.307
Neg logit contributions
 pause-0.234
 arrest-0.232
 spend-0.230
 victory-0.229
 disl-0.226
Token other
Token ID584
Feature activation+5.17e-01

Pos logit contributions
 disl+0.721
 hey+0.704
pieces+0.701
n+0.701
ric+0.680
Neg logit contributions
 whispered-0.705
 needed-0.701
 meant-0.670
 said-0.616
 faded-0.611
Token people
Token ID661
Feature activation+4.49e-01

Pos logit contributions
pieces+0.538
 disl+0.525
 arrest+0.523
 hey+0.519
n+0.511
Neg logit contributions
 whispered-0.700
 needed-0.654
 meant-0.654
 faded-0.620
 spoke-0.615
Tokenâ
Token ID22940
Feature activation+5.05e-01

Pos logit contributions
n+0.582
 disl+0.581
pieces+0.576
 hey+0.560
 spend+0.560
Neg logit contributions
 whispered-0.466
 meant-0.443
 needed-0.436
 faded-0.416
 Jones-0.400
Token
Token ID26391
Feature activation+5.50e-01

Pos logit contributions
n+0.491
 disl+0.486
 hey+0.473
pieces+0.461
ric+0.453
Neg logit contributions
 whispered-0.744
 needed-0.734
 meant-0.703
 faded-0.674
 said-0.657
Token
Token ID8151
Feature activation+5.76e-01

Pos logit contributions
 disl+0.739
 hey+0.727
pieces+0.725
n+0.709
iu+0.698
Neg logit contributions
 whispered-0.593
 needed-0.590
 meant-0.552
 said-0.517
 spoke-0.501
Tokens
Token ID82
Feature activation+4.49e-01

Pos logit contributions
n+0.825
ric+0.796
 disl+0.768
pieces+0.758
 hey+0.739
Neg logit contributions
 whispered-0.586
 needed-0.529
 meant-0.525
 faded-0.524
 blank-0.505
Token property
Token ID3119
Feature activation+4.48e-01

Pos logit contributions
 disl+0.552
n+0.540
 hey+0.532
ric+0.525
pieces+0.525
Neg logit contributions
 whispered-0.535
 needed-0.509
 meant-0.497
 faded-0.463
 spoke-0.457
Token.
Token ID13
Feature activation+4.01e-01

Pos logit contributions
n+0.581
 disl+0.568
 hey+0.543
pieces+0.542
ric+0.541
Neg logit contributions
 whispered-0.492
 needed-0.459
 faded-0.429
 meant-0.417
 spoke-0.416
Token We
Token ID775
Feature activation+3.66e-01

Pos logit contributions
 disl+0.504
n+0.502
 hey+0.481
pieces+0.480
ric+0.479
Neg logit contributions
 whispered-0.473
 needed-0.466
 meant-0.447
 faded-0.421
 spoke-0.410
Token should
Token ID815
Feature activation+3.66e-01

Pos logit contributions
 disl+0.389
 hey+0.377
n+0.377
pieces+0.376
ric+0.363
Neg logit contributions
 whispered-0.498
 needed-0.494
 meant-0.476
 said-0.441
 faded-0.439
Token
Token ID129
Feature activation+5.89e-01

Pos logit contributions
 disl+0.983
 hey+0.975
pieces+0.972
latable+0.940
iu+0.932
Neg logit contributions
 whispered-0.583
 needed-0.572
 meant-0.523
 said-0.507
 spoke-0.473
Token
Token ID241
Feature activation+5.68e-01

Pos logit contributions
 disl+0.527
n+0.511
 hey+0.495
pieces+0.493
ric+0.479
Neg logit contributions
 whispered-0.909
 needed-0.908
 meant-0.876
 faded-0.821
 said-0.814
TokenWhere
Token ID8496
Feature activation+5.09e-01

Pos logit contributions
 disl+0.831
pieces+0.828
 hey+0.822
 Majesty+0.807
 arrest+0.805
Neg logit contributions
 needed-0.544
 whispered-0.529
 meant-0.516
 said-0.463
 faded-0.433
Tokenâ
Token ID22940
Feature activation+5.00e-01

Pos logit contributions
 disl+0.643
iu+0.638
 arrest+0.629
pieces+0.629
 hey+0.628
Neg logit contributions
 needed-0.618
 whispered-0.609
 meant-0.608
 said-0.562
 spoke-0.518
Token
Token ID26391
Feature activation+5.46e-01

Pos logit contributions
 disl+0.376
pieces+0.369
 hey+0.365
 arrest+0.347
ric+0.344
Neg logit contributions
 whispered-0.837
 needed-0.828
 meant-0.811
 said-0.765
 spoke-0.754
Token
Token ID8151
Feature activation+5.75e-01

Pos logit contributions
 disl+0.760
 hey+0.753
pieces+0.748
iu+0.725
latable+0.724
Neg logit contributions
 whispered-0.570
 needed-0.569
 meant-0.529
 said-0.503
 spoke-0.486
Tokens
Token ID82
Feature activation+4.41e-01

Pos logit contributions
n+0.824
ric+0.795
pieces+0.761
 disl+0.760
ieved+0.731
Neg logit contributions
 whispered-0.589
 faded-0.529
 needed-0.525
 blank-0.500
 meant-0.489
Token my
Token ID616
Feature activation+4.40e-01

Pos logit contributions
 disl+0.499
pieces+0.493
iu+0.485
 hey+0.484
ric+0.475
Neg logit contributions
 needed-0.580
 whispered-0.576
 meant-0.563
 said-0.525
 spoke-0.503
Token jelly
Token ID27644
Feature activation+4.24e-01

Pos logit contributions
 disl+0.469
n+0.457
 hey+0.455
pieces+0.452
ric+0.441
Neg logit contributions
 whispered-0.596
 needed-0.589
 meant-0.570
 faded-0.526
 spoke-0.524
Token?
Token ID30
Feature activation+3.89e-01

Pos logit contributions
n+0.363
 disl+0.363
pieces+0.338
 hey+0.334
ric+0.331
Neg logit contributions
 whispered-0.648
 needed-0.634
 meant-0.610
 faded-0.585
 spoke-0.580
Tokenâ
Token ID22940
Feature activation+3.98e-01

Pos logit contributions
n+0.413
 disl+0.397
pieces+0.390
 hey+0.379
ric+0.379
Neg logit contributions
 whispered-0.537
 needed-0.533
 meant-0.506
 faded-0.497
chy-0.483
Token
Token ID129
Feature activation+5.89e-01

Pos logit contributions
 disl+0.983
 hey+0.975
pieces+0.972
latable+0.940
iu+0.932
Neg logit contributions
 whispered-0.583
 needed-0.572
 meant-0.523
 said-0.507
 spoke-0.473
Token
Token ID241
Feature activation+5.68e-01

Pos logit contributions
 disl+0.527
n+0.511
 hey+0.495
pieces+0.493
ric+0.479
Neg logit contributions
 whispered-0.909
 needed-0.908
 meant-0.876
 faded-0.821
 said-0.814
TokenIt
Token ID1026
Feature activation+4.70e-01

Pos logit contributions
 disl+0.831
pieces+0.828
 hey+0.822
 Majesty+0.807
 arrest+0.805
Neg logit contributions
 needed-0.544
 whispered-0.529
 meant-0.516
 said-0.463
 faded-0.433
Tokenâ
Token ID22940
Feature activation+5.00e-01

Pos logit contributions
 disl+0.587
n+0.578
pieces+0.569
 hey+0.567
ric+0.556
Neg logit contributions
 whispered-0.548
 needed-0.539
 meant-0.513
 faded-0.478
 said-0.470
Token
Token ID26391
Feature activation+5.46e-01

Pos logit contributions
 disl+0.383
n+0.370
 hey+0.365
pieces+0.363
ric+0.347
Neg logit contributions
 whispered-0.833
 needed-0.827
 meant-0.801
 faded-0.754
 said-0.753
Token
Token ID8151
Feature activation+5.75e-01

Pos logit contributions
 hey+0.648
pieces+0.641
 disl+0.640
iu+0.611
latable+0.611
Neg logit contributions
 whispered-0.671
 needed-0.671
 meant-0.625
 said-0.608
 spoke-0.584
Tokens
Token ID82
Feature activation+4.42e-01

Pos logit contributions
n+0.830
ric+0.817
 disl+0.777
pieces+0.763
ieved+0.734
Neg logit contributions
 whispered-0.594
 needed-0.524
 faded-0.519
 meant-0.498
 blank-0.493
Token a
Token ID257
Feature activation+4.23e-01

Pos logit contributions
 disl+0.505
n+0.491
 hey+0.490
pieces+0.489
ric+0.474
Neg logit contributions
 whispered-0.567
 needed-0.561
 meant-0.539
 said-0.497
 faded-0.496
Token surprise
Token ID5975
Feature activation+4.07e-01

Pos logit contributions
 disl+0.533
pieces+0.518
 hey+0.517
n+0.517
 blast+0.517
Neg logit contributions
 whispered-0.484
 needed-0.472
 meant-0.463
 spoke-0.414
 faded-0.413
Token!
Token ID0
Feature activation+3.90e-01

Pos logit contributions
n+0.412
pieces+0.379
 disl+0.375
 hey+0.369
 blast+0.369
Neg logit contributions
 whispered-0.562
 needed-0.558
 meant-0.520
 faded-0.514
 spoke-0.504
Tokenâ
Token ID22940
Feature activation+3.98e-01

Pos logit contributions
 disl+0.444
 hey+0.430
pieces+0.427
n+0.420
ric+0.416
Neg logit contributions
 whispered-0.505
 needed-0.498
 meant-0.481
 said-0.447
 spoke-0.436
Token
Token ID129
Feature activation+5.89e-01

Pos logit contributions
 disl+0.983
 hey+0.975
pieces+0.972
latable+0.940
iu+0.932
Neg logit contributions
 whispered-0.583
 needed-0.572
 meant-0.523
 said-0.507
 spoke-0.473
Token
Token ID241
Feature activation+5.68e-01

Pos logit contributions
 disl+0.527
n+0.511
 hey+0.495
pieces+0.493
ric+0.479
Neg logit contributions
 whispered-0.909
 needed-0.908
 meant-0.876
 faded-0.821
 said-0.814
TokenIt
Token ID1026
Feature activation+4.70e-01

Pos logit contributions
 disl+0.831
pieces+0.828
 hey+0.822
 Majesty+0.807
 arrest+0.805
Neg logit contributions
 needed-0.544
 whispered-0.529
 meant-0.516
 said-0.463
 faded-0.433
Tokenâ
Token ID22940
Feature activation+5.00e-01

Pos logit contributions
 disl+0.587
n+0.578
pieces+0.569
 hey+0.567
ric+0.556
Neg logit contributions
 whispered-0.548
 needed-0.539
 meant-0.513
 faded-0.478
 said-0.470
Token
Token ID26391
Feature activation+5.46e-01

Pos logit contributions
 disl+0.383
n+0.370
 hey+0.365
pieces+0.363
ric+0.347
Neg logit contributions
 whispered-0.833
 needed-0.827
 meant-0.801
 faded-0.754
 said-0.753
Token
Token ID8151
Feature activation+5.75e-01

Pos logit contributions
 hey+0.648
pieces+0.641
 disl+0.640
iu+0.611
latable+0.611
Neg logit contributions
 whispered-0.671
 needed-0.671
 meant-0.625
 said-0.608
 spoke-0.584
Tokens
Token ID82
Feature activation+4.42e-01

Pos logit contributions
n+0.830
ric+0.817
 disl+0.777
pieces+0.763
ieved+0.734
Neg logit contributions
 whispered-0.594
 needed-0.524
 faded-0.519
 meant-0.498
 blank-0.493
Token a
Token ID257
Feature activation+4.23e-01

Pos logit contributions
 disl+0.505
n+0.491
 hey+0.490
pieces+0.489
ric+0.474
Neg logit contributions
 whispered-0.567
 needed-0.561
 meant-0.539
 said-0.497
 faded-0.496
Token books
Token ID3835
Feature activation+4.08e-01

Pos logit contributions
 disl+0.533
pieces+0.518
 hey+0.517
n+0.517
 blast+0.517
Neg logit contributions
 whispered-0.484
 needed-0.472
 meant-0.463
 spoke-0.414
 faded-0.413
Tokenhe
Token ID258
Feature activation+4.27e-01

Pos logit contributions
n+0.425
 blast+0.400
 spend+0.397
 victory+0.375
 pause+0.371
Neg logit contributions
 whispered-0.535
 needed-0.513
 Jones-0.491
 spoke-0.484
 White-0.476
Tokenlf
Token ID1652
Feature activation+4.26e-01

Pos logit contributions
 disl+0.521
n+0.511
pieces+0.506
 hey+0.505
ric+0.489
Neg logit contributions
 whispered-0.512
 needed-0.506
 meant-0.485
 faded-0.445
 said-0.442
Token said
Token ID531
Feature activation-2.20e-01

Pos logit contributions
 whispered+0.182
 needed+0.179
 meant+0.167
 said+0.149
 faded+0.145
Neg logit contributions
 disl-0.365
 hey-0.358
pieces-0.357
n-0.357
ric-0.353
Token goodbye
Token ID24829
Feature activation-2.23e-01

Pos logit contributions
 whispered+0.295
 needed+0.292
 meant+0.281
 said+0.263
 faded+0.260
Neg logit contributions
 disl-0.240
pieces-0.234
 hey-0.234
ric-0.233
n-0.233
Token.
Token ID13
Feature activation-2.03e-01

Pos logit contributions
 meant+0.260
 whispered+0.258
 needed+0.256
 spoke+0.228
 faded+0.226
Neg logit contributions
n-0.273
 disl-0.264
 hey-0.259
iu-0.252
 victory-0.250
Token She
Token ID1375
Feature activation-2.26e-01

Pos logit contributions
 whispered+0.226
 needed+0.221
 meant+0.219
 faded+0.197
 spoke+0.194
Neg logit contributions
 disl-0.257
n-0.253
pieces-0.247
 blast-0.242
 spend-0.242
Token walked
Token ID6807
Feature activation-2.41e-01

Pos logit contributions
 meant+0.186
 needed+0.185
 whispered+0.180
 Jones+0.176
 dark+0.167
Neg logit contributions
n-0.335
 blast-0.325
 spend-0.321
 disl-0.312
ext-0.310
Token safely
Token ID11512
Feature activation-2.81e-01

Pos logit contributions
 needed+0.334
 whispered+0.332
 meant+0.323
 faded+0.295
 spoke+0.293
Neg logit contributions
n-0.244
 disl-0.244
 hey-0.242
pieces-0.242
 blast-0.231
Token home
Token ID1363
Feature activation-2.67e-01

Pos logit contributions
 meant+0.428
 needed+0.419
 whispered+0.401
 smelled+0.386
chy+0.377
Neg logit contributions
n-0.251
 blast-0.242
 spend-0.238
 arrest-0.238
 join-0.238
Token,
Token ID11
Feature activation-2.38e-01

Pos logit contributions
 meant+0.333
 needed+0.322
 whispered+0.315
chy+0.288
Fuel+0.287
Neg logit contributions
n-0.304
 arrest-0.297
 disl-0.288
 victory-0.286
 hey-0.286
Token feeling
Token ID4203
Feature activation-2.50e-01

Pos logit contributions
 meant+0.288
 needed+0.278
 whispered+0.265
chy+0.250
 Jones+0.250
Neg logit contributions
n-0.273
 spend-0.270
 blast-0.268
 pause-0.261
 disl-0.252
Token happy
Token ID3772
Feature activation-2.47e-01

Pos logit contributions
 whispered+0.253
 needed+0.248
 meant+0.236
 spoke+0.211
 said+0.211
Neg logit contributions
 disl-0.351
 hey-0.340
n-0.339
pieces-0.337
 blast-0.332
Token that
Token ID326
Feature activation-2.49e-01

Pos logit contributions
 whispered+0.287
 meant+0.275
 needed+0.273
chy+0.248
 spoke+0.246
Neg logit contributions
 victory-0.286
n-0.283
 disl-0.282
 hey-0.281
 interests-0.280
Token.
Token ID13
Feature activation-2.04e-01

Pos logit contributions
 whispered+0.279
 needed+0.268
 meant+0.249
 spoke+0.244
 bored+0.240
Neg logit contributions
n-0.299
pieces-0.294
 victory-0.279
 blast-0.266
 arrest-0.266
Token This
Token ID770
Feature activation-2.42e-01

Pos logit contributions
 whispered+0.203
 needed+0.201
 meant+0.191
 said+0.171
 faded+0.170
Neg logit contributions
 disl-0.291
n-0.285
 hey-0.285
pieces-0.285
ric-0.278
Token time
Token ID640
Feature activation-2.39e-01

Pos logit contributions
 whispered+0.359
 needed+0.332
 Jones+0.324
 spoke+0.318
 bored+0.312
Neg logit contributions
 blast-0.224
n-0.220
 victory-0.217
 spend-0.211
 interests-0.206
Token,
Token ID11
Feature activation-2.31e-01

Pos logit contributions
 whispered+0.324
 needed+0.308
 meant+0.301
chy+0.298
Fuel+0.297
Neg logit contributions
n-0.237
 blast-0.222
 spend-0.221
 our-0.221
 victory-0.215
Token Rita
Token ID42605
Feature activation-2.43e-01

Pos logit contributions
 whispered+0.216
 needed+0.211
 meant+0.200
 said+0.182
 faded+0.176
Neg logit contributions
 disl-0.361
 hey-0.346
pieces-0.345
n-0.343
ric-0.341
Token successfully
Token ID7675
Feature activation-2.76e-01

Pos logit contributions
 dark+0.226
 whispered+0.224
 meant+0.214
 faded+0.210
 bored+0.205
Neg logit contributions
n-0.346
 spend-0.325
 disl-0.322
 blast-0.322
 arrest-0.318
Token measured
Token ID8630
Feature activation-2.56e-01

Pos logit contributions
 whispered+0.332
 Jones+0.315
 needed+0.311
 meant+0.306
 White+0.301
Neg logit contributions
n-0.334
 blast-0.311
 arrest-0.307
 disl-0.304
 spend-0.303
Token the
Token ID262
Feature activation-2.43e-01

Pos logit contributions
 whispered+0.356
 needed+0.355
 spoke+0.328
 meant+0.325
 faded+0.321
Neg logit contributions
n-0.259
pieces-0.253
ric-0.246
 hey-0.243
 victory-0.242
Token cushion
Token ID33579
Feature activation-2.52e-01

Pos logit contributions
 whispered+0.345
 needed+0.327
 meant+0.314
 spoke+0.314
 faded+0.298
Neg logit contributions
 victory-0.253
 blast-0.253
 hey-0.245
n-0.245
pieces-0.245
Token very
Token ID845
Feature activation-2.47e-01

Pos logit contributions
 whispered+0.306
 needed+0.290
 meant+0.282
 spoke+0.268
 faded+0.262
Neg logit contributions
pieces-0.306
n-0.304
 victory-0.289
 spend-0.284
 hey-0.280
Token carefully
Token ID7773
Feature activation-2.65e-01

Pos logit contributions
 whispered+0.270
 needed+0.261
 meant+0.254
 spoke+0.232
 said+0.225
Neg logit contributions
 hey-0.328
n-0.327
 disl-0.325
pieces-0.320
iu-0.317
Token bird
Token ID6512
Feature activation-2.29e-01

Pos logit contributions
 whispered+0.264
 meant+0.259
 needed+0.257
 Jones+0.242
 spoke+0.229
Neg logit contributions
 hey-0.266
 blast-0.264
 pause-0.254
n-0.254
 disl-0.248
Token flew
Token ID13112
Feature activation-2.33e-01

Pos logit contributions
 Jones+0.230
 aisle+0.206
 blank+0.203
 meant+0.193
Ancient+0.192
Neg logit contributions
n-0.337
 pause-0.291
 arrest-0.284
 hey-0.284
 spend-0.283
Token away
Token ID1497
Feature activation-2.40e-01

Pos logit contributions
 needed+0.370
 whispered+0.364
 meant+0.355
 faded+0.333
 spoke+0.333
Neg logit contributions
n-0.176
 victory-0.175
 pause-0.175
 hey-0.167
 our-0.164
Token quickly
Token ID2952
Feature activation-2.45e-01

Pos logit contributions
Fuel+0.347
 meant+0.342
book+0.338
 Lover+0.334
ison+0.329
Neg logit contributions
 victory-0.199
 pause-0.184
 arrest-0.182
 our-0.177
 how-0.175
Token and
Token ID290
Feature activation-2.32e-01

Pos logit contributions
 meant+0.285
 needed+0.281
 whispered+0.274
 shone+0.265
 smelled+0.263
Neg logit contributions
 arrest-0.306
n-0.295
 spend-0.279
 pause-0.277
 blast-0.274
Token safely
Token ID11512
Feature activation-2.75e-01

Pos logit contributions
 blank+0.274
elly+0.269
book+0.262
 Jones+0.260
books+0.260
Neg logit contributions
n-0.278
 hey-0.266
 our-0.262
 arrest-0.258
 spend-0.251
Token.
Token ID13
Feature activation-2.10e-01

Pos logit contributions
isha+0.385
 Jones+0.374
 blank+0.372
ison+0.361
book+0.361
Neg logit contributions
n-0.301
 escape-0.253
 stay-0.253
 arrest-0.249
 join-0.241
Token
Token ID220
Feature activation-1.94e-01

Pos logit contributions
 whispered+0.305
 meant+0.305
 faded+0.289
 needed+0.286
chy+0.279
Neg logit contributions
n-0.197
 disl-0.193
 spend-0.180
 hey-0.178
 arrest-0.177
Token
Token ID198
Feature activation-2.04e-01

Pos logit contributions
 whispered+0.179
 needed+0.177
 meant+0.168
 faded+0.150
 said+0.148
Neg logit contributions
 disl-0.290
n-0.285
 hey-0.284
pieces-0.283
ric-0.277
Token
Token ID198
Feature activation-2.06e-01

Pos logit contributions
 meant+0.270
 needed+0.270
 whispered+0.260
 faded+0.253
 shone+0.240
Neg logit contributions
n-0.245
 blast-0.215
 disl-0.214
ext-0.211
 join-0.203
TokenFrom
Token ID4863
Feature activation-2.29e-01

Pos logit contributions
 whispered+0.305
 needed+0.303
 meant+0.298
 Jones+0.280
 faded+0.272
Neg logit contributions
n-0.205
 disl-0.195
 hey-0.194
pieces-0.188
 blast-0.187
Token always
Token ID1464
Feature activation-2.34e-01

Pos logit contributions
 dark+0.265
 whispered+0.265
 dusty+0.253
 aisle+0.252
Fuel+0.245
Neg logit contributions
n-0.264
 arrest-0.245
 spend-0.245
 disl-0.244
ric-0.242
Token put
Token ID1234
Feature activation-2.33e-01

Pos logit contributions
 aisle+0.232
chy+0.221
 whispered+0.214
 pods+0.205
vine+0.204
Neg logit contributions
 spend-0.321
 disl-0.290
 blast-0.290
 hey-0.288
 interests-0.286
Token her
Token ID607
Feature activation-2.32e-01

Pos logit contributions
 needed+0.259
 whispered+0.258
 meant+0.245
 faded+0.226
 spoke+0.222
Neg logit contributions
pieces-0.303
 disl-0.301
n-0.299
 hey-0.295
 blast-0.288
Token notebook
Token ID20922
Feature activation-2.38e-01

Pos logit contributions
 whispered+0.214
 needed+0.212
 meant+0.196
 said+0.178
 faded+0.173
Neg logit contributions
 disl-0.354
 hey-0.350
n-0.347
ric-0.344
pieces-0.343
Token away
Token ID1497
Feature activation-2.43e-01

Pos logit contributions
 whispered+0.291
 needed+0.286
 meant+0.284
 said+0.254
 spoke+0.254
Neg logit contributions
n-0.278
pieces-0.270
 hey-0.263
 interests-0.260
 blast-0.260
Token safely
Token ID11512
Feature activation-2.75e-01

Pos logit contributions
 whispered+0.304
 needed+0.287
Fuel+0.287
 meant+0.286
 spoke+0.282
Neg logit contributions
 victory-0.248
n-0.236
 interests-0.235
 pause-0.235
 how-0.228
Token when
Token ID618
Feature activation-2.28e-01

Pos logit contributions
 whispered+0.325
 meant+0.314
 needed+0.301
Dead+0.299
Fuel+0.295
Neg logit contributions
n-0.322
 spend-0.297
pieces-0.296
 victory-0.285
 interests-0.283
Token she
Token ID673
Feature activation-2.41e-01

Pos logit contributions
 whispered+0.191
 needed+0.180
 meant+0.174
 said+0.150
 faded+0.149
Neg logit contributions
 disl-0.362
n-0.358
 hey-0.358
pieces-0.356
 arrest-0.347
Token had
Token ID550
Feature activation-2.39e-01

Pos logit contributions
Fuel+0.234
 Jones+0.228
 White+0.226
 whispered+0.225
Gas+0.219
Neg logit contributions
 spend-0.314
n-0.312
 arrest-0.307
 blast-0.294
 practice-0.286
Token finished
Token ID5201
Feature activation-2.56e-01

Pos logit contributions
 whispered+0.306
 needed+0.301
 meant+0.293
chy+0.286
 Jones+0.285
Neg logit contributions
n-0.258
 disl-0.255
 interests-0.245
pieces-0.244
 spend-0.241
Token using
Token ID1262
Feature activation-2.52e-01

Pos logit contributions
 whispered+0.289
 needed+0.284
 meant+0.273
 spoke+0.250
 faded+0.250
Neg logit contributions
 disl-0.333
pieces-0.328
n-0.322
 interests-0.307
 spend-0.306
Token
Token ID198
Feature activation-2.07e-01

Pos logit contributions
 whispered+0.309
 needed+0.304
chy+0.296
 shone+0.294
 faded+0.286
Neg logit contributions
 disl-0.206
 blast-0.191
n-0.187
 yeah-0.179
 join-0.176
Token
Token ID198
Feature activation-2.09e-01

Pos logit contributions
 whispered+0.195
 needed+0.193
 meant+0.183
 faded+0.163
 spoke+0.160
Neg logit contributions
 disl-0.309
n-0.303
 hey-0.302
pieces-0.299
ric-0.293
Token"
Token ID1
Feature activation-2.27e-01

Pos logit contributions
 whispered+0.235
 needed+0.233
 meant+0.220
 said+0.210
 spoke+0.200
Neg logit contributions
 disl-0.275
pieces-0.275
ric-0.272
n-0.268
 hey-0.264
TokenHelp
Token ID22087
Feature activation-2.56e-01

Pos logit contributions
 whispered+0.323
 needed+0.322
 meant+0.308
 Jones+0.296
chy+0.290
Neg logit contributions
 disl-0.229
 hey-0.227
n-0.217
 blast-0.209
 arrest-0.205
Token!
Token ID0
Feature activation-2.25e-01

Pos logit contributions
 whispered+0.269
 needed+0.251
 said+0.251
 meant+0.245
 dark+0.224
Neg logit contributions
 disl-0.384
pieces-0.363
ric-0.354
latable-0.353
 Majesty-0.344
Token Help
Token ID10478
Feature activation-2.75e-01

Pos logit contributions
 needed+0.293
 whispered+0.288
 meant+0.275
 said+0.260
 faded+0.252
Neg logit contributions
pieces-0.257
 disl-0.251
n-0.243
 interests-0.239
latable-0.239
Token!"
Token ID2474
Feature activation-2.26e-01

Pos logit contributions
 whispered+0.439
 needed+0.437
 meant+0.423
 said+0.398
 faded+0.394
Neg logit contributions
 disl-0.232
pieces-0.222
 hey-0.221
n-0.221
ric-0.211
Token Sam
Token ID3409
Feature activation-2.62e-01

Pos logit contributions
 whispered+0.303
 needed+0.300
chy+0.290
 meant+0.290
 shone+0.283
Neg logit contributions
 blast-0.233
 disl-0.229
 hey-0.227
 arrest-0.223
n-0.222
Token said
Token ID531
Feature activation-2.52e-01

Pos logit contributions
 blank+0.240
 Jones+0.234
 aisle+0.233
 faded+0.227
 shone+0.224
Neg logit contributions
n-0.377
 pause-0.346
 hey-0.339
 punish-0.331
 arrest-0.318
Token.
Token ID13
Feature activation-2.34e-01

Pos logit contributions
 whispered+0.254
 needed+0.253
 meant+0.247
 faded+0.225
 shone+0.223
Neg logit contributions
n-0.340
 hey-0.339
iu-0.329
 disl-0.319
pieces-0.319
Token He
Token ID679
Feature activation-2.55e-01

Pos logit contributions
 shone+0.336
chy+0.331
 whispered+0.326
 needed+0.317
 faded+0.316
Neg logit contributions
 arrest-0.214
 blast-0.204
n-0.198
 disl-0.191
han-0.187
Token.
Token ID13
Feature activation-2.00e-01

Pos logit contributions
 whispered+0.273
 needed+0.267
 meant+0.256
 White+0.238
 smelled+0.238
Neg logit contributions
pieces-0.292
n-0.286
 join-0.280
 victory-0.280
 blast-0.275
Token Jane
Token ID12091
Feature activation-2.36e-01

Pos logit contributions
 whispered+0.289
 needed+0.283
 meant+0.274
 faded+0.263
chy+0.259
Neg logit contributions
 disl-0.196
n-0.195
 hey-0.189
pieces-0.189
iu-0.177
Token nodded
Token ID14464
Feature activation-2.31e-01

Pos logit contributions
 dark+0.239
 Jones+0.233
 aisle+0.225
 needed+0.224
 meant+0.221
Neg logit contributions
n-0.325
 spend-0.281
 blast-0.278
ieved-0.274
 practice-0.273
Token,
Token ID11
Feature activation-2.29e-01

Pos logit contributions
 needed+0.298
 aisle+0.291
 whispered+0.283
 meant+0.280
Fuel+0.273
Neg logit contributions
 join-0.243
 pause-0.240
 spend-0.236
n-0.227
ap-0.225
Token and
Token ID290
Feature activation-2.35e-01

Pos logit contributions
 needed+0.252
 meant+0.249
 whispered+0.239
chy+0.225
 aisle+0.220
Neg logit contributions
n-0.289
 disl-0.288
 hey-0.280
pieces-0.279
 spend-0.277
Token thus
Token ID4145
Feature activation-2.75e-01

Pos logit contributions
 needed+0.243
 whispered+0.242
 meant+0.234
chy+0.215
 faded+0.209
Neg logit contributions
n-0.312
pieces-0.309
 hey-0.309
 disl-0.305
 spend-0.297
Token began
Token ID2540
Feature activation-2.40e-01

Pos logit contributions
chy+0.330
oky+0.328
elly+0.324
Fuel+0.320
mail+0.311
Neg logit contributions
n-0.318
 blast-0.302
pieces-0.294
 victory-0.293
 join-0.287
Token the
Token ID262
Feature activation-2.41e-01

Pos logit contributions
 whispered+0.360
 needed+0.358
 meant+0.344
chy+0.323
 aisle+0.317
Neg logit contributions
 disl-0.211
 join-0.207
pieces-0.206
 spend-0.205
 victory-0.204
Token game
Token ID983
Feature activation-2.54e-01

Pos logit contributions
 whispered+0.370
 needed+0.367
 meant+0.351
 spoke+0.332
 faded+0.331
Neg logit contributions
pieces-0.216
 victory-0.212
 blast-0.211
 disl-0.209
 hey-0.208
Token!
Token ID0
Feature activation-2.21e-01

Pos logit contributions
 whispered+0.321
 needed+0.310
 meant+0.297
 Jones+0.281
Fuel+0.277
Neg logit contributions
pieces-0.291
 join-0.291
 victory-0.287
 blast-0.280
n-0.278
Token Everyone
Token ID11075
Feature activation-2.54e-01

Pos logit contributions
 whispered+0.252
 needed+0.246
 meant+0.236
 faded+0.219
 spoke+0.214
Neg logit contributions
 disl-0.285
n-0.280
pieces-0.278
 hey-0.277
ric-0.266
Token him
Token ID683
Feature activation-2.41e-01

Pos logit contributions
 whispered+0.246
 needed+0.234
 meant+0.228
 said+0.217
 spoke+0.205
Neg logit contributions
 disl-0.314
ric-0.309
pieces-0.299
latable-0.297
 hey-0.295
Token how
Token ID703
Feature activation-2.10e-01

Pos logit contributions
 whispered+0.361
 needed+0.357
 meant+0.345
 said+0.329
 spoke+0.312
Neg logit contributions
iu-0.226
 hey-0.221
 blast-0.220
 disl-0.218
ric-0.212
Token to
Token ID284
Feature activation-2.32e-01

Pos logit contributions
 whispered+0.258
 meant+0.245
 needed+0.244
 faded+0.222
 Jones+0.221
Neg logit contributions
 disl-0.257
n-0.247
 hey-0.241
pieces-0.240
 spend-0.232
Token remove
Token ID4781
Feature activation-2.45e-01

Pos logit contributions
 whispered+0.318
 needed+0.311
 meant+0.293
 said+0.288
 faded+0.272
Neg logit contributions
pieces-0.270
 disl-0.262
 Majesty-0.257
iu-0.254
 hey-0.248
Token it
Token ID340
Feature activation-2.47e-01

Pos logit contributions
 whispered+0.374
 needed+0.370
 meant+0.352
 spoke+0.351
 faded+0.343
Neg logit contributions
n-0.216
pieces-0.204
 disl-0.199
 victory-0.198
 our-0.198
Token safely
Token ID11512
Feature activation-2.74e-01

Pos logit contributions
 whispered+0.326
 spoke+0.324
 needed+0.320
 faded+0.300
 shone+0.297
Neg logit contributions
n-0.277
 victory-0.247
 our-0.243
 join-0.233
 pause-0.230
Token.
Token ID13
Feature activation-2.07e-01

Pos logit contributions
 whispered+0.300
 needed+0.294
 meant+0.276
 spoke+0.273
 smelled+0.265
Neg logit contributions
n-0.367
pieces-0.343
 victory-0.332
 spend-0.327
 join-0.326
Token
Token ID220
Feature activation-1.92e-01

Pos logit contributions
 whispered+0.269
 meant+0.265
 needed+0.259
 faded+0.247
chy+0.246
Neg logit contributions
 disl-0.233
n-0.231
 spend-0.214
 hey-0.212
 yeah-0.210
Token
Token ID198
Feature activation-2.02e-01

Pos logit contributions
 whispered+0.189
 needed+0.186
 meant+0.176
 faded+0.162
 spoke+0.158
Neg logit contributions
 disl-0.276
n-0.275
 hey-0.270
pieces-0.268
ric-0.265
Token
Token ID198
Feature activation-2.04e-01

Pos logit contributions
 needed+0.222
 whispered+0.217
 meant+0.213
 faded+0.197
chy+0.185
Neg logit contributions
n-0.279
 disl-0.270
 hey-0.260
 blast-0.257
pieces-0.257
TokenHis
Token ID6653
Feature activation-2.52e-01

Pos logit contributions
 whispered+0.237
 needed+0.234
 meant+0.224
 said+0.206
 faded+0.204
Neg logit contributions
 disl-0.259
pieces-0.253
 hey-0.253
n-0.250
ric-0.248
Token and
Token ID290
Feature activation-2.28e-01

Pos logit contributions
 whispered+0.300
 needed+0.297
 meant+0.296
chy+0.285
 spoke+0.268
Neg logit contributions
 victory-0.260
 join-0.252
n-0.250
pieces-0.246
 spend-0.244
Token opened
Token ID4721
Feature activation-2.33e-01

Pos logit contributions
 whispered+0.255
 needed+0.249
 meant+0.232
 said+0.227
 spoke+0.214
Neg logit contributions
 disl-0.314
 hey-0.310
pieces-0.307
ric-0.305
iu-0.303
Token the
Token ID262
Feature activation-2.32e-01

Pos logit contributions
 whispered+0.261
 needed+0.248
 meant+0.241
 said+0.231
 dark+0.220
Neg logit contributions
 disl-0.317
 hey-0.312
pieces-0.303
n-0.302
iu-0.298
Token box
Token ID3091
Feature activation-2.44e-01

Pos logit contributions
 whispered+0.213
 needed+0.211
 meant+0.199
 said+0.179
 faded+0.179
Neg logit contributions
 disl-0.355
n-0.345
 hey-0.344
pieces-0.342
iu-0.336
Token.
Token ID13
Feature activation-2.09e-01

Pos logit contributions
 needed+0.325
chy+0.325
 meant+0.320
 White+0.309
opy+0.308
Neg logit contributions
pieces-0.230
ap-0.225
n-0.223
 join-0.222
 victory-0.221
Token Cheap
Token ID41087
Feature activation-2.74e-01

Pos logit contributions
 whispered+0.233
 needed+0.232
 meant+0.219
 spoke+0.200
 faded+0.199
Neg logit contributions
 disl-0.277
n-0.267
pieces-0.267
 hey-0.266
 blast-0.261
Token flew
Token ID13112
Feature activation-2.48e-01

Pos logit contributions
 whispered+0.349
 needed+0.338
 meant+0.315
 spoke+0.304
 Jones+0.301
Neg logit contributions
 disl-0.314
n-0.312
pieces-0.302
ric-0.299
 interests-0.297
Token out
Token ID503
Feature activation-2.51e-01

Pos logit contributions
 whispered+0.339
 needed+0.332
 meant+0.314
 spoke+0.299
 Jones+0.296
Neg logit contributions
 victory-0.251
n-0.248
 our-0.244
nels-0.238
pieces-0.236
Token and
Token ID290
Feature activation-2.44e-01

Pos logit contributions
 whispered+0.382
 needed+0.379
 meant+0.375
 spoke+0.348
 Jones+0.346
Neg logit contributions
 join-0.214
 our-0.214
pieces-0.208
 victory-0.206
 blast-0.205
Token circled
Token ID48149
Feature activation-2.60e-01

Pos logit contributions
 whispered+0.226
 needed+0.222
 meant+0.210
 said+0.189
 faded+0.186
Neg logit contributions
 disl-0.370
 hey-0.362
n-0.361
pieces-0.360
ric-0.354
Token around
Token ID1088
Feature activation-2.54e-01

Pos logit contributions
 whispered+0.388
 needed+0.382
 meant+0.350
chy+0.347
 spoke+0.344
Neg logit contributions
 victory-0.229
 our-0.225
n-0.220
pieces-0.210
nels-0.206
Token it
Token ID340
Feature activation-2.57e-01

Pos logit contributions
 whispered+0.264
 needed+0.263
 meant+0.247
 said+0.231
 faded+0.229
Neg logit contributions
 disl-0.342
n-0.327
pieces-0.326
ric-0.326
 hey-0.326
Token was
Token ID373
Feature activation-2.49e-01

Pos logit contributions
 whispered+0.215
 needed+0.206
 meant+0.193
 Jones+0.175
 spoke+0.167
Neg logit contributions
n-0.401
 hey-0.387
 blast-0.386
 disl-0.385
pieces-0.381
Token time
Token ID640
Feature activation-2.48e-01

Pos logit contributions
 needed+0.315
 whispered+0.313
 meant+0.303
 dark+0.284
 faded+0.280
Neg logit contributions
 disl-0.310
pieces-0.298
n-0.290
 hey-0.289
ric-0.281
Token to
Token ID284
Feature activation-2.51e-01

Pos logit contributions
 whispered+0.324
 needed+0.318
 meant+0.311
 faded+0.287
 spoke+0.287
Neg logit contributions
n-0.268
 disl-0.265
 hey-0.260
pieces-0.259
iu-0.253
Token go
Token ID467
Feature activation-2.60e-01

Pos logit contributions
 said+0.279
 whispered+0.276
 dark+0.261
 meant+0.243
 closed+0.236
Neg logit contributions
 interests-0.368
 disl-0.364
earned-0.364
 yeah-0.361
pieces-0.360
Token home
Token ID1363
Feature activation-2.74e-01

Pos logit contributions
 needed+0.226
 whispered+0.223
 said+0.210
 dark+0.197
 meant+0.189
Neg logit contributions
 disl-0.456
 hey-0.436
 Majesty-0.433
pieces-0.428
iu-0.425
Token.
Token ID13
Feature activation-2.24e-01

Pos logit contributions
 needed+0.283
 whispered+0.277
 meant+0.265
 spoke+0.249
 shone+0.248
Neg logit contributions
n-0.364
 disl-0.361
 hey-0.351
pieces-0.345
 spend-0.342
Token Tom
Token ID4186
Feature activation-2.58e-01

Pos logit contributions
 whispered+0.230
 needed+0.227
 meant+0.216
 said+0.194
 faded+0.193
Neg logit contributions
 disl-0.319
 hey-0.311
n-0.311
pieces-0.310
ric-0.302
Token and
Token ID290
Feature activation-2.53e-01

Pos logit contributions
 pods+0.279
oky+0.279
 darker+0.266
Pr+0.263
elly+0.263
Neg logit contributions
n-0.334
 blast-0.298
 pause-0.291
 spend-0.288
 stay-0.285
Token Anna
Token ID11735
Feature activation-2.62e-01

Pos logit contributions
 whispered+0.304
 needed+0.301
 meant+0.289
 said+0.264
 faded+0.264
Neg logit contributions
 disl-0.314
n-0.305
 hey-0.305
pieces-0.305
ric-0.295
Token invited
Token ID9392
Feature activation-2.61e-01

Pos logit contributions
elly+0.327
oky+0.322
 crooked+0.304
 blank+0.303
 meant+0.302
Neg logit contributions
 spend-0.322
n-0.309
 stay-0.306
 join-0.305
 pause-0.302
Token how
Token ID703
Feature activation-2.10e-01

Pos logit contributions
 needed+0.385
 whispered+0.385
 meant+0.372
 said+0.351
 faded+0.344
Neg logit contributions
 hey-0.196
 disl-0.194
iu-0.192
n-0.188
pieces-0.187
Token to
Token ID284
Feature activation-2.32e-01

Pos logit contributions
 needed+0.237
 whispered+0.225
 meant+0.214
 said+0.200
 faded+0.195
Neg logit contributions
ric-0.281
pieces-0.279
 hey-0.278
n-0.277
iu-0.276
Token fold
Token ID5591
Feature activation-2.39e-01

Pos logit contributions
 whispered+0.276
 said+0.256
 needed+0.250
 meant+0.235
chy+0.226
Neg logit contributions
iu-0.310
pieces-0.307
earned-0.303
latable-0.296
rd-0.292
Token his
Token ID465
Feature activation-2.35e-01

Pos logit contributions
 whispered+0.315
 needed+0.314
 meant+0.300
 faded+0.279
 spoke+0.279
Neg logit contributions
 disl-0.262
pieces-0.256
n-0.255
 hey-0.247
 interests-0.241
Token wings
Token ID12098
Feature activation-2.33e-01

Pos logit contributions
 whispered+0.340
 needed+0.336
 meant+0.331
 faded+0.308
 spoke+0.305
Neg logit contributions
 disl-0.226
n-0.219
pieces-0.218
 interests-0.212
 blast-0.210
Token properly
Token ID6105
Feature activation-2.73e-01

Pos logit contributions
 whispered+0.243
 needed+0.240
 meant+0.223
 spoke+0.209
 faded+0.206
Neg logit contributions
 disl-0.312
n-0.311
pieces-0.306
 victory-0.298
 spend-0.295
Token so
Token ID523
Feature activation-2.29e-01

Pos logit contributions
 whispered+0.315
 needed+0.313
 spoke+0.293
 meant+0.291
 smelled+0.287
Neg logit contributions
n-0.340
 victory-0.320
 disl-0.311
pieces-0.311
 spend-0.305
Token he
Token ID339
Feature activation-2.43e-01

Pos logit contributions
 whispered+0.243
 needed+0.233
 meant+0.227
 spoke+0.203
 said+0.201
Neg logit contributions
 disl-0.315
 hey-0.309
n-0.306
pieces-0.304
 victory-0.295
Token could
Token ID714
Feature activation-2.35e-01

Pos logit contributions
 whispered+0.161
 Jones+0.156
 aisle+0.150
 dark+0.148
 meant+0.143
Neg logit contributions
n-0.408
 blast-0.379
 disl-0.378
 interests-0.378
 spend-0.376
Token fly
Token ID6129
Feature activation-2.48e-01

Pos logit contributions
 whispered+0.311
 needed+0.292
 said+0.289
chy+0.277
 meant+0.276
Neg logit contributions
iu-0.283
 disl-0.281
pieces-0.279
latable-0.268
nels-0.267
Token.
Token ID13
Feature activation-2.11e-01

Pos logit contributions
 whispered+0.251
 needed+0.248
 meant+0.237
 said+0.217
 faded+0.215
Neg logit contributions
 hey-0.351
 disl-0.350
iu-0.344
pieces-0.342
n-0.340
Token different
Token ID1180
Feature activation-2.40e-01

Pos logit contributions
 whispered+0.267
 needed+0.264
 meant+0.258
 spoke+0.233
 said+0.228
Neg logit contributions
 disl-0.294
 blast-0.280
n-0.280
ric-0.278
 Majesty-0.278
Token body
Token ID1767
Feature activation-2.54e-01

Pos logit contributions
 whispered+0.294
 needed+0.289
 meant+0.275
 faded+0.259
 spoke+0.258
Neg logit contributions
 disl-0.283
n-0.274
 interests-0.269
pieces-0.267
ric-0.264
Token and
Token ID290
Feature activation-2.40e-01

Pos logit contributions
 whispered+0.288
 needed+0.278
 meant+0.266
 spoke+0.254
 faded+0.246
Neg logit contributions
 disl-0.323
n-0.323
ric-0.310
pieces-0.306
 hey-0.301
Token a
Token ID257
Feature activation-2.39e-01

Pos logit contributions
 whispered+0.273
 needed+0.272
 meant+0.258
 said+0.238
 faded+0.235
Neg logit contributions
 hey-0.310
pieces-0.308
 disl-0.308
n-0.306
iu-0.299
Token different
Token ID1180
Feature activation-2.44e-01

Pos logit contributions
 whispered+0.264
 needed+0.262
 meant+0.256
 spoke+0.229
 said+0.225
Neg logit contributions
 disl-0.310
 blast-0.294
 Majesty-0.294
ric-0.293
n-0.291
Token life
Token ID1204
Feature activation-2.73e-01

Pos logit contributions
 whispered+0.301
 needed+0.296
 meant+0.284
 faded+0.263
 spoke+0.263
Neg logit contributions
 disl-0.289
n-0.279
pieces-0.274
ric-0.270
 interests-0.270
Token.
Token ID13
Feature activation-2.19e-01

Pos logit contributions
 whispered+0.291
 needed+0.287
 meant+0.272
 faded+0.247
 said+0.247
Neg logit contributions
 disl-0.373
n-0.363
pieces-0.361
 hey-0.358
ric-0.350
Token It
Token ID632
Feature activation-2.50e-01

Pos logit contributions
 whispered+0.249
 needed+0.241
 meant+0.239
 said+0.217
 Jones+0.207
Neg logit contributions
 disl-0.282
pieces-0.278
 hey-0.276
 arrest-0.275
n-0.271
Token can
Token ID460
Feature activation-2.50e-01

Pos logit contributions
 whispered+0.279
 needed+0.276
 meant+0.263
 faded+0.241
 said+0.238
Neg logit contributions
 disl-0.323
n-0.316
 hey-0.313
pieces-0.311
ric-0.305
Token act
Token ID719
Feature activation-2.63e-01

Pos logit contributions
 whispered+0.350
 needed+0.348
 meant+0.335
 said+0.311
 faded+0.310
Neg logit contributions
 disl-0.257
n-0.249
 hey-0.249
pieces-0.248
ric-0.239
Token in
Token ID287
Feature activation-2.55e-01

Pos logit contributions
 needed+0.332
 whispered+0.330
 meant+0.315
 Jones+0.296
 spoke+0.296
Neg logit contributions
 disl-0.302
n-0.296
pieces-0.293
 blast-0.284
 interests-0.284
Token Soon
Token ID15894
Feature activation-2.54e-01

Pos logit contributions
 needed+0.214
 whispered+0.214
 meant+0.204
 said+0.184
 faded+0.181
Neg logit contributions
 disl-0.285
 hey-0.278
pieces-0.277
n-0.277
iu-0.268
Token they
Token ID484
Feature activation-2.40e-01

Pos logit contributions
 needed+0.245
 whispered+0.233
 meant+0.230
 said+0.213
 faded+0.206
Neg logit contributions
 disl-0.395
ric-0.387
pieces-0.375
 hey-0.374
n-0.365
Token created
Token ID2727
Feature activation-2.52e-01

Pos logit contributions
 whispered+0.229
 meant+0.202
 needed+0.197
 Jones+0.196
 White+0.189
Neg logit contributions
 blast-0.348
pieces-0.347
n-0.345
 spend-0.344
 hey-0.340
Token a
Token ID257
Feature activation-2.38e-01

Pos logit contributions
 needed+0.252
 whispered+0.250
 meant+0.244
 said+0.230
 smelled+0.226
Neg logit contributions
 disl-0.368
 hey-0.354
 arrest-0.348
n-0.342
iu-0.341
Token delicious
Token ID12625
Feature activation-2.50e-01

Pos logit contributions
 needed+0.253
 whispered+0.252
 meant+0.235
 said+0.222
 smelled+0.217
Neg logit contributions
 disl-0.326
n-0.317
 hey-0.314
pieces-0.308
 arrest-0.307
Token treat
Token ID2190
Feature activation-2.73e-01

Pos logit contributions
 needed+0.273
 smelled+0.255
 whispered+0.254
 meant+0.254
 said+0.252
Neg logit contributions
 disl-0.343
 hey-0.331
pieces-0.328
ric-0.323
 interests-0.318
Token.
Token ID13
Feature activation-2.18e-01

Pos logit contributions
 whispered+0.336
 needed+0.325
 meant+0.304
 faded+0.296
 spoke+0.292
Neg logit contributions
n-0.326
pieces-0.325
 hey-0.305
 disl-0.305
 blast-0.302
Token Billy
Token ID15890
Feature activation-2.50e-01

Pos logit contributions
 whispered+0.250
 needed+0.249
 meant+0.238
 said+0.217
 faded+0.215
Neg logit contributions
 disl-0.281
 hey-0.275
pieces-0.274
n-0.273
ric-0.265
Token was
Token ID373
Feature activation-2.57e-01

Pos logit contributions
 whispered+0.247
 dark+0.243
 faded+0.235
 meant+0.233
 Jones+0.232
Neg logit contributions
n-0.332
 disl-0.319
 spend-0.305
 blast-0.297
pieces-0.295
Token so
Token ID523
Feature activation-2.45e-01

Pos logit contributions
 needed+0.232
 whispered+0.226
 meant+0.220
 said+0.203
 smelled+0.191
Neg logit contributions
 disl-0.411
nels-0.398
 hey-0.396
 arrest-0.391
n-0.385
Token proud
Token ID6613
Feature activation-2.45e-01

Pos logit contributions
 whispered+0.239
 needed+0.231
 meant+0.222
 faded+0.200
 spoke+0.199
Neg logit contributions
 disl-0.358
n-0.349
 hey-0.348
pieces-0.347
 blast-0.338
Token glad
Token ID9675
Feature activation-2.21e-01

Pos logit contributions
 whispered+0.126
 needed+0.124
 meant+0.110
 said+0.093
 faded+0.089
Neg logit contributions
 disl-0.419
pieces-0.419
n-0.415
 hey-0.413
ric-0.409
Token to
Token ID284
Feature activation-2.30e-01

Pos logit contributions
 whispered+0.282
 needed+0.276
 meant+0.268
 faded+0.245
 spoke+0.243
Neg logit contributions
 disl-0.252
n-0.248
 hey-0.246
pieces-0.239
 arrest-0.236
Token get
Token ID651
Feature activation-2.43e-01

Pos logit contributions
 whispered+0.290
 needed+0.288
 meant+0.278
 said+0.273
 dark+0.256
Neg logit contributions
pieces-0.298
latable-0.286
 interests-0.281
iu-0.277
nels-0.276
Token back
Token ID736
Feature activation-2.39e-01

Pos logit contributions
 whispered+0.286
 needed+0.286
 meant+0.268
 said+0.256
 spoke+0.245
Neg logit contributions
 disl-0.317
pieces-0.309
ric-0.307
 hey-0.303
latable-0.303
Token home
Token ID1363
Feature activation-2.54e-01

Pos logit contributions
 whispered+0.246
 needed+0.244
 meant+0.228
 said+0.213
 dark+0.204
Neg logit contributions
 disl-0.357
 hey-0.344
pieces-0.344
iu-0.336
ric-0.333
Token safely
Token ID11512
Feature activation-2.73e-01

Pos logit contributions
 whispered+0.281
 needed+0.271
 meant+0.269
 spoke+0.246
 shone+0.239
Neg logit contributions
n-0.325
 disl-0.312
 victory-0.312
 arrest-0.311
 hey-0.305
Token!
Token ID0
Feature activation-2.10e-01

Pos logit contributions
 whispered+0.295
 meant+0.289
 needed+0.288
 smelled+0.271
 shone+0.263
Neg logit contributions
n-0.360
 arrest-0.333
 victory-0.326
 spend-0.325
pieces-0.315
Token<|endoftext|>
Token ID50256
Feature activation-1.91e-01

Pos logit contributions
 whispered+0.223
 needed+0.212
 meant+0.205
 faded+0.194
 spoke+0.187
Neg logit contributions
 disl-0.286
n-0.284
 hey-0.275
ric-0.272
pieces-0.271
TokenOnce
Token ID7454
Feature activation-2.12e-01

Pos logit contributions
 whispered+0.242
 needed+0.231
 faded+0.221
 meant+0.220
chy+0.211
Neg logit contributions
n-0.226
 blast-0.205
ric-0.200
 join-0.198
ext-0.190
Token upon
Token ID2402
Feature activation-2.01e-01

Pos logit contributions
 needed+0.201
 whispered+0.199
 meant+0.187
 said+0.172
 faded+0.163
Neg logit contributions
 disl-0.320
pieces-0.312
 hey-0.310
n-0.307
iu-0.302
Token a
Token ID257
Feature activation-2.34e-01

Pos logit contributions
 needed+0.107
 dark+0.107
 whispered+0.102
 said+0.095
 meant+0.094
Neg logit contributions
 Majesty-0.427
nels-0.412
pieces-0.411
 disl-0.410
han-0.409
Token you
Token ID345
Feature activation-2.39e-01

Pos logit contributions
 said+0.260
 needed+0.246
 whispered+0.239
s+0.221
 meant+0.219
Neg logit contributions
ric-0.323
 Majesty-0.317
 disl-0.314
latable-0.304
 former-0.294
Token,
Token ID11
Feature activation-2.30e-01

Pos logit contributions
 whispered+0.156
 needed+0.156
 said+0.154
 meant+0.139
 dark+0.124
Neg logit contributions
 disl-0.454
pieces-0.427
ric-0.418
 interests-0.414
n-0.414
Token Ben
Token ID3932
Feature activation-2.42e-01

Pos logit contributions
 said+0.186
 whispered+0.177
 needed+0.159
 meant+0.154
 dark+0.137
Neg logit contributions
earned-0.417
pieces-0.414
ric-0.411
 occasions-0.408
 interests-0.402
Token.
Token ID13
Feature activation-2.15e-01

Pos logit contributions
 whispered+0.332
 needed+0.329
 meant+0.317
 said+0.294
 faded+0.293
Neg logit contributions
 disl-0.256
n-0.248
 hey-0.248
pieces-0.248
ric-0.239
Token You
Token ID921
Feature activation-2.34e-01

Pos logit contributions
 needed+0.226
 meant+0.209
 whispered+0.201
 Jones+0.200
 White+0.196
Neg logit contributions
pieces-0.312
ieved-0.288
n-0.285
 disl-0.284
 interests-0.282
Token saved
Token ID7448
Feature activation-2.73e-01

Pos logit contributions
 whispered+0.328
 needed+0.322
 meant+0.310
 faded+0.292
 said+0.288
Neg logit contributions
 disl-0.240
n-0.235
 hey-0.235
pieces-0.229
 blast-0.224
Token me
Token ID502
Feature activation-2.58e-01

Pos logit contributions
 whispered+0.323
 dark+0.321
 said+0.321
 meant+0.313
 needed+0.312
Neg logit contributions
pieces-0.344
 disl-0.340
earned-0.325
 yeah-0.316
 Majesty-0.314
Token,"
Token ID553
Feature activation-2.22e-01

Pos logit contributions
 whispered+0.298
 needed+0.298
 meant+0.283
 faded+0.262
 spoke+0.259
Neg logit contributions
 disl-0.326
n-0.325
 hey-0.318
pieces-0.317
 arrest-0.307
Token Sam
Token ID3409
Feature activation-2.56e-01

Pos logit contributions
 whispered+0.249
 needed+0.246
 meant+0.236
 faded+0.214
 said+0.212
Neg logit contributions
 disl-0.293
 hey-0.285
n-0.285
pieces-0.285
ric-0.276
Token says
Token ID1139
Feature activation-2.48e-01

Pos logit contributions
 whispered+0.178
 needed+0.175
 meant+0.163
 said+0.138
 faded+0.137
Neg logit contributions
 disl-0.443
 hey-0.435
n-0.435
pieces-0.434
ric-0.425
Token.
Token ID13
Feature activation-2.28e-01

Pos logit contributions
 whispered+0.204
 needed+0.202
 meant+0.190
 faded+0.165
 said+0.163
Neg logit contributions
 disl-0.395
n-0.389
 hey-0.388
pieces-0.388
iu-0.379
Token back
Token ID736
Feature activation-2.47e-01

Pos logit contributions
 needed+0.226
 whispered+0.215
 meant+0.200
 dark+0.194
 hungry+0.187
Neg logit contributions
 disl-0.398
iu-0.371
latable-0.363
n-0.361
 hey-0.355
Token.
Token ID13
Feature activation-2.10e-01

Pos logit contributions
 whispered+0.267
 needed+0.261
 meant+0.251
 faded+0.228
 said+0.226
Neg logit contributions
 disl-0.329
n-0.325
pieces-0.324
 hey-0.324
ric-0.311
Token He
Token ID679
Feature activation-2.37e-01

Pos logit contributions
 whispered+0.228
 needed+0.224
 meant+0.215
 faded+0.198
 spoke+0.192
Neg logit contributions
 disl-0.279
n-0.276
 hey-0.273
pieces-0.268
ric-0.266
Token had
Token ID550
Feature activation-2.41e-01

Pos logit contributions
 Jones+0.247
 dark+0.246
 blank+0.245
 aisle+0.243
oky+0.239
Neg logit contributions
n-0.312
 spend-0.278
 blast-0.268
ieved-0.265
 pause-0.263
Token to
Token ID284
Feature activation-2.50e-01

Pos logit contributions
 whispered+0.327
 aisle+0.326
 Jones+0.313
 needed+0.312
opy+0.304
Neg logit contributions
n-0.246
 spend-0.232
 victory-0.228
 disl-0.226
 interests-0.219
Token leave
Token ID2666
Feature activation-2.72e-01

Pos logit contributions
 whispered+0.344
 said+0.327
 meant+0.312
 needed+0.303
 closed+0.278
Neg logit contributions
pieces-0.314
ric-0.293
nels-0.283
 disl-0.280
ext-0.273
Token the
Token ID262
Feature activation-2.47e-01

Pos logit contributions
 whispered+0.382
 needed+0.367
 meant+0.362
 faded+0.352
chy+0.350
Neg logit contributions
n-0.268
 hey-0.257
 victory-0.255
 arrest-0.254
 disl-0.247
Token wheat
Token ID17135
Feature activation-2.66e-01

Pos logit contributions
 whispered+0.313
 needed+0.295
 meant+0.293
 spoke+0.277
 faded+0.276
Neg logit contributions
 disl-0.284
n-0.283
 hey-0.279
pieces-0.273
 arrest-0.271
Token behind
Token ID2157
Feature activation-2.55e-01

Pos logit contributions
 whispered+0.310
 meant+0.302
 needed+0.297
 spoke+0.271
 faded+0.265
Neg logit contributions
n-0.324
pieces-0.313
 victory-0.312
 arrest-0.312
 disl-0.311
Token,
Token ID11
Feature activation-2.46e-01

Pos logit contributions
 whispered+0.371
 meant+0.338
chy+0.336
 smelled+0.334
 faded+0.332
Neg logit contributions
n-0.253
 arrest-0.237
 victory-0.229
 join-0.220
 our-0.212
Token even
Token ID772
Feature activation-2.62e-01

Pos logit contributions
 whispered+0.258
 meant+0.250
 needed+0.250
 spoke+0.224
 faded+0.222
Neg logit contributions
 disl-0.320
n-0.316
 blast-0.310
 arrest-0.308
 hey-0.308
Token Sam
Token ID3409
Feature activation-2.42e-01

Pos logit contributions
 whispered+0.328
 needed+0.324
 meant+0.307
 faded+0.305
chy+0.296
Neg logit contributions
pieces-0.227
 disl-0.227
 hey-0.226
n-0.224
 victory-0.212
Token if
Token ID611
Feature activation-2.42e-01

Pos logit contributions
 whispered+0.328
 needed+0.319
 faded+0.308
 meant+0.306
 dark+0.301
Neg logit contributions
n-0.251
 disl-0.249
pieces-0.244
 join-0.242
 spend-0.241
Token he
Token ID339
Feature activation-2.46e-01

Pos logit contributions
 needed+0.184
 whispered+0.180
 meant+0.168
 said+0.154
 spoke+0.146
Neg logit contributions
 disl-0.411
ric-0.410
latable-0.409
pieces-0.406
iu-0.399
Token understood
Token ID7247
Feature activation-2.55e-01

Pos logit contributions
 aisle+0.208
 dark+0.193
 blank+0.185
 whispered+0.184
 dusty+0.175
Neg logit contributions
n-0.368
 spend-0.365
 disl-0.358
 interests-0.355
 join-0.348
Token the
Token ID262
Feature activation-2.41e-01

Pos logit contributions
 whispered+0.337
 needed+0.336
 meant+0.317
 faded+0.304
 spoke+0.297
Neg logit contributions
 disl-0.281
n-0.273
pieces-0.272
 hey-0.266
ric-0.259
Token number
Token ID1271
Feature activation-2.72e-01

Pos logit contributions
 whispered+0.308
 needed+0.304
 meant+0.289
 faded+0.271
 said+0.268
Neg logit contributions
 disl-0.281
 hey-0.273
n-0.272
pieces-0.271
 arrest-0.259
Token rules
Token ID3173
Feature activation-2.57e-01

Pos logit contributions
 whispered+0.313
 needed+0.293
 faded+0.288
chy+0.285
 smelled+0.274
Neg logit contributions
pieces-0.350
 victory-0.324
n-0.321
 spend-0.309
 disl-0.308
Token.
Token ID13
Feature activation-2.20e-01

Pos logit contributions
 whispered+0.318
 smelled+0.297
 shone+0.294
 needed+0.294
 faded+0.293
Neg logit contributions
n-0.285
 join-0.261
pieces-0.254
 victory-0.249
 spend-0.243
Token Sam
Token ID3409
Feature activation-2.57e-01

Pos logit contributions
 whispered+0.296
 needed+0.287
 meant+0.277
 faded+0.276
chy+0.269
Neg logit contributions
 disl-0.247
pieces-0.233
n-0.231
 hey-0.227
ric-0.222
Token nodded
Token ID14464
Feature activation-2.50e-01

Pos logit contributions
 dark+0.222
chy+0.212
 dusty+0.204
 blank+0.199
 faded+0.199
Neg logit contributions
n-0.373
 disl-0.355
 spend-0.349
ieved-0.342
 interests-0.336
Token his
Token ID465
Feature activation-2.58e-01

Pos logit contributions
 needed+0.304
 aisle+0.300
 whispered+0.292
 meant+0.289
chy+0.280
Neg logit contributions
 pause-0.276
 spend-0.269
 join-0.268
ap-0.264
 interests-0.264
Token the
Token ID262
Feature activation-2.22e-01

Pos logit contributions
 whispered+0.287
 needed+0.286
 meant+0.277
 faded+0.254
chy+0.251
Neg logit contributions
 hey-0.238
n-0.235
 disl-0.234
pieces-0.233
 blast-0.224
Token pilot
Token ID8022
Feature activation-2.23e-01

Pos logit contributions
 whispered+0.274
 needed+0.270
 meant+0.262
 faded+0.237
 said+0.236
Neg logit contributions
 hey-0.267
 disl-0.263
n-0.262
 blast-0.259
pieces-0.259
Token for
Token ID329
Feature activation-2.36e-01

Pos logit contributions
 needed+0.265
 whispered+0.264
 meant+0.260
 faded+0.235
 smelled+0.234
Neg logit contributions
pieces-0.272
n-0.269
 hey-0.258
 spend-0.257
 disl-0.254
Token flying
Token ID7348
Feature activation-2.34e-01

Pos logit contributions
 whispered+0.299
 needed+0.296
 meant+0.285
 said+0.262
 faded+0.261
Neg logit contributions
 disl-0.276
 hey-0.268
n-0.268
pieces-0.267
ric-0.259
Token them
Token ID606
Feature activation-2.47e-01

Pos logit contributions
 needed+0.308
 whispered+0.307
 meant+0.292
 spoke+0.269
 faded+0.268
Neg logit contributions
 disl-0.254
n-0.250
 hey-0.244
 victory-0.243
pieces-0.243
Token safely
Token ID11512
Feature activation-2.72e-01

Pos logit contributions
Fuel+0.345
 needed+0.341
 meant+0.337
Ancient+0.335
 whispered+0.332
Neg logit contributions
n-0.238
 victory-0.237
 join-0.229
 pause-0.214
 spend-0.211
Token back
Token ID736
Feature activation-2.43e-01

Pos logit contributions
 needed+0.328
 meant+0.326
 whispered+0.321
 smelled+0.318
Fuel+0.300
Neg logit contributions
n-0.318
 victory-0.302
ap-0.287
 join-0.287
 spend-0.285
Token home
Token ID1363
Feature activation-2.59e-01

Pos logit contributions
 needed+0.279
 whispered+0.278
 meant+0.270
 spoke+0.245
 faded+0.243
Neg logit contributions
n-0.296
 disl-0.292
pieces-0.289
ric-0.288
 hey-0.284
Token and
Token ID290
Feature activation-2.34e-01

Pos logit contributions
 whispered+0.294
 meant+0.290
 needed+0.286
 spoke+0.267
 shone+0.257
Neg logit contributions
n-0.321
 victory-0.306
 disl-0.299
 arrest-0.298
 spend-0.295
Token everyone
Token ID2506
Feature activation-2.51e-01

Pos logit contributions
 whispered+0.247
 needed+0.245
 meant+0.234
 faded+0.210
 spoke+0.207
Neg logit contributions
 disl-0.316
 hey-0.311
n-0.310
pieces-0.307
ric-0.298
Token hugged
Token ID41130
Feature activation-2.51e-01

Pos logit contributions
 whispered+0.232
 Jones+0.231
 meant+0.229
 aisle+0.225
 needed+0.222
Neg logit contributions
n-0.340
 hey-0.326
 join-0.325
 disl-0.323
 spend-0.319
Token gift
Token ID6979
Feature activation-2.38e-01

Pos logit contributions
 whispered+0.274
 needed+0.265
 meant+0.255
 spoke+0.238
 said+0.235
Neg logit contributions
 disl-0.270
n-0.269
 hey-0.266
 victory-0.263
 blast-0.260
Token of
Token ID286
Feature activation-2.28e-01

Pos logit contributions
 whispered+0.229
 bored+0.199
 needed+0.198
 meant+0.190
 Jones+0.185
Neg logit contributions
n-0.342
 victory-0.313
 spend-0.311
 blast-0.308
 hey-0.305
Token running
Token ID2491
Feature activation-2.33e-01

Pos logit contributions
 whispered+0.262
 needed+0.262
 meant+0.255
 said+0.235
 faded+0.224
Neg logit contributions
 hey-0.298
pieces-0.297
 disl-0.291
iu-0.284
n-0.283
Token and
Token ID290
Feature activation-2.32e-01

Pos logit contributions
 whispered+0.263
 needed+0.248
 meant+0.245
 spoke+0.221
 Jones+0.221
Neg logit contributions
 disl-0.297
n-0.296
 arrest-0.288
pieces-0.287
 victory-0.285
Token jumping
Token ID14284
Feature activation-2.40e-01

Pos logit contributions
 whispered+0.291
 needed+0.278
 meant+0.272
 spoke+0.249
 faded+0.249
Neg logit contributions
 disl-0.269
n-0.262
 hey-0.256
 arrest-0.255
ric-0.254
Token freely
Token ID12748
Feature activation-2.72e-01

Pos logit contributions
 whispered+0.317
 needed+0.306
 meant+0.295
 Jones+0.281
 spoke+0.272
Neg logit contributions
n-0.264
 victory-0.259
 disl-0.248
 blast-0.242
pieces-0.241
Token.
Token ID13
Feature activation-2.13e-01

Pos logit contributions
 whispered+0.306
 needed+0.295
 meant+0.283
 spoke+0.257
 faded+0.256
Neg logit contributions
n-0.357
 disl-0.351
pieces-0.335
 hey-0.332
ric-0.329
Token<|endoftext|>
Token ID50256
Feature activation-1.96e-01

Pos logit contributions
 needed+0.174
 whispered+0.167
 meant+0.159
 said+0.141
 Jones+0.133
Neg logit contributions
pieces-0.351
 disl-0.350
 hey-0.346
n-0.337
nels-0.335
TokenOnce
Token ID7454
Feature activation-2.17e-01

Pos logit contributions
 whispered+0.232
 needed+0.211
 faded+0.208
 meant+0.206
 crooked+0.199
Neg logit contributions
n-0.252
ric-0.224
 blast-0.224
 join-0.221
 hey-0.208
Token upon
Token ID2402
Feature activation-2.07e-01

Pos logit contributions
 needed+0.206
 whispered+0.202
 meant+0.190
 said+0.176
 faded+0.166
Neg logit contributions
 disl-0.328
pieces-0.320
 hey-0.319
n-0.315
 arrest-0.311
Token a
Token ID257
Feature activation-2.38e-01

Pos logit contributions
 needed+0.145
 whispered+0.141
 meant+0.133
 dark+0.131
 said+0.126
Neg logit contributions
 Majesty-0.390
 disl-0.387
nels-0.380
pieces-0.377
han-0.376
TokenThey
Token ID2990
Feature activation-2.29e-01

Pos logit contributions
 whispered+0.238
 needed+0.225
 said+0.213
 meant+0.211
 faded+0.200
Neg logit contributions
 disl-0.265
n-0.258
pieces-0.255
ric-0.251
iu-0.249
Token washed
Token ID18989
Feature activation-2.57e-01

Pos logit contributions
 meant+0.241
 whispered+0.233
 Jones+0.231
 needed+0.222
 White+0.218
Neg logit contributions
 spend-0.292
 blast-0.289
 hey-0.288
n-0.280
iu-0.275
Token the
Token ID262
Feature activation-2.40e-01

Pos logit contributions
 whispered+0.284
 needed+0.281
 meant+0.271
 said+0.250
 faded+0.239
Neg logit contributions
 disl-0.351
 hey-0.337
pieces-0.336
n-0.336
ric-0.329
Token fruits
Token ID15921
Feature activation-2.53e-01

Pos logit contributions
 needed+0.319
 whispered+0.307
 meant+0.295
 said+0.283
 smelled+0.279
Neg logit contributions
 disl-0.274
n-0.269
pieces-0.265
 hey-0.262
latable-0.260
Token and
Token ID290
Feature activation-2.43e-01

Pos logit contributions
 whispered+0.267
 needed+0.258
 meant+0.251
 spoke+0.240
 faded+0.224
Neg logit contributions
n-0.334
pieces-0.318
 victory-0.313
iu-0.312
 pause-0.310
Token cut
Token ID2005
Feature activation-2.72e-01

Pos logit contributions
 whispered+0.240
 needed+0.240
 meant+0.225
 said+0.205
 faded+0.200
Neg logit contributions
 disl-0.356
pieces-0.347
 hey-0.347
n-0.345
ric-0.339
Token them
Token ID606
Feature activation-2.64e-01

Pos logit contributions
 whispered+0.426
 needed+0.417
 meant+0.395
 spoke+0.386
 faded+0.380
Neg logit contributions
n-0.234
pieces-0.231
 hey-0.222
 disl-0.218
ric-0.217
Token.
Token ID13
Feature activation-2.24e-01

Pos logit contributions
 whispered+0.340
 needed+0.336
 meant+0.322
 spoke+0.306
 faded+0.299
Neg logit contributions
n-0.303
pieces-0.288
 disl-0.287
 hey-0.280
 victory-0.279
Token They
Token ID1119
Feature activation-2.44e-01

Pos logit contributions
 whispered+0.297
 needed+0.292
 meant+0.283
 spoke+0.274
chy+0.260
Neg logit contributions
 blast-0.250
 disl-0.247
latable-0.239
 interests-0.238
n-0.232
Token put
Token ID1234
Feature activation-2.61e-01

Pos logit contributions
 Jones+0.270
elly+0.248
 White+0.245
 aisle+0.241
 wicked+0.237
Neg logit contributions
 spend-0.306
n-0.285
 blast-0.285
 hey-0.273
 unite-0.271
Token them
Token ID606
Feature activation-2.72e-01

Pos logit contributions
 needed+0.372
 whispered+0.367
 meant+0.358
 said+0.338
 faded+0.329
Neg logit contributions
 disl-0.274
 hey-0.255
pieces-0.253
n-0.251
 Majesty-0.249
Token the
Token ID262
Feature activation-2.25e-01

Pos logit contributions
 whispered+0.269
 needed+0.262
 meant+0.256
 said+0.240
 faded+0.224
Neg logit contributions
 hey-0.297
 disl-0.295
n-0.292
pieces-0.291
 Majesty-0.288
Token mineral
Token ID18352
Feature activation-2.39e-01

Pos logit contributions
 whispered+0.275
 needed+0.267
 meant+0.258
 faded+0.239
 spoke+0.238
Neg logit contributions
 disl-0.278
n-0.267
 hey-0.260
pieces-0.258
 interests-0.254
Token and
Token ID290
Feature activation-2.28e-01

Pos logit contributions
 needed+0.219
 whispered+0.218
 meant+0.203
 said+0.188
 faded+0.177
Neg logit contributions
 disl-0.360
 hey-0.359
ric-0.353
pieces-0.351
n-0.350
Token play
Token ID711
Feature activation-2.42e-01

Pos logit contributions
 said+0.309
 needed+0.282
 whispered+0.270
 had+0.247
 meant+0.243
Neg logit contributions
ocol-0.368
 Deal-0.364
ric-0.345
 Majesty-0.339
latable-0.338
Token together
Token ID1978
Feature activation-2.41e-01

Pos logit contributions
 whispered+0.242
 needed+0.241
 meant+0.221
 said+0.220
 dark+0.202
Neg logit contributions
ric-0.360
 disl-0.346
iu-0.338
 Majesty-0.333
 hey-0.331
Token nicely
Token ID16576
Feature activation-2.72e-01

Pos logit contributions
 whispered+0.301
 needed+0.288
 meant+0.286
 spoke+0.264
 faded+0.260
Neg logit contributions
n-0.294
 victory-0.280
pieces-0.277
 hey-0.270
 join-0.269
Token.
Token ID13
Feature activation-2.08e-01

Pos logit contributions
 whispered+0.309
 needed+0.292
 meant+0.282
 faded+0.276
 spoke+0.271
Neg logit contributions
n-0.356
 disl-0.344
pieces-0.343
 victory-0.328
 interests-0.327
Token
Token ID198
Feature activation-2.03e-01

Pos logit contributions
 whispered+0.215
 needed+0.210
 meant+0.202
 faded+0.183
 spoke+0.180
Neg logit contributions
 disl-0.294
n-0.288
 hey-0.284
pieces-0.284
ric-0.278
Token
Token ID198
Feature activation-2.05e-01

Pos logit contributions
 whispered+0.213
 needed+0.213
 meant+0.204
 faded+0.189
 spoke+0.181
Neg logit contributions
n-0.281
 disl-0.277
pieces-0.267
 hey-0.266
 blast-0.263
TokenThe
Token ID464
Feature activation-2.41e-01

Pos logit contributions
 whispered+0.267
 needed+0.264
 meant+0.254
 said+0.236
 faded+0.234
Neg logit contributions
 disl-0.232
pieces-0.225
 hey-0.225
n-0.223
ric-0.220
Token moral
Token ID6573
Feature activation-2.33e-01

Pos logit contributions
 whispered+0.285
 needed+0.274
 meant+0.264
 spoke+0.248
 said+0.242
Neg logit contributions
 disl-0.295
pieces-0.291
 hey-0.290
n-0.288
 victory-0.284
Token garden
Token ID11376
Feature activation+4.70e-01

Pos logit contributions
 disl+0.609
pieces+0.590
n+0.590
 hey+0.589
ric+0.585
Neg logit contributions
 whispered-0.663
 needed-0.648
 meant-0.632
 spoke-0.589
 faded-0.582
Token and
Token ID290
Feature activation+4.18e-01

Pos logit contributions
 disl+0.580
n+0.578
ric+0.565
pieces+0.560
 spend+0.559
Neg logit contributions
 whispered-0.521
 meant-0.469
 needed-0.465
 faded-0.450
 spoke-0.449
Token admire
Token ID21099
Feature activation+3.96e-01

Pos logit contributions
 disl+0.620
 hey+0.606
n+0.605
pieces+0.604
ric+0.586
Neg logit contributions
 whispered-0.387
 needed-0.382
 meant-0.364
 faded-0.324
 spoke-0.317
Token his
Token ID465
Feature activation+3.94e-01

Pos logit contributions
 disl+0.449
 hey+0.444
pieces+0.443
n+0.440
iu+0.425
Neg logit contributions
 whispered-0.509
 needed-0.506
 meant-0.488
 said-0.454
 faded-0.445
Token sun
Token ID4252
Feature activation+3.86e-01

Pos logit contributions
 disl+0.449
pieces+0.429
n+0.422
 hey+0.413
 interests+0.412
Neg logit contributions
 whispered-0.513
 needed-0.502
 meant-0.491
 faded-0.453
 spoke-0.452
Tokenfl
Token ID2704
Feature activation+5.33e-01

Pos logit contributions
n+0.515
pieces+0.510
ric+0.506
 disl+0.502
 spend+0.493
Neg logit contributions
 whispered-0.406
 needed-0.395
 meant-0.366
 spoke-0.342
 faded-0.340
Tokenowers
Token ID3618
Feature activation+3.77e-01

Pos logit contributions
 disl+0.860
 hey+0.842
n+0.839
pieces+0.834
iu+0.816
Neg logit contributions
 whispered-0.434
 needed-0.431
 meant-0.404
 said-0.356
 faded-0.348
Token.
Token ID13
Feature activation+3.61e-01

Pos logit contributions
n+0.474
pieces+0.464
ric+0.451
 spend+0.448
 disl+0.447
Neg logit contributions
 whispered-0.412
 needed-0.392
 meant-0.371
 faded-0.362
 spoke-0.356
Token They
Token ID1119
Feature activation+3.37e-01

Pos logit contributions
 disl+0.542
n+0.528
pieces+0.524
 hey+0.523
ric+0.514
Neg logit contributions
 whispered-0.336
 needed-0.332
 meant-0.315
 faded-0.282
 spoke-0.277
Token also
Token ID635
Feature activation+3.27e-01

Pos logit contributions
 disl+0.378
 hey+0.367
 spend+0.365
n+0.363
pieces+0.362
Neg logit contributions
 whispered-0.420
 needed-0.414
 meant-0.410
 faded-0.378
 Jones-0.371
Token see
Token ID766
Feature activation+3.00e-01

Pos logit contributions
 disl+0.420
pieces+0.396
 spend+0.393
 hey+0.393
n+0.391
Neg logit contributions
 whispered-0.365
 needed-0.361
 meant-0.353
 faded-0.322
 Jones-0.319
Token't
Token ID470
Feature activation+5.83e-01

Pos logit contributions
 whispered+0.278
 needed+0.271
 meant+0.261
 faded+0.240
chy+0.237
Neg logit contributions
 disl-0.281
n-0.276
 hey-0.274
pieces-0.266
 yeah-0.260
Token know
Token ID760
Feature activation+4.73e-01

Pos logit contributions
latable+0.673
ric+0.657
nels+0.648
pieces+0.645
iu+0.633
Neg logit contributions
 needed-0.816
 whispered-0.812
 meant-0.786
 said-0.755
 spoke-0.707
Token the
Token ID262
Feature activation+4.73e-01

Pos logit contributions
 disl+0.410
n+0.395
 hey+0.394
pieces+0.394
ric+0.376
Neg logit contributions
 whispered-0.736
 needed-0.730
 meant-0.707
 faded-0.661
 said-0.661
Token rh
Token ID9529
Feature activation+5.33e-01

Pos logit contributions
 disl+0.617
n+0.600
 hey+0.599
pieces+0.597
ric+0.588
Neg logit contributions
 whispered-0.520
 needed-0.502
 meant-0.480
 faded-0.445
 spoke-0.445
Tokenin
Token ID259
Feature activation+4.54e-01

Pos logit contributions
 disl+0.802
n+0.771
pieces+0.763
 hey+0.762
latable+0.750
Neg logit contributions
 needed-0.511
 whispered-0.504
 meant-0.476
 said-0.429
 faded-0.422
Tokenoc
Token ID420
Feature activation+5.77e-01

Pos logit contributions
n+0.381
 disl+0.336
 victory+0.315
pieces+0.308
earned+0.302
Neg logit contributions
 whispered-0.729
 faded-0.682
 needed-0.661
 blank-0.660
 meant-0.660
Tokeneros
Token ID27498
Feature activation+4.32e-01

Pos logit contributions
 blast+0.565
 disl+0.564
n+0.552
pieces+0.551
latable+0.542
Neg logit contributions
 needed-0.821
 whispered-0.799
 meant-0.781
 said-0.729
 Jones-0.721
Token was
Token ID373
Feature activation+3.84e-01

Pos logit contributions
n+0.589
 disl+0.530
 spend+0.522
pieces+0.519
 arrest+0.519
Neg logit contributions
 blank-0.433
 whispered-0.428
 faded-0.394
 dark-0.394
 aisle-0.390
Token wild
Token ID4295
Feature activation+3.79e-01

Pos logit contributions
n+0.456
 victory+0.447
 disl+0.447
ric+0.432
 arrest+0.431
Neg logit contributions
 whispered-0.444
 needed-0.410
 meant-0.394
 Jones-0.382
 aisle-0.380
Token.
Token ID13
Feature activation+3.80e-01

Pos logit contributions
 disl+0.544
n+0.527
 hey+0.523
pieces+0.520
 arrest+0.513
Neg logit contributions
 whispered-0.374
 needed-0.359
 meant-0.333
 faded-0.312
 spoke-0.310
Token Can
Token ID1680
Feature activation+3.68e-01

Pos logit contributions
 disl+0.449
n+0.447
 hey+0.419
ric+0.417
pieces+0.412
Neg logit contributions
 needed-0.460
 whispered-0.460
 meant-0.446
 faded-0.436
chy-0.434
Token
Token ID129
Feature activation+7.19e-01

Pos logit contributions
 needed+0.193
 whispered+0.193
 meant+0.183
chy+0.162
 faded+0.159
Neg logit contributions
 hey-0.375
pieces-0.369
 disl-0.367
n-0.362
iu-0.357
Token
Token ID241
Feature activation+6.40e-01

Pos logit contributions
pieces+0.714
 interests+0.707
 yeah+0.661
 hey+0.659
 victory+0.642
Neg logit contributions
 whispered-1.041
 said-0.980
 meant-0.977
 needed-0.976
 spoke-0.888
TokenIt
Token ID1026
Feature activation+5.06e-01

Pos logit contributions
 disl+0.963
pieces+0.953
 hey+0.953
 Majesty+0.950
 arrest+0.933
Neg logit contributions
 needed-0.588
 whispered-0.574
 meant-0.569
 said-0.503
 spoke-0.466
Tokenâ
Token ID22940
Feature activation+5.35e-01

Pos logit contributions
n+0.607
 disl+0.606
pieces+0.579
ric+0.569
 hey+0.565
Neg logit contributions
 whispered-0.608
 needed-0.575
 faded-0.546
 meant-0.543
 spoke-0.524
Token
Token ID26391
Feature activation+5.77e-01

Pos logit contributions
 disl+0.458
pieces+0.440
 hey+0.439
n+0.431
ric+0.417
Neg logit contributions
 whispered-0.841
 needed-0.836
 meant-0.811
 said-0.760
 faded-0.752
Token
Token ID8151
Feature activation+6.05e-01

Pos logit contributions
 hey+0.654
pieces+0.650
 disl+0.648
iu+0.618
latable+0.617
Neg logit contributions
 needed-0.740
 whispered-0.736
 meant-0.692
 said-0.667
 spoke-0.643
Tokens
Token ID82
Feature activation+4.66e-01

Pos logit contributions
n+0.895
ric+0.885
 disl+0.826
pieces+0.815
ieved+0.796
Neg logit contributions
 whispered-0.605
 faded-0.537
 needed-0.525
 blank-0.511
 meant-0.500
Token a
Token ID257
Feature activation+4.44e-01

Pos logit contributions
 disl+0.526
n+0.517
 hey+0.501
pieces+0.500
ric+0.499
Neg logit contributions
 whispered-0.608
 needed-0.592
 meant-0.568
 faded-0.534
 spoke-0.531
Token magic
Token ID5536
Feature activation+4.06e-01

Pos logit contributions
 disl+0.572
 blast+0.555
 hey+0.554
pieces+0.554
n+0.552
Neg logit contributions
 whispered-0.496
 needed-0.479
 meant-0.473
 spoke-0.421
 faded-0.419
Token book
Token ID1492
Feature activation+3.91e-01

Pos logit contributions
 disl+0.479
n+0.477
 blast+0.460
pieces+0.460
 Majesty+0.458
Neg logit contributions
 whispered-0.485
 needed-0.470
 meant-0.456
 spoke-0.418
 bored-0.412
Token!
Token ID0
Feature activation+3.88e-01

Pos logit contributions
n+0.420
 join+0.354
pieces+0.353
 blast+0.348
 disl+0.346
Neg logit contributions
 whispered-0.498
 needed-0.471
 bored-0.454
 Jones-0.451
 faded-0.450
Token
Token ID26391
Feature activation+7.57e-01

Pos logit contributions
 whispered+0.219
 meant+0.216
 needed+0.213
 faded+0.209
 blank+0.204
Neg logit contributions
n-0.306
 disl-0.290
 arrest-0.287
 spend-0.287
 hey-0.285
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenLet
Token ID5756
Feature activation+4.94e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.17e-01

Pos logit contributions
ric+0.777
iu+0.737
latable+0.733
 hey+0.715
 disl+0.712
Neg logit contributions
 meant-0.511
 whispered-0.505
 said-0.503
 needed-0.492
 spoke-0.425
Token
Token ID26391
Feature activation+5.61e-01

Pos logit contributions
 disl+0.397
 hey+0.396
pieces+0.395
 Majesty+0.389
 arrest+0.381
Neg logit contributions
 whispered-0.830
 needed-0.820
 meant-0.811
 said-0.763
 spoke-0.749
Token
Token ID8151
Feature activation+5.91e-01

Pos logit contributions
 hey+0.690
pieces+0.671
 disl+0.668
latable+0.664
iu+0.657
Neg logit contributions
 needed-0.681
 whispered-0.666
 said-0.626
 meant-0.621
 spoke-0.587
Tokens
Token ID82
Feature activation+4.52e-01

Pos logit contributions
n+0.908
ric+0.876
pieces+0.867
 disl+0.862
 hey+0.836
Neg logit contributions
 whispered-0.545
 needed-0.498
 meant-0.475
 faded-0.471
 dark-0.433
Token play
Token ID711
Feature activation+4.18e-01

Pos logit contributions
iu+0.512
latable+0.486
 disl+0.469
pieces+0.467
ext+0.463
Neg logit contributions
 needed-0.637
 whispered-0.627
 meant-0.625
 said-0.617
 spoke-0.544
Token a
Token ID257
Feature activation+4.10e-01

Pos logit contributions
 disl+0.554
 hey+0.534
iu+0.532
ric+0.531
pieces+0.528
Neg logit contributions
 whispered-0.470
 needed-0.454
 meant-0.437
 said-0.419
 spoke-0.389
Token joke
Token ID9707
Feature activation+3.92e-01

Pos logit contributions
 disl+0.511
 hey+0.494
n+0.493
pieces+0.491
ric+0.481
Neg logit contributions
 whispered-0.487
 needed-0.483
 meant-0.457
 said-0.426
 faded-0.421
Token
Token ID129
Feature activation+6.60e-01

Pos logit contributions
pieces+1.037
 hey+1.019
 disl+1.006
latable+1.000
nels+0.990
Neg logit contributions
 whispered-0.819
 needed-0.813
 said-0.765
 meant-0.728
 spoke-0.673
Token
Token ID241
Feature activation+6.13e-01

Pos logit contributions
pieces+0.605
 hey+0.600
 disl+0.574
 interests+0.569
 blast+0.566
Neg logit contributions
 whispered-0.992
 needed-0.965
 meant-0.944
 said-0.903
 spoke-0.868
TokenWho
Token ID8241
Feature activation+4.95e-01

Pos logit contributions
 disl+0.869
pieces+0.859
 hey+0.854
n+0.833
iu+0.831
Neg logit contributions
 needed-0.611
 whispered-0.604
 meant-0.583
 said-0.524
 faded-0.502
Tokenâ
Token ID22940
Feature activation+5.09e-01

Pos logit contributions
 disl+0.640
n+0.624
 hey+0.624
pieces+0.623
ric+0.603
Neg logit contributions
 whispered-0.557
 needed-0.550
 meant-0.525
 faded-0.479
 said-0.478
Token
Token ID26391
Feature activation+5.56e-01

Pos logit contributions
 disl+0.401
pieces+0.389
 hey+0.387
n+0.376
 arrest+0.366
Neg logit contributions
 whispered-0.830
 needed-0.825
 meant-0.805
 said-0.755
 spoke-0.747
Token
Token ID8151
Feature activation+5.86e-01

Pos logit contributions
 disl+0.661
 hey+0.650
pieces+0.649
n+0.632
iu+0.625
Neg logit contributions
 whispered-0.684
 needed-0.683
 meant-0.649
 said-0.606
 spoke-0.594
Tokens
Token ID82
Feature activation+4.50e-01

Pos logit contributions
n+0.881
ric+0.873
 disl+0.832
pieces+0.828
 hey+0.791
Neg logit contributions
 whispered-0.559
 faded-0.498
 needed-0.489
 blank-0.458
 meant-0.456
Token there
Token ID612
Feature activation+4.29e-01

Pos logit contributions
 disl+0.458
pieces+0.445
 hey+0.441
n+0.438
iu+0.430
Neg logit contributions
 whispered-0.635
 needed-0.635
 meant-0.614
 said-0.572
 spoke-0.560
Token?
Token ID30
Feature activation+4.02e-01

Pos logit contributions
n+0.366
ric+0.344
 hey+0.332
 disl+0.332
 spend+0.331
Neg logit contributions
 whispered-0.673
 needed-0.639
 faded-0.633
 meant-0.605
 spoke-0.603
Tokenâ
Token ID22940
Feature activation+4.09e-01

Pos logit contributions
n+0.435
 disl+0.424
pieces+0.417
 hey+0.414
ric+0.406
Neg logit contributions
 whispered-0.542
 needed-0.531
 meant-0.512
 faded-0.499
chy-0.480
Token
Token ID26391
Feature activation+4.57e-01

Pos logit contributions
n+0.271
 disl+0.261
ric+0.227
 hey+0.218
iu+0.209
Neg logit contributions
 needed-0.749
 whispered-0.742
 meant-0.711
 faded-0.710
 smelled-0.683
Token Liam
Token ID31684
Feature activation+3.63e-01

Pos logit contributions
 disl+0.479
pieces+0.469
 hey+0.469
n+0.469
ric+0.456
Neg logit contributions
 whispered-0.360
 needed-0.355
 meant-0.338
 said-0.308
 faded-0.301
Token and
Token ID290
Feature activation+3.26e-01

Pos logit contributions
n+0.452
pieces+0.446
 disl+0.434
 spend+0.429
 hey+0.428
Neg logit contributions
 faded-0.409
 whispered-0.403
 needed-0.397
 meant-0.396
 shone-0.393
Token asked
Token ID1965
Feature activation+3.28e-01

Pos logit contributions
 disl+0.554
n+0.549
pieces+0.537
 hey+0.537
 spend+0.528
Neg logit contributions
 needed-0.213
 whispered-0.206
 meant-0.197
chy-0.176
 faded-0.173
Token him
Token ID683
Feature activation+3.00e-01

Pos logit contributions
n+0.313
 hey+0.305
pieces+0.298
 interests+0.291
 disl+0.290
Neg logit contributions
 whispered-0.467
 needed-0.458
 faded-0.448
 meant-0.445
 smelled-0.439
Token to
Token ID284
Feature activation+3.00e-01

Pos logit contributions
 disl+0.310
n+0.308
pieces+0.306
 hey+0.300
 interests+0.288
Neg logit contributions
 whispered-0.410
 needed-0.406
 meant-0.391
 faded-0.375
 spoke-0.360
Token help
Token ID1037
Feature activation+3.00e-01

Pos logit contributions
pieces+0.328
ric+0.327
 disl+0.322
latable+0.314
nels+0.313
Neg logit contributions
 whispered-0.435
 needed-0.414
 said-0.411
 meant-0.401
 spoke-0.377
Token her
Token ID607
Feature activation+2.97e-01

Pos logit contributions
n+0.332
pieces+0.311
 spend+0.309
 hey+0.308
 join+0.297
Neg logit contributions
 needed-0.397
 whispered-0.397
 meant-0.381
 faded-0.381
 shone-0.362
Token think
Token ID892
Feature activation+2.94e-01

Pos logit contributions
 disl+0.442
 hey+0.431
latable+0.424
iu+0.423
ric+0.419
Neg logit contributions
 whispered-0.308
 needed-0.303
 meant-0.288
 said-0.265
 spoke-0.241
Token of
Token ID286
Feature activation+2.92e-01

Pos logit contributions
n+0.311
pieces+0.299
 victory+0.279
 spend+0.266
 join+0.265
Neg logit contributions
 whispered-0.393
 Jones-0.367
 needed-0.366
 spoke-0.362
 smelled-0.354
Token a
Token ID257
Feature activation+2.88e-01

Pos logit contributions
n+0.273
 disl+0.271
 victory+0.263
pieces+0.254
 arrest+0.253
Neg logit contributions
 whispered-0.425
 needed-0.418
 meant-0.400
 spoke-0.393
 faded-0.389
Token creative
Token ID7325
Feature activation+3.04e-01

Pos logit contributions
 victory+0.334
 blast+0.320
 Majesty+0.317
pieces+0.313
 pause+0.312
Neg logit contributions
 whispered-0.361
 meant-0.355
 needed-0.348
 spoke-0.321
 said-0.315
Token He
Token ID679
Feature activation+3.81e-01

Pos logit contributions
 disl+0.487
n+0.483
pieces+0.460
 hey+0.445
ric+0.437
Neg logit contributions
 whispered-0.490
 needed-0.475
 meant-0.471
 faded-0.464
chy-0.455
Token dropped
Token ID5710
Feature activation+3.34e-01

Pos logit contributions
n+0.546
 spend+0.510
 hey+0.495
 disl+0.494
 blast+0.492
Neg logit contributions
 Jones-0.312
 aisle-0.310
 meant-0.290
 White-0.288
 whispered-0.287
Token the
Token ID262
Feature activation+3.59e-01

Pos logit contributions
 disl+0.332
n+0.324
 arrest+0.322
 victory+0.310
 hey+0.310
Neg logit contributions
 whispered-0.466
 needed-0.458
 meant-0.435
 faded-0.419
 spoke-0.412
Token dolls
Token ID36062
Feature activation+3.44e-01

Pos logit contributions
 disl+0.460
pieces+0.455
n+0.451
 hey+0.451
ric+0.447
Neg logit contributions
 whispered-0.399
 needed-0.382
 meant-0.372
 spoke-0.343
 faded-0.341
Token and
Token ID290
Feature activation+3.24e-01

Pos logit contributions
n+0.407
 arrest+0.403
pieces+0.389
 pause+0.375
 victory+0.373
Neg logit contributions
 whispered-0.391
 needed-0.384
 meant-0.371
 White-0.368
 spoke-0.351
Token tried
Token ID3088
Feature activation+3.16e-01

Pos logit contributions
 arrest+0.443
 disl+0.434
pieces+0.429
n+0.426
 spend+0.422
Neg logit contributions
 whispered-0.290
 meant-0.290
 needed-0.278
 Jones-0.270
chy-0.264
Token to
Token ID284
Feature activation+3.06e-01

Pos logit contributions
 disl+0.357
n+0.348
 hey+0.344
pieces+0.344
ric+0.335
Neg logit contributions
 whispered-0.411
 needed-0.408
 meant-0.392
 faded-0.363
 said-0.358
Token run
Token ID1057
Feature activation+2.95e-01

Pos logit contributions
pieces+0.385
ric+0.383
iu+0.377
 Majesty+0.374
 interests+0.372
Neg logit contributions
 whispered-0.399
 said-0.378
 meant-0.359
 needed-0.341
chy-0.332
Token away
Token ID1497
Feature activation+2.89e-01

Pos logit contributions
 disl+0.360
pieces+0.345
 hey+0.342
ric+0.341
iu+0.339
Neg logit contributions
 whispered-0.365
 needed-0.362
 meant-0.347
 said-0.325
 faded-0.316
Token.
Token ID13
Feature activation+3.03e-01

Pos logit contributions
n+0.361
 disl+0.356
 victory+0.354
 arrest+0.343
 hey+0.341
Neg logit contributions
 needed-0.332
 whispered-0.320
 meant-0.303
 shone-0.285
 spoke-0.285
Token But
Token ID887
Feature activation+2.78e-01

Pos logit contributions
n+0.345
 disl+0.325
 hey+0.300
 yeah+0.294
 join+0.292
Neg logit contributions
 whispered-0.400
 needed-0.395
 meant-0.394
 faded-0.386
 shone-0.385
Token."
Token ID526
Feature activation+3.92e-01

Pos logit contributions
n+0.509
 disl+0.502
pieces+0.497
 spend+0.486
 interests+0.482
Neg logit contributions
 whispered-0.437
 needed-0.431
 meant-0.420
 spoke-0.389
 faded-0.379
Token
Token ID220
Feature activation+4.22e-01

Pos logit contributions
n+0.486
 disl+0.479
pieces+0.467
 hey+0.457
 spend+0.444
Neg logit contributions
 needed-0.456
 meant-0.451
 whispered-0.442
 faded-0.426
chy-0.424
Token
Token ID198
Feature activation+3.78e-01

Pos logit contributions
n+0.589
 disl+0.588
pieces+0.574
 hey+0.570
ric+0.566
Neg logit contributions
 whispered-0.426
 needed-0.424
 meant-0.400
 faded-0.373
 spoke-0.362
Token
Token ID198
Feature activation+3.66e-01

Pos logit contributions
n+0.517
 disl+0.513
pieces+0.501
 hey+0.494
ric+0.477
Neg logit contributions
 needed-0.406
 whispered-0.397
 meant-0.389
 faded-0.358
chy-0.339
TokenTom
Token ID13787
Feature activation+3.44e-01

Pos logit contributions
 disl+0.449
n+0.438
 hey+0.437
pieces+0.436
ric+0.421
Neg logit contributions
 whispered-0.441
 needed-0.438
 meant-0.420
 faded-0.384
 said-0.383
Tokenmy
Token ID1820
Feature activation+3.44e-01

Pos logit contributions
n+0.402
ieved+0.377
 spend+0.374
ric+0.361
 interests+0.360
Neg logit contributions
 dark-0.380
 faded-0.373
 needed-0.371
 aisle-0.370
 whispered-0.367
Token played
Token ID2826
Feature activation+3.15e-01

Pos logit contributions
n+0.504
 disl+0.477
 spend+0.466
 hey+0.460
 Majesty+0.455
Neg logit contributions
 meant-0.295
 needed-0.295
 whispered-0.289
 dark-0.283
 aisle-0.273
Token with
Token ID351
Feature activation+3.04e-01

Pos logit contributions
 disl+0.408
 hey+0.396
pieces+0.395
n+0.394
ric+0.387
Neg logit contributions
 whispered-0.360
 needed-0.355
 meant-0.340
 said-0.312
 faded-0.308
Token Mitt
Token ID16627
Feature activation+3.55e-01

Pos logit contributions
pieces+0.346
 disl+0.342
n+0.339
 arrest+0.331
 hey+0.331
Neg logit contributions
 whispered-0.377
 needed-0.374
 meant-0.358
 faded-0.333
 spoke-0.332
Tokenens
Token ID641
Feature activation+2.96e-01

Pos logit contributions
n+0.471
 hey+0.458
 disl+0.450
 exercise+0.439
 goodness+0.434
Neg logit contributions
 whispered-0.373
 blank-0.362
 needed-0.347
 faded-0.344
 dusty-0.332
Token for
Token ID329
Feature activation+2.75e-01

Pos logit contributions
pieces+0.267
n+0.259
 exercise+0.258
 spend+0.256
 victory+0.256
Neg logit contributions
 whispered-0.421
 meant-0.420
 faded-0.408
mail-0.398
 spoke-0.398
Token had
Token ID550
Feature activation+3.38e-01

Pos logit contributions
n+0.539
 disl+0.521
 spend+0.519
pieces+0.509
 hey+0.508
Neg logit contributions
 whispered-0.290
 meant-0.288
 needed-0.276
 Jones-0.263
 faded-0.256
Token never
Token ID1239
Feature activation+3.23e-01

Pos logit contributions
n+0.342
 disl+0.327
 arrest+0.321
 blast+0.320
pieces+0.312
Neg logit contributions
 whispered-0.463
 needed-0.449
 meant-0.444
 Jones-0.421
 aisle-0.415
Token seen
Token ID1775
Feature activation+3.37e-01

Pos logit contributions
 disl+0.339
n+0.325
 spend+0.320
pieces+0.317
 interests+0.311
Neg logit contributions
 whispered-0.423
 needed-0.402
 meant-0.393
 Jones-0.390
 White-0.382
Token one
Token ID530
Feature activation+3.21e-01

Pos logit contributions
 disl+0.289
n+0.287
pieces+0.285
 hey+0.277
ric+0.271
Neg logit contributions
 whispered-0.520
 needed-0.515
 meant-0.494
 faded-0.467
 said-0.465
Token like
Token ID588
Feature activation+3.02e-01

Pos logit contributions
n+0.367
pieces+0.363
 victory+0.348
 yeah+0.340
 hey+0.335
Neg logit contributions
 whispered-0.405
 needed-0.382
 meant-0.380
 spoke-0.362
 faded-0.359
Token it
Token ID340
Feature activation+2.86e-01

Pos logit contributions
n+0.235
 disl+0.224
pieces+0.218
 arrest+0.214
 victory+0.212
Neg logit contributions
 whispered-0.501
 needed-0.483
 meant-0.476
 faded-0.445
 spoke-0.445
Token before
Token ID878
Feature activation+2.92e-01

Pos logit contributions
n+0.253
 join+0.205
 how+0.202
 victory+0.195
 pause+0.191
Neg logit contributions
 whispered-0.448
Fuel-0.439
Gas-0.433
Factory-0.431
 woman-0.425
Token.
Token ID13
Feature activation+2.96e-01

Pos logit contributions
 arrest+0.229
n+0.226
 our+0.212
 explore+0.212
 victory+0.211
Neg logit contributions
 whispered-0.434
Ancient-0.404
 Lover-0.396
 meant-0.396
Fuel-0.395
Token But
Token ID887
Feature activation+2.70e-01

Pos logit contributions
 disl+0.436
n+0.427
 hey+0.426
pieces+0.426
ric+0.415
Neg logit contributions
 whispered-0.283
 needed-0.279
 meant-0.264
 said-0.236
 faded-0.235
Token Turtle
Token ID33137
Feature activation+2.58e-01

Pos logit contributions
n+0.324
 disl+0.317
 hey+0.315
pieces+0.309
 spend+0.302
Neg logit contributions
 whispered-0.325
 needed-0.316
 meant-0.309
 faded-0.286
 spoke-0.280
Token was
Token ID373
Feature activation+2.35e-01

Pos logit contributions
n+0.347
 disl+0.290
 spend+0.279
ieved+0.271
 hey+0.262
Neg logit contributions
 blank-0.283
 dark-0.270
 aisle-0.269
 faded-0.260
 Jones-0.256
Token
Token ID198
Feature activation+4.02e-01

Pos logit contributions
n+0.566
 disl+0.560
pieces+0.551
 hey+0.537
ric+0.537
Neg logit contributions
 needed-0.445
 whispered-0.441
 meant-0.426
 faded-0.394
chy-0.378
TokenThe
Token ID464
Feature activation+3.74e-01

Pos logit contributions
 disl+0.465
n+0.454
pieces+0.451
 hey+0.450
ric+0.433
Neg logit contributions
 whispered-0.513
 needed-0.511
 meant-0.492
 faded-0.452
 said-0.449
Token bus
Token ID1323
Feature activation+3.57e-01

Pos logit contributions
 disl+0.474
 hey+0.470
n+0.467
pieces+0.462
 blast+0.450
Neg logit contributions
 whispered-0.428
 needed-0.421
 meant-0.412
 faded-0.369
 spoke-0.366
Token driver
Token ID4639
Feature activation+3.49e-01

Pos logit contributions
n+0.411
 pause+0.370
 spend+0.366
 hey+0.365
 blast+0.363
Neg logit contributions
 basil-0.433
 Jones-0.423
 pods-0.420
chy-0.409
 whispered-0.408
Token smiled
Token ID13541
Feature activation+3.38e-01

Pos logit contributions
n+0.470
 spend+0.440
 pause+0.418
pieces+0.416
 hey+0.415
Neg logit contributions
 pods-0.352
chy-0.333
 basil-0.331
 blank-0.326
 Jones-0.320
Token back
Token ID736
Feature activation+3.15e-01

Pos logit contributions
 victory+0.390
 pause+0.386
 Majesty+0.384
 hey+0.383
ap+0.381
Neg logit contributions
 needed-0.391
 whispered-0.389
 meant-0.360
 aisle-0.350
 spoke-0.346
Token as
Token ID355
Feature activation+3.29e-01

Pos logit contributions
n+0.442
 disl+0.440
pieces+0.435
 hey+0.433
ric+0.430
Neg logit contributions
 whispered-0.300
 needed-0.299
 meant-0.284
 spoke-0.256
 faded-0.254
Token he
Token ID339
Feature activation+3.02e-01

Pos logit contributions
 disl+0.435
n+0.433
 hey+0.431
 victory+0.419
pieces+0.417
Neg logit contributions
 whispered-0.343
 needed-0.323
 meant-0.310
 spoke-0.281
 faded-0.281
Token opened
Token ID4721
Feature activation+2.95e-01

Pos logit contributions
n+0.355
 blast+0.326
 spend+0.317
 arrest+0.300
 pause+0.296
Neg logit contributions
 Jones-0.342
chy-0.334
book-0.333
 pods-0.332
 blank-0.330
Token the
Token ID262
Feature activation+2.97e-01

Pos logit contributions
 disl+0.359
n+0.350
pieces+0.349
 hey+0.348
ric+0.338
Neg logit contributions
 whispered-0.355
 needed-0.353
 meant-0.338
 faded-0.308
 said-0.308
Token door
Token ID3420
Feature activation+2.77e-01

Pos logit contributions
 disl+0.414
pieces+0.406
n+0.405
 hey+0.405
ric+0.396
Neg logit contributions
 whispered-0.306
 needed-0.303
 meant-0.290
 faded-0.258
 said-0.258
Token The
Token ID383
Feature activation+8.51e-02

Pos logit contributions
n+0.132
 disl+0.124
 arrest+0.120
 blast+0.116
 spend+0.114
Neg logit contributions
 needed-0.144
 whispered-0.143
 faded-0.136
 meant-0.136
chy-0.129
Token bug
Token ID5434
Feature activation+8.99e-02

Pos logit contributions
 disl+0.112
n+0.109
 hey+0.108
pieces+0.107
 blast+0.104
Neg logit contributions
 whispered-0.096
 needed-0.093
 meant-0.089
 faded-0.082
 spoke-0.081
Token felt
Token ID2936
Feature activation+7.28e-02

Pos logit contributions
n+0.131
 disl+0.118
 interests+0.115
 pause+0.113
 how+0.113
Neg logit contributions
 Jones-0.095
 aisle-0.085
Ancient-0.075
oky-0.074
Fuel-0.074
Token better
Token ID1365
Feature activation+6.88e-02

Pos logit contributions
 disl+0.095
n+0.089
 victory+0.087
 hey+0.086
 pause+0.085
Neg logit contributions
 whispered-0.081
 needed-0.079
 Jones-0.077
 aisle-0.075
 meant-0.073
Token and
Token ID290
Feature activation+6.87e-02

Pos logit contributions
n+0.085
 hey+0.083
 arrest+0.079
 disl+0.078
iu+0.077
Neg logit contributions
 needed-0.074
 meant-0.074
 whispered-0.074
 spoke-0.070
 shone-0.069
Token said
Token ID531
Feature activation+7.27e-02

Pos logit contributions
 disl+0.103
n+0.102
 hey+0.099
 spend+0.097
pieces+0.096
Neg logit contributions
 needed-0.057
 whispered-0.057
 meant-0.054
 Jones-0.052
 aisle-0.050
Token thank
Token ID5875
Feature activation+6.97e-02

Pos logit contributions
 disl+0.091
n+0.090
 hey+0.089
pieces+0.087
 yeah+0.085
Neg logit contributions
 whispered-0.087
 needed-0.085
 meant-0.082
 aisle-0.074
 faded-0.074
Token you
Token ID345
Feature activation+5.85e-02

Pos logit contributions
 disl+0.039
n+0.037
 hey+0.037
pieces+0.037
ric+0.034
Neg logit contributions
 whispered-0.130
 needed-0.130
 meant-0.126
 faded-0.119
 said-0.119
Token.
Token ID13
Feature activation+8.06e-02

Pos logit contributions
n+0.083
pieces+0.083
 disl+0.083
 hey+0.081
iu+0.078
Neg logit contributions
 whispered-0.059
 needed-0.057
 meant-0.055
 faded-0.050
 spoke-0.049
Token
Token ID198
Feature activation+8.41e-02

Pos logit contributions
 disl+0.088
n+0.088
 hey+0.083
pieces+0.082
ric+0.081
Neg logit contributions
 needed-0.107
 whispered-0.107
 meant-0.102
 faded-0.097
 spoke-0.094
Token
Token ID198
Feature activation+8.10e-02

Pos logit contributions
n+0.114
 disl+0.107
 blast+0.104
 hey+0.103
pieces+0.103
Neg logit contributions
 needed-0.092
 whispered-0.091
 meant-0.087
 faded-0.084
 spoke-0.078
Token He
Token ID679
Feature activation+5.11e-02

Pos logit contributions
 join+0.076
 GO+0.073
 spend+0.072
latable+0.070
 yeah+0.070
Neg logit contributions
 whispered-0.110
 faded-0.102
chy-0.100
 needed-0.099
 wil-0.099
Token decided
Token ID3066
Feature activation+3.94e-02

Pos logit contributions
 spend+0.062
n+0.061
 blast+0.055
 pause+0.052
 interests+0.051
Neg logit contributions
lessness-0.055
 Jones-0.055
 sky-0.052
 White-0.051
 bored-0.051
Token to
Token ID284
Feature activation+3.29e-02

Pos logit contributions
 disl+0.045
n+0.044
 hey+0.044
pieces+0.043
 arrest+0.042
Neg logit contributions
 whispered-0.050
 needed-0.050
 meant-0.048
 faded-0.044
 said-0.044
Token make
Token ID787
Feature activation+2.71e-02

Pos logit contributions
ric+0.040
pieces+0.038
 disl+0.038
ext+0.037
iu+0.037
Neg logit contributions
 whispered-0.042
 needed-0.042
 meant-0.042
 said-0.040
 spoke-0.037
Token a
Token ID257
Feature activation+3.68e-02

Pos logit contributions
 disl+0.036
 hey+0.035
ric+0.035
pieces+0.035
n+0.034
Neg logit contributions
 whispered-0.030
 needed-0.030
 meant-0.029
 said-0.027
 dark-0.025
Token special
Token ID2041
Feature activation+2.47e-02

Pos logit contributions
 disl+0.043
pieces+0.042
 hey+0.042
n+0.042
 blast+0.041
Neg logit contributions
 whispered-0.046
 needed-0.046
 meant-0.044
 faded-0.040
 spoke-0.040
Token pie
Token ID2508
Feature activation+2.88e-02

Pos logit contributions
 disl+0.025
pieces+0.025
n+0.025
 hey+0.024
ric+0.023
Neg logit contributions
 whispered-0.035
 needed-0.034
 meant-0.033
 faded-0.031
 said-0.031
Token.
Token ID13
Feature activation+5.30e-02

Pos logit contributions
n+0.039
pieces+0.039
 disl+0.039
 hey+0.038
 blast+0.038
Neg logit contributions
 whispered-0.030
 needed-0.029
 meant-0.027
 faded-0.026
 spoke-0.025
Token So
Token ID1406
Feature activation+3.76e-02

Pos logit contributions
n+0.063
 join+0.060
 blast+0.060
 GO+0.059
 disl+0.059
Neg logit contributions
 whispered-0.065
 needed-0.061
 faded-0.061
chy-0.059
 meant-0.058
Token he
Token ID339
Feature activation+1.19e-02

Pos logit contributions
n+0.046
 victory+0.045
 blast+0.044
 spend+0.044
 hey+0.044
Neg logit contributions
 whispered-0.045
 meant-0.042
 needed-0.042
chy-0.042
 faded-0.039
Token got
Token ID1392
Feature activation+3.76e-03

Pos logit contributions
n+0.016
 blast+0.015
 spend+0.015
iu+0.014
pieces+0.014
Neg logit contributions
 Jones-0.011
 whispered-0.011
 meant-0.011
 dark-0.011
 bored-0.011
Token hive
Token ID35881
Feature activation+1.76e-01

Pos logit contributions
 hey+0.205
 Majesty+0.200
ric+0.198
n+0.195
 pause+0.191
Neg logit contributions
 whispered-0.193
 meant-0.187
 needed-0.185
 faded-0.176
 spoke-0.170
Token.
Token ID13
Feature activation+1.75e-01

Pos logit contributions
n+0.226
pieces+0.221
 spend+0.210
 join+0.210
 disl+0.204
Neg logit contributions
 needed-0.189
 meant-0.184
 whispered-0.184
 faded-0.171
 spoke-0.163
Token She
Token ID1375
Feature activation+1.54e-01

Pos logit contributions
n+0.215
pieces+0.209
 disl+0.208
 hey+0.203
 spend+0.200
Neg logit contributions
 whispered-0.204
 needed-0.202
 meant-0.199
 faded-0.187
chy-0.182
Token waved
Token ID26834
Feature activation+1.42e-01

Pos logit contributions
n+0.223
 spend+0.209
 blast+0.208
 pause+0.201
pieces+0.200
Neg logit contributions
 meant-0.133
 Jones-0.130
 needed-0.125
 whispered-0.123
 dark-0.119
Token her
Token ID607
Feature activation+1.36e-01

Pos logit contributions
 hey+0.132
 disl+0.132
pieces+0.128
n+0.128
 victory+0.126
Neg logit contributions
 needed-0.207
 whispered-0.201
 meant-0.199
chy-0.180
 spoke-0.179
Token hands
Token ID2832
Feature activation+1.34e-01

Pos logit contributions
 disl+0.148
n+0.147
 hey+0.145
pieces+0.145
ric+0.142
Neg logit contributions
 whispered-0.179
 needed-0.178
 meant-0.173
 spoke-0.158
 faded-0.158
Token and
Token ID290
Feature activation+1.26e-01

Pos logit contributions
 join+0.118
 pause+0.117
ap+0.116
n+0.115
 victory+0.114
Neg logit contributions
 needed-0.191
 meant-0.187
 whispered-0.170
 smoked-0.169
Ancient-0.168
Token the
Token ID262
Feature activation+1.30e-01

Pos logit contributions
n+0.184
 blast+0.178
 disl+0.178
 spend+0.177
pieces+0.176
Neg logit contributions
 needed-0.109
 meant-0.107
 whispered-0.099
chy-0.098
 aisle-0.092
Token bees
Token ID17002
Feature activation+1.30e-01

Pos logit contributions
n+0.165
 hey+0.164
 Majesty+0.163
pieces+0.161
 pause+0.158
Neg logit contributions
 whispered-0.144
 meant-0.140
 needed-0.137
 faded-0.126
 spoke-0.123
Token started
Token ID2067
Feature activation+1.16e-01

Pos logit contributions
n+0.161
 join+0.155
pieces+0.150
 pause+0.146
 spend+0.145
Neg logit contributions
 Jones-0.145
 meant-0.142
 needed-0.138
Fuel-0.133
Ancient-0.132
Token to
Token ID284
Feature activation+1.10e-01

Pos logit contributions
n+0.101
 join+0.100
 disl+0.099
ric+0.095
 spend+0.094
Neg logit contributions
 needed-0.175
 whispered-0.169
 meant-0.168
 aisle-0.157
chy-0.152
Token were
Token ID547
Feature activation+6.58e-03

Pos logit contributions
 pause+0.026
 join+0.026
 spend+0.025
n+0.025
 hey+0.025
Neg logit contributions
chy-0.023
 aisle-0.023
 dusty-0.023
 meant-0.023
elly-0.022
Token amazed
Token ID24562
Feature activation+2.22e-02

Pos logit contributions
n+0.010
 disl+0.010
ric+0.010
 hey+0.010
pieces+0.010
Neg logit contributions
 needed-0.006
 whispered-0.006
 meant-0.005
 faded-0.005
chy-0.005
Token and
Token ID290
Feature activation+5.67e-03

Pos logit contributions
 disl+0.033
 hey+0.033
n+0.032
pieces+0.032
nels+0.031
Neg logit contributions
 whispered-0.020
 needed-0.020
 meant-0.018
 faded-0.017
 spoke-0.016
Token they
Token ID484
Feature activation+6.58e-04

Pos logit contributions
 disl+0.010
n+0.010
pieces+0.010
 spend+0.010
 join+0.009
Neg logit contributions
 aisle-0.003
chy-0.003
 Jones-0.003
 needed-0.003
 meant-0.003
Token all
Token ID477
Feature activation-5.24e-03

Pos logit contributions
 spend+0.001
 pause+0.001
n+0.001
pieces+0.001
 join+0.001
Neg logit contributions
 aisle-0.001
 Jones-0.001
chy-0.001
elly-0.001
 meant-0.001
Token stopped
Token ID5025
Feature activation-7.01e-03

Pos logit contributions
 needed+0.004
 Jones+0.004
 whispered+0.004
 meant+0.004
 aisle+0.004
Neg logit contributions
 disl-0.008
n-0.008
pieces-0.008
 spend-0.008
 interests-0.007
Token to
Token ID284
Feature activation-1.16e-02

Pos logit contributions
 whispered+0.008
 needed+0.007
 meant+0.007
 said+0.006
 faded+0.006
Neg logit contributions
 disl-0.010
n-0.010
pieces-0.009
 hey-0.009
latable-0.009
Token watch
Token ID2342
Feature activation-2.30e-02

Pos logit contributions
 whispered+0.016
 said+0.015
 needed+0.014
 spoke+0.014
 faded+0.014
Neg logit contributions
ext-0.014
immers-0.014
nels-0.014
han-0.013
iu-0.013
Token the
Token ID262
Feature activation-9.87e-03

Pos logit contributions
 needed+0.034
 whispered+0.033
 meant+0.032
chy+0.030
 smelled+0.030
Neg logit contributions
pieces-0.023
 victory-0.021
n-0.021
 blast-0.021
 spend-0.021
Token bird
Token ID6512
Feature activation-1.19e-02

Pos logit contributions
 whispered+0.010
 needed+0.010
 meant+0.010
 spoke+0.009
 said+0.009
Neg logit contributions
 disl-0.014
 hey-0.013
n-0.013
pieces-0.013
 blast-0.013
Token for
Token ID329
Feature activation-2.71e-02

Pos logit contributions
 needed+0.013
 whispered+0.013
 meant+0.013
 faded+0.011
 Jones+0.011
Neg logit contributions
pieces-0.016
n-0.015
 disl-0.015
ric-0.015
 hey-0.015
Token."
Token ID526
Feature activation+2.56e-01

Pos logit contributions
 disl+0.290
n+0.285
pieces+0.282
 hey+0.282
ric+0.274
Neg logit contributions
 whispered-0.355
 needed-0.352
 meant-0.338
 faded-0.313
 said-0.312
Token
Token ID198
Feature activation+2.49e-01

Pos logit contributions
n+0.298
 disl+0.286
pieces+0.274
 join+0.265
 arrest+0.264
Neg logit contributions
 needed-0.330
 meant-0.325
 whispered-0.313
chy-0.312
 faded-0.307
Token
Token ID198
Feature activation+2.44e-01

Pos logit contributions
n+0.343
 disl+0.336
pieces+0.328
 hey+0.325
ric+0.317
Neg logit contributions
 needed-0.271
 whispered-0.262
 meant-0.259
 faded-0.237
chy-0.225
TokenThe
Token ID464
Feature activation+2.15e-01

Pos logit contributions
 disl+0.274
n+0.271
pieces+0.267
 hey+0.266
 blast+0.254
Neg logit contributions
 needed-0.324
 whispered-0.320
 meant-0.311
 faded-0.286
 said-0.279
Token mouse
Token ID10211
Feature activation+2.23e-01

Pos logit contributions
 hey+0.268
n+0.265
pieces+0.262
 disl+0.260
 victory+0.259
Neg logit contributions
 whispered-0.245
 needed-0.241
 meant-0.238
 faded-0.213
 spoke-0.211
Token followed
Token ID3940
Feature activation+1.87e-01

Pos logit contributions
n+0.317
 blast+0.274
 spend+0.270
 disl+0.268
 pause+0.265
Neg logit contributions
 Jones-0.211
 dark-0.208
 bored-0.206
 meant-0.203
 needed-0.200
Token Benny
Token ID44275
Feature activation+2.07e-01

Pos logit contributions
n+0.212
pieces+0.209
 disl+0.208
 hey+0.207
 arrest+0.205
Neg logit contributions
 needed-0.241
 whispered-0.239
 meant-0.232
 faded-0.214
 shone-0.210
Token as
Token ID355
Feature activation+2.03e-01

Pos logit contributions
n+0.244
pieces+0.222
 spend+0.220
 pause+0.217
 join+0.216
Neg logit contributions
 meant-0.257
 needed-0.255
 whispered-0.252
 faded-0.245
 smelled-0.236
Token he
Token ID339
Feature activation+1.81e-01

Pos logit contributions
 disl+0.311
 hey+0.307
n+0.307
pieces+0.304
 arrest+0.297
Neg logit contributions
 whispered-0.178
 needed-0.175
 meant-0.166
 faded-0.144
 spoke-0.144
Token led
Token ID2957
Feature activation+1.83e-01

Pos logit contributions
n+0.240
 blast+0.237
 spend+0.219
 arrest+0.217
 join+0.215
Neg logit contributions
chy-0.184
 Jones-0.182
 meant-0.178
 bored-0.177
 dark-0.175
Token the
Token ID262
Feature activation+1.80e-01

Pos logit contributions
 disl+0.213
 hey+0.204
pieces+0.202
ric+0.202
n+0.196
Neg logit contributions
 whispered-0.238
 needed-0.236
 meant-0.227
 said-0.214
 spoke-0.208
Token to
Token ID284
Feature activation-1.78e-01

Pos logit contributions
 needed+0.185
 whispered+0.185
 meant+0.173
 said+0.166
 spoke+0.157
Neg logit contributions
ric-0.244
pieces-0.243
 hey-0.235
 disl-0.234
latable-0.233
Token play
Token ID711
Feature activation-1.83e-01

Pos logit contributions
 said+0.195
 whispered+0.192
 needed+0.175
 closed+0.173
 meant+0.172
Neg logit contributions
pieces-0.285
ocol-0.267
iu-0.254
ieved-0.248
 disl-0.243
Token with
Token ID351
Feature activation-1.79e-01

Pos logit contributions
 whispered+0.179
 needed+0.173
 said+0.166
 meant+0.158
 dark+0.155
Neg logit contributions
ric-0.293
 Majesty-0.274
iu-0.273
 disl-0.271
 hey-0.264
Token Tim
Token ID5045
Feature activation-1.80e-01

Pos logit contributions
 needed+0.226
 whispered+0.222
 meant+0.212
 said+0.201
 spoke+0.192
Neg logit contributions
 disl-0.212
 hey-0.209
ric-0.208
pieces-0.207
n-0.206
Tokenmy
Token ID1820
Feature activation-1.83e-01

Pos logit contributions
 whispered+0.195
 needed+0.193
 meant+0.183
 faded+0.169
 said+0.166
Neg logit contributions
 disl-0.241
n-0.238
pieces-0.237
 hey-0.236
ric-0.230
Token's
Token ID338
Feature activation-1.71e-01

Pos logit contributions
 whispered+0.231
 faded+0.225
 meant+0.218
 Jones+0.217
 shone+0.216
Neg logit contributions
pieces-0.218
n-0.208
 victory-0.189
 interests-0.189
 spend-0.181
Token toy
Token ID13373
Feature activation-1.94e-01

Pos logit contributions
 needed+0.191
 whispered+0.190
 meant+0.179
 said+0.165
 faded+0.162
Neg logit contributions
 disl-0.225
 hey-0.223
pieces-0.222
n-0.221
ric-0.217
Token cars
Token ID5006
Feature activation-2.01e-01

Pos logit contributions
 whispered+0.257
 meant+0.246
 needed+0.237
 faded+0.233
 smelled+0.230
Neg logit contributions
pieces-0.209
 hey-0.205
 arrest-0.204
 victory-0.202
nels-0.197
Token instead
Token ID2427
Feature activation-1.93e-01

Pos logit contributions
 whispered+0.246
 needed+0.240
 meant+0.234
 faded+0.227
 Jones+0.226
Neg logit contributions
n-0.229
pieces-0.225
 victory-0.217
 hey-0.213
 interests-0.209
Token.
Token ID13
Feature activation-1.60e-01

Pos logit contributions
 whispered+0.258
 smelled+0.244
 shone+0.238
Ancient+0.237
 Jones+0.235
Neg logit contributions
n-0.221
pieces-0.186
 victory-0.179
 spend-0.178
 join-0.175
Token They
Token ID1119
Feature activation-1.84e-01

Pos logit contributions
 whispered+0.207
 needed+0.201
 meant+0.195
 faded+0.183
 spoke+0.179
Neg logit contributions
 disl-0.187
n-0.179
 hey-0.177
pieces-0.176
 arrest-0.170
Token toy
Token ID13373
Feature activation-1.46e-01

Pos logit contributions
 whispered+0.153
 needed+0.149
 meant+0.142
 spoke+0.136
 faded+0.130
Neg logit contributions
n-0.157
 hey-0.154
 victory-0.154
 disl-0.154
latable-0.152
Token.
Token ID13
Feature activation-1.09e-01

Pos logit contributions
 whispered+0.190
 meant+0.182
 blank+0.177
 smelled+0.175
 spoke+0.174
Neg logit contributions
 victory-0.138
 our-0.136
 join-0.136
n-0.134
 arrest-0.133
Token Finally
Token ID9461
Feature activation-1.40e-01

Pos logit contributions
 whispered+0.121
 needed+0.117
 meant+0.114
 spoke+0.103
 faded+0.103
Neg logit contributions
 disl-0.144
n-0.142
 hey-0.141
pieces-0.140
ric-0.138
Token,
Token ID11
Feature activation-1.37e-01

Pos logit contributions
 whispered+0.133
 needed+0.132
 meant+0.123
 said+0.113
 faded+0.111
Neg logit contributions
 disl-0.213
pieces-0.207
 hey-0.205
n-0.204
ric-0.203
Token they
Token ID484
Feature activation-1.47e-01

Pos logit contributions
 needed+0.123
 whispered+0.121
 meant+0.110
 said+0.104
 faded+0.100
Neg logit contributions
 disl-0.221
pieces-0.214
 hey-0.211
ric-0.208
 interests-0.207
Token found
Token ID1043
Feature activation-1.58e-01

Pos logit contributions
 whispered+0.145
 meant+0.143
 needed+0.136
chy+0.130
 White+0.124
Neg logit contributions
 blast-0.199
n-0.197
 hey-0.192
 spend-0.192
 pause-0.187
Token it
Token ID340
Feature activation-1.65e-01

Pos logit contributions
 whispered+0.237
 needed+0.230
 meant+0.224
chy+0.215
 spoke+0.213
Neg logit contributions
n-0.140
 hey-0.138
 victory-0.134
 arrest-0.134
 disl-0.132
Token!
Token ID0
Feature activation-1.28e-01

Pos logit contributions
Dead+0.233
 rhy+0.225
Ancient+0.221
 White+0.217
 whispered+0.217
Neg logit contributions
n-0.156
latable-0.129
 how-0.128
 explore-0.127
 stay-0.126
Token The
Token ID383
Feature activation-1.53e-01

Pos logit contributions
 whispered+0.142
 needed+0.138
 meant+0.134
 spoke+0.121
 faded+0.121
Neg logit contributions
 disl-0.167
n-0.166
 hey-0.162
pieces-0.162
ric-0.159
Token little
Token ID1310
Feature activation-1.56e-01

Pos logit contributions
 whispered+0.170
 needed+0.168
 meant+0.160
 said+0.146
 faded+0.145
Neg logit contributions
 disl-0.205
pieces-0.201
 hey-0.200
n-0.200
ric-0.195
Token girl
Token ID2576
Feature activation-1.59e-01

Pos logit contributions
 whispered+0.122
 needed+0.120
 meant+0.112
 faded+0.097
 said+0.097
Neg logit contributions
 disl-0.256
 hey-0.251
n-0.251
pieces-0.250
ric-0.244
Token ears
Token ID11368
Feature activation-2.12e-01

Pos logit contributions
 whispered+0.280
 meant+0.276
 needed+0.268
 spoke+0.254
 faded+0.245
Neg logit contributions
 Majesty-0.217
n-0.213
 disl-0.212
 pause-0.209
 interests-0.208
Token felt
Token ID2936
Feature activation-2.14e-01

Pos logit contributions
 Jones+0.278
Ancient+0.239
 White+0.237
 aisle+0.232
 Green+0.231
Neg logit contributions
n-0.250
 pause-0.246
 spend-0.227
 stay-0.224
 stop-0.212
Token deaf
Token ID21815
Feature activation-2.10e-01

Pos logit contributions
 whispered+0.209
 needed+0.183
 Jones+0.180
 aisle+0.176
 meant+0.176
Neg logit contributions
 disl-0.310
 hey-0.293
n-0.290
 victory-0.287
 pause-0.284
Token.
Token ID13
Feature activation-1.83e-01

Pos logit contributions
 whispered+0.300
Ancient+0.285
Fuel+0.281
 aisle+0.280
 spoke+0.278
Neg logit contributions
 how-0.193
 victory-0.170
 join-0.170
 interests-0.166
n-0.165
Token
Token ID220
Feature activation-1.71e-01

Pos logit contributions
 whispered+0.186
 needed+0.181
 meant+0.175
 faded+0.160
 spoke+0.155
Neg logit contributions
 disl-0.260
n-0.257
 hey-0.251
pieces-0.249
ric-0.246
Token
Token ID198
Feature activation-1.80e-01

Pos logit contributions
 whispered+0.176
 needed+0.172
 meant+0.163
 faded+0.155
 spoke+0.150
Neg logit contributions
n-0.236
 disl-0.233
 hey-0.227
pieces-0.227
ric-0.226
Token
Token ID198
Feature activation-1.82e-01

Pos logit contributions
 whispered+0.165
 needed+0.163
 meant+0.154
 faded+0.137
 said+0.136
Neg logit contributions
 disl-0.271
n-0.266
 hey-0.265
pieces-0.265
ric-0.258
TokenB
Token ID33
Feature activation-2.11e-01

Pos logit contributions
 whispered+0.192
 needed+0.192
 meant+0.179
 said+0.165
 faded+0.162
Neg logit contributions
 disl-0.254
pieces-0.248
 hey-0.247
ric-0.243
n-0.243
Tokenenny
Token ID11870
Feature activation-2.29e-01

Pos logit contributions
 needed+0.198
 whispered+0.190
 meant+0.181
 faded+0.161
 said+0.159
Neg logit contributions
 disl-0.319
pieces-0.316
 hey-0.308
n-0.304
ric-0.301
Token decided
Token ID3066
Feature activation-2.25e-01

Pos logit contributions
 meant+0.235
 Jones+0.226
 blank+0.220
 aisle+0.219
 faded+0.218
Neg logit contributions
n-0.297
 spend-0.272
 hey-0.271
 pause-0.266
 arrest-0.260
Token to
Token ID284
Feature activation-2.30e-01

Pos logit contributions
 whispered+0.295
 needed+0.291
 meant+0.285
 faded+0.262
 spoke+0.255
Neg logit contributions
 disl-0.250
n-0.243
 hey-0.239
 arrest-0.229
pieces-0.229
Token bee
Token ID20697
Feature activation-1.44e-01

Pos logit contributions
 whispered+0.156
 needed+0.153
 meant+0.146
 faded+0.139
 spoke+0.138
Neg logit contributions
 victory-0.162
 disl-0.162
 pause-0.157
 Majesty-0.156
n-0.154
Token's
Token ID338
Feature activation-1.30e-01

Pos logit contributions
 aisle+0.170
 whispered+0.166
Ancient+0.165
Fuel+0.163
 meant+0.162
Neg logit contributions
pieces-0.166
n-0.164
 victory-0.148
 disl-0.147
 spend-0.145
Token answer
Token ID3280
Feature activation-1.64e-01

Pos logit contributions
 whispered+0.199
 needed+0.190
 meant+0.188
 aisle+0.183
 spoke+0.179
Neg logit contributions
 disl-0.124
 interests-0.114
 victory-0.113
n-0.107
 respect-0.105
Token.
Token ID13
Feature activation-1.17e-01

Pos logit contributions
 aisle+0.212
 whispered+0.209
 needed+0.207
Ancient+0.203
Factory+0.202
Neg logit contributions
n-0.161
 how-0.160
pieces-0.152
 victory-0.149
 respect-0.147
Token He
Token ID679
Feature activation-1.42e-01

Pos logit contributions
 whispered+0.145
 needed+0.138
 faded+0.131
 meant+0.130
 spoke+0.129
Neg logit contributions
 disl-0.141
 Majesty-0.134
n-0.132
 yeah-0.130
 arrest-0.125
Token thanked
Token ID26280
Feature activation-1.45e-01

Pos logit contributions
 aisle+0.142
 Jones+0.135
 dark+0.127
oky+0.127
lessness+0.127
Neg logit contributions
n-0.192
 spend-0.190
 blast-0.177
 join-0.173
 how-0.172
Token her
Token ID607
Feature activation-1.54e-01

Pos logit contributions
 whispered+0.164
 needed+0.162
 meant+0.155
 said+0.143
 faded+0.140
Neg logit contributions
 disl-0.190
pieces-0.185
 hey-0.184
n-0.184
ric-0.179
Token and
Token ID290
Feature activation-1.52e-01

Pos logit contributions
 needed+0.184
 whispered+0.178
 meant+0.178
 faded+0.165
 spoke+0.164
Neg logit contributions
pieces-0.192
 victory-0.185
n-0.184
 Majesty-0.183
 hey-0.182
Token went
Token ID1816
Feature activation-1.66e-01

Pos logit contributions
 needed+0.112
 meant+0.110
chy+0.106
 whispered+0.104
 aisle+0.095
Neg logit contributions
n-0.239
 spend-0.236
 disl-0.233
 hey-0.231
 blast-0.230
Token back
Token ID736
Feature activation-1.67e-01

Pos logit contributions
 whispered+0.184
 needed+0.167
 said+0.160
 meant+0.152
 faded+0.150
Neg logit contributions
 disl-0.244
 hey-0.232
ric-0.229
pieces-0.219
 interests-0.219
Token home
Token ID1363
Feature activation-1.83e-01

Pos logit contributions
 whispered+0.209
 needed+0.206
 meant+0.196
 said+0.182
 faded+0.179
Neg logit contributions
 disl-0.202
 hey-0.197
pieces-0.195
n-0.195
ric-0.190
TokenHis
Token ID6653
Feature activation-1.09e-01

Pos logit contributions
 whispered+0.063
 needed+0.059
 meant+0.057
 said+0.056
 spoke+0.052
Neg logit contributions
ric-0.098
pieces-0.097
 disl-0.097
 hey-0.096
 Majesty-0.094
Token mum
Token ID25682
Feature activation-1.03e-01

Pos logit contributions
 whispered+0.150
 needed+0.149
 meant+0.143
 said+0.136
 faded+0.131
Neg logit contributions
 disl-0.118
pieces-0.117
 hey-0.115
n-0.115
ric-0.112
Token said
Token ID531
Feature activation-9.88e-02

Pos logit contributions
chy+0.094
 pods+0.091
 dark+0.090
 dusty+0.090
 faded+0.088
Neg logit contributions
n-0.162
 pause-0.135
 how-0.133
 hey-0.132
ieved-0.130
Token,
Token ID11
Feature activation-1.02e-01

Pos logit contributions
 whispered+0.141
 meant+0.138
 needed+0.135
 faded+0.130
 aisle+0.127
Neg logit contributions
n-0.097
 hey-0.090
 yeah-0.089
ap-0.085
 pause-0.084
Token "
Token ID366
Feature activation-9.97e-02

Pos logit contributions
 faded+0.158
able+0.154
 whispered+0.150
 meant+0.146
 needed+0.144
Neg logit contributions
n-0.095
 how-0.084
 yeah-0.077
 wow-0.076
 hey-0.076
TokenNo
Token ID2949
Feature activation-1.17e-01

Pos logit contributions
 whispered+0.146
 needed+0.145
 meant+0.140
 faded+0.131
 said+0.130
Neg logit contributions
 disl-0.097
n-0.094
 hey-0.094
pieces-0.093
ric-0.089
Token,
Token ID11
Feature activation-1.10e-01

Pos logit contributions
 whispered+0.085
 said+0.084
 meant+0.081
 needed+0.080
 dark+0.064
Neg logit contributions
 disl-0.209
ric-0.205
pieces-0.205
 hey-0.201
 Majesty-0.201
Token not
Token ID407
Feature activation-1.14e-01

Pos logit contributions
 said+0.106
 meant+0.102
 whispered+0.095
 needed+0.091
 dark+0.078
Neg logit contributions
ric-0.186
 Amen-0.178
 Majesty-0.178
pieces-0.177
 disl-0.173
Token today
Token ID1909
Feature activation-1.28e-01

Pos logit contributions
 whispered+0.130
 needed+0.130
 meant+0.128
 dark+0.122
 said+0.119
Neg logit contributions
pieces-0.153
 disl-0.148
 hey-0.146
iu-0.143
ric-0.143
Token,
Token ID11
Feature activation-1.16e-01

Pos logit contributions
 whispered+0.157
 needed+0.143
 meant+0.139
 spoke+0.138
 shone+0.136
Neg logit contributions
n-0.155
 our-0.144
 spend-0.143
 victory-0.143
 arrest-0.141
Token Tom
Token ID4186
Feature activation-1.24e-01

Pos logit contributions
 said+0.093
 meant+0.092
 whispered+0.083
 needed+0.078
 dark+0.067
Neg logit contributions
ric-0.212
 Amen-0.202
pieces-0.200
 Majesty-0.199
 disl-0.197