TOP ACTIVATIONS
MAX = 4.463

It is their teacher . She comes to see what is
comes to their room . She is not happy .
. Sammy wanted to touch the staff , but Lucy didn
comes to their room . She sees them fighting over the
what they are doing . She sees them in the hat
comes to the box . She says , " Stop ,
is Lily 's mom . She has come to pick them
Mom hears them . She comes to the box .
comes to their room . She is not happy .
Mom hears them . She comes to see what is
comes to the yard . She sees the box and the
Sara and Ben fighting . She comes to see what is
and comes to see . She is not happy . She
Mom hears them . She comes to their room .
Mom hears them . She comes to their room .
comes to the room . She sees Anna and Ben struggle
It is their mom . She is holding a shiny key
decided to go and touch the staff , even though he
and comes to see . She sees the broken comb and
It is their mom . She comes into the room .

MIN ACTIVATIONS
MIN = -4.124

steak . While his mom was busy preparing the
maybe I can eat it !" He looked and looked to
The waves are too strong !" But no one was there
to walk home . So , he asked his friend for
the tunnel and be friends !" Mom o thought about it
of the hill . So , he kept running .
. But , because Bob was so lazy
felt a burning sensation on his paw and quickly pulled it
fell down . He hurt his leg and started to suffer
my fell down and hurt his knee . Max bark ed
the big , mean cat !" Spot 's friends
things . Eventually , Henry became the most knowledgeable
Mama ! A red balloon !" M ama smiled
he had a cut on his cheek . It was red
He felt a tug on his fishing line and pulled up
to jump . But then , Bob o fell into the
away . The bunny enjoyed his steak and the huge tree
ground and started to stretch his muscles . Suddenly he heard
ac he . He felt his head spin . He felt
my fell down and hurt his knee .

INTERVAL 2.316 - 4.463
CONTAINS 0.381%

and angry . She did not like the boy . She
She goes to them . She says , " Hi ,
a big bush with red flowers . They wanted to cut
So , Lily took the purse and put her favorite
find anything . Then she went outside and searched between the

INTERVAL 0.169 - 2.316
CONTAINS 40.942%

and kisses them . She is proud of them .
In the end , Sally and Jill were so tired that
They loved it and asked Tim my how he made it
� � Who are you ? â € The
€ The bear thought that was a great idea so they

INTERVAL -1.977 - 0.169
CONTAINS 57.692%

sorry , mum . I learned my lesson . I don
height . Tom 's tower is big in width . You
, " Is the bad man going to jail ?" His
He asked a bird . The bird said " That
sits next to Sam on the slide . "

INTERVAL -4.124 - -1.977
CONTAINS 0.985%

sad . She played with him in the park and at
grab a few peanuts . But the peanut tree was very
storm raged on . Winds whipped and rain fell heavily .
that they both slept for a long , long time and
day , Max was tired but happy . He went home
Token It
Token ID632
Feature activation+3.64e+00

Pos logit contributions
ples+5.604
ptions+5.319
ount+5.246
ydia+5.001
 blo+4.996
Neg logit contributions
 tum-3.948
 His-3.253
night-3.167
 lung-3.106
 steak-3.063
Token is
Token ID318
Feature activation+1.87e+00

Pos logit contributions
ples+7.680
ount+7.105
ptions+7.055
ydia+6.946
 Jem+6.874
Neg logit contributions
 tum-5.330
night-4.695
 lung-4.353
 leg-4.248
 steak-4.182
Token their
Token ID511
Feature activation+3.99e-01

Pos logit contributions
ples+4.027
ount+3.716
 paints+3.656
 angles+3.511
 violet+3.510
Neg logit contributions
 tum-2.376
 thirst-1.979
night-1.893
 steak-1.866
 lung-1.817
Token teacher
Token ID4701
Feature activation+2.45e+00

Pos logit contributions
ples+0.895
ount+0.839
 Jas+0.833
yl+0.812
 paints+0.809
Neg logit contributions
 tum-0.462
 thirst-0.371
night-0.367
 steak-0.335
 lung-0.321
Token.
Token ID13
Feature activation+2.82e+00

Pos logit contributions
ples+4.775
ount+4.590
ydia+4.555
 Jas+4.457
 violet+4.423
Neg logit contributions
 tum-3.610
night-2.924
 steak-2.880
 lung-2.828
 thirst-2.823
Token She
Token ID1375
Feature activation+4.46e+00

Pos logit contributions
ples+6.735
ptions+6.362
ount+6.168
 blo+6.032
 paints+5.998
Neg logit contributions
 tum-3.406
 His-2.845
night-2.527
His-2.426
 steak-2.393
Token comes
Token ID2058
Feature activation+4.60e-01

Pos logit contributions
ples+9.258
ount+8.839
ydia+8.678
ptions+8.619
 Jem+8.361
Neg logit contributions
 tum-6.816
night-5.839
 lung-5.822
 sled-5.579
 drowned-5.522
Token to
Token ID284
Feature activation+1.04e+00

Pos logit contributions
ples+0.866
ount+0.804
 Jas+0.796
 paints+0.776
yl+0.774
Neg logit contributions
 tum-0.682
night-0.582
 thirst-0.582
 steak-0.541
 lung-0.525
Token see
Token ID766
Feature activation+2.21e+00

Pos logit contributions
ples+1.907
ount+1.763
 Jas+1.688
 blo+1.662
 paints+1.656
Neg logit contributions
 tum-1.617
night-1.399
 thirst-1.390
 steak-1.321
 dinner-1.268
Token what
Token ID644
Feature activation+1.62e+00

Pos logit contributions
ples+3.560
ount+3.158
ydia+3.098
 Jas+3.045
iddling+3.019
Neg logit contributions
 tum-4.015
 steak-3.539
night-3.415
 dinner-3.385
 thirst-3.383
Token is
Token ID318
Feature activation+3.47e+00

Pos logit contributions
ples+2.639
ount+2.367
 Jas+2.219
ydia+2.196
 violet+2.171
Neg logit contributions
 tum-3.082
night-2.665
 steak-2.625
 thirst-2.497
 lung-2.476
Token comes
Token ID2058
Feature activation+8.28e-01

Pos logit contributions
ples+7.302
ount+6.696
ydia+6.428
umes+6.303
od+6.299
Neg logit contributions
 tum-7.806
night-6.741
 sled-6.696
 leg-6.671
 muscle-6.614
Token to
Token ID284
Feature activation+1.29e+00

Pos logit contributions
ples+1.720
ount+1.613
 Jas+1.574
yl+1.548
 paints+1.547
Neg logit contributions
 tum-1.074
night-0.890
 thirst-0.884
 steak-0.819
 lung-0.787
Token their
Token ID511
Feature activation+6.93e-01

Pos logit contributions
ples+2.234
ount+2.061
 Jas+1.937
 blo+1.934
 paints+1.922
Neg logit contributions
 tum-2.148
 thirst-1.858
night-1.836
 steak-1.761
 dinner-1.728
Token room
Token ID2119
Feature activation+7.82e-01

Pos logit contributions
ples+1.044
ount+0.952
 Jas+0.929
yl+0.902
ydia+0.887
Neg logit contributions
 tum-1.324
 thirst-1.158
night-1.144
 steak-1.101
 lung-1.076
Token.
Token ID13
Feature activation+2.42e+00

Pos logit contributions
ples+1.345
ount+1.239
 Jas+1.225
yl+1.188
ydia+1.185
Neg logit contributions
 tum-1.316
night-1.121
 thirst-1.119
 steak-1.067
 lung-1.029
Token She
Token ID1375
Feature activation+4.30e+00

Pos logit contributions
ples+3.960
ount+3.513
ptions+3.417
 angles+3.274
 blo+3.251
Neg logit contributions
 tum-4.841
night-3.896
 steak-3.733
 leg-3.731
time-3.718
Token is
Token ID318
Feature activation+2.42e+00

Pos logit contributions
ples+7.925
ount+7.540
od+7.468
 written+6.977
ydia+6.970
Neg logit contributions
 tum-7.932
 leg-6.617
 sled-6.436
night-6.223
 muscle-6.206
Token not
Token ID407
Feature activation+1.79e+00

Pos logit contributions
ples+6.112
ount+5.561
od+5.336
 Kay+5.330
 violet+5.330
Neg logit contributions
 tum-2.570
 thirst-1.935
 leg-1.911
 steak-1.756
 dinner-1.740
Token happy
Token ID3772
Feature activation+1.20e+00

Pos logit contributions
ples+3.666
ount+3.354
 violet+3.252
 paints+3.220
 Jas+3.213
Neg logit contributions
 tum-2.541
 thirst-2.135
 steak-2.023
night-2.006
 dinner-1.973
Token.
Token ID13
Feature activation+2.23e+00

Pos logit contributions
ples+2.130
ount+1.944
 Jas+1.922
yl+1.897
ydia+1.895
Neg logit contributions
 tum-1.985
night-1.691
 thirst-1.658
 steak-1.610
 dinner-1.540
Token
Token ID198
Feature activation+1.86e+00

Pos logit contributions
ples+4.273
ount+3.788
ptions+3.745
 blo+3.653
 Jas+3.619
Neg logit contributions
 tum-3.761
night-2.918
 steak-2.778
time-2.772
 leg-2.757
Token.
Token ID13
Feature activation-5.74e-02

Pos logit contributions
ples+0.860
ount+0.793
 Jas+0.787
yl+0.764
 paints+0.760
Neg logit contributions
 tum-0.880
night-0.768
 thirst-0.767
 steak-0.725
 lung-0.702
Token Sammy
Token ID44067
Feature activation+1.75e-01

Pos logit contributions
 tum+0.090
night+0.078
 thirst+0.078
 steak+0.074
 lung+0.071
Neg logit contributions
ples-0.103
 Jas-0.096
ount-0.095
 paints-0.092
yl-0.091
Token wanted
Token ID2227
Feature activation+1.24e-02

Pos logit contributions
ples+0.286
 Jas+0.262
ount+0.262
yl+0.253
 paints+0.253
Neg logit contributions
 tum-0.302
 thirst-0.265
night-0.263
 steak-0.248
 lung-0.241
Token to
Token ID284
Feature activation-3.87e-02

Pos logit contributions
ples+0.024
 Jas+0.022
ount+0.022
 paints+0.021
yl+0.021
Neg logit contributions
 tum-0.018
 thirst-0.015
night-0.015
 steak-0.014
 lung-0.014
Token touch
Token ID3638
Feature activation+6.15e-01

Pos logit contributions
 tum+0.066
 thirst+0.058
night+0.058
 steak+0.054
 lung+0.053
Neg logit contributions
ples-0.064
 Jas-0.059
ount-0.059
 paints-0.057
yl-0.057
Token the
Token ID262
Feature activation+4.30e+00

Pos logit contributions
ples+1.044
ount+0.966
 Jas+0.955
yl+0.926
 paints+0.917
Neg logit contributions
 tum-1.026
 thirst-0.894
night-0.879
 steak-0.846
 lung-0.818
Token staff
Token ID3085
Feature activation-5.87e-01

Pos logit contributions
ples+8.074
 Jas+7.222
ount+7.133
iddling+6.999
bugs+6.820
Neg logit contributions
 tum-8.344
 job-7.028
 dinner-6.898
 thirst-6.782
 leg-6.671
Token,
Token ID11
Feature activation-3.89e-01

Pos logit contributions
 tum+0.966
 thirst+0.850
night+0.848
 steak+0.773
 lung+0.754
Neg logit contributions
ples-1.003
 Jas-0.922
ount-0.920
 paints-0.903
 blo-0.896
Token but
Token ID475
Feature activation-4.44e-01

Pos logit contributions
 tum+0.635
night+0.570
 thirst+0.560
 steak+0.518
 lung+0.506
Neg logit contributions
ples-0.660
 Jas-0.617
ount-0.606
 paints-0.598
 Kay-0.585
Token Lucy
Token ID22162
Feature activation+2.18e-02

Pos logit contributions
 tum+0.624
night+0.539
 thirst+0.526
 steak+0.488
 lung+0.470
Neg logit contributions
ples-0.857
 Jas-0.821
ount-0.795
yl-0.788
 Kay-0.784
Token didn
Token ID1422
Feature activation-4.15e-01

Pos logit contributions
ples+0.035
 Jas+0.032
ount+0.032
yl+0.031
 paints+0.031
Neg logit contributions
 tum-0.038
 thirst-0.033
night-0.033
 steak-0.031
 lung-0.030
Token comes
Token ID2058
Feature activation+7.69e-01

Pos logit contributions
ples+7.586
ount+6.983
od+6.673
ydia+6.615
umes+6.540
Neg logit contributions
 tum-7.691
night-6.411
 leg-6.398
 drowned-6.287
 sled-6.279
Token to
Token ID284
Feature activation+1.00e+00

Pos logit contributions
ples+1.603
ount+1.504
 Jas+1.472
yl+1.444
 paints+1.443
Neg logit contributions
 tum-0.996
night-0.825
 thirst-0.823
 steak-0.760
 lung-0.727
Token their
Token ID511
Feature activation+7.24e-01

Pos logit contributions
ples+1.710
ount+1.582
 Jas+1.518
yl+1.487
 blo+1.482
Neg logit contributions
 tum-1.684
 thirst-1.464
night-1.454
 steak-1.397
 dinner-1.344
Token room
Token ID2119
Feature activation+8.58e-01

Pos logit contributions
ples+1.083
ount+0.992
 Jas+0.969
yl+0.937
ydia+0.921
Neg logit contributions
 tum-1.389
 thirst-1.215
night-1.199
 steak-1.158
 lung-1.129
Token.
Token ID13
Feature activation+1.81e+00

Pos logit contributions
ples+1.469
ount+1.354
 Jas+1.339
yl+1.299
ydia+1.297
Neg logit contributions
 tum-1.451
 thirst-1.236
night-1.231
 steak-1.172
 lung-1.130
Token She
Token ID1375
Feature activation+4.28e+00

Pos logit contributions
ples+2.402
ount+2.115
yl+1.998
ptions+1.988
 Jas+1.976
Neg logit contributions
 tum-4.054
night-3.454
 thirst-3.365
 steak-3.321
 lung-3.243
Token sees
Token ID7224
Feature activation+1.96e+00

Pos logit contributions
ples+7.725
ount+7.381
od+7.217
ydia+6.708
 written+6.697
Neg logit contributions
 tum-8.214
 leg-6.743
 drowned-6.401
 steak-6.390
 sled-6.352
Token them
Token ID606
Feature activation+9.78e-01

Pos logit contributions
ples+3.680
ount+3.369
 Jas+3.281
yl+3.273
ydia+3.194
Neg logit contributions
 tum-3.112
 steak-2.709
 thirst-2.579
night-2.522
 dinner-2.505
Token fighting
Token ID4330
Feature activation+6.42e-01

Pos logit contributions
ples+1.493
 Jas+1.363
ount+1.349
yl+1.320
ydia+1.319
Neg logit contributions
 tum-1.809
 thirst-1.593
night-1.562
 steak-1.519
 lung-1.483
Token over
Token ID625
Feature activation+6.57e-01

Pos logit contributions
ples+1.028
ount+0.944
 Jas+0.939
yl+0.914
 paints+0.897
Neg logit contributions
 tum-1.135
night-0.997
 thirst-0.986
 steak-0.947
 lung-0.914
Token the
Token ID262
Feature activation+6.45e-01

Pos logit contributions
ples+1.143
ount+1.066
 Jas+1.046
yl+1.016
ydia+1.008
Neg logit contributions
 tum-1.076
 thirst-0.934
night-0.915
 steak-0.887
 lung-0.850
Token what
Token ID644
Feature activation+1.41e+00

Pos logit contributions
ples+2.939
ydia+2.593
 Jas+2.524
yl+2.512
ount+2.493
Neg logit contributions
 tum-4.191
 thirst-3.601
 steak-3.553
night-3.495
 dinner-3.435
Token they
Token ID484
Feature activation+2.85e+00

Pos logit contributions
ples+2.646
ount+2.432
 Jas+2.309
ydia+2.308
ptions+2.270
Neg logit contributions
 tum-2.328
night-1.914
 steak-1.884
 thirst-1.858
 lung-1.775
Token are
Token ID389
Feature activation+2.37e+00

Pos logit contributions
ples+5.196
 Jas+5.103
ount+4.896
yl+4.797
ydia+4.775
Neg logit contributions
 tum-4.494
night-3.918
 thirst-3.829
 drowned-3.637
uch-3.602
Token doing
Token ID1804
Feature activation+7.89e-01

Pos logit contributions
ples+3.493
 Jas+3.141
ydia+3.108
ount+3.100
 Kay+3.021
Neg logit contributions
 tum-4.590
 thirst-4.050
 dinner-3.999
night-3.914
 steak-3.878
Token.
Token ID13
Feature activation+1.95e+00

Pos logit contributions
ples+1.327
ount+1.220
 Jas+1.208
yl+1.175
 paints+1.168
Neg logit contributions
 tum-1.320
night-1.162
 thirst-1.145
 steak-1.088
 lung-1.051
Token She
Token ID1375
Feature activation+4.26e+00

Pos logit contributions
ples+2.557
ptions+2.127
ount+2.117
yl+2.102
 Jas+2.020
Neg logit contributions
 tum-4.529
 thirst-3.764
night-3.743
 steak-3.622
 stomach-3.519
Token sees
Token ID7224
Feature activation+2.01e+00

Pos logit contributions
ples+7.403
ount+7.312
od+6.645
 Jem+6.483
ydia+6.403
Neg logit contributions
 tum-7.816
night-6.332
 leg-6.307
 steak-6.276
 sled-6.194
Token them
Token ID606
Feature activation+1.24e+00

Pos logit contributions
ples+3.578
ydia+3.116
ount+3.109
 Jas+3.105
yl+3.041
Neg logit contributions
 tum-3.437
 steak-2.927
 thirst-2.878
night-2.746
 dinner-2.694
Token in
Token ID287
Feature activation+1.67e+00

Pos logit contributions
ples+1.940
ount+1.791
 Jas+1.788
ydia+1.740
yl+1.727
Neg logit contributions
 tum-2.207
 thirst-1.938
night-1.905
 steak-1.843
 lung-1.819
Token the
Token ID262
Feature activation+1.73e+00

Pos logit contributions
ples+3.159
ount+3.095
 Jas+2.800
 paints+2.788
yl+2.772
Neg logit contributions
 tum-2.543
 steak-2.125
 thirst-2.124
night-2.035
 lung-2.031
Token hat
Token ID6877
Feature activation+3.72e-01

Pos logit contributions
ples+2.746
ount+2.555
 Jas+2.444
yl+2.344
 Kay+2.314
Neg logit contributions
 tum-3.328
night-2.838
 thirst-2.801
 lung-2.681
 steak-2.679
Token comes
Token ID2058
Feature activation+6.75e-01

Pos logit contributions
ples+7.906
ount+7.163
umes+6.957
ydia+6.931
od+6.893
Neg logit contributions
 tum-7.740
night-6.513
 leg-6.312
 drowned-6.238
 sled-6.103
Token to
Token ID284
Feature activation+1.02e+00

Pos logit contributions
ples+1.391
ount+1.304
 Jas+1.281
yl+1.257
 paints+1.254
Neg logit contributions
 tum-0.887
night-0.736
 thirst-0.735
 steak-0.677
 lung-0.651
Token the
Token ID262
Feature activation+9.92e-01

Pos logit contributions
ples+1.729
ount+1.585
 Jas+1.517
 paints+1.493
 blo+1.490
Neg logit contributions
 tum-1.733
 thirst-1.507
night-1.491
 steak-1.434
 dinner-1.381
Token box
Token ID3091
Feature activation+1.08e+00

Pos logit contributions
ples+1.467
ount+1.340
 Jas+1.326
yl+1.271
 Kay+1.234
Neg logit contributions
 tum-1.924
 thirst-1.689
night-1.659
 steak-1.643
 lung-1.604
Token.
Token ID13
Feature activation+2.35e+00

Pos logit contributions
ples+1.926
ount+1.785
 Jas+1.756
ydia+1.695
yl+1.693
Neg logit contributions
 tum-1.757
night-1.497
 thirst-1.485
 steak-1.433
 lung-1.375
Token She
Token ID1375
Feature activation+4.26e+00

Pos logit contributions
ples+3.328
ount+2.917
ptions+2.859
yl+2.706
 blo+2.642
Neg logit contributions
 tum-5.258
night-4.355
 thirst-4.190
 steak-4.157
 leg-4.114
Token says
Token ID1139
Feature activation+1.23e+00

Pos logit contributions
ples+7.762
ount+7.257
od+7.118
ydia+6.709
 Jas+6.532
Neg logit contributions
 tum-8.104
 leg-6.541
 sled-6.434
 steak-6.403
night-6.225
Token,
Token ID11
Feature activation+1.78e+00

Pos logit contributions
ples+2.673
ount+2.464
 Jas+2.406
ydia+2.387
ptions+2.352
Neg logit contributions
 tum-1.651
night-1.275
 thirst-1.262
 steak-1.231
 lung-1.133
Token "
Token ID366
Feature activation+1.73e+00

Pos logit contributions
ples+3.025
ount+2.792
 Jas+2.707
ydia+2.703
ptions+2.654
Neg logit contributions
 tum-3.187
 thirst-2.555
 steak-2.510
night-2.499
 lung-2.433
TokenStop
Token ID19485
Feature activation+1.65e+00

Pos logit contributions
ples+2.498
ount+2.221
 Jas+2.158
 paints+2.110
 blo+2.066
Neg logit contributions
 tum-3.306
night-2.953
 steak-2.946
 thirst-2.927
 lung-2.809
Token,
Token ID11
Feature activation+5.31e-01

Pos logit contributions
ples+3.063
ount+2.940
ydia+2.823
 Jas+2.789
 paints+2.733
Neg logit contributions
 tum-2.548
 thirst-2.133
night-2.117
 steak-2.075
 lung-1.998
Token is
Token ID318
Feature activation+1.96e+00

Pos logit contributions
ples+7.641
ptions+7.183
ount+7.145
ydia+6.980
 Jem+6.932
Neg logit contributions
 tum-5.370
night-4.670
 lung-4.334
 leg-4.165
 steak-4.132
Token Lily
Token ID20037
Feature activation+1.34e+00

Pos logit contributions
ples+4.329
ount+4.013
 paints+3.929
ptions+3.817
 angles+3.796
Neg logit contributions
 tum-2.427
 thirst-2.009
night-1.895
 steak-1.885
 dinner-1.821
Token's
Token ID338
Feature activation+9.37e-01

Pos logit contributions
ples+2.909
ount+2.795
 Jas+2.776
ydia+2.680
 paints+2.648
Neg logit contributions
 tum-1.652
night-1.312
 thirst-1.302
 lung-1.228
 steak-1.203
Token mom
Token ID1995
Feature activation+3.16e+00

Pos logit contributions
ples+1.763
ount+1.655
 Jas+1.616
yl+1.545
ptions+1.543
Neg logit contributions
 tum-1.492
 thirst-1.252
night-1.226
 steak-1.167
 lung-1.144
Token.
Token ID13
Feature activation+3.02e+00

Pos logit contributions
ydia+6.700
ples+6.601
 Jas+6.521
ount+6.520
 ey+6.266
Neg logit contributions
 tum-4.102
 lung-3.187
night-3.149
 steak-3.088
 thirst-3.077
Token She
Token ID1375
Feature activation+4.26e+00

Pos logit contributions
ples+7.048
ptions+6.683
 angles+6.320
ount+6.304
els+6.228
Neg logit contributions
 tum-4.237
 His-3.206
night-2.975
 thirst-2.743
 leg-2.666
Token has
Token ID468
Feature activation+1.42e+00

Pos logit contributions
ples+8.062
ount+7.853
ptions+7.730
ydia+7.668
 Jem+7.445
Neg logit contributions
 tum-7.158
night-6.246
 lung-5.992
 steak-5.704
 thirst-5.539
Token come
Token ID1282
Feature activation+8.59e-01

Pos logit contributions
ples+2.814
ount+2.663
 Jas+2.513
yl+2.459
ydia+2.438
Neg logit contributions
 tum-2.145
 thirst-1.733
 steak-1.688
night-1.687
 lung-1.604
Token to
Token ID284
Feature activation+5.81e-01

Pos logit contributions
ples+1.789
ount+1.674
 Jas+1.654
 paints+1.616
yl+1.610
Neg logit contributions
 tum-1.103
night-0.928
 thirst-0.905
 steak-0.832
 lung-0.812
Token pick
Token ID2298
Feature activation+1.52e+00

Pos logit contributions
ples+0.994
ount+0.914
 Jas+0.900
yl+0.875
 paints+0.871
Neg logit contributions
 tum-0.976
 thirst-0.849
night-0.840
 steak-0.794
 lung-0.768
Token them
Token ID606
Feature activation+1.01e-01

Pos logit contributions
ples+2.733
ount+2.495
 Jas+2.491
ptions+2.380
yl+2.368
Neg logit contributions
 tum-2.494
night-2.192
 thirst-2.133
 steak-2.106
 dinner-1.994
Token
Token ID198
Feature activation-4.72e-01

Pos logit contributions
 tum+0.208
 thirst+0.182
night+0.181
 steak+0.167
 lung+0.162
Neg logit contributions
ples-0.236
 Jas-0.221
ount-0.220
yl-0.213
 paints-0.213
TokenMom
Token ID29252
Feature activation+2.80e+00

Pos logit contributions
 tum+0.719
night+0.618
 thirst+0.610
 steak+0.555
 lung+0.543
Neg logit contributions
ples-0.888
 Jas-0.832
ount-0.827
yl-0.812
 paints-0.806
Token hears
Token ID23180
Feature activation+9.91e-01

Pos logit contributions
ples+4.192
 Kay+4.001
ount+3.980
ydia+3.889
yl+3.831
Neg logit contributions
 tum-5.354
night-4.642
 steak-4.567
 dinner-4.312
 drowned-4.280
Token them
Token ID606
Feature activation+6.17e-01

Pos logit contributions
ples+1.390
ount+1.258
 Jas+1.198
 paints+1.190
 blo+1.183
Neg logit contributions
 tum-1.974
 thirst-1.740
night-1.739
 steak-1.694
 dinner-1.623
Token.
Token ID13
Feature activation+2.43e+00

Pos logit contributions
ples+1.050
 Jas+0.962
ount+0.959
yl+0.937
 paints+0.927
Neg logit contributions
 tum-1.033
night-0.893
 thirst-0.892
 steak-0.842
 lung-0.818
Token She
Token ID1375
Feature activation+4.25e+00

Pos logit contributions
ples+3.366
ount+2.967
ptions+2.967
yl+2.785
 angles+2.750
Neg logit contributions
 tum-5.466
night-4.566
 thirst-4.395
 leg-4.323
 steak-4.316
Token comes
Token ID2058
Feature activation+6.75e-01

Pos logit contributions
ples+7.906
ount+7.163
umes+6.957
ydia+6.931
od+6.893
Neg logit contributions
 tum-7.740
night-6.513
 leg-6.312
 drowned-6.238
 sled-6.103
Token to
Token ID284
Feature activation+1.02e+00

Pos logit contributions
ples+1.391
ount+1.304
 Jas+1.281
yl+1.257
 paints+1.254
Neg logit contributions
 tum-0.887
night-0.736
 thirst-0.735
 steak-0.677
 lung-0.651
Token the
Token ID262
Feature activation+9.92e-01

Pos logit contributions
ples+1.729
ount+1.585
 Jas+1.517
 paints+1.493
 blo+1.490
Neg logit contributions
 tum-1.733
 thirst-1.507
night-1.491
 steak-1.434
 dinner-1.381
Token box
Token ID3091
Feature activation+1.08e+00

Pos logit contributions
ples+1.467
ount+1.340
 Jas+1.326
yl+1.271
 Kay+1.234
Neg logit contributions
 tum-1.924
 thirst-1.689
night-1.659
 steak-1.643
 lung-1.604
Token.
Token ID13
Feature activation+2.35e+00

Pos logit contributions
ples+1.926
ount+1.785
 Jas+1.756
ydia+1.695
yl+1.693
Neg logit contributions
 tum-1.757
night-1.497
 thirst-1.485
 steak-1.433
 lung-1.375
Token comes
Token ID2058
Feature activation+7.08e-01

Pos logit contributions
ples+7.240
ount+6.792
 Jem+6.344
ydia+6.330
od+6.275
Neg logit contributions
 tum-6.892
night-5.952
 muscle-5.836
 leg-5.802
 sled-5.734
Token to
Token ID284
Feature activation+1.46e+00

Pos logit contributions
ples+1.405
ount+1.315
 Jas+1.287
yl+1.262
 paints+1.262
Neg logit contributions
 tum-0.979
night-0.822
 thirst-0.821
 steak-0.762
 lung-0.734
Token their
Token ID511
Feature activation+6.11e-01

Pos logit contributions
ples+2.669
ount+2.470
 blo+2.332
 paints+2.293
 Jas+2.292
Neg logit contributions
 tum-2.298
 thirst-1.987
night-1.942
 steak-1.855
 dinner-1.849
Token room
Token ID2119
Feature activation+7.22e-01

Pos logit contributions
ples+0.902
ount+0.823
 Jas+0.808
yl+0.780
ydia+0.763
Neg logit contributions
 tum-1.180
 thirst-1.037
night-1.027
 steak-0.988
 lung-0.963
Token.
Token ID13
Feature activation+2.19e+00

Pos logit contributions
ples+1.282
ount+1.186
 Jas+1.174
yl+1.138
 paints+1.133
Neg logit contributions
 tum-1.168
 thirst-0.993
night-0.992
 steak-0.939
 lung-0.905
Token She
Token ID1375
Feature activation+4.23e+00

Pos logit contributions
ples+3.724
ount+3.362
ptions+3.195
 blo+3.097
yl+3.097
Neg logit contributions
 tum-4.061
night-3.288
 thirst-3.186
time-3.124
 leg-3.121
Token is
Token ID318
Feature activation+2.49e+00

Pos logit contributions
ples+7.818
ount+7.449
od+7.224
 written+6.815
ydia+6.748
Neg logit contributions
 tum-7.688
 leg-6.396
 sled-6.161
 muscle-6.070
 drowned-6.065
Token not
Token ID407
Feature activation+1.91e+00

Pos logit contributions
ples+6.537
ount+6.001
od+5.741
 violet+5.733
 Kay+5.731
Neg logit contributions
 tum-2.352
 thirst-1.748
 leg-1.693
 steak-1.560
 dinner-1.519
Token happy
Token ID3772
Feature activation+1.43e+00

Pos logit contributions
ples+3.977
ount+3.650
 violet+3.527
 paints+3.515
 Jas+3.488
Neg logit contributions
 tum-2.607
 thirst-2.208
 steak-2.106
night-2.062
 dinner-2.042
Token.
Token ID13
Feature activation+2.15e+00

Pos logit contributions
ples+2.613
ount+2.389
 Jas+2.349
yl+2.326
 violet+2.323
Neg logit contributions
 tum-2.312
night-1.947
 thirst-1.906
 steak-1.866
 dinner-1.796
Token
Token ID198
Feature activation+1.65e+00

Pos logit contributions
ples+4.177
ount+3.730
ptions+3.644
 blo+3.575
 Jas+3.534
Neg logit contributions
 tum-3.469
night-2.704
 thirst-2.605
time-2.576
 steak-2.549
Token
Token ID198
Feature activation-4.00e-01

Pos logit contributions
 tum+0.541
 thirst+0.478
night+0.477
 steak+0.440
 lung+0.426
Neg logit contributions
ples-0.585
 Jas-0.550
ount-0.544
yl-0.529
 paints-0.529
TokenMom
Token ID29252
Feature activation+2.78e+00

Pos logit contributions
 tum+0.653
 thirst+0.568
night+0.566
 steak+0.519
 lung+0.507
Neg logit contributions
ples-0.709
 Jas-0.658
ount-0.651
yl-0.643
 paints-0.637
Token hears
Token ID23180
Feature activation+1.19e+00

Pos logit contributions
ples+4.426
ount+4.199
 Kay+4.168
ydia+4.026
 Jas+4.006
Neg logit contributions
 tum-5.029
night-4.326
 steak-4.277
 leg-4.053
 dinner-4.049
Token them
Token ID606
Feature activation+5.77e-01

Pos logit contributions
ples+1.739
ount+1.590
 Jas+1.504
 blo+1.499
yl+1.493
Neg logit contributions
 tum-2.306
night-2.034
 thirst-2.017
 steak-1.996
 dinner-1.907
Token.
Token ID13
Feature activation+2.44e+00

Pos logit contributions
ples+0.980
 Jas+0.897
ount+0.894
yl+0.872
 paints+0.866
Neg logit contributions
 tum-0.969
night-0.837
 thirst-0.837
 steak-0.789
 lung-0.768
Token She
Token ID1375
Feature activation+4.22e+00

Pos logit contributions
ples+3.494
ount+3.092
ptions+3.006
yl+2.846
 angles+2.839
Neg logit contributions
 tum-5.300
night-4.431
time-4.249
 leg-4.247
 steak-4.236
Token comes
Token ID2058
Feature activation+8.56e-01

Pos logit contributions
ples+7.813
ount+7.109
od+6.803
 Jem+6.687
umes+6.684
Neg logit contributions
 tum-7.625
night-6.550
 leg-6.434
 drowned-6.321
 muscle-6.316
Token to
Token ID284
Feature activation+1.19e+00

Pos logit contributions
ples+1.798
ount+1.688
 Jas+1.647
 paints+1.618
yl+1.618
Neg logit contributions
 tum-1.094
night-0.903
 thirst-0.898
 steak-0.829
 lung-0.797
Token see
Token ID766
Feature activation+2.29e+00

Pos logit contributions
ples+2.047
ount+1.889
 Jas+1.786
 blo+1.773
 paints+1.764
Neg logit contributions
 tum-2.007
 thirst-1.741
night-1.731
 steak-1.658
 dinner-1.610
Token what
Token ID644
Feature activation+1.76e+00

Pos logit contributions
ples+3.346
ount+2.918
ydia+2.886
yl+2.873
 Jas+2.867
Neg logit contributions
 tum-4.631
 steak-4.095
 dinner-3.927
night-3.913
 thirst-3.899
Token is
Token ID318
Feature activation+2.95e+00

Pos logit contributions
ples+2.952
ount+2.676
 Jas+2.470
ydia+2.459
 violet+2.457
Neg logit contributions
 tum-3.312
night-2.813
 steak-2.787
time-2.647
 dinner-2.646
Token comes
Token ID2058
Feature activation+9.29e-01

Pos logit contributions
ples+7.262
ount+6.653
ydia+6.480
ptions+6.229
od+6.212
Neg logit contributions
 tum-7.545
night-6.401
 sled-6.263
 leg-6.186
 drowned-6.133
Token to
Token ID284
Feature activation+9.70e-01

Pos logit contributions
ples+1.933
ount+1.817
 Jas+1.761
 paints+1.735
yl+1.732
Neg logit contributions
 tum-1.200
night-0.992
 thirst-0.986
 steak-0.911
 lung-0.878
Token the
Token ID262
Feature activation+1.70e+00

Pos logit contributions
ples+1.584
ount+1.454
 Jas+1.397
 paints+1.362
 blo+1.361
Neg logit contributions
 tum-1.703
 thirst-1.488
night-1.484
 steak-1.422
 lung-1.370
Token yard
Token ID12699
Feature activation+1.40e+00

Pos logit contributions
ples+2.808
ount+2.584
 Jas+2.514
yl+2.419
 Kay+2.384
Neg logit contributions
 tum-3.226
 steak-2.719
 thirst-2.688
 dinner-2.626
 lung-2.624
Token.
Token ID13
Feature activation+1.75e+00

Pos logit contributions
ples+2.554
ount+2.375
 Jas+2.342
ydia+2.317
yl+2.286
Neg logit contributions
 tum-2.221
 thirst-1.851
night-1.842
 steak-1.771
 leg-1.730
Token She
Token ID1375
Feature activation+4.22e+00

Pos logit contributions
ples+2.143
ount+1.888
ptions+1.768
yl+1.746
 Jas+1.717
Neg logit contributions
 tum-4.065
night-3.505
 thirst-3.381
 steak-3.356
 lung-3.286
Token sees
Token ID7224
Feature activation+1.59e+00

Pos logit contributions
ples+7.542
ount+7.122
od+6.991
ydia+6.686
 written+6.600
Neg logit contributions
 tum-7.989
 sled-6.639
 leg-6.563
 drowned-6.345
 steak-6.298
Token the
Token ID262
Feature activation+2.01e+00

Pos logit contributions
ples+2.892
ount+2.653
 Jas+2.619
yl+2.569
ydia+2.562
Neg logit contributions
 tum-2.552
 steak-2.222
 thirst-2.126
night-2.122
 dinner-2.025
Token box
Token ID3091
Feature activation+2.18e+00

Pos logit contributions
ples+3.735
ount+3.474
 Jas+3.352
ain+3.248
ydia+3.203
Neg logit contributions
 tum-3.522
 steak-2.938
 thirst-2.880
 lung-2.807
 dinner-2.761
Token and
Token ID290
Feature activation+3.39e+00

Pos logit contributions
ples+4.382
ount+4.131
ydia+4.128
 Jas+4.050
ptions+3.970
Neg logit contributions
 tum-3.107
 leg-2.462
 thirst-2.439
night-2.413
 steak-2.382
Token the
Token ID262
Feature activation+1.67e+00

Pos logit contributions
ples+8.173
ount+7.822
yl+7.309
ydia+7.256
od+7.209
Neg logit contributions
 tum-3.939
 steak-2.972
 leg-2.852
 sled-2.793
 dinner-2.777
Token Sara
Token ID24799
Feature activation+2.03e-01

Pos logit contributions
ples+1.398
ount+1.237
 Jas+1.209
 paints+1.192
yl+1.174
Neg logit contributions
 tum-2.076
 thirst-1.835
night-1.820
 steak-1.779
 lung-1.701
Token and
Token ID290
Feature activation+1.68e+00

Pos logit contributions
ples+0.361
ount+0.334
 Jas+0.334
yl+0.323
 paints+0.323
Neg logit contributions
 tum-0.321
 thirst-0.278
night-0.276
 steak-0.259
 lung-0.251
Token Ben
Token ID3932
Feature activation+6.32e-01

Pos logit contributions
ples+2.810
ount+2.543
yl+2.517
 Jas+2.461
ydia+2.441
Neg logit contributions
 tum-3.086
night-2.662
 thirst-2.644
 steak-2.507
 lung-2.437
Token fighting
Token ID4330
Feature activation+7.77e-01

Pos logit contributions
ples+1.009
 Jas+0.922
ount+0.919
yl+0.895
 paints+0.882
Neg logit contributions
 tum-1.123
 thirst-0.981
night-0.976
 steak-0.923
 lung-0.903
Token.
Token ID13
Feature activation+1.65e+00

Pos logit contributions
ples+1.248
ount+1.140
 Jas+1.132
yl+1.106
 paints+1.084
Neg logit contributions
 tum-1.377
night-1.219
 thirst-1.197
 steak-1.146
 lung-1.105
Token She
Token ID1375
Feature activation+4.20e+00

Pos logit contributions
ples+2.048
ount+1.729
ptions+1.683
yl+1.653
 Jas+1.631
Neg logit contributions
 tum-3.808
 thirst-3.248
night-3.227
 steak-3.136
 lung-3.064
Token comes
Token ID2058
Feature activation+7.14e-01

Pos logit contributions
ples+8.389
ount+7.646
 Jem+7.481
illian+7.434
 Jas+7.389
Neg logit contributions
 tum-6.776
night-5.670
 leg-5.627
 muscle-5.619
 drowned-5.477
Token to
Token ID284
Feature activation+8.84e-01

Pos logit contributions
ples+1.454
ount+1.363
 Jas+1.339
yl+1.310
 paints+1.308
Neg logit contributions
 tum-0.955
night-0.798
 thirst-0.797
 steak-0.731
 lung-0.707
Token see
Token ID766
Feature activation+2.00e+00

Pos logit contributions
ples+1.478
ount+1.344
 Jas+1.302
 paints+1.277
yl+1.271
Neg logit contributions
 tum-1.535
 thirst-1.341
night-1.323
 steak-1.261
 lung-1.217
Token what
Token ID644
Feature activation+1.30e+00

Pos logit contributions
ples+3.026
 Jas+2.587
ydia+2.558
ptions+2.534
yl+2.529
Neg logit contributions
 tum-4.045
 thirst-3.469
 steak-3.438
night-3.329
 lung-3.269
Token is
Token ID318
Feature activation+2.79e+00

Pos logit contributions
ples+2.262
ount+2.026
 Jas+1.953
ydia+1.910
ptions+1.903
Neg logit contributions
 tum-2.346
night-1.973
 steak-1.941
 thirst-1.916
 lung-1.833
Token and
Token ID290
Feature activation+3.34e+00

Pos logit contributions
ples+1.028
 Jas+0.946
ount+0.943
yl+0.920
 paints+0.912
Neg logit contributions
 tum-0.921
night-0.790
 thirst-0.789
 steak-0.742
 lung-0.719
Token comes
Token ID2058
Feature activation+1.32e+00

Pos logit contributions
ples+4.928
ount+4.439
ydia+4.297
 written+4.041
 blo+4.034
Neg logit contributions
 tum-6.999
night-6.228
 leg-5.778
 sled-5.768
 muscle-5.738
Token to
Token ID284
Feature activation+1.18e+00

Pos logit contributions
ples+2.925
ount+2.764
 violet+2.660
 Jas+2.657
ydia+2.648
Neg logit contributions
 tum-1.535
night-1.236
 thirst-1.216
 steak-1.110
 leg-1.088
Token see
Token ID766
Feature activation+2.53e+00

Pos logit contributions
ples+2.001
ount+1.848
 Jas+1.736
 blo+1.730
 paints+1.729
Neg logit contributions
 tum-2.004
 thirst-1.745
night-1.738
 steak-1.658
 dinner-1.623
Token.
Token ID13
Feature activation+1.88e+00

Pos logit contributions
ples+3.739
ount+3.336
 violet+3.257
yl+3.248
ydia+3.228
Neg logit contributions
 tum-5.086
 steak-4.385
 dinner-4.332
 thirst-4.254
night-4.247
Token She
Token ID1375
Feature activation+4.20e+00

Pos logit contributions
ples+2.343
ount+2.040
ptions+1.877
yl+1.873
 blo+1.861
Neg logit contributions
 tum-4.358
night-3.727
 thirst-3.609
 steak-3.556
 leg-3.514
Token is
Token ID318
Feature activation+2.29e+00

Pos logit contributions
ples+8.180
ount+7.573
od+7.383
ydia+7.044
 written+6.936
Neg logit contributions
 tum-7.388
 leg-5.931
night-5.733
 drowned-5.697
 sled-5.637
Token not
Token ID407
Feature activation+1.92e+00

Pos logit contributions
ples+5.903
ount+5.403
 violet+5.182
 Jas+5.179
od+5.159
Neg logit contributions
 tum-2.209
 thirst-1.685
 leg-1.652
 steak-1.512
 dinner-1.464
Token happy
Token ID3772
Feature activation+1.50e+00

Pos logit contributions
ples+4.016
ount+3.675
 violet+3.548
 paints+3.545
 Jas+3.523
Neg logit contributions
 tum-2.607
 thirst-2.194
 steak-2.114
night-2.048
 dinner-2.040
Token.
Token ID13
Feature activation+2.03e+00

Pos logit contributions
ples+2.772
ount+2.539
 Jas+2.490
 violet+2.475
ydia+2.462
Neg logit contributions
 tum-2.403
night-2.023
 thirst-1.963
 steak-1.938
 dinner-1.864
Token She
Token ID1375
Feature activation+3.95e+00

Pos logit contributions
ples+2.640
ount+2.253
ptions+2.132
 blo+2.114
yl+2.083
Neg logit contributions
 tum-4.664
night-3.925
 thirst-3.794
 leg-3.736
 steak-3.724
Token
Token ID198
Feature activation-4.17e-01

Pos logit contributions
 tum+0.817
night+0.729
 thirst+0.727
 steak+0.670
 lung+0.650
Neg logit contributions
ples-0.833
 Jas-0.794
ount-0.780
yl-0.761
 paints-0.754
TokenMom
Token ID29252
Feature activation+2.71e+00

Pos logit contributions
 tum+0.654
night+0.565
 thirst+0.560
 steak+0.518
 lung+0.503
Neg logit contributions
ples-0.765
 Jas-0.717
ount-0.708
yl-0.698
 paints-0.687
Token hears
Token ID23180
Feature activation+1.43e+00

Pos logit contributions
ples+3.968
ount+3.824
 Kay+3.781
ydia+3.658
Kay+3.609
Neg logit contributions
 tum-5.238
night-4.544
 steak-4.482
 sled-4.269
 dinner-4.239
Token them
Token ID606
Feature activation+5.41e-01

Pos logit contributions
ples+2.252
ount+2.057
 blo+1.967
 paints+1.939
yl+1.927
Neg logit contributions
 tum-2.621
night-2.277
 thirst-2.266
 steak-2.238
 dinner-2.173
Token.
Token ID13
Feature activation+2.60e+00

Pos logit contributions
ples+0.947
 Jas+0.869
ount+0.868
yl+0.844
 paints+0.839
Neg logit contributions
 tum-0.878
 thirst-0.756
night-0.756
 steak-0.713
 lung-0.693
Token She
Token ID1375
Feature activation+4.20e+00

Pos logit contributions
ples+3.904
ount+3.481
ptions+3.408
 angles+3.276
yl+3.154
Neg logit contributions
 tum-5.500
night-4.523
 leg-4.393
time-4.363
 thirst-4.310
Token comes
Token ID2058
Feature activation+8.28e-01

Pos logit contributions
ples+7.302
ount+6.696
ydia+6.428
umes+6.303
od+6.299
Neg logit contributions
 tum-7.806
night-6.741
 sled-6.696
 leg-6.671
 muscle-6.614
Token to
Token ID284
Feature activation+1.29e+00

Pos logit contributions
ples+1.720
ount+1.613
 Jas+1.574
yl+1.548
 paints+1.547
Neg logit contributions
 tum-1.074
night-0.890
 thirst-0.884
 steak-0.819
 lung-0.787
Token their
Token ID511
Feature activation+6.93e-01

Pos logit contributions
ples+2.234
ount+2.061
 Jas+1.937
 blo+1.934
 paints+1.922
Neg logit contributions
 tum-2.148
 thirst-1.858
night-1.836
 steak-1.761
 dinner-1.728
Token room
Token ID2119
Feature activation+7.82e-01

Pos logit contributions
ples+1.044
ount+0.952
 Jas+0.929
yl+0.902
ydia+0.887
Neg logit contributions
 tum-1.324
 thirst-1.158
night-1.144
 steak-1.101
 lung-1.076
Token.
Token ID13
Feature activation+2.42e+00

Pos logit contributions
ples+1.345
ount+1.239
 Jas+1.225
yl+1.188
ydia+1.185
Neg logit contributions
 tum-1.316
night-1.121
 thirst-1.119
 steak-1.067
 lung-1.029
Token
Token ID198
Feature activation-8.46e-01

Pos logit contributions
 tum+0.687
 thirst+0.612
night+0.611
 steak+0.561
 lung+0.546
Neg logit contributions
ples-0.724
 Jas-0.684
ount-0.676
 paints-0.656
yl-0.656
TokenMom
Token ID29252
Feature activation+2.73e+00

Pos logit contributions
 tum+1.466
night+1.293
 thirst+1.275
 steak+1.164
 lung+1.146
Neg logit contributions
ples-1.422
 Jas-1.302
ount-1.291
 paints-1.278
yl-1.275
Token hears
Token ID23180
Feature activation+7.29e-01

Pos logit contributions
ples+4.149
ount+3.991
 Kay+3.971
ydia+3.817
 Jas+3.809
Neg logit contributions
 tum-5.103
night-4.388
 steak-4.386
 dinner-4.142
 leg-4.096
Token them
Token ID606
Feature activation+6.04e-01

Pos logit contributions
ples+0.993
ount+0.903
 Jas+0.877
yl+0.854
 paints+0.848
Neg logit contributions
 tum-1.472
 thirst-1.308
night-1.308
 steak-1.266
 lung-1.217
Token.
Token ID13
Feature activation+1.97e+00

Pos logit contributions
ples+1.031
 Jas+0.945
ount+0.942
yl+0.918
 paints+0.910
Neg logit contributions
 tum-1.010
 thirst-0.873
night-0.872
 steak-0.824
 lung-0.800
Token She
Token ID1375
Feature activation+4.19e+00

Pos logit contributions
ples+2.518
ount+2.205
ptions+2.118
yl+2.084
 Jas+2.030
Neg logit contributions
 tum-4.518
night-3.858
 thirst-3.757
 steak-3.715
 leg-3.633
Token comes
Token ID2058
Feature activation+7.69e-01

Pos logit contributions
ples+7.586
ount+6.983
od+6.673
ydia+6.615
umes+6.540
Neg logit contributions
 tum-7.691
night-6.411
 leg-6.398
 drowned-6.287
 sled-6.279
Token to
Token ID284
Feature activation+1.00e+00

Pos logit contributions
ples+1.603
ount+1.504
 Jas+1.472
yl+1.444
 paints+1.443
Neg logit contributions
 tum-0.996
night-0.825
 thirst-0.823
 steak-0.760
 lung-0.727
Token their
Token ID511
Feature activation+7.24e-01

Pos logit contributions
ples+1.710
ount+1.582
 Jas+1.518
yl+1.487
 blo+1.482
Neg logit contributions
 tum-1.684
 thirst-1.464
night-1.454
 steak-1.397
 dinner-1.344
Token room
Token ID2119
Feature activation+8.58e-01

Pos logit contributions
ples+1.083
ount+0.992
 Jas+0.969
yl+0.937
ydia+0.921
Neg logit contributions
 tum-1.389
 thirst-1.215
night-1.199
 steak-1.158
 lung-1.129
Token.
Token ID13
Feature activation+1.81e+00

Pos logit contributions
ples+1.469
ount+1.354
 Jas+1.339
yl+1.299
ydia+1.297
Neg logit contributions
 tum-1.451
 thirst-1.236
night-1.231
 steak-1.172
 lung-1.130
Token comes
Token ID2058
Feature activation+6.82e-01

Pos logit contributions
ples+7.147
ount+6.561
ydia+6.400
 written+6.228
od+6.194
Neg logit contributions
 tum-7.572
 sled-6.379
 leg-6.349
night-6.334
 muscle-6.170
Token to
Token ID284
Feature activation+1.00e+00

Pos logit contributions
ples+1.378
ount+1.292
 Jas+1.265
 paints+1.242
yl+1.240
Neg logit contributions
 tum-0.921
night-0.773
 thirst-0.771
 steak-0.711
 lung-0.684
Token the
Token ID262
Feature activation+9.13e-01

Pos logit contributions
ples+1.650
ount+1.498
 Jas+1.446
 paints+1.419
ydia+1.412
Neg logit contributions
 tum-1.748
 thirst-1.532
night-1.529
 steak-1.459
 dinner-1.409
Token room
Token ID2119
Feature activation+1.42e+00

Pos logit contributions
ples+1.341
ount+1.217
 Jas+1.210
yl+1.168
 Kay+1.129
Neg logit contributions
 tum-1.797
 thirst-1.567
night-1.539
 steak-1.519
 lung-1.483
Token.
Token ID13
Feature activation+1.86e+00

Pos logit contributions
ples+2.592
ount+2.405
 Jas+2.346
ydia+2.324
yl+2.299
Neg logit contributions
 tum-2.271
night-1.894
 thirst-1.878
 steak-1.793
 leg-1.745
Token She
Token ID1375
Feature activation+4.18e+00

Pos logit contributions
ples+2.631
ount+2.276
ptions+2.188
ydia+2.186
 Jas+2.157
Neg logit contributions
 tum-3.999
night-3.397
 thirst-3.311
 steak-3.243
 lung-3.232
Token sees
Token ID7224
Feature activation+1.91e+00

Pos logit contributions
ples+7.436
ount+7.051
od+6.971
ydia+6.525
 written+6.445
Neg logit contributions
 tum-8.097
 leg-6.635
 sled-6.451
night-6.262
 steak-6.230
Token Anna
Token ID11735
Feature activation+1.23e+00

Pos logit contributions
ples+3.577
ount+3.207
ydia+3.146
 Jas+3.141
yl+3.087
Neg logit contributions
 tum-3.019
 steak-2.611
 thirst-2.499
night-2.457
 dinner-2.425
Token and
Token ID290
Feature activation+1.72e+00

Pos logit contributions
ples+2.413
 Jas+2.296
ount+2.278
 Kay+2.210
ydia+2.204
Neg logit contributions
 tum-1.723
 thirst-1.435
night-1.415
 steak-1.373
 lung-1.356
Token Ben
Token ID3932
Feature activation+1.12e+00

Pos logit contributions
ples+3.468
ount+3.257
yl+3.149
ydia+3.068
 Jas+3.058
Neg logit contributions
 tum-2.538
night-2.071
 thirst-2.064
 steak-2.043
 lung-1.925
Token struggle
Token ID6531
Feature activation+8.10e-01

Pos logit contributions
ples+1.827
 Jas+1.698
ount+1.677
yl+1.640
ydia+1.625
Neg logit contributions
 tum-1.969
 thirst-1.715
night-1.672
 steak-1.620
 lung-1.600
Token It
Token ID632
Feature activation+3.55e+00

Pos logit contributions
ples+4.121
ptions+3.841
ount+3.823
 Jas+3.682
ydia+3.670
Neg logit contributions
 tum-3.366
night-2.838
 lung-2.740
 steak-2.703
 His-2.658
Token is
Token ID318
Feature activation+1.98e+00

Pos logit contributions
ples+7.288
ount+6.803
ptions+6.759
ydia+6.713
 Jem+6.603
Neg logit contributions
 tum-5.278
night-4.595
 lung-4.351
 leg-4.202
 steak-4.162
Token their
Token ID511
Feature activation+9.04e-01

Pos logit contributions
ples+4.213
ount+3.908
 paints+3.834
ptions+3.685
iddling+3.677
Neg logit contributions
 tum-2.572
 thirst-2.151
 steak-2.069
night-2.060
 lung-2.018
Token mom
Token ID1995
Feature activation+2.89e+00

Pos logit contributions
ples+2.260
ount+2.124
 Jas+2.069
ptions+2.020
ydia+2.007
Neg logit contributions
 tum-0.892
 thirst-0.641
night-0.641
 steak-0.597
 lung-0.571
Token.
Token ID13
Feature activation+2.61e+00

Pos logit contributions
ydia+5.848
ples+5.847
ount+5.692
 Jas+5.659
 paints+5.568
Neg logit contributions
 tum-3.876
 lung-3.093
 steak-3.060
night-3.037
 thirst-2.995
Token She
Token ID1375
Feature activation+4.18e+00

Pos logit contributions
ples+5.020
ptions+4.723
ount+4.562
 paints+4.367
ydia+4.336
Neg logit contributions
 tum-4.453
night-3.579
 His-3.508
 lung-3.407
 steak-3.356
Token is
Token ID318
Feature activation+2.95e+00

Pos logit contributions
ples+7.905
ount+7.628
ptions+7.534
ydia+7.462
 Jem+7.217
Neg logit contributions
 tum-6.974
 lung-6.063
night-5.974
 sled-5.773
 steak-5.725
Token holding
Token ID4769
Feature activation+7.76e-01

Pos logit contributions
ples+6.397
ount+5.696
pill+5.508
 paints+5.482
 angles+5.386
Neg logit contributions
 tum-4.098
 thirst-3.404
 leg-3.372
 steak-3.318
 dinner-3.286
Token a
Token ID257
Feature activation+1.28e+00

Pos logit contributions
ples+1.383
ount+1.294
 Jas+1.265
yl+1.233
 paints+1.224
Neg logit contributions
 tum-1.236
 thirst-1.079
night-1.063
 steak-1.020
 lung-0.980
Token shiny
Token ID22441
Feature activation+1.07e+00

Pos logit contributions
ples+2.244
ount+2.119
 Jas+2.054
ydia+1.924
ain+1.895
Neg logit contributions
 tum-2.233
 steak-1.865
 thirst-1.855
night-1.822
 lung-1.793
Token key
Token ID1994
Feature activation+1.06e+00

Pos logit contributions
ples+1.812
ount+1.711
 Jas+1.679
ydia+1.571
 Kay+1.548
Neg logit contributions
 tum-1.887
night-1.621
 thirst-1.616
 steak-1.586
 lung-1.533
Token decided
Token ID3066
Feature activation+3.24e-01

Pos logit contributions
 tum+0.125
 thirst+0.110
night+0.110
 steak+0.104
 lung+0.101
Neg logit contributions
ples-0.108
 Jas-0.099
ount-0.098
yl-0.096
 paints-0.095
Token to
Token ID284
Feature activation-1.49e-01

Pos logit contributions
ples+0.648
ount+0.604
 Jas+0.602
 paints+0.585
yl+0.584
Neg logit contributions
 tum-0.444
 thirst-0.374
night-0.372
 steak-0.345
 lung-0.331
Token go
Token ID467
Feature activation+1.67e+00

Pos logit contributions
 tum+0.247
 thirst+0.216
night+0.215
 steak+0.203
 lung+0.196
Neg logit contributions
ples-0.251
 Jas-0.233
ount-0.231
 paints-0.225
yl-0.224
Token and
Token ID290
Feature activation-1.36e-01

Pos logit contributions
ples+3.627
ount+3.277
 Jas+3.251
ptions+3.186
 paints+3.170
Neg logit contributions
 tum-2.138
 thirst-1.793
night-1.782
 steak-1.659
 leg-1.636
Token touch
Token ID3638
Feature activation+8.15e-01

Pos logit contributions
 tum+0.241
 thirst+0.213
night+0.212
 steak+0.201
 lung+0.195
Neg logit contributions
ples-0.214
 Jas-0.198
ount-0.196
 paints-0.192
yl-0.190
Token the
Token ID262
Feature activation+4.18e+00

Pos logit contributions
ples+1.389
ount+1.296
 Jas+1.271
yl+1.237
 paints+1.225
Neg logit contributions
 tum-1.352
 thirst-1.181
night-1.159
 steak-1.113
 lung-1.078
Token staff
Token ID3085
Feature activation-3.76e-01

Pos logit contributions
ples+7.597
 Jas+7.017
ount+6.918
iddling+6.749
 typing+6.616
Neg logit contributions
 tum-8.046
 thirst-6.705
 dinner-6.485
 job-6.445
 lesson-6.379
Token,
Token ID11
Feature activation-1.89e-01

Pos logit contributions
 tum+0.645
 thirst+0.568
night+0.567
 steak+0.526
 lung+0.511
Neg logit contributions
ples-0.616
 Jas-0.565
ount-0.563
 paints-0.550
yl-0.545
Token even
Token ID772
Feature activation+2.15e-01

Pos logit contributions
 tum+0.306
night+0.269
 thirst+0.267
 steak+0.250
 lung+0.242
Neg logit contributions
ples-0.327
 Jas-0.303
ount-0.300
 paints-0.294
yl-0.290
Token though
Token ID996
Feature activation+2.58e-01

Pos logit contributions
ples+0.391
ount+0.363
 Jas+0.362
yl+0.351
 paints+0.351
Neg logit contributions
 tum-0.331
 thirst-0.286
night-0.284
 steak-0.266
 lung-0.257
Token he
Token ID339
Feature activation-3.18e-01

Pos logit contributions
ples+0.468
ount+0.433
 Jas+0.432
yl+0.419
 paints+0.419
Neg logit contributions
 tum-0.400
 thirst-0.345
night-0.343
 steak-0.322
 lung-0.311
Token and
Token ID290
Feature activation+2.97e+00

Pos logit contributions
ples+0.910
 Jas+0.838
ount+0.836
yl+0.814
 paints+0.808
Neg logit contributions
 tum-0.834
 thirst-0.717
night-0.717
 steak-0.673
 lung-0.653
Token comes
Token ID2058
Feature activation+1.42e+00

Pos logit contributions
ples+4.455
ount+4.003
ydia+3.897
yl+3.778
 written+3.736
Neg logit contributions
 tum-6.115
night-5.374
 sled-5.054
 steak-5.036
 thirst-5.028
Token to
Token ID284
Feature activation+1.14e+00

Pos logit contributions
ples+3.155
ount+2.979
 violet+2.873
 Jas+2.860
yl+2.855
Neg logit contributions
 tum-1.664
night-1.318
 thirst-1.302
 steak-1.178
 leg-1.173
Token see
Token ID766
Feature activation+1.89e+00

Pos logit contributions
ples+1.899
ount+1.748
 Jas+1.656
 paints+1.640
yl+1.635
Neg logit contributions
 tum-1.988
 thirst-1.730
night-1.729
 steak-1.635
 dinner-1.598
Token.
Token ID13
Feature activation+1.76e+00

Pos logit contributions
ples+2.609
ount+2.319
yl+2.297
 Jas+2.284
ydia+2.282
Neg logit contributions
 tum-3.933
night-3.433
 steak-3.422
 thirst-3.404
 dinner-3.325
Token She
Token ID1375
Feature activation+4.18e+00

Pos logit contributions
ples+2.416
ount+2.157
ptions+2.033
 Jas+2.007
yl+1.996
Neg logit contributions
 tum-3.830
night-3.280
 thirst-3.187
 steak-3.109
 lung-3.062
Token sees
Token ID7224
Feature activation+1.85e+00

Pos logit contributions
ples+7.681
ount+7.236
od+7.191
 written+6.834
ydia+6.681
Neg logit contributions
 tum-7.678
 leg-6.354
 drowned-6.173
 sled-6.138
night-5.971
Token the
Token ID262
Feature activation+2.18e+00

Pos logit contributions
ples+3.390
ount+3.085
 Jas+3.029
yl+3.014
ydia+2.966
Neg logit contributions
 tum-2.966
 steak-2.565
 thirst-2.472
night-2.450
 dinner-2.366
Token broken
Token ID5445
Feature activation+1.88e+00

Pos logit contributions
ples+3.814
ount+3.716
 Jas+3.637
yl+3.305
 Kay+3.300
Neg logit contributions
 tum-4.033
 steak-3.317
 lung-3.305
 thirst-3.266
 leg-3.233
Token comb
Token ID1974
Feature activation+6.13e-01

Pos logit contributions
ount+3.119
ples+3.077
 Jas+2.986
ptions+2.748
ydia+2.712
Neg logit contributions
 tum-3.463
 lung-2.913
 thirst-2.876
 steak-2.868
night-2.865
Token and
Token ID290
Feature activation+2.57e+00

Pos logit contributions
ples+1.051
ount+0.974
 Jas+0.971
yl+0.932
ydia+0.924
Neg logit contributions
 tum-1.014
 thirst-0.874
night-0.873
 steak-0.828
 lung-0.810
Token It
Token ID632
Feature activation+3.40e+00

Pos logit contributions
ples+3.825
ptions+3.529
ount+3.408
 paints+3.362
 Jas+3.352
Neg logit contributions
 tum-3.648
night-3.010
 thirst-2.928
 steak-2.902
 lung-2.858
Token is
Token ID318
Feature activation+1.79e+00

Pos logit contributions
ples+6.973
ount+6.495
ptions+6.414
ydia+6.319
 Jem+6.214
Neg logit contributions
 tum-5.223
night-4.525
 lung-4.208
 steak-4.091
 leg-4.035
Token their
Token ID511
Feature activation-1.08e+00

Pos logit contributions
ples+3.905
ount+3.629
 paints+3.558
 Jas+3.434
ptions+3.431
Neg logit contributions
 tum-2.248
 thirst-1.869
night-1.776
 steak-1.737
 lung-1.701
Token mom
Token ID1995
Feature activation+2.63e+00

Pos logit contributions
 tum+1.450
 thirst+1.358
night+1.330
 steak+1.147
 lung+1.117
Neg logit contributions
 paints-2.004
ples-1.984
 Jas-1.961
yl-1.882
 blo-1.879
Token.
Token ID13
Feature activation+2.55e+00

Pos logit contributions
ples+5.238
ydia+5.145
ount+5.023
 Jas+5.006
 violet+4.936
Neg logit contributions
 tum-3.721
night-2.944
 lung-2.940
 thirst-2.927
 steak-2.922
Token She
Token ID1375
Feature activation+4.18e+00

Pos logit contributions
ples+4.547
ptions+4.187
ount+3.927
 paints+3.853
iddling+3.809
Neg logit contributions
 tum-4.956
night-3.947
 His-3.800
 thirst-3.776
 steak-3.766
Token comes
Token ID2058
Feature activation+2.95e-01

Pos logit contributions
ples+8.288
ount+7.966
ptions+7.883
ydia+7.821
 Jas+7.479
Neg logit contributions
 tum-7.016
 lung-5.935
night-5.910
 steak-5.621
 drowned-5.582
Token into
Token ID656
Feature activation+1.26e+00

Pos logit contributions
ples+0.564
ount+0.524
 Jas+0.522
yl+0.507
 paints+0.507
Neg logit contributions
 tum-0.432
 thirst-0.368
night-0.366
 steak-0.341
 lung-0.330
Token the
Token ID262
Feature activation+4.70e-01

Pos logit contributions
ples+2.320
ount+2.181
 Jas+2.063
ydia+2.051
 paints+2.038
Neg logit contributions
 tum-1.946
 thirst-1.664
night-1.633
 steak-1.620
 lung-1.566
Token room
Token ID2119
Feature activation+1.43e+00

Pos logit contributions
ples+0.674
ount+0.609
 Jas+0.604
yl+0.580
ydia+0.567
Neg logit contributions
 tum-0.928
 thirst-0.814
night-0.811
 steak-0.782
 lung-0.761
Token.
Token ID13
Feature activation+2.06e+00

Pos logit contributions
ples+2.608
ount+2.436
ydia+2.392
 Jas+2.389
yl+2.316
Neg logit contributions
 tum-2.283
night-1.901
 thirst-1.878
 steak-1.813
 lung-1.741
Token steak
Token ID26320
Feature activation-1.94e+00

Pos logit contributions
 thirst+4.055
 tum+3.842
night+3.453
 drowned+3.372
edly+3.370
Neg logit contributions
 paints-5.829
 written-5.581
 mirror-5.537
ples-5.457
 typing-5.442
Token.
Token ID13
Feature activation+3.43e-01

Pos logit contributions
 tum+3.904
 thirst+3.606
night+3.543
bye+3.289
 lung+3.271
Neg logit contributions
ples-2.488
 blo-2.320
 paints-2.267
ount-2.230
 written-2.227
Token
Token ID198
Feature activation-3.98e-02

Pos logit contributions
ples+0.654
ount+0.605
 Jas+0.604
yl+0.589
 paints+0.586
Neg logit contributions
 tum-0.506
 thirst-0.430
night-0.427
 steak-0.398
 lung-0.382
Token
Token ID198
Feature activation+4.46e-02

Pos logit contributions
 tum+0.062
 thirst+0.054
night+0.053
 steak+0.050
 lung+0.048
Neg logit contributions
ples-0.071
 Jas-0.067
ount-0.066
yl-0.065
 paints-0.064
TokenWhile
Token ID3633
Feature activation-1.04e+00

Pos logit contributions
ples+0.094
ount+0.088
 Jas+0.088
yl+0.085
 paints+0.085
Neg logit contributions
 tum-0.057
 thirst-0.047
night-0.047
 steak-0.043
 lung-0.041
Token his
Token ID465
Feature activation-4.12e+00

Pos logit contributions
 tum+1.643
night+1.471
 thirst+1.431
 lung+1.311
 steak+1.298
Neg logit contributions
ples-1.796
 Jas-1.757
yl-1.684
 paints-1.681
 blo-1.667
Token mom
Token ID1995
Feature activation-1.37e+00

Pos logit contributions
 thirst+6.118
 alas+6.004
Sadly+5.939
bye+5.869
night+5.790
Neg logit contributions
 paints-7.094
 mirror-6.674
 paintings-6.673
 typing-6.597
 makeup-6.419
Token was
Token ID373
Feature activation-7.86e-01

Pos logit contributions
 tum+2.708
 thirst+2.494
night+2.456
 steak+2.272
 lung+2.245
Neg logit contributions
ples-1.906
 paints-1.681
 Jas-1.678
yl-1.648
 written-1.613
Token busy
Token ID8179
Feature activation-1.22e+00

Pos logit contributions
 tum+1.651
night+1.510
 thirst+1.508
bye+1.418
 lung+1.402
Neg logit contributions
ples-0.946
ount-0.841
 Jas-0.840
 paints-0.820
 blo-0.812
Token preparing
Token ID10629
Feature activation-1.37e+00

Pos logit contributions
 tum+2.410
 thirst+2.218
night+2.169
bye+2.051
 lung+2.013
Neg logit contributions
ples-1.633
 Jas-1.431
 blo-1.430
 paints-1.423
ount-1.404
Token the
Token ID262
Feature activation-2.74e+00

Pos logit contributions
 tum+2.346
 thirst+2.114
night+2.063
bye+1.952
 drowned+1.897
Neg logit contributions
ples-2.129
 paints-2.013
 Jas-2.000
 blo-1.972
yl-1.919
Token maybe
Token ID3863
Feature activation-1.11e+00

Pos logit contributions
 tum+1.842
night+1.696
 thirst+1.678
 drowned+1.479
 steak+1.470
Neg logit contributions
ples-1.745
 Jas-1.618
ount-1.600
 blo-1.579
 paints-1.561
Token I
Token ID314
Feature activation+8.15e-01

Pos logit contributions
 tum+1.836
 thirst+1.648
night+1.621
 drowned+1.417
 lung+1.413
Neg logit contributions
ples-1.850
 paints-1.693
 blo-1.686
 Jas-1.685
yl-1.651
Token can
Token ID460
Feature activation-6.87e-01

Pos logit contributions
ples+1.193
 Jas+1.104
ount+1.072
yl+1.044
ydia+1.039
Neg logit contributions
 tum-1.567
 thirst-1.395
night-1.373
 steak-1.296
 lung-1.280
Token eat
Token ID4483
Feature activation-1.17e+00

Pos logit contributions
 tum+1.133
night+1.026
 thirst+1.014
 steak+0.951
 lung+0.912
Neg logit contributions
ples-1.139
ount-1.055
 paints-1.034
 Jas-1.028
 blo-1.023
Token it
Token ID340
Feature activation-1.78e+00

Pos logit contributions
 tum+2.101
night+1.897
 thirst+1.883
 lung+1.709
bye+1.679
Neg logit contributions
ples-1.769
 blo-1.639
ount-1.624
 paints-1.611
 Jas-1.581
Token!"
Token ID2474
Feature activation-4.07e+00

Pos logit contributions
 tum+3.285
 thirst+3.001
night+2.924
 lung+2.744
 steak+2.662
Neg logit contributions
ples-2.641
 blo-2.520
ount-2.520
 paints-2.382
 Jas-2.364
Token He
Token ID679
Feature activation-1.72e+00

Pos logit contributions
uch+7.957
 footing+7.952
 thirst+7.912
 tum+7.332
 current+7.216
Neg logit contributions
 She-5.251
 Kay-5.056
 mirror-4.876
ples-4.833
 Paula-4.732
Token looked
Token ID3114
Feature activation+2.31e-01

Pos logit contributions
 tum+3.369
 thirst+3.072
night+3.038
 lung+2.921
 steak+2.881
Neg logit contributions
ples-2.296
 Jas-2.097
yl-2.083
 paints-2.076
 violet-2.073
Token and
Token ID290
Feature activation+1.75e-01

Pos logit contributions
ples+0.363
 Jas+0.332
ount+0.332
yl+0.320
 paints+0.319
Neg logit contributions
 tum-0.414
 thirst-0.365
night-0.363
 steak-0.343
 lung-0.334
Token looked
Token ID3114
Feature activation+5.12e-01

Pos logit contributions
ples+0.223
 Jas+0.199
ount+0.199
 paints+0.190
yl+0.190
Neg logit contributions
 tum-0.369
 thirst-0.331
night-0.330
 steak-0.315
 lung-0.308
Token to
Token ID284
Feature activation+8.14e-02

Pos logit contributions
ples+0.880
 Jas+0.815
ount+0.814
yl+0.789
 paints+0.781
Neg logit contributions
 tum-0.857
 thirst-0.745
night-0.740
 steak-0.696
 lung-0.666
TokenThe
Token ID464
Feature activation-1.55e+00

Pos logit contributions
 tum+5.003
night+4.616
 thirst+4.448
 steak+4.288
 dinner+4.138
Neg logit contributions
ples-2.276
ount-2.103
 Jas-2.037
Mel-1.956
ptions-1.943
Token waves
Token ID9813
Feature activation-1.11e+00

Pos logit contributions
 tum+2.379
night+2.231
 thirst+2.142
 steak+1.899
edly+1.894
Neg logit contributions
ples-2.611
 blo-2.525
 paints-2.518
 Jas-2.489
ount-2.415
Token are
Token ID389
Feature activation-1.37e+00

Pos logit contributions
 tum+2.310
 thirst+2.121
night+2.104
 steak+2.029
 lung+1.936
Neg logit contributions
ples-1.357
ount-1.224
 Jas-1.183
yl-1.173
 paints-1.165
Token too
Token ID1165
Feature activation-2.33e+00

Pos logit contributions
 tum+2.479
night+2.321
 thirst+2.198
 steak+2.074
bye+2.044
Neg logit contributions
ples-2.055
 blo-1.894
ount-1.843
 Jas-1.833
 paints-1.799
Token strong
Token ID1913
Feature activation-9.77e-01

Pos logit contributions
 tum+4.455
 thirst+4.130
night+4.097
edly+3.859
 steak+3.616
Neg logit contributions
ples-3.299
yl-3.006
 blo-2.929
grown-2.907
 Jas-2.860
Token!"
Token ID2474
Feature activation-4.05e+00

Pos logit contributions
 tum+1.889
 thirst+1.723
night+1.607
 steak+1.525
 lung+1.513
Neg logit contributions
ples-1.430
ount-1.275
 Jas-1.252
yl-1.207
 paints-1.206
Token But
Token ID887
Feature activation-2.72e+00

Pos logit contributions
uch+7.423
 steak+7.112
night+6.938
 thirst+6.877
 lung+6.817
Neg logit contributions
 Jas-6.505
 Kay-5.675
 paints-5.519
 Jenn-5.366
 drawn-5.327
Token no
Token ID645
Feature activation-2.19e+00

Pos logit contributions
 tum+4.594
night+4.588
 steak+4.174
ages+4.159
 thirst+4.149
Neg logit contributions
 Jas-4.194
 Kay-3.954
ples-3.763
yl-3.749
 Jenn-3.691
Token one
Token ID530
Feature activation-5.98e-01

Pos logit contributions
 tum+3.947
 thirst+3.531
edly+3.386
night+3.280
ages+3.258
Neg logit contributions
ples-3.479
yl-3.244
 Jas-3.243
 paints-3.121
 Kay-3.072
Token was
Token ID373
Feature activation-4.87e-01

Pos logit contributions
 tum+1.289
 thirst+1.181
night+1.168
 steak+1.116
 lung+1.088
Neg logit contributions
ples-0.697
 Jas-0.624
yl-0.617
ount-0.598
 paints-0.592
Token there
Token ID612
Feature activation-6.74e-01

Pos logit contributions
 tum+0.898
night+0.796
 thirst+0.785
 steak+0.732
 lung+0.713
Neg logit contributions
ples-0.739
 Jas-0.667
ount-0.666
yl-0.649
 blo-0.647
Token to
Token ID284
Feature activation-8.42e-01

Pos logit contributions
 tum+2.402
 thirst+2.265
night+2.175
 lung+1.998
uch+1.966
Neg logit contributions
ples-2.489
 Jas-2.359
 Kay-2.335
ount-2.326
 paints-2.309
Token walk
Token ID2513
Feature activation-1.04e+00

Pos logit contributions
 tum+1.403
night+1.247
 thirst+1.237
 steak+1.171
 lung+1.104
Neg logit contributions
ples-1.388
 Jas-1.288
ount-1.284
 blo-1.243
 Kay-1.240
Token home
Token ID1363
Feature activation-1.33e+00

Pos logit contributions
 tum+1.881
 thirst+1.678
night+1.636
 steak+1.527
 lung+1.513
Neg logit contributions
ples-1.589
ount-1.475
 Jas-1.468
yl-1.410
 blo-1.391
Token.
Token ID13
Feature activation-2.32e+00

Pos logit contributions
 tum+2.649
 thirst+2.397
night+2.382
 steak+2.220
 lung+2.172
Neg logit contributions
ples-1.766
 Jas-1.665
ount-1.640
yl-1.585
 blo-1.570
Token So
Token ID1406
Feature activation-3.33e+00

Pos logit contributions
 tum+3.337
 thirst+3.196
night+3.186
 lung+2.945
uch+2.928
Neg logit contributions
ples-4.015
 Kay-3.933
ount-3.929
 Jas-3.908
 paints-3.647
Token,
Token ID11
Feature activation-4.03e+00

Pos logit contributions
 tum+6.636
 thirst+6.552
night+6.350
 lung+6.262
uch+6.110
Neg logit contributions
 Kay-4.315
yl-4.113
 Jas-3.705
 blo-3.626
 Jem-3.542
Token he
Token ID339
Feature activation-1.38e+00

Pos logit contributions
night+7.173
 tum+7.130
 thirst+7.025
 lung+6.658
uch+6.533
Neg logit contributions
 Kay-6.659
 blo-5.912
yl-5.742
 Jas-5.670
 Sylvia-5.566
Token asked
Token ID1965
Feature activation-2.56e+00

Pos logit contributions
 tum+2.851
 thirst+2.562
night+2.493
 steak+2.438
 lung+2.327
Neg logit contributions
ples-1.738
yl-1.594
 Jas-1.591
ount-1.529
 paints-1.525
Token his
Token ID465
Feature activation-7.29e-01

Pos logit contributions
 tum+3.943
 thirst+3.568
uch+3.459
night+3.366
 lung+3.111
Neg logit contributions
 Jas-4.462
ples-4.389
 Kay-4.369
yl-4.274
 blo-4.173
Token friend
Token ID1545
Feature activation-3.88e-01

Pos logit contributions
 tum+0.922
night+0.817
 thirst+0.805
 steak+0.738
 lung+0.699
Neg logit contributions
ples-1.460
 Jas-1.388
yl-1.362
ount-1.355
 paints-1.352
Token for
Token ID329
Feature activation-2.48e+00

Pos logit contributions
 tum+0.618
 thirst+0.553
night+0.542
 steak+0.504
 lung+0.497
Neg logit contributions
ples-0.662
 Jas-0.625
ount-0.621
yl-0.601
 paints-0.598
Token the
Token ID262
Feature activation+3.39e-01

Pos logit contributions
 tum+0.512
night+0.449
 thirst+0.447
 steak+0.401
 lung+0.396
Neg logit contributions
ples-0.565
 Jas-0.524
ount-0.520
 paints-0.516
yl-0.507
Token tunnel
Token ID13275
Feature activation-4.12e-01

Pos logit contributions
ples+0.524
ount+0.476
 Jas+0.475
yl+0.457
 paints+0.456
Neg logit contributions
 tum-0.623
 thirst-0.550
night-0.546
 steak-0.517
 lung-0.504
Token and
Token ID290
Feature activation-7.11e-01

Pos logit contributions
 tum+0.664
 thirst+0.584
night+0.575
 steak+0.537
 lung+0.529
Neg logit contributions
ples-0.712
ount-0.663
 Jas-0.659
 paints-0.641
yl-0.640
Token be
Token ID307
Feature activation+1.51e+00

Pos logit contributions
 tum+1.290
night+1.154
 thirst+1.145
 steak+1.085
 lung+1.041
Neg logit contributions
ples-1.067
 Jas-1.002
 paints-0.983
ount-0.975
 blo-0.960
Token friends
Token ID2460
Feature activation-1.06e+00

Pos logit contributions
ples+3.179
ount+3.048
 Jas+2.913
yl+2.868
 paints+2.859
Neg logit contributions
 tum-1.960
 thirst-1.728
night-1.581
 steak-1.513
 dinner-1.463
Token!"
Token ID2474
Feature activation-3.97e+00

Pos logit contributions
 tum+1.914
 thirst+1.716
night+1.663
 steak+1.538
 lung+1.527
Neg logit contributions
ples-1.632
ount-1.455
 Jas-1.453
yl-1.430
 blo-1.419
Token Mom
Token ID11254
Feature activation+5.61e-02

Pos logit contributions
 footing+7.750
night+7.365
uch+7.334
 thirst+7.271
 drowned+7.133
Neg logit contributions
 Jas-5.354
 violet-5.296
 paints-5.182
 Kay-5.011
ples-5.002
Tokeno
Token ID78
Feature activation-9.25e-01

Pos logit contributions
ples+0.085
 Jas+0.077
ount+0.077
yl+0.075
 paints+0.074
Neg logit contributions
 tum-0.104
 thirst-0.092
night-0.091
 steak-0.086
 lung-0.084
Token thought
Token ID1807
Feature activation-4.55e-01

Pos logit contributions
 tum+1.741
 thirst+1.569
night+1.526
 steak+1.476
 lung+1.458
Neg logit contributions
ples-1.350
 Jas-1.242
yl-1.224
ount-1.210
 paints-1.198
Token about
Token ID546
Feature activation+6.65e-02

Pos logit contributions
 tum+0.798
 thirst+0.701
night+0.691
 steak+0.648
 lung+0.645
Neg logit contributions
ples-0.724
 Jas-0.677
ount-0.662
 paints-0.649
yl-0.646
Token it
Token ID340
Feature activation-9.76e-01

Pos logit contributions
ples+0.107
 Jas+0.099
ount+0.098
yl+0.095
 paints+0.095
Neg logit contributions
 tum-0.116
 thirst-0.102
night-0.102
 steak-0.095
 lung-0.093
Token of
Token ID286
Feature activation-2.27e-01

Pos logit contributions
 tum+0.677
 thirst+0.613
night+0.606
 steak+0.570
 lung+0.555
Neg logit contributions
ples-0.531
 Jas-0.485
ount-0.482
yl-0.467
 paints-0.465
Token the
Token ID262
Feature activation-5.05e-01

Pos logit contributions
 tum+0.411
night+0.367
 thirst+0.361
 steak+0.335
 lung+0.330
Neg logit contributions
ples-0.350
 Jas-0.325
ount-0.317
 paints-0.312
yl-0.310
Token hill
Token ID12788
Feature activation-5.57e-01

Pos logit contributions
 tum+0.703
 thirst+0.622
night+0.619
 steak+0.545
 lung+0.532
Neg logit contributions
ples-0.960
 Jas-0.908
ount-0.906
 paints-0.906
 blo-0.896
Token.
Token ID13
Feature activation-2.69e+00

Pos logit contributions
 tum+1.082
 thirst+0.971
night+0.961
 steak+0.912
 lung+0.892
Neg logit contributions
ples-0.786
 Jas-0.721
ount-0.716
yl-0.688
 paints-0.676
Token So
Token ID1406
Feature activation-2.95e+00

Pos logit contributions
 thirst+4.115
 tum+3.985
night+3.861
 lung+3.712
 footing+3.662
Neg logit contributions
ount-4.199
ples-4.136
 Jas-4.121
 Kay-4.103
 paints-3.972
Token,
Token ID11
Feature activation-3.91e+00

Pos logit contributions
 thirst+6.048
 tum+6.007
night+5.856
 lung+5.612
edly+5.568
Neg logit contributions
yl-3.288
 Kay-3.276
 Jas-2.921
 blo-2.899
ples-2.810
Token he
Token ID339
Feature activation-1.44e+00

Pos logit contributions
 thirst+7.005
night+6.878
 tum+6.825
edly+6.357
 forehead+6.170
Neg logit contributions
 Kay-5.520
 blo-5.077
 written-5.074
 Carrie-5.043
 drawn-5.029
Token kept
Token ID4030
Feature activation+1.00e+00

Pos logit contributions
 tum+2.892
 thirst+2.570
night+2.545
 steak+2.452
 lung+2.381
Neg logit contributions
ples-1.973
 Jas-1.861
yl-1.835
 paints-1.786
ount-1.732
Token running
Token ID2491
Feature activation+6.52e-02

Pos logit contributions
ples+1.430
ount+1.329
 Jas+1.248
yl+1.199
ydia+1.198
Neg logit contributions
 tum-1.967
 thirst-1.760
night-1.732
 steak-1.704
 lung-1.654
Token.
Token ID13
Feature activation-2.29e+00

Pos logit contributions
ples+0.121
 Jas+0.113
ount+0.113
yl+0.109
 paints+0.109
Neg logit contributions
 tum-0.098
 thirst-0.084
night-0.083
 steak-0.078
 lung-0.075
Token
Token ID198
Feature activation-1.05e+00

Pos logit contributions
 tum+4.041
 thirst+3.951
night+3.844
 lung+3.596
 steak+3.503
Neg logit contributions
ples-3.241
ount-3.224
 Jas-3.089
 Kay-3.023
 paints-2.944
Token.
Token ID13
Feature activation-1.58e+00

Pos logit contributions
ples+0.564
ount+0.517
 Jas+0.516
yl+0.498
 paints+0.497
Neg logit contributions
 tum-0.633
 thirst-0.558
night-0.554
 steak-0.525
 lung-0.509
Token
Token ID220
Feature activation-7.37e-01

Pos logit contributions
 tum+2.532
 thirst+2.361
night+2.274
 lung+2.172
 steak+2.100
Neg logit contributions
ples-2.601
ount-2.546
 Jas-2.513
 Kay-2.490
 violet-2.349
Token
Token ID198
Feature activation-9.10e-01

Pos logit contributions
 tum+1.240
 thirst+1.117
night+1.110
 steak+1.026
 lung+1.011
Neg logit contributions
ples-1.204
ount-1.162
 Jas-1.154
 Kay-1.087
 paints-1.084
Token
Token ID198
Feature activation-1.66e+00

Pos logit contributions
 tum+1.584
 thirst+1.435
night+1.422
 lung+1.305
 steak+1.298
Neg logit contributions
ples-1.436
 Jas-1.384
ount-1.380
yl-1.332
 Kay-1.322
TokenBut
Token ID1537
Feature activation-3.01e+00

Pos logit contributions
 tum+2.671
 thirst+2.346
night+2.325
 lung+2.113
 steak+1.986
Neg logit contributions
ples-2.975
ount-2.898
Kay-2.811
 Jas-2.791
yl-2.764
Token,
Token ID11
Feature activation-3.91e+00

Pos logit contributions
 tum+5.141
night+4.630
 thirst+4.499
 lung+4.183
ages+4.094
Neg logit contributions
 Kay-4.801
yl-4.756
 blo-4.598
ples-4.556
 Jas-4.448
Token because
Token ID780
Feature activation-2.66e+00

Pos logit contributions
 tum+8.462
night+8.300
 thirst+7.851
uch+7.448
 lung+7.376
Neg logit contributions
 Kay-4.616
 written-4.220
 blo-4.219
yl-4.188
grown-4.173
Token Bob
Token ID5811
Feature activation-1.27e+00

Pos logit contributions
 tum+5.334
night+5.092
 thirst+4.950
edly+4.436
bye+4.389
Neg logit contributions
 Jas-3.346
 Kay-3.342
 blo-3.268
ples-3.209
 paints-3.108
Token was
Token ID373
Feature activation-3.25e-01

Pos logit contributions
 tum+2.349
night+2.102
 thirst+2.079
 steak+1.952
bye+1.890
Neg logit contributions
ples-1.909
 paints-1.712
yl-1.700
 Jas-1.691
ount-1.658
Token so
Token ID523
Feature activation-1.30e+00

Pos logit contributions
 tum+0.555
night+0.493
 thirst+0.485
 steak+0.456
 lung+0.443
Neg logit contributions
ples-0.531
 Jas-0.489
ount-0.487
yl-0.471
 blo-0.470
Token lazy
Token ID16931
Feature activation-5.14e-02

Pos logit contributions
 tum+2.383
night+2.214
 thirst+2.120
 steak+2.053
 lung+2.010
Neg logit contributions
ples-1.830
ount-1.711
 blo-1.684
 Jas-1.678
yl-1.668
Token felt
Token ID2936
Feature activation-1.47e+00

Pos logit contributions
 tum+3.643
night+3.357
 thirst+3.332
 steak+3.193
bye+3.112
Neg logit contributions
ples-2.366
 paints-2.137
yl-2.078
 blo-2.069
 Jas-2.044
Token a
Token ID257
Feature activation-2.55e+00

Pos logit contributions
 tum+2.623
night+2.499
 thirst+2.310
bye+2.281
 steak+2.259
Neg logit contributions
ples-2.048
 blo-1.945
ount-1.916
 paints-1.831
 Jas-1.825
Token burning
Token ID9482
Feature activation-1.48e+00

Pos logit contributions
 tum+3.979
night+3.878
bye+3.605
 thirst+3.561
 steak+3.483
Neg logit contributions
ples-4.060
 blo-3.862
yl-3.727
 violet-3.663
 mirror-3.585
Token sensation
Token ID18098
Feature activation-1.03e+00

Pos logit contributions
 tum+2.292
night+2.069
 thirst+2.046
bye+1.888
 steak+1.843
Neg logit contributions
ples-2.540
 blo-2.329
 paints-2.288
 Jas-2.275
yl-2.224
Token on
Token ID319
Feature activation-2.26e+00

Pos logit contributions
 tum+1.890
 thirst+1.709
night+1.682
 steak+1.591
 lung+1.549
Neg logit contributions
ples-1.579
 Jas-1.402
ount-1.386
 blo-1.367
 paints-1.358
Token his
Token ID465
Feature activation-3.89e+00

Pos logit contributions
 tum+3.990
night+3.778
 thirst+3.623
bye+3.497
uch+3.471
Neg logit contributions
ples-3.418
 Jas-3.127
 Kay-3.064
els-3.057
 paints-3.017
Token paw
Token ID41827
Feature activation-1.73e+00

Pos logit contributions
night+6.402
bye+5.978
 alas+5.583
 tum+5.579
time+5.572
Neg logit contributions
ples-6.794
 paints-6.612
 blo-6.132
els-6.070
 mirror-5.951
Token and
Token ID290
Feature activation-1.80e+00

Pos logit contributions
 tum+3.190
night+3.041
 thirst+3.019
bye+2.755
 steak+2.727
Neg logit contributions
ples-2.502
 blo-2.177
 paints-2.176
 written-2.138
ount-2.103
Token quickly
Token ID2952
Feature activation-1.06e+00

Pos logit contributions
 tum+3.083
night+2.870
 thirst+2.759
bye+2.617
uch+2.592
Neg logit contributions
ples-2.778
 Jas-2.639
 blo-2.594
 paints-2.582
 written-2.481
Token pulled
Token ID5954
Feature activation-1.63e+00

Pos logit contributions
 tum+1.911
night+1.747
 thirst+1.707
 steak+1.665
 lung+1.616
Neg logit contributions
ples-1.587
 Jas-1.396
 paints-1.381
 blo-1.379
ount-1.359
Token it
Token ID340
Feature activation-1.65e+00

Pos logit contributions
 tum+3.284
 thirst+2.992
night+2.972
bye+2.777
 steak+2.700
Neg logit contributions
ples-2.185
 Jas-2.012
 paints-1.915
els-1.876
yl-1.872
Token fell
Token ID3214
Feature activation-1.34e+00

Pos logit contributions
 tum+3.505
night+3.221
 thirst+3.213
 steak+3.078
 dinner+2.905
Neg logit contributions
ples-2.246
 paints-1.891
 drawn-1.864
yl-1.853
 Jas-1.848
Token down
Token ID866
Feature activation-2.30e-01

Pos logit contributions
 tum+2.505
 thirst+2.309
night+2.207
 steak+2.159
 dinner+1.999
Neg logit contributions
ples-2.028
 Jas-1.841
 blo-1.815
ount-1.811
 paints-1.780
Token.
Token ID13
Feature activation-1.12e+00

Pos logit contributions
 tum+0.419
 thirst+0.372
night+0.367
 steak+0.346
 lung+0.336
Neg logit contributions
ples-0.355
 Jas-0.329
ount-0.326
yl-0.317
 paints-0.312
Token He
Token ID679
Feature activation-2.32e+00

Pos logit contributions
 tum+1.707
 thirst+1.603
night+1.540
 lung+1.425
 steak+1.411
Neg logit contributions
ples-1.920
 Jas-1.890
 Kay-1.802
ount-1.788
 paints-1.788
Token hurt
Token ID5938
Feature activation-2.25e+00

Pos logit contributions
 tum+4.593
night+4.353
 thirst+4.254
 steak+4.033
bye+3.994
Neg logit contributions
ples-3.037
 paints-2.870
yl-2.740
 drawn-2.721
 written-2.689
Token his
Token ID465
Feature activation-3.89e+00

Pos logit contributions
 tum+3.290
night+3.014
 thirst+2.888
uch+2.706
bye+2.616
Neg logit contributions
 Jas-4.262
ples-4.156
 Kay-4.031
 mirror-3.898
yl-3.898
Token leg
Token ID1232
Feature activation-1.13e+00

Pos logit contributions
night+5.862
bye+5.294
 tum+5.174
time+4.983
 thirst+4.786
Neg logit contributions
 paints-6.806
yl-6.763
 mirror-6.675
ples-6.588
els-6.523
Token and
Token ID290
Feature activation-1.92e+00

Pos logit contributions
 tum+2.006
 thirst+1.856
night+1.834
 steak+1.724
bye+1.619
Neg logit contributions
ples-1.754
 blo-1.545
 paints-1.544
yl-1.521
 drawn-1.515
Token started
Token ID2067
Feature activation-4.68e-01

Pos logit contributions
 tum+3.165
night+2.987
 thirst+2.946
 steak+2.724
bye+2.696
Neg logit contributions
ples-2.956
 Jas-2.846
 paints-2.823
 written-2.768
 drawn-2.749
Token to
Token ID284
Feature activation-9.55e-01

Pos logit contributions
 tum+0.715
 thirst+0.635
night+0.631
 steak+0.589
 lung+0.555
Neg logit contributions
ples-0.839
 Jas-0.778
ount-0.764
 paints-0.762
yl-0.752
Token suffer
Token ID8659
Feature activation-1.21e+00

Pos logit contributions
 tum+1.605
night+1.442
 thirst+1.428
 steak+1.352
bye+1.288
Neg logit contributions
ples-1.556
 paints-1.436
 Jas-1.431
 blo-1.425
ount-1.416
Tokenmy
Token ID1820
Feature activation-1.20e+00

Pos logit contributions
 tum+0.758
 thirst+0.705
night+0.694
 steak+0.671
 lung+0.653
Neg logit contributions
ples-0.230
 Jas-0.190
ount-0.182
 paints-0.177
yl-0.176
Token fell
Token ID3214
Feature activation-1.29e+00

Pos logit contributions
 tum+2.374
 thirst+2.265
night+2.234
 steak+2.095
bye+2.042
Neg logit contributions
ples-1.555
 Jas-1.417
yl-1.385
 paints-1.381
 drawn-1.335
Token down
Token ID866
Feature activation-5.23e-01

Pos logit contributions
 tum+2.419
 thirst+2.259
night+2.110
 steak+2.066
 lung+1.948
Neg logit contributions
ples-1.921
 Jas-1.758
 blo-1.737
ount-1.708
 paints-1.707
Token and
Token ID290
Feature activation-1.51e+00

Pos logit contributions
 tum+0.901
 thirst+0.800
night+0.773
 steak+0.728
 lung+0.710
Neg logit contributions
ples-0.848
 Jas-0.797
ount-0.785
yl-0.775
 blo-0.767
Token hurt
Token ID5938
Feature activation-1.83e+00

Pos logit contributions
 tum+2.406
 thirst+2.226
night+2.189
 steak+2.099
bye+1.994
Neg logit contributions
ples-2.435
 Jas-2.319
 paints-2.299
 blo-2.276
 drawn-2.258
Token his
Token ID465
Feature activation-3.89e+00

Pos logit contributions
 tum+2.470
night+2.070
 thirst+2.051
uch+1.878
bye+1.814
Neg logit contributions
 Jas-3.697
ples-3.695
 Kay-3.463
grown-3.388
yl-3.377
Token knee
Token ID10329
Feature activation-1.17e+00

Pos logit contributions
night+5.085
bye+4.789
His+4.601
Sadly+4.553
By+4.505
Neg logit contributions
 mirror-7.226
 paints-7.039
 makeup-7.034
 jewelry-6.849
 blo-6.694
Token.
Token ID13
Feature activation-1.50e+00

Pos logit contributions
 tum+2.305
 thirst+2.158
night+2.114
 steak+2.053
 lung+1.919
Neg logit contributions
ples-1.552
 blo-1.339
ount-1.335
 Jas-1.332
 paints-1.324
Token Max
Token ID5436
Feature activation-2.00e+00

Pos logit contributions
 tum+1.989
 thirst+1.963
night+1.812
 steak+1.660
bye+1.648
Neg logit contributions
ples-2.778
 Jas-2.732
 paints-2.612
 Kay-2.609
ydia-2.573
Token bark
Token ID21405
Feature activation-7.59e-01

Pos logit contributions
 tum+3.710
night+3.516
 thirst+3.500
 steak+3.382
bye+3.324
Neg logit contributions
ples-2.691
 paints-2.409
 blo-2.397
yl-2.393
 drawn-2.374
Tokened
Token ID276
Feature activation-1.36e+00

Pos logit contributions
 tum+1.357
 thirst+1.212
night+1.175
 steak+1.164
 lung+1.128
Neg logit contributions
ples-1.188
ount-1.051
 Jas-1.051
yl-1.021
 paints-1.008
Token the
Token ID262
Feature activation-1.32e-01

Pos logit contributions
 tum+0.570
night+0.503
 thirst+0.501
 steak+0.465
 lung+0.454
Neg logit contributions
ples-0.498
 Jas-0.458
ount-0.455
yl-0.443
 paints-0.442
Token big
Token ID1263
Feature activation+6.71e-01

Pos logit contributions
 tum+0.223
 thirst+0.198
night+0.198
 steak+0.183
 lung+0.177
Neg logit contributions
ples-0.220
 Jas-0.200
ount-0.199
 paints-0.197
yl-0.193
Token,
Token ID11
Feature activation+5.22e-01

Pos logit contributions
ples+1.212
ount+1.151
 Jas+1.134
yl+1.093
ydia+1.084
Neg logit contributions
 tum-1.063
 thirst-0.896
night-0.890
 steak-0.858
 lung-0.835
Token mean
Token ID1612
Feature activation+1.13e+00

Pos logit contributions
ples+0.853
ount+0.783
 Jas+0.778
yl+0.751
 paints+0.746
Neg logit contributions
 tum-0.914
 thirst-0.796
night-0.785
 steak-0.752
 lung-0.732
Token cat
Token ID3797
Feature activation-5.94e-01

Pos logit contributions
ples+2.319
ount+2.188
 Jas+2.131
 blo+2.075
 paints+2.065
Neg logit contributions
 tum-1.601
night-1.300
 thirst-1.270
 steak-1.238
 lung-1.214
Token!"
Token ID2474
Feature activation-3.87e+00

Pos logit contributions
 tum+1.107
 thirst+0.995
night+0.980
 steak+0.927
 lung+0.895
Neg logit contributions
ples-0.879
ount-0.796
 Jas-0.787
yl-0.770
 paints-0.759
Token
Token ID198
Feature activation-2.60e+00

Pos logit contributions
night+8.126
 thirst+7.721
 tum+7.562
uch+7.456
bye+7.295
Neg logit contributions
 Kay-4.651
 Jas-4.360
 Jem-4.127
 Carrie-4.057
 She-4.034
Token
Token ID198
Feature activation-3.36e+00

Pos logit contributions
 tum+5.579
 thirst+5.471
night+5.328
 steak+4.875
 lung+4.798
Neg logit contributions
 Jas-2.974
Kay-2.930
yl-2.911
ount-2.892
ples-2.860
TokenSpot
Token ID32565
Feature activation-1.35e+00

Pos logit contributions
 tum+6.121
night+5.743
 thirst+5.522
 steak+4.776
 stomach+4.750
Neg logit contributions
Kay-5.494
ples-5.407
Mel-5.222
ount-5.118
yl-5.060
Token's
Token ID338
Feature activation-1.29e+00

Pos logit contributions
 tum+2.621
night+2.390
 thirst+2.346
 steak+2.295
 lung+2.182
Neg logit contributions
ples-1.888
 Jas-1.712
yl-1.686
ount-1.645
 Kay-1.607
Token friends
Token ID2460
Feature activation-3.84e-01

Pos logit contributions
 tum+1.717
night+1.629
 thirst+1.549
 steak+1.480
bye+1.384
Neg logit contributions
ples-2.356
 Jas-2.317
yl-2.268
 paints-2.231
ydia-2.214
Token things
Token ID1243
Feature activation+3.12e-01

Pos logit contributions
ples+0.375
ount+0.344
 Jas+0.344
yl+0.332
 paints+0.331
Neg logit contributions
 tum-0.397
 thirst-0.348
night-0.346
 steak-0.327
 lung-0.318
Token.
Token ID13
Feature activation-1.68e+00

Pos logit contributions
ples+0.538
 Jas+0.496
ount+0.496
yl+0.479
 paints+0.478
Neg logit contributions
 tum-0.510
 thirst-0.443
night-0.442
 steak-0.416
 lung-0.403
Token
Token ID198
Feature activation-3.55e-01

Pos logit contributions
 tum+3.068
 thirst+2.805
night+2.722
uch+2.574
 leg+2.530
Neg logit contributions
ples-2.490
 Jas-2.279
ount-2.241
 Kay-2.200
 paints-2.171
Token
Token ID198
Feature activation-1.37e+00

Pos logit contributions
 tum+0.509
 thirst+0.438
night+0.437
 steak+0.396
 lung+0.390
Neg logit contributions
ples-0.672
 Jas-0.636
ount-0.632
yl-0.616
 paints-0.612
TokenEventually
Token ID28724
Feature activation-2.62e+00

Pos logit contributions
 tum+2.313
 thirst+1.989
night+1.930
 lung+1.731
 leg+1.712
Neg logit contributions
ples-2.372
ount-2.191
 Jas-2.184
 blo-2.167
yl-2.163
Token,
Token ID11
Feature activation-3.86e+00

Pos logit contributions
 tum+4.976
 thirst+4.421
uch+4.298
bye+4.213
 hind+4.210
Neg logit contributions
ples-3.637
 blo-3.415
 Jas-3.396
 Kay-3.374
yl-3.325
Token Henry
Token ID8616
Feature activation-7.87e-02

Pos logit contributions
 tum+7.939
uch+7.148
 thirst+7.115
erved+7.033
 knee+7.021
Neg logit contributions
 blo-5.051
 Kay-4.920
 written-4.862
ples-4.643
 Jem-4.592
Token became
Token ID2627
Feature activation+1.07e-01

Pos logit contributions
 tum+0.139
 thirst+0.123
night+0.121
 steak+0.115
 lung+0.111
Neg logit contributions
ples-0.125
 Jas-0.114
ount-0.114
yl-0.111
 paints-0.111
Token the
Token ID262
Feature activation-9.22e-02

Pos logit contributions
ples+0.186
 Jas+0.172
ount+0.172
 paints+0.166
yl+0.166
Neg logit contributions
 tum-0.175
 thirst-0.152
night-0.151
 steak-0.142
 lung-0.137
Token most
Token ID749
Feature activation+3.23e-01

Pos logit contributions
 tum+0.139
 thirst+0.120
night+0.119
 steak+0.111
 lung+0.107
Neg logit contributions
ples-0.170
 Jas-0.160
ount-0.158
 paints-0.155
yl-0.154
Token knowledgeable
Token ID29549
Feature activation+8.29e-01

Pos logit contributions
ples+0.508
ount+0.465
 Jas+0.464
yl+0.447
 paints+0.445
Neg logit contributions
 tum-0.586
 thirst-0.517
night-0.514
 steak-0.487
 lung-0.473
Token Mama
Token ID40084
Feature activation+1.21e+00

Pos logit contributions
 tum+0.694
 thirst+0.611
night+0.600
 steak+0.514
 lung+0.506
Neg logit contributions
ples-1.045
 Jas-1.007
ount-0.983
yl-0.975
 Kay-0.974
Token!
Token ID0
Feature activation-6.23e-01

Pos logit contributions
ples+2.144
ount+2.008
 Jas+1.973
 paints+1.943
ydia+1.933
Neg logit contributions
 tum-1.943
night-1.666
 thirst-1.627
 steak-1.601
 lung-1.537
Token A
Token ID317
Feature activation-4.66e-01

Pos logit contributions
 tum+1.196
 thirst+1.098
night+1.081
 steak+0.996
 lung+0.976
Neg logit contributions
ples-0.880
 Jas-0.807
ount-0.807
 paints-0.771
 Kay-0.767
Token red
Token ID2266
Feature activation-1.04e+00

Pos logit contributions
 tum+0.685
 thirst+0.610
night+0.598
 steak+0.526
 lung+0.517
Neg logit contributions
ples-0.858
 Jas-0.789
 paints-0.785
yl-0.784
ount-0.776
Token balloon
Token ID21190
Feature activation+7.31e-01

Pos logit contributions
 tum+1.423
 thirst+1.289
night+1.231
 drowned+1.084
 steak+1.055
Neg logit contributions
ples-1.968
 paints-1.854
yl-1.844
 violet-1.803
 blo-1.803
Token!"
Token ID2474
Feature activation-3.86e+00

Pos logit contributions
ples+1.208
 Jas+1.123
ount+1.115
ydia+1.072
yl+1.065
Neg logit contributions
 tum-1.252
night-1.093
 thirst-1.079
 steak-1.034
 lung-1.001
Token
Token ID198
Feature activation-2.91e+00

Pos logit contributions
uch+8.021
 thirst+7.595
 footing+7.440
 tum+7.368
 drowned+7.230
Neg logit contributions
 Kay-4.645
 She-4.271
ples-4.247
 mirror-4.231
 blo-4.211
Token
Token ID198
Feature activation-3.28e+00

Pos logit contributions
 tum+6.205
 thirst+6.157
night+5.802
uch+5.373
 current+5.347
Neg logit contributions
Kay-3.328
 Kay-3.292
 Jas-3.274
yl-3.179
ples-3.162
TokenM
Token ID44
Feature activation-1.82e+00

Pos logit contributions
 tum+6.097
 thirst+5.657
night+5.353
uch+5.068
 elbow+4.927
Neg logit contributions
ples-4.993
Kay-4.808
 blo-4.722
 Kay-4.613
ivia-4.612
Tokenama
Token ID1689
Feature activation+8.71e-01

Pos logit contributions
 tum+2.448
 thirst+2.146
night+2.066
 drowned+1.907
 recharge+1.877
Neg logit contributions
ples-3.529
yl-3.448
 Jas-3.366
ount-3.350
els-3.334
Token smiled
Token ID13541
Feature activation-2.42e-01

Pos logit contributions
ples+1.437
ount+1.347
 Jas+1.309
 paints+1.267
ydia+1.265
Neg logit contributions
 tum-1.490
night-1.295
 thirst-1.284
 steak-1.227
 lung-1.182
Token he
Token ID339
Feature activation+1.14e-01

Pos logit contributions
 tum+0.383
 thirst+0.335
night+0.333
 steak+0.300
 lung+0.296
Neg logit contributions
ples-0.449
 Jas-0.427
ount-0.420
 paints-0.413
yl-0.407
Token had
Token ID550
Feature activation-1.11e+00

Pos logit contributions
ples+0.181
ount+0.165
 Jas+0.165
 paints+0.159
yl+0.159
Neg logit contributions
 tum-0.203
 thirst-0.179
night-0.178
 steak-0.168
 lung-0.163
Token a
Token ID257
Feature activation-6.98e-01

Pos logit contributions
 tum+1.843
night+1.705
 thirst+1.651
 steak+1.557
bye+1.533
Neg logit contributions
ples-1.772
 Jas-1.668
 paints-1.662
 blo-1.610
ount-1.604
Token cut
Token ID2005
Feature activation-1.02e+00

Pos logit contributions
 tum+1.131
night+1.042
 thirst+1.028
bye+0.930
 steak+0.928
Neg logit contributions
ples-1.128
 paints-1.057
 Jas-1.047
yl-1.040
 blo-1.026
Token on
Token ID319
Feature activation-5.27e-01

Pos logit contributions
 tum+1.865
 thirst+1.692
night+1.657
 steak+1.539
 lung+1.486
Neg logit contributions
ples-1.531
 Jas-1.386
ount-1.379
 paints-1.370
yl-1.340
Token his
Token ID465
Feature activation-3.84e+00

Pos logit contributions
 tum+0.728
night+0.629
 thirst+0.624
 steak+0.548
bye+0.539
Neg logit contributions
ples-1.033
 Jas-0.968
 paints-0.950
ount-0.944
yl-0.933
Token cheek
Token ID19353
Feature activation-9.77e-01

Pos logit contributions
night+6.742
time+6.238
bye+6.232
 thirst+6.130
 alas+5.978
Neg logit contributions
 paints-6.302
 mirror-6.189
 blo-5.696
ples-5.540
 wardrobe-5.517
Token.
Token ID13
Feature activation+4.81e-01

Pos logit contributions
 tum+1.921
 thirst+1.776
night+1.707
 steak+1.643
 current+1.561
Neg logit contributions
ples-1.363
 Jas-1.206
ount-1.205
yl-1.165
 paints-1.163
Token It
Token ID632
Feature activation-4.31e-01

Pos logit contributions
ples+0.707
ount+0.634
 Jas+0.632
yl+0.613
 paints+0.608
Neg logit contributions
 tum-0.923
 thirst-0.809
night-0.807
 steak-0.774
 lung-0.749
Token was
Token ID373
Feature activation-2.18e-01

Pos logit contributions
 tum+0.797
 thirst+0.717
night+0.715
 steak+0.673
 lung+0.647
Neg logit contributions
ples-0.645
 Jas-0.583
ount-0.579
 paints-0.576
yl-0.567
Token red
Token ID2266
Feature activation-8.06e-01

Pos logit contributions
 tum+0.334
night+0.293
 thirst+0.289
 steak+0.266
 lung+0.260
Neg logit contributions
ples-0.394
 Jas-0.365
ount-0.362
 paints-0.354
 blo-0.351
Token He
Token ID679
Feature activation-1.20e+00

Pos logit contributions
 tum+2.747
 thirst+2.576
night+2.529
uch+2.404
 drowned+2.344
Neg logit contributions
ples-2.473
 Jas-2.314
ount-2.307
 paints-2.244
 Kay-2.241
Token felt
Token ID2936
Feature activation-7.36e-01

Pos logit contributions
 tum+2.628
night+2.400
 thirst+2.399
 steak+2.250
time+2.197
Neg logit contributions
ples-1.394
 Jas-1.186
yl-1.180
 paints-1.162
ount-1.151
Token a
Token ID257
Feature activation-2.13e+00

Pos logit contributions
 tum+1.266
night+1.141
 thirst+1.104
 steak+1.047
bye+1.021
Neg logit contributions
ples-1.149
ount-1.057
 Jas-1.045
 blo-1.035
 paints-1.018
Token tug
Token ID27762
Feature activation-6.46e-01

Pos logit contributions
 tum+3.356
night+3.151
 thirst+3.019
bye+2.891
 steak+2.832
Neg logit contributions
ples-3.477
 blo-3.242
yl-3.180
 mirror-3.030
 violet-3.030
Token on
Token ID319
Feature activation-1.61e+00

Pos logit contributions
 tum+1.238
 thirst+1.095
night+1.075
 steak+1.011
 lung+0.982
Neg logit contributions
ples-0.962
 Jas-0.859
yl-0.850
ount-0.847
 paints-0.823
Token his
Token ID465
Feature activation-3.84e+00

Pos logit contributions
 tum+2.362
 thirst+2.112
night+2.079
uch+1.984
bye+1.924
Neg logit contributions
ples-2.900
 paints-2.681
 Jas-2.679
 Kay-2.670
 blo-2.614
Token fishing
Token ID12478
Feature activation-2.85e+00

Pos logit contributions
night+6.021
 thirst+5.766
 alas+5.615
 tum+5.609
time+5.594
Neg logit contributions
ples-6.150
 paints-6.146
 mirror-5.884
 blo-5.871
 prints-5.809
Token line
Token ID1627
Feature activation-4.61e-01

Pos logit contributions
night+4.196
 tum+4.054
 thirst+3.946
 drowned+3.530
uch+3.400
Neg logit contributions
ples-4.898
yl-4.733
 blo-4.580
 paints-4.568
 Jas-4.481
Token and
Token ID290
Feature activation-9.00e-01

Pos logit contributions
 tum+0.779
 thirst+0.695
night+0.682
 steak+0.640
 lung+0.625
Neg logit contributions
ples-0.768
 Jas-0.710
ount-0.708
yl-0.682
 paints-0.681
Token pulled
Token ID5954
Feature activation-6.75e-01

Pos logit contributions
 tum+1.775
night+1.604
 thirst+1.590
 steak+1.499
 lung+1.463
Neg logit contributions
ples-1.220
 Jas-1.118
 paints-1.070
 blo-1.067
ount-1.062
Token up
Token ID510
Feature activation-5.50e-01

Pos logit contributions
 tum+1.225
 thirst+1.086
night+1.074
 steak+0.991
 lung+0.975
Neg logit contributions
ples-1.057
 Jas-0.980
ount-0.942
yl-0.936
 paints-0.928
Token to
Token ID284
Feature activation-6.70e-01

Pos logit contributions
 tum+2.536
 thirst+2.345
night+2.180
 lung+2.130
 steak+2.130
Neg logit contributions
ples-2.504
 Jas-2.438
 Kay-2.334
 written-2.314
yl-2.305
Token jump
Token ID4391
Feature activation-4.54e-01

Pos logit contributions
 tum+1.107
night+0.973
 thirst+0.970
 steak+0.918
 lung+0.868
Neg logit contributions
ples-1.113
 Jas-1.070
ount-1.036
 Kay-1.016
 paints-1.014
Token.
Token ID13
Feature activation-2.58e+00

Pos logit contributions
 tum+0.822
 thirst+0.729
night+0.715
 steak+0.682
 lung+0.665
Neg logit contributions
ples-0.698
 Jas-0.653
ount-0.649
yl-0.622
 paints-0.617
Token But
Token ID887
Feature activation-3.12e+00

Pos logit contributions
 tum+4.103
 thirst+3.879
night+3.764
uch+3.718
 lung+3.705
Neg logit contributions
 Kay-4.132
ount-4.120
ples-4.059
 Jas-3.974
 violet-3.754
Token then
Token ID788
Feature activation-2.20e+00

Pos logit contributions
 tum+5.547
night+5.173
 thirst+4.760
ages+4.616
 lung+4.545
Neg logit contributions
 Kay-4.779
 blo-4.733
yl-4.673
 Jas-4.373
ples-4.310
Token,
Token ID11
Feature activation-3.81e+00

Pos logit contributions
 tum+4.086
 thirst+3.778
night+3.658
 hind+3.342
edly+3.325
Neg logit contributions
yl-2.957
 Kay-2.941
ples-2.922
 blo-2.857
 Jas-2.837
Token Bob
Token ID5811
Feature activation-3.86e-01

Pos logit contributions
 tum+7.876
night+7.361
 thirst+7.156
 hind+6.934
ages+6.823
Neg logit contributions
 Kay-5.226
 blo-4.575
yl-4.390
 Jas-4.365
grown-4.363
Tokeno
Token ID78
Feature activation-1.06e+00

Pos logit contributions
 tum+0.664
 thirst+0.590
night+0.585
 steak+0.553
 lung+0.527
Neg logit contributions
ples-0.636
 Jas-0.583
ount-0.578
yl-0.573
 paints-0.555
Token fell
Token ID3214
Feature activation-1.25e+00

Pos logit contributions
 tum+2.077
 thirst+1.910
night+1.898
 steak+1.820
 lung+1.731
Neg logit contributions
ples-1.435
 Jas-1.311
yl-1.284
ount-1.248
 paints-1.237
Token into
Token ID656
Feature activation-1.25e+00

Pos logit contributions
 tum+2.287
 thirst+2.101
night+2.001
 steak+1.970
 lung+1.841
Neg logit contributions
ples-1.903
 Jas-1.790
 blo-1.740
ount-1.718
 paints-1.698
Token the
Token ID262
Feature activation-1.06e+00

Pos logit contributions
 tum+2.347
night+2.215
 thirst+2.100
bye+2.041
uch+1.944
Neg logit contributions
 Jas-1.754
ples-1.732
yl-1.647
 Kay-1.629
 paints-1.581
Token away
Token ID1497
Feature activation-6.20e-01

Pos logit contributions
 tum+1.073
 thirst+0.936
night+0.910
 steak+0.872
 lung+0.862
Neg logit contributions
ples-1.138
 Jas-1.065
ount-1.048
 Kay-1.024
yl-1.022
Token.
Token ID13
Feature activation-6.98e-01

Pos logit contributions
 tum+1.061
night+0.939
 thirst+0.937
 steak+0.891
 lung+0.859
Neg logit contributions
ples-1.007
 Jas-0.920
ount-0.918
 blo-0.910
yl-0.903
Token The
Token ID383
Feature activation-1.25e+00

Pos logit contributions
 tum+0.999
 thirst+0.879
night+0.854
 steak+0.805
 lung+0.784
Neg logit contributions
ples-1.322
 Jas-1.249
ount-1.242
 Kay-1.202
 paints-1.183
Token bunny
Token ID44915
Feature activation-1.57e+00

Pos logit contributions
 tum+2.065
night+1.840
 thirst+1.815
 steak+1.608
time+1.602
Neg logit contributions
ples-2.045
 blo-1.926
 paints-1.880
 violet-1.862
yl-1.844
Token enjoyed
Token ID8359
Feature activation-3.64e-01

Pos logit contributions
 tum+2.971
night+2.684
 thirst+2.651
 steak+2.604
 lung+2.436
Neg logit contributions
ples-2.252
 blo-2.075
yl-2.001
 paints-1.953
 Jas-1.950
Token his
Token ID465
Feature activation-3.80e+00

Pos logit contributions
 tum+0.659
night+0.580
 thirst+0.576
 steak+0.530
 lung+0.525
Neg logit contributions
ples-0.566
 Jas-0.524
ount-0.508
 blo-0.506
yl-0.504
Token steak
Token ID26320
Feature activation-2.34e+00

Pos logit contributions
 tum+6.168
 thirst+5.633
 drowned+5.602
night+5.559
edly+5.431
Neg logit contributions
 paints-5.997
 blo-5.988
ples-5.776
 violet-5.736
 freshly-5.613
Token and
Token ID290
Feature activation-1.13e+00

Pos logit contributions
 tum+4.373
night+3.967
 thirst+3.794
bye+3.646
 drowned+3.575
Neg logit contributions
 blo-3.447
ples-3.372
 written-3.007
 paints-2.961
yl-2.946
Token the
Token ID262
Feature activation-1.23e+00

Pos logit contributions
 tum+2.058
night+1.833
 thirst+1.810
 steak+1.727
 lung+1.669
Neg logit contributions
ples-1.707
 blo-1.592
 Jas-1.580
 Kay-1.528
 paints-1.503
Token huge
Token ID3236
Feature activation-1.08e+00

Pos logit contributions
 tum+2.111
night+1.898
 thirst+1.880
time+1.657
 steak+1.656
Neg logit contributions
ples-1.902
 blo-1.820
 paints-1.767
 violet-1.746
yl-1.707
Token tree
Token ID5509
Feature activation-9.39e-01

Pos logit contributions
 tum+1.696
 thirst+1.528
night+1.493
 steak+1.328
 drowned+1.321
Neg logit contributions
ples-1.824
 blo-1.680
 paints-1.645
 violet-1.625
yl-1.611
Token ground
Token ID2323
Feature activation-1.37e+00

Pos logit contributions
 tum+5.165
 thirst+4.997
night+4.967
time+4.497
 drowned+4.357
Neg logit contributions
ples-3.424
 blo-3.415
 paints-3.242
 violet-3.211
 mirror-3.011
Token and
Token ID290
Feature activation-2.63e+00

Pos logit contributions
 tum+2.604
 thirst+2.355
night+2.342
 steak+2.213
 lung+2.188
Neg logit contributions
ples-1.939
 blo-1.796
 paints-1.746
 Jas-1.741
ount-1.702
Token started
Token ID2067
Feature activation-1.12e+00

Pos logit contributions
 tum+5.122
 thirst+4.656
night+4.439
time+4.427
 team+4.373
Neg logit contributions
 paints-3.476
 blo-3.414
ples-3.393
 written-3.210
 mirror-3.189
Token to
Token ID284
Feature activation-1.89e+00

Pos logit contributions
 tum+1.887
 thirst+1.704
night+1.646
 steak+1.524
bye+1.500
Neg logit contributions
ples-1.825
 blo-1.700
 paints-1.689
 Jas-1.659
 violet-1.610
Token stretch
Token ID7539
Feature activation-1.73e+00

Pos logit contributions
 tum+3.153
night+2.778
 thirst+2.770
bye+2.518
 steak+2.508
Neg logit contributions
ples-3.102
 blo-3.031
 paints-2.992
 violet-2.849
 mirror-2.842
Token his
Token ID465
Feature activation-3.79e+00

Pos logit contributions
 tum+3.361
night+3.013
 thirst+3.008
 steak+2.791
uch+2.780
Neg logit contributions
ples-2.432
 Jas-2.355
 blo-2.249
 paints-2.170
 Kay-2.168
Token muscles
Token ID12749
Feature activation-1.29e+00

Pos logit contributions
night+6.262
 thirst+5.895
bye+5.884
 tum+5.853
 unfortunately+5.750
Neg logit contributions
 blo-6.233
 paints-6.087
 mirror-5.843
ples-5.747
 ey-5.662
Token.
Token ID13
Feature activation-1.65e+00

Pos logit contributions
 tum+2.577
 thirst+2.339
night+2.305
 steak+2.166
 lung+2.102
Neg logit contributions
ples-1.723
 blo-1.580
ount-1.523
 paints-1.522
 Jas-1.495
Token Suddenly
Token ID24975
Feature activation-1.43e+00

Pos logit contributions
 tum+2.565
 thirst+2.407
night+2.336
 steak+2.205
uch+2.175
Neg logit contributions
ples-2.855
 Jas-2.704
ount-2.660
 paints-2.593
 violet-2.577
Token he
Token ID339
Feature activation-1.54e+00

Pos logit contributions
 tum+2.450
 thirst+2.314
night+2.254
 steak+2.039
edly+2.019
Neg logit contributions
ples-2.141
 Jas-2.058
yl-2.027
 Kay-1.974
 paints-1.964
Token heard
Token ID2982
Feature activation-7.03e-01

Pos logit contributions
 tum+3.304
night+3.145
 thirst+3.126
 steak+2.969
bye+2.900
Neg logit contributions
ples-1.738
 drawn-1.505
yl-1.499
 blo-1.497
 paints-1.496
Token ac
Token ID936
Feature activation-1.04e+00

Pos logit contributions
 tum+3.200
night+3.040
 thirst+2.982
bye+2.787
 steak+2.752
Neg logit contributions
ples-2.843
 blo-2.561
 paints-2.440
ount-2.404
 drawn-2.396
Tokenhe
Token ID258
Feature activation-1.86e+00

Pos logit contributions
 tum+2.149
 thirst+1.907
night+1.907
 steak+1.785
 lung+1.724
Neg logit contributions
ples-1.386
ount-1.237
yl-1.218
 Jas-1.189
 paints-1.184
Token.
Token ID13
Feature activation-7.81e-01

Pos logit contributions
 tum+3.768
night+3.496
 thirst+3.373
bye+3.232
 steak+3.143
Neg logit contributions
ples-2.360
 blo-2.139
 Jas-2.134
yl-2.062
ount-2.029
Token He
Token ID679
Feature activation-1.42e+00

Pos logit contributions
 tum+1.101
 thirst+0.981
night+0.976
 steak+0.888
 lung+0.847
Neg logit contributions
ples-1.467
 Jas-1.416
ount-1.393
 paints-1.347
 Kay-1.346
Token felt
Token ID2936
Feature activation-1.21e+00

Pos logit contributions
 tum+2.567
night+2.361
 thirst+2.326
 steak+2.245
bye+2.179
Neg logit contributions
ples-2.003
 paints-1.837
 drawn-1.787
 Jas-1.786
 blo-1.785
Token his
Token ID465
Feature activation-3.79e+00

Pos logit contributions
 tum+1.899
night+1.776
 thirst+1.680
bye+1.634
 steak+1.616
Neg logit contributions
ples-1.937
 blo-1.803
ount-1.781
 paints-1.764
 drawn-1.749
Token head
Token ID1182
Feature activation-8.87e-01

Pos logit contributions
night+5.346
His+5.334
bye+5.185
 steak+4.905
edly+4.694
Neg logit contributions
 blo-6.737
 paints-6.593
 mirror-6.447
 makeup-6.160
ples-5.906
Token spin
Token ID7906
Feature activation-9.91e-01

Pos logit contributions
 tum+1.513
night+1.366
 thirst+1.360
 steak+1.279
bye+1.197
Neg logit contributions
ples-1.428
ount-1.283
 blo-1.260
 Jas-1.244
 paints-1.241
Token.
Token ID13
Feature activation-7.09e-01

Pos logit contributions
 tum+1.871
 thirst+1.686
night+1.630
 steak+1.587
 lung+1.532
Neg logit contributions
ples-1.460
ount-1.325
 Jas-1.314
 paints-1.267
yl-1.255
Token He
Token ID679
Feature activation-1.31e+00

Pos logit contributions
 tum+0.988
 thirst+0.878
night+0.867
 steak+0.795
 lung+0.760
Neg logit contributions
ples-1.347
 Jas-1.299
ount-1.279
 paints-1.238
 Kay-1.234
Token felt
Token ID2936
Feature activation-8.96e-01

Pos logit contributions
 tum+2.309
night+2.112
 thirst+2.062
 steak+2.024
bye+1.932
Neg logit contributions
ples-1.955
 paints-1.787
 Jas-1.764
 blo-1.739
 drawn-1.730
Tokenmy
Token ID1820
Feature activation-2.39e+00

Pos logit contributions
 tum+1.030
 thirst+0.953
night+0.941
 steak+0.910
 lung+0.885
Neg logit contributions
ples-0.298
 Jas-0.243
ount-0.231
 paints-0.230
yl-0.225
Token fell
Token ID3214
Feature activation-1.35e+00

Pos logit contributions
 tum+4.458
night+4.330
 thirst+4.329
bye+4.070
 steak+3.958
Neg logit contributions
 paints-3.130
ples-3.129
 drawn-2.922
 written-2.881
yl-2.839
Token down
Token ID866
Feature activation-9.05e-01

Pos logit contributions
 tum+2.535
 thirst+2.363
night+2.238
 steak+2.182
 lung+2.051
Neg logit contributions
ples-1.936
 Jas-1.801
 blo-1.781
 paints-1.741
yl-1.735
Token and
Token ID290
Feature activation-1.82e+00

Pos logit contributions
 tum+1.621
 thirst+1.458
night+1.405
 steak+1.321
 lung+1.294
Neg logit contributions
ples-1.377
 Jas-1.323
yl-1.300
 blo-1.275
ount-1.271
Token hurt
Token ID5938
Feature activation-1.84e+00

Pos logit contributions
 tum+2.889
night+2.710
 thirst+2.699
 steak+2.558
bye+2.485
Neg logit contributions
ples-2.825
 Jas-2.718
 paints-2.705
 blo-2.704
 drawn-2.663
Token his
Token ID465
Feature activation-3.78e+00

Pos logit contributions
 tum+2.482
night+2.172
 thirst+2.085
 steak+1.902
uch+1.893
Neg logit contributions
ples-3.619
 Jas-3.613
 Kay-3.405
yl-3.355
grown-3.353
Token knee
Token ID10329
Feature activation-1.46e+00

Pos logit contributions
night+5.278
bye+4.741
 steak+4.696
 thirst+4.632
 tum+4.619
Neg logit contributions
 mirror-6.715
 blo-6.418
 paints-6.409
 makeup-6.334
ples-6.228
Token.
Token ID13
Feature activation-1.18e+00

Pos logit contributions
 tum+2.938
 thirst+2.794
night+2.733
 steak+2.656
 lung+2.468
Neg logit contributions
ples-1.826
 blo-1.573
ount-1.546
 drawn-1.544
 written-1.542
Token
Token ID220
Feature activation-8.71e-01

Pos logit contributions
 tum+1.951
 thirst+1.855
night+1.853
 steak+1.716
 lung+1.656
Neg logit contributions
ples-1.804
 Jas-1.775
 Kay-1.707
ount-1.703
yl-1.651
Token
Token ID198
Feature activation-6.42e-01

Pos logit contributions
 tum+1.479
 thirst+1.359
night+1.348
 steak+1.253
 lung+1.216
Neg logit contributions
ples-1.381
ount-1.333
 Jas-1.332
yl-1.255
 paints-1.247
Token
Token ID198
Feature activation-1.21e+00

Pos logit contributions
 tum+1.075
 thirst+0.978
night+0.962
 steak+0.896
 lung+0.872
Neg logit contributions
ples-1.046
 Jas-1.008
ount-0.985
yl-0.977
 Kay-0.953
Token and
Token ID290
Feature activation+5.03e-01

Pos logit contributions
 tum+0.422
 thirst+0.379
night+0.372
 steak+0.353
 lung+0.343
Neg logit contributions
ples-0.361
ount-0.331
 Jas-0.329
 paints-0.319
yl-0.319
Token angry
Token ID7954
Feature activation+9.37e-01

Pos logit contributions
ples+0.752
ount+0.679
 Jas+0.677
yl+0.655
 paints+0.646
Neg logit contributions
 tum-0.961
 thirst-0.853
night-0.849
 steak-0.800
 lung-0.780
Token.
Token ID13
Feature activation+1.65e+00

Pos logit contributions
ples+1.630
 Jas+1.508
ount+1.503
 paints+1.461
yl+1.455
Neg logit contributions
 tum-1.572
night-1.351
 thirst-1.332
 steak-1.263
 lung-1.228
Token She
Token ID1375
Feature activation+7.47e-01

Pos logit contributions
ples+1.904
ount+1.654
yl+1.608
 Jas+1.528
ptions+1.519
Neg logit contributions
 tum-3.865
night-3.407
 thirst-3.377
 steak-3.199
 leg-3.180
Token did
Token ID750
Feature activation+6.16e-01

Pos logit contributions
ples+1.222
ount+1.130
 Jas+1.112
yl+1.073
 paints+1.066
Neg logit contributions
 tum-1.314
 thirst-1.146
night-1.143
 steak-1.073
 lung-1.045
Token not
Token ID407
Feature activation+2.38e+00

Pos logit contributions
ples+1.449
ount+1.355
 Jas+1.350
 paints+1.313
yl+1.311
Neg logit contributions
 tum-0.650
 thirst-0.518
night-0.513
 steak-0.466
 lung-0.435
Token like
Token ID588
Feature activation+8.14e-01

Pos logit contributions
ples+4.275
 Jas+3.897
 paints+3.815
ount+3.787
els+3.739
Neg logit contributions
 tum-4.054
night-3.665
 thirst-3.518
 steak-3.362
 leg-3.253
Token the
Token ID262
Feature activation+4.99e-01

Pos logit contributions
ples+1.386
ount+1.273
 Jas+1.251
yl+1.217
ydia+1.199
Neg logit contributions
 tum-1.392
 thirst-1.193
night-1.189
 steak-1.149
 lung-1.108
Token boy
Token ID2933
Feature activation+2.78e-01

Pos logit contributions
ples+0.903
ount+0.833
 Jas+0.829
yl+0.799
 paints+0.792
Neg logit contributions
 tum-0.804
 thirst-0.688
night-0.684
 steak-0.650
 lung-0.632
Token.
Token ID13
Feature activation+8.82e-01

Pos logit contributions
ples+0.460
 Jas+0.423
ount+0.422
yl+0.407
 paints+0.407
Neg logit contributions
 tum-0.477
 thirst-0.417
night-0.414
 steak-0.391
 lung-0.380
Token She
Token ID1375
Feature activation+6.71e-01

Pos logit contributions
ples+0.877
ount+0.746
 Jas+0.727
yl+0.723
 paints+0.680
Neg logit contributions
 tum-2.135
night-1.916
 thirst-1.916
 steak-1.845
 lung-1.819
Token She
Token ID1375
Feature activation+3.10e+00

Pos logit contributions
ples+1.644
ount+1.433
ptions+1.395
yl+1.377
 Jas+1.363
Neg logit contributions
 tum-3.388
night-3.002
 thirst-2.900
 steak-2.873
 lung-2.815
Token goes
Token ID2925
Feature activation+2.08e+00

Pos logit contributions
ples+4.905
 Jas+4.581
ptions+4.402
yl+4.376
ount+4.371
Neg logit contributions
 tum-5.959
night-5.238
 steak-5.049
 thirst-5.012
 lung-4.903
Token to
Token ID284
Feature activation+3.55e-01

Pos logit contributions
ples+4.972
 Jas+4.421
ount+4.379
od+4.356
 violet+4.337
Neg logit contributions
 tum-2.283
night-1.905
 thirst-1.772
 steak-1.630
 leg-1.607
Token them
Token ID606
Feature activation+5.64e-01

Pos logit contributions
ples+0.552
ount+0.502
 Jas+0.498
yl+0.481
 paints+0.480
Neg logit contributions
 tum-0.648
 thirst-0.571
night-0.569
 steak-0.542
 lung-0.525
Token.
Token ID13
Feature activation+1.18e+00

Pos logit contributions
ples+0.972
 Jas+0.895
ount+0.891
yl+0.864
 paints+0.857
Neg logit contributions
 tum-0.933
 thirst-0.807
night-0.805
 steak-0.761
 lung-0.739
Token She
Token ID1375
Feature activation+2.73e+00

Pos logit contributions
ples+1.326
ount+1.152
yl+1.116
 Jas+1.109
ptions+1.087
Neg logit contributions
 tum-2.732
night-2.436
 thirst-2.370
 steak-2.335
 lung-2.275
Token says
Token ID1139
Feature activation+6.26e-01

Pos logit contributions
ples+4.139
 Jas+3.718
ount+3.650
yl+3.558
ptions+3.552
Neg logit contributions
 tum-5.454
night-4.674
 steak-4.635
 thirst-4.567
 lung-4.490
Token,
Token ID11
Feature activation+1.15e+00

Pos logit contributions
ples+1.250
ount+1.158
 Jas+1.148
 paints+1.111
ydia+1.110
Neg logit contributions
 tum-0.889
night-0.734
 thirst-0.731
 steak-0.690
 lung-0.656
Token "
Token ID366
Feature activation+1.13e+00

Pos logit contributions
ples+1.779
ount+1.631
 Jas+1.612
yl+1.555
ydia+1.552
Neg logit contributions
 tum-2.167
 thirst-1.837
night-1.822
 steak-1.771
 lung-1.730
TokenHi
Token ID17250
Feature activation+7.39e-01

Pos logit contributions
ples+1.457
ount+1.294
 Jas+1.292
 paints+1.215
yl+1.213
Neg logit contributions
 tum-2.354
night-2.152
 thirst-2.095
 steak-2.083
 lung-2.018
Token,
Token ID11
Feature activation+1.33e+00

Pos logit contributions
ples+1.477
ount+1.374
 Jas+1.355
ydia+1.323
 paints+1.319
Neg logit contributions
 tum-1.038
night-0.868
 thirst-0.862
 steak-0.822
 lung-0.776
Token a
Token ID257
Feature activation+1.36e+00

Pos logit contributions
 tum+0.025
 thirst+0.022
night+0.022
 steak+0.020
 lung+0.019
Neg logit contributions
ples-0.031
 Jas-0.029
ount-0.029
 paints-0.028
yl-0.028
Token big
Token ID1263
Feature activation+1.30e+00

Pos logit contributions
ples+2.487
ount+2.340
 Jas+2.266
ydia+2.189
ptions+2.127
Neg logit contributions
 tum-2.274
 steak-1.910
 thirst-1.870
 lung-1.821
night-1.817
Token bush
Token ID24484
Feature activation+1.74e+00

Pos logit contributions
ples+2.482
ount+2.389
 Jas+2.305
ydia+2.245
ptions+2.189
Neg logit contributions
 tum-2.008
 steak-1.669
 thirst-1.654
night-1.618
 lung-1.571
Token with
Token ID351
Feature activation+1.47e+00

Pos logit contributions
ples+3.532
 Jas+3.286
ount+3.270
ydia+3.224
ptions+3.174
Neg logit contributions
 tum-2.522
 steak-2.022
 thirst-2.020
night-2.016
time-1.872
Token red
Token ID2266
Feature activation+6.45e-01

Pos logit contributions
ples+2.983
ount+2.843
 Jas+2.773
yl+2.712
ydia+2.711
Neg logit contributions
 tum-2.118
 thirst-1.740
 steak-1.657
night-1.608
 lung-1.515
Token flowers
Token ID12734
Feature activation+2.41e+00

Pos logit contributions
ples+1.118
ount+1.043
 Jas+1.043
yl+0.998
ydia+0.996
Neg logit contributions
 tum-1.070
 thirst-0.915
night-0.908
 steak-0.866
 lung-0.831
Token.
Token ID13
Feature activation-3.88e-01

Pos logit contributions
ples+5.069
 Jas+4.982
ydia+4.899
ount+4.828
ptions+4.780
Neg logit contributions
 tum-3.257
 thirst-2.558
 steak-2.532
 dinner-2.415
 lung-2.395
Token They
Token ID1119
Feature activation+1.29e+00

Pos logit contributions
 tum+0.680
 thirst+0.613
night+0.610
 steak+0.564
 lung+0.561
Neg logit contributions
ples-0.607
 Jas-0.564
ount-0.559
 paints-0.543
 Kay-0.532
Token wanted
Token ID2227
Feature activation+7.81e-01

Pos logit contributions
ples+1.999
ount+1.848
 Jas+1.734
yl+1.728
ydia+1.715
Neg logit contributions
 tum-2.416
 thirst-2.100
night-2.017
 steak-1.966
 lung-1.893
Token to
Token ID284
Feature activation+9.42e-01

Pos logit contributions
ples+1.666
ount+1.560
 Jas+1.541
ydia+1.494
yl+1.488
Neg logit contributions
 tum-1.017
 thirst-0.822
night-0.809
 steak-0.778
 lung-0.724
Token cut
Token ID2005
Feature activation+4.80e-01

Pos logit contributions
ples+1.629
ount+1.485
 Jas+1.459
yl+1.420
ydia+1.412
Neg logit contributions
 tum-1.625
 thirst-1.386
night-1.346
 steak-1.276
 lung-1.243
Token
Token ID198
Feature activation+9.12e-01

Pos logit contributions
ples+1.136
ount+1.045
 Jas+1.041
 paints+1.019
yl+1.008
Neg logit contributions
 tum-0.829
night-0.691
 thirst-0.690
 steak-0.644
 lung-0.616
TokenSo
Token ID2396
Feature activation-3.29e-01

Pos logit contributions
ples+1.737
ount+1.609
 Jas+1.607
 paints+1.558
yl+1.554
Neg logit contributions
 tum-1.351
 thirst-1.146
night-1.144
 steak-1.049
time-1.023
Token,
Token ID11
Feature activation+3.70e-02

Pos logit contributions
 tum+0.523
 thirst+0.462
night+0.453
 steak+0.436
 lung+0.426
Neg logit contributions
ples-0.570
 Jas-0.536
yl-0.521
ount-0.518
 paints-0.514
Token Lily
Token ID20037
Feature activation-2.47e-01

Pos logit contributions
ples+0.051
 Jas+0.046
ount+0.046
 paints+0.044
yl+0.044
Neg logit contributions
 tum-0.074
 thirst-0.066
night-0.066
 steak-0.063
 lung-0.061
Token took
Token ID1718
Feature activation+3.95e-01

Pos logit contributions
 tum+0.428
 thirst+0.379
night+0.369
 steak+0.353
 lung+0.341
Neg logit contributions
ples-0.403
 Jas-0.370
ount-0.365
 paints-0.362
yl-0.358
Token the
Token ID262
Feature activation+2.47e+00

Pos logit contributions
ples+0.669
ount+0.617
 Jas+0.613
yl+0.591
 paints+0.589
Neg logit contributions
 tum-0.666
 thirst-0.580
night-0.580
 steak-0.545
 lung-0.527
Token purse
Token ID27517
Feature activation-9.47e-01

Pos logit contributions
ples+4.062
ount+3.619
 Jas+3.536
od+3.313
ptions+3.201
Neg logit contributions
 tum-5.343
 thirst-4.284
night-4.216
 leg-4.210
 steak-4.112
Token and
Token ID290
Feature activation+9.76e-02

Pos logit contributions
 tum+1.653
 thirst+1.519
night+1.440
 steak+1.363
 lung+1.349
Neg logit contributions
ples-1.466
ount-1.392
 paints-1.385
yl-1.380
 Jas-1.380
Token put
Token ID1234
Feature activation+1.83e+00

Pos logit contributions
ples+0.153
 Jas+0.141
ount+0.140
 paints+0.136
yl+0.135
Neg logit contributions
 tum-0.174
 thirst-0.154
night-0.153
 steak-0.145
 lung-0.141
Token her
Token ID607
Feature activation+9.17e-01

Pos logit contributions
ples+3.571
ount+3.411
 Jas+3.279
ydia+3.164
ptions+3.158
Neg logit contributions
 tum-2.761
 thirst-2.334
night-2.288
 steak-2.229
 dinner-2.160
Token favorite
Token ID4004
Feature activation+1.19e+00

Pos logit contributions
ples+1.486
ount+1.366
 Jas+1.365
ydia+1.285
yl+1.273
Neg logit contributions
 tum-1.682
 thirst-1.462
night-1.414
 steak-1.352
 lung-1.326
Token find
Token ID1064
Feature activation+1.91e+00

Pos logit contributions
ples+1.046
ount+0.950
 Jas+0.938
 paints+0.909
yl+0.893
Neg logit contributions
 tum-1.282
 thirst-1.129
night-1.110
 steak-1.060
 lung-1.046
Token anything
Token ID1997
Feature activation+1.48e+00

Pos logit contributions
ples+3.929
ount+3.756
 Jas+3.673
 typing+3.519
ydia+3.508
Neg logit contributions
 tum-2.736
 thirst-2.306
night-2.163
 steak-2.076
 lung-2.036
Token.
Token ID13
Feature activation+8.24e-02

Pos logit contributions
ples+2.783
 Jas+2.592
ount+2.557
ydia+2.471
ptions+2.455
Neg logit contributions
 tum-2.341
 thirst-1.955
night-1.922
 steak-1.883
 dinner-1.794
Token Then
Token ID3244
Feature activation-4.64e-01

Pos logit contributions
ples+0.105
 Jas+0.094
ount+0.094
 paints+0.090
yl+0.089
Neg logit contributions
 tum-0.172
 thirst-0.155
night-0.154
 steak-0.147
 lung-0.144
Token she
Token ID673
Feature activation+1.42e-01

Pos logit contributions
 tum+0.873
 thirst+0.784
night+0.777
 steak+0.745
 lung+0.729
Neg logit contributions
ples-0.658
 Jas-0.611
yl-0.592
 Kay-0.586
 blo-0.582
Token went
Token ID1816
Feature activation+2.49e+00

Pos logit contributions
ples+0.170
 Jas+0.151
ount+0.151
yl+0.144
 paints+0.144
Neg logit contributions
 tum-0.309
 thirst-0.279
night-0.278
 steak-0.266
 lung-0.260
Token outside
Token ID2354
Feature activation+8.47e-02

Pos logit contributions
ples+6.209
ount+5.695
od+5.436
 Jas+5.427
ptions+5.389
Neg logit contributions
 tum-2.758
night-2.251
 dinner-2.114
 thirst-2.062
 steak-2.014
Token and
Token ID290
Feature activation+1.00e+00

Pos logit contributions
ples+0.156
 Jas+0.146
ount+0.145
yl+0.141
 paints+0.141
Neg logit contributions
 tum-0.128
 thirst-0.110
night-0.109
 steak-0.102
 lung-0.099
Token searched
Token ID16499
Feature activation+1.27e+00

Pos logit contributions
ples+1.319
ount+1.186
 Jas+1.127
ydia+1.080
yl+1.079
Neg logit contributions
 tum-2.123
 thirst-1.911
night-1.901
 steak-1.762
 lung-1.736
Token between
Token ID1022
Feature activation+1.78e+00

Pos logit contributions
ples+2.588
ount+2.450
 Jas+2.360
ydia+2.355
yl+2.332
Neg logit contributions
 tum-1.728
 thirst-1.466
night-1.439
 steak-1.329
 dinner-1.297
Token the
Token ID262
Feature activation+1.23e+00

Pos logit contributions
ples+2.976
ount+2.863
 violet+2.583
yl+2.574
ydia+2.542
Neg logit contributions
 tum-3.175
 thirst-2.764
 steak-2.635
 dinner-2.610
night-2.585
Token and
Token ID290
Feature activation+5.37e-01

Pos logit contributions
 tum+1.417
 thirst+1.270
night+1.228
 lung+1.155
 steak+1.144
Neg logit contributions
ples-1.288
 Jas-1.189
ount-1.179
 paints-1.176
 blo-1.152
Token kisses
Token ID36544
Feature activation-3.85e-01

Pos logit contributions
ples+0.673
ount+0.601
 Jas+0.593
yl+0.575
ydia+0.554
Neg logit contributions
 tum-1.147
 thirst-1.028
night-1.028
 steak-0.991
 lung-0.960
Token them
Token ID606
Feature activation-1.64e+00

Pos logit contributions
 tum+0.907
 thirst+0.835
night+0.815
 steak+0.784
 lung+0.777
Neg logit contributions
ples-0.386
 Jas-0.342
ount-0.331
 paints-0.316
yl-0.314
Token.
Token ID13
Feature activation+9.22e-01

Pos logit contributions
 tum+3.388
 thirst+3.064
night+2.914
 lung+2.877
 current+2.869
Neg logit contributions
ples-2.025
 paints-1.834
ount-1.830
 blo-1.826
 Jas-1.796
Token She
Token ID1375
Feature activation+1.55e+00

Pos logit contributions
ples+1.191
ount+1.057
yl+1.045
 Jas+1.026
ptions+0.988
Neg logit contributions
 tum-1.980
night-1.736
 thirst-1.734
 steak-1.675
 lung-1.612
Token is
Token ID318
Feature activation+1.27e+00

Pos logit contributions
ples+2.211
ount+1.979
 Jas+1.976
ptions+1.917
yl+1.910
Neg logit contributions
 tum-3.147
night-2.775
 thirst-2.747
 steak-2.717
 lung-2.598
Token proud
Token ID6613
Feature activation+2.63e-01

Pos logit contributions
ples+2.397
ount+2.188
 Jas+2.138
yl+2.116
 paints+2.098
Neg logit contributions
 tum-2.009
 thirst-1.745
night-1.644
 steak-1.636
 lung-1.544
Token of
Token ID286
Feature activation+8.21e-01

Pos logit contributions
ples+0.432
ount+0.396
 Jas+0.395
yl+0.382
 paints+0.381
Neg logit contributions
 tum-0.457
 thirst-0.400
night-0.398
 steak-0.376
 lung-0.365
Token them
Token ID606
Feature activation-6.50e-01

Pos logit contributions
ples+1.062
ount+0.945
 Jas+0.918
 paints+0.892
yl+0.886
Neg logit contributions
 tum-1.711
 thirst-1.553
night-1.532
 steak-1.488
 lung-1.437
Token.
Token ID13
Feature activation+1.44e+00

Pos logit contributions
 tum+1.243
 thirst+1.113
night+1.098
 steak+1.027
 lung+1.017
Neg logit contributions
ples-0.925
ount-0.846
 Jas-0.834
 paints-0.817
 blo-0.814
Token
Token ID198
Feature activation+9.12e-01

Pos logit contributions
ples+2.412
yl+2.180
ount+2.171
ptions+2.111
 Jas+2.101
Neg logit contributions
 tum-2.597
night-2.184
 thirst-2.170
 steak-2.108
 lung-1.993
TokenIn
Token ID818
Feature activation+1.14e+00

Pos logit contributions
ples+0.890
ount+0.832
 Jas+0.830
 paints+0.810
yl+0.808
Neg logit contributions
 tum-0.506
 thirst-0.417
night-0.415
 steak-0.381
 lung-0.361
Token the
Token ID262
Feature activation-9.46e-02

Pos logit contributions
ples+2.078
ount+1.985
 Jas+1.820
yl+1.782
 paints+1.780
Neg logit contributions
 tum-1.806
 thirst-1.556
night-1.515
 steak-1.472
 dinner-1.420
Token end
Token ID886
Feature activation+2.28e+00

Pos logit contributions
 tum+0.138
 thirst+0.118
night+0.118
 steak+0.109
 lung+0.106
Neg logit contributions
ples-0.180
 Jas-0.168
ount-0.167
 paints-0.163
yl-0.162
Token,
Token ID11
Feature activation+8.19e-01

Pos logit contributions
ples+5.570
ount+5.100
ydia+4.964
ptions+4.920
 violet+4.873
Neg logit contributions
 tum-2.488
time-1.769
night-1.721
 team-1.686
 thirst-1.659
Token Sally
Token ID25737
Feature activation+1.95e-01

Pos logit contributions
ples+1.298
ount+1.177
 Jas+1.119
yl+1.098
 paints+1.096
Neg logit contributions
 tum-1.497
 thirst-1.321
night-1.302
 steak-1.229
 dinner-1.193
Token and
Token ID290
Feature activation+3.91e-01

Pos logit contributions
ples+0.321
 Jas+0.295
ount+0.295
yl+0.284
 paints+0.284
Neg logit contributions
 tum-0.335
 thirst-0.293
night-0.292
 steak-0.275
 lung-0.267
Token Jill
Token ID22772
Feature activation-1.08e-01

Pos logit contributions
ples+0.436
ount+0.385
 Jas+0.377
yl+0.357
 paints+0.356
Neg logit contributions
 tum-0.897
night-0.813
 thirst-0.811
 steak-0.777
 lung-0.760
Token were
Token ID547
Feature activation+8.53e-01

Pos logit contributions
 tum+0.185
 thirst+0.163
night+0.161
 steak+0.154
 lung+0.149
Neg logit contributions
ples-0.176
 Jas-0.163
ount-0.162
 paints-0.157
yl-0.156
Token so
Token ID523
Feature activation+3.96e-02

Pos logit contributions
ples+1.512
ount+1.392
 Jas+1.363
 paints+1.346
yl+1.343
Neg logit contributions
 tum-1.380
 thirst-1.195
night-1.168
 steak-1.107
 lung-1.075
Token tired
Token ID10032
Feature activation+7.18e-02

Pos logit contributions
ples+0.066
 Jas+0.061
ount+0.061
yl+0.059
 paints+0.059
Neg logit contributions
 tum-0.067
 thirst-0.059
night-0.059
 steak-0.055
 lung-0.054
Token that
Token ID326
Feature activation+1.38e-02

Pos logit contributions
ples+0.114
 Jas+0.105
ount+0.105
yl+0.101
 paints+0.101
Neg logit contributions
 tum-0.127
 thirst-0.112
night-0.111
 steak-0.105
 lung-0.102
Token They
Token ID1119
Feature activation+4.38e-01

Pos logit contributions
 tum+0.261
 thirst+0.228
night+0.227
 steak+0.211
 lung+0.205
Neg logit contributions
ples-0.311
 Jas-0.293
ount-0.292
 paints-0.280
 Kay-0.279
Token loved
Token ID6151
Feature activation+6.80e-02

Pos logit contributions
ples+0.763
ount+0.705
 Jas+0.701
yl+0.681
 paints+0.674
Neg logit contributions
 tum-0.715
 thirst-0.621
night-0.615
 steak-0.578
 lung-0.560
Token it
Token ID340
Feature activation-1.63e+00

Pos logit contributions
ples+0.112
 Jas+0.103
ount+0.102
 paints+0.099
yl+0.099
Neg logit contributions
 tum-0.117
 thirst-0.103
night-0.103
 steak-0.096
 lung-0.094
Token and
Token ID290
Feature activation-4.14e-02

Pos logit contributions
 tum+3.107
 thirst+2.766
night+2.697
 steak+2.522
 lung+2.504
Neg logit contributions
ples-2.389
 blo-2.224
 Jas-2.175
 paints-2.120
yl-2.107
Token asked
Token ID1965
Feature activation-6.53e-01

Pos logit contributions
 tum+0.069
 thirst+0.060
night+0.060
 steak+0.057
 lung+0.055
Neg logit contributions
ples-0.070
 Jas-0.065
ount-0.064
 paints-0.062
yl-0.062
Token Tim
Token ID5045
Feature activation+6.32e-01

Pos logit contributions
 tum+1.084
 thirst+0.952
night+0.941
 steak+0.883
 lung+0.879
Neg logit contributions
ples-1.105
 Jas-1.052
ount-1.013
 Kay-0.989
yl-0.987
Tokenmy
Token ID1820
Feature activation+2.03e-01

Pos logit contributions
ples+0.627
 Jas+0.537
ount+0.535
yl+0.504
 paints+0.495
Neg logit contributions
 tum-1.514
night-1.374
 thirst-1.368
 steak-1.320
 lung-1.289
Token how
Token ID703
Feature activation-3.02e-01

Pos logit contributions
ples+0.357
ount+0.330
 Jas+0.330
yl+0.319
 paints+0.319
Neg logit contributions
 tum-0.324
 thirst-0.281
night-0.279
 steak-0.262
 lung-0.254
Token he
Token ID339
Feature activation-5.05e-01

Pos logit contributions
 tum+0.449
 thirst+0.388
night+0.384
 steak+0.357
 lung+0.352
Neg logit contributions
ples-0.555
 Jas-0.523
ount-0.516
 paints-0.506
yl-0.503
Token made
Token ID925
Feature activation+3.57e-01

Pos logit contributions
 tum+0.991
 thirst+0.877
night+0.869
 steak+0.835
 lung+0.812
Neg logit contributions
ples-0.708
 Jas-0.627
 paints-0.626
ount-0.622
yl-0.610
Token it
Token ID340
Feature activation-4.94e-01

Pos logit contributions
ples+0.623
ount+0.576
 Jas+0.571
yl+0.554
 paints+0.553
Neg logit contributions
 tum-0.583
 thirst-0.506
night-0.503
 steak-0.477
 lung-0.459
Token�
Token ID129
Feature activation+1.33e+00

Pos logit contributions
ples+1.069
ount+0.993
 Jas+0.986
 paints+0.957
yl+0.949
Neg logit contributions
 tum-0.934
 thirst-0.793
night-0.791
 steak-0.742
 lung-0.720
Token�
Token ID241
Feature activation+1.71e+00

Pos logit contributions
ples+2.114
 Jas+1.968
ount+1.947
 paints+1.944
yl+1.889
Neg logit contributions
 tum-2.340
 thirst-2.091
night-2.056
 steak-1.923
 lung-1.892
TokenWho
Token ID8241
Feature activation+1.33e-01

Pos logit contributions
ples+3.045
ount+2.845
els+2.715
ptions+2.714
 paints+2.710
Neg logit contributions
 tum-2.774
night-2.394
 thirst-2.381
 steak-2.285
bye-2.189
Token are
Token ID389
Feature activation+1.35e+00

Pos logit contributions
ples+0.199
 Jas+0.181
ount+0.181
 paints+0.175
yl+0.174
Neg logit contributions
 tum-0.246
 thirst-0.218
night-0.217
 steak-0.205
 lung-0.200
Token you
Token ID345
Feature activation+1.71e+00

Pos logit contributions
ples+2.869
ount+2.634
 Jas+2.590
 paints+2.514
ydia+2.506
Neg logit contributions
 tum-1.755
 thirst-1.436
night-1.433
 steak-1.372
 lung-1.298
Token?
Token ID30
Feature activation+9.96e-01

Pos logit contributions
ples+3.261
ount+3.063
 Jas+3.040
 paints+2.865
ptions+2.862
Neg logit contributions
 tum-2.583
 thirst-2.230
night-2.194
 lung-2.107
 steak-2.077
Tokenâ
Token ID22940
Feature activation+1.20e+00

Pos logit contributions
ples+1.807
 Jas+1.663
ount+1.659
yl+1.622
 paints+1.619
Neg logit contributions
 tum-1.543
night-1.315
 thirst-1.306
 steak-1.249
 lung-1.224
Token€
Token ID26391
Feature activation+7.74e-01

Pos logit contributions
ples+2.270
 Jas+2.137
ount+2.110
 paints+2.087
yl+2.035
Neg logit contributions
 tum-1.721
night-1.577
 thirst-1.519
bye-1.381
 steak-1.376
Token
Token ID198
Feature activation+6.04e-01

Pos logit contributions
ples+1.503
ount+1.401
 Jas+1.385
 paints+1.355
ydia+1.346
Neg logit contributions
 tum-1.124
 thirst-0.928
night-0.927
 steak-0.872
 lung-0.836
Token
Token ID198
Feature activation+1.07e+00

Pos logit contributions
ples+1.189
ount+1.097
 Jas+1.085
 paints+1.068
yl+1.053
Neg logit contributions
 tum-0.860
night-0.719
 thirst-0.719
 steak-0.676
 lung-0.644
TokenThe
Token ID464
Feature activation+7.98e-01

Pos logit contributions
ples+2.273
ount+2.109
 Jas+2.099
 paints+2.087
yl+2.034
Neg logit contributions
 tum-1.311
night-1.079
 thirst-1.078
 steak-1.016
time-0.967
Token€
Token ID26391
Feature activation+7.54e-01

Pos logit contributions
ples+1.147
 Jas+1.068
ount+1.064
 paints+1.040
yl+1.033
Neg logit contributions
 tum-0.917
night-0.807
 thirst-0.796
 steak-0.730
 lung-0.704
Token The
Token ID383
Feature activation-2.13e-02

Pos logit contributions
ples+1.475
ount+1.374
 Jas+1.356
 paints+1.330
yl+1.323
Neg logit contributions
 tum-1.079
 thirst-0.894
night-0.888
 steak-0.839
 lung-0.805
Token bear
Token ID6842
Feature activation-3.36e-01

Pos logit contributions
 tum+0.036
night+0.031
 thirst+0.031
 steak+0.029
 lung+0.028
Neg logit contributions
ples-0.036
 Jas-0.033
ount-0.033
 paints-0.032
yl-0.032
Token thought
Token ID1807
Feature activation-2.54e-03

Pos logit contributions
 tum+0.625
 thirst+0.556
night+0.549
 steak+0.525
 lung+0.511
Neg logit contributions
ples-0.496
 Jas-0.450
ount-0.445
 paints-0.438
yl-0.437
Token that
Token ID326
Feature activation+2.42e-01

Pos logit contributions
 tum+0.004
 thirst+0.004
night+0.004
 steak+0.004
 lung+0.004
Neg logit contributions
ples-0.004
 Jas-0.004
ount-0.004
yl-0.004
 paints-0.004
Token was
Token ID373
Feature activation+1.34e+00

Pos logit contributions
ples+0.389
ount+0.357
 Jas+0.357
yl+0.343
 paints+0.343
Neg logit contributions
 tum-0.425
 thirst-0.373
night-0.371
 steak-0.351
 lung-0.341
Token a
Token ID257
Feature activation-1.12e-01

Pos logit contributions
ples+2.539
ount+2.381
 Jas+2.332
 paints+2.286
yl+2.279
Neg logit contributions
 tum-2.029
 thirst-1.765
night-1.701
 steak-1.655
 lung-1.578
Token great
Token ID1049
Feature activation+1.92e-01

Pos logit contributions
 tum+0.190
 thirst+0.166
night+0.166
 steak+0.155
 lung+0.150
Neg logit contributions
ples-0.184
 Jas-0.168
ount-0.167
yl-0.165
 paints-0.163
Token idea
Token ID2126
Feature activation+6.94e-01

Pos logit contributions
ples+0.340
 Jas+0.314
ount+0.314
yl+0.304
 paints+0.303
Neg logit contributions
 tum-0.305
 thirst-0.265
night-0.263
 steak-0.247
 lung-0.239
Token so
Token ID523
Feature activation-1.40e-01

Pos logit contributions
ples+1.128
ount+1.035
 Jas+1.026
 paints+0.997
yl+0.982
Neg logit contributions
 tum-1.208
night-1.059
 thirst-1.047
 steak-1.002
 lung-0.963
Token they
Token ID484
Feature activation+7.09e-02

Pos logit contributions
 tum+0.212
 thirst+0.183
night+0.180
 steak+0.168
 lung+0.164
Neg logit contributions
ples-0.257
 Jas-0.242
ount-0.238
yl-0.234
 Kay-0.232
Token sorry
Token ID7926
Feature activation+2.69e-01

Pos logit contributions
ples+2.491
ount+2.272
 Jas+2.200
 paints+2.195
ptions+2.186
Neg logit contributions
 tum-1.936
 thirst-1.678
night-1.584
 steak-1.525
 lung-1.487
Token,
Token ID11
Feature activation+6.64e-02

Pos logit contributions
ples+0.511
ount+0.474
 Jas+0.473
 paints+0.459
yl+0.459
Neg logit contributions
 tum-0.397
 thirst-0.339
night-0.337
 steak-0.314
 lung-0.303
Token mum
Token ID25682
Feature activation+1.83e+00

Pos logit contributions
ples+0.121
 Jas+0.113
ount+0.112
yl+0.109
 paints+0.109
Neg logit contributions
 tum-0.101
 thirst-0.088
night-0.088
 steak-0.081
 lung-0.078
Token.
Token ID13
Feature activation+2.98e-01

Pos logit contributions
ples+3.443
ount+3.333
 Jas+3.215
ptions+3.197
ydia+3.182
Neg logit contributions
 tum-2.761
night-2.417
 thirst-2.273
 steak-2.216
 lung-2.139
Token I
Token ID314
Feature activation-1.10e-01

Pos logit contributions
ples+0.509
ount+0.470
 Jas+0.469
yl+0.453
 paints+0.452
Neg logit contributions
 tum-0.494
 thirst-0.430
night-0.428
 steak-0.401
 lung-0.390
Token learned
Token ID4499
Feature activation-6.09e-01

Pos logit contributions
 tum+0.194
 thirst+0.171
night+0.169
 steak+0.162
 lung+0.156
Neg logit contributions
ples-0.175
 Jas-0.161
ount-0.161
yl-0.156
 paints-0.155
Token my
Token ID616
Feature activation-2.57e-01

Pos logit contributions
 tum+1.045
 thirst+0.941
night+0.917
 steak+0.833
 lung+0.826
Neg logit contributions
ples-0.967
 Jas-0.921
 paints-0.908
ount-0.894
 Kay-0.888
Token lesson
Token ID11483
Feature activation-8.00e-03

Pos logit contributions
 tum+0.185
 thirst+0.133
night+0.129
 steak+0.103
 lung+0.092
Neg logit contributions
ples-0.672
 Jas-0.643
 paints-0.636
ount-0.636
yl-0.632
Token.
Token ID13
Feature activation+1.23e-01

Pos logit contributions
 tum+0.015
 thirst+0.013
night+0.013
 steak+0.012
 lung+0.012
Neg logit contributions
ples-0.012
 Jas-0.011
ount-0.011
yl-0.011
 paints-0.011
Token I
Token ID314
Feature activation+1.99e-01

Pos logit contributions
ples+0.202
 Jas+0.186
ount+0.185
 paints+0.179
yl+0.179
Neg logit contributions
 tum-0.210
 thirst-0.184
night-0.182
 steak-0.173
 lung-0.167
Token don
Token ID836
Feature activation+3.34e-01

Pos logit contributions
ples+0.336
ount+0.309
 Jas+0.309
yl+0.298
 paints+0.298
Neg logit contributions
 tum-0.335
 thirst-0.292
night-0.291
 steak-0.274
 lung-0.266
Token height
Token ID6001
Feature activation+2.61e-01

Pos logit contributions
ples+0.518
ount+0.478
 Jas+0.474
yl+0.458
 paints+0.457
Neg logit contributions
 tum-0.540
 thirst-0.473
night-0.469
 steak-0.446
 lung-0.433
Token.
Token ID13
Feature activation+9.67e-01

Pos logit contributions
ples+0.442
ount+0.407
 Jas+0.407
yl+0.393
 paints+0.392
Neg logit contributions
 tum-0.433
 thirst-0.378
night-0.375
 steak-0.354
 lung-0.343
Token Tom
Token ID4186
Feature activation+1.04e+00

Pos logit contributions
ples+1.293
ount+1.131
 Jas+1.113
yl+1.109
 paints+1.086
Neg logit contributions
 tum-2.003
 thirst-1.759
night-1.738
 steak-1.692
 lung-1.652
Token's
Token ID338
Feature activation-1.40e+00

Pos logit contributions
ples+1.417
 Jas+1.316
ount+1.304
yl+1.262
ydia+1.252
Neg logit contributions
 tum-2.080
night-1.835
 thirst-1.818
 steak-1.778
 lung-1.745
Token tower
Token ID10580
Feature activation+4.48e-02

Pos logit contributions
 tum+1.931
 thirst+1.830
night+1.820
 steak+1.564
 drowned+1.562
Neg logit contributions
ples-2.512
 paints-2.506
 Jas-2.418
 mirror-2.328
 blo-2.304
Token is
Token ID318
Feature activation+1.56e-01

Pos logit contributions
ples+0.056
ount+0.050
 Jas+0.050
 paints+0.048
yl+0.048
Neg logit contributions
 tum-0.094
 thirst-0.085
night-0.084
 steak-0.080
 lung-0.078
Token big
Token ID1263
Feature activation+5.09e-01

Pos logit contributions
ples+0.258
 Jas+0.237
ount+0.237
 paints+0.229
yl+0.228
Neg logit contributions
 tum-0.267
 thirst-0.234
night-0.233
 steak-0.219
 lung-0.213
Token in
Token ID287
Feature activation-1.78e-01

Pos logit contributions
ples+0.933
ount+0.875
 Jas+0.872
yl+0.844
ydia+0.834
Neg logit contributions
 tum-0.776
night-0.662
 thirst-0.659
 steak-0.628
 lung-0.606
Token width
Token ID9647
Feature activation+2.26e-01

Pos logit contributions
 tum+0.320
night+0.283
 thirst+0.283
 steak+0.260
 lung+0.254
Neg logit contributions
ples-0.282
 Jas-0.260
ount-0.254
 paints-0.250
yl-0.249
Token.
Token ID13
Feature activation+2.16e-01

Pos logit contributions
ples+0.364
ount+0.334
 Jas+0.334
yl+0.322
 paints+0.321
Neg logit contributions
 tum-0.393
 thirst-0.345
night-0.343
 steak-0.324
 lung-0.315
Token You
Token ID921
Feature activation-1.03e-01

Pos logit contributions
ples+0.422
ount+0.393
 Jas+0.393
yl+0.381
 paints+0.381
Neg logit contributions
 tum-0.305
 thirst-0.259
night-0.257
 steak-0.239
 lung-0.230
Token,
Token ID11
Feature activation-3.42e-01

Pos logit contributions
 tum+1.604
 thirst+1.436
night+1.380
uch+1.294
 lung+1.276
Neg logit contributions
ples-1.808
 Jas-1.737
ount-1.663
 Kay-1.659
yl-1.656
Token "
Token ID366
Feature activation-2.56e-01

Pos logit contributions
 tum+0.724
 thirst+0.656
night+0.656
 steak+0.620
 lung+0.607
Neg logit contributions
ples-0.423
 Jas-0.384
ount-0.375
yl-0.362
 paints-0.357
TokenIs
Token ID3792
Feature activation-3.25e-01

Pos logit contributions
 tum+0.569
 thirst+0.514
night+0.511
 steak+0.482
 lung+0.472
Neg logit contributions
ples-0.297
ount-0.266
 Jas-0.265
yl-0.253
 paints-0.251
Token the
Token ID262
Feature activation+1.22e+00

Pos logit contributions
 tum+0.563
 thirst+0.502
night+0.493
 steak+0.456
 lung+0.450
Neg logit contributions
ples-0.516
 Jas-0.483
ount-0.478
 paints-0.470
yl-0.468
Token bad
Token ID2089
Feature activation+1.14e+00

Pos logit contributions
ples+2.083
ount+1.895
 Jas+1.886
 violet+1.752
 paints+1.749
Neg logit contributions
 tum-2.171
night-1.818
 steak-1.811
 thirst-1.808
 lung-1.759
Token man
Token ID582
Feature activation-3.42e-01

Pos logit contributions
ples+2.081
ount+1.911
 Jas+1.898
 paints+1.810
 violet+1.801
Neg logit contributions
 tum-1.869
night-1.597
 steak-1.539
 thirst-1.537
 lung-1.482
Token going
Token ID1016
Feature activation+2.18e-01

Pos logit contributions
 tum+0.606
 thirst+0.541
night+0.532
 steak+0.498
 lung+0.485
Neg logit contributions
ples-0.534
ount-0.493
 Jas-0.486
yl-0.473
 paints-0.473
Token to
Token ID284
Feature activation-8.16e-01

Pos logit contributions
ples+0.395
ount+0.366
 Jas+0.366
yl+0.354
 paints+0.354
Neg logit contributions
 tum-0.338
 thirst-0.291
night-0.290
 steak-0.271
 lung-0.262
Token jail
Token ID7356
Feature activation-5.73e-01

Pos logit contributions
 tum+1.387
night+1.253
 thirst+1.238
 steak+1.134
 lung+1.081
Neg logit contributions
ples-1.294
 Jas-1.229
ount-1.220
 Kay-1.194
 paints-1.183
Token?"
Token ID1701
Feature activation-2.06e+00

Pos logit contributions
 tum+1.043
 thirst+0.938
night+0.909
 steak+0.852
 lung+0.840
Neg logit contributions
ples-0.876
ount-0.812
 Jas-0.810
 paints-0.782
yl-0.778
Token His
Token ID2399
Feature activation-9.74e-01

Pos logit contributions
 tum+3.751
 thirst+3.547
night+3.434
uch+3.377
 lung+3.232
Neg logit contributions
ples-2.855
 Jas-2.842
 Kay-2.800
 Jem-2.681
ount-2.661
Token
Token ID198
Feature activation-2.09e-01

Pos logit contributions
ples+0.556
ount+0.515
 Jas+0.514
 paints+0.500
yl+0.498
Neg logit contributions
 tum-0.439
 thirst-0.375
night-0.372
 steak-0.348
 lung-0.336
TokenHe
Token ID1544
Feature activation-3.29e-01

Pos logit contributions
 tum+0.308
 thirst+0.263
night+0.261
 steak+0.240
 lung+0.232
Neg logit contributions
ples-0.400
 Jas-0.377
ount-0.374
yl-0.365
ydia-0.361
Token asked
Token ID1965
Feature activation-2.26e-01

Pos logit contributions
 tum+0.636
 thirst+0.568
night+0.563
 steak+0.538
 lung+0.524
Neg logit contributions
ples-0.470
 Jas-0.430
ount-0.417
yl-0.416
 paints-0.412
Token a
Token ID257
Feature activation+9.52e-01

Pos logit contributions
 tum+0.269
 thirst+0.225
night+0.221
 steak+0.200
 lung+0.194
Neg logit contributions
ples-0.487
 Jas-0.468
ount-0.458
yl-0.453
 paints-0.448
Token bird
Token ID6512
Feature activation+6.97e-01

Pos logit contributions
ples+1.469
ount+1.350
 Jas+1.315
 paints+1.268
ydia+1.254
Neg logit contributions
 tum-1.788
 thirst-1.547
night-1.537
 steak-1.496
 lung-1.459
Token.
Token ID13
Feature activation-7.06e-01

Pos logit contributions
ples+1.187
 Jas+1.092
ount+1.085
 paints+1.045
yl+1.037
Neg logit contributions
 tum-1.173
 thirst-1.018
night-1.017
 steak-0.957
 lung-0.934
Token The
Token ID383
Feature activation-6.49e-01

Pos logit contributions
 tum+1.108
 thirst+0.984
night+0.982
 steak+0.898
 lung+0.862
Neg logit contributions
ples-1.249
 Jas-1.181
ount-1.159
yl-1.143
 Kay-1.137
Token bird
Token ID6512
Feature activation-4.02e-05

Pos logit contributions
 tum+1.112
night+1.021
 thirst+0.996
 steak+0.914
 lung+0.866
Neg logit contributions
ples-1.006
yl-0.943
 paints-0.943
 Jas-0.935
 blo-0.929
Token said
Token ID531
Feature activation-1.89e-01

Pos logit contributions
 tum+0.000
 thirst+0.000
night+0.000
 steak+0.000
 lung+0.000
Neg logit contributions
ples-0.000
 Jas-0.000
ount-0.000
yl-0.000
 paints-0.000
Token "
Token ID366
Feature activation-3.09e-01

Pos logit contributions
 tum+0.294
 thirst+0.258
night+0.257
 steak+0.236
 lung+0.230
Neg logit contributions
ples-0.334
 Jas-0.311
ount-0.308
yl-0.304
 paints-0.302
TokenThat
Token ID2504
Feature activation-8.28e-01

Pos logit contributions
 tum+0.645
 thirst+0.578
night+0.571
 steak+0.544
 lung+0.527
Neg logit contributions
ples-0.403
ount-0.361
 Jas-0.356
yl-0.349
 paints-0.345
Token sits
Token ID10718
Feature activation+1.84e-01

Pos logit contributions
ples+1.795
ount+1.594
 Jas+1.571
yl+1.553
ydia+1.521
Neg logit contributions
 tum-2.565
night-2.250
 thirst-2.240
 steak-2.154
 lung-2.112
Token next
Token ID1306
Feature activation+7.19e-01

Pos logit contributions
ples+0.298
 Jas+0.273
ount+0.273
yl+0.263
 paints+0.263
Neg logit contributions
 tum-0.321
 thirst-0.282
night-0.280
 steak-0.265
 lung-0.257
Token to
Token ID284
Feature activation+2.46e-01

Pos logit contributions
ples+1.503
ount+1.407
 Jas+1.384
ydia+1.343
 paints+1.342
Neg logit contributions
 tum-0.949
night-0.790
 thirst-0.765
 steak-0.725
 lung-0.685
Token Sam
Token ID3409
Feature activation+3.20e-01

Pos logit contributions
ples+0.405
ount+0.372
 Jas+0.371
yl+0.358
 paints+0.357
Neg logit contributions
 tum-0.427
 thirst-0.374
night-0.372
 steak-0.352
 lung-0.341
Token on
Token ID319
Feature activation-1.90e-01

Pos logit contributions
ples+0.540
 Jas+0.497
ount+0.497
yl+0.480
 paints+0.478
Neg logit contributions
 tum-0.536
 thirst-0.466
night-0.465
 steak-0.438
 lung-0.425
Token the
Token ID262
Feature activation-6.53e-01

Pos logit contributions
 tum+0.341
 thirst+0.304
night+0.300
 steak+0.279
 lung+0.273
Neg logit contributions
ples-0.294
 Jas-0.276
ount-0.266
yl-0.263
 paints-0.263
Token slide
Token ID10649
Feature activation-2.79e-01

Pos logit contributions
 tum+1.234
night+1.150
 thirst+1.138
 steak+1.062
 lung+1.049
Neg logit contributions
ples-0.880
 Jas-0.838
 paints-0.830
ount-0.821
 Kay-0.796
Token.
Token ID13
Feature activation+5.12e-01

Pos logit contributions
 tum+0.516
 thirst+0.465
night+0.458
 steak+0.433
 lung+0.424
Neg logit contributions
ples-0.416
 Jas-0.384
ount-0.382
 paints-0.369
yl-0.367
Token
Token ID198
Feature activation+1.30e-01

Pos logit contributions
ples+0.665
ount+0.592
 Jas+0.583
yl+0.568
 paints+0.558
Neg logit contributions
 tum-1.068
 thirst-0.950
night-0.948
 steak-0.905
 lung-0.881
Token
Token ID198
Feature activation+1.15e-01

Pos logit contributions
ples+0.239
 Jas+0.223
ount+0.222
yl+0.216
 paints+0.215
Neg logit contributions
 tum-0.198
 thirst-0.171
night-0.169
 steak-0.158
 lung-0.153
Token"
Token ID1
Feature activation+1.87e-01

Pos logit contributions
ples+0.201
 Jas+0.186
ount+0.185
yl+0.179
 paints+0.179
Neg logit contributions
 tum-0.189
 thirst-0.164
night-0.163
 steak-0.153
 lung-0.148
Token sad
Token ID6507
Feature activation-7.94e-01

Pos logit contributions
 tum+0.849
 thirst+0.747
night+0.742
 steak+0.716
 lung+0.694
Neg logit contributions
ples-0.718
 Jas-0.664
ount-0.645
 blo-0.635
 paints-0.627
Token.
Token ID13
Feature activation-7.41e-01

Pos logit contributions
 tum+1.531
 thirst+1.391
night+1.336
 steak+1.296
 lung+1.274
Neg logit contributions
ples-1.102
 Jas-1.002
ount-1.001
 blo-0.963
 paints-0.954
Token She
Token ID1375
Feature activation-5.09e-01

Pos logit contributions
 tum+1.253
 thirst+1.135
night+1.123
 steak+1.031
 lung+1.030
Neg logit contributions
ples-1.178
 Jas-1.116
ount-1.108
 Kay-1.083
 paints-1.068
Token played
Token ID2826
Feature activation-7.70e-01

Pos logit contributions
 tum+0.975
 thirst+0.873
night+0.864
 steak+0.832
 lung+0.802
Neg logit contributions
ples-0.720
 Jas-0.656
ount-0.643
 paints-0.637
yl-0.626
Token with
Token ID351
Feature activation-9.67e-01

Pos logit contributions
 tum+1.314
 thirst+1.142
night+1.135
 lung+1.059
 steak+1.041
Neg logit contributions
ples-1.246
 Jas-1.193
ount-1.168
yl-1.133
ydia-1.127
Token him
Token ID683
Feature activation-2.29e+00

Pos logit contributions
 tum+1.615
night+1.430
 thirst+1.416
 lung+1.305
 steak+1.262
Neg logit contributions
ples-1.556
 Jas-1.485
 Kay-1.474
ount-1.447
 blo-1.432
Token in
Token ID287
Feature activation+7.94e-01

Pos logit contributions
 tum+4.348
night+3.880
 thirst+3.851
 current+3.585
bye+3.559
Neg logit contributions
ples-3.158
 blo-3.110
 paints-3.044
yl-2.882
 written-2.843
Token the
Token ID262
Feature activation+2.30e-01

Pos logit contributions
ples+1.466
ount+1.383
 Jas+1.339
yl+1.310
 paints+1.296
Neg logit contributions
 tum-1.238
 thirst-1.059
night-1.034
 steak-1.012
 lung-0.970
Token park
Token ID3952
Feature activation-9.13e-01

Pos logit contributions
ples+0.306
ount+0.275
 Jas+0.275
yl+0.262
 paints+0.262
Neg logit contributions
 tum-0.471
 thirst-0.421
night-0.419
 steak-0.400
 lung-0.391
Token and
Token ID290
Feature activation-2.33e-01

Pos logit contributions
 tum+1.556
 thirst+1.383
night+1.374
 steak+1.279
 lung+1.273
Neg logit contributions
ples-1.476
 Jas-1.362
ount-1.336
 blo-1.324
 paints-1.316
Token at
Token ID379
Feature activation-1.58e-02

Pos logit contributions
 tum+0.411
 thirst+0.364
night+0.362
 steak+0.341
 lung+0.334
Neg logit contributions
ples-0.367
 Jas-0.341
ount-0.336
 paints-0.327
yl-0.327
Token grab
Token ID5552
Feature activation-3.06e-01

Pos logit contributions
 tum+0.608
night+0.546
 thirst+0.543
 steak+0.511
 lung+0.492
Neg logit contributions
ples-0.563
ount-0.521
 Jas-0.513
 paints-0.505
yl-0.499
Token a
Token ID257
Feature activation+8.97e-02

Pos logit contributions
 tum+0.510
night+0.446
 thirst+0.444
 steak+0.408
 lung+0.402
Neg logit contributions
ples-0.515
ount-0.471
 Jas-0.470
yl-0.460
 paints-0.457
Token few
Token ID1178
Feature activation+8.94e-01

Pos logit contributions
ples+0.179
ount+0.167
 Jas+0.166
yl+0.162
 paints+0.162
Neg logit contributions
 tum-0.122
 thirst-0.104
night-0.103
 steak-0.095
 lung-0.091
Token peanuts
Token ID45612
Feature activation-4.75e-01

Pos logit contributions
ples+1.641
 Jas+1.547
ount+1.529
yl+1.474
 Kay+1.465
Neg logit contributions
 tum-1.416
 thirst-1.211
night-1.170
 steak-1.120
 lung-1.059
Token.
Token ID13
Feature activation-2.12e+00

Pos logit contributions
 tum+0.858
 thirst+0.772
night+0.756
 steak+0.701
 lung+0.696
Neg logit contributions
ples-0.722
ount-0.672
 Jas-0.655
 paints-0.644
yl-0.640
Token But
Token ID887
Feature activation-2.90e+00

Pos logit contributions
 tum+3.375
 thirst+3.189
uch+2.975
night+2.923
 drowned+2.734
Neg logit contributions
ples-3.566
 Kay-3.461
 Jas-3.400
ount-3.342
 Jem-3.285
Token the
Token ID262
Feature activation+3.71e-01

Pos logit contributions
 tum+5.683
night+5.136
 thirst+5.080
uch+4.772
bye+4.565
Neg logit contributions
 Kay-3.907
yl-3.774
ples-3.685
 Jem-3.652
 blo-3.619
Token peanut
Token ID26636
Feature activation-7.71e-01

Pos logit contributions
ples+0.679
ount+0.629
 Jas+0.629
yl+0.605
 paints+0.605
Neg logit contributions
 tum-0.583
 thirst-0.500
night-0.492
 steak-0.467
 lung-0.450
Token tree
Token ID5509
Feature activation-1.23e+00

Pos logit contributions
 tum+1.400
night+1.273
 thirst+1.261
 steak+1.164
 lung+1.144
Neg logit contributions
ples-1.134
yl-1.035
 paints-1.024
 Jas-1.022
 blo-1.017
Token was
Token ID373
Feature activation-3.48e-01

Pos logit contributions
 tum+2.235
night+2.068
 thirst+2.022
 steak+1.914
 lung+1.846
Neg logit contributions
ples-1.794
yl-1.656
 Jas-1.621
 paints-1.620
ount-1.609
Token very
Token ID845
Feature activation-7.52e-01

Pos logit contributions
 tum+0.553
night+0.490
 thirst+0.478
 steak+0.445
 lung+0.435
Neg logit contributions
ples-0.610
 Jas-0.559
ount-0.558
 blo-0.548
yl-0.545
Token storm
Token ID6388
Feature activation+2.43e-01

Pos logit contributions
 tum+1.563
 thirst+1.385
night+1.381
 steak+1.252
uch+1.227
Neg logit contributions
ples-1.667
 paints-1.546
 Jas-1.535
 blo-1.533
ount-1.487
Token raged
Token ID49000
Feature activation-1.53e-02

Pos logit contributions
ples+0.400
ount+0.368
 Jas+0.367
yl+0.354
 paints+0.354
Neg logit contributions
 tum-0.415
 thirst-0.364
night-0.362
 steak-0.341
 lung-0.331
Token on
Token ID319
Feature activation-5.25e-01

Pos logit contributions
 tum+0.030
 thirst+0.026
night+0.026
 steak+0.025
 lung+0.024
Neg logit contributions
ples-0.022
 Jas-0.020
ount-0.020
yl-0.019
 paints-0.019
Token.
Token ID13
Feature activation-8.06e-01

Pos logit contributions
 tum+0.876
 thirst+0.767
night+0.755
 steak+0.710
 lung+0.686
Neg logit contributions
ples-0.896
 Jas-0.821
ount-0.803
yl-0.788
 paints-0.785
Token Winds
Token ID22608
Feature activation-1.14e-01

Pos logit contributions
 tum+1.325
night+1.192
 thirst+1.172
 steak+1.115
 lung+1.081
Neg logit contributions
ples-1.356
 Jas-1.291
ount-1.278
 Kay-1.227
 paints-1.199
Token whipped
Token ID30712
Feature activation-2.03e+00

Pos logit contributions
 tum+0.202
 thirst+0.177
night+0.176
 steak+0.167
 lung+0.161
Neg logit contributions
ples-0.182
ount-0.165
 Jas-0.165
 paints-0.159
yl-0.159
Token and
Token ID290
Feature activation-1.21e+00

Pos logit contributions
 tum+3.927
 thirst+3.396
night+3.256
 lung+3.182
 steak+3.141
Neg logit contributions
ples-3.074
 Jas-2.763
 blo-2.706
 paints-2.706
ount-2.700
Token rain
Token ID6290
Feature activation-5.84e-01

Pos logit contributions
 tum+2.113
night+1.863
 thirst+1.848
 steak+1.785
 lung+1.708
Neg logit contributions
ples-1.916
 Jas-1.793
 blo-1.730
 paints-1.730
ount-1.706
Token fell
Token ID3214
Feature activation-7.39e-01

Pos logit contributions
 tum+1.070
 thirst+0.956
night+0.931
 steak+0.922
 lung+0.884
Neg logit contributions
ples-0.897
 Jas-0.812
ount-0.810
yl-0.775
 paints-0.775
Token heavily
Token ID7272
Feature activation+2.75e-02

Pos logit contributions
 tum+1.313
 thirst+1.179
night+1.142
 steak+1.109
 lung+1.053
Neg logit contributions
ples-1.179
 Jas-1.069
ount-1.065
 blo-1.032
 paints-1.031
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
ples+0.044
ount+0.040
 Jas+0.040
yl+0.038
 paints+0.038
Neg logit contributions
 tum-0.049
 thirst-0.043
night-0.043
 steak-0.040
 lung-0.039
Token that
Token ID326
Feature activation-1.37e+00

Pos logit contributions
 tum+1.460
 thirst+1.282
night+1.273
 steak+1.206
 lung+1.181
Neg logit contributions
ples-1.341
 Jas-1.216
 blo-1.208
ount-1.187
yl-1.170
Token they
Token ID484
Feature activation-1.18e+00

Pos logit contributions
 tum+2.137
 thirst+1.909
night+1.875
 steak+1.753
 hind+1.694
Neg logit contributions
ples-2.377
 Jas-2.277
 blo-2.195
 Kay-2.171
 paints-2.156
Token both
Token ID1111
Feature activation-1.29e+00

Pos logit contributions
 tum+2.168
night+1.897
 thirst+1.888
 steak+1.872
 lung+1.782
Neg logit contributions
ples-1.753
 Jas-1.629
ount-1.584
 blo-1.578
yl-1.544
Token slept
Token ID21256
Feature activation-1.88e+00

Pos logit contributions
 tum+2.407
 steak+2.134
 thirst+2.134
night+2.104
 lung+2.017
Neg logit contributions
ples-1.869
 Jas-1.692
 blo-1.648
ount-1.623
 paints-1.614
Token for
Token ID329
Feature activation-1.52e+00

Pos logit contributions
 tum+3.282
 thirst+2.755
night+2.680
 steak+2.659
 lung+2.576
Neg logit contributions
ples-3.144
 blo-2.877
 Jas-2.873
yl-2.756
ount-2.732
Token a
Token ID257
Feature activation-2.24e+00

Pos logit contributions
 tum+2.576
night+2.210
 thirst+2.145
 steak+2.031
 current+2.025
Neg logit contributions
ples-2.556
 Jas-2.373
 blo-2.303
yl-2.260
 Kay-2.252
Token long
Token ID890
Feature activation-1.71e+00

Pos logit contributions
 tum+3.067
 thirst+2.753
night+2.735
 steak+2.511
edly+2.493
Neg logit contributions
ples-4.003
 blo-3.998
yl-3.961
 violet-3.750
 Kay-3.719
Token,
Token ID11
Feature activation-9.84e-01

Pos logit contributions
 tum+2.461
 thirst+2.191
night+2.118
 steak+2.017
 lung+1.979
Neg logit contributions
ples-3.083
 blo-2.853
 Jas-2.851
yl-2.816
 paints-2.789
Token long
Token ID890
Feature activation-1.16e+00

Pos logit contributions
 tum+1.613
night+1.424
 thirst+1.404
 steak+1.315
 lung+1.291
Neg logit contributions
ples-1.646
 Jas-1.532
 blo-1.504
 paints-1.494
ount-1.472
Token time
Token ID640
Feature activation-5.63e-01

Pos logit contributions
 tum+1.592
 thirst+1.386
night+1.374
 steak+1.270
 lung+1.217
Neg logit contributions
ples-2.230
 Jas-2.052
 blo-2.049
 paints-2.041
yl-1.997
Token and
Token ID290
Feature activation-1.70e+00

Pos logit contributions
 tum+0.988
 thirst+0.870
night+0.864
 steak+0.820
 lung+0.794
Neg logit contributions
ples-0.907
 Jas-0.831
ount-0.812
 blo-0.803
 paints-0.796
Token day
Token ID1110
Feature activation-1.36e+00

Pos logit contributions
 tum+3.116
 thirst+2.903
night+2.775
edly+2.597
 lung+2.391
Neg logit contributions
 blo-3.619
 paints-3.611
ples-3.541
yl-3.516
 Jas-3.464
Token,
Token ID11
Feature activation-2.48e+00

Pos logit contributions
 tum+2.322
 thirst+2.198
night+2.096
 steak+1.977
 lung+1.926
Neg logit contributions
ples-2.157
 Jas-2.081
ount-1.989
yl-1.974
 Kay-1.959
Token Max
Token ID5436
Feature activation-1.30e+00

Pos logit contributions
 tum+4.597
 thirst+4.400
night+4.268
edly+4.037
bye+3.916
Neg logit contributions
 Jas-3.626
ples-3.600
 Melissa-3.502
 typing-3.488
 paints-3.403
Token was
Token ID373
Feature activation-1.28e+00

Pos logit contributions
 tum+2.391
 thirst+2.191
night+2.117
 steak+2.021
time+1.920
Neg logit contributions
ples-1.914
 Jas-1.725
 paints-1.706
yl-1.670
 blo-1.659
Token tired
Token ID10032
Feature activation-1.59e+00

Pos logit contributions
 tum+2.211
night+2.065
 thirst+2.002
 steak+1.908
bye+1.842
Neg logit contributions
ples-1.927
 Jas-1.821
 blo-1.785
ount-1.762
 drawn-1.749
Token but
Token ID475
Feature activation-2.24e+00

Pos logit contributions
 tum+2.843
 thirst+2.735
night+2.645
 steak+2.458
uch+2.443
Neg logit contributions
ples-2.235
 Jas-2.100
ount-2.085
 blo-2.052
yl-2.001
Token happy
Token ID3772
Feature activation-1.31e+00

Pos logit contributions
 tum+4.320
night+4.051
 thirst+3.926
uch+3.777
time+3.687
Neg logit contributions
 Jas-2.689
ples-2.681
 blo-2.674
 written-2.672
 drawn-2.619
Token.
Token ID13
Feature activation-1.71e+00

Pos logit contributions
 tum+2.571
 thirst+2.409
night+2.268
uch+2.177
 steak+2.175
Neg logit contributions
ples-1.700
 Jas-1.640
yl-1.545
 written-1.543
 blo-1.528
Token He
Token ID679
Feature activation-1.98e+00

Pos logit contributions
 tum+2.552
 thirst+2.532
night+2.206
uch+2.168
 steak+2.135
Neg logit contributions
ples-2.975
 Jas-2.870
ydia-2.722
 paints-2.713
ount-2.691
Token went
Token ID1816
Feature activation-2.07e-01

Pos logit contributions
 tum+4.090
 thirst+3.719
night+3.594
time+3.547
 steak+3.505
Neg logit contributions
ples-2.495
 blo-2.226
 paints-2.211
 written-2.177
 Jas-2.163
Token home
Token ID1363
Feature activation-1.67e+00

Pos logit contributions
 tum+0.309
 thirst+0.269
night+0.261
 steak+0.244
 lung+0.238
Neg logit contributions
ples-0.381
ount-0.358
 Jas-0.357
yl-0.349
 paints-0.348