TOP ACTIVATIONS
MAX = 3.857

The kind stranger explained that cl apping can help during emergencies
The old man gr umbled and shook his head
he could help . Per c ival took out a pair
boy named Tim found the sad golf ball . Tim was
was so happy ! He gle efully filled his bowl with
The merchant heard the cl apping and saw little Abby
Mom my saw the excited gle am in his eye .
, " What are you playing with , Tim ?" Tim
get sick ." The mole gr umbled but eventually went to
said goodbye . The person gr umbled a small thank you
. They were happy to be pirates and sharks , and
came back inside . Spot gr umbled , and ran off
!" The cat gr ump ily me owed and
my explained to her that cl apping is a way to
mess . They learned that decor ating was fun , but
M olly heard some cl apping and went to investigate
, Tim my learned to cl ap for himself and not
with a club . The big boy was hitting trees with
L ily and Ben cl apped their hands and gave
boy , " What 's that ?" "

MIN ACTIVATIONS
MIN = -4.086

the chair and reaches for the top books . She grabs
in this seat and take a long nap . â
, so it might take a few times for it to
slippery and she couldn 't get to the top . So
. Mom my and Daddy then told her that
shelf . He climbed up the chair , and he could
rolled down the hill . When it was time to go
sit down and rest for a bit ." L
a good place to take a nap . As
to jump . One day , Megan went to the park
excited and quickly scrambled up the chair .
he lf . She pushes the chair to the books he
the covers , ready for a cos y sleep .
's sit down and take a break ." They found a
so happy that he reached the popcorn . But then ,
now tall enough to reach the mixer . He smiled and
on the chair and grab the book . They are happy
top of the hill . When they got there they saw
and pretended for hours . When it was time to go
chair so she could reach the onion on the top shelf

INTERVAL 1.871 - 3.857
CONTAINS 0.636%

of her for making a new friend and not feeling guilty
Then , one of the boys saw a shiny ladder
But his mum knew how to comfort him . She hugged
He saw a little boy playing with a huge , pink
many colors . Tom likes the shiny pen a lot .

INTERVAL -0.114 - 1.871
CONTAINS 55.382%

all become kind and take care of nature , it would
finally she pushed the bedroom door open , she sighed in
were happy and smiled . The sun smiled too . <|endoftext|>
he found a friend who also liked empty apples . <|endoftext|>
up with you ." L ily didn 't care

INTERVAL -2.100 - -0.114
CONTAINS 43.713%

dress that her mom my bought for her . It was
mud and placed it on a nearby tree branch . The
but he did not want to be the only one to
?" Tom and Mia look at Mom . They
banner that said , " We are family ." They hung

INTERVAL -4.086 - -2.100
CONTAINS 0.269%

. The End . <|endoftext|> Once upon a
but to also stay safe . E lla
, I want to pick a
He took his truck for a drive and it felt so
! Let 's go for a ride . â
TokenThe
Token ID464
Feature activation+8.34e-01

Pos logit contributions
 syrup+1.534
 kil+1.319
 marsh+1.316
Pen+1.308
 Pen+1.288
Neg logit contributions
ate-2.230
amed-2.139
 army-2.076
ating-2.013
room-2.004
Token kind
Token ID1611
Feature activation+1.50e+00

Pos logit contributions
ate+1.644
 army+1.544
roc+1.525
amed+1.505
ating+1.467
Neg logit contributions
 syrup-1.255
Pen-1.189
 Pen-1.145
 marsh-1.108
 Frost-1.072
Token stranger
Token ID16195
Feature activation-2.64e-01

Pos logit contributions
ate+3.131
 army+2.991
amed+2.954
roc+2.949
ating+2.858
Neg logit contributions
 syrup-2.019
 Pen-1.921
Pen-1.918
 Megan-1.808
 Lu-1.803
Token explained
Token ID4893
Feature activation-7.94e-02

Pos logit contributions
 syrup+0.417
Pen+0.396
 Pen+0.373
 marsh+0.370
 kil+0.362
Neg logit contributions
ate-0.488
 army-0.470
amed-0.450
roc-0.443
ating-0.433
Token that
Token ID326
Feature activation-1.91e-01

Pos logit contributions
 syrup+0.128
Pen+0.122
 Pen+0.115
 marsh+0.113
 kil+0.111
Neg logit contributions
ate-0.145
 army-0.139
amed-0.132
roc-0.132
ating-0.128
Token cl
Token ID537
Feature activation+3.86e+00

Pos logit contributions
 syrup+0.250
Pen+0.234
 Pen+0.218
 marsh+0.214
 kil+0.210
Neg logit contributions
ate-0.405
 army-0.391
amed-0.373
roc-0.371
ating-0.365
Tokenapping
Token ID5912
Feature activation+1.88e+00

Pos logit contributions
 army+5.323
ate+5.034
 alien+4.740
ully+4.558
roc+4.534
Neg logit contributions
 syrup-7.866
 kil-7.730
 Megan-7.415
Pen-7.303
 tug-7.202
Token can
Token ID460
Feature activation+9.17e-01

Pos logit contributions
ate+4.059
 army+3.863
roc+3.730
amed+3.704
ating+3.563
Neg logit contributions
 syrup-2.516
Pen-2.263
 Pen-2.241
 might-2.193
 marsh-2.165
Token help
Token ID1037
Feature activation+4.19e-01

Pos logit contributions
ate+1.887
 army+1.843
amed+1.776
roc+1.764
ating+1.702
Neg logit contributions
 syrup-1.259
Pen-1.173
 Pen-1.115
 marsh-1.091
 Frost-1.055
Token during
Token ID1141
Feature activation+4.33e-01

Pos logit contributions
ate+0.807
 army+0.776
amed+0.747
roc+0.736
ating+0.717
Neg logit contributions
 syrup-0.634
Pen-0.602
 Pen-0.577
 marsh-0.560
 Willow-0.544
Token emergencies
Token ID36047
Feature activation+1.35e+00

Pos logit contributions
ate+0.839
 army+0.801
amed+0.778
roc+0.770
ating+0.748
Neg logit contributions
 syrup-0.654
Pen-0.613
 Pen-0.589
 marsh-0.578
 Frost-0.558
Token
Token ID198
Feature activation+1.14e-01

Pos logit contributions
 syrup+0.209
Pen+0.196
 Pen+0.185
 marsh+0.183
 kil+0.179
Neg logit contributions
ate-0.255
 army-0.244
amed-0.232
roc-0.231
ating-0.225
Token
Token ID198
Feature activation+5.99e-01

Pos logit contributions
ate+0.215
 army+0.206
amed+0.197
roc+0.195
ating+0.191
Neg logit contributions
 syrup-0.180
Pen-0.169
 Pen-0.161
 marsh-0.158
 kil-0.154
TokenThe
Token ID464
Feature activation+1.12e+00

Pos logit contributions
ate+1.017
 army+0.978
amed+0.931
roc+0.920
ating+0.898
Neg logit contributions
 syrup-1.026
Pen-1.002
 Pen-0.959
 marsh-0.932
 Frost-0.915
Token old
Token ID1468
Feature activation+1.41e+00

Pos logit contributions
ate+1.763
amed+1.622
 army+1.609
ating+1.566
roc+1.546
Neg logit contributions
 syrup-2.138
Pen-1.998
 Pen-1.954
 marsh-1.912
 Frost-1.884
Token man
Token ID582
Feature activation+9.08e-01

Pos logit contributions
ate+2.340
amed+2.240
 army+2.222
roc+2.126
ating+2.106
Neg logit contributions
 syrup-2.520
 Pen-2.331
Pen-2.326
 Frost-2.233
 marsh-2.212
Token gr
Token ID1036
Feature activation+3.77e+00

Pos logit contributions
ate+1.755
 army+1.691
amed+1.615
roc+1.612
ating+1.567
Neg logit contributions
 syrup-1.352
Pen-1.274
 Pen-1.243
 Frost-1.191
 Willow-1.177
Tokenumbled
Token ID11137
Feature activation+1.36e+00

Pos logit contributions
 army+5.826
ate+5.220
 alien+4.984
roc+4.748
 friendships+4.678
Neg logit contributions
 syrup-7.040
Pen-7.002
 kil-6.938
 Pen-6.770
 Lu-6.768
Token and
Token ID290
Feature activation+4.57e-01

Pos logit contributions
ate+2.710
 army+2.545
roc+2.527
amed+2.472
ating+2.407
Neg logit contributions
 syrup-1.989
Pen-1.829
 Pen-1.826
 Frost-1.746
 marsh-1.744
Token shook
Token ID14682
Feature activation+9.17e-01

Pos logit contributions
ate+0.926
 army+0.888
amed+0.858
roc+0.851
ating+0.831
Neg logit contributions
 syrup-0.646
Pen-0.606
 Pen-0.583
 marsh-0.564
 Frost-0.550
Token his
Token ID465
Feature activation+8.57e-01

Pos logit contributions
ate+1.599
 army+1.517
amed+1.474
roc+1.453
ating+1.414
Neg logit contributions
 syrup-1.554
Pen-1.477
 Pen-1.453
 marsh-1.404
 Frost-1.397
Token head
Token ID1182
Feature activation+2.00e-01

Pos logit contributions
ate+1.391
 army+1.292
amed+1.260
roc+1.235
ating+1.230
Neg logit contributions
 syrup-1.579
Pen-1.496
 Pen-1.465
 marsh-1.418
 kil-1.417
Token he
Token ID339
Feature activation-7.84e-01

Pos logit contributions
ate+0.469
 army+0.454
amed+0.439
roc+0.435
ating+0.427
Neg logit contributions
 syrup-0.223
Pen-0.206
 Pen-0.193
 marsh-0.187
 kil-0.179
Token could
Token ID714
Feature activation+5.11e-01

Pos logit contributions
 syrup+0.867
Pen+0.805
 marsh+0.745
 Pen+0.728
 Willow+0.662
Neg logit contributions
ate-1.872
 army-1.820
amed-1.750
roc-1.749
ating-1.696
Token help
Token ID1037
Feature activation-8.29e-02

Pos logit contributions
ate+0.950
 army+0.913
amed+0.888
roc+0.876
ating+0.850
Neg logit contributions
 syrup-0.797
Pen-0.753
 Pen-0.724
 marsh-0.709
 Frost-0.692
Token.
Token ID13
Feature activation+1.48e+00

Pos logit contributions
 syrup+0.130
Pen+0.124
 Pen+0.117
 marsh+0.115
 kil+0.111
Neg logit contributions
ate-0.156
 army-0.150
roc-0.142
amed-0.142
ating-0.139
Token Per
Token ID2448
Feature activation+3.34e-01

Pos logit contributions
ate+2.769
amed+2.658
 army+2.650
ating+2.536
roc+2.514
Neg logit contributions
 syrup-2.258
Pen-2.141
 Pen-2.132
 Willow-2.062
 Frost-2.048
Tokenc
Token ID66
Feature activation+3.74e+00

Pos logit contributions
ate+0.597
 army+0.573
amed+0.547
roc+0.541
ating+0.526
Neg logit contributions
 syrup-0.551
Pen-0.524
 Pen-0.502
 marsh-0.489
 kil-0.481
Tokenival
Token ID2473
Feature activation+1.17e+00

Pos logit contributions
ate+7.375
 army+7.308
room+6.683
 alien+6.497
amed+6.490
Neg logit contributions
 might-5.643
 almost-5.267
 nearly-5.224
 syrup-5.009
 Frost-4.769
Token took
Token ID1718
Feature activation+1.75e-01

Pos logit contributions
ate+2.176
 army+2.103
amed+2.035
roc+2.000
ating+1.910
Neg logit contributions
 syrup-1.812
Pen-1.697
 Pen-1.692
 Frost-1.638
 Willow-1.617
Token out
Token ID503
Feature activation+5.44e-01

Pos logit contributions
ate+0.328
 army+0.315
amed+0.302
roc+0.299
ating+0.292
Neg logit contributions
 syrup-0.273
Pen-0.258
 Pen-0.247
 marsh-0.241
 kil-0.235
Token a
Token ID257
Feature activation-6.09e-01

Pos logit contributions
ate+0.993
 army+0.945
amed+0.914
roc+0.903
ating+0.883
Neg logit contributions
 syrup-0.886
Pen-0.829
 Pen-0.802
 marsh-0.790
 Frost-0.771
Token pair
Token ID5166
Feature activation+2.85e-01

Pos logit contributions
 syrup+1.055
Pen+1.015
 Pen+0.973
 marsh+0.931
 Frost+0.926
Neg logit contributions
 army-1.020
ate-1.017
roc-0.928
amed-0.926
ating-0.881
Token boy
Token ID2933
Feature activation+1.08e+00

Pos logit contributions
ate+5.090
amed+4.970
ating+4.903
ined+4.584
 army+4.532
Neg logit contributions
 syrup-3.877
 Pen-3.610
Pen-3.482
 Frost-3.468
 Lu-3.394
Token named
Token ID3706
Feature activation+1.07e+00

Pos logit contributions
ate+1.898
 army+1.773
amed+1.753
roc+1.732
ating+1.669
Neg logit contributions
 syrup-1.797
Pen-1.694
 Pen-1.667
 marsh-1.606
 Frost-1.600
Token Tim
Token ID5045
Feature activation+1.15e+00

Pos logit contributions
ate+2.025
 army+1.940
amed+1.933
ating+1.848
roc+1.818
Neg logit contributions
 syrup-1.632
Pen-1.575
 Pen-1.562
 Frost-1.514
 marsh-1.471
Token found
Token ID1043
Feature activation+1.64e+00

Pos logit contributions
ate+2.299
 army+2.193
roc+2.134
amed+2.129
ating+2.071
Neg logit contributions
 syrup-1.618
Pen-1.501
 Pen-1.466
 Frost-1.389
 marsh-1.386
Token the
Token ID262
Feature activation+2.66e+00

Pos logit contributions
ate+3.560
 army+3.455
amed+3.351
roc+3.344
ating+3.308
Neg logit contributions
 syrup-2.053
 Frost-1.905
Pen-1.865
 Pen-1.853
 marsh-1.786
Token sad
Token ID6507
Feature activation+3.73e+00

Pos logit contributions
ate+5.193
ating+5.030
 army+4.886
amed+4.882
isdom+4.802
Neg logit contributions
 syrup-4.106
 might-3.807
 tug-3.519
 almost-3.448
 bit-3.430
Token golf
Token ID13126
Feature activation+2.24e+00

Pos logit contributions
ate+8.075
amed+7.985
ating+7.965
ined+7.351
 army+7.332
Neg logit contributions
 syrup-5.599
 bit-5.062
 tug-4.993
 sled-4.813
 might-4.743
Token ball
Token ID2613
Feature activation+9.07e-02

Pos logit contributions
ate+4.185
roc+3.923
 army+3.832
amed+3.805
isdom+3.745
Neg logit contributions
 syrup-3.518
 might-3.280
Pen-3.071
 nib-3.055
 tug-3.049
Token.
Token ID13
Feature activation+9.85e-01

Pos logit contributions
ate+0.183
 army+0.177
amed+0.170
roc+0.168
ating+0.165
Neg logit contributions
 syrup-0.128
Pen-0.121
 Pen-0.115
 marsh-0.112
 kil-0.109
Token Tim
Token ID5045
Feature activation+1.43e+00

Pos logit contributions
ate+1.770
 army+1.675
amed+1.652
ating+1.583
roc+1.577
Neg logit contributions
 syrup-1.640
Pen-1.545
 Pen-1.520
 marsh-1.472
 Frost-1.460
Token was
Token ID373
Feature activation+2.19e+00

Pos logit contributions
ate+3.072
 army+2.965
amed+2.842
roc+2.828
ating+2.816
Neg logit contributions
 syrup-1.824
Pen-1.669
 Pen-1.644
 might-1.590
 Frost-1.558
Token was
Token ID373
Feature activation+2.86e-01

Pos logit contributions
 syrup+0.214
Pen+0.204
 Pen+0.194
 marsh+0.190
 kil+0.186
Neg logit contributions
ate-0.235
 army-0.224
amed-0.213
roc-0.211
ating-0.207
Token so
Token ID523
Feature activation+5.28e-01

Pos logit contributions
ate+0.564
 army+0.542
amed+0.520
roc+0.518
ating+0.505
Neg logit contributions
 syrup-0.422
Pen-0.398
 Pen-0.380
 marsh-0.370
 Frost-0.360
Token happy
Token ID3772
Feature activation-2.96e-01

Pos logit contributions
ate+0.903
 army+0.868
amed+0.824
roc+0.824
ating+0.801
Neg logit contributions
 syrup-0.920
Pen-0.878
 Pen-0.846
 marsh-0.829
 Frost-0.813
Token!
Token ID0
Feature activation-3.42e-01

Pos logit contributions
 syrup+0.482
Pen+0.463
 Pen+0.435
 marsh+0.430
 kil+0.417
Neg logit contributions
ate-0.546
 army-0.522
amed-0.490
roc-0.487
ating-0.483
Token He
Token ID679
Feature activation-5.81e-01

Pos logit contributions
 syrup+0.620
Pen+0.584
 Pen+0.556
 marsh+0.554
 kil+0.542
Neg logit contributions
ate-0.564
 army-0.533
amed-0.502
roc-0.499
ating-0.478
Token gle
Token ID26852
Feature activation+3.69e+00

Pos logit contributions
 syrup+0.859
Pen+0.829
 Pen+0.777
 marsh+0.759
 kil+0.753
Neg logit contributions
ate-1.134
 army-1.092
amed-1.050
roc-1.014
ating-0.997
Tokenefully
Token ID7549
Feature activation-1.75e-01

Pos logit contributions
 army+6.204
 disl+5.999
ully+5.556
isdom+5.457
 alien+5.388
Neg logit contributions
Pen-6.394
 syrup-6.390
 Frost-6.230
 kil-6.172
 Megan-6.079
Token filled
Token ID5901
Feature activation+1.34e+00

Pos logit contributions
 syrup+0.246
Pen+0.236
 Pen+0.222
 marsh+0.215
 kil+0.210
Neg logit contributions
ate-0.355
 army-0.343
amed-0.331
roc-0.323
ating-0.318
Token his
Token ID465
Feature activation+6.63e-01

Pos logit contributions
 army+2.294
ate+2.278
roc+2.174
amed+2.164
ully+2.056
Neg logit contributions
 syrup-2.277
Pen-2.127
 Pen-2.094
 Frost-2.059
 marsh-2.022
Token bowl
Token ID9396
Feature activation+5.56e-01

Pos logit contributions
ate+1.305
 army+1.246
amed+1.211
roc+1.199
ating+1.172
Neg logit contributions
 syrup-0.988
Pen-0.914
 Pen-0.879
 marsh-0.876
 Frost-0.844
Token with
Token ID351
Feature activation-1.82e-01

Pos logit contributions
ate+0.876
 army+0.841
roc+0.803
amed+0.790
ating+0.761
Neg logit contributions
 syrup-1.033
Pen-0.973
 Pen-0.942
 marsh-0.930
 Frost-0.908
Token
Token ID198
Feature activation-4.11e-02

Pos logit contributions
 syrup+0.369
Pen+0.340
 marsh+0.325
 Pen+0.324
 kil+0.321
Neg logit contributions
ate-0.409
 army-0.382
amed-0.372
roc-0.363
ating-0.356
TokenThe
Token ID464
Feature activation+1.11e+00

Pos logit contributions
 syrup+0.062
Pen+0.058
 Pen+0.055
 marsh+0.054
 kil+0.053
Neg logit contributions
ate-0.080
 army-0.076
amed-0.074
roc-0.073
ating-0.071
Token merchant
Token ID20729
Feature activation+1.72e+00

Pos logit contributions
ate+2.021
roc+1.866
 army+1.859
amed+1.847
ating+1.764
Neg logit contributions
 syrup-1.902
Pen-1.758
 marsh-1.728
 Pen-1.686
 Frost-1.610
Token heard
Token ID2982
Feature activation+2.05e-01

Pos logit contributions
ate+3.513
 army+3.340
roc+3.280
amed+3.235
ating+3.114
Neg logit contributions
 syrup-2.393
 Pen-2.194
Pen-2.192
 Frost-2.070
 marsh-2.062
Token the
Token ID262
Feature activation+1.09e+00

Pos logit contributions
ate+0.397
 army+0.381
amed+0.366
roc+0.363
ating+0.354
Neg logit contributions
 syrup-0.308
Pen-0.291
 Pen-0.277
 marsh-0.271
 kil-0.263
Token cl
Token ID537
Feature activation+3.69e+00

Pos logit contributions
ate+1.737
 army+1.550
amed+1.543
roc+1.499
ating+1.474
Neg logit contributions
 syrup-2.132
Pen-1.975
 marsh-1.927
 Pen-1.914
 kil-1.858
Tokenapping
Token ID5912
Feature activation+1.51e+00

Pos logit contributions
ate+3.982
 army+3.951
roc+3.643
 disl+3.355
amed+3.244
Neg logit contributions
 syrup-8.531
 kil-8.063
 Megan-7.928
Pen-7.922
abel-7.884
Token and
Token ID290
Feature activation+4.95e-01

Pos logit contributions
ate+2.880
 army+2.628
roc+2.621
amed+2.612
ating+2.487
Neg logit contributions
 syrup-2.464
Pen-2.204
 Pen-2.191
 marsh-2.158
 Willow-2.067
Token saw
Token ID2497
Feature activation+5.23e-01

Pos logit contributions
ate+1.022
 army+0.984
amed+0.944
roc+0.941
ating+0.918
Neg logit contributions
 syrup-0.686
Pen-0.643
 Pen-0.618
 marsh-0.598
 Frost-0.584
Token little
Token ID1310
Feature activation+3.45e-01

Pos logit contributions
ate+1.053
 army+1.025
amed+0.981
roc+0.973
ating+0.947
Neg logit contributions
 syrup-0.747
Pen-0.707
 Pen-0.680
 marsh-0.661
 Frost-0.640
Token Abby
Token ID49220
Feature activation-5.71e-01

Pos logit contributions
ate+0.580
 army+0.551
amed+0.529
roc+0.519
ating+0.507
Neg logit contributions
 syrup-0.609
Pen-0.577
 Pen-0.557
 marsh-0.546
 kil-0.533
TokenMom
Token ID29252
Feature activation+4.74e-01

Pos logit contributions
ate+0.157
 army+0.149
amed+0.143
roc+0.141
ating+0.137
Neg logit contributions
 syrup-0.161
Pen-0.153
 Pen-0.147
 marsh-0.144
 kil-0.141
Tokenmy
Token ID1820
Feature activation+1.21e+00

Pos logit contributions
ate+0.774
 army+0.743
roc+0.699
amed+0.698
ating+0.675
Neg logit contributions
 syrup-0.857
Pen-0.810
 Pen-0.781
 marsh-0.764
 Frost-0.750
Token saw
Token ID2497
Feature activation+9.44e-01

Pos logit contributions
ate+2.519
 army+2.435
roc+2.355
amed+2.303
ating+2.276
Neg logit contributions
 syrup-1.656
Pen-1.530
 Pen-1.503
 Frost-1.443
 marsh-1.421
Token the
Token ID262
Feature activation+1.06e-01

Pos logit contributions
ate+1.921
 army+1.878
amed+1.821
roc+1.801
ating+1.745
Neg logit contributions
 syrup-1.359
Pen-1.213
 Pen-1.182
 marsh-1.172
 Frost-1.149
Token excited
Token ID6568
Feature activation+1.14e+00

Pos logit contributions
ate+0.165
 army+0.158
amed+0.149
roc+0.147
ating+0.142
Neg logit contributions
 syrup-0.203
Pen-0.194
 Pen-0.187
 marsh-0.183
 kil-0.179
Token gle
Token ID26852
Feature activation+3.68e+00

Pos logit contributions
ate+1.761
amed+1.632
 army+1.606
roc+1.561
ating+1.527
Neg logit contributions
 syrup-2.149
Pen-2.014
 Pen-1.982
 marsh-1.944
 kil-1.911
Tokenam
Token ID321
Feature activation+1.73e+00

Pos logit contributions
 army+4.799
 alien+4.606
 disl+4.404
ully+4.102
isdom+4.046
Neg logit contributions
 kil-7.474
Pen-7.470
 syrup-7.420
 Megan-7.348
 Pen-7.073
Token in
Token ID287
Feature activation-4.80e-01

Pos logit contributions
ate+3.375
 army+3.212
roc+3.126
amed+3.067
ating+2.978
Neg logit contributions
 syrup-2.519
Pen-2.354
 Pen-2.342
 nearly-2.234
 marsh-2.216
Token his
Token ID465
Feature activation+6.40e-01

Pos logit contributions
 syrup+0.840
Pen+0.817
 Pen+0.767
 marsh+0.755
 kil+0.747
Neg logit contributions
ate-0.807
 army-0.781
amed-0.731
roc-0.718
room-0.702
Token eye
Token ID4151
Feature activation+9.75e-01

Pos logit contributions
ate+1.233
 army+1.158
amed+1.141
roc+1.123
ating+1.103
Neg logit contributions
 syrup-0.983
Pen-0.928
 Pen-0.903
 marsh-0.881
 Frost-0.861
Token.
Token ID13
Feature activation+6.17e-01

Pos logit contributions
ate+1.824
 army+1.731
roc+1.706
amed+1.690
ating+1.604
Neg logit contributions
 syrup-1.561
Pen-1.422
 Pen-1.398
 marsh-1.376
 kil-1.333
Token,
Token ID11
Feature activation-8.86e-02

Pos logit contributions
ate+2.554
 army+2.446
amed+2.411
ating+2.266
roc+2.266
Neg logit contributions
 syrup-2.336
Pen-2.207
 Pen-2.202
 Frost-2.139
 marsh-2.116
Token "
Token ID366
Feature activation-1.74e-02

Pos logit contributions
 syrup+0.157
Pen+0.149
 Pen+0.142
 marsh+0.140
 kil+0.137
Neg logit contributions
ate-0.148
 army-0.142
amed-0.135
roc-0.133
ating-0.129
TokenWhat
Token ID2061
Feature activation+1.35e+00

Pos logit contributions
 syrup+0.025
Pen+0.023
 Pen+0.022
 marsh+0.022
 kil+0.021
Neg logit contributions
ate-0.035
 army-0.034
amed-0.033
roc-0.033
ating-0.032
Token are
Token ID389
Feature activation+3.06e-01

Pos logit contributions
ate+2.709
amed+2.617
 army+2.590
roc+2.500
ating+2.480
Neg logit contributions
 syrup-1.891
Pen-1.773
 Pen-1.734
 Frost-1.671
 might-1.665
Token you
Token ID345
Feature activation+3.09e+00

Pos logit contributions
ate+0.624
 army+0.599
amed+0.580
roc+0.571
ating+0.559
Neg logit contributions
 syrup-0.433
Pen-0.407
 Pen-0.388
 marsh-0.378
 Frost-0.367
Token playing
Token ID2712
Feature activation+3.65e+00

Pos logit contributions
ate+5.892
roc+5.508
ating+5.505
amed+5.428
 army+5.410
Neg logit contributions
 syrup-4.975
 almost-4.778
 nearly-4.586
 nib-4.457
 Pen-4.416
Token with
Token ID351
Feature activation+2.24e+00

Pos logit contributions
amed+7.450
roc+6.983
ate+6.969
ully+6.964
 disl+6.962
Neg logit contributions
 syrup-5.172
 Frost-4.936
 nib-4.899
 Will-4.855
 almost-4.824
Token,
Token ID11
Feature activation+9.06e-01

Pos logit contributions
amed+4.096
ate+4.090
 army+3.821
roc+3.806
ating+3.793
Neg logit contributions
 syrup-3.687
 marsh-3.273
 Frost-3.260
Pen-3.233
 Pen-3.156
Token Tim
Token ID5045
Feature activation+1.72e+00

Pos logit contributions
ate+1.621
 army+1.548
amed+1.526
ating+1.476
roc+1.447
Neg logit contributions
 syrup-1.509
Pen-1.410
 Pen-1.369
 marsh-1.346
 Frost-1.333
Token?"
Token ID1701
Feature activation+1.44e+00

Pos logit contributions
ate+2.779
amed+2.724
 army+2.675
roc+2.546
ating+2.466
Neg logit contributions
 syrup-3.093
Pen-2.884
 Pen-2.817
 Frost-2.718
 kil-2.679
Token Tim
Token ID5045
Feature activation+1.05e+00

Pos logit contributions
ate+2.524
 army+2.486
amed+2.410
roc+2.317
ating+2.309
Neg logit contributions
 syrup-2.447
Pen-2.311
 Pen-2.293
 Lu-2.213
 Frost-2.198
Token get
Token ID651
Feature activation-9.41e-01

Pos logit contributions
 syrup+0.576
Pen+0.553
 Pen+0.522
 kil+0.506
 marsh+0.505
Neg logit contributions
ate-0.788
 army-0.754
amed-0.716
roc-0.714
ating-0.695
Token sick
Token ID6639
Feature activation+2.72e-01

Pos logit contributions
Pen+1.451
 syrup+1.439
 Pen+1.381
 kil+1.315
 marsh+1.302
Neg logit contributions
ate-1.745
 army-1.681
amed-1.618
roc-1.554
ating-1.538
Token."
Token ID526
Feature activation+1.03e-02

Pos logit contributions
ate+0.562
 army+0.541
amed+0.522
roc+0.518
ating+0.505
Neg logit contributions
 syrup-0.372
Pen-0.348
 Pen-0.331
 marsh-0.323
 kil-0.312
Token The
Token ID383
Feature activation+1.09e+00

Pos logit contributions
ate+0.019
 army+0.018
amed+0.018
roc+0.017
ating+0.017
Neg logit contributions
 syrup-0.016
Pen-0.015
 Pen-0.015
 marsh-0.014
 kil-0.014
Token mole
Token ID9411
Feature activation-1.88e-01

Pos logit contributions
ate+1.828
 army+1.693
amed+1.667
ating+1.620
roc+1.618
Neg logit contributions
 syrup-1.950
Pen-1.832
 Pen-1.779
 marsh-1.776
 Frost-1.741
Token gr
Token ID1036
Feature activation+3.62e+00

Pos logit contributions
 syrup+0.290
Pen+0.275
 Pen+0.261
 marsh+0.257
 kil+0.252
Neg logit contributions
ate-0.354
 army-0.341
amed-0.328
roc-0.324
ating-0.316
Tokenumbled
Token ID11137
Feature activation+3.66e-01

Pos logit contributions
 army+5.763
ate+5.274
 alien+4.967
roc+4.491
 photographer+4.489
Neg logit contributions
 syrup-7.029
Pen-6.947
 Pen-6.653
 Megan-6.635
 Lu-6.428
Token but
Token ID475
Feature activation-1.35e-01

Pos logit contributions
ate+0.633
 army+0.603
amed+0.575
roc+0.573
ating+0.554
Neg logit contributions
 syrup-0.629
Pen-0.595
 Pen-0.575
 marsh-0.564
 Frost-0.550
Token eventually
Token ID4191
Feature activation-1.33e-01

Pos logit contributions
 syrup+0.182
Pen+0.170
 Pen+0.159
 marsh+0.158
 kil+0.153
Neg logit contributions
ate-0.283
 army-0.274
amed-0.265
roc-0.260
ating-0.255
Token went
Token ID1816
Feature activation-1.81e-01

Pos logit contributions
 syrup+0.207
Pen+0.198
 Pen+0.186
 marsh+0.183
 kil+0.179
Neg logit contributions
ate-0.252
 army-0.243
amed-0.234
roc-0.229
ating-0.225
Token to
Token ID284
Feature activation-6.19e-01

Pos logit contributions
 syrup+0.269
Pen+0.257
 Pen+0.244
 marsh+0.238
 kil+0.234
Neg logit contributions
ate-0.354
 army-0.341
amed-0.328
roc-0.321
ating-0.316
Token said
Token ID531
Feature activation+1.12e+00

Pos logit contributions
ate+1.277
 army+1.228
amed+1.192
roc+1.181
ating+1.155
Neg logit contributions
 syrup-0.812
Pen-0.753
 Pen-0.729
 marsh-0.701
 Frost-0.686
Token goodbye
Token ID24829
Feature activation+2.25e-02

Pos logit contributions
ate+1.977
 army+1.869
amed+1.821
roc+1.797
ating+1.730
Neg logit contributions
 syrup-1.885
Pen-1.795
 Pen-1.772
 marsh-1.727
 Frost-1.710
Token.
Token ID13
Feature activation+1.91e-01

Pos logit contributions
ate+0.048
 army+0.047
amed+0.045
roc+0.044
ating+0.044
Neg logit contributions
 syrup-0.029
Pen-0.027
 Pen-0.026
 marsh-0.025
 kil-0.024
Token The
Token ID383
Feature activation+1.30e+00

Pos logit contributions
ate+0.332
 army+0.317
amed+0.303
roc+0.300
ating+0.292
Neg logit contributions
 syrup-0.325
Pen-0.309
 Pen-0.296
 marsh-0.290
 kil-0.283
Token person
Token ID1048
Feature activation-1.23e-01

Pos logit contributions
ate+2.145
amed+1.959
 army+1.942
ating+1.919
roc+1.912
Neg logit contributions
 syrup-2.367
Pen-2.240
 Pen-2.188
 marsh-2.131
 tug-2.117
Token gr
Token ID1036
Feature activation+3.61e+00

Pos logit contributions
 syrup+0.187
Pen+0.177
 Pen+0.167
 marsh+0.165
 kil+0.160
Neg logit contributions
ate-0.237
 army-0.229
amed-0.220
roc-0.215
ating-0.212
Tokenumbled
Token ID11137
Feature activation+6.76e-01

Pos logit contributions
 army+5.698
ate+5.305
 alien+5.079
roc+4.720
amed+4.528
Neg logit contributions
 syrup-6.464
Pen-6.406
 kil-6.379
 Pen-6.225
 Megan-6.154
Token a
Token ID257
Feature activation-5.40e-01

Pos logit contributions
ate+1.296
 army+1.232
roc+1.194
amed+1.183
ating+1.146
Neg logit contributions
 syrup-1.035
Pen-0.970
 Pen-0.944
 marsh-0.918
 Frost-0.901
Token small
Token ID1402
Feature activation+4.08e-01

Pos logit contributions
 syrup+0.725
Pen+0.687
 Pen+0.649
 marsh+0.622
 Frost+0.616
Neg logit contributions
 army-1.099
ate-1.095
amed-1.037
roc-1.024
ating-0.989
Token thank
Token ID5875
Feature activation-1.80e-01

Pos logit contributions
ate+0.784
 army+0.743
amed+0.721
roc+0.712
ating+0.698
Neg logit contributions
 syrup-0.624
Pen-0.587
 Pen-0.564
 marsh-0.551
 kil-0.535
Token you
Token ID345
Feature activation+2.59e-01

Pos logit contributions
 syrup+0.269
Pen+0.251
 Pen+0.238
 marsh+0.237
 kil+0.225
Neg logit contributions
ate-0.351
 army-0.339
roc-0.322
amed-0.322
ating-0.316
Token.
Token ID13
Feature activation+5.70e-01

Pos logit contributions
ate+2.267
 army+2.176
amed+2.156
roc+2.123
ating+2.044
Neg logit contributions
 syrup-1.166
Pen-1.034
 Pen-1.007
 marsh-0.988
 kil-0.968
Token They
Token ID1119
Feature activation-2.87e-01

Pos logit contributions
ate+0.930
 army+0.896
amed+0.862
roc+0.840
ating+0.829
Neg logit contributions
 syrup-1.023
Pen-0.982
 Pen-0.952
 marsh-0.923
 Frost-0.914
Token were
Token ID547
Feature activation+1.17e+00

Pos logit contributions
 syrup+0.436
Pen+0.416
 Pen+0.395
 marsh+0.387
 kil+0.379
Neg logit contributions
ate-0.552
 army-0.534
amed-0.509
roc-0.505
ating-0.485
Token happy
Token ID3772
Feature activation+4.33e-01

Pos logit contributions
ate+2.189
 army+2.076
roc+2.058
amed+1.987
ating+1.981
Neg logit contributions
 syrup-1.832
Pen-1.716
 Pen-1.676
 Frost-1.632
 marsh-1.627
Token to
Token ID284
Feature activation+1.91e+00

Pos logit contributions
ate+0.845
 army+0.808
amed+0.783
roc+0.774
ating+0.752
Neg logit contributions
 syrup-0.654
Pen-0.611
 Pen-0.589
 marsh-0.576
 Frost-0.564
Token be
Token ID307
Feature activation+3.61e+00

Pos logit contributions
ate+4.397
amed+4.246
 army+4.206
ating+4.141
roc+4.057
Neg logit contributions
 syrup-2.208
 tug-1.960
Pen-1.948
 Pen-1.840
 Frost-1.831
Token pirates
Token ID27516
Feature activation+1.20e+00

Pos logit contributions
ate+8.138
roc+7.973
ully+7.626
isdom+7.491
ini+7.303
Neg logit contributions
 syrup-4.181
 almost-4.129
 bit-4.020
 tug-4.012
 Frost-3.830
Token and
Token ID290
Feature activation+1.58e+00

Pos logit contributions
ate+2.265
amed+2.097
 army+2.093
roc+2.073
ating+1.990
Neg logit contributions
 syrup-1.871
Pen-1.748
 Pen-1.724
 marsh-1.669
 Frost-1.666
Token sharks
Token ID27476
Feature activation+3.84e-01

Pos logit contributions
ate+3.582
amed+3.373
 army+3.319
roc+3.315
ating+3.314
Neg logit contributions
 syrup-1.928
Pen-1.736
 Pen-1.720
 marsh-1.666
 Frost-1.661
Token,
Token ID11
Feature activation+1.26e+00

Pos logit contributions
ate+0.789
 army+0.757
amed+0.732
roc+0.728
ating+0.707
Neg logit contributions
 syrup-0.531
Pen-0.497
 Pen-0.475
 marsh-0.464
 Frost-0.451
Token and
Token ID290
Feature activation+1.26e+00

Pos logit contributions
ate+2.640
amed+2.473
 army+2.462
roc+2.412
ating+2.391
Neg logit contributions
 syrup-1.781
Pen-1.610
 Pen-1.581
 marsh-1.554
 Frost-1.547
Token came
Token ID1625
Feature activation-3.39e-01

Pos logit contributions
 syrup+0.600
Pen+0.578
 Pen+0.543
 marsh+0.525
 kil+0.515
Neg logit contributions
ate-0.833
 army-0.805
amed-0.767
roc-0.756
ating-0.745
Token back
Token ID736
Feature activation-2.79e-01

Pos logit contributions
 syrup+0.500
Pen+0.471
 Pen+0.444
 marsh+0.436
 kil+0.424
Neg logit contributions
ate-0.677
 army-0.651
amed-0.626
roc-0.611
ating-0.603
Token inside
Token ID2641
Feature activation-5.06e-01

Pos logit contributions
 syrup+0.404
Pen+0.383
 Pen+0.358
 marsh+0.351
 kil+0.342
Neg logit contributions
ate-0.563
 army-0.541
amed-0.515
roc-0.509
ating-0.502
Token.
Token ID13
Feature activation+1.15e+00

Pos logit contributions
 syrup+0.723
Pen+0.697
 Pen+0.644
 marsh+0.631
 Willow+0.613
Neg logit contributions
ate-1.013
 army-0.989
amed-0.919
roc-0.915
ating-0.910
Token Spot
Token ID15899
Feature activation+1.46e-01

Pos logit contributions
ate+2.216
 army+2.105
amed+2.069
ating+2.014
roc+1.977
Neg logit contributions
 syrup-1.732
Pen-1.668
 Pen-1.648
 Frost-1.578
 marsh-1.539
Token gr
Token ID1036
Feature activation+3.60e+00

Pos logit contributions
ate+0.277
 army+0.266
amed+0.255
roc+0.252
ating+0.246
Neg logit contributions
 syrup-0.224
Pen-0.212
 Pen-0.202
 marsh-0.198
 kil-0.193
Tokenumbled
Token ID11137
Feature activation+2.34e-01

Pos logit contributions
 army+5.585
ate+5.510
 alien+4.983
 disl+4.647
roc+4.629
Neg logit contributions
 syrup-6.583
Pen-6.539
 Megan-6.228
 Pen-6.215
 Lu-6.181
Token,
Token ID11
Feature activation-5.53e-02

Pos logit contributions
ate+0.441
 army+0.423
amed+0.405
roc+0.402
ating+0.392
Neg logit contributions
 syrup-0.367
Pen-0.347
 Pen-0.332
 marsh-0.325
 Frost-0.316
Token and
Token ID290
Feature activation-3.35e-02

Pos logit contributions
 syrup+0.083
Pen+0.079
 Pen+0.075
 marsh+0.073
 kil+0.072
Neg logit contributions
ate-0.106
 army-0.103
amed-0.099
roc-0.097
ating-0.095
Token ran
Token ID4966
Feature activation-8.65e-01

Pos logit contributions
 syrup+0.046
Pen+0.043
 Pen+0.040
 marsh+0.040
 kil+0.039
Neg logit contributions
ate-0.069
 army-0.067
amed-0.065
roc-0.064
ating-0.063
Token off
Token ID572
Feature activation-5.87e-01

Pos logit contributions
 syrup+1.343
Pen+1.311
 Pen+1.213
 kil+1.196
 marsh+1.164
Neg logit contributions
ate-1.637
 army-1.596
amed-1.561
roc-1.467
ating-1.460
Token!"
Token ID2474
Feature activation+1.11e+00

Pos logit contributions
 syrup+0.497
Pen+0.460
 Pen+0.430
 marsh+0.421
 kil+0.410
Neg logit contributions
ate-0.813
 army-0.783
amed-0.751
ating-0.736
roc-0.735
Token
Token ID198
Feature activation+1.21e+00

Pos logit contributions
ate+2.276
 army+2.212
amed+2.122
roc+2.112
ating+2.079
Neg logit contributions
 syrup-1.541
Pen-1.449
 Pen-1.414
 marsh-1.335
 Frost-1.330
Token
Token ID198
Feature activation+1.73e+00

Pos logit contributions
ate+2.481
 army+2.475
roc+2.338
amed+2.334
ating+2.306
Neg logit contributions
 syrup-1.613
Pen-1.584
 Pen-1.508
 Frost-1.440
 marsh-1.422
TokenThe
Token ID464
Feature activation+1.38e+00

Pos logit contributions
ate+3.024
 army+3.020
roc+2.800
amed+2.763
ating+2.755
Neg logit contributions
Pen-2.860
 syrup-2.806
 Pen-2.647
 tug-2.579
 Willow-2.484
Token cat
Token ID3797
Feature activation+2.54e-01

Pos logit contributions
ate+2.402
 army+2.156
roc+2.152
amed+2.146
ating+2.120
Neg logit contributions
 syrup-2.480
Pen-2.347
 Pen-2.278
 marsh-2.246
 Frost-2.161
Token gr
Token ID1036
Feature activation+3.59e+00

Pos logit contributions
ate+0.500
 army+0.480
amed+0.461
roc+0.457
ating+0.447
Neg logit contributions
 syrup-0.372
Pen-0.351
 Pen-0.335
 marsh-0.327
 Frost-0.317
Tokenump
Token ID931
Feature activation+1.63e+00

Pos logit contributions
 army+5.707
ate+5.376
 alien+4.796
roc+4.475
 friendships+4.408
Neg logit contributions
 syrup-6.816
Pen-6.698
 Megan-6.368
 Pen-6.348
 Lu-6.262
Tokenily
Token ID813
Feature activation+2.31e-01

Pos logit contributions
ate+2.742
 army+2.569
roc+2.491
amed+2.434
ating+2.287
Neg logit contributions
 syrup-2.988
Pen-2.750
 Pen-2.739
 Megan-2.614
 marsh-2.606
Token me
Token ID502
Feature activation+1.07e-01

Pos logit contributions
ate+0.441
 army+0.424
amed+0.406
roc+0.403
ating+0.393
Neg logit contributions
 syrup-0.353
Pen-0.333
 Pen-0.319
 marsh-0.311
 kil-0.302
Tokenowed
Token ID6972
Feature activation+8.34e-01

Pos logit contributions
ate+0.181
 army+0.173
amed+0.166
roc+0.164
ating+0.159
Neg logit contributions
 syrup-0.188
Pen-0.179
 Pen-0.171
 marsh-0.168
 kil-0.164
Token and
Token ID290
Feature activation-9.09e-02

Pos logit contributions
ate+1.637
 army+1.555
roc+1.503
amed+1.481
ating+1.452
Neg logit contributions
 syrup-1.257
Pen-1.176
 Pen-1.139
 marsh-1.110
 Frost-1.088
Tokenmy
Token ID1820
Feature activation+4.07e-01

Pos logit contributions
ate+0.193
 army+0.185
amed+0.177
roc+0.175
ating+0.170
Neg logit contributions
 syrup-0.189
Pen-0.180
 Pen-0.173
 marsh-0.170
 kil-0.166
Token explained
Token ID4893
Feature activation+5.28e-01

Pos logit contributions
ate+0.782
 army+0.752
amed+0.717
roc+0.716
ating+0.697
Neg logit contributions
 syrup-0.624
Pen-0.586
 Pen-0.563
 marsh-0.548
 Frost-0.536
Token to
Token ID284
Feature activation+1.03e+00

Pos logit contributions
ate+0.937
 army+0.891
amed+0.859
roc+0.841
ating+0.822
Neg logit contributions
 syrup-0.896
Pen-0.841
 Pen-0.816
 marsh-0.799
 Frost-0.781
Token her
Token ID607
Feature activation+8.66e-01

Pos logit contributions
ate+2.080
 army+2.004
amed+1.959
roc+1.895
ating+1.881
Neg logit contributions
 syrup-1.493
Pen-1.391
 Pen-1.339
 marsh-1.278
 Willow-1.265
Token that
Token ID326
Feature activation+7.86e-01

Pos logit contributions
ate+1.708
 army+1.639
amed+1.594
roc+1.577
ating+1.528
Neg logit contributions
 syrup-1.293
Pen-1.194
 Pen-1.176
 marsh-1.134
 Willow-1.114
Token cl
Token ID537
Feature activation+3.57e+00

Pos logit contributions
ate+1.757
 army+1.688
amed+1.663
roc+1.643
ating+1.590
Neg logit contributions
 syrup-0.964
Pen-0.875
 Pen-0.856
 marsh-0.830
 Willow-0.792
Tokenapping
Token ID5912
Feature activation+1.86e+00

Pos logit contributions
 army+4.278
ate+3.970
 alien+3.823
ully+3.709
amed+3.708
Neg logit contributions
 syrup-8.088
 kil-7.553
 Megan-7.344
Pen-7.299
abel-7.195
Token is
Token ID318
Feature activation+3.60e-01

Pos logit contributions
ate+3.800
 army+3.632
roc+3.529
amed+3.479
ating+3.347
Neg logit contributions
 syrup-2.702
Pen-2.395
 Pen-2.379
 marsh-2.336
 might-2.297
Token a
Token ID257
Feature activation-6.41e-01

Pos logit contributions
ate+0.700
 army+0.675
amed+0.646
roc+0.643
ating+0.626
Neg logit contributions
 syrup-0.540
Pen-0.506
 Pen-0.485
 marsh-0.476
 kil-0.459
Token way
Token ID835
Feature activation+5.85e-01

Pos logit contributions
 syrup+0.922
Pen+0.894
 Pen+0.848
 Frost+0.794
 marsh+0.791
Neg logit contributions
ate-1.261
 army-1.246
amed-1.165
roc-1.161
ating-1.119
Token to
Token ID284
Feature activation+1.32e+00

Pos logit contributions
ate+1.232
 army+1.182
amed+1.144
roc+1.142
ating+1.106
Neg logit contributions
 syrup-0.780
Pen-0.728
 Pen-0.698
 marsh-0.681
 Willow-0.653
Token mess
Token ID2085
Feature activation+4.76e-01

Pos logit contributions
ate+1.157
 army+1.103
amed+1.073
roc+1.053
ating+1.024
Neg logit contributions
 syrup-1.164
Pen-1.077
 Pen-1.041
 marsh-1.035
 kil-1.004
Token.
Token ID13
Feature activation+8.05e-03

Pos logit contributions
ate+0.983
 army+0.950
amed+0.915
roc+0.913
ating+0.879
Neg logit contributions
 syrup-0.654
Pen-0.606
 Pen-0.581
 marsh-0.565
 Frost-0.550
Token They
Token ID1119
Feature activation-9.54e-01

Pos logit contributions
ate+0.012
 army+0.012
amed+0.011
roc+0.011
ating+0.011
Neg logit contributions
 syrup-0.015
Pen-0.015
 Pen-0.014
 marsh-0.014
 kil-0.014
Token learned
Token ID4499
Feature activation+5.00e-01

Pos logit contributions
 syrup+1.559
Pen+1.544
 Pen+1.448
 kil+1.434
 marsh+1.420
Neg logit contributions
ate-1.711
 army-1.640
amed-1.606
roc-1.518
room-1.470
Token that
Token ID326
Feature activation+5.24e-01

Pos logit contributions
ate+0.985
 army+0.957
amed+0.921
roc+0.911
ating+0.891
Neg logit contributions
 syrup-0.739
Pen-0.693
 Pen-0.666
 marsh-0.648
 Frost-0.640
Token decor
Token ID11705
Feature activation+3.56e+00

Pos logit contributions
ate+0.971
 army+0.932
amed+0.907
roc+0.888
ating+0.863
Neg logit contributions
 syrup-0.839
Pen-0.785
 Pen-0.760
 marsh-0.745
 Frost-0.729
Tokenating
Token ID803
Feature activation+1.65e+00

Pos logit contributions
 army+3.299
 alien+2.549
roc+2.190
 photographer+2.130
ully+2.123
Neg logit contributions
 syrup-9.837
 might-9.106
 marsh-9.086
 Frost-9.029
 almost-8.903
Token was
Token ID373
Feature activation-2.29e-01

Pos logit contributions
ate+3.264
 army+3.226
amed+3.171
roc+3.094
ully+2.934
Neg logit contributions
 syrup-2.398
 marsh-2.168
 Pen-2.137
Pen-2.134
 Frost-2.092
Token fun
Token ID1257
Feature activation+1.43e+00

Pos logit contributions
 syrup+0.376
Pen+0.363
 Pen+0.346
 marsh+0.334
 Willow+0.329
Neg logit contributions
ate-0.413
 army-0.391
amed-0.374
roc-0.369
ating-0.362
Token,
Token ID11
Feature activation+1.23e+00

Pos logit contributions
ate+2.721
 army+2.662
roc+2.624
amed+2.614
ully+2.439
Neg logit contributions
 syrup-2.171
 marsh-1.954
 Pen-1.920
 might-1.909
Pen-1.900
Token but
Token ID475
Feature activation+1.09e+00

Pos logit contributions
ate+3.086
 army+2.957
amed+2.947
roc+2.885
ating+2.864
Neg logit contributions
 syrup-1.212
Pen-1.038
 Pen-1.010
 marsh-0.972
 Frost-0.964
Token
Token ID198
Feature activation-6.41e-01

Pos logit contributions
 syrup+1.636
Pen+1.498
 kil+1.458
 marsh+1.452
 Pen+1.428
Neg logit contributions
ate-1.551
 army-1.406
amed-1.368
room-1.331
roc-1.310
TokenM
Token ID44
Feature activation+4.68e-01

Pos logit contributions
 syrup+0.914
Pen+0.835
 Pen+0.807
 marsh+0.806
 kil+0.798
Neg logit contributions
ate-1.307
 army-1.237
amed-1.215
roc-1.166
ating-1.153
Tokenolly
Token ID5098
Feature activation+9.34e-01

Pos logit contributions
ate+0.719
 army+0.689
amed+0.656
roc+0.648
ating+0.624
Neg logit contributions
 syrup-0.896
Pen-0.855
 Pen-0.824
 marsh-0.805
 Willow-0.788
Token heard
Token ID2982
Feature activation+1.64e+00

Pos logit contributions
ate+1.856
 army+1.790
roc+1.724
amed+1.709
ating+1.679
Neg logit contributions
 syrup-1.330
Pen-1.276
 Pen-1.256
 marsh-1.173
 Willow-1.160
Token some
Token ID617
Feature activation+1.63e+00

Pos logit contributions
ate+3.407
 army+3.231
roc+3.181
ating+3.145
amed+3.128
Neg logit contributions
 syrup-2.346
Pen-2.164
 Pen-2.145
 marsh-2.053
 Willow-2.023
Token cl
Token ID537
Feature activation+3.55e+00

Pos logit contributions
ate+2.758
 army+2.500
amed+2.455
ating+2.387
roc+2.386
Neg logit contributions
 syrup-2.999
Pen-2.810
 Pen-2.721
 marsh-2.699
 kil-2.639
Tokenapping
Token ID5912
Feature activation+2.09e+00

Pos logit contributions
ate+3.442
 army+3.430
roc+3.105
ating+2.908
 alien+2.801
Neg logit contributions
 syrup-8.445
 Megan-8.168
Pen-8.046
 Pen-7.874
 kil-7.847
Token and
Token ID290
Feature activation+1.28e+00

Pos logit contributions
ate+4.246
roc+4.007
 army+3.980
amed+3.824
ating+3.750
Neg logit contributions
 syrup-3.032
 Pen-2.794
Pen-2.765
 marsh-2.609
 Megan-2.598
Token went
Token ID1816
Feature activation+9.75e-02

Pos logit contributions
ate+2.727
 army+2.616
roc+2.531
amed+2.496
ating+2.471
Neg logit contributions
 syrup-1.677
Pen-1.585
 Pen-1.570
 marsh-1.481
 Willow-1.472
Token to
Token ID284
Feature activation-6.34e-03

Pos logit contributions
ate+0.203
 army+0.195
amed+0.188
roc+0.186
ating+0.182
Neg logit contributions
 syrup-0.134
Pen-0.126
 Pen-0.119
 marsh-0.116
 kil-0.113
Token investigate
Token ID9161
Feature activation-3.09e-01

Pos logit contributions
 syrup+0.010
Pen+0.009
 Pen+0.009
 marsh+0.009
 kil+0.008
Neg logit contributions
ate-0.012
 army-0.012
amed-0.011
roc-0.011
ating-0.011
Token,
Token ID11
Feature activation-2.70e-01

Pos logit contributions
ate+1.611
 army+1.507
amed+1.482
roc+1.446
ating+1.338
Neg logit contributions
 syrup-2.175
Pen-2.088
 Pen-2.070
 marsh-2.023
 Willow-1.974
Token Tim
Token ID5045
Feature activation+1.33e+00

Pos logit contributions
 syrup+0.408
Pen+0.384
 Pen+0.362
 marsh+0.356
 kil+0.352
Neg logit contributions
ate-0.517
 army-0.499
roc-0.473
amed-0.471
ating-0.460
Tokenmy
Token ID1820
Feature activation+1.32e+00

Pos logit contributions
ate+2.459
 army+2.398
amed+2.277
roc+2.266
ating+2.160
Neg logit contributions
 syrup-2.141
Pen-1.981
 Pen-1.930
 marsh-1.826
 Willow-1.815
Token learned
Token ID4499
Feature activation+1.10e+00

Pos logit contributions
ate+2.628
 army+2.576
roc+2.493
amed+2.465
room+2.348
Neg logit contributions
 syrup-1.911
Pen-1.735
 Pen-1.707
 marsh-1.657
 Frost-1.602
Token to
Token ID284
Feature activation+1.19e+00

Pos logit contributions
ate+2.264
 army+2.224
amed+2.167
roc+2.134
ating+2.084
Neg logit contributions
 syrup-1.507
Pen-1.369
 Pen-1.326
 marsh-1.290
 Frost-1.289
Token cl
Token ID537
Feature activation+3.54e+00

Pos logit contributions
ate+2.294
 army+2.278
amed+2.222
roc+2.172
ating+2.094
Neg logit contributions
 syrup-1.740
Pen-1.619
 Pen-1.554
 marsh-1.503
 Frost-1.478
Tokenap
Token ID499
Feature activation+1.82e+00

Pos logit contributions
 army+3.245
 alien+3.024
roc+2.649
amed+2.629
ully+2.614
Neg logit contributions
 syrup-8.843
 kil-8.536
Pen-8.437
 tug-8.375
 Megan-8.115
Token for
Token ID329
Feature activation+4.80e-01

Pos logit contributions
ate+3.500
 army+3.406
roc+3.272
amed+3.266
ating+3.133
Neg logit contributions
 syrup-2.784
Pen-2.533
 Pen-2.493
 marsh-2.426
 Frost-2.410
Token himself
Token ID2241
Feature activation+3.57e-01

Pos logit contributions
ate+0.838
 army+0.797
amed+0.774
roc+0.758
ating+0.737
Neg logit contributions
 syrup-0.819
Pen-0.774
 Pen-0.747
 marsh-0.731
 kil-0.715
Token and
Token ID290
Feature activation+1.39e+00

Pos logit contributions
ate+0.678
 army+0.650
amed+0.626
roc+0.621
ating+0.602
Neg logit contributions
 syrup-0.549
Pen-0.517
 Pen-0.496
 marsh-0.484
 kil-0.472
Token not
Token ID407
Feature activation-2.90e-01

Pos logit contributions
ate+2.974
 army+2.878
amed+2.830
roc+2.820
ating+2.710
Neg logit contributions
 syrup-1.800
Pen-1.625
 Pen-1.603
 marsh-1.540
 Frost-1.512
Token with
Token ID351
Feature activation+1.17e+00

Pos logit contributions
ate+0.188
 army+0.180
amed+0.170
roc+0.168
ating+0.163
Neg logit contributions
 syrup-0.219
Pen-0.209
 Pen-0.201
 marsh-0.198
 kil-0.193
Token a
Token ID257
Feature activation+1.12e+00

Pos logit contributions
ate+2.397
 army+2.217
ating+2.193
roc+2.192
amed+2.153
Neg logit contributions
 syrup-1.723
Pen-1.557
 Pen-1.488
 marsh-1.488
 Frost-1.445
Token club
Token ID3430
Feature activation+8.31e-01

Pos logit contributions
ate+2.168
amed+1.991
 army+1.989
ating+1.969
roc+1.941
Neg logit contributions
 syrup-1.790
Pen-1.621
 marsh-1.577
 Pen-1.555
 tug-1.526
Token.
Token ID13
Feature activation+9.52e-01

Pos logit contributions
ate+1.814
 army+1.719
roc+1.707
amed+1.676
ating+1.636
Neg logit contributions
 syrup-1.050
Pen-0.942
 Pen-0.902
 marsh-0.886
 kil-0.872
Token The
Token ID383
Feature activation+2.77e+00

Pos logit contributions
ate+1.623
 army+1.552
amed+1.511
ating+1.461
roc+1.446
Neg logit contributions
 syrup-1.667
Pen-1.567
 Pen-1.536
 marsh-1.514
 Frost-1.505
Token big
Token ID1263
Feature activation+3.53e+00

Pos logit contributions
ate+5.308
ating+5.198
isdom+5.115
 army+5.043
ined+4.936
Neg logit contributions
 syrup-4.031
 might-3.965
 bit-3.718
 tug-3.641
 almost-3.574
Token boy
Token ID2933
Feature activation+3.03e-01

Pos logit contributions
ate+8.445
ating+8.039
amed+7.943
isdom+7.722
ined+7.719
Neg logit contributions
 bit-4.275
 syrup-4.174
 might-4.106
 tug-4.093
 sled-3.972
Token was
Token ID373
Feature activation+2.11e+00

Pos logit contributions
ate+0.619
 army+0.595
amed+0.573
roc+0.567
ating+0.555
Neg logit contributions
 syrup-0.427
Pen-0.400
 Pen-0.382
 marsh-0.371
 Frost-0.362
Token hitting
Token ID9008
Feature activation+1.62e+00

Pos logit contributions
ate+4.935
ating+4.695
 army+4.596
roc+4.592
isdom+4.425
Neg logit contributions
 syrup-2.495
 Pen-2.315
Pen-2.263
 Frost-2.235
 almost-2.227
Token trees
Token ID7150
Feature activation+3.81e-01

Pos logit contributions
ate+3.466
 army+3.359
amed+3.260
ating+3.203
roc+3.073
Neg logit contributions
 syrup-2.142
 Frost-1.885
Pen-1.850
 marsh-1.834
 Pen-1.793
Token with
Token ID351
Feature activation+1.11e+00

Pos logit contributions
ate+0.622
 army+0.592
amed+0.564
roc+0.560
ating+0.541
Neg logit contributions
 syrup-0.690
Pen-0.655
 Pen-0.633
 marsh-0.621
 Frost-0.607
Token
Token ID198
Feature activation+1.27e+00

Pos logit contributions
ate+2.099
 army+2.080
amed+1.986
roc+1.965
ating+1.952
Neg logit contributions
 syrup-1.419
Pen-1.373
 Pen-1.298
 Frost-1.251
 marsh-1.241
TokenL
Token ID43
Feature activation+1.07e+00

Pos logit contributions
ate+2.401
 army+2.382
amed+2.236
roc+2.224
ating+2.204
Neg logit contributions
 syrup-1.929
Pen-1.912
 Pen-1.761
 Willow-1.669
 Frost-1.668
Tokenily
Token ID813
Feature activation+1.23e+00

Pos logit contributions
ate+1.646
 army+1.644
amed+1.531
roc+1.521
ating+1.474
Neg logit contributions
 syrup-2.021
Pen-1.937
 Pen-1.871
 kil-1.787
 Frost-1.784
Token and
Token ID290
Feature activation+9.44e-01

Pos logit contributions
ate+2.217
 army+2.200
roc+2.102
amed+2.079
ating+2.036
Neg logit contributions
 syrup-1.988
Pen-1.838
 Pen-1.817
 Frost-1.737
 marsh-1.725
Token Ben
Token ID3932
Feature activation+3.53e-01

Pos logit contributions
ate+2.053
amed+1.953
 army+1.953
ating+1.881
roc+1.869
Neg logit contributions
 syrup-1.265
Pen-1.157
 Pen-1.130
 marsh-1.071
 Frost-1.068
Token cl
Token ID537
Feature activation+3.52e+00

Pos logit contributions
ate+0.692
 army+0.666
amed+0.639
roc+0.635
ating+0.621
Neg logit contributions
 syrup-0.523
Pen-0.491
 Pen-0.470
 marsh-0.459
 Frost-0.446
Tokenapped
Token ID6320
Feature activation+1.52e+00

Pos logit contributions
 army+3.135
ate+3.099
ating+2.981
ully+2.768
 disl+2.690
Neg logit contributions
 syrup-8.471
 Frost-8.372
Pen-8.349
 Megan-8.320
 Pen-8.257
Token their
Token ID511
Feature activation+3.45e-01

Pos logit contributions
ate+2.889
 army+2.761
roc+2.718
amed+2.666
ating+2.661
Neg logit contributions
 syrup-2.374
Pen-2.153
 Pen-2.119
 Frost-2.110
 marsh-2.069
Token hands
Token ID2832
Feature activation-2.04e-01

Pos logit contributions
ate+0.693
 army+0.665
amed+0.638
roc+0.634
ating+0.621
Neg logit contributions
 syrup-0.501
Pen-0.472
 Pen-0.451
 marsh-0.439
 Frost-0.425
Token and
Token ID290
Feature activation+1.34e+00

Pos logit contributions
 syrup+0.338
Pen+0.326
 Pen+0.310
 marsh+0.302
 kil+0.296
Neg logit contributions
ate-0.360
 army-0.348
amed-0.334
roc-0.323
ating-0.318
Token gave
Token ID2921
Feature activation-4.61e-01

Pos logit contributions
ate+2.894
 army+2.769
roc+2.675
ating+2.618
amed+2.604
Neg logit contributions
 syrup-1.828
Pen-1.636
 Pen-1.632
 marsh-1.591
 Frost-1.568
Token boy
Token ID2933
Feature activation+7.45e-01

Pos logit contributions
ate+2.274
 army+2.168
amed+2.168
ating+2.092
roc+2.088
Neg logit contributions
 syrup-2.088
Pen-1.947
 Frost-1.889
 Pen-1.888
 marsh-1.883
Token,
Token ID11
Feature activation-4.71e-01

Pos logit contributions
ate+1.283
 army+1.219
amed+1.193
roc+1.141
ating+1.137
Neg logit contributions
 syrup-1.272
Pen-1.213
 Pen-1.183
 Frost-1.152
 marsh-1.138
Token "
Token ID366
Feature activation-2.58e-02

Pos logit contributions
 syrup+0.870
Pen+0.825
 marsh+0.783
 Pen+0.782
 kil+0.769
Neg logit contributions
ate-0.744
 army-0.714
amed-0.674
roc-0.661
ating-0.636
TokenWhat
Token ID2061
Feature activation+1.58e+00

Pos logit contributions
 syrup+0.033
Pen+0.031
 Pen+0.030
 marsh+0.029
 kil+0.028
Neg logit contributions
ate-0.055
 army-0.054
amed-0.052
roc-0.051
ating-0.050
Token's
Token ID338
Feature activation+9.60e-01

Pos logit contributions
ate+3.116
amed+3.088
 army+3.008
roc+2.894
ating+2.881
Neg logit contributions
 syrup-2.241
Pen-2.108
 Pen-2.075
 might-2.022
 Frost-2.001
Token that
Token ID326
Feature activation+3.52e+00

Pos logit contributions
ate+1.688
 army+1.611
amed+1.585
roc+1.554
ating+1.507
Neg logit contributions
 syrup-1.612
Pen-1.522
 Pen-1.472
 Frost-1.445
 marsh-1.441
Token?"
Token ID1701
Feature activation+3.07e-01

Pos logit contributions
amed+7.298
ate+6.803
ating+6.657
ined+6.338
roc+6.334
Neg logit contributions
 syrup-5.708
 might-5.423
 Frost-5.252
 almost-5.193
 marsh-5.075
Token
Token ID220
Feature activation+4.49e-01

Pos logit contributions
ate+0.565
 army+0.542
amed+0.521
roc+0.514
ating+0.502
Neg logit contributions
 syrup-0.494
Pen-0.468
 Pen-0.450
 marsh-0.439
 Frost-0.428
Token
Token ID198
Feature activation+9.82e-01

Pos logit contributions
ate+0.866
 army+0.842
amed+0.807
roc+0.799
ating+0.781
Neg logit contributions
 syrup-0.665
Pen-0.632
 Pen-0.605
 marsh-0.588
 Frost-0.575
Token
Token ID198
Feature activation+1.36e+00

Pos logit contributions
ate+1.996
 army+1.964
amed+1.877
roc+1.858
ating+1.826
Neg logit contributions
 syrup-1.355
Pen-1.325
 Pen-1.259
 Frost-1.193
 marsh-1.190
Token"
Token ID1
Feature activation+8.44e-01

Pos logit contributions
ate+2.327
 army+2.252
amed+2.161
roc+2.108
ating+2.082
Neg logit contributions
Pen-2.331
 syrup-2.321
 Pen-2.206
 Frost-2.098
 Willow-2.093
Token the
Token ID262
Feature activation-3.01e+00

Pos logit contributions
 syrup+2.354
Pen+2.315
 kil+2.195
 Pen+2.119
 marsh+2.099
Neg logit contributions
ate-2.483
 army-2.271
amed-2.142
room-2.137
ating-2.130
Token chair
Token ID5118
Feature activation-1.05e-01

Pos logit contributions
 syrup+5.155
 Frost+5.040
Pen+4.882
 spe+4.750
 marsh+4.624
Neg logit contributions
 army-5.203
ate-5.042
amed-4.843
room-4.732
ating-4.518
Token and
Token ID290
Feature activation-2.23e+00

Pos logit contributions
 syrup+0.177
Pen+0.170
 Pen+0.161
 marsh+0.159
 Frost+0.154
Neg logit contributions
ate-0.184
 army-0.176
amed-0.168
roc-0.164
ating-0.162
Token reaches
Token ID12229
Feature activation-2.05e+00

Pos logit contributions
Pen+3.606
 syrup+3.525
 Pen+3.208
 kil+3.159
 marsh+3.132
Neg logit contributions
ate-4.231
 army-3.915
amed-3.567
ating-3.507
room-3.374
Token for
Token ID329
Feature activation-2.77e+00

Pos logit contributions
 syrup+3.558
Pen+3.549
 marsh+3.149
 Pen+3.106
 Frost+3.102
Neg logit contributions
ate-3.799
 army-3.491
amed-3.229
ating-3.197
room-3.195
Token the
Token ID262
Feature activation-4.09e+00

Pos logit contributions
Pen+5.427
 syrup+5.319
 kil+4.953
fy+4.900
 Pen+4.898
Neg logit contributions
ate-4.196
 army-3.948
room-3.678
ating-3.403
 models-3.364
Token top
Token ID1353
Feature activation-2.18e+00

Pos logit contributions
 Frost+7.898
Pen+7.718
 syrup+7.613
 dev+7.457
 spe+7.187
Neg logit contributions
 army-6.899
room-6.185
ate-6.184
 office-5.867
amed-5.601
Token books
Token ID3835
Feature activation-1.52e+00

Pos logit contributions
 syrup+3.859
Pen+3.782
 Frost+3.536
 marsh+3.460
 Pen+3.440
Neg logit contributions
ate-3.785
room-3.523
 army-3.474
ating-3.363
amed-3.218
Token.
Token ID13
Feature activation-3.29e-01

Pos logit contributions
 syrup+2.479
Pen+2.388
 marsh+2.244
 Pen+2.181
 Frost+2.163
Neg logit contributions
ate-2.848
 army-2.656
room-2.619
amed-2.587
ating-2.503
Token She
Token ID1375
Feature activation-2.00e+00

Pos logit contributions
 syrup+0.598
Pen+0.569
 Pen+0.541
 kil+0.537
 marsh+0.536
Neg logit contributions
ate-0.540
 army-0.504
amed-0.487
roc-0.477
room-0.463
Token grabs
Token ID22378
Feature activation-1.63e+00

Pos logit contributions
 syrup+3.367
Pen+3.356
 Pen+3.060
 marsh+3.011
 Willow+2.963
Neg logit contributions
ate-3.666
 army-3.345
amed-3.072
roc-2.993
ating-2.966
Token in
Token ID287
Feature activation-5.91e-01

Pos logit contributions
 syrup+0.565
Pen+0.538
 Pen+0.513
 marsh+0.505
 kil+0.492
Neg logit contributions
ate-0.601
 army-0.569
amed-0.544
roc-0.524
ating-0.524
Token this
Token ID428
Feature activation-1.80e+00

Pos logit contributions
 syrup+1.021
Pen+0.982
 Pen+0.927
 marsh+0.908
 kil+0.890
Neg logit contributions
ate-1.024
 army-0.991
amed-0.904
roc-0.892
ating-0.887
Token seat
Token ID5852
Feature activation-6.10e-02

Pos logit contributions
 syrup+3.369
Pen+3.178
 Pen+3.003
 marsh+2.953
 Frost+2.939
Neg logit contributions
ate-3.093
 army-2.821
ating-2.574
room-2.509
amed-2.482
Token and
Token ID290
Feature activation-7.19e-01

Pos logit contributions
 syrup+0.093
Pen+0.088
 Pen+0.083
 marsh+0.082
 kil+0.080
Neg logit contributions
ate-0.117
 army-0.112
amed-0.107
roc-0.106
ating-0.104
Token take
Token ID1011
Feature activation-8.32e-01

Pos logit contributions
 syrup+1.021
Pen+0.963
 Pen+0.899
 kil+0.890
 marsh+0.879
Neg logit contributions
ate-1.453
 army-1.399
amed-1.309
roc-1.289
ating-1.275
Token a
Token ID257
Feature activation-3.94e+00

Pos logit contributions
 syrup+1.387
Pen+1.343
 Pen+1.277
 marsh+1.229
 kil+1.221
Neg logit contributions
ate-1.492
 army-1.402
amed-1.340
ating-1.294
roc-1.290
Token long
Token ID890
Feature activation-2.27e+00

Pos logit contributions
 syrup+6.530
Pen+6.339
 Pen+6.102
 kil+6.027
 degrees+6.005
Neg logit contributions
 army-6.891
ate-6.685
ap-6.311
 club-6.070
 network-6.010
Token nap
Token ID25422
Feature activation-2.04e-01

Pos logit contributions
 syrup+3.456
Pen+3.449
 kil+3.215
 Pen+3.193
 Frost+3.178
Neg logit contributions
ate-4.076
 army-4.038
room-3.742
amed-3.677
ap-3.574
Token.
Token ID13
Feature activation-1.30e+00

Pos logit contributions
 syrup+0.279
Pen+0.262
 Pen+0.247
 marsh+0.242
 Frost+0.235
Neg logit contributions
ate-0.423
 army-0.408
amed-0.391
roc-0.382
ating-0.381
Tokenâ
Token ID22940
Feature activation+1.40e+00

Pos logit contributions
 syrup+2.426
Pen+2.196
 tug+2.092
 kil+2.075
 marsh+2.067
Neg logit contributions
ate-2.136
 army-1.952
amed-1.841
roc-1.834
room-1.785
Token
Token ID26391
Feature activation-3.96e-01

Pos logit contributions
ate+2.442
 army+2.440
amed+2.252
roc+2.229
 alien+2.138
Neg logit contributions
 syrup-2.298
Pen-2.234
 Pen-2.120
 kil-2.025
 Frost-2.017
Token,
Token ID11
Feature activation-4.61e-01

Pos logit contributions
 syrup+1.151
Pen+1.094
 Pen+1.003
 marsh+0.995
 kil+0.980
Neg logit contributions
ate-1.567
 army-1.502
amed-1.421
ating-1.402
roc-1.387
Token so
Token ID523
Feature activation-9.02e-02

Pos logit contributions
 syrup+0.710
Pen+0.679
 Pen+0.635
 marsh+0.628
 kil+0.622
Neg logit contributions
ate-0.867
 army-0.829
roc-0.788
amed-0.785
ating-0.765
Token it
Token ID340
Feature activation-1.09e+00

Pos logit contributions
 syrup+0.120
Pen+0.112
 Pen+0.106
 marsh+0.103
 kil+0.100
Neg logit contributions
ate-0.192
 army-0.184
amed-0.176
roc-0.176
ating-0.172
Token might
Token ID1244
Feature activation-1.33e+00

Pos logit contributions
 syrup+1.616
Pen+1.503
 marsh+1.390
 Pen+1.388
 kil+1.364
Neg logit contributions
ate-2.205
 army-2.068
ating-1.987
amed-1.975
roc-1.934
Token take
Token ID1011
Feature activation-1.12e-01

Pos logit contributions
 syrup+2.250
Pen+2.159
 Pen+2.046
 kil+2.028
 marsh+1.996
Neg logit contributions
ate-2.239
 army-2.183
roc-2.009
amed-1.964
ating-1.907
Token a
Token ID257
Feature activation-3.67e+00

Pos logit contributions
 syrup+0.172
Pen+0.163
 Pen+0.155
 marsh+0.151
 kil+0.149
Neg logit contributions
ate-0.212
 army-0.204
amed-0.195
roc-0.193
ating-0.189
Token few
Token ID1178
Feature activation-2.19e+00

Pos logit contributions
 syrup+5.797
Pen+5.655
 Megan+5.473
 Pen+5.440
 kil+5.359
Neg logit contributions
 army-6.578
 alien-6.038
 network-5.946
ate-5.886
ap-5.736
Token times
Token ID1661
Feature activation+3.76e-05

Pos logit contributions
 syrup+3.327
Pen+3.312
 Frost+3.186
 Pen+3.117
 Willow+2.983
Neg logit contributions
ate-4.115
 army-3.963
ating-3.772
amed-3.703
ined-3.537
Token for
Token ID329
Feature activation-3.95e-01

Pos logit contributions
ate+0.000
 army+0.000
amed+0.000
roc+0.000
ating+0.000
Neg logit contributions
 syrup-0.000
Pen-0.000
 Pen-0.000
 marsh-0.000
 kil-0.000
Token it
Token ID340
Feature activation-6.02e-01

Pos logit contributions
 syrup+0.646
Pen+0.614
 Pen+0.583
 marsh+0.573
 kil+0.568
Neg logit contributions
ate-0.708
 army-0.679
amed-0.638
roc-0.632
ating-0.623
Token to
Token ID284
Feature activation-3.96e-01

Pos logit contributions
 syrup+0.827
Pen+0.777
 Pen+0.723
 marsh+0.706
 kil+0.697
Neg logit contributions
ate-1.243
 army-1.198
amed-1.149
ating-1.122
roc-1.114
Token slippery
Token ID32911
Feature activation-7.91e-01

Pos logit contributions
 syrup+3.723
Pen+3.705
 Pen+3.392
 kil+3.347
 Willow+3.334
Neg logit contributions
ate-5.073
amed-4.909
 army-4.885
ined-4.667
ating-4.568
Token and
Token ID290
Feature activation-1.64e+00

Pos logit contributions
 syrup+1.293
Pen+1.277
 Pen+1.200
 marsh+1.148
 Willow+1.148
Neg logit contributions
ate-1.417
 army-1.352
amed-1.294
ating-1.278
room-1.242
Token she
Token ID673
Feature activation-2.23e+00

Pos logit contributions
 syrup+2.315
Pen+2.298
 kil+2.151
 Pen+2.130
 Frost+2.062
Neg logit contributions
ate-3.101
 army-3.011
amed-2.974
ating-2.840
roc-2.826
Token couldn
Token ID3521
Feature activation-9.49e-01

Pos logit contributions
 syrup+3.280
Pen+3.202
 marsh+2.985
 kil+2.954
 Pen+2.940
Neg logit contributions
amed-4.157
ate-4.145
 army-4.110
ined-3.855
roc-3.836
Token't
Token ID470
Feature activation-2.41e+00

Pos logit contributions
 syrup+1.425
Pen+1.302
 kil+1.237
 marsh+1.234
 Pen+1.223
Neg logit contributions
ate-1.868
 army-1.736
amed-1.683
room-1.671
ating-1.660
Token get
Token ID651
Feature activation-3.63e+00

Pos logit contributions
 syrup+4.093
Pen+3.713
 marsh+3.529
 Willow+3.404
 Pen+3.399
Neg logit contributions
ate-4.639
 army-4.392
room-4.222
ating-4.005
ully-3.918
Token to
Token ID284
Feature activation-2.03e+00

Pos logit contributions
Pen+7.317
 syrup+7.144
 kil+6.874
 Pen+6.762
 marsh+6.688
Neg logit contributions
ate-5.601
amed-5.265
 army-4.890
ating-4.824
ined-4.777
Token the
Token ID262
Feature activation-1.91e+00

Pos logit contributions
 syrup+3.530
Pen+3.393
 kil+3.290
 Pen+3.101
 Willow+3.099
Neg logit contributions
ate-3.526
 army-3.277
room-3.021
roc-2.989
ating-2.916
Token top
Token ID1353
Feature activation-2.10e+00

Pos logit contributions
 syrup+3.573
Pen+3.440
 kil+3.321
 Pen+3.317
 Frost+3.300
Neg logit contributions
 army-2.980
ate-2.923
roc-2.667
amed-2.606
room-2.587
Token.
Token ID13
Feature activation+2.97e-01

Pos logit contributions
 syrup+3.505
Pen+3.431
 Frost+3.232
 Pen+3.156
 marsh+3.101
Neg logit contributions
ate-3.686
 army-3.622
room-3.483
ating-3.382
amed-3.241
Token So
Token ID1406
Feature activation+3.89e-01

Pos logit contributions
ate+0.556
 army+0.537
amed+0.512
roc+0.508
ating+0.497
Neg logit contributions
 syrup-0.462
Pen-0.438
 Pen-0.420
 marsh-0.408
 kil-0.397
Token.
Token ID13
Feature activation-1.83e+00

Pos logit contributions
 syrup+1.422
Pen+1.398
 Pen+1.293
 marsh+1.276
 Frost+1.261
Neg logit contributions
ate-1.654
 army-1.554
ating-1.507
amed-1.506
room-1.454
Token
Token ID198
Feature activation-9.55e-01

Pos logit contributions
 syrup+3.777
Pen+3.474
 kil+3.397
 marsh+3.375
 tug+3.351
Neg logit contributions
ate-2.748
amed-2.292
 army-2.285
room-2.280
roc-2.241
Token
Token ID198
Feature activation-1.51e+00

Pos logit contributions
 syrup+1.728
Pen+1.567
 kil+1.521
 marsh+1.511
 Pen+1.491
Neg logit contributions
ate-1.637
 army-1.453
amed-1.450
roc-1.392
room-1.387
TokenMom
Token ID29252
Feature activation-5.07e-01

Pos logit contributions
 syrup+2.534
 marsh+2.227
Pen+2.208
 kil+2.187
 Frost+2.181
Neg logit contributions
ate-2.835
amed-2.586
 army-2.489
ating-2.471
room-2.468
Tokenmy
Token ID1820
Feature activation+2.34e-01

Pos logit contributions
 syrup+1.001
Pen+0.981
 Pen+0.942
 marsh+0.933
 kil+0.917
Neg logit contributions
ate-0.730
 army-0.678
amed-0.672
roc-0.632
ating-0.623
Token and
Token ID290
Feature activation-3.63e+00

Pos logit contributions
ate+0.455
 army+0.438
amed+0.420
roc+0.416
ating+0.406
Neg logit contributions
 syrup-0.350
Pen-0.330
 Pen-0.315
 marsh-0.308
 kil-0.298
Token Daddy
Token ID26926
Feature activation-8.19e-01

Pos logit contributions
 marsh+7.570
Pen+7.487
 kil+7.457
 syrup+7.411
fy+7.166
Neg logit contributions
ate-5.157
amed-4.457
apped-4.278
 army-4.277
roc-4.215
Token then
Token ID788
Feature activation-1.87e-01

Pos logit contributions
 syrup+1.227
Pen+1.214
 Pen+1.124
 marsh+1.105
 kil+1.087
Neg logit contributions
ate-1.554
 army-1.533
amed-1.510
room-1.414
roc-1.408
Token told
Token ID1297
Feature activation-2.06e+00

Pos logit contributions
 syrup+0.280
Pen+0.268
 Pen+0.251
 marsh+0.248
 kil+0.242
Neg logit contributions
ate-0.364
 army-0.351
amed-0.338
roc-0.331
ating-0.323
Token her
Token ID607
Feature activation-6.28e-01

Pos logit contributions
 syrup+3.917
 kil+3.840
Pen+3.835
 marsh+3.558
 sled+3.536
Neg logit contributions
ate-3.091
room-2.772
 army-2.770
roc-2.724
amed-2.683
Token that
Token ID326
Feature activation-5.93e-01

Pos logit contributions
 syrup+0.957
Pen+0.926
 Pen+0.853
 marsh+0.847
 kil+0.836
Neg logit contributions
ate-1.211
 army-1.154
amed-1.098
roc-1.091
ating-1.076
Token shelf
Token ID18316
Feature activation-2.98e-01

Pos logit contributions
 syrup+4.971
Pen+4.953
 Megan+4.684
 Pen+4.675
 Frost+4.672
Neg logit contributions
 army-4.280
ate-3.874
room-3.685
amed-3.454
roc-3.354
Token.
Token ID13
Feature activation+1.80e-01

Pos logit contributions
 syrup+0.419
Pen+0.399
 Pen+0.372
 marsh+0.366
 Willow+0.353
Neg logit contributions
ate-0.607
 army-0.589
amed-0.564
roc-0.552
ating-0.546
Token He
Token ID679
Feature activation-1.04e+00

Pos logit contributions
ate+0.315
 army+0.301
amed+0.288
roc+0.285
ating+0.277
Neg logit contributions
 syrup-0.307
Pen-0.291
 Pen-0.280
 marsh-0.274
 kil-0.267
Token climbed
Token ID19952
Feature activation-2.16e+00

Pos logit contributions
 syrup+1.741
Pen+1.678
 Pen+1.571
 marsh+1.559
 kil+1.528
Neg logit contributions
ate-1.857
 army-1.747
amed-1.730
roc-1.642
ating-1.605
Token up
Token ID510
Feature activation-1.70e+00

Pos logit contributions
 syrup+4.294
Pen+4.182
 kil+3.995
 Pen+3.795
 tug+3.765
Neg logit contributions
ate-3.449
amed-3.088
ating-2.937
 army-2.820
room-2.732
Token the
Token ID262
Feature activation-3.62e+00

Pos logit contributions
 syrup+3.429
Pen+3.304
 kil+3.147
 marsh+3.026
 Pen+2.977
Neg logit contributions
ate-2.574
 army-2.247
amed-2.188
ating-2.187
room-2.158
Token chair
Token ID5118
Feature activation+9.07e-02

Pos logit contributions
 syrup+5.610
 Frost+5.481
Pen+5.269
 dev+5.231
 marsh+5.028
Neg logit contributions
ate-7.131
 army-6.875
ating-6.527
room-6.478
amed-6.313
Token,
Token ID11
Feature activation-1.51e+00

Pos logit contributions
ate+0.146
 army+0.140
amed+0.133
roc+0.131
ating+0.127
Neg logit contributions
 syrup-0.165
Pen-0.158
 Pen-0.152
 marsh-0.149
 kil-0.146
Token and
Token ID290
Feature activation-1.18e+00

Pos logit contributions
 syrup+2.693
Pen+2.640
 kil+2.444
 marsh+2.441
 Pen+2.416
Neg logit contributions
ate-2.479
amed-2.371
 army-2.350
ating-2.167
ined-2.060
Token he
Token ID339
Feature activation-1.23e+00

Pos logit contributions
 syrup+1.725
Pen+1.653
 kil+1.510
 Pen+1.495
 marsh+1.483
Neg logit contributions
ate-2.349
 army-2.224
amed-2.223
ating-2.095
roc-2.090
Token could
Token ID714
Feature activation-1.76e+00

Pos logit contributions
 syrup+1.954
Pen+1.884
 Pen+1.735
 marsh+1.716
 kil+1.700
Neg logit contributions
ate-2.298
amed-2.169
 army-2.166
roc-2.022
ating-2.000
Token rolled
Token ID11686
Feature activation-7.43e-01

Pos logit contributions
 syrup+2.511
Pen+2.460
 Pen+2.375
 kil+2.323
 Willow+2.290
Neg logit contributions
ate-2.280
 army-2.185
amed-2.173
ating-2.022
room-1.998
Token down
Token ID866
Feature activation-7.60e-01

Pos logit contributions
 syrup+1.315
Pen+1.254
 Pen+1.176
 Willow+1.137
 kil+1.137
Neg logit contributions
ate-1.281
 army-1.218
amed-1.168
roc-1.126
ating-1.123
Token the
Token ID262
Feature activation-8.77e-01

Pos logit contributions
 syrup+1.295
Pen+1.227
 Pen+1.133
 kil+1.115
 marsh+1.104
Neg logit contributions
ate-1.360
 army-1.290
amed-1.220
roc-1.199
ating-1.195
Token hill
Token ID12788
Feature activation+2.98e-01

Pos logit contributions
 syrup+1.337
Pen+1.241
 Pen+1.184
 Willow+1.156
 kil+1.142
Neg logit contributions
ate-1.696
 army-1.649
amed-1.543
ating-1.524
roc-1.513
Token.
Token ID13
Feature activation-1.87e+00

Pos logit contributions
ate+0.626
 army+0.603
amed+0.581
roc+0.577
ating+0.563
Neg logit contributions
 syrup-0.397
Pen-0.371
 Pen-0.353
 marsh-0.344
 Frost-0.332
Token When
Token ID1649
Feature activation-3.62e+00

Pos logit contributions
 syrup+3.898
Pen+3.711
 kil+3.645
 tug+3.525
 Pen+3.429
Neg logit contributions
ate-2.596
 army-2.278
amed-2.244
room-2.145
roc-2.090
Token it
Token ID340
Feature activation-6.27e-01

Pos logit contributions
 kil+6.772
 syrup+6.585
fy+6.367
Pen+6.150
 pounds+5.864
Neg logit contributions
ate-5.886
 army-5.472
ating-5.148
ap-4.983
 nation-4.968
Token was
Token ID373
Feature activation-2.36e-01

Pos logit contributions
 syrup+1.064
Pen+1.012
 Pen+0.956
 kil+0.947
 marsh+0.935
Neg logit contributions
ate-1.106
 army-1.035
amed-1.031
ating-0.984
roc-0.970
Token time
Token ID640
Feature activation-1.67e+00

Pos logit contributions
 syrup+0.315
Pen+0.300
 Pen+0.284
 kil+0.271
 marsh+0.269
Neg logit contributions
ate-0.497
 army-0.477
amed-0.466
roc-0.450
ating-0.444
Token to
Token ID284
Feature activation+1.25e-01

Pos logit contributions
Pen+2.706
 syrup+2.661
 kil+2.457
 marsh+2.418
 Pen+2.398
Neg logit contributions
ate-3.048
 army-2.980
amed-2.979
ating-2.737
apped-2.717
Token go
Token ID467
Feature activation-3.14e-01

Pos logit contributions
ate+0.265
 army+0.255
amed+0.245
roc+0.243
ating+0.238
Neg logit contributions
 syrup-0.168
Pen-0.158
 Pen-0.149
 marsh-0.145
 kil-0.141
Token sit
Token ID1650
Feature activation+1.56e-01

Pos logit contributions
ate+0.033
 army+0.032
amed+0.030
roc+0.030
ating+0.029
Neg logit contributions
 syrup-0.033
Pen-0.031
 Pen-0.030
 marsh-0.029
 kil-0.029
Token down
Token ID866
Feature activation-5.13e-01

Pos logit contributions
ate+0.318
 army+0.306
amed+0.295
roc+0.292
ating+0.286
Neg logit contributions
 syrup-0.220
Pen-0.207
 Pen-0.197
 marsh-0.192
 kil-0.186
Token and
Token ID290
Feature activation-1.12e+00

Pos logit contributions
 syrup+0.878
Pen+0.835
 Pen+0.789
 marsh+0.773
 kil+0.770
Neg logit contributions
ate-0.888
 army-0.846
amed-0.798
ating-0.777
roc-0.769
Token rest
Token ID1334
Feature activation-2.77e-01

Pos logit contributions
 syrup+1.531
Pen+1.469
 kil+1.349
 Pen+1.342
 marsh+1.293
Neg logit contributions
ate-2.306
 army-2.217
amed-2.074
roc-2.031
ating-2.031
Token for
Token ID329
Feature activation-1.50e+00

Pos logit contributions
 syrup+0.406
Pen+0.383
 Pen+0.360
 marsh+0.351
 kil+0.347
Neg logit contributions
ate-0.547
 army-0.525
amed-0.506
ating-0.493
roc-0.488
Token a
Token ID257
Feature activation-3.59e+00

Pos logit contributions
 syrup+2.588
Pen+2.481
 Pen+2.330
 kil+2.308
 marsh+2.267
Neg logit contributions
ate-2.560
 army-2.444
amed-2.215
ating-2.195
roc-2.123
Token bit
Token ID1643
Feature activation-1.80e+00

Pos logit contributions
 syrup+5.973
 kil+5.591
Pen+5.549
 Pen+5.384
 Megan+5.180
Neg logit contributions
 army-6.291
 alien-5.694
 nation-5.691
ap-5.581
ate-5.552
Token."
Token ID526
Feature activation-9.56e-02

Pos logit contributions
Pen+3.046
 syrup+3.021
 Pen+2.878
 kil+2.738
 Willow+2.725
Neg logit contributions
ate-3.088
 army-2.932
amed-2.741
ating-2.706
ap-2.610
Token
Token ID198
Feature activation-1.75e-01

Pos logit contributions
 syrup+0.151
Pen+0.142
 Pen+0.135
 marsh+0.134
 kil+0.131
Neg logit contributions
ate-0.177
 army-0.170
amed-0.162
roc-0.161
ating-0.156
Token
Token ID198
Feature activation+6.35e-02

Pos logit contributions
 syrup+0.284
Pen+0.264
 Pen+0.253
 marsh+0.251
 kil+0.246
Neg logit contributions
ate-0.322
 army-0.305
amed-0.294
roc-0.291
ating-0.281
TokenL
Token ID43
Feature activation+7.41e-01

Pos logit contributions
ate+0.111
 army+0.106
amed+0.101
roc+0.100
ating+0.097
Neg logit contributions
 syrup-0.108
Pen-0.102
 Pen-0.098
 marsh-0.096
 kil-0.094
Token a
Token ID257
Feature activation-2.18e+00

Pos logit contributions
 syrup+0.973
Pen+0.953
 Pen+0.901
 kil+0.863
 Willow+0.859
Neg logit contributions
ate-1.248
 army-1.190
amed-1.156
ating-1.106
roc-1.103
Token good
Token ID922
Feature activation-1.39e+00

Pos logit contributions
 syrup+4.213
Pen+4.006
 Pen+3.882
 kil+3.856
 Willow+3.786
Neg logit contributions
 army-3.355
ate-3.235
roc-2.933
 alien-2.806
room-2.764
Token place
Token ID1295
Feature activation-7.14e-01

Pos logit contributions
 syrup+2.728
Pen+2.670
 kil+2.507
 Pen+2.489
 Willow+2.454
Neg logit contributions
ate-2.115
 army-1.997
roc-1.789
room-1.771
ating-1.731
Token to
Token ID284
Feature activation-1.16e+00

Pos logit contributions
 syrup+0.965
Pen+0.949
 Pen+0.866
 marsh+0.836
 Willow+0.825
Neg logit contributions
ate-1.477
 army-1.437
amed-1.358
roc-1.333
ating-1.324
Token take
Token ID1011
Feature activation-1.19e+00

Pos logit contributions
 syrup+1.540
Pen+1.436
 kil+1.343
 Pen+1.334
 Willow+1.290
Neg logit contributions
ate-2.493
 army-2.354
room-2.211
ating-2.195
amed-2.183
Token a
Token ID257
Feature activation-3.56e+00

Pos logit contributions
 syrup+2.052
Pen+2.009
 Pen+1.882
 kil+1.792
 Willow+1.791
Neg logit contributions
ate-2.071
 army-1.972
amed-1.840
roc-1.822
ating-1.818
Token nap
Token ID25422
Feature activation+2.78e-01

Pos logit contributions
 syrup+6.099
Pen+5.866
 kil+5.690
 Pen+5.643
 Rudy+5.325
Neg logit contributions
 army-6.151
ate-6.000
 network-5.716
ap-5.713
 club-5.617
Token.
Token ID13
Feature activation-3.46e-01

Pos logit contributions
ate+0.590
 army+0.569
amed+0.549
roc+0.546
ating+0.532
Neg logit contributions
 syrup-0.367
Pen-0.343
 Pen-0.326
 marsh-0.317
 kil-0.306
Token
Token ID198
Feature activation-6.60e-01

Pos logit contributions
 syrup+0.579
Pen+0.546
 Pen+0.517
 marsh+0.510
 kil+0.502
Neg logit contributions
ate-0.629
 army-0.589
roc-0.568
amed-0.566
ating-0.545
Token
Token ID198
Feature activation-5.74e-01

Pos logit contributions
 syrup+1.156
Pen+1.069
 Pen+1.022
 marsh+1.022
 kil+1.006
Neg logit contributions
ate-1.156
 army-1.062
amed-1.026
roc-1.016
room-0.995
TokenAs
Token ID1722
Feature activation-4.57e-01

Pos logit contributions
 syrup+1.011
Pen+0.950
 Pen+0.929
 marsh+0.899
 Willow+0.887
Neg logit contributions
ate-1.006
 army-0.940
amed-0.923
roc-0.907
ating-0.877
Token to
Token ID284
Feature activation+1.30e+00

Pos logit contributions
 syrup+0.133
Pen+0.126
 Pen+0.118
 marsh+0.114
 kil+0.112
Neg logit contributions
ate-0.237
 army-0.230
roc-0.219
amed-0.219
ating-0.214
Token jump
Token ID4391
Feature activation+2.27e-01

Pos logit contributions
ate+2.597
amed+2.543
 army+2.490
roc+2.453
ating+2.409
Neg logit contributions
 syrup-1.905
Pen-1.719
 marsh-1.679
 Pen-1.657
 tug-1.577
Token.
Token ID13
Feature activation-2.37e+00

Pos logit contributions
ate+0.432
 army+0.415
amed+0.398
roc+0.395
ating+0.385
Neg logit contributions
 syrup-0.351
Pen-0.332
 Pen-0.317
 marsh-0.310
 kil-0.301
Token One
Token ID1881
Feature activation-2.45e-01

Pos logit contributions
 syrup+4.589
 kil+4.022
 tug+4.000
 marsh+3.902
 bit+3.871
Neg logit contributions
ate-4.018
room-3.438
 army-3.393
ap-3.281
roc-3.243
Token day
Token ID1110
Feature activation-2.57e+00

Pos logit contributions
 syrup+0.430
Pen+0.407
 Pen+0.388
 marsh+0.384
 kil+0.379
Neg logit contributions
ate-0.418
 army-0.398
amed-0.380
roc-0.373
ating-0.366
Token,
Token ID11
Feature activation-3.56e+00

Pos logit contributions
 syrup+6.796
Pen+6.303
 kil+6.249
 marsh+5.999
fy+5.962
Neg logit contributions
ate-2.285
ap-1.769
ating-1.760
amed-1.720
room-1.711
Token Megan
Token ID27616
Feature activation-7.18e-02

Pos logit contributions
 syrup+6.635
 kil+6.364
fy+5.827
 marsh+5.725
 spe+5.641
Neg logit contributions
ate-6.421
room-5.330
apped-5.251
ating-5.220
amed-5.210
Token went
Token ID1816
Feature activation-2.18e-01

Pos logit contributions
 syrup+0.113
Pen+0.107
 Pen+0.101
 marsh+0.100
 kil+0.097
Neg logit contributions
ate-0.135
 army-0.129
amed-0.124
roc-0.122
ating-0.119
Token to
Token ID284
Feature activation-5.57e-02

Pos logit contributions
 syrup+0.282
Pen+0.268
 Pen+0.251
 marsh+0.245
 kil+0.239
Neg logit contributions
ate-0.470
 army-0.453
amed-0.438
roc-0.426
ating-0.422
Token the
Token ID262
Feature activation+6.99e-01

Pos logit contributions
 syrup+0.085
Pen+0.080
 Pen+0.076
 marsh+0.075
 kil+0.074
Neg logit contributions
ate-0.108
 army-0.103
amed-0.098
roc-0.098
ating-0.094
Token park
Token ID3952
Feature activation+6.17e-01

Pos logit contributions
ate+0.994
 army+0.940
amed+0.910
roc+0.887
ating+0.869
Neg logit contributions
 syrup-1.421
Pen-1.356
 Pen-1.309
 marsh-1.293
 Frost-1.256
Token excited
Token ID6568
Feature activation-3.03e-01

Pos logit contributions
ate+0.502
 army+0.482
amed+0.461
roc+0.459
ating+0.448
Neg logit contributions
 syrup-0.396
Pen-0.375
 Pen-0.359
 marsh-0.350
 Frost-0.339
Token and
Token ID290
Feature activation-3.89e-01

Pos logit contributions
 syrup+0.473
Pen+0.459
 Pen+0.430
 marsh+0.424
 kil+0.411
Neg logit contributions
ate-0.567
 army-0.549
amed-0.520
roc-0.515
ating-0.508
Token quickly
Token ID2952
Feature activation-6.50e-01

Pos logit contributions
 syrup+0.555
Pen+0.526
 Pen+0.489
 marsh+0.482
 kil+0.480
Neg logit contributions
ate-0.780
 army-0.757
amed-0.732
roc-0.708
ating-0.696
Token scrambled
Token ID34351
Feature activation-1.76e+00

Pos logit contributions
 syrup+0.842
Pen+0.827
 Pen+0.761
 marsh+0.730
 kil+0.723
Neg logit contributions
ate-1.375
 army-1.337
amed-1.303
roc-1.242
ating-1.239
Token up
Token ID510
Feature activation-1.52e+00

Pos logit contributions
Pen+3.186
 syrup+3.154
 kil+2.911
 Pen+2.827
 Willow+2.785
Neg logit contributions
ate-3.035
 army-2.723
amed-2.703
ating-2.626
room-2.481
Token the
Token ID262
Feature activation-3.55e+00

Pos logit contributions
 syrup+2.692
Pen+2.655
 kil+2.423
 Pen+2.352
 Willow+2.345
Neg logit contributions
ate-2.589
 army-2.448
amed-2.296
ating-2.267
room-2.233
Token chair
Token ID5118
Feature activation+5.21e-03

Pos logit contributions
 syrup+5.377
 Frost+5.226
 dev+5.187
Pen+5.005
 nib+4.977
Neg logit contributions
ate-7.101
 army-6.771
ating-6.617
room-6.296
amed-6.288
Token.
Token ID13
Feature activation-6.64e-01

Pos logit contributions
ate+0.011
 army+0.010
amed+0.010
roc+0.010
ating+0.010
Neg logit contributions
 syrup-0.007
Pen-0.007
 Pen-0.006
 marsh-0.006
 kil-0.006
Token
Token ID220
Feature activation+1.89e-01

Pos logit contributions
 syrup+1.070
Pen+1.002
 Pen+0.932
 marsh+0.930
 kil+0.925
Neg logit contributions
ate-1.230
 army-1.168
amed-1.127
roc-1.114
room-1.078
Token
Token ID198
Feature activation-9.98e-02

Pos logit contributions
ate+0.357
 army+0.343
amed+0.329
roc+0.326
ating+0.318
Neg logit contributions
 syrup-0.291
Pen-0.275
 Pen-0.262
 marsh-0.256
 kil-0.249
Token
Token ID198
Feature activation-2.57e-01

Pos logit contributions
 syrup+0.160
Pen+0.150
 Pen+0.143
 marsh+0.141
 kil+0.138
Neg logit contributions
ate-0.185
 army-0.175
amed-0.170
roc-0.167
ating-0.162
Tokenhe
Token ID258
Feature activation-2.84e-01

Pos logit contributions
 syrup+0.731
Pen+0.690
 Pen+0.651
 marsh+0.650
 kil+0.634
Neg logit contributions
ate-0.832
 army-0.788
amed-0.751
room-0.736
ating-0.729
Tokenlf
Token ID1652
Feature activation-4.16e-01

Pos logit contributions
 syrup+0.658
Pen+0.635
 Pen+0.609
 marsh+0.607
 Frost+0.597
Neg logit contributions
ate-0.327
 army-0.301
amed-0.275
ating-0.269
roc-0.265
Token.
Token ID13
Feature activation-4.47e-02

Pos logit contributions
 syrup+0.582
Pen+0.546
 Pen+0.507
 marsh+0.506
 kil+0.487
Neg logit contributions
ate-0.860
 army-0.828
amed-0.787
ating-0.773
roc-0.771
Token She
Token ID1375
Feature activation-1.91e+00

Pos logit contributions
 syrup+0.083
Pen+0.079
 Pen+0.075
 marsh+0.075
 kil+0.073
Neg logit contributions
ate-0.072
 army-0.069
amed-0.065
roc-0.065
ating-0.062
Token pushes
Token ID20070
Feature activation-1.23e+00

Pos logit contributions
 syrup+3.071
Pen+2.995
 marsh+2.751
 Pen+2.745
 Frost+2.670
Neg logit contributions
ate-3.720
 army-3.480
amed-3.212
roc-3.144
 nations-3.073
Token the
Token ID262
Feature activation-3.54e+00

Pos logit contributions
 syrup+2.009
Pen+1.942
 marsh+1.843
 Pen+1.824
 kil+1.781
Neg logit contributions
ate-2.258
amed-2.060
 army-2.048
room-1.998
roc-1.987
Token chair
Token ID5118
Feature activation-6.63e-01

Pos logit contributions
 Frost+5.746
 syrup+5.653
 dev+5.446
Pen+5.305
 spe+5.200
Neg logit contributions
 army-6.615
ate-6.439
room-6.173
amed-6.138
ating-5.750
Token to
Token ID284
Feature activation-1.74e+00

Pos logit contributions
 syrup+1.028
Pen+0.985
 Pen+0.921
 marsh+0.911
 Frost+0.878
Neg logit contributions
ate-1.264
 army-1.213
amed-1.170
roc-1.132
room-1.127
Token the
Token ID262
Feature activation-2.76e+00

Pos logit contributions
 syrup+3.096
Pen+2.895
 kil+2.777
 marsh+2.763
 Willow+2.664
Neg logit contributions
ate-3.056
 army-2.806
room-2.699
roc-2.569
amed-2.563
Token books
Token ID3835
Feature activation-2.25e-01

Pos logit contributions
 syrup+5.064
 Frost+4.871
Pen+4.768
 marsh+4.573
 Pen+4.542
Neg logit contributions
 army-4.629
ate-4.355
room-4.125
amed-3.999
ating-3.762
Tokenhe
Token ID258
Feature activation-5.76e-01

Pos logit contributions
 syrup+0.389
Pen+0.371
 Pen+0.353
 marsh+0.350
 kil+0.341
Neg logit contributions
ate-0.387
 army-0.368
amed-0.352
roc-0.340
ating-0.338
Token the
Token ID262
Feature activation-1.11e+00

Pos logit contributions
 syrup+1.965
Pen+1.893
 kil+1.784
 Pen+1.752
 Frost+1.741
Neg logit contributions
ate-1.906
 army-1.827
amed-1.679
room-1.658
ating-1.642
Token covers
Token ID8698
Feature activation+8.45e-02

Pos logit contributions
 syrup+1.861
Pen+1.822
 Pen+1.740
 kil+1.715
 tug+1.696
Neg logit contributions
ate-1.868
 army-1.822
amed-1.670
roc-1.630
 alien-1.606
Token,
Token ID11
Feature activation-4.04e-01

Pos logit contributions
ate+0.179
 army+0.172
amed+0.166
roc+0.164
ating+0.162
Neg logit contributions
 syrup-0.112
Pen-0.105
 Pen-0.100
 marsh-0.097
 kil-0.094
Token ready
Token ID3492
Feature activation-1.33e+00

Pos logit contributions
 syrup+0.635
Pen+0.611
 Pen+0.575
 marsh+0.559
 kil+0.559
Neg logit contributions
ate-0.749
 army-0.717
amed-0.690
roc-0.666
ating-0.662
Token for
Token ID329
Feature activation-1.42e+00

Pos logit contributions
 syrup+1.911
Pen+1.892
 Pen+1.720
 kil+1.683
 Frost+1.666
Neg logit contributions
ate-2.660
 army-2.505
amed-2.427
ating-2.392
roc-2.302
Token a
Token ID257
Feature activation-3.53e+00

Pos logit contributions
 syrup+2.428
Pen+2.389
 Pen+2.212
 kil+2.206
 marsh+2.123
Neg logit contributions
ate-2.451
 army-2.340
ating-2.113
amed-2.107
room-2.057
Token cos
Token ID8615
Feature activation-1.18e-01

Pos logit contributions
 syrup+6.540
Pen+6.167
 kil+6.072
 Megan+5.998
 Pen+5.944
Neg logit contributions
 army-5.812
 network-5.192
ate-5.152
 alien-5.088
 nation-5.088
Tokeny
Token ID88
Feature activation-1.83e+00

Pos logit contributions
 syrup+0.238
Pen+0.228
 Pen+0.220
 marsh+0.218
 kil+0.213
Neg logit contributions
ate-0.167
 army-0.157
amed-0.148
roc-0.146
ating-0.142
Token sleep
Token ID3993
Feature activation-3.98e-01

Pos logit contributions
 syrup+3.321
Pen+3.229
 kil+3.134
 Pen+3.026
 Frost+2.974
Neg logit contributions
 army-2.881
ate-2.754
room-2.508
 alien-2.481
ating-2.425
Token.
Token ID13
Feature activation-1.19e+00

Pos logit contributions
 syrup+0.545
Pen+0.516
 Pen+0.481
 marsh+0.472
 Frost+0.463
Neg logit contributions
ate-0.821
 army-0.798
amed-0.763
ating-0.750
roc-0.738
Token
Token ID220
Feature activation-5.76e-02

Pos logit contributions
 syrup+2.121
Pen+1.969
 kil+1.962
 tug+1.903
 Pen+1.855
Neg logit contributions
ate-1.997
 army-1.813
amed-1.725
room-1.700
roc-1.689
Token's
Token ID338
Feature activation-8.17e-01

Pos logit contributions
ate+0.940
 army+0.899
amed+0.858
roc+0.848
ating+0.827
Neg logit contributions
 syrup-0.919
Pen-0.875
 Pen-0.843
 marsh-0.820
 Willow-0.800
Token sit
Token ID1650
Feature activation+5.75e-01

Pos logit contributions
 syrup+1.176
Pen+1.121
 Pen+1.049
 kil+1.012
 marsh+1.001
Neg logit contributions
ate-1.674
 army-1.595
roc-1.478
amed-1.469
room-1.456
Token down
Token ID866
Feature activation-6.68e-01

Pos logit contributions
ate+1.221
 army+1.189
roc+1.144
amed+1.141
ating+1.103
Neg logit contributions
 syrup-0.753
Pen-0.707
 Pen-0.670
 marsh-0.649
 Frost-0.636
Token and
Token ID290
Feature activation-1.42e+00

Pos logit contributions
 syrup+1.229
Pen+1.166
 Pen+1.108
 marsh+1.098
 kil+1.095
Neg logit contributions
ate-1.089
 army-1.021
amed-0.965
ating-0.940
room-0.924
Token take
Token ID1011
Feature activation-1.04e+00

Pos logit contributions
 syrup+1.958
Pen+1.848
 Pen+1.718
 kil+1.715
 marsh+1.638
Neg logit contributions
ate-2.922
 army-2.819
amed-2.591
roc-2.580
ating-2.530
Token a
Token ID257
Feature activation-3.52e+00

Pos logit contributions
 syrup+1.738
Pen+1.678
 Pen+1.591
 kil+1.575
 marsh+1.543
Neg logit contributions
ate-1.830
 army-1.715
amed-1.658
roc-1.589
ating-1.570
Token break
Token ID2270
Feature activation-7.17e-01

Pos logit contributions
 syrup+5.228
Pen+5.197
 Megan+5.050
 Pen+5.041
 kil+4.829
Neg logit contributions
 army-6.480
ate-6.119
ully-5.847
 alien-5.736
ap-5.616
Token."
Token ID526
Feature activation-1.29e-01

Pos logit contributions
 syrup+0.957
Pen+0.903
 Pen+0.846
 marsh+0.821
 tug+0.800
Neg logit contributions
ate-1.529
 army-1.445
amed-1.390
ating-1.379
roc-1.364
Token They
Token ID1119
Feature activation-2.11e+00

Pos logit contributions
 syrup+0.222
Pen+0.213
 Pen+0.203
 marsh+0.199
 kil+0.198
Neg logit contributions
ate-0.222
 army-0.210
amed-0.202
roc-0.199
ating-0.193
Token found
Token ID1043
Feature activation-7.96e-02

Pos logit contributions
Pen+3.382
 syrup+3.355
 kil+3.288
 Pen+3.127
 marsh+3.015
Neg logit contributions
ate-3.797
amed-3.712
 army-3.605
ully-3.307
room-3.282
Token a
Token ID257
Feature activation-6.32e-01

Pos logit contributions
 syrup+0.129
Pen+0.122
 Pen+0.116
 marsh+0.114
 kil+0.113
Neg logit contributions
ate-0.145
 army-0.139
amed-0.133
roc-0.131
ating-0.128
Token so
Token ID523
Feature activation+7.54e-01

Pos logit contributions
 syrup+0.417
Pen+0.393
 Pen+0.372
 marsh+0.367
 kil+0.357
Neg logit contributions
ate-0.517
 army-0.496
amed-0.488
roc-0.466
ating-0.457
Token happy
Token ID3772
Feature activation-4.72e-01

Pos logit contributions
ate+1.301
 army+1.264
roc+1.188
amed+1.169
ating+1.157
Neg logit contributions
 syrup-1.294
Pen-1.249
 Pen-1.205
 marsh-1.173
 Frost-1.157
Token that
Token ID326
Feature activation-3.40e-01

Pos logit contributions
 syrup+0.807
Pen+0.783
 Pen+0.731
 marsh+0.722
 Willow+0.699
Neg logit contributions
ate-0.836
 army-0.796
roc-0.743
amed-0.741
ating-0.741
Token he
Token ID339
Feature activation-6.61e-01

Pos logit contributions
 syrup+0.408
Pen+0.387
 Pen+0.353
 marsh+0.340
 kil+0.330
Neg logit contributions
ate-0.774
 army-0.753
amed-0.710
roc-0.708
ating-0.699
Token reached
Token ID4251
Feature activation-2.33e+00

Pos logit contributions
 syrup+0.885
Pen+0.838
 Pen+0.785
 marsh+0.779
 kil+0.748
Neg logit contributions
ate-1.398
 army-1.347
amed-1.332
roc-1.267
ating-1.259
Token the
Token ID262
Feature activation-3.50e+00

Pos logit contributions
Pen+4.489
 syrup+4.442
 spe+4.062
 marsh+4.023
 Willow+3.976
Neg logit contributions
ate-3.919
 army-3.601
ating-3.175
amed-3.115
room-3.019
Token popcorn
Token ID38914
Feature activation-9.95e-01

Pos logit contributions
Pen+6.591
 Lu+6.418
 dev+6.365
 Willow+6.302
 Megan+6.300
Neg logit contributions
 army-6.015
ate-5.633
room-5.204
ating-4.868
amed-4.775
Token.
Token ID13
Feature activation+2.40e-01

Pos logit contributions
 syrup+1.766
Pen+1.742
 Pen+1.610
 marsh+1.589
 kil+1.579
Neg logit contributions
ate-1.644
 army-1.584
room-1.454
amed-1.437
ating-1.433
Token But
Token ID887
Feature activation+3.80e-01

Pos logit contributions
ate+0.402
 army+0.384
amed+0.366
roc+0.362
ating+0.352
Neg logit contributions
 syrup-0.428
Pen-0.407
 Pen-0.392
 marsh-0.384
 kil-0.375
Token then
Token ID788
Feature activation+6.46e-01

Pos logit contributions
ate+0.833
 army+0.804
amed+0.778
roc+0.771
ating+0.753
Neg logit contributions
 syrup-0.478
Pen-0.444
 Pen-0.422
 marsh-0.410
 Frost-0.398
Token,
Token ID11
Feature activation-2.05e-01

Pos logit contributions
ate+1.167
 army+1.103
amed+1.079
roc+1.050
ating+1.031
Neg logit contributions
 syrup-1.068
Pen-0.997
 Pen-0.972
 marsh-0.959
 Frost-0.943
Token now
Token ID783
Feature activation-5.82e-01

Pos logit contributions
 syrup+2.441
Pen+2.373
 Pen+2.187
 marsh+2.185
 Willow+2.173
Neg logit contributions
ate-2.494
amed-2.462
 army-2.387
ating-2.127
ined-2.123
Token tall
Token ID7331
Feature activation-5.54e-01

Pos logit contributions
 syrup+0.917
Pen+0.885
 Pen+0.829
 marsh+0.809
 Willow+0.803
Neg logit contributions
ate-1.087
 army-1.055
amed-1.013
roc-0.961
ating-0.956
Token enough
Token ID1576
Feature activation-1.51e+00

Pos logit contributions
 syrup+1.014
Pen+0.995
 Pen+0.938
 marsh+0.918
 Willow+0.906
Neg logit contributions
ate-0.881
 army-0.842
amed-0.782
ating-0.777
roc-0.774
Token to
Token ID284
Feature activation-1.62e+00

Pos logit contributions
Pen+2.492
 syrup+2.430
 Pen+2.249
 Willow+2.233
 marsh+2.148
Neg logit contributions
ate-2.745
 army-2.641
amed-2.436
ating-2.436
room-2.371
Token reach
Token ID3151
Feature activation-1.83e+00

Pos logit contributions
 syrup+2.108
Pen+1.999
 Pen+1.841
 Willow+1.835
 marsh+1.783
Neg logit contributions
ate-3.570
 army-3.366
room-3.233
ating-3.174
roc-3.140
Token the
Token ID262
Feature activation-3.49e+00

Pos logit contributions
Pen+3.226
 syrup+3.114
 Pen+2.909
 Willow+2.794
 tug+2.733
Neg logit contributions
ate-3.286
 army-2.980
ating-2.799
roc-2.717
room-2.695
Token mixer
Token ID33938
Feature activation-4.40e-01

Pos logit contributions
Pen+6.970
 Frost+6.766
 Pen+6.513
 Willow+6.437
 nib+6.380
Neg logit contributions
ate-5.498
room-5.386
 army-5.346
ating-4.919
ini-4.892
Token.
Token ID13
Feature activation-2.63e-01

Pos logit contributions
 syrup+0.668
Pen+0.644
 Pen+0.602
 marsh+0.592
 Willow+0.574
Neg logit contributions
ate-0.849
 army-0.814
amed-0.771
roc-0.759
ating-0.757
Token He
Token ID679
Feature activation-1.58e+00

Pos logit contributions
 syrup+0.478
Pen+0.459
 Pen+0.436
 marsh+0.430
 Willow+0.418
Neg logit contributions
ate-0.437
 army-0.411
amed-0.393
roc-0.390
room-0.372
Token smiled
Token ID13541
Feature activation+8.66e-01

Pos logit contributions
 syrup+2.477
Pen+2.393
 marsh+2.242
 Pen+2.193
 kil+2.191
Neg logit contributions
ate-3.036
amed-2.834
 army-2.814
roc-2.642
room-2.605
Token and
Token ID290
Feature activation-9.38e-02

Pos logit contributions
ate+1.586
 army+1.532
roc+1.483
amed+1.460
ating+1.431
Neg logit contributions
 syrup-1.385
Pen-1.296
 Pen-1.264
 Frost-1.228
 marsh-1.224
Token on
Token ID319
Feature activation-8.20e-01

Pos logit contributions
 syrup+3.789
Pen+3.665
 kil+3.483
 Pen+3.332
 dev+3.323
Neg logit contributions
ate-3.335
ating-2.851
 army-2.816
amed-2.813
room-2.777
Token the
Token ID262
Feature activation-2.71e+00

Pos logit contributions
 syrup+1.316
Pen+1.252
 Pen+1.158
 kil+1.149
 marsh+1.135
Neg logit contributions
ate-1.544
 army-1.444
roc-1.369
amed-1.359
room-1.358
Token chair
Token ID5118
Feature activation-1.90e-01

Pos logit contributions
 syrup+4.336
Pen+4.220
 dev+4.004
 Frost+3.958
 spe+3.932
Neg logit contributions
ate-4.941
 army-4.782
room-4.523
amed-4.308
ating-4.304
Token and
Token ID290
Feature activation-2.57e+00

Pos logit contributions
 syrup+0.340
Pen+0.325
 Pen+0.308
 marsh+0.305
 kil+0.298
Neg logit contributions
ate-0.316
 army-0.302
amed-0.286
roc-0.281
ating-0.276
Token grab
Token ID5552
Feature activation-1.75e+00

Pos logit contributions
 syrup+3.951
Pen+3.864
 kil+3.456
 Pen+3.416
 marsh+3.357
Neg logit contributions
ate-5.025
 army-4.808
ap-4.263
amed-4.245
room-4.197
Token the
Token ID262
Feature activation-3.49e+00

Pos logit contributions
Pen+3.087
 syrup+3.081
 Pen+2.794
 Willow+2.763
 nearly+2.702
Neg logit contributions
ate-3.129
 army-2.741
roc-2.674
amed-2.668
room-2.654
Token book
Token ID1492
Feature activation-1.10e+00

Pos logit contributions
Pen+6.389
 dev+6.359
 syrup+6.084
 nearly+5.974
 Lu+5.961
Neg logit contributions
 army-5.794
ate-5.763
room-5.309
amed-4.784
ating-4.749
Token.
Token ID13
Feature activation+1.72e-01

Pos logit contributions
 syrup+1.764
Pen+1.676
 marsh+1.561
 Pen+1.544
 Willow+1.493
Neg logit contributions
ate-2.125
 army-1.969
amed-1.895
room-1.879
ating-1.829
Token They
Token ID1119
Feature activation-2.43e+00

Pos logit contributions
ate+0.283
 army+0.269
amed+0.256
roc+0.254
ating+0.246
Neg logit contributions
 syrup-0.312
Pen-0.297
 Pen-0.286
 marsh-0.280
 kil-0.274
Token are
Token ID389
Feature activation-1.28e+00

Pos logit contributions
 syrup+3.864
Pen+3.815
 Pen+3.501
 kil+3.472
 Willow+3.445
Neg logit contributions
ate-4.548
 army-4.371
room-3.992
roc-3.938
amed-3.857
Token happy
Token ID3772
Feature activation-8.17e-01

Pos logit contributions
 syrup+2.276
Pen+2.227
 Pen+2.076
 Willow+2.052
 marsh+2.023
Neg logit contributions
ate-2.109
 army-2.039
amed-2.032
roc-1.793
ating-1.792
Token top
Token ID1353
Feature activation-4.55e-01

Pos logit contributions
 syrup+2.387
Pen+2.268
 Pen+2.193
 Willow+2.128
 kil+2.126
Neg logit contributions
 army-1.745
ate-1.728
roc-1.519
amed-1.510
ating-1.440
Token of
Token ID286
Feature activation-1.84e+00

Pos logit contributions
 syrup+0.781
Pen+0.748
 Pen+0.712
 marsh+0.693
 Frost+0.687
Neg logit contributions
ate-0.787
 army-0.755
amed-0.708
ating-0.698
roc-0.695
Token the
Token ID262
Feature activation-1.91e+00

Pos logit contributions
 syrup+3.401
Pen+3.189
fy+3.115
 kil+3.075
 nearly+3.002
Neg logit contributions
ate-3.002
 army-2.851
room-2.469
ating-2.458
roc-2.454
Token hill
Token ID12788
Feature activation+6.87e-01

Pos logit contributions
 syrup+3.485
Pen+3.273
 Pen+3.188
 Willow+3.114
 Megan+3.108
Neg logit contributions
ate-3.168
 army-3.160
ating-2.744
roc-2.725
amed-2.699
Token.
Token ID13
Feature activation-2.27e+00

Pos logit contributions
ate+1.441
 army+1.387
roc+1.344
amed+1.340
ating+1.292
Neg logit contributions
 syrup-0.910
Pen-0.850
 Pen-0.821
 marsh-0.794
 Frost-0.767
Token When
Token ID1649
Feature activation-3.48e+00

Pos logit contributions
 syrup+5.259
Pen+4.904
 kil+4.825
 tug+4.755
 nearly+4.610
Neg logit contributions
ate-2.829
 army-2.370
room-2.310
apped-2.178
amed-2.173
Token they
Token ID484
Feature activation-1.46e+00

Pos logit contributions
 syrup+6.392
 kil+6.284
Pen+6.003
fy+5.971
ches+5.577
Neg logit contributions
ate-5.826
 army-5.259
ating-5.112
ap-4.824
room-4.814
Token got
Token ID1392
Feature activation-1.13e+00

Pos logit contributions
 syrup+1.996
Pen+1.930
 Pen+1.751
 marsh+1.734
 kil+1.715
Neg logit contributions
ate-3.077
 army-2.963
amed-2.927
roc-2.753
room-2.739
Token there
Token ID612
Feature activation-3.99e-02

Pos logit contributions
 syrup+1.850
Pen+1.744
 kil+1.637
 Pen+1.613
 marsh+1.596
Neg logit contributions
ate-2.133
 army-2.007
amed-1.997
ating-1.853
roc-1.845
Token they
Token ID484
Feature activation-2.45e-01

Pos logit contributions
 syrup+0.086
Pen+0.082
 Pen+0.079
 marsh+0.078
 kil+0.077
Neg logit contributions
ate-0.052
 army-0.049
amed-0.046
roc-0.045
ating-0.044
Token saw
Token ID2497
Feature activation-6.91e-01

Pos logit contributions
 syrup+0.415
Pen+0.398
 Pen+0.377
 marsh+0.369
 kil+0.362
Neg logit contributions
ate-0.432
 army-0.415
amed-0.397
roc-0.386
ating-0.377
Token and
Token ID290
Feature activation-4.92e-01

Pos logit contributions
ate+1.397
 army+1.325
roc+1.285
amed+1.266
ating+1.236
Neg logit contributions
 syrup-1.273
Pen-1.199
 Pen-1.171
 marsh-1.143
 Frost-1.128
Token pretended
Token ID36766
Feature activation-6.99e-01

Pos logit contributions
 syrup+0.733
Pen+0.700
 Pen+0.650
 kil+0.648
 marsh+0.639
Neg logit contributions
ate-0.944
 army-0.913
amed-0.888
roc-0.865
ating-0.844
Token for
Token ID329
Feature activation-1.58e+00

Pos logit contributions
 syrup+0.938
Pen+0.888
 kil+0.806
 Pen+0.802
 marsh+0.796
Neg logit contributions
ate-1.454
 army-1.413
amed-1.360
roc-1.335
ating-1.330
Token hours
Token ID2250
Feature activation-5.10e-01

Pos logit contributions
 syrup+2.797
Pen+2.579
 kil+2.475
 nearly+2.398
 marsh+2.367
Neg logit contributions
ate-2.681
 army-2.539
roc-2.389
room-2.342
ating-2.325
Token.
Token ID13
Feature activation-1.96e+00

Pos logit contributions
 syrup+0.903
Pen+0.868
 Pen+0.813
 marsh+0.796
 kil+0.794
Neg logit contributions
ate-0.842
 army-0.808
amed-0.775
ating-0.759
room-0.743
Token When
Token ID1649
Feature activation-3.47e+00

Pos logit contributions
 syrup+4.147
Pen+3.843
 kil+3.814
 tug+3.708
 nearly+3.586
Neg logit contributions
ate-2.762
 army-2.343
amed-2.303
room-2.288
roc-2.266
Token it
Token ID340
Feature activation+3.51e-01

Pos logit contributions
 kil+6.613
 syrup+6.579
Pen+5.985
fy+5.969
ches+5.864
Neg logit contributions
ate-5.808
ating-5.133
ap-4.897
 army-4.838
room-4.738
Token was
Token ID373
Feature activation+7.42e-01

Pos logit contributions
ate+0.672
 army+0.647
amed+0.619
roc+0.616
ating+0.597
Neg logit contributions
 syrup-0.539
Pen-0.508
 Pen-0.488
 marsh-0.477
 Frost-0.463
Token time
Token ID640
Feature activation-1.01e+00

Pos logit contributions
ate+1.691
 army+1.631
roc+1.588
amed+1.557
ating+1.541
Neg logit contributions
 syrup-0.883
Pen-0.797
 marsh-0.764
 Pen-0.759
 Frost-0.714
Token to
Token ID284
Feature activation+8.65e-01

Pos logit contributions
 syrup+1.471
Pen+1.448
 marsh+1.317
 kil+1.304
 Pen+1.297
Neg logit contributions
ate-1.970
 army-1.930
amed-1.894
ating-1.788
roc-1.766
Token go
Token ID467
Feature activation+9.40e-01

Pos logit contributions
ate+1.885
 army+1.837
amed+1.815
roc+1.777
ating+1.740
Neg logit contributions
 syrup-1.083
Pen-0.983
 Pen-0.942
 marsh-0.936
 Frost-0.899
Token chair
Token ID5118
Feature activation+4.03e-01

Pos logit contributions
 syrup+2.336
Pen+2.220
 Pen+2.117
 Frost+2.088
 Willow+2.071
Neg logit contributions
 army-2.702
ate-2.603
room-2.342
amed-2.322
 alien-2.301
Token so
Token ID523
Feature activation-1.46e+00

Pos logit contributions
ate+0.822
 army+0.788
amed+0.762
roc+0.759
ating+0.738
Neg logit contributions
 syrup-0.564
Pen-0.528
 Pen-0.506
 marsh-0.490
 Frost-0.475
Token she
Token ID673
Feature activation-1.66e+00

Pos logit contributions
 syrup+2.128
Pen+1.918
 kil+1.887
 marsh+1.784
 Pen+1.757
Neg logit contributions
ate-2.935
 army-2.760
amed-2.653
ating-2.649
roc-2.616
Token could
Token ID714
Feature activation-1.94e+00

Pos logit contributions
 syrup+2.323
Pen+2.237
 kil+2.065
 Pen+2.049
 marsh+2.030
Neg logit contributions
ate-3.507
 army-3.240
amed-3.183
ating-2.980
roc-2.964
Token reach
Token ID3151
Feature activation-2.12e+00

Pos logit contributions
 syrup+3.011
Pen+2.865
 kil+2.662
 Pen+2.598
 marsh+2.589
Neg logit contributions
ate-3.922
 army-3.572
room-3.407
ap-3.283
ating-3.210
Token the
Token ID262
Feature activation-3.47e+00

Pos logit contributions
Pen+3.833
 syrup+3.650
 Pen+3.482
 kil+3.467
 tug+3.346
Neg logit contributions
ate-3.660
 army-3.183
ating-2.972
amed-2.887
room-2.853
Token onion
Token ID21670
Feature activation-1.11e+00

Pos logit contributions
Pen+6.229
 Frost+6.072
 dev+5.975
 Willow+5.873
 Megan+5.869
Neg logit contributions
 army-5.976
ate-5.794
room-5.458
 office-5.118
ating-4.963
Token on
Token ID319
Feature activation-7.82e-01

Pos logit contributions
 syrup+1.687
Pen+1.636
 Pen+1.503
 marsh+1.494
 Willow+1.462
Neg logit contributions
ate-2.156
 army-2.068
ating-1.918
amed-1.906
room-1.892
Token the
Token ID262
Feature activation-1.83e+00

Pos logit contributions
 syrup+1.265
Pen+1.203
 kil+1.136
 Pen+1.122
 marsh+1.119
Neg logit contributions
ate-1.452
 army-1.340
room-1.269
amed-1.266
roc-1.250
Token top
Token ID1353
Feature activation-1.08e+00

Pos logit contributions
 syrup+3.070
Pen+2.984
 tug+2.849
 Frost+2.792
 Willow+2.790
Neg logit contributions
 army-3.073
ate-3.020
amed-2.720
room-2.707
ating-2.646
Token shelf
Token ID18316
Feature activation-3.61e-01

Pos logit contributions
 syrup+1.817
Pen+1.755
 Pen+1.650
 Frost+1.646
 marsh+1.625
Neg logit contributions
ate-1.911
 army-1.790
room-1.716
amed-1.670
ating-1.669
Token of
Token ID286
Feature activation+1.04e-01

Pos logit contributions
 syrup+1.188
Pen+1.179
 Pen+1.111
 marsh+1.104
 kil+1.069
Neg logit contributions
ate-1.069
 army-1.035
amed-0.967
roc-0.941
ating-0.938
Token her
Token ID607
Feature activation+3.36e-01

Pos logit contributions
ate+0.186
 army+0.178
amed+0.170
roc+0.169
ating+0.165
Neg logit contributions
 syrup-0.171
Pen-0.162
 Pen-0.155
 marsh-0.153
 kil-0.149
Token for
Token ID329
Feature activation+1.09e+00

Pos logit contributions
ate+0.661
 army+0.633
amed+0.611
roc+0.605
ating+0.590
Neg logit contributions
 syrup-0.501
Pen-0.470
 Pen-0.451
 marsh-0.439
 kil-0.428
Token making
Token ID1642
Feature activation+1.52e+00

Pos logit contributions
ate+1.871
 army+1.756
roc+1.716
amed+1.712
ating+1.616
Neg logit contributions
 syrup-1.924
Pen-1.829
 Pen-1.791
 Frost-1.703
 Willow-1.698
Token a
Token ID257
Feature activation+1.53e+00

Pos logit contributions
ate+3.039
 army+2.915
amed+2.886
roc+2.851
ating+2.762
Neg logit contributions
 syrup-2.257
Pen-2.034
 Pen-2.007
 marsh-1.917
 Frost-1.893
Token new
Token ID649
Feature activation+2.56e+00

Pos logit contributions
ate+2.768
amed+2.618
 army+2.536
ating+2.513
roc+2.452
Neg logit contributions
 syrup-2.511
Pen-2.342
 Pen-2.270
 marsh-2.250
 Frost-2.212
Token friend
Token ID1545
Feature activation+4.85e-01

Pos logit contributions
ate+4.029
amed+3.966
roc+3.759
ating+3.716
ined+3.650
Neg logit contributions
 syrup-4.545
 Pen-4.358
Pen-4.278
 bit-4.277
 might-4.258
Token and
Token ID290
Feature activation+1.77e+00

Pos logit contributions
ate+0.984
 army+0.940
amed+0.912
roc+0.903
ating+0.881
Neg logit contributions
 syrup-0.688
Pen-0.642
 Pen-0.617
 marsh-0.597
 kil-0.582
Token not
Token ID407
Feature activation+2.59e-02

Pos logit contributions
ate+3.927
roc+3.770
 army+3.758
amed+3.679
ating+3.615
Neg logit contributions
 syrup-2.212
 Pen-1.954
Pen-1.942
 marsh-1.863
 Willow-1.835
Token feeling
Token ID4203
Feature activation+4.18e-01

Pos logit contributions
ate+0.044
 army+0.042
amed+0.040
roc+0.040
ating+0.039
Neg logit contributions
 syrup-0.045
Pen-0.043
 Pen-0.041
 marsh-0.040
 kil-0.039
Token guilty
Token ID6717
Feature activation+4.57e-03

Pos logit contributions
ate+0.688
 army+0.653
amed+0.626
roc+0.621
ating+0.601
Neg logit contributions
 syrup-0.750
Pen-0.713
 Pen-0.690
 marsh-0.675
 kil-0.660
Token
Token ID198
Feature activation+8.02e-01

Pos logit contributions
ate+0.154
 army+0.147
amed+0.141
roc+0.140
ating+0.136
Neg logit contributions
 syrup-0.129
Pen-0.121
 Pen-0.116
 marsh-0.114
 kil-0.111
TokenThen
Token ID6423
Feature activation+6.98e-01

Pos logit contributions
ate+1.519
 army+1.462
amed+1.386
roc+1.380
ating+1.355
Neg logit contributions
 syrup-1.239
Pen-1.190
 Pen-1.118
 marsh-1.099
 kil-1.075
Token,
Token ID11
Feature activation+5.86e-01

Pos logit contributions
ate+1.033
 army+0.961
amed+0.926
roc+0.908
ating+0.879
Neg logit contributions
 syrup-1.378
Pen-1.311
 Pen-1.282
 marsh-1.276
 Frost-1.250
Token one
Token ID530
Feature activation-4.17e-01

Pos logit contributions
ate+1.217
 army+1.164
amed+1.128
roc+1.117
ating+1.094
Neg logit contributions
 syrup-0.806
Pen-0.750
 Pen-0.720
 marsh-0.714
 Frost-0.691
Token of
Token ID286
Feature activation+1.06e+00

Pos logit contributions
 syrup+0.749
Pen+0.721
 Pen+0.688
 marsh+0.674
 Willow+0.661
Neg logit contributions
ate-0.684
 army-0.677
amed-0.626
roc-0.625
ating-0.598
Token the
Token ID262
Feature activation+2.24e+00

Pos logit contributions
ate+2.232
 army+2.167
amed+2.084
roc+2.068
ating+2.022
Neg logit contributions
 syrup-1.407
Pen-1.306
 Pen-1.256
 marsh-1.230
 Frost-1.199
Token boys
Token ID6510
Feature activation+6.93e-01

Pos logit contributions
ate+3.777
ating+3.281
amed+3.241
isdom+3.107
roc+3.068
Neg logit contributions
 syrup-4.301
 tug-3.899
 marsh-3.867
 kil-3.813
 might-3.789
Token saw
Token ID2497
Feature activation+2.31e+00

Pos logit contributions
ate+1.419
 army+1.340
amed+1.304
roc+1.299
ating+1.270
Neg logit contributions
 syrup-0.970
Pen-0.894
 Pen-0.867
 marsh-0.846
 kil-0.817
Token a
Token ID257
Feature activation+3.27e-01

Pos logit contributions
ate+5.272
 army+5.073
amed+5.026
roc+5.016
ating+4.840
Neg logit contributions
 syrup-2.734
 marsh-2.506
 Frost-2.476
 Pen-2.315
 tug-2.308
Token shiny
Token ID22441
Feature activation+1.25e+00

Pos logit contributions
ate+0.564
 army+0.536
amed+0.515
roc+0.508
ating+0.497
Neg logit contributions
 syrup-0.559
Pen-0.530
 Pen-0.510
 marsh-0.502
 kil-0.488
Token ladder
Token ID18002
Feature activation-1.17e-01

Pos logit contributions
ate+2.266
amed+2.156
 army+2.120
roc+2.081
ating+2.029
Neg logit contributions
 syrup-2.065
Pen-1.848
 marsh-1.826
 Pen-1.800
 kil-1.784
TokenBut
Token ID1537
Feature activation+9.44e-02

Pos logit contributions
 syrup+0.145
Pen+0.137
 Pen+0.133
 marsh+0.131
 kil+0.127
Neg logit contributions
ate-0.137
 army-0.130
amed-0.125
roc-0.123
ating-0.119
Token his
Token ID465
Feature activation+5.78e-02

Pos logit contributions
ate+0.193
 army+0.187
amed+0.179
roc+0.177
ating+0.174
Neg logit contributions
 syrup-0.132
Pen-0.125
 Pen-0.118
 marsh-0.115
 kil-0.112
Token mum
Token ID25682
Feature activation+5.34e-01

Pos logit contributions
ate+0.107
 army+0.103
amed+0.098
roc+0.097
ating+0.095
Neg logit contributions
 syrup-0.092
Pen-0.088
 Pen-0.084
 marsh-0.082
 kil-0.080
Token knew
Token ID2993
Feature activation-6.86e-01

Pos logit contributions
ate+1.069
 army+1.026
amed+0.983
roc+0.979
ating+0.951
Neg logit contributions
 syrup-0.770
Pen-0.719
 Pen-0.691
 marsh-0.668
 Willow-0.653
Token how
Token ID703
Feature activation-3.34e-01

Pos logit contributions
 syrup+0.976
Pen+0.932
 marsh+0.852
 Pen+0.851
 kil+0.821
Neg logit contributions
ate-1.410
 army-1.345
ating-1.272
amed-1.271
roc-1.257
Token to
Token ID284
Feature activation+1.93e+00

Pos logit contributions
 syrup+0.462
Pen+0.437
 Pen+0.412
 marsh+0.403
 Willow+0.388
Neg logit contributions
ate-0.695
 army-0.675
amed-0.640
roc-0.638
ating-0.627
Token comfort
Token ID4467
Feature activation+1.11e+00

Pos logit contributions
amed+4.020
ate+3.933
 army+3.922
roc+3.844
ating+3.665
Neg logit contributions
 syrup-2.601
Pen-2.350
 tug-2.331
 Pen-2.265
 might-2.241
Token him
Token ID683
Feature activation+2.22e-01

Pos logit contributions
ate+2.264
 army+2.197
amed+2.108
roc+2.071
ating+1.996
Neg logit contributions
 syrup-1.569
Pen-1.477
 Pen-1.434
 Frost-1.353
 Willow-1.352
Token.
Token ID13
Feature activation+5.33e-01

Pos logit contributions
ate+0.438
 army+0.422
amed+0.405
roc+0.401
ating+0.392
Neg logit contributions
 syrup-0.325
Pen-0.306
 Pen-0.292
 marsh-0.285
 kil-0.276
Token She
Token ID1375
Feature activation+1.55e-02

Pos logit contributions
ate+0.912
 army+0.872
amed+0.833
roc+0.820
ating+0.806
Neg logit contributions
 syrup-0.925
Pen-0.882
 Pen-0.855
 marsh-0.825
 Willow-0.813
Token hugged
Token ID41130
Feature activation+1.64e-01

Pos logit contributions
ate+0.030
 army+0.028
amed+0.027
roc+0.027
ating+0.026
Neg logit contributions
 syrup-0.023
Pen-0.022
 Pen-0.021
 marsh-0.021
 kil-0.020
Token He
Token ID679
Feature activation-4.28e-01

Pos logit contributions
ate+0.257
 army+0.245
amed+0.233
roc+0.231
ating+0.224
Neg logit contributions
 syrup-0.272
Pen-0.260
 Pen-0.249
 marsh-0.245
 kil-0.239
Token saw
Token ID2497
Feature activation+1.42e+00

Pos logit contributions
 syrup+0.703
Pen+0.670
 Pen+0.630
 marsh+0.624
 kil+0.612
Neg logit contributions
ate-0.777
 army-0.748
amed-0.718
roc-0.696
ating-0.680
Token a
Token ID257
Feature activation+3.93e-01

Pos logit contributions
ate+2.933
 army+2.804
amed+2.734
ating+2.698
roc+2.694
Neg logit contributions
 syrup-2.050
Pen-1.787
 marsh-1.778
 Pen-1.773
 Frost-1.772
Token little
Token ID1310
Feature activation+8.20e-01

Pos logit contributions
ate+0.678
 army+0.644
amed+0.619
roc+0.609
ating+0.598
Neg logit contributions
 syrup-0.677
Pen-0.641
 Pen-0.616
 marsh-0.607
 kil-0.590
Token boy
Token ID2933
Feature activation+6.77e-02

Pos logit contributions
ate+1.448
 army+1.339
amed+1.320
ating+1.278
roc+1.275
Neg logit contributions
 syrup-1.405
Pen-1.320
 Pen-1.283
 marsh-1.255
 kil-1.223
Token playing
Token ID2712
Feature activation+1.93e+00

Pos logit contributions
ate+0.118
 army+0.114
amed+0.108
roc+0.107
ating+0.104
Neg logit contributions
 syrup-0.115
Pen-0.109
 Pen-0.104
 marsh-0.102
 kil-0.099
Token with
Token ID351
Feature activation+6.59e-01

Pos logit contributions
ate+3.442
amed+3.205
roc+3.191
 army+3.185
ating+3.079
Neg logit contributions
 syrup-3.191
 tug-2.938
Pen-2.872
 Pen-2.867
 Frost-2.861
Token a
Token ID257
Feature activation+4.94e-01

Pos logit contributions
ate+1.257
 army+1.196
amed+1.159
roc+1.147
ating+1.126
Neg logit contributions
 syrup-1.028
Pen-0.942
 Pen-0.909
 marsh-0.909
 Frost-0.880
Token huge
Token ID3236
Feature activation+1.67e+00

Pos logit contributions
ate+0.853
 army+0.803
amed+0.774
roc+0.764
ating+0.752
Neg logit contributions
 syrup-0.858
Pen-0.807
 Pen-0.775
 marsh-0.767
 kil-0.745
Token,
Token ID11
Feature activation+1.32e+00

Pos logit contributions
ate+3.308
amed+3.007
ating+2.938
 army+2.918
roc+2.883
Neg logit contributions
 syrup-2.735
 tug-2.381
 marsh-2.353
Pen-2.327
 Pen-2.308
Token pink
Token ID11398
Feature activation+1.07e+00

Pos logit contributions
ate+2.604
 army+2.383
roc+2.369
ating+2.350
amed+2.346
Neg logit contributions
 syrup-2.082
Pen-1.873
 marsh-1.828
 Pen-1.820
 tug-1.788
Token many
Token ID867
Feature activation+1.09e+00

Pos logit contributions
ate+0.545
 army+0.523
amed+0.501
roc+0.497
ating+0.486
Neg logit contributions
 syrup-0.469
Pen-0.442
 Pen-0.423
 marsh-0.415
 kil-0.404
Token colors
Token ID7577
Feature activation+1.20e+00

Pos logit contributions
ate+1.740
 army+1.666
amed+1.595
roc+1.579
ating+1.521
Neg logit contributions
 syrup-2.072
Pen-1.846
 marsh-1.823
 kil-1.812
 Pen-1.781
Token.
Token ID13
Feature activation+3.90e-01

Pos logit contributions
 army+2.211
ate+2.208
roc+2.148
amed+2.061
ating+1.969
Neg logit contributions
 syrup-1.927
Pen-1.672
 marsh-1.669
 Pen-1.637
 kil-1.613
Token Tom
Token ID4186
Feature activation+6.11e-01

Pos logit contributions
ate+0.732
 army+0.706
amed+0.679
roc+0.669
ating+0.657
Neg logit contributions
 syrup-0.614
Pen-0.579
 Pen-0.557
 marsh-0.542
 Frost-0.531
Token likes
Token ID7832
Feature activation+6.51e-01

Pos logit contributions
ate+1.171
 army+1.139
amed+1.098
roc+1.086
ating+1.062
Neg logit contributions
 syrup-0.918
Pen-0.856
 Pen-0.827
 marsh-0.798
 Frost-0.790
Token the
Token ID262
Feature activation+2.17e+00

Pos logit contributions
ate+1.414
 army+1.363
amed+1.343
roc+1.308
ating+1.290
Neg logit contributions
 syrup-0.842
Pen-0.751
 Pen-0.718
 marsh-0.705
 Frost-0.686
Token shiny
Token ID22441
Feature activation+2.25e+00

Pos logit contributions
ate+3.603
amed+3.495
 army+3.404
ating+3.372
roc+3.297
Neg logit contributions
 syrup-4.130
 might-3.558
 marsh-3.385
 kil-3.343
 nib-3.313
Token pen
Token ID3112
Feature activation+7.70e-01

Pos logit contributions
amed+3.698
ate+3.644
 army+3.551
roc+3.415
room+3.312
Neg logit contributions
 syrup-4.363
 might-3.627
 kil-3.561
 marsh-3.547
 tug-3.443
Token a
Token ID257
Feature activation-2.24e-01

Pos logit contributions
ate+1.664
 army+1.603
roc+1.572
amed+1.561
ating+1.506
Neg logit contributions
 syrup-0.992
Pen-0.878
 Pen-0.848
 marsh-0.835
 kil-0.814
Token lot
Token ID1256
Feature activation+1.90e-01

Pos logit contributions
 syrup+0.317
Pen+0.307
 Pen+0.292
 marsh+0.277
 Willow+0.271
Neg logit contributions
ate-0.452
 army-0.435
amed-0.414
roc-0.410
ating-0.401
Token.
Token ID13
Feature activation+2.46e-01

Pos logit contributions
ate+0.407
 army+0.393
amed+0.379
roc+0.375
ating+0.367
Neg logit contributions
 syrup-0.247
Pen-0.231
 Pen-0.218
 marsh-0.212
 kil-0.205
Token all
Token ID477
Feature activation-2.34e-01

Pos logit contributions
 syrup+1.853
Pen+1.811
 Pen+1.665
 kil+1.649
 marsh+1.611
Neg logit contributions
ate-2.769
 army-2.675
amed-2.530
roc-2.483
ating-2.472
Token become
Token ID1716
Feature activation+9.49e-02

Pos logit contributions
 syrup+0.352
Pen+0.333
 Pen+0.314
 marsh+0.307
 kil+0.303
Neg logit contributions
ate-0.455
 army-0.436
amed-0.416
roc-0.411
ating-0.407
Token kind
Token ID1611
Feature activation+1.31e+00

Pos logit contributions
ate+0.176
 army+0.168
amed+0.161
roc+0.159
ating+0.155
Neg logit contributions
 syrup-0.151
Pen-0.144
 Pen-0.137
 marsh-0.134
 kil-0.131
Token and
Token ID290
Feature activation+7.09e-01

Pos logit contributions
ate+2.483
 army+2.371
amed+2.307
roc+2.302
ating+2.192
Neg logit contributions
 syrup-2.038
Pen-1.876
 Pen-1.853
 marsh-1.795
 Willow-1.749
Token take
Token ID1011
Feature activation+4.68e-01

Pos logit contributions
ate+1.285
 army+1.236
amed+1.191
roc+1.184
ating+1.135
Neg logit contributions
 syrup-1.158
Pen-1.089
 Pen-1.053
 marsh-1.030
 Frost-1.004
Token care
Token ID1337
Feature activation+1.21e+00

Pos logit contributions
ate+1.020
 army+0.989
amed+0.951
roc+0.947
ating+0.923
Neg logit contributions
 syrup-0.592
Pen-0.549
 Pen-0.522
 marsh-0.508
 Frost-0.487
Token of
Token ID286
Feature activation+4.09e-01

Pos logit contributions
ate+2.362
 army+2.330
roc+2.271
amed+2.197
ating+2.098
Neg logit contributions
 syrup-1.800
Pen-1.672
 Pen-1.620
 marsh-1.588
 Willow-1.549
Token nature
Token ID3450
Feature activation+3.43e-01

Pos logit contributions
ate+0.796
 army+0.768
amed+0.742
roc+0.733
ating+0.711
Neg logit contributions
 syrup-0.608
Pen-0.573
 Pen-0.549
 marsh-0.537
 Frost-0.519
Token,
Token ID11
Feature activation+4.96e-01

Pos logit contributions
ate+0.646
 army+0.620
amed+0.596
roc+0.591
ating+0.573
Neg logit contributions
 syrup-0.529
Pen-0.501
 Pen-0.481
 marsh-0.469
 Frost-0.454
Token it
Token ID340
Feature activation-3.23e-01

Pos logit contributions
ate+1.095
 army+1.057
amed+1.027
roc+1.018
ating+0.993
Neg logit contributions
 syrup-0.611
Pen-0.565
 Pen-0.539
 marsh-0.524
 Willow-0.501
Token would
Token ID561
Feature activation+4.10e-02

Pos logit contributions
 syrup+0.357
Pen+0.328
 Pen+0.301
 marsh+0.294
 kil+0.288
Neg logit contributions
ate-0.757
 army-0.726
amed-0.704
ating-0.693
roc-0.693
Token finally
Token ID3443
Feature activation-2.32e-02

Pos logit contributions
 syrup+2.938
Pen+2.765
 kil+2.702
 marsh+2.498
 Pen+2.448
Neg logit contributions
ate-3.173
 army-2.935
ating-2.882
amed-2.811
roc-2.739
Token she
Token ID673
Feature activation-3.20e-01

Pos logit contributions
 syrup+0.041
Pen+0.039
 Pen+0.037
 marsh+0.036
 kil+0.036
Neg logit contributions
ate-0.039
 army-0.038
amed-0.036
roc-0.035
ating-0.035
Token pushed
Token ID7121
Feature activation-9.18e-01

Pos logit contributions
 syrup+0.429
Pen+0.407
 Pen+0.379
 marsh+0.371
 kil+0.357
Neg logit contributions
ate-0.683
 army-0.657
amed-0.635
roc-0.621
ating-0.615
Token the
Token ID262
Feature activation-6.10e-01

Pos logit contributions
 syrup+1.528
Pen+1.448
 Pen+1.344
 kil+1.340
 marsh+1.338
Neg logit contributions
ate-1.659
 army-1.564
amed-1.507
ating-1.465
roc-1.458
Token bedroom
Token ID14043
Feature activation+3.34e-01

Pos logit contributions
 syrup+1.133
Pen+1.083
 Pen+1.028
 kil+1.008
 marsh+1.007
Neg logit contributions
ate-0.987
 army-0.953
amed-0.890
ating-0.858
roc-0.848
Token door
Token ID3420
Feature activation+3.09e-01

Pos logit contributions
ate+0.515
 army+0.489
amed+0.465
roc+0.461
ating+0.445
Neg logit contributions
 syrup-0.632
Pen-0.603
 Pen-0.584
 marsh-0.573
 kil-0.559
Token open
Token ID1280
Feature activation+1.67e-02

Pos logit contributions
ate+0.514
 army+0.492
amed+0.467
roc+0.465
ating+0.449
Neg logit contributions
 syrup-0.547
Pen-0.520
 Pen-0.502
 marsh-0.492
 kil-0.479
Token,
Token ID11
Feature activation-5.05e-01

Pos logit contributions
ate+0.022
 army+0.021
amed+0.020
roc+0.019
ating+0.019
Neg logit contributions
 syrup-0.035
Pen-0.034
 Pen-0.033
 marsh-0.032
 kil-0.032
Token she
Token ID673
Feature activation+4.06e-03

Pos logit contributions
 syrup+0.708
Pen+0.659
 kil+0.604
 Pen+0.602
 marsh+0.600
Neg logit contributions
ate-1.043
 army-0.992
amed-0.961
roc-0.936
ating-0.933
Token sighed
Token ID21893
Feature activation+1.56e-01

Pos logit contributions
ate+0.008
 army+0.008
amed+0.007
roc+0.007
ating+0.007
Neg logit contributions
 syrup-0.006
Pen-0.006
 Pen-0.005
 marsh-0.005
 kil-0.005
Token in
Token ID287
Feature activation-5.27e-01

Pos logit contributions
ate+0.304
 army+0.292
amed+0.280
roc+0.277
ating+0.271
Neg logit contributions
 syrup-0.231
Pen-0.218
 Pen-0.208
 marsh-0.203
 kil-0.197
Token were
Token ID547
Feature activation+3.15e-01

Pos logit contributions
 syrup+0.397
Pen+0.381
 Pen+0.362
 marsh+0.354
 kil+0.351
Neg logit contributions
ate-0.444
 army-0.428
amed-0.410
roc-0.402
ating-0.388
Token happy
Token ID3772
Feature activation-5.01e-01

Pos logit contributions
ate+0.551
 army+0.528
amed+0.504
roc+0.501
ating+0.487
Neg logit contributions
 syrup-0.534
Pen-0.508
 Pen-0.488
 marsh-0.478
 Frost-0.467
Token and
Token ID290
Feature activation+1.25e+00

Pos logit contributions
 syrup+0.847
Pen+0.820
 Pen+0.767
 marsh+0.754
 kil+0.743
Neg logit contributions
ate-0.892
 army-0.860
amed-0.799
roc-0.794
ating-0.788
Token smiled
Token ID13541
Feature activation+6.40e-01

Pos logit contributions
ate+2.490
 army+2.370
roc+2.302
amed+2.244
ating+2.232
Neg logit contributions
 syrup-1.833
Pen-1.692
 Pen-1.656
 Frost-1.630
 marsh-1.597
Token.
Token ID13
Feature activation+3.34e-01

Pos logit contributions
ate+1.311
 army+1.263
roc+1.226
amed+1.216
ating+1.188
Neg logit contributions
 syrup-0.879
Pen-0.823
 Pen-0.789
 marsh-0.767
 Frost-0.756
Token The
Token ID383
Feature activation+2.77e-01

Pos logit contributions
ate+0.567
 army+0.543
amed+0.518
roc+0.512
ating+0.500
Neg logit contributions
 syrup-0.586
Pen-0.558
 Pen-0.537
 marsh-0.528
 Frost-0.516
Token sun
Token ID4252
Feature activation-3.02e-01

Pos logit contributions
ate+0.430
 army+0.408
amed+0.389
roc+0.384
ating+0.372
Neg logit contributions
 syrup-0.530
Pen-0.505
 Pen-0.487
 marsh-0.479
 kil-0.468
Token smiled
Token ID13541
Feature activation+1.06e+00

Pos logit contributions
 syrup+0.484
Pen+0.465
 Pen+0.439
 kil+0.429
 marsh+0.428
Neg logit contributions
ate-0.552
 army-0.524
amed-0.508
ating-0.493
roc-0.491
Token too
Token ID1165
Feature activation+5.01e-01

Pos logit contributions
ate+1.984
 army+1.910
roc+1.883
amed+1.805
ating+1.781
Neg logit contributions
 syrup-1.670
Pen-1.543
 Pen-1.494
 marsh-1.477
 Frost-1.458
Token.
Token ID13
Feature activation+7.50e-01

Pos logit contributions
ate+1.066
 army+1.033
amed+0.996
roc+0.993
ating+0.961
Neg logit contributions
 syrup-0.662
Pen-0.610
 Pen-0.584
 marsh-0.576
 Frost-0.555
Token<|endoftext|>
Token ID50256
Feature activation-1.64e-01

Pos logit contributions
ate+1.329
 army+1.289
amed+1.236
roc+1.225
ating+1.207
Neg logit contributions
 syrup-1.253
Pen-1.184
 Pen-1.145
 marsh-1.129
 Frost-1.109
Token he
Token ID339
Feature activation-6.16e-01

Pos logit contributions
 syrup+0.885
Pen+0.836
 Pen+0.765
 marsh+0.762
 kil+0.749
Neg logit contributions
ate-1.243
 army-1.209
amed-1.125
roc-1.121
ating-1.117
Token found
Token ID1043
Feature activation-2.83e-02

Pos logit contributions
 syrup+0.863
Pen+0.811
 marsh+0.767
 Pen+0.758
 kil+0.728
Neg logit contributions
ate-1.274
 army-1.228
amed-1.206
ating-1.164
roc-1.145
Token a
Token ID257
Feature activation-8.93e-01

Pos logit contributions
 syrup+0.046
Pen+0.044
 Pen+0.042
 marsh+0.041
 kil+0.040
Neg logit contributions
ate-0.052
 army-0.049
amed-0.047
roc-0.047
ating-0.046
Token friend
Token ID1545
Feature activation-9.24e-01

Pos logit contributions
 syrup+1.570
Pen+1.523
 Pen+1.457
 Frost+1.410
 kil+1.406
Neg logit contributions
ate-1.455
 army-1.416
roc-1.317
amed-1.313
ating-1.247
Token who
Token ID508
Feature activation-3.64e-01

Pos logit contributions
 syrup+1.404
Pen+1.338
 marsh+1.244
 Pen+1.236
 kil+1.191
Neg logit contributions
ate-1.751
 army-1.736
amed-1.621
ating-1.582
roc-1.581
Token also
Token ID635
Feature activation+8.32e-01

Pos logit contributions
 syrup+0.573
Pen+0.541
 Pen+0.510
 marsh+0.506
 kil+0.493
Neg logit contributions
ate-0.683
 army-0.657
amed-0.627
roc-0.613
ating-0.609
Token liked
Token ID8288
Feature activation-1.11e+00

Pos logit contributions
ate+1.709
 army+1.632
roc+1.588
amed+1.558
ating+1.531
Neg logit contributions
 syrup-1.183
Pen-1.092
 Pen-1.053
 marsh-1.030
 kil-0.991
Token empty
Token ID6565
Feature activation+1.56e-01

Pos logit contributions
 syrup+1.665
Pen+1.626
 Pen+1.499
 kil+1.481
 marsh+1.472
Neg logit contributions
ate-2.091
 army-2.040
roc-1.918
ating-1.882
amed-1.861
Token apples
Token ID22514
Feature activation-7.49e-01

Pos logit contributions
ate+0.293
 army+0.282
amed+0.270
roc+0.267
ating+0.261
Neg logit contributions
 syrup-0.243
Pen-0.230
 Pen-0.220
 marsh-0.215
 kil-0.209
Token.
Token ID13
Feature activation-4.78e-01

Pos logit contributions
 syrup+1.091
Pen+1.057
 Pen+0.985
 marsh+0.957
 Willow+0.947
Neg logit contributions
ate-1.456
 army-1.420
amed-1.331
ating-1.324
roc-1.295
Token<|endoftext|>
Token ID50256
Feature activation-5.70e-01

Pos logit contributions
 syrup+0.934
Pen+0.881
 marsh+0.840
 Pen+0.839
 kil+0.832
Neg logit contributions
ate-0.716
 army-0.675
amed-0.631
roc-0.629
ating-0.599
Token up
Token ID510
Feature activation-9.46e-02

Pos logit contributions
ate+0.471
 army+0.452
amed+0.434
roc+0.430
ating+0.419
Neg logit contributions
 syrup-0.391
Pen-0.369
 Pen-0.354
 marsh-0.345
 kil-0.336
Token with
Token ID351
Feature activation-1.09e-01

Pos logit contributions
 syrup+0.127
Pen+0.120
 Pen+0.113
 marsh+0.110
 kil+0.106
Neg logit contributions
ate-0.198
 army-0.192
amed-0.183
roc-0.181
ating-0.178
Token you
Token ID345
Feature activation-6.89e-01

Pos logit contributions
 syrup+0.168
Pen+0.160
 Pen+0.152
 marsh+0.149
 kil+0.145
Neg logit contributions
ate-0.206
 army-0.198
amed-0.188
roc-0.187
ating-0.183
Token."
Token ID526
Feature activation+5.00e-01

Pos logit contributions
 syrup+0.935
Pen+0.879
 Pen+0.818
 marsh+0.805
 kil+0.776
Neg logit contributions
ate-1.434
 army-1.407
amed-1.319
roc-1.296
ating-1.287
Token
Token ID198
Feature activation-9.78e-02

Pos logit contributions
ate+0.999
 army+0.970
amed+0.926
roc+0.919
ating+0.902
Neg logit contributions
 syrup-0.724
Pen-0.677
 Pen-0.648
 marsh-0.629
 Frost-0.615
Token
Token ID198
Feature activation+3.46e-01

Pos logit contributions
 syrup+0.157
Pen+0.148
 Pen+0.142
 marsh+0.140
 kil+0.137
Neg logit contributions
ate-0.181
 army-0.171
amed-0.166
roc-0.163
ating-0.159
TokenL
Token ID43
Feature activation+1.75e+00

Pos logit contributions
ate+0.610
 army+0.587
amed+0.558
roc+0.553
ating+0.539
Neg logit contributions
 syrup-0.583
Pen-0.556
 Pen-0.531
 marsh-0.518
 kil-0.506
Tokenily
Token ID813
Feature activation+1.47e+00

Pos logit contributions
 army+2.869
ate+2.812
roc+2.663
amed+2.605
ating+2.508
Neg logit contributions
 syrup-3.099
Pen-2.954
 Pen-2.819
 Megan-2.731
 kil-2.693
Token didn
Token ID1422
Feature activation-9.24e-02

Pos logit contributions
ate+2.964
 army+2.917
roc+2.764
ating+2.674
amed+2.670
Neg logit contributions
 syrup-2.111
Pen-1.948
 Pen-1.918
 tug-1.841
 Willow-1.797
Token't
Token ID470
Feature activation-2.85e-01

Pos logit contributions
 syrup+0.123
Pen+0.115
 Pen+0.109
 marsh+0.106
 kil+0.103
Neg logit contributions
ate-0.195
 army-0.187
amed-0.180
roc-0.178
ating-0.175
Token care
Token ID1337
Feature activation+5.99e-01

Pos logit contributions
 syrup+0.438
Pen+0.421
 Pen+0.400
 marsh+0.394
 Frost+0.378
Neg logit contributions
ate-0.542
 army-0.524
roc-0.496
amed-0.496
ating-0.477
Token dress
Token ID6576
Feature activation+3.59e-01

Pos logit contributions
ate+2.144
amed+2.030
 army+2.008
roc+1.996
ating+1.954
Neg logit contributions
 syrup-2.195
Pen-2.046
 Pen-2.014
 marsh-1.969
 Willow-1.915
Token that
Token ID326
Feature activation+2.40e-01

Pos logit contributions
ate+0.662
 army+0.636
amed+0.609
roc+0.608
ating+0.587
Neg logit contributions
 syrup-0.574
Pen-0.539
 Pen-0.517
 marsh-0.507
 kil-0.491
Token her
Token ID607
Feature activation+2.34e-01

Pos logit contributions
ate+0.540
 army+0.522
amed+0.504
roc+0.501
ating+0.490
Neg logit contributions
 syrup-0.287
Pen-0.266
 Pen-0.251
 marsh-0.243
 Frost-0.234
Token mom
Token ID1995
Feature activation-1.07e-02

Pos logit contributions
ate+0.425
 army+0.408
amed+0.390
roc+0.387
ating+0.376
Neg logit contributions
 syrup-0.381
Pen-0.361
 Pen-0.346
 marsh-0.338
 kil-0.329
Tokenmy
Token ID1820
Feature activation-3.74e-02

Pos logit contributions
 syrup+0.020
Pen+0.019
 Pen+0.018
 marsh+0.018
 kil+0.017
Neg logit contributions
ate-0.017
 army-0.016
amed-0.016
roc-0.015
ating-0.015
Token bought
Token ID5839
Feature activation-3.67e-01

Pos logit contributions
 syrup+0.054
Pen+0.051
 Pen+0.049
 marsh+0.048
 kil+0.047
Neg logit contributions
ate-0.075
 army-0.071
amed-0.069
roc-0.068
ating-0.067
Token for
Token ID329
Feature activation-1.51e-01

Pos logit contributions
 syrup+0.535
Pen+0.522
 Pen+0.489
 marsh+0.479
 kil+0.467
Neg logit contributions
ate-0.723
 army-0.691
amed-0.663
roc-0.649
ating-0.645
Token her
Token ID607
Feature activation-8.22e-01

Pos logit contributions
 syrup+0.252
Pen+0.244
 Pen+0.233
 marsh+0.228
 kil+0.225
Neg logit contributions
ate-0.264
 army-0.253
amed-0.239
roc-0.237
ating-0.232
Token.
Token ID13
Feature activation-9.03e-02

Pos logit contributions
 syrup+1.201
Pen+1.170
 Pen+1.082
 marsh+1.073
 kil+1.047
Neg logit contributions
ate-1.638
 army-1.583
amed-1.488
roc-1.447
ating-1.442
Token It
Token ID632
Feature activation-1.39e-01

Pos logit contributions
 syrup+0.135
Pen+0.127
 Pen+0.121
 marsh+0.118
 kil+0.117
Neg logit contributions
ate-0.176
 army-0.168
amed-0.161
roc-0.160
ating-0.155
Token was
Token ID373
Feature activation-8.05e-02

Pos logit contributions
 syrup+0.226
Pen+0.215
 Pen+0.204
 marsh+0.201
 kil+0.200
Neg logit contributions
ate-0.254
 army-0.240
amed-0.233
roc-0.226
ating-0.224
Token mud
Token ID17492
Feature activation+2.87e-01

Pos logit contributions
 syrup+1.409
Pen+1.372
 Pen+1.311
 kil+1.259
 marsh+1.255
Neg logit contributions
 army-1.243
ate-1.222
roc-1.131
amed-1.125
ating-1.053
Token and
Token ID290
Feature activation-4.57e-02

Pos logit contributions
ate+0.548
 army+0.525
amed+0.504
roc+0.500
ating+0.487
Neg logit contributions
 syrup-0.440
Pen-0.414
 Pen-0.397
 marsh-0.389
 kil-0.377
Token placed
Token ID4624
Feature activation-8.22e-01

Pos logit contributions
 syrup+0.061
Pen+0.058
 Pen+0.054
 marsh+0.052
 kil+0.051
Neg logit contributions
ate-0.096
 army-0.093
amed-0.090
roc-0.088
ating-0.087
Token it
Token ID340
Feature activation-4.37e-01

Pos logit contributions
 syrup+1.078
Pen+1.044
 Pen+0.963
 marsh+0.921
 kil+0.911
Neg logit contributions
ate-1.747
 army-1.673
amed-1.639
roc-1.588
ating-1.577
Token on
Token ID319
Feature activation-5.19e-01

Pos logit contributions
 syrup+0.668
Pen+0.634
 Pen+0.598
 marsh+0.580
 Willow+0.570
Neg logit contributions
ate-0.845
 army-0.814
amed-0.798
ating-0.763
roc-0.748
Token a
Token ID257
Feature activation-1.38e+00

Pos logit contributions
 syrup+0.842
Pen+0.813
 Pen+0.763
 marsh+0.738
 kil+0.735
Neg logit contributions
ate-0.943
 army-0.905
amed-0.847
roc-0.840
ating-0.827
Token nearby
Token ID6716
Feature activation-2.41e-01

Pos logit contributions
 syrup+2.327
Pen+2.227
 Pen+2.146
 Willow+2.070
 Lu+2.067
Neg logit contributions
 army-2.383
ate-2.296
amed-2.124
roc-2.112
 alien-2.053
Token tree
Token ID5509
Feature activation-2.39e-01

Pos logit contributions
 syrup+0.400
Pen+0.382
 Pen+0.365
 marsh+0.354
 Willow+0.349
Neg logit contributions
ate-0.423
 army-0.413
roc-0.385
amed-0.385
ating-0.371
Token branch
Token ID8478
Feature activation+1.38e-01

Pos logit contributions
 syrup+0.337
Pen+0.319
 Pen+0.300
 marsh+0.292
 Willow+0.285
Neg logit contributions
ate-0.482
 army-0.469
amed-0.448
roc-0.439
ating-0.434
Token.
Token ID13
Feature activation-6.63e-01

Pos logit contributions
ate+0.291
 army+0.281
amed+0.271
roc+0.268
ating+0.263
Neg logit contributions
 syrup-0.183
Pen-0.171
 Pen-0.162
 marsh-0.157
 kil-0.152
Token The
Token ID383
Feature activation-3.96e-01

Pos logit contributions
 syrup+1.290
Pen+1.224
 Pen+1.157
 marsh+1.149
 kil+1.135
Neg logit contributions
ate-1.009
 army-0.949
amed-0.898
roc-0.877
room-0.839
Token but
Token ID475
Feature activation-3.69e-02

Pos logit contributions
 syrup+0.506
Pen+0.478
 Pen+0.446
 marsh+0.435
 kil+0.432
Neg logit contributions
ate-0.748
 army-0.731
amed-0.696
roc-0.683
ating-0.668
Token he
Token ID339
Feature activation-5.84e-01

Pos logit contributions
 syrup+0.045
Pen+0.042
 Pen+0.039
 marsh+0.038
 kil+0.037
Neg logit contributions
ate-0.081
 army-0.079
amed-0.076
roc-0.075
ating-0.074
Token did
Token ID750
Feature activation-1.06e+00

Pos logit contributions
 syrup+0.850
Pen+0.797
 Pen+0.743
 marsh+0.739
 kil+0.715
Neg logit contributions
ate-1.163
 army-1.123
amed-1.075
roc-1.057
ating-1.034
Token not
Token ID407
Feature activation-9.25e-01

Pos logit contributions
 syrup+1.966
Pen+1.877
 Pen+1.765
 marsh+1.716
 Willow+1.703
Neg logit contributions
ate-1.750
 army-1.645
amed-1.508
ating-1.504
roc-1.474
Token want
Token ID765
Feature activation-1.27e+00

Pos logit contributions
 syrup+1.407
Pen+1.338
 Pen+1.249
 marsh+1.244
 Willow+1.207
Neg logit contributions
ate-1.794
 army-1.771
roc-1.629
amed-1.622
ating-1.571
Token to
Token ID284
Feature activation-4.03e-01

Pos logit contributions
 syrup+1.778
Pen+1.771
 Pen+1.611
 marsh+1.560
 kil+1.555
Neg logit contributions
ate-2.523
 army-2.492
roc-2.300
amed-2.261
ating-2.230
Token be
Token ID307
Feature activation+4.68e-01

Pos logit contributions
 syrup+0.594
Pen+0.562
 Pen+0.530
 marsh+0.520
 kil+0.510
Neg logit contributions
ate-0.799
 army-0.770
roc-0.726
amed-0.715
ating-0.704
Token the
Token ID262
Feature activation-6.76e-01

Pos logit contributions
ate+0.889
 army+0.848
amed+0.816
roc+0.813
ating+0.791
Neg logit contributions
 syrup-0.722
Pen-0.683
 Pen-0.656
 marsh-0.640
 Frost-0.624
Token only
Token ID691
Feature activation-7.12e-01

Pos logit contributions
 syrup+1.251
Pen+1.215
 Pen+1.167
 Willow+1.124
 marsh+1.118
Neg logit contributions
 army-1.049
ate-1.048
roc-0.945
amed-0.936
ating-0.896
Token one
Token ID530
Feature activation-1.98e+00

Pos logit contributions
 syrup+1.141
Pen+1.094
 Pen+1.022
 marsh+1.001
 Willow+0.990
Neg logit contributions
ate-1.291
 army-1.269
roc-1.169
amed-1.167
ating-1.144
Token to
Token ID284
Feature activation+4.74e-01

Pos logit contributions
 syrup+3.296
Pen+3.093
 Willow+2.885
 marsh+2.867
 Pen+2.815
Neg logit contributions
 army-3.648
ate-3.538
ating-3.216
roc-3.124
amed-3.120
Token?"
Token ID1701
Feature activation+3.90e-01

Pos logit contributions
 syrup+0.561
Pen+0.543
 Pen+0.524
 marsh+0.515
 kil+0.507
Neg logit contributions
ate-0.419
 army-0.392
amed-0.372
roc-0.372
ating-0.356
Token
Token ID198
Feature activation+2.82e-01

Pos logit contributions
ate+0.790
 army+0.759
amed+0.733
roc+0.725
ating+0.711
Neg logit contributions
 syrup-0.554
Pen-0.523
 Pen-0.500
 marsh-0.484
 Frost-0.471
Token
Token ID198
Feature activation+2.42e-01

Pos logit contributions
ate+0.537
 army+0.517
amed+0.496
roc+0.491
ating+0.480
Neg logit contributions
 syrup-0.431
Pen-0.410
 Pen-0.391
 marsh-0.381
 Frost-0.371
TokenTom
Token ID13787
Feature activation+2.38e-01

Pos logit contributions
ate+0.407
 army+0.389
amed+0.371
roc+0.367
ating+0.357
Neg logit contributions
 syrup-0.426
Pen-0.407
 Pen-0.391
 marsh-0.382
 kil-0.373
Token and
Token ID290
Feature activation+9.32e-02

Pos logit contributions
ate+0.435
 army+0.417
amed+0.399
roc+0.395
ating+0.385
Neg logit contributions
 syrup-0.384
Pen-0.364
 Pen-0.349
 marsh-0.341
 Frost-0.332
Token Mia
Token ID32189
Feature activation-2.92e-01

Pos logit contributions
ate+0.189
 army+0.182
amed+0.175
roc+0.174
ating+0.169
Neg logit contributions
 syrup-0.132
Pen-0.124
 Pen-0.117
 marsh-0.115
 kil-0.112
Token look
Token ID804
Feature activation+5.13e-01

Pos logit contributions
 syrup+0.440
Pen+0.418
 Pen+0.395
 marsh+0.385
 kil+0.380
Neg logit contributions
ate-0.560
 army-0.544
amed-0.520
roc-0.510
ating-0.497
Token at
Token ID379
Feature activation+2.77e-02

Pos logit contributions
ate+0.942
 army+0.899
amed+0.860
roc+0.859
ating+0.840
Neg logit contributions
 syrup-0.827
Pen-0.775
 Pen-0.747
 marsh-0.738
 Frost-0.718
Token Mom
Token ID11254
Feature activation-2.78e-01

Pos logit contributions
ate+0.052
 army+0.050
amed+0.048
roc+0.047
ating+0.046
Neg logit contributions
 syrup-0.043
Pen-0.041
 Pen-0.039
 marsh-0.038
 kil-0.038
Token.
Token ID13
Feature activation-3.51e-02

Pos logit contributions
 syrup+0.425
Pen+0.407
 Pen+0.384
 marsh+0.379
 kil+0.366
Neg logit contributions
ate-0.529
 army-0.506
amed-0.483
roc-0.475
ating-0.472
Token They
Token ID1119
Feature activation+5.13e-01

Pos logit contributions
 syrup+0.067
Pen+0.064
 Pen+0.061
 marsh+0.060
 kil+0.059
Neg logit contributions
ate-0.054
 army-0.051
amed-0.049
roc-0.048
ating-0.046
Token banner
Token ID17625
Feature activation+4.29e-02

Pos logit contributions
ate+0.913
 army+0.881
amed+0.850
roc+0.842
ating+0.812
Neg logit contributions
 syrup-0.792
Pen-0.737
 Pen-0.709
 marsh-0.704
 Willow-0.676
Token that
Token ID326
Feature activation+2.61e-01

Pos logit contributions
ate+0.075
 army+0.072
amed+0.069
roc+0.067
ating+0.066
Neg logit contributions
 syrup-0.072
Pen-0.069
 Pen-0.066
 marsh-0.065
 kil-0.063
Token said
Token ID531
Feature activation+3.19e-01

Pos logit contributions
ate+0.528
 army+0.510
amed+0.490
roc+0.486
ating+0.474
Neg logit contributions
 syrup-0.368
Pen-0.344
 Pen-0.328
 marsh-0.321
 Frost-0.310
Token,
Token ID11
Feature activation-4.14e-01

Pos logit contributions
ate+0.560
 army+0.538
amed+0.514
roc+0.510
ating+0.494
Neg logit contributions
 syrup-0.539
Pen-0.510
 Pen-0.491
 marsh-0.482
 Frost-0.468
Token "
Token ID366
Feature activation-4.86e-01

Pos logit contributions
 syrup+0.741
Pen+0.715
 Pen+0.680
 marsh+0.665
 kil+0.661
Neg logit contributions
ate-0.671
 army-0.640
amed-0.605
roc-0.589
ating-0.580
TokenWe
Token ID1135
Feature activation-7.69e-01

Pos logit contributions
 syrup+0.814
Pen+0.769
 Pen+0.749
 marsh+0.738
 kil+0.722
Neg logit contributions
ate-0.870
 army-0.797
amed-0.796
roc-0.764
ating-0.757
Token are
Token ID389
Feature activation-6.45e-01

Pos logit contributions
Pen+1.073
 syrup+1.070
 Pen+1.019
 Willow+0.950
 kil+0.949
Neg logit contributions
ate-1.574
 army-1.516
amed-1.456
roc-1.413
ating-1.377
Token family
Token ID1641
Feature activation-6.34e-01

Pos logit contributions
 syrup+1.024
Pen+0.981
 Pen+0.924
 kil+0.904
 marsh+0.893
Neg logit contributions
ate-1.220
 army-1.155
amed-1.114
roc-1.061
ating-1.060
Token."
Token ID526
Feature activation+5.83e-01

Pos logit contributions
 syrup+1.052
Pen+1.011
 Pen+0.957
 marsh+0.941
 kil+0.920
Neg logit contributions
ate-1.119
 army-1.083
amed-1.017
roc-0.998
ating-0.995
Token They
Token ID1119
Feature activation-2.84e-01

Pos logit contributions
ate+1.151
 army+1.120
amed+1.074
roc+1.069
ating+1.048
Neg logit contributions
 syrup-0.870
Pen-0.808
 Pen-0.779
 marsh-0.765
 Frost-0.746
Token hung
Token ID9174
Feature activation+4.71e-01

Pos logit contributions
 syrup+0.422
Pen+0.407
 Pen+0.385
 marsh+0.374
 kil+0.373
Neg logit contributions
ate-0.554
 army-0.528
amed-0.516
roc-0.494
ating-0.487
Token.
Token ID13
Feature activation-6.43e-01

Pos logit contributions
 syrup+0.198
Pen+0.184
 Pen+0.173
 marsh+0.169
 kil+0.162
Neg logit contributions
ate-0.349
 army-0.337
amed-0.322
roc-0.320
ating-0.316
Token
Token ID220
Feature activation+5.44e-01

Pos logit contributions
 syrup+1.338
Pen+1.264
 kil+1.222
 marsh+1.207
 Pen+1.198
Neg logit contributions
ate-0.900
 army-0.835
amed-0.800
roc-0.779
ating-0.738
Token
Token ID198
Feature activation-6.32e-02

Pos logit contributions
ate+1.054
 army+1.033
roc+0.980
amed+0.979
ating+0.955
Neg logit contributions
 syrup-0.796
Pen-0.766
 Pen-0.734
 marsh-0.708
 Frost-0.695
Token
Token ID198
Feature activation-3.83e-02

Pos logit contributions
 syrup+0.103
Pen+0.096
 Pen+0.092
 marsh+0.091
 kil+0.090
Neg logit contributions
ate-0.116
 army-0.110
amed-0.107
roc-0.104
ating-0.102
TokenThe
Token ID464
Feature activation-4.43e-02

Pos logit contributions
 syrup+0.071
Pen+0.067
 Pen+0.065
 marsh+0.064
 kil+0.063
Neg logit contributions
ate-0.062
 army-0.058
amed-0.056
roc-0.055
ating-0.053
Token End
Token ID5268
Feature activation-2.16e+00

Pos logit contributions
 syrup+0.080
Pen+0.076
 Pen+0.073
 marsh+0.071
 kil+0.071
Neg logit contributions
ate-0.073
 army-0.070
amed-0.066
roc-0.065
ating-0.064
Token.
Token ID13
Feature activation+4.60e-02

Pos logit contributions
 syrup+3.794
Pen+3.625
 Pen+3.395
 marsh+3.294
 Willow+3.227
Neg logit contributions
ate-3.886
 army-3.611
room-3.415
roc-3.352
ating-3.331
Token<|endoftext|>
Token ID50256
Feature activation-1.41e-01

Pos logit contributions
ate+0.086
 army+0.082
amed+0.079
roc+0.078
ating+0.076
Neg logit contributions
 syrup-0.073
Pen-0.068
 Pen-0.065
 marsh-0.064
 kil-0.063
TokenOnce
Token ID7454
Feature activation+9.78e-02

Pos logit contributions
 syrup+0.226
Pen+0.213
 Pen+0.203
 marsh+0.199
 kil+0.196
Neg logit contributions
ate-0.261
 army-0.247
amed-0.238
roc-0.232
ating-0.229
Token upon
Token ID2402
Feature activation-3.91e-01

Pos logit contributions
ate+0.222
 army+0.215
amed+0.207
roc+0.206
ating+0.202
Neg logit contributions
 syrup-0.114
Pen-0.106
 Pen-0.099
 marsh-0.096
 kil-0.093
Token a
Token ID257
Feature activation-8.41e-01

Pos logit contributions
 syrup+0.670
Pen+0.633
 Pen+0.608
 marsh+0.599
 kil+0.596
Neg logit contributions
ate-0.678
 army-0.643
amed-0.614
roc-0.593
room-0.582
Token but
Token ID475
Feature activation-1.53e+00

Pos logit contributions
 syrup+2.985
Pen+2.954
 kil+2.750
 marsh+2.679
 Pen+2.650
Neg logit contributions
ate-3.475
 army-3.261
amed-3.052
ating-3.004
 friendships-2.956
Token to
Token ID284
Feature activation+4.54e-01

Pos logit contributions
 syrup+2.113
Pen+1.971
 kil+1.886
 marsh+1.864
 Pen+1.751
Neg logit contributions
ate-3.185
 army-2.896
ating-2.840
amed-2.810
roc-2.805
Token also
Token ID635
Feature activation+7.54e-01

Pos logit contributions
ate+0.910
 army+0.880
amed+0.851
roc+0.840
ating+0.821
Neg logit contributions
 syrup-0.652
Pen-0.613
 Pen-0.586
 marsh-0.568
 Frost-0.551
Token stay
Token ID2652
Feature activation+9.98e-01

Pos logit contributions
ate+1.647
 army+1.589
amed+1.556
roc+1.535
ating+1.513
Neg logit contributions
 syrup-0.943
Pen-0.870
 Pen-0.835
 marsh-0.805
 Frost-0.782
Token safe
Token ID3338
Feature activation+5.03e-01

Pos logit contributions
ate+1.995
 army+1.935
roc+1.919
amed+1.848
room+1.806
Neg logit contributions
 syrup-1.410
Pen-1.303
 Pen-1.254
 marsh-1.227
 tug-1.190
Token.
Token ID13
Feature activation-2.43e+00

Pos logit contributions
ate+0.998
 army+0.962
roc+0.928
amed+0.927
ating+0.890
Neg logit contributions
 syrup-0.728
Pen-0.682
 Pen-0.656
 marsh-0.638
 Frost-0.619
Token
Token ID220
Feature activation-8.29e-01

Pos logit contributions
 syrup+4.724
 kil+4.248
Pen+4.180
 tug+4.090
 marsh+4.039
Neg logit contributions
ate-4.016
amed-3.434
room-3.412
apped-3.335
 army-3.315
Token
Token ID198
Feature activation-1.70e+00

Pos logit contributions
 syrup+1.434
Pen+1.307
 kil+1.246
 marsh+1.246
 Pen+1.232
Neg logit contributions
ate-1.502
amed-1.350
 army-1.324
roc-1.275
ating-1.272
Token
Token ID198
Feature activation-1.97e+00

Pos logit contributions
 syrup+3.282
 kil+2.978
Pen+2.902
 marsh+2.872
 Pen+2.807
Neg logit contributions
ate-2.790
amed-2.522
room-2.321
ating-2.284
 army-2.281
TokenE
Token ID36
Feature activation-1.38e+00

Pos logit contributions
 syrup+3.325
 kil+2.966
 marsh+2.931
 Pen+2.856
 Frost+2.835
Neg logit contributions
ate-3.786
amed-3.509
ating-3.305
room-3.245
ined-3.226
Tokenlla
Token ID8466
Feature activation+2.35e-01

Pos logit contributions
 syrup+1.946
 marsh+1.715
Pen+1.698
 kil+1.613
 Pen+1.609
Neg logit contributions
ate-3.010
ating-2.784
amed-2.708
roc-2.692
 army-2.648
Token,
Token ID11
Feature activation-1.12e+00

Pos logit contributions
ate+0.537
 army+0.506
amed+0.480
roc+0.469
ating+0.452
Neg logit contributions
 syrup-0.831
Pen-0.794
 Pen-0.773
 marsh-0.758
 Frost-0.745
Token
Token ID6184
Feature activation-1.74e-01

Pos logit contributions
 syrup+2.242
Pen+2.078
 marsh+2.007
 kil+1.988
 Pen+1.909
Neg logit contributions
ate-1.623
 army-1.564
amed-1.488
roc-1.437
room-1.386
Token
Token ID95
Feature activation+5.51e-01

Pos logit contributions
 syrup+0.282
Pen+0.264
 Pen+0.253
 marsh+0.250
 kil+0.243
Neg logit contributions
ate-0.318
 army-0.304
roc-0.289
amed-0.288
ating-0.282
Token
Token ID26391
Feature activation+4.54e-01

Pos logit contributions
ate+0.918
 army+0.888
amed+0.842
roc+0.835
ating+0.807
Neg logit contributions
 syrup-0.958
Pen-0.926
 Pen-0.890
 marsh-0.857
 kil-0.846
Token
Token ID129
Feature activation+3.53e-01

Pos logit contributions
ate+0.777
 army+0.746
amed+0.711
roc+0.702
ating+0.681
Neg logit contributions
 syrup-0.776
Pen-0.743
 Pen-0.715
 marsh-0.699
 Willow-0.682
Token
Token ID241
Feature activation-2.17e+00

Pos logit contributions
ate+0.589
 army+0.564
amed+0.538
roc+0.531
ating+0.517
Neg logit contributions
 syrup-0.621
Pen-0.594
 Pen-0.571
 marsh-0.557
 Frost-0.545
TokenI
Token ID40
Feature activation-8.54e-01

Pos logit contributions
 syrup+3.628
 marsh+3.353
Pen+3.323
 kil+3.238
 Pen+3.221
Neg logit contributions
ate-3.989
amed-3.631
room-3.555
 army-3.536
roc-3.432
Token want
Token ID765
Feature activation-6.15e-01

Pos logit contributions
 syrup+1.215
Pen+1.180
 Pen+1.121
 marsh+1.062
 Willow+1.056
Neg logit contributions
ate-1.773
 army-1.652
roc-1.592
amed-1.588
ating-1.531
Token to
Token ID284
Feature activation+1.97e-01

Pos logit contributions
 syrup+0.815
Pen+0.795
 Pen+0.744
 marsh+0.723
 kil+0.714
Neg logit contributions
ate-1.289
 army-1.228
amed-1.171
roc-1.167
ating-1.131
Token pick
Token ID2298
Feature activation-1.90e-01

Pos logit contributions
ate+0.430
 army+0.415
amed+0.400
roc+0.397
ating+0.389
Neg logit contributions
 syrup-0.248
Pen-0.231
 Pen-0.218
 marsh-0.212
 kil-0.204
Token a
Token ID257
Feature activation-1.81e+00

Pos logit contributions
 syrup+0.274
Pen+0.262
 Pen+0.248
 marsh+0.241
 kil+0.236
Neg logit contributions
ate-0.377
 army-0.357
amed-0.344
roc-0.342
ating-0.334
Token He
Token ID679
Feature activation-1.04e+00

Pos logit contributions
 syrup+1.567
Pen+1.478
 marsh+1.393
 Pen+1.380
 kil+1.365
Neg logit contributions
ate-1.495
 army-1.401
amed-1.338
roc-1.327
room-1.259
Token took
Token ID1718
Feature activation-6.02e-01

Pos logit contributions
 syrup+1.554
Pen+1.494
 Pen+1.378
 marsh+1.371
 kil+1.349
Neg logit contributions
ate-2.050
 army-1.987
amed-1.923
roc-1.820
ating-1.805
Token his
Token ID465
Feature activation-4.82e-01

Pos logit contributions
 syrup+0.845
Pen+0.827
 Pen+0.769
 marsh+0.735
 kil+0.721
Neg logit contributions
ate-1.245
 army-1.185
amed-1.153
roc-1.108
ating-1.103
Token truck
Token ID7779
Feature activation+3.98e-01

Pos logit contributions
 syrup+0.914
Pen+0.888
 Pen+0.847
 marsh+0.815
 Willow+0.811
Neg logit contributions
ate-0.736
 army-0.732
amed-0.667
roc-0.657
ating-0.637
Token for
Token ID329
Feature activation-1.08e+00

Pos logit contributions
ate+0.732
 army+0.699
amed+0.669
roc+0.666
ating+0.649
Neg logit contributions
 syrup-0.637
Pen-0.600
 Pen-0.577
 marsh-0.565
 Frost-0.552
Token a
Token ID257
Feature activation-2.22e+00

Pos logit contributions
 syrup+1.769
Pen+1.762
 Pen+1.648
 marsh+1.590
 kil+1.563
Neg logit contributions
ate-1.919
 army-1.891
amed-1.749
roc-1.691
ating-1.682
Token drive
Token ID3708
Feature activation+2.26e-01

Pos logit contributions
 syrup+3.353
Pen+3.207
 Pen+3.022
 kil+2.929
 Megan+2.886
Neg logit contributions
 army-4.296
ate-3.893
 alien-3.885
amed-3.781
ully-3.771
Token and
Token ID290
Feature activation-1.87e-01

Pos logit contributions
ate+0.402
 army+0.385
amed+0.368
roc+0.364
ating+0.354
Neg logit contributions
 syrup-0.377
Pen-0.358
 Pen-0.343
 marsh-0.336
 kil-0.328
Token it
Token ID340
Feature activation-1.38e-01

Pos logit contributions
 syrup+0.250
Pen+0.239
 Pen+0.223
 marsh+0.215
 kil+0.213
Neg logit contributions
ate-0.393
 army-0.380
amed-0.366
roc-0.357
ating-0.353
Token felt
Token ID2936
Feature activation-4.00e-01

Pos logit contributions
 syrup+0.203
Pen+0.194
 Pen+0.182
 marsh+0.177
 kil+0.174
Neg logit contributions
ate-0.273
 army-0.262
amed-0.253
roc-0.246
ating-0.244
Token so
Token ID523
Feature activation+1.52e-01

Pos logit contributions
 syrup+0.578
Pen+0.562
 Pen+0.530
 marsh+0.501
 Willow+0.500
Neg logit contributions
ate-0.789
 army-0.765
amed-0.738
roc-0.712
ating-0.700
Token!
Token ID0
Feature activation-6.54e-01

Pos logit contributions
 syrup+0.806
Pen+0.764
 Pen+0.713
 marsh+0.710
 kil+0.682
Neg logit contributions
ate-1.026
 army-0.985
amed-0.942
ating-0.912
roc-0.910
Token Let
Token ID3914
Feature activation+1.61e-01

Pos logit contributions
 syrup+1.108
Pen+1.033
 marsh+0.996
 Pen+0.987
 kil+0.963
Neg logit contributions
ate-1.147
 army-1.077
amed-1.034
roc-1.015
room-0.990
Token's
Token ID338
Feature activation+1.85e-01

Pos logit contributions
ate+0.273
 army+0.261
amed+0.249
roc+0.246
ating+0.239
Neg logit contributions
 syrup-0.283
Pen-0.270
 Pen-0.259
 marsh-0.254
 kil-0.248
Token go
Token ID467
Feature activation-7.74e-01

Pos logit contributions
ate+0.389
 army+0.375
amed+0.361
roc+0.358
ating+0.350
Neg logit contributions
 syrup-0.249
Pen-0.233
 Pen-0.221
 marsh-0.215
 kil-0.208
Token for
Token ID329
Feature activation-1.07e+00

Pos logit contributions
 syrup+1.292
Pen+1.251
 Pen+1.186
 marsh+1.156
 kil+1.138
Neg logit contributions
ate-1.410
 army-1.338
amed-1.266
ating-1.220
room-1.208
Token a
Token ID257
Feature activation-2.58e+00

Pos logit contributions
 syrup+1.859
Pen+1.825
 Pen+1.736
 kil+1.701
 marsh+1.668
Neg logit contributions
ate-1.792
 army-1.731
amed-1.586
roc-1.543
ating-1.506
Token ride
Token ID6594
Feature activation+5.32e-02

Pos logit contributions
 syrup+3.941
Pen+3.809
 Pen+3.659
 kil+3.540
 Megan+3.450
Neg logit contributions
 army-5.035
ate-4.564
 alien-4.516
ully-4.383
 nation-4.255
Token.
Token ID13
Feature activation-8.11e-01

Pos logit contributions
ate+0.104
 army+0.100
amed+0.096
roc+0.094
ating+0.093
Neg logit contributions
 syrup-0.079
Pen-0.075
 Pen-0.071
 marsh-0.070
 kil-0.068
Tokenâ
Token ID22940
Feature activation+1.04e+00

Pos logit contributions
 syrup+1.394
Pen+1.297
 marsh+1.242
 Pen+1.234
 kil+1.207
Neg logit contributions
ate-1.409
 army-1.326
amed-1.259
roc-1.248
room-1.226
Token
Token ID26391
Feature activation-3.79e-01

Pos logit contributions
ate+1.853
 army+1.834
amed+1.703
roc+1.702
ating+1.623
Neg logit contributions
 syrup-1.690
Pen-1.627
 Pen-1.553
 marsh-1.476
 kil-1.472
Token
Token ID198
Feature activation-7.05e-01

Pos logit contributions
 syrup+0.604
Pen+0.563
 Pen+0.535
 marsh+0.525
 kil+0.512
Neg logit contributions
ate-0.712
 army-0.679
amed-0.650
roc-0.646
ating-0.629