TOP ACTIVATIONS
MAX = 4.125

. They wished they had not fought . They wished they
. They wished they had not broken the scale and the
. They wished they had not fought over their toys .
. They wished they had not played with the blocks .
. They wished they had not taken the pasta . They
. They wished they had not disagreed . But
. They wished they had not destroyed their toys and their
. They wished they had not opened the box . They
. They wish they did not sell it . They are
. They wish they had not played with the paper .
two cubes . One cube was very boring , and it
. They wished they had not pushed the shelf . They
. They wished they had not made such a big tower
. They wished they had not cut the map . They
my 's mom found out what had happened and she was
too late . The man had cut all the corn and
problem . Ben 's pants are too big for him .
. They wished they had not been sour . They wished
on the news that there was going to be a big
toys . But their toys are messy . They are everywhere

MIN ACTIVATIONS
MIN = -4.711

. She would fly up and up , then come back
flew over hills , mountains , rivers and trees . The
Every day , she would soar up high into the sky
the butterfly . She flew around in circles and then the
pond . She flew higher and higher , feeling proud of
. She flew over hills , mountains , rivers and trees
soared high over the trees , buildings and people . Everything
! She flew ever higher and higher until she reached the
would go on adventures , exploring the world around her .
safe . She soared higher and higher until she could touch
Sally watched it fly higher and higher . Then , she
same cozy blanket . They sat in the sun , chatting
free . She flew higher and higher into the sky .
day on , Dove practiced flying every day and soon became
sky . He looked up and saw the big , blue
the tree . She would fly around and play with her
became great friends . They explored the peaceful garden together and
sky . He soared up and through the clouds , feeling
off ! He flew higher and higher and was amazed at
tw irled around the sun and spun in the breeze .

INTERVAL 1.916 - 4.125
CONTAINS 1.679%

day , Lily and her friends were playing hide and seek
in the ocean with his family . One day , the
. He had a pet dog called Bos co . Bos
truck driver was angry and hurt . He had a cut
, until suddenly the bag broke . All of his toys

INTERVAL -0.293 - 1.916
CONTAINS 61.244%

" No , I don 't have any ," Lily
making different animals and shapes with the clay . Suddenly ,
too because it was no longer rotting away . The bird
!" he said excited ly . " Wow , Tim my
He had to decide . He could wish for a puppy

INTERVAL -2.502 - -0.293
CONTAINS 36.543%

and tw irled in the light and smiled .
take turns . They have fun and laugh together . They
the bird . She learned that even if someone doesn 't
They are like us . They can wave and talk and
the buttons together . The little girl did as she was

INTERVAL -4.711 - -2.502
CONTAINS 0.534%

pond nearby . Sharon walked closer and saw a little girl
Sally . She liked to explore the world . One day
One day , Lily was exploring when she found a big
Gr ace started walking home , making sure not
" Yes , let 's sing together !" They sang and
Token.
Token ID13
Feature activation+8.93e-01

Pos logit contributions
 tweet+1.477
 bou+1.424
 anytime+1.415
 forests+1.400
 galaxies+1.351
Neg logit contributions
 stolen-0.908
illing-0.885
 supposed-0.884
 delayed-0.851
 thrown-0.822
Token They
Token ID1119
Feature activation+1.13e+00

Pos logit contributions
 stolen+1.356
 supposed+1.289
 robbed+1.229
illing+1.225
 thrown+1.208
Neg logit contributions
 anytime-1.474
 tweet-1.472
 bou-1.429
 forests-1.414
 valleys-1.348
Token wished
Token ID16555
Feature activation-6.34e-02

Pos logit contributions
illing+1.661
 supposed+1.629
 stolen+1.607
 robbed+1.527
 sale+1.523
Neg logit contributions
 tweet-1.852
 bou-1.724
 anytime-1.723
 forests-1.659
 sings-1.637
Token they
Token ID484
Feature activation+1.22e+00

Pos logit contributions
 tweet+0.099
 anytime+0.097
 bou+0.094
 forests+0.092
 galaxies+0.089
Neg logit contributions
 stolen-0.098
 supposed-0.094
illing-0.093
 delayed-0.090
 robbed-0.089
Token had
Token ID550
Feature activation+1.88e+00

Pos logit contributions
illing+1.781
 stolen+1.718
 supposed+1.711
 Timothy+1.654
omo+1.637
Neg logit contributions
 anytime-1.968
 tweet-1.915
 forests-1.819
 bou-1.765
 galaxies-1.763
Token not
Token ID407
Feature activation+3.96e+00

Pos logit contributions
illing+2.902
 Timothy+2.645
 paramed+2.542
omo+2.538
 stolen+2.512
Neg logit contributions
 anytime-3.074
 tweet-3.031
 forests-2.834
 bou-2.821
 sings-2.738
Token fought
Token ID8350
Feature activation+1.56e+00

Pos logit contributions
 Timothy+7.257
 paramed+6.776
illing+6.475
omm+6.415
omo+6.366
Neg logit contributions
 tweet-6.016
 anytime-5.826
 sings-5.571
 forests-5.457
 bou-5.319
Token.
Token ID13
Feature activation+6.08e-01

Pos logit contributions
 stolen+3.222
illing+3.128
 supposed+3.055
omo+3.039
 paramed+3.024
Neg logit contributions
 anytime-1.803
 tweet-1.615
 forests-1.503
 bou-1.442
 wherever-1.421
Token They
Token ID1119
Feature activation+1.54e+00

Pos logit contributions
 stolen+0.873
 supposed+0.836
illing+0.812
 robbed+0.787
 delayed+0.780
Neg logit contributions
 anytime-1.040
 tweet-1.032
 bou-0.992
 forests-0.989
 galaxies-0.939
Token wished
Token ID16555
Feature activation-2.72e-02

Pos logit contributions
illing+2.602
 supposed+2.552
 stolen+2.408
 Timothy+2.342
omo+2.318
Neg logit contributions
 tweet-2.208
 anytime-2.132
 bou-2.119
 forests-2.057
 galaxies-1.992
Token they
Token ID484
Feature activation+1.44e+00

Pos logit contributions
 tweet+0.042
 anytime+0.041
 bou+0.040
 forests+0.039
 galaxies+0.038
Neg logit contributions
 stolen-0.042
 supposed-0.041
illing-0.041
 delayed-0.039
 robbed-0.039
Token.
Token ID13
Feature activation+6.28e-01

Pos logit contributions
 stolen+1.089
 supposed+1.064
illing+1.063
omo+1.019
 robbed+1.008
Neg logit contributions
 anytime-0.998
 tweet-0.955
 bou-0.922
 forests-0.894
 galaxies-0.845
Token They
Token ID1119
Feature activation+1.26e+00

Pos logit contributions
 stolen+0.978
 supposed+0.938
illing+0.924
 delayed+0.889
 robbed+0.887
Neg logit contributions
 anytime-0.987
 tweet-0.978
 forests-0.930
 bou-0.923
 galaxies-0.891
Token wished
Token ID16555
Feature activation+2.52e-01

Pos logit contributions
illing+2.043
 supposed+1.972
 stolen+1.905
omo+1.865
 Timothy+1.833
Neg logit contributions
 tweet-1.871
 anytime-1.814
 bou-1.761
 forests-1.742
 galaxies-1.698
Token they
Token ID484
Feature activation+1.64e+00

Pos logit contributions
 stolen+0.414
 supposed+0.402
illing+0.400
 delayed+0.384
 robbed+0.380
Neg logit contributions
 tweet-0.370
 anytime-0.364
 bou-0.348
 forests-0.342
 galaxies-0.327
Token had
Token ID550
Feature activation+2.02e+00

Pos logit contributions
illing+2.560
 supposed+2.378
omo+2.368
 stolen+2.366
 Timothy+2.287
Neg logit contributions
 anytime-2.564
 tweet-2.423
 forests-2.306
 wherever-2.288
 galaxies-2.247
Token not
Token ID407
Feature activation+3.94e+00

Pos logit contributions
illing+3.202
omo+2.843
 Timothy+2.813
 paramed+2.736
 stolen+2.718
Neg logit contributions
 anytime-3.271
 tweet-3.241
 bou-2.979
 forests-2.938
 sings-2.883
Token broken
Token ID5445
Feature activation+6.48e-01

Pos logit contributions
 Timothy+6.786
illing+6.625
omo+6.406
 paramed+6.318
omm+6.200
Neg logit contributions
 tweet-6.133
 anytime-5.981
 sings-5.477
 forests-5.418
 again-5.288
Token the
Token ID262
Feature activation-8.57e-01

Pos logit contributions
 stolen+1.174
illing+1.173
 supposed+1.148
 delayed+1.111
 robbed+1.096
Neg logit contributions
 anytime-0.837
 tweet-0.800
 bou-0.788
 forests-0.784
 galaxies-0.735
Token scale
Token ID5046
Feature activation+1.09e+00

Pos logit contributions
 tweet+1.342
 anytime+1.296
 bou+1.263
 forests+1.217
 galaxies+1.179
Neg logit contributions
 stolen-1.382
 supposed-1.322
illing-1.277
 delayed-1.269
 unfair-1.250
Token and
Token ID290
Feature activation+1.80e+00

Pos logit contributions
 stolen+1.749
illing+1.723
 supposed+1.715
omo+1.664
 Timothy+1.623
Neg logit contributions
 anytime-1.673
 tweet-1.595
 bou-1.528
 forests-1.474
 wherever-1.401
Token the
Token ID262
Feature activation-1.56e-01

Pos logit contributions
illing+3.397
omo+3.207
 supposed+3.161
 stolen+3.135
 Timothy+3.083
Neg logit contributions
 anytime-2.307
 tweet-2.267
 forests-2.118
 bou-2.060
 galaxies-2.018
Token.
Token ID13
Feature activation+1.02e+00

Pos logit contributions
 tweet+0.260
 anytime+0.252
 bou+0.244
 forests+0.237
 galaxies+0.225
Neg logit contributions
 stolen-0.387
 supposed-0.377
illing-0.374
 delayed-0.365
 thrown-0.360
Token They
Token ID1119
Feature activation+1.17e+00

Pos logit contributions
 stolen+1.573
 supposed+1.496
illing+1.447
 delayed+1.410
 robbed+1.409
Neg logit contributions
 anytime-1.664
 tweet-1.622
 forests-1.568
 bou-1.551
 galaxies-1.498
Token wished
Token ID16555
Feature activation+1.28e-01

Pos logit contributions
illing+1.756
 supposed+1.688
 stolen+1.647
omo+1.627
 Timothy+1.573
Neg logit contributions
 tweet-1.892
 anytime-1.780
 bou-1.760
 forests-1.717
 sings-1.681
Token they
Token ID484
Feature activation+1.73e+00

Pos logit contributions
 stolen+0.206
 supposed+0.200
illing+0.199
 delayed+0.192
 robbed+0.189
Neg logit contributions
 tweet-0.191
 anytime-0.187
 bou-0.180
 forests-0.176
 galaxies-0.169
Token had
Token ID550
Feature activation+2.12e+00

Pos logit contributions
illing+2.710
 supposed+2.510
 stolen+2.489
omo+2.482
 Timothy+2.423
Neg logit contributions
 anytime-2.701
 tweet-2.551
 forests-2.437
 wherever-2.410
 galaxies-2.361
Token not
Token ID407
Feature activation+3.93e+00

Pos logit contributions
illing+3.463
omo+3.082
 Timothy+3.078
 paramed+2.928
 supposed+2.925
Neg logit contributions
 anytime-3.297
 tweet-3.275
 bou-2.988
 forests-2.941
 sings-2.908
Token fought
Token ID8350
Feature activation+1.63e+00

Pos logit contributions
 Timothy+7.157
illing+6.759
omo+6.717
 paramed+6.468
omm+6.435
Neg logit contributions
 tweet-5.869
 anytime-5.602
 sings-5.188
 forests-5.054
 watches-4.883
Token over
Token ID625
Feature activation+4.90e-01

Pos logit contributions
 stolen+3.446
illing+3.391
 supposed+3.302
omo+3.302
 robbed+3.230
Neg logit contributions
 anytime-1.750
 tweet-1.560
 forests-1.418
 wherever-1.371
 bou-1.370
Token their
Token ID511
Feature activation-3.66e-01

Pos logit contributions
 stolen+0.888
illing+0.877
 supposed+0.873
 delayed+0.839
omo+0.834
Neg logit contributions
 anytime-0.625
 tweet-0.618
 bou-0.587
 forests-0.583
 galaxies-0.548
Token toys
Token ID14958
Feature activation+6.69e-01

Pos logit contributions
 tweet+0.639
 anytime+0.620
 bou+0.607
 forests+0.593
 galaxies+0.573
Neg logit contributions
 stolen-0.504
 supposed-0.484
illing-0.476
 delayed-0.461
 robbed-0.452
Token.
Token ID13
Feature activation+6.28e-01

Pos logit contributions
 stolen+1.089
 supposed+1.064
illing+1.063
omo+1.019
 robbed+1.008
Neg logit contributions
 anytime-0.998
 tweet-0.955
 bou-0.922
 forests-0.894
 galaxies-0.845
Token.
Token ID13
Feature activation+1.23e+00

Pos logit contributions
 tweet+0.541
 anytime+0.524
 bou+0.516
 forests+0.504
 galaxies+0.482
Neg logit contributions
 stolen-0.573
 supposed-0.558
illing-0.554
 delayed-0.537
 robbed-0.526
Token They
Token ID1119
Feature activation+1.33e+00

Pos logit contributions
 stolen+2.009
 supposed+1.896
 robbed+1.803
 delayed+1.780
illing+1.778
Neg logit contributions
 anytime-1.964
 tweet-1.886
 bou-1.851
 forests-1.815
 galaxies-1.773
Token wished
Token ID16555
Feature activation-5.09e-01

Pos logit contributions
illing+2.159
 supposed+2.108
 stolen+2.056
 robbed+1.977
omo+1.955
Neg logit contributions
 tweet-2.037
 bou-1.908
 anytime-1.885
 sings-1.757
 forests-1.748
Token they
Token ID484
Feature activation+1.54e+00

Pos logit contributions
 tweet+0.822
 anytime+0.797
 bou+0.789
 forests+0.766
 galaxies+0.736
Neg logit contributions
 stolen-0.756
 supposed-0.730
illing-0.722
 delayed-0.705
 thrown-0.692
Token had
Token ID550
Feature activation+1.86e+00

Pos logit contributions
illing+2.367
 supposed+2.236
 stolen+2.212
omo+2.176
 Timothy+2.127
Neg logit contributions
 anytime-2.455
 tweet-2.333
 forests-2.189
 wherever-2.145
 bou-2.132
Token not
Token ID407
Feature activation+3.91e+00

Pos logit contributions
illing+2.952
omo+2.648
 Timothy+2.606
 supposed+2.568
 paramed+2.552
Neg logit contributions
 anytime-2.959
 tweet-2.945
 bou-2.703
 forests-2.651
 sings-2.603
Token played
Token ID2826
Feature activation+1.55e-02

Pos logit contributions
 Timothy+6.621
illing+6.544
omo+6.394
 paramed+6.381
omm+6.333
Neg logit contributions
 tweet-6.362
 anytime-6.093
 sings-5.566
 forests-5.524
 bou-5.381
Token with
Token ID351
Feature activation+3.99e-01

Pos logit contributions
 stolen+0.029
 supposed+0.028
illing+0.028
 delayed+0.027
 robbed+0.027
Neg logit contributions
 tweet-0.019
 anytime-0.019
 bou-0.018
 forests-0.018
 galaxies-0.017
Token the
Token ID262
Feature activation+6.34e-01

Pos logit contributions
 stolen+0.720
illing+0.712
 supposed+0.711
 delayed+0.681
omo+0.673
Neg logit contributions
 tweet-0.512
 anytime-0.509
 bou-0.486
 forests-0.476
 galaxies-0.447
Token blocks
Token ID7021
Feature activation+5.34e-01

Pos logit contributions
 stolen+1.007
illing+0.993
 supposed+0.993
 delayed+0.948
 thrown+0.940
Neg logit contributions
 tweet-0.962
 anytime-0.940
 bou-0.911
 forests-0.897
 galaxies-0.844
Token.
Token ID13
Feature activation+7.53e-01

Pos logit contributions
 stolen+0.891
 supposed+0.871
illing+0.862
omo+0.827
 delayed+0.826
Neg logit contributions
 anytime-0.773
 tweet-0.752
 bou-0.720
 forests-0.696
 galaxies-0.660
Token.
Token ID13
Feature activation+1.18e+00

Pos logit contributions
 stolen+0.171
 supposed+0.165
illing+0.163
 delayed+0.157
 robbed+0.156
Neg logit contributions
 tweet-0.183
 anytime-0.180
 bou-0.174
 forests-0.170
 galaxies-0.163
Token They
Token ID1119
Feature activation+1.59e+00

Pos logit contributions
 stolen+1.838
 supposed+1.809
illing+1.729
 robbed+1.674
 thrown+1.644
Neg logit contributions
 anytime-1.878
 tweet-1.873
 forests-1.805
 bou-1.760
 valleys-1.736
Token wished
Token ID16555
Feature activation+7.38e-02

Pos logit contributions
illing+2.466
 supposed+2.314
omo+2.220
 stolen+2.178
 sale+2.152
Neg logit contributions
 tweet-2.585
 bou-2.385
 anytime-2.378
 forests-2.312
 sings-2.273
Token they
Token ID484
Feature activation+2.06e+00

Pos logit contributions
 stolen+0.118
 supposed+0.114
illing+0.113
 delayed+0.109
 robbed+0.108
Neg logit contributions
 tweet-0.112
 anytime-0.110
 bou-0.105
 forests-0.104
 galaxies-0.099
Token had
Token ID550
Feature activation+2.35e+00

Pos logit contributions
illing+3.479
omo+3.153
 supposed+3.133
 paramed+3.088
 sale+3.051
Neg logit contributions
 anytime-3.088
 tweet-2.923
 wherever-2.781
 forests-2.747
 galaxies-2.587
Token not
Token ID407
Feature activation+3.91e+00

Pos logit contributions
illing+3.836
omo+3.326
 paramed+3.218
attered+3.217
 Timothy+3.202
Neg logit contributions
 tweet-3.816
 anytime-3.793
 bou-3.466
 forests-3.420
 sings-3.334
Token taken
Token ID2077
Feature activation+5.49e-01

Pos logit contributions
illing+6.865
 paramed+6.783
 Timothy+6.698
omo+6.649
omm+6.534
Neg logit contributions
 tweet-6.152
 anytime-6.026
 again-5.355
 forests-5.343
 sings-5.272
Token the
Token ID262
Feature activation+2.78e-01

Pos logit contributions
illing+1.039
 stolen+1.035
 supposed+1.022
 delayed+0.991
omo+0.988
Neg logit contributions
 anytime-0.655
 tweet-0.645
 bou-0.611
 forests-0.608
 galaxies-0.561
Token pasta
Token ID26296
Feature activation+1.00e+00

Pos logit contributions
 stolen+0.282
 supposed+0.273
illing+0.272
 delayed+0.256
 thrown+0.251
Neg logit contributions
 tweet-0.570
 anytime-0.562
 bou-0.553
 forests-0.542
 galaxies-0.526
Token.
Token ID13
Feature activation+1.14e+00

Pos logit contributions
 stolen+1.536
 supposed+1.504
illing+1.499
omo+1.444
 Timothy+1.431
Neg logit contributions
 anytime-1.599
 tweet-1.506
 bou-1.485
 forests-1.420
 wherever-1.356
Token They
Token ID1119
Feature activation+1.40e+00

Pos logit contributions
 stolen+1.789
 supposed+1.726
illing+1.681
 robbed+1.628
 thrown+1.602
Neg logit contributions
 anytime-1.801
 tweet-1.763
 forests-1.710
 valleys-1.641
 bou-1.635
Token.
Token ID13
Feature activation+1.28e+00

Pos logit contributions
 stolen+1.218
 supposed+1.155
illing+1.128
 robbed+1.101
 delayed+1.096
Neg logit contributions
 anytime-1.280
 tweet-1.262
 bou-1.182
 forests-1.178
 galaxies-1.122
Token They
Token ID1119
Feature activation+1.31e+00

Pos logit contributions
 stolen+2.178
 supposed+2.083
illing+1.991
 robbed+1.981
 delayed+1.969
Neg logit contributions
 anytime-1.894
 tweet-1.828
 forests-1.793
 bou-1.749
 valleys-1.698
Token wished
Token ID16555
Feature activation+1.77e-02

Pos logit contributions
illing+1.966
 supposed+1.877
 stolen+1.811
omo+1.802
 robbed+1.741
Neg logit contributions
 tweet-2.140
 anytime-2.006
 bou-2.002
 forests-1.959
 sings-1.919
Token they
Token ID484
Feature activation+1.62e+00

Pos logit contributions
 stolen+0.028
 supposed+0.027
illing+0.027
 delayed+0.026
 robbed+0.026
Neg logit contributions
 tweet-0.027
 anytime-0.027
 bou-0.026
 forests-0.025
 galaxies-0.024
Token had
Token ID550
Feature activation+2.10e+00

Pos logit contributions
illing+2.569
 supposed+2.372
 stolen+2.345
omo+2.328
 paramed+2.310
Neg logit contributions
 anytime-2.569
 tweet-2.424
 forests-2.311
 wherever-2.264
 galaxies-2.215
Token not
Token ID407
Feature activation+3.89e+00

Pos logit contributions
illing+3.401
omo+2.983
 Timothy+2.941
 paramed+2.922
attered+2.843
Neg logit contributions
 anytime-3.398
 tweet-3.331
 bou-3.107
 forests-3.059
 sings-2.981
Token disagreed
Token ID27884
Feature activation+2.20e+00

Pos logit contributions
illing+7.017
 Timothy+6.949
omo+6.862
 paramed+6.744
omm+6.498
Neg logit contributions
 anytime-5.887
 tweet-5.827
 forests-5.296
 sings-5.291
 wherever-5.042
Token.
Token ID13
Feature activation+1.05e+00

Pos logit contributions
 stolen+4.391
 paramed+4.076
illing+4.066
 robbed+4.048
 supposed+4.038
Neg logit contributions
 anytime-2.928
 tweet-2.438
 until-2.337
 wherever-2.321
 again-2.278
Token
Token ID198
Feature activation+3.04e-03

Pos logit contributions
 stolen+1.994
 supposed+1.923
illing+1.867
 robbed+1.836
 delayed+1.829
Neg logit contributions
 anytime-1.335
 tweet-1.295
 forests-1.237
 bou-1.206
 valleys-1.150
Token
Token ID198
Feature activation+1.56e+00

Pos logit contributions
 stolen+0.006
 supposed+0.006
illing+0.006
 delayed+0.005
 robbed+0.005
Neg logit contributions
 tweet-0.004
 anytime-0.004
 bou-0.003
 forests-0.003
 galaxies-0.003
TokenBut
Token ID1537
Feature activation+6.26e-01

Pos logit contributions
 stolen+2.888
 supposed+2.823
 unfair+2.718
illing+2.642
 robbed+2.622
Neg logit contributions
 anytime-2.099
 tweet-1.947
 forests-1.863
 bou-1.808
 until-1.724
Token.
Token ID13
Feature activation+2.87e-01

Pos logit contributions
 stolen+0.150
 supposed+0.145
illing+0.143
 delayed+0.138
 robbed+0.136
Neg logit contributions
 tweet-0.179
 anytime-0.178
 bou-0.171
 forests-0.168
 galaxies-0.162
Token They
Token ID1119
Feature activation+1.13e+00

Pos logit contributions
 stolen+0.457
 supposed+0.439
illing+0.435
 delayed+0.421
 robbed+0.419
Neg logit contributions
 tweet-0.439
 anytime-0.434
 bou-0.414
 forests-0.411
 galaxies-0.393
Token wished
Token ID16555
Feature activation-3.99e-01

Pos logit contributions
illing+1.873
 supposed+1.828
 stolen+1.774
 Timothy+1.683
 robbed+1.677
Neg logit contributions
 tweet-1.643
 anytime-1.624
 bou-1.588
 forests-1.534
 galaxies-1.500
Token they
Token ID484
Feature activation+9.02e-01

Pos logit contributions
 tweet+0.643
 anytime+0.626
 bou+0.615
 forests+0.599
 galaxies+0.574
Neg logit contributions
 stolen-0.595
 supposed-0.574
illing-0.566
 delayed-0.553
 thrown-0.543
Token had
Token ID550
Feature activation+1.65e+00

Pos logit contributions
illing+1.295
 stolen+1.256
 supposed+1.240
omo+1.192
 Timothy+1.167
Neg logit contributions
 anytime-1.499
 tweet-1.460
 forests-1.376
 bou-1.358
 galaxies-1.342
Token not
Token ID407
Feature activation+3.85e+00

Pos logit contributions
illing+2.563
 stolen+2.246
 supposed+2.241
 Timothy+2.241
omo+2.232
Neg logit contributions
 anytime-2.693
 tweet-2.638
 bou-2.499
 forests-2.477
 sings-2.365
Token destroyed
Token ID6572
Feature activation+1.52e-01

Pos logit contributions
illing+6.477
 Timothy+6.384
 paramed+6.243
omo+6.029
omm+5.948
Neg logit contributions
 tweet-6.230
 anytime-6.180
 forests-5.809
 sings-5.625
 again-5.458
Token their
Token ID511
Feature activation+3.43e-01

Pos logit contributions
 stolen+0.278
illing+0.271
 supposed+0.270
 delayed+0.261
 robbed+0.258
Neg logit contributions
 tweet-0.191
 anytime-0.191
 bou-0.182
 forests-0.178
 galaxies-0.168
Token toys
Token ID14958
Feature activation+4.96e-01

Pos logit contributions
 stolen+0.561
illing+0.557
 supposed+0.556
 delayed+0.529
 robbed+0.520
Neg logit contributions
 tweet-0.493
 anytime-0.488
 bou-0.474
 forests-0.462
 galaxies-0.437
Token and
Token ID290
Feature activation+1.23e+00

Pos logit contributions
 stolen+0.783
 supposed+0.765
illing+0.765
omo+0.727
 delayed+0.724
Neg logit contributions
 anytime-0.758
 tweet-0.732
 bou-0.706
 forests-0.689
 galaxies-0.654
Token their
Token ID511
Feature activation+6.42e-01

Pos logit contributions
illing+1.956
 stolen+1.841
 supposed+1.831
omo+1.799
 Timothy+1.744
Neg logit contributions
 anytime-1.906
 tweet-1.882
 forests-1.779
 bou-1.767
 galaxies-1.722
Token.
Token ID13
Feature activation+1.18e+00

Pos logit contributions
 stolen+1.616
 supposed+1.558
illing+1.503
 robbed+1.469
omo+1.457
Neg logit contributions
 anytime-1.560
 tweet-1.544
 bou-1.432
 forests-1.423
 until-1.344
Token They
Token ID1119
Feature activation+1.34e+00

Pos logit contributions
 stolen+1.867
 supposed+1.797
illing+1.757
 robbed+1.706
 thrown+1.666
Neg logit contributions
 anytime-1.855
 tweet-1.832
 forests-1.768
 bou-1.758
 galaxies-1.688
Token wished
Token ID16555
Feature activation+2.18e-01

Pos logit contributions
illing+2.165
 supposed+2.086
 stolen+2.009
omo+1.962
 sale+1.939
Neg logit contributions
 tweet-2.051
 bou-1.932
 anytime-1.893
 forests-1.811
 sings-1.797
Token they
Token ID484
Feature activation+1.42e+00

Pos logit contributions
 stolen+0.356
 supposed+0.345
illing+0.345
 delayed+0.330
 robbed+0.327
Neg logit contributions
 tweet-0.322
 anytime-0.317
 bou-0.304
 forests-0.297
 galaxies-0.285
Token had
Token ID550
Feature activation+2.19e+00

Pos logit contributions
illing+2.132
 supposed+1.987
 stolen+1.980
omo+1.941
 Timothy+1.907
Neg logit contributions
 anytime-2.280
 tweet-2.202
 forests-2.065
 bou-2.019
 wherever-2.017
Token not
Token ID407
Feature activation+3.83e+00

Pos logit contributions
illing+3.625
omo+3.166
 Timothy+3.102
 supposed+3.017
attered+3.012
Neg logit contributions
 tweet-3.451
 anytime-3.448
 bou-3.201
 forests-3.089
 sings-3.056
Token opened
Token ID4721
Feature activation+1.24e-01

Pos logit contributions
illing+6.725
 Timothy+6.511
omo+6.389
 paramed+6.168
omm+6.083
Neg logit contributions
 tweet-6.067
 anytime-5.828
 sings-5.395
 bou-5.368
 forests-5.265
Token the
Token ID262
Feature activation+2.86e-01

Pos logit contributions
 stolen+0.226
illing+0.221
 supposed+0.221
 delayed+0.213
 robbed+0.210
Neg logit contributions
 tweet-0.155
 anytime-0.154
 bou-0.147
 forests-0.143
 galaxies-0.136
Token box
Token ID3091
Feature activation+8.75e-01

Pos logit contributions
 stolen+0.514
illing+0.509
 supposed+0.506
 delayed+0.488
 thrown+0.484
Neg logit contributions
 tweet-0.370
 anytime-0.360
 bou-0.354
 forests-0.341
 galaxies-0.324
Token.
Token ID13
Feature activation+9.96e-01

Pos logit contributions
 stolen+1.372
 supposed+1.357
illing+1.344
omo+1.289
 delayed+1.276
Neg logit contributions
 anytime-1.367
 tweet-1.294
 bou-1.274
 forests-1.229
 galaxies-1.165
Token They
Token ID1119
Feature activation+1.24e+00

Pos logit contributions
 stolen+1.554
 supposed+1.497
illing+1.472
 robbed+1.409
 thrown+1.385
Neg logit contributions
 anytime-1.584
 tweet-1.561
 forests-1.497
 bou-1.479
 galaxies-1.436
Token.
Token ID13
Feature activation+7.98e-01

Pos logit contributions
 tweet+1.648
 bou+1.599
 anytime+1.594
 forests+1.566
 galaxies+1.498
Neg logit contributions
 stolen-1.156
 supposed-1.140
illing-1.127
 delayed-1.103
 thrown-1.051
Token They
Token ID1119
Feature activation-3.83e-02

Pos logit contributions
 stolen+1.534
 supposed+1.479
illing+1.445
 robbed+1.418
 delayed+1.412
Neg logit contributions
 anytime-0.997
 tweet-0.991
 forests-0.919
 bou-0.911
 galaxies-0.861
Token wish
Token ID4601
Feature activation-5.27e-01

Pos logit contributions
 tweet+0.050
 anytime+0.049
 bou+0.047
 forests+0.046
 galaxies+0.044
Neg logit contributions
 stolen-0.068
 supposed-0.066
illing-0.066
 delayed-0.064
 robbed-0.063
Token they
Token ID484
Feature activation+7.60e-01

Pos logit contributions
 tweet+0.856
 anytime+0.838
 bou+0.824
 forests+0.803
 galaxies+0.769
Neg logit contributions
 stolen-0.772
 supposed-0.744
illing-0.734
 delayed-0.718
 thrown-0.707
Token did
Token ID750
Feature activation+1.91e-01

Pos logit contributions
illing+1.043
 stolen+1.032
 supposed+1.007
omo+0.975
 Timothy+0.970
Neg logit contributions
 anytime-1.290
 tweet-1.271
 forests-1.194
 bou-1.183
 galaxies-1.158
Token not
Token ID407
Feature activation+3.82e+00

Pos logit contributions
 stolen+0.199
illing+0.194
 supposed+0.189
 delayed+0.178
omo+0.175
Neg logit contributions
 tweet-0.389
 anytime-0.388
 bou-0.374
 forests-0.369
 galaxies-0.358
Token sell
Token ID3677
Feature activation+6.98e-01

Pos logit contributions
illing+7.732
 Timothy+7.655
omo+7.629
attered+7.412
 paramed+7.408
Neg logit contributions
 tweet-4.938
 anytime-4.818
 sings-4.136
 wherever-4.024
 flap-4.017
Token it
Token ID340
Feature activation+1.29e+00

Pos logit contributions
illing+1.287
 stolen+1.273
 supposed+1.260
 delayed+1.218
omo+1.214
Neg logit contributions
 anytime-0.886
 tweet-0.841
 bou-0.822
 forests-0.806
 galaxies-0.744
Token.
Token ID13
Feature activation+4.49e-01

Pos logit contributions
illing+2.162
 stolen+2.160
omo+2.119
 supposed+2.083
 robbed+2.028
Neg logit contributions
 anytime-1.913
 tweet-1.795
 bou-1.691
 forests-1.638
 wherever-1.619
Token They
Token ID1119
Feature activation-2.71e-02

Pos logit contributions
 stolen+0.785
 supposed+0.754
illing+0.745
 delayed+0.722
 robbed+0.721
Neg logit contributions
 tweet-0.621
 anytime-0.620
 forests-0.580
 bou-0.579
 galaxies-0.552
Token are
Token ID389
Feature activation+1.62e+00

Pos logit contributions
 tweet+0.032
 anytime+0.031
 bou+0.030
 forests+0.029
 galaxies+0.027
Neg logit contributions
 stolen-0.052
 supposed-0.051
illing-0.051
 delayed-0.049
 robbed-0.049
Token.
Token ID13
Feature activation+5.59e-01

Pos logit contributions
 stolen+0.004
 supposed+0.004
illing+0.004
 delayed+0.004
 robbed+0.004
Neg logit contributions
 tweet-0.005
 anytime-0.005
 bou-0.005
 forests-0.005
 galaxies-0.005
Token They
Token ID1119
Feature activation-9.06e-02

Pos logit contributions
 stolen+0.993
 supposed+0.962
illing+0.941
 robbed+0.917
 delayed+0.916
Neg logit contributions
 tweet-0.772
 anytime-0.759
 bou-0.712
 forests-0.712
 galaxies-0.672
Token wish
Token ID4601
Feature activation-1.78e-01

Pos logit contributions
 tweet+0.120
 anytime+0.117
 bou+0.113
 forests+0.109
 galaxies+0.104
Neg logit contributions
 stolen-0.161
 supposed-0.157
illing-0.155
 delayed-0.151
 robbed-0.149
Token they
Token ID484
Feature activation+9.36e-01

Pos logit contributions
 tweet+0.277
 anytime+0.271
 bou+0.263
 forests+0.257
 galaxies+0.247
Neg logit contributions
 stolen-0.275
 supposed-0.266
illing-0.263
 delayed-0.255
 robbed-0.251
Token had
Token ID550
Feature activation+2.27e+00

Pos logit contributions
illing+1.329
 stolen+1.292
 supposed+1.270
omo+1.242
 Timothy+1.241
Neg logit contributions
 anytime-1.552
 tweet-1.522
 forests-1.423
 bou-1.409
 galaxies-1.391
Token not
Token ID407
Feature activation+3.79e+00

Pos logit contributions
illing+3.688
 Timothy+3.182
omo+3.182
 paramed+3.173
 supposed+3.095
Neg logit contributions
 anytime-3.730
 tweet-3.594
 bou-3.418
 forests-3.292
 sings-3.261
Token played
Token ID2826
Feature activation+8.96e-01

Pos logit contributions
 Timothy+6.302
illing+6.280
omo+6.116
 paramed+6.016
omm+5.762
Neg logit contributions
 tweet-6.216
 anytime-6.106
 sings-5.563
 bou-5.457
 wherever-5.413
Token with
Token ID351
Feature activation+8.44e-01

Pos logit contributions
 stolen+1.726
 supposed+1.708
illing+1.704
omo+1.655
 delayed+1.636
Neg logit contributions
 anytime-1.072
 tweet-1.022
 bou-0.980
 forests-0.948
 galaxies-0.879
Token the
Token ID262
Feature activation-3.31e-01

Pos logit contributions
illing+1.548
 supposed+1.523
 stolen+1.517
 delayed+1.463
omo+1.441
Neg logit contributions
 anytime-1.083
 tweet-1.062
 bou-1.026
 forests-1.008
 galaxies-0.948
Token paper
Token ID3348
Feature activation+5.26e-01

Pos logit contributions
 tweet+0.509
 anytime+0.499
 bou+0.480
 forests+0.471
 galaxies+0.451
Neg logit contributions
 stolen-0.525
 supposed-0.506
illing-0.498
 delayed-0.485
 robbed-0.479
Token.
Token ID13
Feature activation+4.78e-01

Pos logit contributions
 stolen+0.763
 supposed+0.746
illing+0.739
 delayed+0.708
omo+0.694
Neg logit contributions
 anytime-0.865
 tweet-0.864
 bou-0.836
 forests-0.807
 galaxies-0.777
Token two
Token ID734
Feature activation+9.72e-01

Pos logit contributions
 stolen+1.183
 supposed+1.133
illing+1.118
 robbed+1.086
 delayed+1.080
Neg logit contributions
 tweet-1.236
 anytime-1.227
 bou-1.204
 forests-1.185
 galaxies-1.112
Token cubes
Token ID34896
Feature activation+2.74e+00

Pos logit contributions
 stolen+1.677
 supposed+1.661
illing+1.613
 delayed+1.586
 robbed+1.572
Neg logit contributions
 tweet-1.342
 anytime-1.309
 bou-1.305
 forests-1.257
 galaxies-1.162
Token.
Token ID13
Feature activation+2.64e+00

Pos logit contributions
 unfair+5.123
 stolen+4.988
 supposed+4.869
 delayed+4.856
 cancelled+4.839
Neg logit contributions
 anytime-3.629
 tweet-3.433
 bou-3.357
 forests-3.136
 wherever-3.100
Token One
Token ID1881
Feature activation+2.22e+00

Pos logit contributions
 supposed+5.017
 stolen+4.905
 disob+4.603
illing+4.593
 robbed+4.585
Neg logit contributions
 anytime-3.902
 tweet-3.789
 forests-3.674
 bou-3.552
 Moon-3.531
Token cube
Token ID23441
Feature activation+2.03e+00

Pos logit contributions
 supposed+4.274
 stolen+4.233
 Timothy+4.045
 robbed+4.017
olded+4.002
Neg logit contributions
 anytime-2.722
 tweet-2.530
 bou-2.375
 forests-2.365
 moment-2.306
Token was
Token ID373
Feature activation+3.78e+00

Pos logit contributions
 supposed+4.110
 stolen+3.988
illing+3.926
 unfair+3.874
 cancelled+3.841
Neg logit contributions
 anytime-2.305
 tweet-2.145
 forests-1.981
 bou-1.952
 sings-1.922
Token very
Token ID845
Feature activation+2.92e+00

Pos logit contributions
olded+9.045
illing+8.990
beh+8.841
 supposed+8.731
 cancelled+8.699
Neg logit contributions
 forests-3.133
 bou-3.099
 higher-3.006
 until-2.723
 mountains-2.674
Token boring
Token ID14262
Feature activation+1.89e+00

Pos logit contributions
 supposed+6.131
 robbed+6.092
 stolen+6.047
 cancelled+6.015
olded+5.980
Neg logit contributions
 bou-2.842
 forests-2.754
 anytime-2.669
 tweet-2.590
 higher-2.533
Token,
Token ID11
Feature activation+1.31e+00

Pos logit contributions
 stolen+4.163
illing+3.988
omo+3.982
 paramed+3.957
 robbed+3.955
Neg logit contributions
 anytime-1.820
 forests-1.619
 galaxies-1.508
 bou-1.477
 tweet-1.471
Token and
Token ID290
Feature activation+1.59e+00

Pos logit contributions
 stolen+2.668
 supposed+2.642
illing+2.610
 robbed+2.516
 delayed+2.498
Neg logit contributions
 anytime-1.434
 tweet-1.415
 forests-1.332
 bou-1.316
 galaxies-1.205
Token it
Token ID340
Feature activation+2.14e+00

Pos logit contributions
illing+2.969
 supposed+2.947
 stolen+2.927
 robbed+2.779
 delayed+2.730
Neg logit contributions
 tweet-2.000
 anytime-1.991
 forests-1.859
 bou-1.799
 galaxies-1.762
Token.
Token ID13
Feature activation+3.76e-01

Pos logit contributions
 stolen+1.806
 supposed+1.764
illing+1.680
 robbed+1.661
omo+1.643
Neg logit contributions
 anytime-1.716
 tweet-1.697
 bou-1.588
 forests-1.522
 until-1.505
Token They
Token ID1119
Feature activation+8.74e-01

Pos logit contributions
 stolen+0.541
 supposed+0.518
illing+0.503
 delayed+0.491
 robbed+0.486
Neg logit contributions
 anytime-0.634
 tweet-0.634
 bou-0.607
 forests-0.601
 galaxies-0.577
Token wished
Token ID16555
Feature activation-6.43e-01

Pos logit contributions
illing+1.331
 supposed+1.316
 stolen+1.299
 robbed+1.228
omo+1.217
Neg logit contributions
 tweet-1.379
 bou-1.310
 anytime-1.308
 forests-1.228
 sings-1.204
Token they
Token ID484
Feature activation+1.14e+00

Pos logit contributions
 tweet+1.059
 anytime+1.020
 bou+1.018
 forests+0.984
 galaxies+0.943
Neg logit contributions
 stolen-0.937
 supposed-0.905
illing-0.897
 delayed-0.879
 thrown-0.855
Token had
Token ID550
Feature activation+1.95e+00

Pos logit contributions
illing+1.698
 stolen+1.632
 supposed+1.630
omo+1.575
 Timothy+1.538
Neg logit contributions
 anytime-1.841
 tweet-1.786
 forests-1.677
 bou-1.641
 galaxies-1.639
Token not
Token ID407
Feature activation+3.78e+00

Pos logit contributions
illing+3.228
omo+2.923
 Timothy+2.871
 paramed+2.797
 stolen+2.796
Neg logit contributions
 anytime-2.974
 tweet-2.957
 bou-2.710
 forests-2.668
 sings-2.640
Token pushed
Token ID7121
Feature activation+2.61e-01

Pos logit contributions
 Timothy+6.509
illing+6.399
omo+6.357
 paramed+6.305
omm+6.158
Neg logit contributions
 tweet-5.824
 anytime-5.601
 sings-5.301
 forests-5.033
 wherever-4.949
Token the
Token ID262
Feature activation-1.15e+00

Pos logit contributions
 stolen+0.474
 supposed+0.461
illing+0.461
 delayed+0.446
 robbed+0.441
Neg logit contributions
 tweet-0.335
 anytime-0.333
 bou-0.316
 forests-0.311
 galaxies-0.293
Token shelf
Token ID18316
Feature activation+9.30e-01

Pos logit contributions
 tweet+1.815
 bou+1.715
 anytime+1.710
 forests+1.632
 galaxies+1.594
Neg logit contributions
 stolen-1.837
 supposed-1.763
illing-1.693
 delayed-1.677
 robbed-1.665
Token.
Token ID13
Feature activation+3.85e-01

Pos logit contributions
 stolen+1.743
 supposed+1.727
illing+1.705
omo+1.660
 delayed+1.635
Neg logit contributions
 anytime-1.171
 tweet-1.095
 bou-1.053
 forests-1.011
 galaxies-0.945
Token They
Token ID1119
Feature activation+1.18e+00

Pos logit contributions
 stolen+0.587
 supposed+0.564
illing+0.552
 delayed+0.533
 robbed+0.531
Neg logit contributions
 anytime-0.613
 tweet-0.612
 bou-0.582
 forests-0.580
 galaxies-0.557
Token.
Token ID13
Feature activation+6.67e-01

Pos logit contributions
 stolen+1.887
 supposed+1.848
illing+1.759
 Timothy+1.730
 robbed+1.728
Neg logit contributions
 anytime-1.769
 tweet-1.756
 bou-1.634
 forests-1.581
 until-1.562
Token They
Token ID1119
Feature activation+8.92e-01

Pos logit contributions
 stolen+0.969
 supposed+0.914
illing+0.886
 delayed+0.870
 robbed+0.867
Neg logit contributions
 anytime-1.134
 tweet-1.119
 forests-1.078
 bou-1.076
 galaxies-1.041
Token wished
Token ID16555
Feature activation-3.08e-01

Pos logit contributions
illing+1.422
 supposed+1.388
 stolen+1.383
omo+1.312
 robbed+1.309
Neg logit contributions
 tweet-1.345
 anytime-1.261
 bou-1.259
 forests-1.188
 sings-1.158
Token they
Token ID484
Feature activation+1.27e+00

Pos logit contributions
 tweet+0.489
 anytime+0.476
 bou+0.467
 forests+0.454
 galaxies+0.435
Neg logit contributions
 stolen-0.467
 supposed-0.452
illing-0.448
 delayed-0.435
 thrown-0.428
Token had
Token ID550
Feature activation+2.00e+00

Pos logit contributions
illing+1.848
 stolen+1.768
 supposed+1.746
omo+1.700
 sale+1.676
Neg logit contributions
 anytime-2.070
 tweet-1.992
 forests-1.875
 galaxies-1.831
 wherever-1.829
Token not
Token ID407
Feature activation+3.76e+00

Pos logit contributions
illing+3.317
omo+2.945
 Timothy+2.931
 stolen+2.876
 supposed+2.857
Neg logit contributions
 anytime-3.042
 tweet-2.999
 bou-2.723
 forests-2.680
 sings-2.645
Token made
Token ID925
Feature activation+9.94e-01

Pos logit contributions
illing+6.600
 Timothy+6.459
 paramed+6.203
omo+6.117
 paramedics+5.943
Neg logit contributions
 tweet-5.935
 anytime-5.601
 sings-5.192
 bou-4.969
 forests-4.913
Token such
Token ID884
Feature activation+1.46e+00

Pos logit contributions
 stolen+1.712
illing+1.692
 supposed+1.620
 delayed+1.600
 robbed+1.595
Neg logit contributions
 anytime-1.396
 tweet-1.340
 bou-1.306
 forests-1.273
 galaxies-1.213
Token a
Token ID257
Feature activation-5.33e-01

Pos logit contributions
illing+2.379
 stolen+2.278
 supposed+2.182
 driven+2.144
 robbed+2.143
Neg logit contributions
 anytime-2.178
 tweet-2.173
 bou-2.119
 forests-2.041
 galaxies-2.034
Token big
Token ID1263
Feature activation-6.76e-01

Pos logit contributions
 tweet+0.917
 anytime+0.888
 bou+0.869
 forests+0.859
 galaxies+0.830
Neg logit contributions
 stolen-0.753
 supposed-0.724
illing-0.701
 delayed-0.688
 unfair-0.676
Token tower
Token ID10580
Feature activation+7.14e-01

Pos logit contributions
 tweet+1.237
 anytime+1.194
 bou+1.168
 forests+1.156
 galaxies+1.122
Neg logit contributions
 stolen-0.880
 supposed-0.832
illing-0.815
 delayed-0.795
 robbed-0.785
Token.
Token ID13
Feature activation+5.93e-01

Pos logit contributions
 tweet+0.041
 anytime+0.040
 bou+0.039
 forests+0.038
 galaxies+0.037
Neg logit contributions
 stolen-0.031
 supposed-0.029
illing-0.029
 delayed-0.028
 robbed-0.028
Token They
Token ID1119
Feature activation+1.34e+00

Pos logit contributions
 stolen+0.842
 supposed+0.802
illing+0.782
 robbed+0.758
 delayed+0.757
Neg logit contributions
 tweet-1.020
 anytime-1.007
 bou-0.968
 forests-0.967
 galaxies-0.926
Token wished
Token ID16555
Feature activation-1.09e-01

Pos logit contributions
illing+2.106
 supposed+2.013
 stolen+1.967
omo+1.927
 sale+1.892
Neg logit contributions
 tweet-2.113
 bou-1.952
 anytime-1.927
 forests-1.847
 sings-1.827
Token they
Token ID484
Feature activation+1.86e+00

Pos logit contributions
 tweet+0.168
 anytime+0.164
 bou+0.159
 forests+0.156
 galaxies+0.149
Neg logit contributions
 stolen-0.170
 supposed-0.165
illing-0.163
 delayed-0.158
 robbed-0.156
Token had
Token ID550
Feature activation+2.51e+00

Pos logit contributions
illing+2.973
omo+2.709
 supposed+2.703
 stolen+2.678
 sale+2.669
Neg logit contributions
 anytime-2.855
 tweet-2.719
 forests-2.565
 wherever-2.534
 galaxies-2.480
Token not
Token ID407
Feature activation+3.76e+00

Pos logit contributions
illing+4.272
omo+3.649
 Timothy+3.648
 paramed+3.523
attered+3.520
Neg logit contributions
 anytime-3.938
 tweet-3.898
 bou-3.488
 sings-3.462
 forests-3.417
Token cut
Token ID2005
Feature activation+4.87e-01

Pos logit contributions
illing+6.397
 Timothy+6.259
omo+6.089
omm+5.943
 paramed+5.923
Neg logit contributions
 tweet-6.158
 anytime-5.709
 sings-5.386
 bou-5.113
 wherever-4.973
Token the
Token ID262
Feature activation-1.04e+00

Pos logit contributions
 stolen+0.923
illing+0.908
 supposed+0.903
 delayed+0.873
omo+0.861
Neg logit contributions
 anytime-0.590
 tweet-0.583
 bou-0.551
 forests-0.545
 galaxies-0.505
Token map
Token ID3975
Feature activation+9.70e-01

Pos logit contributions
 tweet+1.333
 anytime+1.328
 bou+1.224
 forests+1.174
 galaxies+1.112
Neg logit contributions
 stolen-2.009
 supposed-1.890
 robbed-1.848
 delayed-1.826
illing-1.826
Token.
Token ID13
Feature activation+7.96e-01

Pos logit contributions
 stolen+1.642
illing+1.632
 supposed+1.618
omo+1.573
 delayed+1.541
Neg logit contributions
 anytime-1.384
 tweet-1.334
 bou-1.283
 forests-1.228
 galaxies-1.158
Token They
Token ID1119
Feature activation+1.49e+00

Pos logit contributions
 stolen+1.186
 supposed+1.131
illing+1.104
 robbed+1.067
 delayed+1.060
Neg logit contributions
 tweet-1.319
 anytime-1.317
 forests-1.257
 bou-1.235
 galaxies-1.206
Tokenmy
Token ID1820
Feature activation+1.18e+00

Pos logit contributions
 stolen+2.213
illing+2.208
 supposed+2.129
 sale+2.119
 unfair+2.100
Neg logit contributions
 anytime-1.762
 tweet-1.708
 forests-1.599
 bou-1.562
 sings-1.515
Token's
Token ID338
Feature activation+1.65e+00

Pos logit contributions
illing+2.127
 supposed+2.117
 stolen+2.105
 unfair+2.029
 Timothy+2.022
Neg logit contributions
 anytime-1.464
 tweet-1.401
 forests-1.361
 bou-1.332
 sings-1.275
Token mom
Token ID1995
Feature activation+2.04e+00

Pos logit contributions
illing+2.668
 supposed+2.648
 stolen+2.500
 delayed+2.481
 thrown+2.452
Neg logit contributions
 tweet-2.407
 anytime-2.404
 bou-2.365
 forests-2.262
 wherever-2.111
Token found
Token ID1043
Feature activation+1.60e+00

Pos logit contributions
illing+3.329
 stolen+3.236
 supposed+3.207
 unfair+3.196
 Timothy+3.095
Neg logit contributions
 anytime-3.008
 tweet-2.845
 bou-2.747
 forests-2.593
 sings-2.559
Token out
Token ID503
Feature activation+1.23e+00

Pos logit contributions
illing+3.053
omo+2.928
 supposed+2.848
 stolen+2.823
 delayed+2.770
Neg logit contributions
 anytime-2.039
 tweet-1.991
 bou-1.885
 forests-1.873
 wherever-1.769
Token what
Token ID644
Feature activation+3.75e+00

Pos logit contributions
illing+2.219
 stolen+2.187
 supposed+2.174
omo+2.168
 delayed+2.102
Neg logit contributions
 tweet-1.562
 anytime-1.546
 bou-1.446
 forests-1.435
 galaxies-1.350
Token had
Token ID550
Feature activation+2.42e+00

Pos logit contributions
illing+7.855
olded+7.364
ction+7.353
ooked+7.346
omo+7.285
Neg logit contributions
 anytime-4.385
 tweet-4.022
 forests-3.837
 sings-3.781
 wherever-3.734
Token happened
Token ID3022
Feature activation+4.49e-01

Pos logit contributions
illing+2.771
omo+2.651
 paramed+2.409
 cigarette+2.268
 mercury+2.189
Neg logit contributions
 anytime-4.871
 tweet-4.803
 bou-4.720
 wherever-4.610
 forests-4.548
Token and
Token ID290
Feature activation+1.39e+00

Pos logit contributions
 stolen+0.753
illing+0.725
 supposed+0.719
omo+0.706
 delayed+0.703
Neg logit contributions
 anytime-0.630
 tweet-0.623
 bou-0.592
 forests-0.583
 galaxies-0.551
Token she
Token ID673
Feature activation+1.26e+00

Pos logit contributions
illing+2.339
 supposed+2.230
 stolen+2.144
omo+2.138
 unfair+2.052
Neg logit contributions
 anytime-2.053
 tweet-2.029
 bou-1.949
 forests-1.850
 galaxies-1.725
Token was
Token ID373
Feature activation+2.10e+00

Pos logit contributions
illing+1.953
 supposed+1.880
 stolen+1.824
 unfair+1.769
omo+1.742
Neg logit contributions
 anytime-1.963
 tweet-1.957
 bou-1.907
 forests-1.780
 sings-1.712
Token too
Token ID1165
Feature activation+1.51e+00

Pos logit contributions
illing+2.710
 stolen+2.669
omo+2.647
 Timothy+2.588
 paramed+2.550
Neg logit contributions
 anytime-2.224
 tweet-2.180
 higher-2.045
 forests-2.006
 bou-1.865
Token late
Token ID2739
Feature activation+4.46e-01

Pos logit contributions
 stolen+2.721
 supposed+2.578
 robbed+2.524
illing+2.515
 sale+2.494
Neg logit contributions
 anytime-1.959
 tweet-1.869
 forests-1.766
 bou-1.691
 valleys-1.674
Token.
Token ID13
Feature activation+1.58e+00

Pos logit contributions
 stolen+0.642
 supposed+0.618
illing+0.616
 delayed+0.589
omo+0.587
Neg logit contributions
 anytime-0.738
 tweet-0.738
 bou-0.697
 forests-0.691
 galaxies-0.660
Token The
Token ID383
Feature activation+1.37e-01

Pos logit contributions
 supposed+2.952
 stolen+2.945
illing+2.834
 robbed+2.687
 delayed+2.682
Neg logit contributions
 anytime-2.103
 tweet-1.966
 forests-1.873
 valleys-1.760
 bou-1.728
Token man
Token ID582
Feature activation+1.33e+00

Pos logit contributions
 stolen+0.251
 supposed+0.246
illing+0.244
 delayed+0.236
 thrown+0.234
Neg logit contributions
 tweet-0.175
 anytime-0.170
 bou-0.164
 forests-0.161
 galaxies-0.152
Token had
Token ID550
Feature activation+3.71e+00

Pos logit contributions
 supposed+2.100
illing+2.057
 stolen+1.978
omo+1.926
 Timothy+1.925
Neg logit contributions
 tweet-2.034
 anytime-2.013
 forests-1.854
 bou-1.837
 sings-1.769
Token cut
Token ID2005
Feature activation+9.06e-01

Pos logit contributions
illing+7.641
omo+6.592
attered+6.589
 supposed+6.403
 Timothy+6.292
Neg logit contributions
 anytime-4.944
 tweet-4.722
 forests-4.640
 wherever-4.335
 bou-4.233
Token all
Token ID477
Feature activation+1.40e+00

Pos logit contributions
 stolen+1.620
 supposed+1.612
illing+1.598
 delayed+1.529
omo+1.513
Neg logit contributions
 anytime-1.175
 tweet-1.163
 forests-1.100
 bou-1.073
 valleys-1.019
Token the
Token ID262
Feature activation-1.44e+00

Pos logit contributions
 stolen+2.501
illing+2.434
 supposed+2.406
 delayed+2.385
 robbed+2.357
Neg logit contributions
 anytime-1.872
 tweet-1.777
 forests-1.744
 bou-1.683
 valleys-1.616
Token corn
Token ID11676
Feature activation+5.94e-02

Pos logit contributions
 tweet+2.260
 anytime+2.187
 bou+2.080
 galaxies+1.958
 forests+1.949
Neg logit contributions
 stolen-2.419
 supposed-2.309
illing-2.259
 delayed-2.231
 burned-2.198
Token and
Token ID290
Feature activation+1.21e+00

Pos logit contributions
 stolen+0.117
 supposed+0.114
illing+0.113
 delayed+0.110
 robbed+0.109
Neg logit contributions
 tweet-0.067
 anytime-0.066
 bou-0.062
 forests-0.061
 galaxies-0.057
Token problem
Token ID1917
Feature activation+2.91e+00

Pos logit contributions
 stolen+0.642
 supposed+0.635
illing+0.632
 delayed+0.595
 thrown+0.591
Neg logit contributions
 tweet-0.779
 anytime-0.746
 bou-0.741
 forests-0.718
 galaxies-0.682
Token.
Token ID13
Feature activation+2.23e+00

Pos logit contributions
 stolen+5.681
 paramed+5.675
 unfair+5.503
omo+5.375
illing+5.285
Neg logit contributions
 anytime-3.990
 forests-3.323
 tweet-3.290
 until-3.265
 wherever-3.129
Token Ben
Token ID3932
Feature activation+2.86e-01

Pos logit contributions
 supposed+3.969
 stolen+3.952
 delayed+3.529
 thrown+3.512
 cancelled+3.504
Neg logit contributions
 anytime-3.271
 tweet-3.176
 forests-3.163
 galaxies-2.931
 valleys-2.905
Token's
Token ID338
Feature activation+5.00e-01

Pos logit contributions
 stolen+0.540
 supposed+0.530
illing+0.527
 delayed+0.506
omo+0.505
Neg logit contributions
 tweet-0.340
 anytime-0.334
 bou-0.314
 forests-0.314
 galaxies-0.294
Token pants
Token ID12581
Feature activation+1.44e+00

Pos logit contributions
 stolen+0.766
illing+0.751
 supposed+0.749
 delayed+0.712
 robbed+0.709
Neg logit contributions
 tweet-0.782
 anytime-0.767
 bou-0.739
 forests-0.726
 galaxies-0.692
Token are
Token ID389
Feature activation+3.71e+00

Pos logit contributions
illing+2.452
 stolen+2.353
 supposed+2.346
omo+2.341
 paramed+2.280
Neg logit contributions
 tweet-2.024
 anytime-2.020
 forests-1.839
 bou-1.838
 sings-1.766
Token too
Token ID1165
Feature activation+3.28e+00

Pos logit contributions
illing+7.523
 paramed+7.044
 Timothy+6.842
ird+6.789
olded+6.673
Neg logit contributions
 higher-4.534
 until-4.473
 tweet-4.448
 anytime-4.441
 wherever-4.435
Token big
Token ID1263
Feature activation+8.66e-01

Pos logit contributions
 blamed+6.792
ction+6.718
 supposed+6.682
 paramed+6.658
 Timothy+6.622
Neg logit contributions
 anytime-3.330
 higher-3.276
 bou-2.970
 tweet-2.850
 forests-2.741
Token for
Token ID329
Feature activation+2.13e+00

Pos logit contributions
 stolen+1.944
illing+1.914
omo+1.886
 supposed+1.874
 delayed+1.849
Neg logit contributions
 anytime-0.722
 tweet-0.700
 bou-0.631
 forests-0.622
 galaxies-0.558
Token him
Token ID683
Feature activation+2.26e+00

Pos logit contributions
 stolen+3.834
 supposed+3.739
illing+3.718
attered+3.653
olded+3.645
Neg logit contributions
 anytime-3.065
 forests-2.930
 tweet-2.930
 bou-2.668
 deserts-2.597
Token.
Token ID13
Feature activation+2.07e+00

Pos logit contributions
omo+4.251
 paramed+4.247
illing+4.216
 stolen+4.108
 unfair+4.083
Neg logit contributions
 anytime-2.985
 tweet-2.748
 wherever-2.655
 until-2.515
 forests-2.453
Token.
Token ID13
Feature activation+3.03e-01

Pos logit contributions
 stolen+2.291
illing+2.283
 supposed+2.219
 robbed+2.143
 delayed+2.140
Neg logit contributions
 anytime-2.193
 tweet-2.102
 bou-2.044
 forests-1.990
 galaxies-1.936
Token They
Token ID1119
Feature activation+1.06e+00

Pos logit contributions
 stolen+0.484
 supposed+0.464
illing+0.461
 delayed+0.445
 robbed+0.443
Neg logit contributions
 tweet-0.461
 anytime-0.457
 bou-0.433
 forests-0.432
 galaxies-0.414
Token wished
Token ID16555
Feature activation-2.74e-01

Pos logit contributions
illing+1.739
 supposed+1.693
 stolen+1.662
 Timothy+1.570
 robbed+1.567
Neg logit contributions
 tweet-1.536
 anytime-1.519
 bou-1.474
 forests-1.428
 galaxies-1.398
Token they
Token ID484
Feature activation+9.35e-01

Pos logit contributions
 tweet+0.433
 anytime+0.423
 bou+0.413
 forests+0.403
 galaxies+0.386
Neg logit contributions
 stolen-0.417
 supposed-0.403
illing-0.398
 delayed-0.388
 thrown-0.381
Token had
Token ID550
Feature activation+1.52e+00

Pos logit contributions
illing+1.329
 stolen+1.287
 supposed+1.268
omo+1.230
 Timothy+1.199
Neg logit contributions
 anytime-1.563
 tweet-1.525
 forests-1.437
 bou-1.416
 galaxies-1.402
Token not
Token ID407
Feature activation+3.71e+00

Pos logit contributions
illing+2.327
 stolen+2.072
omo+2.059
 supposed+2.053
 Timothy+2.051
Neg logit contributions
 anytime-2.488
 tweet-2.450
 bou-2.319
 forests-2.290
 sings-2.190
Token been
Token ID587
Feature activation+2.58e+00

Pos logit contributions
illing+6.043
 Timothy+5.865
 paramed+5.686
omo+5.646
ooked+5.423
Neg logit contributions
 tweet-6.183
 anytime-6.099
 forests-5.712
 sings-5.564
 bou-5.474
Token sour
Token ID11348
Feature activation+9.53e-01

Pos logit contributions
illing+4.595
omo+4.508
 Timothy+4.411
 paramed+4.159
 stolen+4.130
Neg logit contributions
 anytime-3.639
 tweet-3.425
 higher-3.263
 forests-3.202
 bou-3.101
Token.
Token ID13
Feature activation+4.78e-01

Pos logit contributions
 stolen+1.878
 supposed+1.843
illing+1.804
omo+1.764
 thrown+1.762
Neg logit contributions
 anytime-1.109
 tweet-1.047
 bou-0.992
 forests-0.965
 galaxies-0.901
Token They
Token ID1119
Feature activation+1.03e+00

Pos logit contributions
 stolen+0.765
 supposed+0.729
illing+0.725
 delayed+0.699
 robbed+0.698
Neg logit contributions
 tweet-0.727
 anytime-0.723
 forests-0.682
 bou-0.679
 galaxies-0.655
Token wished
Token ID16555
Feature activation-3.86e-01

Pos logit contributions
illing+1.655
 supposed+1.605
 stolen+1.585
omo+1.497
 Timothy+1.492
Neg logit contributions
 tweet-1.549
 anytime-1.526
 bou-1.472
 forests-1.443
 galaxies-1.412
Token on
Token ID319
Feature activation+8.39e-01

Pos logit contributions
 stolen+3.866
 supposed+3.819
illing+3.735
omo+3.697
 robbed+3.577
Neg logit contributions
 tweet-2.710
 anytime-2.678
 forests-2.652
 bou-2.599
 galaxies-2.546
Token the
Token ID262
Feature activation+1.14e+00

Pos logit contributions
 stolen+1.590
illing+1.570
 supposed+1.565
 delayed+1.500
omo+1.492
Neg logit contributions
 tweet-0.999
 anytime-0.984
 bou-0.969
 forests-0.964
 galaxies-0.891
Token news
Token ID1705
Feature activation+1.68e+00

Pos logit contributions
 stolen+1.808
illing+1.760
omo+1.716
 robbed+1.692
 supposed+1.687
Neg logit contributions
 tweet-1.721
 bou-1.646
 anytime-1.640
 forests-1.609
 galaxies-1.546
Token that
Token ID326
Feature activation+2.63e+00

Pos logit contributions
 stolen+2.812
omo+2.775
 supposed+2.656
illing+2.637
 Timothy+2.582
Neg logit contributions
 anytime-2.415
 tweet-2.367
 bou-2.340
 forests-2.236
 sings-2.134
Token there
Token ID612
Feature activation+2.51e+00

Pos logit contributions
illing+4.511
omo+4.220
 stolen+4.175
 supposed+4.175
ction+4.065
Neg logit contributions
 anytime-3.853
 tweet-3.789
 forests-3.720
 sings-3.698
 galaxies-3.612
Token was
Token ID373
Feature activation+3.69e+00

Pos logit contributions
 stolen+5.036
illing+5.007
 supposed+4.942
olded+4.839
 robbed+4.810
Neg logit contributions
 anytime-2.994
 tweet-2.827
 forests-2.694
 bou-2.581
 sings-2.543
Token going
Token ID1016
Feature activation+2.60e+00

Pos logit contributions
illing+8.591
olded+7.882
attered+7.793
issors+7.655
 supposed+7.582
Neg logit contributions
 anytime-4.264
 forests-4.190
 galaxies-3.679
 cities-3.643
 again-3.596
Token to
Token ID284
Feature activation+1.37e+00

Pos logit contributions
omo+5.639
illing+5.379
 stolen+5.250
 blamed+5.216
olded+5.146
Neg logit contributions
 anytime-2.754
 forests-2.553
 tweet-2.498
 higher-2.439
 wherever-2.432
Token be
Token ID307
Feature activation+2.00e+00

Pos logit contributions
illing+2.434
 stolen+2.369
 supposed+2.298
olded+2.294
omo+2.256
Neg logit contributions
 tweet-1.916
 anytime-1.899
 bou-1.786
 forests-1.761
 galaxies-1.621
Token a
Token ID257
Feature activation+4.31e-01

Pos logit contributions
illing+3.332
 stolen+3.163
omo+3.128
 supposed+3.076
 Timothy+3.065
Neg logit contributions
 anytime-3.085
 forests-2.882
 tweet-2.707
 bou-2.679
 galaxies-2.657
Token big
Token ID1263
Feature activation-1.06e-01

Pos logit contributions
 stolen+0.615
illing+0.608
 supposed+0.602
 thrown+0.570
omo+0.567
Neg logit contributions
 tweet-0.710
 anytime-0.691
 bou-0.679
 forests-0.655
 galaxies-0.625
Token toys
Token ID14958
Feature activation+1.43e+00

Pos logit contributions
illing+2.057
 supposed+2.005
 stolen+1.969
 delayed+1.942
 robbed+1.887
Neg logit contributions
 anytime-1.982
 tweet-1.953
 bou-1.952
 forests-1.845
 wherever-1.723
Token.
Token ID13
Feature activation+2.20e+00

Pos logit contributions
 stolen+2.313
illing+2.275
 supposed+2.269
omo+2.154
 Timothy+2.151
Neg logit contributions
 anytime-2.281
 tweet-2.044
 bou-2.021
 forests-1.938
 wherever-1.930
Token But
Token ID887
Feature activation+1.55e+00

Pos logit contributions
 supposed+3.875
 stolen+3.802
 robbed+3.552
illing+3.481
 delayed+3.450
Neg logit contributions
 anytime-3.382
 tweet-3.201
 forests-3.175
 bou-3.042
 nods-3.027
Token their
Token ID511
Feature activation+1.10e+00

Pos logit contributions
 supposed+2.576
illing+2.567
 stolen+2.472
 paramed+2.325
 robbed+2.315
Neg logit contributions
 anytime-2.473
 tweet-2.361
 forests-2.168
 bou-2.141
 wherever-2.096
Token toys
Token ID14958
Feature activation+1.71e+00

Pos logit contributions
illing+1.931
 supposed+1.868
 stolen+1.867
 Timothy+1.797
 delayed+1.784
Neg logit contributions
 anytime-1.484
 tweet-1.471
 bou-1.426
 forests-1.383
 galaxies-1.289
Token are
Token ID389
Feature activation+3.68e+00

Pos logit contributions
illing+3.619
 supposed+3.492
 stolen+3.446
omo+3.411
 paramed+3.402
Neg logit contributions
 tweet-1.748
 anytime-1.747
 bou-1.500
 sings-1.462
 wherever-1.450
Token messy
Token ID24097
Feature activation+1.27e+00

Pos logit contributions
illing+7.478
 Timothy+7.335
 paramed+7.129
Ice+6.803
olded+6.699
Neg logit contributions
 anytime-4.520
 higher-4.488
 tweet-4.402
 wherever-4.142
 until-4.137
Token.
Token ID13
Feature activation+3.19e+00

Pos logit contributions
 stolen+2.322
illing+2.231
 supposed+2.184
 paramed+2.158
 robbed+2.142
Neg logit contributions
 anytime-1.690
 forests-1.493
 tweet-1.491
 bou-1.478
 galaxies-1.423
Token They
Token ID1119
Feature activation+9.36e-01

Pos logit contributions
 supposed+6.607
 stolen+6.289
 delayed+5.692
 robbed+5.684
 cancelled+5.579
Neg logit contributions
 anytime-4.507
 tweet-4.443
 forests-4.181
 nods-4.169
 cities-3.909
Token are
Token ID389
Feature activation+2.74e+00

Pos logit contributions
 stolen+1.608
illing+1.603
 supposed+1.587
omo+1.526
 Timothy+1.514
Neg logit contributions
 tweet-1.273
 anytime-1.233
 bou-1.175
 forests-1.146
 galaxies-1.109
Token everywhere
Token ID8347
Feature activation+7.89e-01

Pos logit contributions
 Timothy+4.621
illing+4.603
Ice+4.319
 stolen+4.241
 paramed+4.191
Neg logit contributions
 anytime-3.862
 tweet-3.856
 until-3.749
 wherever-3.629
 bou-3.604
Token.
Token ID13
Feature activation-2.08e+00

Pos logit contributions
 tweet+4.060
 bou+3.925
 anytime+3.737
 forests+3.695
 galaxies+3.679
Neg logit contributions
 supposed-1.927
illing-1.927
 stolen-1.817
 delayed-1.721
 eaten-1.716
Token She
Token ID1375
Feature activation-2.69e+00

Pos logit contributions
 tweet+3.615
 bou+3.428
 anytime+3.361
 galaxies+3.342
 forests+3.340
Neg logit contributions
illing-2.771
 stolen-2.749
 Tommy-2.721
 supposed-2.649
omo-2.616
Token would
Token ID561
Feature activation-2.74e+00

Pos logit contributions
 bou+5.628
 tweet+5.529
 anytime+5.383
 forests+5.074
 cities+4.971
Neg logit contributions
 stolen-3.162
 thrown-3.033
 sold-2.988
 fought-2.977
 delayed-2.923
Token fly
Token ID6129
Feature activation-4.04e+00

Pos logit contributions
 anytime+4.247
 bou+4.082
 forests+3.994
 galaxies+3.932
 tweet+3.893
Neg logit contributions
 stolen-4.331
 supposed-4.185
 disob-4.131
 thrown-4.101
 delayed-3.977
Token up
Token ID510
Feature activation-2.93e+00

Pos logit contributions
 Majesty+7.862
 tweet+7.655
 bou+7.495
ig+7.273
 galaxies+7.127
Neg logit contributions
 stolen-5.329
 thrown-5.118
 eaten-5.116
 supposed-4.892
 ruined-4.821
Token and
Token ID290
Feature activation-4.71e+00

Pos logit contributions
 tweet+4.485
 bou+4.204
 anytime+4.085
 Majesty+3.968
 galaxies+3.956
Neg logit contributions
 stolen-4.886
 thrown-4.702
 delayed-4.701
 supposed-4.642
 eaten-4.601
Token up
Token ID510
Feature activation-2.74e+00

Pos logit contributions
 bou+9.530
 anytime+8.505
 galaxies+8.054
ig+7.976
Summer+7.974
Neg logit contributions
 stolen-6.675
 fought-6.666
 stole-6.462
 hurt-6.374
 driven-6.364
Token,
Token ID11
Feature activation-3.91e+00

Pos logit contributions
 tweet+4.201
 bou+4.050
 galaxies+3.851
 anytime+3.778
 forests+3.714
Neg logit contributions
 stolen-4.363
 supposed-4.261
 delayed-4.247
 thrown-4.219
 eaten-4.117
Token then
Token ID788
Feature activation-3.30e+00

Pos logit contributions
 bou+6.230
 anytime+6.024
ig+5.988
 galaxies+5.909
 tweet+5.799
Neg logit contributions
 stolen-6.088
 supposed-5.922
 selling-5.736
 thrown-5.735
 delayed-5.723
Token come
Token ID1282
Feature activation-3.48e+00

Pos logit contributions
 bou+5.227
 anytime+4.895
 galaxies+4.678
 tweet+4.655
 forests+4.558
Neg logit contributions
 stolen-5.170
 delayed-4.963
 disob-4.935
 supposed-4.876
 fought-4.844
Token back
Token ID736
Feature activation-2.99e+00

Pos logit contributions
 bou+6.247
 galaxies+5.845
 sings+5.798
 tweet+5.726
 forests+5.595
Neg logit contributions
 supposed-5.243
 stolen-4.911
 driven-4.677
 delayed-4.630
illing-4.459
Token flew
Token ID13112
Feature activation-3.84e+00

Pos logit contributions
 bou+6.804
 tweet+6.668
 anytime+6.209
 forests+5.742
 cities+5.738
Neg logit contributions
 stolen-4.042
 delayed-3.914
 sold-3.746
 burned-3.725
 thrown-3.665
Token over
Token ID625
Feature activation-3.28e+00

Pos logit contributions
 tweet+7.038
 sings+6.855
 Majesty+6.571
 bou+6.510
 galaxies+6.284
Neg logit contributions
 stolen-5.504
 delayed-5.185
 thrown-5.178
 Tommy-5.121
 supposed-5.039
Token hills
Token ID18639
Feature activation-2.73e+00

Pos logit contributions
 tweet+5.154
 bou+4.825
 Majesty+4.696
 anytime+4.669
 sings+4.631
Neg logit contributions
 stolen-5.396
 thrown-4.940
 delayed-4.924
 sale-4.892
 burned-4.871
Token,
Token ID11
Feature activation-4.51e+00

Pos logit contributions
 tweet+4.176
 bou+3.862
 sings+3.825
 anytime+3.651
 galaxies+3.572
Neg logit contributions
 stolen-4.385
illing-4.290
 sale-4.246
 supposed-4.189
 thrown-4.179
Token mountains
Token ID12269
Feature activation-2.36e+00

Pos logit contributions
 bou+9.032
 tweet+8.843
 anytime+8.705
 Majesty+8.431
 galaxies+7.931
Neg logit contributions
 stolen-6.601
 thrown-6.421
 delayed-6.333
 driven-6.313
 supposed-6.100
Token,
Token ID11
Feature activation-4.59e+00

Pos logit contributions
 tweet+3.576
 bou+3.418
 anytime+3.308
 sings+3.264
 galaxies+3.149
Neg logit contributions
 stolen-3.769
 supposed-3.640
illing-3.630
 sale-3.569
 delayed-3.556
Token rivers
Token ID18180
Feature activation-1.61e+00

Pos logit contributions
 bou+8.154
 tweet+7.917
 anytime+7.738
 Majesty+7.407
 Sparkle+6.961
Neg logit contributions
 stolen-7.692
 driven-7.644
 delayed-7.569
 thrown-7.555
 supposed-7.311
Token and
Token ID290
Feature activation-3.51e+00

Pos logit contributions
 tweet+2.022
 anytime+1.921
 bou+1.906
 sings+1.753
 galaxies+1.751
Neg logit contributions
 stolen-2.946
 supposed-2.881
illing-2.866
 delayed-2.778
 thrown-2.771
Token trees
Token ID7150
Feature activation-1.63e+00

Pos logit contributions
 bou+4.394
 anytime+4.296
 tweet+4.237
ig+4.031
 nods+3.663
Neg logit contributions
 stolen-7.073
 delayed-6.883
 thrown-6.778
 driven-6.774
 been-6.707
Token.
Token ID13
Feature activation-3.22e+00

Pos logit contributions
 tweet+3.166
 anytime+3.063
 bou+3.050
 galaxies+2.922
 sings+2.870
Neg logit contributions
 stolen-1.876
 supposed-1.820
illing-1.752
 thrown-1.725
 delayed-1.701
Token The
Token ID383
Feature activation-2.35e+00

Pos logit contributions
 bou+6.721
 anytime+6.324
 tweet+6.264
cast+5.899
 higher+5.603
Neg logit contributions
illing-3.639
 supposed-3.558
 delayed-3.182
 driven-3.106
 Their-3.084
Token Every
Token ID3887
Feature activation-7.73e-01

Pos logit contributions
 bou+6.474
 tweet+6.239
 anytime+6.051
cast+5.586
 higher+5.525
Neg logit contributions
illing-3.195
 Tommy-3.078
omo-3.063
 Their-2.967
 supposed-2.935
Token day
Token ID1110
Feature activation-1.69e+00

Pos logit contributions
 tweet+1.222
 anytime+1.196
 bou+1.180
 forests+1.140
 galaxies+1.108
Neg logit contributions
 stolen-1.160
illing-1.137
 supposed-1.135
 delayed-1.092
 thrown-1.065
Token,
Token ID11
Feature activation-3.03e+00

Pos logit contributions
 bou+2.533
 tweet+2.472
 anytime+2.357
 forests+2.235
 galaxies+2.186
Neg logit contributions
 stolen-2.644
illing-2.625
 delayed-2.613
 supposed-2.595
 Tommy-2.544
Token she
Token ID673
Feature activation-1.96e+00

Pos logit contributions
 bou+4.899
 tweet+4.536
 anytime+4.398
cast+4.092
 Majesty+4.037
Neg logit contributions
 delayed-4.528
 Tommy-4.450
 supposed-4.441
illing-4.431
 disturbed-4.317
Token would
Token ID561
Feature activation-1.68e+00

Pos logit contributions
 tweet+3.021
 bou+2.944
 anytime+2.922
 forests+2.743
 galaxies+2.684
Neg logit contributions
 stolen-3.196
 delayed-3.124
 burned-3.038
 sold-3.013
 thrown-2.993
Token soar
Token ID48701
Feature activation-4.53e+00

Pos logit contributions
 anytime+2.697
 tweet+2.615
 forests+2.590
 bou+2.587
 galaxies+2.536
Neg logit contributions
 stolen-2.483
 supposed-2.380
 disob-2.352
illing-2.332
 delayed-2.311
Token up
Token ID510
Feature activation-2.83e+00

Pos logit contributions
 Majesty+9.657
 tweet+9.282
 bou+8.731
 sings+8.650
 nods+8.380
Neg logit contributions
 badly-5.273
 ruined-5.259
 thrown-5.202
 stolen-5.087
 delayed-5.046
Token high
Token ID1029
Feature activation-2.76e+00

Pos logit contributions
 tweet+4.674
 bou+4.336
 anytime+4.255
 Majesty+4.203
 sings+3.978
Neg logit contributions
 stolen-4.334
 supposed-4.279
 delayed-4.274
 burned-4.245
 thrown-4.216
Token into
Token ID656
Feature activation-8.51e-01

Pos logit contributions
 tweet+4.951
 bou+4.468
 anytime+4.191
 sings+4.085
 Majesty+3.903
Neg logit contributions
 stolen-4.045
 delayed-4.042
illing-4.006
 burned-3.949
 supposed-3.875
Token the
Token ID262
Feature activation-1.89e+00

Pos logit contributions
 tweet+1.162
 anytime+1.145
 bou+1.082
 forests+1.024
 galaxies+0.997
Neg logit contributions
 stolen-1.472
 supposed-1.420
illing-1.393
 thrown-1.378
 delayed-1.369
Token sky
Token ID6766
Feature activation-2.69e+00

Pos logit contributions
 anytime+2.590
 tweet+2.574
 bou+2.415
 galaxies+2.098
 forests+2.088
Neg logit contributions
 stolen-3.329
 delayed-3.219
illing-3.197
 ruined-3.187
 burned-3.164
Token the
Token ID262
Feature activation-5.05e-01

Pos logit contributions
 tweet+0.118
 anytime+0.116
 bou+0.112
 forests+0.110
 galaxies+0.105
Neg logit contributions
 stolen-0.127
 supposed-0.123
illing-0.122
 delayed-0.118
 robbed-0.116
Token butterfly
Token ID35113
Feature activation-7.76e-01

Pos logit contributions
 tweet+0.689
 anytime+0.681
 bou+0.647
 forests+0.636
 galaxies+0.610
Neg logit contributions
 stolen-0.877
 supposed-0.845
illing-0.837
 delayed-0.820
 unfair-0.808
Token.
Token ID13
Feature activation-1.22e+00

Pos logit contributions
 tweet+1.353
 anytime+1.325
 bou+1.312
 forests+1.281
 galaxies+1.243
Neg logit contributions
 stolen-1.028
 supposed-0.993
illing-0.970
 delayed-0.950
 thrown-0.936
Token She
Token ID1375
Feature activation-2.92e+00

Pos logit contributions
 tweet+2.047
 bou+1.986
 anytime+1.957
 forests+1.902
 galaxies+1.869
Neg logit contributions
illing-1.670
 stolen-1.654
 supposed-1.654
 Timothy-1.569
 delayed-1.561
Token flew
Token ID13112
Feature activation-3.45e+00

Pos logit contributions
 bou+5.850
 tweet+5.761
 anytime+5.521
 forests+5.306
 cities+5.179
Neg logit contributions
 stolen-3.579
 thrown-3.565
 delayed-3.492
 fought-3.447
 burned-3.373
Token around
Token ID1088
Feature activation-4.53e+00

Pos logit contributions
 tweet+5.584
 sings+5.571
 galaxies+5.391
 Majesty+5.260
 bou+5.240
Neg logit contributions
 stolen-5.194
 Tommy-5.038
 thrown-5.004
 Timothy-4.995
 supposed-4.949
Token in
Token ID287
Feature activation-2.54e+00

Pos logit contributions
 bou+8.320
 galaxies+8.267
 Majesty+8.201
 sings+8.005
 tweet+7.959
Neg logit contributions
 sale-5.616
 stolen-5.512
 eaten-5.483
 thrown-5.465
 been-5.297
Token circles
Token ID13332
Feature activation-2.35e+00

Pos logit contributions
 tweet+3.713
 anytime+3.572
 bou+3.505
cast+3.441
 sings+3.391
Neg logit contributions
 stolen-3.999
 supposed-3.863
 thrown-3.859
 Tommy-3.791
 sale-3.781
Token and
Token ID290
Feature activation-3.74e+00

Pos logit contributions
 tweet+3.073
 anytime+2.996
 bou+2.850
 forests+2.705
 sings+2.696
Neg logit contributions
illing-4.027
 stolen-4.021
 supposed-3.946
 thrown-3.929
 delayed-3.838
Token then
Token ID788
Feature activation-3.08e+00

Pos logit contributions
 bou+7.127
 anytime+6.997
 tweet+6.794
ig+6.426
 Him+6.414
Neg logit contributions
 stolen-4.894
 thrown-4.867
 fought-4.824
 supposed-4.640
 delayed-4.596
Token the
Token ID262
Feature activation-8.75e-01

Pos logit contributions
 bou+5.402
 anytime+5.319
 tweet+5.187
 galaxies+4.974
 forests+4.884
Neg logit contributions
 stolen-4.309
 delayed-4.130
 thrown-4.053
 fought-3.947
 eaten-3.927
Token pond
Token ID16723
Feature activation-9.35e-01

Pos logit contributions
 tweet+0.221
 anytime+0.215
 bou+0.207
 forests+0.201
 galaxies+0.191
Neg logit contributions
 stolen-0.328
 supposed-0.319
illing-0.317
 delayed-0.309
 robbed-0.305
Token.
Token ID13
Feature activation-1.99e+00

Pos logit contributions
 tweet+1.664
 anytime+1.616
 bou+1.573
 forests+1.546
 galaxies+1.520
Neg logit contributions
 stolen-1.216
illing-1.164
 supposed-1.163
 thrown-1.111
 delayed-1.107
Token She
Token ID1375
Feature activation-3.31e+00

Pos logit contributions
 tweet+3.200
 bou+3.166
 anytime+3.112
 forests+3.024
 galaxies+3.013
Neg logit contributions
illing-2.858
 supposed-2.800
 Timothy-2.774
 stolen-2.718
 Tommy-2.667
Token flew
Token ID13112
Feature activation-2.85e+00

Pos logit contributions
 bou+6.811
 tweet+6.787
 anytime+6.499
 forests+6.397
 Him+6.366
Neg logit contributions
 thrown-4.015
 stolen-3.919
 delayed-3.757
 ruined-3.748
 fought-3.737
Token higher
Token ID2440
Feature activation-3.79e+00

Pos logit contributions
 tweet+4.043
 sings+3.946
 anytime+3.848
 galaxies+3.743
 bou+3.711
Neg logit contributions
 stolen-4.839
 Timothy-4.682
 Tommy-4.602
 thrown-4.593
 supposed-4.542
Token and
Token ID290
Feature activation-4.51e+00

Pos logit contributions
 tweet+6.135
 bou+5.535
 sings+5.468
 anytime+5.226
 galaxies+5.149
Neg logit contributions
 thrown-6.067
 stolen-5.956
 burned-5.875
 delayed-5.834
illing-5.814
Token higher
Token ID2440
Feature activation-2.95e+00

Pos logit contributions
 bou+6.969
 tweet+6.831
 Him+6.383
 Majesty+6.161
 anytime+5.982
Neg logit contributions
 delayed-8.584
 thrown-8.442
 sold-8.405
 stolen-8.375
 driven-8.196
Token,
Token ID11
Feature activation-4.14e+00

Pos logit contributions
 tweet+4.650
 bou+4.414
 anytime+4.267
 sings+4.194
 galaxies+4.147
Neg logit contributions
 stolen-4.571
 thrown-4.437
 supposed-4.376
 delayed-4.369
 burned-4.309
Token feeling
Token ID4203
Feature activation-1.69e+00

Pos logit contributions
 bou+7.309
 tweet+7.182
 galaxies+7.138
 anytime+6.803
 Him+6.660
Neg logit contributions
 stolen-6.348
 thrown-6.306
 supposed-6.301
 delayed-6.008
 driven-5.914
Token proud
Token ID6613
Feature activation-6.48e-01

Pos logit contributions
 tweet+2.705
 anytime+2.623
 bou+2.620
 forests+2.450
 galaxies+2.391
Neg logit contributions
 stolen-2.576
 supposed-2.535
 delayed-2.460
 thrown-2.447
illing-2.408
Token of
Token ID286
Feature activation-1.75e+00

Pos logit contributions
 tweet+1.253
 anytime+1.210
 bou+1.200
 forests+1.169
 galaxies+1.130
Neg logit contributions
 stolen-0.758
 supposed-0.731
illing-0.729
 delayed-0.691
 thrown-0.679
Token.
Token ID13
Feature activation-2.67e+00

Pos logit contributions
 tweet+0.531
 anytime+0.518
 bou+0.508
 forests+0.496
 galaxies+0.481
Neg logit contributions
 stolen-0.380
 supposed-0.365
illing-0.360
 delayed-0.348
 thrown-0.341
Token She
Token ID1375
Feature activation-3.16e+00

Pos logit contributions
 bou+5.648
 tweet+5.431
 anytime+5.302
cast+5.045
 higher+4.878
Neg logit contributions
illing-2.811
 supposed-2.762
 delayed-2.448
 stolen-2.390
 unfair-2.388
Token flew
Token ID13112
Feature activation-3.84e+00

Pos logit contributions
 bou+6.804
 tweet+6.668
 anytime+6.209
 forests+5.742
 cities+5.738
Neg logit contributions
 stolen-4.042
 delayed-3.914
 sold-3.746
 burned-3.725
 thrown-3.665
Token over
Token ID625
Feature activation-3.28e+00

Pos logit contributions
 tweet+7.038
 sings+6.855
 Majesty+6.571
 bou+6.510
 galaxies+6.284
Neg logit contributions
 stolen-5.504
 delayed-5.185
 thrown-5.178
 Tommy-5.121
 supposed-5.039
Token hills
Token ID18639
Feature activation-2.73e+00

Pos logit contributions
 tweet+5.154
 bou+4.825
 Majesty+4.696
 anytime+4.669
 sings+4.631
Neg logit contributions
 stolen-5.396
 thrown-4.940
 delayed-4.924
 sale-4.892
 burned-4.871
Token,
Token ID11
Feature activation-4.51e+00

Pos logit contributions
 tweet+4.176
 bou+3.862
 sings+3.825
 anytime+3.651
 galaxies+3.572
Neg logit contributions
 stolen-4.385
illing-4.290
 sale-4.246
 supposed-4.189
 thrown-4.179
Token mountains
Token ID12269
Feature activation-2.36e+00

Pos logit contributions
 bou+9.032
 tweet+8.843
 anytime+8.705
 Majesty+8.431
 galaxies+7.931
Neg logit contributions
 stolen-6.601
 thrown-6.421
 delayed-6.333
 driven-6.313
 supposed-6.100
Token,
Token ID11
Feature activation-4.59e+00

Pos logit contributions
 tweet+3.576
 bou+3.418
 anytime+3.308
 sings+3.264
 galaxies+3.149
Neg logit contributions
 stolen-3.769
 supposed-3.640
illing-3.630
 sale-3.569
 delayed-3.556
Token rivers
Token ID18180
Feature activation-1.61e+00

Pos logit contributions
 bou+8.154
 tweet+7.917
 anytime+7.738
 Majesty+7.407
 Sparkle+6.961
Neg logit contributions
 stolen-7.692
 driven-7.644
 delayed-7.569
 thrown-7.555
 supposed-7.311
Token and
Token ID290
Feature activation-3.51e+00

Pos logit contributions
 tweet+2.022
 anytime+1.921
 bou+1.906
 sings+1.753
 galaxies+1.751
Neg logit contributions
 stolen-2.946
 supposed-2.881
illing-2.866
 delayed-2.778
 thrown-2.771
Token trees
Token ID7150
Feature activation-1.63e+00

Pos logit contributions
 bou+4.394
 anytime+4.296
 tweet+4.237
ig+4.031
 nods+3.663
Neg logit contributions
 stolen-7.073
 delayed-6.883
 thrown-6.778
 driven-6.774
 been-6.707
Token soared
Token ID35292
Feature activation-3.82e+00

Pos logit contributions
 tweet+4.764
 bou+4.627
 anytime+4.465
 forests+4.226
 galaxies+4.175
Neg logit contributions
 stolen-3.019
 delayed-2.939
 thrown-2.839
 sold-2.789
 burned-2.788
Token high
Token ID1029
Feature activation-3.40e+00

Pos logit contributions
 tweet+6.256
 sings+6.017
 Majesty+5.788
 anytime+5.701
 bou+5.602
Neg logit contributions
 stolen-5.878
 delayed-5.690
 thrown-5.685
 Tommy-5.669
 Timothy-5.591
Token over
Token ID625
Feature activation-2.38e+00

Pos logit contributions
 tweet+6.278
 bou+5.605
 sings+5.464
 anytime+5.358
 nods+5.149
Neg logit contributions
 delayed-4.960
 stolen-4.935
 burned-4.816
 sold-4.665
 sale-4.653
Token the
Token ID262
Feature activation-2.65e+00

Pos logit contributions
 tweet+3.709
 bou+3.406
 anytime+3.393
 sings+3.236
 Majesty+3.205
Neg logit contributions
 stolen-3.730
 burned-3.591
 supposed-3.546
 thrown-3.536
 delayed-3.524
Token trees
Token ID7150
Feature activation-2.51e+00

Pos logit contributions
 tweet+3.920
 anytime+3.863
 bou+3.670
 nods+3.164
 galaxies+3.145
Neg logit contributions
 stolen-4.457
 burned-4.413
 ruined-4.375
 delayed-4.373
 wrecked-4.236
Token,
Token ID11
Feature activation-4.48e+00

Pos logit contributions
 tweet+3.379
 bou+3.201
 anytime+3.132
 galaxies+3.129
 sings+3.029
Neg logit contributions
illing-4.295
 supposed-4.289
 stolen-4.288
 thrown-4.140
 delayed-4.108
Token buildings
Token ID6832
Feature activation-1.87e+00

Pos logit contributions
 tweet+8.747
 anytime+8.688
 bou+8.525
 galaxies+8.332
 Majesty+8.316
Neg logit contributions
 stolen-6.355
 thrown-6.221
 delayed-6.157
 burned-6.104
 eaten-6.045
Token and
Token ID290
Feature activation-3.00e+00

Pos logit contributions
 tweet+2.406
 anytime+2.266
 bou+2.255
 galaxies+2.115
 sings+2.057
Neg logit contributions
 stolen-3.413
 supposed-3.326
illing-3.256
 thrown-3.227
 delayed-3.207
Token people
Token ID661
Feature activation-6.62e-01

Pos logit contributions
 anytime+4.338
 tweet+4.305
 bou+4.237
ig+3.996
 Majesty+3.664
Neg logit contributions
 stolen-5.223
 thrown-5.058
 delayed-5.041
 burned-4.918
 supposed-4.907
Token.
Token ID13
Feature activation-6.08e-01

Pos logit contributions
 tweet+0.952
 anytime+0.931
 bou+0.902
 forests+0.873
 galaxies+0.842
Neg logit contributions
 stolen-1.100
 supposed-1.070
illing-1.053
 delayed-1.025
 thrown-1.018
Token Everything
Token ID11391
Feature activation-2.19e+00

Pos logit contributions
 tweet+1.058
 anytime+1.029
 bou+1.010
 forests+0.981
 galaxies+0.947
Neg logit contributions
 stolen-0.823
 supposed-0.807
illing-0.791
 delayed-0.763
 unfair-0.748
Token!
Token ID0
Feature activation-5.91e-01

Pos logit contributions
 tweet+2.181
 anytime+2.108
 bou+2.078
 forests+2.008
 galaxies+1.956
Neg logit contributions
 stolen-1.596
 supposed-1.559
illing-1.535
 delayed-1.495
 thrown-1.458
Token She
Token ID1375
Feature activation-2.84e+00

Pos logit contributions
 tweet+0.896
 anytime+0.866
 bou+0.856
 forests+0.825
 galaxies+0.795
Neg logit contributions
 stolen-0.924
 supposed-0.904
illing-0.902
 delayed-0.867
 thrown-0.855
Token flew
Token ID13112
Feature activation-2.98e+00

Pos logit contributions
 bou+5.758
 tweet+5.705
 anytime+5.533
 forests+5.141
 cities+4.984
Neg logit contributions
 stolen-3.398
 thrown-3.286
 delayed-3.282
 ruined-3.212
 burned-3.178
Token ever
Token ID1683
Feature activation-2.59e+00

Pos logit contributions
 tweet+4.417
 sings+4.253
 anytime+4.211
 bou+4.021
 galaxies+3.880
Neg logit contributions
 stolen-4.992
 delayed-4.717
 Timothy-4.679
 Tommy-4.671
 supposed-4.659
Token higher
Token ID2440
Feature activation-3.58e+00

Pos logit contributions
 tweet+4.589
 bou+4.539
 anytime+4.364
 forests+4.126
 sings+4.040
Neg logit contributions
 supposed-3.719
 stolen-3.711
 delayed-3.586
 forgotten-3.330
 thrown-3.289
Token and
Token ID290
Feature activation-4.46e+00

Pos logit contributions
 tweet+5.762
 bou+5.145
 sings+4.906
 anytime+4.840
 Majesty+4.495
Neg logit contributions
 stolen-5.940
 supposed-5.781
 thrown-5.769
 delayed-5.755
 burned-5.652
Token higher
Token ID2440
Feature activation-2.95e+00

Pos logit contributions
 bou+7.913
 tweet+7.543
 anytime+6.539
ig+6.485
 Him+6.294
Neg logit contributions
 delayed-7.940
 stolen-7.602
 sold-7.472
 forgotten-7.340
 thrown-7.332
Token until
Token ID1566
Feature activation-1.55e+00

Pos logit contributions
 tweet+5.001
 bou+4.686
 anytime+4.525
 sings+4.412
 galaxies+4.178
Neg logit contributions
 stolen-4.331
 supposed-4.249
 delayed-4.137
 thrown-4.126
 burned-4.034
Token she
Token ID673
Feature activation-1.63e+00

Pos logit contributions
 tweet+2.290
 bou+2.190
 anytime+2.156
 forests+2.011
 sings+1.992
Neg logit contributions
 stolen-2.569
 delayed-2.451
 supposed-2.400
 thrown-2.393
 Timothy-2.366
Token reached
Token ID4251
Feature activation-2.43e+00

Pos logit contributions
 tweet+3.059
 bou+3.001
 anytime+2.882
 forests+2.808
 galaxies+2.718
Neg logit contributions
 stolen-2.109
 delayed-2.011
 thrown-1.964
 supposed-1.955
 ruined-1.911
Token the
Token ID262
Feature activation-1.48e+00

Pos logit contributions
 tweet+4.572
 bou+4.302
 anytime+4.212
 sings+3.884
 nods+3.836
Neg logit contributions
 stolen-3.181
 supposed-3.144
 thrown-2.930
 delayed-2.892
 Tommy-2.891
Token would
Token ID561
Feature activation-1.20e+00

Pos logit contributions
 tweet+3.046
 bou+2.997
 anytime+2.890
 forests+2.774
 valleys+2.665
Neg logit contributions
 stolen-2.894
 delayed-2.783
 supposed-2.749
 sold-2.732
 burned-2.709
Token go
Token ID467
Feature activation-2.94e+00

Pos logit contributions
 anytime+1.675
 tweet+1.653
 bou+1.610
 forests+1.582
 galaxies+1.514
Neg logit contributions
 stolen-2.037
 supposed-1.955
illing-1.907
 delayed-1.892
 disob-1.889
Token on
Token ID319
Feature activation-1.47e+00

Pos logit contributions
 tweet+5.960
 bou+5.735
 galaxies+5.608
 nods+5.589
 Majesty+5.463
Neg logit contributions
 stolen-3.699
 supposed-3.542
 thrown-3.221
 delayed-3.172
 driven-3.151
Token adventures
Token ID17545
Feature activation-1.75e+00

Pos logit contributions
 tweet+2.804
 anytime+2.628
 bou+2.596
 forests+2.427
 galaxies+2.408
Neg logit contributions
 stolen-1.908
 supposed-1.814
 thrown-1.713
 delayed-1.696
 sale-1.691
Token,
Token ID11
Feature activation-2.80e+00

Pos logit contributions
 tweet+3.075
 bou+2.869
 anytime+2.815
 forests+2.685
 galaxies+2.636
Neg logit contributions
 stolen-2.465
 supposed-2.411
illing-2.334
 thrown-2.313
 sale-2.310
Token exploring
Token ID13504
Feature activation-4.46e+00

Pos logit contributions
 bou+4.739
 tweet+4.642
 anytime+4.626
 galaxies+4.284
ig+4.278
Neg logit contributions
 stolen-4.094
 supposed-3.942
 thrown-3.799
 delayed-3.733
 sold-3.658
Token the
Token ID262
Feature activation-3.86e+00

Pos logit contributions
 tweet+10.228
 Majesty+9.748
 sings+9.125
ig+9.090
als+9.013
Neg logit contributions
 sale-4.816
 stolen-4.782
 selling-4.507
 supposed-4.491
 taken-4.389
Token world
Token ID995
Feature activation-2.43e+00

Pos logit contributions
 tweet+6.838
 anytime+6.172
 bou+6.121
 nods+5.429
 wow+5.245
Neg logit contributions
 stolen-6.700
 supposed-6.165
 rude-6.153
 delayed-6.077
 garbage-5.969
Token around
Token ID1088
Feature activation-2.73e+00

Pos logit contributions
 tweet+4.877
 bou+4.611
 anytime+4.353
 Majesty+4.316
 galaxies+4.210
Neg logit contributions
 supposed-2.868
illing-2.834
 stolen-2.833
 sold-2.720
 sale-2.633
Token her
Token ID607
Feature activation-3.31e+00

Pos logit contributions
 tweet+5.063
 bou+4.844
 anytime+4.541
 nods+4.455
 Majesty+4.432
Neg logit contributions
 stolen-3.444
 supposed-3.420
 sale-3.311
 Tommy-3.249
 disob-3.216
Token.
Token ID13
Feature activation-1.47e+00

Pos logit contributions
 tweet+7.889
 bou+7.358
 anytime+7.057
 sings+6.937
 galaxies+6.774
Neg logit contributions
 stolen-2.762
 supposed-2.667
 delayed-2.406
 burned-2.392
illing-2.364
Token safe
Token ID3338
Feature activation+1.79e-02

Pos logit contributions
 stolen+0.247
illing+0.240
 supposed+0.238
 delayed+0.229
 robbed+0.227
Neg logit contributions
 tweet-0.250
 anytime-0.245
 bou-0.238
 forests-0.232
 galaxies-0.221
Token.
Token ID13
Feature activation-2.36e-01

Pos logit contributions
 stolen+0.029
 supposed+0.028
illing+0.028
 delayed+0.027
 robbed+0.027
Neg logit contributions
 tweet-0.026
 anytime-0.026
 bou-0.025
 forests-0.025
 galaxies-0.023
Token She
Token ID1375
Feature activation-2.01e+00

Pos logit contributions
 tweet+0.370
 anytime+0.361
 bou+0.352
 forests+0.343
 galaxies+0.329
Neg logit contributions
 stolen-0.365
 supposed-0.353
illing-0.351
 delayed-0.340
 robbed-0.334
Token soared
Token ID35292
Feature activation-3.56e+00

Pos logit contributions
 tweet+3.718
 bou+3.671
 anytime+3.616
 forests+3.541
 galaxies+3.418
Neg logit contributions
 stolen-2.681
 thrown-2.531
 delayed-2.499
 ruined-2.443
 burned-2.438
Token higher
Token ID2440
Feature activation-3.82e+00

Pos logit contributions
 tweet+5.656
 sings+5.499
 bou+5.170
 Majesty+5.021
 anytime+4.960
Neg logit contributions
 stolen-5.777
 delayed-5.605
 thrown-5.587
 Tommy-5.488
 Timothy-5.486
Token and
Token ID290
Feature activation-4.44e+00

Pos logit contributions
 tweet+6.735
 bou+5.972
 sings+5.715
 Majesty+5.581
 anytime+5.370
Neg logit contributions
 thrown-5.992
 burned-5.962
 delayed-5.933
illing-5.919
 stolen-5.859
Token higher
Token ID2440
Feature activation-3.12e+00

Pos logit contributions
 bou+6.278
 tweet+6.006
 Majesty+5.388
 galaxies+5.164
 Him+5.097
Neg logit contributions
 delayed-9.254
 stolen-8.964
 thrown-8.954
 sold-8.940
 driven-8.830
Token until
Token ID1566
Feature activation-1.49e+00

Pos logit contributions
 tweet+5.547
 bou+5.317
 sings+4.947
 anytime+4.912
 galaxies+4.812
Neg logit contributions
 supposed-4.306
 stolen-4.295
 delayed-4.250
 burned-4.219
 thrown-4.198
Token she
Token ID673
Feature activation-1.28e+00

Pos logit contributions
 tweet+2.153
 bou+2.082
 anytime+2.024
 galaxies+1.904
 forests+1.897
Neg logit contributions
 stolen-2.514
 delayed-2.420
 thrown-2.379
 supposed-2.368
 burned-2.367
Token could
Token ID714
Feature activation-2.65e+00

Pos logit contributions
 tweet+2.354
 bou+2.302
 anytime+2.241
 forests+2.196
 galaxies+2.143
Neg logit contributions
 stolen-1.669
 delayed-1.558
 supposed-1.557
 thrown-1.549
 ruined-1.495
Token touch
Token ID3638
Feature activation-1.71e+00

Pos logit contributions
 bou+5.227
 forests+5.072
 galaxies+5.030
 anytime+5.003
 tweet+4.982
Neg logit contributions
 supposed-3.219
 stolen-3.150
 thrown-3.025
 delayed-3.019
 disob-2.949
Token Sally
Token ID25737
Feature activation-1.61e+00

Pos logit contributions
 tweet+0.245
 anytime+0.241
 bou+0.235
 forests+0.230
 galaxies+0.223
Neg logit contributions
 stolen-0.168
 supposed-0.161
illing-0.160
 delayed-0.154
 robbed-0.151
Token watched
Token ID7342
Feature activation-1.92e+00

Pos logit contributions
 tweet+2.809
 bou+2.743
 anytime+2.734
 forests+2.655
 galaxies+2.607
Neg logit contributions
 stolen-2.208
 delayed-2.087
 supposed-2.056
 thrown-2.049
illing-2.036
Token it
Token ID340
Feature activation-2.88e+00

Pos logit contributions
 tweet+2.725
 anytime+2.708
 forests+2.686
 galaxies+2.672
 bou+2.643
Neg logit contributions
 stolen-2.981
illing-2.886
 supposed-2.875
 Tommy-2.855
 thrown-2.841
Token fly
Token ID6129
Feature activation-3.17e+00

Pos logit contributions
 forests+4.059
 anytime+4.015
 galaxies+3.999
 valleys+3.924
 bou+3.893
Neg logit contributions
 supposed-4.488
illing-4.466
 stolen-4.380
 thrown-4.371
 burned-4.263
Token higher
Token ID2440
Feature activation-3.09e+00

Pos logit contributions
 anytime+4.356
 tweet+4.353
 galaxies+4.302
 Majesty+4.257
 forests+4.236
Neg logit contributions
 stolen-5.257
 thrown-5.059
 eaten-4.899
 supposed-4.881
 burned-4.872
Token and
Token ID290
Feature activation-4.40e+00

Pos logit contributions
 tweet+4.898
 sings+4.298
 bou+4.251
 anytime+4.251
 galaxies+4.088
Neg logit contributions
 stolen-4.889
illing-4.871
 burned-4.818
 thrown-4.785
 supposed-4.754
Token higher
Token ID2440
Feature activation-2.53e+00

Pos logit contributions
 bou+6.287
 tweet+5.930
 anytime+5.765
ig+5.702
 galaxies+5.669
Neg logit contributions
 delayed-8.510
 stolen-8.413
 disturbed-8.136
 thrown-8.115
 burned-8.097
Token.
Token ID13
Feature activation-6.66e-01

Pos logit contributions
 tweet+4.925
 anytime+4.673
 bou+4.643
 galaxies+4.507
 sings+4.483
Neg logit contributions
 stolen-2.942
 supposed-2.815
 thrown-2.753
 burned-2.734
illing-2.723
Token Then
Token ID3244
Feature activation-1.13e+00

Pos logit contributions
 tweet+1.038
 anytime+1.009
 bou+0.981
 forests+0.962
 galaxies+0.941
Neg logit contributions
 stolen-1.020
illing-0.995
 supposed-0.995
 delayed-0.948
 thrown-0.942
Token,
Token ID11
Feature activation-1.41e+00

Pos logit contributions
 tweet+1.596
 anytime+1.558
 bou+1.500
 forests+1.483
 galaxies+1.471
Neg logit contributions
 stolen-1.912
 delayed-1.793
 supposed-1.784
 thrown-1.760
 Timothy-1.756
Token she
Token ID673
Feature activation-2.09e+00

Pos logit contributions
 tweet+2.233
 anytime+2.161
 forests+2.088
 bou+2.084
 galaxies+2.084
Neg logit contributions
 stolen-2.193
 delayed-2.044
 supposed-2.031
 thrown-2.013
 Timothy-2.012
Token same
Token ID976
Feature activation-1.24e+00

Pos logit contributions
 tweet+3.072
 anytime+2.941
 forests+2.721
 galaxies+2.638
 bou+2.636
Neg logit contributions
 stolen-3.998
 supposed-3.840
 delayed-3.775
 ruined-3.766
omo-3.721
Token cozy
Token ID37438
Feature activation-2.09e+00

Pos logit contributions
 tweet+1.723
 anytime+1.623
 forests+1.561
 bou+1.531
 galaxies+1.494
Neg logit contributions
 stolen-2.160
 supposed-2.071
illing-2.043
 delayed-2.014
omo-2.006
Token blanket
Token ID18447
Feature activation-4.38e-01

Pos logit contributions
 tweet+2.760
 anytime+2.439
 bou+2.347
 forests+2.322
 galaxies+2.274
Neg logit contributions
 stolen-3.798
omo-3.740
illing-3.671
 burned-3.631
 supposed-3.597
Token.
Token ID13
Feature activation-2.19e+00

Pos logit contributions
 tweet+0.725
 anytime+0.702
 bou+0.687
 forests+0.676
 galaxies+0.651
Neg logit contributions
 stolen-0.630
 supposed-0.610
illing-0.600
 delayed-0.581
 thrown-0.573
Token They
Token ID1119
Feature activation-8.81e-01

Pos logit contributions
 tweet+4.231
 bou+3.994
 forests+3.841
 anytime+3.834
 galaxies+3.727
Neg logit contributions
 supposed-2.665
illing-2.608
 Tommy-2.559
 stolen-2.484
 sold-2.414
Token sat
Token ID3332
Feature activation-4.40e+00

Pos logit contributions
 tweet+1.531
 anytime+1.503
 bou+1.453
 forests+1.452
 galaxies+1.396
Neg logit contributions
 stolen-1.212
 supposed-1.139
 delayed-1.120
illing-1.113
 thrown-1.103
Token in
Token ID287
Feature activation-2.23e+00

Pos logit contributions
 tweet+8.334
 galaxies+8.064
 bou+7.459
 sings+7.394
 Majesty+7.102
Neg logit contributions
 burned-6.836
 sold-6.742
 delayed-6.724
 Tommy-6.643
 stolen-6.606
Token the
Token ID262
Feature activation-1.53e+00

Pos logit contributions
 tweet+3.715
 anytime+3.417
 galaxies+3.381
 forests+3.324
 Majesty+3.217
Neg logit contributions
 stolen-3.371
 Tommy-3.204
 sale-3.146
 thrown-3.121
 ruined-3.112
Token sun
Token ID4252
Feature activation-2.38e+00

Pos logit contributions
 tweet+2.268
 anytime+2.142
 bou+2.020
 galaxies+1.951
 forests+1.925
Neg logit contributions
 stolen-2.581
 burned-2.418
 delayed-2.414
 ruined-2.404
omo-2.400
Token,
Token ID11
Feature activation-2.86e+00

Pos logit contributions
 tweet+3.178
 galaxies+2.873
 anytime+2.811
 bou+2.788
 forests+2.767
Neg logit contributions
 stolen-4.234
illing-4.179
 supposed-4.165
 delayed-4.149
 burned-4.087
Token chatting
Token ID32046
Feature activation-2.82e+00

Pos logit contributions
 tweet+5.694
 anytime+5.209
 bou+5.195
 galaxies+5.170
 forests+5.017
Neg logit contributions
 stolen-3.820
 thrown-3.589
 delayed-3.568
 sold-3.544
 eaten-3.518
Token free
Token ID1479
Feature activation-9.41e-01

Pos logit contributions
 tweet+1.452
 anytime+1.422
 bou+1.392
 forests+1.352
 galaxies+1.311
Neg logit contributions
 stolen-1.377
 supposed-1.324
 delayed-1.289
illing-1.284
 thrown-1.271
Token.
Token ID13
Feature activation-1.36e+00

Pos logit contributions
 tweet+1.600
 anytime+1.521
 bou+1.519
 forests+1.465
 galaxies+1.434
Neg logit contributions
 stolen-1.338
 supposed-1.313
illing-1.275
 delayed-1.248
 thrown-1.221
Token She
Token ID1375
Feature activation-2.91e+00

Pos logit contributions
 tweet+2.312
 anytime+2.297
 bou+2.254
 forests+2.152
 galaxies+2.133
Neg logit contributions
 stolen-1.812
illing-1.767
 supposed-1.749
 delayed-1.713
 unfair-1.678
Token flew
Token ID13112
Feature activation-3.07e+00

Pos logit contributions
 bou+5.991
 tweet+5.858
 anytime+5.734
 forests+5.354
 cities+5.342
Neg logit contributions
 stolen-3.573
 thrown-3.445
 delayed-3.416
 fought-3.400
 burned-3.394
Token higher
Token ID2440
Feature activation-3.97e+00

Pos logit contributions
 tweet+5.044
 sings+4.922
 galaxies+4.793
 Majesty+4.765
 bou+4.754
Neg logit contributions
 stolen-4.644
 delayed-4.453
 supposed-4.406
 thrown-4.384
 eaten-4.362
Token and
Token ID290
Feature activation-4.40e+00

Pos logit contributions
 tweet+7.094
 bou+6.390
 Majesty+6.066
 sings+6.047
 anytime+5.984
Neg logit contributions
 burned-6.002
 thrown-5.979
 delayed-5.831
 supposed-5.806
illing-5.787
Token higher
Token ID2440
Feature activation-3.21e+00

Pos logit contributions
 bou+6.392
 tweet+6.208
 Him+5.607
 galaxies+5.596
 Majesty+5.579
Neg logit contributions
 delayed-8.835
 stolen-8.701
 thrown-8.598
 sold-8.481
 burned-8.271
Token into
Token ID656
Feature activation-1.31e+00

Pos logit contributions
 tweet+7.117
 bou+6.855
 anytime+6.615
 galaxies+6.514
 sings+6.510
Neg logit contributions
 supposed-3.012
 stolen-2.923
 burned-2.898
 thrown-2.883
 delayed-2.836
Token the
Token ID262
Feature activation-1.52e+00

Pos logit contributions
 tweet+1.915
 anytime+1.866
 bou+1.787
 galaxies+1.656
 sings+1.649
Neg logit contributions
 stolen-2.148
 supposed-2.052
 thrown-2.048
 burned-1.982
 delayed-1.978
Token sky
Token ID6766
Feature activation-1.79e+00

Pos logit contributions
 anytime+1.987
 tweet+1.968
 bou+1.871
 forests+1.665
 galaxies+1.654
Neg logit contributions
 stolen-2.760
 delayed-2.646
illing-2.626
 ruined-2.594
 burned-2.594
Token.
Token ID13
Feature activation-1.96e+00

Pos logit contributions
 tweet+3.698
 anytime+3.591
 bou+3.573
 forests+3.433
 galaxies+3.426
Neg logit contributions
 stolen-1.786
 supposed-1.734
illing-1.701
 delayed-1.670
 thrown-1.654
Token day
Token ID1110
Feature activation-7.45e-01

Pos logit contributions
 bou+5.026
 galaxies+4.853
 tweet+4.849
 forests+4.813
 valleys+4.789
Neg logit contributions
 supposed-3.072
 stolen-3.000
illing-2.820
 delayed-2.738
 eaten-2.704
Token on
Token ID319
Feature activation-1.11e+00

Pos logit contributions
 tweet+0.971
 bou+0.956
 anytime+0.956
 forests+0.915
 galaxies+0.895
Neg logit contributions
 stolen-1.323
 supposed-1.303
illing-1.278
 delayed-1.245
 thrown-1.233
Token,
Token ID11
Feature activation-1.93e+00

Pos logit contributions
 tweet+1.548
 bou+1.487
 anytime+1.484
 forests+1.446
 galaxies+1.440
Neg logit contributions
 stolen-1.911
 supposed-1.819
 Timothy-1.771
 delayed-1.769
illing-1.767
Token Dove
Token ID48595
Feature activation-2.18e+00

Pos logit contributions
 tweet+2.907
 bou+2.855
 galaxies+2.841
 anytime+2.817
 forests+2.796
Neg logit contributions
 stolen-3.036
 supposed-2.808
 thrown-2.778
 delayed-2.776
 Tommy-2.765
Token practiced
Token ID19893
Feature activation-3.67e+00

Pos logit contributions
 bou+4.073
 tweet+4.006
 anytime+3.856
 forests+3.774
 cities+3.675
Neg logit contributions
 stolen-2.820
 supposed-2.780
 delayed-2.720
 eaten-2.649
 ruined-2.647
Token flying
Token ID7348
Feature activation-4.38e+00

Pos logit contributions
 bou+6.246
als+6.096
 tweet+5.967
 anytime+5.807
 galaxies+5.668
Neg logit contributions
 supposed-5.765
 thrown-5.380
 sale-5.233
 eaten-5.193
 stolen-5.180
Token every
Token ID790
Feature activation-1.18e+00

Pos logit contributions
 Majesty+8.617
 bou+8.277
 tweet+8.160
 sings+7.912
 galaxies+7.689
Neg logit contributions
 thrown-5.449
 sale-5.194
 supposed-5.172
 badly-5.119
 eaten-5.070
Token day
Token ID1110
Feature activation-1.59e+00

Pos logit contributions
 tweet+1.927
 anytime+1.900
 bou+1.888
 forests+1.813
 galaxies+1.785
Neg logit contributions
 stolen-1.719
 supposed-1.663
illing-1.640
 delayed-1.611
 thrown-1.582
Token and
Token ID290
Feature activation-3.73e+00

Pos logit contributions
 tweet+2.876
 bou+2.779
 anytime+2.689
 forests+2.609
 galaxies+2.592
Neg logit contributions
 supposed-2.094
illing-2.054
 stolen-2.045
 thrown-1.917
 delayed-1.895
Token soon
Token ID2582
Feature activation-1.32e+00

Pos logit contributions
 bou+8.491
 tweet+7.653
 anytime+7.557
 galaxies+7.324
 forests+7.277
Neg logit contributions
 supposed-4.345
 stolen-4.142
 thrown-4.094
 fought-4.042
 eaten-3.971
Token became
Token ID2627
Feature activation-3.42e-01

Pos logit contributions
 tweet+1.929
 bou+1.860
 anytime+1.843
 forests+1.738
 galaxies+1.712
Neg logit contributions
 stolen-2.165
 supposed-2.090
 delayed-2.043
 thrown-2.041
 eaten-1.979
Token sky
Token ID6766
Feature activation-9.45e-01

Pos logit contributions
 anytime+1.361
 tweet+1.342
 bou+1.273
 forests+1.159
 galaxies+1.150
Neg logit contributions
 stolen-2.174
 supposed-2.097
illing-2.082
 delayed-2.062
 thrown-2.039
Token.
Token ID13
Feature activation-1.09e+00

Pos logit contributions
 tweet+1.859
 bou+1.793
 anytime+1.770
 forests+1.716
 galaxies+1.682
Neg logit contributions
 stolen-1.068
 supposed-1.050
illing-1.050
 delayed-0.989
 thrown-0.968
Token He
Token ID679
Feature activation-2.45e+00

Pos logit contributions
 tweet+1.944
 bou+1.904
 anytime+1.891
 forests+1.829
 galaxies+1.800
Neg logit contributions
 stolen-1.427
 supposed-1.351
illing-1.342
 delayed-1.313
 sold-1.298
Token looked
Token ID3114
Feature activation-3.63e+00

Pos logit contributions
 tweet+4.792
 bou+4.748
 anytime+4.710
 forests+4.492
 cities+4.462
Neg logit contributions
 stolen-3.109
 delayed-2.961
 burned-2.916
 thrown-2.905
 sold-2.852
Token up
Token ID510
Feature activation-1.84e+00

Pos logit contributions
 tweet+6.467
 anytime+6.200
 forests+5.902
 bou+5.870
 galaxies+5.819
Neg logit contributions
 stolen-5.888
 supposed-5.671
 delayed-5.326
 ruined-5.209
 unfair-5.189
Token and
Token ID290
Feature activation-4.38e+00

Pos logit contributions
 tweet+3.110
 bou+2.927
 anytime+2.834
 forests+2.715
 galaxies+2.692
Neg logit contributions
 stolen-2.774
illing-2.702
 supposed-2.656
 delayed-2.647
 burned-2.552
Token saw
Token ID2497
Feature activation-8.03e-01

Pos logit contributions
 bou+9.661
 tweet+9.561
 anytime+9.456
ig+8.926
 galaxies+8.879
Neg logit contributions
 stolen-5.374
 delayed-5.089
 disturbed-5.045
 burned-5.042
 Tommy-4.939
Token the
Token ID262
Feature activation-1.24e+00

Pos logit contributions
 tweet+1.423
 anytime+1.420
 bou+1.382
 forests+1.323
 galaxies+1.284
Neg logit contributions
 stolen-1.067
 supposed-1.009
illing-0.972
 delayed-0.967
 thrown-0.963
Token big
Token ID1263
Feature activation-1.63e+00

Pos logit contributions
 anytime+1.682
 tweet+1.610
 bou+1.545
 forests+1.469
 galaxies+1.417
Neg logit contributions
 stolen-2.238
 supposed-2.184
 delayed-2.115
illing-2.098
 ruined-2.065
Token,
Token ID11
Feature activation-1.68e+00

Pos logit contributions
 anytime+2.512
 tweet+2.349
 bou+2.289
 galaxies+2.222
 forests+2.205
Neg logit contributions
 stolen-2.603
 supposed-2.435
 burned-2.412
 ruined-2.398
illing-2.397
Token blue
Token ID4171
Feature activation-7.12e-01

Pos logit contributions
 anytime+3.172
 tweet+2.980
 bou+2.865
 forests+2.856
 galaxies+2.822
Neg logit contributions
 stolen-2.253
 supposed-2.070
 unfair-2.061
 delayed-2.021
 ruined-1.995
Token the
Token ID262
Feature activation-2.24e+00

Pos logit contributions
 stolen+0.013
 supposed+0.013
illing+0.013
 delayed+0.012
 robbed+0.012
Neg logit contributions
 tweet-0.009
 anytime-0.008
 bou-0.008
 forests-0.008
 galaxies-0.007
Token tree
Token ID5509
Feature activation-1.38e+00

Pos logit contributions
 anytime+3.500
 tweet+3.431
 bou+3.350
 galaxies+3.222
cast+3.034
Neg logit contributions
 stolen-3.556
 delayed-3.365
 supposed-3.354
illing-3.327
 burned-3.324
Token.
Token ID13
Feature activation-1.20e+00

Pos logit contributions
 tweet+2.142
 bou+2.035
 anytime+2.014
 forests+1.935
 galaxies+1.928
Neg logit contributions
 stolen-2.177
 supposed-2.138
illing-2.109
 delayed-2.040
 thrown-2.016
Token She
Token ID1375
Feature activation-2.44e+00

Pos logit contributions
 anytime+2.084
 tweet+2.060
 bou+2.033
 forests+1.971
 galaxies+1.947
Neg logit contributions
 stolen-1.527
illing-1.500
 supposed-1.470
 unfair-1.421
 sale-1.411
Token would
Token ID561
Feature activation-2.07e+00

Pos logit contributions
 tweet+4.444
 bou+4.419
 anytime+4.248
 forests+4.079
 galaxies+4.036
Neg logit contributions
 stolen-3.388
 delayed-3.318
 burned-3.240
 sold-3.231
 disob-3.187
Token fly
Token ID6129
Feature activation-4.38e+00

Pos logit contributions
 anytime+3.463
 bou+3.364
 tweet+3.352
 forests+3.323
 galaxies+3.272
Neg logit contributions
 stolen-2.957
 supposed-2.866
 disob-2.837
 delayed-2.773
illing-2.731
Token around
Token ID1088
Feature activation-3.65e+00

Pos logit contributions
 Majesty+9.574
 tweet+9.163
 bou+9.054
 galaxies+8.850
 sings+8.832
Neg logit contributions
 eaten-4.845
 stolen-4.797
 thrown-4.598
 ruined-4.467
 delayed-4.326
Token and
Token ID290
Feature activation-4.13e+00

Pos logit contributions
 bou+6.104
 tweet+5.929
 galaxies+5.821
 Majesty+5.813
 nods+5.720
Neg logit contributions
 eaten-5.239
 sale-5.231
 stolen-5.216
 supposed-5.201
 burned-5.159
Token play
Token ID711
Feature activation-3.60e+00

Pos logit contributions
 bou+7.912
 anytime+7.397
 galaxies+7.036
 tweet+7.014
ig+6.931
Neg logit contributions
 stolen-5.896
 burned-5.828
 fought-5.750
 disob-5.680
 eaten-5.623
Token with
Token ID351
Feature activation-1.26e+00

Pos logit contributions
 tweet+6.569
 galaxies+6.528
 bou+6.413
 sings+6.253
 Majesty+6.212
Neg logit contributions
 sale-4.686
 thrown-4.413
 eaten-4.345
 stolen-4.310
 supposed-4.216
Token her
Token ID607
Feature activation-2.21e+00

Pos logit contributions
 tweet+1.902
 anytime+1.825
 bou+1.801
 forests+1.761
 galaxies+1.740
Neg logit contributions
 stolen-2.015
 supposed-1.910
 thrown-1.881
 burned-1.842
 ruined-1.834
Token became
Token ID2627
Feature activation+2.73e-01

Pos logit contributions
 tweet+1.120
 anytime+1.103
 bou+1.069
 forests+1.048
 galaxies+1.020
Neg logit contributions
 stolen-0.926
 supposed-0.872
illing-0.861
 delayed-0.852
 thrown-0.844
Token great
Token ID1049
Feature activation-6.67e-01

Pos logit contributions
 stolen+0.473
illing+0.460
 supposed+0.458
 delayed+0.441
 robbed+0.440
Neg logit contributions
 tweet-0.374
 anytime-0.370
 bou-0.358
 forests-0.350
 galaxies-0.330
Token friends
Token ID2460
Feature activation-3.31e-01

Pos logit contributions
 tweet+1.065
 anytime+1.018
 bou+1.003
 forests+0.968
 galaxies+0.929
Neg logit contributions
 stolen-1.028
 supposed-1.009
illing-0.991
 delayed-0.962
 thrown-0.946
Token.
Token ID13
Feature activation+1.65e-02

Pos logit contributions
 tweet+0.574
 anytime+0.558
 bou+0.550
 forests+0.535
 galaxies+0.516
Neg logit contributions
 stolen-0.454
 supposed-0.438
illing-0.433
 delayed-0.417
 thrown-0.410
Token They
Token ID1119
Feature activation-1.16e+00

Pos logit contributions
 stolen+0.022
 supposed+0.021
illing+0.020
 delayed+0.020
 robbed+0.019
Neg logit contributions
 tweet-0.030
 anytime-0.029
 bou-0.028
 forests-0.028
 galaxies-0.027
Token explored
Token ID18782
Feature activation-4.35e+00

Pos logit contributions
 tweet+2.134
 anytime+2.115
 bou+2.055
 forests+2.005
 galaxies+1.949
Neg logit contributions
 stolen-1.467
 supposed-1.357
 delayed-1.338
 thrown-1.325
 sold-1.322
Token the
Token ID262
Feature activation-3.04e+00

Pos logit contributions
 tweet+9.163
 Majesty+9.050
 sings+8.496
 bou+8.107
 anytime+8.082
Neg logit contributions
 stolen-5.336
 sold-5.259
 Tim-4.947
 sale-4.937
 thrown-4.885
Token peaceful
Token ID12309
Feature activation-2.88e+00

Pos logit contributions
 tweet+5.252
 anytime+4.724
 sings+4.531
 galaxies+4.523
 nods+4.519
Neg logit contributions
 stolen-4.959
 supposed-4.693
 delayed-4.527
 ruined-4.513
 thrown-4.383
Token garden
Token ID11376
Feature activation-2.43e+00

Pos logit contributions
 tweet+3.996
 anytime+3.713
 bou+3.530
 sings+3.334
 galaxies+3.323
Neg logit contributions
 stolen-5.175
illing-4.920
 burned-4.839
 thrown-4.790
 supposed-4.776
Token together
Token ID1978
Feature activation-2.18e+00

Pos logit contributions
 tweet+3.194
 bou+2.912
 anytime+2.736
 sings+2.728
 galaxies+2.621
Neg logit contributions
 stolen-4.451
 supposed-4.380
illing-4.366
 sold-4.314
 sale-4.239
Token and
Token ID290
Feature activation-2.14e+00

Pos logit contributions
 tweet+3.120
 bou+2.926
 anytime+2.753
 sings+2.734
 galaxies+2.659
Neg logit contributions
 stolen-3.687
 supposed-3.627
illing-3.620
 sale-3.501
 thrown-3.489
Token sky
Token ID6766
Feature activation-4.28e-01

Pos logit contributions
 tweet+0.568
 anytime+0.563
 bou+0.528
 forests+0.498
 galaxies+0.479
Neg logit contributions
 stolen-1.071
 supposed-1.035
illing-1.030
 delayed-1.012
 thrown-1.000
Token.
Token ID13
Feature activation-1.76e+00

Pos logit contributions
 tweet+0.789
 anytime+0.772
 bou+0.758
 forests+0.738
 galaxies+0.717
Neg logit contributions
 stolen-0.535
illing-0.515
 supposed-0.515
 delayed-0.490
 thrown-0.483
Token He
Token ID679
Feature activation-1.24e+00

Pos logit contributions
 bou+3.808
 tweet+3.706
 anytime+3.605
 forests+3.413
 galaxies+3.403
Neg logit contributions
illing-1.638
 stolen-1.630
 supposed-1.558
 delayed-1.531
 Tommy-1.508
Token soared
Token ID35292
Feature activation-3.30e+00

Pos logit contributions
 tweet+2.197
 anytime+2.173
 bou+2.161
 forests+2.038
 galaxies+1.987
Neg logit contributions
 stolen-1.731
 delayed-1.605
 thrown-1.585
 supposed-1.580
 burned-1.549
Token up
Token ID510
Feature activation-2.52e+00

Pos logit contributions
 tweet+5.504
 bou+5.238
 sings+5.119
 anytime+5.049
 Majesty+4.886
Neg logit contributions
 stolen-5.133
 delayed-4.904
 thrown-4.886
 supposed-4.842
 eaten-4.783
Token and
Token ID290
Feature activation-4.35e+00

Pos logit contributions
 tweet+3.819
 bou+3.606
 anytime+3.539
 Majesty+3.355
 sings+3.299
Neg logit contributions
 stolen-4.198
 burned-4.077
 thrown-4.070
 delayed-4.063
 supposed-4.047
Token through
Token ID832
Feature activation-1.90e+00

Pos logit contributions
 bou+9.077
 anytime+8.736
 tweet+8.700
 Him+8.307
 Majesty+8.170
Neg logit contributions
 stolen-5.836
 delayed-5.690
 driven-5.619
 fought-5.600
 thrown-5.564
Token the
Token ID262
Feature activation-1.86e+00

Pos logit contributions
 tweet+2.994
 bou+2.916
 anytime+2.824
 sings+2.654
 nods+2.535
Neg logit contributions
 stolen-2.931
 thrown-2.753
 burned-2.727
 supposed-2.701
 sale-2.665
Token clouds
Token ID15114
Feature activation-9.73e-01

Pos logit contributions
 anytime+2.506
 tweet+2.503
 bou+2.304
 sings+2.039
 galaxies+2.010
Neg logit contributions
 stolen-3.330
illing-3.223
 supposed-3.206
 delayed-3.150
 ruined-3.119
Token,
Token ID11
Feature activation-3.55e+00

Pos logit contributions
 tweet+1.685
 anytime+1.647
 bou+1.630
 forests+1.547
 galaxies+1.516
Neg logit contributions
 stolen-1.326
illing-1.279
 supposed-1.278
 delayed-1.230
 thrown-1.224
Token feeling
Token ID4203
Feature activation-1.16e+00

Pos logit contributions
 bou+7.236
 anytime+6.861
 tweet+6.704
 galaxies+6.383
ig+6.313
Neg logit contributions
 stolen-4.804
 supposed-4.552
 delayed-4.523
 thrown-4.488
 burned-4.472
Token off
Token ID572
Feature activation-8.09e-01

Pos logit contributions
 tweet+1.955
 bou+1.901
 anytime+1.897
 forests+1.820
 galaxies+1.762
Neg logit contributions
 supposed-1.619
 stolen-1.614
 delayed-1.528
 thrown-1.493
illing-1.491
Token!
Token ID0
Feature activation-1.41e-01

Pos logit contributions
 tweet+1.131
 anytime+1.089
 bou+1.076
 forests+1.029
 galaxies+0.993
Neg logit contributions
 stolen-1.389
 supposed-1.336
illing-1.307
 delayed-1.298
 thrown-1.289
Token He
Token ID679
Feature activation-2.68e+00

Pos logit contributions
 tweet+0.208
 anytime+0.203
 bou+0.197
 forests+0.192
 galaxies+0.184
Neg logit contributions
 stolen-0.231
 supposed-0.224
illing-0.222
 delayed-0.215
 robbed-0.212
Token flew
Token ID13112
Feature activation-2.65e+00

Pos logit contributions
 bou+5.273
 tweet+5.210
 anytime+5.093
 forests+4.889
 cities+4.768
Neg logit contributions
 stolen-3.298
 delayed-3.187
 thrown-3.146
 burned-3.058
 sold-3.058
Token higher
Token ID2440
Feature activation-3.93e+00

Pos logit contributions
 tweet+3.862
 sings+3.636
 bou+3.602
 anytime+3.557
 galaxies+3.352
Neg logit contributions
 stolen-4.496
 delayed-4.264
 Timothy-4.251
 thrown-4.218
 supposed-4.216
Token and
Token ID290
Feature activation-4.33e+00

Pos logit contributions
 tweet+6.652
 bou+5.977
 sings+5.621
 anytime+5.524
 Majesty+5.337
Neg logit contributions
 thrown-6.313
 stolen-6.275
 delayed-6.219
 burned-6.195
 driven-6.027
Token higher
Token ID2440
Feature activation-2.96e+00

Pos logit contributions
 bou+7.176
 tweet+6.777
 anytime+5.997
 Him+5.853
 Majesty+5.792
Neg logit contributions
 delayed-8.221
 stolen-7.926
 thrown-7.901
 driven-7.781
 sold-7.685
Token and
Token ID290
Feature activation-4.11e+00

Pos logit contributions
 tweet+4.748
 bou+4.525
 anytime+4.261
 sings+4.155
 galaxies+3.945
Neg logit contributions
 stolen-4.654
 delayed-4.484
 thrown-4.467
 supposed-4.457
 burned-4.394
Token was
Token ID373
Feature activation-1.60e+00

Pos logit contributions
 bou+6.792
 tweet+6.360
 anytime+5.566
 Him+5.277
 sings+5.242
Neg logit contributions
 delayed-7.635
 stolen-7.326
 thrown-7.205
 sold-7.148
 driven-7.078
Token amazed
Token ID24562
Feature activation-3.75e-01

Pos logit contributions
 tweet+2.805
 bou+2.764
 anytime+2.696
 forests+2.580
 galaxies+2.511
Neg logit contributions
 supposed-2.291
 stolen-2.235
 delayed-2.136
 thrown-2.095
 driven-2.032
Token at
Token ID379
Feature activation-1.57e+00

Pos logit contributions
 tweet+0.596
 anytime+0.576
 bou+0.564
 forests+0.549
 galaxies+0.526
Neg logit contributions
 stolen-0.571
 supposed-0.551
illing-0.548
 delayed-0.531
 thrown-0.519
Token tw
Token ID665
Feature activation-2.83e+00

Pos logit contributions
 bou+5.793
 tweet+5.588
 anytime+5.547
 forests+5.289
 cities+5.217
Neg logit contributions
 delayed-3.435
 burned-3.332
 thrown-3.296
 stolen-3.260
 disturbed-3.190
Tokenirled
Token ID49376
Feature activation-3.92e+00

Pos logit contributions
 tweet+4.942
 anytime+4.832
 bou+4.620
 forests+4.614
 cities+4.568
Neg logit contributions
omo-4.054
illing-4.000
 thrown-3.968
 stolen-3.845
 supposed-3.838
Token around
Token ID1088
Feature activation-3.28e+00

Pos logit contributions
 tweet+6.788
 Majesty+6.531
 anytime+6.304
 bou+6.227
 sings+6.041
Neg logit contributions
 Tommy-5.596
 thrown-5.566
 Tim-5.472
 delayed-5.414
 stolen-5.407
Token the
Token ID262
Feature activation-3.27e+00

Pos logit contributions
 bou+5.427
 anytime+5.346
 tweet+5.327
 Majesty+5.199
 galaxies+5.142
Neg logit contributions
 burned-4.366
 stolen-4.333
 thrown-4.321
 sale-4.320
 eaten-4.258
Token sun
Token ID4252
Feature activation-2.60e+00

Pos logit contributions
 anytime+5.228
 tweet+5.165
 bou+4.777
 galaxies+4.469
cast+4.410
Neg logit contributions
 stolen-5.206
 ruined-5.086
 burned-5.060
 delayed-5.005
 supposed-5.005
Token and
Token ID290
Feature activation-4.33e+00

Pos logit contributions
 tweet+3.927
 anytime+3.884
 bou+3.818
 galaxies+3.774
 forests+3.745
Neg logit contributions
illing-3.827
 supposed-3.818
 burned-3.792
 stolen-3.771
 delayed-3.768
Token spun
Token ID26843
Feature activation-3.52e+00

Pos logit contributions
 bou+9.051
 anytime+8.747
 tweet+8.463
 Majesty+8.110
ig+8.044
Neg logit contributions
 thrown-5.374
 burned-5.330
 delayed-5.195
 disturbed-5.179
 spilled-5.173
Token in
Token ID287
Feature activation-2.16e+00

Pos logit contributions
 tweet+6.116
 anytime+5.878
 bou+5.519
 Majesty+5.479
 sings+5.437
Neg logit contributions
 thrown-4.805
 stolen-4.653
 supposed-4.582
 Tommy-4.571
 delayed-4.498
Token the
Token ID262
Feature activation-2.46e+00

Pos logit contributions
 tweet+3.316
 anytime+3.218
 bou+3.015
 sings+3.003
 galaxies+2.955
Neg logit contributions
 stolen-3.215
 thrown-3.132
 supposed-3.082
 Tommy-3.070
 sale-3.060
Token breeze
Token ID28633
Feature activation-7.25e-01

Pos logit contributions
 anytime+3.953
 tweet+3.671
 bou+3.470
cast+3.274
 galaxies+3.246
Neg logit contributions
 stolen-3.840
 ruined-3.740
 burned-3.682
 delayed-3.662
illing-3.626
Token.
Token ID13
Feature activation-9.09e-01

Pos logit contributions
 tweet+1.352
 anytime+1.319
 bou+1.299
 forests+1.273
 galaxies+1.239
Neg logit contributions
 stolen-0.878
 supposed-0.856
illing-0.848
 delayed-0.811
 thrown-0.798
Token day
Token ID1110
Feature activation+6.34e-01

Pos logit contributions
 stolen+1.001
 supposed+0.955
 robbed+0.936
illing+0.935
 delayed+0.928
Neg logit contributions
 tweet-0.762
 anytime-0.747
 bou-0.706
 forests-0.687
 galaxies-0.644
Token,
Token ID11
Feature activation+5.93e-01

Pos logit contributions
 stolen+1.319
 supposed+1.291
illing+1.277
 delayed+1.247
 robbed+1.244
Neg logit contributions
 tweet-0.663
 anytime-0.648
 bou-0.600
 forests-0.571
 sings-0.522
Token Lily
Token ID20037
Feature activation+5.94e-02

Pos logit contributions
 stolen+1.037
 supposed+1.030
illing+1.020
 robbed+0.967
 delayed+0.966
Neg logit contributions
 tweet-0.818
 anytime-0.808
 bou-0.779
 forests-0.736
 galaxies-0.696
Token and
Token ID290
Feature activation-1.35e-01

Pos logit contributions
 stolen+0.101
 supposed+0.098
illing+0.097
 delayed+0.094
 robbed+0.093
Neg logit contributions
 tweet-0.082
 anytime-0.081
 bou-0.078
 forests-0.076
 galaxies-0.073
Token her
Token ID607
Feature activation-1.42e-01

Pos logit contributions
 tweet+0.181
 anytime+0.177
 bou+0.171
 forests+0.166
 galaxies+0.158
Neg logit contributions
 stolen-0.242
 supposed-0.235
illing-0.233
 delayed-0.227
 robbed-0.224
Token friends
Token ID2460
Feature activation+2.15e+00

Pos logit contributions
 tweet+0.218
 anytime+0.213
 bou+0.207
 forests+0.202
 galaxies+0.194
Neg logit contributions
 stolen-0.224
 supposed-0.217
illing-0.215
 delayed-0.208
 robbed-0.205
Token were
Token ID547
Feature activation+7.83e-01

Pos logit contributions
illing+3.485
omo+3.413
 supposed+3.398
 stolen+3.317
 mercury+3.278
Neg logit contributions
 tweet-3.166
 anytime-3.131
 bou-3.080
 forests-2.861
 wherever-2.795
Token playing
Token ID2712
Feature activation-1.27e+00

Pos logit contributions
 stolen+1.011
illing+0.961
omo+0.936
 supposed+0.930
 unfair+0.914
Neg logit contributions
 tweet-1.405
 anytime-1.387
 forests-1.338
 bou-1.337
 sings-1.253
Token hide
Token ID7808
Feature activation-1.47e+00

Pos logit contributions
 tweet+1.883
 anytime+1.822
 bou+1.810
 forests+1.759
 galaxies+1.758
Neg logit contributions
 stolen-2.038
illing-1.944
 supposed-1.932
 thrown-1.925
 delayed-1.912
Token and
Token ID290
Feature activation-6.57e-01

Pos logit contributions
 tweet+2.093
 bou+1.867
 anytime+1.862
 forests+1.773
 galaxies+1.746
Neg logit contributions
 stolen-2.604
illing-2.561
 supposed-2.515
 delayed-2.472
 thrown-2.458
Token seek
Token ID5380
Feature activation+4.85e-01

Pos logit contributions
 tweet+0.936
 anytime+0.913
 bou+0.888
 forests+0.863
 galaxies+0.835
Neg logit contributions
 stolen-1.110
 supposed-1.065
illing-1.038
 delayed-1.033
 thrown-1.022
Token in
Token ID287
Feature activation+3.81e-01

Pos logit contributions
 tweet+0.392
 anytime+0.380
 bou+0.370
 forests+0.358
 galaxies+0.343
Neg logit contributions
 stolen-0.528
 supposed-0.513
illing-0.508
 delayed-0.495
 thrown-0.489
Token the
Token ID262
Feature activation-3.31e-01

Pos logit contributions
 stolen+0.633
illing+0.617
 supposed+0.616
 delayed+0.594
omo+0.585
Neg logit contributions
 tweet-0.546
 anytime-0.541
 forests-0.521
 bou-0.516
 galaxies-0.493
Token ocean
Token ID9151
Feature activation+8.22e-01

Pos logit contributions
 tweet+0.453
 anytime+0.441
 bou+0.426
 forests+0.409
 galaxies+0.394
Neg logit contributions
 stolen-0.578
 supposed-0.557
illing-0.554
 delayed-0.539
 robbed-0.532
Token with
Token ID351
Feature activation+1.45e+00

Pos logit contributions
 stolen+1.317
 supposed+1.278
 Timothy+1.231
 unfair+1.225
illing+1.220
Neg logit contributions
 anytime-1.255
 tweet-1.202
 forests-1.185
 bou-1.159
 galaxies-1.108
Token his
Token ID465
Feature activation+4.51e-02

Pos logit contributions
illing+2.073
 stolen+2.038
 supposed+2.037
 delayed+1.983
 robbed+1.955
Neg logit contributions
 tweet-2.460
 anytime-2.454
 forests-2.419
 bou-2.416
 galaxies-2.320
Token family
Token ID1641
Feature activation+2.00e+00

Pos logit contributions
 stolen+0.071
 supposed+0.069
illing+0.069
 delayed+0.066
 robbed+0.065
Neg logit contributions
 tweet-0.069
 anytime-0.068
 bou-0.066
 forests-0.064
 galaxies-0.061
Token.
Token ID13
Feature activation+8.32e-01

Pos logit contributions
 stolen+3.596
 unfair+3.455
 supposed+3.427
 delayed+3.404
 robbed+3.339
Neg logit contributions
 anytime-2.982
 wherever-2.683
 bou-2.670
 forests-2.665
 tweet-2.597
Token One
Token ID1881
Feature activation+2.67e-01

Pos logit contributions
 stolen+1.280
 supposed+1.226
illing+1.183
 robbed+1.168
 delayed+1.160
Neg logit contributions
 anytime-1.353
 tweet-1.325
 forests-1.292
 bou-1.269
 galaxies-1.234
Token day
Token ID1110
Feature activation+1.11e+00

Pos logit contributions
 stolen+0.457
 supposed+0.442
illing+0.432
 delayed+0.424
 robbed+0.423
Neg logit contributions
 tweet-0.372
 anytime-0.366
 bou-0.350
 forests-0.344
 galaxies-0.326
Token,
Token ID11
Feature activation+1.23e+00

Pos logit contributions
 stolen+2.397
 supposed+2.362
illing+2.322
 unfair+2.269
 robbed+2.266
Neg logit contributions
 tweet-1.057
 anytime-1.052
 bou-0.931
 forests-0.927
 wherever-0.861
Token the
Token ID262
Feature activation+3.41e-01

Pos logit contributions
 stolen+1.794
 supposed+1.777
illing+1.752
 robbed+1.648
 disob+1.630
Neg logit contributions
 anytime-2.093
 tweet-2.012
 forests-1.954
 bou-1.911
 wherever-1.905
Token.
Token ID13
Feature activation+1.54e+00

Pos logit contributions
 stolen+2.252
 supposed+2.172
 unfair+2.114
illing+2.100
omo+2.073
Neg logit contributions
 anytime-1.754
 tweet-1.648
 forests-1.561
 bou-1.532
 wherever-1.461
Token He
Token ID679
Feature activation+5.88e-01

Pos logit contributions
 stolen+2.305
 supposed+2.172
 robbed+2.078
 thrown+2.060
 disob+2.058
Neg logit contributions
 anytime-2.717
 forests-2.606
 galaxies-2.489
 tweet-2.487
 valleys-2.461
Token had
Token ID550
Feature activation+2.02e+00

Pos logit contributions
 stolen+0.935
 supposed+0.915
illing+0.909
 robbed+0.864
 delayed+0.861
Neg logit contributions
 tweet-0.872
 anytime-0.870
 forests-0.823
 bou-0.816
 galaxies-0.786
Token a
Token ID257
Feature activation+6.39e-01

Pos logit contributions
illing+3.511
 stolen+3.274
omo+3.231
 supposed+3.213
olded+3.174
Neg logit contributions
 forests-2.789
 anytime-2.750
 tweet-2.687
 galaxies-2.602
 valleys-2.521
Token pet
Token ID4273
Feature activation+5.40e-01

Pos logit contributions
illing+1.029
 stolen+1.027
 supposed+1.021
 delayed+0.973
 thrown+0.968
Neg logit contributions
 tweet-0.937
 bou-0.904
 anytime-0.902
 forests-0.849
 galaxies-0.811
Token dog
Token ID3290
Feature activation+2.09e+00

Pos logit contributions
 stolen+0.939
 supposed+0.934
illing+0.910
 delayed+0.889
 thrown+0.876
Neg logit contributions
 tweet-0.748
 anytime-0.726
 bou-0.703
 forests-0.673
 galaxies-0.633
Token called
Token ID1444
Feature activation+1.19e+00

Pos logit contributions
 supposed+3.890
 stolen+3.864
illing+3.816
 delayed+3.713
omo+3.678
Neg logit contributions
 anytime-2.811
 tweet-2.706
 bou-2.461
 forests-2.434
 Sparkle-2.430
Token Bos
Token ID14548
Feature activation+6.56e-01

Pos logit contributions
 stolen+1.893
illing+1.844
 supposed+1.835
 delayed+1.814
omo+1.780
Neg logit contributions
 tweet-1.851
 anytime-1.787
 bou-1.694
 forests-1.673
 Sparkle-1.594
Tokenco
Token ID1073
Feature activation+1.64e+00

Pos logit contributions
 stolen+1.106
 supposed+1.062
illing+1.040
 delayed+1.026
 thrown+1.021
Neg logit contributions
 anytime-0.936
 tweet-0.935
 bou-0.881
 forests-0.871
 galaxies-0.822
Token.
Token ID13
Feature activation+1.23e+00

Pos logit contributions
 stolen+2.679
 delayed+2.556
 supposed+2.548
 unfair+2.541
 robbed+2.538
Neg logit contributions
 anytime-2.508
 tweet-2.326
 forests-2.230
 bou-2.190
 wherever-2.110
Token Bos
Token ID14548
Feature activation+6.06e-01

Pos logit contributions
 stolen+1.903
 supposed+1.836
illing+1.735
 robbed+1.728
 thrown+1.698
Neg logit contributions
 anytime-2.078
 tweet-1.969
 forests-1.960
 valleys-1.853
 bou-1.847
Token truck
Token ID7779
Feature activation+9.04e-01

Pos logit contributions
 supposed+1.324
illing+1.318
 stolen+1.304
 delayed+1.217
 thrown+1.211
Neg logit contributions
 tweet-1.761
 anytime-1.693
 forests-1.659
 bou-1.648
 cities-1.566
Token driver
Token ID4639
Feature activation+6.73e-01

Pos logit contributions
 supposed+1.322
 stolen+1.256
illing+1.254
 robbed+1.145
 thrown+1.142
Neg logit contributions
 tweet-1.532
 anytime-1.483
 bou-1.429
 forests-1.406
 galaxies-1.327
Token was
Token ID373
Feature activation+2.14e+00

Pos logit contributions
 supposed+1.259
 stolen+1.238
illing+1.214
 delayed+1.151
 robbed+1.150
Neg logit contributions
 tweet-0.840
 anytime-0.815
 bou-0.770
 forests-0.747
 galaxies-0.698
Token angry
Token ID7954
Feature activation+6.35e-02

Pos logit contributions
illing+3.631
 supposed+3.429
 stolen+3.411
attered+3.407
 Timothy+3.300
Neg logit contributions
 tweet-2.902
 anytime-2.880
 forests-2.729
 bou-2.719
 higher-2.642
Token and
Token ID290
Feature activation+1.77e+00

Pos logit contributions
 stolen+0.142
 supposed+0.139
illing+0.137
 delayed+0.135
 robbed+0.133
Neg logit contributions
 tweet-0.055
 anytime-0.054
 bou-0.050
 forests-0.048
 galaxies-0.044
Token hurt
Token ID5938
Feature activation+2.75e+00

Pos logit contributions
illing+2.658
 supposed+2.619
 stolen+2.419
omo+2.374
ction+2.328
Neg logit contributions
 tweet-2.994
 anytime-2.771
 bou-2.655
 forests-2.622
 sings-2.608
Token.
Token ID13
Feature activation+1.29e+00

Pos logit contributions
 supposed+5.523
illing+5.349
 stolen+5.321
 robbed+4.996
omo+4.975
Neg logit contributions
 tweet-3.606
 anytime-3.491
 until-3.272
 forests-3.239
 bou-3.059
Token He
Token ID679
Feature activation+1.05e+00

Pos logit contributions
 stolen+1.822
 supposed+1.789
illing+1.754
 delayed+1.646
 robbed+1.637
Neg logit contributions
 tweet-2.207
 anytime-2.190
 forests-2.078
 bou-2.029
 valleys-1.986
Token had
Token ID550
Feature activation+3.14e+00

Pos logit contributions
illing+1.516
 supposed+1.500
 stolen+1.419
omo+1.355
 Timothy+1.337
Neg logit contributions
 tweet-1.722
 anytime-1.661
 bou-1.614
 forests-1.582
 sings-1.535
Token a
Token ID257
Feature activation+8.29e-01

Pos logit contributions
illing+6.080
omo+5.296
attered+5.146
 supposed+5.027
ooked+4.908
Neg logit contributions
 tweet-4.532
 anytime-4.244
 forests-4.210
 bou-3.996
 hopes-3.913
Token cut
Token ID2005
Feature activation+1.08e+00

Pos logit contributions
illing+1.223
 supposed+1.187
 stolen+1.168
 delayed+1.123
omo+1.096
Neg logit contributions
 tweet-1.383
 bou-1.312
 anytime-1.278
 forests-1.230
 sings-1.169
Token,
Token ID11
Feature activation-6.73e-01

Pos logit contributions
 stolen+0.057
 supposed+0.055
illing+0.054
 delayed+0.053
 robbed+0.052
Neg logit contributions
 tweet-0.049
 anytime-0.048
 bou-0.046
 forests-0.045
 galaxies-0.043
Token until
Token ID1566
Feature activation+8.43e-01

Pos logit contributions
 tweet+0.936
 bou+0.892
 anytime+0.890
 forests+0.850
 galaxies+0.817
Neg logit contributions
 stolen-1.167
 supposed-1.129
illing-1.108
 delayed-1.095
 thrown-1.081
Token suddenly
Token ID6451
Feature activation+1.01e+00

Pos logit contributions
 stolen+1.464
illing+1.458
 supposed+1.450
 robbed+1.356
 unfair+1.352
Neg logit contributions
 anytime-1.171
 tweet-1.127
 forests-1.083
 bou-1.043
 galaxies-1.018
Token the
Token ID262
Feature activation+1.21e+00

Pos logit contributions
illing+1.757
 supposed+1.748
 stolen+1.741
omo+1.605
 robbed+1.604
Neg logit contributions
 anytime-1.406
 tweet-1.395
 forests-1.356
 galaxies-1.291
 bou-1.278
Token bag
Token ID6131
Feature activation+1.87e+00

Pos logit contributions
 supposed+2.286
illing+2.260
 stolen+2.244
olded+2.127
 delayed+2.126
Neg logit contributions
 tweet-1.494
 anytime-1.406
 bou-1.402
 forests-1.388
 galaxies-1.244
Token broke
Token ID6265
Feature activation+2.01e+00

Pos logit contributions
 supposed+2.791
illing+2.675
 stolen+2.598
 unfair+2.545
 Timothy+2.454
Neg logit contributions
 tweet-3.060
 anytime-3.007
 bou-2.891
 forests-2.846
 valleys-2.689
Token.
Token ID13
Feature activation+1.49e+00

Pos logit contributions
illing+4.366
 stolen+4.349
 supposed+4.241
 unfair+4.237
 paramed+4.175
Neg logit contributions
 anytime-2.012
 forests-1.862
 tweet-1.772
 valleys-1.742
 bou-1.721
Token All
Token ID1439
Feature activation+1.40e+00

Pos logit contributions
 stolen+3.096
 supposed+2.985
illing+2.918
 robbed+2.869
 delayed+2.864
Neg logit contributions
 anytime-1.648
 tweet-1.541
 forests-1.531
 valleys-1.418
 galaxies-1.368
Token of
Token ID286
Feature activation+1.07e+00

Pos logit contributions
 stolen+2.499
 supposed+2.446
illing+2.430
 sale+2.328
omo+2.324
Neg logit contributions
 anytime-1.892
 tweet-1.811
 forests-1.779
 galaxies-1.706
 valleys-1.660
Token his
Token ID465
Feature activation+3.44e-01

Pos logit contributions
 stolen+1.793
 supposed+1.777
illing+1.769
 delayed+1.674
 robbed+1.664
Neg logit contributions
 anytime-1.550
 tweet-1.510
 forests-1.498
 bou-1.436
 galaxies-1.413
Token toys
Token ID14958
Feature activation+1.11e+00

Pos logit contributions
 stolen+0.562
 supposed+0.552
illing+0.551
 delayed+0.531
 robbed+0.523
Neg logit contributions
 tweet-0.498
 anytime-0.488
 bou-0.470
 forests-0.463
 galaxies-0.440
Token
Token ID198
Feature activation+9.76e-01

Pos logit contributions
 stolen+0.346
 supposed+0.337
illing+0.331
 delayed+0.325
 robbed+0.322
Neg logit contributions
 tweet-0.202
 anytime-0.201
 bou-0.189
 forests-0.184
 galaxies-0.171
Token"
Token ID1
Feature activation+2.63e-01

Pos logit contributions
 stolen+1.925
 supposed+1.907
illing+1.779
 delayed+1.772
 robbed+1.765
Neg logit contributions
 anytime-1.163
 tweet-1.117
 bou-1.050
 forests-1.038
 valleys-0.947
TokenNo
Token ID2949
Feature activation+8.59e-01

Pos logit contributions
 stolen+0.483
 supposed+0.472
illing+0.463
 delayed+0.450
 robbed+0.449
Neg logit contributions
 tweet-0.337
 anytime-0.331
 bou-0.317
 forests-0.306
 galaxies-0.290
Token,
Token ID11
Feature activation-4.99e-01

Pos logit contributions
 stolen+1.800
illing+1.782
 supposed+1.764
 delayed+1.724
 robbed+1.721
Neg logit contributions
 tweet-0.912
 anytime-0.900
 bou-0.843
 forests-0.775
 sings-0.716
Token I
Token ID314
Feature activation+1.04e+00

Pos logit contributions
 tweet+0.681
 anytime+0.668
 bou+0.649
 forests+0.642
 galaxies+0.612
Neg logit contributions
 stolen-0.867
 supposed-0.831
illing-0.819
 delayed-0.807
 thrown-0.797
Token don
Token ID836
Feature activation+1.27e+00

Pos logit contributions
illing+1.869
 stolen+1.856
 supposed+1.819
 Timothy+1.759
 robbed+1.756
Neg logit contributions
 anytime-1.337
 tweet-1.295
 bou-1.184
 forests-1.104
 sings-1.090
Token't
Token ID470
Feature activation+1.87e+00

Pos logit contributions
 stolen+1.594
illing+1.445
 supposed+1.415
 robbed+1.410
 delayed+1.407
Neg logit contributions
 anytime-2.414
 tweet-2.399
 bou-2.227
 forests-2.205
 sings-2.160
Token have
Token ID423
Feature activation+1.21e+00

Pos logit contributions
 stolen+3.348
illing+3.299
 robbed+3.286
 paramed+3.238
 supposed+3.192
Neg logit contributions
 anytime-2.536
 tweet-2.428
 bou-2.204
 wherever-2.055
 until-2.015
Token any
Token ID597
Feature activation+1.15e+00

Pos logit contributions
illing+1.945
 stolen+1.834
 supposed+1.826
omo+1.818
 paramed+1.775
Neg logit contributions
 anytime-1.863
 tweet-1.793
 bou-1.737
 forests-1.711
 wherever-1.601
Token,"
Token ID553
Feature activation+6.06e-01

Pos logit contributions
illing+1.607
 supposed+1.572
 stolen+1.561
 delayed+1.523
 driven+1.486
Neg logit contributions
 anytime-1.988
 tweet-1.914
 bou-1.887
 forests-1.782
 wherever-1.699
Token Lily
Token ID20037
Feature activation+1.18e+00

Pos logit contributions
 supposed+0.821
 stolen+0.821
illing+0.797
 robbed+0.748
 delayed+0.745
Neg logit contributions
 tweet-1.067
 anytime-1.059
 bou-1.025
 forests-0.976
 sings-0.945
Token making
Token ID1642
Feature activation-2.32e-01

Pos logit contributions
 stolen+1.628
 supposed+1.562
illing+1.559
omo+1.546
 robbed+1.520
Neg logit contributions
 tweet-1.494
 anytime-1.454
 bou-1.438
 forests-1.429
 valleys-1.316
Token different
Token ID1180
Feature activation-7.93e-01

Pos logit contributions
 tweet+0.367
 anytime+0.359
 bou+0.347
 forests+0.340
 galaxies+0.327
Neg logit contributions
 stolen-0.353
 supposed-0.341
illing-0.337
 delayed-0.327
 thrown-0.322
Token animals
Token ID4695
Feature activation-1.62e-01

Pos logit contributions
 tweet+1.120
 anytime+1.064
 bou+1.026
 forests+0.995
 galaxies+0.945
Neg logit contributions
 stolen-1.354
 supposed-1.322
illing-1.317
 delayed-1.264
 thrown-1.251
Token and
Token ID290
Feature activation+1.92e-01

Pos logit contributions
 tweet+0.201
 anytime+0.196
 bou+0.189
 forests+0.183
 galaxies+0.174
Neg logit contributions
 stolen-0.302
 supposed-0.293
illing-0.291
 delayed-0.284
 robbed-0.280
Token shapes
Token ID15268
Feature activation+6.83e-01

Pos logit contributions
 stolen+0.272
 supposed+0.263
illing+0.263
 delayed+0.251
 robbed+0.249
Neg logit contributions
 tweet-0.323
 anytime-0.317
 bou-0.309
 forests-0.303
 galaxies-0.291
Token with
Token ID351
Feature activation+8.49e-01

Pos logit contributions
 stolen+1.168
 supposed+1.126
illing+1.113
 robbed+1.092
 delayed+1.091
Neg logit contributions
 tweet-0.969
 anytime-0.946
 bou-0.927
 forests-0.892
 galaxies-0.852
Token the
Token ID262
Feature activation+2.86e-01

Pos logit contributions
illing+1.512
 stolen+1.500
 supposed+1.499
 delayed+1.447
omo+1.432
Neg logit contributions
 tweet-1.136
 bou-1.093
 anytime-1.083
 forests-1.067
 galaxies-1.004
Token clay
Token ID21558
Feature activation+5.02e-01

Pos logit contributions
 stolen+0.412
illing+0.399
 supposed+0.399
 delayed+0.384
 robbed+0.378
Neg logit contributions
 tweet-0.475
 anytime-0.461
 bou-0.453
 forests-0.443
 galaxies-0.426
Token.
Token ID13
Feature activation+9.51e-01

Pos logit contributions
 stolen+0.779
 supposed+0.757
illing+0.750
 delayed+0.726
 robbed+0.718
Neg logit contributions
 tweet-0.784
 anytime-0.768
 bou-0.748
 forests-0.727
 galaxies-0.696
Token Suddenly
Token ID24975
Feature activation+3.45e-01

Pos logit contributions
 stolen+1.474
 supposed+1.448
illing+1.388
 disob+1.364
 robbed+1.353
Neg logit contributions
 anytime-1.514
 tweet-1.496
 bou-1.482
 forests-1.416
 valleys-1.333
Token,
Token ID11
Feature activation+1.14e+00

Pos logit contributions
 stolen+0.689
 supposed+0.683
illing+0.679
 delayed+0.649
 robbed+0.645
Neg logit contributions
 tweet-0.383
 anytime-0.370
 bou-0.356
 forests-0.346
 galaxies-0.320
Token too
Token ID1165
Feature activation+1.85e-01

Pos logit contributions
 tweet+1.797
 bou+1.731
 forests+1.685
 anytime+1.675
 galaxies+1.610
Neg logit contributions
 stolen-1.554
 supposed-1.545
illing-1.490
 delayed-1.478
 thrown-1.439
Token because
Token ID780
Feature activation+1.58e-01

Pos logit contributions
 stolen+0.263
 supposed+0.251
illing+0.249
 delayed+0.239
 robbed+0.238
Neg logit contributions
 tweet-0.310
 anytime-0.308
 bou-0.296
 forests-0.291
 galaxies-0.280
Token it
Token ID340
Feature activation-4.92e-01

Pos logit contributions
 stolen+0.289
illing+0.280
 supposed+0.279
 delayed+0.270
 robbed+0.269
Neg logit contributions
 tweet-0.201
 anytime-0.201
 bou-0.189
 forests-0.186
 galaxies-0.177
Token was
Token ID373
Feature activation+9.10e-01

Pos logit contributions
 tweet+0.882
 anytime+0.860
 bou+0.848
 forests+0.825
 galaxies+0.791
Neg logit contributions
 stolen-0.655
 supposed-0.634
illing-0.618
 delayed-0.608
 thrown-0.591
Token no
Token ID645
Feature activation+1.36e+00

Pos logit contributions
 stolen+1.585
illing+1.573
 supposed+1.495
 robbed+1.490
olded+1.469
Neg logit contributions
 anytime-1.223
 tweet-1.212
 bou-1.166
 forests-1.123
 galaxies-1.057
Token longer
Token ID2392
Feature activation+1.77e+00

Pos logit contributions
 stolen+2.572
illing+2.558
 supposed+2.542
 robbed+2.461
olded+2.391
Neg logit contributions
 anytime-1.678
 tweet-1.617
 bou-1.595
 forests-1.461
 galaxies-1.382
Token rotting
Token ID44494
Feature activation+4.23e-01

Pos logit contributions
 stolen+2.343
omo+2.285
illing+2.268
 Timothy+2.251
 sale+2.208
Neg logit contributions
 tweet-2.995
 anytime-2.989
 forests-2.915
 bou-2.909
 galaxies-2.814
Token away
Token ID1497
Feature activation+7.33e-02

Pos logit contributions
 stolen+0.679
 supposed+0.659
illing+0.656
omo+0.632
 delayed+0.627
Neg logit contributions
 anytime-0.630
 tweet-0.629
 bou-0.603
 forests-0.593
 galaxies-0.562
Token.
Token ID13
Feature activation-1.17e+00

Pos logit contributions
 stolen+0.105
 supposed+0.101
illing+0.101
 delayed+0.097
 robbed+0.095
Neg logit contributions
 tweet-0.122
 anytime-0.120
 bou-0.116
 forests-0.114
 galaxies-0.110
Token The
Token ID383
Feature activation-5.79e-01

Pos logit contributions
 tweet+1.737
 bou+1.685
 anytime+1.640
 forests+1.621
 galaxies+1.555
Neg logit contributions
illing-1.806
 stolen-1.806
 supposed-1.803
omo-1.739
 Timothy-1.721
Token bird
Token ID6512
Feature activation-1.71e+00

Pos logit contributions
 tweet+0.801
 anytime+0.797
 bou+0.761
 forests+0.745
 galaxies+0.718
Neg logit contributions
 stolen-0.996
 supposed-0.952
illing-0.951
 delayed-0.936
omo-0.918
Token!"
Token ID2474
Feature activation+7.82e-01

Pos logit contributions
 stolen+0.783
illing+0.765
 supposed+0.763
 delayed+0.741
omo+0.733
Neg logit contributions
 tweet-0.475
 anytime-0.473
 bou-0.452
 forests-0.431
 galaxies-0.408
Token he
Token ID339
Feature activation+7.08e-01

Pos logit contributions
 stolen+1.090
 supposed+1.060
illing+1.052
 delayed+1.000
 robbed+0.996
Neg logit contributions
 anytime-1.352
 tweet-1.339
 bou-1.295
 forests-1.259
 galaxies-1.213
Token said
Token ID531
Feature activation+1.12e+00

Pos logit contributions
 stolen+1.018
 supposed+1.009
illing+0.997
 delayed+0.938
 robbed+0.937
Neg logit contributions
 tweet-1.172
 anytime-1.148
 bou-1.133
 forests-1.076
 galaxies-1.050
Token excited
Token ID6568
Feature activation+4.20e-01

Pos logit contributions
 stolen+1.970
 supposed+1.916
illing+1.905
omo+1.873
 delayed+1.843
Neg logit contributions
 tweet-1.530
 anytime-1.527
 bou-1.456
 forests-1.321
 wherever-1.306
Tokenly
Token ID306
Feature activation+1.79e+00

Pos logit contributions
 stolen+0.795
 supposed+0.774
illing+0.764
 delayed+0.749
 robbed+0.745
Neg logit contributions
 tweet-0.512
 anytime-0.503
 bou-0.483
 forests-0.459
 galaxies-0.436
Token.
Token ID13
Feature activation+9.40e-01

Pos logit contributions
 stolen+3.233
 supposed+3.172
illing+3.101
 unfair+3.077
omo+3.051
Neg logit contributions
 anytime-2.387
 tweet-2.299
 bou-2.271
 until-2.088
 forests-2.064
Token "
Token ID366
Feature activation-1.14e+00

Pos logit contributions
 stolen+1.797
 supposed+1.743
illing+1.716
 robbed+1.666
 delayed+1.665
Neg logit contributions
 anytime-1.189
 tweet-1.146
 bou-1.092
 forests-1.049
 galaxies-0.974
TokenWow
Token ID22017
Feature activation-1.34e+00

Pos logit contributions
 tweet+1.884
 anytime+1.839
 bou+1.826
 forests+1.732
 sings+1.670
Neg logit contributions
 stolen-1.624
illing-1.604
 supposed-1.595
 delayed-1.527
 driven-1.500
Token,
Token ID11
Feature activation-4.31e-01

Pos logit contributions
 tweet+1.954
 bou+1.904
 forests+1.902
 anytime+1.901
 galaxies+1.829
Neg logit contributions
 stolen-2.170
 supposed-2.120
 delayed-2.029
 Tommy-2.014
 ruined-2.012
Token Tim
Token ID5045
Feature activation+1.20e-01

Pos logit contributions
 tweet+0.860
 anytime+0.851
 bou+0.833
 forests+0.822
 galaxies+0.793
Neg logit contributions
 stolen-0.479
 supposed-0.451
illing-0.433
 delayed-0.427
 thrown-0.416
Tokenmy
Token ID1820
Feature activation-2.20e-01

Pos logit contributions
 stolen+0.183
 supposed+0.176
illing+0.176
 delayed+0.170
 robbed+0.168
Neg logit contributions
 tweet-0.189
 anytime-0.185
 bou-0.179
 forests-0.174
 galaxies-0.167
Token He
Token ID679
Feature activation-3.79e-01

Pos logit contributions
 tweet+0.171
 anytime+0.168
 bou+0.164
 forests+0.161
 galaxies+0.155
Neg logit contributions
 stolen-0.124
 supposed-0.120
illing-0.118
 delayed-0.114
 robbed-0.112
Token had
Token ID550
Feature activation+1.51e+00

Pos logit contributions
 tweet+0.615
 anytime+0.601
 bou+0.586
 forests+0.571
 galaxies+0.548
Neg logit contributions
 stolen-0.567
 supposed-0.539
illing-0.534
 delayed-0.525
 thrown-0.517
Token to
Token ID284
Feature activation+5.13e-01

Pos logit contributions
illing+2.432
omo+2.164
 supposed+2.161
 Timothy+2.153
 stolen+2.144
Neg logit contributions
 forests-2.334
 anytime-2.313
 tweet-2.310
 bou-2.227
 galaxies-2.210
Token decide
Token ID5409
Feature activation-1.45e-01

Pos logit contributions
 stolen+0.874
illing+0.859
 supposed+0.856
 delayed+0.818
 robbed+0.813
Neg logit contributions
 tweet-0.719
 anytime-0.703
 bou-0.680
 forests-0.653
 galaxies-0.622
Token.
Token ID13
Feature activation+7.68e-02

Pos logit contributions
 tweet+0.192
 anytime+0.186
 bou+0.181
 forests+0.175
 galaxies+0.167
Neg logit contributions
 stolen-0.259
 supposed-0.251
illing-0.249
 delayed-0.243
 robbed-0.239
Token He
Token ID679
Feature activation+8.44e-02

Pos logit contributions
 stolen+0.104
 supposed+0.100
illing+0.098
 delayed+0.095
 robbed+0.093
Neg logit contributions
 tweet-0.135
 anytime-0.134
 bou-0.129
 forests-0.127
 galaxies-0.122
Token could
Token ID714
Feature activation-4.42e-02

Pos logit contributions
 stolen+0.117
 supposed+0.114
illing+0.113
 delayed+0.107
 robbed+0.106
Neg logit contributions
 tweet-0.144
 anytime-0.142
 bou-0.137
 forests-0.135
 galaxies-0.130
Token wish
Token ID4601
Feature activation-1.32e+00

Pos logit contributions
 tweet+0.068
 anytime+0.067
 bou+0.065
 forests+0.063
 galaxies+0.061
Neg logit contributions
 stolen-0.069
 supposed-0.067
illing-0.067
 delayed-0.064
 robbed-0.063
Token for
Token ID329
Feature activation-5.53e-01

Pos logit contributions
 tweet+2.075
 bou+2.041
 anytime+1.999
 forests+1.947
 galaxies+1.862
Neg logit contributions
 stolen-1.993
 supposed-1.926
 thrown-1.850
illing-1.846
 delayed-1.836
Token a
Token ID257
Feature activation-8.47e-02

Pos logit contributions
 tweet+0.860
 anytime+0.823
 bou+0.813
 forests+0.783
 galaxies+0.754
Neg logit contributions
 stolen-0.878
 supposed-0.831
illing-0.807
 delayed-0.803
 thrown-0.799
Token puppy
Token ID26188
Feature activation+8.85e-01

Pos logit contributions
 tweet+0.126
 anytime+0.123
 bou+0.119
 forests+0.116
 galaxies+0.111
Neg logit contributions
 stolen-0.137
 supposed-0.133
illing-0.132
 delayed-0.128
 robbed-0.126
Token and
Token ID290
Feature activation-2.52e+00

Pos logit contributions
 tweet+5.007
 anytime+4.749
 forests+4.512
 sings+4.475
 bou+4.380
Neg logit contributions
 stolen-4.397
illing-4.320
 delayed-4.304
 sold-4.194
 thrown-4.184
Token tw
Token ID665
Feature activation-1.71e+00

Pos logit contributions
 tweet+4.031
 anytime+3.953
 bou+3.913
 forests+3.640
 cities+3.505
Neg logit contributions
 stolen-3.829
 thrown-3.794
 delayed-3.779
 burned-3.772
 supposed-3.658
Tokenirled
Token ID49376
Feature activation-2.98e+00

Pos logit contributions
 tweet+2.806
 anytime+2.675
 forests+2.571
 bou+2.565
 cities+2.471
Neg logit contributions
omo-2.548
illing-2.543
 stolen-2.531
 supposed-2.486
 thrown-2.472
Token in
Token ID287
Feature activation-2.00e+00

Pos logit contributions
 tweet+4.692
 anytime+4.315
 sings+4.006
 forests+3.926
 bou+3.908
Neg logit contributions
 stolen-4.601
 Tommy-4.594
 thrown-4.551
 burned-4.477
 delayed-4.435
Token the
Token ID262
Feature activation-1.58e+00

Pos logit contributions
 tweet+3.156
 anytime+2.976
 bou+2.754
 sings+2.739
 galaxies+2.731
Neg logit contributions
 stolen-3.036
 supposed-2.902
 thrown-2.871
 Tommy-2.866
 sale-2.853
Token light
Token ID1657
Feature activation-9.09e-01

Pos logit contributions
 anytime+2.278
 tweet+2.276
 bou+2.066
 forests+1.921
 galaxies+1.871
Neg logit contributions
 stolen-2.661
 burned-2.562
 ruined-2.522
 delayed-2.511
illing-2.504
Token and
Token ID290
Feature activation-2.96e+00

Pos logit contributions
 tweet+1.542
 anytime+1.492
 bou+1.439
 forests+1.412
 galaxies+1.362
Neg logit contributions
 stolen-1.255
illing-1.217
 supposed-1.203
 delayed-1.170
 thrown-1.148
Token smiled
Token ID13541
Feature activation-1.38e+00

Pos logit contributions
 tweet+5.480
 bou+5.380
 anytime+5.327
 forests+4.933
 cities+4.747
Neg logit contributions
 stolen-3.995
 delayed-3.921
 burned-3.913
 thrown-3.888
 supposed-3.820
Token.
Token ID13
Feature activation-3.96e-01

Pos logit contributions
 tweet+2.587
 anytime+2.471
 bou+2.443
 forests+2.417
 sings+2.332
Neg logit contributions
 stolen-1.619
 supposed-1.595
illing-1.588
 delayed-1.569
 unfair-1.481
Token
Token ID198
Feature activation-9.42e-01

Pos logit contributions
 tweet+0.618
 anytime+0.600
 bou+0.586
 forests+0.570
 galaxies+0.548
Neg logit contributions
 stolen-0.604
 supposed-0.590
illing-0.588
 delayed-0.564
 thrown-0.556
Token
Token ID198
Feature activation+9.89e-03

Pos logit contributions
 tweet+1.334
 bou+1.254
 anytime+1.245
 forests+1.193
 galaxies+1.162
Neg logit contributions
 stolen-1.580
illing-1.578
 supposed-1.572
 thrown-1.508
omo-1.499
Token take
Token ID1011
Feature activation+9.57e-02

Pos logit contributions
 tweet+3.289
 bou+3.200
 anytime+3.149
 galaxies+3.106
 forests+3.093
Neg logit contributions
 stolen-2.691
 thrown-2.533
 Tommy-2.521
 delayed-2.495
 sold-2.467
Token turns
Token ID4962
Feature activation+1.15e+00

Pos logit contributions
 stolen+0.182
 supposed+0.178
illing+0.177
 delayed+0.172
 robbed+0.170
Neg logit contributions
 tweet-0.116
 anytime-0.113
 bou-0.108
 forests-0.105
 galaxies-0.099
Token.
Token ID13
Feature activation-2.10e-01

Pos logit contributions
 stolen+1.857
illing+1.763
 supposed+1.754
omo+1.743
 delayed+1.735
Neg logit contributions
 anytime-1.763
 tweet-1.739
 bou-1.642
 forests-1.628
 valleys-1.518
Token They
Token ID1119
Feature activation-1.38e+00

Pos logit contributions
 tweet+0.397
 anytime+0.390
 bou+0.381
 forests+0.373
 galaxies+0.361
Neg logit contributions
 stolen-0.256
 supposed-0.245
illing-0.243
 delayed-0.233
 thrown-0.228
Token have
Token ID423
Feature activation+9.11e-01

Pos logit contributions
 tweet+2.189
 anytime+2.127
 bou+2.101
 forests+2.097
 galaxies+2.060
Neg logit contributions
 stolen-2.075
 supposed-1.972
 thrown-1.944
 delayed-1.944
 disob-1.926
Token fun
Token ID1257
Feature activation-1.27e+00

Pos logit contributions
illing+1.417
 stolen+1.395
 supposed+1.366
 robbed+1.323
 delayed+1.296
Neg logit contributions
 tweet-1.400
 anytime-1.398
 forests-1.365
 bou-1.361
 galaxies-1.271
Token and
Token ID290
Feature activation-1.53e+00

Pos logit contributions
 tweet+2.161
 anytime+2.066
 bou+2.064
 forests+2.007
 galaxies+2.003
Neg logit contributions
 stolen-1.751
 supposed-1.741
illing-1.726
 thrown-1.702
 delayed-1.633
Token laugh
Token ID6487
Feature activation-1.28e+00

Pos logit contributions
 tweet+2.210
 bou+2.181
 anytime+2.149
 forests+2.117
 galaxies+2.034
Neg logit contributions
 stolen-2.546
 supposed-2.469
 thrown-2.422
 delayed-2.376
 disob-2.365
Token together
Token ID1978
Feature activation-1.24e-01

Pos logit contributions
 tweet+2.233
 bou+2.132
 anytime+2.093
 forests+2.089
 galaxies+2.002
Neg logit contributions
 stolen-1.699
 supposed-1.680
illing-1.680
 delayed-1.640
 thrown-1.582
Token.
Token ID13
Feature activation-5.38e-01

Pos logit contributions
 tweet+0.222
 anytime+0.218
 bou+0.213
 forests+0.209
 galaxies+0.201
Neg logit contributions
 stolen-0.163
 supposed-0.157
illing-0.155
 delayed-0.149
 robbed-0.146
Token They
Token ID1119
Feature activation-1.59e+00

Pos logit contributions
 tweet+1.043
 anytime+1.022
 bou+1.012
 forests+0.977
 galaxies+0.948
Neg logit contributions
 stolen-0.623
 supposed-0.603
illing-0.600
 delayed-0.571
 thrown-0.563
Token the
Token ID262
Feature activation-8.89e-01

Pos logit contributions
 tweet+0.751
 anytime+0.728
 bou+0.706
 forests+0.698
 galaxies+0.674
Neg logit contributions
 stolen-0.963
 supposed-0.937
illing-0.919
 delayed-0.896
 thrown-0.891
Token bird
Token ID6512
Feature activation-9.76e-01

Pos logit contributions
 anytime+1.156
 tweet+1.130
 bou+1.077
 forests+1.064
 galaxies+1.039
Neg logit contributions
 stolen-1.634
 supposed-1.550
illing-1.520
 delayed-1.516
 thrown-1.500
Token.
Token ID13
Feature activation-1.69e+00

Pos logit contributions
 tweet+1.447
 anytime+1.411
 forests+1.402
 bou+1.391
 galaxies+1.361
Neg logit contributions
 stolen-1.541
 supposed-1.494
illing-1.479
 delayed-1.423
 thrown-1.417
Token She
Token ID1375
Feature activation-8.51e-01

Pos logit contributions
 tweet+2.990
 bou+2.796
 forests+2.741
 galaxies+2.711
 anytime+2.706
Neg logit contributions
 Tommy-2.242
 stolen-2.231
 supposed-2.230
illing-2.209
omo-2.181
Token learned
Token ID4499
Feature activation-8.25e-01

Pos logit contributions
 tweet+1.493
 anytime+1.438
 bou+1.404
 forests+1.403
 galaxies+1.350
Neg logit contributions
 stolen-1.178
 supposed-1.108
 delayed-1.092
illing-1.088
 thrown-1.084
Token that
Token ID326
Feature activation-1.46e+00

Pos logit contributions
 tweet+1.382
 anytime+1.331
 bou+1.304
 forests+1.288
 valleys+1.237
Neg logit contributions
 stolen-1.215
 supposed-1.185
illing-1.133
 delayed-1.119
 unfair-1.114
Token even
Token ID772
Feature activation-1.62e+00

Pos logit contributions
 tweet+2.240
 bou+2.152
 anytime+2.128
 forests+2.062
 galaxies+2.000
Neg logit contributions
 stolen-2.288
 supposed-2.269
illing-2.172
 delayed-2.150
 thrown-2.136
Token if
Token ID611
Feature activation-6.46e-01

Pos logit contributions
 tweet+2.556
 bou+2.488
 anytime+2.421
 forests+2.345
 galaxies+2.247
Neg logit contributions
 stolen-2.613
 supposed-2.538
 delayed-2.460
illing-2.416
 unfair-2.409
Token someone
Token ID2130
Feature activation+2.59e-01

Pos logit contributions
 tweet+0.885
 anytime+0.861
 bou+0.847
 forests+0.811
 galaxies+0.781
Neg logit contributions
 stolen-1.126
 supposed-1.080
illing-1.054
 delayed-1.054
 thrown-1.032
Token doesn
Token ID1595
Feature activation+8.52e-01

Pos logit contributions
 stolen+0.426
 supposed+0.414
illing+0.414
omo+0.399
 delayed+0.397
Neg logit contributions
 tweet-0.371
 anytime-0.364
 bou-0.347
 forests-0.339
 galaxies-0.324
Token't
Token ID470
Feature activation+1.05e+00

Pos logit contributions
 stolen+0.989
 supposed+0.908
illing+0.884
 robbed+0.875
 delayed+0.853
Neg logit contributions
 tweet-1.713
 anytime-1.697
 bou-1.618
 forests-1.583
 galaxies-1.530
Token They
Token ID1119
Feature activation-1.75e-01

Pos logit contributions
 stolen+1.297
 supposed+1.244
illing+1.201
 delayed+1.187
 robbed+1.187
Neg logit contributions
 tweet-1.026
 anytime-1.021
 bou-0.950
 forests-0.938
 galaxies-0.888
Token are
Token ID389
Feature activation+9.93e-01

Pos logit contributions
 tweet+0.246
 anytime+0.241
 bou+0.233
 forests+0.228
 galaxies+0.217
Neg logit contributions
 stolen-0.295
 supposed-0.285
illing-0.283
 delayed-0.276
 thrown-0.271
Token like
Token ID588
Feature activation+8.25e-01

Pos logit contributions
 stolen+1.534
illing+1.530
 supposed+1.459
 Timothy+1.437
 robbed+1.431
Neg logit contributions
 tweet-1.517
 anytime-1.469
 bou-1.443
 forests-1.398
 sings-1.337
Token us
Token ID514
Feature activation+1.27e+00

Pos logit contributions
 stolen+1.358
illing+1.343
 supposed+1.326
 delayed+1.280
omo+1.273
Neg logit contributions
 tweet-1.216
 anytime-1.170
 bou-1.138
 forests-1.123
 galaxies-1.062
Token.
Token ID13
Feature activation+6.52e-01

Pos logit contributions
 stolen+1.987
illing+1.905
 supposed+1.889
 robbed+1.852
 Timothy+1.830
Neg logit contributions
 anytime-2.047
 tweet-2.030
 bou-1.893
 forests-1.817
 sings-1.751
Token They
Token ID1119
Feature activation-4.37e-01

Pos logit contributions
 stolen+1.181
 supposed+1.134
illing+1.101
 delayed+1.087
 robbed+1.085
Neg logit contributions
 tweet-0.884
 anytime-0.876
 bou-0.814
 forests-0.805
 galaxies-0.758
Token can
Token ID460
Feature activation-5.50e-01

Pos logit contributions
 tweet+0.624
 anytime+0.615
 bou+0.595
 forests+0.584
 galaxies+0.558
Neg logit contributions
 stolen-0.726
 supposed-0.699
illing-0.689
 delayed-0.679
 thrown-0.668
Token wave
Token ID6769
Feature activation-1.74e+00

Pos logit contributions
 tweet+0.856
 anytime+0.844
 bou+0.829
 forests+0.814
 galaxies+0.783
Neg logit contributions
 stolen-0.837
 supposed-0.801
illing-0.792
 delayed-0.780
 thrown-0.769
Token and
Token ID290
Feature activation-1.82e+00

Pos logit contributions
 tweet+2.720
 bou+2.687
 forests+2.673
 anytime+2.648
 galaxies+2.615
Neg logit contributions
 stolen-2.571
 supposed-2.404
 unfair-2.387
illing-2.383
 delayed-2.374
Token talk
Token ID1561
Feature activation-1.32e+00

Pos logit contributions
 anytime+2.663
 bou+2.622
 tweet+2.567
 forests+2.526
 galaxies+2.458
Neg logit contributions
 stolen-2.936
 thrown-2.749
 supposed-2.705
 delayed-2.701
 disob-2.693
Token and
Token ID290
Feature activation-1.39e+00

Pos logit contributions
 tweet+2.430
 bou+2.380
 anytime+2.316
 forests+2.306
 galaxies+2.223
Neg logit contributions
 stolen-1.627
 supposed-1.591
illing-1.547
 delayed-1.528
 unfair-1.486
Token the
Token ID262
Feature activation-1.49e-01

Pos logit contributions
 tweet+0.597
 anytime+0.578
 bou+0.565
 forests+0.551
 galaxies+0.530
Neg logit contributions
 stolen-0.618
 supposed-0.594
illing-0.585
 delayed-0.572
 thrown-0.566
Token buttons
Token ID12163
Feature activation+6.25e-02

Pos logit contributions
 tweet+0.232
 anytime+0.227
 bou+0.220
 forests+0.215
 galaxies+0.206
Neg logit contributions
 stolen-0.232
 supposed-0.224
illing-0.222
 delayed-0.216
 robbed-0.212
Token together
Token ID1978
Feature activation-4.31e-02

Pos logit contributions
 stolen+0.108
 supposed+0.105
illing+0.104
 delayed+0.101
 robbed+0.100
Neg logit contributions
 tweet-0.085
 anytime-0.084
 bou-0.080
 forests-0.078
 galaxies-0.074
Token.
Token ID13
Feature activation-2.06e-02

Pos logit contributions
 tweet+0.074
 anytime+0.073
 bou+0.071
 forests+0.070
 galaxies+0.067
Neg logit contributions
 stolen-0.059
 supposed-0.057
illing-0.056
 delayed-0.054
 robbed-0.053
Token The
Token ID383
Feature activation+4.68e-03

Pos logit contributions
 tweet+0.030
 anytime+0.029
 bou+0.028
 forests+0.027
 galaxies+0.026
Neg logit contributions
 stolen-0.035
 supposed-0.033
illing-0.033
 delayed-0.032
 robbed-0.032
Token little
Token ID1310
Feature activation-5.57e-01

Pos logit contributions
 stolen+0.007
 supposed+0.007
illing+0.007
 delayed+0.007
 thrown+0.007
Neg logit contributions
 tweet-0.007
 anytime-0.007
 bou-0.007
 forests-0.007
 galaxies-0.006
Token girl
Token ID2576
Feature activation-1.01e+00

Pos logit contributions
 tweet+0.599
 anytime+0.580
 bou+0.555
 forests+0.532
 galaxies+0.505
Neg logit contributions
 stolen-1.131
illing-1.088
 supposed-1.082
 delayed-1.074
 robbed-1.053
Token did
Token ID750
Feature activation-1.91e-01

Pos logit contributions
 tweet+1.676
 anytime+1.603
 forests+1.581
 bou+1.580
 galaxies+1.529
Neg logit contributions
 stolen-1.487
 delayed-1.389
 supposed-1.387
illing-1.384
 thrown-1.362
Token as
Token ID355
Feature activation-2.77e-01

Pos logit contributions
 tweet+0.327
 anytime+0.320
 bou+0.313
 forests+0.306
 galaxies+0.295
Neg logit contributions
 stolen-0.268
 supposed-0.258
illing-0.255
 delayed-0.247
 thrown-0.242
Token she
Token ID673
Feature activation-7.20e-01

Pos logit contributions
 tweet+0.381
 anytime+0.369
 bou+0.360
 forests+0.350
 galaxies+0.333
Neg logit contributions
 stolen-0.481
 supposed-0.468
illing-0.463
 delayed-0.451
 thrown-0.444
Token was
Token ID373
Feature activation+6.38e-01

Pos logit contributions
 tweet+1.337
 anytime+1.290
 bou+1.276
 forests+1.250
 galaxies+1.212
Neg logit contributions
 stolen-0.914
 supposed-0.850
illing-0.849
 delayed-0.844
 thrown-0.830
Token pond
Token ID16723
Feature activation+1.16e+00

Pos logit contributions
 stolen+1.061
 supposed+1.051
illing+1.047
 delayed+0.993
omo+0.991
Neg logit contributions
 tweet-0.962
 anytime-0.918
 bou-0.899
 forests-0.897
 galaxies-0.843
Token nearby
Token ID6716
Feature activation+1.03e+00

Pos logit contributions
 stolen+2.003
 supposed+1.979
illing+1.944
omo+1.898
 delayed+1.871
Neg logit contributions
 tweet-1.613
 anytime-1.610
 bou-1.534
 forests-1.490
 galaxies-1.378
Token.
Token ID13
Feature activation-4.58e-01

Pos logit contributions
 stolen+1.596
 supposed+1.561
illing+1.547
omo+1.503
 delayed+1.481
Neg logit contributions
 tweet-1.621
 anytime-1.606
 bou-1.537
 forests-1.519
 galaxies-1.422
Token Sharon
Token ID26918
Feature activation+9.13e-02

Pos logit contributions
 tweet+0.642
 anytime+0.617
 bou+0.608
 forests+0.585
 galaxies+0.566
Neg logit contributions
 stolen-0.778
 supposed-0.755
illing-0.752
 delayed-0.733
 thrown-0.718
Token walked
Token ID6807
Feature activation-1.37e+00

Pos logit contributions
 stolen+0.131
 supposed+0.128
illing+0.125
 delayed+0.120
 robbed+0.119
Neg logit contributions
 tweet-0.152
 anytime-0.148
 bou-0.144
 forests-0.142
 galaxies-0.136
Token closer
Token ID5699
Feature activation-2.71e+00

Pos logit contributions
 tweet+1.802
 anytime+1.672
 bou+1.649
 galaxies+1.581
 forests+1.572
Neg logit contributions
 stolen-2.548
illing-2.449
 supposed-2.426
 delayed-2.418
 thrown-2.352
Token and
Token ID290
Feature activation-2.55e+00

Pos logit contributions
 tweet+4.783
 bou+4.578
 anytime+4.441
 galaxies+4.398
 sings+4.367
Neg logit contributions
 stolen-3.712
illing-3.533
 delayed-3.490
 sale-3.446
 supposed-3.440
Token saw
Token ID2497
Feature activation-2.01e-01

Pos logit contributions
 bou+4.537
 tweet+4.530
 anytime+4.336
 galaxies+4.220
 forests+4.148
Neg logit contributions
 stolen-3.711
 delayed-3.597
 sold-3.365
 robbed-3.357
 disob-3.342
Token a
Token ID257
Feature activation+1.27e-01

Pos logit contributions
 tweet+0.315
 anytime+0.309
 bou+0.300
 forests+0.293
 galaxies+0.281
Neg logit contributions
 stolen-0.309
 supposed-0.298
illing-0.295
 delayed-0.287
 robbed-0.282
Token little
Token ID1310
Feature activation-1.73e-01

Pos logit contributions
 stolen+0.205
 supposed+0.200
illing+0.198
 delayed+0.191
 thrown+0.189
Neg logit contributions
 tweet-0.190
 anytime-0.183
 bou-0.180
 forests-0.174
 galaxies-0.166
Token girl
Token ID2576
Feature activation+8.80e-01

Pos logit contributions
 tweet+0.214
 anytime+0.209
 bou+0.201
 forests+0.195
 galaxies+0.185
Neg logit contributions
 stolen-0.321
 supposed-0.311
illing-0.309
 delayed-0.301
 robbed-0.298
Token Sally
Token ID25737
Feature activation-2.12e-01

Pos logit contributions
 tweet+1.400
 anytime+1.374
 forests+1.348
 bou+1.344
 galaxies+1.321
Neg logit contributions
 stolen-0.727
 supposed-0.703
illing-0.673
 delayed-0.650
 thrown-0.639
Token.
Token ID13
Feature activation+6.95e-02

Pos logit contributions
 tweet+0.386
 anytime+0.378
 bou+0.369
 forests+0.362
 galaxies+0.350
Neg logit contributions
 stolen-0.273
 supposed-0.263
illing-0.259
 delayed-0.250
 thrown-0.245
Token She
Token ID1375
Feature activation+3.18e-01

Pos logit contributions
 stolen+0.094
 supposed+0.090
illing+0.089
 delayed+0.086
 robbed+0.085
Neg logit contributions
 tweet-0.123
 anytime-0.120
 bou-0.118
 forests-0.115
 galaxies-0.110
Token liked
Token ID8288
Feature activation+3.14e-01

Pos logit contributions
 stolen+0.554
illing+0.538
 supposed+0.537
 robbed+0.516
 delayed+0.516
Neg logit contributions
 tweet-0.430
 anytime-0.418
 bou-0.399
 forests-0.394
 galaxies-0.375
Token to
Token ID284
Feature activation+8.25e-01

Pos logit contributions
 stolen+0.500
illing+0.482
 supposed+0.480
 delayed+0.467
 robbed+0.464
Neg logit contributions
 tweet-0.471
 anytime-0.461
 bou-0.449
 forests-0.441
 galaxies-0.417
Token explore
Token ID7301
Feature activation-3.16e+00

Pos logit contributions
 stolen+1.537
 supposed+1.496
illing+1.484
 robbed+1.464
 delayed+1.444
Neg logit contributions
 tweet-1.076
 anytime-1.013
 bou-0.995
 forests-0.942
 sings-0.895
Token the
Token ID262
Feature activation-1.70e+00

Pos logit contributions
 tweet+6.098
 Majesty+5.504
 anytime+5.439
 bou+5.293
 sings+5.038
Neg logit contributions
 stolen-4.104
 Tommy-4.089
 garbage-4.019
 supposed-3.998
 thrown-3.889
Token world
Token ID995
Feature activation-1.86e+00

Pos logit contributions
 tweet+2.393
 anytime+2.253
 bou+2.071
 sings+1.922
 nods+1.899
Neg logit contributions
 stolen-3.087
 supposed-3.019
illing-2.997
 delayed-2.860
 garbage-2.839
Token.
Token ID13
Feature activation-6.28e-01

Pos logit contributions
 tweet+2.773
 bou+2.524
 anytime+2.466
 Majesty+2.330
 galaxies+2.289
Neg logit contributions
 stolen-3.144
 supposed-3.140
illing-3.103
 sold-2.996
 thrown-2.977
Token One
Token ID1881
Feature activation+4.95e-02

Pos logit contributions
 tweet+1.165
 anytime+1.129
 bou+1.106
 forests+1.090
 galaxies+1.070
Neg logit contributions
 stolen-0.778
 supposed-0.757
illing-0.754
 delayed-0.709
omo-0.704
Token day
Token ID1110
Feature activation+9.74e-01

Pos logit contributions
 stolen+0.082
 supposed+0.079
illing+0.079
 delayed+0.077
 robbed+0.076
Neg logit contributions
 tweet-0.072
 anytime-0.070
 bou-0.068
 forests-0.066
 galaxies-0.063
Token One
Token ID1881
Feature activation+7.76e-02

Pos logit contributions
 tweet+2.413
 bou+2.290
 anytime+2.284
 forests+2.239
 galaxies+2.215
Neg logit contributions
 stolen-1.214
 supposed-1.209
illing-1.195
omo-1.143
 Tommy-1.135
Token day
Token ID1110
Feature activation+1.01e+00

Pos logit contributions
 stolen+0.130
 supposed+0.126
illing+0.124
 delayed+0.121
 robbed+0.120
Neg logit contributions
 tweet-0.111
 anytime-0.109
 bou-0.105
 forests-0.103
 galaxies-0.098
Token,
Token ID11
Feature activation+2.41e-01

Pos logit contributions
 stolen+2.199
 supposed+2.162
illing+2.138
 unfair+2.095
 robbed+2.076
Neg logit contributions
 anytime-0.988
 tweet-0.984
 bou-0.867
 forests-0.828
 wherever-0.789
Token Lily
Token ID20037
Feature activation+1.43e-01

Pos logit contributions
 stolen+0.411
 supposed+0.403
illing+0.400
 delayed+0.383
 robbed+0.380
Neg logit contributions
 tweet-0.341
 anytime-0.338
 bou-0.324
 forests-0.312
 galaxies-0.296
Token was
Token ID373
Feature activation+2.81e-01

Pos logit contributions
 stolen+0.216
 supposed+0.210
illing+0.207
 delayed+0.199
 robbed+0.197
Neg logit contributions
 tweet-0.224
 anytime-0.223
 bou-0.215
 forests-0.212
 galaxies-0.202
Token exploring
Token ID13504
Feature activation-3.07e+00

Pos logit contributions
 stolen+0.357
illing+0.339
 supposed+0.336
omo+0.323
 delayed+0.322
Neg logit contributions
 tweet-0.513
 anytime-0.504
 bou-0.488
 forests-0.485
 galaxies-0.463
Token when
Token ID618
Feature activation+3.56e-01

Pos logit contributions
 tweet+6.002
 bou+5.489
 Majesty+5.423
 sings+5.234
 anytime+5.223
Neg logit contributions
 stolen-3.994
 supposed-3.966
illing-3.806
 thrown-3.769
 Tommy-3.747
Token she
Token ID673
Feature activation-9.49e-02

Pos logit contributions
 stolen+0.682
illing+0.667
 supposed+0.664
 delayed+0.640
 robbed+0.636
Neg logit contributions
 tweet-0.420
 anytime-0.419
 forests-0.392
 bou-0.390
 galaxies-0.366
Token found
Token ID1043
Feature activation+1.62e-01

Pos logit contributions
 tweet+0.139
 anytime+0.136
 bou+0.132
 forests+0.129
 galaxies+0.123
Neg logit contributions
 stolen-0.156
 supposed-0.151
illing-0.150
 delayed-0.145
 robbed-0.143
Token a
Token ID257
Feature activation+9.14e-03

Pos logit contributions
 stolen+0.230
illing+0.224
 supposed+0.222
 delayed+0.213
omo+0.210
Neg logit contributions
 tweet-0.272
 anytime-0.267
 bou-0.260
 forests-0.255
 galaxies-0.245
Token big
Token ID1263
Feature activation-3.01e-01

Pos logit contributions
 stolen+0.015
 supposed+0.015
illing+0.015
 delayed+0.014
 robbed+0.014
Neg logit contributions
 tweet-0.013
 anytime-0.013
 bou-0.013
 forests-0.012
 galaxies-0.012
Token
Token ID198
Feature activation-1.80e+00

Pos logit contributions
 tweet+2.644
 bou+2.515
 anytime+2.464
 forests+2.392
 galaxies+2.348
Neg logit contributions
illing-2.271
 stolen-2.264
 supposed-2.256
 Tommy-2.187
 Timothy-2.162
Token
Token ID198
Feature activation-1.80e+00

Pos logit contributions
 tweet+2.932
 bou+2.781
 anytime+2.671
 forests+2.638
 galaxies+2.620
Neg logit contributions
 supposed-2.699
illing-2.686
 stolen-2.622
 thrown-2.556
 Timothy-2.514
TokenGr
Token ID8642
Feature activation-9.85e-01

Pos logit contributions
 tweet+2.933
 bou+2.827
 anytime+2.779
 forests+2.626
 sings+2.556
Neg logit contributions
illing-2.693
 stolen-2.600
 supposed-2.586
 delayed-2.554
 disob-2.533
Tokenace
Token ID558
Feature activation-1.01e-01

Pos logit contributions
 tweet+1.696
 anytime+1.630
 bou+1.619
 forests+1.576
 valleys+1.490
Neg logit contributions
 supposed-1.373
illing-1.363
 stolen-1.348
 delayed-1.267
 robbed-1.249
Token started
Token ID2067
Feature activation+8.04e-01

Pos logit contributions
 tweet+0.161
 anytime+0.158
 bou+0.154
 forests+0.150
 galaxies+0.144
Neg logit contributions
 stolen-0.152
 supposed-0.146
illing-0.145
 delayed-0.140
 robbed-0.138
Token walking
Token ID6155
Feature activation-2.53e+00

Pos logit contributions
 stolen+1.304
illing+1.271
 supposed+1.261
 unfair+1.233
omo+1.232
Neg logit contributions
 tweet-1.168
 anytime-1.168
 bou-1.127
 forests-1.108
 valleys-1.016
Token home
Token ID1363
Feature activation-6.59e-01

Pos logit contributions
 tweet+4.175
 sings+3.908
 bou+3.772
 galaxies+3.621
 forests+3.476
Neg logit contributions
 stolen-3.925
illing-3.720
 delayed-3.684
 Timothy-3.670
 Tommy-3.661
Token,
Token ID11
Feature activation-1.46e+00

Pos logit contributions
 tweet+0.878
 anytime+0.820
 bou+0.817
 forests+0.792
 galaxies+0.763
Neg logit contributions
 stolen-1.187
 supposed-1.158
illing-1.140
 delayed-1.122
 thrown-1.104
Token making
Token ID1642
Feature activation-2.80e-01

Pos logit contributions
 tweet+2.335
 anytime+2.215
 bou+2.201
 forests+2.112
 galaxies+2.100
Neg logit contributions
 stolen-2.312
 supposed-2.191
 delayed-2.161
 thrown-2.138
illing-2.095
Token sure
Token ID1654
Feature activation-2.95e-01

Pos logit contributions
 tweet+0.411
 anytime+0.400
 bou+0.386
 forests+0.378
 galaxies+0.362
Neg logit contributions
 stolen-0.459
 supposed-0.444
illing-0.438
 delayed-0.427
 thrown-0.422
Token not
Token ID407
Feature activation+1.51e+00

Pos logit contributions
 tweet+0.392
 anytime+0.379
 bou+0.369
 forests+0.359
 galaxies+0.341
Neg logit contributions
 stolen-0.526
 supposed-0.509
illing-0.501
 delayed-0.493
 robbed-0.485
Token "
Token ID366
Feature activation-1.17e+00

Pos logit contributions
 tweet+0.122
 anytime+0.119
 bou+0.115
 forests+0.112
 galaxies+0.107
Neg logit contributions
 stolen-0.147
 supposed-0.143
illing-0.141
 delayed-0.137
 robbed-0.136
TokenYes
Token ID5297
Feature activation+2.86e-01

Pos logit contributions
 tweet+1.756
 bou+1.716
 anytime+1.675
 forests+1.662
 galaxies+1.622
Neg logit contributions
 stolen-1.842
illing-1.821
 supposed-1.758
 delayed-1.754
 disob-1.731
Token,
Token ID11
Feature activation-2.25e-01

Pos logit contributions
 stolen+0.569
 supposed+0.557
illing+0.555
 delayed+0.537
 robbed+0.534
Neg logit contributions
 tweet-0.328
 anytime-0.322
 bou-0.304
 forests-0.286
 galaxies-0.267
Token let
Token ID1309
Feature activation+3.37e-01

Pos logit contributions
 tweet+0.317
 anytime+0.310
 bou+0.301
 forests+0.295
 galaxies+0.282
Neg logit contributions
 stolen-0.382
 supposed-0.370
illing-0.366
 delayed-0.357
 thrown-0.351
Token's
Token ID338
Feature activation-1.88e+00

Pos logit contributions
 stolen+0.578
illing+0.564
 supposed+0.560
 delayed+0.541
 robbed+0.537
Neg logit contributions
 tweet-0.475
 anytime-0.468
 bou-0.449
 forests-0.434
 sings-0.406
Token sing
Token ID1702
Feature activation-2.77e+00

Pos logit contributions
 forests+2.301
 bou+2.298
 anytime+2.292
 tweet+2.257
 valleys+2.228
Neg logit contributions
 supposed-3.391
 stolen-3.364
 disob-3.300
 delayed-3.270
 ruined-3.262
Token together
Token ID1978
Feature activation-1.80e+00

Pos logit contributions
 forests+4.912
 valleys+4.864
 tweet+4.838
 bou+4.810
 galaxies+4.701
Neg logit contributions
 supposed-3.391
 sale-3.325
 unfair-3.298
 stolen-3.297
illing-3.247
Token!"
Token ID2474
Feature activation+3.83e-01

Pos logit contributions
 tweet+3.120
 bou+3.017
 forests+2.973
 valleys+2.870
 galaxies+2.869
Neg logit contributions
 supposed-2.376
illing-2.328
 stolen-2.305
 delayed-2.243
 sale-2.232
Token They
Token ID1119
Feature activation-1.79e+00

Pos logit contributions
 stolen+0.674
 supposed+0.655
illing+0.641
 delayed+0.628
 robbed+0.627
Neg logit contributions
 tweet-0.533
 anytime-0.529
 bou-0.504
 forests-0.483
 galaxies-0.451
Token sang
Token ID25889
Feature activation-2.48e+00

Pos logit contributions
 anytime+2.903
 tweet+2.845
 bou+2.704
 forests+2.700
 galaxies+2.560
Neg logit contributions
 stolen-2.659
 delayed-2.574
 thrown-2.513
 ruined-2.501
 fought-2.496
Token and
Token ID290
Feature activation-2.08e+00

Pos logit contributions
 tweet+4.267
 anytime+4.060
 bou+3.988
 forests+3.983
 valleys+3.894
Neg logit contributions
illing-3.235
 supposed-3.190
 sale-3.183
 Tommy-3.173
 stolen-3.162