TOP ACTIVATIONS
MAX = 4.668

. How big do you want it to be ?" her
. " How can you like it ? It is not
, Lily ! Would you like to help me bake a
. One day , Lily wanted to make a special cake
butter in it . She wanted to try it , so
. One day , she wanted to make a special picture
. She was very good at it . Every day she
's mom my showed her a beautiful picture of a flower
bowl . Mom said she needed a measure of milk to
" Why don 't you like them ?" Tommy replied ,
kind of cookie do you want to bake , Lily ?"
hour . What do you want to do while we wait
the kitchen table . Sue wanted the sugar very much ,
?" Lily didn 't know what " we igh " meant
sweet ie . Do you understand ?" L ily
with the laundry . Lucy thought she was very useful when
my . They always used a mixer to mix the ingredients
." Lily didn 't know what " fold " meant ,
" What do you want , Ben ?" Lily asked
the oven . Do you want to do it ?"

MIN ACTIVATIONS
MIN = -4.071

the park . Ben laughed with them . He liked to
right thing and they sw am off together to find some
the p uddle squ ealing with delight . Water spl ashed
cat was terrified and started running away . In
the enclosure . John laughed with excitement and cl apped his
louder and soon everyone started running , screaming , and pan
jet took off , the little boy screamed with joy .
ear . Everyone congratulated the little boy on such an important
her too and started to swim away . The little girl
and waved goodbye to the little girl as he sailed away
goodbye and then they started running back to the park on
together and soon they were running and diving into the pond
Tom and Sam feel sad and angry . They are jealous
't seem to bother the little boy and his dad .
like hours , and the little girl got awfully hot .
ily and Max feel sad and angry . They do not
and the ambulance people left . Anna and Ben went back
the bug started to fight . They shouted and chased each
glad to see the cops . They are sorry to be
sad . They feel angry and sad . They have no

INTERVAL 2.483 - 4.668
CONTAINS 0.719%

cart ! I 'll always remember to take good care of
?" asked Lily . " Sure , sweet
big and strong . I want to try it ."
a toy that she really wanted , but her mom said
cup cakes . She put lots of tasty ingredients into a

INTERVAL 0.299 - 2.483
CONTAINS 36.742%

. Sue told her mom , " I want to swing
by one with determination . After he finished all
. The fairy waved her wand and the bunny grew back
in the next town . When she arrived ,
photo albums . " These are pictures of when I was

INTERVAL -1.886 - 0.299
CONTAINS 60.255%

gave the gift to Sam , Sam was very happy too
was there . He was all alone . He started to
say anything either . Tom was scared . Tom
, so she threw it away . From that
saw lots of y ummy treats ! The fly couldn 't

INTERVAL -4.071 - -1.886
CONTAINS 2.284%

!" he shouted . Ben and Lily were very
the monster and they screamed too . The parents ran to
was the only one who felt so scared . Every night
. He caught them and grabbed the com bs from their
<|endoftext|> Tom and Sam are friends . They like to play
Token.
Token ID13
Feature activation+6.94e-01

Pos logit contributions
 goalie+0.987
irms+0.970
 villains+0.907
awks+0.898
 thrill+0.868
Neg logit contributions
pron-1.371
 recipe-1.366
ons-1.287
 mixer-1.197
 baking-1.150
Token How
Token ID1374
Feature activation+1.89e+00

Pos logit contributions
 recipe+1.628
pron+1.623
ons+1.548
 mixer+1.432
 baking+1.384
Neg logit contributions
 goalie-0.851
irms-0.838
 villains-0.777
awks-0.748
ashes-0.714
Token big
Token ID1263
Feature activation+1.25e+00

Pos logit contributions
 recipe+4.127
pron+4.087
ons+3.872
 mixer+3.605
 baking+3.439
Neg logit contributions
 goalie-2.660
irms-2.510
 villains-2.449
awks-2.305
 thrill-2.294
Token do
Token ID466
Feature activation+2.00e+00

Pos logit contributions
pron+2.464
 recipe+2.447
ons+2.281
innie+2.118
 mixer+2.087
Neg logit contributions
 goalie-1.913
irms-1.867
 villains-1.742
awks-1.734
 thrill-1.706
Token you
Token ID345
Feature activation+1.69e+00

Pos logit contributions
pron+4.010
 recipe+4.002
 mixer+3.523
ons+3.490
 baking+3.284
Neg logit contributions
 villains-3.200
 goalie-2.926
irms-2.915
 villain-2.761
gers-2.653
Token want
Token ID765
Feature activation+4.67e+00

Pos logit contributions
pron+3.673
 recipe+3.661
ons+3.367
 mixer+3.182
innie+3.022
Neg logit contributions
 goalie-2.332
irms-2.254
 villains-2.114
awks-2.088
ashes-2.083
Token it
Token ID340
Feature activation+3.12e+00

Pos logit contributions
pron+11.364
 recipe+10.265
ants+9.484
ama+9.362
aman+9.294
Neg logit contributions
 like-7.176
irms-6.298
 He-6.199
 villains-6.066
 goalie-5.762
Token to
Token ID284
Feature activation+2.46e+00

Pos logit contributions
 recipe+6.062
pron+6.039
ons+5.436
innie+5.373
 mixer+5.318
Neg logit contributions
 goalie-5.099
 like-4.875
irms-4.830
ashes-4.708
vious-4.564
Token be
Token ID307
Feature activation+1.89e+00

Pos logit contributions
pron+5.168
 recipe+5.143
ons+4.782
 mixer+4.513
innie+4.338
Neg logit contributions
 goalie-3.617
irms-3.341
 villains-3.265
 thrill-3.208
awks-3.132
Token?"
Token ID1701
Feature activation+1.93e+00

Pos logit contributions
 recipe+3.916
pron+3.898
ons+3.703
 mixer+3.525
innie+3.266
Neg logit contributions
 goalie-2.720
irms-2.702
 villains-2.512
awks-2.465
 victorious-2.334
Token her
Token ID607
Feature activation+4.30e-01

Pos logit contributions
 recipe+3.648
pron+3.476
ons+3.058
 mixer+2.861
 baking+2.729
Neg logit contributions
irms-3.412
 goalie-3.264
 villains-3.174
awks-3.136
 thrill-3.031
Token.
Token ID13
Feature activation-6.58e-02

Pos logit contributions
pron+1.097
 recipe+1.075
ons+1.004
 mixer+0.934
innie+0.882
Neg logit contributions
 goalie-0.876
irms-0.853
awks-0.794
 villains-0.790
ashes-0.763
Token "
Token ID366
Feature activation-5.34e-01

Pos logit contributions
 goalie+0.077
irms+0.076
 villains+0.070
awks+0.068
 thrill+0.065
Neg logit contributions
pron-0.158
 recipe-0.156
ons-0.148
 mixer-0.139
 baking-0.135
TokenHow
Token ID2437
Feature activation+8.10e-01

Pos logit contributions
 goalie+0.877
irms+0.867
 villains+0.819
awks+0.803
 thrill+0.775
Neg logit contributions
pron-1.039
 recipe-1.018
ons-0.952
 mixer-0.886
 baking-0.848
Token can
Token ID460
Feature activation+2.49e+00

Pos logit contributions
pron+1.711
 recipe+1.708
ons+1.598
 mixer+1.497
 baking+1.435
Neg logit contributions
 goalie-1.173
irms-1.137
 villains-1.075
awks-1.034
 thrill-1.008
Token you
Token ID345
Feature activation+2.21e+00

Pos logit contributions
pron+5.578
 recipe+5.365
ons+4.750
innie+4.576
 mixer+4.538
Neg logit contributions
 goalie-3.494
 villains-3.374
irms-3.262
awks-3.085
ashes-2.977
Token like
Token ID588
Feature activation+4.44e+00

Pos logit contributions
pron+5.324
 recipe+5.171
ons+5.039
 mixer+4.664
innie+4.561
Neg logit contributions
 goalie-2.743
irms-2.524
awks-2.442
ashes-2.402
 thrill-2.301
Token it
Token ID340
Feature activation+2.71e+00

Pos logit contributions
pron+12.010
 recipe+10.567
innie+10.356
ama+9.785
lf+9.779
Neg logit contributions
 like-5.979
awks-5.619
 goalie-5.537
irms-5.510
ashes-5.221
Token?
Token ID30
Feature activation+1.50e+00

Pos logit contributions
pron+5.130
 recipe+5.045
ons+4.952
innie+4.524
 mixer+4.368
Neg logit contributions
 goalie-4.845
irms-4.456
awks-4.252
ashes-4.124
 like-4.042
Token It
Token ID632
Feature activation+1.37e+00

Pos logit contributions
pron+3.590
 recipe+3.541
ons+3.357
 mixer+3.042
lf+2.983
Neg logit contributions
 goalie-1.903
irms-1.813
awks-1.643
 villains-1.612
ashes-1.604
Token is
Token ID318
Feature activation+1.48e+00

Pos logit contributions
pron+2.803
 recipe+2.671
ons+2.486
innie+2.300
 mixer+2.287
Neg logit contributions
 goalie-2.121
irms-2.072
awks-1.996
 villains-1.949
ashes-1.923
Token not
Token ID407
Feature activation+1.45e+00

Pos logit contributions
pron+3.056
 recipe+3.000
ons+2.974
 mixer+2.655
innie+2.608
Neg logit contributions
 goalie-2.195
irms-2.136
awks-2.008
 villains-1.932
 thrill-1.908
Token,
Token ID11
Feature activation+2.44e+00

Pos logit contributions
 recipe+2.805
pron+2.788
ons+2.558
 mixer+2.282
 baking+2.191
Neg logit contributions
 goalie-2.569
irms-2.513
awks-2.300
cer-2.289
 villains-2.248
Token Lily
Token ID20037
Feature activation-9.10e-01

Pos logit contributions
 recipe+5.060
pron+4.899
ons+4.852
ants+4.124
 mixer+4.107
Neg logit contributions
 goalie-3.540
irms-3.430
 villain-3.293
 villains-3.282
 thrill-3.265
Token!
Token ID0
Feature activation+7.74e-01

Pos logit contributions
 goalie+1.616
irms+1.594
 villains+1.528
awks+1.502
ashes+1.473
Neg logit contributions
pron-1.653
 recipe-1.621
ons-1.530
 mixer-1.400
 baking-1.335
Token Would
Token ID10928
Feature activation+3.01e+00

Pos logit contributions
 recipe+1.722
pron+1.708
ons+1.624
 mixer+1.483
 baking+1.426
Neg logit contributions
 goalie-1.054
irms-1.019
 villains-0.961
awks-0.919
ashes-0.883
Token you
Token ID345
Feature activation+7.26e-01

Pos logit contributions
pron+7.436
 recipe+7.414
 mixer+6.299
ons+6.060
itten+5.848
Neg logit contributions
 villains-3.552
irms-3.352
 goalie-3.296
Goal-3.148
 victorious-3.147
Token like
Token ID588
Feature activation+4.44e+00

Pos logit contributions
 recipe+2.275
pron+2.266
ons+2.185
 mixer+2.094
 baking+2.006
Neg logit contributions
 goalie-0.306
irms-0.288
awks-0.199
 villains-0.193
ashes-0.181
Token to
Token ID284
Feature activation+3.02e+00

Pos logit contributions
pron+11.017
 recipe+10.074
ama+9.995
innie+9.772
ants+9.719
Neg logit contributions
 like-6.233
 pirates-5.631
irms-5.486
 pirate-5.198
 sailors-5.189
Token help
Token ID1037
Feature activation+3.73e+00

Pos logit contributions
pron+6.795
 recipe+6.681
ons+6.274
innie+5.920
 mixer+5.885
Neg logit contributions
 goalie-4.064
ashes-3.606
irms-3.605
 like-3.594
 thrill-3.490
Token me
Token ID502
Feature activation+3.32e+00

Pos logit contributions
 recipe+7.527
pron+7.518
 mixer+6.557
ons+6.532
itten+6.294
Neg logit contributions
 goalie-5.816
irms-5.709
 like-5.547
 villain-5.311
 villains-5.289
Token bake
Token ID28450
Feature activation+3.22e+00

Pos logit contributions
pron+6.975
 recipe+6.819
ons+6.529
innie+6.113
 mixer+5.971
Neg logit contributions
 goalie-5.004
irms-4.710
 like-4.424
awks-4.423
ashes-4.402
Token a
Token ID257
Feature activation+3.46e+00

Pos logit contributions
pron+7.409
 recipe+7.386
ons+6.668
innie+6.608
itten+6.405
Neg logit contributions
irms-4.200
 goalie-3.898
awks-3.585
 like-3.499
 pirate-3.321
Token.
Token ID13
Feature activation+1.92e+00

Pos logit contributions
pron+4.948
 recipe+4.822
ons+4.672
 mixer+4.320
innie+4.256
Neg logit contributions
 goalie-1.617
irms-1.484
cer-1.353
 villains-1.263
ashes-1.251
Token One
Token ID1881
Feature activation+2.88e+00

Pos logit contributions
pron+5.112
 recipe+5.078
ons+5.054
 mixer+4.436
iflower+4.398
Neg logit contributions
irms-1.740
 goalie-1.731
ashes-1.477
awks-1.409
 villains-1.403
Token day
Token ID1110
Feature activation+1.98e+00

Pos logit contributions
 recipe+6.751
ons+6.668
pron+6.647
ants+5.788
innie+5.768
Neg logit contributions
 goalie-3.118
irms-3.061
ashes-3.045
 thrill-2.967
cer-2.759
Token,
Token ID11
Feature activation+1.40e+00

Pos logit contributions
 recipe+3.993
pron+3.926
ons+3.418
 mixer+3.182
innie+3.032
Neg logit contributions
 goalie-3.219
irms-3.143
cer-2.958
 villains-2.911
ashes-2.905
Token Lily
Token ID20037
Feature activation+2.12e+00

Pos logit contributions
 recipe+3.127
pron+3.010
ons+2.964
 mixer+2.629
innie+2.607
Neg logit contributions
 goalie-1.829
irms-1.767
 villains-1.614
 thrill-1.611
cer-1.568
Token wanted
Token ID2227
Feature activation+4.40e+00

Pos logit contributions
pron+4.624
 recipe+4.567
ons+4.093
 mixer+3.853
innie+3.827
Neg logit contributions
 goalie-2.862
irms-2.797
cer-2.699
ashes-2.635
 villains-2.617
Token to
Token ID284
Feature activation+2.94e+00

Pos logit contributions
pron+10.614
itten+9.204
isy+8.912
ama+8.897
lf+8.845
Neg logit contributions
 pirates-5.976
 like-5.768
 goalie-5.698
ashes-5.486
irms-5.435
Token make
Token ID787
Feature activation+1.86e+00

Pos logit contributions
pron+6.287
 recipe+5.848
ons+5.512
innie+5.227
itten+5.118
Neg logit contributions
 goalie-4.394
ashes-4.097
irms-3.970
 villains-3.922
 want-3.836
Token a
Token ID257
Feature activation+1.90e+00

Pos logit contributions
pron+4.029
 recipe+3.809
ons+3.545
 mixer+3.250
innie+3.221
Neg logit contributions
irms-2.610
 goalie-2.535
awks-2.404
ashes-2.281
 villains-2.263
Token special
Token ID2041
Feature activation+2.76e+00

Pos logit contributions
ons+2.762
pron+2.691
 recipe+2.576
innie+2.243
 brushing+2.112
Neg logit contributions
 goalie-3.823
irms-3.692
awks-3.606
ashes-3.555
 thrill-3.528
Token cake
Token ID12187
Feature activation+1.54e+00

Pos logit contributions
ons+3.972
pron+3.771
ants+3.367
 recipe+3.318
innie+3.242
Neg logit contributions
 goalie-5.517
irms-5.297
ashes-5.254
 pirate-5.198
 thrill-5.176
Token butter
Token ID9215
Feature activation+1.40e+00

Pos logit contributions
pron+4.048
 recipe+3.914
ons+3.647
 mixer+3.344
innie+3.321
Neg logit contributions
 goalie-2.157
irms-2.150
 villains-1.980
 thrill-1.924
awks-1.922
Token in
Token ID287
Feature activation+1.85e+00

Pos logit contributions
pron+3.074
 recipe+3.036
ons+2.805
 mixer+2.663
innie+2.646
Neg logit contributions
 goalie-1.882
irms-1.789
cer-1.709
awks-1.640
ashes-1.630
Token it
Token ID340
Feature activation+1.78e+00

Pos logit contributions
pron+3.714
 recipe+3.607
ons+3.291
 mixer+3.041
innie+2.963
Neg logit contributions
irms-2.812
 goalie-2.759
 thrill-2.637
awks-2.584
cer-2.580
Token.
Token ID13
Feature activation+1.98e+00

Pos logit contributions
pron+4.710
 recipe+4.669
ons+4.357
 mixer+4.144
innie+4.127
Neg logit contributions
irms-1.507
 goalie-1.465
cer-1.296
awks-1.282
ashes-1.244
Token She
Token ID1375
Feature activation+2.48e+00

Pos logit contributions
pron+4.839
 recipe+4.733
ons+4.542
 mixer+4.103
 baking+3.990
Neg logit contributions
irms-2.319
 goalie-2.284
 He-2.136
 villains-2.080
ashes-1.996
Token wanted
Token ID2227
Feature activation+4.40e+00

Pos logit contributions
 recipe+6.938
pron+6.665
ons+6.353
 mixer+6.100
 baking+6.029
Neg logit contributions
 goalie-1.954
 villains-1.939
irms-1.774
 He-1.595
cer-1.593
Token to
Token ID284
Feature activation+2.53e+00

Pos logit contributions
pron+11.707
 recipe+10.608
innie+10.563
lf+10.533
ama+10.088
Neg logit contributions
 like-5.436
irms-4.757
 goalie-4.676
 pirates-4.633
 hope-4.610
Token try
Token ID1949
Feature activation+4.08e+00

Pos logit contributions
pron+5.618
 recipe+5.600
ons+5.109
 mixer+4.846
innie+4.829
Neg logit contributions
 goalie-3.379
irms-3.216
ashes-3.180
 thrill-3.025
awks-3.013
Token it
Token ID340
Feature activation+2.00e+00

Pos logit contributions
pron+10.339
 recipe+10.140
innie+8.947
omm+8.930
ons+8.718
Neg logit contributions
irms-4.609
 goalie-4.461
ks-4.420
awks-4.264
 like-4.204
Token,
Token ID11
Feature activation+1.69e+00

Pos logit contributions
pron+5.114
 recipe+4.907
ons+4.522
 mixer+4.319
innie+4.130
Neg logit contributions
 goalie-2.240
irms-2.184
ashes-2.063
awks-1.990
cer-1.921
Token so
Token ID523
Feature activation+1.74e+00

Pos logit contributions
pron+4.064
 recipe+4.003
ons+3.812
innie+3.541
 mixer+3.482
Neg logit contributions
 goalie-2.059
irms-2.016
awks-1.918
 villains-1.828
 thrill-1.779
Token.
Token ID13
Feature activation+2.11e+00

Pos logit contributions
pron+5.611
 recipe+5.455
ons+5.151
 mixer+4.908
innie+4.862
Neg logit contributions
 goalie-2.018
irms-1.932
cer-1.795
ashes-1.795
ks-1.736
Token One
Token ID1881
Feature activation+2.24e+00

Pos logit contributions
pron+5.159
 recipe+5.146
ons+4.961
 mixer+4.347
 baking+4.329
Neg logit contributions
irms-2.616
 goalie-2.555
awks-2.258
ashes-2.203
 He-2.178
Token day
Token ID1110
Feature activation+2.02e+00

Pos logit contributions
 recipe+5.131
pron+4.898
ons+4.816
 mixer+4.314
 baking+4.162
Neg logit contributions
 goalie-2.813
irms-2.728
 thrill-2.625
cer-2.596
 villains-2.519
Token,
Token ID11
Feature activation+1.03e+00

Pos logit contributions
 recipe+4.144
pron+4.072
ons+3.472
 mixer+3.296
innie+3.072
Neg logit contributions
 goalie-3.227
irms-3.217
 villains-2.977
cer-2.976
ashes-2.930
Token she
Token ID673
Feature activation+1.12e+00

Pos logit contributions
 recipe+1.796
pron+1.729
ons+1.636
 mixer+1.453
innie+1.369
Neg logit contributions
 goalie-1.871
irms-1.852
 villains-1.738
 thrill-1.707
awks-1.699
Token wanted
Token ID2227
Feature activation+4.39e+00

Pos logit contributions
 recipe+2.381
pron+2.337
ons+2.127
 mixer+2.032
 baking+1.947
Neg logit contributions
 goalie-1.597
irms-1.562
 villains-1.478
cer-1.455
awks-1.412
Token to
Token ID284
Feature activation+2.28e+00

Pos logit contributions
pron+10.395
 recipe+9.189
lf+8.737
ama+8.688
quet+8.613
Neg logit contributions
 goalie-6.312
irms-5.959
 pirates-5.913
 like-5.777
 villains-5.554
Token make
Token ID787
Feature activation+1.43e+00

Pos logit contributions
pron+4.688
 recipe+4.606
ons+4.121
 mixer+3.842
innie+3.776
Neg logit contributions
 goalie-3.495
 villains-3.240
irms-3.227
 thrill-3.043
ashes-3.009
Token a
Token ID257
Feature activation+1.18e+00

Pos logit contributions
pron+3.117
 recipe+3.029
ons+2.766
 mixer+2.603
innie+2.484
Neg logit contributions
irms-1.988
 goalie-1.964
awks-1.816
 villains-1.773
ashes-1.691
Token special
Token ID2041
Feature activation+2.23e+00

Pos logit contributions
pron+1.972
 recipe+1.956
ons+1.944
 mixer+1.648
innie+1.604
Neg logit contributions
 goalie-2.148
irms-2.049
awks-1.961
 villains-1.961
 thrill-1.931
Token picture
Token ID4286
Feature activation+1.34e+00

Pos logit contributions
ons+3.695
pron+3.590
 recipe+3.442
ants+3.104
innie+3.046
Neg logit contributions
 goalie-4.055
irms-3.783
 pirate-3.707
 thrill-3.680
cer-3.635
Token.
Token ID13
Feature activation+6.36e-01

Pos logit contributions
pron+3.582
 recipe+3.424
ons+3.176
 mixer+3.007
innie+2.972
Neg logit contributions
irms-1.986
 goalie-1.932
awks-1.784
ashes-1.772
cer-1.701
Token She
Token ID1375
Feature activation+3.20e+00

Pos logit contributions
pron+1.342
 recipe+1.305
ons+1.237
 mixer+1.130
 baking+1.094
Neg logit contributions
 goalie-0.943
irms-0.930
 villains-0.857
awks-0.854
ashes-0.840
Token was
Token ID373
Feature activation+2.00e+00

Pos logit contributions
 recipe+7.684
pron+7.499
ons+7.140
ama+6.607
 mixer+6.502
Neg logit contributions
 villains-3.231
 hope-3.220
 goalie-3.155
ks-3.149
mer-3.059
Token very
Token ID845
Feature activation+1.62e+00

Pos logit contributions
pron+4.499
 recipe+4.483
ons+4.440
 mixer+3.854
innie+3.811
Neg logit contributions
 goalie-2.448
irms-2.361
 thrill-2.141
awks-2.117
ashes-2.081
Token good
Token ID922
Feature activation-6.65e-01

Pos logit contributions
 recipe+3.915
pron+3.907
ons+3.881
 mixer+3.454
innie+3.330
Neg logit contributions
irms-1.728
 goalie-1.697
ashes-1.606
 thrill-1.521
awks-1.517
Token at
Token ID379
Feature activation+4.38e+00

Pos logit contributions
 goalie+1.006
irms+1.002
 villains+0.943
awks+0.931
 thrill+0.896
Neg logit contributions
 recipe-1.378
pron-1.375
ons-1.278
 mixer-1.207
 baking-1.169
Token it
Token ID340
Feature activation+1.42e+00

Pos logit contributions
pron+9.639
itten+8.964
ama+8.473
ants+8.383
ittens+8.366
Neg logit contributions
 hope-6.397
 goalie-6.025
 hopes-5.914
 pirates-5.887
irms-5.786
Token.
Token ID13
Feature activation+7.84e-01

Pos logit contributions
pron+3.361
 recipe+3.226
ons+3.111
 mixer+2.923
innie+2.775
Neg logit contributions
 goalie-1.713
irms-1.594
 villains-1.527
cer-1.456
ashes-1.446
Token Every
Token ID3887
Feature activation+2.00e+00

Pos logit contributions
pron+1.707
 recipe+1.665
ons+1.560
 mixer+1.430
 baking+1.392
Neg logit contributions
 goalie-1.113
irms-1.102
awks-1.003
 villains-1.001
ashes-0.981
Token day
Token ID1110
Feature activation+1.27e+00

Pos logit contributions
pron+4.230
 recipe+4.155
ons+4.121
ants+3.638
innie+3.614
Neg logit contributions
 goalie-2.694
irms-2.574
 thrill-2.406
ashes-2.406
 villains-2.388
Token she
Token ID673
Feature activation+2.24e+00

Pos logit contributions
pron+2.205
 recipe+2.113
ons+1.891
 mixer+1.721
innie+1.644
Neg logit contributions
 goalie-2.356
irms-2.278
 thrill-2.195
 villains-2.187
cer-2.147
Token's
Token ID338
Feature activation+2.39e+00

Pos logit contributions
 recipe+4.300
pron+4.268
ons+3.733
 mixer+3.534
innie+3.524
Neg logit contributions
 goalie-2.750
irms-2.696
cer-2.557
ashes-2.522
awks-2.450
Token mom
Token ID1995
Feature activation+3.15e+00

Pos logit contributions
pron+4.544
 recipe+4.501
ons+4.488
innie+3.690
ants+3.609
Neg logit contributions
 goalie-3.991
irms-3.650
 thrill-3.521
ashes-3.507
 teammate-3.387
Tokenmy
Token ID1820
Feature activation+3.30e+00

Pos logit contributions
 recipe+6.724
pron+6.475
ons+5.847
ants+5.564
 mixer+5.563
Neg logit contributions
 goalie-4.275
cer-4.129
irms-4.117
ny-4.062
 like-4.006
Token showed
Token ID3751
Feature activation+1.14e+00

Pos logit contributions
 recipe+7.645
pron+7.392
ons+6.617
 mixer+6.290
innie+6.259
Neg logit contributions
 goalie-3.928
 want-3.753
irms-3.732
cer-3.727
 like-3.697
Token her
Token ID607
Feature activation+1.85e+00

Pos logit contributions
pron+1.912
 recipe+1.800
ons+1.586
 mixer+1.473
 baking+1.421
Neg logit contributions
irms-2.258
 goalie-2.169
awks-2.114
ashes-2.069
 villains-2.060
Token a
Token ID257
Feature activation+4.33e+00

Pos logit contributions
pron+3.904
 recipe+3.751
ons+3.488
innie+3.172
 mixer+3.109
Neg logit contributions
 goalie-2.633
irms-2.581
 villains-2.333
awks-2.301
ashes-2.284
Token beautiful
Token ID4950
Feature activation+3.36e+00

Pos logit contributions
ons+10.888
ants+10.273
ridge+9.837
innie+9.725
 recipe+9.571
Neg logit contributions
 goalie-4.776
 thrill-4.408
 pirate-4.347
 boy-4.060
 lonely-3.985
Token picture
Token ID4286
Feature activation+2.09e+00

Pos logit contributions
ons+5.841
 recipe+5.604
ants+5.525
pron+5.438
innie+5.261
Neg logit contributions
 thrill-5.567
 goalie-5.465
 pirate-5.420
cer-5.067
irms-5.041
Token of
Token ID286
Feature activation+8.31e-01

Pos logit contributions
pron+5.001
 recipe+4.837
ons+4.425
innie+4.349
 mixer+4.257
Neg logit contributions
 goalie-2.338
 villains-2.204
irms-2.141
ashes-2.084
ks-1.987
Token a
Token ID257
Feature activation+1.72e+00

Pos logit contributions
pron+1.818
 recipe+1.793
ons+1.673
 mixer+1.557
innie+1.482
Neg logit contributions
 goalie-1.142
irms-1.104
 villains-1.061
awks-1.009
 thrill-0.992
Token flower
Token ID15061
Feature activation+1.32e+00

Pos logit contributions
ons+3.195
 recipe+3.181
pron+3.126
innie+2.725
ants+2.647
Neg logit contributions
 goalie-2.847
 villains-2.646
irms-2.543
 pirate-2.486
 thrill-2.466
Token bowl
Token ID9396
Feature activation+1.06e+00

Pos logit contributions
pron+3.410
 recipe+3.395
ons+3.215
innie+2.949
ants+2.779
Neg logit contributions
irms-4.864
 goalie-4.807
 thrill-4.618
awks-4.531
ashes-4.431
Token.
Token ID13
Feature activation+1.18e+00

Pos logit contributions
pron+2.624
 recipe+2.549
ons+2.466
 mixer+2.279
innie+2.257
Neg logit contributions
 goalie-1.161
irms-1.093
ashes-1.012
cer-0.971
 villains-0.953
Token Mom
Token ID11254
Feature activation+3.23e+00

Pos logit contributions
pron+2.651
 recipe+2.642
ons+2.506
 mixer+2.256
 baking+2.214
Neg logit contributions
irms-1.626
 goalie-1.590
awks-1.404
ashes-1.402
 villains-1.362
Token said
Token ID531
Feature activation+3.32e+00

Pos logit contributions
 recipe+7.466
pron+7.115
ons+6.614
 baking+6.091
 canvas+6.052
Neg logit contributions
 goalie-3.798
 cheered-3.794
ks-3.664
cer-3.605
ny-3.601
Token she
Token ID673
Feature activation+1.95e+00

Pos logit contributions
pron+7.229
 recipe+7.028
ons+6.633
innie+6.177
iflower+5.837
Neg logit contributions
 goalie-4.753
 villains-4.492
irms-4.440
 pirate-4.369
ks-4.310
Token needed
Token ID2622
Feature activation+4.31e+00

Pos logit contributions
 recipe+4.157
pron+4.125
ons+3.849
 mixer+3.487
 baking+3.409
Neg logit contributions
 goalie-2.679
irms-2.602
 villains-2.577
ks-2.482
ashes-2.421
Token a
Token ID257
Feature activation+3.41e+00

Pos logit contributions
pron+11.202
lf+10.440
 recipe+10.032
ons+9.897
innie+9.741
Neg logit contributions
 hope-5.080
 like-4.810
irms-4.623
ashes-4.359
ks-4.275
Token measure
Token ID3953
Feature activation+2.55e+00

Pos logit contributions
ons+7.839
ants+6.644
lf+6.538
 recipe+6.534
pron+6.375
Neg logit contributions
 goalie-4.973
 thrill-4.736
ashes-4.445
 hope-4.430
irms-4.417
Token of
Token ID286
Feature activation+2.94e+00

Pos logit contributions
pron+6.021
 recipe+5.597
ons+5.460
innie+5.031
lf+4.851
Neg logit contributions
 goalie-3.068
irms-2.952
ashes-2.819
ks-2.808
cer-2.764
Token milk
Token ID7545
Feature activation+2.66e+00

Pos logit contributions
pron+6.409
 recipe+5.858
ons+5.821
innie+5.361
ama+5.204
Neg logit contributions
irms-4.265
 goalie-4.048
 thrill-4.041
 hope-3.946
awks-3.873
Token to
Token ID284
Feature activation+2.95e+00

Pos logit contributions
pron+6.681
ons+6.385
 recipe+6.255
innie+5.847
lf+5.561
Neg logit contributions
 goalie-2.857
ashes-2.674
cer-2.580
irms-2.554
 like-2.397
Token "
Token ID366
Feature activation+7.91e-01

Pos logit contributions
 recipe+3.044
ons+3.038
pron+2.954
 mixer+2.362
innie+2.339
Neg logit contributions
 goalie-3.724
irms-3.511
 thrill-3.306
 villains-3.301
 victorious-3.296
TokenWhy
Token ID5195
Feature activation+1.06e+00

Pos logit contributions
 recipe+1.729
pron+1.715
ons+1.661
 mixer+1.504
 baking+1.474
Neg logit contributions
 goalie-1.099
irms-1.066
 villains-0.974
awks-0.966
cer-0.951
Token don
Token ID836
Feature activation+1.76e+00

Pos logit contributions
 recipe+2.266
pron+2.264
ons+2.090
 mixer+1.948
 baking+1.896
Neg logit contributions
 goalie-1.549
irms-1.490
 villains-1.402
cer-1.332
awks-1.324
Token't
Token ID470
Feature activation+1.22e+00

Pos logit contributions
 recipe+2.511
pron+2.482
ons+1.948
 mixer+1.842
 baking+1.747
Neg logit contributions
 goalie-3.819
 villains-3.697
irms-3.684
 villain-3.498
 victorious-3.477
Token you
Token ID345
Feature activation+1.81e+00

Pos logit contributions
pron+2.315
 recipe+2.231
ons+2.055
 mixer+1.871
innie+1.818
Neg logit contributions
 goalie-2.114
irms-2.014
 villains-1.954
awks-1.867
cer-1.865
Token like
Token ID588
Feature activation+4.31e+00

Pos logit contributions
pron+4.266
 recipe+4.197
ons+3.996
 mixer+3.677
innie+3.597
Neg logit contributions
 goalie-2.356
irms-2.227
awks-2.049
ashes-2.029
 thrill-1.989
Token them
Token ID606
Feature activation+2.94e+00

Pos logit contributions
pron+11.042
innie+10.081
 recipe+9.804
ons+9.801
ama+9.759
Neg logit contributions
irms-5.529
 like-5.427
 goalie-5.167
awks-4.999
ashes-4.998
Token?"
Token ID1701
Feature activation+8.68e-02

Pos logit contributions
pron+5.772
 recipe+5.750
ons+5.560
innie+5.094
 mixer+4.770
Neg logit contributions
 goalie-5.073
irms-4.739
awks-4.399
ashes-4.362
 like-4.324
Token Tommy
Token ID19919
Feature activation+8.52e-01

Pos logit contributions
pron+0.192
 recipe+0.191
ons+0.179
 mixer+0.168
 baking+0.162
Neg logit contributions
 goalie-0.118
irms-0.117
 villains-0.108
awks-0.106
 thrill-0.102
Token replied
Token ID8712
Feature activation+1.05e+00

Pos logit contributions
 recipe+1.578
pron+1.574
ons+1.482
 mixer+1.330
 baking+1.300
Neg logit contributions
 goalie-1.448
irms-1.423
 villains-1.325
awks-1.322
cer-1.280
Token,
Token ID11
Feature activation+1.75e+00

Pos logit contributions
 recipe+1.810
pron+1.797
ons+1.661
 mixer+1.472
innie+1.404
Neg logit contributions
 goalie-1.925
irms-1.884
 villains-1.786
 thrill-1.735
awks-1.726
Token kind
Token ID1611
Feature activation+2.24e+00

Pos logit contributions
pron+4.809
 recipe+4.612
ons+4.296
 mixer+4.001
innie+3.967
Neg logit contributions
 goalie-3.361
 villains-3.063
irms-3.004
cer-2.821
ks-2.712
Token of
Token ID286
Feature activation+3.26e+00

Pos logit contributions
pron+4.696
 recipe+4.499
ons+4.192
innie+4.019
 mixer+3.867
Neg logit contributions
 goalie-3.369
irms-3.104
awks-2.909
 villains-2.905
 thrill-2.876
Token cookie
Token ID19751
Feature activation+3.06e+00

Pos logit contributions
pron+5.174
ants+4.399
ons+4.368
 recipe+4.363
innie+3.994
Neg logit contributions
irms-6.671
 goalie-6.516
awks-6.320
 thrill-6.308
ashes-6.281
Token do
Token ID466
Feature activation+2.42e+00

Pos logit contributions
pron+6.108
 recipe+5.852
ons+5.521
innie+5.507
ants+5.100
Neg logit contributions
 goalie-4.841
irms-4.576
ashes-4.275
 villain-4.266
ger-4.225
Token you
Token ID345
Feature activation+2.62e+00

Pos logit contributions
pron+4.829
 recipe+4.590
ons+4.363
 mixer+4.235
innie+4.015
Neg logit contributions
 villains-3.808
 goalie-3.720
irms-3.599
gers-3.382
ashes-3.377
Token want
Token ID765
Feature activation+4.29e+00

Pos logit contributions
pron+8.370
 recipe+8.042
ons+7.993
innie+7.462
lf+7.426
Neg logit contributions
 goalie-1.140
irms-1.081
ashes-1.040
 want-0.973
awks-0.939
Token to
Token ID284
Feature activation+3.56e+00

Pos logit contributions
pron+9.942
 recipe+8.885
itten+8.614
innie+8.570
ants+8.518
Neg logit contributions
 like-6.749
irms-5.978
 goalie-5.896
 want-5.653
ashes-5.478
Token bake
Token ID28450
Feature activation+2.94e+00

Pos logit contributions
pron+8.476
 recipe+8.011
ons+7.882
itten+7.272
innie+7.212
Neg logit contributions
 like-4.633
 goalie-4.497
 want-4.288
irms-4.275
ashes-4.065
Token,
Token ID11
Feature activation+1.69e+00

Pos logit contributions
pron+6.057
 recipe+5.901
ons+5.417
innie+5.343
itten+5.136
Neg logit contributions
irms-4.580
 goalie-4.513
awks-4.117
ashes-4.022
 like-3.960
Token Lily
Token ID20037
Feature activation+1.96e-01

Pos logit contributions
 recipe+3.462
ons+3.446
pron+3.426
 mixer+2.988
innie+2.950
Neg logit contributions
 goalie-2.535
irms-2.439
 villains-2.324
awks-2.218
 thrill-2.190
Token?"
Token ID1701
Feature activation+1.84e+00

Pos logit contributions
pron+0.266
 recipe+0.263
ons+0.239
 mixer+0.214
 baking+0.201
Neg logit contributions
 goalie-0.428
irms-0.423
 villains-0.405
awks-0.400
cer-0.391
Token hour
Token ID1711
Feature activation+2.14e+00

Pos logit contributions
pron+3.996
ons+3.776
 recipe+3.758
 mixer+3.112
 brushing+3.081
Neg logit contributions
 goalie-3.698
irms-3.697
 thrill-3.583
awks-3.448
ashes-3.402
Token.
Token ID13
Feature activation+2.10e-01

Pos logit contributions
pron+5.376
 recipe+5.127
ons+4.900
 mixer+4.596
innie+4.595
Neg logit contributions
 goalie-2.437
irms-2.220
ashes-2.148
cer-2.032
ks-2.031
Token What
Token ID1867
Feature activation+2.06e+00

Pos logit contributions
pron+0.484
 recipe+0.481
ons+0.455
 mixer+0.426
 baking+0.410
Neg logit contributions
 goalie-0.264
irms-0.260
 villains-0.241
awks-0.235
ashes-0.223
Token do
Token ID466
Feature activation+1.30e+00

Pos logit contributions
pron+4.694
 recipe+4.495
ons+4.254
 mixer+4.027
innie+3.924
Neg logit contributions
 goalie-2.923
 villains-2.668
irms-2.619
ks-2.411
 goals-2.355
Token you
Token ID345
Feature activation+7.93e-01

Pos logit contributions
pron+2.478
 recipe+2.406
ons+2.191
 mixer+2.095
 baking+1.946
Neg logit contributions
 goalie-2.145
irms-2.095
 villains-2.090
awks-1.908
cer-1.897
Token want
Token ID765
Feature activation+4.27e+00

Pos logit contributions
pron+2.264
 recipe+2.233
ons+2.149
 mixer+2.036
innie+1.952
Neg logit contributions
 goalie-0.567
irms-0.547
awks-0.462
 villains-0.459
ashes-0.436
Token to
Token ID284
Feature activation+2.46e+00

Pos logit contributions
pron+9.670
 recipe+8.417
itten+8.407
innie+8.249
ons+8.026
Neg logit contributions
 like-6.654
 goalie-6.174
irms-5.912
 pirates-5.687
ashes-5.633
Token do
Token ID466
Feature activation+2.20e+00

Pos logit contributions
pron+4.991
 recipe+4.855
ons+4.742
innie+4.226
 mixer+4.179
Neg logit contributions
 goalie-3.835
irms-3.554
 thrill-3.483
ashes-3.316
awks-3.287
Token while
Token ID981
Feature activation+2.12e+00

Pos logit contributions
pron+3.966
 recipe+3.842
ons+3.723
 mixer+3.369
itten+3.327
Neg logit contributions
 goalie-3.798
irms-3.635
 villains-3.489
cer-3.351
ashes-3.338
Token we
Token ID356
Feature activation+1.15e+00

Pos logit contributions
pron+4.066
 recipe+3.975
ons+3.715
innie+3.393
 mixer+3.344
Neg logit contributions
 goalie-3.605
irms-3.371
 villains-3.260
 sailors-3.095
awks-3.082
Token wait
Token ID4043
Feature activation+1.66e+00

Pos logit contributions
pron+2.709
 recipe+2.675
ons+2.554
 mixer+2.353
innie+2.298
Neg logit contributions
 goalie-1.405
irms-1.357
ashes-1.236
awks-1.222
 villains-1.213
Token the
Token ID262
Feature activation+2.87e+00

Pos logit contributions
pron+1.891
 recipe+1.843
ons+1.722
 mixer+1.578
 baking+1.512
Neg logit contributions
 goalie-1.417
irms-1.382
 villains-1.266
awks-1.266
 thrill-1.244
Token kitchen
Token ID9592
Feature activation+1.35e+00

Pos logit contributions
pron+5.086
ons+5.050
 recipe+4.709
innie+4.229
ants+4.043
Neg logit contributions
 goalie-5.293
irms-4.825
 villains-4.811
ashes-4.763
 seeker-4.538
Token table
Token ID3084
Feature activation+6.97e-01

Pos logit contributions
pron+2.699
 recipe+2.504
ons+2.503
innie+2.260
 mixer+2.211
Neg logit contributions
 goalie-2.126
irms-2.072
ashes-1.945
awks-1.921
 villains-1.919
Token.
Token ID13
Feature activation+1.52e+00

Pos logit contributions
pron+1.691
 recipe+1.650
ons+1.574
 mixer+1.483
innie+1.444
Neg logit contributions
 goalie-0.791
irms-0.764
 villains-0.696
awks-0.676
ashes-0.671
Token Sue
Token ID26089
Feature activation+2.38e+00

Pos logit contributions
pron+3.705
 recipe+3.587
ons+3.295
 mixer+3.084
 baking+2.992
Neg logit contributions
irms-1.942
 goalie-1.899
awks-1.787
ashes-1.752
 villains-1.630
Token wanted
Token ID2227
Feature activation+4.27e+00

Pos logit contributions
pron+6.555
 recipe+6.493
ons+5.867
 mixer+5.658
 baking+5.612
Neg logit contributions
 goalie-2.064
irms-2.058
 villains-1.889
ks-1.808
awks-1.762
Token the
Token ID262
Feature activation+3.43e+00

Pos logit contributions
pron+10.943
 recipe+9.627
lf+9.280
ants+8.720
innie+8.586
Neg logit contributions
 like-6.024
irms-5.723
 hope-5.486
ashes-5.398
 He-5.377
Token sugar
Token ID7543
Feature activation+2.01e+00

Pos logit contributions
pron+6.585
ons+6.567
 recipe+6.180
ants+6.024
lf+5.891
Neg logit contributions
irms-5.505
 thrill-5.400
ashes-5.343
 goalie-5.252
 like-5.064
Token very
Token ID845
Feature activation+3.62e-01

Pos logit contributions
pron+4.560
 recipe+4.248
ons+4.107
innie+3.811
 mixer+3.674
Neg logit contributions
 goalie-2.736
irms-2.644
ashes-2.507
cer-2.416
awks-2.364
Token much
Token ID881
Feature activation+1.11e-01

Pos logit contributions
pron+0.765
 recipe+0.757
ons+0.709
 mixer+0.662
 baking+0.640
Neg logit contributions
 goalie-0.525
irms-0.516
 villains-0.486
awks-0.472
ashes-0.462
Token,
Token ID11
Feature activation+1.66e+00

Pos logit contributions
pron+0.219
 recipe+0.214
ons+0.201
 mixer+0.186
 baking+0.179
Neg logit contributions
 goalie-0.177
irms-0.176
 villains-0.164
awks-0.162
 thrill-0.157
Token?"
Token ID1701
Feature activation+6.78e-01

Pos logit contributions
pron+4.439
 recipe+4.407
ons+3.851
 mixer+3.796
innie+3.596
Neg logit contributions
 goalie-3.148
irms-3.116
 villains-2.850
awks-2.846
cer-2.825
Token Lily
Token ID20037
Feature activation+1.83e+00

Pos logit contributions
pron+1.471
 recipe+1.464
ons+1.356
 mixer+1.276
 baking+1.229
Neg logit contributions
irms-0.941
 goalie-0.924
awks-0.858
 villains-0.848
ashes-0.836
Token didn
Token ID1422
Feature activation+1.68e+00

Pos logit contributions
 recipe+3.995
pron+3.969
ons+3.707
 baking+3.464
 mixer+3.461
Neg logit contributions
 goalie-2.476
irms-2.424
cer-2.227
 villains-2.196
awks-2.106
Token't
Token ID470
Feature activation-1.17e+00

Pos logit contributions
pron+2.355
 recipe+2.311
ons+2.001
 mixer+1.754
 baking+1.621
Neg logit contributions
 goalie-3.743
irms-3.574
 villains-3.505
 victorious-3.412
awks-3.356
Token know
Token ID760
Feature activation+2.61e+00

Pos logit contributions
 goalie+1.635
irms+1.632
awks+1.496
 villains+1.485
cer+1.440
Neg logit contributions
pron-2.533
 recipe-2.525
ons-2.337
 mixer-2.201
 baking-2.136
Token what
Token ID644
Feature activation+4.26e+00

Pos logit contributions
pron+6.091
 recipe+5.831
 mixer+4.791
ons+4.737
 baking+4.515
Neg logit contributions
irms-3.650
ks-3.447
 goalie-3.443
 thrill-3.396
awks-3.303
Token "
Token ID366
Feature activation+2.02e+00

Pos logit contributions
pron+10.352
 recipe+9.923
ants+9.099
ons+9.071
cakes+8.733
Neg logit contributions
 villains-5.649
 hope-5.622
ks-4.983
 like-4.979
 hopes-4.961
Tokenwe
Token ID732
Feature activation+1.68e+00

Pos logit contributions
 recipe+3.935
pron+3.854
ons+3.469
 mixer+3.366
 baking+3.265
Neg logit contributions
 goalie-3.395
irms-3.220
cer-3.141
 villains-3.091
awks-3.075
Tokenigh
Token ID394
Feature activation+2.53e+00

Pos logit contributions
pron+3.926
 recipe+3.921
ons+3.459
 mixer+3.348
 baking+3.261
Neg logit contributions
 goalie-2.128
irms-2.080
 villains-1.927
awks-1.924
ashes-1.856
Token"
Token ID1
Feature activation+1.61e+00

Pos logit contributions
pron+5.280
 recipe+5.123
 mixer+4.401
ons+4.374
innie+4.218
Neg logit contributions
 goalie-4.208
irms-3.747
cer-3.744
 villains-3.697
ashes-3.605
Token meant
Token ID4001
Feature activation+1.92e+00

Pos logit contributions
pron+3.434
 recipe+3.379
ons+3.125
 mixer+2.983
 baking+2.864
Neg logit contributions
 goalie-2.341
irms-2.307
 villains-2.158
awks-2.151
ashes-2.105
Token sweet
Token ID6029
Feature activation+2.52e-01

Pos logit contributions
pron+2.639
 recipe+2.620
ons+2.561
 mixer+2.277
innie+2.244
Neg logit contributions
 goalie-1.826
irms-1.756
 villains-1.696
awks-1.630
 thrill-1.608
Tokenie
Token ID494
Feature activation+2.19e-01

Pos logit contributions
pron+0.362
 recipe+0.352
ons+0.322
 mixer+0.288
 baking+0.271
Neg logit contributions
 goalie-0.533
irms-0.530
 villains-0.507
awks-0.503
 thrill-0.490
Token.
Token ID13
Feature activation+4.55e-01

Pos logit contributions
pron+0.483
 recipe+0.477
ons+0.449
 mixer+0.422
 baking+0.404
Neg logit contributions
 goalie-0.294
irms-0.288
 villains-0.272
awks-0.264
ashes-0.254
Token Do
Token ID2141
Feature activation+1.50e+00

Pos logit contributions
pron+1.044
 recipe+1.035
ons+0.985
 mixer+0.910
 baking+0.877
Neg logit contributions
 goalie-0.587
irms-0.573
 villains-0.537
awks-0.522
ashes-0.495
Token you
Token ID345
Feature activation-1.00e+00

Pos logit contributions
pron+3.000
 recipe+3.000
ons+2.611
 mixer+2.580
 baking+2.420
Neg logit contributions
 goalie-2.435
 villains-2.386
irms-2.336
awks-2.217
 thrill-2.131
Token understand
Token ID1833
Feature activation+4.25e+00

Pos logit contributions
 goalie+0.983
irms+0.974
 villains+0.874
awks+0.835
cer+0.821
Neg logit contributions
pron-2.610
 recipe-2.575
ons-2.453
 mixer-2.310
 baking-2.277
Token?"
Token ID1701
Feature activation+1.20e+00

Pos logit contributions
pron+9.736
 recipe+9.401
ons+8.814
innie+8.456
gary+8.381
Neg logit contributions
 goalie-6.459
 like-6.228
irms-5.783
 sailors-5.690
 villain-5.589
Token
Token ID198
Feature activation+6.14e-01

Pos logit contributions
 recipe+2.707
pron+2.692
ons+2.477
 mixer+2.348
 baking+2.268
Neg logit contributions
irms-1.583
 goalie-1.513
awks-1.430
 villains-1.395
ashes-1.376
Token
Token ID198
Feature activation+9.12e-01

Pos logit contributions
pron+1.303
 recipe+1.299
ons+1.192
 mixer+1.138
 baking+1.094
Neg logit contributions
irms-0.864
 goalie-0.858
 villains-0.803
awks-0.788
ashes-0.756
TokenL
Token ID43
Feature activation-1.65e-01

Pos logit contributions
 recipe+2.025
pron+1.985
ons+1.891
 mixer+1.790
 baking+1.725
Neg logit contributions
 goalie-1.232
irms-1.224
 villains-1.149
awks-1.100
ashes-1.079
Tokenily
Token ID813
Feature activation+9.47e-01

Pos logit contributions
 goalie+0.216
irms+0.214
 villains+0.198
awks+0.194
 thrill+0.186
Neg logit contributions
pron-0.375
 recipe-0.371
ons-0.350
 mixer-0.329
 baking-0.318
Token with
Token ID351
Feature activation+2.73e+00

Pos logit contributions
pron+7.540
 recipe+7.330
ons+6.946
innie+6.573
 mixer+6.308
Neg logit contributions
 like-3.877
 goalie-3.824
cer-3.639
ashes-3.624
irms-3.622
Token the
Token ID262
Feature activation+1.97e+00

Pos logit contributions
pron+6.030
 recipe+5.986
ons+5.461
innie+5.065
itten+4.987
Neg logit contributions
irms-3.965
 goalie-3.516
 hope-3.400
 thrill-3.364
awks-3.328
Token laundry
Token ID25724
Feature activation+1.79e+00

Pos logit contributions
pron+3.516
ons+3.369
 recipe+3.248
innie+2.908
lf+2.830
Neg logit contributions
 goalie-3.477
irms-3.434
 thrill-3.283
 villains-3.232
awks-3.231
Token.
Token ID13
Feature activation+2.21e+00

Pos logit contributions
pron+4.517
 recipe+4.355
ons+4.263
innie+4.028
 mixer+3.879
Neg logit contributions
 goalie-1.687
irms-1.641
ashes-1.532
 villains-1.436
cer-1.418
Token Lucy
Token ID22162
Feature activation+2.23e+00

Pos logit contributions
pron+5.665
 recipe+5.552
ons+5.351
 mixer+4.703
iflower+4.648
Neg logit contributions
irms-2.453
 goalie-2.318
ashes-2.074
awks-2.038
 thrill-1.991
Token thought
Token ID1807
Feature activation+4.24e+00

Pos logit contributions
 recipe+5.459
pron+5.384
ons+4.948
 mixer+4.633
 baking+4.492
Neg logit contributions
 goalie-2.449
irms-2.415
 villains-2.316
cer-2.219
ks-2.205
Token she
Token ID673
Feature activation+7.12e-01

Pos logit contributions
pron+10.723
 recipe+10.514
ons+9.208
innie+9.166
itten+8.960
Neg logit contributions
 like-4.593
irms-4.466
 villains-4.436
 goals-4.336
 hope-4.334
Token was
Token ID373
Feature activation+1.59e-02

Pos logit contributions
pron+1.306
 recipe+1.303
ons+1.191
 mixer+1.103
 baking+1.062
Neg logit contributions
 goalie-1.213
irms-1.197
 villains-1.160
awks-1.102
cer-1.091
Token very
Token ID845
Feature activation-8.81e-02

Pos logit contributions
pron+0.032
 recipe+0.032
ons+0.030
 mixer+0.028
 baking+0.026
Neg logit contributions
 goalie-0.024
irms-0.024
 villains-0.023
awks-0.022
 thrill-0.021
Token useful
Token ID4465
Feature activation-5.15e-01

Pos logit contributions
 goalie+0.132
irms+0.131
 villains+0.123
awks+0.120
 thrill+0.116
Neg logit contributions
pron-0.182
 recipe-0.180
ons-0.168
 mixer-0.157
 baking-0.151
Token when
Token ID618
Feature activation+1.95e+00

Pos logit contributions
 goalie+0.831
irms+0.822
 villains+0.768
awks+0.768
 thrill+0.742
Neg logit contributions
 recipe-1.013
pron-1.008
ons-0.940
 mixer-0.878
 baking-0.845
Tokenmy
Token ID1820
Feature activation+2.03e+00

Pos logit contributions
pron+4.760
 recipe+4.596
ons+4.374
 mixer+4.018
innie+3.979
Neg logit contributions
 goalie-2.453
irms-2.345
cer-2.194
ashes-2.060
awks-2.022
Token.
Token ID13
Feature activation+1.77e+00

Pos logit contributions
pron+5.248
 recipe+5.107
ons+5.034
 mixer+4.599
innie+4.519
Neg logit contributions
 goalie-1.951
irms-1.714
cer-1.619
 villains-1.462
ashes-1.456
Token They
Token ID1119
Feature activation+3.09e+00

Pos logit contributions
pron+4.385
 recipe+4.346
ons+4.338
 mixer+3.769
iflower+3.679
Neg logit contributions
 goalie-1.923
irms-1.913
ashes-1.697
awks-1.619
 villains-1.572
Token always
Token ID1464
Feature activation+2.00e+00

Pos logit contributions
 recipe+6.965
pron+6.927
ons+6.277
ama+5.813
itten+5.748
Neg logit contributions
 goalie-3.931
 hope-3.828
 like-3.708
irms-3.668
 villains-3.624
Token used
Token ID973
Feature activation+2.90e+00

Pos logit contributions
 recipe+4.365
pron+4.337
ons+4.067
 mixer+3.743
innie+3.580
Neg logit contributions
irms-2.676
 goalie-2.560
 villains-2.394
awks-2.358
ks-2.357
Token a
Token ID257
Feature activation+4.24e+00

Pos logit contributions
pron+6.801
 recipe+6.227
ons+5.814
innie+5.717
ama+5.705
Neg logit contributions
irms-3.673
 goalie-3.399
ashes-3.383
awks-3.379
ks-3.340
Token mixer
Token ID33938
Feature activation+1.67e+00

Pos logit contributions
ons+9.748
innie+9.077
 recipe+8.642
lf+8.626
ants+8.616
Neg logit contributions
 thrill-5.299
ashes-5.246
 goalie-5.063
irms-4.822
 pirate-4.704
Token to
Token ID284
Feature activation+3.05e+00

Pos logit contributions
pron+2.696
 recipe+2.416
ons+2.251
 mixer+1.993
innie+1.944
Neg logit contributions
 goalie-3.357
ashes-3.253
irms-3.211
 villains-3.112
cer-3.079
Token mix
Token ID5022
Feature activation+2.36e+00

Pos logit contributions
pron+5.683
 recipe+5.666
ons+5.395
ama+4.911
innie+4.871
Neg logit contributions
irms-4.870
ashes-4.798
 thrill-4.787
 goalie-4.778
 like-4.654
Token the
Token ID262
Feature activation+2.51e+00

Pos logit contributions
pron+5.142
 recipe+4.734
ons+4.297
innie+4.100
lf+3.980
Neg logit contributions
irms-3.418
ashes-3.355
 goalie-3.226
awks-3.118
ks-3.047
Token ingredients
Token ID9391
Feature activation+1.86e+00

Pos logit contributions
pron+3.648
ons+3.249
 recipe+3.139
innie+2.915
lf+2.774
Neg logit contributions
ashes-5.198
 goalie-5.131
irms-5.111
 thrill-4.972
awks-4.916
Token."
Token ID526
Feature activation+4.60e-01

Pos logit contributions
pron+1.836
 recipe+1.828
ons+1.730
 mixer+1.596
innie+1.507
Neg logit contributions
 goalie-1.328
irms-1.318
awks-1.215
 villains-1.204
ashes-1.163
Token Lily
Token ID20037
Feature activation+1.36e+00

Pos logit contributions
pron+0.988
 recipe+0.979
ons+0.918
 mixer+0.852
 baking+0.819
Neg logit contributions
irms-0.659
 goalie-0.645
awks-0.603
 villains-0.597
ashes-0.582
Token didn
Token ID1422
Feature activation+1.64e+00

Pos logit contributions
 recipe+2.932
pron+2.861
ons+2.692
 mixer+2.515
 baking+2.482
Neg logit contributions
 goalie-1.898
irms-1.879
 villains-1.717
cer-1.680
awks-1.669
Token't
Token ID470
Feature activation-1.64e+00

Pos logit contributions
pron+2.298
 recipe+2.255
ons+1.935
 mixer+1.720
 baking+1.597
Neg logit contributions
 goalie-3.643
irms-3.522
 villains-3.430
 victorious-3.297
awks-3.285
Token know
Token ID760
Feature activation+2.18e+00

Pos logit contributions
 goalie+2.248
irms+2.222
awks+2.069
 villains+2.027
cer+1.982
Neg logit contributions
pron-3.668
 recipe-3.641
ons-3.330
 mixer-3.170
 baking-3.104
Token what
Token ID644
Feature activation+4.23e+00

Pos logit contributions
pron+4.974
 recipe+4.795
ons+4.038
 mixer+3.928
 baking+3.728
Neg logit contributions
irms-3.128
 goalie-2.971
 villains-2.825
 thrill-2.806
ks-2.790
Token "
Token ID366
Feature activation+1.89e+00

Pos logit contributions
pron+9.508
 recipe+9.434
ons+8.717
ants+8.367
quet+8.301
Neg logit contributions
 villains-6.274
 hope-5.699
 like-5.654
irms-5.464
ks-5.340
Tokenfold
Token ID11379
Feature activation+1.90e+00

Pos logit contributions
 recipe+3.693
pron+3.619
ons+3.315
 mixer+3.179
 baking+3.072
Neg logit contributions
 goalie-3.137
irms-3.023
cer-2.900
awks-2.883
ashes-2.830
Token"
Token ID1
Feature activation+1.22e+00

Pos logit contributions
pron+3.470
 recipe+3.461
ons+3.048
 mixer+2.874
 baking+2.799
Neg logit contributions
 goalie-3.353
irms-3.214
cer-3.087
 villains-3.022
ashes-2.978
Token meant
Token ID4001
Feature activation+1.54e+00

Pos logit contributions
pron+2.503
 recipe+2.484
ons+2.313
 mixer+2.178
 baking+2.097
Neg logit contributions
irms-1.811
 goalie-1.810
 villains-1.678
awks-1.666
ashes-1.614
Token,
Token ID11
Feature activation+1.15e+00

Pos logit contributions
pron+3.236
 recipe+3.073
ons+3.013
 mixer+2.663
innie+2.552
Neg logit contributions
 goalie-2.254
irms-2.163
 villains-2.085
 thrill-1.974
cer-1.941
Token
Token ID198
Feature activation+1.03e+00

Pos logit contributions
 recipe+2.347
pron+2.340
ons+2.180
 mixer+2.062
 baking+1.989
Neg logit contributions
 goalie-1.413
irms-1.394
 villains-1.289
ashes-1.253
awks-1.234
Token"
Token ID1
Feature activation+2.01e+00

Pos logit contributions
 recipe+1.839
pron+1.792
ons+1.708
 mixer+1.564
 baking+1.479
Neg logit contributions
 goalie-1.836
irms-1.790
 villains-1.677
cer-1.656
ashes-1.650
TokenWhat
Token ID2061
Feature activation+2.06e+00

Pos logit contributions
 recipe+4.452
ons+4.440
pron+4.389
 mixer+3.876
 baking+3.778
Neg logit contributions
 goalie-2.715
irms-2.589
cer-2.373
 thrill-2.281
Goal-2.276
Token do
Token ID466
Feature activation+1.69e+00

Pos logit contributions
pron+4.492
 recipe+4.263
ons+3.979
 mixer+3.695
innie+3.659
Neg logit contributions
 goalie-3.118
irms-2.858
 villains-2.819
cer-2.677
ks-2.615
Token you
Token ID345
Feature activation+6.81e-01

Pos logit contributions
pron+3.268
 recipe+3.145
ons+2.907
 mixer+2.745
innie+2.559
Neg logit contributions
 villains-2.770
 goalie-2.755
irms-2.679
awks-2.447
ashes-2.418
Token want
Token ID765
Feature activation+4.22e+00

Pos logit contributions
pron+2.110
 recipe+2.081
ons+2.015
 mixer+1.905
 baking+1.832
Neg logit contributions
 goalie-0.332
irms-0.327
awks-0.246
 villains-0.226
 thrill-0.213
Token,
Token ID11
Feature activation+2.81e-01

Pos logit contributions
pron+9.794
 recipe+8.292
innie+8.270
ants+8.014
itten+7.985
Neg logit contributions
 like-6.385
 goalie-5.782
ger-5.629
irms-5.609
 He-5.596
Token Ben
Token ID3932
Feature activation+8.37e-01

Pos logit contributions
pron+0.519
 recipe+0.514
ons+0.489
 mixer+0.439
 baking+0.422
Neg logit contributions
 goalie-0.482
irms-0.474
 villains-0.443
awks-0.438
 thrill-0.428
Token?"
Token ID1701
Feature activation+3.46e-01

Pos logit contributions
pron+1.230
 recipe+1.211
ons+1.132
 mixer+1.015
 baking+0.955
Neg logit contributions
 goalie-1.728
irms-1.701
 villains-1.599
cer-1.595
awks-1.591
Token Lily
Token ID20037
Feature activation+1.49e+00

Pos logit contributions
pron+0.688
 recipe+0.686
ons+0.647
 mixer+0.591
 baking+0.570
Neg logit contributions
 goalie-0.537
irms-0.536
 villains-0.498
awks-0.491
 thrill-0.476
Token asked
Token ID1965
Feature activation+5.22e-01

Pos logit contributions
pron+3.080
 recipe+3.031
ons+2.889
 mixer+2.609
 baking+2.571
Neg logit contributions
 goalie-2.246
irms-2.225
awks-2.015
 villains-2.010
cer-1.947
Token the
Token ID262
Feature activation+1.17e+00

Pos logit contributions
pron+2.602
 recipe+2.547
ons+2.467
 mixer+2.213
innie+2.149
Neg logit contributions
 goalie-1.803
irms-1.740
 thrill-1.588
cer-1.578
awks-1.572
Token oven
Token ID14361
Feature activation+1.43e+00

Pos logit contributions
pron+1.321
 recipe+1.271
ons+1.257
innie+0.971
 mixer+0.903
Neg logit contributions
irms-2.762
 goalie-2.761
awks-2.624
 thrill-2.617
ashes-2.609
Token.
Token ID13
Feature activation-2.75e-01

Pos logit contributions
pron+3.517
 recipe+3.371
ons+3.336
innie+3.073
 mixer+3.041
Neg logit contributions
 goalie-1.591
irms-1.557
ashes-1.448
awks-1.377
 villains-1.306
Token Do
Token ID2141
Feature activation+1.50e+00

Pos logit contributions
 goalie+0.347
irms+0.342
 villains+0.318
awks+0.311
 thrill+0.297
Neg logit contributions
pron-0.632
 recipe-0.626
ons-0.589
 mixer-0.557
 baking-0.539
Token you
Token ID345
Feature activation-3.60e-01

Pos logit contributions
pron+3.003
 recipe+2.979
ons+2.653
 mixer+2.539
 baking+2.369
Neg logit contributions
 goalie-2.479
irms-2.345
 villains-2.325
awks-2.152
 thrill-2.092
Token want
Token ID765
Feature activation+4.22e+00

Pos logit contributions
 goalie+0.269
irms+0.263
 villains+0.232
awks+0.221
 thrill+0.200
Neg logit contributions
pron-1.017
 recipe-1.007
ons-0.961
 mixer-0.916
 baking-0.895
Token to
Token ID284
Feature activation+2.09e+00

Pos logit contributions
pron+10.558
 recipe+9.521
itten+9.257
innie+9.132
lf+9.004
Neg logit contributions
 like-5.215
 goalie-5.020
irms-4.954
 pirates-4.779
ashes-4.760
Token do
Token ID466
Feature activation+2.49e+00

Pos logit contributions
pron+4.919
 recipe+4.815
ons+4.708
innie+4.204
 mixer+4.143
Neg logit contributions
 goalie-2.649
irms-2.437
ashes-2.409
 thrill-2.300
awks-2.256
Token it
Token ID340
Feature activation+2.31e+00

Pos logit contributions
pron+4.903
 recipe+4.900
ons+4.485
innie+4.177
itten+4.102
Neg logit contributions
 goalie-3.759
irms-3.737
 villains-3.574
 pirate-3.467
ashes-3.403
Token?"
Token ID1701
Feature activation-7.27e-01

Pos logit contributions
pron+4.517
 recipe+4.480
ons+4.424
 mixer+3.981
innie+3.938
Neg logit contributions
 goalie-3.791
irms-3.474
ashes-3.241
cer-3.158
 victorious-3.144
Token
Token ID198
Feature activation-7.06e-01

Pos logit contributions
 goalie+1.249
irms+1.234
 villains+1.169
awks+1.154
 thrill+1.131
Neg logit contributions
pron-1.364
 recipe-1.346
ons-1.253
 mixer-1.171
 baking-1.112
Token the
Token ID262
Feature activation-1.04e-01

Pos logit contributions
irms+2.288
 goalie+2.256
 villains+2.149
awks+2.143
 thrill+2.085
Neg logit contributions
 recipe-2.085
pron-2.014
ons-1.877
 baking-1.770
 mixer-1.770
Token park
Token ID3952
Feature activation-1.36e+00

Pos logit contributions
 goalie+0.181
irms+0.180
 villains+0.170
awks+0.167
 thrill+0.162
Neg logit contributions
pron-0.191
 recipe-0.188
ons-0.175
 mixer-0.162
 baking-0.155
Token.
Token ID13
Feature activation-1.92e+00

Pos logit contributions
irms+2.224
 goalie+2.137
awks+2.031
 villains+2.003
 thrill+1.949
Neg logit contributions
 recipe-2.722
pron-2.619
ons-2.456
 mixer-2.316
 baking-2.310
Token Ben
Token ID3932
Feature activation-3.21e-01

Pos logit contributions
irms+2.802
 goalie+2.796
cer+2.720
 villains+2.684
awks+2.626
Neg logit contributions
pron-3.902
 recipe-3.840
ons-3.553
 mixer-3.442
 baking-3.399
Token laughed
Token ID13818
Feature activation-2.16e+00

Pos logit contributions
 goalie+0.528
irms+0.525
 villains+0.495
awks+0.488
 thrill+0.472
Neg logit contributions
pron-0.612
 recipe-0.604
ons-0.565
 mixer-0.524
 baking-0.502
Token with
Token ID351
Feature activation-4.07e+00

Pos logit contributions
irms+4.547
 goalie+4.429
 villains+4.420
awks+4.336
ashes+4.219
Neg logit contributions
pron-3.142
 recipe-3.041
ons-2.727
 mixer-2.587
 baking-2.520
Token them
Token ID606
Feature activation-1.33e+00

Pos logit contributions
mer+8.847
 goalie+8.776
cer+8.611
irms+8.377
ashes+8.366
Neg logit contributions
ons-6.080
 recipe-5.756
pron-5.430
 mixer-5.162
on-4.939
Token.
Token ID13
Feature activation-1.90e+00

Pos logit contributions
irms+2.203
 goalie+2.113
 villains+2.042
awks+2.032
cer+1.932
Neg logit contributions
pron-2.531
 recipe-2.517
ons-2.337
 mixer-2.209
 baking-2.165
Token He
Token ID679
Feature activation-1.57e+00

Pos logit contributions
irms+3.215
 goalie+3.211
cer+3.142
 villains+3.104
awks+3.097
Neg logit contributions
pron-3.449
 recipe-3.388
ons-3.091
 mixer-2.970
 baking-2.839
Token liked
Token ID8288
Feature activation+5.96e-01

Pos logit contributions
irms+2.933
 goalie+2.876
awks+2.729
 villains+2.669
ashes+2.610
Neg logit contributions
pron-2.811
 recipe-2.672
ons-2.520
 mixer-2.308
 baking-2.195
Token to
Token ID284
Feature activation-1.03e+00

Pos logit contributions
pron+1.126
 recipe+1.090
ons+1.024
 mixer+0.942
innie+0.898
Neg logit contributions
 goalie-1.015
irms-0.994
 villains-0.959
awks-0.928
ashes-0.898
Token right
Token ID826
Feature activation-2.52e+00

Pos logit contributions
irms+3.694
 villains+3.381
awks+3.343
 goalie+3.342
ashes+3.278
Neg logit contributions
 recipe-4.581
pron-4.455
 mixer-3.960
ons-3.913
 baking-3.756
Token thing
Token ID1517
Feature activation-8.44e-01

Pos logit contributions
irms+5.124
 goalie+4.841
awks+4.834
 villains+4.808
ashes+4.682
Neg logit contributions
 recipe-4.449
pron-4.129
ons-3.696
 mixer-3.625
 baking-3.429
Token and
Token ID290
Feature activation-1.96e+00

Pos logit contributions
irms+1.517
 goalie+1.490
awks+1.409
 villains+1.404
 thrill+1.360
Neg logit contributions
pron-1.502
 recipe-1.502
ons-1.404
 mixer-1.291
 baking-1.238
Token they
Token ID484
Feature activation-1.82e+00

Pos logit contributions
irms+3.897
 goalie+3.752
awks+3.600
 thrill+3.500
 villains+3.492
Neg logit contributions
 recipe-3.301
pron-3.246
ons-3.072
 mixer-2.754
 baking-2.736
Token sw
Token ID1509
Feature activation-1.84e+00

Pos logit contributions
irms+3.261
 goalie+3.101
awks+2.983
 villains+2.905
ashes+2.886
Neg logit contributions
pron-3.362
ons-3.205
 recipe-3.186
 mixer-2.850
 baking-2.757
Tokenam
Token ID321
Feature activation-4.07e+00

Pos logit contributions
irms+2.794
 villains+2.762
 goalie+2.735
 thrill+2.616
 villain+2.514
Neg logit contributions
pron-3.649
 recipe-3.627
ons-3.569
 mixer-3.177
 baking-3.027
Token off
Token ID572
Feature activation-1.33e+00

Pos logit contributions
irms+11.641
 villains+10.912
awks+10.655
 goalie+10.513
 villain+10.431
Neg logit contributions
 recipe-3.831
ons-3.426
pron-3.171
 baking-3.113
 to-2.767
Token together
Token ID1978
Feature activation-8.16e-01

Pos logit contributions
irms+3.060
 goalie+3.006
 villains+2.916
awks+2.858
 thrill+2.788
Neg logit contributions
 recipe-1.719
pron-1.690
ons-1.558
 baking-1.404
 mixer-1.391
Token to
Token ID284
Feature activation-1.48e+00

Pos logit contributions
irms+1.402
 goalie+1.371
 villains+1.306
awks+1.299
 thrill+1.239
Neg logit contributions
 recipe-1.517
pron-1.498
ons-1.401
 mixer-1.294
 baking-1.283
Token find
Token ID1064
Feature activation-1.64e+00

Pos logit contributions
irms+2.536
 goalie+2.482
 villains+2.346
awks+2.338
ashes+2.281
Neg logit contributions
 recipe-2.762
pron-2.707
ons-2.529
 baking-2.402
 mixer-2.378
Token some
Token ID617
Feature activation-1.49e-02

Pos logit contributions
 goalie+3.189
irms+3.096
 villains+3.041
awks+2.880
ashes+2.836
Neg logit contributions
 recipe-2.707
ons-2.637
pron-2.584
 mixer-2.363
 baking-2.257
Token the
Token ID262
Feature activation-1.68e+00

Pos logit contributions
 goalie+3.684
irms+3.649
 villains+3.591
awks+3.510
ashes+3.444
Neg logit contributions
 recipe-2.808
pron-2.746
ons-2.603
 mixer-2.500
 baking-2.380
Token p
Token ID279
Feature activation-5.61e-01

Pos logit contributions
 goalie+3.157
irms+3.053
 villains+2.999
awks+2.865
cer+2.797
Neg logit contributions
pron-2.929
 recipe-2.896
 mixer-2.700
ons-2.672
 baking-2.475
Tokenuddle
Token ID24500
Feature activation-1.08e+00

Pos logit contributions
 goalie+1.002
irms+0.981
 villains+0.948
awks+0.909
 thrill+0.895
Neg logit contributions
pron-1.001
 recipe-0.990
ons-0.927
 mixer-0.860
 baking-0.812
Token squ
Token ID2809
Feature activation-2.11e+00

Pos logit contributions
irms+1.630
 goalie+1.620
 villains+1.528
awks+1.508
 thrill+1.416
Neg logit contributions
 recipe-2.249
pron-2.220
ons-2.084
 mixer-1.991
 baking-1.921
Tokenealing
Token ID26919
Feature activation-1.30e+00

Pos logit contributions
 villains+3.170
 goalie+3.001
 villain+2.715
irms+2.630
 thrill+2.516
Neg logit contributions
pron-4.475
 recipe-4.349
ons-4.343
 mixer-4.068
 sugar-3.882
Token with
Token ID351
Feature activation-4.04e+00

Pos logit contributions
irms+2.174
 goalie+2.103
 villains+2.041
awks+2.031
 thrill+1.885
Neg logit contributions
 recipe-2.509
pron-2.469
ons-2.271
 mixer-2.201
 baking-2.137
Token delight
Token ID10974
Feature activation-6.72e-01

Pos logit contributions
 goalie+8.934
 villains+8.752
cer+8.243
mer+8.132
irms+8.088
Neg logit contributions
ons-6.322
pron-5.941
 mixer-5.877
 baking-5.808
 recipe-5.754
Token.
Token ID13
Feature activation-8.13e-01

Pos logit contributions
 goalie+0.940
irms+0.938
 villains+0.876
awks+0.861
 thrill+0.815
Neg logit contributions
pron-1.458
 recipe-1.457
ons-1.361
 mixer-1.282
 baking-1.247
Token Water
Token ID5638
Feature activation+3.49e-01

Pos logit contributions
 goalie+1.033
irms+1.028
 villains+0.966
awks+0.945
 thrill+0.918
Neg logit contributions
pron-1.859
 recipe-1.839
ons-1.732
 mixer-1.648
 baking-1.588
Token spl
Token ID4328
Feature activation-9.98e-01

Pos logit contributions
pron+0.674
 recipe+0.664
ons+0.615
 mixer+0.568
 baking+0.546
Neg logit contributions
 goalie-0.573
irms-0.566
 villains-0.533
awks-0.526
ashes-0.515
Tokenashed
Token ID5263
Feature activation-1.20e+00

Pos logit contributions
 goalie+1.040
 villains+0.976
irms+0.974
awks+0.857
 thrill+0.839
Neg logit contributions
pron-2.524
 recipe-2.464
ons-2.383
 mixer-2.255
 baking-2.181
Token cat
Token ID3797
Feature activation-5.44e-01

Pos logit contributions
irms+2.458
 goalie+2.431
 villains+2.296
awks+2.283
ashes+2.259
Neg logit contributions
 recipe-2.368
pron-2.367
ons-2.185
 mixer-2.070
 baking-1.963
Token was
Token ID373
Feature activation-1.67e+00

Pos logit contributions
 goalie+0.988
irms+0.988
awks+0.919
 villains+0.918
 thrill+0.888
Neg logit contributions
pron-0.943
 recipe-0.933
ons-0.877
 mixer-0.799
 baking-0.764
Token terrified
Token ID22144
Feature activation-1.44e+00

Pos logit contributions
irms+2.655
 goalie+2.612
awks+2.411
 villains+2.404
ashes+2.385
Neg logit contributions
 recipe-3.259
pron-3.259
ons-3.033
 baking-2.918
 mixer-2.862
Token and
Token ID290
Feature activation-1.93e+00

Pos logit contributions
 goalie+2.910
irms+2.892
 villains+2.731
awks+2.665
 thrill+2.615
Neg logit contributions
 recipe-2.229
pron-2.162
ons-2.060
 mixer-1.893
 baking-1.838
Token started
Token ID2067
Feature activation-7.77e-01

Pos logit contributions
irms+3.744
 goalie+3.736
awks+3.412
ashes+3.411
cer+3.315
Neg logit contributions
pron-3.176
 recipe-3.117
ons-3.015
 mixer-2.733
 baking-2.702
Token running
Token ID2491
Feature activation-4.01e+00

Pos logit contributions
 goalie+1.905
irms+1.892
 villains+1.794
awks+1.791
 thrill+1.766
Neg logit contributions
 recipe-0.870
pron-0.848
ons-0.753
 mixer-0.665
 baking-0.636
Token away
Token ID1497
Feature activation-2.31e+00

Pos logit contributions
irms+9.772
 goalie+9.653
awks+9.091
 thrill+8.771
Goal+8.757
Neg logit contributions
 recipe-4.991
ons-4.346
 baking-4.129
 mixer-3.982
pron-3.977
Token.
Token ID13
Feature activation-3.15e-01

Pos logit contributions
irms+4.585
 goalie+4.430
awks+4.214
 villains+4.075
Goal+4.045
Neg logit contributions
 recipe-3.931
pron-3.519
 baking-3.370
ons-3.358
 brushing-3.180
Token
Token ID198
Feature activation+9.83e-01

Pos logit contributions
 goalie+0.446
irms+0.440
 villains+0.411
awks+0.402
 thrill+0.388
Neg logit contributions
pron-0.681
 recipe-0.675
ons-0.636
 mixer-0.596
 baking-0.575
Token
Token ID198
Feature activation+5.79e-01

Pos logit contributions
pron+2.207
 recipe+2.198
ons+1.988
 mixer+1.944
 baking+1.851
Neg logit contributions
 goalie-1.316
irms-1.312
 villains-1.267
awks-1.205
ashes-1.164
TokenIn
Token ID818
Feature activation-8.36e-01

Pos logit contributions
 recipe+1.036
pron+1.033
ons+0.948
 mixer+0.897
 baking+0.842
Neg logit contributions
 goalie-1.039
irms-1.013
 villains-0.998
awks-0.954
 thrill-0.934
Token the
Token ID262
Feature activation-4.44e-01

Pos logit contributions
 goalie+4.289
irms+4.173
 villains+4.149
awks+3.858
 villain+3.756
Neg logit contributions
 recipe-3.684
pron-3.654
ons-3.471
 mixer-3.260
 baking-3.259
Token enclosure
Token ID30685
Feature activation-9.21e-01

Pos logit contributions
 goalie+0.749
irms+0.743
 villains+0.702
awks+0.688
ashes+0.663
Neg logit contributions
pron-0.842
 recipe-0.839
ons-0.772
 mixer-0.726
 baking-0.696
Token.
Token ID13
Feature activation-4.53e-02

Pos logit contributions
irms+1.340
 goalie+1.325
 villains+1.252
awks+1.213
 thrill+1.145
Neg logit contributions
 recipe-1.956
pron-1.939
ons-1.820
 mixer-1.717
 baking-1.680
Token John
Token ID1757
Feature activation+1.71e-01

Pos logit contributions
 goalie+0.068
irms+0.068
 villains+0.064
awks+0.062
 thrill+0.060
Neg logit contributions
pron-0.093
 recipe-0.092
ons-0.086
 mixer-0.080
 baking-0.077
Token laughed
Token ID13818
Feature activation-1.57e+00

Pos logit contributions
pron+0.319
 recipe+0.316
ons+0.293
 mixer+0.272
 baking+0.260
Neg logit contributions
 goalie-0.288
irms-0.283
 villains-0.269
awks-0.265
 thrill-0.257
Token with
Token ID351
Feature activation-4.00e+00

Pos logit contributions
irms+3.164
 goalie+3.111
 villains+3.040
awks+2.979
ashes+2.884
Neg logit contributions
pron-2.514
 recipe-2.429
ons-2.177
 mixer-2.107
 baking-2.043
Token excitement
Token ID14067
Feature activation-9.50e-02

Pos logit contributions
 goalie+6.879
ashes+6.776
mer+6.628
cer+6.343
irms+6.291
Neg logit contributions
ons-8.072
pron-7.981
 recipe-7.822
 mixer-7.593
 baking-7.360
Token and
Token ID290
Feature activation-1.69e+00

Pos logit contributions
 goalie+0.156
irms+0.154
 villains+0.146
awks+0.143
 thrill+0.139
Neg logit contributions
pron-0.182
 recipe-0.180
ons-0.168
 mixer-0.156
 baking-0.149
Token cl
Token ID537
Feature activation-1.88e+00

Pos logit contributions
 goalie+3.145
irms+3.143
awks+2.935
 villains+2.880
ashes+2.828
Neg logit contributions
pron-2.975
 recipe-2.882
ons-2.797
 mixer-2.518
 baking-2.494
Tokenapped
Token ID6320
Feature activation-2.22e+00

Pos logit contributions
irms+2.572
 villains+2.555
 goalie+2.552
awks+2.351
 villain+2.254
Neg logit contributions
pron-4.171
 recipe-4.031
ons-3.965
 mixer-3.723
ama-3.587
Token his
Token ID465
Feature activation-1.30e+00

Pos logit contributions
irms+4.018
 goalie+3.992
awks+3.762
 villains+3.755
ashes+3.610
Neg logit contributions
pron-4.116
 recipe-4.091
ons-3.657
 baking-3.460
 mixer-3.441
Token louder
Token ID27089
Feature activation-1.95e+00

Pos logit contributions
 goalie+0.709
irms+0.697
 villains+0.651
awks+0.633
ashes+0.613
Neg logit contributions
pron-1.077
 recipe-1.067
ons-1.004
 mixer-0.948
 baking-0.913
Token and
Token ID290
Feature activation-5.46e-01

Pos logit contributions
 goalie+4.068
irms+4.030
awks+3.728
 villains+3.678
Goal+3.587
Neg logit contributions
 recipe-2.996
pron-2.871
ons-2.731
 baking-2.535
 mixer-2.490
Token soon
Token ID2582
Feature activation-9.57e-01

Pos logit contributions
 goalie+0.883
irms+0.874
 villains+0.824
awks+0.803
ashes+0.781
Neg logit contributions
pron-1.062
 recipe-1.053
ons-0.983
 mixer-0.923
 baking-0.887
Token everyone
Token ID2506
Feature activation-1.79e+00

Pos logit contributions
 goalie+1.844
irms+1.833
 villains+1.713
awks+1.700
ashes+1.658
Neg logit contributions
pron-1.575
 recipe-1.550
ons-1.498
 mixer-1.334
 baking-1.289
Token started
Token ID2067
Feature activation-1.30e+00

Pos logit contributions
irms+3.870
 goalie+3.799
awks+3.636
 villains+3.470
ashes+3.399
Neg logit contributions
pron-2.565
 recipe-2.540
ons-2.515
 mixer-2.245
 baking-2.109
Token running
Token ID2491
Feature activation-3.94e+00

Pos logit contributions
 goalie+3.344
irms+3.340
awks+3.169
 villains+3.161
 thrill+3.108
Neg logit contributions
 recipe-1.289
pron-1.228
ons-1.102
 mixer-0.987
 baking-0.960
Token,
Token ID11
Feature activation-7.78e-01

Pos logit contributions
irms+9.569
 goalie+9.399
awks+9.057
 villains+8.435
 thrill+8.387
Neg logit contributions
 recipe-4.856
ons-4.427
 baking-4.235
 mixer-4.224
pron-4.105
Token screaming
Token ID14788
Feature activation-1.88e+00

Pos logit contributions
 goalie+1.355
irms+1.349
 villains+1.269
awks+1.242
cer+1.206
Neg logit contributions
pron-1.418
 recipe-1.401
ons-1.294
 mixer-1.220
 baking-1.182
Token,
Token ID11
Feature activation-5.42e-01

Pos logit contributions
 goalie+3.401
irms+3.370
awks+3.124
 villains+3.114
 thrill+2.962
Neg logit contributions
 recipe-3.364
pron-3.310
ons-3.056
 mixer-2.892
 baking-2.875
Token and
Token ID290
Feature activation-6.27e-01

Pos logit contributions
 goalie+0.970
irms+0.963
 villains+0.913
awks+0.892
ashes+0.866
Neg logit contributions
pron-0.964
 recipe-0.950
ons-0.875
 mixer-0.818
 baking-0.789
Token pan
Token ID3425
Feature activation-6.91e-01

Pos logit contributions
 goalie+1.089
irms+1.081
 villains+1.020
awks+0.999
ashes+0.967
Neg logit contributions
pron-1.135
 recipe-1.133
ons-1.053
 mixer-0.984
 baking-0.953
Token jet
Token ID12644
Feature activation+3.70e-01

Pos logit contributions
irms+1.968
 goalie+1.932
awks+1.828
 villains+1.796
ashes+1.780
Neg logit contributions
pron-1.836
 recipe-1.766
ons-1.648
 mixer-1.553
 baking-1.464
Token took
Token ID1718
Feature activation-1.77e+00

Pos logit contributions
pron+0.696
 recipe+0.690
ons+0.632
 mixer+0.591
 baking+0.563
Neg logit contributions
 goalie-0.624
irms-0.605
 villains-0.584
awks-0.571
 thrill-0.557
Token off
Token ID572
Feature activation-6.50e-01

Pos logit contributions
 goalie+3.806
irms+3.756
 villains+3.538
awks+3.509
ashes+3.447
Neg logit contributions
ons-2.481
 recipe-2.431
pron-2.412
 mixer-2.118
 baking-2.075
Token,
Token ID11
Feature activation-1.54e+00

Pos logit contributions
 goalie+1.298
irms+1.295
 villains+1.227
awks+1.214
ashes+1.172
Neg logit contributions
pron-1.017
 recipe-1.009
ons-0.947
 mixer-0.862
 baking-0.825
Token the
Token ID262
Feature activation-1.76e+00

Pos logit contributions
irms+3.107
 goalie+3.002
awks+2.918
 villains+2.901
ashes+2.847
Neg logit contributions
pron-2.456
 recipe-2.391
ons-2.257
 mixer-2.134
 baking-2.078
Token little
Token ID1310
Feature activation-3.94e+00

Pos logit contributions
irms+3.724
 goalie+3.572
awks+3.475
ashes+3.399
 villains+3.359
Neg logit contributions
pron-2.806
 recipe-2.634
ons-2.420
 mixer-2.346
 baking-2.184
Token boy
Token ID2933
Feature activation+4.13e-01

Pos logit contributions
irms+6.433
 villains+6.224
awks+6.111
ks+6.089
 victorious+5.609
Neg logit contributions
 mixer-8.403
pron-8.121
 recipe-7.836
 sugar-7.509
ons-7.390
Token screamed
Token ID25421
Feature activation-9.55e-01

Pos logit contributions
pron+0.794
 recipe+0.788
ons+0.718
 mixer+0.674
 baking+0.644
Neg logit contributions
 goalie-0.685
irms-0.668
 villains-0.639
awks-0.626
cer-0.611
Token with
Token ID351
Feature activation-3.10e+00

Pos logit contributions
 goalie+1.634
irms+1.633
 villains+1.551
awks+1.531
ashes+1.471
Neg logit contributions
pron-1.774
 recipe-1.742
ons-1.587
 mixer-1.520
 baking-1.487
Token joy
Token ID8716
Feature activation-4.65e-01

Pos logit contributions
 goalie+5.402
ashes+5.142
cer+5.065
mer+5.022
irms+4.847
Neg logit contributions
pron-5.884
ons-5.847
 recipe-5.677
 mixer-5.324
 baking-5.215
Token.
Token ID13
Feature activation-7.30e-02

Pos logit contributions
 goalie+0.649
irms+0.645
 villains+0.601
awks+0.591
ashes+0.560
Neg logit contributions
pron-1.017
 recipe-1.008
ons-0.945
 mixer-0.890
 baking-0.860
Token ear
Token ID1027
Feature activation+3.95e-01

Pos logit contributions
 goalie+2.136
irms+2.116
awks+1.974
ashes+1.940
 villains+1.936
Neg logit contributions
 recipe-2.469
pron-2.467
ons-2.329
 baking-2.129
 mixer-2.128
Token.
Token ID13
Feature activation-1.36e-01

Pos logit contributions
pron+0.711
 recipe+0.705
ons+0.649
 mixer+0.604
 baking+0.569
Neg logit contributions
 goalie-0.690
irms-0.676
 villains-0.644
awks-0.632
cer-0.618
Token Everyone
Token ID11075
Feature activation-1.45e-01

Pos logit contributions
 goalie+0.119
irms+0.117
 villains+0.105
awks+0.101
 thrill+0.094
Neg logit contributions
pron-0.367
 recipe-0.364
ons-0.347
 mixer-0.330
 baking-0.320
Token congratulated
Token ID47630
Feature activation-1.86e+00

Pos logit contributions
 goalie+0.253
irms+0.251
 villains+0.238
awks+0.234
 thrill+0.227
Neg logit contributions
pron-0.263
 recipe-0.260
ons-0.241
 mixer-0.223
 baking-0.213
Token the
Token ID262
Feature activation-2.84e+00

Pos logit contributions
 goalie+3.107
irms+3.077
 villains+2.835
awks+2.795
ashes+2.787
Neg logit contributions
 recipe-3.550
ons-3.497
pron-3.481
 mixer-3.152
 baking-3.098
Token little
Token ID1310
Feature activation-3.90e+00

Pos logit contributions
irms+5.289
 goalie+5.116
awks+4.926
 villains+4.670
cer+4.608
Neg logit contributions
 recipe-5.719
pron-5.601
 mixer-5.072
ons-4.942
 baking-4.790
Token boy
Token ID2933
Feature activation-1.16e+00

Pos logit contributions
irms+6.531
awks+6.104
 goalie+5.659
 villains+5.621
 thrill+5.500
Neg logit contributions
 recipe-8.964
 mixer-8.563
pron-8.104
 sugar-7.560
 baking-7.509
Token on
Token ID319
Feature activation-4.11e-01

Pos logit contributions
irms+2.204
 goalie+2.171
 villains+2.053
awks+2.036
 thrill+1.979
Neg logit contributions
 recipe-1.934
pron-1.923
ons-1.831
 mixer-1.636
 baking-1.623
Token such
Token ID884
Feature activation-2.63e+00

Pos logit contributions
 goalie+0.607
irms+0.606
 villains+0.564
awks+0.558
ashes+0.536
Neg logit contributions
pron-0.864
 recipe-0.857
ons-0.802
 mixer-0.750
 baking-0.729
Token an
Token ID281
Feature activation-1.60e-01

Pos logit contributions
irms+5.405
awks+5.173
ashes+5.070
 goalie+4.977
cer+4.476
Neg logit contributions
pron-4.751
 recipe-4.599
ons-4.012
 baking-3.923
 mixer-3.836
Token important
Token ID1593
Feature activation-1.52e+00

Pos logit contributions
 goalie+0.240
irms+0.238
 villains+0.223
awks+0.219
 thrill+0.211
Neg logit contributions
pron-0.328
 recipe-0.325
ons-0.304
 mixer-0.284
 baking-0.274
Token her
Token ID607
Feature activation-1.42e+00

Pos logit contributions
 goalie+0.406
irms+0.401
 villains+0.377
awks+0.371
 thrill+0.359
Neg logit contributions
pron-0.499
 recipe-0.495
ons-0.465
 mixer-0.431
 baking-0.414
Token too
Token ID1165
Feature activation-1.09e+00

Pos logit contributions
irms+2.729
 goalie+2.604
 villains+2.598
awks+2.507
ashes+2.419
Neg logit contributions
 recipe-2.422
pron-2.377
ons-2.132
 mixer-2.064
 baking-2.008
Token and
Token ID290
Feature activation-4.14e-01

Pos logit contributions
irms+1.983
 goalie+1.970
 villains+1.898
awks+1.818
 thrill+1.766
Neg logit contributions
 recipe-1.862
pron-1.858
ons-1.767
 mixer-1.607
 baking-1.573
Token started
Token ID2067
Feature activation+3.09e-01

Pos logit contributions
 goalie+0.688
irms+0.683
 villains+0.639
awks+0.631
 thrill+0.611
Neg logit contributions
pron-0.789
 recipe-0.780
ons-0.727
 mixer-0.676
 baking-0.651
Token to
Token ID284
Feature activation-1.51e+00

Pos logit contributions
pron+0.395
 recipe+0.378
ons+0.346
 mixer+0.302
 baking+0.272
Neg logit contributions
 goalie-0.713
irms-0.707
 villains-0.679
awks-0.674
ashes-0.659
Token swim
Token ID9422
Feature activation-3.90e+00

Pos logit contributions
irms+2.524
 goalie+2.495
 villains+2.464
awks+2.291
cer+2.243
Neg logit contributions
 recipe-2.838
pron-2.781
ons-2.481
 mixer-2.464
 baking-2.424
Token away
Token ID1497
Feature activation-2.02e+00

Pos logit contributions
irms+9.623
 villains+9.319
 goalie+9.139
awks+9.000
 thrill+8.757
Neg logit contributions
 recipe-4.917
pron-4.513
 baking-3.857
ons-3.849
 mixer-3.777
Token.
Token ID13
Feature activation-2.10e-01

Pos logit contributions
irms+3.705
 goalie+3.581
 villains+3.426
awks+3.383
Goal+3.228
Neg logit contributions
 recipe-3.713
pron-3.457
ons-3.202
 baking-3.176
 mixer-3.071
Token The
Token ID383
Feature activation-2.45e+00

Pos logit contributions
 goalie+0.338
irms+0.334
 villains+0.315
awks+0.309
 thrill+0.299
Neg logit contributions
pron-0.413
 recipe-0.408
ons-0.381
 mixer-0.355
 baking-0.340
Token little
Token ID1310
Feature activation-2.54e+00

Pos logit contributions
irms+5.491
 goalie+5.266
awks+5.127
 villains+5.028
ashes+4.825
Neg logit contributions
pron-3.753
 recipe-3.670
 mixer-3.043
ons-2.908
 baking-2.847
Token girl
Token ID2576
Feature activation-5.51e-01

Pos logit contributions
irms+4.362
 goalie+4.229
 villains+4.203
awks+3.892
 victorious+3.749
Neg logit contributions
 recipe-5.012
pron-4.854
 mixer-4.439
ons-4.286
 sugar-4.179
Token and
Token ID290
Feature activation-7.84e-01

Pos logit contributions
irms+3.583
 villains+3.517
 goalie+3.480
awks+3.358
ashes+3.293
Neg logit contributions
pron-2.881
 recipe-2.802
ons-2.389
 mixer-2.334
 baking-2.280
Token waved
Token ID26834
Feature activation-1.95e+00

Pos logit contributions
 goalie+1.350
irms+1.339
 villains+1.259
awks+1.243
ashes+1.196
Neg logit contributions
pron-1.459
 recipe-1.434
ons-1.350
 mixer-1.239
 baking-1.197
Token goodbye
Token ID24829
Feature activation-7.82e-01

Pos logit contributions
 goalie+3.269
irms+3.237
 villains+3.081
awks+2.964
ashes+2.855
Neg logit contributions
pron-3.784
 recipe-3.762
ons-3.404
 mixer-3.240
 baking-3.159
Token to
Token ID284
Feature activation-1.67e+00

Pos logit contributions
irms+1.566
 goalie+1.565
 villains+1.480
awks+1.457
 thrill+1.411
Neg logit contributions
pron-1.230
 recipe-1.222
ons-1.092
 mixer-1.007
 baking-0.978
Token the
Token ID262
Feature activation-2.62e+00

Pos logit contributions
 goalie+3.518
irms+3.456
awks+3.232
 villains+3.228
ashes+3.213
Neg logit contributions
 recipe-2.558
pron-2.510
ons-2.360
 mixer-2.105
 baking-2.087
Token little
Token ID1310
Feature activation-3.90e+00

Pos logit contributions
irms+4.850
 goalie+4.605
ashes+4.444
awks+4.433
cer+4.067
Neg logit contributions
pron-5.190
 recipe-5.162
ons-4.363
 mixer-4.297
 baking-4.122
Token girl
Token ID2576
Feature activation-7.01e-01

Pos logit contributions
irms+7.839
awks+7.293
 goalie+7.056
 villains+6.860
ashes+6.746
Neg logit contributions
 recipe-7.706
pron-7.493
 mixer-6.506
 sugar-6.152
 baking-6.091
Token as
Token ID355
Feature activation-1.41e+00

Pos logit contributions
irms+1.120
 goalie+1.118
 villains+1.050
awks+1.026
 thrill+0.991
Neg logit contributions
pron-1.373
 recipe-1.367
ons-1.284
 mixer-1.186
 baking-1.158
Token he
Token ID339
Feature activation-1.32e-01

Pos logit contributions
irms+3.263
 goalie+3.203
awks+3.057
ashes+3.027
 villains+2.981
Neg logit contributions
pron-1.867
 recipe-1.807
ons-1.709
 mixer-1.459
 baking-1.455
Token sailed
Token ID33835
Feature activation-1.73e+00

Pos logit contributions
 goalie+0.202
irms+0.200
 villains+0.188
awks+0.185
 thrill+0.178
Neg logit contributions
pron-0.268
 recipe-0.266
ons-0.249
 mixer-0.232
 baking-0.223
Token away
Token ID1497
Feature activation-1.34e+00

Pos logit contributions
irms+2.634
 goalie+2.622
 villains+2.454
awks+2.375
ashes+2.290
Neg logit contributions
 recipe-3.656
pron-3.633
ons-3.445
 mixer-3.126
 baking-3.097
Token goodbye
Token ID24829
Feature activation-6.32e-01

Pos logit contributions
 goalie+5.587
irms+5.442
 villains+5.330
cer+5.048
awks+5.011
Neg logit contributions
 recipe-3.995
pron-3.960
 mixer-3.517
ons-3.497
 baking-3.363
Token and
Token ID290
Feature activation-2.22e+00

Pos logit contributions
 goalie+1.395
irms+1.391
 villains+1.330
awks+1.305
 thrill+1.277
Neg logit contributions
pron-0.850
 recipe-0.845
ons-0.749
 mixer-0.680
 baking-0.649
Token then
Token ID788
Feature activation-1.63e+00

Pos logit contributions
 goalie+4.448
irms+4.352
awks+4.101
ashes+4.024
 villains+3.927
Neg logit contributions
 recipe-3.567
pron-3.543
ons-3.427
 mixer-3.066
 baking-3.025
Token they
Token ID484
Feature activation-1.81e+00

Pos logit contributions
irms+2.984
 goalie+2.978
awks+2.772
 villains+2.739
ashes+2.684
Neg logit contributions
 recipe-2.764
pron-2.754
ons-2.658
 mixer-2.420
 baking-2.380
Token started
Token ID2067
Feature activation-6.89e-01

Pos logit contributions
irms+3.195
 goalie+3.130
awks+2.955
 villains+2.929
ashes+2.879
Neg logit contributions
pron-3.248
 recipe-3.108
ons-3.071
 mixer-2.768
 baking-2.673
Token running
Token ID2491
Feature activation-3.88e+00

Pos logit contributions
 goalie+1.677
irms+1.670
 villains+1.604
awks+1.587
 thrill+1.558
Neg logit contributions
 recipe-0.775
pron-0.762
ons-0.678
 mixer-0.600
 baking-0.576
Token back
Token ID736
Feature activation-1.20e+00

Pos logit contributions
irms+10.017
 goalie+9.831
 villains+9.585
awks+9.553
 villain+9.199
Neg logit contributions
 recipe-3.968
ons-3.620
 baking-3.545
 mixer-3.399
 measuring-3.157
Token to
Token ID284
Feature activation-1.77e+00

Pos logit contributions
 goalie+2.888
irms+2.865
 villains+2.789
awks+2.725
 thrill+2.712
Neg logit contributions
 recipe-1.384
pron-1.302
ons-1.241
 mixer-1.103
 baking-1.063
Token the
Token ID262
Feature activation-1.27e+00

Pos logit contributions
 goalie+3.319
irms+3.283
 villains+3.116
awks+3.075
ashes+3.016
Neg logit contributions
 recipe-2.987
pron-2.863
ons-2.686
 baking-2.678
 mixer-2.602
Token park
Token ID3952
Feature activation-1.69e+00

Pos logit contributions
irms+1.899
 goalie+1.875
 villains+1.759
awks+1.703
 thrill+1.620
Neg logit contributions
 recipe-2.690
pron-2.646
ons-2.439
 mixer-2.405
 baking-2.322
Token on
Token ID319
Feature activation-1.24e+00

Pos logit contributions
irms+3.025
 goalie+2.901
awks+2.750
 villains+2.745
 thrill+2.654
Neg logit contributions
 recipe-3.142
pron-3.003
ons-2.827
 baking-2.694
 mixer-2.676
Token together
Token ID1978
Feature activation+1.14e+00

Pos logit contributions
 goalie+0.413
irms+0.410
 villains+0.383
awks+0.377
 thrill+0.362
Neg logit contributions
pron-0.609
 recipe-0.605
ons-0.569
 mixer-0.532
 baking-0.514
Token and
Token ID290
Feature activation-5.67e-01

Pos logit contributions
pron+2.780
 recipe+2.712
ons+2.606
 mixer+2.432
innie+2.345
Neg logit contributions
 goalie-1.339
irms-1.226
 villains-1.137
cer-1.110
ashes-1.105
Token soon
Token ID2582
Feature activation-1.16e+00

Pos logit contributions
irms+1.090
 goalie+1.089
 villains+1.028
awks+1.024
 thrill+0.994
Neg logit contributions
pron-0.924
 recipe-0.922
ons-0.850
 mixer-0.782
 baking-0.745
Token they
Token ID484
Feature activation-1.48e+00

Pos logit contributions
irms+2.325
 goalie+2.303
awks+2.195
 villains+2.177
 thrill+2.121
Neg logit contributions
 recipe-1.834
pron-1.803
ons-1.729
 mixer-1.522
 baking-1.478
Token were
Token ID547
Feature activation-1.66e+00

Pos logit contributions
irms+2.824
 goalie+2.717
awks+2.645
 villains+2.623
ashes+2.476
Neg logit contributions
pron-2.520
 recipe-2.449
ons-2.367
 mixer-2.153
 baking-2.046
Token running
Token ID2491
Feature activation-3.88e+00

Pos logit contributions
irms+3.175
 villains+3.047
 goalie+3.038
awks+2.961
 thrill+2.808
Neg logit contributions
pron-2.822
 recipe-2.783
ons-2.561
 baking-2.493
 mixer-2.412
Token and
Token ID290
Feature activation-4.06e-01

Pos logit contributions
irms+9.992
 goalie+9.538
 villains+9.522
awks+9.503
 thrill+9.235
Neg logit contributions
 recipe-4.181
ons-3.914
pron-3.818
 baking-3.738
 mixer-3.653
Token diving
Token ID23186
Feature activation-2.69e+00

Pos logit contributions
 goalie+0.747
irms+0.743
 villains+0.706
awks+0.697
 thrill+0.674
Neg logit contributions
pron-0.697
 recipe-0.692
ons-0.641
 mixer-0.592
 baking-0.569
Token into
Token ID656
Feature activation-1.54e+00

Pos logit contributions
irms+5.380
 goalie+5.273
 villains+5.165
awks+5.103
 thrill+4.862
Neg logit contributions
 recipe-4.413
pron-4.171
ons-4.014
 mixer-3.671
 baking-3.618
Token the
Token ID262
Feature activation-2.10e+00

Pos logit contributions
 goalie+3.040
irms+3.008
 villains+2.909
awks+2.837
ashes+2.781
Neg logit contributions
 recipe-2.435
pron-2.340
ons-2.280
 mixer-2.131
 baking-2.074
Token pond
Token ID16723
Feature activation-1.01e+00

Pos logit contributions
 goalie+3.441
irms+3.438
 villains+3.404
awks+3.099
ks+3.042
Neg logit contributions
pron-4.167
 recipe-4.109
ons-3.748
 mixer-3.726
 baking-3.600
TokenTom
Token ID13787
Feature activation-1.18e+00

Pos logit contributions
irms+1.667
 goalie+1.638
awks+1.562
 villains+1.546
ashes+1.516
Neg logit contributions
pron-1.541
 recipe-1.496
ons-1.398
 mixer-1.276
 baking-1.200
Token and
Token ID290
Feature activation-1.76e+00

Pos logit contributions
 goalie+1.985
 villains+1.946
irms+1.942
 thrill+1.843
awks+1.821
Neg logit contributions
pron-2.181
 recipe-2.180
ons-2.074
 mixer-1.916
 baking-1.784
Token Sam
Token ID3409
Feature activation-2.10e+00

Pos logit contributions
 goalie+2.877
irms+2.823
awks+2.684
ashes+2.643
 thrill+2.628
Neg logit contributions
 recipe-3.402
pron-3.384
ons-3.199
 mixer-3.037
 baking-2.848
Token feel
Token ID1254
Feature activation-3.06e+00

Pos logit contributions
 goalie+3.640
 villains+3.565
irms+3.525
cer+3.364
awks+3.225
Neg logit contributions
pron-3.623
 recipe-3.606
ons-3.597
 mixer-3.389
 baking-3.140
Token sad
Token ID6507
Feature activation-1.49e+00

Pos logit contributions
 goalie+5.331
cer+4.931
irms+4.613
 villains+4.575
ashes+4.482
Neg logit contributions
ons-5.726
pron-5.636
 recipe-5.472
 mixer-5.368
 baking-5.260
Token and
Token ID290
Feature activation-3.87e+00

Pos logit contributions
 goalie+2.461
irms+2.402
 villains+2.259
awks+2.254
cer+2.185
Neg logit contributions
 recipe-2.907
pron-2.808
ons-2.638
 mixer-2.521
 baking-2.458
Token angry
Token ID7954
Feature activation-3.24e+00

Pos logit contributions
 goalie+7.158
cer+6.894
irms+6.182
 loser+6.155
 villains+6.064
Neg logit contributions
 recipe-7.173
pron-7.122
ons-6.873
 baking-6.751
 mixer-6.689
Token.
Token ID13
Feature activation-2.29e+00

Pos logit contributions
 goalie+6.597
irms+6.507
 thrill+6.268
Goal+6.205
awks+6.178
Neg logit contributions
 recipe-5.145
ons-4.782
 mixer-4.484
 baking-4.408
pron-4.333
Token They
Token ID1119
Feature activation-2.85e+00

Pos logit contributions
irms+2.726
 goalie+2.681
 villains+2.518
 thrill+2.510
awks+2.475
Neg logit contributions
 recipe-5.358
pron-5.194
 mixer-5.084
ons-4.935
 baking-4.738
Token are
Token ID389
Feature activation-2.50e+00

Pos logit contributions
 goalie+5.265
cer+5.116
irms+5.087
 villains+4.900
mer+4.763
Neg logit contributions
pron-4.826
 recipe-4.682
ons-4.528
 mixer-4.362
 comb-4.271
Token jealous
Token ID19354
Feature activation-1.86e+00

Pos logit contributions
irms+4.721
 goalie+4.686
cer+4.553
awks+4.546
ashes+4.452
Neg logit contributions
pron-4.279
 recipe-4.143
 baking-3.790
ons-3.776
 mixer-3.734
Token't
Token ID470
Feature activation-4.50e-01

Pos logit contributions
pron+1.106
 recipe+1.090
ons+0.949
 mixer+0.829
 baking+0.754
Neg logit contributions
 goalie-2.098
irms-2.035
 villains-2.003
awks-1.940
 thrill-1.930
Token seem
Token ID1283
Feature activation-9.99e-01

Pos logit contributions
 goalie+0.726
irms+0.722
 villains+0.671
awks+0.666
cer+0.639
Neg logit contributions
pron-0.883
 recipe-0.877
ons-0.812
 mixer-0.760
 baking-0.734
Token to
Token ID284
Feature activation-9.01e-01

Pos logit contributions
irms+1.992
 goalie+1.989
 villains+1.902
awks+1.860
cer+1.807
Neg logit contributions
 recipe-1.567
pron-1.565
ons-1.364
 mixer-1.296
 baking-1.272
Token bother
Token ID11393
Feature activation+1.69e-01

Pos logit contributions
irms+1.604
 goalie+1.579
 villains+1.489
awks+1.476
cer+1.414
Neg logit contributions
 recipe-1.613
pron-1.606
ons-1.460
 mixer-1.381
 baking-1.357
Token the
Token ID262
Feature activation-2.62e+00

Pos logit contributions
pron+0.338
 recipe+0.332
ons+0.310
 mixer+0.288
 baking+0.275
Neg logit contributions
 goalie-0.268
irms-0.264
 villains-0.250
awks-0.245
 thrill-0.236
Token little
Token ID1310
Feature activation-3.86e+00

Pos logit contributions
irms+5.130
 goalie+4.799
awks+4.626
 villains+4.539
ashes+4.493
Neg logit contributions
pron-4.774
 recipe-4.720
 mixer-4.261
ons-4.145
 baking-4.047
Token boy
Token ID2933
Feature activation-5.06e-01

Pos logit contributions
irms+6.371
 villains+5.749
awks+5.634
 goalie+5.526
ks+5.306
Neg logit contributions
 recipe-8.456
 mixer-8.396
pron-8.306
ons-7.733
 baking-7.728
Token and
Token ID290
Feature activation-1.10e+00

Pos logit contributions
 goalie+0.763
irms+0.762
 villains+0.711
awks+0.698
 thrill+0.673
Neg logit contributions
pron-1.034
 recipe-1.030
ons-0.964
 mixer-0.901
 baking-0.876
Token his
Token ID465
Feature activation-1.33e+00

Pos logit contributions
 goalie+1.927
irms+1.926
 villains+1.794
awks+1.790
ashes+1.736
Neg logit contributions
pron-2.002
 recipe-1.986
ons-1.848
 mixer-1.720
 baking-1.675
Token dad
Token ID9955
Feature activation-2.26e-02

Pos logit contributions
irms+2.726
 goalie+2.708
awks+2.552
 villains+2.547
ashes+2.493
Neg logit contributions
pron-2.114
 recipe-2.073
ons-1.845
 mixer-1.776
 baking-1.666
Token.
Token ID13
Feature activation-6.77e-01

Pos logit contributions
 goalie+0.033
irms+0.032
 villains+0.030
awks+0.030
 thrill+0.028
Neg logit contributions
pron-0.048
 recipe-0.047
ons-0.044
 mixer-0.042
 baking-0.040
Token like
Token ID588
Feature activation-2.37e-01

Pos logit contributions
irms+3.957
 goalie+3.830
 villains+3.534
awks+3.514
cer+3.319
Neg logit contributions
 recipe-3.996
pron-3.960
ons-3.540
 baking-3.446
 mixer-3.332
Token hours
Token ID2250
Feature activation+3.58e-02

Pos logit contributions
 goalie+0.415
irms+0.412
 villains+0.389
awks+0.384
 thrill+0.371
Neg logit contributions
pron-0.429
 recipe-0.424
ons-0.394
 mixer-0.364
 baking-0.348
Token,
Token ID11
Feature activation-1.62e+00

Pos logit contributions
pron+0.067
 recipe+0.067
ons+0.062
 mixer+0.058
 baking+0.055
Neg logit contributions
 goalie-0.060
irms-0.059
 villains-0.056
awks-0.055
 thrill-0.053
Token and
Token ID290
Feature activation-1.02e+00

Pos logit contributions
irms+2.962
 goalie+2.910
 villains+2.740
awks+2.676
ashes+2.612
Neg logit contributions
 recipe-2.841
pron-2.819
ons-2.625
 baking-2.470
 mixer-2.412
Token the
Token ID262
Feature activation-2.51e+00

Pos logit contributions
 goalie+1.748
irms+1.737
awks+1.596
 villains+1.595
ashes+1.554
Neg logit contributions
 recipe-1.884
pron-1.883
ons-1.775
 mixer-1.611
 baking-1.607
Token little
Token ID1310
Feature activation-3.83e+00

Pos logit contributions
irms+4.719
 goalie+4.428
awks+4.328
ashes+4.162
 villains+3.989
Neg logit contributions
pron-4.894
 recipe-4.785
 mixer-4.108
ons-4.100
 baking-4.018
Token girl
Token ID2576
Feature activation-4.87e-01

Pos logit contributions
irms+7.246
awks+6.588
 goalie+6.473
ashes+6.106
 villains+6.012
Neg logit contributions
 recipe-8.009
pron-7.473
 mixer-7.030
 sugar-6.709
 baking-6.695
Token got
Token ID1392
Feature activation-1.65e+00

Pos logit contributions
irms+0.850
 goalie+0.849
 villains+0.795
awks+0.788
 thrill+0.760
Neg logit contributions
pron-0.880
 recipe-0.869
ons-0.817
 mixer-0.749
 baking-0.720
Token awfully
Token ID45421
Feature activation-2.49e+00

Pos logit contributions
irms+3.485
 goalie+3.482
 villains+3.331
awks+3.215
 thrill+3.158
Neg logit contributions
 recipe-2.431
pron-2.407
ons-2.308
 mixer-2.063
 baking-2.055
Token hot
Token ID3024
Feature activation-1.64e+00

Pos logit contributions
 goalie+4.342
irms+4.332
 villains+4.085
awks+3.835
Goal+3.828
Neg logit contributions
pron-4.587
 recipe-4.516
ons-4.361
 mixer-4.086
 baking-4.080
Token.
Token ID13
Feature activation+4.81e-02

Pos logit contributions
 goalie+3.154
 villains+3.045
irms+3.045
awks+2.893
 thrill+2.759
Neg logit contributions
 recipe-2.818
pron-2.743
ons-2.597
 mixer-2.478
 baking-2.383
Tokenily
Token ID813
Feature activation-9.22e-01

Pos logit contributions
 goalie+2.163
irms+2.126
awks+2.009
 villains+1.974
 thrill+1.935
Neg logit contributions
pron-2.876
 recipe-2.849
ons-2.768
 mixer-2.497
innie-2.422
Token and
Token ID290
Feature activation-6.04e-01

Pos logit contributions
 goalie+1.617
irms+1.592
 villains+1.569
awks+1.491
 thrill+1.482
Neg logit contributions
 recipe-1.625
pron-1.623
ons-1.517
 mixer-1.420
 baking-1.332
Token Max
Token ID5436
Feature activation-2.05e+00

Pos logit contributions
 goalie+0.753
irms+0.739
 villains+0.687
awks+0.678
ashes+0.661
Neg logit contributions
pron-1.383
 recipe-1.367
ons-1.310
 mixer-1.228
 baking-1.183
Token feel
Token ID1254
Feature activation-3.33e+00

Pos logit contributions
 goalie+3.414
irms+3.377
 villains+3.373
cer+3.164
awks+3.123
Neg logit contributions
 recipe-3.664
pron-3.661
ons-3.464
 mixer-3.319
 baking-3.200
Token sad
Token ID6507
Feature activation-1.08e+00

Pos logit contributions
 goalie+5.958
cer+5.543
irms+5.112
 villains+5.084
mer+5.002
Neg logit contributions
ons-6.078
pron-6.034
 recipe-5.813
 baking-5.621
 mixer-5.610
Token and
Token ID290
Feature activation-3.82e+00

Pos logit contributions
 goalie+1.678
irms+1.628
 villains+1.537
awks+1.515
cer+1.475
Neg logit contributions
 recipe-2.201
pron-2.160
ons-2.033
 mixer-1.917
 baking-1.868
Token angry
Token ID7954
Feature activation-2.75e+00

Pos logit contributions
 goalie+6.498
cer+6.304
mer+5.570
 villains+5.537
 loser+5.446
Neg logit contributions
 recipe-7.468
pron-7.405
 baking-7.116
ons-7.060
 mixer-6.864
Token.
Token ID13
Feature activation-2.69e+00

Pos logit contributions
 goalie+5.216
irms+5.051
 thrill+4.898
 villains+4.845
cer+4.796
Neg logit contributions
 recipe-4.683
ons-4.417
pron-4.065
 mixer-4.039
 baking-4.026
Token They
Token ID1119
Feature activation-2.66e+00

Pos logit contributions
 goalie+3.397
 thrill+3.177
irms+3.131
cer+3.053
awks+2.948
Neg logit contributions
 recipe-6.091
pron-5.853
 mixer-5.675
ons-5.664
 baking-5.359
Token do
Token ID466
Feature activation-8.98e-01

Pos logit contributions
 goalie+4.926
cer+4.775
irms+4.736
 villains+4.586
mer+4.465
Neg logit contributions
pron-4.557
 recipe-4.389
ons-4.164
 mixer-3.988
 comb-3.932
Token not
Token ID407
Feature activation-2.55e+00

Pos logit contributions
 goalie+1.951
irms+1.946
 villains+1.842
awks+1.831
cer+1.795
Neg logit contributions
pron-1.278
 recipe-1.265
ons-1.157
 mixer-1.024
 baking-1.005
Token and
Token ID290
Feature activation-1.35e+00

Pos logit contributions
irms+1.067
 goalie+1.067
 villains+1.003
awks+0.992
ashes+0.954
Neg logit contributions
pron-1.084
 recipe-1.070
ons-0.998
 mixer-0.922
 baking-0.879
Token the
Token ID262
Feature activation-7.24e-01

Pos logit contributions
irms+1.772
 goalie+1.746
 villains+1.656
awks+1.656
ashes+1.589
Neg logit contributions
pron-3.047
 recipe-3.004
ons-2.811
 mixer-2.682
 baking-2.595
Token ambulance
Token ID22536
Feature activation-5.42e-01

Pos logit contributions
 goalie+1.125
irms+1.124
awks+1.043
 villains+1.036
cer+1.005
Neg logit contributions
pron-1.460
 recipe-1.451
ons-1.333
 mixer-1.280
 baking-1.224
Token people
Token ID661
Feature activation-6.15e-01

Pos logit contributions
 goalie+0.862
irms+0.858
 villains+0.799
awks+0.794
 thrill+0.756
Neg logit contributions
pron-1.065
 recipe-1.055
ons-0.990
 mixer-0.924
 baking-0.887
Token left
Token ID1364
Feature activation-1.23e+00

Pos logit contributions
 goalie+0.989
irms+0.986
 villains+0.918
awks+0.911
ashes+0.868
Neg logit contributions
pron-1.206
 recipe-1.188
ons-1.129
 mixer-1.046
 baking-1.007
Token.
Token ID13
Feature activation-3.81e+00

Pos logit contributions
 goalie+2.244
irms+2.229
awks+2.100
 villains+2.099
 thrill+2.022
Neg logit contributions
 recipe-2.160
pron-2.135
ons-2.027
 mixer-1.877
 baking-1.826
Token Anna
Token ID11735
Feature activation-6.22e-01

Pos logit contributions
cer+6.004
 goalie+5.721
awks+5.631
irms+5.627
 afloat+5.512
Neg logit contributions
 recipe-8.336
pron-7.913
 mixer-7.119
 flour-6.941
 baking-6.917
Token and
Token ID290
Feature activation-1.44e+00

Pos logit contributions
 goalie+1.093
irms+1.092
 villains+1.029
awks+1.016
 thrill+0.986
Neg logit contributions
pron-1.114
 recipe-1.099
ons-1.030
 mixer-0.946
 baking-0.904
Token Ben
Token ID3932
Feature activation-1.52e+00

Pos logit contributions
irms+1.683
 goalie+1.677
 villains+1.648
awks+1.557
ashes+1.522
Neg logit contributions
pron-3.333
 recipe-3.310
ons-3.157
 mixer-2.996
 baking-2.919
Token went
Token ID1816
Feature activation-1.46e+00

Pos logit contributions
irms+2.637
 goalie+2.631
 villains+2.460
awks+2.448
ashes+2.353
Neg logit contributions
pron-2.751
 recipe-2.708
ons-2.615
 mixer-2.396
 baking-2.286
Token back
Token ID736
Feature activation+1.54e-01

Pos logit contributions
 goalie+3.258
irms+3.193
awks+3.026
 villains+3.023
 thrill+3.000
Neg logit contributions
 recipe-2.012
pron-1.934
ons-1.785
 mixer-1.628
 baking-1.603
Token the
Token ID262
Feature activation-3.30e-01

Pos logit contributions
 goalie+2.690
irms+2.585
 villains+2.480
awks+2.456
 thrill+2.428
Neg logit contributions
pron-3.196
 recipe-3.188
ons-3.003
 mixer-2.804
 baking-2.700
Token bug
Token ID5434
Feature activation-1.13e+00

Pos logit contributions
 goalie+0.604
irms+0.603
 villains+0.571
awks+0.563
 thrill+0.546
Neg logit contributions
pron-0.574
 recipe-0.569
ons-0.526
 mixer-0.487
 baking-0.462
Token started
Token ID2067
Feature activation-8.97e-01

Pos logit contributions
 goalie+1.894
irms+1.859
 villains+1.728
awks+1.697
cer+1.676
Neg logit contributions
pron-2.109
 recipe-2.109
ons-2.021
 mixer-1.844
 baking-1.777
Token to
Token ID284
Feature activation-1.07e+00

Pos logit contributions
 goalie+2.311
irms+2.295
 villains+2.212
awks+2.179
 thrill+2.159
Neg logit contributions
 recipe-0.889
pron-0.846
ons-0.745
 mixer-0.660
 baking-0.632
Token fight
Token ID1907
Feature activation-2.06e+00

Pos logit contributions
 goalie+1.585
irms+1.565
 villains+1.484
awks+1.424
cer+1.407
Neg logit contributions
 recipe-2.211
pron-2.193
ons-2.016
 mixer-1.936
 baking-1.900
Token.
Token ID13
Feature activation-3.81e+00

Pos logit contributions
irms+3.489
 goalie+3.481
 villains+3.397
awks+3.336
 thrill+3.268
Neg logit contributions
 recipe-3.710
ons-3.540
pron-3.519
 mixer-3.200
 baking-3.190
Token They
Token ID1119
Feature activation-2.11e+00

Pos logit contributions
cer+5.577
irms+5.329
 thrill+5.233
 goalie+4.915
awks+4.803
Neg logit contributions
 recipe-8.332
pron-7.656
 mixer-7.380
ons-7.268
 baking-7.235
Token shouted
Token ID17293
Feature activation-7.94e-01

Pos logit contributions
 goalie+3.838
irms+3.810
awks+3.544
ashes+3.491
 villains+3.488
Neg logit contributions
pron-3.616
ons-3.504
 recipe-3.490
 mixer-3.162
 comb-3.029
Token and
Token ID290
Feature activation-2.08e+00

Pos logit contributions
 goalie+1.475
irms+1.459
 villains+1.397
awks+1.370
cer+1.341
Neg logit contributions
pron-1.320
 recipe-1.319
ons-1.220
 mixer-1.134
 baking-1.091
Token chased
Token ID26172
Feature activation-2.18e+00

Pos logit contributions
 goalie+3.633
irms+3.428
cer+3.255
 villains+3.237
awks+3.234
Neg logit contributions
pron-3.845
 recipe-3.771
ons-3.597
 mixer-3.394
 baking-3.251
Token each
Token ID1123
Feature activation-8.80e-01

Pos logit contributions
 goalie+3.926
irms+3.829
 villains+3.811
 thrill+3.787
cer+3.604
Neg logit contributions
 recipe-3.590
ons-3.528
pron-3.315
 mixer-3.183
 baking-2.996
Token glad
Token ID9675
Feature activation-1.53e+00

Pos logit contributions
irms+3.014
 goalie+2.994
awks+2.832
 villains+2.793
ashes+2.770
Neg logit contributions
pron-3.707
 recipe-3.594
ons-3.296
 baking-3.224
 mixer-3.165
Token to
Token ID284
Feature activation-8.46e-01

Pos logit contributions
irms+3.185
 goalie+3.175
awks+2.969
 villains+2.940
cer+2.922
Neg logit contributions
 recipe-2.376
pron-2.272
ons-2.151
 mixer-1.994
 baking-1.954
Token see
Token ID766
Feature activation-1.25e+00

Pos logit contributions
irms+1.226
 goalie+1.225
 villains+1.119
awks+1.111
cer+1.082
Neg logit contributions
 recipe-1.775
pron-1.774
ons-1.666
 mixer-1.556
 baking-1.525
Token the
Token ID262
Feature activation-1.19e+00

Pos logit contributions
 goalie+2.072
irms+2.034
 villains+1.938
awks+1.882
 thrill+1.847
Neg logit contributions
 recipe-2.387
pron-2.349
ons-2.300
 mixer-2.148
 baking-2.041
Token cops
Token ID14073
Feature activation-1.67e+00

Pos logit contributions
 goalie+2.045
irms+2.032
 villains+1.878
awks+1.865
cer+1.862
Neg logit contributions
 recipe-2.295
pron-2.255
ons-2.058
 mixer-1.976
 baking-1.901
Token.
Token ID13
Feature activation-3.78e+00

Pos logit contributions
irms+2.729
 goalie+2.623
awks+2.474
 villains+2.466
 thrill+2.361
Neg logit contributions
 recipe-3.321
pron-3.239
ons-3.084
 mixer-2.889
 baking-2.865
Token They
Token ID1119
Feature activation-2.29e+00

Pos logit contributions
cer+5.750
irms+5.526
 goalie+5.441
 thrill+5.256
awks+5.208
Neg logit contributions
 recipe-7.875
pron-7.397
 mixer-7.237
 flour-7.099
 baking-6.891
Token are
Token ID389
Feature activation-1.70e+00

Pos logit contributions
irms+3.996
 goalie+3.986
cer+3.800
 villains+3.731
awks+3.644
Neg logit contributions
 recipe-4.144
pron-4.125
ons-3.853
 mixer-3.594
 baking-3.453
Token sorry
Token ID7926
Feature activation-1.21e+00

Pos logit contributions
irms+2.822
 goalie+2.794
awks+2.645
 villains+2.616
ashes+2.597
Neg logit contributions
pron-3.361
 recipe-3.251
ons-2.995
 baking-2.875
 mixer-2.861
Token to
Token ID284
Feature activation-1.06e+00

Pos logit contributions
 goalie+2.274
irms+2.263
awks+2.123
cer+2.073
 thrill+2.071
Neg logit contributions
 recipe-2.048
pron-2.006
ons-1.885
 mixer-1.764
 baking-1.702
Token be
Token ID307
Feature activation-2.21e+00

Pos logit contributions
irms+1.701
 goalie+1.699
 villains+1.586
awks+1.554
 thrill+1.542
Neg logit contributions
 recipe-2.029
pron-2.027
ons-1.906
 mixer-1.799
 baking-1.748
Token sad
Token ID6507
Feature activation-1.66e+00

Pos logit contributions
irms+4.157
 goalie+4.080
cer+3.946
awks+3.887
ashes+3.836
Neg logit contributions
pron-4.677
 recipe-4.529
ons-4.461
 baking-4.282
 brushing-4.095
Token.
Token ID13
Feature activation-3.60e+00

Pos logit contributions
 goalie+2.780
irms+2.708
awks+2.540
 villains+2.534
cer+2.467
Neg logit contributions
 recipe-3.200
pron-3.076
ons-2.947
 mixer-2.783
 baking-2.737
Token They
Token ID1119
Feature activation-3.05e+00

Pos logit contributions
 thrill+5.267
irms+5.135
 goalie+5.130
cer+5.049
 villains+4.923
Neg logit contributions
 recipe-7.572
 mixer-7.187
pron-6.945
 flour-6.670
 baking-6.603
Token feel
Token ID1254
Feature activation-3.45e+00

Pos logit contributions
 goalie+5.812
cer+5.730
irms+5.582
mer+5.227
 teammate+5.172
Neg logit contributions
pron-5.059
 recipe-4.964
ons-4.777
 comb-4.514
 mixer-4.455
Token angry
Token ID7954
Feature activation-2.75e+00

Pos logit contributions
 goalie+6.579
cer+6.189
irms+5.755
 teammate+5.637
mer+5.630
Neg logit contributions
ons-6.057
pron-5.919
 recipe-5.772
 baking-5.693
 butter-5.501
Token and
Token ID290
Feature activation-3.78e+00

Pos logit contributions
irms+5.636
 goalie+5.596
 thrill+5.441
awks+5.310
 villains+5.307
Neg logit contributions
 recipe-4.132
ons-3.997
 baking-3.600
pron-3.558
 mixer-3.527
Token sad
Token ID6507
Feature activation-2.00e+00

Pos logit contributions
 goalie+6.781
cer+6.605
irms+6.041
 teammate+5.754
 villains+5.530
Neg logit contributions
 recipe-7.069
pron-7.042
ons-6.791
 baking-6.785
 mixer-6.497
Token.
Token ID13
Feature activation-3.02e+00

Pos logit contributions
 goalie+3.478
irms+3.341
awks+3.157
cer+3.117
 villains+3.082
Neg logit contributions
 recipe-3.846
pron-3.614
ons-3.494
 mixer-3.305
 baking-3.217
Token They
Token ID1119
Feature activation-2.94e+00

Pos logit contributions
 goalie+4.507
 thrill+4.445
irms+4.399
awks+4.166
cer+4.166
Neg logit contributions
 recipe-6.167
pron-5.961
 mixer-5.877
ons-5.744
 flour-5.456
Token have
Token ID423
Feature activation-1.66e+00

Pos logit contributions
 goalie+5.541
cer+5.415
irms+5.270
mer+5.018
 villains+4.981
Neg logit contributions
pron-4.912
 recipe-4.908
ons-4.642
 mixer-4.396
 comb-4.356
Token no
Token ID645
Feature activation-1.12e+00

Pos logit contributions
irms+2.983
 goalie+2.971
awks+2.822
 thrill+2.792
cer+2.766
Neg logit contributions
 recipe-2.905
pron-2.821
ons-2.704
 mixer-2.486
 baking-2.413
Token cart
Token ID6383
Feature activation+7.21e-01

Pos logit contributions
pron+1.806
 recipe+1.758
ons+1.703
 mixer+1.387
 baking+1.373
Neg logit contributions
 goalie-2.258
irms-2.257
awks-2.098
 thrill-2.086
 villains-2.082
Token!
Token ID0
Feature activation-3.52e-01

Pos logit contributions
pron+1.564
 recipe+1.543
ons+1.427
 mixer+1.354
 baking+1.297
Neg logit contributions
 goalie-0.994
irms-0.956
 villains-0.905
cer-0.874
awks-0.870
Token I
Token ID314
Feature activation-1.49e+00

Pos logit contributions
 goalie+0.544
irms+0.540
 villains+0.505
awks+0.499
 thrill+0.482
Neg logit contributions
pron-0.709
 recipe-0.700
ons-0.654
 mixer-0.613
 baking-0.588
Token'll
Token ID1183
Feature activation+4.04e-01

Pos logit contributions
 goalie+2.708
irms+2.679
awks+2.486
 villains+2.485
ashes+2.328
Neg logit contributions
pron-2.764
 recipe-2.690
ons-2.475
 mixer-2.312
 baking-2.243
Token always
Token ID1464
Feature activation-1.06e-01

Pos logit contributions
pron+0.754
 recipe+0.748
ons+0.697
 mixer+0.643
 baking+0.609
Neg logit contributions
 goalie-0.681
irms-0.664
 villains-0.641
awks-0.621
 thrill-0.614
Token remember
Token ID3505
Feature activation+2.69e+00

Pos logit contributions
 goalie+0.178
irms+0.176
 villains+0.166
awks+0.163
 thrill+0.158
Neg logit contributions
pron-0.200
 recipe-0.198
ons-0.184
 mixer-0.171
 baking-0.164
Token to
Token ID284
Feature activation+5.82e-02

Pos logit contributions
pron+5.736
 recipe+5.577
ons+4.980
 mixer+4.913
innie+4.637
Neg logit contributions
 goalie-3.775
irms-3.774
 villains-3.649
awks-3.490
cer-3.485
Token take
Token ID1011
Feature activation+6.69e-02

Pos logit contributions
pron+0.107
 recipe+0.106
ons+0.098
 mixer+0.091
 baking+0.086
Neg logit contributions
 goalie-0.100
irms-0.099
 villains-0.094
awks-0.092
 thrill-0.090
Token good
Token ID922
Feature activation-6.50e-01

Pos logit contributions
pron+0.138
 recipe+0.137
ons+0.127
 mixer+0.119
 baking+0.115
Neg logit contributions
 goalie-0.100
irms-0.100
 villains-0.093
awks-0.092
 thrill-0.088
Token care
Token ID1337
Feature activation+1.25e+00

Pos logit contributions
 goalie+0.801
irms+0.767
 villains+0.724
awks+0.699
 thrill+0.670
Neg logit contributions
 recipe-1.526
pron-1.520
ons-1.461
 mixer-1.366
 baking-1.322
Token of
Token ID286
Feature activation+1.61e+00

Pos logit contributions
pron+2.364
 recipe+2.334
ons+2.164
 mixer+2.059
 baking+1.926
Neg logit contributions
 goalie-2.096
irms-1.998
 villains-1.949
cer-1.892
ashes-1.884
Token?"
Token ID1701
Feature activation+3.16e-01

Pos logit contributions
pron+3.029
 recipe+2.976
ons+2.729
 mixer+2.511
innie+2.480
Neg logit contributions
 goalie-3.187
irms-3.185
ashes-2.964
awks-2.962
cer-2.915
Token asked
Token ID1965
Feature activation+5.89e-01

Pos logit contributions
pron+0.731
 recipe+0.729
ons+0.681
 mixer+0.640
 baking+0.619
Neg logit contributions
irms-0.393
 goalie-0.387
 villains-0.360
awks-0.352
 thrill-0.333
Token Lily
Token ID20037
Feature activation+2.19e-01

Pos logit contributions
pron+1.414
 recipe+1.411
ons+1.315
 mixer+1.237
 baking+1.198
Neg logit contributions
irms-0.682
 goalie-0.673
 villains-0.609
awks-0.592
cer-0.568
Token.
Token ID13
Feature activation+2.59e+00

Pos logit contributions
pron+0.509
 recipe+0.503
ons+0.477
 mixer+0.447
 baking+0.431
Neg logit contributions
 goalie-0.270
irms-0.267
 villains-0.244
awks-0.239
 thrill-0.227
Token
Token ID220
Feature activation+1.07e+00

Pos logit contributions
 recipe+6.429
pron+6.157
ons+5.918
 mixer+5.240
 baking+5.165
Neg logit contributions
irms-3.307
 He-3.122
 goalie-3.008
awks-2.810
 villains-2.684
Token
Token ID198
Feature activation+2.49e+00

Pos logit contributions
pron+2.399
 recipe+2.392
ons+2.154
 mixer+2.074
 baking+2.013
Neg logit contributions
irms-1.441
 goalie-1.419
awks-1.301
 villains-1.281
ashes-1.257
Token
Token ID198
Feature activation+2.72e+00

Pos logit contributions
 recipe+6.706
pron+6.473
ons+5.887
 mixer+5.813
 baking+5.726
Neg logit contributions
irms-2.464
 goalie-2.246
 He-2.166
awks-2.081
 villains-1.995
Token"
Token ID1
Feature activation+1.60e+00

Pos logit contributions
 recipe+4.694
ons+4.238
pron+4.154
 mixer+3.794
 baking+3.651
Neg logit contributions
 goalie-5.399
irms-5.348
cer-5.016
 villains-4.989
ashes-4.845
TokenSure
Token ID19457
Feature activation+5.40e-01

Pos logit contributions
 recipe+3.614
pron+3.583
ons+3.575
 baking+3.152
 mixer+3.142
Neg logit contributions
 goalie-2.028
irms-1.993
 villains-1.897
ashes-1.815
 thrill-1.795
Token,
Token ID11
Feature activation+1.22e+00

Pos logit contributions
pron+0.878
 recipe+0.863
ons+0.798
 mixer+0.717
 baking+0.681
Neg logit contributions
 goalie-1.049
irms-1.041
 villains-0.999
awks-0.975
 thrill-0.957
Token sweet
Token ID6029
Feature activation+7.62e-01

Pos logit contributions
pron+2.620
 recipe+2.619
ons+2.507
 mixer+2.288
 baking+2.191
Neg logit contributions
 goalie-1.676
irms-1.646
 villains-1.573
awks-1.485
 thrill-1.476
Token big
Token ID1263
Feature activation-8.10e-01

Pos logit contributions
pron+0.411
 recipe+0.409
ons+0.381
 mixer+0.355
 baking+0.341
Neg logit contributions
 goalie-0.296
irms-0.294
 villains-0.279
awks-0.271
 thrill-0.262
Token and
Token ID290
Feature activation+1.67e-01

Pos logit contributions
 goalie+1.435
irms+1.429
 villains+1.344
awks+1.341
 thrill+1.309
Neg logit contributions
 recipe-1.446
pron-1.436
ons-1.340
 mixer-1.245
 baking-1.195
Token strong
Token ID1913
Feature activation-1.12e+00

Pos logit contributions
pron+0.343
 recipe+0.341
ons+0.319
 mixer+0.296
 baking+0.284
Neg logit contributions
 goalie-0.251
irms-0.249
 villains-0.235
awks-0.230
 thrill-0.222
Token.
Token ID13
Feature activation-8.62e-02

Pos logit contributions
irms+1.794
 goalie+1.784
awks+1.688
 villains+1.669
 thrill+1.633
Neg logit contributions
 recipe-2.171
pron-2.147
ons-2.034
 mixer-1.906
 baking-1.841
Token I
Token ID314
Feature activation-1.11e+00

Pos logit contributions
 goalie+0.125
irms+0.124
 villains+0.116
awks+0.113
 thrill+0.109
Neg logit contributions
pron-0.182
 recipe-0.180
ons-0.169
 mixer-0.158
 baking-0.152
Token want
Token ID765
Feature activation+2.69e+00

Pos logit contributions
 goalie+1.673
irms+1.662
 villains+1.543
awks+1.528
cer+1.450
Neg logit contributions
pron-2.324
 recipe-2.284
ons-2.124
 mixer-2.022
 baking-1.950
Token to
Token ID284
Feature activation+8.08e-01

Pos logit contributions
pron+5.178
 recipe+4.678
ons+4.134
innie+4.023
itten+3.872
Neg logit contributions
 goalie-4.819
 villains-4.783
irms-4.708
 sailors-4.572
ks-4.408
Token try
Token ID1949
Feature activation+2.56e+00

Pos logit contributions
pron+1.786
 recipe+1.771
ons+1.651
 mixer+1.542
 baking+1.478
Neg logit contributions
 goalie-1.122
irms-1.062
 villains-1.040
awks-0.987
 thrill-0.973
Token it
Token ID340
Feature activation+1.50e+00

Pos logit contributions
pron+5.487
 recipe+5.416
ons+4.834
innie+4.618
 mixer+4.604
Neg logit contributions
 goalie-3.956
irms-3.857
 villains-3.568
awks-3.558
ks-3.512
Token."
Token ID526
Feature activation-7.95e-01

Pos logit contributions
pron+3.740
 recipe+3.638
ons+3.447
 mixer+3.269
innie+3.155
Neg logit contributions
 goalie-1.783
irms-1.660
 villains-1.570
ashes-1.525
awks-1.514
Token
Token ID198
Feature activation-7.51e-02

Pos logit contributions
 goalie+1.373
irms+1.358
 villains+1.282
awks+1.275
 thrill+1.237
Neg logit contributions
pron-1.493
 recipe-1.468
ons-1.356
 mixer-1.275
 baking-1.204
Token a
Token ID257
Feature activation+6.77e-01

Pos logit contributions
 goalie+0.625
irms+0.618
 villains+0.587
awks+0.579
 thrill+0.562
Neg logit contributions
pron-0.590
 recipe-0.584
ons-0.544
 mixer-0.501
 baking-0.476
Token toy
Token ID13373
Feature activation-5.04e-01

Pos logit contributions
 recipe+1.272
pron+1.269
ons+1.235
 mixer+1.091
 baking+1.051
Neg logit contributions
 goalie-1.117
irms-1.061
 villains-1.022
 thrill-0.997
awks-0.993
Token that
Token ID326
Feature activation+2.77e-01

Pos logit contributions
irms+0.896
 goalie+0.886
 villains+0.838
awks+0.834
 thrill+0.805
Neg logit contributions
pron-0.899
 recipe-0.892
ons-0.826
 mixer-0.774
 baking-0.735
Token she
Token ID673
Feature activation-5.28e-01

Pos logit contributions
pron+0.464
 recipe+0.464
ons+0.423
 mixer+0.388
 baking+0.371
Neg logit contributions
 goalie-0.515
irms-0.510
 villains-0.488
awks-0.478
 thrill-0.468
Token really
Token ID1107
Feature activation-1.12e+00

Pos logit contributions
irms+0.848
 goalie+0.847
awks+0.783
 villains+0.782
ashes+0.755
Neg logit contributions
pron-1.052
 recipe-1.025
ons-0.967
 mixer-0.894
 baking-0.861
Token wanted
Token ID2227
Feature activation+2.50e+00

Pos logit contributions
 goalie+1.680
irms+1.656
 villains+1.518
awks+1.510
cer+1.436
Neg logit contributions
pron-2.348
 recipe-2.281
ons-2.153
 mixer-2.006
 baking-1.942
Token,
Token ID11
Feature activation+1.14e+00

Pos logit contributions
pron+5.646
 recipe+5.292
ons+4.741
 mixer+4.606
innie+4.455
Neg logit contributions
 villains-3.419
 goalie-3.393
ks-3.378
irms-3.373
 like-3.144
Token but
Token ID475
Feature activation+5.37e-01

Pos logit contributions
pron+2.542
 recipe+2.535
ons+2.389
 mixer+2.185
 baking+2.119
Neg logit contributions
 goalie-1.537
irms-1.464
 villains-1.410
awks-1.368
 thrill-1.323
Token her
Token ID607
Feature activation+3.07e-01

Pos logit contributions
 recipe+0.836
pron+0.828
ons+0.755
 mixer+0.678
 baking+0.640
Neg logit contributions
 goalie-1.093
irms-1.072
 villains-1.027
awks-1.002
cer-0.985
Token mom
Token ID1995
Feature activation+1.99e+00

Pos logit contributions
pron+0.482
 recipe+0.480
ons+0.441
 mixer+0.393
 baking+0.379
Neg logit contributions
 goalie-0.614
irms-0.603
 villains-0.577
awks-0.564
 thrill-0.556
Token said
Token ID531
Feature activation+2.00e+00

Pos logit contributions
 recipe+4.198
pron+4.080
ons+3.784
 mixer+3.512
 baking+3.423
Neg logit contributions
 goalie-2.823
irms-2.769
cer-2.617
awks-2.538
 thrill-2.491
Token cup
Token ID6508
Feature activation+1.70e+00

Pos logit contributions
pron+1.293
 recipe+1.258
ons+1.158
 mixer+1.098
 baking+1.032
Neg logit contributions
irms-1.102
 goalie-1.096
 villains-1.034
awks-1.020
 thrill-0.968
Tokencakes
Token ID37263
Feature activation+1.91e+00

Pos logit contributions
pron+2.189
 recipe+2.009
ons+1.779
 mixer+1.644
innie+1.534
Neg logit contributions
irms-3.782
 goalie-3.765
ashes-3.710
awks-3.597
 villains-3.581
Token.
Token ID13
Feature activation+2.20e+00

Pos logit contributions
pron+4.511
 recipe+4.282
ons+4.173
innie+3.912
 mixer+3.891
Neg logit contributions
 goalie-2.236
ashes-2.108
irms-2.107
awks-1.936
 thrill-1.892
Token She
Token ID1375
Feature activation+2.53e+00

Pos logit contributions
pron+5.635
 recipe+5.199
ons+5.133
 mixer+4.590
 baking+4.454
Neg logit contributions
 goalie-2.544
irms-2.514
 villains-2.473
 He-2.404
ashes-2.387
Token put
Token ID1234
Feature activation+2.67e+00

Pos logit contributions
 recipe+5.633
pron+5.557
ons+5.161
 mixer+4.842
 baking+4.748
Neg logit contributions
 goalie-3.140
 villains-3.039
irms-2.981
cer-2.981
 pirate-2.866
Token lots
Token ID6041
Feature activation+2.59e+00

Pos logit contributions
 recipe+6.023
pron+5.996
ons+5.296
innie+5.104
ama+4.995
Neg logit contributions
irms-3.408
 goalie-3.077
ashes-3.068
awks-3.025
 thrill-2.926
Token of
Token ID286
Feature activation+2.43e+00

Pos logit contributions
pron+5.701
 recipe+5.526
ons+4.812
innie+4.794
 mixer+4.712
Neg logit contributions
ashes-3.676
irms-3.495
 thrill-3.247
vious-3.220
awks-3.193
Token tasty
Token ID25103
Feature activation+3.32e+00

Pos logit contributions
pron+4.651
 recipe+4.456
ama+3.759
ons+3.703
 mixer+3.640
Neg logit contributions
irms-4.161
awks-3.863
ashes-3.757
 thrill-3.715
 villains-3.474
Token ingredients
Token ID9391
Feature activation+2.81e+00

Pos logit contributions
pron+6.875
 recipe+6.081
ama+5.484
innie+5.421
lf+5.098
Neg logit contributions
ashes-5.664
irms-5.587
awks-5.185
gers-5.071
 adventurers-4.808
Token into
Token ID656
Feature activation+3.02e+00

Pos logit contributions
pron+6.683
 recipe+6.239
ons+5.791
innie+5.610
 mixer+5.503
Neg logit contributions
ashes-3.312
 goalie-3.230
irms-3.225
ks-3.209
cer-3.173
Token a
Token ID257
Feature activation+2.80e+00

Pos logit contributions
pron+6.996
 recipe+6.545
ons+6.255
itten+5.888
innie+5.887
Neg logit contributions
irms-3.745
 goalie-3.529
ks-3.525
awks-3.438
 thrill-3.405
Token.
Token ID13
Feature activation+1.25e+00

Pos logit contributions
pron+0.478
 recipe+0.469
ons+0.440
 mixer+0.413
 baking+0.397
Neg logit contributions
 goalie-0.306
irms-0.301
 villains-0.283
awks-0.276
ashes-0.267
Token Sue
Token ID26089
Feature activation+1.53e+00

Pos logit contributions
pron+3.049
 recipe+2.894
ons+2.703
 mixer+2.512
 baking+2.446
Neg logit contributions
irms-1.629
 goalie-1.595
awks-1.506
ashes-1.499
 villains-1.427
Token told
Token ID1297
Feature activation+1.12e+00

Pos logit contributions
 recipe+3.563
pron+3.547
ons+3.133
 mixer+2.993
 baking+2.991
Neg logit contributions
 goalie-1.994
irms-1.962
 villains-1.824
awks-1.776
cer-1.729
Token her
Token ID607
Feature activation+1.26e+00

Pos logit contributions
pron+2.094
 recipe+2.010
ons+1.774
 mixer+1.665
 baking+1.581
Neg logit contributions
irms-2.077
 goalie-2.038
awks-1.950
 villains-1.901
 thrill-1.868
Token mom
Token ID1995
Feature activation+1.85e+00

Pos logit contributions
pron+2.256
 recipe+2.155
ons+1.991
 mixer+1.806
 baking+1.755
Neg logit contributions
 goalie-2.333
 villains-2.252
irms-2.197
awks-2.050
 thrill-2.040
Token,
Token ID11
Feature activation+1.86e+00

Pos logit contributions
pron+4.061
 recipe+3.775
ons+3.459
 mixer+3.247
innie+3.200
Neg logit contributions
 goalie-2.707
irms-2.622
ks-2.465
 villains-2.458
awks-2.354
Token "
Token ID366
Feature activation+1.17e+00

Pos logit contributions
pron+3.177
 recipe+3.052
ons+2.917
 mixer+2.371
innie+2.335
Neg logit contributions
 goalie-3.587
irms-3.534
 villains-3.365
awks-3.292
ks-3.239
TokenI
Token ID40
Feature activation+3.43e-01

Pos logit contributions
 recipe+2.517
pron+2.472
ons+2.349
 mixer+2.161
 baking+2.100
Neg logit contributions
 goalie-1.756
irms-1.694
awks-1.570
cer-1.527
 villains-1.516
Token want
Token ID765
Feature activation+2.74e+00

Pos logit contributions
pron+0.839
 recipe+0.835
ons+0.791
 mixer+0.742
 baking+0.719
Neg logit contributions
 goalie-0.389
irms-0.382
 villains-0.347
awks-0.342
ashes-0.327
Token to
Token ID284
Feature activation+9.83e-01

Pos logit contributions
pron+5.639
 recipe+5.204
innie+4.504
ons+4.409
ama+4.146
Neg logit contributions
 goalie-5.129
irms-4.808
awks-4.569
ashes-4.538
 villains-4.536
Token swing
Token ID9628
Feature activation+2.86e-01

Pos logit contributions
pron+1.926
 recipe+1.903
ons+1.748
 mixer+1.602
innie+1.546
Neg logit contributions
 goalie-1.668
irms-1.581
 villains-1.495
awks-1.487
ashes-1.481
Token by
Token ID416
Feature activation+5.66e-01

Pos logit contributions
pron+1.050
 recipe+1.032
ons+0.977
 mixer+0.905
 baking+0.868
Neg logit contributions
 goalie-0.723
irms-0.696
 villains-0.659
awks-0.635
cer-0.628
Token one
Token ID530
Feature activation+6.52e-01

Pos logit contributions
pron+1.238
 recipe+1.210
ons+1.149
 mixer+1.067
 baking+1.015
Neg logit contributions
 goalie-0.799
irms-0.774
 villains-0.733
awks-0.699
cer-0.696
Token with
Token ID351
Feature activation+2.83e-01

Pos logit contributions
pron+1.495
 recipe+1.460
ons+1.391
 mixer+1.296
 baking+1.244
Neg logit contributions
 goalie-0.851
irms-0.811
 villains-0.759
awks-0.734
 thrill-0.724
Token determination
Token ID12123
Feature activation+5.03e-01

Pos logit contributions
pron+0.669
 recipe+0.663
ons+0.622
 mixer+0.588
 baking+0.567
Neg logit contributions
 goalie-0.341
irms-0.336
 villains-0.310
awks-0.300
 thrill-0.291
Token.
Token ID13
Feature activation-3.36e-01

Pos logit contributions
pron+1.157
 recipe+1.136
ons+1.070
 mixer+1.015
 baking+0.972
Neg logit contributions
 goalie-0.628
irms-0.612
 villains-0.575
awks-0.549
 thrill-0.536
Token
Token ID198
Feature activation+7.39e-01

Pos logit contributions
 goalie+0.480
irms+0.474
 villains+0.445
awks+0.437
 thrill+0.419
Neg logit contributions
pron-0.722
 recipe-0.715
ons-0.674
 mixer-0.632
 baking-0.608
Token
Token ID198
Feature activation+7.27e-01

Pos logit contributions
pron+1.579
 recipe+1.576
ons+1.437
 mixer+1.384
 baking+1.330
Neg logit contributions
 goalie-1.028
irms-1.023
 villains-0.964
awks-0.929
ashes-0.895
TokenAfter
Token ID3260
Feature activation+8.69e-01

Pos logit contributions
 recipe+1.308
pron+1.297
ons+1.208
 mixer+1.136
 baking+1.074
Neg logit contributions
 goalie-1.299
irms-1.245
 villains-1.232
 thrill-1.156
awks-1.153
Token he
Token ID339
Feature activation+1.88e-01

Pos logit contributions
pron+1.899
 recipe+1.867
ons+1.720
 mixer+1.631
 baking+1.516
Neg logit contributions
 goalie-1.204
 villains-1.167
irms-1.160
 thrill-1.055
awks-1.036
Token finished
Token ID5201
Feature activation+2.11e+00

Pos logit contributions
pron+0.333
 recipe+0.328
ons+0.302
 mixer+0.279
 baking+0.265
Neg logit contributions
 goalie-0.339
irms-0.334
 villains-0.320
awks-0.310
 thrill-0.304
Token all
Token ID477
Feature activation+5.08e-01

Pos logit contributions
pron+4.942
 recipe+4.843
ons+4.505
 mixer+4.222
itten+4.212
Neg logit contributions
irms-2.471
 goalie-2.469
 villains-2.212
cer-2.112
 pirate-2.100
Token.
Token ID13
Feature activation+1.94e-01

Pos logit contributions
pron+1.344
 recipe+1.322
ons+1.212
 mixer+1.150
innie+1.070
Neg logit contributions
 goalie-1.211
irms-1.181
 villains-1.140
awks-1.121
cer-1.097
Token The
Token ID383
Feature activation-9.40e-02

Pos logit contributions
pron+0.428
 recipe+0.425
ons+0.399
 mixer+0.375
 baking+0.361
Neg logit contributions
 goalie-0.263
irms-0.259
 villains-0.241
awks-0.238
 thrill-0.226
Token fairy
Token ID25607
Feature activation+1.31e+00

Pos logit contributions
 goalie+0.153
irms+0.151
 villains+0.143
awks+0.140
 thrill+0.136
Neg logit contributions
pron-0.184
 recipe-0.182
ons-0.170
 mixer-0.158
 baking-0.151
Token waved
Token ID26834
Feature activation-1.19e+00

Pos logit contributions
 recipe+2.623
pron+2.622
ons+2.355
 mixer+2.316
 baking+2.170
Neg logit contributions
 goalie-1.967
 villains-1.884
irms-1.867
cer-1.819
awks-1.791
Token her
Token ID607
Feature activation+7.51e-01

Pos logit contributions
 goalie+2.304
irms+2.281
 villains+2.192
awks+2.093
ashes+2.082
Neg logit contributions
pron-1.921
 recipe-1.899
ons-1.754
 mixer-1.600
 baking-1.546
Token wand
Token ID11569
Feature activation+5.40e-01

Pos logit contributions
pron+1.607
 recipe+1.586
ons+1.517
 mixer+1.402
 baking+1.331
Neg logit contributions
 goalie-1.084
irms-1.046
 villains-0.981
awks-0.973
 thrill-0.964
Token and
Token ID290
Feature activation+6.40e-02

Pos logit contributions
pron+1.081
 recipe+1.069
ons+0.996
 mixer+0.937
 baking+0.888
Neg logit contributions
 goalie-0.838
irms-0.820
 villains-0.774
awks-0.766
 thrill-0.750
Token the
Token ID262
Feature activation-1.00e-01

Pos logit contributions
pron+0.122
 recipe+0.121
ons+0.112
 mixer+0.104
 baking+0.099
Neg logit contributions
 goalie-0.106
irms-0.105
 villains-0.100
awks-0.098
 thrill-0.095
Token bunny
Token ID44915
Feature activation+3.29e-01

Pos logit contributions
 goalie+0.165
irms+0.163
 villains+0.154
awks+0.151
 thrill+0.147
Neg logit contributions
pron-0.194
 recipe-0.191
ons-0.179
 mixer-0.166
 baking-0.159
Token grew
Token ID6348
Feature activation-5.35e-02

Pos logit contributions
pron+0.650
 recipe+0.641
ons+0.594
 mixer+0.561
 baking+0.532
Neg logit contributions
 goalie-0.521
irms-0.510
 villains-0.491
awks-0.476
 thrill-0.465
Token back
Token ID736
Feature activation+2.51e-01

Pos logit contributions
 goalie+0.085
irms+0.084
 villains+0.079
awks+0.078
 thrill+0.075
Neg logit contributions
pron-0.106
 recipe-0.105
ons-0.098
 mixer-0.091
 baking-0.087
Token in
Token ID287
Feature activation+2.26e-01

Pos logit contributions
pron+2.696
 recipe+2.592
ons+2.473
 mixer+2.262
innie+2.176
Neg logit contributions
irms-1.937
 goalie-1.905
awks-1.801
 villains-1.754
 thrill-1.697
Token the
Token ID262
Feature activation+3.08e-01

Pos logit contributions
pron+0.427
 recipe+0.417
ons+0.387
 mixer+0.358
 baking+0.341
Neg logit contributions
 goalie-0.383
irms-0.383
 villains-0.359
awks-0.356
 thrill-0.344
Token next
Token ID1306
Feature activation+1.44e+00

Pos logit contributions
pron+0.556
 recipe+0.551
ons+0.516
 mixer+0.465
 baking+0.448
Neg logit contributions
 goalie-0.541
irms-0.539
 villains-0.508
awks-0.504
 thrill-0.489
Token town
Token ID3240
Feature activation+4.89e-01

Pos logit contributions
pron+2.522
ons+2.377
 recipe+2.367
 mixer+1.969
innie+1.960
Neg logit contributions
 goalie-2.581
irms-2.506
 villains-2.414
ashes-2.366
awks-2.325
Token.
Token ID13
Feature activation+1.26e+00

Pos logit contributions
pron+1.069
 recipe+1.038
ons+0.985
 mixer+0.918
 baking+0.872
Neg logit contributions
 goalie-0.686
irms-0.661
 villains-0.629
awks-0.608
ashes-0.594
Token
Token ID198
Feature activation+1.06e+00

Pos logit contributions
pron+2.920
 recipe+2.863
ons+2.691
 mixer+2.465
 baking+2.381
Neg logit contributions
 goalie-1.649
irms-1.641
awks-1.506
ashes-1.472
 villains-1.463
Token
Token ID198
Feature activation+1.53e+00

Pos logit contributions
pron+2.375
 recipe+2.357
ons+2.138
 mixer+2.073
 baking+2.003
Neg logit contributions
 goalie-1.378
irms-1.364
 villains-1.302
awks-1.245
ashes-1.194
TokenWhen
Token ID2215
Feature activation+1.42e+00

Pos logit contributions
 recipe+2.920
pron+2.858
ons+2.668
 mixer+2.540
 baking+2.382
Neg logit contributions
 goalie-2.621
 villains-2.527
irms-2.470
awks-2.325
 thrill-2.300
Token she
Token ID673
Feature activation+1.41e+00

Pos logit contributions
pron+2.779
 recipe+2.734
ons+2.558
 mixer+2.290
innie+2.229
Neg logit contributions
 goalie-2.318
irms-2.188
 villains-2.136
 thrill-2.085
awks-2.046
Token arrived
Token ID5284
Feature activation+9.63e-01

Pos logit contributions
 recipe+2.831
pron+2.799
ons+2.548
 mixer+2.371
innie+2.251
Neg logit contributions
 goalie-2.165
irms-2.114
 villains-1.996
awks-1.985
cer-1.978
Token,
Token ID11
Feature activation+7.63e-01

Pos logit contributions
pron+2.060
 recipe+2.015
ons+1.876
 mixer+1.723
innie+1.636
Neg logit contributions
 goalie-1.434
irms-1.415
awks-1.304
 villains-1.268
cer-1.257
Token photo
Token ID4590
Feature activation+1.53e+00

Pos logit contributions
pron+2.996
 recipe+2.778
ons+2.598
innie+2.338
 mixer+2.323
Neg logit contributions
irms-2.488
 goalie-2.419
 villains-2.407
awks-2.311
ashes-2.238
Token albums
Token ID16788
Feature activation+1.50e+00

Pos logit contributions
pron+3.288
 recipe+3.178
ons+2.914
 mixer+2.781
innie+2.750
Neg logit contributions
 goalie-2.112
irms-2.032
 villains-2.026
ashes-1.912
awks-1.870
Token.
Token ID13
Feature activation+1.63e+00

Pos logit contributions
pron+3.525
 recipe+3.363
ons+3.179
 mixer+3.002
innie+2.970
Neg logit contributions
 goalie-1.799
irms-1.705
 villains-1.645
ashes-1.588
awks-1.576
Token "
Token ID366
Feature activation+7.18e-01

Pos logit contributions
ons+3.227
pron+3.225
 recipe+3.219
 mixer+2.752
innie+2.647
Neg logit contributions
 goalie-2.506
irms-2.478
 villains-2.373
ashes-2.299
awks-2.266
TokenThese
Token ID4711
Feature activation+3.60e-01

Pos logit contributions
pron+1.587
 recipe+1.585
ons+1.502
 mixer+1.397
 baking+1.350
Neg logit contributions
 goalie-0.967
irms-0.926
 villains-0.890
awks-0.853
 thrill-0.830
Token are
Token ID389
Feature activation+2.10e+00

Pos logit contributions
pron+0.779
 recipe+0.760
ons+0.704
 mixer+0.662
 baking+0.637
Neg logit contributions
 goalie-0.510
irms-0.503
 villains-0.473
awks-0.468
ashes-0.449
Token pictures
Token ID5986
Feature activation+1.72e+00

Pos logit contributions
pron+4.704
 recipe+4.407
ons+4.200
 mixer+3.957
innie+3.814
Neg logit contributions
 goalie-2.948
 villains-2.844
irms-2.716
awks-2.637
ks-2.595
Token of
Token ID286
Feature activation+1.55e+00

Pos logit contributions
pron+3.418
 recipe+3.288
ons+2.872
 mixer+2.840
innie+2.840
Neg logit contributions
 goalie-2.740
irms-2.619
 villains-2.617
ashes-2.497
awks-2.486
Token when
Token ID618
Feature activation+7.50e-01

Pos logit contributions
pron+3.422
 recipe+3.229
ons+2.988
 mixer+2.826
innie+2.786
Neg logit contributions
 goalie-2.216
 villains-2.145
irms-2.123
awks-2.039
 thrill-1.941
Token I
Token ID314
Feature activation+5.44e-01

Pos logit contributions
pron+1.418
 recipe+1.386
ons+1.281
 mixer+1.186
innie+1.114
Neg logit contributions
 goalie-1.263
irms-1.226
 villains-1.212
awks-1.146
 thrill-1.112
Token was
Token ID373
Feature activation+7.90e-01

Pos logit contributions
pron+1.135
 recipe+1.126
ons+1.055
 mixer+0.986
 baking+0.938
Neg logit contributions
 goalie-0.789
irms-0.778
 villains-0.732
awks-0.726
ashes-0.702
Token gave
Token ID2921
Feature activation-2.43e-01

Pos logit contributions
pron+1.639
 recipe+1.639
ons+1.485
 mixer+1.401
 baking+1.320
Neg logit contributions
 goalie-1.475
irms-1.439
 villains-1.367
awks-1.341
cer-1.338
Token the
Token ID262
Feature activation-2.36e-01

Pos logit contributions
 goalie+0.420
irms+0.415
 villains+0.395
awks+0.387
 thrill+0.376
Neg logit contributions
pron-0.448
 recipe-0.444
ons-0.416
 mixer-0.383
 baking-0.366
Token gift
Token ID6979
Feature activation+3.91e-01

Pos logit contributions
 goalie+0.467
irms+0.464
 villains+0.442
awks+0.436
 thrill+0.423
Neg logit contributions
pron-0.379
 recipe-0.375
ons-0.343
 mixer-0.314
 baking-0.298
Token to
Token ID284
Feature activation-1.95e-02

Pos logit contributions
pron+0.635
 recipe+0.620
ons+0.565
 mixer+0.520
 baking+0.492
Neg logit contributions
 goalie-0.768
irms-0.747
 villains-0.721
awks-0.705
cer-0.689
Token Sam
Token ID3409
Feature activation+1.03e+00

Pos logit contributions
 goalie+0.025
irms+0.024
 villains+0.023
awks+0.022
 thrill+0.021
Neg logit contributions
pron-0.045
 recipe-0.045
ons-0.042
 mixer-0.040
 baking-0.038
Token,
Token ID11
Feature activation-5.05e-01

Pos logit contributions
pron+1.877
 recipe+1.817
ons+1.686
 mixer+1.560
 baking+1.451
Neg logit contributions
 goalie-1.792
irms-1.762
awks-1.653
 villains-1.651
ashes-1.611
Token Sam
Token ID3409
Feature activation+8.80e-01

Pos logit contributions
irms+0.759
 goalie+0.759
 villains+0.705
awks+0.702
ashes+0.677
Neg logit contributions
pron-1.041
 recipe-1.026
ons-0.975
 mixer-0.911
 baking-0.876
Token was
Token ID373
Feature activation+4.10e-02

Pos logit contributions
pron+1.655
 recipe+1.621
ons+1.484
 mixer+1.396
 baking+1.328
Neg logit contributions
 goalie-1.471
irms-1.431
 villains-1.364
awks-1.340
cer-1.319
Token very
Token ID845
Feature activation-4.36e-02

Pos logit contributions
pron+0.088
 recipe+0.088
ons+0.083
 mixer+0.077
 baking+0.074
Neg logit contributions
 goalie-0.058
irms-0.057
 villains-0.053
awks-0.052
 thrill-0.050
Token happy
Token ID3772
Feature activation-1.17e+00

Pos logit contributions
 goalie+0.055
irms+0.054
 villains+0.050
awks+0.049
 thrill+0.047
Neg logit contributions
pron-0.100
 recipe-0.100
ons-0.094
 mixer-0.089
 baking-0.085
Token too
Token ID1165
Feature activation-9.98e-01

Pos logit contributions
irms+2.039
 goalie+2.006
awks+1.915
 villains+1.877
 thrill+1.838
Neg logit contributions
 recipe-2.242
pron-2.194
ons-2.055
 mixer-1.906
 baking-1.870
Token was
Token ID373
Feature activation-7.06e-02

Pos logit contributions
pron+1.034
 recipe+1.023
ons+0.941
 mixer+0.877
 baking+0.841
Neg logit contributions
 goalie-0.890
irms-0.870
 villains-0.843
awks-0.807
 thrill-0.800
Token there
Token ID612
Feature activation+7.17e-01

Pos logit contributions
 goalie+0.115
irms+0.114
 villains+0.107
awks+0.105
 thrill+0.102
Neg logit contributions
pron-0.137
 recipe-0.135
ons-0.126
 mixer-0.117
 baking-0.112
Token.
Token ID13
Feature activation+4.71e-01

Pos logit contributions
pron+1.702
 recipe+1.668
ons+1.584
 mixer+1.487
innie+1.426
Neg logit contributions
 goalie-0.860
irms-0.820
 villains-0.781
awks-0.732
 thrill-0.726
Token He
Token ID679
Feature activation+3.90e-01

Pos logit contributions
pron+1.334
 recipe+1.319
ons+1.266
 mixer+1.191
 baking+1.160
Neg logit contributions
 goalie-0.358
irms-0.349
 villains-0.315
awks-0.291
ashes-0.267
Token was
Token ID373
Feature activation-1.37e-01

Pos logit contributions
 recipe+0.789
pron+0.787
ons+0.729
 mixer+0.680
 baking+0.654
Neg logit contributions
 goalie-0.598
irms-0.594
 villains-0.565
awks-0.549
 thrill-0.531
Token all
Token ID477
Feature activation-3.13e-01

Pos logit contributions
 goalie+0.189
irms+0.187
 villains+0.174
awks+0.171
 thrill+0.164
Neg logit contributions
pron-0.299
 recipe-0.296
ons-0.279
 mixer-0.261
 baking-0.252
Token alone
Token ID3436
Feature activation-1.05e+00

Pos logit contributions
 goalie+0.269
irms+0.264
 villains+0.235
awks+0.227
 thrill+0.209
Neg logit contributions
pron-0.852
 recipe-0.846
ons-0.805
 mixer-0.767
 baking-0.747
Token.
Token ID13
Feature activation+2.62e-01

Pos logit contributions
irms+1.793
 goalie+1.786
awks+1.686
 villains+1.665
ashes+1.614
Neg logit contributions
pron-1.932
 recipe-1.915
ons-1.795
 mixer-1.664
 baking-1.630
Token He
Token ID679
Feature activation-5.36e-01

Pos logit contributions
pron+0.744
 recipe+0.736
ons+0.703
 mixer+0.667
 baking+0.649
Neg logit contributions
 goalie-0.194
irms-0.191
 villains-0.172
awks-0.159
 thrill-0.145
Token started
Token ID2067
Feature activation+5.92e-01

Pos logit contributions
 goalie+0.929
irms+0.916
 villains+0.856
awks+0.853
ashes+0.830
Neg logit contributions
pron-0.990
 recipe-0.971
ons-0.912
 mixer-0.843
 baking-0.806
Token to
Token ID284
Feature activation-7.19e-01

Pos logit contributions
pron+0.797
 recipe+0.755
ons+0.698
 mixer+0.609
 baking+0.547
Neg logit contributions
 goalie-1.328
irms-1.325
 villains-1.278
awks-1.256
ashes-1.224
Token say
Token ID910
Feature activation-5.38e-02

Pos logit contributions
 goalie+0.146
irms+0.145
 villains+0.135
awks+0.133
 thrill+0.127
Neg logit contributions
pron-0.223
 recipe-0.221
ons-0.208
 mixer-0.194
 baking-0.187
Token anything
Token ID1997
Feature activation-1.49e-01

Pos logit contributions
 goalie+0.079
irms+0.078
 villains+0.074
awks+0.072
 thrill+0.069
Neg logit contributions
pron-0.112
 recipe-0.111
ons-0.104
 mixer-0.097
 baking-0.093
Token either
Token ID2035
Feature activation+1.65e-01

Pos logit contributions
 goalie+0.232
irms+0.230
 villains+0.216
awks+0.212
 thrill+0.205
Neg logit contributions
pron-0.299
 recipe-0.296
ons-0.277
 mixer-0.258
 baking-0.248
Token.
Token ID13
Feature activation-4.26e-01

Pos logit contributions
pron+0.364
 recipe+0.358
ons+0.337
 mixer+0.315
 baking+0.305
Neg logit contributions
 goalie-0.224
irms-0.219
 villains-0.205
awks-0.201
 thrill-0.193
Token Tom
Token ID4186
Feature activation-4.42e-01

Pos logit contributions
 goalie+0.625
irms+0.622
 villains+0.581
awks+0.572
 thrill+0.552
Neg logit contributions
pron-0.890
 recipe-0.879
ons-0.823
 mixer-0.774
 baking-0.743
Token was
Token ID373
Feature activation-1.21e+00

Pos logit contributions
 goalie+0.762
irms+0.756
 villains+0.717
awks+0.704
 thrill+0.684
Neg logit contributions
pron-0.812
 recipe-0.803
ons-0.748
 mixer-0.693
 baking-0.661
Token scared
Token ID12008
Feature activation-1.35e+00

Pos logit contributions
 goalie+1.494
irms+1.471
 villains+1.378
awks+1.334
cer+1.332
Neg logit contributions
pron-2.835
 recipe-2.802
ons-2.606
 mixer-2.513
 baking-2.489
Token.
Token ID13
Feature activation-1.41e+00

Pos logit contributions
 goalie+2.413
irms+2.368
 villains+2.270
awks+2.209
cer+2.161
Neg logit contributions
 recipe-2.425
pron-2.381
ons-2.278
 mixer-2.145
 baking-2.044
Token
Token ID198
Feature activation-2.19e-01

Pos logit contributions
irms+1.550
 goalie+1.538
awks+1.413
 villains+1.404
cer+1.370
Neg logit contributions
pron-3.481
 recipe-3.434
ons-3.230
 mixer-3.128
 baking-2.998
Token
Token ID198
Feature activation-8.04e-01

Pos logit contributions
 goalie+0.355
irms+0.351
 villains+0.331
awks+0.326
 thrill+0.315
Neg logit contributions
pron-0.427
 recipe-0.422
ons-0.395
 mixer-0.366
 baking-0.351
TokenTom
Token ID13787
Feature activation-3.42e-01

Pos logit contributions
irms+1.942
 goalie+1.935
awks+1.856
 villains+1.844
 thrill+1.809
Neg logit contributions
pron-0.943
 recipe-0.904
ons-0.825
 mixer-0.707
 baking-0.643
Token,
Token ID11
Feature activation-5.71e-02

Pos logit contributions
pron+1.914
 recipe+1.869
ons+1.666
 mixer+1.608
innie+1.522
Neg logit contributions
 goalie-1.498
irms-1.472
 villains-1.442
awks-1.383
cer-1.346
Token so
Token ID523
Feature activation-3.84e-01

Pos logit contributions
 goalie+0.092
irms+0.091
 villains+0.086
awks+0.085
 thrill+0.082
Neg logit contributions
pron-0.111
 recipe-0.110
ons-0.103
 mixer-0.096
 baking-0.092
Token she
Token ID673
Feature activation-1.61e-01

Pos logit contributions
 goalie+0.894
irms+0.887
 villains+0.849
awks+0.841
ashes+0.822
Neg logit contributions
pron-0.480
 recipe-0.471
ons-0.426
 mixer-0.373
 baking-0.352
Token threw
Token ID9617
Feature activation-4.01e-01

Pos logit contributions
 goalie+0.271
irms+0.269
 villains+0.254
awks+0.250
 thrill+0.242
Neg logit contributions
pron-0.303
 recipe-0.299
ons-0.279
 mixer-0.258
 baking-0.247
Token it
Token ID340
Feature activation+3.22e-01

Pos logit contributions
 goalie+0.800
irms+0.789
 villains+0.760
awks+0.742
 thrill+0.724
Neg logit contributions
pron-0.627
 recipe-0.621
ons-0.578
 mixer-0.525
 baking-0.496
Token away
Token ID1497
Feature activation-1.14e-01

Pos logit contributions
pron+0.712
 recipe+0.701
ons+0.658
 mixer+0.618
 baking+0.594
Neg logit contributions
 goalie-0.439
irms-0.432
 villains-0.400
awks-0.397
ashes-0.383
Token.
Token ID13
Feature activation+8.38e-03

Pos logit contributions
 goalie+0.186
irms+0.185
 villains+0.174
awks+0.171
 thrill+0.166
Neg logit contributions
pron-0.220
 recipe-0.217
ons-0.203
 mixer-0.188
 baking-0.180
Token
Token ID198
Feature activation+4.99e-01

Pos logit contributions
pron+0.017
 recipe+0.017
ons+0.016
 mixer+0.015
 baking+0.014
Neg logit contributions
 goalie-0.013
irms-0.012
 villains-0.012
awks-0.011
 thrill-0.011
Token
Token ID198
Feature activation+5.31e-01

Pos logit contributions
 recipe+1.038
pron+1.036
ons+0.949
 mixer+0.906
 baking+0.871
Neg logit contributions
 goalie-0.737
irms-0.734
 villains-0.693
awks-0.675
ashes-0.647
TokenFrom
Token ID4863
Feature activation-7.45e-01

Pos logit contributions
 recipe+0.972
pron+0.963
ons+0.889
 mixer+0.835
 baking+0.794
Neg logit contributions
 goalie-0.929
irms-0.908
 villains-0.880
awks-0.846
 thrill-0.823
Token that
Token ID326
Feature activation+1.28e+00

Pos logit contributions
 goalie+1.560
irms+1.532
 villains+1.487
awks+1.440
ashes+1.413
Neg logit contributions
pron-1.113
 recipe-1.084
ons-1.013
 mixer-0.914
 baking-0.865
Token saw
Token ID2497
Feature activation-4.65e-02

Pos logit contributions
pron+1.218
 recipe+1.213
ons+1.123
 mixer+1.053
 baking+0.994
Neg logit contributions
 goalie-0.873
irms-0.864
 villains-0.813
awks-0.798
cer-0.775
Token lots
Token ID6041
Feature activation-1.01e-01

Pos logit contributions
 goalie+0.073
irms+0.072
 villains+0.068
awks+0.067
 thrill+0.064
Neg logit contributions
pron-0.094
 recipe-0.093
ons-0.086
 mixer-0.080
 baking-0.077
Token of
Token ID286
Feature activation-6.06e-02

Pos logit contributions
 goalie+0.193
irms+0.192
 villains+0.182
awks+0.180
 thrill+0.175
Neg logit contributions
pron-0.166
 recipe-0.164
ons-0.151
 mixer-0.138
 baking-0.131
Token y
Token ID331
Feature activation+8.87e-01

Pos logit contributions
 goalie+0.099
irms+0.098
 villains+0.092
awks+0.091
 thrill+0.088
Neg logit contributions
pron-0.117
 recipe-0.116
ons-0.108
 mixer-0.100
 baking-0.096
Tokenummy
Token ID13513
Feature activation+7.31e-01

Pos logit contributions
pron+0.810
 recipe+0.728
ons+0.573
 mixer+0.498
 baking+0.413
Neg logit contributions
irms-2.415
 goalie-2.410
awks-2.334
ashes-2.302
 villains-2.278
Token treats
Token ID18432
Feature activation+6.36e-02

Pos logit contributions
pron+1.267
 recipe+1.197
ons+1.079
 mixer+1.009
 baking+0.942
Neg logit contributions
irms-1.349
 goalie-1.338
 villains-1.280
awks-1.273
ashes-1.250
Token!
Token ID0
Feature activation-2.37e-01

Pos logit contributions
pron+0.144
 recipe+0.141
ons+0.134
 mixer+0.126
 baking+0.121
Neg logit contributions
 goalie-0.083
irms-0.081
 villains-0.076
awks-0.074
ashes-0.071
Token The
Token ID383
Feature activation-2.39e-01

Pos logit contributions
 goalie+0.341
irms+0.337
 villains+0.315
awks+0.309
 thrill+0.298
Neg logit contributions
pron-0.504
 recipe-0.499
ons-0.471
 mixer-0.441
 baking-0.424
Token fly
Token ID6129
Feature activation+2.97e-01

Pos logit contributions
 goalie+0.389
irms+0.386
 villains+0.363
awks+0.358
 thrill+0.344
Neg logit contributions
pron-0.465
 recipe-0.460
ons-0.428
 mixer-0.399
 baking-0.384
Token couldn
Token ID3521
Feature activation+1.51e+00

Pos logit contributions
pron+0.540
 recipe+0.534
ons+0.494
 mixer+0.459
 baking+0.435
Neg logit contributions
 goalie-0.518
irms-0.503
 villains-0.486
awks-0.475
 thrill-0.465
Token't
Token ID470
Feature activation-3.03e-01

Pos logit contributions
pron+2.085
 recipe+2.032
ons+1.704
 mixer+1.576
 baking+1.406
Neg logit contributions
 goalie-3.390
irms-3.240
 villains-3.194
awks-3.147
ashes-3.134
Token!"
Token ID2474
Feature activation-1.50e+00

Pos logit contributions
 goalie+0.308
irms+0.305
 villains+0.287
awks+0.282
 thrill+0.272
Neg logit contributions
pron-0.381
 recipe-0.377
ons-0.353
 mixer-0.328
 baking-0.315
Token he
Token ID339
Feature activation-2.69e-01

Pos logit contributions
 goalie+2.765
irms+2.664
 villains+2.508
cer+2.496
 thrill+2.462
Neg logit contributions
pron-2.640
 recipe-2.628
ons-2.552
 mixer-2.349
 baking-2.193
Token shouted
Token ID17293
Feature activation-1.81e+00

Pos logit contributions
 goalie+0.445
irms+0.440
 villains+0.414
awks+0.408
 thrill+0.395
Neg logit contributions
pron-0.513
 recipe-0.509
ons-0.475
 mixer-0.441
 baking-0.422
Token.
Token ID13
Feature activation-8.61e-01

Pos logit contributions
 goalie+2.985
irms+2.942
 villains+2.822
awks+2.730
 thrill+2.722
Neg logit contributions
 recipe-3.458
pron-3.350
ons-3.207
 mixer-3.083
 baking-2.984
Token
Token ID198
Feature activation-9.74e-01

Pos logit contributions
 goalie+1.485
irms+1.475
awks+1.380
 villains+1.375
cer+1.346
Neg logit contributions
pron-1.593
 recipe-1.570
ons-1.465
 mixer-1.368
 baking-1.292
Token
Token ID198
Feature activation-2.20e+00

Pos logit contributions
 goalie+1.765
irms+1.727
 villains+1.622
awks+1.616
 thrill+1.595
Neg logit contributions
pron-1.712
 recipe-1.678
ons-1.624
 mixer-1.450
 baking-1.378
TokenBen
Token ID11696
Feature activation-6.46e-02

Pos logit contributions
irms+5.054
 goalie+4.946
awks+4.869
 thrill+4.722
ashes+4.661
Neg logit contributions
pron-2.994
 recipe-2.711
ons-2.652
 mixer-2.174
on-2.048
Token and
Token ID290
Feature activation-1.68e+00

Pos logit contributions
 goalie+0.107
irms+0.106
 villains+0.100
awks+0.099
 thrill+0.095
Neg logit contributions
pron-0.123
 recipe-0.121
ons-0.113
 mixer-0.105
 baking-0.101
Token Lily
Token ID20037
Feature activation-7.64e-01

Pos logit contributions
 goalie+2.806
irms+2.765
awks+2.617
ashes+2.532
 thrill+2.502
Neg logit contributions
pron-3.285
 recipe-3.193
ons-2.961
 mixer-2.815
 baking-2.663
Token were
Token ID547
Feature activation-2.33e+00

Pos logit contributions
 goalie+1.243
irms+1.242
 villains+1.177
awks+1.144
ashes+1.099
Neg logit contributions
pron-1.458
 recipe-1.441
ons-1.372
 mixer-1.282
 baking-1.219
Token very
Token ID845
Feature activation-1.97e+00

Pos logit contributions
irms+3.461
 goalie+3.432
 villains+3.214
awks+3.178
cer+3.177
Neg logit contributions
pron-4.692
 recipe-4.636
ons-4.607
 baking-4.391
 mixer-4.320
Token the
Token ID262
Feature activation-1.60e+00

Pos logit contributions
 goalie+1.105
irms+1.091
 villains+1.038
awks+1.021
 thrill+0.992
Neg logit contributions
pron-1.116
 recipe-1.111
ons-1.054
 mixer-0.971
 baking-0.924
Token monster
Token ID9234
Feature activation-6.34e-01

Pos logit contributions
irms+2.468
 goalie+2.422
awks+2.282
 villains+2.253
ashes+2.205
Neg logit contributions
pron-3.355
 recipe-3.353
ons-3.025
 mixer-2.971
 baking-2.829
Token and
Token ID290
Feature activation-9.95e-01

Pos logit contributions
irms+1.071
 goalie+1.070
 villains+1.002
awks+0.989
 thrill+0.955
Neg logit contributions
pron-1.179
 recipe-1.176
ons-1.103
 mixer-1.019
 baking-0.988
Token they
Token ID484
Feature activation-4.50e-01

Pos logit contributions
 goalie+1.763
irms+1.762
awks+1.636
 villains+1.608
ashes+1.595
Neg logit contributions
pron-1.788
 recipe-1.778
ons-1.683
 mixer-1.543
 baking-1.487
Token screamed
Token ID25421
Feature activation-9.09e-01

Pos logit contributions
 goalie+0.734
irms+0.730
 villains+0.682
awks+0.675
ashes+0.652
Neg logit contributions
pron-0.869
 recipe-0.858
ons-0.810
 mixer-0.750
 baking-0.721
Token too
Token ID1165
Feature activation-2.01e+00

Pos logit contributions
 goalie+1.447
irms+1.437
 villains+1.353
awks+1.336
 thrill+1.275
Neg logit contributions
pron-1.783
 recipe-1.771
ons-1.636
 mixer-1.558
 baking-1.526
Token.
Token ID13
Feature activation-1.35e+00

Pos logit contributions
 goalie+3.434
irms+3.352
awks+3.122
 villains+3.082
cer+3.075
Neg logit contributions
 recipe-3.767
pron-3.675
ons-3.547
 baking-3.284
 mixer-3.275
Token The
Token ID383
Feature activation-2.02e+00

Pos logit contributions
 goalie+1.788
irms+1.764
 villains+1.660
awks+1.600
 thrill+1.596
Neg logit contributions
pron-2.976
 recipe-2.974
ons-2.781
 mixer-2.704
 baking-2.588
Token parents
Token ID3397
Feature activation+6.22e-01

Pos logit contributions
irms+2.773
 goalie+2.716
awks+2.625
ashes+2.531
cer+2.490
Neg logit contributions
pron-4.636
 recipe-4.567
ons-4.231
 mixer-4.201
 baking-3.937
Token ran
Token ID4966
Feature activation-1.50e+00

Pos logit contributions
pron+1.266
 recipe+1.256
ons+1.145
 mixer+1.067
 baking+1.033
Neg logit contributions
 goalie-0.949
irms-0.926
 villains-0.886
awks-0.857
 thrill-0.849
Token to
Token ID284
Feature activation-1.03e+00

Pos logit contributions
 goalie+3.218
irms+3.178
 villains+3.074
awks+2.996
 thrill+2.981
Neg logit contributions
 recipe-2.122
pron-2.065
ons-1.880
 mixer-1.785
 baking-1.698
Token was
Token ID373
Feature activation-4.91e-01

Pos logit contributions
 goalie+1.100
irms+1.096
awks+1.031
 villains+1.023
ashes+0.997
Neg logit contributions
pron-1.011
 recipe-0.988
ons-0.923
 mixer-0.848
 baking-0.810
Token the
Token ID262
Feature activation-1.20e+00

Pos logit contributions
 goalie+0.737
irms+0.733
 villains+0.680
awks+0.677
ashes+0.651
Neg logit contributions
pron-1.020
 recipe-1.006
ons-0.936
 mixer-0.884
 baking-0.861
Token only
Token ID691
Feature activation-1.42e+00

Pos logit contributions
irms+1.995
 goalie+1.932
awks+1.851
 villains+1.822
ashes+1.771
Neg logit contributions
pron-2.393
 recipe-2.379
 mixer-2.083
ons-2.064
 baking-1.972
Token one
Token ID530
Feature activation-3.24e-01

Pos logit contributions
irms+2.509
 goalie+2.477
awks+2.362
 villains+2.292
ashes+2.225
Neg logit contributions
 recipe-2.693
pron-2.686
ons-2.371
 mixer-2.347
 baking-2.242
Token who
Token ID508
Feature activation-7.45e-01

Pos logit contributions
 goalie+0.525
irms+0.524
 villains+0.490
awks+0.486
 thrill+0.467
Neg logit contributions
pron-0.631
 recipe-0.626
ons-0.582
 mixer-0.545
 baking-0.522
Token felt
Token ID2936
Feature activation-2.20e+00

Pos logit contributions
 goalie+1.299
irms+1.290
awks+1.204
 villains+1.194
 thrill+1.161
Neg logit contributions
pron-1.374
 recipe-1.351
ons-1.286
 mixer-1.173
 baking-1.126
Token so
Token ID523
Feature activation-1.52e+00

Pos logit contributions
 goalie+3.649
irms+3.464
awks+3.224
cer+3.188
 villains+3.130
Neg logit contributions
pron-4.384
 recipe-4.317
ons-4.052
 baking-3.919
 mixer-3.844
Token scared
Token ID12008
Feature activation-1.35e+00

Pos logit contributions
 goalie+2.351
irms+2.202
cer+2.026
 villains+2.016
awks+2.011
Neg logit contributions
pron-3.171
 recipe-3.107
ons-2.984
 mixer-2.753
 baking-2.744
Token.
Token ID13
Feature activation+2.42e-01

Pos logit contributions
 goalie+2.321
irms+2.317
awks+2.180
 villains+2.105
 thrill+2.086
Neg logit contributions
 recipe-2.480
pron-2.461
ons-2.394
 mixer-2.214
 baking-2.118
Token Every
Token ID3887
Feature activation-5.63e-02

Pos logit contributions
pron+0.527
 recipe+0.522
ons+0.489
 mixer+0.455
 baking+0.441
Neg logit contributions
 goalie-0.339
irms-0.335
 villains-0.313
awks-0.306
 thrill-0.292
Token night
Token ID1755
Feature activation+5.52e-01

Pos logit contributions
 goalie+0.097
irms+0.096
 villains+0.091
awks+0.089
 thrill+0.087
Neg logit contributions
pron-0.104
 recipe-0.103
ons-0.096
 mixer-0.089
 baking-0.085
Token.
Token ID13
Feature activation-1.70e+00

Pos logit contributions
irms+3.605
 goalie+3.555
awks+3.424
 villains+3.386
 thrill+3.296
Neg logit contributions
 recipe-3.927
pron-3.661
ons-3.502
 mixer-3.361
 baking-3.343
Token He
Token ID679
Feature activation-1.32e+00

Pos logit contributions
 goalie+1.855
irms+1.847
cer+1.708
awks+1.669
 thrill+1.621
Neg logit contributions
 recipe-4.132
pron-4.113
ons-3.894
 mixer-3.775
 baking-3.606
Token caught
Token ID4978
Feature activation-1.56e+00

Pos logit contributions
 goalie+2.387
irms+2.381
awks+2.219
 villains+2.177
 thrill+2.139
Neg logit contributions
pron-2.376
 recipe-2.324
ons-2.193
 mixer-2.019
 baking-1.913
Token them
Token ID606
Feature activation-6.34e-01

Pos logit contributions
 goalie+2.976
irms+2.960
 villains+2.919
awks+2.758
 thrill+2.739
Neg logit contributions
 recipe-2.504
pron-2.452
ons-2.412
 mixer-2.256
 baking-2.106
Token and
Token ID290
Feature activation-7.70e-01

Pos logit contributions
irms+1.075
 goalie+1.067
 villains+1.013
awks+0.996
 thrill+0.955
Neg logit contributions
pron-1.171
 recipe-1.166
ons-1.090
 mixer-1.012
 baking-0.978
Token grabbed
Token ID13176
Feature activation-2.11e+00

Pos logit contributions
 goalie+1.368
irms+1.342
 villains+1.264
awks+1.253
ashes+1.212
Neg logit contributions
pron-1.397
 recipe-1.384
ons-1.291
 mixer-1.194
 baking-1.150
Token the
Token ID262
Feature activation+5.46e-01

Pos logit contributions
 goalie+4.339
 villains+4.253
irms+4.168
awks+3.954
 thrill+3.915
Neg logit contributions
 recipe-3.207
ons-3.114
pron-3.089
 mixer-2.937
 flour-2.711
Token com
Token ID401
Feature activation-4.49e-01

Pos logit contributions
pron+0.887
 recipe+0.877
ons+0.825
 mixer+0.727
 baking+0.697
Neg logit contributions
 goalie-1.058
irms-1.043
 villains-0.988
awks-0.981
 thrill-0.978
Tokenbs
Token ID1443
Feature activation-7.08e-01

Pos logit contributions
 goalie+0.671
irms+0.656
 villains+0.622
awks+0.600
 thrill+0.583
Neg logit contributions
pron-0.938
 recipe-0.927
ons-0.872
 mixer-0.811
 baking-0.778
Token from
Token ID422
Feature activation-1.46e+00

Pos logit contributions
irms+1.075
 goalie+1.064
 villains+1.008
awks+0.986
 thrill+0.947
Neg logit contributions
 recipe-1.437
pron-1.430
ons-1.346
 mixer-1.266
 baking-1.217
Token their
Token ID511
Feature activation+2.61e-01

Pos logit contributions
 goalie+2.586
irms+2.517
 villains+2.430
awks+2.352
ashes+2.318
Neg logit contributions
 recipe-2.636
pron-2.624
ons-2.426
 mixer-2.321
 baking-2.247
Token<|endoftext|>
Token ID50256
Feature activation-7.64e-01

Pos logit contributions
 goalie+3.079
irms+3.071
awks+2.940
 villains+2.920
 thrill+2.884
Neg logit contributions
pron-2.965
 recipe-2.915
ons-2.651
 mixer-2.557
 baking-2.488
TokenTom
Token ID13787
Feature activation-1.15e-01

Pos logit contributions
 goalie+1.240
irms+1.232
 villains+1.176
awks+1.152
 thrill+1.126
Neg logit contributions
pron-1.456
 recipe-1.451
ons-1.393
 mixer-1.275
 baking-1.225
Token and
Token ID290
Feature activation-1.58e+00

Pos logit contributions
 goalie+0.220
irms+0.218
 villains+0.208
awks+0.205
 thrill+0.199
Neg logit contributions
pron-0.190
 recipe-0.187
ons-0.172
 mixer-0.158
 baking-0.150
Token Sam
Token ID3409
Feature activation-1.40e+00

Pos logit contributions
irms+2.752
 goalie+2.709
awks+2.561
 thrill+2.507
 villains+2.486
Neg logit contributions
pron-2.922
 recipe-2.870
ons-2.803
 mixer-2.581
 baking-2.461
Token are
Token ID389
Feature activation-5.04e-01

Pos logit contributions
 goalie+2.191
irms+2.159
 villains+2.091
awks+2.003
 thrill+1.936
Neg logit contributions
pron-2.745
 recipe-2.726
ons-2.693
 mixer-2.469
 baking-2.378
Token friends
Token ID2460
Feature activation-2.24e+00

Pos logit contributions
 goalie+0.775
irms+0.770
 villains+0.735
awks+0.713
 thrill+0.693
Neg logit contributions
pron-1.032
 recipe-1.020
ons-0.962
 mixer-0.904
 baking-0.875
Token.
Token ID13
Feature activation-1.63e+00

Pos logit contributions
 goalie+3.954
irms+3.908
 villains+3.884
awks+3.692
ashes+3.613
Neg logit contributions
pron-4.329
 recipe-4.133
ons-4.075
 mixer-3.735
 baking-3.613
Token They
Token ID1119
Feature activation-2.07e+00

Pos logit contributions
 goalie+1.670
 villains+1.650
irms+1.634
 thrill+1.471
awks+1.467
Neg logit contributions
pron-4.046
 recipe-4.032
ons-3.855
 mixer-3.751
 baking-3.622
Token like
Token ID588
Feature activation+1.52e+00

Pos logit contributions
irms+2.578
 goalie+2.441
awks+2.316
 villains+2.292
cer+2.135
Neg logit contributions
pron-4.985
 recipe-4.802
ons-4.661
 mixer-4.450
 baking-4.411
Token to
Token ID284
Feature activation+1.84e-01

Pos logit contributions
pron+2.312
 recipe+2.160
ons+2.024
innie+1.784
 mixer+1.744
Neg logit contributions
 goalie-3.289
 villains-3.140
irms-3.101
ashes-3.003
awks-2.948
Token play
Token ID711
Feature activation-6.68e-01

Pos logit contributions
pron+0.417
 recipe+0.414
ons+0.389
 mixer+0.365
 baking+0.352
Neg logit contributions
 goalie-0.240
irms-0.234
 villains-0.219
awks-0.212
ashes-0.204