TOP ACTIVATIONS
MAX = 3.909

Tim loved to play all day . One day , he
I can 't get out !" cried Blue . He tried
Max loved to play all day long . He had many
uffy loved to play all day and drink milk from his
the middle of the swamp ! Blue w aded
Tom loved to play all day long . One day ,
pe b ble I found !" said Blue . " Wow
Max loved to play all day long with his friends in
He liked to play all day in the sunshine . One
through the pages of all the different magazines . She looked
play with her toys all day long . One day ,
? They look so amazing !" " Yes ,
we breathe . Blue sw am up to
play with her toys all day long . One day ,
She would eat candy all day and all night . One
toy cars and trucks all day long . One day ,
Lily loved to play all day long and never wanted to
play with his toys all day long . One day ,
play with his toys all day long . One day ,
illy liked to play all day long . She was very

MIN ACTIVATIONS
MIN = -4.209

S ara offered her purple bubble wand to Max
butterfly flew away , Joe and Linda decided to share a
share a hug . Joe and Linda closed their eyes and
on , Mom my ! Hur ry up !" he said
tea and cookies to the living room . They sit on
They go back to the living room and pick up the
to bat he first . Hur ry up , or the
Bob said , " Mr . Ant , why is the
surrender if it is the right thing to do , without
with the ball in the living room . They forget what
Mom comes to the living room and sees their bridge
and sad . She wants her tool back . She dives
like a toy ! Joe and Linda were so amazed and
the f auc et . The water came out and Lily
surrender if it is the right thing to do !"
happily . Then Rose showed her how to turn it off
f auc et to ' spl ash ' around and have
together . Joe and Linda then kept walking in
were walking down the street . Sandy was feeling cold ,
" It 's alive ! Hur ray !" The grape j

INTERVAL 1.880 - 3.909
CONTAINS 0.371%

had worked . After lots of hugs , her family was
It was old and had many colors . Tim did not
things around . It would carry sticks and leaves to make
. Let 's go home ." They thanked the nice man
the queen had taken away the peace of his people .

INTERVAL -0.150 - 1.880
CONTAINS 58.741%

are welcome , Lily ," Max says . " You are
flew away , and the village lived happily ever after .
. He bark ed at them and waved his tail .
you like him ?" Anna and Ben hug Spot
the lab . After sailing back to shore ,

INTERVAL -2.179 - -0.150
CONTAINS 40.756%

and suddenly his own misery seemed to have gone . They
. He runs towards the barn . Anna and
him and said , " Good night , snow man .
eating it herself . Mrs . Smith was so surprised and
it either . Then Max had a good idea . He

INTERVAL -4.209 - -2.179
CONTAINS 0.132%

in the corner of the living room . She didn 't
were crashing and the se ag ull s were squ aw
ows and blankets ." " Okay ," Mia said
they do not like mosquitoes . Mos quit oes are small
other man . He loved his pipe and he could never
Token Tim
Token ID5045
Feature activation+7.57e-01

Pos logit contributions
riage+0.361
ipples+0.354
ry+0.346
flower+0.344
ac+0.338
Neg logit contributions
hy-0.358
Mic-0.319
 rides-0.279
Red-0.276
Roger-0.270
Token loved
Token ID6151
Feature activation+1.50e-01

Pos logit contributions
hy+1.502
Mic+1.356
 rides+1.176
Red+1.171
Roger+1.169
Neg logit contributions
riage-1.194
flower-1.183
ipples-1.183
ry-1.168
andy-1.141
Token to
Token ID284
Feature activation+1.06e-01

Pos logit contributions
hy+0.337
Mic+0.306
Red+0.276
 rides+0.273
Roger+0.270
Neg logit contributions
riage-0.201
ipples-0.196
flower-0.190
ry-0.189
 department-0.185
Token play
Token ID711
Feature activation+1.45e+00

Pos logit contributions
hy+0.216
Mic+0.197
Red+0.174
 rides+0.173
Roger+0.172
Neg logit contributions
riage-0.162
ipples-0.158
ry-0.153
flower-0.153
andy-0.150
Token all
Token ID477
Feature activation+3.00e-01

Pos logit contributions
hy+2.919
Mic+2.472
Robert+2.178
nam+2.175
Roger+2.136
Neg logit contributions
ipples-2.339
ry-2.327
flower-2.267
riage-2.244
andy-2.240
Token day
Token ID1110
Feature activation+3.91e+00

Pos logit contributions
hy+0.676
Mic+0.622
Red+0.554
 rides+0.553
Roger+0.550
Neg logit contributions
riage-0.397
ipples-0.391
flower-0.378
ry-0.378
ac-0.369
Token.
Token ID13
Feature activation-3.84e-01

Pos logit contributions
hy+10.413
Mic+9.533
Red+9.028
Blue+8.707
Bow+8.472
Neg logit contributions
ry-5.171
 again-4.872
andy-4.644
ipples-4.565
en-4.441
Token One
Token ID1881
Feature activation-8.39e-01

Pos logit contributions
riage+0.705
ipples+0.691
ry+0.676
flower+0.671
ac+0.662
Neg logit contributions
hy-0.677
Mic-0.601
 rides-0.526
Red-0.521
Roger-0.506
Token day
Token ID1110
Feature activation+1.20e+00

Pos logit contributions
riage+1.331
ipples+1.296
 department+1.253
ry+1.250
flower+1.225
Neg logit contributions
hy-1.713
Mic-1.517
 rides-1.374
Red-1.346
Roger-1.316
Token,
Token ID11
Feature activation+7.90e-01

Pos logit contributions
hy+2.547
Mic+2.355
Red+2.095
Robert+2.043
 rides+2.041
Neg logit contributions
ry-1.784
riage-1.755
ipples-1.685
flower-1.676
andy-1.658
Token he
Token ID339
Feature activation-5.87e-01

Pos logit contributions
hy+1.456
Mic+1.280
 rides+1.107
Red+1.104
Robert+1.083
Neg logit contributions
riage-1.394
ipples-1.361
flower-1.358
ry-1.339
 department-1.318
Token I
Token ID314
Feature activation-1.15e+00

Pos logit contributions
riage+0.683
ipples+0.680
ry+0.656
flower+0.649
andy+0.641
Neg logit contributions
hy-0.847
Mic-0.764
 rides-0.678
Red-0.672
Roger-0.663
Token can
Token ID460
Feature activation+3.45e-01

Pos logit contributions
riage+1.848
ipples+1.747
flower+1.679
 department+1.639
andy+1.636
Neg logit contributions
hy-2.352
Mic-2.088
 rides-1.911
Red-1.840
Roger-1.804
Token't
Token ID470
Feature activation+3.24e-01

Pos logit contributions
hy+0.543
Mic+0.485
 rides+0.407
Red+0.405
Roger+0.398
Neg logit contributions
riage-0.687
ipples-0.681
ry-0.669
flower-0.668
andy-0.652
Token get
Token ID651
Feature activation+3.60e-01

Pos logit contributions
hy+0.634
Mic+0.571
Red+0.498
 rides+0.498
Roger+0.491
Neg logit contributions
riage-0.531
ipples-0.525
ry-0.512
flower-0.510
ac-0.492
Token out
Token ID503
Feature activation-8.94e-01

Pos logit contributions
hy+0.691
Mic+0.624
Red+0.543
 rides+0.541
Roger+0.528
Neg logit contributions
riage-0.603
ipples-0.592
flower-0.577
ry-0.576
 department-0.563
Token!"
Token ID2474
Feature activation+3.86e+00

Pos logit contributions
riage+1.572
ipples+1.538
flower+1.486
ry+1.476
andy+1.474
Neg logit contributions
hy-1.645
Mic-1.427
 rides-1.300
Red-1.241
Roger-1.233
Token cried
Token ID16896
Feature activation+1.75e+00

Pos logit contributions
hy+8.623
Mic+7.437
 blonde+6.820
hell+6.803
nam+6.613
Neg logit contributions
 Duck-6.177
 again-6.010
 Mama-5.617
antly-5.519
 department-5.454
Token Blue
Token ID4518
Feature activation-1.55e-01

Pos logit contributions
hy+3.411
Mic+2.962
 rides+2.741
Robert+2.666
nam+2.608
Neg logit contributions
 department-2.786
 again-2.743
riage-2.721
ipples-2.618
flower-2.614
Token.
Token ID13
Feature activation+2.14e+00

Pos logit contributions
riage+0.247
ipples+0.241
ry+0.235
flower+0.233
ac+0.229
Neg logit contributions
hy-0.310
Mic-0.280
 rides-0.249
Red-0.246
Roger-0.242
Token He
Token ID679
Feature activation-3.49e-01

Pos logit contributions
hy+5.169
Mic+4.473
 rides+4.153
nam+4.151
hell+4.102
Neg logit contributions
 department-2.647
riage-2.552
 Duck-2.505
 again-2.394
flower-2.359
Token tried
Token ID3088
Feature activation-9.76e-01

Pos logit contributions
riage+0.573
ipples+0.557
ry+0.539
flower+0.537
ac+0.526
Neg logit contributions
hy-0.688
Mic-0.621
 rides-0.553
Red-0.545
Roger-0.535
Token Max
Token ID5436
Feature activation+1.74e+00

Pos logit contributions
hy+2.409
Mic+2.050
ve+1.879
nam+1.872
 rides+1.802
Neg logit contributions
riage-2.686
 department-2.595
flower-2.548
ipples-2.543
andy-2.516
Token loved
Token ID6151
Feature activation+6.17e-01

Pos logit contributions
hy+4.372
Mic+4.055
nam+3.743
Robert+3.642
Roger+3.627
Neg logit contributions
flower-1.802
ry-1.760
riage-1.749
ipples-1.730
andy-1.678
Token to
Token ID284
Feature activation+3.31e-01

Pos logit contributions
hy+1.486
Mic+1.347
Red+1.246
Roger+1.212
Robert+1.200
Neg logit contributions
riage-0.705
ipples-0.705
flower-0.682
 department-0.672
andy-0.656
Token play
Token ID711
Feature activation+1.13e+00

Pos logit contributions
hy+0.648
Mic+0.589
Red+0.522
Roger+0.515
 rides+0.508
Neg logit contributions
riage-0.533
ipples-0.524
flower-0.510
ry-0.503
andy-0.498
Token all
Token ID477
Feature activation+2.67e-01

Pos logit contributions
hy+2.249
Mic+1.900
nam+1.720
Red+1.681
Robert+1.665
Neg logit contributions
ipples-1.859
ry-1.786
ac-1.784
riage-1.783
flower-1.776
Token day
Token ID1110
Feature activation+3.83e+00

Pos logit contributions
hy+0.599
Mic+0.550
Red+0.491
 rides+0.491
Roger+0.488
Neg logit contributions
riage-0.355
ipples-0.352
flower-0.339
ry-0.338
ac-0.332
Token long
Token ID890
Feature activation+2.58e+00

Pos logit contributions
hy+10.220
Mic+8.996
Red+8.640
Robert+8.003
Blue+7.969
Neg logit contributions
ry-4.974
ipples-4.636
andy-4.590
 again-4.582
 boil-4.400
Token.
Token ID13
Feature activation+4.08e-01

Pos logit contributions
hy+6.748
Mic+6.211
Red+5.664
Robert+5.601
Roger+5.598
Neg logit contributions
 again-2.713
ry-2.680
ipples-2.642
en-2.622
ac-2.437
Token He
Token ID679
Feature activation+1.40e-01

Pos logit contributions
hy+0.783
Mic+0.697
 rides+0.613
Red+0.607
Roger+0.603
Neg logit contributions
riage-0.678
ipples-0.658
ry-0.640
flower-0.637
andy-0.633
Token had
Token ID550
Feature activation+1.17e+00

Pos logit contributions
hy+0.328
Mic+0.300
 rides+0.270
Red+0.270
Roger+0.267
Neg logit contributions
riage-0.172
ipples-0.170
ry-0.164
flower-0.164
ac-0.160
Token many
Token ID867
Feature activation+1.82e+00

Pos logit contributions
hy+2.463
Mic+2.217
Red+1.983
Roger+1.949
Robert+1.919
Neg logit contributions
ipples-1.668
ac-1.604
riage-1.556
flower-1.532
 department-1.528
Tokenuffy
Token ID15352
Feature activation+3.56e-01

Pos logit contributions
riage+0.513
ipples+0.497
ry+0.480
flower+0.478
 department+0.469
Neg logit contributions
hy-0.709
Mic-0.644
 rides-0.575
Red-0.571
Roger-0.561
Token loved
Token ID6151
Feature activation+1.46e-01

Pos logit contributions
hy+0.692
Mic+0.625
Red+0.543
 rides+0.541
Roger+0.535
Neg logit contributions
riage-0.580
ipples-0.568
ry-0.562
flower-0.557
andy-0.549
Token to
Token ID284
Feature activation-1.02e-01

Pos logit contributions
hy+0.338
Mic+0.309
Red+0.279
 rides+0.277
Roger+0.274
Neg logit contributions
riage-0.186
ipples-0.182
flower-0.175
ry-0.175
 department-0.171
Token play
Token ID711
Feature activation+1.03e+00

Pos logit contributions
riage+0.162
ipples+0.159
ry+0.154
flower+0.153
ac+0.150
Neg logit contributions
hy-0.205
Mic-0.185
 rides-0.164
Red-0.163
Roger-0.160
Token all
Token ID477
Feature activation-1.10e-01

Pos logit contributions
hy+2.014
Mic+1.718
nam+1.512
Robert+1.484
Red+1.480
Neg logit contributions
ipples-1.690
ry-1.667
riage-1.657
flower-1.633
ac-1.630
Token day
Token ID1110
Feature activation+3.72e+00

Pos logit contributions
riage+0.155
ipples+0.151
ry+0.146
flower+0.145
ac+0.142
Neg logit contributions
hy-0.241
Mic-0.220
 rides-0.197
Red-0.196
Roger-0.193
Token and
Token ID290
Feature activation+1.84e-01

Pos logit contributions
hy+9.893
Mic+9.066
Red+8.445
Blue+7.996
Bow+7.914
Neg logit contributions
ry-4.876
 again-4.724
andy-4.491
ipples-4.273
 M-4.220
Token drink
Token ID4144
Feature activation+5.39e-01

Pos logit contributions
hy+0.364
Mic+0.331
Red+0.292
 rides+0.289
Roger+0.286
Neg logit contributions
riage-0.294
ipples-0.286
flower-0.278
ry-0.278
andy-0.271
Token milk
Token ID7545
Feature activation+7.40e-01

Pos logit contributions
hy+1.154
Mic+1.052
Red+0.929
 rides+0.920
Roger+0.917
Neg logit contributions
ipples-0.778
riage-0.768
ry-0.753
flower-0.741
andy-0.735
Token from
Token ID422
Feature activation+3.48e-01

Pos logit contributions
hy+1.614
Mic+1.466
Red+1.306
 rides+1.290
Roger+1.289
Neg logit contributions
ipples-1.056
flower-1.037
riage-1.035
ry-1.032
 department-0.979
Token his
Token ID465
Feature activation+5.99e-01

Pos logit contributions
hy+0.741
Mic+0.673
Red+0.597
 rides+0.594
Roger+0.587
Neg logit contributions
riage-0.508
ipples-0.494
ry-0.480
flower-0.479
andy-0.470
Token the
Token ID262
Feature activation-8.14e-01

Pos logit contributions
hy+0.141
Mic+0.129
 rides+0.117
Red+0.116
Roger+0.114
Neg logit contributions
riage-0.074
ipples-0.072
flower-0.070
ry-0.069
ac-0.067
Token middle
Token ID3504
Feature activation-7.75e-01

Pos logit contributions
riage+1.342
ipples+1.291
ry+1.277
ac+1.247
flower+1.244
Neg logit contributions
hy-1.614
Mic-1.434
 rides-1.281
Red-1.266
Roger-1.227
Token of
Token ID286
Feature activation-1.47e-01

Pos logit contributions
riage+1.246
ipples+1.187
 department+1.152
flower+1.148
ry+1.140
Neg logit contributions
hy-1.563
Mic-1.363
 rides-1.251
Red-1.197
Robert-1.180
Token the
Token ID262
Feature activation-7.67e-01

Pos logit contributions
riage+0.187
ipples+0.182
ry+0.176
flower+0.174
ac+0.170
Neg logit contributions
hy-0.344
Mic-0.314
 rides-0.285
Red-0.283
Roger-0.278
Token swamp
Token ID26837
Feature activation+1.55e-01

Pos logit contributions
riage+0.865
ipples+0.806
ry+0.793
ac+0.777
flower+0.774
Neg logit contributions
hy-1.915
Mic-1.763
 rides-1.626
Red-1.602
Roger-1.571
Token!
Token ID0
Feature activation+3.70e+00

Pos logit contributions
hy+0.313
Mic+0.288
Red+0.253
 rides+0.251
Roger+0.249
Neg logit contributions
riage-0.241
ipples-0.237
ry-0.233
flower-0.229
 department-0.224
Token
Token ID198
Feature activation+2.12e+00

Pos logit contributions
hy+8.637
Mic+7.283
hell+7.127
nam+7.105
thel+7.045
Neg logit contributions
 again-4.938
 department-4.901
 Duck-4.887
riage-4.555
 Sometimes-4.385
Token
Token ID198
Feature activation+3.27e+00

Pos logit contributions
hy+5.221
nam+4.358
Mic+4.347
 brace+4.140
 blonde+4.134
Neg logit contributions
ull-2.425
ry-2.413
ucks-2.366
ac-2.341
 again-2.327
TokenBlue
Token ID14573
Feature activation-1.85e-01

Pos logit contributions
hy+7.497
nam+5.918
con+5.867
 blonde+5.774
 brace+5.633
Neg logit contributions
ac-4.278
 department-4.183
 again-4.145
riage-4.142
 end-4.135
Token w
Token ID266
Feature activation+4.64e-01

Pos logit contributions
riage+0.266
ipples+0.259
ry+0.251
flower+0.249
ac+0.244
Neg logit contributions
hy-0.399
Mic-0.363
 rides-0.327
Red-0.323
Roger-0.318
Tokenaded
Token ID5286
Feature activation-6.10e-01

Pos logit contributions
hy+1.075
Mic+0.977
 rides+0.884
Red+0.880
Roger+0.866
Neg logit contributions
ipples-0.602
riage-0.580
andy-0.572
ry-0.567
ac-0.567
Token Tom
Token ID4186
Feature activation+1.30e+00

Pos logit contributions
hy+0.726
Mic+0.660
Red+0.568
Roger+0.562
 rides+0.560
Neg logit contributions
riage-0.763
ipples-0.743
flower-0.718
ry-0.714
 department-0.709
Token loved
Token ID6151
Feature activation+2.78e-01

Pos logit contributions
hy+2.803
Mic+2.585
Roger+2.274
nam+2.271
Robert+2.258
Neg logit contributions
flower-1.834
riage-1.821
ipples-1.810
ry-1.803
ac-1.759
Token to
Token ID284
Feature activation+6.70e-02

Pos logit contributions
hy+0.646
Mic+0.587
Red+0.533
 rides+0.525
Roger+0.524
Neg logit contributions
riage-0.346
ipples-0.340
flower-0.328
ry-0.324
 department-0.322
Token play
Token ID711
Feature activation+4.75e-01

Pos logit contributions
hy+0.137
Mic+0.125
Red+0.110
 rides+0.110
Roger+0.109
Neg logit contributions
riage-0.103
ipples-0.101
flower-0.097
ry-0.097
andy-0.095
Token all
Token ID477
Feature activation+8.21e-02

Pos logit contributions
hy+0.856
Mic+0.739
Red+0.634
 rides+0.632
Roger+0.626
Neg logit contributions
riage-0.845
ipples-0.843
ry-0.827
flower-0.815
ac-0.808
Token day
Token ID1110
Feature activation+3.68e+00

Pos logit contributions
hy+0.182
Mic+0.167
 rides+0.149
Red+0.149
Roger+0.147
Neg logit contributions
riage-0.112
ipples-0.110
ry-0.106
flower-0.106
ac-0.104
Token long
Token ID890
Feature activation+2.23e+00

Pos logit contributions
hy+9.813
Mic+8.839
Red+8.302
Blue+7.850
Robert+7.767
Neg logit contributions
ry-4.639
 again-4.297
ipples-4.231
andy-4.097
 boil-4.078
Token.
Token ID13
Feature activation+8.45e-02

Pos logit contributions
hy+5.618
Mic+5.263
Red+4.675
Roger+4.644
Robert+4.627
Neg logit contributions
ry-2.462
ipples-2.397
 again-2.375
en-2.330
ac-2.263
Token One
Token ID1881
Feature activation-7.64e-01

Pos logit contributions
hy+0.152
Mic+0.137
 rides+0.118
Red+0.118
Roger+0.116
Neg logit contributions
riage-0.151
ipples-0.148
ry-0.143
flower-0.142
andy-0.140
Token day
Token ID1110
Feature activation+5.31e-01

Pos logit contributions
riage+1.206
ipples+1.170
ry+1.135
 department+1.120
flower+1.115
Neg logit contributions
hy-1.566
Mic-1.399
 rides-1.255
Red-1.237
Roger-1.206
Token,
Token ID11
Feature activation+7.76e-01

Pos logit contributions
hy+1.029
Mic+0.956
Red+0.835
 rides+0.818
Roger+0.818
Neg logit contributions
riage-0.872
ry-0.847
ipples-0.836
flower-0.817
ac-0.815
Token pe
Token ID613
Feature activation+2.23e-02

Pos logit contributions
riage+1.553
ipples+1.518
 department+1.471
ry+1.453
ac+1.437
Neg logit contributions
hy-1.575
Mic-1.355
 rides-1.244
Red-1.185
 brace-1.157
Tokenb
Token ID65
Feature activation+1.66e-01

Pos logit contributions
hy+0.041
Mic+0.037
 rides+0.032
Red+0.032
Roger+0.032
Neg logit contributions
riage-0.039
ipples-0.038
ry-0.037
flower-0.037
ac-0.036
Tokenble
Token ID903
Feature activation+1.40e+00

Pos logit contributions
hy+0.469
Mic+0.439
 rides+0.404
Red+0.404
Roger+0.399
Neg logit contributions
riage-0.126
ipples-0.122
ry-0.117
flower-0.116
ac-0.111
Token I
Token ID314
Feature activation-5.15e-01

Pos logit contributions
hy+2.917
Mic+2.708
Red+2.391
Roger+2.349
Robert+2.306
Neg logit contributions
ry-2.204
flower-2.141
riage-2.141
ipples-2.111
ac-2.043
Token found
Token ID1043
Feature activation+1.36e+00

Pos logit contributions
riage+0.754
ipples+0.732
ry+0.704
flower+0.703
 department+0.693
Neg logit contributions
hy-1.098
Mic-0.988
 rides-0.903
Red-0.880
Roger-0.862
Token!"
Token ID2474
Feature activation+3.67e+00

Pos logit contributions
hy+2.882
Mic+2.693
Red+2.366
Roger+2.312
 blonde+2.284
Neg logit contributions
riage-2.021
ry-1.970
ac-1.942
flower-1.916
ipples-1.909
Token said
Token ID531
Feature activation+1.84e+00

Pos logit contributions
hy+8.240
Mic+7.060
hell+6.799
 blonde+6.651
nam+6.423
Neg logit contributions
 Duck-5.554
 again-5.266
riage-4.963
 Mama-4.886
 M-4.849
Token Blue
Token ID4518
Feature activation-1.08e-01

Pos logit contributions
hy+2.566
Mic+2.168
 rides+1.845
hell+1.810
 blonde+1.783
Neg logit contributions
riage-3.943
flower-3.749
 department-3.743
 Duck-3.703
 again-3.674
Token.
Token ID13
Feature activation+2.19e+00

Pos logit contributions
riage+0.171
ipples+0.167
ry+0.163
flower+0.162
ac+0.158
Neg logit contributions
hy-0.218
Mic-0.197
 rides-0.176
Red-0.174
Roger-0.171
Token "
Token ID366
Feature activation+1.20e+00

Pos logit contributions
hy+5.608
Mic+4.885
hell+4.610
nam+4.603
 blonde+4.529
Neg logit contributions
 department-2.263
riage-2.190
 Duck-2.096
flower-2.076
 again-2.032
TokenWow
Token ID22017
Feature activation+2.35e+00

Pos logit contributions
hy+2.732
Mic+2.377
 rides+2.273
 Robert+2.158
nam+2.148
Neg logit contributions
ipples-1.504
riage-1.503
flower-1.492
ac-1.418
andy-1.361
Token Max
Token ID5436
Feature activation+1.58e+00

Pos logit contributions
hy+2.433
Mic+2.099
ve+1.906
nam+1.869
Roger+1.830
Neg logit contributions
riage-2.757
 department-2.617
ipples-2.579
flower-2.575
andy-2.562
Token loved
Token ID6151
Feature activation+5.41e-01

Pos logit contributions
hy+3.548
Mic+3.257
nam+2.946
Robert+2.918
Roger+2.891
Neg logit contributions
riage-1.992
flower-1.978
ry-1.954
ipples-1.917
andy-1.897
Token to
Token ID284
Feature activation+5.64e-01

Pos logit contributions
hy+1.275
Mic+1.155
Red+1.063
Roger+1.037
Robert+1.029
Neg logit contributions
riage-0.646
ipples-0.642
flower-0.623
 department-0.611
andy-0.602
Token play
Token ID711
Feature activation+1.14e+00

Pos logit contributions
hy+1.139
Mic+1.047
Red+0.934
Roger+0.926
Robert+0.916
Neg logit contributions
riage-0.857
ipples-0.843
flower-0.826
ry-0.807
andy-0.805
Token all
Token ID477
Feature activation+8.26e-02

Pos logit contributions
hy+2.231
Mic+1.881
nam+1.689
Red+1.662
Robert+1.660
Neg logit contributions
ipples-1.872
flower-1.810
ry-1.807
riage-1.805
ac-1.798
Token day
Token ID1110
Feature activation+3.67e+00

Pos logit contributions
hy+0.183
Mic+0.167
 rides+0.150
Red+0.149
Roger+0.148
Neg logit contributions
riage-0.113
ipples-0.111
ry-0.107
flower-0.107
ac-0.104
Token long
Token ID890
Feature activation+2.43e+00

Pos logit contributions
hy+9.717
Mic+8.394
Red+8.165
Robert+7.558
Blue+7.544
Neg logit contributions
ry-4.671
ipples-4.422
andy-4.414
 again-4.358
 boil-4.159
Token with
Token ID351
Feature activation+1.26e+00

Pos logit contributions
hy+6.315
Mic+5.817
Red+5.273
Robert+5.221
Roger+5.219
Neg logit contributions
 again-2.517
ry-2.512
ipples-2.471
en-2.453
ac-2.315
Token his
Token ID465
Feature activation+1.79e+00

Pos logit contributions
hy+2.827
Mic+2.505
Red+2.317
Roger+2.277
Robert+2.265
Neg logit contributions
ipples-1.666
ac-1.624
flower-1.622
andy-1.618
riage-1.578
Token friends
Token ID2460
Feature activation+1.47e-01

Pos logit contributions
hy+4.827
Mic+4.405
Red+4.224
Roger+4.156
Robert+4.145
Neg logit contributions
flower-1.684
ipples-1.630
andy-1.624
ac-1.540
 department-1.502
Token in
Token ID287
Feature activation+2.45e-01

Pos logit contributions
hy+0.297
Mic+0.268
 rides+0.237
Red+0.237
Roger+0.232
Neg logit contributions
riage-0.231
ipples-0.226
ry-0.223
flower-0.222
ac-0.216
Token He
Token ID679
Feature activation+6.66e-01

Pos logit contributions
riage+0.550
ipples+0.540
ry+0.529
flower+0.525
ac+0.519
Neg logit contributions
hy-0.448
Mic-0.392
 rides-0.338
Red-0.333
Roger-0.325
Token liked
Token ID8288
Feature activation+2.10e-01

Pos logit contributions
hy+1.309
Mic+1.193
 rides+1.028
Robert+1.027
Red+1.026
Neg logit contributions
riage-1.054
ipples-1.044
flower-1.027
ry-1.022
andy-1.008
Token to
Token ID284
Feature activation+8.02e-02

Pos logit contributions
hy+0.489
Mic+0.448
Red+0.405
 rides+0.399
Roger+0.396
Neg logit contributions
riage-0.262
ipples-0.257
flower-0.248
ry-0.246
 department-0.242
Token play
Token ID711
Feature activation+5.75e-01

Pos logit contributions
hy+0.162
Mic+0.147
Red+0.130
 rides+0.130
Roger+0.128
Neg logit contributions
riage-0.125
ipples-0.122
flower-0.119
ry-0.118
andy-0.116
Token all
Token ID477
Feature activation-4.22e-01

Pos logit contributions
hy+1.077
Mic+0.935
Red+0.805
Robert+0.798
Roger+0.794
Neg logit contributions
ipples-0.995
riage-0.990
ry-0.975
flower-0.961
ac-0.947
Token day
Token ID1110
Feature activation+3.63e+00

Pos logit contributions
riage+0.618
ipples+0.598
ry+0.580
flower+0.575
 department+0.573
Neg logit contributions
hy-0.901
Mic-0.813
 rides-0.739
Red-0.726
Roger-0.711
Token in
Token ID287
Feature activation-4.46e-01

Pos logit contributions
hy+9.961
Mic+9.053
Red+8.435
Blue+8.042
Bow+8.000
Neg logit contributions
ry-4.291
 again-3.956
ipples-3.855
 boil-3.697
en-3.695
Token the
Token ID262
Feature activation+7.44e-02

Pos logit contributions
riage+0.586
ipples+0.569
ry+0.548
flower+0.544
 department+0.535
Neg logit contributions
hy-1.025
Mic-0.935
 rides-0.858
Red-0.838
Roger-0.829
Token sunshine
Token ID34488
Feature activation+2.33e-01

Pos logit contributions
hy+0.157
Mic+0.142
 rides+0.126
Red+0.126
Roger+0.124
Neg logit contributions
riage-0.111
ipples-0.109
flower-0.106
ry-0.105
andy-0.104
Token.
Token ID13
Feature activation-5.85e-01

Pos logit contributions
hy+0.479
Mic+0.439
Red+0.387
 rides+0.383
Roger+0.381
Neg logit contributions
riage-0.354
ipples-0.351
flower-0.343
ry-0.342
ac-0.328
Token One
Token ID1881
Feature activation-1.00e+00

Pos logit contributions
riage+1.083
ipples+1.063
ry+1.040
flower+1.038
ac+1.024
Neg logit contributions
hy-1.027
Mic-0.908
 rides-0.803
Red-0.791
Roger-0.767
Token through
Token ID832
Feature activation+1.24e-01

Pos logit contributions
riage+0.601
ipples+0.582
ry+0.565
flower+0.560
ac+0.550
Neg logit contributions
hy-0.840
Mic-0.754
 rides-0.684
Red-0.674
Roger-0.662
Token the
Token ID262
Feature activation+2.44e+00

Pos logit contributions
hy+0.266
Mic+0.244
Red+0.215
 rides+0.215
Roger+0.212
Neg logit contributions
riage-0.178
ipples-0.173
ry-0.170
flower-0.168
ac-0.164
Token pages
Token ID5468
Feature activation+7.07e-01

Pos logit contributions
hy+4.172
Mic+3.996
Roger+3.521
Robert+3.450
Bow+3.439
Neg logit contributions
riage-4.262
ac-4.188
 department-4.187
flower-4.102
liness-4.046
Token of
Token ID286
Feature activation+1.08e+00

Pos logit contributions
hy+1.499
Mic+1.434
Red+1.245
Roger+1.237
 rides+1.226
Neg logit contributions
ipples-1.010
ry-0.999
riage-0.995
flower-0.965
ac-0.941
Token all
Token ID477
Feature activation-6.88e-01

Pos logit contributions
hy+2.546
Mic+2.360
Red+2.116
Roger+2.068
 rides+2.060
Neg logit contributions
riage-1.319
ipples-1.235
flower-1.226
andy-1.182
 department-1.181
Token the
Token ID262
Feature activation+3.62e+00

Pos logit contributions
riage+1.088
ipples+1.065
ry+1.021
 department+1.012
flower+1.010
Neg logit contributions
hy-1.402
Mic-1.228
 rides-1.134
Red-1.096
Roger-1.074
Token different
Token ID1180
Feature activation+1.47e+00

Pos logit contributions
hy+7.279
Mic+6.718
Roger+6.119
Blue+6.012
Robert+5.934
Neg logit contributions
ac-6.146
en-5.898
liness-5.735
riage-5.712
 department-5.581
Token magazines
Token ID16695
Feature activation+1.82e+00

Pos logit contributions
hy+3.183
Mic+3.093
Roger+2.711
Red+2.695
Blue+2.642
Neg logit contributions
riage-2.082
ry-2.015
en-2.006
ac-1.993
ipples-1.990
Token.
Token ID13
Feature activation+9.92e-01

Pos logit contributions
Mic+4.139
hy+4.072
Red+3.505
Roger+3.494
Robert+3.414
Neg logit contributions
ry-2.414
riage-2.348
flower-2.329
ipples-2.295
 again-2.204
Token She
Token ID1375
Feature activation+5.90e-01

Pos logit contributions
hy+1.879
Mic+1.703
 rides+1.483
Roger+1.424
Red+1.422
Neg logit contributions
riage-1.691
ry-1.602
ipples-1.576
 department-1.570
flower-1.553
Token looked
Token ID3114
Feature activation+8.21e-01

Pos logit contributions
hy+1.308
Mic+1.213
Red+1.072
Roger+1.063
 rides+1.047
Neg logit contributions
riage-0.796
ipples-0.785
ry-0.784
flower-0.762
andy-0.752
Token play
Token ID711
Feature activation+1.38e+00

Pos logit contributions
hy+0.615
Mic+0.561
Red+0.497
Roger+0.491
 rides+0.491
Neg logit contributions
riage-0.459
ipples-0.448
flower-0.438
ry-0.435
andy-0.428
Token with
Token ID351
Feature activation+1.53e+00

Pos logit contributions
hy+2.755
Mic+2.323
Robert+2.052
Roger+2.012
nam+2.011
Neg logit contributions
flower-2.235
ry-2.233
ipples-2.232
riage-2.183
andy-2.166
Token her
Token ID607
Feature activation+1.43e+00

Pos logit contributions
hy+3.309
Mic+2.854
Robert+2.585
Roger+2.576
Red+2.556
Neg logit contributions
flower-2.287
ry-2.214
andy-2.213
en-2.165
ipples-2.165
Token toys
Token ID14958
Feature activation+1.39e+00

Pos logit contributions
hy+3.004
Mic+2.619
Robert+2.374
Red+2.357
Roger+2.341
Neg logit contributions
flower-2.298
ry-2.151
riage-2.129
ipples-2.114
andy-2.111
Token all
Token ID477
Feature activation+1.46e-01

Pos logit contributions
hy+3.103
Mic+2.877
Roger+2.543
Robert+2.527
Red+2.491
Neg logit contributions
flower-1.907
ry-1.903
ipples-1.846
riage-1.842
ac-1.790
Token day
Token ID1110
Feature activation+3.62e+00

Pos logit contributions
hy+0.326
Mic+0.300
 rides+0.268
Red+0.267
Roger+0.264
Neg logit contributions
riage-0.200
ipples-0.195
flower-0.189
ry-0.189
ac-0.184
Token long
Token ID890
Feature activation+2.18e+00

Pos logit contributions
hy+8.870
Mic+8.054
Red+7.492
Bow+7.185
Blue+7.141
Neg logit contributions
ry-5.211
andy-4.907
 again-4.860
ipples-4.680
ac-4.459
Token.
Token ID13
Feature activation+9.00e-02

Pos logit contributions
hy+5.373
Mic+5.174
Roger+4.623
Red+4.608
Robert+4.563
Neg logit contributions
ry-2.449
 again-2.366
en-2.344
ac-2.294
ipples-2.285
Token One
Token ID1881
Feature activation-7.21e-01

Pos logit contributions
hy+0.177
Mic+0.158
 rides+0.140
Red+0.138
Roger+0.137
Neg logit contributions
riage-0.147
ipples-0.143
ry-0.140
flower-0.140
andy-0.137
Token day
Token ID1110
Feature activation+1.42e+00

Pos logit contributions
riage+1.123
ipples+1.102
ry+1.048
 department+1.047
flower+1.034
Neg logit contributions
hy-1.495
Mic-1.325
 rides-1.200
Red-1.179
Roger-1.150
Token,
Token ID11
Feature activation+9.92e-01

Pos logit contributions
hy+3.082
Mic+2.843
Red+2.497
Robert+2.472
 rides+2.465
Neg logit contributions
ry-2.072
riage-2.028
andy-1.932
flower-1.923
ac-1.881
Token?
Token ID30
Feature activation-5.12e-01

Pos logit contributions
riage+1.130
ipples+1.121
ry+1.042
flower+1.037
 department+1.035
Neg logit contributions
hy-1.967
Mic-1.762
 rides-1.627
Red-1.584
Roger-1.571
Token They
Token ID1119
Feature activation+2.65e-01

Pos logit contributions
riage+0.874
ipples+0.869
ry+0.837
flower+0.834
andy+0.825
Neg logit contributions
hy-0.958
Mic-0.853
Red-0.754
 rides-0.754
Roger-0.734
Token look
Token ID804
Feature activation+9.40e-01

Pos logit contributions
hy+0.557
Mic+0.513
 rides+0.450
Red+0.449
Roger+0.444
Neg logit contributions
riage-0.394
ipples-0.382
ry-0.373
flower-0.373
ac-0.365
Token so
Token ID523
Feature activation+1.14e+00

Pos logit contributions
hy+1.704
Mic+1.573
 rides+1.317
Red+1.307
Roger+1.297
Neg logit contributions
riage-1.673
ry-1.604
ipples-1.602
ac-1.595
 department-1.554
Token amazing
Token ID4998
Feature activation+4.87e-01

Pos logit contributions
hy+1.896
Mic+1.793
Roger+1.435
 rides+1.422
Robert+1.416
Neg logit contributions
riage-2.193
ipples-2.145
ry-2.106
ac-2.077
andy-2.047
Token!"
Token ID2474
Feature activation+3.62e+00

Pos logit contributions
hy+0.887
Mic+0.790
Red+0.676
 rides+0.670
Roger+0.654
Neg logit contributions
riage-0.883
ry-0.854
ipples-0.852
flower-0.833
ac-0.820
Token
Token ID198
Feature activation+2.19e+00

Pos logit contributions
hy+8.246
Mic+6.685
hell+6.458
Red+5.939
 brace+5.870
Neg logit contributions
 Mama-5.902
 again-5.700
 Duck-5.529
 Mum-5.496
 M-5.393
Token
Token ID198
Feature activation+1.74e+00

Pos logit contributions
hy+5.318
Mic+4.261
 brace+4.114
 rides+4.095
nam+4.062
Neg logit contributions
riage-2.681
ry-2.670
ucks-2.635
ull-2.626
 again-2.618
Token"
Token ID1
Feature activation-3.61e-04

Pos logit contributions
hy+3.713
Mic+2.897
 rides+2.820
con+2.721
 brace+2.718
Neg logit contributions
riage-2.657
ac-2.562
ry-2.532
 department-2.531
ipples-2.500
TokenYes
Token ID5297
Feature activation+7.19e-01

Pos logit contributions
riage+0.001
ipples+0.001
flower+0.001
ry+0.001
ac+0.001
Neg logit contributions
hy-0.001
Mic-0.001
 rides-0.001
Red-0.001
Roger-0.001
Token,
Token ID11
Feature activation+4.26e-01

Pos logit contributions
hy+1.319
Mic+1.191
 rides+1.051
Red+1.024
nam+0.985
Neg logit contributions
riage-1.276
ry-1.230
 department-1.205
ipples-1.203
flower-1.194
Token we
Token ID356
Feature activation+2.56e-01

Pos logit contributions
riage+1.084
ipples+1.054
ry+1.018
flower+1.014
andy+1.008
Neg logit contributions
hy-1.293
Mic-1.133
 rides-1.033
Red-1.004
Roger-0.988
Token breathe
Token ID18044
Feature activation-3.23e-01

Pos logit contributions
hy+0.528
Mic+0.482
 rides+0.425
Red+0.425
Roger+0.418
Neg logit contributions
riage-0.388
ipples-0.383
ry-0.372
flower-0.368
ac-0.361
Token.
Token ID13
Feature activation+2.50e+00

Pos logit contributions
riage+0.521
ipples+0.507
flower+0.493
ry+0.492
andy+0.482
Neg logit contributions
hy-0.634
Mic-0.571
 rides-0.511
Red-0.504
Roger-0.494
Token
Token ID220
Feature activation+3.94e-01

Pos logit contributions
hy+5.659
Mic+4.695
 blonde+4.400
nam+4.383
hell+4.335
Neg logit contributions
riage-3.552
 department-3.402
flower-3.378
 again-3.367
 Duck-3.365
Token
Token ID198
Feature activation+1.56e+00

Pos logit contributions
hy+0.776
Mic+0.689
 rides+0.615
Red+0.601
Roger+0.591
Neg logit contributions
riage-0.635
ipples-0.630
ry-0.619
flower-0.601
ac-0.601
Token
Token ID198
Feature activation+3.60e+00

Pos logit contributions
hy+3.557
Mic+3.070
nam+2.934
 rides+2.792
 blonde+2.783
Neg logit contributions
ry-2.040
riage-2.021
andy-2.016
ipples-2.015
ull-1.953
TokenBlue
Token ID14573
Feature activation-1.28e+00

Pos logit contributions
hy+7.359
nam+5.622
con+5.490
dream+5.379
 blonde+5.349
Neg logit contributions
flower-5.784
 department-5.741
 end-5.719
ac-5.631
 Duck-5.576
Token sw
Token ID1509
Feature activation-1.96e+00

Pos logit contributions
riage+1.389
ipples+1.221
 department+1.097
ours+1.092
ucks+1.086
Neg logit contributions
hy-3.300
Mic-3.098
 rides-2.777
Red-2.769
Roger-2.745
Tokenam
Token ID321
Feature activation-5.04e-01

Pos logit contributions
riage+2.654
 department+2.307
flower+2.291
ipples+2.268
ry+2.196
Neg logit contributions
hy-4.564
Mic-4.172
 rides-3.886
Red-3.683
 Bow-3.642
Token up
Token ID510
Feature activation+3.07e-02

Pos logit contributions
riage+0.779
ipples+0.752
ry+0.726
flower+0.725
andy+0.709
Neg logit contributions
hy-1.048
Mic-0.946
 rides-0.852
Red-0.838
Roger-0.830
Token to
Token ID284
Feature activation+5.80e-01

Pos logit contributions
hy+0.069
Mic+0.063
 rides+0.057
Red+0.056
Roger+0.055
Neg logit contributions
riage-0.041
ipples-0.040
ry-0.039
flower-0.039
 department-0.038
Token play
Token ID711
Feature activation+1.12e+00

Pos logit contributions
hy+0.666
Mic+0.608
Red+0.540
Roger+0.532
 rides+0.531
Neg logit contributions
riage-0.486
ipples-0.473
flower-0.460
ry-0.456
andy-0.452
Token with
Token ID351
Feature activation+1.92e+00

Pos logit contributions
hy+2.149
Mic+1.800
nam+1.571
Robert+1.567
Red+1.562
Neg logit contributions
ipples-1.862
ry-1.845
riage-1.840
flower-1.836
andy-1.814
Token her
Token ID607
Feature activation+1.74e+00

Pos logit contributions
hy+4.310
Mic+3.702
Robert+3.426
Roger+3.411
Red+3.368
Neg logit contributions
andy-2.674
flower-2.630
en-2.559
ipples-2.513
ucks-2.475
Token toys
Token ID14958
Feature activation+8.89e-01

Pos logit contributions
hy+3.848
Mic+3.341
Robert+3.069
Red+3.007
Roger+2.998
Neg logit contributions
flower-2.643
 department-2.406
ipples-2.391
ry-2.383
andy-2.373
Token all
Token ID477
Feature activation-4.98e-01

Pos logit contributions
hy+1.909
Mic+1.736
Red+1.530
Roger+1.523
Robert+1.511
Neg logit contributions
ry-1.283
flower-1.281
ipples-1.277
riage-1.275
ac-1.233
Token day
Token ID1110
Feature activation+3.58e+00

Pos logit contributions
riage+0.738
ipples+0.719
ry+0.696
 department+0.690
flower+0.686
Neg logit contributions
hy-1.053
Mic-0.947
 rides-0.858
Red-0.844
Roger-0.828
Token long
Token ID890
Feature activation+1.95e+00

Pos logit contributions
hy+8.006
Mic+6.787
Red+6.479
Bow+6.127
Robert+6.099
Neg logit contributions
ry-5.918
andy-5.667
 again-5.543
 boil-5.506
ipples-5.469
Token.
Token ID13
Feature activation-2.56e-01

Pos logit contributions
hy+4.726
Mic+4.521
Roger+4.027
Red+4.020
Robert+3.992
Neg logit contributions
ry-2.245
en-2.130
ipples-2.125
ac-2.110
 again-2.104
Token One
Token ID1881
Feature activation-6.23e-01

Pos logit contributions
riage+0.443
ipples+0.435
ry+0.423
flower+0.420
ac+0.413
Neg logit contributions
hy-0.479
Mic-0.429
 rides-0.378
Red-0.374
Roger-0.366
Token day
Token ID1110
Feature activation+1.10e+00

Pos logit contributions
riage+0.959
ipples+0.935
ry+0.902
 department+0.886
flower+0.884
Neg logit contributions
hy-1.298
Mic-1.164
 rides-1.043
Red-1.028
Roger-1.008
Token,
Token ID11
Feature activation+4.04e-01

Pos logit contributions
hy+2.317
Mic+2.132
Red+1.892
 rides+1.875
Robert+1.857
Neg logit contributions
ry-1.658
riage-1.615
flower-1.553
andy-1.550
ac-1.517
Token She
Token ID1375
Feature activation+9.29e-01

Pos logit contributions
riage+1.086
ipples+1.076
ry+1.049
flower+1.041
ac+1.028
Neg logit contributions
hy-0.766
Mic-0.661
 rides-0.564
Red-0.561
Roger-0.531
Token would
Token ID561
Feature activation+1.16e+00

Pos logit contributions
hy+2.041
Mic+1.823
Red+1.632
Roger+1.630
 rides+1.621
Neg logit contributions
riage-1.268
ipples-1.239
ry-1.236
flower-1.236
ac-1.227
Token eat
Token ID4483
Feature activation+2.00e+00

Pos logit contributions
hy+2.242
Mic+2.022
Red+1.820
Roger+1.802
Robert+1.753
Neg logit contributions
riage-1.848
ipples-1.818
flower-1.811
ac-1.769
andy-1.757
Token candy
Token ID18550
Feature activation+2.21e+00

Pos logit contributions
hy+4.154
Mic+3.406
powers+3.006
Robert+3.003
Roger+2.998
Neg logit contributions
 again-3.170
flower-3.157
riage-3.145
andy-3.114
en-3.107
Token all
Token ID477
Feature activation-4.19e-01

Pos logit contributions
hy+5.253
Mic+4.641
Roger+4.214
 blonde+4.191
Red+4.179
Neg logit contributions
ry-2.968
flower-2.904
 again-2.875
ipples-2.821
ac-2.777
Token day
Token ID1110
Feature activation+3.57e+00

Pos logit contributions
riage+0.608
ipples+0.591
ry+0.571
flower+0.565
 department+0.562
Neg logit contributions
hy-0.894
Mic-0.811
 rides-0.732
Red-0.722
Roger-0.710
Token and
Token ID290
Feature activation+1.10e+00

Pos logit contributions
hy+8.784
Mic+7.417
Red+7.398
Bow+6.996
Blue+6.787
Neg logit contributions
ry-5.375
andy-5.021
 boil-4.946
 again-4.871
ipples-4.852
Token all
Token ID477
Feature activation-6.62e-01

Pos logit contributions
hy+2.293
Mic+2.093
Red+1.886
Roger+1.851
Robert+1.797
Neg logit contributions
riage-1.668
ipples-1.606
flower-1.604
ry-1.559
andy-1.545
Token night
Token ID1755
Feature activation+3.04e+00

Pos logit contributions
riage+1.031
ipples+1.015
ry+0.980
 department+0.974
flower+0.969
Neg logit contributions
hy-1.363
Mic-1.203
 rides-1.099
Red-1.066
nam-1.045
Token.
Token ID13
Feature activation+2.00e-01

Pos logit contributions
hy+7.603
Mic+7.179
Robert+6.345
Red+6.280
Bow+6.273
Neg logit contributions
ry-4.098
ipples-3.766
 boil-3.591
 again-3.516
en-3.468
Token One
Token ID1881
Feature activation-5.02e-01

Pos logit contributions
hy+0.383
Mic+0.342
 rides+0.301
Red+0.296
Roger+0.295
Neg logit contributions
riage-0.336
ipples-0.326
ry-0.320
flower-0.319
andy-0.315
Token toy
Token ID13373
Feature activation+4.89e-01

Pos logit contributions
hy+2.560
Mic+2.256
Red+2.034
Robert+2.017
Roger+1.996
Neg logit contributions
flower-1.834
ipples-1.759
riage-1.753
 department-1.735
ry-1.728
Token cars
Token ID5006
Feature activation+1.83e+00

Pos logit contributions
hy+0.981
Mic+0.877
Red+0.770
Robert+0.769
Roger+0.768
Neg logit contributions
riage-0.769
ry-0.755
flower-0.750
ipples-0.748
 department-0.733
Token and
Token ID290
Feature activation+9.02e-01

Pos logit contributions
hy+4.044
Mic+3.856
Roger+3.503
Robert+3.441
 blonde+3.427
Neg logit contributions
flower-2.456
ry-2.399
ipples-2.383
ac-2.369
 again-2.280
Token trucks
Token ID13960
Feature activation+2.11e+00

Pos logit contributions
hy+1.832
Mic+1.623
Red+1.452
Roger+1.447
Robert+1.442
Neg logit contributions
riage-1.389
flower-1.342
 department-1.328
ipples-1.327
ric-1.302
Token all
Token ID477
Feature activation-4.50e-01

Pos logit contributions
Mic+4.850
hy+4.777
Roger+4.336
Robert+4.206
 blonde+4.181
Neg logit contributions
ipples-2.632
flower-2.583
ry-2.495
 department-2.483
 again-2.474
Token day
Token ID1110
Feature activation+3.57e+00

Pos logit contributions
riage+0.669
ipples+0.650
ry+0.634
flower+0.624
 department+0.621
Neg logit contributions
hy-0.957
Mic-0.857
 rides-0.779
Red-0.766
Roger-0.751
Token long
Token ID890
Feature activation+1.87e+00

Pos logit contributions
hy+7.448
Mic+6.922
Red+6.197
Blue+5.919
Bow+5.837
Neg logit contributions
ry-5.823
ipples-5.762
 again-5.601
ull-5.388
andy-5.310
Token.
Token ID13
Feature activation-1.98e-01

Pos logit contributions
hy+4.427
Mic+4.280
Red+3.792
Roger+3.783
Robert+3.768
Neg logit contributions
ry-2.181
ipples-2.157
en-2.110
 again-2.074
riage-2.072
Token One
Token ID1881
Feature activation-9.17e-01

Pos logit contributions
riage+0.338
ipples+0.331
ry+0.323
flower+0.321
ac+0.315
Neg logit contributions
hy-0.374
Mic-0.336
 rides-0.296
Red-0.293
Roger-0.287
Token day
Token ID1110
Feature activation+1.26e+00

Pos logit contributions
riage+1.472
ipples+1.442
ry+1.374
 department+1.372
flower+1.356
Neg logit contributions
hy-1.868
Mic-1.643
 rides-1.494
Red-1.463
Roger-1.426
Token,
Token ID11
Feature activation+8.32e-01

Pos logit contributions
hy+2.702
Mic+2.476
Red+2.199
 rides+2.159
Robert+2.147
Neg logit contributions
ry-1.849
riage-1.824
flower-1.746
ipples-1.731
 department-1.727
Token Lily
Token ID20037
Feature activation+1.11e+00

Pos logit contributions
riage+0.957
ipples+0.943
ry+0.923
flower+0.914
ac+0.905
Neg logit contributions
hy-0.656
Mic-0.568
Red-0.475
 rides-0.473
Roger-0.453
Token loved
Token ID6151
Feature activation+1.84e-01

Pos logit contributions
hy+2.174
Mic+1.958
Roger+1.704
Robert+1.703
 rides+1.693
Neg logit contributions
riage-1.769
flower-1.768
ry-1.752
ipples-1.734
ac-1.696
Token to
Token ID284
Feature activation+1.04e-01

Pos logit contributions
hy+0.413
Mic+0.376
Red+0.338
 rides+0.335
Roger+0.333
Neg logit contributions
riage-0.245
ipples-0.237
flower-0.232
ry-0.230
 department-0.225
Token play
Token ID711
Feature activation+9.60e-01

Pos logit contributions
hy+0.212
Mic+0.193
Red+0.171
 rides+0.170
Roger+0.168
Neg logit contributions
riage-0.160
ipples-0.156
ry-0.152
flower-0.151
andy-0.148
Token all
Token ID477
Feature activation-1.57e-01

Pos logit contributions
hy+1.814
Mic+1.544
Robert+1.342
Red+1.330
Roger+1.326
Neg logit contributions
ry-1.623
ipples-1.620
riage-1.609
flower-1.608
andy-1.576
Token day
Token ID1110
Feature activation+3.55e+00

Pos logit contributions
riage+0.222
ipples+0.216
ry+0.209
flower+0.208
ac+0.203
Neg logit contributions
hy-0.341
Mic-0.310
 rides-0.279
Red-0.276
Roger-0.272
Token long
Token ID890
Feature activation+2.41e+00

Pos logit contributions
hy+9.069
Mic+8.164
Red+7.618
Bow+7.352
Blue+7.339
Neg logit contributions
ry-4.903
andy-4.415
 again-4.370
 boil-4.251
ipples-4.244
Token and
Token ID290
Feature activation+2.13e-01

Pos logit contributions
hy+6.286
Mic+5.869
Roger+5.270
Robert+5.264
Red+5.219
Neg logit contributions
ry-2.524
 again-2.495
en-2.462
ipples-2.282
andy-2.233
Token never
Token ID1239
Feature activation-5.71e-01

Pos logit contributions
hy+0.408
Mic+0.369
 rides+0.324
Red+0.322
Roger+0.318
Neg logit contributions
riage-0.359
ipples-0.347
ry-0.340
flower-0.340
ac-0.333
Token wanted
Token ID2227
Feature activation+1.14e-01

Pos logit contributions
riage+0.964
ipples+0.947
ry+0.921
flower+0.914
andy+0.906
Neg logit contributions
hy-1.093
Mic-0.957
 rides-0.868
Red-0.846
Roger-0.818
Token to
Token ID284
Feature activation-6.95e-01

Pos logit contributions
hy+0.261
Mic+0.241
Red+0.215
 rides+0.215
Roger+0.212
Neg logit contributions
riage-0.145
ipples-0.141
ry-0.137
flower-0.135
 department-0.134
Token play
Token ID711
Feature activation+4.74e-01

Pos logit contributions
riage+0.306
ipples+0.299
ry+0.291
flower+0.289
ac+0.283
Neg logit contributions
hy-0.406
Mic-0.367
 rides-0.328
Red-0.324
Roger-0.319
Token with
Token ID351
Feature activation+1.58e-01

Pos logit contributions
hy+0.858
Mic+0.747
Red+0.632
 rides+0.631
Robert+0.628
Neg logit contributions
riage-0.842
ipples-0.840
ry-0.824
flower-0.817
ac-0.804
Token his
Token ID465
Feature activation+7.49e-01

Pos logit contributions
hy+0.326
Mic+0.293
 rides+0.258
Red+0.258
Roger+0.254
Neg logit contributions
riage-0.243
ipples-0.239
ry-0.234
flower-0.232
andy-0.227
Token toys
Token ID14958
Feature activation+7.50e-01

Pos logit contributions
hy+1.528
Mic+1.340
Red+1.198
 rides+1.191
Robert+1.188
Neg logit contributions
flower-1.155
riage-1.144
ipples-1.140
ry-1.119
 department-1.113
Token all
Token ID477
Feature activation+4.95e-02

Pos logit contributions
hy+1.627
Mic+1.484
Roger+1.294
Red+1.293
Robert+1.288
Neg logit contributions
riage-1.060
ry-1.053
ipples-1.052
flower-1.034
ac-1.009
Token day
Token ID1110
Feature activation+3.55e+00

Pos logit contributions
hy+0.110
Mic+0.100
 rides+0.090
Red+0.089
Roger+0.088
Neg logit contributions
riage-0.069
ipples-0.067
ry-0.065
flower-0.065
ac-0.063
Token long
Token ID890
Feature activation+1.77e+00

Pos logit contributions
hy+8.636
Mic+7.798
Red+7.134
Blue+6.915
Bow+6.893
Neg logit contributions
ry-5.126
ipples-4.708
 again-4.698
andy-4.676
 boil-4.417
Token.
Token ID13
Feature activation-3.50e-01

Pos logit contributions
hy+4.177
Mic+3.948
Red+3.491
Roger+3.460
Robert+3.447
Neg logit contributions
ry-2.150
ipples-2.093
riage-2.068
ac-2.048
en-2.024
Token One
Token ID1881
Feature activation-1.00e+00

Pos logit contributions
riage+0.618
ipples+0.608
ry+0.592
flower+0.588
ac+0.580
Neg logit contributions
hy-0.638
Mic-0.572
 rides-0.502
Red-0.498
Roger-0.487
Token day
Token ID1110
Feature activation+1.39e+00

Pos logit contributions
riage+1.648
ipples+1.617
 department+1.545
ry+1.533
ric+1.525
Neg logit contributions
hy-2.014
Mic-1.776
 rides-1.610
Red-1.576
Roger-1.534
Token,
Token ID11
Feature activation+8.14e-01

Pos logit contributions
hy+3.001
Mic+2.763
Red+2.445
 rides+2.397
nam+2.358
Neg logit contributions
ry-2.040
riage-1.987
 department-1.890
flower-1.887
ipples-1.885
Token play
Token ID711
Feature activation+4.74e-01

Pos logit contributions
riage+0.306
ipples+0.299
ry+0.291
flower+0.289
ac+0.283
Neg logit contributions
hy-0.406
Mic-0.367
 rides-0.328
Red-0.324
Roger-0.319
Token with
Token ID351
Feature activation+1.58e-01

Pos logit contributions
hy+0.858
Mic+0.747
Red+0.632
 rides+0.631
Robert+0.628
Neg logit contributions
riage-0.842
ipples-0.840
ry-0.824
flower-0.817
ac-0.804
Token his
Token ID465
Feature activation+7.49e-01

Pos logit contributions
hy+0.326
Mic+0.293
 rides+0.258
Red+0.258
Roger+0.254
Neg logit contributions
riage-0.243
ipples-0.239
ry-0.234
flower-0.232
andy-0.227
Token toys
Token ID14958
Feature activation+7.50e-01

Pos logit contributions
hy+1.528
Mic+1.340
Red+1.198
 rides+1.191
Robert+1.188
Neg logit contributions
flower-1.155
riage-1.144
ipples-1.140
ry-1.119
 department-1.113
Token all
Token ID477
Feature activation+4.95e-02

Pos logit contributions
hy+1.627
Mic+1.484
Roger+1.294
Red+1.293
Robert+1.288
Neg logit contributions
riage-1.060
ry-1.053
ipples-1.052
flower-1.034
ac-1.009
Token day
Token ID1110
Feature activation+3.55e+00

Pos logit contributions
hy+0.110
Mic+0.100
 rides+0.090
Red+0.089
Roger+0.088
Neg logit contributions
riage-0.069
ipples-0.067
ry-0.065
flower-0.065
ac-0.063
Token long
Token ID890
Feature activation+1.77e+00

Pos logit contributions
hy+8.636
Mic+7.798
Red+7.134
Blue+6.915
Bow+6.893
Neg logit contributions
ry-5.126
ipples-4.708
 again-4.698
andy-4.676
 boil-4.417
Token.
Token ID13
Feature activation-3.50e-01

Pos logit contributions
hy+4.177
Mic+3.948
Red+3.491
Roger+3.460
Robert+3.447
Neg logit contributions
ry-2.150
ipples-2.093
riage-2.068
ac-2.048
en-2.024
Token One
Token ID1881
Feature activation-1.00e+00

Pos logit contributions
riage+0.618
ipples+0.608
ry+0.592
flower+0.588
ac+0.580
Neg logit contributions
hy-0.638
Mic-0.572
 rides-0.502
Red-0.498
Roger-0.487
Token day
Token ID1110
Feature activation+1.39e+00

Pos logit contributions
riage+1.648
ipples+1.617
 department+1.545
ry+1.533
ric+1.525
Neg logit contributions
hy-2.014
Mic-1.776
 rides-1.610
Red-1.576
Roger-1.534
Token,
Token ID11
Feature activation+8.14e-01

Pos logit contributions
hy+3.001
Mic+2.763
Red+2.445
 rides+2.397
nam+2.358
Neg logit contributions
ry-2.040
riage-1.987
 department-1.890
flower-1.887
ipples-1.885
Tokenilly
Token ID6548
Feature activation+6.26e-01

Pos logit contributions
riage+0.506
ipples+0.489
ry+0.468
flower+0.466
 department+0.457
Neg logit contributions
hy-0.927
Mic-0.852
 rides-0.772
Red-0.764
Roger-0.754
Token liked
Token ID8288
Feature activation+2.84e-01

Pos logit contributions
hy+1.199
Mic+1.101
Red+0.947
 rides+0.946
Roger+0.942
Neg logit contributions
riage-1.016
ry-0.995
ipples-0.994
flower-0.992
andy-0.973
Token to
Token ID284
Feature activation-1.46e-01

Pos logit contributions
hy+0.663
Mic+0.608
Red+0.550
 rides+0.541
Roger+0.540
Neg logit contributions
riage-0.350
ipples-0.342
flower-0.332
ry-0.327
 department-0.323
Token play
Token ID711
Feature activation+6.02e-01

Pos logit contributions
riage+0.224
ipples+0.219
ry+0.212
flower+0.211
ac+0.207
Neg logit contributions
hy-0.300
Mic-0.272
 rides-0.243
Red-0.240
Roger-0.236
Token all
Token ID477
Feature activation-2.11e-01

Pos logit contributions
hy+1.096
Mic+0.946
Red+0.812
Robert+0.801
Roger+0.801
Neg logit contributions
ipples-1.063
riage-1.061
ry-1.048
flower-1.036
andy-1.025
Token day
Token ID1110
Feature activation+3.55e+00

Pos logit contributions
riage+0.300
ipples+0.292
ry+0.283
flower+0.281
 department+0.275
Neg logit contributions
hy-0.457
Mic-0.416
 rides-0.375
Red-0.371
Roger-0.364
Token long
Token ID890
Feature activation+2.31e+00

Pos logit contributions
hy+8.842
Mic+8.141
Red+7.570
Blue+7.225
Roger+7.041
Neg logit contributions
ry-4.993
andy-4.610
 again-4.579
en-4.550
ipples-4.502
Token.
Token ID13
Feature activation-6.89e-01

Pos logit contributions
hy+5.838
Mic+5.486
Red+4.912
Roger+4.874
Robert+4.868
Neg logit contributions
ry-2.590
en-2.537
 again-2.469
ipples-2.450
andy-2.368
Token She
Token ID1375
Feature activation-1.91e-02

Pos logit contributions
riage+1.354
ipples+1.339
ry+1.300
flower+1.294
 department+1.288
Neg logit contributions
hy-1.130
Mic-0.988
 rides-0.866
Red-0.846
Roger-0.816
Token was
Token ID373
Feature activation-3.98e-01

Pos logit contributions
riage+0.029
ipples+0.029
ry+0.028
flower+0.028
ac+0.027
Neg logit contributions
hy-0.039
Mic-0.035
 rides-0.031
Red-0.031
Roger-0.031
Token very
Token ID845
Feature activation+1.10e+00

Pos logit contributions
riage+0.748
ipples+0.728
ry+0.718
flower+0.706
andy+0.696
Neg logit contributions
hy-0.693
Mic-0.610
 rides-0.537
Red-0.526
Roger-0.514
Token
Token ID198
Feature activation+1.63e-01

Pos logit contributions
hy+2.212
Mic+1.963
nam+1.719
 rides+1.712
Red+1.695
Neg logit contributions
riage-1.482
ry-1.452
ipples-1.437
ucks-1.417
andy-1.407
Token
Token ID198
Feature activation-2.64e-01

Pos logit contributions
hy+0.312
Mic+0.278
 rides+0.245
Red+0.242
Roger+0.238
Neg logit contributions
riage-0.271
ipples-0.266
ry-0.262
flower-0.256
andy-0.254
TokenS
Token ID50
Feature activation-1.14e+00

Pos logit contributions
riage+0.396
ipples+0.386
ry+0.373
flower+0.372
ac+0.364
Neg logit contributions
hy-0.561
Mic-0.511
 rides-0.458
Red-0.455
Roger-0.446
Tokenara
Token ID3301
Feature activation-7.54e-01

Pos logit contributions
riage+1.261
ipples+1.247
flower+1.168
ac+1.128
ry+1.123
Neg logit contributions
hy-2.862
Mic-2.580
 rides-2.444
Red-2.392
Roger-2.317
Token offered
Token ID4438
Feature activation-1.58e+00

Pos logit contributions
riage+1.371
ipples+1.311
ry+1.257
flower+1.252
ric+1.245
Neg logit contributions
hy-1.359
Mic-1.213
 rides-1.084
Red-1.057
Roger-1.022
Token her
Token ID607
Feature activation-4.21e+00

Pos logit contributions
riage+2.589
ipples+2.562
ry+2.434
flower+2.423
ric+2.360
Neg logit contributions
hy-3.221
Mic-2.896
 rides-2.719
 Bow-2.621
Red-2.579
Token purple
Token ID14032
Feature activation-2.80e+00

Pos logit contributions
riage+9.835
ipples+8.619
ry+8.541
 department+8.514
pour+8.470
Neg logit contributions
 long-6.397
hy-6.179
 brace-5.970
 rides-5.878
 curly-5.407
Token bubble
Token ID14310
Feature activation-1.68e+00

Pos logit contributions
riage+4.445
ipples+3.967
 department+3.686
iz+3.659
ac+3.641
Neg logit contributions
hy-5.874
Mic-5.317
 rides-5.145
 brace-5.137
 long-5.023
Token wand
Token ID11569
Feature activation-1.02e+00

Pos logit contributions
riage+2.995
ipples+2.688
ric+2.605
ry+2.602
flower+2.590
Neg logit contributions
hy-3.237
Mic-2.760
 rides-2.588
 brace-2.521
hell-2.464
Token to
Token ID284
Feature activation-9.13e-01

Pos logit contributions
riage+1.867
ipples+1.797
andy+1.718
flower+1.717
ry+1.716
Neg logit contributions
hy-1.870
Mic-1.586
 rides-1.457
Red-1.391
 Bow-1.373
Token Max
Token ID5436
Feature activation-2.50e-01

Pos logit contributions
riage+1.297
ipples+1.259
ry+1.226
flower+1.209
ac+1.194
Neg logit contributions
hy-1.996
Mic-1.824
 rides-1.661
Red-1.641
Roger-1.600
Token butterfly
Token ID35113
Feature activation+6.63e-01

Pos logit contributions
hy+1.319
Mic+1.188
Red+1.072
 rides+1.048
Roger+1.045
Neg logit contributions
riage-0.908
flower-0.883
ipples-0.877
andy-0.844
ac-0.836
Token flew
Token ID13112
Feature activation-6.87e-01

Pos logit contributions
hy+1.329
Mic+1.228
Red+1.083
nam+1.060
Roger+1.058
Neg logit contributions
riage-1.035
ry-1.013
ipples-1.012
flower-1.004
andy-0.983
Token away
Token ID1497
Feature activation-1.87e-01

Pos logit contributions
riage+0.963
ipples+0.929
ry+0.900
flower+0.883
ac+0.869
Neg logit contributions
hy-1.530
Mic-1.374
 rides-1.253
Red-1.238
Roger-1.227
Token,
Token ID11
Feature activation-2.77e+00

Pos logit contributions
riage+0.330
ipples+0.323
ry+0.315
flower+0.313
ac+0.308
Neg logit contributions
hy-0.342
Mic-0.305
 rides-0.268
Red-0.265
Roger-0.260
Token Joe
Token ID5689
Feature activation+1.80e-01

Pos logit contributions
riage+6.133
ucks+6.047
ac+5.962
ric+5.930
flower+5.913
Neg logit contributions
hy-3.637
 long-3.136
 rides-2.986
 Bow-2.975
 Robert-2.935
Token and
Token ID290
Feature activation-3.73e+00

Pos logit contributions
hy+0.349
Mic+0.315
Red+0.276
 rides+0.275
Roger+0.270
Neg logit contributions
riage-0.292
ipples-0.288
ry-0.283
flower-0.282
andy-0.275
Token Linda
Token ID21776
Feature activation+8.41e-01

Pos logit contributions
riage+5.600
ipples+5.487
ric+5.435
flower+5.158
 department+5.146
Neg logit contributions
hy-8.356
Mic-7.379
 long-7.308
 Ellie-7.163
 Bow-7.006
Token decided
Token ID3066
Feature activation-4.75e-01

Pos logit contributions
hy+1.725
Mic+1.619
Red+1.420
Roger+1.395
Robert+1.371
Neg logit contributions
riage-1.259
ipples-1.232
ry-1.221
flower-1.210
andy-1.179
Token to
Token ID284
Feature activation-2.46e-01

Pos logit contributions
riage+0.672
ipples+0.664
flower+0.640
ry+0.638
ac+0.626
Neg logit contributions
hy-1.040
Mic-0.935
 rides-0.862
Red-0.839
Roger-0.833
Token share
Token ID2648
Feature activation-2.83e-01

Pos logit contributions
riage+0.347
ipples+0.338
ry+0.328
flower+0.325
ac+0.318
Neg logit contributions
hy-0.535
Mic-0.487
 rides-0.439
Red-0.434
Roger-0.427
Token a
Token ID257
Feature activation+6.75e-01

Pos logit contributions
riage+0.358
ipples+0.348
ry+0.336
flower+0.333
 department+0.325
Neg logit contributions
hy-0.658
Mic-0.603
 rides-0.550
Red-0.543
Roger-0.535
Token share
Token ID2648
Feature activation-2.83e-01

Pos logit contributions
riage+0.347
ipples+0.338
ry+0.328
flower+0.325
ac+0.318
Neg logit contributions
hy-0.535
Mic-0.487
 rides-0.439
Red-0.434
Roger-0.427
Token a
Token ID257
Feature activation+6.75e-01

Pos logit contributions
riage+0.358
ipples+0.348
ry+0.336
flower+0.333
 department+0.325
Neg logit contributions
hy-0.658
Mic-0.603
 rides-0.550
Red-0.543
Roger-0.535
Token hug
Token ID16225
Feature activation-1.15e+00

Pos logit contributions
hy+1.270
Mic+1.120
Red+0.990
 rides+0.983
Roger+0.965
Neg logit contributions
riage-1.135
ipples-1.119
ry-1.113
flower-1.110
ac-1.100
Token.
Token ID13
Feature activation-2.48e+00

Pos logit contributions
riage+1.999
ipples+1.990
andy+1.908
ry+1.894
ric+1.878
Neg logit contributions
hy-2.170
Mic-1.853
 rides-1.719
Red-1.638
 Bow-1.617
Token Joe
Token ID5689
Feature activation+2.93e-01

Pos logit contributions
riage+4.894
flower+4.818
ac+4.722
ipples+4.637
ric+4.496
Neg logit contributions
hy-4.184
Mic-3.518
 rides-3.317
 Robert-3.207
 Bow-3.140
Token and
Token ID290
Feature activation-3.71e+00

Pos logit contributions
hy+0.584
Mic+0.528
Red+0.464
 rides+0.461
Roger+0.456
Neg logit contributions
riage-0.462
ipples-0.455
ry-0.450
flower-0.449
andy-0.436
Token Linda
Token ID21776
Feature activation+6.20e-01

Pos logit contributions
riage+5.590
ipples+5.391
ric+5.288
flower+5.035
 department+4.981
Neg logit contributions
hy-8.342
Mic-7.548
 long-7.338
 Ellie-7.168
 brace-7.120
Token closed
Token ID4838
Feature activation-8.20e-01

Pos logit contributions
hy+1.253
Mic+1.153
Red+1.012
Roger+1.002
 rides+0.985
Neg logit contributions
riage-0.954
ipples-0.930
ry-0.920
flower-0.915
andy-0.896
Token their
Token ID511
Feature activation+9.82e-02

Pos logit contributions
riage+1.029
ipples+1.005
flower+0.953
ry+0.941
ac+0.927
Neg logit contributions
hy-1.937
Mic-1.752
 rides-1.637
Roger-1.588
Red-1.584
Token eyes
Token ID2951
Feature activation+8.56e-01

Pos logit contributions
hy+0.253
Mic+0.234
Red+0.213
 rides+0.212
Roger+0.210
Neg logit contributions
riage-0.100
ipples-0.096
ry-0.093
flower-0.092
ac-0.090
Token and
Token ID290
Feature activation+8.37e-01

Pos logit contributions
hy+1.777
Mic+1.680
Red+1.463
Roger+1.445
 rides+1.436
Neg logit contributions
ry-1.256
ipples-1.243
riage-1.241
flower-1.211
 department-1.199
Token on
Token ID319
Feature activation-3.03e-01

Pos logit contributions
riage+2.044
ipples+2.036
ry+1.932
andy+1.919
ac+1.911
Neg logit contributions
hy-1.813
Mic-1.599
 rides-1.451
Red-1.388
Roger-1.368
Token,
Token ID11
Feature activation-6.14e-01

Pos logit contributions
riage+0.547
ipples+0.539
ry+0.522
flower+0.521
ac+0.512
Neg logit contributions
hy-0.545
Mic-0.484
 rides-0.427
Red-0.419
Roger-0.412
Token Mom
Token ID11254
Feature activation-1.11e+00

Pos logit contributions
riage+1.051
ipples+1.033
ry+1.003
flower+0.999
ac+0.977
Neg logit contributions
hy-1.158
Mic-1.034
 rides-0.923
Red-0.911
Roger-0.888
Tokenmy
Token ID1820
Feature activation+3.43e-01

Pos logit contributions
ipples+1.411
riage+1.393
ac+1.295
ry+1.288
andy+1.264
Neg logit contributions
hy-2.579
Mic-2.334
 rides-2.149
Red-2.106
Roger-2.048
Token!
Token ID0
Feature activation-1.24e+00

Pos logit contributions
hy+0.664
Mic+0.603
Red+0.527
 rides+0.525
Roger+0.517
Neg logit contributions
riage-0.566
ry-0.547
ipples-0.545
flower-0.535
 department-0.523
Token Hur
Token ID10077
Feature activation-3.68e+00

Pos logit contributions
ipples+2.314
riage+2.271
andy+2.233
ry+2.224
flower+2.212
Neg logit contributions
hy-2.223
Mic-1.897
Red-1.696
 rides-1.663
 Bow-1.618
Tokenry
Token ID563
Feature activation-1.98e+00

Pos logit contributions
ipples+4.948
ac+4.155
ull+3.992
andy+3.750
ainer+3.695
Neg logit contributions
hy-9.593
 rides-7.860
ve-7.591
Mic-7.561
Red-7.514
Token up
Token ID510
Feature activation-1.52e-01

Pos logit contributions
ipples+3.622
riage+3.606
flower+3.474
ac+3.409
ric+3.324
Neg logit contributions
hy-3.575
 rides-3.077
Mic-2.982
Red-2.729
 Robert-2.725
Token!"
Token ID2474
Feature activation+1.88e+00

Pos logit contributions
riage+0.256
ipples+0.251
ry+0.244
flower+0.243
ac+0.238
Neg logit contributions
hy-0.289
Mic-0.260
 rides-0.229
Red-0.227
Roger-0.222
Token he
Token ID339
Feature activation+3.67e-01

Pos logit contributions
hy+4.090
Mic+3.900
Red+3.384
Robert+3.380
Roger+3.342
Neg logit contributions
ry-2.731
ipples-2.599
flower-2.570
riage-2.527
 M-2.515
Token said
Token ID531
Feature activation+4.94e-01

Pos logit contributions
hy+0.603
Mic+0.538
Red+0.455
 rides+0.449
Roger+0.443
Neg logit contributions
riage-0.704
ipples-0.697
ry-0.689
flower-0.677
 department-0.673
Token tea
Token ID8887
Feature activation-1.01e+00

Pos logit contributions
riage+2.909
ipples+2.866
ull+2.715
ry+2.680
flower+2.679
Neg logit contributions
hy-3.839
 rides-3.267
Mic-3.257
nam-3.066
 brace-3.025
Token and
Token ID290
Feature activation-7.83e-01

Pos logit contributions
riage+1.740
ipples+1.652
ry+1.586
ric+1.584
flower+1.572
Neg logit contributions
hy-1.908
Mic-1.691
 rides-1.547
Red-1.467
 brace-1.430
Token cookies
Token ID14746
Feature activation-4.97e-01

Pos logit contributions
riage+1.170
ipples+1.140
ry+1.108
flower+1.100
andy+1.081
Neg logit contributions
hy-1.638
Mic-1.469
 rides-1.369
Red-1.304
nam-1.302
Token to
Token ID284
Feature activation-7.37e-01

Pos logit contributions
riage+0.828
ipples+0.803
ry+0.776
flower+0.773
andy+0.767
Neg logit contributions
hy-0.959
Mic-0.853
 rides-0.768
Red-0.750
Roger-0.734
Token the
Token ID262
Feature activation-1.28e+00

Pos logit contributions
riage+0.987
ipples+0.964
flower+0.936
ry+0.935
 department+0.909
Neg logit contributions
hy-1.658
Mic-1.518
 rides-1.390
Red-1.374
nam-1.342
Token living
Token ID2877
Feature activation-3.60e+00

Pos logit contributions
riage+1.823
ipples+1.762
ry+1.719
ac+1.713
flower+1.686
Neg logit contributions
hy-2.793
Mic-2.473
 rides-2.337
Red-2.270
Roger-2.182
Token room
Token ID2119
Feature activation-1.26e+00

Pos logit contributions
riage+6.517
 department+6.303
flower+6.083
ipples+6.050
andy+6.023
Neg logit contributions
hy-6.267
 rides-5.861
nam-5.274
 wra-5.232
 long-5.163
Token.
Token ID13
Feature activation-8.65e-01

Pos logit contributions
riage+2.257
ipples+2.186
flower+2.138
andy+2.137
ric+2.095
Neg logit contributions
hy-2.315
Mic-1.881
 rides-1.844
Red-1.686
nam-1.672
Token They
Token ID1119
Feature activation-3.38e-01

Pos logit contributions
riage+1.419
ipples+1.393
flower+1.374
ry+1.359
ac+1.343
Neg logit contributions
hy-1.714
Mic-1.526
 rides-1.387
Red-1.381
 Bow-1.324
Token sit
Token ID1650
Feature activation-2.04e+00

Pos logit contributions
riage+0.511
ipples+0.498
ry+0.484
flower+0.481
ac+0.471
Neg logit contributions
hy-0.703
Mic-0.632
 rides-0.573
Red-0.562
Roger-0.550
Token on
Token ID319
Feature activation-9.34e-01

Pos logit contributions
riage+4.312
ipples+4.185
andy+4.108
 department+4.069
ric+4.009
Neg logit contributions
hy-2.953
Mic-2.534
 rides-2.350
nam-2.292
Red-2.249
TokenThey
Token ID2990
Feature activation-1.93e-01

Pos logit contributions
hy+0.312
Mic+0.282
 rides+0.253
Red+0.250
Roger+0.246
Neg logit contributions
riage-0.209
ipples-0.206
ry-0.200
flower-0.198
andy-0.195
Token go
Token ID467
Feature activation-1.39e+00

Pos logit contributions
riage+0.293
ipples+0.286
ry+0.278
flower+0.276
ac+0.271
Neg logit contributions
hy-0.398
Mic-0.360
 rides-0.323
Red-0.319
Roger-0.313
Token back
Token ID736
Feature activation-9.34e-01

Pos logit contributions
riage+2.160
ipples+2.114
andy+2.030
flower+2.015
ac+1.993
Neg logit contributions
hy-2.890
Mic-2.520
 rides-2.466
Red-2.284
Robert-2.256
Token to
Token ID284
Feature activation-1.45e+00

Pos logit contributions
riage+1.435
ipples+1.403
flower+1.332
andy+1.323
ac+1.315
Neg logit contributions
hy-1.922
Mic-1.722
 rides-1.622
Red-1.529
Roger-1.502
Token the
Token ID262
Feature activation-2.19e+00

Pos logit contributions
ipples+2.097
riage+2.096
ac+2.052
ry+2.000
flower+1.998
Neg logit contributions
hy-3.084
Mic-2.797
 rides-2.634
 Robert-2.546
nam-2.516
Token living
Token ID2877
Feature activation-3.56e+00

Pos logit contributions
riage+3.534
ipples+3.340
ac+3.327
flower+3.220
ry+3.158
Neg logit contributions
hy-4.353
Mic-3.992
 rides-3.840
Red-3.581
 long-3.573
Token room
Token ID2119
Feature activation-4.92e-01

Pos logit contributions
riage+6.752
 department+6.429
andy+6.176
flower+6.114
ric+6.080
Neg logit contributions
hy-5.753
 rides-5.641
nam-5.412
Mic-5.095
 long-4.989
Token and
Token ID290
Feature activation-4.31e-01

Pos logit contributions
riage+0.843
ipples+0.822
flower+0.798
ry+0.790
andy+0.785
Neg logit contributions
hy-0.927
Mic-0.811
 rides-0.738
Red-0.713
Roger-0.698
Token pick
Token ID2298
Feature activation-1.33e+00

Pos logit contributions
riage+0.654
ipples+0.644
flower+0.621
ry+0.619
ac+0.610
Neg logit contributions
hy-0.893
Mic-0.798
 rides-0.731
Red-0.710
Roger-0.697
Token up
Token ID510
Feature activation-1.04e+00

Pos logit contributions
riage+2.082
ipples+1.976
flower+1.913
 department+1.898
ric+1.893
Neg logit contributions
hy-2.699
Mic-2.434
 rides-2.342
Red-2.215
 Robert-2.173
Token the
Token ID262
Feature activation-1.70e+00

Pos logit contributions
riage+1.375
ipples+1.332
flower+1.287
ry+1.264
ac+1.246
Neg logit contributions
hy-2.394
Mic-2.139
 rides-2.053
Red-1.952
nam-1.930
Token to
Token ID284
Feature activation-4.40e-02

Pos logit contributions
riage+1.092
ipples+1.071
flower+1.037
ry+1.036
andy+1.013
Neg logit contributions
hy-1.615
Mic-1.453
 rides-1.345
Red-1.298
Roger-1.290
Token bat
Token ID7365
Feature activation-2.89e-02

Pos logit contributions
riage+0.056
ipples+0.054
ry+0.052
flower+0.052
ac+0.051
Neg logit contributions
hy-0.102
Mic-0.093
 rides-0.084
Red-0.084
Roger-0.082
Tokenhe
Token ID258
Feature activation-1.70e-02

Pos logit contributions
riage+0.076
ipples+0.075
ry+0.074
flower+0.074
ac+0.073
Neg logit contributions
hy-0.027
Mic-0.022
 rides-0.016
Red-0.015
Roger-0.015
Token first
Token ID717
Feature activation-5.81e-01

Pos logit contributions
riage+0.027
ipples+0.027
ry+0.026
flower+0.026
ac+0.025
Neg logit contributions
hy-0.034
Mic-0.031
 rides-0.027
Red-0.027
Roger-0.026
Token.
Token ID13
Feature activation-6.64e-01

Pos logit contributions
riage+0.944
ipples+0.929
flower+0.900
ry+0.889
 department+0.882
Neg logit contributions
hy-1.131
Mic-1.008
 rides-0.913
Red-0.894
Roger-0.878
Token Hur
Token ID10077
Feature activation-3.56e+00

Pos logit contributions
riage+1.151
ipples+1.147
flower+1.115
ry+1.109
 department+1.091
Neg logit contributions
hy-1.228
Mic-1.089
 rides-0.960
Red-0.958
Roger-0.926
Tokenry
Token ID563
Feature activation-2.02e+00

Pos logit contributions
ipples+3.454
ac+2.942
ull+2.689
flower+2.637
andy+2.544
Neg logit contributions
hy-10.097
 rides-8.686
cases-8.510
Mic-8.404
ve-8.284
Token up
Token ID510
Feature activation-6.71e-01

Pos logit contributions
riage+3.720
ipples+3.693
flower+3.564
ac+3.446
ric+3.412
Neg logit contributions
hy-3.537
Mic-2.999
 rides-2.973
Red-2.708
 Robert-2.700
Token,
Token ID11
Feature activation-3.21e-01

Pos logit contributions
riage+1.108
ipples+1.091
flower+1.068
ry+1.039
 department+1.033
Neg logit contributions
hy-1.282
Mic-1.147
 rides-1.039
Red-1.010
Roger-1.000
Token or
Token ID393
Feature activation+1.27e-01

Pos logit contributions
riage+0.497
ipples+0.485
ry+0.471
flower+0.470
 department+0.459
Neg logit contributions
hy-0.654
Mic-0.587
 rides-0.529
Red-0.521
Roger-0.512
Token the
Token ID262
Feature activation-4.50e-01

Pos logit contributions
hy+0.260
Mic+0.236
 rides+0.209
Red+0.208
Roger+0.204
Neg logit contributions
riage-0.196
ipples-0.189
ry-0.186
flower-0.183
ac-0.180
TokenBob
Token ID18861
Feature activation-4.98e-01

Pos logit contributions
hy+2.581
Mic+2.241
nam+2.117
 rides+2.069
 brace+2.019
Neg logit contributions
riage-1.752
ipples-1.706
flower-1.670
ry-1.655
andy-1.630
Token said
Token ID531
Feature activation+9.30e-02

Pos logit contributions
riage+0.915
ipples+0.892
ry+0.861
flower+0.857
ac+0.849
Neg logit contributions
hy-0.879
Mic-0.784
 rides-0.688
Red-0.674
Roger-0.662
Token,
Token ID11
Feature activation+1.93e-01

Pos logit contributions
hy+0.176
Mic+0.159
 rides+0.140
Red+0.139
Roger+0.136
Neg logit contributions
riage-0.157
ipples-0.153
ry-0.151
flower-0.149
 department-0.147
Token "
Token ID366
Feature activation+6.21e-02

Pos logit contributions
hy+0.451
Mic+0.412
 rides+0.377
Red+0.372
Roger+0.363
Neg logit contributions
riage-0.239
ipples-0.228
ry-0.224
flower-0.223
 department-0.219
TokenMr
Token ID5246
Feature activation-1.06e+00

Pos logit contributions
hy+0.121
Mic+0.108
 rides+0.097
Red+0.094
Roger+0.092
Neg logit contributions
riage-0.103
ipples-0.101
flower-0.099
ry-0.098
ac-0.096
Token.
Token ID13
Feature activation-3.54e+00

Pos logit contributions
riage+1.788
ipples+1.732
ry+1.678
ac+1.669
flower+1.650
Neg logit contributions
hy-2.036
Mic-1.793
 rides-1.581
Red-1.565
 Bow-1.531
Token Ant
Token ID3738
Feature activation-4.94e-01

Pos logit contributions
riage+5.020
ipples+4.745
ucks+4.460
ac+4.401
en+4.345
Neg logit contributions
hy-8.095
Mic-7.638
 Stretch-6.722
 Bow-6.661
 long-6.532
Token,
Token ID11
Feature activation-4.63e-01

Pos logit contributions
riage+0.851
ipples+0.830
flower+0.800
ry+0.798
ac+0.789
Neg logit contributions
hy-0.925
Mic-0.817
 rides-0.731
Red-0.714
Roger-0.700
Token why
Token ID1521
Feature activation-2.55e-01

Pos logit contributions
riage+0.726
ipples+0.715
ry+0.689
flower+0.686
ac+0.673
Neg logit contributions
hy-0.940
Mic-0.842
 rides-0.757
Red-0.743
Roger-0.734
Token is
Token ID318
Feature activation+2.72e-01

Pos logit contributions
riage+0.368
ipples+0.359
ry+0.347
flower+0.345
ac+0.338
Neg logit contributions
hy-0.550
Mic-0.498
 rides-0.450
Red-0.444
Roger-0.437
Token the
Token ID262
Feature activation-1.30e+00

Pos logit contributions
hy+0.568
Mic+0.517
 rides+0.460
Red+0.456
Roger+0.445
Neg logit contributions
riage-0.408
ipples-0.394
ry-0.382
flower-0.381
 department-0.375
Token surrender
Token ID16908
Feature activation-3.42e-01

Pos logit contributions
riage+0.187
ipples+0.182
ry+0.176
flower+0.174
ac+0.170
Neg logit contributions
hy-0.330
Mic-0.301
 rides-0.273
Red-0.270
Roger-0.266
Token if
Token ID611
Feature activation-3.07e-01

Pos logit contributions
riage+0.557
ipples+0.544
flower+0.527
ry+0.525
 department+0.515
Neg logit contributions
hy-0.670
Mic-0.602
 rides-0.537
Red-0.529
Roger-0.521
Token it
Token ID340
Feature activation-5.30e-02

Pos logit contributions
riage+0.655
ipples+0.646
ry+0.631
flower+0.631
ac+0.620
Neg logit contributions
hy-0.442
Mic-0.381
 rides-0.324
Red-0.316
Roger-0.309
Token is
Token ID318
Feature activation-2.32e-01

Pos logit contributions
riage+0.067
ipples+0.065
ry+0.063
flower+0.062
ac+0.061
Neg logit contributions
hy-0.123
Mic-0.113
 rides-0.102
Red-0.102
Roger-0.100
Token the
Token ID262
Feature activation-1.60e+00

Pos logit contributions
riage+0.383
ipples+0.374
ry+0.365
flower+0.362
andy+0.355
Neg logit contributions
hy-0.451
Mic-0.405
 rides-0.360
Red-0.356
Roger-0.350
Token right
Token ID826
Feature activation-3.53e+00

Pos logit contributions
ry+2.502
riage+2.477
ipples+2.453
ucks+2.375
flower+2.352
Neg logit contributions
hy-3.304
Mic-2.892
 rides-2.696
Red-2.579
nam-2.529
Token thing
Token ID1517
Feature activation+4.16e-01

Pos logit contributions
ipples+5.434
andy+5.305
riage+5.246
ric+5.209
ucks+5.095
Neg logit contributions
hy-7.290
 rides-6.140
 long-5.990
 Robert-5.685
 than-5.685
Token to
Token ID284
Feature activation+7.95e-01

Pos logit contributions
hy+0.901
Mic+0.834
Red+0.738
 rides+0.734
Roger+0.723
Neg logit contributions
riage-0.589
ry-0.571
ipples-0.568
flower-0.556
 department-0.546
Token do
Token ID466
Feature activation+3.39e-01

Pos logit contributions
hy+1.793
Mic+1.647
Red+1.464
 rides+1.442
nam+1.435
Neg logit contributions
riage-1.092
ipples-1.048
ry-1.000
flower-0.999
andy-0.983
Token,
Token ID11
Feature activation-7.36e-01

Pos logit contributions
hy+0.706
Mic+0.647
Red+0.564
 rides+0.559
Roger+0.552
Neg logit contributions
riage-0.515
ipples-0.499
ry-0.493
ac-0.478
flower-0.474
Token without
Token ID1231
Feature activation+4.63e-01

Pos logit contributions
riage+1.281
ipples+1.248
ry+1.214
flower+1.211
andy+1.189
Neg logit contributions
hy-1.359
Mic-1.210
 rides-1.084
Red-1.060
Roger-1.044
Token with
Token ID351
Feature activation-1.07e+00

Pos logit contributions
riage+2.335
ipples+2.228
flower+2.183
ry+2.174
andy+2.174
Neg logit contributions
hy-1.682
Mic-1.485
 rides-1.355
Red-1.292
Roger-1.230
Token the
Token ID262
Feature activation-3.99e-01

Pos logit contributions
riage+1.520
ipples+1.442
ry+1.398
flower+1.390
andy+1.373
Neg logit contributions
hy-2.334
Mic-2.138
 rides-2.001
nam-1.951
Red-1.944
Token ball
Token ID2613
Feature activation-9.03e-01

Pos logit contributions
riage+0.667
ipples+0.647
ry+0.635
flower+0.624
ac+0.616
Neg logit contributions
hy-0.777
Mic-0.697
 rides-0.623
Red-0.616
Roger-0.601
Token in
Token ID287
Feature activation-1.60e+00

Pos logit contributions
riage+1.630
ipples+1.561
ry+1.504
flower+1.504
andy+1.497
Neg logit contributions
hy-1.627
Mic-1.443
 rides-1.309
Red-1.265
Roger-1.213
Token the
Token ID262
Feature activation-1.73e+00

Pos logit contributions
riage+2.451
ipples+2.361
 department+2.322
flower+2.297
ry+2.294
Neg logit contributions
hy-3.314
Mic-3.046
 rides-2.816
Red-2.768
nam-2.712
Token living
Token ID2877
Feature activation-3.52e+00

Pos logit contributions
riage+3.007
ipples+2.890
ry+2.874
ull+2.742
ac+2.736
Neg logit contributions
hy-3.433
Mic-3.124
 rides-2.863
Red-2.818
Blue-2.623
Token room
Token ID2119
Feature activation-1.59e+00

Pos logit contributions
riage+6.597
 department+6.310
flower+6.176
ipples+6.107
andy+5.908
Neg logit contributions
hy-6.257
 rides-5.770
Mic-5.228
nam-5.096
 long-5.037
Token.
Token ID13
Feature activation-1.82e+00

Pos logit contributions
riage+2.924
ipples+2.851
flower+2.808
ric+2.804
andy+2.779
Neg logit contributions
hy-2.801
 rides-2.255
Mic-2.247
Red-2.034
nam-2.002
Token They
Token ID1119
Feature activation+2.79e-01

Pos logit contributions
riage+3.436
ac+3.392
ry+3.323
ipples+3.301
flower+3.277
Neg logit contributions
hy-3.212
Mic-2.705
 rides-2.507
Red-2.452
 Robert-2.387
Token forget
Token ID6044
Feature activation-1.66e+00

Pos logit contributions
hy+0.567
Mic+0.516
Red+0.452
 rides+0.450
Roger+0.449
Neg logit contributions
riage-0.434
ipples-0.427
ry-0.413
flower-0.412
andy-0.402
Token what
Token ID644
Feature activation-2.01e+00

Pos logit contributions
ipples+3.127
flower+3.058
riage+3.035
ac+2.929
andy+2.909
Neg logit contributions
hy-2.956
Mic-2.512
 rides-2.501
 Bow-2.349
 Robert-2.331
Token
Token ID198
Feature activation-5.62e-01

Pos logit contributions
riage+0.565
ipples+0.549
flower+0.535
ry+0.534
 department+0.523
Neg logit contributions
hy-0.712
Mic-0.643
 rides-0.575
Red-0.568
Roger-0.558
TokenMom
Token ID29252
Feature activation-9.86e-01

Pos logit contributions
riage+0.856
ipples+0.825
flower+0.805
ry+0.803
 department+0.793
Neg logit contributions
hy-1.172
Mic-1.073
Red-0.959
 rides-0.958
Roger-0.939
Token comes
Token ID2058
Feature activation-1.58e+00

Pos logit contributions
riage+1.665
ipples+1.619
flower+1.554
ac+1.548
ry+1.545
Neg logit contributions
hy-1.893
Mic-1.661
 rides-1.588
Red-1.470
Robert-1.442
Token to
Token ID284
Feature activation-7.75e-01

Pos logit contributions
riage+2.508
ipples+2.349
flower+2.256
ric+2.226
andy+2.217
Neg logit contributions
hy-3.234
Mic-2.922
 rides-2.746
Roger-2.651
Robert-2.647
Token the
Token ID262
Feature activation-2.46e+00

Pos logit contributions
ipples+1.075
riage+1.068
ry+1.034
flower+1.027
ac+1.011
Neg logit contributions
hy-1.701
Mic-1.553
 rides-1.424
Red-1.394
Roger-1.382
Token living
Token ID2877
Feature activation-3.51e+00

Pos logit contributions
riage+4.426
ipples+4.283
ac+4.239
flower+4.151
andy+4.111
Neg logit contributions
hy-4.623
 rides-4.066
Mic-4.053
Red-3.830
 long-3.601
Token room
Token ID2119
Feature activation-5.18e-01

Pos logit contributions
riage+6.583
 department+6.188
andy+6.042
flower+6.028
ipples+5.996
Neg logit contributions
hy-5.801
 rides-5.640
nam-5.209
Mic-5.114
 long-4.886
Token and
Token ID290
Feature activation-3.16e-01

Pos logit contributions
riage+0.890
ipples+0.873
flower+0.844
ry+0.836
andy+0.830
Neg logit contributions
hy-0.972
Mic-0.848
 rides-0.774
Red-0.747
Roger-0.731
Token sees
Token ID7224
Feature activation-8.42e-01

Pos logit contributions
riage+0.486
ipples+0.478
flower+0.460
ry+0.460
ac+0.452
Neg logit contributions
hy-0.649
Mic-0.584
 rides-0.530
Red-0.519
Roger-0.509
Token their
Token ID511
Feature activation-2.19e+00

Pos logit contributions
riage+1.128
ipples+1.095
flower+1.066
ry+1.040
ac+1.040
Neg logit contributions
hy-1.906
Mic-1.713
 rides-1.619
Red-1.570
Roger-1.535
Token bridge
Token ID7696
Feature activation-1.13e+00

Pos logit contributions
riage+4.076
flower+3.793
ipples+3.775
ry+3.758
ac+3.702
Neg logit contributions
hy-4.052
 rides-3.392
Mic-3.252
 long-3.101
 brace-2.958
Token and
Token ID290
Feature activation+5.59e-01

Pos logit contributions
riage+1.170
ipples+1.133
flower+1.102
ry+1.086
ac+1.079
Neg logit contributions
hy-1.258
Mic-1.102
 rides-1.008
Red-0.971
Roger-0.942
Token sad
Token ID6507
Feature activation-7.29e-01

Pos logit contributions
hy+1.015
Mic+0.948
Red+0.807
Roger+0.803
Robert+0.797
Neg logit contributions
riage-0.966
ipples-0.953
ry-0.931
flower-0.920
 department-0.906
Token.
Token ID13
Feature activation-6.43e-02

Pos logit contributions
riage+1.249
ipples+1.217
flower+1.188
ac+1.167
ry+1.160
Neg logit contributions
hy-1.384
Mic-1.212
 rides-1.109
Red-1.077
Roger-1.034
Token She
Token ID1375
Feature activation+7.49e-01

Pos logit contributions
riage+0.136
ipples+0.134
ry+0.131
flower+0.131
andy+0.129
Neg logit contributions
hy-0.095
Mic-0.083
 rides-0.070
Red-0.069
Roger-0.067
Token wants
Token ID3382
Feature activation-1.30e+00

Pos logit contributions
hy+1.674
Mic+1.554
Roger+1.376
Red+1.369
Robert+1.367
Neg logit contributions
riage-0.993
ipples-0.983
ry-0.943
andy-0.936
ric-0.915
Token her
Token ID607
Feature activation-3.50e+00

Pos logit contributions
riage+2.136
ipples+2.073
ry+2.025
flower+2.018
ac+1.963
Neg logit contributions
hy-2.648
Mic-2.364
 rides-2.204
 Bow-2.096
Red-2.087
Token tool
Token ID2891
Feature activation-1.28e+00

Pos logit contributions
riage+6.148
ric+5.755
ipples+5.397
ucks+5.248
 department+5.104
Neg logit contributions
hy-6.707
 long-6.651
Mic-6.222
 rides-6.150
powers-5.881
Token back
Token ID736
Feature activation-8.09e-01

Pos logit contributions
riage+2.016
ipples+1.891
ry+1.838
ric+1.820
andy+1.794
Neg logit contributions
hy-2.675
Mic-2.398
 rides-2.207
Red-2.112
 brace-2.066
Token.
Token ID13
Feature activation+4.67e-01

Pos logit contributions
riage+1.396
ipples+1.369
flower+1.306
ry+1.291
andy+1.289
Neg logit contributions
hy-1.515
Mic-1.338
 rides-1.212
Red-1.171
Roger-1.151
Token She
Token ID1375
Feature activation+4.02e-01

Pos logit contributions
hy+0.743
Mic+0.653
 rides+0.559
Red+0.545
Roger+0.543
Neg logit contributions
riage-0.931
ipples-0.908
ry-0.892
andy-0.885
flower-0.885
Token dives
Token ID46638
Feature activation-1.47e+00

Pos logit contributions
hy+0.860
Mic+0.786
Red+0.693
Roger+0.690
Robert+0.683
Neg logit contributions
riage-0.581
ipples-0.566
ry-0.554
andy-0.544
flower-0.538
Token like
Token ID588
Feature activation-3.06e-01

Pos logit contributions
hy+0.450
Mic+0.407
 rides+0.347
Red+0.346
Roger+0.338
Neg logit contributions
riage-0.479
ipples-0.471
ry-0.459
flower-0.456
ac-0.452
Token a
Token ID257
Feature activation-4.11e-01

Pos logit contributions
riage+0.478
ipples+0.466
ry+0.455
flower+0.451
ac+0.443
Neg logit contributions
hy-0.618
Mic-0.556
 rides-0.499
Red-0.492
Roger-0.484
Token toy
Token ID13373
Feature activation+4.33e-01

Pos logit contributions
riage+0.681
ipples+0.665
ry+0.648
flower+0.637
ac+0.632
Neg logit contributions
hy-0.795
Mic-0.717
 rides-0.634
Red-0.629
Roger-0.617
Token!
Token ID0
Feature activation-2.32e+00

Pos logit contributions
hy+0.889
Mic+0.823
Red+0.723
 rides+0.719
Roger+0.717
Neg logit contributions
riage-0.652
ipples-0.634
flower-0.631
ry-0.625
 department-0.606
Token Joe
Token ID5689
Feature activation+4.79e-01

Pos logit contributions
riage+4.284
flower+4.142
ac+4.078
ipples+4.035
ric+3.944
Neg logit contributions
hy-4.141
Mic-3.626
 rides-3.386
 Bow-3.265
 Robert-3.246
Token and
Token ID290
Feature activation-3.50e+00

Pos logit contributions
hy+0.968
Mic+0.876
Red+0.772
 rides+0.769
Roger+0.757
Neg logit contributions
ipples-0.733
riage-0.731
ry-0.729
flower-0.726
andy-0.710
Token Linda
Token ID21776
Feature activation+1.03e+00

Pos logit contributions
riage+5.100
ipples+5.026
ric+4.732
 department+4.576
flower+4.478
Neg logit contributions
hy-8.005
Mic-7.360
 long-6.924
 Bow-6.795
 Ellie-6.772
Token were
Token ID547
Feature activation-1.01e-01

Pos logit contributions
hy+2.168
Mic+2.050
Red+1.793
Roger+1.770
Robert+1.736
Neg logit contributions
riage-1.464
ipples-1.456
ry-1.445
flower-1.424
andy-1.389
Token so
Token ID523
Feature activation+9.99e-01

Pos logit contributions
riage+0.195
ipples+0.192
ry+0.187
flower+0.186
ac+0.183
Neg logit contributions
hy-0.170
Mic-0.150
 rides-0.130
Red-0.128
Roger-0.125
Token amazed
Token ID24562
Feature activation-3.30e-01

Pos logit contributions
hy+1.729
Mic+1.620
 rides+1.351
Red+1.350
Roger+1.328
Neg logit contributions
ipples-1.903
riage-1.817
flower-1.792
ac-1.785
ry-1.783
Token and
Token ID290
Feature activation-7.40e-01

Pos logit contributions
riage+0.544
ipples+0.530
ry+0.513
flower+0.512
ac+0.505
Neg logit contributions
hy-0.641
Mic-0.575
 rides-0.513
Red-0.505
Roger-0.497
Token the
Token ID262
Feature activation-1.95e+00

Pos logit contributions
riage+0.995
ipples+0.986
ry+0.939
flower+0.938
ac+0.921
Neg logit contributions
hy-1.434
Mic-1.293
 rides-1.187
Red-1.153
Roger-1.138
Token f
Token ID277
Feature activation-2.31e+00

Pos logit contributions
riage+2.503
ac+2.236
ipples+2.221
flower+2.197
ry+2.184
Neg logit contributions
hy-4.726
Mic-4.075
 rides-4.032
 long-3.909
Red-3.806
Tokenauc
Token ID14272
Feature activation-1.71e+00

Pos logit contributions
riage+2.165
flower+1.973
ac+1.760
ipples+1.752
 department+1.715
Neg logit contributions
hy-6.405
Mic-5.585
 rides-5.388
looking-5.357
Red-5.293
Tokenet
Token ID316
Feature activation-1.86e+00

Pos logit contributions
riage+2.093
ipples+1.996
ry+1.955
flower+1.948
andy+1.891
Neg logit contributions
hy-4.112
Mic-3.703
 rides-3.495
Red-3.359
 Bow-3.211
Token.
Token ID13
Feature activation-1.07e+00

Pos logit contributions
riage+3.397
ipples+3.311
andy+3.160
ucks+3.144
ac+3.137
Neg logit contributions
hy-3.443
Mic-2.799
 rides-2.686
Red-2.518
 Bow-2.419
Token The
Token ID383
Feature activation-3.49e+00

Pos logit contributions
ipples+2.207
riage+2.182
flower+2.122
ac+2.087
ry+2.079
Neg logit contributions
hy-1.707
Mic-1.457
 rides-1.291
Red-1.272
 Robert-1.199
Token water
Token ID1660
Feature activation-2.30e+00

Pos logit contributions
riage+6.544
ric+6.016
ry+5.955
ipples+5.807
ill+5.750
Neg logit contributions
hy-6.952
 long-6.071
Mic-5.637
 rides-5.632
 curly-5.537
Token came
Token ID1625
Feature activation-1.10e+00

Pos logit contributions
riage+3.892
ipples+3.449
ucks+3.446
ric+3.437
andy+3.411
Neg logit contributions
hy-4.664
Mic-3.863
 rides-3.833
 brace-3.523
Red-3.510
Token out
Token ID503
Feature activation-1.60e+00

Pos logit contributions
riage+1.845
ipples+1.726
ry+1.656
ac+1.620
andy+1.618
Neg logit contributions
hy-2.162
Mic-1.948
Red-1.720
 rides-1.714
Roger-1.687
Token and
Token ID290
Feature activation-1.58e+00

Pos logit contributions
riage+3.162
ipples+2.977
ric+2.924
flower+2.867
ours+2.857
Neg logit contributions
hy-2.757
Mic-2.339
 rides-2.057
Red-2.017
 long-2.014
Token Lily
Token ID20037
Feature activation-1.22e+00

Pos logit contributions
ipples+2.769
riage+2.741
ric+2.629
andy+2.627
ac+2.625
Neg logit contributions
hy-2.940
Mic-2.627
 rides-2.408
Red-2.374
nam-2.362
Token surrender
Token ID16908
Feature activation-5.89e-01

Pos logit contributions
hy+0.698
Mic+0.643
 rides+0.582
Red+0.576
Roger+0.568
Neg logit contributions
riage-0.388
ipples-0.375
ry-0.361
flower-0.360
andy-0.351
Token if
Token ID611
Feature activation-1.95e-01

Pos logit contributions
riage+0.959
ipples+0.939
flower+0.906
ry+0.899
 department+0.888
Neg logit contributions
hy-1.147
Mic-1.030
 rides-0.924
Red-0.908
Roger-0.895
Token it
Token ID340
Feature activation+9.79e-02

Pos logit contributions
riage+0.420
ipples+0.413
ry+0.404
flower+0.403
ac+0.397
Neg logit contributions
hy-0.278
Mic-0.240
 rides-0.202
Red-0.198
Roger-0.193
Token is
Token ID318
Feature activation-7.67e-02

Pos logit contributions
hy+0.236
Mic+0.218
Red+0.196
 rides+0.196
Roger+0.194
Neg logit contributions
riage-0.115
ipples-0.111
ry-0.108
flower-0.107
ac-0.104
Token the
Token ID262
Feature activation-1.97e+00

Pos logit contributions
riage+0.122
ipples+0.119
ry+0.116
flower+0.115
ac+0.113
Neg logit contributions
hy-0.153
Mic-0.138
 rides-0.123
Red-0.122
Roger-0.119
Token right
Token ID826
Feature activation-3.48e+00

Pos logit contributions
ry+3.169
riage+3.125
ipples+3.073
ucks+3.009
ull+2.965
Neg logit contributions
hy-4.052
Mic-3.472
 rides-3.264
 long-3.146
Red-3.122
Token thing
Token ID1517
Feature activation+4.44e-01

Pos logit contributions
ipples+5.200
riage+5.098
andy+5.065
ric+5.021
 department+4.852
Neg logit contributions
hy-7.212
 rides-6.041
 long-5.942
 blonde-5.679
 curly-5.667
Token to
Token ID284
Feature activation+8.93e-01

Pos logit contributions
hy+0.970
Mic+0.897
Red+0.795
 rides+0.791
Roger+0.780
Neg logit contributions
riage-0.623
ry-0.599
ipples-0.599
flower-0.587
 department-0.578
Token do
Token ID466
Feature activation+4.67e-01

Pos logit contributions
hy+1.978
Mic+1.829
Red+1.615
 rides+1.594
nam+1.590
Neg logit contributions
riage-1.259
ipples-1.199
ry-1.163
flower-1.156
andy-1.133
Token!"
Token ID2474
Feature activation+3.75e-01

Pos logit contributions
hy+0.987
Mic+0.908
Red+0.790
 rides+0.785
Roger+0.775
Neg logit contributions
riage-0.695
ipples-0.666
ry-0.659
ac-0.641
flower-0.634
Token
Token ID198
Feature activation-3.38e-02

Pos logit contributions
hy+0.734
Mic+0.655
 rides+0.581
Red+0.577
Roger+0.557
Neg logit contributions
riage-0.617
ipples-0.599
ry-0.594
andy-0.584
flower-0.582
Token happily
Token ID18177
Feature activation+9.88e-01

Pos logit contributions
hy+2.183
Mic+2.025
Roger+1.761
Robert+1.759
Red+1.758
Neg logit contributions
ipples-1.347
riage-1.336
ry-1.335
flower-1.296
ac-1.270
Token.
Token ID13
Feature activation-4.96e-01

Pos logit contributions
hy+2.173
Mic+2.085
Red+1.819
Roger+1.802
Robert+1.778
Neg logit contributions
ry-1.377
ipples-1.369
riage-1.354
flower-1.315
ac-1.277
Token Then
Token ID3244
Feature activation-7.08e-01

Pos logit contributions
riage+0.994
ipples+0.988
flower+0.958
ry+0.955
ac+0.942
Neg logit contributions
hy-0.793
Mic-0.692
 rides-0.603
Red-0.593
Roger-0.569
Token Rose
Token ID8049
Feature activation-1.41e+00

Pos logit contributions
ipples+1.403
riage+1.389
flower+1.333
ry+1.323
ac+1.322
Neg logit contributions
hy-1.158
Mic-1.010
 rides-0.897
Red-0.863
Roger-0.845
Token showed
Token ID3751
Feature activation-6.90e-01

Pos logit contributions
riage+2.554
ipples+2.446
ric+2.372
flower+2.326
ac+2.319
Neg logit contributions
hy-2.549
Mic-2.305
 rides-2.163
Red-2.009
Roger-1.964
Token her
Token ID607
Feature activation-3.48e+00

Pos logit contributions
riage+1.292
ipples+1.281
flower+1.237
ry+1.220
 department+1.197
Neg logit contributions
hy-1.203
Mic-1.070
 rides-0.961
Red-0.926
Roger-0.902
Token how
Token ID703
Feature activation-7.82e-01

Pos logit contributions
riage+8.095
ipples+7.252
ucks+7.048
flower+7.030
 department+6.944
Neg logit contributions
hy-5.492
 rides-4.752
 long-4.557
 brace-4.260
powers-4.096
Token to
Token ID284
Feature activation-1.73e+00

Pos logit contributions
riage+1.175
ipples+1.125
flower+1.105
ry+1.087
 department+1.072
Neg logit contributions
hy-1.669
Mic-1.487
 rides-1.369
Red-1.332
Roger-1.314
Token turn
Token ID1210
Feature activation-1.43e+00

Pos logit contributions
riage+2.701
 department+2.511
ac+2.449
ipples+2.449
ry+2.408
Neg logit contributions
hy-3.635
Mic-3.183
 rides-3.021
Red-2.924
Blue-2.832
Token it
Token ID340
Feature activation-1.87e-02

Pos logit contributions
riage+2.375
ipples+2.209
 department+2.147
flower+2.137
ac+2.103
Neg logit contributions
hy-2.836
Mic-2.481
 rides-2.397
Red-2.268
 brace-2.236
Token off
Token ID572
Feature activation-1.63e-01

Pos logit contributions
riage+0.029
ipples+0.029
ry+0.028
flower+0.028
ac+0.027
Neg logit contributions
hy-0.038
Mic-0.034
 rides-0.031
Red-0.030
Roger-0.030
Token f
Token ID277
Feature activation-2.76e+00

Pos logit contributions
riage+2.929
ac+2.748
ipples+2.664
ry+2.648
ucks+2.632
Neg logit contributions
hy-4.440
Mic-3.980
 rides-3.788
Red-3.709
Roger-3.590
Tokenauc
Token ID14272
Feature activation-2.58e+00

Pos logit contributions
riage+2.910
 department+2.409
flower+2.370
 mistake+2.186
ac+2.178
Neg logit contributions
hy-7.474
Mic-6.428
looking-6.368
 rides-6.181
Red-6.164
Tokenet
Token ID316
Feature activation-2.63e+00

Pos logit contributions
riage+3.489
 department+3.279
ipples+3.200
flower+3.150
ric+3.125
Neg logit contributions
hy-5.817
Mic-5.132
 rides-4.930
Red-4.690
hell-4.592
Token to
Token ID284
Feature activation-1.66e+00

Pos logit contributions
riage+4.336
ipples+4.095
 department+4.034
ucks+3.938
ac+3.884
Neg logit contributions
hy-5.336
Mic-4.367
 rides-4.348
Red-3.952
dream-3.907
Token '
Token ID705
Feature activation-4.95e-01

Pos logit contributions
riage+2.707
ipples+2.636
flower+2.612
 department+2.602
ac+2.568
Neg logit contributions
hy-3.282
Mic-2.868
 rides-2.752
Red-2.607
Roger-2.546
Tokenspl
Token ID22018
Feature activation-3.48e+00

Pos logit contributions
riage+0.780
ipples+0.764
ry+0.742
flower+0.732
 department+0.730
Neg logit contributions
hy-0.987
Mic-0.892
 rides-0.796
Red-0.791
Roger-0.775
Tokenash
Token ID1077
Feature activation-2.25e+00

Pos logit contributions
riage+4.683
 department+4.666
ipples+4.102
 mistake+4.076
flower+4.052
Neg logit contributions
hy-8.185
Mic-7.264
hell-6.911
looking-6.763
Red-6.727
Token'
Token ID6
Feature activation-8.95e-01

Pos logit contributions
ipples+3.833
riage+3.755
 department+3.614
ucks+3.586
andy+3.506
Neg logit contributions
hy-4.291
Mic-3.741
Roger-3.376
 rides-3.324
Red-3.319
Token around
Token ID1088
Feature activation-4.41e-01

Pos logit contributions
riage+1.453
ipples+1.411
ry+1.358
flower+1.354
andy+1.345
Neg logit contributions
hy-1.766
Mic-1.554
 rides-1.420
Red-1.382
Roger-1.351
Token and
Token ID290
Feature activation-2.96e-01

Pos logit contributions
riage+0.722
ipples+0.708
ry+0.682
flower+0.680
 department+0.669
Neg logit contributions
hy-0.857
Mic-0.770
 rides-0.686
Red-0.676
Roger-0.667
Token have
Token ID423
Feature activation-7.17e-01

Pos logit contributions
riage+0.438
ipples+0.428
ry+0.414
flower+0.412
ac+0.404
Neg logit contributions
hy-0.623
Mic-0.563
 rides-0.508
Red-0.500
Roger-0.492
Token together
Token ID1978
Feature activation-1.05e-01

Pos logit contributions
hy+1.932
Mic+1.800
Red+1.598
 rides+1.567
Roger+1.552
Neg logit contributions
ipples-1.241
riage-1.241
ry-1.224
flower-1.200
ac-1.183
Token.
Token ID13
Feature activation-2.21e+00

Pos logit contributions
riage+0.165
ipples+0.161
ry+0.157
flower+0.156
ac+0.153
Neg logit contributions
hy-0.213
Mic-0.193
 rides-0.171
Red-0.170
Roger-0.167
Token
Token ID198
Feature activation-1.02e+00

Pos logit contributions
riage+4.780
flower+4.624
ac+4.621
ipples+4.509
 department+4.497
Neg logit contributions
hy-3.138
Mic-2.782
 rides-2.515
 Bow-2.420
Red-2.383
Token
Token ID198
Feature activation-2.12e+00

Pos logit contributions
riage+2.020
ipples+1.938
flower+1.929
 department+1.891
ry+1.860
Neg logit contributions
hy-1.652
Mic-1.498
 rides-1.304
Red-1.288
Roger-1.272
TokenJoe
Token ID19585
Feature activation-2.46e-01

Pos logit contributions
riage+4.199
flower+3.911
 department+3.890
ipples+3.868
ry+3.812
Neg logit contributions
hy-3.555
Mic-3.314
Red-2.923
 rides-2.846
Roger-2.837
Token and
Token ID290
Feature activation-3.47e+00

Pos logit contributions
riage+0.419
ipples+0.409
ry+0.397
flower+0.395
ac+0.388
Neg logit contributions
hy-0.465
Mic-0.417
 rides-0.369
Red-0.364
Roger-0.357
Token Linda
Token ID21776
Feature activation+1.31e+00

Pos logit contributions
riage+4.806
ipples+4.583
 department+4.350
ric+4.283
ucks+4.148
Neg logit contributions
hy-8.153
Mic-7.573
 long-7.007
 Ellie-6.918
nam-6.865
Token then
Token ID788
Feature activation-1.42e-03

Pos logit contributions
hy+2.784
Mic+2.637
Red+2.303
Roger+2.303
Robert+2.247
Neg logit contributions
riage-1.886
ry-1.833
flower-1.810
ipples-1.808
andy-1.761
Token kept
Token ID4030
Feature activation-1.62e-01

Pos logit contributions
riage+0.002
ipples+0.002
ry+0.002
flower+0.002
ac+0.002
Neg logit contributions
hy-0.003
Mic-0.003
 rides-0.002
Red-0.002
Roger-0.002
Token walking
Token ID6155
Feature activation-8.37e-01

Pos logit contributions
riage+0.261
ipples+0.255
ry+0.248
flower+0.247
ac+0.242
Neg logit contributions
hy-0.322
Mic-0.290
 rides-0.258
Red-0.255
Roger-0.251
Token in
Token ID287
Feature activation-1.02e+00

Pos logit contributions
riage+1.471
ipples+1.431
flower+1.386
ry+1.385
andy+1.382
Neg logit contributions
hy-1.563
Mic-1.351
 rides-1.245
Red-1.189
Roger-1.173
Token were
Token ID547
Feature activation+2.48e-01

Pos logit contributions
hy+0.270
Mic+0.245
 rides+0.215
Red+0.214
Roger+0.211
Neg logit contributions
riage-0.220
ipples-0.214
ry-0.210
flower-0.210
andy-0.204
Token walking
Token ID6155
Feature activation-2.82e-01

Pos logit contributions
hy+0.432
Mic+0.387
Red+0.333
 rides+0.330
Roger+0.321
Neg logit contributions
riage-0.457
ipples-0.450
ry-0.441
flower-0.439
ac-0.436
Token down
Token ID866
Feature activation-4.02e-01

Pos logit contributions
riage+0.442
ipples+0.432
ry+0.419
flower+0.416
andy+0.409
Neg logit contributions
hy-0.573
Mic-0.516
 rides-0.463
Red-0.456
Roger-0.449
Token the
Token ID262
Feature activation+1.71e+00

Pos logit contributions
riage+0.519
ipples+0.505
ry+0.487
flower+0.484
ac+0.475
Neg logit contributions
hy-0.930
Mic-0.844
 rides-0.776
Red-0.761
Roger-0.753
Token street
Token ID4675
Feature activation+4.31e-01

Pos logit contributions
hy+3.012
Mic+2.773
 blonde+2.280
hell+2.269
Roger+2.260
Neg logit contributions
riage-3.070
flower-3.062
ipples-3.038
ry-3.014
andy-2.984
Token.
Token ID13
Feature activation-3.46e+00

Pos logit contributions
hy+0.889
Mic+0.840
Red+0.730
Roger+0.719
 rides+0.715
Neg logit contributions
riage-0.648
ipples-0.636
ry-0.626
flower-0.617
 department-0.600
Token Sandy
Token ID17581
Feature activation+1.50e+00

Pos logit contributions
riage+7.472
flower+7.293
 department+7.235
ac+7.199
 mistake+7.045
Neg logit contributions
hy-5.242
 long-4.158
 blonde-4.049
Mic-3.984
 curly-3.983
Token was
Token ID373
Feature activation+1.56e-01

Pos logit contributions
hy+3.826
Mic+3.606
nam+3.201
Red+3.180
Robert+3.141
Neg logit contributions
ry-1.610
flower-1.502
andy-1.499
riage-1.492
ipples-1.478
Token feeling
Token ID4203
Feature activation+6.31e-01

Pos logit contributions
hy+0.275
Mic+0.249
 rides+0.214
Red+0.213
Roger+0.208
Neg logit contributions
riage-0.280
ipples-0.278
flower-0.271
ry-0.268
ac-0.264
Token cold
Token ID4692
Feature activation+1.21e+00

Pos logit contributions
hy+1.133
Mic+1.050
 rides+0.885
Red+0.880
Roger+0.857
Neg logit contributions
ipples-1.110
riage-1.099
ry-1.058
flower-1.054
ac-1.047
Token,
Token ID11
Feature activation+7.15e-01

Pos logit contributions
hy+2.572
Mic+2.445
Red+2.117
Robert+2.082
Roger+2.065
Neg logit contributions
ipples-1.745
flower-1.737
riage-1.717
ry-1.693
 department-1.656
Token "
Token ID366
Feature activation-2.27e-01

Pos logit contributions
riage+0.086
ipples+0.084
ry+0.081
flower+0.080
ac+0.078
Neg logit contributions
hy-0.151
Mic-0.138
 rides-0.125
Red-0.123
Roger-0.121
TokenIt
Token ID1026
Feature activation-1.11e+00

Pos logit contributions
riage+0.370
ipples+0.361
ry+0.353
flower+0.349
 department+0.343
Neg logit contributions
hy-0.445
Mic-0.401
 rides-0.355
Red-0.352
Roger-0.346
Token's
Token ID338
Feature activation-9.53e-01

Pos logit contributions
riage+1.711
ipples+1.662
flower+1.572
ry+1.566
ucks+1.559
Neg logit contributions
hy-2.263
Mic-2.024
 rides-1.867
Red-1.812
Roger-1.776
Token alive
Token ID6776
Feature activation-7.96e-01

Pos logit contributions
riage+1.785
ipples+1.756
ry+1.744
flower+1.713
 department+1.700
Neg logit contributions
hy-1.623
Mic-1.433
 rides-1.258
Red-1.240
Roger-1.224
Token!
Token ID0
Feature activation-1.29e+00

Pos logit contributions
riage+1.379
ipples+1.350
ry+1.305
flower+1.288
andy+1.277
Neg logit contributions
hy-1.477
Mic-1.306
 rides-1.167
Red-1.146
Roger-1.132
Token Hur
Token ID10077
Feature activation-3.45e+00

Pos logit contributions
ipples+2.444
riage+2.381
flower+2.340
ry+2.333
andy+2.308
Neg logit contributions
hy-2.264
Mic-1.972
Red-1.739
 rides-1.695
Roger-1.677
Tokenray
Token ID2433
Feature activation-9.13e-01

Pos logit contributions
ipples+4.527
ac+3.954
ull+3.869
ucks+3.603
riage+3.596
Neg logit contributions
hy-8.723
ve-7.053
Mic-7.005
 rides-6.680
Red-6.664
Token!"
Token ID2474
Feature activation+6.76e-01

Pos logit contributions
riage+1.558
ipples+1.528
flower+1.470
andy+1.443
ry+1.437
Neg logit contributions
hy-1.706
Mic-1.504
 rides-1.377
Red-1.320
Roger-1.303
Token The
Token ID383
Feature activation-1.51e+00

Pos logit contributions
hy+1.335
Mic+1.196
 rides+1.053
Red+1.049
nam+1.014
Neg logit contributions
riage-1.092
ipples-1.087
ry-1.071
flower-1.045
andy-1.038
Token grape
Token ID30777
Feature activation-4.12e-01

Pos logit contributions
riage+2.227
ry+2.132
ipples+2.049
ac+2.003
ucks+1.994
Neg logit contributions
hy-3.236
Mic-3.074
 rides-2.740
Red-2.655
Roger-2.645
Token j
Token ID474
Feature activation-1.04e-01

Pos logit contributions
riage+0.609
ipples+0.585
ry+0.568
flower+0.562
ac+0.554
Neg logit contributions
hy-0.869
Mic-0.787
 rides-0.711
Red-0.698
Roger-0.688
Token had
Token ID550
Feature activation+4.64e-01

Pos logit contributions
hy+1.523
Mic+1.454
Red+1.255
Roger+1.228
 rides+1.225
Neg logit contributions
riage-1.171
ipples-1.151
ry-1.147
flower-1.120
 department-1.100
Token worked
Token ID3111
Feature activation+9.95e-01

Pos logit contributions
hy+0.934
Mic+0.861
Red+0.753
 rides+0.743
Roger+0.730
Neg logit contributions
riage-0.720
ipples-0.705
flower-0.679
ry-0.679
 department-0.677
Token.
Token ID13
Feature activation+3.55e-01

Pos logit contributions
hy+2.114
Mic+1.978
Red+1.716
 rides+1.685
Roger+1.667
Neg logit contributions
riage-1.433
ry-1.403
ipples-1.389
ac-1.360
 department-1.340
Token After
Token ID2293
Feature activation+1.55e-01

Pos logit contributions
hy+0.668
Mic+0.592
 rides+0.526
Red+0.516
Roger+0.505
Neg logit contributions
riage-0.609
ipples-0.591
ry-0.583
flower-0.575
andy-0.573
Token lots
Token ID6041
Feature activation+2.34e+00

Pos logit contributions
hy+0.293
Mic+0.265
Red+0.231
 rides+0.230
Roger+0.225
Neg logit contributions
riage-0.264
ipples-0.257
ry-0.251
flower-0.249
 department-0.246
Token of
Token ID286
Feature activation+2.30e+00

Pos logit contributions
hy+5.901
Mic+5.898
nam+5.523
Red+5.300
Roger+5.129
Neg logit contributions
ry-2.519
andy-2.431
en-2.389
ipples-2.377
riage-2.263
Token hugs
Token ID40657
Feature activation+5.42e-01

Pos logit contributions
hy+4.829
Mic+4.664
Red+3.916
Blue+3.897
Bow+3.766
Neg logit contributions
riage-3.450
ac-3.337
ipples-3.304
 department-3.285
 mistake-3.255
Token,
Token ID11
Feature activation-1.78e-01

Pos logit contributions
hy+1.071
Mic+0.994
Red+0.869
 rides+0.858
Roger+0.848
Neg logit contributions
riage-0.853
ipples-0.825
ry-0.823
flower-0.821
 department-0.807
Token her
Token ID607
Feature activation+1.56e-01

Pos logit contributions
riage+0.383
ipples+0.376
ry+0.368
flower+0.367
ac+0.361
Neg logit contributions
hy-0.258
Mic-0.223
 rides-0.188
Red-0.185
Roger-0.180
Token family
Token ID1641
Feature activation-3.32e-01

Pos logit contributions
hy+0.405
Mic+0.375
Red+0.342
 rides+0.340
Roger+0.336
Neg logit contributions
riage-0.157
ipples-0.151
ry-0.147
flower-0.145
andy-0.141
Token was
Token ID373
Feature activation-9.74e-01

Pos logit contributions
riage+0.542
ipples+0.528
ry+0.511
flower+0.509
ac+0.501
Neg logit contributions
hy-0.652
Mic-0.584
 rides-0.523
Red-0.513
Roger-0.506
Token It
Token ID632
Feature activation+8.61e-03

Pos logit contributions
riage+0.678
ipples+0.667
ry+0.651
flower+0.648
ac+0.638
Neg logit contributions
hy-0.634
Mic-0.561
 rides-0.490
Red-0.485
Roger-0.472
Token was
Token ID373
Feature activation+2.43e-01

Pos logit contributions
hy+0.020
Mic+0.019
 rides+0.017
Red+0.017
Roger+0.017
Neg logit contributions
riage-0.010
ipples-0.010
ry-0.010
flower-0.010
 department-0.009
Token old
Token ID1468
Feature activation+8.88e-01

Pos logit contributions
hy+0.431
Mic+0.389
 rides+0.338
Red+0.335
Roger+0.330
Neg logit contributions
riage-0.436
ipples-0.427
flower-0.417
ry-0.414
ac-0.412
Token and
Token ID290
Feature activation+7.71e-01

Pos logit contributions
hy+1.864
Mic+1.732
Red+1.516
Roger+1.513
Robert+1.502
Neg logit contributions
ry-1.289
riage-1.288
flower-1.287
ipples-1.278
 department-1.257
Token had
Token ID550
Feature activation+3.32e-01

Pos logit contributions
hy+1.357
Mic+1.242
 rides+1.085
Red+1.059
Robert+1.053
Neg logit contributions
riage-1.397
flower-1.319
ipples-1.317
ry-1.316
ac-1.296
Token many
Token ID867
Feature activation+2.06e+00

Pos logit contributions
hy+0.603
Mic+0.552
Red+0.472
Roger+0.470
 rides+0.469
Neg logit contributions
riage-0.574
ipples-0.565
ac-0.544
 department-0.544
flower-0.544
Token colors
Token ID7577
Feature activation-1.40e-01

Pos logit contributions
hy+4.646
Mic+4.237
Roger+3.776
Robert+3.725
Red+3.682
Neg logit contributions
ac-2.800
riage-2.746
ry-2.703
flower-2.654
ipples-2.654
Token.
Token ID13
Feature activation+2.62e-01

Pos logit contributions
riage+0.228
ipples+0.223
ry+0.217
flower+0.216
ac+0.211
Neg logit contributions
hy-0.274
Mic-0.247
 rides-0.219
Red-0.217
Roger-0.213
Token Tim
Token ID5045
Feature activation+2.53e-01

Pos logit contributions
hy+0.511
Mic+0.459
 rides+0.405
Red+0.399
Roger+0.396
Neg logit contributions
riage-0.429
ipples-0.418
ry-0.407
flower-0.403
andy-0.398
Token did
Token ID750
Feature activation+1.70e-01

Pos logit contributions
hy+0.466
Mic+0.416
 rides+0.360
Red+0.357
Roger+0.356
Neg logit contributions
riage-0.440
ipples-0.431
flower-0.426
ry-0.425
ac-0.416
Token not
Token ID407
Feature activation+4.58e-01

Pos logit contributions
hy+0.345
Mic+0.316
Red+0.275
 rides+0.273
Roger+0.272
Neg logit contributions
riage-0.269
ipples-0.261
ry-0.254
flower-0.249
ac-0.247
Token things
Token ID1243
Feature activation+1.99e+00

Pos logit contributions
hy+6.222
Mic+5.634
Robert+5.262
Roger+5.018
Red+5.002
Neg logit contributions
ac-3.001
ipples-2.872
en-2.846
ry-2.844
ull-2.737
Token around
Token ID1088
Feature activation+6.66e-01

Pos logit contributions
hy+4.766
Mic+4.461
Robert+3.954
Red+3.893
 blonde+3.881
Neg logit contributions
ry-2.672
ipples-2.489
ac-2.436
flower-2.433
riage-2.396
Token.
Token ID13
Feature activation-3.32e-01

Pos logit contributions
hy+1.558
Mic+1.387
Red+1.257
 rides+1.245
Robert+1.227
Neg logit contributions
riage-0.851
ipples-0.834
ry-0.834
flower-0.829
ac-0.809
Token It
Token ID632
Feature activation-4.97e-01

Pos logit contributions
riage+0.560
ipples+0.549
ry+0.535
flower+0.532
ac+0.523
Neg logit contributions
hy-0.635
Mic-0.569
 rides-0.504
Red-0.500
Roger-0.489
Token would
Token ID561
Feature activation+7.45e-01

Pos logit contributions
riage+0.755
ipples+0.724
ry+0.702
flower+0.696
andy+0.685
Neg logit contributions
hy-1.044
Mic-0.946
 rides-0.855
Red-0.840
Roger-0.822
Token carry
Token ID3283
Feature activation+2.00e+00

Pos logit contributions
hy+1.612
Mic+1.493
Red+1.329
Robert+1.318
 rides+1.318
Neg logit contributions
riage-1.021
ipples-1.018
flower-0.986
ry-0.950
ac-0.949
Token sticks
Token ID16461
Feature activation+1.06e+00

Pos logit contributions
hy+4.780
Mic+4.359
Robert+3.954
Roger+3.876
Red+3.824
Neg logit contributions
ac-2.555
ipples-2.433
ry-2.413
riage-2.373
ull-2.354
Token and
Token ID290
Feature activation+1.14e+00

Pos logit contributions
hy+2.248
Mic+2.072
Red+1.836
Robert+1.807
Roger+1.804
Neg logit contributions
ry-1.560
ipples-1.552
ac-1.516
flower-1.509
riage-1.487
Token leaves
Token ID5667
Feature activation+4.76e-01

Pos logit contributions
hy+2.493
Mic+2.231
Robert+2.048
Red+2.027
Roger+2.023
Neg logit contributions
ipples-1.578
riage-1.570
flower-1.555
ry-1.521
ac-1.494
Token to
Token ID284
Feature activation+1.52e-01

Pos logit contributions
hy+1.002
Mic+0.921
Red+0.814
 rides+0.811
Roger+0.804
Neg logit contributions
ipples-0.695
riage-0.685
ry-0.683
flower-0.673
ac-0.656
Token make
Token ID787
Feature activation+5.79e-01

Pos logit contributions
hy+0.339
Mic+0.308
 rides+0.275
Red+0.275
Roger+0.272
Neg logit contributions
riage-0.207
ipples-0.203
ry-0.196
flower-0.195
ac-0.190
Token.
Token ID13
Feature activation-5.37e-01

Pos logit contributions
riage+1.233
ipples+1.225
flower+1.163
ry+1.147
andy+1.143
Neg logit contributions
hy-1.530
Mic-1.357
 rides-1.245
Red-1.203
Roger-1.186
Token Let
Token ID3914
Feature activation-2.82e-01

Pos logit contributions
ipples+0.914
riage+0.913
ry+0.879
flower+0.875
andy+0.866
Neg logit contributions
hy-1.005
Mic-0.892
 rides-0.786
Red-0.784
Roger-0.763
Token's
Token ID338
Feature activation-7.74e-01

Pos logit contributions
riage+0.427
ipples+0.419
ry+0.405
flower+0.402
ac+0.393
Neg logit contributions
hy-0.588
Mic-0.535
 rides-0.481
Red-0.474
Roger-0.467
Token go
Token ID467
Feature activation-9.93e-01

Pos logit contributions
riage+1.025
ipples+1.003
ry+0.978
flower+0.962
 department+0.944
Neg logit contributions
hy-1.759
Mic-1.596
 rides-1.492
Red-1.445
Roger-1.424
Token home
Token ID1363
Feature activation-1.01e+00

Pos logit contributions
riage+1.510
ipples+1.475
flower+1.439
andy+1.396
ac+1.393
Neg logit contributions
hy-2.069
Mic-1.813
 rides-1.745
Red-1.656
Roger-1.618
Token."
Token ID526
Feature activation+1.90e+00

Pos logit contributions
riage+1.779
ipples+1.745
flower+1.718
andy+1.682
ric+1.651
Neg logit contributions
hy-1.816
Mic-1.570
 rides-1.474
Red-1.408
Roger-1.363
Token They
Token ID1119
Feature activation+1.93e-01

Pos logit contributions
hy+4.399
Mic+3.808
nam+3.415
Red+3.397
 rides+3.369
Neg logit contributions
riage-2.634
 department-2.531
ucks-2.441
ry-2.440
 again-2.394
Token thanked
Token ID26280
Feature activation+1.42e+00

Pos logit contributions
hy+0.382
Mic+0.345
Red+0.304
 rides+0.303
Roger+0.299
Neg logit contributions
riage-0.308
ipples-0.301
ry-0.294
flower-0.292
ac-0.288
Token the
Token ID262
Feature activation+6.24e-01

Pos logit contributions
hy+3.400
Mic+2.972
Red+2.698
 rides+2.663
 blonde+2.651
Neg logit contributions
riage-1.840
 department-1.665
flower-1.662
andy-1.647
ry-1.640
Token nice
Token ID3621
Feature activation-8.74e-04

Pos logit contributions
hy+1.340
Mic+1.232
Red+1.079
 rides+1.076
Roger+1.075
Neg logit contributions
riage-0.899
ipples-0.867
 department-0.853
flower-0.845
ac-0.817
Token man
Token ID582
Feature activation-1.72e-01

Pos logit contributions
riage+0.001
ipples+0.001
ry+0.001
flower+0.001
 department+0.001
Neg logit contributions
hy-0.002
Mic-0.002
 rides-0.001
Red-0.001
Roger-0.001
Token the
Token ID262
Feature activation+1.00e+00

Pos logit contributions
hy+2.238
Mic+2.062
Red+1.834
Roger+1.778
 rides+1.758
Neg logit contributions
riage-1.658
ipples-1.553
flower-1.531
ry-1.509
ac-1.507
Token queen
Token ID16599
Feature activation+4.45e-01

Pos logit contributions
hy+2.076
Mic+1.879
Red+1.699
 rides+1.658
Roger+1.655
Neg logit contributions
riage-1.535
ipples-1.514
flower-1.486
ac-1.466
 department-1.452
Token had
Token ID550
Feature activation+7.76e-01

Pos logit contributions
hy+0.992
Mic+0.923
Red+0.822
 rides+0.811
Roger+0.811
Neg logit contributions
riage-0.596
ry-0.576
ipples-0.576
flower-0.572
ac-0.548
Token taken
Token ID2077
Feature activation+1.31e+00

Pos logit contributions
hy+1.553
Mic+1.461
Red+1.284
Roger+1.249
 rides+1.238
Neg logit contributions
riage-1.170
ipples-1.166
ac-1.115
flower-1.113
 department-1.108
Token away
Token ID1497
Feature activation+1.09e-01

Pos logit contributions
hy+2.969
Mic+2.760
Red+2.480
Roger+2.414
Robert+2.396
Neg logit contributions
riage-1.663
ac-1.660
ipples-1.635
ry-1.631
 department-1.573
Token the
Token ID262
Feature activation+2.00e+00

Pos logit contributions
hy+0.229
Mic+0.210
Red+0.186
 rides+0.186
Roger+0.182
Neg logit contributions
riage-0.160
ipples-0.156
ry-0.152
flower-0.151
ac-0.148
Token peace
Token ID4167
Feature activation+2.91e-01

Pos logit contributions
hy+4.461
Mic+4.166
Red+3.840
Roger+3.733
 rides+3.646
Neg logit contributions
riage-2.803
flower-2.701
ac-2.683
ipples-2.655
 department-2.605
Token of
Token ID286
Feature activation+8.01e-01

Pos logit contributions
hy+0.603
Mic+0.553
Red+0.492
 rides+0.488
Roger+0.480
Neg logit contributions
riage-0.436
ipples-0.429
flower-0.421
ry-0.420
ac-0.407
Token his
Token ID465
Feature activation+1.89e+00

Pos logit contributions
hy+1.796
Mic+1.650
Red+1.502
 rides+1.446
Roger+1.445
Neg logit contributions
riage-1.086
ipples-1.043
flower-1.023
ac-1.007
ry-0.997
Token people
Token ID661
Feature activation+1.33e+00

Pos logit contributions
hy+4.278
Mic+4.111
Red+3.764
Roger+3.651
Robert+3.546
Neg logit contributions
riage-2.591
ry-2.537
ipples-2.522
flower-2.492
ac-2.486
Token.
Token ID13
Feature activation+9.68e-01

Pos logit contributions
hy+2.883
Mic+2.762
Red+2.453
Roger+2.374
 rides+2.333
Neg logit contributions
ry-1.852
riage-1.847
flower-1.801
ipples-1.782
ac-1.709
Token are
Token ID389
Feature activation+6.64e-02

Pos logit contributions
riage+0.224
ipples+0.219
ry+0.213
flower+0.212
ac+0.207
Neg logit contributions
hy-0.284
Mic-0.257
 rides-0.229
Red-0.226
Roger-0.222
Token welcome
Token ID7062
Feature activation+4.05e-01

Pos logit contributions
hy+0.137
Mic+0.124
 rides+0.111
Red+0.109
Roger+0.108
Neg logit contributions
riage-0.102
ipples-0.099
flower-0.096
ry-0.095
ac-0.094
Token,
Token ID11
Feature activation-8.22e-01

Pos logit contributions
hy+0.774
Mic+0.699
 rides+0.617
Red+0.616
Roger+0.598
Neg logit contributions
riage-0.678
ipples-0.647
ry-0.644
flower-0.643
 department-0.636
Token Lily
Token ID20037
Feature activation-8.76e-02

Pos logit contributions
riage+1.291
ipples+1.278
ry+1.251
flower+1.220
ric+1.187
Neg logit contributions
hy-1.669
Mic-1.525
Red-1.362
 rides-1.353
Roger-1.315
Token,"
Token ID553
Feature activation+1.20e+00

Pos logit contributions
riage+0.148
ipples+0.145
ry+0.141
flower+0.140
ac+0.138
Neg logit contributions
hy-0.166
Mic-0.149
 rides-0.132
Red-0.130
Roger-0.128
Token Max
Token ID5436
Feature activation+1.88e-01

Pos logit contributions
hy+2.159
Mic+1.863
Roger+1.607
Robert+1.606
 blonde+1.581
Neg logit contributions
riage-2.188
andy-2.117
flower-2.100
ry-2.090
 department-2.079
Token says
Token ID1139
Feature activation+3.16e-01

Pos logit contributions
hy+0.360
Mic+0.325
Red+0.283
 rides+0.280
Roger+0.279
Neg logit contributions
riage-0.312
ipples-0.305
ry-0.297
flower-0.296
andy-0.292
Token.
Token ID13
Feature activation-7.53e-01

Pos logit contributions
hy+0.662
Mic+0.609
 rides+0.540
Red+0.538
Roger+0.533
Neg logit contributions
riage-0.465
ipples-0.456
ry-0.452
flower-0.439
 department-0.434
Token "
Token ID366
Feature activation+7.79e-01

Pos logit contributions
riage+1.154
ipples+1.118
flower+1.082
ry+1.079
ac+1.057
Neg logit contributions
hy-1.578
Mic-1.447
Red-1.287
 rides-1.284
Roger-1.256
TokenYou
Token ID1639
Feature activation+4.57e-01

Pos logit contributions
hy+1.547
Mic+1.343
 rides+1.216
Roger+1.160
 blonde+1.156
Neg logit contributions
ipples-1.271
riage-1.251
flower-1.232
ac-1.197
ry-1.177
Token are
Token ID389
Feature activation+3.22e-01

Pos logit contributions
hy+0.962
Mic+0.877
 rides+0.775
Red+0.773
Roger+0.769
Neg logit contributions
riage-0.679
ipples-0.667
ac-0.640
flower-0.637
ry-0.634
Token flew
Token ID13112
Feature activation-3.04e-02

Pos logit contributions
riage+0.533
ipples+0.520
ry+0.501
flower+0.500
ac+0.490
Neg logit contributions
hy-0.742
Mic-0.669
 rides-0.603
Red-0.593
Roger-0.584
Token away
Token ID1497
Feature activation-1.62e-01

Pos logit contributions
riage+0.037
ipples+0.036
ry+0.035
flower+0.035
ac+0.034
Neg logit contributions
hy-0.072
Mic-0.066
 rides-0.060
Red-0.059
Roger-0.059
Token,
Token ID11
Feature activation+4.96e-01

Pos logit contributions
riage+0.264
ipples+0.258
ry+0.251
flower+0.250
ac+0.245
Neg logit contributions
hy-0.317
Mic-0.285
 rides-0.253
Red-0.250
Roger-0.246
Token and
Token ID290
Feature activation-5.53e-01

Pos logit contributions
hy+0.949
Mic+0.868
Red+0.759
 rides+0.749
Roger+0.732
Neg logit contributions
riage-0.824
ipples-0.814
ry-0.793
flower-0.792
ac-0.772
Token the
Token ID262
Feature activation-6.54e-01

Pos logit contributions
riage+0.805
ipples+0.792
flower+0.761
ry+0.760
andy+0.744
Neg logit contributions
hy-1.181
Mic-1.062
 rides-0.966
Red-0.943
Roger-0.936
Token village
Token ID7404
Feature activation+5.94e-02

Pos logit contributions
riage+1.082
ipples+1.059
ry+1.048
 department+1.026
flower+1.018
Neg logit contributions
hy-1.278
Mic-1.131
 rides-1.034
Red-0.999
Roger-0.985
Token lived
Token ID5615
Feature activation-5.47e-01

Pos logit contributions
hy+0.123
Mic+0.113
 rides+0.100
Red+0.099
Roger+0.097
Neg logit contributions
riage-0.089
ipples-0.087
ry-0.085
flower-0.085
ac-0.083
Token happily
Token ID18177
Feature activation+9.26e-01

Pos logit contributions
riage+1.063
ipples+1.033
ry+1.015
flower+1.006
andy+0.996
Neg logit contributions
hy-0.906
Mic-0.792
 rides-0.693
Red-0.680
Roger-0.671
Token ever
Token ID1683
Feature activation+9.52e-01

Pos logit contributions
hy+1.592
Mic+1.524
Red+1.323
Roger+1.260
 rides+1.258
Neg logit contributions
riage-1.692
ipples-1.674
ry-1.644
flower-1.633
ac-1.620
Token after
Token ID706
Feature activation+6.28e-01

Pos logit contributions
hy+1.993
Mic+1.841
Red+1.676
 rides+1.640
Roger+1.602
Neg logit contributions
riage-1.451
ipples-1.413
ry-1.361
flower-1.349
ac-1.346
Token.
Token ID13
Feature activation-4.99e-02

Pos logit contributions
hy+1.353
Mic+1.263
Red+1.113
 rides+1.076
Roger+1.071
Neg logit contributions
riage-0.908
ipples-0.889
ry-0.879
flower-0.845
andy-0.839
Token.
Token ID13
Feature activation+5.62e-01

Pos logit contributions
hy+0.121
Mic+0.110
 rides+0.096
Red+0.095
Roger+0.093
Neg logit contributions
riage-0.115
ipples-0.112
ry-0.110
flower-0.110
andy-0.107
Token He
Token ID679
Feature activation+4.44e-01

Pos logit contributions
hy+1.103
Mic+1.001
 rides+0.899
Red+0.891
Roger+0.866
Neg logit contributions
riage-0.887
ipples-0.843
flower-0.833
 department-0.832
ry-0.832
Token bark
Token ID21405
Feature activation+4.63e-01

Pos logit contributions
hy+0.975
Mic+0.891
 rides+0.803
Red+0.797
Roger+0.779
Neg logit contributions
riage-0.601
ipples-0.587
ry-0.581
flower-0.579
andy-0.570
Tokened
Token ID276
Feature activation+3.67e-01

Pos logit contributions
hy+0.893
Mic+0.824
 rides+0.721
Red+0.719
Roger+0.705
Neg logit contributions
ipples-0.743
riage-0.738
ry-0.728
flower-0.723
 department-0.710
Token at
Token ID379
Feature activation+1.03e+00

Pos logit contributions
hy+0.712
Mic+0.651
Red+0.571
 rides+0.569
Roger+0.555
Neg logit contributions
riage-0.592
ipples-0.583
ry-0.578
flower-0.573
 department-0.565
Token them
Token ID606
Feature activation+5.09e-01

Pos logit contributions
hy+2.276
Mic+2.026
Red+1.853
 rides+1.810
nam+1.790
Neg logit contributions
riage-1.407
flower-1.321
ipples-1.309
andy-1.303
ry-1.295
Token and
Token ID290
Feature activation+5.14e-01

Pos logit contributions
hy+1.011
Mic+0.939
Red+0.821
 rides+0.808
Roger+0.803
Neg logit contributions
riage-0.793
ipples-0.788
ry-0.788
flower-0.774
andy-0.754
Token waved
Token ID26834
Feature activation+1.84e-01

Pos logit contributions
hy+0.923
Mic+0.836
Red+0.725
 rides+0.716
Roger+0.700
Neg logit contributions
riage-0.912
ipples-0.891
ry-0.884
flower-0.879
 department-0.861
Token his
Token ID465
Feature activation+1.46e+00

Pos logit contributions
hy+0.358
Mic+0.328
 rides+0.288
Red+0.287
Roger+0.280
Neg logit contributions
riage-0.298
ipples-0.292
ry-0.288
flower-0.283
ac-0.280
Token tail
Token ID7894
Feature activation+7.73e-02

Pos logit contributions
hy+3.229
Mic+3.057
Red+2.826
 rides+2.778
Robert+2.727
Neg logit contributions
ac-1.924
flower-1.907
ipples-1.899
ry-1.889
 department-1.854
Token.
Token ID13
Feature activation+6.42e-01

Pos logit contributions
hy+0.154
Mic+0.140
 rides+0.124
Red+0.124
Roger+0.121
Neg logit contributions
riage-0.121
ipples-0.120
ry-0.117
flower-0.116
ac-0.114
Token you
Token ID345
Feature activation-1.91e-01

Pos logit contributions
hy+1.366
Mic+1.198
Red+0.993
Roger+0.979
 rides+0.965
Neg logit contributions
riage-1.883
andy-1.800
ipples-1.795
ac-1.772
ry-1.763
Token like
Token ID588
Feature activation+2.68e-01

Pos logit contributions
riage+0.329
ipples+0.322
ry+0.314
flower+0.312
ac+0.306
Neg logit contributions
hy-0.360
Mic-0.322
 rides-0.284
Red-0.281
Roger-0.275
Token him
Token ID683
Feature activation-3.94e-01

Pos logit contributions
hy+0.602
Mic+0.551
Red+0.491
 rides+0.485
Roger+0.483
Neg logit contributions
riage-0.360
ipples-0.349
ry-0.339
flower-0.339
andy-0.331
Token?"
Token ID1701
Feature activation+7.62e-01

Pos logit contributions
riage+0.473
ipples+0.462
ry+0.441
flower+0.435
 department+0.427
Neg logit contributions
hy-0.941
Mic-0.862
 rides-0.790
Red-0.779
Roger-0.768
Token
Token ID198
Feature activation+3.73e-01

Pos logit contributions
hy+1.568
Mic+1.407
 rides+1.258
Red+1.249
Roger+1.216
Neg logit contributions
riage-1.164
ipples-1.111
ry-1.110
andy-1.100
flower-1.096
Token
Token ID198
Feature activation+3.53e-01

Pos logit contributions
hy+0.726
Mic+0.642
 rides+0.575
Red+0.561
nam+0.553
Neg logit contributions
riage-0.605
ipples-0.592
ry-0.583
andy-0.576
flower-0.571
TokenAnna
Token ID31160
Feature activation-2.24e-01

Pos logit contributions
hy+0.717
Mic+0.635
 rides+0.580
Red+0.558
Roger+0.550
Neg logit contributions
riage-0.547
ipples-0.534
ry-0.522
flower-0.519
andy-0.513
Token and
Token ID290
Feature activation+5.05e-01

Pos logit contributions
riage+0.358
ipples+0.350
ry+0.339
flower+0.338
ac+0.331
Neg logit contributions
hy-0.445
Mic-0.400
 rides-0.357
Red-0.353
Roger-0.346
Token Ben
Token ID3932
Feature activation+4.47e-01

Pos logit contributions
hy+1.128
Mic+1.030
Red+0.933
 rides+0.922
Roger+0.914
Neg logit contributions
riage-0.688
flower-0.645
ipples-0.644
ry-0.633
andy-0.628
Token hug
Token ID16225
Feature activation-2.42e-01

Pos logit contributions
hy+0.957
Mic+0.886
Red+0.782
 rides+0.775
Roger+0.772
Neg logit contributions
riage-0.642
ipples-0.623
flower-0.607
ry-0.606
andy-0.592
Token Spot
Token ID15899
Feature activation-2.21e-01

Pos logit contributions
riage+0.344
ipples+0.336
ry+0.325
flower+0.323
ac+0.316
Neg logit contributions
hy-0.528
Mic-0.480
 rides-0.433
Red-0.428
Roger-0.421
Token the
Token ID262
Feature activation-3.26e-01

Pos logit contributions
riage+0.031
ipples+0.030
ry+0.030
flower+0.029
ac+0.029
Neg logit contributions
hy-0.053
Mic-0.049
 rides-0.044
Red-0.043
Roger-0.043
Token lab
Token ID2248
Feature activation+5.12e-01

Pos logit contributions
riage+0.457
ipples+0.445
ry+0.431
flower+0.430
ac+0.419
Neg logit contributions
hy-0.717
Mic-0.654
 rides-0.595
Red-0.586
Roger-0.576
Token.
Token ID13
Feature activation+6.47e-02

Pos logit contributions
hy+1.072
Mic+1.002
Red+0.874
Roger+0.860
 rides+0.856
Neg logit contributions
riage-0.761
ipples-0.736
ry-0.734
flower-0.711
 department-0.705
Token
Token ID198
Feature activation+4.13e-01

Pos logit contributions
hy+0.119
Mic+0.105
 rides+0.092
Red+0.091
Roger+0.089
Neg logit contributions
riage-0.114
ipples-0.112
ry-0.110
flower-0.109
andy-0.108
Token
Token ID198
Feature activation-2.42e-01

Pos logit contributions
hy+0.821
Mic+0.720
 rides+0.647
Red+0.634
nam+0.628
Neg logit contributions
riage-0.662
ipples-0.656
ry-0.648
andy-0.639
ac-0.625
TokenAfter
Token ID3260
Feature activation+2.04e-02

Pos logit contributions
riage+0.441
ipples+0.432
ry+0.421
flower+0.420
 department+0.411
Neg logit contributions
hy-0.432
Mic-0.386
 rides-0.336
Red-0.334
Roger-0.327
Token sailing
Token ID28499
Feature activation-6.72e-01

Pos logit contributions
hy+0.039
Mic+0.035
Red+0.031
 rides+0.031
Roger+0.030
Neg logit contributions
riage-0.034
ipples-0.034
ry-0.033
flower-0.032
 department-0.032
Token back
Token ID736
Feature activation-2.63e-02

Pos logit contributions
riage+1.195
ipples+1.182
flower+1.144
ry+1.132
andy+1.117
Neg logit contributions
hy-1.222
Mic-1.061
 rides-0.956
Red-0.923
Roger-0.919
Token to
Token ID284
Feature activation-1.29e+00

Pos logit contributions
riage+0.035
ipples+0.034
ry+0.033
flower+0.033
andy+0.032
Neg logit contributions
hy-0.059
Mic-0.054
Red-0.049
 rides-0.049
Roger-0.048
Token shore
Token ID15191
Feature activation-8.48e-01

Pos logit contributions
ipples+1.929
riage+1.891
flower+1.871
ac+1.828
 department+1.814
Neg logit contributions
hy-2.752
Mic-2.399
 rides-2.348
 Bow-2.201
 Robert-2.187
Token,
Token ID11
Feature activation+2.29e-01

Pos logit contributions
riage+1.610
ipples+1.579
flower+1.537
ac+1.514
ry+1.500
Neg logit contributions
hy-1.429
Mic-1.219
 rides-1.115
Roger-1.052
Red-1.046
Token and
Token ID290
Feature activation+3.91e-01

Pos logit contributions
hy+1.061
Mic+0.968
Red+0.852
 rides+0.849
Roger+0.826
Neg logit contributions
riage-0.863
ipples-0.855
flower-0.846
ry-0.835
andy-0.811
Token suddenly
Token ID6451
Feature activation+5.83e-01

Pos logit contributions
hy+0.717
Mic+0.646
Red+0.563
 rides+0.556
Roger+0.544
Neg logit contributions
riage-0.686
ipples-0.672
ry-0.663
flower-0.661
 department-0.642
Token his
Token ID465
Feature activation+3.70e-01

Pos logit contributions
hy+1.115
Mic+1.024
Red+0.895
 rides+0.866
Roger+0.864
Neg logit contributions
riage-0.964
ipples-0.948
ry-0.938
flower-0.923
 department-0.904
Token own
Token ID898
Feature activation+1.25e-01

Pos logit contributions
hy+0.783
Mic+0.726
Red+0.646
 rides+0.639
Roger+0.633
Neg logit contributions
riage-0.535
ipples-0.526
flower-0.512
ry-0.508
andy-0.500
Token misery
Token ID24672
Feature activation-1.12e-02

Pos logit contributions
hy+0.246
Mic+0.224
Red+0.197
 rides+0.197
Roger+0.193
Neg logit contributions
riage-0.201
ipples-0.198
flower-0.193
ry-0.192
andy-0.188
Token seemed
Token ID3947
Feature activation-5.22e-01

Pos logit contributions
riage+0.016
ipples+0.016
ry+0.015
flower+0.015
 department+0.015
Neg logit contributions
hy-0.024
Mic-0.022
 rides-0.019
Red-0.019
Roger-0.019
Token to
Token ID284
Feature activation+2.53e-01

Pos logit contributions
riage+0.817
ipples+0.793
ry+0.772
flower+0.765
andy+0.752
Neg logit contributions
hy-1.066
Mic-0.946
 rides-0.854
Red-0.840
Roger-0.830
Token have
Token ID423
Feature activation-8.85e-03

Pos logit contributions
hy+0.570
Mic+0.528
 rides+0.475
Red+0.470
Roger+0.464
Neg logit contributions
riage-0.334
ipples-0.327
flower-0.316
ry-0.312
ac-0.305
Token gone
Token ID3750
Feature activation-7.88e-02

Pos logit contributions
riage+0.014
ipples+0.013
ry+0.013
flower+0.013
ac+0.013
Neg logit contributions
hy-0.018
Mic-0.016
 rides-0.014
Red-0.014
Roger-0.014
Token.
Token ID13
Feature activation+7.90e-01

Pos logit contributions
riage+0.118
ipples+0.115
ry+0.112
flower+0.111
ac+0.109
Neg logit contributions
hy-0.165
Mic-0.150
 rides-0.133
Red-0.132
Roger-0.130
Token They
Token ID1119
Feature activation+4.76e-03

Pos logit contributions
hy+1.564
Mic+1.393
Red+1.239
 rides+1.234
nam+1.206
Neg logit contributions
riage-1.287
ipples-1.256
ry-1.217
andy-1.206
flower-1.195
Token.
Token ID13
Feature activation+6.13e-01

Pos logit contributions
hy+1.955
Mic+1.846
Red+1.629
nam+1.593
Roger+1.587
Neg logit contributions
riage-1.254
ipples-1.234
ry-1.219
flower-1.192
 department-1.189
Token He
Token ID679
Feature activation+3.13e-01

Pos logit contributions
hy+1.055
Mic+0.930
 rides+0.799
Red+0.787
nam+0.771
Neg logit contributions
riage-1.144
ipples-1.126
ry-1.105
 department-1.087
ac-1.071
Token runs
Token ID4539
Feature activation+2.72e-01

Pos logit contributions
hy+0.664
Mic+0.607
Red+0.538
Roger+0.530
 rides+0.526
Neg logit contributions
riage-0.456
ipples-0.448
ry-0.437
flower-0.429
 department-0.427
Token towards
Token ID3371
Feature activation-3.69e-01

Pos logit contributions
hy+0.621
Mic+0.572
Red+0.511
 rides+0.509
Roger+0.503
Neg logit contributions
riage-0.348
ipples-0.340
ry-0.333
 department-0.328
flower-0.326
Token the
Token ID262
Feature activation+6.10e-01

Pos logit contributions
riage+0.457
ipples+0.445
ry+0.427
flower+0.426
ac+0.413
Neg logit contributions
hy-0.870
Mic-0.799
 rides-0.731
Red-0.719
Roger-0.712
Token barn
Token ID25203
Feature activation-4.46e-01

Pos logit contributions
hy+1.190
Mic+1.076
Red+0.940
 rides+0.925
nam+0.924
Neg logit contributions
riage-1.002
ipples-0.983
flower-0.964
 department-0.956
ry-0.936
Token.
Token ID13
Feature activation+5.54e-01

Pos logit contributions
riage+0.750
ipples+0.728
flower+0.705
ry+0.705
ac+0.697
Neg logit contributions
hy-0.854
Mic-0.754
 rides-0.682
Red-0.659
Roger-0.649
Token
Token ID198
Feature activation+4.61e-01

Pos logit contributions
hy+0.999
Mic+0.892
 rides+0.768
Red+0.760
nam+0.745
Neg logit contributions
riage-0.988
ipples-0.971
ry-0.951
 department-0.935
ucks-0.918
Token
Token ID198
Feature activation+6.06e-01

Pos logit contributions
hy+0.924
Mic+0.821
 rides+0.727
Red+0.714
nam+0.711
Neg logit contributions
riage-0.727
ipples-0.724
ry-0.713
andy-0.691
ac-0.683
TokenAnna
Token ID31160
Feature activation+2.58e-01

Pos logit contributions
hy+1.336
Mic+1.188
 rides+1.083
Red+1.042
nam+1.035
Neg logit contributions
riage-0.852
ipples-0.846
ry-0.837
ac-0.808
 department-0.791
Token and
Token ID290
Feature activation+5.94e-01

Pos logit contributions
hy+0.517
Mic+0.470
Red+0.415
Roger+0.407
 rides+0.406
Neg logit contributions
riage-0.402
ipples-0.399
ry-0.395
flower-0.387
 department-0.382
Token him
Token ID683
Feature activation+1.87e-01

Pos logit contributions
riage+0.025
ipples+0.024
ry+0.023
flower+0.023
andy+0.022
Neg logit contributions
hy-0.050
Mic-0.046
 rides-0.041
Red-0.041
Roger-0.040
Token and
Token ID290
Feature activation+6.45e-01

Pos logit contributions
hy+0.366
Mic+0.335
Red+0.292
 rides+0.291
Roger+0.287
Neg logit contributions
riage-0.305
ipples-0.296
ry-0.294
flower-0.292
 department-0.285
Token said
Token ID531
Feature activation+3.38e-01

Pos logit contributions
hy+1.180
Mic+1.077
Red+0.935
 rides+0.907
Roger+0.902
Neg logit contributions
riage-1.132
ipples-1.097
ry-1.081
flower-1.074
 department-1.065
Token,
Token ID11
Feature activation-1.40e-01

Pos logit contributions
hy+0.633
Mic+0.581
Red+0.504
 rides+0.501
Roger+0.490
Neg logit contributions
riage-0.576
ry-0.555
ipples-0.554
flower-0.538
ac-0.535
Token "
Token ID366
Feature activation+8.18e-01

Pos logit contributions
riage+0.184
ipples+0.179
ry+0.173
flower+0.172
ac+0.167
Neg logit contributions
hy-0.317
Mic-0.290
 rides-0.262
Red-0.260
Roger-0.256
TokenGood
Token ID10248
Feature activation-1.56e-01

Pos logit contributions
hy+1.625
Mic+1.451
 rides+1.303
nam+1.261
Red+1.239
Neg logit contributions
riage-1.338
ipples-1.291
flower-1.288
ry-1.231
ac-1.220
Token night
Token ID1755
Feature activation+2.71e-01

Pos logit contributions
riage+0.204
ipples+0.198
ry+0.192
flower+0.190
ac+0.186
Neg logit contributions
hy-0.360
Mic-0.329
 rides-0.298
Red-0.295
Roger-0.291
Token,
Token ID11
Feature activation+3.11e-01

Pos logit contributions
hy+0.510
Mic+0.466
Red+0.406
 rides+0.400
Roger+0.394
Neg logit contributions
riage-0.464
ipples-0.452
ry-0.447
flower-0.440
 department-0.433
Token snow
Token ID6729
Feature activation-7.13e-01

Pos logit contributions
hy+0.613
Mic+0.556
Red+0.484
 rides+0.483
Roger+0.473
Neg logit contributions
riage-0.513
ipples-0.494
ry-0.484
flower-0.479
ac-0.472
Tokenman
Token ID805
Feature activation-5.51e-01

Pos logit contributions
riage+0.770
ipples+0.739
ry+0.699
ac+0.690
andy+0.676
Neg logit contributions
hy-1.793
Mic-1.638
 rides-1.514
Red-1.488
Roger-1.475
Token.
Token ID13
Feature activation+6.82e-02

Pos logit contributions
riage+0.937
ipples+0.923
ry+0.882
flower+0.880
andy+0.870
Neg logit contributions
hy-1.043
Mic-0.921
 rides-0.822
Red-0.805
Roger-0.792
Token eating
Token ID6600
Feature activation-2.98e-01

Pos logit contributions
riage+1.337
ipples+1.304
ry+1.258
 department+1.242
flower+1.240
Neg logit contributions
hy-1.317
Mic-1.180
 rides-1.036
Red-1.008
Roger-1.001
Token it
Token ID340
Feature activation+4.83e-01

Pos logit contributions
riage+0.455
ipples+0.443
ry+0.430
flower+0.428
 department+0.419
Neg logit contributions
hy-0.614
Mic-0.555
 rides-0.498
Red-0.492
Roger-0.484
Token herself
Token ID5223
Feature activation+7.51e-01

Pos logit contributions
hy+1.067
Mic+0.984
Red+0.871
 rides+0.867
Roger+0.860
Neg logit contributions
riage-0.664
ipples-0.652
ry-0.640
flower-0.634
ac-0.614
Token.
Token ID13
Feature activation+6.32e-01

Pos logit contributions
hy+1.608
Mic+1.509
Red+1.328
Roger+1.307
 rides+1.293
Neg logit contributions
riage-1.078
ipples-1.060
ry-1.055
flower-1.032
ac-1.008
Token Mrs
Token ID9074
Feature activation+3.74e-01

Pos logit contributions
hy+1.277
Mic+1.167
 rides+1.032
Red+1.018
Roger+1.001
Neg logit contributions
riage-1.001
ipples-0.944
ry-0.934
 department-0.926
flower-0.920
Token.
Token ID13
Feature activation-3.49e-01

Pos logit contributions
hy+0.774
Mic+0.709
 rides+0.632
Red+0.626
Roger+0.613
Neg logit contributions
riage-0.573
ipples-0.547
flower-0.544
ry-0.541
 department-0.528
Token Smith
Token ID4176
Feature activation-7.07e-01

Pos logit contributions
riage+0.673
ipples+0.665
ry+0.648
flower+0.642
ac+0.636
Neg logit contributions
hy-0.578
Mic-0.511
 rides-0.438
Red-0.435
Roger-0.427
Token was
Token ID373
Feature activation-4.96e-01

Pos logit contributions
riage+1.155
ipples+1.113
ry+1.059
flower+1.051
ac+1.043
Neg logit contributions
hy-1.402
Mic-1.263
 rides-1.121
Red-1.104
Roger-1.090
Token so
Token ID523
Feature activation+5.06e-01

Pos logit contributions
riage+0.970
ipples+0.948
ry+0.934
flower+0.922
andy+0.908
Neg logit contributions
hy-0.824
Mic-0.714
 rides-0.623
Red-0.611
Roger-0.603
Token surprised
Token ID6655
Feature activation+5.89e-01

Pos logit contributions
hy+0.825
Mic+0.750
 rides+0.649
Red+0.631
Roger+0.614
Neg logit contributions
riage-0.975
ipples-0.969
ac-0.937
ry-0.936
flower-0.936
Token and
Token ID290
Feature activation+4.86e-01

Pos logit contributions
hy+1.216
Mic+1.122
Red+0.990
 rides+0.977
nam+0.964
Neg logit contributions
riage-0.893
ry-0.872
ipples-0.869
flower-0.857
andy-0.854
Token it
Token ID340
Feature activation-2.19e-02

Pos logit contributions
riage+0.395
ipples+0.385
ry+0.371
flower+0.370
ac+0.361
Neg logit contributions
hy-0.633
Mic-0.576
 rides-0.522
Red-0.515
Roger-0.508
Token either
Token ID2035
Feature activation-1.19e+00

Pos logit contributions
riage+0.034
ipples+0.033
ry+0.032
flower+0.032
ac+0.031
Neg logit contributions
hy-0.045
Mic-0.040
 rides-0.036
Red-0.036
Roger-0.035
Token.
Token ID13
Feature activation-1.26e-01

Pos logit contributions
riage+2.034
ipples+2.005
 department+1.908
flower+1.893
ry+1.889
Neg logit contributions
hy-2.244
Mic-1.923
 rides-1.755
Red-1.696
 Robert-1.672
Token Then
Token ID3244
Feature activation-3.52e-01

Pos logit contributions
riage+0.208
ipples+0.203
ry+0.198
flower+0.196
ac+0.193
Neg logit contributions
hy-0.245
Mic-0.220
 rides-0.195
Red-0.193
Roger-0.189
Token Max
Token ID5436
Feature activation-3.43e-01

Pos logit contributions
riage+0.607
ipples+0.598
ry+0.579
flower+0.577
ac+0.567
Neg logit contributions
hy-0.653
Mic-0.582
 rides-0.517
Red-0.507
Roger-0.500
Token had
Token ID550
Feature activation-7.07e-01

Pos logit contributions
riage+0.534
ipples+0.520
ry+0.502
flower+0.500
ac+0.491
Neg logit contributions
hy-0.700
Mic-0.633
 rides-0.566
Red-0.558
Roger-0.550
Token a
Token ID257
Feature activation-4.88e-02

Pos logit contributions
riage+1.140
ipples+1.106
flower+1.084
ry+1.084
andy+1.044
Neg logit contributions
hy-1.424
Mic-1.277
 rides-1.158
Roger-1.122
Red-1.121
Token good
Token ID922
Feature activation-1.49e-02

Pos logit contributions
riage+0.072
ipples+0.070
ry+0.068
flower+0.068
 department+0.066
Neg logit contributions
hy-0.103
Mic-0.093
 rides-0.084
Red-0.083
Roger-0.081
Token idea
Token ID2126
Feature activation+4.94e-03

Pos logit contributions
riage+0.020
ipples+0.020
ry+0.019
flower+0.019
ac+0.019
Neg logit contributions
hy-0.033
Mic-0.030
 rides-0.027
Red-0.027
Roger-0.027
Token.
Token ID13
Feature activation-8.62e-02

Pos logit contributions
hy+0.010
Mic+0.009
 rides+0.008
Red+0.008
Roger+0.008
Neg logit contributions
riage-0.008
ipples-0.008
ry-0.007
flower-0.007
ac-0.007
Token He
Token ID679
Feature activation-1.84e-01

Pos logit contributions
riage+0.170
ipples+0.166
ry+0.163
flower+0.162
 department+0.159
Neg logit contributions
hy-0.140
Mic-0.123
 rides-0.106
Red-0.104
Roger-0.102
Token in
Token ID287
Feature activation-4.40e-02

Pos logit contributions
hy+1.108
Mic+1.030
Red+0.891
 rides+0.872
Roger+0.872
Neg logit contributions
riage-0.994
ry-0.978
flower-0.974
ipples-0.973
ac-0.933
Token the
Token ID262
Feature activation-8.01e-03

Pos logit contributions
riage+0.067
ipples+0.065
ry+0.063
flower+0.063
ac+0.062
Neg logit contributions
hy-0.091
Mic-0.083
 rides-0.074
Red-0.073
Roger-0.072
Token corner
Token ID5228
Feature activation-1.94e-01

Pos logit contributions
riage+0.011
ipples+0.011
ry+0.011
flower+0.011
ac+0.010
Neg logit contributions
hy-0.017
Mic-0.016
 rides-0.014
Red-0.014
Roger-0.014
Token of
Token ID286
Feature activation+5.71e-02

Pos logit contributions
riage+0.292
ipples+0.285
ry+0.276
flower+0.274
ac+0.268
Neg logit contributions
hy-0.405
Mic-0.367
 rides-0.329
Red-0.325
Roger-0.319
Token the
Token ID262
Feature activation-2.47e-02

Pos logit contributions
hy+0.120
Mic+0.109
 rides+0.097
Red+0.097
Roger+0.095
Neg logit contributions
riage-0.085
ipples-0.083
ry-0.080
flower-0.080
ac-0.078
Token living
Token ID2877
Feature activation-2.95e+00

Pos logit contributions
riage+0.030
ipples+0.030
ry+0.029
flower+0.028
ac+0.028
Neg logit contributions
hy-0.059
Mic-0.054
 rides-0.049
Red-0.048
Roger-0.048
Token room
Token ID2119
Feature activation+1.60e-01

Pos logit contributions
riage+5.002
ipples+4.641
 department+4.626
andy+4.491
flower+4.435
Neg logit contributions
hy-5.743
 rides-5.037
nam-4.852
Mic-4.823
 blonde-4.708
Token.
Token ID13
Feature activation+6.10e-01

Pos logit contributions
hy+0.325
Mic+0.301
Red+0.264
 rides+0.262
Roger+0.260
Neg logit contributions
riage-0.245
ipples-0.240
ry-0.236
flower-0.232
 department-0.226
Token She
Token ID1375
Feature activation+2.47e-01

Pos logit contributions
hy+1.102
Mic+0.979
 rides+0.862
Red+0.835
Roger+0.832
Neg logit contributions
riage-1.079
ipples-1.045
ry-1.043
andy-1.022
flower-1.022
Token didn
Token ID1422
Feature activation+4.64e-01

Pos logit contributions
hy+0.555
Mic+0.504
 rides+0.452
Red+0.451
Roger+0.445
Neg logit contributions
riage-0.327
ipples-0.322
ry-0.319
flower-0.313
andy-0.310
Token't
Token ID470
Feature activation+5.30e-01

Pos logit contributions
hy+0.722
Mic+0.642
 rides+0.537
Red+0.529
nam+0.524
Neg logit contributions
riage-0.921
ipples-0.912
ry-0.896
flower-0.881
andy-0.878
Token were
Token ID547
Feature activation-8.90e-01

Pos logit contributions
riage+0.206
ipples+0.202
ry+0.196
flower+0.195
ac+0.191
Neg logit contributions
hy-0.265
Mic-0.240
 rides-0.213
Red-0.211
Roger-0.208
Token crashing
Token ID21899
Feature activation-6.58e-01

Pos logit contributions
riage+1.751
ry+1.673
ipples+1.652
flower+1.621
andy+1.606
Neg logit contributions
hy-1.514
Mic-1.301
 rides-1.159
Red-1.116
nam-1.103
Token and
Token ID290
Feature activation-7.60e-01

Pos logit contributions
riage+1.070
ipples+1.033
ry+0.997
andy+0.984
flower+0.983
Neg logit contributions
hy-1.326
Mic-1.168
 rides-1.055
Red-1.046
Roger-1.016
Token the
Token ID262
Feature activation-6.44e-01

Pos logit contributions
riage+1.152
ipples+1.103
ry+1.096
ac+1.076
flower+1.060
Neg logit contributions
hy-1.585
Mic-1.446
 rides-1.300
Red-1.282
Roger-1.271
Token se
Token ID384
Feature activation+2.19e-01

Pos logit contributions
riage+0.842
ry+0.800
ipples+0.796
flower+0.774
 department+0.768
Neg logit contributions
hy-1.484
Mic-1.341
 rides-1.245
Red-1.211
Roger-1.195
Tokenag
Token ID363
Feature activation-2.54e+00

Pos logit contributions
hy+0.334
Mic+0.296
 rides+0.250
Red+0.246
Roger+0.244
Neg logit contributions
riage-0.447
ipples-0.443
ry-0.432
ac-0.428
flower-0.428
Tokenull
Token ID724
Feature activation+5.84e-01

Pos logit contributions
riage+2.030
 department+1.701
ours+1.284
ipples+1.254
flower+1.208
Neg logit contributions
hy-7.493
Mic-6.743
hell-6.527
looking-6.516
 rides-6.413
Tokens
Token ID82
Feature activation+4.25e-01

Pos logit contributions
hy+1.234
Mic+1.131
Robert+0.996
Red+0.993
 rides+0.991
Neg logit contributions
ipples-0.847
ry-0.846
flower-0.836
riage-0.829
ac-0.796
Token were
Token ID547
Feature activation-3.65e-01

Pos logit contributions
hy+0.797
Mic+0.739
Red+0.640
 rides+0.638
Robert+0.628
Neg logit contributions
ipples-0.708
riage-0.698
ry-0.689
flower-0.686
 department-0.665
Token squ
Token ID2809
Feature activation+1.10e-01

Pos logit contributions
riage+0.678
ipples+0.655
ry+0.645
flower+0.639
andy+0.629
Neg logit contributions
hy-0.646
Mic-0.570
 rides-0.501
Red-0.492
Roger-0.484
Tokenaw
Token ID707
Feature activation+6.27e-01

Pos logit contributions
hy+0.238
Mic+0.217
 rides+0.194
Red+0.192
Roger+0.191
Neg logit contributions
riage-0.157
ipples-0.157
ry-0.151
flower-0.149
ac-0.148
Tokenows
Token ID1666
Feature activation-4.17e-01

Pos logit contributions
hy+0.235
Mic+0.211
 rides+0.179
Red+0.179
Roger+0.177
Neg logit contributions
riage-0.260
ipples-0.257
ry-0.252
flower-0.248
 department-0.246
Token and
Token ID290
Feature activation-4.59e-01

Pos logit contributions
riage+0.672
ipples+0.655
ry+0.636
flower+0.632
andy+0.626
Neg logit contributions
hy-0.825
Mic-0.735
 rides-0.663
Red-0.652
Roger-0.639
Token blankets
Token ID34794
Feature activation+1.16e-01

Pos logit contributions
riage+0.766
ipples+0.750
ry+0.731
flower+0.725
andy+0.717
Neg logit contributions
hy-0.879
Mic-0.788
 rides-0.707
Red-0.690
Roger-0.675
Token."
Token ID526
Feature activation-1.95e+00

Pos logit contributions
hy+0.250
Mic+0.230
 rides+0.204
Red+0.203
Roger+0.201
Neg logit contributions
riage-0.165
ipples-0.161
ry-0.157
flower-0.155
ac-0.152
Token
Token ID198
Feature activation-1.99e+00

Pos logit contributions
riage+3.741
flower+3.584
ric+3.441
ac+3.396
ipples+3.361
Neg logit contributions
hy-3.322
Mic-2.887
 rides-2.744
 Robert-2.615
 Bow-2.583
Token
Token ID198
Feature activation-2.65e+00

Pos logit contributions
riage+4.533
flower+4.342
 department+4.196
ipples+4.194
ric+4.148
Neg logit contributions
hy-2.693
Mic-2.386
Red-2.078
 rides-2.059
Roger-1.996
Token"
Token ID1
Feature activation-1.72e-02

Pos logit contributions
riage+5.190
ric+5.116
flower+4.752
 department+4.751
ipples+4.548
Neg logit contributions
hy-4.453
Mic-4.120
Red-3.755
Blue-3.698
Robert-3.591
TokenOkay
Token ID16454
Feature activation+3.99e-02

Pos logit contributions
riage+0.027
ipples+0.027
ry+0.026
flower+0.026
ac+0.025
Neg logit contributions
hy-0.035
Mic-0.031
 rides-0.028
Red-0.027
Roger-0.027
Token,"
Token ID553
Feature activation-7.95e-01

Pos logit contributions
hy+0.067
Mic+0.059
 rides+0.051
Red+0.050
Roger+0.049
Neg logit contributions
riage-0.077
ipples-0.076
ry-0.074
flower-0.074
ac-0.073
Token Mia
Token ID32189
Feature activation+1.11e+00

Pos logit contributions
riage+0.795
ipples+0.738
flower+0.720
ry+0.711
ric+0.698
Neg logit contributions
hy-2.073
Mic-1.906
 rides-1.772
Red-1.737
Roger-1.721
Token said
Token ID531
Feature activation+3.93e-01

Pos logit contributions
hy+2.003
Mic+1.828
Red+1.594
Roger+1.566
 rides+1.542
Neg logit contributions
ry-1.973
ipples-1.941
riage-1.922
flower-1.916
ac-1.913
Token they
Token ID484
Feature activation+3.80e-01

Pos logit contributions
ipples+2.154
 department+2.077
riage+2.076
ucks+2.002
flower+1.997
Neg logit contributions
hy-2.945
Mic-2.540
 rides-2.414
nam-2.380
Red-2.322
Token do
Token ID466
Feature activation-1.97e+00

Pos logit contributions
hy+0.823
Mic+0.762
Red+0.675
 rides+0.671
Roger+0.666
Neg logit contributions
riage-0.536
ipples-0.529
ry-0.507
flower-0.506
ac-0.497
Token not
Token ID407
Feature activation-4.76e-01

Pos logit contributions
ipples+4.015
flower+3.968
riage+3.918
 department+3.888
ucks+3.845
Neg logit contributions
hy-3.103
 rides-2.630
Mic-2.505
 long-2.317
 Bow-2.305
Token like
Token ID588
Feature activation-1.22e+00

Pos logit contributions
riage+0.912
ipples+0.888
ry+0.876
flower+0.869
 department+0.860
Neg logit contributions
hy-0.799
Mic-0.697
 rides-0.614
Red-0.600
Roger-0.585
Token mosquitoes
Token ID35573
Feature activation+1.34e-01

Pos logit contributions
riage+1.901
ipples+1.806
ry+1.770
ac+1.769
flower+1.753
Neg logit contributions
hy-2.552
Mic-2.227
 rides-2.169
Red-2.002
 Bow-1.996
Token.
Token ID13
Feature activation-2.49e+00

Pos logit contributions
hy+0.271
Mic+0.247
Red+0.217
 rides+0.217
Roger+0.214
Neg logit contributions
riage-0.206
ipples-0.203
ry-0.198
flower-0.197
 department-0.192
Token Mos
Token ID5826
Feature activation+5.76e-01

Pos logit contributions
riage+4.937
flower+4.848
 department+4.842
ipples+4.733
ry+4.708
Neg logit contributions
hy-4.191
Mic-3.495
Red-3.319
 rides-3.266
 blonde-3.255
Tokenquit
Token ID47391
Feature activation-1.58e+00

Pos logit contributions
hy+0.900
Mic+0.778
 rides+0.650
nam+0.649
Red+0.648
Neg logit contributions
ipples-1.165
riage-1.156
andy-1.121
flower-1.118
ry-1.117
Tokenoes
Token ID3028
Feature activation+6.33e-02

Pos logit contributions
riage+1.508
ipples+1.408
 department+1.337
ry+1.285
flower+1.276
Neg logit contributions
hy-4.243
Mic-3.709
 rides-3.532
Red-3.508
Blue-3.409
Token are
Token ID389
Feature activation-9.22e-01

Pos logit contributions
hy+0.131
Mic+0.119
 rides+0.105
Red+0.105
Roger+0.104
Neg logit contributions
riage-0.095
ipples-0.094
ry-0.091
flower-0.090
ac-0.089
Token small
Token ID1402
Feature activation-1.76e-01

Pos logit contributions
riage+1.851
ipples+1.772
ry+1.762
flower+1.737
andy+1.709
Neg logit contributions
hy-1.499
Mic-1.297
 rides-1.136
Red-1.096
Roger-1.071
Token other
Token ID584
Feature activation-8.08e-01

Pos logit contributions
riage+1.524
ipples+1.487
ry+1.473
flower+1.439
ac+1.437
Neg logit contributions
hy-1.670
Mic-1.466
 rides-1.327
Red-1.282
Roger-1.255
Token man
Token ID582
Feature activation+6.06e-01

Pos logit contributions
riage+1.270
ipples+1.247
flower+1.209
ry+1.202
 department+1.185
Neg logit contributions
hy-1.633
Mic-1.444
 rides-1.337
Red-1.280
Roger-1.248
Token.
Token ID13
Feature activation-3.89e-01

Pos logit contributions
hy+1.266
Mic+1.185
Red+1.050
Roger+1.028
 rides+1.026
Neg logit contributions
riage-0.890
ry-0.866
ipples-0.858
flower-0.841
ac-0.830
Token He
Token ID679
Feature activation-6.92e-01

Pos logit contributions
riage+0.772
ipples+0.759
ry+0.742
flower+0.742
ac+0.728
Neg logit contributions
hy-0.630
Mic-0.552
 rides-0.477
Red-0.470
Roger-0.458
Token loved
Token ID6151
Feature activation-2.60e-01

Pos logit contributions
riage+1.021
ipples+0.982
ry+0.941
flower+0.940
 department+0.922
Neg logit contributions
hy-1.476
Mic-1.329
 rides-1.225
Red-1.188
Roger-1.168
Token his
Token ID465
Feature activation-2.34e+00

Pos logit contributions
riage+0.406
ipples+0.397
ry+0.386
flower+0.383
ac+0.375
Neg logit contributions
hy-0.530
Mic-0.479
 rides-0.430
Red-0.423
Roger-0.416
Token pipe
Token ID12656
Feature activation+9.15e-01

Pos logit contributions
riage+4.064
ucks+3.894
ipples+3.800
ours+3.701
ry+3.678
Neg logit contributions
hy-4.528
Mic-4.014
 rides-3.766
 long-3.753
 Bow-3.489
Token and
Token ID290
Feature activation-3.88e-01

Pos logit contributions
hy+1.864
Mic+1.741
Red+1.536
Roger+1.507
Robert+1.497
Neg logit contributions
ipples-1.357
flower-1.342
riage-1.338
ry-1.336
 department-1.302
Token he
Token ID339
Feature activation-4.29e-01

Pos logit contributions
riage+0.682
ipples+0.668
ry+0.650
flower+0.646
ac+0.636
Neg logit contributions
hy-0.714
Mic-0.632
 rides-0.565
Red-0.550
Roger-0.541
Token could
Token ID714
Feature activation-3.84e-01

Pos logit contributions
riage+0.622
ipples+0.602
ry+0.581
flower+0.578
 department+0.568
Neg logit contributions
hy-0.922
Mic-0.833
 rides-0.759
Red-0.745
Roger-0.732
Token never
Token ID1239
Feature activation-1.45e-02

Pos logit contributions
riage+0.601
ipples+0.583
ry+0.567
flower+0.562
ac+0.556
Neg logit contributions
hy-0.783
Mic-0.703
 rides-0.635
Red-0.622
Roger-0.612