TOP ACTIVATIONS
MAX = -1.000

Tom went to his bed and hid under the covers .
Tom went to his bed and hid under the covers
. Tom went to his bed and hid under
Tom went to his bed and hid under the
talking , table ?" The table did not say anything either
you talking , table ?" The table did not say anything
, table ?" The table did not say anything either .
table ?" The table did not say anything either . Tom
either . Tom was scared . Tom went to
anything either . Tom was scared . Tom went
not say anything either . Tom was scared .
say anything either . Tom was scared . Tom
Tom was scared . Tom went to his bed
. Tom was scared . Tom went to his
was scared . Tom went to his bed and
scared . Tom went to his bed and hid
did not say anything either . Tom was scared .
table did not say anything either . Tom was scared .
?" The table did not say anything either . Tom was
The table did not say anything either . Tom was scared

MIN ACTIVATIONS
MIN = -1.000

Tom went to his bed and hid under the covers .
Tom went to his bed and hid under the covers
. Tom went to his bed and hid under
Tom went to his bed and hid under the
talking , table ?" The table did not say anything either
you talking , table ?" The table did not say anything
, table ?" The table did not say anything either .
table ?" The table did not say anything either . Tom
either . Tom was scared . Tom went to
anything either . Tom was scared . Tom went
not say anything either . Tom was scared .
say anything either . Tom was scared . Tom
Tom was scared . Tom went to his bed
. Tom was scared . Tom went to his
was scared . Tom went to his bed and
scared . Tom went to his bed and hid
did not say anything either . Tom was scared .
table did not say anything either . Tom was scared .
?" The table did not say anything either . Tom was
The table did not say anything either . Tom was scared

INTERVAL -1.000 - -1.000
CONTAINS 0.000%

TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 and+1.532
,+1.464
.+1.245
 the+1.099
 a+0.937
Neg logit contributions
lor-6.934
 former-6.905
umm-6.900
ocol-6.876
arians-6.766
Token covers
Token ID8698
Feature activation-1.00e+00

Pos logit contributions
 and+4.412
,+4.369
.+4.146
 the+4.127
 a+3.973
Neg logit contributions
lor-4.047
umm-4.031
 former-3.973
ocol-3.943
arians-3.864
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.763
,+0.705
 the+0.464
.+0.458
 a+0.320
Neg logit contributions
lor-7.725
umm-7.663
 former-7.591
ocol-7.545
arians-7.521
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 and+1.532
,+1.464
.+1.245
 the+1.099
 a+0.937
Neg logit contributions
lor-6.934
 former-6.905
umm-6.900
ocol-6.876
arians-6.766
Token covers
Token ID8698
Feature activation-1.00e+00

Pos logit contributions
 and+4.412
,+4.369
.+4.146
 the+4.127
 a+3.973
Neg logit contributions
lor-4.047
umm-4.031
 former-3.973
ocol-3.943
arians-3.864
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 and+1.532
,+1.464
.+1.245
 the+1.099
 a+0.937
Neg logit contributions
lor-6.934
 former-6.905
umm-6.900
ocol-6.876
arians-6.766
Token talking
Token ID3375
Feature activation-1.00e+00

Pos logit contributions
 and+3.731
,+3.640
.+3.455
 the+3.427
 a+3.243
Neg logit contributions
lor-4.752
 former-4.710
umm-4.675
arians-4.612
ocol-4.595
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
 and+2.784
,+2.732
.+2.526
 the+2.511
 a+2.325
Neg logit contributions
lor-5.684
 former-5.632
umm-5.619
arians-5.534
osity-5.507
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token you
Token ID345
Feature activation-1.00e+00

Pos logit contributions
 and+3.236
,+3.182
.+3.016
 the+2.955
 a+2.812
Neg logit contributions
lor-5.173
umm-5.125
 former-5.110
arians-5.105
osity-5.021
Token talking
Token ID3375
Feature activation-1.00e+00

Pos logit contributions
 and+3.731
,+3.640
.+3.455
 the+3.427
 a+3.243
Neg logit contributions
lor-4.752
 former-4.710
umm-4.675
arians-4.612
ocol-4.595
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
 and+2.784
,+2.732
.+2.526
 the+2.511
 a+2.325
Neg logit contributions
lor-5.684
 former-5.632
umm-5.619
arians-5.534
osity-5.507
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
 and+2.784
,+2.732
.+2.526
 the+2.511
 a+2.325
Neg logit contributions
lor-5.684
 former-5.632
umm-5.619
arians-5.534
osity-5.507
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 and+1.532
,+1.464
.+1.245
 the+1.099
 a+0.937
Neg logit contributions
lor-6.934
 former-6.905
umm-6.900
ocol-6.876
arians-6.766
Token covers
Token ID8698
Feature activation-1.00e+00

Pos logit contributions
 and+4.412
,+4.369
.+4.146
 the+4.127
 a+3.973
Neg logit contributions
lor-4.047
umm-4.031
 former-3.973
ocol-3.943
arians-3.864
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.763
,+0.705
 the+0.464
.+0.458
 a+0.320
Neg logit contributions
lor-7.725
umm-7.663
 former-7.591
ocol-7.545
arians-7.521
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 and+1.532
,+1.464
.+1.245
 the+1.099
 a+0.937
Neg logit contributions
lor-6.934
 former-6.905
umm-6.900
ocol-6.876
arians-6.766
Token covers
Token ID8698
Feature activation-1.00e+00

Pos logit contributions
 and+4.412
,+4.369
.+4.146
 the+4.127
 a+3.973
Neg logit contributions
lor-4.047
umm-4.031
 former-3.973
ocol-3.943
arians-3.864
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token under
Token ID739
Feature activation-1.00e+00

Pos logit contributions
 and+2.962
,+2.908
.+2.664
 the+2.613
 a+2.469
Neg logit contributions
lor-5.494
umm-5.473
 former-5.442
ocol-5.434
arians-5.377
Token the
Token ID262
Feature activation-1.00e+00

Pos logit contributions
 and+1.532
,+1.464
.+1.245
 the+1.099
 a+0.937
Neg logit contributions
lor-6.934
 former-6.905
umm-6.900
ocol-6.876
arians-6.766
Token talking
Token ID3375
Feature activation-1.00e+00

Pos logit contributions
 and+3.731
,+3.640
.+3.455
 the+3.427
 a+3.243
Neg logit contributions
lor-4.752
 former-4.710
umm-4.675
arians-4.612
ocol-4.595
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
 and+2.784
,+2.732
.+2.526
 the+2.511
 a+2.325
Neg logit contributions
lor-5.684
 former-5.632
umm-5.619
arians-5.534
osity-5.507
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token you
Token ID345
Feature activation-1.00e+00

Pos logit contributions
 and+3.236
,+3.182
.+3.016
 the+2.955
 a+2.812
Neg logit contributions
lor-5.173
umm-5.125
 former-5.110
arians-5.105
osity-5.021
Token talking
Token ID3375
Feature activation-1.00e+00

Pos logit contributions
 and+3.731
,+3.640
.+3.455
 the+3.427
 a+3.243
Neg logit contributions
lor-4.752
 former-4.710
umm-4.675
arians-4.612
ocol-4.595
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
 and+2.784
,+2.732
.+2.526
 the+2.511
 a+2.325
Neg logit contributions
lor-5.684
 former-5.632
umm-5.619
arians-5.534
osity-5.507
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token,
Token ID11
Feature activation-1.00e+00

Pos logit contributions
 and+2.784
,+2.732
.+2.526
 the+2.511
 a+2.325
Neg logit contributions
lor-5.684
 former-5.632
umm-5.619
arians-5.534
osity-5.507
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+3.706
,+3.662
.+3.499
 the+3.411
 a+3.251
Neg logit contributions
 former-4.722
lor-4.713
umm-4.708
arians-4.616
ocol-4.607
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+1.383
,+1.315
.+1.093
 the+1.036
 a+0.878
Neg logit contributions
lor-7.158
 former-7.033
umm-7.013
ocol-6.947
arians-6.932
TokenTom
Token ID13787
Feature activation-1.00e+00

Pos logit contributions
 and+3.677
,+3.604
.+3.407
 the+3.350
 a+3.179
Neg logit contributions
lor-4.939
umm-4.833
 former-4.816
bek-4.763
arians-4.758
Token went
Token ID1816
Feature activation-1.00e+00

Pos logit contributions
 and+2.939
,+2.912
.+2.760
 the+2.713
 a+2.539
Neg logit contributions
lor-5.478
 former-5.367
umm-5.354
arians-5.351
bek-5.259
Token to
Token ID284
Feature activation-1.00e+00

Pos logit contributions
 and+1.862
,+1.803
.+1.607
 the+1.556
 a+1.392
Neg logit contributions
lor-6.621
 former-6.546
umm-6.532
arians-6.474
ocol-6.445
Token his
Token ID465
Feature activation-1.00e+00

Pos logit contributions
 and+2.354
,+2.288
.+2.084
 the+1.945
 a+1.809
Neg logit contributions
lor-6.112
 former-6.112
umm-6.066
ocol-6.032
arians-5.999
Token bed
Token ID3996
Feature activation-1.00e+00

Pos logit contributions
 and+3.782
,+3.719
.+3.530
 the+3.482
 a+3.302
Neg logit contributions
 former-4.731
umm-4.693
lor-4.688
arians-4.604
ocol-4.531
Token and
Token ID290
Feature activation-1.00e+00

Pos logit contributions
 and+1.139
,+1.086
 the+0.914
.+0.887
 a+0.730
Neg logit contributions
lor-7.320
 former-7.251
umm-7.228
arians-7.134
osity-7.048
Token hid
Token ID24519
Feature activation-1.00e+00

Pos logit contributions
 and+3.558
,+3.501
.+3.318
 the+3.211
 a+3.043
Neg logit contributions
lor-4.923
 former-4.888
umm-4.804
arians-4.768
bek-4.736
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 and+3.032
,+2.964
.+2.749
 the+2.691
 a+2.515
Neg logit contributions
lor-5.502
umm-5.475
 former-5.442
arians-5.345
ocol-5.328
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.497
,+0.436
 the+0.265
.+0.227
 a+0.129
Neg logit contributions
umm-7.852
lor-7.782
ocol-7.742
 former-7.733
arians-7.676
Token?"
Token ID1701
Feature activation-1.00e+00

Pos logit contributions
 and+2.151
,+2.084
 the+1.905
.+1.900
 a+1.721
Neg logit contributions
lor-6.287
umm-6.255
 former-6.216
arians-6.130
ocol-6.122
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token The
Token ID383
Feature activation-1.00e+00

Pos logit contributions
 and+3.108
,+3.029
.+2.844
 the+2.755
 a+2.608
Neg logit contributions
lor-5.346
 former-5.315
umm-5.288
ocol-5.238
arians-5.209
Token table
Token ID3084
Feature activation-1.00e+00

Pos logit contributions
 and+4.158
,+4.099
.+3.898
 the+3.837
 a+3.655
Neg logit contributions
umm-4.334
 former-4.311
lor-4.306
ocol-4.239
arians-4.210
Token did
Token ID750
Feature activation-1.00e+00

Pos logit contributions
 and+3.733
,+3.669
.+3.492
 the+3.470
 a+3.288
Neg logit contributions
lor-4.721
umm-4.663
 former-4.625
ocol-4.545
arians-4.542
Token not
Token ID407
Feature activation-1.00e+00

Pos logit contributions
 and+2.363
,+2.283
.+2.090
 the+2.049
 a+1.864
Neg logit contributions
lor-6.097
 former-6.066
umm-5.999
ocol-5.945
arians-5.938
Token say
Token ID910
Feature activation-1.00e+00

Pos logit contributions
 and+3.886
,+3.820
.+3.615
 the+3.593
 a+3.401
Neg logit contributions
umm-4.547
 former-4.535
lor-4.516
ocol-4.483
arians-4.433
Token anything
Token ID1997
Feature activation-1.00e+00

Pos logit contributions
 and+4.160
,+4.023
.+3.852
 the+3.789
 a+3.614
Neg logit contributions
lor-4.396
umm-4.290
 former-4.262
ocol-4.234
arians-4.192
Token either
Token ID2035
Feature activation-1.00e+00

Pos logit contributions
 and+1.844
,+1.770
.+1.562
 the+1.557
 a+1.402
Neg logit contributions
lor-6.589
umm-6.532
 former-6.524
arians-6.459
ocol-6.439
Token.
Token ID13
Feature activation-1.00e+00

Pos logit contributions
 and+0.886
,+0.802
 the+0.624
.+0.598
 a+0.466
Neg logit contributions
lor-7.539
umm-7.485
 former-7.416
ocol-7.385
arians-7.368
Token Tom
Token ID4186
Feature activation-1.00e+00

Pos logit contributions
 and+2.929
,+2.867
.+2.660
 the+2.593
 a+2.420
Neg logit contributions
lor-5.613
 former-5.544
umm-5.540
arians-5.442
ocol-5.441
Token was
Token ID373
Feature activation-1.00e+00

Pos logit contributions
 and+2.977
,+2.950
.+2.785
 the+2.736
 a+2.562
Neg logit contributions
lor-5.440
 former-5.369
umm-5.348
arians-5.316
ocol-5.227
Token scared
Token ID12008
Feature activation-1.00e+00

Pos logit contributions
 and+3.632
,+3.580
.+3.379
 the+3.294
 a+3.089
Neg logit contributions
 former-4.832
lor-4.824
umm-4.770
ocol-4.714
arians-4.700