TOP ACTIVATIONS
MAX = 3.354

. It landed on the other side of the fence ,
!" Ben sits on the other chair and says , "
Sad ie sat down on the floor and began to cry
saw something shiny on the other side of the room .
delicious looking cake on the other side .
the robe and sat on the throne . They laughed and
They both sat down on the couch and watched a show
cookies . They sat on the floor and ate them .
living room and sat on the couch . Lily turned on
was her grandma on the other side . Grand ma said
. He climbed on the other end and tried to balance
saw their ball on the other side of the yard .
a shiny box on the other side of the river .
. They all sit on the bench and smile . The
apr ons and sat on the floor . "
picks one and sits on the floor . "
saw the blocks scattered on the floor and the children yelling
agree . They sit on the couch and hold hands .
. He played with cars and balls and blocks , but
about a cave on the other side of the forest .

MIN ACTIVATIONS
MIN = -3.602

hed the pain . She sn ugg led in her mom
the knot was . He sn uck up to the knot
be craft y . He sn uck around until he was
own blue pillow . She sn ugg led up and went
tried it out . She sn ugg led into the soft
steal a haircut ! He sn uck into a salon and
delicious dinner she made ! From that day on , the
didn 't listen . She sn uck into the zoo at
loved it very much . From that day on , Lily
banana he ever tasted ! From that day on , Max
the DVD player . They sn ugg led up on the
steal her toys . He sn uck around the corner and
It was very refreshing ! From that day on , Lily
the salad so much . From then on , Fl oppy
in her arms . She sn ugg led the doll close
were so happy . They sn ugg led up on the
a little better . She sn ugg led up in her
he was welcome . They sn ugg led in her heart
before bed , she would sn uggle up and give her
a very special reward . From then on , the girl

INTERVAL 1.615 - 3.354
CONTAINS 0.361%

homes . They get off the bus and hug their moms
lines and circles on a new paper . She said ,
loved the color yellow because it was so bright and happy
little girl was so happy that she quickly got in the
oons . Mom gives them each a bowl of soup .

INTERVAL -0.124 - 1.615
CONTAINS 58.869%

problems . One day , a lady came to
was happy to see the child , so he waved at
back . But the balls are too many . They fall
now ." Tim my said , " Okay , mom my
block . They did not fight anymore . They played with

INTERVAL -1.863 - -0.124
CONTAINS 40.614%

love , it 's important to be kind and generous ,"
hug their mom again and go to fix their cars .
, stars , and animals . They used many colors ,
she opened the box and inside , there was a beautiful
you can draw a frog too . Here , take a

INTERVAL -3.602 - -1.863
CONTAINS 0.155%

view . From that day on , the
matching leaves and flowers . From then on , she loved
his nap . He thought n aps were un important and
became very famous . People from all around the world bought
his help . From that day on , King
Token.
Token ID13
Feature activation-4.64e-01

Pos logit contributions
 eyel+0.037
cel+0.033
 moral+0.031
 floor+0.031
 fore+0.031
Neg logit contributions
ised-0.075
eday-0.063
rous-0.063
 that-0.061
und-0.060
Token It
Token ID632
Feature activation+1.03e+00

Pos logit contributions
ised+0.785
eday+0.638
rous+0.623
 that+0.595
und+0.578
Neg logit contributions
 eyel-0.667
cel-0.617
 moral-0.593
 floor-0.591
 fore-0.577
Token landed
Token ID11406
Feature activation+7.43e-01

Pos logit contributions
 eyel+1.316
cel+1.271
 moral+1.185
quin+1.168
 fore+1.148
Neg logit contributions
ised-1.873
 that-1.561
eday-1.502
rous-1.478
und-1.441
Token on
Token ID319
Feature activation+3.76e-01

Pos logit contributions
 eyel+1.005
cel+0.950
 moral+0.906
 floor+0.877
quin+0.868
Neg logit contributions
ised-1.310
 that-1.071
eday-1.051
rous-1.040
aw-1.018
Token the
Token ID262
Feature activation+2.09e+00

Pos logit contributions
 eyel+0.514
cel+0.488
 moral+0.463
 fore+0.453
 floor+0.451
Neg logit contributions
ised-0.651
eday-0.524
 that-0.521
rous-0.519
und-0.491
Token other
Token ID584
Feature activation+3.35e+00

Pos logit contributions
 eyel+2.540
cel+2.463
 Si+2.361
quin+2.307
Beck+2.280
Neg logit contributions
ised-3.853
 that-3.405
eday-3.037
rous-3.029
aw-3.018
Token side
Token ID1735
Feature activation+1.97e-01

Pos logit contributions
Beck+5.074
 eyel+5.029
cel+4.986
quin+4.862
 Si+4.787
Neg logit contributions
ised-5.478
 that-5.154
aw-4.330
eday-4.251
 goodness-4.110
Token of
Token ID286
Feature activation-1.42e-01

Pos logit contributions
 eyel+0.288
cel+0.273
 moral+0.261
 floor+0.256
 fore+0.256
Neg logit contributions
ised-0.320
eday-0.255
rous-0.252
 that-0.249
und-0.237
Token the
Token ID262
Feature activation+1.51e+00

Pos logit contributions
ised+0.249
eday+0.202
rous+0.200
 that+0.194
und+0.190
Neg logit contributions
 eyel-0.192
cel-0.178
 moral-0.173
 floor-0.171
 fore-0.170
Token fence
Token ID13990
Feature activation+6.53e-03

Pos logit contributions
 eyel+2.027
cel+1.963
 moral+1.772
quin+1.769
 Si+1.757
Neg logit contributions
ised-2.647
 that-2.240
rous-2.141
eday-2.086
aw-2.013
Token,
Token ID11
Feature activation+8.27e-01

Pos logit contributions
 eyel+0.007
cel+0.007
 moral+0.006
 floor+0.006
 fore+0.006
Neg logit contributions
ised-0.013
eday-0.011
rous-0.011
 that-0.011
und-0.010
Token!"
Token ID2474
Feature activation+3.66e-01

Pos logit contributions
 eyel+0.001
cel+0.001
 moral+0.001
 floor+0.001
 fore+0.001
Neg logit contributions
ised-0.002
rous-0.001
eday-0.001
 that-0.001
und-0.001
Token Ben
Token ID3932
Feature activation+6.72e-01

Pos logit contributions
 eyel+0.491
cel+0.467
 moral+0.442
 fore+0.436
 floor+0.433
Neg logit contributions
ised-0.642
eday-0.519
 that-0.516
rous-0.515
und-0.488
Token sits
Token ID10718
Feature activation+1.70e-01

Pos logit contributions
 eyel+1.042
cel+1.001
 moral+0.946
 fore+0.929
 floor+0.928
Neg logit contributions
ised-1.024
eday-0.811
 that-0.806
rous-0.791
aw-0.763
Token on
Token ID319
Feature activation+6.04e-01

Pos logit contributions
 eyel+0.215
cel+0.201
 moral+0.191
 floor+0.187
 fore+0.186
Neg logit contributions
ised-0.314
eday-0.257
rous-0.255
 that-0.252
und-0.242
Token the
Token ID262
Feature activation+2.33e+00

Pos logit contributions
 eyel+0.875
cel+0.837
 moral+0.789
 fore+0.774
 floor+0.764
Neg logit contributions
ised-0.988
 that-0.798
eday-0.791
rous-0.777
und-0.734
Token other
Token ID584
Feature activation+3.28e+00

Pos logit contributions
 eyel+2.717
cel+2.665
quin+2.494
Beck+2.492
 Si+2.473
Neg logit contributions
ised-4.237
 that-4.017
eday-3.550
rous-3.409
aw-3.380
Token chair
Token ID5118
Feature activation-1.55e-01

Pos logit contributions
cel+4.865
Beck+4.837
 eyel+4.805
quin+4.551
 cor+4.444
Neg logit contributions
ised-5.206
 that-4.999
eday-4.352
aw-4.145
 goodness-3.769
Token and
Token ID290
Feature activation+9.36e-01

Pos logit contributions
ised+0.305
rous+0.253
eday+0.253
 that+0.243
und+0.239
Neg logit contributions
 eyel-0.175
cel-0.160
 moral-0.153
 floor-0.153
 fore-0.149
Token says
Token ID1139
Feature activation+1.88e-02

Pos logit contributions
 eyel+1.267
cel+1.217
 moral+1.168
 fore+1.143
 floor+1.111
Neg logit contributions
ised-1.581
 that-1.324
eday-1.311
rous-1.278
aw-1.208
Token,
Token ID11
Feature activation+1.46e-01

Pos logit contributions
 eyel+0.027
cel+0.025
 moral+0.024
 fore+0.023
 floor+0.023
Neg logit contributions
ised-0.032
eday-0.026
rous-0.025
 that-0.025
und-0.024
Token "
Token ID366
Feature activation-2.27e-02

Pos logit contributions
 eyel+0.164
cel+0.152
 moral+0.144
 floor+0.140
 fore+0.140
Neg logit contributions
ised-0.287
eday-0.238
rous-0.237
 that-0.233
und-0.225
TokenSad
Token ID26699
Feature activation+1.01e+00

Pos logit contributions
ised+0.584
eday+0.480
rous+0.472
 that+0.455
aw+0.438
Neg logit contributions
 eyel-0.545
cel-0.519
 moral-0.504
 fore-0.487
 floor-0.483
Tokenie
Token ID494
Feature activation-1.03e-01

Pos logit contributions
 eyel+1.603
cel+1.546
 magn+1.462
 moral+1.456
 fore+1.448
Neg logit contributions
ised-1.549
 that-1.214
eday-1.176
rous-1.124
und-1.098
Token sat
Token ID3332
Feature activation-6.59e-01

Pos logit contributions
ised+0.206
eday+0.173
rous+0.172
 that+0.167
und+0.163
Neg logit contributions
 eyel-0.113
cel-0.105
 moral-0.099
 floor-0.097
 fore-0.097
Token down
Token ID866
Feature activation+3.40e-01

Pos logit contributions
ised+1.164
eday+0.964
rous+0.961
 that+0.906
und+0.905
Neg logit contributions
 eyel-0.846
cel-0.809
 moral-0.775
 floor-0.763
 fore-0.751
Token on
Token ID319
Feature activation+4.44e-01

Pos logit contributions
 eyel+0.439
cel+0.414
 moral+0.391
 floor+0.384
 fore+0.380
Neg logit contributions
ised-0.613
eday-0.500
 that-0.495
rous-0.494
und-0.469
Token the
Token ID262
Feature activation+3.22e+00

Pos logit contributions
 eyel+0.623
cel+0.590
 moral+0.565
 fore+0.550
 floor+0.549
Neg logit contributions
ised-0.747
eday-0.600
 that-0.599
rous-0.593
und-0.561
Token floor
Token ID4314
Feature activation-1.28e-02

Pos logit contributions
quin+3.269
 moral+3.080
 Si+3.011
 eyel+2.918
Beck+2.871
Neg logit contributions
ised-6.424
 that-6.266
eday-5.301
rous-5.244
Deal-5.096
Token and
Token ID290
Feature activation+1.83e-03

Pos logit contributions
ised+0.024
eday+0.020
rous+0.020
 that+0.019
und+0.019
Neg logit contributions
 eyel-0.015
cel-0.014
 moral-0.013
 floor-0.013
 fore-0.013
Token began
Token ID2540
Feature activation+4.02e-01

Pos logit contributions
 eyel+0.002
cel+0.002
 moral+0.002
 floor+0.002
 fore+0.002
Neg logit contributions
ised-0.003
eday-0.003
rous-0.003
 that-0.002
und-0.002
Token to
Token ID284
Feature activation+3.54e-01

Pos logit contributions
 eyel+0.724
cel+0.700
 moral+0.675
 floor+0.662
 fore+0.660
Neg logit contributions
ised-0.511
eday-0.377
 that-0.376
rous-0.376
und-0.345
Token cry
Token ID3960
Feature activation+3.97e-01

Pos logit contributions
 eyel+0.419
cel+0.395
 moral+0.375
 fore+0.365
 floor+0.363
Neg logit contributions
ised-0.669
eday-0.554
rous-0.549
 that-0.544
und-0.521
Token saw
Token ID2497
Feature activation-4.64e-01

Pos logit contributions
 eyel+1.290
cel+1.196
 moral+1.195
 fore+1.151
quin+1.132
Neg logit contributions
ised-1.761
 that-1.459
rous-1.387
aw-1.386
eday-1.368
Token something
Token ID1223
Feature activation-7.65e-02

Pos logit contributions
ised+0.881
eday+0.737
rous+0.725
und+0.686
Deal+0.681
Neg logit contributions
 eyel-0.564
cel-0.524
 moral-0.504
 floor-0.502
 fore-0.490
Token shiny
Token ID22441
Feature activation-1.43e+00

Pos logit contributions
ised+0.133
eday+0.109
rous+0.107
 that+0.103
und+0.102
Neg logit contributions
 eyel-0.104
cel-0.097
 moral-0.093
 floor-0.092
 fore-0.091
Token on
Token ID319
Feature activation-2.21e-01

Pos logit contributions
ised+2.829
rous+2.474
eday+2.419
und+2.270
Deal+2.250
Neg logit contributions
 eyel-1.580
 floor-1.435
cel-1.407
 moral-1.356
nob-1.355
Token the
Token ID262
Feature activation+2.53e+00

Pos logit contributions
ised+0.398
eday+0.326
rous+0.325
 that+0.310
und+0.307
Neg logit contributions
 eyel-0.283
cel-0.264
 moral-0.257
 floor-0.254
 fore-0.248
Token other
Token ID584
Feature activation+3.21e+00

Pos logit contributions
 eyel+2.127
cel+2.107
quin+2.074
 Si+1.978
 cor+1.900
Neg logit contributions
ised-5.465
 that-5.142
eday-4.692
rous-4.493
izing-4.348
Token side
Token ID1735
Feature activation+6.31e-01

Pos logit contributions
quin+4.488
 eyel+4.473
cel+4.416
 cor+4.404
Beck+4.258
Neg logit contributions
ised-5.301
 that-5.106
eday-4.371
aw-4.021
here-4.003
Token of
Token ID286
Feature activation-3.24e-02

Pos logit contributions
 eyel+0.960
cel+0.922
 moral+0.874
 fore+0.852
 floor+0.834
Neg logit contributions
ised-0.992
 that-0.798
eday-0.785
rous-0.775
und-0.736
Token the
Token ID262
Feature activation+1.87e+00

Pos logit contributions
ised+0.057
eday+0.046
rous+0.046
 that+0.044
und+0.043
Neg logit contributions
 eyel-0.043
cel-0.041
 moral-0.039
 floor-0.039
 fore-0.038
Token room
Token ID2119
Feature activation+7.54e-01

Pos logit contributions
 eyel+2.027
cel+1.978
quin+1.833
 moral+1.793
 cor+1.787
Neg logit contributions
ised-3.663
 that-3.300
eday-3.084
rous-3.082
und-2.867
Token.
Token ID13
Feature activation+3.69e-01

Pos logit contributions
 eyel+0.854
cel+0.809
 moral+0.763
 fore+0.731
 magn+0.709
Neg logit contributions
ised-1.471
 that-1.254
eday-1.217
rous-1.197
aw-1.170
Token delicious
Token ID12625
Feature activation-7.60e-01

Pos logit contributions
ised+0.210
eday+0.174
rous+0.171
 that+0.165
und+0.161
Neg logit contributions
 eyel-0.151
cel-0.141
 moral-0.136
 floor-0.134
 fore-0.132
Token looking
Token ID2045
Feature activation+8.64e-05

Pos logit contributions
ised+1.381
eday+1.148
rous+1.125
und+1.058
Fine+1.045
Neg logit contributions
 eyel-0.998
cel-0.923
 floor-0.914
 moral-0.890
 magn-0.873
Token cake
Token ID12187
Feature activation+4.29e-01

Pos logit contributions
 eyel+0.000
cel+0.000
 moral+0.000
 floor+0.000
 fore+0.000
Neg logit contributions
ised-0.000
eday-0.000
rous-0.000
 that-0.000
und-0.000
Token on
Token ID319
Feature activation-1.62e-02

Pos logit contributions
 eyel+0.546
cel+0.512
 moral+0.494
 fore+0.480
 floor+0.471
Neg logit contributions
ised-0.776
 that-0.636
eday-0.634
rous-0.626
und-0.599
Token the
Token ID262
Feature activation+1.99e+00

Pos logit contributions
ised+0.028
eday+0.023
rous+0.023
 that+0.022
und+0.021
Neg logit contributions
 eyel-0.022
cel-0.021
 moral-0.020
 floor-0.020
 fore-0.020
Token other
Token ID584
Feature activation+3.19e+00

Pos logit contributions
 eyel+1.889
cel+1.786
quin+1.684
 moral+1.610
 Si+1.590
Neg logit contributions
ised-4.142
 that-3.844
eday-3.511
rous-3.445
aw-3.399
Token side
Token ID1735
Feature activation+7.01e-01

Pos logit contributions
 eyel+4.529
cel+4.491
quin+4.387
 moral+4.132
Beck+4.121
Neg logit contributions
ised-5.314
 that-5.261
aw-4.485
eday-4.186
 goodness-4.108
Token.
Token ID13
Feature activation-2.54e-01

Pos logit contributions
 eyel+0.995
cel+0.959
 moral+0.906
 fore+0.880
 floor+0.856
Neg logit contributions
ised-1.170
 that-0.956
eday-0.938
rous-0.934
und-0.886
Token
Token ID220
Feature activation+4.14e-01

Pos logit contributions
ised+0.461
eday+0.380
rous+0.376
 that+0.361
und+0.352
Neg logit contributions
 eyel-0.328
cel-0.309
 moral-0.289
 floor-0.288
 fore-0.282
Token
Token ID198
Feature activation-1.11e-01

Pos logit contributions
 eyel+0.558
cel+0.523
 moral+0.496
 fore+0.488
 floor+0.487
Neg logit contributions
ised-0.724
eday-0.583
rous-0.581
 that-0.580
und-0.555
Token
Token ID198
Feature activation-2.10e-01

Pos logit contributions
ised+0.207
eday+0.171
rous+0.169
 that+0.164
und+0.159
Neg logit contributions
 eyel-0.136
cel-0.129
 moral-0.122
 floor-0.120
 fore-0.119
Token the
Token ID262
Feature activation+1.56e+00

Pos logit contributions
 eyel+1.927
cel+1.872
 moral+1.796
 fore+1.744
quin+1.682
Neg logit contributions
ised-2.276
 that-1.929
eday-1.849
rous-1.804
aw-1.758
Token robe
Token ID33192
Feature activation-1.19e-01

Pos logit contributions
 eyel+1.660
cel+1.647
nob+1.478
quin+1.471
 moral+1.452
Neg logit contributions
ised-2.973
 that-2.619
eday-2.501
rous-2.500
aw-2.431
Token and
Token ID290
Feature activation+8.18e-01

Pos logit contributions
ised+0.236
eday+0.196
rous+0.196
 that+0.188
und+0.185
Neg logit contributions
 eyel-0.135
cel-0.125
 moral-0.118
 floor-0.117
 fore-0.115
Token sat
Token ID3332
Feature activation-7.58e-01

Pos logit contributions
 eyel+1.122
cel+1.069
 moral+1.011
 fore+0.986
 floor+0.979
Neg logit contributions
ised-1.390
 that-1.151
eday-1.134
rous-1.113
aw-1.058
Token on
Token ID319
Feature activation-5.74e-02

Pos logit contributions
ised+1.420
eday+1.169
rous+1.167
und+1.102
 that+1.100
Neg logit contributions
 eyel-0.912
cel-0.860
 moral-0.828
 floor-0.826
 fore-0.806
Token the
Token ID262
Feature activation+3.18e+00

Pos logit contributions
ised+0.100
eday+0.081
rous+0.080
 that+0.077
und+0.076
Neg logit contributions
 eyel-0.078
cel-0.073
 moral-0.070
 floor-0.069
 fore-0.069
Token throne
Token ID19262
Feature activation+1.60e-01

Pos logit contributions
quin+3.084
cel+2.953
Beck+2.927
 Si+2.900
 eyel+2.826
Neg logit contributions
 that-6.391
ised-6.150
eday-5.206
aw-5.075
rous-5.020
Token.
Token ID13
Feature activation-6.35e-01

Pos logit contributions
 eyel+0.166
cel+0.153
 moral+0.144
 floor+0.140
 fore+0.140
Neg logit contributions
ised-0.328
eday-0.275
rous-0.274
 that-0.270
und-0.261
Token They
Token ID1119
Feature activation+6.69e-01

Pos logit contributions
ised+1.156
eday+0.938
rous+0.933
und+0.876
 that+0.875
Neg logit contributions
 eyel-0.835
cel-0.773
 moral-0.754
 floor-0.745
 fore-0.725
Token laughed
Token ID13818
Feature activation+1.74e-01

Pos logit contributions
 eyel+0.925
cel+0.868
 moral+0.838
 fore+0.821
 floor+0.811
Neg logit contributions
ised-1.131
eday-0.920
 that-0.920
rous-0.901
und-0.858
Token and
Token ID290
Feature activation+6.16e-01

Pos logit contributions
 eyel+0.209
cel+0.195
 moral+0.185
 floor+0.181
 fore+0.180
Neg logit contributions
ised-0.331
eday-0.272
rous-0.271
 that-0.267
und-0.257
Token They
Token ID1119
Feature activation-2.15e-01

Pos logit contributions
ised+0.587
rous+0.493
eday+0.492
 that+0.471
und+0.462
Neg logit contributions
 eyel-0.320
cel-0.298
 moral-0.278
 floor-0.275
 fore-0.269
Token both
Token ID1111
Feature activation-1.05e-02

Pos logit contributions
ised+0.396
eday+0.325
rous+0.325
 that+0.313
und+0.305
Neg logit contributions
 eyel-0.270
cel-0.253
 moral-0.240
 fore-0.236
 floor-0.235
Token sat
Token ID3332
Feature activation-1.26e+00

Pos logit contributions
ised+0.019
eday+0.016
rous+0.016
 that+0.015
und+0.015
Neg logit contributions
 eyel-0.013
cel-0.012
 moral-0.012
 floor-0.012
 fore-0.012
Token down
Token ID866
Feature activation+1.66e-01

Pos logit contributions
ised+2.172
eday+1.832
rous+1.809
uable+1.745
 that+1.716
Neg logit contributions
 eyel-1.582
cel-1.563
 moral-1.469
 floor-1.463
 fore-1.436
Token on
Token ID319
Feature activation+4.84e-02

Pos logit contributions
 eyel+0.224
cel+0.211
 moral+0.201
 floor+0.197
 fore+0.196
Neg logit contributions
ised-0.291
eday-0.235
rous-0.233
 that-0.230
und-0.221
Token the
Token ID262
Feature activation+3.14e+00

Pos logit contributions
 eyel+0.066
cel+0.062
 moral+0.059
 floor+0.058
 fore+0.058
Neg logit contributions
ised-0.084
eday-0.068
rous-0.068
 that-0.066
und-0.064
Token couch
Token ID18507
Feature activation-5.78e-01

Pos logit contributions
 eyel+3.393
quin+3.229
 Si+3.144
 moral+3.070
Beck+3.066
Neg logit contributions
ised-6.314
 that-6.104
eday-5.138
rous-5.035
izing-4.873
Token and
Token ID290
Feature activation+5.73e-01

Pos logit contributions
ised+1.128
rous+0.950
eday+0.937
amine+0.884
 that+0.882
Neg logit contributions
 eyel-0.662
cel-0.604
 floor-0.594
 moral-0.566
 fore-0.562
Token watched
Token ID7342
Feature activation-4.65e-01

Pos logit contributions
 eyel+0.831
cel+0.783
 moral+0.763
 fore+0.739
 floor+0.734
Neg logit contributions
ised-0.937
eday-0.752
 that-0.749
rous-0.740
und-0.697
Token a
Token ID257
Feature activation-1.30e-01

Pos logit contributions
ised+0.870
eday+0.722
rous+0.720
und+0.682
Deal+0.673
Neg logit contributions
 eyel-0.581
cel-0.527
 moral-0.510
 floor-0.508
 fore-0.494
Token show
Token ID905
Feature activation-5.86e-01

Pos logit contributions
ised+0.257
eday+0.215
rous+0.213
 that+0.207
und+0.203
Neg logit contributions
 eyel-0.146
cel-0.135
 moral-0.130
 floor-0.127
 fore-0.125
Token cookies
Token ID14746
Feature activation+5.87e-01

Pos logit contributions
 eyel+1.275
cel+1.180
 fore+1.099
nob+1.093
 moral+1.088
Neg logit contributions
ised-1.648
 that-1.366
eday-1.324
rous-1.319
und-1.241
Token.
Token ID13
Feature activation-9.96e-01

Pos logit contributions
 eyel+0.691
cel+0.650
 moral+0.601
 fore+0.581
quin+0.574
Neg logit contributions
ised-1.129
 that-0.951
eday-0.933
rous-0.921
und-0.880
Token They
Token ID1119
Feature activation+2.01e-01

Pos logit contributions
ised+1.931
eday+1.602
rous+1.596
und+1.500
amine+1.486
Neg logit contributions
 eyel-1.210
 floor-1.098
 moral-1.082
cel-1.068
 fore-1.013
Token sat
Token ID3332
Feature activation-6.72e-01

Pos logit contributions
 eyel+0.273
cel+0.257
 moral+0.246
 floor+0.240
 fore+0.240
Neg logit contributions
ised-0.350
eday-0.283
rous-0.281
 that-0.277
und-0.265
Token on
Token ID319
Feature activation-8.40e-02

Pos logit contributions
ised+1.287
eday+1.078
rous+1.071
und+1.019
 that+1.015
Neg logit contributions
 eyel-0.757
cel-0.719
 moral-0.709
 floor-0.695
 fore-0.676
Token the
Token ID262
Feature activation+3.10e+00

Pos logit contributions
ised+0.147
eday+0.119
rous+0.118
 that+0.114
und+0.112
Neg logit contributions
 eyel-0.113
cel-0.106
 moral-0.102
 floor-0.101
 fore-0.100
Token floor
Token ID4314
Feature activation+7.64e-02

Pos logit contributions
 eyel+2.985
 Si+2.941
cel+2.824
quin+2.788
Beck+2.676
Neg logit contributions
 that-6.283
ised-6.197
arl-4.977
izing-4.939
aw-4.938
Token and
Token ID290
Feature activation+1.05e+00

Pos logit contributions
 eyel+0.090
cel+0.084
 moral+0.079
 floor+0.078
 fore+0.077
Neg logit contributions
ised-0.146
eday-0.121
rous-0.120
 that-0.117
und-0.114
Token ate
Token ID15063
Feature activation-5.97e-01

Pos logit contributions
 eyel+1.553
cel+1.477
 moral+1.462
 fore+1.402
nob+1.368
Neg logit contributions
ised-1.664
 that-1.375
eday-1.343
rous-1.303
aw-1.234
Token them
Token ID606
Feature activation+1.01e+00

Pos logit contributions
ised+1.129
eday+0.941
rous+0.935
und+0.894
amine+0.869
Neg logit contributions
 eyel-0.718
 moral-0.654
 floor-0.650
cel-0.649
 fore-0.623
Token.
Token ID13
Feature activation-5.05e-01

Pos logit contributions
 eyel+1.403
cel+1.370
 moral+1.256
 fore+1.248
quin+1.245
Neg logit contributions
ised-1.737
 that-1.461
eday-1.386
rous-1.368
aw-1.313
Token living
Token ID2877
Feature activation-8.02e-01

Pos logit contributions
 eyel+0.165
cel+0.153
 moral+0.144
 floor+0.141
 fore+0.140
Neg logit contributions
ised-0.301
eday-0.251
rous-0.250
 that-0.246
und-0.238
Token room
Token ID2119
Feature activation-1.11e-01

Pos logit contributions
ised+1.757
eday+1.519
rous+1.486
 that+1.459
uable+1.442
Neg logit contributions
 eyel-0.678
cel-0.643
 floor-0.638
 hallway-0.592
 fore-0.591
Token and
Token ID290
Feature activation+6.26e-01

Pos logit contributions
ised+0.212
eday+0.175
rous+0.175
 that+0.168
und+0.166
Neg logit contributions
 eyel-0.130
cel-0.121
 moral-0.115
 floor-0.114
 fore-0.113
Token sat
Token ID3332
Feature activation-7.63e-01

Pos logit contributions
 eyel+0.893
cel+0.849
 moral+0.821
 fore+0.791
 floor+0.788
Neg logit contributions
ised-1.031
eday-0.836
 that-0.836
rous-0.827
und-0.772
Token on
Token ID319
Feature activation-1.29e-01

Pos logit contributions
ised+1.413
eday+1.176
rous+1.165
und+1.112
 that+1.104
Neg logit contributions
 eyel-0.922
cel-0.882
 moral-0.840
 floor-0.838
 fore-0.811
Token the
Token ID262
Feature activation+3.06e+00

Pos logit contributions
ised+0.227
eday+0.185
rous+0.183
 that+0.176
und+0.173
Neg logit contributions
 eyel-0.172
cel-0.162
 moral-0.155
 floor-0.153
 fore-0.151
Token couch
Token ID18507
Feature activation-3.45e-01

Pos logit contributions
 eyel+3.198
quin+2.882
 moral+2.882
 Si+2.871
Beck+2.808
Neg logit contributions
ised-6.009
 that-5.950
rous-4.849
eday-4.802
izing-4.776
Token.
Token ID13
Feature activation-5.94e-01

Pos logit contributions
ised+0.738
rous+0.629
eday+0.626
 that+0.596
und+0.593
Neg logit contributions
 eyel-0.334
cel-0.300
 floor-0.290
 moral-0.278
 fore-0.272
Token Lily
Token ID20037
Feature activation+3.44e-01

Pos logit contributions
ised+1.075
eday+0.881
rous+0.866
 that+0.824
und+0.819
Neg logit contributions
 eyel-0.795
cel-0.732
 moral-0.708
 floor-0.700
 fore-0.678
Token turned
Token ID2900
Feature activation-2.35e-01

Pos logit contributions
 eyel+0.474
cel+0.444
 moral+0.429
 fore+0.417
 floor+0.415
Neg logit contributions
ised-0.588
eday-0.474
rous-0.469
 that-0.467
und-0.444
Token on
Token ID319
Feature activation-1.10e+00

Pos logit contributions
ised+0.416
eday+0.343
rous+0.337
 that+0.321
und+0.319
Neg logit contributions
 eyel-0.311
cel-0.290
 moral-0.280
 floor-0.277
 fore-0.276
Token was
Token ID373
Feature activation+2.97e-01

Pos logit contributions
 eyel+0.361
cel+0.333
 moral+0.312
 floor+0.303
 fore+0.302
Neg logit contributions
ised-0.749
eday-0.630
 that-0.626
rous-0.624
und-0.598
Token her
Token ID607
Feature activation-5.24e-01

Pos logit contributions
 eyel+0.409
cel+0.389
 moral+0.368
 floor+0.361
 fore+0.360
Neg logit contributions
ised-0.510
eday-0.412
rous-0.409
 that-0.407
und-0.384
Token grandma
Token ID49890
Feature activation-4.33e-02

Pos logit contributions
ised+0.979
eday+0.798
rous+0.785
und+0.772
aw+0.751
Neg logit contributions
 eyel-0.679
 moral-0.612
cel-0.607
 floor-0.588
 fore-0.582
Token on
Token ID319
Feature activation-5.25e-01

Pos logit contributions
ised+0.089
eday+0.075
rous+0.074
 that+0.072
und+0.071
Neg logit contributions
 eyel-0.046
cel-0.042
 moral-0.039
 floor-0.038
 fore-0.038
Token the
Token ID262
Feature activation+2.18e+00

Pos logit contributions
ised+0.955
eday+0.780
rous+0.772
Deal+0.734
 that+0.732
Neg logit contributions
 eyel-0.672
cel-0.619
 moral-0.611
 floor-0.607
 fore-0.597
Token other
Token ID584
Feature activation+3.05e+00

Pos logit contributions
 eyel+2.374
quin+2.319
cel+2.259
 cor+2.182
Beck+2.142
Neg logit contributions
ised-4.159
 that-3.851
eday-3.489
rous-3.444
Deal-3.266
Token side
Token ID1735
Feature activation-2.97e-01

Pos logit contributions
 cor+4.072
 eyel+4.023
quin+3.881
Beck+3.861
 moral+3.798
Neg logit contributions
ised-5.328
 that-4.967
eday-4.311
aw-4.081
rous-4.006
Token.
Token ID13
Feature activation+6.86e-02

Pos logit contributions
ised+0.545
eday+0.444
rous+0.442
 that+0.420
und+0.417
Neg logit contributions
 eyel-0.374
cel-0.345
 floor-0.334
 moral-0.331
 fore-0.327
Token Grand
Token ID5675
Feature activation-1.82e-01

Pos logit contributions
 eyel+0.101
cel+0.095
 moral+0.091
 floor+0.090
 fore+0.089
Neg logit contributions
ised-0.112
eday-0.089
rous-0.088
 that-0.086
und-0.083
Tokenma
Token ID2611
Feature activation-1.73e-01

Pos logit contributions
ised+0.381
eday+0.318
rous+0.318
 that+0.311
und+0.304
Neg logit contributions
 eyel-0.182
cel-0.169
 moral-0.159
 fore-0.155
 floor-0.153
Token said
Token ID531
Feature activation-6.18e-01

Pos logit contributions
ised+0.314
eday+0.256
rous+0.256
 that+0.244
und+0.241
Neg logit contributions
 eyel-0.222
cel-0.210
 moral-0.200
 fore-0.195
 floor-0.195
Token.
Token ID13
Feature activation-7.59e-01

Pos logit contributions
ised+0.708
rous+0.601
eday+0.597
und+0.567
 that+0.565
Neg logit contributions
 eyel-0.324
cel-0.290
 floor-0.281
 moral-0.276
 fore-0.269
Token He
Token ID679
Feature activation+1.90e-01

Pos logit contributions
ised+1.514
eday+1.260
rous+1.252
 that+1.170
und+1.166
Neg logit contributions
 eyel-0.879
cel-0.778
 moral-0.777
 floor-0.774
 fore-0.737
Token climbed
Token ID19952
Feature activation+9.86e-01

Pos logit contributions
 eyel+0.259
cel+0.244
 moral+0.233
 floor+0.228
 fore+0.227
Neg logit contributions
ised-0.329
eday-0.265
rous-0.263
 that-0.260
und-0.249
Token on
Token ID319
Feature activation+3.65e-01

Pos logit contributions
 eyel+1.432
cel+1.373
 moral+1.287
 fore+1.250
 floor+1.243
Neg logit contributions
ised-1.628
 that-1.361
eday-1.307
rous-1.274
aw-1.227
Token the
Token ID262
Feature activation+1.78e+00

Pos logit contributions
 eyel+0.519
cel+0.493
 moral+0.468
 fore+0.459
 floor+0.457
Neg logit contributions
ised-0.610
eday-0.489
 that-0.485
rous-0.484
und-0.457
Token other
Token ID584
Feature activation+3.04e+00

Pos logit contributions
 eyel+2.529
cel+2.262
quin+2.198
Beck+2.177
 moral+2.165
Neg logit contributions
ised-3.007
 that-2.702
aw-2.354
eday-2.318
rous-2.311
Token end
Token ID886
Feature activation+8.97e-01

Pos logit contributions
 eyel+4.471
Beck+4.165
cel+4.128
quin+4.025
 moral+3.850
Neg logit contributions
ised-5.177
 that-4.779
aw-4.361
eday-4.113
rous-3.949
Token and
Token ID290
Feature activation+5.81e-01

Pos logit contributions
 eyel+1.411
cel+1.381
 moral+1.274
 fore+1.245
quin+1.233
Neg logit contributions
ised-1.365
 that-1.118
rous-1.042
eday-1.038
aw-1.008
Token tried
Token ID3088
Feature activation+4.04e-01

Pos logit contributions
 eyel+0.820
cel+0.769
 moral+0.745
 fore+0.716
 floor+0.714
Neg logit contributions
ised-0.984
 that-0.794
eday-0.792
rous-0.782
aw-0.743
Token to
Token ID284
Feature activation-1.10e-01

Pos logit contributions
 eyel+0.729
cel+0.702
 moral+0.674
 fore+0.664
 floor+0.663
Neg logit contributions
ised-0.518
 that-0.386
eday-0.383
rous-0.382
und-0.350
Token balance
Token ID5236
Feature activation+1.03e+00

Pos logit contributions
ised+0.196
rous+0.159
eday+0.159
 that+0.154
und+0.150
Neg logit contributions
 eyel-0.145
cel-0.136
 moral-0.130
 floor-0.129
 fore-0.127
Token saw
Token ID2497
Feature activation-1.19e+00

Pos logit contributions
 eyel+0.116
cel+0.110
 moral+0.105
 floor+0.104
 fore+0.103
Neg logit contributions
ised-0.110
eday-0.086
rous-0.085
 that-0.083
und-0.080
Token their
Token ID511
Feature activation-8.38e-01

Pos logit contributions
ised+2.334
eday+1.994
rous+1.928
und+1.905
Deal+1.881
Neg logit contributions
 eyel-1.327
 moral-1.242
 other-1.211
 floor-1.207
 fore-1.191
Token ball
Token ID2613
Feature activation+1.49e-01

Pos logit contributions
ised+1.650
eday+1.403
rous+1.351
und+1.337
amine+1.319
Neg logit contributions
 eyel-0.961
 floor-0.870
 moral-0.868
 fore-0.826
cel-0.817
Token on
Token ID319
Feature activation-2.45e-01

Pos logit contributions
 eyel+0.166
cel+0.155
 moral+0.146
 floor+0.142
 fore+0.142
Neg logit contributions
ised-0.296
eday-0.246
rous-0.244
 that-0.241
und-0.233
Token the
Token ID262
Feature activation+2.12e+00

Pos logit contributions
ised+0.432
eday+0.353
rous+0.352
 that+0.334
und+0.331
Neg logit contributions
 eyel-0.325
cel-0.302
 moral-0.296
 floor-0.294
 fore-0.287
Token other
Token ID584
Feature activation+3.00e+00

Pos logit contributions
 eyel+2.322
cel+2.180
quin+2.133
 Si+2.009
 moral+1.988
Neg logit contributions
ised-4.149
 that-3.857
eday-3.367
rous-3.344
aw-3.275
Token side
Token ID1735
Feature activation+4.72e-02

Pos logit contributions
 eyel+4.234
cel+4.170
quin+4.159
Beck+4.037
 moral+3.910
Neg logit contributions
ised-5.192
 that-4.731
aw-4.038
eday-3.941
rous-3.834
Token of
Token ID286
Feature activation-1.04e-01

Pos logit contributions
 eyel+0.053
cel+0.049
 moral+0.046
 floor+0.045
 fore+0.045
Neg logit contributions
ised-0.093
eday-0.078
rous-0.077
 that-0.075
und-0.073
Token the
Token ID262
Feature activation+4.46e-01

Pos logit contributions
ised+0.182
eday+0.147
rous+0.146
 that+0.142
und+0.138
Neg logit contributions
 eyel-0.140
cel-0.131
 moral-0.127
 floor-0.126
 fore-0.125
Token yard
Token ID12699
Feature activation+1.66e-02

Pos logit contributions
 eyel+0.560
cel+0.525
 moral+0.494
 fore+0.481
quin+0.474
Neg logit contributions
ised-0.829
rous-0.678
 that-0.677
eday-0.677
und-0.642
Token.
Token ID13
Feature activation-1.25e+00

Pos logit contributions
 eyel+0.017
cel+0.015
 moral+0.014
 floor+0.014
 fore+0.014
Neg logit contributions
ised-0.035
eday-0.029
rous-0.029
 that-0.028
und-0.028
Token a
Token ID257
Feature activation-3.54e-01

Pos logit contributions
ised+0.209
eday+0.172
rous+0.169
 that+0.163
und+0.161
Neg logit contributions
 eyel-0.150
cel-0.139
 moral-0.135
 floor-0.132
 fore-0.132
Token shiny
Token ID22441
Feature activation-5.22e-01

Pos logit contributions
ised+0.651
eday+0.545
rous+0.527
und+0.508
 that+0.506
Neg logit contributions
 eyel-0.443
cel-0.417
 moral-0.408
 floor-0.398
 fore-0.390
Token box
Token ID3091
Feature activation-2.88e-01

Pos logit contributions
ised+0.952
eday+0.790
rous+0.779
und+0.742
 that+0.737
Neg logit contributions
 eyel-0.654
cel-0.602
 moral-0.595
 floor-0.593
 fore-0.569
Token on
Token ID319
Feature activation-8.70e-02

Pos logit contributions
ised+0.516
eday+0.420
rous+0.418
und+0.393
 that+0.392
Neg logit contributions
 eyel-0.379
cel-0.353
 moral-0.341
 floor-0.338
 fore-0.332
Token the
Token ID262
Feature activation+2.22e+00

Pos logit contributions
ised+0.151
eday+0.123
rous+0.122
 that+0.117
und+0.115
Neg logit contributions
 eyel-0.118
cel-0.110
 moral-0.107
 floor-0.105
 fore-0.104
Token other
Token ID584
Feature activation+3.00e+00

Pos logit contributions
 eyel+3.003
cel+2.814
quin+2.672
 Si+2.533
Beck+2.512
Neg logit contributions
ised-3.895
 that-3.556
eday-3.210
rous-3.106
aw-2.928
Token side
Token ID1735
Feature activation+4.62e-01

Pos logit contributions
 eyel+4.549
cel+4.308
Beck+4.218
quin+4.097
 magn+3.906
Neg logit contributions
ised-5.266
 that-4.731
eday-4.200
aw-4.055
rous-3.787
Token of
Token ID286
Feature activation-8.78e-02

Pos logit contributions
 eyel+0.679
cel+0.650
 moral+0.613
 fore+0.599
 floor+0.592
Neg logit contributions
ised-0.753
eday-0.601
 that-0.600
rous-0.596
und-0.562
Token the
Token ID262
Feature activation+8.31e-01

Pos logit contributions
ised+0.154
eday+0.125
rous+0.123
 that+0.120
und+0.117
Neg logit contributions
 eyel-0.119
cel-0.111
 moral-0.108
 floor-0.107
 fore-0.105
Token river
Token ID7850
Feature activation+5.17e-01

Pos logit contributions
 eyel+1.233
cel+1.161
 moral+1.070
 fore+1.060
quin+1.054
Neg logit contributions
ised-1.379
 that-1.109
rous-1.098
eday-1.087
und-1.017
Token.
Token ID13
Feature activation-1.21e+00

Pos logit contributions
 eyel+0.566
cel+0.532
 moral+0.495
 fore+0.478
 floor+0.465
Neg logit contributions
ised-1.035
 that-0.866
eday-0.861
rous-0.853
und-0.821
Token.
Token ID13
Feature activation-8.32e-01

Pos logit contributions
ised+0.255
eday+0.213
rous+0.211
 that+0.206
und+0.202
Neg logit contributions
 eyel-0.131
cel-0.121
 moral-0.114
 floor-0.112
 fore-0.111
Token They
Token ID1119
Feature activation-1.52e-01

Pos logit contributions
ised+1.607
rous+1.301
eday+1.300
und+1.222
 that+1.220
Neg logit contributions
 eyel-1.018
cel-0.921
 moral-0.911
 floor-0.886
 fore-0.870
Token all
Token ID477
Feature activation-8.03e-01

Pos logit contributions
ised+0.271
eday+0.218
rous+0.217
 that+0.210
und+0.205
Neg logit contributions
 eyel-0.201
cel-0.189
 moral-0.181
 fore-0.177
 floor-0.177
Token sit
Token ID1650
Feature activation-2.95e-01

Pos logit contributions
ised+1.568
eday+1.261
rous+1.251
und+1.202
 that+1.193
Neg logit contributions
 eyel-0.938
 moral-0.857
cel-0.845
 floor-0.839
 fore-0.838
Token on
Token ID319
Feature activation+2.72e-01

Pos logit contributions
ised+0.498
eday+0.404
rous+0.398
 that+0.390
und+0.376
Neg logit contributions
 eyel-0.410
cel-0.387
 moral-0.373
 fore-0.364
 floor-0.362
Token the
Token ID262
Feature activation+2.98e+00

Pos logit contributions
 eyel+0.376
cel+0.356
 moral+0.338
 floor+0.332
 fore+0.331
Neg logit contributions
ised-0.466
eday-0.376
rous-0.373
 that-0.370
und-0.351
Token bench
Token ID7624
Feature activation-1.25e-01

Pos logit contributions
 eyel+3.942
cel+3.884
Beck+3.782
quin+3.661
 moral+3.578
Neg logit contributions
ised-4.870
 that-4.663
aw-4.029
rous-3.919
amine-3.821
Token and
Token ID290
Feature activation+7.27e-01

Pos logit contributions
ised+0.246
rous+0.204
eday+0.204
 that+0.196
und+0.193
Neg logit contributions
 eyel-0.140
cel-0.129
 moral-0.123
 floor-0.122
 fore-0.120
Token smile
Token ID8212
Feature activation-1.83e-01

Pos logit contributions
 eyel+1.060
cel+1.017
 moral+0.972
 fore+0.953
 floor+0.951
Neg logit contributions
ised-1.154
eday-0.947
 that-0.938
rous-0.921
und-0.857
Token.
Token ID13
Feature activation-4.20e-01

Pos logit contributions
ised+0.379
eday+0.320
rous+0.318
 that+0.306
und+0.302
Neg logit contributions
 eyel-0.189
cel-0.173
 moral-0.165
 floor-0.159
 fore-0.158
Token The
Token ID383
Feature activation+2.78e-01

Pos logit contributions
ised+0.772
eday+0.629
rous+0.623
 that+0.595
und+0.586
Neg logit contributions
 eyel-0.542
cel-0.500
 moral-0.486
 floor-0.473
 fore-0.468
Token apr
Token ID46593
Feature activation-9.04e-02

Pos logit contributions
 eyel+1.891
cel+1.843
 moral+1.779
nob+1.720
quin+1.708
Neg logit contributions
ised-2.504
rous-2.044
 that-2.033
eday-1.958
amine-1.834
Tokenons
Token ID684
Feature activation+2.81e-01

Pos logit contributions
ised+0.173
eday+0.144
rous+0.142
 that+0.140
und+0.135
Neg logit contributions
 eyel-0.105
cel-0.099
 moral-0.093
 floor-0.092
 fore-0.091
Token and
Token ID290
Feature activation+1.40e+00

Pos logit contributions
 eyel+0.364
cel+0.341
 moral+0.327
 fore+0.317
 floor+0.316
Neg logit contributions
ised-0.506
eday-0.411
rous-0.408
 that-0.406
und-0.387
Token sat
Token ID3332
Feature activation-1.08e+00

Pos logit contributions
 eyel+1.998
cel+1.941
 moral+1.885
 fore+1.815
nob+1.789
Neg logit contributions
ised-2.286
 that-1.868
eday-1.792
rous-1.777
aw-1.683
Token on
Token ID319
Feature activation+1.26e-01

Pos logit contributions
ised+2.086
eday+1.761
rous+1.752
und+1.677
Fine+1.669
Neg logit contributions
 eyel-1.180
cel-1.143
 floor-1.085
 moral-1.076
 fore-1.049
Token the
Token ID262
Feature activation+2.97e+00

Pos logit contributions
 eyel+0.169
cel+0.159
 moral+0.152
 floor+0.149
 fore+0.149
Neg logit contributions
ised-0.221
eday-0.179
rous-0.178
 that-0.175
und-0.168
Token floor
Token ID4314
Feature activation+7.75e-02

Pos logit contributions
quin+3.145
 eyel+3.103
cel+3.029
Beck+2.923
 moral+2.876
Neg logit contributions
ised-5.658
 that-5.588
rous-4.751
eday-4.645
aw-4.505
Token.
Token ID13
Feature activation+3.97e-01

Pos logit contributions
 eyel+0.082
cel+0.076
 moral+0.071
 floor+0.070
 fore+0.069
Neg logit contributions
ised-0.157
eday-0.132
rous-0.131
 that-0.128
und-0.125
Token
Token ID198
Feature activation+8.20e-02

Pos logit contributions
 eyel+0.560
cel+0.526
 moral+0.503
 fore+0.498
 floor+0.492
Neg logit contributions
ised-0.672
eday-0.536
rous-0.532
 that-0.531
und-0.503
Token
Token ID198
Feature activation+3.41e-01

Pos logit contributions
 eyel+0.104
cel+0.097
 moral+0.093
 floor+0.091
 fore+0.091
Neg logit contributions
ised-0.149
eday-0.122
rous-0.121
 that-0.119
und-0.114
Token"
Token ID1
Feature activation+1.20e+00

Pos logit contributions
 eyel+0.524
cel+0.494
 moral+0.472
 floor+0.466
 fore+0.466
Neg logit contributions
ised-0.538
eday-0.421
rous-0.417
 that-0.412
und-0.392
Token picks
Token ID11103
Feature activation-2.76e-02

Pos logit contributions
 eyel+1.469
cel+1.407
 fore+1.320
nob+1.310
 moral+1.307
Neg logit contributions
ised-1.550
 that-1.287
eday-1.279
rous-1.250
aw-1.201
Token one
Token ID530
Feature activation+2.93e-01

Pos logit contributions
ised+0.050
eday+0.040
rous+0.040
 that+0.039
und+0.038
Neg logit contributions
 eyel-0.036
cel-0.034
 moral-0.032
 floor-0.032
 fore-0.031
Token and
Token ID290
Feature activation+4.42e-01

Pos logit contributions
 eyel+0.381
cel+0.359
 moral+0.338
 fore+0.333
 floor+0.331
Neg logit contributions
ised-0.526
eday-0.431
rous-0.426
 that-0.425
und-0.402
Token sits
Token ID10718
Feature activation-1.70e-01

Pos logit contributions
 eyel+0.600
cel+0.569
 moral+0.538
 fore+0.532
 floor+0.527
Neg logit contributions
ised-0.764
eday-0.622
 that-0.617
rous-0.615
aw-0.578
Token on
Token ID319
Feature activation+2.08e-01

Pos logit contributions
ised+0.316
eday+0.258
rous+0.257
 that+0.249
und+0.245
Neg logit contributions
 eyel-0.212
cel-0.201
 moral-0.192
 floor-0.188
 fore-0.186
Token the
Token ID262
Feature activation+2.97e+00

Pos logit contributions
 eyel+0.296
cel+0.280
 moral+0.268
 fore+0.263
 floor+0.262
Neg logit contributions
ised-0.345
eday-0.277
rous-0.274
 that-0.271
und-0.258
Token floor
Token ID4314
Feature activation+2.58e-02

Pos logit contributions
 eyel+2.623
cel+2.615
Beck+2.604
 Si+2.529
quin+2.510
Neg logit contributions
 that-5.994
ised-5.933
eday-5.074
aw-4.985
Deal-4.891
Token.
Token ID13
Feature activation-5.43e-01

Pos logit contributions
 eyel+0.026
cel+0.024
 moral+0.023
 floor+0.022
 fore+0.022
Neg logit contributions
ised-0.053
eday-0.045
rous-0.045
 that-0.044
und-0.043
Token
Token ID198
Feature activation+1.60e-02

Pos logit contributions
ised+0.938
eday+0.747
rous+0.742
 that+0.693
und+0.687
Neg logit contributions
 eyel-0.781
cel-0.722
 moral-0.699
 floor-0.688
 magn-0.667
Token
Token ID198
Feature activation+1.38e-01

Pos logit contributions
 eyel+0.020
cel+0.019
 moral+0.018
 floor+0.018
 fore+0.017
Neg logit contributions
ised-0.029
eday-0.024
rous-0.024
 that-0.023
und-0.023
Token"
Token ID1
Feature activation+5.02e-01

Pos logit contributions
 eyel+0.200
cel+0.189
 moral+0.181
 floor+0.178
 fore+0.177
Neg logit contributions
ised-0.227
eday-0.181
rous-0.180
 that-0.176
und-0.169
Token saw
Token ID2497
Feature activation-8.48e-01

Pos logit contributions
ised+0.666
eday+0.546
rous+0.538
 that+0.506
und+0.504
Neg logit contributions
 eyel-0.542
cel-0.519
 moral-0.497
 fore-0.493
 floor-0.487
Token the
Token ID262
Feature activation+1.40e+00

Pos logit contributions
ised+1.516
eday+1.249
rous+1.223
Deal+1.180
amine+1.159
Neg logit contributions
 eyel-1.094
cel-1.040
 floor-1.026
 moral-1.026
 fore-0.994
Token blocks
Token ID7021
Feature activation+9.20e-02

Pos logit contributions
 eyel+1.681
cel+1.574
Beck+1.420
nob+1.418
quin+1.387
Neg logit contributions
ised-2.689
 that-2.319
rous-2.184
eday-2.178
aw-2.074
Token scattered
Token ID16830
Feature activation-4.01e-01

Pos logit contributions
 eyel+0.111
cel+0.103
 moral+0.098
 floor+0.096
 fore+0.096
Neg logit contributions
ised-0.174
eday-0.143
rous-0.142
 that-0.139
und-0.135
Token on
Token ID319
Feature activation-2.47e-01

Pos logit contributions
ised+0.732
eday+0.613
rous+0.606
 that+0.573
und+0.563
Neg logit contributions
 eyel-0.498
cel-0.457
 floor-0.453
 fore-0.446
 moral-0.445
Token the
Token ID262
Feature activation+2.95e+00

Pos logit contributions
ised+0.435
eday+0.357
rous+0.355
 that+0.337
und+0.331
Neg logit contributions
 eyel-0.327
cel-0.304
 floor-0.296
 moral-0.295
 fore-0.288
Token floor
Token ID4314
Feature activation+1.74e-01

Pos logit contributions
 eyel+2.077
 Si+1.911
Beck+1.902
cel+1.873
quin+1.850
Neg logit contributions
ised-6.893
 that-6.587
eday-5.826
und-5.612
aw-5.588
Token and
Token ID290
Feature activation+1.04e+00

Pos logit contributions
 eyel+0.191
cel+0.177
 moral+0.167
 floor+0.162
 fore+0.162
Neg logit contributions
ised-0.346
eday-0.289
rous-0.286
 that-0.283
und-0.273
Token the
Token ID262
Feature activation+1.83e+00

Pos logit contributions
 eyel+1.558
cel+1.474
 moral+1.420
 fore+1.350
nob+1.347
Neg logit contributions
ised-1.678
 that-1.404
eday-1.312
rous-1.296
und-1.241
Token children
Token ID1751
Feature activation-5.15e-02

Pos logit contributions
 eyel+1.844
cel+1.756
Beck+1.580
 Si+1.562
 moral+1.547
Neg logit contributions
ised-3.739
 that-3.269
eday-3.045
rous-2.993
aw-2.969
Token yelling
Token ID22187
Feature activation+8.07e-01

Pos logit contributions
ised+0.104
eday+0.088
rous+0.087
 that+0.084
und+0.082
Neg logit contributions
 eyel-0.055
cel-0.051
 moral-0.048
 floor-0.048
 fore-0.047
Token agree
Token ID4236
Feature activation-8.64e-02

Pos logit contributions
ised+0.562
rous+0.455
eday+0.452
 that+0.436
und+0.431
Neg logit contributions
 eyel-0.396
cel-0.369
 moral-0.355
 floor-0.346
 fore-0.345
Token.
Token ID13
Feature activation-6.16e-01

Pos logit contributions
ised+0.179
eday+0.150
rous+0.149
 that+0.144
und+0.142
Neg logit contributions
 eyel-0.090
cel-0.083
 moral-0.079
 floor-0.077
 fore-0.076
Token They
Token ID1119
Feature activation-1.44e-01

Pos logit contributions
ised+1.148
rous+0.951
eday+0.945
 that+0.894
und+0.889
Neg logit contributions
 eyel-0.786
cel-0.720
 moral-0.711
 floor-0.703
 fore-0.670
Token sit
Token ID1650
Feature activation-6.67e-01

Pos logit contributions
ised+0.260
eday+0.209
rous+0.209
 that+0.202
und+0.198
Neg logit contributions
 eyel-0.188
cel-0.176
 moral-0.170
 floor-0.165
 fore-0.165
Token on
Token ID319
Feature activation+4.31e-01

Pos logit contributions
ised+1.259
eday+1.051
rous+1.043
 that+1.006
und+0.999
Neg logit contributions
 eyel-0.780
cel-0.743
 moral-0.720
 fore-0.691
 floor-0.686
Token the
Token ID262
Feature activation+2.95e+00

Pos logit contributions
 eyel+0.608
cel+0.577
 moral+0.545
 fore+0.533
 floor+0.531
Neg logit contributions
ised-0.731
eday-0.587
 that-0.584
rous-0.582
und-0.546
Token couch
Token ID18507
Feature activation-4.94e-01

Pos logit contributions
 eyel+2.896
quin+2.769
cel+2.757
Beck+2.722
 Si+2.690
Neg logit contributions
ised-5.826
 that-5.818
eday-4.907
rous-4.809
here-4.614
Token and
Token ID290
Feature activation+1.03e+00

Pos logit contributions
ised+0.992
rous+0.833
eday+0.822
und+0.783
 that+0.779
Neg logit contributions
 eyel-0.545
cel-0.492
 floor-0.488
 moral-0.477
 fore-0.464
Token hold
Token ID1745
Feature activation+5.12e-01

Pos logit contributions
 eyel+1.558
cel+1.509
 moral+1.442
 fore+1.420
 floor+1.396
Neg logit contributions
ised-1.543
eday-1.271
 that-1.271
rous-1.225
Deal-1.120
Token hands
Token ID2832
Feature activation+9.26e-01

Pos logit contributions
 eyel+0.668
cel+0.642
 moral+0.601
 floor+0.591
 fore+0.581
Neg logit contributions
ised-0.919
eday-0.747
 that-0.746
rous-0.746
aw-0.703
Token.
Token ID13
Feature activation-4.84e-01

Pos logit contributions
 eyel+1.085
cel+1.058
 moral+0.985
 fore+0.953
 floor+0.934
Neg logit contributions
ised-1.775
 that-1.509
eday-1.436
rous-1.423
aw-1.386
Token.
Token ID13
Feature activation+3.43e-02

Pos logit contributions
ised+0.586
eday+0.479
rous+0.469
 that+0.440
und+0.437
Neg logit contributions
 eyel-0.523
cel-0.501
 floor-0.473
 moral-0.473
nob-0.460
Token He
Token ID679
Feature activation+6.69e-01

Pos logit contributions
 eyel+0.042
cel+0.039
 moral+0.038
 floor+0.037
 fore+0.036
Neg logit contributions
ised-0.065
eday-0.053
rous-0.053
 that-0.052
und-0.050
Token played
Token ID2826
Feature activation-7.13e-01

Pos logit contributions
 eyel+0.827
cel+0.779
 moral+0.740
 fore+0.724
 floor+0.713
Neg logit contributions
ised-1.222
 that-1.011
eday-1.011
rous-0.998
und-0.952
Token with
Token ID351
Feature activation+1.55e-02

Pos logit contributions
ised+1.275
eday+1.068
rous+1.050
und+0.979
uable+0.961
Neg logit contributions
 eyel-0.919
cel-0.884
 floor-0.854
 moral-0.843
nob-0.816
Token cars
Token ID5006
Feature activation+1.45e-01

Pos logit contributions
 eyel+0.014
cel+0.012
 moral+0.011
 floor+0.011
 fore+0.011
Neg logit contributions
ised-0.035
eday-0.030
rous-0.029
 that-0.029
und-0.028
Token and
Token ID290
Feature activation+2.95e+00

Pos logit contributions
 eyel+0.179
cel+0.167
 moral+0.159
 floor+0.155
 fore+0.155
Neg logit contributions
ised-0.272
eday-0.224
rous-0.222
 that-0.219
und-0.211
Token balls
Token ID11333
Feature activation-1.87e-01

Pos logit contributions
 eyel+4.305
 fore+4.136
cel+4.059
nob+3.848
 moral+3.730
Neg logit contributions
ised-4.658
 that-4.349
eday-3.690
amine-3.471
Fine-3.447
Token and
Token ID290
Feature activation+2.47e+00

Pos logit contributions
ised+0.352
eday+0.291
rous+0.290
 that+0.274
und+0.272
Neg logit contributions
 eyel-0.228
cel-0.211
 moral-0.203
 floor-0.201
 fore-0.197
Token blocks
Token ID7021
Feature activation-6.43e-01

Pos logit contributions
 eyel+3.428
cel+3.167
 fore+3.161
nob+3.018
 moral+2.928
Neg logit contributions
ised-4.096
 that-3.825
eday-3.372
aw-3.201
Deal-3.170
Token,
Token ID11
Feature activation+1.20e+00

Pos logit contributions
ised+1.347
rous+1.145
eday+1.141
und+1.067
amine+1.062
Neg logit contributions
 eyel-0.645
cel-0.574
 floor-0.572
 moral-0.552
 fore-0.542
Token but
Token ID475
Feature activation+6.31e-01

Pos logit contributions
 eyel+1.520
cel+1.391
 fore+1.348
 moral+1.333
nob+1.286
Neg logit contributions
ised-2.192
 that-1.908
eday-1.825
rous-1.741
aw-1.717
Token about
Token ID546
Feature activation+1.47e-01

Pos logit contributions
 eyel+0.818
cel+0.777
 moral+0.728
 floor+0.713
 fore+0.713
Neg logit contributions
ised-0.962
 that-0.786
rous-0.762
eday-0.758
aw-0.733
Token a
Token ID257
Feature activation-8.72e-01

Pos logit contributions
 eyel+0.196
cel+0.184
 moral+0.176
 floor+0.172
 fore+0.172
Neg logit contributions
ised-0.260
eday-0.211
rous-0.209
 that-0.206
und-0.198
Token cave
Token ID11527
Feature activation-1.11e+00

Pos logit contributions
ised+1.690
eday+1.415
rous+1.343
und+1.320
aw+1.290
Neg logit contributions
 eyel-1.052
cel-0.991
 moral-0.984
 floor-0.954
 magn-0.950
Token on
Token ID319
Feature activation-1.85e-01

Pos logit contributions
ised+1.982
rous+1.614
eday+1.603
uable+1.512
Fine+1.510
Neg logit contributions
 eyel-1.430
 floor-1.352
cel-1.349
 moral-1.300
nob-1.297
Token the
Token ID262
Feature activation+2.24e+00

Pos logit contributions
ised+0.332
eday+0.269
rous+0.269
 that+0.256
und+0.254
Neg logit contributions
 eyel-0.241
cel-0.224
 moral-0.218
 floor-0.216
 fore-0.212
Token other
Token ID584
Feature activation+2.94e+00

Pos logit contributions
 eyel+3.055
quin+2.891
cel+2.837
 fore+2.672
 Si+2.611
Neg logit contributions
ised-3.807
 that-3.613
eday-3.167
rous-3.091
Deal-2.882
Token side
Token ID1735
Feature activation+1.10e-01

Pos logit contributions
 eyel+4.122
cel+4.005
quin+3.909
 fore+3.699
Beck+3.660
Neg logit contributions
ised-5.106
 that-4.698
eday-4.186
rous-3.870
aw-3.860
Token of
Token ID286
Feature activation+1.85e-01

Pos logit contributions
 eyel+0.157
cel+0.148
 moral+0.142
 floor+0.140
 fore+0.139
Neg logit contributions
ised-0.182
eday-0.146
rous-0.145
 that-0.142
und-0.136
Token the
Token ID262
Feature activation+1.34e+00

Pos logit contributions
 eyel+0.244
cel+0.229
 moral+0.218
 floor+0.213
 fore+0.213
Neg logit contributions
ised-0.331
eday-0.270
rous-0.268
 that-0.264
und-0.253
Token forest
Token ID8222
Feature activation-3.17e-02

Pos logit contributions
 eyel+1.903
cel+1.783
quin+1.660
 fore+1.641
 moral+1.610
Neg logit contributions
ised-2.292
 that-1.942
rous-1.865
eday-1.818
und-1.743
Token.
Token ID13
Feature activation-2.94e-01

Pos logit contributions
ised+0.066
eday+0.056
rous+0.055
 that+0.054
und+0.053
Neg logit contributions
 eyel-0.032
cel-0.029
 moral-0.028
 floor-0.027
 fore-0.027
Tokenhed
Token ID704
Feature activation-2.43e-01

Pos logit contributions
ised+0.762
rous+0.627
eday+0.619
 that+0.584
und+0.572
Neg logit contributions
 eyel-0.564
cel-0.535
 fore-0.507
 moral-0.504
 floor-0.497
Token the
Token ID262
Feature activation+3.44e-01

Pos logit contributions
ised+0.441
eday+0.362
rous+0.358
 that+0.344
und+0.338
Neg logit contributions
 eyel-0.314
cel-0.289
 moral-0.278
 floor-0.276
 fore-0.275
Token pain
Token ID2356
Feature activation+7.67e-02

Pos logit contributions
 eyel+0.467
cel+0.445
 moral+0.419
 fore+0.409
 floor+0.409
Neg logit contributions
ised-0.592
eday-0.479
rous-0.475
 that-0.470
und-0.446
Token.
Token ID13
Feature activation-4.76e-01

Pos logit contributions
 eyel+0.090
cel+0.083
 moral+0.079
 floor+0.078
 fore+0.077
Neg logit contributions
ised-0.147
eday-0.122
rous-0.121
 that-0.118
und-0.115
Token She
Token ID1375
Feature activation+1.65e-01

Pos logit contributions
ised+0.862
eday+0.708
rous+0.703
 that+0.674
und+0.653
Neg logit contributions
 eyel-0.622
cel-0.579
 moral-0.549
 floor-0.542
 fore-0.531
Token sn
Token ID3013
Feature activation-3.60e+00

Pos logit contributions
 eyel+0.209
cel+0.196
 moral+0.186
 floor+0.182
 fore+0.182
Neg logit contributions
ised-0.303
eday-0.248
rous-0.246
 that-0.243
und-0.233
Tokenugg
Token ID6837
Feature activation-8.24e-01

Pos logit contributions
ised+4.177
rous+3.597
izing+3.379
uable+3.250
eday+3.077
Neg logit contributions
 eyel-6.653
 fore-6.236
cel-6.172
 moral-5.845
nob-5.804
Tokenled
Token ID992
Feature activation-1.23e-01

Pos logit contributions
ised+1.357
rous+1.154
eday+1.147
 that+1.065
und+1.054
Neg logit contributions
 eyel-1.134
cel-1.084
 floor-1.053
 moral-1.025
 fore-1.019
Token in
Token ID287
Feature activation-1.20e-01

Pos logit contributions
ised+0.216
eday+0.176
rous+0.174
 that+0.169
und+0.165
Neg logit contributions
 eyel-0.164
cel-0.152
 moral-0.146
 floor-0.144
 fore-0.144
Token her
Token ID607
Feature activation+3.03e-02

Pos logit contributions
ised+0.237
eday+0.196
rous+0.195
 that+0.189
und+0.186
Neg logit contributions
 eyel-0.139
cel-0.127
 moral-0.122
 floor-0.120
 fore-0.120
Token mom
Token ID1995
Feature activation-2.34e-01

Pos logit contributions
 eyel+0.034
cel+0.031
 moral+0.029
 floor+0.029
 fore+0.029
Neg logit contributions
ised-0.060
eday-0.050
rous-0.050
 that-0.049
und-0.048
Token the
Token ID262
Feature activation-1.15e+00

Pos logit contributions
ised+0.032
eday+0.026
rous+0.026
 that+0.025
und+0.025
Neg logit contributions
 eyel-0.022
cel-0.021
 moral-0.020
 floor-0.020
 fore-0.020
Token knot
Token ID29654
Feature activation+2.62e-02

Pos logit contributions
ised+1.455
rous+1.140
eday+1.134
und+1.099
aw+1.011
Neg logit contributions
 eyel-2.108
 moral-2.049
 floor-2.008
cel-1.963
 other-1.949
Token was
Token ID373
Feature activation+1.40e-01

Pos logit contributions
 eyel+0.029
cel+0.027
 moral+0.026
 floor+0.025
 fore+0.025
Neg logit contributions
ised-0.052
eday-0.043
rous-0.043
 that-0.042
und-0.041
Token.
Token ID13
Feature activation-1.31e+00

Pos logit contributions
 eyel+0.174
cel+0.163
 moral+0.155
 floor+0.152
 fore+0.151
Neg logit contributions
ised-0.260
eday-0.213
rous-0.212
 that-0.209
und-0.201
Token He
Token ID679
Feature activation-5.10e-01

Pos logit contributions
ised+2.555
rous+2.218
eday+2.178
und+2.000
 that+1.968
Neg logit contributions
 eyel-1.529
cel-1.443
 floor-1.388
 moral-1.374
nob-1.329
Token sn
Token ID3013
Feature activation-3.40e+00

Pos logit contributions
ised+0.946
eday+0.786
rous+0.784
 that+0.754
und+0.739
Neg logit contributions
 eyel-0.639
cel-0.591
 moral-0.573
 floor-0.559
 fore-0.557
Tokenuck
Token ID1347
Feature activation+5.64e-01

Pos logit contributions
ised+3.798
rous+3.091
izing+2.892
 that+2.660
eday+2.636
Neg logit contributions
cel-6.383
 eyel-6.375
nob-6.346
any-5.762
 fore-5.695
Token up
Token ID510
Feature activation+7.02e-01

Pos logit contributions
 eyel+0.741
cel+0.701
 moral+0.672
 fore+0.647
 floor+0.643
Neg logit contributions
ised-1.013
 that-0.823
eday-0.817
rous-0.811
und-0.770
Token to
Token ID284
Feature activation+8.68e-01

Pos logit contributions
 eyel+1.054
cel+1.026
 moral+0.958
 floor+0.943
quin+0.937
Neg logit contributions
ised-1.122
 that-0.897
eday-0.886
rous-0.878
und-0.815
Token the
Token ID262
Feature activation-9.66e-01

Pos logit contributions
 eyel+1.279
cel+1.201
 moral+1.150
 fore+1.148
 floor+1.128
Neg logit contributions
ised-1.388
 that-1.155
eday-1.129
rous-1.103
und-1.025
Token knot
Token ID29654
Feature activation+1.90e-01

Pos logit contributions
ised+1.219
eday+0.953
rous+0.944
und+0.898
aw+0.817
Neg logit contributions
 eyel-1.784
 floor-1.712
 moral-1.711
cel-1.676
 fore-1.623
Token be
Token ID307
Feature activation+1.29e-01

Pos logit contributions
ised+0.256
eday+0.204
rous+0.204
 that+0.197
und+0.193
Neg logit contributions
 eyel-0.213
cel-0.198
 moral-0.193
 floor-0.188
 fore-0.188
Token craft
Token ID5977
Feature activation-5.94e-01

Pos logit contributions
 eyel+0.197
cel+0.187
 moral+0.179
 floor+0.176
 fore+0.176
Neg logit contributions
ised-0.203
eday-0.160
rous-0.159
 that-0.156
und-0.149
Tokeny
Token ID88
Feature activation-7.65e-01

Pos logit contributions
ised+1.070
eday+0.893
rous+0.884
 that+0.841
und+0.833
Neg logit contributions
 eyel-0.743
cel-0.693
 floor-0.669
 moral-0.664
 fore-0.656
Token.
Token ID13
Feature activation-5.61e-01

Pos logit contributions
ised+1.487
eday+1.231
rous+1.226
amine+1.152
und+1.150
Neg logit contributions
 eyel-0.886
cel-0.813
 floor-0.793
 moral-0.783
 fore-0.776
Token He
Token ID679
Feature activation-1.56e-01

Pos logit contributions
ised+1.081
rous+0.906
eday+0.902
 that+0.864
und+0.834
Neg logit contributions
 eyel-0.656
cel-0.609
 floor-0.586
 moral-0.581
 fore-0.565
Token sn
Token ID3013
Feature activation-3.39e+00

Pos logit contributions
ised+0.285
eday+0.235
rous+0.233
 that+0.227
und+0.220
Neg logit contributions
 eyel-0.199
cel-0.186
 moral-0.178
 floor-0.175
 fore-0.174
Tokenuck
Token ID1347
Feature activation+2.58e-01

Pos logit contributions
ised+3.719
rous+3.143
uable+3.113
izing+2.975
eday+2.890
Neg logit contributions
 eyel-6.059
 fore-5.799
cel-5.668
nob-5.608
 moral-5.431
Token around
Token ID1088
Feature activation+1.14e-01

Pos logit contributions
 eyel+0.327
cel+0.308
 moral+0.293
 floor+0.285
 fore+0.284
Neg logit contributions
ised-0.475
eday-0.388
rous-0.385
 that-0.382
und-0.366
Token until
Token ID1566
Feature activation+4.94e-01

Pos logit contributions
 eyel+0.143
cel+0.134
 moral+0.127
 floor+0.124
 fore+0.124
Neg logit contributions
ised-0.211
eday-0.173
rous-0.172
 that-0.169
und-0.163
Token he
Token ID339
Feature activation+1.87e-01

Pos logit contributions
 eyel+0.607
cel+0.577
 moral+0.540
 fore+0.526
 floor+0.526
Neg logit contributions
ised-0.932
 that-0.766
eday-0.765
rous-0.757
und-0.716
Token was
Token ID373
Feature activation+5.62e-01

Pos logit contributions
 eyel+0.239
cel+0.224
 moral+0.213
 floor+0.209
 fore+0.208
Neg logit contributions
ised-0.340
eday-0.278
rous-0.276
 that-0.272
und-0.262
Token own
Token ID898
Feature activation-1.95e-01

Pos logit contributions
ised+1.660
eday+1.390
und+1.382
rous+1.359
aw+1.356
Neg logit contributions
 eyel-0.891
 moral-0.804
cel-0.781
 floor-0.772
 magn-0.751
Token blue
Token ID4171
Feature activation+1.55e-01

Pos logit contributions
ised+0.403
eday+0.341
rous+0.337
 that+0.327
und+0.327
Neg logit contributions
 eyel-0.202
cel-0.184
 moral-0.177
 floor-0.172
 fore-0.169
Token pillow
Token ID28774
Feature activation-1.05e-01

Pos logit contributions
 eyel+0.142
cel+0.129
 moral+0.120
 floor+0.117
 fore+0.116
Neg logit contributions
ised-0.338
eday-0.287
rous-0.285
 that-0.282
und-0.273
Token.
Token ID13
Feature activation-1.51e-01

Pos logit contributions
ised+0.211
eday+0.176
rous+0.175
 that+0.169
und+0.167
Neg logit contributions
 eyel-0.115
cel-0.105
 moral-0.100
 floor-0.098
 fore-0.098
Token She
Token ID1375
Feature activation+2.08e-01

Pos logit contributions
ised+0.235
rous+0.183
eday+0.183
 that+0.175
und+0.169
Neg logit contributions
 eyel-0.238
cel-0.225
 moral-0.215
 floor-0.213
 fore-0.211
Token sn
Token ID3013
Feature activation-3.38e+00

Pos logit contributions
 eyel+0.262
cel+0.246
 moral+0.234
 floor+0.229
 fore+0.228
Neg logit contributions
ised-0.380
eday-0.311
rous-0.309
 that-0.305
und-0.293
Tokenugg
Token ID6837
Feature activation-1.01e+00

Pos logit contributions
ised+4.137
rous+3.313
izing+3.213
uable+3.111
amine+2.840
Neg logit contributions
 eyel-6.238
cel-5.919
nob-5.776
 fore-5.645
any-5.449
Tokenled
Token ID992
Feature activation+1.66e-01

Pos logit contributions
ised+1.684
rous+1.450
eday+1.429
und+1.321
 that+1.314
Neg logit contributions
 eyel-1.379
cel-1.322
 floor-1.270
 fore-1.236
 moral-1.228
Token up
Token ID510
Feature activation+4.95e-01

Pos logit contributions
 eyel+0.234
cel+0.221
 moral+0.211
 floor+0.207
 fore+0.206
Neg logit contributions
ised-0.282
eday-0.227
rous-0.225
 that-0.221
und-0.212
Token and
Token ID290
Feature activation-2.03e-02

Pos logit contributions
 eyel+0.700
cel+0.672
 moral+0.640
 floor+0.625
quin+0.618
Neg logit contributions
ised-0.828
eday-0.666
 that-0.663
rous-0.662
und-0.620
Token went
Token ID1816
Feature activation+4.56e-01

Pos logit contributions
ised+0.037
eday+0.031
rous+0.030
 that+0.030
und+0.029
Neg logit contributions
 eyel-0.026
cel-0.024
 moral-0.023
 floor-0.022
 fore-0.022
Token tried
Token ID3088
Feature activation+1.38e+00

Pos logit contributions
 eyel+0.117
cel+0.109
 moral+0.103
 floor+0.101
 fore+0.101
Neg logit contributions
ised-0.178
eday-0.147
rous-0.146
 that-0.143
und-0.138
Token it
Token ID340
Feature activation+2.53e-01

Pos logit contributions
 eyel+2.729
cel+2.692
 moral+2.574
quin+2.540
 fore+2.501
Neg logit contributions
ised-1.427
 that-1.188
rous-1.011
eday-0.986
und-0.880
Token out
Token ID503
Feature activation-2.96e-01

Pos logit contributions
 eyel+0.306
cel+0.286
 moral+0.272
 floor+0.265
 fore+0.265
Neg logit contributions
ised-0.476
eday-0.392
rous-0.388
 that-0.386
und-0.370
Token.
Token ID13
Feature activation-3.43e-01

Pos logit contributions
ised+0.585
eday+0.491
rous+0.489
 that+0.462
und+0.461
Neg logit contributions
 eyel-0.333
cel-0.307
 moral-0.291
 floor-0.286
 fore-0.286
Token She
Token ID1375
Feature activation+3.66e-01

Pos logit contributions
ised+0.567
rous+0.459
eday+0.456
 that+0.434
aw+0.427
Neg logit contributions
 eyel-0.500
cel-0.466
 floor-0.448
 moral-0.444
 fore-0.435
Token sn
Token ID3013
Feature activation-3.36e+00

Pos logit contributions
 eyel+0.474
cel+0.446
 moral+0.424
 floor+0.413
 fore+0.412
Neg logit contributions
ised-0.658
eday-0.536
rous-0.530
 that-0.528
und-0.502
Tokenugg
Token ID6837
Feature activation-6.20e-01

Pos logit contributions
ised+3.877
rous+3.264
izing+3.098
uable+3.002
 that+2.849
Neg logit contributions
 eyel-6.323
cel-6.069
nob-5.945
 fore-5.645
any-5.623
Tokenled
Token ID992
Feature activation+1.51e-01

Pos logit contributions
ised+0.997
eday+0.834
rous+0.830
 that+0.789
und+0.760
Neg logit contributions
 eyel-0.878
cel-0.842
 floor-0.808
 moral-0.791
 fore-0.786
Token into
Token ID656
Feature activation+4.42e-01

Pos logit contributions
 eyel+0.206
cel+0.194
 moral+0.185
 floor+0.182
 fore+0.181
Neg logit contributions
ised-0.260
eday-0.210
rous-0.209
 that-0.205
und-0.197
Token the
Token ID262
Feature activation-2.37e-01

Pos logit contributions
 eyel+0.606
cel+0.575
 moral+0.548
 fore+0.534
 floor+0.529
Neg logit contributions
ised-0.764
 that-0.612
eday-0.611
rous-0.609
und-0.575
Token soft
Token ID2705
Feature activation+2.91e-01

Pos logit contributions
ised+0.424
eday+0.347
rous+0.345
 that+0.333
und+0.328
Neg logit contributions
 eyel-0.312
cel-0.291
 floor-0.283
 moral-0.281
 fore-0.273
Token steal
Token ID8711
Feature activation+6.85e-01

Pos logit contributions
ised+0.062
eday+0.049
rous+0.049
 that+0.048
und+0.046
Neg logit contributions
 eyel-0.051
cel-0.048
 moral-0.046
 floor-0.045
 fore-0.045
Token a
Token ID257
Feature activation-5.66e-02

Pos logit contributions
 eyel+0.913
cel+0.873
 moral+0.801
quin+0.787
 fore+0.784
Neg logit contributions
ised-1.228
 that-1.011
rous-0.992
eday-0.984
aw-0.953
Token haircut
Token ID45548
Feature activation-1.69e-01

Pos logit contributions
ised+0.098
eday+0.079
rous+0.078
 that+0.076
und+0.074
Neg logit contributions
 eyel-0.077
cel-0.073
 moral-0.070
 floor-0.069
 fore-0.068
Token!
Token ID0
Feature activation-1.69e-01

Pos logit contributions
ised+0.331
rous+0.275
eday+0.274
 that+0.262
und+0.259
Neg logit contributions
 eyel-0.191
cel-0.178
 floor-0.168
 moral-0.167
 fore-0.163
Token He
Token ID679
Feature activation-4.84e-01

Pos logit contributions
ised+0.311
rous+0.254
eday+0.253
 that+0.246
und+0.237
Neg logit contributions
 eyel-0.214
cel-0.199
 moral-0.192
 floor-0.189
 fore-0.185
Token sn
Token ID3013
Feature activation-3.36e+00

Pos logit contributions
ised+0.886
rous+0.731
eday+0.721
 that+0.696
und+0.681
Neg logit contributions
 eyel-0.622
cel-0.576
 moral-0.560
 floor-0.551
 fore-0.547
Tokenuck
Token ID1347
Feature activation+1.89e-01

Pos logit contributions
ised+3.857
izing+3.213
rous+3.203
uable+2.777
 that+2.672
Neg logit contributions
 eyel-6.117
nob-5.960
cel-5.876
 fore-5.608
any-5.541
Token into
Token ID656
Feature activation-1.21e-01

Pos logit contributions
 eyel+0.252
cel+0.237
 moral+0.226
 floor+0.221
 fore+0.221
Neg logit contributions
ised-0.334
eday-0.271
rous-0.268
 that-0.265
und-0.254
Token a
Token ID257
Feature activation-5.06e-01

Pos logit contributions
ised+0.236
eday+0.198
rous+0.196
 that+0.190
und+0.186
Neg logit contributions
 eyel-0.139
cel-0.129
 moral-0.124
 floor-0.123
 fore-0.121
Token salon
Token ID46608
Feature activation+2.58e-01

Pos logit contributions
ised+0.990
eday+0.834
rous+0.806
 that+0.778
und+0.772
Neg logit contributions
 eyel-0.590
cel-0.544
 floor-0.535
 moral-0.521
 fore-0.512
Token and
Token ID290
Feature activation-1.53e-01

Pos logit contributions
 eyel+0.341
cel+0.320
 moral+0.307
 fore+0.299
 floor+0.298
Neg logit contributions
ised-0.457
eday-0.372
rous-0.368
 that-0.367
und-0.349
Token delicious
Token ID12625
Feature activation-2.81e-01

Pos logit contributions
 eyel+0.842
cel+0.806
 moral+0.738
quin+0.736
 fore+0.734
Neg logit contributions
ised-1.004
eday-0.810
 that-0.803
rous-0.802
und-0.732
Token dinner
Token ID8073
Feature activation+6.21e-02

Pos logit contributions
ised+0.509
eday+0.420
rous+0.415
 that+0.399
und+0.397
Neg logit contributions
 eyel-0.359
cel-0.336
 moral-0.323
 floor-0.323
 fore-0.313
Token she
Token ID673
Feature activation+1.55e-01

Pos logit contributions
 eyel+0.074
cel+0.069
 moral+0.065
 floor+0.064
 fore+0.064
Neg logit contributions
ised-0.118
eday-0.098
rous-0.097
 that-0.095
und-0.092
Token made
Token ID925
Feature activation+1.16e+00

Pos logit contributions
 eyel+0.183
cel+0.170
 moral+0.162
 floor+0.158
 fore+0.157
Neg logit contributions
ised-0.299
eday-0.247
rous-0.245
 that-0.242
und-0.233
Token!
Token ID0
Feature activation+2.30e-01

Pos logit contributions
 eyel+1.552
cel+1.465
 moral+1.370
quin+1.346
 magn+1.330
Neg logit contributions
ised-2.057
 that-1.783
eday-1.655
rous-1.642
aw-1.575
Token From
Token ID3574
Feature activation-3.35e+00

Pos logit contributions
 eyel+0.318
cel+0.300
 moral+0.286
 floor+0.281
 fore+0.281
Neg logit contributions
ised-0.393
eday-0.317
rous-0.314
 that-0.311
und-0.298
Token that
Token ID326
Feature activation+5.38e-01

Pos logit contributions
ised+3.881
und+3.631
Fine+3.304
unch+3.226
itting+3.164
Neg logit contributions
 moral-6.269
 other-5.916
 fore-5.660
 floor-5.495
 cor-5.416
Token day
Token ID1110
Feature activation+4.64e-01

Pos logit contributions
 eyel+0.784
cel+0.738
 floor+0.684
 moral+0.683
 fore+0.682
Neg logit contributions
ised-0.893
eday-0.718
 that-0.714
rous-0.712
und-0.664
Token on
Token ID319
Feature activation+2.40e-01

Pos logit contributions
 eyel+0.581
cel+0.539
 floor+0.513
 moral+0.512
 fore+0.501
Neg logit contributions
ised-0.858
eday-0.703
 that-0.699
rous-0.695
und-0.667
Token,
Token ID11
Feature activation+8.98e-01

Pos logit contributions
 eyel+0.347
cel+0.329
 moral+0.315
 floor+0.308
 fore+0.308
Neg logit contributions
ised-0.398
eday-0.318
rous-0.315
 that-0.312
und-0.298
Token the
Token ID262
Feature activation+7.25e-01

Pos logit contributions
 eyel+1.343
cel+1.266
 fore+1.199
 floor+1.184
nob+1.182
Neg logit contributions
ised-1.465
 that-1.206
eday-1.175
rous-1.162
und-1.070
Token didn
Token ID1422
Feature activation+7.65e-01

Pos logit contributions
ised+0.179
eday+0.147
rous+0.147
 that+0.143
und+0.140
Neg logit contributions
 eyel-0.110
cel-0.102
 moral-0.099
 fore-0.096
 floor-0.096
Token't
Token ID470
Feature activation+8.55e-01

Pos logit contributions
 eyel+1.013
cel+0.959
 magn+0.885
 floor+0.882
 fore+0.876
Neg logit contributions
ised-1.356
 that-1.124
eday-1.109
rous-1.081
und-1.035
Token listen
Token ID6004
Feature activation-1.07e-01

Pos logit contributions
 eyel+1.413
cel+1.360
 floor+1.269
 fore+1.258
 moral+1.252
Neg logit contributions
ised-1.237
 that-0.994
eday-0.976
rous-0.948
amine-0.881
Token.
Token ID13
Feature activation-4.06e-01

Pos logit contributions
ised+0.218
eday+0.183
rous+0.182
 that+0.176
und+0.174
Neg logit contributions
 eyel-0.113
cel-0.103
 moral-0.099
 floor-0.096
 fore-0.096
Token She
Token ID1375
Feature activation+2.31e-01

Pos logit contributions
ised+0.683
rous+0.554
eday+0.548
 that+0.527
und+0.516
Neg logit contributions
 eyel-0.581
cel-0.537
 floor-0.520
 moral-0.520
 fore-0.505
Token sn
Token ID3013
Feature activation-3.33e+00

Pos logit contributions
 eyel+0.315
cel+0.297
 moral+0.282
 floor+0.277
 fore+0.276
Neg logit contributions
ised-0.402
eday-0.325
rous-0.321
 that-0.318
und-0.303
Tokenuck
Token ID1347
Feature activation+4.25e-01

Pos logit contributions
ised+3.622
rous+3.137
izing+2.896
uable+2.805
 that+2.609
Neg logit contributions
 eyel-5.910
 fore-5.804
cel-5.754
nob-5.667
 moral-5.664
Token into
Token ID656
Feature activation-1.97e-02

Pos logit contributions
 eyel+0.572
cel+0.540
 moral+0.516
 floor+0.500
 fore+0.499
Neg logit contributions
ised-0.749
eday-0.606
 that-0.600
rous-0.599
und-0.567
Token the
Token ID262
Feature activation-9.91e-01

Pos logit contributions
ised+0.035
eday+0.028
rous+0.028
 that+0.027
und+0.026
Neg logit contributions
 eyel-0.027
cel-0.025
 moral-0.024
 floor-0.024
 fore-0.023
Token zoo
Token ID26626
Feature activation-4.32e-01

Pos logit contributions
ised+1.971
eday+1.614
rous+1.588
und+1.542
Deal+1.510
Neg logit contributions
 eyel-1.132
 floor-1.105
 moral-1.087
cel-1.076
 fore-1.038
Token at
Token ID379
Feature activation+6.31e-02

Pos logit contributions
ised+0.845
eday+0.703
rous+0.702
und+0.664
 that+0.657
Neg logit contributions
 eyel-0.491
cel-0.456
 floor-0.452
 fore-0.433
 moral-0.432
Token loved
Token ID6151
Feature activation-1.52e+00

Pos logit contributions
ised+1.414
rous+1.204
eday+1.190
und+1.114
 that+1.106
Neg logit contributions
 eyel-0.720
cel-0.691
 fore-0.626
 floor-0.625
 moral-0.621
Token it
Token ID340
Feature activation+5.47e-01

Pos logit contributions
ised+3.077
eday+2.505
rous+2.432
und+2.421
aw+2.336
Neg logit contributions
 eyel-1.842
cel-1.631
 floor-1.615
 moral-1.588
 other-1.576
Token very
Token ID845
Feature activation+4.68e-01

Pos logit contributions
 eyel+0.681
cel+0.646
 moral+0.614
 fore+0.597
quin+0.596
Neg logit contributions
ised-1.001
 that-0.833
eday-0.823
rous-0.811
und-0.769
Token much
Token ID881
Feature activation-4.83e-01

Pos logit contributions
 eyel+0.619
cel+0.591
 moral+0.557
 floor+0.548
 fore+0.548
Neg logit contributions
ised-0.835
 that-0.674
eday-0.672
rous-0.671
aw-0.629
Token.
Token ID13
Feature activation-4.46e-01

Pos logit contributions
ised+1.054
rous+0.900
eday+0.890
und+0.860
amine+0.832
Neg logit contributions
 eyel-0.451
cel-0.407
 moral-0.383
 floor-0.380
 fore-0.375
Token From
Token ID3574
Feature activation-3.32e+00

Pos logit contributions
ised+0.834
eday+0.689
rous+0.688
 that+0.650
und+0.640
Neg logit contributions
 eyel-0.551
cel-0.520
 moral-0.491
 floor-0.490
 fore-0.470
Token that
Token ID326
Feature activation+5.87e-01

Pos logit contributions
ised+3.425
und+3.155
Deal+2.999
unch+2.802
Fine+2.792
Neg logit contributions
 moral-6.435
 other-6.125
 fore-5.870
 floor-5.834
 eyel-5.638
Token day
Token ID1110
Feature activation+3.77e-01

Pos logit contributions
 eyel+0.860
cel+0.807
 moral+0.749
 floor+0.749
 fore+0.747
Neg logit contributions
ised-0.969
eday-0.780
 that-0.779
rous-0.770
und-0.721
Token on
Token ID319
Feature activation+3.17e-01

Pos logit contributions
 eyel+0.463
cel+0.431
 moral+0.410
 floor+0.407
 fore+0.401
Neg logit contributions
ised-0.702
eday-0.577
 that-0.572
rous-0.571
und-0.546
Token,
Token ID11
Feature activation+8.58e-01

Pos logit contributions
 eyel+0.466
cel+0.441
 moral+0.423
 floor+0.415
 fore+0.415
Neg logit contributions
ised-0.523
eday-0.415
 that-0.411
rous-0.410
und-0.388
Token Lily
Token ID20037
Feature activation+6.71e-01

Pos logit contributions
 eyel+1.129
cel+1.065
 fore+1.020
nob+1.009
 moral+0.989
Neg logit contributions
ised-1.534
 that-1.291
eday-1.252
rous-1.231
und-1.141
Token banana
Token ID25996
Feature activation+3.50e-01

Pos logit contributions
 eyel+0.473
cel+0.450
 moral+0.422
 fore+0.417
 floor+0.410
Neg logit contributions
ised-0.636
eday-0.520
 that-0.515
rous-0.514
und-0.486
Token he
Token ID339
Feature activation+1.42e-01

Pos logit contributions
 eyel+0.402
cel+0.376
 moral+0.356
 fore+0.343
 floor+0.342
Neg logit contributions
ised-0.678
eday-0.562
 that-0.561
rous-0.558
und-0.533
Token ever
Token ID1683
Feature activation+1.75e-01

Pos logit contributions
 eyel+0.186
cel+0.175
 moral+0.167
 floor+0.163
 fore+0.163
Neg logit contributions
ised-0.253
eday-0.206
rous-0.205
 that-0.201
und-0.194
Token tasted
Token ID29187
Feature activation-4.22e-01

Pos logit contributions
 eyel+0.206
cel+0.191
 moral+0.181
 floor+0.177
 fore+0.177
Neg logit contributions
ised-0.337
eday-0.279
rous-0.277
 that-0.273
und-0.263
Token!
Token ID0
Feature activation+3.85e-01

Pos logit contributions
ised+0.861
eday+0.725
rous+0.720
und+0.687
 that+0.677
Neg logit contributions
 eyel-0.442
cel-0.402
 moral-0.393
 floor-0.386
 fore-0.377
Token From
Token ID3574
Feature activation-3.32e+00

Pos logit contributions
 eyel+0.534
cel+0.502
 moral+0.480
 fore+0.474
 floor+0.469
Neg logit contributions
ised-0.657
eday-0.530
 that-0.524
rous-0.522
und-0.498
Token that
Token ID326
Feature activation+5.90e-01

Pos logit contributions
ised+3.550
und+3.218
itting+2.971
Deal+2.917
Fine+2.908
Neg logit contributions
 moral-6.301
 other-6.093
 floor-5.811
's-5.585
 cor-5.544
Token day
Token ID1110
Feature activation+1.65e-01

Pos logit contributions
 eyel+0.864
cel+0.815
 fore+0.756
quin+0.753
 moral+0.753
Neg logit contributions
ised-0.970
eday-0.782
 that-0.779
rous-0.767
und-0.724
Token on
Token ID319
Feature activation+1.50e-01

Pos logit contributions
 eyel+0.196
cel+0.183
 moral+0.173
 floor+0.170
 fore+0.169
Neg logit contributions
ised-0.312
eday-0.258
rous-0.256
 that-0.252
und-0.243
Token,
Token ID11
Feature activation+8.28e-02

Pos logit contributions
 eyel+0.216
cel+0.204
 moral+0.195
 floor+0.192
 fore+0.191
Neg logit contributions
ised-0.251
eday-0.201
rous-0.200
 that-0.196
und-0.188
Token Max
Token ID5436
Feature activation+3.90e-01

Pos logit contributions
 eyel+0.130
cel+0.124
 moral+0.119
 floor+0.117
 fore+0.117
Neg logit contributions
ised-0.127
eday-0.099
rous-0.098
 that-0.095
und-0.092
Token the
Token ID262
Feature activation-2.44e-01

Pos logit contributions
 eyel+0.349
cel+0.329
 moral+0.314
 floor+0.307
 fore+0.306
Neg logit contributions
ised-0.441
eday-0.356
rous-0.353
 that-0.350
und-0.334
Token DVD
Token ID12490
Feature activation+1.54e-01

Pos logit contributions
ised+0.486
eday+0.401
rous+0.398
 that+0.387
und+0.381
Neg logit contributions
 eyel-0.279
cel-0.258
 moral-0.246
 floor-0.246
 fore-0.242
Token player
Token ID2137
Feature activation+1.17e-01

Pos logit contributions
 eyel+0.194
cel+0.182
 moral+0.173
 floor+0.169
 fore+0.169
Neg logit contributions
ised-0.283
eday-0.232
rous-0.230
 that-0.227
und-0.218
Token.
Token ID13
Feature activation-1.37e-01

Pos logit contributions
 eyel+0.119
cel+0.110
 moral+0.103
 floor+0.100
 fore+0.100
Neg logit contributions
ised-0.243
eday-0.204
rous-0.203
 that-0.200
und-0.194
Token They
Token ID1119
Feature activation-1.65e-01

Pos logit contributions
ised+0.202
eday+0.156
rous+0.156
 that+0.150
und+0.143
Neg logit contributions
 eyel-0.223
cel-0.212
 moral-0.204
 fore-0.201
 floor-0.201
Token sn
Token ID3013
Feature activation-3.31e+00

Pos logit contributions
ised+0.310
eday+0.255
rous+0.255
 that+0.246
und+0.239
Neg logit contributions
 eyel-0.201
cel-0.189
 moral-0.178
 floor-0.176
 fore-0.175
Tokenugg
Token ID6837
Feature activation-8.73e-01

Pos logit contributions
ised+3.827
rous+3.351
uable+3.177
izing+3.059
 afar+2.954
Neg logit contributions
 eyel-6.077
cel-5.780
any-5.515
 fore-5.488
nob-5.370
Tokenled
Token ID992
Feature activation-9.55e-02

Pos logit contributions
ised+1.420
rous+1.218
eday+1.208
 that+1.135
Deal+1.104
Neg logit contributions
 eyel-1.196
cel-1.154
 floor-1.112
 fore-1.091
 moral-1.066
Token up
Token ID510
Feature activation+2.20e-01

Pos logit contributions
ised+0.164
eday+0.133
rous+0.131
 that+0.128
und+0.124
Neg logit contributions
 eyel-0.131
cel-0.122
 moral-0.117
 floor-0.116
 fore-0.116
Token on
Token ID319
Feature activation-5.39e-01

Pos logit contributions
 eyel+0.306
cel+0.289
 moral+0.276
 floor+0.270
 fore+0.269
Neg logit contributions
ised-0.375
eday-0.302
rous-0.300
 that-0.296
und-0.283
Token the
Token ID262
Feature activation+9.44e-01

Pos logit contributions
ised+0.991
eday+0.814
rous+0.805
Fine+0.760
und+0.759
Neg logit contributions
 eyel-0.686
cel-0.631
 floor-0.609
 moral-0.606
 fore-0.601
Token steal
Token ID8711
Feature activation+5.85e-01

Pos logit contributions
ised+0.571
eday+0.455
rous+0.453
 that+0.431
aw+0.427
Neg logit contributions
 eyel-0.464
cel-0.428
 moral-0.420
 floor-0.417
 fore-0.407
Token her
Token ID607
Feature activation-2.97e-01

Pos logit contributions
 eyel+0.784
cel+0.755
 moral+0.697
 fore+0.684
quin+0.682
Neg logit contributions
ised-1.032
 that-0.843
rous-0.833
eday-0.831
aw-0.793
Token toys
Token ID14958
Feature activation-1.16e-01

Pos logit contributions
ised+0.535
eday+0.436
rous+0.434
 that+0.415
und+0.410
Neg logit contributions
 eyel-0.388
cel-0.357
 moral-0.350
 floor-0.345
 fore-0.340
Token.
Token ID13
Feature activation-1.11e+00

Pos logit contributions
ised+0.238
eday+0.199
rous+0.199
 that+0.192
und+0.189
Neg logit contributions
 eyel-0.119
cel-0.109
 moral-0.104
 floor-0.103
 fore-0.101
Token He
Token ID679
Feature activation-2.80e-01

Pos logit contributions
ised+2.058
rous+1.783
eday+1.761
und+1.615
aw+1.602
Neg logit contributions
 eyel-1.376
cel-1.286
 floor-1.252
 moral-1.197
's-1.195
Token sn
Token ID3013
Feature activation-3.31e+00

Pos logit contributions
ised+0.515
rous+0.424
eday+0.422
 that+0.405
und+0.395
Neg logit contributions
 eyel-0.358
cel-0.334
 moral-0.319
 floor-0.315
 fore-0.313
Tokenuck
Token ID1347
Feature activation+5.65e-01

Pos logit contributions
ised+4.121
rous+3.528
izing+3.335
uable+2.898
eday+2.875
Neg logit contributions
nob-5.639
 eyel-5.626
cel-5.549
 fore-5.209
 moral-5.158
Token around
Token ID1088
Feature activation+2.92e-01

Pos logit contributions
 eyel+0.763
cel+0.730
 moral+0.694
 fore+0.670
 floor+0.664
Neg logit contributions
ised-0.988
 that-0.801
eday-0.799
rous-0.788
und-0.750
Token the
Token ID262
Feature activation-4.54e-01

Pos logit contributions
 eyel+0.380
cel+0.359
 moral+0.339
 floor+0.331
 fore+0.331
Neg logit contributions
ised-0.525
eday-0.429
rous-0.425
 that-0.423
und-0.403
Token corner
Token ID5228
Feature activation-1.64e-01

Pos logit contributions
ised+0.889
eday+0.739
rous+0.730
und+0.697
 that+0.695
Neg logit contributions
 eyel-0.526
 floor-0.497
cel-0.491
 moral-0.482
 fore-0.464
Token and
Token ID290
Feature activation-2.22e-01

Pos logit contributions
ised+0.301
eday+0.247
rous+0.246
 that+0.236
und+0.232
Neg logit contributions
 eyel-0.205
cel-0.190
 moral-0.183
 floor-0.183
 fore-0.180
Token It
Token ID632
Feature activation+1.59e+00

Pos logit contributions
 eyel+0.636
cel+0.606
 moral+0.588
 fore+0.579
 floor+0.570
Neg logit contributions
ised-0.640
eday-0.504
 that-0.498
rous-0.496
und-0.469
Token was
Token ID373
Feature activation-1.52e-01

Pos logit contributions
 eyel+1.865
cel+1.824
quin+1.739
 moral+1.686
 magn+1.657
Neg logit contributions
ised-3.067
 that-2.584
rous-2.421
eday-2.414
und-2.329
Token very
Token ID845
Feature activation-5.37e-01

Pos logit contributions
ised+0.255
eday+0.205
rous+0.202
 that+0.194
und+0.192
Neg logit contributions
 eyel-0.218
cel-0.203
 moral-0.197
 floor-0.195
 fore-0.193
Token refreshing
Token ID23056
Feature activation+1.72e-01

Pos logit contributions
ised+0.914
eday+0.761
rous+0.726
und+0.710
 that+0.701
Neg logit contributions
 eyel-0.757
 moral-0.685
 floor-0.676
cel-0.670
 fore-0.659
Token!
Token ID0
Feature activation+7.52e-01

Pos logit contributions
 eyel+0.226
cel+0.212
 moral+0.202
 floor+0.198
 fore+0.197
Neg logit contributions
ised-0.306
eday-0.249
rous-0.247
 that-0.244
und-0.234
Token From
Token ID3574
Feature activation-3.30e+00

Pos logit contributions
 eyel+1.081
cel+1.026
 moral+1.000
 fore+0.987
 cor+0.967
Neg logit contributions
ised-1.233
 that-0.997
eday-0.978
rous-0.962
und-0.929
Token that
Token ID326
Feature activation+9.78e-01

Pos logit contributions
ised+4.021
und+3.697
Deal+3.371
unch+3.355
itting+3.318
Neg logit contributions
 moral-5.766
 other-5.606
 floor-5.266
 fore-5.222
 eyel-5.219
Token day
Token ID1110
Feature activation+2.93e-01

Pos logit contributions
 eyel+1.527
cel+1.443
quin+1.348
 fore+1.335
nob+1.332
Neg logit contributions
ised-1.524
 that-1.258
eday-1.217
rous-1.200
und-1.109
Token on
Token ID319
Feature activation+3.82e-01

Pos logit contributions
 eyel+0.355
cel+0.331
 moral+0.315
 floor+0.311
 fore+0.307
Neg logit contributions
ised-0.550
eday-0.453
rous-0.449
 that-0.447
und-0.428
Token,
Token ID11
Feature activation+8.14e-01

Pos logit contributions
 eyel+0.567
cel+0.537
 moral+0.517
 fore+0.506
 floor+0.504
Neg logit contributions
ised-0.623
eday-0.494
 that-0.491
rous-0.486
und-0.462
Token Lily
Token ID20037
Feature activation+7.38e-01

Pos logit contributions
 eyel+1.072
cel+1.005
 fore+0.965
nob+0.942
 moral+0.936
Neg logit contributions
ised-1.450
 that-1.227
eday-1.190
rous-1.163
und-1.089
Token the
Token ID262
Feature activation+8.59e-01

Pos logit contributions
ised+2.585
eday+2.052
rous+2.033
und+1.926
Fine+1.911
Neg logit contributions
 eyel-1.911
 floor-1.765
cel-1.718
 moral-1.692
 fore-1.680
Token salad
Token ID20698
Feature activation-2.37e-01

Pos logit contributions
 eyel+1.187
cel+1.158
quin+1.047
 fore+1.040
nob+1.032
Neg logit contributions
ised-1.466
 that-1.193
rous-1.185
eday-1.179
Deal-1.090
Token so
Token ID523
Feature activation-3.70e-01

Pos logit contributions
ised+0.452
eday+0.372
rous+0.370
und+0.352
 that+0.352
Neg logit contributions
 eyel-0.285
cel-0.263
 moral-0.251
 floor-0.250
 fore-0.247
Token much
Token ID881
Feature activation-1.30e+00

Pos logit contributions
ised+0.706
eday+0.594
rous+0.585
und+0.554
 that+0.550
Neg logit contributions
 eyel-0.440
cel-0.405
 moral-0.390
 floor-0.379
 fore-0.371
Token.
Token ID13
Feature activation-5.68e-01

Pos logit contributions
ised+2.299
rous+1.901
eday+1.851
und+1.716
amine+1.678
Neg logit contributions
 eyel-1.801
 floor-1.689
cel-1.661
 fore-1.608
 moral-1.595
Token From
Token ID3574
Feature activation-3.28e+00

Pos logit contributions
ised+1.108
rous+0.935
eday+0.925
 that+0.869
aw+0.856
Neg logit contributions
 eyel-0.661
cel-0.620
 moral-0.577
 floor-0.567
quin-0.548
Token then
Token ID788
Feature activation-6.53e-01

Pos logit contributions
ised+3.172
und+2.784
Deal+2.573
Fine+2.436
amine+2.426
Neg logit contributions
 moral-6.693
 other-6.384
 floor-6.092
 fore-6.000
 eyel-5.925
Token on
Token ID319
Feature activation+4.29e-02

Pos logit contributions
ised+1.344
rous+1.128
eday+1.121
 that+1.063
aw+1.061
Neg logit contributions
 eyel-0.674
cel-0.631
 moral-0.612
 fore-0.575
 floor-0.560
Token,
Token ID11
Feature activation+1.23e-01

Pos logit contributions
 eyel+0.061
cel+0.057
 moral+0.055
 floor+0.054
 fore+0.053
Neg logit contributions
ised-0.072
eday-0.058
rous-0.058
 that-0.056
und-0.054
Token Fl
Token ID1610
Feature activation-4.52e-01

Pos logit contributions
 eyel+0.154
cel+0.144
 moral+0.137
 floor+0.134
 fore+0.134
Neg logit contributions
ised-0.228
eday-0.187
rous-0.186
 that-0.183
und-0.176
Tokenoppy
Token ID26696
Feature activation+7.57e-01

Pos logit contributions
ised+0.687
eday+0.550
rous+0.534
 that+0.527
und+0.496
Neg logit contributions
 eyel-0.729
cel-0.683
 moral-0.659
 floor-0.637
 magn-0.636
Token in
Token ID287
Feature activation+2.70e-01

Pos logit contributions
ised+0.869
rous+0.743
eday+0.735
und+0.701
 that+0.691
Neg logit contributions
 eyel-0.394
cel-0.357
 floor-0.331
 moral-0.320
 fore-0.319
Token her
Token ID607
Feature activation+7.46e-01

Pos logit contributions
 eyel+0.340
cel+0.320
 moral+0.302
 fore+0.296
 floor+0.295
Neg logit contributions
ised-0.501
eday-0.410
rous-0.409
 that-0.404
und-0.386
Token arms
Token ID5101
Feature activation+8.83e-02

Pos logit contributions
 eyel+0.988
cel+0.964
 moral+0.933
quin+0.896
 fore+0.889
Neg logit contributions
ised-1.288
rous-1.037
 that-1.037
eday-1.026
aw-0.982
Token.
Token ID13
Feature activation+9.81e-02

Pos logit contributions
 eyel+0.090
cel+0.083
 moral+0.077
 floor+0.076
 fore+0.075
Neg logit contributions
ised-0.183
eday-0.154
rous-0.153
 that-0.150
und-0.146
Token She
Token ID1375
Feature activation+2.36e-01

Pos logit contributions
 eyel+0.154
cel+0.146
 moral+0.140
 floor+0.138
 fore+0.137
Neg logit contributions
ised-0.151
eday-0.118
rous-0.117
 that-0.114
und-0.109
Token sn
Token ID3013
Feature activation-3.28e+00

Pos logit contributions
 eyel+0.321
cel+0.303
 moral+0.289
 floor+0.283
 fore+0.282
Neg logit contributions
ised-0.409
eday-0.330
rous-0.327
 that-0.324
und-0.309
Tokenugg
Token ID6837
Feature activation-7.00e-01

Pos logit contributions
ised+3.850
rous+3.130
izing+2.990
uable+2.851
und+2.716
Neg logit contributions
 eyel-6.063
cel-5.938
nob-5.592
any-5.520
 fore-5.376
Tokenled
Token ID992
Feature activation+4.96e-02

Pos logit contributions
ised+1.133
eday+0.950
rous+0.949
 that+0.893
und+0.872
Neg logit contributions
 eyel-0.985
cel-0.949
 floor-0.905
 moral-0.888
 fore-0.874
Token the
Token ID262
Feature activation+4.88e-01

Pos logit contributions
 eyel+0.069
cel+0.065
 moral+0.062
 floor+0.061
 fore+0.061
Neg logit contributions
ised-0.084
eday-0.068
rous-0.067
 that-0.066
und-0.064
Token doll
Token ID3654
Feature activation+2.79e-02

Pos logit contributions
 eyel+0.489
cel+0.457
 moral+0.427
 fore+0.412
quin+0.406
Neg logit contributions
ised-1.010
eday-0.853
 that-0.845
rous-0.840
und-0.802
Token close
Token ID1969
Feature activation-1.27e+00

Pos logit contributions
 eyel+0.038
cel+0.036
 moral+0.034
 floor+0.034
 fore+0.033
Neg logit contributions
ised-0.048
eday-0.039
rous-0.039
 that-0.038
und-0.036
Token were
Token ID547
Feature activation-5.66e-01

Pos logit contributions
ised+0.531
rous+0.440
eday+0.437
 that+0.420
und+0.408
Neg logit contributions
 eyel-0.359
cel-0.341
 moral-0.325
 floor-0.321
 fore-0.318
Token so
Token ID523
Feature activation-9.63e-01

Pos logit contributions
ised+1.055
eday+0.866
rous+0.861
und+0.821
 that+0.814
Neg logit contributions
 eyel-0.710
cel-0.653
 fore-0.629
 floor-0.628
 moral-0.626
Token happy
Token ID3772
Feature activation-6.57e-01

Pos logit contributions
ised+1.806
eday+1.524
rous+1.473
und+1.426
 that+1.414
Neg logit contributions
 eyel-1.175
cel-1.062
 moral-1.054
 fore-1.045
 floor-1.034
Token.
Token ID13
Feature activation+2.36e-01

Pos logit contributions
ised+1.147
eday+0.938
rous+0.937
amine+0.868
und+0.862
Neg logit contributions
 eyel-0.879
cel-0.819
 floor-0.797
 moral-0.782
 fore-0.774
Token They
Token ID1119
Feature activation-2.58e-01

Pos logit contributions
 eyel+0.347
cel+0.329
 moral+0.315
 fore+0.309
 floor+0.308
Neg logit contributions
ised-0.384
eday-0.305
rous-0.302
 that-0.298
und-0.284
Token sn
Token ID3013
Feature activation-3.27e+00

Pos logit contributions
ised+0.470
rous+0.385
eday+0.385
 that+0.368
und+0.361
Neg logit contributions
 eyel-0.326
cel-0.309
 moral-0.294
 floor-0.289
 fore-0.288
Tokenugg
Token ID6837
Feature activation-7.32e-01

Pos logit contributions
ised+3.760
rous+3.384
uable+3.110
izing+3.072
 afar+2.954
Neg logit contributions
 eyel-5.912
cel-5.839
any-5.589
nob-5.365
 fore-5.352
Tokenled
Token ID992
Feature activation-7.79e-02

Pos logit contributions
ised+1.188
eday+1.006
rous+1.006
 that+0.945
und+0.918
Neg logit contributions
 eyel-1.012
cel-0.976
 floor-0.946
 moral-0.916
 fore-0.915
Token up
Token ID510
Feature activation+5.78e-01

Pos logit contributions
ised+0.136
eday+0.111
rous+0.109
 that+0.107
und+0.103
Neg logit contributions
 eyel-0.105
cel-0.098
 moral-0.093
 floor-0.093
 fore-0.092
Token on
Token ID319
Feature activation-2.12e-01

Pos logit contributions
 eyel+0.834
cel+0.802
 moral+0.761
 floor+0.736
quin+0.731
Neg logit contributions
ised-0.960
eday-0.765
 that-0.763
rous-0.758
und-0.718
Token the
Token ID262
Feature activation+1.76e+00

Pos logit contributions
ised+0.378
eday+0.309
rous+0.306
 that+0.293
und+0.289
Neg logit contributions
 eyel-0.282
cel-0.262
 moral-0.252
 floor-0.251
 fore-0.248
Token a
Token ID257
Feature activation+1.49e-02

Pos logit contributions
ised+1.228
eday+1.026
rous+0.999
und+0.991
Fine+0.942
Neg logit contributions
 eyel-0.784
 moral-0.702
cel-0.702
 fore-0.685
 floor-0.668
Token little
Token ID1310
Feature activation-3.54e-01

Pos logit contributions
 eyel+0.021
cel+0.019
 moral+0.018
 floor+0.018
 fore+0.018
Neg logit contributions
ised-0.026
eday-0.021
rous-0.021
 that-0.020
und-0.019
Token better
Token ID1365
Feature activation-1.04e-01

Pos logit contributions
ised+0.706
eday+0.595
rous+0.588
und+0.571
 that+0.565
Neg logit contributions
 eyel-0.399
cel-0.357
 moral-0.351
 fore-0.339
 floor-0.337
Token.
Token ID13
Feature activation-1.52e-01

Pos logit contributions
ised+0.206
eday+0.172
rous+0.171
 that+0.165
und+0.163
Neg logit contributions
 eyel-0.118
cel-0.109
 moral-0.104
 floor-0.101
 fore-0.101
Token She
Token ID1375
Feature activation+7.01e-01

Pos logit contributions
ised+0.234
rous+0.183
eday+0.182
 that+0.175
und+0.169
Neg logit contributions
 eyel-0.241
cel-0.227
 moral-0.218
 floor-0.216
 fore-0.213
Token sn
Token ID3013
Feature activation-3.26e+00

Pos logit contributions
 eyel+0.865
cel+0.818
 moral+0.761
 floor+0.745
nob+0.744
Neg logit contributions
ised-1.298
 that-1.074
eday-1.067
rous-1.054
aw-1.011
Tokenugg
Token ID6837
Feature activation-7.19e-01

Pos logit contributions
ised+4.158
rous+3.315
izing+3.274
uable+2.992
 afar+2.867
Neg logit contributions
 eyel-5.800
cel-5.558
 fore-5.386
nob-5.276
 moral-5.257
Tokenled
Token ID992
Feature activation-7.23e-02

Pos logit contributions
ised+1.176
rous+0.988
eday+0.980
 that+0.921
und+0.910
Neg logit contributions
 eyel-1.005
cel-0.964
 floor-0.922
 moral-0.912
 fore-0.900
Token up
Token ID510
Feature activation+4.63e-01

Pos logit contributions
ised+0.126
eday+0.102
rous+0.101
 that+0.098
und+0.096
Neg logit contributions
 eyel-0.098
cel-0.092
 moral-0.088
 floor-0.087
 fore-0.086
Token in
Token ID287
Feature activation-7.82e-02

Pos logit contributions
 eyel+0.655
cel+0.628
 moral+0.597
 floor+0.583
 fore+0.578
Neg logit contributions
ised-0.775
eday-0.623
rous-0.619
 that-0.618
und-0.580
Token her
Token ID607
Feature activation+4.26e-01

Pos logit contributions
ised+0.165
eday+0.139
rous+0.137
 that+0.134
und+0.132
Neg logit contributions
 eyel-0.080
cel-0.072
 moral-0.069
 floor-0.067
 fore-0.067
Token he
Token ID339
Feature activation-9.43e-01

Pos logit contributions
ised+1.343
rous+1.106
eday+1.093
und+1.038
amine+1.001
Neg logit contributions
 eyel-0.918
cel-0.824
 moral-0.816
 fore-0.791
 floor-0.760
Token was
Token ID373
Feature activation+4.68e-01

Pos logit contributions
ised+1.899
rous+1.586
eday+1.553
und+1.494
izing+1.450
Neg logit contributions
 eyel-1.087
cel-1.021
 moral-0.969
 fore-0.947
 floor-0.921
Token welcome
Token ID7062
Feature activation-5.08e-01

Pos logit contributions
 eyel+0.664
cel+0.630
 moral+0.599
 floor+0.595
 fore+0.585
Neg logit contributions
ised-0.777
eday-0.631
 that-0.629
rous-0.626
aw-0.582
Token.
Token ID13
Feature activation-1.50e-02

Pos logit contributions
ised+0.951
eday+0.768
rous+0.768
und+0.725
 that+0.712
Neg logit contributions
 eyel-0.643
cel-0.603
 moral-0.570
 fore-0.557
 floor-0.544
Token They
Token ID1119
Feature activation-1.39e-01

Pos logit contributions
ised+0.025
eday+0.020
rous+0.020
 that+0.019
und+0.018
Neg logit contributions
 eyel-0.022
cel-0.021
 moral-0.020
 floor-0.019
 fore-0.019
Token sn
Token ID3013
Feature activation-3.25e+00

Pos logit contributions
ised+0.240
eday+0.193
rous+0.193
 that+0.186
und+0.182
Neg logit contributions
 eyel-0.192
cel-0.180
 moral-0.173
 fore-0.169
 floor-0.169
Tokenugg
Token ID6837
Feature activation-9.39e-01

Pos logit contributions
ised+3.817
izing+3.223
rous+3.136
uable+3.119
amine+2.800
Neg logit contributions
 eyel-5.994
 fore-5.530
cel-5.497
nob-5.307
any-5.302
Tokenled
Token ID992
Feature activation+1.26e-01

Pos logit contributions
ised+1.557
rous+1.325
eday+1.309
 that+1.226
und+1.208
Neg logit contributions
 eyel-1.294
cel-1.216
 floor-1.170
 fore-1.160
 moral-1.155
Token in
Token ID287
Feature activation-1.63e-01

Pos logit contributions
 eyel+0.182
cel+0.171
 moral+0.164
 floor+0.161
 fore+0.161
Neg logit contributions
ised-0.210
eday-0.168
rous-0.167
 that-0.164
und-0.157
Token her
Token ID607
Feature activation+6.76e-01

Pos logit contributions
ised+0.296
eday+0.241
rous+0.238
 that+0.230
und+0.227
Neg logit contributions
 eyel-0.214
cel-0.199
 moral-0.192
 floor-0.189
 fore-0.187
Token heart
Token ID2612
Feature activation+2.27e-01

Pos logit contributions
 eyel+0.789
cel+0.749
 moral+0.712
quin+0.683
 fore+0.675
Neg logit contributions
ised-1.299
 that-1.074
eday-1.064
rous-1.061
und-1.001
Token before
Token ID878
Feature activation-7.22e-02

Pos logit contributions
 eyel+0.261
cel+0.245
 moral+0.232
 fore+0.227
 floor+0.227
Neg logit contributions
ised-0.389
eday-0.320
rous-0.317
 that-0.314
und-0.300
Token bed
Token ID3996
Feature activation-5.47e-01

Pos logit contributions
ised+0.173
eday+0.149
rous+0.149
 that+0.145
und+0.143
Neg logit contributions
 eyel-0.052
cel-0.046
 moral-0.042
 floor-0.040
 fore-0.039
Token,
Token ID11
Feature activation+4.05e-01

Pos logit contributions
ised+0.940
rous+0.773
eday+0.753
aw+0.707
und+0.706
Neg logit contributions
 eyel-0.762
cel-0.718
 moral-0.682
 floor-0.680
 fore-0.679
Token she
Token ID673
Feature activation+5.99e-01

Pos logit contributions
 eyel+0.552
cel+0.522
 moral+0.498
 fore+0.491
 floor+0.486
Neg logit contributions
ised-0.697
eday-0.567
 that-0.564
rous-0.558
und-0.524
Token would
Token ID561
Feature activation-8.49e-02

Pos logit contributions
 eyel+0.788
cel+0.745
 moral+0.712
 fore+0.697
 floor+0.697
Neg logit contributions
ised-1.039
eday-0.854
 that-0.852
rous-0.841
aw-0.797
Token sn
Token ID3013
Feature activation-3.25e+00

Pos logit contributions
ised+0.155
rous+0.125
eday+0.125
 that+0.121
und+0.119
Neg logit contributions
 eyel-0.110
cel-0.102
 moral-0.098
 floor-0.096
 fore-0.095
Tokenuggle
Token ID33498
Feature activation-2.76e-03

Pos logit contributions
ised+4.139
Fine+3.073
izing+3.020
rous+3.007
und+2.820
Neg logit contributions
 eyel-6.204
cel-6.022
nob-5.874
 moral-5.646
 babys-5.554
Token up
Token ID510
Feature activation+9.47e-01

Pos logit contributions
ised+0.005
eday+0.004
rous+0.004
 that+0.004
und+0.004
Neg logit contributions
 eyel-0.004
cel-0.004
 moral-0.003
 floor-0.003
 fore-0.003
Token and
Token ID290
Feature activation+4.67e-01

Pos logit contributions
 eyel+1.419
cel+1.378
 moral+1.301
quin+1.277
 floor+1.264
Neg logit contributions
ised-1.501
 that-1.224
eday-1.195
rous-1.186
und-1.101
Token give
Token ID1577
Feature activation+7.45e-02

Pos logit contributions
 eyel+0.616
cel+0.581
 moral+0.557
 fore+0.547
 floor+0.543
Neg logit contributions
ised-0.820
eday-0.671
rous-0.661
 that-0.661
und-0.619
Token her
Token ID607
Feature activation-7.76e-02

Pos logit contributions
 eyel+0.124
cel+0.118
 moral+0.114
 floor+0.112
 fore+0.112
Neg logit contributions
ised-0.107
eday-0.083
rous-0.082
 that-0.079
und-0.076
Token a
Token ID257
Feature activation+1.15e-01

Pos logit contributions
 eyel+1.115
cel+1.061
 moral+0.999
 floor+0.995
quin+0.974
Neg logit contributions
ised-1.266
 that-1.047
rous-1.025
eday-1.021
aw-0.967
Token very
Token ID845
Feature activation-8.67e-03

Pos logit contributions
 eyel+0.153
cel+0.144
 moral+0.137
 floor+0.134
 fore+0.134
Neg logit contributions
ised-0.203
eday-0.165
rous-0.163
 that-0.160
und-0.154
Token special
Token ID2041
Feature activation+2.49e-01

Pos logit contributions
ised+0.015
eday+0.012
rous+0.012
 that+0.012
und+0.011
Neg logit contributions
 eyel-0.012
cel-0.011
 moral-0.011
 floor-0.011
 fore-0.011
Token reward
Token ID6721
Feature activation+2.94e-01

Pos logit contributions
 eyel+0.346
cel+0.328
 moral+0.311
 floor+0.305
 fore+0.304
Neg logit contributions
ised-0.424
eday-0.341
rous-0.339
 that-0.335
und-0.318
Token.
Token ID13
Feature activation+2.49e-01

Pos logit contributions
 eyel+0.329
cel+0.305
 moral+0.289
 floor+0.280
 fore+0.279
Neg logit contributions
ised-0.575
eday-0.480
 that-0.475
rous-0.475
und-0.454
Token From
Token ID3574
Feature activation-3.25e+00

Pos logit contributions
 eyel+0.339
cel+0.319
 moral+0.304
 floor+0.298
 fore+0.298
Neg logit contributions
ised-0.431
eday-0.350
rous-0.345
 that-0.342
und-0.328
Token then
Token ID788
Feature activation-3.52e-01

Pos logit contributions
ised+3.682
und+3.374
unch+3.257
itting+3.130
Fine+3.058
Neg logit contributions
 moral-5.910
 other-5.609
 fore-5.382
 floor-5.291
's-5.220
Token on
Token ID319
Feature activation+4.85e-01

Pos logit contributions
ised+0.703
rous+0.586
eday+0.583
 that+0.559
aw+0.549
Neg logit contributions
 eyel-0.384
cel-0.364
 moral-0.346
 fore-0.334
 floor-0.326
Token,
Token ID11
Feature activation+3.53e-01

Pos logit contributions
 eyel+0.725
cel+0.688
 moral+0.658
 fore+0.647
 floor+0.645
Neg logit contributions
ised-0.786
 that-0.623
eday-0.622
rous-0.609
und-0.583
Token the
Token ID262
Feature activation+3.65e-01

Pos logit contributions
 eyel+0.505
cel+0.475
 moral+0.452
 floor+0.447
 fore+0.446
Neg logit contributions
ised-0.594
eday-0.478
rous-0.472
 that-0.472
und-0.444
Token girl
Token ID2576
Feature activation+4.92e-01

Pos logit contributions
 eyel+0.505
cel+0.474
 moral+0.444
 floor+0.441
 fore+0.440
Neg logit contributions
ised-0.638
eday-0.513
rous-0.509
 that-0.505
aw-0.474
Token homes
Token ID5682
Feature activation-1.92e-01

Pos logit contributions
ised+0.688
eday+0.558
rous+0.550
 that+0.537
und+0.535
Neg logit contributions
 eyel-0.454
 moral-0.413
cel-0.412
 floor-0.401
 fore-0.394
Token.
Token ID13
Feature activation+1.17e-01

Pos logit contributions
ised+0.403
rous+0.339
eday+0.339
 that+0.325
und+0.322
Neg logit contributions
 eyel-0.191
cel-0.174
 moral-0.167
 floor-0.164
 fore-0.162
Token They
Token ID1119
Feature activation+4.19e-01

Pos logit contributions
 eyel+0.165
cel+0.156
 moral+0.149
 floor+0.147
 fore+0.146
Neg logit contributions
ised-0.198
eday-0.159
rous-0.158
 that-0.155
und-0.149
Token get
Token ID651
Feature activation+2.14e-01

Pos logit contributions
 eyel+0.564
cel+0.530
 moral+0.502
 floor+0.495
 fore+0.493
Neg logit contributions
ised-0.723
eday-0.594
rous-0.587
 that-0.585
und-0.553
Token off
Token ID572
Feature activation+1.65e+00

Pos logit contributions
 eyel+0.268
cel+0.252
 moral+0.239
 floor+0.234
 fore+0.233
Neg logit contributions
ised-0.395
eday-0.324
rous-0.322
 that-0.318
und-0.304
Token the
Token ID262
Feature activation+1.63e+00

Pos logit contributions
 eyel+2.619
cel+2.588
 magn+2.310
nob+2.295
 moral+2.287
Neg logit contributions
ised-2.501
 that-2.218
eday-1.966
rous-1.961
aw-1.867
Token bus
Token ID1323
Feature activation+4.63e-01

Pos logit contributions
 eyel+2.117
cel+2.077
quin+1.910
Beck+1.899
nob+1.824
Neg logit contributions
ised-2.780
 that-2.442
eday-2.388
rous-2.337
Deal-2.168
Token and
Token ID290
Feature activation+7.21e-01

Pos logit contributions
 eyel+0.528
cel+0.492
 moral+0.461
 fore+0.446
 floor+0.443
Neg logit contributions
ised-0.901
eday-0.754
 that-0.752
rous-0.739
und-0.709
Token hug
Token ID16225
Feature activation+6.93e-01

Pos logit contributions
 eyel+1.050
cel+1.011
 moral+0.950
 floor+0.927
 fore+0.926
Neg logit contributions
ised-1.156
eday-0.948
 that-0.944
rous-0.930
und-0.863
Token their
Token ID511
Feature activation+1.02e+00

Pos logit contributions
 eyel+0.902
cel+0.864
 moral+0.800
 floor+0.790
 fore+0.790
Neg logit contributions
ised-1.241
 that-1.027
eday-1.011
rous-0.991
aw-0.960
Token moms
Token ID34015
Feature activation+1.21e+00

Pos logit contributions
 eyel+1.288
cel+1.276
nob+1.168
quin+1.159
 fore+1.152
Neg logit contributions
ised-1.815
 that-1.485
eday-1.484
rous-1.461
aw-1.380
Token lines
Token ID3951
Feature activation+2.28e-01

Pos logit contributions
 eyel+0.793
cel+0.758
 moral+0.694
 fore+0.691
nob+0.685
Neg logit contributions
ised-1.260
 that-1.045
rous-1.025
eday-1.016
aw-0.985
Token and
Token ID290
Feature activation+1.17e+00

Pos logit contributions
 eyel+0.281
cel+0.263
 moral+0.250
 floor+0.243
 fore+0.243
Neg logit contributions
ised-0.422
eday-0.346
rous-0.343
 that-0.341
und-0.326
Token circles
Token ID13332
Feature activation-5.06e-02

Pos logit contributions
 eyel+1.270
cel+1.216
 moral+1.145
 fore+1.116
nob+1.104
Neg logit contributions
ised-2.305
 that-1.987
eday-1.867
rous-1.867
aw-1.838
Token on
Token ID319
Feature activation+6.40e-01

Pos logit contributions
ised+0.099
eday+0.083
rous+0.082
 that+0.080
und+0.078
Neg logit contributions
 eyel-0.057
cel-0.053
 moral-0.050
 floor-0.049
 fore-0.049
Token a
Token ID257
Feature activation+1.38e+00

Pos logit contributions
 eyel+0.989
cel+0.944
 moral+0.904
 fore+0.893
nob+0.867
Neg logit contributions
ised-0.985
 that-0.782
rous-0.757
eday-0.757
und-0.718
Token new
Token ID649
Feature activation+2.39e+00

Pos logit contributions
 eyel+1.790
cel+1.669
 moral+1.621
 fore+1.599
quin+1.548
Neg logit contributions
ised-2.405
 that-2.068
rous-1.920
aw-1.905
amine-1.844
Token paper
Token ID3348
Feature activation+3.62e-01

Pos logit contributions
cel+3.025
 eyel+3.009
 fore+2.905
quin+2.880
 moral+2.786
Neg logit contributions
ised-4.217
 that-3.719
rous-3.271
eday-3.213
aw-3.211
Token.
Token ID13
Feature activation-4.32e-01

Pos logit contributions
 eyel+0.385
cel+0.357
 moral+0.336
 fore+0.327
 floor+0.322
Neg logit contributions
ised-0.728
eday-0.607
 that-0.606
rous-0.602
und-0.578
Token She
Token ID1375
Feature activation+9.84e-01

Pos logit contributions
ised+0.765
eday+0.627
rous+0.619
 that+0.584
und+0.577
Neg logit contributions
 eyel-0.584
cel-0.546
 moral-0.527
 floor-0.517
 fore-0.503
Token said
Token ID531
Feature activation-5.65e-01

Pos logit contributions
 eyel+1.386
cel+1.313
 moral+1.229
 fore+1.212
nob+1.208
Neg logit contributions
ised-1.664
 that-1.359
eday-1.316
rous-1.306
aw-1.263
Token,
Token ID11
Feature activation+4.13e-01

Pos logit contributions
ised+1.013
eday+0.834
rous+0.821
und+0.774
Deal+0.753
Neg logit contributions
 eyel-0.755
cel-0.692
 moral-0.677
 fore-0.653
 floor-0.652
Token loved
Token ID6151
Feature activation-1.04e+00

Pos logit contributions
 eyel+0.930
cel+0.886
 moral+0.854
 fore+0.839
 floor+0.835
Neg logit contributions
ised-0.900
 that-0.714
eday-0.700
rous-0.693
und-0.651
Token the
Token ID262
Feature activation+9.34e-01

Pos logit contributions
ised+1.711
eday+1.303
rous+1.255
und+1.228
aw+1.192
Neg logit contributions
 eyel-1.590
 floor-1.475
cel-1.474
 moral-1.448
 fore-1.425
Token color
Token ID3124
Feature activation+5.58e-01

Pos logit contributions
 eyel+1.208
cel+1.181
quin+1.080
 moral+1.076
 fore+1.074
Neg logit contributions
ised-1.633
 that-1.359
eday-1.350
rous-1.347
aw-1.270
Token yellow
Token ID7872
Feature activation+8.39e-01

Pos logit contributions
 eyel+0.748
cel+0.713
 moral+0.677
 fore+0.662
quin+0.655
Neg logit contributions
ised-0.974
 that-0.801
eday-0.785
rous-0.782
aw-0.736
Token because
Token ID780
Feature activation+6.49e-01

Pos logit contributions
 eyel+1.007
cel+0.991
 moral+0.921
 fore+0.911
nob+0.882
Neg logit contributions
ised-1.564
 that-1.310
eday-1.285
rous-1.260
aw-1.230
Token it
Token ID340
Feature activation+1.90e+00

Pos logit contributions
 eyel+1.133
cel+1.083
 moral+1.024
 fore+1.021
 floor+1.012
Neg logit contributions
ised-0.881
 that-0.690
eday-0.675
rous-0.670
aw-0.599
Token was
Token ID373
Feature activation-1.00e-01

Pos logit contributions
 eyel+2.677
 moral+2.488
quin+2.424
cel+2.390
 fore+2.304
Neg logit contributions
ised-3.214
 that-2.779
eday-2.555
rous-2.537
amine-2.438
Token so
Token ID523
Feature activation+5.58e-03

Pos logit contributions
ised+0.186
eday+0.152
rous+0.150
 that+0.146
und+0.144
Neg logit contributions
 eyel-0.126
cel-0.117
 moral-0.112
 floor-0.111
 fore-0.109
Token bright
Token ID6016
Feature activation-1.01e+00

Pos logit contributions
 eyel+0.008
cel+0.008
 moral+0.008
 floor+0.008
 fore+0.007
Neg logit contributions
ised-0.009
eday-0.007
rous-0.007
 that-0.007
und-0.007
Token and
Token ID290
Feature activation+3.12e-01

Pos logit contributions
ised+2.089
eday+1.777
rous+1.771
und+1.688
Fine+1.668
Neg logit contributions
 eyel-1.011
cel-0.929
 floor-0.910
 moral-0.853
 fore-0.835
Token happy
Token ID3772
Feature activation+2.32e-01

Pos logit contributions
 eyel+0.445
cel+0.419
 moral+0.402
 fore+0.393
 floor+0.392
Neg logit contributions
ised-0.522
eday-0.417
rous-0.415
 that-0.412
und-0.387
Token little
Token ID1310
Feature activation-1.88e-01

Pos logit contributions
 eyel+0.503
cel+0.475
 moral+0.443
 fore+0.438
 floor+0.437
Neg logit contributions
ised-0.646
eday-0.518
rous-0.514
 that-0.512
und-0.481
Token girl
Token ID2576
Feature activation+1.11e-01

Pos logit contributions
ised+0.356
eday+0.296
rous+0.294
 that+0.287
und+0.282
Neg logit contributions
 eyel-0.227
cel-0.209
 moral-0.202
 floor-0.199
 fore-0.194
Token was
Token ID373
Feature activation-2.12e-01

Pos logit contributions
 eyel+0.127
cel+0.118
 moral+0.112
 floor+0.110
 fore+0.109
Neg logit contributions
ised-0.216
eday-0.179
rous-0.178
 that-0.175
und-0.169
Token so
Token ID523
Feature activation-9.26e-01

Pos logit contributions
ised+0.366
eday+0.295
rous+0.292
 that+0.281
und+0.278
Neg logit contributions
 eyel-0.294
cel-0.274
 moral-0.264
 fore-0.262
 floor-0.261
Token happy
Token ID3772
Feature activation-6.91e-01

Pos logit contributions
ised+1.759
eday+1.499
rous+1.438
und+1.413
 that+1.384
Neg logit contributions
 eyel-1.105
 moral-1.009
 fore-0.995
cel-0.958
 floor-0.940
Token that
Token ID326
Feature activation+1.75e+00

Pos logit contributions
ised+1.293
eday+1.073
rous+1.073
amine+1.000
und+0.997
Neg logit contributions
 eyel-0.852
cel-0.776
 floor-0.752
 moral-0.745
 fore-0.730
Token she
Token ID673
Feature activation+2.87e-01

Pos logit contributions
 eyel+2.843
cel+2.767
nob+2.608
quin+2.559
 fore+2.557
Neg logit contributions
ised-2.589
 that-2.374
eday-2.119
rous-2.081
aw-1.831
Token quickly
Token ID2952
Feature activation-2.35e-01

Pos logit contributions
 eyel+0.362
cel+0.338
 moral+0.323
 floor+0.315
 fore+0.313
Neg logit contributions
ised-0.527
eday-0.431
rous-0.427
 that-0.425
und-0.406
Token got
Token ID1392
Feature activation-9.16e-02

Pos logit contributions
ised+0.441
eday+0.365
rous+0.363
 that+0.348
und+0.340
Neg logit contributions
 eyel-0.288
cel-0.268
 moral-0.253
 fore-0.252
 floor-0.252
Token in
Token ID287
Feature activation-2.07e-01

Pos logit contributions
ised+0.136
eday+0.106
rous+0.103
 that+0.099
und+0.098
Neg logit contributions
 eyel-0.148
cel-0.140
 moral-0.136
 fore-0.134
 floor-0.133
Token the
Token ID262
Feature activation-4.39e-01

Pos logit contributions
ised+0.377
eday+0.307
rous+0.304
 that+0.292
und+0.287
Neg logit contributions
 eyel-0.270
cel-0.250
 moral-0.242
 fore-0.239
 floor-0.238
Tokenoons
Token ID13022
Feature activation+8.09e-01

Pos logit contributions
ised+0.630
eday+0.542
rous+0.534
 that+0.524
und+0.510
Neg logit contributions
 eyel-0.260
cel-0.230
 floor-0.221
 moral-0.220
 fore-0.213
Token.
Token ID13
Feature activation-8.92e-01

Pos logit contributions
 eyel+1.004
cel+0.963
 moral+0.875
quin+0.850
 fore+0.845
Neg logit contributions
ised-1.514
 that-1.279
eday-1.240
rous-1.234
aw-1.168
Token Mom
Token ID11254
Feature activation+2.34e-02

Pos logit contributions
ised+1.719
eday+1.432
rous+1.411
und+1.340
 that+1.309
Neg logit contributions
 eyel-1.110
cel-1.001
 floor-0.986
 moral-0.974
 py-0.927
Token gives
Token ID3607
Feature activation-3.71e-01

Pos logit contributions
 eyel+0.029
cel+0.027
 moral+0.025
 floor+0.025
 fore+0.025
Neg logit contributions
ised-0.044
eday-0.036
rous-0.036
 that-0.035
und-0.034
Token them
Token ID606
Feature activation+7.70e-01

Pos logit contributions
ised+0.715
eday+0.584
rous+0.579
 that+0.560
und+0.560
Neg logit contributions
 eyel-0.446
cel-0.407
 moral-0.396
 floor-0.391
 fore-0.387
Token each
Token ID1123
Feature activation+1.69e+00

Pos logit contributions
 eyel+1.270
cel+1.231
 moral+1.155
 floor+1.147
quin+1.146
Neg logit contributions
ised-1.095
 that-0.879
rous-0.855
eday-0.854
aw-0.793
Token a
Token ID257
Feature activation+3.06e-01

Pos logit contributions
cel+2.674
 eyel+2.622
nob+2.366
 hallway+2.363
quin+2.294
Neg logit contributions
ised-2.560
 that-2.303
rous-2.068
aw-2.015
eday-2.002
Token bowl
Token ID9396
Feature activation+2.70e-01

Pos logit contributions
 eyel+0.377
cel+0.355
 moral+0.333
 floor+0.327
 fore+0.327
Neg logit contributions
ised-0.567
eday-0.464
rous-0.463
 that-0.459
und-0.438
Token of
Token ID286
Feature activation+3.63e-01

Pos logit contributions
 eyel+0.333
cel+0.314
 moral+0.297
 floor+0.289
 fore+0.288
Neg logit contributions
ised-0.499
eday-0.411
rous-0.407
 that-0.406
und-0.387
Token soup
Token ID17141
Feature activation+1.70e-01

Pos logit contributions
 eyel+0.457
cel+0.434
 moral+0.406
 fore+0.398
 floor+0.396
Neg logit contributions
ised-0.666
eday-0.544
rous-0.544
 that-0.539
und-0.510
Token.
Token ID13
Feature activation-5.63e-01

Pos logit contributions
 eyel+0.189
cel+0.176
 moral+0.166
 floor+0.162
 fore+0.161
Neg logit contributions
ised-0.337
eday-0.280
rous-0.278
 that-0.275
und-0.265
Token problems
Token ID2761
Feature activation-1.66e-01

Pos logit contributions
ised+0.789
rous+0.672
eday+0.671
 that+0.634
und+0.634
Neg logit contributions
 eyel-0.366
cel-0.338
 moral-0.319
 floor-0.319
 fore-0.313
Token.
Token ID13
Feature activation+4.63e-01

Pos logit contributions
ised+0.348
eday+0.294
rous+0.294
 that+0.282
und+0.278
Neg logit contributions
 eyel-0.162
cel-0.149
 moral-0.140
 floor-0.138
 fore-0.138
Token
Token ID198
Feature activation-1.40e-01

Pos logit contributions
 eyel+0.636
cel+0.601
 moral+0.562
 fore+0.559
 floor+0.553
Neg logit contributions
ised-0.798
eday-0.651
rous-0.640
 that-0.637
und-0.608
Token
Token ID198
Feature activation+7.42e-01

Pos logit contributions
ised+0.261
eday+0.215
rous+0.213
 that+0.207
und+0.201
Neg logit contributions
 eyel-0.170
cel-0.159
 moral-0.156
 floor-0.150
 fore-0.149
TokenOne
Token ID3198
Feature activation+4.19e-01

Pos logit contributions
 eyel+1.041
cel+0.979
 magn+0.901
 floor+0.892
 fore+0.891
Neg logit contributions
ised-1.288
eday-1.058
rous-1.033
 that-1.033
aw-0.989
Token day
Token ID1110
Feature activation+5.93e-01

Pos logit contributions
 eyel+0.604
cel+0.575
 moral+0.533
 fore+0.529
 floor+0.529
Neg logit contributions
ised-0.706
eday-0.565
rous-0.561
 that-0.561
und-0.522
Token,
Token ID11
Feature activation-5.86e-01

Pos logit contributions
 eyel+0.903
cel+0.867
 moral+0.821
 floor+0.811
 fore+0.808
Neg logit contributions
ised-0.941
 that-0.749
eday-0.740
rous-0.724
und-0.692
Token a
Token ID257
Feature activation-2.79e-01

Pos logit contributions
ised+0.982
rous+0.792
eday+0.781
und+0.748
 that+0.725
Neg logit contributions
 eyel-0.821
 moral-0.754
cel-0.751
 floor-0.740
 fore-0.724
Token lady
Token ID10846
Feature activation-1.02e-01

Pos logit contributions
ised+0.502
eday+0.416
rous+0.407
und+0.392
 that+0.390
Neg logit contributions
 eyel-0.364
cel-0.340
 moral-0.333
 fore-0.323
 floor-0.323
Token came
Token ID1625
Feature activation+6.07e-01

Pos logit contributions
ised+0.204
eday+0.170
rous+0.169
 that+0.164
und+0.159
Neg logit contributions
 eyel-0.112
cel-0.105
 moral-0.098
 fore-0.096
 floor-0.096
Token to
Token ID284
Feature activation+3.14e-02

Pos logit contributions
 eyel+1.102
cel+1.067
 moral+1.016
 floor+1.002
 fore+0.990
Neg logit contributions
ised-0.783
rous-0.578
 that-0.578
eday-0.576
und-0.537
Token was
Token ID373
Feature activation-5.49e-01

Pos logit contributions
ised+0.447
eday+0.369
rous+0.366
 that+0.349
und+0.343
Neg logit contributions
 eyel-0.329
cel-0.309
 moral-0.298
 fore-0.290
 floor-0.289
Token happy
Token ID3772
Feature activation-1.14e+00

Pos logit contributions
ised+1.039
eday+0.856
rous+0.843
und+0.802
 that+0.797
Neg logit contributions
 eyel-0.686
cel-0.624
 moral-0.598
 fore-0.598
 floor-0.594
Token to
Token ID284
Feature activation+9.37e-01

Pos logit contributions
ised+2.029
eday+1.653
rous+1.637
Deal+1.539
und+1.535
Neg logit contributions
 eyel-1.534
cel-1.409
 floor-1.408
 fore-1.370
 moral-1.362
Token see
Token ID766
Feature activation-6.44e-02

Pos logit contributions
 eyel+1.639
cel+1.571
 floor+1.487
 fore+1.476
 moral+1.474
Neg logit contributions
ised-1.244
eday-0.985
 that-0.984
rous-0.943
und-0.866
Token the
Token ID262
Feature activation+9.08e-02

Pos logit contributions
ised+0.109
eday+0.088
rous+0.087
 that+0.084
und+0.082
Neg logit contributions
 eyel-0.090
cel-0.085
 moral-0.082
 floor-0.080
 fore-0.080
Token child
Token ID1200
Feature activation+1.16e-01

Pos logit contributions
 eyel+0.120
cel+0.112
 moral+0.107
 floor+0.105
 fore+0.105
Neg logit contributions
ised-0.164
eday-0.134
rous-0.133
 that-0.130
und-0.126
Token,
Token ID11
Feature activation+2.05e-01

Pos logit contributions
 eyel+0.136
cel+0.127
 moral+0.120
 floor+0.118
 fore+0.117
Neg logit contributions
ised-0.221
eday-0.183
rous-0.181
 that-0.178
und-0.172
Token so
Token ID523
Feature activation-1.18e+00

Pos logit contributions
 eyel+0.282
cel+0.266
 moral+0.254
 floor+0.249
 fore+0.248
Neg logit contributions
ised-0.353
eday-0.285
rous-0.283
 that-0.279
und-0.267
Token he
Token ID339
Feature activation+1.50e-01

Pos logit contributions
ised+2.713
eday+2.360
rous+2.359
und+2.259
amine+2.222
Neg logit contributions
 eyel-0.950
 moral-0.871
 other-0.811
cel-0.798
 cor-0.794
Token waved
Token ID26834
Feature activation-3.10e-01

Pos logit contributions
 eyel+0.201
cel+0.189
 moral+0.181
 floor+0.177
 fore+0.177
Neg logit contributions
ised-0.263
eday-0.213
rous-0.212
 that-0.208
und-0.200
Token at
Token ID379
Feature activation-6.93e-01

Pos logit contributions
ised+0.498
eday+0.399
rous+0.389
und+0.372
 that+0.369
Neg logit contributions
 eyel-0.467
cel-0.432
 moral-0.423
 fore-0.417
 floor-0.408
Token back
Token ID736
Feature activation+2.12e-01

Pos logit contributions
 eyel+0.920
cel+0.892
 moral+0.814
 floor+0.802
quin+0.802
Neg logit contributions
ised-1.103
 that-0.904
eday-0.887
rous-0.879
aw-0.827
Token.
Token ID13
Feature activation-8.02e-01

Pos logit contributions
 eyel+0.279
cel+0.263
 moral+0.250
 floor+0.244
 fore+0.244
Neg logit contributions
ised-0.378
eday-0.308
rous-0.305
 that-0.301
und-0.289
Token But
Token ID887
Feature activation+9.85e-01

Pos logit contributions
ised+1.562
rous+1.320
eday+1.304
amine+1.198
und+1.195
Neg logit contributions
 eyel-0.938
cel-0.861
 moral-0.844
 floor-0.842
 fore-0.809
Token the
Token ID262
Feature activation+2.36e-01

Pos logit contributions
 eyel+1.588
cel+1.534
 moral+1.418
quin+1.414
 floor+1.403
Neg logit contributions
ised-1.491
 that-1.252
eday-1.186
rous-1.133
aw-1.066
Token balls
Token ID11333
Feature activation-7.78e-01

Pos logit contributions
 eyel+0.335
cel+0.317
 moral+0.301
 fore+0.296
 floor+0.295
Neg logit contributions
ised-0.398
eday-0.319
rous-0.316
 that-0.313
und-0.298
Token are
Token ID389
Feature activation+7.16e-02

Pos logit contributions
ised+1.565
rous+1.310
eday+1.292
und+1.210
amine+1.200
Neg logit contributions
 eyel-0.849
 moral-0.782
 floor-0.765
 fore-0.757
cel-0.748
Token too
Token ID1165
Feature activation-4.22e-01

Pos logit contributions
 eyel+0.092
cel+0.086
 moral+0.082
 floor+0.080
 fore+0.080
Neg logit contributions
ised-0.130
eday-0.106
rous-0.106
 that-0.103
und-0.100
Token many
Token ID867
Feature activation-2.00e-01

Pos logit contributions
ised+0.693
eday+0.571
rous+0.555
 that+0.529
und+0.526
Neg logit contributions
 eyel-0.601
 moral-0.566
cel-0.556
 fore-0.550
 floor-0.548
Token.
Token ID13
Feature activation-2.12e-01

Pos logit contributions
ised+0.418
eday+0.354
rous+0.351
 that+0.336
und+0.333
Neg logit contributions
 eyel-0.202
cel-0.186
 moral-0.178
 floor-0.175
 fore-0.173
Token They
Token ID1119
Feature activation-9.21e-01

Pos logit contributions
ised+0.369
eday+0.301
rous+0.300
 that+0.288
und+0.278
Neg logit contributions
 eyel-0.286
cel-0.270
 moral-0.261
 floor-0.255
 fore-0.254
Token fall
Token ID2121
Feature activation-4.22e-01

Pos logit contributions
ised+1.828
rous+1.502
eday+1.473
aw+1.395
und+1.392
Neg logit contributions
 eyel-1.090
cel-0.995
 floor-0.964
 moral-0.959
 magn-0.926
Token now
Token ID783
Feature activation-6.12e-01

Pos logit contributions
 eyel+0.702
cel+0.663
 moral+0.632
 fore+0.617
 floor+0.612
Neg logit contributions
ised-0.786
eday-0.630
 that-0.626
rous-0.618
und-0.586
Token."
Token ID526
Feature activation-7.52e-01

Pos logit contributions
ised+1.267
rous+1.082
eday+1.053
aw+1.008
und+0.994
Neg logit contributions
 eyel-0.626
cel-0.591
 floor-0.557
 moral-0.546
 fore-0.530
Token Tim
Token ID5045
Feature activation-2.01e-01

Pos logit contributions
ised+1.275
rous+1.044
eday+1.029
aw+0.956
und+0.935
Neg logit contributions
 eyel-1.096
cel-1.035
 floor-0.989
 moral-0.946
nob-0.929
Tokenmy
Token ID1820
Feature activation+7.62e-01

Pos logit contributions
ised+0.467
rous+0.399
eday+0.399
 that+0.392
und+0.381
Neg logit contributions
 eyel-0.158
cel-0.141
 moral-0.132
 floor-0.127
 fore-0.125
Token said
Token ID531
Feature activation+8.68e-01

Pos logit contributions
 eyel+1.100
cel+1.051
 moral+0.988
quin+0.977
 floor+0.971
Neg logit contributions
ised-1.256
 that-1.000
eday-0.988
rous-0.965
aw-0.926
Token,
Token ID11
Feature activation+9.28e-01

Pos logit contributions
 eyel+1.294
cel+1.259
 moral+1.202
 floor+1.187
quin+1.183
Neg logit contributions
ised-1.373
 that-1.132
eday-1.069
rous-1.028
aw-0.980
Token "
Token ID366
Feature activation-2.51e-01

Pos logit contributions
 eyel+1.170
cel+1.091
 moral+1.053
nob+1.027
 fore+1.024
Neg logit contributions
ised-1.665
 that-1.418
eday-1.359
rous-1.302
und-1.266
TokenOkay
Token ID16454
Feature activation-4.56e-01

Pos logit contributions
ised+0.333
eday+0.247
rous+0.247
 that+0.239
und+0.229
Neg logit contributions
 eyel-0.447
cel-0.427
 moral-0.415
 floor-0.407
 fore-0.406
Token,
Token ID11
Feature activation-1.76e-01

Pos logit contributions
ised+0.819
rous+0.679
eday+0.662
und+0.638
aw+0.632
Neg logit contributions
 eyel-0.604
cel-0.570
 moral-0.542
 floor-0.534
 fore-0.519
Token mom
Token ID1995
Feature activation-8.87e-02

Pos logit contributions
ised+0.339
rous+0.280
eday+0.279
 that+0.266
und+0.266
Neg logit contributions
 eyel-0.207
cel-0.194
 moral-0.186
 floor-0.181
 fore-0.179
Tokenmy
Token ID1820
Feature activation-8.50e-01

Pos logit contributions
ised+0.204
rous+0.174
eday+0.173
 that+0.169
und+0.166
Neg logit contributions
 eyel-0.071
cel-0.065
 moral-0.060
 floor-0.058
 fore-0.057
Token block
Token ID2512
Feature activation+8.04e-01

Pos logit contributions
cel+1.916
 eyel+1.912
 fore+1.748
Beck+1.722
 moral+1.681
Neg logit contributions
ised-2.416
 that-2.054
eday-1.958
rous-1.917
aw-1.869
Token.
Token ID13
Feature activation-3.40e-01

Pos logit contributions
 eyel+1.073
cel+1.058
quin+0.960
 moral+0.958
 floor+0.934
Neg logit contributions
ised-1.394
 that-1.174
eday-1.127
rous-1.115
aw-1.076
Token They
Token ID1119
Feature activation-1.02e-02

Pos logit contributions
ised+0.609
rous+0.496
eday+0.495
 that+0.470
und+0.463
Neg logit contributions
 eyel-0.455
cel-0.422
 moral-0.411
 floor-0.398
 fore-0.396
Token did
Token ID750
Feature activation-1.89e-01

Pos logit contributions
ised+0.017
eday+0.014
rous+0.014
 that+0.014
und+0.013
Neg logit contributions
 eyel-0.014
cel-0.013
 moral-0.013
 floor-0.012
 fore-0.012
Token not
Token ID407
Feature activation+1.23e-01

Pos logit contributions
ised+0.408
eday+0.345
rous+0.343
und+0.331
 that+0.330
Neg logit contributions
 eyel-0.182
cel-0.162
 moral-0.159
 floor-0.151
 fore-0.150
Token fight
Token ID1907
Feature activation+3.41e-01

Pos logit contributions
 eyel+0.186
cel+0.176
 moral+0.169
 floor+0.166
 fore+0.166
Neg logit contributions
ised-0.197
eday-0.156
rous-0.155
 that-0.151
und-0.145
Token anymore
Token ID7471
Feature activation+1.49e-01

Pos logit contributions
 eyel+0.457
cel+0.434
 moral+0.406
 floor+0.399
 fore+0.399
Neg logit contributions
ised-0.599
eday-0.485
 that-0.482
rous-0.481
und-0.455
Token.
Token ID13
Feature activation+8.21e-02

Pos logit contributions
 eyel+0.154
cel+0.142
 moral+0.133
 floor+0.130
 fore+0.129
Neg logit contributions
ised-0.309
eday-0.259
rous-0.258
 that-0.254
und-0.246
Token They
Token ID1119
Feature activation-2.08e-01

Pos logit contributions
 eyel+0.113
cel+0.107
 moral+0.102
 floor+0.100
 fore+0.100
Neg logit contributions
ised-0.142
eday-0.115
rous-0.114
 that-0.111
und-0.107
Token played
Token ID2826
Feature activation-8.28e-01

Pos logit contributions
ised+0.376
rous+0.306
eday+0.306
 that+0.294
und+0.289
Neg logit contributions
 eyel-0.269
cel-0.250
 moral-0.242
 fore-0.236
 floor-0.235
Token with
Token ID351
Feature activation-3.64e-01

Pos logit contributions
ised+1.460
eday+1.218
rous+1.203
und+1.129
amine+1.120
Neg logit contributions
 eyel-1.088
 moral-1.028
 floor-1.015
cel-0.998
 fore-0.975
Token love
Token ID1842
Feature activation-7.43e-01

Pos logit contributions
 eyel+0.811
cel+0.768
 moral+0.716
 floor+0.706
quin+0.705
Neg logit contributions
ised-0.909
eday-0.734
 that-0.727
rous-0.724
und-0.680
Token,
Token ID11
Feature activation-4.36e-01

Pos logit contributions
ised+1.340
rous+1.109
eday+1.096
aw+1.026
und+1.025
Neg logit contributions
 eyel-0.921
cel-0.885
 moral-0.857
 floor-0.848
 fore-0.837
Token it
Token ID340
Feature activation-3.40e-02

Pos logit contributions
ised+0.751
rous+0.610
eday+0.605
und+0.572
aw+0.566
Neg logit contributions
 eyel-0.582
cel-0.559
 moral-0.552
 floor-0.532
 fore-0.529
Token's
Token ID338
Feature activation-5.76e-01

Pos logit contributions
ised+0.067
eday+0.056
rous+0.056
 that+0.054
und+0.053
Neg logit contributions
 eyel-0.038
cel-0.035
 moral-0.033
 fore-0.032
 floor-0.032
Token important
Token ID1593
Feature activation-8.50e-01

Pos logit contributions
ised+1.044
rous+0.842
eday+0.838
und+0.810
aw+0.801
Neg logit contributions
 eyel-0.736
 moral-0.688
cel-0.684
 fore-0.674
 floor-0.656
Token to
Token ID284
Feature activation-1.92e-01

Pos logit contributions
ised+1.279
rous+1.011
eday+0.971
und+0.923
aw+0.922
Neg logit contributions
 eyel-1.358
 moral-1.264
cel-1.262
 fore-1.250
 floor-1.245
Token be
Token ID307
Feature activation-2.31e-01

Pos logit contributions
ised+0.354
rous+0.287
eday+0.287
 that+0.277
und+0.272
Neg logit contributions
 eyel-0.240
cel-0.225
 moral-0.218
 fore-0.212
 floor-0.210
Token kind
Token ID1611
Feature activation-1.34e+00

Pos logit contributions
ised+0.379
eday+0.305
rous+0.298
 that+0.289
und+0.284
Neg logit contributions
 eyel-0.331
cel-0.314
 moral-0.303
 fore-0.301
 floor-0.295
Token and
Token ID290
Feature activation-7.19e-01

Pos logit contributions
ised+2.531
rous+2.185
eday+2.123
und+1.996
amine+1.985
Neg logit contributions
 eyel-1.519
 moral-1.429
 floor-1.404
 fore-1.401
cel-1.367
Token generous
Token ID14431
Feature activation-1.10e+00

Pos logit contributions
ised+1.252
eday+0.994
rous+0.993
und+0.973
aw+0.960
Neg logit contributions
 eyel-0.985
 moral-0.918
cel-0.906
 fore-0.903
 floor-0.869
Token,"
Token ID553
Feature activation-4.02e-01

Pos logit contributions
ised+1.990
rous+1.687
eday+1.634
und+1.570
amine+1.533
Neg logit contributions
 eyel-1.359
 moral-1.264
cel-1.236
 fore-1.235
 floor-1.226
Token hug
Token ID16225
Feature activation+1.19e+00

Pos logit contributions
ised+0.550
rous+0.450
eday+0.448
 that+0.432
und+0.427
Neg logit contributions
 eyel-0.342
cel-0.317
 moral-0.308
 fore-0.299
 floor-0.296
Token their
Token ID511
Feature activation+8.44e-01

Pos logit contributions
 eyel+1.732
cel+1.717
 magn+1.565
nob+1.564
quin+1.558
Neg logit contributions
ised-2.004
 that-1.645
eday-1.589
rous-1.522
aw-1.490
Token mom
Token ID1995
Feature activation+8.81e-01

Pos logit contributions
 eyel+0.989
cel+0.954
nob+0.862
quin+0.850
 fore+0.842
Neg logit contributions
ised-1.625
eday-1.335
 that-1.333
rous-1.321
aw-1.237
Token again
Token ID757
Feature activation+4.27e-01

Pos logit contributions
 eyel+1.151
cel+1.089
 moral+1.022
 floor+1.010
quin+1.000
Neg logit contributions
ised-1.564
 that-1.316
eday-1.276
rous-1.239
aw-1.227
Token and
Token ID290
Feature activation+2.54e-01

Pos logit contributions
 eyel+0.533
cel+0.499
 moral+0.465
 floor+0.460
 fore+0.455
Neg logit contributions
ised-0.795
 that-0.648
eday-0.647
rous-0.642
und-0.611
Token go
Token ID467
Feature activation-8.28e-01

Pos logit contributions
 eyel+0.338
cel+0.319
 moral+0.302
 floor+0.296
 fore+0.294
Neg logit contributions
ised-0.449
eday-0.366
rous-0.362
 that-0.359
und-0.341
Token to
Token ID284
Feature activation-1.71e-01

Pos logit contributions
ised+1.558
eday+1.259
rous+1.251
und+1.185
izing+1.178
Neg logit contributions
 eyel-1.026
cel-0.976
 fore-0.931
 moral-0.927
 floor-0.919
Token fix
Token ID4259
Feature activation+1.53e-01

Pos logit contributions
ised+0.308
eday+0.248
rous+0.246
 that+0.238
aw+0.236
Neg logit contributions
 eyel-0.226
cel-0.211
 moral-0.204
 floor-0.201
 fore-0.198
Token their
Token ID511
Feature activation+3.81e-01

Pos logit contributions
 eyel+0.197
cel+0.185
 moral+0.176
 floor+0.172
 fore+0.172
Neg logit contributions
ised-0.275
eday-0.225
rous-0.223
 that-0.220
und-0.211
Token cars
Token ID5006
Feature activation+3.24e-01

Pos logit contributions
 eyel+0.406
cel+0.377
 moral+0.345
 fore+0.339
 floor+0.336
Neg logit contributions
ised-0.777
eday-0.648
rous-0.644
 that-0.637
und-0.612
Token.
Token ID13
Feature activation+1.77e-01

Pos logit contributions
 eyel+0.366
cel+0.340
 moral+0.319
 fore+0.309
 floor+0.309
Neg logit contributions
ised-0.640
eday-0.532
 that-0.527
rous-0.525
und-0.503
Token,
Token ID11
Feature activation+2.15e+00

Pos logit contributions
 eyel+0.836
cel+0.778
 moral+0.751
 fore+0.736
 floor+0.726
Neg logit contributions
ised-0.981
 that-0.808
eday-0.781
rous-0.773
und-0.741
Token stars
Token ID5788
Feature activation+2.38e-01

Pos logit contributions
 eyel+2.886
cel+2.705
 fore+2.614
nob+2.580
any+2.542
Neg logit contributions
ised-3.801
 that-3.433
eday-3.033
rous-3.024
aw-2.997
Token,
Token ID11
Feature activation+1.97e+00

Pos logit contributions
 eyel+0.325
cel+0.305
 moral+0.292
 fore+0.286
 floor+0.285
Neg logit contributions
ised-0.410
eday-0.331
rous-0.327
 that-0.326
und-0.311
Token and
Token ID290
Feature activation+1.24e+00

Pos logit contributions
 eyel+2.759
cel+2.621
nob+2.474
 fore+2.439
any+2.429
Neg logit contributions
ised-3.302
 that-2.939
aw-2.695
eday-2.615
rous-2.568
Token animals
Token ID4695
Feature activation-3.50e-01

Pos logit contributions
 eyel+1.504
cel+1.478
nob+1.356
 fore+1.328
 moral+1.327
Neg logit contributions
ised-2.257
 that-1.953
rous-1.856
eday-1.836
aw-1.774
Token.
Token ID13
Feature activation-3.16e-01

Pos logit contributions
ised+0.752
eday+0.635
rous+0.631
 that+0.597
und+0.596
Neg logit contributions
 eyel-0.335
cel-0.311
 moral-0.288
 floor-0.287
 fore-0.279
Token They
Token ID1119
Feature activation+4.34e-01

Pos logit contributions
ised+0.586
eday+0.481
rous+0.477
 that+0.464
und+0.455
Neg logit contributions
 eyel-0.396
cel-0.362
 moral-0.357
 floor-0.348
 fore-0.336
Token used
Token ID973
Feature activation-7.14e-01

Pos logit contributions
 eyel+0.560
cel+0.527
 moral+0.499
 fore+0.493
 floor+0.486
Neg logit contributions
ised-0.781
eday-0.638
 that-0.632
rous-0.632
und-0.598
Token many
Token ID867
Feature activation+9.11e-01

Pos logit contributions
ised+1.402
eday+1.191
rous+1.145
Deal+1.119
Fine+1.116
Neg logit contributions
 eyel-0.831
cel-0.729
 floor-0.728
 moral-0.721
 fore-0.716
Token colors
Token ID7577
Feature activation-2.89e-01

Pos logit contributions
 eyel+1.084
cel+1.025
 moral+0.954
nob+0.938
 fore+0.932
Neg logit contributions
ised-1.741
 that-1.429
rous-1.422
eday-1.373
aw-1.350
Token,
Token ID11
Feature activation+1.00e+00

Pos logit contributions
ised+0.504
eday+0.412
rous+0.408
und+0.381
 that+0.380
Neg logit contributions
 eyel-0.389
cel-0.361
 moral-0.349
 floor-0.347
 fore-0.342
Token she
Token ID673
Feature activation+1.72e-01

Pos logit contributions
 eyel+1.471
cel+1.405
 moral+1.311
 fore+1.309
quin+1.303
Neg logit contributions
ised-1.731
 that-1.473
eday-1.414
rous-1.386
und-1.268
Token opened
Token ID4721
Feature activation-1.76e-01

Pos logit contributions
 eyel+0.214
cel+0.200
 moral+0.191
 floor+0.186
 fore+0.186
Neg logit contributions
ised-0.319
eday-0.262
rous-0.260
 that-0.256
und-0.246
Token the
Token ID262
Feature activation+3.94e-01

Pos logit contributions
ised+0.310
eday+0.253
rous+0.249
 that+0.240
und+0.237
Neg logit contributions
 eyel-0.234
cel-0.219
 moral-0.210
 floor-0.207
 fore-0.205
Token box
Token ID3091
Feature activation-4.40e-01

Pos logit contributions
 eyel+0.517
cel+0.486
 moral+0.460
 fore+0.450
 floor+0.450
Neg logit contributions
ised-0.704
eday-0.571
rous-0.569
 that-0.565
und-0.535
Token and
Token ID290
Feature activation+1.35e-01

Pos logit contributions
ised+0.848
eday+0.714
rous+0.707
und+0.668
 that+0.664
Neg logit contributions
 eyel-0.503
cel-0.462
 floor-0.449
 moral-0.442
 fore-0.433
Token inside
Token ID2641
Feature activation-2.56e-01

Pos logit contributions
 eyel+0.183
cel+0.172
 moral+0.165
 floor+0.162
 fore+0.161
Neg logit contributions
ised-0.235
eday-0.190
rous-0.189
 that-0.186
und-0.178
Token,
Token ID11
Feature activation+3.47e-01

Pos logit contributions
ised+0.487
eday+0.405
rous+0.401
 that+0.381
und+0.380
Neg logit contributions
 eyel-0.304
cel-0.283
 moral-0.269
 floor-0.266
 fore-0.262
Token there
Token ID612
Feature activation+1.47e-01

Pos logit contributions
 eyel+0.468
cel+0.440
 moral+0.423
 fore+0.411
 floor+0.410
Neg logit contributions
ised-0.606
eday-0.491
rous-0.486
 that-0.485
und-0.459
Token was
Token ID373
Feature activation+1.93e-01

Pos logit contributions
 eyel+0.154
cel+0.142
 moral+0.134
 floor+0.131
 fore+0.130
Neg logit contributions
ised-0.300
eday-0.251
rous-0.249
 that-0.246
und-0.238
Token a
Token ID257
Feature activation+1.21e-01

Pos logit contributions
 eyel+0.253
cel+0.238
 moral+0.227
 floor+0.222
 fore+0.222
Neg logit contributions
ised-0.342
eday-0.278
rous-0.276
 that-0.272
und-0.261
Token beautiful
Token ID4950
Feature activation+1.94e-03

Pos logit contributions
 eyel+0.165
cel+0.155
 moral+0.148
 floor+0.145
 fore+0.145
Neg logit contributions
ised-0.210
eday-0.170
rous-0.169
 that-0.166
und-0.159
Token you
Token ID345
Feature activation-1.01e+00

Pos logit contributions
ised+0.298
rous+0.233
eday+0.233
und+0.219
 that+0.218
Neg logit contributions
 eyel-0.285
cel-0.270
 moral-0.261
 floor-0.256
 fore-0.256
Token can
Token ID460
Feature activation-2.86e-01

Pos logit contributions
ised+1.731
eday+1.353
rous+1.347
und+1.264
izing+1.263
Neg logit contributions
 eyel-1.459
cel-1.421
 moral-1.406
 magn-1.313
 floor-1.312
Token draw
Token ID3197
Feature activation+4.48e-01

Pos logit contributions
ised+0.558
eday+0.458
rous+0.457
 that+0.437
und+0.435
Neg logit contributions
 eyel-0.337
cel-0.312
 moral-0.298
 floor-0.290
nob-0.281
Token a
Token ID257
Feature activation+3.26e-01

Pos logit contributions
 eyel+0.636
cel+0.601
 moral+0.574
 fore+0.566
 floor+0.565
Neg logit contributions
ised-0.747
 that-0.598
eday-0.595
rous-0.594
aw-0.561
Token frog
Token ID21264
Feature activation-6.57e-01

Pos logit contributions
 eyel+0.428
cel+0.404
 moral+0.381
 fore+0.377
 floor+0.374
Neg logit contributions
ised-0.575
eday-0.467
rous-0.466
 that-0.462
und-0.440
Token too
Token ID1165
Feature activation-1.01e+00

Pos logit contributions
ised+1.185
rous+0.959
eday+0.957
und+0.894
amine+0.876
Neg logit contributions
 eyel-0.855
cel-0.804
 moral-0.774
 floor-0.769
nob-0.740
Token.
Token ID13
Feature activation+5.34e-02

Pos logit contributions
ised+1.736
eday+1.444
rous+1.444
und+1.319
amine+1.306
Neg logit contributions
 eyel-1.347
 floor-1.244
 moral-1.234
cel-1.229
nob-1.161
Token Here
Token ID3423
Feature activation+2.96e-01

Pos logit contributions
 eyel+0.080
cel+0.076
 moral+0.072
 floor+0.071
 fore+0.071
Neg logit contributions
ised-0.086
eday-0.068
rous-0.067
 that-0.066
und-0.063
Token,
Token ID11
Feature activation-5.31e-01

Pos logit contributions
 eyel+0.448
cel+0.424
 moral+0.407
 floor+0.401
 fore+0.400
Neg logit contributions
ised-0.468
eday-0.371
rous-0.366
 that-0.366
und-0.343
Token take
Token ID1011
Feature activation-6.77e-01

Pos logit contributions
ised+0.963
rous+0.770
eday+0.765
und+0.726
aw+0.722
Neg logit contributions
 eyel-0.695
cel-0.646
 moral-0.641
 fore-0.617
 floor-0.616
Token a
Token ID257
Feature activation-1.09e-01

Pos logit contributions
ised+1.148
eday+0.917
rous+0.895
Deal+0.850
und+0.844
Neg logit contributions
 eyel-0.965
cel-0.896
 moral-0.874
 fore-0.864
 floor-0.856
Token view
Token ID1570
Feature activation+4.46e-01

Pos logit contributions
 eyel+0.958
cel+0.924
 moral+0.852
 fore+0.842
nob+0.841
Neg logit contributions
ised-1.170
rous-0.947
 that-0.933
eday-0.931
aw-0.873
Token.
Token ID13
Feature activation+2.04e-02

Pos logit contributions
 eyel+0.561
cel+0.530
 moral+0.502
 fore+0.487
 floor+0.486
Neg logit contributions
ised-0.820
 that-0.672
eday-0.668
rous-0.665
aw-0.629
Token
Token ID220
Feature activation+2.15e-01

Pos logit contributions
 eyel+0.027
cel+0.025
 moral+0.024
 floor+0.023
 fore+0.023
Neg logit contributions
ised-0.037
eday-0.030
rous-0.030
 that-0.029
und-0.028
Token
Token ID198
Feature activation-7.95e-02

Pos logit contributions
 eyel+0.279
cel+0.261
 moral+0.249
 floor+0.244
 fore+0.243
Neg logit contributions
ised-0.385
eday-0.313
rous-0.311
 that-0.307
und-0.295
Token
Token ID198
Feature activation+2.49e-02

Pos logit contributions
ised+0.147
eday+0.121
rous+0.119
 that+0.117
und+0.113
Neg logit contributions
 eyel-0.099
cel-0.093
 moral-0.090
 floor-0.087
 fore-0.086
TokenFrom
Token ID4863
Feature activation-2.41e+00

Pos logit contributions
 eyel+0.039
cel+0.037
 moral+0.036
 floor+0.035
 fore+0.035
Neg logit contributions
ised-0.038
eday-0.030
rous-0.030
 that-0.029
und-0.028
Token that
Token ID326
Feature activation+7.47e-01

Pos logit contributions
ised+2.354
und+1.854
eday+1.727
rous+1.691
Deal+1.656
Neg logit contributions
 moral-4.759
 eyel-4.655
 fore-4.494
 floor-4.483
 other-4.476
Token day
Token ID1110
Feature activation+6.39e-01

Pos logit contributions
 eyel+1.116
cel+1.065
 floor+0.979
nob+0.978
 moral+0.975
Neg logit contributions
ised-1.204
 that-0.979
eday-0.969
rous-0.958
und-0.895
Token on
Token ID319
Feature activation+1.07e-02

Pos logit contributions
 eyel+0.824
cel+0.771
 floor+0.737
 moral+0.735
 fore+0.718
Neg logit contributions
ised-1.159
 that-0.947
eday-0.941
rous-0.924
und-0.899
Token,
Token ID11
Feature activation+8.61e-02

Pos logit contributions
 eyel+0.015
cel+0.014
 moral+0.014
 floor+0.013
 fore+0.013
Neg logit contributions
ised-0.018
eday-0.015
rous-0.015
 that-0.014
und-0.014
Token the
Token ID262
Feature activation+3.18e-01

Pos logit contributions
 eyel+0.121
cel+0.114
 moral+0.109
 floor+0.107
 fore+0.107
Neg logit contributions
ised-0.146
eday-0.117
rous-0.116
 that-0.114
und-0.109
Token matching
Token ID12336
Feature activation-1.14e-01

Pos logit contributions
ised+0.218
eday+0.178
rous+0.176
 that+0.171
und+0.168
Neg logit contributions
 eyel-0.168
cel-0.156
 moral-0.151
 floor-0.148
 fore-0.147
Token leaves
Token ID5667
Feature activation-2.49e-01

Pos logit contributions
ised+0.212
eday+0.175
rous+0.173
 that+0.169
und+0.165
Neg logit contributions
 eyel-0.142
cel-0.131
 moral-0.126
 floor-0.125
 fore-0.123
Token and
Token ID290
Feature activation+1.48e-01

Pos logit contributions
ised+0.462
eday+0.382
rous+0.379
 that+0.360
und+0.357
Neg logit contributions
 eyel-0.307
cel-0.285
 moral-0.271
 floor-0.270
 fore-0.268
Token flowers
Token ID12734
Feature activation-3.61e-01

Pos logit contributions
 eyel+0.187
cel+0.175
 moral+0.167
 floor+0.163
 fore+0.163
Neg logit contributions
ised-0.271
eday-0.222
rous-0.220
 that-0.217
und-0.209
Token.
Token ID13
Feature activation+3.18e-01

Pos logit contributions
ised+0.722
eday+0.605
rous+0.600
 that+0.567
und+0.566
Neg logit contributions
 eyel-0.395
cel-0.364
 moral-0.341
 floor-0.339
 fore-0.337
Token From
Token ID3574
Feature activation-2.83e+00

Pos logit contributions
 eyel+0.507
cel+0.481
 moral+0.463
 fore+0.456
 floor+0.456
Neg logit contributions
ised-0.477
eday-0.373
rous-0.367
 that-0.364
und-0.344
Token then
Token ID788
Feature activation-1.73e-01

Pos logit contributions
ised+3.613
und+3.242
amine+2.960
unch+2.948
Deal+2.947
Neg logit contributions
 moral-4.667
 eyel-4.503
 other-4.487
 fore-4.396
 floor-4.323
Token on
Token ID319
Feature activation+1.79e-01

Pos logit contributions
ised+0.340
eday+0.282
rous+0.281
 that+0.272
und+0.264
Neg logit contributions
 eyel-0.195
cel-0.183
 moral-0.173
 fore-0.168
 floor-0.166
Token,
Token ID11
Feature activation+6.76e-01

Pos logit contributions
 eyel+0.256
cel+0.241
 moral+0.231
 floor+0.227
 fore+0.226
Neg logit contributions
ised-0.299
eday-0.240
rous-0.238
 that-0.234
und-0.224
Token she
Token ID673
Feature activation+6.72e-01

Pos logit contributions
 eyel+0.935
cel+0.872
 moral+0.827
 fore+0.821
 floor+0.821
Neg logit contributions
ised-1.179
 that-0.960
eday-0.956
rous-0.942
und-0.882
Token loved
Token ID6151
Feature activation-1.53e+00

Pos logit contributions
 eyel+0.942
cel+0.877
 moral+0.838
 floor+0.825
 fore+0.812
Neg logit contributions
ised-1.136
eday-0.928
 that-0.920
rous-0.911
und-0.870
Token his
Token ID465
Feature activation+3.50e-01

Pos logit contributions
 eyel+0.247
cel+0.232
 moral+0.221
 floor+0.217
 fore+0.216
Neg logit contributions
ised-0.338
eday-0.275
rous-0.273
 that-0.270
und-0.258
Token nap
Token ID25422
Feature activation+5.22e-02

Pos logit contributions
 eyel+0.402
cel+0.381
 moral+0.357
 floor+0.347
 fore+0.347
Neg logit contributions
ised-0.673
eday-0.558
rous-0.555
 that-0.550
und-0.523
Token.
Token ID13
Feature activation-1.55e-01

Pos logit contributions
 eyel+0.060
cel+0.056
 moral+0.053
 floor+0.052
 fore+0.051
Neg logit contributions
ised-0.101
eday-0.084
rous-0.084
 that-0.082
und-0.079
Token He
Token ID679
Feature activation+2.87e-01

Pos logit contributions
ised+0.294
rous+0.242
eday+0.241
 that+0.234
und+0.228
Neg logit contributions
 eyel-0.187
cel-0.172
 moral-0.165
 floor-0.164
 fore-0.161
Token thought
Token ID1807
Feature activation-1.21e-01

Pos logit contributions
 eyel+0.367
cel+0.346
 moral+0.328
 floor+0.321
 fore+0.321
Neg logit contributions
ised-0.517
eday-0.423
rous-0.418
 that-0.414
und-0.396
Token n
Token ID299
Feature activation-2.41e+00

Pos logit contributions
ised+0.191
eday+0.151
rous+0.150
 that+0.142
und+0.140
Neg logit contributions
 eyel-0.185
cel-0.174
 moral-0.168
 floor-0.167
 fore-0.166
Tokenaps
Token ID1686
Feature activation-7.14e-01

Pos logit contributions
ised+4.250
eday+3.640
 that+3.552
rous+3.542
Fine+3.483
Neg logit contributions
 eyel-3.029
cel-2.864
nob-2.703
Beck-2.695
quin-2.653
Token were
Token ID547
Feature activation-3.65e-01

Pos logit contributions
ised+1.338
eday+1.115
rous+1.111
 that+1.035
und+1.033
Neg logit contributions
 eyel-0.856
cel-0.779
 fore-0.772
 moral-0.768
 floor-0.737
Token un
Token ID555
Feature activation-9.73e-01

Pos logit contributions
ised+0.647
eday+0.523
rous+0.520
und+0.492
 that+0.488
Neg logit contributions
 eyel-0.491
cel-0.453
 moral-0.442
 floor-0.439
 fore-0.436
Tokenimportant
Token ID18049
Feature activation-7.02e-01

Pos logit contributions
ised+1.617
eday+1.315
rous+1.310
 that+1.303
aw+1.239
Neg logit contributions
 eyel-1.367
cel-1.341
nob-1.249
 floor-1.206
 fore-1.193
Token and
Token ID290
Feature activation-4.57e-01

Pos logit contributions
ised+1.344
rous+1.137
eday+1.121
und+1.042
Deal+1.037
Neg logit contributions
 eyel-0.837
cel-0.771
 floor-0.744
 fore-0.730
 moral-0.724
Token became
Token ID2627
Feature activation+1.58e-01

Pos logit contributions
 eyel+1.320
cel+1.220
 moral+1.156
quin+1.124
 floor+1.115
Neg logit contributions
ised-1.799
 that-1.518
eday-1.479
rous-1.443
und-1.391
Token very
Token ID845
Feature activation-2.80e-01

Pos logit contributions
 eyel+0.193
cel+0.181
 moral+0.171
 floor+0.168
 fore+0.167
Neg logit contributions
ised-0.297
eday-0.245
rous-0.243
 that-0.239
und-0.230
Token famous
Token ID5863
Feature activation-2.82e-01

Pos logit contributions
ised+0.487
eday+0.395
rous+0.390
 that+0.378
und+0.377
Neg logit contributions
 eyel-0.381
 moral-0.350
cel-0.349
 fore-0.338
 floor-0.335
Token.
Token ID13
Feature activation+3.17e-01

Pos logit contributions
ised+0.551
rous+0.458
eday+0.456
 that+0.433
und+0.433
Neg logit contributions
 eyel-0.322
cel-0.299
 moral-0.282
 floor-0.282
 fore-0.278
Token People
Token ID4380
Feature activation-5.62e-01

Pos logit contributions
 eyel+0.432
cel+0.406
 moral+0.386
 floor+0.379
 fore+0.377
Neg logit contributions
ised-0.552
eday-0.448
rous-0.441
 that-0.440
und-0.421
Token from
Token ID422
Feature activation-1.87e+00

Pos logit contributions
ised+1.045
rous+0.859
eday+0.857
 that+0.801
und+0.798
Neg logit contributions
 eyel-0.670
cel-0.640
 fore-0.618
 moral-0.617
 floor-0.599
Token all
Token ID477
Feature activation-1.93e-01

Pos logit contributions
ised+3.531
und+2.892
Deal+2.855
uck+2.831
Fine+2.822
Neg logit contributions
 eyel-2.213
 moral-2.146
 fore-2.111
 floor-2.074
 other-1.996
Token around
Token ID1088
Feature activation+1.23e-01

Pos logit contributions
ised+0.356
eday+0.290
rous+0.288
 that+0.277
und+0.272
Neg logit contributions
 eyel-0.243
cel-0.228
 moral-0.220
 fore-0.217
 floor-0.215
Token the
Token ID262
Feature activation-4.28e-01

Pos logit contributions
 eyel+0.205
cel+0.195
 moral+0.188
 floor+0.186
 fore+0.185
Neg logit contributions
ised-0.176
eday-0.135
rous-0.133
 that-0.130
und-0.124
Token world
Token ID995
Feature activation-7.27e-01

Pos logit contributions
ised+0.759
eday+0.620
rous+0.609
und+0.583
 that+0.582
Neg logit contributions
 eyel-0.566
cel-0.547
 floor-0.531
 moral-0.530
 fore-0.514
Token bought
Token ID5839
Feature activation-4.24e-01

Pos logit contributions
ised+1.266
eday+1.029
rous+1.026
Fine+0.953
und+0.942
Neg logit contributions
 eyel-0.957
cel-0.922
 fore-0.884
 floor-0.873
 moral-0.859
Token his
Token ID465
Feature activation-6.32e-01

Pos logit contributions
ised+1.068
rous+0.885
eday+0.879
aw+0.844
und+0.837
Neg logit contributions
 eyel-0.582
cel-0.544
 fore-0.530
 moral-0.525
 floor-0.499
Token help
Token ID1037
Feature activation+2.23e-01

Pos logit contributions
ised+1.155
eday+0.936
rous+0.931
und+0.916
aw+0.892
Neg logit contributions
 eyel-0.829
 moral-0.770
 fore-0.750
cel-0.750
 floor-0.714
Token.
Token ID13
Feature activation+4.69e-01

Pos logit contributions
 eyel+0.239
cel+0.222
 moral+0.209
 floor+0.203
 fore+0.202
Neg logit contributions
ised-0.452
eday-0.378
rous-0.374
 that-0.372
und-0.358
Token
Token ID198
Feature activation-9.22e-02

Pos logit contributions
 eyel+0.627
cel+0.587
 moral+0.556
 floor+0.553
 fore+0.547
Neg logit contributions
ised-0.829
eday-0.671
 that-0.663
rous-0.660
und-0.634
Token
Token ID198
Feature activation-3.16e-02

Pos logit contributions
ised+0.167
eday+0.137
rous+0.136
 that+0.132
und+0.128
Neg logit contributions
 eyel-0.117
cel-0.111
 moral-0.107
 fore-0.104
 floor-0.103
TokenFrom
Token ID4863
Feature activation-2.49e+00

Pos logit contributions
ised+0.055
eday+0.045
rous+0.044
 that+0.043
und+0.042
Neg logit contributions
 eyel-0.043
cel-0.040
 moral-0.039
 fore-0.038
 floor-0.038
Token that
Token ID326
Feature activation+5.81e-01

Pos logit contributions
ised+2.264
und+1.763
rous+1.737
eday+1.686
Fine+1.681
Neg logit contributions
 moral-5.067
 fore-4.799
 eyel-4.795
cel-4.691
 cor-4.649
Token day
Token ID1110
Feature activation+6.40e-01

Pos logit contributions
 eyel+0.849
cel+0.799
 floor+0.746
 moral+0.741
 fore+0.734
Neg logit contributions
ised-0.964
 that-0.774
eday-0.773
rous-0.767
und-0.721
Token on
Token ID319
Feature activation+4.33e-01

Pos logit contributions
 eyel+0.829
cel+0.766
 floor+0.741
 moral+0.734
 fore+0.711
Neg logit contributions
ised-1.160
 that-0.945
eday-0.945
rous-0.925
und-0.901
Token,
Token ID11
Feature activation+5.93e-01

Pos logit contributions
 eyel+0.638
cel+0.604
 moral+0.577
 floor+0.570
 fore+0.563
Neg logit contributions
ised-0.709
eday-0.562
 that-0.559
rous-0.553
und-0.527
Token King
Token ID2677
Feature activation+4.37e-02

Pos logit contributions
 eyel+0.843
cel+0.786
 floor+0.750
 moral+0.746
 fore+0.738
Neg logit contributions
ised-1.012
eday-0.810
 that-0.808
rous-0.797
und-0.761