TOP ACTIVATIONS
MAX = 15.749

j olly cat named Wh isk ers . Wh isk ers
happier . Mr . Wh isk ers went outside and when
's pet cat , Wh isk ers , jumped up and
a black cat named Wh isk ers . Wh isk ers
nos y cat , Wh isk ers , saw her searching
a lazy cat named Wh isk ers . Wh isk ers
gr umpy cat , Wh isk ers .
cat named Mr . Wh isk ers . Mr . Wh
a long time . Wh isk ers learned that sometimes it
a quiet mouse named Wh isk ers . Wh isk ers
a friendly cat named Wh isk ers . Wh isk ers
the string . But Wh isk ers got all tangled up
Wh isk ers . Wh isk ers loved to eat tuna
clever k itty named Wh isk ers . Wh isk ers
Mr . Wh isk ers dreams were filled with
Wh isk ers . Wh isk ers loved to walk around
an adorable cat named Wh isk ers . Lily and Wh
to the mailbox . Wh isk ers followed her and watched
ers . Mr . Wh isk ers loved to study .
a lazy cat named Wh isk ers . Wh isk ers

MIN ACTIVATIONS
MIN = -14.955

learned that just because someone waves at you , it does
to surf on the big waves at the beach . One
spend the day chasing the waves . Mom my
water . They saw the waves and the shells . They
and the dog made BIG waves in the sea and had
sea . They enjoyed the waves and the feeling of being
face , and watching the waves . One
The wind blew and the waves rocked , but with Jane
. He looked at the waves again and saw that they
that , he watched the waves with a tw inkle in
into the water . The waves took his boat away ,
runs to the fence and waves at the bull . The
some people surfing in the waves . He thought it looked
would splash and make big waves with his hands .
in the water . The waves felt gentle as they rolled
for more fun in the waves . <|endoftext|> Once upon a
them could really ride the waves , but they all had
D ucky . D ucky waves back with its wing .
hand as they watched the waves crashing against the shore .
Lily wanted to feel the waves on her feet , so

INTERVAL 8.073 - 15.749
CONTAINS 1.190%

me a hug ?" Mom my said , " Of course
." Sam my said , " I can
L ila ." L ila says , " Thank
Lisa . It asked her if she wanted to play .
whispered to him about the needle . The park ranger thanked

INTERVAL 0.397 - 8.073
CONTAINS 42.816%

reminder . She walked away . Ben and Lily
them . They helped them make a bigger snow man .
to do . He tried to fly away , but the
something . He remembered a funny looking tree at the top
to jump and sing ! M andy and Martin were so

INTERVAL -7.279 - 0.397
CONTAINS 54.850%

time , there was a shy little girl named Lily .
. They are happy . They are friends . <|endoftext|> Tom
his family enjoy mel ons all the time . They know
between playing and doing your job . Doing something you are
is playing with it . He feels angry . He wants

INTERVAL -14.955 - -7.279
CONTAINS 1.143%

what was inside it . Brad decided to ask his best
time , there were two friends . Their names were Lisa
An ice cream party ! Everyone was so happy and the
The bird was happy and thanked the bunny . The bunny
L ily showed her mum the line and mum said
Token j
Token ID474
Feature activation-3.42e-02

Pos logit contributions
isk+0.094
 whether+0.090
ety+0.088
my+0.087
coat+0.082
Neg logit contributions
 waves-0.059
 victorious-0.057
 shark-0.055
 screams-0.055
 plunge-0.053
Tokenolly
Token ID5098
Feature activation-7.46e+00

Pos logit contributions
 waves+0.001
 victorious+0.001
 screams+0.001
 cheers+0.001
 sinks+0.001
Neg logit contributions
isk-0.002
ety-0.002
my-0.002
coat-0.002
 whether-0.002
Token cat
Token ID3797
Feature activation+5.11e+00

Pos logit contributions
 waves+0.338
 shark+0.328
 victorious+0.322
 screams+0.305
 cliff+0.303
Neg logit contributions
isk-0.509
 whether-0.486
ety-0.469
my-0.463
coat-0.436
Token named
Token ID3706
Feature activation+2.98e+00

Pos logit contributions
isk+0.276
 whether+0.255
ety+0.245
coat+0.231
wn+0.226
Neg logit contributions
 shark-0.301
 waves-0.290
 boat-0.289
 cliff-0.271
 whale-0.270
Token Wh
Token ID854
Feature activation+1.17e+01

Pos logit contributions
isk+0.150
 whether+0.147
epad+0.135
tr+0.133
coat+0.132
Neg logit contributions
 shark-0.159
 waves-0.159
 boat-0.152
 shore-0.150
ls-0.150
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.266
Which+0.196
 blonde+0.195
 Which+0.182
 hers+0.173
Neg logit contributions
ls-0.879
 victorious-0.871
 screams-0.870
ars-0.858
 waves-0.857
Tokeners
Token ID364
Feature activation+3.45e-01

Pos logit contributions
 whether+1.144
isk+1.091
ety+1.079
coat+1.035
my+1.026
Neg logit contributions
 victorious-0.601
 waves-0.590
ls-0.545
 shark-0.541
 cliff-0.508
Token.
Token ID13
Feature activation+5.01e-01

Pos logit contributions
isk+0.024
wn+0.024
ety+0.024
 whether+0.024
coat+0.024
Neg logit contributions
 waves-0.011
 boat-0.011
 shark-0.011
 victorious-0.010
 shore-0.010
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
 whether+0.023
epad+0.022
ellery+0.021
ption+0.021
 drew+0.020
Neg logit contributions
 waves-0.029
 winds-0.027
 current-0.027
 boat-0.027
 victorious-0.026
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.289
 blonde+0.210
epad+0.204
 blond+0.196
rings+0.191
Neg logit contributions
ls-0.916
 victorious-0.875
 waves-0.858
 screams-0.856
ars-0.853
Tokeners
Token ID364
Feature activation+2.71e-01

Pos logit contributions
 whether+1.137
isk+1.091
ety+1.079
coat+1.034
my+1.022
Neg logit contributions
 victorious-0.589
 waves-0.580
ls-0.539
 shark-0.535
 losers-0.499
Token happier
Token ID23030
Feature activation-5.23e+00

Pos logit contributions
my+0.127
 whether+0.127
isk+0.116
 makeup+0.112
ety+0.111
Neg logit contributions
aro-0.129
 cliff-0.125
 shark-0.125
 lighthouse-0.120
 plunge-0.119
Token.
Token ID13
Feature activation+4.62e-01

Pos logit contributions
 victorious+0.353
 waves+0.352
 shark+0.346
 plunge+0.343
 screams+0.341
Neg logit contributions
 whether-0.219
isk-0.219
my-0.201
ety-0.199
ine-0.174
Token Mr
Token ID1770
Feature activation+1.15e+00

Pos logit contributions
isk+0.030
 whether+0.029
ety+0.028
my+0.027
coat+0.027
Neg logit contributions
 waves-0.021
 victorious-0.021
 shark-0.021
 screams-0.020
 cheers-0.019
Token.
Token ID13
Feature activation+4.60e-01

Pos logit contributions
 whether+0.066
isk+0.063
my+0.058
ety+0.058
coat+0.057
Neg logit contributions
 waves-0.062
 shark-0.061
 victorious-0.061
ls-0.059
 screams-0.058
Token Wh
Token ID854
Feature activation+1.13e+01

Pos logit contributions
 whether+0.018
ety+0.016
my+0.015
isk+0.015
 drew+0.015
Neg logit contributions
 victorious-0.032
 shark-0.032
 waves-0.031
 current-0.031
 shore-0.030
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.113
 drew+0.015
isk+0.003
ety+0.002
tr+0.001
Neg logit contributions
 victorious-1.131
 waves-1.111
 shark-1.079
ls-1.078
 screams-1.074
Tokeners
Token ID364
Feature activation+2.45e-01

Pos logit contributions
isk+0.961
 whether+0.894
ety+0.863
my+0.858
coat+0.793
Neg logit contributions
 waves-0.838
 shark-0.817
 victorious-0.792
 screams-0.771
 cliff-0.766
Token went
Token ID1816
Feature activation-1.68e+00

Pos logit contributions
isk+0.013
my+0.013
 whether+0.012
ety+0.011
 drew+0.011
Neg logit contributions
 shark-0.014
 waves-0.014
rier-0.014
ls-0.014
 victorious-0.014
Token outside
Token ID2354
Feature activation+1.12e+00

Pos logit contributions
 waves+0.073
 shark+0.071
 victorious+0.070
 screams+0.065
 cliff+0.065
Neg logit contributions
isk-0.118
 whether-0.113
ety-0.110
my-0.107
coat-0.101
Token and
Token ID290
Feature activation+6.29e-02

Pos logit contributions
isk+0.055
 whether+0.051
ety+0.050
my+0.047
coat+0.044
Neg logit contributions
 waves-0.072
 shark-0.071
 victorious-0.070
 screams-0.067
 cliff-0.067
Token when
Token ID618
Feature activation+5.67e+00

Pos logit contributions
isk+0.003
 whether+0.003
my+0.003
ety+0.003
 drew+0.003
Neg logit contributions
 waves-0.004
 victorious-0.004
 shark-0.004
rier-0.004
 screams-0.004
Token's
Token ID338
Feature activation+8.04e-01

Pos logit contributions
 waves+0.144
 shark+0.141
 victorious+0.140
 screams+0.134
 cliff+0.133
Neg logit contributions
isk-0.118
 whether-0.111
ety-0.106
my-0.105
coat-0.096
Token pet
Token ID4273
Feature activation+2.88e-01

Pos logit contributions
isk+0.049
my+0.048
 whether+0.047
ety+0.046
coat+0.045
Neg logit contributions
 victorious-0.040
 waves-0.038
 shark-0.037
 cliff-0.037
ls-0.037
Token cat
Token ID3797
Feature activation+5.04e+00

Pos logit contributions
isk+0.018
 whether+0.017
ety+0.016
my+0.015
coat+0.014
Neg logit contributions
 waves-0.016
 shark-0.015
 victorious-0.014
 boat-0.014
 screams-0.014
Token,
Token ID11
Feature activation+8.91e-01

Pos logit contributions
isk+0.301
 whether+0.296
ety+0.277
my+0.255
coat+0.250
Neg logit contributions
 waves-0.275
 shark-0.264
 screams-0.258
 victorious-0.248
ls-0.246
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.047
 whether+0.043
ety+0.042
my+0.039
coat+0.038
Neg logit contributions
 waves-0.055
 shark-0.053
 victorious-0.052
 screams-0.051
 cheers-0.050
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.304
Which+0.215
Do+0.172
 Which+0.170
 drew+0.162
Neg logit contributions
 screams-0.932
 waves-0.911
 victorious-0.891
ls-0.863
 losers-0.859
Tokeners
Token ID364
Feature activation+2.01e-01

Pos logit contributions
 whether+1.124
isk+1.089
ety+1.075
my+1.006
coat+1.006
Neg logit contributions
 victorious-0.630
 waves-0.623
 shark-0.581
ls-0.562
 screams-0.547
Token,
Token ID11
Feature activation+8.98e-01

Pos logit contributions
isk+0.011
 whether+0.011
ety+0.010
my+0.010
coat+0.009
Neg logit contributions
 waves-0.012
 shark-0.011
 victorious-0.011
 screams-0.011
 cliff-0.011
Token jumped
Token ID11687
Feature activation-2.32e+00

Pos logit contributions
isk+0.050
 whether+0.046
ety+0.045
my+0.042
coat+0.041
Neg logit contributions
 waves-0.054
 shark-0.052
 victorious-0.050
 screams-0.050
 boat-0.048
Token up
Token ID510
Feature activation-4.16e+00

Pos logit contributions
 waves+0.113
 shark+0.111
 victorious+0.110
 cliff+0.105
 lighthouse+0.104
Neg logit contributions
isk-0.147
 whether-0.139
my-0.136
ety-0.134
coat-0.127
Token and
Token ID290
Feature activation+7.20e-03

Pos logit contributions
 waves+0.259
 shark+0.252
 victorious+0.250
 screams+0.240
 cliff+0.239
Neg logit contributions
isk-0.216
 whether-0.203
ety-0.194
my-0.189
coat-0.174
Token a
Token ID257
Feature activation+1.37e+00

Pos logit contributions
isk+0.008
 whether+0.007
ety+0.007
coat+0.006
my+0.006
Neg logit contributions
 waves-0.006
 shark-0.006
 victorious-0.006
 boat-0.006
 shore-0.006
Token black
Token ID2042
Feature activation+5.28e+00

Pos logit contributions
isk+0.095
 whether+0.091
ety+0.089
my+0.088
coat+0.082
Neg logit contributions
 waves-0.058
 victorious-0.057
 screams-0.054
 shark-0.054
 plunge-0.053
Token cat
Token ID3797
Feature activation+4.97e+00

Pos logit contributions
isk+0.361
 whether+0.351
ety+0.326
my+0.316
coat+0.302
Neg logit contributions
 shark-0.258
 waves-0.253
 boat-0.236
 cliff-0.232
 current-0.221
Token named
Token ID3706
Feature activation+3.03e+00

Pos logit contributions
isk+0.271
 whether+0.248
ety+0.242
coat+0.226
my+0.216
Neg logit contributions
 shark-0.304
 waves-0.289
 boat-0.287
 cliff-0.273
 whale-0.269
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.174
 whether+0.171
my+0.162
ety+0.156
epad+0.152
Neg logit contributions
 shark-0.147
 screams-0.145
 waves-0.144
ls-0.136
 boat-0.136
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.279
Which+0.229
 Which+0.194
 blonde+0.191
 gloss+0.188
Neg logit contributions
 screams-0.871
ars-0.868
ls-0.860
 victorious-0.849
 waves-0.836
Tokeners
Token ID364
Feature activation+1.66e-01

Pos logit contributions
 whether+1.142
isk+1.096
ety+1.081
coat+1.036
my+1.024
Neg logit contributions
 victorious-0.608
 waves-0.605
ls-0.564
 shark-0.563
 screams-0.525
Token.
Token ID13
Feature activation+2.97e-01

Pos logit contributions
isk+0.012
ety+0.012
coat+0.012
wn+0.012
 whether+0.012
Neg logit contributions
 shark-0.005
 waves-0.005
 boat-0.005
 victorious-0.005
 wave-0.005
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
 whether+0.013
isk+0.012
ety+0.011
epad+0.011
ellery+0.011
Neg logit contributions
 waves-0.019
 winds-0.018
 victorious-0.018
 current-0.017
 losers-0.017
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.254
epad+0.184
rings+0.156
 blonde+0.155
 drew+0.149
Neg logit contributions
ls-0.973
 waves-0.926
 victorious-0.917
 screams-0.912
ars-0.903
Tokeners
Token ID364
Feature activation+1.44e-01

Pos logit contributions
 whether+1.116
isk+1.084
ety+1.058
coat+1.015
my+1.004
Neg logit contributions
 waves-0.620
 victorious-0.613
 shark-0.581
ls-0.573
 screams-0.534
Token nos
Token ID43630
Feature activation+9.14e-01

Pos logit contributions
isk+0.056
 whether+0.054
ety+0.052
my+0.051
coat+0.048
Neg logit contributions
 waves-0.028
 shark-0.026
 victorious-0.025
 screams-0.024
 boat-0.024
Tokeny
Token ID88
Feature activation+7.23e+00

Pos logit contributions
isk+0.043
 whether+0.041
 knit+0.035
tr+0.033
coat+0.032
Neg logit contributions
 waves-0.062
 screams-0.059
 shark-0.058
ls-0.057
 winds-0.055
Token cat
Token ID3797
Feature activation+5.00e+00

Pos logit contributions
isk+0.456
 whether+0.432
ety+0.405
my+0.392
coat+0.374
Neg logit contributions
 waves-0.383
 shark-0.369
 screams-0.355
 victorious-0.347
 boat-0.345
Token,
Token ID11
Feature activation+1.10e+00

Pos logit contributions
 whether+0.351
isk+0.349
ety+0.339
coat+0.320
epad+0.317
Neg logit contributions
 waves-0.194
 screams-0.192
 shark-0.192
ls-0.184
 boat-0.178
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.065
ety+0.061
coat+0.058
tr+0.057
epad+0.057
Neg logit contributions
 waves-0.056
 screams-0.056
 shark-0.053
 terrified-0.050
 cheers-0.049
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.304
Which+0.280
shirt+0.208
 Which+0.207
 which+0.205
Neg logit contributions
 screams-0.862
ls-0.827
ars-0.814
 victorious-0.805
 waves-0.800
Tokeners
Token ID364
Feature activation+3.02e-01

Pos logit contributions
 whether+1.161
ety+1.123
isk+1.117
coat+1.063
ine+1.054
Neg logit contributions
 victorious-0.585
 waves-0.560
ls-0.535
 shark-0.515
 screams-0.498
Token,
Token ID11
Feature activation+1.09e+00

Pos logit contributions
isk+0.020
ety+0.020
 whether+0.019
tr+0.018
my+0.018
Neg logit contributions
 waves-0.013
 victorious-0.013
 screams-0.013
 shark-0.013
 boat-0.012
Token saw
Token ID2497
Feature activation+1.76e+00

Pos logit contributions
isk+0.066
ety+0.061
tr+0.060
ine+0.057
coat+0.057
Neg logit contributions
 waves-0.058
 shark-0.053
 screams-0.052
 wave-0.050
 boat-0.048
Token her
Token ID607
Feature activation+5.18e+00

Pos logit contributions
coat+0.109
tr+0.103
isk+0.102
ety+0.102
epad+0.100
Neg logit contributions
 waves-0.095
ls-0.090
 screams-0.089
 shark-0.087
 cheers-0.080
Token searching
Token ID10342
Feature activation+2.70e+00

Pos logit contributions
isk+0.272
 whether+0.253
ety+0.236
my+0.224
tr+0.212
Neg logit contributions
 waves-0.326
 shark-0.315
 screams-0.306
ls-0.302
 victorious-0.302
Token a
Token ID257
Feature activation+1.41e+00

Pos logit contributions
isk+0.006
 whether+0.006
ety+0.006
coat+0.005
my+0.005
Neg logit contributions
 waves-0.006
 shark-0.006
 victorious-0.005
 refreshed-0.005
 boat-0.005
Token lazy
Token ID16931
Feature activation-1.23e+00

Pos logit contributions
isk+0.097
 whether+0.093
ety+0.091
my+0.090
coat+0.084
Neg logit contributions
 waves-0.061
 victorious-0.059
 screams-0.056
 shark-0.056
 plunge-0.055
Token cat
Token ID3797
Feature activation+5.12e+00

Pos logit contributions
 waves+0.065
 victorious+0.065
 shark+0.062
 screams+0.061
 plunge+0.059
Neg logit contributions
isk-0.073
 whether-0.068
ety-0.066
my-0.066
coat-0.062
Token named
Token ID3706
Feature activation+3.08e+00

Pos logit contributions
isk+0.277
 whether+0.258
ety+0.251
coat+0.234
ptions+0.229
Neg logit contributions
 shark-0.300
 waves-0.286
 boat-0.285
 cliff-0.270
 wave-0.265
Token Wh
Token ID854
Feature activation+1.16e+01

Pos logit contributions
isk+0.158
 whether+0.152
 drew+0.141
epad+0.140
ety+0.139
Neg logit contributions
 shark-0.162
 waves-0.157
 screams-0.157
 losers-0.151
 plunge-0.150
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.254
Which+0.205
 blonde+0.183
 Which+0.177
 hers+0.173
Neg logit contributions
 screams-0.893
 victorious-0.886
ars-0.874
ls-0.871
 waves-0.868
Tokeners
Token ID364
Feature activation+2.86e-01

Pos logit contributions
 whether+1.099
isk+1.069
ety+1.048
my+0.994
coat+0.993
Neg logit contributions
 victorious-0.644
 waves-0.640
 shark-0.597
ls-0.579
 screams-0.560
Token.
Token ID13
Feature activation+5.19e-01

Pos logit contributions
wn+0.021
isk+0.021
coat+0.021
ety+0.020
 whether+0.020
Neg logit contributions
 boat-0.008
 shark-0.008
 waves-0.008
 victorious-0.007
 wave-0.007
Token Wh
Token ID854
Feature activation+1.16e+01

Pos logit contributions
 whether+0.023
epad+0.022
ellery+0.021
ety+0.021
isk+0.021
Neg logit contributions
 waves-0.031
 winds-0.030
 current-0.029
 losers-0.029
 screams-0.029
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.247
 blonde+0.159
epad+0.155
rings+0.153
 drew+0.149
Neg logit contributions
ls-0.963
 victorious-0.936
 waves-0.929
 screams-0.924
ars-0.902
Tokeners
Token ID364
Feature activation+2.34e-01

Pos logit contributions
 whether+1.080
isk+1.061
ety+1.032
my+0.980
coat+0.980
Neg logit contributions
 waves-0.646
 victorious-0.642
 shark-0.610
ls-0.585
 screams-0.563
Token gr
Token ID1036
Feature activation+1.53e+00

Pos logit contributions
isk+0.365
 whether+0.350
ety+0.339
my+0.336
coat+0.313
Neg logit contributions
 waves-0.217
 shark-0.208
 victorious-0.207
 screams-0.194
 cliff-0.194
Tokenumpy
Token ID32152
Feature activation-2.31e+00

Pos logit contributions
isk+0.096
 whether+0.092
ety+0.088
my+0.087
coat+0.082
Neg logit contributions
 waves-0.078
 shark-0.076
 victorious-0.075
 screams-0.071
 cliff-0.070
Token cat
Token ID3797
Feature activation+5.04e+00

Pos logit contributions
 waves+0.114
 shark+0.110
 victorious+0.107
 screams+0.103
 cliff+0.101
Neg logit contributions
isk-0.151
 whether-0.144
ety-0.138
my-0.136
coat-0.128
Token,
Token ID11
Feature activation+1.03e+00

Pos logit contributions
isk+0.310
 whether+0.306
coat+0.295
ety+0.293
organ+0.293
Neg logit contributions
 boat-0.233
 waves-0.232
 shark-0.228
 victorious-0.213
ls-0.211
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.051
ety+0.046
 whether+0.045
my+0.043
coat+0.042
Neg logit contributions
 waves-0.064
 shark-0.064
 victorious-0.063
 screams-0.062
 shore-0.060
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
Which+0.293
 whether+0.287
shirt+0.239
 Which+0.238
 blonde+0.220
Neg logit contributions
 screams-0.844
an-0.810
ls-0.801
ble-0.794
 waves-0.793
Tokeners
Token ID364
Feature activation+3.00e-01

Pos logit contributions
 whether+1.163
ety+1.121
isk+1.116
coat+1.083
ine+1.051
Neg logit contributions
 victorious-0.577
 waves-0.552
ls-0.526
 shark-0.504
 screams-0.476
Token.
Token ID13
Feature activation+4.72e-01

Pos logit contributions
isk+0.019
 whether+0.018
ety+0.018
my+0.017
tr+0.017
Neg logit contributions
 waves-0.014
 victorious-0.014
 shark-0.014
 screams-0.014
 boat-0.013
Token
Token ID220
Feature activation+4.36e-01

Pos logit contributions
 whether+0.034
isk+0.033
epad+0.032
ellery+0.032
ety+0.032
Neg logit contributions
 waves-0.016
 victorious-0.015
 boat-0.014
ls-0.014
 refreshed-0.014
Token
Token ID198
Feature activation-6.02e-01

Pos logit contributions
isk+0.028
 whether+0.027
ety+0.026
coat+0.025
my+0.024
Neg logit contributions
 victorious-0.019
 waves-0.019
 shark-0.018
ls-0.018
 screams-0.018
Token
Token ID198
Feature activation-5.95e-01

Pos logit contributions
ls+0.022
 victorious+0.022
 waves+0.022
 shark+0.021
 screams+0.021
Neg logit contributions
isk-0.044
 whether-0.041
ety-0.040
coat-0.038
tr-0.038
Token cat
Token ID3797
Feature activation+4.94e+00

Pos logit contributions
isk+0.241
 whether+0.230
ety+0.218
my+0.211
coat+0.200
Neg logit contributions
 shark-0.179
 waves-0.179
 boat-0.164
 cliff-0.164
 victorious-0.160
Token named
Token ID3706
Feature activation+2.92e+00

Pos logit contributions
isk+0.274
 whether+0.248
ety+0.242
coat+0.225
ptions+0.217
Neg logit contributions
 shark-0.294
 waves-0.281
 boat-0.281
 cliff-0.266
 whale-0.261
Token Mr
Token ID1770
Feature activation+1.11e+00

Pos logit contributions
isk+0.163
 whether+0.158
my+0.147
ety+0.145
coat+0.139
Neg logit contributions
 shark-0.149
 waves-0.142
 screams-0.142
 cliff-0.134
 plunge-0.134
Token.
Token ID13
Feature activation+1.23e-01

Pos logit contributions
 whether+0.076
isk+0.073
coat+0.069
shirt+0.065
 drew+0.064
Neg logit contributions
 shark-0.046
 waves-0.043
ls-0.042
 screams-0.041
 victorious-0.041
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
 whether+0.007
isk+0.007
ety+0.006
my+0.006
tr+0.006
Neg logit contributions
 shark-0.007
 victorious-0.006
 waves-0.006
 current-0.006
 shore-0.006
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.263
Which+0.199
rings+0.189
 Which+0.163
 drew+0.162
Neg logit contributions
 screams-0.895
ls-0.892
 shark-0.874
 victorious-0.874
 waves-0.872
Tokeners
Token ID364
Feature activation+1.15e-01

Pos logit contributions
 whether+1.116
isk+1.081
ety+1.054
coat+1.007
my+1.000
Neg logit contributions
 waves-0.629
 victorious-0.628
 shark-0.600
ls-0.583
 screams-0.552
Token.
Token ID13
Feature activation+1.26e-01

Pos logit contributions
isk+0.008
 whether+0.008
ety+0.007
coat+0.007
wn+0.007
Neg logit contributions
 waves-0.005
 shark-0.005
 boat-0.005
 wave-0.004
 screams-0.004
Token Mr
Token ID1770
Feature activation+1.11e+00

Pos logit contributions
 whether+0.008
isk+0.008
ety+0.007
ine+0.007
ellery+0.007
Neg logit contributions
 waves-0.006
 shark-0.005
 current-0.005
 cheers-0.005
 victorious-0.005
Token.
Token ID13
Feature activation+1.18e-01

Pos logit contributions
 whether+0.070
isk+0.067
coat+0.061
ety+0.059
my+0.059
Neg logit contributions
 waves-0.054
 shark-0.054
ls-0.052
 victorious-0.051
 screams-0.051
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
 whether+0.005
shirt+0.005
 drew+0.005
isk+0.004
ety+0.004
Neg logit contributions
 shark-0.007
 current-0.007
 victorious-0.007
 waves-0.007
 shore-0.007
Token a
Token ID257
Feature activation+1.46e+00

Pos logit contributions
isk+0.140
 whether+0.134
ety+0.131
my+0.127
coat+0.120
Neg logit contributions
 waves-0.131
 shark-0.130
 victorious-0.128
 screams-0.121
 cliff-0.120
Token long
Token ID890
Feature activation-4.25e-01

Pos logit contributions
isk+0.094
 whether+0.089
ety+0.086
my+0.085
coat+0.079
Neg logit contributions
 waves-0.072
 shark-0.071
 victorious-0.070
 screams-0.066
 cliff-0.066
Token time
Token ID640
Feature activation-1.71e+00

Pos logit contributions
 waves+0.018
 shark+0.018
 current+0.017
 boat+0.017
 cliff+0.017
Neg logit contributions
isk-0.029
 whether-0.029
ety-0.028
my-0.026
tr-0.025
Token.
Token ID13
Feature activation+4.17e-01

Pos logit contributions
 waves+0.109
 shark+0.106
 victorious+0.105
 screams+0.101
 cliff+0.101
Neg logit contributions
isk-0.085
 whether-0.081
ety-0.076
my-0.076
coat-0.069
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.023
 whether+0.023
ety+0.021
my+0.021
coat+0.020
Neg logit contributions
 victorious-0.023
 waves-0.022
 shark-0.021
 screams-0.021
 cheers-0.021
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.172
coat+0.067
 drew+0.061
 makeup+0.055
 blonde+0.055
Neg logit contributions
 victorious-1.092
ls-1.065
 waves-1.046
 screams-1.031
 losers-1.017
Tokeners
Token ID364
Feature activation+3.00e-01

Pos logit contributions
isk+0.993
 whether+0.959
ety+0.922
my+0.903
coat+0.855
Neg logit contributions
 waves-0.775
 victorious-0.753
 shark-0.749
 screams-0.704
 cliff-0.697
Token learned
Token ID4499
Feature activation-9.07e-01

Pos logit contributions
isk+0.015
my+0.013
 whether+0.012
ety+0.012
 drew+0.011
Neg logit contributions
 waves-0.020
 shark-0.020
 victorious-0.019
 cliff-0.019
 plunge-0.019
Token that
Token ID326
Feature activation+2.09e-01

Pos logit contributions
 waves+0.041
 shark+0.040
 victorious+0.040
 screams+0.037
 boat+0.036
Neg logit contributions
isk-0.062
ety-0.058
 whether-0.057
my-0.056
coat-0.055
Token sometimes
Token ID3360
Feature activation+3.83e+00

Pos logit contributions
isk+0.013
ety+0.012
 whether+0.012
coat+0.011
my+0.011
Neg logit contributions
 shark-0.011
 waves-0.011
 victorious-0.010
 cliff-0.010
 screams-0.010
Token it
Token ID340
Feature activation+2.66e+00

Pos logit contributions
 whether+0.222
my+0.209
isk+0.208
ety+0.197
ine+0.178
Neg logit contributions
 victorious-0.211
 waves-0.201
 shark-0.191
rier-0.190
 lighthouse-0.186
Token a
Token ID257
Feature activation+1.20e+00

Pos logit contributions
isk+0.011
coat+0.010
ety+0.010
ine+0.010
 whether+0.010
Neg logit contributions
 victorious-0.008
 waves-0.008
 shark-0.007
 boat-0.007
 screams-0.007
Token quiet
Token ID5897
Feature activation-6.59e-01

Pos logit contributions
isk+0.091
 whether+0.087
ety+0.084
my+0.083
coat+0.079
Neg logit contributions
 waves-0.046
 shark-0.045
 victorious-0.043
 cliff-0.041
 boat-0.040
Token mouse
Token ID10211
Feature activation+3.29e+00

Pos logit contributions
 waves+0.031
 shark+0.031
 cliff+0.029
 boat+0.029
 victorious+0.028
Neg logit contributions
isk-0.045
 whether-0.043
ety-0.041
my-0.039
coat-0.037
Token named
Token ID3706
Feature activation+3.00e+00

Pos logit contributions
isk+0.168
 whether+0.147
ety+0.146
coat+0.136
my+0.134
Neg logit contributions
 shark-0.213
 waves-0.210
 boat-0.207
 wave-0.198
 cliff-0.198
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
isk+0.159
 whether+0.151
ety+0.146
my+0.141
coat+0.138
Neg logit contributions
 waves-0.177
 shark-0.176
 victorious-0.168
 screams-0.167
 cliff-0.163
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.302
shirt+0.246
 hers+0.229
Which+0.226
 blonde+0.221
Neg logit contributions
 waves-0.851
ls-0.840
 screams-0.833
 victorious-0.831
 shark-0.812
Tokeners
Token ID364
Feature activation+2.58e-01

Pos logit contributions
 whether+1.082
isk+1.071
ety+1.036
my+0.989
coat+0.981
Neg logit contributions
 waves-0.658
 victorious-0.652
 shark-0.614
ls-0.608
 cliff-0.577
Token.
Token ID13
Feature activation+4.15e-01

Pos logit contributions
isk+0.018
coat+0.017
ety+0.017
 whether+0.016
shirt+0.016
Neg logit contributions
 waves-0.010
 boat-0.010
 shark-0.010
 victorious-0.010
 shore-0.009
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
isk+0.024
 whether+0.024
ety+0.023
my+0.021
ellery+0.021
Neg logit contributions
 waves-0.021
 screams-0.020
 shark-0.020
 victorious-0.020
 cheers-0.019
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.233
epad+0.168
 hers+0.160
shirt+0.154
 blonde+0.149
Neg logit contributions
ls-0.952
 waves-0.914
 victorious-0.910
 screams-0.902
 shark-0.879
Tokeners
Token ID364
Feature activation+2.40e-01

Pos logit contributions
 whether+1.078
isk+1.067
ety+1.037
my+0.985
coat+0.978
Neg logit contributions
 waves-0.648
 victorious-0.642
 shark-0.610
ls-0.593
 cliff-0.564
Token a
Token ID257
Feature activation+1.24e+00

Pos logit contributions
 waves+0.004
 shark+0.004
 victorious+0.004
 refreshed+0.003
 boat+0.003
Neg logit contributions
isk-0.005
 whether-0.004
ety-0.004
coat-0.004
my-0.004
Token friendly
Token ID8030
Feature activation-1.07e+00

Pos logit contributions
isk+0.087
 whether+0.083
ety+0.081
my+0.080
coat+0.075
Neg logit contributions
 waves-0.053
 victorious-0.051
 shark-0.049
 screams-0.049
 cheers-0.047
Token cat
Token ID3797
Feature activation+5.02e+00

Pos logit contributions
 waves+0.052
 shark+0.051
 victorious+0.047
 boat+0.047
 cliff+0.046
Neg logit contributions
isk-0.071
 whether-0.069
ety-0.065
my-0.064
coat-0.060
Token named
Token ID3706
Feature activation+2.94e+00

Pos logit contributions
isk+0.284
 whether+0.262
ety+0.256
coat+0.246
wn+0.244
Neg logit contributions
 shark-0.282
 boat-0.268
 waves-0.265
 whale-0.253
 wave-0.249
Token Wh
Token ID854
Feature activation+1.16e+01

Pos logit contributions
isk+0.163
 whether+0.160
epad+0.142
my+0.142
ety+0.141
Neg logit contributions
 shark-0.140
 waves-0.136
 screams-0.135
 boat-0.130
 we-0.129
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.267
Which+0.219
 blonde+0.199
 Which+0.190
rings+0.178
Neg logit contributions
ars-0.857
 victorious-0.854
 screams-0.853
ls-0.850
 waves-0.827
Tokeners
Token ID364
Feature activation+2.92e-01

Pos logit contributions
 whether+1.146
isk+1.090
ety+1.077
coat+1.032
my+1.019
Neg logit contributions
 victorious-0.601
 waves-0.587
ls-0.546
 shark-0.537
 screams-0.507
Token.
Token ID13
Feature activation+2.88e-01

Pos logit contributions
isk+0.021
wn+0.021
coat+0.021
ety+0.021
 whether+0.020
Neg logit contributions
 waves-0.009
 boat-0.009
 shark-0.009
 victorious-0.008
 wave-0.008
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
 whether+0.013
ellery+0.012
epad+0.012
isk+0.012
ety+0.012
Neg logit contributions
 waves-0.017
 winds-0.016
 current-0.016
 victorious-0.016
 losers-0.016
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.261
epad+0.190
 blonde+0.177
 blond+0.166
rings+0.165
Neg logit contributions
ls-0.953
 victorious-0.908
 waves-0.898
ars-0.879
 screams-0.878
Tokeners
Token ID364
Feature activation+2.42e-01

Pos logit contributions
 whether+1.125
isk+1.083
ety+1.063
coat+1.018
my+1.003
Neg logit contributions
 victorious-0.599
 waves-0.598
 shark-0.553
ls-0.551
 screams-0.509
Token the
Token ID262
Feature activation-1.30e+00

Pos logit contributions
 waves+0.096
 screams+0.089
 victorious+0.088
 shark+0.088
ls+0.087
Neg logit contributions
isk-0.117
 whether-0.114
ety-0.110
coat-0.108
tr-0.104
Token string
Token ID4731
Feature activation+1.47e+00

Pos logit contributions
 victorious+0.078
 waves+0.075
 shark+0.075
 current+0.072
ls+0.071
Neg logit contributions
isk-0.065
my-0.063
 whether-0.063
ety-0.061
tr-0.059
Token.
Token ID13
Feature activation+3.13e-01

Pos logit contributions
isk+0.081
 whether+0.076
ety+0.073
my+0.071
coat+0.066
Neg logit contributions
 waves-0.086
 shark-0.084
 victorious-0.083
 screams-0.080
 cliff-0.079
Token But
Token ID887
Feature activation+9.43e-01

Pos logit contributions
isk+0.015
 whether+0.014
ety+0.014
my+0.014
ine+0.012
Neg logit contributions
 waves-0.020
 victorious-0.019
 shark-0.019
 screams-0.019
 current-0.018
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.061
ety+0.057
 whether+0.057
my+0.054
coat+0.053
Neg logit contributions
 victorious-0.042
 shark-0.041
 boat-0.041
 waves-0.041
 current-0.040
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.270
Which+0.192
Do+0.157
 hers+0.135
 drew+0.132
Neg logit contributions
 screams-0.928
 victorious-0.905
ls-0.900
 losers-0.893
 waves-0.881
Tokeners
Token ID364
Feature activation+1.49e-01

Pos logit contributions
 whether+1.091
isk+1.070
ety+1.057
my+0.995
coat+0.992
Neg logit contributions
 victorious-0.640
 waves-0.632
 shark-0.592
ls-0.573
 screams-0.560
Token got
Token ID1392
Feature activation+2.61e+00

Pos logit contributions
isk+0.008
 whether+0.007
my+0.007
ety+0.007
coat+0.007
Neg logit contributions
 waves-0.009
 shark-0.009
 victorious-0.009
 cliff-0.008
 lighthouse-0.008
Token all
Token ID477
Feature activation-4.10e+00

Pos logit contributions
isk+0.191
 whether+0.181
ety+0.175
my+0.172
coat+0.164
Neg logit contributions
 waves-0.108
 shark-0.104
 victorious-0.103
 screams-0.097
 current-0.095
Token tangled
Token ID41431
Feature activation+4.17e+00

Pos logit contributions
 shark+0.219
 waves+0.214
 victorious+0.211
 lighthouse+0.207
 cliff+0.203
Neg logit contributions
isk-0.237
my-0.233
 whether-0.228
ety-0.227
coat-0.212
Token up
Token ID510
Feature activation-4.22e+00

Pos logit contributions
isk+0.299
 whether+0.280
ety+0.268
my+0.259
tr+0.252
Neg logit contributions
 waves-0.180
 shark-0.170
 victorious-0.169
ls-0.168
rier-0.163
Token Wh
Token ID854
Feature activation+1.17e+01

Pos logit contributions
isk+0.150
 whether+0.147
epad+0.135
tr+0.133
coat+0.132
Neg logit contributions
 shark-0.159
 waves-0.159
 boat-0.152
 shore-0.150
ls-0.150
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.266
Which+0.196
 blonde+0.195
 Which+0.182
 hers+0.173
Neg logit contributions
ls-0.879
 victorious-0.871
 screams-0.870
ars-0.858
 waves-0.857
Tokeners
Token ID364
Feature activation+3.45e-01

Pos logit contributions
 whether+1.144
isk+1.091
ety+1.079
coat+1.035
my+1.026
Neg logit contributions
 victorious-0.601
 waves-0.590
ls-0.545
 shark-0.541
 cliff-0.508
Token.
Token ID13
Feature activation+5.01e-01

Pos logit contributions
isk+0.024
wn+0.024
ety+0.024
 whether+0.024
coat+0.024
Neg logit contributions
 waves-0.011
 boat-0.011
 shark-0.011
 victorious-0.010
 shore-0.010
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
 whether+0.023
epad+0.022
ellery+0.021
ption+0.021
 drew+0.020
Neg logit contributions
 waves-0.029
 winds-0.027
 current-0.027
 boat-0.027
 victorious-0.026
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.289
 blonde+0.210
epad+0.204
 blond+0.196
rings+0.191
Neg logit contributions
ls-0.916
 victorious-0.875
 waves-0.858
 screams-0.856
ars-0.853
Tokeners
Token ID364
Feature activation+2.71e-01

Pos logit contributions
 whether+1.137
isk+1.091
ety+1.079
coat+1.034
my+1.022
Neg logit contributions
 victorious-0.589
 waves-0.580
ls-0.539
 shark-0.535
 losers-0.499
Token loved
Token ID6151
Feature activation+1.37e+00

Pos logit contributions
isk+0.015
 whether+0.014
ety+0.013
coat+0.012
my+0.012
Neg logit contributions
 waves-0.016
 shark-0.015
 boat-0.015
 screams-0.014
 victorious-0.014
Token to
Token ID284
Feature activation+1.01e+00

Pos logit contributions
isk+0.080
 whether+0.069
ety+0.068
ine+0.067
my+0.066
Neg logit contributions
 waves-0.084
 shark-0.080
 shore-0.074
 boat-0.073
 cliff-0.073
Token eat
Token ID4483
Feature activation+3.02e+00

Pos logit contributions
isk+0.062
 whether+0.059
ety+0.058
my+0.055
tr+0.051
Neg logit contributions
 waves-0.057
 shark-0.052
 wave-0.052
 boat-0.051
 cliff-0.050
Token tuna
Token ID38883
Feature activation+1.59e+00

Pos logit contributions
isk+0.206
ety+0.191
 whether+0.187
my+0.183
coat+0.177
Neg logit contributions
 waves-0.154
 shark-0.147
ls-0.136
 shore-0.133
 boat-0.133
Token clever
Token ID14169
Feature activation-1.11e+00

Pos logit contributions
isk+0.073
 whether+0.070
ety+0.068
my+0.067
coat+0.063
Neg logit contributions
 waves-0.045
 victorious-0.043
 shark-0.042
 screams-0.041
 plunge-0.040
Token k
Token ID479
Feature activation-3.63e+00

Pos logit contributions
 waves+0.050
 shark+0.049
 victorious+0.047
 cliff+0.045
 screams+0.045
Neg logit contributions
isk-0.076
 whether-0.072
ety-0.070
my-0.069
coat-0.065
Tokenitty
Token ID9760
Feature activation+7.39e+00

Pos logit contributions
 waves+0.213
 shark+0.207
 victorious+0.205
 screams+0.197
 cliff+0.195
Neg logit contributions
isk-0.201
 whether-0.189
ety-0.183
my-0.180
coat-0.166
Token named
Token ID3706
Feature activation+2.74e+00

Pos logit contributions
isk+0.349
 whether+0.323
ety+0.306
my+0.285
coat+0.279
Neg logit contributions
 shark-0.491
 waves-0.486
 boat-0.475
 victorious-0.457
 cliff-0.452
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
isk+0.147
 whether+0.144
 drew+0.128
my+0.128
ety+0.125
Neg logit contributions
 waves-0.142
 shark-0.140
 plunge-0.140
 cliff-0.133
 victorious-0.133
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.239
 drew+0.168
 makeup+0.156
 lipstick+0.152
Which+0.150
Neg logit contributions
 victorious-0.930
ls-0.921
 screams-0.918
ars-0.901
 waves-0.889
Tokeners
Token ID364
Feature activation+1.30e-01

Pos logit contributions
 whether+1.066
isk+1.051
ety+1.013
my+0.973
coat+0.957
Neg logit contributions
 waves-0.672
 victorious-0.670
 shark-0.634
ls-0.616
 screams-0.598
Token.
Token ID13
Feature activation+7.50e-02

Pos logit contributions
isk+0.009
coat+0.009
 whether+0.009
wn+0.009
ety+0.009
Neg logit contributions
 boat-0.004
 waves-0.004
 shark-0.004
 wave-0.004
 victorious-0.004
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
 whether+0.004
epad+0.004
ellery+0.004
isk+0.003
ety+0.003
Neg logit contributions
 waves-0.004
 winds-0.004
 victorious-0.004
 cheers-0.004
 current-0.004
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.237
epad+0.167
 drew+0.152
rings+0.136
 hers+0.132
Neg logit contributions
ls-0.973
 victorious-0.946
 screams-0.925
 waves-0.925
ars-0.908
Tokeners
Token ID364
Feature activation+9.43e-02

Pos logit contributions
 whether+1.068
isk+1.051
ety+1.016
my+0.973
coat+0.959
Neg logit contributions
 waves-0.662
 victorious-0.655
 shark-0.623
ls-0.605
 screams-0.582
Token
Token ID198
Feature activation-7.35e-01

Pos logit contributions
isk+0.022
 whether+0.021
ety+0.021
my+0.020
coat+0.019
Neg logit contributions
 waves-0.018
 victorious-0.018
 shark-0.018
 screams-0.016
 cheers-0.016
Token
Token ID198
Feature activation-7.31e-01

Pos logit contributions
 waves+0.037
 shark+0.036
 victorious+0.035
 screams+0.033
 cliff+0.033
Neg logit contributions
isk-0.047
 whether-0.045
ety-0.043
my-0.042
coat-0.040
TokenMr
Token ID5246
Feature activation+1.74e+00

Pos logit contributions
 victorious+0.039
 waves+0.038
 shark+0.038
 cliff+0.035
 current+0.035
Neg logit contributions
isk-0.043
 whether-0.041
ety-0.040
my-0.039
coat-0.037
Token.
Token ID13
Feature activation+3.73e-01

Pos logit contributions
 whether+0.105
isk+0.100
coat+0.091
my+0.090
ety+0.089
Neg logit contributions
 waves-0.089
 victorious-0.088
 shark-0.086
 screams-0.083
ls-0.083
Token Wh
Token ID854
Feature activation+1.13e+01

Pos logit contributions
 whether+0.018
ety+0.016
isk+0.016
 drew+0.015
my+0.015
Neg logit contributions
 shark-0.023
 victorious-0.022
 waves-0.022
 shore-0.022
 current-0.022
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.146
 drew+0.048
 knit+0.028
tr+0.028
 makeup+0.024
Neg logit contributions
 victorious-1.094
 waves-1.065
ls-1.036
 screams-1.035
 shark-1.025
Tokeners
Token ID364
Feature activation+1.88e-01

Pos logit contributions
isk+0.980
 whether+0.935
ety+0.900
my+0.884
coat+0.831
Neg logit contributions
 waves-0.794
 shark-0.771
 victorious-0.765
 screams-0.724
 cliff-0.718
Token dreams
Token ID10625
Feature activation-9.06e-02

Pos logit contributions
isk+0.010
my+0.009
 whether+0.008
ety+0.008
 drew+0.008
Neg logit contributions
 shark-0.012
 waves-0.011
rier-0.011
 victorious-0.011
ls-0.011
Token were
Token ID547
Feature activation-1.34e+00

Pos logit contributions
 waves+0.005
 shark+0.005
 victorious+0.004
 screams+0.004
 cliff+0.004
Neg logit contributions
isk-0.006
 whether-0.005
my-0.005
ety-0.005
coat-0.005
Token filled
Token ID5901
Feature activation-5.89e-01

Pos logit contributions
 waves+0.072
 shark+0.071
 screams+0.069
 cliff+0.068
 victorious+0.068
Neg logit contributions
isk-0.078
 whether-0.076
my-0.074
ety-0.070
coat-0.065
Token with
Token ID351
Feature activation+2.16e+00

Pos logit contributions
 waves+0.037
 shark+0.036
 victorious+0.036
 screams+0.035
 cliff+0.034
Neg logit contributions
isk-0.030
 whether-0.028
ety-0.026
my-0.026
coat-0.024
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.174
 whether+0.171
my+0.162
ety+0.156
epad+0.152
Neg logit contributions
 shark-0.147
 screams-0.145
 waves-0.144
ls-0.136
 boat-0.136
Tokenisk
Token ID1984
Feature activation+1.57e+01

Pos logit contributions
 whether+0.279
Which+0.229
 Which+0.194
 blonde+0.191
 gloss+0.188
Neg logit contributions
 screams-0.871
ars-0.868
ls-0.860
 victorious-0.849
 waves-0.836
Tokeners
Token ID364
Feature activation+1.66e-01

Pos logit contributions
 whether+1.142
isk+1.096
ety+1.081
coat+1.036
my+1.024
Neg logit contributions
 victorious-0.608
 waves-0.605
ls-0.564
 shark-0.563
 screams-0.525
Token.
Token ID13
Feature activation+2.97e-01

Pos logit contributions
isk+0.012
ety+0.012
coat+0.012
wn+0.012
 whether+0.012
Neg logit contributions
 shark-0.005
 waves-0.005
 boat-0.005
 victorious-0.005
 wave-0.005
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
 whether+0.013
isk+0.012
ety+0.011
epad+0.011
ellery+0.011
Neg logit contributions
 waves-0.019
 winds-0.018
 victorious-0.018
 current-0.017
 losers-0.017
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.254
epad+0.184
rings+0.156
 blonde+0.155
 drew+0.149
Neg logit contributions
ls-0.973
 waves-0.926
 victorious-0.917
 screams-0.912
ars-0.903
Tokeners
Token ID364
Feature activation+1.44e-01

Pos logit contributions
 whether+1.116
isk+1.084
ety+1.058
coat+1.015
my+1.004
Neg logit contributions
 waves-0.620
 victorious-0.613
 shark-0.581
ls-0.573
 screams-0.534
Token loved
Token ID6151
Feature activation+1.37e+00

Pos logit contributions
isk+0.008
 whether+0.008
ety+0.007
coat+0.007
my+0.006
Neg logit contributions
 waves-0.008
 shark-0.008
 boat-0.008
 wave-0.007
 screams-0.007
Token to
Token ID284
Feature activation+9.87e-01

Pos logit contributions
isk+0.084
 whether+0.074
ine+0.073
ety+0.071
my+0.070
Neg logit contributions
 waves-0.080
 shark-0.076
 boat-0.070
 shore-0.069
 cliff-0.068
Token walk
Token ID2513
Feature activation-3.68e-01

Pos logit contributions
isk+0.063
 whether+0.061
ety+0.060
my+0.057
tr+0.052
Neg logit contributions
 waves-0.053
 wave-0.049
 shark-0.048
 boat-0.047
 cliff-0.046
Token around
Token ID1088
Feature activation-2.55e+00

Pos logit contributions
 waves+0.017
 shark+0.016
ls+0.016
 boat+0.015
 victorious+0.015
Neg logit contributions
isk-0.026
 whether-0.024
ety-0.023
my-0.022
ine-0.022
Token an
Token ID281
Feature activation+1.01e+00

Pos logit contributions
 waves+0.023
 screams+0.020
 refreshed+0.020
 shark+0.020
 victorious+0.020
Neg logit contributions
isk-0.032
 whether-0.028
ety-0.026
my-0.025
ine-0.025
Token adorable
Token ID29012
Feature activation+6.63e-01

Pos logit contributions
isk+0.073
 whether+0.069
ety+0.067
my+0.065
coat+0.063
Neg logit contributions
 waves-0.043
 shark-0.040
 boat-0.038
 cliff-0.038
 victorious-0.037
Token cat
Token ID3797
Feature activation+4.94e+00

Pos logit contributions
isk+0.047
 whether+0.044
ety+0.041
my+0.039
If+0.037
Neg logit contributions
 shark-0.033
 boat-0.032
 waves-0.032
 wave-0.029
 cliff-0.029
Token named
Token ID3706
Feature activation+2.91e+00

Pos logit contributions
isk+0.328
ptions+0.307
 whether+0.305
epad+0.298
organ+0.290
Neg logit contributions
 boat-0.214
 shark-0.211
 waves-0.186
 wave-0.178
 whale-0.176
Token Wh
Token ID854
Feature activation+1.15e+01

Pos logit contributions
isk+0.157
 whether+0.155
ety+0.152
my+0.148
coat+0.145
Neg logit contributions
 shark-0.138
 waves-0.134
 screams-0.132
 losers-0.126
 cheers-0.125
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.310
Which+0.262
 blonde+0.221
 Which+0.218
shirt+0.215
Neg logit contributions
 screams-0.788
ars-0.775
ls-0.771
 victorious-0.767
 ourselves-0.759
Tokeners
Token ID364
Feature activation+1.94e-01

Pos logit contributions
 whether+1.216
ety+1.131
isk+1.130
ine+1.087
coat+1.080
Neg logit contributions
 victorious-0.519
 waves-0.493
ls-0.485
 shark-0.438
 losers-0.420
Token.
Token ID13
Feature activation+3.93e-01

Pos logit contributions
wn+0.015
ety+0.014
isk+0.014
 whether+0.014
coat+0.014
Neg logit contributions
 boat-0.005
 victorious-0.005
 waves-0.005
 shark-0.005
 shore-0.004
Token Lily
Token ID20037
Feature activation+4.90e+00

Pos logit contributions
epad+0.023
 whether+0.022
ellery+0.022
ption+0.021
itting+0.021
Neg logit contributions
 waves-0.017
 boat-0.016
 floats-0.016
 cheers-0.016
 Everyone-0.016
Token and
Token ID290
Feature activation+1.09e-01

Pos logit contributions
isk+0.305
 whether+0.296
ety+0.256
epad+0.255
coat+0.253
Neg logit contributions
 waves-0.244
 screams-0.223
 shark-0.220
 boat-0.218
ls-0.208
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
epad+0.005
 whether+0.005
 pad+0.004
coat+0.004
isk+0.004
Neg logit contributions
 waves-0.007
 boat-0.006
 screams-0.006
 shark-0.006
 wave-0.006
Token to
Token ID284
Feature activation+1.05e+00

Pos logit contributions
 waves+0.031
 victorious+0.031
 shark+0.029
 shore+0.028
 boat+0.028
Neg logit contributions
isk-0.042
 whether-0.039
ety-0.038
my-0.036
tr-0.035
Token the
Token ID262
Feature activation-1.07e+00

Pos logit contributions
isk+0.063
ety+0.062
 whether+0.059
my+0.057
coat+0.057
Neg logit contributions
 victorious-0.056
 waves-0.055
 shark-0.054
 shore-0.052
 screams-0.052
Token mailbox
Token ID37282
Feature activation+2.14e+00

Pos logit contributions
 waves+0.065
 shark+0.064
 victorious+0.063
 screams+0.060
ls+0.060
Neg logit contributions
isk-0.056
 whether-0.052
ety-0.050
my-0.050
tr-0.046
Token.
Token ID13
Feature activation+4.89e-01

Pos logit contributions
isk+0.132
 whether+0.122
ety+0.121
my+0.109
coat+0.108
Neg logit contributions
 waves-0.112
 victorious-0.109
 boat-0.106
 shark-0.104
 cheers-0.102
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
isk+0.031
 whether+0.031
ety+0.029
ptions+0.028
my+0.027
Neg logit contributions
 victorious-0.020
 waves-0.020
 screams-0.020
 cheers-0.019
 Everyone-0.019
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.233
 blonde+0.146
 drew+0.143
 blond+0.141
 makeup+0.128
Neg logit contributions
 victorious-0.971
ls-0.965
 waves-0.916
 losers-0.910
 screams-0.903
Tokeners
Token ID364
Feature activation+2.78e-01

Pos logit contributions
 whether+1.104
isk+1.067
ety+1.059
coat+1.005
my+1.000
Neg logit contributions
 victorious-0.639
 waves-0.620
 shark-0.580
ls-0.565
 screams-0.539
Token followed
Token ID3940
Feature activation-2.40e+00

Pos logit contributions
isk+0.014
 whether+0.013
my+0.013
ety+0.012
 drew+0.011
Neg logit contributions
 waves-0.017
 shark-0.017
 victorious-0.017
 cliff-0.016
 screams-0.016
Token her
Token ID607
Feature activation+5.18e+00

Pos logit contributions
 victorious+0.104
 waves+0.099
 shark+0.095
 boat+0.093
 screams+0.092
Neg logit contributions
ety-0.162
isk-0.161
 whether-0.154
tr-0.152
coat-0.151
Token and
Token ID290
Feature activation+1.72e-01

Pos logit contributions
isk+0.298
 whether+0.277
ety+0.270
my+0.260
coat+0.252
Neg logit contributions
 waves-0.288
 victorious-0.287
 shark-0.275
 screams-0.272
 boat-0.270
Token watched
Token ID7342
Feature activation-4.66e-01

Pos logit contributions
isk+0.009
 whether+0.009
ety+0.008
my+0.008
coat+0.008
Neg logit contributions
 waves-0.010
 victorious-0.010
 shark-0.010
 boat-0.010
 screams-0.009
Tokeners
Token ID364
Feature activation+1.15e-01

Pos logit contributions
 whether+1.116
isk+1.081
ety+1.054
coat+1.007
my+1.000
Neg logit contributions
 waves-0.629
 victorious-0.628
 shark-0.600
ls-0.583
 screams-0.552
Token.
Token ID13
Feature activation+1.26e-01

Pos logit contributions
isk+0.008
 whether+0.008
ety+0.007
coat+0.007
wn+0.007
Neg logit contributions
 waves-0.005
 shark-0.005
 boat-0.005
 wave-0.004
 screams-0.004
Token Mr
Token ID1770
Feature activation+1.11e+00

Pos logit contributions
 whether+0.008
isk+0.008
ety+0.007
ine+0.007
ellery+0.007
Neg logit contributions
 waves-0.006
 shark-0.005
 current-0.005
 cheers-0.005
 victorious-0.005
Token.
Token ID13
Feature activation+1.18e-01

Pos logit contributions
 whether+0.070
isk+0.067
coat+0.061
ety+0.059
my+0.059
Neg logit contributions
 waves-0.054
 shark-0.054
ls-0.052
 victorious-0.051
 screams-0.051
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
 whether+0.005
shirt+0.005
 drew+0.005
isk+0.004
ety+0.004
Neg logit contributions
 shark-0.007
 current-0.007
 victorious-0.007
 waves-0.007
 shore-0.007
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.233
rings+0.123
Which+0.122
epad+0.118
 hers+0.110
Neg logit contributions
ls-0.974
 waves-0.969
 victorious-0.953
 screams-0.952
 shark-0.941
Tokeners
Token ID364
Feature activation+6.72e-02

Pos logit contributions
 whether+1.085
isk+1.065
ety+1.025
my+0.981
coat+0.978
Neg logit contributions
 waves-0.654
 victorious-0.649
 shark-0.628
ls-0.601
 screams-0.573
Token loved
Token ID6151
Feature activation+1.23e+00

Pos logit contributions
isk+0.004
 whether+0.003
ety+0.003
coat+0.003
my+0.003
Neg logit contributions
 waves-0.004
 shark-0.004
 boat-0.004
 victorious-0.004
 screams-0.004
Token to
Token ID284
Feature activation+7.99e-01

Pos logit contributions
isk+0.077
ety+0.068
ine+0.065
 whether+0.065
coat+0.064
Neg logit contributions
 waves-0.070
 shark-0.067
 shore-0.061
 boat-0.059
 cliff-0.059
Token study
Token ID2050
Feature activation+7.73e+00

Pos logit contributions
isk+0.050
ety+0.049
 whether+0.048
my+0.046
tr+0.042
Neg logit contributions
 waves-0.042
 wave-0.040
 shark-0.039
 boat-0.038
 cliff-0.038
Token.
Token ID13
Feature activation+1.25e-01

Pos logit contributions
isk+0.471
ety+0.452
 whether+0.414
my+0.400
coat+0.391
Neg logit contributions
 waves-0.438
 victorious-0.411
 shark-0.409
ls-0.394
 cheers-0.386
Token a
Token ID257
Feature activation+1.10e+00

Pos logit contributions
 waves+0.006
 shark+0.006
 victorious+0.006
 refreshed+0.006
 shore+0.005
Neg logit contributions
isk-0.008
 whether-0.007
ety-0.007
coat-0.006
my-0.006
Token lazy
Token ID16931
Feature activation-1.50e+00

Pos logit contributions
isk+0.076
 whether+0.073
ety+0.071
my+0.071
coat+0.066
Neg logit contributions
 waves-0.047
 victorious-0.045
 shark-0.044
 screams-0.043
 plunge-0.042
Token cat
Token ID3797
Feature activation+4.86e+00

Pos logit contributions
 waves+0.078
 victorious+0.077
 shark+0.074
 screams+0.073
 plunge+0.071
Neg logit contributions
isk-0.090
 whether-0.085
ety-0.082
my-0.082
coat-0.077
Token named
Token ID3706
Feature activation+2.75e+00

Pos logit contributions
isk+0.270
 whether+0.249
ety+0.244
coat+0.231
ptions+0.228
Neg logit contributions
 shark-0.277
 boat-0.264
 waves-0.260
 whale-0.244
 wave-0.244
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
isk+0.143
 whether+0.140
my+0.131
ety+0.130
 drew+0.128
Neg logit contributions
 shark-0.145
 waves-0.139
 screams-0.137
 losers-0.134
 plunge-0.133
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.251
Which+0.194
 blonde+0.174
 drew+0.169
 Which+0.167
Neg logit contributions
 victorious-0.886
ls-0.883
 screams-0.874
ars-0.867
 waves-0.836
Tokeners
Token ID364
Feature activation+7.61e-02

Pos logit contributions
 whether+1.115
isk+1.071
ety+1.057
coat+1.009
my+1.002
Neg logit contributions
 victorious-0.632
 waves-0.618
 shark-0.576
ls-0.573
 screams-0.545
Token.
Token ID13
Feature activation+1.82e-01

Pos logit contributions
wn+0.006
isk+0.005
coat+0.005
ety+0.005
 whether+0.005
Neg logit contributions
 boat-0.002
 shark-0.002
 waves-0.002
 victorious-0.002
 wave-0.002
Token Wh
Token ID854
Feature activation+1.14e+01

Pos logit contributions
 whether+0.008
epad+0.008
ellery+0.008
ety+0.007
isk+0.007
Neg logit contributions
 waves-0.011
 winds-0.010
 current-0.010
 victorious-0.010
 losers-0.010
Tokenisk
Token ID1984
Feature activation+1.56e+01

Pos logit contributions
 whether+0.251
epad+0.172
 blonde+0.168
rings+0.159
 blond+0.157
Neg logit contributions
ls-0.962
 victorious-0.923
 waves-0.907
 screams-0.893
ars-0.882
Tokeners
Token ID364
Feature activation+6.86e-02

Pos logit contributions
 whether+1.089
isk+1.059
ety+1.034
coat+0.986
my+0.983
Neg logit contributions
 waves-0.640
 victorious-0.638
 shark-0.602
ls-0.586
 screams-0.557
Token learned
Token ID4499
Feature activation-1.76e+00

Pos logit contributions
rier+0.131
 plunge+0.129
 shark+0.128
 lighthouse+0.126
 cliff+0.125
Neg logit contributions
isk-0.082
my-0.079
 whether-0.071
 drew-0.068
ety-0.067
Token that
Token ID326
Feature activation-7.22e-01

Pos logit contributions
 waves+0.101
 shark+0.098
 victorious+0.097
 screams+0.093
 cliff+0.091
Neg logit contributions
isk-0.102
 whether-0.094
ety-0.093
my-0.090
coat-0.085
Token just
Token ID655
Feature activation-2.51e+00

Pos logit contributions
 waves+0.040
 shark+0.039
 victorious+0.039
 screams+0.037
 cliff+0.037
Neg logit contributions
isk-0.043
 whether-0.040
ety-0.039
my-0.038
coat-0.035
Token because
Token ID780
Feature activation+5.70e+00

Pos logit contributions
 waves+0.153
 shark+0.148
 victorious+0.147
 screams+0.142
 current+0.141
Neg logit contributions
isk-0.135
 whether-0.120
ety-0.119
my-0.114
coat-0.112
Token someone
Token ID2130
Feature activation+1.72e-01

Pos logit contributions
 whether+0.315
my+0.294
isk+0.288
ety+0.275
 drew+0.254
Neg logit contributions
 victorious-0.319
 waves-0.311
 screams-0.300
rier-0.297
 shark-0.297
Token waves
Token ID9813
Feature activation-1.29e+01

Pos logit contributions
isk+0.008
 whether+0.008
my+0.008
ety+0.008
 drew+0.007
Neg logit contributions
 victorious-0.011
 waves-0.011
 current-0.011
 shark-0.010
rier-0.010
Token at
Token ID379
Feature activation-8.21e-01

Pos logit contributions
 shark+0.752
 waves+0.751
 cliff+0.742
 victorious+0.739
 lighthouse+0.726
Neg logit contributions
isk-0.659
 whether-0.654
my-0.636
ety-0.615
coat-0.565
Token you
Token ID345
Feature activation-6.92e+00

Pos logit contributions
 waves+0.046
 victorious+0.045
 shark+0.045
 screams+0.043
 current+0.042
Neg logit contributions
isk-0.047
 whether-0.045
my-0.043
ety-0.043
ine-0.038
Token,
Token ID11
Feature activation+7.92e-01

Pos logit contributions
 waves+0.440
 shark+0.433
 victorious+0.425
 cliff+0.412
 current+0.409
Neg logit contributions
isk-0.344
 whether-0.330
my-0.310
ety-0.307
coat-0.272
Token it
Token ID340
Feature activation+2.41e+00

Pos logit contributions
 whether+0.033
isk+0.032
my+0.029
ety+0.028
coat+0.024
Neg logit contributions
 waves-0.055
 shark-0.055
 victorious-0.055
 screams-0.053
ls-0.053
Token does
Token ID857
Feature activation+4.85e-01

Pos logit contributions
my+0.100
isk+0.090
 whether+0.087
c+0.080
ety+0.080
Neg logit contributions
rier-0.167
 lighthouse-0.166
 shark-0.166
 victorious-0.165
 shore-0.164
Token to
Token ID284
Feature activation+7.67e-01

Pos logit contributions
isk+0.053
 whether+0.045
ety+0.045
coat+0.043
my+0.043
Neg logit contributions
 waves-0.063
 shark-0.059
 shore-0.055
 screams-0.055
 cliff-0.054
Token surf
Token ID9053
Feature activation-2.23e+00

Pos logit contributions
isk+0.048
ety+0.044
 whether+0.044
my+0.041
tr+0.039
Neg logit contributions
 waves-0.043
 wave-0.040
 shark-0.039
 boat-0.039
 cliff-0.037
Token on
Token ID319
Feature activation-1.09e+00

Pos logit contributions
 victorious+0.136
 shark+0.135
 screams+0.134
 current+0.134
 waves+0.133
Neg logit contributions
isk-0.110
 whether-0.105
my-0.105
ety-0.100
coat-0.094
Token the
Token ID262
Feature activation-1.41e+00

Pos logit contributions
 waves+0.058
 shark+0.055
 victorious+0.054
 screams+0.052
 cliff+0.052
Neg logit contributions
isk-0.068
 whether-0.064
ety-0.062
my-0.061
coat-0.057
Token big
Token ID1263
Feature activation-4.82e+00

Pos logit contributions
 victorious+0.063
ls+0.062
 terrified+0.060
rier+0.059
 Sorry+0.059
Neg logit contributions
my-0.085
isk-0.082
 whether-0.079
ine-0.077
coat-0.077
Token waves
Token ID9813
Feature activation-1.29e+01

Pos logit contributions
 victorious+0.238
rier+0.217
 terrified+0.212
ls+0.211
 drown+0.210
Neg logit contributions
my-0.308
isk-0.284
 whether-0.283
ety-0.276
coat-0.267
Token at
Token ID379
Feature activation-8.16e-01

Pos logit contributions
 waves+0.802
 shark+0.788
 victorious+0.782
 screams+0.753
 cliff+0.745
Neg logit contributions
isk-0.650
 whether-0.615
ety-0.587
my-0.581
coat-0.533
Token the
Token ID262
Feature activation-1.33e+00

Pos logit contributions
 victorious+0.042
 shark+0.042
 waves+0.042
 screams+0.040
 current+0.040
Neg logit contributions
isk-0.048
 whether-0.046
my-0.045
ety-0.045
 drew-0.040
Token beach
Token ID10481
Feature activation-5.66e+00

Pos logit contributions
ls+0.057
 victorious+0.055
aro+0.053
 screams+0.053
 terrified+0.052
Neg logit contributions
my-0.086
isk-0.083
 whether-0.082
ety-0.078
 library-0.076
Token.
Token ID13
Feature activation+8.52e-01

Pos logit contributions
 waves+0.353
 shark+0.346
 victorious+0.343
 screams+0.330
 cliff+0.326
Neg logit contributions
isk-0.290
 whether-0.272
ety-0.259
my-0.255
coat-0.235
Token One
Token ID1881
Feature activation+5.13e+00

Pos logit contributions
isk+0.047
 whether+0.041
coat+0.040
organized+0.040
organ+0.040
Neg logit contributions
 waves-0.046
 cheers-0.042
 cliff-0.042
 boat-0.042
 shark-0.042
Token spend
Token ID4341
Feature activation-7.12e-01

Pos logit contributions
 whether+0.044
isk+0.043
my+0.041
ine+0.039
 drew+0.038
Neg logit contributions
ls-0.057
rier-0.056
 victorious-0.055
 shark-0.054
 waves-0.052
Token the
Token ID262
Feature activation-1.73e+00

Pos logit contributions
 waves+0.039
 shark+0.038
 victorious+0.038
 screams+0.036
 cliff+0.036
Neg logit contributions
isk-0.041
 whether-0.039
ety-0.038
my-0.037
coat-0.035
Token day
Token ID1110
Feature activation+1.20e+00

Pos logit contributions
 victorious+0.093
 shark+0.088
 waves+0.087
 lighthouse+0.085
rier+0.084
Neg logit contributions
isk-0.097
my-0.096
 whether-0.095
coat-0.092
ety-0.088
Token chasing
Token ID20023
Feature activation-1.39e+00

Pos logit contributions
 whether+0.053
my+0.052
isk+0.051
coat+0.049
ety+0.047
Neg logit contributions
 shark-0.077
 waves-0.075
 victorious-0.074
 current-0.074
 lighthouse-0.073
Token the
Token ID262
Feature activation-1.81e+00

Pos logit contributions
 waves+0.076
 victorious+0.075
 shark+0.074
 lighthouse+0.071
 cliff+0.071
Neg logit contributions
isk-0.079
 whether-0.076
ety-0.072
my-0.071
coat-0.066
Token waves
Token ID9813
Feature activation-1.28e+01

Pos logit contributions
 victorious+0.091
 waves+0.084
ls+0.082
 screams+0.082
 sinks+0.081
Neg logit contributions
isk-0.106
my-0.104
 whether-0.103
ety-0.098
coat-0.098
Token.
Token ID13
Feature activation+2.66e-01

Pos logit contributions
 waves+0.812
 victorious+0.800
 shark+0.797
 screams+0.780
 lighthouse+0.766
Neg logit contributions
isk-0.605
 whether-0.589
my-0.567
ety-0.551
coat-0.520
Token
Token ID220
Feature activation+1.97e-01

Pos logit contributions
isk+0.015
 whether+0.014
ety+0.014
my+0.013
coat+0.012
Neg logit contributions
 waves-0.015
 shark-0.015
 victorious-0.015
 screams-0.014
 cliff-0.014
Token
Token ID198
Feature activation-7.37e-01

Pos logit contributions
isk+0.011
 whether+0.010
ety+0.010
my+0.010
coat+0.009
Neg logit contributions
 waves-0.012
 shark-0.011
 victorious-0.011
 screams-0.011
 cliff-0.011
TokenMom
Token ID29252
Feature activation+1.32e-02

Pos logit contributions
 waves+0.039
 victorious+0.038
 shark+0.038
 screams+0.036
 cheers+0.035
Neg logit contributions
isk-0.044
 whether-0.042
ety-0.040
my-0.039
coat-0.037
Tokenmy
Token ID1820
Feature activation+1.06e+01

Pos logit contributions
isk+0.000
 whether+0.000
ety+0.000
my+0.000
coat+0.000
Neg logit contributions
 waves-0.001
 shark-0.001
 victorious-0.001
 screams-0.001
 cliff-0.001
Token water
Token ID1660
Feature activation-4.21e+00

Pos logit contributions
 victorious+0.085
ls+0.080
 screams+0.076
 losers+0.075
 shark+0.074
Neg logit contributions
my-0.118
isk-0.114
 whether-0.114
ety-0.112
 makeup-0.109
Token.
Token ID13
Feature activation+4.35e-02

Pos logit contributions
 victorious+0.267
 shark+0.260
 losers+0.252
 screams+0.249
 waves+0.246
Neg logit contributions
isk-0.189
my-0.182
 whether-0.181
ety-0.175
coat-0.168
Token They
Token ID1119
Feature activation-2.33e+00

Pos logit contributions
isk+0.002
 whether+0.002
ety+0.002
my+0.002
coat+0.002
Neg logit contributions
 waves-0.003
 shark-0.003
 victorious-0.002
 screams-0.002
 cheers-0.002
Token saw
Token ID2497
Feature activation+1.49e+00

Pos logit contributions
rier+0.145
 victorious+0.143
 shark+0.140
ls+0.136
 waves+0.132
Neg logit contributions
isk-0.113
 whether-0.110
my-0.109
 drew-0.105
ety-0.102
Token the
Token ID262
Feature activation-1.51e+00

Pos logit contributions
isk+0.077
 whether+0.076
my+0.071
ety+0.071
coat+0.063
Neg logit contributions
 victorious-0.087
 waves-0.085
 shark-0.085
 cliff-0.081
 screams-0.081
Token waves
Token ID9813
Feature activation-1.28e+01

Pos logit contributions
 victorious+0.069
vation+0.063
 losers+0.060
 screams+0.060
 cheers+0.059
Neg logit contributions
my-0.091
 whether-0.089
ety-0.088
 makeup-0.086
isk-0.084
Token and
Token ID290
Feature activation+4.78e-02

Pos logit contributions
 shark+0.818
 victorious+0.816
 waves+0.812
 screams+0.788
 cliff+0.780
Neg logit contributions
isk-0.571
 whether-0.561
my-0.550
ety-0.524
coat-0.482
Token the
Token ID262
Feature activation-1.31e+00

Pos logit contributions
 whether+0.003
isk+0.003
ety+0.003
my+0.003
 drew+0.003
Neg logit contributions
 victorious-0.002
rier-0.002
 shark-0.002
ls-0.002
 lighthouse-0.002
Token shells
Token ID19679
Feature activation-1.27e+00

Pos logit contributions
 victorious+0.054
vation+0.048
 losers+0.047
rier+0.045
ls+0.044
Neg logit contributions
my-0.084
ety-0.080
 makeup-0.080
 whether-0.079
coat-0.076
Token.
Token ID13
Feature activation+4.14e-01

Pos logit contributions
 victorious+0.081
 shark+0.081
 losers+0.078
 screams+0.077
 waves+0.077
Neg logit contributions
 whether-0.054
my-0.053
isk-0.052
ety-0.050
coat-0.048
Token They
Token ID1119
Feature activation-2.05e+00

Pos logit contributions
isk+0.025
 whether+0.023
ety+0.022
my+0.022
coat+0.020
Neg logit contributions
 waves-0.023
 shark-0.022
 victorious-0.022
 screams-0.021
 cliff-0.021
Token and
Token ID290
Feature activation-1.52e-01

Pos logit contributions
 victorious+0.107
 shark+0.107
rier+0.106
 waves+0.103
 plunge+0.103
Neg logit contributions
isk-0.069
my-0.068
 whether-0.066
ety-0.063
 drew-0.062
Token the
Token ID262
Feature activation-1.56e+00

Pos logit contributions
 waves+0.010
 shark+0.010
 victorious+0.010
 screams+0.009
 cliff+0.009
Neg logit contributions
isk-0.007
 whether-0.007
ety-0.006
my-0.006
coat-0.006
Token dog
Token ID3290
Feature activation-2.40e+00

Pos logit contributions
 victorious+0.076
 waves+0.072
 screams+0.069
 shark+0.069
ls+0.069
Neg logit contributions
isk-0.096
my-0.091
 whether-0.091
ety-0.091
coat-0.085
Token made
Token ID925
Feature activation+2.92e+00

Pos logit contributions
rier+0.156
 shark+0.150
 victorious+0.148
 lighthouse+0.144
 cliff+0.142
Neg logit contributions
isk-0.116
my-0.110
 whether-0.109
ety-0.104
 drew-0.099
Token BIG
Token ID26746
Feature activation-3.37e+00

Pos logit contributions
isk+0.158
 whether+0.150
my+0.143
ety+0.141
coat+0.127
Neg logit contributions
 waves-0.169
 shark-0.168
 victorious-0.166
 cliff-0.161
 lighthouse-0.159
Token waves
Token ID9813
Feature activation-1.28e+01

Pos logit contributions
 victorious+0.158
 lighthouse+0.149
 cliff+0.146
vation+0.146
rier+0.137
Neg logit contributions
 whether-0.208
my-0.202
isk-0.197
ety-0.186
 makeup-0.185
Token in
Token ID287
Feature activation-3.71e+00

Pos logit contributions
 shark+0.861
 waves+0.856
 victorious+0.852
 cliff+0.843
 lighthouse+0.841
Neg logit contributions
 whether-0.526
isk-0.525
my-0.503
ety-0.488
coat-0.458
Token the
Token ID262
Feature activation-1.40e+00

Pos logit contributions
 waves+0.180
 shark+0.175
 victorious+0.173
 screams+0.165
 cliff+0.163
Neg logit contributions
isk-0.240
 whether-0.231
ety-0.220
my-0.219
coat-0.203
Token sea
Token ID5417
Feature activation-7.22e+00

Pos logit contributions
keys+0.066
 victorious+0.065
rier+0.064
aro+0.062
 losers+0.062
Neg logit contributions
my-0.086
 makeup-0.078
ety-0.078
c-0.076
 drew-0.074
Token and
Token ID290
Feature activation-6.10e-02

Pos logit contributions
 shark+0.511
 victorious+0.508
 screams+0.501
 waves+0.501
 losers+0.490
Neg logit contributions
 whether-0.275
isk-0.273
my-0.264
ety-0.247
ine-0.222
Token had
Token ID550
Feature activation-7.44e-01

Pos logit contributions
 victorious+0.003
 shark+0.003
rier+0.003
 waves+0.003
ls+0.003
Neg logit contributions
isk-0.004
 whether-0.004
my-0.003
ety-0.003
 drew-0.003
Token sea
Token ID5417
Feature activation-7.26e+00

Pos logit contributions
 waves+0.101
 shark+0.099
 victorious+0.099
 screams+0.095
 cliff+0.093
Neg logit contributions
isk-0.118
 whether-0.111
my-0.109
ety-0.107
coat-0.099
Token.
Token ID13
Feature activation-2.96e-02

Pos logit contributions
 screams+0.534
 losers+0.528
 victorious+0.512
 hawk+0.503
 plunge+0.497
Neg logit contributions
 whether-0.260
my-0.253
isk-0.234
ety-0.216
ine-0.209
Token They
Token ID1119
Feature activation-2.45e+00

Pos logit contributions
 waves+0.002
 shark+0.002
 victorious+0.002
 screams+0.001
 current+0.001
Neg logit contributions
isk-0.002
 whether-0.002
ety-0.002
my-0.002
coat-0.001
Token enjoyed
Token ID8359
Feature activation-1.83e+00

Pos logit contributions
rier+0.169
 victorious+0.163
ls+0.162
 shark+0.161
 screams+0.160
Neg logit contributions
isk-0.104
 whether-0.101
my-0.100
 drew-0.097
ety-0.092
Token the
Token ID262
Feature activation-1.53e+00

Pos logit contributions
 waves+0.091
 shark+0.084
 victorious+0.083
 shore+0.080
 boat+0.079
Neg logit contributions
isk-0.125
ety-0.118
 whether-0.114
my-0.112
tr-0.108
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
 victorious+0.073
 screams+0.069
rier+0.069
 losers+0.067
 sinks+0.066
Neg logit contributions
my-0.094
 whether-0.091
isk-0.091
ety-0.084
coat-0.084
Token and
Token ID290
Feature activation-3.32e-01

Pos logit contributions
 shark+0.850
 waves+0.844
 victorious+0.839
 screams+0.836
 cliff+0.823
Neg logit contributions
isk-0.535
my-0.519
 whether-0.518
ety-0.489
coat-0.467
Token the
Token ID262
Feature activation-1.38e+00

Pos logit contributions
 shark+0.018
 victorious+0.017
 waves+0.017
rier+0.017
 screams+0.017
Neg logit contributions
isk-0.019
 whether-0.019
my-0.018
ety-0.018
 drew-0.016
Token feeling
Token ID4203
Feature activation-2.55e+00

Pos logit contributions
 victorious+0.057
rier+0.053
ls+0.053
 screams+0.053
 losers+0.053
Neg logit contributions
my-0.091
 whether-0.083
ety-0.082
isk-0.080
 makeup-0.080
Token of
Token ID286
Feature activation-5.09e+00

Pos logit contributions
 shark+0.139
 screams+0.139
 current+0.136
 cliff+0.136
 plunge+0.135
Neg logit contributions
 whether-0.125
my-0.123
 drew-0.121
isk-0.120
c-0.118
Token being
Token ID852
Feature activation+1.40e+00

Pos logit contributions
 shark+0.246
ls+0.243
 victorious+0.242
 waves+0.237
rier+0.236
Neg logit contributions
isk-0.309
 whether-0.308
my-0.286
ety-0.279
 drew-0.260
Token face
Token ID1986
Feature activation-4.37e+00

Pos logit contributions
isk+0.256
 whether+0.254
my+0.254
ety+0.244
tr+0.237
Neg logit contributions
 shark-0.230
 victorious-0.226
 lighthouse-0.213
 losers-0.212
 waves-0.211
Token,
Token ID11
Feature activation+4.78e-01

Pos logit contributions
 waves+0.255
 shark+0.241
 victorious+0.240
 screams+0.232
 cheers+0.228
Neg logit contributions
isk-0.249
 whether-0.230
ety-0.224
my-0.212
coat-0.201
Token and
Token ID290
Feature activation-1.15e-01

Pos logit contributions
isk+0.025
 whether+0.023
ety+0.023
my+0.021
coat+0.020
Neg logit contributions
 waves-0.030
 shark-0.028
 victorious-0.028
 screams-0.027
 cliff-0.027
Token watching
Token ID4964
Feature activation+3.24e+00

Pos logit contributions
 waves+0.007
 shark+0.007
 victorious+0.007
 cliff+0.007
 current+0.007
Neg logit contributions
isk-0.006
 whether-0.005
my-0.005
ety-0.005
coat-0.004
Token the
Token ID262
Feature activation-1.30e+00

Pos logit contributions
isk+0.173
 whether+0.165
my+0.156
ety+0.155
coat+0.141
Neg logit contributions
 waves-0.191
 shark-0.187
 victorious-0.187
 screams-0.178
 current-0.177
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
 victorious+0.060
 shark+0.055
 sinks+0.054
 drown+0.053
 losers+0.053
Neg logit contributions
my-0.079
isk-0.076
 whether-0.076
ety-0.074
 makeup-0.072
Token.
Token ID13
Feature activation+4.64e-01

Pos logit contributions
 shark+0.759
 waves+0.747
 victorious+0.747
 screams+0.717
 lighthouse+0.717
Neg logit contributions
isk-0.645
 whether-0.621
my-0.615
ety-0.592
coat-0.551
Token
Token ID220
Feature activation+8.41e-02

Pos logit contributions
isk+0.023
 whether+0.021
ety+0.020
my+0.020
coat+0.019
Neg logit contributions
 waves-0.029
 shark-0.028
 victorious-0.028
 cheers-0.027
 screams-0.027
Token
Token ID198
Feature activation-1.03e+00

Pos logit contributions
isk+0.005
 whether+0.005
ety+0.004
my+0.004
coat+0.004
Neg logit contributions
 waves-0.005
 victorious-0.004
 shark-0.004
 screams-0.004
 cliff-0.004
Token
Token ID198
Feature activation-1.00e+00

Pos logit contributions
 victorious+0.046
 waves+0.045
 shark+0.044
 cheers+0.043
 screams+0.043
Neg logit contributions
isk-0.068
 whether-0.064
ety-0.063
coat-0.060
tr-0.059
TokenOne
Token ID3198
Feature activation+7.07e+00

Pos logit contributions
 victorious+0.054
 waves+0.052
 shark+0.051
 cliff+0.051
 cheers+0.050
Neg logit contributions
isk-0.057
 whether-0.052
ety-0.052
my-0.050
coat-0.049
TokenThe
Token ID464
Feature activation-1.25e+00

Pos logit contributions
 waves+0.057
 victorious+0.055
 shark+0.055
 screams+0.052
 cliff+0.052
Neg logit contributions
isk-0.058
 whether-0.054
ety-0.053
my-0.051
coat-0.048
Token wind
Token ID2344
Feature activation-5.25e+00

Pos logit contributions
ls+0.080
rier+0.076
 Sorry+0.075
keys+0.074
vation+0.073
Neg logit contributions
my-0.053
One-0.044
Today-0.042
 one-0.041
 whether-0.038
Token blew
Token ID17948
Feature activation-1.92e+00

Pos logit contributions
 cliff+0.343
 victorious+0.335
rier+0.329
 plunge+0.328
 refreshed+0.327
Neg logit contributions
my-0.225
isk-0.204
 drew-0.202
tr-0.195
ine-0.185
Token and
Token ID290
Feature activation-3.85e-01

Pos logit contributions
 waves+0.106
 shark+0.104
 victorious+0.102
 cliff+0.098
 screams+0.098
Neg logit contributions
isk-0.112
 whether-0.106
my-0.102
ety-0.101
coat-0.093
Token the
Token ID262
Feature activation-1.64e+00

Pos logit contributions
 victorious+0.022
 waves+0.021
 shark+0.021
rier+0.021
 screams+0.021
Neg logit contributions
isk-0.021
 whether-0.020
my-0.020
ety-0.019
 drew-0.018
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
 Sorry+0.086
ls+0.081
keys+0.080
aro+0.077
opher+0.073
Neg logit contributions
my-0.081
 one-0.074
 color-0.072
 pen-0.071
 rag-0.070
Token rocked
Token ID36872
Feature activation-1.75e+00

Pos logit contributions
 shark+0.746
 cliff+0.720
 waves+0.718
 victorious+0.709
 screams+0.694
Neg logit contributions
isk-0.639
 whether-0.592
my-0.588
coat-0.588
 drew-0.586
Token,
Token ID11
Feature activation+4.68e-01

Pos logit contributions
 waves+0.089
 victorious+0.088
 shark+0.084
 screams+0.080
 current+0.080
Neg logit contributions
isk-0.110
 whether-0.102
ety-0.101
my-0.095
coat-0.092
Token but
Token ID475
Feature activation+2.42e+00

Pos logit contributions
isk+0.021
 whether+0.020
my+0.019
ety+0.019
coat+0.017
Neg logit contributions
 waves-0.031
 shark-0.031
 victorious-0.030
 screams-0.029
 cliff-0.029
Token with
Token ID351
Feature activation+1.85e+00

Pos logit contributions
 whether+0.118
isk+0.113
my+0.110
ety+0.103
 drew+0.094
Neg logit contributions
 waves-0.156
 shark-0.149
 victorious-0.147
 screams-0.146
ls-0.143
Token Jane
Token ID12091
Feature activation-3.00e+00

Pos logit contributions
isk+0.110
 whether+0.103
ety+0.101
my+0.097
coat+0.092
Neg logit contributions
 waves-0.102
 shark-0.099
 victorious-0.098
 screams-0.093
 cliff-0.092
Token.
Token ID13
Feature activation+1.76e-01

Pos logit contributions
isk+0.153
 whether+0.142
ety+0.138
my+0.132
coat+0.126
Neg logit contributions
 waves-0.160
 shark-0.156
 victorious-0.156
 screams-0.149
 current-0.146
Token He
Token ID679
Feature activation-2.65e+00

Pos logit contributions
isk+0.013
 whether+0.012
ety+0.012
coat+0.012
 knit+0.011
Neg logit contributions
 waves-0.007
 screams-0.006
 shark-0.006
 victorious-0.006
 cheers-0.006
Token looked
Token ID3114
Feature activation+1.11e+00

Pos logit contributions
rier+0.169
 victorious+0.164
 plunge+0.164
 shark+0.161
 cliff+0.160
Neg logit contributions
isk-0.132
 whether-0.124
my-0.123
ety-0.121
 drew-0.114
Token at
Token ID379
Feature activation-7.86e-01

Pos logit contributions
isk+0.063
 whether+0.061
my+0.058
ety+0.057
coat+0.052
Neg logit contributions
 waves-0.063
 shark-0.061
 victorious-0.060
 screams-0.059
 cliff-0.059
Token the
Token ID262
Feature activation-1.41e+00

Pos logit contributions
 waves+0.040
 shark+0.039
 victorious+0.038
 screams+0.036
 cliff+0.036
Neg logit contributions
isk-0.050
 whether-0.047
ety-0.046
my-0.044
coat-0.043
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
aro+0.053
 victorious+0.053
 losers+0.052
keys+0.052
rier+0.048
Neg logit contributions
my-0.087
 whether-0.083
 which-0.080
 makeup-0.080
 drew-0.079
Token again
Token ID757
Feature activation-2.67e+00

Pos logit contributions
 waves+0.809
 shark+0.791
 victorious+0.784
 screams+0.756
 cliff+0.751
Neg logit contributions
isk-0.628
 whether-0.591
ety-0.562
my-0.557
coat-0.508
Token and
Token ID290
Feature activation-2.27e-01

Pos logit contributions
 waves+0.169
 shark+0.165
 victorious+0.164
 screams+0.157
 cliff+0.157
Neg logit contributions
isk-0.136
 whether-0.125
ety-0.120
my-0.117
coat-0.110
Token saw
Token ID2497
Feature activation+1.49e+00

Pos logit contributions
 waves+0.014
 shark+0.013
 victorious+0.013
 screams+0.013
 cliff+0.012
Neg logit contributions
isk-0.012
 whether-0.011
ety-0.011
my-0.011
coat-0.010
Token that
Token ID326
Feature activation-1.76e-01

Pos logit contributions
isk+0.081
 whether+0.080
my+0.075
ety+0.074
tr+0.067
Neg logit contributions
 waves-0.084
 victorious-0.083
 shark-0.083
 screams-0.080
 cliff-0.079
Token they
Token ID484
Feature activation-4.99e+00

Pos logit contributions
 waves+0.009
 victorious+0.009
 shark+0.009
 screams+0.009
 cliff+0.009
Neg logit contributions
isk-0.010
 whether-0.010
my-0.009
ety-0.009
coat-0.008
Token that
Token ID326
Feature activation-3.56e-01

Pos logit contributions
isk+0.098
 whether+0.093
ety+0.088
my+0.087
coat+0.080
Neg logit contributions
 waves-0.105
 shark-0.102
 victorious-0.101
 screams-0.097
 cliff-0.096
Token,
Token ID11
Feature activation+2.82e-01

Pos logit contributions
 waves+0.022
 victorious+0.022
 shark+0.022
 screams+0.021
 lighthouse+0.020
Neg logit contributions
isk-0.017
 whether-0.017
my-0.017
ety-0.016
ine-0.015
Token he
Token ID339
Feature activation-2.61e+00

Pos logit contributions
isk+0.015
 whether+0.015
my+0.014
ety+0.014
ine+0.012
Neg logit contributions
 waves-0.016
 shark-0.016
 victorious-0.016
 screams-0.015
 current-0.015
Token watched
Token ID7342
Feature activation-8.06e-01

Pos logit contributions
rier+0.175
 shark+0.171
 victorious+0.169
 plunge+0.169
 current+0.165
Neg logit contributions
isk-0.115
my-0.113
 drew-0.108
 whether-0.108
ety-0.102
Token the
Token ID262
Feature activation-1.65e+00

Pos logit contributions
 waves+0.047
 shark+0.046
 victorious+0.046
 cliff+0.044
 screams+0.044
Neg logit contributions
isk-0.043
 whether-0.042
my-0.039
ety-0.037
ine-0.034
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
 victorious+0.077
ls+0.074
 waves+0.073
 shark+0.073
 sinks+0.070
Neg logit contributions
isk-0.100
my-0.097
 whether-0.096
tr-0.091
ety-0.091
Token with
Token ID351
Feature activation+1.93e+00

Pos logit contributions
 shark+0.699
 waves+0.694
 victorious+0.685
 screams+0.672
ls+0.665
Neg logit contributions
isk-0.686
 whether-0.674
my-0.658
ety-0.626
coat-0.590
Token a
Token ID257
Feature activation+9.84e-01

Pos logit contributions
 whether+0.104
isk+0.102
my+0.098
ety+0.090
 drew+0.087
Neg logit contributions
 victorious-0.103
 waves-0.103
 shark-0.102
ls-0.102
 cliff-0.099
Token tw
Token ID665
Feature activation-1.66e+00

Pos logit contributions
 whether+0.055
my+0.055
 pad+0.052
ine+0.050
coat+0.050
Neg logit contributions
ls-0.046
 waves-0.045
aro-0.043
 sinks-0.043
 winds-0.043
Tokeninkle
Token ID19894
Feature activation-5.48e+00

Pos logit contributions
 shark+0.073
 victorious+0.068
 screams+0.067
 cheers+0.067
 waves+0.067
Neg logit contributions
ine-0.111
isk-0.108
coat-0.104
 whether-0.103
c-0.103
Token in
Token ID287
Feature activation-3.31e+00

Pos logit contributions
 waves+0.302
 shark+0.296
 victorious+0.296
 screams+0.286
 cliff+0.282
Neg logit contributions
isk-0.308
 whether-0.296
my-0.283
ety-0.281
coat-0.262
Token into
Token ID656
Feature activation-6.45e+00

Pos logit contributions
 waves+0.217
 shark+0.216
 victorious+0.213
 cliff+0.207
 screams+0.201
Neg logit contributions
isk-0.197
 whether-0.195
my-0.192
ety-0.186
coat-0.178
Token the
Token ID262
Feature activation-1.67e+00

Pos logit contributions
 waves+0.334
 victorious+0.324
 shark+0.321
 cliff+0.303
ls+0.301
Neg logit contributions
isk-0.388
 whether-0.372
my-0.351
ety-0.348
coat-0.322
Token water
Token ID1660
Feature activation-3.96e+00

Pos logit contributions
keys+0.086
vere+0.081
aro+0.080
ls+0.080
 victorious+0.079
Neg logit contributions
my-0.093
tr-0.087
 closet-0.081
 makeup-0.080
coat-0.079
Token.
Token ID13
Feature activation+1.98e-01

Pos logit contributions
 victorious+0.283
 waves+0.278
 shark+0.276
 screams+0.269
 losers+0.267
Neg logit contributions
isk-0.148
my-0.147
 whether-0.142
ety-0.138
coat-0.128
Token The
Token ID383
Feature activation+2.23e-01

Pos logit contributions
isk+0.011
 whether+0.010
ety+0.010
my+0.010
coat+0.009
Neg logit contributions
 waves-0.012
 shark-0.011
 victorious-0.011
 screams-0.011
 cliff-0.011
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
my+0.014
 whether+0.013
isk+0.013
ety+0.013
 makeup+0.013
Neg logit contributions
 victorious-0.008
vation-0.008
ls-0.008
 losers-0.008
 screams-0.008
Token took
Token ID1718
Feature activation+2.19e+00

Pos logit contributions
 waves+0.808
 shark+0.806
 cliff+0.802
 victorious+0.802
rier+0.779
Neg logit contributions
isk-0.536
my-0.513
 whether-0.513
ety-0.487
coat-0.479
Token his
Token ID465
Feature activation-4.07e-01

Pos logit contributions
isk+0.109
my+0.103
 whether+0.099
ety+0.095
 her+0.087
Neg logit contributions
 cliff-0.133
 shark-0.132
 waves-0.130
 victorious-0.127
 lighthouse-0.127
Token boat
Token ID8848
Feature activation-8.24e+00

Pos logit contributions
 victorious+0.014
rier+0.014
vation+0.014
 losers+0.013
 cheers+0.013
Neg logit contributions
 whether-0.026
my-0.026
ety-0.025
 makeup-0.025
 pen-0.025
Token away
Token ID1497
Feature activation-2.80e+00

Pos logit contributions
 shark+0.495
 waves+0.485
 screams+0.475
 cheers+0.468
 victorious+0.463
Neg logit contributions
isk-0.429
my-0.415
 whether-0.401
ety-0.387
coat-0.372
Token,
Token ID11
Feature activation+7.80e-01

Pos logit contributions
 waves+0.203
 shark+0.199
 victorious+0.193
 screams+0.192
 cheers+0.191
Neg logit contributions
 whether-0.110
isk-0.106
my-0.099
ety-0.094
coat-0.088
Token runs
Token ID4539
Feature activation-4.75e+00

Pos logit contributions
isk+0.121
 whether+0.118
my+0.115
ety+0.110
 drew+0.104
Neg logit contributions
 victorious-0.132
 shark-0.131
 waves-0.127
rier-0.127
 lighthouse-0.126
Token to
Token ID284
Feature activation+3.05e-01

Pos logit contributions
 waves+0.277
 victorious+0.270
 shark+0.266
 screams+0.254
 shore+0.249
Neg logit contributions
isk-0.274
 whether-0.246
ety-0.245
my-0.235
ine-0.226
Token the
Token ID262
Feature activation-1.64e+00

Pos logit contributions
isk+0.017
 whether+0.016
ety+0.016
my+0.015
coat+0.014
Neg logit contributions
 waves-0.017
 shark-0.017
 victorious-0.017
 screams-0.016
 cliff-0.016
Token fence
Token ID13990
Feature activation-9.01e-01

Pos logit contributions
 victorious+0.078
 waves+0.076
ls+0.075
 shark+0.074
 screams+0.072
Neg logit contributions
isk-0.100
my-0.095
 whether-0.095
ety-0.092
tr-0.090
Token and
Token ID290
Feature activation+3.52e-02

Pos logit contributions
 waves+0.053
 shark+0.051
 victorious+0.051
 boat+0.049
 cliff+0.049
Neg logit contributions
isk-0.050
 whether-0.045
ety-0.045
my-0.041
coat-0.039
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
isk+0.002
 whether+0.002
ety+0.002
my+0.002
coat+0.002
Neg logit contributions
 victorious-0.002
 shark-0.002
 waves-0.002
rier-0.002
 current-0.002
Token at
Token ID379
Feature activation-1.05e+00

Pos logit contributions
 waves+0.712
 shark+0.690
 victorious+0.683
 screams+0.655
 cliff+0.646
Neg logit contributions
isk-0.735
 whether-0.688
ety-0.661
my-0.647
coat-0.601
Token the
Token ID262
Feature activation-1.58e+00

Pos logit contributions
 waves+0.068
 shark+0.067
 victorious+0.066
 screams+0.064
 cliff+0.063
Neg logit contributions
isk-0.051
 whether-0.047
ety-0.045
my-0.044
coat-0.040
Token bull
Token ID6473
Feature activation-5.63e+00

Pos logit contributions
 victorious+0.073
ls+0.072
 shark+0.071
 waves+0.068
 refreshed+0.067
Neg logit contributions
isk-0.096
 whether-0.094
my-0.093
tr-0.089
ety-0.089
Token.
Token ID13
Feature activation+5.83e-01

Pos logit contributions
 waves+0.365
 shark+0.359
 victorious+0.358
 lighthouse+0.341
 current+0.341
Neg logit contributions
isk-0.269
 whether-0.254
my-0.242
ety-0.238
coat-0.211
Token The
Token ID383
Feature activation+7.97e-02

Pos logit contributions
isk+0.028
 whether+0.025
ety+0.024
my+0.024
coat+0.022
Neg logit contributions
 waves-0.039
 shark-0.038
 victorious-0.037
 screams-0.037
 cheers-0.036
Token some
Token ID617
Feature activation+3.44e+00

Pos logit contributions
isk+0.072
 whether+0.067
ety+0.064
my+0.062
coat+0.058
Neg logit contributions
 waves-0.097
 shark-0.095
 victorious-0.093
 screams-0.090
 cliff-0.089
Token people
Token ID661
Feature activation-2.76e+00

Pos logit contributions
 whether+0.213
my+0.207
ety+0.199
isk+0.198
 makeup+0.198
Neg logit contributions
 victorious-0.164
 drown-0.143
 screams-0.141
 lighthouse-0.140
 cliff-0.138
Token surfing
Token ID36254
Feature activation-5.84e-01

Pos logit contributions
 waves+0.170
 shark+0.167
 victorious+0.166
 screams+0.159
 current+0.159
Neg logit contributions
isk-0.141
 whether-0.134
ety-0.127
my-0.127
coat-0.115
Token in
Token ID287
Feature activation-3.37e+00

Pos logit contributions
 victorious+0.037
 shark+0.036
 waves+0.036
 screams+0.035
 current+0.035
Neg logit contributions
isk-0.028
 whether-0.026
my-0.026
ety-0.025
coat-0.024
Token the
Token ID262
Feature activation-1.55e+00

Pos logit contributions
 waves+0.155
 shark+0.150
 victorious+0.147
 cliff+0.139
 screams+0.138
Neg logit contributions
isk-0.232
 whether-0.218
ety-0.214
my-0.209
coat-0.199
Token waves
Token ID9813
Feature activation-1.27e+01

Pos logit contributions
keys+0.082
aro+0.082
 victorious+0.080
rier+0.078
vere+0.078
Neg logit contributions
my-0.090
ety-0.078
 drew-0.078
 makeup-0.075
coat-0.073
Token.
Token ID13
Feature activation+2.01e-02

Pos logit contributions
 waves+0.782
 shark+0.764
 victorious+0.756
 screams+0.726
 cliff+0.721
Neg logit contributions
isk-0.656
 whether-0.616
ety-0.588
my-0.577
coat-0.532
Token He
Token ID679
Feature activation-2.79e+00

Pos logit contributions
isk+0.001
 whether+0.001
ety+0.001
my+0.001
ine+0.001
Neg logit contributions
 waves-0.001
 shark-0.001
 current-0.001
 boat-0.001
 screams-0.001
Token thought
Token ID1807
Feature activation+2.54e+00

Pos logit contributions
rier+0.208
 victorious+0.206
 plunge+0.200
 cliff+0.194
 shark+0.193
Neg logit contributions
isk-0.106
my-0.102
 whether-0.101
ety-0.099
 drew-0.091
Token it
Token ID340
Feature activation+2.21e+00

Pos logit contributions
 whether+0.154
isk+0.154
my+0.141
ety+0.141
coat+0.132
Neg logit contributions
 waves-0.127
 victorious-0.126
 shark-0.122
 screams-0.118
 current-0.117
Token looked
Token ID3114
Feature activation+1.02e+00

Pos logit contributions
isk+0.116
 whether+0.109
ety+0.104
my+0.101
coat+0.094
Neg logit contributions
 waves-0.137
 shark-0.133
 victorious-0.130
 screams-0.126
 cliff-0.125
Token would
Token ID561
Feature activation+3.23e+00

Pos logit contributions
 waves+0.176
 shark+0.173
 victorious+0.166
 boat+0.164
 screams+0.162
Neg logit contributions
isk-0.119
 whether-0.110
ety-0.103
my-0.099
coat-0.093
Token splash
Token ID22870
Feature activation-7.18e+00

Pos logit contributions
isk+0.192
 whether+0.182
ety+0.176
my+0.172
coat+0.161
Neg logit contributions
 waves-0.175
 shark-0.170
 victorious-0.167
 screams-0.160
 cliff-0.159
Token and
Token ID290
Feature activation+1.94e-01

Pos logit contributions
 waves+0.384
 victorious+0.382
 shark+0.376
 screams+0.366
 current+0.359
Neg logit contributions
isk-0.422
 whether-0.404
my-0.387
ety-0.386
coat-0.351
Token make
Token ID787
Feature activation+1.39e+00

Pos logit contributions
isk+0.012
 whether+0.012
my+0.012
ety+0.012
 drew+0.011
Neg logit contributions
 victorious-0.009
 waves-0.009
rier-0.009
ls-0.009
 shark-0.009
Token big
Token ID1263
Feature activation-4.63e+00

Pos logit contributions
isk+0.085
 whether+0.083
my+0.078
ety+0.075
 drew+0.071
Neg logit contributions
 victorious-0.068
 shark-0.067
 cliff-0.065
 waves-0.065
 current-0.065
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
 victorious+0.227
 cliff+0.212
vation+0.212
 plunge+0.209
aro+0.207
Neg logit contributions
my-0.273
 whether-0.269
isk-0.256
 makeup-0.242
ety-0.238
Token with
Token ID351
Feature activation+2.29e+00

Pos logit contributions
 waves+0.850
 victorious+0.849
 shark+0.842
 screams+0.833
 cliff+0.832
Neg logit contributions
isk-0.544
 whether-0.520
my-0.504
ety-0.487
coat-0.449
Token his
Token ID465
Feature activation-2.85e-01

Pos logit contributions
isk+0.114
 whether+0.113
my+0.109
ety+0.105
 makeup+0.095
Neg logit contributions
 victorious-0.137
 waves-0.132
 shark-0.131
 screams-0.131
 cliff-0.130
Token hands
Token ID2832
Feature activation-4.38e+00

Pos logit contributions
 victorious+0.019
rier+0.018
 cliff+0.018
vation+0.018
 plunge+0.018
Neg logit contributions
my-0.012
ety-0.011
 makeup-0.011
isk-0.011
coat-0.011
Token.
Token ID13
Feature activation+7.02e-01

Pos logit contributions
 waves+0.269
 shark+0.263
 victorious+0.260
 screams+0.250
 cliff+0.248
Neg logit contributions
isk-0.229
 whether-0.215
ety-0.205
my-0.201
coat-0.186
Token
Token ID220
Feature activation+2.91e-01

Pos logit contributions
isk+0.038
 whether+0.034
ety+0.034
coat+0.033
ine+0.031
Neg logit contributions
 waves-0.041
 shark-0.039
 victorious-0.038
 boat-0.038
 current-0.037
Token in
Token ID287
Feature activation-3.49e+00

Pos logit contributions
isk+0.014
 whether+0.014
my+0.014
ety+0.014
coat+0.013
Neg logit contributions
 victorious-0.015
 shark-0.015
 waves-0.014
 cliff-0.014
 lighthouse-0.014
Token the
Token ID262
Feature activation-1.63e+00

Pos logit contributions
 waves+0.174
 shark+0.173
 victorious+0.172
 screams+0.163
 current+0.161
Neg logit contributions
isk-0.211
 whether-0.209
my-0.198
ety-0.195
coat-0.180
Token water
Token ID1660
Feature activation-3.99e+00

Pos logit contributions
keys+0.095
vere+0.088
 victorious+0.085
 losers+0.085
aro+0.084
Neg logit contributions
my-0.093
ety-0.079
 makeup-0.078
 her-0.075
c-0.072
Token.
Token ID13
Feature activation-3.51e-02

Pos logit contributions
 shark+0.263
 waves+0.263
 victorious+0.261
 screams+0.252
 cliff+0.248
Neg logit contributions
isk-0.181
 whether-0.171
my-0.164
ety-0.162
coat-0.151
Token The
Token ID383
Feature activation+3.50e-02

Pos logit contributions
 waves+0.002
 shark+0.002
 victorious+0.002
 screams+0.002
 cliff+0.002
Neg logit contributions
isk-0.002
 whether-0.002
ety-0.002
my-0.002
coat-0.001
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
my+0.002
ety+0.002
 whether+0.002
 makeup+0.002
 pad+0.002
Neg logit contributions
 victorious-0.002
keys-0.002
vere-0.002
ls-0.002
 drown-0.002
Token felt
Token ID2936
Feature activation-3.51e+00

Pos logit contributions
 shark+0.759
 waves+0.754
 victorious+0.754
 cliff+0.734
 screams+0.722
Neg logit contributions
isk-0.620
 whether-0.588
my-0.581
ety-0.564
coat-0.557
Token gentle
Token ID10296
Feature activation-2.90e-01

Pos logit contributions
ls+0.207
 shark+0.205
 cliff+0.204
 waves+0.202
 screams+0.199
Neg logit contributions
 whether-0.175
isk-0.162
my-0.156
ety-0.151
coat-0.147
Token as
Token ID355
Feature activation+1.21e+00

Pos logit contributions
 waves+0.019
 shark+0.018
 victorious+0.018
 screams+0.018
 cliff+0.018
Neg logit contributions
isk-0.014
 whether-0.013
ety-0.012
my-0.012
coat-0.011
Token they
Token ID484
Feature activation-5.12e+00

Pos logit contributions
 whether+0.073
ety+0.065
my+0.064
isk+0.064
 hers+0.058
Neg logit contributions
 screams-0.062
 victorious-0.062
 shark-0.062
 waves-0.061
 current-0.059
Token rolled
Token ID11686
Feature activation-1.68e+00

Pos logit contributions
 victorious+0.302
 shark+0.301
 waves+0.300
rier+0.295
 screams+0.291
Neg logit contributions
isk-0.270
 whether-0.251
ety-0.251
my-0.245
coat-0.230
Token for
Token ID329
Feature activation+1.89e+00

Pos logit contributions
 waves+0.348
 shark+0.343
 victorious+0.340
 screams+0.334
 current+0.327
Neg logit contributions
isk-0.269
 whether-0.255
my-0.239
ety-0.238
coat-0.217
Token more
Token ID517
Feature activation+5.32e-01

Pos logit contributions
isk+0.102
 whether+0.096
my+0.088
ety+0.085
ine+0.080
Neg logit contributions
 waves-0.113
 shark-0.110
 victorious-0.107
 screams-0.107
ls-0.107
Token fun
Token ID1257
Feature activation-2.94e-01

Pos logit contributions
 whether+0.031
isk+0.028
my+0.027
ety+0.026
coat+0.025
Neg logit contributions
 victorious-0.028
 shark-0.027
 waves-0.027
 screams-0.027
 cliff-0.026
Token in
Token ID287
Feature activation-3.41e+00

Pos logit contributions
 victorious+0.020
 shark+0.019
 waves+0.019
 screams+0.019
 current+0.019
Neg logit contributions
 whether-0.013
isk-0.012
my-0.012
ety-0.011
coat-0.010
Token the
Token ID262
Feature activation-1.61e+00

Pos logit contributions
 waves+0.182
 shark+0.175
 victorious+0.173
 screams+0.169
 current+0.165
Neg logit contributions
isk-0.206
 whether-0.197
my-0.185
ety-0.184
ine-0.168
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
 screams+0.075
ls+0.072
 victorious+0.070
keys+0.070
 losers+0.069
Neg logit contributions
my-0.100
 whether-0.093
ety-0.091
isk-0.090
 makeup-0.087
Token.
Token ID13
Feature activation+2.61e-01

Pos logit contributions
 waves+0.832
 shark+0.820
 victorious+0.815
 screams+0.808
 cliff+0.787
Neg logit contributions
isk-0.581
 whether-0.550
my-0.538
ety-0.523
coat-0.482
Token<|endoftext|>
Token ID50256
Feature activation-1.62e+00

Pos logit contributions
isk+0.015
 whether+0.014
ety+0.013
my+0.013
coat+0.012
Neg logit contributions
 waves-0.015
 shark-0.015
 victorious-0.014
 screams-0.014
 cliff-0.014
TokenOnce
Token ID7454
Feature activation+2.06e+00

Pos logit contributions
 waves+0.088
 shark+0.085
 victorious+0.084
 screams+0.080
 cliff+0.080
Neg logit contributions
isk-0.096
 whether-0.091
ety-0.088
my-0.086
coat-0.081
Token upon
Token ID2402
Feature activation+1.42e+00

Pos logit contributions
isk+0.073
 whether+0.067
ety+0.064
my+0.064
coat+0.052
Neg logit contributions
 waves-0.161
 shark-0.156
 victorious-0.156
 screams-0.152
 current-0.151
Token a
Token ID257
Feature activation+1.11e+00

Pos logit contributions
isk+0.089
ety+0.079
 whether+0.079
coat+0.074
my+0.073
Neg logit contributions
 waves-0.076
 shark-0.073
 victorious-0.073
 screams-0.069
 refreshed-0.068
Token them
Token ID606
Feature activation-1.73e+00

Pos logit contributions
 waves+0.278
 victorious+0.273
 shark+0.271
 screams+0.259
 lighthouse+0.257
Neg logit contributions
isk-0.327
 whether-0.318
my-0.302
ety-0.297
ine-0.270
Token could
Token ID714
Feature activation+4.40e-01

Pos logit contributions
rier+0.101
 victorious+0.101
 shark+0.101
 current+0.099
 cliff+0.098
Neg logit contributions
 whether-0.084
isk-0.082
my-0.082
 drew-0.076
ety-0.076
Token really
Token ID1107
Feature activation-1.15e+00

Pos logit contributions
 whether+0.023
isk+0.022
my+0.021
ine+0.019
ety+0.019
Neg logit contributions
 shark-0.026
ls-0.026
 waves-0.026
 cliff-0.024
 lighthouse-0.024
Token ride
Token ID6594
Feature activation+8.05e-01

Pos logit contributions
 shark+0.061
 waves+0.059
 victorious+0.058
 lighthouse+0.058
rier+0.058
Neg logit contributions
 whether-0.063
isk-0.061
my-0.059
ety-0.056
 drew-0.056
Token the
Token ID262
Feature activation-1.69e+00

Pos logit contributions
isk+0.046
 whether+0.044
my+0.042
ety+0.042
coat+0.038
Neg logit contributions
 waves-0.044
 shark-0.043
 victorious-0.043
 screams-0.041
 current-0.041
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
 victorious+0.077
ls+0.076
 screams+0.074
 drown+0.074
 lighthouse+0.073
Neg logit contributions
my-0.097
 whether-0.094
isk-0.091
coat-0.091
ety-0.088
Token,
Token ID11
Feature activation+6.40e-01

Pos logit contributions
 waves+0.820
 shark+0.810
 victorious+0.807
 screams+0.798
 lighthouse+0.780
Neg logit contributions
isk-0.561
 whether-0.558
my-0.546
ety-0.522
coat-0.484
Token but
Token ID475
Feature activation+2.54e+00

Pos logit contributions
isk+0.031
 whether+0.031
my+0.028
ety+0.028
coat+0.025
Neg logit contributions
 waves-0.040
 shark-0.040
 victorious-0.039
 screams-0.038
ls-0.038
Token they
Token ID484
Feature activation-4.93e+00

Pos logit contributions
isk+0.150
 whether+0.147
my+0.139
ety+0.138
ine+0.126
Neg logit contributions
 waves-0.134
 victorious-0.129
 shark-0.128
 screams-0.126
 cliff-0.122
Token all
Token ID477
Feature activation-4.23e+00

Pos logit contributions
 waves+0.311
 shark+0.306
 victorious+0.305
rier+0.303
ls+0.293
Neg logit contributions
isk-0.238
 whether-0.225
my-0.219
ety-0.205
 drew-0.196
Token had
Token ID550
Feature activation-6.59e-01

Pos logit contributions
 shark+0.261
 waves+0.256
 victorious+0.255
rier+0.254
ls+0.247
Neg logit contributions
isk-0.200
my-0.199
 whether-0.197
ety-0.184
 drew-0.178
Token D
Token ID360
Feature activation+3.08e+00

Pos logit contributions
isk+0.049
 whether+0.046
ety+0.045
my+0.044
coat+0.042
Neg logit contributions
 waves-0.040
 shark-0.039
 victorious-0.039
 screams-0.037
 cliff-0.037
Tokenucky
Token ID5309
Feature activation-1.50e-01

Pos logit contributions
 whether+0.164
isk+0.139
ety+0.131
 hers+0.124
ellery+0.123
Neg logit contributions
 screams-0.178
 waves-0.177
 cheers-0.165
 wave-0.165
 victorious-0.164
Token.
Token ID13
Feature activation+2.33e-01

Pos logit contributions
 waves+0.010
 shark+0.010
 victorious+0.010
 cliff+0.009
 screams+0.009
Neg logit contributions
isk-0.007
 whether-0.007
my-0.006
ety-0.006
ine-0.006
Token D
Token ID360
Feature activation+3.08e+00

Pos logit contributions
isk+0.014
 whether+0.013
ety+0.013
my+0.012
coat+0.012
Neg logit contributions
 waves-0.013
 shark-0.012
 victorious-0.012
 cheers-0.011
 screams-0.011
Tokenucky
Token ID5309
Feature activation-1.18e-01

Pos logit contributions
 whether+0.164
isk+0.150
ety+0.133
ellery+0.125
 hers+0.125
Neg logit contributions
 waves-0.184
 screams-0.179
 cheers-0.171
 victorious-0.167
 shark-0.165
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
 victorious+0.007
rier+0.007
 lighthouse+0.007
 shark+0.007
 plunge+0.007
Neg logit contributions
isk-0.006
my-0.006
 whether-0.006
ety-0.005
ine-0.005
Token back
Token ID736
Feature activation-6.28e+00

Pos logit contributions
 waves+0.704
 shark+0.696
 victorious+0.690
 cliff+0.674
 screams+0.663
Neg logit contributions
isk-0.693
 whether-0.676
my-0.644
ety-0.638
coat-0.597
Token with
Token ID351
Feature activation+1.98e+00

Pos logit contributions
 waves+0.423
 victorious+0.423
 shark+0.421
 current+0.408
 screams+0.405
Neg logit contributions
isk-0.266
 whether-0.264
my-0.255
ety-0.255
coat-0.224
Token its
Token ID663
Feature activation-1.46e-01

Pos logit contributions
isk+0.099
 whether+0.093
ety+0.089
my+0.086
coat+0.080
Neg logit contributions
 waves-0.127
 shark-0.124
 victorious-0.122
 screams-0.117
 current-0.116
Token wing
Token ID8539
Feature activation-1.67e+00

Pos logit contributions
 victorious+0.008
vation+0.007
rier+0.007
 lighthouse+0.007
 cheers+0.007
Neg logit contributions
 whether-0.008
ety-0.008
my-0.007
coat-0.007
 makeup-0.007
Token.
Token ID13
Feature activation+4.55e-01

Pos logit contributions
 waves+0.098
 shark+0.095
 victorious+0.095
 screams+0.090
 current+0.090
Neg logit contributions
isk-0.091
 whether-0.086
ety-0.082
my-0.080
coat-0.073
Token hand
Token ID1021
Feature activation-2.12e+00

Pos logit contributions
isk+0.032
 whether+0.031
my+0.031
coat+0.030
 makeup+0.029
Neg logit contributions
ls-0.031
 victorious-0.030
 cliff-0.030
 waves-0.030
 shark-0.030
Token as
Token ID355
Feature activation+1.33e+00

Pos logit contributions
 waves+0.124
 shark+0.119
 victorious+0.119
 screams+0.114
 boat+0.114
Neg logit contributions
isk-0.120
 whether-0.110
ety-0.107
my-0.101
coat-0.096
Token they
Token ID484
Feature activation-5.05e+00

Pos logit contributions
 whether+0.079
isk+0.077
my+0.074
ety+0.073
coat+0.067
Neg logit contributions
 waves-0.069
 shark-0.068
 victorious-0.066
 current-0.065
 screams-0.065
Token watched
Token ID7342
Feature activation-6.19e-01

Pos logit contributions
rier+0.342
 victorious+0.325
 shark+0.324
 screams+0.322
 cliff+0.314
Neg logit contributions
isk-0.234
my-0.218
 whether-0.208
ety-0.206
 drew-0.200
Token the
Token ID262
Feature activation-1.47e+00

Pos logit contributions
 waves+0.037
 shark+0.037
 victorious+0.037
 cliff+0.036
 current+0.035
Neg logit contributions
 whether-0.031
isk-0.030
my-0.028
ety-0.026
ine-0.025
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
 victorious+0.059
 drown+0.057
ls+0.054
rier+0.054
aro+0.053
Neg logit contributions
my-0.087
 makeup-0.081
 whether-0.078
 library-0.075
 jewelry-0.074
Token crashing
Token ID21899
Feature activation-4.98e+00

Pos logit contributions
 shark+0.712
 victorious+0.691
 screams+0.682
 waves+0.680
 cliff+0.677
Neg logit contributions
isk-0.651
my-0.647
 whether-0.635
ety-0.602
coat-0.575
Token against
Token ID1028
Feature activation-3.57e-01

Pos logit contributions
 shark+0.274
 victorious+0.272
 screams+0.270
 waves+0.265
 losers+0.264
Neg logit contributions
isk-0.248
my-0.238
 whether-0.237
coat-0.235
ety-0.232
Token the
Token ID262
Feature activation-1.47e+00

Pos logit contributions
 waves+0.018
 shark+0.018
 victorious+0.017
 screams+0.017
 cliff+0.016
Neg logit contributions
isk-0.022
 whether-0.021
ety-0.020
my-0.020
coat-0.019
Token shore
Token ID15191
Feature activation-9.99e+00

Pos logit contributions
 victorious+0.056
 losers+0.055
ls+0.055
keys+0.055
aro+0.053
Neg logit contributions
my-0.095
 makeup-0.085
 pad-0.084
 which-0.082
 whether-0.082
Token.
Token ID13
Feature activation-1.02e-01

Pos logit contributions
 shark+0.688
 victorious+0.677
 waves+0.672
 losers+0.666
 screams+0.664
Neg logit contributions
isk-0.409
 whether-0.395
my-0.390
ety-0.362
ine-0.343
Token Lily
Token ID20037
Feature activation+4.81e+00

Pos logit contributions
isk+0.054
coat+0.047
 whether+0.046
tr+0.046
ety+0.046
Neg logit contributions
 waves-0.037
 boat-0.036
 shark-0.036
 we-0.035
 victorious-0.034
Token wanted
Token ID2227
Feature activation+2.18e+00

Pos logit contributions
isk+0.243
 whether+0.227
ety+0.219
my+0.216
coat+0.197
Neg logit contributions
 waves-0.303
 shark-0.298
 victorious-0.296
 cliff-0.282
 screams-0.281
Token to
Token ID284
Feature activation+8.60e-01

Pos logit contributions
isk+0.128
 whether+0.105
ety+0.103
tr+0.099
ine+0.099
Neg logit contributions
 waves-0.137
 shark-0.129
 shore-0.123
 cliff-0.123
 boat-0.121
Token feel
Token ID1254
Feature activation-1.54e+00

Pos logit contributions
 whether+0.037
isk+0.035
my+0.035
ety+0.033
coat+0.031
Neg logit contributions
rier-0.063
ls-0.063
 victorious-0.059
 shark-0.057
 winds-0.055
Token the
Token ID262
Feature activation-1.32e+00

Pos logit contributions
 shark+0.077
 victorious+0.077
 waves+0.076
 lighthouse+0.074
 screams+0.072
Neg logit contributions
 whether-0.092
isk-0.088
my-0.085
ety-0.085
coat-0.077
Token waves
Token ID9813
Feature activation-1.26e+01

Pos logit contributions
 victorious+0.075
keys+0.073
ls+0.066
eros+0.063
vation+0.063
Neg logit contributions
my-0.073
 color-0.065
 whether-0.064
ety-0.063
coat-0.062
Token on
Token ID319
Feature activation-8.54e-01

Pos logit contributions
 waves+0.778
 shark+0.762
 victorious+0.753
 screams+0.727
 lighthouse+0.719
Neg logit contributions
isk-0.645
 whether-0.617
ety-0.593
my-0.591
coat-0.543
Token her
Token ID607
Feature activation+4.96e+00

Pos logit contributions
 waves+0.072
 shark+0.069
 cliff+0.068
 screams+0.068
 victorious+0.068
Neg logit contributions
isk-0.028
 whether-0.024
ety-0.023
my-0.021
coat-0.020
Token feet
Token ID3625
Feature activation-4.63e+00

Pos logit contributions
 whether+0.215
ine+0.209
my+0.209
ety+0.200
 her+0.197
Neg logit contributions
 victorious-0.315
 losers-0.306
 disappears-0.288
athed-0.282
 hawk-0.278
Token,
Token ID11
Feature activation+7.61e-01

Pos logit contributions
 waves+0.242
 victorious+0.230
 shark+0.229
 boat+0.227
 cliff+0.220
Neg logit contributions
isk-0.287
 whether-0.254
ety-0.248
my-0.233
 drew-0.229
Token so
Token ID523
Feature activation-3.95e+00

Pos logit contributions
isk+0.043
 whether+0.038
ety+0.038
my+0.036
tr+0.035
Neg logit contributions
 waves-0.045
 shark-0.043
 victorious-0.042
 screams-0.042
 current-0.041
Token me
Token ID502
Feature activation-4.83e+00

Pos logit contributions
isk+0.180
 whether+0.172
ety+0.163
my+0.159
coat+0.157
Neg logit contributions
 waves-0.104
 shark-0.098
 victorious-0.096
 screams-0.092
ls-0.090
Token a
Token ID257
Feature activation+1.37e+00

Pos logit contributions
 waves+0.288
 shark+0.277
 victorious+0.274
 boat+0.260
 cliff+0.257
Neg logit contributions
isk-0.264
 whether-0.246
ety-0.236
my-0.225
coat-0.216
Token hug
Token ID16225
Feature activation-3.53e+00

Pos logit contributions
isk+0.086
 whether+0.082
ety+0.079
my+0.078
coat+0.073
Neg logit contributions
 waves-0.070
 shark-0.067
 victorious-0.067
 screams-0.063
 cliff-0.063
Token?"
Token ID1701
Feature activation-4.83e+00

Pos logit contributions
 waves+0.114
 shark+0.107
 boat+0.097
 wave+0.096
 victorious+0.096
Neg logit contributions
isk-0.288
 whether-0.266
ety-0.256
coat-0.238
ine-0.238
Token Mom
Token ID11254
Feature activation-1.63e+00

Pos logit contributions
 waves+0.336
 shark+0.329
 victorious+0.326
 screams+0.318
 cheers+0.312
Neg logit contributions
isk-0.219
 whether-0.197
ety-0.184
my-0.176
coat-0.168
Tokenmy
Token ID1820
Feature activation+1.11e+01

Pos logit contributions
 waves+0.126
 cheers+0.122
 screams+0.122
 wave+0.119
 shark+0.117
Neg logit contributions
isk-0.052
 whether-0.045
artments-0.035
tr-0.035
ellery-0.034
Token said
Token ID531
Feature activation-2.46e+00

Pos logit contributions
isk+0.842
 whether+0.800
ety+0.766
coat+0.733
tr+0.733
Neg logit contributions
 waves-0.425
 shark-0.405
 victorious-0.403
 screams-0.392
 cheers-0.384
Token,
Token ID11
Feature activation+1.15e+00

Pos logit contributions
 victorious+0.130
 waves+0.130
 shark+0.127
 screams+0.123
rier+0.122
Neg logit contributions
isk-0.150
 whether-0.137
ety-0.137
my-0.130
tr-0.124
Token "
Token ID366
Feature activation+5.13e+00

Pos logit contributions
isk+0.057
ety+0.053
 whether+0.051
my+0.049
tr+0.046
Neg logit contributions
 waves-0.074
 victorious-0.073
 shark-0.072
 screams-0.069
 refreshed-0.068
TokenOf
Token ID5189
Feature activation+2.18e+00

Pos logit contributions
isk+0.287
 whether+0.263
ety+0.249
my+0.245
ine+0.231
Neg logit contributions
 waves-0.300
 shark-0.291
 victorious-0.291
 screams-0.277
 cheers-0.274
Token course
Token ID1781
Feature activation-2.12e+00

Pos logit contributions
isk+0.139
 whether+0.132
ety+0.128
my+0.127
coat+0.117
Neg logit contributions
 waves-0.110
 shark-0.108
 victorious-0.106
 screams-0.100
 cliff-0.100
Token."
Token ID526
Feature activation-4.00e+00

Pos logit contributions
 whether+0.183
isk+0.182
my+0.175
ety+0.166
coat+0.155
Neg logit contributions
 waves-0.217
 victorious-0.215
 shark-0.213
 lighthouse-0.207
 cliff-0.205
Token
Token ID220
Feature activation+2.44e-01

Pos logit contributions
 waves+0.201
 victorious+0.186
 shark+0.185
 screams+0.178
 cheers+0.177
Neg logit contributions
isk-0.253
 whether-0.241
ety-0.226
coat-0.220
tr-0.218
Token
Token ID198
Feature activation-8.42e-01

Pos logit contributions
isk+0.013
 whether+0.012
my+0.011
ety+0.011
coat+0.010
Neg logit contributions
 waves-0.015
 shark-0.015
 victorious-0.015
 cliff-0.014
 current-0.014
Token
Token ID198
Feature activation-8.35e-01

Pos logit contributions
 waves+0.048
 shark+0.047
 victorious+0.047
 screams+0.045
 cliff+0.044
Neg logit contributions
isk-0.047
 whether-0.045
ety-0.043
my-0.042
coat-0.039
TokenSam
Token ID16305
Feature activation+4.53e+00

Pos logit contributions
 waves+0.054
 victorious+0.052
 shark+0.050
 current+0.050
 boat+0.049
Neg logit contributions
isk-0.040
 whether-0.036
ety-0.035
coat-0.034
tr-0.033
Tokenmy
Token ID1820
Feature activation+1.03e+01

Pos logit contributions
 whether+0.071
isk+0.066
ety+0.038
coat+0.033
tr+0.032
Neg logit contributions
 waves-0.439
 shark-0.423
 victorious-0.417
 boat-0.413
 screams-0.410
Token said
Token ID531
Feature activation-2.90e+00

Pos logit contributions
isk+0.538
my+0.506
ety+0.489
 whether+0.471
 drew+0.454
Neg logit contributions
rier-0.620
 victorious-0.619
 plunge-0.611
 waves-0.607
 shark-0.605
Token,
Token ID11
Feature activation+6.46e-01

Pos logit contributions
 waves+0.180
 shark+0.172
 victorious+0.171
 cliff+0.167
 screams+0.166
Neg logit contributions
isk-0.147
 whether-0.144
ety-0.133
my-0.132
coat-0.122
Token "
Token ID366
Feature activation+4.80e+00

Pos logit contributions
isk+0.032
ety+0.029
 whether+0.029
my+0.028
coat+0.026
Neg logit contributions
 waves-0.041
 shark-0.041
 victorious-0.041
 boat-0.039
 current-0.038
TokenI
Token ID40
Feature activation-3.32e+00

Pos logit contributions
isk+0.259
 whether+0.232
my+0.213
ety+0.213
coat+0.208
Neg logit contributions
 waves-0.295
 shark-0.284
 victorious-0.280
 current-0.276
 cliff-0.272
Token can
Token ID460
Feature activation+2.56e+00

Pos logit contributions
 waves+0.165
 shark+0.164
 victorious+0.161
 screams+0.153
 boat+0.150
Neg logit contributions
isk-0.212
 whether-0.198
ety-0.192
my-0.186
coat-0.179
Token L
Token ID406
Feature activation+8.78e+00

Pos logit contributions
isk+0.064
 whether+0.059
ety+0.057
my+0.055
coat+0.052
Neg logit contributions
 waves-0.074
 shark-0.073
 victorious-0.071
 screams-0.068
 cliff-0.068
Tokenila
Token ID10102
Feature activation+1.93e+00

Pos logit contributions
isk+0.534
 whether+0.511
ety+0.482
my+0.477
coat+0.446
Neg logit contributions
 waves-0.463
 shark-0.450
 victorious-0.442
 cliff-0.426
 screams-0.424
Token."
Token ID526
Feature activation-3.27e+00

Pos logit contributions
isk+0.123
 whether+0.116
ety+0.111
my+0.108
coat+0.104
Neg logit contributions
 waves-0.097
 shark-0.094
 victorious-0.091
 screams-0.088
 boat-0.087
Token
Token ID198
Feature activation-2.91e-01

Pos logit contributions
 waves+0.134
 screams+0.125
 cheers+0.124
 shark+0.116
ls+0.115
Neg logit contributions
isk-0.255
 whether-0.223
ety-0.215
tr-0.212
coat-0.209
Token
Token ID198
Feature activation-2.89e-01

Pos logit contributions
 waves+0.015
 shark+0.015
 victorious+0.015
 screams+0.014
 cheers+0.014
Neg logit contributions
isk-0.018
 whether-0.017
ety-0.016
my-0.016
coat-0.015
TokenL
Token ID43
Feature activation+1.01e+01

Pos logit contributions
 waves+0.017
 screams+0.016
 victorious+0.016
 cheers+0.016
 cliff+0.016
Neg logit contributions
isk-0.016
 whether-0.015
ety-0.014
tr-0.013
my-0.013
Tokenila
Token ID10102
Feature activation+1.92e+00

Pos logit contributions
 whether+0.615
isk+0.608
ety+0.568
my+0.555
tr+0.543
Neg logit contributions
 waves-0.518
 screams-0.497
ls-0.478
 victorious-0.478
 cheers-0.459
Token says
Token ID1139
Feature activation-4.34e+00

Pos logit contributions
isk+0.129
my+0.128
 whether+0.122
ety+0.119
 drew+0.116
Neg logit contributions
 victorious-0.084
 shark-0.081
rier-0.081
 lighthouse-0.081
 current-0.079
Token,
Token ID11
Feature activation+1.19e+00

Pos logit contributions
 waves+0.263
 victorious+0.256
 shark+0.256
 cliff+0.245
 current+0.244
Neg logit contributions
isk-0.220
 whether-0.216
my-0.203
ety-0.202
coat-0.187
Token "
Token ID366
Feature activation+5.26e+00

Pos logit contributions
isk+0.067
ety+0.062
 whether+0.060
my+0.058
ine+0.054
Neg logit contributions
 waves-0.070
 shark-0.069
 victorious-0.068
 cheers-0.064
 screams-0.064
TokenThank
Token ID10449
Feature activation+1.26e+00

Pos logit contributions
isk+0.359
 whether+0.329
my+0.307
ety+0.301
ine+0.290
Neg logit contributions
 waves-0.265
 shark-0.244
 cheers-0.242
 screams-0.239
 victorious-0.239
Token Lisa
Token ID15378
Feature activation+2.25e+00

Pos logit contributions
isk+0.046
ety+0.043
coat+0.041
 whether+0.040
tr+0.039
Neg logit contributions
 waves-0.069
 victorious-0.067
 cheers-0.066
 screams-0.066
 shark-0.065
Token.
Token ID13
Feature activation+5.32e-01

Pos logit contributions
isk+0.138
ety+0.129
 whether+0.128
coat+0.124
tr+0.117
Neg logit contributions
 waves-0.112
 victorious-0.110
 shark-0.109
 screams-0.106
 cheers-0.104
Token It
Token ID632
Feature activation+3.85e+00

Pos logit contributions
isk+0.030
ety+0.028
 whether+0.027
coat+0.026
my+0.025
Neg logit contributions
 waves-0.031
 victorious-0.031
 cheers-0.030
 screams-0.030
 shark-0.029
Token asked
Token ID1965
Feature activation-1.42e+00

Pos logit contributions
isk+0.203
 whether+0.192
my+0.184
ety+0.182
ine+0.164
Neg logit contributions
 waves-0.231
 shark-0.226
 victorious-0.225
 cliff-0.218
 current-0.215
Token her
Token ID607
Feature activation+5.19e+00

Pos logit contributions
 waves+0.081
 victorious+0.081
 cheers+0.080
 screams+0.078
 shark+0.076
Neg logit contributions
isk-0.081
ety-0.078
coat-0.076
tr-0.072
 whether-0.066
Token if
Token ID611
Feature activation+9.47e+00

Pos logit contributions
isk+0.252
ety+0.223
 whether+0.222
my+0.210
coat+0.207
Neg logit contributions
 waves-0.345
 shark-0.339
 victorious-0.333
 boat-0.320
 cliff-0.319
Token she
Token ID673
Feature activation+2.36e+00

Pos logit contributions
isk+0.433
 whether+0.405
ety+0.383
my+0.376
coat+0.341
Neg logit contributions
 waves-0.640
 shark-0.627
 victorious-0.621
 screams-0.599
 cliff-0.595
Token wanted
Token ID2227
Feature activation+2.57e+00

Pos logit contributions
isk+0.134
 whether+0.123
ety+0.117
my+0.112
coat+0.112
Neg logit contributions
 waves-0.136
 shark-0.132
 victorious-0.129
 screams-0.126
 wave-0.122
Token to
Token ID284
Feature activation+1.12e+00

Pos logit contributions
isk+0.122
 whether+0.108
ety+0.106
my+0.101
coat+0.097
Neg logit contributions
 waves-0.176
 shark-0.172
 victorious-0.169
 screams-0.164
 cliff-0.162
Token play
Token ID711
Feature activation+2.61e+00

Pos logit contributions
isk+0.059
 whether+0.058
my+0.056
ety+0.054
 drew+0.050
Neg logit contributions
 shark-0.066
ls-0.065
 victorious-0.064
 waves-0.064
rier-0.063
Token.
Token ID13
Feature activation+6.69e-01

Pos logit contributions
isk+0.159
ety+0.148
 whether+0.146
my+0.140
tr+0.133
Neg logit contributions
 waves-0.141
 shark-0.139
 victorious-0.136
 boat-0.129
 cliff-0.126
Token whispered
Token ID25029
Feature activation+1.39e+00

Pos logit contributions
 waves+0.008
 shark+0.008
 victorious+0.008
 screams+0.007
 cliff+0.007
Neg logit contributions
isk-0.008
 whether-0.007
ety-0.007
my-0.007
coat-0.006
Token to
Token ID284
Feature activation+9.50e-01

Pos logit contributions
isk+0.086
ety+0.080
 whether+0.079
my+0.077
coat+0.073
Neg logit contributions
 waves-0.073
 shark-0.072
 victorious-0.070
 boat-0.066
 screams-0.066
Token him
Token ID683
Feature activation+1.66e+00

Pos logit contributions
isk+0.050
 whether+0.048
ety+0.048
my+0.046
coat+0.045
Neg logit contributions
 waves-0.057
 victorious-0.054
 shark-0.054
 screams-0.051
 cheers-0.051
Token about
Token ID546
Feature activation+2.19e+00

Pos logit contributions
ety+0.099
isk+0.097
coat+0.095
 whether+0.090
my+0.088
Neg logit contributions
 victorious-0.086
 waves-0.084
 shark-0.083
 boat-0.080
rier-0.078
Token the
Token ID262
Feature activation-1.22e+00

Pos logit contributions
ety+0.147
isk+0.147
tr+0.144
coat+0.144
my+0.137
Neg logit contributions
 shark-0.096
 waves-0.096
 shore-0.089
 victorious-0.087
 boat-0.087
Token needle
Token ID17598
Feature activation+9.59e+00

Pos logit contributions
 waves+0.076
 screams+0.070
 boat+0.070
 cliff+0.070
 shark+0.070
Neg logit contributions
isk-0.069
 whether-0.063
ety-0.059
ine-0.055
my-0.055
Token.
Token ID13
Feature activation+8.84e-02

Pos logit contributions
isk+0.516
 whether+0.481
ety+0.465
my+0.449
coat+0.419
Neg logit contributions
 waves-0.576
 shark-0.562
 victorious-0.555
 screams-0.533
 cliff-0.529
Token The
Token ID383
Feature activation+3.39e-01

Pos logit contributions
isk+0.005
 whether+0.004
ety+0.004
my+0.004
coat+0.004
Neg logit contributions
 waves-0.005
 victorious-0.005
 shark-0.005
 cheers-0.005
 screams-0.005
Token park
Token ID3952
Feature activation-1.60e+00

Pos logit contributions
isk+0.022
 whether+0.022
ety+0.021
my+0.020
ine+0.020
Neg logit contributions
 waves-0.017
 victorious-0.016
 shark-0.016
 current-0.015
 screams-0.015
Token ranger
Token ID43570
Feature activation+1.01e-01

Pos logit contributions
 waves+0.078
 shark+0.077
 victorious+0.075
 screams+0.073
 cliff+0.072
Neg logit contributions
isk-0.103
 whether-0.096
my-0.093
ety-0.093
coat-0.087
Token thanked
Token ID26280
Feature activation-9.26e+00

Pos logit contributions
isk+0.006
 whether+0.005
ety+0.005
my+0.005
coat+0.005
Neg logit contributions
 waves-0.006
 shark-0.006
 victorious-0.006
 screams-0.005
 cliff-0.005
Token reminder
Token ID15438
Feature activation+1.03e+00

Pos logit contributions
isk+0.098
 whether+0.093
ety+0.090
my+0.089
coat+0.083
Neg logit contributions
 waves-0.070
 shark-0.067
 victorious-0.066
 screams-0.063
 cliff-0.062
Token.
Token ID13
Feature activation+5.10e-01

Pos logit contributions
isk+0.056
 whether+0.053
ety+0.051
my+0.050
coat+0.046
Neg logit contributions
 waves-0.061
 shark-0.059
 victorious-0.059
 screams-0.056
 cliff-0.056
Token She
Token ID1375
Feature activation+2.84e+00

Pos logit contributions
isk+0.028
 whether+0.026
ety+0.025
my+0.025
coat+0.024
Neg logit contributions
 waves-0.030
 victorious-0.029
 shark-0.028
 cheers-0.028
 screams-0.027
Token walked
Token ID6807
Feature activation-6.43e-01

Pos logit contributions
isk+0.158
 whether+0.147
ety+0.141
my+0.141
coat+0.128
Neg logit contributions
 waves-0.164
 shark-0.163
 victorious-0.160
 cliff-0.156
 current-0.154
Token away
Token ID1497
Feature activation-2.71e+00

Pos logit contributions
 waves+0.028
 victorious+0.027
 shark+0.026
ls+0.024
 cheers+0.024
Neg logit contributions
isk-0.047
 whether-0.044
ety-0.043
my-0.041
tr-0.040
Token.
Token ID13
Feature activation+5.20e-01

Pos logit contributions
 waves+0.175
 shark+0.171
 victorious+0.163
 plunge+0.162
 cliff+0.161
Neg logit contributions
isk-0.133
 whether-0.127
ety-0.118
my-0.116
coat-0.106
Token
Token ID198
Feature activation-4.65e-01

Pos logit contributions
isk+0.032
 whether+0.029
ety+0.029
my+0.029
coat+0.028
Neg logit contributions
 victorious-0.027
 waves-0.026
 cheers-0.025
 shark-0.025
 refreshed-0.025
Token
Token ID198
Feature activation-4.55e-01

Pos logit contributions
 waves+0.026
 shark+0.025
 victorious+0.025
 screams+0.024
 cliff+0.024
Neg logit contributions
isk-0.027
 whether-0.025
ety-0.024
my-0.024
coat-0.022
TokenBen
Token ID11696
Feature activation+3.10e+00

Pos logit contributions
 victorious+0.029
 waves+0.028
 shark+0.027
ls+0.027
 cliff+0.027
Neg logit contributions
isk-0.022
 whether-0.020
ety-0.019
coat-0.019
my-0.018
Token and
Token ID290
Feature activation+2.16e-01

Pos logit contributions
isk+0.131
my+0.118
 whether+0.113
ety+0.112
 drew+0.101
Neg logit contributions
 shark-0.219
 plunge-0.215
 waves-0.214
rier-0.214
 victorious-0.213
Token Lily
Token ID20037
Feature activation+5.05e+00

Pos logit contributions
isk+0.008
 whether+0.008
ety+0.007
my+0.007
coat+0.007
Neg logit contributions
 waves-0.016
 shark-0.016
 victorious-0.015
 screams-0.015
 current-0.015
Token them
Token ID606
Feature activation-1.39e+00

Pos logit contributions
 waves+0.269
 victorious+0.263
 shark+0.263
 cliff+0.255
 screams+0.251
Neg logit contributions
isk-0.254
 whether-0.240
ety-0.234
my-0.232
 drew-0.208
Token.
Token ID13
Feature activation+4.42e-01

Pos logit contributions
 waves+0.092
 shark+0.092
 victorious+0.090
 current+0.089
 cliff+0.088
Neg logit contributions
isk-0.063
 whether-0.059
my-0.057
ety-0.055
coat-0.050
Token They
Token ID1119
Feature activation-2.27e+00

Pos logit contributions
isk+0.028
 whether+0.026
ety+0.026
my+0.025
coat+0.024
Neg logit contributions
 waves-0.024
 shark-0.022
 victorious-0.022
 cheers-0.021
 screams-0.021
Token helped
Token ID4193
Feature activation+1.25e+00

Pos logit contributions
 waves+0.135
 shark+0.135
 victorious+0.134
rier+0.130
 cliff+0.129
Neg logit contributions
isk-0.121
 whether-0.112
my-0.109
ety-0.108
coat-0.099
Token them
Token ID606
Feature activation-1.39e+00

Pos logit contributions
isk+0.082
 whether+0.078
ety+0.074
my+0.072
tr+0.072
Neg logit contributions
 waves-0.061
 shark-0.057
 victorious-0.055
 boat-0.054
 shore-0.054
Token make
Token ID787
Feature activation+1.49e+00

Pos logit contributions
 waves+0.081
 shark+0.079
 victorious+0.078
 screams+0.075
 cliff+0.075
Neg logit contributions
isk-0.077
 whether-0.073
ety-0.070
my-0.069
coat-0.064
Token a
Token ID257
Feature activation+1.20e+00

Pos logit contributions
isk+0.072
 whether+0.070
my+0.067
ety+0.063
ine+0.058
Neg logit contributions
 victorious-0.092
 shark-0.091
 cliff-0.089
 lighthouse-0.089
 current-0.088
Token bigger
Token ID5749
Feature activation-1.47e+00

Pos logit contributions
isk+0.059
my+0.058
tr+0.055
 whether+0.055
ety+0.051
Neg logit contributions
aro-0.070
 winds-0.069
 plunge-0.067
ls-0.067
ued-0.066
Token snow
Token ID6729
Feature activation+4.13e+00

Pos logit contributions
 victorious+0.088
 shark+0.085
 waves+0.083
 plunge+0.083
ls+0.083
Neg logit contributions
isk-0.075
 whether-0.072
my-0.071
ety-0.066
tr-0.066
Tokenman
Token ID805
Feature activation+3.37e-01

Pos logit contributions
isk+0.232
tr+0.210
my+0.206
 whether+0.201
ety+0.195
Neg logit contributions
 victorious-0.237
 waves-0.230
 shark-0.227
 plunge-0.223
 cliff-0.223
Token.
Token ID13
Feature activation+4.19e-01

Pos logit contributions
isk+0.016
 whether+0.015
my+0.014
ety+0.014
tr+0.013
Neg logit contributions
 waves-0.022
 shark-0.022
 victorious-0.021
 cliff-0.021
 current-0.021
Token to
Token ID284
Feature activation+9.28e-01

Pos logit contributions
isk+0.554
ety+0.486
my+0.451
coat+0.450
 whether+0.443
Neg logit contributions
 shark-0.482
 waves-0.456
 boat-0.451
 screams-0.436
 cliff-0.431
Token do
Token ID466
Feature activation+1.31e+00

Pos logit contributions
isk+0.066
 whether+0.063
ety+0.061
my+0.058
coat+0.056
Neg logit contributions
 waves-0.040
 shark-0.039
 screams-0.037
 plunge-0.036
 boat-0.036
Token.
Token ID13
Feature activation+3.27e-01

Pos logit contributions
isk+0.085
ety+0.079
 whether+0.075
my+0.073
coat+0.072
Neg logit contributions
 waves-0.064
 shark-0.064
 victorious-0.062
 screams-0.061
ls-0.060
Token He
Token ID679
Feature activation-2.38e+00

Pos logit contributions
isk+0.021
 whether+0.020
ety+0.019
my+0.019
coat+0.018
Neg logit contributions
 waves-0.017
 victorious-0.016
 shark-0.016
 screams-0.016
 current-0.016
Token tried
Token ID3088
Feature activation+3.63e-01

Pos logit contributions
 waves+0.132
 shark+0.126
 screams+0.123
 victorious+0.123
 current+0.119
Neg logit contributions
isk-0.141
 whether-0.134
ety-0.126
my-0.122
coat-0.120
Token to
Token ID284
Feature activation+9.26e-01

Pos logit contributions
isk+0.024
ety+0.021
coat+0.021
tr+0.021
 drew+0.020
Neg logit contributions
 victorious-0.018
 shark-0.018
 waves-0.017
 shore-0.017
 screams-0.017
Token fly
Token ID6129
Feature activation-1.68e+00

Pos logit contributions
isk+0.055
 whether+0.052
ety+0.050
my+0.049
coat+0.046
Neg logit contributions
 waves-0.051
 shark-0.049
 victorious-0.049
 screams-0.047
 cliff-0.046
Token away
Token ID1497
Feature activation-2.73e+00

Pos logit contributions
 waves+0.074
 victorious+0.072
 shark+0.072
 screams+0.066
 cliff+0.066
Neg logit contributions
isk-0.118
 whether-0.111
ety-0.109
my-0.106
coat-0.102
Token,
Token ID11
Feature activation+8.51e-01

Pos logit contributions
 waves+0.165
 shark+0.160
 victorious+0.157
 screams+0.152
 cliff+0.150
Neg logit contributions
isk-0.145
 whether-0.140
ety-0.130
my-0.130
coat-0.119
Token but
Token ID475
Feature activation+2.76e+00

Pos logit contributions
isk+0.040
ety+0.035
 whether+0.034
my+0.032
coat+0.031
Neg logit contributions
 waves-0.058
 victorious-0.057
 shark-0.056
 screams-0.055
 current-0.055
Token the
Token ID262
Feature activation-1.55e+00

Pos logit contributions
isk+0.168
 whether+0.165
my+0.156
ety+0.155
tr+0.139
Neg logit contributions
 waves-0.145
 shark-0.138
 victorious-0.136
 screams-0.132
 cheers-0.132
Token something
Token ID1223
Feature activation+5.66e+00

Pos logit contributions
 waves+0.061
 shark+0.060
 victorious+0.057
 shore+0.055
 boat+0.055
Neg logit contributions
isk-0.070
ety-0.065
coat-0.063
my-0.061
tr-0.061
Token.
Token ID13
Feature activation+5.08e-01

Pos logit contributions
isk+0.376
ety+0.341
 whether+0.332
my+0.322
coat+0.316
Neg logit contributions
 waves-0.286
 shark-0.281
 victorious-0.272
 boat-0.260
 screams-0.260
Token He
Token ID679
Feature activation-2.22e+00

Pos logit contributions
isk+0.033
 whether+0.031
ety+0.030
my+0.029
coat+0.029
Neg logit contributions
 waves-0.025
 shark-0.025
 victorious-0.025
 screams-0.023
 current-0.023
Token remembered
Token ID12086
Feature activation-1.10e+00

Pos logit contributions
 waves+0.137
 shark+0.136
 victorious+0.134
 cliff+0.131
 plunge+0.129
Neg logit contributions
isk-0.112
 whether-0.106
my-0.102
ety-0.101
ine-0.092
Token a
Token ID257
Feature activation+1.48e+00

Pos logit contributions
 waves+0.061
 shark+0.059
 victorious+0.057
 boat+0.055
 shore+0.054
Neg logit contributions
isk-0.067
ety-0.061
 whether-0.059
coat-0.058
my-0.058
Token funny
Token ID8258
Feature activation+1.06e+00

Pos logit contributions
isk+0.104
 whether+0.097
ety+0.095
my+0.093
coat+0.088
Neg logit contributions
 waves-0.067
 shark-0.066
 victorious-0.064
 cliff-0.061
 screams-0.061
Token looking
Token ID2045
Feature activation+3.94e+00

Pos logit contributions
isk+0.079
ety+0.075
 whether+0.074
my+0.070
coat+0.065
Neg logit contributions
 waves-0.045
 shark-0.045
 boat-0.041
 screams-0.041
 wave-0.041
Token tree
Token ID5509
Feature activation-9.71e-01

Pos logit contributions
isk+0.244
 whether+0.231
ety+0.222
my+0.220
coat+0.206
Neg logit contributions
 waves-0.203
 shark-0.197
 victorious-0.195
 screams-0.186
 cliff-0.184
Token at
Token ID379
Feature activation-4.67e-01

Pos logit contributions
 waves+0.054
 shark+0.053
 victorious+0.053
 boat+0.050
 cliff+0.050
Neg logit contributions
isk-0.057
 whether-0.053
ety-0.051
my-0.049
coat-0.047
Token the
Token ID262
Feature activation-1.10e+00

Pos logit contributions
 waves+0.021
 shark+0.021
 victorious+0.020
 cliff+0.019
 screams+0.019
Neg logit contributions
isk-0.032
 whether-0.030
ety-0.030
my-0.028
coat-0.028
Token top
Token ID1353
Feature activation-4.72e+00

Pos logit contributions
 waves+0.043
 shark+0.042
 victorious+0.041
 screams+0.038
 cliff+0.038
Neg logit contributions
isk-0.082
 whether-0.079
ety-0.076
my-0.075
coat-0.071
Token to
Token ID284
Feature activation+9.34e-01

Pos logit contributions
 waves+0.379
 shark+0.367
 victorious+0.363
 screams+0.354
 current+0.347
Neg logit contributions
isk-0.269
 whether-0.242
ety-0.240
my-0.228
coat-0.212
Token jump
Token ID4391
Feature activation-1.55e+00

Pos logit contributions
isk+0.058
 whether+0.054
my+0.052
ety+0.051
coat+0.048
Neg logit contributions
 waves-0.047
 victorious-0.047
 shark-0.047
ls-0.045
 screams-0.044
Token and
Token ID290
Feature activation-1.32e-01

Pos logit contributions
 waves+0.088
 shark+0.087
 victorious+0.086
 screams+0.083
 cliff+0.082
Neg logit contributions
isk-0.084
 whether-0.081
my-0.078
ety-0.077
coat-0.072
Token sing
Token ID1702
Feature activation+1.40e+00

Pos logit contributions
 victorious+0.007
rier+0.006
 waves+0.006
 shark+0.006
 lighthouse+0.006
Neg logit contributions
 whether-0.008
isk-0.008
my-0.007
ety-0.007
 drew-0.007
Token!
Token ID0
Feature activation-2.59e+00

Pos logit contributions
isk+0.082
 whether+0.076
ety+0.076
my+0.072
coat+0.068
Neg logit contributions
 waves-0.079
 victorious-0.077
 shark-0.076
 screams-0.072
 cheers-0.072
Token M
Token ID337
Feature activation+2.00e+00

Pos logit contributions
 waves+0.153
 shark+0.150
 victorious+0.147
 cliff+0.142
 screams+0.142
Neg logit contributions
isk-0.140
 whether-0.132
ety-0.127
my-0.125
coat-0.115
Tokenandy
Token ID10757
Feature activation-1.06e+00

Pos logit contributions
isk+0.130
ety+0.118
 whether+0.116
my+0.115
ine+0.107
Neg logit contributions
 victorious-0.104
 waves-0.102
 shark-0.098
 lighthouse-0.097
 cliff-0.095
Token and
Token ID290
Feature activation-1.05e-01

Pos logit contributions
rier+0.072
 victorious+0.070
 shark+0.070
 lighthouse+0.069
 current+0.067
Neg logit contributions
my-0.044
isk-0.044
 drew-0.042
ine-0.042
 whether-0.042
Token Martin
Token ID5780
Feature activation-8.34e+00

Pos logit contributions
 waves+0.007
 shark+0.007
 victorious+0.007
 screams+0.006
 cliff+0.006
Neg logit contributions
isk-0.005
 whether-0.005
ety-0.005
my-0.004
coat-0.004
Token were
Token ID547
Feature activation-1.41e+00

Pos logit contributions
 shark+0.484
 victorious+0.479
 waves+0.479
rier+0.464
 cliff+0.457
Neg logit contributions
isk-0.446
 whether-0.422
my-0.417
ety-0.409
 drew-0.390
Token so
Token ID523
Feature activation-3.76e+00

Pos logit contributions
 shark+0.086
ls+0.083
 current+0.080
 waves+0.078
 lighthouse+0.078
Neg logit contributions
my-0.072
 whether-0.071
isk-0.068
ine-0.063
 her-0.063
Token time
Token ID640
Feature activation-1.94e+00

Pos logit contributions
isk+0.084
 whether+0.080
ety+0.077
my+0.074
coat+0.071
Neg logit contributions
 waves-0.048
 shark-0.047
 victorious-0.045
 screams-0.045
 cliff-0.043
Token,
Token ID11
Feature activation+8.21e-01

Pos logit contributions
 shark+0.093
 waves+0.091
 victorious+0.089
ls+0.086
 cliff+0.086
Neg logit contributions
isk-0.126
 whether-0.117
ety-0.114
my-0.109
coat-0.107
Token there
Token ID612
Feature activation+2.13e+00

Pos logit contributions
isk+0.057
 whether+0.051
coat+0.049
ety+0.049
ine+0.047
Neg logit contributions
 shark-0.038
 waves-0.035
ls-0.035
 boat-0.035
 victorious-0.035
Token was
Token ID373
Feature activation-5.08e-02

Pos logit contributions
isk+0.136
 whether+0.129
ety+0.125
coat+0.118
my+0.114
Neg logit contributions
 waves-0.107
 shark-0.106
 screams-0.099
 victorious-0.097
 cliff-0.097
Token a
Token ID257
Feature activation+1.20e+00

Pos logit contributions
 waves+0.003
 shark+0.002
 victorious+0.002
 refreshed+0.002
 boat+0.002
Neg logit contributions
isk-0.003
 whether-0.003
ety-0.003
coat-0.003
ine-0.003
Token shy
Token ID15800
Feature activation-2.14e-01

Pos logit contributions
isk+0.083
 whether+0.080
ety+0.078
my+0.077
coat+0.072
Neg logit contributions
 waves-0.052
 victorious-0.050
 shark-0.048
 screams-0.048
 plunge-0.047
Token little
Token ID1310
Feature activation-3.69e+00

Pos logit contributions
 waves+0.010
 shark+0.009
 victorious+0.009
 screams+0.009
 cliff+0.009
Neg logit contributions
isk-0.015
 whether-0.014
ety-0.013
my-0.013
coat-0.013
Token girl
Token ID2576
Feature activation-9.88e-01

Pos logit contributions
 victorious+0.200
 waves+0.195
 screams+0.185
 cheers+0.180
 shark+0.180
Neg logit contributions
isk-0.210
ety-0.203
my-0.201
 whether-0.201
coat-0.186
Token named
Token ID3706
Feature activation+2.87e+00

Pos logit contributions
 boat+0.044
 wave+0.042
 waves+0.041
 shark+0.041
 whale+0.041
Neg logit contributions
isk-0.066
 whether-0.058
books-0.057
organ-0.056
coat-0.056
Token Lily
Token ID20037
Feature activation+4.83e+00

Pos logit contributions
isk+0.140
 whether+0.134
ine+0.128
ety+0.114
kit+0.114
Neg logit contributions
 refreshed-0.157
 plunge-0.156
 we-0.154
 waves-0.153
ls-0.153
Token.
Token ID13
Feature activation+2.54e-01

Pos logit contributions
isk+0.407
coat+0.374
ptions+0.362
 whether+0.359
 gloss+0.357
Neg logit contributions
 waves-0.124
 wave-0.117
 boat-0.115
 shark-0.103
 screams-0.096
Token.
Token ID13
Feature activation+9.83e-02

Pos logit contributions
 waves+0.113
 shark+0.112
 victorious+0.111
 screams+0.108
 current+0.105
Neg logit contributions
isk-0.085
 whether-0.081
my-0.076
ety-0.075
coat-0.069
Token They
Token ID1119
Feature activation-2.34e+00

Pos logit contributions
isk+0.005
 whether+0.005
ety+0.005
my+0.005
coat+0.005
Neg logit contributions
 waves-0.006
 shark-0.006
 victorious-0.006
 screams-0.005
 cliff-0.005
Token are
Token ID389
Feature activation-1.66e+00

Pos logit contributions
 shark+0.127
 waves+0.127
 victorious+0.125
 screams+0.118
 current+0.118
Neg logit contributions
isk-0.138
 whether-0.130
ety-0.124
my-0.124
coat-0.115
Token happy
Token ID3772
Feature activation+1.49e-01

Pos logit contributions
 shark+0.082
 waves+0.080
 cliff+0.078
 plunge+0.078
 victorious+0.076
Neg logit contributions
isk-0.104
 whether-0.103
my-0.099
ety-0.097
coat-0.090
Token.
Token ID13
Feature activation+5.92e-01

Pos logit contributions
 whether+0.006
isk+0.005
my+0.005
ety+0.005
 drew+0.004
Neg logit contributions
 victorious-0.011
 plunge-0.011
 waves-0.010
 current-0.010
 cliff-0.010
Token They
Token ID1119
Feature activation-2.15e+00

Pos logit contributions
isk+0.033
 whether+0.031
ety+0.030
my+0.030
coat+0.028
Neg logit contributions
 waves-0.034
 shark-0.033
 victorious-0.033
 screams-0.032
 cliff-0.031
Token are
Token ID389
Feature activation-1.42e+00

Pos logit contributions
 shark+0.130
rier+0.126
 waves+0.122
 victorious+0.122
ls+0.121
Neg logit contributions
isk-0.118
 whether-0.107
my-0.106
ety-0.104
 drew-0.097
Token friends
Token ID2460
Feature activation-7.15e+00

Pos logit contributions
 cliff+0.068
 shark+0.068
 screams+0.068
 plunge+0.068
 current+0.067
Neg logit contributions
 whether-0.089
isk-0.087
my-0.085
ety-0.083
 her-0.078
Token.
Token ID13
Feature activation+7.24e-01

Pos logit contributions
 current+0.563
 plunge+0.560
 waves+0.555
 cliff+0.553
 shark+0.543
Neg logit contributions
 whether-0.245
my-0.243
isk-0.216
ety-0.208
coat-0.171
Token<|endoftext|>
Token ID50256
Feature activation-1.48e+00

Pos logit contributions
isk+0.040
 whether+0.038
ety+0.037
my+0.036
coat+0.033
Neg logit contributions
 waves-0.042
 shark-0.041
 victorious-0.041
 screams-0.039
 cliff-0.039
TokenTom
Token ID13787
Feature activation+4.19e+00

Pos logit contributions
 waves+0.081
 victorious+0.079
 shark+0.079
 current+0.075
 screams+0.075
Neg logit contributions
isk-0.086
 whether-0.081
my-0.079
ety-0.079
coat-0.071
Token his
Token ID465
Feature activation-5.25e-01

Pos logit contributions
 waves+0.010
 victorious+0.010
 shark+0.009
 screams+0.009
ls+0.009
Neg logit contributions
isk-0.009
 whether-0.008
my-0.008
ety-0.008
ine-0.007
Token family
Token ID1641
Feature activation-4.88e+00

Pos logit contributions
 waves+0.023
 shark+0.023
 victorious+0.022
 screams+0.021
 cliff+0.021
Neg logit contributions
isk-0.036
 whether-0.034
my-0.034
ety-0.033
ine-0.030
Token enjoy
Token ID2883
Feature activation-1.26e+00

Pos logit contributions
 shark+0.309
 waves+0.302
 victorious+0.297
 current+0.296
rier+0.296
Neg logit contributions
isk-0.245
my-0.235
 whether-0.223
ety-0.216
 drew-0.209
Token mel
Token ID7758
Feature activation+7.10e+00

Pos logit contributions
 waves+0.059
 victorious+0.055
 shark+0.055
 shore+0.052
 cliff+0.052
Neg logit contributions
isk-0.088
ety-0.082
 whether-0.081
my-0.079
coat-0.077
Tokenons
Token ID684
Feature activation+5.73e+00

Pos logit contributions
isk+0.383
 whether+0.362
ety+0.342
my+0.325
coat+0.317
Neg logit contributions
 waves-0.424
 victorious-0.410
 shark-0.401
ls-0.389
 cliff-0.384
Token all
Token ID477
Feature activation-4.09e+00

Pos logit contributions
isk+0.287
my+0.278
 whether+0.271
ety+0.262
coat+0.240
Neg logit contributions
 shark-0.356
 waves-0.353
 current-0.350
 screams-0.350
 cliff-0.341
Token the
Token ID262
Feature activation-1.46e+00

Pos logit contributions
 waves+0.225
 shark+0.223
 victorious+0.217
 screams+0.212
 current+0.210
Neg logit contributions
isk-0.238
 whether-0.229
my-0.224
ety-0.219
ine-0.197
Token time
Token ID640
Feature activation-1.96e+00

Pos logit contributions
 shark+0.091
 waves+0.086
 current+0.086
 victorious+0.086
 screams+0.085
Neg logit contributions
isk-0.074
my-0.073
 whether-0.073
ety-0.067
ine-0.062
Token.
Token ID13
Feature activation+1.22e-01

Pos logit contributions
 shark+0.128
 waves+0.125
 victorious+0.124
 cliff+0.123
 screams+0.123
Neg logit contributions
 whether-0.088
isk-0.086
my-0.081
ety-0.079
ine-0.072
Token They
Token ID1119
Feature activation-2.29e+00

Pos logit contributions
isk+0.007
 whether+0.006
my+0.006
ety+0.006
tr+0.005
Neg logit contributions
 waves-0.008
 shark-0.007
 victorious-0.007
 screams-0.007
ls-0.007
Token know
Token ID760
Feature activation+3.07e+00

Pos logit contributions
 shark+0.133
 waves+0.130
rier+0.128
 screams+0.127
 current+0.127
Neg logit contributions
isk-0.130
my-0.123
 whether-0.119
ety-0.114
 drew-0.107
Token between
Token ID1022
Feature activation-2.51e+00

Pos logit contributions
 waves+0.157
 shark+0.152
 victorious+0.151
 screams+0.145
 cliff+0.142
Neg logit contributions
isk-0.140
 whether-0.129
ety-0.127
my-0.121
coat-0.114
Token playing
Token ID2712
Feature activation+2.58e+00

Pos logit contributions
 victorious+0.119
 waves+0.118
 lighthouse+0.117
 shark+0.116
 cliff+0.114
Neg logit contributions
 whether-0.160
isk-0.147
my-0.146
ety-0.136
ine-0.131
Token and
Token ID290
Feature activation+1.92e-02

Pos logit contributions
isk+0.139
 whether+0.131
ety+0.126
my+0.123
coat+0.114
Neg logit contributions
 waves-0.154
 shark-0.151
 victorious-0.149
 screams-0.143
 cliff-0.142
Token doing
Token ID1804
Feature activation+2.12e+00

Pos logit contributions
isk+0.001
 whether+0.001
my+0.001
ety+0.001
 drew+0.001
Neg logit contributions
 waves-0.001
 shark-0.001
 victorious-0.001
 screams-0.001
ls-0.001
Token your
Token ID534
Feature activation-1.98e+00

Pos logit contributions
isk+0.122
 whether+0.119
my+0.110
ety+0.108
ine+0.101
Neg logit contributions
 waves-0.116
 shark-0.115
 victorious-0.110
 current-0.108
 cliff-0.108
Token job
Token ID1693
Feature activation+2.89e-01

Pos logit contributions
 shark+0.084
 lighthouse+0.084
 waves+0.084
 victorious+0.080
 current+0.078
Neg logit contributions
isk-0.135
 whether-0.133
my-0.125
ine-0.120
ety-0.118
Token.
Token ID13
Feature activation+4.49e-01

Pos logit contributions
isk+0.015
 whether+0.014
my+0.013
ety+0.013
ine+0.012
Neg logit contributions
 waves-0.018
 shark-0.018
 victorious-0.017
 cliff-0.017
 current-0.017
Token Doing
Token ID25327
Feature activation-1.63e-01

Pos logit contributions
isk+0.023
 whether+0.022
my+0.020
ety+0.020
ine+0.018
Neg logit contributions
 waves-0.029
 shark-0.027
 victorious-0.027
ls-0.027
 cliff-0.026
Token something
Token ID1223
Feature activation+5.41e+00

Pos logit contributions
 waves+0.009
 shark+0.009
 victorious+0.009
 cliff+0.009
 current+0.009
Neg logit contributions
isk-0.009
 whether-0.009
my-0.008
ety-0.008
ine-0.007
Token you
Token ID345
Feature activation-6.81e+00

Pos logit contributions
 whether+0.289
my+0.259
isk+0.234
 which+0.230
 makeup+0.228
Neg logit contributions
 current-0.320
 waves-0.311
 shark-0.303
 lighthouse-0.303
 cliff-0.299
Token are
Token ID389
Feature activation-1.42e+00

Pos logit contributions
 shark+0.413
 waves+0.406
 victorious+0.388
 lighthouse+0.385
rier+0.385
Neg logit contributions
isk-0.366
 whether-0.346
my-0.339
ety-0.317
 drew-0.301
Token is
Token ID318
Feature activation-8.58e-01

Pos logit contributions
isk+0.042
 whether+0.041
ety+0.039
my+0.038
coat+0.037
Neg logit contributions
 waves-0.029
 shark-0.028
 victorious-0.028
 screams-0.027
 boat-0.026
Token playing
Token ID2712
Feature activation+2.27e+00

Pos logit contributions
 waves+0.048
 shark+0.047
 cliff+0.046
 current+0.045
 victorious+0.045
Neg logit contributions
isk-0.049
 whether-0.046
my-0.044
ety-0.043
coat-0.039
Token with
Token ID351
Feature activation+2.10e+00

Pos logit contributions
isk+0.135
ety+0.130
 whether+0.125
my+0.119
ine+0.114
Neg logit contributions
 waves-0.126
 victorious-0.124
 shark-0.122
 boat-0.115
 cheers-0.112
Token it
Token ID340
Feature activation+2.68e+00

Pos logit contributions
isk+0.155
ety+0.147
 whether+0.142
tr+0.139
coat+0.139
Neg logit contributions
 waves-0.090
 shark-0.085
 victorious-0.081
 boat-0.077
 screams-0.077
Token.
Token ID13
Feature activation+1.03e+00

Pos logit contributions
isk+0.154
 whether+0.141
ety+0.140
my+0.132
coat+0.128
Neg logit contributions
 waves-0.152
 shark-0.149
 victorious-0.148
 boat-0.140
 screams-0.140
Token He
Token ID679
Feature activation-2.24e+00

Pos logit contributions
isk+0.077
 whether+0.072
ety+0.071
my+0.069
coat+0.069
Neg logit contributions
 waves-0.044
 shark-0.042
 victorious-0.041
 screams-0.039
 current-0.038
Token feels
Token ID5300
Feature activation-3.15e+00

Pos logit contributions
 victorious+0.120
rier+0.118
 shark+0.117
 plunge+0.116
 current+0.115
Neg logit contributions
isk-0.128
 whether-0.122
my-0.119
ety-0.113
 drew-0.108
Token angry
Token ID7954
Feature activation-2.67e+00

Pos logit contributions
 waves+0.148
 shark+0.145
 cliff+0.141
 victorious+0.140
ls+0.136
Neg logit contributions
isk-0.200
 whether-0.198
my-0.185
ety-0.183
coat-0.170
Token.
Token ID13
Feature activation+1.04e+00

Pos logit contributions
 waves+0.143
 shark+0.135
 screams+0.129
 victorious+0.128
 cheers+0.126
Neg logit contributions
isk-0.176
ety-0.159
coat-0.154
 whether-0.150
tr-0.144
Token He
Token ID679
Feature activation-2.25e+00

Pos logit contributions
isk+0.082
ety+0.076
 whether+0.075
coat+0.074
my+0.073
Neg logit contributions
 waves-0.041
 shark-0.040
 victorious-0.038
 screams-0.035
 current-0.035
Token wants
Token ID3382
Feature activation+1.19e+00

Pos logit contributions
 victorious+0.117
 shark+0.115
 lighthouse+0.112
rier+0.112
 cliff+0.112
Neg logit contributions
isk-0.134
 whether-0.127
my-0.122
ety-0.120
 drew-0.113
Token what
Token ID644
Feature activation+8.92e+00

Pos logit contributions
isk+0.070
ety+0.063
coat+0.058
tr+0.055
my+0.052
Neg logit contributions
 waves-0.106
 shark-0.104
 victorious-0.101
 boat-0.101
 shore-0.098
Token was
Token ID373
Feature activation+3.30e-01

Pos logit contributions
isk+0.588
ety+0.533
 whether+0.531
my+0.508
coat+0.502
Neg logit contributions
 waves-0.453
 shark-0.444
 victorious-0.425
 boat-0.410
 cliff-0.410
Token inside
Token ID2641
Feature activation+2.77e+00

Pos logit contributions
isk+0.021
ety+0.019
 whether+0.018
my+0.018
coat+0.017
Neg logit contributions
 waves-0.018
 shark-0.017
 victorious-0.017
 cliff-0.016
 boat-0.016
Token it
Token ID340
Feature activation+2.79e+00

Pos logit contributions
isk+0.152
 whether+0.145
my+0.137
ety+0.137
coat+0.123
Neg logit contributions
 waves-0.162
 shark-0.158
 victorious-0.156
 screams-0.150
 cliff-0.148
Token.
Token ID13
Feature activation+5.80e-01

Pos logit contributions
isk+0.162
ety+0.146
 whether+0.146
coat+0.137
my+0.137
Neg logit contributions
 waves-0.159
 shark-0.153
 victorious-0.152
 cliff-0.145
 boat-0.145
Token Brad
Token ID8114
Feature activation-7.58e+00

Pos logit contributions
isk+0.037
 whether+0.034
ety+0.034
my+0.033
coat+0.032
Neg logit contributions
 waves-0.029
 victorious-0.028
 shark-0.028
 screams-0.027
 current-0.026
Token decided
Token ID3066
Feature activation+8.86e-01

Pos logit contributions
 waves+0.447
 shark+0.435
 victorious+0.430
 screams+0.413
 cliff+0.410
Neg logit contributions
isk-0.415
 whether-0.390
ety-0.374
my-0.367
coat-0.341
Token to
Token ID284
Feature activation+1.33e+00

Pos logit contributions
isk+0.046
ety+0.041
 whether+0.041
my+0.040
coat+0.038
Neg logit contributions
 waves-0.056
 victorious-0.054
 shark-0.054
 screams-0.052
 cliff-0.051
Token ask
Token ID1265
Feature activation+3.16e+00

Pos logit contributions
isk+0.070
 whether+0.067
my+0.064
ety+0.061
 drew+0.057
Neg logit contributions
 shark-0.079
 victorious-0.079
 waves-0.078
ls-0.077
rier-0.074
Token his
Token ID465
Feature activation-1.90e-01

Pos logit contributions
isk+0.198
ety+0.191
 whether+0.182
my+0.176
coat+0.174
Neg logit contributions
 waves-0.161
 shark-0.159
 victorious-0.157
 boat-0.146
 screams-0.146
Token best
Token ID1266
Feature activation-5.37e+00

Pos logit contributions
 waves+0.008
 victorious+0.008
 shark+0.008
 cliff+0.008
 current+0.008
Neg logit contributions
isk-0.013
 whether-0.013
my-0.012
ety-0.012
ine-0.011
Token time
Token ID640
Feature activation-2.01e+00

Pos logit contributions
isk+0.069
 whether+0.066
ety+0.063
my+0.062
coat+0.059
Neg logit contributions
 waves-0.051
 shark-0.050
 victorious-0.049
 screams-0.047
 cliff-0.046
Token,
Token ID11
Feature activation+8.17e-01

Pos logit contributions
 waves+0.102
 shark+0.101
 victorious+0.098
 cliff+0.094
 screams+0.094
Neg logit contributions
isk-0.125
 whether-0.117
ety-0.113
my-0.109
coat-0.105
Token there
Token ID612
Feature activation+2.15e+00

Pos logit contributions
isk+0.056
 whether+0.051
coat+0.047
ety+0.047
ine+0.046
Neg logit contributions
 shark-0.037
 victorious-0.036
ls-0.036
 waves-0.035
 boat-0.035
Token were
Token ID547
Feature activation-1.48e+00

Pos logit contributions
isk+0.136
 whether+0.129
ety+0.124
coat+0.118
my+0.115
Neg logit contributions
 waves-0.108
 shark-0.107
 victorious-0.099
 screams-0.099
 cliff-0.099
Token two
Token ID734
Feature activation-4.34e+00

Pos logit contributions
 victorious+0.079
 waves+0.076
 shark+0.076
rier+0.074
 screams+0.074
Neg logit contributions
my-0.085
ety-0.084
isk-0.082
 whether-0.081
 drew-0.074
Token friends
Token ID2460
Feature activation-7.37e+00

Pos logit contributions
 victorious+0.159
 shark+0.150
 waves+0.146
 cliff+0.146
 lighthouse+0.145
Neg logit contributions
isk-0.333
my-0.326
 whether-0.317
ety-0.315
ine-0.289
Token.
Token ID13
Feature activation+3.89e-01

Pos logit contributions
 waves+0.523
 shark+0.515
 victorious+0.505
 boat+0.491
 screams+0.488
Neg logit contributions
isk-0.316
 whether-0.285
ety-0.274
my-0.255
coat-0.252
Token Their
Token ID5334
Feature activation-1.03e+00

Pos logit contributions
isk+0.027
 whether+0.026
coat+0.025
ety+0.023
my+0.023
Neg logit contributions
 shark-0.016
 waves-0.015
 cheers-0.014
 floats-0.014
 fishes-0.014
Token names
Token ID3891
Feature activation+5.94e+00

Pos logit contributions
 waves+0.067
 shark+0.066
 victorious+0.066
 screams+0.063
 cliff+0.063
Neg logit contributions
isk-0.050
 whether-0.046
ety-0.044
my-0.043
coat-0.039
Token were
Token ID547
Feature activation-1.32e+00

Pos logit contributions
isk+0.339
 whether+0.319
my+0.310
ety+0.306
coat+0.281
Neg logit contributions
 waves-0.331
 shark-0.324
 victorious-0.321
 cliff-0.309
 screams-0.308
Token Lisa
Token ID15378
Feature activation+2.16e+00

Pos logit contributions
 waves+0.086
 victorious+0.085
 shark+0.084
 current+0.082
 screams+0.082
Neg logit contributions
isk-0.063
 whether-0.060
my-0.057
ety-0.057
 drew-0.049
Token An
Token ID1052
Feature activation+1.16e+00

Pos logit contributions
 waves+0.160
 victorious+0.158
 shark+0.156
 cheers+0.153
 screams+0.152
Neg logit contributions
isk-0.129
 whether-0.119
ety-0.115
my-0.109
coat-0.102
Token ice
Token ID4771
Feature activation+1.68e+00

Pos logit contributions
isk+0.082
 whether+0.078
ety+0.076
my+0.074
coat+0.071
Neg logit contributions
 waves-0.051
 shark-0.049
 victorious-0.048
 screams-0.046
 cliff-0.045
Token cream
Token ID8566
Feature activation+1.31e+00

Pos logit contributions
isk+0.101
 whether+0.096
ety+0.092
my+0.091
coat+0.084
Neg logit contributions
 waves-0.091
 shark-0.088
 victorious-0.087
 screams-0.083
 cliff-0.082
Token party
Token ID2151
Feature activation-2.52e+00

Pos logit contributions
isk+0.084
 whether+0.080
ety+0.077
my+0.075
coat+0.071
Neg logit contributions
 waves-0.065
 shark-0.063
 victorious-0.062
 screams-0.059
 cliff-0.059
Token!
Token ID0
Feature activation-2.51e+00

Pos logit contributions
 waves+0.142
 shark+0.136
 victorious+0.136
 screams+0.130
 cliff+0.128
Neg logit contributions
isk-0.145
 whether-0.137
ety-0.133
my-0.128
coat-0.118
Token Everyone
Token ID11075
Feature activation-8.14e+00

Pos logit contributions
 waves+0.156
 victorious+0.153
 shark+0.151
 screams+0.146
 cheers+0.145
Neg logit contributions
isk-0.130
 whether-0.121
ety-0.116
my-0.112
coat-0.105
Token was
Token ID373
Feature activation+5.70e-02

Pos logit contributions
 current+0.556
rier+0.549
 victorious+0.536
 drown+0.530
 shark+0.529
Neg logit contributions
my-0.350
 whether-0.334
 drew-0.318
isk-0.318
ety-0.295
Token so
Token ID523
Feature activation-3.82e+00

Pos logit contributions
isk+0.003
 whether+0.003
my+0.003
ety+0.003
ine+0.003
Neg logit contributions
 shark-0.003
 waves-0.003
 current-0.003
ls-0.003
 cliff-0.003
Token happy
Token ID3772
Feature activation+3.94e-01

Pos logit contributions
 shark+0.243
 current+0.243
 hawk+0.234
 plunge+0.232
 cliff+0.225
Neg logit contributions
 whether-0.181
my-0.173
 her-0.158
isk-0.156
 drew-0.154
Token and
Token ID290
Feature activation-3.83e-02

Pos logit contributions
isk+0.019
 whether+0.019
ety+0.018
my+0.018
coat+0.016
Neg logit contributions
 waves-0.025
 shark-0.024
 victorious-0.024
 current-0.023
 cliff-0.023
Token the
Token ID262
Feature activation-1.42e+00

Pos logit contributions
 shark+0.002
 hawk+0.002
 cliff+0.002
 current+0.002
rier+0.002
Neg logit contributions
 whether-0.002
my-0.002
isk-0.002
 her-0.002
 drew-0.002
Token The
Token ID383
Feature activation+9.45e-02

Pos logit contributions
isk+0.005
 whether+0.005
ety+0.005
my+0.005
coat+0.004
Neg logit contributions
 waves-0.006
 shark-0.006
 victorious-0.006
 screams-0.006
 cliff-0.006
Token bird
Token ID6512
Feature activation-1.09e+00

Pos logit contributions
my+0.005
isk+0.005
ety+0.004
 whether+0.004
coat+0.004
Neg logit contributions
 victorious-0.005
 waves-0.005
 screams-0.005
ls-0.005
 cheers-0.005
Token was
Token ID373
Feature activation-1.26e-01

Pos logit contributions
 shark+0.067
 waves+0.067
 victorious+0.066
 screams+0.063
 cliff+0.063
Neg logit contributions
isk-0.056
 whether-0.052
my-0.052
ety-0.050
 drew-0.046
Token happy
Token ID3772
Feature activation+3.71e-01

Pos logit contributions
 shark+0.007
 waves+0.007
ls+0.006
 screams+0.006
 cliff+0.006
Neg logit contributions
isk-0.008
 whether-0.008
my-0.007
ety-0.007
ine-0.007
Token and
Token ID290
Feature activation-1.87e-01

Pos logit contributions
isk+0.017
 whether+0.016
ety+0.015
my+0.015
coat+0.013
Neg logit contributions
 waves-0.025
 shark-0.024
 victorious-0.024
 screams-0.024
 cliff-0.023
Token thanked
Token ID26280
Feature activation-9.15e+00

Pos logit contributions
rier+0.011
 shark+0.011
 plunge+0.010
 cliff+0.010
 screams+0.010
Neg logit contributions
 drew-0.009
 whether-0.009
isk-0.009
my-0.009
 her-0.009
Token the
Token ID262
Feature activation-1.67e+00

Pos logit contributions
 waves+0.486
 victorious+0.465
 shark+0.464
 screams+0.453
 cliff+0.443
Neg logit contributions
isk-0.553
 whether-0.527
my-0.495
ety-0.492
ine-0.448
Token bunny
Token ID44915
Feature activation-3.54e+00

Pos logit contributions
 waves+0.091
 victorious+0.091
ls+0.088
 screams+0.088
 sinks+0.085
Neg logit contributions
isk-0.089
my-0.087
ety-0.084
 whether-0.083
coat-0.076
Token.
Token ID13
Feature activation+1.11e-01

Pos logit contributions
 waves+0.228
 shark+0.223
 victorious+0.219
 screams+0.213
 cliff+0.211
Neg logit contributions
isk-0.174
 whether-0.163
ety-0.155
my-0.154
coat-0.138
Token The
Token ID383
Feature activation+6.09e-02

Pos logit contributions
isk+0.005
 whether+0.005
ety+0.005
my+0.005
 drew+0.004
Neg logit contributions
 waves-0.007
 shark-0.007
 victorious-0.007
 plunge-0.007
 lighthouse-0.007
Token bunny
Token ID44915
Feature activation-3.54e+00

Pos logit contributions
my+0.003
isk+0.003
ety+0.003
 whether+0.003
 knit+0.003
Neg logit contributions
 waves-0.003
 victorious-0.003
 screams-0.003
ls-0.003
vation-0.003
Token
Token ID198
Feature activation-6.35e-01

Pos logit contributions
 victorious+0.030
 waves+0.030
 shark+0.029
 screams+0.028
ls+0.027
Neg logit contributions
isk-0.042
 whether-0.039
ety-0.038
my-0.036
coat-0.036
TokenL
Token ID43
Feature activation+9.92e+00

Pos logit contributions
 victorious+0.035
 waves+0.033
 refreshed+0.033
 losers+0.032
Everyone+0.032
Neg logit contributions
isk-0.035
 whether-0.032
ety-0.031
my-0.029
coat-0.028
Tokenily
Token ID813
Feature activation+2.39e+00

Pos logit contributions
 whether+0.563
isk+0.544
ety+0.510
tr+0.500
coat+0.488
Neg logit contributions
 victorious-0.494
 waves-0.487
ls-0.482
 screams-0.478
 losers-0.470
Token showed
Token ID3751
Feature activation+2.82e+00

Pos logit contributions
isk+0.140
 whether+0.132
ety+0.124
coat+0.115
my+0.113
Neg logit contributions
 waves-0.137
 victorious-0.125
 screams-0.125
 shark-0.123
 cheers-0.122
Token her
Token ID607
Feature activation+4.95e+00

Pos logit contributions
isk+0.074
 whether+0.064
ety+0.058
my+0.056
coat+0.046
Neg logit contributions
 waves-0.248
 shark-0.244
 victorious-0.242
 screams-0.236
 cliff-0.234
Token mum
Token ID25682
Feature activation-7.74e+00

Pos logit contributions
 whether+0.334
my+0.325
isk+0.325
ety+0.316
 makeup+0.300
Neg logit contributions
 victorious-0.202
rier-0.194
 shark-0.192
 losers-0.190
 current-0.189
Token the
Token ID262
Feature activation-1.58e+00

Pos logit contributions
 waves+0.321
 shark+0.268
 boat+0.264
 cheers+0.262
 wave+0.249
Neg logit contributions
isk-0.574
tr-0.558
coat-0.552
ety-0.540
 whether-0.524
Token line
Token ID1627
Feature activation-3.84e+00

Pos logit contributions
 waves+0.075
 victorious+0.068
 boat+0.067
 cliff+0.066
 screams+0.066
Neg logit contributions
isk-0.114
 whether-0.105
ety-0.102
my-0.098
coat-0.091
Token and
Token ID290
Feature activation-6.34e-02

Pos logit contributions
 waves+0.208
 victorious+0.199
 shark+0.194
 screams+0.187
 boat+0.187
Neg logit contributions
isk-0.236
 whether-0.218
ety-0.212
my-0.200
coat-0.191
Token mum
Token ID25682
Feature activation-7.75e+00

Pos logit contributions
 victorious+0.003
 waves+0.003
 shark+0.003
 cliff+0.003
 current+0.003
Neg logit contributions
isk-0.004
 whether-0.004
my-0.004
ety-0.003
 drew-0.003
Token said
Token ID531
Feature activation-2.73e+00

Pos logit contributions
 waves+0.418
 shark+0.385
 victorious+0.384
 cheers+0.374
 screams+0.369
Neg logit contributions
isk-0.486
 whether-0.455
ety-0.441
coat-0.407
my-0.404