TOP ACTIVATIONS
MAX = 5.102

They are different colors and sizes . Some are red ,
trucks of different colors and sizes . He liked to make
are in different shapes and sizes . They are in a
it with holes of different sizes . I will make it
shapes by their colors and sizes and made their own puzzles
clay of all kinds and sizes and shapes . They are
They were different colors and sizes . Lily picked a pink
cars of different colors and sizes . He likes to make
cars of different colors and sizes . They like to drive
with the different colors and sizes . They are very proud
were all different colors and sizes . She wondered how many
cars of different colors and sizes . He likes to make
the blocks , compare the sizes , and sort the colors
were of different colors and sizes . They filled the air
they 're different colors and sizes ." Max started
fish of different colors and sizes . They swim and splash
truck . They were different sizes and colors . Some had
three shapes , all different sizes . The shape
prints in different colors and sizes . The
cars of different colors and sizes . They make a line

MIN ACTIVATIONS
MIN = -7.343

there was a three - year - old who found a
was an average three - year - old boy . He
saw a little 3 - year - old boy . He
olly was a three - year old girl who loved to
Joe , a 3 - year - old boy . He
Pat was a 3 - year - old boy . He
a very adventurous three - year - old . She was
. Two curious three - year - old children , Bob
there was a 4 - year old girl named Sus ie
buried , the 3 - year - old and the parent
was an attractive three - year - old who loved reading
there was a three - year - old boy called Tommy
and their little three - year - old daughter . They
ley was a three - year - old who loved to
The three - year - old girl was fascinated
A 3 - year - old toddler walked up
mum and her three - year - old daughter went to
She told her three - year - old daughter to attach
on , the three - year - old started testing and
, a little three - year - old boy named John

INTERVAL 1.990 - 5.102
CONTAINS 1.455%

, she looked at her wall and smiled . It was
will be in trouble !" Anna said . " We were
's shoulder . " That 's very kind of you to
and angry . He does not like Mia . He does
's go play . Let 's make a snow man friend

INTERVAL -1.121 - 1.990
CONTAINS 90.879%

for the surprise . Come to the centre of the room
The moral of the story is that it 's always best
until , eventually , a flock of birds flew out of
could have a turn and Jane said yes .
it was too wind y and the waves were too big

INTERVAL -4.232 - -1.121
CONTAINS 7.650%

mom my asked her to help with the grocery shopping .
" Just then , the rabbit heard a
it with both hands . It was too heavy for her
pulled a giant carrot . It was bigger than the other
She saw a plane pass by in the sky one day

INTERVAL -7.343 - -4.232
CONTAINS 0.016%

ball . The 3 - year - old was so excited
. She tried to sh oo him away , but it
ate too much ice - cream already . Tim my got
kinds of delicious ice - cream flavors . " Can I
Alex for some ice - cream . He smiled and was
Token They
Token ID1119
Feature activation-4.33e-01

Pos logit contributions
cks+0.026
 lair+0.026
 pads+0.025
 necks+0.024
 floor+0.024
Neg logit contributions
 reward-0.039
 plunge-0.038
 help-0.037
ar-0.037
less-0.036
Token are
Token ID389
Feature activation+2.58e-01

Pos logit contributions
 help+0.028
 reward+0.023
 plunge+0.023
ar+0.021
f+0.020
Neg logit contributions
 lair-0.045
llers-0.042
cks-0.042
 pads-0.041
 floor-0.040
Token different
Token ID1180
Feature activation+5.03e-01

Pos logit contributions
 lair+0.021
opl+0.021
llers+0.021
 pads+0.020
 floor+0.020
Neg logit contributions
 help-0.019
 plunge-0.016
 Bub-0.016
 reward-0.016
less-0.015
Token colors
Token ID7577
Feature activation+1.34e+00

Pos logit contributions
 lair+0.044
cks+0.043
 pads+0.041
 mess+0.041
 floor+0.040
Neg logit contributions
 help-0.034
 reward-0.033
 plunge-0.032
ar-0.032
less-0.031
Token and
Token ID290
Feature activation+4.57e-01

Pos logit contributions
 lair+0.141
cks+0.132
llers+0.130
 mess+0.130
 necks+0.128
Neg logit contributions
 help-0.068
less-0.065
iest-0.060
 reward-0.060
 of-0.060
Token sizes
Token ID10620
Feature activation+5.10e+00

Pos logit contributions
 lair+0.034
cks+0.030
llers+0.029
 floor+0.029
opl+0.029
Neg logit contributions
 help-0.042
 reward-0.038
 plunge-0.035
ar-0.035
f-0.034
Token.
Token ID13
Feature activation+5.22e-01

Pos logit contributions
 lair+0.520
cks+0.484
llers+0.478
 mess+0.472
 pads+0.458
Neg logit contributions
 help-0.301
 reward-0.274
ar-0.260
less-0.256
 plunge-0.248
Token Some
Token ID2773
Feature activation-1.09e+00

Pos logit contributions
cks+0.034
 lair+0.034
 pads+0.033
 necks+0.032
 floor+0.032
Neg logit contributions
 reward-0.046
 plunge-0.044
 help-0.044
ar-0.043
less-0.042
Token are
Token ID389
Feature activation+3.40e-01

Pos logit contributions
 help+0.070
 reward+0.069
ar+0.067
 plunge+0.066
bye+0.063
Neg logit contributions
 lair-0.098
cks-0.097
 mess-0.094
 pads-0.092
 floor-0.091
Token red
Token ID2266
Feature activation+1.04e+00

Pos logit contributions
 lair+0.028
cks+0.026
opl+0.026
llers+0.026
 floor+0.025
Neg logit contributions
 help-0.026
 reward-0.024
less-0.023
 plunge-0.023
ar-0.023
Token,
Token ID11
Feature activation+3.09e-02

Pos logit contributions
cks+0.092
 lair+0.091
 mess+0.090
 pads+0.090
 necks+0.088
Neg logit contributions
 reward-0.066
 help-0.064
ar-0.063
 plunge-0.062
Sure-0.062
Token trucks
Token ID13960
Feature activation+9.10e-01

Pos logit contributions
 lair+0.036
cks+0.036
 mess+0.034
 pads+0.033
 floor+0.032
Neg logit contributions
 help-0.047
 reward-0.046
ar-0.044
 plunge-0.044
bye-0.042
Token of
Token ID286
Feature activation-7.15e-01

Pos logit contributions
 lair+0.074
cks+0.074
 mess+0.071
 pads+0.069
 floor+0.069
Neg logit contributions
 help-0.066
 reward-0.065
ar-0.063
 plunge-0.061
bye-0.060
Token different
Token ID1180
Feature activation+4.79e-01

Pos logit contributions
ar+0.049
bye+0.048
 reward+0.048
 plunge+0.047
less+0.044
Neg logit contributions
 mess-0.061
 lair-0.061
cks-0.060
 floor-0.059
 sizes-0.058
Token colors
Token ID7577
Feature activation+1.24e+00

Pos logit contributions
 lair+0.030
cks+0.030
 mess+0.028
 pads+0.028
 floor+0.027
Neg logit contributions
 help-0.043
 reward-0.043
ar-0.042
 plunge-0.041
bye-0.041
Token and
Token ID290
Feature activation+5.13e-01

Pos logit contributions
 lair+0.112
cks+0.110
 mess+0.105
 necks+0.102
 web+0.100
Neg logit contributions
 help-0.080
 reward-0.075
ar-0.074
less-0.073
ted-0.072
Token sizes
Token ID10620
Feature activation+5.00e+00

Pos logit contributions
 lair+0.024
cks+0.024
 mess+0.020
 pads+0.019
llers+0.019
Neg logit contributions
 help-0.058
 reward-0.055
ar-0.052
 plunge-0.051
bye-0.051
Token.
Token ID13
Feature activation+5.50e-01

Pos logit contributions
 lair+0.462
cks+0.460
 mess+0.434
llers+0.426
 necks+0.421
Neg logit contributions
 help-0.321
 reward-0.305
ar-0.293
less-0.283
 plunge-0.279
Token He
Token ID679
Feature activation+3.78e-01

Pos logit contributions
 lair+0.042
cks+0.042
 mess+0.040
 pads+0.039
 floor+0.039
Neg logit contributions
 help-0.043
 reward-0.042
ar-0.041
 plunge-0.040
Sure-0.040
Token liked
Token ID8288
Feature activation+3.75e-01

Pos logit contributions
cks+0.035
 lair+0.033
opl+0.031
 mess+0.030
 web+0.030
Neg logit contributions
 help-0.026
 reward-0.024
ar-0.024
Sure-0.023
f-0.023
Token to
Token ID284
Feature activation+3.13e-01

Pos logit contributions
cks+0.041
 lair+0.040
 pads+0.039
 mess+0.038
 necks+0.038
Neg logit contributions
 help-0.019
 reward-0.016
ted-0.015
ar-0.014
 Bub-0.014
Token make
Token ID787
Feature activation+1.21e+00

Pos logit contributions
 lair+0.031
cks+0.030
 pads+0.029
 mess+0.028
 floor+0.027
Neg logit contributions
 help-0.020
 reward-0.018
ar-0.017
bye-0.016
 plunge-0.016
Token are
Token ID389
Feature activation+1.82e-01

Pos logit contributions
 help+0.036
 reward+0.029
 plunge+0.025
f+0.024
Sure+0.023
Neg logit contributions
 lair-0.065
llers-0.061
opl-0.060
 pads-0.059
cks-0.059
Token in
Token ID287
Feature activation+1.61e-01

Pos logit contributions
opl+0.017
 lair+0.017
llers+0.016
storms+0.016
 pads+0.015
Neg logit contributions
 help-0.012
 reward-0.011
 Bub-0.009
 souven-0.009
less-0.009
Token different
Token ID1180
Feature activation+4.38e-01

Pos logit contributions
 lair+0.016
cks+0.015
 mess+0.015
 pads+0.015
 floor+0.015
Neg logit contributions
 help-0.009
 reward-0.009
ar-0.009
 plunge-0.009
bye-0.008
Token shapes
Token ID15268
Feature activation+2.71e+00

Pos logit contributions
cks+0.027
 lair+0.027
llers+0.025
 pads+0.025
opl+0.024
Neg logit contributions
 help-0.040
 reward-0.040
 plunge-0.037
ar-0.037
less-0.036
Token and
Token ID290
Feature activation+3.99e-01

Pos logit contributions
 lair+0.292
llers+0.289
cks+0.277
die+0.273
opl+0.272
Neg logit contributions
 help-0.128
 reward-0.108
less-0.108
 of-0.104
ar-0.103
Token sizes
Token ID10620
Feature activation+5.00e+00

Pos logit contributions
 lair+0.024
cks+0.022
opl+0.021
llers+0.020
 pads+0.019
Neg logit contributions
 help-0.040
 reward-0.038
 plunge-0.035
 souven-0.034
ar-0.034
Token.
Token ID13
Feature activation+3.15e-01

Pos logit contributions
 lair+0.534
cks+0.500
llers+0.497
opl+0.472
 pads+0.471
Neg logit contributions
 help-0.266
 reward-0.237
less-0.230
ar-0.218
f-0.211
Token They
Token ID1119
Feature activation-5.28e-01

Pos logit contributions
 pads+0.018
 lair+0.018
cks+0.018
 floor+0.017
 necks+0.017
Neg logit contributions
 plunge-0.030
 reward-0.029
ar-0.029
 help-0.029
iest-0.028
Token are
Token ID389
Feature activation+2.25e-01

Pos logit contributions
 help+0.034
 reward+0.028
f+0.023
 plunge+0.022
Sure+0.022
Neg logit contributions
 lair-0.059
llers-0.055
opl-0.055
cks-0.054
 pads-0.054
Token in
Token ID287
Feature activation+1.72e-01

Pos logit contributions
opl+0.020
 lair+0.020
llers+0.019
storms+0.019
cks+0.018
Neg logit contributions
 help-0.016
 reward-0.014
 Bub-0.012
 souven-0.012
less-0.012
Token a
Token ID257
Feature activation-3.00e-01

Pos logit contributions
 lair+0.016
cks+0.015
 pads+0.015
 mess+0.015
 floor+0.014
Neg logit contributions
 help-0.011
 reward-0.011
ar-0.010
 plunge-0.010
Sure-0.010
Token it
Token ID340
Feature activation-8.30e-01

Pos logit contributions
 lair+0.135
cks+0.130
 pads+0.129
 necks+0.126
 mess+0.126
Neg logit contributions
 help-0.059
 reward-0.056
 plunge-0.052
ar-0.050
bye-0.046
Token with
Token ID351
Feature activation+6.92e-01

Pos logit contributions
 help+0.055
 reward+0.052
 plunge+0.046
ar+0.044
bye+0.044
Neg logit contributions
cks-0.077
 lair-0.077
 necks-0.074
 pads-0.073
llers-0.072
Token holes
Token ID10421
Feature activation+2.27e+00

Pos logit contributions
 lair+0.064
cks+0.063
 mess+0.062
 pads+0.060
 floor+0.060
Neg logit contributions
 reward-0.042
 help-0.042
ar-0.041
 plunge-0.040
bye-0.038
Token of
Token ID286
Feature activation-7.80e-01

Pos logit contributions
 lair+0.185
cks+0.181
 mess+0.174
 pads+0.170
llers+0.168
Neg logit contributions
 help-0.166
 reward-0.162
ar-0.158
 plunge-0.155
less-0.151
Token different
Token ID1180
Feature activation+3.88e-01

Pos logit contributions
 help+0.047
 reward+0.044
 plunge+0.041
ar+0.038
ted+0.036
Neg logit contributions
 lair-0.076
cks-0.074
 pads-0.074
llers-0.070
 mess-0.070
Token sizes
Token ID10620
Feature activation+4.96e+00

Pos logit contributions
 lair+0.027
cks+0.025
 mess+0.024
 pads+0.023
llers+0.023
Neg logit contributions
 help-0.034
 reward-0.033
 plunge-0.032
ar-0.032
bye-0.030
Token.
Token ID13
Feature activation+3.04e-01

Pos logit contributions
 lair+0.457
cks+0.434
 mess+0.414
llers+0.411
 frown+0.403
Neg logit contributions
 help-0.333
 reward-0.318
ar-0.304
 plunge-0.295
less-0.293
Token I
Token ID314
Feature activation+3.23e-01

Pos logit contributions
 lair+0.024
cks+0.024
 pads+0.024
 floor+0.023
 mess+0.023
Neg logit contributions
 reward-0.022
 help-0.021
ar-0.021
 plunge-0.021
less-0.020
Token will
Token ID481
Feature activation-1.05e+00

Pos logit contributions
 lair+0.029
cks+0.029
 mess+0.028
 pads+0.027
 floor+0.027
Neg logit contributions
 help-0.020
 reward-0.020
ar-0.020
 plunge-0.020
bye-0.019
Token make
Token ID787
Feature activation+1.22e+00

Pos logit contributions
 help+0.070
 reward+0.065
ar+0.055
 plunge+0.055
bye+0.053
Neg logit contributions
 lair-0.104
cks-0.102
 pads-0.101
 necks-0.095
 mess-0.094
Token it
Token ID340
Feature activation-8.23e-01

Pos logit contributions
 lair+0.130
cks+0.126
 pads+0.125
 necks+0.122
 mess+0.121
Neg logit contributions
 help-0.062
 reward-0.058
 plunge-0.054
ar-0.053
bye-0.048
Token shapes
Token ID15268
Feature activation+2.70e+00

Pos logit contributions
 lair+0.031
cks+0.030
 pads+0.029
 mess+0.028
llers+0.028
Neg logit contributions
 help-0.039
 reward-0.038
 plunge-0.037
ar-0.036
Sure-0.035
Token by
Token ID416
Feature activation-1.28e+00

Pos logit contributions
 lair+0.263
cks+0.261
llers+0.250
 mess+0.248
 pads+0.247
Neg logit contributions
 help-0.158
 reward-0.152
ar-0.145
less-0.141
 plunge-0.140
Token their
Token ID511
Feature activation-5.12e-02

Pos logit contributions
 help+0.085
 reward+0.082
ar+0.078
 plunge+0.077
less+0.074
Neg logit contributions
 lair-0.115
cks-0.109
 pads-0.108
 mess-0.107
llers-0.103
Token colors
Token ID7577
Feature activation+1.20e+00

Pos logit contributions
 help+0.005
 reward+0.005
ar+0.004
 plunge+0.004
bye+0.004
Neg logit contributions
 lair-0.003
cks-0.003
 mess-0.003
 pads-0.003
 floor-0.003
Token and
Token ID290
Feature activation+2.64e-01

Pos logit contributions
 lair+0.101
cks+0.099
 mess+0.093
 pads+0.091
llers+0.090
Neg logit contributions
 help-0.085
 reward-0.084
ar-0.081
 plunge-0.079
less-0.079
Token sizes
Token ID10620
Feature activation+4.96e+00

Pos logit contributions
 lair+0.021
cks+0.021
 mess+0.019
opl+0.019
 pads+0.019
Neg logit contributions
 help-0.020
 reward-0.020
ar-0.018
 plunge-0.018
bye-0.018
Token and
Token ID290
Feature activation+2.79e-01

Pos logit contributions
 lair+0.455
cks+0.445
 mess+0.422
 pads+0.421
llers+0.415
Neg logit contributions
 help-0.322
 reward-0.312
ar-0.298
 plunge-0.293
less-0.291
Token made
Token ID925
Feature activation+3.87e-01

Pos logit contributions
cks+0.023
 lair+0.023
 mess+0.021
 pads+0.020
llers+0.020
Neg logit contributions
 help-0.021
 reward-0.021
ar-0.019
 plunge-0.019
bye-0.018
Token their
Token ID511
Feature activation-5.65e-02

Pos logit contributions
 lair+0.035
cks+0.035
 mess+0.034
 pads+0.034
 floor+0.033
Neg logit contributions
 help-0.024
 reward-0.024
ar-0.023
 plunge-0.022
Sure-0.021
Token own
Token ID898
Feature activation-3.84e-01

Pos logit contributions
 help+0.005
 reward+0.005
 plunge+0.004
f+0.004
bye+0.004
Neg logit contributions
cks-0.004
 lair-0.004
llers-0.004
 pads-0.004
opl-0.004
Token puzzles
Token ID24367
Feature activation+1.65e+00

Pos logit contributions
 help+0.031
 reward+0.031
 plunge+0.028
ar+0.027
less+0.027
Neg logit contributions
cks-0.029
 lair-0.028
 pads-0.028
llers-0.027
 necks-0.026
Token clay
Token ID21558
Feature activation+9.63e-01

Pos logit contributions
 lair+0.150
cks+0.148
 mess+0.143
 pads+0.142
 floor+0.140
Neg logit contributions
 help-0.091
 reward-0.090
ar-0.087
 plunge-0.086
bye-0.082
Token of
Token ID286
Feature activation-7.44e-01

Pos logit contributions
 lair+0.085
cks+0.083
 pads+0.080
 mess+0.079
 necks+0.078
Neg logit contributions
 help-0.065
 reward-0.063
ar-0.060
 plunge-0.059
ted-0.057
Token all
Token ID477
Feature activation-6.82e-02

Pos logit contributions
 help+0.041
 reward+0.041
ar+0.039
 plunge+0.038
bye+0.037
Neg logit contributions
 lair-0.073
cks-0.073
 mess-0.071
 pads-0.069
 floor-0.069
Token kinds
Token ID6982
Feature activation+7.52e-01

Pos logit contributions
 help+0.003
 reward+0.002
 plunge+0.002
iest+0.002
f+0.002
Neg logit contributions
 lair-0.008
 pads-0.008
 necks-0.007
cks-0.007
llers-0.007
Token and
Token ID290
Feature activation+4.24e-01

Pos logit contributions
 lair+0.098
cks+0.096
 pads+0.093
 mess+0.092
llers+0.092
Neg logit contributions
 help-0.020
 reward-0.017
ar-0.016
less-0.015
f-0.013
Token sizes
Token ID10620
Feature activation+4.95e+00

Pos logit contributions
 lair+0.024
cks+0.022
 pads+0.020
opl+0.019
llers+0.019
Neg logit contributions
 help-0.045
 reward-0.043
 plunge-0.040
f-0.039
ar-0.039
Token and
Token ID290
Feature activation+4.08e-01

Pos logit contributions
 lair+0.490
cks+0.465
llers+0.448
 mess+0.439
 necks+0.439
Neg logit contributions
 help-0.301
 reward-0.279
less-0.265
ar-0.264
 plunge-0.251
Token shapes
Token ID15268
Feature activation+2.67e+00

Pos logit contributions
 lair+0.026
cks+0.025
opl+0.022
 pads+0.022
llers+0.022
Neg logit contributions
 help-0.040
 reward-0.038
 plunge-0.034
ar-0.034
f-0.033
Token.
Token ID13
Feature activation+3.73e-01

Pos logit contributions
 lair+0.268
cks+0.258
llers+0.256
 necks+0.251
 mess+0.248
Neg logit contributions
 help-0.155
 reward-0.142
less-0.135
ar-0.132
iest-0.123
Token They
Token ID1119
Feature activation-4.24e-01

Pos logit contributions
 lair+0.027
cks+0.026
 pads+0.025
 mess+0.025
 floor+0.025
Neg logit contributions
 reward-0.031
 help-0.030
 plunge-0.030
ar-0.030
less-0.028
Token are
Token ID389
Feature activation+2.20e-01

Pos logit contributions
 help+0.027
 reward+0.019
 plunge+0.017
Sure+0.016
f+0.016
Neg logit contributions
 lair-0.049
 pads-0.046
llers-0.045
cks-0.045
opl-0.044
Token They
Token ID1119
Feature activation-4.61e-01

Pos logit contributions
 lair+0.033
 floor+0.032
cks+0.032
 pads+0.031
 mess+0.031
Neg logit contributions
 help-0.040
 reward-0.040
 plunge-0.040
ar-0.039
f-0.038
Token were
Token ID547
Feature activation+2.07e-01

Pos logit contributions
 help+0.032
 reward+0.029
Sure+0.028
ar+0.027
 plunge+0.027
Neg logit contributions
 lair-0.043
cks-0.041
 mess-0.039
opl-0.039
llers-0.038
Token different
Token ID1180
Feature activation+3.62e-01

Pos logit contributions
 lair+0.018
opl+0.017
urious+0.017
llers+0.016
 pads+0.016
Neg logit contributions
 help-0.015
 reward-0.012
 plunge-0.012
less-0.012
 Bub-0.011
Token colors
Token ID7577
Feature activation+1.17e+00

Pos logit contributions
cks+0.028
 lair+0.028
 mess+0.026
 pads+0.026
 floor+0.025
Neg logit contributions
 help-0.028
 reward-0.027
ar-0.026
 plunge-0.026
less-0.026
Token and
Token ID290
Feature activation+3.74e-01

Pos logit contributions
 lair+0.114
cks+0.108
 mess+0.100
 necks+0.100
llers+0.100
Neg logit contributions
 help-0.070
less-0.068
 reward-0.064
iest-0.063
ar-0.061
Token sizes
Token ID10620
Feature activation+4.94e+00

Pos logit contributions
 lair+0.020
cks+0.018
 mess+0.016
llers+0.015
 pads+0.015
Neg logit contributions
 help-0.041
 reward-0.039
ar-0.037
 plunge-0.037
less-0.037
Token.
Token ID13
Feature activation+4.69e-01

Pos logit contributions
 lair+0.477
cks+0.460
llers+0.433
 mess+0.433
 pads+0.424
Neg logit contributions
 help-0.313
 reward-0.292
ar-0.276
less-0.275
 plunge-0.268
Token Lily
Token ID20037
Feature activation+7.98e-01

Pos logit contributions
 lair+0.030
 floor+0.030
cks+0.029
 pads+0.029
 mess+0.028
Neg logit contributions
 reward-0.041
 help-0.041
 plunge-0.041
ar-0.040
bye-0.039
Token picked
Token ID6497
Feature activation-6.15e-01

Pos logit contributions
cks+0.093
 lair+0.090
opl+0.089
ric+0.083
 fault+0.080
Neg logit contributions
ed-0.035
 help-0.033
ar-0.031
ted-0.031
f-0.029
Token a
Token ID257
Feature activation-3.07e-01

Pos logit contributions
 reward+0.041
bye+0.040
ar+0.039
 help+0.039
 plunge+0.037
Neg logit contributions
cks-0.053
 pads-0.053
 neck-0.050
 mess-0.050
 goals-0.049
Token pink
Token ID11398
Feature activation+9.87e-01

Pos logit contributions
 help+0.025
 reward+0.025
ar+0.025
 plunge+0.024
Sure+0.024
Neg logit contributions
 lair-0.022
cks-0.022
 mess-0.020
 pads-0.020
 floor-0.020
Token cars
Token ID5006
Feature activation-4.20e-01

Pos logit contributions
 lair+0.076
cks+0.075
 pads+0.069
 mess+0.069
llers+0.068
Neg logit contributions
 help-0.085
 reward-0.081
ar-0.078
 plunge-0.077
ted-0.074
Token of
Token ID286
Feature activation-9.00e-01

Pos logit contributions
 reward+0.027
 help+0.026
 plunge+0.026
ar+0.026
Sure+0.025
Neg logit contributions
 lair-0.037
cks-0.037
 mess-0.036
 pads-0.036
 floor-0.035
Token different
Token ID1180
Feature activation+3.14e-01

Pos logit contributions
ar+0.071
 reward+0.070
bye+0.069
 plunge+0.069
less+0.067
Neg logit contributions
cks-0.067
 mess-0.066
 lair-0.066
 floor-0.065
 sizes-0.064
Token colors
Token ID7577
Feature activation+1.08e+00

Pos logit contributions
 lair+0.019
cks+0.019
 mess+0.017
 pads+0.017
llers+0.016
Neg logit contributions
 help-0.029
 reward-0.029
ar-0.028
 plunge-0.028
bye-0.027
Token and
Token ID290
Feature activation+3.74e-01

Pos logit contributions
 lair+0.106
cks+0.099
 mess+0.096
 necks+0.093
 web+0.093
Neg logit contributions
 help-0.065
ar-0.058
iest-0.057
 reward-0.057
less-0.056
Token sizes
Token ID10620
Feature activation+4.92e+00

Pos logit contributions
 lair+0.021
cks+0.019
 mess+0.016
irling+0.016
llers+0.016
Neg logit contributions
 help-0.041
 reward-0.038
ar-0.035
 Bub-0.035
f-0.035
Token.
Token ID13
Feature activation+4.21e-01

Pos logit contributions
 lair+0.475
cks+0.452
 mess+0.439
llers+0.431
 necks+0.423
Neg logit contributions
 help-0.309
 reward-0.286
ar-0.279
f-0.262
less-0.261
Token He
Token ID679
Feature activation+2.21e-01

Pos logit contributions
 lair+0.031
cks+0.030
 mess+0.029
 pads+0.029
 floor+0.028
Neg logit contributions
 help-0.034
 reward-0.034
ar-0.033
 plunge-0.033
bye-0.032
Token likes
Token ID7832
Feature activation+1.17e+00

Pos logit contributions
 lair+0.016
cks+0.016
llers+0.015
ivering+0.015
cking+0.015
Neg logit contributions
 help-0.019
 reward-0.017
ar-0.016
 Bub-0.016
f-0.015
Token to
Token ID284
Feature activation+1.63e-01

Pos logit contributions
cks+0.121
 lair+0.120
 pads+0.113
 mess+0.113
 necks+0.111
Neg logit contributions
 help-0.062
 reward-0.055
ar-0.051
ted-0.050
 plunge-0.050
Token make
Token ID787
Feature activation+1.01e+00

Pos logit contributions
 lair+0.015
cks+0.014
 pads+0.014
 mess+0.014
 floor+0.013
Neg logit contributions
 help-0.011
 reward-0.011
ar-0.010
 plunge-0.010
Sure-0.010
Token cars
Token ID5006
Feature activation-3.93e-01

Pos logit contributions
 lair+0.078
cks+0.077
 mess+0.072
 pads+0.072
 floor+0.070
Neg logit contributions
 help-0.090
 reward-0.088
ar-0.085
 plunge-0.084
Sure-0.081
Token of
Token ID286
Feature activation-9.03e-01

Pos logit contributions
 plunge+0.026
 reward+0.025
ar+0.024
Sure+0.024
 help+0.023
Neg logit contributions
 pads-0.034
 mess-0.034
cks-0.034
 lair-0.033
 floor-0.032
Token different
Token ID1180
Feature activation+3.60e-01

Pos logit contributions
ar+0.070
 reward+0.069
 plunge+0.068
bye+0.068
less+0.065
Neg logit contributions
 mess-0.069
cks-0.069
 lair-0.068
 floor-0.066
 sizes-0.066
Token colors
Token ID7577
Feature activation+1.08e+00

Pos logit contributions
 lair+0.022
cks+0.021
 mess+0.020
 pads+0.019
 floor+0.019
Neg logit contributions
 help-0.034
 reward-0.033
ar-0.033
 plunge-0.032
bye-0.031
Token and
Token ID290
Feature activation+3.92e-01

Pos logit contributions
 lair+0.101
cks+0.095
 mess+0.090
 necks+0.088
 pads+0.087
Neg logit contributions
 help-0.070
 reward-0.065
ar-0.064
less-0.063
iest-0.061
Token sizes
Token ID10620
Feature activation+4.92e+00

Pos logit contributions
 lair+0.018
cks+0.016
 mess+0.013
llers+0.013
opl+0.012
Neg logit contributions
 help-0.047
 reward-0.044
ar-0.041
f-0.041
 plunge-0.040
Token.
Token ID13
Feature activation+5.16e-01

Pos logit contributions
 lair+0.455
cks+0.436
 mess+0.420
llers+0.408
 pads+0.407
Neg logit contributions
 help-0.323
 reward-0.308
ar-0.300
 plunge-0.285
less-0.283
Token They
Token ID1119
Feature activation-4.19e-01

Pos logit contributions
 lair+0.042
cks+0.041
 mess+0.040
 pads+0.039
 floor+0.039
Neg logit contributions
 help-0.038
 reward-0.037
ar-0.036
 plunge-0.036
bye-0.035
Token like
Token ID588
Feature activation-3.24e-01

Pos logit contributions
 help+0.026
 reward+0.022
f+0.019
Sure+0.019
ar+0.018
Neg logit contributions
 lair-0.045
cks-0.044
llers-0.042
 pads-0.041
opl-0.040
Token to
Token ID284
Feature activation+2.49e-01

Pos logit contributions
 help+0.018
 reward+0.016
less+0.014
 plunge+0.014
Sure+0.014
Neg logit contributions
cks-0.034
 lair-0.034
 pads-0.032
 mess-0.031
 necks-0.031
Token drive
Token ID3708
Feature activation-5.88e-01

Pos logit contributions
 lair+0.023
cks+0.022
 pads+0.021
 mess+0.021
 floor+0.021
Neg logit contributions
 help-0.016
 reward-0.016
ar-0.015
 plunge-0.015
bye-0.014
Token with
Token ID351
Feature activation+6.92e-01

Pos logit contributions
 lair+0.254
cks+0.246
llers+0.239
 pads+0.237
 necks+0.236
Neg logit contributions
 help-0.163
 reward-0.151
ar-0.143
less-0.142
 plunge-0.133
Token the
Token ID262
Feature activation+2.98e-01

Pos logit contributions
 lair+0.061
cks+0.061
 mess+0.059
 pads+0.058
 floor+0.058
Neg logit contributions
 reward-0.045
 help-0.044
ar-0.044
 plunge-0.043
bye-0.041
Token different
Token ID1180
Feature activation+3.96e-01

Pos logit contributions
cks+0.025
 lair+0.025
llers+0.023
 pads+0.023
 mess+0.023
Neg logit contributions
 help-0.022
 reward-0.021
 plunge-0.021
ar-0.020
less-0.019
Token colors
Token ID7577
Feature activation+1.15e+00

Pos logit contributions
 lair+0.028
cks+0.028
 pads+0.025
llers+0.025
 mess+0.025
Neg logit contributions
 help-0.034
 reward-0.033
 plunge-0.032
ar-0.031
bye-0.031
Token and
Token ID290
Feature activation+2.40e-01

Pos logit contributions
 lair+0.116
cks+0.108
 mess+0.104
llers+0.103
 necks+0.103
Neg logit contributions
 help-0.067
 reward-0.064
less-0.061
ar-0.060
 plunge-0.058
Token sizes
Token ID10620
Feature activation+4.92e+00

Pos logit contributions
 lair+0.015
cks+0.013
opl+0.013
llers+0.012
 pads+0.012
Neg logit contributions
 help-0.024
 reward-0.023
 plunge-0.021
 souven-0.021
less-0.020
Token.
Token ID13
Feature activation+1.75e-01

Pos logit contributions
 lair+0.473
cks+0.449
 mess+0.429
llers+0.428
 pads+0.426
Neg logit contributions
 help-0.309
 reward-0.295
ar-0.279
less-0.276
 plunge-0.271
Token They
Token ID1119
Feature activation-5.93e-01

Pos logit contributions
 lair+0.012
cks+0.012
 pads+0.012
 mess+0.011
 floor+0.011
Neg logit contributions
 reward-0.015
 help-0.015
 plunge-0.014
ar-0.014
bye-0.014
Token are
Token ID389
Feature activation+2.11e-01

Pos logit contributions
 help+0.038
 reward+0.031
f+0.027
 plunge+0.025
Sure+0.024
Neg logit contributions
 lair-0.066
llers-0.062
cks-0.062
 pads-0.061
opl-0.059
Token very
Token ID845
Feature activation+8.75e-02

Pos logit contributions
 lair+0.021
opl+0.020
 pads+0.020
cks+0.019
urious+0.019
Neg logit contributions
 help-0.014
 reward-0.011
less-0.010
 Bub-0.009
 souven-0.009
Token proud
Token ID6613
Feature activation-1.43e-01

Pos logit contributions
 lair+0.007
cks+0.007
 pads+0.007
 necks+0.006
 mess+0.006
Neg logit contributions
 help-0.007
 reward-0.007
ar-0.006
less-0.006
 plunge-0.006
Token were
Token ID547
Feature activation+9.30e-02

Pos logit contributions
 help+0.026
 reward+0.022
 plunge+0.022
f+0.021
ar+0.021
Neg logit contributions
 lair-0.055
cks-0.051
 pads-0.049
opl-0.049
 mess-0.049
Token all
Token ID477
Feature activation-2.68e-01

Pos logit contributions
 pads+0.008
 lair+0.008
cks+0.007
 necks+0.007
opl+0.007
Neg logit contributions
 help-0.006
 plunge-0.005
less-0.005
 reward-0.005
 Bub-0.005
Token different
Token ID1180
Feature activation+2.63e-01

Pos logit contributions
 help+0.020
less+0.018
 plunge+0.017
 reward+0.017
iest+0.016
Neg logit contributions
 lair-0.022
 necks-0.021
 pads-0.021
cks-0.021
llers-0.020
Token colors
Token ID7577
Feature activation+1.09e+00

Pos logit contributions
 lair+0.017
cks+0.017
 pads+0.016
 mess+0.016
 floor+0.015
Neg logit contributions
 help-0.023
 reward-0.023
 plunge-0.022
ar-0.022
less-0.022
Token and
Token ID290
Feature activation+2.23e-01

Pos logit contributions
 lair+0.109
cks+0.104
 necks+0.098
 mess+0.097
llers+0.095
Neg logit contributions
 help-0.063
less-0.061
 reward-0.058
iest-0.057
 plunge-0.056
Token sizes
Token ID10620
Feature activation+4.91e+00

Pos logit contributions
 lair+0.014
cks+0.013
opl+0.011
 mess+0.011
 pads+0.011
Neg logit contributions
 help-0.022
 reward-0.021
 plunge-0.020
less-0.020
ar-0.019
Token.
Token ID13
Feature activation+3.33e-01

Pos logit contributions
 lair+0.480
cks+0.459
 mess+0.433
llers+0.430
 necks+0.427
Neg logit contributions
 help-0.307
 reward-0.287
less-0.273
ar-0.272
 plunge-0.268
Token She
Token ID1375
Feature activation+9.32e-01

Pos logit contributions
cks+0.027
 lair+0.027
 mess+0.025
 floor+0.024
 necks+0.024
Neg logit contributions
 help-0.025
 reward-0.024
Sure-0.024
ar-0.024
 plunge-0.023
Token wondered
Token ID14028
Feature activation+2.20e+00

Pos logit contributions
cks+0.095
 lair+0.092
opl+0.090
 fault+0.083
 mess+0.081
Neg logit contributions
 help-0.054
 reward-0.051
Sure-0.050
ar-0.050
ted-0.048
Token how
Token ID703
Feature activation+4.11e-01

Pos logit contributions
 lair+0.187
cks+0.185
 mess+0.179
 pads+0.176
 floor+0.174
Neg logit contributions
 reward-0.149
 help-0.149
ar-0.145
 plunge-0.142
bye-0.138
Token many
Token ID867
Feature activation+9.85e-01

Pos logit contributions
 lair+0.038
cks+0.038
 mess+0.037
 pads+0.036
 floor+0.036
Neg logit contributions
 reward-0.024
 help-0.024
ar-0.024
 plunge-0.023
bye-0.022
Token cars
Token ID5006
Feature activation-4.30e-01

Pos logit contributions
 lair+0.081
cks+0.078
 pads+0.072
llers+0.071
 mess+0.070
Neg logit contributions
 help-0.086
 reward-0.081
ar-0.077
 plunge-0.075
Sure-0.074
Token of
Token ID286
Feature activation-9.28e-01

Pos logit contributions
 reward+0.026
 plunge+0.026
 help+0.025
ar+0.025
Sure+0.024
Neg logit contributions
cks-0.039
 lair-0.039
 mess-0.038
 pads-0.038
 floor-0.037
Token different
Token ID1180
Feature activation+3.22e-01

Pos logit contributions
 reward+0.060
ar+0.060
bye+0.058
 plunge+0.058
 help+0.056
Neg logit contributions
cks-0.082
 mess-0.081
 lair-0.081
 floor-0.080
 sizes-0.078
Token colors
Token ID7577
Feature activation+1.07e+00

Pos logit contributions
 lair+0.020
cks+0.019
 mess+0.017
 pads+0.017
llers+0.016
Neg logit contributions
 help-0.030
 reward-0.029
ar-0.029
 plunge-0.028
bye-0.028
Token and
Token ID290
Feature activation+3.56e-01

Pos logit contributions
 lair+0.107
cks+0.099
 mess+0.095
 necks+0.095
 web+0.094
Neg logit contributions
 help-0.063
iest-0.057
 reward-0.057
ar-0.057
less-0.055
Token sizes
Token ID10620
Feature activation+4.90e+00

Pos logit contributions
 lair+0.017
cks+0.015
irling+0.012
iver+0.012
ivering+0.012
Neg logit contributions
 help-0.042
 reward-0.039
 Bub-0.036
ar-0.036
f-0.035
Token.
Token ID13
Feature activation+3.94e-01

Pos logit contributions
 lair+0.476
cks+0.452
 mess+0.438
llers+0.431
 necks+0.424
Neg logit contributions
 help-0.305
 reward-0.283
ar-0.275
f-0.260
less-0.259
Token He
Token ID679
Feature activation+2.40e-01

Pos logit contributions
 lair+0.032
cks+0.032
 mess+0.031
 pads+0.030
 floor+0.029
Neg logit contributions
 help-0.029
 reward-0.028
ar-0.027
Sure-0.027
 plunge-0.026
Token likes
Token ID7832
Feature activation+1.18e+00

Pos logit contributions
 lair+0.018
cks+0.018
llers+0.017
ivering+0.017
iver+0.016
Neg logit contributions
 help-0.020
 reward-0.017
ar-0.017
 Bub-0.016
f-0.016
Token to
Token ID284
Feature activation+1.92e-01

Pos logit contributions
 lair+0.121
cks+0.121
 pads+0.114
 mess+0.114
 necks+0.112
Neg logit contributions
 help-0.064
 reward-0.056
ted-0.052
ar-0.052
 Bub-0.052
Token make
Token ID787
Feature activation+1.03e+00

Pos logit contributions
 lair+0.017
cks+0.017
 pads+0.016
 mess+0.016
 floor+0.016
Neg logit contributions
 help-0.013
 reward-0.013
ar-0.012
 plunge-0.012
Sure-0.011
Token the
Token ID262
Feature activation+2.43e-01

Pos logit contributions
 lair+0.158
cks+0.155
 mess+0.151
 pads+0.150
 floor+0.147
Neg logit contributions
 help-0.096
 reward-0.094
ar-0.090
 plunge-0.088
bye-0.084
Token blocks
Token ID7021
Feature activation+1.06e+00

Pos logit contributions
cks+0.018
 lair+0.016
llers+0.016
 pads+0.016
 necks+0.015
Neg logit contributions
 help-0.021
 reward-0.021
 plunge-0.019
ar-0.018
f-0.018
Token,
Token ID11
Feature activation-1.95e-01

Pos logit contributions
 lair+0.098
cks+0.097
 mess+0.094
 pads+0.092
 floor+0.091
Neg logit contributions
 help-0.065
 reward-0.064
ar-0.062
 plunge-0.060
bye-0.058
Token compare
Token ID8996
Feature activation+7.98e-01

Pos logit contributions
 help+0.015
 reward+0.014
ar+0.014
 plunge+0.014
Sure+0.013
Neg logit contributions
 lair-0.016
cks-0.016
 mess-0.015
 pads-0.015
 floor-0.014
Token the
Token ID262
Feature activation+2.54e-01

Pos logit contributions
cks+0.072
 lair+0.072
 mess+0.070
 pads+0.069
 floor+0.068
Neg logit contributions
 reward-0.050
 help-0.050
ar-0.048
 plunge-0.047
bye-0.046
Token sizes
Token ID10620
Feature activation+4.90e+00

Pos logit contributions
cks+0.019
 lair+0.018
llers+0.017
 pads+0.017
 necks+0.016
Neg logit contributions
 help-0.022
 reward-0.021
 plunge-0.019
f-0.018
ar-0.018
Token,
Token ID11
Feature activation-2.24e-01

Pos logit contributions
 lair+0.488
cks+0.481
 mess+0.467
 pads+0.461
 floor+0.455
Neg logit contributions
 help-0.266
 reward-0.263
ar-0.252
 plunge-0.247
bye-0.236
Token and
Token ID290
Feature activation+2.01e-01

Pos logit contributions
 help+0.018
 reward+0.018
ar+0.017
 plunge+0.017
bye+0.016
Neg logit contributions
 lair-0.017
cks-0.016
 mess-0.015
 pads-0.015
 floor-0.015
Token sort
Token ID3297
Feature activation+9.20e-01

Pos logit contributions
 lair+0.015
cks+0.014
 pads+0.013
 mess+0.013
opl+0.013
Neg logit contributions
 help-0.017
 reward-0.016
 plunge-0.015
ar-0.014
f-0.014
Token the
Token ID262
Feature activation+2.70e-01

Pos logit contributions
cks+0.079
 mess+0.078
 sizes+0.076
 floor+0.076
 lair+0.075
Neg logit contributions
 reward-0.061
ar-0.060
 plunge-0.059
 help-0.058
Sure-0.057
Token colors
Token ID7577
Feature activation+1.17e+00

Pos logit contributions
cks+0.019
 lair+0.019
llers+0.018
 pads+0.017
irling+0.017
Neg logit contributions
 help-0.023
 reward-0.022
 plunge-0.021
f-0.020
ar-0.020
Token were
Token ID547
Feature activation+1.82e-01

Pos logit contributions
 help+0.042
 reward+0.040
 plunge+0.038
ar+0.038
ted+0.038
Neg logit contributions
 lair-0.053
cks-0.050
 pads-0.048
 mess-0.048
opl-0.048
Token of
Token ID286
Feature activation-8.54e-01

Pos logit contributions
 lair+0.014
opl+0.014
 pads+0.013
llers+0.013
urious+0.012
Neg logit contributions
 help-0.014
 reward-0.012
 plunge-0.012
less-0.012
 souven-0.011
Token different
Token ID1180
Feature activation+3.25e-01

Pos logit contributions
ar+0.064
 reward+0.064
 help+0.063
 plunge+0.061
bye+0.061
Neg logit contributions
 lair-0.067
cks-0.066
 mess-0.065
 floor-0.062
 pads-0.062
Token colors
Token ID7577
Feature activation+1.12e+00

Pos logit contributions
 lair+0.020
cks+0.020
 pads+0.019
 mess+0.018
llers+0.018
Neg logit contributions
 help-0.030
 reward-0.029
 plunge-0.028
ar-0.028
less-0.028
Token and
Token ID290
Feature activation+2.09e-01

Pos logit contributions
 lair+0.109
cks+0.104
 necks+0.098
 mess+0.097
llers+0.096
Neg logit contributions
 help-0.065
less-0.063
 reward-0.060
iest-0.059
ar-0.059
Token sizes
Token ID10620
Feature activation+4.90e+00

Pos logit contributions
 lair+0.013
cks+0.012
 mess+0.011
 pads+0.011
 floor+0.011
Neg logit contributions
 help-0.020
 reward-0.019
ar-0.018
 plunge-0.018
bye-0.018
Token.
Token ID13
Feature activation+2.25e-01

Pos logit contributions
 lair+0.493
cks+0.474
llers+0.451
 mess+0.447
 necks+0.443
Neg logit contributions
 help-0.291
 reward-0.270
less-0.254
ar-0.253
 plunge-0.244
Token They
Token ID1119
Feature activation-5.66e-01

Pos logit contributions
 lair+0.017
 pads+0.017
cks+0.017
 floor+0.017
 mess+0.016
Neg logit contributions
 reward-0.017
 help-0.017
 plunge-0.017
bye-0.016
ar-0.016
Token filled
Token ID5901
Feature activation+2.49e-01

Pos logit contributions
 help+0.040
 reward+0.038
ar+0.036
 plunge+0.035
f+0.035
Neg logit contributions
 lair-0.050
cks-0.048
opl-0.046
 pads-0.045
 mess-0.045
Token the
Token ID262
Feature activation+3.37e-01

Pos logit contributions
cks+0.023
 mess+0.022
 lair+0.022
 floor+0.021
 necks+0.021
Neg logit contributions
 reward-0.016
 help-0.016
ar-0.015
 plunge-0.015
bye-0.015
Token air
Token ID1633
Feature activation-3.43e-01

Pos logit contributions
 lair+0.022
cks+0.022
 pads+0.020
 mess+0.020
llers+0.020
Neg logit contributions
 help-0.030
 reward-0.030
ar-0.029
 plunge-0.029
bye-0.028
Token they
Token ID484
Feature activation-5.68e-01

Pos logit contributions
ar+0.032
 reward+0.031
 plunge+0.031
 help+0.030
bye+0.029
Neg logit contributions
 lair-0.051
cks-0.051
 mess-0.051
 floor-0.050
 pads-0.049
Token're
Token ID821
Feature activation-9.93e-01

Pos logit contributions
 help+0.036
 reward+0.034
ar+0.032
 plunge+0.032
f+0.031
Neg logit contributions
 lair-0.055
cks-0.053
 pads-0.051
 mess-0.050
 floor-0.049
Token different
Token ID1180
Feature activation+3.24e-01

Pos logit contributions
 help+0.075
 reward+0.070
 plunge+0.062
less+0.061
ar+0.061
Neg logit contributions
 lair-0.084
 pads-0.080
cks-0.079
 necks-0.077
llers-0.074
Token colors
Token ID7577
Feature activation+1.14e+00

Pos logit contributions
cks+0.028
 lair+0.028
 pads+0.026
 mess+0.026
 necks+0.026
Neg logit contributions
 help-0.022
 reward-0.021
ar-0.020
 plunge-0.020
less-0.020
Token and
Token ID290
Feature activation+1.13e-01

Pos logit contributions
 lair+0.110
cks+0.106
 necks+0.100
 mess+0.098
 frown+0.098
Neg logit contributions
 help-0.071
 reward-0.066
less-0.066
iest-0.064
ar-0.062
Token sizes
Token ID10620
Feature activation+4.89e+00

Pos logit contributions
 lair+0.008
cks+0.007
 mess+0.007
 pads+0.007
 floor+0.007
Neg logit contributions
 help-0.010
 reward-0.010
ar-0.009
 plunge-0.009
bye-0.009
Token."
Token ID526
Feature activation+4.72e-01

Pos logit contributions
 lair+0.431
cks+0.417
 mess+0.394
llers+0.387
 pads+0.387
Neg logit contributions
 help-0.343
 reward-0.330
ar-0.314
 plunge-0.308
less-0.305
Token
Token ID198
Feature activation+1.93e-01

Pos logit contributions
 lair+0.045
cks+0.044
 mess+0.042
 floor+0.041
 pads+0.041
Neg logit contributions
 help-0.029
 reward-0.028
ar-0.027
Sure-0.027
 plunge-0.026
Token
Token ID198
Feature activation+2.04e-01

Pos logit contributions
 lair+0.017
cks+0.017
 mess+0.016
 pads+0.016
 floor+0.016
Neg logit contributions
 help-0.013
 reward-0.013
ar-0.012
 plunge-0.012
Sure-0.011
TokenMax
Token ID11518
Feature activation-1.84e-01

Pos logit contributions
 lair+0.023
cks+0.022
 mess+0.021
 sizes+0.021
 fault+0.021
Neg logit contributions
Sure-0.010
 help-0.009
 reward-0.009
ar-0.009
f-0.008
Token started
Token ID2067
Feature activation-3.50e-01

Pos logit contributions
 help+0.012
 reward+0.012
ar+0.012
Sure+0.011
bye+0.011
Neg logit contributions
cks-0.017
 lair-0.017
 mess-0.015
 pads-0.015
opl-0.015
Token fish
Token ID5916
Feature activation-6.84e-01

Pos logit contributions
 lair+0.100
cks+0.096
 pads+0.094
llers+0.092
 mess+0.091
Neg logit contributions
 help-0.065
 reward-0.063
 plunge-0.061
ar-0.059
ted-0.055
Token of
Token ID286
Feature activation-8.67e-01

Pos logit contributions
 help+0.044
 reward+0.041
ar+0.041
 plunge+0.040
less+0.038
Neg logit contributions
 lair-0.062
llers-0.060
cks-0.059
 mess-0.059
 floor-0.058
Token different
Token ID1180
Feature activation+3.20e-01

Pos logit contributions
 reward+0.053
 help+0.053
ar+0.053
 plunge+0.051
bye+0.050
Neg logit contributions
 lair-0.079
cks-0.079
 mess-0.077
 floor-0.075
 pads-0.074
Token colors
Token ID7577
Feature activation+1.13e+00

Pos logit contributions
 lair+0.020
cks+0.019
 pads+0.018
 mess+0.018
llers+0.017
Neg logit contributions
 help-0.029
 reward-0.029
 plunge-0.029
ar-0.028
ted-0.027
Token and
Token ID290
Feature activation+3.60e-01

Pos logit contributions
 lair+0.120
cks+0.106
llers+0.104
 mess+0.104
 necks+0.104
Neg logit contributions
 help-0.061
iest-0.055
 reward-0.055
less-0.054
 plunge-0.052
Token sizes
Token ID10620
Feature activation+4.89e+00

Pos logit contributions
 lair+0.023
opl+0.019
cks+0.019
ivering+0.019
iver+0.018
Neg logit contributions
 help-0.037
 reward-0.034
 plunge-0.032
 Bub-0.031
ar-0.030
Token.
Token ID13
Feature activation+8.27e-02

Pos logit contributions
 lair+0.525
cks+0.477
llers+0.476
 necks+0.460
 pads+0.458
Neg logit contributions
 help-0.269
 reward-0.243
less-0.228
 plunge-0.221
ar-0.218
Token They
Token ID1119
Feature activation-5.16e-01

Pos logit contributions
 lair+0.006
cks+0.006
 mess+0.006
 pads+0.006
 floor+0.006
Neg logit contributions
 help-0.006
 reward-0.006
ar-0.006
 plunge-0.006
Sure-0.006
Token swim
Token ID9422
Feature activation-2.78e-01

Pos logit contributions
 help+0.036
 plunge+0.028
 reward+0.027
Sure+0.025
f+0.024
Neg logit contributions
 lair-0.057
llers-0.051
opl-0.051
cks-0.051
 pads-0.050
Token and
Token ID290
Feature activation+3.09e-01

Pos logit contributions
 reward+0.022
 help+0.022
 plunge+0.021
ar+0.021
bye+0.020
Neg logit contributions
cks-0.020
 lair-0.020
 pads-0.020
 mess-0.019
 necks-0.019
Token splash
Token ID22870
Feature activation+7.26e-01

Pos logit contributions
 lair+0.029
cks+0.027
llers+0.027
 mess+0.026
opl+0.026
Neg logit contributions
 help-0.022
 reward-0.020
 plunge-0.019
ar-0.018
f-0.018
Token truck
Token ID7779
Feature activation-1.27e+00

Pos logit contributions
cks+0.013
 lair+0.013
llers+0.012
 mess+0.012
 pads+0.012
Neg logit contributions
 help-0.020
 reward-0.019
ar-0.019
 plunge-0.018
bye-0.018
Token.
Token ID13
Feature activation+3.68e-01

Pos logit contributions
 help+0.082
 reward+0.078
ar+0.078
bye+0.075
 plunge+0.074
Neg logit contributions
 lair-0.113
cks-0.112
 mess-0.109
 pads-0.104
llers-0.104
Token They
Token ID1119
Feature activation-3.49e-01

Pos logit contributions
 lair+0.031
cks+0.031
 mess+0.030
 pads+0.029
 floor+0.028
Neg logit contributions
 help-0.025
 reward-0.025
ar-0.025
 plunge-0.024
Sure-0.024
Token were
Token ID547
Feature activation+3.37e-01

Pos logit contributions
 help+0.025
 reward+0.023
ar+0.022
 plunge+0.021
f+0.021
Neg logit contributions
 lair-0.031
cks-0.030
 mess-0.029
 pads-0.029
llers-0.029
Token different
Token ID1180
Feature activation+4.87e-01

Pos logit contributions
 lair+0.029
cks+0.027
 pads+0.027
 mess+0.027
 floor+0.027
Neg logit contributions
 help-0.024
 reward-0.022
 plunge-0.021
ar-0.021
bye-0.020
Token sizes
Token ID10620
Feature activation+4.88e+00

Pos logit contributions
cks+0.039
 lair+0.038
 pads+0.037
 mess+0.036
 floor+0.035
Neg logit contributions
 help-0.037
 reward-0.036
ar-0.034
less-0.033
 plunge-0.033
Token and
Token ID290
Feature activation+3.90e-01

Pos logit contributions
 lair+0.482
cks+0.473
 mess+0.445
llers+0.439
 necks+0.436
Neg logit contributions
 help-0.291
 reward-0.272
ar-0.263
less-0.259
 plunge-0.248
Token colors
Token ID7577
Feature activation+1.24e+00

Pos logit contributions
 lair+0.011
cks+0.010
 mess+0.009
 pads+0.009
 floor+0.008
Neg logit contributions
 help-0.049
 reward-0.049
ar-0.048
 plunge-0.048
bye-0.047
Token.
Token ID13
Feature activation+5.86e-01

Pos logit contributions
 lair+0.116
cks+0.112
 mess+0.109
 necks+0.106
 pads+0.105
Neg logit contributions
 help-0.077
 reward-0.073
ar-0.072
less-0.070
 plunge-0.069
Token Some
Token ID2773
Feature activation-1.13e+00

Pos logit contributions
 lair+0.045
cks+0.044
 pads+0.043
 mess+0.042
 floor+0.042
Neg logit contributions
 reward-0.045
 help-0.045
 plunge-0.044
ar-0.043
bye-0.041
Token had
Token ID550
Feature activation+5.62e-01

Pos logit contributions
 reward+0.071
 help+0.069
ar+0.069
 plunge+0.069
Sure+0.066
Neg logit contributions
cks-0.100
 lair-0.100
 mess-0.098
 pads-0.097
 floor-0.096
Token three
Token ID1115
Feature activation-1.11e+00

Pos logit contributions
 lair+0.021
cks+0.021
 pads+0.020
 mess+0.020
 necks+0.019
Neg logit contributions
 help-0.021
 reward-0.021
ar-0.019
 plunge-0.019
less-0.019
Token shapes
Token ID15268
Feature activation+2.56e+00

Pos logit contributions
 help+0.099
ar+0.096
 reward+0.096
f+0.090
 plunge+0.089
Neg logit contributions
 lair-0.076
cks-0.072
 pads-0.068
llers-0.066
 floor-0.063
Token,
Token ID11
Feature activation-1.68e-01

Pos logit contributions
 lair+0.273
cks+0.271
 necks+0.256
 mess+0.256
 pads+0.256
Neg logit contributions
 help-0.131
 reward-0.122
less-0.114
ar-0.114
 plunge-0.113
Token all
Token ID477
Feature activation-2.84e-01

Pos logit contributions
 help+0.013
 reward+0.012
 plunge+0.012
ar+0.012
ted+0.011
Neg logit contributions
 lair-0.014
cks-0.013
 mess-0.012
 pads-0.012
llers-0.012
Token different
Token ID1180
Feature activation+2.90e-01

Pos logit contributions
 help+0.013
 reward+0.011
less+0.011
 plunge+0.010
iest+0.010
Neg logit contributions
 lair-0.032
cks-0.031
 necks-0.031
 pads-0.030
llers-0.030
Token sizes
Token ID10620
Feature activation+4.88e+00

Pos logit contributions
 lair+0.017
cks+0.017
 mess+0.016
 pads+0.016
 floor+0.015
Neg logit contributions
 help-0.027
 reward-0.027
ar-0.026
 plunge-0.026
less-0.025
Token.
Token ID13
Feature activation+2.29e-01

Pos logit contributions
 lair+0.474
cks+0.464
 mess+0.444
 pads+0.436
 floor+0.432
Neg logit contributions
 help-0.289
 reward-0.280
ar-0.269
 plunge-0.263
less-0.257
Token
Token ID198
Feature activation+2.71e-01

Pos logit contributions
 lair+0.020
cks+0.019
 mess+0.019
 pads+0.019
 floor+0.018
Neg logit contributions
 help-0.015
 reward-0.015
 plunge-0.015
ar-0.015
bye-0.014
Token
Token ID198
Feature activation+3.17e-01

Pos logit contributions
 lair+0.024
cks+0.024
 mess+0.023
 pads+0.023
 floor+0.022
Neg logit contributions
 help-0.018
 reward-0.017
ar-0.017
 plunge-0.017
Sure-0.016
TokenThe
Token ID464
Feature activation-5.61e-01

Pos logit contributions
 lair+0.026
cks+0.026
 mess+0.025
 pads+0.024
 floor+0.024
Neg logit contributions
 help-0.023
 reward-0.023
ar-0.022
 plunge-0.022
Sure-0.022
Token shape
Token ID5485
Feature activation+1.85e+00

Pos logit contributions
ar+0.042
ted+0.040
 plunge+0.040
bye+0.038
less+0.038
Neg logit contributions
 mess-0.045
 floor-0.045
 lair-0.044
 frown-0.042
 neck-0.041
Token prints
Token ID20842
Feature activation-2.79e-01

Pos logit contributions
 reward+0.027
 help+0.027
ar+0.027
 plunge+0.026
bye+0.026
Neg logit contributions
cks-0.022
 lair-0.022
 mess-0.021
 pads-0.021
 floor-0.020
Token in
Token ID287
Feature activation-5.59e-02

Pos logit contributions
 reward+0.019
 plunge+0.019
 help+0.019
Sure+0.018
ar+0.018
Neg logit contributions
 lair-0.023
 pads-0.023
cks-0.022
 mess-0.022
 floor-0.022
Token different
Token ID1180
Feature activation+2.97e-01

Pos logit contributions
ar+0.003
 reward+0.003
 plunge+0.003
 help+0.003
bye+0.003
Neg logit contributions
 lair-0.005
cks-0.005
 mess-0.005
 floor-0.005
 necks-0.005
Token colors
Token ID7577
Feature activation+1.11e+00

Pos logit contributions
 lair+0.016
cks+0.016
 mess+0.015
 pads+0.015
 floor+0.015
Neg logit contributions
 help-0.029
 reward-0.029
ar-0.028
 plunge-0.028
bye-0.027
Token and
Token ID290
Feature activation+1.84e-01

Pos logit contributions
 lair+0.094
cks+0.092
 mess+0.087
 pads+0.084
 necks+0.084
Neg logit contributions
 help-0.078
 reward-0.077
ar-0.075
less-0.073
 plunge-0.072
Token sizes
Token ID10620
Feature activation+4.88e+00

Pos logit contributions
 lair+0.010
cks+0.010
 mess+0.010
 pads+0.010
 floor+0.010
Neg logit contributions
 reward-0.018
ar-0.017
 help-0.017
 plunge-0.017
Sure-0.017
Token.
Token ID13
Feature activation+2.42e-01

Pos logit contributions
 lair+0.434
cks+0.425
 mess+0.407
 pads+0.400
 floor+0.395
Neg logit contributions
 help-0.325
 reward-0.318
ar-0.307
 plunge-0.301
less-0.294
Token
Token ID220
Feature activation-3.73e-01

Pos logit contributions
cks+0.018
 lair+0.018
 pads+0.018
 floor+0.017
 mess+0.017
Neg logit contributions
 reward-0.018
 help-0.018
ar-0.018
 plunge-0.018
bye-0.017
Token
Token ID198
Feature activation+3.07e-01

Pos logit contributions
 help+0.025
 reward+0.025
ar+0.024
 plunge+0.024
bye+0.023
Neg logit contributions
 lair-0.032
cks-0.032
 mess-0.031
 pads-0.030
 floor-0.030
Token
Token ID198
Feature activation+3.38e-01

Pos logit contributions
 lair+0.026
cks+0.026
 mess+0.025
 pads+0.024
 floor+0.024
Neg logit contributions
 reward-0.021
 help-0.021
ar-0.021
 plunge-0.020
bye-0.019
TokenThe
Token ID464
Feature activation-5.32e-01

Pos logit contributions
 lair+0.026
cks+0.026
 mess+0.025
 pads+0.025
 floor+0.024
Neg logit contributions
 help-0.026
 reward-0.025
ar-0.025
 plunge-0.024
bye-0.024
Token cars
Token ID5006
Feature activation-4.67e-01

Pos logit contributions
 lair+0.073
cks+0.072
 mess+0.068
 pads+0.067
 floor+0.066
Neg logit contributions
 help-0.091
 reward-0.089
ar-0.087
 plunge-0.085
Sure-0.082
Token of
Token ID286
Feature activation-9.45e-01

Pos logit contributions
 plunge+0.031
 reward+0.030
 help+0.028
ar+0.028
Sure+0.027
Neg logit contributions
cks-0.041
 mess-0.041
 pads-0.040
 lair-0.040
 floor-0.039
Token different
Token ID1180
Feature activation+3.33e-01

Pos logit contributions
ar+0.070
 reward+0.070
 plunge+0.068
bye+0.068
 help+0.066
Neg logit contributions
cks-0.074
 lair-0.074
 mess-0.074
 floor-0.071
 pads-0.070
Token colors
Token ID7577
Feature activation+9.98e-01

Pos logit contributions
 lair+0.021
cks+0.021
 mess+0.019
 pads+0.019
 floor+0.018
Neg logit contributions
 help-0.031
 reward-0.030
ar-0.029
 plunge-0.029
bye-0.028
Token and
Token ID290
Feature activation+3.62e-01

Pos logit contributions
 lair+0.093
cks+0.088
 mess+0.084
 necks+0.082
llers+0.081
Neg logit contributions
 help-0.064
ar-0.059
 reward-0.059
less-0.058
iest-0.056
Token sizes
Token ID10620
Feature activation+4.88e+00

Pos logit contributions
 lair+0.016
cks+0.013
 mess+0.011
llers+0.011
opl+0.010
Neg logit contributions
 help-0.044
 reward-0.041
ar-0.039
f-0.039
 plunge-0.038
Token.
Token ID13
Feature activation+5.17e-01

Pos logit contributions
 lair+0.456
cks+0.434
 mess+0.419
llers+0.409
 pads+0.405
Neg logit contributions
 help-0.317
 reward-0.300
ar-0.293
 plunge-0.276
less-0.276
Token They
Token ID1119
Feature activation-4.23e-01

Pos logit contributions
 lair+0.042
cks+0.042
 mess+0.040
 pads+0.040
 floor+0.039
Neg logit contributions
 help-0.037
 reward-0.037
ar-0.036
 plunge-0.035
bye-0.034
Token make
Token ID787
Feature activation+1.06e+00

Pos logit contributions
 help+0.026
 reward+0.022
f+0.019
ar+0.019
 plunge+0.018
Neg logit contributions
 lair-0.046
cks-0.044
llers-0.043
 pads-0.041
opl-0.040
Token a
Token ID257
Feature activation-2.61e-01

Pos logit contributions
 lair+0.080
cks+0.077
 pads+0.075
 necks+0.072
 mess+0.071
Neg logit contributions
 help-0.084
 reward-0.079
 plunge-0.077
ar-0.076
f-0.074
Token line
Token ID1627
Feature activation+5.65e-01

Pos logit contributions
 help+0.019
 reward+0.019
ar+0.017
Sure+0.017
 plunge+0.017
Neg logit contributions
cks-0.022
 lair-0.021
 pads-0.020
llers-0.020
 necks-0.019
Token there
Token ID612
Feature activation-7.78e-01

Pos logit contributions
 lair+0.096
cks+0.095
 necks+0.091
 pads+0.091
 mess+0.090
Neg logit contributions
 help-0.049
 reward-0.047
ar-0.046
less-0.045
 plunge-0.045
Token was
Token ID373
Feature activation-8.46e-01

Pos logit contributions
 reward+0.052
 help+0.051
ar+0.050
 plunge+0.049
bye+0.048
Neg logit contributions
 lair-0.066
cks-0.066
 mess-0.064
 pads-0.063
 floor-0.063
Token a
Token ID257
Feature activation-9.23e-01

Pos logit contributions
ar+0.048
 reward+0.048
 help+0.045
 plunge+0.044
Sure+0.043
Neg logit contributions
cks-0.081
 lair-0.080
 pads-0.079
 mess-0.078
 floor-0.078
Token three
Token ID1115
Feature activation-1.57e+00

Pos logit contributions
 help+0.058
 reward+0.058
ar+0.058
 plunge+0.057
ted+0.055
Neg logit contributions
 lair-0.081
 mess-0.081
cks-0.079
 floor-0.078
 pads-0.076
Token-
Token ID12
Feature activation-3.91e+00

Pos logit contributions
 help+0.078
 reward+0.076
ar+0.074
 plunge+0.069
bye+0.068
Neg logit contributions
 lair-0.165
cks-0.163
 pads-0.158
 mess-0.156
llers-0.154
Tokenyear
Token ID1941
Feature activation-7.30e+00

Pos logit contributions
 reward+0.154
 help+0.149
ar+0.143
 plunge+0.141
bye+0.127
Neg logit contributions
 lair-0.449
cks-0.447
 mess-0.430
 pads-0.428
 floor-0.424
Token-
Token ID12
Feature activation-3.86e+00

Pos logit contributions
 help+0.277
 reward+0.272
ar+0.258
 plunge+0.249
bye+0.238
Neg logit contributions
 lair-0.843
cks-0.839
 mess-0.816
 pads-0.808
 floor-0.798
Tokenold
Token ID727
Feature activation-1.36e+00

Pos logit contributions
 reward+0.278
 help+0.275
ar+0.268
 plunge+0.266
bye+0.251
Neg logit contributions
 lair-0.317
cks-0.313
 mess-0.298
 pads-0.296
 floor-0.292
Token who
Token ID508
Feature activation-1.69e+00

Pos logit contributions
 reward+0.111
 plunge+0.111
ar+0.108
 help+0.108
less+0.103
Neg logit contributions
 lair-0.097
cks-0.092
 pads-0.091
 mess-0.091
 neck-0.090
Token found
Token ID1043
Feature activation-9.24e-01

Pos logit contributions
 help+0.134
 reward+0.133
ar+0.130
 plunge+0.127
bye+0.124
Neg logit contributions
 lair-0.126
cks-0.125
 mess-0.120
 pads-0.118
 floor-0.116
Token a
Token ID257
Feature activation-6.90e-01

Pos logit contributions
 reward+0.055
ar+0.054
 plunge+0.051
bye+0.051
 help+0.051
Neg logit contributions
 lair-0.089
cks-0.087
 floor-0.084
 mess-0.083
 neck-0.082
Token was
Token ID373
Feature activation-4.99e-01

Pos logit contributions
 lair+0.050
cks+0.047
 sizes+0.043
 web+0.043
 pads+0.043
Neg logit contributions
 help-0.039
bye-0.037
 reward-0.037
ted-0.036
ar-0.036
Token an
Token ID281
Feature activation+9.80e-02

Pos logit contributions
 help+0.031
 plunge+0.028
 reward+0.028
ar+0.027
ted+0.026
Neg logit contributions
 lair-0.045
 sizes-0.045
cks-0.044
 necks-0.043
llers-0.043
Token average
Token ID2811
Feature activation-1.35e+00

Pos logit contributions
 lair+0.007
cks+0.007
 pads+0.007
 mess+0.007
 floor+0.006
Neg logit contributions
 help-0.008
 reward-0.008
ar-0.008
 plunge-0.008
bye-0.008
Token three
Token ID1115
Feature activation-1.51e+00

Pos logit contributions
 plunge+0.087
ar+0.086
 reward+0.086
 help+0.086
ted+0.084
Neg logit contributions
 lair-0.115
 floor-0.111
 mess-0.111
 neck-0.111
 pads-0.107
Token-
Token ID12
Feature activation-3.93e+00

Pos logit contributions
 help+0.078
 reward+0.073
ar+0.073
f+0.068
bye+0.066
Neg logit contributions
 lair-0.159
cks-0.157
 pads-0.150
llers-0.148
 mess-0.147
Tokenyear
Token ID1941
Feature activation-7.18e+00

Pos logit contributions
 reward+0.154
 help+0.140
ar+0.136
 plunge+0.135
bye+0.116
Neg logit contributions
 lair-0.458
cks-0.455
 pads-0.439
 mess-0.435
 floor-0.431
Token-
Token ID12
Feature activation-3.89e+00

Pos logit contributions
 help+0.254
 reward+0.246
ar+0.234
 plunge+0.219
bye+0.218
Neg logit contributions
cks-0.852
 lair-0.850
 mess-0.823
 pads-0.818
 floor-0.802
Tokenold
Token ID727
Feature activation-1.39e+00

Pos logit contributions
 reward+0.295
 help+0.286
 plunge+0.281
ar+0.280
ted+0.263
Neg logit contributions
 lair-0.308
cks-0.301
 pads-0.288
 mess-0.286
 floor-0.281
Token boy
Token ID2933
Feature activation-1.84e-02

Pos logit contributions
 reward+0.109
 help+0.107
 plunge+0.107
ar+0.106
Sure+0.101
Neg logit contributions
 lair-0.104
cks-0.099
 pads-0.098
 mess-0.097
 floor-0.096
Token.
Token ID13
Feature activation-3.00e-01

Pos logit contributions
 help+0.001
 reward+0.001
ar+0.001
 plunge+0.001
bye+0.001
Neg logit contributions
 lair-0.002
cks-0.002
 mess-0.001
 floor-0.001
 pads-0.001
Token He
Token ID679
Feature activation-3.72e-01

Pos logit contributions
 help+0.021
bye+0.020
ar+0.020
 reward+0.020
Sure+0.019
Neg logit contributions
cks-0.025
 mess-0.025
 lair-0.025
 necks-0.023
 web-0.023
Token saw
Token ID2497
Feature activation-5.60e-01

Pos logit contributions
cks+0.003
 lair+0.003
llers+0.002
 sizes+0.002
 mess+0.002
Neg logit contributions
 help-0.002
 reward-0.002
ar-0.002
bye-0.002
 plunge-0.002
Token a
Token ID257
Feature activation-8.12e-01

Pos logit contributions
 reward+0.027
 help+0.027
ar+0.026
 plunge+0.026
bye+0.024
Neg logit contributions
 lair-0.058
cks-0.058
 mess-0.056
 pads-0.055
 floor-0.055
Token little
Token ID1310
Feature activation-6.75e-01

Pos logit contributions
 help+0.061
 reward+0.061
ar+0.059
 plunge+0.058
bye+0.056
Neg logit contributions
 lair-0.063
cks-0.063
 mess-0.060
 pads-0.059
 floor-0.058
Token 3
Token ID513
Feature activation-1.80e+00

Pos logit contributions
 help+0.042
 reward+0.042
ar+0.042
 plunge+0.040
bye+0.039
Neg logit contributions
 lair-0.061
 mess-0.059
cks-0.058
 floor-0.057
 pads-0.056
Token-
Token ID12
Feature activation-3.94e+00

Pos logit contributions
 help+0.085
ar+0.082
 reward+0.079
f+0.076
bye+0.072
Neg logit contributions
 lair-0.194
 pads-0.192
cks-0.190
llers-0.185
 sizes-0.180
Tokenyear
Token ID1941
Feature activation-7.14e+00

Pos logit contributions
 reward+0.165
ar+0.147
 plunge+0.141
 help+0.138
bye+0.121
Neg logit contributions
 lair-0.451
cks-0.449
 mess-0.418
 pads-0.414
 floor-0.409
Token-
Token ID12
Feature activation-3.87e+00

Pos logit contributions
 help+0.263
 reward+0.257
ar+0.245
 plunge+0.234
bye+0.225
Neg logit contributions
 lair-0.834
cks-0.830
 mess-0.807
 pads-0.802
 floor-0.791
Tokenold
Token ID727
Feature activation-1.29e+00

Pos logit contributions
 reward+0.305
 help+0.291
 plunge+0.289
ar+0.288
less+0.265
Neg logit contributions
 lair-0.299
cks-0.291
 mess-0.272
 pads-0.269
 floor-0.264
Token boy
Token ID2933
Feature activation+1.57e-01

Pos logit contributions
 reward+0.089
ar+0.088
 plunge+0.087
 help+0.085
Sure+0.083
Neg logit contributions
 lair-0.108
 mess-0.102
 pads-0.100
cks-0.100
 floor-0.098
Token.
Token ID13
Feature activation-1.14e-01

Pos logit contributions
 lair+0.011
cks+0.011
 mess+0.010
 pads+0.010
 floor+0.010
Neg logit contributions
 reward-0.013
 help-0.013
ar-0.013
 plunge-0.013
bye-0.013
Token He
Token ID679
Feature activation-8.77e-02

Pos logit contributions
 help+0.008
 reward+0.008
ar+0.007
 plunge+0.007
Sure+0.007
Neg logit contributions
cks-0.010
 lair-0.010
 mess-0.009
 pads-0.009
 floor-0.009
Tokenolly
Token ID5098
Feature activation-1.55e+00

Pos logit contributions
 help+0.073
 reward+0.072
 plunge+0.072
ar+0.071
ted+0.068
Neg logit contributions
 lair-0.091
cks-0.089
 mess-0.088
 floor-0.086
 frown-0.084
Token was
Token ID373
Feature activation-3.88e-01

Pos logit contributions
 help+0.112
 reward+0.110
ar+0.110
bye+0.105
ted+0.104
Neg logit contributions
 lair-0.130
cks-0.127
 mess-0.119
 pads-0.117
 sizes-0.116
Token a
Token ID257
Feature activation-6.48e-01

Pos logit contributions
 help+0.024
 reward+0.023
 plunge+0.022
ar+0.021
less+0.021
Neg logit contributions
 lair-0.036
cks-0.035
 sizes-0.034
 mess-0.033
 necks-0.033
Token three
Token ID1115
Feature activation-1.33e+00

Pos logit contributions
 help+0.047
 reward+0.046
ar+0.045
 plunge+0.045
bye+0.043
Neg logit contributions
 lair-0.053
cks-0.052
 mess-0.050
 pads-0.049
 floor-0.049
Token-
Token ID12
Feature activation-3.78e+00

Pos logit contributions
 help+0.065
 reward+0.061
ar+0.060
f+0.056
bye+0.054
Neg logit contributions
 lair-0.143
cks-0.141
 pads-0.137
llers-0.135
 mess-0.132
Tokenyear
Token ID1941
Feature activation-7.05e+00

Pos logit contributions
 reward+0.144
 help+0.131
ar+0.129
 plunge+0.126
bye+0.110
Neg logit contributions
 lair-0.443
cks-0.443
 mess-0.421
 pads-0.420
 floor-0.416
Token old
Token ID1468
Feature activation-3.80e-01

Pos logit contributions
 help+0.260
 reward+0.256
ar+0.242
 plunge+0.232
bye+0.221
Neg logit contributions
 lair-0.823
cks-0.819
 mess-0.796
 pads-0.789
 floor-0.780
Token girl
Token ID2576
Feature activation-5.66e-01

Pos logit contributions
 help+0.022
 reward+0.021
ar+0.019
 plunge+0.019
bye+0.019
Neg logit contributions
cks-0.038
 lair-0.037
 pads-0.036
 mess-0.035
llers-0.035
Token who
Token ID508
Feature activation-1.60e+00

Pos logit contributions
 help+0.038
 reward+0.036
ar+0.036
 plunge+0.033
bye+0.033
Neg logit contributions
 lair-0.050
cks-0.050
 mess-0.048
 pads-0.046
llers-0.046
Token loved
Token ID6151
Feature activation-7.42e-01

Pos logit contributions
 help+0.118
 reward+0.113
ar+0.112
bye+0.106
f+0.106
Neg logit contributions
 lair-0.133
cks-0.130
 pads-0.123
 mess-0.123
 necks-0.120
Token to
Token ID284
Feature activation-1.92e-01

Pos logit contributions
 help+0.042
 reward+0.038
 plunge+0.035
ted+0.034
less+0.033
Neg logit contributions
cks-0.078
 lair-0.075
 pads-0.072
 mess-0.070
 necks-0.070
Token Joe
Token ID5689
Feature activation+1.75e-01

Pos logit contributions
 plunge+0.029
 reward+0.028
iest+0.027
bye+0.026
 help+0.026
Neg logit contributions
 frown-0.030
 neck-0.028
 lair-0.028
cks-0.028
 mess-0.028
Token,
Token ID11
Feature activation-6.06e-01

Pos logit contributions
 lair+0.014
cks+0.014
 mess+0.014
 pads+0.014
 floor+0.013
Neg logit contributions
 help-0.012
 reward-0.012
ar-0.012
 plunge-0.012
bye-0.011
Token a
Token ID257
Feature activation-7.13e-01

Pos logit contributions
 help+0.046
 reward+0.044
ar+0.042
 plunge+0.042
Sure+0.042
Neg logit contributions
 lair-0.051
cks-0.049
 mess-0.046
 pads-0.046
 floor-0.045
Token 3
Token ID513
Feature activation-1.62e+00

Pos logit contributions
 help+0.052
ar+0.052
 reward+0.052
 plunge+0.051
ted+0.051
Neg logit contributions
 lair-0.055
 mess-0.054
cks-0.052
 floor-0.052
 pads-0.050
Token-
Token ID12
Feature activation-3.80e+00

Pos logit contributions
 help+0.090
 reward+0.087
ar+0.086
 plunge+0.081
bye+0.079
Neg logit contributions
 lair-0.162
cks-0.158
 pads-0.155
 mess-0.153
llers-0.150
Tokenyear
Token ID1941
Feature activation-7.02e+00

Pos logit contributions
 reward+0.182
 plunge+0.153
ar+0.150
 help+0.149
less+0.128
Neg logit contributions
cks-0.410
 lair-0.390
 pads-0.384
opl-0.366
cking-0.364
Token-
Token ID12
Feature activation-3.79e+00

Pos logit contributions
 help+0.268
 reward+0.266
ar+0.250
 plunge+0.247
Sure+0.228
Neg logit contributions
 lair-0.808
cks-0.801
 mess-0.781
 pads-0.770
 floor-0.766
Tokenold
Token ID727
Feature activation-1.21e+00

Pos logit contributions
 reward+0.327
 plunge+0.310
 help+0.304
ar+0.298
ted+0.285
Neg logit contributions
cks-0.253
 lair-0.250
 pads-0.236
opl-0.226
 necks-0.219
Token boy
Token ID2933
Feature activation+2.38e-01

Pos logit contributions
 reward+0.083
 plunge+0.083
ar+0.082
 help+0.079
Sure+0.078
Neg logit contributions
 lair-0.097
 mess-0.093
cks-0.091
 pads-0.090
 neck-0.090
Token.
Token ID13
Feature activation-2.52e-01

Pos logit contributions
 lair+0.019
cks+0.019
 mess+0.018
 pads+0.018
 floor+0.017
Neg logit contributions
 reward-0.018
 help-0.017
 plunge-0.017
ar-0.017
bye-0.016
Token He
Token ID679
Feature activation-3.03e-01

Pos logit contributions
 help+0.018
 reward+0.018
ar+0.017
 plunge+0.017
bye+0.016
Neg logit contributions
 lair-0.021
cks-0.021
 mess-0.020
 pads-0.020
 floor-0.019
TokenPat
Token ID12130
Feature activation-5.72e-01

Pos logit contributions
cks+0.023
 lair+0.023
 pads+0.023
 floor+0.022
 mess+0.022
Neg logit contributions
 reward-0.020
 help-0.020
 plunge-0.019
ar-0.019
bye-0.018
Token was
Token ID373
Feature activation-3.04e-01

Pos logit contributions
 help+0.035
ar+0.035
 reward+0.034
 plunge+0.033
ted+0.033
Neg logit contributions
 lair-0.053
cks-0.052
 mess-0.050
 floor-0.049
 pads-0.048
Token a
Token ID257
Feature activation-6.56e-01

Pos logit contributions
 help+0.020
 reward+0.018
 plunge+0.018
bye+0.017
Sure+0.017
Neg logit contributions
 lair-0.027
cks-0.027
 sizes-0.026
llers-0.026
 necks-0.026
Token 3
Token ID513
Feature activation-1.68e+00

Pos logit contributions
 help+0.045
 reward+0.045
ar+0.044
 plunge+0.043
bye+0.042
Neg logit contributions
 lair-0.055
cks-0.055
 mess-0.053
 pads-0.052
 floor-0.051
Token-
Token ID12
Feature activation-3.74e+00

Pos logit contributions
 help+0.088
ar+0.083
 reward+0.083
f+0.077
bye+0.077
Neg logit contributions
 lair-0.173
cks-0.170
 pads-0.166
 mess-0.163
llers-0.162
Tokenyear
Token ID1941
Feature activation-7.01e+00

Pos logit contributions
 reward+0.176
ar+0.160
 help+0.157
 plunge+0.156
bye+0.138
Neg logit contributions
cks-0.406
 lair-0.405
 pads-0.383
 mess-0.380
 floor-0.374
Token-
Token ID12
Feature activation-3.74e+00

Pos logit contributions
 help+0.267
 reward+0.264
ar+0.249
 plunge+0.242
bye+0.226
Neg logit contributions
 lair-0.809
cks-0.803
 mess-0.782
 pads-0.774
 floor-0.766
Tokenold
Token ID727
Feature activation-1.35e+00

Pos logit contributions
 reward+0.290
 help+0.279
 plunge+0.277
ar+0.277
ted+0.258
Neg logit contributions
 lair-0.288
cks-0.285
 pads-0.268
 mess-0.266
 floor-0.259
Token boy
Token ID2933
Feature activation+8.12e-02

Pos logit contributions
 reward+0.105
 plunge+0.103
ar+0.102
 help+0.102
Sure+0.096
Neg logit contributions
 lair-0.101
cks-0.095
 pads-0.095
 mess-0.095
 floor-0.094
Token.
Token ID13
Feature activation-3.48e-01

Pos logit contributions
 lair+0.006
cks+0.006
 mess+0.006
 pads+0.006
 floor+0.005
Neg logit contributions
 help-0.006
 reward-0.006
ar-0.006
 plunge-0.006
bye-0.006
Token He
Token ID679
Feature activation-3.91e-01

Pos logit contributions
 help+0.023
 reward+0.022
bye+0.021
ar+0.021
Sure+0.021
Neg logit contributions
cks-0.032
 mess-0.032
 lair-0.031
irling-0.029
cking-0.029
Token a
Token ID257
Feature activation-6.96e-01

Pos logit contributions
 help+0.018
 reward+0.017
 plunge+0.017
ar+0.016
bye+0.016
Neg logit contributions
 lair-0.028
cks-0.028
 mess-0.027
 pads-0.026
 necks-0.026
Token very
Token ID845
Feature activation-5.03e-01

Pos logit contributions
 help+0.051
 reward+0.051
ar+0.050
 plunge+0.049
ted+0.047
Neg logit contributions
 lair-0.055
cks-0.054
 mess-0.053
 pads-0.051
 floor-0.051
Token adventurous
Token ID39770
Feature activation-7.94e-01

Pos logit contributions
 reward+0.046
ar+0.046
 plunge+0.046
 help+0.046
ted+0.045
Neg logit contributions
 lair-0.029
 mess-0.029
cks-0.028
 floor-0.028
 frown-0.027
Token three
Token ID1115
Feature activation-1.41e+00

Pos logit contributions
 plunge+0.054
ar+0.053
 help+0.053
 reward+0.053
ted+0.052
Neg logit contributions
 lair-0.064
 mess-0.064
 floor-0.063
 neck-0.060
 frown-0.059
Token-
Token ID12
Feature activation-3.73e+00

Pos logit contributions
 help+0.085
 reward+0.083
ar+0.081
 plunge+0.076
bye+0.075
Neg logit contributions
 lair-0.134
cks-0.132
 pads-0.126
 mess-0.126
llers-0.123
Tokenyear
Token ID1941
Feature activation-6.96e+00

Pos logit contributions
 reward+0.223
 help+0.200
 plunge+0.199
ar+0.192
ted+0.180
Neg logit contributions
 lair-0.361
cks-0.358
 pads-0.343
opl-0.331
 floor-0.326
Token-
Token ID12
Feature activation-3.74e+00

Pos logit contributions
 help+0.276
 reward+0.274
ar+0.258
 plunge+0.257
Sure+0.239
Neg logit contributions
 lair-0.792
cks-0.781
 mess-0.764
 pads-0.755
 floor-0.751
Tokenold
Token ID727
Feature activation-1.29e+00

Pos logit contributions
 reward+0.310
 help+0.296
 plunge+0.295
ar+0.288
ted+0.277
Neg logit contributions
 lair-0.271
cks-0.266
 pads-0.254
 mess-0.242
 floor-0.241
Token.
Token ID13
Feature activation-4.33e-01

Pos logit contributions
 reward+0.102
 plunge+0.102
 help+0.100
ar+0.099
Sure+0.095
Neg logit contributions
 lair-0.095
 pads-0.091
cks-0.088
 mess-0.087
 neck-0.087
Token She
Token ID1375
Feature activation+2.79e-01

Pos logit contributions
bye+0.027
 help+0.027
 reward+0.026
 Enjoy+0.025
Sure+0.025
Neg logit contributions
cks-0.043
 mess-0.040
cking-0.039
 lair-0.038
irling-0.037
Token was
Token ID373
Feature activation-2.17e-01

Pos logit contributions
cks+0.026
 lair+0.026
llers+0.024
 mess+0.024
 necks+0.023
Neg logit contributions
 help-0.018
ar-0.017
 reward-0.016
bye-0.016
f-0.016
Token.
Token ID13
Feature activation-3.17e-01

Pos logit contributions
bye+0.004
 help+0.004
ar+0.004
 reward+0.003
less+0.003
Neg logit contributions
 lair-0.008
cks-0.008
 mess-0.008
 frown-0.008
llers-0.008
Token Two
Token ID4930
Feature activation-2.98e-01

Pos logit contributions
 help+0.023
 reward+0.023
ar+0.023
 plunge+0.023
less+0.022
Neg logit contributions
 lair-0.025
cks-0.024
 floor-0.023
 pads-0.023
llers-0.023
Token curious
Token ID11040
Feature activation+1.04e+00

Pos logit contributions
 help+0.019
 reward+0.019
ar+0.018
 plunge+0.018
bye+0.017
Neg logit contributions
 lair-0.027
cks-0.027
 mess-0.026
 pads-0.025
 floor-0.025
Token three
Token ID1115
Feature activation-1.29e+00

Pos logit contributions
 lair+0.072
 floor+0.068
 mess+0.067
 pads+0.065
cks+0.065
Neg logit contributions
ar-0.089
 reward-0.088
 help-0.087
 plunge-0.086
Sure-0.086
Token-
Token ID12
Feature activation-3.70e+00

Pos logit contributions
 help+0.064
ar+0.062
 reward+0.059
f+0.056
bye+0.053
Neg logit contributions
 lair-0.137
 pads-0.130
cks-0.129
 mess-0.126
 floor-0.124
Tokenyear
Token ID1941
Feature activation-6.95e+00

Pos logit contributions
 reward+0.235
 plunge+0.218
 help+0.210
ar+0.205
less+0.196
Neg logit contributions
cks-0.347
 lair-0.342
 pads-0.328
 sizes-0.319
opl-0.314
Token-
Token ID12
Feature activation-3.71e+00

Pos logit contributions
 help+0.287
 reward+0.286
ar+0.269
 plunge+0.268
Sure+0.252
Neg logit contributions
 lair-0.780
cks-0.772
 mess-0.750
 pads-0.742
 floor-0.737
Tokenold
Token ID727
Feature activation-1.26e+00

Pos logit contributions
 reward+0.302
 plunge+0.288
 help+0.288
ar+0.281
less+0.270
Neg logit contributions
 lair-0.275
cks-0.273
 pads-0.258
llers-0.248
 mess-0.247
Token children
Token ID1751
Feature activation-8.83e-01

Pos logit contributions
 reward+0.100
 plunge+0.100
 help+0.095
ar+0.095
Sure+0.093
Neg logit contributions
 lair-0.093
 pads-0.092
cks-0.088
 sizes-0.088
 necks-0.087
Token,
Token ID11
Feature activation-6.16e-01

Pos logit contributions
 reward+0.050
 help+0.048
ar+0.047
 plunge+0.046
Sure+0.045
Neg logit contributions
cks-0.086
 lair-0.086
 mess-0.082
 pads-0.082
 floor-0.080
Token Bob
Token ID5811
Feature activation-1.01e-01

Pos logit contributions
 help+0.040
 reward+0.039
ar+0.038
 plunge+0.037
bye+0.036
Neg logit contributions
 lair-0.055
cks-0.054
 mess-0.053
 pads-0.052
 floor-0.051
Token there
Token ID612
Feature activation-4.20e-01

Pos logit contributions
 lair+0.134
cks+0.133
 mess+0.128
 pads+0.127
 floor+0.125
Neg logit contributions
 help-0.086
 reward-0.084
ar-0.082
 plunge-0.081
less-0.077
Token was
Token ID373
Feature activation-4.43e-01

Pos logit contributions
 reward+0.033
 help+0.033
ar+0.032
 plunge+0.032
bye+0.031
Neg logit contributions
 lair-0.029
cks-0.029
 mess-0.029
 floor-0.028
 frown-0.028
Token a
Token ID257
Feature activation-7.14e-01

Pos logit contributions
 reward+0.023
 help+0.023
ar+0.022
 plunge+0.021
Sure+0.020
Neg logit contributions
 lair-0.045
cks-0.045
 mess-0.043
 pads-0.043
 floor-0.043
Token 4
Token ID604
Feature activation-1.75e+00

Pos logit contributions
 help+0.043
 reward+0.042
ar+0.041
 plunge+0.041
bye+0.039
Neg logit contributions
 lair-0.066
cks-0.065
 mess-0.064
 pads-0.062
 floor-0.062
Token-
Token ID12
Feature activation-3.65e+00

Pos logit contributions
 reward+0.093
 help+0.093
 plunge+0.093
ar+0.087
Sure+0.086
Neg logit contributions
cks-0.172
 lair-0.171
 mess-0.168
 floor-0.167
 necks-0.164
Tokenyear
Token ID1941
Feature activation-6.95e+00

Pos logit contributions
 reward+0.271
 help+0.261
 plunge+0.258
ar+0.255
ted+0.243
Neg logit contributions
cks-0.289
 lair-0.286
 pads-0.273
 mess-0.267
 floor-0.267
Token old
Token ID1468
Feature activation-3.55e-01

Pos logit contributions
 help+0.281
 reward+0.278
ar+0.263
 plunge+0.261
Sure+0.244
Neg logit contributions
 lair-0.781
cks-0.772
 mess-0.756
 pads-0.744
 floor-0.742
Token girl
Token ID2576
Feature activation-5.83e-01

Pos logit contributions
 help+0.025
 reward+0.025
ar+0.024
 plunge+0.023
bye+0.023
Neg logit contributions
cks-0.030
 lair-0.030
 pads-0.028
 mess-0.028
llers-0.027
Token named
Token ID3706
Feature activation-2.93e+00

Pos logit contributions
 help+0.031
 reward+0.030
ar+0.029
 plunge+0.028
bye+0.028
Neg logit contributions
 lair-0.059
cks-0.058
 mess-0.057
 pads-0.056
 floor-0.055
Token Sus
Token ID8932
Feature activation-6.82e-01

Pos logit contributions
 reward+0.184
 help+0.184
ar+0.178
 plunge+0.175
bye+0.168
Neg logit contributions
 lair-0.265
cks-0.263
 mess-0.256
 pads-0.252
 floor-0.250
Tokenie
Token ID494
Feature activation-1.04e+00

Pos logit contributions
 help+0.027
f+0.025
less+0.025
ted+0.024
 reward+0.024
Neg logit contributions
 lair-0.078
cks-0.075
 pads-0.072
llers-0.072
 necks-0.071
Token buried
Token ID11694
Feature activation-9.62e-01

Pos logit contributions
 help+0.005
 reward+0.005
 plunge+0.005
ar+0.005
less+0.005
Neg logit contributions
 lair-0.006
cks-0.006
 necks-0.006
 pads-0.006
 mess-0.006
Token,
Token ID11
Feature activation-3.53e-01

Pos logit contributions
 reward+0.063
 help+0.061
bye+0.060
Sure+0.059
 plunge+0.059
Neg logit contributions
 mess-0.084
cks-0.083
 lair-0.083
 floor-0.082
 pads-0.079
Token the
Token ID262
Feature activation+4.05e-02

Pos logit contributions
ar+0.022
 help+0.022
 reward+0.022
bye+0.021
 plunge+0.021
Neg logit contributions
 lair-0.031
 mess-0.030
 floor-0.030
 neck-0.029
cks-0.029
Token 3
Token ID513
Feature activation-1.54e+00

Pos logit contributions
 lair+0.003
cks+0.003
 mess+0.003
 pads+0.003
 floor+0.003
Neg logit contributions
 help-0.003
 reward-0.003
ar-0.003
 plunge-0.003
bye-0.003
Token-
Token ID12
Feature activation-3.80e+00

Pos logit contributions
 help+0.080
ar+0.078
 reward+0.078
 plunge+0.072
bye+0.072
Neg logit contributions
 lair-0.158
cks-0.156
 pads-0.151
 mess-0.149
llers-0.146
Tokenyear
Token ID1941
Feature activation-6.94e+00

Pos logit contributions
 reward+0.205
 help+0.193
 plunge+0.186
ar+0.178
ted+0.159
Neg logit contributions
cks-0.381
 lair-0.380
 pads-0.367
 mess-0.353
 floor-0.352
Token-
Token ID12
Feature activation-3.79e+00

Pos logit contributions
 help+0.263
 reward+0.259
ar+0.247
 plunge+0.237
bye+0.227
Neg logit contributions
 lair-0.803
cks-0.799
 mess-0.776
 pads-0.769
 floor-0.759
Tokenold
Token ID727
Feature activation-1.23e+00

Pos logit contributions
 reward+0.308
 help+0.298
 plunge+0.293
ar+0.283
ted+0.271
Neg logit contributions
 lair-0.278
cks-0.273
 pads-0.262
 mess-0.251
llers-0.249
Token and
Token ID290
Feature activation-1.02e-01

Pos logit contributions
 reward+0.089
 help+0.089
 plunge+0.089
ar+0.086
less+0.083
Neg logit contributions
 lair-0.097
 mess-0.093
 pads-0.092
 floor-0.092
cks-0.091
Token the
Token ID262
Feature activation+3.74e-02

Pos logit contributions
 help+0.007
 reward+0.006
ar+0.006
 plunge+0.006
bye+0.006
Neg logit contributions
 lair-0.009
cks-0.009
 mess-0.009
 pads-0.009
 floor-0.009
Token parent
Token ID2560
Feature activation-3.72e-01

Pos logit contributions
 lair+0.003
 mess+0.003
 floor+0.003
 pads+0.003
cks+0.003
Neg logit contributions
ar-0.003
 reward-0.003
 help-0.003
 plunge-0.002
bye-0.002
Token was
Token ID373
Feature activation-1.62e-01

Pos logit contributions
cks+0.044
 lair+0.044
 mess+0.040
 pads+0.039
opl+0.039
Neg logit contributions
 help-0.031
ar-0.030
 reward-0.030
bye-0.029
Sure-0.029
Token an
Token ID281
Feature activation+1.52e-01

Pos logit contributions
 help+0.010
 reward+0.010
 plunge+0.010
ar+0.009
Sure+0.009
Neg logit contributions
 lair-0.015
cks-0.015
 mess-0.014
 pads-0.014
 necks-0.014
Token attractive
Token ID10966
Feature activation-7.11e-01

Pos logit contributions
 lair+0.010
cks+0.010
 mess+0.010
 pads+0.009
 floor+0.009
Neg logit contributions
 help-0.013
 reward-0.013
ar-0.013
 plunge-0.013
bye-0.012
Token three
Token ID1115
Feature activation-1.28e+00

Pos logit contributions
ar+0.052
 help+0.051
 plunge+0.051
 reward+0.050
ted+0.050
Neg logit contributions
 lair-0.056
 mess-0.055
 floor-0.053
 neck-0.051
 frown-0.051
Token-
Token ID12
Feature activation-3.79e+00

Pos logit contributions
 help+0.054
 reward+0.050
ar+0.046
f+0.045
iest+0.044
Neg logit contributions
 lair-0.150
cks-0.149
llers-0.144
 pads-0.143
ric-0.136
Tokenyear
Token ID1941
Feature activation-6.94e+00

Pos logit contributions
 reward+0.155
 plunge+0.145
 help+0.145
ar+0.137
ted+0.119
Neg logit contributions
cks-0.432
 lair-0.428
 pads-0.413
 mess-0.407
 floor-0.403
Token-
Token ID12
Feature activation-3.78e+00

Pos logit contributions
 help+0.230
 reward+0.224
ar+0.211
bye+0.201
 plunge+0.190
Neg logit contributions
cks-0.842
 lair-0.835
 mess-0.809
 pads-0.809
llers-0.791
Tokenold
Token ID727
Feature activation-1.19e+00

Pos logit contributions
 reward+0.295
 help+0.288
 plunge+0.288
ar+0.280
ted+0.265
Neg logit contributions
 lair-0.285
cks-0.284
 pads-0.271
 mess-0.265
 floor-0.261
Token who
Token ID508
Feature activation-1.47e+00

Pos logit contributions
 reward+0.078
 help+0.077
ar+0.076
 plunge+0.076
Sure+0.071
Neg logit contributions
 lair-0.103
cks-0.100
 mess-0.099
 pads-0.098
 floor-0.097
Token loved
Token ID6151
Feature activation-6.38e-01

Pos logit contributions
 reward+0.122
 help+0.122
ar+0.118
 plunge+0.118
bye+0.113
Neg logit contributions
 lair-0.103
cks-0.102
 mess-0.098
 pads-0.096
 floor-0.095
Token reading
Token ID3555
Feature activation+1.62e+00

Pos logit contributions
 help+0.032
 reward+0.029
less+0.026
 plunge+0.026
ted+0.026
Neg logit contributions
cks-0.070
 lair-0.067
 pads-0.065
 mess-0.063
 necks-0.063
Token there
Token ID612
Feature activation-4.10e-01

Pos logit contributions
 reward+0.059
 help+0.056
ar+0.055
 plunge+0.054
bye+0.054
Neg logit contributions
cks-0.072
 lair-0.069
 mess-0.068
 pads-0.067
 floor-0.066
Token was
Token ID373
Feature activation-4.26e-01

Pos logit contributions
 help+0.028
 reward+0.028
ar+0.027
 plunge+0.027
bye+0.026
Neg logit contributions
 lair-0.035
cks-0.034
 mess-0.033
 pads-0.032
 floor-0.032
Token a
Token ID257
Feature activation-8.35e-01

Pos logit contributions
ar+0.029
 reward+0.028
 plunge+0.027
 help+0.026
less+0.025
Neg logit contributions
 lair-0.036
cks-0.036
 floor-0.036
 pads-0.036
 mess-0.036
Token three
Token ID1115
Feature activation-1.39e+00

Pos logit contributions
 plunge+0.060
ar+0.059
ted+0.058
 help+0.058
 reward+0.058
Neg logit contributions
 mess-0.068
 lair-0.066
 floor-0.064
 frown-0.063
 neck-0.063
Token-
Token ID12
Feature activation-3.65e+00

Pos logit contributions
 help+0.068
 reward+0.066
ar+0.064
 plunge+0.058
bye+0.058
Neg logit contributions
 lair-0.147
cks-0.145
 pads-0.140
 mess-0.138
llers-0.137
Tokenyear
Token ID1941
Feature activation-6.94e+00

Pos logit contributions
 reward+0.145
 plunge+0.129
ar+0.128
 help+0.125
less+0.108
Neg logit contributions
 lair-0.425
cks-0.423
 pads-0.400
 mess-0.400
 floor-0.398
Token-
Token ID12
Feature activation-3.65e+00

Pos logit contributions
 help+0.268
 reward+0.265
ar+0.250
 plunge+0.243
bye+0.228
Neg logit contributions
 lair-0.797
cks-0.791
 mess-0.770
 pads-0.762
 floor-0.754
Tokenold
Token ID727
Feature activation-1.34e+00

Pos logit contributions
 reward+0.275
 help+0.265
 plunge+0.264
ar+0.260
ted+0.245
Neg logit contributions
 lair-0.292
cks-0.288
 pads-0.272
 mess-0.272
 floor-0.267
Token boy
Token ID2933
Feature activation+1.62e-01

Pos logit contributions
 plunge+0.105
 reward+0.099
ar+0.097
less+0.094
 help+0.093
Neg logit contributions
 lair-0.104
 neck-0.101
 mess-0.098
 pads-0.097
cks-0.096
Token called
Token ID1444
Feature activation-1.22e+00

Pos logit contributions
 lair+0.015
cks+0.015
 mess+0.015
 pads+0.014
 floor+0.014
Neg logit contributions
 reward-0.010
 plunge-0.009
 help-0.009
ar-0.009
bye-0.009
Token Tommy
Token ID19919
Feature activation-6.19e-01

Pos logit contributions
ar+0.098
ted+0.097
Sure+0.097
bye+0.095
 reward+0.094
Neg logit contributions
 lair-0.082
 floor-0.077
 frown-0.074
cks-0.073
 neck-0.072
Token and
Token ID290
Feature activation-2.95e-01

Pos logit contributions
 lair+0.082
cks+0.082
 mess+0.079
 pads+0.078
 floor+0.077
Neg logit contributions
 reward-0.054
 help-0.054
ar-0.052
 plunge-0.052
Sure-0.049
Token their
Token ID511
Feature activation-5.90e-01

Pos logit contributions
ar+0.020
 plunge+0.019
 reward+0.019
less+0.018
bye+0.018
Neg logit contributions
cks-0.025
 lair-0.025
 mess-0.024
 frown-0.024
 neck-0.024
Token little
Token ID1310
Feature activation-7.11e-01

Pos logit contributions
ar+0.056
ted+0.048
f+0.046
 plunge+0.046
less+0.046
Neg logit contributions
 mess-0.041
 floor-0.040
 lair-0.040
 frown-0.038
 neck-0.036
Token three
Token ID1115
Feature activation-1.34e+00

Pos logit contributions
ar+0.061
 reward+0.057
 plunge+0.055
ted+0.054
Sure+0.054
Neg logit contributions
 mess-0.049
 lair-0.049
 floor-0.047
 frown-0.047
 neck-0.047
Token-
Token ID12
Feature activation-3.61e+00

Pos logit contributions
 help+0.068
 reward+0.066
ar+0.064
 plunge+0.060
bye+0.059
Neg logit contributions
 lair-0.141
cks-0.138
 mess-0.133
 pads-0.133
 floor-0.130
Tokenyear
Token ID1941
Feature activation-6.94e+00

Pos logit contributions
 reward+0.198
ar+0.176
 plunge+0.174
 help+0.151
less+0.136
Neg logit contributions
cks-0.376
 lair-0.365
 pads-0.348
 sizes-0.345
 floor-0.341
Token-
Token ID12
Feature activation-3.62e+00

Pos logit contributions
 reward+0.305
 help+0.300
 plunge+0.295
ar+0.289
Sure+0.282
Neg logit contributions
 lair-0.757
cks-0.745
 mess-0.731
 floor-0.720
 pads-0.718
Tokenold
Token ID727
Feature activation-1.27e+00

Pos logit contributions
 reward+0.311
 plunge+0.298
ar+0.291
 help+0.281
less+0.267
Neg logit contributions
cks-0.254
 lair-0.253
 pads-0.239
 sizes-0.230
 mess-0.229
Token daughter
Token ID4957
Feature activation-2.02e+00

Pos logit contributions
 plunge+0.112
 reward+0.109
ar+0.108
Sure+0.104
 help+0.101
Neg logit contributions
 lair-0.082
 pads-0.079
 mess-0.079
 neck-0.078
 necks-0.078
Token.
Token ID13
Feature activation-4.12e-01

Pos logit contributions
 plunge+0.135
 reward+0.133
ar+0.130
Sure+0.126
 help+0.125
Neg logit contributions
 lair-0.173
cks-0.169
 pads-0.166
 mess-0.166
 floor-0.163
Token They
Token ID1119
Feature activation-1.19e+00

Pos logit contributions
 help+0.026
 reward+0.024
 Bub+0.024
Sure+0.024
 Enjoy+0.023
Neg logit contributions
cks-0.039
 lair-0.038
 mess-0.038
 pads-0.036
 necks-0.035
Tokenley
Token ID1636
Feature activation-2.91e+00

Pos logit contributions
 help+0.071
 reward+0.070
 plunge+0.069
ar+0.066
Sure+0.065
Neg logit contributions
 lair-0.065
cks-0.063
 pads-0.060
 mess-0.060
 necks-0.058
Token was
Token ID373
Feature activation-3.02e-01

Pos logit contributions
 help+0.198
ar+0.190
 reward+0.188
ted+0.179
bye+0.179
Neg logit contributions
cks-0.255
 lair-0.252
 mess-0.235
 pads-0.233
 sizes-0.232
Token a
Token ID257
Feature activation-6.88e-01

Pos logit contributions
 help+0.019
 reward+0.017
 plunge+0.016
ar+0.016
Sure+0.016
Neg logit contributions
 lair-0.028
cks-0.028
 sizes-0.028
 necks-0.027
 mess-0.026
Token three
Token ID1115
Feature activation-1.40e+00

Pos logit contributions
 help+0.047
 reward+0.047
ar+0.046
 plunge+0.045
bye+0.043
Neg logit contributions
 lair-0.058
cks-0.058
 mess-0.055
 pads-0.055
 floor-0.054
Token-
Token ID12
Feature activation-3.64e+00

Pos logit contributions
 help+0.071
 reward+0.068
ar+0.067
f+0.062
bye+0.060
Neg logit contributions
 lair-0.148
cks-0.145
 pads-0.140
llers-0.138
 mess-0.136
Tokenyear
Token ID1941
Feature activation-6.93e+00

Pos logit contributions
 reward+0.178
 plunge+0.162
ar+0.161
 help+0.157
bye+0.137
Neg logit contributions
 lair-0.389
cks-0.382
 pads-0.364
 mess-0.357
 necks-0.351
Token-
Token ID12
Feature activation-3.64e+00

Pos logit contributions
 help+0.258
 reward+0.255
ar+0.241
 plunge+0.233
bye+0.218
Neg logit contributions
 lair-0.806
cks-0.801
 mess-0.780
 pads-0.772
 floor-0.764
Tokenold
Token ID727
Feature activation-1.25e+00

Pos logit contributions
 reward+0.292
 plunge+0.284
 help+0.281
ar+0.278
ted+0.262
Neg logit contributions
 lair-0.272
cks-0.263
 pads-0.251
 mess-0.247
 necks-0.240
Token who
Token ID508
Feature activation-1.54e+00

Pos logit contributions
 reward+0.089
 plunge+0.089
ar+0.087
 help+0.087
Sure+0.082
Neg logit contributions
 lair-0.101
 pads-0.096
 mess-0.095
cks-0.094
 neck-0.093
Token loved
Token ID6151
Feature activation-6.95e-01

Pos logit contributions
 help+0.114
 reward+0.110
ar+0.109
bye+0.104
f+0.102
Neg logit contributions
 lair-0.125
cks-0.125
 mess-0.118
 pads-0.116
 necks-0.114
Token to
Token ID284
Feature activation-1.36e-01

Pos logit contributions
 help+0.041
 reward+0.038
 plunge+0.035
bye+0.034
ted+0.033
Neg logit contributions
cks-0.071
 lair-0.068
 pads-0.065
 mess-0.063
 necks-0.063
Token
Token ID198
Feature activation+2.29e-02

Pos logit contributions
 help+0.015
 reward+0.015
ar+0.014
 plunge+0.014
bye+0.013
Neg logit contributions
 lair-0.017
cks-0.017
 pads-0.016
 mess-0.016
 floor-0.016
Token
Token ID198
Feature activation+1.20e-01

Pos logit contributions
 lair+0.002
cks+0.002
 mess+0.002
 pads+0.002
 floor+0.002
Neg logit contributions
 help-0.002
 reward-0.002
ar-0.001
 plunge-0.001
bye-0.001
TokenThe
Token ID464
Feature activation-9.02e-01

Pos logit contributions
 lair+0.009
cks+0.009
 mess+0.009
 pads+0.009
 floor+0.008
Neg logit contributions
 help-0.009
 reward-0.009
ar-0.009
 plunge-0.009
bye-0.009
Token three
Token ID1115
Feature activation-1.28e+00

Pos logit contributions
ar+0.070
ted+0.067
less+0.067
 help+0.064
 plunge+0.064
Neg logit contributions
 floor-0.067
 lair-0.067
 mess-0.066
 neck-0.064
 frown-0.062
Token-
Token ID12
Feature activation-3.77e+00

Pos logit contributions
 help+0.074
 reward+0.073
ar+0.071
 plunge+0.069
bye+0.066
Neg logit contributions
 lair-0.123
cks-0.121
 mess-0.117
 pads-0.116
 floor-0.114
Tokenyear
Token ID1941
Feature activation-6.93e+00

Pos logit contributions
 reward+0.165
 help+0.149
 plunge+0.145
ar+0.141
less+0.123
Neg logit contributions
cks-0.422
 lair-0.420
 pads-0.402
 necks-0.390
 floor-0.388
Token-
Token ID12
Feature activation-3.76e+00

Pos logit contributions
 help+0.292
 reward+0.289
ar+0.273
 plunge+0.269
Sure+0.251
Neg logit contributions
 lair-0.773
cks-0.765
 mess-0.745
 pads-0.736
 floor-0.732
Tokenold
Token ID727
Feature activation-1.14e+00

Pos logit contributions
 reward+0.303
 help+0.292
 plunge+0.288
ar+0.281
ted+0.268
Neg logit contributions
 lair-0.280
cks-0.275
 pads-0.259
llers-0.251
 necks-0.250
Token girl
Token ID2576
Feature activation-4.71e-01

Pos logit contributions
 reward+0.087
 plunge+0.085
 help+0.084
ar+0.082
less+0.079
Neg logit contributions
 lair-0.087
 pads-0.083
 mess-0.083
 floor-0.082
cks-0.082
Token was
Token ID373
Feature activation-1.82e-01

Pos logit contributions
 help+0.034
 reward+0.033
ar+0.033
 plunge+0.032
bye+0.031
Neg logit contributions
 lair-0.039
cks-0.039
 mess-0.037
 pads-0.036
 floor-0.036
Token fascinated
Token ID30103
Feature activation+1.35e+00

Pos logit contributions
 help+0.013
 reward+0.013
ar+0.012
 plunge+0.012
less+0.012
Neg logit contributions
cks-0.015
 lair-0.015
 pads-0.014
 mess-0.014
 necks-0.014
Token
Token ID198
Feature activation-3.19e-01

Pos logit contributions
 reward+0.027
 help+0.027
ar+0.026
 plunge+0.026
bye+0.024
Neg logit contributions
 lair-0.033
cks-0.032
 floor-0.031
 mess-0.031
 pads-0.031
Token
Token ID198
Feature activation-2.54e-01

Pos logit contributions
 help+0.022
 reward+0.022
ar+0.021
 plunge+0.021
bye+0.020
Neg logit contributions
 lair-0.027
cks-0.027
 mess-0.026
 pads-0.025
 floor-0.025
TokenA
Token ID32
Feature activation-2.51e+00

Pos logit contributions
 help+0.017
 reward+0.017
ar+0.016
 plunge+0.016
Sure+0.015
Neg logit contributions
 lair-0.023
cks-0.022
 mess-0.022
 pads-0.021
 floor-0.021
Token 3
Token ID513
Feature activation-1.51e+00

Pos logit contributions
ted+0.242
 plunge+0.226
ed+0.210
ar+0.210
 Enjoy+0.205
Neg logit contributions
 mess-0.167
 floor-0.151
 frown-0.148
 web-0.141
 lair-0.134
Token-
Token ID12
Feature activation-3.66e+00

Pos logit contributions
 reward+0.087
 help+0.087
 plunge+0.083
ar+0.083
Sure+0.078
Neg logit contributions
 lair-0.143
cks-0.143
 mess-0.138
 floor-0.135
 pads-0.135
Tokenyear
Token ID1941
Feature activation-6.93e+00

Pos logit contributions
 reward+0.158
 plunge+0.142
 help+0.137
ar+0.137
less+0.124
Neg logit contributions
cks-0.417
 lair-0.413
 pads-0.387
 mess-0.384
 floor-0.380
Token-
Token ID12
Feature activation-3.66e+00

Pos logit contributions
 help+0.285
 reward+0.283
ar+0.268
 plunge+0.266
Sure+0.247
Neg logit contributions
 lair-0.777
cks-0.769
 mess-0.750
 pads-0.739
 floor-0.736
Tokenold
Token ID727
Feature activation-1.25e+00

Pos logit contributions
 reward+0.285
 help+0.276
 plunge+0.274
ar+0.268
less+0.257
Neg logit contributions
 lair-0.284
cks-0.281
 pads-0.262
 mess-0.261
 floor-0.255
Token toddler
Token ID30773
Feature activation-4.10e-01

Pos logit contributions
 plunge+0.105
 reward+0.101
less+0.097
ar+0.095
Sure+0.094
Neg logit contributions
 lair-0.087
 mess-0.080
 frown-0.079
 neck-0.079
 wondered-0.077
Token walked
Token ID6807
Feature activation-1.25e+00

Pos logit contributions
 reward+0.029
 plunge+0.029
bye+0.027
Sure+0.027
less+0.026
Neg logit contributions
 lair-0.033
cks-0.031
 frown-0.031
 neck-0.031
 mess-0.031
Token up
Token ID510
Feature activation-1.96e-01

Pos logit contributions
 reward+0.101
ar+0.094
 help+0.093
 plunge+0.091
bye+0.087
Neg logit contributions
 lair-0.095
 mess-0.090
cks-0.090
 pads-0.089
 floor-0.088
Token mum
Token ID25682
Feature activation+1.70e-01

Pos logit contributions
cks+0.106
 lair+0.100
 necks+0.098
 mess+0.098
llers+0.096
Neg logit contributions
 help-0.033
 reward-0.026
 plunge-0.023
less-0.022
bye-0.022
Token and
Token ID290
Feature activation-6.30e-02

Pos logit contributions
 lair+0.014
cks+0.014
 mess+0.013
 pads+0.013
 floor+0.013
Neg logit contributions
 help-0.012
 reward-0.012
ar-0.012
 plunge-0.011
bye-0.011
Token her
Token ID607
Feature activation+1.21e+00

Pos logit contributions
 reward+0.004
 help+0.004
ar+0.004
 plunge+0.004
bye+0.004
Neg logit contributions
 lair-0.005
cks-0.005
 mess-0.005
 pads-0.005
 floor-0.005
Token three
Token ID1115
Feature activation-1.17e+00

Pos logit contributions
 lair+0.107
cks+0.105
 mess+0.103
 pads+0.102
 floor+0.101
Neg logit contributions
 help-0.076
 reward-0.076
ar-0.075
 plunge-0.073
bye-0.071
Token-
Token ID12
Feature activation-3.70e+00

Pos logit contributions
 help+0.082
 reward+0.080
ar+0.079
 plunge+0.075
f+0.074
Neg logit contributions
 lair-0.100
cks-0.097
 mess-0.093
 pads-0.092
llers-0.091
Tokenyear
Token ID1941
Feature activation-6.93e+00

Pos logit contributions
 reward+0.216
 plunge+0.198
ar+0.195
 help+0.192
bye+0.175
Neg logit contributions
cks-0.354
 lair-0.347
 pads-0.342
 necks-0.329
 sizes-0.328
Token-
Token ID12
Feature activation-3.64e+00

Pos logit contributions
 help+0.279
 reward+0.278
 plunge+0.262
ar+0.262
Sure+0.244
Neg logit contributions
 lair-0.783
cks-0.774
 mess-0.757
 pads-0.746
 floor-0.743
Tokenold
Token ID727
Feature activation-1.01e+00

Pos logit contributions
 reward+0.309
 plunge+0.293
ar+0.290
 help+0.289
Sure+0.269
Neg logit contributions
cks-0.245
 lair-0.245
 pads-0.237
 sizes-0.230
llers-0.228
Token daughter
Token ID4957
Feature activation-1.67e+00

Pos logit contributions
 reward+0.069
ar+0.068
 plunge+0.068
 help+0.065
Sure+0.065
Neg logit contributions
 lair-0.082
 pads-0.080
 mess-0.079
cks-0.078
 floor-0.078
Token went
Token ID1816
Feature activation-6.24e-01

Pos logit contributions
 help+0.121
 reward+0.121
ar+0.118
 plunge+0.116
bye+0.112
Neg logit contributions
 lair-0.134
cks-0.132
 mess-0.127
 pads-0.126
 floor-0.124
Token to
Token ID284
Feature activation-1.21e-01

Pos logit contributions
 reward+0.041
bye+0.041
ar+0.040
 plunge+0.039
Sure+0.038
Neg logit contributions
 lair-0.051
 mess-0.051
 floor-0.049
 pads-0.049
cks-0.049
Token She
Token ID1375
Feature activation+2.64e-01

Pos logit contributions
 help+0.024
 reward+0.024
ar+0.023
 plunge+0.023
bye+0.022
Neg logit contributions
 lair-0.032
cks-0.032
 mess-0.031
 pads-0.031
 floor-0.030
Token told
Token ID1297
Feature activation-7.92e-01

Pos logit contributions
 lair+0.019
cks+0.019
 mess+0.019
 floor+0.018
 pads+0.018
Neg logit contributions
 help-0.021
 reward-0.021
 plunge-0.020
ar-0.020
bye-0.019
Token her
Token ID607
Feature activation+1.07e+00

Pos logit contributions
ar+0.062
 reward+0.059
Sure+0.058
 plunge+0.057
less+0.057
Neg logit contributions
 mess-0.061
 floor-0.059
 frown-0.058
cks-0.057
 lair-0.056
Token three
Token ID1115
Feature activation-1.38e+00

Pos logit contributions
 mess+0.084
 lair+0.084
 floor+0.081
 pads+0.079
 frown+0.079
Neg logit contributions
ar-0.079
 plunge-0.075
ted-0.073
bye-0.072
 reward-0.072
Token-
Token ID12
Feature activation-3.74e+00

Pos logit contributions
 help+0.073
 reward+0.073
ar+0.071
 plunge+0.066
bye+0.064
Neg logit contributions
 lair-0.140
cks-0.139
 pads-0.133
 mess-0.132
 floor-0.129
Tokenyear
Token ID1941
Feature activation-6.92e+00

Pos logit contributions
 reward+0.239
 plunge+0.232
 help+0.207
ar+0.202
less+0.177
Neg logit contributions
 pads-0.325
cks-0.320
 lair-0.317
ents-0.310
vice-0.300
Token-
Token ID12
Feature activation-3.74e+00

Pos logit contributions
 help+0.298
 reward+0.296
 plunge+0.287
ar+0.279
Sure+0.269
Neg logit contributions
 lair-0.761
cks-0.747
 mess-0.734
 floor-0.725
 pads-0.725
Tokenold
Token ID727
Feature activation-1.22e+00

Pos logit contributions
 plunge+0.361
 reward+0.360
 help+0.335
ar+0.334
ted+0.323
Neg logit contributions
 pads-0.201
 lair-0.200
cks-0.192
opl-0.180
 necks-0.179
Token daughter
Token ID4957
Feature activation-2.02e+00

Pos logit contributions
 plunge+0.084
ar+0.083
 reward+0.083
 help+0.077
less+0.075
Neg logit contributions
 lair-0.099
 pads-0.096
 mess-0.096
 floor-0.094
 neck-0.093
Token to
Token ID284
Feature activation-1.60e-01

Pos logit contributions
 reward+0.112
ar+0.111
 plunge+0.110
 help+0.107
bye+0.104
Neg logit contributions
 lair-0.192
 pads-0.186
cks-0.185
 mess-0.185
 floor-0.181
Token attach
Token ID10199
Feature activation-9.15e-01

Pos logit contributions
ar+0.013
ted+0.012
less+0.012
iest+0.011
ed+0.011
Neg logit contributions
 mess-0.012
 frown-0.012
 floor-0.011
 neck-0.011
 lair-0.011
Token on
Token ID319
Feature activation+3.53e-01

Pos logit contributions
 pads+0.028
cks+0.026
 goals+0.026
 neck+0.025
 beating+0.024
Neg logit contributions
 plunge-0.030
 reward-0.027
bye-0.026
Sure-0.026
 help-0.026
Token,
Token ID11
Feature activation-5.91e-02

Pos logit contributions
cks+0.034
 lair+0.034
 mess+0.033
 floor+0.033
 pads+0.033
Neg logit contributions
 reward-0.019
 help-0.019
 plunge-0.018
bye-0.018
ar-0.018
Token the
Token ID262
Feature activation+1.86e-01

Pos logit contributions
ar+0.004
 plunge+0.004
 reward+0.004
 help+0.004
less+0.004
Neg logit contributions
 lair-0.005
 pads-0.005
 mess-0.005
cks-0.005
 sizes-0.005
Token three
Token ID1115
Feature activation-1.24e+00

Pos logit contributions
 mess+0.014
 floor+0.014
 lair+0.013
 frown+0.013
 neck+0.013
Neg logit contributions
ar-0.015
ted-0.014
 plunge-0.014
Sure-0.014
less-0.013
Token-
Token ID12
Feature activation-3.68e+00

Pos logit contributions
 help+0.070
 reward+0.070
 plunge+0.067
ar+0.067
bye+0.064
Neg logit contributions
 lair-0.119
cks-0.118
 mess-0.115
 pads-0.113
 floor-0.112
Tokenyear
Token ID1941
Feature activation-6.92e+00

Pos logit contributions
 reward+0.199
 plunge+0.192
 help+0.187
ar+0.172
ted+0.156
Neg logit contributions
cks-0.360
 pads-0.354
 lair-0.352
llers-0.337
opl-0.337
Token-
Token ID12
Feature activation-3.69e+00

Pos logit contributions
 help+0.302
 reward+0.301
 plunge+0.293
Sure+0.279
ar+0.278
Neg logit contributions
 lair-0.750
cks-0.737
 mess-0.720
 floor-0.714
 pads-0.713
Tokenold
Token ID727
Feature activation-1.12e+00

Pos logit contributions
 reward+0.307
 plunge+0.301
 help+0.293
ar+0.280
ted+0.272
Neg logit contributions
cks-0.250
 lair-0.247
 pads-0.245
opl-0.231
llers-0.231
Token started
Token ID2067
Feature activation-6.97e-01

Pos logit contributions
 plunge+0.091
 reward+0.091
 help+0.090
ar+0.087
Sure+0.086
Neg logit contributions
 lair-0.077
 pads-0.075
 mess-0.074
 floor-0.073
cks-0.073
Token testing
Token ID4856
Feature activation+2.54e-01

Pos logit contributions
 help+0.037
 reward+0.033
less+0.031
ar+0.031
 plunge+0.031
Neg logit contributions
cks-0.074
 lair-0.072
 mess-0.070
 pads-0.069
 floor-0.068
Token and
Token ID290
Feature activation-9.83e-03

Pos logit contributions
 lair+0.023
cks+0.022
 mess+0.022
 pads+0.021
 floor+0.021
Neg logit contributions
 help-0.016
 reward-0.016
ar-0.016
 plunge-0.016
bye-0.015
Token,
Token ID11
Feature activation-4.97e-01

Pos logit contributions
 plunge+0.105
bye+0.101
Sure+0.098
 reward+0.098
ar+0.097
Neg logit contributions
 frown-0.080
 mess-0.080
 floor-0.079
 neck-0.074
 reading-0.074
Token a
Token ID257
Feature activation-6.74e-01

Pos logit contributions
ar+0.044
 plunge+0.042
bye+0.042
or+0.041
less+0.039
Neg logit contributions
 floor-0.032
 frown-0.031
 neck-0.031
 mess-0.031
 uneven-0.029
Token little
Token ID1310
Feature activation-7.73e-01

Pos logit contributions
ted+0.062
 plunge+0.059
ar+0.057
bye+0.056
f+0.055
Neg logit contributions
 mess-0.045
 frown-0.043
 floor-0.041
 lair-0.040
 web-0.039
Token three
Token ID1115
Feature activation-1.28e+00

Pos logit contributions
 reward+0.051
ar+0.050
 plunge+0.049
 help+0.048
f+0.046
Neg logit contributions
 mess-0.066
 lair-0.065
 floor-0.063
 frown-0.063
 neck-0.061
Token-
Token ID12
Feature activation-3.61e+00

Pos logit contributions
 reward+0.062
 help+0.061
 plunge+0.059
ar+0.058
bye+0.055
Neg logit contributions
 lair-0.133
cks-0.133
 mess-0.130
 floor-0.127
 pads-0.127
Tokenyear
Token ID1941
Feature activation-6.91e+00

Pos logit contributions
 reward+0.185
 plunge+0.168
ar+0.159
 help+0.152
less+0.141
Neg logit contributions
cks-0.382
 lair-0.375
 pads-0.356
 floor-0.344
 mess-0.344
Token-
Token ID12
Feature activation-3.61e+00

Pos logit contributions
 help+0.270
 reward+0.269
ar+0.253
 plunge+0.252
Sure+0.234
Neg logit contributions
 lair-0.790
cks-0.780
 mess-0.762
 pads-0.752
 floor-0.748
Tokenold
Token ID727
Feature activation-1.26e+00

Pos logit contributions
 reward+0.313
 plunge+0.303
 help+0.294
ar+0.290
less+0.278
Neg logit contributions
 lair-0.249
cks-0.247
 pads-0.230
 mess-0.222
 floor-0.218
Token boy
Token ID2933
Feature activation+2.78e-01

Pos logit contributions
 plunge+0.099
 reward+0.096
ar+0.091
less+0.090
Sure+0.089
Neg logit contributions
 lair-0.094
 mess-0.088
 pads-0.086
 neck-0.085
 frown-0.085
Token named
Token ID3706
Feature activation-2.60e+00

Pos logit contributions
 lair+0.021
cks+0.021
 pads+0.020
 mess+0.020
 frown+0.020
Neg logit contributions
 reward-0.021
 plunge-0.021
Sure-0.020
less-0.019
bye-0.019
Token John
Token ID1757
Feature activation+7.77e-01

Pos logit contributions
 reward+0.173
ar+0.168
 help+0.164
bye+0.159
 plunge+0.159
Neg logit contributions
 lair-0.229
cks-0.222
 mess-0.220
 floor-0.216
 pads-0.211
Token,
Token ID11
Feature activation-1.97e-01

Pos logit contributions
 mess+0.062
 lair+0.061
cks+0.061
 floor+0.061
 pads+0.059
Neg logit contributions
 reward-0.039
ar-0.038
 help-0.038
 plunge-0.037
bye-0.036
Token she
Token ID673
Feature activation+7.39e-01

Pos logit contributions
ar+0.017
bye+0.016
 reward+0.016
 plunge+0.016
less+0.016
Neg logit contributions
 mess-0.013
 floor-0.013
 frown-0.012
 lair-0.012
 neck-0.012
Token looked
Token ID3114
Feature activation+1.01e-01

Pos logit contributions
 lair+0.077
cks+0.077
opl+0.075
 pads+0.069
 sizes+0.069
Neg logit contributions
 help-0.043
 reward-0.038
f-0.037
Sure-0.036
ar-0.035
Token at
Token ID379
Feature activation+7.75e-02

Pos logit contributions
 lair+0.009
cks+0.009
 pads+0.009
 mess+0.009
 floor+0.008
Neg logit contributions
 help-0.007
 reward-0.006
ar-0.006
 plunge-0.006
Sure-0.006
Token her
Token ID607
Feature activation+1.09e+00

Pos logit contributions
 lair+0.008
cks+0.008
 pads+0.007
 mess+0.007
 necks+0.007
Neg logit contributions
 help-0.004
 reward-0.004
 plunge-0.004
ar-0.004
ted-0.004
Token wall
Token ID3355
Feature activation+2.55e+00

Pos logit contributions
cks+0.103
 lair+0.101
 pads+0.098
 sizes+0.095
 necks+0.090
Neg logit contributions
 help-0.070
 reward-0.070
 plunge-0.060
ar-0.059
f-0.059
Token and
Token ID290
Feature activation-5.87e-02

Pos logit contributions
cks+0.268
 lair+0.262
die+0.254
 pads+0.248
opl+0.242
Neg logit contributions
 help-0.136
less-0.126
f-0.121
 reward-0.119
ar-0.116
Token smiled
Token ID13541
Feature activation-1.04e+00

Pos logit contributions
 help+0.004
ted+0.003
 reward+0.003
Sure+0.003
ar+0.003
Neg logit contributions
 lair-0.006
cks-0.005
opl-0.005
 pads-0.005
 fault-0.005
Token.
Token ID13
Feature activation-9.31e-03

Pos logit contributions
 help+0.065
 reward+0.063
less+0.059
bye+0.058
ar+0.058
Neg logit contributions
cks-0.099
 lair-0.097
 pads-0.092
 necks-0.091
 mess-0.091
Token It
Token ID632
Feature activation-1.84e+00

Pos logit contributions
 help+0.001
 reward+0.001
ar+0.001
Sure+0.001
 plunge+0.001
Neg logit contributions
cks-0.001
 lair-0.001
 mess-0.001
 pads-0.001
 necks-0.001
Token was
Token ID373
Feature activation-8.63e-04

Pos logit contributions
 help+0.086
 reward+0.074
Sure+0.071
ar+0.071
f+0.067
Neg logit contributions
 lair-0.217
cks-0.210
 necks-0.200
 pads-0.197
 goals-0.193
Token will
Token ID481
Feature activation-1.01e+00

Pos logit contributions
 help+0.036
 reward+0.035
ar+0.033
 plunge+0.032
bye+0.031
Neg logit contributions
 lair-0.050
cks-0.050
 pads-0.047
 mess-0.047
 floor-0.047
Token be
Token ID307
Feature activation+4.06e-01

Pos logit contributions
 help+0.079
 reward+0.075
 plunge+0.070
ar+0.070
bye+0.068
Neg logit contributions
 lair-0.084
cks-0.083
 pads-0.080
 mess-0.076
 necks-0.076
Token in
Token ID287
Feature activation+9.47e-02

Pos logit contributions
 lair+0.036
llers+0.035
 sizes+0.035
 pads+0.035
 floor+0.035
Neg logit contributions
 help-0.027
 reward-0.025
ar-0.024
 plunge-0.023
less-0.022
Token trouble
Token ID5876
Feature activation+1.82e+00

Pos logit contributions
cks+0.009
 pads+0.009
 lair+0.009
llers+0.008
 necks+0.008
Neg logit contributions
 help-0.006
 reward-0.006
 plunge-0.005
ar-0.005
less-0.005
Token!"
Token ID2474
Feature activation+9.29e-01

Pos logit contributions
 lair+0.160
cks+0.158
 pads+0.152
 mess+0.151
 floor+0.148
Neg logit contributions
 help-0.122
 reward-0.119
ar-0.116
 plunge-0.114
bye-0.110
Token Anna
Token ID11735
Feature activation+2.05e+00

Pos logit contributions
 lair+0.083
cks+0.082
 pads+0.081
llers+0.078
 sizes+0.077
Neg logit contributions
 help-0.066
 reward-0.062
ar-0.058
Sure-0.057
 plunge-0.055
Token said
Token ID531
Feature activation-1.41e-01

Pos logit contributions
 lair+0.208
cks+0.204
opl+0.197
 mess+0.188
 pads+0.188
Neg logit contributions
 help-0.120
ar-0.112
 reward-0.110
f-0.105
Sure-0.105
Token.
Token ID13
Feature activation+3.37e-01

Pos logit contributions
 reward+0.011
ar+0.011
 plunge+0.011
 help+0.011
bye+0.011
Neg logit contributions
cks-0.010
 lair-0.010
 mess-0.010
 pads-0.009
 floor-0.009
Token "
Token ID366
Feature activation+1.62e-01

Pos logit contributions
 lair+0.035
cks+0.035
 mess+0.034
opl+0.034
 sizes+0.034
Neg logit contributions
 help-0.019
Sure-0.017
 reward-0.016
ted-0.016
ar-0.015
TokenWe
Token ID1135
Feature activation-1.65e+00

Pos logit contributions
 lair+0.016
 mess+0.015
cks+0.015
 pads+0.015
opl+0.015
Neg logit contributions
Sure-0.009
 help-0.009
bye-0.008
f-0.008
 reward-0.008
Token were
Token ID547
Feature activation+2.56e-01

Pos logit contributions
 reward+0.110
 help+0.108
ar+0.107
 plunge+0.106
bye+0.103
Neg logit contributions
 lair-0.141
cks-0.140
 mess-0.137
 pads-0.134
 floor-0.131
Token's
Token ID338
Feature activation+2.09e+00

Pos logit contributions
cks+0.059
 lair+0.059
 mess+0.055
 pads+0.052
llers+0.052
Neg logit contributions
 help-0.107
 reward-0.105
ar-0.102
 plunge-0.101
less-0.098
Token shoulder
Token ID8163
Feature activation+1.11e-01

Pos logit contributions
 lair+0.137
cks+0.136
 mess+0.129
 pads+0.127
 floor+0.124
Neg logit contributions
 help-0.186
 reward-0.184
ar-0.180
 plunge-0.178
bye-0.173
Token.
Token ID13
Feature activation-1.60e-02

Pos logit contributions
cks+0.009
 lair+0.009
 mess+0.008
 pads+0.008
 floor+0.008
Neg logit contributions
 help-0.008
 reward-0.008
ar-0.008
 plunge-0.008
bye-0.008
Token "
Token ID366
Feature activation+6.79e-02

Pos logit contributions
ar+0.001
 plunge+0.001
 reward+0.001
 help+0.001
bye+0.001
Neg logit contributions
 lair-0.001
 floor-0.001
llers-0.001
 pads-0.001
 mess-0.001
TokenThat
Token ID2504
Feature activation-6.47e-01

Pos logit contributions
 lair+0.005
cks+0.005
 pads+0.005
 mess+0.005
 floor+0.005
Neg logit contributions
 reward-0.005
 help-0.005
ar-0.005
 plunge-0.005
bye-0.005
Token's
Token ID338
Feature activation+2.07e+00

Pos logit contributions
 help+0.061
 reward+0.061
 plunge+0.059
bye+0.058
ar+0.057
Neg logit contributions
cks-0.045
 lair-0.042
 necks-0.036
 pads-0.036
llers-0.035
Token very
Token ID845
Feature activation-1.12e-01

Pos logit contributions
cks+0.156
 mess+0.152
 lair+0.149
 floor+0.145
 frown+0.144
Neg logit contributions
ar-0.158
 plunge-0.158
 reward-0.155
ted-0.155
 help-0.154
Token kind
Token ID1611
Feature activation-9.25e-01

Pos logit contributions
ted+0.010
 plunge+0.010
 reward+0.010
ar+0.010
 help+0.010
Neg logit contributions
 mess-0.007
cks-0.007
 lair-0.007
 floor-0.006
 frown-0.006
Token of
Token ID286
Feature activation-1.01e+00

Pos logit contributions
ar+0.034
 reward+0.034
 plunge+0.033
 help+0.033
Sure+0.031
Neg logit contributions
 lair-0.105
 mess-0.103
 pads-0.103
 floor-0.102
 sizes-0.101
Token you
Token ID345
Feature activation-1.39e+00

Pos logit contributions
ar+0.063
Sure+0.061
 reward+0.060
 plunge+0.059
less+0.058
Neg logit contributions
cks-0.091
 mess-0.090
 lair-0.088
 floor-0.087
 sizes-0.086
Token to
Token ID284
Feature activation+1.02e-01

Pos logit contributions
 reward+0.093
ar+0.091
 plunge+0.089
Sure+0.089
 help+0.089
Neg logit contributions
 lair-0.115
 mess-0.114
cks-0.112
 pads-0.111
 floor-0.111
Token and
Token ID290
Feature activation+2.76e-01

Pos logit contributions
cks+0.076
 lair+0.075
 pads+0.071
llers+0.071
 necks+0.070
Neg logit contributions
 help-0.041
 reward-0.034
less-0.032
ar-0.032
Sure-0.029
Token angry
Token ID7954
Feature activation+6.57e-02

Pos logit contributions
 lair+0.019
cks+0.018
 sizes+0.017
 pads+0.017
llers+0.017
Neg logit contributions
 help-0.025
 reward-0.023
ted-0.023
bye-0.023
 plunge-0.022
Token.
Token ID13
Feature activation+1.36e-01

Pos logit contributions
 lair+0.006
cks+0.006
 pads+0.006
 mess+0.006
 necks+0.006
Neg logit contributions
 help-0.004
 reward-0.004
ar-0.004
bye-0.004
ted-0.004
Token He
Token ID679
Feature activation+2.15e-01

Pos logit contributions
cks+0.013
 mess+0.013
 pads+0.013
 lair+0.013
cking+0.013
Neg logit contributions
 help-0.008
 Bub-0.007
Sure-0.007
ar-0.007
 Enjoy-0.007
Token does
Token ID857
Feature activation+6.30e-02

Pos logit contributions
 lair+0.015
llers+0.015
 sizes+0.015
cks+0.015
opl+0.014
Neg logit contributions
 help-0.018
ted-0.017
 reward-0.016
ar-0.016
 Enjoy-0.015
Token not
Token ID407
Feature activation+3.17e+00

Pos logit contributions
 pads+0.004
 sizes+0.004
 lair+0.004
 necks+0.004
cks+0.004
Neg logit contributions
 help-0.006
 reward-0.006
ar-0.005
Sure-0.005
f-0.005
Token like
Token ID588
Feature activation-3.40e-01

Pos logit contributions
 lair+0.282
cks+0.274
llers+0.265
 necks+0.260
 floor+0.258
Neg logit contributions
 help-0.235
 reward-0.221
ar-0.203
bye-0.200
f-0.198
Token Mia
Token ID32189
Feature activation+1.59e+00

Pos logit contributions
 help+0.020
 reward+0.017
ted+0.016
 Bub+0.015
 plunge+0.015
Neg logit contributions
 lair-0.034
cks-0.033
 pads-0.033
 necks-0.032
llers-0.031
Token.
Token ID13
Feature activation+1.43e-01

Pos logit contributions
 lair+0.132
cks+0.128
llers+0.128
 sizes+0.117
 necks+0.117
Neg logit contributions
 help-0.122
 reward-0.109
f-0.107
ar-0.105
less-0.104
Token He
Token ID679
Feature activation+2.31e-01

Pos logit contributions
 lair+0.013
 mess+0.013
cks+0.013
 pads+0.013
llers+0.012
Neg logit contributions
 help-0.009
ar-0.008
Sure-0.008
 Bub-0.008
 reward-0.008
Token does
Token ID857
Feature activation+7.30e-02

Pos logit contributions
 lair+0.017
llers+0.016
cks+0.015
 sizes+0.015
 pads+0.015
Neg logit contributions
 help-0.020
ar-0.019
ted-0.018
 reward-0.018
f-0.017
Token's
Token ID338
Feature activation+2.34e+00

Pos logit contributions
 mess+0.014
cks+0.013
 pads+0.013
 lair+0.012
 floor+0.012
Neg logit contributions
 reward-0.060
ar-0.060
 help-0.059
 plunge-0.059
less-0.059
Token go
Token ID467
Feature activation+3.52e-01

Pos logit contributions
 lair+0.241
cks+0.239
 mess+0.229
 pads+0.229
 floor+0.225
Neg logit contributions
 help-0.125
 reward-0.121
ar-0.113
 plunge-0.113
bye-0.107
Token play
Token ID711
Feature activation-9.57e-02

Pos logit contributions
 mess+0.021
 lair+0.021
 pads+0.021
cks+0.020
 floor+0.019
Neg logit contributions
 reward-0.029
ted-0.028
less-0.028
 plunge-0.028
ar-0.027
Token.
Token ID13
Feature activation+2.60e-01

Pos logit contributions
 reward+0.007
 plunge+0.007
less+0.007
bye+0.007
ted+0.007
Neg logit contributions
 lair-0.007
 mess-0.007
cks-0.007
 web-0.007
 floor-0.007
Token Let
Token ID3914
Feature activation+4.55e-01

Pos logit contributions
 lair+0.018
 pads+0.018
cks+0.018
 necks+0.017
 sizes+0.017
Neg logit contributions
 plunge-0.021
 reward-0.021
ar-0.020
less-0.020
 help-0.020
Token's
Token ID338
Feature activation+2.32e+00

Pos logit contributions
 mess+0.012
cks+0.011
 pads+0.011
 lair+0.010
 floor+0.010
Neg logit contributions
 reward-0.057
 help-0.057
ar-0.057
 plunge-0.056
less-0.056
Token make
Token ID787
Feature activation+1.25e+00

Pos logit contributions
 lair+0.237
cks+0.235
 mess+0.225
 pads+0.224
 floor+0.221
Neg logit contributions
 help-0.125
 reward-0.122
ar-0.114
 plunge-0.114
bye-0.108
Token a
Token ID257
Feature activation-2.44e-01

Pos logit contributions
 lair+0.126
cks+0.125
 mess+0.122
 pads+0.120
 floor+0.119
Neg logit contributions
 help-0.065
 reward-0.065
ar-0.062
 plunge-0.061
bye-0.058
Token snow
Token ID6729
Feature activation+5.95e-01

Pos logit contributions
 help+0.022
ar+0.022
 reward+0.022
 plunge+0.021
ted+0.021
Neg logit contributions
 lair-0.015
 mess-0.015
cks-0.015
 floor-0.014
 pads-0.014
Tokenman
Token ID805
Feature activation+4.57e-01

Pos logit contributions
 lair+0.046
cks+0.045
 mess+0.044
 pads+0.043
 floor+0.042
Neg logit contributions
 help-0.045
 reward-0.045
ar-0.044
 plunge-0.044
Sure-0.042
Token friend
Token ID1545
Feature activation-1.03e+00

Pos logit contributions
 lair+0.040
cks+0.040
 mess+0.038
 pads+0.038
 floor+0.037
Neg logit contributions
 help-0.031
 reward-0.030
ar-0.029
 plunge-0.028
bye-0.028
Token for
Token ID329
Feature activation-1.60e+00

Pos logit contributions
 lair+0.052
cks+0.051
 mess+0.051
 pads+0.050
 floor+0.049
Neg logit contributions
 reward-0.030
 help-0.029
ar-0.029
 plunge-0.028
Sure-0.027
Token the
Token ID262
Feature activation+1.32e-01

Pos logit contributions
Sure+0.114
ar+0.111
less+0.108
f+0.108
ted+0.103
Neg logit contributions
 mess-0.137
 lair-0.133
 floor-0.132
cks-0.127
 frown-0.120
Token surprise
Token ID5975
Feature activation+1.70e+00

Pos logit contributions
 mess+0.011
 floor+0.011
 lair+0.010
 pads+0.010
 frown+0.009
Neg logit contributions
ar-0.010
ted-0.009
Sure-0.009
less-0.009
f-0.009
Token.
Token ID13
Feature activation-2.80e-02

Pos logit contributions
 lair+0.144
cks+0.141
 mess+0.138
 pads+0.136
 floor+0.134
Neg logit contributions
 help-0.116
 reward-0.115
ar-0.113
 plunge-0.111
Sure-0.107
Token Come
Token ID7911
Feature activation-8.98e-01

Pos logit contributions
iest+0.002
 plunge+0.002
ar+0.002
less+0.002
 reward+0.002
Neg logit contributions
 lair-0.002
 pads-0.002
llers-0.002
 floor-0.002
cks-0.002
Token to
Token ID284
Feature activation+8.36e-02

Pos logit contributions
iest+0.067
bye+0.065
 plunge+0.065
 reward+0.063
ted+0.062
Neg logit contributions
 mess-0.062
 lair-0.061
 frown-0.061
 neck-0.060
cks-0.059
Token the
Token ID262
Feature activation+1.24e-01

Pos logit contributions
 lair+0.007
 mess+0.007
cks+0.007
 floor+0.007
 frown+0.007
Neg logit contributions
ar-0.006
less-0.005
f-0.005
 plunge-0.005
iest-0.005
Token centre
Token ID7372
Feature activation-1.37e+00

Pos logit contributions
 floor+0.010
 mess+0.010
 lair+0.009
 frown+0.008
 pads+0.008
Neg logit contributions
ar-0.010
f-0.009
ted-0.009
less-0.009
 help-0.009
Token of
Token ID286
Feature activation-1.00e+00

Pos logit contributions
 help+0.057
 reward+0.054
ar+0.049
 plunge+0.048
bye+0.047
Neg logit contributions
cks-0.159
 lair-0.157
 pads-0.151
 necks-0.149
llers-0.149
Token the
Token ID262
Feature activation+8.41e-02

Pos logit contributions
ar+0.066
 reward+0.064
Sure+0.063
 plunge+0.062
 help+0.062
Neg logit contributions
 lair-0.088
 mess-0.087
cks-0.086
 floor-0.083
 pads-0.079
Token room
Token ID2119
Feature activation+1.86e+00

Pos logit contributions
 mess+0.007
 lair+0.007
 floor+0.007
 pads+0.006
 goals+0.006
Neg logit contributions
ar-0.006
 help-0.006
ted-0.006
 plunge-0.006
less-0.006
Token The
Token ID383
Feature activation-8.51e-01

Pos logit contributions
cks+0.049
 lair+0.048
 mess+0.048
 pads+0.046
 floor+0.046
Neg logit contributions
 help-0.032
 reward-0.032
ar-0.031
Sure-0.030
bye-0.029
Token moral
Token ID6573
Feature activation-9.36e-01

Pos logit contributions
ar+0.064
 plunge+0.063
ted+0.061
less+0.059
bye+0.058
Neg logit contributions
 lair-0.069
 mess-0.069
 floor-0.066
 frown-0.064
 web-0.062
Token of
Token ID286
Feature activation-5.80e-01

Pos logit contributions
 plunge+0.044
 reward+0.042
ar+0.041
f+0.039
ted+0.039
Neg logit contributions
 lair-0.096
 floor-0.093
 mess-0.093
 goals-0.090
 pads-0.090
Token the
Token ID262
Feature activation+4.86e-01

Pos logit contributions
 reward+0.033
ar+0.033
 help+0.032
 plunge+0.032
Sure+0.030
Neg logit contributions
cks-0.056
 lair-0.055
 mess-0.053
 floor-0.052
 pads-0.051
Token story
Token ID1621
Feature activation+1.78e+00

Pos logit contributions
 mess+0.025
 lair+0.024
 floor+0.023
 frown+0.022
 web+0.021
Neg logit contributions
ar-0.049
ted-0.049
less-0.048
 plunge-0.048
bye-0.047
Token is
Token ID318
Feature activation+8.74e-01

Pos logit contributions
 lair+0.128
cks+0.127
 pads+0.119
 mess+0.118
 necks+0.116
Neg logit contributions
 help-0.148
 reward-0.147
ar-0.143
 plunge-0.139
bye-0.138
Token that
Token ID326
Feature activation-1.65e-01

Pos logit contributions
 lair+0.076
cks+0.073
 mess+0.072
 floor+0.070
llers+0.069
Neg logit contributions
 reward-0.056
 plunge-0.055
ar-0.055
ted-0.054
 help-0.053
Token it
Token ID340
Feature activation-7.86e-01

Pos logit contributions
 help+0.010
 reward+0.009
 plunge+0.009
bye+0.008
ar+0.008
Neg logit contributions
cks-0.016
 lair-0.016
 pads-0.016
 necks-0.015
 mess-0.015
Token's
Token ID338
Feature activation+2.03e+00

Pos logit contributions
 help+0.077
 reward+0.072
ar+0.069
 plunge+0.068
f+0.065
Neg logit contributions
 lair-0.052
cks-0.050
 necks-0.047
 pads-0.046
 neck-0.045
Token always
Token ID1464
Feature activation+8.82e-01

Pos logit contributions
 lair+0.131
cks+0.129
 mess+0.123
 pads+0.120
 floor+0.118
Neg logit contributions
 help-0.182
 reward-0.181
ar-0.178
 plunge-0.176
bye-0.171
Token best
Token ID1266
Feature activation+1.35e+00

Pos logit contributions
 lair+0.079
cks+0.078
 mess+0.074
 pads+0.074
 floor+0.073
Neg logit contributions
 help-0.059
 reward-0.058
ar-0.055
 plunge-0.054
bye-0.052
Token until
Token ID1566
Feature activation+2.21e-01

Pos logit contributions
 help+0.020
ar+0.018
 reward+0.018
 plunge+0.017
less+0.017
Neg logit contributions
 lair-0.034
cks-0.033
 pads-0.032
 mess-0.032
 floor-0.031
Token,
Token ID11
Feature activation-6.10e-02

Pos logit contributions
 mess+0.016
 floor+0.016
 frown+0.015
 lair+0.014
llers+0.014
Neg logit contributions
ar-0.016
 plunge-0.016
bye-0.016
Sure-0.015
f-0.015
Token eventually
Token ID4191
Feature activation+9.91e-02

Pos logit contributions
ar+0.004
 reward+0.004
 help+0.004
 plunge+0.004
bye+0.004
Neg logit contributions
 lair-0.005
 mess-0.005
cks-0.005
 floor-0.005
 pads-0.005
Token,
Token ID11
Feature activation-7.52e-02

Pos logit contributions
 mess+0.009
 lair+0.009
 floor+0.008
 frown+0.008
cks+0.008
Neg logit contributions
Sure-0.006
 plunge-0.006
 help-0.006
bye-0.006
 reward-0.006
Token a
Token ID257
Feature activation-2.69e-01

Pos logit contributions
ar+0.004
 help+0.004
 reward+0.004
 plunge+0.004
bye+0.004
Neg logit contributions
 lair-0.007
 mess-0.007
cks-0.007
 floor-0.007
 pads-0.007
Token flock
Token ID29556
Feature activation-4.13e-01

Pos logit contributions
ted+0.022
ar+0.022
 plunge+0.021
 help+0.021
 reward+0.021
Neg logit contributions
 mess-0.019
 lair-0.019
 floor-0.018
 frown-0.018
 web-0.017
Token of
Token ID286
Feature activation-7.24e-01

Pos logit contributions
 reward+0.019
 help+0.017
ar+0.016
 plunge+0.016
Sure+0.016
Neg logit contributions
cks-0.043
 lair-0.042
 floor-0.042
 mess-0.041
 pads-0.041
Token birds
Token ID10087
Feature activation+7.35e-01

Pos logit contributions
ar+0.062
 reward+0.061
less+0.057
 help+0.057
Sure+0.057
Neg logit contributions
 lair-0.050
 mess-0.049
cks-0.048
 floor-0.047
 pads-0.047
Token flew
Token ID13112
Feature activation+3.59e-01

Pos logit contributions
 lair+0.057
cks+0.055
 mess+0.053
 floor+0.052
 pads+0.052
Neg logit contributions
 help-0.057
 reward-0.055
ar-0.055
 plunge-0.055
ted-0.053
Token out
Token ID503
Feature activation-9.66e-01

Pos logit contributions
 pads+0.028
cks+0.028
 lair+0.028
 mess+0.027
 necks+0.026
Neg logit contributions
 reward-0.027
 help-0.025
ar-0.025
 plunge-0.024
 Enjoy-0.023
Token of
Token ID286
Feature activation-7.98e-01

Pos logit contributions
 help+0.043
 reward+0.043
ar+0.041
 plunge+0.040
bye+0.037
Neg logit contributions
 lair-0.105
cks-0.104
 mess-0.102
 pads-0.100
 floor-0.100
Token could
Token ID714
Feature activation-3.64e-01

Pos logit contributions
 reward+0.045
 help+0.045
ar+0.044
 plunge+0.044
bye+0.042
Neg logit contributions
 lair-0.043
cks-0.043
 mess-0.041
 pads-0.041
 floor-0.040
Token have
Token ID423
Feature activation+2.26e-01

Pos logit contributions
ar+0.023
 reward+0.022
 plunge+0.022
ted+0.022
 help+0.022
Neg logit contributions
 lair-0.031
cks-0.031
 mess-0.030
 pads-0.030
 floor-0.029
Token a
Token ID257
Feature activation-4.26e-01

Pos logit contributions
cks+0.020
 lair+0.020
 mess+0.020
 pads+0.020
 necks+0.019
Neg logit contributions
 reward-0.013
ar-0.013
 help-0.013
 plunge-0.013
bye-0.013
Token turn
Token ID1210
Feature activation+5.38e-01

Pos logit contributions
ar+0.031
 help+0.031
 reward+0.030
 plunge+0.030
ted+0.030
Neg logit contributions
 lair-0.034
 mess-0.033
cks-0.032
 pads-0.032
 floor-0.031
Token and
Token ID290
Feature activation+1.82e-01

Pos logit contributions
 lair+0.048
cks+0.047
 mess+0.046
 pads+0.046
 floor+0.045
Neg logit contributions
 reward-0.034
 help-0.033
ar-0.032
 plunge-0.032
bye-0.031
Token Jane
Token ID12091
Feature activation+1.52e+00

Pos logit contributions
 lair+0.016
 mess+0.016
cks+0.016
 pads+0.016
 floor+0.015
Neg logit contributions
ar-0.011
 reward-0.011
 plunge-0.011
less-0.010
 help-0.010
Token said
Token ID531
Feature activation-1.53e-01

Pos logit contributions
 lair+0.136
cks+0.134
 mess+0.128
 floor+0.125
 pads+0.123
Neg logit contributions
 help-0.103
 reward-0.101
ar-0.099
Sure-0.093
f-0.092
Token yes
Token ID3763
Feature activation-1.47e+00

Pos logit contributions
 plunge+0.007
 reward+0.006
ar+0.006
bye+0.006
ted+0.006
Neg logit contributions
 neck-0.015
 pads-0.015
 frown-0.015
 goals-0.015
 lair-0.015
Token.
Token ID13
Feature activation+1.53e-01

Pos logit contributions
 help+0.099
 reward+0.098
ar+0.095
 plunge+0.095
bye+0.090
Neg logit contributions
 lair-0.126
cks-0.124
 mess-0.121
 pads-0.119
 floor-0.118
Token
Token ID220
Feature activation-4.47e-01

Pos logit contributions
 lair+0.012
cks+0.012
 mess+0.012
 pads+0.012
 floor+0.012
Neg logit contributions
 help-0.011
 reward-0.011
ar-0.011
 plunge-0.010
bye-0.010
Token
Token ID198
Feature activation+1.63e-01

Pos logit contributions
 help+0.030
 reward+0.029
ar+0.029
 plunge+0.028
bye+0.027
Neg logit contributions
 lair-0.039
cks-0.038
 mess-0.037
 pads-0.037
 floor-0.036
Token it
Token ID340
Feature activation-1.06e+00

Pos logit contributions
 lair+0.060
 floor+0.057
 mess+0.057
cks+0.057
 pads+0.056
Neg logit contributions
 reward-0.049
ar-0.049
 help-0.049
 plunge-0.048
less-0.047
Token was
Token ID373
Feature activation-1.54e-01

Pos logit contributions
 help+0.062
 reward+0.061
ar+0.061
 plunge+0.058
bye+0.055
Neg logit contributions
cks-0.104
 lair-0.102
 pads-0.097
 necks-0.097
llers-0.094
Token too
Token ID1165
Feature activation-3.38e-01

Pos logit contributions
 help+0.008
 reward+0.007
less+0.006
 plunge+0.006
ar+0.006
Neg logit contributions
cks-0.017
 necks-0.016
 lair-0.016
 wrists-0.016
 pads-0.016
Token wind
Token ID2344
Feature activation+1.29e-01

Pos logit contributions
 help+0.026
 reward+0.025
ar+0.025
 plunge+0.024
bye+0.024
Neg logit contributions
 lair-0.026
cks-0.026
 mess-0.025
 pads-0.025
 floor-0.024
Tokeny
Token ID88
Feature activation+6.04e-01

Pos logit contributions
 lair+0.010
cks+0.009
 mess+0.009
 necks+0.009
 floor+0.009
Neg logit contributions
 help-0.010
less-0.010
ar-0.010
 reward-0.010
 plunge-0.010
Token and
Token ID290
Feature activation-1.40e-01

Pos logit contributions
 lair+0.057
cks+0.057
 mess+0.054
 necks+0.053
 floor+0.053
Neg logit contributions
 help-0.036
 reward-0.035
ar-0.034
 plunge-0.033
less-0.032
Token the
Token ID262
Feature activation-2.32e-02

Pos logit contributions
 help+0.010
 reward+0.009
 plunge+0.008
ted+0.008
ar+0.008
Neg logit contributions
 lair-0.013
cks-0.013
 necks-0.012
 pads-0.012
opl-0.011
Token waves
Token ID9813
Feature activation-3.49e-01

Pos logit contributions
ar+0.002
 reward+0.002
Sure+0.002
ted+0.002
less+0.002
Neg logit contributions
 lair-0.002
 floor-0.002
 mess-0.002
 neck-0.002
 frown-0.002
Token were
Token ID547
Feature activation-2.37e-01

Pos logit contributions
 reward+0.025
 help+0.024
 plunge+0.024
ar+0.024
Sure+0.024
Neg logit contributions
cks-0.027
 mess-0.027
 lair-0.027
 pads-0.027
 floor-0.026
Token too
Token ID1165
Feature activation-3.32e-01

Pos logit contributions
 help+0.017
less+0.015
 reward+0.015
 plunge+0.014
ted+0.013
Neg logit contributions
 lair-0.021
cks-0.020
 necks-0.020
 pads-0.019
 wrists-0.019
Token big
Token ID1263
Feature activation+1.23e+00

Pos logit contributions
 help+0.026
 reward+0.025
ar+0.025
 plunge+0.024
bye+0.024
Neg logit contributions
 lair-0.025
cks-0.025
 mess-0.024
 pads-0.024
 floor-0.023
Token mom
Token ID1995
Feature activation-4.76e-03

Pos logit contributions
 mess+0.071
 floor+0.068
 lair+0.068
 frown+0.067
 web+0.064
Neg logit contributions
ar-0.083
 plunge-0.077
ted-0.074
Sure-0.073
f-0.073
Tokenmy
Token ID1820
Feature activation-2.43e-01

Pos logit contributions
 help+0.000
 reward+0.000
ar+0.000
bye+0.000
Sure+0.000
Neg logit contributions
 lair-0.000
cks-0.000
 necks-0.000
 pads-0.000
 goals-0.000
Token asked
Token ID1965
Feature activation+6.99e-01

Pos logit contributions
 help+0.018
 reward+0.017
ar+0.017
 plunge+0.017
bye+0.016
Neg logit contributions
 lair-0.020
cks-0.020
 mess-0.019
 pads-0.019
 floor-0.018
Token her
Token ID607
Feature activation+9.67e-01

Pos logit contributions
 mess+0.043
 floor+0.038
 frown+0.037
 pads+0.032
cks+0.031
Neg logit contributions
 plunge-0.071
ar-0.068
 reward-0.067
iest-0.064
bye-0.063
Token to
Token ID284
Feature activation-8.72e-02

Pos logit contributions
 mess+0.073
 lair+0.071
 frown+0.069
 floor+0.067
 neck+0.064
Neg logit contributions
ar-0.076
 plunge-0.074
ted-0.073
bye-0.070
 reward-0.065
Token help
Token ID1037
Feature activation-2.10e+00

Pos logit contributions
ted+0.007
ar+0.007
ed+0.006
less+0.006
rav+0.006
Neg logit contributions
 floor-0.007
 mess-0.006
 frown-0.006
 neck-0.006
 lair-0.006
Token with
Token ID351
Feature activation+5.22e-01

Pos logit contributions
ar+0.186
 plunge+0.182
bye+0.181
 reward+0.176
ted+0.174
Neg logit contributions
 mess-0.142
 floor-0.140
cks-0.129
 frown-0.125
 neck-0.121
Token the
Token ID262
Feature activation-1.95e-01

Pos logit contributions
 mess+0.039
 floor+0.036
 frown+0.035
cks+0.033
 uneven+0.033
Neg logit contributions
ar-0.047
bye-0.045
 plunge-0.043
Sure-0.041
rav-0.040
Token grocery
Token ID16918
Feature activation-1.47e-01

Pos logit contributions
ar+0.015
bye+0.014
 reward+0.014
ted+0.014
Sure+0.014
Neg logit contributions
 mess-0.015
 lair-0.015
 floor-0.015
 frown-0.014
cks-0.014
Token shopping
Token ID9735
Feature activation-1.67e+00

Pos logit contributions
ar+0.011
 plunge+0.011
bye+0.010
ted+0.010
 reward+0.010
Neg logit contributions
 mess-0.012
 floor-0.012
 lair-0.011
 pads-0.011
cks-0.011
Token.
Token ID13
Feature activation-3.06e-01

Pos logit contributions
 plunge+0.146
ar+0.146
ted+0.141
bye+0.137
 Bub+0.136
Neg logit contributions
 floor-0.112
 mess-0.109
 lair-0.095
 pads-0.093
 sizes-0.091
Token"
Token ID1
Feature activation-9.04e-01

Pos logit contributions
cks+0.105
 lair+0.102
 mess+0.099
llers+0.097
 pads+0.097
Neg logit contributions
 help-0.076
 reward-0.073
ar-0.072
 plunge-0.070
bye-0.069
Token
Token ID220
Feature activation-7.05e-01

Pos logit contributions
 reward+0.065
 help+0.064
ar+0.063
 plunge+0.063
Sure+0.058
Neg logit contributions
 lair-0.073
 pads-0.072
cks-0.071
 mess-0.069
 floor-0.069
Token
Token ID198
Feature activation-1.59e-01

Pos logit contributions
 help+0.048
 reward+0.048
ar+0.046
 plunge+0.046
bye+0.044
Neg logit contributions
 lair-0.060
cks-0.059
 mess-0.057
 pads-0.057
 floor-0.056
Token
Token ID198
Feature activation-1.07e-01

Pos logit contributions
 reward+0.013
 help+0.013
 plunge+0.013
less+0.013
ar+0.012
Neg logit contributions
 lair-0.011
cks-0.011
 floor-0.010
 pads-0.010
llers-0.010
TokenJust
Token ID5703
Feature activation-1.33e+00

Pos logit contributions
 help+0.009
 reward+0.009
ar+0.009
 plunge+0.009
Sure+0.009
Neg logit contributions
 lair-0.008
cks-0.007
 mess-0.007
 pads-0.007
 floor-0.007
Token then
Token ID788
Feature activation-1.74e+00

Pos logit contributions
 reward+0.094
ar+0.090
less+0.090
 plunge+0.089
Sure+0.089
Neg logit contributions
 lair-0.106
cks-0.106
 floor-0.102
 pads-0.102
 neck-0.101
Token,
Token ID11
Feature activation-1.83e-01

Pos logit contributions
 reward+0.121
 plunge+0.120
bye+0.118
Sure+0.116
less+0.112
Neg logit contributions
 lair-0.139
 frown-0.136
 neck-0.128
cks-0.123
 mess-0.123
Token the
Token ID262
Feature activation+1.02e-01

Pos logit contributions
rav+0.018
ar+0.018
bye+0.018
less+0.017
ay+0.017
Neg logit contributions
 frown-0.009
 neck-0.008
 not-0.007
 also-0.007
 mess-0.007
Token rabbit
Token ID22746
Feature activation-8.41e-01

Pos logit contributions
 lair+0.008
 floor+0.008
 mess+0.008
 pads+0.008
 neck+0.007
Neg logit contributions
ar-0.008
 reward-0.008
f-0.007
less-0.007
 help-0.007
Token heard
Token ID2982
Feature activation-8.69e-03

Pos logit contributions
 help+0.061
ar+0.060
 reward+0.059
 plunge+0.056
ted+0.055
Neg logit contributions
cks-0.071
 lair-0.069
 mess-0.066
 floor-0.064
 sizes-0.064
Token a
Token ID257
Feature activation-5.52e-01

Pos logit contributions
 reward+0.001
 plunge+0.001
ar+0.001
Sure+0.001
bye+0.001
Neg logit contributions
 lair-0.001
 frown-0.001
 floor-0.001
 mess-0.001
cks-0.001
Token it
Token ID340
Feature activation-8.98e-01

Pos logit contributions
bye+0.026
ar+0.025
 plunge+0.024
less+0.024
ival+0.023
Neg logit contributions
 frown-0.020
cks-0.020
 neck-0.019
 wrists-0.018
 teeth-0.018
Token with
Token ID351
Feature activation+6.23e-01

Pos logit contributions
 reward+0.065
ar+0.064
 help+0.064
 plunge+0.062
bye+0.060
Neg logit contributions
 lair-0.071
 mess-0.069
cks-0.069
 floor-0.069
 pads-0.068
Token both
Token ID1111
Feature activation-1.33e-01

Pos logit contributions
cks+0.047
 mess+0.046
 frown+0.045
 lair+0.045
 pads+0.045
Neg logit contributions
ar-0.049
 reward-0.046
 plunge-0.046
bye-0.045
Sure-0.044
Token hands
Token ID2832
Feature activation-6.49e-01

Pos logit contributions
 help+0.010
 reward+0.009
 plunge+0.008
less+0.008
Sure+0.008
Neg logit contributions
 lair-0.013
cks-0.012
llers-0.011
 pads-0.011
opl-0.011
Token.
Token ID13
Feature activation+2.70e-02

Pos logit contributions
 help+0.040
 reward+0.039
ar+0.037
 plunge+0.037
less+0.036
Neg logit contributions
 lair-0.061
cks-0.060
 mess-0.058
 pads-0.057
 floor-0.056
Token It
Token ID632
Feature activation-1.77e+00

Pos logit contributions
 lair+0.002
cks+0.002
 mess+0.002
 pads+0.002
 floor+0.002
Neg logit contributions
 help-0.002
 reward-0.002
ar-0.002
 plunge-0.002
Sure-0.002
Token was
Token ID373
Feature activation+1.94e-01

Pos logit contributions
 help+0.108
 reward+0.100
ar+0.094
Sure+0.092
ted+0.091
Neg logit contributions
 lair-0.183
cks-0.180
opl-0.170
 pads-0.164
 goals-0.161
Token too
Token ID1165
Feature activation-1.12e-01

Pos logit contributions
 lair+0.018
cks+0.017
 pads+0.016
 necks+0.016
llers+0.016
Neg logit contributions
 help-0.013
 reward-0.013
less-0.011
 plunge-0.011
ar-0.011
Token heavy
Token ID4334
Feature activation+2.86e-01

Pos logit contributions
 help+0.008
 reward+0.007
less+0.007
ar+0.007
bye+0.007
Neg logit contributions
 lair-0.010
cks-0.010
 pads-0.009
 necks-0.009
llers-0.009
Token for
Token ID329
Feature activation-1.54e+00

Pos logit contributions
 lair+0.030
cks+0.030
 mess+0.029
 pads+0.029
 floor+0.029
Neg logit contributions
 reward-0.013
 help-0.013
ar-0.013
 plunge-0.013
Sure-0.012
Token her
Token ID607
Feature activation+1.24e+00

Pos logit contributions
ar+0.144
bye+0.141
Sure+0.139
iest+0.134
 plunge+0.134
Neg logit contributions
 mess-0.103
 floor-0.099
cks-0.094
 neck-0.093
 frown-0.091
Token pulled
Token ID5954
Feature activation-6.58e-01

Pos logit contributions
 lair+0.017
cks+0.017
 mess+0.016
 pads+0.016
 floor+0.016
Neg logit contributions
 help-0.017
 reward-0.016
ar-0.016
 plunge-0.016
bye-0.015
Token a
Token ID257
Feature activation-6.18e-01

Pos logit contributions
 reward+0.041
bye+0.041
ar+0.040
Sure+0.040
 help+0.038
Neg logit contributions
 mess-0.059
 lair-0.058
cks-0.058
 floor-0.057
 pads-0.056
Token giant
Token ID6175
Feature activation+1.96e+00

Pos logit contributions
 help+0.048
ar+0.048
 reward+0.047
bye+0.047
ted+0.047
Neg logit contributions
 mess-0.045
 lair-0.045
 floor-0.043
cks-0.043
 pads-0.041
Token carrot
Token ID40562
Feature activation-1.28e-01

Pos logit contributions
cks+0.155
 lair+0.152
 pads+0.146
llers+0.145
 necks+0.143
Neg logit contributions
 help-0.153
 reward-0.150
ar-0.141
 plunge-0.140
less-0.132
Token.
Token ID13
Feature activation-2.31e-01

Pos logit contributions
 help+0.008
 reward+0.008
ar+0.008
 plunge+0.008
less+0.007
Neg logit contributions
cks-0.012
 lair-0.012
 mess-0.011
llers-0.011
 pads-0.011
Token It
Token ID632
Feature activation-2.10e+00

Pos logit contributions
 reward+0.019
 help+0.018
ar+0.018
 plunge+0.018
less+0.018
Neg logit contributions
 lair-0.016
cks-0.016
 floor-0.015
 pads-0.015
llers-0.015
Token was
Token ID373
Feature activation-3.27e-02

Pos logit contributions
 help+0.129
 reward+0.124
ar+0.123
 plunge+0.115
ted+0.113
Neg logit contributions
 lair-0.200
cks-0.196
 pads-0.186
 mess-0.186
 necks-0.185
Token bigger
Token ID5749
Feature activation+3.90e-01

Pos logit contributions
 help+0.002
 reward+0.002
 plunge+0.002
ar+0.002
less+0.002
Neg logit contributions
cks-0.003
 lair-0.003
 pads-0.003
llers-0.003
 mess-0.003
Token than
Token ID621
Feature activation+2.43e+00

Pos logit contributions
cks+0.031
llers+0.028
 lair+0.027
opl+0.026
 frown+0.026
Neg logit contributions
 help-0.030
ar-0.029
 reward-0.029
less-0.028
ted-0.025
Token the
Token ID262
Feature activation+6.87e-02

Pos logit contributions
 lair+0.202
cks+0.194
 floor+0.194
 sizes+0.191
 mess+0.189
Neg logit contributions
Sure-0.168
ar-0.168
bye-0.164
 reward-0.163
 help-0.154
Token other
Token ID584
Feature activation-5.30e-01

Pos logit contributions
cks+0.006
 lair+0.005
llers+0.005
 pads+0.005
 mess+0.005
Neg logit contributions
 help-0.005
 reward-0.005
 plunge-0.005
ar-0.005
bye-0.004
Token She
Token ID1375
Feature activation+3.03e-01

Pos logit contributions
 help+0.020
 reward+0.020
ar+0.019
 plunge+0.019
Sure+0.018
Neg logit contributions
 lair-0.020
cks-0.020
 mess-0.019
 pads-0.019
 necks-0.018
Token saw
Token ID2497
Feature activation-9.74e-02

Pos logit contributions
 lair+0.024
cks+0.024
 mess+0.022
 pads+0.022
llers+0.021
Neg logit contributions
 help-0.024
 reward-0.023
ar-0.023
bye-0.022
Sure-0.022
Token a
Token ID257
Feature activation-6.27e-01

Pos logit contributions
 reward+0.005
ar+0.005
 help+0.005
 plunge+0.005
bye+0.005
Neg logit contributions
 lair-0.010
cks-0.010
 mess-0.010
 floor-0.009
 pads-0.009
Token plane
Token ID6614
Feature activation-2.14e-01

Pos logit contributions
ar+0.050
 help+0.050
 reward+0.049
 plunge+0.049
ted+0.048
Neg logit contributions
 mess-0.045
 lair-0.045
 floor-0.043
cks-0.042
 frown-0.041
Token pass
Token ID1208
Feature activation+2.38e-01

Pos logit contributions
 help+0.011
 reward+0.010
ar+0.010
bye+0.010
 plunge+0.010
Neg logit contributions
 lair-0.022
 pads-0.021
cks-0.021
 necks-0.021
 mess-0.021
Token by
Token ID416
Feature activation-1.66e+00

Pos logit contributions
cks+0.020
 pads+0.019
 lair+0.019
 floor+0.019
 mess+0.019
Neg logit contributions
 reward-0.017
 help-0.017
ar-0.016
 plunge-0.016
Sure-0.015
Token in
Token ID287
Feature activation-1.74e-01

Pos logit contributions
 help+0.116
 reward+0.114
ar+0.110
 plunge+0.109
bye+0.105
Neg logit contributions
 lair-0.140
cks-0.138
 mess-0.133
 pads-0.131
 floor-0.129
Token the
Token ID262
Feature activation-1.28e-01

Pos logit contributions
 reward+0.010
 help+0.010
ar+0.010
 plunge+0.010
bye+0.009
Neg logit contributions
 lair-0.016
cks-0.016
 mess-0.016
 floor-0.016
 pads-0.016
Token sky
Token ID6766
Feature activation+1.62e-01

Pos logit contributions
ar+0.011
 reward+0.011
 help+0.011
 plunge+0.011
less+0.011
Neg logit contributions
 lair-0.008
 mess-0.008
 floor-0.008
cks-0.008
 pads-0.008
Token one
Token ID530
Feature activation-7.97e-01

Pos logit contributions
 lair+0.017
cks+0.017
 mess+0.016
llers+0.016
 pads+0.015
Neg logit contributions
 help-0.008
bye-0.008
 reward-0.008
less-0.007
ar-0.007
Token day
Token ID1110
Feature activation-1.18e+00

Pos logit contributions
ar+0.044
 help+0.044
 reward+0.044
 plunge+0.043
ted+0.041
Neg logit contributions
cks-0.076
 lair-0.076
 mess-0.074
 pads-0.073
 floor-0.073
Token ball
Token ID2613
Feature activation+6.79e-01

Pos logit contributions
 help+0.185
 reward+0.185
ar+0.179
 plunge+0.177
bye+0.171
Neg logit contributions
 lair-0.181
cks-0.180
 mess-0.170
 pads-0.168
 floor-0.165
Token.
Token ID13
Feature activation+7.81e-02

Pos logit contributions
cks+0.063
 lair+0.063
 mess+0.060
 pads+0.059
 necks+0.058
Neg logit contributions
 help-0.042
 reward-0.041
ar-0.040
 plunge-0.039
bye-0.038
Token The
Token ID383
Feature activation-9.87e-01

Pos logit contributions
 lair+0.007
cks+0.007
 mess+0.006
 pads+0.006
 floor+0.006
Neg logit contributions
 reward-0.005
 help-0.005
ar-0.005
 plunge-0.005
bye-0.005
Token 3
Token ID513
Feature activation-1.13e+00

Pos logit contributions
ar+0.072
ted+0.071
 plunge+0.071
 help+0.068
less+0.068
Neg logit contributions
 mess-0.078
 lair-0.078
 floor-0.077
 frown-0.073
 neck-0.073
Token-
Token ID12
Feature activation-3.48e+00

Pos logit contributions
 help+0.060
 reward+0.060
ar+0.057
 plunge+0.056
bye+0.053
Neg logit contributions
 lair-0.113
cks-0.112
 mess-0.109
 pads-0.108
 floor-0.107
Tokenyear
Token ID1941
Feature activation-6.70e+00

Pos logit contributions
 plunge+0.201
 reward+0.199
 help+0.189
ar+0.161
 Bub+0.143
Neg logit contributions
cks-0.324
ents-0.308
 pads-0.304
aron-0.291
thia-0.286
Token-
Token ID12
Feature activation-3.47e+00

Pos logit contributions
 help+0.288
 reward+0.282
 plunge+0.277
ar+0.264
Sure+0.258
Neg logit contributions
 lair-0.737
cks-0.727
 mess-0.713
 floor-0.701
 pads-0.699
Tokenold
Token ID727
Feature activation-1.00e+00

Pos logit contributions
 plunge+0.331
 reward+0.320
 help+0.315
ar+0.289
 Bub+0.283
Neg logit contributions
cks-0.196
 pads-0.190
opl-0.178
ents-0.174
 lair-0.173
Token was
Token ID373
Feature activation+2.42e-01

Pos logit contributions
 plunge+0.079
 help+0.079
 reward+0.078
ar+0.076
less+0.074
Neg logit contributions
 lair-0.072
 mess-0.069
 pads-0.068
cks-0.068
 floor-0.067
Token so
Token ID523
Feature activation+1.02e+00

Pos logit contributions
 lair+0.022
cks+0.022
 mess+0.021
 necks+0.021
 floor+0.020
Neg logit contributions
 help-0.016
 reward-0.015
less-0.014
 plunge-0.014
ar-0.014
Token excited
Token ID6568
Feature activation+5.51e-01

Pos logit contributions
 lair+0.073
cks+0.071
 mess+0.067
 pads+0.067
 floor+0.066
Neg logit contributions
 help-0.086
 reward-0.085
ar-0.082
 plunge-0.082
bye-0.080
Token.
Token ID13
Feature activation-2.20e-01

Pos logit contributions
 help+0.031
 reward+0.030
ar+0.029
 plunge+0.029
bye+0.028
Neg logit contributions
 lair-0.040
cks-0.039
 mess-0.038
 pads-0.037
 floor-0.037
Token She
Token ID1375
Feature activation+4.45e-01

Pos logit contributions
 help+0.017
 reward+0.017
ar+0.016
 plunge+0.016
bye+0.016
Neg logit contributions
cks-0.017
 lair-0.017
 mess-0.016
 pads-0.016
 necks-0.015
Token tried
Token ID3088
Feature activation-6.84e-01

Pos logit contributions
cks+0.038
 lair+0.037
 pads+0.035
 mess+0.035
opl+0.034
Neg logit contributions
 help-0.032
 reward-0.032
ar-0.031
Sure-0.029
f-0.029
Token to
Token ID284
Feature activation+4.95e-03

Pos logit contributions
 help+0.040
 reward+0.036
 plunge+0.033
ar+0.033
less+0.032
Neg logit contributions
cks-0.072
 lair-0.070
 pads-0.067
 sizes-0.064
 necks-0.064
Token sh
Token ID427
Feature activation+5.28e-01

Pos logit contributions
 lair+0.000
cks+0.000
 pads+0.000
 mess+0.000
llers+0.000
Neg logit contributions
 help-0.000
 reward-0.000
 plunge-0.000
ar-0.000
Sure-0.000
Tokenoo
Token ID2238
Feature activation-4.94e+00

Pos logit contributions
cks+0.049
 lair+0.048
 mess+0.046
llers+0.045
 pads+0.044
Neg logit contributions
 reward-0.033
 help-0.033
 plunge-0.031
ar-0.030
bye-0.030
Token him
Token ID683
Feature activation-1.53e+00

Pos logit contributions
 reward+0.320
bye+0.299
 plunge+0.294
 help+0.291
Sure+0.289
Neg logit contributions
cks-0.429
 mess-0.423
 floor-0.413
llers-0.409
 pads-0.404
Token away
Token ID1497
Feature activation-1.99e+00

Pos logit contributions
 help+0.085
 reward+0.084
ar+0.081
 plunge+0.080
bye+0.075
Neg logit contributions
 lair-0.150
cks-0.149
 mess-0.144
 pads-0.142
 floor-0.141
Token,
Token ID11
Feature activation-4.37e-01

Pos logit contributions
 help+0.091
 reward+0.089
ar+0.083
 plunge+0.080
less+0.076
Neg logit contributions
cks-0.221
 lair-0.220
 pads-0.211
 mess-0.207
 necks-0.206
Token but
Token ID475
Feature activation+7.93e-01

Pos logit contributions
 help+0.035
 reward+0.032
Sure+0.029
 plunge+0.029
ar+0.029
Neg logit contributions
cks-0.037
 lair-0.037
 pads-0.034
llers-0.033
opl-0.033
Token it
Token ID340
Feature activation-1.02e+00

Pos logit contributions
 lair+0.064
 mess+0.064
 floor+0.063
 frown+0.062
cks+0.061
Neg logit contributions
ar-0.054
 plunge-0.053
 reward-0.053
bye-0.052
 help-0.051
Token ate
Token ID15063
Feature activation-1.93e-01

Pos logit contributions
cks+0.019
 lair+0.019
 necks+0.018
 pads+0.018
 goals+0.017
Neg logit contributions
 help-0.014
 reward-0.013
ar-0.013
bye-0.012
f-0.012
Token too
Token ID1165
Feature activation-1.72e-01

Pos logit contributions
ar+0.013
 reward+0.012
 help+0.012
 plunge+0.012
bye+0.012
Neg logit contributions
 mess-0.017
cks-0.017
 floor-0.016
 lair-0.016
 pads-0.016
Token much
Token ID881
Feature activation-1.14e-01

Pos logit contributions
 plunge+0.015
ar+0.015
 reward+0.015
 help+0.014
ted+0.014
Neg logit contributions
 mess-0.011
 lair-0.011
cks-0.011
 floor-0.011
 pads-0.010
Token ice
Token ID4771
Feature activation-1.89e+00

Pos logit contributions
ar+0.008
 plunge+0.008
 reward+0.008
 help+0.008
bye+0.008
Neg logit contributions
 mess-0.009
cks-0.009
 floor-0.009
 pads-0.008
 lair-0.008
Token-
Token ID12
Feature activation-3.52e+00

Pos logit contributions
 plunge+0.112
 help+0.111
 reward+0.110
ar+0.103
Sure+0.103
Neg logit contributions
cks-0.175
 mess-0.174
 pads-0.170
 floor-0.170
 lair-0.168
Tokencream
Token ID36277
Feature activation-4.57e+00

Pos logit contributions
 reward+0.264
 help+0.256
 plunge+0.255
mate+0.235
ar+0.230
Neg logit contributions
cks-0.278
llers-0.275
 pads-0.265
 mess-0.254
 necks-0.253
Token already
Token ID1541
Feature activation-1.65e+00

Pos logit contributions
 plunge+0.362
 help+0.354
 reward+0.349
ar+0.348
Sure+0.335
Neg logit contributions
 floor-0.327
 mess-0.325
 pads-0.317
 lair-0.309
 sizes-0.303
Token.
Token ID13
Feature activation+8.49e-03

Pos logit contributions
 reward+0.124
 help+0.123
ar+0.121
 plunge+0.121
Sure+0.119
Neg logit contributions
 mess-0.125
 lair-0.123
 floor-0.122
cks-0.121
 pads-0.120
Token Tim
Token ID5045
Feature activation-2.79e-01

Pos logit contributions
 lair+0.001
cks+0.001
 mess+0.001
 pads+0.001
 floor+0.001
Neg logit contributions
 help-0.000
 reward-0.000
ar-0.000
 plunge-0.000
Sure-0.000
Tokenmy
Token ID1820
Feature activation-2.50e-02

Pos logit contributions
 help+0.015
Sure+0.014
bye+0.014
 reward+0.014
less+0.013
Neg logit contributions
cks-0.028
 lair-0.027
 mesmer-0.025
 web-0.025
 wrists-0.025
Token got
Token ID1392
Feature activation+1.41e-03

Pos logit contributions
 help+0.001
Sure+0.001
 reward+0.001
bye+0.001
ar+0.001
Neg logit contributions
cks-0.003
 lair-0.003
opl-0.003
 sizes-0.003
 pads-0.003
Token kinds
Token ID6982
Feature activation+4.72e-01

Pos logit contributions
 help+0.007
 reward+0.006
ted+0.006
 plunge+0.006
bye+0.006
Neg logit contributions
 lair-0.020
cks-0.020
 necks-0.019
 pads-0.019
 mess-0.019
Token of
Token ID286
Feature activation-1.01e+00

Pos logit contributions
 lair+0.055
cks+0.054
 mess+0.054
 floor+0.053
 pads+0.053
Neg logit contributions
 reward-0.017
 help-0.016
 plunge-0.016
ar-0.016
bye-0.014
Token delicious
Token ID12625
Feature activation-1.69e-01

Pos logit contributions
 help+0.073
 reward+0.072
ar+0.070
 plunge+0.069
bye+0.066
Neg logit contributions
 lair-0.083
cks-0.082
 mess-0.079
 pads-0.078
 floor-0.077
Token ice
Token ID4771
Feature activation-1.86e+00

Pos logit contributions
 help+0.011
 reward+0.011
ar+0.010
 plunge+0.010
bye+0.010
Neg logit contributions
 lair-0.015
cks-0.015
 mess-0.015
 pads-0.014
 floor-0.014
Token-
Token ID12
Feature activation-3.59e+00

Pos logit contributions
 help+0.093
 reward+0.092
 plunge+0.089
ar+0.086
Sure+0.083
Neg logit contributions
cks-0.191
 lair-0.190
 mess-0.189
 pads-0.187
 floor-0.185
Tokencream
Token ID36277
Feature activation-4.56e+00

Pos logit contributions
 reward+0.267
 help+0.259
 plunge+0.257
mate+0.236
ar+0.229
Neg logit contributions
cks-0.286
 pads-0.282
llers-0.282
 necks-0.269
 sizes-0.268
Token flavors
Token ID17361
Feature activation+7.75e-01

Pos logit contributions
 help+0.347
 plunge+0.347
 reward+0.344
ar+0.337
Sure+0.329
Neg logit contributions
 floor-0.337
 mess-0.337
 pads-0.334
 lair-0.334
cks-0.329
Token.
Token ID13
Feature activation+1.88e-01

Pos logit contributions
cks+0.066
 lair+0.066
 pads+0.065
 mess+0.065
 floor+0.064
Neg logit contributions
 help-0.051
 reward-0.051
 plunge-0.050
ar-0.049
Sure-0.047
Token "
Token ID366
Feature activation+1.22e-01

Pos logit contributions
cks+0.015
 lair+0.015
 mess+0.014
 pads+0.014
 necks+0.014
Neg logit contributions
 help-0.014
 reward-0.014
ar-0.013
Sure-0.013
 plunge-0.013
TokenCan
Token ID6090
Feature activation-4.93e-01

Pos logit contributions
 lair+0.011
cks+0.011
 mess+0.010
 pads+0.010
 floor+0.010
Neg logit contributions
 help-0.008
 reward-0.008
ar-0.008
 plunge-0.008
bye-0.007
Token I
Token ID314
Feature activation+1.67e-01

Pos logit contributions
less+0.033
ar+0.033
 reward+0.033
 plunge+0.033
 help+0.031
Neg logit contributions
llers-0.038
 floor-0.037
 mess-0.037
 lair-0.037
 pads-0.037
Token Alex
Token ID4422
Feature activation+2.31e+00

Pos logit contributions
bye+0.007
 reward+0.007
less+0.007
ar+0.007
 plunge+0.007
Neg logit contributions
cks-0.006
 floor-0.005
 mess-0.005
 lair-0.005
 frown-0.005
Token for
Token ID329
Feature activation-1.53e+00

Pos logit contributions
 lair+0.177
cks+0.174
 mess+0.167
 pads+0.163
 floor+0.162
Neg logit contributions
 help-0.180
 reward-0.178
ar-0.173
 plunge-0.171
bye-0.165
Token some
Token ID617
Feature activation-2.28e-01

Pos logit contributions
ar+0.105
bye+0.104
ted+0.097
 reward+0.096
 plunge+0.095
Neg logit contributions
 mess-0.131
cks-0.127
 lair-0.127
 floor-0.127
 neck-0.124
Token ice
Token ID4771
Feature activation-1.83e+00

Pos logit contributions
ar+0.016
bye+0.015
 reward+0.015
Sure+0.015
 plunge+0.015
Neg logit contributions
 mess-0.019
 lair-0.019
 floor-0.018
cks-0.018
 pads-0.018
Token-
Token ID12
Feature activation-3.61e+00

Pos logit contributions
 reward+0.125
 plunge+0.124
 help+0.122
ar+0.117
Sure+0.117
Neg logit contributions
cks-0.149
 pads-0.149
 mess-0.148
 lair-0.147
 floor-0.146
Tokencream
Token ID36277
Feature activation-4.49e+00

Pos logit contributions
 reward+0.288
 plunge+0.270
 help+0.263
mate+0.246
 awaited+0.231
Neg logit contributions
 pads-0.263
cks-0.253
llers-0.251
 necks-0.247
opl-0.222
Token.
Token ID13
Feature activation+1.81e-01

Pos logit contributions
 reward+0.331
 help+0.328
ar+0.326
 plunge+0.324
Sure+0.309
Neg logit contributions
 floor-0.338
 pads-0.323
 mess-0.321
 lair-0.321
 sizes-0.302
Token He
Token ID679
Feature activation+8.24e-02

Pos logit contributions
cks+0.016
 lair+0.016
 mess+0.015
 necks+0.015
 pads+0.015
Neg logit contributions
 help-0.012
 reward-0.012
Sure-0.012
 plunge-0.011
ar-0.011
Token smiled
Token ID13541
Feature activation-8.15e-01

Pos logit contributions
cks+0.007
 lair+0.007
 mess+0.007
 pads+0.006
 floor+0.006
Neg logit contributions
 help-0.006
 reward-0.006
ar-0.006
Sure-0.005
 plunge-0.005
Token and
Token ID290
Feature activation+1.93e-01

Pos logit contributions
 help+0.054
 reward+0.053
less+0.050
ar+0.050
bye+0.049
Neg logit contributions
cks-0.076
 lair-0.072
 mess-0.069
 floor-0.068
 pads-0.067
Token was
Token ID373
Feature activation+3.56e-01

Pos logit contributions
 lair+0.015
cks+0.015
 mess+0.015
 pads+0.014
 floor+0.014
Neg logit contributions
 help-0.014
 reward-0.014
ar-0.014
 plunge-0.014
bye-0.013